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/baseline/wandb/run-20240218_170205-iu28me1d
wandb: Run `wandb offline` to turn off syncing.
wandb: Syncing run auspicious-lamp-5
wandb: ⭐️ View project at https://wandb.ai/laurence_r/colorful-llama
wandb: 🚀 View run at https://wandb.ai/laurence_r/colorful-llama/runs/iu28me1d
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.

Downloading readme:   0%|          | 0.00/11.6k [00:00<?, ?B/s]
Downloading readme: 100%|██████████| 11.6k/11.6k [00:00<00:00, 8.91MB/s]

Downloading data files:   0%|          | 0/1 [00:00<?, ?it/s]

Downloading data:   0%|          | 0.00/44.3M [00:00<?, ?B/s]

Downloading data:   9%|▉         | 4.19M/44.3M [00:00<00:04, 8.26MB/s]

Downloading data:  28%|██▊       | 12.6M/44.3M [00:00<00:01, 18.0MB/s]

Downloading data:  47%|████▋     | 21.0M/44.3M [00:01<00:01, 20.3MB/s]

Downloading data:  66%|██████▋   | 29.4M/44.3M [00:01<00:00, 25.2MB/s]

Downloading data:  85%|████████▌ | 37.7M/44.3M [00:01<00:00, 25.6MB/s]

Downloading data: 100%|██████████| 44.3M/44.3M [00:01<00:00, 29.2MB/s]
Downloading data: 100%|██████████| 44.3M/44.3M [00:01<00:00, 23.6MB/s]

Downloading data files: 100%|██████████| 1/1 [00:01<00:00,  1.88s/it]
Downloading data files: 100%|██████████| 1/1 [00:01<00:00,  1.88s/it]

Extracting data files:   0%|          | 0/1 [00:00<?, ?it/s]
Extracting data files: 100%|██████████| 1/1 [00:00<00:00, 600.73it/s]

Generating train split: 0 examples [00:00, ? examples/s]
Generating train split: 51760 examples [00:00, 76320.61 examples/s]
Generating train split: 51760 examples [00:00, 74445.25 examples/s]
Dataset and Sampler are initialized.

  0%|          | 0/6470 [00:00<?, ?it/s]
1|1|Loss: 1.501872181892395:   0%|          | 0/6470 [00:40<?, ?it/s]
1|1|Loss: 1.501872181892395:   0%|          | 1/6470 [01:10<127:11:31, 70.78s/it]
1|2|Loss: 1.294187307357788:   0%|          | 1/6470 [02:44<127:11:31, 70.78s/it]
1|2|Loss: 1.294187307357788:   0%|          | 2/6470 [02:44<151:42:16, 84.44s/it]
1|3|Loss: 1.2840849161148071:   0%|          | 2/6470 [02:45<151:42:16, 84.44s/it]
1|3|Loss: 1.2840849161148071:   0%|          | 3/6470 [02:45<83:10:46, 46.30s/it] 
1|4|Loss: 1.3659697771072388:   0%|          | 3/6470 [02:45<83:10:46, 46.30s/it]
1|4|Loss: 1.3659697771072388:   0%|          | 4/6470 [02:46<50:41:02, 28.22s/it]
1|5|Loss: 1.2847164869308472:   0%|          | 4/6470 [02:46<50:41:02, 28.22s/it]
1|5|Loss: 1.2847164869308472:   0%|          | 5/6470 [02:46<32:48:20, 18.27s/it]
1|6|Loss: 1.4036566019058228:   0%|          | 5/6470 [02:47<32:48:20, 18.27s/it]
1|6|Loss: 1.4036566019058228:   0%|          | 6/6470 [02:47<21:59:25, 12.25s/it]
1|7|Loss: 1.6473394632339478:   0%|          | 6/6470 [02:47<21:59:25, 12.25s/it]
1|7|Loss: 1.6473394632339478:   0%|          | 7/6470 [02:48<15:12:05,  8.47s/it]
1|8|Loss: 1.3102757930755615:   0%|          | 7/6470 [02:48<15:12:05,  8.47s/it]
1|8|Loss: 1.3102757930755615:   0%|          | 8/6470 [02:48<10:48:36,  6.02s/it]
1|9|Loss: 1.533294677734375:   0%|          | 8/6470 [02:49<10:48:36,  6.02s/it] 
1|9|Loss: 1.533294677734375:   0%|          | 9/6470 [02:49<7:50:40,  4.37s/it] 
1|10|Loss: 1.3988475799560547:   0%|          | 9/6470 [02:49<7:50:40,  4.37s/it]
1|10|Loss: 1.3988475799560547:   0%|          | 10/6470 [02:50<5:48:27,  3.24s/it]
1|11|Loss: 1.4548633098602295:   0%|          | 10/6470 [02:50<5:48:27,  3.24s/it]
1|11|Loss: 1.4548633098602295:   0%|          | 11/6470 [02:51<4:26:29,  2.48s/it]
1|12|Loss: 1.2906246185302734:   0%|          | 11/6470 [02:51<4:26:29,  2.48s/it]
1|12|Loss: 1.2906246185302734:   0%|          | 12/6470 [02:51<3:32:52,  1.98s/it]
1|13|Loss: 1.4396921396255493:   0%|          | 12/6470 [02:52<3:32:52,  1.98s/it]
1|13|Loss: 1.4396921396255493:   0%|          | 13/6470 [02:52<2:53:07,  1.61s/it]
1|14|Loss: 1.5497629642486572:   0%|          | 13/6470 [02:52<2:53:07,  1.61s/it]
1|14|Loss: 1.5497629642486572:   0%|          | 14/6470 [02:53<2:18:31,  1.29s/it]
1|15|Loss: 1.2908661365509033:   0%|          | 14/6470 [02:53<2:18:31,  1.29s/it]
1|15|Loss: 1.2908661365509033:   0%|          | 15/6470 [02:53<1:56:21,  1.08s/it]
1|16|Loss: 1.3362915515899658:   0%|          | 15/6470 [02:54<1:56:21,  1.08s/it]
1|16|Loss: 1.3362915515899658:   0%|          | 16/6470 [02:54<1:44:24,  1.03it/s]
1|17|Loss: 0.9975453019142151:   0%|          | 16/6470 [02:54<1:44:24,  1.03it/s]
1|17|Loss: 0.9975453019142151:   0%|          | 17/6470 [02:55<1:42:05,  1.05it/s]
1|18|Loss: 1.2147175073623657:   0%|          | 17/6470 [02:55<1:42:05,  1.05it/s]
1|18|Loss: 1.2147175073623657:   0%|          | 18/6470 [02:56<1:36:38,  1.11it/s]
1|19|Loss: 1.4077593088150024:   0%|          | 18/6470 [02:56<1:36:38,  1.11it/s]
1|19|Loss: 1.4077593088150024:   0%|          | 19/6470 [02:56<1:28:39,  1.21it/s]
1|20|Loss: 1.244315505027771:   0%|          | 19/6470 [02:57<1:28:39,  1.21it/s] 
1|20|Loss: 1.244315505027771:   0%|          | 20/6470 [02:57<1:27:37,  1.23it/s]
1|21|Loss: 1.4736154079437256:   0%|          | 20/6470 [02:58<1:27:37,  1.23it/s]
1|21|Loss: 1.4736154079437256:   0%|          | 21/6470 [02:58<1:25:29,  1.26it/s]
1|22|Loss: 1.3675404787063599:   0%|          | 21/6470 [02:58<1:25:29,  1.26it/s]
1|22|Loss: 1.3675404787063599:   0%|          | 22/6470 [02:59<1:20:42,  1.33it/s]
1|23|Loss: 1.2888811826705933:   0%|          | 22/6470 [02:59<1:20:42,  1.33it/s]
1|23|Loss: 1.2888811826705933:   0%|          | 23/6470 [02:59<1:18:29,  1.37it/s]
1|24|Loss: 1.3863017559051514:   0%|          | 23/6470 [03:00<1:18:29,  1.37it/s]
1|24|Loss: 1.3863017559051514:   0%|          | 24/6470 [03:00<1:17:16,  1.39it/s]
1|25|Loss: 1.3230516910552979:   0%|          | 24/6470 [03:00<1:17:16,  1.39it/s]
1|25|Loss: 1.3230516910552979:   0%|          | 25/6470 [03:01<1:19:28,  1.35it/s]
1|26|Loss: 1.4430614709854126:   0%|          | 25/6470 [03:01<1:19:28,  1.35it/s]
1|26|Loss: 1.4430614709854126:   0%|          | 26/6470 [03:01<1:19:23,  1.35it/s]
1|27|Loss: 1.3046883344650269:   0%|          | 26/6470 [03:02<1:19:23,  1.35it/s]
1|27|Loss: 1.3046883344650269:   0%|          | 27/6470 [03:02<1:19:07,  1.36it/s]
1|28|Loss: 1.338098406791687:   0%|          | 27/6470 [03:03<1:19:07,  1.36it/s] 
1|28|Loss: 1.338098406791687:   0%|          | 28/6470 [03:03<1:17:56,  1.38it/s]
1|29|Loss: 1.564326524734497:   0%|          | 28/6470 [03:03<1:17:56,  1.38it/s]
1|29|Loss: 1.564326524734497:   0%|          | 29/6470 [03:03<1:13:24,  1.46it/s]
1|30|Loss: 1.5711710453033447:   0%|          | 29/6470 [03:04<1:13:24,  1.46it/s]
1|30|Loss: 1.5711710453033447:   0%|          | 30/6470 [03:04<1:08:02,  1.58it/s]
1|31|Loss: 1.2188606262207031:   0%|          | 30/6470 [03:04<1:08:02,  1.58it/s]
1|31|Loss: 1.2188606262207031:   0%|          | 31/6470 [03:05<1:09:30,  1.54it/s]
1|32|Loss: 1.3671900033950806:   0%|          | 31/6470 [03:05<1:09:30,  1.54it/s]
1|32|Loss: 1.3671900033950806:   0%|          | 32/6470 [03:05<1:15:13,  1.43it/s]
1|33|Loss: 1.4637751579284668:   0%|          | 32/6470 [03:06<1:15:13,  1.43it/s]
1|33|Loss: 1.4637751579284668:   1%|          | 33/6470 [03:06<1:12:15,  1.48it/s]
1|34|Loss: 1.2203490734100342:   1%|          | 33/6470 [03:06<1:12:15,  1.48it/s]
1|34|Loss: 1.2203490734100342:   1%|          | 34/6470 [03:07<1:15:46,  1.42it/s]
1|35|Loss: 1.1782612800598145:   1%|          | 34/6470 [03:07<1:15:46,  1.42it/s]
1|35|Loss: 1.1782612800598145:   1%|          | 35/6470 [03:08<1:20:48,  1.33it/s]
1|36|Loss: 1.4127483367919922:   1%|          | 35/6470 [03:08<1:20:48,  1.33it/s]
1|36|Loss: 1.4127483367919922:   1%|          | 36/6470 [03:08<1:20:54,  1.33it/s]
1|37|Loss: 1.198026418685913:   1%|          | 36/6470 [03:09<1:20:54,  1.33it/s] 
1|37|Loss: 1.198026418685913:   1%|          | 37/6470 [03:09<1:23:32,  1.28it/s]
1|38|Loss: 1.371580958366394:   1%|          | 37/6470 [03:10<1:23:32,  1.28it/s]
1|38|Loss: 1.371580958366394:   1%|          | 38/6470 [03:10<1:24:42,  1.27it/s]
1|39|Loss: 1.5584805011749268:   1%|          | 38/6470 [03:11<1:24:42,  1.27it/s]
1|39|Loss: 1.5584805011749268:   1%|          | 39/6470 [03:11<1:24:23,  1.27it/s]
1|40|Loss: 1.4363775253295898:   1%|          | 39/6470 [03:11<1:24:23,  1.27it/s]
1|40|Loss: 1.4363775253295898:   1%|          | 40/6470 [03:12<1:26:06,  1.24it/s]
1|41|Loss: 1.3500676155090332:   1%|          | 40/6470 [03:12<1:26:06,  1.24it/s]
1|41|Loss: 1.3500676155090332:   1%|          | 41/6470 [03:12<1:22:05,  1.31it/s]
1|42|Loss: 1.5524228811264038:   1%|          | 41/6470 [03:13<1:22:05,  1.31it/s]
1|42|Loss: 1.5524228811264038:   1%|          | 42/6470 [03:13<1:17:37,  1.38it/s]
1|43|Loss: 1.3458174467086792:   1%|          | 42/6470 [03:13<1:17:37,  1.38it/s]
1|43|Loss: 1.3458174467086792:   1%|          | 43/6470 [03:14<1:16:11,  1.41it/s]
1|44|Loss: 1.3237074613571167:   1%|          | 43/6470 [03:14<1:16:11,  1.41it/s]
1|44|Loss: 1.3237074613571167:   1%|          | 44/6470 [03:14<1:10:38,  1.52it/s]
1|45|Loss: 1.2205528020858765:   1%|          | 44/6470 [03:15<1:10:38,  1.52it/s]
1|45|Loss: 1.2205528020858765:   1%|          | 45/6470 [03:15<1:13:34,  1.46it/s]
1|46|Loss: 1.5594590902328491:   1%|          | 45/6470 [03:15<1:13:34,  1.46it/s]
1|46|Loss: 1.5594590902328491:   1%|          | 46/6470 [03:16<1:10:43,  1.51it/s]
1|47|Loss: 1.077263593673706:   1%|          | 46/6470 [03:16<1:10:43,  1.51it/s] 
1|47|Loss: 1.077263593673706:   1%|          | 47/6470 [03:16<1:13:47,  1.45it/s]
1|48|Loss: 1.6814899444580078:   1%|          | 47/6470 [03:17<1:13:47,  1.45it/s]
1|48|Loss: 1.6814899444580078:   1%|          | 48/6470 [03:17<1:12:51,  1.47it/s]
1|49|Loss: 1.1635417938232422:   1%|          | 48/6470 [03:17<1:12:51,  1.47it/s]
1|49|Loss: 1.1635417938232422:   1%|          | 49/6470 [03:18<1:18:59,  1.35it/s]
1|50|Loss: 1.515518069267273:   1%|          | 49/6470 [03:18<1:18:59,  1.35it/s] 
1|50|Loss: 1.515518069267273:   1%|          | 50/6470 [03:19<1:24:11,  1.27it/s]
1|51|Loss: 1.2879090309143066:   1%|          | 50/6470 [03:19<1:24:11,  1.27it/s]
1|51|Loss: 1.2879090309143066:   1%|          | 51/6470 [03:20<1:22:58,  1.29it/s]
1|52|Loss: 1.3329590559005737:   1%|          | 51/6470 [03:20<1:22:58,  1.29it/s]
1|52|Loss: 1.3329590559005737:   1%|          | 52/6470 [03:20<1:25:27,  1.25it/s]
1|53|Loss: 1.5004225969314575:   1%|          | 52/6470 [03:21<1:25:27,  1.25it/s]
1|53|Loss: 1.5004225969314575:   1%|          | 53/6470 [03:21<1:24:13,  1.27it/s]
1|54|Loss: 1.6034446954727173:   1%|          | 53/6470 [03:22<1:24:13,  1.27it/s]
1|54|Loss: 1.6034446954727173:   1%|          | 54/6470 [03:22<1:19:20,  1.35it/s]
1|55|Loss: 1.319648265838623:   1%|          | 54/6470 [03:22<1:19:20,  1.35it/s] 
1|55|Loss: 1.319648265838623:   1%|          | 55/6470 [03:23<1:17:58,  1.37it/s]
1|56|Loss: 1.1424591541290283:   1%|          | 55/6470 [03:23<1:17:58,  1.37it/s]
1|56|Loss: 1.1424591541290283:   1%|          | 56/6470 [03:23<1:20:19,  1.33it/s]
1|57|Loss: 1.508151888847351:   1%|          | 56/6470 [03:24<1:20:19,  1.33it/s] 
1|57|Loss: 1.508151888847351:   1%|          | 57/6470 [03:24<1:15:22,  1.42it/s]
1|58|Loss: 1.3826780319213867:   1%|          | 57/6470 [03:24<1:15:22,  1.42it/s]
1|58|Loss: 1.3826780319213867:   1%|          | 58/6470 [03:25<1:12:17,  1.48it/s]
1|59|Loss: 1.2150849103927612:   1%|          | 58/6470 [03:25<1:12:17,  1.48it/s]
1|59|Loss: 1.2150849103927612:   1%|          | 59/6470 [03:25<1:13:08,  1.46it/s]
1|60|Loss: 1.3185876607894897:   1%|          | 59/6470 [03:26<1:13:08,  1.46it/s]
1|60|Loss: 1.3185876607894897:   1%|          | 60/6470 [03:26<1:17:40,  1.38it/s]
1|61|Loss: 1.048214077949524:   1%|          | 60/6470 [03:26<1:17:40,  1.38it/s] 
1|61|Loss: 1.048214077949524:   1%|          | 61/6470 [03:27<1:19:13,  1.35it/s]
1|62|Loss: 1.3661811351776123:   1%|          | 61/6470 [03:27<1:19:13,  1.35it/s]
1|62|Loss: 1.3661811351776123:   1%|          | 62/6470 [03:27<1:16:08,  1.40it/s]
1|63|Loss: 1.2318936586380005:   1%|          | 62/6470 [03:28<1:16:08,  1.40it/s]
1|63|Loss: 1.2318936586380005:   1%|          | 63/6470 [03:28<1:13:33,  1.45it/s]
1|64|Loss: 1.2985602617263794:   1%|          | 63/6470 [03:29<1:13:33,  1.45it/s]
1|64|Loss: 1.2985602617263794:   1%|          | 64/6470 [03:29<1:17:58,  1.37it/s]
1|65|Loss: 1.2766863107681274:   1%|          | 64/6470 [03:29<1:17:58,  1.37it/s]
1|65|Loss: 1.2766863107681274:   1%|          | 65/6470 [03:30<1:20:54,  1.32it/s]
1|66|Loss: 1.3334022760391235:   1%|          | 65/6470 [03:30<1:20:54,  1.32it/s]
1|66|Loss: 1.3334022760391235:   1%|          | 66/6470 [03:30<1:19:40,  1.34it/s]
1|67|Loss: 1.560469627380371:   1%|          | 66/6470 [03:31<1:19:40,  1.34it/s] 
1|67|Loss: 1.560469627380371:   1%|          | 67/6470 [03:31<1:17:49,  1.37it/s]
1|68|Loss: 1.2271676063537598:   1%|          | 67/6470 [03:32<1:17:49,  1.37it/s]
1|68|Loss: 1.2271676063537598:   1%|          | 68/6470 [03:32<1:20:57,  1.32it/s]
1|69|Loss: 1.2802866697311401:   1%|          | 68/6470 [03:32<1:20:57,  1.32it/s]
1|69|Loss: 1.2802866697311401:   1%|          | 69/6470 [03:33<1:18:11,  1.36it/s]
1|70|Loss: 1.3294180631637573:   1%|          | 69/6470 [03:33<1:18:11,  1.36it/s]
1|70|Loss: 1.3294180631637573:   1%|          | 70/6470 [03:33<1:16:23,  1.40it/s]
1|71|Loss: 1.4568904638290405:   1%|          | 70/6470 [03:34<1:16:23,  1.40it/s]
1|71|Loss: 1.4568904638290405:   1%|          | 71/6470 [03:34<1:18:50,  1.35it/s]
1|72|Loss: 1.372212290763855:   1%|          | 71/6470 [03:35<1:18:50,  1.35it/s] 
1|72|Loss: 1.372212290763855:   1%|          | 72/6470 [03:35<1:18:16,  1.36it/s]
1|73|Loss: 1.422770619392395:   1%|          | 72/6470 [03:35<1:18:16,  1.36it/s]
1|73|Loss: 1.422770619392395:   1%|          | 73/6470 [03:36<1:18:12,  1.36it/s]
1|74|Loss: 1.3050442934036255:   1%|          | 73/6470 [03:36<1:18:12,  1.36it/s]
1|74|Loss: 1.3050442934036255:   1%|          | 74/6470 [03:36<1:15:17,  1.42it/s]
1|75|Loss: 1.244707703590393:   1%|          | 74/6470 [03:37<1:15:17,  1.42it/s] 
1|75|Loss: 1.244707703590393:   1%|          | 75/6470 [03:37<1:15:40,  1.41it/s]
1|76|Loss: 1.218338131904602:   1%|          | 75/6470 [03:37<1:15:40,  1.41it/s]
1|76|Loss: 1.218338131904602:   1%|          | 76/6470 [03:38<1:21:04,  1.31it/s]
1|77|Loss: 1.454959750175476:   1%|          | 76/6470 [03:38<1:21:04,  1.31it/s]
1|77|Loss: 1.454959750175476:   1%|          | 77/6470 [03:38<1:17:22,  1.38it/s]
1|78|Loss: 1.4058265686035156:   1%|          | 77/6470 [03:39<1:17:22,  1.38it/s]
1|78|Loss: 1.4058265686035156:   1%|          | 78/6470 [03:39<1:15:36,  1.41it/s]
1|79|Loss: 1.446523904800415:   1%|          | 78/6470 [03:40<1:15:36,  1.41it/s] 
1|79|Loss: 1.446523904800415:   1%|          | 79/6470 [03:40<1:13:31,  1.45it/s]
1|80|Loss: 1.3090981245040894:   1%|          | 79/6470 [03:40<1:13:31,  1.45it/s]
1|80|Loss: 1.3090981245040894:   1%|          | 80/6470 [03:40<1:13:17,  1.45it/s]
1|81|Loss: 1.4209113121032715:   1%|          | 80/6470 [03:41<1:13:17,  1.45it/s]
1|81|Loss: 1.4209113121032715:   1%|▏         | 81/6470 [03:41<1:09:38,  1.53it/s]
1|82|Loss: 1.268842339515686:   1%|▏         | 81/6470 [03:41<1:09:38,  1.53it/s] 
1|82|Loss: 1.268842339515686:   1%|▏         | 82/6470 [03:42<1:07:16,  1.58it/s]
1|83|Loss: 1.3746179342269897:   1%|▏         | 82/6470 [03:42<1:07:16,  1.58it/s]
1|83|Loss: 1.3746179342269897:   1%|▏         | 83/6470 [03:42<1:06:23,  1.60it/s]
1|84|Loss: 1.2939040660858154:   1%|▏         | 83/6470 [03:43<1:06:23,  1.60it/s]
1|84|Loss: 1.2939040660858154:   1%|▏         | 84/6470 [03:43<1:09:01,  1.54it/s]
1|85|Loss: 1.193249225616455:   1%|▏         | 84/6470 [03:43<1:09:01,  1.54it/s] 
1|85|Loss: 1.193249225616455:   1%|▏         | 85/6470 [03:44<1:09:12,  1.54it/s]
1|86|Loss: 1.252859115600586:   1%|▏         | 85/6470 [03:44<1:09:12,  1.54it/s]
1|86|Loss: 1.252859115600586:   1%|▏         | 86/6470 [03:44<1:10:38,  1.51it/s]
1|87|Loss: 1.4807300567626953:   1%|▏         | 86/6470 [03:45<1:10:38,  1.51it/s]
1|87|Loss: 1.4807300567626953:   1%|▏         | 87/6470 [03:45<1:08:51,  1.54it/s]
1|88|Loss: 1.2177728414535522:   1%|▏         | 87/6470 [03:45<1:08:51,  1.54it/s]
1|88|Loss: 1.2177728414535522:   1%|▏         | 88/6470 [03:46<1:08:07,  1.56it/s]
1|89|Loss: 1.3290730714797974:   1%|▏         | 88/6470 [03:46<1:08:07,  1.56it/s]
1|89|Loss: 1.3290730714797974:   1%|▏         | 89/6470 [03:46<1:10:20,  1.51it/s]
1|90|Loss: 1.2620337009429932:   1%|▏         | 89/6470 [03:47<1:10:20,  1.51it/s]
1|90|Loss: 1.2620337009429932:   1%|▏         | 90/6470 [03:47<1:10:11,  1.51it/s]
1|91|Loss: 1.0900152921676636:   1%|▏         | 90/6470 [03:47<1:10:11,  1.51it/s]
1|91|Loss: 1.0900152921676636:   1%|▏         | 91/6470 [03:48<1:16:35,  1.39it/s]
1|92|Loss: 1.247839093208313:   1%|▏         | 91/6470 [03:48<1:16:35,  1.39it/s] 
1|92|Loss: 1.247839093208313:   1%|▏         | 92/6470 [03:49<1:21:10,  1.31it/s]
1|93|Loss: 1.2043358087539673:   1%|▏         | 92/6470 [03:49<1:21:10,  1.31it/s]
1|93|Loss: 1.2043358087539673:   1%|▏         | 93/6470 [03:49<1:21:23,  1.31it/s]
1|94|Loss: 1.5492631196975708:   1%|▏         | 93/6470 [03:50<1:21:23,  1.31it/s]
1|94|Loss: 1.5492631196975708:   1%|▏         | 94/6470 [03:50<1:15:10,  1.41it/s]
1|95|Loss: 1.4053503274917603:   1%|▏         | 94/6470 [03:50<1:15:10,  1.41it/s]
1|95|Loss: 1.4053503274917603:   1%|▏         | 95/6470 [03:51<1:10:37,  1.50it/s]
1|96|Loss: 1.3689202070236206:   1%|▏         | 95/6470 [03:51<1:10:37,  1.50it/s]
1|96|Loss: 1.3689202070236206:   1%|▏         | 96/6470 [03:51<1:07:38,  1.57it/s]
1|97|Loss: 1.6626442670822144:   1%|▏         | 96/6470 [03:51<1:07:38,  1.57it/s]
1|97|Loss: 1.6626442670822144:   1%|▏         | 97/6470 [03:52<1:04:22,  1.65it/s]
1|98|Loss: 1.872829556465149:   1%|▏         | 97/6470 [03:52<1:04:22,  1.65it/s] 
1|98|Loss: 1.872829556465149:   2%|▏         | 98/6470 [03:52<58:07,  1.83it/s]  
1|99|Loss: 1.462458848953247:   2%|▏         | 98/6470 [03:52<58:07,  1.83it/s]
1|99|Loss: 1.462458848953247:   2%|▏         | 99/6470 [03:53<1:01:02,  1.74it/s]
1|100|Loss: 1.2091480493545532:   2%|▏         | 99/6470 [03:53<1:01:02,  1.74it/s]
1|100|Loss: 1.2091480493545532:   2%|▏         | 100/6470 [03:53<1:08:06,  1.56it/s]
1|101|Loss: 1.2614188194274902:   2%|▏         | 100/6470 [03:54<1:08:06,  1.56it/s]
1|101|Loss: 1.2614188194274902:   2%|▏         | 101/6470 [03:54<1:12:02,  1.47it/s]
1|102|Loss: 1.3228704929351807:   2%|▏         | 101/6470 [03:55<1:12:02,  1.47it/s]
1|102|Loss: 1.3228704929351807:   2%|▏         | 102/6470 [03:55<1:17:16,  1.37it/s]
1|103|Loss: 1.499592900276184:   2%|▏         | 102/6470 [03:55<1:17:16,  1.37it/s] 
1|103|Loss: 1.499592900276184:   2%|▏         | 103/6470 [03:56<1:14:08,  1.43it/s]
1|104|Loss: 1.3651882410049438:   2%|▏         | 103/6470 [03:56<1:14:08,  1.43it/s]
1|104|Loss: 1.3651882410049438:   2%|▏         | 104/6470 [03:56<1:15:45,  1.40it/s]
1|105|Loss: 1.5454292297363281:   2%|▏         | 104/6470 [03:57<1:15:45,  1.40it/s]
1|105|Loss: 1.5454292297363281:   2%|▏         | 105/6470 [03:57<1:14:19,  1.43it/s]
1|106|Loss: 1.3916610479354858:   2%|▏         | 105/6470 [03:58<1:14:19,  1.43it/s]
1|106|Loss: 1.3916610479354858:   2%|▏         | 106/6470 [03:58<1:14:46,  1.42it/s]
1|107|Loss: 1.2795780897140503:   2%|▏         | 106/6470 [03:58<1:14:46,  1.42it/s]
1|107|Loss: 1.2795780897140503:   2%|▏         | 107/6470 [03:59<1:14:53,  1.42it/s]
1|108|Loss: 1.209450364112854:   2%|▏         | 107/6470 [03:59<1:14:53,  1.42it/s] 
1|108|Loss: 1.209450364112854:   2%|▏         | 108/6470 [03:59<1:11:15,  1.49it/s]
1|109|Loss: 1.6037557125091553:   2%|▏         | 108/6470 [03:59<1:11:15,  1.49it/s]
1|109|Loss: 1.6037557125091553:   2%|▏         | 109/6470 [04:00<1:06:59,  1.58it/s]
1|110|Loss: 1.3208732604980469:   2%|▏         | 109/6470 [04:00<1:06:59,  1.58it/s]
1|110|Loss: 1.3208732604980469:   2%|▏         | 110/6470 [04:00<1:11:53,  1.47it/s]
1|111|Loss: 1.0508133172988892:   2%|▏         | 110/6470 [04:01<1:11:53,  1.47it/s]
1|111|Loss: 1.0508133172988892:   2%|▏         | 111/6470 [04:01<1:14:49,  1.42it/s]
1|112|Loss: 1.4829890727996826:   2%|▏         | 111/6470 [04:02<1:14:49,  1.42it/s]
1|112|Loss: 1.4829890727996826:   2%|▏         | 112/6470 [04:02<1:12:49,  1.46it/s]
1|113|Loss: 1.4814391136169434:   2%|▏         | 112/6470 [04:02<1:12:49,  1.46it/s]
1|113|Loss: 1.4814391136169434:   2%|▏         | 113/6470 [04:02<1:07:37,  1.57it/s]
1|114|Loss: 1.3033864498138428:   2%|▏         | 113/6470 [04:03<1:07:37,  1.57it/s]
1|114|Loss: 1.3033864498138428:   2%|▏         | 114/6470 [04:03<1:06:38,  1.59it/s]
1|115|Loss: 1.4121286869049072:   2%|▏         | 114/6470 [04:03<1:06:38,  1.59it/s]
1|115|Loss: 1.4121286869049072:   2%|▏         | 115/6470 [04:04<1:06:19,  1.60it/s]
1|116|Loss: 1.1951618194580078:   2%|▏         | 115/6470 [04:04<1:06:19,  1.60it/s]
1|116|Loss: 1.1951618194580078:   2%|▏         | 116/6470 [04:04<1:06:02,  1.60it/s]
1|117|Loss: 1.3016176223754883:   2%|▏         | 116/6470 [04:05<1:06:02,  1.60it/s]
1|117|Loss: 1.3016176223754883:   2%|▏         | 117/6470 [04:05<1:11:41,  1.48it/s]
1|118|Loss: 1.2518032789230347:   2%|▏         | 117/6470 [04:05<1:11:41,  1.48it/s]
1|118|Loss: 1.2518032789230347:   2%|▏         | 118/6470 [04:06<1:13:19,  1.44it/s]
1|119|Loss: 1.1773594617843628:   2%|▏         | 118/6470 [04:06<1:13:19,  1.44it/s]
1|119|Loss: 1.1773594617843628:   2%|▏         | 119/6470 [04:06<1:13:33,  1.44it/s]
1|120|Loss: 1.3703696727752686:   2%|▏         | 119/6470 [04:07<1:13:33,  1.44it/s]
1|120|Loss: 1.3703696727752686:   2%|▏         | 120/6470 [04:07<1:14:56,  1.41it/s]
1|121|Loss: 1.3716299533843994:   2%|▏         | 120/6470 [04:08<1:14:56,  1.41it/s]
1|121|Loss: 1.3716299533843994:   2%|▏         | 121/6470 [04:08<1:13:33,  1.44it/s]
1|122|Loss: 1.2628839015960693:   2%|▏         | 121/6470 [04:08<1:13:33,  1.44it/s]
1|122|Loss: 1.2628839015960693:   2%|▏         | 122/6470 [04:09<1:15:24,  1.40it/s]
1|123|Loss: 1.2962628602981567:   2%|▏         | 122/6470 [04:09<1:15:24,  1.40it/s]
1|123|Loss: 1.2962628602981567:   2%|▏         | 123/6470 [04:09<1:11:48,  1.47it/s]
1|124|Loss: 1.5098304748535156:   2%|▏         | 123/6470 [04:10<1:11:48,  1.47it/s]
1|124|Loss: 1.5098304748535156:   2%|▏         | 124/6470 [04:10<1:07:38,  1.56it/s]
1|125|Loss: 1.2199630737304688:   2%|▏         | 124/6470 [04:10<1:07:38,  1.56it/s]
1|125|Loss: 1.2199630737304688:   2%|▏         | 125/6470 [04:11<1:12:46,  1.45it/s]
1|126|Loss: 1.4863845109939575:   2%|▏         | 125/6470 [04:11<1:12:46,  1.45it/s]
1|126|Loss: 1.4863845109939575:   2%|▏         | 126/6470 [04:11<1:12:56,  1.45it/s]
1|127|Loss: 1.2724088430404663:   2%|▏         | 126/6470 [04:12<1:12:56,  1.45it/s]
1|127|Loss: 1.2724088430404663:   2%|▏         | 127/6470 [04:12<1:11:31,  1.48it/s]
1|128|Loss: 1.4528114795684814:   2%|▏         | 127/6470 [04:12<1:11:31,  1.48it/s]
1|128|Loss: 1.4528114795684814:   2%|▏         | 128/6470 [04:13<1:09:04,  1.53it/s]
1|129|Loss: 1.4307026863098145:   2%|▏         | 128/6470 [04:13<1:09:04,  1.53it/s]
1|129|Loss: 1.4307026863098145:   2%|▏         | 129/6470 [04:13<1:04:59,  1.63it/s]
1|130|Loss: 1.2629714012145996:   2%|▏         | 129/6470 [04:13<1:04:59,  1.63it/s]
1|130|Loss: 1.2629714012145996:   2%|▏         | 130/6470 [04:14<1:09:40,  1.52it/s]
1|131|Loss: 1.2863285541534424:   2%|▏         | 130/6470 [04:14<1:09:40,  1.52it/s]
1|131|Loss: 1.2863285541534424:   2%|▏         | 131/6470 [04:15<1:12:01,  1.47it/s]
1|132|Loss: 1.2628964185714722:   2%|▏         | 131/6470 [04:15<1:12:01,  1.47it/s]
1|132|Loss: 1.2628964185714722:   2%|▏         | 132/6470 [04:15<1:10:44,  1.49it/s]
1|133|Loss: 1.201406478881836:   2%|▏         | 132/6470 [04:16<1:10:44,  1.49it/s] 
1|133|Loss: 1.201406478881836:   2%|▏         | 133/6470 [04:16<1:09:40,  1.52it/s]
1|134|Loss: 1.3734631538391113:   2%|▏         | 133/6470 [04:16<1:09:40,  1.52it/s]
1|134|Loss: 1.3734631538391113:   2%|▏         | 134/6470 [04:17<1:13:41,  1.43it/s]
1|135|Loss: 1.210668921470642:   2%|▏         | 134/6470 [04:17<1:13:41,  1.43it/s] 
1|135|Loss: 1.210668921470642:   2%|▏         | 135/6470 [04:17<1:13:03,  1.45it/s]
1|136|Loss: 1.4851224422454834:   2%|▏         | 135/6470 [04:18<1:13:03,  1.45it/s]
1|136|Loss: 1.4851224422454834:   2%|▏         | 136/6470 [04:18<1:10:46,  1.49it/s]
1|137|Loss: 1.4348951578140259:   2%|▏         | 136/6470 [04:18<1:10:46,  1.49it/s]
1|137|Loss: 1.4348951578140259:   2%|▏         | 137/6470 [04:19<1:08:24,  1.54it/s]
1|138|Loss: 1.132756233215332:   2%|▏         | 137/6470 [04:19<1:08:24,  1.54it/s] 
1|138|Loss: 1.132756233215332:   2%|▏         | 138/6470 [04:19<1:10:12,  1.50it/s]
1|139|Loss: 1.3337385654449463:   2%|▏         | 138/6470 [04:20<1:10:12,  1.50it/s]
1|139|Loss: 1.3337385654449463:   2%|▏         | 139/6470 [04:20<1:10:02,  1.51it/s]
1|140|Loss: 1.3698749542236328:   2%|▏         | 139/6470 [04:20<1:10:02,  1.51it/s]
1|140|Loss: 1.3698749542236328:   2%|▏         | 140/6470 [04:21<1:11:56,  1.47it/s]
1|141|Loss: 1.2947603464126587:   2%|▏         | 140/6470 [04:21<1:11:56,  1.47it/s]
1|141|Loss: 1.2947603464126587:   2%|▏         | 141/6470 [04:21<1:09:42,  1.51it/s]
1|142|Loss: 1.179447889328003:   2%|▏         | 141/6470 [04:22<1:09:42,  1.51it/s] 
1|142|Loss: 1.179447889328003:   2%|▏         | 142/6470 [04:22<1:11:26,  1.48it/s]
1|143|Loss: 1.2740366458892822:   2%|▏         | 142/6470 [04:22<1:11:26,  1.48it/s]
1|143|Loss: 1.2740366458892822:   2%|▏         | 143/6470 [04:23<1:12:49,  1.45it/s]
1|144|Loss: 1.5121281147003174:   2%|▏         | 143/6470 [04:23<1:12:49,  1.45it/s]
1|144|Loss: 1.5121281147003174:   2%|▏         | 144/6470 [04:23<1:11:31,  1.47it/s]
1|145|Loss: 1.1694388389587402:   2%|▏         | 144/6470 [04:24<1:11:31,  1.47it/s]
1|145|Loss: 1.1694388389587402:   2%|▏         | 145/6470 [04:24<1:11:57,  1.47it/s]
1|146|Loss: 1.3643230199813843:   2%|▏         | 145/6470 [04:24<1:11:57,  1.47it/s]
1|146|Loss: 1.3643230199813843:   2%|▏         | 146/6470 [04:25<1:13:16,  1.44it/s]
1|147|Loss: 1.3971120119094849:   2%|▏         | 146/6470 [04:25<1:13:16,  1.44it/s]
1|147|Loss: 1.3971120119094849:   2%|▏         | 147/6470 [04:26<1:20:02,  1.32it/s]
1|148|Loss: 1.2944772243499756:   2%|▏         | 147/6470 [04:26<1:20:02,  1.32it/s]
1|148|Loss: 1.2944772243499756:   2%|▏         | 148/6470 [04:26<1:20:44,  1.30it/s]
1|149|Loss: 1.375208854675293:   2%|▏         | 148/6470 [04:27<1:20:44,  1.30it/s] 
1|149|Loss: 1.375208854675293:   2%|▏         | 149/6470 [04:27<1:14:32,  1.41it/s]
1|150|Loss: 1.4542605876922607:   2%|▏         | 149/6470 [04:27<1:14:32,  1.41it/s]
1|150|Loss: 1.4542605876922607:   2%|▏         | 150/6470 [04:28<1:15:42,  1.39it/s]
1|151|Loss: 1.380807876586914:   2%|▏         | 150/6470 [04:28<1:15:42,  1.39it/s] 
1|151|Loss: 1.380807876586914:   2%|▏         | 151/6470 [04:28<1:12:29,  1.45it/s]
1|152|Loss: 1.5891449451446533:   2%|▏         | 151/6470 [04:29<1:12:29,  1.45it/s]
1|152|Loss: 1.5891449451446533:   2%|▏         | 152/6470 [04:29<1:09:51,  1.51it/s]
1|153|Loss: 1.1299705505371094:   2%|▏         | 152/6470 [04:29<1:09:51,  1.51it/s]
1|153|Loss: 1.1299705505371094:   2%|▏         | 153/6470 [04:30<1:09:00,  1.53it/s]
1|154|Loss: 1.2028658390045166:   2%|▏         | 153/6470 [04:30<1:09:00,  1.53it/s]
1|154|Loss: 1.2028658390045166:   2%|▏         | 154/6470 [04:30<1:13:30,  1.43it/s]
1|155|Loss: 1.4365357160568237:   2%|▏         | 154/6470 [04:31<1:13:30,  1.43it/s]
1|155|Loss: 1.4365357160568237:   2%|▏         | 155/6470 [04:31<1:16:57,  1.37it/s]
1|156|Loss: 1.4958025217056274:   2%|▏         | 155/6470 [04:32<1:16:57,  1.37it/s]
1|156|Loss: 1.4958025217056274:   2%|▏         | 156/6470 [04:32<1:14:31,  1.41it/s]
1|157|Loss: 1.6851342916488647:   2%|▏         | 156/6470 [04:32<1:14:31,  1.41it/s]
1|157|Loss: 1.6851342916488647:   2%|▏         | 157/6470 [04:32<1:08:49,  1.53it/s]
1|158|Loss: 1.167479395866394:   2%|▏         | 157/6470 [04:33<1:08:49,  1.53it/s] 
1|158|Loss: 1.167479395866394:   2%|▏         | 158/6470 [04:33<1:10:06,  1.50it/s]
1|159|Loss: 1.2851593494415283:   2%|▏         | 158/6470 [04:33<1:10:06,  1.50it/s]
1|159|Loss: 1.2851593494415283:   2%|▏         | 159/6470 [04:34<1:11:33,  1.47it/s]
1|160|Loss: 1.4679739475250244:   2%|▏         | 159/6470 [04:34<1:11:33,  1.47it/s]
1|160|Loss: 1.4679739475250244:   2%|▏         | 160/6470 [04:34<1:08:58,  1.52it/s]
1|161|Loss: 1.2028859853744507:   2%|▏         | 160/6470 [04:35<1:08:58,  1.52it/s]
1|161|Loss: 1.2028859853744507:   2%|▏         | 161/6470 [04:35<1:07:36,  1.56it/s]
1|162|Loss: 1.2261171340942383:   2%|▏         | 161/6470 [04:35<1:07:36,  1.56it/s]
1|162|Loss: 1.2261171340942383:   3%|▎         | 162/6470 [04:36<1:12:36,  1.45it/s]
1|163|Loss: 1.3658605813980103:   3%|▎         | 162/6470 [04:36<1:12:36,  1.45it/s]
1|163|Loss: 1.3658605813980103:   3%|▎         | 163/6470 [04:37<1:17:09,  1.36it/s]
1|164|Loss: 1.436338186264038:   3%|▎         | 163/6470 [04:37<1:17:09,  1.36it/s] 
1|164|Loss: 1.436338186264038:   3%|▎         | 164/6470 [04:37<1:12:01,  1.46it/s]
1|165|Loss: 1.033538579940796:   3%|▎         | 164/6470 [04:38<1:12:01,  1.46it/s]
1|165|Loss: 1.033538579940796:   3%|▎         | 165/6470 [04:38<1:14:34,  1.41it/s]
1|166|Loss: 1.2447739839553833:   3%|▎         | 165/6470 [04:38<1:14:34,  1.41it/s]
1|166|Loss: 1.2447739839553833:   3%|▎         | 166/6470 [04:39<1:16:13,  1.38it/s]
1|167|Loss: 1.45057213306427:   3%|▎         | 166/6470 [04:39<1:16:13,  1.38it/s]  
1|167|Loss: 1.45057213306427:   3%|▎         | 167/6470 [04:39<1:15:45,  1.39it/s]
1|168|Loss: 1.1512336730957031:   3%|▎         | 167/6470 [04:40<1:15:45,  1.39it/s]
1|168|Loss: 1.1512336730957031:   3%|▎         | 168/6470 [04:40<1:16:26,  1.37it/s]
1|169|Loss: 1.2304646968841553:   3%|▎         | 168/6470 [04:41<1:16:26,  1.37it/s]
1|169|Loss: 1.2304646968841553:   3%|▎         | 169/6470 [04:41<1:19:49,  1.32it/s]
1|170|Loss: 1.172959327697754:   3%|▎         | 169/6470 [04:41<1:19:49,  1.32it/s] 
1|170|Loss: 1.172959327697754:   3%|▎         | 170/6470 [04:42<1:19:57,  1.31it/s]
1|171|Loss: 1.1312774419784546:   3%|▎         | 170/6470 [04:42<1:19:57,  1.31it/s]
1|171|Loss: 1.1312774419784546:   3%|▎         | 171/6470 [04:43<1:21:26,  1.29it/s]
1|172|Loss: 1.4174717664718628:   3%|▎         | 171/6470 [04:43<1:21:26,  1.29it/s]
1|172|Loss: 1.4174717664718628:   3%|▎         | 172/6470 [04:43<1:18:22,  1.34it/s]
1|173|Loss: 1.2712520360946655:   3%|▎         | 172/6470 [04:44<1:18:22,  1.34it/s]
1|173|Loss: 1.2712520360946655:   3%|▎         | 173/6470 [04:44<1:21:08,  1.29it/s]
1|174|Loss: 1.1434900760650635:   3%|▎         | 173/6470 [04:45<1:21:08,  1.29it/s]
1|174|Loss: 1.1434900760650635:   3%|▎         | 174/6470 [04:45<1:20:08,  1.31it/s]
1|175|Loss: 1.3928738832473755:   3%|▎         | 174/6470 [04:45<1:20:08,  1.31it/s]
1|175|Loss: 1.3928738832473755:   3%|▎         | 175/6470 [04:46<1:18:12,  1.34it/s]
1|176|Loss: 1.517192006111145:   3%|▎         | 175/6470 [04:46<1:18:12,  1.34it/s] 
1|176|Loss: 1.517192006111145:   3%|▎         | 176/6470 [04:46<1:15:13,  1.39it/s]
1|177|Loss: 1.2423410415649414:   3%|▎         | 176/6470 [04:47<1:15:13,  1.39it/s]
1|177|Loss: 1.2423410415649414:   3%|▎         | 177/6470 [04:47<1:15:18,  1.39it/s]
1|178|Loss: 1.1844696998596191:   3%|▎         | 177/6470 [04:47<1:15:18,  1.39it/s]
1|178|Loss: 1.1844696998596191:   3%|▎         | 178/6470 [04:48<1:19:22,  1.32it/s]
1|179|Loss: 1.0748305320739746:   3%|▎         | 178/6470 [04:48<1:19:22,  1.32it/s]
1|179|Loss: 1.0748305320739746:   3%|▎         | 179/6470 [04:49<1:24:07,  1.25it/s]
1|180|Loss: 1.1905250549316406:   3%|▎         | 179/6470 [04:49<1:24:07,  1.25it/s]
1|180|Loss: 1.1905250549316406:   3%|▎         | 180/6470 [04:50<1:24:43,  1.24it/s]
1|181|Loss: 1.3341058492660522:   3%|▎         | 180/6470 [04:50<1:24:43,  1.24it/s]
1|181|Loss: 1.3341058492660522:   3%|▎         | 181/6470 [04:50<1:24:31,  1.24it/s]
1|182|Loss: 1.2830432653427124:   3%|▎         | 181/6470 [04:51<1:24:31,  1.24it/s]
1|182|Loss: 1.2830432653427124:   3%|▎         | 182/6470 [04:51<1:20:30,  1.30it/s]
1|183|Loss: 1.070534110069275:   3%|▎         | 182/6470 [04:51<1:20:30,  1.30it/s] 
1|183|Loss: 1.070534110069275:   3%|▎         | 183/6470 [04:52<1:19:10,  1.32it/s]
1|184|Loss: 1.0906908512115479:   3%|▎         | 183/6470 [04:52<1:19:10,  1.32it/s]
1|184|Loss: 1.0906908512115479:   3%|▎         | 184/6470 [04:52<1:20:00,  1.31it/s]
1|185|Loss: 1.1085684299468994:   3%|▎         | 184/6470 [04:53<1:20:00,  1.31it/s]
1|185|Loss: 1.1085684299468994:   3%|▎         | 185/6470 [04:53<1:21:08,  1.29it/s]
1|186|Loss: 1.4465501308441162:   3%|▎         | 185/6470 [04:54<1:21:08,  1.29it/s]
1|186|Loss: 1.4465501308441162:   3%|▎         | 186/6470 [04:54<1:16:27,  1.37it/s]
1|187|Loss: 1.2801481485366821:   3%|▎         | 186/6470 [04:54<1:16:27,  1.37it/s]
1|187|Loss: 1.2801481485366821:   3%|▎         | 187/6470 [04:55<1:15:48,  1.38it/s]
1|188|Loss: 1.3013108968734741:   3%|▎         | 187/6470 [04:55<1:15:48,  1.38it/s]
1|188|Loss: 1.3013108968734741:   3%|▎         | 188/6470 [04:55<1:13:32,  1.42it/s]
1|189|Loss: 1.3577964305877686:   3%|▎         | 188/6470 [04:56<1:13:32,  1.42it/s]
1|189|Loss: 1.3577964305877686:   3%|▎         | 189/6470 [04:56<1:14:03,  1.41it/s]
1|190|Loss: 1.1642266511917114:   3%|▎         | 189/6470 [04:56<1:14:03,  1.41it/s]
1|190|Loss: 1.1642266511917114:   3%|▎         | 190/6470 [04:57<1:16:46,  1.36it/s]
1|191|Loss: 1.2750104665756226:   3%|▎         | 190/6470 [04:57<1:16:46,  1.36it/s]
1|191|Loss: 1.2750104665756226:   3%|▎         | 191/6470 [04:58<1:16:31,  1.37it/s]
1|192|Loss: 1.1127116680145264:   3%|▎         | 191/6470 [04:58<1:16:31,  1.37it/s]
1|192|Loss: 1.1127116680145264:   3%|▎         | 192/6470 [04:58<1:20:07,  1.31it/s]
1|193|Loss: 1.4488301277160645:   3%|▎         | 192/6470 [04:59<1:20:07,  1.31it/s]
1|193|Loss: 1.4488301277160645:   3%|▎         | 193/6470 [04:59<1:17:45,  1.35it/s]
1|194|Loss: 1.2580441236495972:   3%|▎         | 193/6470 [04:59<1:17:45,  1.35it/s]
1|194|Loss: 1.2580441236495972:   3%|▎         | 194/6470 [05:00<1:18:29,  1.33it/s]
1|195|Loss: 1.2827473878860474:   3%|▎         | 194/6470 [05:00<1:18:29,  1.33it/s]
1|195|Loss: 1.2827473878860474:   3%|▎         | 195/6470 [05:00<1:14:22,  1.41it/s]
1|196|Loss: 1.177286148071289:   3%|▎         | 195/6470 [05:01<1:14:22,  1.41it/s] 
1|196|Loss: 1.177286148071289:   3%|▎         | 196/6470 [05:01<1:14:17,  1.41it/s]
1|197|Loss: 1.2547335624694824:   3%|▎         | 196/6470 [05:02<1:14:17,  1.41it/s]
1|197|Loss: 1.2547335624694824:   3%|▎         | 197/6470 [05:02<1:17:28,  1.35it/s]
1|198|Loss: 1.6762093305587769:   3%|▎         | 197/6470 [05:02<1:17:28,  1.35it/s]
1|198|Loss: 1.6762093305587769:   3%|▎         | 198/6470 [05:03<1:18:15,  1.34it/s]
1|199|Loss: 1.355857253074646:   3%|▎         | 198/6470 [05:03<1:18:15,  1.34it/s] 
1|199|Loss: 1.355857253074646:   3%|▎         | 199/6470 [05:03<1:15:34,  1.38it/s]
1|200|Loss: 1.3695404529571533:   3%|▎         | 199/6470 [05:04<1:15:34,  1.38it/s]
1|200|Loss: 1.3695404529571533:   3%|▎         | 200/6470 [05:04<1:12:47,  1.44it/s]
1|201|Loss: 1.4407002925872803:   3%|▎         | 200/6470 [05:04<1:12:47,  1.44it/s]
1|201|Loss: 1.4407002925872803:   3%|▎         | 201/6470 [05:05<1:10:15,  1.49it/s]
1|202|Loss: 1.3362013101577759:   3%|▎         | 201/6470 [05:05<1:10:15,  1.49it/s]
1|202|Loss: 1.3362013101577759:   3%|▎         | 202/6470 [05:05<1:11:43,  1.46it/s]
1|203|Loss: 1.5756019353866577:   3%|▎         | 202/6470 [05:06<1:11:43,  1.46it/s]
1|203|Loss: 1.5756019353866577:   3%|▎         | 203/6470 [05:06<1:09:20,  1.51it/s]
1|204|Loss: 1.3988173007965088:   3%|▎         | 203/6470 [05:06<1:09:20,  1.51it/s]
1|204|Loss: 1.3988173007965088:   3%|▎         | 204/6470 [05:07<1:06:54,  1.56it/s]
1|205|Loss: 1.4267522096633911:   3%|▎         | 204/6470 [05:07<1:06:54,  1.56it/s]
1|205|Loss: 1.4267522096633911:   3%|▎         | 205/6470 [05:07<1:05:55,  1.58it/s]
1|206|Loss: 1.6430060863494873:   3%|▎         | 205/6470 [05:07<1:05:55,  1.58it/s]
1|206|Loss: 1.6430060863494873:   3%|▎         | 206/6470 [05:08<1:01:15,  1.70it/s]
1|207|Loss: 1.2847709655761719:   3%|▎         | 206/6470 [05:08<1:01:15,  1.70it/s]
1|207|Loss: 1.2847709655761719:   3%|▎         | 207/6470 [05:08<1:05:11,  1.60it/s]
1|208|Loss: 1.1946009397506714:   3%|▎         | 207/6470 [05:09<1:05:11,  1.60it/s]
1|208|Loss: 1.1946009397506714:   3%|▎         | 208/6470 [05:09<1:08:35,  1.52it/s]
1|209|Loss: 1.222614049911499:   3%|▎         | 208/6470 [05:09<1:08:35,  1.52it/s] 
1|209|Loss: 1.222614049911499:   3%|▎         | 209/6470 [05:10<1:08:21,  1.53it/s]
1|210|Loss: 1.25221848487854:   3%|▎         | 209/6470 [05:10<1:08:21,  1.53it/s] 
1|210|Loss: 1.25221848487854:   3%|▎         | 210/6470 [05:11<1:11:46,  1.45it/s]
1|211|Loss: 1.3751674890518188:   3%|▎         | 210/6470 [05:11<1:11:46,  1.45it/s]
1|211|Loss: 1.3751674890518188:   3%|▎         | 211/6470 [05:11<1:13:30,  1.42it/s]
1|212|Loss: 1.3351000547409058:   3%|▎         | 211/6470 [05:12<1:13:30,  1.42it/s]
1|212|Loss: 1.3351000547409058:   3%|▎         | 212/6470 [05:12<1:14:51,  1.39it/s]
1|213|Loss: 1.1749532222747803:   3%|▎         | 212/6470 [05:12<1:14:51,  1.39it/s]
1|213|Loss: 1.1749532222747803:   3%|▎         | 213/6470 [05:13<1:17:30,  1.35it/s]
1|214|Loss: 1.1924127340316772:   3%|▎         | 213/6470 [05:13<1:17:30,  1.35it/s]
1|214|Loss: 1.1924127340316772:   3%|▎         | 214/6470 [05:14<1:17:43,  1.34it/s]
1|215|Loss: 1.2313134670257568:   3%|▎         | 214/6470 [05:14<1:17:43,  1.34it/s]
1|215|Loss: 1.2313134670257568:   3%|▎         | 215/6470 [05:14<1:21:06,  1.29it/s]
1|216|Loss: 1.2051020860671997:   3%|▎         | 215/6470 [05:15<1:21:06,  1.29it/s]
1|216|Loss: 1.2051020860671997:   3%|▎         | 216/6470 [05:15<1:19:59,  1.30it/s]
1|217|Loss: 1.2506113052368164:   3%|▎         | 216/6470 [05:16<1:19:59,  1.30it/s]
1|217|Loss: 1.2506113052368164:   3%|▎         | 217/6470 [05:16<1:20:00,  1.30it/s]
1|218|Loss: 1.4374511241912842:   3%|▎         | 217/6470 [05:16<1:20:00,  1.30it/s]
1|218|Loss: 1.4374511241912842:   3%|▎         | 218/6470 [05:17<1:21:38,  1.28it/s]
1|219|Loss: 0.9240778684616089:   3%|▎         | 218/6470 [05:17<1:21:38,  1.28it/s]
1|219|Loss: 0.9240778684616089:   3%|▎         | 219/6470 [05:18<1:24:14,  1.24it/s]
1|220|Loss: 1.2402100563049316:   3%|▎         | 219/6470 [05:18<1:24:14,  1.24it/s]
1|220|Loss: 1.2402100563049316:   3%|▎         | 220/6470 [05:18<1:25:59,  1.21it/s]
1|221|Loss: 1.360840916633606:   3%|▎         | 220/6470 [05:19<1:25:59,  1.21it/s] 
1|221|Loss: 1.360840916633606:   3%|▎         | 221/6470 [05:19<1:20:24,  1.30it/s]
1|222|Loss: 1.3486422300338745:   3%|▎         | 221/6470 [05:19<1:20:24,  1.30it/s]
1|222|Loss: 1.3486422300338745:   3%|▎         | 222/6470 [05:20<1:15:08,  1.39it/s]
1|223|Loss: 1.301328420639038:   3%|▎         | 222/6470 [05:20<1:15:08,  1.39it/s] 
1|223|Loss: 1.301328420639038:   3%|▎         | 223/6470 [05:20<1:14:01,  1.41it/s]
1|224|Loss: 1.0772467851638794:   3%|▎         | 223/6470 [05:21<1:14:01,  1.41it/s]
1|224|Loss: 1.0772467851638794:   3%|▎         | 224/6470 [05:21<1:15:04,  1.39it/s]
1|225|Loss: 1.236762285232544:   3%|▎         | 224/6470 [05:22<1:15:04,  1.39it/s] 
1|225|Loss: 1.236762285232544:   3%|▎         | 225/6470 [05:22<1:16:09,  1.37it/s]
1|226|Loss: 1.3466229438781738:   3%|▎         | 225/6470 [05:22<1:16:09,  1.37it/s]
1|226|Loss: 1.3466229438781738:   3%|▎         | 226/6470 [05:23<1:16:22,  1.36it/s]
1|227|Loss: 1.316007137298584:   3%|▎         | 226/6470 [05:23<1:16:22,  1.36it/s] 
1|227|Loss: 1.316007137298584:   4%|▎         | 227/6470 [05:23<1:16:16,  1.36it/s]
1|228|Loss: 1.331756830215454:   4%|▎         | 227/6470 [05:24<1:16:16,  1.36it/s]
1|228|Loss: 1.331756830215454:   4%|▎         | 228/6470 [05:24<1:15:24,  1.38it/s]
1|229|Loss: 1.3552696704864502:   4%|▎         | 228/6470 [05:24<1:15:24,  1.38it/s]
1|229|Loss: 1.3552696704864502:   4%|▎         | 229/6470 [05:25<1:11:18,  1.46it/s]
1|230|Loss: 1.2483043670654297:   4%|▎         | 229/6470 [05:25<1:11:18,  1.46it/s]
1|230|Loss: 1.2483043670654297:   4%|▎         | 230/6470 [05:25<1:09:25,  1.50it/s]
1|231|Loss: 1.1527363061904907:   4%|▎         | 230/6470 [05:26<1:09:25,  1.50it/s]
1|231|Loss: 1.1527363061904907:   4%|▎         | 231/6470 [05:26<1:11:21,  1.46it/s]
1|232|Loss: 1.366928219795227:   4%|▎         | 231/6470 [05:26<1:11:21,  1.46it/s] 
1|232|Loss: 1.366928219795227:   4%|▎         | 232/6470 [05:27<1:08:57,  1.51it/s]
1|233|Loss: 1.2979382276535034:   4%|▎         | 232/6470 [05:27<1:08:57,  1.51it/s]
1|233|Loss: 1.2979382276535034:   4%|▎         | 233/6470 [05:27<1:06:22,  1.57it/s]
1|234|Loss: 1.334466576576233:   4%|▎         | 233/6470 [05:28<1:06:22,  1.57it/s] 
1|234|Loss: 1.334466576576233:   4%|▎         | 234/6470 [05:28<1:04:33,  1.61it/s]
1|235|Loss: 1.3990062475204468:   4%|▎         | 234/6470 [05:28<1:04:33,  1.61it/s]
1|235|Loss: 1.3990062475204468:   4%|▎         | 235/6470 [05:29<1:07:06,  1.55it/s]
1|236|Loss: 1.1887798309326172:   4%|▎         | 235/6470 [05:29<1:07:06,  1.55it/s]
1|236|Loss: 1.1887798309326172:   4%|▎         | 236/6470 [05:29<1:09:45,  1.49it/s]
1|237|Loss: 1.0802572965621948:   4%|▎         | 236/6470 [05:30<1:09:45,  1.49it/s]
1|237|Loss: 1.0802572965621948:   4%|▎         | 237/6470 [05:30<1:11:00,  1.46it/s]
1|238|Loss: 1.343638300895691:   4%|▎         | 237/6470 [05:30<1:11:00,  1.46it/s] 
1|238|Loss: 1.343638300895691:   4%|▎         | 238/6470 [05:31<1:09:33,  1.49it/s]
1|239|Loss: 1.260638952255249:   4%|▎         | 238/6470 [05:31<1:09:33,  1.49it/s]
1|239|Loss: 1.260638952255249:   4%|▎         | 239/6470 [05:31<1:05:16,  1.59it/s]
1|240|Loss: 1.2829277515411377:   4%|▎         | 239/6470 [05:31<1:05:16,  1.59it/s]
1|240|Loss: 1.2829277515411377:   4%|▎         | 240/6470 [05:32<1:06:57,  1.55it/s]
1|241|Loss: 1.1320308446884155:   4%|▎         | 240/6470 [05:32<1:06:57,  1.55it/s]
1|241|Loss: 1.1320308446884155:   4%|▎         | 241/6470 [05:33<1:09:35,  1.49it/s]
1|242|Loss: 1.1754003763198853:   4%|▎         | 241/6470 [05:33<1:09:35,  1.49it/s]
1|242|Loss: 1.1754003763198853:   4%|▎         | 242/6470 [05:33<1:13:55,  1.40it/s]
1|243|Loss: 1.275800347328186:   4%|▎         | 242/6470 [05:34<1:13:55,  1.40it/s] 
1|243|Loss: 1.275800347328186:   4%|▍         | 243/6470 [05:34<1:14:32,  1.39it/s]
1|244|Loss: 1.4527626037597656:   4%|▍         | 243/6470 [05:34<1:14:32,  1.39it/s]
1|244|Loss: 1.4527626037597656:   4%|▍         | 244/6470 [05:35<1:14:43,  1.39it/s]
1|245|Loss: 1.2754632234573364:   4%|▍         | 244/6470 [05:35<1:14:43,  1.39it/s]
1|245|Loss: 1.2754632234573364:   4%|▍         | 245/6470 [05:36<1:14:18,  1.40it/s]
1|246|Loss: 1.2195122241973877:   4%|▍         | 245/6470 [05:36<1:14:18,  1.40it/s]
1|246|Loss: 1.2195122241973877:   4%|▍         | 246/6470 [05:36<1:14:09,  1.40it/s]
1|247|Loss: 1.3502459526062012:   4%|▍         | 246/6470 [05:37<1:14:09,  1.40it/s]
1|247|Loss: 1.3502459526062012:   4%|▍         | 247/6470 [05:37<1:10:01,  1.48it/s]
1|248|Loss: 1.2765721082687378:   4%|▍         | 247/6470 [05:37<1:10:01,  1.48it/s]
1|248|Loss: 1.2765721082687378:   4%|▍         | 248/6470 [05:38<1:11:08,  1.46it/s]
1|249|Loss: 1.2928112745285034:   4%|▍         | 248/6470 [05:38<1:11:08,  1.46it/s]
1|249|Loss: 1.2928112745285034:   4%|▍         | 249/6470 [05:38<1:17:08,  1.34it/s]
1|250|Loss: 1.3234200477600098:   4%|▍         | 249/6470 [05:39<1:17:08,  1.34it/s]
1|250|Loss: 1.3234200477600098:   4%|▍         | 250/6470 [05:39<1:13:30,  1.41it/s]
1|251|Loss: 1.7209373712539673:   4%|▍         | 250/6470 [05:39<1:13:30,  1.41it/s]
1|251|Loss: 1.7209373712539673:   4%|▍         | 251/6470 [05:40<1:07:34,  1.53it/s]
1|252|Loss: 1.1700419187545776:   4%|▍         | 251/6470 [05:40<1:07:34,  1.53it/s]
1|252|Loss: 1.1700419187545776:   4%|▍         | 252/6470 [05:40<1:13:16,  1.41it/s]
1|253|Loss: 1.2771199941635132:   4%|▍         | 252/6470 [05:41<1:13:16,  1.41it/s]
1|253|Loss: 1.2771199941635132:   4%|▍         | 253/6470 [05:41<1:12:56,  1.42it/s]
1|254|Loss: 1.368299126625061:   4%|▍         | 253/6470 [05:41<1:12:56,  1.42it/s] 
1|254|Loss: 1.368299126625061:   4%|▍         | 254/6470 [05:42<1:07:47,  1.53it/s]
1|255|Loss: 1.461413025856018:   4%|▍         | 254/6470 [05:42<1:07:47,  1.53it/s]
1|255|Loss: 1.461413025856018:   4%|▍         | 255/6470 [05:42<1:06:36,  1.56it/s]
1|256|Loss: 1.3709592819213867:   4%|▍         | 255/6470 [05:43<1:06:36,  1.56it/s]
1|256|Loss: 1.3709592819213867:   4%|▍         | 256/6470 [05:43<1:06:35,  1.56it/s]
1|257|Loss: 1.2692316770553589:   4%|▍         | 256/6470 [05:43<1:06:35,  1.56it/s]
1|257|Loss: 1.2692316770553589:   4%|▍         | 257/6470 [05:44<1:06:32,  1.56it/s]
1|258|Loss: 1.0986154079437256:   4%|▍         | 257/6470 [05:44<1:06:32,  1.56it/s]
1|258|Loss: 1.0986154079437256:   4%|▍         | 258/6470 [05:44<1:08:51,  1.50it/s]
1|259|Loss: 1.3830231428146362:   4%|▍         | 258/6470 [05:45<1:08:51,  1.50it/s]
1|259|Loss: 1.3830231428146362:   4%|▍         | 259/6470 [05:45<1:10:53,  1.46it/s]
1|260|Loss: 1.1333473920822144:   4%|▍         | 259/6470 [05:45<1:10:53,  1.46it/s]
1|260|Loss: 1.1333473920822144:   4%|▍         | 260/6470 [05:46<1:09:29,  1.49it/s]
1|261|Loss: 1.2531956434249878:   4%|▍         | 260/6470 [05:46<1:09:29,  1.49it/s]
1|261|Loss: 1.2531956434249878:   4%|▍         | 261/6470 [05:46<1:13:56,  1.40it/s]
1|262|Loss: 1.3335753679275513:   4%|▍         | 261/6470 [05:47<1:13:56,  1.40it/s]
1|262|Loss: 1.3335753679275513:   4%|▍         | 262/6470 [05:47<1:12:50,  1.42it/s]
1|263|Loss: 1.5151257514953613:   4%|▍         | 262/6470 [05:47<1:12:50,  1.42it/s]
1|263|Loss: 1.5151257514953613:   4%|▍         | 263/6470 [05:48<1:10:01,  1.48it/s]
1|264|Loss: 1.2880055904388428:   4%|▍         | 263/6470 [05:48<1:10:01,  1.48it/s]
1|264|Loss: 1.2880055904388428:   4%|▍         | 264/6470 [05:48<1:05:38,  1.58it/s]
1|265|Loss: 1.5309641361236572:   4%|▍         | 264/6470 [05:49<1:05:38,  1.58it/s]
1|265|Loss: 1.5309641361236572:   4%|▍         | 265/6470 [05:49<1:04:25,  1.61it/s]
1|266|Loss: 1.1315760612487793:   4%|▍         | 265/6470 [05:49<1:04:25,  1.61it/s]
1|266|Loss: 1.1315760612487793:   4%|▍         | 266/6470 [05:49<1:03:35,  1.63it/s]
1|267|Loss: 1.6537033319473267:   4%|▍         | 266/6470 [05:50<1:03:35,  1.63it/s]
1|267|Loss: 1.6537033319473267:   4%|▍         | 267/6470 [05:50<57:46,  1.79it/s]  
1|268|Loss: 1.2103497982025146:   4%|▍         | 267/6470 [05:50<57:46,  1.79it/s]
1|268|Loss: 1.2103497982025146:   4%|▍         | 268/6470 [05:50<59:01,  1.75it/s]
1|269|Loss: 1.1896796226501465:   4%|▍         | 268/6470 [05:51<59:01,  1.75it/s]
1|269|Loss: 1.1896796226501465:   4%|▍         | 269/6470 [05:51<1:06:36,  1.55it/s]
1|270|Loss: 1.3327720165252686:   4%|▍         | 269/6470 [05:52<1:06:36,  1.55it/s]
1|270|Loss: 1.3327720165252686:   4%|▍         | 270/6470 [05:52<1:09:15,  1.49it/s]
1|271|Loss: 1.2321929931640625:   4%|▍         | 270/6470 [05:52<1:09:15,  1.49it/s]
1|271|Loss: 1.2321929931640625:   4%|▍         | 271/6470 [05:53<1:07:01,  1.54it/s]
1|272|Loss: 1.5345379114151:   4%|▍         | 271/6470 [05:53<1:07:01,  1.54it/s]   
1|272|Loss: 1.5345379114151:   4%|▍         | 272/6470 [05:53<1:05:08,  1.59it/s]
1|273|Loss: 1.2574695348739624:   4%|▍         | 272/6470 [05:54<1:05:08,  1.59it/s]
1|273|Loss: 1.2574695348739624:   4%|▍         | 273/6470 [05:54<1:11:54,  1.44it/s]
1|274|Loss: 1.4051153659820557:   4%|▍         | 273/6470 [05:55<1:11:54,  1.44it/s]
1|274|Loss: 1.4051153659820557:   4%|▍         | 274/6470 [05:55<1:14:04,  1.39it/s]
1|275|Loss: 1.4258472919464111:   4%|▍         | 274/6470 [05:55<1:14:04,  1.39it/s]
1|275|Loss: 1.4258472919464111:   4%|▍         | 275/6470 [05:55<1:11:47,  1.44it/s]
1|276|Loss: 1.2913427352905273:   4%|▍         | 275/6470 [05:56<1:11:47,  1.44it/s]
1|276|Loss: 1.2913427352905273:   4%|▍         | 276/6470 [05:56<1:08:13,  1.51it/s]
1|277|Loss: 1.3828582763671875:   4%|▍         | 276/6470 [05:56<1:08:13,  1.51it/s]
1|277|Loss: 1.3828582763671875:   4%|▍         | 277/6470 [05:57<1:03:29,  1.63it/s]
1|278|Loss: 1.2207684516906738:   4%|▍         | 277/6470 [05:57<1:03:29,  1.63it/s]
1|278|Loss: 1.2207684516906738:   4%|▍         | 278/6470 [05:57<1:04:56,  1.59it/s]
1|279|Loss: 1.1913237571716309:   4%|▍         | 278/6470 [05:58<1:04:56,  1.59it/s]
1|279|Loss: 1.1913237571716309:   4%|▍         | 279/6470 [05:58<1:09:56,  1.48it/s]
1|280|Loss: 1.2058327198028564:   4%|▍         | 279/6470 [05:58<1:09:56,  1.48it/s]
1|280|Loss: 1.2058327198028564:   4%|▍         | 280/6470 [05:59<1:12:02,  1.43it/s]
1|281|Loss: 1.405735969543457:   4%|▍         | 280/6470 [05:59<1:12:02,  1.43it/s] 
1|281|Loss: 1.405735969543457:   4%|▍         | 281/6470 [06:00<1:16:04,  1.36it/s]
1|282|Loss: 0.9377135038375854:   4%|▍         | 281/6470 [06:00<1:16:04,  1.36it/s]
1|282|Loss: 0.9377135038375854:   4%|▍         | 282/6470 [06:00<1:20:34,  1.28it/s]
1|283|Loss: 1.4367562532424927:   4%|▍         | 282/6470 [06:01<1:20:34,  1.28it/s]
1|283|Loss: 1.4367562532424927:   4%|▍         | 283/6470 [06:01<1:16:24,  1.35it/s]
1|284|Loss: 1.4348706007003784:   4%|▍         | 283/6470 [06:02<1:16:24,  1.35it/s]
1|284|Loss: 1.4348706007003784:   4%|▍         | 284/6470 [06:02<1:17:30,  1.33it/s]
1|285|Loss: 1.1190663576126099:   4%|▍         | 284/6470 [06:02<1:17:30,  1.33it/s]
1|285|Loss: 1.1190663576126099:   4%|▍         | 285/6470 [06:03<1:19:57,  1.29it/s]
1|286|Loss: 1.3794854879379272:   4%|▍         | 285/6470 [06:03<1:19:57,  1.29it/s]
1|286|Loss: 1.3794854879379272:   4%|▍         | 286/6470 [06:03<1:17:09,  1.34it/s]
1|287|Loss: 1.2587838172912598:   4%|▍         | 286/6470 [06:04<1:17:09,  1.34it/s]
1|287|Loss: 1.2587838172912598:   4%|▍         | 287/6470 [06:04<1:15:15,  1.37it/s]
1|288|Loss: 1.2372562885284424:   4%|▍         | 287/6470 [06:05<1:15:15,  1.37it/s]
1|288|Loss: 1.2372562885284424:   4%|▍         | 288/6470 [06:05<1:15:07,  1.37it/s]
1|289|Loss: 1.0736284255981445:   4%|▍         | 288/6470 [06:05<1:15:07,  1.37it/s]
1|289|Loss: 1.0736284255981445:   4%|▍         | 289/6470 [06:06<1:15:28,  1.37it/s]
1|290|Loss: 1.276488184928894:   4%|▍         | 289/6470 [06:06<1:15:28,  1.37it/s] 
1|290|Loss: 1.276488184928894:   4%|▍         | 290/6470 [06:06<1:12:51,  1.41it/s]
1|291|Loss: 1.3055402040481567:   4%|▍         | 290/6470 [06:07<1:12:51,  1.41it/s]
1|291|Loss: 1.3055402040481567:   4%|▍         | 291/6470 [06:07<1:08:21,  1.51it/s]
1|292|Loss: 1.5214674472808838:   4%|▍         | 291/6470 [06:07<1:08:21,  1.51it/s]
1|292|Loss: 1.5214674472808838:   5%|▍         | 292/6470 [06:07<1:04:57,  1.59it/s]
1|293|Loss: 1.610094428062439:   5%|▍         | 292/6470 [06:08<1:04:57,  1.59it/s] 
1|293|Loss: 1.610094428062439:   5%|▍         | 293/6470 [06:08<1:00:02,  1.71it/s]
1|294|Loss: 1.4485501050949097:   5%|▍         | 293/6470 [06:08<1:00:02,  1.71it/s]
1|294|Loss: 1.4485501050949097:   5%|▍         | 294/6470 [06:08<59:47,  1.72it/s]  
1|295|Loss: 1.289219856262207:   5%|▍         | 294/6470 [06:09<59:47,  1.72it/s] 
1|295|Loss: 1.289219856262207:   5%|▍         | 295/6470 [06:09<1:00:53,  1.69it/s]
1|296|Loss: 1.4210039377212524:   5%|▍         | 295/6470 [06:09<1:00:53,  1.69it/s]
1|296|Loss: 1.4210039377212524:   5%|▍         | 296/6470 [06:09<57:49,  1.78it/s]  
1|297|Loss: 1.3701624870300293:   5%|▍         | 296/6470 [06:10<57:49,  1.78it/s]
1|297|Loss: 1.3701624870300293:   5%|▍         | 297/6470 [06:10<54:18,  1.89it/s]
1|298|Loss: 1.1961414813995361:   5%|▍         | 297/6470 [06:10<54:18,  1.89it/s]
1|298|Loss: 1.1961414813995361:   5%|▍         | 298/6470 [06:11<56:36,  1.82it/s]
1|299|Loss: 1.226709246635437:   5%|▍         | 298/6470 [06:11<56:36,  1.82it/s] 
1|299|Loss: 1.226709246635437:   5%|▍         | 299/6470 [06:11<1:05:09,  1.58it/s]
1|300|Loss: 1.0040615797042847:   5%|▍         | 299/6470 [06:12<1:05:09,  1.58it/s]
1|300|Loss: 1.0040615797042847:   5%|▍         | 300/6470 [06:12<1:13:39,  1.40it/s]
1|301|Loss: 1.1114482879638672:   5%|▍         | 300/6470 [06:13<1:13:39,  1.40it/s]
1|301|Loss: 1.1114482879638672:   5%|▍         | 301/6470 [06:13<1:18:30,  1.31it/s]
1|302|Loss: 1.2942475080490112:   5%|▍         | 301/6470 [06:14<1:18:30,  1.31it/s]
1|302|Loss: 1.2942475080490112:   5%|▍         | 302/6470 [06:14<1:15:55,  1.35it/s]
1|303|Loss: 0.9825357794761658:   5%|▍         | 302/6470 [06:14<1:15:55,  1.35it/s]
1|303|Loss: 0.9825357794761658:   5%|▍         | 303/6470 [06:15<1:20:33,  1.28it/s]
1|304|Loss: 1.0166223049163818:   5%|▍         | 303/6470 [06:15<1:20:33,  1.28it/s]
1|304|Loss: 1.0166223049163818:   5%|▍         | 304/6470 [06:16<1:25:02,  1.21it/s]
1|305|Loss: 1.1658656597137451:   5%|▍         | 304/6470 [06:16<1:25:02,  1.21it/s]
1|305|Loss: 1.1658656597137451:   5%|▍         | 305/6470 [06:16<1:22:31,  1.24it/s]
1|306|Loss: 1.2069302797317505:   5%|▍         | 305/6470 [06:17<1:22:31,  1.24it/s]
1|306|Loss: 1.2069302797317505:   5%|▍         | 306/6470 [06:17<1:19:17,  1.30it/s]
1|307|Loss: 1.198851227760315:   5%|▍         | 306/6470 [06:18<1:19:17,  1.30it/s] 
1|307|Loss: 1.198851227760315:   5%|▍         | 307/6470 [06:18<1:18:00,  1.32it/s]
1|308|Loss: 1.3331738710403442:   5%|▍         | 307/6470 [06:18<1:18:00,  1.32it/s]
1|308|Loss: 1.3331738710403442:   5%|▍         | 308/6470 [06:18<1:12:08,  1.42it/s]
1|309|Loss: 1.022324562072754:   5%|▍         | 308/6470 [06:19<1:12:08,  1.42it/s] 
1|309|Loss: 1.022324562072754:   5%|▍         | 309/6470 [06:19<1:13:15,  1.40it/s]
1|310|Loss: 1.1991866827011108:   5%|▍         | 309/6470 [06:20<1:13:15,  1.40it/s]
1|310|Loss: 1.1991866827011108:   5%|▍         | 310/6470 [06:20<1:13:26,  1.40it/s]
1|311|Loss: 1.2213404178619385:   5%|▍         | 310/6470 [06:20<1:13:26,  1.40it/s]
1|311|Loss: 1.2213404178619385:   5%|▍         | 311/6470 [06:21<1:11:32,  1.43it/s]
1|312|Loss: 1.4135411977767944:   5%|▍         | 311/6470 [06:21<1:11:32,  1.43it/s]
1|312|Loss: 1.4135411977767944:   5%|▍         | 312/6470 [06:21<1:12:14,  1.42it/s]
1|313|Loss: 1.1301888227462769:   5%|▍         | 312/6470 [06:22<1:12:14,  1.42it/s]
1|313|Loss: 1.1301888227462769:   5%|▍         | 313/6470 [06:22<1:12:29,  1.42it/s]
1|314|Loss: 1.2536240816116333:   5%|▍         | 313/6470 [06:22<1:12:29,  1.42it/s]
1|314|Loss: 1.2536240816116333:   5%|▍         | 314/6470 [06:23<1:15:29,  1.36it/s]
1|315|Loss: 1.1891279220581055:   5%|▍         | 314/6470 [06:23<1:15:29,  1.36it/s]
1|315|Loss: 1.1891279220581055:   5%|▍         | 315/6470 [06:24<1:16:08,  1.35it/s]
1|316|Loss: 1.2879643440246582:   5%|▍         | 315/6470 [06:24<1:16:08,  1.35it/s]
1|316|Loss: 1.2879643440246582:   5%|▍         | 316/6470 [06:24<1:12:10,  1.42it/s]
1|317|Loss: 1.341320276260376:   5%|▍         | 316/6470 [06:24<1:12:10,  1.42it/s] 
1|317|Loss: 1.341320276260376:   5%|▍         | 317/6470 [06:25<1:08:06,  1.51it/s]
1|318|Loss: 1.1247832775115967:   5%|▍         | 317/6470 [06:25<1:08:06,  1.51it/s]
1|318|Loss: 1.1247832775115967:   5%|▍         | 318/6470 [06:25<1:11:16,  1.44it/s]
1|319|Loss: 1.152326226234436:   5%|▍         | 318/6470 [06:26<1:11:16,  1.44it/s] 
1|319|Loss: 1.152326226234436:   5%|▍         | 319/6470 [06:26<1:09:26,  1.48it/s]
1|320|Loss: 1.375433325767517:   5%|▍         | 319/6470 [06:26<1:09:26,  1.48it/s]
1|320|Loss: 1.375433325767517:   5%|▍         | 320/6470 [06:27<1:05:13,  1.57it/s]
1|321|Loss: 1.3029160499572754:   5%|▍         | 320/6470 [06:27<1:05:13,  1.57it/s]
1|321|Loss: 1.3029160499572754:   5%|▍         | 321/6470 [06:27<1:03:58,  1.60it/s]
1|322|Loss: 1.1885597705841064:   5%|▍         | 321/6470 [06:28<1:03:58,  1.60it/s]
1|322|Loss: 1.1885597705841064:   5%|▍         | 322/6470 [06:28<1:09:56,  1.47it/s]
1|323|Loss: 1.2683749198913574:   5%|▍         | 322/6470 [06:28<1:09:56,  1.47it/s]
1|323|Loss: 1.2683749198913574:   5%|▍         | 323/6470 [06:29<1:12:37,  1.41it/s]
1|324|Loss: 1.4690191745758057:   5%|▍         | 323/6470 [06:29<1:12:37,  1.41it/s]
1|324|Loss: 1.4690191745758057:   5%|▌         | 324/6470 [06:29<1:04:16,  1.59it/s]
1|325|Loss: 1.4062166213989258:   5%|▌         | 324/6470 [06:30<1:04:16,  1.59it/s]
1|325|Loss: 1.4062166213989258:   5%|▌         | 325/6470 [06:30<1:06:46,  1.53it/s]
1|326|Loss: 1.145735263824463:   5%|▌         | 325/6470 [06:30<1:06:46,  1.53it/s] 
1|326|Loss: 1.145735263824463:   5%|▌         | 326/6470 [06:31<1:07:34,  1.52it/s]
1|327|Loss: 1.0849088430404663:   5%|▌         | 326/6470 [06:31<1:07:34,  1.52it/s]
1|327|Loss: 1.0849088430404663:   5%|▌         | 327/6470 [06:31<1:10:47,  1.45it/s]
1|328|Loss: 1.5089105367660522:   5%|▌         | 327/6470 [06:32<1:10:47,  1.45it/s]
1|328|Loss: 1.5089105367660522:   5%|▌         | 328/6470 [06:32<1:09:35,  1.47it/s]
1|329|Loss: 1.0213963985443115:   5%|▌         | 328/6470 [06:33<1:09:35,  1.47it/s]
1|329|Loss: 1.0213963985443115:   5%|▌         | 329/6470 [06:33<1:17:06,  1.33it/s]
1|330|Loss: 1.2407810688018799:   5%|▌         | 329/6470 [06:33<1:17:06,  1.33it/s]
1|330|Loss: 1.2407810688018799:   5%|▌         | 330/6470 [06:34<1:19:54,  1.28it/s]
1|331|Loss: 1.3872628211975098:   5%|▌         | 330/6470 [06:34<1:19:54,  1.28it/s]
1|331|Loss: 1.3872628211975098:   5%|▌         | 331/6470 [06:34<1:15:05,  1.36it/s]
1|332|Loss: 1.4676109552383423:   5%|▌         | 331/6470 [06:35<1:15:05,  1.36it/s]
1|332|Loss: 1.4676109552383423:   5%|▌         | 332/6470 [06:35<1:14:19,  1.38it/s]
1|333|Loss: 1.1926707029342651:   5%|▌         | 332/6470 [06:36<1:14:19,  1.38it/s]
1|333|Loss: 1.1926707029342651:   5%|▌         | 333/6470 [06:36<1:11:57,  1.42it/s]
1|334|Loss: 1.3099162578582764:   5%|▌         | 333/6470 [06:36<1:11:57,  1.42it/s]
1|334|Loss: 1.3099162578582764:   5%|▌         | 334/6470 [06:36<1:09:55,  1.46it/s]
1|335|Loss: 1.2313638925552368:   5%|▌         | 334/6470 [06:37<1:09:55,  1.46it/s]
1|335|Loss: 1.2313638925552368:   5%|▌         | 335/6470 [06:37<1:11:10,  1.44it/s]
1|336|Loss: 1.3509770631790161:   5%|▌         | 335/6470 [06:38<1:11:10,  1.44it/s]
1|336|Loss: 1.3509770631790161:   5%|▌         | 336/6470 [06:38<1:14:47,  1.37it/s]
1|337|Loss: 1.1787068843841553:   5%|▌         | 336/6470 [06:38<1:14:47,  1.37it/s]
1|337|Loss: 1.1787068843841553:   5%|▌         | 337/6470 [06:39<1:17:48,  1.31it/s]
1|338|Loss: 1.1764075756072998:   5%|▌         | 337/6470 [06:39<1:17:48,  1.31it/s]
1|338|Loss: 1.1764075756072998:   5%|▌         | 338/6470 [06:40<1:19:16,  1.29it/s]
1|339|Loss: 1.1819838285446167:   5%|▌         | 338/6470 [06:40<1:19:16,  1.29it/s]
1|339|Loss: 1.1819838285446167:   5%|▌         | 339/6470 [06:40<1:13:54,  1.38it/s]
1|340|Loss: 1.60573410987854:   5%|▌         | 339/6470 [06:41<1:13:54,  1.38it/s]  
1|340|Loss: 1.60573410987854:   5%|▌         | 340/6470 [06:41<1:08:02,  1.50it/s]
1|341|Loss: 1.3671678304672241:   5%|▌         | 340/6470 [06:41<1:08:02,  1.50it/s]
1|341|Loss: 1.3671678304672241:   5%|▌         | 341/6470 [06:41<1:03:14,  1.62it/s]
1|342|Loss: 1.2638461589813232:   5%|▌         | 341/6470 [06:42<1:03:14,  1.62it/s]
1|342|Loss: 1.2638461589813232:   5%|▌         | 342/6470 [06:42<1:01:21,  1.66it/s]
1|343|Loss: 1.1971787214279175:   5%|▌         | 342/6470 [06:42<1:01:21,  1.66it/s]
1|343|Loss: 1.1971787214279175:   5%|▌         | 343/6470 [06:43<1:07:15,  1.52it/s]
1|344|Loss: 1.3027539253234863:   5%|▌         | 343/6470 [06:43<1:07:15,  1.52it/s]
1|344|Loss: 1.3027539253234863:   5%|▌         | 344/6470 [06:43<1:07:29,  1.51it/s]
1|345|Loss: 1.2893364429473877:   5%|▌         | 344/6470 [06:44<1:07:29,  1.51it/s]
1|345|Loss: 1.2893364429473877:   5%|▌         | 345/6470 [06:44<1:02:01,  1.65it/s]
1|346|Loss: 1.140641212463379:   5%|▌         | 345/6470 [06:44<1:02:01,  1.65it/s] 
1|346|Loss: 1.140641212463379:   5%|▌         | 346/6470 [06:44<1:02:01,  1.65it/s]
1|347|Loss: 1.1314475536346436:   5%|▌         | 346/6470 [06:45<1:02:01,  1.65it/s]
1|347|Loss: 1.1314475536346436:   5%|▌         | 347/6470 [06:45<1:08:31,  1.49it/s]
1|348|Loss: 1.1735670566558838:   5%|▌         | 347/6470 [06:46<1:08:31,  1.49it/s]
1|348|Loss: 1.1735670566558838:   5%|▌         | 348/6470 [06:46<1:14:26,  1.37it/s]
1|349|Loss: 1.317746639251709:   5%|▌         | 348/6470 [06:46<1:14:26,  1.37it/s] 
1|349|Loss: 1.317746639251709:   5%|▌         | 349/6470 [06:47<1:13:15,  1.39it/s]
1|350|Loss: 1.2674064636230469:   5%|▌         | 349/6470 [06:47<1:13:15,  1.39it/s]
1|350|Loss: 1.2674064636230469:   5%|▌         | 350/6470 [06:47<1:10:54,  1.44it/s]
1|351|Loss: 1.0896795988082886:   5%|▌         | 350/6470 [06:48<1:10:54,  1.44it/s]
1|351|Loss: 1.0896795988082886:   5%|▌         | 351/6470 [06:48<1:10:43,  1.44it/s]
1|352|Loss: 1.2520761489868164:   5%|▌         | 351/6470 [06:49<1:10:43,  1.44it/s]
1|352|Loss: 1.2520761489868164:   5%|▌         | 352/6470 [06:49<1:14:44,  1.36it/s]
1|353|Loss: 1.2013920545578003:   5%|▌         | 352/6470 [06:49<1:14:44,  1.36it/s]
1|353|Loss: 1.2013920545578003:   5%|▌         | 353/6470 [06:50<1:15:01,  1.36it/s]
1|354|Loss: 1.2570728063583374:   5%|▌         | 353/6470 [06:50<1:15:01,  1.36it/s]
1|354|Loss: 1.2570728063583374:   5%|▌         | 354/6470 [06:50<1:13:52,  1.38it/s]
1|355|Loss: 1.2690064907073975:   5%|▌         | 354/6470 [06:51<1:13:52,  1.38it/s]
1|355|Loss: 1.2690064907073975:   5%|▌         | 355/6470 [06:51<1:11:03,  1.43it/s]
1|356|Loss: 1.3677946329116821:   5%|▌         | 355/6470 [06:51<1:11:03,  1.43it/s]
1|356|Loss: 1.3677946329116821:   6%|▌         | 356/6470 [06:52<1:07:46,  1.50it/s]
1|357|Loss: 1.1965328454971313:   6%|▌         | 356/6470 [06:52<1:07:46,  1.50it/s]
1|357|Loss: 1.1965328454971313:   6%|▌         | 357/6470 [06:52<1:04:03,  1.59it/s]
1|358|Loss: 1.1244163513183594:   6%|▌         | 357/6470 [06:52<1:04:03,  1.59it/s]
1|358|Loss: 1.1244163513183594:   6%|▌         | 358/6470 [06:53<1:06:39,  1.53it/s]
1|359|Loss: 1.3173991441726685:   6%|▌         | 358/6470 [06:53<1:06:39,  1.53it/s]
1|359|Loss: 1.3173991441726685:   6%|▌         | 359/6470 [06:54<1:07:12,  1.52it/s]
1|360|Loss: 1.2448787689208984:   6%|▌         | 359/6470 [06:54<1:07:12,  1.52it/s]
1|360|Loss: 1.2448787689208984:   6%|▌         | 360/6470 [06:54<1:04:22,  1.58it/s]
1|361|Loss: 1.3107434511184692:   6%|▌         | 360/6470 [06:54<1:04:22,  1.58it/s]
1|361|Loss: 1.3107434511184692:   6%|▌         | 361/6470 [06:55<1:04:05,  1.59it/s]
1|362|Loss: 1.3756775856018066:   6%|▌         | 361/6470 [06:55<1:04:05,  1.59it/s]
1|362|Loss: 1.3756775856018066:   6%|▌         | 362/6470 [06:55<1:01:42,  1.65it/s]
1|363|Loss: 1.2032537460327148:   6%|▌         | 362/6470 [06:56<1:01:42,  1.65it/s]
1|363|Loss: 1.2032537460327148:   6%|▌         | 363/6470 [06:56<58:18,  1.75it/s]  
1|364|Loss: 1.2324084043502808:   6%|▌         | 363/6470 [06:56<58:18,  1.75it/s]
1|364|Loss: 1.2324084043502808:   6%|▌         | 364/6470 [06:56<1:03:06,  1.61it/s]
1|365|Loss: 1.1917823553085327:   6%|▌         | 364/6470 [06:57<1:03:06,  1.61it/s]
1|365|Loss: 1.1917823553085327:   6%|▌         | 365/6470 [06:57<1:08:24,  1.49it/s]
1|366|Loss: 1.209226369857788:   6%|▌         | 365/6470 [06:58<1:08:24,  1.49it/s] 
1|366|Loss: 1.209226369857788:   6%|▌         | 366/6470 [06:58<1:10:49,  1.44it/s]
1|367|Loss: 1.3974193334579468:   6%|▌         | 366/6470 [06:58<1:10:49,  1.44it/s]
1|367|Loss: 1.3974193334579468:   6%|▌         | 367/6470 [06:59<1:09:09,  1.47it/s]
1|368|Loss: 1.2126377820968628:   6%|▌         | 367/6470 [06:59<1:09:09,  1.47it/s]
1|368|Loss: 1.2126377820968628:   6%|▌         | 368/6470 [06:59<1:12:52,  1.40it/s]
1|369|Loss: 1.2940183877944946:   6%|▌         | 368/6470 [07:00<1:12:52,  1.40it/s]
1|369|Loss: 1.2940183877944946:   6%|▌         | 369/6470 [07:00<1:14:13,  1.37it/s]
1|370|Loss: 1.2038605213165283:   6%|▌         | 369/6470 [07:01<1:14:13,  1.37it/s]
1|370|Loss: 1.2038605213165283:   6%|▌         | 370/6470 [07:01<1:18:00,  1.30it/s]
1|371|Loss: 1.3611022233963013:   6%|▌         | 370/6470 [07:01<1:18:00,  1.30it/s]
1|371|Loss: 1.3611022233963013:   6%|▌         | 371/6470 [07:02<1:14:22,  1.37it/s]
1|372|Loss: 1.004569411277771:   6%|▌         | 371/6470 [07:02<1:14:22,  1.37it/s] 
1|372|Loss: 1.004569411277771:   6%|▌         | 372/6470 [07:02<1:13:16,  1.39it/s]
1|373|Loss: 1.1385151147842407:   6%|▌         | 372/6470 [07:03<1:13:16,  1.39it/s]
1|373|Loss: 1.1385151147842407:   6%|▌         | 373/6470 [07:03<1:12:51,  1.39it/s]
1|374|Loss: 1.2663779258728027:   6%|▌         | 373/6470 [07:04<1:12:51,  1.39it/s]
1|374|Loss: 1.2663779258728027:   6%|▌         | 374/6470 [07:04<1:10:50,  1.43it/s]
1|375|Loss: 1.2965853214263916:   6%|▌         | 374/6470 [07:04<1:10:50,  1.43it/s]
1|375|Loss: 1.2965853214263916:   6%|▌         | 375/6470 [07:05<1:11:40,  1.42it/s]
1|376|Loss: 1.3740673065185547:   6%|▌         | 375/6470 [07:05<1:11:40,  1.42it/s]
1|376|Loss: 1.3740673065185547:   6%|▌         | 376/6470 [07:05<1:10:01,  1.45it/s]
1|377|Loss: 1.1489170789718628:   6%|▌         | 376/6470 [07:06<1:10:01,  1.45it/s]
1|377|Loss: 1.1489170789718628:   6%|▌         | 377/6470 [07:06<1:07:51,  1.50it/s]
1|378|Loss: 1.1176950931549072:   6%|▌         | 377/6470 [07:06<1:07:51,  1.50it/s]
1|378|Loss: 1.1176950931549072:   6%|▌         | 378/6470 [07:06<1:08:30,  1.48it/s]
1|379|Loss: 1.0783978700637817:   6%|▌         | 378/6470 [07:07<1:08:30,  1.48it/s]
1|379|Loss: 1.0783978700637817:   6%|▌         | 379/6470 [07:07<1:07:40,  1.50it/s]
1|380|Loss: 1.1559051275253296:   6%|▌         | 379/6470 [07:07<1:07:40,  1.50it/s]
1|380|Loss: 1.1559051275253296:   6%|▌         | 380/6470 [07:08<1:08:27,  1.48it/s]
1|381|Loss: 1.2246294021606445:   6%|▌         | 380/6470 [07:08<1:08:27,  1.48it/s]
1|381|Loss: 1.2246294021606445:   6%|▌         | 381/6470 [07:09<1:10:17,  1.44it/s]
1|382|Loss: 1.2591577768325806:   6%|▌         | 381/6470 [07:09<1:10:17,  1.44it/s]
1|382|Loss: 1.2591577768325806:   6%|▌         | 382/6470 [07:09<1:09:18,  1.46it/s]
1|383|Loss: 1.4158653020858765:   6%|▌         | 382/6470 [07:10<1:09:18,  1.46it/s]
1|383|Loss: 1.4158653020858765:   6%|▌         | 383/6470 [07:10<1:01:37,  1.65it/s]
1|384|Loss: 1.1979233026504517:   6%|▌         | 383/6470 [07:10<1:01:37,  1.65it/s]
1|384|Loss: 1.1979233026504517:   6%|▌         | 384/6470 [07:10<1:01:15,  1.66it/s]
1|385|Loss: 1.3442063331604004:   6%|▌         | 384/6470 [07:11<1:01:15,  1.66it/s]
1|385|Loss: 1.3442063331604004:   6%|▌         | 385/6470 [07:11<1:00:05,  1.69it/s]
1|386|Loss: 1.176762580871582:   6%|▌         | 385/6470 [07:11<1:00:05,  1.69it/s] 
1|386|Loss: 1.176762580871582:   6%|▌         | 386/6470 [07:11<1:00:29,  1.68it/s]
1|387|Loss: 1.1514540910720825:   6%|▌         | 386/6470 [07:12<1:00:29,  1.68it/s]
1|387|Loss: 1.1514540910720825:   6%|▌         | 387/6470 [07:12<1:03:07,  1.61it/s]
1|388|Loss: 1.2774467468261719:   6%|▌         | 387/6470 [07:13<1:03:07,  1.61it/s]
1|388|Loss: 1.2774467468261719:   6%|▌         | 388/6470 [07:13<1:09:06,  1.47it/s]
1|389|Loss: 1.118950366973877:   6%|▌         | 388/6470 [07:13<1:09:06,  1.47it/s] 
1|389|Loss: 1.118950366973877:   6%|▌         | 389/6470 [07:14<1:12:12,  1.40it/s]
1|390|Loss: 1.2402061223983765:   6%|▌         | 389/6470 [07:14<1:12:12,  1.40it/s]
1|390|Loss: 1.2402061223983765:   6%|▌         | 390/6470 [07:14<1:09:29,  1.46it/s]
1|391|Loss: 1.0308796167373657:   6%|▌         | 390/6470 [07:15<1:09:29,  1.46it/s]
1|391|Loss: 1.0308796167373657:   6%|▌         | 391/6470 [07:15<1:10:37,  1.43it/s]
1|392|Loss: 1.2940081357955933:   6%|▌         | 391/6470 [07:15<1:10:37,  1.43it/s]
1|392|Loss: 1.2940081357955933:   6%|▌         | 392/6470 [07:16<1:06:01,  1.53it/s]
1|393|Loss: 0.9996537566184998:   6%|▌         | 392/6470 [07:16<1:06:01,  1.53it/s]
1|393|Loss: 0.9996537566184998:   6%|▌         | 393/6470 [07:16<1:06:14,  1.53it/s]
1|394|Loss: 0.9345635771751404:   6%|▌         | 393/6470 [07:17<1:06:14,  1.53it/s]
1|394|Loss: 0.9345635771751404:   6%|▌         | 394/6470 [07:17<1:10:05,  1.44it/s]
1|395|Loss: 1.2607102394104004:   6%|▌         | 394/6470 [07:17<1:10:05,  1.44it/s]
1|395|Loss: 1.2607102394104004:   6%|▌         | 395/6470 [07:18<1:06:23,  1.52it/s]
1|396|Loss: 1.0172826051712036:   6%|▌         | 395/6470 [07:18<1:06:23,  1.52it/s]
1|396|Loss: 1.0172826051712036:   6%|▌         | 396/6470 [07:18<1:10:51,  1.43it/s]
1|397|Loss: 1.585935115814209:   6%|▌         | 396/6470 [07:19<1:10:51,  1.43it/s] 
1|397|Loss: 1.585935115814209:   6%|▌         | 397/6470 [07:19<1:06:53,  1.51it/s]
1|398|Loss: 1.0754450559616089:   6%|▌         | 397/6470 [07:19<1:06:53,  1.51it/s]
1|398|Loss: 1.0754450559616089:   6%|▌         | 398/6470 [07:20<1:07:14,  1.51it/s]
1|399|Loss: 1.1370693445205688:   6%|▌         | 398/6470 [07:20<1:07:14,  1.51it/s]
1|399|Loss: 1.1370693445205688:   6%|▌         | 399/6470 [07:20<1:10:59,  1.43it/s]
1|400|Loss: 1.178529977798462:   6%|▌         | 399/6470 [07:21<1:10:59,  1.43it/s] 
1|400|Loss: 1.178529977798462:   6%|▌         | 400/6470 [07:21<1:11:45,  1.41it/s]
1|401|Loss: 0.9993912577629089:   6%|▌         | 400/6470 [07:22<1:11:45,  1.41it/s]
1|401|Loss: 0.9993912577629089:   6%|▌         | 401/6470 [07:22<1:15:22,  1.34it/s]
1|402|Loss: 1.1466341018676758:   6%|▌         | 401/6470 [07:22<1:15:22,  1.34it/s]
1|402|Loss: 1.1466341018676758:   6%|▌         | 402/6470 [07:23<1:11:55,  1.41it/s]
1|403|Loss: 1.0082811117172241:   6%|▌         | 402/6470 [07:23<1:11:55,  1.41it/s]
1|403|Loss: 1.0082811117172241:   6%|▌         | 403/6470 [07:23<1:06:10,  1.53it/s]
1|404|Loss: 1.035779356956482:   6%|▌         | 403/6470 [07:24<1:06:10,  1.53it/s] 
1|404|Loss: 1.035779356956482:   6%|▌         | 404/6470 [07:24<1:07:32,  1.50it/s]
1|405|Loss: 1.2911452054977417:   6%|▌         | 404/6470 [07:24<1:07:32,  1.50it/s]
1|405|Loss: 1.2911452054977417:   6%|▋         | 405/6470 [07:24<1:05:20,  1.55it/s]
1|406|Loss: 1.1968138217926025:   6%|▋         | 405/6470 [07:25<1:05:20,  1.55it/s]
1|406|Loss: 1.1968138217926025:   6%|▋         | 406/6470 [07:25<1:09:02,  1.46it/s]
1|407|Loss: 1.1762042045593262:   6%|▋         | 406/6470 [07:26<1:09:02,  1.46it/s]
1|407|Loss: 1.1762042045593262:   6%|▋         | 407/6470 [07:26<1:12:33,  1.39it/s]
1|408|Loss: 1.3169623613357544:   6%|▋         | 407/6470 [07:26<1:12:33,  1.39it/s]
1|408|Loss: 1.3169623613357544:   6%|▋         | 408/6470 [07:27<1:12:39,  1.39it/s]
1|409|Loss: 1.199285626411438:   6%|▋         | 408/6470 [07:27<1:12:39,  1.39it/s] 
1|409|Loss: 1.199285626411438:   6%|▋         | 409/6470 [07:27<1:12:10,  1.40it/s]
1|410|Loss: 1.2885135412216187:   6%|▋         | 409/6470 [07:28<1:12:10,  1.40it/s]
1|410|Loss: 1.2885135412216187:   6%|▋         | 410/6470 [07:28<1:09:46,  1.45it/s]
1|411|Loss: 1.4221009016036987:   6%|▋         | 410/6470 [07:28<1:09:46,  1.45it/s]
1|411|Loss: 1.4221009016036987:   6%|▋         | 411/6470 [07:29<1:04:53,  1.56it/s]
1|412|Loss: 1.1979960203170776:   6%|▋         | 411/6470 [07:29<1:04:53,  1.56it/s]
1|412|Loss: 1.1979960203170776:   6%|▋         | 412/6470 [07:29<1:03:23,  1.59it/s]
1|413|Loss: 1.1178839206695557:   6%|▋         | 412/6470 [07:30<1:03:23,  1.59it/s]
1|413|Loss: 1.1178839206695557:   6%|▋         | 413/6470 [07:30<1:03:01,  1.60it/s]
1|414|Loss: 1.185197114944458:   6%|▋         | 413/6470 [07:30<1:03:01,  1.60it/s] 
1|414|Loss: 1.185197114944458:   6%|▋         | 414/6470 [07:30<1:02:42,  1.61it/s]
1|415|Loss: 1.0747129917144775:   6%|▋         | 414/6470 [07:31<1:02:42,  1.61it/s]
1|415|Loss: 1.0747129917144775:   6%|▋         | 415/6470 [07:31<1:04:41,  1.56it/s]
1|416|Loss: 1.0996121168136597:   6%|▋         | 415/6470 [07:32<1:04:41,  1.56it/s]
1|416|Loss: 1.0996121168136597:   6%|▋         | 416/6470 [07:32<1:07:37,  1.49it/s]
1|417|Loss: 1.3493067026138306:   6%|▋         | 416/6470 [07:32<1:07:37,  1.49it/s]
1|417|Loss: 1.3493067026138306:   6%|▋         | 417/6470 [07:32<1:05:50,  1.53it/s]
1|418|Loss: 1.3505847454071045:   6%|▋         | 417/6470 [07:33<1:05:50,  1.53it/s]
1|418|Loss: 1.3505847454071045:   6%|▋         | 418/6470 [07:33<1:10:11,  1.44it/s]
1|419|Loss: 1.300553560256958:   6%|▋         | 418/6470 [07:34<1:10:11,  1.44it/s] 
1|419|Loss: 1.300553560256958:   6%|▋         | 419/6470 [07:34<1:11:29,  1.41it/s]
1|420|Loss: 1.0621099472045898:   6%|▋         | 419/6470 [07:34<1:11:29,  1.41it/s]
1|420|Loss: 1.0621099472045898:   6%|▋         | 420/6470 [07:35<1:14:26,  1.35it/s]
1|421|Loss: 1.0736680030822754:   6%|▋         | 420/6470 [07:35<1:14:26,  1.35it/s]
1|421|Loss: 1.0736680030822754:   7%|▋         | 421/6470 [07:36<1:17:17,  1.30it/s]
1|422|Loss: 1.5240401029586792:   7%|▋         | 421/6470 [07:36<1:17:17,  1.30it/s]
1|422|Loss: 1.5240401029586792:   7%|▋         | 422/6470 [07:36<1:08:31,  1.47it/s]
1|423|Loss: 1.3472259044647217:   7%|▋         | 422/6470 [07:36<1:08:31,  1.47it/s]
1|423|Loss: 1.3472259044647217:   7%|▋         | 423/6470 [07:37<1:03:49,  1.58it/s]
1|424|Loss: 1.1661628484725952:   7%|▋         | 423/6470 [07:37<1:03:49,  1.58it/s]
1|424|Loss: 1.1661628484725952:   7%|▋         | 424/6470 [07:37<1:05:44,  1.53it/s]
1|425|Loss: 1.2088699340820312:   7%|▋         | 424/6470 [07:38<1:05:44,  1.53it/s]
1|425|Loss: 1.2088699340820312:   7%|▋         | 425/6470 [07:38<1:06:02,  1.53it/s]
1|426|Loss: 1.1932766437530518:   7%|▋         | 425/6470 [07:38<1:06:02,  1.53it/s]
1|426|Loss: 1.1932766437530518:   7%|▋         | 426/6470 [07:39<1:05:32,  1.54it/s]
1|427|Loss: 1.3138556480407715:   7%|▋         | 426/6470 [07:39<1:05:32,  1.54it/s]
1|427|Loss: 1.3138556480407715:   7%|▋         | 427/6470 [07:39<1:06:50,  1.51it/s]
1|428|Loss: 0.9512643218040466:   7%|▋         | 427/6470 [07:40<1:06:50,  1.51it/s]
1|428|Loss: 0.9512643218040466:   7%|▋         | 428/6470 [07:40<1:08:36,  1.47it/s]
1|429|Loss: 1.145135760307312:   7%|▋         | 428/6470 [07:40<1:08:36,  1.47it/s] 
1|429|Loss: 1.145135760307312:   7%|▋         | 429/6470 [07:41<1:05:38,  1.53it/s]
1|430|Loss: 1.372530460357666:   7%|▋         | 429/6470 [07:41<1:05:38,  1.53it/s]
1|430|Loss: 1.372530460357666:   7%|▋         | 430/6470 [07:41<1:03:02,  1.60it/s]
1|431|Loss: 1.2762504816055298:   7%|▋         | 430/6470 [07:42<1:03:02,  1.60it/s]
1|431|Loss: 1.2762504816055298:   7%|▋         | 431/6470 [07:42<1:03:14,  1.59it/s]
1|432|Loss: 1.0836436748504639:   7%|▋         | 431/6470 [07:42<1:03:14,  1.59it/s]
1|432|Loss: 1.0836436748504639:   7%|▋         | 432/6470 [07:43<1:09:29,  1.45it/s]
1|433|Loss: 1.2061936855316162:   7%|▋         | 432/6470 [07:43<1:09:29,  1.45it/s]
1|433|Loss: 1.2061936855316162:   7%|▋         | 433/6470 [07:43<1:12:13,  1.39it/s]
1|434|Loss: 1.2460633516311646:   7%|▋         | 433/6470 [07:44<1:12:13,  1.39it/s]
1|434|Loss: 1.2460633516311646:   7%|▋         | 434/6470 [07:44<1:08:35,  1.47it/s]
1|435|Loss: 1.1703052520751953:   7%|▋         | 434/6470 [07:44<1:08:35,  1.47it/s]
1|435|Loss: 1.1703052520751953:   7%|▋         | 435/6470 [07:45<1:10:51,  1.42it/s]
1|436|Loss: 1.3180665969848633:   7%|▋         | 435/6470 [07:45<1:10:51,  1.42it/s]
1|436|Loss: 1.3180665969848633:   7%|▋         | 436/6470 [07:45<1:09:54,  1.44it/s]
1|437|Loss: 1.122549295425415:   7%|▋         | 436/6470 [07:46<1:09:54,  1.44it/s] 
1|437|Loss: 1.122549295425415:   7%|▋         | 437/6470 [07:46<1:06:55,  1.50it/s]
1|438|Loss: 1.0194505453109741:   7%|▋         | 437/6470 [07:46<1:06:55,  1.50it/s]
1|438|Loss: 1.0194505453109741:   7%|▋         | 438/6470 [07:47<1:05:50,  1.53it/s]
1|439|Loss: 1.2945764064788818:   7%|▋         | 438/6470 [07:47<1:05:50,  1.53it/s]
1|439|Loss: 1.2945764064788818:   7%|▋         | 439/6470 [07:47<1:09:51,  1.44it/s]
1|440|Loss: 1.3368616104125977:   7%|▋         | 439/6470 [07:48<1:09:51,  1.44it/s]
1|440|Loss: 1.3368616104125977:   7%|▋         | 440/6470 [07:48<1:14:26,  1.35it/s]
1|441|Loss: 1.240148663520813:   7%|▋         | 440/6470 [07:49<1:14:26,  1.35it/s] 
1|441|Loss: 1.240148663520813:   7%|▋         | 441/6470 [07:49<1:15:01,  1.34it/s]
1|442|Loss: 1.2041194438934326:   7%|▋         | 441/6470 [07:50<1:15:01,  1.34it/s]
1|442|Loss: 1.2041194438934326:   7%|▋         | 442/6470 [07:50<1:15:47,  1.33it/s]
1|443|Loss: 1.2848894596099854:   7%|▋         | 442/6470 [07:50<1:15:47,  1.33it/s]
1|443|Loss: 1.2848894596099854:   7%|▋         | 443/6470 [07:51<1:17:55,  1.29it/s]
1|444|Loss: 1.13716459274292:   7%|▋         | 443/6470 [07:51<1:17:55,  1.29it/s]  
1|444|Loss: 1.13716459274292:   7%|▋         | 444/6470 [07:52<1:21:25,  1.23it/s]
1|445|Loss: 1.0459630489349365:   7%|▋         | 444/6470 [07:52<1:21:25,  1.23it/s]
1|445|Loss: 1.0459630489349365:   7%|▋         | 445/6470 [07:52<1:20:28,  1.25it/s]
1|446|Loss: 1.0521581172943115:   7%|▋         | 445/6470 [07:53<1:20:28,  1.25it/s]
1|446|Loss: 1.0521581172943115:   7%|▋         | 446/6470 [07:53<1:16:30,  1.31it/s]
1|447|Loss: 1.176719069480896:   7%|▋         | 446/6470 [07:53<1:16:30,  1.31it/s] 
1|447|Loss: 1.176719069480896:   7%|▋         | 447/6470 [07:54<1:10:41,  1.42it/s]
1|448|Loss: 1.1600182056427002:   7%|▋         | 447/6470 [07:54<1:10:41,  1.42it/s]
1|448|Loss: 1.1600182056427002:   7%|▋         | 448/6470 [07:54<1:09:50,  1.44it/s]
1|449|Loss: 1.2696970701217651:   7%|▋         | 448/6470 [07:55<1:09:50,  1.44it/s]
1|449|Loss: 1.2696970701217651:   7%|▋         | 449/6470 [07:55<1:07:55,  1.48it/s]
1|450|Loss: 1.2959234714508057:   7%|▋         | 449/6470 [07:55<1:07:55,  1.48it/s]
1|450|Loss: 1.2959234714508057:   7%|▋         | 450/6470 [07:56<1:08:10,  1.47it/s]
1|451|Loss: 1.4718602895736694:   7%|▋         | 450/6470 [07:56<1:08:10,  1.47it/s]
1|451|Loss: 1.4718602895736694:   7%|▋         | 451/6470 [07:56<1:04:07,  1.56it/s]
1|452|Loss: 1.4701029062271118:   7%|▋         | 451/6470 [07:56<1:04:07,  1.56it/s]
1|452|Loss: 1.4701029062271118:   7%|▋         | 452/6470 [07:57<56:34,  1.77it/s]  
1|453|Loss: 1.200043797492981:   7%|▋         | 452/6470 [07:57<56:34,  1.77it/s] 
1|453|Loss: 1.200043797492981:   7%|▋         | 453/6470 [07:57<55:45,  1.80it/s]
1|454|Loss: 1.0267598628997803:   7%|▋         | 453/6470 [07:57<55:45,  1.80it/s]
1|454|Loss: 1.0267598628997803:   7%|▋         | 454/6470 [07:58<58:05,  1.73it/s]
1|455|Loss: 1.3133217096328735:   7%|▋         | 454/6470 [07:58<58:05,  1.73it/s]
1|455|Loss: 1.3133217096328735:   7%|▋         | 455/6470 [07:59<1:05:48,  1.52it/s]
1|456|Loss: 1.2938973903656006:   7%|▋         | 455/6470 [07:59<1:05:48,  1.52it/s]
1|456|Loss: 1.2938973903656006:   7%|▋         | 456/6470 [07:59<1:06:55,  1.50it/s]
1|457|Loss: 1.2561198472976685:   7%|▋         | 456/6470 [08:00<1:06:55,  1.50it/s]
1|457|Loss: 1.2561198472976685:   7%|▋         | 457/6470 [08:00<1:06:09,  1.51it/s]
1|458|Loss: 1.1848063468933105:   7%|▋         | 457/6470 [08:00<1:06:09,  1.51it/s]
1|458|Loss: 1.1848063468933105:   7%|▋         | 458/6470 [08:01<1:08:09,  1.47it/s]
1|459|Loss: 1.234032392501831:   7%|▋         | 458/6470 [08:01<1:08:09,  1.47it/s] 
1|459|Loss: 1.234032392501831:   7%|▋         | 459/6470 [08:01<1:07:17,  1.49it/s]
1|460|Loss: 1.1397278308868408:   7%|▋         | 459/6470 [08:02<1:07:17,  1.49it/s]
1|460|Loss: 1.1397278308868408:   7%|▋         | 460/6470 [08:02<1:11:14,  1.41it/s]
1|461|Loss: 1.1592432260513306:   7%|▋         | 460/6470 [08:02<1:11:14,  1.41it/s]
1|461|Loss: 1.1592432260513306:   7%|▋         | 461/6470 [08:03<1:10:40,  1.42it/s]
1|462|Loss: 1.3649178743362427:   7%|▋         | 461/6470 [08:03<1:10:40,  1.42it/s]
1|462|Loss: 1.3649178743362427:   7%|▋         | 462/6470 [08:03<1:10:46,  1.41it/s]
1|463|Loss: 1.1260297298431396:   7%|▋         | 462/6470 [08:04<1:10:46,  1.41it/s]
1|463|Loss: 1.1260297298431396:   7%|▋         | 463/6470 [08:04<1:08:42,  1.46it/s]
1|464|Loss: 1.106885313987732:   7%|▋         | 463/6470 [08:04<1:08:42,  1.46it/s] 
1|464|Loss: 1.106885313987732:   7%|▋         | 464/6470 [08:05<1:06:34,  1.50it/s]
1|465|Loss: 0.9172400832176208:   7%|▋         | 464/6470 [08:05<1:06:34,  1.50it/s]
1|465|Loss: 0.9172400832176208:   7%|▋         | 465/6470 [08:05<1:09:19,  1.44it/s]
1|466|Loss: 1.192630410194397:   7%|▋         | 465/6470 [08:06<1:09:19,  1.44it/s] 
1|466|Loss: 1.192630410194397:   7%|▋         | 466/6470 [08:06<1:11:25,  1.40it/s]
1|467|Loss: 1.2001636028289795:   7%|▋         | 466/6470 [08:07<1:11:25,  1.40it/s]
1|467|Loss: 1.2001636028289795:   7%|▋         | 467/6470 [08:07<1:12:39,  1.38it/s]
1|468|Loss: 1.254255771636963:   7%|▋         | 467/6470 [08:07<1:12:39,  1.38it/s] 
1|468|Loss: 1.254255771636963:   7%|▋         | 468/6470 [08:08<1:07:45,  1.48it/s]
1|469|Loss: 1.116076111793518:   7%|▋         | 468/6470 [08:08<1:07:45,  1.48it/s]
1|469|Loss: 1.116076111793518:   7%|▋         | 469/6470 [08:08<1:04:12,  1.56it/s]
1|470|Loss: 1.3046807050704956:   7%|▋         | 469/6470 [08:08<1:04:12,  1.56it/s]
1|470|Loss: 1.3046807050704956:   7%|▋         | 470/6470 [08:09<1:02:06,  1.61it/s]
1|471|Loss: 1.126526117324829:   7%|▋         | 470/6470 [08:09<1:02:06,  1.61it/s] 
1|471|Loss: 1.126526117324829:   7%|▋         | 471/6470 [08:09<1:03:51,  1.57it/s]
1|472|Loss: 1.0037232637405396:   7%|▋         | 471/6470 [08:10<1:03:51,  1.57it/s]
1|472|Loss: 1.0037232637405396:   7%|▋         | 472/6470 [08:10<1:08:13,  1.47it/s]
1|473|Loss: 1.2121248245239258:   7%|▋         | 472/6470 [08:11<1:08:13,  1.47it/s]
1|473|Loss: 1.2121248245239258:   7%|▋         | 473/6470 [08:11<1:06:30,  1.50it/s]
1|474|Loss: 1.2850936651229858:   7%|▋         | 473/6470 [08:11<1:06:30,  1.50it/s]
1|474|Loss: 1.2850936651229858:   7%|▋         | 474/6470 [08:11<1:06:55,  1.49it/s]
1|475|Loss: 1.2854052782058716:   7%|▋         | 474/6470 [08:12<1:06:55,  1.49it/s]
1|475|Loss: 1.2854052782058716:   7%|▋         | 475/6470 [08:12<1:07:57,  1.47it/s]
1|476|Loss: 1.069892168045044:   7%|▋         | 475/6470 [08:13<1:07:57,  1.47it/s] 
1|476|Loss: 1.069892168045044:   7%|▋         | 476/6470 [08:13<1:07:11,  1.49it/s]
1|477|Loss: 0.9199583530426025:   7%|▋         | 476/6470 [08:13<1:07:11,  1.49it/s]
1|477|Loss: 0.9199583530426025:   7%|▋         | 477/6470 [08:14<1:08:29,  1.46it/s]
1|478|Loss: 1.0860291719436646:   7%|▋         | 477/6470 [08:14<1:08:29,  1.46it/s]
1|478|Loss: 1.0860291719436646:   7%|▋         | 478/6470 [08:14<1:14:27,  1.34it/s]
1|479|Loss: 1.200560450553894:   7%|▋         | 478/6470 [08:15<1:14:27,  1.34it/s] 
1|479|Loss: 1.200560450553894:   7%|▋         | 479/6470 [08:15<1:14:33,  1.34it/s]
1|480|Loss: 1.111122488975525:   7%|▋         | 479/6470 [08:16<1:14:33,  1.34it/s]
1|480|Loss: 1.111122488975525:   7%|▋         | 480/6470 [08:16<1:10:04,  1.42it/s]
1|481|Loss: 1.2669854164123535:   7%|▋         | 480/6470 [08:16<1:10:04,  1.42it/s]
1|481|Loss: 1.2669854164123535:   7%|▋         | 481/6470 [08:16<1:07:13,  1.48it/s]
1|482|Loss: 1.1153059005737305:   7%|▋         | 481/6470 [08:17<1:07:13,  1.48it/s]
1|482|Loss: 1.1153059005737305:   7%|▋         | 482/6470 [08:17<1:07:36,  1.48it/s]
1|483|Loss: 1.141824722290039:   7%|▋         | 482/6470 [08:17<1:07:36,  1.48it/s] 
1|483|Loss: 1.141824722290039:   7%|▋         | 483/6470 [08:18<1:06:46,  1.49it/s]
1|484|Loss: 1.0268516540527344:   7%|▋         | 483/6470 [08:18<1:06:46,  1.49it/s]
1|484|Loss: 1.0268516540527344:   7%|▋         | 484/6470 [08:18<1:08:22,  1.46it/s]
1|485|Loss: 1.1134719848632812:   7%|▋         | 484/6470 [08:19<1:08:22,  1.46it/s]
1|485|Loss: 1.1134719848632812:   7%|▋         | 485/6470 [08:19<1:09:18,  1.44it/s]
1|486|Loss: 1.227873682975769:   7%|▋         | 485/6470 [08:20<1:09:18,  1.44it/s] 
1|486|Loss: 1.227873682975769:   8%|▊         | 486/6470 [08:20<1:08:09,  1.46it/s]
1|487|Loss: 1.06171715259552:   8%|▊         | 486/6470 [08:20<1:08:09,  1.46it/s] 
1|487|Loss: 1.06171715259552:   8%|▊         | 487/6470 [08:20<1:03:40,  1.57it/s]
1|488|Loss: 1.1887303590774536:   8%|▊         | 487/6470 [08:21<1:03:40,  1.57it/s]
1|488|Loss: 1.1887303590774536:   8%|▊         | 488/6470 [08:21<1:02:45,  1.59it/s]
1|489|Loss: 1.125168800354004:   8%|▊         | 488/6470 [08:21<1:02:45,  1.59it/s] 
1|489|Loss: 1.125168800354004:   8%|▊         | 489/6470 [08:22<1:09:33,  1.43it/s]
1|490|Loss: 1.2840709686279297:   8%|▊         | 489/6470 [08:22<1:09:33,  1.43it/s]
1|490|Loss: 1.2840709686279297:   8%|▊         | 490/6470 [08:23<1:09:32,  1.43it/s]
1|491|Loss: 1.2993943691253662:   8%|▊         | 490/6470 [08:23<1:09:32,  1.43it/s]
1|491|Loss: 1.2993943691253662:   8%|▊         | 491/6470 [08:23<1:07:39,  1.47it/s]
1|492|Loss: 0.956386387348175:   8%|▊         | 491/6470 [08:24<1:07:39,  1.47it/s] 
1|492|Loss: 0.956386387348175:   8%|▊         | 492/6470 [08:24<1:12:21,  1.38it/s]
1|493|Loss: 1.2740155458450317:   8%|▊         | 492/6470 [08:24<1:12:21,  1.38it/s]
1|493|Loss: 1.2740155458450317:   8%|▊         | 493/6470 [08:25<1:13:29,  1.36it/s]
1|494|Loss: 1.2301833629608154:   8%|▊         | 493/6470 [08:25<1:13:29,  1.36it/s]
1|494|Loss: 1.2301833629608154:   8%|▊         | 494/6470 [08:25<1:10:36,  1.41it/s]
1|495|Loss: 1.134412169456482:   8%|▊         | 494/6470 [08:26<1:10:36,  1.41it/s] 
1|495|Loss: 1.134412169456482:   8%|▊         | 495/6470 [08:26<1:06:36,  1.50it/s]
1|496|Loss: 1.1320557594299316:   8%|▊         | 495/6470 [08:26<1:06:36,  1.50it/s]
1|496|Loss: 1.1320557594299316:   8%|▊         | 496/6470 [08:27<1:09:11,  1.44it/s]
1|497|Loss: 1.055912971496582:   8%|▊         | 496/6470 [08:27<1:09:11,  1.44it/s] 
1|497|Loss: 1.055912971496582:   8%|▊         | 497/6470 [08:28<1:13:00,  1.36it/s]
1|498|Loss: 1.3718894720077515:   8%|▊         | 497/6470 [08:28<1:13:00,  1.36it/s]
1|498|Loss: 1.3718894720077515:   8%|▊         | 498/6470 [08:28<1:09:18,  1.44it/s]
1|499|Loss: 1.1325652599334717:   8%|▊         | 498/6470 [08:29<1:09:18,  1.44it/s]
1|499|Loss: 1.1325652599334717:   8%|▊         | 499/6470 [08:29<1:11:29,  1.39it/s]
1|500|Loss: 1.2875940799713135:   8%|▊         | 499/6470 [08:29<1:11:29,  1.39it/s]
1|500|Loss: 1.2875940799713135:   8%|▊         | 500/6470 [08:29<1:05:51,  1.51it/s]
1|501|Loss: 1.0785681009292603:   8%|▊         | 500/6470 [08:30<1:05:51,  1.51it/s]
1|501|Loss: 1.0785681009292603:   8%|▊         | 501/6470 [08:30<1:03:52,  1.56it/s]
1|502|Loss: 1.0945123434066772:   8%|▊         | 501/6470 [08:30<1:03:52,  1.56it/s]
1|502|Loss: 1.0945123434066772:   8%|▊         | 502/6470 [08:31<1:03:48,  1.56it/s]
1|503|Loss: 1.3051714897155762:   8%|▊         | 502/6470 [08:31<1:03:48,  1.56it/s]
1|503|Loss: 1.3051714897155762:   8%|▊         | 503/6470 [08:31<1:00:38,  1.64it/s]
1|504|Loss: 0.9425746202468872:   8%|▊         | 503/6470 [08:32<1:00:38,  1.64it/s]
1|504|Loss: 0.9425746202468872:   8%|▊         | 504/6470 [08:32<1:05:11,  1.53it/s]
1|505|Loss: 1.1038827896118164:   8%|▊         | 504/6470 [08:32<1:05:11,  1.53it/s]
1|505|Loss: 1.1038827896118164:   8%|▊         | 505/6470 [08:33<1:07:17,  1.48it/s]
1|506|Loss: 1.1107332706451416:   8%|▊         | 505/6470 [08:33<1:07:17,  1.48it/s]
1|506|Loss: 1.1107332706451416:   8%|▊         | 506/6470 [08:33<1:04:59,  1.53it/s]
1|507|Loss: 1.0326164960861206:   8%|▊         | 506/6470 [08:34<1:04:59,  1.53it/s]
1|507|Loss: 1.0326164960861206:   8%|▊         | 507/6470 [08:34<1:10:48,  1.40it/s]
1|508|Loss: 1.5797902345657349:   8%|▊         | 507/6470 [08:35<1:10:48,  1.40it/s]
1|508|Loss: 1.5797902345657349:   8%|▊         | 508/6470 [08:35<1:03:40,  1.56it/s]
1|509|Loss: 1.3467686176300049:   8%|▊         | 508/6470 [08:35<1:03:40,  1.56it/s]
1|509|Loss: 1.3467686176300049:   8%|▊         | 509/6470 [08:35<1:04:19,  1.54it/s]
1|510|Loss: 1.298073649406433:   8%|▊         | 509/6470 [08:36<1:04:19,  1.54it/s] 
1|510|Loss: 1.298073649406433:   8%|▊         | 510/6470 [08:36<1:04:54,  1.53it/s]
1|511|Loss: 1.2772705554962158:   8%|▊         | 510/6470 [08:36<1:04:54,  1.53it/s]
1|511|Loss: 1.2772705554962158:   8%|▊         | 511/6470 [08:37<1:02:21,  1.59it/s]
1|512|Loss: 1.2062886953353882:   8%|▊         | 511/6470 [08:37<1:02:21,  1.59it/s]
1|512|Loss: 1.2062886953353882:   8%|▊         | 512/6470 [08:37<1:07:19,  1.47it/s]
1|513|Loss: 1.0365983247756958:   8%|▊         | 512/6470 [08:38<1:07:19,  1.47it/s]
1|513|Loss: 1.0365983247756958:   8%|▊         | 513/6470 [08:38<1:11:06,  1.40it/s]
1|514|Loss: 1.1301730871200562:   8%|▊         | 513/6470 [08:39<1:11:06,  1.40it/s]
1|514|Loss: 1.1301730871200562:   8%|▊         | 514/6470 [08:39<1:09:15,  1.43it/s]
1|515|Loss: 1.2922606468200684:   8%|▊         | 514/6470 [08:39<1:09:15,  1.43it/s]
1|515|Loss: 1.2922606468200684:   8%|▊         | 515/6470 [08:39<1:06:47,  1.49it/s]
1|516|Loss: 1.166774868965149:   8%|▊         | 515/6470 [08:40<1:06:47,  1.49it/s] 
1|516|Loss: 1.166774868965149:   8%|▊         | 516/6470 [08:40<1:08:20,  1.45it/s]
1|517|Loss: 1.098609447479248:   8%|▊         | 516/6470 [08:41<1:08:20,  1.45it/s]
1|517|Loss: 1.098609447479248:   8%|▊         | 517/6470 [08:41<1:12:49,  1.36it/s]
1|518|Loss: 1.0661067962646484:   8%|▊         | 517/6470 [08:41<1:12:49,  1.36it/s]
1|518|Loss: 1.0661067962646484:   8%|▊         | 518/6470 [08:42<1:15:58,  1.31it/s]
1|519|Loss: 1.2072657346725464:   8%|▊         | 518/6470 [08:42<1:15:58,  1.31it/s]
1|519|Loss: 1.2072657346725464:   8%|▊         | 519/6470 [08:43<1:14:45,  1.33it/s]
1|520|Loss: 1.0900235176086426:   8%|▊         | 519/6470 [08:43<1:14:45,  1.33it/s]
1|520|Loss: 1.0900235176086426:   8%|▊         | 520/6470 [08:43<1:15:14,  1.32it/s]
1|521|Loss: 1.3186769485473633:   8%|▊         | 520/6470 [08:44<1:15:14,  1.32it/s]
1|521|Loss: 1.3186769485473633:   8%|▊         | 521/6470 [08:44<1:11:20,  1.39it/s]
1|522|Loss: 1.062718152999878:   8%|▊         | 521/6470 [08:44<1:11:20,  1.39it/s] 
1|522|Loss: 1.062718152999878:   8%|▊         | 522/6470 [08:45<1:13:18,  1.35it/s]
1|523|Loss: 1.096472978591919:   8%|▊         | 522/6470 [08:45<1:13:18,  1.35it/s]
1|523|Loss: 1.096472978591919:   8%|▊         | 523/6470 [08:45<1:11:33,  1.39it/s]
1|524|Loss: 1.413132905960083:   8%|▊         | 523/6470 [08:46<1:11:33,  1.39it/s]
1|524|Loss: 1.413132905960083:   8%|▊         | 524/6470 [08:46<1:08:27,  1.45it/s]
1|525|Loss: 1.1234831809997559:   8%|▊         | 524/6470 [08:46<1:08:27,  1.45it/s]
1|525|Loss: 1.1234831809997559:   8%|▊         | 525/6470 [08:47<1:09:04,  1.43it/s]
1|526|Loss: 1.0480700731277466:   8%|▊         | 525/6470 [08:47<1:09:04,  1.43it/s]
1|526|Loss: 1.0480700731277466:   8%|▊         | 526/6470 [08:48<1:14:24,  1.33it/s]
1|527|Loss: 1.0819194316864014:   8%|▊         | 526/6470 [08:48<1:14:24,  1.33it/s]
1|527|Loss: 1.0819194316864014:   8%|▊         | 527/6470 [08:48<1:12:57,  1.36it/s]
1|528|Loss: 1.0102719068527222:   8%|▊         | 527/6470 [08:49<1:12:57,  1.36it/s]
1|528|Loss: 1.0102719068527222:   8%|▊         | 528/6470 [08:49<1:12:19,  1.37it/s]
1|529|Loss: 1.0674521923065186:   8%|▊         | 528/6470 [08:49<1:12:19,  1.37it/s]
1|529|Loss: 1.0674521923065186:   8%|▊         | 529/6470 [08:50<1:12:46,  1.36it/s]
1|530|Loss: 1.1755025386810303:   8%|▊         | 529/6470 [08:50<1:12:46,  1.36it/s]
1|530|Loss: 1.1755025386810303:   8%|▊         | 530/6470 [08:51<1:13:06,  1.35it/s]
1|531|Loss: 1.14215087890625:   8%|▊         | 530/6470 [08:51<1:13:06,  1.35it/s]  
1|531|Loss: 1.14215087890625:   8%|▊         | 531/6470 [08:51<1:14:34,  1.33it/s]
1|532|Loss: 1.1268692016601562:   8%|▊         | 531/6470 [08:52<1:14:34,  1.33it/s]
1|532|Loss: 1.1268692016601562:   8%|▊         | 532/6470 [08:52<1:12:20,  1.37it/s]
1|533|Loss: 1.1979808807373047:   8%|▊         | 532/6470 [08:52<1:12:20,  1.37it/s]
1|533|Loss: 1.1979808807373047:   8%|▊         | 533/6470 [08:53<1:12:09,  1.37it/s]
1|534|Loss: 1.0182433128356934:   8%|▊         | 533/6470 [08:53<1:12:09,  1.37it/s]
1|534|Loss: 1.0182433128356934:   8%|▊         | 534/6470 [08:54<1:17:49,  1.27it/s]
1|535|Loss: 1.1474143266677856:   8%|▊         | 534/6470 [08:54<1:17:49,  1.27it/s]
1|535|Loss: 1.1474143266677856:   8%|▊         | 535/6470 [08:54<1:15:36,  1.31it/s]
1|536|Loss: 1.1162666082382202:   8%|▊         | 535/6470 [08:55<1:15:36,  1.31it/s]
1|536|Loss: 1.1162666082382202:   8%|▊         | 536/6470 [08:55<1:16:09,  1.30it/s]
1|537|Loss: 1.355203628540039:   8%|▊         | 536/6470 [08:55<1:16:09,  1.30it/s] 
1|537|Loss: 1.355203628540039:   8%|▊         | 537/6470 [08:56<1:09:34,  1.42it/s]
1|538|Loss: 1.1514074802398682:   8%|▊         | 537/6470 [08:56<1:09:34,  1.42it/s]
1|538|Loss: 1.1514074802398682:   8%|▊         | 538/6470 [08:56<1:02:17,  1.59it/s]
1|539|Loss: 1.238270878791809:   8%|▊         | 538/6470 [08:56<1:02:17,  1.59it/s] 
1|539|Loss: 1.238270878791809:   8%|▊         | 539/6470 [08:57<1:01:17,  1.61it/s]
1|540|Loss: 1.0493465662002563:   8%|▊         | 539/6470 [08:57<1:01:17,  1.61it/s]
1|540|Loss: 1.0493465662002563:   8%|▊         | 540/6470 [08:58<1:07:01,  1.47it/s]
1|541|Loss: 1.0718668699264526:   8%|▊         | 540/6470 [08:58<1:07:01,  1.47it/s]
1|541|Loss: 1.0718668699264526:   8%|▊         | 541/6470 [08:58<1:06:16,  1.49it/s]
1|542|Loss: 1.1093252897262573:   8%|▊         | 541/6470 [08:59<1:06:16,  1.49it/s]
1|542|Loss: 1.1093252897262573:   8%|▊         | 542/6470 [08:59<1:08:27,  1.44it/s]
1|543|Loss: 1.3542654514312744:   8%|▊         | 542/6470 [08:59<1:08:27,  1.44it/s]
1|543|Loss: 1.3542654514312744:   8%|▊         | 543/6470 [09:00<1:09:12,  1.43it/s]
1|544|Loss: 1.2717223167419434:   8%|▊         | 543/6470 [09:00<1:09:12,  1.43it/s]
1|544|Loss: 1.2717223167419434:   8%|▊         | 544/6470 [09:00<1:07:39,  1.46it/s]
1|545|Loss: 1.1915556192398071:   8%|▊         | 544/6470 [09:01<1:07:39,  1.46it/s]
1|545|Loss: 1.1915556192398071:   8%|▊         | 545/6470 [09:01<1:02:31,  1.58it/s]
1|546|Loss: 1.3368592262268066:   8%|▊         | 545/6470 [09:01<1:02:31,  1.58it/s]
1|546|Loss: 1.3368592262268066:   8%|▊         | 546/6470 [09:01<1:01:36,  1.60it/s]
1|547|Loss: 1.1443687677383423:   8%|▊         | 546/6470 [09:02<1:01:36,  1.60it/s]
1|547|Loss: 1.1443687677383423:   8%|▊         | 547/6470 [09:02<1:05:40,  1.50it/s]
1|548|Loss: 1.18953275680542:   8%|▊         | 547/6470 [09:03<1:05:40,  1.50it/s]  
1|548|Loss: 1.18953275680542:   8%|▊         | 548/6470 [09:03<1:05:48,  1.50it/s]
1|549|Loss: 1.2853422164916992:   8%|▊         | 548/6470 [09:03<1:05:48,  1.50it/s]
1|549|Loss: 1.2853422164916992:   8%|▊         | 549/6470 [09:04<1:06:23,  1.49it/s]
1|550|Loss: 1.28239905834198:   8%|▊         | 549/6470 [09:04<1:06:23,  1.49it/s]  
1|550|Loss: 1.28239905834198:   9%|▊         | 550/6470 [09:04<1:05:18,  1.51it/s]
1|551|Loss: 1.1946358680725098:   9%|▊         | 550/6470 [09:05<1:05:18,  1.51it/s]
1|551|Loss: 1.1946358680725098:   9%|▊         | 551/6470 [09:05<1:09:42,  1.42it/s]
1|552|Loss: 1.1612634658813477:   9%|▊         | 551/6470 [09:05<1:09:42,  1.42it/s]
1|552|Loss: 1.1612634658813477:   9%|▊         | 552/6470 [09:06<1:11:12,  1.39it/s]
1|553|Loss: 1.0990650653839111:   9%|▊         | 552/6470 [09:06<1:11:12,  1.39it/s]
1|553|Loss: 1.0990650653839111:   9%|▊         | 553/6470 [09:06<1:06:09,  1.49it/s]
1|554|Loss: 1.1054662466049194:   9%|▊         | 553/6470 [09:07<1:06:09,  1.49it/s]
1|554|Loss: 1.1054662466049194:   9%|▊         | 554/6470 [09:07<1:04:33,  1.53it/s]
1|555|Loss: 1.1034200191497803:   9%|▊         | 554/6470 [09:07<1:04:33,  1.53it/s]
1|555|Loss: 1.1034200191497803:   9%|▊         | 555/6470 [09:08<1:05:45,  1.50it/s]
1|556|Loss: 1.139624834060669:   9%|▊         | 555/6470 [09:08<1:05:45,  1.50it/s] 
1|556|Loss: 1.139624834060669:   9%|▊         | 556/6470 [09:08<1:07:34,  1.46it/s]
1|557|Loss: 1.1743484735488892:   9%|▊         | 556/6470 [09:09<1:07:34,  1.46it/s]
1|557|Loss: 1.1743484735488892:   9%|▊         | 557/6470 [09:09<1:11:49,  1.37it/s]
1|558|Loss: 1.2859939336776733:   9%|▊         | 557/6470 [09:10<1:11:49,  1.37it/s]
1|558|Loss: 1.2859939336776733:   9%|▊         | 558/6470 [09:10<1:12:46,  1.35it/s]
1|559|Loss: 1.4179097414016724:   9%|▊         | 558/6470 [09:10<1:12:46,  1.35it/s]
1|559|Loss: 1.4179097414016724:   9%|▊         | 559/6470 [09:10<1:07:18,  1.46it/s]
1|560|Loss: 1.0926891565322876:   9%|▊         | 559/6470 [09:11<1:07:18,  1.46it/s]
1|560|Loss: 1.0926891565322876:   9%|▊         | 560/6470 [09:11<1:10:35,  1.40it/s]
1|561|Loss: 1.1514192819595337:   9%|▊         | 560/6470 [09:12<1:10:35,  1.40it/s]
1|561|Loss: 1.1514192819595337:   9%|▊         | 561/6470 [09:12<1:08:29,  1.44it/s]
1|562|Loss: 1.0571529865264893:   9%|▊         | 561/6470 [09:12<1:08:29,  1.44it/s]
1|562|Loss: 1.0571529865264893:   9%|▊         | 562/6470 [09:13<1:12:14,  1.36it/s]
1|563|Loss: 1.2468929290771484:   9%|▊         | 562/6470 [09:13<1:12:14,  1.36it/s]
1|563|Loss: 1.2468929290771484:   9%|▊         | 563/6470 [09:13<1:12:32,  1.36it/s]
1|564|Loss: 1.314945936203003:   9%|▊         | 563/6470 [09:14<1:12:32,  1.36it/s] 
1|564|Loss: 1.314945936203003:   9%|▊         | 564/6470 [09:14<1:08:14,  1.44it/s]
1|565|Loss: 1.081529974937439:   9%|▊         | 564/6470 [09:14<1:08:14,  1.44it/s]
1|565|Loss: 1.081529974937439:   9%|▊         | 565/6470 [09:15<1:04:41,  1.52it/s]
1|566|Loss: 1.2387714385986328:   9%|▊         | 565/6470 [09:15<1:04:41,  1.52it/s]
1|566|Loss: 1.2387714385986328:   9%|▊         | 566/6470 [09:15<1:03:46,  1.54it/s]
1|567|Loss: 1.1495847702026367:   9%|▊         | 566/6470 [09:16<1:03:46,  1.54it/s]
1|567|Loss: 1.1495847702026367:   9%|▉         | 567/6470 [09:16<1:03:09,  1.56it/s]
1|568|Loss: 1.0960711240768433:   9%|▉         | 567/6470 [09:16<1:03:09,  1.56it/s]
1|568|Loss: 1.0960711240768433:   9%|▉         | 568/6470 [09:17<1:02:29,  1.57it/s]
1|569|Loss: 1.2997150421142578:   9%|▉         | 568/6470 [09:17<1:02:29,  1.57it/s]
1|569|Loss: 1.2997150421142578:   9%|▉         | 569/6470 [09:17<1:01:02,  1.61it/s]
1|570|Loss: 0.926150381565094:   9%|▉         | 569/6470 [09:18<1:01:02,  1.61it/s] 
1|570|Loss: 0.926150381565094:   9%|▉         | 570/6470 [09:18<1:03:55,  1.54it/s]
1|571|Loss: 1.11149263381958:   9%|▉         | 570/6470 [09:18<1:03:55,  1.54it/s] 
1|571|Loss: 1.11149263381958:   9%|▉         | 571/6470 [09:19<1:07:51,  1.45it/s]
1|572|Loss: 1.1173585653305054:   9%|▉         | 571/6470 [09:19<1:07:51,  1.45it/s]
1|572|Loss: 1.1173585653305054:   9%|▉         | 572/6470 [09:19<1:05:49,  1.49it/s]
1|573|Loss: 0.9863489270210266:   9%|▉         | 572/6470 [09:20<1:05:49,  1.49it/s]
1|573|Loss: 0.9863489270210266:   9%|▉         | 573/6470 [09:20<1:09:00,  1.42it/s]
1|574|Loss: 1.2803291082382202:   9%|▉         | 573/6470 [09:20<1:09:00,  1.42it/s]
1|574|Loss: 1.2803291082382202:   9%|▉         | 574/6470 [09:21<1:07:52,  1.45it/s]
1|575|Loss: 1.0790892839431763:   9%|▉         | 574/6470 [09:21<1:07:52,  1.45it/s]
1|575|Loss: 1.0790892839431763:   9%|▉         | 575/6470 [09:21<1:07:36,  1.45it/s]
1|576|Loss: 1.188490390777588:   9%|▉         | 575/6470 [09:22<1:07:36,  1.45it/s] 
1|576|Loss: 1.188490390777588:   9%|▉         | 576/6470 [09:22<1:06:28,  1.48it/s]
1|577|Loss: 0.9717691540718079:   9%|▉         | 576/6470 [09:22<1:06:28,  1.48it/s]
1|577|Loss: 0.9717691540718079:   9%|▉         | 577/6470 [09:23<1:05:20,  1.50it/s]
1|578|Loss: 1.1470011472702026:   9%|▉         | 577/6470 [09:23<1:05:20,  1.50it/s]
1|578|Loss: 1.1470011472702026:   9%|▉         | 578/6470 [09:23<1:04:09,  1.53it/s]
1|579|Loss: 1.1897389888763428:   9%|▉         | 578/6470 [09:24<1:04:09,  1.53it/s]
1|579|Loss: 1.1897389888763428:   9%|▉         | 579/6470 [09:24<1:08:44,  1.43it/s]
1|580|Loss: 1.1237685680389404:   9%|▉         | 579/6470 [09:24<1:08:44,  1.43it/s]
1|580|Loss: 1.1237685680389404:   9%|▉         | 580/6470 [09:25<1:07:55,  1.45it/s]
1|581|Loss: 1.0613223314285278:   9%|▉         | 580/6470 [09:25<1:07:55,  1.45it/s]
1|581|Loss: 1.0613223314285278:   9%|▉         | 581/6470 [09:25<1:06:10,  1.48it/s]
1|582|Loss: 1.0544039011001587:   9%|▉         | 581/6470 [09:26<1:06:10,  1.48it/s]
1|582|Loss: 1.0544039011001587:   9%|▉         | 582/6470 [09:26<1:05:22,  1.50it/s]
1|583|Loss: 1.171762466430664:   9%|▉         | 582/6470 [09:26<1:05:22,  1.50it/s] 
1|583|Loss: 1.171762466430664:   9%|▉         | 583/6470 [09:27<1:05:58,  1.49it/s]
1|584|Loss: 1.136134386062622:   9%|▉         | 583/6470 [09:27<1:05:58,  1.49it/s]
1|584|Loss: 1.136134386062622:   9%|▉         | 584/6470 [09:27<1:07:45,  1.45it/s]
1|585|Loss: 1.1822131872177124:   9%|▉         | 584/6470 [09:28<1:07:45,  1.45it/s]
1|585|Loss: 1.1822131872177124:   9%|▉         | 585/6470 [09:28<1:09:34,  1.41it/s]
1|586|Loss: 1.1400004625320435:   9%|▉         | 585/6470 [09:29<1:09:34,  1.41it/s]
1|586|Loss: 1.1400004625320435:   9%|▉         | 586/6470 [09:29<1:11:50,  1.36it/s]
1|587|Loss: 1.1134916543960571:   9%|▉         | 586/6470 [09:29<1:11:50,  1.36it/s]
1|587|Loss: 1.1134916543960571:   9%|▉         | 587/6470 [09:30<1:12:20,  1.36it/s]
1|588|Loss: 0.9890490174293518:   9%|▉         | 587/6470 [09:30<1:12:20,  1.36it/s]
1|588|Loss: 0.9890490174293518:   9%|▉         | 588/6470 [09:31<1:15:47,  1.29it/s]
1|589|Loss: 1.0857549905776978:   9%|▉         | 588/6470 [09:31<1:15:47,  1.29it/s]
1|589|Loss: 1.0857549905776978:   9%|▉         | 589/6470 [09:31<1:15:47,  1.29it/s]
1|590|Loss: 1.1528085470199585:   9%|▉         | 589/6470 [09:32<1:15:47,  1.29it/s]
1|590|Loss: 1.1528085470199585:   9%|▉         | 590/6470 [09:32<1:14:39,  1.31it/s]
1|591|Loss: 1.0767099857330322:   9%|▉         | 590/6470 [09:33<1:14:39,  1.31it/s]
1|591|Loss: 1.0767099857330322:   9%|▉         | 591/6470 [09:33<1:14:12,  1.32it/s]
1|592|Loss: 1.1586835384368896:   9%|▉         | 591/6470 [09:33<1:14:12,  1.32it/s]
1|592|Loss: 1.1586835384368896:   9%|▉         | 592/6470 [09:34<1:15:00,  1.31it/s]
1|593|Loss: 1.1967973709106445:   9%|▉         | 592/6470 [09:34<1:15:00,  1.31it/s]
1|593|Loss: 1.1967973709106445:   9%|▉         | 593/6470 [09:34<1:14:51,  1.31it/s]
1|594|Loss: 1.190259337425232:   9%|▉         | 593/6470 [09:35<1:14:51,  1.31it/s] 
1|594|Loss: 1.190259337425232:   9%|▉         | 594/6470 [09:35<1:11:32,  1.37it/s]
1|595|Loss: 1.0811126232147217:   9%|▉         | 594/6470 [09:35<1:11:32,  1.37it/s]
1|595|Loss: 1.0811126232147217:   9%|▉         | 595/6470 [09:36<1:08:34,  1.43it/s]
1|596|Loss: 0.9484679102897644:   9%|▉         | 595/6470 [09:36<1:08:34,  1.43it/s]
1|596|Loss: 0.9484679102897644:   9%|▉         | 596/6470 [09:36<1:08:59,  1.42it/s]
1|597|Loss: 1.0742805004119873:   9%|▉         | 596/6470 [09:37<1:08:59,  1.42it/s]
1|597|Loss: 1.0742805004119873:   9%|▉         | 597/6470 [09:37<1:05:34,  1.49it/s]
1|598|Loss: 1.1676361560821533:   9%|▉         | 597/6470 [09:37<1:05:34,  1.49it/s]
1|598|Loss: 1.1676361560821533:   9%|▉         | 598/6470 [09:38<1:04:21,  1.52it/s]
1|599|Loss: 0.9131771326065063:   9%|▉         | 598/6470 [09:38<1:04:21,  1.52it/s]
1|599|Loss: 0.9131771326065063:   9%|▉         | 599/6470 [09:38<1:04:21,  1.52it/s]
1|600|Loss: 1.2024353742599487:   9%|▉         | 599/6470 [09:39<1:04:21,  1.52it/s]
1|600|Loss: 1.2024353742599487:   9%|▉         | 600/6470 [09:39<1:01:49,  1.58it/s]
1|601|Loss: 1.228110671043396:   9%|▉         | 600/6470 [09:39<1:01:49,  1.58it/s] 
1|601|Loss: 1.228110671043396:   9%|▉         | 601/6470 [09:40<1:03:40,  1.54it/s]
1|602|Loss: 0.9464154243469238:   9%|▉         | 601/6470 [09:40<1:03:40,  1.54it/s]
1|602|Loss: 0.9464154243469238:   9%|▉         | 602/6470 [09:40<1:08:58,  1.42it/s]
1|603|Loss: 1.089167594909668:   9%|▉         | 602/6470 [09:41<1:08:58,  1.42it/s] 
1|603|Loss: 1.089167594909668:   9%|▉         | 603/6470 [09:41<1:13:23,  1.33it/s]
1|604|Loss: 1.266722321510315:   9%|▉         | 603/6470 [09:42<1:13:23,  1.33it/s]
1|604|Loss: 1.266722321510315:   9%|▉         | 604/6470 [09:42<1:11:32,  1.37it/s]
1|605|Loss: 1.0595539808273315:   9%|▉         | 604/6470 [09:42<1:11:32,  1.37it/s]
1|605|Loss: 1.0595539808273315:   9%|▉         | 605/6470 [09:43<1:11:34,  1.37it/s]
1|606|Loss: 1.1438761949539185:   9%|▉         | 605/6470 [09:43<1:11:34,  1.37it/s]
1|606|Loss: 1.1438761949539185:   9%|▉         | 606/6470 [09:43<1:11:49,  1.36it/s]
1|607|Loss: 1.1598361730575562:   9%|▉         | 606/6470 [09:44<1:11:49,  1.36it/s]
1|607|Loss: 1.1598361730575562:   9%|▉         | 607/6470 [09:44<1:09:47,  1.40it/s]
1|608|Loss: 1.1334850788116455:   9%|▉         | 607/6470 [09:44<1:09:47,  1.40it/s]
1|608|Loss: 1.1334850788116455:   9%|▉         | 608/6470 [09:45<1:09:02,  1.41it/s]
1|609|Loss: 1.2593058347702026:   9%|▉         | 608/6470 [09:45<1:09:02,  1.41it/s]
1|609|Loss: 1.2593058347702026:   9%|▉         | 609/6470 [09:45<1:06:11,  1.48it/s]
1|610|Loss: 1.0619786977767944:   9%|▉         | 609/6470 [09:46<1:06:11,  1.48it/s]
1|610|Loss: 1.0619786977767944:   9%|▉         | 610/6470 [09:46<1:09:47,  1.40it/s]
1|611|Loss: 1.21816885471344:   9%|▉         | 610/6470 [09:47<1:09:47,  1.40it/s]  
1|611|Loss: 1.21816885471344:   9%|▉         | 611/6470 [09:47<1:10:29,  1.39it/s]
1|612|Loss: 0.9021587371826172:   9%|▉         | 611/6470 [09:47<1:10:29,  1.39it/s]
1|612|Loss: 0.9021587371826172:   9%|▉         | 612/6470 [09:48<1:10:37,  1.38it/s]
1|613|Loss: 1.0593279600143433:   9%|▉         | 612/6470 [09:48<1:10:37,  1.38it/s]
1|613|Loss: 1.0593279600143433:   9%|▉         | 613/6470 [09:48<1:13:14,  1.33it/s]
1|614|Loss: 1.1399427652359009:   9%|▉         | 613/6470 [09:49<1:13:14,  1.33it/s]
1|614|Loss: 1.1399427652359009:   9%|▉         | 614/6470 [09:49<1:13:21,  1.33it/s]
1|615|Loss: 1.2006806135177612:   9%|▉         | 614/6470 [09:50<1:13:21,  1.33it/s]
1|615|Loss: 1.2006806135177612:  10%|▉         | 615/6470 [09:50<1:10:34,  1.38it/s]
1|616|Loss: 0.9953798055648804:  10%|▉         | 615/6470 [09:50<1:10:34,  1.38it/s]
1|616|Loss: 0.9953798055648804:  10%|▉         | 616/6470 [09:51<1:10:54,  1.38it/s]
1|617|Loss: 1.3242225646972656:  10%|▉         | 616/6470 [09:51<1:10:54,  1.38it/s]
1|617|Loss: 1.3242225646972656:  10%|▉         | 617/6470 [09:51<1:07:02,  1.46it/s]
1|618|Loss: 1.0693385601043701:  10%|▉         | 617/6470 [09:52<1:07:02,  1.46it/s]
1|618|Loss: 1.0693385601043701:  10%|▉         | 618/6470 [09:52<1:04:32,  1.51it/s]
1|619|Loss: 1.1189814805984497:  10%|▉         | 618/6470 [09:52<1:04:32,  1.51it/s]
1|619|Loss: 1.1189814805984497:  10%|▉         | 619/6470 [09:52<1:03:06,  1.55it/s]
1|620|Loss: 1.0718653202056885:  10%|▉         | 619/6470 [09:53<1:03:06,  1.55it/s]
1|620|Loss: 1.0718653202056885:  10%|▉         | 620/6470 [09:53<1:06:14,  1.47it/s]
1|621|Loss: 1.0543354749679565:  10%|▉         | 620/6470 [09:54<1:06:14,  1.47it/s]
1|621|Loss: 1.0543354749679565:  10%|▉         | 621/6470 [09:54<1:10:35,  1.38it/s]
1|622|Loss: 1.0377856492996216:  10%|▉         | 621/6470 [09:54<1:10:35,  1.38it/s]
1|622|Loss: 1.0377856492996216:  10%|▉         | 622/6470 [09:55<1:09:16,  1.41it/s]
1|623|Loss: 1.3902024030685425:  10%|▉         | 622/6470 [09:55<1:09:16,  1.41it/s]
1|623|Loss: 1.3902024030685425:  10%|▉         | 623/6470 [09:55<1:03:30,  1.53it/s]
1|624|Loss: 1.3117750883102417:  10%|▉         | 623/6470 [09:56<1:03:30,  1.53it/s]
1|624|Loss: 1.3117750883102417:  10%|▉         | 624/6470 [09:56<1:01:35,  1.58it/s]
1|625|Loss: 1.0625884532928467:  10%|▉         | 624/6470 [09:56<1:01:35,  1.58it/s]
1|625|Loss: 1.0625884532928467:  10%|▉         | 625/6470 [09:56<1:03:07,  1.54it/s]
1|626|Loss: 1.292049765586853:  10%|▉         | 625/6470 [09:57<1:03:07,  1.54it/s] 
1|626|Loss: 1.292049765586853:  10%|▉         | 626/6470 [09:57<1:01:06,  1.59it/s]
1|627|Loss: 1.1513820886611938:  10%|▉         | 626/6470 [09:57<1:01:06,  1.59it/s]
1|627|Loss: 1.1513820886611938:  10%|▉         | 627/6470 [09:58<1:03:31,  1.53it/s]
1|628|Loss: 1.1119765043258667:  10%|▉         | 627/6470 [09:58<1:03:31,  1.53it/s]
1|628|Loss: 1.1119765043258667:  10%|▉         | 628/6470 [09:58<1:06:27,  1.47it/s]
1|629|Loss: 1.1923441886901855:  10%|▉         | 628/6470 [09:59<1:06:27,  1.47it/s]
1|629|Loss: 1.1923441886901855:  10%|▉         | 629/6470 [09:59<1:05:22,  1.49it/s]
1|630|Loss: 1.085185170173645:  10%|▉         | 629/6470 [10:00<1:05:22,  1.49it/s] 
1|630|Loss: 1.085185170173645:  10%|▉         | 630/6470 [10:00<1:05:31,  1.49it/s]
1|631|Loss: 1.002663493156433:  10%|▉         | 630/6470 [10:00<1:05:31,  1.49it/s]
1|631|Loss: 1.002663493156433:  10%|▉         | 631/6470 [10:01<1:09:53,  1.39it/s]
1|632|Loss: 1.2233359813690186:  10%|▉         | 631/6470 [10:01<1:09:53,  1.39it/s]
1|632|Loss: 1.2233359813690186:  10%|▉         | 632/6470 [10:01<1:07:50,  1.43it/s]
1|633|Loss: 1.0255742073059082:  10%|▉         | 632/6470 [10:02<1:07:50,  1.43it/s]
1|633|Loss: 1.0255742073059082:  10%|▉         | 633/6470 [10:02<1:06:58,  1.45it/s]
1|634|Loss: 1.1972098350524902:  10%|▉         | 633/6470 [10:02<1:06:58,  1.45it/s]
1|634|Loss: 1.1972098350524902:  10%|▉         | 634/6470 [10:03<1:05:52,  1.48it/s]
1|635|Loss: 1.1946887969970703:  10%|▉         | 634/6470 [10:03<1:05:52,  1.48it/s]
1|635|Loss: 1.1946887969970703:  10%|▉         | 635/6470 [10:03<1:09:16,  1.40it/s]
1|636|Loss: 1.1409558057785034:  10%|▉         | 635/6470 [10:04<1:09:16,  1.40it/s]
1|636|Loss: 1.1409558057785034:  10%|▉         | 636/6470 [10:04<1:05:22,  1.49it/s]
1|637|Loss: 1.184794306755066:  10%|▉         | 636/6470 [10:04<1:05:22,  1.49it/s] 
1|637|Loss: 1.184794306755066:  10%|▉         | 637/6470 [10:05<1:06:01,  1.47it/s]
1|638|Loss: 1.1869792938232422:  10%|▉         | 637/6470 [10:05<1:06:01,  1.47it/s]
1|638|Loss: 1.1869792938232422:  10%|▉         | 638/6470 [10:05<1:04:08,  1.52it/s]
1|639|Loss: 1.2154253721237183:  10%|▉         | 638/6470 [10:06<1:04:08,  1.52it/s]
1|639|Loss: 1.2154253721237183:  10%|▉         | 639/6470 [10:06<1:08:20,  1.42it/s]
1|640|Loss: 1.2434566020965576:  10%|▉         | 639/6470 [10:07<1:08:20,  1.42it/s]
1|640|Loss: 1.2434566020965576:  10%|▉         | 640/6470 [10:07<1:07:48,  1.43it/s]
1|641|Loss: 1.1204988956451416:  10%|▉         | 640/6470 [10:07<1:07:48,  1.43it/s]
1|641|Loss: 1.1204988956451416:  10%|▉         | 641/6470 [10:08<1:09:50,  1.39it/s]
1|642|Loss: 1.129950761795044:  10%|▉         | 641/6470 [10:08<1:09:50,  1.39it/s] 
1|642|Loss: 1.129950761795044:  10%|▉         | 642/6470 [10:08<1:07:53,  1.43it/s]
1|643|Loss: 1.1545569896697998:  10%|▉         | 642/6470 [10:09<1:07:53,  1.43it/s]
1|643|Loss: 1.1545569896697998:  10%|▉         | 643/6470 [10:09<1:05:18,  1.49it/s]
1|644|Loss: 1.256805658340454:  10%|▉         | 643/6470 [10:09<1:05:18,  1.49it/s] 
1|644|Loss: 1.256805658340454:  10%|▉         | 644/6470 [10:09<1:02:33,  1.55it/s]
1|645|Loss: 1.044122338294983:  10%|▉         | 644/6470 [10:10<1:02:33,  1.55it/s]
1|645|Loss: 1.044122338294983:  10%|▉         | 645/6470 [10:10<1:06:43,  1.46it/s]
1|646|Loss: 1.0549626350402832:  10%|▉         | 645/6470 [10:11<1:06:43,  1.46it/s]
1|646|Loss: 1.0549626350402832:  10%|▉         | 646/6470 [10:11<1:05:19,  1.49it/s]
1|647|Loss: 1.25277841091156:  10%|▉         | 646/6470 [10:11<1:05:19,  1.49it/s]  
1|647|Loss: 1.25277841091156:  10%|█         | 647/6470 [10:11<1:03:24,  1.53it/s]
1|648|Loss: 1.193165898323059:  10%|█         | 647/6470 [10:12<1:03:24,  1.53it/s]
1|648|Loss: 1.193165898323059:  10%|█         | 648/6470 [10:12<1:04:15,  1.51it/s]
1|649|Loss: 1.3103605508804321:  10%|█         | 648/6470 [10:13<1:04:15,  1.51it/s]
1|649|Loss: 1.3103605508804321:  10%|█         | 649/6470 [10:13<1:05:56,  1.47it/s]
1|650|Loss: 1.1938669681549072:  10%|█         | 649/6470 [10:13<1:05:56,  1.47it/s]
1|650|Loss: 1.1938669681549072:  10%|█         | 650/6470 [10:14<1:06:35,  1.46it/s]
1|651|Loss: 1.125494360923767:  10%|█         | 650/6470 [10:14<1:06:35,  1.46it/s] 
1|651|Loss: 1.125494360923767:  10%|█         | 651/6470 [10:14<1:07:00,  1.45it/s]
1|652|Loss: 0.9973143339157104:  10%|█         | 651/6470 [10:15<1:07:00,  1.45it/s]
1|652|Loss: 0.9973143339157104:  10%|█         | 652/6470 [10:15<1:09:40,  1.39it/s]
1|653|Loss: 1.2276419401168823:  10%|█         | 652/6470 [10:15<1:09:40,  1.39it/s]
1|653|Loss: 1.2276419401168823:  10%|█         | 653/6470 [10:16<1:11:53,  1.35it/s]
1|654|Loss: 1.249210238456726:  10%|█         | 653/6470 [10:16<1:11:53,  1.35it/s] 
1|654|Loss: 1.249210238456726:  10%|█         | 654/6470 [10:16<1:08:28,  1.42it/s]
1|655|Loss: 1.0727806091308594:  10%|█         | 654/6470 [10:17<1:08:28,  1.42it/s]
1|655|Loss: 1.0727806091308594:  10%|█         | 655/6470 [10:17<1:12:00,  1.35it/s]
1|656|Loss: 1.2785555124282837:  10%|█         | 655/6470 [10:18<1:12:00,  1.35it/s]
1|656|Loss: 1.2785555124282837:  10%|█         | 656/6470 [10:18<1:10:09,  1.38it/s]
1|657|Loss: 0.9539545774459839:  10%|█         | 656/6470 [10:18<1:10:09,  1.38it/s]
1|657|Loss: 0.9539545774459839:  10%|█         | 657/6470 [10:19<1:12:58,  1.33it/s]
1|658|Loss: 1.1852043867111206:  10%|█         | 657/6470 [10:19<1:12:58,  1.33it/s]
1|658|Loss: 1.1852043867111206:  10%|█         | 658/6470 [10:19<1:10:49,  1.37it/s]
1|659|Loss: 1.29999840259552:  10%|█         | 658/6470 [10:20<1:10:49,  1.37it/s]  
1|659|Loss: 1.29999840259552:  10%|█         | 659/6470 [10:20<1:06:24,  1.46it/s]
1|660|Loss: 1.2740048170089722:  10%|█         | 659/6470 [10:20<1:06:24,  1.46it/s]
1|660|Loss: 1.2740048170089722:  10%|█         | 660/6470 [10:21<1:07:01,  1.44it/s]
1|661|Loss: 1.0513477325439453:  10%|█         | 660/6470 [10:21<1:07:01,  1.44it/s]
1|661|Loss: 1.0513477325439453:  10%|█         | 661/6470 [10:21<1:07:41,  1.43it/s]
1|662|Loss: 1.1259392499923706:  10%|█         | 661/6470 [10:22<1:07:41,  1.43it/s]
1|662|Loss: 1.1259392499923706:  10%|█         | 662/6470 [10:22<1:05:57,  1.47it/s]
1|663|Loss: 1.0879566669464111:  10%|█         | 662/6470 [10:22<1:05:57,  1.47it/s]
1|663|Loss: 1.0879566669464111:  10%|█         | 663/6470 [10:23<1:03:09,  1.53it/s]
1|664|Loss: 1.023406982421875:  10%|█         | 663/6470 [10:23<1:03:09,  1.53it/s] 
1|664|Loss: 1.023406982421875:  10%|█         | 664/6470 [10:23<1:03:51,  1.52it/s]
1|665|Loss: 1.1319814920425415:  10%|█         | 664/6470 [10:24<1:03:51,  1.52it/s]
1|665|Loss: 1.1319814920425415:  10%|█         | 665/6470 [10:24<1:04:58,  1.49it/s]
1|666|Loss: 1.1193220615386963:  10%|█         | 665/6470 [10:24<1:04:58,  1.49it/s]
1|666|Loss: 1.1193220615386963:  10%|█         | 666/6470 [10:25<1:01:39,  1.57it/s]
1|667|Loss: 0.9994519352912903:  10%|█         | 666/6470 [10:25<1:01:39,  1.57it/s]
1|667|Loss: 0.9994519352912903:  10%|█         | 667/6470 [10:25<1:03:29,  1.52it/s]
1|668|Loss: 1.0172537565231323:  10%|█         | 667/6470 [10:26<1:03:29,  1.52it/s]
1|668|Loss: 1.0172537565231323:  10%|█         | 668/6470 [10:26<1:07:31,  1.43it/s]
1|669|Loss: 0.9781010150909424:  10%|█         | 668/6470 [10:27<1:07:31,  1.43it/s]
1|669|Loss: 0.9781010150909424:  10%|█         | 669/6470 [10:27<1:09:06,  1.40it/s]
1|670|Loss: 0.9778783917427063:  10%|█         | 669/6470 [10:27<1:09:06,  1.40it/s]
1|670|Loss: 0.9778783917427063:  10%|█         | 670/6470 [10:27<1:06:59,  1.44it/s]
1|671|Loss: 1.0775612592697144:  10%|█         | 670/6470 [10:28<1:06:59,  1.44it/s]
1|671|Loss: 1.0775612592697144:  10%|█         | 671/6470 [10:28<1:06:54,  1.44it/s]
1|672|Loss: 1.1191529035568237:  10%|█         | 671/6470 [10:29<1:06:54,  1.44it/s]
1|672|Loss: 1.1191529035568237:  10%|█         | 672/6470 [10:29<1:02:47,  1.54it/s]
1|673|Loss: 1.1343330144882202:  10%|█         | 672/6470 [10:29<1:02:47,  1.54it/s]
1|673|Loss: 1.1343330144882202:  10%|█         | 673/6470 [10:29<1:01:51,  1.56it/s]
1|674|Loss: 1.060062050819397:  10%|█         | 673/6470 [10:30<1:01:51,  1.56it/s] 
1|674|Loss: 1.060062050819397:  10%|█         | 674/6470 [10:30<1:02:06,  1.56it/s]
1|675|Loss: 1.4060571193695068:  10%|█         | 674/6470 [10:30<1:02:06,  1.56it/s]
1|675|Loss: 1.4060571193695068:  10%|█         | 675/6470 [10:31<1:03:36,  1.52it/s]
1|676|Loss: 0.961757242679596:  10%|█         | 675/6470 [10:31<1:03:36,  1.52it/s] 
1|676|Loss: 0.961757242679596:  10%|█         | 676/6470 [10:31<1:02:54,  1.54it/s]
1|677|Loss: 0.8749459385871887:  10%|█         | 676/6470 [10:32<1:02:54,  1.54it/s]
1|677|Loss: 0.8749459385871887:  10%|█         | 677/6470 [10:32<1:02:19,  1.55it/s]
1|678|Loss: 1.143034815788269:  10%|█         | 677/6470 [10:32<1:02:19,  1.55it/s] 
1|678|Loss: 1.143034815788269:  10%|█         | 678/6470 [10:33<1:04:13,  1.50it/s]
1|679|Loss: 1.0072534084320068:  10%|█         | 678/6470 [10:33<1:04:13,  1.50it/s]
1|679|Loss: 1.0072534084320068:  10%|█         | 679/6470 [10:34<1:10:21,  1.37it/s]
1|680|Loss: 1.5619356632232666:  10%|█         | 679/6470 [10:34<1:10:21,  1.37it/s]
1|680|Loss: 1.5619356632232666:  11%|█         | 680/6470 [10:34<1:06:51,  1.44it/s]
1|681|Loss: 1.0101277828216553:  11%|█         | 680/6470 [10:35<1:06:51,  1.44it/s]
1|681|Loss: 1.0101277828216553:  11%|█         | 681/6470 [10:35<1:06:02,  1.46it/s]
1|682|Loss: 1.1380811929702759:  11%|█         | 681/6470 [10:35<1:06:02,  1.46it/s]
1|682|Loss: 1.1380811929702759:  11%|█         | 682/6470 [10:36<1:09:59,  1.38it/s]
1|683|Loss: 1.1540786027908325:  11%|█         | 682/6470 [10:36<1:09:59,  1.38it/s]
1|683|Loss: 1.1540786027908325:  11%|█         | 683/6470 [10:36<1:11:00,  1.36it/s]
1|684|Loss: 1.3904603719711304:  11%|█         | 683/6470 [10:37<1:11:00,  1.36it/s]
1|684|Loss: 1.3904603719711304:  11%|█         | 684/6470 [10:37<1:08:04,  1.42it/s]
1|685|Loss: 0.9457705616950989:  11%|█         | 684/6470 [10:37<1:08:04,  1.42it/s]
1|685|Loss: 0.9457705616950989:  11%|█         | 685/6470 [10:38<1:11:25,  1.35it/s]
1|686|Loss: 0.8964731097221375:  11%|█         | 685/6470 [10:38<1:11:25,  1.35it/s]
1|686|Loss: 0.8964731097221375:  11%|█         | 686/6470 [10:39<1:11:27,  1.35it/s]
1|687|Loss: 1.218815565109253:  11%|█         | 686/6470 [10:39<1:11:27,  1.35it/s] 
1|687|Loss: 1.218815565109253:  11%|█         | 687/6470 [10:39<1:08:53,  1.40it/s]
1|688|Loss: 1.0066821575164795:  11%|█         | 687/6470 [10:40<1:08:53,  1.40it/s]
1|688|Loss: 1.0066821575164795:  11%|█         | 688/6470 [10:40<1:09:13,  1.39it/s]
1|689|Loss: 1.2835619449615479:  11%|█         | 688/6470 [10:40<1:09:13,  1.39it/s]
1|689|Loss: 1.2835619449615479:  11%|█         | 689/6470 [10:41<1:06:25,  1.45it/s]
1|690|Loss: 1.1035375595092773:  11%|█         | 689/6470 [10:41<1:06:25,  1.45it/s]
1|690|Loss: 1.1035375595092773:  11%|█         | 690/6470 [10:41<1:09:48,  1.38it/s]
1|691|Loss: 1.0049740076065063:  11%|█         | 690/6470 [10:42<1:09:48,  1.38it/s]
1|691|Loss: 1.0049740076065063:  11%|█         | 691/6470 [10:42<1:10:12,  1.37it/s]
1|692|Loss: 1.002964735031128:  11%|█         | 691/6470 [10:43<1:10:12,  1.37it/s] 
1|692|Loss: 1.002964735031128:  11%|█         | 692/6470 [10:43<1:10:14,  1.37it/s]
1|693|Loss: 1.2018330097198486:  11%|█         | 692/6470 [10:43<1:10:14,  1.37it/s]
1|693|Loss: 1.2018330097198486:  11%|█         | 693/6470 [10:44<1:06:48,  1.44it/s]
1|694|Loss: 1.1418278217315674:  11%|█         | 693/6470 [10:44<1:06:48,  1.44it/s]
1|694|Loss: 1.1418278217315674:  11%|█         | 694/6470 [10:44<1:05:01,  1.48it/s]
1|695|Loss: 1.1718525886535645:  11%|█         | 694/6470 [10:44<1:05:01,  1.48it/s]
1|695|Loss: 1.1718525886535645:  11%|█         | 695/6470 [10:45<1:01:02,  1.58it/s]
1|696|Loss: 1.0871695280075073:  11%|█         | 695/6470 [10:45<1:01:02,  1.58it/s]
1|696|Loss: 1.0871695280075073:  11%|█         | 696/6470 [10:45<1:05:39,  1.47it/s]
1|697|Loss: 1.334412932395935:  11%|█         | 696/6470 [10:46<1:05:39,  1.47it/s] 
1|697|Loss: 1.334412932395935:  11%|█         | 697/6470 [10:46<1:04:40,  1.49it/s]
1|698|Loss: 1.0566445589065552:  11%|█         | 697/6470 [10:46<1:04:40,  1.49it/s]
1|698|Loss: 1.0566445589065552:  11%|█         | 698/6470 [10:47<1:02:46,  1.53it/s]
1|699|Loss: 1.0504440069198608:  11%|█         | 698/6470 [10:47<1:02:46,  1.53it/s]
1|699|Loss: 1.0504440069198608:  11%|█         | 699/6470 [10:48<1:07:44,  1.42it/s]
1|700|Loss: 1.2549618482589722:  11%|█         | 699/6470 [10:48<1:07:44,  1.42it/s]
1|700|Loss: 1.2549618482589722:  11%|█         | 700/6470 [10:48<1:03:10,  1.52it/s]
1|701|Loss: 1.1487045288085938:  11%|█         | 700/6470 [10:48<1:03:10,  1.52it/s]
1|701|Loss: 1.1487045288085938:  11%|█         | 701/6470 [10:49<1:05:40,  1.46it/s]
1|702|Loss: 1.0785202980041504:  11%|█         | 701/6470 [10:49<1:05:40,  1.46it/s]
1|702|Loss: 1.0785202980041504:  11%|█         | 702/6470 [10:50<1:11:54,  1.34it/s]
1|703|Loss: 1.3155895471572876:  11%|█         | 702/6470 [10:50<1:11:54,  1.34it/s]
1|703|Loss: 1.3155895471572876:  11%|█         | 703/6470 [10:50<1:09:48,  1.38it/s]
1|704|Loss: 1.0795619487762451:  11%|█         | 703/6470 [10:51<1:09:48,  1.38it/s]
1|704|Loss: 1.0795619487762451:  11%|█         | 704/6470 [10:51<1:09:03,  1.39it/s]
1|705|Loss: 1.203439474105835:  11%|█         | 704/6470 [10:52<1:09:03,  1.39it/s] 
1|705|Loss: 1.203439474105835:  11%|█         | 705/6470 [10:52<1:09:25,  1.38it/s]
1|706|Loss: 1.0852973461151123:  11%|█         | 705/6470 [10:52<1:09:25,  1.38it/s]
1|706|Loss: 1.0852973461151123:  11%|█         | 706/6470 [10:52<1:06:32,  1.44it/s]
1|707|Loss: 1.2361741065979004:  11%|█         | 706/6470 [10:53<1:06:32,  1.44it/s]
1|707|Loss: 1.2361741065979004:  11%|█         | 707/6470 [10:53<1:04:31,  1.49it/s]
1|708|Loss: 1.1792900562286377:  11%|█         | 707/6470 [10:53<1:04:31,  1.49it/s]
1|708|Loss: 1.1792900562286377:  11%|█         | 708/6470 [10:54<1:03:43,  1.51it/s]
1|709|Loss: 1.0778586864471436:  11%|█         | 708/6470 [10:54<1:03:43,  1.51it/s]
1|709|Loss: 1.0778586864471436:  11%|█         | 709/6470 [10:54<1:05:39,  1.46it/s]
1|710|Loss: 1.0790326595306396:  11%|█         | 709/6470 [10:55<1:05:39,  1.46it/s]
1|710|Loss: 1.0790326595306396:  11%|█         | 710/6470 [10:55<1:07:17,  1.43it/s]
1|711|Loss: 1.01951003074646:  11%|█         | 710/6470 [10:56<1:07:17,  1.43it/s]  
1|711|Loss: 1.01951003074646:  11%|█         | 711/6470 [10:56<1:10:32,  1.36it/s]
1|712|Loss: 1.0857747793197632:  11%|█         | 711/6470 [10:56<1:10:32,  1.36it/s]
1|712|Loss: 1.0857747793197632:  11%|█         | 712/6470 [10:57<1:14:00,  1.30it/s]
1|713|Loss: 1.0809234380722046:  11%|█         | 712/6470 [10:57<1:14:00,  1.30it/s]
1|713|Loss: 1.0809234380722046:  11%|█         | 713/6470 [10:58<1:11:44,  1.34it/s]
1|714|Loss: 1.039058804512024:  11%|█         | 713/6470 [10:58<1:11:44,  1.34it/s] 
1|714|Loss: 1.039058804512024:  11%|█         | 714/6470 [10:58<1:06:05,  1.45it/s]
1|715|Loss: 1.0789936780929565:  11%|█         | 714/6470 [10:58<1:06:05,  1.45it/s]
1|715|Loss: 1.0789936780929565:  11%|█         | 715/6470 [10:59<1:06:46,  1.44it/s]
1|716|Loss: 1.0945981740951538:  11%|█         | 715/6470 [10:59<1:06:46,  1.44it/s]
1|716|Loss: 1.0945981740951538:  11%|█         | 716/6470 [11:00<1:06:06,  1.45it/s]
1|717|Loss: 1.1160200834274292:  11%|█         | 716/6470 [11:00<1:06:06,  1.45it/s]
1|717|Loss: 1.1160200834274292:  11%|█         | 717/6470 [11:00<1:06:55,  1.43it/s]
1|718|Loss: 1.366604208946228:  11%|█         | 717/6470 [11:01<1:06:55,  1.43it/s] 
1|718|Loss: 1.366604208946228:  11%|█         | 718/6470 [11:01<1:02:46,  1.53it/s]
1|719|Loss: 1.182383418083191:  11%|█         | 718/6470 [11:01<1:02:46,  1.53it/s]
1|719|Loss: 1.182383418083191:  11%|█         | 719/6470 [11:01<1:03:43,  1.50it/s]
1|720|Loss: 1.0206797122955322:  11%|█         | 719/6470 [11:02<1:03:43,  1.50it/s]
1|720|Loss: 1.0206797122955322:  11%|█         | 720/6470 [11:02<1:05:44,  1.46it/s]
1|721|Loss: 1.0586133003234863:  11%|█         | 720/6470 [11:03<1:05:44,  1.46it/s]
1|721|Loss: 1.0586133003234863:  11%|█         | 721/6470 [11:03<1:04:55,  1.48it/s]
1|722|Loss: 1.2386404275894165:  11%|█         | 721/6470 [11:03<1:04:55,  1.48it/s]
1|722|Loss: 1.2386404275894165:  11%|█         | 722/6470 [11:03<1:01:43,  1.55it/s]
1|723|Loss: 1.224717617034912:  11%|█         | 722/6470 [11:04<1:01:43,  1.55it/s] 
1|723|Loss: 1.224717617034912:  11%|█         | 723/6470 [11:04<1:02:34,  1.53it/s]
1|724|Loss: 1.1928021907806396:  11%|█         | 723/6470 [11:05<1:02:34,  1.53it/s]
1|724|Loss: 1.1928021907806396:  11%|█         | 724/6470 [11:05<1:05:50,  1.45it/s]
1|725|Loss: 1.053471326828003:  11%|█         | 724/6470 [11:05<1:05:50,  1.45it/s] 
1|725|Loss: 1.053471326828003:  11%|█         | 725/6470 [11:06<1:08:03,  1.41it/s]
1|726|Loss: 1.1775856018066406:  11%|█         | 725/6470 [11:06<1:08:03,  1.41it/s]
1|726|Loss: 1.1775856018066406:  11%|█         | 726/6470 [11:06<1:03:51,  1.50it/s]
1|727|Loss: 1.278551697731018:  11%|█         | 726/6470 [11:07<1:03:51,  1.50it/s] 
1|727|Loss: 1.278551697731018:  11%|█         | 727/6470 [11:07<1:00:12,  1.59it/s]
1|728|Loss: 1.3129513263702393:  11%|█         | 727/6470 [11:07<1:00:12,  1.59it/s]
1|728|Loss: 1.3129513263702393:  11%|█▏        | 728/6470 [11:07<58:32,  1.63it/s]  
1|729|Loss: 0.9763526320457458:  11%|█▏        | 728/6470 [11:08<58:32,  1.63it/s]
1|729|Loss: 0.9763526320457458:  11%|█▏        | 729/6470 [11:08<57:03,  1.68it/s]
1|730|Loss: 1.2794743776321411:  11%|█▏        | 729/6470 [11:08<57:03,  1.68it/s]
1|730|Loss: 1.2794743776321411:  11%|█▏        | 730/6470 [11:09<1:00:12,  1.59it/s]
1|731|Loss: 1.0831624269485474:  11%|█▏        | 730/6470 [11:09<1:00:12,  1.59it/s]
1|731|Loss: 1.0831624269485474:  11%|█▏        | 731/6470 [11:09<1:00:41,  1.58it/s]
1|732|Loss: 1.0295460224151611:  11%|█▏        | 731/6470 [11:10<1:00:41,  1.58it/s]
1|732|Loss: 1.0295460224151611:  11%|█▏        | 732/6470 [11:10<1:00:05,  1.59it/s]
1|733|Loss: 1.1258317232131958:  11%|█▏        | 732/6470 [11:10<1:00:05,  1.59it/s]
1|733|Loss: 1.1258317232131958:  11%|█▏        | 733/6470 [11:11<1:02:27,  1.53it/s]
1|734|Loss: 1.2211556434631348:  11%|█▏        | 733/6470 [11:11<1:02:27,  1.53it/s]
1|734|Loss: 1.2211556434631348:  11%|█▏        | 734/6470 [11:11<1:06:30,  1.44it/s]
1|735|Loss: 0.9938421249389648:  11%|█▏        | 734/6470 [11:12<1:06:30,  1.44it/s]
1|735|Loss: 0.9938421249389648:  11%|█▏        | 735/6470 [11:12<1:08:38,  1.39it/s]
1|736|Loss: 1.2769322395324707:  11%|█▏        | 735/6470 [11:12<1:08:38,  1.39it/s]
1|736|Loss: 1.2769322395324707:  11%|█▏        | 736/6470 [11:13<1:05:10,  1.47it/s]
1|737|Loss: 0.9793127775192261:  11%|█▏        | 736/6470 [11:13<1:05:10,  1.47it/s]
1|737|Loss: 0.9793127775192261:  11%|█▏        | 737/6470 [11:13<1:02:19,  1.53it/s]
1|738|Loss: 1.0859978199005127:  11%|█▏        | 737/6470 [11:14<1:02:19,  1.53it/s]
1|738|Loss: 1.0859978199005127:  11%|█▏        | 738/6470 [11:14<1:04:55,  1.47it/s]
1|739|Loss: 1.1598137617111206:  11%|█▏        | 738/6470 [11:14<1:04:55,  1.47it/s]
1|739|Loss: 1.1598137617111206:  11%|█▏        | 739/6470 [11:15<1:06:53,  1.43it/s]
1|740|Loss: 1.0059925317764282:  11%|█▏        | 739/6470 [11:15<1:06:53,  1.43it/s]
1|740|Loss: 1.0059925317764282:  11%|█▏        | 740/6470 [11:15<1:05:46,  1.45it/s]
1|741|Loss: 1.0808545351028442:  11%|█▏        | 740/6470 [11:16<1:05:46,  1.45it/s]
1|741|Loss: 1.0808545351028442:  11%|█▏        | 741/6470 [11:16<1:09:18,  1.38it/s]
1|742|Loss: 0.9560370445251465:  11%|█▏        | 741/6470 [11:17<1:09:18,  1.38it/s]
1|742|Loss: 0.9560370445251465:  11%|█▏        | 742/6470 [11:17<1:10:26,  1.36it/s]
1|743|Loss: 1.1314889192581177:  11%|█▏        | 742/6470 [11:17<1:10:26,  1.36it/s]
1|743|Loss: 1.1314889192581177:  11%|█▏        | 743/6470 [11:18<1:03:56,  1.49it/s]
1|744|Loss: 0.964620053768158:  11%|█▏        | 743/6470 [11:18<1:03:56,  1.49it/s] 
1|744|Loss: 0.964620053768158:  11%|█▏        | 744/6470 [11:18<1:07:22,  1.42it/s]
1|745|Loss: 1.2689205408096313:  11%|█▏        | 744/6470 [11:19<1:07:22,  1.42it/s]
1|745|Loss: 1.2689205408096313:  12%|█▏        | 745/6470 [11:19<1:07:01,  1.42it/s]
1|746|Loss: 1.045446753501892:  12%|█▏        | 745/6470 [11:19<1:07:01,  1.42it/s] 
1|746|Loss: 1.045446753501892:  12%|█▏        | 746/6470 [11:20<1:05:10,  1.46it/s]
1|747|Loss: 1.1030100584030151:  12%|█▏        | 746/6470 [11:20<1:05:10,  1.46it/s]
1|747|Loss: 1.1030100584030151:  12%|█▏        | 747/6470 [11:20<1:02:32,  1.53it/s]
1|748|Loss: 1.2116612195968628:  12%|█▏        | 747/6470 [11:21<1:02:32,  1.53it/s]
1|748|Loss: 1.2116612195968628:  12%|█▏        | 748/6470 [11:21<1:03:06,  1.51it/s]
1|749|Loss: 1.0200743675231934:  12%|█▏        | 748/6470 [11:21<1:03:06,  1.51it/s]
1|749|Loss: 1.0200743675231934:  12%|█▏        | 749/6470 [11:22<1:04:19,  1.48it/s]
1|750|Loss: 1.0337198972702026:  12%|█▏        | 749/6470 [11:22<1:04:19,  1.48it/s]
1|750|Loss: 1.0337198972702026:  12%|█▏        | 750/6470 [11:22<1:06:08,  1.44it/s]
1|751|Loss: 1.0835601091384888:  12%|█▏        | 750/6470 [11:23<1:06:08,  1.44it/s]
1|751|Loss: 1.0835601091384888:  12%|█▏        | 751/6470 [11:23<1:04:49,  1.47it/s]
1|752|Loss: 1.0155856609344482:  12%|█▏        | 751/6470 [11:23<1:04:49,  1.47it/s]
1|752|Loss: 1.0155856609344482:  12%|█▏        | 752/6470 [11:24<1:07:52,  1.40it/s]
1|753|Loss: 1.0540999174118042:  12%|█▏        | 752/6470 [11:24<1:07:52,  1.40it/s]
1|753|Loss: 1.0540999174118042:  12%|█▏        | 753/6470 [11:25<1:10:35,  1.35it/s]
1|754|Loss: 1.1412550210952759:  12%|█▏        | 753/6470 [11:25<1:10:35,  1.35it/s]
1|754|Loss: 1.1412550210952759:  12%|█▏        | 754/6470 [11:25<1:13:24,  1.30it/s]
1|755|Loss: 0.9396622180938721:  12%|█▏        | 754/6470 [11:26<1:13:24,  1.30it/s]
1|755|Loss: 0.9396622180938721:  12%|█▏        | 755/6470 [11:26<1:11:09,  1.34it/s]
1|756|Loss: 1.1518348455429077:  12%|█▏        | 755/6470 [11:27<1:11:09,  1.34it/s]
1|756|Loss: 1.1518348455429077:  12%|█▏        | 756/6470 [11:27<1:10:32,  1.35it/s]
1|757|Loss: 0.8764671683311462:  12%|█▏        | 756/6470 [11:27<1:10:32,  1.35it/s]
1|757|Loss: 0.8764671683311462:  12%|█▏        | 757/6470 [11:28<1:08:24,  1.39it/s]
1|758|Loss: 1.2347283363342285:  12%|█▏        | 757/6470 [11:28<1:08:24,  1.39it/s]
1|758|Loss: 1.2347283363342285:  12%|█▏        | 758/6470 [11:28<1:07:35,  1.41it/s]
1|759|Loss: 1.1895276308059692:  12%|█▏        | 758/6470 [11:29<1:07:35,  1.41it/s]
1|759|Loss: 1.1895276308059692:  12%|█▏        | 759/6470 [11:29<1:04:32,  1.47it/s]
1|760|Loss: 1.0224566459655762:  12%|█▏        | 759/6470 [11:29<1:04:32,  1.47it/s]
1|760|Loss: 1.0224566459655762:  12%|█▏        | 760/6470 [11:29<1:02:59,  1.51it/s]
1|761|Loss: 1.2527947425842285:  12%|█▏        | 760/6470 [11:30<1:02:59,  1.51it/s]
1|761|Loss: 1.2527947425842285:  12%|█▏        | 761/6470 [11:30<1:03:38,  1.50it/s]
1|762|Loss: 1.113275170326233:  12%|█▏        | 761/6470 [11:31<1:03:38,  1.50it/s] 
1|762|Loss: 1.113275170326233:  12%|█▏        | 762/6470 [11:31<1:03:03,  1.51it/s]
1|763|Loss: 1.0189718008041382:  12%|█▏        | 762/6470 [11:31<1:03:03,  1.51it/s]
1|763|Loss: 1.0189718008041382:  12%|█▏        | 763/6470 [11:32<1:04:54,  1.47it/s]
1|764|Loss: 1.0813807249069214:  12%|█▏        | 763/6470 [11:32<1:04:54,  1.47it/s]
1|764|Loss: 1.0813807249069214:  12%|█▏        | 764/6470 [11:32<1:09:50,  1.36it/s]
1|765|Loss: 0.96345454454422:  12%|█▏        | 764/6470 [11:33<1:09:50,  1.36it/s]  
1|765|Loss: 0.96345454454422:  12%|█▏        | 765/6470 [11:33<1:11:05,  1.34it/s]
1|766|Loss: 0.9793746471405029:  12%|█▏        | 765/6470 [11:34<1:11:05,  1.34it/s]
1|766|Loss: 0.9793746471405029:  12%|█▏        | 766/6470 [11:34<1:14:14,  1.28it/s]
1|767|Loss: 1.0371865034103394:  12%|█▏        | 766/6470 [11:34<1:14:14,  1.28it/s]
1|767|Loss: 1.0371865034103394:  12%|█▏        | 767/6470 [11:35<1:14:23,  1.28it/s]
1|768|Loss: 1.0183773040771484:  12%|█▏        | 767/6470 [11:35<1:14:23,  1.28it/s]
1|768|Loss: 1.0183773040771484:  12%|█▏        | 768/6470 [11:36<1:15:46,  1.25it/s]
1|769|Loss: 1.1513481140136719:  12%|█▏        | 768/6470 [11:36<1:15:46,  1.25it/s]
1|769|Loss: 1.1513481140136719:  12%|█▏        | 769/6470 [11:36<1:12:37,  1.31it/s]
1|770|Loss: 1.1859248876571655:  12%|█▏        | 769/6470 [11:37<1:12:37,  1.31it/s]
1|770|Loss: 1.1859248876571655:  12%|█▏        | 770/6470 [11:37<1:09:01,  1.38it/s]
1|771|Loss: 1.0531821250915527:  12%|█▏        | 770/6470 [11:37<1:09:01,  1.38it/s]
1|771|Loss: 1.0531821250915527:  12%|█▏        | 771/6470 [11:38<1:10:23,  1.35it/s]
1|772|Loss: 1.0650665760040283:  12%|█▏        | 771/6470 [11:38<1:10:23,  1.35it/s]
1|772|Loss: 1.0650665760040283:  12%|█▏        | 772/6470 [11:38<1:10:37,  1.34it/s]
1|773|Loss: 1.1805627346038818:  12%|█▏        | 772/6470 [11:39<1:10:37,  1.34it/s]
1|773|Loss: 1.1805627346038818:  12%|█▏        | 773/6470 [11:39<1:07:50,  1.40it/s]
1|774|Loss: 1.0592759847640991:  12%|█▏        | 773/6470 [11:40<1:07:50,  1.40it/s]
1|774|Loss: 1.0592759847640991:  12%|█▏        | 774/6470 [11:40<1:06:59,  1.42it/s]
1|775|Loss: 0.977271318435669:  12%|█▏        | 774/6470 [11:40<1:06:59,  1.42it/s] 
1|775|Loss: 0.977271318435669:  12%|█▏        | 775/6470 [11:41<1:08:15,  1.39it/s]
1|776|Loss: 1.1473947763442993:  12%|█▏        | 775/6470 [11:41<1:08:15,  1.39it/s]
1|776|Loss: 1.1473947763442993:  12%|█▏        | 776/6470 [11:41<1:08:59,  1.38it/s]
1|777|Loss: 1.079344630241394:  12%|█▏        | 776/6470 [11:42<1:08:59,  1.38it/s] 
1|777|Loss: 1.079344630241394:  12%|█▏        | 777/6470 [11:42<1:09:28,  1.37it/s]
1|778|Loss: 1.1260184049606323:  12%|█▏        | 777/6470 [11:42<1:09:28,  1.37it/s]
1|778|Loss: 1.1260184049606323:  12%|█▏        | 778/6470 [11:43<1:09:59,  1.36it/s]
1|779|Loss: 0.992436945438385:  12%|█▏        | 778/6470 [11:43<1:09:59,  1.36it/s] 
1|779|Loss: 0.992436945438385:  12%|█▏        | 779/6470 [11:44<1:12:05,  1.32it/s]
1|780|Loss: 1.3761320114135742:  12%|█▏        | 779/6470 [11:44<1:12:05,  1.32it/s]
1|780|Loss: 1.3761320114135742:  12%|█▏        | 780/6470 [11:44<1:02:39,  1.51it/s]
1|781|Loss: 1.0816563367843628:  12%|█▏        | 780/6470 [11:44<1:02:39,  1.51it/s]
1|781|Loss: 1.0816563367843628:  12%|█▏        | 781/6470 [11:45<58:37,  1.62it/s]  
1|782|Loss: 1.1980633735656738:  12%|█▏        | 781/6470 [11:45<58:37,  1.62it/s]
1|782|Loss: 1.1980633735656738:  12%|█▏        | 782/6470 [11:45<1:01:01,  1.55it/s]
1|783|Loss: 1.323357105255127:  12%|█▏        | 782/6470 [11:46<1:01:01,  1.55it/s] 
1|783|Loss: 1.323357105255127:  12%|█▏        | 783/6470 [11:46<59:23,  1.60it/s]  
1|784|Loss: 1.4489418268203735:  12%|█▏        | 783/6470 [11:46<59:23,  1.60it/s]
1|784|Loss: 1.4489418268203735:  12%|█▏        | 784/6470 [11:46<55:05,  1.72it/s]
1|785|Loss: 0.9080103039741516:  12%|█▏        | 784/6470 [11:47<55:05,  1.72it/s]
1|785|Loss: 0.9080103039741516:  12%|█▏        | 785/6470 [11:47<58:53,  1.61it/s]
1|786|Loss: 1.1150301694869995:  12%|█▏        | 785/6470 [11:47<58:53,  1.61it/s]
1|786|Loss: 1.1150301694869995:  12%|█▏        | 786/6470 [11:48<1:04:12,  1.48it/s]
1|787|Loss: 1.143954873085022:  12%|█▏        | 786/6470 [11:48<1:04:12,  1.48it/s] 
1|787|Loss: 1.143954873085022:  12%|█▏        | 787/6470 [11:48<1:02:08,  1.52it/s]
1|788|Loss: 1.0606255531311035:  12%|█▏        | 787/6470 [11:49<1:02:08,  1.52it/s]
1|788|Loss: 1.0606255531311035:  12%|█▏        | 788/6470 [11:49<1:05:31,  1.45it/s]
1|789|Loss: 1.0349531173706055:  12%|█▏        | 788/6470 [11:50<1:05:31,  1.45it/s]
1|789|Loss: 1.0349531173706055:  12%|█▏        | 789/6470 [11:50<1:07:45,  1.40it/s]
1|790|Loss: 1.0102829933166504:  12%|█▏        | 789/6470 [11:50<1:07:45,  1.40it/s]
1|790|Loss: 1.0102829933166504:  12%|█▏        | 790/6470 [11:51<1:12:32,  1.30it/s]
1|791|Loss: 1.0630847215652466:  12%|█▏        | 790/6470 [11:51<1:12:32,  1.30it/s]
1|791|Loss: 1.0630847215652466:  12%|█▏        | 791/6470 [11:52<1:13:07,  1.29it/s]
1|792|Loss: 1.0087182521820068:  12%|█▏        | 791/6470 [11:52<1:13:07,  1.29it/s]
1|792|Loss: 1.0087182521820068:  12%|█▏        | 792/6470 [11:52<1:05:56,  1.44it/s]
1|793|Loss: 0.9510853290557861:  12%|█▏        | 792/6470 [11:53<1:05:56,  1.44it/s]
1|793|Loss: 0.9510853290557861:  12%|█▏        | 793/6470 [11:53<1:05:33,  1.44it/s]
1|794|Loss: 1.164180040359497:  12%|█▏        | 793/6470 [11:53<1:05:33,  1.44it/s] 
1|794|Loss: 1.164180040359497:  12%|█▏        | 794/6470 [11:54<1:04:23,  1.47it/s]
1|795|Loss: 1.1971873044967651:  12%|█▏        | 794/6470 [11:54<1:04:23,  1.47it/s]
1|795|Loss: 1.1971873044967651:  12%|█▏        | 795/6470 [11:54<1:01:02,  1.55it/s]
1|796|Loss: 1.1808420419692993:  12%|█▏        | 795/6470 [11:54<1:01:02,  1.55it/s]
1|796|Loss: 1.1808420419692993:  12%|█▏        | 796/6470 [11:55<58:27,  1.62it/s]  
1|797|Loss: 1.076591968536377:  12%|█▏        | 796/6470 [11:55<58:27,  1.62it/s] 
1|797|Loss: 1.076591968536377:  12%|█▏        | 797/6470 [11:55<1:01:01,  1.55it/s]
1|798|Loss: 1.2167619466781616:  12%|█▏        | 797/6470 [11:56<1:01:01,  1.55it/s]
1|798|Loss: 1.2167619466781616:  12%|█▏        | 798/6470 [11:56<57:21,  1.65it/s]  
1|799|Loss: 1.147735834121704:  12%|█▏        | 798/6470 [11:56<57:21,  1.65it/s] 
1|799|Loss: 1.147735834121704:  12%|█▏        | 799/6470 [11:57<1:02:30,  1.51it/s]
1|800|Loss: 1.169230580329895:  12%|█▏        | 799/6470 [11:57<1:02:30,  1.51it/s]
1|800|Loss: 1.169230580329895:  12%|█▏        | 800/6470 [11:57<1:05:42,  1.44it/s]
1|801|Loss: 1.228893518447876:  12%|█▏        | 800/6470 [11:58<1:05:42,  1.44it/s]
1|801|Loss: 1.228893518447876:  12%|█▏        | 801/6470 [11:58<1:06:09,  1.43it/s]
1|802|Loss: 1.1150459051132202:  12%|█▏        | 801/6470 [11:59<1:06:09,  1.43it/s]
1|802|Loss: 1.1150459051132202:  12%|█▏        | 802/6470 [11:59<1:06:09,  1.43it/s]
1|803|Loss: 1.0298268795013428:  12%|█▏        | 802/6470 [11:59<1:06:09,  1.43it/s]
1|803|Loss: 1.0298268795013428:  12%|█▏        | 803/6470 [12:00<1:07:04,  1.41it/s]
1|804|Loss: 1.079980492591858:  12%|█▏        | 803/6470 [12:00<1:07:04,  1.41it/s] 
1|804|Loss: 1.079980492591858:  12%|█▏        | 804/6470 [12:00<1:07:43,  1.39it/s]
1|805|Loss: 1.0459009408950806:  12%|█▏        | 804/6470 [12:01<1:07:43,  1.39it/s]
1|805|Loss: 1.0459009408950806:  12%|█▏        | 805/6470 [12:01<1:07:36,  1.40it/s]
1|806|Loss: 0.8922942280769348:  12%|█▏        | 805/6470 [12:01<1:07:36,  1.40it/s]
1|806|Loss: 0.8922942280769348:  12%|█▏        | 806/6470 [12:02<1:12:06,  1.31it/s]
1|807|Loss: 1.2488703727722168:  12%|█▏        | 806/6470 [12:02<1:12:06,  1.31it/s]
1|807|Loss: 1.2488703727722168:  12%|█▏        | 807/6470 [12:02<1:06:09,  1.43it/s]
1|808|Loss: 0.9962544441223145:  12%|█▏        | 807/6470 [12:03<1:06:09,  1.43it/s]
1|808|Loss: 0.9962544441223145:  12%|█▏        | 808/6470 [12:03<1:15:19,  1.25it/s]
1|809|Loss: 1.1404856443405151:  12%|█▏        | 808/6470 [12:04<1:15:19,  1.25it/s]
1|809|Loss: 1.1404856443405151:  13%|█▎        | 809/6470 [12:04<1:15:17,  1.25it/s]
1|810|Loss: 1.107366681098938:  13%|█▎        | 809/6470 [12:05<1:15:17,  1.25it/s] 
1|810|Loss: 1.107366681098938:  13%|█▎        | 810/6470 [12:05<1:08:07,  1.38it/s]
1|811|Loss: 1.042220115661621:  13%|█▎        | 810/6470 [12:05<1:08:07,  1.38it/s]
1|811|Loss: 1.042220115661621:  13%|█▎        | 811/6470 [12:06<1:07:32,  1.40it/s]
1|812|Loss: 1.0693563222885132:  13%|█▎        | 811/6470 [12:06<1:07:32,  1.40it/s]
1|812|Loss: 1.0693563222885132:  13%|█▎        | 812/6470 [12:06<1:04:49,  1.45it/s]
1|813|Loss: 1.2332863807678223:  13%|█▎        | 812/6470 [12:06<1:04:49,  1.45it/s]
1|813|Loss: 1.2332863807678223:  13%|█▎        | 813/6470 [12:07<59:36,  1.58it/s]  
1|814|Loss: 1.2006430625915527:  13%|█▎        | 813/6470 [12:07<59:36,  1.58it/s]
1|814|Loss: 1.2006430625915527:  13%|█▎        | 814/6470 [12:07<55:57,  1.68it/s]
1|815|Loss: 1.0766702890396118:  13%|█▎        | 814/6470 [12:07<55:57,  1.68it/s]
1|815|Loss: 1.0766702890396118:  13%|█▎        | 815/6470 [12:08<56:28,  1.67it/s]
1|816|Loss: 1.1037094593048096:  13%|█▎        | 815/6470 [12:08<56:28,  1.67it/s]
1|816|Loss: 1.1037094593048096:  13%|█▎        | 816/6470 [12:08<57:46,  1.63it/s]
1|817|Loss: 0.9172113537788391:  13%|█▎        | 816/6470 [12:09<57:46,  1.63it/s]
1|817|Loss: 0.9172113537788391:  13%|█▎        | 817/6470 [12:09<1:00:55,  1.55it/s]
1|818|Loss: 1.1669867038726807:  13%|█▎        | 817/6470 [12:10<1:00:55,  1.55it/s]
1|818|Loss: 1.1669867038726807:  13%|█▎        | 818/6470 [12:10<1:03:25,  1.49it/s]
1|819|Loss: 1.1700674295425415:  13%|█▎        | 818/6470 [12:10<1:03:25,  1.49it/s]
1|819|Loss: 1.1700674295425415:  13%|█▎        | 819/6470 [12:10<59:19,  1.59it/s]  
1|820|Loss: 1.0248063802719116:  13%|█▎        | 819/6470 [12:11<59:19,  1.59it/s]
1|820|Loss: 1.0248063802719116:  13%|█▎        | 820/6470 [12:11<1:00:42,  1.55it/s]
1|821|Loss: 1.1507065296173096:  13%|█▎        | 820/6470 [12:12<1:00:42,  1.55it/s]
1|821|Loss: 1.1507065296173096:  13%|█▎        | 821/6470 [12:12<1:04:34,  1.46it/s]
1|822|Loss: 0.9659947156906128:  13%|█▎        | 821/6470 [12:12<1:04:34,  1.46it/s]
1|822|Loss: 0.9659947156906128:  13%|█▎        | 822/6470 [12:13<1:03:49,  1.47it/s]
1|823|Loss: 1.0169172286987305:  13%|█▎        | 822/6470 [12:13<1:03:49,  1.47it/s]
1|823|Loss: 1.0169172286987305:  13%|█▎        | 823/6470 [12:13<1:01:24,  1.53it/s]
1|824|Loss: 1.115131139755249:  13%|█▎        | 823/6470 [12:14<1:01:24,  1.53it/s] 
1|824|Loss: 1.115131139755249:  13%|█▎        | 824/6470 [12:14<1:04:54,  1.45it/s]
1|825|Loss: 1.1166988611221313:  13%|█▎        | 824/6470 [12:14<1:04:54,  1.45it/s]
1|825|Loss: 1.1166988611221313:  13%|█▎        | 825/6470 [12:15<1:04:49,  1.45it/s]
1|826|Loss: 1.071473240852356:  13%|█▎        | 825/6470 [12:15<1:04:49,  1.45it/s] 
1|826|Loss: 1.071473240852356:  13%|█▎        | 826/6470 [12:15<1:09:01,  1.36it/s]
1|827|Loss: 0.9719586968421936:  13%|█▎        | 826/6470 [12:16<1:09:01,  1.36it/s]
1|827|Loss: 0.9719586968421936:  13%|█▎        | 827/6470 [12:16<1:09:23,  1.36it/s]
1|828|Loss: 1.183970332145691:  13%|█▎        | 827/6470 [12:16<1:09:23,  1.36it/s] 
1|828|Loss: 1.183970332145691:  13%|█▎        | 828/6470 [12:17<1:02:50,  1.50it/s]
1|829|Loss: 1.216166377067566:  13%|█▎        | 828/6470 [12:17<1:02:50,  1.50it/s]
1|829|Loss: 1.216166377067566:  13%|█▎        | 829/6470 [12:17<1:02:49,  1.50it/s]
1|830|Loss: 1.162567138671875:  13%|█▎        | 829/6470 [12:18<1:02:49,  1.50it/s]
1|830|Loss: 1.162567138671875:  13%|█▎        | 830/6470 [12:18<1:04:40,  1.45it/s]
1|831|Loss: 1.2508612871170044:  13%|█▎        | 830/6470 [12:18<1:04:40,  1.45it/s]
1|831|Loss: 1.2508612871170044:  13%|█▎        | 831/6470 [12:19<1:06:02,  1.42it/s]
1|832|Loss: 0.9084734320640564:  13%|█▎        | 831/6470 [12:19<1:06:02,  1.42it/s]
1|832|Loss: 0.9084734320640564:  13%|█▎        | 832/6470 [12:20<1:06:10,  1.42it/s]
1|833|Loss: 1.095147967338562:  13%|█▎        | 832/6470 [12:20<1:06:10,  1.42it/s] 
1|833|Loss: 1.095147967338562:  13%|█▎        | 833/6470 [12:20<1:09:56,  1.34it/s]
1|834|Loss: 1.0826823711395264:  13%|█▎        | 833/6470 [12:21<1:09:56,  1.34it/s]
1|834|Loss: 1.0826823711395264:  13%|█▎        | 834/6470 [12:21<1:07:08,  1.40it/s]
1|835|Loss: 1.164286494255066:  13%|█▎        | 834/6470 [12:21<1:07:08,  1.40it/s] 
1|835|Loss: 1.164286494255066:  13%|█▎        | 835/6470 [12:22<1:07:00,  1.40it/s]
1|836|Loss: 1.0347577333450317:  13%|█▎        | 835/6470 [12:22<1:07:00,  1.40it/s]
1|836|Loss: 1.0347577333450317:  13%|█▎        | 836/6470 [12:22<1:08:02,  1.38it/s]
1|837|Loss: 1.0650930404663086:  13%|█▎        | 836/6470 [12:23<1:08:02,  1.38it/s]
1|837|Loss: 1.0650930404663086:  13%|█▎        | 837/6470 [12:23<1:07:52,  1.38it/s]
1|838|Loss: 1.1740392446517944:  13%|█▎        | 837/6470 [12:24<1:07:52,  1.38it/s]
1|838|Loss: 1.1740392446517944:  13%|█▎        | 838/6470 [12:24<1:07:58,  1.38it/s]
1|839|Loss: 1.2305891513824463:  13%|█▎        | 838/6470 [12:24<1:07:58,  1.38it/s]
1|839|Loss: 1.2305891513824463:  13%|█▎        | 839/6470 [12:25<1:04:57,  1.44it/s]
1|840|Loss: 1.02494478225708:  13%|█▎        | 839/6470 [12:25<1:04:57,  1.44it/s]  
1|840|Loss: 1.02494478225708:  13%|█▎        | 840/6470 [12:25<1:05:29,  1.43it/s]
1|841|Loss: 1.0903648138046265:  13%|█▎        | 840/6470 [12:26<1:05:29,  1.43it/s]
1|841|Loss: 1.0903648138046265:  13%|█▎        | 841/6470 [12:26<1:08:01,  1.38it/s]
1|842|Loss: 1.081053614616394:  13%|█▎        | 841/6470 [12:26<1:08:01,  1.38it/s] 
1|842|Loss: 1.081053614616394:  13%|█▎        | 842/6470 [12:27<1:11:15,  1.32it/s]
1|843|Loss: 1.0168159008026123:  13%|█▎        | 842/6470 [12:27<1:11:15,  1.32it/s]
1|843|Loss: 1.0168159008026123:  13%|█▎        | 843/6470 [12:28<1:15:16,  1.25it/s]
1|844|Loss: 1.083613395690918:  13%|█▎        | 843/6470 [12:28<1:15:16,  1.25it/s] 
1|844|Loss: 1.083613395690918:  13%|█▎        | 844/6470 [12:29<1:14:33,  1.26it/s]
1|845|Loss: 1.0493839979171753:  13%|█▎        | 844/6470 [12:29<1:14:33,  1.26it/s]
1|845|Loss: 1.0493839979171753:  13%|█▎        | 845/6470 [12:29<1:11:03,  1.32it/s]
1|846|Loss: 1.0357590913772583:  13%|█▎        | 845/6470 [12:30<1:11:03,  1.32it/s]
1|846|Loss: 1.0357590913772583:  13%|█▎        | 846/6470 [12:30<1:09:18,  1.35it/s]
1|847|Loss: 1.0443230867385864:  13%|█▎        | 846/6470 [12:30<1:09:18,  1.35it/s]
1|847|Loss: 1.0443230867385864:  13%|█▎        | 847/6470 [12:31<1:08:58,  1.36it/s]
1|848|Loss: 1.0635473728179932:  13%|█▎        | 847/6470 [12:31<1:08:58,  1.36it/s]
1|848|Loss: 1.0635473728179932:  13%|█▎        | 848/6470 [12:31<1:08:46,  1.36it/s]
1|849|Loss: 1.1425281763076782:  13%|█▎        | 848/6470 [12:32<1:08:46,  1.36it/s]
1|849|Loss: 1.1425281763076782:  13%|█▎        | 849/6470 [12:32<1:09:54,  1.34it/s]
1|850|Loss: 1.2719329595565796:  13%|█▎        | 849/6470 [12:33<1:09:54,  1.34it/s]
1|850|Loss: 1.2719329595565796:  13%|█▎        | 850/6470 [12:33<1:11:53,  1.30it/s]
1|851|Loss: 1.2396552562713623:  13%|█▎        | 850/6470 [12:33<1:11:53,  1.30it/s]
1|851|Loss: 1.2396552562713623:  13%|█▎        | 851/6470 [12:34<1:09:17,  1.35it/s]
1|852|Loss: 1.2339179515838623:  13%|█▎        | 851/6470 [12:34<1:09:17,  1.35it/s]
1|852|Loss: 1.2339179515838623:  13%|█▎        | 852/6470 [12:34<1:10:42,  1.32it/s]
1|853|Loss: 0.953738272190094:  13%|█▎        | 852/6470 [12:35<1:10:42,  1.32it/s] 
1|853|Loss: 0.953738272190094:  13%|█▎        | 853/6470 [12:35<1:11:05,  1.32it/s]
1|854|Loss: 1.0844060182571411:  13%|█▎        | 853/6470 [12:36<1:11:05,  1.32it/s]
1|854|Loss: 1.0844060182571411:  13%|█▎        | 854/6470 [12:36<1:10:40,  1.32it/s]
1|855|Loss: 1.0278843641281128:  13%|█▎        | 854/6470 [12:36<1:10:40,  1.32it/s]
1|855|Loss: 1.0278843641281128:  13%|█▎        | 855/6470 [12:37<1:08:09,  1.37it/s]
1|856|Loss: 1.2219645977020264:  13%|█▎        | 855/6470 [12:37<1:08:09,  1.37it/s]
1|856|Loss: 1.2219645977020264:  13%|█▎        | 856/6470 [12:37<1:02:49,  1.49it/s]
1|857|Loss: 1.1803991794586182:  13%|█▎        | 856/6470 [12:37<1:02:49,  1.49it/s]
1|857|Loss: 1.1803991794586182:  13%|█▎        | 857/6470 [12:38<57:00,  1.64it/s]  
1|858|Loss: 1.1063932180404663:  13%|█▎        | 857/6470 [12:38<57:00,  1.64it/s]
1|858|Loss: 1.1063932180404663:  13%|█▎        | 858/6470 [12:38<59:11,  1.58it/s]
1|859|Loss: 1.1662777662277222:  13%|█▎        | 858/6470 [12:39<59:11,  1.58it/s]
1|859|Loss: 1.1662777662277222:  13%|█▎        | 859/6470 [12:39<59:45,  1.56it/s]
1|860|Loss: 1.0740002393722534:  13%|█▎        | 859/6470 [12:39<59:45,  1.56it/s]
1|860|Loss: 1.0740002393722534:  13%|█▎        | 860/6470 [12:40<56:59,  1.64it/s]
1|861|Loss: 1.0692830085754395:  13%|█▎        | 860/6470 [12:40<56:59,  1.64it/s]
1|861|Loss: 1.0692830085754395:  13%|█▎        | 861/6470 [12:40<59:28,  1.57it/s]
1|862|Loss: 1.1336199045181274:  13%|█▎        | 861/6470 [12:41<59:28,  1.57it/s]
1|862|Loss: 1.1336199045181274:  13%|█▎        | 862/6470 [12:41<1:01:41,  1.52it/s]
1|863|Loss: 0.9125655889511108:  13%|█▎        | 862/6470 [12:41<1:01:41,  1.52it/s]
1|863|Loss: 0.9125655889511108:  13%|█▎        | 863/6470 [12:42<1:00:08,  1.55it/s]
1|864|Loss: 1.1774429082870483:  13%|█▎        | 863/6470 [12:42<1:00:08,  1.55it/s]
1|864|Loss: 1.1774429082870483:  13%|█▎        | 864/6470 [12:42<59:52,  1.56it/s]  
1|865|Loss: 1.052677869796753:  13%|█▎        | 864/6470 [12:43<59:52,  1.56it/s] 
1|865|Loss: 1.052677869796753:  13%|█▎        | 865/6470 [12:43<1:03:18,  1.48it/s]
1|866|Loss: 1.0521894693374634:  13%|█▎        | 865/6470 [12:43<1:03:18,  1.48it/s]
1|866|Loss: 1.0521894693374634:  13%|█▎        | 866/6470 [12:44<1:06:33,  1.40it/s]
1|867|Loss: 1.1330255270004272:  13%|█▎        | 866/6470 [12:44<1:06:33,  1.40it/s]
1|867|Loss: 1.1330255270004272:  13%|█▎        | 867/6470 [12:45<1:09:41,  1.34it/s]
1|868|Loss: 1.0366673469543457:  13%|█▎        | 867/6470 [12:45<1:09:41,  1.34it/s]
1|868|Loss: 1.0366673469543457:  13%|█▎        | 868/6470 [12:45<1:07:27,  1.38it/s]
1|869|Loss: 1.1033912897109985:  13%|█▎        | 868/6470 [12:46<1:07:27,  1.38it/s]
1|869|Loss: 1.1033912897109985:  13%|█▎        | 869/6470 [12:46<1:06:57,  1.39it/s]
1|870|Loss: 1.018615484237671:  13%|█▎        | 869/6470 [12:46<1:06:57,  1.39it/s] 
1|870|Loss: 1.018615484237671:  13%|█▎        | 870/6470 [12:47<1:11:21,  1.31it/s]
1|871|Loss: 1.1399909257888794:  13%|█▎        | 870/6470 [12:47<1:11:21,  1.31it/s]
1|871|Loss: 1.1399909257888794:  13%|█▎        | 871/6470 [12:48<1:10:48,  1.32it/s]
1|872|Loss: 1.0287827253341675:  13%|█▎        | 871/6470 [12:48<1:10:48,  1.32it/s]
1|872|Loss: 1.0287827253341675:  13%|█▎        | 872/6470 [12:48<1:09:07,  1.35it/s]
1|873|Loss: 0.927412211894989:  13%|█▎        | 872/6470 [12:49<1:09:07,  1.35it/s] 
1|873|Loss: 0.927412211894989:  13%|█▎        | 873/6470 [12:49<1:10:46,  1.32it/s]
1|874|Loss: 1.2566474676132202:  13%|█▎        | 873/6470 [12:49<1:10:46,  1.32it/s]
1|874|Loss: 1.2566474676132202:  14%|█▎        | 874/6470 [12:50<1:13:16,  1.27it/s]
1|875|Loss: 1.0606224536895752:  14%|█▎        | 874/6470 [12:50<1:13:16,  1.27it/s]
1|875|Loss: 1.0606224536895752:  14%|█▎        | 875/6470 [12:51<1:15:48,  1.23it/s]
1|876|Loss: 1.1303253173828125:  14%|█▎        | 875/6470 [12:51<1:15:48,  1.23it/s]
1|876|Loss: 1.1303253173828125:  14%|█▎        | 876/6470 [12:51<1:11:21,  1.31it/s]
1|877|Loss: 1.2552155256271362:  14%|█▎        | 876/6470 [12:52<1:11:21,  1.31it/s]
1|877|Loss: 1.2552155256271362:  14%|█▎        | 877/6470 [12:52<1:08:45,  1.36it/s]
1|878|Loss: 1.0813648700714111:  14%|█▎        | 877/6470 [12:53<1:08:45,  1.36it/s]
1|878|Loss: 1.0813648700714111:  14%|█▎        | 878/6470 [12:53<1:11:11,  1.31it/s]
1|879|Loss: 1.1261136531829834:  14%|█▎        | 878/6470 [12:53<1:11:11,  1.31it/s]
1|879|Loss: 1.1261136531829834:  14%|█▎        | 879/6470 [12:54<1:07:58,  1.37it/s]
1|880|Loss: 1.208727240562439:  14%|█▎        | 879/6470 [12:54<1:07:58,  1.37it/s] 
1|880|Loss: 1.208727240562439:  14%|█▎        | 880/6470 [12:54<1:02:08,  1.50it/s]
1|881|Loss: 1.2084423303604126:  14%|█▎        | 880/6470 [12:54<1:02:08,  1.50it/s]
1|881|Loss: 1.2084423303604126:  14%|█▎        | 881/6470 [12:55<59:26,  1.57it/s]  
1|882|Loss: 1.0731455087661743:  14%|█▎        | 881/6470 [12:55<59:26,  1.57it/s]
1|882|Loss: 1.0731455087661743:  14%|█▎        | 882/6470 [12:55<1:03:52,  1.46it/s]
1|883|Loss: 1.2915804386138916:  14%|█▎        | 882/6470 [12:56<1:03:52,  1.46it/s]
1|883|Loss: 1.2915804386138916:  14%|█▎        | 883/6470 [12:56<1:05:03,  1.43it/s]
1|884|Loss: 1.1015796661376953:  14%|█▎        | 883/6470 [12:57<1:05:03,  1.43it/s]
1|884|Loss: 1.1015796661376953:  14%|█▎        | 884/6470 [12:57<1:08:25,  1.36it/s]
1|885|Loss: 1.177191138267517:  14%|█▎        | 884/6470 [12:57<1:08:25,  1.36it/s] 
1|885|Loss: 1.177191138267517:  14%|█▎        | 885/6470 [12:58<1:04:17,  1.45it/s]
1|886|Loss: 1.01010262966156:  14%|█▎        | 885/6470 [12:58<1:04:17,  1.45it/s] 
1|886|Loss: 1.01010262966156:  14%|█▎        | 886/6470 [12:58<1:06:28,  1.40it/s]
1|887|Loss: 0.9348987936973572:  14%|█▎        | 886/6470 [12:59<1:06:28,  1.40it/s]
1|887|Loss: 0.9348987936973572:  14%|█▎        | 887/6470 [12:59<1:04:12,  1.45it/s]
1|888|Loss: 1.214795708656311:  14%|█▎        | 887/6470 [12:59<1:04:12,  1.45it/s] 
1|888|Loss: 1.214795708656311:  14%|█▎        | 888/6470 [13:00<1:04:56,  1.43it/s]
1|889|Loss: 0.9616501927375793:  14%|█▎        | 888/6470 [13:00<1:04:56,  1.43it/s]
1|889|Loss: 0.9616501927375793:  14%|█▎        | 889/6470 [13:00<1:07:33,  1.38it/s]
1|890|Loss: 1.1710186004638672:  14%|█▎        | 889/6470 [13:01<1:07:33,  1.38it/s]
1|890|Loss: 1.1710186004638672:  14%|█▍        | 890/6470 [13:01<1:05:39,  1.42it/s]
1|891|Loss: 1.0487185716629028:  14%|█▍        | 890/6470 [13:02<1:05:39,  1.42it/s]
1|891|Loss: 1.0487185716629028:  14%|█▍        | 891/6470 [13:02<1:03:09,  1.47it/s]
1|892|Loss: 1.1571762561798096:  14%|█▍        | 891/6470 [13:02<1:03:09,  1.47it/s]
1|892|Loss: 1.1571762561798096:  14%|█▍        | 892/6470 [13:02<1:01:57,  1.50it/s]
1|893|Loss: 0.8037158250808716:  14%|█▍        | 892/6470 [13:03<1:01:57,  1.50it/s]
1|893|Loss: 0.8037158250808716:  14%|█▍        | 893/6470 [13:03<1:07:17,  1.38it/s]
1|894|Loss: 1.1640655994415283:  14%|█▍        | 893/6470 [13:04<1:07:17,  1.38it/s]
1|894|Loss: 1.1640655994415283:  14%|█▍        | 894/6470 [13:04<1:05:00,  1.43it/s]
1|895|Loss: 1.1189380884170532:  14%|█▍        | 894/6470 [13:04<1:05:00,  1.43it/s]
1|895|Loss: 1.1189380884170532:  14%|█▍        | 895/6470 [13:05<1:03:01,  1.47it/s]
1|896|Loss: 1.1246339082717896:  14%|█▍        | 895/6470 [13:05<1:03:01,  1.47it/s]
1|896|Loss: 1.1246339082717896:  14%|█▍        | 896/6470 [13:05<1:04:22,  1.44it/s]
1|897|Loss: 1.1049097776412964:  14%|█▍        | 896/6470 [13:06<1:04:22,  1.44it/s]
1|897|Loss: 1.1049097776412964:  14%|█▍        | 897/6470 [13:06<1:05:54,  1.41it/s]
1|898|Loss: 1.0544078350067139:  14%|█▍        | 897/6470 [13:06<1:05:54,  1.41it/s]
1|898|Loss: 1.0544078350067139:  14%|█▍        | 898/6470 [13:07<1:06:55,  1.39it/s]
1|899|Loss: 1.013214111328125:  14%|█▍        | 898/6470 [13:07<1:06:55,  1.39it/s] 
1|899|Loss: 1.013214111328125:  14%|█▍        | 899/6470 [13:08<1:08:50,  1.35it/s]
1|900|Loss: 0.9703436493873596:  14%|█▍        | 899/6470 [13:08<1:08:50,  1.35it/s]
1|900|Loss: 0.9703436493873596:  14%|█▍        | 900/6470 [13:08<1:06:44,  1.39it/s]
1|901|Loss: 1.2073884010314941:  14%|█▍        | 900/6470 [13:09<1:06:44,  1.39it/s]
1|901|Loss: 1.2073884010314941:  14%|█▍        | 901/6470 [13:09<1:04:15,  1.44it/s]
1|902|Loss: 1.0931881666183472:  14%|█▍        | 901/6470 [13:09<1:04:15,  1.44it/s]
1|902|Loss: 1.0931881666183472:  14%|█▍        | 902/6470 [13:10<1:05:09,  1.42it/s]
1|903|Loss: 1.2118185758590698:  14%|█▍        | 902/6470 [13:10<1:05:09,  1.42it/s]
1|903|Loss: 1.2118185758590698:  14%|█▍        | 903/6470 [13:10<1:03:40,  1.46it/s]
1|904|Loss: 0.9367146492004395:  14%|█▍        | 903/6470 [13:11<1:03:40,  1.46it/s]
1|904|Loss: 0.9367146492004395:  14%|█▍        | 904/6470 [13:11<1:06:06,  1.40it/s]
1|905|Loss: 0.9830425381660461:  14%|█▍        | 904/6470 [13:11<1:06:06,  1.40it/s]
1|905|Loss: 0.9830425381660461:  14%|█▍        | 905/6470 [13:12<1:06:30,  1.39it/s]
1|906|Loss: 1.2295796871185303:  14%|█▍        | 905/6470 [13:12<1:06:30,  1.39it/s]
1|906|Loss: 1.2295796871185303:  14%|█▍        | 906/6470 [13:12<1:00:29,  1.53it/s]
1|907|Loss: 1.123783826828003:  14%|█▍        | 906/6470 [13:13<1:00:29,  1.53it/s] 
1|907|Loss: 1.123783826828003:  14%|█▍        | 907/6470 [13:13<1:02:04,  1.49it/s]
1|908|Loss: 0.976739227771759:  14%|█▍        | 907/6470 [13:13<1:02:04,  1.49it/s]
1|908|Loss: 0.976739227771759:  14%|█▍        | 908/6470 [13:14<1:02:11,  1.49it/s]
1|909|Loss: 1.105756402015686:  14%|█▍        | 908/6470 [13:14<1:02:11,  1.49it/s]
1|909|Loss: 1.105756402015686:  14%|█▍        | 909/6470 [13:14<1:00:49,  1.52it/s]
1|910|Loss: 1.213313341140747:  14%|█▍        | 909/6470 [13:15<1:00:49,  1.52it/s]
1|910|Loss: 1.213313341140747:  14%|█▍        | 910/6470 [13:15<1:01:42,  1.50it/s]
1|911|Loss: 1.130263328552246:  14%|█▍        | 910/6470 [13:15<1:01:42,  1.50it/s]
1|911|Loss: 1.130263328552246:  14%|█▍        | 911/6470 [13:16<1:04:41,  1.43it/s]
1|912|Loss: 1.0066781044006348:  14%|█▍        | 911/6470 [13:16<1:04:41,  1.43it/s]
1|912|Loss: 1.0066781044006348:  14%|█▍        | 912/6470 [13:16<1:06:01,  1.40it/s]
1|913|Loss: 1.1026757955551147:  14%|█▍        | 912/6470 [13:17<1:06:01,  1.40it/s]
1|913|Loss: 1.1026757955551147:  14%|█▍        | 913/6470 [13:17<1:05:52,  1.41it/s]
1|914|Loss: 1.0786226987838745:  14%|█▍        | 913/6470 [13:18<1:05:52,  1.41it/s]
1|914|Loss: 1.0786226987838745:  14%|█▍        | 914/6470 [13:18<1:03:40,  1.45it/s]
1|915|Loss: 1.2349640130996704:  14%|█▍        | 914/6470 [13:18<1:03:40,  1.45it/s]
1|915|Loss: 1.2349640130996704:  14%|█▍        | 915/6470 [13:18<1:04:09,  1.44it/s]
1|916|Loss: 0.9538163542747498:  14%|█▍        | 915/6470 [13:19<1:04:09,  1.44it/s]
1|916|Loss: 0.9538163542747498:  14%|█▍        | 916/6470 [13:19<1:03:14,  1.46it/s]
1|917|Loss: 1.2236617803573608:  14%|█▍        | 916/6470 [13:19<1:03:14,  1.46it/s]
1|917|Loss: 1.2236617803573608:  14%|█▍        | 917/6470 [13:20<1:01:28,  1.51it/s]
1|918|Loss: 1.126599669456482:  14%|█▍        | 917/6470 [13:20<1:01:28,  1.51it/s] 
1|918|Loss: 1.126599669456482:  14%|█▍        | 918/6470 [13:20<1:00:07,  1.54it/s]
1|919|Loss: 1.2296618223190308:  14%|█▍        | 918/6470 [13:21<1:00:07,  1.54it/s]
1|919|Loss: 1.2296618223190308:  14%|█▍        | 919/6470 [13:21<56:48,  1.63it/s]  
1|920|Loss: 0.9533365964889526:  14%|█▍        | 919/6470 [13:21<56:48,  1.63it/s]
1|920|Loss: 0.9533365964889526:  14%|█▍        | 920/6470 [13:22<58:11,  1.59it/s]
1|921|Loss: 0.9504565596580505:  14%|█▍        | 920/6470 [13:22<58:11,  1.59it/s]
1|921|Loss: 0.9504565596580505:  14%|█▍        | 921/6470 [13:22<1:02:22,  1.48it/s]
1|922|Loss: 1.0064085721969604:  14%|█▍        | 921/6470 [13:23<1:02:22,  1.48it/s]
1|922|Loss: 1.0064085721969604:  14%|█▍        | 922/6470 [13:23<1:04:19,  1.44it/s]
1|923|Loss: 1.1699706315994263:  14%|█▍        | 922/6470 [13:23<1:04:19,  1.44it/s]
1|923|Loss: 1.1699706315994263:  14%|█▍        | 923/6470 [13:24<1:02:42,  1.47it/s]
1|924|Loss: 0.8863333463668823:  14%|█▍        | 923/6470 [13:24<1:02:42,  1.47it/s]
1|924|Loss: 0.8863333463668823:  14%|█▍        | 924/6470 [13:24<1:01:31,  1.50it/s]
1|925|Loss: 1.0164517164230347:  14%|█▍        | 924/6470 [13:25<1:01:31,  1.50it/s]
1|925|Loss: 1.0164517164230347:  14%|█▍        | 925/6470 [13:25<1:04:19,  1.44it/s]
1|926|Loss: 1.0523065328598022:  14%|█▍        | 925/6470 [13:26<1:04:19,  1.44it/s]
1|926|Loss: 1.0523065328598022:  14%|█▍        | 926/6470 [13:26<1:07:55,  1.36it/s]
1|927|Loss: 1.3973701000213623:  14%|█▍        | 926/6470 [13:26<1:07:55,  1.36it/s]
1|927|Loss: 1.3973701000213623:  14%|█▍        | 927/6470 [13:27<1:08:24,  1.35it/s]
1|928|Loss: 1.1352227926254272:  14%|█▍        | 927/6470 [13:27<1:08:24,  1.35it/s]
1|928|Loss: 1.1352227926254272:  14%|█▍        | 928/6470 [13:28<1:10:53,  1.30it/s]
1|929|Loss: 1.1476490497589111:  14%|█▍        | 928/6470 [13:28<1:10:53,  1.30it/s]
1|929|Loss: 1.1476490497589111:  14%|█▍        | 929/6470 [13:28<1:11:05,  1.30it/s]
1|930|Loss: 0.9218441843986511:  14%|█▍        | 929/6470 [13:29<1:11:05,  1.30it/s]
1|930|Loss: 0.9218441843986511:  14%|█▍        | 930/6470 [13:29<1:10:12,  1.32it/s]
1|931|Loss: 1.07887864112854:  14%|█▍        | 930/6470 [13:29<1:10:12,  1.32it/s]  
1|931|Loss: 1.07887864112854:  14%|█▍        | 931/6470 [13:30<1:07:05,  1.38it/s]
1|932|Loss: 1.1767569780349731:  14%|█▍        | 931/6470 [13:30<1:07:05,  1.38it/s]
1|932|Loss: 1.1767569780349731:  14%|█▍        | 932/6470 [13:30<1:04:35,  1.43it/s]
1|933|Loss: 1.0335402488708496:  14%|█▍        | 932/6470 [13:31<1:04:35,  1.43it/s]
1|933|Loss: 1.0335402488708496:  14%|█▍        | 933/6470 [13:31<1:07:09,  1.37it/s]
1|934|Loss: 1.3257572650909424:  14%|█▍        | 933/6470 [13:32<1:07:09,  1.37it/s]
1|934|Loss: 1.3257572650909424:  14%|█▍        | 934/6470 [13:32<1:10:47,  1.30it/s]
1|935|Loss: 1.031105399131775:  14%|█▍        | 934/6470 [13:32<1:10:47,  1.30it/s] 
1|935|Loss: 1.031105399131775:  14%|█▍        | 935/6470 [13:33<1:07:32,  1.37it/s]
1|936|Loss: 0.9365658760070801:  14%|█▍        | 935/6470 [13:33<1:07:32,  1.37it/s]
1|936|Loss: 0.9365658760070801:  14%|█▍        | 936/6470 [13:33<1:07:06,  1.37it/s]
1|937|Loss: 1.093421220779419:  14%|█▍        | 936/6470 [13:34<1:07:06,  1.37it/s] 
1|937|Loss: 1.093421220779419:  14%|█▍        | 937/6470 [13:34<1:04:46,  1.42it/s]
1|938|Loss: 1.0405230522155762:  14%|█▍        | 937/6470 [13:34<1:04:46,  1.42it/s]
1|938|Loss: 1.0405230522155762:  14%|█▍        | 938/6470 [13:35<1:04:43,  1.42it/s]
1|939|Loss: 1.1379899978637695:  14%|█▍        | 938/6470 [13:35<1:04:43,  1.42it/s]
1|939|Loss: 1.1379899978637695:  15%|█▍        | 939/6470 [13:35<1:05:38,  1.40it/s]
1|940|Loss: 1.0362982749938965:  15%|█▍        | 939/6470 [13:36<1:05:38,  1.40it/s]
1|940|Loss: 1.0362982749938965:  15%|█▍        | 940/6470 [13:36<1:05:30,  1.41it/s]
1|941|Loss: 1.097165822982788:  15%|█▍        | 940/6470 [13:37<1:05:30,  1.41it/s] 
1|941|Loss: 1.097165822982788:  15%|█▍        | 941/6470 [13:37<1:05:50,  1.40it/s]
1|942|Loss: 1.15883207321167:  15%|█▍        | 941/6470 [13:37<1:05:50,  1.40it/s] 
1|942|Loss: 1.15883207321167:  15%|█▍        | 942/6470 [13:38<1:06:18,  1.39it/s]
1|943|Loss: 1.1008800268173218:  15%|█▍        | 942/6470 [13:38<1:06:18,  1.39it/s]
1|943|Loss: 1.1008800268173218:  15%|█▍        | 943/6470 [13:38<1:04:35,  1.43it/s]
1|944|Loss: 1.0682549476623535:  15%|█▍        | 943/6470 [13:39<1:04:35,  1.43it/s]
1|944|Loss: 1.0682549476623535:  15%|█▍        | 944/6470 [13:39<1:04:49,  1.42it/s]
1|945|Loss: 1.1066521406173706:  15%|█▍        | 944/6470 [13:39<1:04:49,  1.42it/s]
1|945|Loss: 1.1066521406173706:  15%|█▍        | 945/6470 [13:40<1:05:34,  1.40it/s]
1|946|Loss: 1.0821106433868408:  15%|█▍        | 945/6470 [13:40<1:05:34,  1.40it/s]
1|946|Loss: 1.0821106433868408:  15%|█▍        | 946/6470 [13:40<1:07:35,  1.36it/s]
1|947|Loss: 1.0740877389907837:  15%|█▍        | 946/6470 [13:41<1:07:35,  1.36it/s]
1|947|Loss: 1.0740877389907837:  15%|█▍        | 947/6470 [13:41<1:07:20,  1.37it/s]
1|948|Loss: 1.1270029544830322:  15%|█▍        | 947/6470 [13:42<1:07:20,  1.37it/s]
1|948|Loss: 1.1270029544830322:  15%|█▍        | 948/6470 [13:42<1:03:07,  1.46it/s]
1|949|Loss: 1.1678701639175415:  15%|█▍        | 948/6470 [13:42<1:03:07,  1.46it/s]
1|949|Loss: 1.1678701639175415:  15%|█▍        | 949/6470 [13:42<55:54,  1.65it/s]  
1|950|Loss: 1.1737583875656128:  15%|█▍        | 949/6470 [13:43<55:54,  1.65it/s]
1|950|Loss: 1.1737583875656128:  15%|█▍        | 950/6470 [13:43<58:51,  1.56it/s]
1|951|Loss: 1.0575284957885742:  15%|█▍        | 950/6470 [13:43<58:51,  1.56it/s]
1|951|Loss: 1.0575284957885742:  15%|█▍        | 951/6470 [13:44<1:04:31,  1.43it/s]
1|952|Loss: 1.114144206047058:  15%|█▍        | 951/6470 [13:44<1:04:31,  1.43it/s] 
1|952|Loss: 1.114144206047058:  15%|█▍        | 952/6470 [13:45<1:08:50,  1.34it/s]
1|953|Loss: 1.0112916231155396:  15%|█▍        | 952/6470 [13:45<1:08:50,  1.34it/s]
1|953|Loss: 1.0112916231155396:  15%|█▍        | 953/6470 [13:45<1:06:32,  1.38it/s]
1|954|Loss: 0.9964562654495239:  15%|█▍        | 953/6470 [13:46<1:06:32,  1.38it/s]
1|954|Loss: 0.9964562654495239:  15%|█▍        | 954/6470 [13:46<1:06:16,  1.39it/s]
1|955|Loss: 1.1346182823181152:  15%|█▍        | 954/6470 [13:46<1:06:16,  1.39it/s]
1|955|Loss: 1.1346182823181152:  15%|█▍        | 955/6470 [13:47<1:02:17,  1.48it/s]
1|956|Loss: 1.02870512008667:  15%|█▍        | 955/6470 [13:47<1:02:17,  1.48it/s]  
1|956|Loss: 1.02870512008667:  15%|█▍        | 956/6470 [13:48<1:08:34,  1.34it/s]
1|957|Loss: 1.048519253730774:  15%|█▍        | 956/6470 [13:48<1:08:34,  1.34it/s]
1|957|Loss: 1.048519253730774:  15%|█▍        | 957/6470 [13:48<1:09:25,  1.32it/s]
1|958|Loss: 1.2139424085617065:  15%|█▍        | 957/6470 [13:49<1:09:25,  1.32it/s]
1|958|Loss: 1.2139424085617065:  15%|█▍        | 958/6470 [13:49<1:03:39,  1.44it/s]
1|959|Loss: 1.0382745265960693:  15%|█▍        | 958/6470 [13:49<1:03:39,  1.44it/s]
1|959|Loss: 1.0382745265960693:  15%|█▍        | 959/6470 [13:50<1:05:02,  1.41it/s]
1|960|Loss: 1.2782924175262451:  15%|█▍        | 959/6470 [13:50<1:05:02,  1.41it/s]
1|960|Loss: 1.2782924175262451:  15%|█▍        | 960/6470 [13:50<1:02:02,  1.48it/s]
1|961|Loss: 0.8575905561447144:  15%|█▍        | 960/6470 [13:51<1:02:02,  1.48it/s]
1|961|Loss: 0.8575905561447144:  15%|█▍        | 961/6470 [13:51<1:00:00,  1.53it/s]
1|962|Loss: 1.0734682083129883:  15%|█▍        | 961/6470 [13:51<1:00:00,  1.53it/s]
1|962|Loss: 1.0734682083129883:  15%|█▍        | 962/6470 [13:51<58:01,  1.58it/s]  
1|963|Loss: 1.2200413942337036:  15%|█▍        | 962/6470 [13:52<58:01,  1.58it/s]
1|963|Loss: 1.2200413942337036:  15%|█▍        | 963/6470 [13:52<59:12,  1.55it/s]
1|964|Loss: 1.150681972503662:  15%|█▍        | 963/6470 [13:52<59:12,  1.55it/s] 
1|964|Loss: 1.150681972503662:  15%|█▍        | 964/6470 [13:53<1:00:49,  1.51it/s]
1|965|Loss: 0.9830031991004944:  15%|█▍        | 964/6470 [13:53<1:00:49,  1.51it/s]
1|965|Loss: 0.9830031991004944:  15%|█▍        | 965/6470 [13:54<1:03:49,  1.44it/s]
1|966|Loss: 1.0009649991989136:  15%|█▍        | 965/6470 [13:54<1:03:49,  1.44it/s]
1|966|Loss: 1.0009649991989136:  15%|█▍        | 966/6470 [13:54<1:05:15,  1.41it/s]
1|967|Loss: 1.0490119457244873:  15%|█▍        | 966/6470 [13:55<1:05:15,  1.41it/s]
1|967|Loss: 1.0490119457244873:  15%|█▍        | 967/6470 [13:55<1:07:10,  1.37it/s]
1|968|Loss: 1.3194806575775146:  15%|█▍        | 967/6470 [13:55<1:07:10,  1.37it/s]
1|968|Loss: 1.3194806575775146:  15%|█▍        | 968/6470 [13:56<1:02:15,  1.47it/s]
1|969|Loss: 0.957187294960022:  15%|█▍        | 968/6470 [13:56<1:02:15,  1.47it/s] 
1|969|Loss: 0.957187294960022:  15%|█▍        | 969/6470 [13:56<59:42,  1.54it/s]  
1|970|Loss: 1.0290367603302002:  15%|█▍        | 969/6470 [13:57<59:42,  1.54it/s]
1|970|Loss: 1.0290367603302002:  15%|█▍        | 970/6470 [13:57<1:00:27,  1.52it/s]
1|971|Loss: 1.1636805534362793:  15%|█▍        | 970/6470 [13:57<1:00:27,  1.52it/s]
1|971|Loss: 1.1636805534362793:  15%|█▌        | 971/6470 [13:58<1:02:11,  1.47it/s]
1|972|Loss: 1.0768628120422363:  15%|█▌        | 971/6470 [13:58<1:02:11,  1.47it/s]
1|972|Loss: 1.0768628120422363:  15%|█▌        | 972/6470 [13:58<1:04:50,  1.41it/s]
1|973|Loss: 0.9910776615142822:  15%|█▌        | 972/6470 [13:59<1:04:50,  1.41it/s]
1|973|Loss: 0.9910776615142822:  15%|█▌        | 973/6470 [13:59<1:02:49,  1.46it/s]
1|974|Loss: 1.052248477935791:  15%|█▌        | 973/6470 [13:59<1:02:49,  1.46it/s] 
1|974|Loss: 1.052248477935791:  15%|█▌        | 974/6470 [14:00<1:04:30,  1.42it/s]
1|975|Loss: 1.0589715242385864:  15%|█▌        | 974/6470 [14:00<1:04:30,  1.42it/s]
1|975|Loss: 1.0589715242385864:  15%|█▌        | 975/6470 [14:00<1:03:37,  1.44it/s]
1|976|Loss: 1.1511242389678955:  15%|█▌        | 975/6470 [14:01<1:03:37,  1.44it/s]
1|976|Loss: 1.1511242389678955:  15%|█▌        | 976/6470 [14:01<1:00:27,  1.51it/s]
1|977|Loss: 1.069651484489441:  15%|█▌        | 976/6470 [14:01<1:00:27,  1.51it/s] 
1|977|Loss: 1.069651484489441:  15%|█▌        | 977/6470 [14:02<58:25,  1.57it/s]  
1|978|Loss: 1.257338047027588:  15%|█▌        | 977/6470 [14:02<58:25,  1.57it/s]
1|978|Loss: 1.257338047027588:  15%|█▌        | 978/6470 [14:02<58:12,  1.57it/s]
1|979|Loss: 1.0867080688476562:  15%|█▌        | 978/6470 [14:03<58:12,  1.57it/s]
1|979|Loss: 1.0867080688476562:  15%|█▌        | 979/6470 [14:03<1:00:44,  1.51it/s]
1|980|Loss: 1.1170332431793213:  15%|█▌        | 979/6470 [14:03<1:00:44,  1.51it/s]
1|980|Loss: 1.1170332431793213:  15%|█▌        | 980/6470 [14:04<1:00:49,  1.50it/s]
1|981|Loss: 0.9703027606010437:  15%|█▌        | 980/6470 [14:04<1:00:49,  1.50it/s]
1|981|Loss: 0.9703027606010437:  15%|█▌        | 981/6470 [14:04<1:00:12,  1.52it/s]
1|982|Loss: 1.063031554222107:  15%|█▌        | 981/6470 [14:05<1:00:12,  1.52it/s] 
1|982|Loss: 1.063031554222107:  15%|█▌        | 982/6470 [14:05<55:50,  1.64it/s]  
1|983|Loss: 0.9075513482093811:  15%|█▌        | 982/6470 [14:05<55:50,  1.64it/s]
1|983|Loss: 0.9075513482093811:  15%|█▌        | 983/6470 [14:06<1:00:06,  1.52it/s]
1|984|Loss: 0.9490240216255188:  15%|█▌        | 983/6470 [14:06<1:00:06,  1.52it/s]
1|984|Loss: 0.9490240216255188:  15%|█▌        | 984/6470 [14:06<1:06:59,  1.36it/s]
1|985|Loss: 1.0755716562271118:  15%|█▌        | 984/6470 [14:07<1:06:59,  1.36it/s]
1|985|Loss: 1.0755716562271118:  15%|█▌        | 985/6470 [14:07<1:04:40,  1.41it/s]
1|986|Loss: 1.160711646080017:  15%|█▌        | 985/6470 [14:07<1:04:40,  1.41it/s] 
1|986|Loss: 1.160711646080017:  15%|█▌        | 986/6470 [14:08<1:02:28,  1.46it/s]
1|987|Loss: 1.0412073135375977:  15%|█▌        | 986/6470 [14:08<1:02:28,  1.46it/s]
1|987|Loss: 1.0412073135375977:  15%|█▌        | 987/6470 [14:08<1:03:44,  1.43it/s]
1|988|Loss: 1.0941849946975708:  15%|█▌        | 987/6470 [14:09<1:03:44,  1.43it/s]
1|988|Loss: 1.0941849946975708:  15%|█▌        | 988/6470 [14:09<1:05:08,  1.40it/s]
1|989|Loss: 1.2467230558395386:  15%|█▌        | 988/6470 [14:10<1:05:08,  1.40it/s]
1|989|Loss: 1.2467230558395386:  15%|█▌        | 989/6470 [14:10<1:03:54,  1.43it/s]
1|990|Loss: 1.1068845987319946:  15%|█▌        | 989/6470 [14:10<1:03:54,  1.43it/s]
1|990|Loss: 1.1068845987319946:  15%|█▌        | 990/6470 [14:11<1:04:13,  1.42it/s]
1|991|Loss: 1.02450430393219:  15%|█▌        | 990/6470 [14:11<1:04:13,  1.42it/s]  
1|991|Loss: 1.02450430393219:  15%|█▌        | 991/6470 [14:11<1:01:04,  1.50it/s]
1|992|Loss: 1.0344008207321167:  15%|█▌        | 991/6470 [14:11<1:01:04,  1.50it/s]
1|992|Loss: 1.0344008207321167:  15%|█▌        | 992/6470 [14:12<59:42,  1.53it/s]  
1|993|Loss: 1.115338683128357:  15%|█▌        | 992/6470 [14:12<59:42,  1.53it/s] 
1|993|Loss: 1.115338683128357:  15%|█▌        | 993/6470 [14:12<1:00:51,  1.50it/s]
1|994|Loss: 1.0867670774459839:  15%|█▌        | 993/6470 [14:13<1:00:51,  1.50it/s]
1|994|Loss: 1.0867670774459839:  15%|█▌        | 994/6470 [14:13<1:00:25,  1.51it/s]
1|995|Loss: 1.0999376773834229:  15%|█▌        | 994/6470 [14:13<1:00:25,  1.51it/s]
1|995|Loss: 1.0999376773834229:  15%|█▌        | 995/6470 [14:14<57:49,  1.58it/s]  
1|996|Loss: 0.7817753553390503:  15%|█▌        | 995/6470 [14:14<57:49,  1.58it/s]
1|996|Loss: 0.7817753553390503:  15%|█▌        | 996/6470 [14:14<58:51,  1.55it/s]
1|997|Loss: 1.201208233833313:  15%|█▌        | 996/6470 [14:15<58:51,  1.55it/s] 
1|997|Loss: 1.201208233833313:  15%|█▌        | 997/6470 [14:15<1:03:44,  1.43it/s]
1|998|Loss: 1.1135131120681763:  15%|█▌        | 997/6470 [14:16<1:03:44,  1.43it/s]
1|998|Loss: 1.1135131120681763:  15%|█▌        | 998/6470 [14:16<1:00:06,  1.52it/s]
1|999|Loss: 1.0603351593017578:  15%|█▌        | 998/6470 [14:16<1:00:06,  1.52it/s]
1|999|Loss: 1.0603351593017578:  15%|█▌        | 999/6470 [14:16<58:11,  1.57it/s]  
1|1000|Loss: 1.042678952217102:  15%|█▌        | 999/6470 [14:17<58:11,  1.57it/s]
1|1000|Loss: 1.042678952217102:  15%|█▌        | 1000/6470 [14:17<57:53,  1.57it/s]
1|1001|Loss: 1.0243862867355347:  15%|█▌        | 1000/6470 [14:17<57:53,  1.57it/s]
1|1001|Loss: 1.0243862867355347:  15%|█▌        | 1001/6470 [14:18<1:01:30,  1.48it/s]
1|1002|Loss: 1.0682642459869385:  15%|█▌        | 1001/6470 [14:18<1:01:30,  1.48it/s]
1|1002|Loss: 1.0682642459869385:  15%|█▌        | 1002/6470 [14:18<1:01:29,  1.48it/s]
1|1003|Loss: 1.2685452699661255:  15%|█▌        | 1002/6470 [14:19<1:01:29,  1.48it/s]
1|1003|Loss: 1.2685452699661255:  16%|█▌        | 1003/6470 [14:19<58:01,  1.57it/s]  
1|1004|Loss: 1.0185039043426514:  16%|█▌        | 1003/6470 [14:19<58:01,  1.57it/s]
1|1004|Loss: 1.0185039043426514:  16%|█▌        | 1004/6470 [14:20<56:27,  1.61it/s]
1|1005|Loss: 1.074981689453125:  16%|█▌        | 1004/6470 [14:20<56:27,  1.61it/s] 
1|1005|Loss: 1.074981689453125:  16%|█▌        | 1005/6470 [14:20<52:31,  1.73it/s]
1|1006|Loss: 1.0470560789108276:  16%|█▌        | 1005/6470 [14:20<52:31,  1.73it/s]
1|1006|Loss: 1.0470560789108276:  16%|█▌        | 1006/6470 [14:21<54:24,  1.67it/s]
1|1007|Loss: 1.227932095527649:  16%|█▌        | 1006/6470 [14:21<54:24,  1.67it/s] 
1|1007|Loss: 1.227932095527649:  16%|█▌        | 1007/6470 [14:21<55:23,  1.64it/s]
1|1008|Loss: 1.1261563301086426:  16%|█▌        | 1007/6470 [14:22<55:23,  1.64it/s]
1|1008|Loss: 1.1261563301086426:  16%|█▌        | 1008/6470 [14:22<54:33,  1.67it/s]
1|1009|Loss: 0.8357652425765991:  16%|█▌        | 1008/6470 [14:22<54:33,  1.67it/s]
1|1009|Loss: 0.8357652425765991:  16%|█▌        | 1009/6470 [14:23<59:25,  1.53it/s]
1|1010|Loss: 1.0835363864898682:  16%|█▌        | 1009/6470 [14:23<59:25,  1.53it/s]
1|1010|Loss: 1.0835363864898682:  16%|█▌        | 1010/6470 [14:23<58:31,  1.56it/s]
1|1011|Loss: 1.0440982580184937:  16%|█▌        | 1010/6470 [14:24<58:31,  1.56it/s]
1|1011|Loss: 1.0440982580184937:  16%|█▌        | 1011/6470 [14:24<1:03:31,  1.43it/s]
1|1012|Loss: 1.1608638763427734:  16%|█▌        | 1011/6470 [14:24<1:03:31,  1.43it/s]
1|1012|Loss: 1.1608638763427734:  16%|█▌        | 1012/6470 [14:25<1:02:17,  1.46it/s]
1|1013|Loss: 0.9657968878746033:  16%|█▌        | 1012/6470 [14:25<1:02:17,  1.46it/s]
1|1013|Loss: 0.9657968878746033:  16%|█▌        | 1013/6470 [14:25<1:03:14,  1.44it/s]
1|1014|Loss: 1.0329277515411377:  16%|█▌        | 1013/6470 [14:26<1:03:14,  1.44it/s]
1|1014|Loss: 1.0329277515411377:  16%|█▌        | 1014/6470 [14:26<1:01:54,  1.47it/s]
1|1015|Loss: 0.9632685780525208:  16%|█▌        | 1014/6470 [14:26<1:01:54,  1.47it/s]
1|1015|Loss: 0.9632685780525208:  16%|█▌        | 1015/6470 [14:27<1:00:45,  1.50it/s]
1|1016|Loss: 1.04743492603302:  16%|█▌        | 1015/6470 [14:27<1:00:45,  1.50it/s]  
1|1016|Loss: 1.04743492603302:  16%|█▌        | 1016/6470 [14:28<1:03:48,  1.42it/s]
1|1017|Loss: 1.0908840894699097:  16%|█▌        | 1016/6470 [14:28<1:03:48,  1.42it/s]
1|1017|Loss: 1.0908840894699097:  16%|█▌        | 1017/6470 [14:28<1:08:49,  1.32it/s]
1|1018|Loss: 0.9473347067832947:  16%|█▌        | 1017/6470 [14:29<1:08:49,  1.32it/s]
1|1018|Loss: 0.9473347067832947:  16%|█▌        | 1018/6470 [14:29<1:10:26,  1.29it/s]
1|1019|Loss: 0.9475164413452148:  16%|█▌        | 1018/6470 [14:30<1:10:26,  1.29it/s]
1|1019|Loss: 0.9475164413452148:  16%|█▌        | 1019/6470 [14:30<1:07:46,  1.34it/s]
1|1020|Loss: 1.0641177892684937:  16%|█▌        | 1019/6470 [14:30<1:07:46,  1.34it/s]
1|1020|Loss: 1.0641177892684937:  16%|█▌        | 1020/6470 [14:31<1:03:37,  1.43it/s]
1|1021|Loss: 1.0476783514022827:  16%|█▌        | 1020/6470 [14:31<1:03:37,  1.43it/s]
1|1021|Loss: 1.0476783514022827:  16%|█▌        | 1021/6470 [14:31<1:01:21,  1.48it/s]
1|1022|Loss: 0.8967710733413696:  16%|█▌        | 1021/6470 [14:32<1:01:21,  1.48it/s]
1|1022|Loss: 0.8967710733413696:  16%|█▌        | 1022/6470 [14:32<1:03:34,  1.43it/s]
1|1023|Loss: 0.9944382905960083:  16%|█▌        | 1022/6470 [14:32<1:03:34,  1.43it/s]
1|1023|Loss: 0.9944382905960083:  16%|█▌        | 1023/6470 [14:33<1:02:25,  1.45it/s]
1|1024|Loss: 1.1898835897445679:  16%|█▌        | 1023/6470 [14:33<1:02:25,  1.45it/s]
1|1024|Loss: 1.1898835897445679:  16%|█▌        | 1024/6470 [14:33<1:02:15,  1.46it/s]
1|1025|Loss: 0.943461000919342:  16%|█▌        | 1024/6470 [14:34<1:02:15,  1.46it/s] 
1|1025|Loss: 0.943461000919342:  16%|█▌        | 1025/6470 [14:34<1:02:36,  1.45it/s]
1|1026|Loss: 1.0617258548736572:  16%|█▌        | 1025/6470 [14:34<1:02:36,  1.45it/s]
1|1026|Loss: 1.0617258548736572:  16%|█▌        | 1026/6470 [14:35<1:01:35,  1.47it/s]
1|1027|Loss: 1.157589077949524:  16%|█▌        | 1026/6470 [14:35<1:01:35,  1.47it/s] 
1|1027|Loss: 1.157589077949524:  16%|█▌        | 1027/6470 [14:35<1:00:28,  1.50it/s]
1|1028|Loss: 0.8976144194602966:  16%|█▌        | 1027/6470 [14:36<1:00:28,  1.50it/s]
1|1028|Loss: 0.8976144194602966:  16%|█▌        | 1028/6470 [14:36<1:04:44,  1.40it/s]
1|1029|Loss: 1.107070803642273:  16%|█▌        | 1028/6470 [14:36<1:04:44,  1.40it/s] 
1|1029|Loss: 1.107070803642273:  16%|█▌        | 1029/6470 [14:37<1:03:04,  1.44it/s]
1|1030|Loss: 0.9805428385734558:  16%|█▌        | 1029/6470 [14:37<1:03:04,  1.44it/s]
1|1030|Loss: 0.9805428385734558:  16%|█▌        | 1030/6470 [14:37<58:31,  1.55it/s]  
1|1031|Loss: 1.1816056966781616:  16%|█▌        | 1030/6470 [14:38<58:31,  1.55it/s]
1|1031|Loss: 1.1816056966781616:  16%|█▌        | 1031/6470 [14:38<56:36,  1.60it/s]
1|1032|Loss: 1.1296693086624146:  16%|█▌        | 1031/6470 [14:38<56:36,  1.60it/s]
1|1032|Loss: 1.1296693086624146:  16%|█▌        | 1032/6470 [14:39<1:02:04,  1.46it/s]
1|1033|Loss: 1.1429848670959473:  16%|█▌        | 1032/6470 [14:39<1:02:04,  1.46it/s]
1|1033|Loss: 1.1429848670959473:  16%|█▌        | 1033/6470 [14:39<1:04:34,  1.40it/s]
1|1034|Loss: 1.048328161239624:  16%|█▌        | 1033/6470 [14:40<1:04:34,  1.40it/s] 
1|1034|Loss: 1.048328161239624:  16%|█▌        | 1034/6470 [14:40<1:01:52,  1.46it/s]
1|1035|Loss: 1.1850007772445679:  16%|█▌        | 1034/6470 [14:40<1:01:52,  1.46it/s]
1|1035|Loss: 1.1850007772445679:  16%|█▌        | 1035/6470 [14:41<1:03:41,  1.42it/s]
1|1036|Loss: 0.8796418905258179:  16%|█▌        | 1035/6470 [14:41<1:03:41,  1.42it/s]
1|1036|Loss: 0.8796418905258179:  16%|█▌        | 1036/6470 [14:42<1:04:56,  1.39it/s]
1|1037|Loss: 1.1277716159820557:  16%|█▌        | 1036/6470 [14:42<1:04:56,  1.39it/s]
1|1037|Loss: 1.1277716159820557:  16%|█▌        | 1037/6470 [14:42<1:02:56,  1.44it/s]
1|1038|Loss: 1.1349616050720215:  16%|█▌        | 1037/6470 [14:43<1:02:56,  1.44it/s]
1|1038|Loss: 1.1349616050720215:  16%|█▌        | 1038/6470 [14:43<1:04:21,  1.41it/s]
1|1039|Loss: 0.8550883531570435:  16%|█▌        | 1038/6470 [14:43<1:04:21,  1.41it/s]
1|1039|Loss: 0.8550883531570435:  16%|█▌        | 1039/6470 [14:44<1:01:50,  1.46it/s]
1|1040|Loss: 1.1315593719482422:  16%|█▌        | 1039/6470 [14:44<1:01:50,  1.46it/s]
1|1040|Loss: 1.1315593719482422:  16%|█▌        | 1040/6470 [14:44<1:04:25,  1.40it/s]
1|1041|Loss: 1.0063499212265015:  16%|█▌        | 1040/6470 [14:45<1:04:25,  1.40it/s]
1|1041|Loss: 1.0063499212265015:  16%|█▌        | 1041/6470 [14:45<1:05:46,  1.38it/s]
1|1042|Loss: 0.941787838935852:  16%|█▌        | 1041/6470 [14:45<1:05:46,  1.38it/s] 
1|1042|Loss: 0.941787838935852:  16%|█▌        | 1042/6470 [14:46<1:05:38,  1.38it/s]
1|1043|Loss: 1.0415775775909424:  16%|█▌        | 1042/6470 [14:46<1:05:38,  1.38it/s]
1|1043|Loss: 1.0415775775909424:  16%|█▌        | 1043/6470 [14:47<1:08:31,  1.32it/s]
1|1044|Loss: 1.0539336204528809:  16%|█▌        | 1043/6470 [14:47<1:08:31,  1.32it/s]
1|1044|Loss: 1.0539336204528809:  16%|█▌        | 1044/6470 [14:47<1:06:10,  1.37it/s]
1|1045|Loss: 0.9591525793075562:  16%|█▌        | 1044/6470 [14:48<1:06:10,  1.37it/s]
1|1045|Loss: 0.9591525793075562:  16%|█▌        | 1045/6470 [14:48<1:03:34,  1.42it/s]
1|1046|Loss: 1.037527084350586:  16%|█▌        | 1045/6470 [14:48<1:03:34,  1.42it/s] 
1|1046|Loss: 1.037527084350586:  16%|█▌        | 1046/6470 [14:49<1:00:33,  1.49it/s]
1|1047|Loss: 1.1109280586242676:  16%|█▌        | 1046/6470 [14:49<1:00:33,  1.49it/s]
1|1047|Loss: 1.1109280586242676:  16%|█▌        | 1047/6470 [14:49<1:01:40,  1.47it/s]
1|1048|Loss: 1.0117744207382202:  16%|█▌        | 1047/6470 [14:50<1:01:40,  1.47it/s]
1|1048|Loss: 1.0117744207382202:  16%|█▌        | 1048/6470 [14:50<1:02:33,  1.44it/s]
1|1049|Loss: 1.16339910030365:  16%|█▌        | 1048/6470 [14:50<1:02:33,  1.44it/s]  
1|1049|Loss: 1.16339910030365:  16%|█▌        | 1049/6470 [14:51<1:01:03,  1.48it/s]
1|1050|Loss: 1.0579789876937866:  16%|█▌        | 1049/6470 [14:51<1:01:03,  1.48it/s]
1|1050|Loss: 1.0579789876937866:  16%|█▌        | 1050/6470 [14:51<1:01:09,  1.48it/s]
1|1051|Loss: 0.9202536344528198:  16%|█▌        | 1050/6470 [14:52<1:01:09,  1.48it/s]
1|1051|Loss: 0.9202536344528198:  16%|█▌        | 1051/6470 [14:52<1:07:18,  1.34it/s]
1|1052|Loss: 0.9963213205337524:  16%|█▌        | 1051/6470 [14:53<1:07:18,  1.34it/s]
1|1052|Loss: 0.9963213205337524:  16%|█▋        | 1052/6470 [14:53<1:07:47,  1.33it/s]
1|1053|Loss: 1.2257786989212036:  16%|█▋        | 1052/6470 [14:53<1:07:47,  1.33it/s]
1|1053|Loss: 1.2257786989212036:  16%|█▋        | 1053/6470 [14:53<1:00:55,  1.48it/s]
1|1054|Loss: 1.0961140394210815:  16%|█▋        | 1053/6470 [14:54<1:00:55,  1.48it/s]
1|1054|Loss: 1.0961140394210815:  16%|█▋        | 1054/6470 [14:54<1:00:46,  1.49it/s]
1|1055|Loss: 1.0010536909103394:  16%|█▋        | 1054/6470 [14:54<1:00:46,  1.49it/s]
1|1055|Loss: 1.0010536909103394:  16%|█▋        | 1055/6470 [14:55<59:56,  1.51it/s]  
1|1056|Loss: 1.0883105993270874:  16%|█▋        | 1055/6470 [14:55<59:56,  1.51it/s]
1|1056|Loss: 1.0883105993270874:  16%|█▋        | 1056/6470 [14:55<59:08,  1.53it/s]
1|1057|Loss: 1.0937068462371826:  16%|█▋        | 1056/6470 [14:56<59:08,  1.53it/s]
1|1057|Loss: 1.0937068462371826:  16%|█▋        | 1057/6470 [14:56<55:15,  1.63it/s]
1|1058|Loss: 1.1066951751708984:  16%|█▋        | 1057/6470 [14:56<55:15,  1.63it/s]
1|1058|Loss: 1.1066951751708984:  16%|█▋        | 1058/6470 [14:56<54:28,  1.66it/s]
1|1059|Loss: 0.9114157557487488:  16%|█▋        | 1058/6470 [14:57<54:28,  1.66it/s]
1|1059|Loss: 0.9114157557487488:  16%|█▋        | 1059/6470 [14:57<54:40,  1.65it/s]
1|1060|Loss: 0.9230543375015259:  16%|█▋        | 1059/6470 [14:57<54:40,  1.65it/s]
1|1060|Loss: 0.9230543375015259:  16%|█▋        | 1060/6470 [14:58<55:32,  1.62it/s]
1|1061|Loss: 1.0085591077804565:  16%|█▋        | 1060/6470 [14:58<55:32,  1.62it/s]
1|1061|Loss: 1.0085591077804565:  16%|█▋        | 1061/6470 [14:59<1:00:23,  1.49it/s]
1|1062|Loss: 0.9890912175178528:  16%|█▋        | 1061/6470 [14:59<1:00:23,  1.49it/s]
1|1062|Loss: 0.9890912175178528:  16%|█▋        | 1062/6470 [14:59<1:05:52,  1.37it/s]
1|1063|Loss: 1.0703641176223755:  16%|█▋        | 1062/6470 [15:00<1:05:52,  1.37it/s]
1|1063|Loss: 1.0703641176223755:  16%|█▋        | 1063/6470 [15:00<1:10:02,  1.29it/s]
1|1064|Loss: 1.253926396369934:  16%|█▋        | 1063/6470 [15:01<1:10:02,  1.29it/s] 
1|1064|Loss: 1.253926396369934:  16%|█▋        | 1064/6470 [15:01<1:05:26,  1.38it/s]
1|1065|Loss: 0.9756343364715576:  16%|█▋        | 1064/6470 [15:01<1:05:26,  1.38it/s]
1|1065|Loss: 0.9756343364715576:  16%|█▋        | 1065/6470 [15:01<1:02:03,  1.45it/s]
1|1066|Loss: 1.025640845298767:  16%|█▋        | 1065/6470 [15:02<1:02:03,  1.45it/s] 
1|1066|Loss: 1.025640845298767:  16%|█▋        | 1066/6470 [15:02<1:00:46,  1.48it/s]
1|1067|Loss: 0.725438117980957:  16%|█▋        | 1066/6470 [15:03<1:00:46,  1.48it/s]
1|1067|Loss: 0.725438117980957:  16%|█▋        | 1067/6470 [15:03<1:05:05,  1.38it/s]
1|1068|Loss: 1.0870436429977417:  16%|█▋        | 1067/6470 [15:03<1:05:05,  1.38it/s]
1|1068|Loss: 1.0870436429977417:  17%|█▋        | 1068/6470 [15:04<1:02:04,  1.45it/s]
1|1069|Loss: 1.1130627393722534:  17%|█▋        | 1068/6470 [15:04<1:02:04,  1.45it/s]
1|1069|Loss: 1.1130627393722534:  17%|█▋        | 1069/6470 [15:04<58:44,  1.53it/s]  
1|1070|Loss: 1.0230252742767334:  17%|█▋        | 1069/6470 [15:04<58:44,  1.53it/s]
1|1070|Loss: 1.0230252742767334:  17%|█▋        | 1070/6470 [15:05<58:02,  1.55it/s]
1|1071|Loss: 0.9919255971908569:  17%|█▋        | 1070/6470 [15:05<58:02,  1.55it/s]
1|1071|Loss: 0.9919255971908569:  17%|█▋        | 1071/6470 [15:05<56:05,  1.60it/s]
1|1072|Loss: 1.1146938800811768:  17%|█▋        | 1071/6470 [15:06<56:05,  1.60it/s]
1|1072|Loss: 1.1146938800811768:  17%|█▋        | 1072/6470 [15:06<56:08,  1.60it/s]
1|1073|Loss: 0.8965005874633789:  17%|█▋        | 1072/6470 [15:06<56:08,  1.60it/s]
1|1073|Loss: 0.8965005874633789:  17%|█▋        | 1073/6470 [15:07<57:56,  1.55it/s]
1|1074|Loss: 1.0663214921951294:  17%|█▋        | 1073/6470 [15:07<57:56,  1.55it/s]
1|1074|Loss: 1.0663214921951294:  17%|█▋        | 1074/6470 [15:07<56:53,  1.58it/s]
1|1075|Loss: 0.8838537931442261:  17%|█▋        | 1074/6470 [15:08<56:53,  1.58it/s]
1|1075|Loss: 0.8838537931442261:  17%|█▋        | 1075/6470 [15:08<56:55,  1.58it/s]
1|1076|Loss: 0.967423141002655:  17%|█▋        | 1075/6470 [15:08<56:55,  1.58it/s] 
1|1076|Loss: 0.967423141002655:  17%|█▋        | 1076/6470 [15:09<59:33,  1.51it/s]
1|1077|Loss: 1.2635711431503296:  17%|█▋        | 1076/6470 [15:09<59:33,  1.51it/s]
1|1077|Loss: 1.2635711431503296:  17%|█▋        | 1077/6470 [15:09<59:42,  1.51it/s]
1|1078|Loss: 1.0644168853759766:  17%|█▋        | 1077/6470 [15:10<59:42,  1.51it/s]
1|1078|Loss: 1.0644168853759766:  17%|█▋        | 1078/6470 [15:10<1:12:37,  1.24it/s]
1|1079|Loss: 1.1315053701400757:  17%|█▋        | 1078/6470 [15:11<1:12:37,  1.24it/s]
1|1079|Loss: 1.1315053701400757:  17%|█▋        | 1079/6470 [15:11<1:08:48,  1.31it/s]
1|1080|Loss: 0.9028557538986206:  17%|█▋        | 1079/6470 [15:11<1:08:48,  1.31it/s]
1|1080|Loss: 0.9028557538986206:  17%|█▋        | 1080/6470 [15:12<1:06:06,  1.36it/s]
1|1081|Loss: 0.8514738082885742:  17%|█▋        | 1080/6470 [15:12<1:06:06,  1.36it/s]
1|1081|Loss: 0.8514738082885742:  17%|█▋        | 1081/6470 [15:12<1:03:58,  1.40it/s]
1|1082|Loss: 1.2285223007202148:  17%|█▋        | 1081/6470 [15:13<1:03:58,  1.40it/s]
1|1082|Loss: 1.2285223007202148:  17%|█▋        | 1082/6470 [15:13<1:06:28,  1.35it/s]
1|1083|Loss: 1.013617992401123:  17%|█▋        | 1082/6470 [15:14<1:06:28,  1.35it/s] 
1|1083|Loss: 1.013617992401123:  17%|█▋        | 1083/6470 [15:14<1:07:03,  1.34it/s]
1|1084|Loss: 1.0779482126235962:  17%|█▋        | 1083/6470 [15:14<1:07:03,  1.34it/s]
1|1084|Loss: 1.0779482126235962:  17%|█▋        | 1084/6470 [15:15<1:06:10,  1.36it/s]
1|1085|Loss: 1.2895479202270508:  17%|█▋        | 1084/6470 [15:15<1:06:10,  1.36it/s]
1|1085|Loss: 1.2895479202270508:  17%|█▋        | 1085/6470 [15:15<1:03:53,  1.40it/s]
1|1086|Loss: 1.2332513332366943:  17%|█▋        | 1085/6470 [15:16<1:03:53,  1.40it/s]
1|1086|Loss: 1.2332513332366943:  17%|█▋        | 1086/6470 [15:16<1:00:03,  1.49it/s]
1|1087|Loss: 1.1122628450393677:  17%|█▋        | 1086/6470 [15:16<1:00:03,  1.49it/s]
1|1087|Loss: 1.1122628450393677:  17%|█▋        | 1087/6470 [15:17<1:00:49,  1.47it/s]
1|1088|Loss: 0.9596121311187744:  17%|█▋        | 1087/6470 [15:17<1:00:49,  1.47it/s]
1|1088|Loss: 0.9596121311187744:  17%|█▋        | 1088/6470 [15:17<1:02:21,  1.44it/s]
1|1089|Loss: 1.1199009418487549:  17%|█▋        | 1088/6470 [15:18<1:02:21,  1.44it/s]
1|1089|Loss: 1.1199009418487549:  17%|█▋        | 1089/6470 [15:18<58:04,  1.54it/s]  
1|1090|Loss: 1.193495512008667:  17%|█▋        | 1089/6470 [15:18<58:04,  1.54it/s] 
1|1090|Loss: 1.193495512008667:  17%|█▋        | 1090/6470 [15:19<58:12,  1.54it/s]
1|1091|Loss: 1.0409873723983765:  17%|█▋        | 1090/6470 [15:19<58:12,  1.54it/s]
1|1091|Loss: 1.0409873723983765:  17%|█▋        | 1091/6470 [15:19<1:02:52,  1.43it/s]
1|1092|Loss: 1.0345934629440308:  17%|█▋        | 1091/6470 [15:20<1:02:52,  1.43it/s]
1|1092|Loss: 1.0345934629440308:  17%|█▋        | 1092/6470 [15:20<1:04:02,  1.40it/s]
1|1093|Loss: 1.2754281759262085:  17%|█▋        | 1092/6470 [15:20<1:04:02,  1.40it/s]
1|1093|Loss: 1.2754281759262085:  17%|█▋        | 1093/6470 [15:21<59:39,  1.50it/s]  
1|1094|Loss: 0.9681150317192078:  17%|█▋        | 1093/6470 [15:21<59:39,  1.50it/s]
1|1094|Loss: 0.9681150317192078:  17%|█▋        | 1094/6470 [15:21<56:17,  1.59it/s]
1|1095|Loss: 1.1857154369354248:  17%|█▋        | 1094/6470 [15:22<56:17,  1.59it/s]
1|1095|Loss: 1.1857154369354248:  17%|█▋        | 1095/6470 [15:22<52:07,  1.72it/s]
1|1096|Loss: 1.0797991752624512:  17%|█▋        | 1095/6470 [15:22<52:07,  1.72it/s]
1|1096|Loss: 1.0797991752624512:  17%|█▋        | 1096/6470 [15:22<57:06,  1.57it/s]
1|1097|Loss: 1.0127811431884766:  17%|█▋        | 1096/6470 [15:23<57:06,  1.57it/s]
1|1097|Loss: 1.0127811431884766:  17%|█▋        | 1097/6470 [15:23<1:01:49,  1.45it/s]
1|1098|Loss: 1.0063765048980713:  17%|█▋        | 1097/6470 [15:24<1:01:49,  1.45it/s]
1|1098|Loss: 1.0063765048980713:  17%|█▋        | 1098/6470 [15:24<58:38,  1.53it/s]  
1|1099|Loss: 0.9278144836425781:  17%|█▋        | 1098/6470 [15:24<58:38,  1.53it/s]
1|1099|Loss: 0.9278144836425781:  17%|█▋        | 1099/6470 [15:24<57:40,  1.55it/s]
1|1100|Loss: 1.1079175472259521:  17%|█▋        | 1099/6470 [15:25<57:40,  1.55it/s]
1|1100|Loss: 1.1079175472259521:  17%|█▋        | 1100/6470 [15:25<57:39,  1.55it/s]
1|1101|Loss: 0.975405752658844:  17%|█▋        | 1100/6470 [15:25<57:39,  1.55it/s] 
1|1101|Loss: 0.975405752658844:  17%|█▋        | 1101/6470 [15:26<58:48,  1.52it/s]
1|1102|Loss: 1.0267329216003418:  17%|█▋        | 1101/6470 [15:26<58:48,  1.52it/s]
1|1102|Loss: 1.0267329216003418:  17%|█▋        | 1102/6470 [15:26<58:42,  1.52it/s]
1|1103|Loss: 1.0568568706512451:  17%|█▋        | 1102/6470 [15:27<58:42,  1.52it/s]
1|1103|Loss: 1.0568568706512451:  17%|█▋        | 1103/6470 [15:27<57:49,  1.55it/s]
1|1104|Loss: 0.9360050559043884:  17%|█▋        | 1103/6470 [15:27<57:49,  1.55it/s]
1|1104|Loss: 0.9360050559043884:  17%|█▋        | 1104/6470 [15:28<59:30,  1.50it/s]
1|1105|Loss: 0.9775288701057434:  17%|█▋        | 1104/6470 [15:28<59:30,  1.50it/s]
1|1105|Loss: 0.9775288701057434:  17%|█▋        | 1105/6470 [15:28<57:32,  1.55it/s]
1|1106|Loss: 0.9660511612892151:  17%|█▋        | 1105/6470 [15:29<57:32,  1.55it/s]
1|1106|Loss: 0.9660511612892151:  17%|█▋        | 1106/6470 [15:29<1:00:12,  1.48it/s]
1|1107|Loss: 1.071194052696228:  17%|█▋        | 1106/6470 [15:30<1:00:12,  1.48it/s] 
1|1107|Loss: 1.071194052696228:  17%|█▋        | 1107/6470 [15:30<59:52,  1.49it/s]  
1|1108|Loss: 1.0633471012115479:  17%|█▋        | 1107/6470 [15:30<59:52,  1.49it/s]
1|1108|Loss: 1.0633471012115479:  17%|█▋        | 1108/6470 [15:30<57:50,  1.54it/s]
1|1109|Loss: 1.0503593683242798:  17%|█▋        | 1108/6470 [15:31<57:50,  1.54it/s]
1|1109|Loss: 1.0503593683242798:  17%|█▋        | 1109/6470 [15:31<57:27,  1.56it/s]
1|1110|Loss: 0.9880589842796326:  17%|█▋        | 1109/6470 [15:31<57:27,  1.56it/s]
1|1110|Loss: 0.9880589842796326:  17%|█▋        | 1110/6470 [15:32<58:40,  1.52it/s]
1|1111|Loss: 0.9097805023193359:  17%|█▋        | 1110/6470 [15:32<58:40,  1.52it/s]
1|1111|Loss: 0.9097805023193359:  17%|█▋        | 1111/6470 [15:32<1:00:36,  1.47it/s]
1|1112|Loss: 0.9280546307563782:  17%|█▋        | 1111/6470 [15:33<1:00:36,  1.47it/s]
1|1112|Loss: 0.9280546307563782:  17%|█▋        | 1112/6470 [15:33<1:02:03,  1.44it/s]
1|1113|Loss: 0.8770710229873657:  17%|█▋        | 1112/6470 [15:34<1:02:03,  1.44it/s]
1|1113|Loss: 0.8770710229873657:  17%|█▋        | 1113/6470 [15:34<1:06:08,  1.35it/s]
1|1114|Loss: 0.9433586001396179:  17%|█▋        | 1113/6470 [15:34<1:06:08,  1.35it/s]
1|1114|Loss: 0.9433586001396179:  17%|█▋        | 1114/6470 [15:35<1:08:47,  1.30it/s]
1|1115|Loss: 0.9918433427810669:  17%|█▋        | 1114/6470 [15:35<1:08:47,  1.30it/s]
1|1115|Loss: 0.9918433427810669:  17%|█▋        | 1115/6470 [15:36<1:07:30,  1.32it/s]
1|1116|Loss: 1.083701729774475:  17%|█▋        | 1115/6470 [15:36<1:07:30,  1.32it/s] 
1|1116|Loss: 1.083701729774475:  17%|█▋        | 1116/6470 [15:36<1:03:21,  1.41it/s]
1|1117|Loss: 1.147010326385498:  17%|█▋        | 1116/6470 [15:37<1:03:21,  1.41it/s]
1|1117|Loss: 1.147010326385498:  17%|█▋        | 1117/6470 [15:37<1:01:17,  1.46it/s]
1|1118|Loss: 1.111899733543396:  17%|█▋        | 1117/6470 [15:37<1:01:17,  1.46it/s]
1|1118|Loss: 1.111899733543396:  17%|█▋        | 1118/6470 [15:37<58:11,  1.53it/s]  
1|1119|Loss: 1.1918028593063354:  17%|█▋        | 1118/6470 [15:38<58:11,  1.53it/s]
1|1119|Loss: 1.1918028593063354:  17%|█▋        | 1119/6470 [15:38<54:34,  1.63it/s]
1|1120|Loss: 0.9618305563926697:  17%|█▋        | 1119/6470 [15:38<54:34,  1.63it/s]
1|1120|Loss: 0.9618305563926697:  17%|█▋        | 1120/6470 [15:39<56:57,  1.57it/s]
1|1121|Loss: 1.1164755821228027:  17%|█▋        | 1120/6470 [15:39<56:57,  1.57it/s]
1|1121|Loss: 1.1164755821228027:  17%|█▋        | 1121/6470 [15:39<57:16,  1.56it/s]
1|1122|Loss: 1.1633780002593994:  17%|█▋        | 1121/6470 [15:40<57:16,  1.56it/s]
1|1122|Loss: 1.1633780002593994:  17%|█▋        | 1122/6470 [15:40<54:35,  1.63it/s]
1|1123|Loss: 1.0686006546020508:  17%|█▋        | 1122/6470 [15:40<54:35,  1.63it/s]
1|1123|Loss: 1.0686006546020508:  17%|█▋        | 1123/6470 [15:40<55:53,  1.59it/s]
1|1124|Loss: 1.1228711605072021:  17%|█▋        | 1123/6470 [15:41<55:53,  1.59it/s]
1|1124|Loss: 1.1228711605072021:  17%|█▋        | 1124/6470 [15:41<54:45,  1.63it/s]
1|1125|Loss: 0.9716393351554871:  17%|█▋        | 1124/6470 [15:41<54:45,  1.63it/s]
1|1125|Loss: 0.9716393351554871:  17%|█▋        | 1125/6470 [15:42<56:23,  1.58it/s]
1|1126|Loss: 1.0792207717895508:  17%|█▋        | 1125/6470 [15:42<56:23,  1.58it/s]
1|1126|Loss: 1.0792207717895508:  17%|█▋        | 1126/6470 [15:42<58:24,  1.52it/s]
1|1127|Loss: 1.0242856740951538:  17%|█▋        | 1126/6470 [15:43<58:24,  1.52it/s]
1|1127|Loss: 1.0242856740951538:  17%|█▋        | 1127/6470 [15:43<57:55,  1.54it/s]
1|1128|Loss: 1.0860174894332886:  17%|█▋        | 1127/6470 [15:43<57:55,  1.54it/s]
1|1128|Loss: 1.0860174894332886:  17%|█▋        | 1128/6470 [15:44<56:55,  1.56it/s]
1|1129|Loss: 1.005293369293213:  17%|█▋        | 1128/6470 [15:44<56:55,  1.56it/s] 
1|1129|Loss: 1.005293369293213:  17%|█▋        | 1129/6470 [15:44<56:43,  1.57it/s]
1|1130|Loss: 0.8970772624015808:  17%|█▋        | 1129/6470 [15:45<56:43,  1.57it/s]
1|1130|Loss: 0.8970772624015808:  17%|█▋        | 1130/6470 [15:45<58:02,  1.53it/s]
1|1131|Loss: 1.084214210510254:  17%|█▋        | 1130/6470 [15:45<58:02,  1.53it/s] 
1|1131|Loss: 1.084214210510254:  17%|█▋        | 1131/6470 [15:46<58:01,  1.53it/s]
1|1132|Loss: 0.967593252658844:  17%|█▋        | 1131/6470 [15:46<58:01,  1.53it/s]
1|1132|Loss: 0.967593252658844:  17%|█▋        | 1132/6470 [15:46<1:00:09,  1.48it/s]
1|1133|Loss: 1.0228705406188965:  17%|█▋        | 1132/6470 [15:47<1:00:09,  1.48it/s]
1|1133|Loss: 1.0228705406188965:  18%|█▊        | 1133/6470 [15:47<1:02:19,  1.43it/s]
1|1134|Loss: 0.870646595954895:  18%|█▊        | 1133/6470 [15:48<1:02:19,  1.43it/s] 
1|1134|Loss: 0.870646595954895:  18%|█▊        | 1134/6470 [15:48<1:08:09,  1.30it/s]
1|1135|Loss: 1.0964224338531494:  18%|█▊        | 1134/6470 [15:48<1:08:09,  1.30it/s]
1|1135|Loss: 1.0964224338531494:  18%|█▊        | 1135/6470 [15:49<1:06:34,  1.34it/s]
1|1136|Loss: 1.1170862913131714:  18%|█▊        | 1135/6470 [15:49<1:06:34,  1.34it/s]
1|1136|Loss: 1.1170862913131714:  18%|█▊        | 1136/6470 [15:49<1:03:33,  1.40it/s]
1|1137|Loss: 0.9823451638221741:  18%|█▊        | 1136/6470 [15:50<1:03:33,  1.40it/s]
1|1137|Loss: 0.9823451638221741:  18%|█▊        | 1137/6470 [15:50<1:03:12,  1.41it/s]
1|1138|Loss: 1.0515040159225464:  18%|█▊        | 1137/6470 [15:50<1:03:12,  1.41it/s]
1|1138|Loss: 1.0515040159225464:  18%|█▊        | 1138/6470 [15:51<1:01:16,  1.45it/s]
1|1139|Loss: 0.9770782589912415:  18%|█▊        | 1138/6470 [15:51<1:01:16,  1.45it/s]
1|1139|Loss: 0.9770782589912415:  18%|█▊        | 1139/6470 [15:51<1:01:02,  1.46it/s]
1|1140|Loss: 0.9936997890472412:  18%|█▊        | 1139/6470 [15:52<1:01:02,  1.46it/s]
1|1140|Loss: 0.9936997890472412:  18%|█▊        | 1140/6470 [15:52<57:36,  1.54it/s]  
1|1141|Loss: 0.8933444619178772:  18%|█▊        | 1140/6470 [15:52<57:36,  1.54it/s]
1|1141|Loss: 0.8933444619178772:  18%|█▊        | 1141/6470 [15:53<56:14,  1.58it/s]
1|1142|Loss: 1.2158781290054321:  18%|█▊        | 1141/6470 [15:53<56:14,  1.58it/s]
1|1142|Loss: 1.2158781290054321:  18%|█▊        | 1142/6470 [15:53<54:44,  1.62it/s]
1|1143|Loss: 1.1051586866378784:  18%|█▊        | 1142/6470 [15:53<54:44,  1.62it/s]
1|1143|Loss: 1.1051586866378784:  18%|█▊        | 1143/6470 [15:54<49:34,  1.79it/s]
1|1144|Loss: 0.9562793374061584:  18%|█▊        | 1143/6470 [15:54<49:34,  1.79it/s]
1|1144|Loss: 0.9562793374061584:  18%|█▊        | 1144/6470 [15:54<51:52,  1.71it/s]
1|1145|Loss: 1.0401995182037354:  18%|█▊        | 1144/6470 [15:55<51:52,  1.71it/s]
1|1145|Loss: 1.0401995182037354:  18%|█▊        | 1145/6470 [15:55<53:11,  1.67it/s]
1|1146|Loss: 1.1253783702850342:  18%|█▊        | 1145/6470 [15:55<53:11,  1.67it/s]
1|1146|Loss: 1.1253783702850342:  18%|█▊        | 1146/6470 [15:56<55:59,  1.58it/s]
1|1147|Loss: 1.123337745666504:  18%|█▊        | 1146/6470 [15:56<55:59,  1.58it/s] 
1|1147|Loss: 1.123337745666504:  18%|█▊        | 1147/6470 [15:56<1:00:04,  1.48it/s]
1|1148|Loss: 1.0467448234558105:  18%|█▊        | 1147/6470 [15:57<1:00:04,  1.48it/s]
1|1148|Loss: 1.0467448234558105:  18%|█▊        | 1148/6470 [15:57<1:01:20,  1.45it/s]
1|1149|Loss: 0.9152586460113525:  18%|█▊        | 1148/6470 [15:57<1:01:20,  1.45it/s]
1|1149|Loss: 0.9152586460113525:  18%|█▊        | 1149/6470 [15:58<1:01:29,  1.44it/s]
1|1150|Loss: 1.3233531713485718:  18%|█▊        | 1149/6470 [15:58<1:01:29,  1.44it/s]
1|1150|Loss: 1.3233531713485718:  18%|█▊        | 1150/6470 [15:59<1:14:30,  1.19it/s]
1|1151|Loss: 1.0335294008255005:  18%|█▊        | 1150/6470 [15:59<1:14:30,  1.19it/s]
1|1151|Loss: 1.0335294008255005:  18%|█▊        | 1151/6470 [16:00<1:14:07,  1.20it/s]
1|1152|Loss: 0.986289381980896:  18%|█▊        | 1151/6470 [16:00<1:14:07,  1.20it/s] 
1|1152|Loss: 0.986289381980896:  18%|█▊        | 1152/6470 [16:01<1:13:44,  1.20it/s]
1|1153|Loss: 1.1155781745910645:  18%|█▊        | 1152/6470 [16:01<1:13:44,  1.20it/s]
1|1153|Loss: 1.1155781745910645:  18%|█▊        | 1153/6470 [16:01<1:08:33,  1.29it/s]
1|1154|Loss: 1.1008269786834717:  18%|█▊        | 1153/6470 [16:02<1:08:33,  1.29it/s]
1|1154|Loss: 1.1008269786834717:  18%|█▊        | 1154/6470 [16:02<1:04:37,  1.37it/s]
1|1155|Loss: 1.2088607549667358:  18%|█▊        | 1154/6470 [16:02<1:04:37,  1.37it/s]
1|1155|Loss: 1.2088607549667358:  18%|█▊        | 1155/6470 [16:02<1:01:12,  1.45it/s]
1|1156|Loss: 1.08772873878479:  18%|█▊        | 1155/6470 [16:03<1:01:12,  1.45it/s]  
1|1156|Loss: 1.08772873878479:  18%|█▊        | 1156/6470 [16:03<57:15,  1.55it/s]  
1|1157|Loss: 1.10489022731781:  18%|█▊        | 1156/6470 [16:03<57:15,  1.55it/s]
1|1157|Loss: 1.10489022731781:  18%|█▊        | 1157/6470 [16:04<56:00,  1.58it/s]
1|1158|Loss: 1.1136388778686523:  18%|█▊        | 1157/6470 [16:04<56:00,  1.58it/s]
1|1158|Loss: 1.1136388778686523:  18%|█▊        | 1158/6470 [16:04<56:09,  1.58it/s]
1|1159|Loss: 1.1322059631347656:  18%|█▊        | 1158/6470 [16:05<56:09,  1.58it/s]
1|1159|Loss: 1.1322059631347656:  18%|█▊        | 1159/6470 [16:05<56:20,  1.57it/s]
1|1160|Loss: 1.157622218132019:  18%|█▊        | 1159/6470 [16:05<56:20,  1.57it/s] 
1|1160|Loss: 1.157622218132019:  18%|█▊        | 1160/6470 [16:05<50:08,  1.76it/s]
1|1161|Loss: 0.9230551719665527:  18%|█▊        | 1160/6470 [16:06<50:08,  1.76it/s]
1|1161|Loss: 0.9230551719665527:  18%|█▊        | 1161/6470 [16:06<52:23,  1.69it/s]
1|1162|Loss: 0.9391269683837891:  18%|█▊        | 1161/6470 [16:06<52:23,  1.69it/s]
1|1162|Loss: 0.9391269683837891:  18%|█▊        | 1162/6470 [16:07<58:58,  1.50it/s]
1|1163|Loss: 0.9363862872123718:  18%|█▊        | 1162/6470 [16:07<58:58,  1.50it/s]
1|1163|Loss: 0.9363862872123718:  18%|█▊        | 1163/6470 [16:07<59:28,  1.49it/s]
1|1164|Loss: 1.1256455183029175:  18%|█▊        | 1163/6470 [16:08<59:28,  1.49it/s]
1|1164|Loss: 1.1256455183029175:  18%|█▊        | 1164/6470 [16:08<59:40,  1.48it/s]
1|1165|Loss: 1.0167940855026245:  18%|█▊        | 1164/6470 [16:09<59:40,  1.48it/s]
1|1165|Loss: 1.0167940855026245:  18%|█▊        | 1165/6470 [16:09<58:42,  1.51it/s]
1|1166|Loss: 1.083715558052063:  18%|█▊        | 1165/6470 [16:09<58:42,  1.51it/s] 
1|1166|Loss: 1.083715558052063:  18%|█▊        | 1166/6470 [16:09<55:56,  1.58it/s]
1|1167|Loss: 1.102622151374817:  18%|█▊        | 1166/6470 [16:10<55:56,  1.58it/s]
1|1167|Loss: 1.102622151374817:  18%|█▊        | 1167/6470 [16:10<58:51,  1.50it/s]
1|1168|Loss: 0.962751030921936:  18%|█▊        | 1167/6470 [16:11<58:51,  1.50it/s]
1|1168|Loss: 0.962751030921936:  18%|█▊        | 1168/6470 [16:11<1:02:15,  1.42it/s]
1|1169|Loss: 0.9328747987747192:  18%|█▊        | 1168/6470 [16:11<1:02:15,  1.42it/s]
1|1169|Loss: 0.9328747987747192:  18%|█▊        | 1169/6470 [16:12<1:05:29,  1.35it/s]
1|1170|Loss: 0.7215571999549866:  18%|█▊        | 1169/6470 [16:12<1:05:29,  1.35it/s]
1|1170|Loss: 0.7215571999549866:  18%|█▊        | 1170/6470 [16:13<1:24:10,  1.05it/s]
1|1171|Loss: 1.0405412912368774:  18%|█▊        | 1170/6470 [16:14<1:24:10,  1.05it/s]
1|1171|Loss: 1.0405412912368774:  18%|█▊        | 1171/6470 [16:14<1:23:15,  1.06it/s]
1|1172|Loss: 1.107257604598999:  18%|█▊        | 1171/6470 [16:14<1:23:15,  1.06it/s] 
1|1172|Loss: 1.107257604598999:  18%|█▊        | 1172/6470 [16:15<1:15:10,  1.17it/s]
1|1173|Loss: 1.1149115562438965:  18%|█▊        | 1172/6470 [16:15<1:15:10,  1.17it/s]
1|1173|Loss: 1.1149115562438965:  18%|█▊        | 1173/6470 [16:15<1:09:19,  1.27it/s]
1|1174|Loss: 1.001550555229187:  18%|█▊        | 1173/6470 [16:16<1:09:19,  1.27it/s] 
1|1174|Loss: 1.001550555229187:  18%|█▊        | 1174/6470 [16:16<1:07:14,  1.31it/s]
1|1175|Loss: 1.0192078351974487:  18%|█▊        | 1174/6470 [16:16<1:07:14,  1.31it/s]
1|1175|Loss: 1.0192078351974487:  18%|█▊        | 1175/6470 [16:17<1:06:25,  1.33it/s]
1|1176|Loss: 0.8936951756477356:  18%|█▊        | 1175/6470 [16:17<1:06:25,  1.33it/s]
1|1176|Loss: 0.8936951756477356:  18%|█▊        | 1176/6470 [16:18<1:09:57,  1.26it/s]
1|1177|Loss: 1.1110137701034546:  18%|█▊        | 1176/6470 [16:18<1:09:57,  1.26it/s]
1|1177|Loss: 1.1110137701034546:  18%|█▊        | 1177/6470 [16:18<1:07:09,  1.31it/s]
1|1178|Loss: 1.005528211593628:  18%|█▊        | 1177/6470 [16:19<1:07:09,  1.31it/s] 
1|1178|Loss: 1.005528211593628:  18%|█▊        | 1178/6470 [16:19<1:04:59,  1.36it/s]
1|1179|Loss: 0.9382463693618774:  18%|█▊        | 1178/6470 [16:19<1:04:59,  1.36it/s]
1|1179|Loss: 0.9382463693618774:  18%|█▊        | 1179/6470 [16:20<1:04:07,  1.38it/s]
1|1180|Loss: 1.052920937538147:  18%|█▊        | 1179/6470 [16:20<1:04:07,  1.38it/s] 
1|1180|Loss: 1.052920937538147:  18%|█▊        | 1180/6470 [16:20<1:01:40,  1.43it/s]
1|1181|Loss: 0.9834124445915222:  18%|█▊        | 1180/6470 [16:21<1:01:40,  1.43it/s]
1|1181|Loss: 0.9834124445915222:  18%|█▊        | 1181/6470 [16:21<58:40,  1.50it/s]  
1|1182|Loss: 1.1413886547088623:  18%|█▊        | 1181/6470 [16:21<58:40,  1.50it/s]
1|1182|Loss: 1.1413886547088623:  18%|█▊        | 1182/6470 [16:22<57:39,  1.53it/s]
1|1183|Loss: 1.097656488418579:  18%|█▊        | 1182/6470 [16:22<57:39,  1.53it/s] 
1|1183|Loss: 1.097656488418579:  18%|█▊        | 1183/6470 [16:22<1:01:52,  1.42it/s]
1|1184|Loss: 1.021701455116272:  18%|█▊        | 1183/6470 [16:23<1:01:52,  1.42it/s]
1|1184|Loss: 1.021701455116272:  18%|█▊        | 1184/6470 [16:23<58:12,  1.51it/s]  
1|1185|Loss: 0.9461981654167175:  18%|█▊        | 1184/6470 [16:23<58:12,  1.51it/s]
1|1185|Loss: 0.9461981654167175:  18%|█▊        | 1185/6470 [16:23<54:26,  1.62it/s]
1|1186|Loss: 1.0950493812561035:  18%|█▊        | 1185/6470 [16:24<54:26,  1.62it/s]
1|1186|Loss: 1.0950493812561035:  18%|█▊        | 1186/6470 [16:24<58:58,  1.49it/s]
1|1187|Loss: 1.011697769165039:  18%|█▊        | 1186/6470 [16:25<58:58,  1.49it/s] 
1|1187|Loss: 1.011697769165039:  18%|█▊        | 1187/6470 [16:25<1:03:48,  1.38it/s]
1|1188|Loss: 1.0241092443466187:  18%|█▊        | 1187/6470 [16:26<1:03:48,  1.38it/s]
1|1188|Loss: 1.0241092443466187:  18%|█▊        | 1188/6470 [16:26<1:02:56,  1.40it/s]
1|1189|Loss: 1.1103415489196777:  18%|█▊        | 1188/6470 [16:26<1:02:56,  1.40it/s]
1|1189|Loss: 1.1103415489196777:  18%|█▊        | 1189/6470 [16:26<1:00:27,  1.46it/s]
1|1190|Loss: 0.9438112378120422:  18%|█▊        | 1189/6470 [16:27<1:00:27,  1.46it/s]
1|1190|Loss: 0.9438112378120422:  18%|█▊        | 1190/6470 [16:27<58:02,  1.52it/s]  
1|1191|Loss: 1.1841440200805664:  18%|█▊        | 1190/6470 [16:27<58:02,  1.52it/s]
1|1191|Loss: 1.1841440200805664:  18%|█▊        | 1191/6470 [16:28<53:18,  1.65it/s]
1|1192|Loss: 0.9992234706878662:  18%|█▊        | 1191/6470 [16:28<53:18,  1.65it/s]
1|1192|Loss: 0.9992234706878662:  18%|█▊        | 1192/6470 [16:28<56:11,  1.57it/s]
1|1193|Loss: 1.0882338285446167:  18%|█▊        | 1192/6470 [16:29<56:11,  1.57it/s]
1|1193|Loss: 1.0882338285446167:  18%|█▊        | 1193/6470 [16:29<59:01,  1.49it/s]
1|1194|Loss: 1.1216992139816284:  18%|█▊        | 1193/6470 [16:29<59:01,  1.49it/s]
1|1194|Loss: 1.1216992139816284:  18%|█▊        | 1194/6470 [16:30<56:52,  1.55it/s]
1|1195|Loss: 1.1016701459884644:  18%|█▊        | 1194/6470 [16:30<56:52,  1.55it/s]
1|1195|Loss: 1.1016701459884644:  18%|█▊        | 1195/6470 [16:30<55:51,  1.57it/s]
1|1196|Loss: 1.0504908561706543:  18%|█▊        | 1195/6470 [16:31<55:51,  1.57it/s]
1|1196|Loss: 1.0504908561706543:  18%|█▊        | 1196/6470 [16:31<55:51,  1.57it/s]
1|1197|Loss: 1.0500190258026123:  18%|█▊        | 1196/6470 [16:31<55:51,  1.57it/s]
1|1197|Loss: 1.0500190258026123:  19%|█▊        | 1197/6470 [16:32<1:00:31,  1.45it/s]
1|1198|Loss: 1.148048758506775:  19%|█▊        | 1197/6470 [16:32<1:00:31,  1.45it/s] 
1|1198|Loss: 1.148048758506775:  19%|█▊        | 1198/6470 [16:32<1:00:09,  1.46it/s]
1|1199|Loss: 1.1179524660110474:  19%|█▊        | 1198/6470 [16:33<1:00:09,  1.46it/s]
1|1199|Loss: 1.1179524660110474:  19%|█▊        | 1199/6470 [16:33<58:16,  1.51it/s]  
1|1200|Loss: 0.8286899924278259:  19%|█▊        | 1199/6470 [16:33<58:16,  1.51it/s]
1|1200|Loss: 0.8286899924278259:  19%|█▊        | 1200/6470 [16:34<1:03:14,  1.39it/s]
1|1201|Loss: 0.8796644806861877:  19%|█▊        | 1200/6470 [16:34<1:03:14,  1.39it/s]
1|1201|Loss: 0.8796644806861877:  19%|█▊        | 1201/6470 [16:35<1:05:03,  1.35it/s]
1|1202|Loss: 1.1542803049087524:  19%|█▊        | 1201/6470 [16:35<1:05:03,  1.35it/s]
1|1202|Loss: 1.1542803049087524:  19%|█▊        | 1202/6470 [16:35<1:07:31,  1.30it/s]
1|1203|Loss: 1.0700510740280151:  19%|█▊        | 1202/6470 [16:36<1:07:31,  1.30it/s]
1|1203|Loss: 1.0700510740280151:  19%|█▊        | 1203/6470 [16:36<1:04:21,  1.36it/s]
1|1204|Loss: 0.982672929763794:  19%|█▊        | 1203/6470 [16:36<1:04:21,  1.36it/s] 
1|1204|Loss: 0.982672929763794:  19%|█▊        | 1204/6470 [16:37<1:05:34,  1.34it/s]
1|1205|Loss: 1.0840885639190674:  19%|█▊        | 1204/6470 [16:37<1:05:34,  1.34it/s]
1|1205|Loss: 1.0840885639190674:  19%|█▊        | 1205/6470 [16:38<1:07:16,  1.30it/s]
1|1206|Loss: 0.9870101809501648:  19%|█▊        | 1205/6470 [16:38<1:07:16,  1.30it/s]
1|1206|Loss: 0.9870101809501648:  19%|█▊        | 1206/6470 [16:38<1:06:48,  1.31it/s]
1|1207|Loss: 1.1200286149978638:  19%|█▊        | 1206/6470 [16:39<1:06:48,  1.31it/s]
1|1207|Loss: 1.1200286149978638:  19%|█▊        | 1207/6470 [16:39<1:07:11,  1.31it/s]
1|1208|Loss: 1.0190272331237793:  19%|█▊        | 1207/6470 [16:40<1:07:11,  1.31it/s]
1|1208|Loss: 1.0190272331237793:  19%|█▊        | 1208/6470 [16:40<1:06:32,  1.32it/s]
1|1209|Loss: 0.9659222960472107:  19%|█▊        | 1208/6470 [16:40<1:06:32,  1.32it/s]
1|1209|Loss: 0.9659222960472107:  19%|█▊        | 1209/6470 [16:41<1:05:09,  1.35it/s]
1|1210|Loss: 1.045727014541626:  19%|█▊        | 1209/6470 [16:41<1:05:09,  1.35it/s] 
1|1210|Loss: 1.045727014541626:  19%|█▊        | 1210/6470 [16:41<1:00:16,  1.45it/s]
1|1211|Loss: 1.042616605758667:  19%|█▊        | 1210/6470 [16:42<1:00:16,  1.45it/s]
1|1211|Loss: 1.042616605758667:  19%|█▊        | 1211/6470 [16:42<58:23,  1.50it/s]  
1|1212|Loss: 0.9526625275611877:  19%|█▊        | 1211/6470 [16:42<58:23,  1.50it/s]
1|1212|Loss: 0.9526625275611877:  19%|█▊        | 1212/6470 [16:43<59:35,  1.47it/s]
1|1213|Loss: 1.0181976556777954:  19%|█▊        | 1212/6470 [16:43<59:35,  1.47it/s]
1|1213|Loss: 1.0181976556777954:  19%|█▊        | 1213/6470 [16:43<1:00:54,  1.44it/s]
1|1214|Loss: 0.953036367893219:  19%|█▊        | 1213/6470 [16:44<1:00:54,  1.44it/s] 
1|1214|Loss: 0.953036367893219:  19%|█▉        | 1214/6470 [16:44<1:04:18,  1.36it/s]
1|1215|Loss: 0.9419594407081604:  19%|█▉        | 1214/6470 [16:45<1:04:18,  1.36it/s]
1|1215|Loss: 0.9419594407081604:  19%|█▉        | 1215/6470 [16:45<1:06:11,  1.32it/s]
1|1216|Loss: 1.1238065958023071:  19%|█▉        | 1215/6470 [16:45<1:06:11,  1.32it/s]
1|1216|Loss: 1.1238065958023071:  19%|█▉        | 1216/6470 [16:46<1:06:12,  1.32it/s]
1|1217|Loss: 0.9908724427223206:  19%|█▉        | 1216/6470 [16:46<1:06:12,  1.32it/s]
1|1217|Loss: 0.9908724427223206:  19%|█▉        | 1217/6470 [16:46<1:05:11,  1.34it/s]
1|1218|Loss: 1.151991367340088:  19%|█▉        | 1217/6470 [16:47<1:05:11,  1.34it/s] 
1|1218|Loss: 1.151991367340088:  19%|█▉        | 1218/6470 [16:47<1:00:25,  1.45it/s]
1|1219|Loss: 1.0687940120697021:  19%|█▉        | 1218/6470 [16:47<1:00:25,  1.45it/s]
1|1219|Loss: 1.0687940120697021:  19%|█▉        | 1219/6470 [16:48<58:03,  1.51it/s]  
1|1220|Loss: 1.1893569231033325:  19%|█▉        | 1219/6470 [16:48<58:03,  1.51it/s]
1|1220|Loss: 1.1893569231033325:  19%|█▉        | 1220/6470 [16:48<1:01:47,  1.42it/s]
1|1221|Loss: 0.9681784510612488:  19%|█▉        | 1220/6470 [16:49<1:01:47,  1.42it/s]
1|1221|Loss: 0.9681784510612488:  19%|█▉        | 1221/6470 [16:49<1:03:13,  1.38it/s]
1|1222|Loss: 0.9023383855819702:  19%|█▉        | 1221/6470 [16:49<1:03:13,  1.38it/s]
1|1222|Loss: 0.9023383855819702:  19%|█▉        | 1222/6470 [16:50<1:03:36,  1.37it/s]
1|1223|Loss: 0.9912897944450378:  19%|█▉        | 1222/6470 [16:50<1:03:36,  1.37it/s]
1|1223|Loss: 0.9912897944450378:  19%|█▉        | 1223/6470 [16:51<1:07:28,  1.30it/s]
1|1224|Loss: 1.045356035232544:  19%|█▉        | 1223/6470 [16:51<1:07:28,  1.30it/s] 
1|1224|Loss: 1.045356035232544:  19%|█▉        | 1224/6470 [16:52<1:08:51,  1.27it/s]
1|1225|Loss: 0.9797850847244263:  19%|█▉        | 1224/6470 [16:52<1:08:51,  1.27it/s]
1|1225|Loss: 0.9797850847244263:  19%|█▉        | 1225/6470 [16:52<1:04:31,  1.35it/s]
1|1226|Loss: 1.4648714065551758:  19%|█▉        | 1225/6470 [16:53<1:04:31,  1.35it/s]
1|1226|Loss: 1.4648714065551758:  19%|█▉        | 1226/6470 [16:53<1:06:03,  1.32it/s]
1|1227|Loss: 0.8591429591178894:  19%|█▉        | 1226/6470 [16:53<1:06:03,  1.32it/s]
1|1227|Loss: 0.8591429591178894:  19%|█▉        | 1227/6470 [16:54<1:11:05,  1.23it/s]
1|1228|Loss: 0.8807478547096252:  19%|█▉        | 1227/6470 [16:54<1:11:05,  1.23it/s]
1|1228|Loss: 0.8807478547096252:  19%|█▉        | 1228/6470 [16:55<1:14:28,  1.17it/s]
1|1229|Loss: 1.0492229461669922:  19%|█▉        | 1228/6470 [16:55<1:14:28,  1.17it/s]
1|1229|Loss: 1.0492229461669922:  19%|█▉        | 1229/6470 [16:56<1:12:37,  1.20it/s]
1|1230|Loss: 1.01946222782135:  19%|█▉        | 1229/6470 [16:56<1:12:37,  1.20it/s]  
1|1230|Loss: 1.01946222782135:  19%|█▉        | 1230/6470 [16:56<1:10:11,  1.24it/s]
1|1231|Loss: 1.085723876953125:  19%|█▉        | 1230/6470 [16:57<1:10:11,  1.24it/s]
1|1231|Loss: 1.085723876953125:  19%|█▉        | 1231/6470 [16:57<1:05:26,  1.33it/s]
1|1232|Loss: 1.1519031524658203:  19%|█▉        | 1231/6470 [16:57<1:05:26,  1.33it/s]
1|1232|Loss: 1.1519031524658203:  19%|█▉        | 1232/6470 [16:57<56:23,  1.55it/s]  
1|1233|Loss: 1.054351568222046:  19%|█▉        | 1232/6470 [16:58<56:23,  1.55it/s] 
1|1233|Loss: 1.054351568222046:  19%|█▉        | 1233/6470 [16:58<54:43,  1.59it/s]
1|1234|Loss: 1.067011833190918:  19%|█▉        | 1233/6470 [16:58<54:43,  1.59it/s]
1|1234|Loss: 1.067011833190918:  19%|█▉        | 1234/6470 [16:59<54:30,  1.60it/s]
1|1235|Loss: 1.0272276401519775:  19%|█▉        | 1234/6470 [16:59<54:30,  1.60it/s]
1|1235|Loss: 1.0272276401519775:  19%|█▉        | 1235/6470 [16:59<57:58,  1.51it/s]
1|1236|Loss: 1.0855871438980103:  19%|█▉        | 1235/6470 [17:00<57:58,  1.51it/s]
1|1236|Loss: 1.0855871438980103:  19%|█▉        | 1236/6470 [17:00<52:24,  1.66it/s]
1|1237|Loss: 1.299176573753357:  19%|█▉        | 1236/6470 [17:00<52:24,  1.66it/s] 
1|1237|Loss: 1.299176573753357:  19%|█▉        | 1237/6470 [17:00<48:57,  1.78it/s]
1|1238|Loss: 1.2555863857269287:  19%|█▉        | 1237/6470 [17:01<48:57,  1.78it/s]
1|1238|Loss: 1.2555863857269287:  19%|█▉        | 1238/6470 [17:01<51:27,  1.69it/s]
1|1239|Loss: 0.9895039200782776:  19%|█▉        | 1238/6470 [17:01<51:27,  1.69it/s]
1|1239|Loss: 0.9895039200782776:  19%|█▉        | 1239/6470 [17:01<50:51,  1.71it/s]
1|1240|Loss: 0.9437261819839478:  19%|█▉        | 1239/6470 [17:02<50:51,  1.71it/s]
1|1240|Loss: 0.9437261819839478:  19%|█▉        | 1240/6470 [17:02<54:00,  1.61it/s]
1|1241|Loss: 0.9196011424064636:  19%|█▉        | 1240/6470 [17:03<54:00,  1.61it/s]
1|1241|Loss: 0.9196011424064636:  19%|█▉        | 1241/6470 [17:03<57:12,  1.52it/s]
1|1242|Loss: 1.0332573652267456:  19%|█▉        | 1241/6470 [17:03<57:12,  1.52it/s]
1|1242|Loss: 1.0332573652267456:  19%|█▉        | 1242/6470 [17:03<53:50,  1.62it/s]
1|1243|Loss: 1.1211531162261963:  19%|█▉        | 1242/6470 [17:04<53:50,  1.62it/s]
1|1243|Loss: 1.1211531162261963:  19%|█▉        | 1243/6470 [17:04<53:29,  1.63it/s]
1|1244|Loss: 1.2001334428787231:  19%|█▉        | 1243/6470 [17:04<53:29,  1.63it/s]
1|1244|Loss: 1.2001334428787231:  19%|█▉        | 1244/6470 [17:05<55:30,  1.57it/s]
1|1245|Loss: 0.99931401014328:  19%|█▉        | 1244/6470 [17:05<55:30,  1.57it/s]  
1|1245|Loss: 0.99931401014328:  19%|█▉        | 1245/6470 [17:05<55:29,  1.57it/s]
1|1246|Loss: 1.099960207939148:  19%|█▉        | 1245/6470 [17:06<55:29,  1.57it/s]
1|1246|Loss: 1.099960207939148:  19%|█▉        | 1246/6470 [17:06<54:48,  1.59it/s]
1|1247|Loss: 1.0232114791870117:  19%|█▉        | 1246/6470 [17:06<54:48,  1.59it/s]
1|1247|Loss: 1.0232114791870117:  19%|█▉        | 1247/6470 [17:07<56:13,  1.55it/s]
1|1248|Loss: 1.0869414806365967:  19%|█▉        | 1247/6470 [17:07<56:13,  1.55it/s]
1|1248|Loss: 1.0869414806365967:  19%|█▉        | 1248/6470 [17:07<1:00:29,  1.44it/s]
1|1249|Loss: 1.256838083267212:  19%|█▉        | 1248/6470 [17:08<1:00:29,  1.44it/s] 
1|1249|Loss: 1.256838083267212:  19%|█▉        | 1249/6470 [17:08<1:03:25,  1.37it/s]
1|1250|Loss: 1.0035053491592407:  19%|█▉        | 1249/6470 [17:09<1:03:25,  1.37it/s]
1|1250|Loss: 1.0035053491592407:  19%|█▉        | 1250/6470 [17:09<1:03:45,  1.36it/s]
1|1251|Loss: 1.027467131614685:  19%|█▉        | 1250/6470 [17:09<1:03:45,  1.36it/s] 
1|1251|Loss: 1.027467131614685:  19%|█▉        | 1251/6470 [17:10<1:01:40,  1.41it/s]
1|1252|Loss: 0.9692285060882568:  19%|█▉        | 1251/6470 [17:10<1:01:40,  1.41it/s]
1|1252|Loss: 0.9692285060882568:  19%|█▉        | 1252/6470 [17:10<1:01:23,  1.42it/s]
1|1253|Loss: 1.1658347845077515:  19%|█▉        | 1252/6470 [17:11<1:01:23,  1.42it/s]
1|1253|Loss: 1.1658347845077515:  19%|█▉        | 1253/6470 [17:11<59:35,  1.46it/s]  
1|1254|Loss: 0.8967339992523193:  19%|█▉        | 1253/6470 [17:11<59:35,  1.46it/s]
1|1254|Loss: 0.8967339992523193:  19%|█▉        | 1254/6470 [17:12<1:01:27,  1.41it/s]
1|1255|Loss: 0.9497674703598022:  19%|█▉        | 1254/6470 [17:12<1:01:27,  1.41it/s]
1|1255|Loss: 0.9497674703598022:  19%|█▉        | 1255/6470 [17:12<1:00:09,  1.44it/s]
1|1256|Loss: 1.0054028034210205:  19%|█▉        | 1255/6470 [17:13<1:00:09,  1.44it/s]
1|1256|Loss: 1.0054028034210205:  19%|█▉        | 1256/6470 [17:13<1:02:34,  1.39it/s]
1|1257|Loss: 0.9330791234970093:  19%|█▉        | 1256/6470 [17:14<1:02:34,  1.39it/s]
1|1257|Loss: 0.9330791234970093:  19%|█▉        | 1257/6470 [17:14<1:01:05,  1.42it/s]
1|1258|Loss: 0.9348876476287842:  19%|█▉        | 1257/6470 [17:14<1:01:05,  1.42it/s]
1|1258|Loss: 0.9348876476287842:  19%|█▉        | 1258/6470 [17:15<1:03:49,  1.36it/s]
1|1259|Loss: 0.8638786673545837:  19%|█▉        | 1258/6470 [17:15<1:03:49,  1.36it/s]
1|1259|Loss: 0.8638786673545837:  19%|█▉        | 1259/6470 [17:15<1:02:36,  1.39it/s]
1|1260|Loss: 0.8546451330184937:  19%|█▉        | 1259/6470 [17:16<1:02:36,  1.39it/s]
1|1260|Loss: 0.8546451330184937:  19%|█▉        | 1260/6470 [17:16<1:12:29,  1.20it/s]
1|1261|Loss: 0.9632539749145508:  19%|█▉        | 1260/6470 [17:17<1:12:29,  1.20it/s]
1|1261|Loss: 0.9632539749145508:  19%|█▉        | 1261/6470 [17:17<1:10:35,  1.23it/s]
1|1262|Loss: 1.0814579725265503:  19%|█▉        | 1261/6470 [17:18<1:10:35,  1.23it/s]
1|1262|Loss: 1.0814579725265503:  20%|█▉        | 1262/6470 [17:18<1:07:32,  1.29it/s]
1|1263|Loss: 0.9994091391563416:  20%|█▉        | 1262/6470 [17:18<1:07:32,  1.29it/s]
1|1263|Loss: 0.9994091391563416:  20%|█▉        | 1263/6470 [17:19<1:03:46,  1.36it/s]
1|1264|Loss: 1.0049527883529663:  20%|█▉        | 1263/6470 [17:19<1:03:46,  1.36it/s]
1|1264|Loss: 1.0049527883529663:  20%|█▉        | 1264/6470 [17:19<1:03:21,  1.37it/s]
1|1265|Loss: 1.0680510997772217:  20%|█▉        | 1264/6470 [17:20<1:03:21,  1.37it/s]
1|1265|Loss: 1.0680510997772217:  20%|█▉        | 1265/6470 [17:20<58:56,  1.47it/s]  
1|1266|Loss: 0.9466989040374756:  20%|█▉        | 1265/6470 [17:20<58:56,  1.47it/s]
1|1266|Loss: 0.9466989040374756:  20%|█▉        | 1266/6470 [17:21<1:00:34,  1.43it/s]
1|1267|Loss: 1.0353798866271973:  20%|█▉        | 1266/6470 [17:21<1:00:34,  1.43it/s]
1|1267|Loss: 1.0353798866271973:  20%|█▉        | 1267/6470 [17:21<1:01:08,  1.42it/s]
1|1268|Loss: 1.094165325164795:  20%|█▉        | 1267/6470 [17:22<1:01:08,  1.42it/s] 
1|1268|Loss: 1.094165325164795:  20%|█▉        | 1268/6470 [17:22<1:01:37,  1.41it/s]
1|1269|Loss: 0.9708229303359985:  20%|█▉        | 1268/6470 [17:22<1:01:37,  1.41it/s]
1|1269|Loss: 0.9708229303359985:  20%|█▉        | 1269/6470 [17:23<1:00:19,  1.44it/s]
1|1270|Loss: 0.9450139403343201:  20%|█▉        | 1269/6470 [17:23<1:00:19,  1.44it/s]
1|1270|Loss: 0.9450139403343201:  20%|█▉        | 1270/6470 [17:23<56:04,  1.55it/s]  
1|1271|Loss: 0.8277550935745239:  20%|█▉        | 1270/6470 [17:24<56:04,  1.55it/s]
1|1271|Loss: 0.8277550935745239:  20%|█▉        | 1271/6470 [17:24<1:01:49,  1.40it/s]
1|1272|Loss: 0.9477730989456177:  20%|█▉        | 1271/6470 [17:25<1:01:49,  1.40it/s]
1|1272|Loss: 0.9477730989456177:  20%|█▉        | 1272/6470 [17:25<1:05:37,  1.32it/s]
1|1273|Loss: 1.1074676513671875:  20%|█▉        | 1272/6470 [17:25<1:05:37,  1.32it/s]
1|1273|Loss: 1.1074676513671875:  20%|█▉        | 1273/6470 [17:26<1:06:45,  1.30it/s]
1|1274|Loss: 0.9072520136833191:  20%|█▉        | 1273/6470 [17:26<1:06:45,  1.30it/s]
1|1274|Loss: 0.9072520136833191:  20%|█▉        | 1274/6470 [17:26<1:05:33,  1.32it/s]
1|1275|Loss: 0.9717743992805481:  20%|█▉        | 1274/6470 [17:27<1:05:33,  1.32it/s]
1|1275|Loss: 0.9717743992805481:  20%|█▉        | 1275/6470 [17:27<1:05:08,  1.33it/s]
1|1276|Loss: 1.161085605621338:  20%|█▉        | 1275/6470 [17:28<1:05:08,  1.33it/s] 
1|1276|Loss: 1.161085605621338:  20%|█▉        | 1276/6470 [17:28<1:04:52,  1.33it/s]
1|1277|Loss: 0.9789401292800903:  20%|█▉        | 1276/6470 [17:28<1:04:52,  1.33it/s]
1|1277|Loss: 0.9789401292800903:  20%|█▉        | 1277/6470 [17:29<1:02:16,  1.39it/s]
1|1278|Loss: 0.9836368560791016:  20%|█▉        | 1277/6470 [17:29<1:02:16,  1.39it/s]
1|1278|Loss: 0.9836368560791016:  20%|█▉        | 1278/6470 [17:29<1:04:56,  1.33it/s]
1|1279|Loss: 0.8868250250816345:  20%|█▉        | 1278/6470 [17:30<1:04:56,  1.33it/s]
1|1279|Loss: 0.8868250250816345:  20%|█▉        | 1279/6470 [17:30<1:07:25,  1.28it/s]
1|1280|Loss: 0.9841375350952148:  20%|█▉        | 1279/6470 [17:31<1:07:25,  1.28it/s]
1|1280|Loss: 0.9841375350952148:  20%|█▉        | 1280/6470 [17:31<1:02:18,  1.39it/s]
1|1281|Loss: 0.9713611006736755:  20%|█▉        | 1280/6470 [17:31<1:02:18,  1.39it/s]
1|1281|Loss: 0.9713611006736755:  20%|█▉        | 1281/6470 [17:31<59:05,  1.46it/s]  
1|1282|Loss: 0.8408076167106628:  20%|█▉        | 1281/6470 [17:32<59:05,  1.46it/s]
1|1282|Loss: 0.8408076167106628:  20%|█▉        | 1282/6470 [17:32<57:17,  1.51it/s]
1|1283|Loss: 1.0751198530197144:  20%|█▉        | 1282/6470 [17:32<57:17,  1.51it/s]
1|1283|Loss: 1.0751198530197144:  20%|█▉        | 1283/6470 [17:33<58:36,  1.48it/s]
1|1284|Loss: 0.9749392867088318:  20%|█▉        | 1283/6470 [17:33<58:36,  1.48it/s]
1|1284|Loss: 0.9749392867088318:  20%|█▉        | 1284/6470 [17:33<57:03,  1.51it/s]
1|1285|Loss: 0.8846588134765625:  20%|█▉        | 1284/6470 [17:34<57:03,  1.51it/s]
1|1285|Loss: 0.8846588134765625:  20%|█▉        | 1285/6470 [17:34<1:00:44,  1.42it/s]
1|1286|Loss: 0.95513516664505:  20%|█▉        | 1285/6470 [17:35<1:00:44,  1.42it/s]  
1|1286|Loss: 0.95513516664505:  20%|█▉        | 1286/6470 [17:35<58:06,  1.49it/s]  
1|1287|Loss: 0.9537562131881714:  20%|█▉        | 1286/6470 [17:35<58:06,  1.49it/s]
1|1287|Loss: 0.9537562131881714:  20%|█▉        | 1287/6470 [17:35<57:47,  1.49it/s]
1|1288|Loss: 1.1195460557937622:  20%|█▉        | 1287/6470 [17:36<57:47,  1.49it/s]
1|1288|Loss: 1.1195460557937622:  20%|█▉        | 1288/6470 [17:36<53:25,  1.62it/s]
1|1289|Loss: 0.9185791611671448:  20%|█▉        | 1288/6470 [17:36<53:25,  1.62it/s]
1|1289|Loss: 0.9185791611671448:  20%|█▉        | 1289/6470 [17:37<51:03,  1.69it/s]
1|1290|Loss: 1.0173341035842896:  20%|█▉        | 1289/6470 [17:37<51:03,  1.69it/s]
1|1290|Loss: 1.0173341035842896:  20%|█▉        | 1290/6470 [17:37<53:20,  1.62it/s]
1|1291|Loss: 1.0000925064086914:  20%|█▉        | 1290/6470 [17:38<53:20,  1.62it/s]
1|1291|Loss: 1.0000925064086914:  20%|█▉        | 1291/6470 [17:38<55:41,  1.55it/s]
1|1292|Loss: 0.8054032325744629:  20%|█▉        | 1291/6470 [17:38<55:41,  1.55it/s]
1|1292|Loss: 0.8054032325744629:  20%|█▉        | 1292/6470 [17:39<1:03:19,  1.36it/s]
1|1293|Loss: 1.005487322807312:  20%|█▉        | 1292/6470 [17:39<1:03:19,  1.36it/s] 
1|1293|Loss: 1.005487322807312:  20%|█▉        | 1293/6470 [17:40<1:05:04,  1.33it/s]
1|1294|Loss: 0.9757532477378845:  20%|█▉        | 1293/6470 [17:40<1:05:04,  1.33it/s]
1|1294|Loss: 0.9757532477378845:  20%|██        | 1294/6470 [17:40<1:07:29,  1.28it/s]
1|1295|Loss: 1.061638593673706:  20%|██        | 1294/6470 [17:41<1:07:29,  1.28it/s] 
1|1295|Loss: 1.061638593673706:  20%|██        | 1295/6470 [17:41<1:06:36,  1.29it/s]
1|1296|Loss: 1.0667622089385986:  20%|██        | 1295/6470 [17:42<1:06:36,  1.29it/s]
1|1296|Loss: 1.0667622089385986:  20%|██        | 1296/6470 [17:42<1:04:44,  1.33it/s]
1|1297|Loss: 0.9785624742507935:  20%|██        | 1296/6470 [17:42<1:04:44,  1.33it/s]
1|1297|Loss: 0.9785624742507935:  20%|██        | 1297/6470 [17:43<1:04:20,  1.34it/s]
1|1298|Loss: 1.2303284406661987:  20%|██        | 1297/6470 [17:43<1:04:20,  1.34it/s]
1|1298|Loss: 1.2303284406661987:  20%|██        | 1298/6470 [17:43<1:00:53,  1.42it/s]
1|1299|Loss: 1.0527396202087402:  20%|██        | 1298/6470 [17:44<1:00:53,  1.42it/s]
1|1299|Loss: 1.0527396202087402:  20%|██        | 1299/6470 [17:44<57:33,  1.50it/s]  
1|1300|Loss: 0.9147534966468811:  20%|██        | 1299/6470 [17:44<57:33,  1.50it/s]
1|1300|Loss: 0.9147534966468811:  20%|██        | 1300/6470 [17:45<56:36,  1.52it/s]
1|1301|Loss: 0.9537628889083862:  20%|██        | 1300/6470 [17:45<56:36,  1.52it/s]
1|1301|Loss: 0.9537628889083862:  20%|██        | 1301/6470 [17:45<59:45,  1.44it/s]
1|1302|Loss: 0.9944868683815002:  20%|██        | 1301/6470 [17:46<59:45,  1.44it/s]
1|1302|Loss: 0.9944868683815002:  20%|██        | 1302/6470 [17:46<1:03:04,  1.37it/s]
1|1303|Loss: 0.9612836241722107:  20%|██        | 1302/6470 [17:47<1:03:04,  1.37it/s]
1|1303|Loss: 0.9612836241722107:  20%|██        | 1303/6470 [17:47<1:06:06,  1.30it/s]
1|1304|Loss: 1.0477545261383057:  20%|██        | 1303/6470 [17:47<1:06:06,  1.30it/s]
1|1304|Loss: 1.0477545261383057:  20%|██        | 1304/6470 [17:48<1:01:43,  1.39it/s]
1|1305|Loss: 0.9291325211524963:  20%|██        | 1304/6470 [17:48<1:01:43,  1.39it/s]
1|1305|Loss: 0.9291325211524963:  20%|██        | 1305/6470 [17:48<1:00:39,  1.42it/s]
1|1306|Loss: 1.0819330215454102:  20%|██        | 1305/6470 [17:49<1:00:39,  1.42it/s]
1|1306|Loss: 1.0819330215454102:  20%|██        | 1306/6470 [17:49<1:01:54,  1.39it/s]
1|1307|Loss: 0.759911060333252:  20%|██        | 1306/6470 [17:49<1:01:54,  1.39it/s] 
1|1307|Loss: 0.759911060333252:  20%|██        | 1307/6470 [17:50<1:04:49,  1.33it/s]
1|1308|Loss: 1.0514968633651733:  20%|██        | 1307/6470 [17:50<1:04:49,  1.33it/s]
1|1308|Loss: 1.0514968633651733:  20%|██        | 1308/6470 [17:51<1:03:11,  1.36it/s]
1|1309|Loss: 0.9404452443122864:  20%|██        | 1308/6470 [17:51<1:03:11,  1.36it/s]
1|1309|Loss: 0.9404452443122864:  20%|██        | 1309/6470 [17:51<1:03:04,  1.36it/s]
1|1310|Loss: 1.1136846542358398:  20%|██        | 1309/6470 [17:52<1:03:04,  1.36it/s]
1|1310|Loss: 1.1136846542358398:  20%|██        | 1310/6470 [17:52<1:03:14,  1.36it/s]
1|1311|Loss: 1.059812068939209:  20%|██        | 1310/6470 [17:52<1:03:14,  1.36it/s] 
1|1311|Loss: 1.059812068939209:  20%|██        | 1311/6470 [17:53<59:17,  1.45it/s]  
1|1312|Loss: 1.0356521606445312:  20%|██        | 1311/6470 [17:53<59:17,  1.45it/s]
1|1312|Loss: 1.0356521606445312:  20%|██        | 1312/6470 [17:53<1:04:14,  1.34it/s]
1|1313|Loss: 0.8289414644241333:  20%|██        | 1312/6470 [17:54<1:04:14,  1.34it/s]
1|1313|Loss: 0.8289414644241333:  20%|██        | 1313/6470 [17:54<1:08:13,  1.26it/s]
1|1314|Loss: 1.1219643354415894:  20%|██        | 1313/6470 [17:55<1:08:13,  1.26it/s]
1|1314|Loss: 1.1219643354415894:  20%|██        | 1314/6470 [17:55<1:06:53,  1.28it/s]
1|1315|Loss: 0.9933894872665405:  20%|██        | 1314/6470 [17:55<1:06:53,  1.28it/s]
1|1315|Loss: 0.9933894872665405:  20%|██        | 1315/6470 [17:56<1:02:30,  1.37it/s]
1|1316|Loss: 1.043304443359375:  20%|██        | 1315/6470 [17:56<1:02:30,  1.37it/s] 
1|1316|Loss: 1.043304443359375:  20%|██        | 1316/6470 [17:57<1:05:39,  1.31it/s]
1|1317|Loss: 1.1626099348068237:  20%|██        | 1316/6470 [17:57<1:05:39,  1.31it/s]
1|1317|Loss: 1.1626099348068237:  20%|██        | 1317/6470 [17:57<1:01:43,  1.39it/s]
1|1318|Loss: 0.9306919574737549:  20%|██        | 1317/6470 [17:58<1:01:43,  1.39it/s]
1|1318|Loss: 0.9306919574737549:  20%|██        | 1318/6470 [17:58<1:00:48,  1.41it/s]
1|1319|Loss: 0.907344400882721:  20%|██        | 1318/6470 [17:58<1:00:48,  1.41it/s] 
1|1319|Loss: 0.907344400882721:  20%|██        | 1319/6470 [17:58<59:08,  1.45it/s]  
1|1320|Loss: 0.9152382016181946:  20%|██        | 1319/6470 [17:59<59:08,  1.45it/s]
1|1320|Loss: 0.9152382016181946:  20%|██        | 1320/6470 [17:59<59:40,  1.44it/s]
1|1321|Loss: 1.1380746364593506:  20%|██        | 1320/6470 [18:00<59:40,  1.44it/s]
1|1321|Loss: 1.1380746364593506:  20%|██        | 1321/6470 [18:00<1:01:58,  1.38it/s]
1|1322|Loss: 1.1323262453079224:  20%|██        | 1321/6470 [18:00<1:01:58,  1.38it/s]
1|1322|Loss: 1.1323262453079224:  20%|██        | 1322/6470 [18:01<1:04:35,  1.33it/s]
1|1323|Loss: 0.903632640838623:  20%|██        | 1322/6470 [18:01<1:04:35,  1.33it/s] 
1|1323|Loss: 0.903632640838623:  20%|██        | 1323/6470 [18:02<1:06:37,  1.29it/s]
1|1324|Loss: 0.9444606900215149:  20%|██        | 1323/6470 [18:02<1:06:37,  1.29it/s]
1|1324|Loss: 0.9444606900215149:  20%|██        | 1324/6470 [18:02<1:07:51,  1.26it/s]
1|1325|Loss: 1.0689419507980347:  20%|██        | 1324/6470 [18:03<1:07:51,  1.26it/s]
1|1325|Loss: 1.0689419507980347:  20%|██        | 1325/6470 [18:03<1:08:46,  1.25it/s]
1|1326|Loss: 1.0311074256896973:  20%|██        | 1325/6470 [18:04<1:08:46,  1.25it/s]
1|1326|Loss: 1.0311074256896973:  20%|██        | 1326/6470 [18:04<1:05:10,  1.32it/s]
1|1327|Loss: 1.0224586725234985:  20%|██        | 1326/6470 [18:04<1:05:10,  1.32it/s]
1|1327|Loss: 1.0224586725234985:  21%|██        | 1327/6470 [18:05<1:05:05,  1.32it/s]
1|1328|Loss: 0.8966943621635437:  21%|██        | 1327/6470 [18:05<1:05:05,  1.32it/s]
1|1328|Loss: 0.8966943621635437:  21%|██        | 1328/6470 [18:05<1:05:12,  1.31it/s]
1|1329|Loss: 1.0486900806427002:  21%|██        | 1328/6470 [18:06<1:05:12,  1.31it/s]
1|1329|Loss: 1.0486900806427002:  21%|██        | 1329/6470 [18:06<1:05:49,  1.30it/s]
1|1330|Loss: 0.9584105014801025:  21%|██        | 1329/6470 [18:07<1:05:49,  1.30it/s]
1|1330|Loss: 0.9584105014801025:  21%|██        | 1330/6470 [18:07<1:03:18,  1.35it/s]
1|1331|Loss: 1.0038000345230103:  21%|██        | 1330/6470 [18:07<1:03:18,  1.35it/s]
1|1331|Loss: 1.0038000345230103:  21%|██        | 1331/6470 [18:08<1:02:31,  1.37it/s]
1|1332|Loss: 1.0414162874221802:  21%|██        | 1331/6470 [18:08<1:02:31,  1.37it/s]
1|1332|Loss: 1.0414162874221802:  21%|██        | 1332/6470 [18:08<1:03:45,  1.34it/s]
1|1333|Loss: 0.8619349598884583:  21%|██        | 1332/6470 [18:09<1:03:45,  1.34it/s]
1|1333|Loss: 0.8619349598884583:  21%|██        | 1333/6470 [18:09<1:04:08,  1.33it/s]
1|1334|Loss: 1.1403261423110962:  21%|██        | 1333/6470 [18:10<1:04:08,  1.33it/s]
1|1334|Loss: 1.1403261423110962:  21%|██        | 1334/6470 [18:10<58:43,  1.46it/s]  
1|1335|Loss: 0.936016857624054:  21%|██        | 1334/6470 [18:10<58:43,  1.46it/s] 
1|1335|Loss: 0.936016857624054:  21%|██        | 1335/6470 [18:10<54:54,  1.56it/s]
1|1336|Loss: 0.9780286550521851:  21%|██        | 1335/6470 [18:11<54:54,  1.56it/s]
1|1336|Loss: 0.9780286550521851:  21%|██        | 1336/6470 [18:11<58:21,  1.47it/s]
1|1337|Loss: 0.8981053829193115:  21%|██        | 1336/6470 [18:11<58:21,  1.47it/s]
1|1337|Loss: 0.8981053829193115:  21%|██        | 1337/6470 [18:12<58:11,  1.47it/s]
1|1338|Loss: 0.8373079895973206:  21%|██        | 1337/6470 [18:12<58:11,  1.47it/s]
1|1338|Loss: 0.8373079895973206:  21%|██        | 1338/6470 [18:12<59:04,  1.45it/s]
1|1339|Loss: 1.119066596031189:  21%|██        | 1338/6470 [18:13<59:04,  1.45it/s] 
1|1339|Loss: 1.119066596031189:  21%|██        | 1339/6470 [18:13<56:57,  1.50it/s]
1|1340|Loss: 1.1419397592544556:  21%|██        | 1339/6470 [18:13<56:57,  1.50it/s]
1|1340|Loss: 1.1419397592544556:  21%|██        | 1340/6470 [18:14<58:16,  1.47it/s]
1|1341|Loss: 0.8587824106216431:  21%|██        | 1340/6470 [18:14<58:16,  1.47it/s]
1|1341|Loss: 0.8587824106216431:  21%|██        | 1341/6470 [18:15<1:01:48,  1.38it/s]
1|1342|Loss: 1.032408595085144:  21%|██        | 1341/6470 [18:15<1:01:48,  1.38it/s] 
1|1342|Loss: 1.032408595085144:  21%|██        | 1342/6470 [18:15<1:02:56,  1.36it/s]
1|1343|Loss: 1.0597949028015137:  21%|██        | 1342/6470 [18:16<1:02:56,  1.36it/s]
1|1343|Loss: 1.0597949028015137:  21%|██        | 1343/6470 [18:16<59:56,  1.43it/s]  
1|1344|Loss: 1.0297040939331055:  21%|██        | 1343/6470 [18:16<59:56,  1.43it/s]
1|1344|Loss: 1.0297040939331055:  21%|██        | 1344/6470 [18:17<1:01:43,  1.38it/s]
1|1345|Loss: 0.9916653037071228:  21%|██        | 1344/6470 [18:17<1:01:43,  1.38it/s]
1|1345|Loss: 0.9916653037071228:  21%|██        | 1345/6470 [18:17<58:19,  1.46it/s]  
1|1346|Loss: 0.9443420171737671:  21%|██        | 1345/6470 [18:18<58:19,  1.46it/s]
1|1346|Loss: 0.9443420171737671:  21%|██        | 1346/6470 [18:18<58:43,  1.45it/s]
1|1347|Loss: 0.968770444393158:  21%|██        | 1346/6470 [18:18<58:43,  1.45it/s] 
1|1347|Loss: 0.968770444393158:  21%|██        | 1347/6470 [18:19<57:45,  1.48it/s]
1|1348|Loss: 0.9532867670059204:  21%|██        | 1347/6470 [18:19<57:45,  1.48it/s]
1|1348|Loss: 0.9532867670059204:  21%|██        | 1348/6470 [18:19<58:40,  1.46it/s]
1|1349|Loss: 1.0719983577728271:  21%|██        | 1348/6470 [18:20<58:40,  1.46it/s]
1|1349|Loss: 1.0719983577728271:  21%|██        | 1349/6470 [18:20<55:11,  1.55it/s]
1|1350|Loss: 1.000894546508789:  21%|██        | 1349/6470 [18:20<55:11,  1.55it/s] 
1|1350|Loss: 1.000894546508789:  21%|██        | 1350/6470 [18:21<58:45,  1.45it/s]
1|1351|Loss: 0.9938168525695801:  21%|██        | 1350/6470 [18:21<58:45,  1.45it/s]
1|1351|Loss: 0.9938168525695801:  21%|██        | 1351/6470 [18:22<1:03:29,  1.34it/s]
1|1352|Loss: 1.0261026620864868:  21%|██        | 1351/6470 [18:22<1:03:29,  1.34it/s]
1|1352|Loss: 1.0261026620864868:  21%|██        | 1352/6470 [18:22<1:03:35,  1.34it/s]
1|1353|Loss: 1.0881175994873047:  21%|██        | 1352/6470 [18:23<1:03:35,  1.34it/s]
1|1353|Loss: 1.0881175994873047:  21%|██        | 1353/6470 [18:23<58:52,  1.45it/s]  
1|1354|Loss: 0.9890803694725037:  21%|██        | 1353/6470 [18:23<58:52,  1.45it/s]
1|1354|Loss: 0.9890803694725037:  21%|██        | 1354/6470 [18:24<1:00:48,  1.40it/s]
1|1355|Loss: 1.1290557384490967:  21%|██        | 1354/6470 [18:24<1:00:48,  1.40it/s]
1|1355|Loss: 1.1290557384490967:  21%|██        | 1355/6470 [18:24<1:02:12,  1.37it/s]
1|1356|Loss: 0.9333654642105103:  21%|██        | 1355/6470 [18:25<1:02:12,  1.37it/s]
1|1356|Loss: 0.9333654642105103:  21%|██        | 1356/6470 [18:25<59:15,  1.44it/s]  
1|1357|Loss: 0.9031474590301514:  21%|██        | 1356/6470 [18:25<59:15,  1.44it/s]
1|1357|Loss: 0.9031474590301514:  21%|██        | 1357/6470 [18:26<56:30,  1.51it/s]
1|1358|Loss: 0.9964887499809265:  21%|██        | 1357/6470 [18:26<56:30,  1.51it/s]
1|1358|Loss: 0.9964887499809265:  21%|██        | 1358/6470 [18:26<1:00:03,  1.42it/s]
1|1359|Loss: 0.9679549932479858:  21%|██        | 1358/6470 [18:27<1:00:03,  1.42it/s]
1|1359|Loss: 0.9679549932479858:  21%|██        | 1359/6470 [18:27<1:00:50,  1.40it/s]
1|1360|Loss: 0.9770448207855225:  21%|██        | 1359/6470 [18:28<1:00:50,  1.40it/s]
1|1360|Loss: 0.9770448207855225:  21%|██        | 1360/6470 [18:28<1:03:10,  1.35it/s]
1|1361|Loss: 0.9964250326156616:  21%|██        | 1360/6470 [18:28<1:03:10,  1.35it/s]
1|1361|Loss: 0.9964250326156616:  21%|██        | 1361/6470 [18:28<55:54,  1.52it/s]  
1|1362|Loss: 0.9773707985877991:  21%|██        | 1361/6470 [18:29<55:54,  1.52it/s]
1|1362|Loss: 0.9773707985877991:  21%|██        | 1362/6470 [18:29<53:40,  1.59it/s]
1|1363|Loss: 0.9940769076347351:  21%|██        | 1362/6470 [18:29<53:40,  1.59it/s]
1|1363|Loss: 0.9940769076347351:  21%|██        | 1363/6470 [18:30<55:40,  1.53it/s]
1|1364|Loss: 1.0378522872924805:  21%|██        | 1363/6470 [18:30<55:40,  1.53it/s]
1|1364|Loss: 1.0378522872924805:  21%|██        | 1364/6470 [18:31<1:02:08,  1.37it/s]
1|1365|Loss: 0.9796152114868164:  21%|██        | 1364/6470 [18:31<1:02:08,  1.37it/s]
1|1365|Loss: 0.9796152114868164:  21%|██        | 1365/6470 [18:31<1:03:01,  1.35it/s]
1|1366|Loss: 0.9379270672798157:  21%|██        | 1365/6470 [18:32<1:03:01,  1.35it/s]
1|1366|Loss: 0.9379270672798157:  21%|██        | 1366/6470 [18:32<1:02:43,  1.36it/s]
1|1367|Loss: 0.9249870777130127:  21%|██        | 1366/6470 [18:33<1:02:43,  1.36it/s]
1|1367|Loss: 0.9249870777130127:  21%|██        | 1367/6470 [18:33<1:01:54,  1.37it/s]
1|1368|Loss: 0.9040567874908447:  21%|██        | 1367/6470 [18:33<1:01:54,  1.37it/s]
1|1368|Loss: 0.9040567874908447:  21%|██        | 1368/6470 [18:33<59:25,  1.43it/s]  
1|1369|Loss: 1.1379897594451904:  21%|██        | 1368/6470 [18:34<59:25,  1.43it/s]
1|1369|Loss: 1.1379897594451904:  21%|██        | 1369/6470 [18:34<55:02,  1.54it/s]
1|1370|Loss: 1.0474060773849487:  21%|██        | 1369/6470 [18:34<55:02,  1.54it/s]
1|1370|Loss: 1.0474060773849487:  21%|██        | 1370/6470 [18:35<55:35,  1.53it/s]
1|1371|Loss: 1.0229592323303223:  21%|██        | 1370/6470 [18:35<55:35,  1.53it/s]
1|1371|Loss: 1.0229592323303223:  21%|██        | 1371/6470 [18:35<57:28,  1.48it/s]
1|1372|Loss: 1.1143649816513062:  21%|██        | 1371/6470 [18:36<57:28,  1.48it/s]
1|1372|Loss: 1.1143649816513062:  21%|██        | 1372/6470 [18:36<59:08,  1.44it/s]
1|1373|Loss: 1.0549184083938599:  21%|██        | 1372/6470 [18:37<59:08,  1.44it/s]
1|1373|Loss: 1.0549184083938599:  21%|██        | 1373/6470 [18:37<1:03:51,  1.33it/s]
1|1374|Loss: 1.0000640153884888:  21%|██        | 1373/6470 [18:37<1:03:51,  1.33it/s]
1|1374|Loss: 1.0000640153884888:  21%|██        | 1374/6470 [18:38<1:06:50,  1.27it/s]
1|1375|Loss: 0.9408041834831238:  21%|██        | 1374/6470 [18:38<1:06:50,  1.27it/s]
1|1375|Loss: 0.9408041834831238:  21%|██▏       | 1375/6470 [18:39<1:03:04,  1.35it/s]
1|1376|Loss: 1.0113798379898071:  21%|██▏       | 1375/6470 [18:39<1:03:04,  1.35it/s]
1|1376|Loss: 1.0113798379898071:  21%|██▏       | 1376/6470 [18:39<1:00:06,  1.41it/s]
1|1377|Loss: 0.8035464286804199:  21%|██▏       | 1376/6470 [18:40<1:00:06,  1.41it/s]
1|1377|Loss: 0.8035464286804199:  21%|██▏       | 1377/6470 [18:40<1:02:23,  1.36it/s]
1|1378|Loss: 1.0440869331359863:  21%|██▏       | 1377/6470 [18:40<1:02:23,  1.36it/s]
1|1378|Loss: 1.0440869331359863:  21%|██▏       | 1378/6470 [18:41<1:01:01,  1.39it/s]
1|1379|Loss: 0.9457701444625854:  21%|██▏       | 1378/6470 [18:41<1:01:01,  1.39it/s]
1|1379|Loss: 0.9457701444625854:  21%|██▏       | 1379/6470 [18:41<1:00:54,  1.39it/s]
1|1380|Loss: 1.1672412157058716:  21%|██▏       | 1379/6470 [18:42<1:00:54,  1.39it/s]
1|1380|Loss: 1.1672412157058716:  21%|██▏       | 1380/6470 [18:42<1:02:22,  1.36it/s]
1|1381|Loss: 0.8505274057388306:  21%|██▏       | 1380/6470 [18:43<1:02:22,  1.36it/s]
1|1381|Loss: 0.8505274057388306:  21%|██▏       | 1381/6470 [18:43<1:06:17,  1.28it/s]
1|1382|Loss: 0.933806836605072:  21%|██▏       | 1381/6470 [18:43<1:06:17,  1.28it/s] 
1|1382|Loss: 0.933806836605072:  21%|██▏       | 1382/6470 [18:44<1:05:22,  1.30it/s]
1|1383|Loss: 0.9840201735496521:  21%|██▏       | 1382/6470 [18:44<1:05:22,  1.30it/s]
1|1383|Loss: 0.9840201735496521:  21%|██▏       | 1383/6470 [18:44<1:04:10,  1.32it/s]
1|1384|Loss: 1.0389941930770874:  21%|██▏       | 1383/6470 [18:45<1:04:10,  1.32it/s]
1|1384|Loss: 1.0389941930770874:  21%|██▏       | 1384/6470 [18:45<1:01:35,  1.38it/s]
1|1385|Loss: 1.2031818628311157:  21%|██▏       | 1384/6470 [18:45<1:01:35,  1.38it/s]
1|1385|Loss: 1.2031818628311157:  21%|██▏       | 1385/6470 [18:46<57:58,  1.46it/s]  
1|1386|Loss: 1.0962235927581787:  21%|██▏       | 1385/6470 [18:46<57:58,  1.46it/s]
1|1386|Loss: 1.0962235927581787:  21%|██▏       | 1386/6470 [18:46<56:25,  1.50it/s]
1|1387|Loss: 0.961970865726471:  21%|██▏       | 1386/6470 [18:47<56:25,  1.50it/s] 
1|1387|Loss: 0.961970865726471:  21%|██▏       | 1387/6470 [18:47<55:17,  1.53it/s]
1|1388|Loss: 1.025768756866455:  21%|██▏       | 1387/6470 [18:47<55:17,  1.53it/s]
1|1388|Loss: 1.025768756866455:  21%|██▏       | 1388/6470 [18:47<51:01,  1.66it/s]
1|1389|Loss: 1.0152231454849243:  21%|██▏       | 1388/6470 [18:48<51:01,  1.66it/s]
1|1389|Loss: 1.0152231454849243:  21%|██▏       | 1389/6470 [18:48<50:53,  1.66it/s]
1|1390|Loss: 0.9455534815788269:  21%|██▏       | 1389/6470 [18:48<50:53,  1.66it/s]
1|1390|Loss: 0.9455534815788269:  21%|██▏       | 1390/6470 [18:49<55:49,  1.52it/s]
1|1391|Loss: 0.9705794453620911:  21%|██▏       | 1390/6470 [18:49<55:49,  1.52it/s]
1|1391|Loss: 0.9705794453620911:  21%|██▏       | 1391/6470 [18:50<58:38,  1.44it/s]
1|1392|Loss: 0.9151284098625183:  21%|██▏       | 1391/6470 [18:50<58:38,  1.44it/s]
1|1392|Loss: 0.9151284098625183:  22%|██▏       | 1392/6470 [18:50<58:11,  1.45it/s]
1|1393|Loss: 1.1380630731582642:  22%|██▏       | 1392/6470 [18:51<58:11,  1.45it/s]
1|1393|Loss: 1.1380630731582642:  22%|██▏       | 1393/6470 [18:51<58:22,  1.45it/s]
1|1394|Loss: 1.0776171684265137:  22%|██▏       | 1393/6470 [18:51<58:22,  1.45it/s]
1|1394|Loss: 1.0776171684265137:  22%|██▏       | 1394/6470 [18:52<59:39,  1.42it/s]
1|1395|Loss: 1.0660977363586426:  22%|██▏       | 1394/6470 [18:52<59:39,  1.42it/s]
1|1395|Loss: 1.0660977363586426:  22%|██▏       | 1395/6470 [18:52<58:29,  1.45it/s]
1|1396|Loss: 0.9521462321281433:  22%|██▏       | 1395/6470 [18:53<58:29,  1.45it/s]
1|1396|Loss: 0.9521462321281433:  22%|██▏       | 1396/6470 [18:53<58:09,  1.45it/s]
1|1397|Loss: 0.9852617979049683:  22%|██▏       | 1396/6470 [18:53<58:09,  1.45it/s]
1|1397|Loss: 0.9852617979049683:  22%|██▏       | 1397/6470 [18:54<59:30,  1.42it/s]
1|1398|Loss: 1.177590250968933:  22%|██▏       | 1397/6470 [18:54<59:30,  1.42it/s] 
1|1398|Loss: 1.177590250968933:  22%|██▏       | 1398/6470 [18:54<55:45,  1.52it/s]
1|1399|Loss: 1.1236844062805176:  22%|██▏       | 1398/6470 [18:55<55:45,  1.52it/s]
1|1399|Loss: 1.1236844062805176:  22%|██▏       | 1399/6470 [18:55<54:10,  1.56it/s]
1|1400|Loss: 0.9767428636550903:  22%|██▏       | 1399/6470 [18:55<54:10,  1.56it/s]
1|1400|Loss: 0.9767428636550903:  22%|██▏       | 1400/6470 [18:56<56:13,  1.50it/s]
1|1401|Loss: 1.0996168851852417:  22%|██▏       | 1400/6470 [18:56<56:13,  1.50it/s]
1|1401|Loss: 1.0996168851852417:  22%|██▏       | 1401/6470 [18:56<50:09,  1.68it/s]
1|1402|Loss: 0.958636999130249:  22%|██▏       | 1401/6470 [18:56<50:09,  1.68it/s] 
1|1402|Loss: 0.958636999130249:  22%|██▏       | 1402/6470 [18:57<52:25,  1.61it/s]
1|1403|Loss: 1.106500506401062:  22%|██▏       | 1402/6470 [18:57<52:25,  1.61it/s]
1|1403|Loss: 1.106500506401062:  22%|██▏       | 1403/6470 [18:58<55:47,  1.51it/s]
1|1404|Loss: 0.8972810506820679:  22%|██▏       | 1403/6470 [18:58<55:47,  1.51it/s]
1|1404|Loss: 0.8972810506820679:  22%|██▏       | 1404/6470 [18:58<53:08,  1.59it/s]
1|1405|Loss: 1.0207971334457397:  22%|██▏       | 1404/6470 [18:58<53:08,  1.59it/s]
1|1405|Loss: 1.0207971334457397:  22%|██▏       | 1405/6470 [18:59<51:47,  1.63it/s]
1|1406|Loss: 1.0257682800292969:  22%|██▏       | 1405/6470 [18:59<51:47,  1.63it/s]
1|1406|Loss: 1.0257682800292969:  22%|██▏       | 1406/6470 [18:59<53:41,  1.57it/s]
1|1407|Loss: 0.9418113231658936:  22%|██▏       | 1406/6470 [19:00<53:41,  1.57it/s]
1|1407|Loss: 0.9418113231658936:  22%|██▏       | 1407/6470 [19:00<56:27,  1.49it/s]
1|1408|Loss: 0.9750701785087585:  22%|██▏       | 1407/6470 [19:00<56:27,  1.49it/s]
1|1408|Loss: 0.9750701785087585:  22%|██▏       | 1408/6470 [19:01<56:01,  1.51it/s]
1|1409|Loss: 1.0498210191726685:  22%|██▏       | 1408/6470 [19:01<56:01,  1.51it/s]
1|1409|Loss: 1.0498210191726685:  22%|██▏       | 1409/6470 [19:02<59:05,  1.43it/s]
1|1410|Loss: 1.2361568212509155:  22%|██▏       | 1409/6470 [19:02<59:05,  1.43it/s]
1|1410|Loss: 1.2361568212509155:  22%|██▏       | 1410/6470 [19:02<55:53,  1.51it/s]
1|1411|Loss: 0.9209736585617065:  22%|██▏       | 1410/6470 [19:02<55:53,  1.51it/s]
1|1411|Loss: 0.9209736585617065:  22%|██▏       | 1411/6470 [19:03<54:08,  1.56it/s]
1|1412|Loss: 1.0470860004425049:  22%|██▏       | 1411/6470 [19:03<54:08,  1.56it/s]
1|1412|Loss: 1.0470860004425049:  22%|██▏       | 1412/6470 [19:03<55:41,  1.51it/s]
1|1413|Loss: 0.9516282677650452:  22%|██▏       | 1412/6470 [19:04<55:41,  1.51it/s]
1|1413|Loss: 0.9516282677650452:  22%|██▏       | 1413/6470 [19:04<53:42,  1.57it/s]
1|1414|Loss: 1.0594414472579956:  22%|██▏       | 1413/6470 [19:04<53:42,  1.57it/s]
1|1414|Loss: 1.0594414472579956:  22%|██▏       | 1414/6470 [19:05<58:57,  1.43it/s]
1|1415|Loss: 0.9972872734069824:  22%|██▏       | 1414/6470 [19:05<58:57,  1.43it/s]
1|1415|Loss: 0.9972872734069824:  22%|██▏       | 1415/6470 [19:06<58:46,  1.43it/s]
1|1416|Loss: 1.0386444330215454:  22%|██▏       | 1415/6470 [19:06<58:46,  1.43it/s]
1|1416|Loss: 1.0386444330215454:  22%|██▏       | 1416/6470 [19:06<58:39,  1.44it/s]
1|1417|Loss: 1.0822714567184448:  22%|██▏       | 1416/6470 [19:07<58:39,  1.44it/s]
1|1417|Loss: 1.0822714567184448:  22%|██▏       | 1417/6470 [19:07<1:01:15,  1.37it/s]
1|1418|Loss: 1.0247890949249268:  22%|██▏       | 1417/6470 [19:07<1:01:15,  1.37it/s]
1|1418|Loss: 1.0247890949249268:  22%|██▏       | 1418/6470 [19:08<1:01:08,  1.38it/s]
1|1419|Loss: 1.1092073917388916:  22%|██▏       | 1418/6470 [19:08<1:01:08,  1.38it/s]
1|1419|Loss: 1.1092073917388916:  22%|██▏       | 1419/6470 [19:09<1:03:09,  1.33it/s]
1|1420|Loss: 1.1641024351119995:  22%|██▏       | 1419/6470 [19:09<1:03:09,  1.33it/s]
1|1420|Loss: 1.1641024351119995:  22%|██▏       | 1420/6470 [19:09<1:01:23,  1.37it/s]
1|1421|Loss: 1.0619502067565918:  22%|██▏       | 1420/6470 [19:10<1:01:23,  1.37it/s]
1|1421|Loss: 1.0619502067565918:  22%|██▏       | 1421/6470 [19:10<1:01:18,  1.37it/s]
1|1422|Loss: 1.2193293571472168:  22%|██▏       | 1421/6470 [19:10<1:01:18,  1.37it/s]
1|1422|Loss: 1.2193293571472168:  22%|██▏       | 1422/6470 [19:11<1:03:25,  1.33it/s]
1|1423|Loss: 0.8938845992088318:  22%|██▏       | 1422/6470 [19:11<1:03:25,  1.33it/s]
1|1423|Loss: 0.8938845992088318:  22%|██▏       | 1423/6470 [19:12<1:03:22,  1.33it/s]
1|1424|Loss: 1.0517892837524414:  22%|██▏       | 1423/6470 [19:12<1:03:22,  1.33it/s]
1|1424|Loss: 1.0517892837524414:  22%|██▏       | 1424/6470 [19:12<59:32,  1.41it/s]  
1|1425|Loss: 0.9324597120285034:  22%|██▏       | 1424/6470 [19:13<59:32,  1.41it/s]
1|1425|Loss: 0.9324597120285034:  22%|██▏       | 1425/6470 [19:13<1:01:20,  1.37it/s]
1|1426|Loss: 0.7763325572013855:  22%|██▏       | 1425/6470 [19:13<1:01:20,  1.37it/s]
1|1426|Loss: 0.7763325572013855:  22%|██▏       | 1426/6470 [19:14<1:03:20,  1.33it/s]
1|1427|Loss: 0.9227190017700195:  22%|██▏       | 1426/6470 [19:14<1:03:20,  1.33it/s]
1|1427|Loss: 0.9227190017700195:  22%|██▏       | 1427/6470 [19:14<1:01:25,  1.37it/s]
1|1428|Loss: 0.9163771867752075:  22%|██▏       | 1427/6470 [19:15<1:01:25,  1.37it/s]
1|1428|Loss: 0.9163771867752075:  22%|██▏       | 1428/6470 [19:15<59:07,  1.42it/s]  
1|1429|Loss: 0.9975094795227051:  22%|██▏       | 1428/6470 [19:15<59:07,  1.42it/s]
1|1429|Loss: 0.9975094795227051:  22%|██▏       | 1429/6470 [19:16<1:02:10,  1.35it/s]
1|1430|Loss: 1.0219402313232422:  22%|██▏       | 1429/6470 [19:16<1:02:10,  1.35it/s]
1|1430|Loss: 1.0219402313232422:  22%|██▏       | 1430/6470 [19:17<1:02:11,  1.35it/s]
1|1431|Loss: 1.0008705854415894:  22%|██▏       | 1430/6470 [19:17<1:02:11,  1.35it/s]
1|1431|Loss: 1.0008705854415894:  22%|██▏       | 1431/6470 [19:17<58:37,  1.43it/s]  
1|1432|Loss: 1.1057565212249756:  22%|██▏       | 1431/6470 [19:18<58:37,  1.43it/s]
1|1432|Loss: 1.1057565212249756:  22%|██▏       | 1432/6470 [19:18<58:53,  1.43it/s]
1|1433|Loss: 1.0235315561294556:  22%|██▏       | 1432/6470 [19:18<58:53,  1.43it/s]
1|1433|Loss: 1.0235315561294556:  22%|██▏       | 1433/6470 [19:19<57:31,  1.46it/s]
1|1434|Loss: 1.0615884065628052:  22%|██▏       | 1433/6470 [19:19<57:31,  1.46it/s]
1|1434|Loss: 1.0615884065628052:  22%|██▏       | 1434/6470 [19:19<55:44,  1.51it/s]
1|1435|Loss: 0.9151338338851929:  22%|██▏       | 1434/6470 [19:20<55:44,  1.51it/s]
1|1435|Loss: 0.9151338338851929:  22%|██▏       | 1435/6470 [19:20<54:36,  1.54it/s]
1|1436|Loss: 1.0449298620224:  22%|██▏       | 1435/6470 [19:20<54:36,  1.54it/s]   
1|1436|Loss: 1.0449298620224:  22%|██▏       | 1436/6470 [19:20<51:58,  1.61it/s]
1|1437|Loss: 1.1240806579589844:  22%|██▏       | 1436/6470 [19:21<51:58,  1.61it/s]
1|1437|Loss: 1.1240806579589844:  22%|██▏       | 1437/6470 [19:21<48:06,  1.74it/s]
1|1438|Loss: 0.8859833478927612:  22%|██▏       | 1437/6470 [19:21<48:06,  1.74it/s]
1|1438|Loss: 0.8859833478927612:  22%|██▏       | 1438/6470 [19:21<48:46,  1.72it/s]
1|1439|Loss: 0.7645778059959412:  22%|██▏       | 1438/6470 [19:22<48:46,  1.72it/s]
1|1439|Loss: 0.7645778059959412:  22%|██▏       | 1439/6470 [19:22<55:47,  1.50it/s]
1|1440|Loss: 1.0974839925765991:  22%|██▏       | 1439/6470 [19:23<55:47,  1.50it/s]
1|1440|Loss: 1.0974839925765991:  22%|██▏       | 1440/6470 [19:23<56:33,  1.48it/s]
1|1441|Loss: 0.8991014361381531:  22%|██▏       | 1440/6470 [19:23<56:33,  1.48it/s]
1|1441|Loss: 0.8991014361381531:  22%|██▏       | 1441/6470 [19:24<54:17,  1.54it/s]
1|1442|Loss: 1.1037484407424927:  22%|██▏       | 1441/6470 [19:24<54:17,  1.54it/s]
1|1442|Loss: 1.1037484407424927:  22%|██▏       | 1442/6470 [19:24<51:13,  1.64it/s]
1|1443|Loss: 0.8314723372459412:  22%|██▏       | 1442/6470 [19:24<51:13,  1.64it/s]
1|1443|Loss: 0.8314723372459412:  22%|██▏       | 1443/6470 [19:25<51:14,  1.64it/s]
1|1444|Loss: 1.0246496200561523:  22%|██▏       | 1443/6470 [19:25<51:14,  1.64it/s]
1|1444|Loss: 1.0246496200561523:  22%|██▏       | 1444/6470 [19:25<53:48,  1.56it/s]
1|1445|Loss: 1.0622819662094116:  22%|██▏       | 1444/6470 [19:26<53:48,  1.56it/s]
1|1445|Loss: 1.0622819662094116:  22%|██▏       | 1445/6470 [19:26<51:24,  1.63it/s]
1|1446|Loss: 0.9773045182228088:  22%|██▏       | 1445/6470 [19:26<51:24,  1.63it/s]
1|1446|Loss: 0.9773045182228088:  22%|██▏       | 1446/6470 [19:26<49:29,  1.69it/s]
1|1447|Loss: 1.0476264953613281:  22%|██▏       | 1446/6470 [19:27<49:29,  1.69it/s]
1|1447|Loss: 1.0476264953613281:  22%|██▏       | 1447/6470 [19:27<53:31,  1.56it/s]
1|1448|Loss: 1.1166208982467651:  22%|██▏       | 1447/6470 [19:28<53:31,  1.56it/s]
1|1448|Loss: 1.1166208982467651:  22%|██▏       | 1448/6470 [19:28<50:42,  1.65it/s]
1|1449|Loss: 1.0784265995025635:  22%|██▏       | 1448/6470 [19:28<50:42,  1.65it/s]
1|1449|Loss: 1.0784265995025635:  22%|██▏       | 1449/6470 [19:28<51:39,  1.62it/s]
1|1450|Loss: 1.3467979431152344:  22%|██▏       | 1449/6470 [19:29<51:39,  1.62it/s]
1|1450|Loss: 1.3467979431152344:  22%|██▏       | 1450/6470 [19:29<51:25,  1.63it/s]
1|1451|Loss: 1.0001235008239746:  22%|██▏       | 1450/6470 [19:29<51:25,  1.63it/s]
1|1451|Loss: 1.0001235008239746:  22%|██▏       | 1451/6470 [19:30<56:28,  1.48it/s]
1|1452|Loss: 0.9866844415664673:  22%|██▏       | 1451/6470 [19:30<56:28,  1.48it/s]
1|1452|Loss: 0.9866844415664673:  22%|██▏       | 1452/6470 [19:31<1:01:16,  1.36it/s]
1|1453|Loss: 0.9080758094787598:  22%|██▏       | 1452/6470 [19:31<1:01:16,  1.36it/s]
1|1453|Loss: 0.9080758094787598:  22%|██▏       | 1453/6470 [19:31<58:55,  1.42it/s]  
1|1454|Loss: 1.0505954027175903:  22%|██▏       | 1453/6470 [19:32<58:55,  1.42it/s]
1|1454|Loss: 1.0505954027175903:  22%|██▏       | 1454/6470 [19:32<56:36,  1.48it/s]
1|1455|Loss: 1.0087884664535522:  22%|██▏       | 1454/6470 [19:32<56:36,  1.48it/s]
1|1455|Loss: 1.0087884664535522:  22%|██▏       | 1455/6470 [19:33<55:03,  1.52it/s]
1|1456|Loss: 0.9939430952072144:  22%|██▏       | 1455/6470 [19:33<55:03,  1.52it/s]
1|1456|Loss: 0.9939430952072144:  23%|██▎       | 1456/6470 [19:33<52:34,  1.59it/s]
1|1457|Loss: 0.9080188870429993:  23%|██▎       | 1456/6470 [19:34<52:34,  1.59it/s]
1|1457|Loss: 0.9080188870429993:  23%|██▎       | 1457/6470 [19:34<54:05,  1.54it/s]
1|1458|Loss: 0.9329769611358643:  23%|██▎       | 1457/6470 [19:34<54:05,  1.54it/s]
1|1458|Loss: 0.9329769611358643:  23%|██▎       | 1458/6470 [19:35<57:33,  1.45it/s]
1|1459|Loss: 1.1290397644042969:  23%|██▎       | 1458/6470 [19:35<57:33,  1.45it/s]
1|1459|Loss: 1.1290397644042969:  23%|██▎       | 1459/6470 [19:35<54:32,  1.53it/s]
1|1460|Loss: 1.1452945470809937:  23%|██▎       | 1459/6470 [19:36<54:32,  1.53it/s]
1|1460|Loss: 1.1452945470809937:  23%|██▎       | 1460/6470 [19:36<54:33,  1.53it/s]
1|1461|Loss: 0.9664854407310486:  23%|██▎       | 1460/6470 [19:36<54:33,  1.53it/s]
1|1461|Loss: 0.9664854407310486:  23%|██▎       | 1461/6470 [19:37<58:09,  1.44it/s]
1|1462|Loss: 0.8256308436393738:  23%|██▎       | 1461/6470 [19:37<58:09,  1.44it/s]
1|1462|Loss: 0.8256308436393738:  23%|██▎       | 1462/6470 [19:37<59:36,  1.40it/s]
1|1463|Loss: 0.9872618913650513:  23%|██▎       | 1462/6470 [19:38<59:36,  1.40it/s]
1|1463|Loss: 0.9872618913650513:  23%|██▎       | 1463/6470 [19:38<58:06,  1.44it/s]
1|1464|Loss: 0.9918298721313477:  23%|██▎       | 1463/6470 [19:38<58:06,  1.44it/s]
1|1464|Loss: 0.9918298721313477:  23%|██▎       | 1464/6470 [19:39<57:56,  1.44it/s]
1|1465|Loss: 1.0095796585083008:  23%|██▎       | 1464/6470 [19:39<57:56,  1.44it/s]
1|1465|Loss: 1.0095796585083008:  23%|██▎       | 1465/6470 [19:39<58:04,  1.44it/s]
1|1466|Loss: 1.0894404649734497:  23%|██▎       | 1465/6470 [19:40<58:04,  1.44it/s]
1|1466|Loss: 1.0894404649734497:  23%|██▎       | 1466/6470 [19:40<1:01:14,  1.36it/s]
1|1467|Loss: 1.0150047540664673:  23%|██▎       | 1466/6470 [19:41<1:01:14,  1.36it/s]
1|1467|Loss: 1.0150047540664673:  23%|██▎       | 1467/6470 [19:41<1:01:22,  1.36it/s]
1|1468|Loss: 0.9468971490859985:  23%|██▎       | 1467/6470 [19:41<1:01:22,  1.36it/s]
1|1468|Loss: 0.9468971490859985:  23%|██▎       | 1468/6470 [19:42<58:12,  1.43it/s]  
1|1469|Loss: 1.0781464576721191:  23%|██▎       | 1468/6470 [19:42<58:12,  1.43it/s]
1|1469|Loss: 1.0781464576721191:  23%|██▎       | 1469/6470 [19:42<56:24,  1.48it/s]
1|1470|Loss: 0.9219461679458618:  23%|██▎       | 1469/6470 [19:43<56:24,  1.48it/s]
1|1470|Loss: 0.9219461679458618:  23%|██▎       | 1470/6470 [19:43<54:15,  1.54it/s]
1|1471|Loss: 1.056290626525879:  23%|██▎       | 1470/6470 [19:43<54:15,  1.54it/s] 
1|1471|Loss: 1.056290626525879:  23%|██▎       | 1471/6470 [19:43<53:07,  1.57it/s]
1|1472|Loss: 0.8676488995552063:  23%|██▎       | 1471/6470 [19:44<53:07,  1.57it/s]
1|1472|Loss: 0.8676488995552063:  23%|██▎       | 1472/6470 [19:44<55:09,  1.51it/s]
1|1473|Loss: 0.8441454172134399:  23%|██▎       | 1472/6470 [19:45<55:09,  1.51it/s]
1|1473|Loss: 0.8441454172134399:  23%|██▎       | 1473/6470 [19:45<57:28,  1.45it/s]
1|1474|Loss: 0.9065452814102173:  23%|██▎       | 1473/6470 [19:45<57:28,  1.45it/s]
1|1474|Loss: 0.9065452814102173:  23%|██▎       | 1474/6470 [19:45<54:10,  1.54it/s]
1|1475|Loss: 1.1041651964187622:  23%|██▎       | 1474/6470 [19:46<54:10,  1.54it/s]
1|1475|Loss: 1.1041651964187622:  23%|██▎       | 1475/6470 [19:46<54:31,  1.53it/s]
1|1476|Loss: 0.9615236520767212:  23%|██▎       | 1475/6470 [19:47<54:31,  1.53it/s]
1|1476|Loss: 0.9615236520767212:  23%|██▎       | 1476/6470 [19:47<54:32,  1.53it/s]
1|1477|Loss: 0.8619646430015564:  23%|██▎       | 1476/6470 [19:47<54:32,  1.53it/s]
1|1477|Loss: 0.8619646430015564:  23%|██▎       | 1477/6470 [19:48<55:56,  1.49it/s]
1|1478|Loss: 1.0841299295425415:  23%|██▎       | 1477/6470 [19:48<55:56,  1.49it/s]
1|1478|Loss: 1.0841299295425415:  23%|██▎       | 1478/6470 [19:48<56:51,  1.46it/s]
1|1479|Loss: 0.9968383312225342:  23%|██▎       | 1478/6470 [19:49<56:51,  1.46it/s]
1|1479|Loss: 0.9968383312225342:  23%|██▎       | 1479/6470 [19:49<57:19,  1.45it/s]
1|1480|Loss: 1.045514702796936:  23%|██▎       | 1479/6470 [19:49<57:19,  1.45it/s] 
1|1480|Loss: 1.045514702796936:  23%|██▎       | 1480/6470 [19:50<57:41,  1.44it/s]
1|1481|Loss: 0.8730115294456482:  23%|██▎       | 1480/6470 [19:50<57:41,  1.44it/s]
1|1481|Loss: 0.8730115294456482:  23%|██▎       | 1481/6470 [19:50<55:20,  1.50it/s]
1|1482|Loss: 0.8520846366882324:  23%|██▎       | 1481/6470 [19:51<55:20,  1.50it/s]
1|1482|Loss: 0.8520846366882324:  23%|██▎       | 1482/6470 [19:51<56:32,  1.47it/s]
1|1483|Loss: 0.8859954476356506:  23%|██▎       | 1482/6470 [19:51<56:32,  1.47it/s]
1|1483|Loss: 0.8859954476356506:  23%|██▎       | 1483/6470 [19:52<57:09,  1.45it/s]
1|1484|Loss: 1.0640301704406738:  23%|██▎       | 1483/6470 [19:52<57:09,  1.45it/s]
1|1484|Loss: 1.0640301704406738:  23%|██▎       | 1484/6470 [19:52<54:26,  1.53it/s]
1|1485|Loss: 1.0485459566116333:  23%|██▎       | 1484/6470 [19:53<54:26,  1.53it/s]
1|1485|Loss: 1.0485459566116333:  23%|██▎       | 1485/6470 [19:53<52:34,  1.58it/s]
1|1486|Loss: 0.9824004173278809:  23%|██▎       | 1485/6470 [19:53<52:34,  1.58it/s]
1|1486|Loss: 0.9824004173278809:  23%|██▎       | 1486/6470 [19:53<50:40,  1.64it/s]
1|1487|Loss: 0.8855883479118347:  23%|██▎       | 1486/6470 [19:54<50:40,  1.64it/s]
1|1487|Loss: 0.8855883479118347:  23%|██▎       | 1487/6470 [19:54<51:13,  1.62it/s]
1|1488|Loss: 0.968110978603363:  23%|██▎       | 1487/6470 [19:54<51:13,  1.62it/s] 
1|1488|Loss: 0.968110978603363:  23%|██▎       | 1488/6470 [19:55<49:19,  1.68it/s]
1|1489|Loss: 0.9457629323005676:  23%|██▎       | 1488/6470 [19:55<49:19,  1.68it/s]
1|1489|Loss: 0.9457629323005676:  23%|██▎       | 1489/6470 [19:55<49:23,  1.68it/s]
1|1490|Loss: 0.9649475812911987:  23%|██▎       | 1489/6470 [19:56<49:23,  1.68it/s]
1|1490|Loss: 0.9649475812911987:  23%|██▎       | 1490/6470 [19:56<51:35,  1.61it/s]
1|1491|Loss: 0.8908830881118774:  23%|██▎       | 1490/6470 [19:56<51:35,  1.61it/s]
1|1491|Loss: 0.8908830881118774:  23%|██▎       | 1491/6470 [19:56<52:26,  1.58it/s]
1|1492|Loss: 0.9621607661247253:  23%|██▎       | 1491/6470 [19:57<52:26,  1.58it/s]
1|1492|Loss: 0.9621607661247253:  23%|██▎       | 1492/6470 [19:57<52:45,  1.57it/s]
1|1493|Loss: 0.9792624711990356:  23%|██▎       | 1492/6470 [19:57<52:45,  1.57it/s]
1|1493|Loss: 0.9792624711990356:  23%|██▎       | 1493/6470 [19:58<53:59,  1.54it/s]
1|1494|Loss: 0.9147129058837891:  23%|██▎       | 1493/6470 [19:58<53:59,  1.54it/s]
1|1494|Loss: 0.9147129058837891:  23%|██▎       | 1494/6470 [19:58<53:58,  1.54it/s]
1|1495|Loss: 0.8061798214912415:  23%|██▎       | 1494/6470 [19:59<53:58,  1.54it/s]
1|1495|Loss: 0.8061798214912415:  23%|██▎       | 1495/6470 [19:59<52:25,  1.58it/s]
1|1496|Loss: 1.066145896911621:  23%|██▎       | 1495/6470 [19:59<52:25,  1.58it/s] 
1|1496|Loss: 1.066145896911621:  23%|██▎       | 1496/6470 [20:00<54:16,  1.53it/s]
1|1497|Loss: 0.7897306680679321:  23%|██▎       | 1496/6470 [20:00<54:16,  1.53it/s]
1|1497|Loss: 0.7897306680679321:  23%|██▎       | 1497/6470 [20:00<52:15,  1.59it/s]
1|1498|Loss: 1.0078108310699463:  23%|██▎       | 1497/6470 [20:01<52:15,  1.59it/s]
1|1498|Loss: 1.0078108310699463:  23%|██▎       | 1498/6470 [20:01<53:52,  1.54it/s]
1|1499|Loss: 0.962424635887146:  23%|██▎       | 1498/6470 [20:01<53:52,  1.54it/s] 
1|1499|Loss: 0.962424635887146:  23%|██▎       | 1499/6470 [20:02<57:50,  1.43it/s]
1|1500|Loss: 0.9481469988822937:  23%|██▎       | 1499/6470 [20:02<57:50,  1.43it/s]
1|1500|Loss: 0.9481469988822937:  23%|██▎       | 1500/6470 [20:03<58:31,  1.42it/s]
1|1501|Loss: 1.0204371213912964:  23%|██▎       | 1500/6470 [20:03<58:31,  1.42it/s]
1|1501|Loss: 1.0204371213912964:  23%|██▎       | 1501/6470 [20:03<55:52,  1.48it/s]
1|1502|Loss: 0.9004638195037842:  23%|██▎       | 1501/6470 [20:04<55:52,  1.48it/s]
1|1502|Loss: 0.9004638195037842:  23%|██▎       | 1502/6470 [20:04<59:00,  1.40it/s]
1|1503|Loss: 0.8989526033401489:  23%|██▎       | 1502/6470 [20:04<59:00,  1.40it/s]
1|1503|Loss: 0.8989526033401489:  23%|██▎       | 1503/6470 [20:05<56:13,  1.47it/s]
1|1504|Loss: 0.9085623621940613:  23%|██▎       | 1503/6470 [20:05<56:13,  1.47it/s]
1|1504|Loss: 0.9085623621940613:  23%|██▎       | 1504/6470 [20:05<56:45,  1.46it/s]
1|1505|Loss: 0.8657918572425842:  23%|██▎       | 1504/6470 [20:06<56:45,  1.46it/s]
1|1505|Loss: 0.8657918572425842:  23%|██▎       | 1505/6470 [20:06<54:10,  1.53it/s]
1|1506|Loss: 1.088802695274353:  23%|██▎       | 1505/6470 [20:06<54:10,  1.53it/s] 
1|1506|Loss: 1.088802695274353:  23%|██▎       | 1506/6470 [20:06<52:15,  1.58it/s]
1|1507|Loss: 0.9123253226280212:  23%|██▎       | 1506/6470 [20:07<52:15,  1.58it/s]
1|1507|Loss: 0.9123253226280212:  23%|██▎       | 1507/6470 [20:07<54:19,  1.52it/s]
1|1508|Loss: 1.139706015586853:  23%|██▎       | 1507/6470 [20:08<54:19,  1.52it/s] 
1|1508|Loss: 1.139706015586853:  23%|██▎       | 1508/6470 [20:08<53:13,  1.55it/s]
1|1509|Loss: 0.9638733267784119:  23%|██▎       | 1508/6470 [20:08<53:13,  1.55it/s]
1|1509|Loss: 0.9638733267784119:  23%|██▎       | 1509/6470 [20:08<51:41,  1.60it/s]
1|1510|Loss: 1.0474261045455933:  23%|██▎       | 1509/6470 [20:09<51:41,  1.60it/s]
1|1510|Loss: 1.0474261045455933:  23%|██▎       | 1510/6470 [20:09<53:10,  1.55it/s]
1|1511|Loss: 1.0178886651992798:  23%|██▎       | 1510/6470 [20:09<53:10,  1.55it/s]
1|1511|Loss: 1.0178886651992798:  23%|██▎       | 1511/6470 [20:10<53:38,  1.54it/s]
1|1512|Loss: 0.9523495435714722:  23%|██▎       | 1511/6470 [20:10<53:38,  1.54it/s]
1|1512|Loss: 0.9523495435714722:  23%|██▎       | 1512/6470 [20:10<55:28,  1.49it/s]
1|1513|Loss: 0.8127756118774414:  23%|██▎       | 1512/6470 [20:11<55:28,  1.49it/s]
1|1513|Loss: 0.8127756118774414:  23%|██▎       | 1513/6470 [20:11<1:00:01,  1.38it/s]
1|1514|Loss: 0.9530873894691467:  23%|██▎       | 1513/6470 [20:12<1:00:01,  1.38it/s]
1|1514|Loss: 0.9530873894691467:  23%|██▎       | 1514/6470 [20:12<1:03:06,  1.31it/s]
1|1515|Loss: 0.9540360569953918:  23%|██▎       | 1514/6470 [20:12<1:03:06,  1.31it/s]
1|1515|Loss: 0.9540360569953918:  23%|██▎       | 1515/6470 [20:13<58:34,  1.41it/s]  
1|1516|Loss: 1.0219348669052124:  23%|██▎       | 1515/6470 [20:13<58:34,  1.41it/s]
1|1516|Loss: 1.0219348669052124:  23%|██▎       | 1516/6470 [20:13<57:22,  1.44it/s]
1|1517|Loss: 1.0308223962783813:  23%|██▎       | 1516/6470 [20:14<57:22,  1.44it/s]
1|1517|Loss: 1.0308223962783813:  23%|██▎       | 1517/6470 [20:14<54:04,  1.53it/s]
1|1518|Loss: 1.0354937314987183:  23%|██▎       | 1517/6470 [20:14<54:04,  1.53it/s]
1|1518|Loss: 1.0354937314987183:  23%|██▎       | 1518/6470 [20:15<55:10,  1.50it/s]
1|1519|Loss: 1.067850112915039:  23%|██▎       | 1518/6470 [20:15<55:10,  1.50it/s] 
1|1519|Loss: 1.067850112915039:  23%|██▎       | 1519/6470 [20:15<55:04,  1.50it/s]
1|1520|Loss: 1.0233951807022095:  23%|██▎       | 1519/6470 [20:16<55:04,  1.50it/s]
1|1520|Loss: 1.0233951807022095:  23%|██▎       | 1520/6470 [20:16<53:17,  1.55it/s]
1|1521|Loss: 1.0070313215255737:  23%|██▎       | 1520/6470 [20:16<53:17,  1.55it/s]
1|1521|Loss: 1.0070313215255737:  24%|██▎       | 1521/6470 [20:16<52:23,  1.57it/s]
1|1522|Loss: 1.0309029817581177:  24%|██▎       | 1521/6470 [20:17<52:23,  1.57it/s]
1|1522|Loss: 1.0309029817581177:  24%|██▎       | 1522/6470 [20:17<52:18,  1.58it/s]
1|1523|Loss: 0.916874885559082:  24%|██▎       | 1522/6470 [20:17<52:18,  1.58it/s] 
1|1523|Loss: 0.916874885559082:  24%|██▎       | 1523/6470 [20:18<53:42,  1.54it/s]
1|1524|Loss: 0.8714064359664917:  24%|██▎       | 1523/6470 [20:18<53:42,  1.54it/s]
1|1524|Loss: 0.8714064359664917:  24%|██▎       | 1524/6470 [20:19<56:02,  1.47it/s]
1|1525|Loss: 1.0095576047897339:  24%|██▎       | 1524/6470 [20:19<56:02,  1.47it/s]
1|1525|Loss: 1.0095576047897339:  24%|██▎       | 1525/6470 [20:19<55:51,  1.48it/s]
1|1526|Loss: 0.9258182048797607:  24%|██▎       | 1525/6470 [20:20<55:51,  1.48it/s]
1|1526|Loss: 0.9258182048797607:  24%|██▎       | 1526/6470 [20:20<58:34,  1.41it/s]
1|1527|Loss: 1.006016731262207:  24%|██▎       | 1526/6470 [20:20<58:34,  1.41it/s] 
1|1527|Loss: 1.006016731262207:  24%|██▎       | 1527/6470 [20:21<1:00:48,  1.35it/s]
1|1528|Loss: 0.8922881484031677:  24%|██▎       | 1527/6470 [20:21<1:00:48,  1.35it/s]
1|1528|Loss: 0.8922881484031677:  24%|██▎       | 1528/6470 [20:21<58:54,  1.40it/s]  
1|1529|Loss: 0.9962319731712341:  24%|██▎       | 1528/6470 [20:22<58:54,  1.40it/s]
1|1529|Loss: 0.9962319731712341:  24%|██▎       | 1529/6470 [20:22<58:38,  1.40it/s]
1|1530|Loss: 1.0157415866851807:  24%|██▎       | 1529/6470 [20:23<58:38,  1.40it/s]
1|1530|Loss: 1.0157415866851807:  24%|██▎       | 1530/6470 [20:23<56:03,  1.47it/s]
1|1531|Loss: 1.0057097673416138:  24%|██▎       | 1530/6470 [20:23<56:03,  1.47it/s]
1|1531|Loss: 1.0057097673416138:  24%|██▎       | 1531/6470 [20:23<55:51,  1.47it/s]
1|1532|Loss: 1.054789423942566:  24%|██▎       | 1531/6470 [20:24<55:51,  1.47it/s] 
1|1532|Loss: 1.054789423942566:  24%|██▎       | 1532/6470 [20:24<55:01,  1.50it/s]
1|1533|Loss: 0.9040067195892334:  24%|██▎       | 1532/6470 [20:24<55:01,  1.50it/s]
1|1533|Loss: 0.9040067195892334:  24%|██▎       | 1533/6470 [20:25<53:03,  1.55it/s]
1|1534|Loss: 0.954495370388031:  24%|██▎       | 1533/6470 [20:25<53:03,  1.55it/s] 
1|1534|Loss: 0.954495370388031:  24%|██▎       | 1534/6470 [20:25<52:15,  1.57it/s]
1|1535|Loss: 0.9212403297424316:  24%|██▎       | 1534/6470 [20:26<52:15,  1.57it/s]
1|1535|Loss: 0.9212403297424316:  24%|██▎       | 1535/6470 [20:26<54:12,  1.52it/s]
1|1536|Loss: 0.984150230884552:  24%|██▎       | 1535/6470 [20:26<54:12,  1.52it/s] 
1|1536|Loss: 0.984150230884552:  24%|██▎       | 1536/6470 [20:27<56:08,  1.46it/s]
1|1537|Loss: 1.0998001098632812:  24%|██▎       | 1536/6470 [20:27<56:08,  1.46it/s]
1|1537|Loss: 1.0998001098632812:  24%|██▍       | 1537/6470 [20:27<54:40,  1.50it/s]
1|1538|Loss: 0.9850207567214966:  24%|██▍       | 1537/6470 [20:28<54:40,  1.50it/s]
1|1538|Loss: 0.9850207567214966:  24%|██▍       | 1538/6470 [20:28<53:24,  1.54it/s]
1|1539|Loss: 0.9317248463630676:  24%|██▍       | 1538/6470 [20:28<53:24,  1.54it/s]
1|1539|Loss: 0.9317248463630676:  24%|██▍       | 1539/6470 [20:29<55:17,  1.49it/s]
1|1540|Loss: 1.0213814973831177:  24%|██▍       | 1539/6470 [20:29<55:17,  1.49it/s]
1|1540|Loss: 1.0213814973831177:  24%|██▍       | 1540/6470 [20:29<56:24,  1.46it/s]
1|1541|Loss: 1.037585735321045:  24%|██▍       | 1540/6470 [20:30<56:24,  1.46it/s] 
1|1541|Loss: 1.037585735321045:  24%|██▍       | 1541/6470 [20:30<56:41,  1.45it/s]
1|1542|Loss: 1.0195165872573853:  24%|██▍       | 1541/6470 [20:30<56:41,  1.45it/s]
1|1542|Loss: 1.0195165872573853:  24%|██▍       | 1542/6470 [20:31<53:49,  1.53it/s]
1|1543|Loss: 0.8863467574119568:  24%|██▍       | 1542/6470 [20:31<53:49,  1.53it/s]
1|1543|Loss: 0.8863467574119568:  24%|██▍       | 1543/6470 [20:31<52:56,  1.55it/s]
1|1544|Loss: 1.0732076168060303:  24%|██▍       | 1543/6470 [20:32<52:56,  1.55it/s]
1|1544|Loss: 1.0732076168060303:  24%|██▍       | 1544/6470 [20:32<54:15,  1.51it/s]
1|1545|Loss: 1.0083638429641724:  24%|██▍       | 1544/6470 [20:32<54:15,  1.51it/s]
1|1545|Loss: 1.0083638429641724:  24%|██▍       | 1545/6470 [20:33<56:04,  1.46it/s]
1|1546|Loss: 0.954509973526001:  24%|██▍       | 1545/6470 [20:33<56:04,  1.46it/s] 
1|1546|Loss: 0.954509973526001:  24%|██▍       | 1546/6470 [20:33<55:38,  1.48it/s]
1|1547|Loss: 0.818157434463501:  24%|██▍       | 1546/6470 [20:34<55:38,  1.48it/s]
1|1547|Loss: 0.818157434463501:  24%|██▍       | 1547/6470 [20:34<55:46,  1.47it/s]
1|1548|Loss: 0.9334431290626526:  24%|██▍       | 1547/6470 [20:35<55:46,  1.47it/s]
1|1548|Loss: 0.9334431290626526:  24%|██▍       | 1548/6470 [20:35<56:23,  1.45it/s]
1|1549|Loss: 0.89237380027771:  24%|██▍       | 1548/6470 [20:35<56:23,  1.45it/s]  
1|1549|Loss: 0.89237380027771:  24%|██▍       | 1549/6470 [20:36<56:45,  1.45it/s]
1|1550|Loss: 1.004113793373108:  24%|██▍       | 1549/6470 [20:36<56:45,  1.45it/s]
1|1550|Loss: 1.004113793373108:  24%|██▍       | 1550/6470 [20:36<55:55,  1.47it/s]
1|1551|Loss: 1.1020534038543701:  24%|██▍       | 1550/6470 [20:37<55:55,  1.47it/s]
1|1551|Loss: 1.1020534038543701:  24%|██▍       | 1551/6470 [20:37<53:09,  1.54it/s]
1|1552|Loss: 0.8525952696800232:  24%|██▍       | 1551/6470 [20:37<53:09,  1.54it/s]
1|1552|Loss: 0.8525952696800232:  24%|██▍       | 1552/6470 [20:38<56:20,  1.45it/s]
1|1553|Loss: 1.0344661474227905:  24%|██▍       | 1552/6470 [20:38<56:20,  1.45it/s]
1|1553|Loss: 1.0344661474227905:  24%|██▍       | 1553/6470 [20:38<1:00:16,  1.36it/s]
1|1554|Loss: 1.022240400314331:  24%|██▍       | 1553/6470 [20:39<1:00:16,  1.36it/s] 
1|1554|Loss: 1.022240400314331:  24%|██▍       | 1554/6470 [20:39<1:02:45,  1.31it/s]
1|1555|Loss: 0.9366559982299805:  24%|██▍       | 1554/6470 [20:40<1:02:45,  1.31it/s]
1|1555|Loss: 0.9366559982299805:  24%|██▍       | 1555/6470 [20:40<1:03:36,  1.29it/s]
1|1556|Loss: 0.974003791809082:  24%|██▍       | 1555/6470 [20:40<1:03:36,  1.29it/s] 
1|1556|Loss: 0.974003791809082:  24%|██▍       | 1556/6470 [20:41<59:22,  1.38it/s]  
1|1557|Loss: 1.1058906316757202:  24%|██▍       | 1556/6470 [20:41<59:22,  1.38it/s]
1|1557|Loss: 1.1058906316757202:  24%|██▍       | 1557/6470 [20:41<56:51,  1.44it/s]
1|1558|Loss: 0.9396125674247742:  24%|██▍       | 1557/6470 [20:42<56:51,  1.44it/s]
1|1558|Loss: 0.9396125674247742:  24%|██▍       | 1558/6470 [20:42<57:38,  1.42it/s]
1|1559|Loss: 0.7604709267616272:  24%|██▍       | 1558/6470 [20:42<57:38,  1.42it/s]
1|1559|Loss: 0.7604709267616272:  24%|██▍       | 1559/6470 [20:43<58:36,  1.40it/s]
1|1560|Loss: 1.007506251335144:  24%|██▍       | 1559/6470 [20:43<58:36,  1.40it/s] 
1|1560|Loss: 1.007506251335144:  24%|██▍       | 1560/6470 [20:44<1:01:14,  1.34it/s]
1|1561|Loss: 0.9284932613372803:  24%|██▍       | 1560/6470 [20:44<1:01:14,  1.34it/s]
1|1561|Loss: 0.9284932613372803:  24%|██▍       | 1561/6470 [20:44<59:45,  1.37it/s]  
1|1562|Loss: 0.8066086173057556:  24%|██▍       | 1561/6470 [20:45<59:45,  1.37it/s]
1|1562|Loss: 0.8066086173057556:  24%|██▍       | 1562/6470 [20:45<1:00:20,  1.36it/s]
1|1563|Loss: 0.9322919845581055:  24%|██▍       | 1562/6470 [20:45<1:00:20,  1.36it/s]
1|1563|Loss: 0.9322919845581055:  24%|██▍       | 1563/6470 [20:46<56:59,  1.44it/s]  
1|1564|Loss: 1.0589325428009033:  24%|██▍       | 1563/6470 [20:46<56:59,  1.44it/s]
1|1564|Loss: 1.0589325428009033:  24%|██▍       | 1564/6470 [20:46<56:27,  1.45it/s]
1|1565|Loss: 0.9756066799163818:  24%|██▍       | 1564/6470 [20:47<56:27,  1.45it/s]
1|1565|Loss: 0.9756066799163818:  24%|██▍       | 1565/6470 [20:47<56:46,  1.44it/s]
1|1566|Loss: 1.0630954504013062:  24%|██▍       | 1565/6470 [20:47<56:46,  1.44it/s]
1|1566|Loss: 1.0630954504013062:  24%|██▍       | 1566/6470 [20:48<57:37,  1.42it/s]
1|1567|Loss: 0.9728403091430664:  24%|██▍       | 1566/6470 [20:48<57:37,  1.42it/s]
1|1567|Loss: 0.9728403091430664:  24%|██▍       | 1567/6470 [20:48<59:28,  1.37it/s]
1|1568|Loss: 1.0192992687225342:  24%|██▍       | 1567/6470 [20:49<59:28,  1.37it/s]
1|1568|Loss: 1.0192992687225342:  24%|██▍       | 1568/6470 [20:49<1:00:23,  1.35it/s]
1|1569|Loss: 0.8824864029884338:  24%|██▍       | 1568/6470 [20:50<1:00:23,  1.35it/s]
1|1569|Loss: 0.8824864029884338:  24%|██▍       | 1569/6470 [20:50<57:18,  1.43it/s]  
1|1570|Loss: 0.9367233514785767:  24%|██▍       | 1569/6470 [20:50<57:18,  1.43it/s]
1|1570|Loss: 0.9367233514785767:  24%|██▍       | 1570/6470 [20:51<58:24,  1.40it/s]
1|1571|Loss: 0.9786807298660278:  24%|██▍       | 1570/6470 [20:51<58:24,  1.40it/s]
1|1571|Loss: 0.9786807298660278:  24%|██▍       | 1571/6470 [20:51<54:45,  1.49it/s]
1|1572|Loss: 0.9583168029785156:  24%|██▍       | 1571/6470 [20:52<54:45,  1.49it/s]
1|1572|Loss: 0.9583168029785156:  24%|██▍       | 1572/6470 [20:52<54:26,  1.50it/s]
1|1573|Loss: 1.0247995853424072:  24%|██▍       | 1572/6470 [20:52<54:26,  1.50it/s]
1|1573|Loss: 1.0247995853424072:  24%|██▍       | 1573/6470 [20:53<57:26,  1.42it/s]
1|1574|Loss: 0.9441443085670471:  24%|██▍       | 1573/6470 [20:53<57:26,  1.42it/s]
1|1574|Loss: 0.9441443085670471:  24%|██▍       | 1574/6470 [20:53<58:45,  1.39it/s]
1|1575|Loss: 1.0827339887619019:  24%|██▍       | 1574/6470 [20:54<58:45,  1.39it/s]
1|1575|Loss: 1.0827339887619019:  24%|██▍       | 1575/6470 [20:54<57:22,  1.42it/s]
1|1576|Loss: 1.0806549787521362:  24%|██▍       | 1575/6470 [20:54<57:22,  1.42it/s]
1|1576|Loss: 1.0806549787521362:  24%|██▍       | 1576/6470 [20:55<58:00,  1.41it/s]
1|1577|Loss: 0.9462609887123108:  24%|██▍       | 1576/6470 [20:55<58:00,  1.41it/s]
1|1577|Loss: 0.9462609887123108:  24%|██▍       | 1577/6470 [20:55<55:27,  1.47it/s]
1|1578|Loss: 1.0829497575759888:  24%|██▍       | 1577/6470 [20:56<55:27,  1.47it/s]
1|1578|Loss: 1.0829497575759888:  24%|██▍       | 1578/6470 [20:56<56:20,  1.45it/s]
1|1579|Loss: 0.9615261554718018:  24%|██▍       | 1578/6470 [20:57<56:20,  1.45it/s]
1|1579|Loss: 0.9615261554718018:  24%|██▍       | 1579/6470 [20:57<57:32,  1.42it/s]
1|1580|Loss: 1.032614827156067:  24%|██▍       | 1579/6470 [20:57<57:32,  1.42it/s] 
1|1580|Loss: 1.032614827156067:  24%|██▍       | 1580/6470 [20:57<55:10,  1.48it/s]
1|1581|Loss: 0.9074859023094177:  24%|██▍       | 1580/6470 [20:58<55:10,  1.48it/s]
1|1581|Loss: 0.9074859023094177:  24%|██▍       | 1581/6470 [20:58<55:18,  1.47it/s]
1|1582|Loss: 0.8879357576370239:  24%|██▍       | 1581/6470 [20:58<55:18,  1.47it/s]
1|1582|Loss: 0.8879357576370239:  24%|██▍       | 1582/6470 [20:59<53:21,  1.53it/s]
1|1583|Loss: 1.0450432300567627:  24%|██▍       | 1582/6470 [20:59<53:21,  1.53it/s]
1|1583|Loss: 1.0450432300567627:  24%|██▍       | 1583/6470 [20:59<52:45,  1.54it/s]
1|1584|Loss: 0.9036940336227417:  24%|██▍       | 1583/6470 [21:00<52:45,  1.54it/s]
1|1584|Loss: 0.9036940336227417:  24%|██▍       | 1584/6470 [21:00<49:58,  1.63it/s]
1|1585|Loss: 0.9097764492034912:  24%|██▍       | 1584/6470 [21:00<49:58,  1.63it/s]
1|1585|Loss: 0.9097764492034912:  24%|██▍       | 1585/6470 [21:01<53:46,  1.51it/s]
1|1586|Loss: 0.9937816262245178:  24%|██▍       | 1585/6470 [21:01<53:46,  1.51it/s]
1|1586|Loss: 0.9937816262245178:  25%|██▍       | 1586/6470 [21:01<55:24,  1.47it/s]
1|1587|Loss: 0.8326860070228577:  25%|██▍       | 1586/6470 [21:02<55:24,  1.47it/s]
1|1587|Loss: 0.8326860070228577:  25%|██▍       | 1587/6470 [21:02<56:55,  1.43it/s]
1|1588|Loss: 0.9821273684501648:  25%|██▍       | 1587/6470 [21:02<56:55,  1.43it/s]
1|1588|Loss: 0.9821273684501648:  25%|██▍       | 1588/6470 [21:03<51:09,  1.59it/s]
1|1589|Loss: 0.8675300478935242:  25%|██▍       | 1588/6470 [21:03<51:09,  1.59it/s]
1|1589|Loss: 0.8675300478935242:  25%|██▍       | 1589/6470 [21:03<55:50,  1.46it/s]
1|1590|Loss: 1.0640660524368286:  25%|██▍       | 1589/6470 [21:04<55:50,  1.46it/s]
1|1590|Loss: 1.0640660524368286:  25%|██▍       | 1590/6470 [21:04<57:54,  1.40it/s]
1|1591|Loss: 0.8688236474990845:  25%|██▍       | 1590/6470 [21:05<57:54,  1.40it/s]
1|1591|Loss: 0.8688236474990845:  25%|██▍       | 1591/6470 [21:05<58:26,  1.39it/s]
1|1592|Loss: 0.9711843132972717:  25%|██▍       | 1591/6470 [21:05<58:26,  1.39it/s]
1|1592|Loss: 0.9711843132972717:  25%|██▍       | 1592/6470 [21:05<54:45,  1.48it/s]
1|1593|Loss: 0.8865408897399902:  25%|██▍       | 1592/6470 [21:06<54:45,  1.48it/s]
1|1593|Loss: 0.8865408897399902:  25%|██▍       | 1593/6470 [21:06<52:56,  1.54it/s]
1|1594|Loss: 0.9555874466896057:  25%|██▍       | 1593/6470 [21:06<52:56,  1.54it/s]
1|1594|Loss: 0.9555874466896057:  25%|██▍       | 1594/6470 [21:07<54:19,  1.50it/s]
1|1595|Loss: 0.9167179465293884:  25%|██▍       | 1594/6470 [21:07<54:19,  1.50it/s]
1|1595|Loss: 0.9167179465293884:  25%|██▍       | 1595/6470 [21:08<56:08,  1.45it/s]
1|1596|Loss: 1.0567039251327515:  25%|██▍       | 1595/6470 [21:08<56:08,  1.45it/s]
1|1596|Loss: 1.0567039251327515:  25%|██▍       | 1596/6470 [21:08<57:11,  1.42it/s]
1|1597|Loss: 0.9831646680831909:  25%|██▍       | 1596/6470 [21:09<57:11,  1.42it/s]
1|1597|Loss: 0.9831646680831909:  25%|██▍       | 1597/6470 [21:09<57:21,  1.42it/s]
1|1598|Loss: 1.008005142211914:  25%|██▍       | 1597/6470 [21:09<57:21,  1.42it/s] 
1|1598|Loss: 1.008005142211914:  25%|██▍       | 1598/6470 [21:10<58:03,  1.40it/s]
1|1599|Loss: 0.9488329291343689:  25%|██▍       | 1598/6470 [21:10<58:03,  1.40it/s]
1|1599|Loss: 0.9488329291343689:  25%|██▍       | 1599/6470 [21:10<58:36,  1.39it/s]
1|1600|Loss: 0.890093982219696:  25%|██▍       | 1599/6470 [21:11<58:36,  1.39it/s] 
1|1600|Loss: 0.890093982219696:  25%|██▍       | 1600/6470 [21:11<59:59,  1.35it/s]
1|1601|Loss: 0.9486307501792908:  25%|██▍       | 1600/6470 [21:12<59:59,  1.35it/s]
1|1601|Loss: 0.9486307501792908:  25%|██▍       | 1601/6470 [21:12<55:45,  1.46it/s]
1|1602|Loss: 0.9321985244750977:  25%|██▍       | 1601/6470 [21:12<55:45,  1.46it/s]
1|1602|Loss: 0.9321985244750977:  25%|██▍       | 1602/6470 [21:12<55:08,  1.47it/s]
1|1603|Loss: 1.1642918586730957:  25%|██▍       | 1602/6470 [21:13<55:08,  1.47it/s]
1|1603|Loss: 1.1642918586730957:  25%|██▍       | 1603/6470 [21:13<54:34,  1.49it/s]
1|1604|Loss: 0.9617583751678467:  25%|██▍       | 1603/6470 [21:14<54:34,  1.49it/s]
1|1604|Loss: 0.9617583751678467:  25%|██▍       | 1604/6470 [21:14<55:34,  1.46it/s]
1|1605|Loss: 0.843375563621521:  25%|██▍       | 1604/6470 [21:14<55:34,  1.46it/s] 
1|1605|Loss: 0.843375563621521:  25%|██▍       | 1605/6470 [21:15<57:02,  1.42it/s]
1|1606|Loss: 1.0460973978042603:  25%|██▍       | 1605/6470 [21:15<57:02,  1.42it/s]
1|1606|Loss: 1.0460973978042603:  25%|██▍       | 1606/6470 [21:15<52:28,  1.55it/s]
1|1607|Loss: 0.8719440698623657:  25%|██▍       | 1606/6470 [21:15<52:28,  1.55it/s]
1|1607|Loss: 0.8719440698623657:  25%|██▍       | 1607/6470 [21:16<50:11,  1.61it/s]
1|1608|Loss: 0.9114397764205933:  25%|██▍       | 1607/6470 [21:16<50:11,  1.61it/s]
1|1608|Loss: 0.9114397764205933:  25%|██▍       | 1608/6470 [21:16<54:40,  1.48it/s]
1|1609|Loss: 0.9420299530029297:  25%|██▍       | 1608/6470 [21:17<54:40,  1.48it/s]
1|1609|Loss: 0.9420299530029297:  25%|██▍       | 1609/6470 [21:17<58:33,  1.38it/s]
1|1610|Loss: 0.8275865912437439:  25%|██▍       | 1609/6470 [21:18<58:33,  1.38it/s]
1|1610|Loss: 0.8275865912437439:  25%|██▍       | 1610/6470 [21:18<1:01:37,  1.31it/s]
1|1611|Loss: 1.0532571077346802:  25%|██▍       | 1610/6470 [21:19<1:01:37,  1.31it/s]
1|1611|Loss: 1.0532571077346802:  25%|██▍       | 1611/6470 [21:19<57:26,  1.41it/s]  
1|1612|Loss: 1.0363835096359253:  25%|██▍       | 1611/6470 [21:19<57:26,  1.41it/s]
1|1612|Loss: 1.0363835096359253:  25%|██▍       | 1612/6470 [21:19<55:06,  1.47it/s]
1|1613|Loss: 1.0366666316986084:  25%|██▍       | 1612/6470 [21:20<55:06,  1.47it/s]
1|1613|Loss: 1.0366666316986084:  25%|██▍       | 1613/6470 [21:20<52:01,  1.56it/s]
1|1614|Loss: 0.914881706237793:  25%|██▍       | 1613/6470 [21:20<52:01,  1.56it/s] 
1|1614|Loss: 0.914881706237793:  25%|██▍       | 1614/6470 [21:21<51:25,  1.57it/s]
1|1615|Loss: 0.9120557904243469:  25%|██▍       | 1614/6470 [21:21<51:25,  1.57it/s]
1|1615|Loss: 0.9120557904243469:  25%|██▍       | 1615/6470 [21:21<50:26,  1.60it/s]
1|1616|Loss: 0.9449976682662964:  25%|██▍       | 1615/6470 [21:21<50:26,  1.60it/s]
1|1616|Loss: 0.9449976682662964:  25%|██▍       | 1616/6470 [21:22<48:33,  1.67it/s]
1|1617|Loss: 1.2719931602478027:  25%|██▍       | 1616/6470 [21:22<48:33,  1.67it/s]
1|1617|Loss: 1.2719931602478027:  25%|██▍       | 1617/6470 [21:22<50:48,  1.59it/s]
1|1618|Loss: 1.0814051628112793:  25%|██▍       | 1617/6470 [21:23<50:48,  1.59it/s]
1|1618|Loss: 1.0814051628112793:  25%|██▌       | 1618/6470 [21:23<55:43,  1.45it/s]
1|1619|Loss: 1.1195727586746216:  25%|██▌       | 1618/6470 [21:24<55:43,  1.45it/s]
1|1619|Loss: 1.1195727586746216:  25%|██▌       | 1619/6470 [21:24<55:22,  1.46it/s]
1|1620|Loss: 1.097459077835083:  25%|██▌       | 1619/6470 [21:24<55:22,  1.46it/s] 
1|1620|Loss: 1.097459077835083:  25%|██▌       | 1620/6470 [21:25<56:10,  1.44it/s]
1|1621|Loss: 1.0019941329956055:  25%|██▌       | 1620/6470 [21:25<56:10,  1.44it/s]
1|1621|Loss: 1.0019941329956055:  25%|██▌       | 1621/6470 [21:25<57:13,  1.41it/s]
1|1622|Loss: 1.0067143440246582:  25%|██▌       | 1621/6470 [21:26<57:13,  1.41it/s]
1|1622|Loss: 1.0067143440246582:  25%|██▌       | 1622/6470 [21:26<55:38,  1.45it/s]
1|1623|Loss: 0.9754476547241211:  25%|██▌       | 1622/6470 [21:26<55:38,  1.45it/s]
1|1623|Loss: 0.9754476547241211:  25%|██▌       | 1623/6470 [21:27<56:18,  1.43it/s]
1|1624|Loss: 0.7014505863189697:  25%|██▌       | 1623/6470 [21:27<56:18,  1.43it/s]
1|1624|Loss: 0.7014505863189697:  25%|██▌       | 1624/6470 [21:28<59:31,  1.36it/s]
1|1625|Loss: 1.0197466611862183:  25%|██▌       | 1624/6470 [21:28<59:31,  1.36it/s]
1|1625|Loss: 1.0197466611862183:  25%|██▌       | 1625/6470 [21:28<1:00:21,  1.34it/s]
1|1626|Loss: 0.8345544934272766:  25%|██▌       | 1625/6470 [21:29<1:00:21,  1.34it/s]
1|1626|Loss: 0.8345544934272766:  25%|██▌       | 1626/6470 [21:29<1:02:29,  1.29it/s]
1|1627|Loss: 0.9773324131965637:  25%|██▌       | 1626/6470 [21:30<1:02:29,  1.29it/s]
1|1627|Loss: 0.9773324131965637:  25%|██▌       | 1627/6470 [21:30<59:17,  1.36it/s]  
1|1628|Loss: 0.8698379993438721:  25%|██▌       | 1627/6470 [21:30<59:17,  1.36it/s]
1|1628|Loss: 0.8698379993438721:  25%|██▌       | 1628/6470 [21:31<1:00:35,  1.33it/s]
1|1629|Loss: 0.891185462474823:  25%|██▌       | 1628/6470 [21:31<1:00:35,  1.33it/s] 
1|1629|Loss: 0.891185462474823:  25%|██▌       | 1629/6470 [21:31<56:30,  1.43it/s]  
1|1630|Loss: 0.9805058240890503:  25%|██▌       | 1629/6470 [21:31<56:30,  1.43it/s]
1|1630|Loss: 0.9805058240890503:  25%|██▌       | 1630/6470 [21:32<56:22,  1.43it/s]
1|1631|Loss: 0.9652469754219055:  25%|██▌       | 1630/6470 [21:32<56:22,  1.43it/s]
1|1631|Loss: 0.9652469754219055:  25%|██▌       | 1631/6470 [21:33<58:20,  1.38it/s]
1|1632|Loss: 0.9171798229217529:  25%|██▌       | 1631/6470 [21:33<58:20,  1.38it/s]
1|1632|Loss: 0.9171798229217529:  25%|██▌       | 1632/6470 [21:33<1:02:26,  1.29it/s]
1|1633|Loss: 0.9020169973373413:  25%|██▌       | 1632/6470 [21:34<1:02:26,  1.29it/s]
1|1633|Loss: 0.9020169973373413:  25%|██▌       | 1633/6470 [21:34<1:04:43,  1.25it/s]
1|1634|Loss: 0.977692723274231:  25%|██▌       | 1633/6470 [21:35<1:04:43,  1.25it/s] 
1|1634|Loss: 0.977692723274231:  25%|██▌       | 1634/6470 [21:35<1:01:29,  1.31it/s]
1|1635|Loss: 1.1380807161331177:  25%|██▌       | 1634/6470 [21:35<1:01:29,  1.31it/s]
1|1635|Loss: 1.1380807161331177:  25%|██▌       | 1635/6470 [21:36<55:50,  1.44it/s]  
1|1636|Loss: 1.087667465209961:  25%|██▌       | 1635/6470 [21:36<55:50,  1.44it/s] 
1|1636|Loss: 1.087667465209961:  25%|██▌       | 1636/6470 [21:36<51:21,  1.57it/s]
1|1637|Loss: 1.1554232835769653:  25%|██▌       | 1636/6470 [21:36<51:21,  1.57it/s]
1|1637|Loss: 1.1554232835769653:  25%|██▌       | 1637/6470 [21:37<50:45,  1.59it/s]
1|1638|Loss: 0.9351756572723389:  25%|██▌       | 1637/6470 [21:37<50:45,  1.59it/s]
1|1638|Loss: 0.9351756572723389:  25%|██▌       | 1638/6470 [21:37<52:23,  1.54it/s]
1|1639|Loss: 1.0123523473739624:  25%|██▌       | 1638/6470 [21:38<52:23,  1.54it/s]
1|1639|Loss: 1.0123523473739624:  25%|██▌       | 1639/6470 [21:38<56:53,  1.42it/s]
1|1640|Loss: 1.0945698022842407:  25%|██▌       | 1639/6470 [21:39<56:53,  1.42it/s]
1|1640|Loss: 1.0945698022842407:  25%|██▌       | 1640/6470 [21:39<55:19,  1.46it/s]
1|1641|Loss: 0.9401247501373291:  25%|██▌       | 1640/6470 [21:39<55:19,  1.46it/s]
1|1641|Loss: 0.9401247501373291:  25%|██▌       | 1641/6470 [21:39<53:44,  1.50it/s]
1|1642|Loss: 1.0598347187042236:  25%|██▌       | 1641/6470 [21:40<53:44,  1.50it/s]
1|1642|Loss: 1.0598347187042236:  25%|██▌       | 1642/6470 [21:40<54:22,  1.48it/s]
1|1643|Loss: 0.9421093463897705:  25%|██▌       | 1642/6470 [21:41<54:22,  1.48it/s]
1|1643|Loss: 0.9421093463897705:  25%|██▌       | 1643/6470 [21:41<53:41,  1.50it/s]
1|1644|Loss: 1.054418921470642:  25%|██▌       | 1643/6470 [21:41<53:41,  1.50it/s] 
1|1644|Loss: 1.054418921470642:  25%|██▌       | 1644/6470 [21:41<51:42,  1.56it/s]
1|1645|Loss: 0.9447044134140015:  25%|██▌       | 1644/6470 [21:42<51:42,  1.56it/s]
1|1645|Loss: 0.9447044134140015:  25%|██▌       | 1645/6470 [21:42<53:10,  1.51it/s]
1|1646|Loss: 0.8843106031417847:  25%|██▌       | 1645/6470 [21:43<53:10,  1.51it/s]
1|1646|Loss: 0.8843106031417847:  25%|██▌       | 1646/6470 [21:43<54:10,  1.48it/s]
1|1647|Loss: 0.8487046360969543:  25%|██▌       | 1646/6470 [21:43<54:10,  1.48it/s]
1|1647|Loss: 0.8487046360969543:  25%|██▌       | 1647/6470 [21:43<53:40,  1.50it/s]
1|1648|Loss: 0.865496814250946:  25%|██▌       | 1647/6470 [21:44<53:40,  1.50it/s] 
1|1648|Loss: 0.865496814250946:  25%|██▌       | 1648/6470 [21:44<55:47,  1.44it/s]
1|1649|Loss: 0.9263895750045776:  25%|██▌       | 1648/6470 [21:45<55:47,  1.44it/s]
1|1649|Loss: 0.9263895750045776:  25%|██▌       | 1649/6470 [21:45<56:32,  1.42it/s]
1|1650|Loss: 0.8314595222473145:  25%|██▌       | 1649/6470 [21:45<56:32,  1.42it/s]
1|1650|Loss: 0.8314595222473145:  26%|██▌       | 1650/6470 [21:46<57:30,  1.40it/s]
1|1651|Loss: 1.1522406339645386:  26%|██▌       | 1650/6470 [21:46<57:30,  1.40it/s]
1|1651|Loss: 1.1522406339645386:  26%|██▌       | 1651/6470 [21:46<56:20,  1.43it/s]
1|1652|Loss: 0.9921396970748901:  26%|██▌       | 1651/6470 [21:47<56:20,  1.43it/s]
1|1652|Loss: 0.9921396970748901:  26%|██▌       | 1652/6470 [21:47<56:33,  1.42it/s]
1|1653|Loss: 0.9704993963241577:  26%|██▌       | 1652/6470 [21:47<56:33,  1.42it/s]
1|1653|Loss: 0.9704993963241577:  26%|██▌       | 1653/6470 [21:48<52:27,  1.53it/s]
1|1654|Loss: 1.0288316011428833:  26%|██▌       | 1653/6470 [21:48<52:27,  1.53it/s]
1|1654|Loss: 1.0288316011428833:  26%|██▌       | 1654/6470 [21:48<54:45,  1.47it/s]
1|1655|Loss: 0.986481785774231:  26%|██▌       | 1654/6470 [21:49<54:45,  1.47it/s] 
1|1655|Loss: 0.986481785774231:  26%|██▌       | 1655/6470 [21:49<59:01,  1.36it/s]
1|1656|Loss: 0.9691709876060486:  26%|██▌       | 1655/6470 [21:50<59:01,  1.36it/s]
1|1656|Loss: 0.9691709876060486:  26%|██▌       | 1656/6470 [21:50<1:01:15,  1.31it/s]
1|1657|Loss: 0.9610481858253479:  26%|██▌       | 1656/6470 [21:50<1:01:15,  1.31it/s]
1|1657|Loss: 0.9610481858253479:  26%|██▌       | 1657/6470 [21:51<59:20,  1.35it/s]  
1|1658|Loss: 0.9783110022544861:  26%|██▌       | 1657/6470 [21:51<59:20,  1.35it/s]
1|1658|Loss: 0.9783110022544861:  26%|██▌       | 1658/6470 [21:51<58:52,  1.36it/s]
1|1659|Loss: 0.9986671805381775:  26%|██▌       | 1658/6470 [21:52<58:52,  1.36it/s]
1|1659|Loss: 0.9986671805381775:  26%|██▌       | 1659/6470 [21:52<59:59,  1.34it/s]
1|1660|Loss: 1.0438568592071533:  26%|██▌       | 1659/6470 [21:53<59:59,  1.34it/s]
1|1660|Loss: 1.0438568592071533:  26%|██▌       | 1660/6470 [21:53<1:00:07,  1.33it/s]
1|1661|Loss: 1.0190061330795288:  26%|██▌       | 1660/6470 [21:53<1:00:07,  1.33it/s]
1|1661|Loss: 1.0190061330795288:  26%|██▌       | 1661/6470 [21:54<57:37,  1.39it/s]  
1|1662|Loss: 0.9592573642730713:  26%|██▌       | 1661/6470 [21:54<57:37,  1.39it/s]
1|1662|Loss: 0.9592573642730713:  26%|██▌       | 1662/6470 [21:54<55:15,  1.45it/s]
1|1663|Loss: 0.9531413912773132:  26%|██▌       | 1662/6470 [21:55<55:15,  1.45it/s]
1|1663|Loss: 0.9531413912773132:  26%|██▌       | 1663/6470 [21:55<55:57,  1.43it/s]
1|1664|Loss: 0.702110230922699:  26%|██▌       | 1663/6470 [21:55<55:57,  1.43it/s] 
1|1664|Loss: 0.702110230922699:  26%|██▌       | 1664/6470 [21:56<1:01:25,  1.30it/s]
1|1665|Loss: 0.7486600875854492:  26%|██▌       | 1664/6470 [21:56<1:01:25,  1.30it/s]
1|1665|Loss: 0.7486600875854492:  26%|██▌       | 1665/6470 [21:57<1:04:42,  1.24it/s]
1|1666|Loss: 0.9202219247817993:  26%|██▌       | 1665/6470 [21:57<1:04:42,  1.24it/s]
1|1666|Loss: 0.9202219247817993:  26%|██▌       | 1666/6470 [21:57<1:01:35,  1.30it/s]
1|1667|Loss: 1.0843095779418945:  26%|██▌       | 1666/6470 [21:58<1:01:35,  1.30it/s]
1|1667|Loss: 1.0843095779418945:  26%|██▌       | 1667/6470 [21:58<58:25,  1.37it/s]  
1|1668|Loss: 1.0026451349258423:  26%|██▌       | 1667/6470 [21:58<58:25,  1.37it/s]
1|1668|Loss: 1.0026451349258423:  26%|██▌       | 1668/6470 [21:59<52:40,  1.52it/s]
1|1669|Loss: 0.9779787659645081:  26%|██▌       | 1668/6470 [21:59<52:40,  1.52it/s]
1|1669|Loss: 0.9779787659645081:  26%|██▌       | 1669/6470 [21:59<54:39,  1.46it/s]
1|1670|Loss: 1.016335368156433:  26%|██▌       | 1669/6470 [22:00<54:39,  1.46it/s] 
1|1670|Loss: 1.016335368156433:  26%|██▌       | 1670/6470 [22:00<58:07,  1.38it/s]
1|1671|Loss: 1.0265613794326782:  26%|██▌       | 1670/6470 [22:01<58:07,  1.38it/s]
1|1671|Loss: 1.0265613794326782:  26%|██▌       | 1671/6470 [22:01<1:00:53,  1.31it/s]
1|1672|Loss: 0.8777095675468445:  26%|██▌       | 1671/6470 [22:01<1:00:53,  1.31it/s]
1|1672|Loss: 0.8777095675468445:  26%|██▌       | 1672/6470 [22:02<56:26,  1.42it/s]  
1|1673|Loss: 0.9896978139877319:  26%|██▌       | 1672/6470 [22:02<56:26,  1.42it/s]
1|1673|Loss: 0.9896978139877319:  26%|██▌       | 1673/6470 [22:02<56:10,  1.42it/s]
1|1674|Loss: 1.002036213874817:  26%|██▌       | 1673/6470 [22:03<56:10,  1.42it/s] 
1|1674|Loss: 1.002036213874817:  26%|██▌       | 1674/6470 [22:03<56:37,  1.41it/s]
1|1675|Loss: 0.9601078033447266:  26%|██▌       | 1674/6470 [22:03<56:37,  1.41it/s]
1|1675|Loss: 0.9601078033447266:  26%|██▌       | 1675/6470 [22:04<53:05,  1.51it/s]
1|1676|Loss: 0.9392810463905334:  26%|██▌       | 1675/6470 [22:04<53:05,  1.51it/s]
1|1676|Loss: 0.9392810463905334:  26%|██▌       | 1676/6470 [22:04<51:29,  1.55it/s]
1|1677|Loss: 0.8643057942390442:  26%|██▌       | 1676/6470 [22:05<51:29,  1.55it/s]
1|1677|Loss: 0.8643057942390442:  26%|██▌       | 1677/6470 [22:05<54:58,  1.45it/s]
1|1678|Loss: 0.9657443165779114:  26%|██▌       | 1677/6470 [22:05<54:58,  1.45it/s]
1|1678|Loss: 0.9657443165779114:  26%|██▌       | 1678/6470 [22:06<58:31,  1.36it/s]
1|1679|Loss: 0.9233078360557556:  26%|██▌       | 1678/6470 [22:06<58:31,  1.36it/s]
1|1679|Loss: 0.9233078360557556:  26%|██▌       | 1679/6470 [22:07<1:01:25,  1.30it/s]
1|1680|Loss: 0.9044018983840942:  26%|██▌       | 1679/6470 [22:07<1:01:25,  1.30it/s]
1|1680|Loss: 0.9044018983840942:  26%|██▌       | 1680/6470 [22:08<1:04:02,  1.25it/s]
1|1681|Loss: 1.0078116655349731:  26%|██▌       | 1680/6470 [22:08<1:04:02,  1.25it/s]
1|1681|Loss: 1.0078116655349731:  26%|██▌       | 1681/6470 [22:08<1:03:33,  1.26it/s]
1|1682|Loss: 1.0288186073303223:  26%|██▌       | 1681/6470 [22:09<1:03:33,  1.26it/s]
1|1682|Loss: 1.0288186073303223:  26%|██▌       | 1682/6470 [22:09<1:03:20,  1.26it/s]
1|1683|Loss: 0.948331356048584:  26%|██▌       | 1682/6470 [22:09<1:03:20,  1.26it/s] 
1|1683|Loss: 0.948331356048584:  26%|██▌       | 1683/6470 [22:10<57:56,  1.38it/s]  
1|1684|Loss: 1.019359827041626:  26%|██▌       | 1683/6470 [22:10<57:56,  1.38it/s]
1|1684|Loss: 1.019359827041626:  26%|██▌       | 1684/6470 [22:10<56:29,  1.41it/s]
1|1685|Loss: 1.0038130283355713:  26%|██▌       | 1684/6470 [22:11<56:29,  1.41it/s]
1|1685|Loss: 1.0038130283355713:  26%|██▌       | 1685/6470 [22:11<56:27,  1.41it/s]
1|1686|Loss: 1.0292030572891235:  26%|██▌       | 1685/6470 [22:11<56:27,  1.41it/s]
1|1686|Loss: 1.0292030572891235:  26%|██▌       | 1686/6470 [22:12<55:14,  1.44it/s]
1|1687|Loss: 0.8353661298751831:  26%|██▌       | 1686/6470 [22:12<55:14,  1.44it/s]
1|1687|Loss: 0.8353661298751831:  26%|██▌       | 1687/6470 [22:12<55:08,  1.45it/s]
1|1688|Loss: 1.019112467765808:  26%|██▌       | 1687/6470 [22:13<55:08,  1.45it/s] 
1|1688|Loss: 1.019112467765808:  26%|██▌       | 1688/6470 [22:13<58:16,  1.37it/s]
1|1689|Loss: 1.0749353170394897:  26%|██▌       | 1688/6470 [22:14<58:16,  1.37it/s]
1|1689|Loss: 1.0749353170394897:  26%|██▌       | 1689/6470 [22:14<59:25,  1.34it/s]
1|1690|Loss: 0.8808445334434509:  26%|██▌       | 1689/6470 [22:14<59:25,  1.34it/s]
1|1690|Loss: 0.8808445334434509:  26%|██▌       | 1690/6470 [22:14<51:41,  1.54it/s]
1|1691|Loss: 1.1339644193649292:  26%|██▌       | 1690/6470 [22:15<51:41,  1.54it/s]
1|1691|Loss: 1.1339644193649292:  26%|██▌       | 1691/6470 [22:15<52:38,  1.51it/s]
1|1692|Loss: 0.8981083631515503:  26%|██▌       | 1691/6470 [22:16<52:38,  1.51it/s]
1|1692|Loss: 0.8981083631515503:  26%|██▌       | 1692/6470 [22:16<57:25,  1.39it/s]
1|1693|Loss: 0.9476325511932373:  26%|██▌       | 1692/6470 [22:16<57:25,  1.39it/s]
1|1693|Loss: 0.9476325511932373:  26%|██▌       | 1693/6470 [22:17<1:00:22,  1.32it/s]
1|1694|Loss: 0.9791771173477173:  26%|██▌       | 1693/6470 [22:17<1:00:22,  1.32it/s]
1|1694|Loss: 0.9791771173477173:  26%|██▌       | 1694/6470 [22:18<1:00:08,  1.32it/s]
1|1695|Loss: 0.9112988710403442:  26%|██▌       | 1694/6470 [22:18<1:00:08,  1.32it/s]
1|1695|Loss: 0.9112988710403442:  26%|██▌       | 1695/6470 [22:18<58:50,  1.35it/s]  
1|1696|Loss: 0.9140821695327759:  26%|██▌       | 1695/6470 [22:19<58:50,  1.35it/s]
1|1696|Loss: 0.9140821695327759:  26%|██▌       | 1696/6470 [22:19<59:34,  1.34it/s]
1|1697|Loss: 0.8499082922935486:  26%|██▌       | 1696/6470 [22:19<59:34,  1.34it/s]
1|1697|Loss: 0.8499082922935486:  26%|██▌       | 1697/6470 [22:20<58:51,  1.35it/s]
1|1698|Loss: 1.0504318475723267:  26%|██▌       | 1697/6470 [22:20<58:51,  1.35it/s]
1|1698|Loss: 1.0504318475723267:  26%|██▌       | 1698/6470 [22:21<1:00:14,  1.32it/s]
1|1699|Loss: 0.9688101410865784:  26%|██▌       | 1698/6470 [22:21<1:00:14,  1.32it/s]
1|1699|Loss: 0.9688101410865784:  26%|██▋       | 1699/6470 [22:21<59:57,  1.33it/s]  
1|1700|Loss: 0.9632691740989685:  26%|██▋       | 1699/6470 [22:22<59:57,  1.33it/s]
1|1700|Loss: 0.9632691740989685:  26%|██▋       | 1700/6470 [22:22<1:01:15,  1.30it/s]
1|1701|Loss: 0.8932799100875854:  26%|██▋       | 1700/6470 [22:23<1:01:15,  1.30it/s]
1|1701|Loss: 0.8932799100875854:  26%|██▋       | 1701/6470 [22:23<1:01:19,  1.30it/s]
1|1702|Loss: 1.0207395553588867:  26%|██▋       | 1701/6470 [22:23<1:01:19,  1.30it/s]
1|1702|Loss: 1.0207395553588867:  26%|██▋       | 1702/6470 [22:24<58:39,  1.35it/s]  
1|1703|Loss: 0.9956846833229065:  26%|██▋       | 1702/6470 [22:24<58:39,  1.35it/s]
1|1703|Loss: 0.9956846833229065:  26%|██▋       | 1703/6470 [22:24<55:57,  1.42it/s]
1|1704|Loss: 0.8781957030296326:  26%|██▋       | 1703/6470 [22:25<55:57,  1.42it/s]
1|1704|Loss: 0.8781957030296326:  26%|██▋       | 1704/6470 [22:25<55:43,  1.43it/s]
1|1705|Loss: 0.9608396291732788:  26%|██▋       | 1704/6470 [22:25<55:43,  1.43it/s]
1|1705|Loss: 0.9608396291732788:  26%|██▋       | 1705/6470 [22:26<56:54,  1.40it/s]
1|1706|Loss: 0.9393712282180786:  26%|██▋       | 1705/6470 [22:26<56:54,  1.40it/s]
1|1706|Loss: 0.9393712282180786:  26%|██▋       | 1706/6470 [22:26<55:43,  1.42it/s]
1|1707|Loss: 1.0056122541427612:  26%|██▋       | 1706/6470 [22:27<55:43,  1.42it/s]
1|1707|Loss: 1.0056122541427612:  26%|██▋       | 1707/6470 [22:27<52:50,  1.50it/s]
1|1708|Loss: 0.8212063908576965:  26%|██▋       | 1707/6470 [22:27<52:50,  1.50it/s]
1|1708|Loss: 0.8212063908576965:  26%|██▋       | 1708/6470 [22:28<53:40,  1.48it/s]
1|1709|Loss: 0.9761864542961121:  26%|██▋       | 1708/6470 [22:28<53:40,  1.48it/s]
1|1709|Loss: 0.9761864542961121:  26%|██▋       | 1709/6470 [22:28<54:58,  1.44it/s]
1|1710|Loss: 0.8601637482643127:  26%|██▋       | 1709/6470 [22:29<54:58,  1.44it/s]
1|1710|Loss: 0.8601637482643127:  26%|██▋       | 1710/6470 [22:29<56:04,  1.41it/s]
1|1711|Loss: 0.9128220677375793:  26%|██▋       | 1710/6470 [22:29<56:04,  1.41it/s]
1|1711|Loss: 0.9128220677375793:  26%|██▋       | 1711/6470 [22:30<51:28,  1.54it/s]
1|1712|Loss: 0.9082614183425903:  26%|██▋       | 1711/6470 [22:30<51:28,  1.54it/s]
1|1712|Loss: 0.9082614183425903:  26%|██▋       | 1712/6470 [22:30<49:54,  1.59it/s]
1|1713|Loss: 1.0520503520965576:  26%|██▋       | 1712/6470 [22:31<49:54,  1.59it/s]
1|1713|Loss: 1.0520503520965576:  26%|██▋       | 1713/6470 [22:31<53:32,  1.48it/s]
1|1714|Loss: 0.7113083004951477:  26%|██▋       | 1713/6470 [22:31<53:32,  1.48it/s]
1|1714|Loss: 0.7113083004951477:  26%|██▋       | 1714/6470 [22:32<1:00:29,  1.31it/s]
1|1715|Loss: 0.9587017297744751:  26%|██▋       | 1714/6470 [22:32<1:00:29,  1.31it/s]
1|1715|Loss: 0.9587017297744751:  27%|██▋       | 1715/6470 [22:33<1:03:44,  1.24it/s]
1|1716|Loss: 0.9278392791748047:  27%|██▋       | 1715/6470 [22:33<1:03:44,  1.24it/s]
1|1716|Loss: 0.9278392791748047:  27%|██▋       | 1716/6470 [22:34<1:03:08,  1.25it/s]
1|1717|Loss: 0.882879376411438:  27%|██▋       | 1716/6470 [22:34<1:03:08,  1.25it/s] 
1|1717|Loss: 0.882879376411438:  27%|██▋       | 1717/6470 [22:34<1:04:33,  1.23it/s]
1|1718|Loss: 1.091199517250061:  27%|██▋       | 1717/6470 [22:35<1:04:33,  1.23it/s]
1|1718|Loss: 1.091199517250061:  27%|██▋       | 1718/6470 [22:35<58:14,  1.36it/s]  
1|1719|Loss: 0.9128800630569458:  27%|██▋       | 1718/6470 [22:35<58:14,  1.36it/s]
1|1719|Loss: 0.9128800630569458:  27%|██▋       | 1719/6470 [22:36<56:57,  1.39it/s]
1|1720|Loss: 1.044438362121582:  27%|██▋       | 1719/6470 [22:36<56:57,  1.39it/s] 
1|1720|Loss: 1.044438362121582:  27%|██▋       | 1720/6470 [22:36<55:42,  1.42it/s]
1|1721|Loss: 1.0225270986557007:  27%|██▋       | 1720/6470 [22:37<55:42,  1.42it/s]
1|1721|Loss: 1.0225270986557007:  27%|██▋       | 1721/6470 [22:37<51:59,  1.52it/s]
1|1722|Loss: 1.0159624814987183:  27%|██▋       | 1721/6470 [22:37<51:59,  1.52it/s]
1|1722|Loss: 1.0159624814987183:  27%|██▋       | 1722/6470 [22:37<50:35,  1.56it/s]
1|1723|Loss: 0.872512698173523:  27%|██▋       | 1722/6470 [22:38<50:35,  1.56it/s] 
1|1723|Loss: 0.872512698173523:  27%|██▋       | 1723/6470 [22:38<50:23,  1.57it/s]
1|1724|Loss: 0.9980148673057556:  27%|██▋       | 1723/6470 [22:39<50:23,  1.57it/s]
1|1724|Loss: 0.9980148673057556:  27%|██▋       | 1724/6470 [22:39<54:50,  1.44it/s]
1|1725|Loss: 1.0216002464294434:  27%|██▋       | 1724/6470 [22:39<54:50,  1.44it/s]
1|1725|Loss: 1.0216002464294434:  27%|██▋       | 1725/6470 [22:40<56:34,  1.40it/s]
1|1726|Loss: 0.8879227638244629:  27%|██▋       | 1725/6470 [22:40<56:34,  1.40it/s]
1|1726|Loss: 0.8879227638244629:  27%|██▋       | 1726/6470 [22:40<55:12,  1.43it/s]
1|1727|Loss: 1.0318324565887451:  27%|██▋       | 1726/6470 [22:41<55:12,  1.43it/s]
1|1727|Loss: 1.0318324565887451:  27%|██▋       | 1727/6470 [22:41<53:47,  1.47it/s]
1|1728|Loss: 0.8362719416618347:  27%|██▋       | 1727/6470 [22:41<53:47,  1.47it/s]
1|1728|Loss: 0.8362719416618347:  27%|██▋       | 1728/6470 [22:42<52:23,  1.51it/s]
1|1729|Loss: 0.9058075547218323:  27%|██▋       | 1728/6470 [22:42<52:23,  1.51it/s]
1|1729|Loss: 0.9058075547218323:  27%|██▋       | 1729/6470 [22:42<55:51,  1.41it/s]
1|1730|Loss: 0.8841269016265869:  27%|██▋       | 1729/6470 [22:43<55:51,  1.41it/s]
1|1730|Loss: 0.8841269016265869:  27%|██▋       | 1730/6470 [22:43<51:58,  1.52it/s]
1|1731|Loss: 1.0939831733703613:  27%|██▋       | 1730/6470 [22:43<51:58,  1.52it/s]
1|1731|Loss: 1.0939831733703613:  27%|██▋       | 1731/6470 [22:44<50:32,  1.56it/s]
1|1732|Loss: 0.9490922689437866:  27%|██▋       | 1731/6470 [22:44<50:32,  1.56it/s]
1|1732|Loss: 0.9490922689437866:  27%|██▋       | 1732/6470 [22:44<52:16,  1.51it/s]
1|1733|Loss: 0.9126022458076477:  27%|██▋       | 1732/6470 [22:45<52:16,  1.51it/s]
1|1733|Loss: 0.9126022458076477:  27%|██▋       | 1733/6470 [22:45<53:13,  1.48it/s]
1|1734|Loss: 1.0062437057495117:  27%|██▋       | 1733/6470 [22:45<53:13,  1.48it/s]
1|1734|Loss: 1.0062437057495117:  27%|██▋       | 1734/6470 [22:46<50:47,  1.55it/s]
1|1735|Loss: 1.029200553894043:  27%|██▋       | 1734/6470 [22:46<50:47,  1.55it/s] 
1|1735|Loss: 1.029200553894043:  27%|██▋       | 1735/6470 [22:46<51:21,  1.54it/s]
1|1736|Loss: 0.9409276843070984:  27%|██▋       | 1735/6470 [22:47<51:21,  1.54it/s]
1|1736|Loss: 0.9409276843070984:  27%|██▋       | 1736/6470 [22:47<53:26,  1.48it/s]
1|1737|Loss: 0.8005303144454956:  27%|██▋       | 1736/6470 [22:47<53:26,  1.48it/s]
1|1737|Loss: 0.8005303144454956:  27%|██▋       | 1737/6470 [22:48<53:08,  1.48it/s]
1|1738|Loss: 0.9787078499794006:  27%|██▋       | 1737/6470 [22:48<53:08,  1.48it/s]
1|1738|Loss: 0.9787078499794006:  27%|██▋       | 1738/6470 [22:48<53:30,  1.47it/s]
1|1739|Loss: 0.8634543418884277:  27%|██▋       | 1738/6470 [22:49<53:30,  1.47it/s]
1|1739|Loss: 0.8634543418884277:  27%|██▋       | 1739/6470 [22:49<55:45,  1.41it/s]
1|1740|Loss: 1.0661488771438599:  27%|██▋       | 1739/6470 [22:49<55:45,  1.41it/s]
1|1740|Loss: 1.0661488771438599:  27%|██▋       | 1740/6470 [22:50<55:01,  1.43it/s]
1|1741|Loss: 0.9835332632064819:  27%|██▋       | 1740/6470 [22:50<55:01,  1.43it/s]
1|1741|Loss: 0.9835332632064819:  27%|██▋       | 1741/6470 [22:50<53:33,  1.47it/s]
1|1742|Loss: 0.9595369696617126:  27%|██▋       | 1741/6470 [22:51<53:33,  1.47it/s]
1|1742|Loss: 0.9595369696617126:  27%|██▋       | 1742/6470 [22:51<50:29,  1.56it/s]
1|1743|Loss: 0.9401204586029053:  27%|██▋       | 1742/6470 [22:51<50:29,  1.56it/s]
1|1743|Loss: 0.9401204586029053:  27%|██▋       | 1743/6470 [22:52<52:00,  1.51it/s]
1|1744|Loss: 0.8581715822219849:  27%|██▋       | 1743/6470 [22:52<52:00,  1.51it/s]
1|1744|Loss: 0.8581715822219849:  27%|██▋       | 1744/6470 [22:52<52:14,  1.51it/s]
1|1745|Loss: 1.152756929397583:  27%|██▋       | 1744/6470 [22:53<52:14,  1.51it/s] 
1|1745|Loss: 1.152756929397583:  27%|██▋       | 1745/6470 [22:53<49:59,  1.58it/s]
1|1746|Loss: 0.9595645666122437:  27%|██▋       | 1745/6470 [22:53<49:59,  1.58it/s]
1|1746|Loss: 0.9595645666122437:  27%|██▋       | 1746/6470 [22:54<50:52,  1.55it/s]
1|1747|Loss: 0.9732497930526733:  27%|██▋       | 1746/6470 [22:54<50:52,  1.55it/s]
1|1747|Loss: 0.9732497930526733:  27%|██▋       | 1747/6470 [22:54<55:09,  1.43it/s]
1|1748|Loss: 1.0840593576431274:  27%|██▋       | 1747/6470 [22:55<55:09,  1.43it/s]
1|1748|Loss: 1.0840593576431274:  27%|██▋       | 1748/6470 [22:55<56:06,  1.40it/s]
1|1749|Loss: 0.9278311133384705:  27%|██▋       | 1748/6470 [22:56<56:06,  1.40it/s]
1|1749|Loss: 0.9278311133384705:  27%|██▋       | 1749/6470 [22:56<54:10,  1.45it/s]
1|1750|Loss: 1.0068918466567993:  27%|██▋       | 1749/6470 [22:56<54:10,  1.45it/s]
1|1750|Loss: 1.0068918466567993:  27%|██▋       | 1750/6470 [22:57<56:20,  1.40it/s]
1|1751|Loss: 0.8561215400695801:  27%|██▋       | 1750/6470 [22:57<56:20,  1.40it/s]
1|1751|Loss: 0.8561215400695801:  27%|██▋       | 1751/6470 [22:57<55:09,  1.43it/s]
1|1752|Loss: 0.9378166794776917:  27%|██▋       | 1751/6470 [22:58<55:09,  1.43it/s]
1|1752|Loss: 0.9378166794776917:  27%|██▋       | 1752/6470 [22:58<53:14,  1.48it/s]
1|1753|Loss: 1.014435052871704:  27%|██▋       | 1752/6470 [22:58<53:14,  1.48it/s] 
1|1753|Loss: 1.014435052871704:  27%|██▋       | 1753/6470 [22:58<50:10,  1.57it/s]
1|1754|Loss: 0.9337594509124756:  27%|██▋       | 1753/6470 [22:59<50:10,  1.57it/s]
1|1754|Loss: 0.9337594509124756:  27%|██▋       | 1754/6470 [22:59<47:27,  1.66it/s]
1|1755|Loss: 1.0238934755325317:  27%|██▋       | 1754/6470 [22:59<47:27,  1.66it/s]
1|1755|Loss: 1.0238934755325317:  27%|██▋       | 1755/6470 [22:59<44:17,  1.77it/s]
1|1756|Loss: 1.0058025121688843:  27%|██▋       | 1755/6470 [23:00<44:17,  1.77it/s]
1|1756|Loss: 1.0058025121688843:  27%|██▋       | 1756/6470 [23:00<46:39,  1.68it/s]
1|1757|Loss: 0.872450590133667:  27%|██▋       | 1756/6470 [23:00<46:39,  1.68it/s] 
1|1757|Loss: 0.872450590133667:  27%|██▋       | 1757/6470 [23:01<52:05,  1.51it/s]
1|1758|Loss: 1.0430558919906616:  27%|██▋       | 1757/6470 [23:01<52:05,  1.51it/s]
1|1758|Loss: 1.0430558919906616:  27%|██▋       | 1758/6470 [23:02<52:42,  1.49it/s]
1|1759|Loss: 0.9056647419929504:  27%|██▋       | 1758/6470 [23:02<52:42,  1.49it/s]
1|1759|Loss: 0.9056647419929504:  27%|██▋       | 1759/6470 [23:02<53:10,  1.48it/s]
1|1760|Loss: 0.8783385753631592:  27%|██▋       | 1759/6470 [23:03<53:10,  1.48it/s]
1|1760|Loss: 0.8783385753631592:  27%|██▋       | 1760/6470 [23:03<56:31,  1.39it/s]
1|1761|Loss: 1.2235612869262695:  27%|██▋       | 1760/6470 [23:04<56:31,  1.39it/s]
1|1761|Loss: 1.2235612869262695:  27%|██▋       | 1761/6470 [23:04<59:15,  1.32it/s]
1|1762|Loss: 0.9439942836761475:  27%|██▋       | 1761/6470 [23:04<59:15,  1.32it/s]
1|1762|Loss: 0.9439942836761475:  27%|██▋       | 1762/6470 [23:05<57:31,  1.36it/s]
1|1763|Loss: 0.9984791278839111:  27%|██▋       | 1762/6470 [23:05<57:31,  1.36it/s]
1|1763|Loss: 0.9984791278839111:  27%|██▋       | 1763/6470 [23:05<57:00,  1.38it/s]
1|1764|Loss: 0.9174342155456543:  27%|██▋       | 1763/6470 [23:06<57:00,  1.38it/s]
1|1764|Loss: 0.9174342155456543:  27%|██▋       | 1764/6470 [23:06<58:18,  1.35it/s]
1|1765|Loss: 0.924931526184082:  27%|██▋       | 1764/6470 [23:07<58:18,  1.35it/s] 
1|1765|Loss: 0.924931526184082:  27%|██▋       | 1765/6470 [23:07<58:44,  1.33it/s]
1|1766|Loss: 1.1220924854278564:  27%|██▋       | 1765/6470 [23:07<58:44,  1.33it/s]
1|1766|Loss: 1.1220924854278564:  27%|██▋       | 1766/6470 [23:07<56:22,  1.39it/s]
1|1767|Loss: 1.0012136697769165:  27%|██▋       | 1766/6470 [23:08<56:22,  1.39it/s]
1|1767|Loss: 1.0012136697769165:  27%|██▋       | 1767/6470 [23:08<58:17,  1.34it/s]
1|1768|Loss: 0.8962962031364441:  27%|██▋       | 1767/6470 [23:09<58:17,  1.34it/s]
1|1768|Loss: 0.8962962031364441:  27%|██▋       | 1768/6470 [23:09<1:00:29,  1.30it/s]
1|1769|Loss: 0.9456931352615356:  27%|██▋       | 1768/6470 [23:10<1:00:29,  1.30it/s]
1|1769|Loss: 0.9456931352615356:  27%|██▋       | 1769/6470 [23:10<59:53,  1.31it/s]  
1|1770|Loss: 0.8064587116241455:  27%|██▋       | 1769/6470 [23:10<59:53,  1.31it/s]
1|1770|Loss: 0.8064587116241455:  27%|██▋       | 1770/6470 [23:10<54:48,  1.43it/s]
1|1771|Loss: 0.9969431757926941:  27%|██▋       | 1770/6470 [23:11<54:48,  1.43it/s]
1|1771|Loss: 0.9969431757926941:  27%|██▋       | 1771/6470 [23:11<54:21,  1.44it/s]
1|1772|Loss: 0.9620171189308167:  27%|██▋       | 1771/6470 [23:12<54:21,  1.44it/s]
1|1772|Loss: 0.9620171189308167:  27%|██▋       | 1772/6470 [23:12<54:43,  1.43it/s]
1|1773|Loss: 1.0919866561889648:  27%|██▋       | 1772/6470 [23:12<54:43,  1.43it/s]
1|1773|Loss: 1.0919866561889648:  27%|██▋       | 1773/6470 [23:12<51:12,  1.53it/s]
1|1774|Loss: 1.0289009809494019:  27%|██▋       | 1773/6470 [23:13<51:12,  1.53it/s]
1|1774|Loss: 1.0289009809494019:  27%|██▋       | 1774/6470 [23:13<46:41,  1.68it/s]
1|1775|Loss: 0.8828521370887756:  27%|██▋       | 1774/6470 [23:13<46:41,  1.68it/s]
1|1775|Loss: 0.8828521370887756:  27%|██▋       | 1775/6470 [23:14<48:21,  1.62it/s]
1|1776|Loss: 1.0181584358215332:  27%|██▋       | 1775/6470 [23:14<48:21,  1.62it/s]
1|1776|Loss: 1.0181584358215332:  27%|██▋       | 1776/6470 [23:14<49:32,  1.58it/s]
1|1777|Loss: 1.0118144750595093:  27%|██▋       | 1776/6470 [23:14<49:32,  1.58it/s]
1|1777|Loss: 1.0118144750595093:  27%|██▋       | 1777/6470 [23:15<46:08,  1.70it/s]
1|1778|Loss: 0.9721534252166748:  27%|██▋       | 1777/6470 [23:15<46:08,  1.70it/s]
1|1778|Loss: 0.9721534252166748:  27%|██▋       | 1778/6470 [23:15<46:25,  1.68it/s]
1|1779|Loss: 0.8306066393852234:  27%|██▋       | 1778/6470 [23:16<46:25,  1.68it/s]
1|1779|Loss: 0.8306066393852234:  27%|██▋       | 1779/6470 [23:16<44:03,  1.77it/s]
1|1780|Loss: 1.0134884119033813:  27%|██▋       | 1779/6470 [23:16<44:03,  1.77it/s]
1|1780|Loss: 1.0134884119033813:  28%|██▊       | 1780/6470 [23:17<49:22,  1.58it/s]
1|1781|Loss: 0.8622838854789734:  28%|██▊       | 1780/6470 [23:17<49:22,  1.58it/s]
1|1781|Loss: 0.8622838854789734:  28%|██▊       | 1781/6470 [23:17<54:26,  1.44it/s]
1|1782|Loss: 1.037023663520813:  28%|██▊       | 1781/6470 [23:18<54:26,  1.44it/s] 
1|1782|Loss: 1.037023663520813:  28%|██▊       | 1782/6470 [23:18<55:16,  1.41it/s]
1|1783|Loss: 0.8982945680618286:  28%|██▊       | 1782/6470 [23:18<55:16,  1.41it/s]
1|1783|Loss: 0.8982945680618286:  28%|██▊       | 1783/6470 [23:19<52:39,  1.48it/s]
1|1784|Loss: 0.8970547318458557:  28%|██▊       | 1783/6470 [23:19<52:39,  1.48it/s]
1|1784|Loss: 0.8970547318458557:  28%|██▊       | 1784/6470 [23:19<54:39,  1.43it/s]
1|1785|Loss: 0.9131380319595337:  28%|██▊       | 1784/6470 [23:20<54:39,  1.43it/s]
1|1785|Loss: 0.9131380319595337:  28%|██▊       | 1785/6470 [23:20<55:51,  1.40it/s]
1|1786|Loss: 0.9761377573013306:  28%|██▊       | 1785/6470 [23:21<55:51,  1.40it/s]
1|1786|Loss: 0.9761377573013306:  28%|██▊       | 1786/6470 [23:21<54:14,  1.44it/s]
1|1787|Loss: 1.000423789024353:  28%|██▊       | 1786/6470 [23:21<54:14,  1.44it/s] 
1|1787|Loss: 1.000423789024353:  28%|██▊       | 1787/6470 [23:22<54:25,  1.43it/s]
1|1788|Loss: 1.0464855432510376:  28%|██▊       | 1787/6470 [23:22<54:25,  1.43it/s]
1|1788|Loss: 1.0464855432510376:  28%|██▊       | 1788/6470 [23:22<56:26,  1.38it/s]
1|1789|Loss: 1.2733941078186035:  28%|██▊       | 1788/6470 [23:23<56:26,  1.38it/s]
1|1789|Loss: 1.2733941078186035:  28%|██▊       | 1789/6470 [23:23<1:02:05,  1.26it/s]
1|1790|Loss: 1.0235882997512817:  28%|██▊       | 1789/6470 [23:24<1:02:05,  1.26it/s]
1|1790|Loss: 1.0235882997512817:  28%|██▊       | 1790/6470 [23:24<55:20,  1.41it/s]  
1|1791|Loss: 0.8620622158050537:  28%|██▊       | 1790/6470 [23:24<55:20,  1.41it/s]
1|1791|Loss: 0.8620622158050537:  28%|██▊       | 1791/6470 [23:24<52:33,  1.48it/s]
1|1792|Loss: 1.1031526327133179:  28%|██▊       | 1791/6470 [23:25<52:33,  1.48it/s]
1|1792|Loss: 1.1031526327133179:  28%|██▊       | 1792/6470 [23:25<50:01,  1.56it/s]
1|1793|Loss: 0.9290788769721985:  28%|██▊       | 1792/6470 [23:25<50:01,  1.56it/s]
1|1793|Loss: 0.9290788769721985:  28%|██▊       | 1793/6470 [23:26<49:10,  1.59it/s]
1|1794|Loss: 0.8731568455696106:  28%|██▊       | 1793/6470 [23:26<49:10,  1.59it/s]
1|1794|Loss: 0.8731568455696106:  28%|██▊       | 1794/6470 [23:26<52:29,  1.48it/s]
1|1795|Loss: 0.7994554042816162:  28%|██▊       | 1794/6470 [23:27<52:29,  1.48it/s]
1|1795|Loss: 0.7994554042816162:  28%|██▊       | 1795/6470 [23:27<56:59,  1.37it/s]
1|1796|Loss: 1.0265369415283203:  28%|██▊       | 1795/6470 [23:28<56:59,  1.37it/s]
1|1796|Loss: 1.0265369415283203:  28%|██▊       | 1796/6470 [23:28<58:14,  1.34it/s]
1|1797|Loss: 0.9582359790802002:  28%|██▊       | 1796/6470 [23:28<58:14,  1.34it/s]
1|1797|Loss: 0.9582359790802002:  28%|██▊       | 1797/6470 [23:29<54:44,  1.42it/s]
1|1798|Loss: 1.093831181526184:  28%|██▊       | 1797/6470 [23:29<54:44,  1.42it/s] 
1|1798|Loss: 1.093831181526184:  28%|██▊       | 1798/6470 [23:29<52:54,  1.47it/s]
1|1799|Loss: 0.9079933762550354:  28%|██▊       | 1798/6470 [23:30<52:54,  1.47it/s]
1|1799|Loss: 0.9079933762550354:  28%|██▊       | 1799/6470 [23:30<56:02,  1.39it/s]
1|1800|Loss: 0.9040722250938416:  28%|██▊       | 1799/6470 [23:30<56:02,  1.39it/s]
1|1800|Loss: 0.9040722250938416:  28%|██▊       | 1800/6470 [23:31<54:49,  1.42it/s]
1|1801|Loss: 0.9382557272911072:  28%|██▊       | 1800/6470 [23:31<54:49,  1.42it/s]
1|1801|Loss: 0.9382557272911072:  28%|██▊       | 1801/6470 [23:31<50:36,  1.54it/s]
1|1802|Loss: 0.8907229900360107:  28%|██▊       | 1801/6470 [23:32<50:36,  1.54it/s]
1|1802|Loss: 0.8907229900360107:  28%|██▊       | 1802/6470 [23:32<49:19,  1.58it/s]
1|1803|Loss: 0.7860326170921326:  28%|██▊       | 1802/6470 [23:32<49:19,  1.58it/s]
1|1803|Loss: 0.7860326170921326:  28%|██▊       | 1803/6470 [23:33<53:40,  1.45it/s]
1|1804|Loss: 1.012499451637268:  28%|██▊       | 1803/6470 [23:33<53:40,  1.45it/s] 
1|1804|Loss: 1.012499451637268:  28%|██▊       | 1804/6470 [23:33<53:19,  1.46it/s]
1|1805|Loss: 0.9881854057312012:  28%|██▊       | 1804/6470 [23:34<53:19,  1.46it/s]
1|1805|Loss: 0.9881854057312012:  28%|██▊       | 1805/6470 [23:34<53:18,  1.46it/s]
1|1806|Loss: 0.914564311504364:  28%|██▊       | 1805/6470 [23:34<53:18,  1.46it/s] 
1|1806|Loss: 0.914564311504364:  28%|██▊       | 1806/6470 [23:35<54:12,  1.43it/s]
1|1807|Loss: 1.093202829360962:  28%|██▊       | 1806/6470 [23:35<54:12,  1.43it/s]
1|1807|Loss: 1.093202829360962:  28%|██▊       | 1807/6470 [23:35<51:09,  1.52it/s]
1|1808|Loss: 1.0154352188110352:  28%|██▊       | 1807/6470 [23:36<51:09,  1.52it/s]
1|1808|Loss: 1.0154352188110352:  28%|██▊       | 1808/6470 [23:36<49:43,  1.56it/s]
1|1809|Loss: 0.7508774995803833:  28%|██▊       | 1808/6470 [23:36<49:43,  1.56it/s]
1|1809|Loss: 0.7508774995803833:  28%|██▊       | 1809/6470 [23:37<53:05,  1.46it/s]
1|1810|Loss: 1.1839792728424072:  28%|██▊       | 1809/6470 [23:37<53:05,  1.46it/s]
1|1810|Loss: 1.1839792728424072:  28%|██▊       | 1810/6470 [23:37<51:50,  1.50it/s]
1|1811|Loss: 1.0361510515213013:  28%|██▊       | 1810/6470 [23:38<51:50,  1.50it/s]
1|1811|Loss: 1.0361510515213013:  28%|██▊       | 1811/6470 [23:38<48:51,  1.59it/s]
1|1812|Loss: 0.903917670249939:  28%|██▊       | 1811/6470 [23:38<48:51,  1.59it/s] 
1|1812|Loss: 0.903917670249939:  28%|██▊       | 1812/6470 [23:38<48:31,  1.60it/s]
1|1813|Loss: 0.8795831799507141:  28%|██▊       | 1812/6470 [23:39<48:31,  1.60it/s]
1|1813|Loss: 0.8795831799507141:  28%|██▊       | 1813/6470 [23:39<48:54,  1.59it/s]
1|1814|Loss: 1.0000150203704834:  28%|██▊       | 1813/6470 [23:39<48:54,  1.59it/s]
1|1814|Loss: 1.0000150203704834:  28%|██▊       | 1814/6470 [23:40<47:58,  1.62it/s]
1|1815|Loss: 0.792521059513092:  28%|██▊       | 1814/6470 [23:40<47:58,  1.62it/s] 
1|1815|Loss: 0.792521059513092:  28%|██▊       | 1815/6470 [23:40<47:21,  1.64it/s]
1|1816|Loss: 0.9247449040412903:  28%|██▊       | 1815/6470 [23:41<47:21,  1.64it/s]
1|1816|Loss: 0.9247449040412903:  28%|██▊       | 1816/6470 [23:41<49:05,  1.58it/s]
1|1817|Loss: 0.7916256189346313:  28%|██▊       | 1816/6470 [23:41<49:05,  1.58it/s]
1|1817|Loss: 0.7916256189346313:  28%|██▊       | 1817/6470 [23:42<47:21,  1.64it/s]
1|1818|Loss: 0.9321056008338928:  28%|██▊       | 1817/6470 [23:42<47:21,  1.64it/s]
1|1818|Loss: 0.9321056008338928:  28%|██▊       | 1818/6470 [23:42<45:19,  1.71it/s]
1|1819|Loss: 1.1642857789993286:  28%|██▊       | 1818/6470 [23:42<45:19,  1.71it/s]
1|1819|Loss: 1.1642857789993286:  28%|██▊       | 1819/6470 [23:43<43:13,  1.79it/s]
1|1820|Loss: 0.9749470949172974:  28%|██▊       | 1819/6470 [23:43<43:13,  1.79it/s]
1|1820|Loss: 0.9749470949172974:  28%|██▊       | 1820/6470 [23:43<46:06,  1.68it/s]
1|1821|Loss: 0.939240574836731:  28%|██▊       | 1820/6470 [23:44<46:06,  1.68it/s] 
1|1821|Loss: 0.939240574836731:  28%|██▊       | 1821/6470 [23:44<58:09,  1.33it/s]
1|1822|Loss: 1.0277947187423706:  28%|██▊       | 1821/6470 [23:45<58:09,  1.33it/s]
1|1822|Loss: 1.0277947187423706:  28%|██▊       | 1822/6470 [23:45<56:16,  1.38it/s]
1|1823|Loss: 0.9629814624786377:  28%|██▊       | 1822/6470 [23:45<56:16,  1.38it/s]
1|1823|Loss: 0.9629814624786377:  28%|██▊       | 1823/6470 [23:46<54:53,  1.41it/s]
1|1824|Loss: 0.9674509763717651:  28%|██▊       | 1823/6470 [23:46<54:53,  1.41it/s]
1|1824|Loss: 0.9674509763717651:  28%|██▊       | 1824/6470 [23:46<53:09,  1.46it/s]
1|1825|Loss: 0.89749675989151:  28%|██▊       | 1824/6470 [23:47<53:09,  1.46it/s]  
1|1825|Loss: 0.89749675989151:  28%|██▊       | 1825/6470 [23:47<51:56,  1.49it/s]
1|1826|Loss: 1.162987470626831:  28%|██▊       | 1825/6470 [23:47<51:56,  1.49it/s]
1|1826|Loss: 1.162987470626831:  28%|██▊       | 1826/6470 [23:48<49:15,  1.57it/s]
1|1827|Loss: 0.7833453416824341:  28%|██▊       | 1826/6470 [23:48<49:15,  1.57it/s]
1|1827|Loss: 0.7833453416824341:  28%|██▊       | 1827/6470 [23:48<47:10,  1.64it/s]
1|1828|Loss: 0.9407675862312317:  28%|██▊       | 1827/6470 [23:48<47:10,  1.64it/s]
1|1828|Loss: 0.9407675862312317:  28%|██▊       | 1828/6470 [23:49<48:40,  1.59it/s]
1|1829|Loss: 0.904170036315918:  28%|██▊       | 1828/6470 [23:49<48:40,  1.59it/s] 
1|1829|Loss: 0.904170036315918:  28%|██▊       | 1829/6470 [23:49<50:28,  1.53it/s]
1|1830|Loss: 0.790776252746582:  28%|██▊       | 1829/6470 [23:50<50:28,  1.53it/s]
1|1830|Loss: 0.790776252746582:  28%|██▊       | 1830/6470 [23:50<52:30,  1.47it/s]
1|1831|Loss: 1.043333649635315:  28%|██▊       | 1830/6470 [23:51<52:30,  1.47it/s]
1|1831|Loss: 1.043333649635315:  28%|██▊       | 1831/6470 [23:51<53:19,  1.45it/s]
1|1832|Loss: 0.8883190155029297:  28%|██▊       | 1831/6470 [23:51<53:19,  1.45it/s]
1|1832|Loss: 0.8883190155029297:  28%|██▊       | 1832/6470 [23:52<54:08,  1.43it/s]
1|1833|Loss: 0.9264736175537109:  28%|██▊       | 1832/6470 [23:52<54:08,  1.43it/s]
1|1833|Loss: 0.9264736175537109:  28%|██▊       | 1833/6470 [23:52<52:51,  1.46it/s]
1|1834|Loss: 0.8644790649414062:  28%|██▊       | 1833/6470 [23:53<52:51,  1.46it/s]
1|1834|Loss: 0.8644790649414062:  28%|██▊       | 1834/6470 [23:53<55:10,  1.40it/s]
1|1835|Loss: 0.9244778156280518:  28%|██▊       | 1834/6470 [23:53<55:10,  1.40it/s]
1|1835|Loss: 0.9244778156280518:  28%|██▊       | 1835/6470 [23:54<52:44,  1.46it/s]
1|1836|Loss: 1.0159204006195068:  28%|██▊       | 1835/6470 [23:54<52:44,  1.46it/s]
1|1836|Loss: 1.0159204006195068:  28%|██▊       | 1836/6470 [23:54<53:04,  1.46it/s]
1|1837|Loss: 0.9362011551856995:  28%|██▊       | 1836/6470 [23:55<53:04,  1.46it/s]
1|1837|Loss: 0.9362011551856995:  28%|██▊       | 1837/6470 [23:55<55:44,  1.39it/s]
1|1838|Loss: 1.038243055343628:  28%|██▊       | 1837/6470 [23:56<55:44,  1.39it/s] 
1|1838|Loss: 1.038243055343628:  28%|██▊       | 1838/6470 [23:56<54:21,  1.42it/s]
1|1839|Loss: 1.4505038261413574:  28%|██▊       | 1838/6470 [23:56<54:21,  1.42it/s]
1|1839|Loss: 1.4505038261413574:  28%|██▊       | 1839/6470 [23:57<55:42,  1.39it/s]
1|1840|Loss: 1.0262080430984497:  28%|██▊       | 1839/6470 [23:57<55:42,  1.39it/s]
1|1840|Loss: 1.0262080430984497:  28%|██▊       | 1840/6470 [23:57<57:37,  1.34it/s]
1|1841|Loss: 0.7478979825973511:  28%|██▊       | 1840/6470 [23:58<57:37,  1.34it/s]
1|1841|Loss: 0.7478979825973511:  28%|██▊       | 1841/6470 [23:58<1:00:46,  1.27it/s]
1|1842|Loss: 0.9099830985069275:  28%|██▊       | 1841/6470 [23:59<1:00:46,  1.27it/s]
1|1842|Loss: 0.9099830985069275:  28%|██▊       | 1842/6470 [23:59<1:02:36,  1.23it/s]
1|1843|Loss: 0.9502311944961548:  28%|██▊       | 1842/6470 [24:00<1:02:36,  1.23it/s]
1|1843|Loss: 0.9502311944961548:  28%|██▊       | 1843/6470 [24:00<1:01:52,  1.25it/s]
1|1844|Loss: 0.9387887716293335:  28%|██▊       | 1843/6470 [24:00<1:01:52,  1.25it/s]
1|1844|Loss: 0.9387887716293335:  29%|██▊       | 1844/6470 [24:01<58:56,  1.31it/s]  
1|1845|Loss: 1.077147364616394:  29%|██▊       | 1844/6470 [24:01<58:56,  1.31it/s] 
1|1845|Loss: 1.077147364616394:  29%|██▊       | 1845/6470 [24:01<59:31,  1.30it/s]
1|1846|Loss: 1.0287837982177734:  29%|██▊       | 1845/6470 [24:02<59:31,  1.30it/s]
1|1846|Loss: 1.0287837982177734:  29%|██▊       | 1846/6470 [24:02<56:51,  1.36it/s]
1|1847|Loss: 1.0772215127944946:  29%|██▊       | 1846/6470 [24:02<56:51,  1.36it/s]
1|1847|Loss: 1.0772215127944946:  29%|██▊       | 1847/6470 [24:03<54:24,  1.42it/s]
1|1848|Loss: 0.9426569938659668:  29%|██▊       | 1847/6470 [24:03<54:24,  1.42it/s]
1|1848|Loss: 0.9426569938659668:  29%|██▊       | 1848/6470 [24:03<54:53,  1.40it/s]
1|1849|Loss: 0.949654221534729:  29%|██▊       | 1848/6470 [24:04<54:53,  1.40it/s] 
1|1849|Loss: 0.949654221534729:  29%|██▊       | 1849/6470 [24:04<55:36,  1.38it/s]
1|1850|Loss: 0.8581562638282776:  29%|██▊       | 1849/6470 [24:05<55:36,  1.38it/s]
1|1850|Loss: 0.8581562638282776:  29%|██▊       | 1850/6470 [24:05<53:14,  1.45it/s]
1|1851|Loss: 0.8945392966270447:  29%|██▊       | 1850/6470 [24:05<53:14,  1.45it/s]
1|1851|Loss: 0.8945392966270447:  29%|██▊       | 1851/6470 [24:05<48:23,  1.59it/s]
1|1852|Loss: 0.9900438785552979:  29%|██▊       | 1851/6470 [24:06<48:23,  1.59it/s]
1|1852|Loss: 0.9900438785552979:  29%|██▊       | 1852/6470 [24:06<49:38,  1.55it/s]
1|1853|Loss: 0.8976262211799622:  29%|██▊       | 1852/6470 [24:06<49:38,  1.55it/s]
1|1853|Loss: 0.8976262211799622:  29%|██▊       | 1853/6470 [24:07<49:04,  1.57it/s]
1|1854|Loss: 1.0814990997314453:  29%|██▊       | 1853/6470 [24:07<49:04,  1.57it/s]
1|1854|Loss: 1.0814990997314453:  29%|██▊       | 1854/6470 [24:07<48:50,  1.57it/s]
1|1855|Loss: 0.9545498490333557:  29%|██▊       | 1854/6470 [24:08<48:50,  1.57it/s]
1|1855|Loss: 0.9545498490333557:  29%|██▊       | 1855/6470 [24:08<47:24,  1.62it/s]
1|1856|Loss: 0.8253498673439026:  29%|██▊       | 1855/6470 [24:08<47:24,  1.62it/s]
1|1856|Loss: 0.8253498673439026:  29%|██▊       | 1856/6470 [24:08<48:45,  1.58it/s]
1|1857|Loss: 0.9915469288825989:  29%|██▊       | 1856/6470 [24:09<48:45,  1.58it/s]
1|1857|Loss: 0.9915469288825989:  29%|██▊       | 1857/6470 [24:09<47:08,  1.63it/s]
1|1858|Loss: 1.0100253820419312:  29%|██▊       | 1857/6470 [24:09<47:08,  1.63it/s]
1|1858|Loss: 1.0100253820419312:  29%|██▊       | 1858/6470 [24:10<49:13,  1.56it/s]
1|1859|Loss: 0.8472564816474915:  29%|██▊       | 1858/6470 [24:10<49:13,  1.56it/s]
1|1859|Loss: 0.8472564816474915:  29%|██▊       | 1859/6470 [24:10<49:55,  1.54it/s]
1|1860|Loss: 0.9523147344589233:  29%|██▊       | 1859/6470 [24:11<49:55,  1.54it/s]
1|1860|Loss: 0.9523147344589233:  29%|██▊       | 1860/6470 [24:11<46:50,  1.64it/s]
1|1861|Loss: 0.9415494203567505:  29%|██▊       | 1860/6470 [24:11<46:50,  1.64it/s]
1|1861|Loss: 0.9415494203567505:  29%|██▉       | 1861/6470 [24:12<48:51,  1.57it/s]
1|1862|Loss: 0.9589942097663879:  29%|██▉       | 1861/6470 [24:12<48:51,  1.57it/s]
1|1862|Loss: 0.9589942097663879:  29%|██▉       | 1862/6470 [24:12<53:02,  1.45it/s]
1|1863|Loss: 0.9179354310035706:  29%|██▉       | 1862/6470 [24:13<53:02,  1.45it/s]
1|1863|Loss: 0.9179354310035706:  29%|██▉       | 1863/6470 [24:13<53:54,  1.42it/s]
1|1864|Loss: 0.8877856135368347:  29%|██▉       | 1863/6470 [24:14<53:54,  1.42it/s]
1|1864|Loss: 0.8877856135368347:  29%|██▉       | 1864/6470 [24:14<52:42,  1.46it/s]
1|1865|Loss: 0.9742415547370911:  29%|██▉       | 1864/6470 [24:14<52:42,  1.46it/s]
1|1865|Loss: 0.9742415547370911:  29%|██▉       | 1865/6470 [24:15<52:48,  1.45it/s]
1|1866|Loss: 0.85509192943573:  29%|██▉       | 1865/6470 [24:15<52:48,  1.45it/s]  
1|1866|Loss: 0.85509192943573:  29%|██▉       | 1866/6470 [24:15<50:41,  1.51it/s]
1|1867|Loss: 1.009564995765686:  29%|██▉       | 1866/6470 [24:15<50:41,  1.51it/s]
1|1867|Loss: 1.009564995765686:  29%|██▉       | 1867/6470 [24:16<51:08,  1.50it/s]
1|1868|Loss: 0.851302444934845:  29%|██▉       | 1867/6470 [24:16<51:08,  1.50it/s]
1|1868|Loss: 0.851302444934845:  29%|██▉       | 1868/6470 [24:17<54:54,  1.40it/s]
1|1869|Loss: 1.0480334758758545:  29%|██▉       | 1868/6470 [24:17<54:54,  1.40it/s]
1|1869|Loss: 1.0480334758758545:  29%|██▉       | 1869/6470 [24:17<57:08,  1.34it/s]
1|1870|Loss: 0.8946181535720825:  29%|██▉       | 1869/6470 [24:18<57:08,  1.34it/s]
1|1870|Loss: 0.8946181535720825:  29%|██▉       | 1870/6470 [24:18<57:20,  1.34it/s]
1|1871|Loss: 0.9136441349983215:  29%|██▉       | 1870/6470 [24:19<57:20,  1.34it/s]
1|1871|Loss: 0.9136441349983215:  29%|██▉       | 1871/6470 [24:19<56:38,  1.35it/s]
1|1872|Loss: 0.8404210805892944:  29%|██▉       | 1871/6470 [24:19<56:38,  1.35it/s]
1|1872|Loss: 0.8404210805892944:  29%|██▉       | 1872/6470 [24:20<55:50,  1.37it/s]
1|1873|Loss: 0.9130313396453857:  29%|██▉       | 1872/6470 [24:20<55:50,  1.37it/s]
1|1873|Loss: 0.9130313396453857:  29%|██▉       | 1873/6470 [24:20<55:12,  1.39it/s]
1|1874|Loss: 0.9281908869743347:  29%|██▉       | 1873/6470 [24:21<55:12,  1.39it/s]
1|1874|Loss: 0.9281908869743347:  29%|██▉       | 1874/6470 [24:21<56:58,  1.34it/s]
1|1875|Loss: 1.1266491413116455:  29%|██▉       | 1874/6470 [24:21<56:58,  1.34it/s]
1|1875|Loss: 1.1266491413116455:  29%|██▉       | 1875/6470 [24:22<54:34,  1.40it/s]
1|1876|Loss: 0.9137102365493774:  29%|██▉       | 1875/6470 [24:22<54:34,  1.40it/s]
1|1876|Loss: 0.9137102365493774:  29%|██▉       | 1876/6470 [24:22<52:39,  1.45it/s]
1|1877|Loss: 0.877526581287384:  29%|██▉       | 1876/6470 [24:23<52:39,  1.45it/s] 
1|1877|Loss: 0.877526581287384:  29%|██▉       | 1877/6470 [24:23<54:57,  1.39it/s]
1|1878|Loss: 1.034855604171753:  29%|██▉       | 1877/6470 [24:24<54:57,  1.39it/s]
1|1878|Loss: 1.034855604171753:  29%|██▉       | 1878/6470 [24:24<57:57,  1.32it/s]
1|1879|Loss: 1.0205515623092651:  29%|██▉       | 1878/6470 [24:24<57:57,  1.32it/s]
1|1879|Loss: 1.0205515623092651:  29%|██▉       | 1879/6470 [24:25<58:11,  1.32it/s]
1|1880|Loss: 1.009887456893921:  29%|██▉       | 1879/6470 [24:25<58:11,  1.32it/s] 
1|1880|Loss: 1.009887456893921:  29%|██▉       | 1880/6470 [24:25<54:51,  1.39it/s]
1|1881|Loss: 0.8336966633796692:  29%|██▉       | 1880/6470 [24:26<54:51,  1.39it/s]
1|1881|Loss: 0.8336966633796692:  29%|██▉       | 1881/6470 [24:26<54:26,  1.40it/s]
1|1882|Loss: 0.9582032561302185:  29%|██▉       | 1881/6470 [24:27<54:26,  1.40it/s]
1|1882|Loss: 0.9582032561302185:  29%|██▉       | 1882/6470 [24:27<54:59,  1.39it/s]
1|1883|Loss: 0.9600633978843689:  29%|██▉       | 1882/6470 [24:27<54:59,  1.39it/s]
1|1883|Loss: 0.9600633978843689:  29%|██▉       | 1883/6470 [24:28<54:46,  1.40it/s]
1|1884|Loss: 1.0722333192825317:  29%|██▉       | 1883/6470 [24:28<54:46,  1.40it/s]
1|1884|Loss: 1.0722333192825317:  29%|██▉       | 1884/6470 [24:28<57:06,  1.34it/s]
1|1885|Loss: 0.9193624258041382:  29%|██▉       | 1884/6470 [24:29<57:06,  1.34it/s]
1|1885|Loss: 0.9193624258041382:  29%|██▉       | 1885/6470 [24:29<55:48,  1.37it/s]
1|1886|Loss: 0.951596736907959:  29%|██▉       | 1885/6470 [24:29<55:48,  1.37it/s] 
1|1886|Loss: 0.951596736907959:  29%|██▉       | 1886/6470 [24:30<55:48,  1.37it/s]
1|1887|Loss: 0.8008196353912354:  29%|██▉       | 1886/6470 [24:30<55:48,  1.37it/s]
1|1887|Loss: 0.8008196353912354:  29%|██▉       | 1887/6470 [24:31<56:11,  1.36it/s]
1|1888|Loss: 0.8915743827819824:  29%|██▉       | 1887/6470 [24:31<56:11,  1.36it/s]
1|1888|Loss: 0.8915743827819824:  29%|██▉       | 1888/6470 [24:31<58:25,  1.31it/s]
1|1889|Loss: 0.921349048614502:  29%|██▉       | 1888/6470 [24:32<58:25,  1.31it/s] 
1|1889|Loss: 0.921349048614502:  29%|██▉       | 1889/6470 [24:32<56:21,  1.35it/s]
1|1890|Loss: 0.8785413503646851:  29%|██▉       | 1889/6470 [24:32<56:21,  1.35it/s]
1|1890|Loss: 0.8785413503646851:  29%|██▉       | 1890/6470 [24:33<57:50,  1.32it/s]
1|1891|Loss: 0.8164892792701721:  29%|██▉       | 1890/6470 [24:33<57:50,  1.32it/s]
1|1891|Loss: 0.8164892792701721:  29%|██▉       | 1891/6470 [24:34<58:13,  1.31it/s]
1|1892|Loss: 0.8928046822547913:  29%|██▉       | 1891/6470 [24:34<58:13,  1.31it/s]
1|1892|Loss: 0.8928046822547913:  29%|██▉       | 1892/6470 [24:34<58:04,  1.31it/s]
1|1893|Loss: 0.9311449527740479:  29%|██▉       | 1892/6470 [24:35<58:04,  1.31it/s]
1|1893|Loss: 0.9311449527740479:  29%|██▉       | 1893/6470 [24:35<57:44,  1.32it/s]
1|1894|Loss: 0.8782288432121277:  29%|██▉       | 1893/6470 [24:36<57:44,  1.32it/s]
1|1894|Loss: 0.8782288432121277:  29%|██▉       | 1894/6470 [24:36<55:41,  1.37it/s]
1|1895|Loss: 0.9195489287376404:  29%|██▉       | 1894/6470 [24:36<55:41,  1.37it/s]
1|1895|Loss: 0.9195489287376404:  29%|██▉       | 1895/6470 [24:37<55:16,  1.38it/s]
1|1896|Loss: 0.9429606795310974:  29%|██▉       | 1895/6470 [24:37<55:16,  1.38it/s]
1|1896|Loss: 0.9429606795310974:  29%|██▉       | 1896/6470 [24:37<54:48,  1.39it/s]
1|1897|Loss: 0.9625749588012695:  29%|██▉       | 1896/6470 [24:38<54:48,  1.39it/s]
1|1897|Loss: 0.9625749588012695:  29%|██▉       | 1897/6470 [24:38<54:22,  1.40it/s]
1|1898|Loss: 0.8576498627662659:  29%|██▉       | 1897/6470 [24:38<54:22,  1.40it/s]
1|1898|Loss: 0.8576498627662659:  29%|██▉       | 1898/6470 [24:39<56:53,  1.34it/s]
1|1899|Loss: 1.109336018562317:  29%|██▉       | 1898/6470 [24:39<56:53,  1.34it/s] 
1|1899|Loss: 1.109336018562317:  29%|██▉       | 1899/6470 [24:39<56:33,  1.35it/s]
1|1900|Loss: 0.886821985244751:  29%|██▉       | 1899/6470 [24:40<56:33,  1.35it/s]
1|1900|Loss: 0.886821985244751:  29%|██▉       | 1900/6470 [24:40<1:00:15,  1.26it/s]
1|1901|Loss: 0.9505117535591125:  29%|██▉       | 1900/6470 [24:41<1:00:15,  1.26it/s]
1|1901|Loss: 0.9505117535591125:  29%|██▉       | 1901/6470 [24:41<58:18,  1.31it/s]  
1|1902|Loss: 0.9223541617393494:  29%|██▉       | 1901/6470 [24:41<58:18,  1.31it/s]
1|1902|Loss: 0.9223541617393494:  29%|██▉       | 1902/6470 [24:42<56:34,  1.35it/s]
1|1903|Loss: 0.9540231823921204:  29%|██▉       | 1902/6470 [24:42<56:34,  1.35it/s]
1|1903|Loss: 0.9540231823921204:  29%|██▉       | 1903/6470 [24:42<55:32,  1.37it/s]
1|1904|Loss: 1.019766092300415:  29%|██▉       | 1903/6470 [24:43<55:32,  1.37it/s] 
1|1904|Loss: 1.019766092300415:  29%|██▉       | 1904/6470 [24:43<55:49,  1.36it/s]
1|1905|Loss: 0.9602152705192566:  29%|██▉       | 1904/6470 [24:44<55:49,  1.36it/s]
1|1905|Loss: 0.9602152705192566:  29%|██▉       | 1905/6470 [24:44<56:55,  1.34it/s]
1|1906|Loss: 1.0042487382888794:  29%|██▉       | 1905/6470 [24:44<56:55,  1.34it/s]
1|1906|Loss: 1.0042487382888794:  29%|██▉       | 1906/6470 [24:45<56:58,  1.33it/s]
1|1907|Loss: 0.8732125759124756:  29%|██▉       | 1906/6470 [24:45<56:58,  1.33it/s]
1|1907|Loss: 0.8732125759124756:  29%|██▉       | 1907/6470 [24:46<57:55,  1.31it/s]
1|1908|Loss: 1.0880844593048096:  29%|██▉       | 1907/6470 [24:46<57:55,  1.31it/s]
1|1908|Loss: 1.0880844593048096:  29%|██▉       | 1908/6470 [24:46<58:43,  1.29it/s]
1|1909|Loss: 0.8765236735343933:  29%|██▉       | 1908/6470 [24:47<58:43,  1.29it/s]
1|1909|Loss: 0.8765236735343933:  30%|██▉       | 1909/6470 [24:47<54:49,  1.39it/s]
1|1910|Loss: 0.8942426443099976:  30%|██▉       | 1909/6470 [24:47<54:49,  1.39it/s]
1|1910|Loss: 0.8942426443099976:  30%|██▉       | 1910/6470 [24:48<54:07,  1.40it/s]
1|1911|Loss: 0.9798471927642822:  30%|██▉       | 1910/6470 [24:48<54:07,  1.40it/s]
1|1911|Loss: 0.9798471927642822:  30%|██▉       | 1911/6470 [24:48<49:32,  1.53it/s]
1|1912|Loss: 0.8253383636474609:  30%|██▉       | 1911/6470 [24:48<49:32,  1.53it/s]
1|1912|Loss: 0.8253383636474609:  30%|██▉       | 1912/6470 [24:49<49:28,  1.54it/s]
1|1913|Loss: 0.8222929835319519:  30%|██▉       | 1912/6470 [24:49<49:28,  1.54it/s]
1|1913|Loss: 0.8222929835319519:  30%|██▉       | 1913/6470 [24:49<48:21,  1.57it/s]
1|1914|Loss: 1.0251351594924927:  30%|██▉       | 1913/6470 [24:50<48:21,  1.57it/s]
1|1914|Loss: 1.0251351594924927:  30%|██▉       | 1914/6470 [24:50<49:56,  1.52it/s]
1|1915|Loss: 1.0156499147415161:  30%|██▉       | 1914/6470 [24:50<49:56,  1.52it/s]
1|1915|Loss: 1.0156499147415161:  30%|██▉       | 1915/6470 [24:51<50:01,  1.52it/s]
1|1916|Loss: 0.930706799030304:  30%|██▉       | 1915/6470 [24:51<50:01,  1.52it/s] 
1|1916|Loss: 0.930706799030304:  30%|██▉       | 1916/6470 [24:52<53:30,  1.42it/s]
1|1917|Loss: 0.8844308257102966:  30%|██▉       | 1916/6470 [24:52<53:30,  1.42it/s]
1|1917|Loss: 0.8844308257102966:  30%|██▉       | 1917/6470 [24:52<56:07,  1.35it/s]
1|1918|Loss: 0.9532299041748047:  30%|██▉       | 1917/6470 [24:53<56:07,  1.35it/s]
1|1918|Loss: 0.9532299041748047:  30%|██▉       | 1918/6470 [24:53<54:24,  1.39it/s]
1|1919|Loss: 0.999248206615448:  30%|██▉       | 1918/6470 [24:53<54:24,  1.39it/s] 
1|1919|Loss: 0.999248206615448:  30%|██▉       | 1919/6470 [24:54<53:51,  1.41it/s]
1|1920|Loss: 1.0865120887756348:  30%|██▉       | 1919/6470 [24:54<53:51,  1.41it/s]
1|1920|Loss: 1.0865120887756348:  30%|██▉       | 1920/6470 [24:54<49:10,  1.54it/s]
1|1921|Loss: 0.8748346567153931:  30%|██▉       | 1920/6470 [24:55<49:10,  1.54it/s]
1|1921|Loss: 0.8748346567153931:  30%|██▉       | 1921/6470 [24:55<49:58,  1.52it/s]
1|1922|Loss: 1.0271368026733398:  30%|██▉       | 1921/6470 [24:55<49:58,  1.52it/s]
1|1922|Loss: 1.0271368026733398:  30%|██▉       | 1922/6470 [24:56<53:28,  1.42it/s]
1|1923|Loss: 0.9774577021598816:  30%|██▉       | 1922/6470 [24:56<53:28,  1.42it/s]
1|1923|Loss: 0.9774577021598816:  30%|██▉       | 1923/6470 [24:57<56:28,  1.34it/s]
1|1924|Loss: 0.8422877192497253:  30%|██▉       | 1923/6470 [24:57<56:28,  1.34it/s]
1|1924|Loss: 0.8422877192497253:  30%|██▉       | 1924/6470 [24:57<54:40,  1.39it/s]
1|1925|Loss: 0.7060421109199524:  30%|██▉       | 1924/6470 [24:58<54:40,  1.39it/s]
1|1925|Loss: 0.7060421109199524:  30%|██▉       | 1925/6470 [24:58<50:36,  1.50it/s]
1|1926|Loss: 0.9660942554473877:  30%|██▉       | 1925/6470 [24:58<50:36,  1.50it/s]
1|1926|Loss: 0.9660942554473877:  30%|██▉       | 1926/6470 [24:59<51:22,  1.47it/s]
1|1927|Loss: 1.0679787397384644:  30%|██▉       | 1926/6470 [24:59<51:22,  1.47it/s]
1|1927|Loss: 1.0679787397384644:  30%|██▉       | 1927/6470 [24:59<50:44,  1.49it/s]
1|1928|Loss: 1.164987325668335:  30%|██▉       | 1927/6470 [25:00<50:44,  1.49it/s] 
1|1928|Loss: 1.164987325668335:  30%|██▉       | 1928/6470 [25:00<53:15,  1.42it/s]
1|1929|Loss: 0.7718480825424194:  30%|██▉       | 1928/6470 [25:00<53:15,  1.42it/s]
1|1929|Loss: 0.7718480825424194:  30%|██▉       | 1929/6470 [25:01<53:51,  1.41it/s]
1|1930|Loss: 0.9109219908714294:  30%|██▉       | 1929/6470 [25:01<53:51,  1.41it/s]
1|1930|Loss: 0.9109219908714294:  30%|██▉       | 1930/6470 [25:01<52:09,  1.45it/s]
1|1931|Loss: 0.9943749904632568:  30%|██▉       | 1930/6470 [25:02<52:09,  1.45it/s]
1|1931|Loss: 0.9943749904632568:  30%|██▉       | 1931/6470 [25:02<49:46,  1.52it/s]
1|1932|Loss: 1.0980587005615234:  30%|██▉       | 1931/6470 [25:02<49:46,  1.52it/s]
1|1932|Loss: 1.0980587005615234:  30%|██▉       | 1932/6470 [25:03<49:05,  1.54it/s]
1|1933|Loss: 0.9711645245552063:  30%|██▉       | 1932/6470 [25:03<49:05,  1.54it/s]
1|1933|Loss: 0.9711645245552063:  30%|██▉       | 1933/6470 [25:03<48:35,  1.56it/s]
1|1934|Loss: 0.9868720173835754:  30%|██▉       | 1933/6470 [25:03<48:35,  1.56it/s]
1|1934|Loss: 0.9868720173835754:  30%|██▉       | 1934/6470 [25:04<46:45,  1.62it/s]
1|1935|Loss: 0.9514407515525818:  30%|██▉       | 1934/6470 [25:04<46:45,  1.62it/s]
1|1935|Loss: 0.9514407515525818:  30%|██▉       | 1935/6470 [25:04<49:53,  1.51it/s]
1|1936|Loss: 0.9883326292037964:  30%|██▉       | 1935/6470 [25:05<49:53,  1.51it/s]
1|1936|Loss: 0.9883326292037964:  30%|██▉       | 1936/6470 [25:05<52:21,  1.44it/s]
1|1937|Loss: 1.0669167041778564:  30%|██▉       | 1936/6470 [25:06<52:21,  1.44it/s]
1|1937|Loss: 1.0669167041778564:  30%|██▉       | 1937/6470 [25:06<51:33,  1.47it/s]
1|1938|Loss: 1.0886369943618774:  30%|██▉       | 1937/6470 [25:06<51:33,  1.47it/s]
1|1938|Loss: 1.0886369943618774:  30%|██▉       | 1938/6470 [25:07<51:36,  1.46it/s]
1|1939|Loss: 0.7330355644226074:  30%|██▉       | 1938/6470 [25:07<51:36,  1.46it/s]
1|1939|Loss: 0.7330355644226074:  30%|██▉       | 1939/6470 [25:07<53:49,  1.40it/s]
1|1940|Loss: 0.9086588621139526:  30%|██▉       | 1939/6470 [25:08<53:49,  1.40it/s]
1|1940|Loss: 0.9086588621139526:  30%|██▉       | 1940/6470 [25:08<52:38,  1.43it/s]
1|1941|Loss: 0.9817895293235779:  30%|██▉       | 1940/6470 [25:08<52:38,  1.43it/s]
1|1941|Loss: 0.9817895293235779:  30%|███       | 1941/6470 [25:09<55:25,  1.36it/s]
1|1942|Loss: 0.9004817605018616:  30%|███       | 1941/6470 [25:09<55:25,  1.36it/s]
1|1942|Loss: 0.9004817605018616:  30%|███       | 1942/6470 [25:10<57:10,  1.32it/s]
1|1943|Loss: 1.0250911712646484:  30%|███       | 1942/6470 [25:10<57:10,  1.32it/s]
1|1943|Loss: 1.0250911712646484:  30%|███       | 1943/6470 [25:10<53:01,  1.42it/s]
1|1944|Loss: 0.984062135219574:  30%|███       | 1943/6470 [25:11<53:01,  1.42it/s] 
1|1944|Loss: 0.984062135219574:  30%|███       | 1944/6470 [25:11<54:26,  1.39it/s]
1|1945|Loss: 0.8961085081100464:  30%|███       | 1944/6470 [25:11<54:26,  1.39it/s]
1|1945|Loss: 0.8961085081100464:  30%|███       | 1945/6470 [25:12<57:15,  1.32it/s]
1|1946|Loss: 0.9250594973564148:  30%|███       | 1945/6470 [25:12<57:15,  1.32it/s]
1|1946|Loss: 0.9250594973564148:  30%|███       | 1946/6470 [25:13<56:56,  1.32it/s]
1|1947|Loss: 0.8346607685089111:  30%|███       | 1946/6470 [25:13<56:56,  1.32it/s]
1|1947|Loss: 0.8346607685089111:  30%|███       | 1947/6470 [25:13<55:42,  1.35it/s]
1|1948|Loss: 0.9871773719787598:  30%|███       | 1947/6470 [25:14<55:42,  1.35it/s]
1|1948|Loss: 0.9871773719787598:  30%|███       | 1948/6470 [25:14<54:55,  1.37it/s]
1|1949|Loss: 0.9862026572227478:  30%|███       | 1948/6470 [25:14<54:55,  1.37it/s]
1|1949|Loss: 0.9862026572227478:  30%|███       | 1949/6470 [25:15<54:47,  1.38it/s]
1|1950|Loss: 0.8129074573516846:  30%|███       | 1949/6470 [25:15<54:47,  1.38it/s]
1|1950|Loss: 0.8129074573516846:  30%|███       | 1950/6470 [25:15<55:47,  1.35it/s]
1|1951|Loss: 1.0441066026687622:  30%|███       | 1950/6470 [25:16<55:47,  1.35it/s]
1|1951|Loss: 1.0441066026687622:  30%|███       | 1951/6470 [25:16<52:04,  1.45it/s]
1|1952|Loss: 0.8969953656196594:  30%|███       | 1951/6470 [25:16<52:04,  1.45it/s]
1|1952|Loss: 0.8969953656196594:  30%|███       | 1952/6470 [25:17<54:07,  1.39it/s]
1|1953|Loss: 0.9841578602790833:  30%|███       | 1952/6470 [25:17<54:07,  1.39it/s]
1|1953|Loss: 0.9841578602790833:  30%|███       | 1953/6470 [25:17<50:36,  1.49it/s]
1|1954|Loss: 1.0472866296768188:  30%|███       | 1953/6470 [25:18<50:36,  1.49it/s]
1|1954|Loss: 1.0472866296768188:  30%|███       | 1954/6470 [25:18<48:00,  1.57it/s]
1|1955|Loss: 0.9723093509674072:  30%|███       | 1954/6470 [25:18<48:00,  1.57it/s]
1|1955|Loss: 0.9723093509674072:  30%|███       | 1955/6470 [25:19<46:12,  1.63it/s]
1|1956|Loss: 1.020061731338501:  30%|███       | 1955/6470 [25:19<46:12,  1.63it/s] 
1|1956|Loss: 1.020061731338501:  30%|███       | 1956/6470 [25:19<44:00,  1.71it/s]
1|1957|Loss: 0.8729904890060425:  30%|███       | 1956/6470 [25:19<44:00,  1.71it/s]
1|1957|Loss: 0.8729904890060425:  30%|███       | 1957/6470 [25:20<42:59,  1.75it/s]
1|1958|Loss: 0.9147974252700806:  30%|███       | 1957/6470 [25:20<42:59,  1.75it/s]
1|1958|Loss: 0.9147974252700806:  30%|███       | 1958/6470 [25:20<45:44,  1.64it/s]
1|1959|Loss: 0.8571130633354187:  30%|███       | 1958/6470 [25:21<45:44,  1.64it/s]
1|1959|Loss: 0.8571130633354187:  30%|███       | 1959/6470 [25:21<46:52,  1.60it/s]
1|1960|Loss: 0.9874048829078674:  30%|███       | 1959/6470 [25:21<46:52,  1.60it/s]
1|1960|Loss: 0.9874048829078674:  30%|███       | 1960/6470 [25:22<50:37,  1.48it/s]
1|1961|Loss: 0.8686957955360413:  30%|███       | 1960/6470 [25:22<50:37,  1.48it/s]
1|1961|Loss: 0.8686957955360413:  30%|███       | 1961/6470 [25:22<52:43,  1.43it/s]
1|1962|Loss: 0.8075727820396423:  30%|███       | 1961/6470 [25:23<52:43,  1.43it/s]
1|1962|Loss: 0.8075727820396423:  30%|███       | 1962/6470 [25:23<50:55,  1.48it/s]
1|1963|Loss: 1.032827615737915:  30%|███       | 1962/6470 [25:23<50:55,  1.48it/s] 
1|1963|Loss: 1.032827615737915:  30%|███       | 1963/6470 [25:24<48:36,  1.55it/s]
1|1964|Loss: 1.0475366115570068:  30%|███       | 1963/6470 [25:24<48:36,  1.55it/s]
1|1964|Loss: 1.0475366115570068:  30%|███       | 1964/6470 [25:24<48:09,  1.56it/s]
1|1965|Loss: 1.0357410907745361:  30%|███       | 1964/6470 [25:25<48:09,  1.56it/s]
1|1965|Loss: 1.0357410907745361:  30%|███       | 1965/6470 [25:25<49:50,  1.51it/s]
1|1966|Loss: 0.8237655758857727:  30%|███       | 1965/6470 [25:25<49:50,  1.51it/s]
1|1966|Loss: 0.8237655758857727:  30%|███       | 1966/6470 [25:26<51:35,  1.45it/s]
1|1967|Loss: 0.809187650680542:  30%|███       | 1966/6470 [25:26<51:35,  1.45it/s] 
1|1967|Loss: 0.809187650680542:  30%|███       | 1967/6470 [25:26<51:07,  1.47it/s]
1|1968|Loss: 0.9555515646934509:  30%|███       | 1967/6470 [25:27<51:07,  1.47it/s]
1|1968|Loss: 0.9555515646934509:  30%|███       | 1968/6470 [25:27<51:18,  1.46it/s]
1|1969|Loss: 0.815136730670929:  30%|███       | 1968/6470 [25:28<51:18,  1.46it/s] 
1|1969|Loss: 0.815136730670929:  30%|███       | 1969/6470 [25:28<55:24,  1.35it/s]
1|1970|Loss: 1.0392496585845947:  30%|███       | 1969/6470 [25:28<55:24,  1.35it/s]
1|1970|Loss: 1.0392496585845947:  30%|███       | 1970/6470 [25:29<56:32,  1.33it/s]
1|1971|Loss: 1.016843318939209:  30%|███       | 1970/6470 [25:29<56:32,  1.33it/s] 
1|1971|Loss: 1.016843318939209:  30%|███       | 1971/6470 [25:30<56:05,  1.34it/s]
1|1972|Loss: 1.1138323545455933:  30%|███       | 1971/6470 [25:30<56:05,  1.34it/s]
1|1972|Loss: 1.1138323545455933:  30%|███       | 1972/6470 [25:30<55:46,  1.34it/s]
1|1973|Loss: 0.8675450682640076:  30%|███       | 1972/6470 [25:31<55:46,  1.34it/s]
1|1973|Loss: 0.8675450682640076:  30%|███       | 1973/6470 [25:31<53:26,  1.40it/s]
1|1974|Loss: 0.9853854179382324:  30%|███       | 1973/6470 [25:31<53:26,  1.40it/s]
1|1974|Loss: 0.9853854179382324:  31%|███       | 1974/6470 [25:32<53:42,  1.40it/s]
1|1975|Loss: 0.979555606842041:  31%|███       | 1974/6470 [25:32<53:42,  1.40it/s] 
1|1975|Loss: 0.979555606842041:  31%|███       | 1975/6470 [25:32<55:52,  1.34it/s]
1|1976|Loss: 1.0657305717468262:  31%|███       | 1975/6470 [25:33<55:52,  1.34it/s]
1|1976|Loss: 1.0657305717468262:  31%|███       | 1976/6470 [25:33<55:35,  1.35it/s]
1|1977|Loss: 1.0287679433822632:  31%|███       | 1976/6470 [25:34<55:35,  1.35it/s]
1|1977|Loss: 1.0287679433822632:  31%|███       | 1977/6470 [25:34<51:12,  1.46it/s]
1|1978|Loss: 0.9056035280227661:  31%|███       | 1977/6470 [25:34<51:12,  1.46it/s]
1|1978|Loss: 0.9056035280227661:  31%|███       | 1978/6470 [25:34<51:26,  1.46it/s]
1|1979|Loss: 0.9364812970161438:  31%|███       | 1978/6470 [25:35<51:26,  1.46it/s]
1|1979|Loss: 0.9364812970161438:  31%|███       | 1979/6470 [25:35<50:55,  1.47it/s]
1|1980|Loss: 0.7586348056793213:  31%|███       | 1979/6470 [25:35<50:55,  1.47it/s]
1|1980|Loss: 0.7586348056793213:  31%|███       | 1980/6470 [25:36<49:00,  1.53it/s]
1|1981|Loss: 0.9481635093688965:  31%|███       | 1980/6470 [25:36<49:00,  1.53it/s]
1|1981|Loss: 0.9481635093688965:  31%|███       | 1981/6470 [25:36<52:48,  1.42it/s]
1|1982|Loss: 0.8303936719894409:  31%|███       | 1981/6470 [25:37<52:48,  1.42it/s]
1|1982|Loss: 0.8303936719894409:  31%|███       | 1982/6470 [25:37<57:26,  1.30it/s]
1|1983|Loss: 0.9575793743133545:  31%|███       | 1982/6470 [25:38<57:26,  1.30it/s]
1|1983|Loss: 0.9575793743133545:  31%|███       | 1983/6470 [25:38<57:36,  1.30it/s]
1|1984|Loss: 0.8518202304840088:  31%|███       | 1983/6470 [25:39<57:36,  1.30it/s]
1|1984|Loss: 0.8518202304840088:  31%|███       | 1984/6470 [25:39<58:11,  1.28it/s]
1|1985|Loss: 0.8944377899169922:  31%|███       | 1984/6470 [25:39<58:11,  1.28it/s]
1|1985|Loss: 0.8944377899169922:  31%|███       | 1985/6470 [25:40<58:47,  1.27it/s]
1|1986|Loss: 0.9183884859085083:  31%|███       | 1985/6470 [25:40<58:47,  1.27it/s]
1|1986|Loss: 0.9183884859085083:  31%|███       | 1986/6470 [25:40<53:55,  1.39it/s]
1|1987|Loss: 0.9356866478919983:  31%|███       | 1986/6470 [25:41<53:55,  1.39it/s]
1|1987|Loss: 0.9356866478919983:  31%|███       | 1987/6470 [25:41<51:30,  1.45it/s]
1|1988|Loss: 0.9358175992965698:  31%|███       | 1987/6470 [25:41<51:30,  1.45it/s]
1|1988|Loss: 0.9358175992965698:  31%|███       | 1988/6470 [25:42<52:05,  1.43it/s]
1|1989|Loss: 0.9093244075775146:  31%|███       | 1988/6470 [25:42<52:05,  1.43it/s]
1|1989|Loss: 0.9093244075775146:  31%|███       | 1989/6470 [25:42<52:51,  1.41it/s]
1|1990|Loss: 1.0621025562286377:  31%|███       | 1989/6470 [25:43<52:51,  1.41it/s]
1|1990|Loss: 1.0621025562286377:  31%|███       | 1990/6470 [25:43<54:21,  1.37it/s]
1|1991|Loss: 0.7798917293548584:  31%|███       | 1990/6470 [25:44<54:21,  1.37it/s]
1|1991|Loss: 0.7798917293548584:  31%|███       | 1991/6470 [25:44<54:20,  1.37it/s]
1|1992|Loss: 0.931496262550354:  31%|███       | 1991/6470 [25:44<54:20,  1.37it/s] 
1|1992|Loss: 0.931496262550354:  31%|███       | 1992/6470 [25:44<50:19,  1.48it/s]
1|1993|Loss: 1.0570064783096313:  31%|███       | 1992/6470 [25:45<50:19,  1.48it/s]
1|1993|Loss: 1.0570064783096313:  31%|███       | 1993/6470 [25:45<45:42,  1.63it/s]
1|1994|Loss: 0.96184241771698:  31%|███       | 1993/6470 [25:45<45:42,  1.63it/s]  
1|1994|Loss: 0.96184241771698:  31%|███       | 1994/6470 [25:46<46:37,  1.60it/s]
1|1995|Loss: 0.8813899755477905:  31%|███       | 1994/6470 [25:46<46:37,  1.60it/s]
1|1995|Loss: 0.8813899755477905:  31%|███       | 1995/6470 [25:46<47:25,  1.57it/s]
1|1996|Loss: 0.9961419105529785:  31%|███       | 1995/6470 [25:47<47:25,  1.57it/s]
1|1996|Loss: 0.9961419105529785:  31%|███       | 1996/6470 [25:47<47:07,  1.58it/s]
1|1997|Loss: 1.0111124515533447:  31%|███       | 1996/6470 [25:47<47:07,  1.58it/s]
1|1997|Loss: 1.0111124515533447:  31%|███       | 1997/6470 [25:48<48:50,  1.53it/s]
1|1998|Loss: 0.8617051839828491:  31%|███       | 1997/6470 [25:48<48:50,  1.53it/s]
1|1998|Loss: 0.8617051839828491:  31%|███       | 1998/6470 [25:48<51:35,  1.44it/s]
1|1999|Loss: 0.9849838018417358:  31%|███       | 1998/6470 [25:49<51:35,  1.44it/s]
1|1999|Loss: 0.9849838018417358:  31%|███       | 1999/6470 [25:49<49:12,  1.51it/s]
1|2000|Loss: 1.040968418121338:  31%|███       | 1999/6470 [25:49<49:12,  1.51it/s] 
1|2000|Loss: 1.040968418121338:  31%|███       | 2000/6470 [25:50<47:05,  1.58it/s]
1|2001|Loss: 0.9504256844520569:  31%|███       | 2000/6470 [25:50<47:05,  1.58it/s]
1|2001|Loss: 0.9504256844520569:  31%|███       | 2001/6470 [25:50<48:45,  1.53it/s]
1|2002|Loss: 1.0795975923538208:  31%|███       | 2001/6470 [25:51<48:45,  1.53it/s]
1|2002|Loss: 1.0795975923538208:  31%|███       | 2002/6470 [25:51<43:57,  1.69it/s]
1|2003|Loss: 0.8456236720085144:  31%|███       | 2002/6470 [25:51<43:57,  1.69it/s]
1|2003|Loss: 0.8456236720085144:  31%|███       | 2003/6470 [25:51<45:34,  1.63it/s]
1|2004|Loss: 0.8461346626281738:  31%|███       | 2003/6470 [25:52<45:34,  1.63it/s]
1|2004|Loss: 0.8461346626281738:  31%|███       | 2004/6470 [25:52<50:02,  1.49it/s]
1|2005|Loss: 0.850957453250885:  31%|███       | 2004/6470 [25:53<50:02,  1.49it/s] 
1|2005|Loss: 0.850957453250885:  31%|███       | 2005/6470 [25:53<54:23,  1.37it/s]
1|2006|Loss: 0.9877587556838989:  31%|███       | 2005/6470 [25:53<54:23,  1.37it/s]
1|2006|Loss: 0.9877587556838989:  31%|███       | 2006/6470 [25:54<56:29,  1.32it/s]
1|2007|Loss: 0.9757501482963562:  31%|███       | 2006/6470 [25:54<56:29,  1.32it/s]
1|2007|Loss: 0.9757501482963562:  31%|███       | 2007/6470 [25:54<54:07,  1.37it/s]
1|2008|Loss: 0.948166012763977:  31%|███       | 2007/6470 [25:55<54:07,  1.37it/s] 
1|2008|Loss: 0.948166012763977:  31%|███       | 2008/6470 [25:55<53:46,  1.38it/s]
1|2009|Loss: 1.1432729959487915:  31%|███       | 2008/6470 [25:56<53:46,  1.38it/s]
1|2009|Loss: 1.1432729959487915:  31%|███       | 2009/6470 [25:56<53:26,  1.39it/s]
1|2010|Loss: 0.9979698061943054:  31%|███       | 2009/6470 [25:56<53:26,  1.39it/s]
1|2010|Loss: 0.9979698061943054:  31%|███       | 2010/6470 [25:57<52:03,  1.43it/s]
1|2011|Loss: 0.8585701584815979:  31%|███       | 2010/6470 [25:57<52:03,  1.43it/s]
1|2011|Loss: 0.8585701584815979:  31%|███       | 2011/6470 [25:57<50:43,  1.46it/s]
1|2012|Loss: 0.9895060062408447:  31%|███       | 2011/6470 [25:58<50:43,  1.46it/s]
1|2012|Loss: 0.9895060062408447:  31%|███       | 2012/6470 [25:58<48:48,  1.52it/s]
1|2013|Loss: 1.1270934343338013:  31%|███       | 2012/6470 [25:58<48:48,  1.52it/s]
1|2013|Loss: 1.1270934343338013:  31%|███       | 2013/6470 [25:59<50:46,  1.46it/s]
1|2014|Loss: 1.0158262252807617:  31%|███       | 2013/6470 [25:59<50:46,  1.46it/s]
1|2014|Loss: 1.0158262252807617:  31%|███       | 2014/6470 [25:59<51:47,  1.43it/s]
1|2015|Loss: 0.9213906526565552:  31%|███       | 2014/6470 [26:00<51:47,  1.43it/s]
1|2015|Loss: 0.9213906526565552:  31%|███       | 2015/6470 [26:00<52:19,  1.42it/s]
1|2016|Loss: 0.9719617366790771:  31%|███       | 2015/6470 [26:00<52:19,  1.42it/s]
1|2016|Loss: 0.9719617366790771:  31%|███       | 2016/6470 [26:01<49:00,  1.51it/s]
1|2017|Loss: 0.8633061647415161:  31%|███       | 2016/6470 [26:01<49:00,  1.51it/s]
1|2017|Loss: 0.8633061647415161:  31%|███       | 2017/6470 [26:01<47:59,  1.55it/s]
1|2018|Loss: 0.8171581029891968:  31%|███       | 2017/6470 [26:02<47:59,  1.55it/s]
1|2018|Loss: 0.8171581029891968:  31%|███       | 2018/6470 [26:02<49:27,  1.50it/s]
1|2019|Loss: 1.1712841987609863:  31%|███       | 2018/6470 [26:02<49:27,  1.50it/s]
1|2019|Loss: 1.1712841987609863:  31%|███       | 2019/6470 [26:02<48:10,  1.54it/s]
1|2020|Loss: 1.0099643468856812:  31%|███       | 2019/6470 [26:03<48:10,  1.54it/s]
1|2020|Loss: 1.0099643468856812:  31%|███       | 2020/6470 [26:03<50:22,  1.47it/s]
1|2021|Loss: 1.2149027585983276:  31%|███       | 2020/6470 [26:04<50:22,  1.47it/s]
1|2021|Loss: 1.2149027585983276:  31%|███       | 2021/6470 [26:04<52:50,  1.40it/s]
1|2022|Loss: 0.9042260646820068:  31%|███       | 2021/6470 [26:04<52:50,  1.40it/s]
1|2022|Loss: 0.9042260646820068:  31%|███▏      | 2022/6470 [26:05<47:25,  1.56it/s]
1|2023|Loss: 0.9388560652732849:  31%|███▏      | 2022/6470 [26:05<47:25,  1.56it/s]
1|2023|Loss: 0.9388560652732849:  31%|███▏      | 2023/6470 [26:05<46:30,  1.59it/s]
1|2024|Loss: 0.7536171674728394:  31%|███▏      | 2023/6470 [26:06<46:30,  1.59it/s]
1|2024|Loss: 0.7536171674728394:  31%|███▏      | 2024/6470 [26:06<50:11,  1.48it/s]
1|2025|Loss: 0.7139564752578735:  31%|███▏      | 2024/6470 [26:06<50:11,  1.48it/s]
1|2025|Loss: 0.7139564752578735:  31%|███▏      | 2025/6470 [26:07<51:51,  1.43it/s]
1|2026|Loss: 1.0487388372421265:  31%|███▏      | 2025/6470 [26:07<51:51,  1.43it/s]
1|2026|Loss: 1.0487388372421265:  31%|███▏      | 2026/6470 [26:07<52:07,  1.42it/s]
1|2027|Loss: 0.9601448774337769:  31%|███▏      | 2026/6470 [26:08<52:07,  1.42it/s]
1|2027|Loss: 0.9601448774337769:  31%|███▏      | 2027/6470 [26:08<53:40,  1.38it/s]
1|2028|Loss: 0.8684139847755432:  31%|███▏      | 2027/6470 [26:09<53:40,  1.38it/s]
1|2028|Loss: 0.8684139847755432:  31%|███▏      | 2028/6470 [26:09<52:34,  1.41it/s]
1|2029|Loss: 0.912762463092804:  31%|███▏      | 2028/6470 [26:09<52:34,  1.41it/s] 
1|2029|Loss: 0.912762463092804:  31%|███▏      | 2029/6470 [26:09<49:20,  1.50it/s]
1|2030|Loss: 0.8216986656188965:  31%|███▏      | 2029/6470 [26:10<49:20,  1.50it/s]
1|2030|Loss: 0.8216986656188965:  31%|███▏      | 2030/6470 [26:10<50:14,  1.47it/s]
1|2031|Loss: 0.8820701837539673:  31%|███▏      | 2030/6470 [26:10<50:14,  1.47it/s]
1|2031|Loss: 0.8820701837539673:  31%|███▏      | 2031/6470 [26:11<49:35,  1.49it/s]
1|2032|Loss: 0.8464111089706421:  31%|███▏      | 2031/6470 [26:11<49:35,  1.49it/s]
1|2032|Loss: 0.8464111089706421:  31%|███▏      | 2032/6470 [26:12<53:35,  1.38it/s]
1|2033|Loss: 1.083988070487976:  31%|███▏      | 2032/6470 [26:12<53:35,  1.38it/s] 
1|2033|Loss: 1.083988070487976:  31%|███▏      | 2033/6470 [26:12<56:46,  1.30it/s]
1|2034|Loss: 1.192865252494812:  31%|███▏      | 2033/6470 [26:13<56:46,  1.30it/s]
1|2034|Loss: 1.192865252494812:  31%|███▏      | 2034/6470 [26:13<54:11,  1.36it/s]
1|2035|Loss: 1.1266143321990967:  31%|███▏      | 2034/6470 [26:13<54:11,  1.36it/s]
1|2035|Loss: 1.1266143321990967:  31%|███▏      | 2035/6470 [26:14<53:49,  1.37it/s]
1|2036|Loss: 0.9419426918029785:  31%|███▏      | 2035/6470 [26:14<53:49,  1.37it/s]
1|2036|Loss: 0.9419426918029785:  31%|███▏      | 2036/6470 [26:15<55:06,  1.34it/s]
1|2037|Loss: 1.0494155883789062:  31%|███▏      | 2036/6470 [26:15<55:06,  1.34it/s]
1|2037|Loss: 1.0494155883789062:  31%|███▏      | 2037/6470 [26:15<53:07,  1.39it/s]
1|2038|Loss: 0.7909606695175171:  31%|███▏      | 2037/6470 [26:16<53:07,  1.39it/s]
1|2038|Loss: 0.7909606695175171:  31%|███▏      | 2038/6470 [26:16<53:06,  1.39it/s]
1|2039|Loss: 0.7854965925216675:  31%|███▏      | 2038/6470 [26:16<53:06,  1.39it/s]
1|2039|Loss: 0.7854965925216675:  32%|███▏      | 2039/6470 [26:17<53:49,  1.37it/s]
1|2040|Loss: 0.921786367893219:  32%|███▏      | 2039/6470 [26:17<53:49,  1.37it/s] 
1|2040|Loss: 0.921786367893219:  32%|███▏      | 2040/6470 [26:17<52:03,  1.42it/s]
1|2041|Loss: 0.995691180229187:  32%|███▏      | 2040/6470 [26:18<52:03,  1.42it/s]
1|2041|Loss: 0.995691180229187:  32%|███▏      | 2041/6470 [26:18<48:10,  1.53it/s]
1|2042|Loss: 0.9700915813446045:  32%|███▏      | 2041/6470 [26:18<48:10,  1.53it/s]
1|2042|Loss: 0.9700915813446045:  32%|███▏      | 2042/6470 [26:19<47:16,  1.56it/s]
1|2043|Loss: 0.9329986572265625:  32%|███▏      | 2042/6470 [26:19<47:16,  1.56it/s]
1|2043|Loss: 0.9329986572265625:  32%|███▏      | 2043/6470 [26:19<48:26,  1.52it/s]
1|2044|Loss: 1.0104076862335205:  32%|███▏      | 2043/6470 [26:20<48:26,  1.52it/s]
1|2044|Loss: 1.0104076862335205:  32%|███▏      | 2044/6470 [26:20<52:34,  1.40it/s]
1|2045|Loss: 0.9335626363754272:  32%|███▏      | 2044/6470 [26:21<52:34,  1.40it/s]
1|2045|Loss: 0.9335626363754272:  32%|███▏      | 2045/6470 [26:21<54:02,  1.36it/s]
1|2046|Loss: 0.9506557583808899:  32%|███▏      | 2045/6470 [26:21<54:02,  1.36it/s]
1|2046|Loss: 0.9506557583808899:  32%|███▏      | 2046/6470 [26:22<52:16,  1.41it/s]
1|2047|Loss: 0.875063955783844:  32%|███▏      | 2046/6470 [26:22<52:16,  1.41it/s] 
1|2047|Loss: 0.875063955783844:  32%|███▏      | 2047/6470 [26:22<53:17,  1.38it/s]
1|2048|Loss: 0.8848173022270203:  32%|███▏      | 2047/6470 [26:23<53:17,  1.38it/s]
1|2048|Loss: 0.8848173022270203:  32%|███▏      | 2048/6470 [26:23<55:40,  1.32it/s]
1|2049|Loss: 0.9438304305076599:  32%|███▏      | 2048/6470 [26:23<55:40,  1.32it/s]
1|2049|Loss: 0.9438304305076599:  32%|███▏      | 2049/6470 [26:24<53:35,  1.37it/s]
1|2050|Loss: 0.8948479890823364:  32%|███▏      | 2049/6470 [26:24<53:35,  1.37it/s]
1|2050|Loss: 0.8948479890823364:  32%|███▏      | 2050/6470 [26:25<54:56,  1.34it/s]
1|2051|Loss: 0.8367143869400024:  32%|███▏      | 2050/6470 [26:25<54:56,  1.34it/s]
1|2051|Loss: 0.8367143869400024:  32%|███▏      | 2051/6470 [26:25<56:54,  1.29it/s]
1|2052|Loss: 1.0502948760986328:  32%|███▏      | 2051/6470 [26:26<56:54,  1.29it/s]
1|2052|Loss: 1.0502948760986328:  32%|███▏      | 2052/6470 [26:26<56:54,  1.29it/s]
1|2053|Loss: 1.038169026374817:  32%|███▏      | 2052/6470 [26:27<56:54,  1.29it/s] 
1|2053|Loss: 1.038169026374817:  32%|███▏      | 2053/6470 [26:27<57:19,  1.28it/s]
1|2054|Loss: 0.9281423091888428:  32%|███▏      | 2053/6470 [26:27<57:19,  1.28it/s]
1|2054|Loss: 0.9281423091888428:  32%|███▏      | 2054/6470 [26:28<54:41,  1.35it/s]
1|2055|Loss: 0.921010434627533:  32%|███▏      | 2054/6470 [26:28<54:41,  1.35it/s] 
1|2055|Loss: 0.921010434627533:  32%|███▏      | 2055/6470 [26:28<56:39,  1.30it/s]
1|2056|Loss: 0.9662833213806152:  32%|███▏      | 2055/6470 [26:29<56:39,  1.30it/s]
1|2056|Loss: 0.9662833213806152:  32%|███▏      | 2056/6470 [26:29<58:02,  1.27it/s]
1|2057|Loss: 0.9855544567108154:  32%|███▏      | 2056/6470 [26:30<58:02,  1.27it/s]
1|2057|Loss: 0.9855544567108154:  32%|███▏      | 2057/6470 [26:30<55:41,  1.32it/s]
1|2058|Loss: 0.8965662717819214:  32%|███▏      | 2057/6470 [26:30<55:41,  1.32it/s]
1|2058|Loss: 0.8965662717819214:  32%|███▏      | 2058/6470 [26:31<57:58,  1.27it/s]
1|2059|Loss: 0.8585182428359985:  32%|███▏      | 2058/6470 [26:31<57:58,  1.27it/s]
1|2059|Loss: 0.8585182428359985:  32%|███▏      | 2059/6470 [26:32<55:59,  1.31it/s]
1|2060|Loss: 0.9796169400215149:  32%|███▏      | 2059/6470 [26:32<55:59,  1.31it/s]
1|2060|Loss: 0.9796169400215149:  32%|███▏      | 2060/6470 [26:32<51:43,  1.42it/s]
1|2061|Loss: 0.9039274454116821:  32%|███▏      | 2060/6470 [26:32<51:43,  1.42it/s]
1|2061|Loss: 0.9039274454116821:  32%|███▏      | 2061/6470 [26:33<49:55,  1.47it/s]
1|2062|Loss: 0.8411650061607361:  32%|███▏      | 2061/6470 [26:33<49:55,  1.47it/s]
1|2062|Loss: 0.8411650061607361:  32%|███▏      | 2062/6470 [26:33<52:24,  1.40it/s]
1|2063|Loss: 0.9461972713470459:  32%|███▏      | 2062/6470 [26:34<52:24,  1.40it/s]
1|2063|Loss: 0.9461972713470459:  32%|███▏      | 2063/6470 [26:34<51:24,  1.43it/s]
1|2064|Loss: 0.8084184527397156:  32%|███▏      | 2063/6470 [26:35<51:24,  1.43it/s]
1|2064|Loss: 0.8084184527397156:  32%|███▏      | 2064/6470 [26:35<49:56,  1.47it/s]
1|2065|Loss: 1.07906973361969:  32%|███▏      | 2064/6470 [26:35<49:56,  1.47it/s]  
1|2065|Loss: 1.07906973361969:  32%|███▏      | 2065/6470 [26:35<49:00,  1.50it/s]
1|2066|Loss: 0.9161856770515442:  32%|███▏      | 2065/6470 [26:36<49:00,  1.50it/s]
1|2066|Loss: 0.9161856770515442:  32%|███▏      | 2066/6470 [26:36<47:18,  1.55it/s]
1|2067|Loss: 0.9081671833992004:  32%|███▏      | 2066/6470 [26:36<47:18,  1.55it/s]
1|2067|Loss: 0.9081671833992004:  32%|███▏      | 2067/6470 [26:37<48:33,  1.51it/s]
1|2068|Loss: 1.0120760202407837:  32%|███▏      | 2067/6470 [26:37<48:33,  1.51it/s]
1|2068|Loss: 1.0120760202407837:  32%|███▏      | 2068/6470 [26:38<51:20,  1.43it/s]
1|2069|Loss: 1.0094736814498901:  32%|███▏      | 2068/6470 [26:38<51:20,  1.43it/s]
1|2069|Loss: 1.0094736814498901:  32%|███▏      | 2069/6470 [26:38<50:54,  1.44it/s]
1|2070|Loss: 1.03275465965271:  32%|███▏      | 2069/6470 [26:39<50:54,  1.44it/s]  
1|2070|Loss: 1.03275465965271:  32%|███▏      | 2070/6470 [26:39<53:10,  1.38it/s]
1|2071|Loss: 0.8704652190208435:  32%|███▏      | 2070/6470 [26:39<53:10,  1.38it/s]
1|2071|Loss: 0.8704652190208435:  32%|███▏      | 2071/6470 [26:40<52:18,  1.40it/s]
1|2072|Loss: 0.85561603307724:  32%|███▏      | 2071/6470 [26:40<52:18,  1.40it/s]  
1|2072|Loss: 0.85561603307724:  32%|███▏      | 2072/6470 [26:40<53:36,  1.37it/s]
1|2073|Loss: 0.8574169874191284:  32%|███▏      | 2072/6470 [26:41<53:36,  1.37it/s]
1|2073|Loss: 0.8574169874191284:  32%|███▏      | 2073/6470 [26:41<54:01,  1.36it/s]
1|2074|Loss: 0.9109397530555725:  32%|███▏      | 2073/6470 [26:42<54:01,  1.36it/s]
1|2074|Loss: 0.9109397530555725:  32%|███▏      | 2074/6470 [26:42<53:29,  1.37it/s]
1|2075|Loss: 0.9512082934379578:  32%|███▏      | 2074/6470 [26:42<53:29,  1.37it/s]
1|2075|Loss: 0.9512082934379578:  32%|███▏      | 2075/6470 [26:43<53:26,  1.37it/s]
1|2076|Loss: 0.8797687292098999:  32%|███▏      | 2075/6470 [26:43<53:26,  1.37it/s]
1|2076|Loss: 0.8797687292098999:  32%|███▏      | 2076/6470 [26:43<53:45,  1.36it/s]
1|2077|Loss: 0.9591457843780518:  32%|███▏      | 2076/6470 [26:44<53:45,  1.36it/s]
1|2077|Loss: 0.9591457843780518:  32%|███▏      | 2077/6470 [26:44<55:53,  1.31it/s]
1|2078|Loss: 0.9352208971977234:  32%|███▏      | 2077/6470 [26:45<55:53,  1.31it/s]
1|2078|Loss: 0.9352208971977234:  32%|███▏      | 2078/6470 [26:45<55:16,  1.32it/s]
1|2079|Loss: 0.8369151949882507:  32%|███▏      | 2078/6470 [26:45<55:16,  1.32it/s]
1|2079|Loss: 0.8369151949882507:  32%|███▏      | 2079/6470 [26:46<54:16,  1.35it/s]
1|2080|Loss: 1.0767052173614502:  32%|███▏      | 2079/6470 [26:46<54:16,  1.35it/s]
1|2080|Loss: 1.0767052173614502:  32%|███▏      | 2080/6470 [26:46<53:58,  1.36it/s]
1|2081|Loss: 0.9982185959815979:  32%|███▏      | 2080/6470 [26:47<53:58,  1.36it/s]
1|2081|Loss: 0.9982185959815979:  32%|███▏      | 2081/6470 [26:47<51:53,  1.41it/s]
1|2082|Loss: 0.8835709691047668:  32%|███▏      | 2081/6470 [26:47<51:53,  1.41it/s]
1|2082|Loss: 0.8835709691047668:  32%|███▏      | 2082/6470 [26:48<51:24,  1.42it/s]
1|2083|Loss: 0.901366651058197:  32%|███▏      | 2082/6470 [26:48<51:24,  1.42it/s] 
1|2083|Loss: 0.901366651058197:  32%|███▏      | 2083/6470 [26:49<53:38,  1.36it/s]
1|2084|Loss: 0.8891629576683044:  32%|███▏      | 2083/6470 [26:49<53:38,  1.36it/s]
1|2084|Loss: 0.8891629576683044:  32%|███▏      | 2084/6470 [26:49<53:37,  1.36it/s]
1|2085|Loss: 0.9298378825187683:  32%|███▏      | 2084/6470 [26:50<53:37,  1.36it/s]
1|2085|Loss: 0.9298378825187683:  32%|███▏      | 2085/6470 [26:50<55:06,  1.33it/s]
1|2086|Loss: 0.8462682366371155:  32%|███▏      | 2085/6470 [26:51<55:06,  1.33it/s]
1|2086|Loss: 0.8462682366371155:  32%|███▏      | 2086/6470 [26:51<56:17,  1.30it/s]
1|2087|Loss: 0.8024362921714783:  32%|███▏      | 2086/6470 [26:51<56:17,  1.30it/s]
1|2087|Loss: 0.8024362921714783:  32%|███▏      | 2087/6470 [26:51<49:12,  1.48it/s]
1|2088|Loss: 0.869428813457489:  32%|███▏      | 2087/6470 [26:52<49:12,  1.48it/s] 
1|2088|Loss: 0.869428813457489:  32%|███▏      | 2088/6470 [26:52<46:50,  1.56it/s]
1|2089|Loss: 0.8444538116455078:  32%|███▏      | 2088/6470 [26:52<46:50,  1.56it/s]
1|2089|Loss: 0.8444538116455078:  32%|███▏      | 2089/6470 [26:53<51:06,  1.43it/s]
1|2090|Loss: 0.955674946308136:  32%|███▏      | 2089/6470 [26:53<51:06,  1.43it/s] 
1|2090|Loss: 0.955674946308136:  32%|███▏      | 2090/6470 [26:54<53:52,  1.36it/s]
1|2091|Loss: 0.9330986738204956:  32%|███▏      | 2090/6470 [26:54<53:52,  1.36it/s]
1|2091|Loss: 0.9330986738204956:  32%|███▏      | 2091/6470 [26:54<54:17,  1.34it/s]
1|2092|Loss: 1.0875672101974487:  32%|███▏      | 2091/6470 [26:55<54:17,  1.34it/s]
1|2092|Loss: 1.0875672101974487:  32%|███▏      | 2092/6470 [26:55<52:12,  1.40it/s]
1|2093|Loss: 0.983957052230835:  32%|███▏      | 2092/6470 [26:55<52:12,  1.40it/s] 
1|2093|Loss: 0.983957052230835:  32%|███▏      | 2093/6470 [26:56<51:32,  1.42it/s]
1|2094|Loss: 0.8127906322479248:  32%|███▏      | 2093/6470 [26:56<51:32,  1.42it/s]
1|2094|Loss: 0.8127906322479248:  32%|███▏      | 2094/6470 [26:56<51:52,  1.41it/s]
1|2095|Loss: 0.9859498143196106:  32%|███▏      | 2094/6470 [26:57<51:52,  1.41it/s]
1|2095|Loss: 0.9859498143196106:  32%|███▏      | 2095/6470 [26:57<51:47,  1.41it/s]
1|2096|Loss: 0.9728776812553406:  32%|███▏      | 2095/6470 [26:57<51:47,  1.41it/s]
1|2096|Loss: 0.9728776812553406:  32%|███▏      | 2096/6470 [26:58<47:07,  1.55it/s]
1|2097|Loss: 1.0542635917663574:  32%|███▏      | 2096/6470 [26:58<47:07,  1.55it/s]
1|2097|Loss: 1.0542635917663574:  32%|███▏      | 2097/6470 [26:58<45:58,  1.59it/s]
1|2098|Loss: 0.724266529083252:  32%|███▏      | 2097/6470 [26:59<45:58,  1.59it/s] 
1|2098|Loss: 0.724266529083252:  32%|███▏      | 2098/6470 [26:59<45:47,  1.59it/s]
1|2099|Loss: 1.0539847612380981:  32%|███▏      | 2098/6470 [26:59<45:47,  1.59it/s]
1|2099|Loss: 1.0539847612380981:  32%|███▏      | 2099/6470 [26:59<44:22,  1.64it/s]
1|2100|Loss: 0.9092671275138855:  32%|███▏      | 2099/6470 [27:00<44:22,  1.64it/s]
1|2100|Loss: 0.9092671275138855:  32%|███▏      | 2100/6470 [27:00<44:21,  1.64it/s]
1|2101|Loss: 0.9381438493728638:  32%|███▏      | 2100/6470 [27:00<44:21,  1.64it/s]
1|2101|Loss: 0.9381438493728638:  32%|███▏      | 2101/6470 [27:01<45:56,  1.58it/s]
1|2102|Loss: 0.9724444150924683:  32%|███▏      | 2101/6470 [27:01<45:56,  1.58it/s]
1|2102|Loss: 0.9724444150924683:  32%|███▏      | 2102/6470 [27:01<48:02,  1.52it/s]
1|2103|Loss: 0.9323756098747253:  32%|███▏      | 2102/6470 [27:02<48:02,  1.52it/s]
1|2103|Loss: 0.9323756098747253:  33%|███▎      | 2103/6470 [27:02<50:00,  1.46it/s]
1|2104|Loss: 0.8386945128440857:  33%|███▎      | 2103/6470 [27:03<50:00,  1.46it/s]
1|2104|Loss: 0.8386945128440857:  33%|███▎      | 2104/6470 [27:03<52:47,  1.38it/s]
1|2105|Loss: 1.0189796686172485:  33%|███▎      | 2104/6470 [27:03<52:47,  1.38it/s]
1|2105|Loss: 1.0189796686172485:  33%|███▎      | 2105/6470 [27:04<54:59,  1.32it/s]
1|2106|Loss: 1.1106345653533936:  33%|███▎      | 2105/6470 [27:04<54:59,  1.32it/s]
1|2106|Loss: 1.1106345653533936:  33%|███▎      | 2106/6470 [27:05<54:56,  1.32it/s]
1|2107|Loss: 0.8744032979011536:  33%|███▎      | 2106/6470 [27:05<54:56,  1.32it/s]
1|2107|Loss: 0.8744032979011536:  33%|███▎      | 2107/6470 [27:05<54:23,  1.34it/s]
1|2108|Loss: 0.9506981372833252:  33%|███▎      | 2107/6470 [27:06<54:23,  1.34it/s]
1|2108|Loss: 0.9506981372833252:  33%|███▎      | 2108/6470 [27:06<52:32,  1.38it/s]
1|2109|Loss: 0.7597838640213013:  33%|███▎      | 2108/6470 [27:06<52:32,  1.38it/s]
1|2109|Loss: 0.7597838640213013:  33%|███▎      | 2109/6470 [27:07<53:24,  1.36it/s]
1|2110|Loss: 0.785376250743866:  33%|███▎      | 2109/6470 [27:07<53:24,  1.36it/s] 
1|2110|Loss: 0.785376250743866:  33%|███▎      | 2110/6470 [27:07<53:46,  1.35it/s]
1|2111|Loss: 0.9689894914627075:  33%|███▎      | 2110/6470 [27:08<53:46,  1.35it/s]
1|2111|Loss: 0.9689894914627075:  33%|███▎      | 2111/6470 [27:08<55:15,  1.31it/s]
1|2112|Loss: 0.9668178558349609:  33%|███▎      | 2111/6470 [27:09<55:15,  1.31it/s]
1|2112|Loss: 0.9668178558349609:  33%|███▎      | 2112/6470 [27:09<58:16,  1.25it/s]
1|2113|Loss: 1.0263746976852417:  33%|███▎      | 2112/6470 [27:10<58:16,  1.25it/s]
1|2113|Loss: 1.0263746976852417:  33%|███▎      | 2113/6470 [27:10<57:33,  1.26it/s]
1|2114|Loss: 1.072462558746338:  33%|███▎      | 2113/6470 [27:10<57:33,  1.26it/s] 
1|2114|Loss: 1.072462558746338:  33%|███▎      | 2114/6470 [27:11<57:09,  1.27it/s]
1|2115|Loss: 0.8976778388023376:  33%|███▎      | 2114/6470 [27:11<57:09,  1.27it/s]
1|2115|Loss: 0.8976778388023376:  33%|███▎      | 2115/6470 [27:11<53:03,  1.37it/s]
1|2116|Loss: 0.9511196613311768:  33%|███▎      | 2115/6470 [27:12<53:03,  1.37it/s]
1|2116|Loss: 0.9511196613311768:  33%|███▎      | 2116/6470 [27:12<52:19,  1.39it/s]
1|2117|Loss: 0.9992339015007019:  33%|███▎      | 2116/6470 [27:12<52:19,  1.39it/s]
1|2117|Loss: 0.9992339015007019:  33%|███▎      | 2117/6470 [27:13<53:54,  1.35it/s]
1|2118|Loss: 0.9892381429672241:  33%|███▎      | 2117/6470 [27:13<53:54,  1.35it/s]
1|2118|Loss: 0.9892381429672241:  33%|███▎      | 2118/6470 [27:14<56:06,  1.29it/s]
1|2119|Loss: 0.8333044648170471:  33%|███▎      | 2118/6470 [27:14<56:06,  1.29it/s]
1|2119|Loss: 0.8333044648170471:  33%|███▎      | 2119/6470 [27:14<55:19,  1.31it/s]
1|2120|Loss: 0.8326159715652466:  33%|███▎      | 2119/6470 [27:15<55:19,  1.31it/s]
1|2120|Loss: 0.8326159715652466:  33%|███▎      | 2120/6470 [27:15<52:30,  1.38it/s]
1|2121|Loss: 0.8670979142189026:  33%|███▎      | 2120/6470 [27:15<52:30,  1.38it/s]
1|2121|Loss: 0.8670979142189026:  33%|███▎      | 2121/6470 [27:16<50:30,  1.44it/s]
1|2122|Loss: 0.909019410610199:  33%|███▎      | 2121/6470 [27:16<50:30,  1.44it/s] 
1|2122|Loss: 0.909019410610199:  33%|███▎      | 2122/6470 [27:16<51:11,  1.42it/s]
1|2123|Loss: 0.9363481998443604:  33%|███▎      | 2122/6470 [27:17<51:11,  1.42it/s]
1|2123|Loss: 0.9363481998443604:  33%|███▎      | 2123/6470 [27:17<45:24,  1.60it/s]
1|2124|Loss: 0.9455211162567139:  33%|███▎      | 2123/6470 [27:17<45:24,  1.60it/s]
1|2124|Loss: 0.9455211162567139:  33%|███▎      | 2124/6470 [27:17<43:32,  1.66it/s]
1|2125|Loss: 1.0165858268737793:  33%|███▎      | 2124/6470 [27:18<43:32,  1.66it/s]
1|2125|Loss: 1.0165858268737793:  33%|███▎      | 2125/6470 [27:18<43:04,  1.68it/s]
1|2126|Loss: 0.7687376737594604:  33%|███▎      | 2125/6470 [27:18<43:04,  1.68it/s]
1|2126|Loss: 0.7687376737594604:  33%|███▎      | 2126/6470 [27:19<46:39,  1.55it/s]
1|2127|Loss: 0.9802291393280029:  33%|███▎      | 2126/6470 [27:19<46:39,  1.55it/s]
1|2127|Loss: 0.9802291393280029:  33%|███▎      | 2127/6470 [27:19<45:47,  1.58it/s]
1|2128|Loss: 1.038246989250183:  33%|███▎      | 2127/6470 [27:20<45:47,  1.58it/s] 
1|2128|Loss: 1.038246989250183:  33%|███▎      | 2128/6470 [27:20<44:31,  1.63it/s]
1|2129|Loss: 0.8518499135971069:  33%|███▎      | 2128/6470 [27:20<44:31,  1.63it/s]
1|2129|Loss: 0.8518499135971069:  33%|███▎      | 2129/6470 [27:20<44:49,  1.61it/s]
1|2130|Loss: 0.7693127989768982:  33%|███▎      | 2129/6470 [27:21<44:49,  1.61it/s]
1|2130|Loss: 0.7693127989768982:  33%|███▎      | 2130/6470 [27:21<49:27,  1.46it/s]
1|2131|Loss: 0.9528599977493286:  33%|███▎      | 2130/6470 [27:22<49:27,  1.46it/s]
1|2131|Loss: 0.9528599977493286:  33%|███▎      | 2131/6470 [27:22<52:17,  1.38it/s]
1|2132|Loss: 0.9329562783241272:  33%|███▎      | 2131/6470 [27:23<52:17,  1.38it/s]
1|2132|Loss: 0.9329562783241272:  33%|███▎      | 2132/6470 [27:23<50:54,  1.42it/s]
1|2133|Loss: 1.046807050704956:  33%|███▎      | 2132/6470 [27:23<50:54,  1.42it/s] 
1|2133|Loss: 1.046807050704956:  33%|███▎      | 2133/6470 [27:23<48:22,  1.49it/s]
1|2134|Loss: 0.8673415184020996:  33%|███▎      | 2133/6470 [27:24<48:22,  1.49it/s]
1|2134|Loss: 0.8673415184020996:  33%|███▎      | 2134/6470 [27:24<47:32,  1.52it/s]
1|2135|Loss: 0.7822868824005127:  33%|███▎      | 2134/6470 [27:24<47:32,  1.52it/s]
1|2135|Loss: 0.7822868824005127:  33%|███▎      | 2135/6470 [27:25<50:56,  1.42it/s]
1|2136|Loss: 0.8821377754211426:  33%|███▎      | 2135/6470 [27:25<50:56,  1.42it/s]
1|2136|Loss: 0.8821377754211426:  33%|███▎      | 2136/6470 [27:26<52:10,  1.38it/s]
1|2137|Loss: 0.7867852449417114:  33%|███▎      | 2136/6470 [27:26<52:10,  1.38it/s]
1|2137|Loss: 0.7867852449417114:  33%|███▎      | 2137/6470 [27:26<52:46,  1.37it/s]
1|2138|Loss: 0.9412280917167664:  33%|███▎      | 2137/6470 [27:27<52:46,  1.37it/s]
1|2138|Loss: 0.9412280917167664:  33%|███▎      | 2138/6470 [27:27<55:22,  1.30it/s]
1|2139|Loss: 0.935068666934967:  33%|███▎      | 2138/6470 [27:28<55:22,  1.30it/s] 
1|2139|Loss: 0.935068666934967:  33%|███▎      | 2139/6470 [27:28<52:46,  1.37it/s]
1|2140|Loss: 0.8591226935386658:  33%|███▎      | 2139/6470 [27:28<52:46,  1.37it/s]
1|2140|Loss: 0.8591226935386658:  33%|███▎      | 2140/6470 [27:29<54:35,  1.32it/s]
1|2141|Loss: 1.1396757364273071:  33%|███▎      | 2140/6470 [27:29<54:35,  1.32it/s]
1|2141|Loss: 1.1396757364273071:  33%|███▎      | 2141/6470 [27:29<56:07,  1.29it/s]
1|2142|Loss: 0.9522495865821838:  33%|███▎      | 2141/6470 [27:30<56:07,  1.29it/s]
1|2142|Loss: 0.9522495865821838:  33%|███▎      | 2142/6470 [27:30<53:30,  1.35it/s]
1|2143|Loss: 0.9264615774154663:  33%|███▎      | 2142/6470 [27:30<53:30,  1.35it/s]
1|2143|Loss: 0.9264615774154663:  33%|███▎      | 2143/6470 [27:31<51:11,  1.41it/s]
1|2144|Loss: 0.8581779599189758:  33%|███▎      | 2143/6470 [27:31<51:11,  1.41it/s]
1|2144|Loss: 0.8581779599189758:  33%|███▎      | 2144/6470 [27:31<51:22,  1.40it/s]
1|2145|Loss: 0.9149057865142822:  33%|███▎      | 2144/6470 [27:32<51:22,  1.40it/s]
1|2145|Loss: 0.9149057865142822:  33%|███▎      | 2145/6470 [27:32<50:14,  1.43it/s]
1|2146|Loss: 0.9340204000473022:  33%|███▎      | 2145/6470 [27:33<50:14,  1.43it/s]
1|2146|Loss: 0.9340204000473022:  33%|███▎      | 2146/6470 [27:33<50:10,  1.44it/s]
1|2147|Loss: 0.9896819591522217:  33%|███▎      | 2146/6470 [27:33<50:10,  1.44it/s]
1|2147|Loss: 0.9896819591522217:  33%|███▎      | 2147/6470 [27:34<49:07,  1.47it/s]
1|2148|Loss: 0.8564473986625671:  33%|███▎      | 2147/6470 [27:34<49:07,  1.47it/s]
1|2148|Loss: 0.8564473986625671:  33%|███▎      | 2148/6470 [27:34<52:14,  1.38it/s]
1|2149|Loss: 0.9490959644317627:  33%|███▎      | 2148/6470 [27:35<52:14,  1.38it/s]
1|2149|Loss: 0.9490959644317627:  33%|███▎      | 2149/6470 [27:35<53:22,  1.35it/s]
1|2150|Loss: 1.1215994358062744:  33%|███▎      | 2149/6470 [27:35<53:22,  1.35it/s]
1|2150|Loss: 1.1215994358062744:  33%|███▎      | 2150/6470 [27:36<51:24,  1.40it/s]
1|2151|Loss: 0.876967191696167:  33%|███▎      | 2150/6470 [27:36<51:24,  1.40it/s] 
1|2151|Loss: 0.876967191696167:  33%|███▎      | 2151/6470 [27:36<44:54,  1.60it/s]
1|2152|Loss: 1.153971791267395:  33%|███▎      | 2151/6470 [27:36<44:54,  1.60it/s]
1|2152|Loss: 1.153971791267395:  33%|███▎      | 2152/6470 [27:37<43:51,  1.64it/s]
1|2153|Loss: 1.0555821657180786:  33%|███▎      | 2152/6470 [27:37<43:51,  1.64it/s]
1|2153|Loss: 1.0555821657180786:  33%|███▎      | 2153/6470 [27:37<41:37,  1.73it/s]
1|2154|Loss: 0.9831665754318237:  33%|███▎      | 2153/6470 [27:38<41:37,  1.73it/s]
1|2154|Loss: 0.9831665754318237:  33%|███▎      | 2154/6470 [27:38<43:09,  1.67it/s]
1|2155|Loss: 1.0942355394363403:  33%|███▎      | 2154/6470 [27:38<43:09,  1.67it/s]
1|2155|Loss: 1.0942355394363403:  33%|███▎      | 2155/6470 [27:38<41:00,  1.75it/s]
1|2156|Loss: 0.9209022521972656:  33%|███▎      | 2155/6470 [27:39<41:00,  1.75it/s]
1|2156|Loss: 0.9209022521972656:  33%|███▎      | 2156/6470 [27:39<44:51,  1.60it/s]
1|2157|Loss: 0.9854212999343872:  33%|███▎      | 2156/6470 [27:40<44:51,  1.60it/s]
1|2157|Loss: 0.9854212999343872:  33%|███▎      | 2157/6470 [27:40<47:12,  1.52it/s]
1|2158|Loss: 0.9657532572746277:  33%|███▎      | 2157/6470 [27:40<47:12,  1.52it/s]
1|2158|Loss: 0.9657532572746277:  33%|███▎      | 2158/6470 [27:41<49:02,  1.47it/s]
1|2159|Loss: 0.9938368201255798:  33%|███▎      | 2158/6470 [27:41<49:02,  1.47it/s]
1|2159|Loss: 0.9938368201255798:  33%|███▎      | 2159/6470 [27:41<50:42,  1.42it/s]
1|2160|Loss: 0.9963046312332153:  33%|███▎      | 2159/6470 [27:42<50:42,  1.42it/s]
1|2160|Loss: 0.9963046312332153:  33%|███▎      | 2160/6470 [27:42<48:24,  1.48it/s]
1|2161|Loss: 0.9162420034408569:  33%|███▎      | 2160/6470 [27:42<48:24,  1.48it/s]
1|2161|Loss: 0.9162420034408569:  33%|███▎      | 2161/6470 [27:43<48:55,  1.47it/s]
1|2162|Loss: 0.9477719068527222:  33%|███▎      | 2161/6470 [27:43<48:55,  1.47it/s]
1|2162|Loss: 0.9477719068527222:  33%|███▎      | 2162/6470 [27:43<50:17,  1.43it/s]
1|2163|Loss: 0.7951285243034363:  33%|███▎      | 2162/6470 [27:44<50:17,  1.43it/s]
1|2163|Loss: 0.7951285243034363:  33%|███▎      | 2163/6470 [27:44<52:13,  1.37it/s]
1|2164|Loss: 0.9623699188232422:  33%|███▎      | 2163/6470 [27:45<52:13,  1.37it/s]
1|2164|Loss: 0.9623699188232422:  33%|███▎      | 2164/6470 [27:45<50:45,  1.41it/s]
1|2165|Loss: 0.6059430241584778:  33%|███▎      | 2164/6470 [27:45<50:45,  1.41it/s]
1|2165|Loss: 0.6059430241584778:  33%|███▎      | 2165/6470 [27:46<56:09,  1.28it/s]
1|2166|Loss: 1.072278618812561:  33%|███▎      | 2165/6470 [27:46<56:09,  1.28it/s] 
1|2166|Loss: 1.072278618812561:  33%|███▎      | 2166/6470 [27:47<56:40,  1.27it/s]
1|2167|Loss: 0.929384171962738:  33%|███▎      | 2166/6470 [27:47<56:40,  1.27it/s]
1|2167|Loss: 0.929384171962738:  33%|███▎      | 2167/6470 [27:47<50:41,  1.41it/s]
1|2168|Loss: 0.9717701077461243:  33%|███▎      | 2167/6470 [27:47<50:41,  1.41it/s]
1|2168|Loss: 0.9717701077461243:  34%|███▎      | 2168/6470 [27:48<47:56,  1.50it/s]
1|2169|Loss: 0.9529609680175781:  34%|███▎      | 2168/6470 [27:48<47:56,  1.50it/s]
1|2169|Loss: 0.9529609680175781:  34%|███▎      | 2169/6470 [27:48<45:42,  1.57it/s]
1|2170|Loss: 0.9578691720962524:  34%|███▎      | 2169/6470 [27:49<45:42,  1.57it/s]
1|2170|Loss: 0.9578691720962524:  34%|███▎      | 2170/6470 [27:49<45:33,  1.57it/s]
1|2171|Loss: 1.0229424238204956:  34%|███▎      | 2170/6470 [27:49<45:33,  1.57it/s]
1|2171|Loss: 1.0229424238204956:  34%|███▎      | 2171/6470 [27:50<44:31,  1.61it/s]
1|2172|Loss: 0.838440477848053:  34%|███▎      | 2171/6470 [27:50<44:31,  1.61it/s] 
1|2172|Loss: 0.838440477848053:  34%|███▎      | 2172/6470 [27:50<43:31,  1.65it/s]
1|2173|Loss: 0.9426579475402832:  34%|███▎      | 2172/6470 [27:51<43:31,  1.65it/s]
1|2173|Loss: 0.9426579475402832:  34%|███▎      | 2173/6470 [27:51<47:21,  1.51it/s]
1|2174|Loss: 0.9161986708641052:  34%|███▎      | 2173/6470 [27:51<47:21,  1.51it/s]
1|2174|Loss: 0.9161986708641052:  34%|███▎      | 2174/6470 [27:52<47:54,  1.49it/s]
1|2175|Loss: 0.9660751819610596:  34%|███▎      | 2174/6470 [27:52<47:54,  1.49it/s]
1|2175|Loss: 0.9660751819610596:  34%|███▎      | 2175/6470 [27:52<48:52,  1.46it/s]
1|2176|Loss: 0.9302061200141907:  34%|███▎      | 2175/6470 [27:53<48:52,  1.46it/s]
1|2176|Loss: 0.9302061200141907:  34%|███▎      | 2176/6470 [27:53<52:07,  1.37it/s]
1|2177|Loss: 0.7174329161643982:  34%|███▎      | 2176/6470 [27:54<52:07,  1.37it/s]
1|2177|Loss: 0.7174329161643982:  34%|███▎      | 2177/6470 [27:54<51:20,  1.39it/s]
1|2178|Loss: 1.0734710693359375:  34%|███▎      | 2177/6470 [27:54<51:20,  1.39it/s]
1|2178|Loss: 1.0734710693359375:  34%|███▎      | 2178/6470 [27:54<46:28,  1.54it/s]
1|2179|Loss: 1.034744381904602:  34%|███▎      | 2178/6470 [27:55<46:28,  1.54it/s] 
1|2179|Loss: 1.034744381904602:  34%|███▎      | 2179/6470 [27:55<46:13,  1.55it/s]
1|2180|Loss: 1.2146093845367432:  34%|███▎      | 2179/6470 [27:55<46:13,  1.55it/s]
1|2180|Loss: 1.2146093845367432:  34%|███▎      | 2180/6470 [27:56<46:29,  1.54it/s]
1|2181|Loss: 0.9431340098381042:  34%|███▎      | 2180/6470 [27:56<46:29,  1.54it/s]
1|2181|Loss: 0.9431340098381042:  34%|███▎      | 2181/6470 [27:56<48:08,  1.49it/s]
1|2182|Loss: 0.8040809035301208:  34%|███▎      | 2181/6470 [27:57<48:08,  1.49it/s]
1|2182|Loss: 0.8040809035301208:  34%|███▎      | 2182/6470 [27:57<49:44,  1.44it/s]
1|2183|Loss: 0.8436950445175171:  34%|███▎      | 2182/6470 [27:57<49:44,  1.44it/s]
1|2183|Loss: 0.8436950445175171:  34%|███▎      | 2183/6470 [27:58<48:14,  1.48it/s]
1|2184|Loss: 1.0615462064743042:  34%|███▎      | 2183/6470 [27:58<48:14,  1.48it/s]
1|2184|Loss: 1.0615462064743042:  34%|███▍      | 2184/6470 [27:58<47:01,  1.52it/s]
1|2185|Loss: 0.8887012004852295:  34%|███▍      | 2184/6470 [27:59<47:01,  1.52it/s]
1|2185|Loss: 0.8887012004852295:  34%|███▍      | 2185/6470 [27:59<48:19,  1.48it/s]
1|2186|Loss: 0.8925355672836304:  34%|███▍      | 2185/6470 [27:59<48:19,  1.48it/s]
1|2186|Loss: 0.8925355672836304:  34%|███▍      | 2186/6470 [28:00<49:54,  1.43it/s]
1|2187|Loss: 0.7866711020469666:  34%|███▍      | 2186/6470 [28:00<49:54,  1.43it/s]
1|2187|Loss: 0.7866711020469666:  34%|███▍      | 2187/6470 [28:01<50:48,  1.41it/s]
1|2188|Loss: 0.8845242857933044:  34%|███▍      | 2187/6470 [28:01<50:48,  1.41it/s]
1|2188|Loss: 0.8845242857933044:  34%|███▍      | 2188/6470 [28:01<50:40,  1.41it/s]
1|2189|Loss: 0.9316627383232117:  34%|███▍      | 2188/6470 [28:02<50:40,  1.41it/s]
1|2189|Loss: 0.9316627383232117:  34%|███▍      | 2189/6470 [28:02<49:14,  1.45it/s]
1|2190|Loss: 0.9396742582321167:  34%|███▍      | 2189/6470 [28:02<49:14,  1.45it/s]
1|2190|Loss: 0.9396742582321167:  34%|███▍      | 2190/6470 [28:02<44:48,  1.59it/s]
1|2191|Loss: 1.015643835067749:  34%|███▍      | 2190/6470 [28:03<44:48,  1.59it/s] 
1|2191|Loss: 1.015643835067749:  34%|███▍      | 2191/6470 [28:03<45:54,  1.55it/s]
1|2192|Loss: 0.8608449101448059:  34%|███▍      | 2191/6470 [28:03<45:54,  1.55it/s]
1|2192|Loss: 0.8608449101448059:  34%|███▍      | 2192/6470 [28:04<46:02,  1.55it/s]
1|2193|Loss: 0.9152814745903015:  34%|███▍      | 2192/6470 [28:04<46:02,  1.55it/s]
1|2193|Loss: 0.9152814745903015:  34%|███▍      | 2193/6470 [28:05<49:59,  1.43it/s]
1|2194|Loss: 0.9236084222793579:  34%|███▍      | 2193/6470 [28:05<49:59,  1.43it/s]
1|2194|Loss: 0.9236084222793579:  34%|███▍      | 2194/6470 [28:05<51:23,  1.39it/s]
1|2195|Loss: 0.8871709704399109:  34%|███▍      | 2194/6470 [28:06<51:23,  1.39it/s]
1|2195|Loss: 0.8871709704399109:  34%|███▍      | 2195/6470 [28:06<50:05,  1.42it/s]
1|2196|Loss: 0.8708276152610779:  34%|███▍      | 2195/6470 [28:06<50:05,  1.42it/s]
1|2196|Loss: 0.8708276152610779:  34%|███▍      | 2196/6470 [28:07<47:36,  1.50it/s]
1|2197|Loss: 0.8035515546798706:  34%|███▍      | 2196/6470 [28:07<47:36,  1.50it/s]
1|2197|Loss: 0.8035515546798706:  34%|███▍      | 2197/6470 [28:07<49:59,  1.42it/s]
1|2198|Loss: 0.9435935020446777:  34%|███▍      | 2197/6470 [28:08<49:59,  1.42it/s]
1|2198|Loss: 0.9435935020446777:  34%|███▍      | 2198/6470 [28:08<50:57,  1.40it/s]
1|2199|Loss: 0.8159933686256409:  34%|███▍      | 2198/6470 [28:08<50:57,  1.40it/s]
1|2199|Loss: 0.8159933686256409:  34%|███▍      | 2199/6470 [28:09<49:32,  1.44it/s]
1|2200|Loss: 0.8901349902153015:  34%|███▍      | 2199/6470 [28:09<49:32,  1.44it/s]
1|2200|Loss: 0.8901349902153015:  34%|███▍      | 2200/6470 [28:09<49:22,  1.44it/s]
1|2201|Loss: 0.884562611579895:  34%|███▍      | 2200/6470 [28:10<49:22,  1.44it/s] 
1|2201|Loss: 0.884562611579895:  34%|███▍      | 2201/6470 [28:10<45:46,  1.55it/s]
1|2202|Loss: 0.8465414047241211:  34%|███▍      | 2201/6470 [28:10<45:46,  1.55it/s]
1|2202|Loss: 0.8465414047241211:  34%|███▍      | 2202/6470 [28:11<44:43,  1.59it/s]
1|2203|Loss: 1.0230638980865479:  34%|███▍      | 2202/6470 [28:11<44:43,  1.59it/s]
1|2203|Loss: 1.0230638980865479:  34%|███▍      | 2203/6470 [28:11<45:58,  1.55it/s]
1|2204|Loss: 0.8569469451904297:  34%|███▍      | 2203/6470 [28:12<45:58,  1.55it/s]
1|2204|Loss: 0.8569469451904297:  34%|███▍      | 2204/6470 [28:12<47:42,  1.49it/s]
1|2205|Loss: 0.9006115794181824:  34%|███▍      | 2204/6470 [28:12<47:42,  1.49it/s]
1|2205|Loss: 0.9006115794181824:  34%|███▍      | 2205/6470 [28:13<47:15,  1.50it/s]
1|2206|Loss: 0.7670309543609619:  34%|███▍      | 2205/6470 [28:13<47:15,  1.50it/s]
1|2206|Loss: 0.7670309543609619:  34%|███▍      | 2206/6470 [28:13<48:35,  1.46it/s]
1|2207|Loss: 0.9481944441795349:  34%|███▍      | 2206/6470 [28:14<48:35,  1.46it/s]
1|2207|Loss: 0.9481944441795349:  34%|███▍      | 2207/6470 [28:14<48:17,  1.47it/s]
1|2208|Loss: 1.0141489505767822:  34%|███▍      | 2207/6470 [28:14<48:17,  1.47it/s]
1|2208|Loss: 1.0141489505767822:  34%|███▍      | 2208/6470 [28:15<47:07,  1.51it/s]
1|2209|Loss: 1.0448296070098877:  34%|███▍      | 2208/6470 [28:15<47:07,  1.51it/s]
1|2209|Loss: 1.0448296070098877:  34%|███▍      | 2209/6470 [28:15<44:23,  1.60it/s]
1|2210|Loss: 1.0410035848617554:  34%|███▍      | 2209/6470 [28:16<44:23,  1.60it/s]
1|2210|Loss: 1.0410035848617554:  34%|███▍      | 2210/6470 [28:16<46:47,  1.52it/s]
1|2211|Loss: 0.8446957468986511:  34%|███▍      | 2210/6470 [28:16<46:47,  1.52it/s]
1|2211|Loss: 0.8446957468986511:  34%|███▍      | 2211/6470 [28:16<45:13,  1.57it/s]
1|2212|Loss: 1.0127907991409302:  34%|███▍      | 2211/6470 [28:17<45:13,  1.57it/s]
1|2212|Loss: 1.0127907991409302:  34%|███▍      | 2212/6470 [28:17<46:34,  1.52it/s]
1|2213|Loss: 0.905706524848938:  34%|███▍      | 2212/6470 [28:18<46:34,  1.52it/s] 
1|2213|Loss: 0.905706524848938:  34%|███▍      | 2213/6470 [28:18<50:07,  1.42it/s]
1|2214|Loss: 0.986335039138794:  34%|███▍      | 2213/6470 [28:18<50:07,  1.42it/s]
1|2214|Loss: 0.986335039138794:  34%|███▍      | 2214/6470 [28:19<50:40,  1.40it/s]
1|2215|Loss: 0.9115707278251648:  34%|███▍      | 2214/6470 [28:19<50:40,  1.40it/s]
1|2215|Loss: 0.9115707278251648:  34%|███▍      | 2215/6470 [28:20<52:32,  1.35it/s]
1|2216|Loss: 0.8455702066421509:  34%|███▍      | 2215/6470 [28:20<52:32,  1.35it/s]
1|2216|Loss: 0.8455702066421509:  34%|███▍      | 2216/6470 [28:20<52:52,  1.34it/s]
1|2217|Loss: 0.8839586973190308:  34%|███▍      | 2216/6470 [28:21<52:52,  1.34it/s]
1|2217|Loss: 0.8839586973190308:  34%|███▍      | 2217/6470 [28:21<52:13,  1.36it/s]
1|2218|Loss: 1.0343854427337646:  34%|███▍      | 2217/6470 [28:21<52:13,  1.36it/s]
1|2218|Loss: 1.0343854427337646:  34%|███▍      | 2218/6470 [28:22<51:42,  1.37it/s]
1|2219|Loss: 0.9442387223243713:  34%|███▍      | 2218/6470 [28:22<51:42,  1.37it/s]
1|2219|Loss: 0.9442387223243713:  34%|███▍      | 2219/6470 [28:22<52:00,  1.36it/s]
1|2220|Loss: 0.9038017392158508:  34%|███▍      | 2219/6470 [28:23<52:00,  1.36it/s]
1|2220|Loss: 0.9038017392158508:  34%|███▍      | 2220/6470 [28:23<52:10,  1.36it/s]
1|2221|Loss: 0.9647579193115234:  34%|███▍      | 2220/6470 [28:24<52:10,  1.36it/s]
1|2221|Loss: 0.9647579193115234:  34%|███▍      | 2221/6470 [28:24<49:25,  1.43it/s]
1|2222|Loss: 0.8949912786483765:  34%|███▍      | 2221/6470 [28:24<49:25,  1.43it/s]
1|2222|Loss: 0.8949912786483765:  34%|███▍      | 2222/6470 [28:24<47:57,  1.48it/s]
1|2223|Loss: 0.9337754249572754:  34%|███▍      | 2222/6470 [28:25<47:57,  1.48it/s]
1|2223|Loss: 0.9337754249572754:  34%|███▍      | 2223/6470 [28:25<50:23,  1.40it/s]
1|2224|Loss: 0.7994322180747986:  34%|███▍      | 2223/6470 [28:26<50:23,  1.40it/s]
1|2224|Loss: 0.7994322180747986:  34%|███▍      | 2224/6470 [28:26<52:22,  1.35it/s]
1|2225|Loss: 0.9185943603515625:  34%|███▍      | 2224/6470 [28:26<52:22,  1.35it/s]
1|2225|Loss: 0.9185943603515625:  34%|███▍      | 2225/6470 [28:27<49:39,  1.42it/s]
1|2226|Loss: 0.9888765215873718:  34%|███▍      | 2225/6470 [28:27<49:39,  1.42it/s]
1|2226|Loss: 0.9888765215873718:  34%|███▍      | 2226/6470 [28:27<47:01,  1.50it/s]
1|2227|Loss: 1.008162021636963:  34%|███▍      | 2226/6470 [28:28<47:01,  1.50it/s] 
1|2227|Loss: 1.008162021636963:  34%|███▍      | 2227/6470 [28:28<44:46,  1.58it/s]
1|2228|Loss: 0.9063703417778015:  34%|███▍      | 2227/6470 [28:28<44:46,  1.58it/s]
1|2228|Loss: 0.9063703417778015:  34%|███▍      | 2228/6470 [28:29<50:40,  1.40it/s]
1|2229|Loss: 1.0126268863677979:  34%|███▍      | 2228/6470 [28:29<50:40,  1.40it/s]
1|2229|Loss: 1.0126268863677979:  34%|███▍      | 2229/6470 [28:30<52:27,  1.35it/s]
1|2230|Loss: 0.9139047861099243:  34%|███▍      | 2229/6470 [28:30<52:27,  1.35it/s]
1|2230|Loss: 0.9139047861099243:  34%|███▍      | 2230/6470 [28:30<48:46,  1.45it/s]
1|2231|Loss: 0.9591308236122131:  34%|███▍      | 2230/6470 [28:30<48:46,  1.45it/s]
1|2231|Loss: 0.9591308236122131:  34%|███▍      | 2231/6470 [28:31<49:07,  1.44it/s]
1|2232|Loss: 0.7247664928436279:  34%|███▍      | 2231/6470 [28:31<49:07,  1.44it/s]
1|2232|Loss: 0.7247664928436279:  34%|███▍      | 2232/6470 [28:32<50:23,  1.40it/s]
1|2233|Loss: 0.9626868963241577:  34%|███▍      | 2232/6470 [28:32<50:23,  1.40it/s]
1|2233|Loss: 0.9626868963241577:  35%|███▍      | 2233/6470 [28:32<49:06,  1.44it/s]
1|2234|Loss: 1.032724380493164:  35%|███▍      | 2233/6470 [28:33<49:06,  1.44it/s] 
1|2234|Loss: 1.032724380493164:  35%|███▍      | 2234/6470 [28:33<51:44,  1.36it/s]
1|2235|Loss: 1.0014526844024658:  35%|███▍      | 2234/6470 [28:33<51:44,  1.36it/s]
1|2235|Loss: 1.0014526844024658:  35%|███▍      | 2235/6470 [28:34<53:45,  1.31it/s]
1|2236|Loss: 1.0038126707077026:  35%|███▍      | 2235/6470 [28:34<53:45,  1.31it/s]
1|2236|Loss: 1.0038126707077026:  35%|███▍      | 2236/6470 [28:35<51:59,  1.36it/s]
1|2237|Loss: 0.8280705809593201:  35%|███▍      | 2236/6470 [28:35<51:59,  1.36it/s]
1|2237|Loss: 0.8280705809593201:  35%|███▍      | 2237/6470 [28:35<46:41,  1.51it/s]
1|2238|Loss: 0.9844415783882141:  35%|███▍      | 2237/6470 [28:35<46:41,  1.51it/s]
1|2238|Loss: 0.9844415783882141:  35%|███▍      | 2238/6470 [28:36<43:22,  1.63it/s]
1|2239|Loss: 0.9293302893638611:  35%|███▍      | 2238/6470 [28:36<43:22,  1.63it/s]
1|2239|Loss: 0.9293302893638611:  35%|███▍      | 2239/6470 [28:36<44:33,  1.58it/s]
1|2240|Loss: 0.9202078580856323:  35%|███▍      | 2239/6470 [28:37<44:33,  1.58it/s]
1|2240|Loss: 0.9202078580856323:  35%|███▍      | 2240/6470 [28:37<45:09,  1.56it/s]
1|2241|Loss: 0.8676211833953857:  35%|███▍      | 2240/6470 [28:37<45:09,  1.56it/s]
1|2241|Loss: 0.8676211833953857:  35%|███▍      | 2241/6470 [28:38<47:38,  1.48it/s]
1|2242|Loss: 0.9702730178833008:  35%|███▍      | 2241/6470 [28:38<47:38,  1.48it/s]
1|2242|Loss: 0.9702730178833008:  35%|███▍      | 2242/6470 [28:38<46:35,  1.51it/s]
1|2243|Loss: 0.9825812578201294:  35%|███▍      | 2242/6470 [28:39<46:35,  1.51it/s]
1|2243|Loss: 0.9825812578201294:  35%|███▍      | 2243/6470 [28:39<45:28,  1.55it/s]
1|2244|Loss: 0.9544601440429688:  35%|███▍      | 2243/6470 [28:39<45:28,  1.55it/s]
1|2244|Loss: 0.9544601440429688:  35%|███▍      | 2244/6470 [28:40<49:46,  1.42it/s]
1|2245|Loss: 0.9876524209976196:  35%|███▍      | 2244/6470 [28:40<49:46,  1.42it/s]
1|2245|Loss: 0.9876524209976196:  35%|███▍      | 2245/6470 [28:40<49:34,  1.42it/s]
1|2246|Loss: 0.8645893335342407:  35%|███▍      | 2245/6470 [28:41<49:34,  1.42it/s]
1|2246|Loss: 0.8645893335342407:  35%|███▍      | 2246/6470 [28:41<52:28,  1.34it/s]
1|2247|Loss: 0.931287407875061:  35%|███▍      | 2246/6470 [28:42<52:28,  1.34it/s] 
1|2247|Loss: 0.931287407875061:  35%|███▍      | 2247/6470 [28:42<50:35,  1.39it/s]
1|2248|Loss: 0.771706759929657:  35%|███▍      | 2247/6470 [28:42<50:35,  1.39it/s]
1|2248|Loss: 0.771706759929657:  35%|███▍      | 2248/6470 [28:43<50:11,  1.40it/s]
1|2249|Loss: 0.9208888411521912:  35%|███▍      | 2248/6470 [28:43<50:11,  1.40it/s]
1|2249|Loss: 0.9208888411521912:  35%|███▍      | 2249/6470 [28:43<50:35,  1.39it/s]
1|2250|Loss: 1.0912281274795532:  35%|███▍      | 2249/6470 [28:44<50:35,  1.39it/s]
1|2250|Loss: 1.0912281274795532:  35%|███▍      | 2250/6470 [28:44<47:27,  1.48it/s]
1|2251|Loss: 0.7281948328018188:  35%|███▍      | 2250/6470 [28:44<47:27,  1.48it/s]
1|2251|Loss: 0.7281948328018188:  35%|███▍      | 2251/6470 [28:44<45:49,  1.53it/s]
1|2252|Loss: 1.009874939918518:  35%|███▍      | 2251/6470 [28:45<45:49,  1.53it/s] 
1|2252|Loss: 1.009874939918518:  35%|███▍      | 2252/6470 [28:45<48:41,  1.44it/s]
1|2253|Loss: 0.9586388468742371:  35%|███▍      | 2252/6470 [28:46<48:41,  1.44it/s]
1|2253|Loss: 0.9586388468742371:  35%|███▍      | 2253/6470 [28:46<50:09,  1.40it/s]
1|2254|Loss: 0.897452175617218:  35%|███▍      | 2253/6470 [28:46<50:09,  1.40it/s] 
1|2254|Loss: 0.897452175617218:  35%|███▍      | 2254/6470 [28:47<50:21,  1.40it/s]
1|2255|Loss: 0.8239172101020813:  35%|███▍      | 2254/6470 [28:47<50:21,  1.40it/s]
1|2255|Loss: 0.8239172101020813:  35%|███▍      | 2255/6470 [28:48<50:39,  1.39it/s]
1|2256|Loss: 0.9194836616516113:  35%|███▍      | 2255/6470 [28:48<50:39,  1.39it/s]
1|2256|Loss: 0.9194836616516113:  35%|███▍      | 2256/6470 [28:48<47:11,  1.49it/s]
1|2257|Loss: 0.8978039622306824:  35%|███▍      | 2256/6470 [28:48<47:11,  1.49it/s]
1|2257|Loss: 0.8978039622306824:  35%|███▍      | 2257/6470 [28:49<44:56,  1.56it/s]
1|2258|Loss: 0.891302227973938:  35%|███▍      | 2257/6470 [28:49<44:56,  1.56it/s] 
1|2258|Loss: 0.891302227973938:  35%|███▍      | 2258/6470 [28:49<41:34,  1.69it/s]
1|2259|Loss: 0.9034397006034851:  35%|███▍      | 2258/6470 [28:49<41:34,  1.69it/s]
1|2259|Loss: 0.9034397006034851:  35%|███▍      | 2259/6470 [28:50<41:34,  1.69it/s]
1|2260|Loss: 0.9299790263175964:  35%|███▍      | 2259/6470 [28:50<41:34,  1.69it/s]
1|2260|Loss: 0.9299790263175964:  35%|███▍      | 2260/6470 [28:50<43:41,  1.61it/s]
1|2261|Loss: 0.8950006365776062:  35%|███▍      | 2260/6470 [28:51<43:41,  1.61it/s]
1|2261|Loss: 0.8950006365776062:  35%|███▍      | 2261/6470 [28:51<45:32,  1.54it/s]
1|2262|Loss: 1.0285969972610474:  35%|███▍      | 2261/6470 [28:52<45:32,  1.54it/s]
1|2262|Loss: 1.0285969972610474:  35%|███▍      | 2262/6470 [28:52<48:06,  1.46it/s]
1|2263|Loss: 0.9169690608978271:  35%|███▍      | 2262/6470 [28:52<48:06,  1.46it/s]
1|2263|Loss: 0.9169690608978271:  35%|███▍      | 2263/6470 [28:52<45:40,  1.54it/s]
1|2264|Loss: 0.8005293011665344:  35%|███▍      | 2263/6470 [28:53<45:40,  1.54it/s]
1|2264|Loss: 0.8005293011665344:  35%|███▍      | 2264/6470 [28:53<48:08,  1.46it/s]
1|2265|Loss: 0.8166831731796265:  35%|███▍      | 2264/6470 [28:54<48:08,  1.46it/s]
1|2265|Loss: 0.8166831731796265:  35%|███▌      | 2265/6470 [28:54<52:03,  1.35it/s]
1|2266|Loss: 0.8683290481567383:  35%|███▌      | 2265/6470 [28:55<52:03,  1.35it/s]
1|2266|Loss: 0.8683290481567383:  35%|███▌      | 2266/6470 [28:55<50:49,  1.38it/s]
1|2267|Loss: 1.0912944078445435:  35%|███▌      | 2266/6470 [28:55<50:49,  1.38it/s]
1|2267|Loss: 1.0912944078445435:  35%|███▌      | 2267/6470 [28:55<46:46,  1.50it/s]
1|2268|Loss: 0.8406078219413757:  35%|███▌      | 2267/6470 [28:56<46:46,  1.50it/s]
1|2268|Loss: 0.8406078219413757:  35%|███▌      | 2268/6470 [28:56<45:38,  1.53it/s]
1|2269|Loss: 1.0233802795410156:  35%|███▌      | 2268/6470 [28:56<45:38,  1.53it/s]
1|2269|Loss: 1.0233802795410156:  35%|███▌      | 2269/6470 [28:57<48:08,  1.45it/s]
1|2270|Loss: 0.9668200016021729:  35%|███▌      | 2269/6470 [28:57<48:08,  1.45it/s]
1|2270|Loss: 0.9668200016021729:  35%|███▌      | 2270/6470 [28:57<46:44,  1.50it/s]
1|2271|Loss: 0.8037666082382202:  35%|███▌      | 2270/6470 [28:58<46:44,  1.50it/s]
1|2271|Loss: 0.8037666082382202:  35%|███▌      | 2271/6470 [28:58<45:59,  1.52it/s]
1|2272|Loss: 0.8990451693534851:  35%|███▌      | 2271/6470 [28:58<45:59,  1.52it/s]
1|2272|Loss: 0.8990451693534851:  35%|███▌      | 2272/6470 [28:59<45:43,  1.53it/s]
1|2273|Loss: 0.9516034722328186:  35%|███▌      | 2272/6470 [28:59<45:43,  1.53it/s]
1|2273|Loss: 0.9516034722328186:  35%|███▌      | 2273/6470 [28:59<46:34,  1.50it/s]
1|2274|Loss: 1.006671667098999:  35%|███▌      | 2273/6470 [29:00<46:34,  1.50it/s] 
1|2274|Loss: 1.006671667098999:  35%|███▌      | 2274/6470 [29:00<47:21,  1.48it/s]
1|2275|Loss: 0.9820632338523865:  35%|███▌      | 2274/6470 [29:00<47:21,  1.48it/s]
1|2275|Loss: 0.9820632338523865:  35%|███▌      | 2275/6470 [29:01<50:08,  1.39it/s]
1|2276|Loss: 0.8987724184989929:  35%|███▌      | 2275/6470 [29:01<50:08,  1.39it/s]
1|2276|Loss: 0.8987724184989929:  35%|███▌      | 2276/6470 [29:02<52:27,  1.33it/s]
1|2277|Loss: 1.0286091566085815:  35%|███▌      | 2276/6470 [29:02<52:27,  1.33it/s]
1|2277|Loss: 1.0286091566085815:  35%|███▌      | 2277/6470 [29:02<54:31,  1.28it/s]
1|2278|Loss: 1.3720577955245972:  35%|███▌      | 2277/6470 [29:03<54:31,  1.28it/s]
1|2278|Loss: 1.3720577955245972:  35%|███▌      | 2278/6470 [29:03<52:41,  1.33it/s]
1|2279|Loss: 0.916997492313385:  35%|███▌      | 2278/6470 [29:04<52:41,  1.33it/s] 
1|2279|Loss: 0.916997492313385:  35%|███▌      | 2279/6470 [29:04<52:09,  1.34it/s]
1|2280|Loss: 0.9597383737564087:  35%|███▌      | 2279/6470 [29:04<52:09,  1.34it/s]
1|2280|Loss: 0.9597383737564087:  35%|███▌      | 2280/6470 [29:05<52:22,  1.33it/s]
1|2281|Loss: 0.8624305129051208:  35%|███▌      | 2280/6470 [29:05<52:22,  1.33it/s]
1|2281|Loss: 0.8624305129051208:  35%|███▌      | 2281/6470 [29:05<50:16,  1.39it/s]
1|2282|Loss: 0.9170419573783875:  35%|███▌      | 2281/6470 [29:06<50:16,  1.39it/s]
1|2282|Loss: 0.9170419573783875:  35%|███▌      | 2282/6470 [29:06<49:30,  1.41it/s]
1|2283|Loss: 0.8537890315055847:  35%|███▌      | 2282/6470 [29:06<49:30,  1.41it/s]
1|2283|Loss: 0.8537890315055847:  35%|███▌      | 2283/6470 [29:07<51:19,  1.36it/s]
1|2284|Loss: 0.8112585544586182:  35%|███▌      | 2283/6470 [29:07<51:19,  1.36it/s]
1|2284|Loss: 0.8112585544586182:  35%|███▌      | 2284/6470 [29:07<50:10,  1.39it/s]
1|2285|Loss: 0.8189785480499268:  35%|███▌      | 2284/6470 [29:08<50:10,  1.39it/s]
1|2285|Loss: 0.8189785480499268:  35%|███▌      | 2285/6470 [29:08<49:36,  1.41it/s]
1|2286|Loss: 0.9509873390197754:  35%|███▌      | 2285/6470 [29:09<49:36,  1.41it/s]
1|2286|Loss: 0.9509873390197754:  35%|███▌      | 2286/6470 [29:09<47:17,  1.47it/s]
1|2287|Loss: 0.9132453203201294:  35%|███▌      | 2286/6470 [29:09<47:17,  1.47it/s]
1|2287|Loss: 0.9132453203201294:  35%|███▌      | 2287/6470 [29:09<44:44,  1.56it/s]
1|2288|Loss: 1.0397237539291382:  35%|███▌      | 2287/6470 [29:10<44:44,  1.56it/s]
1|2288|Loss: 1.0397237539291382:  35%|███▌      | 2288/6470 [29:10<46:13,  1.51it/s]
1|2289|Loss: 1.0001790523529053:  35%|███▌      | 2288/6470 [29:10<46:13,  1.51it/s]
1|2289|Loss: 1.0001790523529053:  35%|███▌      | 2289/6470 [29:11<49:23,  1.41it/s]
1|2290|Loss: 0.9119827151298523:  35%|███▌      | 2289/6470 [29:11<49:23,  1.41it/s]
1|2290|Loss: 0.9119827151298523:  35%|███▌      | 2290/6470 [29:12<48:25,  1.44it/s]
1|2291|Loss: 0.8537153601646423:  35%|███▌      | 2290/6470 [29:12<48:25,  1.44it/s]
1|2291|Loss: 0.8537153601646423:  35%|███▌      | 2291/6470 [29:12<49:45,  1.40it/s]
1|2292|Loss: 0.923363447189331:  35%|███▌      | 2291/6470 [29:13<49:45,  1.40it/s] 
1|2292|Loss: 0.923363447189331:  35%|███▌      | 2292/6470 [29:13<47:54,  1.45it/s]
1|2293|Loss: 0.9861743450164795:  35%|███▌      | 2292/6470 [29:13<47:54,  1.45it/s]
1|2293|Loss: 0.9861743450164795:  35%|███▌      | 2293/6470 [29:14<48:15,  1.44it/s]
1|2294|Loss: 0.8488906621932983:  35%|███▌      | 2293/6470 [29:14<48:15,  1.44it/s]
1|2294|Loss: 0.8488906621932983:  35%|███▌      | 2294/6470 [29:14<47:35,  1.46it/s]
1|2295|Loss: 1.0175877809524536:  35%|███▌      | 2294/6470 [29:15<47:35,  1.46it/s]
1|2295|Loss: 1.0175877809524536:  35%|███▌      | 2295/6470 [29:15<46:22,  1.50it/s]
1|2296|Loss: 0.9405497312545776:  35%|███▌      | 2295/6470 [29:15<46:22,  1.50it/s]
1|2296|Loss: 0.9405497312545776:  35%|███▌      | 2296/6470 [29:15<44:49,  1.55it/s]
1|2297|Loss: 0.9738472104072571:  35%|███▌      | 2296/6470 [29:16<44:49,  1.55it/s]
1|2297|Loss: 0.9738472104072571:  36%|███▌      | 2297/6470 [29:16<46:04,  1.51it/s]
1|2298|Loss: 0.8915718793869019:  36%|███▌      | 2297/6470 [29:17<46:04,  1.51it/s]
1|2298|Loss: 0.8915718793869019:  36%|███▌      | 2298/6470 [29:17<50:09,  1.39it/s]
1|2299|Loss: 0.7647697925567627:  36%|███▌      | 2298/6470 [29:18<50:09,  1.39it/s]
1|2299|Loss: 0.7647697925567627:  36%|███▌      | 2299/6470 [29:18<53:35,  1.30it/s]
1|2300|Loss: 0.9314966201782227:  36%|███▌      | 2299/6470 [29:18<53:35,  1.30it/s]
1|2300|Loss: 0.9314966201782227:  36%|███▌      | 2300/6470 [29:19<54:41,  1.27it/s]
1|2301|Loss: 0.8272296190261841:  36%|███▌      | 2300/6470 [29:19<54:41,  1.27it/s]
1|2301|Loss: 0.8272296190261841:  36%|███▌      | 2301/6470 [29:20<54:09,  1.28it/s]
1|2302|Loss: 1.0333229303359985:  36%|███▌      | 2301/6470 [29:20<54:09,  1.28it/s]
1|2302|Loss: 1.0333229303359985:  36%|███▌      | 2302/6470 [29:20<50:20,  1.38it/s]
1|2303|Loss: 1.3729931116104126:  36%|███▌      | 2302/6470 [29:20<50:20,  1.38it/s]
1|2303|Loss: 1.3729931116104126:  36%|███▌      | 2303/6470 [29:21<46:19,  1.50it/s]
1|2304|Loss: 0.8192963004112244:  36%|███▌      | 2303/6470 [29:21<46:19,  1.50it/s]
1|2304|Loss: 0.8192963004112244:  36%|███▌      | 2304/6470 [29:21<46:49,  1.48it/s]
1|2305|Loss: 1.0027509927749634:  36%|███▌      | 2304/6470 [29:22<46:49,  1.48it/s]
1|2305|Loss: 1.0027509927749634:  36%|███▌      | 2305/6470 [29:22<45:32,  1.52it/s]
1|2306|Loss: 0.9038328528404236:  36%|███▌      | 2305/6470 [29:22<45:32,  1.52it/s]
1|2306|Loss: 0.9038328528404236:  36%|███▌      | 2306/6470 [29:23<44:58,  1.54it/s]
1|2307|Loss: 1.0714287757873535:  36%|███▌      | 2306/6470 [29:23<44:58,  1.54it/s]
1|2307|Loss: 1.0714287757873535:  36%|███▌      | 2307/6470 [29:23<48:08,  1.44it/s]
1|2308|Loss: 1.0489832162857056:  36%|███▌      | 2307/6470 [29:24<48:08,  1.44it/s]
1|2308|Loss: 1.0489832162857056:  36%|███▌      | 2308/6470 [29:24<47:30,  1.46it/s]
1|2309|Loss: 1.0058382749557495:  36%|███▌      | 2308/6470 [29:24<47:30,  1.46it/s]
1|2309|Loss: 1.0058382749557495:  36%|███▌      | 2309/6470 [29:25<46:01,  1.51it/s]
1|2310|Loss: 0.9201390743255615:  36%|███▌      | 2309/6470 [29:25<46:01,  1.51it/s]
1|2310|Loss: 0.9201390743255615:  36%|███▌      | 2310/6470 [29:25<44:24,  1.56it/s]
1|2311|Loss: 0.9863746166229248:  36%|███▌      | 2310/6470 [29:26<44:24,  1.56it/s]
1|2311|Loss: 0.9863746166229248:  36%|███▌      | 2311/6470 [29:26<45:41,  1.52it/s]
1|2312|Loss: 0.8975533843040466:  36%|███▌      | 2311/6470 [29:26<45:41,  1.52it/s]
1|2312|Loss: 0.8975533843040466:  36%|███▌      | 2312/6470 [29:27<49:35,  1.40it/s]
1|2313|Loss: 1.0684773921966553:  36%|███▌      | 2312/6470 [29:27<49:35,  1.40it/s]
1|2313|Loss: 1.0684773921966553:  36%|███▌      | 2313/6470 [29:27<47:40,  1.45it/s]
1|2314|Loss: 0.9649009704589844:  36%|███▌      | 2313/6470 [29:28<47:40,  1.45it/s]
1|2314|Loss: 0.9649009704589844:  36%|███▌      | 2314/6470 [29:28<45:22,  1.53it/s]
1|2315|Loss: 0.9589064717292786:  36%|███▌      | 2314/6470 [29:28<45:22,  1.53it/s]
1|2315|Loss: 0.9589064717292786:  36%|███▌      | 2315/6470 [29:29<45:56,  1.51it/s]
1|2316|Loss: 0.8606052994728088:  36%|███▌      | 2315/6470 [29:29<45:56,  1.51it/s]
1|2316|Loss: 0.8606052994728088:  36%|███▌      | 2316/6470 [29:29<43:38,  1.59it/s]
1|2317|Loss: 0.7332234382629395:  36%|███▌      | 2316/6470 [29:30<43:38,  1.59it/s]
1|2317|Loss: 0.7332234382629395:  36%|███▌      | 2317/6470 [29:30<43:17,  1.60it/s]
1|2318|Loss: 0.9683647155761719:  36%|███▌      | 2317/6470 [29:30<43:17,  1.60it/s]
1|2318|Loss: 0.9683647155761719:  36%|███▌      | 2318/6470 [29:31<43:43,  1.58it/s]
1|2319|Loss: 1.0760421752929688:  36%|███▌      | 2318/6470 [29:31<43:43,  1.58it/s]
1|2319|Loss: 1.0760421752929688:  36%|███▌      | 2319/6470 [29:31<45:39,  1.52it/s]
1|2320|Loss: 0.911332905292511:  36%|███▌      | 2319/6470 [29:32<45:39,  1.52it/s] 
1|2320|Loss: 0.911332905292511:  36%|███▌      | 2320/6470 [29:32<46:45,  1.48it/s]
1|2321|Loss: 0.9559910297393799:  36%|███▌      | 2320/6470 [29:32<46:45,  1.48it/s]
1|2321|Loss: 0.9559910297393799:  36%|███▌      | 2321/6470 [29:33<47:16,  1.46it/s]
1|2322|Loss: 0.8342432379722595:  36%|███▌      | 2321/6470 [29:33<47:16,  1.46it/s]
1|2322|Loss: 0.8342432379722595:  36%|███▌      | 2322/6470 [29:33<48:18,  1.43it/s]
1|2323|Loss: 1.0262179374694824:  36%|███▌      | 2322/6470 [29:34<48:18,  1.43it/s]
1|2323|Loss: 1.0262179374694824:  36%|███▌      | 2323/6470 [29:34<49:00,  1.41it/s]
1|2324|Loss: 0.8928865194320679:  36%|███▌      | 2323/6470 [29:35<49:00,  1.41it/s]
1|2324|Loss: 0.8928865194320679:  36%|███▌      | 2324/6470 [29:35<49:31,  1.40it/s]
1|2325|Loss: 1.1284737586975098:  36%|███▌      | 2324/6470 [29:35<49:31,  1.40it/s]
1|2325|Loss: 1.1284737586975098:  36%|███▌      | 2325/6470 [29:36<49:18,  1.40it/s]
1|2326|Loss: 0.8175607919692993:  36%|███▌      | 2325/6470 [29:36<49:18,  1.40it/s]
1|2326|Loss: 0.8175607919692993:  36%|███▌      | 2326/6470 [29:36<51:57,  1.33it/s]
1|2327|Loss: 0.9926287531852722:  36%|███▌      | 2326/6470 [29:37<51:57,  1.33it/s]
1|2327|Loss: 0.9926287531852722:  36%|███▌      | 2327/6470 [29:37<54:19,  1.27it/s]
1|2328|Loss: 0.9352927803993225:  36%|███▌      | 2327/6470 [29:38<54:19,  1.27it/s]
1|2328|Loss: 0.9352927803993225:  36%|███▌      | 2328/6470 [29:38<53:52,  1.28it/s]
1|2329|Loss: 0.7620189785957336:  36%|███▌      | 2328/6470 [29:38<53:52,  1.28it/s]
1|2329|Loss: 0.7620189785957336:  36%|███▌      | 2329/6470 [29:39<56:37,  1.22it/s]
1|2330|Loss: 0.9392302632331848:  36%|███▌      | 2329/6470 [29:39<56:37,  1.22it/s]
1|2330|Loss: 0.9392302632331848:  36%|███▌      | 2330/6470 [29:40<57:51,  1.19it/s]
1|2331|Loss: 0.8280918598175049:  36%|███▌      | 2330/6470 [29:40<57:51,  1.19it/s]
1|2331|Loss: 0.8280918598175049:  36%|███▌      | 2331/6470 [29:40<53:44,  1.28it/s]
1|2332|Loss: 0.9032182693481445:  36%|███▌      | 2331/6470 [29:41<53:44,  1.28it/s]
1|2332|Loss: 0.9032182693481445:  36%|███▌      | 2332/6470 [29:41<51:34,  1.34it/s]
1|2333|Loss: 0.751659631729126:  36%|███▌      | 2332/6470 [29:42<51:34,  1.34it/s] 
1|2333|Loss: 0.751659631729126:  36%|███▌      | 2333/6470 [29:42<51:08,  1.35it/s]
1|2334|Loss: 0.9184220433235168:  36%|███▌      | 2333/6470 [29:42<51:08,  1.35it/s]
1|2334|Loss: 0.9184220433235168:  36%|███▌      | 2334/6470 [29:43<50:23,  1.37it/s]
1|2335|Loss: 0.9530108571052551:  36%|███▌      | 2334/6470 [29:43<50:23,  1.37it/s]
1|2335|Loss: 0.9530108571052551:  36%|███▌      | 2335/6470 [29:43<47:39,  1.45it/s]
1|2336|Loss: 0.906063437461853:  36%|███▌      | 2335/6470 [29:44<47:39,  1.45it/s] 
1|2336|Loss: 0.906063437461853:  36%|███▌      | 2336/6470 [29:44<50:30,  1.36it/s]
1|2337|Loss: 0.8838790655136108:  36%|███▌      | 2336/6470 [29:44<50:30,  1.36it/s]
1|2337|Loss: 0.8838790655136108:  36%|███▌      | 2337/6470 [29:45<49:50,  1.38it/s]
1|2338|Loss: 0.8819066286087036:  36%|███▌      | 2337/6470 [29:45<49:50,  1.38it/s]
1|2338|Loss: 0.8819066286087036:  36%|███▌      | 2338/6470 [29:45<51:22,  1.34it/s]
1|2339|Loss: 0.8607848286628723:  36%|███▌      | 2338/6470 [29:46<51:22,  1.34it/s]
1|2339|Loss: 0.8607848286628723:  36%|███▌      | 2339/6470 [29:46<52:31,  1.31it/s]
1|2340|Loss: 0.9506245255470276:  36%|███▌      | 2339/6470 [29:47<52:31,  1.31it/s]
1|2340|Loss: 0.9506245255470276:  36%|███▌      | 2340/6470 [29:47<49:49,  1.38it/s]
1|2341|Loss: 0.9595605134963989:  36%|███▌      | 2340/6470 [29:47<49:49,  1.38it/s]
1|2341|Loss: 0.9595605134963989:  36%|███▌      | 2341/6470 [29:48<48:48,  1.41it/s]
1|2342|Loss: 0.5811160802841187:  36%|███▌      | 2341/6470 [29:48<48:48,  1.41it/s]
1|2342|Loss: 0.5811160802841187:  36%|███▌      | 2342/6470 [29:49<54:32,  1.26it/s]
1|2343|Loss: 0.9679797291755676:  36%|███▌      | 2342/6470 [29:49<54:32,  1.26it/s]
1|2343|Loss: 0.9679797291755676:  36%|███▌      | 2343/6470 [29:49<51:11,  1.34it/s]
1|2344|Loss: 0.8996878266334534:  36%|███▌      | 2343/6470 [29:50<51:11,  1.34it/s]
1|2344|Loss: 0.8996878266334534:  36%|███▌      | 2344/6470 [29:50<48:30,  1.42it/s]
1|2345|Loss: 0.9148698449134827:  36%|███▌      | 2344/6470 [29:50<48:30,  1.42it/s]
1|2345|Loss: 0.9148698449134827:  36%|███▌      | 2345/6470 [29:50<46:49,  1.47it/s]
1|2346|Loss: 0.9592023491859436:  36%|███▌      | 2345/6470 [29:51<46:49,  1.47it/s]
1|2346|Loss: 0.9592023491859436:  36%|███▋      | 2346/6470 [29:51<49:19,  1.39it/s]
1|2347|Loss: 1.141066312789917:  36%|███▋      | 2346/6470 [29:52<49:19,  1.39it/s] 
1|2347|Loss: 1.141066312789917:  36%|███▋      | 2347/6470 [29:52<50:14,  1.37it/s]
1|2348|Loss: 0.8945297598838806:  36%|███▋      | 2347/6470 [29:52<50:14,  1.37it/s]
1|2348|Loss: 0.8945297598838806:  36%|███▋      | 2348/6470 [29:53<52:15,  1.31it/s]
1|2349|Loss: 0.8686624765396118:  36%|███▋      | 2348/6470 [29:53<52:15,  1.31it/s]
1|2349|Loss: 0.8686624765396118:  36%|███▋      | 2349/6470 [29:54<52:18,  1.31it/s]
1|2350|Loss: 0.8232395052909851:  36%|███▋      | 2349/6470 [29:54<52:18,  1.31it/s]
1|2350|Loss: 0.8232395052909851:  36%|███▋      | 2350/6470 [29:54<48:42,  1.41it/s]
1|2351|Loss: 1.0047849416732788:  36%|███▋      | 2350/6470 [29:55<48:42,  1.41it/s]
1|2351|Loss: 1.0047849416732788:  36%|███▋      | 2351/6470 [29:55<46:57,  1.46it/s]
1|2352|Loss: 0.896956741809845:  36%|███▋      | 2351/6470 [29:55<46:57,  1.46it/s] 
1|2352|Loss: 0.896956741809845:  36%|███▋      | 2352/6470 [29:55<44:05,  1.56it/s]
1|2353|Loss: 0.9850156903266907:  36%|███▋      | 2352/6470 [29:56<44:05,  1.56it/s]
1|2353|Loss: 0.9850156903266907:  36%|███▋      | 2353/6470 [29:56<42:31,  1.61it/s]
1|2354|Loss: 0.9308427572250366:  36%|███▋      | 2353/6470 [29:56<42:31,  1.61it/s]
1|2354|Loss: 0.9308427572250366:  36%|███▋      | 2354/6470 [29:56<40:55,  1.68it/s]
1|2355|Loss: 0.866835355758667:  36%|███▋      | 2354/6470 [29:57<40:55,  1.68it/s] 
1|2355|Loss: 0.866835355758667:  36%|███▋      | 2355/6470 [29:57<38:14,  1.79it/s]
1|2356|Loss: 0.9401305913925171:  36%|███▋      | 2355/6470 [29:57<38:14,  1.79it/s]
1|2356|Loss: 0.9401305913925171:  36%|███▋      | 2356/6470 [29:58<37:57,  1.81it/s]
1|2357|Loss: 0.8715104460716248:  36%|███▋      | 2356/6470 [29:58<37:57,  1.81it/s]
1|2357|Loss: 0.8715104460716248:  36%|███▋      | 2357/6470 [29:58<42:11,  1.62it/s]
1|2358|Loss: 0.8528884053230286:  36%|███▋      | 2357/6470 [29:59<42:11,  1.62it/s]
1|2358|Loss: 0.8528884053230286:  36%|███▋      | 2358/6470 [29:59<44:43,  1.53it/s]
1|2359|Loss: 0.867472767829895:  36%|███▋      | 2358/6470 [29:59<44:43,  1.53it/s] 
1|2359|Loss: 0.867472767829895:  36%|███▋      | 2359/6470 [30:00<44:52,  1.53it/s]
1|2360|Loss: 0.8288082480430603:  36%|███▋      | 2359/6470 [30:00<44:52,  1.53it/s]
1|2360|Loss: 0.8288082480430603:  36%|███▋      | 2360/6470 [30:00<47:42,  1.44it/s]
1|2361|Loss: 0.8438712954521179:  36%|███▋      | 2360/6470 [30:01<47:42,  1.44it/s]
1|2361|Loss: 0.8438712954521179:  36%|███▋      | 2361/6470 [30:01<48:45,  1.40it/s]
1|2362|Loss: 0.8973580598831177:  36%|███▋      | 2361/6470 [30:02<48:45,  1.40it/s]
1|2362|Loss: 0.8973580598831177:  37%|███▋      | 2362/6470 [30:02<49:33,  1.38it/s]
1|2363|Loss: 0.9824182987213135:  37%|███▋      | 2362/6470 [30:02<49:33,  1.38it/s]
1|2363|Loss: 0.9824182987213135:  37%|███▋      | 2363/6470 [30:03<50:43,  1.35it/s]
1|2364|Loss: 0.9316175580024719:  37%|███▋      | 2363/6470 [30:03<50:43,  1.35it/s]
1|2364|Loss: 0.9316175580024719:  37%|███▋      | 2364/6470 [30:03<50:20,  1.36it/s]
1|2365|Loss: 0.9325646758079529:  37%|███▋      | 2364/6470 [30:04<50:20,  1.36it/s]
1|2365|Loss: 0.9325646758079529:  37%|███▋      | 2365/6470 [30:04<50:05,  1.37it/s]
1|2366|Loss: 1.03870689868927:  37%|███▋      | 2365/6470 [30:05<50:05,  1.37it/s]  
1|2366|Loss: 1.03870689868927:  37%|███▋      | 2366/6470 [30:05<52:18,  1.31it/s]
1|2367|Loss: 0.837353527545929:  37%|███▋      | 2366/6470 [30:05<52:18,  1.31it/s]
1|2367|Loss: 0.837353527545929:  37%|███▋      | 2367/6470 [30:06<50:51,  1.34it/s]
1|2368|Loss: 0.8522782921791077:  37%|███▋      | 2367/6470 [30:06<50:51,  1.34it/s]
1|2368|Loss: 0.8522782921791077:  37%|███▋      | 2368/6470 [30:07<52:09,  1.31it/s]
1|2369|Loss: 0.9591068625450134:  37%|███▋      | 2368/6470 [30:07<52:09,  1.31it/s]
1|2369|Loss: 0.9591068625450134:  37%|███▋      | 2369/6470 [30:07<49:45,  1.37it/s]
1|2370|Loss: 1.136928677558899:  37%|███▋      | 2369/6470 [30:07<49:45,  1.37it/s] 
1|2370|Loss: 1.136928677558899:  37%|███▋      | 2370/6470 [30:08<45:31,  1.50it/s]
1|2371|Loss: 0.8900651931762695:  37%|███▋      | 2370/6470 [30:08<45:31,  1.50it/s]
1|2371|Loss: 0.8900651931762695:  37%|███▋      | 2371/6470 [30:08<43:57,  1.55it/s]
1|2372|Loss: 0.7875232696533203:  37%|███▋      | 2371/6470 [30:09<43:57,  1.55it/s]
1|2372|Loss: 0.7875232696533203:  37%|███▋      | 2372/6470 [30:09<42:37,  1.60it/s]
1|2373|Loss: 0.9194402694702148:  37%|███▋      | 2372/6470 [30:09<42:37,  1.60it/s]
1|2373|Loss: 0.9194402694702148:  37%|███▋      | 2373/6470 [30:09<42:17,  1.61it/s]
1|2374|Loss: 1.0086230039596558:  37%|███▋      | 2373/6470 [30:10<42:17,  1.61it/s]
1|2374|Loss: 1.0086230039596558:  37%|███▋      | 2374/6470 [30:10<42:02,  1.62it/s]
1|2375|Loss: 0.9066547155380249:  37%|███▋      | 2374/6470 [30:10<42:02,  1.62it/s]
1|2375|Loss: 0.9066547155380249:  37%|███▋      | 2375/6470 [30:11<41:19,  1.65it/s]
1|2376|Loss: 0.8124906420707703:  37%|███▋      | 2375/6470 [30:11<41:19,  1.65it/s]
1|2376|Loss: 0.8124906420707703:  37%|███▋      | 2376/6470 [30:11<41:30,  1.64it/s]
1|2377|Loss: 0.9005701541900635:  37%|███▋      | 2376/6470 [30:12<41:30,  1.64it/s]
1|2377|Loss: 0.9005701541900635:  37%|███▋      | 2377/6470 [30:12<41:02,  1.66it/s]
1|2378|Loss: 0.9423137903213501:  37%|███▋      | 2377/6470 [30:12<41:02,  1.66it/s]
1|2378|Loss: 0.9423137903213501:  37%|███▋      | 2378/6470 [30:12<41:11,  1.66it/s]
1|2379|Loss: 1.114622712135315:  37%|███▋      | 2378/6470 [30:13<41:11,  1.66it/s] 
1|2379|Loss: 1.114622712135315:  37%|███▋      | 2379/6470 [30:13<40:19,  1.69it/s]
1|2380|Loss: 0.9081757664680481:  37%|███▋      | 2379/6470 [30:13<40:19,  1.69it/s]
1|2380|Loss: 0.9081757664680481:  37%|███▋      | 2380/6470 [30:14<41:00,  1.66it/s]
1|2381|Loss: 1.064145565032959:  37%|███▋      | 2380/6470 [30:14<41:00,  1.66it/s] 
1|2381|Loss: 1.064145565032959:  37%|███▋      | 2381/6470 [30:14<41:33,  1.64it/s]
1|2382|Loss: 1.0430384874343872:  37%|███▋      | 2381/6470 [30:15<41:33,  1.64it/s]
1|2382|Loss: 1.0430384874343872:  37%|███▋      | 2382/6470 [30:15<44:01,  1.55it/s]
1|2383|Loss: 0.9134076237678528:  37%|███▋      | 2382/6470 [30:15<44:01,  1.55it/s]
1|2383|Loss: 0.9134076237678528:  37%|███▋      | 2383/6470 [30:16<46:09,  1.48it/s]
1|2384|Loss: 0.9720957279205322:  37%|███▋      | 2383/6470 [30:16<46:09,  1.48it/s]
1|2384|Loss: 0.9720957279205322:  37%|███▋      | 2384/6470 [30:16<45:36,  1.49it/s]
1|2385|Loss: 0.7641732096672058:  37%|███▋      | 2384/6470 [30:17<45:36,  1.49it/s]
1|2385|Loss: 0.7641732096672058:  37%|███▋      | 2385/6470 [30:17<46:17,  1.47it/s]
1|2386|Loss: 0.9515792727470398:  37%|███▋      | 2385/6470 [30:17<46:17,  1.47it/s]
1|2386|Loss: 0.9515792727470398:  37%|███▋      | 2386/6470 [30:18<44:46,  1.52it/s]
1|2387|Loss: 0.836276113986969:  37%|███▋      | 2386/6470 [30:18<44:46,  1.52it/s] 
1|2387|Loss: 0.836276113986969:  37%|███▋      | 2387/6470 [30:18<45:16,  1.50it/s]
1|2388|Loss: 1.0884250402450562:  37%|███▋      | 2387/6470 [30:19<45:16,  1.50it/s]
1|2388|Loss: 1.0884250402450562:  37%|███▋      | 2388/6470 [30:19<44:02,  1.54it/s]
1|2389|Loss: 0.773638129234314:  37%|███▋      | 2388/6470 [30:19<44:02,  1.54it/s] 
1|2389|Loss: 0.773638129234314:  37%|███▋      | 2389/6470 [30:19<39:09,  1.74it/s]
1|2390|Loss: 0.9024022817611694:  37%|███▋      | 2389/6470 [30:20<39:09,  1.74it/s]
1|2390|Loss: 0.9024022817611694:  37%|███▋      | 2390/6470 [30:20<43:14,  1.57it/s]
1|2391|Loss: 1.0831284523010254:  37%|███▋      | 2390/6470 [30:21<43:14,  1.57it/s]
1|2391|Loss: 1.0831284523010254:  37%|███▋      | 2391/6470 [30:21<41:26,  1.64it/s]
1|2392|Loss: 0.7784236669540405:  37%|███▋      | 2391/6470 [30:21<41:26,  1.64it/s]
1|2392|Loss: 0.7784236669540405:  37%|███▋      | 2392/6470 [30:21<42:39,  1.59it/s]
1|2393|Loss: 0.8745619058609009:  37%|███▋      | 2392/6470 [30:22<42:39,  1.59it/s]
1|2393|Loss: 0.8745619058609009:  37%|███▋      | 2393/6470 [30:22<41:20,  1.64it/s]
1|2394|Loss: 0.880723774433136:  37%|███▋      | 2393/6470 [30:22<41:20,  1.64it/s] 
1|2394|Loss: 0.880723774433136:  37%|███▋      | 2394/6470 [30:23<44:55,  1.51it/s]
1|2395|Loss: 0.8789014220237732:  37%|███▋      | 2394/6470 [30:23<44:55,  1.51it/s]
1|2395|Loss: 0.8789014220237732:  37%|███▋      | 2395/6470 [30:23<45:28,  1.49it/s]
1|2396|Loss: 0.9410341382026672:  37%|███▋      | 2395/6470 [30:24<45:28,  1.49it/s]
1|2396|Loss: 0.9410341382026672:  37%|███▋      | 2396/6470 [30:24<45:54,  1.48it/s]
1|2397|Loss: 0.9117127060890198:  37%|███▋      | 2396/6470 [30:25<45:54,  1.48it/s]
1|2397|Loss: 0.9117127060890198:  37%|███▋      | 2397/6470 [30:25<46:23,  1.46it/s]
1|2398|Loss: 0.8796617984771729:  37%|███▋      | 2397/6470 [30:25<46:23,  1.46it/s]
1|2398|Loss: 0.8796617984771729:  37%|███▋      | 2398/6470 [30:25<45:25,  1.49it/s]
1|2399|Loss: 0.8573959469795227:  37%|███▋      | 2398/6470 [30:26<45:25,  1.49it/s]
1|2399|Loss: 0.8573959469795227:  37%|███▋      | 2399/6470 [30:26<45:47,  1.48it/s]
1|2400|Loss: 0.80184406042099:  37%|███▋      | 2399/6470 [30:27<45:47,  1.48it/s]  
1|2400|Loss: 0.80184406042099:  37%|███▋      | 2400/6470 [30:27<45:02,  1.51it/s]
1|2401|Loss: 0.9651104211807251:  37%|███▋      | 2400/6470 [30:27<45:02,  1.51it/s]
1|2401|Loss: 0.9651104211807251:  37%|███▋      | 2401/6470 [30:27<44:25,  1.53it/s]
1|2402|Loss: 0.9818615913391113:  37%|███▋      | 2401/6470 [30:28<44:25,  1.53it/s]
1|2402|Loss: 0.9818615913391113:  37%|███▋      | 2402/6470 [30:28<43:05,  1.57it/s]
1|2403|Loss: 0.9719618558883667:  37%|███▋      | 2402/6470 [30:28<43:05,  1.57it/s]
1|2403|Loss: 0.9719618558883667:  37%|███▋      | 2403/6470 [30:29<45:35,  1.49it/s]
1|2404|Loss: 1.0028026103973389:  37%|███▋      | 2403/6470 [30:29<45:35,  1.49it/s]
1|2404|Loss: 1.0028026103973389:  37%|███▋      | 2404/6470 [30:30<46:36,  1.45it/s]
1|2405|Loss: 0.9515969753265381:  37%|███▋      | 2404/6470 [30:30<46:36,  1.45it/s]
1|2405|Loss: 0.9515969753265381:  37%|███▋      | 2405/6470 [30:30<50:27,  1.34it/s]
1|2406|Loss: 0.8849353790283203:  37%|███▋      | 2405/6470 [30:31<50:27,  1.34it/s]
1|2406|Loss: 0.8849353790283203:  37%|███▋      | 2406/6470 [30:31<49:34,  1.37it/s]
1|2407|Loss: 1.0298819541931152:  37%|███▋      | 2406/6470 [30:31<49:34,  1.37it/s]
1|2407|Loss: 1.0298819541931152:  37%|███▋      | 2407/6470 [30:32<48:43,  1.39it/s]
1|2408|Loss: 0.9852275848388672:  37%|███▋      | 2407/6470 [30:32<48:43,  1.39it/s]
1|2408|Loss: 0.9852275848388672:  37%|███▋      | 2408/6470 [30:32<47:03,  1.44it/s]
1|2409|Loss: 1.0725802183151245:  37%|███▋      | 2408/6470 [30:33<47:03,  1.44it/s]
1|2409|Loss: 1.0725802183151245:  37%|███▋      | 2409/6470 [30:33<48:31,  1.40it/s]
1|2410|Loss: 0.8938267230987549:  37%|███▋      | 2409/6470 [30:34<48:31,  1.40it/s]
1|2410|Loss: 0.8938267230987549:  37%|███▋      | 2410/6470 [30:34<49:25,  1.37it/s]
1|2411|Loss: 0.8559123277664185:  37%|███▋      | 2410/6470 [30:34<49:25,  1.37it/s]
1|2411|Loss: 0.8559123277664185:  37%|███▋      | 2411/6470 [30:35<46:58,  1.44it/s]
1|2412|Loss: 0.9546063542366028:  37%|███▋      | 2411/6470 [30:35<46:58,  1.44it/s]
1|2412|Loss: 0.9546063542366028:  37%|███▋      | 2412/6470 [30:35<50:02,  1.35it/s]
1|2413|Loss: 0.7485148906707764:  37%|███▋      | 2412/6470 [30:36<50:02,  1.35it/s]
1|2413|Loss: 0.7485148906707764:  37%|███▋      | 2413/6470 [30:36<54:21,  1.24it/s]
1|2414|Loss: 0.8817583918571472:  37%|███▋      | 2413/6470 [30:37<54:21,  1.24it/s]
1|2414|Loss: 0.8817583918571472:  37%|███▋      | 2414/6470 [30:37<56:13,  1.20it/s]
1|2415|Loss: 0.8794458508491516:  37%|███▋      | 2414/6470 [30:38<56:13,  1.20it/s]
1|2415|Loss: 0.8794458508491516:  37%|███▋      | 2415/6470 [30:38<53:06,  1.27it/s]
1|2416|Loss: 0.9077462553977966:  37%|███▋      | 2415/6470 [30:38<53:06,  1.27it/s]
1|2416|Loss: 0.9077462553977966:  37%|███▋      | 2416/6470 [30:39<51:33,  1.31it/s]
1|2417|Loss: 0.954021155834198:  37%|███▋      | 2416/6470 [30:39<51:33,  1.31it/s] 
1|2417|Loss: 0.954021155834198:  37%|███▋      | 2417/6470 [30:39<51:20,  1.32it/s]
1|2418|Loss: 0.8115090131759644:  37%|███▋      | 2417/6470 [30:40<51:20,  1.32it/s]
1|2418|Loss: 0.8115090131759644:  37%|███▋      | 2418/6470 [30:40<49:28,  1.36it/s]
1|2419|Loss: 0.9114105105400085:  37%|███▋      | 2418/6470 [30:40<49:28,  1.36it/s]
1|2419|Loss: 0.9114105105400085:  37%|███▋      | 2419/6470 [30:41<43:25,  1.55it/s]
1|2420|Loss: 0.8513572812080383:  37%|███▋      | 2419/6470 [30:41<43:25,  1.55it/s]
1|2420|Loss: 0.8513572812080383:  37%|███▋      | 2420/6470 [30:41<42:09,  1.60it/s]
1|2421|Loss: 0.833390474319458:  37%|███▋      | 2420/6470 [30:41<42:09,  1.60it/s] 
1|2421|Loss: 0.833390474319458:  37%|███▋      | 2421/6470 [30:42<43:50,  1.54it/s]
1|2422|Loss: 1.008026123046875:  37%|███▋      | 2421/6470 [30:42<43:50,  1.54it/s]
1|2422|Loss: 1.008026123046875:  37%|███▋      | 2422/6470 [30:43<45:03,  1.50it/s]
1|2423|Loss: 1.146299123764038:  37%|███▋      | 2422/6470 [30:43<45:03,  1.50it/s]
1|2423|Loss: 1.146299123764038:  37%|███▋      | 2423/6470 [30:43<47:23,  1.42it/s]
1|2424|Loss: 0.9348265528678894:  37%|███▋      | 2423/6470 [30:44<47:23,  1.42it/s]
1|2424|Loss: 0.9348265528678894:  37%|███▋      | 2424/6470 [30:44<49:51,  1.35it/s]
1|2425|Loss: 0.9598594903945923:  37%|███▋      | 2424/6470 [30:45<49:51,  1.35it/s]
1|2425|Loss: 0.9598594903945923:  37%|███▋      | 2425/6470 [30:45<51:40,  1.30it/s]
1|2426|Loss: 0.8024576306343079:  37%|███▋      | 2425/6470 [30:45<51:40,  1.30it/s]
1|2426|Loss: 0.8024576306343079:  37%|███▋      | 2426/6470 [30:46<49:35,  1.36it/s]
1|2427|Loss: 0.87166428565979:  37%|███▋      | 2426/6470 [30:46<49:35,  1.36it/s]  
1|2427|Loss: 0.87166428565979:  38%|███▊      | 2427/6470 [30:46<45:42,  1.47it/s]
1|2428|Loss: 0.9439324140548706:  38%|███▊      | 2427/6470 [30:47<45:42,  1.47it/s]
1|2428|Loss: 0.9439324140548706:  38%|███▊      | 2428/6470 [30:47<45:52,  1.47it/s]
1|2429|Loss: 1.0423893928527832:  38%|███▊      | 2428/6470 [30:47<45:52,  1.47it/s]
1|2429|Loss: 1.0423893928527832:  38%|███▊      | 2429/6470 [30:48<46:58,  1.43it/s]
1|2430|Loss: 0.8972530961036682:  38%|███▊      | 2429/6470 [30:48<46:58,  1.43it/s]
1|2430|Loss: 0.8972530961036682:  38%|███▊      | 2430/6470 [30:48<48:08,  1.40it/s]
1|2431|Loss: 0.9542222023010254:  38%|███▊      | 2430/6470 [30:49<48:08,  1.40it/s]
1|2431|Loss: 0.9542222023010254:  38%|███▊      | 2431/6470 [30:49<48:40,  1.38it/s]
1|2432|Loss: 1.0692036151885986:  38%|███▊      | 2431/6470 [30:49<48:40,  1.38it/s]
1|2432|Loss: 1.0692036151885986:  38%|███▊      | 2432/6470 [30:50<48:56,  1.38it/s]
1|2433|Loss: 0.7653900384902954:  38%|███▊      | 2432/6470 [30:50<48:56,  1.38it/s]
1|2433|Loss: 0.7653900384902954:  38%|███▊      | 2433/6470 [30:51<52:04,  1.29it/s]
1|2434|Loss: 0.7512733936309814:  38%|███▊      | 2433/6470 [30:51<52:04,  1.29it/s]
1|2434|Loss: 0.7512733936309814:  38%|███▊      | 2434/6470 [30:51<52:19,  1.29it/s]
1|2435|Loss: 0.9752208590507507:  38%|███▊      | 2434/6470 [30:52<52:19,  1.29it/s]
1|2435|Loss: 0.9752208590507507:  38%|███▊      | 2435/6470 [30:52<56:43,  1.19it/s]
1|2436|Loss: 0.9410101175308228:  38%|███▊      | 2435/6470 [30:53<56:43,  1.19it/s]
1|2436|Loss: 0.9410101175308228:  38%|███▊      | 2436/6470 [30:53<56:54,  1.18it/s]
1|2437|Loss: 1.060326337814331:  38%|███▊      | 2436/6470 [30:54<56:54,  1.18it/s] 
1|2437|Loss: 1.060326337814331:  38%|███▊      | 2437/6470 [30:54<53:41,  1.25it/s]
1|2438|Loss: 0.9909352660179138:  38%|███▊      | 2437/6470 [30:54<53:41,  1.25it/s]
1|2438|Loss: 0.9909352660179138:  38%|███▊      | 2438/6470 [30:55<51:23,  1.31it/s]
1|2439|Loss: 0.7890441417694092:  38%|███▊      | 2438/6470 [30:55<51:23,  1.31it/s]
1|2439|Loss: 0.7890441417694092:  38%|███▊      | 2439/6470 [30:56<53:32,  1.25it/s]
1|2440|Loss: 0.9671370983123779:  38%|███▊      | 2439/6470 [30:56<53:32,  1.25it/s]
1|2440|Loss: 0.9671370983123779:  38%|███▊      | 2440/6470 [30:56<52:56,  1.27it/s]
1|2441|Loss: 0.7985000610351562:  38%|███▊      | 2440/6470 [30:57<52:56,  1.27it/s]
1|2441|Loss: 0.7985000610351562:  38%|███▊      | 2441/6470 [30:57<51:51,  1.29it/s]
1|2442|Loss: 0.9333372116088867:  38%|███▊      | 2441/6470 [30:57<51:51,  1.29it/s]
1|2442|Loss: 0.9333372116088867:  38%|███▊      | 2442/6470 [30:58<48:27,  1.39it/s]
1|2443|Loss: 0.9868807196617126:  38%|███▊      | 2442/6470 [30:58<48:27,  1.39it/s]
1|2443|Loss: 0.9868807196617126:  38%|███▊      | 2443/6470 [30:58<50:05,  1.34it/s]
1|2444|Loss: 0.8922023773193359:  38%|███▊      | 2443/6470 [30:59<50:05,  1.34it/s]
1|2444|Loss: 0.8922023773193359:  38%|███▊      | 2444/6470 [30:59<50:17,  1.33it/s]
1|2445|Loss: 0.8642572164535522:  38%|███▊      | 2444/6470 [31:00<50:17,  1.33it/s]
1|2445|Loss: 0.8642572164535522:  38%|███▊      | 2445/6470 [31:00<45:25,  1.48it/s]
1|2446|Loss: 0.8946231603622437:  38%|███▊      | 2445/6470 [31:00<45:25,  1.48it/s]
1|2446|Loss: 0.8946231603622437:  38%|███▊      | 2446/6470 [31:00<42:54,  1.56it/s]
1|2447|Loss: 1.005621075630188:  38%|███▊      | 2446/6470 [31:01<42:54,  1.56it/s] 
1|2447|Loss: 1.005621075630188:  38%|███▊      | 2447/6470 [31:01<44:06,  1.52it/s]
1|2448|Loss: 0.7799608707427979:  38%|███▊      | 2447/6470 [31:01<44:06,  1.52it/s]
1|2448|Loss: 0.7799608707427979:  38%|███▊      | 2448/6470 [31:02<45:33,  1.47it/s]
1|2449|Loss: 0.8800366520881653:  38%|███▊      | 2448/6470 [31:02<45:33,  1.47it/s]
1|2449|Loss: 0.8800366520881653:  38%|███▊      | 2449/6470 [31:02<43:23,  1.54it/s]
1|2450|Loss: 0.762835681438446:  38%|███▊      | 2449/6470 [31:03<43:23,  1.54it/s] 
1|2450|Loss: 0.762835681438446:  38%|███▊      | 2450/6470 [31:03<45:14,  1.48it/s]
1|2451|Loss: 0.7990559339523315:  38%|███▊      | 2450/6470 [31:03<45:14,  1.48it/s]
1|2451|Loss: 0.7990559339523315:  38%|███▊      | 2451/6470 [31:04<46:54,  1.43it/s]
1|2452|Loss: 0.8766656517982483:  38%|███▊      | 2451/6470 [31:04<46:54,  1.43it/s]
1|2452|Loss: 0.8766656517982483:  38%|███▊      | 2452/6470 [31:05<47:39,  1.40it/s]
1|2453|Loss: 0.9365158677101135:  38%|███▊      | 2452/6470 [31:05<47:39,  1.40it/s]
1|2453|Loss: 0.9365158677101135:  38%|███▊      | 2453/6470 [31:05<49:18,  1.36it/s]
1|2454|Loss: 1.08448326587677:  38%|███▊      | 2453/6470 [31:06<49:18,  1.36it/s]  
1|2454|Loss: 1.08448326587677:  38%|███▊      | 2454/6470 [31:06<48:59,  1.37it/s]
1|2455|Loss: 0.8563827872276306:  38%|███▊      | 2454/6470 [31:06<48:59,  1.37it/s]
1|2455|Loss: 0.8563827872276306:  38%|███▊      | 2455/6470 [31:07<48:25,  1.38it/s]
1|2456|Loss: 0.6809148192405701:  38%|███▊      | 2455/6470 [31:07<48:25,  1.38it/s]
1|2456|Loss: 0.6809148192405701:  38%|███▊      | 2456/6470 [31:08<51:31,  1.30it/s]
1|2457|Loss: 0.8387320637702942:  38%|███▊      | 2456/6470 [31:08<51:31,  1.30it/s]
1|2457|Loss: 0.8387320637702942:  38%|███▊      | 2457/6470 [31:08<50:18,  1.33it/s]
1|2458|Loss: 0.8748311996459961:  38%|███▊      | 2457/6470 [31:09<50:18,  1.33it/s]
1|2458|Loss: 0.8748311996459961:  38%|███▊      | 2458/6470 [31:09<49:06,  1.36it/s]
1|2459|Loss: 0.90398770570755:  38%|███▊      | 2458/6470 [31:09<49:06,  1.36it/s]  
1|2459|Loss: 0.90398770570755:  38%|███▊      | 2459/6470 [31:10<47:26,  1.41it/s]
1|2460|Loss: 0.8502048254013062:  38%|███▊      | 2459/6470 [31:10<47:26,  1.41it/s]
1|2460|Loss: 0.8502048254013062:  38%|███▊      | 2460/6470 [31:11<49:55,  1.34it/s]
1|2461|Loss: 0.9947401285171509:  38%|███▊      | 2460/6470 [31:11<49:55,  1.34it/s]
1|2461|Loss: 0.9947401285171509:  38%|███▊      | 2461/6470 [31:11<48:35,  1.38it/s]
1|2462|Loss: 0.9148174524307251:  38%|███▊      | 2461/6470 [31:12<48:35,  1.38it/s]
1|2462|Loss: 0.9148174524307251:  38%|███▊      | 2462/6470 [31:12<47:36,  1.40it/s]
1|2463|Loss: 1.2243834733963013:  38%|███▊      | 2462/6470 [31:12<47:36,  1.40it/s]
1|2463|Loss: 1.2243834733963013:  38%|███▊      | 2463/6470 [31:13<50:55,  1.31it/s]
1|2464|Loss: 1.0592682361602783:  38%|███▊      | 2463/6470 [31:13<50:55,  1.31it/s]
1|2464|Loss: 1.0592682361602783:  38%|███▊      | 2464/6470 [31:13<49:47,  1.34it/s]
1|2465|Loss: 0.9425377249717712:  38%|███▊      | 2464/6470 [31:14<49:47,  1.34it/s]
1|2465|Loss: 0.9425377249717712:  38%|███▊      | 2465/6470 [31:14<48:45,  1.37it/s]
1|2466|Loss: 0.8973172903060913:  38%|███▊      | 2465/6470 [31:15<48:45,  1.37it/s]
1|2466|Loss: 0.8973172903060913:  38%|███▊      | 2466/6470 [31:15<48:08,  1.39it/s]
1|2467|Loss: 0.9661283493041992:  38%|███▊      | 2466/6470 [31:15<48:08,  1.39it/s]
1|2467|Loss: 0.9661283493041992:  38%|███▊      | 2467/6470 [31:16<49:48,  1.34it/s]
1|2468|Loss: 0.9840870499610901:  38%|███▊      | 2467/6470 [31:16<49:48,  1.34it/s]
1|2468|Loss: 0.9840870499610901:  38%|███▊      | 2468/6470 [31:16<50:17,  1.33it/s]
1|2469|Loss: 0.896713376045227:  38%|███▊      | 2468/6470 [31:17<50:17,  1.33it/s] 
1|2469|Loss: 0.896713376045227:  38%|███▊      | 2469/6470 [31:17<50:05,  1.33it/s]
1|2470|Loss: 0.9427600502967834:  38%|███▊      | 2469/6470 [31:18<50:05,  1.33it/s]
1|2470|Loss: 0.9427600502967834:  38%|███▊      | 2470/6470 [31:18<48:19,  1.38it/s]
1|2471|Loss: 0.938098132610321:  38%|███▊      | 2470/6470 [31:18<48:19,  1.38it/s] 
1|2471|Loss: 0.938098132610321:  38%|███▊      | 2471/6470 [31:18<46:22,  1.44it/s]
1|2472|Loss: 0.9160413146018982:  38%|███▊      | 2471/6470 [31:19<46:22,  1.44it/s]
1|2472|Loss: 0.9160413146018982:  38%|███▊      | 2472/6470 [31:19<46:03,  1.45it/s]
1|2473|Loss: 1.0084457397460938:  38%|███▊      | 2472/6470 [31:20<46:03,  1.45it/s]
1|2473|Loss: 1.0084457397460938:  38%|███▊      | 2473/6470 [31:20<46:44,  1.43it/s]
1|2474|Loss: 0.7176798582077026:  38%|███▊      | 2473/6470 [31:20<46:44,  1.43it/s]
1|2474|Loss: 0.7176798582077026:  38%|███▊      | 2474/6470 [31:21<49:32,  1.34it/s]
1|2475|Loss: 0.8127873539924622:  38%|███▊      | 2474/6470 [31:21<49:32,  1.34it/s]
1|2475|Loss: 0.8127873539924622:  38%|███▊      | 2475/6470 [31:21<48:14,  1.38it/s]
1|2476|Loss: 1.0684750080108643:  38%|███▊      | 2475/6470 [31:22<48:14,  1.38it/s]
1|2476|Loss: 1.0684750080108643:  38%|███▊      | 2476/6470 [31:22<47:29,  1.40it/s]
1|2477|Loss: 0.772537350654602:  38%|███▊      | 2476/6470 [31:22<47:29,  1.40it/s] 
1|2477|Loss: 0.772537350654602:  38%|███▊      | 2477/6470 [31:23<46:22,  1.43it/s]
1|2478|Loss: 0.956248939037323:  38%|███▊      | 2477/6470 [31:23<46:22,  1.43it/s]
1|2478|Loss: 0.956248939037323:  38%|███▊      | 2478/6470 [31:23<45:10,  1.47it/s]
1|2479|Loss: 0.9882020354270935:  38%|███▊      | 2478/6470 [31:24<45:10,  1.47it/s]
1|2479|Loss: 0.9882020354270935:  38%|███▊      | 2479/6470 [31:24<41:30,  1.60it/s]
1|2480|Loss: 0.990567684173584:  38%|███▊      | 2479/6470 [31:24<41:30,  1.60it/s] 
1|2480|Loss: 0.990567684173584:  38%|███▊      | 2480/6470 [31:25<41:58,  1.58it/s]
1|2481|Loss: 0.8284244537353516:  38%|███▊      | 2480/6470 [31:25<41:58,  1.58it/s]
1|2481|Loss: 0.8284244537353516:  38%|███▊      | 2481/6470 [31:25<44:14,  1.50it/s]
1|2482|Loss: 0.908280611038208:  38%|███▊      | 2481/6470 [31:26<44:14,  1.50it/s] 
1|2482|Loss: 0.908280611038208:  38%|███▊      | 2482/6470 [31:26<44:24,  1.50it/s]
1|2483|Loss: 0.9174527525901794:  38%|███▊      | 2482/6470 [31:26<44:24,  1.50it/s]
1|2483|Loss: 0.9174527525901794:  38%|███▊      | 2483/6470 [31:27<47:12,  1.41it/s]
1|2484|Loss: 0.9939032196998596:  38%|███▊      | 2483/6470 [31:27<47:12,  1.41it/s]
1|2484|Loss: 0.9939032196998596:  38%|███▊      | 2484/6470 [31:28<48:27,  1.37it/s]
1|2485|Loss: 0.8726032972335815:  38%|███▊      | 2484/6470 [31:28<48:27,  1.37it/s]
1|2485|Loss: 0.8726032972335815:  38%|███▊      | 2485/6470 [31:28<46:44,  1.42it/s]
1|2486|Loss: 0.9440631866455078:  38%|███▊      | 2485/6470 [31:29<46:44,  1.42it/s]
1|2486|Loss: 0.9440631866455078:  38%|███▊      | 2486/6470 [31:29<47:46,  1.39it/s]
1|2487|Loss: 0.7718490958213806:  38%|███▊      | 2486/6470 [31:29<47:46,  1.39it/s]
1|2487|Loss: 0.7718490958213806:  38%|███▊      | 2487/6470 [31:30<48:19,  1.37it/s]
1|2488|Loss: 1.0164837837219238:  38%|███▊      | 2487/6470 [31:30<48:19,  1.37it/s]
1|2488|Loss: 1.0164837837219238:  38%|███▊      | 2488/6470 [31:30<47:06,  1.41it/s]
1|2489|Loss: 1.0134438276290894:  38%|███▊      | 2488/6470 [31:31<47:06,  1.41it/s]
1|2489|Loss: 1.0134438276290894:  38%|███▊      | 2489/6470 [31:31<47:26,  1.40it/s]
1|2490|Loss: 0.8405894637107849:  38%|███▊      | 2489/6470 [31:31<47:26,  1.40it/s]
1|2490|Loss: 0.8405894637107849:  38%|███▊      | 2490/6470 [31:32<46:31,  1.43it/s]
1|2491|Loss: 1.0046340227127075:  38%|███▊      | 2490/6470 [31:32<46:31,  1.43it/s]
1|2491|Loss: 1.0046340227127075:  39%|███▊      | 2491/6470 [31:32<45:17,  1.46it/s]
1|2492|Loss: 0.919185996055603:  39%|███▊      | 2491/6470 [31:33<45:17,  1.46it/s] 
1|2492|Loss: 0.919185996055603:  39%|███▊      | 2492/6470 [31:33<44:00,  1.51it/s]
1|2493|Loss: 0.9185052514076233:  39%|███▊      | 2492/6470 [31:33<44:00,  1.51it/s]
1|2493|Loss: 0.9185052514076233:  39%|███▊      | 2493/6470 [31:34<41:40,  1.59it/s]
1|2494|Loss: 0.9065941572189331:  39%|███▊      | 2493/6470 [31:34<41:40,  1.59it/s]
1|2494|Loss: 0.9065941572189331:  39%|███▊      | 2494/6470 [31:34<42:24,  1.56it/s]
1|2495|Loss: 0.9944256544113159:  39%|███▊      | 2494/6470 [31:35<42:24,  1.56it/s]
1|2495|Loss: 0.9944256544113159:  39%|███▊      | 2495/6470 [31:35<44:23,  1.49it/s]
1|2496|Loss: 1.0255544185638428:  39%|███▊      | 2495/6470 [31:35<44:23,  1.49it/s]
1|2496|Loss: 1.0255544185638428:  39%|███▊      | 2496/6470 [31:36<43:19,  1.53it/s]
1|2497|Loss: 0.9940862655639648:  39%|███▊      | 2496/6470 [31:36<43:19,  1.53it/s]
1|2497|Loss: 0.9940862655639648:  39%|███▊      | 2497/6470 [31:36<44:21,  1.49it/s]
1|2498|Loss: 0.9869235157966614:  39%|███▊      | 2497/6470 [31:37<44:21,  1.49it/s]
1|2498|Loss: 0.9869235157966614:  39%|███▊      | 2498/6470 [31:37<47:16,  1.40it/s]
1|2499|Loss: 0.8852789998054504:  39%|███▊      | 2498/6470 [31:37<47:16,  1.40it/s]
1|2499|Loss: 0.8852789998054504:  39%|███▊      | 2499/6470 [31:38<45:35,  1.45it/s]
1|2500|Loss: 0.7382310032844543:  39%|███▊      | 2499/6470 [31:38<45:35,  1.45it/s]
1|2500|Loss: 0.7382310032844543:  39%|███▊      | 2500/6470 [31:38<45:26,  1.46it/s]
1|2501|Loss: 0.8369526267051697:  39%|███▊      | 2500/6470 [31:39<45:26,  1.46it/s]
1|2501|Loss: 0.8369526267051697:  39%|███▊      | 2501/6470 [31:39<44:45,  1.48it/s]
1|2502|Loss: 0.8898841738700867:  39%|███▊      | 2501/6470 [31:40<44:45,  1.48it/s]
1|2502|Loss: 0.8898841738700867:  39%|███▊      | 2502/6470 [31:40<48:27,  1.36it/s]
1|2503|Loss: 0.7840359210968018:  39%|███▊      | 2502/6470 [31:40<48:27,  1.36it/s]
1|2503|Loss: 0.7840359210968018:  39%|███▊      | 2503/6470 [31:41<50:56,  1.30it/s]
1|2504|Loss: 0.8184634447097778:  39%|███▊      | 2503/6470 [31:41<50:56,  1.30it/s]
1|2504|Loss: 0.8184634447097778:  39%|███▊      | 2504/6470 [31:41<47:46,  1.38it/s]
1|2505|Loss: 0.8776207566261292:  39%|███▊      | 2504/6470 [31:42<47:46,  1.38it/s]
1|2505|Loss: 0.8776207566261292:  39%|███▊      | 2505/6470 [31:42<44:39,  1.48it/s]
1|2506|Loss: 1.042202353477478:  39%|███▊      | 2505/6470 [31:42<44:39,  1.48it/s] 
1|2506|Loss: 1.042202353477478:  39%|███▊      | 2506/6470 [31:42<40:49,  1.62it/s]
1|2507|Loss: 0.7460989356040955:  39%|███▊      | 2506/6470 [31:43<40:49,  1.62it/s]
1|2507|Loss: 0.7460989356040955:  39%|███▊      | 2507/6470 [31:43<40:03,  1.65it/s]
1|2508|Loss: 0.8676872849464417:  39%|███▊      | 2507/6470 [31:43<40:03,  1.65it/s]
1|2508|Loss: 0.8676872849464417:  39%|███▉      | 2508/6470 [31:44<40:37,  1.63it/s]
1|2509|Loss: 0.8080084323883057:  39%|███▉      | 2508/6470 [31:44<40:37,  1.63it/s]
1|2509|Loss: 0.8080084323883057:  39%|███▉      | 2509/6470 [31:44<40:06,  1.65it/s]
1|2510|Loss: 0.912634015083313:  39%|███▉      | 2509/6470 [31:45<40:06,  1.65it/s] 
1|2510|Loss: 0.912634015083313:  39%|███▉      | 2510/6470 [31:45<41:58,  1.57it/s]
1|2511|Loss: 1.0225213766098022:  39%|███▉      | 2510/6470 [31:45<41:58,  1.57it/s]
1|2511|Loss: 1.0225213766098022:  39%|███▉      | 2511/6470 [31:46<41:27,  1.59it/s]
1|2512|Loss: 0.9295746088027954:  39%|███▉      | 2511/6470 [31:46<41:27,  1.59it/s]
1|2512|Loss: 0.9295746088027954:  39%|███▉      | 2512/6470 [31:46<43:13,  1.53it/s]
1|2513|Loss: 0.9152094125747681:  39%|███▉      | 2512/6470 [31:47<43:13,  1.53it/s]
1|2513|Loss: 0.9152094125747681:  39%|███▉      | 2513/6470 [31:47<47:13,  1.40it/s]
1|2514|Loss: 0.7870708703994751:  39%|███▉      | 2513/6470 [31:48<47:13,  1.40it/s]
1|2514|Loss: 0.7870708703994751:  39%|███▉      | 2514/6470 [31:48<50:16,  1.31it/s]
1|2515|Loss: 0.9854653477668762:  39%|███▉      | 2514/6470 [31:48<50:16,  1.31it/s]
1|2515|Loss: 0.9854653477668762:  39%|███▉      | 2515/6470 [31:49<47:03,  1.40it/s]
1|2516|Loss: 0.9299114346504211:  39%|███▉      | 2515/6470 [31:49<47:03,  1.40it/s]
1|2516|Loss: 0.9299114346504211:  39%|███▉      | 2516/6470 [31:49<44:41,  1.47it/s]
1|2517|Loss: 1.005196213722229:  39%|███▉      | 2516/6470 [31:50<44:41,  1.47it/s] 
1|2517|Loss: 1.005196213722229:  39%|███▉      | 2517/6470 [31:50<45:12,  1.46it/s]
1|2518|Loss: 0.9445859789848328:  39%|███▉      | 2517/6470 [31:50<45:12,  1.46it/s]
1|2518|Loss: 0.9445859789848328:  39%|███▉      | 2518/6470 [31:51<46:03,  1.43it/s]
1|2519|Loss: 1.1098582744598389:  39%|███▉      | 2518/6470 [31:51<46:03,  1.43it/s]
1|2519|Loss: 1.1098582744598389:  39%|███▉      | 2519/6470 [31:51<46:55,  1.40it/s]
1|2520|Loss: 0.7939505577087402:  39%|███▉      | 2519/6470 [31:52<46:55,  1.40it/s]
1|2520|Loss: 0.7939505577087402:  39%|███▉      | 2520/6470 [31:52<47:27,  1.39it/s]
1|2521|Loss: 0.7371620535850525:  39%|███▉      | 2520/6470 [31:53<47:27,  1.39it/s]
1|2521|Loss: 0.7371620535850525:  39%|███▉      | 2521/6470 [31:53<51:19,  1.28it/s]
1|2522|Loss: 0.8674529194831848:  39%|███▉      | 2521/6470 [31:54<51:19,  1.28it/s]
1|2522|Loss: 0.8674529194831848:  39%|███▉      | 2522/6470 [31:54<54:28,  1.21it/s]
1|2523|Loss: 0.9026244282722473:  39%|███▉      | 2522/6470 [31:54<54:28,  1.21it/s]
1|2523|Loss: 0.9026244282722473:  39%|███▉      | 2523/6470 [31:55<53:20,  1.23it/s]
1|2524|Loss: 0.8959546089172363:  39%|███▉      | 2523/6470 [31:55<53:20,  1.23it/s]
1|2524|Loss: 0.8959546089172363:  39%|███▉      | 2524/6470 [31:55<49:18,  1.33it/s]
1|2525|Loss: 0.9635500907897949:  39%|███▉      | 2524/6470 [31:56<49:18,  1.33it/s]
1|2525|Loss: 0.9635500907897949:  39%|███▉      | 2525/6470 [31:56<49:53,  1.32it/s]
1|2526|Loss: 1.1160993576049805:  39%|███▉      | 2525/6470 [31:57<49:53,  1.32it/s]
1|2526|Loss: 1.1160993576049805:  39%|███▉      | 2526/6470 [31:57<46:57,  1.40it/s]
1|2527|Loss: 0.9175401926040649:  39%|███▉      | 2526/6470 [31:57<46:57,  1.40it/s]
1|2527|Loss: 0.9175401926040649:  39%|███▉      | 2527/6470 [31:57<44:53,  1.46it/s]
1|2528|Loss: 0.9190106391906738:  39%|███▉      | 2527/6470 [31:58<44:53,  1.46it/s]
1|2528|Loss: 0.9190106391906738:  39%|███▉      | 2528/6470 [31:58<45:53,  1.43it/s]
1|2529|Loss: 0.8976389765739441:  39%|███▉      | 2528/6470 [31:59<45:53,  1.43it/s]
1|2529|Loss: 0.8976389765739441:  39%|███▉      | 2529/6470 [31:59<46:58,  1.40it/s]
1|2530|Loss: 0.9265963435173035:  39%|███▉      | 2529/6470 [31:59<46:58,  1.40it/s]
1|2530|Loss: 0.9265963435173035:  39%|███▉      | 2530/6470 [31:59<43:53,  1.50it/s]
1|2531|Loss: 0.922484278678894:  39%|███▉      | 2530/6470 [32:00<43:53,  1.50it/s] 
1|2531|Loss: 0.922484278678894:  39%|███▉      | 2531/6470 [32:00<42:51,  1.53it/s]
1|2532|Loss: 0.8144326210021973:  39%|███▉      | 2531/6470 [32:00<42:51,  1.53it/s]
1|2532|Loss: 0.8144326210021973:  39%|███▉      | 2532/6470 [32:01<44:01,  1.49it/s]
1|2533|Loss: 0.8266354203224182:  39%|███▉      | 2532/6470 [32:01<44:01,  1.49it/s]
1|2533|Loss: 0.8266354203224182:  39%|███▉      | 2533/6470 [32:02<46:10,  1.42it/s]
1|2534|Loss: 1.1248630285263062:  39%|███▉      | 2533/6470 [32:02<46:10,  1.42it/s]
1|2534|Loss: 1.1248630285263062:  39%|███▉      | 2534/6470 [32:02<47:01,  1.39it/s]
1|2535|Loss: 0.8218284845352173:  39%|███▉      | 2534/6470 [32:03<47:01,  1.39it/s]
1|2535|Loss: 0.8218284845352173:  39%|███▉      | 2535/6470 [32:03<43:48,  1.50it/s]
1|2536|Loss: 0.8192080855369568:  39%|███▉      | 2535/6470 [32:03<43:48,  1.50it/s]
1|2536|Loss: 0.8192080855369568:  39%|███▉      | 2536/6470 [32:03<43:40,  1.50it/s]
1|2537|Loss: 0.8125833868980408:  39%|███▉      | 2536/6470 [32:04<43:40,  1.50it/s]
1|2537|Loss: 0.8125833868980408:  39%|███▉      | 2537/6470 [32:04<46:49,  1.40it/s]
1|2538|Loss: 0.8339543342590332:  39%|███▉      | 2537/6470 [32:05<46:49,  1.40it/s]
1|2538|Loss: 0.8339543342590332:  39%|███▉      | 2538/6470 [32:05<42:16,  1.55it/s]
1|2539|Loss: 1.0246000289916992:  39%|███▉      | 2538/6470 [32:05<42:16,  1.55it/s]
1|2539|Loss: 1.0246000289916992:  39%|███▉      | 2539/6470 [32:05<42:59,  1.52it/s]
1|2540|Loss: 0.8819621205329895:  39%|███▉      | 2539/6470 [32:06<42:59,  1.52it/s]
1|2540|Loss: 0.8819621205329895:  39%|███▉      | 2540/6470 [32:06<46:10,  1.42it/s]
1|2541|Loss: 0.9705080986022949:  39%|███▉      | 2540/6470 [32:07<46:10,  1.42it/s]
1|2541|Loss: 0.9705080986022949:  39%|███▉      | 2541/6470 [32:07<49:20,  1.33it/s]
1|2542|Loss: 0.9392979145050049:  39%|███▉      | 2541/6470 [32:08<49:20,  1.33it/s]
1|2542|Loss: 0.9392979145050049:  39%|███▉      | 2542/6470 [32:08<47:51,  1.37it/s]
1|2543|Loss: 0.8600229024887085:  39%|███▉      | 2542/6470 [32:08<47:51,  1.37it/s]
1|2543|Loss: 0.8600229024887085:  39%|███▉      | 2543/6470 [32:09<46:58,  1.39it/s]
1|2544|Loss: 0.8884831070899963:  39%|███▉      | 2543/6470 [32:09<46:58,  1.39it/s]
1|2544|Loss: 0.8884831070899963:  39%|███▉      | 2544/6470 [32:10<53:00,  1.23it/s]
1|2545|Loss: 0.9652420282363892:  39%|███▉      | 2544/6470 [32:10<53:00,  1.23it/s]
1|2545|Loss: 0.9652420282363892:  39%|███▉      | 2545/6470 [32:10<49:40,  1.32it/s]
1|2546|Loss: 0.9410032033920288:  39%|███▉      | 2545/6470 [32:11<49:40,  1.32it/s]
1|2546|Loss: 0.9410032033920288:  39%|███▉      | 2546/6470 [32:11<48:13,  1.36it/s]
1|2547|Loss: 0.9221549034118652:  39%|███▉      | 2546/6470 [32:11<48:13,  1.36it/s]
1|2547|Loss: 0.9221549034118652:  39%|███▉      | 2547/6470 [32:12<47:32,  1.38it/s]
1|2548|Loss: 0.8589292168617249:  39%|███▉      | 2547/6470 [32:12<47:32,  1.38it/s]
1|2548|Loss: 0.8589292168617249:  39%|███▉      | 2548/6470 [32:12<46:50,  1.40it/s]
1|2549|Loss: 1.1004105806350708:  39%|███▉      | 2548/6470 [32:13<46:50,  1.40it/s]
1|2549|Loss: 1.1004105806350708:  39%|███▉      | 2549/6470 [32:13<45:07,  1.45it/s]
1|2550|Loss: 0.8768010139465332:  39%|███▉      | 2549/6470 [32:13<45:07,  1.45it/s]
1|2550|Loss: 0.8768010139465332:  39%|███▉      | 2550/6470 [32:14<43:07,  1.51it/s]
1|2551|Loss: 0.7762471437454224:  39%|███▉      | 2550/6470 [32:14<43:07,  1.51it/s]
1|2551|Loss: 0.7762471437454224:  39%|███▉      | 2551/6470 [32:14<41:40,  1.57it/s]
1|2552|Loss: 1.0288491249084473:  39%|███▉      | 2551/6470 [32:14<41:40,  1.57it/s]
1|2552|Loss: 1.0288491249084473:  39%|███▉      | 2552/6470 [32:15<41:10,  1.59it/s]
1|2553|Loss: 0.920619785785675:  39%|███▉      | 2552/6470 [32:15<41:10,  1.59it/s] 
1|2553|Loss: 0.920619785785675:  39%|███▉      | 2553/6470 [32:15<39:17,  1.66it/s]
1|2554|Loss: 1.0045795440673828:  39%|███▉      | 2553/6470 [32:16<39:17,  1.66it/s]
1|2554|Loss: 1.0045795440673828:  39%|███▉      | 2554/6470 [32:16<39:29,  1.65it/s]
1|2555|Loss: 0.8570411801338196:  39%|███▉      | 2554/6470 [32:16<39:29,  1.65it/s]
1|2555|Loss: 0.8570411801338196:  39%|███▉      | 2555/6470 [32:16<37:44,  1.73it/s]
1|2556|Loss: 0.8312018513679504:  39%|███▉      | 2555/6470 [32:17<37:44,  1.73it/s]
1|2556|Loss: 0.8312018513679504:  40%|███▉      | 2556/6470 [32:17<38:15,  1.70it/s]
1|2557|Loss: 0.9928911328315735:  40%|███▉      | 2556/6470 [32:17<38:15,  1.70it/s]
1|2557|Loss: 0.9928911328315735:  40%|███▉      | 2557/6470 [32:18<38:19,  1.70it/s]
1|2558|Loss: 0.9726324081420898:  40%|███▉      | 2557/6470 [32:18<38:19,  1.70it/s]
1|2558|Loss: 0.9726324081420898:  40%|███▉      | 2558/6470 [32:18<38:52,  1.68it/s]
1|2559|Loss: 0.6473774313926697:  40%|███▉      | 2558/6470 [32:19<38:52,  1.68it/s]
1|2559|Loss: 0.6473774313926697:  40%|███▉      | 2559/6470 [32:19<39:17,  1.66it/s]
1|2560|Loss: 0.7727296948432922:  40%|███▉      | 2559/6470 [32:19<39:17,  1.66it/s]
1|2560|Loss: 0.7727296948432922:  40%|███▉      | 2560/6470 [32:20<42:32,  1.53it/s]
1|2561|Loss: 0.8989758491516113:  40%|███▉      | 2560/6470 [32:20<42:32,  1.53it/s]
1|2561|Loss: 0.8989758491516113:  40%|███▉      | 2561/6470 [32:20<42:58,  1.52it/s]
1|2562|Loss: 0.9722002744674683:  40%|███▉      | 2561/6470 [32:21<42:58,  1.52it/s]
1|2562|Loss: 0.9722002744674683:  40%|███▉      | 2562/6470 [32:21<41:15,  1.58it/s]
1|2563|Loss: 0.8711938858032227:  40%|███▉      | 2562/6470 [32:21<41:15,  1.58it/s]
1|2563|Loss: 0.8711938858032227:  40%|███▉      | 2563/6470 [32:22<42:58,  1.52it/s]
1|2564|Loss: 0.876983642578125:  40%|███▉      | 2563/6470 [32:22<42:58,  1.52it/s] 
1|2564|Loss: 0.876983642578125:  40%|███▉      | 2564/6470 [32:22<40:11,  1.62it/s]
1|2565|Loss: 0.9557256698608398:  40%|███▉      | 2564/6470 [32:22<40:11,  1.62it/s]
1|2565|Loss: 0.9557256698608398:  40%|███▉      | 2565/6470 [32:23<39:24,  1.65it/s]
1|2566|Loss: 0.8993474841117859:  40%|███▉      | 2565/6470 [32:23<39:24,  1.65it/s]
1|2566|Loss: 0.8993474841117859:  40%|███▉      | 2566/6470 [32:23<41:28,  1.57it/s]
1|2567|Loss: 1.0613490343093872:  40%|███▉      | 2566/6470 [32:24<41:28,  1.57it/s]
1|2567|Loss: 1.0613490343093872:  40%|███▉      | 2567/6470 [32:24<42:53,  1.52it/s]
1|2568|Loss: 0.8830760717391968:  40%|███▉      | 2567/6470 [32:24<42:53,  1.52it/s]
1|2568|Loss: 0.8830760717391968:  40%|███▉      | 2568/6470 [32:25<43:49,  1.48it/s]
1|2569|Loss: 0.8738024830818176:  40%|███▉      | 2568/6470 [32:25<43:49,  1.48it/s]
1|2569|Loss: 0.8738024830818176:  40%|███▉      | 2569/6470 [32:25<43:07,  1.51it/s]
1|2570|Loss: 0.8203929662704468:  40%|███▉      | 2569/6470 [32:26<43:07,  1.51it/s]
1|2570|Loss: 0.8203929662704468:  40%|███▉      | 2570/6470 [32:26<42:34,  1.53it/s]
1|2571|Loss: 0.923638641834259:  40%|███▉      | 2570/6470 [32:26<42:34,  1.53it/s] 
1|2571|Loss: 0.923638641834259:  40%|███▉      | 2571/6470 [32:27<42:17,  1.54it/s]
1|2572|Loss: 0.9834209084510803:  40%|███▉      | 2571/6470 [32:27<42:17,  1.54it/s]
1|2572|Loss: 0.9834209084510803:  40%|███▉      | 2572/6470 [32:27<43:27,  1.49it/s]
1|2573|Loss: 0.8725422024726868:  40%|███▉      | 2572/6470 [32:28<43:27,  1.49it/s]
1|2573|Loss: 0.8725422024726868:  40%|███▉      | 2573/6470 [32:28<44:19,  1.47it/s]
1|2574|Loss: 0.9105613827705383:  40%|███▉      | 2573/6470 [32:29<44:19,  1.47it/s]
1|2574|Loss: 0.9105613827705383:  40%|███▉      | 2574/6470 [32:29<45:05,  1.44it/s]
1|2575|Loss: 0.9410414695739746:  40%|███▉      | 2574/6470 [32:29<45:05,  1.44it/s]
1|2575|Loss: 0.9410414695739746:  40%|███▉      | 2575/6470 [32:29<44:30,  1.46it/s]
1|2576|Loss: 1.0630488395690918:  40%|███▉      | 2575/6470 [32:30<44:30,  1.46it/s]
1|2576|Loss: 1.0630488395690918:  40%|███▉      | 2576/6470 [32:30<45:02,  1.44it/s]
1|2577|Loss: 0.9291390776634216:  40%|███▉      | 2576/6470 [32:31<45:02,  1.44it/s]
1|2577|Loss: 0.9291390776634216:  40%|███▉      | 2577/6470 [32:31<44:18,  1.46it/s]
1|2578|Loss: 0.9698588848114014:  40%|███▉      | 2577/6470 [32:31<44:18,  1.46it/s]
1|2578|Loss: 0.9698588848114014:  40%|███▉      | 2578/6470 [32:32<43:23,  1.50it/s]
1|2579|Loss: 0.8881098031997681:  40%|███▉      | 2578/6470 [32:32<43:23,  1.50it/s]
1|2579|Loss: 0.8881098031997681:  40%|███▉      | 2579/6470 [32:32<42:31,  1.52it/s]
1|2580|Loss: 0.9177286028862:  40%|███▉      | 2579/6470 [32:32<42:31,  1.52it/s]   
1|2580|Loss: 0.9177286028862:  40%|███▉      | 2580/6470 [32:33<42:06,  1.54it/s]
1|2581|Loss: 0.985615074634552:  40%|███▉      | 2580/6470 [32:33<42:06,  1.54it/s]
1|2581|Loss: 0.985615074634552:  40%|███▉      | 2581/6470 [32:33<43:19,  1.50it/s]
1|2582|Loss: 0.8307691812515259:  40%|███▉      | 2581/6470 [32:34<43:19,  1.50it/s]
1|2582|Loss: 0.8307691812515259:  40%|███▉      | 2582/6470 [32:34<44:52,  1.44it/s]
1|2583|Loss: 0.945757269859314:  40%|███▉      | 2582/6470 [32:35<44:52,  1.44it/s] 
1|2583|Loss: 0.945757269859314:  40%|███▉      | 2583/6470 [32:35<48:05,  1.35it/s]
1|2584|Loss: 0.9167461395263672:  40%|███▉      | 2583/6470 [32:35<48:05,  1.35it/s]
1|2584|Loss: 0.9167461395263672:  40%|███▉      | 2584/6470 [32:36<47:17,  1.37it/s]
1|2585|Loss: 0.9388482570648193:  40%|███▉      | 2584/6470 [32:36<47:17,  1.37it/s]
1|2585|Loss: 0.9388482570648193:  40%|███▉      | 2585/6470 [32:36<46:51,  1.38it/s]
1|2586|Loss: 0.7388600707054138:  40%|███▉      | 2585/6470 [32:37<46:51,  1.38it/s]
1|2586|Loss: 0.7388600707054138:  40%|███▉      | 2586/6470 [32:37<46:30,  1.39it/s]
1|2587|Loss: 1.1023163795471191:  40%|███▉      | 2586/6470 [32:38<46:30,  1.39it/s]
1|2587|Loss: 1.1023163795471191:  40%|███▉      | 2587/6470 [32:38<46:21,  1.40it/s]
1|2588|Loss: 0.887172520160675:  40%|███▉      | 2587/6470 [32:38<46:21,  1.40it/s] 
1|2588|Loss: 0.887172520160675:  40%|████      | 2588/6470 [32:38<40:43,  1.59it/s]
1|2589|Loss: 0.80438232421875:  40%|████      | 2588/6470 [32:39<40:43,  1.59it/s] 
1|2589|Loss: 0.80438232421875:  40%|████      | 2589/6470 [32:39<42:09,  1.53it/s]
1|2590|Loss: 0.9863184094429016:  40%|████      | 2589/6470 [32:39<42:09,  1.53it/s]
1|2590|Loss: 0.9863184094429016:  40%|████      | 2590/6470 [32:40<42:45,  1.51it/s]
1|2591|Loss: 0.8265334963798523:  40%|████      | 2590/6470 [32:40<42:45,  1.51it/s]
1|2591|Loss: 0.8265334963798523:  40%|████      | 2591/6470 [32:40<42:19,  1.53it/s]
1|2592|Loss: 0.9506561756134033:  40%|████      | 2591/6470 [32:41<42:19,  1.53it/s]
1|2592|Loss: 0.9506561756134033:  40%|████      | 2592/6470 [32:41<42:55,  1.51it/s]
1|2593|Loss: 0.9233808517456055:  40%|████      | 2592/6470 [32:41<42:55,  1.51it/s]
1|2593|Loss: 0.9233808517456055:  40%|████      | 2593/6470 [32:42<43:44,  1.48it/s]
1|2594|Loss: 0.9331151843070984:  40%|████      | 2593/6470 [32:42<43:44,  1.48it/s]
1|2594|Loss: 0.9331151843070984:  40%|████      | 2594/6470 [32:42<42:11,  1.53it/s]
1|2595|Loss: 0.7482627630233765:  40%|████      | 2594/6470 [32:43<42:11,  1.53it/s]
1|2595|Loss: 0.7482627630233765:  40%|████      | 2595/6470 [32:43<45:27,  1.42it/s]
1|2596|Loss: 0.7338561415672302:  40%|████      | 2595/6470 [32:44<45:27,  1.42it/s]
1|2596|Loss: 0.7338561415672302:  40%|████      | 2596/6470 [32:44<50:40,  1.27it/s]
1|2597|Loss: 1.0309442281723022:  40%|████      | 2596/6470 [32:45<50:40,  1.27it/s]
1|2597|Loss: 1.0309442281723022:  40%|████      | 2597/6470 [32:45<49:06,  1.31it/s]
1|2598|Loss: 1.0561341047286987:  40%|████      | 2597/6470 [32:45<49:06,  1.31it/s]
1|2598|Loss: 1.0561341047286987:  40%|████      | 2598/6470 [32:45<44:43,  1.44it/s]
1|2599|Loss: 0.9804477691650391:  40%|████      | 2598/6470 [32:46<44:43,  1.44it/s]
1|2599|Loss: 0.9804477691650391:  40%|████      | 2599/6470 [32:46<42:05,  1.53it/s]
1|2600|Loss: 0.7378348708152771:  40%|████      | 2599/6470 [32:46<42:05,  1.53it/s]
1|2600|Loss: 0.7378348708152771:  40%|████      | 2600/6470 [32:47<43:23,  1.49it/s]
1|2601|Loss: 1.023887038230896:  40%|████      | 2600/6470 [32:47<43:23,  1.49it/s] 
1|2601|Loss: 1.023887038230896:  40%|████      | 2601/6470 [32:48<46:36,  1.38it/s]
1|2602|Loss: 0.9586111307144165:  40%|████      | 2601/6470 [32:48<46:36,  1.38it/s]
1|2602|Loss: 0.9586111307144165:  40%|████      | 2602/6470 [32:48<45:37,  1.41it/s]
1|2603|Loss: 1.0015153884887695:  40%|████      | 2602/6470 [32:49<45:37,  1.41it/s]
1|2603|Loss: 1.0015153884887695:  40%|████      | 2603/6470 [32:49<44:12,  1.46it/s]
1|2604|Loss: 0.8502570986747742:  40%|████      | 2603/6470 [32:49<44:12,  1.46it/s]
1|2604|Loss: 0.8502570986747742:  40%|████      | 2604/6470 [32:50<46:19,  1.39it/s]
1|2605|Loss: 0.9013723134994507:  40%|████      | 2604/6470 [32:50<46:19,  1.39it/s]
1|2605|Loss: 0.9013723134994507:  40%|████      | 2605/6470 [32:50<45:39,  1.41it/s]
1|2606|Loss: 0.8365240097045898:  40%|████      | 2605/6470 [32:51<45:39,  1.41it/s]
1|2606|Loss: 0.8365240097045898:  40%|████      | 2606/6470 [32:51<47:12,  1.36it/s]
1|2607|Loss: 0.944341778755188:  40%|████      | 2606/6470 [32:51<47:12,  1.36it/s] 
1|2607|Loss: 0.944341778755188:  40%|████      | 2607/6470 [32:52<44:45,  1.44it/s]
1|2608|Loss: 0.9670491814613342:  40%|████      | 2607/6470 [32:52<44:45,  1.44it/s]
1|2608|Loss: 0.9670491814613342:  40%|████      | 2608/6470 [32:52<45:04,  1.43it/s]
1|2609|Loss: 0.8574350476264954:  40%|████      | 2608/6470 [32:53<45:04,  1.43it/s]
1|2609|Loss: 0.8574350476264954:  40%|████      | 2609/6470 [32:53<44:25,  1.45it/s]
1|2610|Loss: 0.8855517506599426:  40%|████      | 2609/6470 [32:53<44:25,  1.45it/s]
1|2610|Loss: 0.8855517506599426:  40%|████      | 2610/6470 [32:54<43:06,  1.49it/s]
1|2611|Loss: 0.9239039421081543:  40%|████      | 2610/6470 [32:54<43:06,  1.49it/s]
1|2611|Loss: 0.9239039421081543:  40%|████      | 2611/6470 [32:54<42:07,  1.53it/s]
1|2612|Loss: 0.8756617903709412:  40%|████      | 2611/6470 [32:55<42:07,  1.53it/s]
1|2612|Loss: 0.8756617903709412:  40%|████      | 2612/6470 [32:55<41:27,  1.55it/s]
1|2613|Loss: 0.9254913926124573:  40%|████      | 2612/6470 [32:55<41:27,  1.55it/s]
1|2613|Loss: 0.9254913926124573:  40%|████      | 2613/6470 [32:56<41:05,  1.56it/s]
1|2614|Loss: 0.8796517848968506:  40%|████      | 2613/6470 [32:56<41:05,  1.56it/s]
1|2614|Loss: 0.8796517848968506:  40%|████      | 2614/6470 [32:56<42:38,  1.51it/s]
1|2615|Loss: 0.7505495548248291:  40%|████      | 2614/6470 [32:57<42:38,  1.51it/s]
1|2615|Loss: 0.7505495548248291:  40%|████      | 2615/6470 [32:57<46:05,  1.39it/s]
1|2616|Loss: 0.9058238863945007:  40%|████      | 2615/6470 [32:58<46:05,  1.39it/s]
1|2616|Loss: 0.9058238863945007:  40%|████      | 2616/6470 [32:58<47:07,  1.36it/s]
1|2617|Loss: 0.8371477723121643:  40%|████      | 2616/6470 [32:58<47:07,  1.36it/s]
1|2617|Loss: 0.8371477723121643:  40%|████      | 2617/6470 [32:59<45:51,  1.40it/s]
1|2618|Loss: 0.8107007145881653:  40%|████      | 2617/6470 [32:59<45:51,  1.40it/s]
1|2618|Loss: 0.8107007145881653:  40%|████      | 2618/6470 [32:59<40:30,  1.58it/s]
1|2619|Loss: 0.8719778060913086:  40%|████      | 2618/6470 [32:59<40:30,  1.58it/s]
1|2619|Loss: 0.8719778060913086:  40%|████      | 2619/6470 [33:00<40:49,  1.57it/s]
1|2620|Loss: 0.8384067416191101:  40%|████      | 2619/6470 [33:00<40:49,  1.57it/s]
1|2620|Loss: 0.8384067416191101:  40%|████      | 2620/6470 [33:00<42:03,  1.53it/s]
1|2621|Loss: 0.8791555762290955:  40%|████      | 2620/6470 [33:01<42:03,  1.53it/s]
1|2621|Loss: 0.8791555762290955:  41%|████      | 2621/6470 [33:01<43:22,  1.48it/s]
1|2622|Loss: 0.9327548742294312:  41%|████      | 2621/6470 [33:02<43:22,  1.48it/s]
1|2622|Loss: 0.9327548742294312:  41%|████      | 2622/6470 [33:02<46:21,  1.38it/s]
1|2623|Loss: 0.8419879674911499:  41%|████      | 2622/6470 [33:02<46:21,  1.38it/s]
1|2623|Loss: 0.8419879674911499:  41%|████      | 2623/6470 [33:03<47:23,  1.35it/s]
1|2624|Loss: 0.9071874618530273:  41%|████      | 2623/6470 [33:03<47:23,  1.35it/s]
1|2624|Loss: 0.9071874618530273:  41%|████      | 2624/6470 [33:03<47:23,  1.35it/s]
1|2625|Loss: 0.8373696804046631:  41%|████      | 2624/6470 [33:04<47:23,  1.35it/s]
1|2625|Loss: 0.8373696804046631:  41%|████      | 2625/6470 [33:04<45:21,  1.41it/s]
1|2626|Loss: 0.8496769070625305:  41%|████      | 2625/6470 [33:04<45:21,  1.41it/s]
1|2626|Loss: 0.8496769070625305:  41%|████      | 2626/6470 [33:05<46:44,  1.37it/s]
1|2627|Loss: 0.9019593000411987:  41%|████      | 2626/6470 [33:05<46:44,  1.37it/s]
1|2627|Loss: 0.9019593000411987:  41%|████      | 2627/6470 [33:05<41:49,  1.53it/s]
1|2628|Loss: 0.7883666753768921:  41%|████      | 2627/6470 [33:06<41:49,  1.53it/s]
1|2628|Loss: 0.7883666753768921:  41%|████      | 2628/6470 [33:06<41:48,  1.53it/s]
1|2629|Loss: 0.8204697370529175:  41%|████      | 2628/6470 [33:06<41:48,  1.53it/s]
1|2629|Loss: 0.8204697370529175:  41%|████      | 2629/6470 [33:07<45:25,  1.41it/s]
1|2630|Loss: 0.9104332327842712:  41%|████      | 2629/6470 [33:07<45:25,  1.41it/s]
1|2630|Loss: 0.9104332327842712:  41%|████      | 2630/6470 [33:08<48:36,  1.32it/s]
1|2631|Loss: 0.8761642575263977:  41%|████      | 2630/6470 [33:08<48:36,  1.32it/s]
1|2631|Loss: 0.8761642575263977:  41%|████      | 2631/6470 [33:08<47:17,  1.35it/s]
1|2632|Loss: 1.0997259616851807:  41%|████      | 2631/6470 [33:09<47:17,  1.35it/s]
1|2632|Loss: 1.0997259616851807:  41%|████      | 2632/6470 [33:09<43:19,  1.48it/s]
1|2633|Loss: 0.7612184286117554:  41%|████      | 2632/6470 [33:09<43:19,  1.48it/s]
1|2633|Loss: 0.7612184286117554:  41%|████      | 2633/6470 [33:10<44:26,  1.44it/s]
1|2634|Loss: 0.8884581327438354:  41%|████      | 2633/6470 [33:10<44:26,  1.44it/s]
1|2634|Loss: 0.8884581327438354:  41%|████      | 2634/6470 [33:10<46:51,  1.36it/s]
1|2635|Loss: 0.9743505120277405:  41%|████      | 2634/6470 [33:11<46:51,  1.36it/s]
1|2635|Loss: 0.9743505120277405:  41%|████      | 2635/6470 [33:11<45:00,  1.42it/s]
1|2636|Loss: 0.7223050594329834:  41%|████      | 2635/6470 [33:11<45:00,  1.42it/s]
1|2636|Loss: 0.7223050594329834:  41%|████      | 2636/6470 [33:12<44:26,  1.44it/s]
1|2637|Loss: 0.9834961295127869:  41%|████      | 2636/6470 [33:12<44:26,  1.44it/s]
1|2637|Loss: 0.9834961295127869:  41%|████      | 2637/6470 [33:13<45:15,  1.41it/s]
1|2638|Loss: 0.923990786075592:  41%|████      | 2637/6470 [33:13<45:15,  1.41it/s] 
1|2638|Loss: 0.923990786075592:  41%|████      | 2638/6470 [33:13<45:56,  1.39it/s]
1|2639|Loss: 0.9837434887886047:  41%|████      | 2638/6470 [33:14<45:56,  1.39it/s]
1|2639|Loss: 0.9837434887886047:  41%|████      | 2639/6470 [33:14<45:49,  1.39it/s]
1|2640|Loss: 0.9161178469657898:  41%|████      | 2639/6470 [33:14<45:49,  1.39it/s]
1|2640|Loss: 0.9161178469657898:  41%|████      | 2640/6470 [33:14<40:13,  1.59it/s]
1|2641|Loss: 0.8859137892723083:  41%|████      | 2640/6470 [33:15<40:13,  1.59it/s]
1|2641|Loss: 0.8859137892723083:  41%|████      | 2641/6470 [33:15<40:48,  1.56it/s]
1|2642|Loss: 0.9480631947517395:  41%|████      | 2641/6470 [33:15<40:48,  1.56it/s]
1|2642|Loss: 0.9480631947517395:  41%|████      | 2642/6470 [33:16<41:16,  1.55it/s]
1|2643|Loss: 0.9540175795555115:  41%|████      | 2642/6470 [33:16<41:16,  1.55it/s]
1|2643|Loss: 0.9540175795555115:  41%|████      | 2643/6470 [33:16<40:39,  1.57it/s]
1|2644|Loss: 0.9905449151992798:  41%|████      | 2643/6470 [33:17<40:39,  1.57it/s]
1|2644|Loss: 0.9905449151992798:  41%|████      | 2644/6470 [33:17<40:39,  1.57it/s]
1|2645|Loss: 0.8065791726112366:  41%|████      | 2644/6470 [33:17<40:39,  1.57it/s]
1|2645|Loss: 0.8065791726112366:  41%|████      | 2645/6470 [33:18<41:45,  1.53it/s]
1|2646|Loss: 0.8991144299507141:  41%|████      | 2645/6470 [33:18<41:45,  1.53it/s]
1|2646|Loss: 0.8991144299507141:  41%|████      | 2646/6470 [33:18<42:38,  1.49it/s]
1|2647|Loss: 0.9767752885818481:  41%|████      | 2646/6470 [33:19<42:38,  1.49it/s]
1|2647|Loss: 0.9767752885818481:  41%|████      | 2647/6470 [33:19<41:58,  1.52it/s]
1|2648|Loss: 0.9073156714439392:  41%|████      | 2647/6470 [33:19<41:58,  1.52it/s]
1|2648|Loss: 0.9073156714439392:  41%|████      | 2648/6470 [33:20<42:56,  1.48it/s]
1|2649|Loss: 0.8234076499938965:  41%|████      | 2648/6470 [33:20<42:56,  1.48it/s]
1|2649|Loss: 0.8234076499938965:  41%|████      | 2649/6470 [33:20<42:20,  1.50it/s]
1|2650|Loss: 0.8560683131217957:  41%|████      | 2649/6470 [33:21<42:20,  1.50it/s]
1|2650|Loss: 0.8560683131217957:  41%|████      | 2650/6470 [33:21<42:42,  1.49it/s]
1|2651|Loss: 0.8312979340553284:  41%|████      | 2650/6470 [33:21<42:42,  1.49it/s]
1|2651|Loss: 0.8312979340553284:  41%|████      | 2651/6470 [33:22<43:58,  1.45it/s]
1|2652|Loss: 0.8609928488731384:  41%|████      | 2651/6470 [33:22<43:58,  1.45it/s]
1|2652|Loss: 0.8609928488731384:  41%|████      | 2652/6470 [33:23<46:44,  1.36it/s]
1|2653|Loss: 0.8746013045310974:  41%|████      | 2652/6470 [33:23<46:44,  1.36it/s]
1|2653|Loss: 0.8746013045310974:  41%|████      | 2653/6470 [33:23<45:56,  1.38it/s]
1|2654|Loss: 0.9118539094924927:  41%|████      | 2653/6470 [33:24<45:56,  1.38it/s]
1|2654|Loss: 0.9118539094924927:  41%|████      | 2654/6470 [33:24<46:42,  1.36it/s]
1|2655|Loss: 0.837245523929596:  41%|████      | 2654/6470 [33:25<46:42,  1.36it/s] 
1|2655|Loss: 0.837245523929596:  41%|████      | 2655/6470 [33:25<47:50,  1.33it/s]
1|2656|Loss: 0.9165735244750977:  41%|████      | 2655/6470 [33:25<47:50,  1.33it/s]
1|2656|Loss: 0.9165735244750977:  41%|████      | 2656/6470 [33:26<45:58,  1.38it/s]
1|2657|Loss: 0.8768436312675476:  41%|████      | 2656/6470 [33:26<45:58,  1.38it/s]
1|2657|Loss: 0.8768436312675476:  41%|████      | 2657/6470 [33:26<43:51,  1.45it/s]
1|2658|Loss: 0.9051793813705444:  41%|████      | 2657/6470 [33:26<43:51,  1.45it/s]
1|2658|Loss: 0.9051793813705444:  41%|████      | 2658/6470 [33:27<42:24,  1.50it/s]
1|2659|Loss: 0.8319350481033325:  41%|████      | 2658/6470 [33:27<42:24,  1.50it/s]
1|2659|Loss: 0.8319350481033325:  41%|████      | 2659/6470 [33:28<44:41,  1.42it/s]
1|2660|Loss: 0.8374797701835632:  41%|████      | 2659/6470 [33:28<44:41,  1.42it/s]
1|2660|Loss: 0.8374797701835632:  41%|████      | 2660/6470 [33:28<47:08,  1.35it/s]
1|2661|Loss: 0.9217232465744019:  41%|████      | 2660/6470 [33:29<47:08,  1.35it/s]
1|2661|Loss: 0.9217232465744019:  41%|████      | 2661/6470 [33:29<47:12,  1.34it/s]
1|2662|Loss: 0.6633924841880798:  41%|████      | 2661/6470 [33:30<47:12,  1.34it/s]
1|2662|Loss: 0.6633924841880798:  41%|████      | 2662/6470 [33:30<46:59,  1.35it/s]
1|2663|Loss: 0.9364070892333984:  41%|████      | 2662/6470 [33:30<46:59,  1.35it/s]
1|2663|Loss: 0.9364070892333984:  41%|████      | 2663/6470 [33:31<46:24,  1.37it/s]
1|2664|Loss: 0.7878021597862244:  41%|████      | 2663/6470 [33:31<46:24,  1.37it/s]
1|2664|Loss: 0.7878021597862244:  41%|████      | 2664/6470 [33:31<44:55,  1.41it/s]
1|2665|Loss: 0.9446655511856079:  41%|████      | 2664/6470 [33:32<44:55,  1.41it/s]
1|2665|Loss: 0.9446655511856079:  41%|████      | 2665/6470 [33:32<45:51,  1.38it/s]
1|2666|Loss: 0.8886846899986267:  41%|████      | 2665/6470 [33:32<45:51,  1.38it/s]
1|2666|Loss: 0.8886846899986267:  41%|████      | 2666/6470 [33:33<43:36,  1.45it/s]
1|2667|Loss: 0.962787389755249:  41%|████      | 2666/6470 [33:33<43:36,  1.45it/s] 
1|2667|Loss: 0.962787389755249:  41%|████      | 2667/6470 [33:33<43:55,  1.44it/s]
1|2668|Loss: 1.0965898036956787:  41%|████      | 2667/6470 [33:34<43:55,  1.44it/s]
1|2668|Loss: 1.0965898036956787:  41%|████      | 2668/6470 [33:34<43:26,  1.46it/s]
1|2669|Loss: 0.9717581272125244:  41%|████      | 2668/6470 [33:34<43:26,  1.46it/s]
1|2669|Loss: 0.9717581272125244:  41%|████▏     | 2669/6470 [33:35<41:29,  1.53it/s]
1|2670|Loss: 0.8154118061065674:  41%|████▏     | 2669/6470 [33:35<41:29,  1.53it/s]
1|2670|Loss: 0.8154118061065674:  41%|████▏     | 2670/6470 [33:35<41:56,  1.51it/s]
1|2671|Loss: 0.8433045148849487:  41%|████▏     | 2670/6470 [33:36<41:56,  1.51it/s]
1|2671|Loss: 0.8433045148849487:  41%|████▏     | 2671/6470 [33:36<40:16,  1.57it/s]
1|2672|Loss: 1.144457459449768:  41%|████▏     | 2671/6470 [33:36<40:16,  1.57it/s] 
1|2672|Loss: 1.144457459449768:  41%|████▏     | 2672/6470 [33:36<40:56,  1.55it/s]
1|2673|Loss: 0.9569698572158813:  41%|████▏     | 2672/6470 [33:37<40:56,  1.55it/s]
1|2673|Loss: 0.9569698572158813:  41%|████▏     | 2673/6470 [33:37<43:22,  1.46it/s]
1|2674|Loss: 0.9204227924346924:  41%|████▏     | 2673/6470 [33:38<43:22,  1.46it/s]
1|2674|Loss: 0.9204227924346924:  41%|████▏     | 2674/6470 [33:38<42:51,  1.48it/s]
1|2675|Loss: 0.7928675413131714:  41%|████▏     | 2674/6470 [33:38<42:51,  1.48it/s]
1|2675|Loss: 0.7928675413131714:  41%|████▏     | 2675/6470 [33:38<39:13,  1.61it/s]
1|2676|Loss: 1.0209852457046509:  41%|████▏     | 2675/6470 [33:39<39:13,  1.61it/s]
1|2676|Loss: 1.0209852457046509:  41%|████▏     | 2676/6470 [33:39<42:55,  1.47it/s]
1|2677|Loss: 0.9417745471000671:  41%|████▏     | 2676/6470 [33:40<42:55,  1.47it/s]
1|2677|Loss: 0.9417745471000671:  41%|████▏     | 2677/6470 [33:40<39:04,  1.62it/s]
1|2678|Loss: 0.9734390377998352:  41%|████▏     | 2677/6470 [33:40<39:04,  1.62it/s]
1|2678|Loss: 0.9734390377998352:  41%|████▏     | 2678/6470 [33:40<39:36,  1.60it/s]
1|2679|Loss: 1.006555199623108:  41%|████▏     | 2678/6470 [33:41<39:36,  1.60it/s] 
1|2679|Loss: 1.006555199623108:  41%|████▏     | 2679/6470 [33:41<41:00,  1.54it/s]
1|2680|Loss: 0.8866027593612671:  41%|████▏     | 2679/6470 [33:41<41:00,  1.54it/s]
1|2680|Loss: 0.8866027593612671:  41%|████▏     | 2680/6470 [33:42<42:06,  1.50it/s]
1|2681|Loss: 0.7819275259971619:  41%|████▏     | 2680/6470 [33:42<42:06,  1.50it/s]
1|2681|Loss: 0.7819275259971619:  41%|████▏     | 2681/6470 [33:43<44:06,  1.43it/s]
1|2682|Loss: 0.9516264200210571:  41%|████▏     | 2681/6470 [33:43<44:06,  1.43it/s]
1|2682|Loss: 0.9516264200210571:  41%|████▏     | 2682/6470 [33:43<41:53,  1.51it/s]
1|2683|Loss: 0.7254716753959656:  41%|████▏     | 2682/6470 [33:43<41:53,  1.51it/s]
1|2683|Loss: 0.7254716753959656:  41%|████▏     | 2683/6470 [33:44<42:34,  1.48it/s]
1|2684|Loss: 0.883431613445282:  41%|████▏     | 2683/6470 [33:44<42:34,  1.48it/s] 
1|2684|Loss: 0.883431613445282:  41%|████▏     | 2684/6470 [33:45<44:44,  1.41it/s]
1|2685|Loss: 0.8854581117630005:  41%|████▏     | 2684/6470 [33:45<44:44,  1.41it/s]
1|2685|Loss: 0.8854581117630005:  41%|████▏     | 2685/6470 [33:45<42:22,  1.49it/s]
1|2686|Loss: 0.8807082176208496:  41%|████▏     | 2685/6470 [33:46<42:22,  1.49it/s]
1|2686|Loss: 0.8807082176208496:  42%|████▏     | 2686/6470 [33:46<42:53,  1.47it/s]
1|2687|Loss: 0.8622177839279175:  42%|████▏     | 2686/6470 [33:46<42:53,  1.47it/s]
1|2687|Loss: 0.8622177839279175:  42%|████▏     | 2687/6470 [33:47<43:55,  1.44it/s]
1|2688|Loss: 0.8038275837898254:  42%|████▏     | 2687/6470 [33:47<43:55,  1.44it/s]
1|2688|Loss: 0.8038275837898254:  42%|████▏     | 2688/6470 [33:47<43:14,  1.46it/s]
1|2689|Loss: 0.8937969207763672:  42%|████▏     | 2688/6470 [33:48<43:14,  1.46it/s]
1|2689|Loss: 0.8937969207763672:  42%|████▏     | 2689/6470 [33:48<46:19,  1.36it/s]
1|2690|Loss: 1.1379070281982422:  42%|████▏     | 2689/6470 [33:49<46:19,  1.36it/s]
1|2690|Loss: 1.1379070281982422:  42%|████▏     | 2690/6470 [33:49<48:19,  1.30it/s]
1|2691|Loss: 0.9206724166870117:  42%|████▏     | 2690/6470 [33:49<48:19,  1.30it/s]
1|2691|Loss: 0.9206724166870117:  42%|████▏     | 2691/6470 [33:50<49:35,  1.27it/s]
1|2692|Loss: 0.8551837205886841:  42%|████▏     | 2691/6470 [33:50<49:35,  1.27it/s]
1|2692|Loss: 0.8551837205886841:  42%|████▏     | 2692/6470 [33:51<49:10,  1.28it/s]
1|2693|Loss: 0.6863724589347839:  42%|████▏     | 2692/6470 [33:51<49:10,  1.28it/s]
1|2693|Loss: 0.6863724589347839:  42%|████▏     | 2693/6470 [33:51<46:10,  1.36it/s]
1|2694|Loss: 1.128207802772522:  42%|████▏     | 2693/6470 [33:52<46:10,  1.36it/s] 
1|2694|Loss: 1.128207802772522:  42%|████▏     | 2694/6470 [33:52<45:01,  1.40it/s]
1|2695|Loss: 0.9940139651298523:  42%|████▏     | 2694/6470 [33:52<45:01,  1.40it/s]
1|2695|Loss: 0.9940139651298523:  42%|████▏     | 2695/6470 [33:53<46:38,  1.35it/s]
1|2696|Loss: 0.9643833041191101:  42%|████▏     | 2695/6470 [33:53<46:38,  1.35it/s]
1|2696|Loss: 0.9643833041191101:  42%|████▏     | 2696/6470 [33:53<45:13,  1.39it/s]
1|2697|Loss: 0.8785470128059387:  42%|████▏     | 2696/6470 [33:54<45:13,  1.39it/s]
1|2697|Loss: 0.8785470128059387:  42%|████▏     | 2697/6470 [33:54<46:27,  1.35it/s]
1|2698|Loss: 0.93543940782547:  42%|████▏     | 2697/6470 [33:55<46:27,  1.35it/s]  
1|2698|Loss: 0.93543940782547:  42%|████▏     | 2698/6470 [33:55<47:36,  1.32it/s]
1|2699|Loss: 0.9343580007553101:  42%|████▏     | 2698/6470 [33:55<47:36,  1.32it/s]
1|2699|Loss: 0.9343580007553101:  42%|████▏     | 2699/6470 [33:56<49:16,  1.28it/s]
1|2700|Loss: 0.9141738414764404:  42%|████▏     | 2699/6470 [33:56<49:16,  1.28it/s]
1|2700|Loss: 0.9141738414764404:  42%|████▏     | 2700/6470 [33:56<47:17,  1.33it/s]
1|2701|Loss: 1.0699195861816406:  42%|████▏     | 2700/6470 [33:57<47:17,  1.33it/s]
1|2701|Loss: 1.0699195861816406:  42%|████▏     | 2701/6470 [33:57<47:10,  1.33it/s]
1|2702|Loss: 0.7544673085212708:  42%|████▏     | 2701/6470 [33:58<47:10,  1.33it/s]
1|2702|Loss: 0.7544673085212708:  42%|████▏     | 2702/6470 [33:58<47:07,  1.33it/s]
1|2703|Loss: 0.9296188354492188:  42%|████▏     | 2702/6470 [33:58<47:07,  1.33it/s]
1|2703|Loss: 0.9296188354492188:  42%|████▏     | 2703/6470 [33:59<47:06,  1.33it/s]
1|2704|Loss: 0.7714064121246338:  42%|████▏     | 2703/6470 [33:59<47:06,  1.33it/s]
1|2704|Loss: 0.7714064121246338:  42%|████▏     | 2704/6470 [33:59<46:28,  1.35it/s]
1|2705|Loss: 0.9182571172714233:  42%|████▏     | 2704/6470 [34:00<46:28,  1.35it/s]
1|2705|Loss: 0.9182571172714233:  42%|████▏     | 2705/6470 [34:00<40:35,  1.55it/s]
1|2706|Loss: 0.6767537593841553:  42%|████▏     | 2705/6470 [34:00<40:35,  1.55it/s]
1|2706|Loss: 0.6767537593841553:  42%|████▏     | 2706/6470 [34:01<44:48,  1.40it/s]
1|2707|Loss: 0.800349771976471:  42%|████▏     | 2706/6470 [34:01<44:48,  1.40it/s] 
1|2707|Loss: 0.800349771976471:  42%|████▏     | 2707/6470 [34:01<43:02,  1.46it/s]
1|2708|Loss: 0.9317166209220886:  42%|████▏     | 2707/6470 [34:02<43:02,  1.46it/s]
1|2708|Loss: 0.9317166209220886:  42%|████▏     | 2708/6470 [34:02<45:23,  1.38it/s]
1|2709|Loss: 0.9667448997497559:  42%|████▏     | 2708/6470 [34:03<45:23,  1.38it/s]
1|2709|Loss: 0.9667448997497559:  42%|████▏     | 2709/6470 [34:03<44:27,  1.41it/s]
1|2710|Loss: 0.9452947974205017:  42%|████▏     | 2709/6470 [34:03<44:27,  1.41it/s]
1|2710|Loss: 0.9452947974205017:  42%|████▏     | 2710/6470 [34:03<42:00,  1.49it/s]
1|2711|Loss: 0.8672528862953186:  42%|████▏     | 2710/6470 [34:04<42:00,  1.49it/s]
1|2711|Loss: 0.8672528862953186:  42%|████▏     | 2711/6470 [34:04<40:55,  1.53it/s]
1|2712|Loss: 0.8259577751159668:  42%|████▏     | 2711/6470 [34:04<40:55,  1.53it/s]
1|2712|Loss: 0.8259577751159668:  42%|████▏     | 2712/6470 [34:05<42:25,  1.48it/s]
1|2713|Loss: 0.8458147644996643:  42%|████▏     | 2712/6470 [34:05<42:25,  1.48it/s]
1|2713|Loss: 0.8458147644996643:  42%|████▏     | 2713/6470 [34:05<43:13,  1.45it/s]
1|2714|Loss: 0.9440702199935913:  42%|████▏     | 2713/6470 [34:06<43:13,  1.45it/s]
1|2714|Loss: 0.9440702199935913:  42%|████▏     | 2714/6470 [34:06<43:52,  1.43it/s]
1|2715|Loss: 0.8994355797767639:  42%|████▏     | 2714/6470 [34:07<43:52,  1.43it/s]
1|2715|Loss: 0.8994355797767639:  42%|████▏     | 2715/6470 [34:07<42:13,  1.48it/s]
1|2716|Loss: 0.8971844911575317:  42%|████▏     | 2715/6470 [34:07<42:13,  1.48it/s]
1|2716|Loss: 0.8971844911575317:  42%|████▏     | 2716/6470 [34:08<46:39,  1.34it/s]
1|2717|Loss: 0.9128604531288147:  42%|████▏     | 2716/6470 [34:08<46:39,  1.34it/s]
1|2717|Loss: 0.9128604531288147:  42%|████▏     | 2717/6470 [34:08<44:55,  1.39it/s]
1|2718|Loss: 0.9134191274642944:  42%|████▏     | 2717/6470 [34:09<44:55,  1.39it/s]
1|2718|Loss: 0.9134191274642944:  42%|████▏     | 2718/6470 [34:09<39:23,  1.59it/s]
1|2719|Loss: 0.929454505443573:  42%|████▏     | 2718/6470 [34:09<39:23,  1.59it/s] 
1|2719|Loss: 0.929454505443573:  42%|████▏     | 2719/6470 [34:09<40:22,  1.55it/s]
1|2720|Loss: 0.8462716341018677:  42%|████▏     | 2719/6470 [34:10<40:22,  1.55it/s]
1|2720|Loss: 0.8462716341018677:  42%|████▏     | 2720/6470 [34:10<39:17,  1.59it/s]
1|2721|Loss: 1.0163434743881226:  42%|████▏     | 2720/6470 [34:10<39:17,  1.59it/s]
1|2721|Loss: 1.0163434743881226:  42%|████▏     | 2721/6470 [34:11<40:11,  1.55it/s]
1|2722|Loss: 0.9381470084190369:  42%|████▏     | 2721/6470 [34:11<40:11,  1.55it/s]
1|2722|Loss: 0.9381470084190369:  42%|████▏     | 2722/6470 [34:11<40:07,  1.56it/s]
1|2723|Loss: 0.9757487773895264:  42%|████▏     | 2722/6470 [34:12<40:07,  1.56it/s]
1|2723|Loss: 0.9757487773895264:  42%|████▏     | 2723/6470 [34:12<38:37,  1.62it/s]
1|2724|Loss: 0.8939334154129028:  42%|████▏     | 2723/6470 [34:12<38:37,  1.62it/s]
1|2724|Loss: 0.8939334154129028:  42%|████▏     | 2724/6470 [34:13<40:08,  1.56it/s]
1|2725|Loss: 0.8315796852111816:  42%|████▏     | 2724/6470 [34:13<40:08,  1.56it/s]
1|2725|Loss: 0.8315796852111816:  42%|████▏     | 2725/6470 [34:13<40:21,  1.55it/s]
1|2726|Loss: 1.0496634244918823:  42%|████▏     | 2725/6470 [34:14<40:21,  1.55it/s]
1|2726|Loss: 1.0496634244918823:  42%|████▏     | 2726/6470 [34:14<38:15,  1.63it/s]
1|2727|Loss: 0.820803701877594:  42%|████▏     | 2726/6470 [34:14<38:15,  1.63it/s] 
1|2727|Loss: 0.820803701877594:  42%|████▏     | 2727/6470 [34:15<39:44,  1.57it/s]
1|2728|Loss: 0.8832170367240906:  42%|████▏     | 2727/6470 [34:15<39:44,  1.57it/s]
1|2728|Loss: 0.8832170367240906:  42%|████▏     | 2728/6470 [34:15<41:36,  1.50it/s]
1|2729|Loss: 0.8583398461341858:  42%|████▏     | 2728/6470 [34:16<41:36,  1.50it/s]
1|2729|Loss: 0.8583398461341858:  42%|████▏     | 2729/6470 [34:16<40:24,  1.54it/s]
1|2730|Loss: 0.7871887683868408:  42%|████▏     | 2729/6470 [34:16<40:24,  1.54it/s]
1|2730|Loss: 0.7871887683868408:  42%|████▏     | 2730/6470 [34:17<41:23,  1.51it/s]
1|2731|Loss: 1.0491083860397339:  42%|████▏     | 2730/6470 [34:17<41:23,  1.51it/s]
1|2731|Loss: 1.0491083860397339:  42%|████▏     | 2731/6470 [34:17<42:21,  1.47it/s]
1|2732|Loss: 0.856845498085022:  42%|████▏     | 2731/6470 [34:18<42:21,  1.47it/s] 
1|2732|Loss: 0.856845498085022:  42%|████▏     | 2732/6470 [34:18<41:50,  1.49it/s]
1|2733|Loss: 1.0144178867340088:  42%|████▏     | 2732/6470 [34:18<41:50,  1.49it/s]
1|2733|Loss: 1.0144178867340088:  42%|████▏     | 2733/6470 [34:19<41:02,  1.52it/s]
1|2734|Loss: 0.9610360860824585:  42%|████▏     | 2733/6470 [34:19<41:02,  1.52it/s]
1|2734|Loss: 0.9610360860824585:  42%|████▏     | 2734/6470 [34:19<40:42,  1.53it/s]
1|2735|Loss: 0.9106225967407227:  42%|████▏     | 2734/6470 [34:20<40:42,  1.53it/s]
1|2735|Loss: 0.9106225967407227:  42%|████▏     | 2735/6470 [34:20<39:09,  1.59it/s]
1|2736|Loss: 0.9746147394180298:  42%|████▏     | 2735/6470 [34:20<39:09,  1.59it/s]
1|2736|Loss: 0.9746147394180298:  42%|████▏     | 2736/6470 [34:20<40:42,  1.53it/s]
1|2737|Loss: 0.9936037659645081:  42%|████▏     | 2736/6470 [34:21<40:42,  1.53it/s]
1|2737|Loss: 0.9936037659645081:  42%|████▏     | 2737/6470 [34:21<41:55,  1.48it/s]
1|2738|Loss: 0.9147449731826782:  42%|████▏     | 2737/6470 [34:22<41:55,  1.48it/s]
1|2738|Loss: 0.9147449731826782:  42%|████▏     | 2738/6470 [34:22<43:16,  1.44it/s]
1|2739|Loss: 0.8311029076576233:  42%|████▏     | 2738/6470 [34:22<43:16,  1.44it/s]
1|2739|Loss: 0.8311029076576233:  42%|████▏     | 2739/6470 [34:23<44:11,  1.41it/s]
1|2740|Loss: 1.0083258152008057:  42%|████▏     | 2739/6470 [34:23<44:11,  1.41it/s]
1|2740|Loss: 1.0083258152008057:  42%|████▏     | 2740/6470 [34:23<44:49,  1.39it/s]
1|2741|Loss: 0.790368378162384:  42%|████▏     | 2740/6470 [34:24<44:49,  1.39it/s] 
1|2741|Loss: 0.790368378162384:  42%|████▏     | 2741/6470 [34:24<41:50,  1.49it/s]
1|2742|Loss: 0.9700278043746948:  42%|████▏     | 2741/6470 [34:24<41:50,  1.49it/s]
1|2742|Loss: 0.9700278043746948:  42%|████▏     | 2742/6470 [34:25<42:15,  1.47it/s]
1|2743|Loss: 0.9308763146400452:  42%|████▏     | 2742/6470 [34:25<42:15,  1.47it/s]
1|2743|Loss: 0.9308763146400452:  42%|████▏     | 2743/6470 [34:25<43:30,  1.43it/s]
1|2744|Loss: 0.8812099695205688:  42%|████▏     | 2743/6470 [34:26<43:30,  1.43it/s]
1|2744|Loss: 0.8812099695205688:  42%|████▏     | 2744/6470 [34:26<42:51,  1.45it/s]
1|2745|Loss: 0.9060776233673096:  42%|████▏     | 2744/6470 [34:26<42:51,  1.45it/s]
1|2745|Loss: 0.9060776233673096:  42%|████▏     | 2745/6470 [34:27<39:11,  1.58it/s]
1|2746|Loss: 0.9195565581321716:  42%|████▏     | 2745/6470 [34:27<39:11,  1.58it/s]
1|2746|Loss: 0.9195565581321716:  42%|████▏     | 2746/6470 [34:27<39:51,  1.56it/s]
1|2747|Loss: 0.9316311478614807:  42%|████▏     | 2746/6470 [34:28<39:51,  1.56it/s]
1|2747|Loss: 0.9316311478614807:  42%|████▏     | 2747/6470 [34:28<43:20,  1.43it/s]
1|2748|Loss: 0.8326635956764221:  42%|████▏     | 2747/6470 [34:29<43:20,  1.43it/s]
1|2748|Loss: 0.8326635956764221:  42%|████▏     | 2748/6470 [34:29<44:40,  1.39it/s]
1|2749|Loss: 1.0543277263641357:  42%|████▏     | 2748/6470 [34:29<44:40,  1.39it/s]
1|2749|Loss: 1.0543277263641357:  42%|████▏     | 2749/6470 [34:30<43:19,  1.43it/s]
1|2750|Loss: 0.8938536047935486:  42%|████▏     | 2749/6470 [34:30<43:19,  1.43it/s]
1|2750|Loss: 0.8938536047935486:  43%|████▎     | 2750/6470 [34:30<41:44,  1.49it/s]
1|2751|Loss: 0.9922500848770142:  43%|████▎     | 2750/6470 [34:31<41:44,  1.49it/s]
1|2751|Loss: 0.9922500848770142:  43%|████▎     | 2751/6470 [34:31<42:34,  1.46it/s]
1|2752|Loss: 0.7249499559402466:  43%|████▎     | 2751/6470 [34:31<42:34,  1.46it/s]
1|2752|Loss: 0.7249499559402466:  43%|████▎     | 2752/6470 [34:31<41:11,  1.50it/s]
1|2753|Loss: 0.894596517086029:  43%|████▎     | 2752/6470 [34:32<41:11,  1.50it/s] 
1|2753|Loss: 0.894596517086029:  43%|████▎     | 2753/6470 [34:32<42:10,  1.47it/s]
1|2754|Loss: 1.006847858428955:  43%|████▎     | 2753/6470 [34:33<42:10,  1.47it/s]
1|2754|Loss: 1.006847858428955:  43%|████▎     | 2754/6470 [34:33<43:13,  1.43it/s]
1|2755|Loss: 0.9402382373809814:  43%|████▎     | 2754/6470 [34:33<43:13,  1.43it/s]
1|2755|Loss: 0.9402382373809814:  43%|████▎     | 2755/6470 [34:34<44:56,  1.38it/s]
1|2756|Loss: 0.9943327903747559:  43%|████▎     | 2755/6470 [34:34<44:56,  1.38it/s]
1|2756|Loss: 0.9943327903747559:  43%|████▎     | 2756/6470 [34:34<45:22,  1.36it/s]
1|2757|Loss: 0.9679815769195557:  43%|████▎     | 2756/6470 [34:35<45:22,  1.36it/s]
1|2757|Loss: 0.9679815769195557:  43%|████▎     | 2757/6470 [34:35<44:10,  1.40it/s]
1|2758|Loss: 0.7227758765220642:  43%|████▎     | 2757/6470 [34:36<44:10,  1.40it/s]
1|2758|Loss: 0.7227758765220642:  43%|████▎     | 2758/6470 [34:36<46:02,  1.34it/s]
1|2759|Loss: 0.8703663349151611:  43%|████▎     | 2758/6470 [34:36<46:02,  1.34it/s]
1|2759|Loss: 0.8703663349151611:  43%|████▎     | 2759/6470 [34:37<48:27,  1.28it/s]
1|2760|Loss: 0.9642302989959717:  43%|████▎     | 2759/6470 [34:37<48:27,  1.28it/s]
1|2760|Loss: 0.9642302989959717:  43%|████▎     | 2760/6470 [34:38<50:30,  1.22it/s]
1|2761|Loss: 0.9663407802581787:  43%|████▎     | 2760/6470 [34:38<50:30,  1.22it/s]
1|2761|Loss: 0.9663407802581787:  43%|████▎     | 2761/6470 [34:39<51:27,  1.20it/s]
1|2762|Loss: 0.8759461641311646:  43%|████▎     | 2761/6470 [34:39<51:27,  1.20it/s]
1|2762|Loss: 0.8759461641311646:  43%|████▎     | 2762/6470 [34:39<48:29,  1.27it/s]
1|2763|Loss: 0.8706518411636353:  43%|████▎     | 2762/6470 [34:40<48:29,  1.27it/s]
1|2763|Loss: 0.8706518411636353:  43%|████▎     | 2763/6470 [34:40<47:19,  1.31it/s]
1|2764|Loss: 0.880525529384613:  43%|████▎     | 2763/6470 [34:40<47:19,  1.31it/s] 
1|2764|Loss: 0.880525529384613:  43%|████▎     | 2764/6470 [34:41<45:07,  1.37it/s]
1|2765|Loss: 1.0488433837890625:  43%|████▎     | 2764/6470 [34:41<45:07,  1.37it/s]
1|2765|Loss: 1.0488433837890625:  43%|████▎     | 2765/6470 [34:41<43:16,  1.43it/s]
1|2766|Loss: 0.8528075218200684:  43%|████▎     | 2765/6470 [34:42<43:16,  1.43it/s]
1|2766|Loss: 0.8528075218200684:  43%|████▎     | 2766/6470 [34:42<40:19,  1.53it/s]
1|2767|Loss: 0.9368643164634705:  43%|████▎     | 2766/6470 [34:42<40:19,  1.53it/s]
1|2767|Loss: 0.9368643164634705:  43%|████▎     | 2767/6470 [34:42<40:53,  1.51it/s]
1|2768|Loss: 1.1260721683502197:  43%|████▎     | 2767/6470 [34:43<40:53,  1.51it/s]
1|2768|Loss: 1.1260721683502197:  43%|████▎     | 2768/6470 [34:43<38:08,  1.62it/s]
1|2769|Loss: 0.878705620765686:  43%|████▎     | 2768/6470 [34:43<38:08,  1.62it/s] 
1|2769|Loss: 0.878705620765686:  43%|████▎     | 2769/6470 [34:44<36:54,  1.67it/s]
1|2770|Loss: 0.8463329672813416:  43%|████▎     | 2769/6470 [34:44<36:54,  1.67it/s]
1|2770|Loss: 0.8463329672813416:  43%|████▎     | 2770/6470 [34:44<37:31,  1.64it/s]
1|2771|Loss: 0.6191433668136597:  43%|████▎     | 2770/6470 [34:45<37:31,  1.64it/s]
1|2771|Loss: 0.6191433668136597:  43%|████▎     | 2771/6470 [34:45<39:49,  1.55it/s]
1|2772|Loss: 0.8332898020744324:  43%|████▎     | 2771/6470 [34:45<39:49,  1.55it/s]
1|2772|Loss: 0.8332898020744324:  43%|████▎     | 2772/6470 [34:46<41:30,  1.48it/s]
1|2773|Loss: 0.9282003045082092:  43%|████▎     | 2772/6470 [34:46<41:30,  1.48it/s]
1|2773|Loss: 0.9282003045082092:  43%|████▎     | 2773/6470 [34:46<42:34,  1.45it/s]
1|2774|Loss: 0.9602617621421814:  43%|████▎     | 2773/6470 [34:47<42:34,  1.45it/s]
1|2774|Loss: 0.9602617621421814:  43%|████▎     | 2774/6470 [34:47<42:00,  1.47it/s]
1|2775|Loss: 0.913640022277832:  43%|████▎     | 2774/6470 [34:47<42:00,  1.47it/s] 
1|2775|Loss: 0.913640022277832:  43%|████▎     | 2775/6470 [34:48<43:41,  1.41it/s]
1|2776|Loss: 0.9997879266738892:  43%|████▎     | 2775/6470 [34:48<43:41,  1.41it/s]
1|2776|Loss: 0.9997879266738892:  43%|████▎     | 2776/6470 [34:48<43:02,  1.43it/s]
1|2777|Loss: 0.9016956686973572:  43%|████▎     | 2776/6470 [34:49<43:02,  1.43it/s]
1|2777|Loss: 0.9016956686973572:  43%|████▎     | 2777/6470 [34:49<40:54,  1.50it/s]
1|2778|Loss: 0.9000774621963501:  43%|████▎     | 2777/6470 [34:49<40:54,  1.50it/s]
1|2778|Loss: 0.9000774621963501:  43%|████▎     | 2778/6470 [34:50<40:20,  1.52it/s]
1|2779|Loss: 1.0207499265670776:  43%|████▎     | 2778/6470 [34:50<40:20,  1.52it/s]
1|2779|Loss: 1.0207499265670776:  43%|████▎     | 2779/6470 [34:51<43:49,  1.40it/s]
1|2780|Loss: 0.9252758026123047:  43%|████▎     | 2779/6470 [34:51<43:49,  1.40it/s]
1|2780|Loss: 0.9252758026123047:  43%|████▎     | 2780/6470 [34:51<44:53,  1.37it/s]
1|2781|Loss: 0.8917979598045349:  43%|████▎     | 2780/6470 [34:52<44:53,  1.37it/s]
1|2781|Loss: 0.8917979598045349:  43%|████▎     | 2781/6470 [34:52<45:09,  1.36it/s]
1|2782|Loss: 0.8741007447242737:  43%|████▎     | 2781/6470 [34:52<45:09,  1.36it/s]
1|2782|Loss: 0.8741007447242737:  43%|████▎     | 2782/6470 [34:53<39:39,  1.55it/s]
1|2783|Loss: 0.8585713505744934:  43%|████▎     | 2782/6470 [34:53<39:39,  1.55it/s]
1|2783|Loss: 0.8585713505744934:  43%|████▎     | 2783/6470 [34:53<40:03,  1.53it/s]
1|2784|Loss: 0.9646031856536865:  43%|████▎     | 2783/6470 [34:54<40:03,  1.53it/s]
1|2784|Loss: 0.9646031856536865:  43%|████▎     | 2784/6470 [34:54<39:52,  1.54it/s]
1|2785|Loss: 0.7484026551246643:  43%|████▎     | 2784/6470 [34:54<39:52,  1.54it/s]
1|2785|Loss: 0.7484026551246643:  43%|████▎     | 2785/6470 [34:55<40:53,  1.50it/s]
1|2786|Loss: 0.8602272868156433:  43%|████▎     | 2785/6470 [34:55<40:53,  1.50it/s]
1|2786|Loss: 0.8602272868156433:  43%|████▎     | 2786/6470 [34:55<40:29,  1.52it/s]
1|2787|Loss: 0.8657581806182861:  43%|████▎     | 2786/6470 [34:56<40:29,  1.52it/s]
1|2787|Loss: 0.8657581806182861:  43%|████▎     | 2787/6470 [34:56<43:30,  1.41it/s]
1|2788|Loss: 0.7934651970863342:  43%|████▎     | 2787/6470 [34:56<43:30,  1.41it/s]
1|2788|Loss: 0.7934651970863342:  43%|████▎     | 2788/6470 [34:57<45:22,  1.35it/s]
1|2789|Loss: 0.985352635383606:  43%|████▎     | 2788/6470 [34:57<45:22,  1.35it/s] 
1|2789|Loss: 0.985352635383606:  43%|████▎     | 2789/6470 [34:57<43:46,  1.40it/s]
1|2790|Loss: 0.940428614616394:  43%|████▎     | 2789/6470 [34:58<43:46,  1.40it/s]
1|2790|Loss: 0.940428614616394:  43%|████▎     | 2790/6470 [34:58<44:31,  1.38it/s]
1|2791|Loss: 0.8204810619354248:  43%|████▎     | 2790/6470 [34:59<44:31,  1.38it/s]
1|2791|Loss: 0.8204810619354248:  43%|████▎     | 2791/6470 [34:59<46:49,  1.31it/s]
1|2792|Loss: 0.9130856394767761:  43%|████▎     | 2791/6470 [34:59<46:49,  1.31it/s]
1|2792|Loss: 0.9130856394767761:  43%|████▎     | 2792/6470 [35:00<45:09,  1.36it/s]
1|2793|Loss: 0.9754824638366699:  43%|████▎     | 2792/6470 [35:00<45:09,  1.36it/s]
1|2793|Loss: 0.9754824638366699:  43%|████▎     | 2793/6470 [35:00<43:11,  1.42it/s]
1|2794|Loss: 1.0503628253936768:  43%|████▎     | 2793/6470 [35:01<43:11,  1.42it/s]
1|2794|Loss: 1.0503628253936768:  43%|████▎     | 2794/6470 [35:01<42:59,  1.43it/s]
1|2795|Loss: 0.877156138420105:  43%|████▎     | 2794/6470 [35:01<42:59,  1.43it/s] 
1|2795|Loss: 0.877156138420105:  43%|████▎     | 2795/6470 [35:02<41:51,  1.46it/s]
1|2796|Loss: 0.9005438089370728:  43%|████▎     | 2795/6470 [35:02<41:51,  1.46it/s]
1|2796|Loss: 0.9005438089370728:  43%|████▎     | 2796/6470 [35:02<42:37,  1.44it/s]
1|2797|Loss: 1.034732460975647:  43%|████▎     | 2796/6470 [35:03<42:37,  1.44it/s] 
1|2797|Loss: 1.034732460975647:  43%|████▎     | 2797/6470 [35:03<41:21,  1.48it/s]
1|2798|Loss: 1.001232385635376:  43%|████▎     | 2797/6470 [35:03<41:21,  1.48it/s]
1|2798|Loss: 1.001232385635376:  43%|████▎     | 2798/6470 [35:04<39:45,  1.54it/s]
1|2799|Loss: 0.8533489108085632:  43%|████▎     | 2798/6470 [35:04<39:45,  1.54it/s]
1|2799|Loss: 0.8533489108085632:  43%|████▎     | 2799/6470 [35:04<40:46,  1.50it/s]
1|2800|Loss: 0.9187789559364319:  43%|████▎     | 2799/6470 [35:05<40:46,  1.50it/s]
1|2800|Loss: 0.9187789559364319:  43%|████▎     | 2800/6470 [35:05<40:28,  1.51it/s]
1|2801|Loss: 1.0148965120315552:  43%|████▎     | 2800/6470 [35:05<40:28,  1.51it/s]
1|2801|Loss: 1.0148965120315552:  43%|████▎     | 2801/6470 [35:06<42:37,  1.43it/s]
1|2802|Loss: 0.9250271320343018:  43%|████▎     | 2801/6470 [35:06<42:37,  1.43it/s]
1|2802|Loss: 0.9250271320343018:  43%|████▎     | 2802/6470 [35:07<45:27,  1.34it/s]
1|2803|Loss: 0.8331087827682495:  43%|████▎     | 2802/6470 [35:07<45:27,  1.34it/s]
1|2803|Loss: 0.8331087827682495:  43%|████▎     | 2803/6470 [35:07<44:30,  1.37it/s]
1|2804|Loss: 0.9774482250213623:  43%|████▎     | 2803/6470 [35:08<44:30,  1.37it/s]
1|2804|Loss: 0.9774482250213623:  43%|████▎     | 2804/6470 [35:08<41:32,  1.47it/s]
1|2805|Loss: 0.8194122314453125:  43%|████▎     | 2804/6470 [35:08<41:32,  1.47it/s]
1|2805|Loss: 0.8194122314453125:  43%|████▎     | 2805/6470 [35:08<38:34,  1.58it/s]
1|2806|Loss: 0.8506395220756531:  43%|████▎     | 2805/6470 [35:09<38:34,  1.58it/s]
1|2806|Loss: 0.8506395220756531:  43%|████▎     | 2806/6470 [35:09<39:34,  1.54it/s]
1|2807|Loss: 0.8844352960586548:  43%|████▎     | 2806/6470 [35:10<39:34,  1.54it/s]
1|2807|Loss: 0.8844352960586548:  43%|████▎     | 2807/6470 [35:10<41:33,  1.47it/s]
1|2808|Loss: 1.1849405765533447:  43%|████▎     | 2807/6470 [35:10<41:33,  1.47it/s]
1|2808|Loss: 1.1849405765533447:  43%|████▎     | 2808/6470 [35:11<44:35,  1.37it/s]
1|2809|Loss: 0.8351300954818726:  43%|████▎     | 2808/6470 [35:11<44:35,  1.37it/s]
1|2809|Loss: 0.8351300954818726:  43%|████▎     | 2809/6470 [35:11<42:30,  1.44it/s]
1|2810|Loss: 1.0634407997131348:  43%|████▎     | 2809/6470 [35:12<42:30,  1.44it/s]
1|2810|Loss: 1.0634407997131348:  43%|████▎     | 2810/6470 [35:12<40:22,  1.51it/s]
1|2811|Loss: 0.735832154750824:  43%|████▎     | 2810/6470 [35:12<40:22,  1.51it/s] 
1|2811|Loss: 0.735832154750824:  43%|████▎     | 2811/6470 [35:13<52:07,  1.17it/s]
1|2812|Loss: 0.8494365215301514:  43%|████▎     | 2811/6470 [35:14<52:07,  1.17it/s]
1|2812|Loss: 0.8494365215301514:  43%|████▎     | 2812/6470 [35:14<52:26,  1.16it/s]
1|2813|Loss: 0.9645867347717285:  43%|████▎     | 2812/6470 [35:14<52:26,  1.16it/s]
1|2813|Loss: 0.9645867347717285:  43%|████▎     | 2813/6470 [35:15<49:58,  1.22it/s]
1|2814|Loss: 0.9088530540466309:  43%|████▎     | 2813/6470 [35:15<49:58,  1.22it/s]
1|2814|Loss: 0.9088530540466309:  43%|████▎     | 2814/6470 [35:15<46:07,  1.32it/s]
1|2815|Loss: 1.0264211893081665:  43%|████▎     | 2814/6470 [35:16<46:07,  1.32it/s]
1|2815|Loss: 1.0264211893081665:  44%|████▎     | 2815/6470 [35:16<45:26,  1.34it/s]
1|2816|Loss: 0.89497971534729:  44%|████▎     | 2815/6470 [35:17<45:26,  1.34it/s]  
1|2816|Loss: 0.89497971534729:  44%|████▎     | 2816/6470 [35:17<44:00,  1.38it/s]
1|2817|Loss: 0.8785104155540466:  44%|████▎     | 2816/6470 [35:17<44:00,  1.38it/s]
1|2817|Loss: 0.8785104155540466:  44%|████▎     | 2817/6470 [35:17<40:33,  1.50it/s]
1|2818|Loss: 0.8665245771408081:  44%|████▎     | 2817/6470 [35:18<40:33,  1.50it/s]
1|2818|Loss: 0.8665245771408081:  44%|████▎     | 2818/6470 [35:18<40:25,  1.51it/s]
1|2819|Loss: 0.9810528755187988:  44%|████▎     | 2818/6470 [35:18<40:25,  1.51it/s]
1|2819|Loss: 0.9810528755187988:  44%|████▎     | 2819/6470 [35:19<37:31,  1.62it/s]
1|2820|Loss: 0.819683849811554:  44%|████▎     | 2819/6470 [35:19<37:31,  1.62it/s] 
1|2820|Loss: 0.819683849811554:  44%|████▎     | 2820/6470 [35:19<36:51,  1.65it/s]
1|2821|Loss: 0.8496038913726807:  44%|████▎     | 2820/6470 [35:19<36:51,  1.65it/s]
1|2821|Loss: 0.8496038913726807:  44%|████▎     | 2821/6470 [35:20<39:01,  1.56it/s]
1|2822|Loss: 0.9351957440376282:  44%|████▎     | 2821/6470 [35:20<39:01,  1.56it/s]
1|2822|Loss: 0.9351957440376282:  44%|████▎     | 2822/6470 [35:20<39:06,  1.55it/s]
1|2823|Loss: 0.9119895696640015:  44%|████▎     | 2822/6470 [35:21<39:06,  1.55it/s]
1|2823|Loss: 0.9119895696640015:  44%|████▎     | 2823/6470 [35:21<38:37,  1.57it/s]
1|2824|Loss: 0.9749087691307068:  44%|████▎     | 2823/6470 [35:21<38:37,  1.57it/s]
1|2824|Loss: 0.9749087691307068:  44%|████▎     | 2824/6470 [35:22<40:51,  1.49it/s]
1|2825|Loss: 0.8855025768280029:  44%|████▎     | 2824/6470 [35:22<40:51,  1.49it/s]
1|2825|Loss: 0.8855025768280029:  44%|████▎     | 2825/6470 [35:23<41:40,  1.46it/s]
1|2826|Loss: 0.8191432356834412:  44%|████▎     | 2825/6470 [35:23<41:40,  1.46it/s]
1|2826|Loss: 0.8191432356834412:  44%|████▎     | 2826/6470 [35:23<40:07,  1.51it/s]
1|2827|Loss: 0.8279743790626526:  44%|████▎     | 2826/6470 [35:24<40:07,  1.51it/s]
1|2827|Loss: 0.8279743790626526:  44%|████▎     | 2827/6470 [35:24<41:46,  1.45it/s]
1|2828|Loss: 0.9557272791862488:  44%|████▎     | 2827/6470 [35:24<41:46,  1.45it/s]
1|2828|Loss: 0.9557272791862488:  44%|████▎     | 2828/6470 [35:24<39:59,  1.52it/s]
1|2829|Loss: 1.009114384651184:  44%|████▎     | 2828/6470 [35:25<39:59,  1.52it/s] 
1|2829|Loss: 1.009114384651184:  44%|████▎     | 2829/6470 [35:25<38:25,  1.58it/s]
1|2830|Loss: 0.7900238633155823:  44%|████▎     | 2829/6470 [35:25<38:25,  1.58it/s]
1|2830|Loss: 0.7900238633155823:  44%|████▎     | 2830/6470 [35:26<38:13,  1.59it/s]
1|2831|Loss: 0.8368332982063293:  44%|████▎     | 2830/6470 [35:26<38:13,  1.59it/s]
1|2831|Loss: 0.8368332982063293:  44%|████▍     | 2831/6470 [35:26<39:25,  1.54it/s]
1|2832|Loss: 0.7766968607902527:  44%|████▍     | 2831/6470 [35:27<39:25,  1.54it/s]
1|2832|Loss: 0.7766968607902527:  44%|████▍     | 2832/6470 [35:27<41:14,  1.47it/s]
1|2833|Loss: 0.7943408489227295:  44%|████▍     | 2832/6470 [35:28<41:14,  1.47it/s]
1|2833|Loss: 0.7943408489227295:  44%|████▍     | 2833/6470 [35:28<40:42,  1.49it/s]
1|2834|Loss: 1.0065215826034546:  44%|████▍     | 2833/6470 [35:28<40:42,  1.49it/s]
1|2834|Loss: 1.0065215826034546:  44%|████▍     | 2834/6470 [35:28<40:53,  1.48it/s]
1|2835|Loss: 0.9752010703086853:  44%|████▍     | 2834/6470 [35:29<40:53,  1.48it/s]
1|2835|Loss: 0.9752010703086853:  44%|████▍     | 2835/6470 [35:29<41:59,  1.44it/s]
1|2836|Loss: 0.9514222145080566:  44%|████▍     | 2835/6470 [35:30<41:59,  1.44it/s]
1|2836|Loss: 0.9514222145080566:  44%|████▍     | 2836/6470 [35:30<44:30,  1.36it/s]
1|2837|Loss: 0.8061133623123169:  44%|████▍     | 2836/6470 [35:30<44:30,  1.36it/s]
1|2837|Loss: 0.8061133623123169:  44%|████▍     | 2837/6470 [35:31<44:59,  1.35it/s]
1|2838|Loss: 0.8825411796569824:  44%|████▍     | 2837/6470 [35:31<44:59,  1.35it/s]
1|2838|Loss: 0.8825411796569824:  44%|████▍     | 2838/6470 [35:31<43:29,  1.39it/s]
1|2839|Loss: 0.8651230931282043:  44%|████▍     | 2838/6470 [35:32<43:29,  1.39it/s]
1|2839|Loss: 0.8651230931282043:  44%|████▍     | 2839/6470 [35:32<44:16,  1.37it/s]
1|2840|Loss: 0.9544497728347778:  44%|████▍     | 2839/6470 [35:33<44:16,  1.37it/s]
1|2840|Loss: 0.9544497728347778:  44%|████▍     | 2840/6470 [35:33<44:07,  1.37it/s]
1|2841|Loss: 0.6696673035621643:  44%|████▍     | 2840/6470 [35:33<44:07,  1.37it/s]
1|2841|Loss: 0.6696673035621643:  44%|████▍     | 2841/6470 [35:34<45:51,  1.32it/s]
1|2842|Loss: 0.8879521489143372:  44%|████▍     | 2841/6470 [35:34<45:51,  1.32it/s]
1|2842|Loss: 0.8879521489143372:  44%|████▍     | 2842/6470 [35:35<46:48,  1.29it/s]
1|2843|Loss: 1.0753450393676758:  44%|████▍     | 2842/6470 [35:35<46:48,  1.29it/s]
1|2843|Loss: 1.0753450393676758:  44%|████▍     | 2843/6470 [35:35<43:11,  1.40it/s]
1|2844|Loss: 0.8548633456230164:  44%|████▍     | 2843/6470 [35:36<43:11,  1.40it/s]
1|2844|Loss: 0.8548633456230164:  44%|████▍     | 2844/6470 [35:36<42:40,  1.42it/s]
1|2845|Loss: 0.9481654167175293:  44%|████▍     | 2844/6470 [35:36<42:40,  1.42it/s]
1|2845|Loss: 0.9481654167175293:  44%|████▍     | 2845/6470 [35:36<41:34,  1.45it/s]
1|2846|Loss: 1.0504865646362305:  44%|████▍     | 2845/6470 [35:37<41:34,  1.45it/s]
1|2846|Loss: 1.0504865646362305:  44%|████▍     | 2846/6470 [35:37<40:15,  1.50it/s]
1|2847|Loss: 0.9577855467796326:  44%|████▍     | 2846/6470 [35:37<40:15,  1.50it/s]
1|2847|Loss: 0.9577855467796326:  44%|████▍     | 2847/6470 [35:38<41:02,  1.47it/s]
1|2848|Loss: 0.8893774747848511:  44%|████▍     | 2847/6470 [35:38<41:02,  1.47it/s]
1|2848|Loss: 0.8893774747848511:  44%|████▍     | 2848/6470 [35:39<41:37,  1.45it/s]
1|2849|Loss: 0.8250121474266052:  44%|████▍     | 2848/6470 [35:39<41:37,  1.45it/s]
1|2849|Loss: 0.8250121474266052:  44%|████▍     | 2849/6470 [35:39<40:12,  1.50it/s]
1|2850|Loss: 1.0245915651321411:  44%|████▍     | 2849/6470 [35:40<40:12,  1.50it/s]
1|2850|Loss: 1.0245915651321411:  44%|████▍     | 2850/6470 [35:40<39:32,  1.53it/s]
1|2851|Loss: 0.7088829278945923:  44%|████▍     | 2850/6470 [35:40<39:32,  1.53it/s]
1|2851|Loss: 0.7088829278945923:  44%|████▍     | 2851/6470 [35:41<42:48,  1.41it/s]
1|2852|Loss: 0.8722010254859924:  44%|████▍     | 2851/6470 [35:41<42:48,  1.41it/s]
1|2852|Loss: 0.8722010254859924:  44%|████▍     | 2852/6470 [35:41<43:29,  1.39it/s]
1|2853|Loss: 0.9534031748771667:  44%|████▍     | 2852/6470 [35:42<43:29,  1.39it/s]
1|2853|Loss: 0.9534031748771667:  44%|████▍     | 2853/6470 [35:42<42:17,  1.43it/s]
1|2854|Loss: 0.8714677691459656:  44%|████▍     | 2853/6470 [35:42<42:17,  1.43it/s]
1|2854|Loss: 0.8714677691459656:  44%|████▍     | 2854/6470 [35:43<42:10,  1.43it/s]
1|2855|Loss: 0.7807362675666809:  44%|████▍     | 2854/6470 [35:43<42:10,  1.43it/s]
1|2855|Loss: 0.7807362675666809:  44%|████▍     | 2855/6470 [35:44<44:03,  1.37it/s]
1|2856|Loss: 0.8320323824882507:  44%|████▍     | 2855/6470 [35:44<44:03,  1.37it/s]
1|2856|Loss: 0.8320323824882507:  44%|████▍     | 2856/6470 [35:44<44:24,  1.36it/s]
1|2857|Loss: 1.072816014289856:  44%|████▍     | 2856/6470 [35:45<44:24,  1.36it/s] 
1|2857|Loss: 1.072816014289856:  44%|████▍     | 2857/6470 [35:45<46:04,  1.31it/s]
1|2858|Loss: 0.9620254039764404:  44%|████▍     | 2857/6470 [35:46<46:04,  1.31it/s]
1|2858|Loss: 0.9620254039764404:  44%|████▍     | 2858/6470 [35:46<46:06,  1.31it/s]
1|2859|Loss: 0.766043484210968:  44%|████▍     | 2858/6470 [35:46<46:06,  1.31it/s] 
1|2859|Loss: 0.766043484210968:  44%|████▍     | 2859/6470 [35:47<45:03,  1.34it/s]
1|2860|Loss: 0.9228910803794861:  44%|████▍     | 2859/6470 [35:47<45:03,  1.34it/s]
1|2860|Loss: 0.9228910803794861:  44%|████▍     | 2860/6470 [35:47<42:27,  1.42it/s]
1|2861|Loss: 0.9723528623580933:  44%|████▍     | 2860/6470 [35:48<42:27,  1.42it/s]
1|2861|Loss: 0.9723528623580933:  44%|████▍     | 2861/6470 [35:48<40:18,  1.49it/s]
1|2862|Loss: 0.9713621139526367:  44%|████▍     | 2861/6470 [35:48<40:18,  1.49it/s]
1|2862|Loss: 0.9713621139526367:  44%|████▍     | 2862/6470 [35:48<39:18,  1.53it/s]
1|2863|Loss: 0.9029933214187622:  44%|████▍     | 2862/6470 [35:49<39:18,  1.53it/s]
1|2863|Loss: 0.9029933214187622:  44%|████▍     | 2863/6470 [35:49<39:57,  1.50it/s]
1|2864|Loss: 0.9716036915779114:  44%|████▍     | 2863/6470 [35:49<39:57,  1.50it/s]
1|2864|Loss: 0.9716036915779114:  44%|████▍     | 2864/6470 [35:50<40:42,  1.48it/s]
1|2865|Loss: 0.866678774356842:  44%|████▍     | 2864/6470 [35:50<40:42,  1.48it/s] 
1|2865|Loss: 0.866678774356842:  44%|████▍     | 2865/6470 [35:50<38:02,  1.58it/s]
1|2866|Loss: 0.9464724063873291:  44%|████▍     | 2865/6470 [35:51<38:02,  1.58it/s]
1|2866|Loss: 0.9464724063873291:  44%|████▍     | 2866/6470 [35:51<40:42,  1.48it/s]
1|2867|Loss: 0.8584901690483093:  44%|████▍     | 2866/6470 [35:51<40:42,  1.48it/s]
1|2867|Loss: 0.8584901690483093:  44%|████▍     | 2867/6470 [35:52<40:44,  1.47it/s]
1|2868|Loss: 0.9288502335548401:  44%|████▍     | 2867/6470 [35:52<40:44,  1.47it/s]
1|2868|Loss: 0.9288502335548401:  44%|████▍     | 2868/6470 [35:52<40:44,  1.47it/s]
1|2869|Loss: 0.8785126209259033:  44%|████▍     | 2868/6470 [35:53<40:44,  1.47it/s]
1|2869|Loss: 0.8785126209259033:  44%|████▍     | 2869/6470 [35:53<38:01,  1.58it/s]
1|2870|Loss: 0.8334490656852722:  44%|████▍     | 2869/6470 [35:53<38:01,  1.58it/s]
1|2870|Loss: 0.8334490656852722:  44%|████▍     | 2870/6470 [35:54<37:32,  1.60it/s]
1|2871|Loss: 0.9838149547576904:  44%|████▍     | 2870/6470 [35:54<37:32,  1.60it/s]
1|2871|Loss: 0.9838149547576904:  44%|████▍     | 2871/6470 [35:54<40:29,  1.48it/s]
1|2872|Loss: 0.8558693528175354:  44%|████▍     | 2871/6470 [35:55<40:29,  1.48it/s]
1|2872|Loss: 0.8558693528175354:  44%|████▍     | 2872/6470 [35:55<42:13,  1.42it/s]
1|2873|Loss: 1.0639537572860718:  44%|████▍     | 2872/6470 [35:56<42:13,  1.42it/s]
1|2873|Loss: 1.0639537572860718:  44%|████▍     | 2873/6470 [35:56<43:42,  1.37it/s]
1|2874|Loss: 0.8597328066825867:  44%|████▍     | 2873/6470 [35:56<43:42,  1.37it/s]
1|2874|Loss: 0.8597328066825867:  44%|████▍     | 2874/6470 [35:57<45:24,  1.32it/s]
1|2875|Loss: 0.7821647524833679:  44%|████▍     | 2874/6470 [35:57<45:24,  1.32it/s]
1|2875|Loss: 0.7821647524833679:  44%|████▍     | 2875/6470 [35:57<42:26,  1.41it/s]
1|2876|Loss: 0.9369701147079468:  44%|████▍     | 2875/6470 [35:58<42:26,  1.41it/s]
1|2876|Loss: 0.9369701147079468:  44%|████▍     | 2876/6470 [35:58<41:37,  1.44it/s]
1|2877|Loss: 0.888425886631012:  44%|████▍     | 2876/6470 [35:58<41:37,  1.44it/s] 
1|2877|Loss: 0.888425886631012:  44%|████▍     | 2877/6470 [35:59<44:20,  1.35it/s]
1|2878|Loss: 0.998268723487854:  44%|████▍     | 2877/6470 [35:59<44:20,  1.35it/s]
1|2878|Loss: 0.998268723487854:  44%|████▍     | 2878/6470 [36:00<46:15,  1.29it/s]
1|2879|Loss: 0.8468444347381592:  44%|████▍     | 2878/6470 [36:00<46:15,  1.29it/s]
1|2879|Loss: 0.8468444347381592:  44%|████▍     | 2879/6470 [36:00<45:24,  1.32it/s]
1|2880|Loss: 0.9281926155090332:  44%|████▍     | 2879/6470 [36:01<45:24,  1.32it/s]
1|2880|Loss: 0.9281926155090332:  45%|████▍     | 2880/6470 [36:01<43:33,  1.37it/s]
1|2881|Loss: 0.908738911151886:  45%|████▍     | 2880/6470 [36:01<43:33,  1.37it/s] 
1|2881|Loss: 0.908738911151886:  45%|████▍     | 2881/6470 [36:02<42:46,  1.40it/s]
1|2882|Loss: 0.8474732041358948:  45%|████▍     | 2881/6470 [36:02<42:46,  1.40it/s]
1|2882|Loss: 0.8474732041358948:  45%|████▍     | 2882/6470 [36:02<38:52,  1.54it/s]
1|2883|Loss: 0.8701045513153076:  45%|████▍     | 2882/6470 [36:03<38:52,  1.54it/s]
1|2883|Loss: 0.8701045513153076:  45%|████▍     | 2883/6470 [36:03<38:44,  1.54it/s]
1|2884|Loss: 0.773377001285553:  45%|████▍     | 2883/6470 [36:03<38:44,  1.54it/s] 
1|2884|Loss: 0.773377001285553:  45%|████▍     | 2884/6470 [36:03<36:51,  1.62it/s]
1|2885|Loss: 0.778567910194397:  45%|████▍     | 2884/6470 [36:04<36:51,  1.62it/s]
1|2885|Loss: 0.778567910194397:  45%|████▍     | 2885/6470 [36:04<36:32,  1.64it/s]
1|2886|Loss: 0.8264123797416687:  45%|████▍     | 2885/6470 [36:04<36:32,  1.64it/s]
1|2886|Loss: 0.8264123797416687:  45%|████▍     | 2886/6470 [36:05<38:00,  1.57it/s]
1|2887|Loss: 0.9755185842514038:  45%|████▍     | 2886/6470 [36:05<38:00,  1.57it/s]
1|2887|Loss: 0.9755185842514038:  45%|████▍     | 2887/6470 [36:06<40:59,  1.46it/s]
1|2888|Loss: 0.7914611101150513:  45%|████▍     | 2887/6470 [36:06<40:59,  1.46it/s]
1|2888|Loss: 0.7914611101150513:  45%|████▍     | 2888/6470 [36:06<42:23,  1.41it/s]
1|2889|Loss: 0.793373703956604:  45%|████▍     | 2888/6470 [36:07<42:23,  1.41it/s] 
1|2889|Loss: 0.793373703956604:  45%|████▍     | 2889/6470 [36:07<44:37,  1.34it/s]
1|2890|Loss: 0.83897864818573:  45%|████▍     | 2889/6470 [36:08<44:37,  1.34it/s] 
1|2890|Loss: 0.83897864818573:  45%|████▍     | 2890/6470 [36:08<43:05,  1.38it/s]
1|2891|Loss: 0.9973966479301453:  45%|████▍     | 2890/6470 [36:08<43:05,  1.38it/s]
1|2891|Loss: 0.9973966479301453:  45%|████▍     | 2891/6470 [36:08<40:32,  1.47it/s]
1|2892|Loss: 0.9867947697639465:  45%|████▍     | 2891/6470 [36:09<40:32,  1.47it/s]
1|2892|Loss: 0.9867947697639465:  45%|████▍     | 2892/6470 [36:09<39:36,  1.51it/s]
1|2893|Loss: 0.855903685092926:  45%|████▍     | 2892/6470 [36:09<39:36,  1.51it/s] 
1|2893|Loss: 0.855903685092926:  45%|████▍     | 2893/6470 [36:10<37:59,  1.57it/s]
1|2894|Loss: 1.0073339939117432:  45%|████▍     | 2893/6470 [36:10<37:59,  1.57it/s]
1|2894|Loss: 1.0073339939117432:  45%|████▍     | 2894/6470 [36:10<37:22,  1.59it/s]
1|2895|Loss: 1.0139573812484741:  45%|████▍     | 2894/6470 [36:11<37:22,  1.59it/s]
1|2895|Loss: 1.0139573812484741:  45%|████▍     | 2895/6470 [36:11<36:37,  1.63it/s]
1|2896|Loss: 0.8259223103523254:  45%|████▍     | 2895/6470 [36:11<36:37,  1.63it/s]
1|2896|Loss: 0.8259223103523254:  45%|████▍     | 2896/6470 [36:11<37:47,  1.58it/s]
1|2897|Loss: 0.8591859340667725:  45%|████▍     | 2896/6470 [36:12<37:47,  1.58it/s]
1|2897|Loss: 0.8591859340667725:  45%|████▍     | 2897/6470 [36:12<39:46,  1.50it/s]
1|2898|Loss: 0.7918113470077515:  45%|████▍     | 2897/6470 [36:13<39:46,  1.50it/s]
1|2898|Loss: 0.7918113470077515:  45%|████▍     | 2898/6470 [36:13<45:38,  1.30it/s]
1|2899|Loss: 0.8047681450843811:  45%|████▍     | 2898/6470 [36:14<45:38,  1.30it/s]
1|2899|Loss: 0.8047681450843811:  45%|████▍     | 2899/6470 [36:14<48:10,  1.24it/s]
1|2900|Loss: 0.8859742879867554:  45%|████▍     | 2899/6470 [36:15<48:10,  1.24it/s]
1|2900|Loss: 0.8859742879867554:  45%|████▍     | 2900/6470 [36:15<47:02,  1.26it/s]
1|2901|Loss: 0.7239360213279724:  45%|████▍     | 2900/6470 [36:15<47:02,  1.26it/s]
1|2901|Loss: 0.7239360213279724:  45%|████▍     | 2901/6470 [36:16<44:08,  1.35it/s]
1|2902|Loss: 0.8565466403961182:  45%|████▍     | 2901/6470 [36:16<44:08,  1.35it/s]
1|2902|Loss: 0.8565466403961182:  45%|████▍     | 2902/6470 [36:16<42:14,  1.41it/s]
1|2903|Loss: 0.8639037609100342:  45%|████▍     | 2902/6470 [36:17<42:14,  1.41it/s]
1|2903|Loss: 0.8639037609100342:  45%|████▍     | 2903/6470 [36:17<41:02,  1.45it/s]
1|2904|Loss: 0.9305960536003113:  45%|████▍     | 2903/6470 [36:17<41:02,  1.45it/s]
1|2904|Loss: 0.9305960536003113:  45%|████▍     | 2904/6470 [36:17<38:30,  1.54it/s]
1|2905|Loss: 0.9813677668571472:  45%|████▍     | 2904/6470 [36:18<38:30,  1.54it/s]
1|2905|Loss: 0.9813677668571472:  45%|████▍     | 2905/6470 [36:18<37:38,  1.58it/s]
1|2906|Loss: 0.9487901329994202:  45%|████▍     | 2905/6470 [36:18<37:38,  1.58it/s]
1|2906|Loss: 0.9487901329994202:  45%|████▍     | 2906/6470 [36:19<36:44,  1.62it/s]
1|2907|Loss: 0.9093155860900879:  45%|████▍     | 2906/6470 [36:19<36:44,  1.62it/s]
1|2907|Loss: 0.9093155860900879:  45%|████▍     | 2907/6470 [36:19<36:59,  1.61it/s]
1|2908|Loss: 0.9458977580070496:  45%|████▍     | 2907/6470 [36:20<36:59,  1.61it/s]
1|2908|Loss: 0.9458977580070496:  45%|████▍     | 2908/6470 [36:20<38:17,  1.55it/s]
1|2909|Loss: 0.8572574257850647:  45%|████▍     | 2908/6470 [36:20<38:17,  1.55it/s]
1|2909|Loss: 0.8572574257850647:  45%|████▍     | 2909/6470 [36:21<39:19,  1.51it/s]
1|2910|Loss: 0.9253103137016296:  45%|████▍     | 2909/6470 [36:21<39:19,  1.51it/s]
1|2910|Loss: 0.9253103137016296:  45%|████▍     | 2910/6470 [36:21<41:25,  1.43it/s]
1|2911|Loss: 0.823117196559906:  45%|████▍     | 2910/6470 [36:22<41:25,  1.43it/s] 
1|2911|Loss: 0.823117196559906:  45%|████▍     | 2911/6470 [36:22<42:18,  1.40it/s]
1|2912|Loss: 0.9078096747398376:  45%|████▍     | 2911/6470 [36:23<42:18,  1.40it/s]
1|2912|Loss: 0.9078096747398376:  45%|████▌     | 2912/6470 [36:23<44:28,  1.33it/s]
1|2913|Loss: 0.8075793981552124:  45%|████▌     | 2912/6470 [36:23<44:28,  1.33it/s]
1|2913|Loss: 0.8075793981552124:  45%|████▌     | 2913/6470 [36:24<44:20,  1.34it/s]
1|2914|Loss: 0.6781139373779297:  45%|████▌     | 2913/6470 [36:24<44:20,  1.34it/s]
1|2914|Loss: 0.6781139373779297:  45%|████▌     | 2914/6470 [36:24<44:08,  1.34it/s]
1|2915|Loss: 0.7681596875190735:  45%|████▌     | 2914/6470 [36:25<44:08,  1.34it/s]
1|2915|Loss: 0.7681596875190735:  45%|████▌     | 2915/6470 [36:25<45:22,  1.31it/s]
1|2916|Loss: 0.7530278563499451:  45%|████▌     | 2915/6470 [36:26<45:22,  1.31it/s]
1|2916|Loss: 0.7530278563499451:  45%|████▌     | 2916/6470 [36:26<46:05,  1.29it/s]
1|2917|Loss: 0.7670744061470032:  45%|████▌     | 2916/6470 [36:26<46:05,  1.29it/s]
1|2917|Loss: 0.7670744061470032:  45%|████▌     | 2917/6470 [36:27<43:53,  1.35it/s]
1|2918|Loss: 0.9140139818191528:  45%|████▌     | 2917/6470 [36:27<43:53,  1.35it/s]
1|2918|Loss: 0.9140139818191528:  45%|████▌     | 2918/6470 [36:27<41:55,  1.41it/s]
1|2919|Loss: 0.8897920250892639:  45%|████▌     | 2918/6470 [36:28<41:55,  1.41it/s]
1|2919|Loss: 0.8897920250892639:  45%|████▌     | 2919/6470 [36:28<43:26,  1.36it/s]
1|2920|Loss: 1.0071128606796265:  45%|████▌     | 2919/6470 [36:28<43:26,  1.36it/s]
1|2920|Loss: 1.0071128606796265:  45%|████▌     | 2920/6470 [36:29<41:28,  1.43it/s]
1|2921|Loss: 0.8680078983306885:  45%|████▌     | 2920/6470 [36:29<41:28,  1.43it/s]
1|2921|Loss: 0.8680078983306885:  45%|████▌     | 2921/6470 [36:30<43:27,  1.36it/s]
1|2922|Loss: 0.9973582625389099:  45%|████▌     | 2921/6470 [36:30<43:27,  1.36it/s]
1|2922|Loss: 0.9973582625389099:  45%|████▌     | 2922/6470 [36:30<43:56,  1.35it/s]
1|2923|Loss: 0.8054072856903076:  45%|████▌     | 2922/6470 [36:31<43:56,  1.35it/s]
1|2923|Loss: 0.8054072856903076:  45%|████▌     | 2923/6470 [36:31<43:53,  1.35it/s]
1|2924|Loss: 0.8896013498306274:  45%|████▌     | 2923/6470 [36:31<43:53,  1.35it/s]
1|2924|Loss: 0.8896013498306274:  45%|████▌     | 2924/6470 [36:32<43:40,  1.35it/s]
1|2925|Loss: 0.94184809923172:  45%|████▌     | 2924/6470 [36:32<43:40,  1.35it/s]  
1|2925|Loss: 0.94184809923172:  45%|████▌     | 2925/6470 [36:32<42:16,  1.40it/s]
1|2926|Loss: 0.7997079491615295:  45%|████▌     | 2925/6470 [36:33<42:16,  1.40it/s]
1|2926|Loss: 0.7997079491615295:  45%|████▌     | 2926/6470 [36:33<39:38,  1.49it/s]
1|2927|Loss: 0.909645140171051:  45%|████▌     | 2926/6470 [36:33<39:38,  1.49it/s] 
1|2927|Loss: 0.909645140171051:  45%|████▌     | 2927/6470 [36:34<38:24,  1.54it/s]
1|2928|Loss: 0.6508360505104065:  45%|████▌     | 2927/6470 [36:34<38:24,  1.54it/s]
1|2928|Loss: 0.6508360505104065:  45%|████▌     | 2928/6470 [36:34<42:36,  1.39it/s]
1|2929|Loss: 0.9925615787506104:  45%|████▌     | 2928/6470 [36:35<42:36,  1.39it/s]
1|2929|Loss: 0.9925615787506104:  45%|████▌     | 2929/6470 [36:35<41:05,  1.44it/s]
1|2930|Loss: 0.8835673332214355:  45%|████▌     | 2929/6470 [36:35<41:05,  1.44it/s]
1|2930|Loss: 0.8835673332214355:  45%|████▌     | 2930/6470 [36:36<38:53,  1.52it/s]
1|2931|Loss: 0.9474667906761169:  45%|████▌     | 2930/6470 [36:36<38:53,  1.52it/s]
1|2931|Loss: 0.9474667906761169:  45%|████▌     | 2931/6470 [36:36<39:20,  1.50it/s]
1|2932|Loss: 1.0672345161437988:  45%|████▌     | 2931/6470 [36:37<39:20,  1.50it/s]
1|2932|Loss: 1.0672345161437988:  45%|████▌     | 2932/6470 [36:37<38:54,  1.52it/s]
1|2933|Loss: 0.7933275103569031:  45%|████▌     | 2932/6470 [36:37<38:54,  1.52it/s]
1|2933|Loss: 0.7933275103569031:  45%|████▌     | 2933/6470 [36:38<40:43,  1.45it/s]
1|2934|Loss: 0.8550230264663696:  45%|████▌     | 2933/6470 [36:38<40:43,  1.45it/s]
1|2934|Loss: 0.8550230264663696:  45%|████▌     | 2934/6470 [36:38<39:39,  1.49it/s]
1|2935|Loss: 0.882992684841156:  45%|████▌     | 2934/6470 [36:39<39:39,  1.49it/s] 
1|2935|Loss: 0.882992684841156:  45%|████▌     | 2935/6470 [36:39<39:45,  1.48it/s]
1|2936|Loss: 0.918266773223877:  45%|████▌     | 2935/6470 [36:39<39:45,  1.48it/s]
1|2936|Loss: 0.918266773223877:  45%|████▌     | 2936/6470 [36:40<39:03,  1.51it/s]
1|2937|Loss: 0.9891505837440491:  45%|████▌     | 2936/6470 [36:40<39:03,  1.51it/s]
1|2937|Loss: 0.9891505837440491:  45%|████▌     | 2937/6470 [36:40<37:34,  1.57it/s]
1|2938|Loss: 0.8292454481124878:  45%|████▌     | 2937/6470 [36:41<37:34,  1.57it/s]
1|2938|Loss: 0.8292454481124878:  45%|████▌     | 2938/6470 [36:41<35:34,  1.66it/s]
1|2939|Loss: 0.9497181177139282:  45%|████▌     | 2938/6470 [36:41<35:34,  1.66it/s]
1|2939|Loss: 0.9497181177139282:  45%|████▌     | 2939/6470 [36:41<34:24,  1.71it/s]
1|2940|Loss: 0.8888933658599854:  45%|████▌     | 2939/6470 [36:42<34:24,  1.71it/s]
1|2940|Loss: 0.8888933658599854:  45%|████▌     | 2940/6470 [36:42<37:25,  1.57it/s]
1|2941|Loss: 0.8698370456695557:  45%|████▌     | 2940/6470 [36:43<37:25,  1.57it/s]
1|2941|Loss: 0.8698370456695557:  45%|████▌     | 2941/6470 [36:43<39:00,  1.51it/s]
1|2942|Loss: 0.8350063562393188:  45%|████▌     | 2941/6470 [36:43<39:00,  1.51it/s]
1|2942|Loss: 0.8350063562393188:  45%|████▌     | 2942/6470 [36:44<41:36,  1.41it/s]
1|2943|Loss: 0.8489174842834473:  45%|████▌     | 2942/6470 [36:44<41:36,  1.41it/s]
1|2943|Loss: 0.8489174842834473:  45%|████▌     | 2943/6470 [36:45<44:32,  1.32it/s]
1|2944|Loss: 0.8404317498207092:  45%|████▌     | 2943/6470 [36:45<44:32,  1.32it/s]
1|2944|Loss: 0.8404317498207092:  46%|████▌     | 2944/6470 [36:45<45:05,  1.30it/s]
1|2945|Loss: 0.9174575209617615:  46%|████▌     | 2944/6470 [36:46<45:05,  1.30it/s]
1|2945|Loss: 0.9174575209617615:  46%|████▌     | 2945/6470 [36:46<41:46,  1.41it/s]
1|2946|Loss: 0.956521213054657:  46%|████▌     | 2945/6470 [36:46<41:46,  1.41it/s] 
1|2946|Loss: 0.956521213054657:  46%|████▌     | 2946/6470 [36:46<39:12,  1.50it/s]
1|2947|Loss: 0.8776801824569702:  46%|████▌     | 2946/6470 [36:47<39:12,  1.50it/s]
1|2947|Loss: 0.8776801824569702:  46%|████▌     | 2947/6470 [36:47<37:36,  1.56it/s]
1|2948|Loss: 0.8225581049919128:  46%|████▌     | 2947/6470 [36:47<37:36,  1.56it/s]
1|2948|Loss: 0.8225581049919128:  46%|████▌     | 2948/6470 [36:48<38:22,  1.53it/s]
1|2949|Loss: 0.9988166689872742:  46%|████▌     | 2948/6470 [36:48<38:22,  1.53it/s]
1|2949|Loss: 0.9988166689872742:  46%|████▌     | 2949/6470 [36:48<39:51,  1.47it/s]
1|2950|Loss: 0.9219571352005005:  46%|████▌     | 2949/6470 [36:49<39:51,  1.47it/s]
1|2950|Loss: 0.9219571352005005:  46%|████▌     | 2950/6470 [36:49<40:59,  1.43it/s]
1|2951|Loss: 0.7764265537261963:  46%|████▌     | 2950/6470 [36:50<40:59,  1.43it/s]
1|2951|Loss: 0.7764265537261963:  46%|████▌     | 2951/6470 [36:50<41:33,  1.41it/s]
1|2952|Loss: 0.706004798412323:  46%|████▌     | 2951/6470 [36:50<41:33,  1.41it/s] 
1|2952|Loss: 0.706004798412323:  46%|████▌     | 2952/6470 [36:51<41:33,  1.41it/s]
1|2953|Loss: 0.967587947845459:  46%|████▌     | 2952/6470 [36:51<41:33,  1.41it/s]
1|2953|Loss: 0.967587947845459:  46%|████▌     | 2953/6470 [36:51<41:27,  1.41it/s]
1|2954|Loss: 0.8318998217582703:  46%|████▌     | 2953/6470 [36:52<41:27,  1.41it/s]
1|2954|Loss: 0.8318998217582703:  46%|████▌     | 2954/6470 [36:52<43:04,  1.36it/s]
1|2955|Loss: 1.1064122915267944:  46%|████▌     | 2954/6470 [36:53<43:04,  1.36it/s]
1|2955|Loss: 1.1064122915267944:  46%|████▌     | 2955/6470 [36:53<38:17,  1.53it/s]
1|2956|Loss: 1.0303246974945068:  46%|████▌     | 2955/6470 [36:53<38:17,  1.53it/s]
1|2956|Loss: 1.0303246974945068:  46%|████▌     | 2956/6470 [36:53<35:58,  1.63it/s]
1|2957|Loss: 0.8851295113563538:  46%|████▌     | 2956/6470 [36:54<35:58,  1.63it/s]
1|2957|Loss: 0.8851295113563538:  46%|████▌     | 2957/6470 [36:54<38:24,  1.52it/s]
1|2958|Loss: 1.0490200519561768:  46%|████▌     | 2957/6470 [36:54<38:24,  1.52it/s]
1|2958|Loss: 1.0490200519561768:  46%|████▌     | 2958/6470 [36:55<39:57,  1.46it/s]
1|2959|Loss: 0.9664930105209351:  46%|████▌     | 2958/6470 [36:55<39:57,  1.46it/s]
1|2959|Loss: 0.9664930105209351:  46%|████▌     | 2959/6470 [36:55<37:36,  1.56it/s]
1|2960|Loss: 0.7161377668380737:  46%|████▌     | 2959/6470 [36:56<37:36,  1.56it/s]
1|2960|Loss: 0.7161377668380737:  46%|████▌     | 2960/6470 [36:56<36:23,  1.61it/s]
1|2961|Loss: 0.9769307971000671:  46%|████▌     | 2960/6470 [36:56<36:23,  1.61it/s]
1|2961|Loss: 0.9769307971000671:  46%|████▌     | 2961/6470 [36:56<35:34,  1.64it/s]
1|2962|Loss: 0.8535147309303284:  46%|████▌     | 2961/6470 [36:57<35:34,  1.64it/s]
1|2962|Loss: 0.8535147309303284:  46%|████▌     | 2962/6470 [36:57<36:48,  1.59it/s]
1|2963|Loss: 0.8170826435089111:  46%|████▌     | 2962/6470 [36:57<36:48,  1.59it/s]
1|2963|Loss: 0.8170826435089111:  46%|████▌     | 2963/6470 [36:58<37:13,  1.57it/s]
1|2964|Loss: 0.9284835457801819:  46%|████▌     | 2963/6470 [36:58<37:13,  1.57it/s]
1|2964|Loss: 0.9284835457801819:  46%|████▌     | 2964/6470 [36:58<36:51,  1.59it/s]
1|2965|Loss: 0.9699643850326538:  46%|████▌     | 2964/6470 [36:59<36:51,  1.59it/s]
1|2965|Loss: 0.9699643850326538:  46%|████▌     | 2965/6470 [36:59<36:56,  1.58it/s]
1|2966|Loss: 0.9826240539550781:  46%|████▌     | 2965/6470 [36:59<36:56,  1.58it/s]
1|2966|Loss: 0.9826240539550781:  46%|████▌     | 2966/6470 [37:00<38:22,  1.52it/s]
1|2967|Loss: 1.2070688009262085:  46%|████▌     | 2966/6470 [37:00<38:22,  1.52it/s]
1|2967|Loss: 1.2070688009262085:  46%|████▌     | 2967/6470 [37:00<34:48,  1.68it/s]
1|2968|Loss: 0.9868966937065125:  46%|████▌     | 2967/6470 [37:00<34:48,  1.68it/s]
1|2968|Loss: 0.9868966937065125:  46%|████▌     | 2968/6470 [37:01<38:03,  1.53it/s]
1|2969|Loss: 0.8824230432510376:  46%|████▌     | 2968/6470 [37:01<38:03,  1.53it/s]
1|2969|Loss: 0.8824230432510376:  46%|████▌     | 2969/6470 [37:02<39:43,  1.47it/s]
1|2970|Loss: 0.8146570324897766:  46%|████▌     | 2969/6470 [37:02<39:43,  1.47it/s]
1|2970|Loss: 0.8146570324897766:  46%|████▌     | 2970/6470 [37:02<40:39,  1.43it/s]
1|2971|Loss: 0.9271450638771057:  46%|████▌     | 2970/6470 [37:03<40:39,  1.43it/s]
1|2971|Loss: 0.9271450638771057:  46%|████▌     | 2971/6470 [37:03<41:29,  1.41it/s]
1|2972|Loss: 0.9052973985671997:  46%|████▌     | 2971/6470 [37:04<41:29,  1.41it/s]
1|2972|Loss: 0.9052973985671997:  46%|████▌     | 2972/6470 [37:04<40:22,  1.44it/s]
1|2973|Loss: 0.8429856300354004:  46%|████▌     | 2972/6470 [37:04<40:22,  1.44it/s]
1|2973|Loss: 0.8429856300354004:  46%|████▌     | 2973/6470 [37:04<40:35,  1.44it/s]
1|2974|Loss: 0.9452435374259949:  46%|████▌     | 2973/6470 [37:05<40:35,  1.44it/s]
1|2974|Loss: 0.9452435374259949:  46%|████▌     | 2974/6470 [37:05<39:35,  1.47it/s]
1|2975|Loss: 0.8155476450920105:  46%|████▌     | 2974/6470 [37:05<39:35,  1.47it/s]
1|2975|Loss: 0.8155476450920105:  46%|████▌     | 2975/6470 [37:06<39:31,  1.47it/s]
1|2976|Loss: 0.8246367573738098:  46%|████▌     | 2975/6470 [37:06<39:31,  1.47it/s]
1|2976|Loss: 0.8246367573738098:  46%|████▌     | 2976/6470 [37:07<40:22,  1.44it/s]
1|2977|Loss: 1.007540225982666:  46%|████▌     | 2976/6470 [37:07<40:22,  1.44it/s] 
1|2977|Loss: 1.007540225982666:  46%|████▌     | 2977/6470 [37:07<39:37,  1.47it/s]
1|2978|Loss: 0.8372316360473633:  46%|████▌     | 2977/6470 [37:08<39:37,  1.47it/s]
1|2978|Loss: 0.8372316360473633:  46%|████▌     | 2978/6470 [37:08<39:59,  1.46it/s]
1|2979|Loss: 0.7654552459716797:  46%|████▌     | 2978/6470 [37:08<39:59,  1.46it/s]
1|2979|Loss: 0.7654552459716797:  46%|████▌     | 2979/6470 [37:09<42:32,  1.37it/s]
1|2980|Loss: 0.9401997923851013:  46%|████▌     | 2979/6470 [37:09<42:32,  1.37it/s]
1|2980|Loss: 0.9401997923851013:  46%|████▌     | 2980/6470 [37:09<41:51,  1.39it/s]
1|2981|Loss: 0.87678462266922:  46%|████▌     | 2980/6470 [37:10<41:51,  1.39it/s]  
1|2981|Loss: 0.87678462266922:  46%|████▌     | 2981/6470 [37:10<37:49,  1.54it/s]
1|2982|Loss: 0.8362562656402588:  46%|████▌     | 2981/6470 [37:10<37:49,  1.54it/s]
1|2982|Loss: 0.8362562656402588:  46%|████▌     | 2982/6470 [37:11<38:04,  1.53it/s]
1|2983|Loss: 0.8631142973899841:  46%|████▌     | 2982/6470 [37:11<38:04,  1.53it/s]
1|2983|Loss: 0.8631142973899841:  46%|████▌     | 2983/6470 [37:11<39:29,  1.47it/s]
1|2984|Loss: 0.798224687576294:  46%|████▌     | 2983/6470 [37:12<39:29,  1.47it/s] 
1|2984|Loss: 0.798224687576294:  46%|████▌     | 2984/6470 [37:12<40:25,  1.44it/s]
1|2985|Loss: 0.9514648914337158:  46%|████▌     | 2984/6470 [37:12<40:25,  1.44it/s]
1|2985|Loss: 0.9514648914337158:  46%|████▌     | 2985/6470 [37:13<41:23,  1.40it/s]
1|2986|Loss: 1.021193265914917:  46%|████▌     | 2985/6470 [37:13<41:23,  1.40it/s] 
1|2986|Loss: 1.021193265914917:  46%|████▌     | 2986/6470 [37:14<43:35,  1.33it/s]
1|2987|Loss: 0.9523642063140869:  46%|████▌     | 2986/6470 [37:14<43:35,  1.33it/s]
1|2987|Loss: 0.9523642063140869:  46%|████▌     | 2987/6470 [37:14<42:36,  1.36it/s]
1|2988|Loss: 0.8476991057395935:  46%|████▌     | 2987/6470 [37:15<42:36,  1.36it/s]
1|2988|Loss: 0.8476991057395935:  46%|████▌     | 2988/6470 [37:15<39:08,  1.48it/s]
1|2989|Loss: 0.9788268208503723:  46%|████▌     | 2988/6470 [37:15<39:08,  1.48it/s]
1|2989|Loss: 0.9788268208503723:  46%|████▌     | 2989/6470 [37:15<38:01,  1.53it/s]
1|2990|Loss: 0.7486473917961121:  46%|████▌     | 2989/6470 [37:16<38:01,  1.53it/s]
1|2990|Loss: 0.7486473917961121:  46%|████▌     | 2990/6470 [37:16<36:48,  1.58it/s]
1|2991|Loss: 0.8443712592124939:  46%|████▌     | 2990/6470 [37:16<36:48,  1.58it/s]
1|2991|Loss: 0.8443712592124939:  46%|████▌     | 2991/6470 [37:17<36:20,  1.60it/s]
1|2992|Loss: 0.7422175407409668:  46%|████▌     | 2991/6470 [37:17<36:20,  1.60it/s]
1|2992|Loss: 0.7422175407409668:  46%|████▌     | 2992/6470 [37:17<38:07,  1.52it/s]
1|2993|Loss: 0.9191187620162964:  46%|████▌     | 2992/6470 [37:18<38:07,  1.52it/s]
1|2993|Loss: 0.9191187620162964:  46%|████▋     | 2993/6470 [37:18<37:57,  1.53it/s]
1|2994|Loss: 0.8868007659912109:  46%|████▋     | 2993/6470 [37:18<37:57,  1.53it/s]
1|2994|Loss: 0.8868007659912109:  46%|████▋     | 2994/6470 [37:19<40:21,  1.44it/s]
1|2995|Loss: 0.9160367846488953:  46%|████▋     | 2994/6470 [37:19<40:21,  1.44it/s]
1|2995|Loss: 0.9160367846488953:  46%|████▋     | 2995/6470 [37:19<38:18,  1.51it/s]
1|2996|Loss: 0.6659001708030701:  46%|████▋     | 2995/6470 [37:20<38:18,  1.51it/s]
1|2996|Loss: 0.6659001708030701:  46%|████▋     | 2996/6470 [37:20<38:51,  1.49it/s]
1|2997|Loss: 0.8167935013771057:  46%|████▋     | 2996/6470 [37:21<38:51,  1.49it/s]
1|2997|Loss: 0.8167935013771057:  46%|████▋     | 2997/6470 [37:21<40:04,  1.44it/s]
1|2998|Loss: 0.9191139340400696:  46%|████▋     | 2997/6470 [37:21<40:04,  1.44it/s]
1|2998|Loss: 0.9191139340400696:  46%|████▋     | 2998/6470 [37:22<39:34,  1.46it/s]
1|2999|Loss: 0.9197768568992615:  46%|████▋     | 2998/6470 [37:22<39:34,  1.46it/s]
1|2999|Loss: 0.9197768568992615:  46%|████▋     | 2999/6470 [37:22<38:30,  1.50it/s]
1|3000|Loss: 1.0440802574157715:  46%|████▋     | 2999/6470 [37:23<38:30,  1.50it/s]
1|3000|Loss: 1.0440802574157715:  46%|████▋     | 3000/6470 [37:23<39:01,  1.48it/s]
1|3001|Loss: 0.7803256511688232:  46%|████▋     | 3000/6470 [37:23<39:01,  1.48it/s]
1|3001|Loss: 0.7803256511688232:  46%|████▋     | 3001/6470 [37:24<41:25,  1.40it/s]
1|3002|Loss: 0.8693642020225525:  46%|████▋     | 3001/6470 [37:24<41:25,  1.40it/s]
1|3002|Loss: 0.8693642020225525:  46%|████▋     | 3002/6470 [37:24<40:55,  1.41it/s]
1|3003|Loss: 0.6980427503585815:  46%|████▋     | 3002/6470 [37:25<40:55,  1.41it/s]
1|3003|Loss: 0.6980427503585815:  46%|████▋     | 3003/6470 [37:25<39:28,  1.46it/s]
1|3004|Loss: 0.8610587120056152:  46%|████▋     | 3003/6470 [37:25<39:28,  1.46it/s]
1|3004|Loss: 0.8610587120056152:  46%|████▋     | 3004/6470 [37:26<37:46,  1.53it/s]
1|3005|Loss: 0.9331641793251038:  46%|████▋     | 3004/6470 [37:26<37:46,  1.53it/s]
1|3005|Loss: 0.9331641793251038:  46%|████▋     | 3005/6470 [37:26<35:07,  1.64it/s]
1|3006|Loss: 0.9502465724945068:  46%|████▋     | 3005/6470 [37:26<35:07,  1.64it/s]
1|3006|Loss: 0.9502465724945068:  46%|████▋     | 3006/6470 [37:27<35:53,  1.61it/s]
1|3007|Loss: 0.8501350283622742:  46%|████▋     | 3006/6470 [37:27<35:53,  1.61it/s]
1|3007|Loss: 0.8501350283622742:  46%|████▋     | 3007/6470 [37:27<32:25,  1.78it/s]
1|3008|Loss: 0.758495032787323:  46%|████▋     | 3007/6470 [37:27<32:25,  1.78it/s] 
1|3008|Loss: 0.758495032787323:  46%|████▋     | 3008/6470 [37:28<32:35,  1.77it/s]
1|3009|Loss: 1.016700267791748:  46%|████▋     | 3008/6470 [37:28<32:35,  1.77it/s]
1|3009|Loss: 1.016700267791748:  47%|████▋     | 3009/6470 [37:28<36:25,  1.58it/s]
1|3010|Loss: 0.8264402151107788:  47%|████▋     | 3009/6470 [37:29<36:25,  1.58it/s]
1|3010|Loss: 0.8264402151107788:  47%|████▋     | 3010/6470 [37:29<38:31,  1.50it/s]
1|3011|Loss: 0.7829431295394897:  47%|████▋     | 3010/6470 [37:30<38:31,  1.50it/s]
1|3011|Loss: 0.7829431295394897:  47%|████▋     | 3011/6470 [37:30<40:26,  1.43it/s]
1|3012|Loss: 0.7696459889411926:  47%|████▋     | 3011/6470 [37:30<40:26,  1.43it/s]
1|3012|Loss: 0.7696459889411926:  47%|████▋     | 3012/6470 [37:31<39:40,  1.45it/s]
1|3013|Loss: 0.7859953045845032:  47%|████▋     | 3012/6470 [37:31<39:40,  1.45it/s]
1|3013|Loss: 0.7859953045845032:  47%|████▋     | 3013/6470 [37:31<42:02,  1.37it/s]
1|3014|Loss: 0.8812236785888672:  47%|████▋     | 3013/6470 [37:32<42:02,  1.37it/s]
1|3014|Loss: 0.8812236785888672:  47%|████▋     | 3014/6470 [37:32<42:43,  1.35it/s]
1|3015|Loss: 0.8878489136695862:  47%|████▋     | 3014/6470 [37:33<42:43,  1.35it/s]
1|3015|Loss: 0.8878489136695862:  47%|████▋     | 3015/6470 [37:33<40:32,  1.42it/s]
1|3016|Loss: 0.9321309924125671:  47%|████▋     | 3015/6470 [37:33<40:32,  1.42it/s]
1|3016|Loss: 0.9321309924125671:  47%|████▋     | 3016/6470 [37:34<40:05,  1.44it/s]
1|3017|Loss: 0.8720106482505798:  47%|████▋     | 3016/6470 [37:34<40:05,  1.44it/s]
1|3017|Loss: 0.8720106482505798:  47%|████▋     | 3017/6470 [37:34<39:04,  1.47it/s]
1|3018|Loss: 0.857177734375:  47%|████▋     | 3017/6470 [37:35<39:04,  1.47it/s]    
1|3018|Loss: 0.857177734375:  47%|████▋     | 3018/6470 [37:35<37:54,  1.52it/s]
1|3019|Loss: 0.9374218583106995:  47%|████▋     | 3018/6470 [37:35<37:54,  1.52it/s]
1|3019|Loss: 0.9374218583106995:  47%|████▋     | 3019/6470 [37:35<36:32,  1.57it/s]
1|3020|Loss: 0.8983727097511292:  47%|████▋     | 3019/6470 [37:36<36:32,  1.57it/s]
1|3020|Loss: 0.8983727097511292:  47%|████▋     | 3020/6470 [37:36<37:48,  1.52it/s]
1|3021|Loss: 0.9971964955329895:  47%|████▋     | 3020/6470 [37:37<37:48,  1.52it/s]
1|3021|Loss: 0.9971964955329895:  47%|████▋     | 3021/6470 [37:37<39:24,  1.46it/s]
1|3022|Loss: 0.8523104786872864:  47%|████▋     | 3021/6470 [37:37<39:24,  1.46it/s]
1|3022|Loss: 0.8523104786872864:  47%|████▋     | 3022/6470 [37:38<40:30,  1.42it/s]
1|3023|Loss: 0.8399763107299805:  47%|████▋     | 3022/6470 [37:38<40:30,  1.42it/s]
1|3023|Loss: 0.8399763107299805:  47%|████▋     | 3023/6470 [37:38<40:42,  1.41it/s]
1|3024|Loss: 0.8931643962860107:  47%|████▋     | 3023/6470 [37:39<40:42,  1.41it/s]
1|3024|Loss: 0.8931643962860107:  47%|████▋     | 3024/6470 [37:39<40:57,  1.40it/s]
1|3025|Loss: 0.7451696395874023:  47%|████▋     | 3024/6470 [37:39<40:57,  1.40it/s]
1|3025|Loss: 0.7451696395874023:  47%|████▋     | 3025/6470 [37:40<39:14,  1.46it/s]
1|3026|Loss: 0.9105889797210693:  47%|████▋     | 3025/6470 [37:40<39:14,  1.46it/s]
1|3026|Loss: 0.9105889797210693:  47%|████▋     | 3026/6470 [37:40<36:27,  1.57it/s]
1|3027|Loss: 0.819031298160553:  47%|████▋     | 3026/6470 [37:41<36:27,  1.57it/s] 
1|3027|Loss: 0.819031298160553:  47%|████▋     | 3027/6470 [37:41<37:30,  1.53it/s]
1|3028|Loss: 0.793059766292572:  47%|████▋     | 3027/6470 [37:41<37:30,  1.53it/s]
1|3028|Loss: 0.793059766292572:  47%|████▋     | 3028/6470 [37:42<38:54,  1.47it/s]
1|3029|Loss: 0.9811059832572937:  47%|████▋     | 3028/6470 [37:42<38:54,  1.47it/s]
1|3029|Loss: 0.9811059832572937:  47%|████▋     | 3029/6470 [37:42<38:14,  1.50it/s]
1|3030|Loss: 0.9015436768531799:  47%|████▋     | 3029/6470 [37:43<38:14,  1.50it/s]
1|3030|Loss: 0.9015436768531799:  47%|████▋     | 3030/6470 [37:43<36:09,  1.59it/s]
1|3031|Loss: 0.9713745713233948:  47%|████▋     | 3030/6470 [37:43<36:09,  1.59it/s]
1|3031|Loss: 0.9713745713233948:  47%|████▋     | 3031/6470 [37:43<37:09,  1.54it/s]
1|3032|Loss: 0.9559535384178162:  47%|████▋     | 3031/6470 [37:44<37:09,  1.54it/s]
1|3032|Loss: 0.9559535384178162:  47%|████▋     | 3032/6470 [37:44<38:17,  1.50it/s]
1|3033|Loss: 0.8792296051979065:  47%|████▋     | 3032/6470 [37:45<38:17,  1.50it/s]
1|3033|Loss: 0.8792296051979065:  47%|████▋     | 3033/6470 [37:45<40:04,  1.43it/s]
1|3034|Loss: 0.9498345255851746:  47%|████▋     | 3033/6470 [37:45<40:04,  1.43it/s]
1|3034|Loss: 0.9498345255851746:  47%|████▋     | 3034/6470 [37:46<40:31,  1.41it/s]
1|3035|Loss: 0.9335074424743652:  47%|████▋     | 3034/6470 [37:46<40:31,  1.41it/s]
1|3035|Loss: 0.9335074424743652:  47%|████▋     | 3035/6470 [37:46<39:26,  1.45it/s]
1|3036|Loss: 0.8261195421218872:  47%|████▋     | 3035/6470 [37:47<39:26,  1.45it/s]
1|3036|Loss: 0.8261195421218872:  47%|████▋     | 3036/6470 [37:47<39:28,  1.45it/s]
1|3037|Loss: 0.8934135437011719:  47%|████▋     | 3036/6470 [37:47<39:28,  1.45it/s]
1|3037|Loss: 0.8934135437011719:  47%|████▋     | 3037/6470 [37:48<36:13,  1.58it/s]
1|3038|Loss: 0.8662059903144836:  47%|████▋     | 3037/6470 [37:48<36:13,  1.58it/s]
1|3038|Loss: 0.8662059903144836:  47%|████▋     | 3038/6470 [37:48<33:27,  1.71it/s]
1|3039|Loss: 0.7842665910720825:  47%|████▋     | 3038/6470 [37:48<33:27,  1.71it/s]
1|3039|Loss: 0.7842665910720825:  47%|████▋     | 3039/6470 [37:49<33:49,  1.69it/s]
1|3040|Loss: 0.8269902467727661:  47%|████▋     | 3039/6470 [37:49<33:49,  1.69it/s]
1|3040|Loss: 0.8269902467727661:  47%|████▋     | 3040/6470 [37:49<35:55,  1.59it/s]
1|3041|Loss: 0.7717976570129395:  47%|████▋     | 3040/6470 [37:50<35:55,  1.59it/s]
1|3041|Loss: 0.7717976570129395:  47%|████▋     | 3041/6470 [37:50<37:37,  1.52it/s]
1|3042|Loss: 0.7370570302009583:  47%|████▋     | 3041/6470 [37:50<37:37,  1.52it/s]
1|3042|Loss: 0.7370570302009583:  47%|████▋     | 3042/6470 [37:51<38:54,  1.47it/s]
1|3043|Loss: 0.8981510996818542:  47%|████▋     | 3042/6470 [37:51<38:54,  1.47it/s]
1|3043|Loss: 0.8981510996818542:  47%|████▋     | 3043/6470 [37:51<39:22,  1.45it/s]
1|3044|Loss: 0.8124613165855408:  47%|████▋     | 3043/6470 [37:52<39:22,  1.45it/s]
1|3044|Loss: 0.8124613165855408:  47%|████▋     | 3044/6470 [37:52<41:50,  1.36it/s]
1|3045|Loss: 0.841969907283783:  47%|████▋     | 3044/6470 [37:53<41:50,  1.36it/s] 
1|3045|Loss: 0.841969907283783:  47%|████▋     | 3045/6470 [37:53<43:48,  1.30it/s]
1|3046|Loss: 0.9477145075798035:  47%|████▋     | 3045/6470 [37:54<43:48,  1.30it/s]
1|3046|Loss: 0.9477145075798035:  47%|████▋     | 3046/6470 [37:54<39:47,  1.43it/s]
1|3047|Loss: 0.9732522368431091:  47%|████▋     | 3046/6470 [37:54<39:47,  1.43it/s]
1|3047|Loss: 0.9732522368431091:  47%|████▋     | 3047/6470 [37:54<37:07,  1.54it/s]
1|3048|Loss: 0.9767276644706726:  47%|████▋     | 3047/6470 [37:55<37:07,  1.54it/s]
1|3048|Loss: 0.9767276644706726:  47%|████▋     | 3048/6470 [37:55<38:08,  1.50it/s]
1|3049|Loss: 0.9031307101249695:  47%|████▋     | 3048/6470 [37:55<38:08,  1.50it/s]
1|3049|Loss: 0.9031307101249695:  47%|████▋     | 3049/6470 [37:56<40:59,  1.39it/s]
1|3050|Loss: 1.0458661317825317:  47%|████▋     | 3049/6470 [37:56<40:59,  1.39it/s]
1|3050|Loss: 1.0458661317825317:  47%|████▋     | 3050/6470 [37:57<42:33,  1.34it/s]
1|3051|Loss: 0.8077490925788879:  47%|████▋     | 3050/6470 [37:57<42:33,  1.34it/s]
1|3051|Loss: 0.8077490925788879:  47%|████▋     | 3051/6470 [37:57<43:37,  1.31it/s]
1|3052|Loss: 0.9780108332633972:  47%|████▋     | 3051/6470 [37:58<43:37,  1.31it/s]
1|3052|Loss: 0.9780108332633972:  47%|████▋     | 3052/6470 [37:58<41:11,  1.38it/s]
1|3053|Loss: 0.7398049831390381:  47%|████▋     | 3052/6470 [37:58<41:11,  1.38it/s]
1|3053|Loss: 0.7398049831390381:  47%|████▋     | 3053/6470 [37:59<40:23,  1.41it/s]
1|3054|Loss: 0.8845813870429993:  47%|████▋     | 3053/6470 [37:59<40:23,  1.41it/s]
1|3054|Loss: 0.8845813870429993:  47%|████▋     | 3054/6470 [37:59<39:10,  1.45it/s]
1|3055|Loss: 1.0429494380950928:  47%|████▋     | 3054/6470 [38:00<39:10,  1.45it/s]
1|3055|Loss: 1.0429494380950928:  47%|████▋     | 3055/6470 [38:00<40:28,  1.41it/s]
1|3056|Loss: 0.9585362672805786:  47%|████▋     | 3055/6470 [38:01<40:28,  1.41it/s]
1|3056|Loss: 0.9585362672805786:  47%|████▋     | 3056/6470 [38:01<41:04,  1.39it/s]
1|3057|Loss: 1.047824501991272:  47%|████▋     | 3056/6470 [38:01<41:04,  1.39it/s] 
1|3057|Loss: 1.047824501991272:  47%|████▋     | 3057/6470 [38:02<41:58,  1.36it/s]
1|3058|Loss: 0.7561486959457397:  47%|████▋     | 3057/6470 [38:02<41:58,  1.36it/s]
1|3058|Loss: 0.7561486959457397:  47%|████▋     | 3058/6470 [38:02<43:47,  1.30it/s]
1|3059|Loss: 0.8113309741020203:  47%|████▋     | 3058/6470 [38:03<43:47,  1.30it/s]
1|3059|Loss: 0.8113309741020203:  47%|████▋     | 3059/6470 [38:03<45:04,  1.26it/s]
1|3060|Loss: 0.8465184569358826:  47%|████▋     | 3059/6470 [38:04<45:04,  1.26it/s]
1|3060|Loss: 0.8465184569358826:  47%|████▋     | 3060/6470 [38:04<44:27,  1.28it/s]
1|3061|Loss: 0.9329276084899902:  47%|████▋     | 3060/6470 [38:04<44:27,  1.28it/s]
1|3061|Loss: 0.9329276084899902:  47%|████▋     | 3061/6470 [38:05<41:40,  1.36it/s]
1|3062|Loss: 0.7568525075912476:  47%|████▋     | 3061/6470 [38:05<41:40,  1.36it/s]
1|3062|Loss: 0.7568525075912476:  47%|████▋     | 3062/6470 [38:05<39:48,  1.43it/s]
1|3063|Loss: 0.8920170664787292:  47%|████▋     | 3062/6470 [38:06<39:48,  1.43it/s]
1|3063|Loss: 0.8920170664787292:  47%|████▋     | 3063/6470 [38:06<38:45,  1.46it/s]
1|3064|Loss: 0.9220932722091675:  47%|████▋     | 3063/6470 [38:06<38:45,  1.46it/s]
1|3064|Loss: 0.9220932722091675:  47%|████▋     | 3064/6470 [38:07<40:09,  1.41it/s]
1|3065|Loss: 0.9546911120414734:  47%|████▋     | 3064/6470 [38:07<40:09,  1.41it/s]
1|3065|Loss: 0.9546911120414734:  47%|████▋     | 3065/6470 [38:07<36:57,  1.54it/s]
1|3066|Loss: 0.935375988483429:  47%|████▋     | 3065/6470 [38:08<36:57,  1.54it/s] 
1|3066|Loss: 0.935375988483429:  47%|████▋     | 3066/6470 [38:08<39:14,  1.45it/s]
1|3067|Loss: 1.0931605100631714:  47%|████▋     | 3066/6470 [38:08<39:14,  1.45it/s]
1|3067|Loss: 1.0931605100631714:  47%|████▋     | 3067/6470 [38:09<37:36,  1.51it/s]
1|3068|Loss: 0.9105042219161987:  47%|████▋     | 3067/6470 [38:09<37:36,  1.51it/s]
1|3068|Loss: 0.9105042219161987:  47%|████▋     | 3068/6470 [38:09<38:14,  1.48it/s]
1|3069|Loss: 0.8848841786384583:  47%|████▋     | 3068/6470 [38:10<38:14,  1.48it/s]
1|3069|Loss: 0.8848841786384583:  47%|████▋     | 3069/6470 [38:10<38:59,  1.45it/s]
1|3070|Loss: 0.8730265498161316:  47%|████▋     | 3069/6470 [38:10<38:59,  1.45it/s]
1|3070|Loss: 0.8730265498161316:  47%|████▋     | 3070/6470 [38:11<36:35,  1.55it/s]
1|3071|Loss: 0.9334171414375305:  47%|████▋     | 3070/6470 [38:11<36:35,  1.55it/s]
1|3071|Loss: 0.9334171414375305:  47%|████▋     | 3071/6470 [38:11<34:49,  1.63it/s]
1|3072|Loss: 0.9482468366622925:  47%|████▋     | 3071/6470 [38:12<34:49,  1.63it/s]
1|3072|Loss: 0.9482468366622925:  47%|████▋     | 3072/6470 [38:12<35:51,  1.58it/s]
1|3073|Loss: 0.8054032325744629:  47%|████▋     | 3072/6470 [38:12<35:51,  1.58it/s]
1|3073|Loss: 0.8054032325744629:  47%|████▋     | 3073/6470 [38:12<35:53,  1.58it/s]
1|3074|Loss: 0.9009592533111572:  47%|████▋     | 3073/6470 [38:13<35:53,  1.58it/s]
1|3074|Loss: 0.9009592533111572:  48%|████▊     | 3074/6470 [38:13<35:51,  1.58it/s]
1|3075|Loss: 0.930359959602356:  48%|████▊     | 3074/6470 [38:14<35:51,  1.58it/s] 
1|3075|Loss: 0.930359959602356:  48%|████▊     | 3075/6470 [38:14<41:02,  1.38it/s]
1|3076|Loss: 0.8105251789093018:  48%|████▊     | 3075/6470 [38:15<41:02,  1.38it/s]
1|3076|Loss: 0.8105251789093018:  48%|████▊     | 3076/6470 [38:15<41:59,  1.35it/s]
1|3077|Loss: 0.917941689491272:  48%|████▊     | 3076/6470 [38:15<41:59,  1.35it/s] 
1|3077|Loss: 0.917941689491272:  48%|████▊     | 3077/6470 [38:15<40:22,  1.40it/s]
1|3078|Loss: 0.9006325006484985:  48%|████▊     | 3077/6470 [38:16<40:22,  1.40it/s]
1|3078|Loss: 0.9006325006484985:  48%|████▊     | 3078/6470 [38:16<41:20,  1.37it/s]
1|3079|Loss: 0.7569280862808228:  48%|████▊     | 3078/6470 [38:17<41:20,  1.37it/s]
1|3079|Loss: 0.7569280862808228:  48%|████▊     | 3079/6470 [38:17<43:10,  1.31it/s]
1|3080|Loss: 0.89430832862854:  48%|████▊     | 3079/6470 [38:17<43:10,  1.31it/s]  
1|3080|Loss: 0.89430832862854:  48%|████▊     | 3080/6470 [38:18<41:53,  1.35it/s]
1|3081|Loss: 0.8470982313156128:  48%|████▊     | 3080/6470 [38:18<41:53,  1.35it/s]
1|3081|Loss: 0.8470982313156128:  48%|████▊     | 3081/6470 [38:19<41:37,  1.36it/s]
1|3082|Loss: 0.9380635023117065:  48%|████▊     | 3081/6470 [38:19<41:37,  1.36it/s]
1|3082|Loss: 0.9380635023117065:  48%|████▊     | 3082/6470 [38:19<42:34,  1.33it/s]
1|3083|Loss: 1.0026220083236694:  48%|████▊     | 3082/6470 [38:20<42:34,  1.33it/s]
1|3083|Loss: 1.0026220083236694:  48%|████▊     | 3083/6470 [38:20<43:19,  1.30it/s]
1|3084|Loss: 0.99863600730896:  48%|████▊     | 3083/6470 [38:21<43:19,  1.30it/s]  
1|3084|Loss: 0.99863600730896:  48%|████▊     | 3084/6470 [38:21<44:16,  1.27it/s]
1|3085|Loss: 0.9130467772483826:  48%|████▊     | 3084/6470 [38:21<44:16,  1.27it/s]
1|3085|Loss: 0.9130467772483826:  48%|████▊     | 3085/6470 [38:22<44:07,  1.28it/s]
1|3086|Loss: 0.9012817144393921:  48%|████▊     | 3085/6470 [38:22<44:07,  1.28it/s]
1|3086|Loss: 0.9012817144393921:  48%|████▊     | 3086/6470 [38:23<44:57,  1.25it/s]
1|3087|Loss: 0.7828668355941772:  48%|████▊     | 3086/6470 [38:23<44:57,  1.25it/s]
1|3087|Loss: 0.7828668355941772:  48%|████▊     | 3087/6470 [38:23<44:37,  1.26it/s]
1|3088|Loss: 0.9014185667037964:  48%|████▊     | 3087/6470 [38:24<44:37,  1.26it/s]
1|3088|Loss: 0.9014185667037964:  48%|████▊     | 3088/6470 [38:24<41:33,  1.36it/s]
1|3089|Loss: 0.7255483865737915:  48%|████▊     | 3088/6470 [38:24<41:33,  1.36it/s]
1|3089|Loss: 0.7255483865737915:  48%|████▊     | 3089/6470 [38:25<40:40,  1.39it/s]
1|3090|Loss: 0.9078587889671326:  48%|████▊     | 3089/6470 [38:25<40:40,  1.39it/s]
1|3090|Loss: 0.9078587889671326:  48%|████▊     | 3090/6470 [38:25<38:22,  1.47it/s]
1|3091|Loss: 0.9372678995132446:  48%|████▊     | 3090/6470 [38:26<38:22,  1.47it/s]
1|3091|Loss: 0.9372678995132446:  48%|████▊     | 3091/6470 [38:26<39:55,  1.41it/s]
1|3092|Loss: 0.8376564383506775:  48%|████▊     | 3091/6470 [38:26<39:55,  1.41it/s]
1|3092|Loss: 0.8376564383506775:  48%|████▊     | 3092/6470 [38:27<39:23,  1.43it/s]
1|3093|Loss: 0.8786811828613281:  48%|████▊     | 3092/6470 [38:27<39:23,  1.43it/s]
1|3093|Loss: 0.8786811828613281:  48%|████▊     | 3093/6470 [38:27<39:36,  1.42it/s]
1|3094|Loss: 0.8887732028961182:  48%|████▊     | 3093/6470 [38:28<39:36,  1.42it/s]
1|3094|Loss: 0.8887732028961182:  48%|████▊     | 3094/6470 [38:28<37:19,  1.51it/s]
1|3095|Loss: 0.9142537117004395:  48%|████▊     | 3094/6470 [38:28<37:19,  1.51it/s]
1|3095|Loss: 0.9142537117004395:  48%|████▊     | 3095/6470 [38:29<36:30,  1.54it/s]
1|3096|Loss: 0.818491518497467:  48%|████▊     | 3095/6470 [38:29<36:30,  1.54it/s] 
1|3096|Loss: 0.818491518497467:  48%|████▊     | 3096/6470 [38:29<37:46,  1.49it/s]
1|3097|Loss: 0.8556926846504211:  48%|████▊     | 3096/6470 [38:30<37:46,  1.49it/s]
1|3097|Loss: 0.8556926846504211:  48%|████▊     | 3097/6470 [38:30<36:44,  1.53it/s]
1|3098|Loss: 0.9269024729728699:  48%|████▊     | 3097/6470 [38:30<36:44,  1.53it/s]
1|3098|Loss: 0.9269024729728699:  48%|████▊     | 3098/6470 [38:31<37:15,  1.51it/s]
1|3099|Loss: 0.9913094639778137:  48%|████▊     | 3098/6470 [38:31<37:15,  1.51it/s]
1|3099|Loss: 0.9913094639778137:  48%|████▊     | 3099/6470 [38:31<39:09,  1.43it/s]
1|3100|Loss: 0.9187909960746765:  48%|████▊     | 3099/6470 [38:32<39:09,  1.43it/s]
1|3100|Loss: 0.9187909960746765:  48%|████▊     | 3100/6470 [38:32<35:14,  1.59it/s]
1|3101|Loss: 0.8421522378921509:  48%|████▊     | 3100/6470 [38:32<35:14,  1.59it/s]
1|3101|Loss: 0.8421522378921509:  48%|████▊     | 3101/6470 [38:32<32:47,  1.71it/s]
1|3102|Loss: 0.9131429195404053:  48%|████▊     | 3101/6470 [38:33<32:47,  1.71it/s]
1|3102|Loss: 0.9131429195404053:  48%|████▊     | 3102/6470 [38:33<36:59,  1.52it/s]
1|3103|Loss: 0.8821814656257629:  48%|████▊     | 3102/6470 [38:34<36:59,  1.52it/s]
1|3103|Loss: 0.8821814656257629:  48%|████▊     | 3103/6470 [38:34<41:24,  1.36it/s]
1|3104|Loss: 0.9242885112762451:  48%|████▊     | 3103/6470 [38:34<41:24,  1.36it/s]
1|3104|Loss: 0.9242885112762451:  48%|████▊     | 3104/6470 [38:35<42:06,  1.33it/s]
1|3105|Loss: 0.9934226870536804:  48%|████▊     | 3104/6470 [38:35<42:06,  1.33it/s]
1|3105|Loss: 0.9934226870536804:  48%|████▊     | 3105/6470 [38:36<41:57,  1.34it/s]
1|3106|Loss: 0.8998716473579407:  48%|████▊     | 3105/6470 [38:36<41:57,  1.34it/s]
1|3106|Loss: 0.8998716473579407:  48%|████▊     | 3106/6470 [38:36<42:59,  1.30it/s]
1|3107|Loss: 0.9609755873680115:  48%|████▊     | 3106/6470 [38:37<42:59,  1.30it/s]
1|3107|Loss: 0.9609755873680115:  48%|████▊     | 3107/6470 [38:37<43:01,  1.30it/s]
1|3108|Loss: 0.7677075862884521:  48%|████▊     | 3107/6470 [38:38<43:01,  1.30it/s]
1|3108|Loss: 0.7677075862884521:  48%|████▊     | 3108/6470 [38:38<42:07,  1.33it/s]
1|3109|Loss: 0.9671157002449036:  48%|████▊     | 3108/6470 [38:38<42:07,  1.33it/s]
1|3109|Loss: 0.9671157002449036:  48%|████▊     | 3109/6470 [38:39<41:55,  1.34it/s]
1|3110|Loss: 0.9159091711044312:  48%|████▊     | 3109/6470 [38:39<41:55,  1.34it/s]
1|3110|Loss: 0.9159091711044312:  48%|████▊     | 3110/6470 [38:39<39:18,  1.42it/s]
1|3111|Loss: 0.9124689698219299:  48%|████▊     | 3110/6470 [38:39<39:18,  1.42it/s]
1|3111|Loss: 0.9124689698219299:  48%|████▊     | 3111/6470 [38:40<34:39,  1.62it/s]
1|3112|Loss: 1.0746647119522095:  48%|████▊     | 3111/6470 [38:40<34:39,  1.62it/s]
1|3112|Loss: 1.0746647119522095:  48%|████▊     | 3112/6470 [38:40<33:11,  1.69it/s]
1|3113|Loss: 0.8494421243667603:  48%|████▊     | 3112/6470 [38:41<33:11,  1.69it/s]
1|3113|Loss: 0.8494421243667603:  48%|████▊     | 3113/6470 [38:41<36:21,  1.54it/s]
1|3114|Loss: 0.8847842216491699:  48%|████▊     | 3113/6470 [38:41<36:21,  1.54it/s]
1|3114|Loss: 0.8847842216491699:  48%|████▊     | 3114/6470 [38:42<38:14,  1.46it/s]
1|3115|Loss: 1.0828348398208618:  48%|████▊     | 3114/6470 [38:42<38:14,  1.46it/s]
1|3115|Loss: 1.0828348398208618:  48%|████▊     | 3115/6470 [38:42<37:59,  1.47it/s]
1|3116|Loss: 0.9174051880836487:  48%|████▊     | 3115/6470 [38:43<37:59,  1.47it/s]
1|3116|Loss: 0.9174051880836487:  48%|████▊     | 3116/6470 [38:43<35:33,  1.57it/s]
1|3117|Loss: 0.8507283329963684:  48%|████▊     | 3116/6470 [38:43<35:33,  1.57it/s]
1|3117|Loss: 0.8507283329963684:  48%|████▊     | 3117/6470 [38:44<36:05,  1.55it/s]
1|3118|Loss: 0.8068951964378357:  48%|████▊     | 3117/6470 [38:44<36:05,  1.55it/s]
1|3118|Loss: 0.8068951964378357:  48%|████▊     | 3118/6470 [38:44<37:32,  1.49it/s]
1|3119|Loss: 0.9239271283149719:  48%|████▊     | 3118/6470 [38:45<37:32,  1.49it/s]
1|3119|Loss: 0.9239271283149719:  48%|████▊     | 3119/6470 [38:45<38:59,  1.43it/s]
1|3120|Loss: 0.9363205432891846:  48%|████▊     | 3119/6470 [38:45<38:59,  1.43it/s]
1|3120|Loss: 0.9363205432891846:  48%|████▊     | 3120/6470 [38:46<38:09,  1.46it/s]
1|3121|Loss: 0.9220535755157471:  48%|████▊     | 3120/6470 [38:46<38:09,  1.46it/s]
1|3121|Loss: 0.9220535755157471:  48%|████▊     | 3121/6470 [38:46<36:28,  1.53it/s]
1|3122|Loss: 0.9772138595581055:  48%|████▊     | 3121/6470 [38:47<36:28,  1.53it/s]
1|3122|Loss: 0.9772138595581055:  48%|████▊     | 3122/6470 [38:47<36:02,  1.55it/s]
1|3123|Loss: 1.0224074125289917:  48%|████▊     | 3122/6470 [38:47<36:02,  1.55it/s]
1|3123|Loss: 1.0224074125289917:  48%|████▊     | 3123/6470 [38:48<36:52,  1.51it/s]
1|3124|Loss: 0.807878851890564:  48%|████▊     | 3123/6470 [38:48<36:52,  1.51it/s] 
1|3124|Loss: 0.807878851890564:  48%|████▊     | 3124/6470 [38:48<39:06,  1.43it/s]
1|3125|Loss: 0.8782882690429688:  48%|████▊     | 3124/6470 [38:49<39:06,  1.43it/s]
1|3125|Loss: 0.8782882690429688:  48%|████▊     | 3125/6470 [38:49<41:13,  1.35it/s]
1|3126|Loss: 0.8781359195709229:  48%|████▊     | 3125/6470 [38:50<41:13,  1.35it/s]
1|3126|Loss: 0.8781359195709229:  48%|████▊     | 3126/6470 [38:50<40:02,  1.39it/s]
1|3127|Loss: 1.048327088356018:  48%|████▊     | 3126/6470 [38:50<40:02,  1.39it/s] 
1|3127|Loss: 1.048327088356018:  48%|████▊     | 3127/6470 [38:51<40:35,  1.37it/s]
1|3128|Loss: 0.8573340773582458:  48%|████▊     | 3127/6470 [38:51<40:35,  1.37it/s]
1|3128|Loss: 0.8573340773582458:  48%|████▊     | 3128/6470 [38:51<41:59,  1.33it/s]
1|3129|Loss: 0.9932562708854675:  48%|████▊     | 3128/6470 [38:52<41:59,  1.33it/s]
1|3129|Loss: 0.9932562708854675:  48%|████▊     | 3129/6470 [38:52<39:29,  1.41it/s]
1|3130|Loss: 0.8969142436981201:  48%|████▊     | 3129/6470 [38:52<39:29,  1.41it/s]
1|3130|Loss: 0.8969142436981201:  48%|████▊     | 3130/6470 [38:53<38:48,  1.43it/s]
1|3131|Loss: 0.8393452167510986:  48%|████▊     | 3130/6470 [38:53<38:48,  1.43it/s]
1|3131|Loss: 0.8393452167510986:  48%|████▊     | 3131/6470 [38:53<37:50,  1.47it/s]
1|3132|Loss: 0.7916637659072876:  48%|████▊     | 3131/6470 [38:54<37:50,  1.47it/s]
1|3132|Loss: 0.7916637659072876:  48%|████▊     | 3132/6470 [38:54<37:58,  1.46it/s]
1|3133|Loss: 1.0313116312026978:  48%|████▊     | 3132/6470 [38:54<37:58,  1.46it/s]
1|3133|Loss: 1.0313116312026978:  48%|████▊     | 3133/6470 [38:55<35:43,  1.56it/s]
1|3134|Loss: 0.8943629860877991:  48%|████▊     | 3133/6470 [38:55<35:43,  1.56it/s]
1|3134|Loss: 0.8943629860877991:  48%|████▊     | 3134/6470 [38:55<36:07,  1.54it/s]
1|3135|Loss: 0.9269159436225891:  48%|████▊     | 3134/6470 [38:56<36:07,  1.54it/s]
1|3135|Loss: 0.9269159436225891:  48%|████▊     | 3135/6470 [38:56<35:52,  1.55it/s]
1|3136|Loss: 1.1207020282745361:  48%|████▊     | 3135/6470 [38:56<35:52,  1.55it/s]
1|3136|Loss: 1.1207020282745361:  48%|████▊     | 3136/6470 [38:57<34:42,  1.60it/s]
1|3137|Loss: 0.814115047454834:  48%|████▊     | 3136/6470 [38:57<34:42,  1.60it/s] 
1|3137|Loss: 0.814115047454834:  48%|████▊     | 3137/6470 [38:57<36:55,  1.50it/s]
1|3138|Loss: 1.0953781604766846:  48%|████▊     | 3137/6470 [38:58<36:55,  1.50it/s]
1|3138|Loss: 1.0953781604766846:  49%|████▊     | 3138/6470 [38:58<36:52,  1.51it/s]
1|3139|Loss: 0.9789281487464905:  49%|████▊     | 3138/6470 [38:58<36:52,  1.51it/s]
1|3139|Loss: 0.9789281487464905:  49%|████▊     | 3139/6470 [38:59<39:05,  1.42it/s]
1|3140|Loss: 0.8392423987388611:  49%|████▊     | 3139/6470 [38:59<39:05,  1.42it/s]
1|3140|Loss: 0.8392423987388611:  49%|████▊     | 3140/6470 [38:59<37:33,  1.48it/s]
1|3141|Loss: 0.9434769749641418:  49%|████▊     | 3140/6470 [39:00<37:33,  1.48it/s]
1|3141|Loss: 0.9434769749641418:  49%|████▊     | 3141/6470 [39:00<37:50,  1.47it/s]
1|3142|Loss: 0.8953817486763:  49%|████▊     | 3141/6470 [39:00<37:50,  1.47it/s]   
1|3142|Loss: 0.8953817486763:  49%|████▊     | 3142/6470 [39:01<38:08,  1.45it/s]
1|3143|Loss: 0.9859516620635986:  49%|████▊     | 3142/6470 [39:01<38:08,  1.45it/s]
1|3143|Loss: 0.9859516620635986:  49%|████▊     | 3143/6470 [39:01<37:32,  1.48it/s]
1|3144|Loss: 0.9890502691268921:  49%|████▊     | 3143/6470 [39:02<37:32,  1.48it/s]
1|3144|Loss: 0.9890502691268921:  49%|████▊     | 3144/6470 [39:02<38:13,  1.45it/s]
1|3145|Loss: 0.9236971735954285:  49%|████▊     | 3144/6470 [39:02<38:13,  1.45it/s]
1|3145|Loss: 0.9236971735954285:  49%|████▊     | 3145/6470 [39:03<36:36,  1.51it/s]
1|3146|Loss: 0.808744490146637:  49%|████▊     | 3145/6470 [39:03<36:36,  1.51it/s] 
1|3146|Loss: 0.808744490146637:  49%|████▊     | 3146/6470 [39:03<34:34,  1.60it/s]
1|3147|Loss: 0.8182618618011475:  49%|████▊     | 3146/6470 [39:04<34:34,  1.60it/s]
1|3147|Loss: 0.8182618618011475:  49%|████▊     | 3147/6470 [39:04<35:12,  1.57it/s]
1|3148|Loss: 0.9492331147193909:  49%|████▊     | 3147/6470 [39:04<35:12,  1.57it/s]
1|3148|Loss: 0.9492331147193909:  49%|████▊     | 3148/6470 [39:05<36:53,  1.50it/s]
1|3149|Loss: 0.9072961211204529:  49%|████▊     | 3148/6470 [39:05<36:53,  1.50it/s]
1|3149|Loss: 0.9072961211204529:  49%|████▊     | 3149/6470 [39:05<38:19,  1.44it/s]
1|3150|Loss: 0.8437842726707458:  49%|████▊     | 3149/6470 [39:06<38:19,  1.44it/s]
1|3150|Loss: 0.8437842726707458:  49%|████▊     | 3150/6470 [39:06<37:37,  1.47it/s]
1|3151|Loss: 0.7740583419799805:  49%|████▊     | 3150/6470 [39:06<37:37,  1.47it/s]
1|3151|Loss: 0.7740583419799805:  49%|████▊     | 3151/6470 [39:07<38:10,  1.45it/s]
1|3152|Loss: 0.9976693987846375:  49%|████▊     | 3151/6470 [39:07<38:10,  1.45it/s]
1|3152|Loss: 0.9976693987846375:  49%|████▊     | 3152/6470 [39:08<40:09,  1.38it/s]
1|3153|Loss: 0.9403814077377319:  49%|████▊     | 3152/6470 [39:08<40:09,  1.38it/s]
1|3153|Loss: 0.9403814077377319:  49%|████▊     | 3153/6470 [39:08<39:22,  1.40it/s]
1|3154|Loss: 1.033448576927185:  49%|████▊     | 3153/6470 [39:09<39:22,  1.40it/s] 
1|3154|Loss: 1.033448576927185:  49%|████▊     | 3154/6470 [39:09<40:06,  1.38it/s]
1|3155|Loss: 0.7831746935844421:  49%|████▊     | 3154/6470 [39:09<40:06,  1.38it/s]
1|3155|Loss: 0.7831746935844421:  49%|████▉     | 3155/6470 [39:10<40:04,  1.38it/s]
1|3156|Loss: 0.8650214076042175:  49%|████▉     | 3155/6470 [39:10<40:04,  1.38it/s]
1|3156|Loss: 0.8650214076042175:  49%|████▉     | 3156/6470 [39:10<37:18,  1.48it/s]
1|3157|Loss: 0.9652993679046631:  49%|████▉     | 3156/6470 [39:11<37:18,  1.48it/s]
1|3157|Loss: 0.9652993679046631:  49%|████▉     | 3157/6470 [39:11<37:33,  1.47it/s]
1|3158|Loss: 0.8223286867141724:  49%|████▉     | 3157/6470 [39:11<37:33,  1.47it/s]
1|3158|Loss: 0.8223286867141724:  49%|████▉     | 3158/6470 [39:12<40:11,  1.37it/s]
1|3159|Loss: 0.9742434024810791:  49%|████▉     | 3158/6470 [39:12<40:11,  1.37it/s]
1|3159|Loss: 0.9742434024810791:  49%|████▉     | 3159/6470 [39:12<39:22,  1.40it/s]
1|3160|Loss: 0.9848880171775818:  49%|████▉     | 3159/6470 [39:13<39:22,  1.40it/s]
1|3160|Loss: 0.9848880171775818:  49%|████▉     | 3160/6470 [39:13<35:35,  1.55it/s]
1|3161|Loss: 0.7853733897209167:  49%|████▉     | 3160/6470 [39:13<35:35,  1.55it/s]
1|3161|Loss: 0.7853733897209167:  49%|████▉     | 3161/6470 [39:14<34:45,  1.59it/s]
1|3162|Loss: 0.9498040080070496:  49%|████▉     | 3161/6470 [39:14<34:45,  1.59it/s]
1|3162|Loss: 0.9498040080070496:  49%|████▉     | 3162/6470 [39:14<32:30,  1.70it/s]
1|3163|Loss: 0.785282552242279:  49%|████▉     | 3162/6470 [39:14<32:30,  1.70it/s] 
1|3163|Loss: 0.785282552242279:  49%|████▉     | 3163/6470 [39:15<32:37,  1.69it/s]
1|3164|Loss: 0.9458848834037781:  49%|████▉     | 3163/6470 [39:15<32:37,  1.69it/s]
1|3164|Loss: 0.9458848834037781:  49%|████▉     | 3164/6470 [39:15<35:55,  1.53it/s]
1|3165|Loss: 0.7366609573364258:  49%|████▉     | 3164/6470 [39:16<35:55,  1.53it/s]
1|3165|Loss: 0.7366609573364258:  49%|████▉     | 3165/6470 [39:16<38:33,  1.43it/s]
1|3166|Loss: 0.6943953037261963:  49%|████▉     | 3165/6470 [39:17<38:33,  1.43it/s]
1|3166|Loss: 0.6943953037261963:  49%|████▉     | 3166/6470 [39:18<48:57,  1.12it/s]
1|3167|Loss: 0.7608519792556763:  49%|████▉     | 3166/6470 [39:18<48:57,  1.12it/s]
1|3167|Loss: 0.7608519792556763:  49%|████▉     | 3167/6470 [39:19<51:01,  1.08it/s]
1|3168|Loss: 0.9000293016433716:  49%|████▉     | 3167/6470 [39:19<51:01,  1.08it/s]
1|3168|Loss: 0.9000293016433716:  49%|████▉     | 3168/6470 [39:19<47:00,  1.17it/s]
1|3169|Loss: 0.9784968495368958:  49%|████▉     | 3168/6470 [39:20<47:00,  1.17it/s]
1|3169|Loss: 0.9784968495368958:  49%|████▉     | 3169/6470 [39:20<44:52,  1.23it/s]
1|3170|Loss: 1.1711087226867676:  49%|████▉     | 3169/6470 [39:20<44:52,  1.23it/s]
1|3170|Loss: 1.1711087226867676:  49%|████▉     | 3170/6470 [39:21<43:40,  1.26it/s]
1|3171|Loss: 0.8667887449264526:  49%|████▉     | 3170/6470 [39:21<43:40,  1.26it/s]
1|3171|Loss: 0.8667887449264526:  49%|████▉     | 3171/6470 [39:22<43:32,  1.26it/s]
1|3172|Loss: 0.9597976207733154:  49%|████▉     | 3171/6470 [39:22<43:32,  1.26it/s]
1|3172|Loss: 0.9597976207733154:  49%|████▉     | 3172/6470 [39:22<42:24,  1.30it/s]
1|3173|Loss: 0.8141230940818787:  49%|████▉     | 3172/6470 [39:23<42:24,  1.30it/s]
1|3173|Loss: 0.8141230940818787:  49%|████▉     | 3173/6470 [39:23<42:28,  1.29it/s]
1|3174|Loss: 0.9191517233848572:  49%|████▉     | 3173/6470 [39:23<42:28,  1.29it/s]
1|3174|Loss: 0.9191517233848572:  49%|████▉     | 3174/6470 [39:24<39:36,  1.39it/s]
1|3175|Loss: 0.9301357865333557:  49%|████▉     | 3174/6470 [39:24<39:36,  1.39it/s]
1|3175|Loss: 0.9301357865333557:  49%|████▉     | 3175/6470 [39:24<39:12,  1.40it/s]
1|3176|Loss: 0.8186145424842834:  49%|████▉     | 3175/6470 [39:25<39:12,  1.40it/s]
1|3176|Loss: 0.8186145424842834:  49%|████▉     | 3176/6470 [39:25<39:47,  1.38it/s]
1|3177|Loss: 0.8951426148414612:  49%|████▉     | 3176/6470 [39:25<39:47,  1.38it/s]
1|3177|Loss: 0.8951426148414612:  49%|████▉     | 3177/6470 [39:26<38:31,  1.42it/s]
1|3178|Loss: 0.8022634387016296:  49%|████▉     | 3177/6470 [39:26<38:31,  1.42it/s]
1|3178|Loss: 0.8022634387016296:  49%|████▉     | 3178/6470 [39:26<38:13,  1.44it/s]
1|3179|Loss: 0.8795554041862488:  49%|████▉     | 3178/6470 [39:27<38:13,  1.44it/s]
1|3179|Loss: 0.8795554041862488:  49%|████▉     | 3179/6470 [39:27<37:10,  1.48it/s]
1|3180|Loss: 0.9009316563606262:  49%|████▉     | 3179/6470 [39:27<37:10,  1.48it/s]
1|3180|Loss: 0.9009316563606262:  49%|████▉     | 3180/6470 [39:28<35:29,  1.55it/s]
1|3181|Loss: 0.8286995887756348:  49%|████▉     | 3180/6470 [39:28<35:29,  1.55it/s]
1|3181|Loss: 0.8286995887756348:  49%|████▉     | 3181/6470 [39:28<34:20,  1.60it/s]
1|3182|Loss: 0.7899064421653748:  49%|████▉     | 3181/6470 [39:29<34:20,  1.60it/s]
1|3182|Loss: 0.7899064421653748:  49%|████▉     | 3182/6470 [39:29<34:09,  1.60it/s]
1|3183|Loss: 0.9821145534515381:  49%|████▉     | 3182/6470 [39:29<34:09,  1.60it/s]
1|3183|Loss: 0.9821145534515381:  49%|████▉     | 3183/6470 [39:29<34:27,  1.59it/s]
1|3184|Loss: 0.8693715333938599:  49%|████▉     | 3183/6470 [39:30<34:27,  1.59it/s]
1|3184|Loss: 0.8693715333938599:  49%|████▉     | 3184/6470 [39:30<37:49,  1.45it/s]
1|3185|Loss: 0.802517831325531:  49%|████▉     | 3184/6470 [39:31<37:49,  1.45it/s] 
1|3185|Loss: 0.802517831325531:  49%|████▉     | 3185/6470 [39:31<40:17,  1.36it/s]
1|3186|Loss: 1.0308582782745361:  49%|████▉     | 3185/6470 [39:32<40:17,  1.36it/s]
1|3186|Loss: 1.0308582782745361:  49%|████▉     | 3186/6470 [39:32<36:53,  1.48it/s]
1|3187|Loss: 0.8631269931793213:  49%|████▉     | 3186/6470 [39:32<36:53,  1.48it/s]
1|3187|Loss: 0.8631269931793213:  49%|████▉     | 3187/6470 [39:32<35:43,  1.53it/s]
1|3188|Loss: 0.9860214591026306:  49%|████▉     | 3187/6470 [39:33<35:43,  1.53it/s]
1|3188|Loss: 0.9860214591026306:  49%|████▉     | 3188/6470 [39:33<36:21,  1.50it/s]
1|3189|Loss: 0.9156122207641602:  49%|████▉     | 3188/6470 [39:33<36:21,  1.50it/s]
1|3189|Loss: 0.9156122207641602:  49%|████▉     | 3189/6470 [39:34<35:51,  1.52it/s]
1|3190|Loss: 1.0181854963302612:  49%|████▉     | 3189/6470 [39:34<35:51,  1.52it/s]
1|3190|Loss: 1.0181854963302612:  49%|████▉     | 3190/6470 [39:34<35:34,  1.54it/s]
1|3191|Loss: 0.9279227256774902:  49%|████▉     | 3190/6470 [39:35<35:34,  1.54it/s]
1|3191|Loss: 0.9279227256774902:  49%|████▉     | 3191/6470 [39:35<33:57,  1.61it/s]
1|3192|Loss: 1.0511788129806519:  49%|████▉     | 3191/6470 [39:35<33:57,  1.61it/s]
1|3192|Loss: 1.0511788129806519:  49%|████▉     | 3192/6470 [39:36<35:03,  1.56it/s]
1|3193|Loss: 0.9404686689376831:  49%|████▉     | 3192/6470 [39:36<35:03,  1.56it/s]
1|3193|Loss: 0.9404686689376831:  49%|████▉     | 3193/6470 [39:36<38:08,  1.43it/s]
1|3194|Loss: 0.8553075790405273:  49%|████▉     | 3193/6470 [39:37<38:08,  1.43it/s]
1|3194|Loss: 0.8553075790405273:  49%|████▉     | 3194/6470 [39:37<38:51,  1.41it/s]
1|3195|Loss: 0.7853206992149353:  49%|████▉     | 3194/6470 [39:38<38:51,  1.41it/s]
1|3195|Loss: 0.7853206992149353:  49%|████▉     | 3195/6470 [39:38<42:22,  1.29it/s]
1|3196|Loss: 0.7768656611442566:  49%|████▉     | 3195/6470 [39:39<42:22,  1.29it/s]
1|3196|Loss: 0.7768656611442566:  49%|████▉     | 3196/6470 [39:39<45:05,  1.21it/s]
1|3197|Loss: 1.0738791227340698:  49%|████▉     | 3196/6470 [39:39<45:05,  1.21it/s]
1|3197|Loss: 1.0738791227340698:  49%|████▉     | 3197/6470 [39:40<42:55,  1.27it/s]
1|3198|Loss: 0.9682085514068604:  49%|████▉     | 3197/6470 [39:40<42:55,  1.27it/s]
1|3198|Loss: 0.9682085514068604:  49%|████▉     | 3198/6470 [39:40<41:22,  1.32it/s]
1|3199|Loss: 0.8914534449577332:  49%|████▉     | 3198/6470 [39:41<41:22,  1.32it/s]
1|3199|Loss: 0.8914534449577332:  49%|████▉     | 3199/6470 [39:41<38:45,  1.41it/s]
1|3200|Loss: 0.9117753505706787:  49%|████▉     | 3199/6470 [39:41<38:45,  1.41it/s]
1|3200|Loss: 0.9117753505706787:  49%|████▉     | 3200/6470 [39:42<38:07,  1.43it/s]
1|3201|Loss: 1.0764130353927612:  49%|████▉     | 3200/6470 [39:42<38:07,  1.43it/s]
1|3201|Loss: 1.0764130353927612:  49%|████▉     | 3201/6470 [39:42<36:40,  1.49it/s]
1|3202|Loss: 0.8619797229766846:  49%|████▉     | 3201/6470 [39:43<36:40,  1.49it/s]
1|3202|Loss: 0.8619797229766846:  49%|████▉     | 3202/6470 [39:43<35:39,  1.53it/s]
1|3203|Loss: 0.824421226978302:  49%|████▉     | 3202/6470 [39:43<35:39,  1.53it/s] 
1|3203|Loss: 0.824421226978302:  50%|████▉     | 3203/6470 [39:43<34:28,  1.58it/s]
1|3204|Loss: 0.826241672039032:  50%|████▉     | 3203/6470 [39:44<34:28,  1.58it/s]
1|3204|Loss: 0.826241672039032:  50%|████▉     | 3204/6470 [39:44<33:29,  1.63it/s]
1|3205|Loss: 0.7202175259590149:  50%|████▉     | 3204/6470 [39:44<33:29,  1.63it/s]
1|3205|Loss: 0.7202175259590149:  50%|████▉     | 3205/6470 [39:45<33:19,  1.63it/s]
1|3206|Loss: 0.9410549998283386:  50%|████▉     | 3205/6470 [39:45<33:19,  1.63it/s]
1|3206|Loss: 0.9410549998283386:  50%|████▉     | 3206/6470 [39:45<34:51,  1.56it/s]
1|3207|Loss: 1.0481089353561401:  50%|████▉     | 3206/6470 [39:46<34:51,  1.56it/s]
1|3207|Loss: 1.0481089353561401:  50%|████▉     | 3207/6470 [39:46<33:57,  1.60it/s]
1|3208|Loss: 1.0214478969573975:  50%|████▉     | 3207/6470 [39:46<33:57,  1.60it/s]
1|3208|Loss: 1.0214478969573975:  50%|████▉     | 3208/6470 [39:47<34:48,  1.56it/s]
1|3209|Loss: 0.8849799633026123:  50%|████▉     | 3208/6470 [39:47<34:48,  1.56it/s]
1|3209|Loss: 0.8849799633026123:  50%|████▉     | 3209/6470 [39:47<35:06,  1.55it/s]
1|3210|Loss: 0.8568769097328186:  50%|████▉     | 3209/6470 [39:48<35:06,  1.55it/s]
1|3210|Loss: 0.8568769097328186:  50%|████▉     | 3210/6470 [39:48<35:50,  1.52it/s]
1|3211|Loss: 1.078737735748291:  50%|████▉     | 3210/6470 [39:48<35:50,  1.52it/s] 
1|3211|Loss: 1.078737735748291:  50%|████▉     | 3211/6470 [39:49<36:57,  1.47it/s]
1|3212|Loss: 0.8711739778518677:  50%|████▉     | 3211/6470 [39:49<36:57,  1.47it/s]
1|3212|Loss: 0.8711739778518677:  50%|████▉     | 3212/6470 [39:49<37:53,  1.43it/s]
1|3213|Loss: 0.7762303352355957:  50%|████▉     | 3212/6470 [39:50<37:53,  1.43it/s]
1|3213|Loss: 0.7762303352355957:  50%|████▉     | 3213/6470 [39:50<38:29,  1.41it/s]
1|3214|Loss: 0.7779678106307983:  50%|████▉     | 3213/6470 [39:51<38:29,  1.41it/s]
1|3214|Loss: 0.7779678106307983:  50%|████▉     | 3214/6470 [39:51<38:24,  1.41it/s]
1|3215|Loss: 1.0607954263687134:  50%|████▉     | 3214/6470 [39:51<38:24,  1.41it/s]
1|3215|Loss: 1.0607954263687134:  50%|████▉     | 3215/6470 [39:52<38:21,  1.41it/s]
1|3216|Loss: 0.8072533011436462:  50%|████▉     | 3215/6470 [39:52<38:21,  1.41it/s]
1|3216|Loss: 0.8072533011436462:  50%|████▉     | 3216/6470 [39:52<39:55,  1.36it/s]
1|3217|Loss: 0.7407875657081604:  50%|████▉     | 3216/6470 [39:53<39:55,  1.36it/s]
1|3217|Loss: 0.7407875657081604:  50%|████▉     | 3217/6470 [39:53<40:17,  1.35it/s]
1|3218|Loss: 0.9865818619728088:  50%|████▉     | 3217/6470 [39:54<40:17,  1.35it/s]
1|3218|Loss: 0.9865818619728088:  50%|████▉     | 3218/6470 [39:54<41:42,  1.30it/s]
1|3219|Loss: 0.8128891587257385:  50%|████▉     | 3218/6470 [39:54<41:42,  1.30it/s]
1|3219|Loss: 0.8128891587257385:  50%|████▉     | 3219/6470 [39:55<41:14,  1.31it/s]
1|3220|Loss: 0.8881962299346924:  50%|████▉     | 3219/6470 [39:55<41:14,  1.31it/s]
1|3220|Loss: 0.8881962299346924:  50%|████▉     | 3220/6470 [39:55<40:09,  1.35it/s]
1|3221|Loss: 0.8715922832489014:  50%|████▉     | 3220/6470 [39:56<40:09,  1.35it/s]
1|3221|Loss: 0.8715922832489014:  50%|████▉     | 3221/6470 [39:56<37:52,  1.43it/s]
1|3222|Loss: 0.7392749786376953:  50%|████▉     | 3221/6470 [39:56<37:52,  1.43it/s]
1|3222|Loss: 0.7392749786376953:  50%|████▉     | 3222/6470 [39:57<41:15,  1.31it/s]
1|3223|Loss: 0.6855273842811584:  50%|████▉     | 3222/6470 [39:57<41:15,  1.31it/s]
1|3223|Loss: 0.6855273842811584:  50%|████▉     | 3223/6470 [39:58<43:05,  1.26it/s]
1|3224|Loss: 0.8555800914764404:  50%|████▉     | 3223/6470 [39:58<43:05,  1.26it/s]
1|3224|Loss: 0.8555800914764404:  50%|████▉     | 3224/6470 [39:59<44:01,  1.23it/s]
1|3225|Loss: 0.8685698509216309:  50%|████▉     | 3224/6470 [39:59<44:01,  1.23it/s]
1|3225|Loss: 0.8685698509216309:  50%|████▉     | 3225/6470 [39:59<41:40,  1.30it/s]
1|3226|Loss: 0.8936987519264221:  50%|████▉     | 3225/6470 [40:00<41:40,  1.30it/s]
1|3226|Loss: 0.8936987519264221:  50%|████▉     | 3226/6470 [40:00<42:02,  1.29it/s]
1|3227|Loss: 0.7404577136039734:  50%|████▉     | 3226/6470 [40:01<42:02,  1.29it/s]
1|3227|Loss: 0.7404577136039734:  50%|████▉     | 3227/6470 [40:01<42:24,  1.27it/s]
1|3228|Loss: 0.93393874168396:  50%|████▉     | 3227/6470 [40:01<42:24,  1.27it/s]  
1|3228|Loss: 0.93393874168396:  50%|████▉     | 3228/6470 [40:02<43:06,  1.25it/s]
1|3229|Loss: 1.0387589931488037:  50%|████▉     | 3228/6470 [40:02<43:06,  1.25it/s]
1|3229|Loss: 1.0387589931488037:  50%|████▉     | 3229/6470 [40:03<43:34,  1.24it/s]
1|3230|Loss: 0.8701047897338867:  50%|████▉     | 3229/6470 [40:03<43:34,  1.24it/s]
1|3230|Loss: 0.8701047897338867:  50%|████▉     | 3230/6470 [40:03<40:24,  1.34it/s]
1|3231|Loss: 0.914585530757904:  50%|████▉     | 3230/6470 [40:03<40:24,  1.34it/s] 
1|3231|Loss: 0.914585530757904:  50%|████▉     | 3231/6470 [40:04<38:07,  1.42it/s]
1|3232|Loss: 0.9049179553985596:  50%|████▉     | 3231/6470 [40:04<38:07,  1.42it/s]
1|3232|Loss: 0.9049179553985596:  50%|████▉     | 3232/6470 [40:04<37:46,  1.43it/s]
1|3233|Loss: 0.8591249585151672:  50%|████▉     | 3232/6470 [40:05<37:46,  1.43it/s]
1|3233|Loss: 0.8591249585151672:  50%|████▉     | 3233/6470 [40:05<39:23,  1.37it/s]
1|3234|Loss: 0.9674823880195618:  50%|████▉     | 3233/6470 [40:06<39:23,  1.37it/s]
1|3234|Loss: 0.9674823880195618:  50%|████▉     | 3234/6470 [40:06<36:11,  1.49it/s]
1|3235|Loss: 0.9816425442695618:  50%|████▉     | 3234/6470 [40:06<36:11,  1.49it/s]
1|3235|Loss: 0.9816425442695618:  50%|█████     | 3235/6470 [40:07<37:46,  1.43it/s]
1|3236|Loss: 0.9554378390312195:  50%|█████     | 3235/6470 [40:07<37:46,  1.43it/s]
1|3236|Loss: 0.9554378390312195:  50%|█████     | 3236/6470 [40:07<39:41,  1.36it/s]
1|3237|Loss: 0.8152260780334473:  50%|█████     | 3236/6470 [40:08<39:41,  1.36it/s]
1|3237|Loss: 0.8152260780334473:  50%|█████     | 3237/6470 [40:08<40:03,  1.35it/s]
1|3238|Loss: 0.7594500780105591:  50%|█████     | 3237/6470 [40:09<40:03,  1.35it/s]
1|3238|Loss: 0.7594500780105591:  50%|█████     | 3238/6470 [40:09<41:56,  1.28it/s]
1|3239|Loss: 0.7359390258789062:  50%|█████     | 3238/6470 [40:09<41:56,  1.28it/s]
1|3239|Loss: 0.7359390258789062:  50%|█████     | 3239/6470 [40:10<41:57,  1.28it/s]
1|3240|Loss: 0.9262040853500366:  50%|█████     | 3239/6470 [40:10<41:57,  1.28it/s]
1|3240|Loss: 0.9262040853500366:  50%|█████     | 3240/6470 [40:10<41:25,  1.30it/s]
1|3241|Loss: 0.8543773889541626:  50%|█████     | 3240/6470 [40:11<41:25,  1.30it/s]
1|3241|Loss: 0.8543773889541626:  50%|█████     | 3241/6470 [40:11<40:32,  1.33it/s]
1|3242|Loss: 0.8740500807762146:  50%|█████     | 3241/6470 [40:12<40:32,  1.33it/s]
1|3242|Loss: 0.8740500807762146:  50%|█████     | 3242/6470 [40:12<40:06,  1.34it/s]
1|3243|Loss: 0.9399924874305725:  50%|█████     | 3242/6470 [40:12<40:06,  1.34it/s]
1|3243|Loss: 0.9399924874305725:  50%|█████     | 3243/6470 [40:13<37:34,  1.43it/s]
1|3244|Loss: 0.7994840741157532:  50%|█████     | 3243/6470 [40:13<37:34,  1.43it/s]
1|3244|Loss: 0.7994840741157532:  50%|█████     | 3244/6470 [40:13<36:00,  1.49it/s]
1|3245|Loss: 0.9109562039375305:  50%|█████     | 3244/6470 [40:14<36:00,  1.49it/s]
1|3245|Loss: 0.9109562039375305:  50%|█████     | 3245/6470 [40:14<37:19,  1.44it/s]
1|3246|Loss: 0.8668702840805054:  50%|█████     | 3245/6470 [40:14<37:19,  1.44it/s]
1|3246|Loss: 0.8668702840805054:  50%|█████     | 3246/6470 [40:15<38:11,  1.41it/s]
1|3247|Loss: 0.9781426787376404:  50%|█████     | 3246/6470 [40:15<38:11,  1.41it/s]
1|3247|Loss: 0.9781426787376404:  50%|█████     | 3247/6470 [40:15<36:34,  1.47it/s]
1|3248|Loss: 0.918278694152832:  50%|█████     | 3247/6470 [40:16<36:34,  1.47it/s] 
1|3248|Loss: 0.918278694152832:  50%|█████     | 3248/6470 [40:16<35:46,  1.50it/s]
1|3249|Loss: 0.8459858894348145:  50%|█████     | 3248/6470 [40:16<35:46,  1.50it/s]
1|3249|Loss: 0.8459858894348145:  50%|█████     | 3249/6470 [40:17<36:41,  1.46it/s]
1|3250|Loss: 0.8899496793746948:  50%|█████     | 3249/6470 [40:17<36:41,  1.46it/s]
1|3250|Loss: 0.8899496793746948:  50%|█████     | 3250/6470 [40:17<36:24,  1.47it/s]
1|3251|Loss: 0.8279872536659241:  50%|█████     | 3250/6470 [40:18<36:24,  1.47it/s]
1|3251|Loss: 0.8279872536659241:  50%|█████     | 3251/6470 [40:18<35:48,  1.50it/s]
1|3252|Loss: 0.9198591709136963:  50%|█████     | 3251/6470 [40:18<35:48,  1.50it/s]
1|3252|Loss: 0.9198591709136963:  50%|█████     | 3252/6470 [40:19<35:19,  1.52it/s]
1|3253|Loss: 0.8163973093032837:  50%|█████     | 3252/6470 [40:19<35:19,  1.52it/s]
1|3253|Loss: 0.8163973093032837:  50%|█████     | 3253/6470 [40:19<35:03,  1.53it/s]
1|3254|Loss: 0.9410485625267029:  50%|█████     | 3253/6470 [40:20<35:03,  1.53it/s]
1|3254|Loss: 0.9410485625267029:  50%|█████     | 3254/6470 [40:20<33:06,  1.62it/s]
1|3255|Loss: 0.9056754112243652:  50%|█████     | 3254/6470 [40:20<33:06,  1.62it/s]
1|3255|Loss: 0.9056754112243652:  50%|█████     | 3255/6470 [40:20<33:02,  1.62it/s]
1|3256|Loss: 0.7568114995956421:  50%|█████     | 3255/6470 [40:21<33:02,  1.62it/s]
1|3256|Loss: 0.7568114995956421:  50%|█████     | 3256/6470 [40:21<34:43,  1.54it/s]
1|3257|Loss: 0.6899177432060242:  50%|█████     | 3256/6470 [40:21<34:43,  1.54it/s]
1|3257|Loss: 0.6899177432060242:  50%|█████     | 3257/6470 [40:22<36:03,  1.49it/s]
1|3258|Loss: 0.8223364353179932:  50%|█████     | 3257/6470 [40:22<36:03,  1.49it/s]
1|3258|Loss: 0.8223364353179932:  50%|█████     | 3258/6470 [40:23<38:21,  1.40it/s]
1|3259|Loss: 0.929845929145813:  50%|█████     | 3258/6470 [40:23<38:21,  1.40it/s] 
1|3259|Loss: 0.929845929145813:  50%|█████     | 3259/6470 [40:23<38:32,  1.39it/s]
1|3260|Loss: 0.8745977878570557:  50%|█████     | 3259/6470 [40:24<38:32,  1.39it/s]
1|3260|Loss: 0.8745977878570557:  50%|█████     | 3260/6470 [40:24<38:13,  1.40it/s]
1|3261|Loss: 0.8800570964813232:  50%|█████     | 3260/6470 [40:24<38:13,  1.40it/s]
1|3261|Loss: 0.8800570964813232:  50%|█████     | 3261/6470 [40:25<35:09,  1.52it/s]
1|3262|Loss: 0.8527927398681641:  50%|█████     | 3261/6470 [40:25<35:09,  1.52it/s]
1|3262|Loss: 0.8527927398681641:  50%|█████     | 3262/6470 [40:25<37:23,  1.43it/s]
1|3263|Loss: 0.8369297981262207:  50%|█████     | 3262/6470 [40:26<37:23,  1.43it/s]
1|3263|Loss: 0.8369297981262207:  50%|█████     | 3263/6470 [40:26<39:44,  1.34it/s]
1|3264|Loss: 0.9223297834396362:  50%|█████     | 3263/6470 [40:27<39:44,  1.34it/s]
1|3264|Loss: 0.9223297834396362:  50%|█████     | 3264/6470 [40:27<37:02,  1.44it/s]
1|3265|Loss: 0.8996921181678772:  50%|█████     | 3264/6470 [40:27<37:02,  1.44it/s]
1|3265|Loss: 0.8996921181678772:  50%|█████     | 3265/6470 [40:28<37:38,  1.42it/s]
1|3266|Loss: 0.9566220641136169:  50%|█████     | 3265/6470 [40:28<37:38,  1.42it/s]
1|3266|Loss: 0.9566220641136169:  50%|█████     | 3266/6470 [40:28<35:31,  1.50it/s]
1|3267|Loss: 0.785445511341095:  50%|█████     | 3266/6470 [40:28<35:31,  1.50it/s] 
1|3267|Loss: 0.785445511341095:  50%|█████     | 3267/6470 [40:29<34:25,  1.55it/s]
1|3268|Loss: 0.9586282968521118:  50%|█████     | 3267/6470 [40:29<34:25,  1.55it/s]
1|3268|Loss: 0.9586282968521118:  51%|█████     | 3268/6470 [40:29<33:08,  1.61it/s]
1|3269|Loss: 0.8986642956733704:  51%|█████     | 3268/6470 [40:30<33:08,  1.61it/s]
1|3269|Loss: 0.8986642956733704:  51%|█████     | 3269/6470 [40:30<36:11,  1.47it/s]
1|3270|Loss: 0.95096755027771:  51%|█████     | 3269/6470 [40:30<36:11,  1.47it/s]  
1|3270|Loss: 0.95096755027771:  51%|█████     | 3270/6470 [40:31<36:14,  1.47it/s]
1|3271|Loss: 1.0284044742584229:  51%|█████     | 3270/6470 [40:31<36:14,  1.47it/s]
1|3271|Loss: 1.0284044742584229:  51%|█████     | 3271/6470 [40:31<35:16,  1.51it/s]
1|3272|Loss: 0.771228015422821:  51%|█████     | 3271/6470 [40:32<35:16,  1.51it/s] 
1|3272|Loss: 0.771228015422821:  51%|█████     | 3272/6470 [40:32<36:13,  1.47it/s]
1|3273|Loss: 0.8649224042892456:  51%|█████     | 3272/6470 [40:33<36:13,  1.47it/s]
1|3273|Loss: 0.8649224042892456:  51%|█████     | 3273/6470 [40:33<38:59,  1.37it/s]
1|3274|Loss: 0.8542137145996094:  51%|█████     | 3273/6470 [40:33<38:59,  1.37it/s]
1|3274|Loss: 0.8542137145996094:  51%|█████     | 3274/6470 [40:34<41:48,  1.27it/s]
1|3275|Loss: 0.784670889377594:  51%|█████     | 3274/6470 [40:34<41:48,  1.27it/s] 
1|3275|Loss: 0.784670889377594:  51%|█████     | 3275/6470 [40:35<41:15,  1.29it/s]
1|3276|Loss: 0.7591564655303955:  51%|█████     | 3275/6470 [40:35<41:15,  1.29it/s]
1|3276|Loss: 0.7591564655303955:  51%|█████     | 3276/6470 [40:35<40:42,  1.31it/s]
1|3277|Loss: 0.8839614987373352:  51%|█████     | 3276/6470 [40:36<40:42,  1.31it/s]
1|3277|Loss: 0.8839614987373352:  51%|█████     | 3277/6470 [40:36<36:44,  1.45it/s]
1|3278|Loss: 0.9784522652626038:  51%|█████     | 3277/6470 [40:36<36:44,  1.45it/s]
1|3278|Loss: 0.9784522652626038:  51%|█████     | 3278/6470 [40:37<36:18,  1.47it/s]
1|3279|Loss: 0.961846649646759:  51%|█████     | 3278/6470 [40:37<36:18,  1.47it/s] 
1|3279|Loss: 0.961846649646759:  51%|█████     | 3279/6470 [40:37<35:36,  1.49it/s]
1|3280|Loss: 0.9624083638191223:  51%|█████     | 3279/6470 [40:38<35:36,  1.49it/s]
1|3280|Loss: 0.9624083638191223:  51%|█████     | 3280/6470 [40:38<37:29,  1.42it/s]
1|3281|Loss: 0.9807196855545044:  51%|█████     | 3280/6470 [40:38<37:29,  1.42it/s]
1|3281|Loss: 0.9807196855545044:  51%|█████     | 3281/6470 [40:39<36:45,  1.45it/s]
1|3282|Loss: 0.8248433470726013:  51%|█████     | 3281/6470 [40:39<36:45,  1.45it/s]
1|3282|Loss: 0.8248433470726013:  51%|█████     | 3282/6470 [40:39<36:44,  1.45it/s]
1|3283|Loss: 0.9171765446662903:  51%|█████     | 3282/6470 [40:40<36:44,  1.45it/s]
1|3283|Loss: 0.9171765446662903:  51%|█████     | 3283/6470 [40:40<38:54,  1.37it/s]
1|3284|Loss: 0.8115974068641663:  51%|█████     | 3283/6470 [40:40<38:54,  1.37it/s]
1|3284|Loss: 0.8115974068641663:  51%|█████     | 3284/6470 [40:41<36:16,  1.46it/s]
1|3285|Loss: 0.7404061555862427:  51%|█████     | 3284/6470 [40:41<36:16,  1.46it/s]
1|3285|Loss: 0.7404061555862427:  51%|█████     | 3285/6470 [40:41<35:03,  1.51it/s]
1|3286|Loss: 0.6909116506576538:  51%|█████     | 3285/6470 [40:42<35:03,  1.51it/s]
1|3286|Loss: 0.6909116506576538:  51%|█████     | 3286/6470 [40:42<37:27,  1.42it/s]
1|3287|Loss: 0.8114699125289917:  51%|█████     | 3286/6470 [40:43<37:27,  1.42it/s]
1|3287|Loss: 0.8114699125289917:  51%|█████     | 3287/6470 [40:43<37:55,  1.40it/s]
1|3288|Loss: 0.9833878874778748:  51%|█████     | 3287/6470 [40:43<37:55,  1.40it/s]
1|3288|Loss: 0.9833878874778748:  51%|█████     | 3288/6470 [40:43<33:23,  1.59it/s]
1|3289|Loss: 0.8540749549865723:  51%|█████     | 3288/6470 [40:44<33:23,  1.59it/s]
1|3289|Loss: 0.8540749549865723:  51%|█████     | 3289/6470 [40:44<32:49,  1.62it/s]
1|3290|Loss: 1.0921920537948608:  51%|█████     | 3289/6470 [40:44<32:49,  1.62it/s]
1|3290|Loss: 1.0921920537948608:  51%|█████     | 3290/6470 [40:45<33:18,  1.59it/s]
1|3291|Loss: 0.9224896430969238:  51%|█████     | 3290/6470 [40:45<33:18,  1.59it/s]
1|3291|Loss: 0.9224896430969238:  51%|█████     | 3291/6470 [40:45<32:49,  1.61it/s]
1|3292|Loss: 0.9208001494407654:  51%|█████     | 3291/6470 [40:45<32:49,  1.61it/s]
1|3292|Loss: 0.9208001494407654:  51%|█████     | 3292/6470 [40:46<32:52,  1.61it/s]
1|3293|Loss: 0.8640801906585693:  51%|█████     | 3292/6470 [40:46<32:52,  1.61it/s]
1|3293|Loss: 0.8640801906585693:  51%|█████     | 3293/6470 [40:47<35:51,  1.48it/s]
1|3294|Loss: 0.8266655802726746:  51%|█████     | 3293/6470 [40:47<35:51,  1.48it/s]
1|3294|Loss: 0.8266655802726746:  51%|█████     | 3294/6470 [40:47<40:05,  1.32it/s]
1|3295|Loss: 0.8623695373535156:  51%|█████     | 3294/6470 [40:48<40:05,  1.32it/s]
1|3295|Loss: 0.8623695373535156:  51%|█████     | 3295/6470 [40:48<38:09,  1.39it/s]
1|3296|Loss: 0.8723543286323547:  51%|█████     | 3295/6470 [40:48<38:09,  1.39it/s]
1|3296|Loss: 0.8723543286323547:  51%|█████     | 3296/6470 [40:49<35:25,  1.49it/s]
1|3297|Loss: 0.8631842136383057:  51%|█████     | 3296/6470 [40:49<35:25,  1.49it/s]
1|3297|Loss: 0.8631842136383057:  51%|█████     | 3297/6470 [40:49<36:07,  1.46it/s]
1|3298|Loss: 0.9987428188323975:  51%|█████     | 3297/6470 [40:50<36:07,  1.46it/s]
1|3298|Loss: 0.9987428188323975:  51%|█████     | 3298/6470 [40:50<36:38,  1.44it/s]
1|3299|Loss: 0.993419349193573:  51%|█████     | 3298/6470 [40:50<36:38,  1.44it/s] 
1|3299|Loss: 0.993419349193573:  51%|█████     | 3299/6470 [40:51<33:38,  1.57it/s]
1|3300|Loss: 0.8079676032066345:  51%|█████     | 3299/6470 [40:51<33:38,  1.57it/s]
1|3300|Loss: 0.8079676032066345:  51%|█████     | 3300/6470 [40:51<32:59,  1.60it/s]
1|3301|Loss: 0.9291841387748718:  51%|█████     | 3300/6470 [40:52<32:59,  1.60it/s]
1|3301|Loss: 0.9291841387748718:  51%|█████     | 3301/6470 [40:52<30:51,  1.71it/s]
1|3302|Loss: 0.8594557046890259:  51%|█████     | 3301/6470 [40:52<30:51,  1.71it/s]
1|3302|Loss: 0.8594557046890259:  51%|█████     | 3302/6470 [40:52<31:50,  1.66it/s]
1|3303|Loss: 1.0401039123535156:  51%|█████     | 3302/6470 [40:53<31:50,  1.66it/s]
1|3303|Loss: 1.0401039123535156:  51%|█████     | 3303/6470 [40:53<35:40,  1.48it/s]
1|3304|Loss: 0.841209352016449:  51%|█████     | 3303/6470 [40:54<35:40,  1.48it/s] 
1|3304|Loss: 0.841209352016449:  51%|█████     | 3304/6470 [40:54<35:18,  1.49it/s]
1|3305|Loss: 0.9794414639472961:  51%|█████     | 3304/6470 [40:54<35:18,  1.49it/s]
1|3305|Loss: 0.9794414639472961:  51%|█████     | 3305/6470 [40:55<37:22,  1.41it/s]
1|3306|Loss: 0.8325828313827515:  51%|█████     | 3305/6470 [40:55<37:22,  1.41it/s]
1|3306|Loss: 0.8325828313827515:  51%|█████     | 3306/6470 [40:55<38:10,  1.38it/s]
1|3307|Loss: 0.6672976016998291:  51%|█████     | 3306/6470 [40:56<38:10,  1.38it/s]
1|3307|Loss: 0.6672976016998291:  51%|█████     | 3307/6470 [40:56<35:36,  1.48it/s]
1|3308|Loss: 0.8749595880508423:  51%|█████     | 3307/6470 [40:56<35:36,  1.48it/s]
1|3308|Loss: 0.8749595880508423:  51%|█████     | 3308/6470 [40:57<36:32,  1.44it/s]
1|3309|Loss: 0.8067463040351868:  51%|█████     | 3308/6470 [40:57<36:32,  1.44it/s]
1|3309|Loss: 0.8067463040351868:  51%|█████     | 3309/6470 [40:57<37:27,  1.41it/s]
1|3310|Loss: 0.8356831073760986:  51%|█████     | 3309/6470 [40:58<37:27,  1.41it/s]
1|3310|Loss: 0.8356831073760986:  51%|█████     | 3310/6470 [40:58<39:02,  1.35it/s]
1|3311|Loss: 0.8203346133232117:  51%|█████     | 3310/6470 [40:59<39:02,  1.35it/s]
1|3311|Loss: 0.8203346133232117:  51%|█████     | 3311/6470 [40:59<41:30,  1.27it/s]
1|3312|Loss: 0.9004449248313904:  51%|█████     | 3311/6470 [41:00<41:30,  1.27it/s]
1|3312|Loss: 0.9004449248313904:  51%|█████     | 3312/6470 [41:00<41:11,  1.28it/s]
1|3313|Loss: 0.9091903567314148:  51%|█████     | 3312/6470 [41:00<41:11,  1.28it/s]
1|3313|Loss: 0.9091903567314148:  51%|█████     | 3313/6470 [41:01<40:04,  1.31it/s]
1|3314|Loss: 0.9241652488708496:  51%|█████     | 3313/6470 [41:01<40:04,  1.31it/s]
1|3314|Loss: 0.9241652488708496:  51%|█████     | 3314/6470 [41:01<37:34,  1.40it/s]
1|3315|Loss: 0.8069487810134888:  51%|█████     | 3314/6470 [41:02<37:34,  1.40it/s]
1|3315|Loss: 0.8069487810134888:  51%|█████     | 3315/6470 [41:02<37:34,  1.40it/s]
1|3316|Loss: 0.8167270421981812:  51%|█████     | 3315/6470 [41:02<37:34,  1.40it/s]
1|3316|Loss: 0.8167270421981812:  51%|█████▏    | 3316/6470 [41:03<35:19,  1.49it/s]
1|3317|Loss: 0.971784770488739:  51%|█████▏    | 3316/6470 [41:03<35:19,  1.49it/s] 
1|3317|Loss: 0.971784770488739:  51%|█████▏    | 3317/6470 [41:03<34:19,  1.53it/s]
1|3318|Loss: 0.9816338419914246:  51%|█████▏    | 3317/6470 [41:04<34:19,  1.53it/s]
1|3318|Loss: 0.9816338419914246:  51%|█████▏    | 3318/6470 [41:04<35:28,  1.48it/s]
1|3319|Loss: 0.980904221534729:  51%|█████▏    | 3318/6470 [41:04<35:28,  1.48it/s] 
1|3319|Loss: 0.980904221534729:  51%|█████▏    | 3319/6470 [41:04<33:01,  1.59it/s]
1|3320|Loss: 0.8211012482643127:  51%|█████▏    | 3319/6470 [41:05<33:01,  1.59it/s]
1|3320|Loss: 0.8211012482643127:  51%|█████▏    | 3320/6470 [41:05<34:51,  1.51it/s]
1|3321|Loss: 0.9930210709571838:  51%|█████▏    | 3320/6470 [41:06<34:51,  1.51it/s]
1|3321|Loss: 0.9930210709571838:  51%|█████▏    | 3321/6470 [41:06<35:52,  1.46it/s]
1|3322|Loss: 0.6707924008369446:  51%|█████▏    | 3321/6470 [41:06<35:52,  1.46it/s]
1|3322|Loss: 0.6707924008369446:  51%|█████▏    | 3322/6470 [41:07<36:46,  1.43it/s]
1|3323|Loss: 0.9206604361534119:  51%|█████▏    | 3322/6470 [41:07<36:46,  1.43it/s]
1|3323|Loss: 0.9206604361534119:  51%|█████▏    | 3323/6470 [41:07<36:12,  1.45it/s]
1|3324|Loss: 0.9138098955154419:  51%|█████▏    | 3323/6470 [41:08<36:12,  1.45it/s]
1|3324|Loss: 0.9138098955154419:  51%|█████▏    | 3324/6470 [41:08<38:12,  1.37it/s]
1|3325|Loss: 0.9201036691665649:  51%|█████▏    | 3324/6470 [41:09<38:12,  1.37it/s]
1|3325|Loss: 0.9201036691665649:  51%|█████▏    | 3325/6470 [41:09<39:43,  1.32it/s]
1|3326|Loss: 0.8881949186325073:  51%|█████▏    | 3325/6470 [41:09<39:43,  1.32it/s]
1|3326|Loss: 0.8881949186325073:  51%|█████▏    | 3326/6470 [41:10<40:19,  1.30it/s]
1|3327|Loss: 0.8549107313156128:  51%|█████▏    | 3326/6470 [41:10<40:19,  1.30it/s]
1|3327|Loss: 0.8549107313156128:  51%|█████▏    | 3327/6470 [41:11<41:32,  1.26it/s]
1|3328|Loss: 0.7990540862083435:  51%|█████▏    | 3327/6470 [41:11<41:32,  1.26it/s]
1|3328|Loss: 0.7990540862083435:  51%|█████▏    | 3328/6470 [41:11<39:39,  1.32it/s]
1|3329|Loss: 0.8515195846557617:  51%|█████▏    | 3328/6470 [41:12<39:39,  1.32it/s]
1|3329|Loss: 0.8515195846557617:  51%|█████▏    | 3329/6470 [41:12<36:57,  1.42it/s]
1|3330|Loss: 0.7517950534820557:  51%|█████▏    | 3329/6470 [41:12<36:57,  1.42it/s]
1|3330|Loss: 0.7517950534820557:  51%|█████▏    | 3330/6470 [41:13<37:07,  1.41it/s]
1|3331|Loss: 0.9043976068496704:  51%|█████▏    | 3330/6470 [41:13<37:07,  1.41it/s]
1|3331|Loss: 0.9043976068496704:  51%|█████▏    | 3331/6470 [41:13<37:13,  1.41it/s]
1|3332|Loss: 1.059046983718872:  51%|█████▏    | 3331/6470 [41:14<37:13,  1.41it/s] 
1|3332|Loss: 1.059046983718872:  51%|█████▏    | 3332/6470 [41:14<37:04,  1.41it/s]
1|3333|Loss: 0.8317908048629761:  51%|█████▏    | 3332/6470 [41:14<37:04,  1.41it/s]
1|3333|Loss: 0.8317908048629761:  52%|█████▏    | 3333/6470 [41:15<37:29,  1.39it/s]
1|3334|Loss: 0.74442058801651:  52%|█████▏    | 3333/6470 [41:15<37:29,  1.39it/s]  
1|3334|Loss: 0.74442058801651:  52%|█████▏    | 3334/6470 [41:15<36:23,  1.44it/s]
1|3335|Loss: 0.9200400114059448:  52%|█████▏    | 3334/6470 [41:16<36:23,  1.44it/s]
1|3335|Loss: 0.9200400114059448:  52%|█████▏    | 3335/6470 [41:16<37:55,  1.38it/s]
1|3336|Loss: 0.8006764054298401:  52%|█████▏    | 3335/6470 [41:17<37:55,  1.38it/s]
1|3336|Loss: 0.8006764054298401:  52%|█████▏    | 3336/6470 [41:17<40:00,  1.31it/s]
1|3337|Loss: 0.9480779767036438:  52%|█████▏    | 3336/6470 [41:17<40:00,  1.31it/s]
1|3337|Loss: 0.9480779767036438:  52%|█████▏    | 3337/6470 [41:18<38:30,  1.36it/s]
1|3338|Loss: 0.8693094253540039:  52%|█████▏    | 3337/6470 [41:18<38:30,  1.36it/s]
1|3338|Loss: 0.8693094253540039:  52%|█████▏    | 3338/6470 [41:18<38:10,  1.37it/s]
1|3339|Loss: 0.7684293985366821:  52%|█████▏    | 3338/6470 [41:19<38:10,  1.37it/s]
1|3339|Loss: 0.7684293985366821:  52%|█████▏    | 3339/6470 [41:19<36:05,  1.45it/s]
1|3340|Loss: 0.6912583708763123:  52%|█████▏    | 3339/6470 [41:19<36:05,  1.45it/s]
1|3340|Loss: 0.6912583708763123:  52%|█████▏    | 3340/6470 [41:20<33:35,  1.55it/s]
1|3341|Loss: 0.8903872966766357:  52%|█████▏    | 3340/6470 [41:20<33:35,  1.55it/s]
1|3341|Loss: 0.8903872966766357:  52%|█████▏    | 3341/6470 [41:20<34:12,  1.52it/s]
1|3342|Loss: 0.834947943687439:  52%|█████▏    | 3341/6470 [41:21<34:12,  1.52it/s] 
1|3342|Loss: 0.834947943687439:  52%|█████▏    | 3342/6470 [41:21<36:10,  1.44it/s]
1|3343|Loss: 0.8158000111579895:  52%|█████▏    | 3342/6470 [41:21<36:10,  1.44it/s]
1|3343|Loss: 0.8158000111579895:  52%|█████▏    | 3343/6470 [41:22<37:00,  1.41it/s]
1|3344|Loss: 0.7761425375938416:  52%|█████▏    | 3343/6470 [41:22<37:00,  1.41it/s]
1|3344|Loss: 0.7761425375938416:  52%|█████▏    | 3344/6470 [41:22<35:24,  1.47it/s]
1|3345|Loss: 0.8849818110466003:  52%|█████▏    | 3344/6470 [41:23<35:24,  1.47it/s]
1|3345|Loss: 0.8849818110466003:  52%|█████▏    | 3345/6470 [41:23<35:26,  1.47it/s]
1|3346|Loss: 0.8023147583007812:  52%|█████▏    | 3345/6470 [41:23<35:26,  1.47it/s]
1|3346|Loss: 0.8023147583007812:  52%|█████▏    | 3346/6470 [41:24<33:49,  1.54it/s]
1|3347|Loss: 0.942069947719574:  52%|█████▏    | 3346/6470 [41:24<33:49,  1.54it/s] 
1|3347|Loss: 0.942069947719574:  52%|█████▏    | 3347/6470 [41:24<36:25,  1.43it/s]
1|3348|Loss: 0.873498260974884:  52%|█████▏    | 3347/6470 [41:25<36:25,  1.43it/s]
1|3348|Loss: 0.873498260974884:  52%|█████▏    | 3348/6470 [41:25<37:08,  1.40it/s]
1|3349|Loss: 0.9295403957366943:  52%|█████▏    | 3348/6470 [41:26<37:08,  1.40it/s]
1|3349|Loss: 0.9295403957366943:  52%|█████▏    | 3349/6470 [41:26<35:56,  1.45it/s]
1|3350|Loss: 0.8355530500411987:  52%|█████▏    | 3349/6470 [41:26<35:56,  1.45it/s]
1|3350|Loss: 0.8355530500411987:  52%|█████▏    | 3350/6470 [41:27<37:03,  1.40it/s]
1|3351|Loss: 0.8821820020675659:  52%|█████▏    | 3350/6470 [41:27<37:03,  1.40it/s]
1|3351|Loss: 0.8821820020675659:  52%|█████▏    | 3351/6470 [41:27<36:27,  1.43it/s]
1|3352|Loss: 0.9630721211433411:  52%|█████▏    | 3351/6470 [41:28<36:27,  1.43it/s]
1|3352|Loss: 0.9630721211433411:  52%|█████▏    | 3352/6470 [41:28<31:49,  1.63it/s]
1|3353|Loss: 0.7597110867500305:  52%|█████▏    | 3352/6470 [41:28<31:49,  1.63it/s]
1|3353|Loss: 0.7597110867500305:  52%|█████▏    | 3353/6470 [41:28<33:36,  1.55it/s]
1|3354|Loss: 0.8040041923522949:  52%|█████▏    | 3353/6470 [41:29<33:36,  1.55it/s]
1|3354|Loss: 0.8040041923522949:  52%|█████▏    | 3354/6470 [41:29<35:26,  1.47it/s]
1|3355|Loss: 0.9117314219474792:  52%|█████▏    | 3354/6470 [41:30<35:26,  1.47it/s]
1|3355|Loss: 0.9117314219474792:  52%|█████▏    | 3355/6470 [41:30<36:56,  1.41it/s]
1|3356|Loss: 0.9675828814506531:  52%|█████▏    | 3355/6470 [41:30<36:56,  1.41it/s]
1|3356|Loss: 0.9675828814506531:  52%|█████▏    | 3356/6470 [41:31<37:10,  1.40it/s]
1|3357|Loss: 1.0047974586486816:  52%|█████▏    | 3356/6470 [41:31<37:10,  1.40it/s]
1|3357|Loss: 1.0047974586486816:  52%|█████▏    | 3357/6470 [41:31<39:03,  1.33it/s]
1|3358|Loss: 0.8732801675796509:  52%|█████▏    | 3357/6470 [41:32<39:03,  1.33it/s]
1|3358|Loss: 0.8732801675796509:  52%|█████▏    | 3358/6470 [41:32<38:53,  1.33it/s]
1|3359|Loss: 0.9011684656143188:  52%|█████▏    | 3358/6470 [41:33<38:53,  1.33it/s]
1|3359|Loss: 0.9011684656143188:  52%|█████▏    | 3359/6470 [41:33<34:08,  1.52it/s]
1|3360|Loss: 0.9462523460388184:  52%|█████▏    | 3359/6470 [41:33<34:08,  1.52it/s]
1|3360|Loss: 0.9462523460388184:  52%|█████▏    | 3360/6470 [41:33<33:56,  1.53it/s]
1|3361|Loss: 0.8328269720077515:  52%|█████▏    | 3360/6470 [41:34<33:56,  1.53it/s]
1|3361|Loss: 0.8328269720077515:  52%|█████▏    | 3361/6470 [41:34<32:13,  1.61it/s]
1|3362|Loss: 0.7030558586120605:  52%|█████▏    | 3361/6470 [41:34<32:13,  1.61it/s]
1|3362|Loss: 0.7030558586120605:  52%|█████▏    | 3362/6470 [41:35<32:51,  1.58it/s]
1|3363|Loss: 0.7411874532699585:  52%|█████▏    | 3362/6470 [41:35<32:51,  1.58it/s]
1|3363|Loss: 0.7411874532699585:  52%|█████▏    | 3363/6470 [41:35<33:51,  1.53it/s]
1|3364|Loss: 0.9357542991638184:  52%|█████▏    | 3363/6470 [41:36<33:51,  1.53it/s]
1|3364|Loss: 0.9357542991638184:  52%|█████▏    | 3364/6470 [41:36<35:00,  1.48it/s]
1|3365|Loss: 0.7047805190086365:  52%|█████▏    | 3364/6470 [41:36<35:00,  1.48it/s]
1|3365|Loss: 0.7047805190086365:  52%|█████▏    | 3365/6470 [41:37<35:33,  1.46it/s]
1|3366|Loss: 0.7873928546905518:  52%|█████▏    | 3365/6470 [41:37<35:33,  1.46it/s]
1|3366|Loss: 0.7873928546905518:  52%|█████▏    | 3366/6470 [41:37<35:41,  1.45it/s]
1|3367|Loss: 1.048128604888916:  52%|█████▏    | 3366/6470 [41:38<35:41,  1.45it/s] 
1|3367|Loss: 1.048128604888916:  52%|█████▏    | 3367/6470 [41:38<33:26,  1.55it/s]
1|3368|Loss: 0.9609824419021606:  52%|█████▏    | 3367/6470 [41:38<33:26,  1.55it/s]
1|3368|Loss: 0.9609824419021606:  52%|█████▏    | 3368/6470 [41:39<32:36,  1.59it/s]
1|3369|Loss: 0.8887341618537903:  52%|█████▏    | 3368/6470 [41:39<32:36,  1.59it/s]
1|3369|Loss: 0.8887341618537903:  52%|█████▏    | 3369/6470 [41:39<34:37,  1.49it/s]
1|3370|Loss: 1.0458950996398926:  52%|█████▏    | 3369/6470 [41:40<34:37,  1.49it/s]
1|3370|Loss: 1.0458950996398926:  52%|█████▏    | 3370/6470 [41:40<34:24,  1.50it/s]
1|3371|Loss: 0.8023089170455933:  52%|█████▏    | 3370/6470 [41:40<34:24,  1.50it/s]
1|3371|Loss: 0.8023089170455933:  52%|█████▏    | 3371/6470 [41:41<35:01,  1.47it/s]
1|3372|Loss: 0.8513322472572327:  52%|█████▏    | 3371/6470 [41:41<35:01,  1.47it/s]
1|3372|Loss: 0.8513322472572327:  52%|█████▏    | 3372/6470 [41:41<35:58,  1.43it/s]
1|3373|Loss: 0.8986790180206299:  52%|█████▏    | 3372/6470 [41:42<35:58,  1.43it/s]
1|3373|Loss: 0.8986790180206299:  52%|█████▏    | 3373/6470 [41:42<36:12,  1.43it/s]
1|3374|Loss: 0.9201943278312683:  52%|█████▏    | 3373/6470 [41:43<36:12,  1.43it/s]
1|3374|Loss: 0.9201943278312683:  52%|█████▏    | 3374/6470 [41:43<37:24,  1.38it/s]
1|3375|Loss: 0.9168022871017456:  52%|█████▏    | 3374/6470 [41:43<37:24,  1.38it/s]
1|3375|Loss: 0.9168022871017456:  52%|█████▏    | 3375/6470 [41:44<37:20,  1.38it/s]
1|3376|Loss: 0.906635582447052:  52%|█████▏    | 3375/6470 [41:44<37:20,  1.38it/s] 
1|3376|Loss: 0.906635582447052:  52%|█████▏    | 3376/6470 [41:44<35:59,  1.43it/s]
1|3377|Loss: 0.9409619569778442:  52%|█████▏    | 3376/6470 [41:45<35:59,  1.43it/s]
1|3377|Loss: 0.9409619569778442:  52%|█████▏    | 3377/6470 [41:45<33:24,  1.54it/s]
1|3378|Loss: 1.0494332313537598:  52%|█████▏    | 3377/6470 [41:45<33:24,  1.54it/s]
1|3378|Loss: 1.0494332313537598:  52%|█████▏    | 3378/6470 [41:45<34:42,  1.48it/s]
1|3379|Loss: 0.9338975548744202:  52%|█████▏    | 3378/6470 [41:46<34:42,  1.48it/s]
1|3379|Loss: 0.9338975548744202:  52%|█████▏    | 3379/6470 [41:46<36:01,  1.43it/s]
1|3380|Loss: 0.8144067525863647:  52%|█████▏    | 3379/6470 [41:47<36:01,  1.43it/s]
1|3380|Loss: 0.8144067525863647:  52%|█████▏    | 3380/6470 [41:47<36:54,  1.40it/s]
1|3381|Loss: 0.8925243616104126:  52%|█████▏    | 3380/6470 [41:47<36:54,  1.40it/s]
1|3381|Loss: 0.8925243616104126:  52%|█████▏    | 3381/6470 [41:48<36:49,  1.40it/s]
1|3382|Loss: 0.8897577524185181:  52%|█████▏    | 3381/6470 [41:48<36:49,  1.40it/s]
1|3382|Loss: 0.8897577524185181:  52%|█████▏    | 3382/6470 [41:48<37:00,  1.39it/s]
1|3383|Loss: 0.7217058539390564:  52%|█████▏    | 3382/6470 [41:49<37:00,  1.39it/s]
1|3383|Loss: 0.7217058539390564:  52%|█████▏    | 3383/6470 [41:49<37:23,  1.38it/s]
1|3384|Loss: 0.903876543045044:  52%|█████▏    | 3383/6470 [41:50<37:23,  1.38it/s] 
1|3384|Loss: 0.903876543045044:  52%|█████▏    | 3384/6470 [41:50<38:30,  1.34it/s]
1|3385|Loss: 0.8604442477226257:  52%|█████▏    | 3384/6470 [41:50<38:30,  1.34it/s]
1|3385|Loss: 0.8604442477226257:  52%|█████▏    | 3385/6470 [41:51<39:46,  1.29it/s]
1|3386|Loss: 0.824748158454895:  52%|█████▏    | 3385/6470 [41:51<39:46,  1.29it/s] 
1|3386|Loss: 0.824748158454895:  52%|█████▏    | 3386/6470 [41:52<40:41,  1.26it/s]
1|3387|Loss: 0.9624724388122559:  52%|█████▏    | 3386/6470 [41:52<40:41,  1.26it/s]
1|3387|Loss: 0.9624724388122559:  52%|█████▏    | 3387/6470 [41:52<38:14,  1.34it/s]
1|3388|Loss: 0.8710059523582458:  52%|█████▏    | 3387/6470 [41:53<38:14,  1.34it/s]
1|3388|Loss: 0.8710059523582458:  52%|█████▏    | 3388/6470 [41:53<37:12,  1.38it/s]
1|3389|Loss: 0.7381716370582581:  52%|█████▏    | 3388/6470 [41:53<37:12,  1.38it/s]
1|3389|Loss: 0.7381716370582581:  52%|█████▏    | 3389/6470 [41:54<35:50,  1.43it/s]
1|3390|Loss: 1.0256706476211548:  52%|█████▏    | 3389/6470 [41:54<35:50,  1.43it/s]
1|3390|Loss: 1.0256706476211548:  52%|█████▏    | 3390/6470 [41:54<36:44,  1.40it/s]
1|3391|Loss: 0.972629964351654:  52%|█████▏    | 3390/6470 [41:55<36:44,  1.40it/s] 
1|3391|Loss: 0.972629964351654:  52%|█████▏    | 3391/6470 [41:55<35:54,  1.43it/s]
1|3392|Loss: 0.8488600850105286:  52%|█████▏    | 3391/6470 [41:55<35:54,  1.43it/s]
1|3392|Loss: 0.8488600850105286:  52%|█████▏    | 3392/6470 [41:56<33:48,  1.52it/s]
1|3393|Loss: 0.8393427133560181:  52%|█████▏    | 3392/6470 [41:56<33:48,  1.52it/s]
1|3393|Loss: 0.8393427133560181:  52%|█████▏    | 3393/6470 [41:56<34:24,  1.49it/s]
1|3394|Loss: 1.0303633213043213:  52%|█████▏    | 3393/6470 [41:57<34:24,  1.49it/s]
1|3394|Loss: 1.0303633213043213:  52%|█████▏    | 3394/6470 [41:57<36:04,  1.42it/s]
1|3395|Loss: 0.9164416193962097:  52%|█████▏    | 3394/6470 [41:57<36:04,  1.42it/s]
1|3395|Loss: 0.9164416193962097:  52%|█████▏    | 3395/6470 [41:58<34:34,  1.48it/s]
1|3396|Loss: 0.8795076608657837:  52%|█████▏    | 3395/6470 [41:58<34:34,  1.48it/s]
1|3396|Loss: 0.8795076608657837:  52%|█████▏    | 3396/6470 [41:58<32:21,  1.58it/s]
1|3397|Loss: 0.7480856776237488:  52%|█████▏    | 3396/6470 [41:59<32:21,  1.58it/s]
1|3397|Loss: 0.7480856776237488:  53%|█████▎    | 3397/6470 [41:59<36:19,  1.41it/s]
1|3398|Loss: 0.814732015132904:  53%|█████▎    | 3397/6470 [42:00<36:19,  1.41it/s] 
1|3398|Loss: 0.814732015132904:  53%|█████▎    | 3398/6470 [42:00<36:01,  1.42it/s]
1|3399|Loss: 0.9641385674476624:  53%|█████▎    | 3398/6470 [42:00<36:01,  1.42it/s]
1|3399|Loss: 0.9641385674476624:  53%|█████▎    | 3399/6470 [42:01<37:40,  1.36it/s]
1|3400|Loss: 0.9672442078590393:  53%|█████▎    | 3399/6470 [42:01<37:40,  1.36it/s]
1|3400|Loss: 0.9672442078590393:  53%|█████▎    | 3400/6470 [42:01<37:39,  1.36it/s]
1|3401|Loss: 0.9289655089378357:  53%|█████▎    | 3400/6470 [42:02<37:39,  1.36it/s]
1|3401|Loss: 0.9289655089378357:  53%|█████▎    | 3401/6470 [42:02<36:03,  1.42it/s]
1|3402|Loss: 0.8996701240539551:  53%|█████▎    | 3401/6470 [42:02<36:03,  1.42it/s]
1|3402|Loss: 0.8996701240539551:  53%|█████▎    | 3402/6470 [42:03<38:20,  1.33it/s]
1|3403|Loss: 0.7583972215652466:  53%|█████▎    | 3402/6470 [42:03<38:20,  1.33it/s]
1|3403|Loss: 0.7583972215652466:  53%|█████▎    | 3403/6470 [42:03<36:54,  1.38it/s]
1|3404|Loss: 1.011590600013733:  53%|█████▎    | 3403/6470 [42:04<36:54,  1.38it/s] 
1|3404|Loss: 1.011590600013733:  53%|█████▎    | 3404/6470 [42:04<35:27,  1.44it/s]
1|3405|Loss: 0.8772821426391602:  53%|█████▎    | 3404/6470 [42:04<35:27,  1.44it/s]
1|3405|Loss: 0.8772821426391602:  53%|█████▎    | 3405/6470 [42:05<34:39,  1.47it/s]
1|3406|Loss: 0.88164883852005:  53%|█████▎    | 3405/6470 [42:05<34:39,  1.47it/s]  
1|3406|Loss: 0.88164883852005:  53%|█████▎    | 3406/6470 [42:06<36:41,  1.39it/s]
1|3407|Loss: 0.9483801126480103:  53%|█████▎    | 3406/6470 [42:06<36:41,  1.39it/s]
1|3407|Loss: 0.9483801126480103:  53%|█████▎    | 3407/6470 [42:06<36:55,  1.38it/s]
1|3408|Loss: 0.7719986438751221:  53%|█████▎    | 3407/6470 [42:07<36:55,  1.38it/s]
1|3408|Loss: 0.7719986438751221:  53%|█████▎    | 3408/6470 [42:07<37:41,  1.35it/s]
1|3409|Loss: 0.8324049115180969:  53%|█████▎    | 3408/6470 [42:07<37:41,  1.35it/s]
1|3409|Loss: 0.8324049115180969:  53%|█████▎    | 3409/6470 [42:08<35:50,  1.42it/s]
1|3410|Loss: 0.8464131355285645:  53%|█████▎    | 3409/6470 [42:08<35:50,  1.42it/s]
1|3410|Loss: 0.8464131355285645:  53%|█████▎    | 3410/6470 [42:08<34:03,  1.50it/s]
1|3411|Loss: 0.9417422413825989:  53%|█████▎    | 3410/6470 [42:09<34:03,  1.50it/s]
1|3411|Loss: 0.9417422413825989:  53%|█████▎    | 3411/6470 [42:09<35:45,  1.43it/s]
1|3412|Loss: 0.9263283610343933:  53%|█████▎    | 3411/6470 [42:09<35:45,  1.43it/s]
1|3412|Loss: 0.9263283610343933:  53%|█████▎    | 3412/6470 [42:10<36:11,  1.41it/s]
1|3413|Loss: 0.8815531730651855:  53%|█████▎    | 3412/6470 [42:10<36:11,  1.41it/s]
1|3413|Loss: 0.8815531730651855:  53%|█████▎    | 3413/6470 [42:10<33:31,  1.52it/s]
1|3414|Loss: 0.9221853613853455:  53%|█████▎    | 3413/6470 [42:11<33:31,  1.52it/s]
1|3414|Loss: 0.9221853613853455:  53%|█████▎    | 3414/6470 [42:11<34:37,  1.47it/s]
1|3415|Loss: 0.8518041968345642:  53%|█████▎    | 3414/6470 [42:12<34:37,  1.47it/s]
1|3415|Loss: 0.8518041968345642:  53%|█████▎    | 3415/6470 [42:12<36:55,  1.38it/s]
1|3416|Loss: 1.026616096496582:  53%|█████▎    | 3415/6470 [42:12<36:55,  1.38it/s] 
1|3416|Loss: 1.026616096496582:  53%|█████▎    | 3416/6470 [42:13<38:02,  1.34it/s]
1|3417|Loss: 0.755081295967102:  53%|█████▎    | 3416/6470 [42:13<38:02,  1.34it/s]
1|3417|Loss: 0.755081295967102:  53%|█████▎    | 3417/6470 [42:14<39:53,  1.28it/s]
1|3418|Loss: 0.7884703874588013:  53%|█████▎    | 3417/6470 [42:14<39:53,  1.28it/s]
1|3418|Loss: 0.7884703874588013:  53%|█████▎    | 3418/6470 [42:14<36:38,  1.39it/s]
1|3419|Loss: 0.9561561346054077:  53%|█████▎    | 3418/6470 [42:14<36:38,  1.39it/s]
1|3419|Loss: 0.9561561346054077:  53%|█████▎    | 3419/6470 [42:15<34:53,  1.46it/s]
1|3420|Loss: 0.8053826689720154:  53%|█████▎    | 3419/6470 [42:15<34:53,  1.46it/s]
1|3420|Loss: 0.8053826689720154:  53%|█████▎    | 3420/6470 [42:16<37:29,  1.36it/s]
1|3421|Loss: 0.9187425971031189:  53%|█████▎    | 3420/6470 [42:16<37:29,  1.36it/s]
1|3421|Loss: 0.9187425971031189:  53%|█████▎    | 3421/6470 [42:16<36:45,  1.38it/s]
1|3422|Loss: 0.9455896615982056:  53%|█████▎    | 3421/6470 [42:17<36:45,  1.38it/s]
1|3422|Loss: 0.9455896615982056:  53%|█████▎    | 3422/6470 [42:17<38:18,  1.33it/s]
1|3423|Loss: 0.9347472786903381:  53%|█████▎    | 3422/6470 [42:17<38:18,  1.33it/s]
1|3423|Loss: 0.9347472786903381:  53%|█████▎    | 3423/6470 [42:18<35:05,  1.45it/s]
1|3424|Loss: 0.856609046459198:  53%|█████▎    | 3423/6470 [42:18<35:05,  1.45it/s] 
1|3424|Loss: 0.856609046459198:  53%|█████▎    | 3424/6470 [42:18<32:59,  1.54it/s]
1|3425|Loss: 0.7896429896354675:  53%|█████▎    | 3424/6470 [42:19<32:59,  1.54it/s]
1|3425|Loss: 0.7896429896354675:  53%|█████▎    | 3425/6470 [42:19<33:48,  1.50it/s]
1|3426|Loss: 0.7709869146347046:  53%|█████▎    | 3425/6470 [42:19<33:48,  1.50it/s]
1|3426|Loss: 0.7709869146347046:  53%|█████▎    | 3426/6470 [42:19<32:38,  1.55it/s]
1|3427|Loss: 0.7938992977142334:  53%|█████▎    | 3426/6470 [42:20<32:38,  1.55it/s]
1|3427|Loss: 0.7938992977142334:  53%|█████▎    | 3427/6470 [42:20<33:27,  1.52it/s]
1|3428|Loss: 0.9624298810958862:  53%|█████▎    | 3427/6470 [42:21<33:27,  1.52it/s]
1|3428|Loss: 0.9624298810958862:  53%|█████▎    | 3428/6470 [42:21<34:16,  1.48it/s]
1|3429|Loss: 0.9227774143218994:  53%|█████▎    | 3428/6470 [42:21<34:16,  1.48it/s]
1|3429|Loss: 0.9227774143218994:  53%|█████▎    | 3429/6470 [42:22<34:46,  1.46it/s]
1|3430|Loss: 0.8755140900611877:  53%|█████▎    | 3429/6470 [42:22<34:46,  1.46it/s]
1|3430|Loss: 0.8755140900611877:  53%|█████▎    | 3430/6470 [42:22<36:02,  1.41it/s]
1|3431|Loss: 0.9608813524246216:  53%|█████▎    | 3430/6470 [42:23<36:02,  1.41it/s]
1|3431|Loss: 0.9608813524246216:  53%|█████▎    | 3431/6470 [42:23<36:35,  1.38it/s]
1|3432|Loss: 0.9864413142204285:  53%|█████▎    | 3431/6470 [42:23<36:35,  1.38it/s]
1|3432|Loss: 0.9864413142204285:  53%|█████▎    | 3432/6470 [42:24<33:17,  1.52it/s]
1|3433|Loss: 0.8425820469856262:  53%|█████▎    | 3432/6470 [42:24<33:17,  1.52it/s]
1|3433|Loss: 0.8425820469856262:  53%|█████▎    | 3433/6470 [42:24<33:07,  1.53it/s]
1|3434|Loss: 0.9226922392845154:  53%|█████▎    | 3433/6470 [42:25<33:07,  1.53it/s]
1|3434|Loss: 0.9226922392845154:  53%|█████▎    | 3434/6470 [42:25<32:50,  1.54it/s]
1|3435|Loss: 0.8690319061279297:  53%|█████▎    | 3434/6470 [42:25<32:50,  1.54it/s]
1|3435|Loss: 0.8690319061279297:  53%|█████▎    | 3435/6470 [42:26<33:48,  1.50it/s]
1|3436|Loss: 0.8798691034317017:  53%|█████▎    | 3435/6470 [42:26<33:48,  1.50it/s]
1|3436|Loss: 0.8798691034317017:  53%|█████▎    | 3436/6470 [42:26<33:46,  1.50it/s]
1|3437|Loss: 0.8744490742683411:  53%|█████▎    | 3436/6470 [42:27<33:46,  1.50it/s]
1|3437|Loss: 0.8744490742683411:  53%|█████▎    | 3437/6470 [42:27<33:22,  1.51it/s]
1|3438|Loss: 0.8928812742233276:  53%|█████▎    | 3437/6470 [42:27<33:22,  1.51it/s]
1|3438|Loss: 0.8928812742233276:  53%|█████▎    | 3438/6470 [42:27<30:50,  1.64it/s]
1|3439|Loss: 0.8800505995750427:  53%|█████▎    | 3438/6470 [42:28<30:50,  1.64it/s]
1|3439|Loss: 0.8800505995750427:  53%|█████▎    | 3439/6470 [42:28<30:31,  1.66it/s]
1|3440|Loss: 0.7501687407493591:  53%|█████▎    | 3439/6470 [42:28<30:31,  1.66it/s]
1|3440|Loss: 0.7501687407493591:  53%|█████▎    | 3440/6470 [42:29<33:16,  1.52it/s]
1|3441|Loss: 0.9200381636619568:  53%|█████▎    | 3440/6470 [42:29<33:16,  1.52it/s]
1|3441|Loss: 0.9200381636619568:  53%|█████▎    | 3441/6470 [42:29<33:14,  1.52it/s]
1|3442|Loss: 0.8288718461990356:  53%|█████▎    | 3441/6470 [42:30<33:14,  1.52it/s]
1|3442|Loss: 0.8288718461990356:  53%|█████▎    | 3442/6470 [42:30<33:34,  1.50it/s]
1|3443|Loss: 0.767301619052887:  53%|█████▎    | 3442/6470 [42:30<33:34,  1.50it/s] 
1|3443|Loss: 0.767301619052887:  53%|█████▎    | 3443/6470 [42:31<31:06,  1.62it/s]
1|3444|Loss: 0.7988037467002869:  53%|█████▎    | 3443/6470 [42:31<31:06,  1.62it/s]
1|3444|Loss: 0.7988037467002869:  53%|█████▎    | 3444/6470 [42:31<34:09,  1.48it/s]
1|3445|Loss: 0.8792692422866821:  53%|█████▎    | 3444/6470 [42:32<34:09,  1.48it/s]
1|3445|Loss: 0.8792692422866821:  53%|█████▎    | 3445/6470 [42:32<35:49,  1.41it/s]
1|3446|Loss: 0.9175115823745728:  53%|█████▎    | 3445/6470 [42:33<35:49,  1.41it/s]
1|3446|Loss: 0.9175115823745728:  53%|█████▎    | 3446/6470 [42:33<34:33,  1.46it/s]
1|3447|Loss: 0.9393665194511414:  53%|█████▎    | 3446/6470 [42:33<34:33,  1.46it/s]
1|3447|Loss: 0.9393665194511414:  53%|█████▎    | 3447/6470 [42:34<33:19,  1.51it/s]
1|3448|Loss: 0.8604122996330261:  53%|█████▎    | 3447/6470 [42:34<33:19,  1.51it/s]
1|3448|Loss: 0.8604122996330261:  53%|█████▎    | 3448/6470 [42:34<32:11,  1.56it/s]
1|3449|Loss: 0.8375179767608643:  53%|█████▎    | 3448/6470 [42:34<32:11,  1.56it/s]
1|3449|Loss: 0.8375179767608643:  53%|█████▎    | 3449/6470 [42:35<33:14,  1.51it/s]
1|3450|Loss: 0.9712283611297607:  53%|█████▎    | 3449/6470 [42:35<33:14,  1.51it/s]
1|3450|Loss: 0.9712283611297607:  53%|█████▎    | 3450/6470 [42:35<32:55,  1.53it/s]
1|3451|Loss: 1.0716869831085205:  53%|█████▎    | 3450/6470 [42:36<32:55,  1.53it/s]
1|3451|Loss: 1.0716869831085205:  53%|█████▎    | 3451/6470 [42:36<32:31,  1.55it/s]
1|3452|Loss: 0.9480223059654236:  53%|█████▎    | 3451/6470 [42:36<32:31,  1.55it/s]
1|3452|Loss: 0.9480223059654236:  53%|█████▎    | 3452/6470 [42:37<34:56,  1.44it/s]
1|3453|Loss: 0.8301595449447632:  53%|█████▎    | 3452/6470 [42:37<34:56,  1.44it/s]
1|3453|Loss: 0.8301595449447632:  53%|█████▎    | 3453/6470 [42:38<34:02,  1.48it/s]
1|3454|Loss: 0.8393855094909668:  53%|█████▎    | 3453/6470 [42:38<34:02,  1.48it/s]
1|3454|Loss: 0.8393855094909668:  53%|█████▎    | 3454/6470 [42:38<33:15,  1.51it/s]
1|3455|Loss: 0.7899149656295776:  53%|█████▎    | 3454/6470 [42:39<33:15,  1.51it/s]
1|3455|Loss: 0.7899149656295776:  53%|█████▎    | 3455/6470 [42:39<34:08,  1.47it/s]
1|3456|Loss: 0.8773525357246399:  53%|█████▎    | 3455/6470 [42:39<34:08,  1.47it/s]
1|3456|Loss: 0.8773525357246399:  53%|█████▎    | 3456/6470 [42:40<36:01,  1.39it/s]
1|3457|Loss: 0.7328630685806274:  53%|█████▎    | 3456/6470 [42:40<36:01,  1.39it/s]
1|3457|Loss: 0.7328630685806274:  53%|█████▎    | 3457/6470 [42:40<34:47,  1.44it/s]
1|3458|Loss: 0.8504423499107361:  53%|█████▎    | 3457/6470 [42:41<34:47,  1.44it/s]
1|3458|Loss: 0.8504423499107361:  53%|█████▎    | 3458/6470 [42:41<33:47,  1.49it/s]
1|3459|Loss: 0.9821746945381165:  53%|█████▎    | 3458/6470 [42:41<33:47,  1.49it/s]
1|3459|Loss: 0.9821746945381165:  53%|█████▎    | 3459/6470 [42:42<34:49,  1.44it/s]
1|3460|Loss: 0.7369720339775085:  53%|█████▎    | 3459/6470 [42:42<34:49,  1.44it/s]
1|3460|Loss: 0.7369720339775085:  53%|█████▎    | 3460/6470 [42:43<37:36,  1.33it/s]
1|3461|Loss: 0.9420551061630249:  53%|█████▎    | 3460/6470 [42:43<37:36,  1.33it/s]
1|3461|Loss: 0.9420551061630249:  53%|█████▎    | 3461/6470 [42:43<39:32,  1.27it/s]
1|3462|Loss: 0.9374114871025085:  53%|█████▎    | 3461/6470 [42:44<39:32,  1.27it/s]
1|3462|Loss: 0.9374114871025085:  54%|█████▎    | 3462/6470 [42:44<39:14,  1.28it/s]
1|3463|Loss: 0.8504136800765991:  54%|█████▎    | 3462/6470 [42:45<39:14,  1.28it/s]
1|3463|Loss: 0.8504136800765991:  54%|█████▎    | 3463/6470 [42:45<37:18,  1.34it/s]
1|3464|Loss: 0.8610395193099976:  54%|█████▎    | 3463/6470 [42:45<37:18,  1.34it/s]
1|3464|Loss: 0.8610395193099976:  54%|█████▎    | 3464/6470 [42:45<34:11,  1.47it/s]
1|3465|Loss: 0.8996338248252869:  54%|█████▎    | 3464/6470 [42:46<34:11,  1.47it/s]
1|3465|Loss: 0.8996338248252869:  54%|█████▎    | 3465/6470 [42:46<32:49,  1.53it/s]
1|3466|Loss: 0.8126634359359741:  54%|█████▎    | 3465/6470 [42:46<32:49,  1.53it/s]
1|3466|Loss: 0.8126634359359741:  54%|█████▎    | 3466/6470 [42:47<31:22,  1.60it/s]
1|3467|Loss: 0.8450654149055481:  54%|█████▎    | 3466/6470 [42:47<31:22,  1.60it/s]
1|3467|Loss: 0.8450654149055481:  54%|█████▎    | 3467/6470 [42:47<32:25,  1.54it/s]
1|3468|Loss: 0.9663988351821899:  54%|█████▎    | 3467/6470 [42:48<32:25,  1.54it/s]
1|3468|Loss: 0.9663988351821899:  54%|█████▎    | 3468/6470 [42:48<33:58,  1.47it/s]
1|3469|Loss: 0.9360848665237427:  54%|█████▎    | 3468/6470 [42:48<33:58,  1.47it/s]
1|3469|Loss: 0.9360848665237427:  54%|█████▎    | 3469/6470 [42:49<35:53,  1.39it/s]
1|3470|Loss: 0.905646562576294:  54%|█████▎    | 3469/6470 [42:49<35:53,  1.39it/s] 
1|3470|Loss: 0.905646562576294:  54%|█████▎    | 3470/6470 [42:50<38:04,  1.31it/s]
1|3471|Loss: 0.961968183517456:  54%|█████▎    | 3470/6470 [42:50<38:04,  1.31it/s]
1|3471|Loss: 0.961968183517456:  54%|█████▎    | 3471/6470 [42:50<35:08,  1.42it/s]
1|3472|Loss: 0.6228652000427246:  54%|█████▎    | 3471/6470 [42:51<35:08,  1.42it/s]
1|3472|Loss: 0.6228652000427246:  54%|█████▎    | 3472/6470 [42:51<34:45,  1.44it/s]
1|3473|Loss: 0.8997130393981934:  54%|█████▎    | 3472/6470 [42:51<34:45,  1.44it/s]
1|3473|Loss: 0.8997130393981934:  54%|█████▎    | 3473/6470 [42:52<35:36,  1.40it/s]
1|3474|Loss: 0.8594674468040466:  54%|█████▎    | 3473/6470 [42:52<35:36,  1.40it/s]
1|3474|Loss: 0.8594674468040466:  54%|█████▎    | 3474/6470 [42:52<34:42,  1.44it/s]
1|3475|Loss: 0.8774824142456055:  54%|█████▎    | 3474/6470 [42:53<34:42,  1.44it/s]
1|3475|Loss: 0.8774824142456055:  54%|█████▎    | 3475/6470 [42:53<34:52,  1.43it/s]
1|3476|Loss: 0.9351164698600769:  54%|█████▎    | 3475/6470 [42:53<34:52,  1.43it/s]
1|3476|Loss: 0.9351164698600769:  54%|█████▎    | 3476/6470 [42:54<33:35,  1.49it/s]
1|3477|Loss: 1.185651183128357:  54%|█████▎    | 3476/6470 [42:54<33:35,  1.49it/s] 
1|3477|Loss: 1.185651183128357:  54%|█████▎    | 3477/6470 [42:54<32:17,  1.54it/s]
1|3478|Loss: 0.8840181827545166:  54%|█████▎    | 3477/6470 [42:55<32:17,  1.54it/s]
1|3478|Loss: 0.8840181827545166:  54%|█████▍    | 3478/6470 [42:55<32:45,  1.52it/s]
1|3479|Loss: 1.0630528926849365:  54%|█████▍    | 3478/6470 [42:55<32:45,  1.52it/s]
1|3479|Loss: 1.0630528926849365:  54%|█████▍    | 3479/6470 [42:56<35:00,  1.42it/s]
1|3480|Loss: 0.7987663745880127:  54%|█████▍    | 3479/6470 [42:56<35:00,  1.42it/s]
1|3480|Loss: 0.7987663745880127:  54%|█████▍    | 3480/6470 [42:56<35:26,  1.41it/s]
1|3481|Loss: 0.907146692276001:  54%|█████▍    | 3480/6470 [42:57<35:26,  1.41it/s] 
1|3481|Loss: 0.907146692276001:  54%|█████▍    | 3481/6470 [42:57<34:35,  1.44it/s]
1|3482|Loss: 0.8410791754722595:  54%|█████▍    | 3481/6470 [42:57<34:35,  1.44it/s]
1|3482|Loss: 0.8410791754722595:  54%|█████▍    | 3482/6470 [42:58<31:58,  1.56it/s]
1|3483|Loss: 0.9356157183647156:  54%|█████▍    | 3482/6470 [42:58<31:58,  1.56it/s]
1|3483|Loss: 0.9356157183647156:  54%|█████▍    | 3483/6470 [42:58<34:02,  1.46it/s]
1|3484|Loss: 0.8883497714996338:  54%|█████▍    | 3483/6470 [42:59<34:02,  1.46it/s]
1|3484|Loss: 0.8883497714996338:  54%|█████▍    | 3484/6470 [42:59<35:20,  1.41it/s]
1|3485|Loss: 1.0127408504486084:  54%|█████▍    | 3484/6470 [43:00<35:20,  1.41it/s]
1|3485|Loss: 1.0127408504486084:  54%|█████▍    | 3485/6470 [43:00<33:17,  1.49it/s]
1|3486|Loss: 0.9033217430114746:  54%|█████▍    | 3485/6470 [43:00<33:17,  1.49it/s]
1|3486|Loss: 0.9033217430114746:  54%|█████▍    | 3486/6470 [43:00<33:14,  1.50it/s]
1|3487|Loss: 0.9355684518814087:  54%|█████▍    | 3486/6470 [43:01<33:14,  1.50it/s]
1|3487|Loss: 0.9355684518814087:  54%|█████▍    | 3487/6470 [43:01<32:17,  1.54it/s]
1|3488|Loss: 0.9924971461296082:  54%|█████▍    | 3487/6470 [43:01<32:17,  1.54it/s]
1|3488|Loss: 0.9924971461296082:  54%|█████▍    | 3488/6470 [43:02<32:42,  1.52it/s]
1|3489|Loss: 0.9181057810783386:  54%|█████▍    | 3488/6470 [43:02<32:42,  1.52it/s]
1|3489|Loss: 0.9181057810783386:  54%|█████▍    | 3489/6470 [43:02<34:33,  1.44it/s]
1|3490|Loss: 0.8377503156661987:  54%|█████▍    | 3489/6470 [43:03<34:33,  1.44it/s]
1|3490|Loss: 0.8377503156661987:  54%|█████▍    | 3490/6470 [43:03<34:58,  1.42it/s]
1|3491|Loss: 0.7956023216247559:  54%|█████▍    | 3490/6470 [43:04<34:58,  1.42it/s]
1|3491|Loss: 0.7956023216247559:  54%|█████▍    | 3491/6470 [43:04<34:53,  1.42it/s]
1|3492|Loss: 0.9469257593154907:  54%|█████▍    | 3491/6470 [43:04<34:53,  1.42it/s]
1|3492|Loss: 0.9469257593154907:  54%|█████▍    | 3492/6470 [43:05<33:32,  1.48it/s]
1|3493|Loss: 0.9699838757514954:  54%|█████▍    | 3492/6470 [43:05<33:32,  1.48it/s]
1|3493|Loss: 0.9699838757514954:  54%|█████▍    | 3493/6470 [43:05<33:57,  1.46it/s]
1|3494|Loss: 0.8839027881622314:  54%|█████▍    | 3493/6470 [43:06<33:57,  1.46it/s]
1|3494|Loss: 0.8839027881622314:  54%|█████▍    | 3494/6470 [43:06<32:02,  1.55it/s]
1|3495|Loss: 0.7897400259971619:  54%|█████▍    | 3494/6470 [43:06<32:02,  1.55it/s]
1|3495|Loss: 0.7897400259971619:  54%|█████▍    | 3495/6470 [43:06<32:37,  1.52it/s]
1|3496|Loss: 0.9585908651351929:  54%|█████▍    | 3495/6470 [43:07<32:37,  1.52it/s]
1|3496|Loss: 0.9585908651351929:  54%|█████▍    | 3496/6470 [43:07<33:26,  1.48it/s]
1|3497|Loss: 0.929777204990387:  54%|█████▍    | 3496/6470 [43:08<33:26,  1.48it/s] 
1|3497|Loss: 0.929777204990387:  54%|█████▍    | 3497/6470 [43:08<33:48,  1.47it/s]
1|3498|Loss: 0.9897226691246033:  54%|█████▍    | 3497/6470 [43:08<33:48,  1.47it/s]
1|3498|Loss: 0.9897226691246033:  54%|█████▍    | 3498/6470 [43:08<32:32,  1.52it/s]
1|3499|Loss: 0.805845320224762:  54%|█████▍    | 3498/6470 [43:09<32:32,  1.52it/s] 
1|3499|Loss: 0.805845320224762:  54%|█████▍    | 3499/6470 [43:09<32:07,  1.54it/s]
1|3500|Loss: 0.7273748517036438:  54%|█████▍    | 3499/6470 [43:09<32:07,  1.54it/s]
1|3500|Loss: 0.7273748517036438:  54%|█████▍    | 3500/6470 [43:10<31:01,  1.60it/s]
1|3501|Loss: 0.8924258351325989:  54%|█████▍    | 3500/6470 [43:10<31:01,  1.60it/s]
1|3501|Loss: 0.8924258351325989:  54%|█████▍    | 3501/6470 [43:10<32:09,  1.54it/s]
1|3502|Loss: 0.9380085468292236:  54%|█████▍    | 3501/6470 [43:11<32:09,  1.54it/s]
1|3502|Loss: 0.9380085468292236:  54%|█████▍    | 3502/6470 [43:11<32:59,  1.50it/s]
1|3503|Loss: 0.7675914764404297:  54%|█████▍    | 3502/6470 [43:11<32:59,  1.50it/s]
1|3503|Loss: 0.7675914764404297:  54%|█████▍    | 3503/6470 [43:12<31:36,  1.56it/s]
1|3504|Loss: 0.9034497141838074:  54%|█████▍    | 3503/6470 [43:12<31:36,  1.56it/s]
1|3504|Loss: 0.9034497141838074:  54%|█████▍    | 3504/6470 [43:12<30:37,  1.61it/s]
1|3505|Loss: 0.8238632082939148:  54%|█████▍    | 3504/6470 [43:13<30:37,  1.61it/s]
1|3505|Loss: 0.8238632082939148:  54%|█████▍    | 3505/6470 [43:13<33:05,  1.49it/s]
1|3506|Loss: 0.8797550797462463:  54%|█████▍    | 3505/6470 [43:13<33:05,  1.49it/s]
1|3506|Loss: 0.8797550797462463:  54%|█████▍    | 3506/6470 [43:14<34:30,  1.43it/s]
1|3507|Loss: 0.9022728204727173:  54%|█████▍    | 3506/6470 [43:14<34:30,  1.43it/s]
1|3507|Loss: 0.9022728204727173:  54%|█████▍    | 3507/6470 [43:14<34:06,  1.45it/s]
1|3508|Loss: 0.805970311164856:  54%|█████▍    | 3507/6470 [43:15<34:06,  1.45it/s] 
1|3508|Loss: 0.805970311164856:  54%|█████▍    | 3508/6470 [43:15<32:02,  1.54it/s]
1|3509|Loss: 0.9625900387763977:  54%|█████▍    | 3508/6470 [43:15<32:02,  1.54it/s]
1|3509|Loss: 0.9625900387763977:  54%|█████▍    | 3509/6470 [43:16<31:31,  1.57it/s]
1|3510|Loss: 0.8873277902603149:  54%|█████▍    | 3509/6470 [43:16<31:31,  1.57it/s]
1|3510|Loss: 0.8873277902603149:  54%|█████▍    | 3510/6470 [43:16<28:36,  1.72it/s]
1|3511|Loss: 0.948311448097229:  54%|█████▍    | 3510/6470 [43:16<28:36,  1.72it/s] 
1|3511|Loss: 0.948311448097229:  54%|█████▍    | 3511/6470 [43:17<30:43,  1.61it/s]
1|3512|Loss: 0.8689574003219604:  54%|█████▍    | 3511/6470 [43:17<30:43,  1.61it/s]
1|3512|Loss: 0.8689574003219604:  54%|█████▍    | 3512/6470 [43:18<32:45,  1.50it/s]
1|3513|Loss: 0.8334242105484009:  54%|█████▍    | 3512/6470 [43:18<32:45,  1.50it/s]
1|3513|Loss: 0.8334242105484009:  54%|█████▍    | 3513/6470 [43:18<31:23,  1.57it/s]
1|3514|Loss: 0.7181820869445801:  54%|█████▍    | 3513/6470 [43:19<31:23,  1.57it/s]
1|3514|Loss: 0.7181820869445801:  54%|█████▍    | 3514/6470 [43:19<32:54,  1.50it/s]
1|3515|Loss: 1.037190318107605:  54%|█████▍    | 3514/6470 [43:19<32:54,  1.50it/s] 
1|3515|Loss: 1.037190318107605:  54%|█████▍    | 3515/6470 [43:20<34:17,  1.44it/s]
1|3516|Loss: 0.7283880114555359:  54%|█████▍    | 3515/6470 [43:20<34:17,  1.44it/s]
1|3516|Loss: 0.7283880114555359:  54%|█████▍    | 3516/6470 [43:20<35:44,  1.38it/s]
1|3517|Loss: 0.8894639611244202:  54%|█████▍    | 3516/6470 [43:21<35:44,  1.38it/s]
1|3517|Loss: 0.8894639611244202:  54%|█████▍    | 3517/6470 [43:21<34:41,  1.42it/s]
1|3518|Loss: 0.9449504613876343:  54%|█████▍    | 3517/6470 [43:21<34:41,  1.42it/s]
1|3518|Loss: 0.9449504613876343:  54%|█████▍    | 3518/6470 [43:22<33:20,  1.48it/s]
1|3519|Loss: 0.8193660974502563:  54%|█████▍    | 3518/6470 [43:22<33:20,  1.48it/s]
1|3519|Loss: 0.8193660974502563:  54%|█████▍    | 3519/6470 [43:22<30:52,  1.59it/s]
1|3520|Loss: 0.863057553768158:  54%|█████▍    | 3519/6470 [43:23<30:52,  1.59it/s] 
1|3520|Loss: 0.863057553768158:  54%|█████▍    | 3520/6470 [43:23<31:41,  1.55it/s]
1|3521|Loss: 1.049672245979309:  54%|█████▍    | 3520/6470 [43:23<31:41,  1.55it/s]
1|3521|Loss: 1.049672245979309:  54%|█████▍    | 3521/6470 [43:24<34:28,  1.43it/s]
1|3522|Loss: 0.8708957433700562:  54%|█████▍    | 3521/6470 [43:24<34:28,  1.43it/s]
1|3522|Loss: 0.8708957433700562:  54%|█████▍    | 3522/6470 [43:24<34:13,  1.44it/s]
1|3523|Loss: 0.8822528719902039:  54%|█████▍    | 3522/6470 [43:25<34:13,  1.44it/s]
1|3523|Loss: 0.8822528719902039:  54%|█████▍    | 3523/6470 [43:25<33:02,  1.49it/s]
1|3524|Loss: 0.9005576968193054:  54%|█████▍    | 3523/6470 [43:25<33:02,  1.49it/s]
1|3524|Loss: 0.9005576968193054:  54%|█████▍    | 3524/6470 [43:26<32:11,  1.53it/s]
1|3525|Loss: 0.8838563561439514:  54%|█████▍    | 3524/6470 [43:26<32:11,  1.53it/s]
1|3525|Loss: 0.8838563561439514:  54%|█████▍    | 3525/6470 [43:26<31:06,  1.58it/s]
1|3526|Loss: 0.856640636920929:  54%|█████▍    | 3525/6470 [43:27<31:06,  1.58it/s] 
1|3526|Loss: 0.856640636920929:  54%|█████▍    | 3526/6470 [43:27<31:00,  1.58it/s]
1|3527|Loss: 0.8079088926315308:  54%|█████▍    | 3526/6470 [43:27<31:00,  1.58it/s]
1|3527|Loss: 0.8079088926315308:  55%|█████▍    | 3527/6470 [43:28<31:51,  1.54it/s]
1|3528|Loss: 0.9460896849632263:  55%|█████▍    | 3527/6470 [43:28<31:51,  1.54it/s]
1|3528|Loss: 0.9460896849632263:  55%|█████▍    | 3528/6470 [43:28<31:50,  1.54it/s]
1|3529|Loss: 0.8020254969596863:  55%|█████▍    | 3528/6470 [43:29<31:50,  1.54it/s]
1|3529|Loss: 0.8020254969596863:  55%|█████▍    | 3529/6470 [43:29<32:23,  1.51it/s]
1|3530|Loss: 0.793748676776886:  55%|█████▍    | 3529/6470 [43:29<32:23,  1.51it/s] 
1|3530|Loss: 0.793748676776886:  55%|█████▍    | 3530/6470 [43:30<33:31,  1.46it/s]
1|3531|Loss: 0.9498530626296997:  55%|█████▍    | 3530/6470 [43:30<33:31,  1.46it/s]
1|3531|Loss: 0.9498530626296997:  55%|█████▍    | 3531/6470 [43:30<35:02,  1.40it/s]
1|3532|Loss: 0.8971603512763977:  55%|█████▍    | 3531/6470 [43:31<35:02,  1.40it/s]
1|3532|Loss: 0.8971603512763977:  55%|█████▍    | 3532/6470 [43:31<32:51,  1.49it/s]
1|3533|Loss: 1.0025445222854614:  55%|█████▍    | 3532/6470 [43:31<32:51,  1.49it/s]
1|3533|Loss: 1.0025445222854614:  55%|█████▍    | 3533/6470 [43:32<31:09,  1.57it/s]
1|3534|Loss: 0.8027268052101135:  55%|█████▍    | 3533/6470 [43:32<31:09,  1.57it/s]
1|3534|Loss: 0.8027268052101135:  55%|█████▍    | 3534/6470 [43:32<30:45,  1.59it/s]
1|3535|Loss: 0.8455636501312256:  55%|█████▍    | 3534/6470 [43:32<30:45,  1.59it/s]
1|3535|Loss: 0.8455636501312256:  55%|█████▍    | 3535/6470 [43:33<30:51,  1.59it/s]
1|3536|Loss: 0.9708218574523926:  55%|█████▍    | 3535/6470 [43:33<30:51,  1.59it/s]
1|3536|Loss: 0.9708218574523926:  55%|█████▍    | 3536/6470 [43:33<31:37,  1.55it/s]
1|3537|Loss: 0.8744354248046875:  55%|█████▍    | 3536/6470 [43:34<31:37,  1.55it/s]
1|3537|Loss: 0.8744354248046875:  55%|█████▍    | 3537/6470 [43:34<32:27,  1.51it/s]
1|3538|Loss: 0.8673597574234009:  55%|█████▍    | 3537/6470 [43:35<32:27,  1.51it/s]
1|3538|Loss: 0.8673597574234009:  55%|█████▍    | 3538/6470 [43:35<31:59,  1.53it/s]
1|3539|Loss: 0.7557685375213623:  55%|█████▍    | 3538/6470 [43:35<31:59,  1.53it/s]
1|3539|Loss: 0.7557685375213623:  55%|█████▍    | 3539/6470 [43:36<32:58,  1.48it/s]
1|3540|Loss: 1.091695785522461:  55%|█████▍    | 3539/6470 [43:36<32:58,  1.48it/s] 
1|3540|Loss: 1.091695785522461:  55%|█████▍    | 3540/6470 [43:36<31:50,  1.53it/s]
1|3541|Loss: 0.8322432041168213:  55%|█████▍    | 3540/6470 [43:36<31:50,  1.53it/s]
1|3541|Loss: 0.8322432041168213:  55%|█████▍    | 3541/6470 [43:37<32:30,  1.50it/s]
1|3542|Loss: 0.8483737707138062:  55%|█████▍    | 3541/6470 [43:37<32:30,  1.50it/s]
1|3542|Loss: 0.8483737707138062:  55%|█████▍    | 3542/6470 [43:37<30:53,  1.58it/s]
1|3543|Loss: 0.8468809723854065:  55%|█████▍    | 3542/6470 [43:38<30:53,  1.58it/s]
1|3543|Loss: 0.8468809723854065:  55%|█████▍    | 3543/6470 [43:38<31:40,  1.54it/s]
1|3544|Loss: 1.019381046295166:  55%|█████▍    | 3543/6470 [43:38<31:40,  1.54it/s] 
1|3544|Loss: 1.019381046295166:  55%|█████▍    | 3544/6470 [43:39<34:04,  1.43it/s]
1|3545|Loss: 0.9791852235794067:  55%|█████▍    | 3544/6470 [43:39<34:04,  1.43it/s]
1|3545|Loss: 0.9791852235794067:  55%|█████▍    | 3545/6470 [43:40<34:29,  1.41it/s]
1|3546|Loss: 1.0191030502319336:  55%|█████▍    | 3545/6470 [43:40<34:29,  1.41it/s]
1|3546|Loss: 1.0191030502319336:  55%|█████▍    | 3546/6470 [43:40<33:26,  1.46it/s]
1|3547|Loss: 1.063492774963379:  55%|█████▍    | 3546/6470 [43:41<33:26,  1.46it/s] 
1|3547|Loss: 1.063492774963379:  55%|█████▍    | 3547/6470 [43:41<33:22,  1.46it/s]
1|3548|Loss: 0.8639171123504639:  55%|█████▍    | 3547/6470 [43:41<33:22,  1.46it/s]
1|3548|Loss: 0.8639171123504639:  55%|█████▍    | 3548/6470 [43:42<34:13,  1.42it/s]
1|3549|Loss: 0.9240386486053467:  55%|█████▍    | 3548/6470 [43:42<34:13,  1.42it/s]
1|3549|Loss: 0.9240386486053467:  55%|█████▍    | 3549/6470 [43:42<35:27,  1.37it/s]
1|3550|Loss: 0.8902140259742737:  55%|█████▍    | 3549/6470 [43:43<35:27,  1.37it/s]
1|3550|Loss: 0.8902140259742737:  55%|█████▍    | 3550/6470 [43:43<37:05,  1.31it/s]
1|3551|Loss: 0.9036090970039368:  55%|█████▍    | 3550/6470 [43:44<37:05,  1.31it/s]
1|3551|Loss: 0.9036090970039368:  55%|█████▍    | 3551/6470 [43:44<38:46,  1.25it/s]
1|3552|Loss: 0.7493352890014648:  55%|█████▍    | 3551/6470 [43:45<38:46,  1.25it/s]
1|3552|Loss: 0.7493352890014648:  55%|█████▍    | 3552/6470 [43:45<37:18,  1.30it/s]
1|3553|Loss: 0.6741914749145508:  55%|█████▍    | 3552/6470 [43:45<37:18,  1.30it/s]
1|3553|Loss: 0.6741914749145508:  55%|█████▍    | 3553/6470 [43:46<36:10,  1.34it/s]
1|3554|Loss: 1.0755621194839478:  55%|█████▍    | 3553/6470 [43:46<36:10,  1.34it/s]
1|3554|Loss: 1.0755621194839478:  55%|█████▍    | 3554/6470 [43:46<35:36,  1.37it/s]
1|3555|Loss: 0.8044922351837158:  55%|█████▍    | 3554/6470 [43:47<35:36,  1.37it/s]
1|3555|Loss: 0.8044922351837158:  55%|█████▍    | 3555/6470 [43:47<36:57,  1.31it/s]
1|3556|Loss: 0.8530213236808777:  55%|█████▍    | 3555/6470 [43:48<36:57,  1.31it/s]
1|3556|Loss: 0.8530213236808777:  55%|█████▍    | 3556/6470 [43:48<38:04,  1.28it/s]
1|3557|Loss: 0.8583178520202637:  55%|█████▍    | 3556/6470 [43:48<38:04,  1.28it/s]
1|3557|Loss: 0.8583178520202637:  55%|█████▍    | 3557/6470 [43:49<36:15,  1.34it/s]
1|3558|Loss: 0.9145883321762085:  55%|█████▍    | 3557/6470 [43:49<36:15,  1.34it/s]
1|3558|Loss: 0.9145883321762085:  55%|█████▍    | 3558/6470 [43:49<35:42,  1.36it/s]
1|3559|Loss: 0.8263086676597595:  55%|█████▍    | 3558/6470 [43:50<35:42,  1.36it/s]
1|3559|Loss: 0.8263086676597595:  55%|█████▌    | 3559/6470 [43:50<33:02,  1.47it/s]
1|3560|Loss: 0.8168945908546448:  55%|█████▌    | 3559/6470 [43:50<33:02,  1.47it/s]
1|3560|Loss: 0.8168945908546448:  55%|█████▌    | 3560/6470 [43:50<32:07,  1.51it/s]
1|3561|Loss: 0.7878068685531616:  55%|█████▌    | 3560/6470 [43:51<32:07,  1.51it/s]
1|3561|Loss: 0.7878068685531616:  55%|█████▌    | 3561/6470 [43:51<30:49,  1.57it/s]
1|3562|Loss: 0.9533231258392334:  55%|█████▌    | 3561/6470 [43:51<30:49,  1.57it/s]
1|3562|Loss: 0.9533231258392334:  55%|█████▌    | 3562/6470 [43:52<30:22,  1.60it/s]
1|3563|Loss: 0.7285933494567871:  55%|█████▌    | 3562/6470 [43:52<30:22,  1.60it/s]
1|3563|Loss: 0.7285933494567871:  55%|█████▌    | 3563/6470 [43:52<30:29,  1.59it/s]
1|3564|Loss: 0.8217044472694397:  55%|█████▌    | 3563/6470 [43:53<30:29,  1.59it/s]
1|3564|Loss: 0.8217044472694397:  55%|█████▌    | 3564/6470 [43:53<30:37,  1.58it/s]
1|3565|Loss: 0.8645932674407959:  55%|█████▌    | 3564/6470 [43:53<30:37,  1.58it/s]
1|3565|Loss: 0.8645932674407959:  55%|█████▌    | 3565/6470 [43:54<31:26,  1.54it/s]
1|3566|Loss: 0.9663687944412231:  55%|█████▌    | 3565/6470 [43:54<31:26,  1.54it/s]
1|3566|Loss: 0.9663687944412231:  55%|█████▌    | 3566/6470 [43:54<28:16,  1.71it/s]
1|3567|Loss: 0.884472131729126:  55%|█████▌    | 3566/6470 [43:54<28:16,  1.71it/s] 
1|3567|Loss: 0.884472131729126:  55%|█████▌    | 3567/6470 [43:55<28:24,  1.70it/s]
1|3568|Loss: 0.8985942602157593:  55%|█████▌    | 3567/6470 [43:55<28:24,  1.70it/s]
1|3568|Loss: 0.8985942602157593:  55%|█████▌    | 3568/6470 [43:55<30:19,  1.60it/s]
1|3569|Loss: 0.912300705909729:  55%|█████▌    | 3568/6470 [43:56<30:19,  1.60it/s] 
1|3569|Loss: 0.912300705909729:  55%|█████▌    | 3569/6470 [43:56<28:38,  1.69it/s]
1|3570|Loss: 0.9130067229270935:  55%|█████▌    | 3569/6470 [43:56<28:38,  1.69it/s]
1|3570|Loss: 0.9130067229270935:  55%|█████▌    | 3570/6470 [43:57<29:59,  1.61it/s]
1|3571|Loss: 0.9460355043411255:  55%|█████▌    | 3570/6470 [43:57<29:59,  1.61it/s]
1|3571|Loss: 0.9460355043411255:  55%|█████▌    | 3571/6470 [43:57<30:31,  1.58it/s]
1|3572|Loss: 0.7347816824913025:  55%|█████▌    | 3571/6470 [43:58<30:31,  1.58it/s]
1|3572|Loss: 0.7347816824913025:  55%|█████▌    | 3572/6470 [43:58<31:31,  1.53it/s]
1|3573|Loss: 0.911481499671936:  55%|█████▌    | 3572/6470 [43:58<31:31,  1.53it/s] 
1|3573|Loss: 0.911481499671936:  55%|█████▌    | 3573/6470 [43:59<31:29,  1.53it/s]
1|3574|Loss: 0.9456428289413452:  55%|█████▌    | 3573/6470 [43:59<31:29,  1.53it/s]
1|3574|Loss: 0.9456428289413452:  55%|█████▌    | 3574/6470 [43:59<32:21,  1.49it/s]
1|3575|Loss: 0.8454697728157043:  55%|█████▌    | 3574/6470 [44:00<32:21,  1.49it/s]
1|3575|Loss: 0.8454697728157043:  55%|█████▌    | 3575/6470 [44:00<31:26,  1.53it/s]
1|3576|Loss: 0.9583995938301086:  55%|█████▌    | 3575/6470 [44:00<31:26,  1.53it/s]
1|3576|Loss: 0.9583995938301086:  55%|█████▌    | 3576/6470 [44:00<30:02,  1.61it/s]
1|3577|Loss: 0.8216676115989685:  55%|█████▌    | 3576/6470 [44:01<30:02,  1.61it/s]
1|3577|Loss: 0.8216676115989685:  55%|█████▌    | 3577/6470 [44:01<29:44,  1.62it/s]
1|3578|Loss: 0.8131595849990845:  55%|█████▌    | 3577/6470 [44:01<29:44,  1.62it/s]
1|3578|Loss: 0.8131595849990845:  55%|█████▌    | 3578/6470 [44:02<32:24,  1.49it/s]
1|3579|Loss: 1.0173194408416748:  55%|█████▌    | 3578/6470 [44:02<32:24,  1.49it/s]
1|3579|Loss: 1.0173194408416748:  55%|█████▌    | 3579/6470 [44:03<32:36,  1.48it/s]
1|3580|Loss: 0.8399364948272705:  55%|█████▌    | 3579/6470 [44:03<32:36,  1.48it/s]
1|3580|Loss: 0.8399364948272705:  55%|█████▌    | 3580/6470 [44:03<32:49,  1.47it/s]
1|3581|Loss: 0.7797150611877441:  55%|█████▌    | 3580/6470 [44:04<32:49,  1.47it/s]
1|3581|Loss: 0.7797150611877441:  55%|█████▌    | 3581/6470 [44:04<33:28,  1.44it/s]
1|3582|Loss: 1.0579519271850586:  55%|█████▌    | 3581/6470 [44:04<33:28,  1.44it/s]
1|3582|Loss: 1.0579519271850586:  55%|█████▌    | 3582/6470 [44:05<32:21,  1.49it/s]
1|3583|Loss: 0.7692457437515259:  55%|█████▌    | 3582/6470 [44:05<32:21,  1.49it/s]
1|3583|Loss: 0.7692457437515259:  55%|█████▌    | 3583/6470 [44:05<32:30,  1.48it/s]
1|3584|Loss: 0.946608304977417:  55%|█████▌    | 3583/6470 [44:06<32:30,  1.48it/s] 
1|3584|Loss: 0.946608304977417:  55%|█████▌    | 3584/6470 [44:06<32:14,  1.49it/s]
1|3585|Loss: 1.0315628051757812:  55%|█████▌    | 3584/6470 [44:06<32:14,  1.49it/s]
1|3585|Loss: 1.0315628051757812:  55%|█████▌    | 3585/6470 [44:07<32:51,  1.46it/s]
1|3586|Loss: 0.7705190777778625:  55%|█████▌    | 3585/6470 [44:07<32:51,  1.46it/s]
1|3586|Loss: 0.7705190777778625:  55%|█████▌    | 3586/6470 [44:07<33:29,  1.44it/s]
1|3587|Loss: 0.9006085991859436:  55%|█████▌    | 3586/6470 [44:08<33:29,  1.44it/s]
1|3587|Loss: 0.9006085991859436:  55%|█████▌    | 3587/6470 [44:08<34:39,  1.39it/s]
1|3588|Loss: 0.9018495678901672:  55%|█████▌    | 3587/6470 [44:09<34:39,  1.39it/s]
1|3588|Loss: 0.9018495678901672:  55%|█████▌    | 3588/6470 [44:09<35:04,  1.37it/s]
1|3589|Loss: 0.8972233533859253:  55%|█████▌    | 3588/6470 [44:09<35:04,  1.37it/s]
1|3589|Loss: 0.8972233533859253:  55%|█████▌    | 3589/6470 [44:10<33:21,  1.44it/s]
1|3590|Loss: 0.8769190907478333:  55%|█████▌    | 3589/6470 [44:10<33:21,  1.44it/s]
1|3590|Loss: 0.8769190907478333:  55%|█████▌    | 3590/6470 [44:10<30:55,  1.55it/s]
1|3591|Loss: 0.9040581583976746:  55%|█████▌    | 3590/6470 [44:10<30:55,  1.55it/s]
1|3591|Loss: 0.9040581583976746:  56%|█████▌    | 3591/6470 [44:11<30:22,  1.58it/s]
1|3592|Loss: 0.93888258934021:  56%|█████▌    | 3591/6470 [44:11<30:22,  1.58it/s]  
1|3592|Loss: 0.93888258934021:  56%|█████▌    | 3592/6470 [44:11<31:36,  1.52it/s]
1|3593|Loss: 0.8129381537437439:  56%|█████▌    | 3592/6470 [44:12<31:36,  1.52it/s]
1|3593|Loss: 0.8129381537437439:  56%|█████▌    | 3593/6470 [44:12<30:36,  1.57it/s]
1|3594|Loss: 0.9067249894142151:  56%|█████▌    | 3593/6470 [44:12<30:36,  1.57it/s]
1|3594|Loss: 0.9067249894142151:  56%|█████▌    | 3594/6470 [44:13<32:16,  1.49it/s]
1|3595|Loss: 1.051076054573059:  56%|█████▌    | 3594/6470 [44:13<32:16,  1.49it/s] 
1|3595|Loss: 1.051076054573059:  56%|█████▌    | 3595/6470 [44:13<33:36,  1.43it/s]
1|3596|Loss: 1.0347508192062378:  56%|█████▌    | 3595/6470 [44:14<33:36,  1.43it/s]
1|3596|Loss: 1.0347508192062378:  56%|█████▌    | 3596/6470 [44:14<32:54,  1.46it/s]
1|3597|Loss: 0.9005930423736572:  56%|█████▌    | 3596/6470 [44:14<32:54,  1.46it/s]
1|3597|Loss: 0.9005930423736572:  56%|█████▌    | 3597/6470 [44:15<33:10,  1.44it/s]
1|3598|Loss: 0.8424292802810669:  56%|█████▌    | 3597/6470 [44:15<33:10,  1.44it/s]
1|3598|Loss: 0.8424292802810669:  56%|█████▌    | 3598/6470 [44:15<32:28,  1.47it/s]
1|3599|Loss: 0.7941784858703613:  56%|█████▌    | 3598/6470 [44:16<32:28,  1.47it/s]
1|3599|Loss: 0.7941784858703613:  56%|█████▌    | 3599/6470 [44:16<31:30,  1.52it/s]
1|3600|Loss: 1.0295029878616333:  56%|█████▌    | 3599/6470 [44:16<31:30,  1.52it/s]
1|3600|Loss: 1.0295029878616333:  56%|█████▌    | 3600/6470 [44:17<29:51,  1.60it/s]
1|3601|Loss: 0.8404898047447205:  56%|█████▌    | 3600/6470 [44:17<29:51,  1.60it/s]
1|3601|Loss: 0.8404898047447205:  56%|█████▌    | 3601/6470 [44:17<30:52,  1.55it/s]
1|3602|Loss: 0.8156454563140869:  56%|█████▌    | 3601/6470 [44:18<30:52,  1.55it/s]
1|3602|Loss: 0.8156454563140869:  56%|█████▌    | 3602/6470 [44:18<31:10,  1.53it/s]
1|3603|Loss: 0.845106303691864:  56%|█████▌    | 3602/6470 [44:18<31:10,  1.53it/s] 
1|3603|Loss: 0.845106303691864:  56%|█████▌    | 3603/6470 [44:19<33:06,  1.44it/s]
1|3604|Loss: 1.0430614948272705:  56%|█████▌    | 3603/6470 [44:19<33:06,  1.44it/s]
1|3604|Loss: 1.0430614948272705:  56%|█████▌    | 3604/6470 [44:19<32:37,  1.46it/s]
1|3605|Loss: 0.8603554368019104:  56%|█████▌    | 3604/6470 [44:20<32:37,  1.46it/s]
1|3605|Loss: 0.8603554368019104:  56%|█████▌    | 3605/6470 [44:20<31:37,  1.51it/s]
1|3606|Loss: 0.729920506477356:  56%|█████▌    | 3605/6470 [44:20<31:37,  1.51it/s] 
1|3606|Loss: 0.729920506477356:  56%|█████▌    | 3606/6470 [44:21<33:18,  1.43it/s]
1|3607|Loss: 0.8425565958023071:  56%|█████▌    | 3606/6470 [44:21<33:18,  1.43it/s]
1|3607|Loss: 0.8425565958023071:  56%|█████▌    | 3607/6470 [44:22<33:42,  1.42it/s]
1|3608|Loss: 0.8553971648216248:  56%|█████▌    | 3607/6470 [44:22<33:42,  1.42it/s]
1|3608|Loss: 0.8553971648216248:  56%|█████▌    | 3608/6470 [44:22<33:50,  1.41it/s]
1|3609|Loss: 0.7451751232147217:  56%|█████▌    | 3608/6470 [44:23<33:50,  1.41it/s]
1|3609|Loss: 0.7451751232147217:  56%|█████▌    | 3609/6470 [44:23<34:02,  1.40it/s]
1|3610|Loss: 0.8594874143600464:  56%|█████▌    | 3609/6470 [44:23<34:02,  1.40it/s]
1|3610|Loss: 0.8594874143600464:  56%|█████▌    | 3610/6470 [44:24<34:16,  1.39it/s]
1|3611|Loss: 0.8723511099815369:  56%|█████▌    | 3610/6470 [44:24<34:16,  1.39it/s]
1|3611|Loss: 0.8723511099815369:  56%|█████▌    | 3611/6470 [44:24<34:22,  1.39it/s]
1|3612|Loss: 0.8358723521232605:  56%|█████▌    | 3611/6470 [44:25<34:22,  1.39it/s]
1|3612|Loss: 0.8358723521232605:  56%|█████▌    | 3612/6470 [44:25<32:24,  1.47it/s]
1|3613|Loss: 0.894483745098114:  56%|█████▌    | 3612/6470 [44:25<32:24,  1.47it/s] 
1|3613|Loss: 0.894483745098114:  56%|█████▌    | 3613/6470 [44:26<34:01,  1.40it/s]
1|3614|Loss: 0.9075085520744324:  56%|█████▌    | 3613/6470 [44:26<34:01,  1.40it/s]
1|3614|Loss: 0.9075085520744324:  56%|█████▌    | 3614/6470 [44:27<33:40,  1.41it/s]
1|3615|Loss: 0.9425169825553894:  56%|█████▌    | 3614/6470 [44:27<33:40,  1.41it/s]
1|3615|Loss: 0.9425169825553894:  56%|█████▌    | 3615/6470 [44:27<31:25,  1.51it/s]
1|3616|Loss: 0.8480769395828247:  56%|█████▌    | 3615/6470 [44:27<31:25,  1.51it/s]
1|3616|Loss: 0.8480769395828247:  56%|█████▌    | 3616/6470 [44:28<32:03,  1.48it/s]
1|3617|Loss: 0.8101336359977722:  56%|█████▌    | 3616/6470 [44:28<32:03,  1.48it/s]
1|3617|Loss: 0.8101336359977722:  56%|█████▌    | 3617/6470 [44:29<36:14,  1.31it/s]
1|3618|Loss: 0.9187829494476318:  56%|█████▌    | 3617/6470 [44:29<36:14,  1.31it/s]
1|3618|Loss: 0.9187829494476318:  56%|█████▌    | 3618/6470 [44:30<36:45,  1.29it/s]
1|3619|Loss: 0.8187087178230286:  56%|█████▌    | 3618/6470 [44:30<36:45,  1.29it/s]
1|3619|Loss: 0.8187087178230286:  56%|█████▌    | 3619/6470 [44:30<36:08,  1.31it/s]
1|3620|Loss: 0.8209431767463684:  56%|█████▌    | 3619/6470 [44:31<36:08,  1.31it/s]
1|3620|Loss: 0.8209431767463684:  56%|█████▌    | 3620/6470 [44:31<35:45,  1.33it/s]
1|3621|Loss: 0.7744821906089783:  56%|█████▌    | 3620/6470 [44:31<35:45,  1.33it/s]
1|3621|Loss: 0.7744821906089783:  56%|█████▌    | 3621/6470 [44:32<36:51,  1.29it/s]
1|3622|Loss: 0.9043393135070801:  56%|█████▌    | 3621/6470 [44:32<36:51,  1.29it/s]
1|3622|Loss: 0.9043393135070801:  56%|█████▌    | 3622/6470 [44:33<36:48,  1.29it/s]
1|3623|Loss: 0.8692184090614319:  56%|█████▌    | 3622/6470 [44:33<36:48,  1.29it/s]
1|3623|Loss: 0.8692184090614319:  56%|█████▌    | 3623/6470 [44:33<37:45,  1.26it/s]
1|3624|Loss: 0.7890514731407166:  56%|█████▌    | 3623/6470 [44:34<37:45,  1.26it/s]
1|3624|Loss: 0.7890514731407166:  56%|█████▌    | 3624/6470 [44:34<37:20,  1.27it/s]
1|3625|Loss: 0.906683087348938:  56%|█████▌    | 3624/6470 [44:35<37:20,  1.27it/s] 
1|3625|Loss: 0.906683087348938:  56%|█████▌    | 3625/6470 [44:35<36:29,  1.30it/s]
1|3626|Loss: 0.803225576877594:  56%|█████▌    | 3625/6470 [44:35<36:29,  1.30it/s]
1|3626|Loss: 0.803225576877594:  56%|█████▌    | 3626/6470 [44:36<34:57,  1.36it/s]
1|3627|Loss: 0.8649328351020813:  56%|█████▌    | 3626/6470 [44:36<34:57,  1.36it/s]
1|3627|Loss: 0.8649328351020813:  56%|█████▌    | 3627/6470 [44:36<33:37,  1.41it/s]
1|3628|Loss: 0.8991295099258423:  56%|█████▌    | 3627/6470 [44:37<33:37,  1.41it/s]
1|3628|Loss: 0.8991295099258423:  56%|█████▌    | 3628/6470 [44:37<32:24,  1.46it/s]
1|3629|Loss: 0.7470288276672363:  56%|█████▌    | 3628/6470 [44:37<32:24,  1.46it/s]
1|3629|Loss: 0.7470288276672363:  56%|█████▌    | 3629/6470 [44:38<32:44,  1.45it/s]
1|3630|Loss: 0.9851428866386414:  56%|█████▌    | 3629/6470 [44:38<32:44,  1.45it/s]
1|3630|Loss: 0.9851428866386414:  56%|█████▌    | 3630/6470 [44:38<31:41,  1.49it/s]
1|3631|Loss: 0.8735321164131165:  56%|█████▌    | 3630/6470 [44:39<31:41,  1.49it/s]
1|3631|Loss: 0.8735321164131165:  56%|█████▌    | 3631/6470 [44:39<30:53,  1.53it/s]
1|3632|Loss: 0.8841672539710999:  56%|█████▌    | 3631/6470 [44:39<30:53,  1.53it/s]
1|3632|Loss: 0.8841672539710999:  56%|█████▌    | 3632/6470 [44:40<31:25,  1.51it/s]
1|3633|Loss: 0.8983956575393677:  56%|█████▌    | 3632/6470 [44:40<31:25,  1.51it/s]
1|3633|Loss: 0.8983956575393677:  56%|█████▌    | 3633/6470 [44:40<31:11,  1.52it/s]
1|3634|Loss: 0.8617305755615234:  56%|█████▌    | 3633/6470 [44:40<31:11,  1.52it/s]
1|3634|Loss: 0.8617305755615234:  56%|█████▌    | 3634/6470 [44:41<28:45,  1.64it/s]
1|3635|Loss: 0.7810915112495422:  56%|█████▌    | 3634/6470 [44:41<28:45,  1.64it/s]
1|3635|Loss: 0.7810915112495422:  56%|█████▌    | 3635/6470 [44:41<29:35,  1.60it/s]
1|3636|Loss: 0.9993866682052612:  56%|█████▌    | 3635/6470 [44:42<29:35,  1.60it/s]
1|3636|Loss: 0.9993866682052612:  56%|█████▌    | 3636/6470 [44:42<29:47,  1.59it/s]
1|3637|Loss: 0.8247438073158264:  56%|█████▌    | 3636/6470 [44:42<29:47,  1.59it/s]
1|3637|Loss: 0.8247438073158264:  56%|█████▌    | 3637/6470 [44:43<29:29,  1.60it/s]
1|3638|Loss: 0.8709563612937927:  56%|█████▌    | 3637/6470 [44:43<29:29,  1.60it/s]
1|3638|Loss: 0.8709563612937927:  56%|█████▌    | 3638/6470 [44:43<29:22,  1.61it/s]
1|3639|Loss: 0.8782950639724731:  56%|█████▌    | 3638/6470 [44:44<29:22,  1.61it/s]
1|3639|Loss: 0.8782950639724731:  56%|█████▌    | 3639/6470 [44:44<30:18,  1.56it/s]
1|3640|Loss: 0.8623751997947693:  56%|█████▌    | 3639/6470 [44:44<30:18,  1.56it/s]
1|3640|Loss: 0.8623751997947693:  56%|█████▋    | 3640/6470 [44:45<33:06,  1.42it/s]
1|3641|Loss: 0.766339123249054:  56%|█████▋    | 3640/6470 [44:45<33:06,  1.42it/s] 
1|3641|Loss: 0.766339123249054:  56%|█████▋    | 3641/6470 [44:46<35:26,  1.33it/s]
1|3642|Loss: 0.8986942172050476:  56%|█████▋    | 3641/6470 [44:46<35:26,  1.33it/s]
1|3642|Loss: 0.8986942172050476:  56%|█████▋    | 3642/6470 [44:46<36:34,  1.29it/s]
1|3643|Loss: 0.9398156404495239:  56%|█████▋    | 3642/6470 [44:47<36:34,  1.29it/s]
1|3643|Loss: 0.9398156404495239:  56%|█████▋    | 3643/6470 [44:47<36:14,  1.30it/s]
1|3644|Loss: 0.8064365386962891:  56%|█████▋    | 3643/6470 [44:48<36:14,  1.30it/s]
1|3644|Loss: 0.8064365386962891:  56%|█████▋    | 3644/6470 [44:48<34:39,  1.36it/s]
1|3645|Loss: 0.7924553155899048:  56%|█████▋    | 3644/6470 [44:48<34:39,  1.36it/s]
1|3645|Loss: 0.7924553155899048:  56%|█████▋    | 3645/6470 [44:48<32:35,  1.44it/s]
1|3646|Loss: 0.8272400498390198:  56%|█████▋    | 3645/6470 [44:49<32:35,  1.44it/s]
1|3646|Loss: 0.8272400498390198:  56%|█████▋    | 3646/6470 [44:49<34:27,  1.37it/s]
1|3647|Loss: 0.8815111517906189:  56%|█████▋    | 3646/6470 [44:50<34:27,  1.37it/s]
1|3647|Loss: 0.8815111517906189:  56%|█████▋    | 3647/6470 [44:50<34:37,  1.36it/s]
1|3648|Loss: 0.8973727822303772:  56%|█████▋    | 3647/6470 [44:50<34:37,  1.36it/s]
1|3648|Loss: 0.8973727822303772:  56%|█████▋    | 3648/6470 [44:51<33:27,  1.41it/s]
1|3649|Loss: 0.827159583568573:  56%|█████▋    | 3648/6470 [44:51<33:27,  1.41it/s] 
1|3649|Loss: 0.827159583568573:  56%|█████▋    | 3649/6470 [44:51<34:29,  1.36it/s]
1|3650|Loss: 1.0093435049057007:  56%|█████▋    | 3649/6470 [44:52<34:29,  1.36it/s]
1|3650|Loss: 1.0093435049057007:  56%|█████▋    | 3650/6470 [44:52<32:16,  1.46it/s]
1|3651|Loss: 0.905390202999115:  56%|█████▋    | 3650/6470 [44:52<32:16,  1.46it/s] 
1|3651|Loss: 0.905390202999115:  56%|█████▋    | 3651/6470 [44:53<30:57,  1.52it/s]
1|3652|Loss: 1.0072612762451172:  56%|█████▋    | 3651/6470 [44:53<30:57,  1.52it/s]
1|3652|Loss: 1.0072612762451172:  56%|█████▋    | 3652/6470 [44:53<32:25,  1.45it/s]
1|3653|Loss: 0.9732156991958618:  56%|█████▋    | 3652/6470 [44:54<32:25,  1.45it/s]
1|3653|Loss: 0.9732156991958618:  56%|█████▋    | 3653/6470 [44:54<31:24,  1.49it/s]
1|3654|Loss: 1.046280026435852:  56%|█████▋    | 3653/6470 [44:54<31:24,  1.49it/s] 
1|3654|Loss: 1.046280026435852:  56%|█████▋    | 3654/6470 [44:55<30:50,  1.52it/s]
1|3655|Loss: 0.8431774377822876:  56%|█████▋    | 3654/6470 [44:55<30:50,  1.52it/s]
1|3655|Loss: 0.8431774377822876:  56%|█████▋    | 3655/6470 [44:55<33:08,  1.42it/s]
1|3656|Loss: 0.7397457361221313:  56%|█████▋    | 3655/6470 [44:56<33:08,  1.42it/s]
1|3656|Loss: 0.7397457361221313:  57%|█████▋    | 3656/6470 [44:56<29:40,  1.58it/s]
1|3657|Loss: 0.7366971969604492:  57%|█████▋    | 3656/6470 [44:56<29:40,  1.58it/s]
1|3657|Loss: 0.7366971969604492:  57%|█████▋    | 3657/6470 [44:56<28:05,  1.67it/s]
1|3658|Loss: 0.9203691482543945:  57%|█████▋    | 3657/6470 [44:57<28:05,  1.67it/s]
1|3658|Loss: 0.9203691482543945:  57%|█████▋    | 3658/6470 [44:57<28:24,  1.65it/s]
1|3659|Loss: 0.858235239982605:  57%|█████▋    | 3658/6470 [44:57<28:24,  1.65it/s] 
1|3659|Loss: 0.858235239982605:  57%|█████▋    | 3659/6470 [44:58<28:35,  1.64it/s]
1|3660|Loss: 0.8095876574516296:  57%|█████▋    | 3659/6470 [44:58<28:35,  1.64it/s]
1|3660|Loss: 0.8095876574516296:  57%|█████▋    | 3660/6470 [44:58<27:55,  1.68it/s]
1|3661|Loss: 0.8622851967811584:  57%|█████▋    | 3660/6470 [44:59<27:55,  1.68it/s]
1|3661|Loss: 0.8622851967811584:  57%|█████▋    | 3661/6470 [44:59<27:35,  1.70it/s]
1|3662|Loss: 0.8534407019615173:  57%|█████▋    | 3661/6470 [44:59<27:35,  1.70it/s]
1|3662|Loss: 0.8534407019615173:  57%|█████▋    | 3662/6470 [45:00<30:32,  1.53it/s]
1|3663|Loss: 0.7076911926269531:  57%|█████▋    | 3662/6470 [45:00<30:32,  1.53it/s]
1|3663|Loss: 0.7076911926269531:  57%|█████▋    | 3663/6470 [45:00<33:41,  1.39it/s]
1|3664|Loss: 0.9149472117424011:  57%|█████▋    | 3663/6470 [45:01<33:41,  1.39it/s]
1|3664|Loss: 0.9149472117424011:  57%|█████▋    | 3664/6470 [45:01<34:07,  1.37it/s]
1|3665|Loss: 0.8810836672782898:  57%|█████▋    | 3664/6470 [45:02<34:07,  1.37it/s]
1|3665|Loss: 0.8810836672782898:  57%|█████▋    | 3665/6470 [45:02<35:22,  1.32it/s]
1|3666|Loss: 0.9220631122589111:  57%|█████▋    | 3665/6470 [45:02<35:22,  1.32it/s]
1|3666|Loss: 0.9220631122589111:  57%|█████▋    | 3666/6470 [45:03<34:03,  1.37it/s]
1|3667|Loss: 0.8645784854888916:  57%|█████▋    | 3666/6470 [45:03<34:03,  1.37it/s]
1|3667|Loss: 0.8645784854888916:  57%|█████▋    | 3667/6470 [45:03<33:24,  1.40it/s]
1|3668|Loss: 0.907987117767334:  57%|█████▋    | 3667/6470 [45:04<33:24,  1.40it/s] 
1|3668|Loss: 0.907987117767334:  57%|█████▋    | 3668/6470 [45:04<34:14,  1.36it/s]
1|3669|Loss: 0.9670490026473999:  57%|█████▋    | 3668/6470 [45:05<34:14,  1.36it/s]
1|3669|Loss: 0.9670490026473999:  57%|█████▋    | 3669/6470 [45:05<32:41,  1.43it/s]
1|3670|Loss: 1.0253690481185913:  57%|█████▋    | 3669/6470 [45:05<32:41,  1.43it/s]
1|3670|Loss: 1.0253690481185913:  57%|█████▋    | 3670/6470 [45:05<31:27,  1.48it/s]
1|3671|Loss: 1.018355131149292:  57%|█████▋    | 3670/6470 [45:06<31:27,  1.48it/s] 
1|3671|Loss: 1.018355131149292:  57%|█████▋    | 3671/6470 [45:06<30:56,  1.51it/s]
1|3672|Loss: 1.0209914445877075:  57%|█████▋    | 3671/6470 [45:06<30:56,  1.51it/s]
1|3672|Loss: 1.0209914445877075:  57%|█████▋    | 3672/6470 [45:07<31:18,  1.49it/s]
1|3673|Loss: 0.8149901628494263:  57%|█████▋    | 3672/6470 [45:07<31:18,  1.49it/s]
1|3673|Loss: 0.8149901628494263:  57%|█████▋    | 3673/6470 [45:08<33:35,  1.39it/s]
1|3674|Loss: 0.8017382025718689:  57%|█████▋    | 3673/6470 [45:08<33:35,  1.39it/s]
1|3674|Loss: 0.8017382025718689:  57%|█████▋    | 3674/6470 [45:08<33:11,  1.40it/s]
1|3675|Loss: 0.8514841198921204:  57%|█████▋    | 3674/6470 [45:09<33:11,  1.40it/s]
1|3675|Loss: 0.8514841198921204:  57%|█████▋    | 3675/6470 [45:09<33:19,  1.40it/s]
1|3676|Loss: 0.7710791826248169:  57%|█████▋    | 3675/6470 [45:09<33:19,  1.40it/s]
1|3676|Loss: 0.7710791826248169:  57%|█████▋    | 3676/6470 [45:10<34:40,  1.34it/s]
1|3677|Loss: 0.8344957232475281:  57%|█████▋    | 3676/6470 [45:10<34:40,  1.34it/s]
1|3677|Loss: 0.8344957232475281:  57%|█████▋    | 3677/6470 [45:10<33:34,  1.39it/s]
1|3678|Loss: 0.8154771327972412:  57%|█████▋    | 3677/6470 [45:11<33:34,  1.39it/s]
1|3678|Loss: 0.8154771327972412:  57%|█████▋    | 3678/6470 [45:11<33:01,  1.41it/s]
1|3679|Loss: 0.993959903717041:  57%|█████▋    | 3678/6470 [45:12<33:01,  1.41it/s] 
1|3679|Loss: 0.993959903717041:  57%|█████▋    | 3679/6470 [45:12<32:57,  1.41it/s]
1|3680|Loss: 0.7571930289268494:  57%|█████▋    | 3679/6470 [45:12<32:57,  1.41it/s]
1|3680|Loss: 0.7571930289268494:  57%|█████▋    | 3680/6470 [45:13<33:13,  1.40it/s]
1|3681|Loss: 0.9012746214866638:  57%|█████▋    | 3680/6470 [45:13<33:13,  1.40it/s]
1|3681|Loss: 0.9012746214866638:  57%|█████▋    | 3681/6470 [45:13<30:43,  1.51it/s]
1|3682|Loss: 0.8687092065811157:  57%|█████▋    | 3681/6470 [45:13<30:43,  1.51it/s]
1|3682|Loss: 0.8687092065811157:  57%|█████▋    | 3682/6470 [45:14<29:56,  1.55it/s]
1|3683|Loss: 0.8818645477294922:  57%|█████▋    | 3682/6470 [45:14<29:56,  1.55it/s]
1|3683|Loss: 0.8818645477294922:  57%|█████▋    | 3683/6470 [45:14<30:37,  1.52it/s]
1|3684|Loss: 0.9538351893424988:  57%|█████▋    | 3683/6470 [45:15<30:37,  1.52it/s]
1|3684|Loss: 0.9538351893424988:  57%|█████▋    | 3684/6470 [45:15<31:39,  1.47it/s]
1|3685|Loss: 0.9183884263038635:  57%|█████▋    | 3684/6470 [45:16<31:39,  1.47it/s]
1|3685|Loss: 0.9183884263038635:  57%|█████▋    | 3685/6470 [45:16<33:02,  1.41it/s]
1|3686|Loss: 0.8700220584869385:  57%|█████▋    | 3685/6470 [45:16<33:02,  1.41it/s]
1|3686|Loss: 0.8700220584869385:  57%|█████▋    | 3686/6470 [45:17<32:29,  1.43it/s]
1|3687|Loss: 0.8277904391288757:  57%|█████▋    | 3686/6470 [45:17<32:29,  1.43it/s]
1|3687|Loss: 0.8277904391288757:  57%|█████▋    | 3687/6470 [45:17<32:40,  1.42it/s]
1|3688|Loss: 0.915437638759613:  57%|█████▋    | 3687/6470 [45:18<32:40,  1.42it/s] 
1|3688|Loss: 0.915437638759613:  57%|█████▋    | 3688/6470 [45:18<34:18,  1.35it/s]
1|3689|Loss: 0.9619128108024597:  57%|█████▋    | 3688/6470 [45:19<34:18,  1.35it/s]
1|3689|Loss: 0.9619128108024597:  57%|█████▋    | 3689/6470 [45:19<32:11,  1.44it/s]
1|3690|Loss: 0.9033725261688232:  57%|█████▋    | 3689/6470 [45:19<32:11,  1.44it/s]
1|3690|Loss: 0.9033725261688232:  57%|█████▋    | 3690/6470 [45:19<30:03,  1.54it/s]
1|3691|Loss: 1.01656174659729:  57%|█████▋    | 3690/6470 [45:20<30:03,  1.54it/s]  
1|3691|Loss: 1.01656174659729:  57%|█████▋    | 3691/6470 [45:20<30:59,  1.49it/s]
1|3692|Loss: 0.9229668378829956:  57%|█████▋    | 3691/6470 [45:20<30:59,  1.49it/s]
1|3692|Loss: 0.9229668378829956:  57%|█████▋    | 3692/6470 [45:21<28:52,  1.60it/s]
1|3693|Loss: 0.8508011102676392:  57%|█████▋    | 3692/6470 [45:21<28:52,  1.60it/s]
1|3693|Loss: 0.8508011102676392:  57%|█████▋    | 3693/6470 [45:21<29:49,  1.55it/s]
1|3694|Loss: 0.8318485617637634:  57%|█████▋    | 3693/6470 [45:22<29:49,  1.55it/s]
1|3694|Loss: 0.8318485617637634:  57%|█████▋    | 3694/6470 [45:22<29:27,  1.57it/s]
1|3695|Loss: 0.8301772475242615:  57%|█████▋    | 3694/6470 [45:22<29:27,  1.57it/s]
1|3695|Loss: 0.8301772475242615:  57%|█████▋    | 3695/6470 [45:22<29:08,  1.59it/s]
1|3696|Loss: 0.925250232219696:  57%|█████▋    | 3695/6470 [45:23<29:08,  1.59it/s] 
1|3696|Loss: 0.925250232219696:  57%|█████▋    | 3696/6470 [45:23<31:40,  1.46it/s]
1|3697|Loss: 0.8027312159538269:  57%|█████▋    | 3696/6470 [45:24<31:40,  1.46it/s]
1|3697|Loss: 0.8027312159538269:  57%|█████▋    | 3697/6470 [45:24<28:52,  1.60it/s]
1|3698|Loss: 0.8869165778160095:  57%|█████▋    | 3697/6470 [45:24<28:52,  1.60it/s]
1|3698|Loss: 0.8869165778160095:  57%|█████▋    | 3698/6470 [45:25<31:06,  1.49it/s]
1|3699|Loss: 0.8795955777168274:  57%|█████▋    | 3698/6470 [45:25<31:06,  1.49it/s]
1|3699|Loss: 0.8795955777168274:  57%|█████▋    | 3699/6470 [45:25<33:55,  1.36it/s]
1|3700|Loss: 1.0611788034439087:  57%|█████▋    | 3699/6470 [45:26<33:55,  1.36it/s]
1|3700|Loss: 1.0611788034439087:  57%|█████▋    | 3700/6470 [45:26<34:35,  1.33it/s]
1|3701|Loss: 0.8108915686607361:  57%|█████▋    | 3700/6470 [45:27<34:35,  1.33it/s]
1|3701|Loss: 0.8108915686607361:  57%|█████▋    | 3701/6470 [45:27<35:44,  1.29it/s]
1|3702|Loss: 0.9908190965652466:  57%|█████▋    | 3701/6470 [45:27<35:44,  1.29it/s]
1|3702|Loss: 0.9908190965652466:  57%|█████▋    | 3702/6470 [45:28<35:40,  1.29it/s]
1|3703|Loss: 0.8486939072608948:  57%|█████▋    | 3702/6470 [45:28<35:40,  1.29it/s]
1|3703|Loss: 0.8486939072608948:  57%|█████▋    | 3703/6470 [45:28<33:51,  1.36it/s]
1|3704|Loss: 0.8553395867347717:  57%|█████▋    | 3703/6470 [45:29<33:51,  1.36it/s]
1|3704|Loss: 0.8553395867347717:  57%|█████▋    | 3704/6470 [45:29<33:27,  1.38it/s]
1|3705|Loss: 0.9115790724754333:  57%|█████▋    | 3704/6470 [45:30<33:27,  1.38it/s]
1|3705|Loss: 0.9115790724754333:  57%|█████▋    | 3705/6470 [45:30<33:06,  1.39it/s]
1|3706|Loss: 0.8039748668670654:  57%|█████▋    | 3705/6470 [45:30<33:06,  1.39it/s]
1|3706|Loss: 0.8039748668670654:  57%|█████▋    | 3706/6470 [45:31<32:53,  1.40it/s]
1|3707|Loss: 0.6983423829078674:  57%|█████▋    | 3706/6470 [45:31<32:53,  1.40it/s]
1|3707|Loss: 0.6983423829078674:  57%|█████▋    | 3707/6470 [45:31<32:58,  1.40it/s]
1|3708|Loss: 0.9498188495635986:  57%|█████▋    | 3707/6470 [45:32<32:58,  1.40it/s]
1|3708|Loss: 0.9498188495635986:  57%|█████▋    | 3708/6470 [45:32<30:24,  1.51it/s]
1|3709|Loss: 0.8212975859642029:  57%|█████▋    | 3708/6470 [45:32<30:24,  1.51it/s]
1|3709|Loss: 0.8212975859642029:  57%|█████▋    | 3709/6470 [45:32<30:22,  1.51it/s]
1|3710|Loss: 0.843712329864502:  57%|█████▋    | 3709/6470 [45:33<30:22,  1.51it/s] 
1|3710|Loss: 0.843712329864502:  57%|█████▋    | 3710/6470 [45:33<32:29,  1.42it/s]
1|3711|Loss: 0.7845388054847717:  57%|█████▋    | 3710/6470 [45:34<32:29,  1.42it/s]
1|3711|Loss: 0.7845388054847717:  57%|█████▋    | 3711/6470 [45:34<30:04,  1.53it/s]
1|3712|Loss: 0.854165256023407:  57%|█████▋    | 3711/6470 [45:34<30:04,  1.53it/s] 
1|3712|Loss: 0.854165256023407:  57%|█████▋    | 3712/6470 [45:34<29:08,  1.58it/s]
1|3713|Loss: 0.8795082569122314:  57%|█████▋    | 3712/6470 [45:35<29:08,  1.58it/s]
1|3713|Loss: 0.8795082569122314:  57%|█████▋    | 3713/6470 [45:35<30:18,  1.52it/s]
1|3714|Loss: 1.0137494802474976:  57%|█████▋    | 3713/6470 [45:36<30:18,  1.52it/s]
1|3714|Loss: 1.0137494802474976:  57%|█████▋    | 3714/6470 [45:36<31:30,  1.46it/s]
1|3715|Loss: 0.8593785166740417:  57%|█████▋    | 3714/6470 [45:36<31:30,  1.46it/s]
1|3715|Loss: 0.8593785166740417:  57%|█████▋    | 3715/6470 [45:36<30:55,  1.49it/s]
1|3716|Loss: 0.9021005630493164:  57%|█████▋    | 3715/6470 [45:37<30:55,  1.49it/s]
1|3716|Loss: 0.9021005630493164:  57%|█████▋    | 3716/6470 [45:37<31:38,  1.45it/s]
1|3717|Loss: 0.8971322178840637:  57%|█████▋    | 3716/6470 [45:38<31:38,  1.45it/s]
1|3717|Loss: 0.8971322178840637:  57%|█████▋    | 3717/6470 [45:38<32:02,  1.43it/s]
1|3718|Loss: 0.9863815307617188:  57%|█████▋    | 3717/6470 [45:38<32:02,  1.43it/s]
1|3718|Loss: 0.9863815307617188:  57%|█████▋    | 3718/6470 [45:39<34:02,  1.35it/s]
1|3719|Loss: 0.7935333251953125:  57%|█████▋    | 3718/6470 [45:39<34:02,  1.35it/s]
1|3719|Loss: 0.7935333251953125:  57%|█████▋    | 3719/6470 [45:40<35:03,  1.31it/s]
1|3720|Loss: 0.9484351873397827:  57%|█████▋    | 3719/6470 [45:40<35:03,  1.31it/s]
1|3720|Loss: 0.9484351873397827:  57%|█████▋    | 3720/6470 [45:40<33:46,  1.36it/s]
1|3721|Loss: 0.7996184825897217:  57%|█████▋    | 3720/6470 [45:41<33:46,  1.36it/s]
1|3721|Loss: 0.7996184825897217:  58%|█████▊    | 3721/6470 [45:41<34:44,  1.32it/s]
1|3722|Loss: 0.7849264740943909:  58%|█████▊    | 3721/6470 [45:42<34:44,  1.32it/s]
1|3722|Loss: 0.7849264740943909:  58%|█████▊    | 3722/6470 [45:42<36:02,  1.27it/s]
1|3723|Loss: 0.923437237739563:  58%|█████▊    | 3722/6470 [45:42<36:02,  1.27it/s] 
1|3723|Loss: 0.923437237739563:  58%|█████▊    | 3723/6470 [45:43<35:23,  1.29it/s]
1|3724|Loss: 0.9488022923469543:  58%|█████▊    | 3723/6470 [45:43<35:23,  1.29it/s]
1|3724|Loss: 0.9488022923469543:  58%|█████▊    | 3724/6470 [45:43<34:22,  1.33it/s]
1|3725|Loss: 0.8781086206436157:  58%|█████▊    | 3724/6470 [45:44<34:22,  1.33it/s]
1|3725|Loss: 0.8781086206436157:  58%|█████▊    | 3725/6470 [45:44<33:38,  1.36it/s]
1|3726|Loss: 0.7950924038887024:  58%|█████▊    | 3725/6470 [45:44<33:38,  1.36it/s]
1|3726|Loss: 0.7950924038887024:  58%|█████▊    | 3726/6470 [45:45<31:58,  1.43it/s]
1|3727|Loss: 0.8523725867271423:  58%|█████▊    | 3726/6470 [45:45<31:58,  1.43it/s]
1|3727|Loss: 0.8523725867271423:  58%|█████▊    | 3727/6470 [45:45<32:05,  1.42it/s]
1|3728|Loss: 1.0256075859069824:  58%|█████▊    | 3727/6470 [45:46<32:05,  1.42it/s]
1|3728|Loss: 1.0256075859069824:  58%|█████▊    | 3728/6470 [45:46<32:32,  1.40it/s]
1|3729|Loss: 0.8048174977302551:  58%|█████▊    | 3728/6470 [45:47<32:32,  1.40it/s]
1|3729|Loss: 0.8048174977302551:  58%|█████▊    | 3729/6470 [45:47<31:37,  1.44it/s]
1|3730|Loss: 0.8557772636413574:  58%|█████▊    | 3729/6470 [45:47<31:37,  1.44it/s]
1|3730|Loss: 0.8557772636413574:  58%|█████▊    | 3730/6470 [45:48<32:51,  1.39it/s]
1|3731|Loss: 1.0789694786071777:  58%|█████▊    | 3730/6470 [45:48<32:51,  1.39it/s]
1|3731|Loss: 1.0789694786071777:  58%|█████▊    | 3731/6470 [45:48<32:03,  1.42it/s]
1|3732|Loss: 0.9165163636207581:  58%|█████▊    | 3731/6470 [45:49<32:03,  1.42it/s]
1|3732|Loss: 0.9165163636207581:  58%|█████▊    | 3732/6470 [45:49<31:10,  1.46it/s]
1|3733|Loss: 0.8106212019920349:  58%|█████▊    | 3732/6470 [45:49<31:10,  1.46it/s]
1|3733|Loss: 0.8106212019920349:  58%|█████▊    | 3733/6470 [45:49<30:24,  1.50it/s]
1|3734|Loss: 1.0109376907348633:  58%|█████▊    | 3733/6470 [45:50<30:24,  1.50it/s]
1|3734|Loss: 1.0109376907348633:  58%|█████▊    | 3734/6470 [45:50<29:18,  1.56it/s]
1|3735|Loss: 0.8134093284606934:  58%|█████▊    | 3734/6470 [45:50<29:18,  1.56it/s]
1|3735|Loss: 0.8134093284606934:  58%|█████▊    | 3735/6470 [45:51<29:50,  1.53it/s]
1|3736|Loss: 1.0633981227874756:  58%|█████▊    | 3735/6470 [45:51<29:50,  1.53it/s]
1|3736|Loss: 1.0633981227874756:  58%|█████▊    | 3736/6470 [45:52<31:30,  1.45it/s]
1|3737|Loss: 0.8052517175674438:  58%|█████▊    | 3736/6470 [45:52<31:30,  1.45it/s]
1|3737|Loss: 0.8052517175674438:  58%|█████▊    | 3737/6470 [45:52<32:14,  1.41it/s]
1|3738|Loss: 0.7880652546882629:  58%|█████▊    | 3737/6470 [45:53<32:14,  1.41it/s]
1|3738|Loss: 0.7880652546882629:  58%|█████▊    | 3738/6470 [45:53<32:46,  1.39it/s]
1|3739|Loss: 0.8344447016716003:  58%|█████▊    | 3738/6470 [45:53<32:46,  1.39it/s]
1|3739|Loss: 0.8344447016716003:  58%|█████▊    | 3739/6470 [45:54<32:04,  1.42it/s]
1|3740|Loss: 0.9421525597572327:  58%|█████▊    | 3739/6470 [45:54<32:04,  1.42it/s]
1|3740|Loss: 0.9421525597572327:  58%|█████▊    | 3740/6470 [45:55<33:43,  1.35it/s]
1|3741|Loss: 0.9037227630615234:  58%|█████▊    | 3740/6470 [45:55<33:43,  1.35it/s]
1|3741|Loss: 0.9037227630615234:  58%|█████▊    | 3741/6470 [45:55<34:39,  1.31it/s]
1|3742|Loss: 0.8533662557601929:  58%|█████▊    | 3741/6470 [45:56<34:39,  1.31it/s]
1|3742|Loss: 0.8533662557601929:  58%|█████▊    | 3742/6470 [45:56<33:29,  1.36it/s]
1|3743|Loss: 0.8679550886154175:  58%|█████▊    | 3742/6470 [45:56<33:29,  1.36it/s]
1|3743|Loss: 0.8679550886154175:  58%|█████▊    | 3743/6470 [45:57<33:15,  1.37it/s]
1|3744|Loss: 0.9093260765075684:  58%|█████▊    | 3743/6470 [45:57<33:15,  1.37it/s]
1|3744|Loss: 0.9093260765075684:  58%|█████▊    | 3744/6470 [45:57<32:14,  1.41it/s]
1|3745|Loss: 0.9096535444259644:  58%|█████▊    | 3744/6470 [45:58<32:14,  1.41it/s]
1|3745|Loss: 0.9096535444259644:  58%|█████▊    | 3745/6470 [45:58<30:15,  1.50it/s]
1|3746|Loss: 0.9256559610366821:  58%|█████▊    | 3745/6470 [45:58<30:15,  1.50it/s]
1|3746|Loss: 0.9256559610366821:  58%|█████▊    | 3746/6470 [45:58<28:14,  1.61it/s]
1|3747|Loss: 0.7592844367027283:  58%|█████▊    | 3746/6470 [45:59<28:14,  1.61it/s]
1|3747|Loss: 0.7592844367027283:  58%|█████▊    | 3747/6470 [45:59<27:50,  1.63it/s]
1|3748|Loss: 0.8426797389984131:  58%|█████▊    | 3747/6470 [45:59<27:50,  1.63it/s]
1|3748|Loss: 0.8426797389984131:  58%|█████▊    | 3748/6470 [46:00<28:50,  1.57it/s]
1|3749|Loss: 0.8897483348846436:  58%|█████▊    | 3748/6470 [46:00<28:50,  1.57it/s]
1|3749|Loss: 0.8897483348846436:  58%|█████▊    | 3749/6470 [46:01<30:16,  1.50it/s]
1|3750|Loss: 1.009880542755127:  58%|█████▊    | 3749/6470 [46:01<30:16,  1.50it/s] 
1|3750|Loss: 1.009880542755127:  58%|█████▊    | 3750/6470 [46:01<31:50,  1.42it/s]
1|3751|Loss: 0.9918757081031799:  58%|█████▊    | 3750/6470 [46:02<31:50,  1.42it/s]
1|3751|Loss: 0.9918757081031799:  58%|█████▊    | 3751/6470 [46:02<30:11,  1.50it/s]
1|3752|Loss: 0.8949715495109558:  58%|█████▊    | 3751/6470 [46:02<30:11,  1.50it/s]
1|3752|Loss: 0.8949715495109558:  58%|█████▊    | 3752/6470 [46:03<31:17,  1.45it/s]
1|3753|Loss: 0.7148922681808472:  58%|█████▊    | 3752/6470 [46:03<31:17,  1.45it/s]
1|3753|Loss: 0.7148922681808472:  58%|█████▊    | 3753/6470 [46:04<34:03,  1.33it/s]
1|3754|Loss: 0.912680447101593:  58%|█████▊    | 3753/6470 [46:04<34:03,  1.33it/s] 
1|3754|Loss: 0.912680447101593:  58%|█████▊    | 3754/6470 [46:04<32:25,  1.40it/s]
1|3755|Loss: 0.8788236379623413:  58%|█████▊    | 3754/6470 [46:04<32:25,  1.40it/s]
1|3755|Loss: 0.8788236379623413:  58%|█████▊    | 3755/6470 [46:05<31:06,  1.45it/s]
1|3756|Loss: 0.9091513156890869:  58%|█████▊    | 3755/6470 [46:05<31:06,  1.45it/s]
1|3756|Loss: 0.9091513156890869:  58%|█████▊    | 3756/6470 [46:05<31:06,  1.45it/s]
1|3757|Loss: 0.8050546050071716:  58%|█████▊    | 3756/6470 [46:06<31:06,  1.45it/s]
1|3757|Loss: 0.8050546050071716:  58%|█████▊    | 3757/6470 [46:06<31:19,  1.44it/s]
1|3758|Loss: 0.8959360122680664:  58%|█████▊    | 3757/6470 [46:07<31:19,  1.44it/s]
1|3758|Loss: 0.8959360122680664:  58%|█████▊    | 3758/6470 [46:07<31:45,  1.42it/s]
1|3759|Loss: 0.917114794254303:  58%|█████▊    | 3758/6470 [46:07<31:45,  1.42it/s] 
1|3759|Loss: 0.917114794254303:  58%|█████▊    | 3759/6470 [46:07<30:23,  1.49it/s]
1|3760|Loss: 0.8735366463661194:  58%|█████▊    | 3759/6470 [46:08<30:23,  1.49it/s]
1|3760|Loss: 0.8735366463661194:  58%|█████▊    | 3760/6470 [46:08<30:29,  1.48it/s]
1|3761|Loss: 0.9859653115272522:  58%|█████▊    | 3760/6470 [46:09<30:29,  1.48it/s]
1|3761|Loss: 0.9859653115272522:  58%|█████▊    | 3761/6470 [46:09<30:50,  1.46it/s]
1|3762|Loss: 0.8738852143287659:  58%|█████▊    | 3761/6470 [46:09<30:50,  1.46it/s]
1|3762|Loss: 0.8738852143287659:  58%|█████▊    | 3762/6470 [46:10<30:07,  1.50it/s]
1|3763|Loss: 0.8630799651145935:  58%|█████▊    | 3762/6470 [46:10<30:07,  1.50it/s]
1|3763|Loss: 0.8630799651145935:  58%|█████▊    | 3763/6470 [46:10<30:44,  1.47it/s]
1|3764|Loss: 1.089378833770752:  58%|█████▊    | 3763/6470 [46:11<30:44,  1.47it/s] 
1|3764|Loss: 1.089378833770752:  58%|█████▊    | 3764/6470 [46:11<29:11,  1.55it/s]
1|3765|Loss: 0.6967682838439941:  58%|█████▊    | 3764/6470 [46:11<29:11,  1.55it/s]
1|3765|Loss: 0.6967682838439941:  58%|█████▊    | 3765/6470 [46:12<31:20,  1.44it/s]
1|3766|Loss: 0.8274701833724976:  58%|█████▊    | 3765/6470 [46:12<31:20,  1.44it/s]
1|3766|Loss: 0.8274701833724976:  58%|█████▊    | 3766/6470 [46:12<31:04,  1.45it/s]
1|3767|Loss: 0.760925829410553:  58%|█████▊    | 3766/6470 [46:13<31:04,  1.45it/s] 
1|3767|Loss: 0.760925829410553:  58%|█████▊    | 3767/6470 [46:13<32:47,  1.37it/s]
1|3768|Loss: 0.8330814838409424:  58%|█████▊    | 3767/6470 [46:13<32:47,  1.37it/s]
1|3768|Loss: 0.8330814838409424:  58%|█████▊    | 3768/6470 [46:14<32:14,  1.40it/s]
1|3769|Loss: 0.893369197845459:  58%|█████▊    | 3768/6470 [46:14<32:14,  1.40it/s] 
1|3769|Loss: 0.893369197845459:  58%|█████▊    | 3769/6470 [46:15<32:56,  1.37it/s]
1|3770|Loss: 0.8852440118789673:  58%|█████▊    | 3769/6470 [46:15<32:56,  1.37it/s]
1|3770|Loss: 0.8852440118789673:  58%|█████▊    | 3770/6470 [46:15<32:44,  1.37it/s]
1|3771|Loss: 0.9226450324058533:  58%|█████▊    | 3770/6470 [46:16<32:44,  1.37it/s]
1|3771|Loss: 0.9226450324058533:  58%|█████▊    | 3771/6470 [46:16<32:46,  1.37it/s]
1|3772|Loss: 0.8561658263206482:  58%|█████▊    | 3771/6470 [46:16<32:46,  1.37it/s]
1|3772|Loss: 0.8561658263206482:  58%|█████▊    | 3772/6470 [46:17<30:25,  1.48it/s]
1|3773|Loss: 0.8295060992240906:  58%|█████▊    | 3772/6470 [46:17<30:25,  1.48it/s]
1|3773|Loss: 0.8295060992240906:  58%|█████▊    | 3773/6470 [46:17<32:32,  1.38it/s]
1|3774|Loss: 0.8575029373168945:  58%|█████▊    | 3773/6470 [46:18<32:32,  1.38it/s]
1|3774|Loss: 0.8575029373168945:  58%|█████▊    | 3774/6470 [46:18<32:12,  1.40it/s]
1|3775|Loss: 0.8681609034538269:  58%|█████▊    | 3774/6470 [46:18<32:12,  1.40it/s]
1|3775|Loss: 0.8681609034538269:  58%|█████▊    | 3775/6470 [46:19<30:08,  1.49it/s]
1|3776|Loss: 0.9070623517036438:  58%|█████▊    | 3775/6470 [46:19<30:08,  1.49it/s]
1|3776|Loss: 0.9070623517036438:  58%|█████▊    | 3776/6470 [46:19<28:47,  1.56it/s]
1|3777|Loss: 0.8071367144584656:  58%|█████▊    | 3776/6470 [46:20<28:47,  1.56it/s]
1|3777|Loss: 0.8071367144584656:  58%|█████▊    | 3777/6470 [46:20<29:35,  1.52it/s]
1|3778|Loss: 0.9404288530349731:  58%|█████▊    | 3777/6470 [46:20<29:35,  1.52it/s]
1|3778|Loss: 0.9404288530349731:  58%|█████▊    | 3778/6470 [46:21<29:18,  1.53it/s]
1|3779|Loss: 0.7938590049743652:  58%|█████▊    | 3778/6470 [46:21<29:18,  1.53it/s]
1|3779|Loss: 0.7938590049743652:  58%|█████▊    | 3779/6470 [46:21<29:41,  1.51it/s]
1|3780|Loss: 0.9876236319541931:  58%|█████▊    | 3779/6470 [46:22<29:41,  1.51it/s]
1|3780|Loss: 0.9876236319541931:  58%|█████▊    | 3780/6470 [46:22<31:47,  1.41it/s]
1|3781|Loss: 0.7588275074958801:  58%|█████▊    | 3780/6470 [46:22<31:47,  1.41it/s]
1|3781|Loss: 0.7588275074958801:  58%|█████▊    | 3781/6470 [46:23<31:31,  1.42it/s]
1|3782|Loss: 0.8425610065460205:  58%|█████▊    | 3781/6470 [46:23<31:31,  1.42it/s]
1|3782|Loss: 0.8425610065460205:  58%|█████▊    | 3782/6470 [46:23<30:22,  1.47it/s]
1|3783|Loss: 0.7400299906730652:  58%|█████▊    | 3782/6470 [46:24<30:22,  1.47it/s]
1|3783|Loss: 0.7400299906730652:  58%|█████▊    | 3783/6470 [46:24<29:45,  1.51it/s]
1|3784|Loss: 0.7751744389533997:  58%|█████▊    | 3783/6470 [46:24<29:45,  1.51it/s]
1|3784|Loss: 0.7751744389533997:  58%|█████▊    | 3784/6470 [46:25<31:51,  1.40it/s]
1|3785|Loss: 0.8455842733383179:  58%|█████▊    | 3784/6470 [46:25<31:51,  1.40it/s]
1|3785|Loss: 0.8455842733383179:  59%|█████▊    | 3785/6470 [46:26<33:51,  1.32it/s]
1|3786|Loss: 0.8133766651153564:  59%|█████▊    | 3785/6470 [46:26<33:51,  1.32it/s]
1|3786|Loss: 0.8133766651153564:  59%|█████▊    | 3786/6470 [46:26<32:46,  1.36it/s]
1|3787|Loss: 0.9517566561698914:  59%|█████▊    | 3786/6470 [46:27<32:46,  1.36it/s]
1|3787|Loss: 0.9517566561698914:  59%|█████▊    | 3787/6470 [46:27<31:11,  1.43it/s]
1|3788|Loss: 0.9408981204032898:  59%|█████▊    | 3787/6470 [46:27<31:11,  1.43it/s]
1|3788|Loss: 0.9408981204032898:  59%|█████▊    | 3788/6470 [46:28<32:20,  1.38it/s]
1|3789|Loss: 0.9095321297645569:  59%|█████▊    | 3788/6470 [46:28<32:20,  1.38it/s]
1|3789|Loss: 0.9095321297645569:  59%|█████▊    | 3789/6470 [46:28<32:26,  1.38it/s]
1|3790|Loss: 0.973224937915802:  59%|█████▊    | 3789/6470 [46:29<32:26,  1.38it/s] 
1|3790|Loss: 0.973224937915802:  59%|█████▊    | 3790/6470 [46:29<33:48,  1.32it/s]
1|3791|Loss: 0.9615946412086487:  59%|█████▊    | 3790/6470 [46:30<33:48,  1.32it/s]
1|3791|Loss: 0.9615946412086487:  59%|█████▊    | 3791/6470 [46:30<33:32,  1.33it/s]
1|3792|Loss: 0.8727173209190369:  59%|█████▊    | 3791/6470 [46:30<33:32,  1.33it/s]
1|3792|Loss: 0.8727173209190369:  59%|█████▊    | 3792/6470 [46:31<32:53,  1.36it/s]
1|3793|Loss: 1.0380853414535522:  59%|█████▊    | 3792/6470 [46:31<32:53,  1.36it/s]
1|3793|Loss: 1.0380853414535522:  59%|█████▊    | 3793/6470 [46:32<34:03,  1.31it/s]
1|3794|Loss: 0.8566075563430786:  59%|█████▊    | 3793/6470 [46:32<34:03,  1.31it/s]
1|3794|Loss: 0.8566075563430786:  59%|█████▊    | 3794/6470 [46:32<35:02,  1.27it/s]
1|3795|Loss: 0.8185091614723206:  59%|█████▊    | 3794/6470 [46:33<35:02,  1.27it/s]
1|3795|Loss: 0.8185091614723206:  59%|█████▊    | 3795/6470 [46:33<32:24,  1.38it/s]
1|3796|Loss: 0.9820413589477539:  59%|█████▊    | 3795/6470 [46:33<32:24,  1.38it/s]
1|3796|Loss: 0.9820413589477539:  59%|█████▊    | 3796/6470 [46:34<30:22,  1.47it/s]
1|3797|Loss: 0.8422873020172119:  59%|█████▊    | 3796/6470 [46:34<30:22,  1.47it/s]
1|3797|Loss: 0.8422873020172119:  59%|█████▊    | 3797/6470 [46:34<28:23,  1.57it/s]
1|3798|Loss: 0.8673165440559387:  59%|█████▊    | 3797/6470 [46:34<28:23,  1.57it/s]
1|3798|Loss: 0.8673165440559387:  59%|█████▊    | 3798/6470 [46:35<26:52,  1.66it/s]
1|3799|Loss: 0.8403865694999695:  59%|█████▊    | 3798/6470 [46:35<26:52,  1.66it/s]
1|3799|Loss: 0.8403865694999695:  59%|█████▊    | 3799/6470 [46:35<27:05,  1.64it/s]
1|3800|Loss: 0.7639811038970947:  59%|█████▊    | 3799/6470 [46:36<27:05,  1.64it/s]
1|3800|Loss: 0.7639811038970947:  59%|█████▊    | 3800/6470 [46:36<24:35,  1.81it/s]
1|3801|Loss: 0.9233323335647583:  59%|█████▊    | 3800/6470 [46:36<24:35,  1.81it/s]
1|3801|Loss: 0.9233323335647583:  59%|█████▊    | 3801/6470 [46:36<25:47,  1.73it/s]
1|3802|Loss: 0.7576603293418884:  59%|█████▊    | 3801/6470 [46:37<25:47,  1.73it/s]
1|3802|Loss: 0.7576603293418884:  59%|█████▉    | 3802/6470 [46:37<28:43,  1.55it/s]
1|3803|Loss: 0.8772386312484741:  59%|█████▉    | 3802/6470 [46:38<28:43,  1.55it/s]
1|3803|Loss: 0.8772386312484741:  59%|█████▉    | 3803/6470 [46:38<28:28,  1.56it/s]
1|3804|Loss: 0.7380266189575195:  59%|█████▉    | 3803/6470 [46:38<28:28,  1.56it/s]
1|3804|Loss: 0.7380266189575195:  59%|█████▉    | 3804/6470 [46:38<27:38,  1.61it/s]
1|3805|Loss: 0.9211946129798889:  59%|█████▉    | 3804/6470 [46:39<27:38,  1.61it/s]
1|3805|Loss: 0.9211946129798889:  59%|█████▉    | 3805/6470 [46:39<29:29,  1.51it/s]
1|3806|Loss: 0.8995823264122009:  59%|█████▉    | 3805/6470 [46:40<29:29,  1.51it/s]
1|3806|Loss: 0.8995823264122009:  59%|█████▉    | 3806/6470 [46:40<31:21,  1.42it/s]
1|3807|Loss: 0.8732395172119141:  59%|█████▉    | 3806/6470 [46:40<31:21,  1.42it/s]
1|3807|Loss: 0.8732395172119141:  59%|█████▉    | 3807/6470 [46:41<31:54,  1.39it/s]
1|3808|Loss: 0.9371775388717651:  59%|█████▉    | 3807/6470 [46:41<31:54,  1.39it/s]
1|3808|Loss: 0.9371775388717651:  59%|█████▉    | 3808/6470 [46:41<30:55,  1.43it/s]
1|3809|Loss: 0.8974522352218628:  59%|█████▉    | 3808/6470 [46:42<30:55,  1.43it/s]
1|3809|Loss: 0.8974522352218628:  59%|█████▉    | 3809/6470 [46:42<29:22,  1.51it/s]
1|3810|Loss: 0.9410993456840515:  59%|█████▉    | 3809/6470 [46:42<29:22,  1.51it/s]
1|3810|Loss: 0.9410993456840515:  59%|█████▉    | 3810/6470 [46:43<29:54,  1.48it/s]
1|3811|Loss: 0.9893743991851807:  59%|█████▉    | 3810/6470 [46:43<29:54,  1.48it/s]
1|3811|Loss: 0.9893743991851807:  59%|█████▉    | 3811/6470 [46:43<30:52,  1.44it/s]
1|3812|Loss: 0.8419321179389954:  59%|█████▉    | 3811/6470 [46:44<30:52,  1.44it/s]
1|3812|Loss: 0.8419321179389954:  59%|█████▉    | 3812/6470 [46:44<28:48,  1.54it/s]
1|3813|Loss: 0.8329582810401917:  59%|█████▉    | 3812/6470 [46:44<28:48,  1.54it/s]
1|3813|Loss: 0.8329582810401917:  59%|█████▉    | 3813/6470 [46:45<30:35,  1.45it/s]
1|3814|Loss: 0.822844922542572:  59%|█████▉    | 3813/6470 [46:45<30:35,  1.45it/s] 
1|3814|Loss: 0.822844922542572:  59%|█████▉    | 3814/6470 [46:45<30:36,  1.45it/s]
1|3815|Loss: 1.0043631792068481:  59%|█████▉    | 3814/6470 [46:46<30:36,  1.45it/s]
1|3815|Loss: 1.0043631792068481:  59%|█████▉    | 3815/6470 [46:46<32:48,  1.35it/s]
1|3816|Loss: 0.9047755002975464:  59%|█████▉    | 3815/6470 [46:47<32:48,  1.35it/s]
1|3816|Loss: 0.9047755002975464:  59%|█████▉    | 3816/6470 [46:47<31:32,  1.40it/s]
1|3817|Loss: 0.952875018119812:  59%|█████▉    | 3816/6470 [46:47<31:32,  1.40it/s] 
1|3817|Loss: 0.952875018119812:  59%|█████▉    | 3817/6470 [46:48<32:50,  1.35it/s]
1|3818|Loss: 0.9748955368995667:  59%|█████▉    | 3817/6470 [46:48<32:50,  1.35it/s]
1|3818|Loss: 0.9748955368995667:  59%|█████▉    | 3818/6470 [46:48<33:05,  1.34it/s]
1|3819|Loss: 0.7538638710975647:  59%|█████▉    | 3818/6470 [46:49<33:05,  1.34it/s]
1|3819|Loss: 0.7538638710975647:  59%|█████▉    | 3819/6470 [46:49<30:39,  1.44it/s]
1|3820|Loss: 0.8426478505134583:  59%|█████▉    | 3819/6470 [46:49<30:39,  1.44it/s]
1|3820|Loss: 0.8426478505134583:  59%|█████▉    | 3820/6470 [46:49<27:37,  1.60it/s]
1|3821|Loss: 0.9624121785163879:  59%|█████▉    | 3820/6470 [46:50<27:37,  1.60it/s]
1|3821|Loss: 0.9624121785163879:  59%|█████▉    | 3821/6470 [46:50<28:13,  1.56it/s]
1|3822|Loss: 0.8580768704414368:  59%|█████▉    | 3821/6470 [46:51<28:13,  1.56it/s]
1|3822|Loss: 0.8580768704414368:  59%|█████▉    | 3822/6470 [46:51<29:14,  1.51it/s]
1|3823|Loss: 0.7719005346298218:  59%|█████▉    | 3822/6470 [46:51<29:14,  1.51it/s]
1|3823|Loss: 0.7719005346298218:  59%|█████▉    | 3823/6470 [46:52<29:44,  1.48it/s]
1|3824|Loss: 0.8459607362747192:  59%|█████▉    | 3823/6470 [46:52<29:44,  1.48it/s]
1|3824|Loss: 0.8459607362747192:  59%|█████▉    | 3824/6470 [46:52<28:01,  1.57it/s]
1|3825|Loss: 0.8737339377403259:  59%|█████▉    | 3824/6470 [46:52<28:01,  1.57it/s]
1|3825|Loss: 0.8737339377403259:  59%|█████▉    | 3825/6470 [46:53<27:29,  1.60it/s]
1|3826|Loss: 0.7840123176574707:  59%|█████▉    | 3825/6470 [46:53<27:29,  1.60it/s]
1|3826|Loss: 0.7840123176574707:  59%|█████▉    | 3826/6470 [46:53<26:58,  1.63it/s]
1|3827|Loss: 0.8768143653869629:  59%|█████▉    | 3826/6470 [46:54<26:58,  1.63it/s]
1|3827|Loss: 0.8768143653869629:  59%|█████▉    | 3827/6470 [46:54<27:15,  1.62it/s]
1|3828|Loss: 0.8154478073120117:  59%|█████▉    | 3827/6470 [46:54<27:15,  1.62it/s]
1|3828|Loss: 0.8154478073120117:  59%|█████▉    | 3828/6470 [46:55<28:20,  1.55it/s]
1|3829|Loss: 0.9149174094200134:  59%|█████▉    | 3828/6470 [46:55<28:20,  1.55it/s]
1|3829|Loss: 0.9149174094200134:  59%|█████▉    | 3829/6470 [46:55<26:28,  1.66it/s]
1|3830|Loss: 0.8172932863235474:  59%|█████▉    | 3829/6470 [46:55<26:28,  1.66it/s]
1|3830|Loss: 0.8172932863235474:  59%|█████▉    | 3830/6470 [46:56<26:26,  1.66it/s]
1|3831|Loss: 0.8828369975090027:  59%|█████▉    | 3830/6470 [46:56<26:26,  1.66it/s]
1|3831|Loss: 0.8828369975090027:  59%|█████▉    | 3831/6470 [46:56<27:39,  1.59it/s]
1|3832|Loss: 0.7600136399269104:  59%|█████▉    | 3831/6470 [46:57<27:39,  1.59it/s]
1|3832|Loss: 0.7600136399269104:  59%|█████▉    | 3832/6470 [46:57<30:49,  1.43it/s]
1|3833|Loss: 0.870713472366333:  59%|█████▉    | 3832/6470 [46:58<30:49,  1.43it/s] 
1|3833|Loss: 0.870713472366333:  59%|█████▉    | 3833/6470 [46:58<29:10,  1.51it/s]
1|3834|Loss: 1.0856152772903442:  59%|█████▉    | 3833/6470 [46:58<29:10,  1.51it/s]
1|3834|Loss: 1.0856152772903442:  59%|█████▉    | 3834/6470 [46:58<26:45,  1.64it/s]
1|3835|Loss: 0.9256187081336975:  59%|█████▉    | 3834/6470 [46:59<26:45,  1.64it/s]
1|3835|Loss: 0.9256187081336975:  59%|█████▉    | 3835/6470 [46:59<25:06,  1.75it/s]
1|3836|Loss: 0.8298324942588806:  59%|█████▉    | 3835/6470 [46:59<25:06,  1.75it/s]
1|3836|Loss: 0.8298324942588806:  59%|█████▉    | 3836/6470 [46:59<25:29,  1.72it/s]
1|3837|Loss: 0.9056946635246277:  59%|█████▉    | 3836/6470 [47:00<25:29,  1.72it/s]
1|3837|Loss: 0.9056946635246277:  59%|█████▉    | 3837/6470 [47:00<27:57,  1.57it/s]
1|3838|Loss: 0.7999029159545898:  59%|█████▉    | 3837/6470 [47:01<27:57,  1.57it/s]
1|3838|Loss: 0.7999029159545898:  59%|█████▉    | 3838/6470 [47:01<29:38,  1.48it/s]
1|3839|Loss: 1.0350940227508545:  59%|█████▉    | 3838/6470 [47:01<29:38,  1.48it/s]
1|3839|Loss: 1.0350940227508545:  59%|█████▉    | 3839/6470 [47:02<30:39,  1.43it/s]
1|3840|Loss: 0.8896244168281555:  59%|█████▉    | 3839/6470 [47:02<30:39,  1.43it/s]
1|3840|Loss: 0.8896244168281555:  59%|█████▉    | 3840/6470 [47:02<31:09,  1.41it/s]
1|3841|Loss: 0.9231296181678772:  59%|█████▉    | 3840/6470 [47:03<31:09,  1.41it/s]
1|3841|Loss: 0.9231296181678772:  59%|█████▉    | 3841/6470 [47:03<32:10,  1.36it/s]
1|3842|Loss: 0.8962142467498779:  59%|█████▉    | 3841/6470 [47:04<32:10,  1.36it/s]
1|3842|Loss: 0.8962142467498779:  59%|█████▉    | 3842/6470 [47:04<32:30,  1.35it/s]
1|3843|Loss: 0.7816992402076721:  59%|█████▉    | 3842/6470 [47:04<32:30,  1.35it/s]
1|3843|Loss: 0.7816992402076721:  59%|█████▉    | 3843/6470 [47:05<33:36,  1.30it/s]
1|3844|Loss: 1.1066945791244507:  59%|█████▉    | 3843/6470 [47:05<33:36,  1.30it/s]
1|3844|Loss: 1.1066945791244507:  59%|█████▉    | 3844/6470 [47:05<32:15,  1.36it/s]
1|3845|Loss: 0.9341320991516113:  59%|█████▉    | 3844/6470 [47:06<32:15,  1.36it/s]
1|3845|Loss: 0.9341320991516113:  59%|█████▉    | 3845/6470 [47:06<30:54,  1.42it/s]
1|3846|Loss: 0.8090534806251526:  59%|█████▉    | 3845/6470 [47:07<30:54,  1.42it/s]
1|3846|Loss: 0.8090534806251526:  59%|█████▉    | 3846/6470 [47:07<32:36,  1.34it/s]
1|3847|Loss: 0.9534263610839844:  59%|█████▉    | 3846/6470 [47:07<32:36,  1.34it/s]
1|3847|Loss: 0.9534263610839844:  59%|█████▉    | 3847/6470 [47:08<32:38,  1.34it/s]
1|3848|Loss: 0.790387749671936:  59%|█████▉    | 3847/6470 [47:08<32:38,  1.34it/s] 
1|3848|Loss: 0.790387749671936:  59%|█████▉    | 3848/6470 [47:08<29:58,  1.46it/s]
1|3849|Loss: 0.9535994529724121:  59%|█████▉    | 3848/6470 [47:09<29:58,  1.46it/s]
1|3849|Loss: 0.9535994529724121:  59%|█████▉    | 3849/6470 [47:09<30:49,  1.42it/s]
1|3850|Loss: 0.6691240072250366:  59%|█████▉    | 3849/6470 [47:10<30:49,  1.42it/s]
1|3850|Loss: 0.6691240072250366:  60%|█████▉    | 3850/6470 [47:10<33:50,  1.29it/s]
1|3851|Loss: 0.7594085335731506:  60%|█████▉    | 3850/6470 [47:10<33:50,  1.29it/s]
1|3851|Loss: 0.7594085335731506:  60%|█████▉    | 3851/6470 [47:11<34:10,  1.28it/s]
1|3852|Loss: 0.9973189830780029:  60%|█████▉    | 3851/6470 [47:11<34:10,  1.28it/s]
1|3852|Loss: 0.9973189830780029:  60%|█████▉    | 3852/6470 [47:11<32:46,  1.33it/s]
1|3853|Loss: 0.7898541688919067:  60%|█████▉    | 3852/6470 [47:12<32:46,  1.33it/s]
1|3853|Loss: 0.7898541688919067:  60%|█████▉    | 3853/6470 [47:12<32:31,  1.34it/s]
1|3854|Loss: 0.8846249580383301:  60%|█████▉    | 3853/6470 [47:13<32:31,  1.34it/s]
1|3854|Loss: 0.8846249580383301:  60%|█████▉    | 3854/6470 [47:13<31:29,  1.38it/s]
1|3855|Loss: 0.8892422914505005:  60%|█████▉    | 3854/6470 [47:13<31:29,  1.38it/s]
1|3855|Loss: 0.8892422914505005:  60%|█████▉    | 3855/6470 [47:14<31:20,  1.39it/s]
1|3856|Loss: 0.8214651942253113:  60%|█████▉    | 3855/6470 [47:14<31:20,  1.39it/s]
1|3856|Loss: 0.8214651942253113:  60%|█████▉    | 3856/6470 [47:14<29:07,  1.50it/s]
1|3857|Loss: 0.8415430188179016:  60%|█████▉    | 3856/6470 [47:14<29:07,  1.50it/s]
1|3857|Loss: 0.8415430188179016:  60%|█████▉    | 3857/6470 [47:15<28:24,  1.53it/s]
1|3858|Loss: 0.8859326243400574:  60%|█████▉    | 3857/6470 [47:15<28:24,  1.53it/s]
1|3858|Loss: 0.8859326243400574:  60%|█████▉    | 3858/6470 [47:15<27:31,  1.58it/s]
1|3859|Loss: 0.8582760095596313:  60%|█████▉    | 3858/6470 [47:16<27:31,  1.58it/s]
1|3859|Loss: 0.8582760095596313:  60%|█████▉    | 3859/6470 [47:16<29:47,  1.46it/s]
1|3860|Loss: 0.9704095721244812:  60%|█████▉    | 3859/6470 [47:17<29:47,  1.46it/s]
1|3860|Loss: 0.9704095721244812:  60%|█████▉    | 3860/6470 [47:17<29:19,  1.48it/s]
1|3861|Loss: 0.848315954208374:  60%|█████▉    | 3860/6470 [47:17<29:19,  1.48it/s] 
1|3861|Loss: 0.848315954208374:  60%|█████▉    | 3861/6470 [47:17<28:27,  1.53it/s]
1|3862|Loss: 0.8797740936279297:  60%|█████▉    | 3861/6470 [47:18<28:27,  1.53it/s]
1|3862|Loss: 0.8797740936279297:  60%|█████▉    | 3862/6470 [47:18<29:17,  1.48it/s]
1|3863|Loss: 0.9112275838851929:  60%|█████▉    | 3862/6470 [47:18<29:17,  1.48it/s]
1|3863|Loss: 0.9112275838851929:  60%|█████▉    | 3863/6470 [47:19<30:06,  1.44it/s]
1|3864|Loss: 0.9545789957046509:  60%|█████▉    | 3863/6470 [47:19<30:06,  1.44it/s]
1|3864|Loss: 0.9545789957046509:  60%|█████▉    | 3864/6470 [47:20<30:51,  1.41it/s]
1|3865|Loss: 1.0710206031799316:  60%|█████▉    | 3864/6470 [47:20<30:51,  1.41it/s]
1|3865|Loss: 1.0710206031799316:  60%|█████▉    | 3865/6470 [47:20<30:54,  1.40it/s]
1|3866|Loss: 0.9487298727035522:  60%|█████▉    | 3865/6470 [47:21<30:54,  1.40it/s]
1|3866|Loss: 0.9487298727035522:  60%|█████▉    | 3866/6470 [47:21<29:38,  1.46it/s]
1|3867|Loss: 0.9220279455184937:  60%|█████▉    | 3866/6470 [47:21<29:38,  1.46it/s]
1|3867|Loss: 0.9220279455184937:  60%|█████▉    | 3867/6470 [47:22<30:57,  1.40it/s]
1|3868|Loss: 0.900169849395752:  60%|█████▉    | 3867/6470 [47:22<30:57,  1.40it/s] 
1|3868|Loss: 0.900169849395752:  60%|█████▉    | 3868/6470 [47:22<29:36,  1.46it/s]
1|3869|Loss: 0.9657220840454102:  60%|█████▉    | 3868/6470 [47:23<29:36,  1.46it/s]
1|3869|Loss: 0.9657220840454102:  60%|█████▉    | 3869/6470 [47:23<27:28,  1.58it/s]
1|3870|Loss: 0.9416596293449402:  60%|█████▉    | 3869/6470 [47:23<27:28,  1.58it/s]
1|3870|Loss: 0.9416596293449402:  60%|█████▉    | 3870/6470 [47:23<27:13,  1.59it/s]
1|3871|Loss: 1.094573736190796:  60%|█████▉    | 3870/6470 [47:24<27:13,  1.59it/s] 
1|3871|Loss: 1.094573736190796:  60%|█████▉    | 3871/6470 [47:24<26:46,  1.62it/s]
1|3872|Loss: 1.000959873199463:  60%|█████▉    | 3871/6470 [47:24<26:46,  1.62it/s]
1|3872|Loss: 1.000959873199463:  60%|█████▉    | 3872/6470 [47:25<27:50,  1.55it/s]
1|3873|Loss: 0.9093652963638306:  60%|█████▉    | 3872/6470 [47:25<27:50,  1.55it/s]
1|3873|Loss: 0.9093652963638306:  60%|█████▉    | 3873/6470 [47:25<28:39,  1.51it/s]
1|3874|Loss: 0.9439826011657715:  60%|█████▉    | 3873/6470 [47:26<28:39,  1.51it/s]
1|3874|Loss: 0.9439826011657715:  60%|█████▉    | 3874/6470 [47:26<29:10,  1.48it/s]
1|3875|Loss: 0.8545646667480469:  60%|█████▉    | 3874/6470 [47:27<29:10,  1.48it/s]
1|3875|Loss: 0.8545646667480469:  60%|█████▉    | 3875/6470 [47:27<29:47,  1.45it/s]
1|3876|Loss: 1.0002318620681763:  60%|█████▉    | 3875/6470 [47:27<29:47,  1.45it/s]
1|3876|Loss: 1.0002318620681763:  60%|█████▉    | 3876/6470 [47:28<30:34,  1.41it/s]
1|3877|Loss: 1.0382944345474243:  60%|█████▉    | 3876/6470 [47:28<30:34,  1.41it/s]
1|3877|Loss: 1.0382944345474243:  60%|█████▉    | 3877/6470 [47:28<31:07,  1.39it/s]
1|3878|Loss: 0.8810660243034363:  60%|█████▉    | 3877/6470 [47:29<31:07,  1.39it/s]
1|3878|Loss: 0.8810660243034363:  60%|█████▉    | 3878/6470 [47:29<30:18,  1.43it/s]
1|3879|Loss: 0.9212263822555542:  60%|█████▉    | 3878/6470 [47:29<30:18,  1.43it/s]
1|3879|Loss: 0.9212263822555542:  60%|█████▉    | 3879/6470 [47:30<29:12,  1.48it/s]
1|3880|Loss: 0.9045220017433167:  60%|█████▉    | 3879/6470 [47:30<29:12,  1.48it/s]
1|3880|Loss: 0.9045220017433167:  60%|█████▉    | 3880/6470 [47:30<29:38,  1.46it/s]
1|3881|Loss: 0.8430073261260986:  60%|█████▉    | 3880/6470 [47:31<29:38,  1.46it/s]
1|3881|Loss: 0.8430073261260986:  60%|█████▉    | 3881/6470 [47:31<30:12,  1.43it/s]
1|3882|Loss: 0.77305668592453:  60%|█████▉    | 3881/6470 [47:31<30:12,  1.43it/s]  
1|3882|Loss: 0.77305668592453:  60%|██████    | 3882/6470 [47:32<29:39,  1.45it/s]
1|3883|Loss: 0.8119856715202332:  60%|██████    | 3882/6470 [47:32<29:39,  1.45it/s]
1|3883|Loss: 0.8119856715202332:  60%|██████    | 3883/6470 [47:32<27:50,  1.55it/s]
1|3884|Loss: 0.9957050085067749:  60%|██████    | 3883/6470 [47:33<27:50,  1.55it/s]
1|3884|Loss: 0.9957050085067749:  60%|██████    | 3884/6470 [47:33<25:33,  1.69it/s]
1|3885|Loss: 0.8935243487358093:  60%|██████    | 3884/6470 [47:33<25:33,  1.69it/s]
1|3885|Loss: 0.8935243487358093:  60%|██████    | 3885/6470 [47:33<27:29,  1.57it/s]
1|3886|Loss: 0.9882535338401794:  60%|██████    | 3885/6470 [47:34<27:29,  1.57it/s]
1|3886|Loss: 0.9882535338401794:  60%|██████    | 3886/6470 [47:34<28:04,  1.53it/s]
1|3887|Loss: 0.7839874625205994:  60%|██████    | 3886/6470 [47:35<28:04,  1.53it/s]
1|3887|Loss: 0.7839874625205994:  60%|██████    | 3887/6470 [47:35<28:29,  1.51it/s]
1|3888|Loss: 1.0150656700134277:  60%|██████    | 3887/6470 [47:35<28:29,  1.51it/s]
1|3888|Loss: 1.0150656700134277:  60%|██████    | 3888/6470 [47:36<28:22,  1.52it/s]
1|3889|Loss: 0.9121130108833313:  60%|██████    | 3888/6470 [47:36<28:22,  1.52it/s]
1|3889|Loss: 0.9121130108833313:  60%|██████    | 3889/6470 [47:36<29:07,  1.48it/s]
1|3890|Loss: 0.897940993309021:  60%|██████    | 3889/6470 [47:37<29:07,  1.48it/s] 
1|3890|Loss: 0.897940993309021:  60%|██████    | 3890/6470 [47:37<27:57,  1.54it/s]
1|3891|Loss: 0.7446257472038269:  60%|██████    | 3890/6470 [47:37<27:57,  1.54it/s]
1|3891|Loss: 0.7446257472038269:  60%|██████    | 3891/6470 [47:37<27:20,  1.57it/s]
1|3892|Loss: 0.6352391839027405:  60%|██████    | 3891/6470 [47:38<27:20,  1.57it/s]
1|3892|Loss: 0.6352391839027405:  60%|██████    | 3892/6470 [47:38<29:16,  1.47it/s]
1|3893|Loss: 0.8524669408798218:  60%|██████    | 3892/6470 [47:39<29:16,  1.47it/s]
1|3893|Loss: 0.8524669408798218:  60%|██████    | 3893/6470 [47:39<30:58,  1.39it/s]
1|3894|Loss: 0.7870229482650757:  60%|██████    | 3893/6470 [47:39<30:58,  1.39it/s]
1|3894|Loss: 0.7870229482650757:  60%|██████    | 3894/6470 [47:40<31:32,  1.36it/s]
1|3895|Loss: 0.8747420310974121:  60%|██████    | 3894/6470 [47:40<31:32,  1.36it/s]
1|3895|Loss: 0.8747420310974121:  60%|██████    | 3895/6470 [47:41<31:14,  1.37it/s]
1|3896|Loss: 0.8295777440071106:  60%|██████    | 3895/6470 [47:41<31:14,  1.37it/s]
1|3896|Loss: 0.8295777440071106:  60%|██████    | 3896/6470 [47:41<28:20,  1.51it/s]
1|3897|Loss: 0.7727958559989929:  60%|██████    | 3896/6470 [47:41<28:20,  1.51it/s]
1|3897|Loss: 0.7727958559989929:  60%|██████    | 3897/6470 [47:42<26:27,  1.62it/s]
1|3898|Loss: 0.9678821563720703:  60%|██████    | 3897/6470 [47:42<26:27,  1.62it/s]
1|3898|Loss: 0.9678821563720703:  60%|██████    | 3898/6470 [47:42<27:30,  1.56it/s]
1|3899|Loss: 0.8208540081977844:  60%|██████    | 3898/6470 [47:43<27:30,  1.56it/s]
1|3899|Loss: 0.8208540081977844:  60%|██████    | 3899/6470 [47:43<29:15,  1.46it/s]
1|3900|Loss: 0.9400943517684937:  60%|██████    | 3899/6470 [47:43<29:15,  1.46it/s]
1|3900|Loss: 0.9400943517684937:  60%|██████    | 3900/6470 [47:44<29:48,  1.44it/s]
1|3901|Loss: 0.9406323432922363:  60%|██████    | 3900/6470 [47:44<29:48,  1.44it/s]
1|3901|Loss: 0.9406323432922363:  60%|██████    | 3901/6470 [47:44<30:13,  1.42it/s]
1|3902|Loss: 0.9408136606216431:  60%|██████    | 3901/6470 [47:45<30:13,  1.42it/s]
1|3902|Loss: 0.9408136606216431:  60%|██████    | 3902/6470 [47:45<30:13,  1.42it/s]
1|3903|Loss: 0.7494222521781921:  60%|██████    | 3902/6470 [47:46<30:13,  1.42it/s]
1|3903|Loss: 0.7494222521781921:  60%|██████    | 3903/6470 [47:46<30:34,  1.40it/s]
1|3904|Loss: 0.9847739338874817:  60%|██████    | 3903/6470 [47:46<30:34,  1.40it/s]
1|3904|Loss: 0.9847739338874817:  60%|██████    | 3904/6470 [47:47<31:53,  1.34it/s]
1|3905|Loss: 0.719638466835022:  60%|██████    | 3904/6470 [47:47<31:53,  1.34it/s] 
1|3905|Loss: 0.719638466835022:  60%|██████    | 3905/6470 [47:47<30:07,  1.42it/s]
1|3906|Loss: 0.9535451531410217:  60%|██████    | 3905/6470 [47:48<30:07,  1.42it/s]
1|3906|Loss: 0.9535451531410217:  60%|██████    | 3906/6470 [47:48<29:06,  1.47it/s]
1|3907|Loss: 0.94905024766922:  60%|██████    | 3906/6470 [47:48<29:06,  1.47it/s]  
1|3907|Loss: 0.94905024766922:  60%|██████    | 3907/6470 [47:49<28:22,  1.51it/s]
1|3908|Loss: 1.072616696357727:  60%|██████    | 3907/6470 [47:49<28:22,  1.51it/s]
1|3908|Loss: 1.072616696357727:  60%|██████    | 3908/6470 [47:49<28:53,  1.48it/s]
1|3909|Loss: 0.8480759859085083:  60%|██████    | 3908/6470 [47:50<28:53,  1.48it/s]
1|3909|Loss: 0.8480759859085083:  60%|██████    | 3909/6470 [47:50<28:43,  1.49it/s]
1|3910|Loss: 0.9414544701576233:  60%|██████    | 3909/6470 [47:50<28:43,  1.49it/s]
1|3910|Loss: 0.9414544701576233:  60%|██████    | 3910/6470 [47:51<29:12,  1.46it/s]
1|3911|Loss: 0.8683910965919495:  60%|██████    | 3910/6470 [47:51<29:12,  1.46it/s]
1|3911|Loss: 0.8683910965919495:  60%|██████    | 3911/6470 [47:51<27:28,  1.55it/s]
1|3912|Loss: 0.9398415684700012:  60%|██████    | 3911/6470 [47:52<27:28,  1.55it/s]
1|3912|Loss: 0.9398415684700012:  60%|██████    | 3912/6470 [47:52<26:23,  1.62it/s]
1|3913|Loss: 0.9046583771705627:  60%|██████    | 3912/6470 [47:52<26:23,  1.62it/s]
1|3913|Loss: 0.9046583771705627:  60%|██████    | 3913/6470 [47:52<27:28,  1.55it/s]
1|3914|Loss: 0.7768775820732117:  60%|██████    | 3913/6470 [47:53<27:28,  1.55it/s]
1|3914|Loss: 0.7768775820732117:  60%|██████    | 3914/6470 [47:53<29:21,  1.45it/s]
1|3915|Loss: 0.8466134071350098:  60%|██████    | 3914/6470 [47:54<29:21,  1.45it/s]
1|3915|Loss: 0.8466134071350098:  61%|██████    | 3915/6470 [47:54<31:03,  1.37it/s]
1|3916|Loss: 0.9293243885040283:  61%|██████    | 3915/6470 [47:54<31:03,  1.37it/s]
1|3916|Loss: 0.9293243885040283:  61%|██████    | 3916/6470 [47:55<29:31,  1.44it/s]
1|3917|Loss: 0.8603810667991638:  61%|██████    | 3916/6470 [47:55<29:31,  1.44it/s]
1|3917|Loss: 0.8603810667991638:  61%|██████    | 3917/6470 [47:55<28:36,  1.49it/s]
1|3918|Loss: 0.9768601059913635:  61%|██████    | 3917/6470 [47:56<28:36,  1.49it/s]
1|3918|Loss: 0.9768601059913635:  61%|██████    | 3918/6470 [47:56<29:09,  1.46it/s]
1|3919|Loss: 0.8936546444892883:  61%|██████    | 3918/6470 [47:56<29:09,  1.46it/s]
1|3919|Loss: 0.8936546444892883:  61%|██████    | 3919/6470 [47:57<29:41,  1.43it/s]
1|3920|Loss: 0.8355370163917542:  61%|██████    | 3919/6470 [47:57<29:41,  1.43it/s]
1|3920|Loss: 0.8355370163917542:  61%|██████    | 3920/6470 [47:58<31:08,  1.36it/s]
1|3921|Loss: 0.8676407933235168:  61%|██████    | 3920/6470 [47:58<31:08,  1.36it/s]
1|3921|Loss: 0.8676407933235168:  61%|██████    | 3921/6470 [47:58<30:15,  1.40it/s]
1|3922|Loss: 0.9590827226638794:  61%|██████    | 3921/6470 [47:59<30:15,  1.40it/s]
1|3922|Loss: 0.9590827226638794:  61%|██████    | 3922/6470 [47:59<30:21,  1.40it/s]
1|3923|Loss: 0.8774512410163879:  61%|██████    | 3922/6470 [47:59<30:21,  1.40it/s]
1|3923|Loss: 0.8774512410163879:  61%|██████    | 3923/6470 [48:00<29:48,  1.42it/s]
1|3924|Loss: 0.9500302076339722:  61%|██████    | 3923/6470 [48:00<29:48,  1.42it/s]
1|3924|Loss: 0.9500302076339722:  61%|██████    | 3924/6470 [48:00<28:21,  1.50it/s]
1|3925|Loss: 0.8317759037017822:  61%|██████    | 3924/6470 [48:01<28:21,  1.50it/s]
1|3925|Loss: 0.8317759037017822:  61%|██████    | 3925/6470 [48:01<29:52,  1.42it/s]
1|3926|Loss: 0.972701370716095:  61%|██████    | 3925/6470 [48:01<29:52,  1.42it/s] 
1|3926|Loss: 0.972701370716095:  61%|██████    | 3926/6470 [48:02<29:19,  1.45it/s]
1|3927|Loss: 1.0173110961914062:  61%|██████    | 3926/6470 [48:02<29:19,  1.45it/s]
1|3927|Loss: 1.0173110961914062:  61%|██████    | 3927/6470 [48:02<29:40,  1.43it/s]
1|3928|Loss: 0.9000983834266663:  61%|██████    | 3927/6470 [48:03<29:40,  1.43it/s]
1|3928|Loss: 0.9000983834266663:  61%|██████    | 3928/6470 [48:03<30:20,  1.40it/s]
1|3929|Loss: 0.8284449577331543:  61%|██████    | 3928/6470 [48:04<30:20,  1.40it/s]
1|3929|Loss: 0.8284449577331543:  61%|██████    | 3929/6470 [48:04<30:47,  1.38it/s]
1|3930|Loss: 0.744078516960144:  61%|██████    | 3929/6470 [48:04<30:47,  1.38it/s] 
1|3930|Loss: 0.744078516960144:  61%|██████    | 3930/6470 [48:05<31:54,  1.33it/s]
1|3931|Loss: 1.089175820350647:  61%|██████    | 3930/6470 [48:05<31:54,  1.33it/s]
1|3931|Loss: 1.089175820350647:  61%|██████    | 3931/6470 [48:05<31:57,  1.32it/s]
1|3932|Loss: 0.7333865761756897:  61%|██████    | 3931/6470 [48:06<31:57,  1.32it/s]
1|3932|Loss: 0.7333865761756897:  61%|██████    | 3932/6470 [48:06<34:04,  1.24it/s]
1|3933|Loss: 0.890916109085083:  61%|██████    | 3932/6470 [48:07<34:04,  1.24it/s] 
1|3933|Loss: 0.890916109085083:  61%|██████    | 3933/6470 [48:07<34:34,  1.22it/s]
1|3934|Loss: 0.8141714334487915:  61%|██████    | 3933/6470 [48:08<34:34,  1.22it/s]
1|3934|Loss: 0.8141714334487915:  61%|██████    | 3934/6470 [48:08<34:23,  1.23it/s]
1|3935|Loss: 0.9310416579246521:  61%|██████    | 3934/6470 [48:08<34:23,  1.23it/s]
1|3935|Loss: 0.9310416579246521:  61%|██████    | 3935/6470 [48:09<33:33,  1.26it/s]
1|3936|Loss: 0.87332683801651:  61%|██████    | 3935/6470 [48:09<33:33,  1.26it/s]  
1|3936|Loss: 0.87332683801651:  61%|██████    | 3936/6470 [48:09<31:56,  1.32it/s]
1|3937|Loss: 0.8004293441772461:  61%|██████    | 3936/6470 [48:10<31:56,  1.32it/s]
1|3937|Loss: 0.8004293441772461:  61%|██████    | 3937/6470 [48:10<31:31,  1.34it/s]
1|3938|Loss: 0.8411502838134766:  61%|██████    | 3937/6470 [48:11<31:31,  1.34it/s]
1|3938|Loss: 0.8411502838134766:  61%|██████    | 3938/6470 [48:11<31:58,  1.32it/s]
1|3939|Loss: 0.8389354348182678:  61%|██████    | 3938/6470 [48:11<31:58,  1.32it/s]
1|3939|Loss: 0.8389354348182678:  61%|██████    | 3939/6470 [48:12<32:55,  1.28it/s]
1|3940|Loss: 0.9379979372024536:  61%|██████    | 3939/6470 [48:12<32:55,  1.28it/s]
1|3940|Loss: 0.9379979372024536:  61%|██████    | 3940/6470 [48:13<32:47,  1.29it/s]
1|3941|Loss: 0.9381771087646484:  61%|██████    | 3940/6470 [48:13<32:47,  1.29it/s]
1|3941|Loss: 0.9381771087646484:  61%|██████    | 3941/6470 [48:13<32:03,  1.32it/s]
1|3942|Loss: 0.7868545651435852:  61%|██████    | 3941/6470 [48:14<32:03,  1.32it/s]
1|3942|Loss: 0.7868545651435852:  61%|██████    | 3942/6470 [48:14<31:23,  1.34it/s]
1|3943|Loss: 0.9228541851043701:  61%|██████    | 3942/6470 [48:14<31:23,  1.34it/s]
1|3943|Loss: 0.9228541851043701:  61%|██████    | 3943/6470 [48:15<31:08,  1.35it/s]
1|3944|Loss: 0.9578526020050049:  61%|██████    | 3943/6470 [48:15<31:08,  1.35it/s]
1|3944|Loss: 0.9578526020050049:  61%|██████    | 3944/6470 [48:15<31:14,  1.35it/s]
1|3945|Loss: 0.8586087226867676:  61%|██████    | 3944/6470 [48:16<31:14,  1.35it/s]
1|3945|Loss: 0.8586087226867676:  61%|██████    | 3945/6470 [48:16<30:18,  1.39it/s]
1|3946|Loss: 0.8867805004119873:  61%|██████    | 3945/6470 [48:17<30:18,  1.39it/s]
1|3946|Loss: 0.8867805004119873:  61%|██████    | 3946/6470 [48:17<28:40,  1.47it/s]
1|3947|Loss: 0.9186087250709534:  61%|██████    | 3946/6470 [48:17<28:40,  1.47it/s]
1|3947|Loss: 0.9186087250709534:  61%|██████    | 3947/6470 [48:17<29:02,  1.45it/s]
1|3948|Loss: 0.8525000810623169:  61%|██████    | 3947/6470 [48:18<29:02,  1.45it/s]
1|3948|Loss: 0.8525000810623169:  61%|██████    | 3948/6470 [48:18<27:45,  1.51it/s]
1|3949|Loss: 1.046113133430481:  61%|██████    | 3948/6470 [48:18<27:45,  1.51it/s] 
1|3949|Loss: 1.046113133430481:  61%|██████    | 3949/6470 [48:19<27:17,  1.54it/s]
1|3950|Loss: 0.9165244698524475:  61%|██████    | 3949/6470 [48:19<27:17,  1.54it/s]
1|3950|Loss: 0.9165244698524475:  61%|██████    | 3950/6470 [48:19<26:31,  1.58it/s]
1|3951|Loss: 0.7577263712882996:  61%|██████    | 3950/6470 [48:20<26:31,  1.58it/s]
1|3951|Loss: 0.7577263712882996:  61%|██████    | 3951/6470 [48:20<26:14,  1.60it/s]
1|3952|Loss: 1.082497000694275:  61%|██████    | 3951/6470 [48:20<26:14,  1.60it/s] 
1|3952|Loss: 1.082497000694275:  61%|██████    | 3952/6470 [48:20<26:13,  1.60it/s]
1|3953|Loss: 0.811988353729248:  61%|██████    | 3952/6470 [48:21<26:13,  1.60it/s]
1|3953|Loss: 0.811988353729248:  61%|██████    | 3953/6470 [48:21<26:24,  1.59it/s]
1|3954|Loss: 1.0168942213058472:  61%|██████    | 3953/6470 [48:21<26:24,  1.59it/s]
1|3954|Loss: 1.0168942213058472:  61%|██████    | 3954/6470 [48:22<24:37,  1.70it/s]
1|3955|Loss: 0.8845004439353943:  61%|██████    | 3954/6470 [48:22<24:37,  1.70it/s]
1|3955|Loss: 0.8845004439353943:  61%|██████    | 3955/6470 [48:22<25:27,  1.65it/s]
1|3956|Loss: 0.8501713275909424:  61%|██████    | 3955/6470 [48:23<25:27,  1.65it/s]
1|3956|Loss: 0.8501713275909424:  61%|██████    | 3956/6470 [48:23<26:45,  1.57it/s]
1|3957|Loss: 0.8284077644348145:  61%|██████    | 3956/6470 [48:23<26:45,  1.57it/s]
1|3957|Loss: 0.8284077644348145:  61%|██████    | 3957/6470 [48:24<27:29,  1.52it/s]
1|3958|Loss: 0.8874557614326477:  61%|██████    | 3957/6470 [48:24<27:29,  1.52it/s]
1|3958|Loss: 0.8874557614326477:  61%|██████    | 3958/6470 [48:24<27:10,  1.54it/s]
1|3959|Loss: 0.9237638115882874:  61%|██████    | 3958/6470 [48:25<27:10,  1.54it/s]
1|3959|Loss: 0.9237638115882874:  61%|██████    | 3959/6470 [48:25<28:50,  1.45it/s]
1|3960|Loss: 0.9716510772705078:  61%|██████    | 3959/6470 [48:25<28:50,  1.45it/s]
1|3960|Loss: 0.9716510772705078:  61%|██████    | 3960/6470 [48:26<27:36,  1.52it/s]
1|3961|Loss: 0.9912325739860535:  61%|██████    | 3960/6470 [48:26<27:36,  1.52it/s]
1|3961|Loss: 0.9912325739860535:  61%|██████    | 3961/6470 [48:26<27:59,  1.49it/s]
1|3962|Loss: 0.8043065667152405:  61%|██████    | 3961/6470 [48:27<27:59,  1.49it/s]
1|3962|Loss: 0.8043065667152405:  61%|██████    | 3962/6470 [48:27<28:30,  1.47it/s]
1|3963|Loss: 0.7388460636138916:  61%|██████    | 3962/6470 [48:27<28:30,  1.47it/s]
1|3963|Loss: 0.7388460636138916:  61%|██████▏   | 3963/6470 [48:28<28:48,  1.45it/s]
1|3964|Loss: 0.730665922164917:  61%|██████▏   | 3963/6470 [48:28<28:48,  1.45it/s] 
1|3964|Loss: 0.730665922164917:  61%|██████▏   | 3964/6470 [48:29<30:16,  1.38it/s]
1|3965|Loss: 0.799545168876648:  61%|██████▏   | 3964/6470 [48:29<30:16,  1.38it/s]
1|3965|Loss: 0.799545168876648:  61%|██████▏   | 3965/6470 [48:29<31:54,  1.31it/s]
1|3966|Loss: 0.8997330665588379:  61%|██████▏   | 3965/6470 [48:30<31:54,  1.31it/s]
1|3966|Loss: 0.8997330665588379:  61%|██████▏   | 3966/6470 [48:30<32:55,  1.27it/s]
1|3967|Loss: 0.8831707835197449:  61%|██████▏   | 3966/6470 [48:31<32:55,  1.27it/s]
1|3967|Loss: 0.8831707835197449:  61%|██████▏   | 3967/6470 [48:31<33:25,  1.25it/s]
1|3968|Loss: 0.9171106815338135:  61%|██████▏   | 3967/6470 [48:32<33:25,  1.25it/s]
1|3968|Loss: 0.9171106815338135:  61%|██████▏   | 3968/6470 [48:32<32:50,  1.27it/s]
1|3969|Loss: 0.8453736901283264:  61%|██████▏   | 3968/6470 [48:32<32:50,  1.27it/s]
1|3969|Loss: 0.8453736901283264:  61%|██████▏   | 3969/6470 [48:33<30:38,  1.36it/s]
1|3970|Loss: 0.9408169984817505:  61%|██████▏   | 3969/6470 [48:33<30:38,  1.36it/s]
1|3970|Loss: 0.9408169984817505:  61%|██████▏   | 3970/6470 [48:33<31:47,  1.31it/s]
1|3971|Loss: 0.8664125204086304:  61%|██████▏   | 3970/6470 [48:34<31:47,  1.31it/s]
1|3971|Loss: 0.8664125204086304:  61%|██████▏   | 3971/6470 [48:34<30:01,  1.39it/s]
1|3972|Loss: 0.7973321676254272:  61%|██████▏   | 3971/6470 [48:34<30:01,  1.39it/s]
1|3972|Loss: 0.7973321676254272:  61%|██████▏   | 3972/6470 [48:35<29:51,  1.39it/s]
1|3973|Loss: 0.7150830030441284:  61%|██████▏   | 3972/6470 [48:35<29:51,  1.39it/s]
1|3973|Loss: 0.7150830030441284:  61%|██████▏   | 3973/6470 [48:35<30:43,  1.35it/s]
1|3974|Loss: 0.9419732093811035:  61%|██████▏   | 3973/6470 [48:36<30:43,  1.35it/s]
1|3974|Loss: 0.9419732093811035:  61%|██████▏   | 3974/6470 [48:36<29:25,  1.41it/s]
1|3975|Loss: 1.0157419443130493:  61%|██████▏   | 3974/6470 [48:36<29:25,  1.41it/s]
1|3975|Loss: 1.0157419443130493:  61%|██████▏   | 3975/6470 [48:37<29:23,  1.41it/s]
1|3976|Loss: 1.0244027376174927:  61%|██████▏   | 3975/6470 [48:37<29:23,  1.41it/s]
1|3976|Loss: 1.0244027376174927:  61%|██████▏   | 3976/6470 [48:38<29:49,  1.39it/s]
1|3977|Loss: 0.828821063041687:  61%|██████▏   | 3976/6470 [48:38<29:49,  1.39it/s] 
1|3977|Loss: 0.828821063041687:  61%|██████▏   | 3977/6470 [48:38<27:17,  1.52it/s]
1|3978|Loss: 0.9370288848876953:  61%|██████▏   | 3977/6470 [48:38<27:17,  1.52it/s]
1|3978|Loss: 0.9370288848876953:  61%|██████▏   | 3978/6470 [48:39<27:10,  1.53it/s]
1|3979|Loss: 0.8963379263877869:  61%|██████▏   | 3978/6470 [48:39<27:10,  1.53it/s]
1|3979|Loss: 0.8963379263877869:  61%|██████▏   | 3979/6470 [48:39<27:53,  1.49it/s]
1|3980|Loss: 0.7708265781402588:  61%|██████▏   | 3979/6470 [48:40<27:53,  1.49it/s]
1|3980|Loss: 0.7708265781402588:  62%|██████▏   | 3980/6470 [48:40<29:42,  1.40it/s]
1|3981|Loss: 0.8615266680717468:  62%|██████▏   | 3980/6470 [48:41<29:42,  1.40it/s]
1|3981|Loss: 0.8615266680717468:  62%|██████▏   | 3981/6470 [48:41<31:18,  1.32it/s]
1|3982|Loss: 0.8297836780548096:  62%|██████▏   | 3981/6470 [48:42<31:18,  1.32it/s]
1|3982|Loss: 0.8297836780548096:  62%|██████▏   | 3982/6470 [48:42<32:35,  1.27it/s]
1|3983|Loss: 0.802502453327179:  62%|██████▏   | 3982/6470 [48:42<32:35,  1.27it/s] 
1|3983|Loss: 0.802502453327179:  62%|██████▏   | 3983/6470 [48:43<33:17,  1.25it/s]
1|3984|Loss: 0.934991180896759:  62%|██████▏   | 3983/6470 [48:43<33:17,  1.25it/s]
1|3984|Loss: 0.934991180896759:  62%|██████▏   | 3984/6470 [48:44<32:40,  1.27it/s]
1|3985|Loss: 0.9588949084281921:  62%|██████▏   | 3984/6470 [48:44<32:40,  1.27it/s]
1|3985|Loss: 0.9588949084281921:  62%|██████▏   | 3985/6470 [48:44<31:58,  1.30it/s]
1|3986|Loss: 0.8901703953742981:  62%|██████▏   | 3985/6470 [48:45<31:58,  1.30it/s]
1|3986|Loss: 0.8901703953742981:  62%|██████▏   | 3986/6470 [48:45<32:01,  1.29it/s]
1|3987|Loss: 1.0104624032974243:  62%|██████▏   | 3986/6470 [48:45<32:01,  1.29it/s]
1|3987|Loss: 1.0104624032974243:  62%|██████▏   | 3987/6470 [48:46<28:55,  1.43it/s]
1|3988|Loss: 0.9367223381996155:  62%|██████▏   | 3987/6470 [48:46<28:55,  1.43it/s]
1|3988|Loss: 0.9367223381996155:  62%|██████▏   | 3988/6470 [48:46<27:27,  1.51it/s]
1|3989|Loss: 0.9493225812911987:  62%|██████▏   | 3988/6470 [48:46<27:27,  1.51it/s]
1|3989|Loss: 0.9493225812911987:  62%|██████▏   | 3989/6470 [48:47<26:08,  1.58it/s]
1|3990|Loss: 1.0107622146606445:  62%|██████▏   | 3989/6470 [48:47<26:08,  1.58it/s]
1|3990|Loss: 1.0107622146606445:  62%|██████▏   | 3990/6470 [48:48<29:32,  1.40it/s]
1|3991|Loss: 0.7922918200492859:  62%|██████▏   | 3990/6470 [48:48<29:32,  1.40it/s]
1|3991|Loss: 0.7922918200492859:  62%|██████▏   | 3991/6470 [48:48<30:31,  1.35it/s]
1|3992|Loss: 0.9046471118927002:  62%|██████▏   | 3991/6470 [48:49<30:31,  1.35it/s]
1|3992|Loss: 0.9046471118927002:  62%|██████▏   | 3992/6470 [48:49<30:28,  1.36it/s]
1|3993|Loss: 0.8094238638877869:  62%|██████▏   | 3992/6470 [48:50<30:28,  1.36it/s]
1|3993|Loss: 0.8094238638877869:  62%|██████▏   | 3993/6470 [48:50<29:18,  1.41it/s]
1|3994|Loss: 0.9305331110954285:  62%|██████▏   | 3993/6470 [48:50<29:18,  1.41it/s]
1|3994|Loss: 0.9305331110954285:  62%|██████▏   | 3994/6470 [48:51<29:18,  1.41it/s]
1|3995|Loss: 0.7686828374862671:  62%|██████▏   | 3994/6470 [48:51<29:18,  1.41it/s]
1|3995|Loss: 0.7686828374862671:  62%|██████▏   | 3995/6470 [48:51<29:15,  1.41it/s]
1|3996|Loss: 0.9352140426635742:  62%|██████▏   | 3995/6470 [48:52<29:15,  1.41it/s]
1|3996|Loss: 0.9352140426635742:  62%|██████▏   | 3996/6470 [48:52<30:41,  1.34it/s]
1|3997|Loss: 0.8397522568702698:  62%|██████▏   | 3996/6470 [48:52<30:41,  1.34it/s]
1|3997|Loss: 0.8397522568702698:  62%|██████▏   | 3997/6470 [48:53<29:09,  1.41it/s]
1|3998|Loss: 0.9467922449111938:  62%|██████▏   | 3997/6470 [48:53<29:09,  1.41it/s]
1|3998|Loss: 0.9467922449111938:  62%|██████▏   | 3998/6470 [48:53<28:03,  1.47it/s]
1|3999|Loss: 1.0048140287399292:  62%|██████▏   | 3998/6470 [48:54<28:03,  1.47it/s]
1|3999|Loss: 1.0048140287399292:  62%|██████▏   | 3999/6470 [48:54<29:27,  1.40it/s]
1|4000|Loss: 0.7329304218292236:  62%|██████▏   | 3999/6470 [48:55<29:27,  1.40it/s]
1|4000|Loss: 0.7329304218292236:  62%|██████▏   | 4000/6470 [48:55<29:50,  1.38it/s]
1|4001|Loss: 0.9382186532020569:  62%|██████▏   | 4000/6470 [48:55<29:50,  1.38it/s]
1|4001|Loss: 0.9382186532020569:  62%|██████▏   | 4001/6470 [48:55<28:19,  1.45it/s]
1|4002|Loss: 1.002257227897644:  62%|██████▏   | 4001/6470 [48:56<28:19,  1.45it/s] 
1|4002|Loss: 1.002257227897644:  62%|██████▏   | 4002/6470 [48:56<27:33,  1.49it/s]
1|4003|Loss: 0.7990860342979431:  62%|██████▏   | 4002/6470 [48:56<27:33,  1.49it/s]
1|4003|Loss: 0.7990860342979431:  62%|██████▏   | 4003/6470 [48:57<25:57,  1.58it/s]
1|4004|Loss: 0.7773952484130859:  62%|██████▏   | 4003/6470 [48:57<25:57,  1.58it/s]
1|4004|Loss: 0.7773952484130859:  62%|██████▏   | 4004/6470 [48:57<27:36,  1.49it/s]
1|4005|Loss: 0.6959975361824036:  62%|██████▏   | 4004/6470 [48:58<27:36,  1.49it/s]
1|4005|Loss: 0.6959975361824036:  62%|██████▏   | 4005/6470 [48:58<30:48,  1.33it/s]
1|4006|Loss: 1.008499026298523:  62%|██████▏   | 4005/6470 [48:59<30:48,  1.33it/s] 
1|4006|Loss: 1.008499026298523:  62%|██████▏   | 4006/6470 [48:59<28:29,  1.44it/s]
1|4007|Loss: 0.977979838848114:  62%|██████▏   | 4006/6470 [48:59<28:29,  1.44it/s]
1|4007|Loss: 0.977979838848114:  62%|██████▏   | 4007/6470 [49:00<28:11,  1.46it/s]
1|4008|Loss: 0.9123522043228149:  62%|██████▏   | 4007/6470 [49:00<28:11,  1.46it/s]
1|4008|Loss: 0.9123522043228149:  62%|██████▏   | 4008/6470 [49:00<28:54,  1.42it/s]
1|4009|Loss: 0.9051404595375061:  62%|██████▏   | 4008/6470 [49:01<28:54,  1.42it/s]
1|4009|Loss: 0.9051404595375061:  62%|██████▏   | 4009/6470 [49:01<27:51,  1.47it/s]
1|4010|Loss: 0.8829293251037598:  62%|██████▏   | 4009/6470 [49:01<27:51,  1.47it/s]
1|4010|Loss: 0.8829293251037598:  62%|██████▏   | 4010/6470 [49:02<27:16,  1.50it/s]
1|4011|Loss: 0.9031222462654114:  62%|██████▏   | 4010/6470 [49:02<27:16,  1.50it/s]
1|4011|Loss: 0.9031222462654114:  62%|██████▏   | 4011/6470 [49:02<26:50,  1.53it/s]
1|4012|Loss: 0.8836982250213623:  62%|██████▏   | 4011/6470 [49:02<26:50,  1.53it/s]
1|4012|Loss: 0.8836982250213623:  62%|██████▏   | 4012/6470 [49:03<25:22,  1.61it/s]
1|4013|Loss: 0.9316641092300415:  62%|██████▏   | 4012/6470 [49:03<25:22,  1.61it/s]
1|4013|Loss: 0.9316641092300415:  62%|██████▏   | 4013/6470 [49:03<26:14,  1.56it/s]
1|4014|Loss: 0.8492269515991211:  62%|██████▏   | 4013/6470 [49:04<26:14,  1.56it/s]
1|4014|Loss: 0.8492269515991211:  62%|██████▏   | 4014/6470 [49:04<26:14,  1.56it/s]
1|4015|Loss: 0.9018512964248657:  62%|██████▏   | 4014/6470 [49:04<26:14,  1.56it/s]
1|4015|Loss: 0.9018512964248657:  62%|██████▏   | 4015/6470 [49:05<28:20,  1.44it/s]
1|4016|Loss: 0.914948046207428:  62%|██████▏   | 4015/6470 [49:05<28:20,  1.44it/s] 
1|4016|Loss: 0.914948046207428:  62%|██████▏   | 4016/6470 [49:06<29:22,  1.39it/s]
1|4017|Loss: 0.837632954120636:  62%|██████▏   | 4016/6470 [49:06<29:22,  1.39it/s]
1|4017|Loss: 0.837632954120636:  62%|██████▏   | 4017/6470 [49:06<29:25,  1.39it/s]
1|4018|Loss: 0.8479116559028625:  62%|██████▏   | 4017/6470 [49:07<29:25,  1.39it/s]
1|4018|Loss: 0.8479116559028625:  62%|██████▏   | 4018/6470 [49:07<30:01,  1.36it/s]
1|4019|Loss: 0.8858100771903992:  62%|██████▏   | 4018/6470 [49:07<30:01,  1.36it/s]
1|4019|Loss: 0.8858100771903992:  62%|██████▏   | 4019/6470 [49:08<28:21,  1.44it/s]
1|4020|Loss: 0.8439595699310303:  62%|██████▏   | 4019/6470 [49:08<28:21,  1.44it/s]
1|4020|Loss: 0.8439595699310303:  62%|██████▏   | 4020/6470 [49:08<27:27,  1.49it/s]
1|4021|Loss: 0.8251270651817322:  62%|██████▏   | 4020/6470 [49:09<27:27,  1.49it/s]
1|4021|Loss: 0.8251270651817322:  62%|██████▏   | 4021/6470 [49:09<29:15,  1.39it/s]
1|4022|Loss: 0.8600276708602905:  62%|██████▏   | 4021/6470 [49:10<29:15,  1.39it/s]
1|4022|Loss: 0.8600276708602905:  62%|██████▏   | 4022/6470 [49:10<30:46,  1.33it/s]
1|4023|Loss: 0.9441016316413879:  62%|██████▏   | 4022/6470 [49:10<30:46,  1.33it/s]
1|4023|Loss: 0.9441016316413879:  62%|██████▏   | 4023/6470 [49:11<30:53,  1.32it/s]
1|4024|Loss: 0.8840287923812866:  62%|██████▏   | 4023/6470 [49:11<30:53,  1.32it/s]
1|4024|Loss: 0.8840287923812866:  62%|██████▏   | 4024/6470 [49:11<28:54,  1.41it/s]
1|4025|Loss: 0.866806149482727:  62%|██████▏   | 4024/6470 [49:12<28:54,  1.41it/s] 
1|4025|Loss: 0.866806149482727:  62%|██████▏   | 4025/6470 [49:12<26:22,  1.55it/s]
1|4026|Loss: 0.9386923313140869:  62%|██████▏   | 4025/6470 [49:12<26:22,  1.55it/s]
1|4026|Loss: 0.9386923313140869:  62%|██████▏   | 4026/6470 [49:12<25:51,  1.57it/s]
1|4027|Loss: 1.034354329109192:  62%|██████▏   | 4026/6470 [49:13<25:51,  1.57it/s] 
1|4027|Loss: 1.034354329109192:  62%|██████▏   | 4027/6470 [49:13<27:30,  1.48it/s]
1|4028|Loss: 0.8962820172309875:  62%|██████▏   | 4027/6470 [49:14<27:30,  1.48it/s]
1|4028|Loss: 0.8962820172309875:  62%|██████▏   | 4028/6470 [49:14<27:20,  1.49it/s]
1|4029|Loss: 0.9171469807624817:  62%|██████▏   | 4028/6470 [49:14<27:20,  1.49it/s]
1|4029|Loss: 0.9171469807624817:  62%|██████▏   | 4029/6470 [49:15<26:52,  1.51it/s]
1|4030|Loss: 0.7983249425888062:  62%|██████▏   | 4029/6470 [49:15<26:52,  1.51it/s]
1|4030|Loss: 0.7983249425888062:  62%|██████▏   | 4030/6470 [49:15<27:42,  1.47it/s]
1|4031|Loss: 0.9633611440658569:  62%|██████▏   | 4030/6470 [49:16<27:42,  1.47it/s]
1|4031|Loss: 0.9633611440658569:  62%|██████▏   | 4031/6470 [49:16<27:21,  1.49it/s]
1|4032|Loss: 1.010230541229248:  62%|██████▏   | 4031/6470 [49:16<27:21,  1.49it/s] 
1|4032|Loss: 1.010230541229248:  62%|██████▏   | 4032/6470 [49:17<26:13,  1.55it/s]
1|4033|Loss: 0.908656895160675:  62%|██████▏   | 4032/6470 [49:17<26:13,  1.55it/s]
1|4033|Loss: 0.908656895160675:  62%|██████▏   | 4033/6470 [49:17<27:27,  1.48it/s]
1|4034|Loss: 0.7322095632553101:  62%|██████▏   | 4033/6470 [49:18<27:27,  1.48it/s]
1|4034|Loss: 0.7322095632553101:  62%|██████▏   | 4034/6470 [49:18<28:55,  1.40it/s]
1|4035|Loss: 0.8111529350280762:  62%|██████▏   | 4034/6470 [49:19<28:55,  1.40it/s]
1|4035|Loss: 0.8111529350280762:  62%|██████▏   | 4035/6470 [49:19<31:34,  1.29it/s]
1|4036|Loss: 0.9250080585479736:  62%|██████▏   | 4035/6470 [49:19<31:34,  1.29it/s]
1|4036|Loss: 0.9250080585479736:  62%|██████▏   | 4036/6470 [49:20<30:44,  1.32it/s]
1|4037|Loss: 0.9074170589447021:  62%|██████▏   | 4036/6470 [49:20<30:44,  1.32it/s]
1|4037|Loss: 0.9074170589447021:  62%|██████▏   | 4037/6470 [49:21<32:02,  1.27it/s]
1|4038|Loss: 1.0541725158691406:  62%|██████▏   | 4037/6470 [49:21<32:02,  1.27it/s]
1|4038|Loss: 1.0541725158691406:  62%|██████▏   | 4038/6470 [49:21<29:39,  1.37it/s]
1|4039|Loss: 0.9659702181816101:  62%|██████▏   | 4038/6470 [49:21<29:39,  1.37it/s]
1|4039|Loss: 0.9659702181816101:  62%|██████▏   | 4039/6470 [49:22<28:10,  1.44it/s]
1|4040|Loss: 0.8288884162902832:  62%|██████▏   | 4039/6470 [49:22<28:10,  1.44it/s]
1|4040|Loss: 0.8288884162902832:  62%|██████▏   | 4040/6470 [49:23<29:37,  1.37it/s]
1|4041|Loss: 0.7391738295555115:  62%|██████▏   | 4040/6470 [49:23<29:37,  1.37it/s]
1|4041|Loss: 0.7391738295555115:  62%|██████▏   | 4041/6470 [49:23<29:48,  1.36it/s]
1|4042|Loss: 0.8882986903190613:  62%|██████▏   | 4041/6470 [49:24<29:48,  1.36it/s]
1|4042|Loss: 0.8882986903190613:  62%|██████▏   | 4042/6470 [49:24<30:15,  1.34it/s]
1|4043|Loss: 0.8648247122764587:  62%|██████▏   | 4042/6470 [49:24<30:15,  1.34it/s]
1|4043|Loss: 0.8648247122764587:  62%|██████▏   | 4043/6470 [49:25<27:35,  1.47it/s]
1|4044|Loss: 0.8369301557540894:  62%|██████▏   | 4043/6470 [49:25<27:35,  1.47it/s]
1|4044|Loss: 0.8369301557540894:  63%|██████▎   | 4044/6470 [49:25<27:24,  1.47it/s]
1|4045|Loss: 0.7811439633369446:  63%|██████▎   | 4044/6470 [49:26<27:24,  1.47it/s]
1|4045|Loss: 0.7811439633369446:  63%|██████▎   | 4045/6470 [49:26<28:07,  1.44it/s]
1|4046|Loss: 0.8723001480102539:  63%|██████▎   | 4045/6470 [49:26<28:07,  1.44it/s]
1|4046|Loss: 0.8723001480102539:  63%|██████▎   | 4046/6470 [49:27<28:38,  1.41it/s]
1|4047|Loss: 1.1199291944503784:  63%|██████▎   | 4046/6470 [49:27<28:38,  1.41it/s]
1|4047|Loss: 1.1199291944503784:  63%|██████▎   | 4047/6470 [49:27<27:56,  1.45it/s]
1|4048|Loss: 0.907187283039093:  63%|██████▎   | 4047/6470 [49:28<27:56,  1.45it/s] 
1|4048|Loss: 0.907187283039093:  63%|██████▎   | 4048/6470 [49:28<27:17,  1.48it/s]
1|4049|Loss: 0.8383620977401733:  63%|██████▎   | 4048/6470 [49:28<27:17,  1.48it/s]
1|4049|Loss: 0.8383620977401733:  63%|██████▎   | 4049/6470 [49:29<27:54,  1.45it/s]
1|4050|Loss: 0.7381194829940796:  63%|██████▎   | 4049/6470 [49:29<27:54,  1.45it/s]
1|4050|Loss: 0.7381194829940796:  63%|██████▎   | 4050/6470 [49:30<29:02,  1.39it/s]
1|4051|Loss: 0.8860694169998169:  63%|██████▎   | 4050/6470 [49:30<29:02,  1.39it/s]
1|4051|Loss: 0.8860694169998169:  63%|██████▎   | 4051/6470 [49:30<27:59,  1.44it/s]
1|4052|Loss: 0.9550063014030457:  63%|██████▎   | 4051/6470 [49:31<27:59,  1.44it/s]
1|4052|Loss: 0.9550063014030457:  63%|██████▎   | 4052/6470 [49:31<26:36,  1.51it/s]
1|4053|Loss: 0.877705991268158:  63%|██████▎   | 4052/6470 [49:31<26:36,  1.51it/s] 
1|4053|Loss: 0.877705991268158:  63%|██████▎   | 4053/6470 [49:31<24:59,  1.61it/s]
1|4054|Loss: 0.8710147142410278:  63%|██████▎   | 4053/6470 [49:32<24:59,  1.61it/s]
1|4054|Loss: 0.8710147142410278:  63%|██████▎   | 4054/6470 [49:32<26:47,  1.50it/s]
1|4055|Loss: 0.9503875374794006:  63%|██████▎   | 4054/6470 [49:32<26:47,  1.50it/s]
1|4055|Loss: 0.9503875374794006:  63%|██████▎   | 4055/6470 [49:33<25:29,  1.58it/s]
1|4056|Loss: 0.6590144038200378:  63%|██████▎   | 4055/6470 [49:33<25:29,  1.58it/s]
1|4056|Loss: 0.6590144038200378:  63%|██████▎   | 4056/6470 [49:33<27:25,  1.47it/s]
1|4057|Loss: 0.9031493067741394:  63%|██████▎   | 4056/6470 [49:34<27:25,  1.47it/s]
1|4057|Loss: 0.9031493067741394:  63%|██████▎   | 4057/6470 [49:34<26:54,  1.49it/s]
1|4058|Loss: 0.9215036630630493:  63%|██████▎   | 4057/6470 [49:34<26:54,  1.49it/s]
1|4058|Loss: 0.9215036630630493:  63%|██████▎   | 4058/6470 [49:35<26:22,  1.52it/s]
1|4059|Loss: 0.8172447085380554:  63%|██████▎   | 4058/6470 [49:35<26:22,  1.52it/s]
1|4059|Loss: 0.8172447085380554:  63%|██████▎   | 4059/6470 [49:35<27:04,  1.48it/s]
1|4060|Loss: 0.9101232886314392:  63%|██████▎   | 4059/6470 [49:36<27:04,  1.48it/s]
1|4060|Loss: 0.9101232886314392:  63%|██████▎   | 4060/6470 [49:36<25:03,  1.60it/s]
1|4061|Loss: 0.9298624992370605:  63%|██████▎   | 4060/6470 [49:36<25:03,  1.60it/s]
1|4061|Loss: 0.9298624992370605:  63%|██████▎   | 4061/6470 [49:36<23:53,  1.68it/s]
1|4062|Loss: 0.8598551154136658:  63%|██████▎   | 4061/6470 [49:37<23:53,  1.68it/s]
1|4062|Loss: 0.8598551154136658:  63%|██████▎   | 4062/6470 [49:37<25:05,  1.60it/s]
1|4063|Loss: 0.9298652410507202:  63%|██████▎   | 4062/6470 [49:38<25:05,  1.60it/s]
1|4063|Loss: 0.9298652410507202:  63%|██████▎   | 4063/6470 [49:38<25:27,  1.58it/s]
1|4064|Loss: 1.0461691617965698:  63%|██████▎   | 4063/6470 [49:38<25:27,  1.58it/s]
1|4064|Loss: 1.0461691617965698:  63%|██████▎   | 4064/6470 [49:39<27:38,  1.45it/s]
1|4065|Loss: 0.7266813516616821:  63%|██████▎   | 4064/6470 [49:39<27:38,  1.45it/s]
1|4065|Loss: 0.7266813516616821:  63%|██████▎   | 4065/6470 [49:39<28:41,  1.40it/s]
1|4066|Loss: 0.9865083694458008:  63%|██████▎   | 4065/6470 [49:40<28:41,  1.40it/s]
1|4066|Loss: 0.9865083694458008:  63%|██████▎   | 4066/6470 [49:40<29:00,  1.38it/s]
1|4067|Loss: 0.7989469170570374:  63%|██████▎   | 4066/6470 [49:41<29:00,  1.38it/s]
1|4067|Loss: 0.7989469170570374:  63%|██████▎   | 4067/6470 [49:41<29:19,  1.37it/s]
1|4068|Loss: 0.9273717403411865:  63%|██████▎   | 4067/6470 [49:41<29:19,  1.37it/s]
1|4068|Loss: 0.9273717403411865:  63%|██████▎   | 4068/6470 [49:42<29:35,  1.35it/s]
1|4069|Loss: 0.9860045313835144:  63%|██████▎   | 4068/6470 [49:42<29:35,  1.35it/s]
1|4069|Loss: 0.9860045313835144:  63%|██████▎   | 4069/6470 [49:42<29:22,  1.36it/s]
1|4070|Loss: 0.9064270853996277:  63%|██████▎   | 4069/6470 [49:43<29:22,  1.36it/s]
1|4070|Loss: 0.9064270853996277:  63%|██████▎   | 4070/6470 [49:43<29:31,  1.35it/s]
1|4071|Loss: 0.8062129616737366:  63%|██████▎   | 4070/6470 [49:43<29:31,  1.35it/s]
1|4071|Loss: 0.8062129616737366:  63%|██████▎   | 4071/6470 [49:44<27:50,  1.44it/s]
1|4072|Loss: 0.9272164702415466:  63%|██████▎   | 4071/6470 [49:44<27:50,  1.44it/s]
1|4072|Loss: 0.9272164702415466:  63%|██████▎   | 4072/6470 [49:45<29:02,  1.38it/s]
1|4073|Loss: 0.9795806407928467:  63%|██████▎   | 4072/6470 [49:45<29:02,  1.38it/s]
1|4073|Loss: 0.9795806407928467:  63%|██████▎   | 4073/6470 [49:45<28:05,  1.42it/s]
1|4074|Loss: 0.9596313834190369:  63%|██████▎   | 4073/6470 [49:46<28:05,  1.42it/s]
1|4074|Loss: 0.9596313834190369:  63%|██████▎   | 4074/6470 [49:46<28:37,  1.40it/s]
1|4075|Loss: 0.8819714784622192:  63%|██████▎   | 4074/6470 [49:46<28:37,  1.40it/s]
1|4075|Loss: 0.8819714784622192:  63%|██████▎   | 4075/6470 [49:47<29:58,  1.33it/s]
1|4076|Loss: 0.9355774521827698:  63%|██████▎   | 4075/6470 [49:47<29:58,  1.33it/s]
1|4076|Loss: 0.9355774521827698:  63%|██████▎   | 4076/6470 [49:47<29:45,  1.34it/s]
1|4077|Loss: 0.9023212790489197:  63%|██████▎   | 4076/6470 [49:48<29:45,  1.34it/s]
1|4077|Loss: 0.9023212790489197:  63%|██████▎   | 4077/6470 [49:48<29:33,  1.35it/s]
1|4078|Loss: 0.8208327889442444:  63%|██████▎   | 4077/6470 [49:49<29:33,  1.35it/s]
1|4078|Loss: 0.8208327889442444:  63%|██████▎   | 4078/6470 [49:49<29:10,  1.37it/s]
1|4079|Loss: 0.7550486326217651:  63%|██████▎   | 4078/6470 [49:49<29:10,  1.37it/s]
1|4079|Loss: 0.7550486326217651:  63%|██████▎   | 4079/6470 [49:50<30:15,  1.32it/s]
1|4080|Loss: 0.9379413723945618:  63%|██████▎   | 4079/6470 [49:50<30:15,  1.32it/s]
1|4080|Loss: 0.9379413723945618:  63%|██████▎   | 4080/6470 [49:51<30:15,  1.32it/s]
1|4081|Loss: 0.9616843461990356:  63%|██████▎   | 4080/6470 [49:51<30:15,  1.32it/s]
1|4081|Loss: 0.9616843461990356:  63%|██████▎   | 4081/6470 [49:51<28:48,  1.38it/s]
1|4082|Loss: 1.0089201927185059:  63%|██████▎   | 4081/6470 [49:52<28:48,  1.38it/s]
1|4082|Loss: 1.0089201927185059:  63%|██████▎   | 4082/6470 [49:52<28:16,  1.41it/s]
1|4083|Loss: 0.8940919041633606:  63%|██████▎   | 4082/6470 [49:52<28:16,  1.41it/s]
1|4083|Loss: 0.8940919041633606:  63%|██████▎   | 4083/6470 [49:52<27:23,  1.45it/s]
1|4084|Loss: 0.8676183819770813:  63%|██████▎   | 4083/6470 [49:53<27:23,  1.45it/s]
1|4084|Loss: 0.8676183819770813:  63%|██████▎   | 4084/6470 [49:53<28:10,  1.41it/s]
1|4085|Loss: 0.9201642870903015:  63%|██████▎   | 4084/6470 [49:54<28:10,  1.41it/s]
1|4085|Loss: 0.9201642870903015:  63%|██████▎   | 4085/6470 [49:54<29:11,  1.36it/s]
1|4086|Loss: 0.9767922163009644:  63%|██████▎   | 4085/6470 [49:54<29:11,  1.36it/s]
1|4086|Loss: 0.9767922163009644:  63%|██████▎   | 4086/6470 [49:55<27:53,  1.42it/s]
1|4087|Loss: 0.8812824487686157:  63%|██████▎   | 4086/6470 [49:55<27:53,  1.42it/s]
1|4087|Loss: 0.8812824487686157:  63%|██████▎   | 4087/6470 [49:55<28:35,  1.39it/s]
1|4088|Loss: 0.8156259059906006:  63%|██████▎   | 4087/6470 [49:56<28:35,  1.39it/s]
1|4088|Loss: 0.8156259059906006:  63%|██████▎   | 4088/6470 [49:56<30:14,  1.31it/s]
1|4089|Loss: 0.7560697197914124:  63%|██████▎   | 4088/6470 [49:57<30:14,  1.31it/s]
1|4089|Loss: 0.7560697197914124:  63%|██████▎   | 4089/6470 [49:57<30:20,  1.31it/s]
1|4090|Loss: 0.9262030124664307:  63%|██████▎   | 4089/6470 [49:57<30:20,  1.31it/s]
1|4090|Loss: 0.9262030124664307:  63%|██████▎   | 4090/6470 [49:58<27:15,  1.45it/s]
1|4091|Loss: 0.8879451155662537:  63%|██████▎   | 4090/6470 [49:58<27:15,  1.45it/s]
1|4091|Loss: 0.8879451155662537:  63%|██████▎   | 4091/6470 [49:58<24:42,  1.61it/s]
1|4092|Loss: 0.9399679899215698:  63%|██████▎   | 4091/6470 [49:58<24:42,  1.61it/s]
1|4092|Loss: 0.9399679899215698:  63%|██████▎   | 4092/6470 [49:59<24:28,  1.62it/s]
1|4093|Loss: 0.7818832993507385:  63%|██████▎   | 4092/6470 [49:59<24:28,  1.62it/s]
1|4093|Loss: 0.7818832993507385:  63%|██████▎   | 4093/6470 [49:59<25:40,  1.54it/s]
1|4094|Loss: 0.9039875268936157:  63%|██████▎   | 4093/6470 [50:00<25:40,  1.54it/s]
1|4094|Loss: 0.9039875268936157:  63%|██████▎   | 4094/6470 [50:00<25:16,  1.57it/s]
1|4095|Loss: 1.0448297262191772:  63%|██████▎   | 4094/6470 [50:00<25:16,  1.57it/s]
1|4095|Loss: 1.0448297262191772:  63%|██████▎   | 4095/6470 [50:01<24:42,  1.60it/s]
1|4096|Loss: 0.8517011404037476:  63%|██████▎   | 4095/6470 [50:01<24:42,  1.60it/s]
1|4096|Loss: 0.8517011404037476:  63%|██████▎   | 4096/6470 [50:01<24:13,  1.63it/s]
1|4097|Loss: 0.922339677810669:  63%|██████▎   | 4096/6470 [50:01<24:13,  1.63it/s] 
1|4097|Loss: 0.922339677810669:  63%|██████▎   | 4097/6470 [50:02<25:03,  1.58it/s]
1|4098|Loss: 1.0434526205062866:  63%|██████▎   | 4097/6470 [50:02<25:03,  1.58it/s]
1|4098|Loss: 1.0434526205062866:  63%|██████▎   | 4098/6470 [50:02<23:32,  1.68it/s]
1|4099|Loss: 0.8759186863899231:  63%|██████▎   | 4098/6470 [50:03<23:32,  1.68it/s]
1|4099|Loss: 0.8759186863899231:  63%|██████▎   | 4099/6470 [50:03<23:34,  1.68it/s]
1|4100|Loss: 1.015013575553894:  63%|██████▎   | 4099/6470 [50:03<23:34,  1.68it/s] 
1|4100|Loss: 1.015013575553894:  63%|██████▎   | 4100/6470 [50:04<24:42,  1.60it/s]
1|4101|Loss: 0.6200164556503296:  63%|██████▎   | 4100/6470 [50:04<24:42,  1.60it/s]
1|4101|Loss: 0.6200164556503296:  63%|██████▎   | 4101/6470 [50:05<28:00,  1.41it/s]
1|4102|Loss: 0.9627087116241455:  63%|██████▎   | 4101/6470 [50:05<28:00,  1.41it/s]
1|4102|Loss: 0.9627087116241455:  63%|██████▎   | 4102/6470 [50:05<28:59,  1.36it/s]
1|4103|Loss: 0.7208795547485352:  63%|██████▎   | 4102/6470 [50:06<28:59,  1.36it/s]
1|4103|Loss: 0.7208795547485352:  63%|██████▎   | 4103/6470 [50:06<30:01,  1.31it/s]
1|4104|Loss: 0.7736868262290955:  63%|██████▎   | 4103/6470 [50:07<30:01,  1.31it/s]
1|4104|Loss: 0.7736868262290955:  63%|██████▎   | 4104/6470 [50:07<30:04,  1.31it/s]
1|4105|Loss: 0.8194817900657654:  63%|██████▎   | 4104/6470 [50:07<30:04,  1.31it/s]
1|4105|Loss: 0.8194817900657654:  63%|██████▎   | 4105/6470 [50:08<31:03,  1.27it/s]
1|4106|Loss: 0.9132674932479858:  63%|██████▎   | 4105/6470 [50:08<31:03,  1.27it/s]
1|4106|Loss: 0.9132674932479858:  63%|██████▎   | 4106/6470 [50:09<30:56,  1.27it/s]
1|4107|Loss: 0.827452540397644:  63%|██████▎   | 4106/6470 [50:09<30:56,  1.27it/s] 
1|4107|Loss: 0.827452540397644:  63%|██████▎   | 4107/6470 [50:09<29:03,  1.36it/s]
1|4108|Loss: 1.0019237995147705:  63%|██████▎   | 4107/6470 [50:10<29:03,  1.36it/s]
1|4108|Loss: 1.0019237995147705:  63%|██████▎   | 4108/6470 [50:10<29:59,  1.31it/s]
1|4109|Loss: 0.980506956577301:  63%|██████▎   | 4108/6470 [50:10<29:59,  1.31it/s] 
1|4109|Loss: 0.980506956577301:  64%|██████▎   | 4109/6470 [50:11<27:58,  1.41it/s]
1|4110|Loss: 0.9786381721496582:  64%|██████▎   | 4109/6470 [50:11<27:58,  1.41it/s]
1|4110|Loss: 0.9786381721496582:  64%|██████▎   | 4110/6470 [50:11<27:38,  1.42it/s]
1|4111|Loss: 0.9392351508140564:  64%|██████▎   | 4110/6470 [50:12<27:38,  1.42it/s]
1|4111|Loss: 0.9392351508140564:  64%|██████▎   | 4111/6470 [50:12<27:07,  1.45it/s]
1|4112|Loss: 1.0668197870254517:  64%|██████▎   | 4111/6470 [50:12<27:07,  1.45it/s]
1|4112|Loss: 1.0668197870254517:  64%|██████▎   | 4112/6470 [50:13<27:07,  1.45it/s]
1|4113|Loss: 0.9545977115631104:  64%|██████▎   | 4112/6470 [50:13<27:07,  1.45it/s]
1|4113|Loss: 0.9545977115631104:  64%|██████▎   | 4113/6470 [50:13<27:15,  1.44it/s]
1|4114|Loss: 0.8270630240440369:  64%|██████▎   | 4113/6470 [50:14<27:15,  1.44it/s]
1|4114|Loss: 0.8270630240440369:  64%|██████▎   | 4114/6470 [50:14<28:46,  1.36it/s]
1|4115|Loss: 0.7557716965675354:  64%|██████▎   | 4114/6470 [50:15<28:46,  1.36it/s]
1|4115|Loss: 0.7557716965675354:  64%|██████▎   | 4115/6470 [50:15<28:52,  1.36it/s]
1|4116|Loss: 0.8930838108062744:  64%|██████▎   | 4115/6470 [50:15<28:52,  1.36it/s]
1|4116|Loss: 0.8930838108062744:  64%|██████▎   | 4116/6470 [50:16<27:52,  1.41it/s]
1|4117|Loss: 0.8938223719596863:  64%|██████▎   | 4116/6470 [50:16<27:52,  1.41it/s]
1|4117|Loss: 0.8938223719596863:  64%|██████▎   | 4117/6470 [50:16<27:03,  1.45it/s]
1|4118|Loss: 0.9476555585861206:  64%|██████▎   | 4117/6470 [50:17<27:03,  1.45it/s]
1|4118|Loss: 0.9476555585861206:  64%|██████▎   | 4118/6470 [50:17<27:22,  1.43it/s]
1|4119|Loss: 0.7957395315170288:  64%|██████▎   | 4118/6470 [50:17<27:22,  1.43it/s]
1|4119|Loss: 0.7957395315170288:  64%|██████▎   | 4119/6470 [50:18<26:55,  1.46it/s]
1|4120|Loss: 0.8008275032043457:  64%|██████▎   | 4119/6470 [50:18<26:55,  1.46it/s]
1|4120|Loss: 0.8008275032043457:  64%|██████▎   | 4120/6470 [50:18<28:20,  1.38it/s]
1|4121|Loss: 0.8020368814468384:  64%|██████▎   | 4120/6470 [50:19<28:20,  1.38it/s]
1|4121|Loss: 0.8020368814468384:  64%|██████▎   | 4121/6470 [50:19<29:52,  1.31it/s]
1|4122|Loss: 0.8029771447181702:  64%|██████▎   | 4121/6470 [50:20<29:52,  1.31it/s]
1|4122|Loss: 0.8029771447181702:  64%|██████▎   | 4122/6470 [50:20<27:44,  1.41it/s]
1|4123|Loss: 0.8892203569412231:  64%|██████▎   | 4122/6470 [50:20<27:44,  1.41it/s]
1|4123|Loss: 0.8892203569412231:  64%|██████▎   | 4123/6470 [50:21<28:06,  1.39it/s]
1|4124|Loss: 0.9189589023590088:  64%|██████▎   | 4123/6470 [50:21<28:06,  1.39it/s]
1|4124|Loss: 0.9189589023590088:  64%|██████▎   | 4124/6470 [50:21<29:40,  1.32it/s]
1|4125|Loss: 0.818631649017334:  64%|██████▎   | 4124/6470 [50:22<29:40,  1.32it/s] 
1|4125|Loss: 0.818631649017334:  64%|██████▍   | 4125/6470 [50:22<29:30,  1.32it/s]
1|4126|Loss: 0.9676548838615417:  64%|██████▍   | 4125/6470 [50:22<29:30,  1.32it/s]
1|4126|Loss: 0.9676548838615417:  64%|██████▍   | 4126/6470 [50:23<28:03,  1.39it/s]
1|4127|Loss: 0.8675468564033508:  64%|██████▍   | 4126/6470 [50:23<28:03,  1.39it/s]
1|4127|Loss: 0.8675468564033508:  64%|██████▍   | 4127/6470 [50:23<27:46,  1.41it/s]
1|4128|Loss: 0.8716317415237427:  64%|██████▍   | 4127/6470 [50:24<27:46,  1.41it/s]
1|4128|Loss: 0.8716317415237427:  64%|██████▍   | 4128/6470 [50:24<28:12,  1.38it/s]
1|4129|Loss: 0.9143925905227661:  64%|██████▍   | 4128/6470 [50:25<28:12,  1.38it/s]
1|4129|Loss: 0.9143925905227661:  64%|██████▍   | 4129/6470 [50:25<28:34,  1.37it/s]
1|4130|Loss: 0.8318120241165161:  64%|██████▍   | 4129/6470 [50:25<28:34,  1.37it/s]
1|4130|Loss: 0.8318120241165161:  64%|██████▍   | 4130/6470 [50:25<26:19,  1.48it/s]
1|4131|Loss: 0.9125789403915405:  64%|██████▍   | 4130/6470 [50:26<26:19,  1.48it/s]
1|4131|Loss: 0.9125789403915405:  64%|██████▍   | 4131/6470 [50:26<25:34,  1.52it/s]
1|4132|Loss: 0.9910270571708679:  64%|██████▍   | 4131/6470 [50:26<25:34,  1.52it/s]
1|4132|Loss: 0.9910270571708679:  64%|██████▍   | 4132/6470 [50:27<25:27,  1.53it/s]
1|4133|Loss: 0.8891376852989197:  64%|██████▍   | 4132/6470 [50:27<25:27,  1.53it/s]
1|4133|Loss: 0.8891376852989197:  64%|██████▍   | 4133/6470 [50:28<27:14,  1.43it/s]
1|4134|Loss: 0.7449901103973389:  64%|██████▍   | 4133/6470 [50:28<27:14,  1.43it/s]
1|4134|Loss: 0.7449901103973389:  64%|██████▍   | 4134/6470 [50:28<26:20,  1.48it/s]
1|4135|Loss: 0.8502436876296997:  64%|██████▍   | 4134/6470 [50:29<26:20,  1.48it/s]
1|4135|Loss: 0.8502436876296997:  64%|██████▍   | 4135/6470 [50:29<26:30,  1.47it/s]
1|4136|Loss: 0.8636080026626587:  64%|██████▍   | 4135/6470 [50:29<26:30,  1.47it/s]
1|4136|Loss: 0.8636080026626587:  64%|██████▍   | 4136/6470 [50:30<28:25,  1.37it/s]
1|4137|Loss: 0.9325541853904724:  64%|██████▍   | 4136/6470 [50:30<28:25,  1.37it/s]
1|4137|Loss: 0.9325541853904724:  64%|██████▍   | 4137/6470 [50:31<30:16,  1.28it/s]
1|4138|Loss: 0.8525739312171936:  64%|██████▍   | 4137/6470 [50:31<30:16,  1.28it/s]
1|4138|Loss: 0.8525739312171936:  64%|██████▍   | 4138/6470 [50:31<29:10,  1.33it/s]
1|4139|Loss: 0.8480270504951477:  64%|██████▍   | 4138/6470 [50:32<29:10,  1.33it/s]
1|4139|Loss: 0.8480270504951477:  64%|██████▍   | 4139/6470 [50:32<29:16,  1.33it/s]
1|4140|Loss: 0.8014576435089111:  64%|██████▍   | 4139/6470 [50:32<29:16,  1.33it/s]
1|4140|Loss: 0.8014576435089111:  64%|██████▍   | 4140/6470 [50:33<28:53,  1.34it/s]
1|4141|Loss: 0.9195029735565186:  64%|██████▍   | 4140/6470 [50:33<28:53,  1.34it/s]
1|4141|Loss: 0.9195029735565186:  64%|██████▍   | 4141/6470 [50:33<27:11,  1.43it/s]
1|4142|Loss: 1.0350794792175293:  64%|██████▍   | 4141/6470 [50:34<27:11,  1.43it/s]
1|4142|Loss: 1.0350794792175293:  64%|██████▍   | 4142/6470 [50:34<27:12,  1.43it/s]
1|4143|Loss: 0.7873928546905518:  64%|██████▍   | 4142/6470 [50:34<27:12,  1.43it/s]
1|4143|Loss: 0.7873928546905518:  64%|██████▍   | 4143/6470 [50:35<24:58,  1.55it/s]
1|4144|Loss: 0.8310429453849792:  64%|██████▍   | 4143/6470 [50:35<24:58,  1.55it/s]
1|4144|Loss: 0.8310429453849792:  64%|██████▍   | 4144/6470 [50:35<25:03,  1.55it/s]
1|4145|Loss: 0.8531205058097839:  64%|██████▍   | 4144/6470 [50:36<25:03,  1.55it/s]
1|4145|Loss: 0.8531205058097839:  64%|██████▍   | 4145/6470 [50:36<25:45,  1.50it/s]
1|4146|Loss: 0.8271358013153076:  64%|██████▍   | 4145/6470 [50:36<25:45,  1.50it/s]
1|4146|Loss: 0.8271358013153076:  64%|██████▍   | 4146/6470 [50:37<25:24,  1.52it/s]
1|4147|Loss: 0.7413002252578735:  64%|██████▍   | 4146/6470 [50:37<25:24,  1.52it/s]
1|4147|Loss: 0.7413002252578735:  64%|██████▍   | 4147/6470 [50:37<25:41,  1.51it/s]
1|4148|Loss: 0.9364974498748779:  64%|██████▍   | 4147/6470 [50:38<25:41,  1.51it/s]
1|4148|Loss: 0.9364974498748779:  64%|██████▍   | 4148/6470 [50:38<26:05,  1.48it/s]
1|4149|Loss: 0.8597540855407715:  64%|██████▍   | 4148/6470 [50:38<26:05,  1.48it/s]
1|4149|Loss: 0.8597540855407715:  64%|██████▍   | 4149/6470 [50:39<25:17,  1.53it/s]
1|4150|Loss: 0.9563940763473511:  64%|██████▍   | 4149/6470 [50:39<25:17,  1.53it/s]
1|4150|Loss: 0.9563940763473511:  64%|██████▍   | 4150/6470 [50:39<27:03,  1.43it/s]
1|4151|Loss: 0.7843660116195679:  64%|██████▍   | 4150/6470 [50:40<27:03,  1.43it/s]
1|4151|Loss: 0.7843660116195679:  64%|██████▍   | 4151/6470 [50:40<24:37,  1.57it/s]
1|4152|Loss: 0.7908682823181152:  64%|██████▍   | 4151/6470 [50:40<24:37,  1.57it/s]
1|4152|Loss: 0.7908682823181152:  64%|██████▍   | 4152/6470 [50:40<24:01,  1.61it/s]
1|4153|Loss: 0.9692894220352173:  64%|██████▍   | 4152/6470 [50:41<24:01,  1.61it/s]
1|4153|Loss: 0.9692894220352173:  64%|██████▍   | 4153/6470 [50:41<24:13,  1.59it/s]
1|4154|Loss: 0.7804641723632812:  64%|██████▍   | 4153/6470 [50:41<24:13,  1.59it/s]
1|4154|Loss: 0.7804641723632812:  64%|██████▍   | 4154/6470 [50:42<25:19,  1.52it/s]
1|4155|Loss: 0.9334569573402405:  64%|██████▍   | 4154/6470 [50:42<25:19,  1.52it/s]
1|4155|Loss: 0.9334569573402405:  64%|██████▍   | 4155/6470 [50:42<25:13,  1.53it/s]
1|4156|Loss: 0.9521845579147339:  64%|██████▍   | 4155/6470 [50:43<25:13,  1.53it/s]
1|4156|Loss: 0.9521845579147339:  64%|██████▍   | 4156/6470 [50:43<24:44,  1.56it/s]
1|4157|Loss: 0.8313787579536438:  64%|██████▍   | 4156/6470 [50:43<24:44,  1.56it/s]
1|4157|Loss: 0.8313787579536438:  64%|██████▍   | 4157/6470 [50:44<24:27,  1.58it/s]
1|4158|Loss: 0.9904213547706604:  64%|██████▍   | 4157/6470 [50:44<24:27,  1.58it/s]
1|4158|Loss: 0.9904213547706604:  64%|██████▍   | 4158/6470 [50:44<25:04,  1.54it/s]
1|4159|Loss: 0.9860058426856995:  64%|██████▍   | 4158/6470 [50:45<25:04,  1.54it/s]
1|4159|Loss: 0.9860058426856995:  64%|██████▍   | 4159/6470 [50:45<23:52,  1.61it/s]
1|4160|Loss: 0.964434802532196:  64%|██████▍   | 4159/6470 [50:45<23:52,  1.61it/s] 
1|4160|Loss: 0.964434802532196:  64%|██████▍   | 4160/6470 [50:46<25:13,  1.53it/s]
1|4161|Loss: 0.8711798191070557:  64%|██████▍   | 4160/6470 [50:46<25:13,  1.53it/s]
1|4161|Loss: 0.8711798191070557:  64%|██████▍   | 4161/6470 [50:46<24:53,  1.55it/s]
1|4162|Loss: 0.891065239906311:  64%|██████▍   | 4161/6470 [50:47<24:53,  1.55it/s] 
1|4162|Loss: 0.891065239906311:  64%|██████▍   | 4162/6470 [50:47<22:59,  1.67it/s]
1|4163|Loss: 0.8363233804702759:  64%|██████▍   | 4162/6470 [50:47<22:59,  1.67it/s]
1|4163|Loss: 0.8363233804702759:  64%|██████▍   | 4163/6470 [50:47<23:59,  1.60it/s]
1|4164|Loss: 0.7882959842681885:  64%|██████▍   | 4163/6470 [50:48<23:59,  1.60it/s]
1|4164|Loss: 0.7882959842681885:  64%|██████▍   | 4164/6470 [50:48<24:30,  1.57it/s]
1|4165|Loss: 0.8582035303115845:  64%|██████▍   | 4164/6470 [50:49<24:30,  1.57it/s]
1|4165|Loss: 0.8582035303115845:  64%|██████▍   | 4165/6470 [50:49<26:34,  1.45it/s]
1|4166|Loss: 0.8170459270477295:  64%|██████▍   | 4165/6470 [50:49<26:34,  1.45it/s]
1|4166|Loss: 0.8170459270477295:  64%|██████▍   | 4166/6470 [50:50<27:12,  1.41it/s]
1|4167|Loss: 0.9097018241882324:  64%|██████▍   | 4166/6470 [50:50<27:12,  1.41it/s]
1|4167|Loss: 0.9097018241882324:  64%|██████▍   | 4167/6470 [50:50<27:09,  1.41it/s]
1|4168|Loss: 0.7986084222793579:  64%|██████▍   | 4167/6470 [50:51<27:09,  1.41it/s]
1|4168|Loss: 0.7986084222793579:  64%|██████▍   | 4168/6470 [50:51<25:16,  1.52it/s]
1|4169|Loss: 0.8988673686981201:  64%|██████▍   | 4168/6470 [50:51<25:16,  1.52it/s]
1|4169|Loss: 0.8988673686981201:  64%|██████▍   | 4169/6470 [50:51<23:35,  1.63it/s]
1|4170|Loss: 0.8216977119445801:  64%|██████▍   | 4169/6470 [50:52<23:35,  1.63it/s]
1|4170|Loss: 0.8216977119445801:  64%|██████▍   | 4170/6470 [50:52<24:35,  1.56it/s]
1|4171|Loss: 0.7323132753372192:  64%|██████▍   | 4170/6470 [50:53<24:35,  1.56it/s]
1|4171|Loss: 0.7323132753372192:  64%|██████▍   | 4171/6470 [50:53<24:55,  1.54it/s]
1|4172|Loss: 0.9917295575141907:  64%|██████▍   | 4171/6470 [50:53<24:55,  1.54it/s]
1|4172|Loss: 0.9917295575141907:  64%|██████▍   | 4172/6470 [50:54<25:22,  1.51it/s]
1|4173|Loss: 0.8280023336410522:  64%|██████▍   | 4172/6470 [50:54<25:22,  1.51it/s]
1|4173|Loss: 0.8280023336410522:  64%|██████▍   | 4173/6470 [50:54<25:04,  1.53it/s]
1|4174|Loss: 0.9074280261993408:  64%|██████▍   | 4173/6470 [50:55<25:04,  1.53it/s]
1|4174|Loss: 0.9074280261993408:  65%|██████▍   | 4174/6470 [50:55<24:48,  1.54it/s]
1|4175|Loss: 0.9090172052383423:  65%|██████▍   | 4174/6470 [50:55<24:48,  1.54it/s]
1|4175|Loss: 0.9090172052383423:  65%|██████▍   | 4175/6470 [50:56<25:42,  1.49it/s]
1|4176|Loss: 0.8420279622077942:  65%|██████▍   | 4175/6470 [50:56<25:42,  1.49it/s]
1|4176|Loss: 0.8420279622077942:  65%|██████▍   | 4176/6470 [50:56<27:18,  1.40it/s]
1|4177|Loss: 0.9471191763877869:  65%|██████▍   | 4176/6470 [50:57<27:18,  1.40it/s]
1|4177|Loss: 0.9471191763877869:  65%|██████▍   | 4177/6470 [50:57<28:20,  1.35it/s]
1|4178|Loss: 0.8874779343605042:  65%|██████▍   | 4177/6470 [50:58<28:20,  1.35it/s]
1|4178|Loss: 0.8874779343605042:  65%|██████▍   | 4178/6470 [50:58<27:02,  1.41it/s]
1|4179|Loss: 0.8379068970680237:  65%|██████▍   | 4178/6470 [50:58<27:02,  1.41it/s]
1|4179|Loss: 0.8379068970680237:  65%|██████▍   | 4179/6470 [50:58<25:34,  1.49it/s]
1|4180|Loss: 0.8066164255142212:  65%|██████▍   | 4179/6470 [50:59<25:34,  1.49it/s]
1|4180|Loss: 0.8066164255142212:  65%|██████▍   | 4180/6470 [50:59<26:30,  1.44it/s]
1|4181|Loss: 0.7932419776916504:  65%|██████▍   | 4180/6470 [51:00<26:30,  1.44it/s]
1|4181|Loss: 0.7932419776916504:  65%|██████▍   | 4181/6470 [51:00<27:55,  1.37it/s]
1|4182|Loss: 0.7060980796813965:  65%|██████▍   | 4181/6470 [51:00<27:55,  1.37it/s]
1|4182|Loss: 0.7060980796813965:  65%|██████▍   | 4182/6470 [51:01<29:17,  1.30it/s]
1|4183|Loss: 0.8109899163246155:  65%|██████▍   | 4182/6470 [51:01<29:17,  1.30it/s]
1|4183|Loss: 0.8109899163246155:  65%|██████▍   | 4183/6470 [51:01<28:26,  1.34it/s]
1|4184|Loss: 0.7852539420127869:  65%|██████▍   | 4183/6470 [51:02<28:26,  1.34it/s]
1|4184|Loss: 0.7852539420127869:  65%|██████▍   | 4184/6470 [51:02<27:13,  1.40it/s]
1|4185|Loss: 0.791719913482666:  65%|██████▍   | 4184/6470 [51:03<27:13,  1.40it/s] 
1|4185|Loss: 0.791719913482666:  65%|██████▍   | 4185/6470 [51:03<27:13,  1.40it/s]
1|4186|Loss: 0.8848317265510559:  65%|██████▍   | 4185/6470 [51:03<27:13,  1.40it/s]
1|4186|Loss: 0.8848317265510559:  65%|██████▍   | 4186/6470 [51:03<26:00,  1.46it/s]
1|4187|Loss: 0.8985175490379333:  65%|██████▍   | 4186/6470 [51:04<26:00,  1.46it/s]
1|4187|Loss: 0.8985175490379333:  65%|██████▍   | 4187/6470 [51:04<27:21,  1.39it/s]
1|4188|Loss: 0.8741370439529419:  65%|██████▍   | 4187/6470 [51:05<27:21,  1.39it/s]
1|4188|Loss: 0.8741370439529419:  65%|██████▍   | 4188/6470 [51:05<26:48,  1.42it/s]
1|4189|Loss: 0.9273993372917175:  65%|██████▍   | 4188/6470 [51:05<26:48,  1.42it/s]
1|4189|Loss: 0.9273993372917175:  65%|██████▍   | 4189/6470 [51:06<25:58,  1.46it/s]
1|4190|Loss: 0.7753210067749023:  65%|██████▍   | 4189/6470 [51:06<25:58,  1.46it/s]
1|4190|Loss: 0.7753210067749023:  65%|██████▍   | 4190/6470 [51:06<27:10,  1.40it/s]
1|4191|Loss: 0.7978107333183289:  65%|██████▍   | 4190/6470 [51:07<27:10,  1.40it/s]
1|4191|Loss: 0.7978107333183289:  65%|██████▍   | 4191/6470 [51:07<26:33,  1.43it/s]
1|4192|Loss: 0.6987009644508362:  65%|██████▍   | 4191/6470 [51:07<26:33,  1.43it/s]
1|4192|Loss: 0.6987009644508362:  65%|██████▍   | 4192/6470 [51:08<26:39,  1.42it/s]
1|4193|Loss: 0.9229795932769775:  65%|██████▍   | 4192/6470 [51:08<26:39,  1.42it/s]
1|4193|Loss: 0.9229795932769775:  65%|██████▍   | 4193/6470 [51:08<26:48,  1.42it/s]
1|4194|Loss: 0.753119170665741:  65%|██████▍   | 4193/6470 [51:09<26:48,  1.42it/s] 
1|4194|Loss: 0.753119170665741:  65%|██████▍   | 4194/6470 [51:09<26:02,  1.46it/s]
1|4195|Loss: 0.7806713581085205:  65%|██████▍   | 4194/6470 [51:09<26:02,  1.46it/s]
1|4195|Loss: 0.7806713581085205:  65%|██████▍   | 4195/6470 [51:10<26:17,  1.44it/s]
1|4196|Loss: 0.7229194641113281:  65%|██████▍   | 4195/6470 [51:10<26:17,  1.44it/s]
1|4196|Loss: 0.7229194641113281:  65%|██████▍   | 4196/6470 [51:10<25:05,  1.51it/s]
1|4197|Loss: 0.9508076906204224:  65%|██████▍   | 4196/6470 [51:11<25:05,  1.51it/s]
1|4197|Loss: 0.9508076906204224:  65%|██████▍   | 4197/6470 [51:11<25:41,  1.47it/s]
1|4198|Loss: 0.9810650944709778:  65%|██████▍   | 4197/6470 [51:12<25:41,  1.47it/s]
1|4198|Loss: 0.9810650944709778:  65%|██████▍   | 4198/6470 [51:12<27:47,  1.36it/s]
1|4199|Loss: 0.9703128933906555:  65%|██████▍   | 4198/6470 [51:12<27:47,  1.36it/s]
1|4199|Loss: 0.9703128933906555:  65%|██████▍   | 4199/6470 [51:13<26:54,  1.41it/s]
1|4200|Loss: 1.0152488946914673:  65%|██████▍   | 4199/6470 [51:13<26:54,  1.41it/s]
1|4200|Loss: 1.0152488946914673:  65%|██████▍   | 4200/6470 [51:13<25:10,  1.50it/s]
1|4201|Loss: 0.9078127145767212:  65%|██████▍   | 4200/6470 [51:14<25:10,  1.50it/s]
1|4201|Loss: 0.9078127145767212:  65%|██████▍   | 4201/6470 [51:14<25:34,  1.48it/s]
1|4202|Loss: 1.0028865337371826:  65%|██████▍   | 4201/6470 [51:14<25:34,  1.48it/s]
1|4202|Loss: 1.0028865337371826:  65%|██████▍   | 4202/6470 [51:15<26:10,  1.44it/s]
1|4203|Loss: 0.8747276663780212:  65%|██████▍   | 4202/6470 [51:15<26:10,  1.44it/s]
1|4203|Loss: 0.8747276663780212:  65%|██████▍   | 4203/6470 [51:15<27:30,  1.37it/s]
1|4204|Loss: 0.8433234691619873:  65%|██████▍   | 4203/6470 [51:16<27:30,  1.37it/s]
1|4204|Loss: 0.8433234691619873:  65%|██████▍   | 4204/6470 [51:16<28:21,  1.33it/s]
1|4205|Loss: 0.8107202053070068:  65%|██████▍   | 4204/6470 [51:17<28:21,  1.33it/s]
1|4205|Loss: 0.8107202053070068:  65%|██████▍   | 4205/6470 [51:17<26:26,  1.43it/s]
1|4206|Loss: 0.7497068643569946:  65%|██████▍   | 4205/6470 [51:17<26:26,  1.43it/s]
1|4206|Loss: 0.7497068643569946:  65%|██████▌   | 4206/6470 [51:17<26:25,  1.43it/s]
1|4207|Loss: 0.7776899933815002:  65%|██████▌   | 4206/6470 [51:18<26:25,  1.43it/s]
1|4207|Loss: 0.7776899933815002:  65%|██████▌   | 4207/6470 [51:18<26:54,  1.40it/s]
1|4208|Loss: 0.9098649024963379:  65%|██████▌   | 4207/6470 [51:19<26:54,  1.40it/s]
1|4208|Loss: 0.9098649024963379:  65%|██████▌   | 4208/6470 [51:19<27:06,  1.39it/s]
1|4209|Loss: 0.8214302659034729:  65%|██████▌   | 4208/6470 [51:19<27:06,  1.39it/s]
1|4209|Loss: 0.8214302659034729:  65%|██████▌   | 4209/6470 [51:20<27:45,  1.36it/s]
1|4210|Loss: 0.8688448667526245:  65%|██████▌   | 4209/6470 [51:20<27:45,  1.36it/s]
1|4210|Loss: 0.8688448667526245:  65%|██████▌   | 4210/6470 [51:20<26:25,  1.43it/s]
1|4211|Loss: 0.7707350850105286:  65%|██████▌   | 4210/6470 [51:21<26:25,  1.43it/s]
1|4211|Loss: 0.7707350850105286:  65%|██████▌   | 4211/6470 [51:21<25:06,  1.50it/s]
1|4212|Loss: 0.7675043344497681:  65%|██████▌   | 4211/6470 [51:21<25:06,  1.50it/s]
1|4212|Loss: 0.7675043344497681:  65%|██████▌   | 4212/6470 [51:22<25:44,  1.46it/s]
1|4213|Loss: 0.9777662754058838:  65%|██████▌   | 4212/6470 [51:22<25:44,  1.46it/s]
1|4213|Loss: 0.9777662754058838:  65%|██████▌   | 4213/6470 [51:22<26:33,  1.42it/s]
1|4214|Loss: 0.8096944689750671:  65%|██████▌   | 4213/6470 [51:23<26:33,  1.42it/s]
1|4214|Loss: 0.8096944689750671:  65%|██████▌   | 4214/6470 [51:23<28:01,  1.34it/s]
1|4215|Loss: 0.742313027381897:  65%|██████▌   | 4214/6470 [51:24<28:01,  1.34it/s] 
1|4215|Loss: 0.742313027381897:  65%|██████▌   | 4215/6470 [51:24<25:56,  1.45it/s]
1|4216|Loss: 0.9147627353668213:  65%|██████▌   | 4215/6470 [51:24<25:56,  1.45it/s]
1|4216|Loss: 0.9147627353668213:  65%|██████▌   | 4216/6470 [51:25<26:51,  1.40it/s]
1|4217|Loss: 0.7301461696624756:  65%|██████▌   | 4216/6470 [51:25<26:51,  1.40it/s]
1|4217|Loss: 0.7301461696624756:  65%|██████▌   | 4217/6470 [51:25<27:06,  1.39it/s]
1|4218|Loss: 0.8110290765762329:  65%|██████▌   | 4217/6470 [51:26<27:06,  1.39it/s]
1|4218|Loss: 0.8110290765762329:  65%|██████▌   | 4218/6470 [51:26<27:15,  1.38it/s]
1|4219|Loss: 0.8980940580368042:  65%|██████▌   | 4218/6470 [51:26<27:15,  1.38it/s]
1|4219|Loss: 0.8980940580368042:  65%|██████▌   | 4219/6470 [51:26<23:41,  1.58it/s]
1|4220|Loss: 0.9153528213500977:  65%|██████▌   | 4219/6470 [51:27<23:41,  1.58it/s]
1|4220|Loss: 0.9153528213500977:  65%|██████▌   | 4220/6470 [51:27<23:45,  1.58it/s]
1|4221|Loss: 0.8884936571121216:  65%|██████▌   | 4220/6470 [51:28<23:45,  1.58it/s]
1|4221|Loss: 0.8884936571121216:  65%|██████▌   | 4221/6470 [51:28<25:02,  1.50it/s]
1|4222|Loss: 0.9656705260276794:  65%|██████▌   | 4221/6470 [51:28<25:02,  1.50it/s]
1|4222|Loss: 0.9656705260276794:  65%|██████▌   | 4222/6470 [51:29<26:54,  1.39it/s]
1|4223|Loss: 0.7975203394889832:  65%|██████▌   | 4222/6470 [51:29<26:54,  1.39it/s]
1|4223|Loss: 0.7975203394889832:  65%|██████▌   | 4223/6470 [51:29<26:36,  1.41it/s]
1|4224|Loss: 0.8125272989273071:  65%|██████▌   | 4223/6470 [51:30<26:36,  1.41it/s]
1|4224|Loss: 0.8125272989273071:  65%|██████▌   | 4224/6470 [51:30<26:25,  1.42it/s]
1|4225|Loss: 1.0029345750808716:  65%|██████▌   | 4224/6470 [51:30<26:25,  1.42it/s]
1|4225|Loss: 1.0029345750808716:  65%|██████▌   | 4225/6470 [51:31<24:28,  1.53it/s]
1|4226|Loss: 0.9024004340171814:  65%|██████▌   | 4225/6470 [51:31<24:28,  1.53it/s]
1|4226|Loss: 0.9024004340171814:  65%|██████▌   | 4226/6470 [51:31<23:22,  1.60it/s]
1|4227|Loss: 0.9662703275680542:  65%|██████▌   | 4226/6470 [51:32<23:22,  1.60it/s]
1|4227|Loss: 0.9662703275680542:  65%|██████▌   | 4227/6470 [51:32<24:01,  1.56it/s]
1|4228|Loss: 0.9204613566398621:  65%|██████▌   | 4227/6470 [51:32<24:01,  1.56it/s]
1|4228|Loss: 0.9204613566398621:  65%|██████▌   | 4228/6470 [51:33<23:57,  1.56it/s]
1|4229|Loss: 0.9398695826530457:  65%|██████▌   | 4228/6470 [51:33<23:57,  1.56it/s]
1|4229|Loss: 0.9398695826530457:  65%|██████▌   | 4229/6470 [51:33<26:07,  1.43it/s]
1|4230|Loss: 0.8756917119026184:  65%|██████▌   | 4229/6470 [51:34<26:07,  1.43it/s]
1|4230|Loss: 0.8756917119026184:  65%|██████▌   | 4230/6470 [51:34<26:38,  1.40it/s]
1|4231|Loss: 0.8452134728431702:  65%|██████▌   | 4230/6470 [51:35<26:38,  1.40it/s]
1|4231|Loss: 0.8452134728431702:  65%|██████▌   | 4231/6470 [51:35<27:22,  1.36it/s]
1|4232|Loss: 0.9620161652565002:  65%|██████▌   | 4231/6470 [51:35<27:22,  1.36it/s]
1|4232|Loss: 0.9620161652565002:  65%|██████▌   | 4232/6470 [51:36<27:19,  1.36it/s]
1|4233|Loss: 0.7929320931434631:  65%|██████▌   | 4232/6470 [51:36<27:19,  1.36it/s]
1|4233|Loss: 0.7929320931434631:  65%|██████▌   | 4233/6470 [51:36<26:17,  1.42it/s]
1|4234|Loss: 0.963710606098175:  65%|██████▌   | 4233/6470 [51:37<26:17,  1.42it/s] 
1|4234|Loss: 0.963710606098175:  65%|██████▌   | 4234/6470 [51:37<26:23,  1.41it/s]
1|4235|Loss: 0.778408408164978:  65%|██████▌   | 4234/6470 [51:37<26:23,  1.41it/s]
1|4235|Loss: 0.778408408164978:  65%|██████▌   | 4235/6470 [51:38<26:24,  1.41it/s]
1|4236|Loss: 0.8894656300544739:  65%|██████▌   | 4235/6470 [51:38<26:24,  1.41it/s]
1|4236|Loss: 0.8894656300544739:  65%|██████▌   | 4236/6470 [51:38<25:30,  1.46it/s]
1|4237|Loss: 1.0258290767669678:  65%|██████▌   | 4236/6470 [51:39<25:30,  1.46it/s]
1|4237|Loss: 1.0258290767669678:  65%|██████▌   | 4237/6470 [51:39<24:22,  1.53it/s]
1|4238|Loss: 0.7392104268074036:  65%|██████▌   | 4237/6470 [51:39<24:22,  1.53it/s]
1|4238|Loss: 0.7392104268074036:  66%|██████▌   | 4238/6470 [51:39<22:54,  1.62it/s]
1|4239|Loss: 0.8285569548606873:  66%|██████▌   | 4238/6470 [51:40<22:54,  1.62it/s]
1|4239|Loss: 0.8285569548606873:  66%|██████▌   | 4239/6470 [51:40<22:39,  1.64it/s]
1|4240|Loss: 0.9479714035987854:  66%|██████▌   | 4239/6470 [51:40<22:39,  1.64it/s]
1|4240|Loss: 0.9479714035987854:  66%|██████▌   | 4240/6470 [51:40<20:35,  1.80it/s]
1|4241|Loss: 0.9161072373390198:  66%|██████▌   | 4240/6470 [51:41<20:35,  1.80it/s]
1|4241|Loss: 0.9161072373390198:  66%|██████▌   | 4241/6470 [51:41<21:46,  1.71it/s]
1|4242|Loss: 0.8885059952735901:  66%|██████▌   | 4241/6470 [51:42<21:46,  1.71it/s]
1|4242|Loss: 0.8885059952735901:  66%|██████▌   | 4242/6470 [51:42<24:33,  1.51it/s]
1|4243|Loss: 0.678900420665741:  66%|██████▌   | 4242/6470 [51:42<24:33,  1.51it/s] 
1|4243|Loss: 0.678900420665741:  66%|██████▌   | 4243/6470 [51:43<26:46,  1.39it/s]
1|4244|Loss: 0.8462406396865845:  66%|██████▌   | 4243/6470 [51:43<26:46,  1.39it/s]
1|4244|Loss: 0.8462406396865845:  66%|██████▌   | 4244/6470 [51:43<24:44,  1.50it/s]
1|4245|Loss: 0.7974564433097839:  66%|██████▌   | 4244/6470 [51:44<24:44,  1.50it/s]
1|4245|Loss: 0.7974564433097839:  66%|██████▌   | 4245/6470 [51:44<25:57,  1.43it/s]
1|4246|Loss: 0.906505823135376:  66%|██████▌   | 4245/6470 [51:45<25:57,  1.43it/s] 
1|4246|Loss: 0.906505823135376:  66%|██████▌   | 4246/6470 [51:45<27:41,  1.34it/s]
1|4247|Loss: 1.0038543939590454:  66%|██████▌   | 4246/6470 [51:45<27:41,  1.34it/s]
1|4247|Loss: 1.0038543939590454:  66%|██████▌   | 4247/6470 [51:46<27:48,  1.33it/s]
1|4248|Loss: 0.8969524502754211:  66%|██████▌   | 4247/6470 [51:46<27:48,  1.33it/s]
1|4248|Loss: 0.8969524502754211:  66%|██████▌   | 4248/6470 [51:46<26:45,  1.38it/s]
1|4249|Loss: 0.7932328581809998:  66%|██████▌   | 4248/6470 [51:47<26:45,  1.38it/s]
1|4249|Loss: 0.7932328581809998:  66%|██████▌   | 4249/6470 [51:47<25:50,  1.43it/s]
1|4250|Loss: 0.9757384061813354:  66%|██████▌   | 4249/6470 [51:47<25:50,  1.43it/s]
1|4250|Loss: 0.9757384061813354:  66%|██████▌   | 4250/6470 [51:48<24:39,  1.50it/s]
1|4251|Loss: 0.7784036993980408:  66%|██████▌   | 4250/6470 [51:48<24:39,  1.50it/s]
1|4251|Loss: 0.7784036993980408:  66%|██████▌   | 4251/6470 [51:48<24:07,  1.53it/s]
1|4252|Loss: 0.9956339001655579:  66%|██████▌   | 4251/6470 [51:49<24:07,  1.53it/s]
1|4252|Loss: 0.9956339001655579:  66%|██████▌   | 4252/6470 [51:49<24:31,  1.51it/s]
1|4253|Loss: 0.9878731369972229:  66%|██████▌   | 4252/6470 [51:49<24:31,  1.51it/s]
1|4253|Loss: 0.9878731369972229:  66%|██████▌   | 4253/6470 [51:50<25:15,  1.46it/s]
1|4254|Loss: 0.9201521277427673:  66%|██████▌   | 4253/6470 [51:50<25:15,  1.46it/s]
1|4254|Loss: 0.9201521277427673:  66%|██████▌   | 4254/6470 [51:50<24:58,  1.48it/s]
1|4255|Loss: 0.8685348033905029:  66%|██████▌   | 4254/6470 [51:51<24:58,  1.48it/s]
1|4255|Loss: 0.8685348033905029:  66%|██████▌   | 4255/6470 [51:51<26:18,  1.40it/s]
1|4256|Loss: 1.148013949394226:  66%|██████▌   | 4255/6470 [51:52<26:18,  1.40it/s] 
1|4256|Loss: 1.148013949394226:  66%|██████▌   | 4256/6470 [51:52<25:47,  1.43it/s]
1|4257|Loss: 1.0076161623001099:  66%|██████▌   | 4256/6470 [51:52<25:47,  1.43it/s]
1|4257|Loss: 1.0076161623001099:  66%|██████▌   | 4257/6470 [51:53<26:09,  1.41it/s]
1|4258|Loss: 0.817399799823761:  66%|██████▌   | 4257/6470 [51:53<26:09,  1.41it/s] 
1|4258|Loss: 0.817399799823761:  66%|██████▌   | 4258/6470 [51:53<25:29,  1.45it/s]
1|4259|Loss: 0.9328724145889282:  66%|██████▌   | 4258/6470 [51:54<25:29,  1.45it/s]
1|4259|Loss: 0.9328724145889282:  66%|██████▌   | 4259/6470 [51:54<25:23,  1.45it/s]
1|4260|Loss: 0.9348020553588867:  66%|██████▌   | 4259/6470 [51:54<25:23,  1.45it/s]
1|4260|Loss: 0.9348020553588867:  66%|██████▌   | 4260/6470 [51:55<25:57,  1.42it/s]
1|4261|Loss: 0.9369016289710999:  66%|██████▌   | 4260/6470 [51:55<25:57,  1.42it/s]
1|4261|Loss: 0.9369016289710999:  66%|██████▌   | 4261/6470 [51:55<25:35,  1.44it/s]
1|4262|Loss: 0.9565413594245911:  66%|██████▌   | 4261/6470 [51:56<25:35,  1.44it/s]
1|4262|Loss: 0.9565413594245911:  66%|██████▌   | 4262/6470 [51:56<25:55,  1.42it/s]
1|4263|Loss: 0.8960047960281372:  66%|██████▌   | 4262/6470 [51:56<25:55,  1.42it/s]
1|4263|Loss: 0.8960047960281372:  66%|██████▌   | 4263/6470 [51:57<24:20,  1.51it/s]
1|4264|Loss: 0.9646735787391663:  66%|██████▌   | 4263/6470 [51:57<24:20,  1.51it/s]
1|4264|Loss: 0.9646735787391663:  66%|██████▌   | 4264/6470 [51:57<23:13,  1.58it/s]
1|4265|Loss: 0.8417624235153198:  66%|██████▌   | 4264/6470 [51:58<23:13,  1.58it/s]
1|4265|Loss: 0.8417624235153198:  66%|██████▌   | 4265/6470 [51:58<25:21,  1.45it/s]
1|4266|Loss: 0.9326418042182922:  66%|██████▌   | 4265/6470 [51:58<25:21,  1.45it/s]
1|4266|Loss: 0.9326418042182922:  66%|██████▌   | 4266/6470 [51:59<24:18,  1.51it/s]
1|4267|Loss: 0.9155212044715881:  66%|██████▌   | 4266/6470 [51:59<24:18,  1.51it/s]
1|4267|Loss: 0.9155212044715881:  66%|██████▌   | 4267/6470 [51:59<25:08,  1.46it/s]
1|4268|Loss: 0.8518511652946472:  66%|██████▌   | 4267/6470 [52:00<25:08,  1.46it/s]
1|4268|Loss: 0.8518511652946472:  66%|██████▌   | 4268/6470 [52:00<25:35,  1.43it/s]
1|4269|Loss: 0.822188675403595:  66%|██████▌   | 4268/6470 [52:00<25:35,  1.43it/s] 
1|4269|Loss: 0.822188675403595:  66%|██████▌   | 4269/6470 [52:01<26:06,  1.40it/s]
1|4270|Loss: 0.716972827911377:  66%|██████▌   | 4269/6470 [52:01<26:06,  1.40it/s]
1|4270|Loss: 0.716972827911377:  66%|██████▌   | 4270/6470 [52:01<25:43,  1.43it/s]
1|4271|Loss: 0.7394770979881287:  66%|██████▌   | 4270/6470 [52:02<25:43,  1.43it/s]
1|4271|Loss: 0.7394770979881287:  66%|██████▌   | 4271/6470 [52:02<24:28,  1.50it/s]
1|4272|Loss: 0.9159237146377563:  66%|██████▌   | 4271/6470 [52:02<24:28,  1.50it/s]
1|4272|Loss: 0.9159237146377563:  66%|██████▌   | 4272/6470 [52:03<24:49,  1.48it/s]
1|4273|Loss: 0.8332586884498596:  66%|██████▌   | 4272/6470 [52:03<24:49,  1.48it/s]
1|4273|Loss: 0.8332586884498596:  66%|██████▌   | 4273/6470 [52:04<26:38,  1.37it/s]
1|4274|Loss: 0.9467800855636597:  66%|██████▌   | 4273/6470 [52:04<26:38,  1.37it/s]
1|4274|Loss: 0.9467800855636597:  66%|██████▌   | 4274/6470 [52:04<27:08,  1.35it/s]
1|4275|Loss: 0.8446454405784607:  66%|██████▌   | 4274/6470 [52:05<27:08,  1.35it/s]
1|4275|Loss: 0.8446454405784607:  66%|██████▌   | 4275/6470 [52:05<26:46,  1.37it/s]
1|4276|Loss: 0.9840449094772339:  66%|██████▌   | 4275/6470 [52:05<26:46,  1.37it/s]
1|4276|Loss: 0.9840449094772339:  66%|██████▌   | 4276/6470 [52:06<24:45,  1.48it/s]
1|4277|Loss: 0.8605136871337891:  66%|██████▌   | 4276/6470 [52:06<24:45,  1.48it/s]
1|4277|Loss: 0.8605136871337891:  66%|██████▌   | 4277/6470 [52:06<24:04,  1.52it/s]
1|4278|Loss: 0.813692033290863:  66%|██████▌   | 4277/6470 [52:07<24:04,  1.52it/s] 
1|4278|Loss: 0.813692033290863:  66%|██████▌   | 4278/6470 [52:07<25:34,  1.43it/s]
1|4279|Loss: 0.9399060010910034:  66%|██████▌   | 4278/6470 [52:07<25:34,  1.43it/s]
1|4279|Loss: 0.9399060010910034:  66%|██████▌   | 4279/6470 [52:08<26:05,  1.40it/s]
1|4280|Loss: 0.7142160534858704:  66%|██████▌   | 4279/6470 [52:08<26:05,  1.40it/s]
1|4280|Loss: 0.7142160534858704:  66%|██████▌   | 4280/6470 [52:08<26:17,  1.39it/s]
1|4281|Loss: 0.7555787563323975:  66%|██████▌   | 4280/6470 [52:09<26:17,  1.39it/s]
1|4281|Loss: 0.7555787563323975:  66%|██████▌   | 4281/6470 [52:09<24:39,  1.48it/s]
1|4282|Loss: 0.8519728779792786:  66%|██████▌   | 4281/6470 [52:09<24:39,  1.48it/s]
1|4282|Loss: 0.8519728779792786:  66%|██████▌   | 4282/6470 [52:10<24:32,  1.49it/s]
1|4283|Loss: 0.7846452593803406:  66%|██████▌   | 4282/6470 [52:10<24:32,  1.49it/s]
1|4283|Loss: 0.7846452593803406:  66%|██████▌   | 4283/6470 [52:10<24:55,  1.46it/s]
1|4284|Loss: 0.8412712812423706:  66%|██████▌   | 4283/6470 [52:11<24:55,  1.46it/s]
1|4284|Loss: 0.8412712812423706:  66%|██████▌   | 4284/6470 [52:11<25:34,  1.42it/s]
1|4285|Loss: 0.6694578528404236:  66%|██████▌   | 4284/6470 [52:12<25:34,  1.42it/s]
1|4285|Loss: 0.6694578528404236:  66%|██████▌   | 4285/6470 [52:12<26:49,  1.36it/s]
1|4286|Loss: 0.8911071419715881:  66%|██████▌   | 4285/6470 [52:12<26:49,  1.36it/s]
1|4286|Loss: 0.8911071419715881:  66%|██████▌   | 4286/6470 [52:13<25:22,  1.43it/s]
1|4287|Loss: 1.018107295036316:  66%|██████▌   | 4286/6470 [52:13<25:22,  1.43it/s] 
1|4287|Loss: 1.018107295036316:  66%|██████▋   | 4287/6470 [52:13<25:23,  1.43it/s]
1|4288|Loss: 0.7972139120101929:  66%|██████▋   | 4287/6470 [52:14<25:23,  1.43it/s]
1|4288|Loss: 0.7972139120101929:  66%|██████▋   | 4288/6470 [52:14<24:49,  1.46it/s]
1|4289|Loss: 1.07292640209198:  66%|██████▋   | 4288/6470 [52:14<24:49,  1.46it/s]  
1|4289|Loss: 1.07292640209198:  66%|██████▋   | 4289/6470 [52:14<22:06,  1.64it/s]
1|4290|Loss: 0.9049798846244812:  66%|██████▋   | 4289/6470 [52:15<22:06,  1.64it/s]
1|4290|Loss: 0.9049798846244812:  66%|██████▋   | 4290/6470 [52:15<22:34,  1.61it/s]
1|4291|Loss: 0.8028280735015869:  66%|██████▋   | 4290/6470 [52:15<22:34,  1.61it/s]
1|4291|Loss: 0.8028280735015869:  66%|██████▋   | 4291/6470 [52:16<23:25,  1.55it/s]
1|4292|Loss: 1.0085259675979614:  66%|██████▋   | 4291/6470 [52:16<23:25,  1.55it/s]
1|4292|Loss: 1.0085259675979614:  66%|██████▋   | 4292/6470 [52:16<23:17,  1.56it/s]
1|4293|Loss: 1.010271668434143:  66%|██████▋   | 4292/6470 [52:17<23:17,  1.56it/s] 
1|4293|Loss: 1.010271668434143:  66%|██████▋   | 4293/6470 [52:17<22:58,  1.58it/s]
1|4294|Loss: 0.8978291749954224:  66%|██████▋   | 4293/6470 [52:17<22:58,  1.58it/s]
1|4294|Loss: 0.8978291749954224:  66%|██████▋   | 4294/6470 [52:18<22:58,  1.58it/s]
1|4295|Loss: 0.8846130967140198:  66%|██████▋   | 4294/6470 [52:18<22:58,  1.58it/s]
1|4295|Loss: 0.8846130967140198:  66%|██████▋   | 4295/6470 [52:18<23:02,  1.57it/s]
1|4296|Loss: 0.8828520178794861:  66%|██████▋   | 4295/6470 [52:19<23:02,  1.57it/s]
1|4296|Loss: 0.8828520178794861:  66%|██████▋   | 4296/6470 [52:19<21:57,  1.65it/s]
1|4297|Loss: 1.0378868579864502:  66%|██████▋   | 4296/6470 [52:19<21:57,  1.65it/s]
1|4297|Loss: 1.0378868579864502:  66%|██████▋   | 4297/6470 [52:20<23:50,  1.52it/s]
1|4298|Loss: 0.965880274772644:  66%|██████▋   | 4297/6470 [52:20<23:50,  1.52it/s] 
1|4298|Loss: 0.965880274772644:  66%|██████▋   | 4298/6470 [52:20<23:39,  1.53it/s]
1|4299|Loss: 0.8394667506217957:  66%|██████▋   | 4298/6470 [52:21<23:39,  1.53it/s]
1|4299|Loss: 0.8394667506217957:  66%|██████▋   | 4299/6470 [52:21<21:48,  1.66it/s]
1|4300|Loss: 0.9616140723228455:  66%|██████▋   | 4299/6470 [52:21<21:48,  1.66it/s]
1|4300|Loss: 0.9616140723228455:  66%|██████▋   | 4300/6470 [52:21<23:36,  1.53it/s]
1|4301|Loss: 0.7298575639724731:  66%|██████▋   | 4300/6470 [52:22<23:36,  1.53it/s]
1|4301|Loss: 0.7298575639724731:  66%|██████▋   | 4301/6470 [52:22<25:40,  1.41it/s]
1|4302|Loss: 1.0208991765975952:  66%|██████▋   | 4301/6470 [52:23<25:40,  1.41it/s]
1|4302|Loss: 1.0208991765975952:  66%|██████▋   | 4302/6470 [52:23<25:56,  1.39it/s]
1|4303|Loss: 1.0695799589157104:  66%|██████▋   | 4302/6470 [52:23<25:56,  1.39it/s]
1|4303|Loss: 1.0695799589157104:  67%|██████▋   | 4303/6470 [52:24<25:44,  1.40it/s]
1|4304|Loss: 0.7950495481491089:  67%|██████▋   | 4303/6470 [52:24<25:44,  1.40it/s]
1|4304|Loss: 0.7950495481491089:  67%|██████▋   | 4304/6470 [52:25<26:37,  1.36it/s]
1|4305|Loss: 0.8978005647659302:  67%|██████▋   | 4304/6470 [52:25<26:37,  1.36it/s]
1|4305|Loss: 0.8978005647659302:  67%|██████▋   | 4305/6470 [52:25<26:46,  1.35it/s]
1|4306|Loss: 0.907467782497406:  67%|██████▋   | 4305/6470 [52:26<26:46,  1.35it/s] 
1|4306|Loss: 0.907467782497406:  67%|██████▋   | 4306/6470 [52:26<26:26,  1.36it/s]
1|4307|Loss: 0.8793654441833496:  67%|██████▋   | 4306/6470 [52:26<26:26,  1.36it/s]
1|4307|Loss: 0.8793654441833496:  67%|██████▋   | 4307/6470 [52:27<26:06,  1.38it/s]
1|4308|Loss: 0.8963635563850403:  67%|██████▋   | 4307/6470 [52:27<26:06,  1.38it/s]
1|4308|Loss: 0.8963635563850403:  67%|██████▋   | 4308/6470 [52:27<26:05,  1.38it/s]
1|4309|Loss: 0.7581762075424194:  67%|██████▋   | 4308/6470 [52:28<26:05,  1.38it/s]
1|4309|Loss: 0.7581762075424194:  67%|██████▋   | 4309/6470 [52:28<25:13,  1.43it/s]
1|4310|Loss: 0.9731542468070984:  67%|██████▋   | 4309/6470 [52:28<25:13,  1.43it/s]
1|4310|Loss: 0.9731542468070984:  67%|██████▋   | 4310/6470 [52:29<25:01,  1.44it/s]
1|4311|Loss: 0.8044500350952148:  67%|██████▋   | 4310/6470 [52:29<25:01,  1.44it/s]
1|4311|Loss: 0.8044500350952148:  67%|██████▋   | 4311/6470 [52:29<24:37,  1.46it/s]
1|4312|Loss: 0.936002790927887:  67%|██████▋   | 4311/6470 [52:30<24:37,  1.46it/s] 
1|4312|Loss: 0.936002790927887:  67%|██████▋   | 4312/6470 [52:30<25:07,  1.43it/s]
1|4313|Loss: 1.1041903495788574:  67%|██████▋   | 4312/6470 [52:31<25:07,  1.43it/s]
1|4313|Loss: 1.1041903495788574:  67%|██████▋   | 4313/6470 [52:31<26:25,  1.36it/s]
1|4314|Loss: 0.733384907245636:  67%|██████▋   | 4313/6470 [52:31<26:25,  1.36it/s] 
1|4314|Loss: 0.733384907245636:  67%|██████▋   | 4314/6470 [52:32<26:24,  1.36it/s]
1|4315|Loss: 0.7497753500938416:  67%|██████▋   | 4314/6470 [52:32<26:24,  1.36it/s]
1|4315|Loss: 0.7497753500938416:  67%|██████▋   | 4315/6470 [52:33<27:21,  1.31it/s]
1|4316|Loss: 0.8679382801055908:  67%|██████▋   | 4315/6470 [52:33<27:21,  1.31it/s]
1|4316|Loss: 0.8679382801055908:  67%|██████▋   | 4316/6470 [52:33<26:23,  1.36it/s]
1|4317|Loss: 0.7959389090538025:  67%|██████▋   | 4316/6470 [52:34<26:23,  1.36it/s]
1|4317|Loss: 0.7959389090538025:  67%|██████▋   | 4317/6470 [52:34<26:02,  1.38it/s]
1|4318|Loss: 0.9131788015365601:  67%|██████▋   | 4317/6470 [52:34<26:02,  1.38it/s]
1|4318|Loss: 0.9131788015365601:  67%|██████▋   | 4318/6470 [52:35<26:18,  1.36it/s]
1|4319|Loss: 1.0176817178726196:  67%|██████▋   | 4318/6470 [52:35<26:18,  1.36it/s]
1|4319|Loss: 1.0176817178726196:  67%|██████▋   | 4319/6470 [52:35<25:10,  1.42it/s]
1|4320|Loss: 0.9965886473655701:  67%|██████▋   | 4319/6470 [52:36<25:10,  1.42it/s]
1|4320|Loss: 0.9965886473655701:  67%|██████▋   | 4320/6470 [52:36<26:01,  1.38it/s]
1|4321|Loss: 1.0119231939315796:  67%|██████▋   | 4320/6470 [52:36<26:01,  1.38it/s]
1|4321|Loss: 1.0119231939315796:  67%|██████▋   | 4321/6470 [52:37<24:45,  1.45it/s]
1|4322|Loss: 0.8744460344314575:  67%|██████▋   | 4321/6470 [52:37<24:45,  1.45it/s]
1|4322|Loss: 0.8744460344314575:  67%|██████▋   | 4322/6470 [52:37<25:41,  1.39it/s]
1|4323|Loss: 0.9492571353912354:  67%|██████▋   | 4322/6470 [52:38<25:41,  1.39it/s]
1|4323|Loss: 0.9492571353912354:  67%|██████▋   | 4323/6470 [52:38<26:52,  1.33it/s]
1|4324|Loss: 0.8409368395805359:  67%|██████▋   | 4323/6470 [52:39<26:52,  1.33it/s]
1|4324|Loss: 0.8409368395805359:  67%|██████▋   | 4324/6470 [52:39<27:23,  1.31it/s]
1|4325|Loss: 0.80698561668396:  67%|██████▋   | 4324/6470 [52:39<27:23,  1.31it/s]  
1|4325|Loss: 0.80698561668396:  67%|██████▋   | 4325/6470 [52:40<26:12,  1.36it/s]
1|4326|Loss: 0.8232244253158569:  67%|██████▋   | 4325/6470 [52:40<26:12,  1.36it/s]
1|4326|Loss: 0.8232244253158569:  67%|██████▋   | 4326/6470 [52:40<26:26,  1.35it/s]
1|4327|Loss: 0.9090315699577332:  67%|██████▋   | 4326/6470 [52:41<26:26,  1.35it/s]
1|4327|Loss: 0.9090315699577332:  67%|██████▋   | 4327/6470 [52:41<25:45,  1.39it/s]
1|4328|Loss: 0.8056192398071289:  67%|██████▋   | 4327/6470 [52:41<25:45,  1.39it/s]
1|4328|Loss: 0.8056192398071289:  67%|██████▋   | 4328/6470 [52:42<23:18,  1.53it/s]
1|4329|Loss: 0.8479959964752197:  67%|██████▋   | 4328/6470 [52:42<23:18,  1.53it/s]
1|4329|Loss: 0.8479959964752197:  67%|██████▋   | 4329/6470 [52:42<23:34,  1.51it/s]
1|4330|Loss: 0.9785720109939575:  67%|██████▋   | 4329/6470 [52:43<23:34,  1.51it/s]
1|4330|Loss: 0.9785720109939575:  67%|██████▋   | 4330/6470 [52:43<22:40,  1.57it/s]
1|4331|Loss: 0.7745765447616577:  67%|██████▋   | 4330/6470 [52:43<22:40,  1.57it/s]
1|4331|Loss: 0.7745765447616577:  67%|██████▋   | 4331/6470 [52:44<22:28,  1.59it/s]
1|4332|Loss: 0.7754210233688354:  67%|██████▋   | 4331/6470 [52:44<22:28,  1.59it/s]
1|4332|Loss: 0.7754210233688354:  67%|██████▋   | 4332/6470 [52:44<24:28,  1.46it/s]
1|4333|Loss: 0.9147557616233826:  67%|██████▋   | 4332/6470 [52:45<24:28,  1.46it/s]
1|4333|Loss: 0.9147557616233826:  67%|██████▋   | 4333/6470 [52:45<25:00,  1.42it/s]
1|4334|Loss: 0.8654595613479614:  67%|██████▋   | 4333/6470 [52:45<25:00,  1.42it/s]
1|4334|Loss: 0.8654595613479614:  67%|██████▋   | 4334/6470 [52:46<25:02,  1.42it/s]
1|4335|Loss: 0.8784641027450562:  67%|██████▋   | 4334/6470 [52:46<25:02,  1.42it/s]
1|4335|Loss: 0.8784641027450562:  67%|██████▋   | 4335/6470 [52:47<25:26,  1.40it/s]
1|4336|Loss: 0.8166356682777405:  67%|██████▋   | 4335/6470 [52:47<25:26,  1.40it/s]
1|4336|Loss: 0.8166356682777405:  67%|██████▋   | 4336/6470 [52:47<24:45,  1.44it/s]
1|4337|Loss: 0.8730224967002869:  67%|██████▋   | 4336/6470 [52:48<24:45,  1.44it/s]
1|4337|Loss: 0.8730224967002869:  67%|██████▋   | 4337/6470 [52:48<24:40,  1.44it/s]
1|4338|Loss: 0.8298159241676331:  67%|██████▋   | 4337/6470 [52:48<24:40,  1.44it/s]
1|4338|Loss: 0.8298159241676331:  67%|██████▋   | 4338/6470 [52:49<26:42,  1.33it/s]
1|4339|Loss: 0.8458185791969299:  67%|██████▋   | 4338/6470 [52:49<26:42,  1.33it/s]
1|4339|Loss: 0.8458185791969299:  67%|██████▋   | 4339/6470 [52:50<26:39,  1.33it/s]
1|4340|Loss: 0.8627545237541199:  67%|██████▋   | 4339/6470 [52:50<26:39,  1.33it/s]
1|4340|Loss: 0.8627545237541199:  67%|██████▋   | 4340/6470 [52:50<25:35,  1.39it/s]
1|4341|Loss: 0.8864638209342957:  67%|██████▋   | 4340/6470 [52:51<25:35,  1.39it/s]
1|4341|Loss: 0.8864638209342957:  67%|██████▋   | 4341/6470 [52:51<24:29,  1.45it/s]
1|4342|Loss: 0.8404070734977722:  67%|██████▋   | 4341/6470 [52:51<24:29,  1.45it/s]
1|4342|Loss: 0.8404070734977722:  67%|██████▋   | 4342/6470 [52:52<25:34,  1.39it/s]
1|4343|Loss: 1.0720434188842773:  67%|██████▋   | 4342/6470 [52:52<25:34,  1.39it/s]
1|4343|Loss: 1.0720434188842773:  67%|██████▋   | 4343/6470 [52:52<24:25,  1.45it/s]
1|4344|Loss: 0.8601149320602417:  67%|██████▋   | 4343/6470 [52:53<24:25,  1.45it/s]
1|4344|Loss: 0.8601149320602417:  67%|██████▋   | 4344/6470 [52:53<24:17,  1.46it/s]
1|4345|Loss: 1.0183510780334473:  67%|██████▋   | 4344/6470 [52:53<24:17,  1.46it/s]
1|4345|Loss: 1.0183510780334473:  67%|██████▋   | 4345/6470 [52:54<24:29,  1.45it/s]
1|4346|Loss: 0.9617326855659485:  67%|██████▋   | 4345/6470 [52:54<24:29,  1.45it/s]
1|4346|Loss: 0.9617326855659485:  67%|██████▋   | 4346/6470 [52:54<22:29,  1.57it/s]
1|4347|Loss: 1.1185919046401978:  67%|██████▋   | 4346/6470 [52:54<22:29,  1.57it/s]
1|4347|Loss: 1.1185919046401978:  67%|██████▋   | 4347/6470 [52:55<21:11,  1.67it/s]
1|4348|Loss: 0.8324468731880188:  67%|██████▋   | 4347/6470 [52:55<21:11,  1.67it/s]
1|4348|Loss: 0.8324468731880188:  67%|██████▋   | 4348/6470 [52:55<23:11,  1.52it/s]
1|4349|Loss: 1.0164235830307007:  67%|██████▋   | 4348/6470 [52:56<23:11,  1.52it/s]
1|4349|Loss: 1.0164235830307007:  67%|██████▋   | 4349/6470 [52:56<25:04,  1.41it/s]
1|4350|Loss: 0.8119766116142273:  67%|██████▋   | 4349/6470 [52:57<25:04,  1.41it/s]
1|4350|Loss: 0.8119766116142273:  67%|██████▋   | 4350/6470 [52:57<25:41,  1.38it/s]
1|4351|Loss: 0.8364260792732239:  67%|██████▋   | 4350/6470 [52:57<25:41,  1.38it/s]
1|4351|Loss: 0.8364260792732239:  67%|██████▋   | 4351/6470 [52:58<23:27,  1.51it/s]
1|4352|Loss: 0.8718574643135071:  67%|██████▋   | 4351/6470 [52:58<23:27,  1.51it/s]
1|4352|Loss: 0.8718574643135071:  67%|██████▋   | 4352/6470 [52:58<24:34,  1.44it/s]
1|4353|Loss: 0.9066554307937622:  67%|██████▋   | 4352/6470 [52:59<24:34,  1.44it/s]
1|4353|Loss: 0.9066554307937622:  67%|██████▋   | 4353/6470 [52:59<23:24,  1.51it/s]
1|4354|Loss: 0.979947566986084:  67%|██████▋   | 4353/6470 [52:59<23:24,  1.51it/s] 
1|4354|Loss: 0.979947566986084:  67%|██████▋   | 4354/6470 [52:59<22:57,  1.54it/s]
1|4355|Loss: 0.9237354397773743:  67%|██████▋   | 4354/6470 [53:00<22:57,  1.54it/s]
1|4355|Loss: 0.9237354397773743:  67%|██████▋   | 4355/6470 [53:00<24:17,  1.45it/s]
1|4356|Loss: 0.7911974787712097:  67%|██████▋   | 4355/6470 [53:01<24:17,  1.45it/s]
1|4356|Loss: 0.7911974787712097:  67%|██████▋   | 4356/6470 [53:01<24:12,  1.46it/s]
1|4357|Loss: 0.8787904381752014:  67%|██████▋   | 4356/6470 [53:01<24:12,  1.46it/s]
1|4357|Loss: 0.8787904381752014:  67%|██████▋   | 4357/6470 [53:02<24:33,  1.43it/s]
1|4358|Loss: 0.863925576210022:  67%|██████▋   | 4357/6470 [53:02<24:33,  1.43it/s] 
1|4358|Loss: 0.863925576210022:  67%|██████▋   | 4358/6470 [53:02<23:38,  1.49it/s]
1|4359|Loss: 0.8377148509025574:  67%|██████▋   | 4358/6470 [53:03<23:38,  1.49it/s]
1|4359|Loss: 0.8377148509025574:  67%|██████▋   | 4359/6470 [53:03<23:41,  1.48it/s]
1|4360|Loss: 0.7840312719345093:  67%|██████▋   | 4359/6470 [53:03<23:41,  1.48it/s]
1|4360|Loss: 0.7840312719345093:  67%|██████▋   | 4360/6470 [53:04<26:06,  1.35it/s]
1|4361|Loss: 0.9615281224250793:  67%|██████▋   | 4360/6470 [53:04<26:06,  1.35it/s]
1|4361|Loss: 0.9615281224250793:  67%|██████▋   | 4361/6470 [53:05<27:01,  1.30it/s]
1|4362|Loss: 0.8884136080741882:  67%|██████▋   | 4361/6470 [53:05<27:01,  1.30it/s]
1|4362|Loss: 0.8884136080741882:  67%|██████▋   | 4362/6470 [53:05<25:16,  1.39it/s]
1|4363|Loss: 0.7778186202049255:  67%|██████▋   | 4362/6470 [53:06<25:16,  1.39it/s]
1|4363|Loss: 0.7778186202049255:  67%|██████▋   | 4363/6470 [53:06<25:08,  1.40it/s]
1|4364|Loss: 0.8781776428222656:  67%|██████▋   | 4363/6470 [53:06<25:08,  1.40it/s]
1|4364|Loss: 0.8781776428222656:  67%|██████▋   | 4364/6470 [53:07<25:58,  1.35it/s]
1|4365|Loss: 0.8975105285644531:  67%|██████▋   | 4364/6470 [53:07<25:58,  1.35it/s]
1|4365|Loss: 0.8975105285644531:  67%|██████▋   | 4365/6470 [53:07<24:19,  1.44it/s]
1|4366|Loss: 0.7969657182693481:  67%|██████▋   | 4365/6470 [53:08<24:19,  1.44it/s]
1|4366|Loss: 0.7969657182693481:  67%|██████▋   | 4366/6470 [53:08<24:14,  1.45it/s]
1|4367|Loss: 0.9558682441711426:  67%|██████▋   | 4366/6470 [53:08<24:14,  1.45it/s]
1|4367|Loss: 0.9558682441711426:  67%|██████▋   | 4367/6470 [53:09<23:24,  1.50it/s]
1|4368|Loss: 0.7275769114494324:  67%|██████▋   | 4367/6470 [53:09<23:24,  1.50it/s]
1|4368|Loss: 0.7275769114494324:  68%|██████▊   | 4368/6470 [53:09<23:46,  1.47it/s]
1|4369|Loss: 0.7590185403823853:  68%|██████▊   | 4368/6470 [53:10<23:46,  1.47it/s]
1|4369|Loss: 0.7590185403823853:  68%|██████▊   | 4369/6470 [53:10<23:23,  1.50it/s]
1|4370|Loss: 0.8978214859962463:  68%|██████▊   | 4369/6470 [53:10<23:23,  1.50it/s]
1|4370|Loss: 0.8978214859962463:  68%|██████▊   | 4370/6470 [53:11<23:31,  1.49it/s]
1|4371|Loss: 0.962810218334198:  68%|██████▊   | 4370/6470 [53:11<23:31,  1.49it/s] 
1|4371|Loss: 0.962810218334198:  68%|██████▊   | 4371/6470 [53:12<25:16,  1.38it/s]
1|4372|Loss: 0.8052807450294495:  68%|██████▊   | 4371/6470 [53:12<25:16,  1.38it/s]
1|4372|Loss: 0.8052807450294495:  68%|██████▊   | 4372/6470 [53:12<25:32,  1.37it/s]
1|4373|Loss: 0.9049555063247681:  68%|██████▊   | 4372/6470 [53:13<25:32,  1.37it/s]
1|4373|Loss: 0.9049555063247681:  68%|██████▊   | 4373/6470 [53:13<24:07,  1.45it/s]
1|4374|Loss: 0.9232043623924255:  68%|██████▊   | 4373/6470 [53:13<24:07,  1.45it/s]
1|4374|Loss: 0.9232043623924255:  68%|██████▊   | 4374/6470 [53:13<22:51,  1.53it/s]
1|4375|Loss: 1.0651758909225464:  68%|██████▊   | 4374/6470 [53:14<22:51,  1.53it/s]
1|4375|Loss: 1.0651758909225464:  68%|██████▊   | 4375/6470 [53:14<22:36,  1.54it/s]
1|4376|Loss: 0.8361368179321289:  68%|██████▊   | 4375/6470 [53:14<22:36,  1.54it/s]
1|4376|Loss: 0.8361368179321289:  68%|██████▊   | 4376/6470 [53:15<23:20,  1.50it/s]
1|4377|Loss: 0.8538641333580017:  68%|██████▊   | 4376/6470 [53:15<23:20,  1.50it/s]
1|4377|Loss: 0.8538641333580017:  68%|██████▊   | 4377/6470 [53:15<23:14,  1.50it/s]
1|4378|Loss: 0.900317370891571:  68%|██████▊   | 4377/6470 [53:16<23:14,  1.50it/s] 
1|4378|Loss: 0.900317370891571:  68%|██████▊   | 4378/6470 [53:16<22:47,  1.53it/s]
1|4379|Loss: 0.8523812890052795:  68%|██████▊   | 4378/6470 [53:16<22:47,  1.53it/s]
1|4379|Loss: 0.8523812890052795:  68%|██████▊   | 4379/6470 [53:17<23:06,  1.51it/s]
1|4380|Loss: 1.0331122875213623:  68%|██████▊   | 4379/6470 [53:17<23:06,  1.51it/s]
1|4380|Loss: 1.0331122875213623:  68%|██████▊   | 4380/6470 [53:17<23:07,  1.51it/s]
1|4381|Loss: 0.9392561316490173:  68%|██████▊   | 4380/6470 [53:18<23:07,  1.51it/s]
1|4381|Loss: 0.9392561316490173:  68%|██████▊   | 4381/6470 [53:18<23:46,  1.46it/s]
1|4382|Loss: 0.7738423347473145:  68%|██████▊   | 4381/6470 [53:19<23:46,  1.46it/s]
1|4382|Loss: 0.7738423347473145:  68%|██████▊   | 4382/6470 [53:19<25:35,  1.36it/s]
1|4383|Loss: 0.9156642556190491:  68%|██████▊   | 4382/6470 [53:19<25:35,  1.36it/s]
1|4383|Loss: 0.9156642556190491:  68%|██████▊   | 4383/6470 [53:20<25:54,  1.34it/s]
1|4384|Loss: 0.9276429414749146:  68%|██████▊   | 4383/6470 [53:20<25:54,  1.34it/s]
1|4384|Loss: 0.9276429414749146:  68%|██████▊   | 4384/6470 [53:20<24:49,  1.40it/s]
1|4385|Loss: 0.8742498755455017:  68%|██████▊   | 4384/6470 [53:21<24:49,  1.40it/s]
1|4385|Loss: 0.8742498755455017:  68%|██████▊   | 4385/6470 [53:21<25:32,  1.36it/s]
1|4386|Loss: 0.930952250957489:  68%|██████▊   | 4385/6470 [53:22<25:32,  1.36it/s] 
1|4386|Loss: 0.930952250957489:  68%|██████▊   | 4386/6470 [53:22<25:47,  1.35it/s]
1|4387|Loss: 0.9142651557922363:  68%|██████▊   | 4386/6470 [53:22<25:47,  1.35it/s]
1|4387|Loss: 0.9142651557922363:  68%|██████▊   | 4387/6470 [53:23<26:16,  1.32it/s]
1|4388|Loss: 0.8962962627410889:  68%|██████▊   | 4387/6470 [53:23<26:16,  1.32it/s]
1|4388|Loss: 0.8962962627410889:  68%|██████▊   | 4388/6470 [53:24<27:18,  1.27it/s]
1|4389|Loss: 0.7766557335853577:  68%|██████▊   | 4388/6470 [53:24<27:18,  1.27it/s]
1|4389|Loss: 0.7766557335853577:  68%|██████▊   | 4389/6470 [53:24<27:04,  1.28it/s]
1|4390|Loss: 0.9705711007118225:  68%|██████▊   | 4389/6470 [53:25<27:04,  1.28it/s]
1|4390|Loss: 0.9705711007118225:  68%|██████▊   | 4390/6470 [53:25<24:17,  1.43it/s]
1|4391|Loss: 0.9731239676475525:  68%|██████▊   | 4390/6470 [53:25<24:17,  1.43it/s]
1|4391|Loss: 0.9731239676475525:  68%|██████▊   | 4391/6470 [53:25<22:28,  1.54it/s]
1|4392|Loss: 0.8559911251068115:  68%|██████▊   | 4391/6470 [53:26<22:28,  1.54it/s]
1|4392|Loss: 0.8559911251068115:  68%|██████▊   | 4392/6470 [53:26<23:44,  1.46it/s]
1|4393|Loss: 0.8433375954627991:  68%|██████▊   | 4392/6470 [53:27<23:44,  1.46it/s]
1|4393|Loss: 0.8433375954627991:  68%|██████▊   | 4393/6470 [53:27<25:10,  1.37it/s]
1|4394|Loss: 0.7557651996612549:  68%|██████▊   | 4393/6470 [53:27<25:10,  1.37it/s]
1|4394|Loss: 0.7557651996612549:  68%|██████▊   | 4394/6470 [53:28<25:21,  1.36it/s]
1|4395|Loss: 0.8896290063858032:  68%|██████▊   | 4394/6470 [53:28<25:21,  1.36it/s]
1|4395|Loss: 0.8896290063858032:  68%|██████▊   | 4395/6470 [53:29<25:09,  1.37it/s]
1|4396|Loss: 0.8350324034690857:  68%|██████▊   | 4395/6470 [53:29<25:09,  1.37it/s]
1|4396|Loss: 0.8350324034690857:  68%|██████▊   | 4396/6470 [53:29<24:59,  1.38it/s]
1|4397|Loss: 0.9026892781257629:  68%|██████▊   | 4396/6470 [53:30<24:59,  1.38it/s]
1|4397|Loss: 0.9026892781257629:  68%|██████▊   | 4397/6470 [53:30<25:50,  1.34it/s]
1|4398|Loss: 0.9352620840072632:  68%|██████▊   | 4397/6470 [53:30<25:50,  1.34it/s]
1|4398|Loss: 0.9352620840072632:  68%|██████▊   | 4398/6470 [53:31<26:43,  1.29it/s]
1|4399|Loss: 0.9260894656181335:  68%|██████▊   | 4398/6470 [53:31<26:43,  1.29it/s]
1|4399|Loss: 0.9260894656181335:  68%|██████▊   | 4399/6470 [53:32<25:36,  1.35it/s]
1|4400|Loss: 0.8295087218284607:  68%|██████▊   | 4399/6470 [53:32<25:36,  1.35it/s]
1|4400|Loss: 0.8295087218284607:  68%|██████▊   | 4400/6470 [53:32<23:58,  1.44it/s]
1|4401|Loss: 0.8270487189292908:  68%|██████▊   | 4400/6470 [53:32<23:58,  1.44it/s]
1|4401|Loss: 0.8270487189292908:  68%|██████▊   | 4401/6470 [53:33<23:18,  1.48it/s]
1|4402|Loss: 1.0024162530899048:  68%|██████▊   | 4401/6470 [53:33<23:18,  1.48it/s]
1|4402|Loss: 1.0024162530899048:  68%|██████▊   | 4402/6470 [53:34<24:47,  1.39it/s]
1|4403|Loss: 0.8199300169944763:  68%|██████▊   | 4402/6470 [53:34<24:47,  1.39it/s]
1|4403|Loss: 0.8199300169944763:  68%|██████▊   | 4403/6470 [53:34<24:59,  1.38it/s]
1|4404|Loss: 0.7970350384712219:  68%|██████▊   | 4403/6470 [53:35<24:59,  1.38it/s]
1|4404|Loss: 0.7970350384712219:  68%|██████▊   | 4404/6470 [53:35<24:14,  1.42it/s]
1|4405|Loss: 0.7813811302185059:  68%|██████▊   | 4404/6470 [53:35<24:14,  1.42it/s]
1|4405|Loss: 0.7813811302185059:  68%|██████▊   | 4405/6470 [53:36<25:07,  1.37it/s]
1|4406|Loss: 0.9311648011207581:  68%|██████▊   | 4405/6470 [53:36<25:07,  1.37it/s]
1|4406|Loss: 0.9311648011207581:  68%|██████▊   | 4406/6470 [53:37<26:36,  1.29it/s]
1|4407|Loss: 0.9058840870857239:  68%|██████▊   | 4406/6470 [53:37<26:36,  1.29it/s]
1|4407|Loss: 0.9058840870857239:  68%|██████▊   | 4407/6470 [53:37<25:17,  1.36it/s]
1|4408|Loss: 0.8384754657745361:  68%|██████▊   | 4407/6470 [53:38<25:17,  1.36it/s]
1|4408|Loss: 0.8384754657745361:  68%|██████▊   | 4408/6470 [53:38<24:08,  1.42it/s]
1|4409|Loss: 0.8121232986450195:  68%|██████▊   | 4408/6470 [53:38<24:08,  1.42it/s]
1|4409|Loss: 0.8121232986450195:  68%|██████▊   | 4409/6470 [53:39<24:04,  1.43it/s]
1|4410|Loss: 0.7863025665283203:  68%|██████▊   | 4409/6470 [53:39<24:04,  1.43it/s]
1|4410|Loss: 0.7863025665283203:  68%|██████▊   | 4410/6470 [53:39<25:51,  1.33it/s]
1|4411|Loss: 0.9689732193946838:  68%|██████▊   | 4410/6470 [53:40<25:51,  1.33it/s]
1|4411|Loss: 0.9689732193946838:  68%|██████▊   | 4411/6470 [53:40<25:20,  1.35it/s]
1|4412|Loss: 0.7555137276649475:  68%|██████▊   | 4411/6470 [53:41<25:20,  1.35it/s]
1|4412|Loss: 0.7555137276649475:  68%|██████▊   | 4412/6470 [53:41<24:52,  1.38it/s]
1|4413|Loss: 0.8950033783912659:  68%|██████▊   | 4412/6470 [53:41<24:52,  1.38it/s]
1|4413|Loss: 0.8950033783912659:  68%|██████▊   | 4413/6470 [53:42<24:08,  1.42it/s]
1|4414|Loss: 0.9281589388847351:  68%|██████▊   | 4413/6470 [53:42<24:08,  1.42it/s]
1|4414|Loss: 0.9281589388847351:  68%|██████▊   | 4414/6470 [53:42<22:12,  1.54it/s]
1|4415|Loss: 0.8887622356414795:  68%|██████▊   | 4414/6470 [53:42<22:12,  1.54it/s]
1|4415|Loss: 0.8887622356414795:  68%|██████▊   | 4415/6470 [53:43<22:17,  1.54it/s]
1|4416|Loss: 0.9391404390335083:  68%|██████▊   | 4415/6470 [53:43<22:17,  1.54it/s]
1|4416|Loss: 0.9391404390335083:  68%|██████▊   | 4416/6470 [53:43<20:47,  1.65it/s]
1|4417|Loss: 0.8643659353256226:  68%|██████▊   | 4416/6470 [53:44<20:47,  1.65it/s]
1|4417|Loss: 0.8643659353256226:  68%|██████▊   | 4417/6470 [53:44<20:33,  1.66it/s]
1|4418|Loss: 0.9904764890670776:  68%|██████▊   | 4417/6470 [53:44<20:33,  1.66it/s]
1|4418|Loss: 0.9904764890670776:  68%|██████▊   | 4418/6470 [53:45<22:12,  1.54it/s]
1|4419|Loss: 0.8437049984931946:  68%|██████▊   | 4418/6470 [53:45<22:12,  1.54it/s]
1|4419|Loss: 0.8437049984931946:  68%|██████▊   | 4419/6470 [53:45<19:48,  1.73it/s]
1|4420|Loss: 0.8676994442939758:  68%|██████▊   | 4419/6470 [53:45<19:48,  1.73it/s]
1|4420|Loss: 0.8676994442939758:  68%|██████▊   | 4420/6470 [53:46<21:40,  1.58it/s]
1|4421|Loss: 1.0584819316864014:  68%|██████▊   | 4420/6470 [53:46<21:40,  1.58it/s]
1|4421|Loss: 1.0584819316864014:  68%|██████▊   | 4421/6470 [53:46<22:56,  1.49it/s]
1|4422|Loss: 0.8592700958251953:  68%|██████▊   | 4421/6470 [53:47<22:56,  1.49it/s]
1|4422|Loss: 0.8592700958251953:  68%|██████▊   | 4422/6470 [53:47<23:36,  1.45it/s]
1|4423|Loss: 0.8344959020614624:  68%|██████▊   | 4422/6470 [53:48<23:36,  1.45it/s]
1|4423|Loss: 0.8344959020614624:  68%|██████▊   | 4423/6470 [53:48<22:35,  1.51it/s]
1|4424|Loss: 0.8460910320281982:  68%|██████▊   | 4423/6470 [53:48<22:35,  1.51it/s]
1|4424|Loss: 0.8460910320281982:  68%|██████▊   | 4424/6470 [53:48<22:15,  1.53it/s]
1|4425|Loss: 0.8787953853607178:  68%|██████▊   | 4424/6470 [53:49<22:15,  1.53it/s]
1|4425|Loss: 0.8787953853607178:  68%|██████▊   | 4425/6470 [53:49<23:01,  1.48it/s]
1|4426|Loss: 0.7985825538635254:  68%|██████▊   | 4425/6470 [53:50<23:01,  1.48it/s]
1|4426|Loss: 0.7985825538635254:  68%|██████▊   | 4426/6470 [53:50<23:27,  1.45it/s]
1|4427|Loss: 0.8013604283332825:  68%|██████▊   | 4426/6470 [53:50<23:27,  1.45it/s]
1|4427|Loss: 0.8013604283332825:  68%|██████▊   | 4427/6470 [53:51<23:34,  1.44it/s]
1|4428|Loss: 0.8369620442390442:  68%|██████▊   | 4427/6470 [53:51<23:34,  1.44it/s]
1|4428|Loss: 0.8369620442390442:  68%|██████▊   | 4428/6470 [53:51<24:33,  1.39it/s]
1|4429|Loss: 0.8010491728782654:  68%|██████▊   | 4428/6470 [53:52<24:33,  1.39it/s]
1|4429|Loss: 0.8010491728782654:  68%|██████▊   | 4429/6470 [53:52<26:43,  1.27it/s]
1|4430|Loss: 0.8508646488189697:  68%|██████▊   | 4429/6470 [53:53<26:43,  1.27it/s]
1|4430|Loss: 0.8508646488189697:  68%|██████▊   | 4430/6470 [53:53<28:34,  1.19it/s]
1|4431|Loss: 0.87685626745224:  68%|██████▊   | 4430/6470 [53:54<28:34,  1.19it/s]  
1|4431|Loss: 0.87685626745224:  68%|██████▊   | 4431/6470 [53:54<27:17,  1.25it/s]
1|4432|Loss: 0.9240691661834717:  68%|██████▊   | 4431/6470 [53:54<27:17,  1.25it/s]
1|4432|Loss: 0.9240691661834717:  69%|██████▊   | 4432/6470 [53:55<24:52,  1.37it/s]
1|4433|Loss: 0.8713054656982422:  69%|██████▊   | 4432/6470 [53:55<24:52,  1.37it/s]
1|4433|Loss: 0.8713054656982422:  69%|██████▊   | 4433/6470 [53:55<23:16,  1.46it/s]
1|4434|Loss: 0.9178429245948792:  69%|██████▊   | 4433/6470 [53:56<23:16,  1.46it/s]
1|4434|Loss: 0.9178429245948792:  69%|██████▊   | 4434/6470 [53:56<23:25,  1.45it/s]
1|4435|Loss: 0.9139417409896851:  69%|██████▊   | 4434/6470 [53:56<23:25,  1.45it/s]
1|4435|Loss: 0.9139417409896851:  69%|██████▊   | 4435/6470 [53:57<23:38,  1.43it/s]
1|4436|Loss: 0.8644551634788513:  69%|██████▊   | 4435/6470 [53:57<23:38,  1.43it/s]
1|4436|Loss: 0.8644551634788513:  69%|██████▊   | 4436/6470 [53:57<22:07,  1.53it/s]
1|4437|Loss: 0.6498979926109314:  69%|██████▊   | 4436/6470 [53:57<22:07,  1.53it/s]
1|4437|Loss: 0.6498979926109314:  69%|██████▊   | 4437/6470 [53:58<20:42,  1.64it/s]
1|4438|Loss: 0.9442899227142334:  69%|██████▊   | 4437/6470 [53:58<20:42,  1.64it/s]
1|4438|Loss: 0.9442899227142334:  69%|██████▊   | 4438/6470 [53:58<20:33,  1.65it/s]
1|4439|Loss: 0.9292105436325073:  69%|██████▊   | 4438/6470 [53:59<20:33,  1.65it/s]
1|4439|Loss: 0.9292105436325073:  69%|██████▊   | 4439/6470 [53:59<21:39,  1.56it/s]
1|4440|Loss: 0.8026872277259827:  69%|██████▊   | 4439/6470 [53:59<21:39,  1.56it/s]
1|4440|Loss: 0.8026872277259827:  69%|██████▊   | 4440/6470 [54:00<21:05,  1.60it/s]
1|4441|Loss: 0.8348469138145447:  69%|██████▊   | 4440/6470 [54:00<21:05,  1.60it/s]
1|4441|Loss: 0.8348469138145447:  69%|██████▊   | 4441/6470 [54:00<20:33,  1.64it/s]
1|4442|Loss: 0.7974638342857361:  69%|██████▊   | 4441/6470 [54:01<20:33,  1.64it/s]
1|4442|Loss: 0.7974638342857361:  69%|██████▊   | 4442/6470 [54:01<21:58,  1.54it/s]
1|4443|Loss: 0.8341847062110901:  69%|██████▊   | 4442/6470 [54:01<21:58,  1.54it/s]
1|4443|Loss: 0.8341847062110901:  69%|██████▊   | 4443/6470 [54:02<22:40,  1.49it/s]
1|4444|Loss: 1.0233567953109741:  69%|██████▊   | 4443/6470 [54:02<22:40,  1.49it/s]
1|4444|Loss: 1.0233567953109741:  69%|██████▊   | 4444/6470 [54:02<23:00,  1.47it/s]
1|4445|Loss: 0.7430759072303772:  69%|██████▊   | 4444/6470 [54:03<23:00,  1.47it/s]
1|4445|Loss: 0.7430759072303772:  69%|██████▊   | 4445/6470 [54:03<23:13,  1.45it/s]
1|4446|Loss: 0.9347018003463745:  69%|██████▊   | 4445/6470 [54:03<23:13,  1.45it/s]
1|4446|Loss: 0.9347018003463745:  69%|██████▊   | 4446/6470 [54:04<23:36,  1.43it/s]
1|4447|Loss: 0.7783463001251221:  69%|██████▊   | 4446/6470 [54:04<23:36,  1.43it/s]
1|4447|Loss: 0.7783463001251221:  69%|██████▊   | 4447/6470 [54:04<22:41,  1.49it/s]
1|4448|Loss: 1.0086932182312012:  69%|██████▊   | 4447/6470 [54:05<22:41,  1.49it/s]
1|4448|Loss: 1.0086932182312012:  69%|██████▊   | 4448/6470 [54:05<22:03,  1.53it/s]
1|4449|Loss: 0.8497886657714844:  69%|██████▊   | 4448/6470 [54:05<22:03,  1.53it/s]
1|4449|Loss: 0.8497886657714844:  69%|██████▉   | 4449/6470 [54:06<22:20,  1.51it/s]
1|4450|Loss: 0.8686273694038391:  69%|██████▉   | 4449/6470 [54:06<22:20,  1.51it/s]
1|4450|Loss: 0.8686273694038391:  69%|██████▉   | 4450/6470 [54:06<21:28,  1.57it/s]
1|4451|Loss: 0.9814825654029846:  69%|██████▉   | 4450/6470 [54:07<21:28,  1.57it/s]
1|4451|Loss: 0.9814825654029846:  69%|██████▉   | 4451/6470 [54:07<22:55,  1.47it/s]
1|4452|Loss: 0.800102710723877:  69%|██████▉   | 4451/6470 [54:07<22:55,  1.47it/s] 
1|4452|Loss: 0.800102710723877:  69%|██████▉   | 4452/6470 [54:08<22:00,  1.53it/s]
1|4453|Loss: 0.889266312122345:  69%|██████▉   | 4452/6470 [54:08<22:00,  1.53it/s]
1|4453|Loss: 0.889266312122345:  69%|██████▉   | 4453/6470 [54:08<22:08,  1.52it/s]
1|4454|Loss: 0.9322779178619385:  69%|██████▉   | 4453/6470 [54:09<22:08,  1.52it/s]
1|4454|Loss: 0.9322779178619385:  69%|██████▉   | 4454/6470 [54:09<22:00,  1.53it/s]
1|4455|Loss: 0.8716958165168762:  69%|██████▉   | 4454/6470 [54:09<22:00,  1.53it/s]
1|4455|Loss: 0.8716958165168762:  69%|██████▉   | 4455/6470 [54:10<24:35,  1.37it/s]
1|4456|Loss: 0.7825829386711121:  69%|██████▉   | 4455/6470 [54:10<24:35,  1.37it/s]
1|4456|Loss: 0.7825829386711121:  69%|██████▉   | 4456/6470 [54:10<24:00,  1.40it/s]
1|4457|Loss: 0.7737136483192444:  69%|██████▉   | 4456/6470 [54:11<24:00,  1.40it/s]
1|4457|Loss: 0.7737136483192444:  69%|██████▉   | 4457/6470 [54:11<24:59,  1.34it/s]
1|4458|Loss: 0.9936097860336304:  69%|██████▉   | 4457/6470 [54:12<24:59,  1.34it/s]
1|4458|Loss: 0.9936097860336304:  69%|██████▉   | 4458/6470 [54:12<24:24,  1.37it/s]
1|4459|Loss: 1.019757866859436:  69%|██████▉   | 4458/6470 [54:12<24:24,  1.37it/s] 
1|4459|Loss: 1.019757866859436:  69%|██████▉   | 4459/6470 [54:13<24:13,  1.38it/s]
1|4460|Loss: 0.9870529174804688:  69%|██████▉   | 4459/6470 [54:13<24:13,  1.38it/s]
1|4460|Loss: 0.9870529174804688:  69%|██████▉   | 4460/6470 [54:13<23:00,  1.46it/s]
1|4461|Loss: 0.9909631609916687:  69%|██████▉   | 4460/6470 [54:14<23:00,  1.46it/s]
1|4461|Loss: 0.9909631609916687:  69%|██████▉   | 4461/6470 [54:14<22:56,  1.46it/s]
1|4462|Loss: 0.7612046599388123:  69%|██████▉   | 4461/6470 [54:14<22:56,  1.46it/s]
1|4462|Loss: 0.7612046599388123:  69%|██████▉   | 4462/6470 [54:15<22:25,  1.49it/s]
1|4463|Loss: 0.8018254041671753:  69%|██████▉   | 4462/6470 [54:15<22:25,  1.49it/s]
1|4463|Loss: 0.8018254041671753:  69%|██████▉   | 4463/6470 [54:15<23:56,  1.40it/s]
1|4464|Loss: 0.8861873745918274:  69%|██████▉   | 4463/6470 [54:16<23:56,  1.40it/s]
1|4464|Loss: 0.8861873745918274:  69%|██████▉   | 4464/6470 [54:16<23:19,  1.43it/s]
1|4465|Loss: 0.8089049458503723:  69%|██████▉   | 4464/6470 [54:16<23:19,  1.43it/s]
1|4465|Loss: 0.8089049458503723:  69%|██████▉   | 4465/6470 [54:17<23:27,  1.42it/s]
1|4466|Loss: 0.7758461833000183:  69%|██████▉   | 4465/6470 [54:17<23:27,  1.42it/s]
1|4466|Loss: 0.7758461833000183:  69%|██████▉   | 4466/6470 [54:17<23:04,  1.45it/s]
1|4467|Loss: 0.8851132392883301:  69%|██████▉   | 4466/6470 [54:18<23:04,  1.45it/s]
1|4467|Loss: 0.8851132392883301:  69%|██████▉   | 4467/6470 [54:18<23:01,  1.45it/s]
1|4468|Loss: 0.8753570914268494:  69%|██████▉   | 4467/6470 [54:19<23:01,  1.45it/s]
1|4468|Loss: 0.8753570914268494:  69%|██████▉   | 4468/6470 [54:19<23:22,  1.43it/s]
1|4469|Loss: 0.7715845704078674:  69%|██████▉   | 4468/6470 [54:19<23:22,  1.43it/s]
1|4469|Loss: 0.7715845704078674:  69%|██████▉   | 4469/6470 [54:20<23:41,  1.41it/s]
1|4470|Loss: 0.9434933066368103:  69%|██████▉   | 4469/6470 [54:20<23:41,  1.41it/s]
1|4470|Loss: 0.9434933066368103:  69%|██████▉   | 4470/6470 [54:20<24:44,  1.35it/s]
1|4471|Loss: 0.8595921397209167:  69%|██████▉   | 4470/6470 [54:21<24:44,  1.35it/s]
1|4471|Loss: 0.8595921397209167:  69%|██████▉   | 4471/6470 [54:21<25:33,  1.30it/s]
1|4472|Loss: 0.8811254501342773:  69%|██████▉   | 4471/6470 [54:22<25:33,  1.30it/s]
1|4472|Loss: 0.8811254501342773:  69%|██████▉   | 4472/6470 [54:22<24:34,  1.36it/s]
1|4473|Loss: 0.9129372239112854:  69%|██████▉   | 4472/6470 [54:22<24:34,  1.36it/s]
1|4473|Loss: 0.9129372239112854:  69%|██████▉   | 4473/6470 [54:23<23:30,  1.42it/s]
1|4474|Loss: 0.8315808176994324:  69%|██████▉   | 4473/6470 [54:23<23:30,  1.42it/s]
1|4474|Loss: 0.8315808176994324:  69%|██████▉   | 4474/6470 [54:23<22:52,  1.45it/s]
1|4475|Loss: 0.8033081889152527:  69%|██████▉   | 4474/6470 [54:24<22:52,  1.45it/s]
1|4475|Loss: 0.8033081889152527:  69%|██████▉   | 4475/6470 [54:24<22:24,  1.48it/s]
1|4476|Loss: 0.8890020251274109:  69%|██████▉   | 4475/6470 [54:24<22:24,  1.48it/s]
1|4476|Loss: 0.8890020251274109:  69%|██████▉   | 4476/6470 [54:24<21:39,  1.53it/s]
1|4477|Loss: 0.8054784536361694:  69%|██████▉   | 4476/6470 [54:25<21:39,  1.53it/s]
1|4477|Loss: 0.8054784536361694:  69%|██████▉   | 4477/6470 [54:25<21:27,  1.55it/s]
1|4478|Loss: 0.670437216758728:  69%|██████▉   | 4477/6470 [54:25<21:27,  1.55it/s] 
1|4478|Loss: 0.670437216758728:  69%|██████▉   | 4478/6470 [54:26<21:13,  1.56it/s]
1|4479|Loss: 0.8080295324325562:  69%|██████▉   | 4478/6470 [54:26<21:13,  1.56it/s]
1|4479|Loss: 0.8080295324325562:  69%|██████▉   | 4479/6470 [54:26<22:44,  1.46it/s]
1|4480|Loss: 0.7932259440422058:  69%|██████▉   | 4479/6470 [54:27<22:44,  1.46it/s]
1|4480|Loss: 0.7932259440422058:  69%|██████▉   | 4480/6470 [54:27<23:10,  1.43it/s]
1|4481|Loss: 0.8785254955291748:  69%|██████▉   | 4480/6470 [54:28<23:10,  1.43it/s]
1|4481|Loss: 0.8785254955291748:  69%|██████▉   | 4481/6470 [54:28<23:16,  1.42it/s]
1|4482|Loss: 0.820464551448822:  69%|██████▉   | 4481/6470 [54:28<23:16,  1.42it/s] 
1|4482|Loss: 0.820464551448822:  69%|██████▉   | 4482/6470 [54:29<23:15,  1.43it/s]
1|4483|Loss: 0.9283361434936523:  69%|██████▉   | 4482/6470 [54:29<23:15,  1.43it/s]
1|4483|Loss: 0.9283361434936523:  69%|██████▉   | 4483/6470 [54:29<24:06,  1.37it/s]
1|4484|Loss: 0.7791478037834167:  69%|██████▉   | 4483/6470 [54:30<24:06,  1.37it/s]
1|4484|Loss: 0.7791478037834167:  69%|██████▉   | 4484/6470 [54:30<24:44,  1.34it/s]
1|4485|Loss: 0.8324199318885803:  69%|██████▉   | 4484/6470 [54:31<24:44,  1.34it/s]
1|4485|Loss: 0.8324199318885803:  69%|██████▉   | 4485/6470 [54:31<22:31,  1.47it/s]
1|4486|Loss: 1.0123618841171265:  69%|██████▉   | 4485/6470 [54:31<22:31,  1.47it/s]
1|4486|Loss: 1.0123618841171265:  69%|██████▉   | 4486/6470 [54:32<23:26,  1.41it/s]
1|4487|Loss: 0.9475047588348389:  69%|██████▉   | 4486/6470 [54:32<23:26,  1.41it/s]
1|4487|Loss: 0.9475047588348389:  69%|██████▉   | 4487/6470 [54:32<22:52,  1.44it/s]
1|4488|Loss: 0.887661874294281:  69%|██████▉   | 4487/6470 [54:33<22:52,  1.44it/s] 
1|4488|Loss: 0.887661874294281:  69%|██████▉   | 4488/6470 [54:33<23:32,  1.40it/s]
1|4489|Loss: 0.9313220977783203:  69%|██████▉   | 4488/6470 [54:33<23:32,  1.40it/s]
1|4489|Loss: 0.9313220977783203:  69%|██████▉   | 4489/6470 [54:34<23:08,  1.43it/s]
1|4490|Loss: 0.9164184927940369:  69%|██████▉   | 4489/6470 [54:34<23:08,  1.43it/s]
1|4490|Loss: 0.9164184927940369:  69%|██████▉   | 4490/6470 [54:34<21:27,  1.54it/s]
1|4491|Loss: 0.8543407320976257:  69%|██████▉   | 4490/6470 [54:35<21:27,  1.54it/s]
1|4491|Loss: 0.8543407320976257:  69%|██████▉   | 4491/6470 [54:35<22:14,  1.48it/s]
1|4492|Loss: 0.9509066343307495:  69%|██████▉   | 4491/6470 [54:35<22:14,  1.48it/s]
1|4492|Loss: 0.9509066343307495:  69%|██████▉   | 4492/6470 [54:35<20:59,  1.57it/s]
1|4493|Loss: 0.8773933053016663:  69%|██████▉   | 4492/6470 [54:36<20:59,  1.57it/s]
1|4493|Loss: 0.8773933053016663:  69%|██████▉   | 4493/6470 [54:36<21:10,  1.56it/s]
1|4494|Loss: 0.9753498435020447:  69%|██████▉   | 4493/6470 [54:36<21:10,  1.56it/s]
1|4494|Loss: 0.9753498435020447:  69%|██████▉   | 4494/6470 [54:37<21:18,  1.55it/s]
1|4495|Loss: 0.8568440079689026:  69%|██████▉   | 4494/6470 [54:37<21:18,  1.55it/s]
1|4495|Loss: 0.8568440079689026:  69%|██████▉   | 4495/6470 [54:37<20:42,  1.59it/s]
1|4496|Loss: 0.8870067596435547:  69%|██████▉   | 4495/6470 [54:38<20:42,  1.59it/s]
1|4496|Loss: 0.8870067596435547:  69%|██████▉   | 4496/6470 [54:38<19:31,  1.69it/s]
1|4497|Loss: 1.1645268201828003:  69%|██████▉   | 4496/6470 [54:38<19:31,  1.69it/s]
1|4497|Loss: 1.1645268201828003:  70%|██████▉   | 4497/6470 [54:38<19:29,  1.69it/s]
1|4498|Loss: 0.9874220490455627:  70%|██████▉   | 4497/6470 [54:39<19:29,  1.69it/s]
1|4498|Loss: 0.9874220490455627:  70%|██████▉   | 4498/6470 [54:39<19:01,  1.73it/s]
1|4499|Loss: 0.9478515386581421:  70%|██████▉   | 4498/6470 [54:39<19:01,  1.73it/s]
1|4499|Loss: 0.9478515386581421:  70%|██████▉   | 4499/6470 [54:40<21:04,  1.56it/s]
1|4500|Loss: 0.9046405553817749:  70%|██████▉   | 4499/6470 [54:40<21:04,  1.56it/s]
1|4500|Loss: 0.9046405553817749:  70%|██████▉   | 4500/6470 [54:40<21:36,  1.52it/s]
1|4501|Loss: 0.9183365702629089:  70%|██████▉   | 4500/6470 [54:41<21:36,  1.52it/s]
1|4501|Loss: 0.9183365702629089:  70%|██████▉   | 4501/6470 [54:41<22:09,  1.48it/s]
1|4502|Loss: 0.8215013146400452:  70%|██████▉   | 4501/6470 [54:42<22:09,  1.48it/s]
1|4502|Loss: 0.8215013146400452:  70%|██████▉   | 4502/6470 [54:42<22:30,  1.46it/s]
1|4503|Loss: 0.7842968702316284:  70%|██████▉   | 4502/6470 [54:42<22:30,  1.46it/s]
1|4503|Loss: 0.7842968702316284:  70%|██████▉   | 4503/6470 [54:43<22:40,  1.45it/s]
1|4504|Loss: 0.9580658674240112:  70%|██████▉   | 4503/6470 [54:43<22:40,  1.45it/s]
1|4504|Loss: 0.9580658674240112:  70%|██████▉   | 4504/6470 [54:43<23:09,  1.42it/s]
1|4505|Loss: 0.8623490333557129:  70%|██████▉   | 4504/6470 [54:44<23:09,  1.42it/s]
1|4505|Loss: 0.8623490333557129:  70%|██████▉   | 4505/6470 [54:44<24:26,  1.34it/s]
1|4506|Loss: 0.8721397519111633:  70%|██████▉   | 4505/6470 [54:45<24:26,  1.34it/s]
1|4506|Loss: 0.8721397519111633:  70%|██████▉   | 4506/6470 [54:45<23:26,  1.40it/s]
1|4507|Loss: 0.8005480170249939:  70%|██████▉   | 4506/6470 [54:45<23:26,  1.40it/s]
1|4507|Loss: 0.8005480170249939:  70%|██████▉   | 4507/6470 [54:46<24:26,  1.34it/s]
1|4508|Loss: 0.8768772482872009:  70%|██████▉   | 4507/6470 [54:46<24:26,  1.34it/s]
1|4508|Loss: 0.8768772482872009:  70%|██████▉   | 4508/6470 [54:46<23:56,  1.37it/s]
1|4509|Loss: 0.9523309469223022:  70%|██████▉   | 4508/6470 [54:47<23:56,  1.37it/s]
1|4509|Loss: 0.9523309469223022:  70%|██████▉   | 4509/6470 [54:47<23:02,  1.42it/s]
1|4510|Loss: 0.8685768246650696:  70%|██████▉   | 4509/6470 [54:47<23:02,  1.42it/s]
1|4510|Loss: 0.8685768246650696:  70%|██████▉   | 4510/6470 [54:48<23:07,  1.41it/s]
1|4511|Loss: 0.8214259743690491:  70%|██████▉   | 4510/6470 [54:48<23:07,  1.41it/s]
1|4511|Loss: 0.8214259743690491:  70%|██████▉   | 4511/6470 [54:48<22:31,  1.45it/s]
1|4512|Loss: 0.8006295561790466:  70%|██████▉   | 4511/6470 [54:49<22:31,  1.45it/s]
1|4512|Loss: 0.8006295561790466:  70%|██████▉   | 4512/6470 [54:49<23:52,  1.37it/s]
1|4513|Loss: 0.8757451772689819:  70%|██████▉   | 4512/6470 [54:50<23:52,  1.37it/s]
1|4513|Loss: 0.8757451772689819:  70%|██████▉   | 4513/6470 [54:50<23:08,  1.41it/s]
1|4514|Loss: 0.8581340909004211:  70%|██████▉   | 4513/6470 [54:50<23:08,  1.41it/s]
1|4514|Loss: 0.8581340909004211:  70%|██████▉   | 4514/6470 [54:51<23:02,  1.42it/s]
1|4515|Loss: 0.8863973021507263:  70%|██████▉   | 4514/6470 [54:51<23:02,  1.42it/s]
1|4515|Loss: 0.8863973021507263:  70%|██████▉   | 4515/6470 [54:51<22:23,  1.46it/s]
1|4516|Loss: 0.8079893589019775:  70%|██████▉   | 4515/6470 [54:52<22:23,  1.46it/s]
1|4516|Loss: 0.8079893589019775:  70%|██████▉   | 4516/6470 [54:52<23:07,  1.41it/s]
1|4517|Loss: 0.9141645431518555:  70%|██████▉   | 4516/6470 [54:52<23:07,  1.41it/s]
1|4517|Loss: 0.9141645431518555:  70%|██████▉   | 4517/6470 [54:53<22:39,  1.44it/s]
1|4518|Loss: 0.9485483169555664:  70%|██████▉   | 4517/6470 [54:53<22:39,  1.44it/s]
1|4518|Loss: 0.9485483169555664:  70%|██████▉   | 4518/6470 [54:53<21:54,  1.49it/s]
1|4519|Loss: 0.7405433654785156:  70%|██████▉   | 4518/6470 [54:54<21:54,  1.49it/s]
1|4519|Loss: 0.7405433654785156:  70%|██████▉   | 4519/6470 [54:54<21:21,  1.52it/s]
1|4520|Loss: 0.9901585578918457:  70%|██████▉   | 4519/6470 [54:54<21:21,  1.52it/s]
1|4520|Loss: 0.9901585578918457:  70%|██████▉   | 4520/6470 [54:55<22:01,  1.48it/s]
1|4521|Loss: 0.8221287727355957:  70%|██████▉   | 4520/6470 [54:55<22:01,  1.48it/s]
1|4521|Loss: 0.8221287727355957:  70%|██████▉   | 4521/6470 [54:55<21:43,  1.50it/s]
1|4522|Loss: 0.9329293966293335:  70%|██████▉   | 4521/6470 [54:56<21:43,  1.50it/s]
1|4522|Loss: 0.9329293966293335:  70%|██████▉   | 4522/6470 [54:56<21:23,  1.52it/s]
1|4523|Loss: 0.8271725177764893:  70%|██████▉   | 4522/6470 [54:56<21:23,  1.52it/s]
1|4523|Loss: 0.8271725177764893:  70%|██████▉   | 4523/6470 [54:56<21:05,  1.54it/s]
1|4524|Loss: 0.8849340677261353:  70%|██████▉   | 4523/6470 [54:57<21:05,  1.54it/s]
1|4524|Loss: 0.8849340677261353:  70%|██████▉   | 4524/6470 [54:57<20:45,  1.56it/s]
1|4525|Loss: 0.8739844560623169:  70%|██████▉   | 4524/6470 [54:57<20:45,  1.56it/s]
1|4525|Loss: 0.8739844560623169:  70%|██████▉   | 4525/6470 [54:58<20:43,  1.56it/s]
1|4526|Loss: 0.7518068552017212:  70%|██████▉   | 4525/6470 [54:58<20:43,  1.56it/s]
1|4526|Loss: 0.7518068552017212:  70%|██████▉   | 4526/6470 [54:58<21:17,  1.52it/s]
1|4527|Loss: 0.7827739119529724:  70%|██████▉   | 4526/6470 [54:59<21:17,  1.52it/s]
1|4527|Loss: 0.7827739119529724:  70%|██████▉   | 4527/6470 [54:59<21:15,  1.52it/s]
1|4528|Loss: 0.9335656762123108:  70%|██████▉   | 4527/6470 [54:59<21:15,  1.52it/s]
1|4528|Loss: 0.9335656762123108:  70%|██████▉   | 4528/6470 [55:00<21:33,  1.50it/s]
1|4529|Loss: 0.8749340772628784:  70%|██████▉   | 4528/6470 [55:00<21:33,  1.50it/s]
1|4529|Loss: 0.8749340772628784:  70%|███████   | 4529/6470 [55:00<21:26,  1.51it/s]
1|4530|Loss: 0.7415632009506226:  70%|███████   | 4529/6470 [55:01<21:26,  1.51it/s]
1|4530|Loss: 0.7415632009506226:  70%|███████   | 4530/6470 [55:01<20:41,  1.56it/s]
1|4531|Loss: 0.8018621206283569:  70%|███████   | 4530/6470 [55:01<20:41,  1.56it/s]
1|4531|Loss: 0.8018621206283569:  70%|███████   | 4531/6470 [55:02<20:03,  1.61it/s]
1|4532|Loss: 0.8818016648292542:  70%|███████   | 4531/6470 [55:02<20:03,  1.61it/s]
1|4532|Loss: 0.8818016648292542:  70%|███████   | 4532/6470 [55:02<20:50,  1.55it/s]
1|4533|Loss: 0.9323683381080627:  70%|███████   | 4532/6470 [55:03<20:50,  1.55it/s]
1|4533|Loss: 0.9323683381080627:  70%|███████   | 4533/6470 [55:03<21:42,  1.49it/s]
1|4534|Loss: 0.9930609464645386:  70%|███████   | 4533/6470 [55:03<21:42,  1.49it/s]
1|4534|Loss: 0.9930609464645386:  70%|███████   | 4534/6470 [55:04<22:20,  1.44it/s]
1|4535|Loss: 0.9474940299987793:  70%|███████   | 4534/6470 [55:04<22:20,  1.44it/s]
1|4535|Loss: 0.9474940299987793:  70%|███████   | 4535/6470 [55:04<22:03,  1.46it/s]
1|4536|Loss: 0.9024746417999268:  70%|███████   | 4535/6470 [55:05<22:03,  1.46it/s]
1|4536|Loss: 0.9024746417999268:  70%|███████   | 4536/6470 [55:05<22:20,  1.44it/s]
1|4537|Loss: 0.8028650283813477:  70%|███████   | 4536/6470 [55:06<22:20,  1.44it/s]
1|4537|Loss: 0.8028650283813477:  70%|███████   | 4537/6470 [55:06<22:28,  1.43it/s]
1|4538|Loss: 0.8036162853240967:  70%|███████   | 4537/6470 [55:06<22:28,  1.43it/s]
1|4538|Loss: 0.8036162853240967:  70%|███████   | 4538/6470 [55:06<21:52,  1.47it/s]
1|4539|Loss: 0.7877128720283508:  70%|███████   | 4538/6470 [55:07<21:52,  1.47it/s]
1|4539|Loss: 0.7877128720283508:  70%|███████   | 4539/6470 [55:07<22:08,  1.45it/s]
1|4540|Loss: 0.9973011016845703:  70%|███████   | 4539/6470 [55:08<22:08,  1.45it/s]
1|4540|Loss: 0.9973011016845703:  70%|███████   | 4540/6470 [55:08<21:42,  1.48it/s]
1|4541|Loss: 0.8461278676986694:  70%|███████   | 4540/6470 [55:08<21:42,  1.48it/s]
1|4541|Loss: 0.8461278676986694:  70%|███████   | 4541/6470 [55:08<20:54,  1.54it/s]
1|4542|Loss: 0.985321581363678:  70%|███████   | 4541/6470 [55:09<20:54,  1.54it/s] 
1|4542|Loss: 0.985321581363678:  70%|███████   | 4542/6470 [55:09<21:24,  1.50it/s]
1|4543|Loss: 0.7780624628067017:  70%|███████   | 4542/6470 [55:09<21:24,  1.50it/s]
1|4543|Loss: 0.7780624628067017:  70%|███████   | 4543/6470 [55:10<20:39,  1.55it/s]
1|4544|Loss: 0.7531249523162842:  70%|███████   | 4543/6470 [55:10<20:39,  1.55it/s]
1|4544|Loss: 0.7531249523162842:  70%|███████   | 4544/6470 [55:10<21:39,  1.48it/s]
1|4545|Loss: 0.7663373351097107:  70%|███████   | 4544/6470 [55:11<21:39,  1.48it/s]
1|4545|Loss: 0.7663373351097107:  70%|███████   | 4545/6470 [55:11<20:45,  1.55it/s]
1|4546|Loss: 0.8347588181495667:  70%|███████   | 4545/6470 [55:11<20:45,  1.55it/s]
1|4546|Loss: 0.8347588181495667:  70%|███████   | 4546/6470 [55:12<21:09,  1.52it/s]
1|4547|Loss: 0.9120720028877258:  70%|███████   | 4546/6470 [55:12<21:09,  1.52it/s]
1|4547|Loss: 0.9120720028877258:  70%|███████   | 4547/6470 [55:12<22:00,  1.46it/s]
1|4548|Loss: 0.7826002836227417:  70%|███████   | 4547/6470 [55:13<22:00,  1.46it/s]
1|4548|Loss: 0.7826002836227417:  70%|███████   | 4548/6470 [55:13<22:16,  1.44it/s]
1|4549|Loss: 0.8688260316848755:  70%|███████   | 4548/6470 [55:14<22:16,  1.44it/s]
1|4549|Loss: 0.8688260316848755:  70%|███████   | 4549/6470 [55:14<23:30,  1.36it/s]
1|4550|Loss: 0.8724309802055359:  70%|███████   | 4549/6470 [55:14<23:30,  1.36it/s]
1|4550|Loss: 0.8724309802055359:  70%|███████   | 4550/6470 [55:15<23:09,  1.38it/s]
1|4551|Loss: 1.030456781387329:  70%|███████   | 4550/6470 [55:15<23:09,  1.38it/s] 
1|4551|Loss: 1.030456781387329:  70%|███████   | 4551/6470 [55:15<22:54,  1.40it/s]
1|4552|Loss: 0.8594261407852173:  70%|███████   | 4551/6470 [55:16<22:54,  1.40it/s]
1|4552|Loss: 0.8594261407852173:  70%|███████   | 4552/6470 [55:16<22:51,  1.40it/s]
1|4553|Loss: 0.9291123151779175:  70%|███████   | 4552/6470 [55:17<22:51,  1.40it/s]
1|4553|Loss: 0.9291123151779175:  70%|███████   | 4553/6470 [55:17<22:55,  1.39it/s]
1|4554|Loss: 0.7922617793083191:  70%|███████   | 4553/6470 [55:17<22:55,  1.39it/s]
1|4554|Loss: 0.7922617793083191:  70%|███████   | 4554/6470 [55:18<23:13,  1.38it/s]
1|4555|Loss: 0.914610743522644:  70%|███████   | 4554/6470 [55:18<23:13,  1.38it/s] 
1|4555|Loss: 0.914610743522644:  70%|███████   | 4555/6470 [55:18<24:05,  1.32it/s]
1|4556|Loss: 0.9396297931671143:  70%|███████   | 4555/6470 [55:19<24:05,  1.32it/s]
1|4556|Loss: 0.9396297931671143:  70%|███████   | 4556/6470 [55:19<24:58,  1.28it/s]
1|4557|Loss: 0.9646593928337097:  70%|███████   | 4556/6470 [55:20<24:58,  1.28it/s]
1|4557|Loss: 0.9646593928337097:  70%|███████   | 4557/6470 [55:20<23:01,  1.38it/s]
1|4558|Loss: 0.941794216632843:  70%|███████   | 4557/6470 [55:20<23:01,  1.38it/s] 
1|4558|Loss: 0.941794216632843:  70%|███████   | 4558/6470 [55:20<21:29,  1.48it/s]
1|4559|Loss: 0.7327790260314941:  70%|███████   | 4558/6470 [55:21<21:29,  1.48it/s]
1|4559|Loss: 0.7327790260314941:  70%|███████   | 4559/6470 [55:21<21:52,  1.46it/s]
1|4560|Loss: 0.7451683878898621:  70%|███████   | 4559/6470 [55:22<21:52,  1.46it/s]
1|4560|Loss: 0.7451683878898621:  70%|███████   | 4560/6470 [55:22<23:43,  1.34it/s]
1|4561|Loss: 0.7384729385375977:  70%|███████   | 4560/6470 [55:23<23:43,  1.34it/s]
1|4561|Loss: 0.7384729385375977:  70%|███████   | 4561/6470 [55:23<25:02,  1.27it/s]
1|4562|Loss: 0.944969892501831:  70%|███████   | 4561/6470 [55:23<25:02,  1.27it/s] 
1|4562|Loss: 0.944969892501831:  71%|███████   | 4562/6470 [55:24<24:09,  1.32it/s]
1|4563|Loss: 0.8403223156929016:  71%|███████   | 4562/6470 [55:24<24:09,  1.32it/s]
1|4563|Loss: 0.8403223156929016:  71%|███████   | 4563/6470 [55:24<23:29,  1.35it/s]
1|4564|Loss: 1.008158802986145:  71%|███████   | 4563/6470 [55:25<23:29,  1.35it/s] 
1|4564|Loss: 1.008158802986145:  71%|███████   | 4564/6470 [55:25<23:32,  1.35it/s]
1|4565|Loss: 0.8236511945724487:  71%|███████   | 4564/6470 [55:25<23:32,  1.35it/s]
1|4565|Loss: 0.8236511945724487:  71%|███████   | 4565/6470 [55:26<23:15,  1.36it/s]
1|4566|Loss: 0.984177827835083:  71%|███████   | 4565/6470 [55:26<23:15,  1.36it/s] 
1|4566|Loss: 0.984177827835083:  71%|███████   | 4566/6470 [55:26<23:12,  1.37it/s]
1|4567|Loss: 0.7775609493255615:  71%|███████   | 4566/6470 [55:27<23:12,  1.37it/s]
1|4567|Loss: 0.7775609493255615:  71%|███████   | 4567/6470 [55:27<23:12,  1.37it/s]
1|4568|Loss: 1.1822624206542969:  71%|███████   | 4567/6470 [55:28<23:12,  1.37it/s]
1|4568|Loss: 1.1822624206542969:  71%|███████   | 4568/6470 [55:28<22:21,  1.42it/s]
1|4569|Loss: 0.8018593788146973:  71%|███████   | 4568/6470 [55:28<22:21,  1.42it/s]
1|4569|Loss: 0.8018593788146973:  71%|███████   | 4569/6470 [55:29<23:06,  1.37it/s]
1|4570|Loss: 0.697326123714447:  71%|███████   | 4569/6470 [55:29<23:06,  1.37it/s] 
1|4570|Loss: 0.697326123714447:  71%|███████   | 4570/6470 [55:29<22:09,  1.43it/s]
1|4571|Loss: 0.6928633451461792:  71%|███████   | 4570/6470 [55:30<22:09,  1.43it/s]
1|4571|Loss: 0.6928633451461792:  71%|███████   | 4571/6470 [55:30<22:14,  1.42it/s]
1|4572|Loss: 0.9240357279777527:  71%|███████   | 4571/6470 [55:30<22:14,  1.42it/s]
1|4572|Loss: 0.9240357279777527:  71%|███████   | 4572/6470 [55:31<22:32,  1.40it/s]
1|4573|Loss: 0.9064918756484985:  71%|███████   | 4572/6470 [55:31<22:32,  1.40it/s]
1|4573|Loss: 0.9064918756484985:  71%|███████   | 4573/6470 [55:31<22:45,  1.39it/s]
1|4574|Loss: 1.0195366144180298:  71%|███████   | 4573/6470 [55:32<22:45,  1.39it/s]
1|4574|Loss: 1.0195366144180298:  71%|███████   | 4574/6470 [55:32<22:52,  1.38it/s]
1|4575|Loss: 0.8920031189918518:  71%|███████   | 4574/6470 [55:33<22:52,  1.38it/s]
1|4575|Loss: 0.8920031189918518:  71%|███████   | 4575/6470 [55:33<22:42,  1.39it/s]
1|4576|Loss: 0.950166642665863:  71%|███████   | 4575/6470 [55:33<22:42,  1.39it/s] 
1|4576|Loss: 0.950166642665863:  71%|███████   | 4576/6470 [55:34<23:43,  1.33it/s]
1|4577|Loss: 0.844283938407898:  71%|███████   | 4576/6470 [55:34<23:43,  1.33it/s]
1|4577|Loss: 0.844283938407898:  71%|███████   | 4577/6470 [55:35<24:35,  1.28it/s]
1|4578|Loss: 1.1068917512893677:  71%|███████   | 4577/6470 [55:35<24:35,  1.28it/s]
1|4578|Loss: 1.1068917512893677:  71%|███████   | 4578/6470 [55:35<25:16,  1.25it/s]
1|4579|Loss: 0.7712344527244568:  71%|███████   | 4578/6470 [55:36<25:16,  1.25it/s]
1|4579|Loss: 0.7712344527244568:  71%|███████   | 4579/6470 [55:36<24:42,  1.28it/s]
1|4580|Loss: 0.9667055606842041:  71%|███████   | 4579/6470 [55:37<24:42,  1.28it/s]
1|4580|Loss: 0.9667055606842041:  71%|███████   | 4580/6470 [55:37<23:19,  1.35it/s]
1|4581|Loss: 0.8296443223953247:  71%|███████   | 4580/6470 [55:37<23:19,  1.35it/s]
1|4581|Loss: 0.8296443223953247:  71%|███████   | 4581/6470 [55:37<22:22,  1.41it/s]
1|4582|Loss: 0.8499690294265747:  71%|███████   | 4581/6470 [55:38<22:22,  1.41it/s]
1|4582|Loss: 0.8499690294265747:  71%|███████   | 4582/6470 [55:38<19:44,  1.59it/s]
1|4583|Loss: 0.9297304749488831:  71%|███████   | 4582/6470 [55:38<19:44,  1.59it/s]
1|4583|Loss: 0.9297304749488831:  71%|███████   | 4583/6470 [55:38<18:55,  1.66it/s]
1|4584|Loss: 0.8250615000724792:  71%|███████   | 4583/6470 [55:39<18:55,  1.66it/s]
1|4584|Loss: 0.8250615000724792:  71%|███████   | 4584/6470 [55:39<19:34,  1.61it/s]
1|4585|Loss: 0.898271381855011:  71%|███████   | 4584/6470 [55:39<19:34,  1.61it/s] 
1|4585|Loss: 0.898271381855011:  71%|███████   | 4585/6470 [55:40<19:40,  1.60it/s]
1|4586|Loss: 0.920398473739624:  71%|███████   | 4585/6470 [55:40<19:40,  1.60it/s]
1|4586|Loss: 0.920398473739624:  71%|███████   | 4586/6470 [55:40<20:12,  1.55it/s]
1|4587|Loss: 1.0395864248275757:  71%|███████   | 4586/6470 [55:41<20:12,  1.55it/s]
1|4587|Loss: 1.0395864248275757:  71%|███████   | 4587/6470 [55:41<20:47,  1.51it/s]
1|4588|Loss: 0.8716154098510742:  71%|███████   | 4587/6470 [55:41<20:47,  1.51it/s]
1|4588|Loss: 0.8716154098510742:  71%|███████   | 4588/6470 [55:42<18:33,  1.69it/s]
1|4589|Loss: 0.7974830865859985:  71%|███████   | 4588/6470 [55:42<18:33,  1.69it/s]
1|4589|Loss: 0.7974830865859985:  71%|███████   | 4589/6470 [55:42<18:22,  1.71it/s]
1|4590|Loss: 0.8600720167160034:  71%|███████   | 4589/6470 [55:43<18:22,  1.71it/s]
1|4590|Loss: 0.8600720167160034:  71%|███████   | 4590/6470 [55:43<20:52,  1.50it/s]
1|4591|Loss: 0.8722589612007141:  71%|███████   | 4590/6470 [55:43<20:52,  1.50it/s]
1|4591|Loss: 0.8722589612007141:  71%|███████   | 4591/6470 [55:44<20:32,  1.52it/s]
1|4592|Loss: 0.8080199956893921:  71%|███████   | 4591/6470 [55:44<20:32,  1.52it/s]
1|4592|Loss: 0.8080199956893921:  71%|███████   | 4592/6470 [55:44<21:07,  1.48it/s]
1|4593|Loss: 0.9020265936851501:  71%|███████   | 4592/6470 [55:45<21:07,  1.48it/s]
1|4593|Loss: 0.9020265936851501:  71%|███████   | 4593/6470 [55:45<20:32,  1.52it/s]
1|4594|Loss: 0.7944474220275879:  71%|███████   | 4593/6470 [55:45<20:32,  1.52it/s]
1|4594|Loss: 0.7944474220275879:  71%|███████   | 4594/6470 [55:46<22:04,  1.42it/s]
1|4595|Loss: 1.0612422227859497:  71%|███████   | 4594/6470 [55:46<22:04,  1.42it/s]
1|4595|Loss: 1.0612422227859497:  71%|███████   | 4595/6470 [55:47<22:39,  1.38it/s]
1|4596|Loss: 0.9226373434066772:  71%|███████   | 4595/6470 [55:47<22:39,  1.38it/s]
1|4596|Loss: 0.9226373434066772:  71%|███████   | 4596/6470 [55:47<21:02,  1.48it/s]
1|4597|Loss: 0.9819161295890808:  71%|███████   | 4596/6470 [55:47<21:02,  1.48it/s]
1|4597|Loss: 0.9819161295890808:  71%|███████   | 4597/6470 [55:48<22:36,  1.38it/s]
1|4598|Loss: 0.9018753170967102:  71%|███████   | 4597/6470 [55:48<22:36,  1.38it/s]
1|4598|Loss: 0.9018753170967102:  71%|███████   | 4598/6470 [55:48<20:28,  1.52it/s]
1|4599|Loss: 0.9068735241889954:  71%|███████   | 4598/6470 [55:49<20:28,  1.52it/s]
1|4599|Loss: 0.9068735241889954:  71%|███████   | 4599/6470 [55:49<19:27,  1.60it/s]
1|4600|Loss: 0.9645358324050903:  71%|███████   | 4599/6470 [55:49<19:27,  1.60it/s]
1|4600|Loss: 0.9645358324050903:  71%|███████   | 4600/6470 [55:50<20:18,  1.53it/s]
1|4601|Loss: 0.8708403706550598:  71%|███████   | 4600/6470 [55:50<20:18,  1.53it/s]
1|4601|Loss: 0.8708403706550598:  71%|███████   | 4601/6470 [55:50<20:00,  1.56it/s]
1|4602|Loss: 0.9372519850730896:  71%|███████   | 4601/6470 [55:51<20:00,  1.56it/s]
1|4602|Loss: 0.9372519850730896:  71%|███████   | 4602/6470 [55:51<21:29,  1.45it/s]
1|4603|Loss: 0.9392922520637512:  71%|███████   | 4602/6470 [55:52<21:29,  1.45it/s]
1|4603|Loss: 0.9392922520637512:  71%|███████   | 4603/6470 [55:52<21:55,  1.42it/s]
1|4604|Loss: 0.8967755436897278:  71%|███████   | 4603/6470 [55:52<21:55,  1.42it/s]
1|4604|Loss: 0.8967755436897278:  71%|███████   | 4604/6470 [55:53<22:17,  1.40it/s]
1|4605|Loss: 0.9178298711776733:  71%|███████   | 4604/6470 [55:53<22:17,  1.40it/s]
1|4605|Loss: 0.9178298711776733:  71%|███████   | 4605/6470 [55:53<22:39,  1.37it/s]
1|4606|Loss: 0.9680280089378357:  71%|███████   | 4605/6470 [55:54<22:39,  1.37it/s]
1|4606|Loss: 0.9680280089378357:  71%|███████   | 4606/6470 [55:54<24:03,  1.29it/s]
1|4607|Loss: 1.043440580368042:  71%|███████   | 4606/6470 [55:55<24:03,  1.29it/s] 
1|4607|Loss: 1.043440580368042:  71%|███████   | 4607/6470 [55:55<23:08,  1.34it/s]
1|4608|Loss: 0.8863426446914673:  71%|███████   | 4607/6470 [55:55<23:08,  1.34it/s]
1|4608|Loss: 0.8863426446914673:  71%|███████   | 4608/6470 [55:56<22:46,  1.36it/s]
1|4609|Loss: 0.8683573007583618:  71%|███████   | 4608/6470 [55:56<22:46,  1.36it/s]
1|4609|Loss: 0.8683573007583618:  71%|███████   | 4609/6470 [55:56<22:35,  1.37it/s]
1|4610|Loss: 0.812255859375:  71%|███████   | 4609/6470 [55:57<22:35,  1.37it/s]    
1|4610|Loss: 0.812255859375:  71%|███████▏  | 4610/6470 [55:57<21:55,  1.41it/s]
1|4611|Loss: 0.9284393191337585:  71%|███████▏  | 4610/6470 [55:57<21:55,  1.41it/s]
1|4611|Loss: 0.9284393191337585:  71%|███████▏  | 4611/6470 [55:58<21:18,  1.45it/s]
1|4612|Loss: 1.1286218166351318:  71%|███████▏  | 4611/6470 [55:58<21:18,  1.45it/s]
1|4612|Loss: 1.1286218166351318:  71%|███████▏  | 4612/6470 [55:58<21:41,  1.43it/s]
1|4613|Loss: 0.8380368947982788:  71%|███████▏  | 4612/6470 [55:59<21:41,  1.43it/s]
1|4613|Loss: 0.8380368947982788:  71%|███████▏  | 4613/6470 [55:59<22:27,  1.38it/s]
1|4614|Loss: 0.6972355246543884:  71%|███████▏  | 4613/6470 [56:00<22:27,  1.38it/s]
1|4614|Loss: 0.6972355246543884:  71%|███████▏  | 4614/6470 [56:00<22:44,  1.36it/s]
1|4615|Loss: 0.814628541469574:  71%|███████▏  | 4614/6470 [56:00<22:44,  1.36it/s] 
1|4615|Loss: 0.814628541469574:  71%|███████▏  | 4615/6470 [56:01<23:15,  1.33it/s]
1|4616|Loss: 0.8687229156494141:  71%|███████▏  | 4615/6470 [56:01<23:15,  1.33it/s]
1|4616|Loss: 0.8687229156494141:  71%|███████▏  | 4616/6470 [56:01<23:12,  1.33it/s]
1|4617|Loss: 0.8477388620376587:  71%|███████▏  | 4616/6470 [56:02<23:12,  1.33it/s]
1|4617|Loss: 0.8477388620376587:  71%|███████▏  | 4617/6470 [56:02<23:09,  1.33it/s]
1|4618|Loss: 0.7625387907028198:  71%|███████▏  | 4617/6470 [56:03<23:09,  1.33it/s]
1|4618|Loss: 0.7625387907028198:  71%|███████▏  | 4618/6470 [56:03<22:52,  1.35it/s]
1|4619|Loss: 0.9248642921447754:  71%|███████▏  | 4618/6470 [56:03<22:52,  1.35it/s]
1|4619|Loss: 0.9248642921447754:  71%|███████▏  | 4619/6470 [56:04<22:49,  1.35it/s]
1|4620|Loss: 0.9699943661689758:  71%|███████▏  | 4619/6470 [56:04<22:49,  1.35it/s]
1|4620|Loss: 0.9699943661689758:  71%|███████▏  | 4620/6470 [56:04<22:49,  1.35it/s]
1|4621|Loss: 0.896651566028595:  71%|███████▏  | 4620/6470 [56:05<22:49,  1.35it/s] 
1|4621|Loss: 0.896651566028595:  71%|███████▏  | 4621/6470 [56:05<23:26,  1.31it/s]
1|4622|Loss: 0.8452739715576172:  71%|███████▏  | 4621/6470 [56:06<23:26,  1.31it/s]
1|4622|Loss: 0.8452739715576172:  71%|███████▏  | 4622/6470 [56:06<22:42,  1.36it/s]
1|4623|Loss: 0.9049623608589172:  71%|███████▏  | 4622/6470 [56:06<22:42,  1.36it/s]
1|4623|Loss: 0.9049623608589172:  71%|███████▏  | 4623/6470 [56:06<21:39,  1.42it/s]
1|4624|Loss: 0.8963571786880493:  71%|███████▏  | 4623/6470 [56:07<21:39,  1.42it/s]
1|4624|Loss: 0.8963571786880493:  71%|███████▏  | 4624/6470 [56:07<20:33,  1.50it/s]
1|4625|Loss: 0.7206270098686218:  71%|███████▏  | 4624/6470 [56:07<20:33,  1.50it/s]
1|4625|Loss: 0.7206270098686218:  71%|███████▏  | 4625/6470 [56:08<19:45,  1.56it/s]
1|4626|Loss: 0.9888915419578552:  71%|███████▏  | 4625/6470 [56:08<19:45,  1.56it/s]
1|4626|Loss: 0.9888915419578552:  71%|███████▏  | 4626/6470 [56:08<20:26,  1.50it/s]
1|4627|Loss: 0.8233246803283691:  71%|███████▏  | 4626/6470 [56:09<20:26,  1.50it/s]
1|4627|Loss: 0.8233246803283691:  72%|███████▏  | 4627/6470 [56:09<20:56,  1.47it/s]
1|4628|Loss: 0.8027830123901367:  72%|███████▏  | 4627/6470 [56:09<20:56,  1.47it/s]
1|4628|Loss: 0.8027830123901367:  72%|███████▏  | 4628/6470 [56:10<21:07,  1.45it/s]
1|4629|Loss: 0.9269448518753052:  72%|███████▏  | 4628/6470 [56:10<21:07,  1.45it/s]
1|4629|Loss: 0.9269448518753052:  72%|███████▏  | 4629/6470 [56:11<21:18,  1.44it/s]
1|4630|Loss: 0.9148225784301758:  72%|███████▏  | 4629/6470 [56:11<21:18,  1.44it/s]
1|4630|Loss: 0.9148225784301758:  72%|███████▏  | 4630/6470 [56:11<22:40,  1.35it/s]
1|4631|Loss: 0.9071445465087891:  72%|███████▏  | 4630/6470 [56:12<22:40,  1.35it/s]
1|4631|Loss: 0.9071445465087891:  72%|███████▏  | 4631/6470 [56:12<21:21,  1.44it/s]
1|4632|Loss: 0.9381188750267029:  72%|███████▏  | 4631/6470 [56:12<21:21,  1.44it/s]
1|4632|Loss: 0.9381188750267029:  72%|███████▏  | 4632/6470 [56:13<21:16,  1.44it/s]
1|4633|Loss: 0.700699508190155:  72%|███████▏  | 4632/6470 [56:13<21:16,  1.44it/s] 
1|4633|Loss: 0.700699508190155:  72%|███████▏  | 4633/6470 [56:13<21:47,  1.41it/s]
1|4634|Loss: 0.8111851811408997:  72%|███████▏  | 4633/6470 [56:14<21:47,  1.41it/s]
1|4634|Loss: 0.8111851811408997:  72%|███████▏  | 4634/6470 [56:14<21:25,  1.43it/s]
1|4635|Loss: 0.933582067489624:  72%|███████▏  | 4634/6470 [56:14<21:25,  1.43it/s] 
1|4635|Loss: 0.933582067489624:  72%|███████▏  | 4635/6470 [56:15<20:43,  1.48it/s]
1|4636|Loss: 0.9496267437934875:  72%|███████▏  | 4635/6470 [56:15<20:43,  1.48it/s]
1|4636|Loss: 0.9496267437934875:  72%|███████▏  | 4636/6470 [56:15<21:07,  1.45it/s]
1|4637|Loss: 0.9912598133087158:  72%|███████▏  | 4636/6470 [56:16<21:07,  1.45it/s]
1|4637|Loss: 0.9912598133087158:  72%|███████▏  | 4637/6470 [56:16<20:52,  1.46it/s]
1|4638|Loss: 0.9276490807533264:  72%|███████▏  | 4637/6470 [56:16<20:52,  1.46it/s]
1|4638|Loss: 0.9276490807533264:  72%|███████▏  | 4638/6470 [56:17<21:00,  1.45it/s]
1|4639|Loss: 0.9212102293968201:  72%|███████▏  | 4638/6470 [56:17<21:00,  1.45it/s]
1|4639|Loss: 0.9212102293968201:  72%|███████▏  | 4639/6470 [56:17<20:31,  1.49it/s]
1|4640|Loss: 0.863307535648346:  72%|███████▏  | 4639/6470 [56:18<20:31,  1.49it/s] 
1|4640|Loss: 0.863307535648346:  72%|███████▏  | 4640/6470 [56:18<19:57,  1.53it/s]
1|4641|Loss: 0.8575432300567627:  72%|███████▏  | 4640/6470 [56:18<19:57,  1.53it/s]
1|4641|Loss: 0.8575432300567627:  72%|███████▏  | 4641/6470 [56:19<20:18,  1.50it/s]
1|4642|Loss: 1.059186577796936:  72%|███████▏  | 4641/6470 [56:19<20:18,  1.50it/s] 
1|4642|Loss: 1.059186577796936:  72%|███████▏  | 4642/6470 [56:19<19:58,  1.53it/s]
1|4643|Loss: 0.950266420841217:  72%|███████▏  | 4642/6470 [56:20<19:58,  1.53it/s]
1|4643|Loss: 0.950266420841217:  72%|███████▏  | 4643/6470 [56:20<20:34,  1.48it/s]
1|4644|Loss: 0.8583388924598694:  72%|███████▏  | 4643/6470 [56:20<20:34,  1.48it/s]
1|4644|Loss: 0.8583388924598694:  72%|███████▏  | 4644/6470 [56:21<21:08,  1.44it/s]
1|4645|Loss: 0.9658843874931335:  72%|███████▏  | 4644/6470 [56:21<21:08,  1.44it/s]
1|4645|Loss: 0.9658843874931335:  72%|███████▏  | 4645/6470 [56:22<21:39,  1.40it/s]
1|4646|Loss: 0.8821271061897278:  72%|███████▏  | 4645/6470 [56:22<21:39,  1.40it/s]
1|4646|Loss: 0.8821271061897278:  72%|███████▏  | 4646/6470 [56:22<22:22,  1.36it/s]
1|4647|Loss: 1.044048547744751:  72%|███████▏  | 4646/6470 [56:23<22:22,  1.36it/s] 
1|4647|Loss: 1.044048547744751:  72%|███████▏  | 4647/6470 [56:23<21:49,  1.39it/s]
1|4648|Loss: 0.96475750207901:  72%|███████▏  | 4647/6470 [56:23<21:49,  1.39it/s] 
1|4648|Loss: 0.96475750207901:  72%|███████▏  | 4648/6470 [56:24<21:07,  1.44it/s]
1|4649|Loss: 0.9163007736206055:  72%|███████▏  | 4648/6470 [56:24<21:07,  1.44it/s]
1|4649|Loss: 0.9163007736206055:  72%|███████▏  | 4649/6470 [56:24<20:38,  1.47it/s]
1|4650|Loss: 0.8386797308921814:  72%|███████▏  | 4649/6470 [56:25<20:38,  1.47it/s]
1|4650|Loss: 0.8386797308921814:  72%|███████▏  | 4650/6470 [56:25<20:42,  1.46it/s]
1|4651|Loss: 0.9572007656097412:  72%|███████▏  | 4650/6470 [56:25<20:42,  1.46it/s]
1|4651|Loss: 0.9572007656097412:  72%|███████▏  | 4651/6470 [56:26<20:54,  1.45it/s]
1|4652|Loss: 0.8583539128303528:  72%|███████▏  | 4651/6470 [56:26<20:54,  1.45it/s]
1|4652|Loss: 0.8583539128303528:  72%|███████▏  | 4652/6470 [56:26<20:59,  1.44it/s]
1|4653|Loss: 0.8514545559883118:  72%|███████▏  | 4652/6470 [56:27<20:59,  1.44it/s]
1|4653|Loss: 0.8514545559883118:  72%|███████▏  | 4653/6470 [56:27<20:28,  1.48it/s]
1|4654|Loss: 0.9139418005943298:  72%|███████▏  | 4653/6470 [56:27<20:28,  1.48it/s]
1|4654|Loss: 0.9139418005943298:  72%|███████▏  | 4654/6470 [56:28<20:45,  1.46it/s]
1|4655|Loss: 0.9773476123809814:  72%|███████▏  | 4654/6470 [56:28<20:45,  1.46it/s]
1|4655|Loss: 0.9773476123809814:  72%|███████▏  | 4655/6470 [56:28<19:24,  1.56it/s]
1|4656|Loss: 0.7582804560661316:  72%|███████▏  | 4655/6470 [56:29<19:24,  1.56it/s]
1|4656|Loss: 0.7582804560661316:  72%|███████▏  | 4656/6470 [56:29<17:14,  1.75it/s]
1|4657|Loss: 0.7872776985168457:  72%|███████▏  | 4656/6470 [56:29<17:14,  1.75it/s]
1|4657|Loss: 0.7872776985168457:  72%|███████▏  | 4657/6470 [56:29<17:29,  1.73it/s]
1|4658|Loss: 0.9122318029403687:  72%|███████▏  | 4657/6470 [56:30<17:29,  1.73it/s]
1|4658|Loss: 0.9122318029403687:  72%|███████▏  | 4658/6470 [56:30<18:44,  1.61it/s]
1|4659|Loss: 0.7730771899223328:  72%|███████▏  | 4658/6470 [56:30<18:44,  1.61it/s]
1|4659|Loss: 0.7730771899223328:  72%|███████▏  | 4659/6470 [56:31<19:43,  1.53it/s]
1|4660|Loss: 0.8375473022460938:  72%|███████▏  | 4659/6470 [56:31<19:43,  1.53it/s]
1|4660|Loss: 0.8375473022460938:  72%|███████▏  | 4660/6470 [56:32<20:48,  1.45it/s]
1|4661|Loss: 0.749177098274231:  72%|███████▏  | 4660/6470 [56:32<20:48,  1.45it/s] 
1|4661|Loss: 0.749177098274231:  72%|███████▏  | 4661/6470 [56:32<18:37,  1.62it/s]
1|4662|Loss: 0.9067788124084473:  72%|███████▏  | 4661/6470 [56:32<18:37,  1.62it/s]
1|4662|Loss: 0.9067788124084473:  72%|███████▏  | 4662/6470 [56:33<18:11,  1.66it/s]
1|4663|Loss: 0.9454882144927979:  72%|███████▏  | 4662/6470 [56:33<18:11,  1.66it/s]
1|4663|Loss: 0.9454882144927979:  72%|███████▏  | 4663/6470 [56:33<18:56,  1.59it/s]
1|4664|Loss: 0.6891574263572693:  72%|███████▏  | 4663/6470 [56:34<18:56,  1.59it/s]
1|4664|Loss: 0.6891574263572693:  72%|███████▏  | 4664/6470 [56:34<20:11,  1.49it/s]
1|4665|Loss: 0.9508921504020691:  72%|███████▏  | 4664/6470 [56:34<20:11,  1.49it/s]
1|4665|Loss: 0.9508921504020691:  72%|███████▏  | 4665/6470 [56:35<21:00,  1.43it/s]
1|4666|Loss: 0.8429191708564758:  72%|███████▏  | 4665/6470 [56:35<21:00,  1.43it/s]
1|4666|Loss: 0.8429191708564758:  72%|███████▏  | 4666/6470 [56:36<23:01,  1.31it/s]
1|4667|Loss: 0.8425065875053406:  72%|███████▏  | 4666/6470 [56:36<23:01,  1.31it/s]
1|4667|Loss: 0.8425065875053406:  72%|███████▏  | 4667/6470 [56:37<23:55,  1.26it/s]
1|4668|Loss: 0.9415316581726074:  72%|███████▏  | 4667/6470 [56:37<23:55,  1.26it/s]
1|4668|Loss: 0.9415316581726074:  72%|███████▏  | 4668/6470 [56:37<22:43,  1.32it/s]
1|4669|Loss: 0.6220775842666626:  72%|███████▏  | 4668/6470 [56:38<22:43,  1.32it/s]
1|4669|Loss: 0.6220775842666626:  72%|███████▏  | 4669/6470 [56:38<23:32,  1.28it/s]
1|4670|Loss: 0.9029566049575806:  72%|███████▏  | 4669/6470 [56:39<23:32,  1.28it/s]
1|4670|Loss: 0.9029566049575806:  72%|███████▏  | 4670/6470 [56:39<23:23,  1.28it/s]
1|4671|Loss: 1.0434048175811768:  72%|███████▏  | 4670/6470 [56:39<23:23,  1.28it/s]
1|4671|Loss: 1.0434048175811768:  72%|███████▏  | 4671/6470 [56:40<23:03,  1.30it/s]
1|4672|Loss: 0.8178321123123169:  72%|███████▏  | 4671/6470 [56:40<23:03,  1.30it/s]
1|4672|Loss: 0.8178321123123169:  72%|███████▏  | 4672/6470 [56:40<22:44,  1.32it/s]
1|4673|Loss: 0.7544633150100708:  72%|███████▏  | 4672/6470 [56:41<22:44,  1.32it/s]
1|4673|Loss: 0.7544633150100708:  72%|███████▏  | 4673/6470 [56:41<20:42,  1.45it/s]
1|4674|Loss: 0.7554419636726379:  72%|███████▏  | 4673/6470 [56:41<20:42,  1.45it/s]
1|4674|Loss: 0.7554419636726379:  72%|███████▏  | 4674/6470 [56:42<22:12,  1.35it/s]
1|4675|Loss: 0.8799586296081543:  72%|███████▏  | 4674/6470 [56:42<22:12,  1.35it/s]
1|4675|Loss: 0.8799586296081543:  72%|███████▏  | 4675/6470 [56:42<20:32,  1.46it/s]
1|4676|Loss: 0.917840838432312:  72%|███████▏  | 4675/6470 [56:43<20:32,  1.46it/s] 
1|4676|Loss: 0.917840838432312:  72%|███████▏  | 4676/6470 [56:43<20:51,  1.43it/s]
1|4677|Loss: 0.7749228477478027:  72%|███████▏  | 4676/6470 [56:43<20:51,  1.43it/s]
1|4677|Loss: 0.7749228477478027:  72%|███████▏  | 4677/6470 [56:44<22:01,  1.36it/s]
1|4678|Loss: 0.9197630286216736:  72%|███████▏  | 4677/6470 [56:44<22:01,  1.36it/s]
1|4678|Loss: 0.9197630286216736:  72%|███████▏  | 4678/6470 [56:45<22:01,  1.36it/s]
1|4679|Loss: 0.7542041540145874:  72%|███████▏  | 4678/6470 [56:45<22:01,  1.36it/s]
1|4679|Loss: 0.7542041540145874:  72%|███████▏  | 4679/6470 [56:45<22:02,  1.35it/s]
1|4680|Loss: 0.9658244848251343:  72%|███████▏  | 4679/6470 [56:46<22:02,  1.35it/s]
1|4680|Loss: 0.9658244848251343:  72%|███████▏  | 4680/6470 [56:46<21:12,  1.41it/s]
1|4681|Loss: 1.0827103853225708:  72%|███████▏  | 4680/6470 [56:46<21:12,  1.41it/s]
1|4681|Loss: 1.0827103853225708:  72%|███████▏  | 4681/6470 [56:47<22:07,  1.35it/s]
1|4682|Loss: 0.7160979509353638:  72%|███████▏  | 4681/6470 [56:47<22:07,  1.35it/s]
1|4682|Loss: 0.7160979509353638:  72%|███████▏  | 4682/6470 [56:47<21:17,  1.40it/s]
1|4683|Loss: 1.059968113899231:  72%|███████▏  | 4682/6470 [56:48<21:17,  1.40it/s] 
1|4683|Loss: 1.059968113899231:  72%|███████▏  | 4683/6470 [56:48<20:58,  1.42it/s]
1|4684|Loss: 0.8342577219009399:  72%|███████▏  | 4683/6470 [56:48<20:58,  1.42it/s]
1|4684|Loss: 0.8342577219009399:  72%|███████▏  | 4684/6470 [56:49<21:17,  1.40it/s]
1|4685|Loss: 0.9406430721282959:  72%|███████▏  | 4684/6470 [56:49<21:17,  1.40it/s]
1|4685|Loss: 0.9406430721282959:  72%|███████▏  | 4685/6470 [56:49<20:51,  1.43it/s]
1|4686|Loss: 0.7630861401557922:  72%|███████▏  | 4685/6470 [56:50<20:51,  1.43it/s]
1|4686|Loss: 0.7630861401557922:  72%|███████▏  | 4686/6470 [56:50<21:38,  1.37it/s]
1|4687|Loss: 0.903980553150177:  72%|███████▏  | 4686/6470 [56:51<21:38,  1.37it/s] 
1|4687|Loss: 0.903980553150177:  72%|███████▏  | 4687/6470 [56:51<22:25,  1.33it/s]
1|4688|Loss: 0.894822359085083:  72%|███████▏  | 4687/6470 [56:52<22:25,  1.33it/s]
1|4688|Loss: 0.894822359085083:  72%|███████▏  | 4688/6470 [56:52<22:25,  1.32it/s]
1|4689|Loss: 0.8398511409759521:  72%|███████▏  | 4688/6470 [56:52<22:25,  1.32it/s]
1|4689|Loss: 0.8398511409759521:  72%|███████▏  | 4689/6470 [56:53<22:15,  1.33it/s]
1|4690|Loss: 0.8251647353172302:  72%|███████▏  | 4689/6470 [56:53<22:15,  1.33it/s]
1|4690|Loss: 0.8251647353172302:  72%|███████▏  | 4690/6470 [56:53<22:45,  1.30it/s]
1|4691|Loss: 0.8940953016281128:  72%|███████▏  | 4690/6470 [56:54<22:45,  1.30it/s]
1|4691|Loss: 0.8940953016281128:  73%|███████▎  | 4691/6470 [56:54<21:35,  1.37it/s]
1|4692|Loss: 0.6522055268287659:  73%|███████▎  | 4691/6470 [56:54<21:35,  1.37it/s]
1|4692|Loss: 0.6522055268287659:  73%|███████▎  | 4692/6470 [56:55<19:22,  1.53it/s]
1|4693|Loss: 0.8686661124229431:  73%|███████▎  | 4692/6470 [56:55<19:22,  1.53it/s]
1|4693|Loss: 0.8686661124229431:  73%|███████▎  | 4693/6470 [56:55<19:34,  1.51it/s]
1|4694|Loss: 1.0310425758361816:  73%|███████▎  | 4693/6470 [56:56<19:34,  1.51it/s]
1|4694|Loss: 1.0310425758361816:  73%|███████▎  | 4694/6470 [56:56<18:49,  1.57it/s]
1|4695|Loss: 0.9828961491584778:  73%|███████▎  | 4694/6470 [56:56<18:49,  1.57it/s]
1|4695|Loss: 0.9828961491584778:  73%|███████▎  | 4695/6470 [56:56<17:20,  1.71it/s]
1|4696|Loss: 0.8426715135574341:  73%|███████▎  | 4695/6470 [56:57<17:20,  1.71it/s]
1|4696|Loss: 0.8426715135574341:  73%|███████▎  | 4696/6470 [56:57<16:46,  1.76it/s]
1|4697|Loss: 0.8141061067581177:  73%|███████▎  | 4696/6470 [56:57<16:46,  1.76it/s]
1|4697|Loss: 0.8141061067581177:  73%|███████▎  | 4697/6470 [56:57<16:37,  1.78it/s]
1|4698|Loss: 1.0490598678588867:  73%|███████▎  | 4697/6470 [56:58<16:37,  1.78it/s]
1|4698|Loss: 1.0490598678588867:  73%|███████▎  | 4698/6470 [56:58<16:59,  1.74it/s]
1|4699|Loss: 0.8887289762496948:  73%|███████▎  | 4698/6470 [56:58<16:59,  1.74it/s]
1|4699|Loss: 0.8887289762496948:  73%|███████▎  | 4699/6470 [56:59<18:43,  1.58it/s]
1|4700|Loss: 1.0252856016159058:  73%|███████▎  | 4699/6470 [56:59<18:43,  1.58it/s]
1|4700|Loss: 1.0252856016159058:  73%|███████▎  | 4700/6470 [56:59<17:56,  1.64it/s]
1|4701|Loss: 0.589645266532898:  73%|███████▎  | 4700/6470 [57:00<17:56,  1.64it/s] 
1|4701|Loss: 0.589645266532898:  73%|███████▎  | 4701/6470 [57:00<19:36,  1.50it/s]
1|4702|Loss: 0.9544192552566528:  73%|███████▎  | 4701/6470 [57:00<19:36,  1.50it/s]
1|4702|Loss: 0.9544192552566528:  73%|███████▎  | 4702/6470 [57:01<19:42,  1.50it/s]
1|4703|Loss: 0.8752082586288452:  73%|███████▎  | 4702/6470 [57:01<19:42,  1.50it/s]
1|4703|Loss: 0.8752082586288452:  73%|███████▎  | 4703/6470 [57:01<19:51,  1.48it/s]
1|4704|Loss: 0.8092360496520996:  73%|███████▎  | 4703/6470 [57:02<19:51,  1.48it/s]
1|4704|Loss: 0.8092360496520996:  73%|███████▎  | 4704/6470 [57:02<20:15,  1.45it/s]
1|4705|Loss: 0.7083555459976196:  73%|███████▎  | 4704/6470 [57:03<20:15,  1.45it/s]
1|4705|Loss: 0.7083555459976196:  73%|███████▎  | 4705/6470 [57:03<21:25,  1.37it/s]
1|4706|Loss: 0.7392383813858032:  73%|███████▎  | 4705/6470 [57:03<21:25,  1.37it/s]
1|4706|Loss: 0.7392383813858032:  73%|███████▎  | 4706/6470 [57:04<20:38,  1.42it/s]
1|4707|Loss: 1.0127955675125122:  73%|███████▎  | 4706/6470 [57:04<20:38,  1.42it/s]
1|4707|Loss: 1.0127955675125122:  73%|███████▎  | 4707/6470 [57:04<20:26,  1.44it/s]
1|4708|Loss: 0.9112609624862671:  73%|███████▎  | 4707/6470 [57:05<20:26,  1.44it/s]
1|4708|Loss: 0.9112609624862671:  73%|███████▎  | 4708/6470 [57:05<21:15,  1.38it/s]
1|4709|Loss: 0.846577525138855:  73%|███████▎  | 4708/6470 [57:05<21:15,  1.38it/s] 
1|4709|Loss: 0.846577525138855:  73%|███████▎  | 4709/6470 [57:06<20:43,  1.42it/s]
1|4710|Loss: 0.7699592113494873:  73%|███████▎  | 4709/6470 [57:06<20:43,  1.42it/s]
1|4710|Loss: 0.7699592113494873:  73%|███████▎  | 4710/6470 [57:06<19:17,  1.52it/s]
1|4711|Loss: 0.8000184297561646:  73%|███████▎  | 4710/6470 [57:07<19:17,  1.52it/s]
1|4711|Loss: 0.8000184297561646:  73%|███████▎  | 4711/6470 [57:07<19:24,  1.51it/s]
1|4712|Loss: 0.9101780652999878:  73%|███████▎  | 4711/6470 [57:07<19:24,  1.51it/s]
1|4712|Loss: 0.9101780652999878:  73%|███████▎  | 4712/6470 [57:08<19:46,  1.48it/s]
1|4713|Loss: 0.9160542488098145:  73%|███████▎  | 4712/6470 [57:08<19:46,  1.48it/s]
1|4713|Loss: 0.9160542488098145:  73%|███████▎  | 4713/6470 [57:08<20:37,  1.42it/s]
1|4714|Loss: 0.978965699672699:  73%|███████▎  | 4713/6470 [57:09<20:37,  1.42it/s] 
1|4714|Loss: 0.978965699672699:  73%|███████▎  | 4714/6470 [57:09<20:20,  1.44it/s]
1|4715|Loss: 0.8610758185386658:  73%|███████▎  | 4714/6470 [57:09<20:20,  1.44it/s]
1|4715|Loss: 0.8610758185386658:  73%|███████▎  | 4715/6470 [57:10<19:42,  1.48it/s]
1|4716|Loss: 0.8678290843963623:  73%|███████▎  | 4715/6470 [57:10<19:42,  1.48it/s]
1|4716|Loss: 0.8678290843963623:  73%|███████▎  | 4716/6470 [57:10<19:53,  1.47it/s]
1|4717|Loss: 0.8137075901031494:  73%|███████▎  | 4716/6470 [57:11<19:53,  1.47it/s]
1|4717|Loss: 0.8137075901031494:  73%|███████▎  | 4717/6470 [57:11<20:58,  1.39it/s]
1|4718|Loss: 0.7927404642105103:  73%|███████▎  | 4717/6470 [57:12<20:58,  1.39it/s]
1|4718|Loss: 0.7927404642105103:  73%|███████▎  | 4718/6470 [57:12<19:44,  1.48it/s]
1|4719|Loss: 0.8880683183670044:  73%|███████▎  | 4718/6470 [57:12<19:44,  1.48it/s]
1|4719|Loss: 0.8880683183670044:  73%|███████▎  | 4719/6470 [57:12<19:10,  1.52it/s]
1|4720|Loss: 0.7730948328971863:  73%|███████▎  | 4719/6470 [57:13<19:10,  1.52it/s]
1|4720|Loss: 0.7730948328971863:  73%|███████▎  | 4720/6470 [57:13<18:35,  1.57it/s]
1|4721|Loss: 0.7084283828735352:  73%|███████▎  | 4720/6470 [57:13<18:35,  1.57it/s]
1|4721|Loss: 0.7084283828735352:  73%|███████▎  | 4721/6470 [57:14<18:56,  1.54it/s]
1|4722|Loss: 0.9779388308525085:  73%|███████▎  | 4721/6470 [57:14<18:56,  1.54it/s]
1|4722|Loss: 0.9779388308525085:  73%|███████▎  | 4722/6470 [57:14<18:58,  1.54it/s]
1|4723|Loss: 0.9016076326370239:  73%|███████▎  | 4722/6470 [57:15<18:58,  1.54it/s]
1|4723|Loss: 0.9016076326370239:  73%|███████▎  | 4723/6470 [57:15<18:29,  1.58it/s]
1|4724|Loss: 0.7883682250976562:  73%|███████▎  | 4723/6470 [57:15<18:29,  1.58it/s]
1|4724|Loss: 0.7883682250976562:  73%|███████▎  | 4724/6470 [57:15<17:30,  1.66it/s]
1|4725|Loss: 0.9537168145179749:  73%|███████▎  | 4724/6470 [57:16<17:30,  1.66it/s]
1|4725|Loss: 0.9537168145179749:  73%|███████▎  | 4725/6470 [57:16<15:49,  1.84it/s]
1|4726|Loss: 0.928317129611969:  73%|███████▎  | 4725/6470 [57:16<15:49,  1.84it/s] 
1|4726|Loss: 0.928317129611969:  73%|███████▎  | 4726/6470 [57:16<15:30,  1.87it/s]
1|4727|Loss: 0.9577086567878723:  73%|███████▎  | 4726/6470 [57:17<15:30,  1.87it/s]
1|4727|Loss: 0.9577086567878723:  73%|███████▎  | 4727/6470 [57:17<14:57,  1.94it/s]
1|4728|Loss: 0.9053152799606323:  73%|███████▎  | 4727/6470 [57:17<14:57,  1.94it/s]
1|4728|Loss: 0.9053152799606323:  73%|███████▎  | 4728/6470 [57:18<16:25,  1.77it/s]
1|4729|Loss: 1.0084959268569946:  73%|███████▎  | 4728/6470 [57:18<16:25,  1.77it/s]
1|4729|Loss: 1.0084959268569946:  73%|███████▎  | 4729/6470 [57:18<17:55,  1.62it/s]
1|4730|Loss: 0.962918758392334:  73%|███████▎  | 4729/6470 [57:19<17:55,  1.62it/s] 
1|4730|Loss: 0.962918758392334:  73%|███████▎  | 4730/6470 [57:19<18:09,  1.60it/s]
1|4731|Loss: 0.5570168495178223:  73%|███████▎  | 4730/6470 [57:19<18:09,  1.60it/s]
1|4731|Loss: 0.5570168495178223:  73%|███████▎  | 4731/6470 [57:20<20:03,  1.44it/s]
1|4732|Loss: 0.8535102009773254:  73%|███████▎  | 4731/6470 [57:20<20:03,  1.44it/s]
1|4732|Loss: 0.8535102009773254:  73%|███████▎  | 4732/6470 [57:20<18:10,  1.59it/s]
1|4733|Loss: 0.819892406463623:  73%|███████▎  | 4732/6470 [57:21<18:10,  1.59it/s] 
1|4733|Loss: 0.819892406463623:  73%|███████▎  | 4733/6470 [57:21<18:37,  1.55it/s]
1|4734|Loss: 1.0473612546920776:  73%|███████▎  | 4733/6470 [57:21<18:37,  1.55it/s]
1|4734|Loss: 1.0473612546920776:  73%|███████▎  | 4734/6470 [57:22<20:18,  1.42it/s]
1|4735|Loss: 0.9005683064460754:  73%|███████▎  | 4734/6470 [57:22<20:18,  1.42it/s]
1|4735|Loss: 0.9005683064460754:  73%|███████▎  | 4735/6470 [57:22<20:36,  1.40it/s]
1|4736|Loss: 0.8834042549133301:  73%|███████▎  | 4735/6470 [57:23<20:36,  1.40it/s]
1|4736|Loss: 0.8834042549133301:  73%|███████▎  | 4736/6470 [57:23<19:53,  1.45it/s]
1|4737|Loss: 0.8905982971191406:  73%|███████▎  | 4736/6470 [57:23<19:53,  1.45it/s]
1|4737|Loss: 0.8905982971191406:  73%|███████▎  | 4737/6470 [57:24<17:30,  1.65it/s]
1|4738|Loss: 0.911840558052063:  73%|███████▎  | 4737/6470 [57:24<17:30,  1.65it/s] 
1|4738|Loss: 0.911840558052063:  73%|███████▎  | 4738/6470 [57:24<18:08,  1.59it/s]
1|4739|Loss: 0.9853696823120117:  73%|███████▎  | 4738/6470 [57:25<18:08,  1.59it/s]
1|4739|Loss: 0.9853696823120117:  73%|███████▎  | 4739/6470 [57:25<17:53,  1.61it/s]
1|4740|Loss: 0.7912002205848694:  73%|███████▎  | 4739/6470 [57:25<17:53,  1.61it/s]
1|4740|Loss: 0.7912002205848694:  73%|███████▎  | 4740/6470 [57:25<17:53,  1.61it/s]
1|4741|Loss: 0.945967972278595:  73%|███████▎  | 4740/6470 [57:26<17:53,  1.61it/s] 
1|4741|Loss: 0.945967972278595:  73%|███████▎  | 4741/6470 [57:26<18:41,  1.54it/s]
1|4742|Loss: 0.8595107793807983:  73%|███████▎  | 4741/6470 [57:27<18:41,  1.54it/s]
1|4742|Loss: 0.8595107793807983:  73%|███████▎  | 4742/6470 [57:27<18:48,  1.53it/s]
1|4743|Loss: 0.8490051031112671:  73%|███████▎  | 4742/6470 [57:27<18:48,  1.53it/s]
1|4743|Loss: 0.8490051031112671:  73%|███████▎  | 4743/6470 [57:27<17:44,  1.62it/s]
1|4744|Loss: 0.9812329411506653:  73%|███████▎  | 4743/6470 [57:28<17:44,  1.62it/s]
1|4744|Loss: 0.9812329411506653:  73%|███████▎  | 4744/6470 [57:28<18:07,  1.59it/s]
1|4745|Loss: 0.7721977829933167:  73%|███████▎  | 4744/6470 [57:28<18:07,  1.59it/s]
1|4745|Loss: 0.7721977829933167:  73%|███████▎  | 4745/6470 [57:29<18:47,  1.53it/s]
1|4746|Loss: 0.9004334211349487:  73%|███████▎  | 4745/6470 [57:29<18:47,  1.53it/s]
1|4746|Loss: 0.9004334211349487:  73%|███████▎  | 4746/6470 [57:30<20:21,  1.41it/s]
1|4747|Loss: 0.7868943810462952:  73%|███████▎  | 4746/6470 [57:30<20:21,  1.41it/s]
1|4747|Loss: 0.7868943810462952:  73%|███████▎  | 4747/6470 [57:30<21:13,  1.35it/s]
1|4748|Loss: 0.9418178200721741:  73%|███████▎  | 4747/6470 [57:31<21:13,  1.35it/s]
1|4748|Loss: 0.9418178200721741:  73%|███████▎  | 4748/6470 [57:31<21:06,  1.36it/s]
1|4749|Loss: 0.8044160008430481:  73%|███████▎  | 4748/6470 [57:31<21:06,  1.36it/s]
1|4749|Loss: 0.8044160008430481:  73%|███████▎  | 4749/6470 [57:32<18:38,  1.54it/s]
1|4750|Loss: 0.9195733070373535:  73%|███████▎  | 4749/6470 [57:32<18:38,  1.54it/s]
1|4750|Loss: 0.9195733070373535:  73%|███████▎  | 4750/6470 [57:32<18:08,  1.58it/s]
1|4751|Loss: 0.8480802774429321:  73%|███████▎  | 4750/6470 [57:32<18:08,  1.58it/s]
1|4751|Loss: 0.8480802774429321:  73%|███████▎  | 4751/6470 [57:33<17:16,  1.66it/s]
1|4752|Loss: 0.8798932433128357:  73%|███████▎  | 4751/6470 [57:33<17:16,  1.66it/s]
1|4752|Loss: 0.8798932433128357:  73%|███████▎  | 4752/6470 [57:33<16:05,  1.78it/s]
1|4753|Loss: 0.8017157912254333:  73%|███████▎  | 4752/6470 [57:33<16:05,  1.78it/s]
1|4753|Loss: 0.8017157912254333:  73%|███████▎  | 4753/6470 [57:34<15:48,  1.81it/s]
1|4754|Loss: 0.877184271812439:  73%|███████▎  | 4753/6470 [57:34<15:48,  1.81it/s] 
1|4754|Loss: 0.877184271812439:  73%|███████▎  | 4754/6470 [57:34<17:54,  1.60it/s]
1|4755|Loss: 0.7436938285827637:  73%|███████▎  | 4754/6470 [57:35<17:54,  1.60it/s]
1|4755|Loss: 0.7436938285827637:  73%|███████▎  | 4755/6470 [57:35<18:24,  1.55it/s]
1|4756|Loss: 0.8626238107681274:  73%|███████▎  | 4755/6470 [57:36<18:24,  1.55it/s]
1|4756|Loss: 0.8626238107681274:  74%|███████▎  | 4756/6470 [57:36<19:37,  1.46it/s]
1|4757|Loss: 0.9627763032913208:  74%|███████▎  | 4756/6470 [57:36<19:37,  1.46it/s]
1|4757|Loss: 0.9627763032913208:  74%|███████▎  | 4757/6470 [57:37<19:05,  1.50it/s]
1|4758|Loss: 0.7903492450714111:  74%|███████▎  | 4757/6470 [57:37<19:05,  1.50it/s]
1|4758|Loss: 0.7903492450714111:  74%|███████▎  | 4758/6470 [57:37<19:45,  1.44it/s]
1|4759|Loss: 0.6529708504676819:  74%|███████▎  | 4758/6470 [57:38<19:45,  1.44it/s]
1|4759|Loss: 0.6529708504676819:  74%|███████▎  | 4759/6470 [57:38<19:37,  1.45it/s]
1|4760|Loss: 0.8362451195716858:  74%|███████▎  | 4759/6470 [57:38<19:37,  1.45it/s]
1|4760|Loss: 0.8362451195716858:  74%|███████▎  | 4760/6470 [57:39<20:17,  1.40it/s]
1|4761|Loss: 0.7889739274978638:  74%|███████▎  | 4760/6470 [57:39<20:17,  1.40it/s]
1|4761|Loss: 0.7889739274978638:  74%|███████▎  | 4761/6470 [57:39<19:58,  1.43it/s]
1|4762|Loss: 0.9954349398612976:  74%|███████▎  | 4761/6470 [57:40<19:58,  1.43it/s]
1|4762|Loss: 0.9954349398612976:  74%|███████▎  | 4762/6470 [57:40<19:17,  1.48it/s]
1|4763|Loss: 0.6625007390975952:  74%|███████▎  | 4762/6470 [57:40<19:17,  1.48it/s]
1|4763|Loss: 0.6625007390975952:  74%|███████▎  | 4763/6470 [57:41<18:27,  1.54it/s]
1|4764|Loss: 0.8488845229148865:  74%|███████▎  | 4763/6470 [57:41<18:27,  1.54it/s]
1|4764|Loss: 0.8488845229148865:  74%|███████▎  | 4764/6470 [57:41<18:07,  1.57it/s]
1|4765|Loss: 0.9018126726150513:  74%|███████▎  | 4764/6470 [57:42<18:07,  1.57it/s]
1|4765|Loss: 0.9018126726150513:  74%|███████▎  | 4765/6470 [57:42<19:33,  1.45it/s]
1|4766|Loss: 0.7148959040641785:  74%|███████▎  | 4765/6470 [57:42<19:33,  1.45it/s]
1|4766|Loss: 0.7148959040641785:  74%|███████▎  | 4766/6470 [57:43<20:17,  1.40it/s]
1|4767|Loss: 0.8916361927986145:  74%|███████▎  | 4766/6470 [57:43<20:17,  1.40it/s]
1|4767|Loss: 0.8916361927986145:  74%|███████▎  | 4767/6470 [57:44<21:16,  1.33it/s]
1|4768|Loss: 0.7657718062400818:  74%|███████▎  | 4767/6470 [57:44<21:16,  1.33it/s]
1|4768|Loss: 0.7657718062400818:  74%|███████▎  | 4768/6470 [57:44<22:00,  1.29it/s]
1|4769|Loss: 0.8313143849372864:  74%|███████▎  | 4768/6470 [57:45<22:00,  1.29it/s]
1|4769|Loss: 0.8313143849372864:  74%|███████▎  | 4769/6470 [57:45<22:39,  1.25it/s]
1|4770|Loss: 0.8724825382232666:  74%|███████▎  | 4769/6470 [57:46<22:39,  1.25it/s]
1|4770|Loss: 0.8724825382232666:  74%|███████▎  | 4770/6470 [57:46<22:23,  1.27it/s]
1|4771|Loss: 0.788666844367981:  74%|███████▎  | 4770/6470 [57:47<22:23,  1.27it/s] 
1|4771|Loss: 0.788666844367981:  74%|███████▎  | 4771/6470 [57:47<22:37,  1.25it/s]
1|4772|Loss: 0.8570693731307983:  74%|███████▎  | 4771/6470 [57:47<22:37,  1.25it/s]
1|4772|Loss: 0.8570693731307983:  74%|███████▍  | 4772/6470 [57:48<22:43,  1.25it/s]
1|4773|Loss: 0.8854859471321106:  74%|███████▍  | 4772/6470 [57:48<22:43,  1.25it/s]
1|4773|Loss: 0.8854859471321106:  74%|███████▍  | 4773/6470 [57:49<22:14,  1.27it/s]
1|4774|Loss: 0.8528282046318054:  74%|███████▍  | 4773/6470 [57:49<22:14,  1.27it/s]
1|4774|Loss: 0.8528282046318054:  74%|███████▍  | 4774/6470 [57:49<20:15,  1.40it/s]
1|4775|Loss: 0.7511588931083679:  74%|███████▍  | 4774/6470 [57:49<20:15,  1.40it/s]
1|4775|Loss: 0.7511588931083679:  74%|███████▍  | 4775/6470 [57:50<19:20,  1.46it/s]
1|4776|Loss: 0.9023510217666626:  74%|███████▍  | 4775/6470 [57:50<19:20,  1.46it/s]
1|4776|Loss: 0.9023510217666626:  74%|███████▍  | 4776/6470 [57:50<19:42,  1.43it/s]
1|4777|Loss: 0.7341166138648987:  74%|███████▍  | 4776/6470 [57:51<19:42,  1.43it/s]
1|4777|Loss: 0.7341166138648987:  74%|███████▍  | 4777/6470 [57:51<20:02,  1.41it/s]
1|4778|Loss: 0.9448323249816895:  74%|███████▍  | 4777/6470 [57:51<20:02,  1.41it/s]
1|4778|Loss: 0.9448323249816895:  74%|███████▍  | 4778/6470 [57:52<19:16,  1.46it/s]
1|4779|Loss: 0.8225622773170471:  74%|███████▍  | 4778/6470 [57:52<19:16,  1.46it/s]
1|4779|Loss: 0.8225622773170471:  74%|███████▍  | 4779/6470 [57:53<20:07,  1.40it/s]
1|4780|Loss: 0.8931870460510254:  74%|███████▍  | 4779/6470 [57:53<20:07,  1.40it/s]
1|4780|Loss: 0.8931870460510254:  74%|███████▍  | 4780/6470 [57:53<19:17,  1.46it/s]
1|4781|Loss: 0.7864986062049866:  74%|███████▍  | 4780/6470 [57:54<19:17,  1.46it/s]
1|4781|Loss: 0.7864986062049866:  74%|███████▍  | 4781/6470 [57:54<19:22,  1.45it/s]
1|4782|Loss: 0.9230383634567261:  74%|███████▍  | 4781/6470 [57:54<19:22,  1.45it/s]
1|4782|Loss: 0.9230383634567261:  74%|███████▍  | 4782/6470 [57:54<18:15,  1.54it/s]
1|4783|Loss: 0.8778474926948547:  74%|███████▍  | 4782/6470 [57:55<18:15,  1.54it/s]
1|4783|Loss: 0.8778474926948547:  74%|███████▍  | 4783/6470 [57:55<16:55,  1.66it/s]
1|4784|Loss: 0.965216338634491:  74%|███████▍  | 4783/6470 [57:55<16:55,  1.66it/s] 
1|4784|Loss: 0.965216338634491:  74%|███████▍  | 4784/6470 [57:56<17:21,  1.62it/s]
1|4785|Loss: 0.8561948537826538:  74%|███████▍  | 4784/6470 [57:56<17:21,  1.62it/s]
1|4785|Loss: 0.8561948537826538:  74%|███████▍  | 4785/6470 [57:56<17:30,  1.60it/s]
1|4786|Loss: 0.9050183296203613:  74%|███████▍  | 4785/6470 [57:57<17:30,  1.60it/s]
1|4786|Loss: 0.9050183296203613:  74%|███████▍  | 4786/6470 [57:57<17:31,  1.60it/s]
1|4787|Loss: 0.8572434782981873:  74%|███████▍  | 4786/6470 [57:57<17:31,  1.60it/s]
1|4787|Loss: 0.8572434782981873:  74%|███████▍  | 4787/6470 [57:58<18:06,  1.55it/s]
1|4788|Loss: 0.8580402731895447:  74%|███████▍  | 4787/6470 [57:58<18:06,  1.55it/s]
1|4788|Loss: 0.8580402731895447:  74%|███████▍  | 4788/6470 [57:58<18:47,  1.49it/s]
1|4789|Loss: 0.9894168376922607:  74%|███████▍  | 4788/6470 [57:59<18:47,  1.49it/s]
1|4789|Loss: 0.9894168376922607:  74%|███████▍  | 4789/6470 [57:59<18:42,  1.50it/s]
1|4790|Loss: 0.998157799243927:  74%|███████▍  | 4789/6470 [57:59<18:42,  1.50it/s] 
1|4790|Loss: 0.998157799243927:  74%|███████▍  | 4790/6470 [58:00<18:53,  1.48it/s]
1|4791|Loss: 1.016774296760559:  74%|███████▍  | 4790/6470 [58:00<18:53,  1.48it/s]
1|4791|Loss: 1.016774296760559:  74%|███████▍  | 4791/6470 [58:00<19:27,  1.44it/s]
1|4792|Loss: 0.8133972883224487:  74%|███████▍  | 4791/6470 [58:01<19:27,  1.44it/s]
1|4792|Loss: 0.8133972883224487:  74%|███████▍  | 4792/6470 [58:01<17:57,  1.56it/s]
1|4793|Loss: 0.9038288593292236:  74%|███████▍  | 4792/6470 [58:01<17:57,  1.56it/s]
1|4793|Loss: 0.9038288593292236:  74%|███████▍  | 4793/6470 [58:02<18:58,  1.47it/s]
1|4794|Loss: 0.8587211966514587:  74%|███████▍  | 4793/6470 [58:02<18:58,  1.47it/s]
1|4794|Loss: 0.8587211966514587:  74%|███████▍  | 4794/6470 [58:02<19:37,  1.42it/s]
1|4795|Loss: 0.9360342025756836:  74%|███████▍  | 4794/6470 [58:03<19:37,  1.42it/s]
1|4795|Loss: 0.9360342025756836:  74%|███████▍  | 4795/6470 [58:03<18:21,  1.52it/s]
1|4796|Loss: 0.8540779948234558:  74%|███████▍  | 4795/6470 [58:03<18:21,  1.52it/s]
1|4796|Loss: 0.8540779948234558:  74%|███████▍  | 4796/6470 [58:04<17:38,  1.58it/s]
1|4797|Loss: 0.6454153656959534:  74%|███████▍  | 4796/6470 [58:04<17:38,  1.58it/s]
1|4797|Loss: 0.6454153656959534:  74%|███████▍  | 4797/6470 [58:04<19:45,  1.41it/s]
1|4798|Loss: 0.8006623983383179:  74%|███████▍  | 4797/6470 [58:05<19:45,  1.41it/s]
1|4798|Loss: 0.8006623983383179:  74%|███████▍  | 4798/6470 [58:05<20:23,  1.37it/s]
1|4799|Loss: 0.8885372877120972:  74%|███████▍  | 4798/6470 [58:06<20:23,  1.37it/s]
1|4799|Loss: 0.8885372877120972:  74%|███████▍  | 4799/6470 [58:06<21:13,  1.31it/s]
1|4800|Loss: 0.9678890109062195:  74%|███████▍  | 4799/6470 [58:06<21:13,  1.31it/s]
1|4800|Loss: 0.9678890109062195:  74%|███████▍  | 4800/6470 [58:07<21:02,  1.32it/s]
1|4801|Loss: 0.9342278838157654:  74%|███████▍  | 4800/6470 [58:07<21:02,  1.32it/s]
1|4801|Loss: 0.9342278838157654:  74%|███████▍  | 4801/6470 [58:07<20:34,  1.35it/s]
1|4802|Loss: 0.8630274534225464:  74%|███████▍  | 4801/6470 [58:08<20:34,  1.35it/s]
1|4802|Loss: 0.8630274534225464:  74%|███████▍  | 4802/6470 [58:08<21:14,  1.31it/s]
1|4803|Loss: 1.022033929824829:  74%|███████▍  | 4802/6470 [58:09<21:14,  1.31it/s] 
1|4803|Loss: 1.022033929824829:  74%|███████▍  | 4803/6470 [58:09<22:06,  1.26it/s]
1|4804|Loss: 0.8798803687095642:  74%|███████▍  | 4803/6470 [58:10<22:06,  1.26it/s]
1|4804|Loss: 0.8798803687095642:  74%|███████▍  | 4804/6470 [58:10<21:16,  1.31it/s]
1|4805|Loss: 0.9036064743995667:  74%|███████▍  | 4804/6470 [58:10<21:16,  1.31it/s]
1|4805|Loss: 0.9036064743995667:  74%|███████▍  | 4805/6470 [58:11<21:39,  1.28it/s]
1|4806|Loss: 0.8564367890357971:  74%|███████▍  | 4805/6470 [58:11<21:39,  1.28it/s]
1|4806|Loss: 0.8564367890357971:  74%|███████▍  | 4806/6470 [58:11<20:46,  1.33it/s]
1|4807|Loss: 0.8140387535095215:  74%|███████▍  | 4806/6470 [58:12<20:46,  1.33it/s]
1|4807|Loss: 0.8140387535095215:  74%|███████▍  | 4807/6470 [58:12<19:48,  1.40it/s]
1|4808|Loss: 0.7532413601875305:  74%|███████▍  | 4807/6470 [58:12<19:48,  1.40it/s]
1|4808|Loss: 0.7532413601875305:  74%|███████▍  | 4808/6470 [58:13<18:44,  1.48it/s]
1|4809|Loss: 0.7097138166427612:  74%|███████▍  | 4808/6470 [58:13<18:44,  1.48it/s]
1|4809|Loss: 0.7097138166427612:  74%|███████▍  | 4809/6470 [58:13<18:17,  1.51it/s]
1|4810|Loss: 0.8285489082336426:  74%|███████▍  | 4809/6470 [58:14<18:17,  1.51it/s]
1|4810|Loss: 0.8285489082336426:  74%|███████▍  | 4810/6470 [58:14<17:13,  1.61it/s]
1|4811|Loss: 0.8507347106933594:  74%|███████▍  | 4810/6470 [58:14<17:13,  1.61it/s]
1|4811|Loss: 0.8507347106933594:  74%|███████▍  | 4811/6470 [58:14<17:32,  1.58it/s]
1|4812|Loss: 0.9158415198326111:  74%|███████▍  | 4811/6470 [58:15<17:32,  1.58it/s]
1|4812|Loss: 0.9158415198326111:  74%|███████▍  | 4812/6470 [58:15<18:18,  1.51it/s]
1|4813|Loss: 0.9246880412101746:  74%|███████▍  | 4812/6470 [58:15<18:18,  1.51it/s]
1|4813|Loss: 0.9246880412101746:  74%|███████▍  | 4813/6470 [58:16<18:42,  1.48it/s]
1|4814|Loss: 0.8695248365402222:  74%|███████▍  | 4813/6470 [58:16<18:42,  1.48it/s]
1|4814|Loss: 0.8695248365402222:  74%|███████▍  | 4814/6470 [58:17<18:56,  1.46it/s]
1|4815|Loss: 0.7273600697517395:  74%|███████▍  | 4814/6470 [58:17<18:56,  1.46it/s]
1|4815|Loss: 0.7273600697517395:  74%|███████▍  | 4815/6470 [58:17<19:23,  1.42it/s]
1|4816|Loss: 0.8019505143165588:  74%|███████▍  | 4815/6470 [58:18<19:23,  1.42it/s]
1|4816|Loss: 0.8019505143165588:  74%|███████▍  | 4816/6470 [58:18<19:47,  1.39it/s]
1|4817|Loss: 0.9934526085853577:  74%|███████▍  | 4816/6470 [58:18<19:47,  1.39it/s]
1|4817|Loss: 0.9934526085853577:  74%|███████▍  | 4817/6470 [58:19<19:47,  1.39it/s]
1|4818|Loss: 0.8019199371337891:  74%|███████▍  | 4817/6470 [58:19<19:47,  1.39it/s]
1|4818|Loss: 0.8019199371337891:  74%|███████▍  | 4818/6470 [58:19<19:08,  1.44it/s]
1|4819|Loss: 0.8982228636741638:  74%|███████▍  | 4818/6470 [58:20<19:08,  1.44it/s]
1|4819|Loss: 0.8982228636741638:  74%|███████▍  | 4819/6470 [58:20<19:21,  1.42it/s]
1|4820|Loss: 0.919289231300354:  74%|███████▍  | 4819/6470 [58:20<19:21,  1.42it/s] 
1|4820|Loss: 0.919289231300354:  74%|███████▍  | 4820/6470 [58:21<18:36,  1.48it/s]
1|4821|Loss: 0.7774238586425781:  74%|███████▍  | 4820/6470 [58:21<18:36,  1.48it/s]
1|4821|Loss: 0.7774238586425781:  75%|███████▍  | 4821/6470 [58:21<17:11,  1.60it/s]
1|4822|Loss: 1.0507338047027588:  75%|███████▍  | 4821/6470 [58:22<17:11,  1.60it/s]
1|4822|Loss: 1.0507338047027588:  75%|███████▍  | 4822/6470 [58:22<17:59,  1.53it/s]
1|4823|Loss: 0.8315331339836121:  75%|███████▍  | 4822/6470 [58:22<17:59,  1.53it/s]
1|4823|Loss: 0.8315331339836121:  75%|███████▍  | 4823/6470 [58:23<18:10,  1.51it/s]
1|4824|Loss: 0.9990626573562622:  75%|███████▍  | 4823/6470 [58:23<18:10,  1.51it/s]
1|4824|Loss: 0.9990626573562622:  75%|███████▍  | 4824/6470 [58:23<17:23,  1.58it/s]
1|4825|Loss: 0.8840771913528442:  75%|███████▍  | 4824/6470 [58:24<17:23,  1.58it/s]
1|4825|Loss: 0.8840771913528442:  75%|███████▍  | 4825/6470 [58:24<18:00,  1.52it/s]
1|4826|Loss: 0.8450015187263489:  75%|███████▍  | 4825/6470 [58:24<18:00,  1.52it/s]
1|4826|Loss: 0.8450015187263489:  75%|███████▍  | 4826/6470 [58:25<18:49,  1.46it/s]
1|4827|Loss: 0.8544036149978638:  75%|███████▍  | 4826/6470 [58:25<18:49,  1.46it/s]
1|4827|Loss: 0.8544036149978638:  75%|███████▍  | 4827/6470 [58:25<19:00,  1.44it/s]
1|4828|Loss: 0.9066424369812012:  75%|███████▍  | 4827/6470 [58:26<19:00,  1.44it/s]
1|4828|Loss: 0.9066424369812012:  75%|███████▍  | 4828/6470 [58:26<20:01,  1.37it/s]
1|4829|Loss: 0.8949037790298462:  75%|███████▍  | 4828/6470 [58:27<20:01,  1.37it/s]
1|4829|Loss: 0.8949037790298462:  75%|███████▍  | 4829/6470 [58:27<20:45,  1.32it/s]
1|4830|Loss: 0.9043378829956055:  75%|███████▍  | 4829/6470 [58:27<20:45,  1.32it/s]
1|4830|Loss: 0.9043378829956055:  75%|███████▍  | 4830/6470 [58:28<21:08,  1.29it/s]
1|4831|Loss: 0.7959444522857666:  75%|███████▍  | 4830/6470 [58:28<21:08,  1.29it/s]
1|4831|Loss: 0.7959444522857666:  75%|███████▍  | 4831/6470 [58:28<20:12,  1.35it/s]
1|4832|Loss: 0.941629946231842:  75%|███████▍  | 4831/6470 [58:29<20:12,  1.35it/s] 
1|4832|Loss: 0.941629946231842:  75%|███████▍  | 4832/6470 [58:29<20:01,  1.36it/s]
1|4833|Loss: 0.7923160195350647:  75%|███████▍  | 4832/6470 [58:30<20:01,  1.36it/s]
1|4833|Loss: 0.7923160195350647:  75%|███████▍  | 4833/6470 [58:30<18:14,  1.50it/s]
1|4834|Loss: 0.9334012866020203:  75%|███████▍  | 4833/6470 [58:30<18:14,  1.50it/s]
1|4834|Loss: 0.9334012866020203:  75%|███████▍  | 4834/6470 [58:30<17:14,  1.58it/s]
1|4835|Loss: 0.8423342108726501:  75%|███████▍  | 4834/6470 [58:31<17:14,  1.58it/s]
1|4835|Loss: 0.8423342108726501:  75%|███████▍  | 4835/6470 [58:31<16:00,  1.70it/s]
1|4836|Loss: 0.8946769833564758:  75%|███████▍  | 4835/6470 [58:31<16:00,  1.70it/s]
1|4836|Loss: 0.8946769833564758:  75%|███████▍  | 4836/6470 [58:32<17:55,  1.52it/s]
1|4837|Loss: 0.7477307319641113:  75%|███████▍  | 4836/6470 [58:32<17:55,  1.52it/s]
1|4837|Loss: 0.7477307319641113:  75%|███████▍  | 4837/6470 [58:32<17:30,  1.55it/s]
1|4838|Loss: 0.7423976063728333:  75%|███████▍  | 4837/6470 [58:33<17:30,  1.55it/s]
1|4838|Loss: 0.7423976063728333:  75%|███████▍  | 4838/6470 [58:33<17:52,  1.52it/s]
1|4839|Loss: 0.947489857673645:  75%|███████▍  | 4838/6470 [58:33<17:52,  1.52it/s] 
1|4839|Loss: 0.947489857673645:  75%|███████▍  | 4839/6470 [58:34<18:57,  1.43it/s]
1|4840|Loss: 0.8243712782859802:  75%|███████▍  | 4839/6470 [58:34<18:57,  1.43it/s]
1|4840|Loss: 0.8243712782859802:  75%|███████▍  | 4840/6470 [58:34<19:23,  1.40it/s]
1|4841|Loss: 0.9021810293197632:  75%|███████▍  | 4840/6470 [58:35<19:23,  1.40it/s]
1|4841|Loss: 0.9021810293197632:  75%|███████▍  | 4841/6470 [58:35<18:30,  1.47it/s]
1|4842|Loss: 0.8639920949935913:  75%|███████▍  | 4841/6470 [58:35<18:30,  1.47it/s]
1|4842|Loss: 0.8639920949935913:  75%|███████▍  | 4842/6470 [58:36<17:44,  1.53it/s]
1|4843|Loss: 0.9200187921524048:  75%|███████▍  | 4842/6470 [58:36<17:44,  1.53it/s]
1|4843|Loss: 0.9200187921524048:  75%|███████▍  | 4843/6470 [58:36<16:47,  1.61it/s]
1|4844|Loss: 0.9361270070075989:  75%|███████▍  | 4843/6470 [58:37<16:47,  1.61it/s]
1|4844|Loss: 0.9361270070075989:  75%|███████▍  | 4844/6470 [58:37<17:45,  1.53it/s]
1|4845|Loss: 0.844096839427948:  75%|███████▍  | 4844/6470 [58:37<17:45,  1.53it/s] 
1|4845|Loss: 0.844096839427948:  75%|███████▍  | 4845/6470 [58:38<18:50,  1.44it/s]
1|4846|Loss: 0.8814341425895691:  75%|███████▍  | 4845/6470 [58:38<18:50,  1.44it/s]
1|4846|Loss: 0.8814341425895691:  75%|███████▍  | 4846/6470 [58:38<19:38,  1.38it/s]
1|4847|Loss: 0.968267023563385:  75%|███████▍  | 4846/6470 [58:39<19:38,  1.38it/s] 
1|4847|Loss: 0.968267023563385:  75%|███████▍  | 4847/6470 [58:39<19:03,  1.42it/s]
1|4848|Loss: 0.8002467751502991:  75%|███████▍  | 4847/6470 [58:39<19:03,  1.42it/s]
1|4848|Loss: 0.8002467751502991:  75%|███████▍  | 4848/6470 [58:40<18:55,  1.43it/s]
1|4849|Loss: 1.0145763158798218:  75%|███████▍  | 4848/6470 [58:40<18:55,  1.43it/s]
1|4849|Loss: 1.0145763158798218:  75%|███████▍  | 4849/6470 [58:41<19:09,  1.41it/s]
1|4850|Loss: 0.960609495639801:  75%|███████▍  | 4849/6470 [58:41<19:09,  1.41it/s] 
1|4850|Loss: 0.960609495639801:  75%|███████▍  | 4850/6470 [58:41<19:20,  1.40it/s]
1|4851|Loss: 0.7991888523101807:  75%|███████▍  | 4850/6470 [58:42<19:20,  1.40it/s]
1|4851|Loss: 0.7991888523101807:  75%|███████▍  | 4851/6470 [58:42<17:49,  1.51it/s]
1|4852|Loss: 0.8343640565872192:  75%|███████▍  | 4851/6470 [58:42<17:49,  1.51it/s]
1|4852|Loss: 0.8343640565872192:  75%|███████▍  | 4852/6470 [58:43<18:33,  1.45it/s]
1|4853|Loss: 0.82004314661026:  75%|███████▍  | 4852/6470 [58:43<18:33,  1.45it/s]  
1|4853|Loss: 0.82004314661026:  75%|███████▌  | 4853/6470 [58:43<19:11,  1.40it/s]
1|4854|Loss: 0.8194140791893005:  75%|███████▌  | 4853/6470 [58:44<19:11,  1.40it/s]
1|4854|Loss: 0.8194140791893005:  75%|███████▌  | 4854/6470 [58:44<18:41,  1.44it/s]
1|4855|Loss: 1.0699748992919922:  75%|███████▌  | 4854/6470 [58:44<18:41,  1.44it/s]
1|4855|Loss: 1.0699748992919922:  75%|███████▌  | 4855/6470 [58:45<19:24,  1.39it/s]
1|4856|Loss: 0.9584594368934631:  75%|███████▌  | 4855/6470 [58:45<19:24,  1.39it/s]
1|4856|Loss: 0.9584594368934631:  75%|███████▌  | 4856/6470 [58:46<19:37,  1.37it/s]
1|4857|Loss: 0.8308720588684082:  75%|███████▌  | 4856/6470 [58:46<19:37,  1.37it/s]
1|4857|Loss: 0.8308720588684082:  75%|███████▌  | 4857/6470 [58:46<19:04,  1.41it/s]
1|4858|Loss: 0.7767503261566162:  75%|███████▌  | 4857/6470 [58:47<19:04,  1.41it/s]
1|4858|Loss: 0.7767503261566162:  75%|███████▌  | 4858/6470 [58:47<18:31,  1.45it/s]
1|4859|Loss: 0.8099526762962341:  75%|███████▌  | 4858/6470 [58:47<18:31,  1.45it/s]
1|4859|Loss: 0.8099526762962341:  75%|███████▌  | 4859/6470 [58:47<17:43,  1.51it/s]
1|4860|Loss: 0.921338677406311:  75%|███████▌  | 4859/6470 [58:48<17:43,  1.51it/s] 
1|4860|Loss: 0.921338677406311:  75%|███████▌  | 4860/6470 [58:48<18:01,  1.49it/s]
1|4861|Loss: 0.5624483227729797:  75%|███████▌  | 4860/6470 [58:49<18:01,  1.49it/s]
1|4861|Loss: 0.5624483227729797:  75%|███████▌  | 4861/6470 [58:49<19:55,  1.35it/s]
1|4862|Loss: 0.6488863229751587:  75%|███████▌  | 4861/6470 [58:49<19:55,  1.35it/s]
1|4862|Loss: 0.6488863229751587:  75%|███████▌  | 4862/6470 [58:50<20:21,  1.32it/s]
1|4863|Loss: 0.8527769446372986:  75%|███████▌  | 4862/6470 [58:50<20:21,  1.32it/s]
1|4863|Loss: 0.8527769446372986:  75%|███████▌  | 4863/6470 [58:50<19:31,  1.37it/s]
1|4864|Loss: 1.066388487815857:  75%|███████▌  | 4863/6470 [58:51<19:31,  1.37it/s] 
1|4864|Loss: 1.066388487815857:  75%|███████▌  | 4864/6470 [58:51<18:36,  1.44it/s]
1|4865|Loss: 0.9028241634368896:  75%|███████▌  | 4864/6470 [58:51<18:36,  1.44it/s]
1|4865|Loss: 0.9028241634368896:  75%|███████▌  | 4865/6470 [58:52<17:06,  1.56it/s]
1|4866|Loss: 0.7834829092025757:  75%|███████▌  | 4865/6470 [58:52<17:06,  1.56it/s]
1|4866|Loss: 0.7834829092025757:  75%|███████▌  | 4866/6470 [58:52<16:15,  1.64it/s]
1|4867|Loss: 0.8094277381896973:  75%|███████▌  | 4866/6470 [58:52<16:15,  1.64it/s]
1|4867|Loss: 0.8094277381896973:  75%|███████▌  | 4867/6470 [58:53<15:49,  1.69it/s]
1|4868|Loss: 0.9683346748352051:  75%|███████▌  | 4867/6470 [58:53<15:49,  1.69it/s]
1|4868|Loss: 0.9683346748352051:  75%|███████▌  | 4868/6470 [58:53<16:08,  1.65it/s]
1|4869|Loss: 0.7877138257026672:  75%|███████▌  | 4868/6470 [58:54<16:08,  1.65it/s]
1|4869|Loss: 0.7877138257026672:  75%|███████▌  | 4869/6470 [58:54<16:07,  1.65it/s]
1|4870|Loss: 0.8156761527061462:  75%|███████▌  | 4869/6470 [58:54<16:07,  1.65it/s]
1|4870|Loss: 0.8156761527061462:  75%|███████▌  | 4870/6470 [58:54<14:38,  1.82it/s]
1|4871|Loss: 1.001308560371399:  75%|███████▌  | 4870/6470 [58:55<14:38,  1.82it/s] 
1|4871|Loss: 1.001308560371399:  75%|███████▌  | 4871/6470 [58:55<14:55,  1.79it/s]
1|4872|Loss: 0.8352639079093933:  75%|███████▌  | 4871/6470 [58:55<14:55,  1.79it/s]
1|4872|Loss: 0.8352639079093933:  75%|███████▌  | 4872/6470 [58:56<15:35,  1.71it/s]
1|4873|Loss: 0.8427357077598572:  75%|███████▌  | 4872/6470 [58:56<15:35,  1.71it/s]
1|4873|Loss: 0.8427357077598572:  75%|███████▌  | 4873/6470 [58:56<16:36,  1.60it/s]
1|4874|Loss: 0.9245933890342712:  75%|███████▌  | 4873/6470 [58:57<16:36,  1.60it/s]
1|4874|Loss: 0.9245933890342712:  75%|███████▌  | 4874/6470 [58:57<17:18,  1.54it/s]
1|4875|Loss: 0.799493670463562:  75%|███████▌  | 4874/6470 [58:57<17:18,  1.54it/s] 
1|4875|Loss: 0.799493670463562:  75%|███████▌  | 4875/6470 [58:58<17:42,  1.50it/s]
1|4876|Loss: 0.7748068571090698:  75%|███████▌  | 4875/6470 [58:58<17:42,  1.50it/s]
1|4876|Loss: 0.7748068571090698:  75%|███████▌  | 4876/6470 [58:58<18:18,  1.45it/s]
1|4877|Loss: 1.1231064796447754:  75%|███████▌  | 4876/6470 [58:59<18:18,  1.45it/s]
1|4877|Loss: 1.1231064796447754:  75%|███████▌  | 4877/6470 [58:59<18:09,  1.46it/s]
1|4878|Loss: 1.0025756359100342:  75%|███████▌  | 4877/6470 [59:00<18:09,  1.46it/s]
1|4878|Loss: 1.0025756359100342:  75%|███████▌  | 4878/6470 [59:00<18:20,  1.45it/s]
1|4879|Loss: 0.7233527302742004:  75%|███████▌  | 4878/6470 [59:00<18:20,  1.45it/s]
1|4879|Loss: 0.7233527302742004:  75%|███████▌  | 4879/6470 [59:01<18:48,  1.41it/s]
1|4880|Loss: 0.6526962518692017:  75%|███████▌  | 4879/6470 [59:01<18:48,  1.41it/s]
1|4880|Loss: 0.6526962518692017:  75%|███████▌  | 4880/6470 [59:01<19:27,  1.36it/s]
1|4881|Loss: 0.9324353933334351:  75%|███████▌  | 4880/6470 [59:02<19:27,  1.36it/s]
1|4881|Loss: 0.9324353933334351:  75%|███████▌  | 4881/6470 [59:02<18:50,  1.41it/s]
1|4882|Loss: 0.8524459600448608:  75%|███████▌  | 4881/6470 [59:02<18:50,  1.41it/s]
1|4882|Loss: 0.8524459600448608:  75%|███████▌  | 4882/6470 [59:03<18:11,  1.45it/s]
1|4883|Loss: 0.9551648497581482:  75%|███████▌  | 4882/6470 [59:03<18:11,  1.45it/s]
1|4883|Loss: 0.9551648497581482:  75%|███████▌  | 4883/6470 [59:03<18:10,  1.45it/s]
1|4884|Loss: 0.8918814063072205:  75%|███████▌  | 4883/6470 [59:04<18:10,  1.45it/s]
1|4884|Loss: 0.8918814063072205:  75%|███████▌  | 4884/6470 [59:04<17:44,  1.49it/s]
1|4885|Loss: 0.6782782077789307:  75%|███████▌  | 4884/6470 [59:04<17:44,  1.49it/s]
1|4885|Loss: 0.6782782077789307:  76%|███████▌  | 4885/6470 [59:05<19:29,  1.36it/s]
1|4886|Loss: 0.9307968020439148:  76%|███████▌  | 4885/6470 [59:05<19:29,  1.36it/s]
1|4886|Loss: 0.9307968020439148:  76%|███████▌  | 4886/6470 [59:06<19:41,  1.34it/s]
1|4887|Loss: 0.9755609035491943:  76%|███████▌  | 4886/6470 [59:06<19:41,  1.34it/s]
1|4887|Loss: 0.9755609035491943:  76%|███████▌  | 4887/6470 [59:06<20:23,  1.29it/s]
1|4888|Loss: 0.9667342305183411:  76%|███████▌  | 4887/6470 [59:07<20:23,  1.29it/s]
1|4888|Loss: 0.9667342305183411:  76%|███████▌  | 4888/6470 [59:07<20:57,  1.26it/s]
1|4889|Loss: 0.880567729473114:  76%|███████▌  | 4888/6470 [59:08<20:57,  1.26it/s] 
1|4889|Loss: 0.880567729473114:  76%|███████▌  | 4889/6470 [59:08<20:26,  1.29it/s]
1|4890|Loss: 1.0410525798797607:  76%|███████▌  | 4889/6470 [59:08<20:26,  1.29it/s]
1|4890|Loss: 1.0410525798797607:  76%|███████▌  | 4890/6470 [59:09<20:12,  1.30it/s]
1|4891|Loss: 0.8397533297538757:  76%|███████▌  | 4890/6470 [59:09<20:12,  1.30it/s]
1|4891|Loss: 0.8397533297538757:  76%|███████▌  | 4891/6470 [59:10<19:58,  1.32it/s]
1|4892|Loss: 0.9206019043922424:  76%|███████▌  | 4891/6470 [59:10<19:58,  1.32it/s]
1|4892|Loss: 0.9206019043922424:  76%|███████▌  | 4892/6470 [59:10<19:35,  1.34it/s]
1|4893|Loss: 0.9638790488243103:  76%|███████▌  | 4892/6470 [59:11<19:35,  1.34it/s]
1|4893|Loss: 0.9638790488243103:  76%|███████▌  | 4893/6470 [59:11<17:52,  1.47it/s]
1|4894|Loss: 0.8674150705337524:  76%|███████▌  | 4893/6470 [59:11<17:52,  1.47it/s]
1|4894|Loss: 0.8674150705337524:  76%|███████▌  | 4894/6470 [59:11<17:17,  1.52it/s]
1|4895|Loss: 0.7861481308937073:  76%|███████▌  | 4894/6470 [59:12<17:17,  1.52it/s]
1|4895|Loss: 0.7861481308937073:  76%|███████▌  | 4895/6470 [59:12<17:32,  1.50it/s]
1|4896|Loss: 0.9298960566520691:  76%|███████▌  | 4895/6470 [59:12<17:32,  1.50it/s]
1|4896|Loss: 0.9298960566520691:  76%|███████▌  | 4896/6470 [59:13<17:45,  1.48it/s]
1|4897|Loss: 0.8273290395736694:  76%|███████▌  | 4896/6470 [59:13<17:45,  1.48it/s]
1|4897|Loss: 0.8273290395736694:  76%|███████▌  | 4897/6470 [59:13<17:56,  1.46it/s]
1|4898|Loss: 0.8495076298713684:  76%|███████▌  | 4897/6470 [59:14<17:56,  1.46it/s]
1|4898|Loss: 0.8495076298713684:  76%|███████▌  | 4898/6470 [59:14<17:39,  1.48it/s]
1|4899|Loss: 0.7738591432571411:  76%|███████▌  | 4898/6470 [59:15<17:39,  1.48it/s]
1|4899|Loss: 0.7738591432571411:  76%|███████▌  | 4899/6470 [59:15<19:24,  1.35it/s]
1|4900|Loss: 0.9908685684204102:  76%|███████▌  | 4899/6470 [59:15<19:24,  1.35it/s]
1|4900|Loss: 0.9908685684204102:  76%|███████▌  | 4900/6470 [59:16<19:52,  1.32it/s]
1|4901|Loss: 0.7801468968391418:  76%|███████▌  | 4900/6470 [59:16<19:52,  1.32it/s]
1|4901|Loss: 0.7801468968391418:  76%|███████▌  | 4901/6470 [59:16<18:42,  1.40it/s]
1|4902|Loss: 0.9946985840797424:  76%|███████▌  | 4901/6470 [59:17<18:42,  1.40it/s]
1|4902|Loss: 0.9946985840797424:  76%|███████▌  | 4902/6470 [59:17<18:58,  1.38it/s]
1|4903|Loss: 0.8402075171470642:  76%|███████▌  | 4902/6470 [59:18<18:58,  1.38it/s]
1|4903|Loss: 0.8402075171470642:  76%|███████▌  | 4903/6470 [59:18<18:36,  1.40it/s]
1|4904|Loss: 0.748818039894104:  76%|███████▌  | 4903/6470 [59:18<18:36,  1.40it/s] 
1|4904|Loss: 0.748818039894104:  76%|███████▌  | 4904/6470 [59:18<17:54,  1.46it/s]
1|4905|Loss: 0.7628677487373352:  76%|███████▌  | 4904/6470 [59:19<17:54,  1.46it/s]
1|4905|Loss: 0.7628677487373352:  76%|███████▌  | 4905/6470 [59:19<17:57,  1.45it/s]
1|4906|Loss: 0.8694214820861816:  76%|███████▌  | 4905/6470 [59:20<17:57,  1.45it/s]
1|4906|Loss: 0.8694214820861816:  76%|███████▌  | 4906/6470 [59:20<18:15,  1.43it/s]
1|4907|Loss: 0.8365292549133301:  76%|███████▌  | 4906/6470 [59:20<18:15,  1.43it/s]
1|4907|Loss: 0.8365292549133301:  76%|███████▌  | 4907/6470 [59:21<20:06,  1.30it/s]
1|4908|Loss: 0.6162055134773254:  76%|███████▌  | 4907/6470 [59:21<20:06,  1.30it/s]
1|4908|Loss: 0.6162055134773254:  76%|███████▌  | 4908/6470 [59:22<22:04,  1.18it/s]
1|4909|Loss: 0.7935760617256165:  76%|███████▌  | 4908/6470 [59:22<22:04,  1.18it/s]
1|4909|Loss: 0.7935760617256165:  76%|███████▌  | 4909/6470 [59:23<20:46,  1.25it/s]
1|4910|Loss: 0.899211049079895:  76%|███████▌  | 4909/6470 [59:23<20:46,  1.25it/s] 
1|4910|Loss: 0.899211049079895:  76%|███████▌  | 4910/6470 [59:23<19:17,  1.35it/s]
1|4911|Loss: 1.0683468580245972:  76%|███████▌  | 4910/6470 [59:24<19:17,  1.35it/s]
1|4911|Loss: 1.0683468580245972:  76%|███████▌  | 4911/6470 [59:24<19:21,  1.34it/s]
1|4912|Loss: 0.9405269026756287:  76%|███████▌  | 4911/6470 [59:24<19:21,  1.34it/s]
1|4912|Loss: 0.9405269026756287:  76%|███████▌  | 4912/6470 [59:25<19:11,  1.35it/s]
1|4913|Loss: 1.037410855293274:  76%|███████▌  | 4912/6470 [59:25<19:11,  1.35it/s] 
1|4913|Loss: 1.037410855293274:  76%|███████▌  | 4913/6470 [59:25<18:55,  1.37it/s]
1|4914|Loss: 0.8758092522621155:  76%|███████▌  | 4913/6470 [59:26<18:55,  1.37it/s]
1|4914|Loss: 0.8758092522621155:  76%|███████▌  | 4914/6470 [59:26<17:35,  1.47it/s]
1|4915|Loss: 0.8973584771156311:  76%|███████▌  | 4914/6470 [59:26<17:35,  1.47it/s]
1|4915|Loss: 0.8973584771156311:  76%|███████▌  | 4915/6470 [59:27<16:57,  1.53it/s]
1|4916|Loss: 0.855546236038208:  76%|███████▌  | 4915/6470 [59:27<16:57,  1.53it/s] 
1|4916|Loss: 0.855546236038208:  76%|███████▌  | 4916/6470 [59:27<17:27,  1.48it/s]
1|4917|Loss: 1.1205772161483765:  76%|███████▌  | 4916/6470 [59:28<17:27,  1.48it/s]
1|4917|Loss: 1.1205772161483765:  76%|███████▌  | 4917/6470 [59:28<17:23,  1.49it/s]
1|4918|Loss: 0.841236412525177:  76%|███████▌  | 4917/6470 [59:28<17:23,  1.49it/s] 
1|4918|Loss: 0.841236412525177:  76%|███████▌  | 4918/6470 [59:29<16:58,  1.52it/s]
1|4919|Loss: 0.9086341261863708:  76%|███████▌  | 4918/6470 [59:29<16:58,  1.52it/s]
1|4919|Loss: 0.9086341261863708:  76%|███████▌  | 4919/6470 [59:29<16:38,  1.55it/s]
1|4920|Loss: 0.7840156555175781:  76%|███████▌  | 4919/6470 [59:30<16:38,  1.55it/s]
1|4920|Loss: 0.7840156555175781:  76%|███████▌  | 4920/6470 [59:30<17:44,  1.46it/s]
1|4921|Loss: 0.8104158639907837:  76%|███████▌  | 4920/6470 [59:30<17:44,  1.46it/s]
1|4921|Loss: 0.8104158639907837:  76%|███████▌  | 4921/6470 [59:31<17:39,  1.46it/s]
1|4922|Loss: 0.8829807639122009:  76%|███████▌  | 4921/6470 [59:31<17:39,  1.46it/s]
1|4922|Loss: 0.8829807639122009:  76%|███████▌  | 4922/6470 [59:31<18:44,  1.38it/s]
1|4923|Loss: 0.8098766207695007:  76%|███████▌  | 4922/6470 [59:32<18:44,  1.38it/s]
1|4923|Loss: 0.8098766207695007:  76%|███████▌  | 4923/6470 [59:32<20:00,  1.29it/s]
1|4924|Loss: 0.8162343502044678:  76%|███████▌  | 4923/6470 [59:33<20:00,  1.29it/s]
1|4924|Loss: 0.8162343502044678:  76%|███████▌  | 4924/6470 [59:33<18:35,  1.39it/s]
1|4925|Loss: 0.9251658916473389:  76%|███████▌  | 4924/6470 [59:33<18:35,  1.39it/s]
1|4925|Loss: 0.9251658916473389:  76%|███████▌  | 4925/6470 [59:34<17:43,  1.45it/s]
1|4926|Loss: 0.8951677680015564:  76%|███████▌  | 4925/6470 [59:34<17:43,  1.45it/s]
1|4926|Loss: 0.8951677680015564:  76%|███████▌  | 4926/6470 [59:34<19:17,  1.33it/s]
1|4927|Loss: 0.8408606648445129:  76%|███████▌  | 4926/6470 [59:35<19:17,  1.33it/s]
1|4927|Loss: 0.8408606648445129:  76%|███████▌  | 4927/6470 [59:35<19:33,  1.32it/s]
1|4928|Loss: 0.9578821063041687:  76%|███████▌  | 4927/6470 [59:36<19:33,  1.32it/s]
1|4928|Loss: 0.9578821063041687:  76%|███████▌  | 4928/6470 [59:36<19:22,  1.33it/s]
1|4929|Loss: 0.7949243187904358:  76%|███████▌  | 4928/6470 [59:36<19:22,  1.33it/s]
1|4929|Loss: 0.7949243187904358:  76%|███████▌  | 4929/6470 [59:37<18:03,  1.42it/s]
1|4930|Loss: 0.8585620522499084:  76%|███████▌  | 4929/6470 [59:37<18:03,  1.42it/s]
1|4930|Loss: 0.8585620522499084:  76%|███████▌  | 4930/6470 [59:37<16:16,  1.58it/s]
1|4931|Loss: 1.26613187789917:  76%|███████▌  | 4930/6470 [59:37<16:16,  1.58it/s]  
1|4931|Loss: 1.26613187789917:  76%|███████▌  | 4931/6470 [59:38<17:17,  1.48it/s]
1|4932|Loss: 1.0665407180786133:  76%|███████▌  | 4931/6470 [59:38<17:17,  1.48it/s]
1|4932|Loss: 1.0665407180786133:  76%|███████▌  | 4932/6470 [59:38<17:25,  1.47it/s]
1|4933|Loss: 0.7761879563331604:  76%|███████▌  | 4932/6470 [59:39<17:25,  1.47it/s]
1|4933|Loss: 0.7761879563331604:  76%|███████▌  | 4933/6470 [59:39<17:39,  1.45it/s]
1|4934|Loss: 0.9812115430831909:  76%|███████▌  | 4933/6470 [59:40<17:39,  1.45it/s]
1|4934|Loss: 0.9812115430831909:  76%|███████▋  | 4934/6470 [59:40<17:46,  1.44it/s]
1|4935|Loss: 0.7723071575164795:  76%|███████▋  | 4934/6470 [59:40<17:46,  1.44it/s]
1|4935|Loss: 0.7723071575164795:  76%|███████▋  | 4935/6470 [59:41<17:49,  1.43it/s]
1|4936|Loss: 1.075871229171753:  76%|███████▋  | 4935/6470 [59:41<17:49,  1.43it/s] 
1|4936|Loss: 1.075871229171753:  76%|███████▋  | 4936/6470 [59:41<16:40,  1.53it/s]
1|4937|Loss: 0.8422460556030273:  76%|███████▋  | 4936/6470 [59:41<16:40,  1.53it/s]
1|4937|Loss: 0.8422460556030273:  76%|███████▋  | 4937/6470 [59:42<16:41,  1.53it/s]
1|4938|Loss: 0.9074966311454773:  76%|███████▋  | 4937/6470 [59:42<16:41,  1.53it/s]
1|4938|Loss: 0.9074966311454773:  76%|███████▋  | 4938/6470 [59:43<17:55,  1.43it/s]
1|4939|Loss: 0.76795893907547:  76%|███████▋  | 4938/6470 [59:43<17:55,  1.43it/s]  
1|4939|Loss: 0.76795893907547:  76%|███████▋  | 4939/6470 [59:43<17:04,  1.49it/s]
1|4940|Loss: 0.7626365423202515:  76%|███████▋  | 4939/6470 [59:44<17:04,  1.49it/s]
1|4940|Loss: 0.7626365423202515:  76%|███████▋  | 4940/6470 [59:44<17:00,  1.50it/s]
1|4941|Loss: 0.9191898107528687:  76%|███████▋  | 4940/6470 [59:44<17:00,  1.50it/s]
1|4941|Loss: 0.9191898107528687:  76%|███████▋  | 4941/6470 [59:44<15:57,  1.60it/s]
1|4942|Loss: 0.8316158652305603:  76%|███████▋  | 4941/6470 [59:45<15:57,  1.60it/s]
1|4942|Loss: 0.8316158652305603:  76%|███████▋  | 4942/6470 [59:45<15:40,  1.62it/s]
1|4943|Loss: 0.8555120229721069:  76%|███████▋  | 4942/6470 [59:45<15:40,  1.62it/s]
1|4943|Loss: 0.8555120229721069:  76%|███████▋  | 4943/6470 [59:46<15:00,  1.70it/s]
1|4944|Loss: 0.95680832862854:  76%|███████▋  | 4943/6470 [59:46<15:00,  1.70it/s]  
1|4944|Loss: 0.95680832862854:  76%|███████▋  | 4944/6470 [59:46<15:43,  1.62it/s]
1|4945|Loss: 0.8326923251152039:  76%|███████▋  | 4944/6470 [59:47<15:43,  1.62it/s]
1|4945|Loss: 0.8326923251152039:  76%|███████▋  | 4945/6470 [59:47<16:24,  1.55it/s]
1|4946|Loss: 0.8923508524894714:  76%|███████▋  | 4945/6470 [59:47<16:24,  1.55it/s]
1|4946|Loss: 0.8923508524894714:  76%|███████▋  | 4946/6470 [59:48<16:16,  1.56it/s]
1|4947|Loss: 0.8820895552635193:  76%|███████▋  | 4946/6470 [59:48<16:16,  1.56it/s]
1|4947|Loss: 0.8820895552635193:  76%|███████▋  | 4947/6470 [59:48<16:02,  1.58it/s]
1|4948|Loss: 1.1584582328796387:  76%|███████▋  | 4947/6470 [59:49<16:02,  1.58it/s]
1|4948|Loss: 1.1584582328796387:  76%|███████▋  | 4948/6470 [59:49<17:30,  1.45it/s]
1|4949|Loss: 0.8727625608444214:  76%|███████▋  | 4948/6470 [59:49<17:30,  1.45it/s]
1|4949|Loss: 0.8727625608444214:  76%|███████▋  | 4949/6470 [59:50<17:54,  1.42it/s]
1|4950|Loss: 0.8442264795303345:  76%|███████▋  | 4949/6470 [59:50<17:54,  1.42it/s]
1|4950|Loss: 0.8442264795303345:  77%|███████▋  | 4950/6470 [59:50<18:03,  1.40it/s]
1|4951|Loss: 0.8109422922134399:  77%|███████▋  | 4950/6470 [59:51<18:03,  1.40it/s]
1|4951|Loss: 0.8109422922134399:  77%|███████▋  | 4951/6470 [59:51<18:00,  1.41it/s]
1|4952|Loss: 0.8670726418495178:  77%|███████▋  | 4951/6470 [59:51<18:00,  1.41it/s]
1|4952|Loss: 0.8670726418495178:  77%|███████▋  | 4952/6470 [59:52<16:51,  1.50it/s]
1|4953|Loss: 0.8802779316902161:  77%|███████▋  | 4952/6470 [59:52<16:51,  1.50it/s]
1|4953|Loss: 0.8802779316902161:  77%|███████▋  | 4953/6470 [59:52<15:21,  1.65it/s]
1|4954|Loss: 1.0678508281707764:  77%|███████▋  | 4953/6470 [59:52<15:21,  1.65it/s]
1|4954|Loss: 1.0678508281707764:  77%|███████▋  | 4954/6470 [59:53<14:56,  1.69it/s]
1|4955|Loss: 0.9880916476249695:  77%|███████▋  | 4954/6470 [59:53<14:56,  1.69it/s]
1|4955|Loss: 0.9880916476249695:  77%|███████▋  | 4955/6470 [59:53<14:06,  1.79it/s]
1|4956|Loss: 0.9820187091827393:  77%|███████▋  | 4955/6470 [59:54<14:06,  1.79it/s]
1|4956|Loss: 0.9820187091827393:  77%|███████▋  | 4956/6470 [59:54<13:43,  1.84it/s]
1|4957|Loss: 0.7892856597900391:  77%|███████▋  | 4956/6470 [59:54<13:43,  1.84it/s]
1|4957|Loss: 0.7892856597900391:  77%|███████▋  | 4957/6470 [59:54<13:37,  1.85it/s]
1|4958|Loss: 0.9231553077697754:  77%|███████▋  | 4957/6470 [59:55<13:37,  1.85it/s]
1|4958|Loss: 0.9231553077697754:  77%|███████▋  | 4958/6470 [59:55<14:05,  1.79it/s]
1|4959|Loss: 0.9446712732315063:  77%|███████▋  | 4958/6470 [59:55<14:05,  1.79it/s]
1|4959|Loss: 0.9446712732315063:  77%|███████▋  | 4959/6470 [59:55<14:30,  1.74it/s]
1|4960|Loss: 0.9131321907043457:  77%|███████▋  | 4959/6470 [59:56<14:30,  1.74it/s]
1|4960|Loss: 0.9131321907043457:  77%|███████▋  | 4960/6470 [59:56<15:19,  1.64it/s]
1|4961|Loss: 0.9253087639808655:  77%|███████▋  | 4960/6470 [59:57<15:19,  1.64it/s]
1|4961|Loss: 0.9253087639808655:  77%|███████▋  | 4961/6470 [59:57<16:01,  1.57it/s]
1|4962|Loss: 0.9134774804115295:  77%|███████▋  | 4961/6470 [59:57<16:01,  1.57it/s]
1|4962|Loss: 0.9134774804115295:  77%|███████▋  | 4962/6470 [59:58<17:03,  1.47it/s]
1|4963|Loss: 1.0122268199920654:  77%|███████▋  | 4962/6470 [59:58<17:03,  1.47it/s]
1|4963|Loss: 1.0122268199920654:  77%|███████▋  | 4963/6470 [59:58<17:58,  1.40it/s]
1|4964|Loss: 0.8742557764053345:  77%|███████▋  | 4963/6470 [59:59<17:58,  1.40it/s]
1|4964|Loss: 0.8742557764053345:  77%|███████▋  | 4964/6470 [59:59<17:33,  1.43it/s]
1|4965|Loss: 0.9449318051338196:  77%|███████▋  | 4964/6470 [1:00:00<17:33,  1.43it/s]
1|4965|Loss: 0.9449318051338196:  77%|███████▋  | 4965/6470 [1:00:00<18:19,  1.37it/s]
1|4966|Loss: 0.826190173625946:  77%|███████▋  | 4965/6470 [1:00:00<18:19,  1.37it/s] 
1|4966|Loss: 0.826190173625946:  77%|███████▋  | 4966/6470 [1:00:00<16:51,  1.49it/s]
1|4967|Loss: 0.9420292377471924:  77%|███████▋  | 4966/6470 [1:00:01<16:51,  1.49it/s]
1|4967|Loss: 0.9420292377471924:  77%|███████▋  | 4967/6470 [1:00:01<15:52,  1.58it/s]
1|4968|Loss: 1.0100579261779785:  77%|███████▋  | 4967/6470 [1:00:01<15:52,  1.58it/s]
1|4968|Loss: 1.0100579261779785:  77%|███████▋  | 4968/6470 [1:00:01<14:17,  1.75it/s]
1|4969|Loss: 0.807231605052948:  77%|███████▋  | 4968/6470 [1:00:02<14:17,  1.75it/s] 
1|4969|Loss: 0.807231605052948:  77%|███████▋  | 4969/6470 [1:00:02<14:21,  1.74it/s]
1|4970|Loss: 0.8800105452537537:  77%|███████▋  | 4969/6470 [1:00:02<14:21,  1.74it/s]
1|4970|Loss: 0.8800105452537537:  77%|███████▋  | 4970/6470 [1:00:03<14:22,  1.74it/s]
1|4971|Loss: 0.9576883316040039:  77%|███████▋  | 4970/6470 [1:00:03<14:22,  1.74it/s]
1|4971|Loss: 0.9576883316040039:  77%|███████▋  | 4971/6470 [1:00:03<14:40,  1.70it/s]
1|4972|Loss: 0.772972583770752:  77%|███████▋  | 4971/6470 [1:00:04<14:40,  1.70it/s] 
1|4972|Loss: 0.772972583770752:  77%|███████▋  | 4972/6470 [1:00:04<14:15,  1.75it/s]
1|4973|Loss: 0.8126717805862427:  77%|███████▋  | 4972/6470 [1:00:04<14:15,  1.75it/s]
1|4973|Loss: 0.8126717805862427:  77%|███████▋  | 4973/6470 [1:00:04<15:06,  1.65it/s]
1|4974|Loss: 0.8730989098548889:  77%|███████▋  | 4973/6470 [1:00:05<15:06,  1.65it/s]
1|4974|Loss: 0.8730989098548889:  77%|███████▋  | 4974/6470 [1:00:05<16:02,  1.55it/s]
1|4975|Loss: 0.9863125085830688:  77%|███████▋  | 4974/6470 [1:00:06<16:02,  1.55it/s]
1|4975|Loss: 0.9863125085830688:  77%|███████▋  | 4975/6470 [1:00:06<16:33,  1.51it/s]
1|4976|Loss: 0.7905118465423584:  77%|███████▋  | 4975/6470 [1:00:06<16:33,  1.51it/s]
1|4976|Loss: 0.7905118465423584:  77%|███████▋  | 4976/6470 [1:00:07<17:05,  1.46it/s]
1|4977|Loss: 0.9837303161621094:  77%|███████▋  | 4976/6470 [1:00:07<17:05,  1.46it/s]
1|4977|Loss: 0.9837303161621094:  77%|███████▋  | 4977/6470 [1:00:07<17:20,  1.44it/s]
1|4978|Loss: 0.8221632838249207:  77%|███████▋  | 4977/6470 [1:00:08<17:20,  1.44it/s]
1|4978|Loss: 0.8221632838249207:  77%|███████▋  | 4978/6470 [1:00:08<17:22,  1.43it/s]
1|4979|Loss: 0.8730732202529907:  77%|███████▋  | 4978/6470 [1:00:08<17:22,  1.43it/s]
1|4979|Loss: 0.8730732202529907:  77%|███████▋  | 4979/6470 [1:00:09<16:19,  1.52it/s]
1|4980|Loss: 0.7984761595726013:  77%|███████▋  | 4979/6470 [1:00:09<16:19,  1.52it/s]
1|4980|Loss: 0.7984761595726013:  77%|███████▋  | 4980/6470 [1:00:09<16:36,  1.50it/s]
1|4981|Loss: 0.9285603761672974:  77%|███████▋  | 4980/6470 [1:00:10<16:36,  1.50it/s]
1|4981|Loss: 0.9285603761672974:  77%|███████▋  | 4981/6470 [1:00:10<16:11,  1.53it/s]
1|4982|Loss: 0.8732925653457642:  77%|███████▋  | 4981/6470 [1:00:10<16:11,  1.53it/s]
1|4982|Loss: 0.8732925653457642:  77%|███████▋  | 4982/6470 [1:00:10<15:18,  1.62it/s]
1|4983|Loss: 0.8461240530014038:  77%|███████▋  | 4982/6470 [1:00:11<15:18,  1.62it/s]
1|4983|Loss: 0.8461240530014038:  77%|███████▋  | 4983/6470 [1:00:11<15:19,  1.62it/s]
1|4984|Loss: 0.8380086421966553:  77%|███████▋  | 4983/6470 [1:00:11<15:19,  1.62it/s]
1|4984|Loss: 0.8380086421966553:  77%|███████▋  | 4984/6470 [1:00:12<15:30,  1.60it/s]
1|4985|Loss: 0.8797754049301147:  77%|███████▋  | 4984/6470 [1:00:12<15:30,  1.60it/s]
1|4985|Loss: 0.8797754049301147:  77%|███████▋  | 4985/6470 [1:00:12<15:36,  1.59it/s]
1|4986|Loss: 0.8375323414802551:  77%|███████▋  | 4985/6470 [1:00:13<15:36,  1.59it/s]
1|4986|Loss: 0.8375323414802551:  77%|███████▋  | 4986/6470 [1:00:13<15:38,  1.58it/s]
1|4987|Loss: 1.0039511919021606:  77%|███████▋  | 4986/6470 [1:00:13<15:38,  1.58it/s]
1|4987|Loss: 1.0039511919021606:  77%|███████▋  | 4987/6470 [1:00:14<16:56,  1.46it/s]
1|4988|Loss: 0.8755490779876709:  77%|███████▋  | 4987/6470 [1:00:14<16:56,  1.46it/s]
1|4988|Loss: 0.8755490779876709:  77%|███████▋  | 4988/6470 [1:00:14<15:53,  1.55it/s]
1|4989|Loss: 0.8303623795509338:  77%|███████▋  | 4988/6470 [1:00:15<15:53,  1.55it/s]
1|4989|Loss: 0.8303623795509338:  77%|███████▋  | 4989/6470 [1:00:15<13:56,  1.77it/s]
1|4990|Loss: 0.8369850516319275:  77%|███████▋  | 4989/6470 [1:00:15<13:56,  1.77it/s]
1|4990|Loss: 0.8369850516319275:  77%|███████▋  | 4990/6470 [1:00:16<15:42,  1.57it/s]
1|4991|Loss: 0.825548529624939:  77%|███████▋  | 4990/6470 [1:00:16<15:42,  1.57it/s] 
1|4991|Loss: 0.825548529624939:  77%|███████▋  | 4991/6470 [1:00:16<16:47,  1.47it/s]
1|4992|Loss: 0.7704933285713196:  77%|███████▋  | 4991/6470 [1:00:17<16:47,  1.47it/s]
1|4992|Loss: 0.7704933285713196:  77%|███████▋  | 4992/6470 [1:00:17<18:57,  1.30it/s]
1|4993|Loss: 0.8162083029747009:  77%|███████▋  | 4992/6470 [1:00:18<18:57,  1.30it/s]
1|4993|Loss: 0.8162083029747009:  77%|███████▋  | 4993/6470 [1:00:18<18:07,  1.36it/s]
1|4994|Loss: 0.7916120886802673:  77%|███████▋  | 4993/6470 [1:00:18<18:07,  1.36it/s]
1|4994|Loss: 0.7916120886802673:  77%|███████▋  | 4994/6470 [1:00:18<16:23,  1.50it/s]
1|4995|Loss: 0.7395947575569153:  77%|███████▋  | 4994/6470 [1:00:19<16:23,  1.50it/s]
1|4995|Loss: 0.7395947575569153:  77%|███████▋  | 4995/6470 [1:00:19<16:32,  1.49it/s]
1|4996|Loss: 0.9113247394561768:  77%|███████▋  | 4995/6470 [1:00:19<16:32,  1.49it/s]
1|4996|Loss: 0.9113247394561768:  77%|███████▋  | 4996/6470 [1:00:20<15:58,  1.54it/s]
1|4997|Loss: 0.8244762420654297:  77%|███████▋  | 4996/6470 [1:00:20<15:58,  1.54it/s]
1|4997|Loss: 0.8244762420654297:  77%|███████▋  | 4997/6470 [1:00:20<15:28,  1.59it/s]
1|4998|Loss: 0.8298619389533997:  77%|███████▋  | 4997/6470 [1:00:21<15:28,  1.59it/s]
1|4998|Loss: 0.8298619389533997:  77%|███████▋  | 4998/6470 [1:00:21<15:20,  1.60it/s]
1|4999|Loss: 0.9445053339004517:  77%|███████▋  | 4998/6470 [1:00:21<15:20,  1.60it/s]
1|4999|Loss: 0.9445053339004517:  77%|███████▋  | 4999/6470 [1:00:22<16:34,  1.48it/s]
1|5000|Loss: 0.8514491319656372:  77%|███████▋  | 4999/6470 [1:00:22<16:34,  1.48it/s]
1|5000|Loss: 0.8514491319656372:  77%|███████▋  | 5000/6470 [1:00:22<16:58,  1.44it/s]
1|5001|Loss: 1.2202361822128296:  77%|███████▋  | 5000/6470 [1:00:23<16:58,  1.44it/s]
1|5001|Loss: 1.2202361822128296:  77%|███████▋  | 5001/6470 [1:00:23<19:27,  1.26it/s]
1|5002|Loss: 0.9095072746276855:  77%|███████▋  | 5001/6470 [1:00:24<19:27,  1.26it/s]
1|5002|Loss: 0.9095072746276855:  77%|███████▋  | 5002/6470 [1:00:24<19:38,  1.25it/s]
1|5003|Loss: 0.7912393808364868:  77%|███████▋  | 5002/6470 [1:00:25<19:38,  1.25it/s]
1|5003|Loss: 0.7912393808364868:  77%|███████▋  | 5003/6470 [1:00:25<19:15,  1.27it/s]
1|5004|Loss: 0.7420560717582703:  77%|███████▋  | 5003/6470 [1:00:25<19:15,  1.27it/s]
1|5004|Loss: 0.7420560717582703:  77%|███████▋  | 5004/6470 [1:00:26<18:59,  1.29it/s]
1|5005|Loss: 0.9944480061531067:  77%|███████▋  | 5004/6470 [1:00:26<18:59,  1.29it/s]
1|5005|Loss: 0.9944480061531067:  77%|███████▋  | 5005/6470 [1:00:27<18:44,  1.30it/s]
1|5006|Loss: 0.9419595003128052:  77%|███████▋  | 5005/6470 [1:00:27<18:44,  1.30it/s]
1|5006|Loss: 0.9419595003128052:  77%|███████▋  | 5006/6470 [1:00:27<17:12,  1.42it/s]
1|5007|Loss: 0.7951980233192444:  77%|███████▋  | 5006/6470 [1:00:27<17:12,  1.42it/s]
1|5007|Loss: 0.7951980233192444:  77%|███████▋  | 5007/6470 [1:00:28<17:04,  1.43it/s]
1|5008|Loss: 0.9996726512908936:  77%|███████▋  | 5007/6470 [1:00:28<17:04,  1.43it/s]
1|5008|Loss: 0.9996726512908936:  77%|███████▋  | 5008/6470 [1:00:28<16:07,  1.51it/s]
1|5009|Loss: 0.9260209202766418:  77%|███████▋  | 5008/6470 [1:00:29<16:07,  1.51it/s]
1|5009|Loss: 0.9260209202766418:  77%|███████▋  | 5009/6470 [1:00:29<16:21,  1.49it/s]
1|5010|Loss: 0.9117066860198975:  77%|███████▋  | 5009/6470 [1:00:30<16:21,  1.49it/s]
1|5010|Loss: 0.9117066860198975:  77%|███████▋  | 5010/6470 [1:00:30<17:33,  1.39it/s]
1|5011|Loss: 0.8883948922157288:  77%|███████▋  | 5010/6470 [1:00:30<17:33,  1.39it/s]
1|5011|Loss: 0.8883948922157288:  77%|███████▋  | 5011/6470 [1:00:31<16:48,  1.45it/s]
1|5012|Loss: 0.6599248647689819:  77%|███████▋  | 5011/6470 [1:00:31<16:48,  1.45it/s]
1|5012|Loss: 0.6599248647689819:  77%|███████▋  | 5012/6470 [1:00:31<17:27,  1.39it/s]
1|5013|Loss: 0.7906686663627625:  77%|███████▋  | 5012/6470 [1:00:32<17:27,  1.39it/s]
1|5013|Loss: 0.7906686663627625:  77%|███████▋  | 5013/6470 [1:00:32<16:26,  1.48it/s]
1|5014|Loss: 0.9693047404289246:  77%|███████▋  | 5013/6470 [1:00:32<16:26,  1.48it/s]
1|5014|Loss: 0.9693047404289246:  77%|███████▋  | 5014/6470 [1:00:32<15:48,  1.53it/s]
1|5015|Loss: 0.8963112831115723:  77%|███████▋  | 5014/6470 [1:00:33<15:48,  1.53it/s]
1|5015|Loss: 0.8963112831115723:  78%|███████▊  | 5015/6470 [1:00:33<16:51,  1.44it/s]
1|5016|Loss: 0.7859644293785095:  78%|███████▊  | 5015/6470 [1:00:34<16:51,  1.44it/s]
1|5016|Loss: 0.7859644293785095:  78%|███████▊  | 5016/6470 [1:00:34<17:41,  1.37it/s]
1|5017|Loss: 0.951230525970459:  78%|███████▊  | 5016/6470 [1:00:35<17:41,  1.37it/s] 
1|5017|Loss: 0.951230525970459:  78%|███████▊  | 5017/6470 [1:00:35<18:12,  1.33it/s]
1|5018|Loss: 0.8908355832099915:  78%|███████▊  | 5017/6470 [1:00:35<18:12,  1.33it/s]
1|5018|Loss: 0.8908355832099915:  78%|███████▊  | 5018/6470 [1:00:36<18:49,  1.29it/s]
1|5019|Loss: 1.059507966041565:  78%|███████▊  | 5018/6470 [1:00:36<18:49,  1.29it/s] 
1|5019|Loss: 1.059507966041565:  78%|███████▊  | 5019/6470 [1:00:37<18:58,  1.27it/s]
1|5020|Loss: 0.9581199288368225:  78%|███████▊  | 5019/6470 [1:00:37<18:58,  1.27it/s]
1|5020|Loss: 0.9581199288368225:  78%|███████▊  | 5020/6470 [1:00:37<18:33,  1.30it/s]
1|5021|Loss: 0.9086836576461792:  78%|███████▊  | 5020/6470 [1:00:38<18:33,  1.30it/s]
1|5021|Loss: 0.9086836576461792:  78%|███████▊  | 5021/6470 [1:00:38<18:23,  1.31it/s]
1|5022|Loss: 0.9344772696495056:  78%|███████▊  | 5021/6470 [1:00:38<18:23,  1.31it/s]
1|5022|Loss: 0.9344772696495056:  78%|███████▊  | 5022/6470 [1:00:39<17:42,  1.36it/s]
1|5023|Loss: 1.0940055847167969:  78%|███████▊  | 5022/6470 [1:00:39<17:42,  1.36it/s]
1|5023|Loss: 1.0940055847167969:  78%|███████▊  | 5023/6470 [1:00:39<18:05,  1.33it/s]
1|5024|Loss: 0.8554907441139221:  78%|███████▊  | 5023/6470 [1:00:40<18:05,  1.33it/s]
1|5024|Loss: 0.8554907441139221:  78%|███████▊  | 5024/6470 [1:00:40<18:51,  1.28it/s]
1|5025|Loss: 1.1022021770477295:  78%|███████▊  | 5024/6470 [1:00:41<18:51,  1.28it/s]
1|5025|Loss: 1.1022021770477295:  78%|███████▊  | 5025/6470 [1:00:41<18:08,  1.33it/s]
1|5026|Loss: 0.8405826091766357:  78%|███████▊  | 5025/6470 [1:00:41<18:08,  1.33it/s]
1|5026|Loss: 0.8405826091766357:  78%|███████▊  | 5026/6470 [1:00:42<16:43,  1.44it/s]
1|5027|Loss: 0.8119829893112183:  78%|███████▊  | 5026/6470 [1:00:42<16:43,  1.44it/s]
1|5027|Loss: 0.8119829893112183:  78%|███████▊  | 5027/6470 [1:00:42<15:25,  1.56it/s]
1|5028|Loss: 0.8626819252967834:  78%|███████▊  | 5027/6470 [1:00:42<15:25,  1.56it/s]
1|5028|Loss: 0.8626819252967834:  78%|███████▊  | 5028/6470 [1:00:43<15:47,  1.52it/s]
1|5029|Loss: 0.8004917502403259:  78%|███████▊  | 5028/6470 [1:00:43<15:47,  1.52it/s]
1|5029|Loss: 0.8004917502403259:  78%|███████▊  | 5029/6470 [1:00:44<17:24,  1.38it/s]
1|5030|Loss: 0.8979095220565796:  78%|███████▊  | 5029/6470 [1:00:44<17:24,  1.38it/s]
1|5030|Loss: 0.8979095220565796:  78%|███████▊  | 5030/6470 [1:00:44<16:20,  1.47it/s]
1|5031|Loss: 0.7644968628883362:  78%|███████▊  | 5030/6470 [1:00:45<16:20,  1.47it/s]
1|5031|Loss: 0.7644968628883362:  78%|███████▊  | 5031/6470 [1:00:45<16:44,  1.43it/s]
1|5032|Loss: 0.8386197090148926:  78%|███████▊  | 5031/6470 [1:00:45<16:44,  1.43it/s]
1|5032|Loss: 0.8386197090148926:  78%|███████▊  | 5032/6470 [1:00:46<17:23,  1.38it/s]
1|5033|Loss: 0.8680099844932556:  78%|███████▊  | 5032/6470 [1:00:46<17:23,  1.38it/s]
1|5033|Loss: 0.8680099844932556:  78%|███████▊  | 5033/6470 [1:00:46<17:23,  1.38it/s]
1|5034|Loss: 0.7811366319656372:  78%|███████▊  | 5033/6470 [1:00:47<17:23,  1.38it/s]
1|5034|Loss: 0.7811366319656372:  78%|███████▊  | 5034/6470 [1:00:47<17:12,  1.39it/s]
1|5035|Loss: 0.9214935898780823:  78%|███████▊  | 5034/6470 [1:00:48<17:12,  1.39it/s]
1|5035|Loss: 0.9214935898780823:  78%|███████▊  | 5035/6470 [1:00:48<17:15,  1.39it/s]
1|5036|Loss: 0.7981665730476379:  78%|███████▊  | 5035/6470 [1:00:48<17:15,  1.39it/s]
1|5036|Loss: 0.7981665730476379:  78%|███████▊  | 5036/6470 [1:00:49<16:31,  1.45it/s]
1|5037|Loss: 0.7742934226989746:  78%|███████▊  | 5036/6470 [1:00:49<16:31,  1.45it/s]
1|5037|Loss: 0.7742934226989746:  78%|███████▊  | 5037/6470 [1:00:49<16:58,  1.41it/s]
1|5038|Loss: 0.824384331703186:  78%|███████▊  | 5037/6470 [1:00:50<16:58,  1.41it/s] 
1|5038|Loss: 0.824384331703186:  78%|███████▊  | 5038/6470 [1:00:50<17:48,  1.34it/s]
1|5039|Loss: 0.9142110347747803:  78%|███████▊  | 5038/6470 [1:00:51<17:48,  1.34it/s]
1|5039|Loss: 0.9142110347747803:  78%|███████▊  | 5039/6470 [1:00:51<17:41,  1.35it/s]
1|5040|Loss: 0.7372260689735413:  78%|███████▊  | 5039/6470 [1:00:51<17:41,  1.35it/s]
1|5040|Loss: 0.7372260689735413:  78%|███████▊  | 5040/6470 [1:00:52<18:18,  1.30it/s]
1|5041|Loss: 0.8598986864089966:  78%|███████▊  | 5040/6470 [1:00:52<18:18,  1.30it/s]
1|5041|Loss: 0.8598986864089966:  78%|███████▊  | 5041/6470 [1:00:52<17:36,  1.35it/s]
1|5042|Loss: 0.82756108045578:  78%|███████▊  | 5041/6470 [1:00:53<17:36,  1.35it/s]  
1|5042|Loss: 0.82756108045578:  78%|███████▊  | 5042/6470 [1:00:53<17:22,  1.37it/s]
1|5043|Loss: 0.9559313058853149:  78%|███████▊  | 5042/6470 [1:00:53<17:22,  1.37it/s]
1|5043|Loss: 0.9559313058853149:  78%|███████▊  | 5043/6470 [1:00:54<16:54,  1.41it/s]
1|5044|Loss: 0.8015545010566711:  78%|███████▊  | 5043/6470 [1:00:54<16:54,  1.41it/s]
1|5044|Loss: 0.8015545010566711:  78%|███████▊  | 5044/6470 [1:00:54<16:54,  1.41it/s]
1|5045|Loss: 0.9621347784996033:  78%|███████▊  | 5044/6470 [1:00:55<16:54,  1.41it/s]
1|5045|Loss: 0.9621347784996033:  78%|███████▊  | 5045/6470 [1:00:55<17:02,  1.39it/s]
1|5046|Loss: 0.9059094786643982:  78%|███████▊  | 5045/6470 [1:00:56<17:02,  1.39it/s]
1|5046|Loss: 0.9059094786643982:  78%|███████▊  | 5046/6470 [1:00:56<17:40,  1.34it/s]
1|5047|Loss: 0.7887072563171387:  78%|███████▊  | 5046/6470 [1:00:56<17:40,  1.34it/s]
1|5047|Loss: 0.7887072563171387:  78%|███████▊  | 5047/6470 [1:00:57<19:12,  1.24it/s]
1|5048|Loss: 0.7648047208786011:  78%|███████▊  | 5047/6470 [1:00:57<19:12,  1.24it/s]
1|5048|Loss: 0.7648047208786011:  78%|███████▊  | 5048/6470 [1:00:58<19:47,  1.20it/s]
1|5049|Loss: 0.9134739637374878:  78%|███████▊  | 5048/6470 [1:00:58<19:47,  1.20it/s]
1|5049|Loss: 0.9134739637374878:  78%|███████▊  | 5049/6470 [1:00:59<19:15,  1.23it/s]
1|5050|Loss: 0.7735822200775146:  78%|███████▊  | 5049/6470 [1:00:59<19:15,  1.23it/s]
1|5050|Loss: 0.7735822200775146:  78%|███████▊  | 5050/6470 [1:00:59<18:07,  1.31it/s]
1|5051|Loss: 0.9635264277458191:  78%|███████▊  | 5050/6470 [1:01:00<18:07,  1.31it/s]
1|5051|Loss: 0.9635264277458191:  78%|███████▊  | 5051/6470 [1:01:00<17:05,  1.38it/s]
1|5052|Loss: 0.8500451445579529:  78%|███████▊  | 5051/6470 [1:01:00<17:05,  1.38it/s]
1|5052|Loss: 0.8500451445579529:  78%|███████▊  | 5052/6470 [1:01:01<17:31,  1.35it/s]
1|5053|Loss: 0.9127874970436096:  78%|███████▊  | 5052/6470 [1:01:01<17:31,  1.35it/s]
1|5053|Loss: 0.9127874970436096:  78%|███████▊  | 5053/6470 [1:01:01<17:43,  1.33it/s]
1|5054|Loss: 0.8158663511276245:  78%|███████▊  | 5053/6470 [1:01:02<17:43,  1.33it/s]
1|5054|Loss: 0.8158663511276245:  78%|███████▊  | 5054/6470 [1:01:02<18:20,  1.29it/s]
1|5055|Loss: 1.048056721687317:  78%|███████▊  | 5054/6470 [1:01:03<18:20,  1.29it/s] 
1|5055|Loss: 1.048056721687317:  78%|███████▊  | 5055/6470 [1:01:03<16:58,  1.39it/s]
1|5056|Loss: 0.8915454149246216:  78%|███████▊  | 5055/6470 [1:01:03<16:58,  1.39it/s]
1|5056|Loss: 0.8915454149246216:  78%|███████▊  | 5056/6470 [1:01:04<17:21,  1.36it/s]
1|5057|Loss: 0.8295878171920776:  78%|███████▊  | 5056/6470 [1:01:04<17:21,  1.36it/s]
1|5057|Loss: 0.8295878171920776:  78%|███████▊  | 5057/6470 [1:01:04<16:55,  1.39it/s]
1|5058|Loss: 0.8005853891372681:  78%|███████▊  | 5057/6470 [1:01:05<16:55,  1.39it/s]
1|5058|Loss: 0.8005853891372681:  78%|███████▊  | 5058/6470 [1:01:05<17:26,  1.35it/s]
1|5059|Loss: 0.991016149520874:  78%|███████▊  | 5058/6470 [1:01:06<17:26,  1.35it/s] 
1|5059|Loss: 0.991016149520874:  78%|███████▊  | 5059/6470 [1:01:06<17:50,  1.32it/s]
1|5060|Loss: 0.8444179892539978:  78%|███████▊  | 5059/6470 [1:01:06<17:50,  1.32it/s]
1|5060|Loss: 0.8444179892539978:  78%|███████▊  | 5060/6470 [1:01:07<17:51,  1.32it/s]
1|5061|Loss: 1.02601957321167:  78%|███████▊  | 5060/6470 [1:01:07<17:51,  1.32it/s]  
1|5061|Loss: 1.02601957321167:  78%|███████▊  | 5061/6470 [1:01:07<17:03,  1.38it/s]
1|5062|Loss: 0.8036522269248962:  78%|███████▊  | 5061/6470 [1:01:08<17:03,  1.38it/s]
1|5062|Loss: 0.8036522269248962:  78%|███████▊  | 5062/6470 [1:01:08<16:45,  1.40it/s]
1|5063|Loss: 0.7514607310295105:  78%|███████▊  | 5062/6470 [1:01:08<16:45,  1.40it/s]
1|5063|Loss: 0.7514607310295105:  78%|███████▊  | 5063/6470 [1:01:09<16:38,  1.41it/s]
1|5064|Loss: 0.8643009066581726:  78%|███████▊  | 5063/6470 [1:01:09<16:38,  1.41it/s]
1|5064|Loss: 0.8643009066581726:  78%|███████▊  | 5064/6470 [1:01:09<17:17,  1.36it/s]
1|5065|Loss: 0.9341582655906677:  78%|███████▊  | 5064/6470 [1:01:10<17:17,  1.36it/s]
1|5065|Loss: 0.9341582655906677:  78%|███████▊  | 5065/6470 [1:01:10<16:21,  1.43it/s]
1|5066|Loss: 0.9236215353012085:  78%|███████▊  | 5065/6470 [1:01:10<16:21,  1.43it/s]
1|5066|Loss: 0.9236215353012085:  78%|███████▊  | 5066/6470 [1:01:11<15:42,  1.49it/s]
1|5067|Loss: 0.9491596817970276:  78%|███████▊  | 5066/6470 [1:01:11<15:42,  1.49it/s]
1|5067|Loss: 0.9491596817970276:  78%|███████▊  | 5067/6470 [1:01:11<15:57,  1.47it/s]
1|5068|Loss: 0.7388758063316345:  78%|███████▊  | 5067/6470 [1:01:12<15:57,  1.47it/s]
1|5068|Loss: 0.7388758063316345:  78%|███████▊  | 5068/6470 [1:01:12<16:47,  1.39it/s]
1|5069|Loss: 0.866234540939331:  78%|███████▊  | 5068/6470 [1:01:13<16:47,  1.39it/s] 
1|5069|Loss: 0.866234540939331:  78%|███████▊  | 5069/6470 [1:01:13<17:20,  1.35it/s]
1|5070|Loss: 0.7675607800483704:  78%|███████▊  | 5069/6470 [1:01:13<17:20,  1.35it/s]
1|5070|Loss: 0.7675607800483704:  78%|███████▊  | 5070/6470 [1:01:14<16:47,  1.39it/s]
1|5071|Loss: 0.7405942678451538:  78%|███████▊  | 5070/6470 [1:01:14<16:47,  1.39it/s]
1|5071|Loss: 0.7405942678451538:  78%|███████▊  | 5071/6470 [1:01:15<17:33,  1.33it/s]
1|5072|Loss: 0.8866127133369446:  78%|███████▊  | 5071/6470 [1:01:15<17:33,  1.33it/s]
1|5072|Loss: 0.8866127133369446:  78%|███████▊  | 5072/6470 [1:01:15<17:04,  1.36it/s]
1|5073|Loss: 0.8873188495635986:  78%|███████▊  | 5072/6470 [1:01:16<17:04,  1.36it/s]
1|5073|Loss: 0.8873188495635986:  78%|███████▊  | 5073/6470 [1:01:16<16:55,  1.38it/s]
1|5074|Loss: 0.8499321341514587:  78%|███████▊  | 5073/6470 [1:01:16<16:55,  1.38it/s]
1|5074|Loss: 0.8499321341514587:  78%|███████▊  | 5074/6470 [1:01:17<16:09,  1.44it/s]
1|5075|Loss: 0.919823944568634:  78%|███████▊  | 5074/6470 [1:01:17<16:09,  1.44it/s] 
1|5075|Loss: 0.919823944568634:  78%|███████▊  | 5075/6470 [1:01:17<14:15,  1.63it/s]
1|5076|Loss: 1.06427800655365:  78%|███████▊  | 5075/6470 [1:01:17<14:15,  1.63it/s] 
1|5076|Loss: 1.06427800655365:  78%|███████▊  | 5076/6470 [1:01:18<15:12,  1.53it/s]
1|5077|Loss: 0.6965404748916626:  78%|███████▊  | 5076/6470 [1:01:18<15:12,  1.53it/s]
1|5077|Loss: 0.6965404748916626:  78%|███████▊  | 5077/6470 [1:01:18<14:42,  1.58it/s]
1|5078|Loss: 0.7969900369644165:  78%|███████▊  | 5077/6470 [1:01:19<14:42,  1.58it/s]
1|5078|Loss: 0.7969900369644165:  78%|███████▊  | 5078/6470 [1:01:19<14:31,  1.60it/s]
1|5079|Loss: 0.950420081615448:  78%|███████▊  | 5078/6470 [1:01:19<14:31,  1.60it/s] 
1|5079|Loss: 0.950420081615448:  79%|███████▊  | 5079/6470 [1:01:20<15:13,  1.52it/s]
1|5080|Loss: 0.9894055724143982:  79%|███████▊  | 5079/6470 [1:01:20<15:13,  1.52it/s]
1|5080|Loss: 0.9894055724143982:  79%|███████▊  | 5080/6470 [1:01:20<16:29,  1.41it/s]
1|5081|Loss: 0.8178212642669678:  79%|███████▊  | 5080/6470 [1:01:21<16:29,  1.41it/s]
1|5081|Loss: 0.8178212642669678:  79%|███████▊  | 5081/6470 [1:01:21<17:29,  1.32it/s]
1|5082|Loss: 0.892636775970459:  79%|███████▊  | 5081/6470 [1:01:22<17:29,  1.32it/s] 
1|5082|Loss: 0.892636775970459:  79%|███████▊  | 5082/6470 [1:01:22<16:22,  1.41it/s]
1|5083|Loss: 0.7950146794319153:  79%|███████▊  | 5082/6470 [1:01:22<16:22,  1.41it/s]
1|5083|Loss: 0.7950146794319153:  79%|███████▊  | 5083/6470 [1:01:23<16:03,  1.44it/s]
1|5084|Loss: 0.9098289608955383:  79%|███████▊  | 5083/6470 [1:01:23<16:03,  1.44it/s]
1|5084|Loss: 0.9098289608955383:  79%|███████▊  | 5084/6470 [1:01:23<16:16,  1.42it/s]
1|5085|Loss: 0.8622385859489441:  79%|███████▊  | 5084/6470 [1:01:24<16:16,  1.42it/s]
1|5085|Loss: 0.8622385859489441:  79%|███████▊  | 5085/6470 [1:01:24<14:54,  1.55it/s]
1|5086|Loss: 0.8050109148025513:  79%|███████▊  | 5085/6470 [1:01:24<14:54,  1.55it/s]
1|5086|Loss: 0.8050109148025513:  79%|███████▊  | 5086/6470 [1:01:24<14:32,  1.59it/s]
1|5087|Loss: 0.8869466185569763:  79%|███████▊  | 5086/6470 [1:01:25<14:32,  1.59it/s]
1|5087|Loss: 0.8869466185569763:  79%|███████▊  | 5087/6470 [1:01:25<15:07,  1.52it/s]
1|5088|Loss: 0.9094829559326172:  79%|███████▊  | 5087/6470 [1:01:25<15:07,  1.52it/s]
1|5088|Loss: 0.9094829559326172:  79%|███████▊  | 5088/6470 [1:01:26<13:42,  1.68it/s]
1|5089|Loss: 0.9001390933990479:  79%|███████▊  | 5088/6470 [1:01:26<13:42,  1.68it/s]
1|5089|Loss: 0.9001390933990479:  79%|███████▊  | 5089/6470 [1:01:26<13:42,  1.68it/s]
1|5090|Loss: 1.0026187896728516:  79%|███████▊  | 5089/6470 [1:01:27<13:42,  1.68it/s]
1|5090|Loss: 1.0026187896728516:  79%|███████▊  | 5090/6470 [1:01:27<13:55,  1.65it/s]
1|5091|Loss: 0.9248251914978027:  79%|███████▊  | 5090/6470 [1:01:27<13:55,  1.65it/s]
1|5091|Loss: 0.9248251914978027:  79%|███████▊  | 5091/6470 [1:01:27<13:25,  1.71it/s]
1|5092|Loss: 0.9483540654182434:  79%|███████▊  | 5091/6470 [1:01:28<13:25,  1.71it/s]
1|5092|Loss: 0.9483540654182434:  79%|███████▊  | 5092/6470 [1:01:28<14:06,  1.63it/s]
1|5093|Loss: 0.8090246915817261:  79%|███████▊  | 5092/6470 [1:01:28<14:06,  1.63it/s]
1|5093|Loss: 0.8090246915817261:  79%|███████▊  | 5093/6470 [1:01:29<14:25,  1.59it/s]
1|5094|Loss: 0.9527907371520996:  79%|███████▊  | 5093/6470 [1:01:29<14:25,  1.59it/s]
1|5094|Loss: 0.9527907371520996:  79%|███████▊  | 5094/6470 [1:01:29<14:07,  1.62it/s]
1|5095|Loss: 0.9352843761444092:  79%|███████▊  | 5094/6470 [1:01:30<14:07,  1.62it/s]
1|5095|Loss: 0.9352843761444092:  79%|███████▊  | 5095/6470 [1:01:30<15:06,  1.52it/s]
1|5096|Loss: 1.0106641054153442:  79%|███████▊  | 5095/6470 [1:01:31<15:06,  1.52it/s]
1|5096|Loss: 1.0106641054153442:  79%|███████▉  | 5096/6470 [1:01:31<16:23,  1.40it/s]
1|5097|Loss: 0.9515552520751953:  79%|███████▉  | 5096/6470 [1:01:31<16:23,  1.40it/s]
1|5097|Loss: 0.9515552520751953:  79%|███████▉  | 5097/6470 [1:01:32<16:51,  1.36it/s]
1|5098|Loss: 0.7925341725349426:  79%|███████▉  | 5097/6470 [1:01:32<16:51,  1.36it/s]
1|5098|Loss: 0.7925341725349426:  79%|███████▉  | 5098/6470 [1:01:32<17:10,  1.33it/s]
1|5099|Loss: 0.8388131260871887:  79%|███████▉  | 5098/6470 [1:01:33<17:10,  1.33it/s]
1|5099|Loss: 0.8388131260871887:  79%|███████▉  | 5099/6470 [1:01:33<16:56,  1.35it/s]
1|5100|Loss: 0.8500706553459167:  79%|███████▉  | 5099/6470 [1:01:34<16:56,  1.35it/s]
1|5100|Loss: 0.8500706553459167:  79%|███████▉  | 5100/6470 [1:01:34<16:50,  1.36it/s]
1|5101|Loss: 0.7936019897460938:  79%|███████▉  | 5100/6470 [1:01:34<16:50,  1.36it/s]
1|5101|Loss: 0.7936019897460938:  79%|███████▉  | 5101/6470 [1:01:35<16:12,  1.41it/s]
1|5102|Loss: 0.8222094774246216:  79%|███████▉  | 5101/6470 [1:01:35<16:12,  1.41it/s]
1|5102|Loss: 0.8222094774246216:  79%|███████▉  | 5102/6470 [1:01:35<15:19,  1.49it/s]
1|5103|Loss: 0.7927947640419006:  79%|███████▉  | 5102/6470 [1:01:35<15:19,  1.49it/s]
1|5103|Loss: 0.7927947640419006:  79%|███████▉  | 5103/6470 [1:01:36<14:39,  1.55it/s]
1|5104|Loss: 0.7173752188682556:  79%|███████▉  | 5103/6470 [1:01:36<14:39,  1.55it/s]
1|5104|Loss: 0.7173752188682556:  79%|███████▉  | 5104/6470 [1:01:36<14:11,  1.60it/s]
1|5105|Loss: 0.7474617958068848:  79%|███████▉  | 5104/6470 [1:01:37<14:11,  1.60it/s]
1|5105|Loss: 0.7474617958068848:  79%|███████▉  | 5105/6470 [1:01:37<15:43,  1.45it/s]
1|5106|Loss: 0.7164650559425354:  79%|███████▉  | 5105/6470 [1:01:38<15:43,  1.45it/s]
1|5106|Loss: 0.7164650559425354:  79%|███████▉  | 5106/6470 [1:01:38<16:14,  1.40it/s]
1|5107|Loss: 0.863009512424469:  79%|███████▉  | 5106/6470 [1:01:38<16:14,  1.40it/s] 
1|5107|Loss: 0.863009512424469:  79%|███████▉  | 5107/6470 [1:01:39<15:45,  1.44it/s]
1|5108|Loss: 0.9486995339393616:  79%|███████▉  | 5107/6470 [1:01:39<15:45,  1.44it/s]
1|5108|Loss: 0.9486995339393616:  79%|███████▉  | 5108/6470 [1:01:39<16:31,  1.37it/s]
1|5109|Loss: 0.770637035369873:  79%|███████▉  | 5108/6470 [1:01:40<16:31,  1.37it/s] 
1|5109|Loss: 0.770637035369873:  79%|███████▉  | 5109/6470 [1:01:40<16:34,  1.37it/s]
1|5110|Loss: 0.8220555186271667:  79%|███████▉  | 5109/6470 [1:01:40<16:34,  1.37it/s]
1|5110|Loss: 0.8220555186271667:  79%|███████▉  | 5110/6470 [1:01:41<16:32,  1.37it/s]
1|5111|Loss: 0.9438043236732483:  79%|███████▉  | 5110/6470 [1:01:41<16:32,  1.37it/s]
1|5111|Loss: 0.9438043236732483:  79%|███████▉  | 5111/6470 [1:01:41<15:55,  1.42it/s]
1|5112|Loss: 0.6936067342758179:  79%|███████▉  | 5111/6470 [1:01:42<15:55,  1.42it/s]
1|5112|Loss: 0.6936067342758179:  79%|███████▉  | 5112/6470 [1:01:42<17:18,  1.31it/s]
1|5113|Loss: 0.8856189250946045:  79%|███████▉  | 5112/6470 [1:01:43<17:18,  1.31it/s]
1|5113|Loss: 0.8856189250946045:  79%|███████▉  | 5113/6470 [1:01:43<17:20,  1.30it/s]
1|5114|Loss: 0.976163387298584:  79%|███████▉  | 5113/6470 [1:01:44<17:20,  1.30it/s] 
1|5114|Loss: 0.976163387298584:  79%|███████▉  | 5114/6470 [1:01:44<16:27,  1.37it/s]
1|5115|Loss: 0.8539575934410095:  79%|███████▉  | 5114/6470 [1:01:44<16:27,  1.37it/s]
1|5115|Loss: 0.8539575934410095:  79%|███████▉  | 5115/6470 [1:01:44<16:10,  1.40it/s]
1|5116|Loss: 0.9185686707496643:  79%|███████▉  | 5115/6470 [1:01:45<16:10,  1.40it/s]
1|5116|Loss: 0.9185686707496643:  79%|███████▉  | 5116/6470 [1:01:45<16:16,  1.39it/s]
1|5117|Loss: 0.7930591106414795:  79%|███████▉  | 5116/6470 [1:01:46<16:16,  1.39it/s]
1|5117|Loss: 0.7930591106414795:  79%|███████▉  | 5117/6470 [1:01:46<15:52,  1.42it/s]
1|5118|Loss: 1.0509066581726074:  79%|███████▉  | 5117/6470 [1:01:46<15:52,  1.42it/s]
1|5118|Loss: 1.0509066581726074:  79%|███████▉  | 5118/6470 [1:01:47<15:26,  1.46it/s]
1|5119|Loss: 0.7578452825546265:  79%|███████▉  | 5118/6470 [1:01:47<15:26,  1.46it/s]
1|5119|Loss: 0.7578452825546265:  79%|███████▉  | 5119/6470 [1:01:47<15:03,  1.50it/s]
1|5120|Loss: 0.8333783149719238:  79%|███████▉  | 5119/6470 [1:01:48<15:03,  1.50it/s]
1|5120|Loss: 0.8333783149719238:  79%|███████▉  | 5120/6470 [1:01:48<15:11,  1.48it/s]
1|5121|Loss: 0.9861017465591431:  79%|███████▉  | 5120/6470 [1:01:48<15:11,  1.48it/s]
1|5121|Loss: 0.9861017465591431:  79%|███████▉  | 5121/6470 [1:01:49<15:37,  1.44it/s]
1|5122|Loss: 0.9121366143226624:  79%|███████▉  | 5121/6470 [1:01:49<15:37,  1.44it/s]
1|5122|Loss: 0.9121366143226624:  79%|███████▉  | 5122/6470 [1:01:49<15:27,  1.45it/s]
1|5123|Loss: 0.8575692176818848:  79%|███████▉  | 5122/6470 [1:01:50<15:27,  1.45it/s]
1|5123|Loss: 0.8575692176818848:  79%|███████▉  | 5123/6470 [1:01:50<16:22,  1.37it/s]
1|5124|Loss: 0.8362845778465271:  79%|███████▉  | 5123/6470 [1:01:51<16:22,  1.37it/s]
1|5124|Loss: 0.8362845778465271:  79%|███████▉  | 5124/6470 [1:01:51<16:36,  1.35it/s]
1|5125|Loss: 0.7479280233383179:  79%|███████▉  | 5124/6470 [1:01:51<16:36,  1.35it/s]
1|5125|Loss: 0.7479280233383179:  79%|███████▉  | 5125/6470 [1:01:52<16:57,  1.32it/s]
1|5126|Loss: 0.8855450749397278:  79%|███████▉  | 5125/6470 [1:01:52<16:57,  1.32it/s]
1|5126|Loss: 0.8855450749397278:  79%|███████▉  | 5126/6470 [1:01:52<15:22,  1.46it/s]
1|5127|Loss: 0.8694313168525696:  79%|███████▉  | 5126/6470 [1:01:52<15:22,  1.46it/s]
1|5127|Loss: 0.8694313168525696:  79%|███████▉  | 5127/6470 [1:01:53<13:47,  1.62it/s]
1|5128|Loss: 0.9674083590507507:  79%|███████▉  | 5127/6470 [1:01:53<13:47,  1.62it/s]
1|5128|Loss: 0.9674083590507507:  79%|███████▉  | 5128/6470 [1:01:53<13:38,  1.64it/s]
1|5129|Loss: 0.8979027271270752:  79%|███████▉  | 5128/6470 [1:01:54<13:38,  1.64it/s]
1|5129|Loss: 0.8979027271270752:  79%|███████▉  | 5129/6470 [1:01:54<14:17,  1.56it/s]
1|5130|Loss: 0.7699915170669556:  79%|███████▉  | 5129/6470 [1:01:54<14:17,  1.56it/s]
1|5130|Loss: 0.7699915170669556:  79%|███████▉  | 5130/6470 [1:01:55<14:58,  1.49it/s]
1|5131|Loss: 0.9348136186599731:  79%|███████▉  | 5130/6470 [1:01:55<14:58,  1.49it/s]
1|5131|Loss: 0.9348136186599731:  79%|███████▉  | 5131/6470 [1:01:55<15:55,  1.40it/s]
1|5132|Loss: 0.812886655330658:  79%|███████▉  | 5131/6470 [1:01:56<15:55,  1.40it/s] 
1|5132|Loss: 0.812886655330658:  79%|███████▉  | 5132/6470 [1:01:56<14:40,  1.52it/s]
1|5133|Loss: 0.8638488054275513:  79%|███████▉  | 5132/6470 [1:01:56<14:40,  1.52it/s]
1|5133|Loss: 0.8638488054275513:  79%|███████▉  | 5133/6470 [1:01:57<14:51,  1.50it/s]
1|5134|Loss: 0.7832931280136108:  79%|███████▉  | 5133/6470 [1:01:57<14:51,  1.50it/s]
1|5134|Loss: 0.7832931280136108:  79%|███████▉  | 5134/6470 [1:01:58<16:02,  1.39it/s]
1|5135|Loss: 0.7647527456283569:  79%|███████▉  | 5134/6470 [1:01:58<16:02,  1.39it/s]
1|5135|Loss: 0.7647527456283569:  79%|███████▉  | 5135/6470 [1:01:58<16:42,  1.33it/s]
1|5136|Loss: 0.8899543285369873:  79%|███████▉  | 5135/6470 [1:01:59<16:42,  1.33it/s]
1|5136|Loss: 0.8899543285369873:  79%|███████▉  | 5136/6470 [1:01:59<15:42,  1.42it/s]
1|5137|Loss: 0.8118466734886169:  79%|███████▉  | 5136/6470 [1:01:59<15:42,  1.42it/s]
1|5137|Loss: 0.8118466734886169:  79%|███████▉  | 5137/6470 [1:02:00<15:37,  1.42it/s]
1|5138|Loss: 0.993265688419342:  79%|███████▉  | 5137/6470 [1:02:00<15:37,  1.42it/s] 
1|5138|Loss: 0.993265688419342:  79%|███████▉  | 5138/6470 [1:02:00<15:54,  1.40it/s]
1|5139|Loss: 0.9635353684425354:  79%|███████▉  | 5138/6470 [1:02:01<15:54,  1.40it/s]
1|5139|Loss: 0.9635353684425354:  79%|███████▉  | 5139/6470 [1:02:01<15:55,  1.39it/s]
1|5140|Loss: 0.8149495720863342:  79%|███████▉  | 5139/6470 [1:02:02<15:55,  1.39it/s]
1|5140|Loss: 0.8149495720863342:  79%|███████▉  | 5140/6470 [1:02:02<15:57,  1.39it/s]
1|5141|Loss: 0.9108932614326477:  79%|███████▉  | 5140/6470 [1:02:02<15:57,  1.39it/s]
1|5141|Loss: 0.9108932614326477:  79%|███████▉  | 5141/6470 [1:02:03<15:51,  1.40it/s]
1|5142|Loss: 0.7885514497756958:  79%|███████▉  | 5141/6470 [1:02:03<15:51,  1.40it/s]
1|5142|Loss: 0.7885514497756958:  79%|███████▉  | 5142/6470 [1:02:03<15:27,  1.43it/s]
1|5143|Loss: 0.7207661271095276:  79%|███████▉  | 5142/6470 [1:02:04<15:27,  1.43it/s]
1|5143|Loss: 0.7207661271095276:  79%|███████▉  | 5143/6470 [1:02:04<14:56,  1.48it/s]
1|5144|Loss: 0.8985890746116638:  79%|███████▉  | 5143/6470 [1:02:04<14:56,  1.48it/s]
1|5144|Loss: 0.8985890746116638:  80%|███████▉  | 5144/6470 [1:02:04<14:31,  1.52it/s]
1|5145|Loss: 0.7914322018623352:  80%|███████▉  | 5144/6470 [1:02:05<14:31,  1.52it/s]
1|5145|Loss: 0.7914322018623352:  80%|███████▉  | 5145/6470 [1:02:05<14:49,  1.49it/s]
1|5146|Loss: 0.7600895166397095:  80%|███████▉  | 5145/6470 [1:02:06<14:49,  1.49it/s]
1|5146|Loss: 0.7600895166397095:  80%|███████▉  | 5146/6470 [1:02:06<16:27,  1.34it/s]
1|5147|Loss: 0.887888491153717:  80%|███████▉  | 5146/6470 [1:02:06<16:27,  1.34it/s] 
1|5147|Loss: 0.887888491153717:  80%|███████▉  | 5147/6470 [1:02:07<15:17,  1.44it/s]
1|5148|Loss: 0.9516593217849731:  80%|███████▉  | 5147/6470 [1:02:07<15:17,  1.44it/s]
1|5148|Loss: 0.9516593217849731:  80%|███████▉  | 5148/6470 [1:02:07<14:17,  1.54it/s]
1|5149|Loss: 0.6816520094871521:  80%|███████▉  | 5148/6470 [1:02:08<14:17,  1.54it/s]
1|5149|Loss: 0.6816520094871521:  80%|███████▉  | 5149/6470 [1:02:08<14:36,  1.51it/s]
1|5150|Loss: 0.8176200985908508:  80%|███████▉  | 5149/6470 [1:02:08<14:36,  1.51it/s]
1|5150|Loss: 0.8176200985908508:  80%|███████▉  | 5150/6470 [1:02:08<13:06,  1.68it/s]
1|5151|Loss: 0.9177833795547485:  80%|███████▉  | 5150/6470 [1:02:09<13:06,  1.68it/s]
1|5151|Loss: 0.9177833795547485:  80%|███████▉  | 5151/6470 [1:02:09<14:07,  1.56it/s]
1|5152|Loss: 0.7408605217933655:  80%|███████▉  | 5151/6470 [1:02:10<14:07,  1.56it/s]
1|5152|Loss: 0.7408605217933655:  80%|███████▉  | 5152/6470 [1:02:10<15:26,  1.42it/s]
1|5153|Loss: 0.8442474007606506:  80%|███████▉  | 5152/6470 [1:02:10<15:26,  1.42it/s]
1|5153|Loss: 0.8442474007606506:  80%|███████▉  | 5153/6470 [1:02:11<15:45,  1.39it/s]
1|5154|Loss: 0.8828058838844299:  80%|███████▉  | 5153/6470 [1:02:11<15:45,  1.39it/s]
1|5154|Loss: 0.8828058838844299:  80%|███████▉  | 5154/6470 [1:02:11<16:20,  1.34it/s]
1|5155|Loss: 0.7580780982971191:  80%|███████▉  | 5154/6470 [1:02:12<16:20,  1.34it/s]
1|5155|Loss: 0.7580780982971191:  80%|███████▉  | 5155/6470 [1:02:12<15:19,  1.43it/s]
1|5156|Loss: 0.9037522673606873:  80%|███████▉  | 5155/6470 [1:02:12<15:19,  1.43it/s]
1|5156|Loss: 0.9037522673606873:  80%|███████▉  | 5156/6470 [1:02:13<15:19,  1.43it/s]
1|5157|Loss: 1.0090796947479248:  80%|███████▉  | 5156/6470 [1:02:13<15:19,  1.43it/s]
1|5157|Loss: 1.0090796947479248:  80%|███████▉  | 5157/6470 [1:02:13<15:29,  1.41it/s]
1|5158|Loss: 0.9112724661827087:  80%|███████▉  | 5157/6470 [1:02:14<15:29,  1.41it/s]
1|5158|Loss: 0.9112724661827087:  80%|███████▉  | 5158/6470 [1:02:14<15:10,  1.44it/s]
1|5159|Loss: 0.8494208455085754:  80%|███████▉  | 5158/6470 [1:02:15<15:10,  1.44it/s]
1|5159|Loss: 0.8494208455085754:  80%|███████▉  | 5159/6470 [1:02:15<14:49,  1.47it/s]
1|5160|Loss: 0.9193517565727234:  80%|███████▉  | 5159/6470 [1:02:15<14:49,  1.47it/s]
1|5160|Loss: 0.9193517565727234:  80%|███████▉  | 5160/6470 [1:02:16<15:03,  1.45it/s]
1|5161|Loss: 0.95644211769104:  80%|███████▉  | 5160/6470 [1:02:16<15:03,  1.45it/s]  
1|5161|Loss: 0.95644211769104:  80%|███████▉  | 5161/6470 [1:02:16<14:52,  1.47it/s]
1|5162|Loss: 0.8313879370689392:  80%|███████▉  | 5161/6470 [1:02:17<14:52,  1.47it/s]
1|5162|Loss: 0.8313879370689392:  80%|███████▉  | 5162/6470 [1:02:17<15:05,  1.45it/s]
1|5163|Loss: 0.9677460193634033:  80%|███████▉  | 5162/6470 [1:02:17<15:05,  1.45it/s]
1|5163|Loss: 0.9677460193634033:  80%|███████▉  | 5163/6470 [1:02:18<15:58,  1.36it/s]
1|5164|Loss: 0.771783709526062:  80%|███████▉  | 5163/6470 [1:02:18<15:58,  1.36it/s] 
1|5164|Loss: 0.771783709526062:  80%|███████▉  | 5164/6470 [1:02:18<15:41,  1.39it/s]
1|5165|Loss: 0.9154845476150513:  80%|███████▉  | 5164/6470 [1:02:19<15:41,  1.39it/s]
1|5165|Loss: 0.9154845476150513:  80%|███████▉  | 5165/6470 [1:02:19<15:28,  1.41it/s]
1|5166|Loss: 0.9367609620094299:  80%|███████▉  | 5165/6470 [1:02:20<15:28,  1.41it/s]
1|5166|Loss: 0.9367609620094299:  80%|███████▉  | 5166/6470 [1:02:20<16:09,  1.34it/s]
1|5167|Loss: 0.9576852917671204:  80%|███████▉  | 5166/6470 [1:02:20<16:09,  1.34it/s]
1|5167|Loss: 0.9576852917671204:  80%|███████▉  | 5167/6470 [1:02:21<15:49,  1.37it/s]
1|5168|Loss: 0.7210361361503601:  80%|███████▉  | 5167/6470 [1:02:21<15:49,  1.37it/s]
1|5168|Loss: 0.7210361361503601:  80%|███████▉  | 5168/6470 [1:02:21<16:12,  1.34it/s]
1|5169|Loss: 0.7603597044944763:  80%|███████▉  | 5168/6470 [1:02:22<16:12,  1.34it/s]
1|5169|Loss: 0.7603597044944763:  80%|███████▉  | 5169/6470 [1:02:22<16:14,  1.34it/s]
1|5170|Loss: 0.8381878733634949:  80%|███████▉  | 5169/6470 [1:02:23<16:14,  1.34it/s]
1|5170|Loss: 0.8381878733634949:  80%|███████▉  | 5170/6470 [1:02:23<16:00,  1.35it/s]
1|5171|Loss: 0.8595187067985535:  80%|███████▉  | 5170/6470 [1:02:23<16:00,  1.35it/s]
1|5171|Loss: 0.8595187067985535:  80%|███████▉  | 5171/6470 [1:02:23<14:36,  1.48it/s]
1|5172|Loss: 0.8412841558456421:  80%|███████▉  | 5171/6470 [1:02:24<14:36,  1.48it/s]
1|5172|Loss: 0.8412841558456421:  80%|███████▉  | 5172/6470 [1:02:24<13:50,  1.56it/s]
1|5173|Loss: 0.8215039372444153:  80%|███████▉  | 5172/6470 [1:02:24<13:50,  1.56it/s]
1|5173|Loss: 0.8215039372444153:  80%|███████▉  | 5173/6470 [1:02:25<13:39,  1.58it/s]
1|5174|Loss: 0.9348657727241516:  80%|███████▉  | 5173/6470 [1:02:25<13:39,  1.58it/s]
1|5174|Loss: 0.9348657727241516:  80%|███████▉  | 5174/6470 [1:02:25<14:46,  1.46it/s]
1|5175|Loss: 0.871038019657135:  80%|███████▉  | 5174/6470 [1:02:26<14:46,  1.46it/s] 
1|5175|Loss: 0.871038019657135:  80%|███████▉  | 5175/6470 [1:02:26<15:55,  1.36it/s]
1|5176|Loss: 0.88741534948349:  80%|███████▉  | 5175/6470 [1:02:27<15:55,  1.36it/s] 
1|5176|Loss: 0.88741534948349:  80%|████████  | 5176/6470 [1:02:27<15:57,  1.35it/s]
1|5177|Loss: 0.853081464767456:  80%|████████  | 5176/6470 [1:02:27<15:57,  1.35it/s]
1|5177|Loss: 0.853081464767456:  80%|████████  | 5177/6470 [1:02:28<15:41,  1.37it/s]
1|5178|Loss: 0.9834412932395935:  80%|████████  | 5177/6470 [1:02:28<15:41,  1.37it/s]
1|5178|Loss: 0.9834412932395935:  80%|████████  | 5178/6470 [1:02:28<15:45,  1.37it/s]
1|5179|Loss: 0.8886076211929321:  80%|████████  | 5178/6470 [1:02:29<15:45,  1.37it/s]
1|5179|Loss: 0.8886076211929321:  80%|████████  | 5179/6470 [1:02:29<13:58,  1.54it/s]
1|5180|Loss: 0.7987744212150574:  80%|████████  | 5179/6470 [1:02:29<13:58,  1.54it/s]
1|5180|Loss: 0.7987744212150574:  80%|████████  | 5180/6470 [1:02:30<14:00,  1.53it/s]
1|5181|Loss: 0.7472088932991028:  80%|████████  | 5180/6470 [1:02:30<14:00,  1.53it/s]
1|5181|Loss: 0.7472088932991028:  80%|████████  | 5181/6470 [1:02:30<13:42,  1.57it/s]
1|5182|Loss: 0.8506583571434021:  80%|████████  | 5181/6470 [1:02:31<13:42,  1.57it/s]
1|5182|Loss: 0.8506583571434021:  80%|████████  | 5182/6470 [1:02:31<15:01,  1.43it/s]
1|5183|Loss: 1.0390095710754395:  80%|████████  | 5182/6470 [1:02:31<15:01,  1.43it/s]
1|5183|Loss: 1.0390095710754395:  80%|████████  | 5183/6470 [1:02:32<15:31,  1.38it/s]
1|5184|Loss: 0.9425917267799377:  80%|████████  | 5183/6470 [1:02:32<15:31,  1.38it/s]
1|5184|Loss: 0.9425917267799377:  80%|████████  | 5184/6470 [1:02:32<15:09,  1.41it/s]
1|5185|Loss: 0.8809981942176819:  80%|████████  | 5184/6470 [1:02:33<15:09,  1.41it/s]
1|5185|Loss: 0.8809981942176819:  80%|████████  | 5185/6470 [1:02:33<15:01,  1.43it/s]
1|5186|Loss: 0.8865236043930054:  80%|████████  | 5185/6470 [1:02:34<15:01,  1.43it/s]
1|5186|Loss: 0.8865236043930054:  80%|████████  | 5186/6470 [1:02:34<15:50,  1.35it/s]
1|5187|Loss: 0.7846564054489136:  80%|████████  | 5186/6470 [1:02:34<15:50,  1.35it/s]
1|5187|Loss: 0.7846564054489136:  80%|████████  | 5187/6470 [1:02:35<14:59,  1.43it/s]
1|5188|Loss: 1.0565283298492432:  80%|████████  | 5187/6470 [1:02:35<14:59,  1.43it/s]
1|5188|Loss: 1.0565283298492432:  80%|████████  | 5188/6470 [1:02:35<15:25,  1.39it/s]
1|5189|Loss: 0.8944863677024841:  80%|████████  | 5188/6470 [1:02:36<15:25,  1.39it/s]
1|5189|Loss: 0.8944863677024841:  80%|████████  | 5189/6470 [1:02:36<15:38,  1.36it/s]
1|5190|Loss: 0.8946670293807983:  80%|████████  | 5189/6470 [1:02:37<15:38,  1.36it/s]
1|5190|Loss: 0.8946670293807983:  80%|████████  | 5190/6470 [1:02:37<15:44,  1.36it/s]
1|5191|Loss: 0.7390651702880859:  80%|████████  | 5190/6470 [1:02:37<15:44,  1.36it/s]
1|5191|Loss: 0.7390651702880859:  80%|████████  | 5191/6470 [1:02:37<14:19,  1.49it/s]
1|5192|Loss: 0.6773764491081238:  80%|████████  | 5191/6470 [1:02:38<14:19,  1.49it/s]
1|5192|Loss: 0.6773764491081238:  80%|████████  | 5192/6470 [1:02:38<13:34,  1.57it/s]
1|5193|Loss: 0.8533177375793457:  80%|████████  | 5192/6470 [1:02:38<13:34,  1.57it/s]
1|5193|Loss: 0.8533177375793457:  80%|████████  | 5193/6470 [1:02:39<13:09,  1.62it/s]
1|5194|Loss: 0.9053296446800232:  80%|████████  | 5193/6470 [1:02:39<13:09,  1.62it/s]
1|5194|Loss: 0.9053296446800232:  80%|████████  | 5194/6470 [1:02:39<13:06,  1.62it/s]
1|5195|Loss: 0.9638469815254211:  80%|████████  | 5194/6470 [1:02:39<13:06,  1.62it/s]
1|5195|Loss: 0.9638469815254211:  80%|████████  | 5195/6470 [1:02:40<13:35,  1.56it/s]
1|5196|Loss: 1.0336980819702148:  80%|████████  | 5195/6470 [1:02:40<13:35,  1.56it/s]
1|5196|Loss: 1.0336980819702148:  80%|████████  | 5196/6470 [1:02:41<14:06,  1.50it/s]
1|5197|Loss: 0.881290853023529:  80%|████████  | 5196/6470 [1:02:41<14:06,  1.50it/s] 
1|5197|Loss: 0.881290853023529:  80%|████████  | 5197/6470 [1:02:41<14:04,  1.51it/s]
1|5198|Loss: 0.8912493586540222:  80%|████████  | 5197/6470 [1:02:42<14:04,  1.51it/s]
1|5198|Loss: 0.8912493586540222:  80%|████████  | 5198/6470 [1:02:42<14:11,  1.49it/s]
1|5199|Loss: 0.8786984086036682:  80%|████████  | 5198/6470 [1:02:42<14:11,  1.49it/s]
1|5199|Loss: 0.8786984086036682:  80%|████████  | 5199/6470 [1:02:43<14:05,  1.50it/s]
1|5200|Loss: 1.0296956300735474:  80%|████████  | 5199/6470 [1:02:43<14:05,  1.50it/s]
1|5200|Loss: 1.0296956300735474:  80%|████████  | 5200/6470 [1:02:43<14:12,  1.49it/s]
1|5201|Loss: 0.7874160408973694:  80%|████████  | 5200/6470 [1:02:44<14:12,  1.49it/s]
1|5201|Loss: 0.7874160408973694:  80%|████████  | 5201/6470 [1:02:44<15:02,  1.41it/s]
1|5202|Loss: 0.8349505066871643:  80%|████████  | 5201/6470 [1:02:44<15:02,  1.41it/s]
1|5202|Loss: 0.8349505066871643:  80%|████████  | 5202/6470 [1:02:45<14:45,  1.43it/s]
1|5203|Loss: 0.8857247233390808:  80%|████████  | 5202/6470 [1:02:45<14:45,  1.43it/s]
1|5203|Loss: 0.8857247233390808:  80%|████████  | 5203/6470 [1:02:45<15:09,  1.39it/s]
1|5204|Loss: 0.8132992386817932:  80%|████████  | 5203/6470 [1:02:46<15:09,  1.39it/s]
1|5204|Loss: 0.8132992386817932:  80%|████████  | 5204/6470 [1:02:46<14:54,  1.42it/s]
1|5205|Loss: 0.8821834325790405:  80%|████████  | 5204/6470 [1:02:47<14:54,  1.42it/s]
1|5205|Loss: 0.8821834325790405:  80%|████████  | 5205/6470 [1:02:47<14:47,  1.43it/s]
1|5206|Loss: 0.8934381008148193:  80%|████████  | 5205/6470 [1:02:47<14:47,  1.43it/s]
1|5206|Loss: 0.8934381008148193:  80%|████████  | 5206/6470 [1:02:47<14:10,  1.49it/s]
1|5207|Loss: 0.8185998797416687:  80%|████████  | 5206/6470 [1:02:48<14:10,  1.49it/s]
1|5207|Loss: 0.8185998797416687:  80%|████████  | 5207/6470 [1:02:48<14:52,  1.42it/s]
1|5208|Loss: 0.8917698860168457:  80%|████████  | 5207/6470 [1:02:49<14:52,  1.42it/s]
1|5208|Loss: 0.8917698860168457:  80%|████████  | 5208/6470 [1:02:49<14:44,  1.43it/s]
1|5209|Loss: 0.8328970074653625:  80%|████████  | 5208/6470 [1:02:49<14:44,  1.43it/s]
1|5209|Loss: 0.8328970074653625:  81%|████████  | 5209/6470 [1:02:50<15:33,  1.35it/s]
1|5210|Loss: 0.8718980550765991:  81%|████████  | 5209/6470 [1:02:50<15:33,  1.35it/s]
1|5210|Loss: 0.8718980550765991:  81%|████████  | 5210/6470 [1:02:51<16:31,  1.27it/s]
1|5211|Loss: 0.6905530691146851:  81%|████████  | 5210/6470 [1:02:51<16:31,  1.27it/s]
1|5211|Loss: 0.6905530691146851:  81%|████████  | 5211/6470 [1:02:51<16:12,  1.29it/s]
1|5212|Loss: 0.7777833938598633:  81%|████████  | 5211/6470 [1:02:52<16:12,  1.29it/s]
1|5212|Loss: 0.7777833938598633:  81%|████████  | 5212/6470 [1:02:52<14:10,  1.48it/s]
1|5213|Loss: 0.7446281313896179:  81%|████████  | 5212/6470 [1:02:52<14:10,  1.48it/s]
1|5213|Loss: 0.7446281313896179:  81%|████████  | 5213/6470 [1:02:52<13:58,  1.50it/s]
1|5214|Loss: 0.8805477023124695:  81%|████████  | 5213/6470 [1:02:53<13:58,  1.50it/s]
1|5214|Loss: 0.8805477023124695:  81%|████████  | 5214/6470 [1:02:53<14:24,  1.45it/s]
1|5215|Loss: 1.0202999114990234:  81%|████████  | 5214/6470 [1:02:54<14:24,  1.45it/s]
1|5215|Loss: 1.0202999114990234:  81%|████████  | 5215/6470 [1:02:54<15:12,  1.38it/s]
1|5216|Loss: 0.9397173523902893:  81%|████████  | 5215/6470 [1:02:54<15:12,  1.38it/s]
1|5216|Loss: 0.9397173523902893:  81%|████████  | 5216/6470 [1:02:55<14:47,  1.41it/s]
1|5217|Loss: 0.8851566314697266:  81%|████████  | 5216/6470 [1:02:55<14:47,  1.41it/s]
1|5217|Loss: 0.8851566314697266:  81%|████████  | 5217/6470 [1:02:55<14:13,  1.47it/s]
1|5218|Loss: 0.8152849078178406:  81%|████████  | 5217/6470 [1:02:56<14:13,  1.47it/s]
1|5218|Loss: 0.8152849078178406:  81%|████████  | 5218/6470 [1:02:56<13:58,  1.49it/s]
1|5219|Loss: 0.8715285062789917:  81%|████████  | 5218/6470 [1:02:56<13:58,  1.49it/s]
1|5219|Loss: 0.8715285062789917:  81%|████████  | 5219/6470 [1:02:57<14:33,  1.43it/s]
1|5220|Loss: 0.8523190021514893:  81%|████████  | 5219/6470 [1:02:57<14:33,  1.43it/s]
1|5220|Loss: 0.8523190021514893:  81%|████████  | 5220/6470 [1:02:57<13:26,  1.55it/s]
1|5221|Loss: 0.8376070261001587:  81%|████████  | 5220/6470 [1:02:58<13:26,  1.55it/s]
1|5221|Loss: 0.8376070261001587:  81%|████████  | 5221/6470 [1:02:58<13:25,  1.55it/s]
1|5222|Loss: 0.8119310140609741:  81%|████████  | 5221/6470 [1:02:58<13:25,  1.55it/s]
1|5222|Loss: 0.8119310140609741:  81%|████████  | 5222/6470 [1:02:59<13:31,  1.54it/s]
1|5223|Loss: 0.8889672756195068:  81%|████████  | 5222/6470 [1:02:59<13:31,  1.54it/s]
1|5223|Loss: 0.8889672756195068:  81%|████████  | 5223/6470 [1:02:59<13:55,  1.49it/s]
1|5224|Loss: 0.8712595105171204:  81%|████████  | 5223/6470 [1:03:00<13:55,  1.49it/s]
1|5224|Loss: 0.8712595105171204:  81%|████████  | 5224/6470 [1:03:00<14:45,  1.41it/s]
1|5225|Loss: 1.033740520477295:  81%|████████  | 5224/6470 [1:03:00<14:45,  1.41it/s] 
1|5225|Loss: 1.033740520477295:  81%|████████  | 5225/6470 [1:03:01<14:01,  1.48it/s]
1|5226|Loss: 0.9135080575942993:  81%|████████  | 5225/6470 [1:03:01<14:01,  1.48it/s]
1|5226|Loss: 0.9135080575942993:  81%|████████  | 5226/6470 [1:03:01<13:34,  1.53it/s]
1|5227|Loss: 0.8710465431213379:  81%|████████  | 5226/6470 [1:03:02<13:34,  1.53it/s]
1|5227|Loss: 0.8710465431213379:  81%|████████  | 5227/6470 [1:03:02<14:01,  1.48it/s]
1|5228|Loss: 0.9552900195121765:  81%|████████  | 5227/6470 [1:03:02<14:01,  1.48it/s]
1|5228|Loss: 0.9552900195121765:  81%|████████  | 5228/6470 [1:03:03<14:25,  1.44it/s]
1|5229|Loss: 0.835061252117157:  81%|████████  | 5228/6470 [1:03:03<14:25,  1.44it/s] 
1|5229|Loss: 0.835061252117157:  81%|████████  | 5229/6470 [1:03:03<13:55,  1.48it/s]
1|5230|Loss: 0.7733603119850159:  81%|████████  | 5229/6470 [1:03:04<13:55,  1.48it/s]
1|5230|Loss: 0.7733603119850159:  81%|████████  | 5230/6470 [1:03:04<13:58,  1.48it/s]
1|5231|Loss: 0.8955612182617188:  81%|████████  | 5230/6470 [1:03:04<13:58,  1.48it/s]
1|5231|Loss: 0.8955612182617188:  81%|████████  | 5231/6470 [1:03:05<13:50,  1.49it/s]
1|5232|Loss: 0.821640133857727:  81%|████████  | 5231/6470 [1:03:05<13:50,  1.49it/s] 
1|5232|Loss: 0.821640133857727:  81%|████████  | 5232/6470 [1:03:05<14:05,  1.46it/s]
1|5233|Loss: 0.8853250741958618:  81%|████████  | 5232/6470 [1:03:06<14:05,  1.46it/s]
1|5233|Loss: 0.8853250741958618:  81%|████████  | 5233/6470 [1:03:06<13:57,  1.48it/s]
1|5234|Loss: 0.9183406829833984:  81%|████████  | 5233/6470 [1:03:06<13:57,  1.48it/s]
1|5234|Loss: 0.9183406829833984:  81%|████████  | 5234/6470 [1:03:07<13:37,  1.51it/s]
1|5235|Loss: 0.8853594064712524:  81%|████████  | 5234/6470 [1:03:07<13:37,  1.51it/s]
1|5235|Loss: 0.8853594064712524:  81%|████████  | 5235/6470 [1:03:07<13:30,  1.52it/s]
1|5236|Loss: 0.7754298448562622:  81%|████████  | 5235/6470 [1:03:08<13:30,  1.52it/s]
1|5236|Loss: 0.7754298448562622:  81%|████████  | 5236/6470 [1:03:08<14:31,  1.42it/s]
1|5237|Loss: 0.895568311214447:  81%|████████  | 5236/6470 [1:03:09<14:31,  1.42it/s] 
1|5237|Loss: 0.895568311214447:  81%|████████  | 5237/6470 [1:03:09<15:22,  1.34it/s]
1|5238|Loss: 0.9672116041183472:  81%|████████  | 5237/6470 [1:03:09<15:22,  1.34it/s]
1|5238|Loss: 0.9672116041183472:  81%|████████  | 5238/6470 [1:03:10<15:32,  1.32it/s]
1|5239|Loss: 0.9058210849761963:  81%|████████  | 5238/6470 [1:03:10<15:32,  1.32it/s]
1|5239|Loss: 0.9058210849761963:  81%|████████  | 5239/6470 [1:03:10<14:03,  1.46it/s]
1|5240|Loss: 0.9351291656494141:  81%|████████  | 5239/6470 [1:03:11<14:03,  1.46it/s]
1|5240|Loss: 0.9351291656494141:  81%|████████  | 5240/6470 [1:03:11<14:33,  1.41it/s]
1|5241|Loss: 0.7918915748596191:  81%|████████  | 5240/6470 [1:03:12<14:33,  1.41it/s]
1|5241|Loss: 0.7918915748596191:  81%|████████  | 5241/6470 [1:03:12<15:33,  1.32it/s]
1|5242|Loss: 0.8917587399482727:  81%|████████  | 5241/6470 [1:03:12<15:33,  1.32it/s]
1|5242|Loss: 0.8917587399482727:  81%|████████  | 5242/6470 [1:03:13<14:35,  1.40it/s]
1|5243|Loss: 0.9601843953132629:  81%|████████  | 5242/6470 [1:03:13<14:35,  1.40it/s]
1|5243|Loss: 0.9601843953132629:  81%|████████  | 5243/6470 [1:03:13<14:15,  1.43it/s]
1|5244|Loss: 0.7799822688102722:  81%|████████  | 5243/6470 [1:03:14<14:15,  1.43it/s]
1|5244|Loss: 0.7799822688102722:  81%|████████  | 5244/6470 [1:03:14<15:03,  1.36it/s]
1|5245|Loss: 0.8850077986717224:  81%|████████  | 5244/6470 [1:03:14<15:03,  1.36it/s]
1|5245|Loss: 0.8850077986717224:  81%|████████  | 5245/6470 [1:03:15<14:47,  1.38it/s]
1|5246|Loss: 0.9863901138305664:  81%|████████  | 5245/6470 [1:03:15<14:47,  1.38it/s]
1|5246|Loss: 0.9863901138305664:  81%|████████  | 5246/6470 [1:03:15<13:57,  1.46it/s]
1|5247|Loss: 0.9419487714767456:  81%|████████  | 5246/6470 [1:03:16<13:57,  1.46it/s]
1|5247|Loss: 0.9419487714767456:  81%|████████  | 5247/6470 [1:03:16<13:37,  1.50it/s]
1|5248|Loss: 0.9055588245391846:  81%|████████  | 5247/6470 [1:03:16<13:37,  1.50it/s]
1|5248|Loss: 0.9055588245391846:  81%|████████  | 5248/6470 [1:03:17<13:07,  1.55it/s]
1|5249|Loss: 0.7652538418769836:  81%|████████  | 5248/6470 [1:03:17<13:07,  1.55it/s]
1|5249|Loss: 0.7652538418769836:  81%|████████  | 5249/6470 [1:03:17<12:21,  1.65it/s]
1|5250|Loss: 0.8164157271385193:  81%|████████  | 5249/6470 [1:03:17<12:21,  1.65it/s]
1|5250|Loss: 0.8164157271385193:  81%|████████  | 5250/6470 [1:03:18<12:55,  1.57it/s]
1|5251|Loss: 0.9016531109809875:  81%|████████  | 5250/6470 [1:03:18<12:55,  1.57it/s]
1|5251|Loss: 0.9016531109809875:  81%|████████  | 5251/6470 [1:03:18<13:03,  1.56it/s]
1|5252|Loss: 0.9512082934379578:  81%|████████  | 5251/6470 [1:03:19<13:03,  1.56it/s]
1|5252|Loss: 0.9512082934379578:  81%|████████  | 5252/6470 [1:03:19<13:19,  1.52it/s]
1|5253|Loss: 0.98354572057724:  81%|████████  | 5252/6470 [1:03:20<13:19,  1.52it/s]  
1|5253|Loss: 0.98354572057724:  81%|████████  | 5253/6470 [1:03:20<14:01,  1.45it/s]
1|5254|Loss: 0.8645049929618835:  81%|████████  | 5253/6470 [1:03:20<14:01,  1.45it/s]
1|5254|Loss: 0.8645049929618835:  81%|████████  | 5254/6470 [1:03:21<13:55,  1.46it/s]
1|5255|Loss: 0.865599513053894:  81%|████████  | 5254/6470 [1:03:21<13:55,  1.46it/s] 
1|5255|Loss: 0.865599513053894:  81%|████████  | 5255/6470 [1:03:21<14:05,  1.44it/s]
1|5256|Loss: 0.8030333518981934:  81%|████████  | 5255/6470 [1:03:22<14:05,  1.44it/s]
1|5256|Loss: 0.8030333518981934:  81%|████████  | 5256/6470 [1:03:22<14:19,  1.41it/s]
1|5257|Loss: 0.9029467105865479:  81%|████████  | 5256/6470 [1:03:22<14:19,  1.41it/s]
1|5257|Loss: 0.9029467105865479:  81%|████████▏ | 5257/6470 [1:03:23<13:06,  1.54it/s]
1|5258|Loss: 0.7845724821090698:  81%|████████▏ | 5257/6470 [1:03:23<13:06,  1.54it/s]
1|5258|Loss: 0.7845724821090698:  81%|████████▏ | 5258/6470 [1:03:23<13:14,  1.52it/s]
1|5259|Loss: 0.8424376845359802:  81%|████████▏ | 5258/6470 [1:03:24<13:14,  1.52it/s]
1|5259|Loss: 0.8424376845359802:  81%|████████▏ | 5259/6470 [1:03:24<13:33,  1.49it/s]
1|5260|Loss: 0.980949342250824:  81%|████████▏ | 5259/6470 [1:03:24<13:33,  1.49it/s] 
1|5260|Loss: 0.980949342250824:  81%|████████▏ | 5260/6470 [1:03:24<12:39,  1.59it/s]
1|5261|Loss: 0.8983522057533264:  81%|████████▏ | 5260/6470 [1:03:25<12:39,  1.59it/s]
1|5261|Loss: 0.8983522057533264:  81%|████████▏ | 5261/6470 [1:03:25<11:53,  1.70it/s]
1|5262|Loss: 0.8710347414016724:  81%|████████▏ | 5261/6470 [1:03:25<11:53,  1.70it/s]
1|5262|Loss: 0.8710347414016724:  81%|████████▏ | 5262/6470 [1:03:26<12:43,  1.58it/s]
1|5263|Loss: 0.8386570811271667:  81%|████████▏ | 5262/6470 [1:03:26<12:43,  1.58it/s]
1|5263|Loss: 0.8386570811271667:  81%|████████▏ | 5263/6470 [1:03:26<13:00,  1.55it/s]
1|5264|Loss: 0.7389230728149414:  81%|████████▏ | 5263/6470 [1:03:27<13:00,  1.55it/s]
1|5264|Loss: 0.7389230728149414:  81%|████████▏ | 5264/6470 [1:03:27<12:57,  1.55it/s]
1|5265|Loss: 0.8217098712921143:  81%|████████▏ | 5264/6470 [1:03:27<12:57,  1.55it/s]
1|5265|Loss: 0.8217098712921143:  81%|████████▏ | 5265/6470 [1:03:28<12:54,  1.56it/s]
1|5266|Loss: 0.8703848123550415:  81%|████████▏ | 5265/6470 [1:03:28<12:54,  1.56it/s]
1|5266|Loss: 0.8703848123550415:  81%|████████▏ | 5266/6470 [1:03:28<12:48,  1.57it/s]
1|5267|Loss: 0.9787384867668152:  81%|████████▏ | 5266/6470 [1:03:29<12:48,  1.57it/s]
1|5267|Loss: 0.9787384867668152:  81%|████████▏ | 5267/6470 [1:03:29<12:50,  1.56it/s]
1|5268|Loss: 0.6512742042541504:  81%|████████▏ | 5267/6470 [1:03:29<12:50,  1.56it/s]
1|5268|Loss: 0.6512742042541504:  81%|████████▏ | 5268/6470 [1:03:30<13:57,  1.44it/s]
1|5269|Loss: 0.9460432529449463:  81%|████████▏ | 5268/6470 [1:03:30<13:57,  1.44it/s]
1|5269|Loss: 0.9460432529449463:  81%|████████▏ | 5269/6470 [1:03:31<14:27,  1.38it/s]
1|5270|Loss: 0.954585611820221:  81%|████████▏ | 5269/6470 [1:03:31<14:27,  1.38it/s] 
1|5270|Loss: 0.954585611820221:  81%|████████▏ | 5270/6470 [1:03:31<14:07,  1.42it/s]
1|5271|Loss: 0.7739176750183105:  81%|████████▏ | 5270/6470 [1:03:32<14:07,  1.42it/s]
1|5271|Loss: 0.7739176750183105:  81%|████████▏ | 5271/6470 [1:03:32<14:54,  1.34it/s]
1|5272|Loss: 0.7965282797813416:  81%|████████▏ | 5271/6470 [1:03:32<14:54,  1.34it/s]
1|5272|Loss: 0.7965282797813416:  81%|████████▏ | 5272/6470 [1:03:33<15:18,  1.30it/s]
1|5273|Loss: 1.1045997142791748:  81%|████████▏ | 5272/6470 [1:03:33<15:18,  1.30it/s]
1|5273|Loss: 1.1045997142791748:  81%|████████▏ | 5273/6470 [1:03:34<15:26,  1.29it/s]
1|5274|Loss: 0.9435999989509583:  81%|████████▏ | 5273/6470 [1:03:34<15:26,  1.29it/s]
1|5274|Loss: 0.9435999989509583:  82%|████████▏ | 5274/6470 [1:03:34<15:33,  1.28it/s]
1|5275|Loss: 0.9008197784423828:  82%|████████▏ | 5274/6470 [1:03:35<15:33,  1.28it/s]
1|5275|Loss: 0.9008197784423828:  82%|████████▏ | 5275/6470 [1:03:35<14:54,  1.34it/s]
1|5276|Loss: 0.9367631077766418:  82%|████████▏ | 5275/6470 [1:03:35<14:54,  1.34it/s]
1|5276|Loss: 0.9367631077766418:  82%|████████▏ | 5276/6470 [1:03:36<14:43,  1.35it/s]
1|5277|Loss: 0.8778561353683472:  82%|████████▏ | 5276/6470 [1:03:36<14:43,  1.35it/s]
1|5277|Loss: 0.8778561353683472:  82%|████████▏ | 5277/6470 [1:03:36<14:17,  1.39it/s]
1|5278|Loss: 0.814889669418335:  82%|████████▏ | 5277/6470 [1:03:37<14:17,  1.39it/s] 
1|5278|Loss: 0.814889669418335:  82%|████████▏ | 5278/6470 [1:03:37<13:34,  1.46it/s]
1|5279|Loss: 0.8115699291229248:  82%|████████▏ | 5278/6470 [1:03:37<13:34,  1.46it/s]
1|5279|Loss: 0.8115699291229248:  82%|████████▏ | 5279/6470 [1:03:38<13:16,  1.50it/s]
1|5280|Loss: 0.753426194190979:  82%|████████▏ | 5279/6470 [1:03:38<13:16,  1.50it/s] 
1|5280|Loss: 0.753426194190979:  82%|████████▏ | 5280/6470 [1:03:38<13:34,  1.46it/s]
1|5281|Loss: 0.8326513171195984:  82%|████████▏ | 5280/6470 [1:03:39<13:34,  1.46it/s]
1|5281|Loss: 0.8326513171195984:  82%|████████▏ | 5281/6470 [1:03:39<13:09,  1.51it/s]
1|5282|Loss: 0.9213008284568787:  82%|████████▏ | 5281/6470 [1:03:39<13:09,  1.51it/s]
1|5282|Loss: 0.9213008284568787:  82%|████████▏ | 5282/6470 [1:03:40<14:06,  1.40it/s]
1|5283|Loss: 0.7509047389030457:  82%|████████▏ | 5282/6470 [1:03:40<14:06,  1.40it/s]
1|5283|Loss: 0.7509047389030457:  82%|████████▏ | 5283/6470 [1:03:41<15:09,  1.30it/s]
1|5284|Loss: 0.758740246295929:  82%|████████▏ | 5283/6470 [1:03:41<15:09,  1.30it/s] 
1|5284|Loss: 0.758740246295929:  82%|████████▏ | 5284/6470 [1:03:42<15:39,  1.26it/s]
1|5285|Loss: 0.9060338735580444:  82%|████████▏ | 5284/6470 [1:03:42<15:39,  1.26it/s]
1|5285|Loss: 0.9060338735580444:  82%|████████▏ | 5285/6470 [1:03:42<15:26,  1.28it/s]
1|5286|Loss: 1.12476646900177:  82%|████████▏ | 5285/6470 [1:03:43<15:26,  1.28it/s]  
1|5286|Loss: 1.12476646900177:  82%|████████▏ | 5286/6470 [1:03:43<15:36,  1.26it/s]
1|5287|Loss: 0.8973408341407776:  82%|████████▏ | 5286/6470 [1:03:44<15:36,  1.26it/s]
1|5287|Loss: 0.8973408341407776:  82%|████████▏ | 5287/6470 [1:03:44<14:56,  1.32it/s]
1|5288|Loss: 0.8955536484718323:  82%|████████▏ | 5287/6470 [1:03:44<14:56,  1.32it/s]
1|5288|Loss: 0.8955536484718323:  82%|████████▏ | 5288/6470 [1:03:44<14:05,  1.40it/s]
1|5289|Loss: 0.9831935167312622:  82%|████████▏ | 5288/6470 [1:03:45<14:05,  1.40it/s]
1|5289|Loss: 0.9831935167312622:  82%|████████▏ | 5289/6470 [1:03:45<13:20,  1.48it/s]
1|5290|Loss: 0.7502105236053467:  82%|████████▏ | 5289/6470 [1:03:46<13:20,  1.48it/s]
1|5290|Loss: 0.7502105236053467:  82%|████████▏ | 5290/6470 [1:03:46<14:12,  1.38it/s]
1|5291|Loss: 0.8377475142478943:  82%|████████▏ | 5290/6470 [1:03:46<14:12,  1.38it/s]
1|5291|Loss: 0.8377475142478943:  82%|████████▏ | 5291/6470 [1:03:47<13:57,  1.41it/s]
1|5292|Loss: 0.8827184438705444:  82%|████████▏ | 5291/6470 [1:03:47<13:57,  1.41it/s]
1|5292|Loss: 0.8827184438705444:  82%|████████▏ | 5292/6470 [1:03:47<14:13,  1.38it/s]
1|5293|Loss: 0.815421462059021:  82%|████████▏ | 5292/6470 [1:03:48<14:13,  1.38it/s] 
1|5293|Loss: 0.815421462059021:  82%|████████▏ | 5293/6470 [1:03:48<14:25,  1.36it/s]
1|5294|Loss: 0.8692747354507446:  82%|████████▏ | 5293/6470 [1:03:48<14:25,  1.36it/s]
1|5294|Loss: 0.8692747354507446:  82%|████████▏ | 5294/6470 [1:03:49<13:53,  1.41it/s]
1|5295|Loss: 0.7681778073310852:  82%|████████▏ | 5294/6470 [1:03:49<13:53,  1.41it/s]
1|5295|Loss: 0.7681778073310852:  82%|████████▏ | 5295/6470 [1:03:49<13:53,  1.41it/s]
1|5296|Loss: 0.8922258019447327:  82%|████████▏ | 5295/6470 [1:03:50<13:53,  1.41it/s]
1|5296|Loss: 0.8922258019447327:  82%|████████▏ | 5296/6470 [1:03:50<14:18,  1.37it/s]
1|5297|Loss: 0.9650758504867554:  82%|████████▏ | 5296/6470 [1:03:51<14:18,  1.37it/s]
1|5297|Loss: 0.9650758504867554:  82%|████████▏ | 5297/6470 [1:03:51<13:54,  1.41it/s]
1|5298|Loss: 0.9299972057342529:  82%|████████▏ | 5297/6470 [1:03:51<13:54,  1.41it/s]
1|5298|Loss: 0.9299972057342529:  82%|████████▏ | 5298/6470 [1:03:52<14:07,  1.38it/s]
1|5299|Loss: 0.8634505271911621:  82%|████████▏ | 5298/6470 [1:03:52<14:07,  1.38it/s]
1|5299|Loss: 0.8634505271911621:  82%|████████▏ | 5299/6470 [1:03:53<14:48,  1.32it/s]
1|5300|Loss: 0.8978568911552429:  82%|████████▏ | 5299/6470 [1:03:53<14:48,  1.32it/s]
1|5300|Loss: 0.8978568911552429:  82%|████████▏ | 5300/6470 [1:03:53<14:03,  1.39it/s]
1|5301|Loss: 0.8882787227630615:  82%|████████▏ | 5300/6470 [1:03:54<14:03,  1.39it/s]
1|5301|Loss: 0.8882787227630615:  82%|████████▏ | 5301/6470 [1:03:54<14:31,  1.34it/s]
1|5302|Loss: 0.8539900779724121:  82%|████████▏ | 5301/6470 [1:03:54<14:31,  1.34it/s]
1|5302|Loss: 0.8539900779724121:  82%|████████▏ | 5302/6470 [1:03:55<14:36,  1.33it/s]
1|5303|Loss: 0.813628613948822:  82%|████████▏ | 5302/6470 [1:03:55<14:36,  1.33it/s] 
1|5303|Loss: 0.813628613948822:  82%|████████▏ | 5303/6470 [1:03:56<15:08,  1.28it/s]
1|5304|Loss: 0.873112142086029:  82%|████████▏ | 5303/6470 [1:03:56<15:08,  1.28it/s]
1|5304|Loss: 0.873112142086029:  82%|████████▏ | 5304/6470 [1:03:56<14:31,  1.34it/s]
1|5305|Loss: 0.9586729407310486:  82%|████████▏ | 5304/6470 [1:03:57<14:31,  1.34it/s]
1|5305|Loss: 0.9586729407310486:  82%|████████▏ | 5305/6470 [1:03:57<14:22,  1.35it/s]
1|5306|Loss: 0.7797635197639465:  82%|████████▏ | 5305/6470 [1:03:57<14:22,  1.35it/s]
1|5306|Loss: 0.7797635197639465:  82%|████████▏ | 5306/6470 [1:03:57<12:43,  1.52it/s]
1|5307|Loss: 0.9157339334487915:  82%|████████▏ | 5306/6470 [1:03:58<12:43,  1.52it/s]
1|5307|Loss: 0.9157339334487915:  82%|████████▏ | 5307/6470 [1:03:58<12:16,  1.58it/s]
1|5308|Loss: 0.7878583669662476:  82%|████████▏ | 5307/6470 [1:03:58<12:16,  1.58it/s]
1|5308|Loss: 0.7878583669662476:  82%|████████▏ | 5308/6470 [1:03:59<13:07,  1.47it/s]
1|5309|Loss: 0.9335578680038452:  82%|████████▏ | 5308/6470 [1:03:59<13:07,  1.47it/s]
1|5309|Loss: 0.9335578680038452:  82%|████████▏ | 5309/6470 [1:04:00<13:37,  1.42it/s]
1|5310|Loss: 0.9054307341575623:  82%|████████▏ | 5309/6470 [1:04:00<13:37,  1.42it/s]
1|5310|Loss: 0.9054307341575623:  82%|████████▏ | 5310/6470 [1:04:00<13:52,  1.39it/s]
1|5311|Loss: 0.8794391751289368:  82%|████████▏ | 5310/6470 [1:04:01<13:52,  1.39it/s]
1|5311|Loss: 0.8794391751289368:  82%|████████▏ | 5311/6470 [1:04:01<12:55,  1.49it/s]
1|5312|Loss: 0.8371227979660034:  82%|████████▏ | 5311/6470 [1:04:01<12:55,  1.49it/s]
1|5312|Loss: 0.8371227979660034:  82%|████████▏ | 5312/6470 [1:04:01<12:34,  1.54it/s]
1|5313|Loss: 0.8276222348213196:  82%|████████▏ | 5312/6470 [1:04:02<12:34,  1.54it/s]
1|5313|Loss: 0.8276222348213196:  82%|████████▏ | 5313/6470 [1:04:02<12:03,  1.60it/s]
1|5314|Loss: 0.7877497673034668:  82%|████████▏ | 5313/6470 [1:04:02<12:03,  1.60it/s]
1|5314|Loss: 0.7877497673034668:  82%|████████▏ | 5314/6470 [1:04:03<12:27,  1.55it/s]
1|5315|Loss: 0.7937505841255188:  82%|████████▏ | 5314/6470 [1:04:03<12:27,  1.55it/s]
1|5315|Loss: 0.7937505841255188:  82%|████████▏ | 5315/6470 [1:04:03<13:04,  1.47it/s]
1|5316|Loss: 0.8111907839775085:  82%|████████▏ | 5315/6470 [1:04:04<13:04,  1.47it/s]
1|5316|Loss: 0.8111907839775085:  82%|████████▏ | 5316/6470 [1:04:04<12:34,  1.53it/s]
1|5317|Loss: 0.9461258053779602:  82%|████████▏ | 5316/6470 [1:04:04<12:34,  1.53it/s]
1|5317|Loss: 0.9461258053779602:  82%|████████▏ | 5317/6470 [1:04:05<12:09,  1.58it/s]
1|5318|Loss: 0.8713018894195557:  82%|████████▏ | 5317/6470 [1:04:05<12:09,  1.58it/s]
1|5318|Loss: 0.8713018894195557:  82%|████████▏ | 5318/6470 [1:04:05<12:33,  1.53it/s]
1|5319|Loss: 0.8873013257980347:  82%|████████▏ | 5318/6470 [1:04:06<12:33,  1.53it/s]
1|5319|Loss: 0.8873013257980347:  82%|████████▏ | 5319/6470 [1:04:06<13:01,  1.47it/s]
1|5320|Loss: 0.8700094223022461:  82%|████████▏ | 5319/6470 [1:04:06<13:01,  1.47it/s]
1|5320|Loss: 0.8700094223022461:  82%|████████▏ | 5320/6470 [1:04:07<13:19,  1.44it/s]
1|5321|Loss: 0.891714334487915:  82%|████████▏ | 5320/6470 [1:04:07<13:19,  1.44it/s] 
1|5321|Loss: 0.891714334487915:  82%|████████▏ | 5321/6470 [1:04:08<13:23,  1.43it/s]
1|5322|Loss: 0.7678689360618591:  82%|████████▏ | 5321/6470 [1:04:08<13:23,  1.43it/s]
1|5322|Loss: 0.7678689360618591:  82%|████████▏ | 5322/6470 [1:04:08<12:41,  1.51it/s]
1|5323|Loss: 0.8398803472518921:  82%|████████▏ | 5322/6470 [1:04:09<12:41,  1.51it/s]
1|5323|Loss: 0.8398803472518921:  82%|████████▏ | 5323/6470 [1:04:09<13:19,  1.43it/s]
1|5324|Loss: 0.8782463669776917:  82%|████████▏ | 5323/6470 [1:04:09<13:19,  1.43it/s]
1|5324|Loss: 0.8782463669776917:  82%|████████▏ | 5324/6470 [1:04:10<13:37,  1.40it/s]
1|5325|Loss: 0.8009297847747803:  82%|████████▏ | 5324/6470 [1:04:10<13:37,  1.40it/s]
1|5325|Loss: 0.8009297847747803:  82%|████████▏ | 5325/6470 [1:04:10<13:46,  1.38it/s]
1|5326|Loss: 0.831712007522583:  82%|████████▏ | 5325/6470 [1:04:11<13:46,  1.38it/s] 
1|5326|Loss: 0.831712007522583:  82%|████████▏ | 5326/6470 [1:04:11<13:20,  1.43it/s]
1|5327|Loss: 0.7879594564437866:  82%|████████▏ | 5326/6470 [1:04:11<13:20,  1.43it/s]
1|5327|Loss: 0.7879594564437866:  82%|████████▏ | 5327/6470 [1:04:12<14:28,  1.32it/s]
1|5328|Loss: 0.9667590260505676:  82%|████████▏ | 5327/6470 [1:04:12<14:28,  1.32it/s]
1|5328|Loss: 0.9667590260505676:  82%|████████▏ | 5328/6470 [1:04:13<13:57,  1.36it/s]
1|5329|Loss: 0.8615735769271851:  82%|████████▏ | 5328/6470 [1:04:13<13:57,  1.36it/s]
1|5329|Loss: 0.8615735769271851:  82%|████████▏ | 5329/6470 [1:04:13<13:21,  1.42it/s]
1|5330|Loss: 0.929576575756073:  82%|████████▏ | 5329/6470 [1:04:14<13:21,  1.42it/s] 
1|5330|Loss: 0.929576575756073:  82%|████████▏ | 5330/6470 [1:04:14<13:51,  1.37it/s]
1|5331|Loss: 0.9795283079147339:  82%|████████▏ | 5330/6470 [1:04:14<13:51,  1.37it/s]
1|5331|Loss: 0.9795283079147339:  82%|████████▏ | 5331/6470 [1:04:15<14:35,  1.30it/s]
1|5332|Loss: 0.8502001166343689:  82%|████████▏ | 5331/6470 [1:04:15<14:35,  1.30it/s]
1|5332|Loss: 0.8502001166343689:  82%|████████▏ | 5332/6470 [1:04:16<13:51,  1.37it/s]
1|5333|Loss: 0.8849241733551025:  82%|████████▏ | 5332/6470 [1:04:16<13:51,  1.37it/s]
1|5333|Loss: 0.8849241733551025:  82%|████████▏ | 5333/6470 [1:04:16<13:16,  1.43it/s]
1|5334|Loss: 0.9561281800270081:  82%|████████▏ | 5333/6470 [1:04:17<13:16,  1.43it/s]
1|5334|Loss: 0.9561281800270081:  82%|████████▏ | 5334/6470 [1:04:17<13:23,  1.41it/s]
1|5335|Loss: 0.9644260406494141:  82%|████████▏ | 5334/6470 [1:04:17<13:23,  1.41it/s]
1|5335|Loss: 0.9644260406494141:  82%|████████▏ | 5335/6470 [1:04:18<13:24,  1.41it/s]
1|5336|Loss: 0.899603545665741:  82%|████████▏ | 5335/6470 [1:04:18<13:24,  1.41it/s] 
1|5336|Loss: 0.899603545665741:  82%|████████▏ | 5336/6470 [1:04:18<13:34,  1.39it/s]
1|5337|Loss: 0.8433761596679688:  82%|████████▏ | 5336/6470 [1:04:19<13:34,  1.39it/s]
1|5337|Loss: 0.8433761596679688:  82%|████████▏ | 5337/6470 [1:04:19<13:55,  1.36it/s]
1|5338|Loss: 0.859080970287323:  82%|████████▏ | 5337/6470 [1:04:20<13:55,  1.36it/s] 
1|5338|Loss: 0.859080970287323:  83%|████████▎ | 5338/6470 [1:04:20<14:31,  1.30it/s]
1|5339|Loss: 1.1126377582550049:  83%|████████▎ | 5338/6470 [1:04:20<14:31,  1.30it/s]
1|5339|Loss: 1.1126377582550049:  83%|████████▎ | 5339/6470 [1:04:21<14:55,  1.26it/s]
1|5340|Loss: 0.7594519257545471:  83%|████████▎ | 5339/6470 [1:04:21<14:55,  1.26it/s]
1|5340|Loss: 0.7594519257545471:  83%|████████▎ | 5340/6470 [1:04:22<15:20,  1.23it/s]
1|5341|Loss: 0.7674204111099243:  83%|████████▎ | 5340/6470 [1:04:22<15:20,  1.23it/s]
1|5341|Loss: 0.7674204111099243:  83%|████████▎ | 5341/6470 [1:04:22<15:21,  1.23it/s]
1|5342|Loss: 1.0129677057266235:  83%|████████▎ | 5341/6470 [1:04:23<15:21,  1.23it/s]
1|5342|Loss: 1.0129677057266235:  83%|████████▎ | 5342/6470 [1:04:23<14:57,  1.26it/s]
1|5343|Loss: 0.7920411825180054:  83%|████████▎ | 5342/6470 [1:04:24<14:57,  1.26it/s]
1|5343|Loss: 0.7920411825180054:  83%|████████▎ | 5343/6470 [1:04:24<14:03,  1.34it/s]
1|5344|Loss: 0.8637703657150269:  83%|████████▎ | 5343/6470 [1:04:24<14:03,  1.34it/s]
1|5344|Loss: 0.8637703657150269:  83%|████████▎ | 5344/6470 [1:04:25<13:47,  1.36it/s]
1|5345|Loss: 0.8599988222122192:  83%|████████▎ | 5344/6470 [1:04:25<13:47,  1.36it/s]
1|5345|Loss: 0.8599988222122192:  83%|████████▎ | 5345/6470 [1:04:25<13:06,  1.43it/s]
1|5346|Loss: 0.9290470480918884:  83%|████████▎ | 5345/6470 [1:04:26<13:06,  1.43it/s]
1|5346|Loss: 0.9290470480918884:  83%|████████▎ | 5346/6470 [1:04:26<12:41,  1.48it/s]
1|5347|Loss: 0.9656027555465698:  83%|████████▎ | 5346/6470 [1:04:26<12:41,  1.48it/s]
1|5347|Loss: 0.9656027555465698:  83%|████████▎ | 5347/6470 [1:04:26<12:04,  1.55it/s]
1|5348|Loss: 0.8451697826385498:  83%|████████▎ | 5347/6470 [1:04:27<12:04,  1.55it/s]
1|5348|Loss: 0.8451697826385498:  83%|████████▎ | 5348/6470 [1:04:27<12:22,  1.51it/s]
1|5349|Loss: 0.7951938509941101:  83%|████████▎ | 5348/6470 [1:04:27<12:22,  1.51it/s]
1|5349|Loss: 0.7951938509941101:  83%|████████▎ | 5349/6470 [1:04:28<12:50,  1.46it/s]
1|5350|Loss: 0.9468727707862854:  83%|████████▎ | 5349/6470 [1:04:28<12:50,  1.46it/s]
1|5350|Loss: 0.9468727707862854:  83%|████████▎ | 5350/6470 [1:04:28<12:36,  1.48it/s]
1|5351|Loss: 0.9757104516029358:  83%|████████▎ | 5350/6470 [1:04:29<12:36,  1.48it/s]
1|5351|Loss: 0.9757104516029358:  83%|████████▎ | 5351/6470 [1:04:29<13:14,  1.41it/s]
1|5352|Loss: 0.8895099759101868:  83%|████████▎ | 5351/6470 [1:04:30<13:14,  1.41it/s]
1|5352|Loss: 0.8895099759101868:  83%|████████▎ | 5352/6470 [1:04:30<13:33,  1.37it/s]
1|5353|Loss: 0.7851269841194153:  83%|████████▎ | 5352/6470 [1:04:30<13:33,  1.37it/s]
1|5353|Loss: 0.7851269841194153:  83%|████████▎ | 5353/6470 [1:04:31<13:30,  1.38it/s]
1|5354|Loss: 0.9262372851371765:  83%|████████▎ | 5353/6470 [1:04:31<13:30,  1.38it/s]
1|5354|Loss: 0.9262372851371765:  83%|████████▎ | 5354/6470 [1:04:31<13:06,  1.42it/s]
1|5355|Loss: 0.8544741272926331:  83%|████████▎ | 5354/6470 [1:04:32<13:06,  1.42it/s]
1|5355|Loss: 0.8544741272926331:  83%|████████▎ | 5355/6470 [1:04:32<11:31,  1.61it/s]
1|5356|Loss: 0.8203319311141968:  83%|████████▎ | 5355/6470 [1:04:32<11:31,  1.61it/s]
1|5356|Loss: 0.8203319311141968:  83%|████████▎ | 5356/6470 [1:04:32<11:36,  1.60it/s]
1|5357|Loss: 0.8479185700416565:  83%|████████▎ | 5356/6470 [1:04:33<11:36,  1.60it/s]
1|5357|Loss: 0.8479185700416565:  83%|████████▎ | 5357/6470 [1:04:33<11:30,  1.61it/s]
1|5358|Loss: 0.8978123664855957:  83%|████████▎ | 5357/6470 [1:04:33<11:30,  1.61it/s]
1|5358|Loss: 0.8978123664855957:  83%|████████▎ | 5358/6470 [1:04:34<11:33,  1.60it/s]
1|5359|Loss: 0.8115937113761902:  83%|████████▎ | 5358/6470 [1:04:34<11:33,  1.60it/s]
1|5359|Loss: 0.8115937113761902:  83%|████████▎ | 5359/6470 [1:04:34<12:06,  1.53it/s]
1|5360|Loss: 0.6950762867927551:  83%|████████▎ | 5359/6470 [1:04:35<12:06,  1.53it/s]
1|5360|Loss: 0.6950762867927551:  83%|████████▎ | 5360/6470 [1:04:35<12:24,  1.49it/s]
1|5361|Loss: 0.8942477107048035:  83%|████████▎ | 5360/6470 [1:04:36<12:24,  1.49it/s]
1|5361|Loss: 0.8942477107048035:  83%|████████▎ | 5361/6470 [1:04:36<13:10,  1.40it/s]
1|5362|Loss: 0.8576471209526062:  83%|████████▎ | 5361/6470 [1:04:36<13:10,  1.40it/s]
1|5362|Loss: 0.8576471209526062:  83%|████████▎ | 5362/6470 [1:04:37<13:23,  1.38it/s]
1|5363|Loss: 0.8073993921279907:  83%|████████▎ | 5362/6470 [1:04:37<13:23,  1.38it/s]
1|5363|Loss: 0.8073993921279907:  83%|████████▎ | 5363/6470 [1:04:37<13:41,  1.35it/s]
1|5364|Loss: 0.8443994522094727:  83%|████████▎ | 5363/6470 [1:04:38<13:41,  1.35it/s]
1|5364|Loss: 0.8443994522094727:  83%|████████▎ | 5364/6470 [1:04:38<14:15,  1.29it/s]
1|5365|Loss: 0.964346170425415:  83%|████████▎ | 5364/6470 [1:04:39<14:15,  1.29it/s] 
1|5365|Loss: 0.964346170425415:  83%|████████▎ | 5365/6470 [1:04:39<14:04,  1.31it/s]
1|5366|Loss: 0.9330943822860718:  83%|████████▎ | 5365/6470 [1:04:39<14:04,  1.31it/s]
1|5366|Loss: 0.9330943822860718:  83%|████████▎ | 5366/6470 [1:04:40<13:10,  1.40it/s]
1|5367|Loss: 0.7791606783866882:  83%|████████▎ | 5366/6470 [1:04:40<13:10,  1.40it/s]
1|5367|Loss: 0.7791606783866882:  83%|████████▎ | 5367/6470 [1:04:40<12:40,  1.45it/s]
1|5368|Loss: 0.9886121153831482:  83%|████████▎ | 5367/6470 [1:04:41<12:40,  1.45it/s]
1|5368|Loss: 0.9886121153831482:  83%|████████▎ | 5368/6470 [1:04:41<13:13,  1.39it/s]
1|5369|Loss: 1.2990822792053223:  83%|████████▎ | 5368/6470 [1:04:42<13:13,  1.39it/s]
1|5369|Loss: 1.2990822792053223:  83%|████████▎ | 5369/6470 [1:04:42<13:53,  1.32it/s]
1|5370|Loss: 0.8413888812065125:  83%|████████▎ | 5369/6470 [1:04:42<13:53,  1.32it/s]
1|5370|Loss: 0.8413888812065125:  83%|████████▎ | 5370/6470 [1:04:43<14:42,  1.25it/s]
1|5371|Loss: 0.9011074900627136:  83%|████████▎ | 5370/6470 [1:04:43<14:42,  1.25it/s]
1|5371|Loss: 0.9011074900627136:  83%|████████▎ | 5371/6470 [1:04:43<13:44,  1.33it/s]
1|5372|Loss: 0.8262965679168701:  83%|████████▎ | 5371/6470 [1:04:44<13:44,  1.33it/s]
1|5372|Loss: 0.8262965679168701:  83%|████████▎ | 5372/6470 [1:04:44<13:32,  1.35it/s]
1|5373|Loss: 0.9095350503921509:  83%|████████▎ | 5372/6470 [1:04:45<13:32,  1.35it/s]
1|5373|Loss: 0.9095350503921509:  83%|████████▎ | 5373/6470 [1:04:45<13:32,  1.35it/s]
1|5374|Loss: 0.9222509860992432:  83%|████████▎ | 5373/6470 [1:04:45<13:32,  1.35it/s]
1|5374|Loss: 0.9222509860992432:  83%|████████▎ | 5374/6470 [1:04:46<13:45,  1.33it/s]
1|5375|Loss: 0.71431964635849:  83%|████████▎ | 5374/6470 [1:04:46<13:45,  1.33it/s]  
1|5375|Loss: 0.71431964635849:  83%|████████▎ | 5375/6470 [1:04:47<15:21,  1.19it/s]
1|5376|Loss: 0.946967601776123:  83%|████████▎ | 5375/6470 [1:04:47<15:21,  1.19it/s]
1|5376|Loss: 0.946967601776123:  83%|████████▎ | 5376/6470 [1:04:47<14:27,  1.26it/s]
1|5377|Loss: 0.7177442908287048:  83%|████████▎ | 5376/6470 [1:04:48<14:27,  1.26it/s]
1|5377|Loss: 0.7177442908287048:  83%|████████▎ | 5377/6470 [1:04:48<12:58,  1.40it/s]
1|5378|Loss: 0.731985867023468:  83%|████████▎ | 5377/6470 [1:04:48<12:58,  1.40it/s] 
1|5378|Loss: 0.731985867023468:  83%|████████▎ | 5378/6470 [1:04:49<12:19,  1.48it/s]
1|5379|Loss: 0.9190090894699097:  83%|████████▎ | 5378/6470 [1:04:49<12:19,  1.48it/s]
1|5379|Loss: 0.9190090894699097:  83%|████████▎ | 5379/6470 [1:04:49<11:47,  1.54it/s]
1|5380|Loss: 0.8490341901779175:  83%|████████▎ | 5379/6470 [1:04:49<11:47,  1.54it/s]
1|5380|Loss: 0.8490341901779175:  83%|████████▎ | 5380/6470 [1:04:50<10:52,  1.67it/s]
1|5381|Loss: 0.8501860499382019:  83%|████████▎ | 5380/6470 [1:04:50<10:52,  1.67it/s]
1|5381|Loss: 0.8501860499382019:  83%|████████▎ | 5381/6470 [1:04:50<10:44,  1.69it/s]
1|5382|Loss: 0.7839671969413757:  83%|████████▎ | 5381/6470 [1:04:51<10:44,  1.69it/s]
1|5382|Loss: 0.7839671969413757:  83%|████████▎ | 5382/6470 [1:04:51<11:21,  1.60it/s]
1|5383|Loss: 0.8921648859977722:  83%|████████▎ | 5382/6470 [1:04:51<11:21,  1.60it/s]
1|5383|Loss: 0.8921648859977722:  83%|████████▎ | 5383/6470 [1:04:52<12:00,  1.51it/s]
1|5384|Loss: 0.9245753884315491:  83%|████████▎ | 5383/6470 [1:04:52<12:00,  1.51it/s]
1|5384|Loss: 0.9245753884315491:  83%|████████▎ | 5384/6470 [1:04:52<12:14,  1.48it/s]
1|5385|Loss: 0.7910866141319275:  83%|████████▎ | 5384/6470 [1:04:53<12:14,  1.48it/s]
1|5385|Loss: 0.7910866141319275:  83%|████████▎ | 5385/6470 [1:04:53<11:58,  1.51it/s]
1|5386|Loss: 1.1563104391098022:  83%|████████▎ | 5385/6470 [1:04:53<11:58,  1.51it/s]
1|5386|Loss: 1.1563104391098022:  83%|████████▎ | 5386/6470 [1:04:54<11:19,  1.60it/s]
1|5387|Loss: 0.8363118171691895:  83%|████████▎ | 5386/6470 [1:04:54<11:19,  1.60it/s]
1|5387|Loss: 0.8363118171691895:  83%|████████▎ | 5387/6470 [1:04:54<11:13,  1.61it/s]
1|5388|Loss: 0.8970206379890442:  83%|████████▎ | 5387/6470 [1:04:55<11:13,  1.61it/s]
1|5388|Loss: 0.8970206379890442:  83%|████████▎ | 5388/6470 [1:04:55<11:47,  1.53it/s]
1|5389|Loss: 1.0723257064819336:  83%|████████▎ | 5388/6470 [1:04:55<11:47,  1.53it/s]
1|5389|Loss: 1.0723257064819336:  83%|████████▎ | 5389/6470 [1:04:56<12:15,  1.47it/s]
1|5390|Loss: 0.7717468738555908:  83%|████████▎ | 5389/6470 [1:04:56<12:15,  1.47it/s]
1|5390|Loss: 0.7717468738555908:  83%|████████▎ | 5390/6470 [1:04:56<12:49,  1.40it/s]
1|5391|Loss: 0.8758296966552734:  83%|████████▎ | 5390/6470 [1:04:57<12:49,  1.40it/s]
1|5391|Loss: 0.8758296966552734:  83%|████████▎ | 5391/6470 [1:04:57<13:19,  1.35it/s]
1|5392|Loss: 0.8133681416511536:  83%|████████▎ | 5391/6470 [1:04:58<13:19,  1.35it/s]
1|5392|Loss: 0.8133681416511536:  83%|████████▎ | 5392/6470 [1:04:58<13:38,  1.32it/s]
1|5393|Loss: 0.9358465075492859:  83%|████████▎ | 5392/6470 [1:04:58<13:38,  1.32it/s]
1|5393|Loss: 0.9358465075492859:  83%|████████▎ | 5393/6470 [1:04:59<13:51,  1.30it/s]
1|5394|Loss: 0.9613810181617737:  83%|████████▎ | 5393/6470 [1:04:59<13:51,  1.30it/s]
1|5394|Loss: 0.9613810181617737:  83%|████████▎ | 5394/6470 [1:05:00<13:46,  1.30it/s]
1|5395|Loss: 1.0164833068847656:  83%|████████▎ | 5394/6470 [1:05:00<13:46,  1.30it/s]
1|5395|Loss: 1.0164833068847656:  83%|████████▎ | 5395/6470 [1:05:00<12:37,  1.42it/s]
1|5396|Loss: 0.8528075814247131:  83%|████████▎ | 5395/6470 [1:05:01<12:37,  1.42it/s]
1|5396|Loss: 0.8528075814247131:  83%|████████▎ | 5396/6470 [1:05:01<12:29,  1.43it/s]
1|5397|Loss: 0.8489939570426941:  83%|████████▎ | 5396/6470 [1:05:01<12:29,  1.43it/s]
1|5397|Loss: 0.8489939570426941:  83%|████████▎ | 5397/6470 [1:05:02<12:44,  1.40it/s]
1|5398|Loss: 0.7971177697181702:  83%|████████▎ | 5397/6470 [1:05:02<12:44,  1.40it/s]
1|5398|Loss: 0.7971177697181702:  83%|████████▎ | 5398/6470 [1:05:02<12:55,  1.38it/s]
1|5399|Loss: 0.9560714960098267:  83%|████████▎ | 5398/6470 [1:05:03<12:55,  1.38it/s]
1|5399|Loss: 0.9560714960098267:  83%|████████▎ | 5399/6470 [1:05:03<13:03,  1.37it/s]
1|5400|Loss: 0.8535621762275696:  83%|████████▎ | 5399/6470 [1:05:03<13:03,  1.37it/s]
1|5400|Loss: 0.8535621762275696:  83%|████████▎ | 5400/6470 [1:05:04<12:36,  1.41it/s]
1|5401|Loss: 0.8945584297180176:  83%|████████▎ | 5400/6470 [1:05:04<12:36,  1.41it/s]
1|5401|Loss: 0.8945584297180176:  83%|████████▎ | 5401/6470 [1:05:04<12:07,  1.47it/s]
1|5402|Loss: 0.8821956515312195:  83%|████████▎ | 5401/6470 [1:05:05<12:07,  1.47it/s]
1|5402|Loss: 0.8821956515312195:  83%|████████▎ | 5402/6470 [1:05:05<12:15,  1.45it/s]
1|5403|Loss: 0.9223924279212952:  83%|████████▎ | 5402/6470 [1:05:05<12:15,  1.45it/s]
1|5403|Loss: 0.9223924279212952:  84%|████████▎ | 5403/6470 [1:05:06<12:06,  1.47it/s]
1|5404|Loss: 0.8644195795059204:  84%|████████▎ | 5403/6470 [1:05:06<12:06,  1.47it/s]
1|5404|Loss: 0.8644195795059204:  84%|████████▎ | 5404/6470 [1:05:06<12:06,  1.47it/s]
1|5405|Loss: 0.9035109877586365:  84%|████████▎ | 5404/6470 [1:05:07<12:06,  1.47it/s]
1|5405|Loss: 0.9035109877586365:  84%|████████▎ | 5405/6470 [1:05:07<11:33,  1.54it/s]
1|5406|Loss: 0.7244824767112732:  84%|████████▎ | 5405/6470 [1:05:07<11:33,  1.54it/s]
1|5406|Loss: 0.7244824767112732:  84%|████████▎ | 5406/6470 [1:05:08<12:13,  1.45it/s]
1|5407|Loss: 1.0069245100021362:  84%|████████▎ | 5406/6470 [1:05:08<12:13,  1.45it/s]
1|5407|Loss: 1.0069245100021362:  84%|████████▎ | 5407/6470 [1:05:09<13:03,  1.36it/s]
1|5408|Loss: 1.0140111446380615:  84%|████████▎ | 5407/6470 [1:05:09<13:03,  1.36it/s]
1|5408|Loss: 1.0140111446380615:  84%|████████▎ | 5408/6470 [1:05:09<13:15,  1.33it/s]
1|5409|Loss: 0.9167605638504028:  84%|████████▎ | 5408/6470 [1:05:10<13:15,  1.33it/s]
1|5409|Loss: 0.9167605638504028:  84%|████████▎ | 5409/6470 [1:05:10<12:25,  1.42it/s]
1|5410|Loss: 0.7566025257110596:  84%|████████▎ | 5409/6470 [1:05:10<12:25,  1.42it/s]
1|5410|Loss: 0.7566025257110596:  84%|████████▎ | 5410/6470 [1:05:11<12:00,  1.47it/s]
1|5411|Loss: 0.9085726737976074:  84%|████████▎ | 5410/6470 [1:05:11<12:00,  1.47it/s]
1|5411|Loss: 0.9085726737976074:  84%|████████▎ | 5411/6470 [1:05:11<12:12,  1.45it/s]
1|5412|Loss: 0.8815089464187622:  84%|████████▎ | 5411/6470 [1:05:12<12:12,  1.45it/s]
1|5412|Loss: 0.8815089464187622:  84%|████████▎ | 5412/6470 [1:05:12<12:28,  1.41it/s]
1|5413|Loss: 0.9059550166130066:  84%|████████▎ | 5412/6470 [1:05:12<12:28,  1.41it/s]
1|5413|Loss: 0.9059550166130066:  84%|████████▎ | 5413/6470 [1:05:13<12:09,  1.45it/s]
1|5414|Loss: 0.8895381093025208:  84%|████████▎ | 5413/6470 [1:05:13<12:09,  1.45it/s]
1|5414|Loss: 0.8895381093025208:  84%|████████▎ | 5414/6470 [1:05:13<11:50,  1.49it/s]
1|5415|Loss: 0.851361095905304:  84%|████████▎ | 5414/6470 [1:05:14<11:50,  1.49it/s] 
1|5415|Loss: 0.851361095905304:  84%|████████▎ | 5415/6470 [1:05:14<11:11,  1.57it/s]
1|5416|Loss: 0.8930613994598389:  84%|████████▎ | 5415/6470 [1:05:14<11:11,  1.57it/s]
1|5416|Loss: 0.8930613994598389:  84%|████████▎ | 5416/6470 [1:05:14<09:54,  1.77it/s]
1|5417|Loss: 0.8899768590927124:  84%|████████▎ | 5416/6470 [1:05:15<09:54,  1.77it/s]
1|5417|Loss: 0.8899768590927124:  84%|████████▎ | 5417/6470 [1:05:15<10:00,  1.75it/s]
1|5418|Loss: 0.8629910349845886:  84%|████████▎ | 5417/6470 [1:05:15<10:00,  1.75it/s]
1|5418|Loss: 0.8629910349845886:  84%|████████▎ | 5418/6470 [1:05:16<10:59,  1.59it/s]
1|5419|Loss: 0.7364707589149475:  84%|████████▎ | 5418/6470 [1:05:16<10:59,  1.59it/s]
1|5419|Loss: 0.7364707589149475:  84%|████████▍ | 5419/6470 [1:05:16<11:09,  1.57it/s]
1|5420|Loss: 0.845704972743988:  84%|████████▍ | 5419/6470 [1:05:17<11:09,  1.57it/s] 
1|5420|Loss: 0.845704972743988:  84%|████████▍ | 5420/6470 [1:05:17<11:38,  1.50it/s]
1|5421|Loss: 0.9048182964324951:  84%|████████▍ | 5420/6470 [1:05:17<11:38,  1.50it/s]
1|5421|Loss: 0.9048182964324951:  84%|████████▍ | 5421/6470 [1:05:18<11:32,  1.51it/s]
1|5422|Loss: 0.9917923808097839:  84%|████████▍ | 5421/6470 [1:05:18<11:32,  1.51it/s]
1|5422|Loss: 0.9917923808097839:  84%|████████▍ | 5422/6470 [1:05:18<11:23,  1.53it/s]
1|5423|Loss: 0.8408410549163818:  84%|████████▍ | 5422/6470 [1:05:19<11:23,  1.53it/s]
1|5423|Loss: 0.8408410549163818:  84%|████████▍ | 5423/6470 [1:05:19<11:34,  1.51it/s]
1|5424|Loss: 0.8810274004936218:  84%|████████▍ | 5423/6470 [1:05:19<11:34,  1.51it/s]
1|5424|Loss: 0.8810274004936218:  84%|████████▍ | 5424/6470 [1:05:20<11:53,  1.47it/s]
1|5425|Loss: 0.8922790884971619:  84%|████████▍ | 5424/6470 [1:05:20<11:53,  1.47it/s]
1|5425|Loss: 0.8922790884971619:  84%|████████▍ | 5425/6470 [1:05:20<11:17,  1.54it/s]
1|5426|Loss: 0.9564359784126282:  84%|████████▍ | 5425/6470 [1:05:21<11:17,  1.54it/s]
1|5426|Loss: 0.9564359784126282:  84%|████████▍ | 5426/6470 [1:05:21<10:45,  1.62it/s]
1|5427|Loss: 0.8736500144004822:  84%|████████▍ | 5426/6470 [1:05:21<10:45,  1.62it/s]
1|5427|Loss: 0.8736500144004822:  84%|████████▍ | 5427/6470 [1:05:22<11:01,  1.58it/s]
1|5428|Loss: 0.9457195997238159:  84%|████████▍ | 5427/6470 [1:05:22<11:01,  1.58it/s]
1|5428|Loss: 0.9457195997238159:  84%|████████▍ | 5428/6470 [1:05:22<10:33,  1.64it/s]
1|5429|Loss: 0.8440502882003784:  84%|████████▍ | 5428/6470 [1:05:22<10:33,  1.64it/s]
1|5429|Loss: 0.8440502882003784:  84%|████████▍ | 5429/6470 [1:05:23<10:50,  1.60it/s]
1|5430|Loss: 0.8301305174827576:  84%|████████▍ | 5429/6470 [1:05:23<10:50,  1.60it/s]
1|5430|Loss: 0.8301305174827576:  84%|████████▍ | 5430/6470 [1:05:23<11:35,  1.50it/s]
1|5431|Loss: 0.8055220246315002:  84%|████████▍ | 5430/6470 [1:05:24<11:35,  1.50it/s]
1|5431|Loss: 0.8055220246315002:  84%|████████▍ | 5431/6470 [1:05:24<11:38,  1.49it/s]
1|5432|Loss: 1.043031930923462:  84%|████████▍ | 5431/6470 [1:05:25<11:38,  1.49it/s] 
1|5432|Loss: 1.043031930923462:  84%|████████▍ | 5432/6470 [1:05:25<12:21,  1.40it/s]
1|5433|Loss: 0.8674148917198181:  84%|████████▍ | 5432/6470 [1:05:25<12:21,  1.40it/s]
1|5433|Loss: 0.8674148917198181:  84%|████████▍ | 5433/6470 [1:05:26<13:12,  1.31it/s]
1|5434|Loss: 0.8322553634643555:  84%|████████▍ | 5433/6470 [1:05:26<13:12,  1.31it/s]
1|5434|Loss: 0.8322553634643555:  84%|████████▍ | 5434/6470 [1:05:27<13:19,  1.30it/s]
1|5435|Loss: 0.9068093299865723:  84%|████████▍ | 5434/6470 [1:05:27<13:19,  1.30it/s]
1|5435|Loss: 0.9068093299865723:  84%|████████▍ | 5435/6470 [1:05:28<13:55,  1.24it/s]
1|5436|Loss: 0.8511094450950623:  84%|████████▍ | 5435/6470 [1:05:28<13:55,  1.24it/s]
1|5436|Loss: 0.8511094450950623:  84%|████████▍ | 5436/6470 [1:05:28<13:06,  1.31it/s]
1|5437|Loss: 0.919765055179596:  84%|████████▍ | 5436/6470 [1:05:29<13:06,  1.31it/s] 
1|5437|Loss: 0.919765055179596:  84%|████████▍ | 5437/6470 [1:05:29<12:52,  1.34it/s]
1|5438|Loss: 0.919460117816925:  84%|████████▍ | 5437/6470 [1:05:29<12:52,  1.34it/s]
1|5438|Loss: 0.919460117816925:  84%|████████▍ | 5438/6470 [1:05:30<12:42,  1.35it/s]
1|5439|Loss: 0.9052947759628296:  84%|████████▍ | 5438/6470 [1:05:30<12:42,  1.35it/s]
1|5439|Loss: 0.9052947759628296:  84%|████████▍ | 5439/6470 [1:05:30<12:10,  1.41it/s]
1|5440|Loss: 0.9502447843551636:  84%|████████▍ | 5439/6470 [1:05:31<12:10,  1.41it/s]
1|5440|Loss: 0.9502447843551636:  84%|████████▍ | 5440/6470 [1:05:31<12:09,  1.41it/s]
1|5441|Loss: 0.9513523578643799:  84%|████████▍ | 5440/6470 [1:05:31<12:09,  1.41it/s]
1|5441|Loss: 0.9513523578643799:  84%|████████▍ | 5441/6470 [1:05:32<12:17,  1.40it/s]
1|5442|Loss: 0.9896445870399475:  84%|████████▍ | 5441/6470 [1:05:32<12:17,  1.40it/s]
1|5442|Loss: 0.9896445870399475:  84%|████████▍ | 5442/6470 [1:05:33<12:40,  1.35it/s]
1|5443|Loss: 0.8005860447883606:  84%|████████▍ | 5442/6470 [1:05:33<12:40,  1.35it/s]
1|5443|Loss: 0.8005860447883606:  84%|████████▍ | 5443/6470 [1:05:33<13:06,  1.31it/s]
1|5444|Loss: 0.9471511840820312:  84%|████████▍ | 5443/6470 [1:05:34<13:06,  1.31it/s]
1|5444|Loss: 0.9471511840820312:  84%|████████▍ | 5444/6470 [1:05:34<12:33,  1.36it/s]
1|5445|Loss: 0.8575320243835449:  84%|████████▍ | 5444/6470 [1:05:34<12:33,  1.36it/s]
1|5445|Loss: 0.8575320243835449:  84%|████████▍ | 5445/6470 [1:05:35<12:46,  1.34it/s]
1|5446|Loss: 0.9036473631858826:  84%|████████▍ | 5445/6470 [1:05:35<12:46,  1.34it/s]
1|5446|Loss: 0.9036473631858826:  84%|████████▍ | 5446/6470 [1:05:36<13:09,  1.30it/s]
1|5447|Loss: 0.904901921749115:  84%|████████▍ | 5446/6470 [1:05:36<13:09,  1.30it/s] 
1|5447|Loss: 0.904901921749115:  84%|████████▍ | 5447/6470 [1:05:36<13:09,  1.30it/s]
1|5448|Loss: 0.9408940076828003:  84%|████████▍ | 5447/6470 [1:05:37<13:09,  1.30it/s]
1|5448|Loss: 0.9408940076828003:  84%|████████▍ | 5448/6470 [1:05:37<13:24,  1.27it/s]
1|5449|Loss: 0.7613483667373657:  84%|████████▍ | 5448/6470 [1:05:38<13:24,  1.27it/s]
1|5449|Loss: 0.7613483667373657:  84%|████████▍ | 5449/6470 [1:05:38<13:58,  1.22it/s]
1|5450|Loss: 0.7915681004524231:  84%|████████▍ | 5449/6470 [1:05:39<13:58,  1.22it/s]
1|5450|Loss: 0.7915681004524231:  84%|████████▍ | 5450/6470 [1:05:39<14:18,  1.19it/s]
1|5451|Loss: 0.8890179395675659:  84%|████████▍ | 5450/6470 [1:05:39<14:18,  1.19it/s]
1|5451|Loss: 0.8890179395675659:  84%|████████▍ | 5451/6470 [1:05:40<13:54,  1.22it/s]
1|5452|Loss: 0.8555928468704224:  84%|████████▍ | 5451/6470 [1:05:40<13:54,  1.22it/s]
1|5452|Loss: 0.8555928468704224:  84%|████████▍ | 5452/6470 [1:05:40<12:52,  1.32it/s]
1|5453|Loss: 0.7620157599449158:  84%|████████▍ | 5452/6470 [1:05:41<12:52,  1.32it/s]
1|5453|Loss: 0.7620157599449158:  84%|████████▍ | 5453/6470 [1:05:41<12:10,  1.39it/s]
1|5454|Loss: 1.0953402519226074:  84%|████████▍ | 5453/6470 [1:05:41<12:10,  1.39it/s]
1|5454|Loss: 1.0953402519226074:  84%|████████▍ | 5454/6470 [1:05:42<11:30,  1.47it/s]
1|5455|Loss: 0.941062867641449:  84%|████████▍ | 5454/6470 [1:05:42<11:30,  1.47it/s] 
1|5455|Loss: 0.941062867641449:  84%|████████▍ | 5455/6470 [1:05:42<11:37,  1.46it/s]
1|5456|Loss: 0.8465799689292908:  84%|████████▍ | 5455/6470 [1:05:43<11:37,  1.46it/s]
1|5456|Loss: 0.8465799689292908:  84%|████████▍ | 5456/6470 [1:05:43<11:29,  1.47it/s]
1|5457|Loss: 0.7085912823677063:  84%|████████▍ | 5456/6470 [1:05:43<11:29,  1.47it/s]
1|5457|Loss: 0.7085912823677063:  84%|████████▍ | 5457/6470 [1:05:44<11:11,  1.51it/s]
1|5458|Loss: 0.881615161895752:  84%|████████▍ | 5457/6470 [1:05:44<11:11,  1.51it/s] 
1|5458|Loss: 0.881615161895752:  84%|████████▍ | 5458/6470 [1:05:44<10:57,  1.54it/s]
1|5459|Loss: 0.8628208637237549:  84%|████████▍ | 5458/6470 [1:05:45<10:57,  1.54it/s]
1|5459|Loss: 0.8628208637237549:  84%|████████▍ | 5459/6470 [1:05:45<11:39,  1.45it/s]
1|5460|Loss: 0.947372317314148:  84%|████████▍ | 5459/6470 [1:05:45<11:39,  1.45it/s] 
1|5460|Loss: 0.947372317314148:  84%|████████▍ | 5460/6470 [1:05:46<11:59,  1.40it/s]
1|5461|Loss: 0.9390504956245422:  84%|████████▍ | 5460/6470 [1:05:46<11:59,  1.40it/s]
1|5461|Loss: 0.9390504956245422:  84%|████████▍ | 5461/6470 [1:05:46<11:44,  1.43it/s]
1|5462|Loss: 0.7950292825698853:  84%|████████▍ | 5461/6470 [1:05:47<11:44,  1.43it/s]
1|5462|Loss: 0.7950292825698853:  84%|████████▍ | 5462/6470 [1:05:47<11:21,  1.48it/s]
1|5463|Loss: 0.8077564835548401:  84%|████████▍ | 5462/6470 [1:05:47<11:21,  1.48it/s]
1|5463|Loss: 0.8077564835548401:  84%|████████▍ | 5463/6470 [1:05:48<11:48,  1.42it/s]
1|5464|Loss: 0.950076699256897:  84%|████████▍ | 5463/6470 [1:05:48<11:48,  1.42it/s] 
1|5464|Loss: 0.950076699256897:  84%|████████▍ | 5464/6470 [1:05:48<11:08,  1.50it/s]
1|5465|Loss: 0.7386429905891418:  84%|████████▍ | 5464/6470 [1:05:49<11:08,  1.50it/s]
1|5465|Loss: 0.7386429905891418:  84%|████████▍ | 5465/6470 [1:05:49<11:45,  1.42it/s]
1|5466|Loss: 0.9905351400375366:  84%|████████▍ | 5465/6470 [1:05:50<11:45,  1.42it/s]
1|5466|Loss: 0.9905351400375366:  84%|████████▍ | 5466/6470 [1:05:50<11:11,  1.49it/s]
1|5467|Loss: 0.8000983595848083:  84%|████████▍ | 5466/6470 [1:05:50<11:11,  1.49it/s]
1|5467|Loss: 0.8000983595848083:  84%|████████▍ | 5467/6470 [1:05:51<11:59,  1.39it/s]
1|5468|Loss: 0.9027941226959229:  84%|████████▍ | 5467/6470 [1:05:51<11:59,  1.39it/s]
1|5468|Loss: 0.9027941226959229:  85%|████████▍ | 5468/6470 [1:05:51<12:07,  1.38it/s]
1|5469|Loss: 0.8662804961204529:  85%|████████▍ | 5468/6470 [1:05:52<12:07,  1.38it/s]
1|5469|Loss: 0.8662804961204529:  85%|████████▍ | 5469/6470 [1:05:52<11:06,  1.50it/s]
1|5470|Loss: 0.8508483171463013:  85%|████████▍ | 5469/6470 [1:05:52<11:06,  1.50it/s]
1|5470|Loss: 0.8508483171463013:  85%|████████▍ | 5470/6470 [1:05:53<11:46,  1.42it/s]
1|5471|Loss: 0.8532856702804565:  85%|████████▍ | 5470/6470 [1:05:53<11:46,  1.42it/s]
1|5471|Loss: 0.8532856702804565:  85%|████████▍ | 5471/6470 [1:05:54<12:37,  1.32it/s]
1|5472|Loss: 0.7773817777633667:  85%|████████▍ | 5471/6470 [1:05:54<12:37,  1.32it/s]
1|5472|Loss: 0.7773817777633667:  85%|████████▍ | 5472/6470 [1:05:54<12:52,  1.29it/s]
1|5473|Loss: 0.9066964983940125:  85%|████████▍ | 5472/6470 [1:05:55<12:52,  1.29it/s]
1|5473|Loss: 0.9066964983940125:  85%|████████▍ | 5473/6470 [1:05:55<11:36,  1.43it/s]
1|5474|Loss: 0.7762141227722168:  85%|████████▍ | 5473/6470 [1:05:55<11:36,  1.43it/s]
1|5474|Loss: 0.7762141227722168:  85%|████████▍ | 5474/6470 [1:05:56<11:30,  1.44it/s]
1|5475|Loss: 0.8224552869796753:  85%|████████▍ | 5474/6470 [1:05:56<11:30,  1.44it/s]
1|5475|Loss: 0.8224552869796753:  85%|████████▍ | 5475/6470 [1:05:56<11:12,  1.48it/s]
1|5476|Loss: 0.8202498555183411:  85%|████████▍ | 5475/6470 [1:05:57<11:12,  1.48it/s]
1|5476|Loss: 0.8202498555183411:  85%|████████▍ | 5476/6470 [1:05:57<11:11,  1.48it/s]
1|5477|Loss: 0.8210176825523376:  85%|████████▍ | 5476/6470 [1:05:57<11:11,  1.48it/s]
1|5477|Loss: 0.8210176825523376:  85%|████████▍ | 5477/6470 [1:05:58<11:48,  1.40it/s]
1|5478|Loss: 0.906923234462738:  85%|████████▍ | 5477/6470 [1:05:58<11:48,  1.40it/s] 
1|5478|Loss: 0.906923234462738:  85%|████████▍ | 5478/6470 [1:05:58<12:16,  1.35it/s]
1|5479|Loss: 0.7869837284088135:  85%|████████▍ | 5478/6470 [1:05:59<12:16,  1.35it/s]
1|5479|Loss: 0.7869837284088135:  85%|████████▍ | 5479/6470 [1:05:59<11:56,  1.38it/s]
1|5480|Loss: 0.8051811456680298:  85%|████████▍ | 5479/6470 [1:06:00<11:56,  1.38it/s]
1|5480|Loss: 0.8051811456680298:  85%|████████▍ | 5480/6470 [1:06:00<12:09,  1.36it/s]
1|5481|Loss: 0.918938934803009:  85%|████████▍ | 5480/6470 [1:06:00<12:09,  1.36it/s] 
1|5481|Loss: 0.918938934803009:  85%|████████▍ | 5481/6470 [1:06:01<12:42,  1.30it/s]
1|5482|Loss: 0.872279942035675:  85%|████████▍ | 5481/6470 [1:06:01<12:42,  1.30it/s]
1|5482|Loss: 0.872279942035675:  85%|████████▍ | 5482/6470 [1:06:02<12:42,  1.30it/s]
1|5483|Loss: 0.8597252368927002:  85%|████████▍ | 5482/6470 [1:06:02<12:42,  1.30it/s]
1|5483|Loss: 0.8597252368927002:  85%|████████▍ | 5483/6470 [1:06:02<11:54,  1.38it/s]
1|5484|Loss: 0.8338554501533508:  85%|████████▍ | 5483/6470 [1:06:03<11:54,  1.38it/s]
1|5484|Loss: 0.8338554501533508:  85%|████████▍ | 5484/6470 [1:06:03<12:11,  1.35it/s]
1|5485|Loss: 0.8980684280395508:  85%|████████▍ | 5484/6470 [1:06:03<12:11,  1.35it/s]
1|5485|Loss: 0.8980684280395508:  85%|████████▍ | 5485/6470 [1:06:04<12:31,  1.31it/s]
1|5486|Loss: 1.0003001689910889:  85%|████████▍ | 5485/6470 [1:06:04<12:31,  1.31it/s]
1|5486|Loss: 1.0003001689910889:  85%|████████▍ | 5486/6470 [1:06:05<12:27,  1.32it/s]
1|5487|Loss: 0.9451776742935181:  85%|████████▍ | 5486/6470 [1:06:05<12:27,  1.32it/s]
1|5487|Loss: 0.9451776742935181:  85%|████████▍ | 5487/6470 [1:06:05<11:52,  1.38it/s]
1|5488|Loss: 0.9122303128242493:  85%|████████▍ | 5487/6470 [1:06:06<11:52,  1.38it/s]
1|5488|Loss: 0.9122303128242493:  85%|████████▍ | 5488/6470 [1:06:06<12:12,  1.34it/s]
1|5489|Loss: 0.7434325814247131:  85%|████████▍ | 5488/6470 [1:06:06<12:12,  1.34it/s]
1|5489|Loss: 0.7434325814247131:  85%|████████▍ | 5489/6470 [1:06:07<11:54,  1.37it/s]
1|5490|Loss: 0.8459274768829346:  85%|████████▍ | 5489/6470 [1:06:07<11:54,  1.37it/s]
1|5490|Loss: 0.8459274768829346:  85%|████████▍ | 5490/6470 [1:06:07<11:01,  1.48it/s]
1|5491|Loss: 0.9397939443588257:  85%|████████▍ | 5490/6470 [1:06:08<11:01,  1.48it/s]
1|5491|Loss: 0.9397939443588257:  85%|████████▍ | 5491/6470 [1:06:08<11:10,  1.46it/s]
1|5492|Loss: 0.7817445993423462:  85%|████████▍ | 5491/6470 [1:06:08<11:10,  1.46it/s]
1|5492|Loss: 0.7817445993423462:  85%|████████▍ | 5492/6470 [1:06:09<11:17,  1.44it/s]
1|5493|Loss: 0.8815966248512268:  85%|████████▍ | 5492/6470 [1:06:09<11:17,  1.44it/s]
1|5493|Loss: 0.8815966248512268:  85%|████████▍ | 5493/6470 [1:06:09<11:29,  1.42it/s]
1|5494|Loss: 0.8482077121734619:  85%|████████▍ | 5493/6470 [1:06:10<11:29,  1.42it/s]
1|5494|Loss: 0.8482077121734619:  85%|████████▍ | 5494/6470 [1:06:10<11:50,  1.37it/s]
1|5495|Loss: 0.836046576499939:  85%|████████▍ | 5494/6470 [1:06:11<11:50,  1.37it/s] 
1|5495|Loss: 0.836046576499939:  85%|████████▍ | 5495/6470 [1:06:11<12:30,  1.30it/s]
1|5496|Loss: 0.8231578469276428:  85%|████████▍ | 5495/6470 [1:06:11<12:30,  1.30it/s]
1|5496|Loss: 0.8231578469276428:  85%|████████▍ | 5496/6470 [1:06:12<12:23,  1.31it/s]
1|5497|Loss: 0.9068213701248169:  85%|████████▍ | 5496/6470 [1:06:12<12:23,  1.31it/s]
1|5497|Loss: 0.9068213701248169:  85%|████████▍ | 5497/6470 [1:06:12<12:14,  1.32it/s]
1|5498|Loss: 0.7843447327613831:  85%|████████▍ | 5497/6470 [1:06:13<12:14,  1.32it/s]
1|5498|Loss: 0.7843447327613831:  85%|████████▍ | 5498/6470 [1:06:13<12:41,  1.28it/s]
1|5499|Loss: 0.833857536315918:  85%|████████▍ | 5498/6470 [1:06:14<12:41,  1.28it/s] 
1|5499|Loss: 0.833857536315918:  85%|████████▍ | 5499/6470 [1:06:14<12:13,  1.32it/s]
1|5500|Loss: 0.9298032522201538:  85%|████████▍ | 5499/6470 [1:06:14<12:13,  1.32it/s]
1|5500|Loss: 0.9298032522201538:  85%|████████▌ | 5500/6470 [1:06:15<12:00,  1.35it/s]
1|5501|Loss: 0.7414204478263855:  85%|████████▌ | 5500/6470 [1:06:15<12:00,  1.35it/s]
1|5501|Loss: 0.7414204478263855:  85%|████████▌ | 5501/6470 [1:06:15<11:30,  1.40it/s]
1|5502|Loss: 0.8819798827171326:  85%|████████▌ | 5501/6470 [1:06:16<11:30,  1.40it/s]
1|5502|Loss: 0.8819798827171326:  85%|████████▌ | 5502/6470 [1:06:16<11:30,  1.40it/s]
1|5503|Loss: 0.7573785185813904:  85%|████████▌ | 5502/6470 [1:06:16<11:30,  1.40it/s]
1|5503|Loss: 0.7573785185813904:  85%|████████▌ | 5503/6470 [1:06:17<11:16,  1.43it/s]
1|5504|Loss: 0.8884493112564087:  85%|████████▌ | 5503/6470 [1:06:17<11:16,  1.43it/s]
1|5504|Loss: 0.8884493112564087:  85%|████████▌ | 5504/6470 [1:06:17<10:29,  1.54it/s]
1|5505|Loss: 0.8608300089836121:  85%|████████▌ | 5504/6470 [1:06:18<10:29,  1.54it/s]
1|5505|Loss: 0.8608300089836121:  85%|████████▌ | 5505/6470 [1:06:18<09:57,  1.62it/s]
1|5506|Loss: 0.9804351925849915:  85%|████████▌ | 5505/6470 [1:06:18<09:57,  1.62it/s]
1|5506|Loss: 0.9804351925849915:  85%|████████▌ | 5506/6470 [1:06:19<10:35,  1.52it/s]
1|5507|Loss: 0.871971607208252:  85%|████████▌ | 5506/6470 [1:06:19<10:35,  1.52it/s] 
1|5507|Loss: 0.871971607208252:  85%|████████▌ | 5507/6470 [1:06:20<13:02,  1.23it/s]
1|5508|Loss: 0.8303465247154236:  85%|████████▌ | 5507/6470 [1:06:20<13:02,  1.23it/s]
1|5508|Loss: 0.8303465247154236:  85%|████████▌ | 5508/6470 [1:06:20<12:06,  1.32it/s]
1|5509|Loss: 0.9517776966094971:  85%|████████▌ | 5508/6470 [1:06:21<12:06,  1.32it/s]
1|5509|Loss: 0.9517776966094971:  85%|████████▌ | 5509/6470 [1:06:21<11:53,  1.35it/s]
1|5510|Loss: 0.7165589928627014:  85%|████████▌ | 5509/6470 [1:06:21<11:53,  1.35it/s]
1|5510|Loss: 0.7165589928627014:  85%|████████▌ | 5510/6470 [1:06:22<11:28,  1.39it/s]
1|5511|Loss: 0.9057242274284363:  85%|████████▌ | 5510/6470 [1:06:22<11:28,  1.39it/s]
1|5511|Loss: 0.9057242274284363:  85%|████████▌ | 5511/6470 [1:06:22<10:50,  1.47it/s]
1|5512|Loss: 0.8606124520301819:  85%|████████▌ | 5511/6470 [1:06:23<10:50,  1.47it/s]
1|5512|Loss: 0.8606124520301819:  85%|████████▌ | 5512/6470 [1:06:23<10:35,  1.51it/s]
1|5513|Loss: 0.7730884552001953:  85%|████████▌ | 5512/6470 [1:06:23<10:35,  1.51it/s]
1|5513|Loss: 0.7730884552001953:  85%|████████▌ | 5513/6470 [1:06:24<11:25,  1.40it/s]
1|5514|Loss: 0.9300671815872192:  85%|████████▌ | 5513/6470 [1:06:24<11:25,  1.40it/s]
1|5514|Loss: 0.9300671815872192:  85%|████████▌ | 5514/6470 [1:06:24<11:05,  1.44it/s]
1|5515|Loss: 0.9759030938148499:  85%|████████▌ | 5514/6470 [1:06:25<11:05,  1.44it/s]
1|5515|Loss: 0.9759030938148499:  85%|████████▌ | 5515/6470 [1:06:25<11:08,  1.43it/s]
1|5516|Loss: 0.8758351802825928:  85%|████████▌ | 5515/6470 [1:06:26<11:08,  1.43it/s]
1|5516|Loss: 0.8758351802825928:  85%|████████▌ | 5516/6470 [1:06:26<11:21,  1.40it/s]
1|5517|Loss: 0.9040080308914185:  85%|████████▌ | 5516/6470 [1:06:26<11:21,  1.40it/s]
1|5517|Loss: 0.9040080308914185:  85%|████████▌ | 5517/6470 [1:06:27<11:34,  1.37it/s]
1|5518|Loss: 0.8767135739326477:  85%|████████▌ | 5517/6470 [1:06:27<11:34,  1.37it/s]
1|5518|Loss: 0.8767135739326477:  85%|████████▌ | 5518/6470 [1:06:27<11:15,  1.41it/s]
1|5519|Loss: 0.7363797426223755:  85%|████████▌ | 5518/6470 [1:06:28<11:15,  1.41it/s]
1|5519|Loss: 0.7363797426223755:  85%|████████▌ | 5519/6470 [1:06:28<11:07,  1.43it/s]
1|5520|Loss: 0.8756759762763977:  85%|████████▌ | 5519/6470 [1:06:28<11:07,  1.43it/s]
1|5520|Loss: 0.8756759762763977:  85%|████████▌ | 5520/6470 [1:06:29<10:52,  1.46it/s]
1|5521|Loss: 0.9240304231643677:  85%|████████▌ | 5520/6470 [1:06:29<10:52,  1.46it/s]
1|5521|Loss: 0.9240304231643677:  85%|████████▌ | 5521/6470 [1:06:29<10:58,  1.44it/s]
1|5522|Loss: 0.8928824067115784:  85%|████████▌ | 5521/6470 [1:06:30<10:58,  1.44it/s]
1|5522|Loss: 0.8928824067115784:  85%|████████▌ | 5522/6470 [1:06:30<11:08,  1.42it/s]
1|5523|Loss: 0.9364868998527527:  85%|████████▌ | 5522/6470 [1:06:30<11:08,  1.42it/s]
1|5523|Loss: 0.9364868998527527:  85%|████████▌ | 5523/6470 [1:06:31<10:25,  1.51it/s]
1|5524|Loss: 0.984933614730835:  85%|████████▌ | 5523/6470 [1:06:31<10:25,  1.51it/s] 
1|5524|Loss: 0.984933614730835:  85%|████████▌ | 5524/6470 [1:06:31<10:08,  1.55it/s]
1|5525|Loss: 0.8413326144218445:  85%|████████▌ | 5524/6470 [1:06:32<10:08,  1.55it/s]
1|5525|Loss: 0.8413326144218445:  85%|████████▌ | 5525/6470 [1:06:32<10:54,  1.44it/s]
1|5526|Loss: 1.0405467748641968:  85%|████████▌ | 5525/6470 [1:06:32<10:54,  1.44it/s]
1|5526|Loss: 1.0405467748641968:  85%|████████▌ | 5526/6470 [1:06:33<10:52,  1.45it/s]
1|5527|Loss: 0.9532079696655273:  85%|████████▌ | 5526/6470 [1:06:33<10:52,  1.45it/s]
1|5527|Loss: 0.9532079696655273:  85%|████████▌ | 5527/6470 [1:06:33<10:30,  1.50it/s]
1|5528|Loss: 0.974187970161438:  85%|████████▌ | 5527/6470 [1:06:34<10:30,  1.50it/s] 
1|5528|Loss: 0.974187970161438:  85%|████████▌ | 5528/6470 [1:06:34<10:08,  1.55it/s]
1|5529|Loss: 0.7877610921859741:  85%|████████▌ | 5528/6470 [1:06:34<10:08,  1.55it/s]
1|5529|Loss: 0.7877610921859741:  85%|████████▌ | 5529/6470 [1:06:35<10:15,  1.53it/s]
1|5530|Loss: 0.8392806053161621:  85%|████████▌ | 5529/6470 [1:06:35<10:15,  1.53it/s]
1|5530|Loss: 0.8392806053161621:  85%|████████▌ | 5530/6470 [1:06:35<11:01,  1.42it/s]
1|5531|Loss: 0.9408966302871704:  85%|████████▌ | 5530/6470 [1:06:36<11:01,  1.42it/s]
1|5531|Loss: 0.9408966302871704:  85%|████████▌ | 5531/6470 [1:06:36<11:34,  1.35it/s]
1|5532|Loss: 0.8441802859306335:  85%|████████▌ | 5531/6470 [1:06:37<11:34,  1.35it/s]
1|5532|Loss: 0.8441802859306335:  86%|████████▌ | 5532/6470 [1:06:37<12:02,  1.30it/s]
1|5533|Loss: 0.9110702276229858:  86%|████████▌ | 5532/6470 [1:06:38<12:02,  1.30it/s]
1|5533|Loss: 0.9110702276229858:  86%|████████▌ | 5533/6470 [1:06:38<11:34,  1.35it/s]
1|5534|Loss: 0.8427850008010864:  86%|████████▌ | 5533/6470 [1:06:38<11:34,  1.35it/s]
1|5534|Loss: 0.8427850008010864:  86%|████████▌ | 5534/6470 [1:06:38<10:29,  1.49it/s]
1|5535|Loss: 0.8621774315834045:  86%|████████▌ | 5534/6470 [1:06:39<10:29,  1.49it/s]
1|5535|Loss: 0.8621774315834045:  86%|████████▌ | 5535/6470 [1:06:39<11:03,  1.41it/s]
1|5536|Loss: 0.9614159464836121:  86%|████████▌ | 5535/6470 [1:06:40<11:03,  1.41it/s]
1|5536|Loss: 0.9614159464836121:  86%|████████▌ | 5536/6470 [1:06:40<11:45,  1.32it/s]
1|5537|Loss: 0.8509452939033508:  86%|████████▌ | 5536/6470 [1:06:40<11:45,  1.32it/s]
1|5537|Loss: 0.8509452939033508:  86%|████████▌ | 5537/6470 [1:06:41<11:23,  1.37it/s]
1|5538|Loss: 1.008784532546997:  86%|████████▌ | 5537/6470 [1:06:41<11:23,  1.37it/s] 
1|5538|Loss: 1.008784532546997:  86%|████████▌ | 5538/6470 [1:06:41<11:08,  1.39it/s]
1|5539|Loss: 0.8865334987640381:  86%|████████▌ | 5538/6470 [1:06:42<11:08,  1.39it/s]
1|5539|Loss: 0.8865334987640381:  86%|████████▌ | 5539/6470 [1:06:42<10:24,  1.49it/s]
1|5540|Loss: 0.7782566547393799:  86%|████████▌ | 5539/6470 [1:06:42<10:24,  1.49it/s]
1|5540|Loss: 0.7782566547393799:  86%|████████▌ | 5540/6470 [1:06:42<09:54,  1.57it/s]
1|5541|Loss: 0.8216784000396729:  86%|████████▌ | 5540/6470 [1:06:43<09:54,  1.57it/s]
1|5541|Loss: 0.8216784000396729:  86%|████████▌ | 5541/6470 [1:06:43<10:11,  1.52it/s]
1|5542|Loss: 0.9268238544464111:  86%|████████▌ | 5541/6470 [1:06:44<10:11,  1.52it/s]
1|5542|Loss: 0.9268238544464111:  86%|████████▌ | 5542/6470 [1:06:44<10:29,  1.47it/s]
1|5543|Loss: 0.7841101288795471:  86%|████████▌ | 5542/6470 [1:06:44<10:29,  1.47it/s]
1|5543|Loss: 0.7841101288795471:  86%|████████▌ | 5543/6470 [1:06:45<11:06,  1.39it/s]
1|5544|Loss: 0.8845157027244568:  86%|████████▌ | 5543/6470 [1:06:45<11:06,  1.39it/s]
1|5544|Loss: 0.8845157027244568:  86%|████████▌ | 5544/6470 [1:06:45<10:30,  1.47it/s]
1|5545|Loss: 0.8976318836212158:  86%|████████▌ | 5544/6470 [1:06:46<10:30,  1.47it/s]
1|5545|Loss: 0.8976318836212158:  86%|████████▌ | 5545/6470 [1:06:46<10:35,  1.46it/s]
1|5546|Loss: 0.8864811062812805:  86%|████████▌ | 5545/6470 [1:06:46<10:35,  1.46it/s]
1|5546|Loss: 0.8864811062812805:  86%|████████▌ | 5546/6470 [1:06:47<11:02,  1.39it/s]
1|5547|Loss: 0.8473713994026184:  86%|████████▌ | 5546/6470 [1:06:47<11:02,  1.39it/s]
1|5547|Loss: 0.8473713994026184:  86%|████████▌ | 5547/6470 [1:06:47<10:52,  1.41it/s]
1|5548|Loss: 0.8174857497215271:  86%|████████▌ | 5547/6470 [1:06:48<10:52,  1.41it/s]
1|5548|Loss: 0.8174857497215271:  86%|████████▌ | 5548/6470 [1:06:48<11:26,  1.34it/s]
1|5549|Loss: 0.7931634783744812:  86%|████████▌ | 5548/6470 [1:06:49<11:26,  1.34it/s]
1|5549|Loss: 0.7931634783744812:  86%|████████▌ | 5549/6470 [1:06:49<11:08,  1.38it/s]
1|5550|Loss: 0.7734025120735168:  86%|████████▌ | 5549/6470 [1:06:49<11:08,  1.38it/s]
1|5550|Loss: 0.7734025120735168:  86%|████████▌ | 5550/6470 [1:06:50<10:37,  1.44it/s]
1|5551|Loss: 0.9681360721588135:  86%|████████▌ | 5550/6470 [1:06:50<10:37,  1.44it/s]
1|5551|Loss: 0.9681360721588135:  86%|████████▌ | 5551/6470 [1:06:50<10:36,  1.44it/s]
1|5552|Loss: 0.8736963868141174:  86%|████████▌ | 5551/6470 [1:06:51<10:36,  1.44it/s]
1|5552|Loss: 0.8736963868141174:  86%|████████▌ | 5552/6470 [1:06:51<09:44,  1.57it/s]
1|5553|Loss: 0.9825080037117004:  86%|████████▌ | 5552/6470 [1:06:51<09:44,  1.57it/s]
1|5553|Loss: 0.9825080037117004:  86%|████████▌ | 5553/6470 [1:06:51<09:30,  1.61it/s]
1|5554|Loss: 0.7674379348754883:  86%|████████▌ | 5553/6470 [1:06:52<09:30,  1.61it/s]
1|5554|Loss: 0.7674379348754883:  86%|████████▌ | 5554/6470 [1:06:52<08:38,  1.77it/s]
1|5555|Loss: 0.7449106574058533:  86%|████████▌ | 5554/6470 [1:06:52<08:38,  1.77it/s]
1|5555|Loss: 0.7449106574058533:  86%|████████▌ | 5555/6470 [1:06:52<09:05,  1.68it/s]
1|5556|Loss: 0.9048666954040527:  86%|████████▌ | 5555/6470 [1:06:53<09:05,  1.68it/s]
1|5556|Loss: 0.9048666954040527:  86%|████████▌ | 5556/6470 [1:06:53<09:46,  1.56it/s]
1|5557|Loss: 0.9225147366523743:  86%|████████▌ | 5556/6470 [1:06:54<09:46,  1.56it/s]
1|5557|Loss: 0.9225147366523743:  86%|████████▌ | 5557/6470 [1:06:54<09:52,  1.54it/s]
1|5558|Loss: 0.8076823949813843:  86%|████████▌ | 5557/6470 [1:06:54<09:52,  1.54it/s]
1|5558|Loss: 0.8076823949813843:  86%|████████▌ | 5558/6470 [1:06:54<09:29,  1.60it/s]
1|5559|Loss: 0.8744947910308838:  86%|████████▌ | 5558/6470 [1:06:55<09:29,  1.60it/s]
1|5559|Loss: 0.8744947910308838:  86%|████████▌ | 5559/6470 [1:06:55<10:07,  1.50it/s]
1|5560|Loss: 0.8146591186523438:  86%|████████▌ | 5559/6470 [1:06:56<10:07,  1.50it/s]
1|5560|Loss: 0.8146591186523438:  86%|████████▌ | 5560/6470 [1:06:56<10:24,  1.46it/s]
1|5561|Loss: 0.8188840746879578:  86%|████████▌ | 5560/6470 [1:06:56<10:24,  1.46it/s]
1|5561|Loss: 0.8188840746879578:  86%|████████▌ | 5561/6470 [1:06:57<10:39,  1.42it/s]
1|5562|Loss: 0.9025527834892273:  86%|████████▌ | 5561/6470 [1:06:57<10:39,  1.42it/s]
1|5562|Loss: 0.9025527834892273:  86%|████████▌ | 5562/6470 [1:06:57<10:42,  1.41it/s]
1|5563|Loss: 0.9202238917350769:  86%|████████▌ | 5562/6470 [1:06:58<10:42,  1.41it/s]
1|5563|Loss: 0.9202238917350769:  86%|████████▌ | 5563/6470 [1:06:58<10:14,  1.48it/s]
1|5564|Loss: 0.8876117467880249:  86%|████████▌ | 5563/6470 [1:06:58<10:14,  1.48it/s]
1|5564|Loss: 0.8876117467880249:  86%|████████▌ | 5564/6470 [1:06:59<11:01,  1.37it/s]
1|5565|Loss: 0.8414136171340942:  86%|████████▌ | 5564/6470 [1:06:59<11:01,  1.37it/s]
1|5565|Loss: 0.8414136171340942:  86%|████████▌ | 5565/6470 [1:07:00<10:48,  1.40it/s]
1|5566|Loss: 0.9661698937416077:  86%|████████▌ | 5565/6470 [1:07:00<10:48,  1.40it/s]
1|5566|Loss: 0.9661698937416077:  86%|████████▌ | 5566/6470 [1:07:00<11:02,  1.37it/s]
1|5567|Loss: 0.828857958316803:  86%|████████▌ | 5566/6470 [1:07:01<11:02,  1.37it/s] 
1|5567|Loss: 0.828857958316803:  86%|████████▌ | 5567/6470 [1:07:01<10:12,  1.47it/s]
1|5568|Loss: 0.9177607297897339:  86%|████████▌ | 5567/6470 [1:07:01<10:12,  1.47it/s]
1|5568|Loss: 0.9177607297897339:  86%|████████▌ | 5568/6470 [1:07:02<09:52,  1.52it/s]
1|5569|Loss: 0.9642060399055481:  86%|████████▌ | 5568/6470 [1:07:02<09:52,  1.52it/s]
1|5569|Loss: 0.9642060399055481:  86%|████████▌ | 5569/6470 [1:07:02<10:01,  1.50it/s]
1|5570|Loss: 0.9808048009872437:  86%|████████▌ | 5569/6470 [1:07:03<10:01,  1.50it/s]
1|5570|Loss: 0.9808048009872437:  86%|████████▌ | 5570/6470 [1:07:03<09:42,  1.55it/s]
1|5571|Loss: 0.8126134872436523:  86%|████████▌ | 5570/6470 [1:07:03<09:42,  1.55it/s]
1|5571|Loss: 0.8126134872436523:  86%|████████▌ | 5571/6470 [1:07:04<09:59,  1.50it/s]
1|5572|Loss: 0.8366894125938416:  86%|████████▌ | 5571/6470 [1:07:04<09:59,  1.50it/s]
1|5572|Loss: 0.8366894125938416:  86%|████████▌ | 5572/6470 [1:07:04<10:18,  1.45it/s]
1|5573|Loss: 0.9490711688995361:  86%|████████▌ | 5572/6470 [1:07:05<10:18,  1.45it/s]
1|5573|Loss: 0.9490711688995361:  86%|████████▌ | 5573/6470 [1:07:05<10:09,  1.47it/s]
1|5574|Loss: 0.8935405611991882:  86%|████████▌ | 5573/6470 [1:07:05<10:09,  1.47it/s]
1|5574|Loss: 0.8935405611991882:  86%|████████▌ | 5574/6470 [1:07:06<10:11,  1.46it/s]
1|5575|Loss: 0.8368317484855652:  86%|████████▌ | 5574/6470 [1:07:06<10:11,  1.46it/s]
1|5575|Loss: 0.8368317484855652:  86%|████████▌ | 5575/6470 [1:07:06<09:34,  1.56it/s]
1|5576|Loss: 0.8860600590705872:  86%|████████▌ | 5575/6470 [1:07:06<09:34,  1.56it/s]
1|5576|Loss: 0.8860600590705872:  86%|████████▌ | 5576/6470 [1:07:07<09:27,  1.58it/s]
1|5577|Loss: 0.9750069975852966:  86%|████████▌ | 5576/6470 [1:07:07<09:27,  1.58it/s]
1|5577|Loss: 0.9750069975852966:  86%|████████▌ | 5577/6470 [1:07:07<09:03,  1.64it/s]
1|5578|Loss: 0.6594671607017517:  86%|████████▌ | 5577/6470 [1:07:08<09:03,  1.64it/s]
1|5578|Loss: 0.6594671607017517:  86%|████████▌ | 5578/6470 [1:07:08<09:05,  1.64it/s]
1|5579|Loss: 0.9085612297058105:  86%|████████▌ | 5578/6470 [1:07:08<09:05,  1.64it/s]
1|5579|Loss: 0.9085612297058105:  86%|████████▌ | 5579/6470 [1:07:09<09:34,  1.55it/s]
1|5580|Loss: 0.7357459664344788:  86%|████████▌ | 5579/6470 [1:07:09<09:34,  1.55it/s]
1|5580|Loss: 0.7357459664344788:  86%|████████▌ | 5580/6470 [1:07:09<09:49,  1.51it/s]
1|5581|Loss: 0.9268833994865417:  86%|████████▌ | 5580/6470 [1:07:10<09:49,  1.51it/s]
1|5581|Loss: 0.9268833994865417:  86%|████████▋ | 5581/6470 [1:07:10<09:45,  1.52it/s]
1|5582|Loss: 0.8817072510719299:  86%|████████▋ | 5581/6470 [1:07:10<09:45,  1.52it/s]
1|5582|Loss: 0.8817072510719299:  86%|████████▋ | 5582/6470 [1:07:11<09:57,  1.49it/s]
1|5583|Loss: 0.7842456102371216:  86%|████████▋ | 5582/6470 [1:07:11<09:57,  1.49it/s]
1|5583|Loss: 0.7842456102371216:  86%|████████▋ | 5583/6470 [1:07:12<10:32,  1.40it/s]
1|5584|Loss: 0.8678653836250305:  86%|████████▋ | 5583/6470 [1:07:12<10:32,  1.40it/s]
1|5584|Loss: 0.8678653836250305:  86%|████████▋ | 5584/6470 [1:07:12<11:08,  1.32it/s]
1|5585|Loss: 0.9239323735237122:  86%|████████▋ | 5584/6470 [1:07:13<11:08,  1.32it/s]
1|5585|Loss: 0.9239323735237122:  86%|████████▋ | 5585/6470 [1:07:13<11:14,  1.31it/s]
1|5586|Loss: 0.9457710385322571:  86%|████████▋ | 5585/6470 [1:07:14<11:14,  1.31it/s]
1|5586|Loss: 0.9457710385322571:  86%|████████▋ | 5586/6470 [1:07:14<10:44,  1.37it/s]
1|5587|Loss: 0.8482246994972229:  86%|████████▋ | 5586/6470 [1:07:14<10:44,  1.37it/s]
1|5587|Loss: 0.8482246994972229:  86%|████████▋ | 5587/6470 [1:07:14<09:38,  1.53it/s]
1|5588|Loss: 1.0017703771591187:  86%|████████▋ | 5587/6470 [1:07:15<09:38,  1.53it/s]
1|5588|Loss: 1.0017703771591187:  86%|████████▋ | 5588/6470 [1:07:15<09:38,  1.53it/s]
1|5589|Loss: 0.8697940111160278:  86%|████████▋ | 5588/6470 [1:07:15<09:38,  1.53it/s]
1|5589|Loss: 0.8697940111160278:  86%|████████▋ | 5589/6470 [1:07:16<09:38,  1.52it/s]
1|5590|Loss: 0.8927117586135864:  86%|████████▋ | 5589/6470 [1:07:16<09:38,  1.52it/s]
1|5590|Loss: 0.8927117586135864:  86%|████████▋ | 5590/6470 [1:07:16<09:05,  1.61it/s]
1|5591|Loss: 1.0765306949615479:  86%|████████▋ | 5590/6470 [1:07:16<09:05,  1.61it/s]
1|5591|Loss: 1.0765306949615479:  86%|████████▋ | 5591/6470 [1:07:17<08:40,  1.69it/s]
1|5592|Loss: 0.8941364884376526:  86%|████████▋ | 5591/6470 [1:07:17<08:40,  1.69it/s]
1|5592|Loss: 0.8941364884376526:  86%|████████▋ | 5592/6470 [1:07:17<08:18,  1.76it/s]
1|5593|Loss: 0.779760479927063:  86%|████████▋ | 5592/6470 [1:07:18<08:18,  1.76it/s] 
1|5593|Loss: 0.779760479927063:  86%|████████▋ | 5593/6470 [1:07:18<08:41,  1.68it/s]
1|5594|Loss: 0.9022844433784485:  86%|████████▋ | 5593/6470 [1:07:18<08:41,  1.68it/s]
1|5594|Loss: 0.9022844433784485:  86%|████████▋ | 5594/6470 [1:07:18<08:44,  1.67it/s]
1|5595|Loss: 0.8269763588905334:  86%|████████▋ | 5594/6470 [1:07:19<08:44,  1.67it/s]
1|5595|Loss: 0.8269763588905334:  86%|████████▋ | 5595/6470 [1:07:19<08:51,  1.65it/s]
1|5596|Loss: 0.8162424564361572:  86%|████████▋ | 5595/6470 [1:07:19<08:51,  1.65it/s]
1|5596|Loss: 0.8162424564361572:  86%|████████▋ | 5596/6470 [1:07:20<08:59,  1.62it/s]
1|5597|Loss: 1.052212119102478:  86%|████████▋ | 5596/6470 [1:07:20<08:59,  1.62it/s] 
1|5597|Loss: 1.052212119102478:  87%|████████▋ | 5597/6470 [1:07:20<09:05,  1.60it/s]
1|5598|Loss: 0.874575674533844:  87%|████████▋ | 5597/6470 [1:07:21<09:05,  1.60it/s]
1|5598|Loss: 0.874575674533844:  87%|████████▋ | 5598/6470 [1:07:21<09:04,  1.60it/s]
1|5599|Loss: 0.8515630960464478:  87%|████████▋ | 5598/6470 [1:07:21<09:04,  1.60it/s]
1|5599|Loss: 0.8515630960464478:  87%|████████▋ | 5599/6470 [1:07:22<09:08,  1.59it/s]
1|5600|Loss: 0.8039666414260864:  87%|████████▋ | 5599/6470 [1:07:22<09:08,  1.59it/s]
1|5600|Loss: 0.8039666414260864:  87%|████████▋ | 5600/6470 [1:07:22<09:32,  1.52it/s]
1|5601|Loss: 0.6765031218528748:  87%|████████▋ | 5600/6470 [1:07:23<09:32,  1.52it/s]
1|5601|Loss: 0.6765031218528748:  87%|████████▋ | 5601/6470 [1:07:23<08:35,  1.69it/s]
1|5602|Loss: 0.8912764191627502:  87%|████████▋ | 5601/6470 [1:07:23<08:35,  1.69it/s]
1|5602|Loss: 0.8912764191627502:  87%|████████▋ | 5602/6470 [1:07:23<08:33,  1.69it/s]
1|5603|Loss: 0.6936696767807007:  87%|████████▋ | 5602/6470 [1:07:24<08:33,  1.69it/s]
1|5603|Loss: 0.6936696767807007:  87%|████████▋ | 5603/6470 [1:07:24<09:03,  1.60it/s]
1|5604|Loss: 0.8914647102355957:  87%|████████▋ | 5603/6470 [1:07:24<09:03,  1.60it/s]
1|5604|Loss: 0.8914647102355957:  87%|████████▋ | 5604/6470 [1:07:25<09:28,  1.52it/s]
1|5605|Loss: 0.7791832089424133:  87%|████████▋ | 5604/6470 [1:07:25<09:28,  1.52it/s]
1|5605|Loss: 0.7791832089424133:  87%|████████▋ | 5605/6470 [1:07:25<09:28,  1.52it/s]
1|5606|Loss: 0.7676680684089661:  87%|████████▋ | 5605/6470 [1:07:26<09:28,  1.52it/s]
1|5606|Loss: 0.7676680684089661:  87%|████████▋ | 5606/6470 [1:07:26<09:04,  1.59it/s]
1|5607|Loss: 0.9354969263076782:  87%|████████▋ | 5606/6470 [1:07:26<09:04,  1.59it/s]
1|5607|Loss: 0.9354969263076782:  87%|████████▋ | 5607/6470 [1:07:27<09:21,  1.54it/s]
1|5608|Loss: 0.752783477306366:  87%|████████▋ | 5607/6470 [1:07:27<09:21,  1.54it/s] 
1|5608|Loss: 0.752783477306366:  87%|████████▋ | 5608/6470 [1:07:27<09:41,  1.48it/s]
1|5609|Loss: 0.964171290397644:  87%|████████▋ | 5608/6470 [1:07:28<09:41,  1.48it/s]
1|5609|Loss: 0.964171290397644:  87%|████████▋ | 5609/6470 [1:07:28<09:49,  1.46it/s]
1|5610|Loss: 0.9397029280662537:  87%|████████▋ | 5609/6470 [1:07:29<09:49,  1.46it/s]
1|5610|Loss: 0.9397029280662537:  87%|████████▋ | 5610/6470 [1:07:29<10:00,  1.43it/s]
1|5611|Loss: 0.8970948457717896:  87%|████████▋ | 5610/6470 [1:07:29<10:00,  1.43it/s]
1|5611|Loss: 0.8970948457717896:  87%|████████▋ | 5611/6470 [1:07:30<09:49,  1.46it/s]
1|5612|Loss: 0.8600445985794067:  87%|████████▋ | 5611/6470 [1:07:30<09:49,  1.46it/s]
1|5612|Loss: 0.8600445985794067:  87%|████████▋ | 5612/6470 [1:07:30<09:17,  1.54it/s]
1|5613|Loss: 0.968795120716095:  87%|████████▋ | 5612/6470 [1:07:31<09:17,  1.54it/s] 
1|5613|Loss: 0.968795120716095:  87%|████████▋ | 5613/6470 [1:07:31<09:49,  1.45it/s]
1|5614|Loss: 0.9686447381973267:  87%|████████▋ | 5613/6470 [1:07:31<09:49,  1.45it/s]
1|5614|Loss: 0.9686447381973267:  87%|████████▋ | 5614/6470 [1:07:32<09:42,  1.47it/s]
1|5615|Loss: 0.8644261360168457:  87%|████████▋ | 5614/6470 [1:07:32<09:42,  1.47it/s]
1|5615|Loss: 0.8644261360168457:  87%|████████▋ | 5615/6470 [1:07:32<09:30,  1.50it/s]
1|5616|Loss: 0.7906611561775208:  87%|████████▋ | 5615/6470 [1:07:33<09:30,  1.50it/s]
1|5616|Loss: 0.7906611561775208:  87%|████████▋ | 5616/6470 [1:07:33<09:12,  1.55it/s]
1|5617|Loss: 0.9014877676963806:  87%|████████▋ | 5616/6470 [1:07:33<09:12,  1.55it/s]
1|5617|Loss: 0.9014877676963806:  87%|████████▋ | 5617/6470 [1:07:34<09:28,  1.50it/s]
1|5618|Loss: 0.9123797416687012:  87%|████████▋ | 5617/6470 [1:07:34<09:28,  1.50it/s]
1|5618|Loss: 0.9123797416687012:  87%|████████▋ | 5618/6470 [1:07:34<09:22,  1.52it/s]
1|5619|Loss: 0.9500530362129211:  87%|████████▋ | 5618/6470 [1:07:35<09:22,  1.52it/s]
1|5619|Loss: 0.9500530362129211:  87%|████████▋ | 5619/6470 [1:07:35<09:54,  1.43it/s]
1|5620|Loss: 0.9964723587036133:  87%|████████▋ | 5619/6470 [1:07:35<09:54,  1.43it/s]
1|5620|Loss: 0.9964723587036133:  87%|████████▋ | 5620/6470 [1:07:36<09:23,  1.51it/s]
1|5621|Loss: 0.9105601906776428:  87%|████████▋ | 5620/6470 [1:07:36<09:23,  1.51it/s]
1|5621|Loss: 0.9105601906776428:  87%|████████▋ | 5621/6470 [1:07:36<09:23,  1.51it/s]
1|5622|Loss: 0.8401753306388855:  87%|████████▋ | 5621/6470 [1:07:37<09:23,  1.51it/s]
1|5622|Loss: 0.8401753306388855:  87%|████████▋ | 5622/6470 [1:07:37<09:20,  1.51it/s]
1|5623|Loss: 0.934196412563324:  87%|████████▋ | 5622/6470 [1:07:37<09:20,  1.51it/s] 
1|5623|Loss: 0.934196412563324:  87%|████████▋ | 5623/6470 [1:07:37<09:09,  1.54it/s]
1|5624|Loss: 0.8945288062095642:  87%|████████▋ | 5623/6470 [1:07:38<09:09,  1.54it/s]
1|5624|Loss: 0.8945288062095642:  87%|████████▋ | 5624/6470 [1:07:38<08:53,  1.59it/s]
1|5625|Loss: 0.8138402700424194:  87%|████████▋ | 5624/6470 [1:07:38<08:53,  1.59it/s]
1|5625|Loss: 0.8138402700424194:  87%|████████▋ | 5625/6470 [1:07:39<09:11,  1.53it/s]
1|5626|Loss: 0.9667274355888367:  87%|████████▋ | 5625/6470 [1:07:39<09:11,  1.53it/s]
1|5626|Loss: 0.9667274355888367:  87%|████████▋ | 5626/6470 [1:07:39<09:32,  1.47it/s]
1|5627|Loss: 0.9357390403747559:  87%|████████▋ | 5626/6470 [1:07:40<09:32,  1.47it/s]
1|5627|Loss: 0.9357390403747559:  87%|████████▋ | 5627/6470 [1:07:40<09:09,  1.54it/s]
1|5628|Loss: 0.7941118478775024:  87%|████████▋ | 5627/6470 [1:07:40<09:09,  1.54it/s]
1|5628|Loss: 0.7941118478775024:  87%|████████▋ | 5628/6470 [1:07:41<09:01,  1.56it/s]
1|5629|Loss: 0.8993614315986633:  87%|████████▋ | 5628/6470 [1:07:41<09:01,  1.56it/s]
1|5629|Loss: 0.8993614315986633:  87%|████████▋ | 5629/6470 [1:07:41<08:55,  1.57it/s]
1|5630|Loss: 0.9195807576179504:  87%|████████▋ | 5629/6470 [1:07:42<08:55,  1.57it/s]
1|5630|Loss: 0.9195807576179504:  87%|████████▋ | 5630/6470 [1:07:42<08:31,  1.64it/s]
1|5631|Loss: 0.825261652469635:  87%|████████▋ | 5630/6470 [1:07:42<08:31,  1.64it/s] 
1|5631|Loss: 0.825261652469635:  87%|████████▋ | 5631/6470 [1:07:42<08:33,  1.63it/s]
1|5632|Loss: 0.8397818207740784:  87%|████████▋ | 5631/6470 [1:07:43<08:33,  1.63it/s]
1|5632|Loss: 0.8397818207740784:  87%|████████▋ | 5632/6470 [1:07:43<08:58,  1.56it/s]
1|5633|Loss: 0.8578817248344421:  87%|████████▋ | 5632/6470 [1:07:44<08:58,  1.56it/s]
1|5633|Loss: 0.8578817248344421:  87%|████████▋ | 5633/6470 [1:07:44<08:52,  1.57it/s]
1|5634|Loss: 0.872748613357544:  87%|████████▋ | 5633/6470 [1:07:44<08:52,  1.57it/s] 
1|5634|Loss: 0.872748613357544:  87%|████████▋ | 5634/6470 [1:07:44<08:37,  1.62it/s]
1|5635|Loss: 0.7880398631095886:  87%|████████▋ | 5634/6470 [1:07:45<08:37,  1.62it/s]
1|5635|Loss: 0.7880398631095886:  87%|████████▋ | 5635/6470 [1:07:45<08:51,  1.57it/s]
1|5636|Loss: 0.7557926177978516:  87%|████████▋ | 5635/6470 [1:07:46<08:51,  1.57it/s]
1|5636|Loss: 0.7557926177978516:  87%|████████▋ | 5636/6470 [1:07:46<09:49,  1.41it/s]
1|5637|Loss: 0.7950145602226257:  87%|████████▋ | 5636/6470 [1:07:46<09:49,  1.41it/s]
1|5637|Loss: 0.7950145602226257:  87%|████████▋ | 5637/6470 [1:07:47<09:57,  1.39it/s]
1|5638|Loss: 0.8397963047027588:  87%|████████▋ | 5637/6470 [1:07:47<09:57,  1.39it/s]
1|5638|Loss: 0.8397963047027588:  87%|████████▋ | 5638/6470 [1:07:47<08:45,  1.58it/s]
1|5639|Loss: 0.7976520657539368:  87%|████████▋ | 5638/6470 [1:07:47<08:45,  1.58it/s]
1|5639|Loss: 0.7976520657539368:  87%|████████▋ | 5639/6470 [1:07:48<08:52,  1.56it/s]
1|5640|Loss: 0.5723680853843689:  87%|████████▋ | 5639/6470 [1:07:49<08:52,  1.56it/s]
1|5640|Loss: 0.5723680853843689:  87%|████████▋ | 5640/6470 [1:07:49<13:19,  1.04it/s]
1|5641|Loss: 0.8025305271148682:  87%|████████▋ | 5640/6470 [1:07:50<13:19,  1.04it/s]
1|5641|Loss: 0.8025305271148682:  87%|████████▋ | 5641/6470 [1:07:50<12:29,  1.11it/s]
1|5642|Loss: 1.0244028568267822:  87%|████████▋ | 5641/6470 [1:07:51<12:29,  1.11it/s]
1|5642|Loss: 1.0244028568267822:  87%|████████▋ | 5642/6470 [1:07:51<11:31,  1.20it/s]
1|5643|Loss: 0.6273019313812256:  87%|████████▋ | 5642/6470 [1:07:51<11:31,  1.20it/s]
1|5643|Loss: 0.6273019313812256:  87%|████████▋ | 5643/6470 [1:07:52<11:20,  1.22it/s]
1|5644|Loss: 0.8423064351081848:  87%|████████▋ | 5643/6470 [1:07:52<11:20,  1.22it/s]
1|5644|Loss: 0.8423064351081848:  87%|████████▋ | 5644/6470 [1:07:53<11:20,  1.21it/s]
1|5645|Loss: 0.9755208492279053:  87%|████████▋ | 5644/6470 [1:07:53<11:20,  1.21it/s]
1|5645|Loss: 0.9755208492279053:  87%|████████▋ | 5645/6470 [1:07:53<11:22,  1.21it/s]
1|5646|Loss: 0.8121342062950134:  87%|████████▋ | 5645/6470 [1:07:54<11:22,  1.21it/s]
1|5646|Loss: 0.8121342062950134:  87%|████████▋ | 5646/6470 [1:07:54<11:21,  1.21it/s]
1|5647|Loss: 0.8473014831542969:  87%|████████▋ | 5646/6470 [1:07:55<11:21,  1.21it/s]
1|5647|Loss: 0.8473014831542969:  87%|████████▋ | 5647/6470 [1:07:55<10:57,  1.25it/s]
1|5648|Loss: 0.8154573440551758:  87%|████████▋ | 5647/6470 [1:07:55<10:57,  1.25it/s]
1|5648|Loss: 0.8154573440551758:  87%|████████▋ | 5648/6470 [1:07:56<09:58,  1.37it/s]
1|5649|Loss: 0.9756056070327759:  87%|████████▋ | 5648/6470 [1:07:56<09:58,  1.37it/s]
1|5649|Loss: 0.9756056070327759:  87%|████████▋ | 5649/6470 [1:07:56<09:49,  1.39it/s]
1|5650|Loss: 0.7744209170341492:  87%|████████▋ | 5649/6470 [1:07:57<09:49,  1.39it/s]
1|5650|Loss: 0.7744209170341492:  87%|████████▋ | 5650/6470 [1:07:57<10:09,  1.34it/s]
1|5651|Loss: 0.92593914270401:  87%|████████▋ | 5650/6470 [1:07:57<10:09,  1.34it/s]  
1|5651|Loss: 0.92593914270401:  87%|████████▋ | 5651/6470 [1:07:58<09:41,  1.41it/s]
1|5652|Loss: 0.9749414324760437:  87%|████████▋ | 5651/6470 [1:07:58<09:41,  1.41it/s]
1|5652|Loss: 0.9749414324760437:  87%|████████▋ | 5652/6470 [1:07:58<09:03,  1.51it/s]
1|5653|Loss: 0.7883327603340149:  87%|████████▋ | 5652/6470 [1:07:59<09:03,  1.51it/s]
1|5653|Loss: 0.7883327603340149:  87%|████████▋ | 5653/6470 [1:07:59<08:26,  1.61it/s]
1|5654|Loss: 0.9278897643089294:  87%|████████▋ | 5653/6470 [1:07:59<08:26,  1.61it/s]
1|5654|Loss: 0.9278897643089294:  87%|████████▋ | 5654/6470 [1:07:59<07:31,  1.81it/s]
1|5655|Loss: 0.9492822289466858:  87%|████████▋ | 5654/6470 [1:07:59<07:31,  1.81it/s]
1|5655|Loss: 0.9492822289466858:  87%|████████▋ | 5655/6470 [1:08:00<07:50,  1.73it/s]
1|5656|Loss: 0.7375360131263733:  87%|████████▋ | 5655/6470 [1:08:00<07:50,  1.73it/s]
1|5656|Loss: 0.7375360131263733:  87%|████████▋ | 5656/6470 [1:08:00<08:26,  1.61it/s]
1|5657|Loss: 0.9950313568115234:  87%|████████▋ | 5656/6470 [1:08:01<08:26,  1.61it/s]
1|5657|Loss: 0.9950313568115234:  87%|████████▋ | 5657/6470 [1:08:01<08:23,  1.62it/s]
1|5658|Loss: 1.0341482162475586:  87%|████████▋ | 5657/6470 [1:08:01<08:23,  1.62it/s]
1|5658|Loss: 1.0341482162475586:  87%|████████▋ | 5658/6470 [1:08:02<08:57,  1.51it/s]
1|5659|Loss: 0.6958920955657959:  87%|████████▋ | 5658/6470 [1:08:02<08:57,  1.51it/s]
1|5659|Loss: 0.6958920955657959:  87%|████████▋ | 5659/6470 [1:08:03<09:01,  1.50it/s]
1|5660|Loss: 0.7831729650497437:  87%|████████▋ | 5659/6470 [1:08:03<09:01,  1.50it/s]
1|5660|Loss: 0.7831729650497437:  87%|████████▋ | 5660/6470 [1:08:03<09:07,  1.48it/s]
1|5661|Loss: 0.7712900638580322:  87%|████████▋ | 5660/6470 [1:08:04<09:07,  1.48it/s]
1|5661|Loss: 0.7712900638580322:  87%|████████▋ | 5661/6470 [1:08:04<08:30,  1.58it/s]
1|5662|Loss: 0.8678763508796692:  87%|████████▋ | 5661/6470 [1:08:04<08:30,  1.58it/s]
1|5662|Loss: 0.8678763508796692:  88%|████████▊ | 5662/6470 [1:08:04<08:02,  1.68it/s]
1|5663|Loss: 0.878034770488739:  88%|████████▊ | 5662/6470 [1:08:05<08:02,  1.68it/s] 
1|5663|Loss: 0.878034770488739:  88%|████████▊ | 5663/6470 [1:08:05<08:24,  1.60it/s]
1|5664|Loss: 0.7762112617492676:  88%|████████▊ | 5663/6470 [1:08:05<08:24,  1.60it/s]
1|5664|Loss: 0.7762112617492676:  88%|████████▊ | 5664/6470 [1:08:06<08:45,  1.53it/s]
1|5665|Loss: 0.8823381662368774:  88%|████████▊ | 5664/6470 [1:08:06<08:45,  1.53it/s]
1|5665|Loss: 0.8823381662368774:  88%|████████▊ | 5665/6470 [1:08:06<08:08,  1.65it/s]
1|5666|Loss: 0.8698083758354187:  88%|████████▊ | 5665/6470 [1:08:07<08:08,  1.65it/s]
1|5666|Loss: 0.8698083758354187:  88%|████████▊ | 5666/6470 [1:08:07<08:23,  1.60it/s]
1|5667|Loss: 0.7541034817695618:  88%|████████▊ | 5666/6470 [1:08:07<08:23,  1.60it/s]
1|5667|Loss: 0.7541034817695618:  88%|████████▊ | 5667/6470 [1:08:08<09:33,  1.40it/s]
1|5668|Loss: 0.8292655348777771:  88%|████████▊ | 5667/6470 [1:08:08<09:33,  1.40it/s]
1|5668|Loss: 0.8292655348777771:  88%|████████▊ | 5668/6470 [1:08:08<09:32,  1.40it/s]
1|5669|Loss: 0.8659718036651611:  88%|████████▊ | 5668/6470 [1:08:09<09:32,  1.40it/s]
1|5669|Loss: 0.8659718036651611:  88%|████████▊ | 5669/6470 [1:08:09<09:43,  1.37it/s]
1|5670|Loss: 0.8739214539527893:  88%|████████▊ | 5669/6470 [1:08:10<09:43,  1.37it/s]
1|5670|Loss: 0.8739214539527893:  88%|████████▊ | 5670/6470 [1:08:10<09:30,  1.40it/s]
1|5671|Loss: 0.7702310681343079:  88%|████████▊ | 5670/6470 [1:08:10<09:30,  1.40it/s]
1|5671|Loss: 0.7702310681343079:  88%|████████▊ | 5671/6470 [1:08:10<08:17,  1.61it/s]
1|5672|Loss: 0.7954409718513489:  88%|████████▊ | 5671/6470 [1:08:11<08:17,  1.61it/s]
1|5672|Loss: 0.7954409718513489:  88%|████████▊ | 5672/6470 [1:08:11<07:59,  1.66it/s]
1|5673|Loss: 0.8968830108642578:  88%|████████▊ | 5672/6470 [1:08:11<07:59,  1.66it/s]
1|5673|Loss: 0.8968830108642578:  88%|████████▊ | 5673/6470 [1:08:12<08:19,  1.60it/s]
1|5674|Loss: 0.8922454714775085:  88%|████████▊ | 5673/6470 [1:08:12<08:19,  1.60it/s]
1|5674|Loss: 0.8922454714775085:  88%|████████▊ | 5674/6470 [1:08:12<08:21,  1.59it/s]
1|5675|Loss: 1.0296343564987183:  88%|████████▊ | 5674/6470 [1:08:13<08:21,  1.59it/s]
1|5675|Loss: 1.0296343564987183:  88%|████████▊ | 5675/6470 [1:08:13<08:51,  1.50it/s]
1|5676|Loss: 0.8214312195777893:  88%|████████▊ | 5675/6470 [1:08:13<08:51,  1.50it/s]
1|5676|Loss: 0.8214312195777893:  88%|████████▊ | 5676/6470 [1:08:14<08:53,  1.49it/s]
1|5677|Loss: 0.9745379686355591:  88%|████████▊ | 5676/6470 [1:08:14<08:53,  1.49it/s]
1|5677|Loss: 0.9745379686355591:  88%|████████▊ | 5677/6470 [1:08:14<08:39,  1.53it/s]
1|5678|Loss: 0.7799127697944641:  88%|████████▊ | 5677/6470 [1:08:15<08:39,  1.53it/s]
1|5678|Loss: 0.7799127697944641:  88%|████████▊ | 5678/6470 [1:08:15<08:29,  1.55it/s]
1|5679|Loss: 0.8689799308776855:  88%|████████▊ | 5678/6470 [1:08:15<08:29,  1.55it/s]
1|5679|Loss: 0.8689799308776855:  88%|████████▊ | 5679/6470 [1:08:15<08:01,  1.64it/s]
1|5680|Loss: 0.8789578080177307:  88%|████████▊ | 5679/6470 [1:08:16<08:01,  1.64it/s]
1|5680|Loss: 0.8789578080177307:  88%|████████▊ | 5680/6470 [1:08:16<08:38,  1.52it/s]
1|5681|Loss: 0.7825606465339661:  88%|████████▊ | 5680/6470 [1:08:17<08:38,  1.52it/s]
1|5681|Loss: 0.7825606465339661:  88%|████████▊ | 5681/6470 [1:08:17<08:20,  1.58it/s]
1|5682|Loss: 0.9631472826004028:  88%|████████▊ | 5681/6470 [1:08:17<08:20,  1.58it/s]
1|5682|Loss: 0.9631472826004028:  88%|████████▊ | 5682/6470 [1:08:17<08:42,  1.51it/s]
1|5683|Loss: 0.8252830505371094:  88%|████████▊ | 5682/6470 [1:08:18<08:42,  1.51it/s]
1|5683|Loss: 0.8252830505371094:  88%|████████▊ | 5683/6470 [1:08:18<08:10,  1.60it/s]
1|5684|Loss: 0.8284103870391846:  88%|████████▊ | 5683/6470 [1:08:18<08:10,  1.60it/s]
1|5684|Loss: 0.8284103870391846:  88%|████████▊ | 5684/6470 [1:08:19<08:20,  1.57it/s]
1|5685|Loss: 0.8202109932899475:  88%|████████▊ | 5684/6470 [1:08:19<08:20,  1.57it/s]
1|5685|Loss: 0.8202109932899475:  88%|████████▊ | 5685/6470 [1:08:19<08:46,  1.49it/s]
1|5686|Loss: 0.7261186242103577:  88%|████████▊ | 5685/6470 [1:08:20<08:46,  1.49it/s]
1|5686|Loss: 0.7261186242103577:  88%|████████▊ | 5686/6470 [1:08:20<08:41,  1.50it/s]
1|5687|Loss: 0.8167402148246765:  88%|████████▊ | 5686/6470 [1:08:20<08:41,  1.50it/s]
1|5687|Loss: 0.8167402148246765:  88%|████████▊ | 5687/6470 [1:08:21<08:45,  1.49it/s]
1|5688|Loss: 0.9426884055137634:  88%|████████▊ | 5687/6470 [1:08:21<08:45,  1.49it/s]
1|5688|Loss: 0.9426884055137634:  88%|████████▊ | 5688/6470 [1:08:21<08:35,  1.52it/s]
1|5689|Loss: 0.9298466444015503:  88%|████████▊ | 5688/6470 [1:08:22<08:35,  1.52it/s]
1|5689|Loss: 0.9298466444015503:  88%|████████▊ | 5689/6470 [1:08:22<08:41,  1.50it/s]
1|5690|Loss: 0.6401250958442688:  88%|████████▊ | 5689/6470 [1:08:23<08:41,  1.50it/s]
1|5690|Loss: 0.6401250958442688:  88%|████████▊ | 5690/6470 [1:08:23<09:36,  1.35it/s]
1|5691|Loss: 0.8701454401016235:  88%|████████▊ | 5690/6470 [1:08:23<09:36,  1.35it/s]
1|5691|Loss: 0.8701454401016235:  88%|████████▊ | 5691/6470 [1:08:24<09:49,  1.32it/s]
1|5692|Loss: 0.9499379992485046:  88%|████████▊ | 5691/6470 [1:08:24<09:49,  1.32it/s]
1|5692|Loss: 0.9499379992485046:  88%|████████▊ | 5692/6470 [1:08:24<09:18,  1.39it/s]
1|5693|Loss: 0.6956105828285217:  88%|████████▊ | 5692/6470 [1:08:25<09:18,  1.39it/s]
1|5693|Loss: 0.6956105828285217:  88%|████████▊ | 5693/6470 [1:08:25<08:27,  1.53it/s]
1|5694|Loss: 0.8901898264884949:  88%|████████▊ | 5693/6470 [1:08:25<08:27,  1.53it/s]
1|5694|Loss: 0.8901898264884949:  88%|████████▊ | 5694/6470 [1:08:26<08:46,  1.47it/s]
1|5695|Loss: 0.6916436553001404:  88%|████████▊ | 5694/6470 [1:08:26<08:46,  1.47it/s]
1|5695|Loss: 0.6916436553001404:  88%|████████▊ | 5695/6470 [1:08:27<09:25,  1.37it/s]
1|5696|Loss: 0.8008260726928711:  88%|████████▊ | 5695/6470 [1:08:27<09:25,  1.37it/s]
1|5696|Loss: 0.8008260726928711:  88%|████████▊ | 5696/6470 [1:08:27<09:05,  1.42it/s]
1|5697|Loss: 0.8822528719902039:  88%|████████▊ | 5696/6470 [1:08:28<09:05,  1.42it/s]
1|5697|Loss: 0.8822528719902039:  88%|████████▊ | 5697/6470 [1:08:28<09:05,  1.42it/s]
1|5698|Loss: 0.9226825833320618:  88%|████████▊ | 5697/6470 [1:08:28<09:05,  1.42it/s]
1|5698|Loss: 0.9226825833320618:  88%|████████▊ | 5698/6470 [1:08:28<08:43,  1.47it/s]
1|5699|Loss: 0.907296359539032:  88%|████████▊ | 5698/6470 [1:08:29<08:43,  1.47it/s] 
1|5699|Loss: 0.907296359539032:  88%|████████▊ | 5699/6470 [1:08:29<08:44,  1.47it/s]
1|5700|Loss: 0.827161967754364:  88%|████████▊ | 5699/6470 [1:08:30<08:44,  1.47it/s]
1|5700|Loss: 0.827161967754364:  88%|████████▊ | 5700/6470 [1:08:30<08:21,  1.53it/s]
1|5701|Loss: 0.9534960985183716:  88%|████████▊ | 5700/6470 [1:08:30<08:21,  1.53it/s]
1|5701|Loss: 0.9534960985183716:  88%|████████▊ | 5701/6470 [1:08:30<08:26,  1.52it/s]
1|5702|Loss: 0.9336437582969666:  88%|████████▊ | 5701/6470 [1:08:31<08:26,  1.52it/s]
1|5702|Loss: 0.9336437582969666:  88%|████████▊ | 5702/6470 [1:08:31<09:01,  1.42it/s]
1|5703|Loss: 0.8742000460624695:  88%|████████▊ | 5702/6470 [1:08:32<09:01,  1.42it/s]
1|5703|Loss: 0.8742000460624695:  88%|████████▊ | 5703/6470 [1:08:32<08:51,  1.44it/s]
1|5704|Loss: 0.813102126121521:  88%|████████▊ | 5703/6470 [1:08:32<08:51,  1.44it/s] 
1|5704|Loss: 0.813102126121521:  88%|████████▊ | 5704/6470 [1:08:33<08:49,  1.45it/s]
1|5705|Loss: 0.8536336421966553:  88%|████████▊ | 5704/6470 [1:08:33<08:49,  1.45it/s]
1|5705|Loss: 0.8536336421966553:  88%|████████▊ | 5705/6470 [1:08:33<08:11,  1.56it/s]
1|5706|Loss: 0.6807370185852051:  88%|████████▊ | 5705/6470 [1:08:33<08:11,  1.56it/s]
1|5706|Loss: 0.6807370185852051:  88%|████████▊ | 5706/6470 [1:08:34<08:14,  1.55it/s]
1|5707|Loss: 0.9129295945167542:  88%|████████▊ | 5706/6470 [1:08:34<08:14,  1.55it/s]
1|5707|Loss: 0.9129295945167542:  88%|████████▊ | 5707/6470 [1:08:35<08:45,  1.45it/s]
1|5708|Loss: 0.9123603105545044:  88%|████████▊ | 5707/6470 [1:08:35<08:45,  1.45it/s]
1|5708|Loss: 0.9123603105545044:  88%|████████▊ | 5708/6470 [1:08:35<08:13,  1.55it/s]
1|5709|Loss: 0.8173456788063049:  88%|████████▊ | 5708/6470 [1:08:35<08:13,  1.55it/s]
1|5709|Loss: 0.8173456788063049:  88%|████████▊ | 5709/6470 [1:08:36<08:13,  1.54it/s]
1|5710|Loss: 0.7997077107429504:  88%|████████▊ | 5709/6470 [1:08:36<08:13,  1.54it/s]
1|5710|Loss: 0.7997077107429504:  88%|████████▊ | 5710/6470 [1:08:37<08:35,  1.47it/s]
1|5711|Loss: 0.7936057448387146:  88%|████████▊ | 5710/6470 [1:08:37<08:35,  1.47it/s]
1|5711|Loss: 0.7936057448387146:  88%|████████▊ | 5711/6470 [1:08:37<08:46,  1.44it/s]
1|5712|Loss: 0.8008151650428772:  88%|████████▊ | 5711/6470 [1:08:38<08:46,  1.44it/s]
1|5712|Loss: 0.8008151650428772:  88%|████████▊ | 5712/6470 [1:08:38<09:33,  1.32it/s]
1|5713|Loss: 0.8192445635795593:  88%|████████▊ | 5712/6470 [1:08:39<09:33,  1.32it/s]
1|5713|Loss: 0.8192445635795593:  88%|████████▊ | 5713/6470 [1:08:39<09:40,  1.30it/s]
1|5714|Loss: 0.8678029775619507:  88%|████████▊ | 5713/6470 [1:08:39<09:40,  1.30it/s]
1|5714|Loss: 0.8678029775619507:  88%|████████▊ | 5714/6470 [1:08:40<09:12,  1.37it/s]
1|5715|Loss: 0.765809178352356:  88%|████████▊ | 5714/6470 [1:08:40<09:12,  1.37it/s] 
1|5715|Loss: 0.765809178352356:  88%|████████▊ | 5715/6470 [1:08:40<09:08,  1.38it/s]
1|5716|Loss: 0.945533275604248:  88%|████████▊ | 5715/6470 [1:08:41<09:08,  1.38it/s]
1|5716|Loss: 0.945533275604248:  88%|████████▊ | 5716/6470 [1:08:41<08:54,  1.41it/s]
1|5717|Loss: 0.9125649333000183:  88%|████████▊ | 5716/6470 [1:08:41<08:54,  1.41it/s]
1|5717|Loss: 0.9125649333000183:  88%|████████▊ | 5717/6470 [1:08:42<08:38,  1.45it/s]
1|5718|Loss: 0.9220001697540283:  88%|████████▊ | 5717/6470 [1:08:42<08:38,  1.45it/s]
1|5718|Loss: 0.9220001697540283:  88%|████████▊ | 5718/6470 [1:08:42<08:43,  1.44it/s]
1|5719|Loss: 0.858129620552063:  88%|████████▊ | 5718/6470 [1:08:43<08:43,  1.44it/s] 
1|5719|Loss: 0.858129620552063:  88%|████████▊ | 5719/6470 [1:08:43<08:52,  1.41it/s]
1|5720|Loss: 0.9721276164054871:  88%|████████▊ | 5719/6470 [1:08:43<08:52,  1.41it/s]
1|5720|Loss: 0.9721276164054871:  88%|████████▊ | 5720/6470 [1:08:44<08:17,  1.51it/s]
1|5721|Loss: 0.7921031713485718:  88%|████████▊ | 5720/6470 [1:08:44<08:17,  1.51it/s]
1|5721|Loss: 0.7921031713485718:  88%|████████▊ | 5721/6470 [1:08:44<07:45,  1.61it/s]
1|5722|Loss: 0.8664982318878174:  88%|████████▊ | 5721/6470 [1:08:44<07:45,  1.61it/s]
1|5722|Loss: 0.8664982318878174:  88%|████████▊ | 5722/6470 [1:08:45<07:15,  1.72it/s]
1|5723|Loss: 0.8719708919525146:  88%|████████▊ | 5722/6470 [1:08:45<07:15,  1.72it/s]
1|5723|Loss: 0.8719708919525146:  88%|████████▊ | 5723/6470 [1:08:45<07:39,  1.62it/s]
1|5724|Loss: 0.8166988492012024:  88%|████████▊ | 5723/6470 [1:08:46<07:39,  1.62it/s]
1|5724|Loss: 0.8166988492012024:  88%|████████▊ | 5724/6470 [1:08:46<07:48,  1.59it/s]
1|5725|Loss: 0.9632593393325806:  88%|████████▊ | 5724/6470 [1:08:46<07:48,  1.59it/s]
1|5725|Loss: 0.9632593393325806:  88%|████████▊ | 5725/6470 [1:08:47<08:28,  1.47it/s]
1|5726|Loss: 0.8787738084793091:  88%|████████▊ | 5725/6470 [1:08:47<08:28,  1.47it/s]
1|5726|Loss: 0.8787738084793091:  89%|████████▊ | 5726/6470 [1:08:48<08:46,  1.41it/s]
1|5727|Loss: 0.9995759725570679:  89%|████████▊ | 5726/6470 [1:08:48<08:46,  1.41it/s]
1|5727|Loss: 0.9995759725570679:  89%|████████▊ | 5727/6470 [1:08:48<09:14,  1.34it/s]
1|5728|Loss: 0.853099524974823:  89%|████████▊ | 5727/6470 [1:08:49<09:14,  1.34it/s] 
1|5728|Loss: 0.853099524974823:  89%|████████▊ | 5728/6470 [1:08:49<09:13,  1.34it/s]
1|5729|Loss: 0.8487344980239868:  89%|████████▊ | 5728/6470 [1:08:50<09:13,  1.34it/s]
1|5729|Loss: 0.8487344980239868:  89%|████████▊ | 5729/6470 [1:08:50<09:08,  1.35it/s]
1|5730|Loss: 0.7845894694328308:  89%|████████▊ | 5729/6470 [1:08:50<09:08,  1.35it/s]
1|5730|Loss: 0.7845894694328308:  89%|████████▊ | 5730/6470 [1:08:51<09:23,  1.31it/s]
1|5731|Loss: 0.8571038842201233:  89%|████████▊ | 5730/6470 [1:08:51<09:23,  1.31it/s]
1|5731|Loss: 0.8571038842201233:  89%|████████▊ | 5731/6470 [1:08:51<08:41,  1.42it/s]
1|5732|Loss: 0.7256606817245483:  89%|████████▊ | 5731/6470 [1:08:52<08:41,  1.42it/s]
1|5732|Loss: 0.7256606817245483:  89%|████████▊ | 5732/6470 [1:08:52<08:31,  1.44it/s]
1|5733|Loss: 0.9908943772315979:  89%|████████▊ | 5732/6470 [1:08:52<08:31,  1.44it/s]
1|5733|Loss: 0.9908943772315979:  89%|████████▊ | 5733/6470 [1:08:53<08:59,  1.37it/s]
1|5734|Loss: 0.8473592400550842:  89%|████████▊ | 5733/6470 [1:08:53<08:59,  1.37it/s]
1|5734|Loss: 0.8473592400550842:  89%|████████▊ | 5734/6470 [1:08:54<09:17,  1.32it/s]
1|5735|Loss: 0.7535650134086609:  89%|████████▊ | 5734/6470 [1:08:54<09:17,  1.32it/s]
1|5735|Loss: 0.7535650134086609:  89%|████████▊ | 5735/6470 [1:08:54<08:09,  1.50it/s]
1|5736|Loss: 0.7742090225219727:  89%|████████▊ | 5735/6470 [1:08:54<08:09,  1.50it/s]
1|5736|Loss: 0.7742090225219727:  89%|████████▊ | 5736/6470 [1:08:55<07:40,  1.59it/s]
1|5737|Loss: 0.7616267800331116:  89%|████████▊ | 5736/6470 [1:08:55<07:40,  1.59it/s]
1|5737|Loss: 0.7616267800331116:  89%|████████▊ | 5737/6470 [1:08:55<07:58,  1.53it/s]
1|5738|Loss: 0.765626072883606:  89%|████████▊ | 5737/6470 [1:08:56<07:58,  1.53it/s] 
1|5738|Loss: 0.765626072883606:  89%|████████▊ | 5738/6470 [1:08:56<08:15,  1.48it/s]
1|5739|Loss: 0.9077848792076111:  89%|████████▊ | 5738/6470 [1:08:56<08:15,  1.48it/s]
1|5739|Loss: 0.9077848792076111:  89%|████████▊ | 5739/6470 [1:08:57<08:12,  1.49it/s]
1|5740|Loss: 0.9657086730003357:  89%|████████▊ | 5739/6470 [1:08:57<08:12,  1.49it/s]
1|5740|Loss: 0.9657086730003357:  89%|████████▊ | 5740/6470 [1:08:57<08:19,  1.46it/s]
1|5741|Loss: 0.9699016213417053:  89%|████████▊ | 5740/6470 [1:08:58<08:19,  1.46it/s]
1|5741|Loss: 0.9699016213417053:  89%|████████▊ | 5741/6470 [1:08:58<08:12,  1.48it/s]
1|5742|Loss: 0.8406485915184021:  89%|████████▊ | 5741/6470 [1:08:58<08:12,  1.48it/s]
1|5742|Loss: 0.8406485915184021:  89%|████████▊ | 5742/6470 [1:08:59<08:14,  1.47it/s]
1|5743|Loss: 0.7759006023406982:  89%|████████▊ | 5742/6470 [1:08:59<08:14,  1.47it/s]
1|5743|Loss: 0.7759006023406982:  89%|████████▉ | 5743/6470 [1:09:00<08:56,  1.36it/s]
1|5744|Loss: 0.8811116218566895:  89%|████████▉ | 5743/6470 [1:09:00<08:56,  1.36it/s]
1|5744|Loss: 0.8811116218566895:  89%|████████▉ | 5744/6470 [1:09:00<08:58,  1.35it/s]
1|5745|Loss: 0.9537948966026306:  89%|████████▉ | 5744/6470 [1:09:01<08:58,  1.35it/s]
1|5745|Loss: 0.9537948966026306:  89%|████████▉ | 5745/6470 [1:09:01<09:06,  1.33it/s]
1|5746|Loss: 0.8394361734390259:  89%|████████▉ | 5745/6470 [1:09:02<09:06,  1.33it/s]
1|5746|Loss: 0.8394361734390259:  89%|████████▉ | 5746/6470 [1:09:02<09:14,  1.30it/s]
1|5747|Loss: 0.8990387916564941:  89%|████████▉ | 5746/6470 [1:09:02<09:14,  1.30it/s]
1|5747|Loss: 0.8990387916564941:  89%|████████▉ | 5747/6470 [1:09:03<08:42,  1.38it/s]
1|5748|Loss: 0.8808637857437134:  89%|████████▉ | 5747/6470 [1:09:03<08:42,  1.38it/s]
1|5748|Loss: 0.8808637857437134:  89%|████████▉ | 5748/6470 [1:09:03<08:05,  1.49it/s]
1|5749|Loss: 1.0633457899093628:  89%|████████▉ | 5748/6470 [1:09:03<08:05,  1.49it/s]
1|5749|Loss: 1.0633457899093628:  89%|████████▉ | 5749/6470 [1:09:04<08:13,  1.46it/s]
1|5750|Loss: 0.8195768594741821:  89%|████████▉ | 5749/6470 [1:09:04<08:13,  1.46it/s]
1|5750|Loss: 0.8195768594741821:  89%|████████▉ | 5750/6470 [1:09:05<08:26,  1.42it/s]
1|5751|Loss: 0.9224521517753601:  89%|████████▉ | 5750/6470 [1:09:05<08:26,  1.42it/s]
1|5751|Loss: 0.9224521517753601:  89%|████████▉ | 5751/6470 [1:09:05<08:14,  1.45it/s]
1|5752|Loss: 0.8632258176803589:  89%|████████▉ | 5751/6470 [1:09:06<08:14,  1.45it/s]
1|5752|Loss: 0.8632258176803589:  89%|████████▉ | 5752/6470 [1:09:06<08:34,  1.40it/s]
1|5753|Loss: 0.9437803626060486:  89%|████████▉ | 5752/6470 [1:09:06<08:34,  1.40it/s]
1|5753|Loss: 0.9437803626060486:  89%|████████▉ | 5753/6470 [1:09:07<08:21,  1.43it/s]
1|5754|Loss: 0.856246829032898:  89%|████████▉ | 5753/6470 [1:09:07<08:21,  1.43it/s] 
1|5754|Loss: 0.856246829032898:  89%|████████▉ | 5754/6470 [1:09:07<07:56,  1.50it/s]
1|5755|Loss: 0.9132044911384583:  89%|████████▉ | 5754/6470 [1:09:08<07:56,  1.50it/s]
1|5755|Loss: 0.9132044911384583:  89%|████████▉ | 5755/6470 [1:09:08<08:07,  1.47it/s]
1|5756|Loss: 0.6813942193984985:  89%|████████▉ | 5755/6470 [1:09:08<08:07,  1.47it/s]
1|5756|Loss: 0.6813942193984985:  89%|████████▉ | 5756/6470 [1:09:09<08:18,  1.43it/s]
1|5757|Loss: 0.767309308052063:  89%|████████▉ | 5756/6470 [1:09:09<08:18,  1.43it/s] 
1|5757|Loss: 0.767309308052063:  89%|████████▉ | 5757/6470 [1:09:09<08:41,  1.37it/s]
1|5758|Loss: 1.094892978668213:  89%|████████▉ | 5757/6470 [1:09:10<08:41,  1.37it/s]
1|5758|Loss: 1.094892978668213:  89%|████████▉ | 5758/6470 [1:09:10<09:17,  1.28it/s]
1|5759|Loss: 0.9055635333061218:  89%|████████▉ | 5758/6470 [1:09:11<09:17,  1.28it/s]
1|5759|Loss: 0.9055635333061218:  89%|████████▉ | 5759/6470 [1:09:11<09:15,  1.28it/s]
1|5760|Loss: 0.8550007343292236:  89%|████████▉ | 5759/6470 [1:09:12<09:15,  1.28it/s]
1|5760|Loss: 0.8550007343292236:  89%|████████▉ | 5760/6470 [1:09:12<09:04,  1.30it/s]
1|5761|Loss: 0.7514339685440063:  89%|████████▉ | 5760/6470 [1:09:12<09:04,  1.30it/s]
1|5761|Loss: 0.7514339685440063:  89%|████████▉ | 5761/6470 [1:09:13<08:53,  1.33it/s]
1|5762|Loss: 0.9227542877197266:  89%|████████▉ | 5761/6470 [1:09:13<08:53,  1.33it/s]
1|5762|Loss: 0.9227542877197266:  89%|████████▉ | 5762/6470 [1:09:13<07:48,  1.51it/s]
1|5763|Loss: 0.7836856245994568:  89%|████████▉ | 5762/6470 [1:09:13<07:48,  1.51it/s]
1|5763|Loss: 0.7836856245994568:  89%|████████▉ | 5763/6470 [1:09:14<07:34,  1.56it/s]
1|5764|Loss: 0.8495383858680725:  89%|████████▉ | 5763/6470 [1:09:14<07:34,  1.56it/s]
1|5764|Loss: 0.8495383858680725:  89%|████████▉ | 5764/6470 [1:09:14<08:04,  1.46it/s]
1|5765|Loss: 0.8869079947471619:  89%|████████▉ | 5764/6470 [1:09:15<08:04,  1.46it/s]
1|5765|Loss: 0.8869079947471619:  89%|████████▉ | 5765/6470 [1:09:15<07:52,  1.49it/s]
1|5766|Loss: 0.7516950368881226:  89%|████████▉ | 5765/6470 [1:09:15<07:52,  1.49it/s]
1|5766|Loss: 0.7516950368881226:  89%|████████▉ | 5766/6470 [1:09:16<07:54,  1.48it/s]
1|5767|Loss: 0.6820705533027649:  89%|████████▉ | 5766/6470 [1:09:16<07:54,  1.48it/s]
1|5767|Loss: 0.6820705533027649:  89%|████████▉ | 5767/6470 [1:09:16<07:27,  1.57it/s]
1|5768|Loss: 1.0490479469299316:  89%|████████▉ | 5767/6470 [1:09:17<07:27,  1.57it/s]
1|5768|Loss: 1.0490479469299316:  89%|████████▉ | 5768/6470 [1:09:17<07:18,  1.60it/s]
1|5769|Loss: 0.9317632913589478:  89%|████████▉ | 5768/6470 [1:09:17<07:18,  1.60it/s]
1|5769|Loss: 0.9317632913589478:  89%|████████▉ | 5769/6470 [1:09:17<06:58,  1.68it/s]
1|5770|Loss: 0.9038999080657959:  89%|████████▉ | 5769/6470 [1:09:18<06:58,  1.68it/s]
1|5770|Loss: 0.9038999080657959:  89%|████████▉ | 5770/6470 [1:09:18<07:40,  1.52it/s]
1|5771|Loss: 0.7711461782455444:  89%|████████▉ | 5770/6470 [1:09:19<07:40,  1.52it/s]
1|5771|Loss: 0.7711461782455444:  89%|████████▉ | 5771/6470 [1:09:19<08:04,  1.44it/s]
1|5772|Loss: 0.8771175742149353:  89%|████████▉ | 5771/6470 [1:09:19<08:04,  1.44it/s]
1|5772|Loss: 0.8771175742149353:  89%|████████▉ | 5772/6470 [1:09:20<08:13,  1.42it/s]
1|5773|Loss: 0.981428325176239:  89%|████████▉ | 5772/6470 [1:09:20<08:13,  1.42it/s] 
1|5773|Loss: 0.981428325176239:  89%|████████▉ | 5773/6470 [1:09:20<07:40,  1.51it/s]
1|5774|Loss: 0.818312406539917:  89%|████████▉ | 5773/6470 [1:09:21<07:40,  1.51it/s]
1|5774|Loss: 0.818312406539917:  89%|████████▉ | 5774/6470 [1:09:21<07:40,  1.51it/s]
1|5775|Loss: 0.9211152791976929:  89%|████████▉ | 5774/6470 [1:09:21<07:40,  1.51it/s]
1|5775|Loss: 0.9211152791976929:  89%|████████▉ | 5775/6470 [1:09:22<07:54,  1.46it/s]
1|5776|Loss: 0.8821254372596741:  89%|████████▉ | 5775/6470 [1:09:22<07:54,  1.46it/s]
1|5776|Loss: 0.8821254372596741:  89%|████████▉ | 5776/6470 [1:09:23<08:15,  1.40it/s]
1|5777|Loss: 0.9659914374351501:  89%|████████▉ | 5776/6470 [1:09:23<08:15,  1.40it/s]
1|5777|Loss: 0.9659914374351501:  89%|████████▉ | 5777/6470 [1:09:23<08:22,  1.38it/s]
1|5778|Loss: 0.7970331311225891:  89%|████████▉ | 5777/6470 [1:09:24<08:22,  1.38it/s]
1|5778|Loss: 0.7970331311225891:  89%|████████▉ | 5778/6470 [1:09:24<08:25,  1.37it/s]
1|5779|Loss: 0.9347736239433289:  89%|████████▉ | 5778/6470 [1:09:24<08:25,  1.37it/s]
1|5779|Loss: 0.9347736239433289:  89%|████████▉ | 5779/6470 [1:09:25<08:07,  1.42it/s]
1|5780|Loss: 0.8545430898666382:  89%|████████▉ | 5779/6470 [1:09:25<08:07,  1.42it/s]
1|5780|Loss: 0.8545430898666382:  89%|████████▉ | 5780/6470 [1:09:25<08:08,  1.41it/s]
1|5781|Loss: 0.805393397808075:  89%|████████▉ | 5780/6470 [1:09:26<08:08,  1.41it/s] 
1|5781|Loss: 0.805393397808075:  89%|████████▉ | 5781/6470 [1:09:26<08:33,  1.34it/s]
1|5782|Loss: 0.9407556056976318:  89%|████████▉ | 5781/6470 [1:09:27<08:33,  1.34it/s]
1|5782|Loss: 0.9407556056976318:  89%|████████▉ | 5782/6470 [1:09:27<08:38,  1.33it/s]
1|5783|Loss: 0.8717358112335205:  89%|████████▉ | 5782/6470 [1:09:27<08:38,  1.33it/s]
1|5783|Loss: 0.8717358112335205:  89%|████████▉ | 5783/6470 [1:09:28<08:46,  1.31it/s]
1|5784|Loss: 0.9424915313720703:  89%|████████▉ | 5783/6470 [1:09:28<08:46,  1.31it/s]
1|5784|Loss: 0.9424915313720703:  89%|████████▉ | 5784/6470 [1:09:29<08:44,  1.31it/s]
1|5785|Loss: 0.8494618535041809:  89%|████████▉ | 5784/6470 [1:09:29<08:44,  1.31it/s]
1|5785|Loss: 0.8494618535041809:  89%|████████▉ | 5785/6470 [1:09:29<08:19,  1.37it/s]
1|5786|Loss: 0.8964577913284302:  89%|████████▉ | 5785/6470 [1:09:30<08:19,  1.37it/s]
1|5786|Loss: 0.8964577913284302:  89%|████████▉ | 5786/6470 [1:09:30<08:17,  1.38it/s]
1|5787|Loss: 0.8639153838157654:  89%|████████▉ | 5786/6470 [1:09:30<08:17,  1.38it/s]
1|5787|Loss: 0.8639153838157654:  89%|████████▉ | 5787/6470 [1:09:31<08:39,  1.32it/s]
1|5788|Loss: 0.9295896291732788:  89%|████████▉ | 5787/6470 [1:09:31<08:39,  1.32it/s]
1|5788|Loss: 0.9295896291732788:  89%|████████▉ | 5788/6470 [1:09:31<08:38,  1.31it/s]
1|5789|Loss: 0.8132440447807312:  89%|████████▉ | 5788/6470 [1:09:32<08:38,  1.31it/s]
1|5789|Loss: 0.8132440447807312:  89%|████████▉ | 5789/6470 [1:09:32<08:02,  1.41it/s]
1|5790|Loss: 0.8190102577209473:  89%|████████▉ | 5789/6470 [1:09:32<08:02,  1.41it/s]
1|5790|Loss: 0.8190102577209473:  89%|████████▉ | 5790/6470 [1:09:33<07:58,  1.42it/s]
1|5791|Loss: 0.7868431806564331:  89%|████████▉ | 5790/6470 [1:09:33<07:58,  1.42it/s]
1|5791|Loss: 0.7868431806564331:  90%|████████▉ | 5791/6470 [1:09:34<08:25,  1.34it/s]
1|5792|Loss: 0.8728171586990356:  90%|████████▉ | 5791/6470 [1:09:34<08:25,  1.34it/s]
1|5792|Loss: 0.8728171586990356:  90%|████████▉ | 5792/6470 [1:09:34<08:32,  1.32it/s]
1|5793|Loss: 0.9042578339576721:  90%|████████▉ | 5792/6470 [1:09:35<08:32,  1.32it/s]
1|5793|Loss: 0.9042578339576721:  90%|████████▉ | 5793/6470 [1:09:35<08:28,  1.33it/s]
1|5794|Loss: 0.9879339337348938:  90%|████████▉ | 5793/6470 [1:09:35<08:28,  1.33it/s]
1|5794|Loss: 0.9879339337348938:  90%|████████▉ | 5794/6470 [1:09:36<07:56,  1.42it/s]
1|5795|Loss: 0.8254796266555786:  90%|████████▉ | 5794/6470 [1:09:36<07:56,  1.42it/s]
1|5795|Loss: 0.8254796266555786:  90%|████████▉ | 5795/6470 [1:09:36<07:54,  1.42it/s]
1|5796|Loss: 0.7528395652770996:  90%|████████▉ | 5795/6470 [1:09:37<07:54,  1.42it/s]
1|5796|Loss: 0.7528395652770996:  90%|████████▉ | 5796/6470 [1:09:37<08:03,  1.39it/s]
1|5797|Loss: 0.9232117533683777:  90%|████████▉ | 5796/6470 [1:09:38<08:03,  1.39it/s]
1|5797|Loss: 0.9232117533683777:  90%|████████▉ | 5797/6470 [1:09:38<07:37,  1.47it/s]
1|5798|Loss: 0.7672416567802429:  90%|████████▉ | 5797/6470 [1:09:38<07:37,  1.47it/s]
1|5798|Loss: 0.7672416567802429:  90%|████████▉ | 5798/6470 [1:09:39<07:49,  1.43it/s]
1|5799|Loss: 0.8127846717834473:  90%|████████▉ | 5798/6470 [1:09:39<07:49,  1.43it/s]
1|5799|Loss: 0.8127846717834473:  90%|████████▉ | 5799/6470 [1:09:39<08:14,  1.36it/s]
1|5800|Loss: 0.7714873552322388:  90%|████████▉ | 5799/6470 [1:09:40<08:14,  1.36it/s]
1|5800|Loss: 0.7714873552322388:  90%|████████▉ | 5800/6470 [1:09:40<08:33,  1.31it/s]
1|5801|Loss: 0.8940387964248657:  90%|████████▉ | 5800/6470 [1:09:41<08:33,  1.31it/s]
1|5801|Loss: 0.8940387964248657:  90%|████████▉ | 5801/6470 [1:09:41<08:31,  1.31it/s]
1|5802|Loss: 0.7826328277587891:  90%|████████▉ | 5801/6470 [1:09:41<08:31,  1.31it/s]
1|5802|Loss: 0.7826328277587891:  90%|████████▉ | 5802/6470 [1:09:42<08:10,  1.36it/s]
1|5803|Loss: 0.9292231798171997:  90%|████████▉ | 5802/6470 [1:09:42<08:10,  1.36it/s]
1|5803|Loss: 0.9292231798171997:  90%|████████▉ | 5803/6470 [1:09:42<08:21,  1.33it/s]
1|5804|Loss: 0.8457748293876648:  90%|████████▉ | 5803/6470 [1:09:43<08:21,  1.33it/s]
1|5804|Loss: 0.8457748293876648:  90%|████████▉ | 5804/6470 [1:09:43<08:17,  1.34it/s]
1|5805|Loss: 0.7388781905174255:  90%|████████▉ | 5804/6470 [1:09:43<08:17,  1.34it/s]
1|5805|Loss: 0.7388781905174255:  90%|████████▉ | 5805/6470 [1:09:44<07:54,  1.40it/s]
1|5806|Loss: 0.8623232245445251:  90%|████████▉ | 5805/6470 [1:09:44<07:54,  1.40it/s]
1|5806|Loss: 0.8623232245445251:  90%|████████▉ | 5806/6470 [1:09:44<07:55,  1.40it/s]
1|5807|Loss: 0.8887543678283691:  90%|████████▉ | 5806/6470 [1:09:45<07:55,  1.40it/s]
1|5807|Loss: 0.8887543678283691:  90%|████████▉ | 5807/6470 [1:09:45<07:34,  1.46it/s]
1|5808|Loss: 1.001578450202942:  90%|████████▉ | 5807/6470 [1:09:45<07:34,  1.46it/s] 
1|5808|Loss: 1.001578450202942:  90%|████████▉ | 5808/6470 [1:09:46<07:37,  1.45it/s]
1|5809|Loss: 0.7910391092300415:  90%|████████▉ | 5808/6470 [1:09:46<07:37,  1.45it/s]
1|5809|Loss: 0.7910391092300415:  90%|████████▉ | 5809/6470 [1:09:47<07:40,  1.44it/s]
1|5810|Loss: 0.8910852074623108:  90%|████████▉ | 5809/6470 [1:09:47<07:40,  1.44it/s]
1|5810|Loss: 0.8910852074623108:  90%|████████▉ | 5810/6470 [1:09:47<07:27,  1.48it/s]
1|5811|Loss: 0.8364191651344299:  90%|████████▉ | 5810/6470 [1:09:48<07:27,  1.48it/s]
1|5811|Loss: 0.8364191651344299:  90%|████████▉ | 5811/6470 [1:09:48<07:31,  1.46it/s]
1|5812|Loss: 1.0056735277175903:  90%|████████▉ | 5811/6470 [1:09:48<07:31,  1.46it/s]
1|5812|Loss: 1.0056735277175903:  90%|████████▉ | 5812/6470 [1:09:49<07:36,  1.44it/s]
1|5813|Loss: 0.9108274579048157:  90%|████████▉ | 5812/6470 [1:09:49<07:36,  1.44it/s]
1|5813|Loss: 0.9108274579048157:  90%|████████▉ | 5813/6470 [1:09:49<07:29,  1.46it/s]
1|5814|Loss: 0.8243260979652405:  90%|████████▉ | 5813/6470 [1:09:50<07:29,  1.46it/s]
1|5814|Loss: 0.8243260979652405:  90%|████████▉ | 5814/6470 [1:09:50<07:29,  1.46it/s]
1|5815|Loss: 0.8487972617149353:  90%|████████▉ | 5814/6470 [1:09:50<07:29,  1.46it/s]
1|5815|Loss: 0.8487972617149353:  90%|████████▉ | 5815/6470 [1:09:51<07:36,  1.44it/s]
1|5816|Loss: 0.7694457173347473:  90%|████████▉ | 5815/6470 [1:09:51<07:36,  1.44it/s]
1|5816|Loss: 0.7694457173347473:  90%|████████▉ | 5816/6470 [1:09:51<07:37,  1.43it/s]
1|5817|Loss: 0.8112543225288391:  90%|████████▉ | 5816/6470 [1:09:52<07:37,  1.43it/s]
1|5817|Loss: 0.8112543225288391:  90%|████████▉ | 5817/6470 [1:09:52<07:09,  1.52it/s]
1|5818|Loss: 0.8693333268165588:  90%|████████▉ | 5817/6470 [1:09:52<07:09,  1.52it/s]
1|5818|Loss: 0.8693333268165588:  90%|████████▉ | 5818/6470 [1:09:53<07:15,  1.50it/s]
1|5819|Loss: 0.922170877456665:  90%|████████▉ | 5818/6470 [1:09:53<07:15,  1.50it/s] 
1|5819|Loss: 0.922170877456665:  90%|████████▉ | 5819/6470 [1:09:53<07:15,  1.50it/s]
1|5820|Loss: 0.788622260093689:  90%|████████▉ | 5819/6470 [1:09:54<07:15,  1.50it/s]
1|5820|Loss: 0.788622260093689:  90%|████████▉ | 5820/6470 [1:09:54<07:23,  1.47it/s]
1|5821|Loss: 0.9113534092903137:  90%|████████▉ | 5820/6470 [1:09:54<07:23,  1.47it/s]
1|5821|Loss: 0.9113534092903137:  90%|████████▉ | 5821/6470 [1:09:55<07:16,  1.49it/s]
1|5822|Loss: 1.0222617387771606:  90%|████████▉ | 5821/6470 [1:09:55<07:16,  1.49it/s]
1|5822|Loss: 1.0222617387771606:  90%|████████▉ | 5822/6470 [1:09:55<06:28,  1.67it/s]
1|5823|Loss: 0.9676439762115479:  90%|████████▉ | 5822/6470 [1:09:55<06:28,  1.67it/s]
1|5823|Loss: 0.9676439762115479:  90%|█████████ | 5823/6470 [1:09:56<06:10,  1.75it/s]
1|5824|Loss: 0.9239910244941711:  90%|█████████ | 5823/6470 [1:09:56<06:10,  1.75it/s]
1|5824|Loss: 0.9239910244941711:  90%|█████████ | 5824/6470 [1:09:56<06:01,  1.79it/s]
1|5825|Loss: 0.6712050437927246:  90%|█████████ | 5824/6470 [1:09:56<06:01,  1.79it/s]
1|5825|Loss: 0.6712050437927246:  90%|█████████ | 5825/6470 [1:09:57<06:24,  1.68it/s]
1|5826|Loss: 0.9313729405403137:  90%|█████████ | 5825/6470 [1:09:57<06:24,  1.68it/s]
1|5826|Loss: 0.9313729405403137:  90%|█████████ | 5826/6470 [1:09:57<06:27,  1.66it/s]
1|5827|Loss: 0.8873819708824158:  90%|█████████ | 5826/6470 [1:09:58<06:27,  1.66it/s]
1|5827|Loss: 0.8873819708824158:  90%|█████████ | 5827/6470 [1:09:58<07:04,  1.51it/s]
1|5828|Loss: 0.8628469109535217:  90%|█████████ | 5827/6470 [1:09:59<07:04,  1.51it/s]
1|5828|Loss: 0.8628469109535217:  90%|█████████ | 5828/6470 [1:09:59<07:46,  1.38it/s]
1|5829|Loss: 0.758100688457489:  90%|█████████ | 5828/6470 [1:10:00<07:46,  1.38it/s] 
1|5829|Loss: 0.758100688457489:  90%|█████████ | 5829/6470 [1:10:00<07:54,  1.35it/s]
1|5830|Loss: 0.8333150148391724:  90%|█████████ | 5829/6470 [1:10:00<07:54,  1.35it/s]
1|5830|Loss: 0.8333150148391724:  90%|█████████ | 5830/6470 [1:10:01<07:38,  1.40it/s]
1|5831|Loss: 0.8022895455360413:  90%|█████████ | 5830/6470 [1:10:01<07:38,  1.40it/s]
1|5831|Loss: 0.8022895455360413:  90%|█████████ | 5831/6470 [1:10:01<07:33,  1.41it/s]
1|5832|Loss: 0.7658157348632812:  90%|█████████ | 5831/6470 [1:10:02<07:33,  1.41it/s]
1|5832|Loss: 0.7658157348632812:  90%|█████████ | 5832/6470 [1:10:02<07:58,  1.33it/s]
1|5833|Loss: 0.738534688949585:  90%|█████████ | 5832/6470 [1:10:02<07:58,  1.33it/s] 
1|5833|Loss: 0.738534688949585:  90%|█████████ | 5833/6470 [1:10:03<07:48,  1.36it/s]
1|5834|Loss: 0.7988851070404053:  90%|█████████ | 5833/6470 [1:10:03<07:48,  1.36it/s]
1|5834|Loss: 0.7988851070404053:  90%|█████████ | 5834/6470 [1:10:03<07:42,  1.37it/s]
1|5835|Loss: 0.7737939953804016:  90%|█████████ | 5834/6470 [1:10:04<07:42,  1.37it/s]
1|5835|Loss: 0.7737939953804016:  90%|█████████ | 5835/6470 [1:10:04<07:42,  1.37it/s]
1|5836|Loss: 1.0028156042099:  90%|█████████ | 5835/6470 [1:10:05<07:42,  1.37it/s]   
1|5836|Loss: 1.0028156042099:  90%|█████████ | 5836/6470 [1:10:05<07:28,  1.41it/s]
1|5837|Loss: 0.7436206340789795:  90%|█████████ | 5836/6470 [1:10:05<07:28,  1.41it/s]
1|5837|Loss: 0.7436206340789795:  90%|█████████ | 5837/6470 [1:10:06<07:25,  1.42it/s]
1|5838|Loss: 0.9079585075378418:  90%|█████████ | 5837/6470 [1:10:06<07:25,  1.42it/s]
1|5838|Loss: 0.9079585075378418:  90%|█████████ | 5838/6470 [1:10:06<07:14,  1.46it/s]
1|5839|Loss: 0.8586742877960205:  90%|█████████ | 5838/6470 [1:10:07<07:14,  1.46it/s]
1|5839|Loss: 0.8586742877960205:  90%|█████████ | 5839/6470 [1:10:07<06:55,  1.52it/s]
1|5840|Loss: 0.9707045555114746:  90%|█████████ | 5839/6470 [1:10:07<06:55,  1.52it/s]
1|5840|Loss: 0.9707045555114746:  90%|█████████ | 5840/6470 [1:10:07<06:21,  1.65it/s]
1|5841|Loss: 0.8548578023910522:  90%|█████████ | 5840/6470 [1:10:08<06:21,  1.65it/s]
1|5841|Loss: 0.8548578023910522:  90%|█████████ | 5841/6470 [1:10:08<06:21,  1.65it/s]
1|5842|Loss: 0.7981232404708862:  90%|█████████ | 5841/6470 [1:10:08<06:21,  1.65it/s]
1|5842|Loss: 0.7981232404708862:  90%|█████████ | 5842/6470 [1:10:08<05:59,  1.75it/s]
1|5843|Loss: 0.8538912534713745:  90%|█████████ | 5842/6470 [1:10:09<05:59,  1.75it/s]
1|5843|Loss: 0.8538912534713745:  90%|█████████ | 5843/6470 [1:10:09<06:32,  1.60it/s]
1|5844|Loss: 0.9046123027801514:  90%|█████████ | 5843/6470 [1:10:10<06:32,  1.60it/s]
1|5844|Loss: 0.9046123027801514:  90%|█████████ | 5844/6470 [1:10:10<06:39,  1.57it/s]
1|5845|Loss: 1.0032684803009033:  90%|█████████ | 5844/6470 [1:10:10<06:39,  1.57it/s]
1|5845|Loss: 1.0032684803009033:  90%|█████████ | 5845/6470 [1:10:10<06:52,  1.51it/s]
1|5846|Loss: 0.8937340974807739:  90%|█████████ | 5845/6470 [1:10:11<06:52,  1.51it/s]
1|5846|Loss: 0.8937340974807739:  90%|█████████ | 5846/6470 [1:10:11<06:52,  1.51it/s]
1|5847|Loss: 0.8020042181015015:  90%|█████████ | 5846/6470 [1:10:12<06:52,  1.51it/s]
1|5847|Loss: 0.8020042181015015:  90%|█████████ | 5847/6470 [1:10:12<06:44,  1.54it/s]
1|5848|Loss: 0.9507359862327576:  90%|█████████ | 5847/6470 [1:10:12<06:44,  1.54it/s]
1|5848|Loss: 0.9507359862327576:  90%|█████████ | 5848/6470 [1:10:12<06:55,  1.50it/s]
1|5849|Loss: 0.7026568651199341:  90%|█████████ | 5848/6470 [1:10:13<06:55,  1.50it/s]
1|5849|Loss: 0.7026568651199341:  90%|█████████ | 5849/6470 [1:10:13<07:02,  1.47it/s]
1|5850|Loss: 0.826176643371582:  90%|█████████ | 5849/6470 [1:10:14<07:02,  1.47it/s] 
1|5850|Loss: 0.826176643371582:  90%|█████████ | 5850/6470 [1:10:14<06:42,  1.54it/s]
1|5851|Loss: 0.8495121002197266:  90%|█████████ | 5850/6470 [1:10:14<06:42,  1.54it/s]
1|5851|Loss: 0.8495121002197266:  90%|█████████ | 5851/6470 [1:10:14<06:51,  1.50it/s]
1|5852|Loss: 0.8777033090591431:  90%|█████████ | 5851/6470 [1:10:15<06:51,  1.50it/s]
1|5852|Loss: 0.8777033090591431:  90%|█████████ | 5852/6470 [1:10:15<06:48,  1.51it/s]
1|5853|Loss: 0.7888604998588562:  90%|█████████ | 5852/6470 [1:10:16<06:48,  1.51it/s]
1|5853|Loss: 0.7888604998588562:  90%|█████████ | 5853/6470 [1:10:16<06:57,  1.48it/s]
1|5854|Loss: 0.8322312831878662:  90%|█████████ | 5853/6470 [1:10:16<06:57,  1.48it/s]
1|5854|Loss: 0.8322312831878662:  90%|█████████ | 5854/6470 [1:10:16<06:40,  1.54it/s]
1|5855|Loss: 0.8670967221260071:  90%|█████████ | 5854/6470 [1:10:17<06:40,  1.54it/s]
1|5855|Loss: 0.8670967221260071:  90%|█████████ | 5855/6470 [1:10:17<06:16,  1.63it/s]
1|5856|Loss: 0.7773480415344238:  90%|█████████ | 5855/6470 [1:10:17<06:16,  1.63it/s]
1|5856|Loss: 0.7773480415344238:  91%|█████████ | 5856/6470 [1:10:18<06:48,  1.50it/s]
1|5857|Loss: 0.9024333357810974:  91%|█████████ | 5856/6470 [1:10:18<06:48,  1.50it/s]
1|5857|Loss: 0.9024333357810974:  91%|█████████ | 5857/6470 [1:10:18<06:24,  1.59it/s]
1|5858|Loss: 0.8153166174888611:  91%|█████████ | 5857/6470 [1:10:19<06:24,  1.59it/s]
1|5858|Loss: 0.8153166174888611:  91%|█████████ | 5858/6470 [1:10:19<06:09,  1.65it/s]
1|5859|Loss: 0.8914175629615784:  91%|█████████ | 5858/6470 [1:10:19<06:09,  1.65it/s]
1|5859|Loss: 0.8914175629615784:  91%|█████████ | 5859/6470 [1:10:20<06:23,  1.59it/s]
1|5860|Loss: 0.8874619603157043:  91%|█████████ | 5859/6470 [1:10:20<06:23,  1.59it/s]
1|5860|Loss: 0.8874619603157043:  91%|█████████ | 5860/6470 [1:10:20<06:27,  1.58it/s]
1|5861|Loss: 0.8584421873092651:  91%|█████████ | 5860/6470 [1:10:21<06:27,  1.58it/s]
1|5861|Loss: 0.8584421873092651:  91%|█████████ | 5861/6470 [1:10:21<06:41,  1.52it/s]
1|5862|Loss: 0.8755143880844116:  91%|█████████ | 5861/6470 [1:10:21<06:41,  1.52it/s]
1|5862|Loss: 0.8755143880844116:  91%|█████████ | 5862/6470 [1:10:22<06:33,  1.55it/s]
1|5863|Loss: 0.8376775979995728:  91%|█████████ | 5862/6470 [1:10:22<06:33,  1.55it/s]
1|5863|Loss: 0.8376775979995728:  91%|█████████ | 5863/6470 [1:10:22<06:25,  1.57it/s]
1|5864|Loss: 0.8256528377532959:  91%|█████████ | 5863/6470 [1:10:22<06:25,  1.57it/s]
1|5864|Loss: 0.8256528377532959:  91%|█████████ | 5864/6470 [1:10:23<06:11,  1.63it/s]
1|5865|Loss: 0.901426374912262:  91%|█████████ | 5864/6470 [1:10:23<06:11,  1.63it/s] 
1|5865|Loss: 0.901426374912262:  91%|█████████ | 5865/6470 [1:10:23<06:28,  1.56it/s]
1|5866|Loss: 0.9469056725502014:  91%|█████████ | 5865/6470 [1:10:24<06:28,  1.56it/s]
1|5866|Loss: 0.9469056725502014:  91%|█████████ | 5866/6470 [1:10:24<06:55,  1.45it/s]
1|5867|Loss: 0.8666154146194458:  91%|█████████ | 5866/6470 [1:10:25<06:55,  1.45it/s]
1|5867|Loss: 0.8666154146194458:  91%|█████████ | 5867/6470 [1:10:25<07:05,  1.42it/s]
1|5868|Loss: 0.9046403765678406:  91%|█████████ | 5867/6470 [1:10:25<07:05,  1.42it/s]
1|5868|Loss: 0.9046403765678406:  91%|█████████ | 5868/6470 [1:10:25<06:36,  1.52it/s]
1|5869|Loss: 1.009285569190979:  91%|█████████ | 5868/6470 [1:10:26<06:36,  1.52it/s] 
1|5869|Loss: 1.009285569190979:  91%|█████████ | 5869/6470 [1:10:26<06:41,  1.50it/s]
1|5870|Loss: 0.7280007600784302:  91%|█████████ | 5869/6470 [1:10:27<06:41,  1.50it/s]
1|5870|Loss: 0.7280007600784302:  91%|█████████ | 5870/6470 [1:10:27<07:10,  1.39it/s]
1|5871|Loss: 0.7697461247444153:  91%|█████████ | 5870/6470 [1:10:27<07:10,  1.39it/s]
1|5871|Loss: 0.7697461247444153:  91%|█████████ | 5871/6470 [1:10:28<07:20,  1.36it/s]
1|5872|Loss: 0.9178997278213501:  91%|█████████ | 5871/6470 [1:10:28<07:20,  1.36it/s]
1|5872|Loss: 0.9178997278213501:  91%|█████████ | 5872/6470 [1:10:28<07:02,  1.41it/s]
1|5873|Loss: 0.6514669060707092:  91%|█████████ | 5872/6470 [1:10:29<07:02,  1.41it/s]
1|5873|Loss: 0.6514669060707092:  91%|█████████ | 5873/6470 [1:10:29<06:48,  1.46it/s]
1|5874|Loss: 0.8160476684570312:  91%|█████████ | 5873/6470 [1:10:29<06:48,  1.46it/s]
1|5874|Loss: 0.8160476684570312:  91%|█████████ | 5874/6470 [1:10:29<05:58,  1.66it/s]
1|5875|Loss: 0.8032687306404114:  91%|█████████ | 5874/6470 [1:10:30<05:58,  1.66it/s]
1|5875|Loss: 0.8032687306404114:  91%|█████████ | 5875/6470 [1:10:30<05:26,  1.82it/s]
1|5876|Loss: 1.0556423664093018:  91%|█████████ | 5875/6470 [1:10:30<05:26,  1.82it/s]
1|5876|Loss: 1.0556423664093018:  91%|█████████ | 5876/6470 [1:10:30<05:21,  1.85it/s]
1|5877|Loss: 0.7744783759117126:  91%|█████████ | 5876/6470 [1:10:31<05:21,  1.85it/s]
1|5877|Loss: 0.7744783759117126:  91%|█████████ | 5877/6470 [1:10:31<06:01,  1.64it/s]
1|5878|Loss: 1.011338233947754:  91%|█████████ | 5877/6470 [1:10:32<06:01,  1.64it/s] 
1|5878|Loss: 1.011338233947754:  91%|█████████ | 5878/6470 [1:10:32<06:39,  1.48it/s]
1|5879|Loss: 0.8457502722740173:  91%|█████████ | 5878/6470 [1:10:32<06:39,  1.48it/s]
1|5879|Loss: 0.8457502722740173:  91%|█████████ | 5879/6470 [1:10:33<07:19,  1.34it/s]
1|5880|Loss: 0.9015862345695496:  91%|█████████ | 5879/6470 [1:10:33<07:19,  1.34it/s]
1|5880|Loss: 0.9015862345695496:  91%|█████████ | 5880/6470 [1:10:34<07:12,  1.37it/s]
1|5881|Loss: 0.9256316423416138:  91%|█████████ | 5880/6470 [1:10:34<07:12,  1.37it/s]
1|5881|Loss: 0.9256316423416138:  91%|█████████ | 5881/6470 [1:10:34<06:45,  1.45it/s]
1|5882|Loss: 0.9966149926185608:  91%|█████████ | 5881/6470 [1:10:35<06:45,  1.45it/s]
1|5882|Loss: 0.9966149926185608:  91%|█████████ | 5882/6470 [1:10:35<06:42,  1.46it/s]
1|5883|Loss: 0.9482852220535278:  91%|█████████ | 5882/6470 [1:10:35<06:42,  1.46it/s]
1|5883|Loss: 0.9482852220535278:  91%|█████████ | 5883/6470 [1:10:35<06:26,  1.52it/s]
1|5884|Loss: 0.769940972328186:  91%|█████████ | 5883/6470 [1:10:36<06:26,  1.52it/s] 
1|5884|Loss: 0.769940972328186:  91%|█████████ | 5884/6470 [1:10:36<06:41,  1.46it/s]
1|5885|Loss: 0.8682019114494324:  91%|█████████ | 5884/6470 [1:10:37<06:41,  1.46it/s]
1|5885|Loss: 0.8682019114494324:  91%|█████████ | 5885/6470 [1:10:37<06:38,  1.47it/s]
1|5886|Loss: 0.8820709586143494:  91%|█████████ | 5885/6470 [1:10:37<06:38,  1.47it/s]
1|5886|Loss: 0.8820709586143494:  91%|█████████ | 5886/6470 [1:10:37<06:21,  1.53it/s]
1|5887|Loss: 0.8110939264297485:  91%|█████████ | 5886/6470 [1:10:38<06:21,  1.53it/s]
1|5887|Loss: 0.8110939264297485:  91%|█████████ | 5887/6470 [1:10:38<06:26,  1.51it/s]
1|5888|Loss: 0.9265823364257812:  91%|█████████ | 5887/6470 [1:10:39<06:26,  1.51it/s]
1|5888|Loss: 0.9265823364257812:  91%|█████████ | 5888/6470 [1:10:39<06:20,  1.53it/s]
1|5889|Loss: 0.7432506084442139:  91%|█████████ | 5888/6470 [1:10:39<06:20,  1.53it/s]
1|5889|Loss: 0.7432506084442139:  91%|█████████ | 5889/6470 [1:10:39<06:12,  1.56it/s]
1|5890|Loss: 0.9662082195281982:  91%|█████████ | 5889/6470 [1:10:40<06:12,  1.56it/s]
1|5890|Loss: 0.9662082195281982:  91%|█████████ | 5890/6470 [1:10:40<05:44,  1.68it/s]
1|5891|Loss: 0.9098268747329712:  91%|█████████ | 5890/6470 [1:10:40<05:44,  1.68it/s]
1|5891|Loss: 0.9098268747329712:  91%|█████████ | 5891/6470 [1:10:41<05:58,  1.62it/s]
1|5892|Loss: 0.8385196328163147:  91%|█████████ | 5891/6470 [1:10:41<05:58,  1.62it/s]
1|5892|Loss: 0.8385196328163147:  91%|█████████ | 5892/6470 [1:10:41<06:18,  1.53it/s]
1|5893|Loss: 0.8886577486991882:  91%|█████████ | 5892/6470 [1:10:42<06:18,  1.53it/s]
1|5893|Loss: 0.8886577486991882:  91%|█████████ | 5893/6470 [1:10:42<06:34,  1.46it/s]
1|5894|Loss: 0.9333469271659851:  91%|█████████ | 5893/6470 [1:10:42<06:34,  1.46it/s]
1|5894|Loss: 0.9333469271659851:  91%|█████████ | 5894/6470 [1:10:43<06:22,  1.51it/s]
1|5895|Loss: 0.9647989273071289:  91%|█████████ | 5894/6470 [1:10:43<06:22,  1.51it/s]
1|5895|Loss: 0.9647989273071289:  91%|█████████ | 5895/6470 [1:10:43<06:36,  1.45it/s]
1|5896|Loss: 0.7951957583427429:  91%|█████████ | 5895/6470 [1:10:44<06:36,  1.45it/s]
1|5896|Loss: 0.7951957583427429:  91%|█████████ | 5896/6470 [1:10:44<06:07,  1.56it/s]
1|5897|Loss: 0.9125171303749084:  91%|█████████ | 5896/6470 [1:10:44<06:07,  1.56it/s]
1|5897|Loss: 0.9125171303749084:  91%|█████████ | 5897/6470 [1:10:45<06:33,  1.46it/s]
1|5898|Loss: 0.7237932085990906:  91%|█████████ | 5897/6470 [1:10:45<06:33,  1.46it/s]
1|5898|Loss: 0.7237932085990906:  91%|█████████ | 5898/6470 [1:10:46<06:49,  1.40it/s]
1|5899|Loss: 0.8988155722618103:  91%|█████████ | 5898/6470 [1:10:46<06:49,  1.40it/s]
1|5899|Loss: 0.8988155722618103:  91%|█████████ | 5899/6470 [1:10:46<06:50,  1.39it/s]
1|5900|Loss: 0.7314512729644775:  91%|█████████ | 5899/6470 [1:10:47<06:50,  1.39it/s]
1|5900|Loss: 0.7314512729644775:  91%|█████████ | 5900/6470 [1:10:47<06:38,  1.43it/s]
1|5901|Loss: 0.8081137537956238:  91%|█████████ | 5900/6470 [1:10:47<06:38,  1.43it/s]
1|5901|Loss: 0.8081137537956238:  91%|█████████ | 5901/6470 [1:10:47<06:02,  1.57it/s]
1|5902|Loss: 1.0177127122879028:  91%|█████████ | 5901/6470 [1:10:48<06:02,  1.57it/s]
1|5902|Loss: 1.0177127122879028:  91%|█████████ | 5902/6470 [1:10:48<05:48,  1.63it/s]
1|5903|Loss: 0.8004328608512878:  91%|█████████ | 5902/6470 [1:10:48<05:48,  1.63it/s]
1|5903|Loss: 0.8004328608512878:  91%|█████████ | 5903/6470 [1:10:49<06:02,  1.56it/s]
1|5904|Loss: 0.8622495532035828:  91%|█████████ | 5903/6470 [1:10:49<06:02,  1.56it/s]
1|5904|Loss: 0.8622495532035828:  91%|█████████▏| 5904/6470 [1:10:49<05:49,  1.62it/s]
1|5905|Loss: 0.8154065608978271:  91%|█████████▏| 5904/6470 [1:10:50<05:49,  1.62it/s]
1|5905|Loss: 0.8154065608978271:  91%|█████████▏| 5905/6470 [1:10:50<05:49,  1.62it/s]
1|5906|Loss: 1.1030429601669312:  91%|█████████▏| 5905/6470 [1:10:50<05:49,  1.62it/s]
1|5906|Loss: 1.1030429601669312:  91%|█████████▏| 5906/6470 [1:10:50<05:35,  1.68it/s]
1|5907|Loss: 0.7620412111282349:  91%|█████████▏| 5906/6470 [1:10:51<05:35,  1.68it/s]
1|5907|Loss: 0.7620412111282349:  91%|█████████▏| 5907/6470 [1:10:51<05:45,  1.63it/s]
1|5908|Loss: 0.9537944793701172:  91%|█████████▏| 5907/6470 [1:10:51<05:45,  1.63it/s]
1|5908|Loss: 0.9537944793701172:  91%|█████████▏| 5908/6470 [1:10:52<05:48,  1.61it/s]
1|5909|Loss: 0.8436739444732666:  91%|█████████▏| 5908/6470 [1:10:52<05:48,  1.61it/s]
1|5909|Loss: 0.8436739444732666:  91%|█████████▏| 5909/6470 [1:10:52<05:47,  1.62it/s]
1|5910|Loss: 0.8268700838088989:  91%|█████████▏| 5909/6470 [1:10:53<05:47,  1.62it/s]
1|5910|Loss: 0.8268700838088989:  91%|█████████▏| 5910/6470 [1:10:53<06:02,  1.55it/s]
1|5911|Loss: 0.850738525390625:  91%|█████████▏| 5910/6470 [1:10:53<06:02,  1.55it/s] 
1|5911|Loss: 0.850738525390625:  91%|█████████▏| 5911/6470 [1:10:54<06:00,  1.55it/s]
1|5912|Loss: 0.9848232865333557:  91%|█████████▏| 5911/6470 [1:10:54<06:00,  1.55it/s]
1|5912|Loss: 0.9848232865333557:  91%|█████████▏| 5912/6470 [1:10:54<05:55,  1.57it/s]
1|5913|Loss: 0.7436897158622742:  91%|█████████▏| 5912/6470 [1:10:55<05:55,  1.57it/s]
1|5913|Loss: 0.7436897158622742:  91%|█████████▏| 5913/6470 [1:10:55<06:15,  1.48it/s]
1|5914|Loss: 0.9844639897346497:  91%|█████████▏| 5913/6470 [1:10:55<06:15,  1.48it/s]
1|5914|Loss: 0.9844639897346497:  91%|█████████▏| 5914/6470 [1:10:56<05:54,  1.57it/s]
1|5915|Loss: 0.7721250653266907:  91%|█████████▏| 5914/6470 [1:10:56<05:54,  1.57it/s]
1|5915|Loss: 0.7721250653266907:  91%|█████████▏| 5915/6470 [1:10:56<06:19,  1.46it/s]
1|5916|Loss: 0.8790200352668762:  91%|█████████▏| 5915/6470 [1:10:57<06:19,  1.46it/s]
1|5916|Loss: 0.8790200352668762:  91%|█████████▏| 5916/6470 [1:10:57<06:35,  1.40it/s]
1|5917|Loss: 0.9554370045661926:  91%|█████████▏| 5916/6470 [1:10:58<06:35,  1.40it/s]
1|5917|Loss: 0.9554370045661926:  91%|█████████▏| 5917/6470 [1:10:58<06:39,  1.38it/s]
1|5918|Loss: 0.9980520606040955:  91%|█████████▏| 5917/6470 [1:10:58<06:39,  1.38it/s]
1|5918|Loss: 0.9980520606040955:  91%|█████████▏| 5918/6470 [1:10:58<06:19,  1.45it/s]
1|5919|Loss: 0.8818650841712952:  91%|█████████▏| 5918/6470 [1:10:59<06:19,  1.45it/s]
1|5919|Loss: 0.8818650841712952:  91%|█████████▏| 5919/6470 [1:10:59<06:37,  1.39it/s]
1|5920|Loss: 0.9389363527297974:  91%|█████████▏| 5919/6470 [1:11:00<06:37,  1.39it/s]
1|5920|Loss: 0.9389363527297974:  91%|█████████▏| 5920/6470 [1:11:00<06:23,  1.44it/s]
1|5921|Loss: 0.9695183038711548:  91%|█████████▏| 5920/6470 [1:11:00<06:23,  1.44it/s]
1|5921|Loss: 0.9695183038711548:  92%|█████████▏| 5921/6470 [1:11:01<06:22,  1.43it/s]
1|5922|Loss: 0.9357810616493225:  92%|█████████▏| 5921/6470 [1:11:01<06:22,  1.43it/s]
1|5922|Loss: 0.9357810616493225:  92%|█████████▏| 5922/6470 [1:11:01<06:07,  1.49it/s]
1|5923|Loss: 0.8959125280380249:  92%|█████████▏| 5922/6470 [1:11:02<06:07,  1.49it/s]
1|5923|Loss: 0.8959125280380249:  92%|█████████▏| 5923/6470 [1:11:02<06:13,  1.46it/s]
1|5924|Loss: 0.8991380929946899:  92%|█████████▏| 5923/6470 [1:11:02<06:13,  1.46it/s]
1|5924|Loss: 0.8991380929946899:  92%|█████████▏| 5924/6470 [1:11:03<06:41,  1.36it/s]
1|5925|Loss: 0.9097520709037781:  92%|█████████▏| 5924/6470 [1:11:03<06:41,  1.36it/s]
1|5925|Loss: 0.9097520709037781:  92%|█████████▏| 5925/6470 [1:11:03<06:31,  1.39it/s]
1|5926|Loss: 0.7329984903335571:  92%|█████████▏| 5925/6470 [1:11:04<06:31,  1.39it/s]
1|5926|Loss: 0.7329984903335571:  92%|█████████▏| 5926/6470 [1:11:04<05:41,  1.59it/s]
1|5927|Loss: 0.7415297031402588:  92%|█████████▏| 5926/6470 [1:11:04<05:41,  1.59it/s]
1|5927|Loss: 0.7415297031402588:  92%|█████████▏| 5927/6470 [1:11:05<06:01,  1.50it/s]
1|5928|Loss: 0.9691221714019775:  92%|█████████▏| 5927/6470 [1:11:05<06:01,  1.50it/s]
1|5928|Loss: 0.9691221714019775:  92%|█████████▏| 5928/6470 [1:11:05<05:57,  1.52it/s]
1|5929|Loss: 0.8729105591773987:  92%|█████████▏| 5928/6470 [1:11:06<05:57,  1.52it/s]
1|5929|Loss: 0.8729105591773987:  92%|█████████▏| 5929/6470 [1:11:06<05:43,  1.57it/s]
1|5930|Loss: 0.7242512702941895:  92%|█████████▏| 5929/6470 [1:11:06<05:43,  1.57it/s]
1|5930|Loss: 0.7242512702941895:  92%|█████████▏| 5930/6470 [1:11:07<05:51,  1.54it/s]
1|5931|Loss: 0.9682654738426208:  92%|█████████▏| 5930/6470 [1:11:07<05:51,  1.54it/s]
1|5931|Loss: 0.9682654738426208:  92%|█████████▏| 5931/6470 [1:11:07<05:59,  1.50it/s]
1|5932|Loss: 0.7665947079658508:  92%|█████████▏| 5931/6470 [1:11:08<05:59,  1.50it/s]
1|5932|Loss: 0.7665947079658508:  92%|█████████▏| 5932/6470 [1:11:08<05:56,  1.51it/s]
1|5933|Loss: 0.9748733043670654:  92%|█████████▏| 5932/6470 [1:11:08<05:56,  1.51it/s]
1|5933|Loss: 0.9748733043670654:  92%|█████████▏| 5933/6470 [1:11:09<06:04,  1.47it/s]
1|5934|Loss: 0.8837670683860779:  92%|█████████▏| 5933/6470 [1:11:09<06:04,  1.47it/s]
1|5934|Loss: 0.8837670683860779:  92%|█████████▏| 5934/6470 [1:11:09<06:21,  1.40it/s]
1|5935|Loss: 0.8532978296279907:  92%|█████████▏| 5934/6470 [1:11:10<06:21,  1.40it/s]
1|5935|Loss: 0.8532978296279907:  92%|█████████▏| 5935/6470 [1:11:10<06:35,  1.35it/s]
1|5936|Loss: 0.9311226606369019:  92%|█████████▏| 5935/6470 [1:11:11<06:35,  1.35it/s]
1|5936|Loss: 0.9311226606369019:  92%|█████████▏| 5936/6470 [1:11:11<06:51,  1.30it/s]
1|5937|Loss: 0.9023812413215637:  92%|█████████▏| 5936/6470 [1:11:11<06:51,  1.30it/s]
1|5937|Loss: 0.9023812413215637:  92%|█████████▏| 5937/6470 [1:11:12<06:38,  1.34it/s]
1|5938|Loss: 0.7462449669837952:  92%|█████████▏| 5937/6470 [1:11:12<06:38,  1.34it/s]
1|5938|Loss: 0.7462449669837952:  92%|█████████▏| 5938/6470 [1:11:12<06:28,  1.37it/s]
1|5939|Loss: 0.8074139356613159:  92%|█████████▏| 5938/6470 [1:11:13<06:28,  1.37it/s]
1|5939|Loss: 0.8074139356613159:  92%|█████████▏| 5939/6470 [1:11:13<06:26,  1.37it/s]
1|5940|Loss: 0.9298887848854065:  92%|█████████▏| 5939/6470 [1:11:14<06:26,  1.37it/s]
1|5940|Loss: 0.9298887848854065:  92%|█████████▏| 5940/6470 [1:11:14<05:59,  1.48it/s]
1|5941|Loss: 0.7294819951057434:  92%|█████████▏| 5940/6470 [1:11:14<05:59,  1.48it/s]
1|5941|Loss: 0.7294819951057434:  92%|█████████▏| 5941/6470 [1:11:14<05:49,  1.51it/s]
1|5942|Loss: 0.7872799038887024:  92%|█████████▏| 5941/6470 [1:11:15<05:49,  1.51it/s]
1|5942|Loss: 0.7872799038887024:  92%|█████████▏| 5942/6470 [1:11:15<06:06,  1.44it/s]
1|5943|Loss: 0.8926282525062561:  92%|█████████▏| 5942/6470 [1:11:16<06:06,  1.44it/s]
1|5943|Loss: 0.8926282525062561:  92%|█████████▏| 5943/6470 [1:11:16<06:14,  1.41it/s]
1|5944|Loss: 0.8804435729980469:  92%|█████████▏| 5943/6470 [1:11:16<06:14,  1.41it/s]
1|5944|Loss: 0.8804435729980469:  92%|█████████▏| 5944/6470 [1:11:17<06:18,  1.39it/s]
1|5945|Loss: 0.8232465386390686:  92%|█████████▏| 5944/6470 [1:11:17<06:18,  1.39it/s]
1|5945|Loss: 0.8232465386390686:  92%|█████████▏| 5945/6470 [1:11:17<06:09,  1.42it/s]
1|5946|Loss: 0.8124077916145325:  92%|█████████▏| 5945/6470 [1:11:18<06:09,  1.42it/s]
1|5946|Loss: 0.8124077916145325:  92%|█████████▏| 5946/6470 [1:11:18<05:51,  1.49it/s]
1|5947|Loss: 0.9743521809577942:  92%|█████████▏| 5946/6470 [1:11:18<05:51,  1.49it/s]
1|5947|Loss: 0.9743521809577942:  92%|█████████▏| 5947/6470 [1:11:19<05:42,  1.53it/s]
1|5948|Loss: 0.8724144101142883:  92%|█████████▏| 5947/6470 [1:11:19<05:42,  1.53it/s]
1|5948|Loss: 0.8724144101142883:  92%|█████████▏| 5948/6470 [1:11:19<05:24,  1.61it/s]
1|5949|Loss: 0.8681336045265198:  92%|█████████▏| 5948/6470 [1:11:19<05:24,  1.61it/s]
1|5949|Loss: 0.8681336045265198:  92%|█████████▏| 5949/6470 [1:11:20<05:20,  1.63it/s]
1|5950|Loss: 0.9567486643791199:  92%|█████████▏| 5949/6470 [1:11:20<05:20,  1.63it/s]
1|5950|Loss: 0.9567486643791199:  92%|█████████▏| 5950/6470 [1:11:20<05:34,  1.56it/s]
1|5951|Loss: 0.7658975720405579:  92%|█████████▏| 5950/6470 [1:11:21<05:34,  1.56it/s]
1|5951|Loss: 0.7658975720405579:  92%|█████████▏| 5951/6470 [1:11:21<05:35,  1.55it/s]
1|5952|Loss: 0.87016361951828:  92%|█████████▏| 5951/6470 [1:11:21<05:35,  1.55it/s]  
1|5952|Loss: 0.87016361951828:  92%|█████████▏| 5952/6470 [1:11:22<05:21,  1.61it/s]
1|5953|Loss: 0.9438866376876831:  92%|█████████▏| 5952/6470 [1:11:22<05:21,  1.61it/s]
1|5953|Loss: 0.9438866376876831:  92%|█████████▏| 5953/6470 [1:11:22<05:34,  1.54it/s]
1|5954|Loss: 0.9001178741455078:  92%|█████████▏| 5953/6470 [1:11:23<05:34,  1.54it/s]
1|5954|Loss: 0.9001178741455078:  92%|█████████▏| 5954/6470 [1:11:23<05:37,  1.53it/s]
1|5955|Loss: 0.8772068023681641:  92%|█████████▏| 5954/6470 [1:11:23<05:37,  1.53it/s]
1|5955|Loss: 0.8772068023681641:  92%|█████████▏| 5955/6470 [1:11:24<05:47,  1.48it/s]
1|5956|Loss: 1.0843253135681152:  92%|█████████▏| 5955/6470 [1:11:24<05:47,  1.48it/s]
1|5956|Loss: 1.0843253135681152:  92%|█████████▏| 5956/6470 [1:11:24<06:03,  1.41it/s]
1|5957|Loss: 0.886254072189331:  92%|█████████▏| 5956/6470 [1:11:25<06:03,  1.41it/s] 
1|5957|Loss: 0.886254072189331:  92%|█████████▏| 5957/6470 [1:11:25<06:20,  1.35it/s]
1|5958|Loss: 0.8232383728027344:  92%|█████████▏| 5957/6470 [1:11:26<06:20,  1.35it/s]
1|5958|Loss: 0.8232383728027344:  92%|█████████▏| 5958/6470 [1:11:26<05:59,  1.42it/s]
1|5959|Loss: 0.9535110592842102:  92%|█████████▏| 5958/6470 [1:11:26<05:59,  1.42it/s]
1|5959|Loss: 0.9535110592842102:  92%|█████████▏| 5959/6470 [1:11:27<05:44,  1.48it/s]
1|5960|Loss: 0.9017998576164246:  92%|█████████▏| 5959/6470 [1:11:27<05:44,  1.48it/s]
1|5960|Loss: 0.9017998576164246:  92%|█████████▏| 5960/6470 [1:11:27<05:46,  1.47it/s]
1|5961|Loss: 0.7904704809188843:  92%|█████████▏| 5960/6470 [1:11:28<05:46,  1.47it/s]
1|5961|Loss: 0.7904704809188843:  92%|█████████▏| 5961/6470 [1:11:28<06:04,  1.40it/s]
1|5962|Loss: 0.8297238945960999:  92%|█████████▏| 5961/6470 [1:11:28<06:04,  1.40it/s]
1|5962|Loss: 0.8297238945960999:  92%|█████████▏| 5962/6470 [1:11:29<05:44,  1.47it/s]
1|5963|Loss: 0.891852080821991:  92%|█████████▏| 5962/6470 [1:11:29<05:44,  1.47it/s] 
1|5963|Loss: 0.891852080821991:  92%|█████████▏| 5963/6470 [1:11:29<05:42,  1.48it/s]
1|5964|Loss: 0.7814680337905884:  92%|█████████▏| 5963/6470 [1:11:30<05:42,  1.48it/s]
1|5964|Loss: 0.7814680337905884:  92%|█████████▏| 5964/6470 [1:11:30<05:52,  1.44it/s]
1|5965|Loss: 0.9258238673210144:  92%|█████████▏| 5964/6470 [1:11:30<05:52,  1.44it/s]
1|5965|Loss: 0.9258238673210144:  92%|█████████▏| 5965/6470 [1:11:31<05:54,  1.42it/s]
1|5966|Loss: 1.0477125644683838:  92%|█████████▏| 5965/6470 [1:11:31<05:54,  1.42it/s]
1|5966|Loss: 1.0477125644683838:  92%|█████████▏| 5966/6470 [1:11:31<05:43,  1.47it/s]
1|5967|Loss: 0.7923480868339539:  92%|█████████▏| 5966/6470 [1:11:32<05:43,  1.47it/s]
1|5967|Loss: 0.7923480868339539:  92%|█████████▏| 5967/6470 [1:11:32<06:06,  1.37it/s]
1|5968|Loss: 0.9059227108955383:  92%|█████████▏| 5967/6470 [1:11:33<06:06,  1.37it/s]
1|5968|Loss: 0.9059227108955383:  92%|█████████▏| 5968/6470 [1:11:33<06:12,  1.35it/s]
1|5969|Loss: 0.9771859645843506:  92%|█████████▏| 5968/6470 [1:11:33<06:12,  1.35it/s]
1|5969|Loss: 0.9771859645843506:  92%|█████████▏| 5969/6470 [1:11:34<05:59,  1.40it/s]
1|5970|Loss: 0.8618882894515991:  92%|█████████▏| 5969/6470 [1:11:34<05:59,  1.40it/s]
1|5970|Loss: 0.8618882894515991:  92%|█████████▏| 5970/6470 [1:11:34<05:59,  1.39it/s]
1|5971|Loss: 0.9989454746246338:  92%|█████████▏| 5970/6470 [1:11:35<05:59,  1.39it/s]
1|5971|Loss: 0.9989454746246338:  92%|█████████▏| 5971/6470 [1:11:35<05:58,  1.39it/s]
1|5972|Loss: 0.8865607380867004:  92%|█████████▏| 5971/6470 [1:11:35<05:58,  1.39it/s]
1|5972|Loss: 0.8865607380867004:  92%|█████████▏| 5972/6470 [1:11:36<05:55,  1.40it/s]
1|5973|Loss: 0.8836011290550232:  92%|█████████▏| 5972/6470 [1:11:36<05:55,  1.40it/s]
1|5973|Loss: 0.8836011290550232:  92%|█████████▏| 5973/6470 [1:11:36<05:45,  1.44it/s]
1|5974|Loss: 0.7780982851982117:  92%|█████████▏| 5973/6470 [1:11:37<05:45,  1.44it/s]
1|5974|Loss: 0.7780982851982117:  92%|█████████▏| 5974/6470 [1:11:37<05:35,  1.48it/s]
1|5975|Loss: 0.8293865919113159:  92%|█████████▏| 5974/6470 [1:11:37<05:35,  1.48it/s]
1|5975|Loss: 0.8293865919113159:  92%|█████████▏| 5975/6470 [1:11:38<05:16,  1.57it/s]
1|5976|Loss: 0.9001176357269287:  92%|█████████▏| 5975/6470 [1:11:38<05:16,  1.57it/s]
1|5976|Loss: 0.9001176357269287:  92%|█████████▏| 5976/6470 [1:11:38<05:19,  1.54it/s]
1|5977|Loss: 1.053829312324524:  92%|█████████▏| 5976/6470 [1:11:39<05:19,  1.54it/s] 
1|5977|Loss: 1.053829312324524:  92%|█████████▏| 5977/6470 [1:11:39<05:32,  1.48it/s]
1|5978|Loss: 1.003678321838379:  92%|█████████▏| 5977/6470 [1:11:39<05:32,  1.48it/s]
1|5978|Loss: 1.003678321838379:  92%|█████████▏| 5978/6470 [1:11:40<05:30,  1.49it/s]
1|5979|Loss: 0.907945990562439:  92%|█████████▏| 5978/6470 [1:11:40<05:30,  1.49it/s]
1|5979|Loss: 0.907945990562439:  92%|█████████▏| 5979/6470 [1:11:40<05:45,  1.42it/s]
1|5980|Loss: 0.8871864080429077:  92%|█████████▏| 5979/6470 [1:11:41<05:45,  1.42it/s]
1|5980|Loss: 0.8871864080429077:  92%|█████████▏| 5980/6470 [1:11:41<05:38,  1.45it/s]
1|5981|Loss: 0.8793598413467407:  92%|█████████▏| 5980/6470 [1:11:41<05:38,  1.45it/s]
1|5981|Loss: 0.8793598413467407:  92%|█████████▏| 5981/6470 [1:11:42<05:29,  1.48it/s]
1|5982|Loss: 0.8508777022361755:  92%|█████████▏| 5981/6470 [1:11:42<05:29,  1.48it/s]
1|5982|Loss: 0.8508777022361755:  92%|█████████▏| 5982/6470 [1:11:42<05:11,  1.57it/s]
1|5983|Loss: 0.9640392065048218:  92%|█████████▏| 5982/6470 [1:11:43<05:11,  1.57it/s]
1|5983|Loss: 0.9640392065048218:  92%|█████████▏| 5983/6470 [1:11:43<05:20,  1.52it/s]
1|5984|Loss: 0.7936012744903564:  92%|█████████▏| 5983/6470 [1:11:43<05:20,  1.52it/s]
1|5984|Loss: 0.7936012744903564:  92%|█████████▏| 5984/6470 [1:11:44<05:38,  1.44it/s]
1|5985|Loss: 0.9472427368164062:  92%|█████████▏| 5984/6470 [1:11:44<05:38,  1.44it/s]
1|5985|Loss: 0.9472427368164062:  93%|█████████▎| 5985/6470 [1:11:44<05:16,  1.53it/s]
1|5986|Loss: 0.9652127027511597:  93%|█████████▎| 5985/6470 [1:11:45<05:16,  1.53it/s]
1|5986|Loss: 0.9652127027511597:  93%|█████████▎| 5986/6470 [1:11:45<04:48,  1.68it/s]
1|5987|Loss: 0.8594562411308289:  93%|█████████▎| 5986/6470 [1:11:45<04:48,  1.68it/s]
1|5987|Loss: 0.8594562411308289:  93%|█████████▎| 5987/6470 [1:11:45<04:58,  1.62it/s]
1|5988|Loss: 0.986650288105011:  93%|█████████▎| 5987/6470 [1:11:46<04:58,  1.62it/s] 
1|5988|Loss: 0.986650288105011:  93%|█████████▎| 5988/6470 [1:11:46<05:01,  1.60it/s]
1|5989|Loss: 0.747902512550354:  93%|█████████▎| 5988/6470 [1:11:47<05:01,  1.60it/s]
1|5989|Loss: 0.747902512550354:  93%|█████████▎| 5989/6470 [1:11:47<05:43,  1.40it/s]
1|5990|Loss: 0.9758610725402832:  93%|█████████▎| 5989/6470 [1:11:47<05:43,  1.40it/s]
1|5990|Loss: 0.9758610725402832:  93%|█████████▎| 5990/6470 [1:11:48<05:35,  1.43it/s]
1|5991|Loss: 0.920428454875946:  93%|█████████▎| 5990/6470 [1:11:48<05:35,  1.43it/s] 
1|5991|Loss: 0.920428454875946:  93%|█████████▎| 5991/6470 [1:11:49<05:53,  1.35it/s]
1|5992|Loss: 0.9217565655708313:  93%|█████████▎| 5991/6470 [1:11:49<05:53,  1.35it/s]
1|5992|Loss: 0.9217565655708313:  93%|█████████▎| 5992/6470 [1:11:49<05:53,  1.35it/s]
1|5993|Loss: 0.9483889937400818:  93%|█████████▎| 5992/6470 [1:11:50<05:53,  1.35it/s]
1|5993|Loss: 0.9483889937400818:  93%|█████████▎| 5993/6470 [1:11:50<05:59,  1.33it/s]
1|5994|Loss: 0.7764500975608826:  93%|█████████▎| 5993/6470 [1:11:50<05:59,  1.33it/s]
1|5994|Loss: 0.7764500975608826:  93%|█████████▎| 5994/6470 [1:11:51<05:16,  1.51it/s]
1|5995|Loss: 0.9283526539802551:  93%|█████████▎| 5994/6470 [1:11:51<05:16,  1.51it/s]
1|5995|Loss: 0.9283526539802551:  93%|█████████▎| 5995/6470 [1:11:51<05:11,  1.53it/s]
1|5996|Loss: 0.9348987936973572:  93%|█████████▎| 5995/6470 [1:11:52<05:11,  1.53it/s]
1|5996|Loss: 0.9348987936973572:  93%|█████████▎| 5996/6470 [1:11:52<05:10,  1.53it/s]
1|5997|Loss: 0.9495856761932373:  93%|█████████▎| 5996/6470 [1:11:52<05:10,  1.53it/s]
1|5997|Loss: 0.9495856761932373:  93%|█████████▎| 5997/6470 [1:11:52<05:14,  1.50it/s]
1|5998|Loss: 0.8039921522140503:  93%|█████████▎| 5997/6470 [1:11:53<05:14,  1.50it/s]
1|5998|Loss: 0.8039921522140503:  93%|█████████▎| 5998/6470 [1:11:53<05:53,  1.34it/s]
1|5999|Loss: 0.9675999283790588:  93%|█████████▎| 5998/6470 [1:11:54<05:53,  1.34it/s]
1|5999|Loss: 0.9675999283790588:  93%|█████████▎| 5999/6470 [1:11:54<06:13,  1.26it/s]
1|6000|Loss: 0.7779848575592041:  93%|█████████▎| 5999/6470 [1:11:55<06:13,  1.26it/s]
1|6000|Loss: 0.7779848575592041:  93%|█████████▎| 6000/6470 [1:11:55<06:08,  1.28it/s]
1|6001|Loss: 0.7904583811759949:  93%|█████████▎| 6000/6470 [1:11:55<06:08,  1.28it/s]
1|6001|Loss: 0.7904583811759949:  93%|█████████▎| 6001/6470 [1:11:56<05:47,  1.35it/s]
1|6002|Loss: 0.8855395317077637:  93%|█████████▎| 6001/6470 [1:11:56<05:47,  1.35it/s]
1|6002|Loss: 0.8855395317077637:  93%|█████████▎| 6002/6470 [1:11:56<05:44,  1.36it/s]
1|6003|Loss: 0.8785482048988342:  93%|█████████▎| 6002/6470 [1:11:57<05:44,  1.36it/s]
1|6003|Loss: 0.8785482048988342:  93%|█████████▎| 6003/6470 [1:11:57<05:46,  1.35it/s]
1|6004|Loss: 0.7152358293533325:  93%|█████████▎| 6003/6470 [1:11:58<05:46,  1.35it/s]
1|6004|Loss: 0.7152358293533325:  93%|█████████▎| 6004/6470 [1:11:58<05:45,  1.35it/s]
1|6005|Loss: 0.9397335052490234:  93%|█████████▎| 6004/6470 [1:11:58<05:45,  1.35it/s]
1|6005|Loss: 0.9397335052490234:  93%|█████████▎| 6005/6470 [1:11:59<05:26,  1.43it/s]
1|6006|Loss: 0.9955875873565674:  93%|█████████▎| 6005/6470 [1:11:59<05:26,  1.43it/s]
1|6006|Loss: 0.9955875873565674:  93%|█████████▎| 6006/6470 [1:11:59<05:02,  1.53it/s]
1|6007|Loss: 0.8302700519561768:  93%|█████████▎| 6006/6470 [1:11:59<05:02,  1.53it/s]
1|6007|Loss: 0.8302700519561768:  93%|█████████▎| 6007/6470 [1:12:00<04:54,  1.57it/s]
1|6008|Loss: 0.9138056635856628:  93%|█████████▎| 6007/6470 [1:12:00<04:54,  1.57it/s]
1|6008|Loss: 0.9138056635856628:  93%|█████████▎| 6008/6470 [1:12:00<04:51,  1.59it/s]
1|6009|Loss: 1.0524845123291016:  93%|█████████▎| 6008/6470 [1:12:01<04:51,  1.59it/s]
1|6009|Loss: 1.0524845123291016:  93%|█████████▎| 6009/6470 [1:12:01<04:44,  1.62it/s]
1|6010|Loss: 1.0264089107513428:  93%|█████████▎| 6009/6470 [1:12:01<04:44,  1.62it/s]
1|6010|Loss: 1.0264089107513428:  93%|█████████▎| 6010/6470 [1:12:02<04:45,  1.61it/s]
1|6011|Loss: 0.7415069341659546:  93%|█████████▎| 6010/6470 [1:12:02<04:45,  1.61it/s]
1|6011|Loss: 0.7415069341659546:  93%|█████████▎| 6011/6470 [1:12:02<05:00,  1.53it/s]
1|6012|Loss: 1.0731732845306396:  93%|█████████▎| 6011/6470 [1:12:03<05:00,  1.53it/s]
1|6012|Loss: 1.0731732845306396:  93%|█████████▎| 6012/6470 [1:12:03<05:01,  1.52it/s]
1|6013|Loss: 0.9698852896690369:  93%|█████████▎| 6012/6470 [1:12:03<05:01,  1.52it/s]
1|6013|Loss: 0.9698852896690369:  93%|█████████▎| 6013/6470 [1:12:04<04:58,  1.53it/s]
1|6014|Loss: 0.9743682742118835:  93%|█████████▎| 6013/6470 [1:12:04<04:58,  1.53it/s]
1|6014|Loss: 0.9743682742118835:  93%|█████████▎| 6014/6470 [1:12:04<04:56,  1.54it/s]
1|6015|Loss: 0.9785534143447876:  93%|█████████▎| 6014/6470 [1:12:05<04:56,  1.54it/s]
1|6015|Loss: 0.9785534143447876:  93%|█████████▎| 6015/6470 [1:12:05<04:54,  1.54it/s]
1|6016|Loss: 0.7380092144012451:  93%|█████████▎| 6015/6470 [1:12:05<04:54,  1.54it/s]
1|6016|Loss: 0.7380092144012451:  93%|█████████▎| 6016/6470 [1:12:06<05:13,  1.45it/s]
1|6017|Loss: 0.933198869228363:  93%|█████████▎| 6016/6470 [1:12:06<05:13,  1.45it/s] 
1|6017|Loss: 0.933198869228363:  93%|█████████▎| 6017/6470 [1:12:07<05:35,  1.35it/s]
1|6018|Loss: 0.9129289984703064:  93%|█████████▎| 6017/6470 [1:12:07<05:35,  1.35it/s]
1|6018|Loss: 0.9129289984703064:  93%|█████████▎| 6018/6470 [1:12:07<05:45,  1.31it/s]
1|6019|Loss: 0.8090026378631592:  93%|█████████▎| 6018/6470 [1:12:08<05:45,  1.31it/s]
1|6019|Loss: 0.8090026378631592:  93%|█████████▎| 6019/6470 [1:12:08<05:39,  1.33it/s]
1|6020|Loss: 0.8621624708175659:  93%|█████████▎| 6019/6470 [1:12:08<05:39,  1.33it/s]
1|6020|Loss: 0.8621624708175659:  93%|█████████▎| 6020/6470 [1:12:09<05:36,  1.34it/s]
1|6021|Loss: 0.9137037992477417:  93%|█████████▎| 6020/6470 [1:12:09<05:36,  1.34it/s]
1|6021|Loss: 0.9137037992477417:  93%|█████████▎| 6021/6470 [1:12:09<05:21,  1.39it/s]
1|6022|Loss: 0.9210100173950195:  93%|█████████▎| 6021/6470 [1:12:10<05:21,  1.39it/s]
1|6022|Loss: 0.9210100173950195:  93%|█████████▎| 6022/6470 [1:12:10<05:30,  1.36it/s]
1|6023|Loss: 0.9484785199165344:  93%|█████████▎| 6022/6470 [1:12:11<05:30,  1.36it/s]
1|6023|Loss: 0.9484785199165344:  93%|█████████▎| 6023/6470 [1:12:11<05:38,  1.32it/s]
1|6024|Loss: 0.9037543535232544:  93%|█████████▎| 6023/6470 [1:12:11<05:38,  1.32it/s]
1|6024|Loss: 0.9037543535232544:  93%|█████████▎| 6024/6470 [1:12:12<05:38,  1.32it/s]
1|6025|Loss: 0.8945208191871643:  93%|█████████▎| 6024/6470 [1:12:12<05:38,  1.32it/s]
1|6025|Loss: 0.8945208191871643:  93%|█████████▎| 6025/6470 [1:12:12<05:22,  1.38it/s]
1|6026|Loss: 0.9040408730506897:  93%|█████████▎| 6025/6470 [1:12:13<05:22,  1.38it/s]
1|6026|Loss: 0.9040408730506897:  93%|█████████▎| 6026/6470 [1:12:13<05:16,  1.40it/s]
1|6027|Loss: 0.7470117807388306:  93%|█████████▎| 6026/6470 [1:12:13<05:16,  1.40it/s]
1|6027|Loss: 0.7470117807388306:  93%|█████████▎| 6027/6470 [1:12:14<04:55,  1.50it/s]
1|6028|Loss: 0.7251259684562683:  93%|█████████▎| 6027/6470 [1:12:14<04:55,  1.50it/s]
1|6028|Loss: 0.7251259684562683:  93%|█████████▎| 6028/6470 [1:12:14<05:13,  1.41it/s]
1|6029|Loss: 0.9061903357505798:  93%|█████████▎| 6028/6470 [1:12:15<05:13,  1.41it/s]
1|6029|Loss: 0.9061903357505798:  93%|█████████▎| 6029/6470 [1:12:15<05:11,  1.42it/s]
1|6030|Loss: 0.8043456673622131:  93%|█████████▎| 6029/6470 [1:12:16<05:11,  1.42it/s]
1|6030|Loss: 0.8043456673622131:  93%|█████████▎| 6030/6470 [1:12:16<05:00,  1.46it/s]
1|6031|Loss: 0.958871603012085:  93%|█████████▎| 6030/6470 [1:12:16<05:00,  1.46it/s] 
1|6031|Loss: 0.958871603012085:  93%|█████████▎| 6031/6470 [1:12:17<05:04,  1.44it/s]
1|6032|Loss: 0.8635683655738831:  93%|█████████▎| 6031/6470 [1:12:17<05:04,  1.44it/s]
1|6032|Loss: 0.8635683655738831:  93%|█████████▎| 6032/6470 [1:12:17<05:10,  1.41it/s]
1|6033|Loss: 0.7686213254928589:  93%|█████████▎| 6032/6470 [1:12:18<05:10,  1.41it/s]
1|6033|Loss: 0.7686213254928589:  93%|█████████▎| 6033/6470 [1:12:18<05:12,  1.40it/s]
1|6034|Loss: 0.913292646408081:  93%|█████████▎| 6033/6470 [1:12:18<05:12,  1.40it/s] 
1|6034|Loss: 0.913292646408081:  93%|█████████▎| 6034/6470 [1:12:19<05:02,  1.44it/s]
1|6035|Loss: 0.7967106699943542:  93%|█████████▎| 6034/6470 [1:12:19<05:02,  1.44it/s]
1|6035|Loss: 0.7967106699943542:  93%|█████████▎| 6035/6470 [1:12:19<04:43,  1.54it/s]
1|6036|Loss: 0.8400903940200806:  93%|█████████▎| 6035/6470 [1:12:19<04:43,  1.54it/s]
1|6036|Loss: 0.8400903940200806:  93%|█████████▎| 6036/6470 [1:12:20<04:18,  1.68it/s]
1|6037|Loss: 0.8965318202972412:  93%|█████████▎| 6036/6470 [1:12:20<04:18,  1.68it/s]
1|6037|Loss: 0.8965318202972412:  93%|█████████▎| 6037/6470 [1:12:20<04:24,  1.64it/s]
1|6038|Loss: 0.8300596475601196:  93%|█████████▎| 6037/6470 [1:12:21<04:24,  1.64it/s]
1|6038|Loss: 0.8300596475601196:  93%|█████████▎| 6038/6470 [1:12:21<04:35,  1.57it/s]
1|6039|Loss: 0.9736506938934326:  93%|█████████▎| 6038/6470 [1:12:21<04:35,  1.57it/s]
1|6039|Loss: 0.9736506938934326:  93%|█████████▎| 6039/6470 [1:12:22<04:43,  1.52it/s]
1|6040|Loss: 0.8938031196594238:  93%|█████████▎| 6039/6470 [1:12:22<04:43,  1.52it/s]
1|6040|Loss: 0.8938031196594238:  93%|█████████▎| 6040/6470 [1:12:22<04:27,  1.61it/s]
1|6041|Loss: 0.8001076579093933:  93%|█████████▎| 6040/6470 [1:12:23<04:27,  1.61it/s]
1|6041|Loss: 0.8001076579093933:  93%|█████████▎| 6041/6470 [1:12:23<04:19,  1.65it/s]
1|6042|Loss: 0.8577454090118408:  93%|█████████▎| 6041/6470 [1:12:23<04:19,  1.65it/s]
1|6042|Loss: 0.8577454090118408:  93%|█████████▎| 6042/6470 [1:12:24<04:32,  1.57it/s]
1|6043|Loss: 0.8258599042892456:  93%|█████████▎| 6042/6470 [1:12:24<04:32,  1.57it/s]
1|6043|Loss: 0.8258599042892456:  93%|█████████▎| 6043/6470 [1:12:24<04:46,  1.49it/s]
1|6044|Loss: 0.8736798167228699:  93%|█████████▎| 6043/6470 [1:12:25<04:46,  1.49it/s]
1|6044|Loss: 0.8736798167228699:  93%|█████████▎| 6044/6470 [1:12:25<05:00,  1.42it/s]
1|6045|Loss: 0.7983571887016296:  93%|█████████▎| 6044/6470 [1:12:25<05:00,  1.42it/s]
1|6045|Loss: 0.7983571887016296:  93%|█████████▎| 6045/6470 [1:12:26<05:04,  1.40it/s]
1|6046|Loss: 0.8630751371383667:  93%|█████████▎| 6045/6470 [1:12:26<05:04,  1.40it/s]
1|6046|Loss: 0.8630751371383667:  93%|█████████▎| 6046/6470 [1:12:26<04:56,  1.43it/s]
1|6047|Loss: 0.9535601139068604:  93%|█████████▎| 6046/6470 [1:12:27<04:56,  1.43it/s]
1|6047|Loss: 0.9535601139068604:  93%|█████████▎| 6047/6470 [1:12:27<04:55,  1.43it/s]
1|6048|Loss: 0.8226783871650696:  93%|█████████▎| 6047/6470 [1:12:28<04:55,  1.43it/s]
1|6048|Loss: 0.8226783871650696:  93%|█████████▎| 6048/6470 [1:12:28<05:12,  1.35it/s]
1|6049|Loss: 0.8238790035247803:  93%|█████████▎| 6048/6470 [1:12:28<05:12,  1.35it/s]
1|6049|Loss: 0.8238790035247803:  93%|█████████▎| 6049/6470 [1:12:29<05:16,  1.33it/s]
1|6050|Loss: 0.8803542256355286:  93%|█████████▎| 6049/6470 [1:12:29<05:16,  1.33it/s]
1|6050|Loss: 0.8803542256355286:  94%|█████████▎| 6050/6470 [1:12:30<05:11,  1.35it/s]
1|6051|Loss: 0.6845608949661255:  94%|█████████▎| 6050/6470 [1:12:30<05:11,  1.35it/s]
1|6051|Loss: 0.6845608949661255:  94%|█████████▎| 6051/6470 [1:12:30<04:40,  1.49it/s]
1|6052|Loss: 0.8999754786491394:  94%|█████████▎| 6051/6470 [1:12:30<04:40,  1.49it/s]
1|6052|Loss: 0.8999754786491394:  94%|█████████▎| 6052/6470 [1:12:31<04:23,  1.58it/s]
1|6053|Loss: 0.9477950930595398:  94%|█████████▎| 6052/6470 [1:12:31<04:23,  1.58it/s]
1|6053|Loss: 0.9477950930595398:  94%|█████████▎| 6053/6470 [1:12:31<04:22,  1.59it/s]
1|6054|Loss: 0.804634153842926:  94%|█████████▎| 6053/6470 [1:12:32<04:22,  1.59it/s] 
1|6054|Loss: 0.804634153842926:  94%|█████████▎| 6054/6470 [1:12:32<04:20,  1.59it/s]
1|6055|Loss: 0.8369730114936829:  94%|█████████▎| 6054/6470 [1:12:32<04:20,  1.59it/s]
1|6055|Loss: 0.8369730114936829:  94%|█████████▎| 6055/6470 [1:12:32<04:19,  1.60it/s]
1|6056|Loss: 0.9795708656311035:  94%|█████████▎| 6055/6470 [1:12:33<04:19,  1.60it/s]
1|6056|Loss: 0.9795708656311035:  94%|█████████▎| 6056/6470 [1:12:33<04:26,  1.55it/s]
1|6057|Loss: 0.8289632797241211:  94%|█████████▎| 6056/6470 [1:12:33<04:26,  1.55it/s]
1|6057|Loss: 0.8289632797241211:  94%|█████████▎| 6057/6470 [1:12:34<04:35,  1.50it/s]
1|6058|Loss: 0.7477396130561829:  94%|█████████▎| 6057/6470 [1:12:34<04:35,  1.50it/s]
1|6058|Loss: 0.7477396130561829:  94%|█████████▎| 6058/6470 [1:12:34<04:31,  1.52it/s]
1|6059|Loss: 0.8908456563949585:  94%|█████████▎| 6058/6470 [1:12:35<04:31,  1.52it/s]
1|6059|Loss: 0.8908456563949585:  94%|█████████▎| 6059/6470 [1:12:35<04:33,  1.50it/s]
1|6060|Loss: 0.8860653042793274:  94%|█████████▎| 6059/6470 [1:12:36<04:33,  1.50it/s]
1|6060|Loss: 0.8860653042793274:  94%|█████████▎| 6060/6470 [1:12:36<04:52,  1.40it/s]
1|6061|Loss: 0.8266409039497375:  94%|█████████▎| 6060/6470 [1:12:36<04:52,  1.40it/s]
1|6061|Loss: 0.8266409039497375:  94%|█████████▎| 6061/6470 [1:12:36<04:21,  1.57it/s]
1|6062|Loss: 0.9161394834518433:  94%|█████████▎| 6061/6470 [1:12:37<04:21,  1.57it/s]
1|6062|Loss: 0.9161394834518433:  94%|█████████▎| 6062/6470 [1:12:37<04:28,  1.52it/s]
1|6063|Loss: 0.9310764670372009:  94%|█████████▎| 6062/6470 [1:12:37<04:28,  1.52it/s]
1|6063|Loss: 0.9310764670372009:  94%|█████████▎| 6063/6470 [1:12:38<04:12,  1.61it/s]
1|6064|Loss: 0.8145059943199158:  94%|█████████▎| 6063/6470 [1:12:38<04:12,  1.61it/s]
1|6064|Loss: 0.8145059943199158:  94%|█████████▎| 6064/6470 [1:12:38<04:07,  1.64it/s]
1|6065|Loss: 0.9529147148132324:  94%|█████████▎| 6064/6470 [1:12:39<04:07,  1.64it/s]
1|6065|Loss: 0.9529147148132324:  94%|█████████▎| 6065/6470 [1:12:39<04:10,  1.61it/s]
1|6066|Loss: 0.850132167339325:  94%|█████████▎| 6065/6470 [1:12:39<04:10,  1.61it/s] 
1|6066|Loss: 0.850132167339325:  94%|█████████▍| 6066/6470 [1:12:40<04:10,  1.61it/s]
1|6067|Loss: 0.8008447289466858:  94%|█████████▍| 6066/6470 [1:12:40<04:10,  1.61it/s]
1|6067|Loss: 0.8008447289466858:  94%|█████████▍| 6067/6470 [1:12:40<04:09,  1.62it/s]
1|6068|Loss: 0.7993484735488892:  94%|█████████▍| 6067/6470 [1:12:41<04:09,  1.62it/s]
1|6068|Loss: 0.7993484735488892:  94%|█████████▍| 6068/6470 [1:12:41<04:16,  1.57it/s]
1|6069|Loss: 0.9285569787025452:  94%|█████████▍| 6068/6470 [1:12:41<04:16,  1.57it/s]
1|6069|Loss: 0.9285569787025452:  94%|█████████▍| 6069/6470 [1:12:41<04:18,  1.55it/s]
1|6070|Loss: 0.7172437906265259:  94%|█████████▍| 6069/6470 [1:12:42<04:18,  1.55it/s]
1|6070|Loss: 0.7172437906265259:  94%|█████████▍| 6070/6470 [1:12:42<04:11,  1.59it/s]
1|6071|Loss: 0.7824311852455139:  94%|█████████▍| 6070/6470 [1:12:42<04:11,  1.59it/s]
1|6071|Loss: 0.7824311852455139:  94%|█████████▍| 6071/6470 [1:12:43<03:59,  1.66it/s]
1|6072|Loss: 0.8178738355636597:  94%|█████████▍| 6071/6470 [1:12:43<03:59,  1.66it/s]
1|6072|Loss: 0.8178738355636597:  94%|█████████▍| 6072/6470 [1:12:43<04:01,  1.65it/s]
1|6073|Loss: 0.9437791705131531:  94%|█████████▍| 6072/6470 [1:12:44<04:01,  1.65it/s]
1|6073|Loss: 0.9437791705131531:  94%|█████████▍| 6073/6470 [1:12:44<03:56,  1.68it/s]
1|6074|Loss: 0.9938421845436096:  94%|█████████▍| 6073/6470 [1:12:44<03:56,  1.68it/s]
1|6074|Loss: 0.9938421845436096:  94%|█████████▍| 6074/6470 [1:12:44<03:53,  1.69it/s]
1|6075|Loss: 0.8025211691856384:  94%|█████████▍| 6074/6470 [1:12:45<03:53,  1.69it/s]
1|6075|Loss: 0.8025211691856384:  94%|█████████▍| 6075/6470 [1:12:45<03:51,  1.71it/s]
1|6076|Loss: 0.8621501326560974:  94%|█████████▍| 6075/6470 [1:12:45<03:51,  1.71it/s]
1|6076|Loss: 0.8621501326560974:  94%|█████████▍| 6076/6470 [1:12:46<04:14,  1.55it/s]
1|6077|Loss: 0.9382591843605042:  94%|█████████▍| 6076/6470 [1:12:46<04:14,  1.55it/s]
1|6077|Loss: 0.9382591843605042:  94%|█████████▍| 6077/6470 [1:12:46<04:18,  1.52it/s]
1|6078|Loss: 0.9153615236282349:  94%|█████████▍| 6077/6470 [1:12:47<04:18,  1.52it/s]
1|6078|Loss: 0.9153615236282349:  94%|█████████▍| 6078/6470 [1:12:47<04:21,  1.50it/s]
1|6079|Loss: 0.8274365067481995:  94%|█████████▍| 6078/6470 [1:12:47<04:21,  1.50it/s]
1|6079|Loss: 0.8274365067481995:  94%|█████████▍| 6079/6470 [1:12:48<04:27,  1.46it/s]
1|6080|Loss: 0.8280414342880249:  94%|█████████▍| 6079/6470 [1:12:48<04:27,  1.46it/s]
1|6080|Loss: 0.8280414342880249:  94%|█████████▍| 6080/6470 [1:12:48<04:21,  1.49it/s]
1|6081|Loss: 0.8417193293571472:  94%|█████████▍| 6080/6470 [1:12:49<04:21,  1.49it/s]
1|6081|Loss: 0.8417193293571472:  94%|█████████▍| 6081/6470 [1:12:49<04:24,  1.47it/s]
1|6082|Loss: 0.8155561089515686:  94%|█████████▍| 6081/6470 [1:12:50<04:24,  1.47it/s]
1|6082|Loss: 0.8155561089515686:  94%|█████████▍| 6082/6470 [1:12:50<04:19,  1.50it/s]
1|6083|Loss: 0.9072251319885254:  94%|█████████▍| 6082/6470 [1:12:50<04:19,  1.50it/s]
1|6083|Loss: 0.9072251319885254:  94%|█████████▍| 6083/6470 [1:12:51<04:28,  1.44it/s]
1|6084|Loss: 0.8272272944450378:  94%|█████████▍| 6083/6470 [1:12:51<04:28,  1.44it/s]
1|6084|Loss: 0.8272272944450378:  94%|█████████▍| 6084/6470 [1:12:51<04:44,  1.36it/s]
1|6085|Loss: 0.8501034379005432:  94%|█████████▍| 6084/6470 [1:12:52<04:44,  1.36it/s]
1|6085|Loss: 0.8501034379005432:  94%|█████████▍| 6085/6470 [1:12:52<05:00,  1.28it/s]
1|6086|Loss: 0.8685507774353027:  94%|█████████▍| 6085/6470 [1:12:53<05:00,  1.28it/s]
1|6086|Loss: 0.8685507774353027:  94%|█████████▍| 6086/6470 [1:12:53<04:50,  1.32it/s]
1|6087|Loss: 0.9614104628562927:  94%|█████████▍| 6086/6470 [1:12:53<04:50,  1.32it/s]
1|6087|Loss: 0.9614104628562927:  94%|█████████▍| 6087/6470 [1:12:54<04:42,  1.36it/s]
1|6088|Loss: 0.8745579719543457:  94%|█████████▍| 6087/6470 [1:12:54<04:42,  1.36it/s]
1|6088|Loss: 0.8745579719543457:  94%|█████████▍| 6088/6470 [1:12:54<04:21,  1.46it/s]
1|6089|Loss: 0.7998963594436646:  94%|█████████▍| 6088/6470 [1:12:55<04:21,  1.46it/s]
1|6089|Loss: 0.7998963594436646:  94%|█████████▍| 6089/6470 [1:12:55<04:35,  1.38it/s]
1|6090|Loss: 0.8856765627861023:  94%|█████████▍| 6089/6470 [1:12:55<04:35,  1.38it/s]
1|6090|Loss: 0.8856765627861023:  94%|█████████▍| 6090/6470 [1:12:56<04:30,  1.40it/s]
1|6091|Loss: 0.9562276005744934:  94%|█████████▍| 6090/6470 [1:12:56<04:30,  1.40it/s]
1|6091|Loss: 0.9562276005744934:  94%|█████████▍| 6091/6470 [1:12:56<04:27,  1.42it/s]
1|6092|Loss: 0.8840279579162598:  94%|█████████▍| 6091/6470 [1:12:57<04:27,  1.42it/s]
1|6092|Loss: 0.8840279579162598:  94%|█████████▍| 6092/6470 [1:12:57<04:20,  1.45it/s]
1|6093|Loss: 0.9228456020355225:  94%|█████████▍| 6092/6470 [1:12:57<04:20,  1.45it/s]
1|6093|Loss: 0.9228456020355225:  94%|█████████▍| 6093/6470 [1:12:58<04:08,  1.52it/s]
1|6094|Loss: 0.9069183468818665:  94%|█████████▍| 6093/6470 [1:12:58<04:08,  1.52it/s]
1|6094|Loss: 0.9069183468818665:  94%|█████████▍| 6094/6470 [1:12:58<04:14,  1.48it/s]
1|6095|Loss: 0.8904217481613159:  94%|█████████▍| 6094/6470 [1:12:59<04:14,  1.48it/s]
1|6095|Loss: 0.8904217481613159:  94%|█████████▍| 6095/6470 [1:12:59<04:18,  1.45it/s]
1|6096|Loss: 0.9180084466934204:  94%|█████████▍| 6095/6470 [1:12:59<04:18,  1.45it/s]
1|6096|Loss: 0.9180084466934204:  94%|█████████▍| 6096/6470 [1:13:00<04:03,  1.53it/s]
1|6097|Loss: 0.8659414649009705:  94%|█████████▍| 6096/6470 [1:13:00<04:03,  1.53it/s]
1|6097|Loss: 0.8659414649009705:  94%|█████████▍| 6097/6470 [1:13:00<03:59,  1.56it/s]
1|6098|Loss: 0.7803740501403809:  94%|█████████▍| 6097/6470 [1:13:01<03:59,  1.56it/s]
1|6098|Loss: 0.7803740501403809:  94%|█████████▍| 6098/6470 [1:13:01<04:04,  1.52it/s]
1|6099|Loss: 0.8447136282920837:  94%|█████████▍| 6098/6470 [1:13:01<04:04,  1.52it/s]
1|6099|Loss: 0.8447136282920837:  94%|█████████▍| 6099/6470 [1:13:02<04:08,  1.49it/s]
1|6100|Loss: 0.8951335549354553:  94%|█████████▍| 6099/6470 [1:13:02<04:08,  1.49it/s]
1|6100|Loss: 0.8951335549354553:  94%|█████████▍| 6100/6470 [1:13:02<03:54,  1.58it/s]
1|6101|Loss: 0.9101538062095642:  94%|█████████▍| 6100/6470 [1:13:03<03:54,  1.58it/s]
1|6101|Loss: 0.9101538062095642:  94%|█████████▍| 6101/6470 [1:13:03<03:57,  1.55it/s]
1|6102|Loss: 0.8041653633117676:  94%|█████████▍| 6101/6470 [1:13:03<03:57,  1.55it/s]
1|6102|Loss: 0.8041653633117676:  94%|█████████▍| 6102/6470 [1:13:04<04:07,  1.49it/s]
1|6103|Loss: 1.018980860710144:  94%|█████████▍| 6102/6470 [1:13:04<04:07,  1.49it/s] 
1|6103|Loss: 1.018980860710144:  94%|█████████▍| 6103/6470 [1:13:04<03:57,  1.54it/s]
1|6104|Loss: 0.9888859391212463:  94%|█████████▍| 6103/6470 [1:13:05<03:57,  1.54it/s]
1|6104|Loss: 0.9888859391212463:  94%|█████████▍| 6104/6470 [1:13:05<03:49,  1.59it/s]
1|6105|Loss: 0.8379462957382202:  94%|█████████▍| 6104/6470 [1:13:05<03:49,  1.59it/s]
1|6105|Loss: 0.8379462957382202:  94%|█████████▍| 6105/6470 [1:13:06<04:07,  1.47it/s]
1|6106|Loss: 0.9397109150886536:  94%|█████████▍| 6105/6470 [1:13:06<04:07,  1.47it/s]
1|6106|Loss: 0.9397109150886536:  94%|█████████▍| 6106/6470 [1:13:06<04:12,  1.44it/s]
1|6107|Loss: 0.9814305901527405:  94%|█████████▍| 6106/6470 [1:13:07<04:12,  1.44it/s]
1|6107|Loss: 0.9814305901527405:  94%|█████████▍| 6107/6470 [1:13:07<03:50,  1.57it/s]
1|6108|Loss: 0.7774531841278076:  94%|█████████▍| 6107/6470 [1:13:07<03:50,  1.57it/s]
1|6108|Loss: 0.7774531841278076:  94%|█████████▍| 6108/6470 [1:13:08<03:58,  1.51it/s]
1|6109|Loss: 0.9472041130065918:  94%|█████████▍| 6108/6470 [1:13:08<03:58,  1.51it/s]
1|6109|Loss: 0.9472041130065918:  94%|█████████▍| 6109/6470 [1:13:08<03:57,  1.52it/s]
1|6110|Loss: 0.8879889249801636:  94%|█████████▍| 6109/6470 [1:13:09<03:57,  1.52it/s]
1|6110|Loss: 0.8879889249801636:  94%|█████████▍| 6110/6470 [1:13:09<03:52,  1.55it/s]
1|6111|Loss: 0.8578348159790039:  94%|█████████▍| 6110/6470 [1:13:09<03:52,  1.55it/s]
1|6111|Loss: 0.8578348159790039:  94%|█████████▍| 6111/6470 [1:13:09<03:48,  1.57it/s]
1|6112|Loss: 0.9116817116737366:  94%|█████████▍| 6111/6470 [1:13:10<03:48,  1.57it/s]
1|6112|Loss: 0.9116817116737366:  94%|█████████▍| 6112/6470 [1:13:10<04:00,  1.49it/s]
1|6113|Loss: 0.6710174083709717:  94%|█████████▍| 6112/6470 [1:13:11<04:00,  1.49it/s]
1|6113|Loss: 0.6710174083709717:  94%|█████████▍| 6113/6470 [1:13:11<03:50,  1.55it/s]
1|6114|Loss: 0.7970826625823975:  94%|█████████▍| 6113/6470 [1:13:11<03:50,  1.55it/s]
1|6114|Loss: 0.7970826625823975:  94%|█████████▍| 6114/6470 [1:13:12<04:09,  1.43it/s]
1|6115|Loss: 0.9061905145645142:  94%|█████████▍| 6114/6470 [1:13:12<04:09,  1.43it/s]
1|6115|Loss: 0.9061905145645142:  95%|█████████▍| 6115/6470 [1:13:12<04:26,  1.33it/s]
1|6116|Loss: 0.8637238144874573:  95%|█████████▍| 6115/6470 [1:13:13<04:26,  1.33it/s]
1|6116|Loss: 0.8637238144874573:  95%|█████████▍| 6116/6470 [1:13:13<04:15,  1.39it/s]
1|6117|Loss: 0.8335697054862976:  95%|█████████▍| 6116/6470 [1:13:14<04:15,  1.39it/s]
1|6117|Loss: 0.8335697054862976:  95%|█████████▍| 6117/6470 [1:13:14<04:27,  1.32it/s]
1|6118|Loss: 0.9708329439163208:  95%|█████████▍| 6117/6470 [1:13:14<04:27,  1.32it/s]
1|6118|Loss: 0.9708329439163208:  95%|█████████▍| 6118/6470 [1:13:15<04:28,  1.31it/s]
1|6119|Loss: 0.8255268931388855:  95%|█████████▍| 6118/6470 [1:13:15<04:28,  1.31it/s]
1|6119|Loss: 0.8255268931388855:  95%|█████████▍| 6119/6470 [1:13:15<04:26,  1.32it/s]
1|6120|Loss: 0.7745412588119507:  95%|█████████▍| 6119/6470 [1:13:16<04:26,  1.32it/s]
1|6120|Loss: 0.7745412588119507:  95%|█████████▍| 6120/6470 [1:13:16<04:16,  1.36it/s]
1|6121|Loss: 0.7354249358177185:  95%|█████████▍| 6120/6470 [1:13:16<04:16,  1.36it/s]
1|6121|Loss: 0.7354249358177185:  95%|█████████▍| 6121/6470 [1:13:17<03:50,  1.51it/s]
1|6122|Loss: 0.7857891321182251:  95%|█████████▍| 6121/6470 [1:13:17<03:50,  1.51it/s]
1|6122|Loss: 0.7857891321182251:  95%|█████████▍| 6122/6470 [1:13:17<03:42,  1.57it/s]
1|6123|Loss: 0.8271631598472595:  95%|█████████▍| 6122/6470 [1:13:18<03:42,  1.57it/s]
1|6123|Loss: 0.8271631598472595:  95%|█████████▍| 6123/6470 [1:13:18<03:48,  1.52it/s]
1|6124|Loss: 0.9194633960723877:  95%|█████████▍| 6123/6470 [1:13:18<03:48,  1.52it/s]
1|6124|Loss: 0.9194633960723877:  95%|█████████▍| 6124/6470 [1:13:19<04:06,  1.41it/s]
1|6125|Loss: 0.8207587599754333:  95%|█████████▍| 6124/6470 [1:13:19<04:06,  1.41it/s]
1|6125|Loss: 0.8207587599754333:  95%|█████████▍| 6125/6470 [1:13:20<04:07,  1.39it/s]
1|6126|Loss: 0.7931690812110901:  95%|█████████▍| 6125/6470 [1:13:20<04:07,  1.39it/s]
1|6126|Loss: 0.7931690812110901:  95%|█████████▍| 6126/6470 [1:13:20<04:12,  1.36it/s]
1|6127|Loss: 0.8143908977508545:  95%|█████████▍| 6126/6470 [1:13:21<04:12,  1.36it/s]
1|6127|Loss: 0.8143908977508545:  95%|█████████▍| 6127/6470 [1:13:21<04:05,  1.40it/s]
1|6128|Loss: 0.9420611262321472:  95%|█████████▍| 6127/6470 [1:13:21<04:05,  1.40it/s]
1|6128|Loss: 0.9420611262321472:  95%|█████████▍| 6128/6470 [1:13:22<03:56,  1.45it/s]
1|6129|Loss: 0.8631210327148438:  95%|█████████▍| 6128/6470 [1:13:22<03:56,  1.45it/s]
1|6129|Loss: 0.8631210327148438:  95%|█████████▍| 6129/6470 [1:13:22<03:40,  1.55it/s]
1|6130|Loss: 0.9410561919212341:  95%|█████████▍| 6129/6470 [1:13:23<03:40,  1.55it/s]
1|6130|Loss: 0.9410561919212341:  95%|█████████▍| 6130/6470 [1:13:23<03:54,  1.45it/s]
1|6131|Loss: 0.9551439881324768:  95%|█████████▍| 6130/6470 [1:13:23<03:54,  1.45it/s]
1|6131|Loss: 0.9551439881324768:  95%|█████████▍| 6131/6470 [1:13:24<04:09,  1.36it/s]
1|6132|Loss: 0.8640396595001221:  95%|█████████▍| 6131/6470 [1:13:24<04:09,  1.36it/s]
1|6132|Loss: 0.8640396595001221:  95%|█████████▍| 6132/6470 [1:13:25<04:09,  1.36it/s]
1|6133|Loss: 0.8896626234054565:  95%|█████████▍| 6132/6470 [1:13:25<04:09,  1.36it/s]
1|6133|Loss: 0.8896626234054565:  95%|█████████▍| 6133/6470 [1:13:25<04:17,  1.31it/s]
1|6134|Loss: 0.7463852763175964:  95%|█████████▍| 6133/6470 [1:13:26<04:17,  1.31it/s]
1|6134|Loss: 0.7463852763175964:  95%|█████████▍| 6134/6470 [1:13:26<04:26,  1.26it/s]
1|6135|Loss: 0.8305814862251282:  95%|█████████▍| 6134/6470 [1:13:27<04:26,  1.26it/s]
1|6135|Loss: 0.8305814862251282:  95%|█████████▍| 6135/6470 [1:13:27<04:19,  1.29it/s]
1|6136|Loss: 0.7664832472801208:  95%|█████████▍| 6135/6470 [1:13:27<04:19,  1.29it/s]
1|6136|Loss: 0.7664832472801208:  95%|█████████▍| 6136/6470 [1:13:28<04:14,  1.31it/s]
1|6137|Loss: 0.8770032525062561:  95%|█████████▍| 6136/6470 [1:13:28<04:14,  1.31it/s]
1|6137|Loss: 0.8770032525062561:  95%|█████████▍| 6137/6470 [1:13:29<04:21,  1.27it/s]
1|6138|Loss: 0.8699995279312134:  95%|█████████▍| 6137/6470 [1:13:29<04:21,  1.27it/s]
1|6138|Loss: 0.8699995279312134:  95%|█████████▍| 6138/6470 [1:13:29<04:06,  1.35it/s]
1|6139|Loss: 0.8395341634750366:  95%|█████████▍| 6138/6470 [1:13:30<04:06,  1.35it/s]
1|6139|Loss: 0.8395341634750366:  95%|█████████▍| 6139/6470 [1:13:30<04:12,  1.31it/s]
1|6140|Loss: 0.8653092384338379:  95%|█████████▍| 6139/6470 [1:13:30<04:12,  1.31it/s]
1|6140|Loss: 0.8653092384338379:  95%|█████████▍| 6140/6470 [1:13:31<04:05,  1.34it/s]
1|6141|Loss: 0.8365610837936401:  95%|█████████▍| 6140/6470 [1:13:31<04:05,  1.34it/s]
1|6141|Loss: 0.8365610837936401:  95%|█████████▍| 6141/6470 [1:13:32<04:16,  1.28it/s]
1|6142|Loss: 0.8330814838409424:  95%|█████████▍| 6141/6470 [1:13:32<04:16,  1.28it/s]
1|6142|Loss: 0.8330814838409424:  95%|█████████▍| 6142/6470 [1:13:32<04:08,  1.32it/s]
1|6143|Loss: 0.9925268292427063:  95%|█████████▍| 6142/6470 [1:13:33<04:08,  1.32it/s]
1|6143|Loss: 0.9925268292427063:  95%|█████████▍| 6143/6470 [1:13:33<04:03,  1.34it/s]
1|6144|Loss: 0.8584946990013123:  95%|█████████▍| 6143/6470 [1:13:33<04:03,  1.34it/s]
1|6144|Loss: 0.8584946990013123:  95%|█████████▍| 6144/6470 [1:13:34<03:59,  1.36it/s]
1|6145|Loss: 0.9985272884368896:  95%|█████████▍| 6144/6470 [1:13:34<03:59,  1.36it/s]
1|6145|Loss: 0.9985272884368896:  95%|█████████▍| 6145/6470 [1:13:34<03:40,  1.48it/s]
1|6146|Loss: 0.8429059982299805:  95%|█████████▍| 6145/6470 [1:13:35<03:40,  1.48it/s]
1|6146|Loss: 0.8429059982299805:  95%|█████████▍| 6146/6470 [1:13:35<03:30,  1.54it/s]
1|6147|Loss: 0.9622750282287598:  95%|█████████▍| 6146/6470 [1:13:35<03:30,  1.54it/s]
1|6147|Loss: 0.9622750282287598:  95%|█████████▌| 6147/6470 [1:13:35<03:16,  1.64it/s]
1|6148|Loss: 0.7752787470817566:  95%|█████████▌| 6147/6470 [1:13:36<03:16,  1.64it/s]
1|6148|Loss: 0.7752787470817566:  95%|█████████▌| 6148/6470 [1:13:36<03:10,  1.69it/s]
1|6149|Loss: 0.7393415570259094:  95%|█████████▌| 6148/6470 [1:13:36<03:10,  1.69it/s]
1|6149|Loss: 0.7393415570259094:  95%|█████████▌| 6149/6470 [1:13:37<03:25,  1.57it/s]
1|6150|Loss: 0.9004617929458618:  95%|█████████▌| 6149/6470 [1:13:37<03:25,  1.57it/s]
1|6150|Loss: 0.9004617929458618:  95%|█████████▌| 6150/6470 [1:13:37<03:06,  1.72it/s]
1|6151|Loss: 0.7932032346725464:  95%|█████████▌| 6150/6470 [1:13:37<03:06,  1.72it/s]
1|6151|Loss: 0.7932032346725464:  95%|█████████▌| 6151/6470 [1:13:38<03:13,  1.65it/s]
1|6152|Loss: 0.8919817209243774:  95%|█████████▌| 6151/6470 [1:13:38<03:13,  1.65it/s]
1|6152|Loss: 0.8919817209243774:  95%|█████████▌| 6152/6470 [1:13:38<03:22,  1.57it/s]
1|6153|Loss: 0.8316062688827515:  95%|█████████▌| 6152/6470 [1:13:39<03:22,  1.57it/s]
1|6153|Loss: 0.8316062688827515:  95%|█████████▌| 6153/6470 [1:13:39<03:30,  1.51it/s]
1|6154|Loss: 0.8531390428543091:  95%|█████████▌| 6153/6470 [1:13:40<03:30,  1.51it/s]
1|6154|Loss: 0.8531390428543091:  95%|█████████▌| 6154/6470 [1:13:40<03:29,  1.51it/s]
1|6155|Loss: 0.8195459842681885:  95%|█████████▌| 6154/6470 [1:13:40<03:29,  1.51it/s]
1|6155|Loss: 0.8195459842681885:  95%|█████████▌| 6155/6470 [1:13:41<03:31,  1.49it/s]
1|6156|Loss: 0.916123628616333:  95%|█████████▌| 6155/6470 [1:13:41<03:31,  1.49it/s] 
1|6156|Loss: 0.916123628616333:  95%|█████████▌| 6156/6470 [1:13:41<03:37,  1.44it/s]
1|6157|Loss: 0.7876414060592651:  95%|█████████▌| 6156/6470 [1:13:42<03:37,  1.44it/s]
1|6157|Loss: 0.7876414060592651:  95%|█████████▌| 6157/6470 [1:13:42<03:48,  1.37it/s]
1|6158|Loss: 1.1396112442016602:  95%|█████████▌| 6157/6470 [1:13:43<03:48,  1.37it/s]
1|6158|Loss: 1.1396112442016602:  95%|█████████▌| 6158/6470 [1:13:43<04:04,  1.28it/s]
1|6159|Loss: 0.9112352728843689:  95%|█████████▌| 6158/6470 [1:13:43<04:04,  1.28it/s]
1|6159|Loss: 0.9112352728843689:  95%|█████████▌| 6159/6470 [1:13:44<03:56,  1.32it/s]
1|6160|Loss: 0.9312596917152405:  95%|█████████▌| 6159/6470 [1:13:44<03:56,  1.32it/s]
1|6160|Loss: 0.9312596917152405:  95%|█████████▌| 6160/6470 [1:13:44<03:52,  1.34it/s]
1|6161|Loss: 0.9405559301376343:  95%|█████████▌| 6160/6470 [1:13:45<03:52,  1.34it/s]
1|6161|Loss: 0.9405559301376343:  95%|█████████▌| 6161/6470 [1:13:45<03:43,  1.38it/s]
1|6162|Loss: 0.8850065469741821:  95%|█████████▌| 6161/6470 [1:13:45<03:43,  1.38it/s]
1|6162|Loss: 0.8850065469741821:  95%|█████████▌| 6162/6470 [1:13:46<03:42,  1.38it/s]
1|6163|Loss: 0.9710427522659302:  95%|█████████▌| 6162/6470 [1:13:46<03:42,  1.38it/s]
1|6163|Loss: 0.9710427522659302:  95%|█████████▌| 6163/6470 [1:13:46<03:21,  1.52it/s]
1|6164|Loss: 0.7831152677536011:  95%|█████████▌| 6163/6470 [1:13:47<03:21,  1.52it/s]
1|6164|Loss: 0.7831152677536011:  95%|█████████▌| 6164/6470 [1:13:47<03:24,  1.50it/s]
1|6165|Loss: 0.8238009214401245:  95%|█████████▌| 6164/6470 [1:13:47<03:24,  1.50it/s]
1|6165|Loss: 0.8238009214401245:  95%|█████████▌| 6165/6470 [1:13:48<03:18,  1.53it/s]
1|6166|Loss: 0.9694544672966003:  95%|█████████▌| 6165/6470 [1:13:48<03:18,  1.53it/s]
1|6166|Loss: 0.9694544672966003:  95%|█████████▌| 6166/6470 [1:13:48<03:02,  1.66it/s]
1|6167|Loss: 0.8225868940353394:  95%|█████████▌| 6166/6470 [1:13:48<03:02,  1.66it/s]
1|6167|Loss: 0.8225868940353394:  95%|█████████▌| 6167/6470 [1:13:49<03:13,  1.57it/s]
1|6168|Loss: 0.8830450177192688:  95%|█████████▌| 6167/6470 [1:13:49<03:13,  1.57it/s]
1|6168|Loss: 0.8830450177192688:  95%|█████████▌| 6168/6470 [1:13:49<03:07,  1.61it/s]
1|6169|Loss: 0.7492138743400574:  95%|█████████▌| 6168/6470 [1:13:50<03:07,  1.61it/s]
1|6169|Loss: 0.7492138743400574:  95%|█████████▌| 6169/6470 [1:13:50<02:54,  1.73it/s]
1|6170|Loss: 0.8159441351890564:  95%|█████████▌| 6169/6470 [1:13:50<02:54,  1.73it/s]
1|6170|Loss: 0.8159441351890564:  95%|█████████▌| 6170/6470 [1:13:50<02:51,  1.75it/s]
1|6171|Loss: 0.7494068741798401:  95%|█████████▌| 6170/6470 [1:13:51<02:51,  1.75it/s]
1|6171|Loss: 0.7494068741798401:  95%|█████████▌| 6171/6470 [1:13:51<03:04,  1.62it/s]
1|6172|Loss: 0.94364994764328:  95%|█████████▌| 6171/6470 [1:13:52<03:04,  1.62it/s]  
1|6172|Loss: 0.94364994764328:  95%|█████████▌| 6172/6470 [1:13:52<03:06,  1.60it/s]
1|6173|Loss: 0.9021450281143188:  95%|█████████▌| 6172/6470 [1:13:52<03:06,  1.60it/s]
1|6173|Loss: 0.9021450281143188:  95%|█████████▌| 6173/6470 [1:13:52<03:01,  1.64it/s]
1|6174|Loss: 0.8360987901687622:  95%|█████████▌| 6173/6470 [1:13:53<03:01,  1.64it/s]
1|6174|Loss: 0.8360987901687622:  95%|█████████▌| 6174/6470 [1:13:53<03:07,  1.58it/s]
1|6175|Loss: 0.83512282371521:  95%|█████████▌| 6174/6470 [1:13:53<03:07,  1.58it/s]  
1|6175|Loss: 0.83512282371521:  95%|█████████▌| 6175/6470 [1:13:54<03:01,  1.62it/s]
1|6176|Loss: 0.8306769728660583:  95%|█████████▌| 6175/6470 [1:13:54<03:01,  1.62it/s]
1|6176|Loss: 0.8306769728660583:  95%|█████████▌| 6176/6470 [1:13:54<03:12,  1.53it/s]
1|6177|Loss: 0.952268660068512:  95%|█████████▌| 6176/6470 [1:13:55<03:12,  1.53it/s] 
1|6177|Loss: 0.952268660068512:  95%|█████████▌| 6177/6470 [1:13:55<03:12,  1.53it/s]
1|6178|Loss: 0.9230161905288696:  95%|█████████▌| 6177/6470 [1:13:55<03:12,  1.53it/s]
1|6178|Loss: 0.9230161905288696:  95%|█████████▌| 6178/6470 [1:13:56<03:00,  1.62it/s]
1|6179|Loss: 0.9622754454612732:  95%|█████████▌| 6178/6470 [1:13:56<03:00,  1.62it/s]
1|6179|Loss: 0.9622754454612732:  96%|█████████▌| 6179/6470 [1:13:56<03:12,  1.51it/s]
1|6180|Loss: 0.9607402086257935:  96%|█████████▌| 6179/6470 [1:13:57<03:12,  1.51it/s]
1|6180|Loss: 0.9607402086257935:  96%|█████████▌| 6180/6470 [1:13:57<03:17,  1.47it/s]
1|6181|Loss: 0.8632169961929321:  96%|█████████▌| 6180/6470 [1:13:57<03:17,  1.47it/s]
1|6181|Loss: 0.8632169961929321:  96%|█████████▌| 6181/6470 [1:13:58<03:12,  1.50it/s]
1|6182|Loss: 0.8669937252998352:  96%|█████████▌| 6181/6470 [1:13:58<03:12,  1.50it/s]
1|6182|Loss: 0.8669937252998352:  96%|█████████▌| 6182/6470 [1:13:58<03:07,  1.53it/s]
1|6183|Loss: 0.7363753318786621:  96%|█████████▌| 6182/6470 [1:13:59<03:07,  1.53it/s]
1|6183|Loss: 0.7363753318786621:  96%|█████████▌| 6183/6470 [1:13:59<03:04,  1.56it/s]
1|6184|Loss: 0.9297311902046204:  96%|█████████▌| 6183/6470 [1:13:59<03:04,  1.56it/s]
1|6184|Loss: 0.9297311902046204:  96%|█████████▌| 6184/6470 [1:14:00<03:10,  1.51it/s]
1|6185|Loss: 0.8381744623184204:  96%|█████████▌| 6184/6470 [1:14:00<03:10,  1.51it/s]
1|6185|Loss: 0.8381744623184204:  96%|█████████▌| 6185/6470 [1:14:00<03:23,  1.40it/s]
1|6186|Loss: 0.9361312389373779:  96%|█████████▌| 6185/6470 [1:14:01<03:23,  1.40it/s]
1|6186|Loss: 0.9361312389373779:  96%|█████████▌| 6186/6470 [1:14:01<03:28,  1.37it/s]
1|6187|Loss: 0.872803270816803:  96%|█████████▌| 6186/6470 [1:14:02<03:28,  1.37it/s] 
1|6187|Loss: 0.872803270816803:  96%|█████████▌| 6187/6470 [1:14:02<03:27,  1.36it/s]
1|6188|Loss: 1.0257556438446045:  96%|█████████▌| 6187/6470 [1:14:02<03:27,  1.36it/s]
1|6188|Loss: 1.0257556438446045:  96%|█████████▌| 6188/6470 [1:14:02<03:01,  1.55it/s]
1|6189|Loss: 1.006485104560852:  96%|█████████▌| 6188/6470 [1:14:03<03:01,  1.55it/s] 
1|6189|Loss: 1.006485104560852:  96%|█████████▌| 6189/6470 [1:14:03<02:56,  1.59it/s]
1|6190|Loss: 0.8587951064109802:  96%|█████████▌| 6189/6470 [1:14:03<02:56,  1.59it/s]
1|6190|Loss: 0.8587951064109802:  96%|█████████▌| 6190/6470 [1:14:04<03:21,  1.39it/s]
1|6191|Loss: 0.9041609764099121:  96%|█████████▌| 6190/6470 [1:14:04<03:21,  1.39it/s]
1|6191|Loss: 0.9041609764099121:  96%|█████████▌| 6191/6470 [1:14:05<03:27,  1.35it/s]
1|6192|Loss: 0.9347096085548401:  96%|█████████▌| 6191/6470 [1:14:05<03:27,  1.35it/s]
1|6192|Loss: 0.9347096085548401:  96%|█████████▌| 6192/6470 [1:14:05<03:25,  1.35it/s]
1|6193|Loss: 0.9449589848518372:  96%|█████████▌| 6192/6470 [1:14:06<03:25,  1.35it/s]
1|6193|Loss: 0.9449589848518372:  96%|█████████▌| 6193/6470 [1:14:06<03:19,  1.39it/s]
1|6194|Loss: 0.7393943667411804:  96%|█████████▌| 6193/6470 [1:14:06<03:19,  1.39it/s]
1|6194|Loss: 0.7393943667411804:  96%|█████████▌| 6194/6470 [1:14:07<03:04,  1.50it/s]
1|6195|Loss: 0.8804562091827393:  96%|█████████▌| 6194/6470 [1:14:07<03:04,  1.50it/s]
1|6195|Loss: 0.8804562091827393:  96%|█████████▌| 6195/6470 [1:14:07<03:03,  1.50it/s]
1|6196|Loss: 0.9053555130958557:  96%|█████████▌| 6195/6470 [1:14:08<03:03,  1.50it/s]
1|6196|Loss: 0.9053555130958557:  96%|█████████▌| 6196/6470 [1:14:08<03:06,  1.47it/s]
1|6197|Loss: 0.9495856761932373:  96%|█████████▌| 6196/6470 [1:14:08<03:06,  1.47it/s]
1|6197|Loss: 0.9495856761932373:  96%|█████████▌| 6197/6470 [1:14:09<03:13,  1.41it/s]
1|6198|Loss: 0.8591424822807312:  96%|█████████▌| 6197/6470 [1:14:09<03:13,  1.41it/s]
1|6198|Loss: 0.8591424822807312:  96%|█████████▌| 6198/6470 [1:14:10<03:14,  1.40it/s]
1|6199|Loss: 0.9456632733345032:  96%|█████████▌| 6198/6470 [1:14:10<03:14,  1.40it/s]
1|6199|Loss: 0.9456632733345032:  96%|█████████▌| 6199/6470 [1:14:10<03:14,  1.40it/s]
1|6200|Loss: 0.9914207458496094:  96%|█████████▌| 6199/6470 [1:14:11<03:14,  1.40it/s]
1|6200|Loss: 0.9914207458496094:  96%|█████████▌| 6200/6470 [1:14:11<03:18,  1.36it/s]
1|6201|Loss: 0.9029398560523987:  96%|█████████▌| 6200/6470 [1:14:11<03:18,  1.36it/s]
1|6201|Loss: 0.9029398560523987:  96%|█████████▌| 6201/6470 [1:14:12<03:16,  1.37it/s]
1|6202|Loss: 0.869556725025177:  96%|█████████▌| 6201/6470 [1:14:12<03:16,  1.37it/s] 
1|6202|Loss: 0.869556725025177:  96%|█████████▌| 6202/6470 [1:14:12<03:09,  1.41it/s]
1|6203|Loss: 0.6856306195259094:  96%|█████████▌| 6202/6470 [1:14:13<03:09,  1.41it/s]
1|6203|Loss: 0.6856306195259094:  96%|█████████▌| 6203/6470 [1:14:13<03:00,  1.48it/s]
1|6204|Loss: 0.862663209438324:  96%|█████████▌| 6203/6470 [1:14:13<03:00,  1.48it/s] 
1|6204|Loss: 0.862663209438324:  96%|█████████▌| 6204/6470 [1:14:14<02:52,  1.54it/s]
1|6205|Loss: 0.9653791189193726:  96%|█████████▌| 6204/6470 [1:14:14<02:52,  1.54it/s]
1|6205|Loss: 0.9653791189193726:  96%|█████████▌| 6205/6470 [1:14:14<03:00,  1.47it/s]
1|6206|Loss: 0.7428271770477295:  96%|█████████▌| 6205/6470 [1:14:15<03:00,  1.47it/s]
1|6206|Loss: 0.7428271770477295:  96%|█████████▌| 6206/6470 [1:14:15<02:41,  1.63it/s]
1|6207|Loss: 0.8056889772415161:  96%|█████████▌| 6206/6470 [1:14:15<02:41,  1.63it/s]
1|6207|Loss: 0.8056889772415161:  96%|█████████▌| 6207/6470 [1:14:16<02:45,  1.58it/s]
1|6208|Loss: 0.9171092510223389:  96%|█████████▌| 6207/6470 [1:14:16<02:45,  1.58it/s]
1|6208|Loss: 0.9171092510223389:  96%|█████████▌| 6208/6470 [1:14:16<02:48,  1.56it/s]
1|6209|Loss: 0.8824916481971741:  96%|█████████▌| 6208/6470 [1:14:17<02:48,  1.56it/s]
1|6209|Loss: 0.8824916481971741:  96%|█████████▌| 6209/6470 [1:14:17<02:51,  1.52it/s]
1|6210|Loss: 0.7653997540473938:  96%|█████████▌| 6209/6470 [1:14:17<02:51,  1.52it/s]
1|6210|Loss: 0.7653997540473938:  96%|█████████▌| 6210/6470 [1:14:17<02:48,  1.54it/s]
1|6211|Loss: 0.9575888514518738:  96%|█████████▌| 6210/6470 [1:14:18<02:48,  1.54it/s]
1|6211|Loss: 0.9575888514518738:  96%|█████████▌| 6211/6470 [1:14:18<02:50,  1.52it/s]
1|6212|Loss: 0.8644118905067444:  96%|█████████▌| 6211/6470 [1:14:19<02:50,  1.52it/s]
1|6212|Loss: 0.8644118905067444:  96%|█████████▌| 6212/6470 [1:14:19<02:46,  1.55it/s]
1|6213|Loss: 0.7204282879829407:  96%|█████████▌| 6212/6470 [1:14:19<02:46,  1.55it/s]
1|6213|Loss: 0.7204282879829407:  96%|█████████▌| 6213/6470 [1:14:20<02:58,  1.44it/s]
1|6214|Loss: 0.8854393362998962:  96%|█████████▌| 6213/6470 [1:14:20<02:58,  1.44it/s]
1|6214|Loss: 0.8854393362998962:  96%|█████████▌| 6214/6470 [1:14:20<02:45,  1.55it/s]
1|6215|Loss: 0.9909536838531494:  96%|█████████▌| 6214/6470 [1:14:20<02:45,  1.55it/s]
1|6215|Loss: 0.9909536838531494:  96%|█████████▌| 6215/6470 [1:14:21<02:41,  1.58it/s]
1|6216|Loss: 0.9620248079299927:  96%|█████████▌| 6215/6470 [1:14:21<02:41,  1.58it/s]
1|6216|Loss: 0.9620248079299927:  96%|█████████▌| 6216/6470 [1:14:22<02:52,  1.47it/s]
1|6217|Loss: 0.9242599606513977:  96%|█████████▌| 6216/6470 [1:14:22<02:52,  1.47it/s]
1|6217|Loss: 0.9242599606513977:  96%|█████████▌| 6217/6470 [1:14:22<03:01,  1.40it/s]
1|6218|Loss: 0.9262014031410217:  96%|█████████▌| 6217/6470 [1:14:23<03:01,  1.40it/s]
1|6218|Loss: 0.9262014031410217:  96%|█████████▌| 6218/6470 [1:14:23<03:11,  1.32it/s]
1|6219|Loss: 0.8452417254447937:  96%|█████████▌| 6218/6470 [1:14:24<03:11,  1.32it/s]
1|6219|Loss: 0.8452417254447937:  96%|█████████▌| 6219/6470 [1:14:24<03:12,  1.31it/s]
1|6220|Loss: 0.8139632344245911:  96%|█████████▌| 6219/6470 [1:14:24<03:12,  1.31it/s]
1|6220|Loss: 0.8139632344245911:  96%|█████████▌| 6220/6470 [1:14:25<03:15,  1.28it/s]
1|6221|Loss: 0.6860737204551697:  96%|█████████▌| 6220/6470 [1:14:25<03:15,  1.28it/s]
1|6221|Loss: 0.6860737204551697:  96%|█████████▌| 6221/6470 [1:14:26<03:18,  1.25it/s]
1|6222|Loss: 0.9644361138343811:  96%|█████████▌| 6221/6470 [1:14:26<03:18,  1.25it/s]
1|6222|Loss: 0.9644361138343811:  96%|█████████▌| 6222/6470 [1:14:26<03:08,  1.31it/s]
1|6223|Loss: 0.9193912744522095:  96%|█████████▌| 6222/6470 [1:14:27<03:08,  1.31it/s]
1|6223|Loss: 0.9193912744522095:  96%|█████████▌| 6223/6470 [1:14:27<02:55,  1.41it/s]
1|6224|Loss: 0.8703950643539429:  96%|█████████▌| 6223/6470 [1:14:27<02:55,  1.41it/s]
1|6224|Loss: 0.8703950643539429:  96%|█████████▌| 6224/6470 [1:14:28<03:03,  1.34it/s]
1|6225|Loss: 0.8176929354667664:  96%|█████████▌| 6224/6470 [1:14:28<03:03,  1.34it/s]
1|6225|Loss: 0.8176929354667664:  96%|█████████▌| 6225/6470 [1:14:28<02:57,  1.38it/s]
1|6226|Loss: 0.7467897534370422:  96%|█████████▌| 6225/6470 [1:14:29<02:57,  1.38it/s]
1|6226|Loss: 0.7467897534370422:  96%|█████████▌| 6226/6470 [1:14:29<03:02,  1.33it/s]
1|6227|Loss: 1.0186823606491089:  96%|█████████▌| 6226/6470 [1:14:30<03:02,  1.33it/s]
1|6227|Loss: 1.0186823606491089:  96%|█████████▌| 6227/6470 [1:14:30<03:13,  1.26it/s]
1|6228|Loss: 0.9292049407958984:  96%|█████████▌| 6227/6470 [1:14:31<03:13,  1.26it/s]
1|6228|Loss: 0.9292049407958984:  96%|█████████▋| 6228/6470 [1:14:31<03:11,  1.27it/s]
1|6229|Loss: 0.8938198685646057:  96%|█████████▋| 6228/6470 [1:14:31<03:11,  1.27it/s]
1|6229|Loss: 0.8938198685646057:  96%|█████████▋| 6229/6470 [1:14:32<03:06,  1.29it/s]
1|6230|Loss: 0.850958526134491:  96%|█████████▋| 6229/6470 [1:14:32<03:06,  1.29it/s] 
1|6230|Loss: 0.850958526134491:  96%|█████████▋| 6230/6470 [1:14:32<02:53,  1.38it/s]
1|6231|Loss: 0.6632795929908752:  96%|█████████▋| 6230/6470 [1:14:33<02:53,  1.38it/s]
1|6231|Loss: 0.6632795929908752:  96%|█████████▋| 6231/6470 [1:14:33<02:51,  1.39it/s]
1|6232|Loss: 0.9340217709541321:  96%|█████████▋| 6231/6470 [1:14:33<02:51,  1.39it/s]
1|6232|Loss: 0.9340217709541321:  96%|█████████▋| 6232/6470 [1:14:34<02:43,  1.46it/s]
1|6233|Loss: 0.7411876916885376:  96%|█████████▋| 6232/6470 [1:14:34<02:43,  1.46it/s]
1|6233|Loss: 0.7411876916885376:  96%|█████████▋| 6233/6470 [1:14:34<02:38,  1.50it/s]
1|6234|Loss: 0.9153867959976196:  96%|█████████▋| 6233/6470 [1:14:35<02:38,  1.50it/s]
1|6234|Loss: 0.9153867959976196:  96%|█████████▋| 6234/6470 [1:14:35<02:39,  1.48it/s]
1|6235|Loss: 0.7549527883529663:  96%|█████████▋| 6234/6470 [1:14:35<02:39,  1.48it/s]
1|6235|Loss: 0.7549527883529663:  96%|█████████▋| 6235/6470 [1:14:36<02:42,  1.45it/s]
1|6236|Loss: 0.9104242920875549:  96%|█████████▋| 6235/6470 [1:14:36<02:42,  1.45it/s]
1|6236|Loss: 0.9104242920875549:  96%|█████████▋| 6236/6470 [1:14:36<02:43,  1.43it/s]
1|6237|Loss: 0.9884971976280212:  96%|█████████▋| 6236/6470 [1:14:37<02:43,  1.43it/s]
1|6237|Loss: 0.9884971976280212:  96%|█████████▋| 6237/6470 [1:14:37<02:45,  1.40it/s]
1|6238|Loss: 0.7847089171409607:  96%|█████████▋| 6237/6470 [1:14:37<02:45,  1.40it/s]
1|6238|Loss: 0.7847089171409607:  96%|█████████▋| 6238/6470 [1:14:38<02:38,  1.47it/s]
1|6239|Loss: 0.8713096976280212:  96%|█████████▋| 6238/6470 [1:14:38<02:38,  1.47it/s]
1|6239|Loss: 0.8713096976280212:  96%|█████████▋| 6239/6470 [1:14:38<02:33,  1.51it/s]
1|6240|Loss: 0.7455461621284485:  96%|█████████▋| 6239/6470 [1:14:39<02:33,  1.51it/s]
1|6240|Loss: 0.7455461621284485:  96%|█████████▋| 6240/6470 [1:14:39<02:36,  1.47it/s]
1|6241|Loss: 0.7343968152999878:  96%|█████████▋| 6240/6470 [1:14:39<02:36,  1.47it/s]
1|6241|Loss: 0.7343968152999878:  96%|█████████▋| 6241/6470 [1:14:40<02:42,  1.41it/s]
1|6242|Loss: 0.8956880569458008:  96%|█████████▋| 6241/6470 [1:14:40<02:42,  1.41it/s]
1|6242|Loss: 0.8956880569458008:  96%|█████████▋| 6242/6470 [1:14:41<02:44,  1.38it/s]
1|6243|Loss: 0.922126054763794:  96%|█████████▋| 6242/6470 [1:14:41<02:44,  1.38it/s] 
1|6243|Loss: 0.922126054763794:  96%|█████████▋| 6243/6470 [1:14:41<02:45,  1.37it/s]
1|6244|Loss: 0.7758954167366028:  96%|█████████▋| 6243/6470 [1:14:42<02:45,  1.37it/s]
1|6244|Loss: 0.7758954167366028:  97%|█████████▋| 6244/6470 [1:14:42<02:37,  1.44it/s]
1|6245|Loss: 0.8320332169532776:  97%|█████████▋| 6244/6470 [1:14:42<02:37,  1.44it/s]
1|6245|Loss: 0.8320332169532776:  97%|█████████▋| 6245/6470 [1:14:43<02:35,  1.44it/s]
1|6246|Loss: 0.820521891117096:  97%|█████████▋| 6245/6470 [1:14:43<02:35,  1.44it/s] 
1|6246|Loss: 0.820521891117096:  97%|█████████▋| 6246/6470 [1:14:43<02:35,  1.44it/s]
1|6247|Loss: 1.0170583724975586:  97%|█████████▋| 6246/6470 [1:14:44<02:35,  1.44it/s]
1|6247|Loss: 1.0170583724975586:  97%|█████████▋| 6247/6470 [1:14:44<02:37,  1.42it/s]
1|6248|Loss: 0.8852568864822388:  97%|█████████▋| 6247/6470 [1:14:44<02:37,  1.42it/s]
1|6248|Loss: 0.8852568864822388:  97%|█████████▋| 6248/6470 [1:14:45<02:37,  1.41it/s]
1|6249|Loss: 0.8560038805007935:  97%|█████████▋| 6248/6470 [1:14:45<02:37,  1.41it/s]
1|6249|Loss: 0.8560038805007935:  97%|█████████▋| 6249/6470 [1:14:45<02:39,  1.39it/s]
1|6250|Loss: 0.7855085730552673:  97%|█████████▋| 6249/6470 [1:14:46<02:39,  1.39it/s]
1|6250|Loss: 0.7855085730552673:  97%|█████████▋| 6250/6470 [1:14:46<02:43,  1.35it/s]
1|6251|Loss: 0.8301974534988403:  97%|█████████▋| 6250/6470 [1:14:47<02:43,  1.35it/s]
1|6251|Loss: 0.8301974534988403:  97%|█████████▋| 6251/6470 [1:14:47<02:28,  1.48it/s]
1|6252|Loss: 0.8087871670722961:  97%|█████████▋| 6251/6470 [1:14:47<02:28,  1.48it/s]
1|6252|Loss: 0.8087871670722961:  97%|█████████▋| 6252/6470 [1:14:47<02:22,  1.53it/s]
1|6253|Loss: 0.878173291683197:  97%|█████████▋| 6252/6470 [1:14:48<02:22,  1.53it/s] 
1|6253|Loss: 0.878173291683197:  97%|█████████▋| 6253/6470 [1:14:48<02:14,  1.61it/s]
1|6254|Loss: 0.7994125485420227:  97%|█████████▋| 6253/6470 [1:14:48<02:14,  1.61it/s]
1|6254|Loss: 0.7994125485420227:  97%|█████████▋| 6254/6470 [1:14:49<02:18,  1.56it/s]
1|6255|Loss: 0.9069346785545349:  97%|█████████▋| 6254/6470 [1:14:49<02:18,  1.56it/s]
1|6255|Loss: 0.9069346785545349:  97%|█████████▋| 6255/6470 [1:14:49<02:29,  1.44it/s]
1|6256|Loss: 0.8095868229866028:  97%|█████████▋| 6255/6470 [1:14:50<02:29,  1.44it/s]
1|6256|Loss: 0.8095868229866028:  97%|█████████▋| 6256/6470 [1:14:50<02:35,  1.38it/s]
1|6257|Loss: 1.015340805053711:  97%|█████████▋| 6256/6470 [1:14:51<02:35,  1.38it/s] 
1|6257|Loss: 1.015340805053711:  97%|█████████▋| 6257/6470 [1:14:51<02:30,  1.42it/s]
1|6258|Loss: 0.8083903789520264:  97%|█████████▋| 6257/6470 [1:14:51<02:30,  1.42it/s]
1|6258|Loss: 0.8083903789520264:  97%|█████████▋| 6258/6470 [1:14:52<02:29,  1.42it/s]
1|6259|Loss: 1.0882595777511597:  97%|█████████▋| 6258/6470 [1:14:52<02:29,  1.42it/s]
1|6259|Loss: 1.0882595777511597:  97%|█████████▋| 6259/6470 [1:14:52<02:25,  1.45it/s]
1|6260|Loss: 0.7901497483253479:  97%|█████████▋| 6259/6470 [1:14:53<02:25,  1.45it/s]
1|6260|Loss: 0.7901497483253479:  97%|█████████▋| 6260/6470 [1:14:53<02:26,  1.43it/s]
1|6261|Loss: 0.8179630041122437:  97%|█████████▋| 6260/6470 [1:14:53<02:26,  1.43it/s]
1|6261|Loss: 0.8179630041122437:  97%|█████████▋| 6261/6470 [1:14:54<02:27,  1.42it/s]
1|6262|Loss: 0.8211147785186768:  97%|█████████▋| 6261/6470 [1:14:54<02:27,  1.42it/s]
1|6262|Loss: 0.8211147785186768:  97%|█████████▋| 6262/6470 [1:14:54<02:17,  1.51it/s]
1|6263|Loss: 0.8780645728111267:  97%|█████████▋| 6262/6470 [1:14:55<02:17,  1.51it/s]
1|6263|Loss: 0.8780645728111267:  97%|█████████▋| 6263/6470 [1:14:55<02:23,  1.45it/s]
1|6264|Loss: 0.8572146892547607:  97%|█████████▋| 6263/6470 [1:14:55<02:23,  1.45it/s]
1|6264|Loss: 0.8572146892547607:  97%|█████████▋| 6264/6470 [1:14:56<02:26,  1.41it/s]
1|6265|Loss: 0.802736222743988:  97%|█████████▋| 6264/6470 [1:14:56<02:26,  1.41it/s] 
1|6265|Loss: 0.802736222743988:  97%|█████████▋| 6265/6470 [1:14:57<02:30,  1.36it/s]
1|6266|Loss: 0.9767656922340393:  97%|█████████▋| 6265/6470 [1:14:57<02:30,  1.36it/s]
1|6266|Loss: 0.9767656922340393:  97%|█████████▋| 6266/6470 [1:14:57<02:28,  1.37it/s]
1|6267|Loss: 0.8884482383728027:  97%|█████████▋| 6266/6470 [1:14:58<02:28,  1.37it/s]
1|6267|Loss: 0.8884482383728027:  97%|█████████▋| 6267/6470 [1:14:58<02:32,  1.33it/s]
1|6268|Loss: 0.9283048510551453:  97%|█████████▋| 6267/6470 [1:14:59<02:32,  1.33it/s]
1|6268|Loss: 0.9283048510551453:  97%|█████████▋| 6268/6470 [1:14:59<02:35,  1.30it/s]
1|6269|Loss: 0.7581987977027893:  97%|█████████▋| 6268/6470 [1:14:59<02:35,  1.30it/s]
1|6269|Loss: 0.7581987977027893:  97%|█████████▋| 6269/6470 [1:15:00<02:39,  1.26it/s]
1|6270|Loss: 1.0021711587905884:  97%|█████████▋| 6269/6470 [1:15:00<02:39,  1.26it/s]
1|6270|Loss: 1.0021711587905884:  97%|█████████▋| 6270/6470 [1:15:00<02:32,  1.31it/s]
1|6271|Loss: 0.8501462340354919:  97%|█████████▋| 6270/6470 [1:15:01<02:32,  1.31it/s]
1|6271|Loss: 0.8501462340354919:  97%|█████████▋| 6271/6470 [1:15:01<02:27,  1.35it/s]
1|6272|Loss: 1.0013060569763184:  97%|█████████▋| 6271/6470 [1:15:02<02:27,  1.35it/s]
1|6272|Loss: 1.0013060569763184:  97%|█████████▋| 6272/6470 [1:15:02<02:26,  1.35it/s]
1|6273|Loss: 0.8808240294456482:  97%|█████████▋| 6272/6470 [1:15:02<02:26,  1.35it/s]
1|6273|Loss: 0.8808240294456482:  97%|█████████▋| 6273/6470 [1:15:02<02:16,  1.45it/s]
1|6274|Loss: 0.7947152853012085:  97%|█████████▋| 6273/6470 [1:15:03<02:16,  1.45it/s]
1|6274|Loss: 0.7947152853012085:  97%|█████████▋| 6274/6470 [1:15:03<02:11,  1.49it/s]
1|6275|Loss: 0.958280622959137:  97%|█████████▋| 6274/6470 [1:15:03<02:11,  1.49it/s] 
1|6275|Loss: 0.958280622959137:  97%|█████████▋| 6275/6470 [1:15:04<02:05,  1.55it/s]
1|6276|Loss: 0.8801151514053345:  97%|█████████▋| 6275/6470 [1:15:04<02:05,  1.55it/s]
1|6276|Loss: 0.8801151514053345:  97%|█████████▋| 6276/6470 [1:15:04<02:07,  1.53it/s]
1|6277|Loss: 0.8453490734100342:  97%|█████████▋| 6276/6470 [1:15:05<02:07,  1.53it/s]
1|6277|Loss: 0.8453490734100342:  97%|█████████▋| 6277/6470 [1:15:05<02:11,  1.47it/s]
1|6278|Loss: 0.8543534874916077:  97%|█████████▋| 6277/6470 [1:15:05<02:11,  1.47it/s]
1|6278|Loss: 0.8543534874916077:  97%|█████████▋| 6278/6470 [1:15:06<02:09,  1.49it/s]
1|6279|Loss: 0.8960772752761841:  97%|█████████▋| 6278/6470 [1:15:06<02:09,  1.49it/s]
1|6279|Loss: 0.8960772752761841:  97%|█████████▋| 6279/6470 [1:15:06<02:11,  1.45it/s]
1|6280|Loss: 0.8348902463912964:  97%|█████████▋| 6279/6470 [1:15:07<02:11,  1.45it/s]
1|6280|Loss: 0.8348902463912964:  97%|█████████▋| 6280/6470 [1:15:07<02:12,  1.43it/s]
1|6281|Loss: 0.9405958652496338:  97%|█████████▋| 6280/6470 [1:15:08<02:12,  1.43it/s]
1|6281|Loss: 0.9405958652496338:  97%|█████████▋| 6281/6470 [1:15:08<02:14,  1.40it/s]
1|6282|Loss: 0.7933928966522217:  97%|█████████▋| 6281/6470 [1:15:08<02:14,  1.40it/s]
1|6282|Loss: 0.7933928966522217:  97%|█████████▋| 6282/6470 [1:15:09<02:11,  1.43it/s]
1|6283|Loss: 0.8968110084533691:  97%|█████████▋| 6282/6470 [1:15:09<02:11,  1.43it/s]
1|6283|Loss: 0.8968110084533691:  97%|█████████▋| 6283/6470 [1:15:09<02:06,  1.48it/s]
1|6284|Loss: 0.9058167934417725:  97%|█████████▋| 6283/6470 [1:15:10<02:06,  1.48it/s]
1|6284|Loss: 0.9058167934417725:  97%|█████████▋| 6284/6470 [1:15:10<02:07,  1.46it/s]
1|6285|Loss: 0.8873332142829895:  97%|█████████▋| 6284/6470 [1:15:10<02:07,  1.46it/s]
1|6285|Loss: 0.8873332142829895:  97%|█████████▋| 6285/6470 [1:15:11<02:15,  1.37it/s]
1|6286|Loss: 0.8916508555412292:  97%|█████████▋| 6285/6470 [1:15:11<02:15,  1.37it/s]
1|6286|Loss: 0.8916508555412292:  97%|█████████▋| 6286/6470 [1:15:12<02:19,  1.32it/s]
1|6287|Loss: 0.8998608589172363:  97%|█████████▋| 6286/6470 [1:15:12<02:19,  1.32it/s]
1|6287|Loss: 0.8998608589172363:  97%|█████████▋| 6287/6470 [1:15:12<02:17,  1.33it/s]
1|6288|Loss: 0.7966484427452087:  97%|█████████▋| 6287/6470 [1:15:13<02:17,  1.33it/s]
1|6288|Loss: 0.7966484427452087:  97%|█████████▋| 6288/6470 [1:15:13<02:11,  1.38it/s]
1|6289|Loss: 0.6244104504585266:  97%|█████████▋| 6288/6470 [1:15:13<02:11,  1.38it/s]
1|6289|Loss: 0.6244104504585266:  97%|█████████▋| 6289/6470 [1:15:14<02:05,  1.44it/s]
1|6290|Loss: 0.9168249368667603:  97%|█████████▋| 6289/6470 [1:15:14<02:05,  1.44it/s]
1|6290|Loss: 0.9168249368667603:  97%|█████████▋| 6290/6470 [1:15:14<02:04,  1.44it/s]
1|6291|Loss: 0.8368865251541138:  97%|█████████▋| 6290/6470 [1:15:15<02:04,  1.44it/s]
1|6291|Loss: 0.8368865251541138:  97%|█████████▋| 6291/6470 [1:15:15<01:59,  1.50it/s]
1|6292|Loss: 0.9614633917808533:  97%|█████████▋| 6291/6470 [1:15:15<01:59,  1.50it/s]
1|6292|Loss: 0.9614633917808533:  97%|█████████▋| 6292/6470 [1:15:16<02:01,  1.46it/s]
1|6293|Loss: 0.8564879298210144:  97%|█████████▋| 6292/6470 [1:15:16<02:01,  1.46it/s]
1|6293|Loss: 0.8564879298210144:  97%|█████████▋| 6293/6470 [1:15:16<02:04,  1.42it/s]
1|6294|Loss: 0.9962416887283325:  97%|█████████▋| 6293/6470 [1:15:17<02:04,  1.42it/s]
1|6294|Loss: 0.9962416887283325:  97%|█████████▋| 6294/6470 [1:15:17<01:56,  1.51it/s]
1|6295|Loss: 0.7962658405303955:  97%|█████████▋| 6294/6470 [1:15:17<01:56,  1.51it/s]
1|6295|Loss: 0.7962658405303955:  97%|█████████▋| 6295/6470 [1:15:18<01:55,  1.52it/s]
1|6296|Loss: 0.816967785358429:  97%|█████████▋| 6295/6470 [1:15:18<01:55,  1.52it/s] 
1|6296|Loss: 0.816967785358429:  97%|█████████▋| 6296/6470 [1:15:18<01:48,  1.60it/s]
1|6297|Loss: 0.8615735173225403:  97%|█████████▋| 6296/6470 [1:15:18<01:48,  1.60it/s]
1|6297|Loss: 0.8615735173225403:  97%|█████████▋| 6297/6470 [1:15:19<01:50,  1.57it/s]
1|6298|Loss: 0.9479919075965881:  97%|█████████▋| 6297/6470 [1:15:19<01:50,  1.57it/s]
1|6298|Loss: 0.9479919075965881:  97%|█████████▋| 6298/6470 [1:15:20<02:00,  1.43it/s]
1|6299|Loss: 0.9637725949287415:  97%|█████████▋| 6298/6470 [1:15:20<02:00,  1.43it/s]
1|6299|Loss: 0.9637725949287415:  97%|█████████▋| 6299/6470 [1:15:20<02:03,  1.39it/s]
1|6300|Loss: 0.851044237613678:  97%|█████████▋| 6299/6470 [1:15:21<02:03,  1.39it/s] 
1|6300|Loss: 0.851044237613678:  97%|█████████▋| 6300/6470 [1:15:21<01:54,  1.49it/s]
1|6301|Loss: 0.9666403532028198:  97%|█████████▋| 6300/6470 [1:15:21<01:54,  1.49it/s]
1|6301|Loss: 0.9666403532028198:  97%|█████████▋| 6301/6470 [1:15:22<01:49,  1.54it/s]
1|6302|Loss: 0.9148784875869751:  97%|█████████▋| 6301/6470 [1:15:22<01:49,  1.54it/s]
1|6302|Loss: 0.9148784875869751:  97%|█████████▋| 6302/6470 [1:15:22<01:47,  1.57it/s]
1|6303|Loss: 0.863889217376709:  97%|█████████▋| 6302/6470 [1:15:23<01:47,  1.57it/s] 
1|6303|Loss: 0.863889217376709:  97%|█████████▋| 6303/6470 [1:15:23<01:46,  1.57it/s]
1|6304|Loss: 0.9773446321487427:  97%|█████████▋| 6303/6470 [1:15:23<01:46,  1.57it/s]
1|6304|Loss: 0.9773446321487427:  97%|█████████▋| 6304/6470 [1:15:23<01:44,  1.58it/s]
1|6305|Loss: 0.8992648720741272:  97%|█████████▋| 6304/6470 [1:15:24<01:44,  1.58it/s]
1|6305|Loss: 0.8992648720741272:  97%|█████████▋| 6305/6470 [1:15:24<01:50,  1.49it/s]
1|6306|Loss: 0.949457049369812:  97%|█████████▋| 6305/6470 [1:15:25<01:50,  1.49it/s] 
1|6306|Loss: 0.949457049369812:  97%|█████████▋| 6306/6470 [1:15:25<01:45,  1.56it/s]
1|6307|Loss: 0.8642650842666626:  97%|█████████▋| 6306/6470 [1:15:25<01:45,  1.56it/s]
1|6307|Loss: 0.8642650842666626:  97%|█████████▋| 6307/6470 [1:15:25<01:43,  1.58it/s]
1|6308|Loss: 0.7903298735618591:  97%|█████████▋| 6307/6470 [1:15:26<01:43,  1.58it/s]
1|6308|Loss: 0.7903298735618591:  97%|█████████▋| 6308/6470 [1:15:26<01:47,  1.51it/s]
1|6309|Loss: 0.8323038220405579:  97%|█████████▋| 6308/6470 [1:15:26<01:47,  1.51it/s]
1|6309|Loss: 0.8323038220405579:  98%|█████████▊| 6309/6470 [1:15:27<01:45,  1.52it/s]
1|6310|Loss: 0.9276056289672852:  98%|█████████▊| 6309/6470 [1:15:27<01:45,  1.52it/s]
1|6310|Loss: 0.9276056289672852:  98%|█████████▊| 6310/6470 [1:15:27<01:46,  1.51it/s]
1|6311|Loss: 0.8297368288040161:  98%|█████████▊| 6310/6470 [1:15:28<01:46,  1.51it/s]
1|6311|Loss: 0.8297368288040161:  98%|█████████▊| 6311/6470 [1:15:28<01:44,  1.52it/s]
1|6312|Loss: 0.8737530708312988:  98%|█████████▊| 6311/6470 [1:15:28<01:44,  1.52it/s]
1|6312|Loss: 0.8737530708312988:  98%|█████████▊| 6312/6470 [1:15:29<01:42,  1.54it/s]
1|6313|Loss: 0.6480215787887573:  98%|█████████▊| 6312/6470 [1:15:29<01:42,  1.54it/s]
1|6313|Loss: 0.6480215787887573:  98%|█████████▊| 6313/6470 [1:15:29<01:37,  1.61it/s]
1|6314|Loss: 0.8629812598228455:  98%|█████████▊| 6313/6470 [1:15:30<01:37,  1.61it/s]
1|6314|Loss: 0.8629812598228455:  98%|█████████▊| 6314/6470 [1:15:30<01:44,  1.49it/s]
1|6315|Loss: 0.8642975091934204:  98%|█████████▊| 6314/6470 [1:15:30<01:44,  1.49it/s]
1|6315|Loss: 0.8642975091934204:  98%|█████████▊| 6315/6470 [1:15:31<01:47,  1.44it/s]
1|6316|Loss: 0.7717471122741699:  98%|█████████▊| 6315/6470 [1:15:31<01:47,  1.44it/s]
1|6316|Loss: 0.7717471122741699:  98%|█████████▊| 6316/6470 [1:15:32<01:48,  1.42it/s]
1|6317|Loss: 0.9299435615539551:  98%|█████████▊| 6316/6470 [1:15:32<01:48,  1.42it/s]
1|6317|Loss: 0.9299435615539551:  98%|█████████▊| 6317/6470 [1:15:32<01:42,  1.49it/s]
1|6318|Loss: 0.8264074921607971:  98%|█████████▊| 6317/6470 [1:15:32<01:42,  1.49it/s]
1|6318|Loss: 0.8264074921607971:  98%|█████████▊| 6318/6470 [1:15:33<01:39,  1.52it/s]
1|6319|Loss: 0.7980147004127502:  98%|█████████▊| 6318/6470 [1:15:33<01:39,  1.52it/s]
1|6319|Loss: 0.7980147004127502:  98%|█████████▊| 6319/6470 [1:15:34<01:46,  1.42it/s]
1|6320|Loss: 0.8184065818786621:  98%|█████████▊| 6319/6470 [1:15:34<01:46,  1.42it/s]
1|6320|Loss: 0.8184065818786621:  98%|█████████▊| 6320/6470 [1:15:34<01:43,  1.45it/s]
1|6321|Loss: 0.7065575122833252:  98%|█████████▊| 6320/6470 [1:15:35<01:43,  1.45it/s]
1|6321|Loss: 0.7065575122833252:  98%|█████████▊| 6321/6470 [1:15:35<01:37,  1.53it/s]
1|6322|Loss: 0.9284718632698059:  98%|█████████▊| 6321/6470 [1:15:35<01:37,  1.53it/s]
1|6322|Loss: 0.9284718632698059:  98%|█████████▊| 6322/6470 [1:15:36<01:39,  1.49it/s]
1|6323|Loss: 0.8414582014083862:  98%|█████████▊| 6322/6470 [1:15:36<01:39,  1.49it/s]
1|6323|Loss: 0.8414582014083862:  98%|█████████▊| 6323/6470 [1:15:36<01:37,  1.50it/s]
1|6324|Loss: 0.9149610996246338:  98%|█████████▊| 6323/6470 [1:15:37<01:37,  1.50it/s]
1|6324|Loss: 0.9149610996246338:  98%|█████████▊| 6324/6470 [1:15:37<01:39,  1.47it/s]
1|6325|Loss: 0.8677464127540588:  98%|█████████▊| 6324/6470 [1:15:37<01:39,  1.47it/s]
1|6325|Loss: 0.8677464127540588:  98%|█████████▊| 6325/6470 [1:15:38<01:40,  1.44it/s]
1|6326|Loss: 0.7659354209899902:  98%|█████████▊| 6325/6470 [1:15:38<01:40,  1.44it/s]
1|6326|Loss: 0.7659354209899902:  98%|█████████▊| 6326/6470 [1:15:38<01:41,  1.42it/s]
1|6327|Loss: 0.8302359580993652:  98%|█████████▊| 6326/6470 [1:15:39<01:41,  1.42it/s]
1|6327|Loss: 0.8302359580993652:  98%|█████████▊| 6327/6470 [1:15:39<01:41,  1.40it/s]
1|6328|Loss: 0.7835891842842102:  98%|█████████▊| 6327/6470 [1:15:39<01:41,  1.40it/s]
1|6328|Loss: 0.7835891842842102:  98%|█████████▊| 6328/6470 [1:15:40<01:42,  1.39it/s]
1|6329|Loss: 0.9779546856880188:  98%|█████████▊| 6328/6470 [1:15:40<01:42,  1.39it/s]
1|6329|Loss: 0.9779546856880188:  98%|█████████▊| 6329/6470 [1:15:40<01:38,  1.43it/s]
1|6330|Loss: 0.9594770669937134:  98%|█████████▊| 6329/6470 [1:15:41<01:38,  1.43it/s]
1|6330|Loss: 0.9594770669937134:  98%|█████████▊| 6330/6470 [1:15:41<01:35,  1.47it/s]
1|6331|Loss: 0.9809725284576416:  98%|█████████▊| 6330/6470 [1:15:41<01:35,  1.47it/s]
1|6331|Loss: 0.9809725284576416:  98%|█████████▊| 6331/6470 [1:15:42<01:36,  1.44it/s]
1|6332|Loss: 0.7892506718635559:  98%|█████████▊| 6331/6470 [1:15:42<01:36,  1.44it/s]
1|6332|Loss: 0.7892506718635559:  98%|█████████▊| 6332/6470 [1:15:42<01:34,  1.46it/s]
1|6333|Loss: 0.6748122572898865:  98%|█████████▊| 6332/6470 [1:15:43<01:34,  1.46it/s]
1|6333|Loss: 0.6748122572898865:  98%|█████████▊| 6333/6470 [1:15:43<01:31,  1.50it/s]
1|6334|Loss: 0.7947266697883606:  98%|█████████▊| 6333/6470 [1:15:43<01:31,  1.50it/s]
1|6334|Loss: 0.7947266697883606:  98%|█████████▊| 6334/6470 [1:15:44<01:25,  1.59it/s]
1|6335|Loss: 0.8986377716064453:  98%|█████████▊| 6334/6470 [1:15:44<01:25,  1.59it/s]
1|6335|Loss: 0.8986377716064453:  98%|█████████▊| 6335/6470 [1:15:44<01:27,  1.54it/s]
1|6336|Loss: 0.7086213231086731:  98%|█████████▊| 6335/6470 [1:15:45<01:27,  1.54it/s]
1|6336|Loss: 0.7086213231086731:  98%|█████████▊| 6336/6470 [1:15:45<01:31,  1.47it/s]
1|6337|Loss: 0.7574805617332458:  98%|█████████▊| 6336/6470 [1:15:45<01:31,  1.47it/s]
1|6337|Loss: 0.7574805617332458:  98%|█████████▊| 6337/6470 [1:15:46<01:29,  1.49it/s]
1|6338|Loss: 0.8854759335517883:  98%|█████████▊| 6337/6470 [1:15:46<01:29,  1.49it/s]
1|6338|Loss: 0.8854759335517883:  98%|█████████▊| 6338/6470 [1:15:46<01:30,  1.46it/s]
1|6339|Loss: 0.9183405041694641:  98%|█████████▊| 6338/6470 [1:15:47<01:30,  1.46it/s]
1|6339|Loss: 0.9183405041694641:  98%|█████████▊| 6339/6470 [1:15:47<01:27,  1.50it/s]
1|6340|Loss: 0.8537173271179199:  98%|█████████▊| 6339/6470 [1:15:47<01:27,  1.50it/s]
1|6340|Loss: 0.8537173271179199:  98%|█████████▊| 6340/6470 [1:15:48<01:29,  1.45it/s]
1|6341|Loss: 0.8353466391563416:  98%|█████████▊| 6340/6470 [1:15:48<01:29,  1.45it/s]
1|6341|Loss: 0.8353466391563416:  98%|█████████▊| 6341/6470 [1:15:49<01:32,  1.39it/s]
1|6342|Loss: 0.8618552684783936:  98%|█████████▊| 6341/6470 [1:15:49<01:32,  1.39it/s]
1|6342|Loss: 0.8618552684783936:  98%|█████████▊| 6342/6470 [1:15:49<01:25,  1.49it/s]
1|6343|Loss: 0.6260345578193665:  98%|█████████▊| 6342/6470 [1:15:50<01:25,  1.49it/s]
1|6343|Loss: 0.6260345578193665:  98%|█████████▊| 6343/6470 [1:15:50<01:34,  1.34it/s]
1|6344|Loss: 0.6891736388206482:  98%|█████████▊| 6343/6470 [1:15:51<01:34,  1.34it/s]
1|6344|Loss: 0.6891736388206482:  98%|█████████▊| 6344/6470 [1:15:51<01:39,  1.27it/s]
1|6345|Loss: 0.916500985622406:  98%|█████████▊| 6344/6470 [1:15:51<01:39,  1.27it/s] 
1|6345|Loss: 0.916500985622406:  98%|█████████▊| 6345/6470 [1:15:52<01:41,  1.23it/s]
1|6346|Loss: 0.9902824759483337:  98%|█████████▊| 6345/6470 [1:15:52<01:41,  1.23it/s]
1|6346|Loss: 0.9902824759483337:  98%|█████████▊| 6346/6470 [1:15:53<01:44,  1.19it/s]
1|6347|Loss: 0.7885977625846863:  98%|█████████▊| 6346/6470 [1:15:53<01:44,  1.19it/s]
1|6347|Loss: 0.7885977625846863:  98%|█████████▊| 6347/6470 [1:15:54<01:41,  1.22it/s]
1|6348|Loss: 0.8511956930160522:  98%|█████████▊| 6347/6470 [1:15:54<01:41,  1.22it/s]
1|6348|Loss: 0.8511956930160522:  98%|█████████▊| 6348/6470 [1:15:54<01:39,  1.23it/s]
1|6349|Loss: 0.8306076526641846:  98%|█████████▊| 6348/6470 [1:15:55<01:39,  1.23it/s]
1|6349|Loss: 0.8306076526641846:  98%|█████████▊| 6349/6470 [1:15:55<01:32,  1.31it/s]
1|6350|Loss: 0.8074870705604553:  98%|█████████▊| 6349/6470 [1:15:55<01:32,  1.31it/s]
1|6350|Loss: 0.8074870705604553:  98%|█████████▊| 6350/6470 [1:15:56<01:31,  1.31it/s]
1|6351|Loss: 1.0383111238479614:  98%|█████████▊| 6350/6470 [1:15:56<01:31,  1.31it/s]
1|6351|Loss: 1.0383111238479614:  98%|█████████▊| 6351/6470 [1:15:56<01:27,  1.36it/s]
1|6352|Loss: 0.904252827167511:  98%|█████████▊| 6351/6470 [1:15:57<01:27,  1.36it/s] 
1|6352|Loss: 0.904252827167511:  98%|█████████▊| 6352/6470 [1:15:57<01:22,  1.44it/s]
1|6353|Loss: 1.0490949153900146:  98%|█████████▊| 6352/6470 [1:15:57<01:22,  1.44it/s]
1|6353|Loss: 1.0490949153900146:  98%|█████████▊| 6353/6470 [1:15:58<01:26,  1.36it/s]
1|6354|Loss: 0.9576819539070129:  98%|█████████▊| 6353/6470 [1:15:58<01:26,  1.36it/s]
1|6354|Loss: 0.9576819539070129:  98%|█████████▊| 6354/6470 [1:15:59<01:26,  1.34it/s]
1|6355|Loss: 0.9799759984016418:  98%|█████████▊| 6354/6470 [1:15:59<01:26,  1.34it/s]
1|6355|Loss: 0.9799759984016418:  98%|█████████▊| 6355/6470 [1:15:59<01:25,  1.35it/s]
1|6356|Loss: 0.7087891101837158:  98%|█████████▊| 6355/6470 [1:16:00<01:25,  1.35it/s]
1|6356|Loss: 0.7087891101837158:  98%|█████████▊| 6356/6470 [1:16:00<01:19,  1.43it/s]
1|6357|Loss: 0.7940430641174316:  98%|█████████▊| 6356/6470 [1:16:00<01:19,  1.43it/s]
1|6357|Loss: 0.7940430641174316:  98%|█████████▊| 6357/6470 [1:16:01<01:19,  1.43it/s]
1|6358|Loss: 0.8622701168060303:  98%|█████████▊| 6357/6470 [1:16:01<01:19,  1.43it/s]
1|6358|Loss: 0.8622701168060303:  98%|█████████▊| 6358/6470 [1:16:01<01:17,  1.45it/s]
1|6359|Loss: 0.9178591370582581:  98%|█████████▊| 6358/6470 [1:16:02<01:17,  1.45it/s]
1|6359|Loss: 0.9178591370582581:  98%|█████████▊| 6359/6470 [1:16:02<01:16,  1.45it/s]
1|6360|Loss: 0.9090416431427002:  98%|█████████▊| 6359/6470 [1:16:02<01:16,  1.45it/s]
1|6360|Loss: 0.9090416431427002:  98%|█████████▊| 6360/6470 [1:16:03<01:09,  1.57it/s]
1|6361|Loss: 0.7461084723472595:  98%|█████████▊| 6360/6470 [1:16:03<01:09,  1.57it/s]
1|6361|Loss: 0.7461084723472595:  98%|█████████▊| 6361/6470 [1:16:03<01:06,  1.64it/s]
1|6362|Loss: 0.8184163570404053:  98%|█████████▊| 6361/6470 [1:16:03<01:06,  1.64it/s]
1|6362|Loss: 0.8184163570404053:  98%|█████████▊| 6362/6470 [1:16:04<01:11,  1.51it/s]
1|6363|Loss: 0.8013101816177368:  98%|█████████▊| 6362/6470 [1:16:04<01:11,  1.51it/s]
1|6363|Loss: 0.8013101816177368:  98%|█████████▊| 6363/6470 [1:16:05<01:14,  1.44it/s]
1|6364|Loss: 0.8395662307739258:  98%|█████████▊| 6363/6470 [1:16:05<01:14,  1.44it/s]
1|6364|Loss: 0.8395662307739258:  98%|█████████▊| 6364/6470 [1:16:05<01:11,  1.49it/s]
1|6365|Loss: 0.8137431740760803:  98%|█████████▊| 6364/6470 [1:16:06<01:11,  1.49it/s]
1|6365|Loss: 0.8137431740760803:  98%|█████████▊| 6365/6470 [1:16:06<01:11,  1.47it/s]
1|6366|Loss: 0.8663618564605713:  98%|█████████▊| 6365/6470 [1:16:06<01:11,  1.47it/s]
1|6366|Loss: 0.8663618564605713:  98%|█████████▊| 6366/6470 [1:16:07<01:10,  1.47it/s]
1|6367|Loss: 0.8314228057861328:  98%|█████████▊| 6366/6470 [1:16:07<01:10,  1.47it/s]
1|6367|Loss: 0.8314228057861328:  98%|█████████▊| 6367/6470 [1:16:07<01:07,  1.53it/s]
1|6368|Loss: 0.9461672902107239:  98%|█████████▊| 6367/6470 [1:16:08<01:07,  1.53it/s]
1|6368|Loss: 0.9461672902107239:  98%|█████████▊| 6368/6470 [1:16:08<01:02,  1.62it/s]
1|6369|Loss: 0.723638117313385:  98%|█████████▊| 6368/6470 [1:16:08<01:02,  1.62it/s] 
1|6369|Loss: 0.723638117313385:  98%|█████████▊| 6369/6470 [1:16:08<01:02,  1.63it/s]
1|6370|Loss: 1.017252802848816:  98%|█████████▊| 6369/6470 [1:16:09<01:02,  1.63it/s]
1|6370|Loss: 1.017252802848816:  98%|█████████▊| 6370/6470 [1:16:09<00:59,  1.68it/s]
1|6371|Loss: 0.9573381543159485:  98%|█████████▊| 6370/6470 [1:16:09<00:59,  1.68it/s]
1|6371|Loss: 0.9573381543159485:  98%|█████████▊| 6371/6470 [1:16:10<01:01,  1.60it/s]
1|6372|Loss: 0.8181744813919067:  98%|█████████▊| 6371/6470 [1:16:10<01:01,  1.60it/s]
1|6372|Loss: 0.8181744813919067:  98%|█████████▊| 6372/6470 [1:16:10<01:04,  1.51it/s]
1|6373|Loss: 0.8348211646080017:  98%|█████████▊| 6372/6470 [1:16:11<01:04,  1.51it/s]
1|6373|Loss: 0.8348211646080017:  99%|█████████▊| 6373/6470 [1:16:11<01:05,  1.48it/s]
1|6374|Loss: 0.875298261642456:  99%|█████████▊| 6373/6470 [1:16:11<01:05,  1.48it/s] 
1|6374|Loss: 0.875298261642456:  99%|█████████▊| 6374/6470 [1:16:12<01:08,  1.41it/s]
1|6375|Loss: 0.7068334817886353:  99%|█████████▊| 6374/6470 [1:16:12<01:08,  1.41it/s]
1|6375|Loss: 0.7068334817886353:  99%|█████████▊| 6375/6470 [1:16:12<01:04,  1.48it/s]
1|6376|Loss: 0.8872255086898804:  99%|█████████▊| 6375/6470 [1:16:13<01:04,  1.48it/s]
1|6376|Loss: 0.8872255086898804:  99%|█████████▊| 6376/6470 [1:16:13<01:01,  1.53it/s]
1|6377|Loss: 0.7918578386306763:  99%|█████████▊| 6376/6470 [1:16:13<01:01,  1.53it/s]
1|6377|Loss: 0.7918578386306763:  99%|█████████▊| 6377/6470 [1:16:14<01:00,  1.54it/s]
1|6378|Loss: 1.0288373231887817:  99%|█████████▊| 6377/6470 [1:16:14<01:00,  1.54it/s]
1|6378|Loss: 1.0288373231887817:  99%|█████████▊| 6378/6470 [1:16:14<01:01,  1.50it/s]
1|6379|Loss: 0.749104380607605:  99%|█████████▊| 6378/6470 [1:16:15<01:01,  1.50it/s] 
1|6379|Loss: 0.749104380607605:  99%|█████████▊| 6379/6470 [1:16:15<01:01,  1.48it/s]
1|6380|Loss: 0.9102834463119507:  99%|█████████▊| 6379/6470 [1:16:15<01:01,  1.48it/s]
1|6380|Loss: 0.9102834463119507:  99%|█████████▊| 6380/6470 [1:16:16<01:02,  1.45it/s]
1|6381|Loss: 0.8286195397377014:  99%|█████████▊| 6380/6470 [1:16:16<01:02,  1.45it/s]
1|6381|Loss: 0.8286195397377014:  99%|█████████▊| 6381/6470 [1:16:17<01:02,  1.42it/s]
1|6382|Loss: 0.7045493721961975:  99%|█████████▊| 6381/6470 [1:16:17<01:02,  1.42it/s]
1|6382|Loss: 0.7045493721961975:  99%|█████████▊| 6382/6470 [1:16:17<00:56,  1.55it/s]
1|6383|Loss: 0.9947232007980347:  99%|█████████▊| 6382/6470 [1:16:17<00:56,  1.55it/s]
1|6383|Loss: 0.9947232007980347:  99%|█████████▊| 6383/6470 [1:16:18<00:58,  1.50it/s]
1|6384|Loss: 1.029383659362793:  99%|█████████▊| 6383/6470 [1:16:18<00:58,  1.50it/s] 
1|6384|Loss: 1.029383659362793:  99%|█████████▊| 6384/6470 [1:16:19<01:01,  1.39it/s]
1|6385|Loss: 0.8556655049324036:  99%|█████████▊| 6384/6470 [1:16:19<01:01,  1.39it/s]
1|6385|Loss: 0.8556655049324036:  99%|█████████▊| 6385/6470 [1:16:19<01:04,  1.32it/s]
1|6386|Loss: 0.8208112120628357:  99%|█████████▊| 6385/6470 [1:16:20<01:04,  1.32it/s]
1|6386|Loss: 0.8208112120628357:  99%|█████████▊| 6386/6470 [1:16:20<01:01,  1.37it/s]
1|6387|Loss: 1.0330461263656616:  99%|█████████▊| 6386/6470 [1:16:20<01:01,  1.37it/s]
1|6387|Loss: 1.0330461263656616:  99%|█████████▊| 6387/6470 [1:16:21<00:58,  1.42it/s]
1|6388|Loss: 0.8748044967651367:  99%|█████████▊| 6387/6470 [1:16:21<00:58,  1.42it/s]
1|6388|Loss: 0.8748044967651367:  99%|█████████▊| 6388/6470 [1:16:21<00:57,  1.43it/s]
1|6389|Loss: 0.8003084063529968:  99%|█████████▊| 6388/6470 [1:16:22<00:57,  1.43it/s]
1|6389|Loss: 0.8003084063529968:  99%|█████████▊| 6389/6470 [1:16:22<00:55,  1.45it/s]
1|6390|Loss: 1.0047916173934937:  99%|█████████▊| 6389/6470 [1:16:22<00:55,  1.45it/s]
1|6390|Loss: 1.0047916173934937:  99%|█████████▉| 6390/6470 [1:16:23<00:52,  1.52it/s]
1|6391|Loss: 0.9401276111602783:  99%|█████████▉| 6390/6470 [1:16:23<00:52,  1.52it/s]
1|6391|Loss: 0.9401276111602783:  99%|█████████▉| 6391/6470 [1:16:23<00:51,  1.55it/s]
1|6392|Loss: 0.9004884362220764:  99%|█████████▉| 6391/6470 [1:16:24<00:51,  1.55it/s]
1|6392|Loss: 0.9004884362220764:  99%|█████████▉| 6392/6470 [1:16:24<00:54,  1.43it/s]
1|6393|Loss: 1.0302817821502686:  99%|█████████▉| 6392/6470 [1:16:25<00:54,  1.43it/s]
1|6393|Loss: 1.0302817821502686:  99%|█████████▉| 6393/6470 [1:16:25<00:55,  1.38it/s]
1|6394|Loss: 0.9670193791389465:  99%|█████████▉| 6393/6470 [1:16:25<00:55,  1.38it/s]
1|6394|Loss: 0.9670193791389465:  99%|█████████▉| 6394/6470 [1:16:26<00:57,  1.33it/s]
1|6395|Loss: 0.9109744429588318:  99%|█████████▉| 6394/6470 [1:16:26<00:57,  1.33it/s]
1|6395|Loss: 0.9109744429588318:  99%|█████████▉| 6395/6470 [1:16:26<00:55,  1.34it/s]
1|6396|Loss: 0.8114997744560242:  99%|█████████▉| 6395/6470 [1:16:27<00:55,  1.34it/s]
1|6396|Loss: 0.8114997744560242:  99%|█████████▉| 6396/6470 [1:16:27<00:52,  1.40it/s]
1|6397|Loss: 0.8351379036903381:  99%|█████████▉| 6396/6470 [1:16:27<00:52,  1.40it/s]
1|6397|Loss: 0.8351379036903381:  99%|█████████▉| 6397/6470 [1:16:28<00:49,  1.48it/s]
1|6398|Loss: 1.1110748052597046:  99%|█████████▉| 6397/6470 [1:16:28<00:49,  1.48it/s]
1|6398|Loss: 1.1110748052597046:  99%|█████████▉| 6398/6470 [1:16:28<00:48,  1.48it/s]
1|6399|Loss: 1.0177329778671265:  99%|█████████▉| 6398/6470 [1:16:29<00:48,  1.48it/s]
1|6399|Loss: 1.0177329778671265:  99%|█████████▉| 6399/6470 [1:16:29<00:49,  1.45it/s]
1|6400|Loss: 1.0142796039581299:  99%|█████████▉| 6399/6470 [1:16:29<00:49,  1.45it/s]
1|6400|Loss: 1.0142796039581299:  99%|█████████▉| 6400/6470 [1:16:30<00:47,  1.48it/s]
1|6401|Loss: 0.7694194912910461:  99%|█████████▉| 6400/6470 [1:16:30<00:47,  1.48it/s]
1|6401|Loss: 0.7694194912910461:  99%|█████████▉| 6401/6470 [1:16:30<00:44,  1.54it/s]
1|6402|Loss: 0.8667630553245544:  99%|█████████▉| 6401/6470 [1:16:31<00:44,  1.54it/s]
1|6402|Loss: 0.8667630553245544:  99%|█████████▉| 6402/6470 [1:16:31<00:46,  1.45it/s]
1|6403|Loss: 0.9309373497962952:  99%|█████████▉| 6402/6470 [1:16:32<00:46,  1.45it/s]
1|6403|Loss: 0.9309373497962952:  99%|█████████▉| 6403/6470 [1:16:32<00:49,  1.36it/s]
1|6404|Loss: 0.7400965094566345:  99%|█████████▉| 6403/6470 [1:16:32<00:49,  1.36it/s]
1|6404|Loss: 0.7400965094566345:  99%|█████████▉| 6404/6470 [1:16:33<00:50,  1.31it/s]
1|6405|Loss: 0.8654297590255737:  99%|█████████▉| 6404/6470 [1:16:33<00:50,  1.31it/s]
1|6405|Loss: 0.8654297590255737:  99%|█████████▉| 6405/6470 [1:16:34<00:50,  1.30it/s]
1|6406|Loss: 0.9540722370147705:  99%|█████████▉| 6405/6470 [1:16:34<00:50,  1.30it/s]
1|6406|Loss: 0.9540722370147705:  99%|█████████▉| 6406/6470 [1:16:34<00:46,  1.38it/s]
1|6407|Loss: 0.8647441267967224:  99%|█████████▉| 6406/6470 [1:16:35<00:46,  1.38it/s]
1|6407|Loss: 0.8647441267967224:  99%|█████████▉| 6407/6470 [1:16:35<00:46,  1.35it/s]
1|6408|Loss: 0.8508633375167847:  99%|█████████▉| 6407/6470 [1:16:35<00:46,  1.35it/s]
1|6408|Loss: 0.8508633375167847:  99%|█████████▉| 6408/6470 [1:16:36<00:44,  1.39it/s]
1|6409|Loss: 0.8399141430854797:  99%|█████████▉| 6408/6470 [1:16:36<00:44,  1.39it/s]
1|6409|Loss: 0.8399141430854797:  99%|█████████▉| 6409/6470 [1:16:36<00:43,  1.40it/s]
1|6410|Loss: 0.841345489025116:  99%|█████████▉| 6409/6470 [1:16:37<00:43,  1.40it/s] 
1|6410|Loss: 0.841345489025116:  99%|█████████▉| 6410/6470 [1:16:37<00:45,  1.33it/s]
1|6411|Loss: 0.8589842915534973:  99%|█████████▉| 6410/6470 [1:16:38<00:45,  1.33it/s]
1|6411|Loss: 0.8589842915534973:  99%|█████████▉| 6411/6470 [1:16:38<00:43,  1.37it/s]
1|6412|Loss: 0.8252794146537781:  99%|█████████▉| 6411/6470 [1:16:38<00:43,  1.37it/s]
1|6412|Loss: 0.8252794146537781:  99%|█████████▉| 6412/6470 [1:16:38<00:40,  1.43it/s]
1|6413|Loss: 0.9230583906173706:  99%|█████████▉| 6412/6470 [1:16:39<00:40,  1.43it/s]
1|6413|Loss: 0.9230583906173706:  99%|█████████▉| 6413/6470 [1:16:39<00:38,  1.47it/s]
1|6414|Loss: 0.9127120971679688:  99%|█████████▉| 6413/6470 [1:16:39<00:38,  1.47it/s]
1|6414|Loss: 0.9127120971679688:  99%|█████████▉| 6414/6470 [1:16:40<00:37,  1.49it/s]
1|6415|Loss: 0.7324219346046448:  99%|█████████▉| 6414/6470 [1:16:40<00:37,  1.49it/s]
1|6415|Loss: 0.7324219346046448:  99%|█████████▉| 6415/6470 [1:16:40<00:34,  1.59it/s]
1|6416|Loss: 0.8612861633300781:  99%|█████████▉| 6415/6470 [1:16:41<00:34,  1.59it/s]
1|6416|Loss: 0.8612861633300781:  99%|█████████▉| 6416/6470 [1:16:41<00:33,  1.60it/s]
1|6417|Loss: 0.9339990019798279:  99%|█████████▉| 6416/6470 [1:16:41<00:33,  1.60it/s]
1|6417|Loss: 0.9339990019798279:  99%|█████████▉| 6417/6470 [1:16:42<00:33,  1.60it/s]
1|6418|Loss: 0.8923978209495544:  99%|█████████▉| 6417/6470 [1:16:42<00:33,  1.60it/s]
1|6418|Loss: 0.8923978209495544:  99%|█████████▉| 6418/6470 [1:16:42<00:34,  1.52it/s]
1|6419|Loss: 0.8736737966537476:  99%|█████████▉| 6418/6470 [1:16:43<00:34,  1.52it/s]
1|6419|Loss: 0.8736737966537476:  99%|█████████▉| 6419/6470 [1:16:43<00:34,  1.47it/s]
1|6420|Loss: 0.8384003043174744:  99%|█████████▉| 6419/6470 [1:16:43<00:34,  1.47it/s]
1|6420|Loss: 0.8384003043174744:  99%|█████████▉| 6420/6470 [1:16:44<00:36,  1.36it/s]
1|6421|Loss: 0.7307854890823364:  99%|█████████▉| 6420/6470 [1:16:44<00:36,  1.36it/s]
1|6421|Loss: 0.7307854890823364:  99%|█████████▉| 6421/6470 [1:16:45<00:38,  1.28it/s]
1|6422|Loss: 0.92928546667099:  99%|█████████▉| 6421/6470 [1:16:45<00:38,  1.28it/s]  
1|6422|Loss: 0.92928546667099:  99%|█████████▉| 6422/6470 [1:16:46<00:37,  1.28it/s]
1|6423|Loss: 0.9091441035270691:  99%|█████████▉| 6422/6470 [1:16:46<00:37,  1.28it/s]
1|6423|Loss: 0.9091441035270691:  99%|█████████▉| 6423/6470 [1:16:46<00:36,  1.30it/s]
1|6424|Loss: 0.752474844455719:  99%|█████████▉| 6423/6470 [1:16:47<00:36,  1.30it/s] 
1|6424|Loss: 0.752474844455719:  99%|█████████▉| 6424/6470 [1:16:47<00:33,  1.36it/s]
1|6425|Loss: 1.0027260780334473:  99%|█████████▉| 6424/6470 [1:16:47<00:33,  1.36it/s]
1|6425|Loss: 1.0027260780334473:  99%|█████████▉| 6425/6470 [1:16:48<00:32,  1.38it/s]
1|6426|Loss: 0.6666315197944641:  99%|█████████▉| 6425/6470 [1:16:48<00:32,  1.38it/s]
1|6426|Loss: 0.6666315197944641:  99%|█████████▉| 6426/6470 [1:16:48<00:31,  1.39it/s]
1|6427|Loss: 1.014092206954956:  99%|█████████▉| 6426/6470 [1:16:49<00:31,  1.39it/s] 
1|6427|Loss: 1.014092206954956:  99%|█████████▉| 6427/6470 [1:16:49<00:32,  1.34it/s]
1|6428|Loss: 0.8239193558692932:  99%|█████████▉| 6427/6470 [1:16:50<00:32,  1.34it/s]
1|6428|Loss: 0.8239193558692932:  99%|█████████▉| 6428/6470 [1:16:50<00:32,  1.28it/s]
1|6429|Loss: 0.823084831237793:  99%|█████████▉| 6428/6470 [1:16:51<00:32,  1.28it/s] 
1|6429|Loss: 0.823084831237793:  99%|█████████▉| 6429/6470 [1:16:51<00:33,  1.23it/s]
1|6430|Loss: 1.0378378629684448:  99%|█████████▉| 6429/6470 [1:16:51<00:33,  1.23it/s]
1|6430|Loss: 1.0378378629684448:  99%|█████████▉| 6430/6470 [1:16:52<00:31,  1.29it/s]
1|6431|Loss: 0.8691393733024597:  99%|█████████▉| 6430/6470 [1:16:52<00:31,  1.29it/s]
1|6431|Loss: 0.8691393733024597:  99%|█████████▉| 6431/6470 [1:16:52<00:29,  1.33it/s]
1|6432|Loss: 0.9105616211891174:  99%|█████████▉| 6431/6470 [1:16:53<00:29,  1.33it/s]
1|6432|Loss: 0.9105616211891174:  99%|█████████▉| 6432/6470 [1:16:53<00:25,  1.48it/s]
1|6433|Loss: 0.9337135553359985:  99%|█████████▉| 6432/6470 [1:16:53<00:25,  1.48it/s]
1|6433|Loss: 0.9337135553359985:  99%|█████████▉| 6433/6470 [1:16:53<00:25,  1.47it/s]
1|6434|Loss: 0.8477503061294556:  99%|█████████▉| 6433/6470 [1:16:54<00:25,  1.47it/s]
1|6434|Loss: 0.8477503061294556:  99%|█████████▉| 6434/6470 [1:16:54<00:24,  1.48it/s]
1|6435|Loss: 0.9180175065994263:  99%|█████████▉| 6434/6470 [1:16:55<00:24,  1.48it/s]
1|6435|Loss: 0.9180175065994263:  99%|█████████▉| 6435/6470 [1:16:55<00:24,  1.45it/s]
1|6436|Loss: 0.9435101747512817:  99%|█████████▉| 6435/6470 [1:16:55<00:24,  1.45it/s]
1|6436|Loss: 0.9435101747512817:  99%|█████████▉| 6436/6470 [1:16:56<00:23,  1.47it/s]
1|6437|Loss: 0.7777802348136902:  99%|█████████▉| 6436/6470 [1:16:56<00:23,  1.47it/s]
1|6437|Loss: 0.7777802348136902:  99%|█████████▉| 6437/6470 [1:16:56<00:21,  1.51it/s]
1|6438|Loss: 0.8781395554542542:  99%|█████████▉| 6437/6470 [1:16:57<00:21,  1.51it/s]
1|6438|Loss: 0.8781395554542542: 100%|█████████▉| 6438/6470 [1:16:57<00:22,  1.41it/s]
1|6439|Loss: 0.8508449196815491: 100%|█████████▉| 6438/6470 [1:16:57<00:22,  1.41it/s]
1|6439|Loss: 0.8508449196815491: 100%|█████████▉| 6439/6470 [1:16:58<00:23,  1.33it/s]
1|6440|Loss: 0.7359904050827026: 100%|█████████▉| 6439/6470 [1:16:58<00:23,  1.33it/s]
1|6440|Loss: 0.7359904050827026: 100%|█████████▉| 6440/6470 [1:16:59<00:22,  1.33it/s]
1|6441|Loss: 0.7836769819259644: 100%|█████████▉| 6440/6470 [1:16:59<00:22,  1.33it/s]
1|6441|Loss: 0.7836769819259644: 100%|█████████▉| 6441/6470 [1:16:59<00:21,  1.35it/s]
1|6442|Loss: 0.8058815002441406: 100%|█████████▉| 6441/6470 [1:17:00<00:21,  1.35it/s]
1|6442|Loss: 0.8058815002441406: 100%|█████████▉| 6442/6470 [1:17:00<00:20,  1.36it/s]
1|6443|Loss: 0.9610121250152588: 100%|█████████▉| 6442/6470 [1:17:00<00:20,  1.36it/s]
1|6443|Loss: 0.9610121250152588: 100%|█████████▉| 6443/6470 [1:17:01<00:19,  1.37it/s]
1|6444|Loss: 0.8697658777236938: 100%|█████████▉| 6443/6470 [1:17:01<00:19,  1.37it/s]
1|6444|Loss: 0.8697658777236938: 100%|█████████▉| 6444/6470 [1:17:01<00:19,  1.37it/s]
1|6445|Loss: 0.8163231611251831: 100%|█████████▉| 6444/6470 [1:17:02<00:19,  1.37it/s]
1|6445|Loss: 0.8163231611251831: 100%|█████████▉| 6445/6470 [1:17:02<00:18,  1.35it/s]
1|6446|Loss: 0.8642125129699707: 100%|█████████▉| 6445/6470 [1:17:03<00:18,  1.35it/s]
1|6446|Loss: 0.8642125129699707: 100%|█████████▉| 6446/6470 [1:17:03<00:17,  1.40it/s]
1|6447|Loss: 0.8168824911117554: 100%|█████████▉| 6446/6470 [1:17:03<00:17,  1.40it/s]
1|6447|Loss: 0.8168824911117554: 100%|█████████▉| 6447/6470 [1:17:04<00:15,  1.45it/s]
1|6448|Loss: 0.8576304912567139: 100%|█████████▉| 6447/6470 [1:17:04<00:15,  1.45it/s]
1|6448|Loss: 0.8576304912567139: 100%|█████████▉| 6448/6470 [1:17:04<00:14,  1.50it/s]
1|6449|Loss: 0.8496704697608948: 100%|█████████▉| 6448/6470 [1:17:05<00:14,  1.50it/s]
1|6449|Loss: 0.8496704697608948: 100%|█████████▉| 6449/6470 [1:17:05<00:14,  1.49it/s]
1|6450|Loss: 0.8409594297409058: 100%|█████████▉| 6449/6470 [1:17:05<00:14,  1.49it/s]
1|6450|Loss: 0.8409594297409058: 100%|█████████▉| 6450/6470 [1:17:05<00:13,  1.51it/s]
1|6451|Loss: 0.7845267653465271: 100%|█████████▉| 6450/6470 [1:17:06<00:13,  1.51it/s]
1|6451|Loss: 0.7845267653465271: 100%|█████████▉| 6451/6470 [1:17:06<00:12,  1.54it/s]
1|6452|Loss: 0.6071706414222717: 100%|█████████▉| 6451/6470 [1:17:06<00:12,  1.54it/s]
1|6452|Loss: 0.6071706414222717: 100%|█████████▉| 6452/6470 [1:17:07<00:11,  1.55it/s]
1|6453|Loss: 0.7821022272109985: 100%|█████████▉| 6452/6470 [1:17:07<00:11,  1.55it/s]
1|6453|Loss: 0.7821022272109985: 100%|█████████▉| 6453/6470 [1:17:07<00:11,  1.46it/s]
1|6454|Loss: 0.6849913001060486: 100%|█████████▉| 6453/6470 [1:17:08<00:11,  1.46it/s]
1|6454|Loss: 0.6849913001060486: 100%|█████████▉| 6454/6470 [1:17:08<00:10,  1.47it/s]
1|6455|Loss: 0.9897137880325317: 100%|█████████▉| 6454/6470 [1:17:09<00:10,  1.47it/s]
1|6455|Loss: 0.9897137880325317: 100%|█████████▉| 6455/6470 [1:17:09<00:10,  1.46it/s]
1|6456|Loss: 0.8926839828491211: 100%|█████████▉| 6455/6470 [1:17:09<00:10,  1.46it/s]
1|6456|Loss: 0.8926839828491211: 100%|█████████▉| 6456/6470 [1:17:10<00:10,  1.39it/s]
1|6457|Loss: 0.9326668381690979: 100%|█████████▉| 6456/6470 [1:17:10<00:10,  1.39it/s]
1|6457|Loss: 0.9326668381690979: 100%|█████████▉| 6457/6470 [1:17:10<00:08,  1.45it/s]
1|6458|Loss: 0.8188270926475525: 100%|█████████▉| 6457/6470 [1:17:11<00:08,  1.45it/s]
1|6458|Loss: 0.8188270926475525: 100%|█████████▉| 6458/6470 [1:17:11<00:08,  1.44it/s]
1|6459|Loss: 0.8251195549964905: 100%|█████████▉| 6458/6470 [1:17:11<00:08,  1.44it/s]
1|6459|Loss: 0.8251195549964905: 100%|█████████▉| 6459/6470 [1:17:12<00:07,  1.43it/s]
1|6460|Loss: 0.8940576910972595: 100%|█████████▉| 6459/6470 [1:17:12<00:07,  1.43it/s]
1|6460|Loss: 0.8940576910972595: 100%|█████████▉| 6460/6470 [1:17:12<00:07,  1.42it/s]
1|6461|Loss: 0.8051562309265137: 100%|█████████▉| 6460/6470 [1:17:13<00:07,  1.42it/s]
1|6461|Loss: 0.8051562309265137: 100%|█████████▉| 6461/6470 [1:17:13<00:05,  1.54it/s]
1|6462|Loss: 0.860571563243866: 100%|█████████▉| 6461/6470 [1:17:13<00:05,  1.54it/s] 
1|6462|Loss: 0.860571563243866: 100%|█████████▉| 6462/6470 [1:17:14<00:05,  1.51it/s]
1|6463|Loss: 0.7669236660003662: 100%|█████████▉| 6462/6470 [1:17:14<00:05,  1.51it/s]
1|6463|Loss: 0.7669236660003662: 100%|█████████▉| 6463/6470 [1:17:14<00:04,  1.45it/s]
1|6464|Loss: 0.8504640460014343: 100%|█████████▉| 6463/6470 [1:17:15<00:04,  1.45it/s]
1|6464|Loss: 0.8504640460014343: 100%|█████████▉| 6464/6470 [1:17:15<00:04,  1.42it/s]
1|6465|Loss: 0.9902701377868652: 100%|█████████▉| 6464/6470 [1:17:16<00:04,  1.42it/s]
1|6465|Loss: 0.9902701377868652: 100%|█████████▉| 6465/6470 [1:17:16<00:03,  1.37it/s]
1|6466|Loss: 0.9333712458610535: 100%|█████████▉| 6465/6470 [1:17:16<00:03,  1.37it/s]
1|6466|Loss: 0.9333712458610535: 100%|█████████▉| 6466/6470 [1:17:17<00:03,  1.33it/s]
1|6467|Loss: 0.8279280662536621: 100%|█████████▉| 6466/6470 [1:17:17<00:03,  1.33it/s]
1|6467|Loss: 0.8279280662536621: 100%|█████████▉| 6467/6470 [1:17:17<00:02,  1.31it/s]
1|6468|Loss: 0.929103672504425: 100%|█████████▉| 6467/6470 [1:17:18<00:02,  1.31it/s] 
1|6468|Loss: 0.929103672504425: 100%|█████████▉| 6468/6470 [1:17:18<00:01,  1.37it/s]
1|6469|Loss: 0.7332636713981628: 100%|█████████▉| 6468/6470 [1:17:18<00:01,  1.37it/s]
1|6469|Loss: 0.7332636713981628: 100%|█████████▉| 6469/6470 [1:17:19<00:00,  1.44it/s]
1|6470|Loss: 0.8265537023544312: 100%|█████████▉| 6469/6470 [1:17:19<00:00,  1.44it/s]
1|6470|Loss: 0.8265537023544312: 100%|██████████| 6470/6470 [1:17:19<00:00,  1.48it/s]
1|6470|Loss: 0.8265537023544312: 100%|██████████| 6470/6470 [1:17:19<00:00,  1.39it/s]
Model checkpoint of size 25705 MB saved to output/alpaca-llama2-baseline/model_0_6470.ckpt

  0%|          | 0/6470 [00:00<?, ?it/s]
2|1|Loss: 1.1222110986709595:   0%|          | 0/6470 [00:00<?, ?it/s]
2|1|Loss: 1.1222110986709595:   0%|          | 1/6470 [00:00<53:16,  2.02it/s]
2|2|Loss: 0.828296422958374:   0%|          | 1/6470 [00:00<53:16,  2.02it/s] 
2|2|Loss: 0.828296422958374:   0%|          | 2/6470 [00:01<1:05:59,  1.63it/s]
2|3|Loss: 0.8462681770324707:   0%|          | 2/6470 [00:01<1:05:59,  1.63it/s]
2|3|Loss: 0.8462681770324707:   0%|          | 3/6470 [00:01<1:15:02,  1.44it/s]
2|4|Loss: 0.7681014537811279:   0%|          | 3/6470 [00:02<1:15:02,  1.44it/s]
2|4|Loss: 0.7681014537811279:   0%|          | 4/6470 [00:02<1:08:48,  1.57it/s]
2|5|Loss: 0.8745959997177124:   0%|          | 4/6470 [00:02<1:08:48,  1.57it/s]
2|5|Loss: 0.8745959997177124:   0%|          | 5/6470 [00:03<1:06:41,  1.62it/s]
2|6|Loss: 0.8619758486747742:   0%|          | 5/6470 [00:03<1:06:41,  1.62it/s]
2|6|Loss: 0.8619758486747742:   0%|          | 6/6470 [00:03<1:09:49,  1.54it/s]
2|7|Loss: 0.9184839129447937:   0%|          | 6/6470 [00:04<1:09:49,  1.54it/s]
2|7|Loss: 0.9184839129447937:   0%|          | 7/6470 [00:04<1:12:59,  1.48it/s]
2|8|Loss: 0.8528720140457153:   0%|          | 7/6470 [00:04<1:12:59,  1.48it/s]
2|8|Loss: 0.8528720140457153:   0%|          | 8/6470 [00:05<1:16:16,  1.41it/s]
2|9|Loss: 0.7252534627914429:   0%|          | 8/6470 [00:05<1:16:16,  1.41it/s]
2|9|Loss: 0.7252534627914429:   0%|          | 9/6470 [00:06<1:17:24,  1.39it/s]
2|10|Loss: 1.0024292469024658:   0%|          | 9/6470 [00:06<1:17:24,  1.39it/s]
2|10|Loss: 1.0024292469024658:   0%|          | 10/6470 [00:06<1:20:25,  1.34it/s]
2|11|Loss: 0.9447494745254517:   0%|          | 10/6470 [00:07<1:20:25,  1.34it/s]
2|11|Loss: 0.9447494745254517:   0%|          | 11/6470 [00:07<1:17:22,  1.39it/s]
2|12|Loss: 0.8989378809928894:   0%|          | 11/6470 [00:07<1:17:22,  1.39it/s]
2|12|Loss: 0.8989378809928894:   0%|          | 12/6470 [00:08<1:16:40,  1.40it/s]
2|13|Loss: 0.9172223806381226:   0%|          | 12/6470 [00:08<1:16:40,  1.40it/s]
2|13|Loss: 0.9172223806381226:   0%|          | 13/6470 [00:09<1:20:55,  1.33it/s]
2|14|Loss: 0.9808290004730225:   0%|          | 13/6470 [00:09<1:20:55,  1.33it/s]
2|14|Loss: 0.9808290004730225:   0%|          | 14/6470 [00:09<1:21:16,  1.32it/s]
2|15|Loss: 0.7544424533843994:   0%|          | 14/6470 [00:10<1:21:16,  1.32it/s]
2|15|Loss: 0.7544424533843994:   0%|          | 15/6470 [00:10<1:19:31,  1.35it/s]
2|16|Loss: 0.8217412829399109:   0%|          | 15/6470 [00:10<1:19:31,  1.35it/s]
2|16|Loss: 0.8217412829399109:   0%|          | 16/6470 [00:11<1:16:32,  1.41it/s]
2|17|Loss: 0.8952864408493042:   0%|          | 16/6470 [00:11<1:16:32,  1.41it/s]
2|17|Loss: 0.8952864408493042:   0%|          | 17/6470 [00:11<1:11:40,  1.50it/s]
2|18|Loss: 0.7890982031822205:   0%|          | 17/6470 [00:12<1:11:40,  1.50it/s]
2|18|Loss: 0.7890982031822205:   0%|          | 18/6470 [00:12<1:12:58,  1.47it/s]
2|19|Loss: 0.9295737147331238:   0%|          | 18/6470 [00:12<1:12:58,  1.47it/s]
2|19|Loss: 0.9295737147331238:   0%|          | 19/6470 [00:13<1:17:24,  1.39it/s]
2|20|Loss: 0.9933003187179565:   0%|          | 19/6470 [00:13<1:17:24,  1.39it/s]
2|20|Loss: 0.9933003187179565:   0%|          | 20/6470 [00:14<1:19:55,  1.35it/s]
2|21|Loss: 0.8285748958587646:   0%|          | 20/6470 [00:14<1:19:55,  1.35it/s]
2|21|Loss: 0.8285748958587646:   0%|          | 21/6470 [00:14<1:17:31,  1.39it/s]
2|22|Loss: 0.8312745094299316:   0%|          | 21/6470 [00:15<1:17:31,  1.39it/s]
2|22|Loss: 0.8312745094299316:   0%|          | 22/6470 [00:15<1:14:42,  1.44it/s]
2|23|Loss: 0.8528504371643066:   0%|          | 22/6470 [00:15<1:14:42,  1.44it/s]
2|23|Loss: 0.8528504371643066:   0%|          | 23/6470 [00:16<1:16:37,  1.40it/s]
2|24|Loss: 0.7993872761726379:   0%|          | 23/6470 [00:16<1:16:37,  1.40it/s]
2|24|Loss: 0.7993872761726379:   0%|          | 24/6470 [00:16<1:10:31,  1.52it/s]
2|25|Loss: 0.7585517168045044:   0%|          | 24/6470 [00:17<1:10:31,  1.52it/s]
2|25|Loss: 0.7585517168045044:   0%|          | 25/6470 [00:17<1:10:08,  1.53it/s]
2|26|Loss: 0.9548993706703186:   0%|          | 25/6470 [00:17<1:10:08,  1.53it/s]
2|26|Loss: 0.9548993706703186:   0%|          | 26/6470 [00:17<1:09:23,  1.55it/s]
2|27|Loss: 0.9889628291130066:   0%|          | 26/6470 [00:18<1:09:23,  1.55it/s]
2|27|Loss: 0.9889628291130066:   0%|          | 27/6470 [00:18<1:02:02,  1.73it/s]
2|28|Loss: 0.8816118240356445:   0%|          | 27/6470 [00:18<1:02:02,  1.73it/s]
2|28|Loss: 0.8816118240356445:   0%|          | 28/6470 [00:19<1:04:37,  1.66it/s]
2|29|Loss: 0.9100712537765503:   0%|          | 28/6470 [00:19<1:04:37,  1.66it/s]
2|29|Loss: 0.9100712537765503:   0%|          | 29/6470 [00:19<1:11:58,  1.49it/s]
2|30|Loss: 0.7306560277938843:   0%|          | 29/6470 [00:20<1:11:58,  1.49it/s]
2|30|Loss: 0.7306560277938843:   0%|          | 30/6470 [00:20<1:17:50,  1.38it/s]
2|31|Loss: 0.7811551690101624:   0%|          | 30/6470 [00:21<1:17:50,  1.38it/s]
2|31|Loss: 0.7811551690101624:   0%|          | 31/6470 [00:21<1:16:32,  1.40it/s]
2|32|Loss: 0.8215075135231018:   0%|          | 31/6470 [00:21<1:16:32,  1.40it/s]
2|32|Loss: 0.8215075135231018:   0%|          | 32/6470 [00:22<1:14:02,  1.45it/s]
2|33|Loss: 0.9483950734138489:   0%|          | 32/6470 [00:22<1:14:02,  1.45it/s]
2|33|Loss: 0.9483950734138489:   1%|          | 33/6470 [00:22<1:12:17,  1.48it/s]
2|34|Loss: 0.730743408203125:   1%|          | 33/6470 [00:23<1:12:17,  1.48it/s] 
2|34|Loss: 0.730743408203125:   1%|          | 34/6470 [00:23<1:13:53,  1.45it/s]
2|35|Loss: 0.8216740489006042:   1%|          | 34/6470 [00:23<1:13:53,  1.45it/s]
2|35|Loss: 0.8216740489006042:   1%|          | 35/6470 [00:23<1:11:33,  1.50it/s]
2|36|Loss: 0.8455624580383301:   1%|          | 35/6470 [00:24<1:11:33,  1.50it/s]
2|36|Loss: 0.8455624580383301:   1%|          | 36/6470 [00:24<1:11:54,  1.49it/s]
2|37|Loss: 0.7511022090911865:   1%|          | 36/6470 [00:25<1:11:54,  1.49it/s]
2|37|Loss: 0.7511022090911865:   1%|          | 37/6470 [00:25<1:08:07,  1.57it/s]
2|38|Loss: 0.6807731986045837:   1%|          | 37/6470 [00:25<1:08:07,  1.57it/s]
2|38|Loss: 0.6807731986045837:   1%|          | 38/6470 [00:25<1:06:38,  1.61it/s]
2|39|Loss: 0.7983988523483276:   1%|          | 38/6470 [00:26<1:06:38,  1.61it/s]
2|39|Loss: 0.7983988523483276:   1%|          | 39/6470 [00:26<1:05:09,  1.65it/s]
2|40|Loss: 0.8818194270133972:   1%|          | 39/6470 [00:26<1:05:09,  1.65it/s]
2|40|Loss: 0.8818194270133972:   1%|          | 40/6470 [00:27<1:09:53,  1.53it/s]
2|41|Loss: 0.8932469487190247:   1%|          | 40/6470 [00:27<1:09:53,  1.53it/s]
2|41|Loss: 0.8932469487190247:   1%|          | 41/6470 [00:27<1:12:18,  1.48it/s]
2|42|Loss: 0.7061992883682251:   1%|          | 41/6470 [00:28<1:12:18,  1.48it/s]
2|42|Loss: 0.7061992883682251:   1%|          | 42/6470 [00:28<1:14:05,  1.45it/s]
2|43|Loss: 0.6768705248832703:   1%|          | 42/6470 [00:29<1:14:05,  1.45it/s]
2|43|Loss: 0.6768705248832703:   1%|          | 43/6470 [00:29<1:17:42,  1.38it/s]
2|44|Loss: 0.9595452547073364:   1%|          | 43/6470 [00:29<1:17:42,  1.38it/s]
2|44|Loss: 0.9595452547073364:   1%|          | 44/6470 [00:30<1:15:31,  1.42it/s]
2|45|Loss: 0.76988685131073:   1%|          | 44/6470 [00:30<1:15:31,  1.42it/s]  
2|45|Loss: 0.76988685131073:   1%|          | 45/6470 [00:30<1:12:25,  1.48it/s]
2|46|Loss: 0.9036888480186462:   1%|          | 45/6470 [00:31<1:12:25,  1.48it/s]
2|46|Loss: 0.9036888480186462:   1%|          | 46/6470 [00:31<1:14:49,  1.43it/s]
2|47|Loss: 0.7726866602897644:   1%|          | 46/6470 [00:31<1:14:49,  1.43it/s]
2|47|Loss: 0.7726866602897644:   1%|          | 47/6470 [00:32<1:21:09,  1.32it/s]
2|48|Loss: 0.7484540343284607:   1%|          | 47/6470 [00:32<1:21:09,  1.32it/s]
2|48|Loss: 0.7484540343284607:   1%|          | 48/6470 [00:33<1:21:28,  1.31it/s]
2|49|Loss: 0.8718453049659729:   1%|          | 48/6470 [00:33<1:21:28,  1.31it/s]
2|49|Loss: 0.8718453049659729:   1%|          | 49/6470 [00:33<1:20:21,  1.33it/s]
2|50|Loss: 0.9323318600654602:   1%|          | 49/6470 [00:34<1:20:21,  1.33it/s]
2|50|Loss: 0.9323318600654602:   1%|          | 50/6470 [00:34<1:16:46,  1.39it/s]
2|51|Loss: 0.9865299463272095:   1%|          | 50/6470 [00:34<1:16:46,  1.39it/s]
2|51|Loss: 0.9865299463272095:   1%|          | 51/6470 [00:35<1:19:59,  1.34it/s]
2|52|Loss: 0.9168194532394409:   1%|          | 51/6470 [00:35<1:19:59,  1.34it/s]
2|52|Loss: 0.9168194532394409:   1%|          | 52/6470 [00:35<1:18:11,  1.37it/s]
2|53|Loss: 0.9386975765228271:   1%|          | 52/6470 [00:36<1:18:11,  1.37it/s]
2|53|Loss: 0.9386975765228271:   1%|          | 53/6470 [00:36<1:13:28,  1.46it/s]
2|54|Loss: 1.081967830657959:   1%|          | 53/6470 [00:36<1:13:28,  1.46it/s] 
2|54|Loss: 1.081967830657959:   1%|          | 54/6470 [00:37<1:16:28,  1.40it/s]
2|55|Loss: 0.8583228588104248:   1%|          | 54/6470 [00:37<1:16:28,  1.40it/s]
2|55|Loss: 0.8583228588104248:   1%|          | 55/6470 [00:37<1:13:01,  1.46it/s]
2|56|Loss: 1.0170472860336304:   1%|          | 55/6470 [00:38<1:13:01,  1.46it/s]
2|56|Loss: 1.0170472860336304:   1%|          | 56/6470 [00:38<1:09:49,  1.53it/s]
2|57|Loss: 1.0743216276168823:   1%|          | 56/6470 [00:38<1:09:49,  1.53it/s]
2|57|Loss: 1.0743216276168823:   1%|          | 57/6470 [00:39<1:06:03,  1.62it/s]
2|58|Loss: 0.9079850316047668:   1%|          | 57/6470 [00:39<1:06:03,  1.62it/s]
2|58|Loss: 0.9079850316047668:   1%|          | 58/6470 [00:39<1:07:57,  1.57it/s]
2|59|Loss: 0.877953052520752:   1%|          | 58/6470 [00:40<1:07:57,  1.57it/s] 
2|59|Loss: 0.877953052520752:   1%|          | 59/6470 [00:40<1:13:36,  1.45it/s]
2|60|Loss: 0.8876432776451111:   1%|          | 59/6470 [00:40<1:13:36,  1.45it/s]
2|60|Loss: 0.8876432776451111:   1%|          | 60/6470 [00:41<1:13:14,  1.46it/s]
2|61|Loss: 0.9066101312637329:   1%|          | 60/6470 [00:41<1:13:14,  1.46it/s]
2|61|Loss: 0.9066101312637329:   1%|          | 61/6470 [00:42<1:16:38,  1.39it/s]
2|62|Loss: 0.8649952411651611:   1%|          | 61/6470 [00:42<1:16:38,  1.39it/s]
2|62|Loss: 0.8649952411651611:   1%|          | 62/6470 [00:42<1:15:01,  1.42it/s]
2|63|Loss: 0.8548246026039124:   1%|          | 62/6470 [00:43<1:15:01,  1.42it/s]
2|63|Loss: 0.8548246026039124:   1%|          | 63/6470 [00:43<1:14:22,  1.44it/s]
2|64|Loss: 0.8725165724754333:   1%|          | 63/6470 [00:43<1:14:22,  1.44it/s]
2|64|Loss: 0.8725165724754333:   1%|          | 64/6470 [00:43<1:11:38,  1.49it/s]
2|65|Loss: 1.0248154401779175:   1%|          | 64/6470 [00:44<1:11:38,  1.49it/s]
2|65|Loss: 1.0248154401779175:   1%|          | 65/6470 [00:44<1:08:34,  1.56it/s]
2|66|Loss: 1.017005205154419:   1%|          | 65/6470 [00:44<1:08:34,  1.56it/s] 
2|66|Loss: 1.017005205154419:   1%|          | 66/6470 [00:45<1:14:03,  1.44it/s]
2|67|Loss: 0.9518964290618896:   1%|          | 66/6470 [00:45<1:14:03,  1.44it/s]
2|67|Loss: 0.9518964290618896:   1%|          | 67/6470 [00:45<1:10:34,  1.51it/s]
2|68|Loss: 0.8170270919799805:   1%|          | 67/6470 [00:46<1:10:34,  1.51it/s]
2|68|Loss: 0.8170270919799805:   1%|          | 68/6470 [00:46<1:05:56,  1.62it/s]
2|69|Loss: 0.7667479515075684:   1%|          | 68/6470 [00:46<1:05:56,  1.62it/s]
2|69|Loss: 0.7667479515075684:   1%|          | 69/6470 [00:47<1:08:16,  1.56it/s]
2|70|Loss: 0.7815788984298706:   1%|          | 69/6470 [00:47<1:08:16,  1.56it/s]
2|70|Loss: 0.7815788984298706:   1%|          | 70/6470 [00:47<1:11:15,  1.50it/s]
2|71|Loss: 0.8774847388267517:   1%|          | 70/6470 [00:48<1:11:15,  1.50it/s]
2|71|Loss: 0.8774847388267517:   1%|          | 71/6470 [00:48<1:12:39,  1.47it/s]
2|72|Loss: 0.9283586144447327:   1%|          | 71/6470 [00:48<1:12:39,  1.47it/s]
2|72|Loss: 0.9283586144447327:   1%|          | 72/6470 [00:49<1:09:55,  1.52it/s]
2|73|Loss: 0.8084491491317749:   1%|          | 72/6470 [00:49<1:09:55,  1.52it/s]
2|73|Loss: 0.8084491491317749:   1%|          | 73/6470 [00:49<1:10:49,  1.51it/s]
2|74|Loss: 0.8966262936592102:   1%|          | 73/6470 [00:50<1:10:49,  1.51it/s]
2|74|Loss: 0.8966262936592102:   1%|          | 74/6470 [00:50<1:10:10,  1.52it/s]
2|75|Loss: 1.1714680194854736:   1%|          | 74/6470 [00:50<1:10:10,  1.52it/s]
2|75|Loss: 1.1714680194854736:   1%|          | 75/6470 [00:51<1:14:29,  1.43it/s]
2|76|Loss: 0.9143086075782776:   1%|          | 75/6470 [00:51<1:14:29,  1.43it/s]
2|76|Loss: 0.9143086075782776:   1%|          | 76/6470 [00:52<1:13:46,  1.44it/s]
2|77|Loss: 0.9478273987770081:   1%|          | 76/6470 [00:52<1:13:46,  1.44it/s]
2|77|Loss: 0.9478273987770081:   1%|          | 77/6470 [00:52<1:10:18,  1.52it/s]
2|78|Loss: 0.8460044860839844:   1%|          | 77/6470 [00:52<1:10:18,  1.52it/s]
2|78|Loss: 0.8460044860839844:   1%|          | 78/6470 [00:53<1:11:44,  1.49it/s]
2|79|Loss: 0.9554065465927124:   1%|          | 78/6470 [00:53<1:11:44,  1.49it/s]
2|79|Loss: 0.9554065465927124:   1%|          | 79/6470 [00:53<1:12:36,  1.47it/s]
2|80|Loss: 0.7831667065620422:   1%|          | 79/6470 [00:54<1:12:36,  1.47it/s]
2|80|Loss: 0.7831667065620422:   1%|          | 80/6470 [00:54<1:11:39,  1.49it/s]
2|81|Loss: 0.7866426110267639:   1%|          | 80/6470 [00:55<1:11:39,  1.49it/s]
2|81|Loss: 0.7866426110267639:   1%|▏         | 81/6470 [00:55<1:13:32,  1.45it/s]
2|82|Loss: 0.8385739922523499:   1%|▏         | 81/6470 [00:55<1:13:32,  1.45it/s]
2|82|Loss: 0.8385739922523499:   1%|▏         | 82/6470 [00:56<1:14:44,  1.42it/s]
2|83|Loss: 0.9785588979721069:   1%|▏         | 82/6470 [00:56<1:14:44,  1.42it/s]
2|83|Loss: 0.9785588979721069:   1%|▏         | 83/6470 [00:56<1:08:19,  1.56it/s]
2|84|Loss: 0.8100414276123047:   1%|▏         | 83/6470 [00:56<1:08:19,  1.56it/s]
2|84|Loss: 0.8100414276123047:   1%|▏         | 84/6470 [00:57<1:04:32,  1.65it/s]
2|85|Loss: 1.1406333446502686:   1%|▏         | 84/6470 [00:57<1:04:32,  1.65it/s]
2|85|Loss: 1.1406333446502686:   1%|▏         | 85/6470 [00:57<1:06:36,  1.60it/s]
2|86|Loss: 0.7713616490364075:   1%|▏         | 85/6470 [00:58<1:06:36,  1.60it/s]
2|86|Loss: 0.7713616490364075:   1%|▏         | 86/6470 [00:58<1:11:05,  1.50it/s]
2|87|Loss: 0.8975955843925476:   1%|▏         | 86/6470 [00:58<1:11:05,  1.50it/s]
2|87|Loss: 0.8975955843925476:   1%|▏         | 87/6470 [00:59<1:11:21,  1.49it/s]
2|88|Loss: 0.8538874983787537:   1%|▏         | 87/6470 [00:59<1:11:21,  1.49it/s]
2|88|Loss: 0.8538874983787537:   1%|▏         | 88/6470 [00:59<1:12:59,  1.46it/s]
2|89|Loss: 0.8832960724830627:   1%|▏         | 88/6470 [01:00<1:12:59,  1.46it/s]
2|89|Loss: 0.8832960724830627:   1%|▏         | 89/6470 [01:00<1:07:30,  1.58it/s]
2|90|Loss: 0.9212388396263123:   1%|▏         | 89/6470 [01:00<1:07:30,  1.58it/s]
2|90|Loss: 0.9212388396263123:   1%|▏         | 90/6470 [01:01<1:11:35,  1.49it/s]
2|91|Loss: 0.9475036859512329:   1%|▏         | 90/6470 [01:01<1:11:35,  1.49it/s]
2|91|Loss: 0.9475036859512329:   1%|▏         | 91/6470 [01:02<1:13:56,  1.44it/s]
2|92|Loss: 0.9394129514694214:   1%|▏         | 91/6470 [01:02<1:13:56,  1.44it/s]
2|92|Loss: 0.9394129514694214:   1%|▏         | 92/6470 [01:02<1:16:38,  1.39it/s]
2|93|Loss: 0.7348156571388245:   1%|▏         | 92/6470 [01:03<1:16:38,  1.39it/s]
2|93|Loss: 0.7348156571388245:   1%|▏         | 93/6470 [01:03<1:20:39,  1.32it/s]
2|94|Loss: 0.8273200392723083:   1%|▏         | 93/6470 [01:03<1:20:39,  1.32it/s]
2|94|Loss: 0.8273200392723083:   1%|▏         | 94/6470 [01:04<1:11:55,  1.48it/s]
2|95|Loss: 0.8202348351478577:   1%|▏         | 94/6470 [01:04<1:11:55,  1.48it/s]
2|95|Loss: 0.8202348351478577:   1%|▏         | 95/6470 [01:04<1:07:46,  1.57it/s]
2|96|Loss: 0.8853794932365417:   1%|▏         | 95/6470 [01:05<1:07:46,  1.57it/s]
2|96|Loss: 0.8853794932365417:   1%|▏         | 96/6470 [01:05<1:08:56,  1.54it/s]
2|97|Loss: 0.7139900922775269:   1%|▏         | 96/6470 [01:05<1:08:56,  1.54it/s]
2|97|Loss: 0.7139900922775269:   1%|▏         | 97/6470 [01:06<1:14:03,  1.43it/s]
2|98|Loss: 0.9158797860145569:   1%|▏         | 97/6470 [01:06<1:14:03,  1.43it/s]
2|98|Loss: 0.9158797860145569:   2%|▏         | 98/6470 [01:06<1:13:38,  1.44it/s]
2|99|Loss: 0.8471909761428833:   2%|▏         | 98/6470 [01:07<1:13:38,  1.44it/s]
2|99|Loss: 0.8471909761428833:   2%|▏         | 99/6470 [01:07<1:11:24,  1.49it/s]
2|100|Loss: 1.0012627840042114:   2%|▏         | 99/6470 [01:07<1:11:24,  1.49it/s]
2|100|Loss: 1.0012627840042114:   2%|▏         | 100/6470 [01:08<1:14:55,  1.42it/s]
2|101|Loss: 0.8464065194129944:   2%|▏         | 100/6470 [01:08<1:14:55,  1.42it/s]
2|101|Loss: 0.8464065194129944:   2%|▏         | 101/6470 [01:09<1:17:54,  1.36it/s]
2|102|Loss: 0.9973405599594116:   2%|▏         | 101/6470 [01:09<1:17:54,  1.36it/s]
2|102|Loss: 0.9973405599594116:   2%|▏         | 102/6470 [01:09<1:18:53,  1.35it/s]
2|103|Loss: 0.864722728729248:   2%|▏         | 102/6470 [01:10<1:18:53,  1.35it/s] 
2|103|Loss: 0.864722728729248:   2%|▏         | 103/6470 [01:10<1:09:23,  1.53it/s]
2|104|Loss: 0.845206081867218:   2%|▏         | 103/6470 [01:10<1:09:23,  1.53it/s]
2|104|Loss: 0.845206081867218:   2%|▏         | 104/6470 [01:10<1:12:11,  1.47it/s]
2|105|Loss: 0.8373737335205078:   2%|▏         | 104/6470 [01:11<1:12:11,  1.47it/s]
2|105|Loss: 0.8373737335205078:   2%|▏         | 105/6470 [01:11<1:10:39,  1.50it/s]
2|106|Loss: 1.023986577987671:   2%|▏         | 105/6470 [01:11<1:10:39,  1.50it/s] 
2|106|Loss: 1.023986577987671:   2%|▏         | 106/6470 [01:12<1:08:46,  1.54it/s]
2|107|Loss: 0.8043748140335083:   2%|▏         | 106/6470 [01:12<1:08:46,  1.54it/s]
2|107|Loss: 0.8043748140335083:   2%|▏         | 107/6470 [01:12<1:10:38,  1.50it/s]
2|108|Loss: 0.9761572480201721:   2%|▏         | 107/6470 [01:13<1:10:38,  1.50it/s]
2|108|Loss: 0.9761572480201721:   2%|▏         | 108/6470 [01:13<1:10:16,  1.51it/s]
2|109|Loss: 0.7940664291381836:   2%|▏         | 108/6470 [01:13<1:10:16,  1.51it/s]
2|109|Loss: 0.7940664291381836:   2%|▏         | 109/6470 [01:14<1:05:30,  1.62it/s]
2|110|Loss: 0.9931761026382446:   2%|▏         | 109/6470 [01:14<1:05:30,  1.62it/s]
2|110|Loss: 0.9931761026382446:   2%|▏         | 110/6470 [01:14<1:07:26,  1.57it/s]
2|111|Loss: 0.9121724963188171:   2%|▏         | 110/6470 [01:15<1:07:26,  1.57it/s]
2|111|Loss: 0.9121724963188171:   2%|▏         | 111/6470 [01:15<1:13:33,  1.44it/s]
2|112|Loss: 0.8852043747901917:   2%|▏         | 111/6470 [01:16<1:13:33,  1.44it/s]
2|112|Loss: 0.8852043747901917:   2%|▏         | 112/6470 [01:16<1:15:33,  1.40it/s]
2|113|Loss: 0.8900025486946106:   2%|▏         | 112/6470 [01:16<1:15:33,  1.40it/s]
2|113|Loss: 0.8900025486946106:   2%|▏         | 113/6470 [01:17<1:15:31,  1.40it/s]
2|114|Loss: 0.6726068258285522:   2%|▏         | 113/6470 [01:17<1:15:31,  1.40it/s]
2|114|Loss: 0.6726068258285522:   2%|▏         | 114/6470 [01:17<1:13:10,  1.45it/s]
2|115|Loss: 0.9808498024940491:   2%|▏         | 114/6470 [01:18<1:13:10,  1.45it/s]
2|115|Loss: 0.9808498024940491:   2%|▏         | 115/6470 [01:18<1:09:38,  1.52it/s]
2|116|Loss: 0.8988686203956604:   2%|▏         | 115/6470 [01:18<1:09:38,  1.52it/s]
2|116|Loss: 0.8988686203956604:   2%|▏         | 116/6470 [01:18<1:08:43,  1.54it/s]
2|117|Loss: 0.8321914672851562:   2%|▏         | 116/6470 [01:19<1:08:43,  1.54it/s]
2|117|Loss: 0.8321914672851562:   2%|▏         | 117/6470 [01:19<1:11:12,  1.49it/s]
2|118|Loss: 1.0103564262390137:   2%|▏         | 117/6470 [01:20<1:11:12,  1.49it/s]
2|118|Loss: 1.0103564262390137:   2%|▏         | 118/6470 [01:20<1:16:01,  1.39it/s]
2|119|Loss: 0.9482384920120239:   2%|▏         | 118/6470 [01:20<1:16:01,  1.39it/s]
2|119|Loss: 0.9482384920120239:   2%|▏         | 119/6470 [01:21<1:20:15,  1.32it/s]
2|120|Loss: 0.8196592330932617:   2%|▏         | 119/6470 [01:21<1:20:15,  1.32it/s]
2|120|Loss: 0.8196592330932617:   2%|▏         | 120/6470 [01:22<1:21:51,  1.29it/s]
2|121|Loss: 0.8491711616516113:   2%|▏         | 120/6470 [01:22<1:21:51,  1.29it/s]
2|121|Loss: 0.8491711616516113:   2%|▏         | 121/6470 [01:22<1:16:29,  1.38it/s]
2|122|Loss: 0.9961144924163818:   2%|▏         | 121/6470 [01:23<1:16:29,  1.38it/s]
2|122|Loss: 0.9961144924163818:   2%|▏         | 122/6470 [01:23<1:12:16,  1.46it/s]
2|123|Loss: 0.9219593405723572:   2%|▏         | 122/6470 [01:23<1:12:16,  1.46it/s]
2|123|Loss: 0.9219593405723572:   2%|▏         | 123/6470 [01:23<1:07:33,  1.57it/s]
2|124|Loss: 0.9801504611968994:   2%|▏         | 123/6470 [01:24<1:07:33,  1.57it/s]
2|124|Loss: 0.9801504611968994:   2%|▏         | 124/6470 [01:24<1:09:26,  1.52it/s]
2|125|Loss: 1.0075325965881348:   2%|▏         | 124/6470 [01:24<1:09:26,  1.52it/s]
2|125|Loss: 1.0075325965881348:   2%|▏         | 125/6470 [01:25<1:12:28,  1.46it/s]
2|126|Loss: 0.773645281791687:   2%|▏         | 125/6470 [01:25<1:12:28,  1.46it/s] 
2|126|Loss: 0.773645281791687:   2%|▏         | 126/6470 [01:25<1:08:35,  1.54it/s]
2|127|Loss: 0.9497044086456299:   2%|▏         | 126/6470 [01:26<1:08:35,  1.54it/s]
2|127|Loss: 0.9497044086456299:   2%|▏         | 127/6470 [01:26<1:09:43,  1.52it/s]
2|128|Loss: 0.9332790970802307:   2%|▏         | 127/6470 [01:26<1:09:43,  1.52it/s]
2|128|Loss: 0.9332790970802307:   2%|▏         | 128/6470 [01:27<1:09:37,  1.52it/s]
2|129|Loss: 0.8716613054275513:   2%|▏         | 128/6470 [01:27<1:09:37,  1.52it/s]
2|129|Loss: 0.8716613054275513:   2%|▏         | 129/6470 [01:27<1:10:33,  1.50it/s]
2|130|Loss: 0.8531779646873474:   2%|▏         | 129/6470 [01:28<1:10:33,  1.50it/s]
2|130|Loss: 0.8531779646873474:   2%|▏         | 130/6470 [01:28<1:15:57,  1.39it/s]
2|131|Loss: 0.9534035325050354:   2%|▏         | 130/6470 [01:29<1:15:57,  1.39it/s]
2|131|Loss: 0.9534035325050354:   2%|▏         | 131/6470 [01:29<1:18:02,  1.35it/s]
2|132|Loss: 0.8515557050704956:   2%|▏         | 131/6470 [01:29<1:18:02,  1.35it/s]
2|132|Loss: 0.8515557050704956:   2%|▏         | 132/6470 [01:30<1:10:57,  1.49it/s]
2|133|Loss: 0.895870566368103:   2%|▏         | 132/6470 [01:30<1:10:57,  1.49it/s] 
2|133|Loss: 0.895870566368103:   2%|▏         | 133/6470 [01:30<1:08:35,  1.54it/s]
2|134|Loss: 0.7956737279891968:   2%|▏         | 133/6470 [01:31<1:08:35,  1.54it/s]
2|134|Loss: 0.7956737279891968:   2%|▏         | 134/6470 [01:31<1:10:40,  1.49it/s]
2|135|Loss: 0.9711563587188721:   2%|▏         | 134/6470 [01:31<1:10:40,  1.49it/s]
2|135|Loss: 0.9711563587188721:   2%|▏         | 135/6470 [01:32<1:15:10,  1.40it/s]
2|136|Loss: 1.0033749341964722:   2%|▏         | 135/6470 [01:32<1:15:10,  1.40it/s]
2|136|Loss: 1.0033749341964722:   2%|▏         | 136/6470 [01:32<1:10:18,  1.50it/s]
2|137|Loss: 0.8018050193786621:   2%|▏         | 136/6470 [01:33<1:10:18,  1.50it/s]
2|137|Loss: 0.8018050193786621:   2%|▏         | 137/6470 [01:33<1:15:26,  1.40it/s]
2|138|Loss: 0.8638302683830261:   2%|▏         | 137/6470 [01:34<1:15:26,  1.40it/s]
2|138|Loss: 0.8638302683830261:   2%|▏         | 138/6470 [01:34<1:16:39,  1.38it/s]
2|139|Loss: 0.8951249718666077:   2%|▏         | 138/6470 [01:34<1:16:39,  1.38it/s]
2|139|Loss: 0.8951249718666077:   2%|▏         | 139/6470 [01:34<1:12:05,  1.46it/s]
2|140|Loss: 0.9327173829078674:   2%|▏         | 139/6470 [01:35<1:12:05,  1.46it/s]
2|140|Loss: 0.9327173829078674:   2%|▏         | 140/6470 [01:35<1:10:17,  1.50it/s]
2|141|Loss: 1.0044538974761963:   2%|▏         | 140/6470 [01:35<1:10:17,  1.50it/s]
2|141|Loss: 1.0044538974761963:   2%|▏         | 141/6470 [01:36<1:08:17,  1.54it/s]
2|142|Loss: 0.8960760235786438:   2%|▏         | 141/6470 [01:36<1:08:17,  1.54it/s]
2|142|Loss: 0.8960760235786438:   2%|▏         | 142/6470 [01:36<1:02:56,  1.68it/s]
2|143|Loss: 0.7740933299064636:   2%|▏         | 142/6470 [01:36<1:02:56,  1.68it/s]
2|143|Loss: 0.7740933299064636:   2%|▏         | 143/6470 [01:37<59:39,  1.77it/s]  
2|144|Loss: 0.9357832670211792:   2%|▏         | 143/6470 [01:37<59:39,  1.77it/s]
2|144|Loss: 0.9357832670211792:   2%|▏         | 144/6470 [01:37<56:29,  1.87it/s]
2|145|Loss: 0.9962561726570129:   2%|▏         | 144/6470 [01:37<56:29,  1.87it/s]
2|145|Loss: 0.9962561726570129:   2%|▏         | 145/6470 [01:38<58:29,  1.80it/s]
2|146|Loss: 0.9216543436050415:   2%|▏         | 145/6470 [01:38<58:29,  1.80it/s]
2|146|Loss: 0.9216543436050415:   2%|▏         | 146/6470 [01:38<1:03:48,  1.65it/s]
2|147|Loss: 0.8958138227462769:   2%|▏         | 146/6470 [01:39<1:03:48,  1.65it/s]
2|147|Loss: 0.8958138227462769:   2%|▏         | 147/6470 [01:39<1:03:20,  1.66it/s]
2|148|Loss: 0.9712257385253906:   2%|▏         | 147/6470 [01:39<1:03:20,  1.66it/s]
2|148|Loss: 0.9712257385253906:   2%|▏         | 148/6470 [01:39<57:07,  1.84it/s]  
2|149|Loss: 0.9553819298744202:   2%|▏         | 148/6470 [01:40<57:07,  1.84it/s]
2|149|Loss: 0.9553819298744202:   2%|▏         | 149/6470 [01:40<1:02:38,  1.68it/s]
2|150|Loss: 1.001118779182434:   2%|▏         | 149/6470 [01:40<1:02:38,  1.68it/s] 
2|150|Loss: 1.001118779182434:   2%|▏         | 150/6470 [01:41<1:03:48,  1.65it/s]
2|151|Loss: 0.797694742679596:   2%|▏         | 150/6470 [01:41<1:03:48,  1.65it/s]
2|151|Loss: 0.797694742679596:   2%|▏         | 151/6470 [01:41<1:06:10,  1.59it/s]
2|152|Loss: 0.9711226224899292:   2%|▏         | 151/6470 [01:42<1:06:10,  1.59it/s]
2|152|Loss: 0.9711226224899292:   2%|▏         | 152/6470 [01:42<1:09:36,  1.51it/s]
2|153|Loss: 0.9565275311470032:   2%|▏         | 152/6470 [01:43<1:09:36,  1.51it/s]
2|153|Loss: 0.9565275311470032:   2%|▏         | 153/6470 [01:43<1:06:31,  1.58it/s]
2|154|Loss: 0.7862483263015747:   2%|▏         | 153/6470 [01:43<1:06:31,  1.58it/s]
2|154|Loss: 0.7862483263015747:   2%|▏         | 154/6470 [01:43<1:04:23,  1.63it/s]
2|155|Loss: 0.8440855741500854:   2%|▏         | 154/6470 [01:44<1:04:23,  1.63it/s]
2|155|Loss: 0.8440855741500854:   2%|▏         | 155/6470 [01:44<1:04:17,  1.64it/s]
2|156|Loss: 0.7403160929679871:   2%|▏         | 155/6470 [01:44<1:04:17,  1.64it/s]
2|156|Loss: 0.7403160929679871:   2%|▏         | 156/6470 [01:45<1:04:35,  1.63it/s]
2|157|Loss: 0.7704368233680725:   2%|▏         | 156/6470 [01:45<1:04:35,  1.63it/s]
2|157|Loss: 0.7704368233680725:   2%|▏         | 157/6470 [01:45<1:03:56,  1.65it/s]
2|158|Loss: 0.875307023525238:   2%|▏         | 157/6470 [01:46<1:03:56,  1.65it/s] 
2|158|Loss: 0.875307023525238:   2%|▏         | 158/6470 [01:46<1:09:24,  1.52it/s]
2|159|Loss: 0.8512083292007446:   2%|▏         | 158/6470 [01:46<1:09:24,  1.52it/s]
2|159|Loss: 0.8512083292007446:   2%|▏         | 159/6470 [01:47<1:13:57,  1.42it/s]
2|160|Loss: 1.008562445640564:   2%|▏         | 159/6470 [01:47<1:13:57,  1.42it/s] 
2|160|Loss: 1.008562445640564:   2%|▏         | 160/6470 [01:47<1:14:34,  1.41it/s]
2|161|Loss: 0.8678188323974609:   2%|▏         | 160/6470 [01:48<1:14:34,  1.41it/s]
2|161|Loss: 0.8678188323974609:   2%|▏         | 161/6470 [01:48<1:15:43,  1.39it/s]
2|162|Loss: 0.8983203768730164:   2%|▏         | 161/6470 [01:49<1:15:43,  1.39it/s]
2|162|Loss: 0.8983203768730164:   3%|▎         | 162/6470 [01:49<1:14:09,  1.42it/s]
2|163|Loss: 0.9962658882141113:   3%|▎         | 162/6470 [01:49<1:14:09,  1.42it/s]
2|163|Loss: 0.9962658882141113:   3%|▎         | 163/6470 [01:50<1:14:29,  1.41it/s]
2|164|Loss: 0.9070631861686707:   3%|▎         | 163/6470 [01:50<1:14:29,  1.41it/s]
2|164|Loss: 0.9070631861686707:   3%|▎         | 164/6470 [01:50<1:12:58,  1.44it/s]
2|165|Loss: 0.8334789872169495:   3%|▎         | 164/6470 [01:51<1:12:58,  1.44it/s]
2|165|Loss: 0.8334789872169495:   3%|▎         | 165/6470 [01:51<1:09:51,  1.50it/s]
2|166|Loss: 0.8344408869743347:   3%|▎         | 165/6470 [01:51<1:09:51,  1.50it/s]
2|166|Loss: 0.8344408869743347:   3%|▎         | 166/6470 [01:51<1:05:24,  1.61it/s]
2|167|Loss: 0.9721439480781555:   3%|▎         | 166/6470 [01:52<1:05:24,  1.61it/s]
2|167|Loss: 0.9721439480781555:   3%|▎         | 167/6470 [01:52<58:13,  1.80it/s]  
2|168|Loss: 0.9460729360580444:   3%|▎         | 167/6470 [01:52<58:13,  1.80it/s]
2|168|Loss: 0.9460729360580444:   3%|▎         | 168/6470 [01:52<59:22,  1.77it/s]
2|169|Loss: 0.9628586769104004:   3%|▎         | 168/6470 [01:53<59:22,  1.77it/s]
2|169|Loss: 0.9628586769104004:   3%|▎         | 169/6470 [01:53<1:05:34,  1.60it/s]
2|170|Loss: 0.8977766633033752:   3%|▎         | 169/6470 [01:53<1:05:34,  1.60it/s]
2|170|Loss: 0.8977766633033752:   3%|▎         | 170/6470 [01:54<1:07:08,  1.56it/s]
2|171|Loss: 0.9629372358322144:   3%|▎         | 170/6470 [01:54<1:07:08,  1.56it/s]
2|171|Loss: 0.9629372358322144:   3%|▎         | 171/6470 [01:55<1:11:17,  1.47it/s]
2|172|Loss: 0.8104771375656128:   3%|▎         | 171/6470 [01:55<1:11:17,  1.47it/s]
2|172|Loss: 0.8104771375656128:   3%|▎         | 172/6470 [01:55<1:16:58,  1.36it/s]
2|173|Loss: 0.9159033298492432:   3%|▎         | 172/6470 [01:56<1:16:58,  1.36it/s]
2|173|Loss: 0.9159033298492432:   3%|▎         | 173/6470 [01:56<1:11:59,  1.46it/s]
2|174|Loss: 0.7453858852386475:   3%|▎         | 173/6470 [01:56<1:11:59,  1.46it/s]
2|174|Loss: 0.7453858852386475:   3%|▎         | 174/6470 [01:57<1:11:02,  1.48it/s]
2|175|Loss: 0.9426302909851074:   3%|▎         | 174/6470 [01:57<1:11:02,  1.48it/s]
2|175|Loss: 0.9426302909851074:   3%|▎         | 175/6470 [01:57<1:12:03,  1.46it/s]
2|176|Loss: 1.0395668745040894:   3%|▎         | 175/6470 [01:58<1:12:03,  1.46it/s]
2|176|Loss: 1.0395668745040894:   3%|▎         | 176/6470 [01:58<1:14:53,  1.40it/s]
2|177|Loss: 0.8027358651161194:   3%|▎         | 176/6470 [01:59<1:14:53,  1.40it/s]
2|177|Loss: 0.8027358651161194:   3%|▎         | 177/6470 [01:59<1:15:15,  1.39it/s]
2|178|Loss: 1.1027381420135498:   3%|▎         | 177/6470 [01:59<1:15:15,  1.39it/s]
2|178|Loss: 1.1027381420135498:   3%|▎         | 178/6470 [01:59<1:11:55,  1.46it/s]
2|179|Loss: 0.9552019238471985:   3%|▎         | 178/6470 [02:00<1:11:55,  1.46it/s]
2|179|Loss: 0.9552019238471985:   3%|▎         | 179/6470 [02:00<1:09:39,  1.51it/s]
2|180|Loss: 0.7392563223838806:   3%|▎         | 179/6470 [02:00<1:09:39,  1.51it/s]
2|180|Loss: 0.7392563223838806:   3%|▎         | 180/6470 [02:01<1:05:25,  1.60it/s]
2|181|Loss: 0.9441819190979004:   3%|▎         | 180/6470 [02:01<1:05:25,  1.60it/s]
2|181|Loss: 0.9441819190979004:   3%|▎         | 181/6470 [02:01<1:05:05,  1.61it/s]
2|182|Loss: 0.821040153503418:   3%|▎         | 181/6470 [02:02<1:05:05,  1.61it/s] 
2|182|Loss: 0.821040153503418:   3%|▎         | 182/6470 [02:02<1:08:46,  1.52it/s]
2|183|Loss: 0.9073567986488342:   3%|▎         | 182/6470 [02:02<1:08:46,  1.52it/s]
2|183|Loss: 0.9073567986488342:   3%|▎         | 183/6470 [02:03<1:14:23,  1.41it/s]
2|184|Loss: 0.7880430817604065:   3%|▎         | 183/6470 [02:03<1:14:23,  1.41it/s]
2|184|Loss: 0.7880430817604065:   3%|▎         | 184/6470 [02:04<1:16:38,  1.37it/s]
2|185|Loss: 0.9887820482254028:   3%|▎         | 184/6470 [02:04<1:16:38,  1.37it/s]
2|185|Loss: 0.9887820482254028:   3%|▎         | 185/6470 [02:04<1:14:09,  1.41it/s]
2|186|Loss: 0.8672689199447632:   3%|▎         | 185/6470 [02:05<1:14:09,  1.41it/s]
2|186|Loss: 0.8672689199447632:   3%|▎         | 186/6470 [02:05<1:15:52,  1.38it/s]
2|187|Loss: 0.9022627472877502:   3%|▎         | 186/6470 [02:05<1:15:52,  1.38it/s]
2|187|Loss: 0.9022627472877502:   3%|▎         | 187/6470 [02:06<1:15:59,  1.38it/s]
2|188|Loss: 0.8391472697257996:   3%|▎         | 187/6470 [02:06<1:15:59,  1.38it/s]
2|188|Loss: 0.8391472697257996:   3%|▎         | 188/6470 [02:06<1:16:36,  1.37it/s]
2|189|Loss: 0.9980823397636414:   3%|▎         | 188/6470 [02:07<1:16:36,  1.37it/s]
2|189|Loss: 0.9980823397636414:   3%|▎         | 189/6470 [02:07<1:14:45,  1.40it/s]
2|190|Loss: 0.751445472240448:   3%|▎         | 189/6470 [02:08<1:14:45,  1.40it/s] 
2|190|Loss: 0.751445472240448:   3%|▎         | 190/6470 [02:08<1:16:27,  1.37it/s]
2|191|Loss: 0.7870774269104004:   3%|▎         | 190/6470 [02:08<1:16:27,  1.37it/s]
2|191|Loss: 0.7870774269104004:   3%|▎         | 191/6470 [02:09<1:14:03,  1.41it/s]
2|192|Loss: 0.7353643178939819:   3%|▎         | 191/6470 [02:09<1:14:03,  1.41it/s]
2|192|Loss: 0.7353643178939819:   3%|▎         | 192/6470 [02:09<1:04:40,  1.62it/s]
2|193|Loss: 0.764154851436615:   3%|▎         | 192/6470 [02:09<1:04:40,  1.62it/s] 
2|193|Loss: 0.764154851436615:   3%|▎         | 193/6470 [02:09<1:00:32,  1.73it/s]
2|194|Loss: 0.822541356086731:   3%|▎         | 193/6470 [02:10<1:00:32,  1.73it/s]
2|194|Loss: 0.822541356086731:   3%|▎         | 194/6470 [02:10<1:07:28,  1.55it/s]
2|195|Loss: 0.7921328544616699:   3%|▎         | 194/6470 [02:11<1:07:28,  1.55it/s]
2|195|Loss: 0.7921328544616699:   3%|▎         | 195/6470 [02:11<1:11:10,  1.47it/s]
2|196|Loss: 0.6398217082023621:   3%|▎         | 195/6470 [02:11<1:11:10,  1.47it/s]
2|196|Loss: 0.6398217082023621:   3%|▎         | 196/6470 [02:12<1:09:39,  1.50it/s]
2|197|Loss: 0.8363714218139648:   3%|▎         | 196/6470 [02:12<1:09:39,  1.50it/s]
2|197|Loss: 0.8363714218139648:   3%|▎         | 197/6470 [02:12<1:06:47,  1.57it/s]
2|198|Loss: 0.7557368874549866:   3%|▎         | 197/6470 [02:13<1:06:47,  1.57it/s]
2|198|Loss: 0.7557368874549866:   3%|▎         | 198/6470 [02:13<1:09:01,  1.51it/s]
2|199|Loss: 0.8875206708908081:   3%|▎         | 198/6470 [02:13<1:09:01,  1.51it/s]
2|199|Loss: 0.8875206708908081:   3%|▎         | 199/6470 [02:14<1:11:47,  1.46it/s]
2|200|Loss: 0.8763412833213806:   3%|▎         | 199/6470 [02:14<1:11:47,  1.46it/s]
2|200|Loss: 0.8763412833213806:   3%|▎         | 200/6470 [02:14<1:10:37,  1.48it/s]
2|201|Loss: 0.8248373866081238:   3%|▎         | 200/6470 [02:15<1:10:37,  1.48it/s]
2|201|Loss: 0.8248373866081238:   3%|▎         | 201/6470 [02:15<1:14:35,  1.40it/s]
2|202|Loss: 0.948196291923523:   3%|▎         | 201/6470 [02:16<1:14:35,  1.40it/s] 
2|202|Loss: 0.948196291923523:   3%|▎         | 202/6470 [02:16<1:17:45,  1.34it/s]
2|203|Loss: 0.9163667559623718:   3%|▎         | 202/6470 [02:16<1:17:45,  1.34it/s]
2|203|Loss: 0.9163667559623718:   3%|▎         | 203/6470 [02:17<1:12:20,  1.44it/s]
2|204|Loss: 0.6457273960113525:   3%|▎         | 203/6470 [02:17<1:12:20,  1.44it/s]
2|204|Loss: 0.6457273960113525:   3%|▎         | 204/6470 [02:17<1:06:56,  1.56it/s]
2|205|Loss: 0.8202836513519287:   3%|▎         | 204/6470 [02:17<1:06:56,  1.56it/s]
2|205|Loss: 0.8202836513519287:   3%|▎         | 205/6470 [02:18<1:07:32,  1.55it/s]
2|206|Loss: 0.8955543041229248:   3%|▎         | 205/6470 [02:18<1:07:32,  1.55it/s]
2|206|Loss: 0.8955543041229248:   3%|▎         | 206/6470 [02:18<1:07:16,  1.55it/s]
2|207|Loss: 0.9880538582801819:   3%|▎         | 206/6470 [02:19<1:07:16,  1.55it/s]
2|207|Loss: 0.9880538582801819:   3%|▎         | 207/6470 [02:19<1:09:26,  1.50it/s]
2|208|Loss: 0.701509952545166:   3%|▎         | 207/6470 [02:20<1:09:26,  1.50it/s] 
2|208|Loss: 0.701509952545166:   3%|▎         | 208/6470 [02:20<1:16:14,  1.37it/s]
2|209|Loss: 0.8972468376159668:   3%|▎         | 208/6470 [02:20<1:16:14,  1.37it/s]
2|209|Loss: 0.8972468376159668:   3%|▎         | 209/6470 [02:21<1:12:29,  1.44it/s]
2|210|Loss: 0.702019989490509:   3%|▎         | 209/6470 [02:21<1:12:29,  1.44it/s] 
2|210|Loss: 0.702019989490509:   3%|▎         | 210/6470 [02:21<1:15:53,  1.37it/s]
2|211|Loss: 0.897336483001709:   3%|▎         | 210/6470 [02:22<1:15:53,  1.37it/s]
2|211|Loss: 0.897336483001709:   3%|▎         | 211/6470 [02:22<1:16:22,  1.37it/s]
2|212|Loss: 0.8193906545639038:   3%|▎         | 211/6470 [02:23<1:16:22,  1.37it/s]
2|212|Loss: 0.8193906545639038:   3%|▎         | 212/6470 [02:23<1:19:54,  1.31it/s]
2|213|Loss: 0.8342573046684265:   3%|▎         | 212/6470 [02:23<1:19:54,  1.31it/s]
2|213|Loss: 0.8342573046684265:   3%|▎         | 213/6470 [02:24<1:17:01,  1.35it/s]
2|214|Loss: 0.9567940831184387:   3%|▎         | 213/6470 [02:24<1:17:01,  1.35it/s]
2|214|Loss: 0.9567940831184387:   3%|▎         | 214/6470 [02:24<1:16:11,  1.37it/s]
2|215|Loss: 1.4358779191970825:   3%|▎         | 214/6470 [02:25<1:16:11,  1.37it/s]
2|215|Loss: 1.4358779191970825:   3%|▎         | 215/6470 [02:25<1:18:00,  1.34it/s]
2|216|Loss: 0.8129514455795288:   3%|▎         | 215/6470 [02:26<1:18:00,  1.34it/s]
2|216|Loss: 0.8129514455795288:   3%|▎         | 216/6470 [02:26<1:21:30,  1.28it/s]
2|217|Loss: 0.8943487405776978:   3%|▎         | 216/6470 [02:26<1:21:30,  1.28it/s]
2|217|Loss: 0.8943487405776978:   3%|▎         | 217/6470 [02:27<1:21:13,  1.28it/s]
2|218|Loss: 0.7944768071174622:   3%|▎         | 217/6470 [02:27<1:21:13,  1.28it/s]
2|218|Loss: 0.7944768071174622:   3%|▎         | 218/6470 [02:28<1:21:39,  1.28it/s]
2|219|Loss: 1.012912392616272:   3%|▎         | 218/6470 [02:28<1:21:39,  1.28it/s] 
2|219|Loss: 1.012912392616272:   3%|▎         | 219/6470 [02:28<1:17:39,  1.34it/s]
2|220|Loss: 0.8864262104034424:   3%|▎         | 219/6470 [02:29<1:17:39,  1.34it/s]
2|220|Loss: 0.8864262104034424:   3%|▎         | 220/6470 [02:29<1:13:21,  1.42it/s]
2|221|Loss: 0.8045491576194763:   3%|▎         | 220/6470 [02:29<1:13:21,  1.42it/s]
2|221|Loss: 0.8045491576194763:   3%|▎         | 221/6470 [02:29<1:12:36,  1.43it/s]
2|222|Loss: 0.8326595425605774:   3%|▎         | 221/6470 [02:30<1:12:36,  1.43it/s]
2|222|Loss: 0.8326595425605774:   3%|▎         | 222/6470 [02:30<1:09:30,  1.50it/s]
2|223|Loss: 0.9880158305168152:   3%|▎         | 222/6470 [02:30<1:09:30,  1.50it/s]
2|223|Loss: 0.9880158305168152:   3%|▎         | 223/6470 [02:31<1:07:55,  1.53it/s]
2|224|Loss: 0.6926364898681641:   3%|▎         | 223/6470 [02:31<1:07:55,  1.53it/s]
2|224|Loss: 0.6926364898681641:   3%|▎         | 224/6470 [02:32<1:14:17,  1.40it/s]
2|225|Loss: 0.9380508661270142:   3%|▎         | 224/6470 [02:32<1:14:17,  1.40it/s]
2|225|Loss: 0.9380508661270142:   3%|▎         | 225/6470 [02:32<1:13:58,  1.41it/s]
2|226|Loss: 0.7684828042984009:   3%|▎         | 225/6470 [02:33<1:13:58,  1.41it/s]
2|226|Loss: 0.7684828042984009:   3%|▎         | 226/6470 [02:33<1:15:30,  1.38it/s]
2|227|Loss: 0.8657495379447937:   3%|▎         | 226/6470 [02:33<1:15:30,  1.38it/s]
2|227|Loss: 0.8657495379447937:   4%|▎         | 227/6470 [02:34<1:10:39,  1.47it/s]
2|228|Loss: 0.9586114287376404:   4%|▎         | 227/6470 [02:34<1:10:39,  1.47it/s]
2|228|Loss: 0.9586114287376404:   4%|▎         | 228/6470 [02:34<1:10:18,  1.48it/s]
2|229|Loss: 0.8260234594345093:   4%|▎         | 228/6470 [02:35<1:10:18,  1.48it/s]
2|229|Loss: 0.8260234594345093:   4%|▎         | 229/6470 [02:35<1:12:24,  1.44it/s]
2|230|Loss: 0.8954121470451355:   4%|▎         | 229/6470 [02:35<1:12:24,  1.44it/s]
2|230|Loss: 0.8954121470451355:   4%|▎         | 230/6470 [02:36<1:14:04,  1.40it/s]
2|231|Loss: 1.0726605653762817:   4%|▎         | 230/6470 [02:36<1:14:04,  1.40it/s]
2|231|Loss: 1.0726605653762817:   4%|▎         | 231/6470 [02:36<1:12:34,  1.43it/s]
2|232|Loss: 0.8408342599868774:   4%|▎         | 231/6470 [02:37<1:12:34,  1.43it/s]
2|232|Loss: 0.8408342599868774:   4%|▎         | 232/6470 [02:37<1:12:14,  1.44it/s]
2|233|Loss: 0.7789819240570068:   4%|▎         | 232/6470 [02:37<1:12:14,  1.44it/s]
2|233|Loss: 0.7789819240570068:   4%|▎         | 233/6470 [02:38<1:06:15,  1.57it/s]
2|234|Loss: 0.99306720495224:   4%|▎         | 233/6470 [02:38<1:06:15,  1.57it/s]  
2|234|Loss: 0.99306720495224:   4%|▎         | 234/6470 [02:38<1:07:20,  1.54it/s]
2|235|Loss: 0.9097546935081482:   4%|▎         | 234/6470 [02:39<1:07:20,  1.54it/s]
2|235|Loss: 0.9097546935081482:   4%|▎         | 235/6470 [02:39<1:07:57,  1.53it/s]
2|236|Loss: 0.795325517654419:   4%|▎         | 235/6470 [02:39<1:07:57,  1.53it/s] 
2|236|Loss: 0.795325517654419:   4%|▎         | 236/6470 [02:39<1:04:44,  1.60it/s]
2|237|Loss: 0.8126881122589111:   4%|▎         | 236/6470 [02:40<1:04:44,  1.60it/s]
2|237|Loss: 0.8126881122589111:   4%|▎         | 237/6470 [02:40<1:07:31,  1.54it/s]
2|238|Loss: 0.9184728264808655:   4%|▎         | 237/6470 [02:41<1:07:31,  1.54it/s]
2|238|Loss: 0.9184728264808655:   4%|▎         | 238/6470 [02:41<1:10:33,  1.47it/s]
2|239|Loss: 0.7208682894706726:   4%|▎         | 238/6470 [02:41<1:10:33,  1.47it/s]
2|239|Loss: 0.7208682894706726:   4%|▎         | 239/6470 [02:42<1:13:58,  1.40it/s]
2|240|Loss: 0.9316745400428772:   4%|▎         | 239/6470 [02:42<1:13:58,  1.40it/s]
2|240|Loss: 0.9316745400428772:   4%|▎         | 240/6470 [02:42<1:15:13,  1.38it/s]
2|241|Loss: 0.7679359912872314:   4%|▎         | 240/6470 [02:43<1:15:13,  1.38it/s]
2|241|Loss: 0.7679359912872314:   4%|▎         | 241/6470 [02:43<1:13:09,  1.42it/s]
2|242|Loss: 0.8681430816650391:   4%|▎         | 241/6470 [02:44<1:13:09,  1.42it/s]
2|242|Loss: 0.8681430816650391:   4%|▎         | 242/6470 [02:44<1:13:54,  1.40it/s]
2|243|Loss: 0.8528693914413452:   4%|▎         | 242/6470 [02:44<1:13:54,  1.40it/s]
2|243|Loss: 0.8528693914413452:   4%|▍         | 243/6470 [02:45<1:12:35,  1.43it/s]
2|244|Loss: 0.7804321050643921:   4%|▍         | 243/6470 [02:45<1:12:35,  1.43it/s]
2|244|Loss: 0.7804321050643921:   4%|▍         | 244/6470 [02:45<1:08:36,  1.51it/s]
2|245|Loss: 0.9278008937835693:   4%|▍         | 244/6470 [02:45<1:08:36,  1.51it/s]
2|245|Loss: 0.9278008937835693:   4%|▍         | 245/6470 [02:46<1:06:56,  1.55it/s]
2|246|Loss: 0.8758989572525024:   4%|▍         | 245/6470 [02:46<1:06:56,  1.55it/s]
2|246|Loss: 0.8758989572525024:   4%|▍         | 246/6470 [02:46<1:08:17,  1.52it/s]
2|247|Loss: 0.80036860704422:   4%|▍         | 246/6470 [02:47<1:08:17,  1.52it/s]  
2|247|Loss: 0.80036860704422:   4%|▍         | 247/6470 [02:47<1:13:58,  1.40it/s]
2|248|Loss: 0.838881254196167:   4%|▍         | 247/6470 [02:48<1:13:58,  1.40it/s]
2|248|Loss: 0.838881254196167:   4%|▍         | 248/6470 [02:48<1:14:54,  1.38it/s]
2|249|Loss: 0.7860874533653259:   4%|▍         | 248/6470 [02:48<1:14:54,  1.38it/s]
2|249|Loss: 0.7860874533653259:   4%|▍         | 249/6470 [02:49<1:18:29,  1.32it/s]
2|250|Loss: 0.9640653133392334:   4%|▍         | 249/6470 [02:49<1:18:29,  1.32it/s]
2|250|Loss: 0.9640653133392334:   4%|▍         | 250/6470 [02:50<1:15:54,  1.37it/s]
2|251|Loss: 0.7800038456916809:   4%|▍         | 250/6470 [02:50<1:15:54,  1.37it/s]
2|251|Loss: 0.7800038456916809:   4%|▍         | 251/6470 [02:50<1:17:28,  1.34it/s]
2|252|Loss: 0.836872935295105:   4%|▍         | 251/6470 [02:51<1:17:28,  1.34it/s] 
2|252|Loss: 0.836872935295105:   4%|▍         | 252/6470 [02:51<1:19:54,  1.30it/s]
2|253|Loss: 1.06581711769104:   4%|▍         | 252/6470 [02:52<1:19:54,  1.30it/s] 
2|253|Loss: 1.06581711769104:   4%|▍         | 253/6470 [02:52<1:23:02,  1.25it/s]
2|254|Loss: 0.835389256477356:   4%|▍         | 253/6470 [02:52<1:23:02,  1.25it/s]
2|254|Loss: 0.835389256477356:   4%|▍         | 254/6470 [02:53<1:21:16,  1.27it/s]
2|255|Loss: 0.8706364035606384:   4%|▍         | 254/6470 [02:53<1:21:16,  1.27it/s]
2|255|Loss: 0.8706364035606384:   4%|▍         | 255/6470 [02:53<1:17:09,  1.34it/s]
2|256|Loss: 0.8495571613311768:   4%|▍         | 255/6470 [02:54<1:17:09,  1.34it/s]
2|256|Loss: 0.8495571613311768:   4%|▍         | 256/6470 [02:54<1:14:03,  1.40it/s]
2|257|Loss: 0.8869730234146118:   4%|▍         | 256/6470 [02:54<1:14:03,  1.40it/s]
2|257|Loss: 0.8869730234146118:   4%|▍         | 257/6470 [02:55<1:18:31,  1.32it/s]
2|258|Loss: 0.9007118940353394:   4%|▍         | 257/6470 [02:55<1:18:31,  1.32it/s]
2|258|Loss: 0.9007118940353394:   4%|▍         | 258/6470 [02:55<1:12:39,  1.42it/s]
2|259|Loss: 0.8730913400650024:   4%|▍         | 258/6470 [02:56<1:12:39,  1.42it/s]
2|259|Loss: 0.8730913400650024:   4%|▍         | 259/6470 [02:56<1:09:04,  1.50it/s]
2|260|Loss: 0.8833642601966858:   4%|▍         | 259/6470 [02:56<1:09:04,  1.50it/s]
2|260|Loss: 0.8833642601966858:   4%|▍         | 260/6470 [02:57<1:09:29,  1.49it/s]
2|261|Loss: 0.745460033416748:   4%|▍         | 260/6470 [02:57<1:09:29,  1.49it/s] 
2|261|Loss: 0.745460033416748:   4%|▍         | 261/6470 [02:57<1:11:39,  1.44it/s]
2|262|Loss: 0.7638862133026123:   4%|▍         | 261/6470 [02:58<1:11:39,  1.44it/s]
2|262|Loss: 0.7638862133026123:   4%|▍         | 262/6470 [02:58<1:09:13,  1.49it/s]
2|263|Loss: 0.8793019652366638:   4%|▍         | 262/6470 [02:59<1:09:13,  1.49it/s]
2|263|Loss: 0.8793019652366638:   4%|▍         | 263/6470 [02:59<1:15:10,  1.38it/s]
2|264|Loss: 0.9151159524917603:   4%|▍         | 263/6470 [02:59<1:15:10,  1.38it/s]
2|264|Loss: 0.9151159524917603:   4%|▍         | 264/6470 [03:00<1:12:55,  1.42it/s]
2|265|Loss: 0.9478726387023926:   4%|▍         | 264/6470 [03:00<1:12:55,  1.42it/s]
2|265|Loss: 0.9478726387023926:   4%|▍         | 265/6470 [03:00<1:14:43,  1.38it/s]
2|266|Loss: 0.9015535712242126:   4%|▍         | 265/6470 [03:01<1:14:43,  1.38it/s]
2|266|Loss: 0.9015535712242126:   4%|▍         | 266/6470 [03:01<1:17:05,  1.34it/s]
2|267|Loss: 0.9221928119659424:   4%|▍         | 266/6470 [03:02<1:17:05,  1.34it/s]
2|267|Loss: 0.9221928119659424:   4%|▍         | 267/6470 [03:02<1:17:40,  1.33it/s]
2|268|Loss: 1.0076345205307007:   4%|▍         | 267/6470 [03:02<1:17:40,  1.33it/s]
2|268|Loss: 1.0076345205307007:   4%|▍         | 268/6470 [03:03<1:15:01,  1.38it/s]
2|269|Loss: 0.8314781188964844:   4%|▍         | 268/6470 [03:03<1:15:01,  1.38it/s]
2|269|Loss: 0.8314781188964844:   4%|▍         | 269/6470 [03:03<1:12:17,  1.43it/s]
2|270|Loss: 0.9478474259376526:   4%|▍         | 269/6470 [03:04<1:12:17,  1.43it/s]
2|270|Loss: 0.9478474259376526:   4%|▍         | 270/6470 [03:04<1:10:37,  1.46it/s]
2|271|Loss: 0.8145402669906616:   4%|▍         | 270/6470 [03:04<1:10:37,  1.46it/s]
2|271|Loss: 0.8145402669906616:   4%|▍         | 271/6470 [03:05<1:13:09,  1.41it/s]
2|272|Loss: 0.7713137865066528:   4%|▍         | 271/6470 [03:05<1:13:09,  1.41it/s]
2|272|Loss: 0.7713137865066528:   4%|▍         | 272/6470 [03:06<1:18:07,  1.32it/s]
2|273|Loss: 0.877913773059845:   4%|▍         | 272/6470 [03:06<1:18:07,  1.32it/s] 
2|273|Loss: 0.877913773059845:   4%|▍         | 273/6470 [03:06<1:13:28,  1.41it/s]
2|274|Loss: 0.969075620174408:   4%|▍         | 273/6470 [03:06<1:13:28,  1.41it/s]
2|274|Loss: 0.969075620174408:   4%|▍         | 274/6470 [03:07<1:10:31,  1.46it/s]
2|275|Loss: 0.8600577116012573:   4%|▍         | 274/6470 [03:07<1:10:31,  1.46it/s]
2|275|Loss: 0.8600577116012573:   4%|▍         | 275/6470 [03:08<1:13:18,  1.41it/s]
2|276|Loss: 0.9404381513595581:   4%|▍         | 275/6470 [03:08<1:13:18,  1.41it/s]
2|276|Loss: 0.9404381513595581:   4%|▍         | 276/6470 [03:08<1:09:28,  1.49it/s]
2|277|Loss: 0.8912635445594788:   4%|▍         | 276/6470 [03:09<1:09:28,  1.49it/s]
2|277|Loss: 0.8912635445594788:   4%|▍         | 277/6470 [03:09<1:13:38,  1.40it/s]
2|278|Loss: 0.8384659290313721:   4%|▍         | 277/6470 [03:09<1:13:38,  1.40it/s]
2|278|Loss: 0.8384659290313721:   4%|▍         | 278/6470 [03:10<1:12:26,  1.42it/s]
2|279|Loss: 0.8418424725532532:   4%|▍         | 278/6470 [03:10<1:12:26,  1.42it/s]
2|279|Loss: 0.8418424725532532:   4%|▍         | 279/6470 [03:10<1:10:09,  1.47it/s]
2|280|Loss: 0.8843050003051758:   4%|▍         | 279/6470 [03:11<1:10:09,  1.47it/s]
2|280|Loss: 0.8843050003051758:   4%|▍         | 280/6470 [03:11<1:11:11,  1.45it/s]
2|281|Loss: 0.9430384039878845:   4%|▍         | 280/6470 [03:11<1:11:11,  1.45it/s]
2|281|Loss: 0.9430384039878845:   4%|▍         | 281/6470 [03:12<1:14:59,  1.38it/s]
2|282|Loss: 0.9291253685951233:   4%|▍         | 281/6470 [03:12<1:14:59,  1.38it/s]
2|282|Loss: 0.9291253685951233:   4%|▍         | 282/6470 [03:12<1:15:41,  1.36it/s]
2|283|Loss: 0.8581163883209229:   4%|▍         | 282/6470 [03:13<1:15:41,  1.36it/s]
2|283|Loss: 0.8581163883209229:   4%|▍         | 283/6470 [03:13<1:10:04,  1.47it/s]
2|284|Loss: 0.8433238863945007:   4%|▍         | 283/6470 [03:13<1:10:04,  1.47it/s]
2|284|Loss: 0.8433238863945007:   4%|▍         | 284/6470 [03:14<1:07:56,  1.52it/s]
2|285|Loss: 0.9124937653541565:   4%|▍         | 284/6470 [03:14<1:07:56,  1.52it/s]
2|285|Loss: 0.9124937653541565:   4%|▍         | 285/6470 [03:14<1:09:56,  1.47it/s]
2|286|Loss: 0.8544236421585083:   4%|▍         | 285/6470 [03:15<1:09:56,  1.47it/s]
2|286|Loss: 0.8544236421585083:   4%|▍         | 286/6470 [03:15<1:07:50,  1.52it/s]
2|287|Loss: 0.9129098653793335:   4%|▍         | 286/6470 [03:15<1:07:50,  1.52it/s]
2|287|Loss: 0.9129098653793335:   4%|▍         | 287/6470 [03:16<1:08:32,  1.50it/s]
2|288|Loss: 0.9513333439826965:   4%|▍         | 287/6470 [03:16<1:08:32,  1.50it/s]
2|288|Loss: 0.9513333439826965:   4%|▍         | 288/6470 [03:16<1:09:48,  1.48it/s]
2|289|Loss: 0.9213574528694153:   4%|▍         | 288/6470 [03:17<1:09:48,  1.48it/s]
2|289|Loss: 0.9213574528694153:   4%|▍         | 289/6470 [03:17<1:08:31,  1.50it/s]
2|290|Loss: 0.8515178561210632:   4%|▍         | 289/6470 [03:17<1:08:31,  1.50it/s]
2|290|Loss: 0.8515178561210632:   4%|▍         | 290/6470 [03:18<1:05:59,  1.56it/s]
2|291|Loss: 0.76064133644104:   4%|▍         | 290/6470 [03:18<1:05:59,  1.56it/s]  
2|291|Loss: 0.76064133644104:   4%|▍         | 291/6470 [03:18<1:04:02,  1.61it/s]
2|292|Loss: 0.9056475162506104:   4%|▍         | 291/6470 [03:19<1:04:02,  1.61it/s]
2|292|Loss: 0.9056475162506104:   5%|▍         | 292/6470 [03:19<1:04:18,  1.60it/s]
2|293|Loss: 0.8958940505981445:   5%|▍         | 292/6470 [03:19<1:04:18,  1.60it/s]
2|293|Loss: 0.8958940505981445:   5%|▍         | 293/6470 [03:20<1:06:34,  1.55it/s]
2|294|Loss: 1.0274574756622314:   5%|▍         | 293/6470 [03:20<1:06:34,  1.55it/s]
2|294|Loss: 1.0274574756622314:   5%|▍         | 294/6470 [03:20<1:11:32,  1.44it/s]
2|295|Loss: 0.9642269611358643:   5%|▍         | 294/6470 [03:21<1:11:32,  1.44it/s]
2|295|Loss: 0.9642269611358643:   5%|▍         | 295/6470 [03:21<1:12:38,  1.42it/s]
2|296|Loss: 0.8227483630180359:   5%|▍         | 295/6470 [03:21<1:12:38,  1.42it/s]
2|296|Loss: 0.8227483630180359:   5%|▍         | 296/6470 [03:22<1:10:28,  1.46it/s]
2|297|Loss: 0.8912458419799805:   5%|▍         | 296/6470 [03:22<1:10:28,  1.46it/s]
2|297|Loss: 0.8912458419799805:   5%|▍         | 297/6470 [03:22<1:10:24,  1.46it/s]
2|298|Loss: 0.78032386302948:   5%|▍         | 297/6470 [03:23<1:10:24,  1.46it/s]  
2|298|Loss: 0.78032386302948:   5%|▍         | 298/6470 [03:23<1:08:47,  1.50it/s]
2|299|Loss: 0.8003853559494019:   5%|▍         | 298/6470 [03:23<1:08:47,  1.50it/s]
2|299|Loss: 0.8003853559494019:   5%|▍         | 299/6470 [03:24<1:07:34,  1.52it/s]
2|300|Loss: 0.9871832728385925:   5%|▍         | 299/6470 [03:24<1:07:34,  1.52it/s]
2|300|Loss: 0.9871832728385925:   5%|▍         | 300/6470 [03:24<1:09:43,  1.47it/s]
2|301|Loss: 0.9319306015968323:   5%|▍         | 300/6470 [03:25<1:09:43,  1.47it/s]
2|301|Loss: 0.9319306015968323:   5%|▍         | 301/6470 [03:25<1:11:33,  1.44it/s]
2|302|Loss: 0.7494020462036133:   5%|▍         | 301/6470 [03:26<1:11:33,  1.44it/s]
2|302|Loss: 0.7494020462036133:   5%|▍         | 302/6470 [03:26<1:14:58,  1.37it/s]
2|303|Loss: 0.8453963994979858:   5%|▍         | 302/6470 [03:26<1:14:58,  1.37it/s]
2|303|Loss: 0.8453963994979858:   5%|▍         | 303/6470 [03:27<1:12:53,  1.41it/s]
2|304|Loss: 0.9886987805366516:   5%|▍         | 303/6470 [03:27<1:12:53,  1.41it/s]
2|304|Loss: 0.9886987805366516:   5%|▍         | 304/6470 [03:27<1:12:16,  1.42it/s]
2|305|Loss: 1.0223801136016846:   5%|▍         | 304/6470 [03:28<1:12:16,  1.42it/s]
2|305|Loss: 1.0223801136016846:   5%|▍         | 305/6470 [03:28<1:15:31,  1.36it/s]
2|306|Loss: 0.8430981636047363:   5%|▍         | 305/6470 [03:28<1:15:31,  1.36it/s]
2|306|Loss: 0.8430981636047363:   5%|▍         | 306/6470 [03:29<1:13:51,  1.39it/s]
2|307|Loss: 0.878296971321106:   5%|▍         | 306/6470 [03:29<1:13:51,  1.39it/s] 
2|307|Loss: 0.878296971321106:   5%|▍         | 307/6470 [03:29<1:10:16,  1.46it/s]
2|308|Loss: 0.9192095398902893:   5%|▍         | 307/6470 [03:30<1:10:16,  1.46it/s]
2|308|Loss: 0.9192095398902893:   5%|▍         | 308/6470 [03:30<1:01:58,  1.66it/s]
2|309|Loss: 0.7857503294944763:   5%|▍         | 308/6470 [03:30<1:01:58,  1.66it/s]
2|309|Loss: 0.7857503294944763:   5%|▍         | 309/6470 [03:31<1:06:37,  1.54it/s]
2|310|Loss: 0.9342963099479675:   5%|▍         | 309/6470 [03:31<1:06:37,  1.54it/s]
2|310|Loss: 0.9342963099479675:   5%|▍         | 310/6470 [03:31<1:10:35,  1.45it/s]
2|311|Loss: 0.9059464335441589:   5%|▍         | 310/6470 [03:32<1:10:35,  1.45it/s]
2|311|Loss: 0.9059464335441589:   5%|▍         | 311/6470 [03:32<1:10:06,  1.46it/s]
2|312|Loss: 0.7834535241127014:   5%|▍         | 311/6470 [03:32<1:10:06,  1.46it/s]
2|312|Loss: 0.7834535241127014:   5%|▍         | 312/6470 [03:33<1:13:33,  1.40it/s]
2|313|Loss: 0.9016486406326294:   5%|▍         | 312/6470 [03:33<1:13:33,  1.40it/s]
2|313|Loss: 0.9016486406326294:   5%|▍         | 313/6470 [03:34<1:14:42,  1.37it/s]
2|314|Loss: 0.8137510418891907:   5%|▍         | 313/6470 [03:34<1:14:42,  1.37it/s]
2|314|Loss: 0.8137510418891907:   5%|▍         | 314/6470 [03:34<1:17:55,  1.32it/s]
2|315|Loss: 0.8415215015411377:   5%|▍         | 314/6470 [03:35<1:17:55,  1.32it/s]
2|315|Loss: 0.8415215015411377:   5%|▍         | 315/6470 [03:35<1:21:48,  1.25it/s]
2|316|Loss: 0.8232675194740295:   5%|▍         | 315/6470 [03:36<1:21:48,  1.25it/s]
2|316|Loss: 0.8232675194740295:   5%|▍         | 316/6470 [03:36<1:21:02,  1.27it/s]
2|317|Loss: 0.8307390213012695:   5%|▍         | 316/6470 [03:36<1:21:02,  1.27it/s]
2|317|Loss: 0.8307390213012695:   5%|▍         | 317/6470 [03:37<1:15:29,  1.36it/s]
2|318|Loss: 0.7935746312141418:   5%|▍         | 317/6470 [03:37<1:15:29,  1.36it/s]
2|318|Loss: 0.7935746312141418:   5%|▍         | 318/6470 [03:37<1:14:34,  1.37it/s]
2|319|Loss: 0.8716599345207214:   5%|▍         | 318/6470 [03:38<1:14:34,  1.37it/s]
2|319|Loss: 0.8716599345207214:   5%|▍         | 319/6470 [03:38<1:14:46,  1.37it/s]
2|320|Loss: 0.9626199007034302:   5%|▍         | 319/6470 [03:38<1:14:46,  1.37it/s]
2|320|Loss: 0.9626199007034302:   5%|▍         | 320/6470 [03:39<1:09:50,  1.47it/s]
2|321|Loss: 0.7928094267845154:   5%|▍         | 320/6470 [03:39<1:09:50,  1.47it/s]
2|321|Loss: 0.7928094267845154:   5%|▍         | 321/6470 [03:39<1:11:25,  1.43it/s]
2|322|Loss: 0.6781179308891296:   5%|▍         | 321/6470 [03:40<1:11:25,  1.43it/s]
2|322|Loss: 0.6781179308891296:   5%|▍         | 322/6470 [03:40<1:14:19,  1.38it/s]
2|323|Loss: 0.9035965204238892:   5%|▍         | 322/6470 [03:41<1:14:19,  1.38it/s]
2|323|Loss: 0.9035965204238892:   5%|▍         | 323/6470 [03:41<1:14:26,  1.38it/s]
2|324|Loss: 0.8346385359764099:   5%|▍         | 323/6470 [03:41<1:14:26,  1.38it/s]
2|324|Loss: 0.8346385359764099:   5%|▌         | 324/6470 [03:42<1:16:24,  1.34it/s]
2|325|Loss: 0.7143082618713379:   5%|▌         | 324/6470 [03:42<1:16:24,  1.34it/s]
2|325|Loss: 0.7143082618713379:   5%|▌         | 325/6470 [03:42<1:16:50,  1.33it/s]
2|326|Loss: 0.7271785736083984:   5%|▌         | 325/6470 [03:43<1:16:50,  1.33it/s]
2|326|Loss: 0.7271785736083984:   5%|▌         | 326/6470 [03:43<1:16:28,  1.34it/s]
2|327|Loss: 0.8279622197151184:   5%|▌         | 326/6470 [03:44<1:16:28,  1.34it/s]
2|327|Loss: 0.8279622197151184:   5%|▌         | 327/6470 [03:44<1:15:19,  1.36it/s]
2|328|Loss: 0.8155394196510315:   5%|▌         | 327/6470 [03:44<1:15:19,  1.36it/s]
2|328|Loss: 0.8155394196510315:   5%|▌         | 328/6470 [03:45<1:12:09,  1.42it/s]
2|329|Loss: 0.8204455375671387:   5%|▌         | 328/6470 [03:45<1:12:09,  1.42it/s]
2|329|Loss: 0.8204455375671387:   5%|▌         | 329/6470 [03:45<1:12:31,  1.41it/s]
2|330|Loss: 0.8434284925460815:   5%|▌         | 329/6470 [03:46<1:12:31,  1.41it/s]
2|330|Loss: 0.8434284925460815:   5%|▌         | 330/6470 [03:46<1:09:46,  1.47it/s]
2|331|Loss: 1.0563199520111084:   5%|▌         | 330/6470 [03:46<1:09:46,  1.47it/s]
2|331|Loss: 1.0563199520111084:   5%|▌         | 331/6470 [03:46<1:08:03,  1.50it/s]
2|332|Loss: 0.8631840348243713:   5%|▌         | 331/6470 [03:47<1:08:03,  1.50it/s]
2|332|Loss: 0.8631840348243713:   5%|▌         | 332/6470 [03:47<1:09:43,  1.47it/s]
2|333|Loss: 0.9068731069564819:   5%|▌         | 332/6470 [03:48<1:09:43,  1.47it/s]
2|333|Loss: 0.9068731069564819:   5%|▌         | 333/6470 [03:48<1:11:24,  1.43it/s]
2|334|Loss: 0.9385631084442139:   5%|▌         | 333/6470 [03:48<1:11:24,  1.43it/s]
2|334|Loss: 0.9385631084442139:   5%|▌         | 334/6470 [03:49<1:07:57,  1.50it/s]
2|335|Loss: 0.9266433119773865:   5%|▌         | 334/6470 [03:49<1:07:57,  1.50it/s]
2|335|Loss: 0.9266433119773865:   5%|▌         | 335/6470 [03:49<1:08:24,  1.49it/s]
2|336|Loss: 0.771018922328949:   5%|▌         | 335/6470 [03:50<1:08:24,  1.49it/s] 
2|336|Loss: 0.771018922328949:   5%|▌         | 336/6470 [03:50<1:04:44,  1.58it/s]
2|337|Loss: 0.8282568454742432:   5%|▌         | 336/6470 [03:50<1:04:44,  1.58it/s]
2|337|Loss: 0.8282568454742432:   5%|▌         | 337/6470 [03:50<59:31,  1.72it/s]  
2|338|Loss: 0.9150883555412292:   5%|▌         | 337/6470 [03:51<59:31,  1.72it/s]
2|338|Loss: 0.9150883555412292:   5%|▌         | 338/6470 [03:51<1:01:34,  1.66it/s]
2|339|Loss: 0.8251320123672485:   5%|▌         | 338/6470 [03:51<1:01:34,  1.66it/s]
2|339|Loss: 0.8251320123672485:   5%|▌         | 339/6470 [03:52<1:04:56,  1.57it/s]
2|340|Loss: 0.8900768756866455:   5%|▌         | 339/6470 [03:52<1:04:56,  1.57it/s]
2|340|Loss: 0.8900768756866455:   5%|▌         | 340/6470 [03:52<58:45,  1.74it/s]  
2|341|Loss: 0.7628061771392822:   5%|▌         | 340/6470 [03:52<58:45,  1.74it/s]
2|341|Loss: 0.7628061771392822:   5%|▌         | 341/6470 [03:53<1:00:55,  1.68it/s]
2|342|Loss: 0.8275128602981567:   5%|▌         | 341/6470 [03:53<1:00:55,  1.68it/s]
2|342|Loss: 0.8275128602981567:   5%|▌         | 342/6470 [03:53<1:02:48,  1.63it/s]
2|343|Loss: 1.001282811164856:   5%|▌         | 342/6470 [03:54<1:02:48,  1.63it/s] 
2|343|Loss: 1.001282811164856:   5%|▌         | 343/6470 [03:54<1:08:20,  1.49it/s]
2|344|Loss: 0.8256542086601257:   5%|▌         | 343/6470 [03:55<1:08:20,  1.49it/s]
2|344|Loss: 0.8256542086601257:   5%|▌         | 344/6470 [03:55<1:12:38,  1.41it/s]
2|345|Loss: 0.7762190699577332:   5%|▌         | 344/6470 [03:55<1:12:38,  1.41it/s]
2|345|Loss: 0.7762190699577332:   5%|▌         | 345/6470 [03:56<1:11:36,  1.43it/s]
2|346|Loss: 0.7764277458190918:   5%|▌         | 345/6470 [03:56<1:11:36,  1.43it/s]
2|346|Loss: 0.7764277458190918:   5%|▌         | 346/6470 [03:56<1:11:26,  1.43it/s]
2|347|Loss: 0.9218212366104126:   5%|▌         | 346/6470 [03:57<1:11:26,  1.43it/s]
2|347|Loss: 0.9218212366104126:   5%|▌         | 347/6470 [03:57<1:09:20,  1.47it/s]
2|348|Loss: 0.7348275780677795:   5%|▌         | 347/6470 [03:57<1:09:20,  1.47it/s]
2|348|Loss: 0.7348275780677795:   5%|▌         | 348/6470 [03:58<1:07:20,  1.52it/s]
2|349|Loss: 1.045248031616211:   5%|▌         | 348/6470 [03:58<1:07:20,  1.52it/s] 
2|349|Loss: 1.045248031616211:   5%|▌         | 349/6470 [03:58<1:05:03,  1.57it/s]
2|350|Loss: 0.7727857828140259:   5%|▌         | 349/6470 [03:59<1:05:03,  1.57it/s]
2|350|Loss: 0.7727857828140259:   5%|▌         | 350/6470 [03:59<1:10:09,  1.45it/s]
2|351|Loss: 1.0657016038894653:   5%|▌         | 350/6470 [03:59<1:10:09,  1.45it/s]
2|351|Loss: 1.0657016038894653:   5%|▌         | 351/6470 [04:00<1:12:51,  1.40it/s]
2|352|Loss: 0.8314216732978821:   5%|▌         | 351/6470 [04:00<1:12:51,  1.40it/s]
2|352|Loss: 0.8314216732978821:   5%|▌         | 352/6470 [04:00<1:13:01,  1.40it/s]
2|353|Loss: 0.7765441536903381:   5%|▌         | 352/6470 [04:01<1:13:01,  1.40it/s]
2|353|Loss: 0.7765441536903381:   5%|▌         | 353/6470 [04:01<1:16:28,  1.33it/s]
2|354|Loss: 0.936051607131958:   5%|▌         | 353/6470 [04:02<1:16:28,  1.33it/s] 
2|354|Loss: 0.936051607131958:   5%|▌         | 354/6470 [04:02<1:16:09,  1.34it/s]
2|355|Loss: 0.9082207679748535:   5%|▌         | 354/6470 [04:02<1:16:09,  1.34it/s]
2|355|Loss: 0.9082207679748535:   5%|▌         | 355/6470 [04:03<1:14:54,  1.36it/s]
2|356|Loss: 0.7316768765449524:   5%|▌         | 355/6470 [04:03<1:14:54,  1.36it/s]
2|356|Loss: 0.7316768765449524:   6%|▌         | 356/6470 [04:03<1:09:31,  1.47it/s]
2|357|Loss: 0.8251968026161194:   6%|▌         | 356/6470 [04:04<1:09:31,  1.47it/s]
2|357|Loss: 0.8251968026161194:   6%|▌         | 357/6470 [04:04<1:10:23,  1.45it/s]
2|358|Loss: 0.8740764260292053:   6%|▌         | 357/6470 [04:04<1:10:23,  1.45it/s]
2|358|Loss: 0.8740764260292053:   6%|▌         | 358/6470 [04:05<1:14:48,  1.36it/s]
2|359|Loss: 0.8408137559890747:   6%|▌         | 358/6470 [04:05<1:14:48,  1.36it/s]
2|359|Loss: 0.8408137559890747:   6%|▌         | 359/6470 [04:06<1:17:49,  1.31it/s]
2|360|Loss: 0.8497029542922974:   6%|▌         | 359/6470 [04:06<1:17:49,  1.31it/s]
2|360|Loss: 0.8497029542922974:   6%|▌         | 360/6470 [04:06<1:17:38,  1.31it/s]
2|361|Loss: 0.9394403696060181:   6%|▌         | 360/6470 [04:07<1:17:38,  1.31it/s]
2|361|Loss: 0.9394403696060181:   6%|▌         | 361/6470 [04:07<1:19:09,  1.29it/s]
2|362|Loss: 0.8790543675422668:   6%|▌         | 361/6470 [04:08<1:19:09,  1.29it/s]
2|362|Loss: 0.8790543675422668:   6%|▌         | 362/6470 [04:08<1:15:45,  1.34it/s]
2|363|Loss: 0.88917076587677:   6%|▌         | 362/6470 [04:08<1:15:45,  1.34it/s]  
2|363|Loss: 0.88917076587677:   6%|▌         | 363/6470 [04:09<1:20:06,  1.27it/s]
2|364|Loss: 0.8048233389854431:   6%|▌         | 363/6470 [04:09<1:20:06,  1.27it/s]
2|364|Loss: 0.8048233389854431:   6%|▌         | 364/6470 [04:10<1:21:19,  1.25it/s]
2|365|Loss: 0.8160268664360046:   6%|▌         | 364/6470 [04:10<1:21:19,  1.25it/s]
2|365|Loss: 0.8160268664360046:   6%|▌         | 365/6470 [04:10<1:19:17,  1.28it/s]
2|366|Loss: 0.8113543391227722:   6%|▌         | 365/6470 [04:11<1:19:17,  1.28it/s]
2|366|Loss: 0.8113543391227722:   6%|▌         | 366/6470 [04:11<1:15:30,  1.35it/s]
2|367|Loss: 0.805283784866333:   6%|▌         | 366/6470 [04:11<1:15:30,  1.35it/s] 
2|367|Loss: 0.805283784866333:   6%|▌         | 367/6470 [04:12<1:14:45,  1.36it/s]
2|368|Loss: 0.8745236992835999:   6%|▌         | 367/6470 [04:12<1:14:45,  1.36it/s]
2|368|Loss: 0.8745236992835999:   6%|▌         | 368/6470 [04:12<1:14:20,  1.37it/s]
2|369|Loss: 0.7720783352851868:   6%|▌         | 368/6470 [04:13<1:14:20,  1.37it/s]
2|369|Loss: 0.7720783352851868:   6%|▌         | 369/6470 [04:13<1:14:42,  1.36it/s]
2|370|Loss: 1.0036906003952026:   6%|▌         | 369/6470 [04:14<1:14:42,  1.36it/s]
2|370|Loss: 1.0036906003952026:   6%|▌         | 370/6470 [04:14<1:12:25,  1.40it/s]
2|371|Loss: 0.7122941613197327:   6%|▌         | 370/6470 [04:14<1:12:25,  1.40it/s]
2|371|Loss: 0.7122941613197327:   6%|▌         | 371/6470 [04:14<1:08:41,  1.48it/s]
2|372|Loss: 0.9578510522842407:   6%|▌         | 371/6470 [04:15<1:08:41,  1.48it/s]
2|372|Loss: 0.9578510522842407:   6%|▌         | 372/6470 [04:15<1:12:35,  1.40it/s]
2|373|Loss: 0.7996901869773865:   6%|▌         | 372/6470 [04:16<1:12:35,  1.40it/s]
2|373|Loss: 0.7996901869773865:   6%|▌         | 373/6470 [04:16<1:09:07,  1.47it/s]
2|374|Loss: 1.0385173559188843:   6%|▌         | 373/6470 [04:16<1:09:07,  1.47it/s]
2|374|Loss: 1.0385173559188843:   6%|▌         | 374/6470 [04:16<1:05:27,  1.55it/s]
2|375|Loss: 0.8498026728630066:   6%|▌         | 374/6470 [04:17<1:05:27,  1.55it/s]
2|375|Loss: 0.8498026728630066:   6%|▌         | 375/6470 [04:17<1:01:41,  1.65it/s]
2|376|Loss: 0.8493112921714783:   6%|▌         | 375/6470 [04:17<1:01:41,  1.65it/s]
2|376|Loss: 0.8493112921714783:   6%|▌         | 376/6470 [04:18<1:04:04,  1.59it/s]
2|377|Loss: 0.8414968848228455:   6%|▌         | 376/6470 [04:18<1:04:04,  1.59it/s]
2|377|Loss: 0.8414968848228455:   6%|▌         | 377/6470 [04:18<1:10:46,  1.43it/s]
2|378|Loss: 0.8416707515716553:   6%|▌         | 377/6470 [04:19<1:10:46,  1.43it/s]
2|378|Loss: 0.8416707515716553:   6%|▌         | 378/6470 [04:19<1:09:07,  1.47it/s]
2|379|Loss: 0.7232309579849243:   6%|▌         | 378/6470 [04:19<1:09:07,  1.47it/s]
2|379|Loss: 0.7232309579849243:   6%|▌         | 379/6470 [04:20<1:10:12,  1.45it/s]
2|380|Loss: 0.8418565988540649:   6%|▌         | 379/6470 [04:20<1:10:12,  1.45it/s]
2|380|Loss: 0.8418565988540649:   6%|▌         | 380/6470 [04:21<1:11:49,  1.41it/s]
2|381|Loss: 0.9183832406997681:   6%|▌         | 380/6470 [04:21<1:11:49,  1.41it/s]
2|381|Loss: 0.9183832406997681:   6%|▌         | 381/6470 [04:21<1:12:12,  1.41it/s]
2|382|Loss: 0.7323178648948669:   6%|▌         | 381/6470 [04:22<1:12:12,  1.41it/s]
2|382|Loss: 0.7323178648948669:   6%|▌         | 382/6470 [04:22<1:09:39,  1.46it/s]
2|383|Loss: 0.8755947947502136:   6%|▌         | 382/6470 [04:22<1:09:39,  1.46it/s]
2|383|Loss: 0.8755947947502136:   6%|▌         | 383/6470 [04:23<1:09:28,  1.46it/s]
2|384|Loss: 0.8428672552108765:   6%|▌         | 383/6470 [04:23<1:09:28,  1.46it/s]
2|384|Loss: 0.8428672552108765:   6%|▌         | 384/6470 [04:23<1:10:36,  1.44it/s]
2|385|Loss: 0.8492439389228821:   6%|▌         | 384/6470 [04:24<1:10:36,  1.44it/s]
2|385|Loss: 0.8492439389228821:   6%|▌         | 385/6470 [04:24<1:14:36,  1.36it/s]
2|386|Loss: 0.7727010250091553:   6%|▌         | 385/6470 [04:25<1:14:36,  1.36it/s]
2|386|Loss: 0.7727010250091553:   6%|▌         | 386/6470 [04:25<1:17:43,  1.30it/s]
2|387|Loss: 0.9636133313179016:   6%|▌         | 386/6470 [04:25<1:17:43,  1.30it/s]
2|387|Loss: 0.9636133313179016:   6%|▌         | 387/6470 [04:26<1:22:59,  1.22it/s]
2|388|Loss: 0.7919507622718811:   6%|▌         | 387/6470 [04:26<1:22:59,  1.22it/s]
2|388|Loss: 0.7919507622718811:   6%|▌         | 388/6470 [04:27<1:19:42,  1.27it/s]
2|389|Loss: 0.8911330103874207:   6%|▌         | 388/6470 [04:27<1:19:42,  1.27it/s]
2|389|Loss: 0.8911330103874207:   6%|▌         | 389/6470 [04:27<1:14:48,  1.35it/s]
2|390|Loss: 0.7663558125495911:   6%|▌         | 389/6470 [04:28<1:14:48,  1.35it/s]
2|390|Loss: 0.7663558125495911:   6%|▌         | 390/6470 [04:28<1:13:58,  1.37it/s]
2|391|Loss: 0.8284533023834229:   6%|▌         | 390/6470 [04:28<1:13:58,  1.37it/s]
2|391|Loss: 0.8284533023834229:   6%|▌         | 391/6470 [04:29<1:11:51,  1.41it/s]
2|392|Loss: 0.7734812498092651:   6%|▌         | 391/6470 [04:29<1:11:51,  1.41it/s]
2|392|Loss: 0.7734812498092651:   6%|▌         | 392/6470 [04:29<1:05:52,  1.54it/s]
2|393|Loss: 0.8449091911315918:   6%|▌         | 392/6470 [04:29<1:05:52,  1.54it/s]
2|393|Loss: 0.8449091911315918:   6%|▌         | 393/6470 [04:30<1:01:21,  1.65it/s]
2|394|Loss: 0.9243055582046509:   6%|▌         | 393/6470 [04:30<1:01:21,  1.65it/s]
2|394|Loss: 0.9243055582046509:   6%|▌         | 394/6470 [04:30<1:03:45,  1.59it/s]
2|395|Loss: 0.9003549814224243:   6%|▌         | 394/6470 [04:31<1:03:45,  1.59it/s]
2|395|Loss: 0.9003549814224243:   6%|▌         | 395/6470 [04:31<1:03:03,  1.61it/s]
2|396|Loss: 0.9803076386451721:   6%|▌         | 395/6470 [04:31<1:03:03,  1.61it/s]
2|396|Loss: 0.9803076386451721:   6%|▌         | 396/6470 [04:31<58:45,  1.72it/s]  
2|397|Loss: 0.8613538146018982:   6%|▌         | 396/6470 [04:32<58:45,  1.72it/s]
2|397|Loss: 0.8613538146018982:   6%|▌         | 397/6470 [04:32<59:17,  1.71it/s]
2|398|Loss: 0.8577275276184082:   6%|▌         | 397/6470 [04:32<59:17,  1.71it/s]
2|398|Loss: 0.8577275276184082:   6%|▌         | 398/6470 [04:33<1:00:21,  1.68it/s]
2|399|Loss: 0.906882107257843:   6%|▌         | 398/6470 [04:33<1:00:21,  1.68it/s] 
2|399|Loss: 0.906882107257843:   6%|▌         | 399/6470 [04:33<1:03:36,  1.59it/s]
2|400|Loss: 0.853376567363739:   6%|▌         | 399/6470 [04:34<1:03:36,  1.59it/s]
2|400|Loss: 0.853376567363739:   6%|▌         | 400/6470 [04:34<1:08:06,  1.49it/s]
2|401|Loss: 0.8930503129959106:   6%|▌         | 400/6470 [04:35<1:08:06,  1.49it/s]
2|401|Loss: 0.8930503129959106:   6%|▌         | 401/6470 [04:35<1:12:57,  1.39it/s]
2|402|Loss: 0.6850959658622742:   6%|▌         | 401/6470 [04:35<1:12:57,  1.39it/s]
2|402|Loss: 0.6850959658622742:   6%|▌         | 402/6470 [04:36<1:11:54,  1.41it/s]
2|403|Loss: 0.8519706726074219:   6%|▌         | 402/6470 [04:36<1:11:54,  1.41it/s]
2|403|Loss: 0.8519706726074219:   6%|▌         | 403/6470 [04:36<1:11:15,  1.42it/s]
2|404|Loss: 0.9617392420768738:   6%|▌         | 403/6470 [04:37<1:11:15,  1.42it/s]
2|404|Loss: 0.9617392420768738:   6%|▌         | 404/6470 [04:37<1:07:29,  1.50it/s]
2|405|Loss: 0.8310704827308655:   6%|▌         | 404/6470 [04:37<1:07:29,  1.50it/s]
2|405|Loss: 0.8310704827308655:   6%|▋         | 405/6470 [04:37<1:04:48,  1.56it/s]
2|406|Loss: 0.6541661620140076:   6%|▋         | 405/6470 [04:38<1:04:48,  1.56it/s]
2|406|Loss: 0.6541661620140076:   6%|▋         | 406/6470 [04:38<1:12:15,  1.40it/s]
2|407|Loss: 0.9057092070579529:   6%|▋         | 406/6470 [04:39<1:12:15,  1.40it/s]
2|407|Loss: 0.9057092070579529:   6%|▋         | 407/6470 [04:39<1:14:08,  1.36it/s]
2|408|Loss: 0.9656165242195129:   6%|▋         | 407/6470 [04:40<1:14:08,  1.36it/s]
2|408|Loss: 0.9656165242195129:   6%|▋         | 408/6470 [04:40<1:10:23,  1.44it/s]
2|409|Loss: 0.9511420726776123:   6%|▋         | 408/6470 [04:40<1:10:23,  1.44it/s]
2|409|Loss: 0.9511420726776123:   6%|▋         | 409/6470 [04:40<1:10:37,  1.43it/s]
2|410|Loss: 0.7849037051200867:   6%|▋         | 409/6470 [04:41<1:10:37,  1.43it/s]
2|410|Loss: 0.7849037051200867:   6%|▋         | 410/6470 [04:41<1:09:29,  1.45it/s]
2|411|Loss: 0.7106785178184509:   6%|▋         | 410/6470 [04:41<1:09:29,  1.45it/s]
2|411|Loss: 0.7106785178184509:   6%|▋         | 411/6470 [04:42<1:06:20,  1.52it/s]
2|412|Loss: 0.9501026272773743:   6%|▋         | 411/6470 [04:42<1:06:20,  1.52it/s]
2|412|Loss: 0.9501026272773743:   6%|▋         | 412/6470 [04:42<1:04:22,  1.57it/s]
2|413|Loss: 0.8399332761764526:   6%|▋         | 412/6470 [04:43<1:04:22,  1.57it/s]
2|413|Loss: 0.8399332761764526:   6%|▋         | 413/6470 [04:43<1:04:07,  1.57it/s]
2|414|Loss: 0.8916819095611572:   6%|▋         | 413/6470 [04:43<1:04:07,  1.57it/s]
2|414|Loss: 0.8916819095611572:   6%|▋         | 414/6470 [04:44<1:06:57,  1.51it/s]
2|415|Loss: 0.830846905708313:   6%|▋         | 414/6470 [04:44<1:06:57,  1.51it/s] 
2|415|Loss: 0.830846905708313:   6%|▋         | 415/6470 [04:44<1:07:09,  1.50it/s]
2|416|Loss: 0.7763550281524658:   6%|▋         | 415/6470 [04:45<1:07:09,  1.50it/s]
2|416|Loss: 0.7763550281524658:   6%|▋         | 416/6470 [04:45<1:04:49,  1.56it/s]
2|417|Loss: 0.9095802903175354:   6%|▋         | 416/6470 [04:45<1:04:49,  1.56it/s]
2|417|Loss: 0.9095802903175354:   6%|▋         | 417/6470 [04:46<1:07:15,  1.50it/s]
2|418|Loss: 0.7863326072692871:   6%|▋         | 417/6470 [04:46<1:07:15,  1.50it/s]
2|418|Loss: 0.7863326072692871:   6%|▋         | 418/6470 [04:46<1:02:43,  1.61it/s]
2|419|Loss: 0.9663293361663818:   6%|▋         | 418/6470 [04:46<1:02:43,  1.61it/s]
2|419|Loss: 0.9663293361663818:   6%|▋         | 419/6470 [04:47<1:01:30,  1.64it/s]
2|420|Loss: 0.9207817912101746:   6%|▋         | 419/6470 [04:47<1:01:30,  1.64it/s]
2|420|Loss: 0.9207817912101746:   6%|▋         | 420/6470 [04:47<1:02:22,  1.62it/s]
2|421|Loss: 0.8637452721595764:   6%|▋         | 420/6470 [04:48<1:02:22,  1.62it/s]
2|421|Loss: 0.8637452721595764:   7%|▋         | 421/6470 [04:48<1:04:32,  1.56it/s]
2|422|Loss: 0.9867960810661316:   7%|▋         | 421/6470 [04:48<1:04:32,  1.56it/s]
2|422|Loss: 0.9867960810661316:   7%|▋         | 422/6470 [04:49<1:02:12,  1.62it/s]
2|423|Loss: 0.8300842642784119:   7%|▋         | 422/6470 [04:49<1:02:12,  1.62it/s]
2|423|Loss: 0.8300842642784119:   7%|▋         | 423/6470 [04:49<1:04:48,  1.56it/s]
2|424|Loss: 0.8684737086296082:   7%|▋         | 423/6470 [04:50<1:04:48,  1.56it/s]
2|424|Loss: 0.8684737086296082:   7%|▋         | 424/6470 [04:50<1:09:29,  1.45it/s]
2|425|Loss: 0.8945059776306152:   7%|▋         | 424/6470 [04:51<1:09:29,  1.45it/s]
2|425|Loss: 0.8945059776306152:   7%|▋         | 425/6470 [04:51<1:09:11,  1.46it/s]
2|426|Loss: 0.8341575860977173:   7%|▋         | 425/6470 [04:51<1:09:11,  1.46it/s]
2|426|Loss: 0.8341575860977173:   7%|▋         | 426/6470 [04:51<1:09:04,  1.46it/s]
2|427|Loss: 0.9368084073066711:   7%|▋         | 426/6470 [04:52<1:09:04,  1.46it/s]
2|427|Loss: 0.9368084073066711:   7%|▋         | 427/6470 [04:52<1:11:35,  1.41it/s]
2|428|Loss: 0.8105531930923462:   7%|▋         | 427/6470 [04:53<1:11:35,  1.41it/s]
2|428|Loss: 0.8105531930923462:   7%|▋         | 428/6470 [04:53<1:14:26,  1.35it/s]
2|429|Loss: 0.8281310796737671:   7%|▋         | 428/6470 [04:54<1:14:26,  1.35it/s]
2|429|Loss: 0.8281310796737671:   7%|▋         | 429/6470 [04:54<1:17:39,  1.30it/s]
2|430|Loss: 0.9747934937477112:   7%|▋         | 429/6470 [04:54<1:17:39,  1.30it/s]
2|430|Loss: 0.9747934937477112:   7%|▋         | 430/6470 [04:55<1:17:27,  1.30it/s]
2|431|Loss: 1.032073974609375:   7%|▋         | 430/6470 [04:55<1:17:27,  1.30it/s] 
2|431|Loss: 1.032073974609375:   7%|▋         | 431/6470 [04:55<1:18:50,  1.28it/s]
2|432|Loss: 0.8987288475036621:   7%|▋         | 431/6470 [04:56<1:18:50,  1.28it/s]
2|432|Loss: 0.8987288475036621:   7%|▋         | 432/6470 [04:56<1:19:17,  1.27it/s]
2|433|Loss: 0.9862043261528015:   7%|▋         | 432/6470 [04:57<1:19:17,  1.27it/s]
2|433|Loss: 0.9862043261528015:   7%|▋         | 433/6470 [04:57<1:18:12,  1.29it/s]
2|434|Loss: 0.7699400782585144:   7%|▋         | 433/6470 [04:57<1:18:12,  1.29it/s]
2|434|Loss: 0.7699400782585144:   7%|▋         | 434/6470 [04:58<1:13:14,  1.37it/s]
2|435|Loss: 0.9348210096359253:   7%|▋         | 434/6470 [04:58<1:13:14,  1.37it/s]
2|435|Loss: 0.9348210096359253:   7%|▋         | 435/6470 [04:58<1:10:16,  1.43it/s]
2|436|Loss: 0.7986642122268677:   7%|▋         | 435/6470 [04:59<1:10:16,  1.43it/s]
2|436|Loss: 0.7986642122268677:   7%|▋         | 436/6470 [04:59<1:08:05,  1.48it/s]
2|437|Loss: 0.8867869973182678:   7%|▋         | 436/6470 [04:59<1:08:05,  1.48it/s]
2|437|Loss: 0.8867869973182678:   7%|▋         | 437/6470 [05:00<1:08:18,  1.47it/s]
2|438|Loss: 0.9295387864112854:   7%|▋         | 437/6470 [05:00<1:08:18,  1.47it/s]
2|438|Loss: 0.9295387864112854:   7%|▋         | 438/6470 [05:00<1:07:31,  1.49it/s]
2|439|Loss: 0.89900803565979:   7%|▋         | 438/6470 [05:01<1:07:31,  1.49it/s]  
2|439|Loss: 0.89900803565979:   7%|▋         | 439/6470 [05:01<1:08:13,  1.47it/s]
2|440|Loss: 1.0088324546813965:   7%|▋         | 439/6470 [05:01<1:08:13,  1.47it/s]
2|440|Loss: 1.0088324546813965:   7%|▋         | 440/6470 [05:02<1:09:46,  1.44it/s]
2|441|Loss: 0.7826581001281738:   7%|▋         | 440/6470 [05:02<1:09:46,  1.44it/s]
2|441|Loss: 0.7826581001281738:   7%|▋         | 441/6470 [05:02<1:08:53,  1.46it/s]
2|442|Loss: 0.8763232827186584:   7%|▋         | 441/6470 [05:03<1:08:53,  1.46it/s]
2|442|Loss: 0.8763232827186584:   7%|▋         | 442/6470 [05:03<1:06:23,  1.51it/s]
2|443|Loss: 0.8689060211181641:   7%|▋         | 442/6470 [05:03<1:06:23,  1.51it/s]
2|443|Loss: 0.8689060211181641:   7%|▋         | 443/6470 [05:04<1:05:30,  1.53it/s]
2|444|Loss: 0.9442675113677979:   7%|▋         | 443/6470 [05:04<1:05:30,  1.53it/s]
2|444|Loss: 0.9442675113677979:   7%|▋         | 444/6470 [05:04<1:09:14,  1.45it/s]
2|445|Loss: 0.8487616777420044:   7%|▋         | 444/6470 [05:05<1:09:14,  1.45it/s]
2|445|Loss: 0.8487616777420044:   7%|▋         | 445/6470 [05:05<1:12:22,  1.39it/s]
2|446|Loss: 0.8044253587722778:   7%|▋         | 445/6470 [05:06<1:12:22,  1.39it/s]
2|446|Loss: 0.8044253587722778:   7%|▋         | 446/6470 [05:06<1:13:21,  1.37it/s]
2|447|Loss: 0.7410529851913452:   7%|▋         | 446/6470 [05:06<1:13:21,  1.37it/s]
2|447|Loss: 0.7410529851913452:   7%|▋         | 447/6470 [05:07<1:13:34,  1.36it/s]
2|448|Loss: 0.7818472981452942:   7%|▋         | 447/6470 [05:07<1:13:34,  1.36it/s]
2|448|Loss: 0.7818472981452942:   7%|▋         | 448/6470 [05:07<1:16:44,  1.31it/s]
2|449|Loss: 0.8447475433349609:   7%|▋         | 448/6470 [05:08<1:16:44,  1.31it/s]
2|449|Loss: 0.8447475433349609:   7%|▋         | 449/6470 [05:08<1:16:49,  1.31it/s]
2|450|Loss: 0.7712019085884094:   7%|▋         | 449/6470 [05:09<1:16:49,  1.31it/s]
2|450|Loss: 0.7712019085884094:   7%|▋         | 450/6470 [05:09<1:12:03,  1.39it/s]
2|451|Loss: 0.938656747341156:   7%|▋         | 450/6470 [05:09<1:12:03,  1.39it/s] 
2|451|Loss: 0.938656747341156:   7%|▋         | 451/6470 [05:10<1:11:40,  1.40it/s]
2|452|Loss: 0.9435773491859436:   7%|▋         | 451/6470 [05:10<1:11:40,  1.40it/s]
2|452|Loss: 0.9435773491859436:   7%|▋         | 452/6470 [05:10<1:07:32,  1.48it/s]
2|453|Loss: 0.8141692280769348:   7%|▋         | 452/6470 [05:11<1:07:32,  1.48it/s]
2|453|Loss: 0.8141692280769348:   7%|▋         | 453/6470 [05:11<1:08:02,  1.47it/s]
2|454|Loss: 0.9413172602653503:   7%|▋         | 453/6470 [05:11<1:08:02,  1.47it/s]
2|454|Loss: 0.9413172602653503:   7%|▋         | 454/6470 [05:12<1:11:57,  1.39it/s]
2|455|Loss: 0.8640973567962646:   7%|▋         | 454/6470 [05:12<1:11:57,  1.39it/s]
2|455|Loss: 0.8640973567962646:   7%|▋         | 455/6470 [05:12<1:15:20,  1.33it/s]
2|456|Loss: 0.8956759572029114:   7%|▋         | 455/6470 [05:13<1:15:20,  1.33it/s]
2|456|Loss: 0.8956759572029114:   7%|▋         | 456/6470 [05:13<1:11:39,  1.40it/s]
2|457|Loss: 0.9060406684875488:   7%|▋         | 456/6470 [05:13<1:11:39,  1.40it/s]
2|457|Loss: 0.9060406684875488:   7%|▋         | 457/6470 [05:14<1:06:14,  1.51it/s]
2|458|Loss: 1.0081013441085815:   7%|▋         | 457/6470 [05:14<1:06:14,  1.51it/s]
2|458|Loss: 1.0081013441085815:   7%|▋         | 458/6470 [05:14<1:06:21,  1.51it/s]
2|459|Loss: 0.9106753468513489:   7%|▋         | 458/6470 [05:15<1:06:21,  1.51it/s]
2|459|Loss: 0.9106753468513489:   7%|▋         | 459/6470 [05:15<1:01:45,  1.62it/s]
2|460|Loss: 0.8816792368888855:   7%|▋         | 459/6470 [05:15<1:01:45,  1.62it/s]
2|460|Loss: 0.8816792368888855:   7%|▋         | 460/6470 [05:15<59:35,  1.68it/s]  
2|461|Loss: 0.8528094291687012:   7%|▋         | 460/6470 [05:16<59:35,  1.68it/s]
2|461|Loss: 0.8528094291687012:   7%|▋         | 461/6470 [05:16<1:04:36,  1.55it/s]
2|462|Loss: 0.7498655915260315:   7%|▋         | 461/6470 [05:17<1:04:36,  1.55it/s]
2|462|Loss: 0.7498655915260315:   7%|▋         | 462/6470 [05:17<1:05:27,  1.53it/s]
2|463|Loss: 0.7713561654090881:   7%|▋         | 462/6470 [05:17<1:05:27,  1.53it/s]
2|463|Loss: 0.7713561654090881:   7%|▋         | 463/6470 [05:17<1:07:06,  1.49it/s]
2|464|Loss: 0.9054463505744934:   7%|▋         | 463/6470 [05:18<1:07:06,  1.49it/s]
2|464|Loss: 0.9054463505744934:   7%|▋         | 464/6470 [05:18<1:06:34,  1.50it/s]
2|465|Loss: 0.8246254920959473:   7%|▋         | 464/6470 [05:19<1:06:34,  1.50it/s]
2|465|Loss: 0.8246254920959473:   7%|▋         | 465/6470 [05:19<1:07:50,  1.48it/s]
2|466|Loss: 0.7785468101501465:   7%|▋         | 465/6470 [05:19<1:07:50,  1.48it/s]
2|466|Loss: 0.7785468101501465:   7%|▋         | 466/6470 [05:19<1:06:40,  1.50it/s]
2|467|Loss: 0.8663789629936218:   7%|▋         | 466/6470 [05:20<1:06:40,  1.50it/s]
2|467|Loss: 0.8663789629936218:   7%|▋         | 467/6470 [05:20<1:08:17,  1.47it/s]
2|468|Loss: 0.7355343699455261:   7%|▋         | 467/6470 [05:21<1:08:17,  1.47it/s]
2|468|Loss: 0.7355343699455261:   7%|▋         | 468/6470 [05:21<1:10:24,  1.42it/s]
2|469|Loss: 1.0284584760665894:   7%|▋         | 468/6470 [05:21<1:10:24,  1.42it/s]
2|469|Loss: 1.0284584760665894:   7%|▋         | 469/6470 [05:22<1:09:17,  1.44it/s]
2|470|Loss: 0.9470440149307251:   7%|▋         | 469/6470 [05:22<1:09:17,  1.44it/s]
2|470|Loss: 0.9470440149307251:   7%|▋         | 470/6470 [05:22<1:12:13,  1.38it/s]
2|471|Loss: 0.7532544732093811:   7%|▋         | 470/6470 [05:23<1:12:13,  1.38it/s]
2|471|Loss: 0.7532544732093811:   7%|▋         | 471/6470 [05:23<1:10:51,  1.41it/s]
2|472|Loss: 0.8391024470329285:   7%|▋         | 471/6470 [05:23<1:10:51,  1.41it/s]
2|472|Loss: 0.8391024470329285:   7%|▋         | 472/6470 [05:24<1:08:17,  1.46it/s]
2|473|Loss: 0.9572260975837708:   7%|▋         | 472/6470 [05:24<1:08:17,  1.46it/s]
2|473|Loss: 0.9572260975837708:   7%|▋         | 473/6470 [05:24<1:04:07,  1.56it/s]
2|474|Loss: 0.9920802712440491:   7%|▋         | 473/6470 [05:25<1:04:07,  1.56it/s]
2|474|Loss: 0.9920802712440491:   7%|▋         | 474/6470 [05:25<1:07:13,  1.49it/s]
2|475|Loss: 0.915749728679657:   7%|▋         | 474/6470 [05:25<1:07:13,  1.49it/s] 
2|475|Loss: 0.915749728679657:   7%|▋         | 475/6470 [05:26<1:05:43,  1.52it/s]
2|476|Loss: 0.8874580264091492:   7%|▋         | 475/6470 [05:26<1:05:43,  1.52it/s]
2|476|Loss: 0.8874580264091492:   7%|▋         | 476/6470 [05:26<1:08:17,  1.46it/s]
2|477|Loss: 0.8752614259719849:   7%|▋         | 476/6470 [05:27<1:08:17,  1.46it/s]
2|477|Loss: 0.8752614259719849:   7%|▋         | 477/6470 [05:27<1:11:53,  1.39it/s]
2|478|Loss: 0.9800330400466919:   7%|▋         | 477/6470 [05:28<1:11:53,  1.39it/s]
2|478|Loss: 0.9800330400466919:   7%|▋         | 478/6470 [05:28<1:10:39,  1.41it/s]
2|479|Loss: 0.8494284749031067:   7%|▋         | 478/6470 [05:28<1:10:39,  1.41it/s]
2|479|Loss: 0.8494284749031067:   7%|▋         | 479/6470 [05:28<1:08:05,  1.47it/s]
2|480|Loss: 0.8963333368301392:   7%|▋         | 479/6470 [05:29<1:08:05,  1.47it/s]
2|480|Loss: 0.8963333368301392:   7%|▋         | 480/6470 [05:29<1:09:28,  1.44it/s]
2|481|Loss: 0.7466398477554321:   7%|▋         | 480/6470 [05:30<1:09:28,  1.44it/s]
2|481|Loss: 0.7466398477554321:   7%|▋         | 481/6470 [05:30<1:08:37,  1.45it/s]
2|482|Loss: 0.8860461115837097:   7%|▋         | 481/6470 [05:30<1:08:37,  1.45it/s]
2|482|Loss: 0.8860461115837097:   7%|▋         | 482/6470 [05:31<1:08:55,  1.45it/s]
2|483|Loss: 0.9801809787750244:   7%|▋         | 482/6470 [05:31<1:08:55,  1.45it/s]
2|483|Loss: 0.9801809787750244:   7%|▋         | 483/6470 [05:31<1:10:03,  1.42it/s]
2|484|Loss: 0.8723405599594116:   7%|▋         | 483/6470 [05:32<1:10:03,  1.42it/s]
2|484|Loss: 0.8723405599594116:   7%|▋         | 484/6470 [05:32<1:08:26,  1.46it/s]
2|485|Loss: 0.8788148164749146:   7%|▋         | 484/6470 [05:32<1:08:26,  1.46it/s]
2|485|Loss: 0.8788148164749146:   7%|▋         | 485/6470 [05:33<1:10:55,  1.41it/s]
2|486|Loss: 0.748382568359375:   7%|▋         | 485/6470 [05:33<1:10:55,  1.41it/s] 
2|486|Loss: 0.748382568359375:   8%|▊         | 486/6470 [05:33<1:09:48,  1.43it/s]
2|487|Loss: 0.8507521748542786:   8%|▊         | 486/6470 [05:34<1:09:48,  1.43it/s]
2|487|Loss: 0.8507521748542786:   8%|▊         | 487/6470 [05:34<1:09:49,  1.43it/s]
2|488|Loss: 0.9374161958694458:   8%|▊         | 487/6470 [05:35<1:09:49,  1.43it/s]
2|488|Loss: 0.9374161958694458:   8%|▊         | 488/6470 [05:35<1:10:51,  1.41it/s]
2|489|Loss: 0.8301988244056702:   8%|▊         | 488/6470 [05:35<1:10:51,  1.41it/s]
2|489|Loss: 0.8301988244056702:   8%|▊         | 489/6470 [05:35<1:09:03,  1.44it/s]
2|490|Loss: 1.112011432647705:   8%|▊         | 489/6470 [05:36<1:09:03,  1.44it/s] 
2|490|Loss: 1.112011432647705:   8%|▊         | 490/6470 [05:36<1:08:55,  1.45it/s]
2|491|Loss: 0.7380251884460449:   8%|▊         | 490/6470 [05:37<1:08:55,  1.45it/s]
2|491|Loss: 0.7380251884460449:   8%|▊         | 491/6470 [05:37<1:04:06,  1.55it/s]
2|492|Loss: 0.9707760810852051:   8%|▊         | 491/6470 [05:37<1:04:06,  1.55it/s]
2|492|Loss: 0.9707760810852051:   8%|▊         | 492/6470 [05:37<1:05:42,  1.52it/s]
2|493|Loss: 0.9260773062705994:   8%|▊         | 492/6470 [05:38<1:05:42,  1.52it/s]
2|493|Loss: 0.9260773062705994:   8%|▊         | 493/6470 [05:38<1:07:37,  1.47it/s]
2|494|Loss: 1.0520570278167725:   8%|▊         | 493/6470 [05:39<1:07:37,  1.47it/s]
2|494|Loss: 1.0520570278167725:   8%|▊         | 494/6470 [05:39<1:06:23,  1.50it/s]
2|495|Loss: 0.9263001084327698:   8%|▊         | 494/6470 [05:39<1:06:23,  1.50it/s]
2|495|Loss: 0.9263001084327698:   8%|▊         | 495/6470 [05:39<1:03:46,  1.56it/s]
2|496|Loss: 0.7517519593238831:   8%|▊         | 495/6470 [05:40<1:03:46,  1.56it/s]
2|496|Loss: 0.7517519593238831:   8%|▊         | 496/6470 [05:40<1:01:53,  1.61it/s]
2|497|Loss: 0.8768041133880615:   8%|▊         | 496/6470 [05:40<1:01:53,  1.61it/s]
2|497|Loss: 0.8768041133880615:   8%|▊         | 497/6470 [05:41<1:02:17,  1.60it/s]
2|498|Loss: 0.8853685855865479:   8%|▊         | 497/6470 [05:41<1:02:17,  1.60it/s]
2|498|Loss: 0.8853685855865479:   8%|▊         | 498/6470 [05:41<1:02:14,  1.60it/s]
2|499|Loss: 0.8171406388282776:   8%|▊         | 498/6470 [05:42<1:02:14,  1.60it/s]
2|499|Loss: 0.8171406388282776:   8%|▊         | 499/6470 [05:42<1:04:17,  1.55it/s]
2|500|Loss: 0.9858442544937134:   8%|▊         | 499/6470 [05:42<1:04:17,  1.55it/s]
2|500|Loss: 0.9858442544937134:   8%|▊         | 500/6470 [05:42<1:02:25,  1.59it/s]
2|501|Loss: 1.0540828704833984:   8%|▊         | 500/6470 [05:43<1:02:25,  1.59it/s]
2|501|Loss: 1.0540828704833984:   8%|▊         | 501/6470 [05:43<1:06:51,  1.49it/s]
2|502|Loss: 0.8894421458244324:   8%|▊         | 501/6470 [05:44<1:06:51,  1.49it/s]
2|502|Loss: 0.8894421458244324:   8%|▊         | 502/6470 [05:44<1:08:36,  1.45it/s]
2|503|Loss: 0.8848018050193787:   8%|▊         | 502/6470 [05:44<1:08:36,  1.45it/s]
2|503|Loss: 0.8848018050193787:   8%|▊         | 503/6470 [05:45<1:06:58,  1.48it/s]
2|504|Loss: 0.8816527128219604:   8%|▊         | 503/6470 [05:45<1:06:58,  1.48it/s]
2|504|Loss: 0.8816527128219604:   8%|▊         | 504/6470 [05:45<1:05:13,  1.52it/s]
2|505|Loss: 0.9092053174972534:   8%|▊         | 504/6470 [05:46<1:05:13,  1.52it/s]
2|505|Loss: 0.9092053174972534:   8%|▊         | 505/6470 [05:46<1:03:04,  1.58it/s]
2|506|Loss: 0.8657637238502502:   8%|▊         | 505/6470 [05:46<1:03:04,  1.58it/s]
2|506|Loss: 0.8657637238502502:   8%|▊         | 506/6470 [05:46<1:01:28,  1.62it/s]
2|507|Loss: 0.7606257796287537:   8%|▊         | 506/6470 [05:47<1:01:28,  1.62it/s]
2|507|Loss: 0.7606257796287537:   8%|▊         | 507/6470 [05:47<1:06:57,  1.48it/s]
2|508|Loss: 0.9442521333694458:   8%|▊         | 507/6470 [05:48<1:06:57,  1.48it/s]
2|508|Loss: 0.9442521333694458:   8%|▊         | 508/6470 [05:48<1:10:51,  1.40it/s]
2|509|Loss: 0.7938998341560364:   8%|▊         | 508/6470 [05:48<1:10:51,  1.40it/s]
2|509|Loss: 0.7938998341560364:   8%|▊         | 509/6470 [05:49<1:15:08,  1.32it/s]
2|510|Loss: 0.9177021384239197:   8%|▊         | 509/6470 [05:49<1:15:08,  1.32it/s]
2|510|Loss: 0.9177021384239197:   8%|▊         | 510/6470 [05:49<1:10:13,  1.41it/s]
2|511|Loss: 0.8134931921958923:   8%|▊         | 510/6470 [05:50<1:10:13,  1.41it/s]
2|511|Loss: 0.8134931921958923:   8%|▊         | 511/6470 [05:50<1:03:48,  1.56it/s]
2|512|Loss: 0.9767682552337646:   8%|▊         | 511/6470 [05:50<1:03:48,  1.56it/s]
2|512|Loss: 0.9767682552337646:   8%|▊         | 512/6470 [05:51<1:04:56,  1.53it/s]
2|513|Loss: 0.8719189167022705:   8%|▊         | 512/6470 [05:51<1:04:56,  1.53it/s]
2|513|Loss: 0.8719189167022705:   8%|▊         | 513/6470 [05:51<1:05:26,  1.52it/s]
2|514|Loss: 1.0139548778533936:   8%|▊         | 513/6470 [05:52<1:05:26,  1.52it/s]
2|514|Loss: 1.0139548778533936:   8%|▊         | 514/6470 [05:52<1:04:10,  1.55it/s]
2|515|Loss: 0.9011058807373047:   8%|▊         | 514/6470 [05:52<1:04:10,  1.55it/s]
2|515|Loss: 0.9011058807373047:   8%|▊         | 515/6470 [05:53<1:03:55,  1.55it/s]
2|516|Loss: 0.8429306745529175:   8%|▊         | 515/6470 [05:53<1:03:55,  1.55it/s]
2|516|Loss: 0.8429306745529175:   8%|▊         | 516/6470 [05:53<1:07:30,  1.47it/s]
2|517|Loss: 0.8335494995117188:   8%|▊         | 516/6470 [05:54<1:07:30,  1.47it/s]
2|517|Loss: 0.8335494995117188:   8%|▊         | 517/6470 [05:54<1:09:06,  1.44it/s]
2|518|Loss: 0.815574049949646:   8%|▊         | 517/6470 [05:54<1:09:06,  1.44it/s] 
2|518|Loss: 0.815574049949646:   8%|▊         | 518/6470 [05:55<1:06:16,  1.50it/s]
2|519|Loss: 0.8499290347099304:   8%|▊         | 518/6470 [05:55<1:06:16,  1.50it/s]
2|519|Loss: 0.8499290347099304:   8%|▊         | 519/6470 [05:55<1:00:42,  1.63it/s]
2|520|Loss: 0.8494347929954529:   8%|▊         | 519/6470 [05:56<1:00:42,  1.63it/s]
2|520|Loss: 0.8494347929954529:   8%|▊         | 520/6470 [05:56<1:05:16,  1.52it/s]
2|521|Loss: 0.8037029504776001:   8%|▊         | 520/6470 [05:56<1:05:16,  1.52it/s]
2|521|Loss: 0.8037029504776001:   8%|▊         | 521/6470 [05:57<1:04:41,  1.53it/s]
2|522|Loss: 0.8072901964187622:   8%|▊         | 521/6470 [05:57<1:04:41,  1.53it/s]
2|522|Loss: 0.8072901964187622:   8%|▊         | 522/6470 [05:57<1:03:54,  1.55it/s]
2|523|Loss: 0.8910719156265259:   8%|▊         | 522/6470 [05:58<1:03:54,  1.55it/s]
2|523|Loss: 0.8910719156265259:   8%|▊         | 523/6470 [05:58<1:07:45,  1.46it/s]
2|524|Loss: 0.8394689559936523:   8%|▊         | 523/6470 [05:58<1:07:45,  1.46it/s]
2|524|Loss: 0.8394689559936523:   8%|▊         | 524/6470 [05:59<1:06:59,  1.48it/s]
2|525|Loss: 0.9058912992477417:   8%|▊         | 524/6470 [05:59<1:06:59,  1.48it/s]
2|525|Loss: 0.9058912992477417:   8%|▊         | 525/6470 [05:59<1:07:14,  1.47it/s]
2|526|Loss: 0.9837677478790283:   8%|▊         | 525/6470 [06:00<1:07:14,  1.47it/s]
2|526|Loss: 0.9837677478790283:   8%|▊         | 526/6470 [06:00<1:03:26,  1.56it/s]
2|527|Loss: 0.7653713822364807:   8%|▊         | 526/6470 [06:00<1:03:26,  1.56it/s]
2|527|Loss: 0.7653713822364807:   8%|▊         | 527/6470 [06:01<1:05:06,  1.52it/s]
2|528|Loss: 0.9599477648735046:   8%|▊         | 527/6470 [06:01<1:05:06,  1.52it/s]
2|528|Loss: 0.9599477648735046:   8%|▊         | 528/6470 [06:01<1:05:06,  1.52it/s]
2|529|Loss: 0.9587617516517639:   8%|▊         | 528/6470 [06:02<1:05:06,  1.52it/s]
2|529|Loss: 0.9587617516517639:   8%|▊         | 529/6470 [06:02<1:04:27,  1.54it/s]
2|530|Loss: 0.8799214363098145:   8%|▊         | 529/6470 [06:02<1:04:27,  1.54it/s]
2|530|Loss: 0.8799214363098145:   8%|▊         | 530/6470 [06:03<1:07:56,  1.46it/s]
2|531|Loss: 0.8274820446968079:   8%|▊         | 530/6470 [06:03<1:07:56,  1.46it/s]
2|531|Loss: 0.8274820446968079:   8%|▊         | 531/6470 [06:03<1:07:49,  1.46it/s]
2|532|Loss: 1.0042351484298706:   8%|▊         | 531/6470 [06:04<1:07:49,  1.46it/s]
2|532|Loss: 1.0042351484298706:   8%|▊         | 532/6470 [06:04<1:06:36,  1.49it/s]
2|533|Loss: 0.792113184928894:   8%|▊         | 532/6470 [06:04<1:06:36,  1.49it/s] 
2|533|Loss: 0.792113184928894:   8%|▊         | 533/6470 [06:05<1:09:33,  1.42it/s]
2|534|Loss: 0.7444930672645569:   8%|▊         | 533/6470 [06:05<1:09:33,  1.42it/s]
2|534|Loss: 0.7444930672645569:   8%|▊         | 534/6470 [06:06<1:15:17,  1.31it/s]
2|535|Loss: 0.9110892415046692:   8%|▊         | 534/6470 [06:06<1:15:17,  1.31it/s]
2|535|Loss: 0.9110892415046692:   8%|▊         | 535/6470 [06:06<1:18:09,  1.27it/s]
2|536|Loss: 0.7390129566192627:   8%|▊         | 535/6470 [06:07<1:18:09,  1.27it/s]
2|536|Loss: 0.7390129566192627:   8%|▊         | 536/6470 [06:07<1:16:38,  1.29it/s]
2|537|Loss: 0.7910435795783997:   8%|▊         | 536/6470 [06:08<1:16:38,  1.29it/s]
2|537|Loss: 0.7910435795783997:   8%|▊         | 537/6470 [06:08<1:15:24,  1.31it/s]
2|538|Loss: 0.9278542399406433:   8%|▊         | 537/6470 [06:08<1:15:24,  1.31it/s]
2|538|Loss: 0.9278542399406433:   8%|▊         | 538/6470 [06:09<1:12:28,  1.36it/s]
2|539|Loss: 0.9798009991645813:   8%|▊         | 538/6470 [06:09<1:12:28,  1.36it/s]
2|539|Loss: 0.9798009991645813:   8%|▊         | 539/6470 [06:09<1:08:15,  1.45it/s]
2|540|Loss: 0.9465713500976562:   8%|▊         | 539/6470 [06:10<1:08:15,  1.45it/s]
2|540|Loss: 0.9465713500976562:   8%|▊         | 540/6470 [06:10<1:11:41,  1.38it/s]
2|541|Loss: 0.8865723609924316:   8%|▊         | 540/6470 [06:10<1:11:41,  1.38it/s]
2|541|Loss: 0.8865723609924316:   8%|▊         | 541/6470 [06:11<1:10:10,  1.41it/s]
2|542|Loss: 1.0715901851654053:   8%|▊         | 541/6470 [06:11<1:10:10,  1.41it/s]
2|542|Loss: 1.0715901851654053:   8%|▊         | 542/6470 [06:11<1:04:48,  1.52it/s]
2|543|Loss: 0.8088449835777283:   8%|▊         | 542/6470 [06:12<1:04:48,  1.52it/s]
2|543|Loss: 0.8088449835777283:   8%|▊         | 543/6470 [06:12<1:08:56,  1.43it/s]
2|544|Loss: 0.8103979229927063:   8%|▊         | 543/6470 [06:12<1:08:56,  1.43it/s]
2|544|Loss: 0.8103979229927063:   8%|▊         | 544/6470 [06:13<1:10:45,  1.40it/s]
2|545|Loss: 0.8147768378257751:   8%|▊         | 544/6470 [06:13<1:10:45,  1.40it/s]
2|545|Loss: 0.8147768378257751:   8%|▊         | 545/6470 [06:13<1:09:02,  1.43it/s]
2|546|Loss: 0.8653382658958435:   8%|▊         | 545/6470 [06:14<1:09:02,  1.43it/s]
2|546|Loss: 0.8653382658958435:   8%|▊         | 546/6470 [06:14<1:09:28,  1.42it/s]
2|547|Loss: 0.8179273009300232:   8%|▊         | 546/6470 [06:15<1:09:28,  1.42it/s]
2|547|Loss: 0.8179273009300232:   8%|▊         | 547/6470 [06:15<1:11:53,  1.37it/s]
2|548|Loss: 0.762437641620636:   8%|▊         | 547/6470 [06:15<1:11:53,  1.37it/s] 
2|548|Loss: 0.762437641620636:   8%|▊         | 548/6470 [06:16<1:10:16,  1.40it/s]
2|549|Loss: 0.8356890082359314:   8%|▊         | 548/6470 [06:16<1:10:16,  1.40it/s]
2|549|Loss: 0.8356890082359314:   8%|▊         | 549/6470 [06:16<1:06:00,  1.49it/s]
2|550|Loss: 0.7655110359191895:   8%|▊         | 549/6470 [06:17<1:06:00,  1.49it/s]
2|550|Loss: 0.7655110359191895:   9%|▊         | 550/6470 [06:17<1:11:19,  1.38it/s]
2|551|Loss: 0.8542758822441101:   9%|▊         | 550/6470 [06:17<1:11:19,  1.38it/s]
2|551|Loss: 0.8542758822441101:   9%|▊         | 551/6470 [06:18<1:13:10,  1.35it/s]
2|552|Loss: 1.0604780912399292:   9%|▊         | 551/6470 [06:18<1:13:10,  1.35it/s]
2|552|Loss: 1.0604780912399292:   9%|▊         | 552/6470 [06:19<1:15:25,  1.31it/s]
2|553|Loss: 0.8695944547653198:   9%|▊         | 552/6470 [06:19<1:15:25,  1.31it/s]
2|553|Loss: 0.8695944547653198:   9%|▊         | 553/6470 [06:19<1:13:07,  1.35it/s]
2|554|Loss: 0.8190985918045044:   9%|▊         | 553/6470 [06:20<1:13:07,  1.35it/s]
2|554|Loss: 0.8190985918045044:   9%|▊         | 554/6470 [06:20<1:07:56,  1.45it/s]
2|555|Loss: 0.8334317207336426:   9%|▊         | 554/6470 [06:20<1:07:56,  1.45it/s]
2|555|Loss: 0.8334317207336426:   9%|▊         | 555/6470 [06:20<1:03:52,  1.54it/s]
2|556|Loss: 0.8267647624015808:   9%|▊         | 555/6470 [06:21<1:03:52,  1.54it/s]
2|556|Loss: 0.8267647624015808:   9%|▊         | 556/6470 [06:21<1:02:27,  1.58it/s]
2|557|Loss: 0.9860395789146423:   9%|▊         | 556/6470 [06:21<1:02:27,  1.58it/s]
2|557|Loss: 0.9860395789146423:   9%|▊         | 557/6470 [06:22<1:05:12,  1.51it/s]
2|558|Loss: 0.8494070172309875:   9%|▊         | 557/6470 [06:22<1:05:12,  1.51it/s]
2|558|Loss: 0.8494070172309875:   9%|▊         | 558/6470 [06:23<1:09:03,  1.43it/s]
2|559|Loss: 0.8745356202125549:   9%|▊         | 558/6470 [06:23<1:09:03,  1.43it/s]
2|559|Loss: 0.8745356202125549:   9%|▊         | 559/6470 [06:23<1:10:29,  1.40it/s]
2|560|Loss: 0.8469673991203308:   9%|▊         | 559/6470 [06:24<1:10:29,  1.40it/s]
2|560|Loss: 0.8469673991203308:   9%|▊         | 560/6470 [06:24<1:08:44,  1.43it/s]
2|561|Loss: 0.9528723955154419:   9%|▊         | 560/6470 [06:24<1:08:44,  1.43it/s]
2|561|Loss: 0.9528723955154419:   9%|▊         | 561/6470 [06:25<1:08:34,  1.44it/s]
2|562|Loss: 0.8678122758865356:   9%|▊         | 561/6470 [06:25<1:08:34,  1.44it/s]
2|562|Loss: 0.8678122758865356:   9%|▊         | 562/6470 [06:25<1:05:12,  1.51it/s]
2|563|Loss: 0.8960767388343811:   9%|▊         | 562/6470 [06:26<1:05:12,  1.51it/s]
2|563|Loss: 0.8960767388343811:   9%|▊         | 563/6470 [06:26<1:06:20,  1.48it/s]
2|564|Loss: 0.7589660882949829:   9%|▊         | 563/6470 [06:26<1:06:20,  1.48it/s]
2|564|Loss: 0.7589660882949829:   9%|▊         | 564/6470 [06:27<1:06:03,  1.49it/s]
2|565|Loss: 0.9501291513442993:   9%|▊         | 564/6470 [06:27<1:06:03,  1.49it/s]
2|565|Loss: 0.9501291513442993:   9%|▊         | 565/6470 [06:27<1:03:30,  1.55it/s]
2|566|Loss: 0.8285417556762695:   9%|▊         | 565/6470 [06:28<1:03:30,  1.55it/s]
2|566|Loss: 0.8285417556762695:   9%|▊         | 566/6470 [06:28<1:07:25,  1.46it/s]
2|567|Loss: 0.8966066837310791:   9%|▊         | 566/6470 [06:28<1:07:25,  1.46it/s]
2|567|Loss: 0.8966066837310791:   9%|▉         | 567/6470 [06:29<1:06:44,  1.47it/s]
2|568|Loss: 0.7966817021369934:   9%|▉         | 567/6470 [06:29<1:06:44,  1.47it/s]
2|568|Loss: 0.7966817021369934:   9%|▉         | 568/6470 [06:29<1:05:21,  1.50it/s]
2|569|Loss: 0.8721045255661011:   9%|▉         | 568/6470 [06:30<1:05:21,  1.50it/s]
2|569|Loss: 0.8721045255661011:   9%|▉         | 569/6470 [06:30<1:04:37,  1.52it/s]
2|570|Loss: 0.8817946910858154:   9%|▉         | 569/6470 [06:30<1:04:37,  1.52it/s]
2|570|Loss: 0.8817946910858154:   9%|▉         | 570/6470 [06:31<1:05:31,  1.50it/s]
2|571|Loss: 0.9788978695869446:   9%|▉         | 570/6470 [06:31<1:05:31,  1.50it/s]
2|571|Loss: 0.9788978695869446:   9%|▉         | 571/6470 [06:31<1:05:17,  1.51it/s]
2|572|Loss: 0.8309928774833679:   9%|▉         | 571/6470 [06:32<1:05:17,  1.51it/s]
2|572|Loss: 0.8309928774833679:   9%|▉         | 572/6470 [06:32<1:06:11,  1.48it/s]
2|573|Loss: 0.9334450364112854:   9%|▉         | 572/6470 [06:32<1:06:11,  1.48it/s]
2|573|Loss: 0.9334450364112854:   9%|▉         | 573/6470 [06:33<1:07:01,  1.47it/s]
2|574|Loss: 0.7611995339393616:   9%|▉         | 573/6470 [06:33<1:07:01,  1.47it/s]
2|574|Loss: 0.7611995339393616:   9%|▉         | 574/6470 [06:33<1:04:33,  1.52it/s]
2|575|Loss: 0.8075372576713562:   9%|▉         | 574/6470 [06:34<1:04:33,  1.52it/s]
2|575|Loss: 0.8075372576713562:   9%|▉         | 575/6470 [06:34<1:03:54,  1.54it/s]
2|576|Loss: 0.816126823425293:   9%|▉         | 575/6470 [06:34<1:03:54,  1.54it/s] 
2|576|Loss: 0.816126823425293:   9%|▉         | 576/6470 [06:34<1:03:19,  1.55it/s]
2|577|Loss: 0.8929146528244019:   9%|▉         | 576/6470 [06:35<1:03:19,  1.55it/s]
2|577|Loss: 0.8929146528244019:   9%|▉         | 577/6470 [06:35<1:05:40,  1.50it/s]
2|578|Loss: 0.7204451560974121:   9%|▉         | 577/6470 [06:36<1:05:40,  1.50it/s]
2|578|Loss: 0.7204451560974121:   9%|▉         | 578/6470 [06:36<1:02:31,  1.57it/s]
2|579|Loss: 0.9880368113517761:   9%|▉         | 578/6470 [06:36<1:02:31,  1.57it/s]
2|579|Loss: 0.9880368113517761:   9%|▉         | 579/6470 [06:36<59:41,  1.64it/s]  
2|580|Loss: 0.7373381853103638:   9%|▉         | 579/6470 [06:37<59:41,  1.64it/s]
2|580|Loss: 0.7373381853103638:   9%|▉         | 580/6470 [06:37<1:05:34,  1.50it/s]
2|581|Loss: 1.0223064422607422:   9%|▉         | 580/6470 [06:38<1:05:34,  1.50it/s]
2|581|Loss: 1.0223064422607422:   9%|▉         | 581/6470 [06:38<1:08:23,  1.44it/s]
2|582|Loss: 0.8250814080238342:   9%|▉         | 581/6470 [06:38<1:08:23,  1.44it/s]
2|582|Loss: 0.8250814080238342:   9%|▉         | 582/6470 [06:39<1:06:44,  1.47it/s]
2|583|Loss: 0.8792083859443665:   9%|▉         | 582/6470 [06:39<1:06:44,  1.47it/s]
2|583|Loss: 0.8792083859443665:   9%|▉         | 583/6470 [06:39<1:10:48,  1.39it/s]
2|584|Loss: 0.9490213990211487:   9%|▉         | 583/6470 [06:40<1:10:48,  1.39it/s]
2|584|Loss: 0.9490213990211487:   9%|▉         | 584/6470 [06:40<1:11:38,  1.37it/s]
2|585|Loss: 0.8585002422332764:   9%|▉         | 584/6470 [06:40<1:11:38,  1.37it/s]
2|585|Loss: 0.8585002422332764:   9%|▉         | 585/6470 [06:41<1:08:04,  1.44it/s]
2|586|Loss: 1.1103590726852417:   9%|▉         | 585/6470 [06:41<1:08:04,  1.44it/s]
2|586|Loss: 1.1103590726852417:   9%|▉         | 586/6470 [06:41<1:07:45,  1.45it/s]
2|587|Loss: 0.9882827401161194:   9%|▉         | 586/6470 [06:42<1:07:45,  1.45it/s]
2|587|Loss: 0.9882827401161194:   9%|▉         | 587/6470 [06:42<1:08:19,  1.44it/s]
2|588|Loss: 0.8530434370040894:   9%|▉         | 587/6470 [06:43<1:08:19,  1.44it/s]
2|588|Loss: 0.8530434370040894:   9%|▉         | 588/6470 [06:43<1:14:41,  1.31it/s]
2|589|Loss: 0.8982570767402649:   9%|▉         | 588/6470 [06:43<1:14:41,  1.31it/s]
2|589|Loss: 0.8982570767402649:   9%|▉         | 589/6470 [06:44<1:08:50,  1.42it/s]
2|590|Loss: 0.753565788269043:   9%|▉         | 589/6470 [06:44<1:08:50,  1.42it/s] 
2|590|Loss: 0.753565788269043:   9%|▉         | 590/6470 [06:44<1:07:11,  1.46it/s]
2|591|Loss: 0.8747369647026062:   9%|▉         | 590/6470 [06:45<1:07:11,  1.46it/s]
2|591|Loss: 0.8747369647026062:   9%|▉         | 591/6470 [06:45<1:06:08,  1.48it/s]
2|592|Loss: 0.8426290154457092:   9%|▉         | 591/6470 [06:45<1:06:08,  1.48it/s]
2|592|Loss: 0.8426290154457092:   9%|▉         | 592/6470 [06:45<1:00:39,  1.62it/s]
2|593|Loss: 0.91900235414505:   9%|▉         | 592/6470 [06:46<1:00:39,  1.62it/s]  
2|593|Loss: 0.91900235414505:   9%|▉         | 593/6470 [06:46<59:48,  1.64it/s]  
2|594|Loss: 1.0235450267791748:   9%|▉         | 593/6470 [06:46<59:48,  1.64it/s]
2|594|Loss: 1.0235450267791748:   9%|▉         | 594/6470 [06:47<1:00:00,  1.63it/s]
2|595|Loss: 0.732390820980072:   9%|▉         | 594/6470 [06:47<1:00:00,  1.63it/s] 
2|595|Loss: 0.732390820980072:   9%|▉         | 595/6470 [06:47<1:00:39,  1.61it/s]
2|596|Loss: 0.969474196434021:   9%|▉         | 595/6470 [06:48<1:00:39,  1.61it/s]
2|596|Loss: 0.969474196434021:   9%|▉         | 596/6470 [06:48<1:00:48,  1.61it/s]
2|597|Loss: 0.9739540815353394:   9%|▉         | 596/6470 [06:48<1:00:48,  1.61it/s]
2|597|Loss: 0.9739540815353394:   9%|▉         | 597/6470 [06:49<1:03:17,  1.55it/s]
2|598|Loss: 0.739090621471405:   9%|▉         | 597/6470 [06:49<1:03:17,  1.55it/s] 
2|598|Loss: 0.739090621471405:   9%|▉         | 598/6470 [06:49<1:03:38,  1.54it/s]
2|599|Loss: 1.0190569162368774:   9%|▉         | 598/6470 [06:50<1:03:38,  1.54it/s]
2|599|Loss: 1.0190569162368774:   9%|▉         | 599/6470 [06:50<1:03:17,  1.55it/s]
2|600|Loss: 0.9901528358459473:   9%|▉         | 599/6470 [06:50<1:03:17,  1.55it/s]
2|600|Loss: 0.9901528358459473:   9%|▉         | 600/6470 [06:51<1:05:20,  1.50it/s]
2|601|Loss: 0.886951208114624:   9%|▉         | 600/6470 [06:51<1:05:20,  1.50it/s] 
2|601|Loss: 0.886951208114624:   9%|▉         | 601/6470 [06:51<1:04:33,  1.52it/s]
2|602|Loss: 0.7653648257255554:   9%|▉         | 601/6470 [06:52<1:04:33,  1.52it/s]
2|602|Loss: 0.7653648257255554:   9%|▉         | 602/6470 [06:52<1:05:44,  1.49it/s]
2|603|Loss: 0.9312692880630493:   9%|▉         | 602/6470 [06:52<1:05:44,  1.49it/s]
2|603|Loss: 0.9312692880630493:   9%|▉         | 603/6470 [06:53<1:05:18,  1.50it/s]
2|604|Loss: 0.8866949677467346:   9%|▉         | 603/6470 [06:53<1:05:18,  1.50it/s]
2|604|Loss: 0.8866949677467346:   9%|▉         | 604/6470 [06:53<1:10:02,  1.40it/s]
2|605|Loss: 0.8367739915847778:   9%|▉         | 604/6470 [06:54<1:10:02,  1.40it/s]
2|605|Loss: 0.8367739915847778:   9%|▉         | 605/6470 [06:54<1:11:55,  1.36it/s]
2|606|Loss: 0.8510379195213318:   9%|▉         | 605/6470 [06:55<1:11:55,  1.36it/s]
2|606|Loss: 0.8510379195213318:   9%|▉         | 606/6470 [06:55<1:15:34,  1.29it/s]
2|607|Loss: 0.9683966636657715:   9%|▉         | 606/6470 [06:55<1:15:34,  1.29it/s]
2|607|Loss: 0.9683966636657715:   9%|▉         | 607/6470 [06:56<1:15:44,  1.29it/s]
2|608|Loss: 0.832654595375061:   9%|▉         | 607/6470 [06:56<1:15:44,  1.29it/s] 
2|608|Loss: 0.832654595375061:   9%|▉         | 608/6470 [06:57<1:14:40,  1.31it/s]
2|609|Loss: 0.8988701105117798:   9%|▉         | 608/6470 [06:57<1:14:40,  1.31it/s]
2|609|Loss: 0.8988701105117798:   9%|▉         | 609/6470 [06:57<1:11:17,  1.37it/s]
2|610|Loss: 0.8021631836891174:   9%|▉         | 609/6470 [06:58<1:11:17,  1.37it/s]
2|610|Loss: 0.8021631836891174:   9%|▉         | 610/6470 [06:58<1:07:09,  1.45it/s]
2|611|Loss: 0.9411912560462952:   9%|▉         | 610/6470 [06:58<1:07:09,  1.45it/s]
2|611|Loss: 0.9411912560462952:   9%|▉         | 611/6470 [06:58<1:07:05,  1.46it/s]
2|612|Loss: 0.7451496720314026:   9%|▉         | 611/6470 [06:59<1:07:05,  1.46it/s]
2|612|Loss: 0.7451496720314026:   9%|▉         | 612/6470 [06:59<1:07:28,  1.45it/s]
2|613|Loss: 0.8858713507652283:   9%|▉         | 612/6470 [07:00<1:07:28,  1.45it/s]
2|613|Loss: 0.8858713507652283:   9%|▉         | 613/6470 [07:00<1:06:11,  1.47it/s]
2|614|Loss: 0.8886173963546753:   9%|▉         | 613/6470 [07:00<1:06:11,  1.47it/s]
2|614|Loss: 0.8886173963546753:   9%|▉         | 614/6470 [07:00<1:06:26,  1.47it/s]
2|615|Loss: 0.8290815949440002:   9%|▉         | 614/6470 [07:01<1:06:26,  1.47it/s]
2|615|Loss: 0.8290815949440002:  10%|▉         | 615/6470 [07:01<1:12:06,  1.35it/s]
2|616|Loss: 0.9636868238449097:  10%|▉         | 615/6470 [07:02<1:12:06,  1.35it/s]
2|616|Loss: 0.9636868238449097:  10%|▉         | 616/6470 [07:02<1:15:27,  1.29it/s]
2|617|Loss: 0.8264467120170593:  10%|▉         | 616/6470 [07:03<1:15:27,  1.29it/s]
2|617|Loss: 0.8264467120170593:  10%|▉         | 617/6470 [07:03<1:15:13,  1.30it/s]
2|618|Loss: 1.0233889818191528:  10%|▉         | 617/6470 [07:03<1:15:13,  1.30it/s]
2|618|Loss: 1.0233889818191528:  10%|▉         | 618/6470 [07:04<1:14:48,  1.30it/s]
2|619|Loss: 1.0618761777877808:  10%|▉         | 618/6470 [07:04<1:14:48,  1.30it/s]
2|619|Loss: 1.0618761777877808:  10%|▉         | 619/6470 [07:04<1:13:23,  1.33it/s]
2|620|Loss: 0.8554785251617432:  10%|▉         | 619/6470 [07:05<1:13:23,  1.33it/s]
2|620|Loss: 0.8554785251617432:  10%|▉         | 620/6470 [07:05<1:07:41,  1.44it/s]
2|621|Loss: 0.7640910148620605:  10%|▉         | 620/6470 [07:05<1:07:41,  1.44it/s]
2|621|Loss: 0.7640910148620605:  10%|▉         | 621/6470 [07:06<1:07:54,  1.44it/s]
2|622|Loss: 0.8676934242248535:  10%|▉         | 621/6470 [07:06<1:07:54,  1.44it/s]
2|622|Loss: 0.8676934242248535:  10%|▉         | 622/6470 [07:06<1:09:05,  1.41it/s]
2|623|Loss: 0.9172429442405701:  10%|▉         | 622/6470 [07:07<1:09:05,  1.41it/s]
2|623|Loss: 0.9172429442405701:  10%|▉         | 623/6470 [07:07<1:09:43,  1.40it/s]
2|624|Loss: 0.792145311832428:  10%|▉         | 623/6470 [07:08<1:09:43,  1.40it/s] 
2|624|Loss: 0.792145311832428:  10%|▉         | 624/6470 [07:08<1:12:16,  1.35it/s]
2|625|Loss: 0.7522200345993042:  10%|▉         | 624/6470 [07:08<1:12:16,  1.35it/s]
2|625|Loss: 0.7522200345993042:  10%|▉         | 625/6470 [07:09<1:09:56,  1.39it/s]
2|626|Loss: 0.8101615309715271:  10%|▉         | 625/6470 [07:09<1:09:56,  1.39it/s]
2|626|Loss: 0.8101615309715271:  10%|▉         | 626/6470 [07:09<1:10:00,  1.39it/s]
2|627|Loss: 0.7910147309303284:  10%|▉         | 626/6470 [07:10<1:10:00,  1.39it/s]
2|627|Loss: 0.7910147309303284:  10%|▉         | 627/6470 [07:10<1:06:12,  1.47it/s]
2|628|Loss: 0.8644894361495972:  10%|▉         | 627/6470 [07:10<1:06:12,  1.47it/s]
2|628|Loss: 0.8644894361495972:  10%|▉         | 628/6470 [07:11<1:06:31,  1.46it/s]
2|629|Loss: 0.859601616859436:  10%|▉         | 628/6470 [07:11<1:06:31,  1.46it/s] 
2|629|Loss: 0.859601616859436:  10%|▉         | 629/6470 [07:11<1:07:09,  1.45it/s]
2|630|Loss: 0.840322732925415:  10%|▉         | 629/6470 [07:12<1:07:09,  1.45it/s]
2|630|Loss: 0.840322732925415:  10%|▉         | 630/6470 [07:12<1:05:41,  1.48it/s]
2|631|Loss: 0.9173578023910522:  10%|▉         | 630/6470 [07:12<1:05:41,  1.48it/s]
2|631|Loss: 0.9173578023910522:  10%|▉         | 631/6470 [07:13<1:04:36,  1.51it/s]
2|632|Loss: 0.8025763630867004:  10%|▉         | 631/6470 [07:13<1:04:36,  1.51it/s]
2|632|Loss: 0.8025763630867004:  10%|▉         | 632/6470 [07:13<56:56,  1.71it/s]  
2|633|Loss: 0.8727220296859741:  10%|▉         | 632/6470 [07:13<56:56,  1.71it/s]
2|633|Loss: 0.8727220296859741:  10%|▉         | 633/6470 [07:14<54:03,  1.80it/s]
2|634|Loss: 0.8771793246269226:  10%|▉         | 633/6470 [07:14<54:03,  1.80it/s]
2|634|Loss: 0.8771793246269226:  10%|▉         | 634/6470 [07:14<53:44,  1.81it/s]
2|635|Loss: 1.0092817544937134:  10%|▉         | 634/6470 [07:14<53:44,  1.81it/s]
2|635|Loss: 1.0092817544937134:  10%|▉         | 635/6470 [07:15<1:00:50,  1.60it/s]
2|636|Loss: 1.0160781145095825:  10%|▉         | 635/6470 [07:15<1:00:50,  1.60it/s]
2|636|Loss: 1.0160781145095825:  10%|▉         | 636/6470 [07:16<1:02:10,  1.56it/s]
2|637|Loss: 0.915552020072937:  10%|▉         | 636/6470 [07:16<1:02:10,  1.56it/s] 
2|637|Loss: 0.915552020072937:  10%|▉         | 637/6470 [07:16<1:06:58,  1.45it/s]
2|638|Loss: 0.947675883769989:  10%|▉         | 637/6470 [07:17<1:06:58,  1.45it/s]
2|638|Loss: 0.947675883769989:  10%|▉         | 638/6470 [07:17<1:05:35,  1.48it/s]
2|639|Loss: 0.7069388031959534:  10%|▉         | 638/6470 [07:17<1:05:35,  1.48it/s]
2|639|Loss: 0.7069388031959534:  10%|▉         | 639/6470 [07:18<1:03:33,  1.53it/s]
2|640|Loss: 0.8239738345146179:  10%|▉         | 639/6470 [07:18<1:03:33,  1.53it/s]
2|640|Loss: 0.8239738345146179:  10%|▉         | 640/6470 [07:18<1:03:04,  1.54it/s]
2|641|Loss: 0.95179682970047:  10%|▉         | 640/6470 [07:19<1:03:04,  1.54it/s]  
2|641|Loss: 0.95179682970047:  10%|▉         | 641/6470 [07:19<1:04:56,  1.50it/s]
2|642|Loss: 0.9435818195343018:  10%|▉         | 641/6470 [07:19<1:04:56,  1.50it/s]
2|642|Loss: 0.9435818195343018:  10%|▉         | 642/6470 [07:20<1:04:04,  1.52it/s]
2|643|Loss: 0.8685361742973328:  10%|▉         | 642/6470 [07:20<1:04:04,  1.52it/s]
2|643|Loss: 0.8685361742973328:  10%|▉         | 643/6470 [07:20<1:03:13,  1.54it/s]
2|644|Loss: 0.7487511038780212:  10%|▉         | 643/6470 [07:21<1:03:13,  1.54it/s]
2|644|Loss: 0.7487511038780212:  10%|▉         | 644/6470 [07:21<1:02:25,  1.56it/s]
2|645|Loss: 0.8489905595779419:  10%|▉         | 644/6470 [07:21<1:02:25,  1.56it/s]
2|645|Loss: 0.8489905595779419:  10%|▉         | 645/6470 [07:22<1:03:40,  1.52it/s]
2|646|Loss: 0.9556794166564941:  10%|▉         | 645/6470 [07:22<1:03:40,  1.52it/s]
2|646|Loss: 0.9556794166564941:  10%|▉         | 646/6470 [07:22<1:02:56,  1.54it/s]
2|647|Loss: 0.8378927111625671:  10%|▉         | 646/6470 [07:23<1:02:56,  1.54it/s]
2|647|Loss: 0.8378927111625671:  10%|█         | 647/6470 [07:23<1:07:35,  1.44it/s]
2|648|Loss: 0.7402419447898865:  10%|█         | 647/6470 [07:23<1:07:35,  1.44it/s]
2|648|Loss: 0.7402419447898865:  10%|█         | 648/6470 [07:24<1:10:35,  1.37it/s]
2|649|Loss: 0.9955861568450928:  10%|█         | 648/6470 [07:24<1:10:35,  1.37it/s]
2|649|Loss: 0.9955861568450928:  10%|█         | 649/6470 [07:25<1:13:22,  1.32it/s]
2|650|Loss: 0.7413692474365234:  10%|█         | 649/6470 [07:25<1:13:22,  1.32it/s]
2|650|Loss: 0.7413692474365234:  10%|█         | 650/6470 [07:25<1:13:41,  1.32it/s]
2|651|Loss: 0.8714982867240906:  10%|█         | 650/6470 [07:26<1:13:41,  1.32it/s]
2|651|Loss: 0.8714982867240906:  10%|█         | 651/6470 [07:26<1:10:57,  1.37it/s]
2|652|Loss: 0.8708413243293762:  10%|█         | 651/6470 [07:26<1:10:57,  1.37it/s]
2|652|Loss: 0.8708413243293762:  10%|█         | 652/6470 [07:27<1:13:27,  1.32it/s]
2|653|Loss: 0.9117733836174011:  10%|█         | 652/6470 [07:27<1:13:27,  1.32it/s]
2|653|Loss: 0.9117733836174011:  10%|█         | 653/6470 [07:28<1:16:13,  1.27it/s]
2|654|Loss: 0.857720673084259:  10%|█         | 653/6470 [07:28<1:16:13,  1.27it/s] 
2|654|Loss: 0.857720673084259:  10%|█         | 654/6470 [07:28<1:15:26,  1.28it/s]
2|655|Loss: 0.8693494200706482:  10%|█         | 654/6470 [07:29<1:15:26,  1.28it/s]
2|655|Loss: 0.8693494200706482:  10%|█         | 655/6470 [07:29<1:10:46,  1.37it/s]
2|656|Loss: 1.0017337799072266:  10%|█         | 655/6470 [07:29<1:10:46,  1.37it/s]
2|656|Loss: 1.0017337799072266:  10%|█         | 656/6470 [07:30<1:12:17,  1.34it/s]
2|657|Loss: 0.8383899927139282:  10%|█         | 656/6470 [07:30<1:12:17,  1.34it/s]
2|657|Loss: 0.8383899927139282:  10%|█         | 657/6470 [07:31<1:14:36,  1.30it/s]
2|658|Loss: 0.8184502720832825:  10%|█         | 657/6470 [07:31<1:14:36,  1.30it/s]
2|658|Loss: 0.8184502720832825:  10%|█         | 658/6470 [07:32<1:16:46,  1.26it/s]
2|659|Loss: 0.8450782299041748:  10%|█         | 658/6470 [07:32<1:16:46,  1.26it/s]
2|659|Loss: 0.8450782299041748:  10%|█         | 659/6470 [07:32<1:13:24,  1.32it/s]
2|660|Loss: 1.079586148262024:  10%|█         | 659/6470 [07:33<1:13:24,  1.32it/s] 
2|660|Loss: 1.079586148262024:  10%|█         | 660/6470 [07:33<1:11:10,  1.36it/s]
2|661|Loss: 0.9062299728393555:  10%|█         | 660/6470 [07:33<1:11:10,  1.36it/s]
2|661|Loss: 0.9062299728393555:  10%|█         | 661/6470 [07:34<1:10:10,  1.38it/s]
2|662|Loss: 0.8254455327987671:  10%|█         | 661/6470 [07:34<1:10:10,  1.38it/s]
2|662|Loss: 0.8254455327987671:  10%|█         | 662/6470 [07:34<1:12:30,  1.33it/s]
2|663|Loss: 0.770217776298523:  10%|█         | 662/6470 [07:35<1:12:30,  1.33it/s] 
2|663|Loss: 0.770217776298523:  10%|█         | 663/6470 [07:35<1:12:55,  1.33it/s]
2|664|Loss: 0.8625494241714478:  10%|█         | 663/6470 [07:36<1:12:55,  1.33it/s]
2|664|Loss: 0.8625494241714478:  10%|█         | 664/6470 [07:36<1:11:50,  1.35it/s]
2|665|Loss: 0.9414742588996887:  10%|█         | 664/6470 [07:36<1:11:50,  1.35it/s]
2|665|Loss: 0.9414742588996887:  10%|█         | 665/6470 [07:36<1:04:51,  1.49it/s]
2|666|Loss: 0.8330842852592468:  10%|█         | 665/6470 [07:37<1:04:51,  1.49it/s]
2|666|Loss: 0.8330842852592468:  10%|█         | 666/6470 [07:37<1:02:20,  1.55it/s]
2|667|Loss: 0.8930432200431824:  10%|█         | 666/6470 [07:37<1:02:20,  1.55it/s]
2|667|Loss: 0.8930432200431824:  10%|█         | 667/6470 [07:37<59:21,  1.63it/s]  
2|668|Loss: 0.8270319104194641:  10%|█         | 667/6470 [07:38<59:21,  1.63it/s]
2|668|Loss: 0.8270319104194641:  10%|█         | 668/6470 [07:38<56:18,  1.72it/s]
2|669|Loss: 0.818798303604126:  10%|█         | 668/6470 [07:38<56:18,  1.72it/s] 
2|669|Loss: 0.818798303604126:  10%|█         | 669/6470 [07:39<1:01:02,  1.58it/s]
2|670|Loss: 0.9385531544685364:  10%|█         | 669/6470 [07:39<1:01:02,  1.58it/s]
2|670|Loss: 0.9385531544685364:  10%|█         | 670/6470 [07:40<1:07:56,  1.42it/s]
2|671|Loss: 0.982272744178772:  10%|█         | 670/6470 [07:40<1:07:56,  1.42it/s] 
2|671|Loss: 0.982272744178772:  10%|█         | 671/6470 [07:40<1:09:57,  1.38it/s]
2|672|Loss: 0.855141818523407:  10%|█         | 671/6470 [07:41<1:09:57,  1.38it/s]
2|672|Loss: 0.855141818523407:  10%|█         | 672/6470 [07:41<1:12:08,  1.34it/s]
2|673|Loss: 0.7866318225860596:  10%|█         | 672/6470 [07:42<1:12:08,  1.34it/s]
2|673|Loss: 0.7866318225860596:  10%|█         | 673/6470 [07:42<1:09:33,  1.39it/s]
2|674|Loss: 0.9980617761611938:  10%|█         | 673/6470 [07:42<1:09:33,  1.39it/s]
2|674|Loss: 0.9980617761611938:  10%|█         | 674/6470 [07:43<1:08:29,  1.41it/s]
2|675|Loss: 0.8764474987983704:  10%|█         | 674/6470 [07:43<1:08:29,  1.41it/s]
2|675|Loss: 0.8764474987983704:  10%|█         | 675/6470 [07:43<1:09:01,  1.40it/s]
2|676|Loss: 0.7747309803962708:  10%|█         | 675/6470 [07:44<1:09:01,  1.40it/s]
2|676|Loss: 0.7747309803962708:  10%|█         | 676/6470 [07:44<1:03:02,  1.53it/s]
2|677|Loss: 0.9225573539733887:  10%|█         | 676/6470 [07:44<1:03:02,  1.53it/s]
2|677|Loss: 0.9225573539733887:  10%|█         | 677/6470 [07:44<1:01:29,  1.57it/s]
2|678|Loss: 0.8365854620933533:  10%|█         | 677/6470 [07:45<1:01:29,  1.57it/s]
2|678|Loss: 0.8365854620933533:  10%|█         | 678/6470 [07:45<1:05:28,  1.47it/s]
2|679|Loss: 0.8431742787361145:  10%|█         | 678/6470 [07:46<1:05:28,  1.47it/s]
2|679|Loss: 0.8431742787361145:  10%|█         | 679/6470 [07:46<1:09:39,  1.39it/s]
2|680|Loss: 0.9124067425727844:  10%|█         | 679/6470 [07:46<1:09:39,  1.39it/s]
2|680|Loss: 0.9124067425727844:  11%|█         | 680/6470 [07:47<1:10:34,  1.37it/s]
2|681|Loss: 0.8626323938369751:  11%|█         | 680/6470 [07:47<1:10:34,  1.37it/s]
2|681|Loss: 0.8626323938369751:  11%|█         | 681/6470 [07:48<1:12:09,  1.34it/s]
2|682|Loss: 0.8897148966789246:  11%|█         | 681/6470 [07:48<1:12:09,  1.34it/s]
2|682|Loss: 0.8897148966789246:  11%|█         | 682/6470 [07:48<1:12:08,  1.34it/s]
2|683|Loss: 0.834433376789093:  11%|█         | 682/6470 [07:49<1:12:08,  1.34it/s] 
2|683|Loss: 0.834433376789093:  11%|█         | 683/6470 [07:49<1:11:30,  1.35it/s]
2|684|Loss: 0.8700276613235474:  11%|█         | 683/6470 [07:49<1:11:30,  1.35it/s]
2|684|Loss: 0.8700276613235474:  11%|█         | 684/6470 [07:50<1:08:58,  1.40it/s]
2|685|Loss: 1.1300499439239502:  11%|█         | 684/6470 [07:50<1:08:58,  1.40it/s]
2|685|Loss: 1.1300499439239502:  11%|█         | 685/6470 [07:50<1:08:11,  1.41it/s]
2|686|Loss: 0.9127532839775085:  11%|█         | 685/6470 [07:51<1:08:11,  1.41it/s]
2|686|Loss: 0.9127532839775085:  11%|█         | 686/6470 [07:51<1:06:37,  1.45it/s]
2|687|Loss: 0.9511237144470215:  11%|█         | 686/6470 [07:51<1:06:37,  1.45it/s]
2|687|Loss: 0.9511237144470215:  11%|█         | 687/6470 [07:52<1:07:05,  1.44it/s]
2|688|Loss: 0.9067524075508118:  11%|█         | 687/6470 [07:52<1:07:05,  1.44it/s]
2|688|Loss: 0.9067524075508118:  11%|█         | 688/6470 [07:53<1:10:50,  1.36it/s]
2|689|Loss: 0.7886248230934143:  11%|█         | 688/6470 [07:53<1:10:50,  1.36it/s]
2|689|Loss: 0.7886248230934143:  11%|█         | 689/6470 [07:53<1:09:34,  1.38it/s]
2|690|Loss: 0.8659549951553345:  11%|█         | 689/6470 [07:54<1:09:34,  1.38it/s]
2|690|Loss: 0.8659549951553345:  11%|█         | 690/6470 [07:54<1:08:48,  1.40it/s]
2|691|Loss: 0.7517262101173401:  11%|█         | 690/6470 [07:54<1:08:48,  1.40it/s]
2|691|Loss: 0.7517262101173401:  11%|█         | 691/6470 [07:55<1:13:12,  1.32it/s]
2|692|Loss: 0.6953121423721313:  11%|█         | 691/6470 [07:55<1:13:12,  1.32it/s]
2|692|Loss: 0.6953121423721313:  11%|█         | 692/6470 [07:55<1:11:30,  1.35it/s]
2|693|Loss: 1.1525568962097168:  11%|█         | 692/6470 [07:56<1:11:30,  1.35it/s]
2|693|Loss: 1.1525568962097168:  11%|█         | 693/6470 [07:56<1:08:39,  1.40it/s]
2|694|Loss: 0.8464725017547607:  11%|█         | 693/6470 [07:56<1:08:39,  1.40it/s]
2|694|Loss: 0.8464725017547607:  11%|█         | 694/6470 [07:57<1:06:09,  1.46it/s]
2|695|Loss: 0.836558997631073:  11%|█         | 694/6470 [07:57<1:06:09,  1.46it/s] 
2|695|Loss: 0.836558997631073:  11%|█         | 695/6470 [07:57<1:06:04,  1.46it/s]
2|696|Loss: 0.8657122254371643:  11%|█         | 695/6470 [07:58<1:06:04,  1.46it/s]
2|696|Loss: 0.8657122254371643:  11%|█         | 696/6470 [07:58<1:05:08,  1.48it/s]
2|697|Loss: 0.8089253902435303:  11%|█         | 696/6470 [07:58<1:05:08,  1.48it/s]
2|697|Loss: 0.8089253902435303:  11%|█         | 697/6470 [07:59<1:00:17,  1.60it/s]
2|698|Loss: 0.7349631786346436:  11%|█         | 697/6470 [07:59<1:00:17,  1.60it/s]
2|698|Loss: 0.7349631786346436:  11%|█         | 698/6470 [07:59<59:40,  1.61it/s]  
2|699|Loss: 0.9383966326713562:  11%|█         | 698/6470 [08:00<59:40,  1.61it/s]
2|699|Loss: 0.9383966326713562:  11%|█         | 699/6470 [08:00<1:00:25,  1.59it/s]
2|700|Loss: 0.8092405200004578:  11%|█         | 699/6470 [08:00<1:00:25,  1.59it/s]
2|700|Loss: 0.8092405200004578:  11%|█         | 700/6470 [08:01<1:05:15,  1.47it/s]
2|701|Loss: 0.8970450162887573:  11%|█         | 700/6470 [08:01<1:05:15,  1.47it/s]
2|701|Loss: 0.8970450162887573:  11%|█         | 701/6470 [08:01<1:05:05,  1.48it/s]
2|702|Loss: 0.7917859554290771:  11%|█         | 701/6470 [08:02<1:05:05,  1.48it/s]
2|702|Loss: 0.7917859554290771:  11%|█         | 702/6470 [08:02<1:12:04,  1.33it/s]
2|703|Loss: 0.7822318077087402:  11%|█         | 702/6470 [08:03<1:12:04,  1.33it/s]
2|703|Loss: 0.7822318077087402:  11%|█         | 703/6470 [08:03<1:11:23,  1.35it/s]
2|704|Loss: 0.9351853132247925:  11%|█         | 703/6470 [08:03<1:11:23,  1.35it/s]
2|704|Loss: 0.9351853132247925:  11%|█         | 704/6470 [08:04<1:09:51,  1.38it/s]
2|705|Loss: 1.0060067176818848:  11%|█         | 704/6470 [08:04<1:09:51,  1.38it/s]
2|705|Loss: 1.0060067176818848:  11%|█         | 705/6470 [08:04<1:03:26,  1.51it/s]
2|706|Loss: 0.8818440437316895:  11%|█         | 705/6470 [08:04<1:03:26,  1.51it/s]
2|706|Loss: 0.8818440437316895:  11%|█         | 706/6470 [08:05<1:01:36,  1.56it/s]
2|707|Loss: 0.9527167081832886:  11%|█         | 706/6470 [08:05<1:01:36,  1.56it/s]
2|707|Loss: 0.9527167081832886:  11%|█         | 707/6470 [08:05<1:01:47,  1.55it/s]
2|708|Loss: 0.7853447198867798:  11%|█         | 707/6470 [08:06<1:01:47,  1.55it/s]
2|708|Loss: 0.7853447198867798:  11%|█         | 708/6470 [08:06<1:06:41,  1.44it/s]
2|709|Loss: 0.9680538773536682:  11%|█         | 708/6470 [08:07<1:06:41,  1.44it/s]
2|709|Loss: 0.9680538773536682:  11%|█         | 709/6470 [08:07<1:10:40,  1.36it/s]
2|710|Loss: 0.8027076125144958:  11%|█         | 709/6470 [08:07<1:10:40,  1.36it/s]
2|710|Loss: 0.8027076125144958:  11%|█         | 710/6470 [08:08<1:10:33,  1.36it/s]
2|711|Loss: 0.8048970103263855:  11%|█         | 710/6470 [08:08<1:10:33,  1.36it/s]
2|711|Loss: 0.8048970103263855:  11%|█         | 711/6470 [08:09<1:12:38,  1.32it/s]
2|712|Loss: 0.8647525310516357:  11%|█         | 711/6470 [08:09<1:12:38,  1.32it/s]
2|712|Loss: 0.8647525310516357:  11%|█         | 712/6470 [08:09<1:12:43,  1.32it/s]
2|713|Loss: 0.9885768294334412:  11%|█         | 712/6470 [08:10<1:12:43,  1.32it/s]
2|713|Loss: 0.9885768294334412:  11%|█         | 713/6470 [08:10<1:13:22,  1.31it/s]
2|714|Loss: 0.8573490381240845:  11%|█         | 713/6470 [08:11<1:13:22,  1.31it/s]
2|714|Loss: 0.8573490381240845:  11%|█         | 714/6470 [08:11<1:12:15,  1.33it/s]
2|715|Loss: 0.8231818079948425:  11%|█         | 714/6470 [08:11<1:12:15,  1.33it/s]
2|715|Loss: 0.8231818079948425:  11%|█         | 715/6470 [08:12<1:16:50,  1.25it/s]
2|716|Loss: 0.8933236002922058:  11%|█         | 715/6470 [08:12<1:16:50,  1.25it/s]
2|716|Loss: 0.8933236002922058:  11%|█         | 716/6470 [08:12<1:09:56,  1.37it/s]
2|717|Loss: 0.8798069357872009:  11%|█         | 716/6470 [08:13<1:09:56,  1.37it/s]
2|717|Loss: 0.8798069357872009:  11%|█         | 717/6470 [08:13<1:06:15,  1.45it/s]
2|718|Loss: 0.6439135670661926:  11%|█         | 717/6470 [08:13<1:06:15,  1.45it/s]
2|718|Loss: 0.6439135670661926:  11%|█         | 718/6470 [08:14<1:12:27,  1.32it/s]
2|719|Loss: 0.8128381967544556:  11%|█         | 718/6470 [08:14<1:12:27,  1.32it/s]
2|719|Loss: 0.8128381967544556:  11%|█         | 719/6470 [08:15<1:13:18,  1.31it/s]
2|720|Loss: 0.7570919394493103:  11%|█         | 719/6470 [08:15<1:13:18,  1.31it/s]
2|720|Loss: 0.7570919394493103:  11%|█         | 720/6470 [08:15<1:12:29,  1.32it/s]
2|721|Loss: 0.78367680311203:  11%|█         | 720/6470 [08:16<1:12:29,  1.32it/s]  
2|721|Loss: 0.78367680311203:  11%|█         | 721/6470 [08:16<1:08:20,  1.40it/s]
2|722|Loss: 0.841757595539093:  11%|█         | 721/6470 [08:16<1:08:20,  1.40it/s]
2|722|Loss: 0.841757595539093:  11%|█         | 722/6470 [08:17<1:11:22,  1.34it/s]
2|723|Loss: 0.8303771615028381:  11%|█         | 722/6470 [08:17<1:11:22,  1.34it/s]
2|723|Loss: 0.8303771615028381:  11%|█         | 723/6470 [08:18<1:11:57,  1.33it/s]
2|724|Loss: 0.7896684408187866:  11%|█         | 723/6470 [08:18<1:11:57,  1.33it/s]
2|724|Loss: 0.7896684408187866:  11%|█         | 724/6470 [08:18<1:10:41,  1.35it/s]
2|725|Loss: 0.6836223006248474:  11%|█         | 724/6470 [08:19<1:10:41,  1.35it/s]
2|725|Loss: 0.6836223006248474:  11%|█         | 725/6470 [08:19<1:15:36,  1.27it/s]
2|726|Loss: 0.8187410235404968:  11%|█         | 725/6470 [08:20<1:15:36,  1.27it/s]
2|726|Loss: 0.8187410235404968:  11%|█         | 726/6470 [08:20<1:18:46,  1.22it/s]
2|727|Loss: 0.83058100938797:  11%|█         | 726/6470 [08:21<1:18:46,  1.22it/s]  
2|727|Loss: 0.83058100938797:  11%|█         | 727/6470 [08:21<1:16:45,  1.25it/s]
2|728|Loss: 0.715625524520874:  11%|█         | 727/6470 [08:21<1:16:45,  1.25it/s]
2|728|Loss: 0.715625524520874:  11%|█▏        | 728/6470 [08:22<1:14:30,  1.28it/s]
2|729|Loss: 0.8736077547073364:  11%|█▏        | 728/6470 [08:22<1:14:30,  1.28it/s]
2|729|Loss: 0.8736077547073364:  11%|█▏        | 729/6470 [08:22<1:10:40,  1.35it/s]
2|730|Loss: 0.9321252703666687:  11%|█▏        | 729/6470 [08:23<1:10:40,  1.35it/s]
2|730|Loss: 0.9321252703666687:  11%|█▏        | 730/6470 [08:23<1:12:41,  1.32it/s]
2|731|Loss: 0.9432487487792969:  11%|█▏        | 730/6470 [08:23<1:12:41,  1.32it/s]
2|731|Loss: 0.9432487487792969:  11%|█▏        | 731/6470 [08:24<1:12:48,  1.31it/s]
2|732|Loss: 0.8052200078964233:  11%|█▏        | 731/6470 [08:24<1:12:48,  1.31it/s]
2|732|Loss: 0.8052200078964233:  11%|█▏        | 732/6470 [08:24<1:09:29,  1.38it/s]
2|733|Loss: 0.8058110475540161:  11%|█▏        | 732/6470 [08:25<1:09:29,  1.38it/s]
2|733|Loss: 0.8058110475540161:  11%|█▏        | 733/6470 [08:25<1:05:32,  1.46it/s]
2|734|Loss: 0.8327360153198242:  11%|█▏        | 733/6470 [08:25<1:05:32,  1.46it/s]
2|734|Loss: 0.8327360153198242:  11%|█▏        | 734/6470 [08:26<1:05:17,  1.46it/s]
2|735|Loss: 0.7838608026504517:  11%|█▏        | 734/6470 [08:26<1:05:17,  1.46it/s]
2|735|Loss: 0.7838608026504517:  11%|█▏        | 735/6470 [08:26<1:04:23,  1.48it/s]
2|736|Loss: 0.9248749613761902:  11%|█▏        | 735/6470 [08:27<1:04:23,  1.48it/s]
2|736|Loss: 0.9248749613761902:  11%|█▏        | 736/6470 [08:27<1:00:25,  1.58it/s]
2|737|Loss: 0.9527706503868103:  11%|█▏        | 736/6470 [08:27<1:00:25,  1.58it/s]
2|737|Loss: 0.9527706503868103:  11%|█▏        | 737/6470 [08:28<1:01:56,  1.54it/s]
2|738|Loss: 0.8523388504981995:  11%|█▏        | 737/6470 [08:28<1:01:56,  1.54it/s]
2|738|Loss: 0.8523388504981995:  11%|█▏        | 738/6470 [08:28<1:01:52,  1.54it/s]
2|739|Loss: 0.888093113899231:  11%|█▏        | 738/6470 [08:29<1:01:52,  1.54it/s] 
2|739|Loss: 0.888093113899231:  11%|█▏        | 739/6470 [08:29<1:05:52,  1.45it/s]
2|740|Loss: 0.8338473439216614:  11%|█▏        | 739/6470 [08:29<1:05:52,  1.45it/s]
2|740|Loss: 0.8338473439216614:  11%|█▏        | 740/6470 [08:30<1:03:57,  1.49it/s]
2|741|Loss: 0.8801956176757812:  11%|█▏        | 740/6470 [08:30<1:03:57,  1.49it/s]
2|741|Loss: 0.8801956176757812:  11%|█▏        | 741/6470 [08:30<1:06:28,  1.44it/s]
2|742|Loss: 0.8399028778076172:  11%|█▏        | 741/6470 [08:31<1:06:28,  1.44it/s]
2|742|Loss: 0.8399028778076172:  11%|█▏        | 742/6470 [08:31<1:05:23,  1.46it/s]
2|743|Loss: 0.926128089427948:  11%|█▏        | 742/6470 [08:31<1:05:23,  1.46it/s] 
2|743|Loss: 0.926128089427948:  11%|█▏        | 743/6470 [08:32<1:05:25,  1.46it/s]
2|744|Loss: 0.9757445454597473:  11%|█▏        | 743/6470 [08:32<1:05:25,  1.46it/s]
2|744|Loss: 0.9757445454597473:  11%|█▏        | 744/6470 [08:32<1:04:29,  1.48it/s]
2|745|Loss: 0.85368812084198:  11%|█▏        | 744/6470 [08:33<1:04:29,  1.48it/s]  
2|745|Loss: 0.85368812084198:  12%|█▏        | 745/6470 [08:33<1:07:02,  1.42it/s]
2|746|Loss: 0.8722054958343506:  12%|█▏        | 745/6470 [08:33<1:07:02,  1.42it/s]
2|746|Loss: 0.8722054958343506:  12%|█▏        | 746/6470 [08:34<1:01:48,  1.54it/s]
2|747|Loss: 0.9475566148757935:  12%|█▏        | 746/6470 [08:34<1:01:48,  1.54it/s]
2|747|Loss: 0.9475566148757935:  12%|█▏        | 747/6470 [08:34<1:00:07,  1.59it/s]
2|748|Loss: 0.8584890365600586:  12%|█▏        | 747/6470 [08:35<1:00:07,  1.59it/s]
2|748|Loss: 0.8584890365600586:  12%|█▏        | 748/6470 [08:35<1:02:29,  1.53it/s]
2|749|Loss: 0.698917806148529:  12%|█▏        | 748/6470 [08:35<1:02:29,  1.53it/s] 
2|749|Loss: 0.698917806148529:  12%|█▏        | 749/6470 [08:36<59:39,  1.60it/s]  
2|750|Loss: 0.8328468203544617:  12%|█▏        | 749/6470 [08:36<59:39,  1.60it/s]
2|750|Loss: 0.8328468203544617:  12%|█▏        | 750/6470 [08:36<57:45,  1.65it/s]
2|751|Loss: 0.8259813785552979:  12%|█▏        | 750/6470 [08:36<57:45,  1.65it/s]
2|751|Loss: 0.8259813785552979:  12%|█▏        | 751/6470 [08:37<58:25,  1.63it/s]
2|752|Loss: 0.7560524940490723:  12%|█▏        | 751/6470 [08:37<58:25,  1.63it/s]
2|752|Loss: 0.7560524940490723:  12%|█▏        | 752/6470 [08:37<59:25,  1.60it/s]
2|753|Loss: 0.8995321989059448:  12%|█▏        | 752/6470 [08:38<59:25,  1.60it/s]
2|753|Loss: 0.8995321989059448:  12%|█▏        | 753/6470 [08:38<1:03:45,  1.49it/s]
2|754|Loss: 0.7910227179527283:  12%|█▏        | 753/6470 [08:39<1:03:45,  1.49it/s]
2|754|Loss: 0.7910227179527283:  12%|█▏        | 754/6470 [08:39<1:06:31,  1.43it/s]
2|755|Loss: 0.9076882004737854:  12%|█▏        | 754/6470 [08:39<1:06:31,  1.43it/s]
2|755|Loss: 0.9076882004737854:  12%|█▏        | 755/6470 [08:40<1:10:33,  1.35it/s]
2|756|Loss: 1.0609370470046997:  12%|█▏        | 755/6470 [08:40<1:10:33,  1.35it/s]
2|756|Loss: 1.0609370470046997:  12%|█▏        | 756/6470 [08:40<1:09:02,  1.38it/s]
2|757|Loss: 0.9123841524124146:  12%|█▏        | 756/6470 [08:41<1:09:02,  1.38it/s]
2|757|Loss: 0.9123841524124146:  12%|█▏        | 757/6470 [08:41<1:06:03,  1.44it/s]
2|758|Loss: 0.9021294116973877:  12%|█▏        | 757/6470 [08:41<1:06:03,  1.44it/s]
2|758|Loss: 0.9021294116973877:  12%|█▏        | 758/6470 [08:42<1:06:51,  1.42it/s]
2|759|Loss: 0.8000668883323669:  12%|█▏        | 758/6470 [08:42<1:06:51,  1.42it/s]
2|759|Loss: 0.8000668883323669:  12%|█▏        | 759/6470 [08:43<1:11:31,  1.33it/s]
2|760|Loss: 0.9543644189834595:  12%|█▏        | 759/6470 [08:43<1:11:31,  1.33it/s]
2|760|Loss: 0.9543644189834595:  12%|█▏        | 760/6470 [08:43<1:12:02,  1.32it/s]
2|761|Loss: 0.9944182634353638:  12%|█▏        | 760/6470 [08:44<1:12:02,  1.32it/s]
2|761|Loss: 0.9944182634353638:  12%|█▏        | 761/6470 [08:44<1:06:04,  1.44it/s]
2|762|Loss: 0.9152994155883789:  12%|█▏        | 761/6470 [08:44<1:06:04,  1.44it/s]
2|762|Loss: 0.9152994155883789:  12%|█▏        | 762/6470 [08:45<1:02:23,  1.52it/s]
2|763|Loss: 0.8096799254417419:  12%|█▏        | 762/6470 [08:45<1:02:23,  1.52it/s]
2|763|Loss: 0.8096799254417419:  12%|█▏        | 763/6470 [08:45<1:01:38,  1.54it/s]
2|764|Loss: 0.8307467103004456:  12%|█▏        | 763/6470 [08:46<1:01:38,  1.54it/s]
2|764|Loss: 0.8307467103004456:  12%|█▏        | 764/6470 [08:46<1:07:49,  1.40it/s]
2|765|Loss: 1.0624483823776245:  12%|█▏        | 764/6470 [08:46<1:07:49,  1.40it/s]
2|765|Loss: 1.0624483823776245:  12%|█▏        | 765/6470 [08:47<1:07:06,  1.42it/s]
2|766|Loss: 0.7092912197113037:  12%|█▏        | 765/6470 [08:47<1:07:06,  1.42it/s]
2|766|Loss: 0.7092912197113037:  12%|█▏        | 766/6470 [08:48<1:12:20,  1.31it/s]
2|767|Loss: 0.7858254313468933:  12%|█▏        | 766/6470 [08:48<1:12:20,  1.31it/s]
2|767|Loss: 0.7858254313468933:  12%|█▏        | 767/6470 [08:48<1:13:19,  1.30it/s]
2|768|Loss: 0.8266905546188354:  12%|█▏        | 767/6470 [08:49<1:13:19,  1.30it/s]
2|768|Loss: 0.8266905546188354:  12%|█▏        | 768/6470 [08:49<1:14:52,  1.27it/s]
2|769|Loss: 0.7471016049385071:  12%|█▏        | 768/6470 [08:50<1:14:52,  1.27it/s]
2|769|Loss: 0.7471016049385071:  12%|█▏        | 769/6470 [08:50<1:07:28,  1.41it/s]
2|770|Loss: 1.0014735460281372:  12%|█▏        | 769/6470 [08:50<1:07:28,  1.41it/s]
2|770|Loss: 1.0014735460281372:  12%|█▏        | 770/6470 [08:50<1:06:30,  1.43it/s]
2|771|Loss: 0.9234108924865723:  12%|█▏        | 770/6470 [08:51<1:06:30,  1.43it/s]
2|771|Loss: 0.9234108924865723:  12%|█▏        | 771/6470 [08:51<1:06:58,  1.42it/s]
2|772|Loss: 0.7992529273033142:  12%|█▏        | 771/6470 [08:51<1:06:58,  1.42it/s]
2|772|Loss: 0.7992529273033142:  12%|█▏        | 772/6470 [08:52<1:03:20,  1.50it/s]
2|773|Loss: 0.9356065988540649:  12%|█▏        | 772/6470 [08:52<1:03:20,  1.50it/s]
2|773|Loss: 0.9356065988540649:  12%|█▏        | 773/6470 [08:52<1:04:17,  1.48it/s]
2|774|Loss: 0.8452093005180359:  12%|█▏        | 773/6470 [08:53<1:04:17,  1.48it/s]
2|774|Loss: 0.8452093005180359:  12%|█▏        | 774/6470 [08:53<1:03:51,  1.49it/s]
2|775|Loss: 0.9431217908859253:  12%|█▏        | 774/6470 [08:54<1:03:51,  1.49it/s]
2|775|Loss: 0.9431217908859253:  12%|█▏        | 775/6470 [08:54<1:08:08,  1.39it/s]
2|776|Loss: 0.957500159740448:  12%|█▏        | 775/6470 [08:54<1:08:08,  1.39it/s] 
2|776|Loss: 0.957500159740448:  12%|█▏        | 776/6470 [08:55<1:09:28,  1.37it/s]
2|777|Loss: 0.7936472296714783:  12%|█▏        | 776/6470 [08:55<1:09:28,  1.37it/s]
2|777|Loss: 0.7936472296714783:  12%|█▏        | 777/6470 [08:55<1:11:13,  1.33it/s]
2|778|Loss: 0.8079007863998413:  12%|█▏        | 777/6470 [08:56<1:11:13,  1.33it/s]
2|778|Loss: 0.8079007863998413:  12%|█▏        | 778/6470 [08:56<1:09:00,  1.37it/s]
2|779|Loss: 0.8751268982887268:  12%|█▏        | 778/6470 [08:57<1:09:00,  1.37it/s]
2|779|Loss: 0.8751268982887268:  12%|█▏        | 779/6470 [08:57<1:08:22,  1.39it/s]
2|780|Loss: 0.9613245129585266:  12%|█▏        | 779/6470 [08:57<1:08:22,  1.39it/s]
2|780|Loss: 0.9613245129585266:  12%|█▏        | 780/6470 [08:57<1:04:56,  1.46it/s]
2|781|Loss: 0.879015326499939:  12%|█▏        | 780/6470 [08:58<1:04:56,  1.46it/s] 
2|781|Loss: 0.879015326499939:  12%|█▏        | 781/6470 [08:58<1:05:41,  1.44it/s]
2|782|Loss: 0.9524204730987549:  12%|█▏        | 781/6470 [08:59<1:05:41,  1.44it/s]
2|782|Loss: 0.9524204730987549:  12%|█▏        | 782/6470 [08:59<1:06:59,  1.42it/s]
2|783|Loss: 1.0130113363265991:  12%|█▏        | 782/6470 [08:59<1:06:59,  1.42it/s]
2|783|Loss: 1.0130113363265991:  12%|█▏        | 783/6470 [09:00<1:07:56,  1.40it/s]
2|784|Loss: 0.8667694926261902:  12%|█▏        | 783/6470 [09:00<1:07:56,  1.40it/s]
2|784|Loss: 0.8667694926261902:  12%|█▏        | 784/6470 [09:00<1:08:50,  1.38it/s]
2|785|Loss: 1.0859098434448242:  12%|█▏        | 784/6470 [09:01<1:08:50,  1.38it/s]
2|785|Loss: 1.0859098434448242:  12%|█▏        | 785/6470 [09:01<1:09:36,  1.36it/s]
2|786|Loss: 0.8714697957038879:  12%|█▏        | 785/6470 [09:02<1:09:36,  1.36it/s]
2|786|Loss: 0.8714697957038879:  12%|█▏        | 786/6470 [09:02<1:09:37,  1.36it/s]
2|787|Loss: 0.8290112614631653:  12%|█▏        | 786/6470 [09:02<1:09:37,  1.36it/s]
2|787|Loss: 0.8290112614631653:  12%|█▏        | 787/6470 [09:03<1:09:36,  1.36it/s]
2|788|Loss: 0.8566744327545166:  12%|█▏        | 787/6470 [09:03<1:09:36,  1.36it/s]
2|788|Loss: 0.8566744327545166:  12%|█▏        | 788/6470 [09:03<1:11:40,  1.32it/s]
2|789|Loss: 1.0063380002975464:  12%|█▏        | 788/6470 [09:04<1:11:40,  1.32it/s]
2|789|Loss: 1.0063380002975464:  12%|█▏        | 789/6470 [09:04<1:11:07,  1.33it/s]
2|790|Loss: 0.8844758868217468:  12%|█▏        | 789/6470 [09:05<1:11:07,  1.33it/s]
2|790|Loss: 0.8844758868217468:  12%|█▏        | 790/6470 [09:05<1:07:46,  1.40it/s]
2|791|Loss: 0.9719710946083069:  12%|█▏        | 790/6470 [09:05<1:07:46,  1.40it/s]
2|791|Loss: 0.9719710946083069:  12%|█▏        | 791/6470 [09:05<1:05:13,  1.45it/s]
2|792|Loss: 0.7167996168136597:  12%|█▏        | 791/6470 [09:06<1:05:13,  1.45it/s]
2|792|Loss: 0.7167996168136597:  12%|█▏        | 792/6470 [09:06<1:08:02,  1.39it/s]
2|793|Loss: 0.9232300519943237:  12%|█▏        | 792/6470 [09:07<1:08:02,  1.39it/s]
2|793|Loss: 0.9232300519943237:  12%|█▏        | 793/6470 [09:07<1:09:17,  1.37it/s]
2|794|Loss: 0.9150962233543396:  12%|█▏        | 793/6470 [09:07<1:09:17,  1.37it/s]
2|794|Loss: 0.9150962233543396:  12%|█▏        | 794/6470 [09:08<1:07:18,  1.41it/s]
2|795|Loss: 0.9807968139648438:  12%|█▏        | 794/6470 [09:08<1:07:18,  1.41it/s]
2|795|Loss: 0.9807968139648438:  12%|█▏        | 795/6470 [09:08<1:06:43,  1.42it/s]
2|796|Loss: 0.8571486473083496:  12%|█▏        | 795/6470 [09:09<1:06:43,  1.42it/s]
2|796|Loss: 0.8571486473083496:  12%|█▏        | 796/6470 [09:09<1:07:33,  1.40it/s]
2|797|Loss: 0.8790916204452515:  12%|█▏        | 796/6470 [09:09<1:07:33,  1.40it/s]
2|797|Loss: 0.8790916204452515:  12%|█▏        | 797/6470 [09:10<1:05:47,  1.44it/s]
2|798|Loss: 0.9211403131484985:  12%|█▏        | 797/6470 [09:10<1:05:47,  1.44it/s]
2|798|Loss: 0.9211403131484985:  12%|█▏        | 798/6470 [09:10<1:03:31,  1.49it/s]
2|799|Loss: 0.8930848240852356:  12%|█▏        | 798/6470 [09:11<1:03:31,  1.49it/s]
2|799|Loss: 0.8930848240852356:  12%|█▏        | 799/6470 [09:11<1:05:12,  1.45it/s]
2|800|Loss: 0.7951031923294067:  12%|█▏        | 799/6470 [09:11<1:05:12,  1.45it/s]
2|800|Loss: 0.7951031923294067:  12%|█▏        | 800/6470 [09:12<1:02:36,  1.51it/s]
2|801|Loss: 0.7662863731384277:  12%|█▏        | 800/6470 [09:12<1:02:36,  1.51it/s]
2|801|Loss: 0.7662863731384277:  12%|█▏        | 801/6470 [09:12<1:04:06,  1.47it/s]
2|802|Loss: 0.8865708112716675:  12%|█▏        | 801/6470 [09:13<1:04:06,  1.47it/s]
2|802|Loss: 0.8865708112716675:  12%|█▏        | 802/6470 [09:13<1:05:52,  1.43it/s]
2|803|Loss: 0.8872359991073608:  12%|█▏        | 802/6470 [09:13<1:05:52,  1.43it/s]
2|803|Loss: 0.8872359991073608:  12%|█▏        | 803/6470 [09:14<1:02:03,  1.52it/s]
2|804|Loss: 0.8026617765426636:  12%|█▏        | 803/6470 [09:14<1:02:03,  1.52it/s]
2|804|Loss: 0.8026617765426636:  12%|█▏        | 804/6470 [09:14<1:01:04,  1.55it/s]
2|805|Loss: 0.8077698945999146:  12%|█▏        | 804/6470 [09:15<1:01:04,  1.55it/s]
2|805|Loss: 0.8077698945999146:  12%|█▏        | 805/6470 [09:15<1:05:18,  1.45it/s]
2|806|Loss: 0.9381405711174011:  12%|█▏        | 805/6470 [09:16<1:05:18,  1.45it/s]
2|806|Loss: 0.9381405711174011:  12%|█▏        | 806/6470 [09:16<1:08:23,  1.38it/s]
2|807|Loss: 0.6316119432449341:  12%|█▏        | 806/6470 [09:16<1:08:23,  1.38it/s]
2|807|Loss: 0.6316119432449341:  12%|█▏        | 807/6470 [09:17<1:07:01,  1.41it/s]
2|808|Loss: 0.7717931866645813:  12%|█▏        | 807/6470 [09:17<1:07:01,  1.41it/s]
2|808|Loss: 0.7717931866645813:  12%|█▏        | 808/6470 [09:17<1:09:21,  1.36it/s]
2|809|Loss: 0.8777455687522888:  12%|█▏        | 808/6470 [09:18<1:09:21,  1.36it/s]
2|809|Loss: 0.8777455687522888:  13%|█▎        | 809/6470 [09:18<1:04:46,  1.46it/s]
2|810|Loss: 0.7201047539710999:  13%|█▎        | 809/6470 [09:18<1:04:46,  1.46it/s]
2|810|Loss: 0.7201047539710999:  13%|█▎        | 810/6470 [09:19<1:05:06,  1.45it/s]
2|811|Loss: 0.811261773109436:  13%|█▎        | 810/6470 [09:19<1:05:06,  1.45it/s] 
2|811|Loss: 0.811261773109436:  13%|█▎        | 811/6470 [09:19<1:06:57,  1.41it/s]
2|812|Loss: 0.8842806220054626:  13%|█▎        | 811/6470 [09:20<1:06:57,  1.41it/s]
2|812|Loss: 0.8842806220054626:  13%|█▎        | 812/6470 [09:20<1:09:36,  1.35it/s]
2|813|Loss: 0.934917151927948:  13%|█▎        | 812/6470 [09:21<1:09:36,  1.35it/s] 
2|813|Loss: 0.934917151927948:  13%|█▎        | 813/6470 [09:21<1:06:18,  1.42it/s]
2|814|Loss: 0.9188082218170166:  13%|█▎        | 813/6470 [09:21<1:06:18,  1.42it/s]
2|814|Loss: 0.9188082218170166:  13%|█▎        | 814/6470 [09:21<1:05:47,  1.43it/s]
2|815|Loss: 0.8946670293807983:  13%|█▎        | 814/6470 [09:22<1:05:47,  1.43it/s]
2|815|Loss: 0.8946670293807983:  13%|█▎        | 815/6470 [09:22<1:06:05,  1.43it/s]
2|816|Loss: 0.8493542075157166:  13%|█▎        | 815/6470 [09:23<1:06:05,  1.43it/s]
2|816|Loss: 0.8493542075157166:  13%|█▎        | 816/6470 [09:23<1:02:37,  1.50it/s]
2|817|Loss: 0.858783483505249:  13%|█▎        | 816/6470 [09:23<1:02:37,  1.50it/s] 
2|817|Loss: 0.858783483505249:  13%|█▎        | 817/6470 [09:24<1:06:14,  1.42it/s]
2|818|Loss: 0.9633691906929016:  13%|█▎        | 817/6470 [09:24<1:06:14,  1.42it/s]
2|818|Loss: 0.9633691906929016:  13%|█▎        | 818/6470 [09:24<1:07:08,  1.40it/s]
2|819|Loss: 0.7630679607391357:  13%|█▎        | 818/6470 [09:25<1:07:08,  1.40it/s]
2|819|Loss: 0.7630679607391357:  13%|█▎        | 819/6470 [09:25<1:09:17,  1.36it/s]
2|820|Loss: 0.786292552947998:  13%|█▎        | 819/6470 [09:25<1:09:17,  1.36it/s] 
2|820|Loss: 0.786292552947998:  13%|█▎        | 820/6470 [09:26<1:04:33,  1.46it/s]
2|821|Loss: 0.8340339660644531:  13%|█▎        | 820/6470 [09:26<1:04:33,  1.46it/s]
2|821|Loss: 0.8340339660644531:  13%|█▎        | 821/6470 [09:26<1:01:08,  1.54it/s]
2|822|Loss: 0.7516632080078125:  13%|█▎        | 821/6470 [09:27<1:01:08,  1.54it/s]
2|822|Loss: 0.7516632080078125:  13%|█▎        | 822/6470 [09:27<1:00:28,  1.56it/s]
2|823|Loss: 0.8293111324310303:  13%|█▎        | 822/6470 [09:27<1:00:28,  1.56it/s]
2|823|Loss: 0.8293111324310303:  13%|█▎        | 823/6470 [09:28<1:02:58,  1.49it/s]
2|824|Loss: 0.987643301486969:  13%|█▎        | 823/6470 [09:28<1:02:58,  1.49it/s] 
2|824|Loss: 0.987643301486969:  13%|█▎        | 824/6470 [09:28<1:02:42,  1.50it/s]
2|825|Loss: 0.8532871007919312:  13%|█▎        | 824/6470 [09:29<1:02:42,  1.50it/s]
2|825|Loss: 0.8532871007919312:  13%|█▎        | 825/6470 [09:29<59:49,  1.57it/s]  
2|826|Loss: 0.8759733438491821:  13%|█▎        | 825/6470 [09:29<59:49,  1.57it/s]
2|826|Loss: 0.8759733438491821:  13%|█▎        | 826/6470 [09:29<59:30,  1.58it/s]
2|827|Loss: 1.0168360471725464:  13%|█▎        | 826/6470 [09:30<59:30,  1.58it/s]
2|827|Loss: 1.0168360471725464:  13%|█▎        | 827/6470 [09:30<59:45,  1.57it/s]
2|828|Loss: 0.8835611939430237:  13%|█▎        | 827/6470 [09:30<59:45,  1.57it/s]
2|828|Loss: 0.8835611939430237:  13%|█▎        | 828/6470 [09:31<58:16,  1.61it/s]
2|829|Loss: 0.8082325458526611:  13%|█▎        | 828/6470 [09:31<58:16,  1.61it/s]
2|829|Loss: 0.8082325458526611:  13%|█▎        | 829/6470 [09:31<59:55,  1.57it/s]
2|830|Loss: 0.8637561798095703:  13%|█▎        | 829/6470 [09:32<59:55,  1.57it/s]
2|830|Loss: 0.8637561798095703:  13%|█▎        | 830/6470 [09:32<1:00:20,  1.56it/s]
2|831|Loss: 0.8051835298538208:  13%|█▎        | 830/6470 [09:32<1:00:20,  1.56it/s]
2|831|Loss: 0.8051835298538208:  13%|█▎        | 831/6470 [09:33<1:01:29,  1.53it/s]
2|832|Loss: 0.8162288069725037:  13%|█▎        | 831/6470 [09:33<1:01:29,  1.53it/s]
2|832|Loss: 0.8162288069725037:  13%|█▎        | 832/6470 [09:33<1:03:25,  1.48it/s]
2|833|Loss: 0.8166840672492981:  13%|█▎        | 832/6470 [09:34<1:03:25,  1.48it/s]
2|833|Loss: 0.8166840672492981:  13%|█▎        | 833/6470 [09:34<1:05:09,  1.44it/s]
2|834|Loss: 0.8722143769264221:  13%|█▎        | 833/6470 [09:35<1:05:09,  1.44it/s]
2|834|Loss: 0.8722143769264221:  13%|█▎        | 834/6470 [09:35<1:04:17,  1.46it/s]
2|835|Loss: 0.8922282457351685:  13%|█▎        | 834/6470 [09:35<1:04:17,  1.46it/s]
2|835|Loss: 0.8922282457351685:  13%|█▎        | 835/6470 [09:35<1:02:22,  1.51it/s]
2|836|Loss: 0.795166015625:  13%|█▎        | 835/6470 [09:36<1:02:22,  1.51it/s]    
2|836|Loss: 0.795166015625:  13%|█▎        | 836/6470 [09:36<1:01:23,  1.53it/s]
2|837|Loss: 0.7530842423439026:  13%|█▎        | 836/6470 [09:36<1:01:23,  1.53it/s]
2|837|Loss: 0.7530842423439026:  13%|█▎        | 837/6470 [09:37<1:00:56,  1.54it/s]
2|838|Loss: 0.8549378514289856:  13%|█▎        | 837/6470 [09:37<1:00:56,  1.54it/s]
2|838|Loss: 0.8549378514289856:  13%|█▎        | 838/6470 [09:37<1:02:05,  1.51it/s]
2|839|Loss: 0.9212294816970825:  13%|█▎        | 838/6470 [09:38<1:02:05,  1.51it/s]
2|839|Loss: 0.9212294816970825:  13%|█▎        | 839/6470 [09:38<1:03:25,  1.48it/s]
2|840|Loss: 0.8427738547325134:  13%|█▎        | 839/6470 [09:38<1:03:25,  1.48it/s]
2|840|Loss: 0.8427738547325134:  13%|█▎        | 840/6470 [09:39<1:02:47,  1.49it/s]
2|841|Loss: 0.9798628091812134:  13%|█▎        | 840/6470 [09:39<1:02:47,  1.49it/s]
2|841|Loss: 0.9798628091812134:  13%|█▎        | 841/6470 [09:39<57:38,  1.63it/s]  
2|842|Loss: 0.9519019722938538:  13%|█▎        | 841/6470 [09:40<57:38,  1.63it/s]
2|842|Loss: 0.9519019722938538:  13%|█▎        | 842/6470 [09:40<59:10,  1.59it/s]
2|843|Loss: 0.9287465810775757:  13%|█▎        | 842/6470 [09:40<59:10,  1.59it/s]
2|843|Loss: 0.9287465810775757:  13%|█▎        | 843/6470 [09:41<1:02:32,  1.50it/s]
2|844|Loss: 0.8127687573432922:  13%|█▎        | 843/6470 [09:41<1:02:32,  1.50it/s]
2|844|Loss: 0.8127687573432922:  13%|█▎        | 844/6470 [09:41<1:04:55,  1.44it/s]
2|845|Loss: 0.8278111815452576:  13%|█▎        | 844/6470 [09:42<1:04:55,  1.44it/s]
2|845|Loss: 0.8278111815452576:  13%|█▎        | 845/6470 [09:42<1:04:07,  1.46it/s]
2|846|Loss: 0.9018900394439697:  13%|█▎        | 845/6470 [09:42<1:04:07,  1.46it/s]
2|846|Loss: 0.9018900394439697:  13%|█▎        | 846/6470 [09:43<1:05:15,  1.44it/s]
2|847|Loss: 0.8210685849189758:  13%|█▎        | 846/6470 [09:43<1:05:15,  1.44it/s]
2|847|Loss: 0.8210685849189758:  13%|█▎        | 847/6470 [09:44<1:08:30,  1.37it/s]
2|848|Loss: 0.705458402633667:  13%|█▎        | 847/6470 [09:44<1:08:30,  1.37it/s] 
2|848|Loss: 0.705458402633667:  13%|█▎        | 848/6470 [09:44<1:08:29,  1.37it/s]
2|849|Loss: 0.8883824944496155:  13%|█▎        | 848/6470 [09:45<1:08:29,  1.37it/s]
2|849|Loss: 0.8883824944496155:  13%|█▎        | 849/6470 [09:45<1:04:50,  1.44it/s]
2|850|Loss: 0.7220157384872437:  13%|█▎        | 849/6470 [09:45<1:04:50,  1.44it/s]
2|850|Loss: 0.7220157384872437:  13%|█▎        | 850/6470 [09:46<1:05:02,  1.44it/s]
2|851|Loss: 1.0349291563034058:  13%|█▎        | 850/6470 [09:46<1:05:02,  1.44it/s]
2|851|Loss: 1.0349291563034058:  13%|█▎        | 851/6470 [09:46<1:05:25,  1.43it/s]
2|852|Loss: 0.959479570388794:  13%|█▎        | 851/6470 [09:47<1:05:25,  1.43it/s] 
2|852|Loss: 0.959479570388794:  13%|█▎        | 852/6470 [09:47<1:03:29,  1.47it/s]
2|853|Loss: 0.8891118764877319:  13%|█▎        | 852/6470 [09:47<1:03:29,  1.47it/s]
2|853|Loss: 0.8891118764877319:  13%|█▎        | 853/6470 [09:48<59:39,  1.57it/s]  
2|854|Loss: 0.8326622247695923:  13%|█▎        | 853/6470 [09:48<59:39,  1.57it/s]
2|854|Loss: 0.8326622247695923:  13%|█▎        | 854/6470 [09:48<1:01:01,  1.53it/s]
2|855|Loss: 0.8887545466423035:  13%|█▎        | 854/6470 [09:49<1:01:01,  1.53it/s]
2|855|Loss: 0.8887545466423035:  13%|█▎        | 855/6470 [09:49<1:02:29,  1.50it/s]
2|856|Loss: 0.8245172500610352:  13%|█▎        | 855/6470 [09:49<1:02:29,  1.50it/s]
2|856|Loss: 0.8245172500610352:  13%|█▎        | 856/6470 [09:50<1:04:11,  1.46it/s]
2|857|Loss: 0.884157657623291:  13%|█▎        | 856/6470 [09:50<1:04:11,  1.46it/s] 
2|857|Loss: 0.884157657623291:  13%|█▎        | 857/6470 [09:50<1:06:46,  1.40it/s]
2|858|Loss: 0.9143074154853821:  13%|█▎        | 857/6470 [09:51<1:06:46,  1.40it/s]
2|858|Loss: 0.9143074154853821:  13%|█▎        | 858/6470 [09:51<1:05:48,  1.42it/s]
2|859|Loss: 0.8967308402061462:  13%|█▎        | 858/6470 [09:52<1:05:48,  1.42it/s]
2|859|Loss: 0.8967308402061462:  13%|█▎        | 859/6470 [09:52<1:08:27,  1.37it/s]
2|860|Loss: 0.9645253419876099:  13%|█▎        | 859/6470 [09:52<1:08:27,  1.37it/s]
2|860|Loss: 0.9645253419876099:  13%|█▎        | 860/6470 [09:53<1:11:31,  1.31it/s]
2|861|Loss: 0.9024553298950195:  13%|█▎        | 860/6470 [09:53<1:11:31,  1.31it/s]
2|861|Loss: 0.9024553298950195:  13%|█▎        | 861/6470 [09:53<1:08:01,  1.37it/s]
2|862|Loss: 0.8623697757720947:  13%|█▎        | 861/6470 [09:54<1:08:01,  1.37it/s]
2|862|Loss: 0.8623697757720947:  13%|█▎        | 862/6470 [09:54<1:09:07,  1.35it/s]
2|863|Loss: 0.8397442102432251:  13%|█▎        | 862/6470 [09:55<1:09:07,  1.35it/s]
2|863|Loss: 0.8397442102432251:  13%|█▎        | 863/6470 [09:55<1:09:18,  1.35it/s]
2|864|Loss: 0.961937665939331:  13%|█▎        | 863/6470 [09:55<1:09:18,  1.35it/s] 
2|864|Loss: 0.961937665939331:  13%|█▎        | 864/6470 [09:56<1:10:50,  1.32it/s]
2|865|Loss: 0.7703088521957397:  13%|█▎        | 864/6470 [09:56<1:10:50,  1.32it/s]
2|865|Loss: 0.7703088521957397:  13%|█▎        | 865/6470 [09:56<1:07:51,  1.38it/s]
2|866|Loss: 0.843856692314148:  13%|█▎        | 865/6470 [09:57<1:07:51,  1.38it/s] 
2|866|Loss: 0.843856692314148:  13%|█▎        | 866/6470 [09:57<1:05:04,  1.44it/s]
2|867|Loss: 0.8307019472122192:  13%|█▎        | 866/6470 [09:57<1:05:04,  1.44it/s]
2|867|Loss: 0.8307019472122192:  13%|█▎        | 867/6470 [09:58<1:05:29,  1.43it/s]
2|868|Loss: 0.8831998109817505:  13%|█▎        | 867/6470 [09:58<1:05:29,  1.43it/s]
2|868|Loss: 0.8831998109817505:  13%|█▎        | 868/6470 [09:58<1:06:16,  1.41it/s]
2|869|Loss: 0.704625129699707:  13%|█▎        | 868/6470 [09:59<1:06:16,  1.41it/s] 
2|869|Loss: 0.704625129699707:  13%|█▎        | 869/6470 [09:59<1:01:50,  1.51it/s]
2|870|Loss: 0.917229175567627:  13%|█▎        | 869/6470 [09:59<1:01:50,  1.51it/s]
2|870|Loss: 0.917229175567627:  13%|█▎        | 870/6470 [09:59<58:22,  1.60it/s]  
2|871|Loss: 0.8096999526023865:  13%|█▎        | 870/6470 [10:00<58:22,  1.60it/s]
2|871|Loss: 0.8096999526023865:  13%|█▎        | 871/6470 [10:00<58:14,  1.60it/s]
2|872|Loss: 0.7936490774154663:  13%|█▎        | 871/6470 [10:00<58:14,  1.60it/s]
2|872|Loss: 0.7936490774154663:  13%|█▎        | 872/6470 [10:01<58:12,  1.60it/s]
2|873|Loss: 0.9056464433670044:  13%|█▎        | 872/6470 [10:01<58:12,  1.60it/s]
2|873|Loss: 0.9056464433670044:  13%|█▎        | 873/6470 [10:01<57:47,  1.61it/s]
2|874|Loss: 0.8085260987281799:  13%|█▎        | 873/6470 [10:02<57:47,  1.61it/s]
2|874|Loss: 0.8085260987281799:  14%|█▎        | 874/6470 [10:02<1:00:45,  1.53it/s]
2|875|Loss: 0.8761924505233765:  14%|█▎        | 874/6470 [10:02<1:00:45,  1.53it/s]
2|875|Loss: 0.8761924505233765:  14%|█▎        | 875/6470 [10:03<58:11,  1.60it/s]  
2|876|Loss: 0.9791994094848633:  14%|█▎        | 875/6470 [10:03<58:11,  1.60it/s]
2|876|Loss: 0.9791994094848633:  14%|█▎        | 876/6470 [10:03<55:53,  1.67it/s]
2|877|Loss: 0.627857506275177:  14%|█▎        | 876/6470 [10:03<55:53,  1.67it/s] 
2|877|Loss: 0.627857506275177:  14%|█▎        | 877/6470 [10:04<53:55,  1.73it/s]
2|878|Loss: 0.952180027961731:  14%|█▎        | 877/6470 [10:04<53:55,  1.73it/s]
2|878|Loss: 0.952180027961731:  14%|█▎        | 878/6470 [10:04<58:38,  1.59it/s]
2|879|Loss: 0.8985949158668518:  14%|█▎        | 878/6470 [10:05<58:38,  1.59it/s]
2|879|Loss: 0.8985949158668518:  14%|█▎        | 879/6470 [10:05<1:01:59,  1.50it/s]
2|880|Loss: 0.797538161277771:  14%|█▎        | 879/6470 [10:06<1:01:59,  1.50it/s] 
2|880|Loss: 0.797538161277771:  14%|█▎        | 880/6470 [10:06<55:56,  1.67it/s]  
2|881|Loss: 0.8297938108444214:  14%|█▎        | 880/6470 [10:06<55:56,  1.67it/s]
2|881|Loss: 0.8297938108444214:  14%|█▎        | 881/6470 [10:06<57:53,  1.61it/s]
2|882|Loss: 0.8597018122673035:  14%|█▎        | 881/6470 [10:07<57:53,  1.61it/s]
2|882|Loss: 0.8597018122673035:  14%|█▎        | 882/6470 [10:07<1:01:22,  1.52it/s]
2|883|Loss: 0.7462663054466248:  14%|█▎        | 882/6470 [10:07<1:01:22,  1.52it/s]
2|883|Loss: 0.7462663054466248:  14%|█▎        | 883/6470 [10:08<1:01:35,  1.51it/s]
2|884|Loss: 0.6896169185638428:  14%|█▎        | 883/6470 [10:08<1:01:35,  1.51it/s]
2|884|Loss: 0.6896169185638428:  14%|█▎        | 884/6470 [10:08<1:00:26,  1.54it/s]
2|885|Loss: 0.7204167246818542:  14%|█▎        | 884/6470 [10:09<1:00:26,  1.54it/s]
2|885|Loss: 0.7204167246818542:  14%|█▎        | 885/6470 [10:09<1:02:20,  1.49it/s]
2|886|Loss: 0.8453384041786194:  14%|█▎        | 885/6470 [10:10<1:02:20,  1.49it/s]
2|886|Loss: 0.8453384041786194:  14%|█▎        | 886/6470 [10:10<1:05:33,  1.42it/s]
2|887|Loss: 0.7880270481109619:  14%|█▎        | 886/6470 [10:10<1:05:33,  1.42it/s]
2|887|Loss: 0.7880270481109619:  14%|█▎        | 887/6470 [10:11<1:04:46,  1.44it/s]
2|888|Loss: 0.9022907614707947:  14%|█▎        | 887/6470 [10:11<1:04:46,  1.44it/s]
2|888|Loss: 0.9022907614707947:  14%|█▎        | 888/6470 [10:11<1:05:37,  1.42it/s]
2|889|Loss: 0.9083759188652039:  14%|█▎        | 888/6470 [10:12<1:05:37,  1.42it/s]
2|889|Loss: 0.9083759188652039:  14%|█▎        | 889/6470 [10:12<1:01:32,  1.51it/s]
2|890|Loss: 0.840735137462616:  14%|█▎        | 889/6470 [10:12<1:01:32,  1.51it/s] 
2|890|Loss: 0.840735137462616:  14%|█▍        | 890/6470 [10:12<59:39,  1.56it/s]  
2|891|Loss: 0.880309522151947:  14%|█▍        | 890/6470 [10:13<59:39,  1.56it/s]
2|891|Loss: 0.880309522151947:  14%|█▍        | 891/6470 [10:13<1:01:38,  1.51it/s]
2|892|Loss: 0.8259516954421997:  14%|█▍        | 891/6470 [10:14<1:01:38,  1.51it/s]
2|892|Loss: 0.8259516954421997:  14%|█▍        | 892/6470 [10:14<1:00:11,  1.54it/s]
2|893|Loss: 1.091580867767334:  14%|█▍        | 892/6470 [10:14<1:00:11,  1.54it/s] 
2|893|Loss: 1.091580867767334:  14%|█▍        | 893/6470 [10:14<57:11,  1.63it/s]  
2|894|Loss: 0.9199557304382324:  14%|█▍        | 893/6470 [10:15<57:11,  1.63it/s]
2|894|Loss: 0.9199557304382324:  14%|█▍        | 894/6470 [10:15<1:01:22,  1.51it/s]
2|895|Loss: 0.9414307475090027:  14%|█▍        | 894/6470 [10:15<1:01:22,  1.51it/s]
2|895|Loss: 0.9414307475090027:  14%|█▍        | 895/6470 [10:16<1:01:56,  1.50it/s]
2|896|Loss: 0.8524763584136963:  14%|█▍        | 895/6470 [10:16<1:01:56,  1.50it/s]
2|896|Loss: 0.8524763584136963:  14%|█▍        | 896/6470 [10:16<1:03:18,  1.47it/s]
2|897|Loss: 0.8144645690917969:  14%|█▍        | 896/6470 [10:17<1:03:18,  1.47it/s]
2|897|Loss: 0.8144645690917969:  14%|█▍        | 897/6470 [10:17<1:02:25,  1.49it/s]
2|898|Loss: 0.8605273962020874:  14%|█▍        | 897/6470 [10:18<1:02:25,  1.49it/s]
2|898|Loss: 0.8605273962020874:  14%|█▍        | 898/6470 [10:18<1:09:47,  1.33it/s]
2|899|Loss: 0.9260589480400085:  14%|█▍        | 898/6470 [10:19<1:09:47,  1.33it/s]
2|899|Loss: 0.9260589480400085:  14%|█▍        | 899/6470 [10:19<1:11:11,  1.30it/s]
2|900|Loss: 1.0168157815933228:  14%|█▍        | 899/6470 [10:19<1:11:11,  1.30it/s]
2|900|Loss: 1.0168157815933228:  14%|█▍        | 900/6470 [10:19<1:07:49,  1.37it/s]
2|901|Loss: 0.9568130970001221:  14%|█▍        | 900/6470 [10:20<1:07:49,  1.37it/s]
2|901|Loss: 0.9568130970001221:  14%|█▍        | 901/6470 [10:20<1:08:47,  1.35it/s]
2|902|Loss: 0.7530354261398315:  14%|█▍        | 901/6470 [10:21<1:08:47,  1.35it/s]
2|902|Loss: 0.7530354261398315:  14%|█▍        | 902/6470 [10:21<1:11:52,  1.29it/s]
2|903|Loss: 0.9038854241371155:  14%|█▍        | 902/6470 [10:22<1:11:52,  1.29it/s]
2|903|Loss: 0.9038854241371155:  14%|█▍        | 903/6470 [10:22<1:09:15,  1.34it/s]
2|904|Loss: 0.8044870495796204:  14%|█▍        | 903/6470 [10:22<1:09:15,  1.34it/s]
2|904|Loss: 0.8044870495796204:  14%|█▍        | 904/6470 [10:22<1:03:16,  1.47it/s]
2|905|Loss: 0.97443687915802:  14%|█▍        | 904/6470 [10:23<1:03:16,  1.47it/s]  
2|905|Loss: 0.97443687915802:  14%|█▍        | 905/6470 [10:23<59:17,  1.56it/s]  
2|906|Loss: 0.916316568851471:  14%|█▍        | 905/6470 [10:23<59:17,  1.56it/s]
2|906|Loss: 0.916316568851471:  14%|█▍        | 906/6470 [10:23<58:19,  1.59it/s]
2|907|Loss: 0.8080894947052002:  14%|█▍        | 906/6470 [10:24<58:19,  1.59it/s]
2|907|Loss: 0.8080894947052002:  14%|█▍        | 907/6470 [10:24<58:10,  1.59it/s]
2|908|Loss: 0.9007362723350525:  14%|█▍        | 907/6470 [10:24<58:10,  1.59it/s]
2|908|Loss: 0.9007362723350525:  14%|█▍        | 908/6470 [10:25<58:16,  1.59it/s]
2|909|Loss: 0.9095894694328308:  14%|█▍        | 908/6470 [10:25<58:16,  1.59it/s]
2|909|Loss: 0.9095894694328308:  14%|█▍        | 909/6470 [10:25<1:00:05,  1.54it/s]
2|910|Loss: 0.7443772554397583:  14%|█▍        | 909/6470 [10:26<1:00:05,  1.54it/s]
2|910|Loss: 0.7443772554397583:  14%|█▍        | 910/6470 [10:26<1:02:42,  1.48it/s]
2|911|Loss: 0.7549542784690857:  14%|█▍        | 910/6470 [10:27<1:02:42,  1.48it/s]
2|911|Loss: 0.7549542784690857:  14%|█▍        | 911/6470 [10:27<1:02:16,  1.49it/s]
2|912|Loss: 0.7026673555374146:  14%|█▍        | 911/6470 [10:27<1:02:16,  1.49it/s]
2|912|Loss: 0.7026673555374146:  14%|█▍        | 912/6470 [10:28<1:03:40,  1.45it/s]
2|913|Loss: 1.0076522827148438:  14%|█▍        | 912/6470 [10:28<1:03:40,  1.45it/s]
2|913|Loss: 1.0076522827148438:  14%|█▍        | 913/6470 [10:28<1:02:29,  1.48it/s]
2|914|Loss: 0.8592099547386169:  14%|█▍        | 913/6470 [10:29<1:02:29,  1.48it/s]
2|914|Loss: 0.8592099547386169:  14%|█▍        | 914/6470 [10:29<58:44,  1.58it/s]  
2|915|Loss: 0.8334286212921143:  14%|█▍        | 914/6470 [10:29<58:44,  1.58it/s]
2|915|Loss: 0.8334286212921143:  14%|█▍        | 915/6470 [10:29<57:00,  1.62it/s]
2|916|Loss: 0.850609540939331:  14%|█▍        | 915/6470 [10:30<57:00,  1.62it/s] 
2|916|Loss: 0.850609540939331:  14%|█▍        | 916/6470 [10:30<58:40,  1.58it/s]
2|917|Loss: 0.9429244995117188:  14%|█▍        | 916/6470 [10:30<58:40,  1.58it/s]
2|917|Loss: 0.9429244995117188:  14%|█▍        | 917/6470 [10:31<58:35,  1.58it/s]
2|918|Loss: 0.8394784927368164:  14%|█▍        | 917/6470 [10:31<58:35,  1.58it/s]
2|918|Loss: 0.8394784927368164:  14%|█▍        | 918/6470 [10:31<59:52,  1.55it/s]
2|919|Loss: 0.8621015548706055:  14%|█▍        | 918/6470 [10:32<59:52,  1.55it/s]
2|919|Loss: 0.8621015548706055:  14%|█▍        | 919/6470 [10:32<1:04:54,  1.43it/s]
2|920|Loss: 0.8237898349761963:  14%|█▍        | 919/6470 [10:33<1:04:54,  1.43it/s]
2|920|Loss: 0.8237898349761963:  14%|█▍        | 920/6470 [10:33<1:06:47,  1.38it/s]
2|921|Loss: 0.8297570943832397:  14%|█▍        | 920/6470 [10:33<1:06:47,  1.38it/s]
2|921|Loss: 0.8297570943832397:  14%|█▍        | 921/6470 [10:34<1:06:31,  1.39it/s]
2|922|Loss: 0.785645067691803:  14%|█▍        | 921/6470 [10:34<1:06:31,  1.39it/s] 
2|922|Loss: 0.785645067691803:  14%|█▍        | 922/6470 [10:34<1:01:43,  1.50it/s]
2|923|Loss: 0.8315834403038025:  14%|█▍        | 922/6470 [10:34<1:01:43,  1.50it/s]
2|923|Loss: 0.8315834403038025:  14%|█▍        | 923/6470 [10:35<1:00:24,  1.53it/s]
2|924|Loss: 0.7967434525489807:  14%|█▍        | 923/6470 [10:35<1:00:24,  1.53it/s]
2|924|Loss: 0.7967434525489807:  14%|█▍        | 924/6470 [10:35<1:00:12,  1.54it/s]
2|925|Loss: 0.8214328289031982:  14%|█▍        | 924/6470 [10:36<1:00:12,  1.54it/s]
2|925|Loss: 0.8214328289031982:  14%|█▍        | 925/6470 [10:36<59:27,  1.55it/s]  
2|926|Loss: 0.9658212065696716:  14%|█▍        | 925/6470 [10:36<59:27,  1.55it/s]
2|926|Loss: 0.9658212065696716:  14%|█▍        | 926/6470 [10:37<1:02:55,  1.47it/s]
2|927|Loss: 0.9315967559814453:  14%|█▍        | 926/6470 [10:37<1:02:55,  1.47it/s]
2|927|Loss: 0.9315967559814453:  14%|█▍        | 927/6470 [10:38<1:05:03,  1.42it/s]
2|928|Loss: 0.9550434350967407:  14%|█▍        | 927/6470 [10:38<1:05:03,  1.42it/s]
2|928|Loss: 0.9550434350967407:  14%|█▍        | 928/6470 [10:38<1:08:10,  1.35it/s]
2|929|Loss: 0.83615642786026:  14%|█▍        | 928/6470 [10:39<1:08:10,  1.35it/s]  
2|929|Loss: 0.83615642786026:  14%|█▍        | 929/6470 [10:39<1:06:29,  1.39it/s]
2|930|Loss: 0.837816059589386:  14%|█▍        | 929/6470 [10:39<1:06:29,  1.39it/s]
2|930|Loss: 0.837816059589386:  14%|█▍        | 930/6470 [10:40<1:06:08,  1.40it/s]
2|931|Loss: 1.0337371826171875:  14%|█▍        | 930/6470 [10:40<1:06:08,  1.40it/s]
2|931|Loss: 1.0337371826171875:  14%|█▍        | 931/6470 [10:40<1:01:38,  1.50it/s]
2|932|Loss: 0.7861993312835693:  14%|█▍        | 931/6470 [10:41<1:01:38,  1.50it/s]
2|932|Loss: 0.7861993312835693:  14%|█▍        | 932/6470 [10:41<1:00:03,  1.54it/s]
2|933|Loss: 0.9221381545066833:  14%|█▍        | 932/6470 [10:41<1:00:03,  1.54it/s]
2|933|Loss: 0.9221381545066833:  14%|█▍        | 933/6470 [10:42<59:21,  1.55it/s]  
2|934|Loss: 0.8562363386154175:  14%|█▍        | 933/6470 [10:42<59:21,  1.55it/s]
2|934|Loss: 0.8562363386154175:  14%|█▍        | 934/6470 [10:42<58:41,  1.57it/s]
2|935|Loss: 0.9496284127235413:  14%|█▍        | 934/6470 [10:43<58:41,  1.57it/s]
2|935|Loss: 0.9496284127235413:  14%|█▍        | 935/6470 [10:43<56:39,  1.63it/s]
2|936|Loss: 0.8194654583930969:  14%|█▍        | 935/6470 [10:43<56:39,  1.63it/s]
2|936|Loss: 0.8194654583930969:  14%|█▍        | 936/6470 [10:43<54:53,  1.68it/s]
2|937|Loss: 0.979320764541626:  14%|█▍        | 936/6470 [10:44<54:53,  1.68it/s] 
2|937|Loss: 0.979320764541626:  14%|█▍        | 937/6470 [10:44<57:58,  1.59it/s]
2|938|Loss: 0.8030669093132019:  14%|█▍        | 937/6470 [10:44<57:58,  1.59it/s]
2|938|Loss: 0.8030669093132019:  14%|█▍        | 938/6470 [10:45<1:02:37,  1.47it/s]
2|939|Loss: 0.8729556202888489:  14%|█▍        | 938/6470 [10:45<1:02:37,  1.47it/s]
2|939|Loss: 0.8729556202888489:  15%|█▍        | 939/6470 [10:46<1:06:18,  1.39it/s]
2|940|Loss: 0.7951804995536804:  15%|█▍        | 939/6470 [10:46<1:06:18,  1.39it/s]
2|940|Loss: 0.7951804995536804:  15%|█▍        | 940/6470 [10:46<1:05:06,  1.42it/s]
2|941|Loss: 0.7829013466835022:  15%|█▍        | 940/6470 [10:47<1:05:06,  1.42it/s]
2|941|Loss: 0.7829013466835022:  15%|█▍        | 941/6470 [10:47<1:05:43,  1.40it/s]
2|942|Loss: 0.8328670859336853:  15%|█▍        | 941/6470 [10:47<1:05:43,  1.40it/s]
2|942|Loss: 0.8328670859336853:  15%|█▍        | 942/6470 [10:48<1:06:21,  1.39it/s]
2|943|Loss: 0.9532520771026611:  15%|█▍        | 942/6470 [10:48<1:06:21,  1.39it/s]
2|943|Loss: 0.9532520771026611:  15%|█▍        | 943/6470 [10:48<1:04:15,  1.43it/s]
2|944|Loss: 0.9189532995223999:  15%|█▍        | 943/6470 [10:49<1:04:15,  1.43it/s]
2|944|Loss: 0.9189532995223999:  15%|█▍        | 944/6470 [10:49<1:02:42,  1.47it/s]
2|945|Loss: 1.0357128381729126:  15%|█▍        | 944/6470 [10:49<1:02:42,  1.47it/s]
2|945|Loss: 1.0357128381729126:  15%|█▍        | 945/6470 [10:50<1:04:13,  1.43it/s]
2|946|Loss: 0.8896239995956421:  15%|█▍        | 945/6470 [10:50<1:04:13,  1.43it/s]
2|946|Loss: 0.8896239995956421:  15%|█▍        | 946/6470 [10:50<1:04:55,  1.42it/s]
2|947|Loss: 0.8443289399147034:  15%|█▍        | 946/6470 [10:51<1:04:55,  1.42it/s]
2|947|Loss: 0.8443289399147034:  15%|█▍        | 947/6470 [10:51<1:00:33,  1.52it/s]
2|948|Loss: 0.8669061660766602:  15%|█▍        | 947/6470 [10:51<1:00:33,  1.52it/s]
2|948|Loss: 0.8669061660766602:  15%|█▍        | 948/6470 [10:52<1:00:42,  1.52it/s]
2|949|Loss: 0.8748357892036438:  15%|█▍        | 948/6470 [10:52<1:00:42,  1.52it/s]
2|949|Loss: 0.8748357892036438:  15%|█▍        | 949/6470 [10:52<59:09,  1.56it/s]  
2|950|Loss: 0.9646034836769104:  15%|█▍        | 949/6470 [10:53<59:09,  1.56it/s]
2|950|Loss: 0.9646034836769104:  15%|█▍        | 950/6470 [10:53<1:03:13,  1.46it/s]
2|951|Loss: 0.9601215124130249:  15%|█▍        | 950/6470 [10:53<1:03:13,  1.46it/s]
2|951|Loss: 0.9601215124130249:  15%|█▍        | 951/6470 [10:54<1:02:03,  1.48it/s]
2|952|Loss: 1.0440529584884644:  15%|█▍        | 951/6470 [10:54<1:02:03,  1.48it/s]
2|952|Loss: 1.0440529584884644:  15%|█▍        | 952/6470 [10:54<1:02:45,  1.47it/s]
2|953|Loss: 1.1476624011993408:  15%|█▍        | 952/6470 [10:55<1:02:45,  1.47it/s]
2|953|Loss: 1.1476624011993408:  15%|█▍        | 953/6470 [10:55<1:04:23,  1.43it/s]
2|954|Loss: 0.8021013140678406:  15%|█▍        | 953/6470 [10:56<1:04:23,  1.43it/s]
2|954|Loss: 0.8021013140678406:  15%|█▍        | 954/6470 [10:56<1:05:23,  1.41it/s]
2|955|Loss: 0.862853467464447:  15%|█▍        | 954/6470 [10:56<1:05:23,  1.41it/s] 
2|955|Loss: 0.862853467464447:  15%|█▍        | 955/6470 [10:56<1:01:28,  1.50it/s]
2|956|Loss: 0.8970034122467041:  15%|█▍        | 955/6470 [10:57<1:01:28,  1.50it/s]
2|956|Loss: 0.8970034122467041:  15%|█▍        | 956/6470 [10:57<1:02:12,  1.48it/s]
2|957|Loss: 0.8931961059570312:  15%|█▍        | 956/6470 [10:58<1:02:12,  1.48it/s]
2|957|Loss: 0.8931961059570312:  15%|█▍        | 957/6470 [10:58<1:01:55,  1.48it/s]
2|958|Loss: 0.854184627532959:  15%|█▍        | 957/6470 [10:58<1:01:55,  1.48it/s] 
2|958|Loss: 0.854184627532959:  15%|█▍        | 958/6470 [10:58<54:50,  1.68it/s]  
2|959|Loss: 0.8516470193862915:  15%|█▍        | 958/6470 [10:59<54:50,  1.68it/s]
2|959|Loss: 0.8516470193862915:  15%|█▍        | 959/6470 [10:59<52:44,  1.74it/s]
2|960|Loss: 0.9076478481292725:  15%|█▍        | 959/6470 [10:59<52:44,  1.74it/s]
2|960|Loss: 0.9076478481292725:  15%|█▍        | 960/6470 [10:59<50:00,  1.84it/s]
2|961|Loss: 0.877601146697998:  15%|█▍        | 960/6470 [11:00<50:00,  1.84it/s] 
2|961|Loss: 0.877601146697998:  15%|█▍        | 961/6470 [11:00<54:46,  1.68it/s]
2|962|Loss: 0.7740684151649475:  15%|█▍        | 961/6470 [11:00<54:46,  1.68it/s]
2|962|Loss: 0.7740684151649475:  15%|█▍        | 962/6470 [11:01<59:15,  1.55it/s]
2|963|Loss: 0.8561927676200867:  15%|█▍        | 962/6470 [11:01<59:15,  1.55it/s]
2|963|Loss: 0.8561927676200867:  15%|█▍        | 963/6470 [11:01<59:29,  1.54it/s]
2|964|Loss: 0.8836103081703186:  15%|█▍        | 963/6470 [11:02<59:29,  1.54it/s]
2|964|Loss: 0.8836103081703186:  15%|█▍        | 964/6470 [11:02<56:17,  1.63it/s]
2|965|Loss: 0.7425274848937988:  15%|█▍        | 964/6470 [11:02<56:17,  1.63it/s]
2|965|Loss: 0.7425274848937988:  15%|█▍        | 965/6470 [11:03<1:00:00,  1.53it/s]
2|966|Loss: 0.7581766843795776:  15%|█▍        | 965/6470 [11:03<1:00:00,  1.53it/s]
2|966|Loss: 0.7581766843795776:  15%|█▍        | 966/6470 [11:03<54:25,  1.69it/s]  
2|967|Loss: 0.722025990486145:  15%|█▍        | 966/6470 [11:03<54:25,  1.69it/s] 
2|967|Loss: 0.722025990486145:  15%|█▍        | 967/6470 [11:04<56:13,  1.63it/s]
2|968|Loss: 0.7497163414955139:  15%|█▍        | 967/6470 [11:04<56:13,  1.63it/s]
2|968|Loss: 0.7497163414955139:  15%|█▍        | 968/6470 [11:05<59:35,  1.54it/s]
2|969|Loss: 0.7991045117378235:  15%|█▍        | 968/6470 [11:05<59:35,  1.54it/s]
2|969|Loss: 0.7991045117378235:  15%|█▍        | 969/6470 [11:05<58:29,  1.57it/s]
2|970|Loss: 0.8713015913963318:  15%|█▍        | 969/6470 [11:05<58:29,  1.57it/s]
2|970|Loss: 0.8713015913963318:  15%|█▍        | 970/6470 [11:06<58:18,  1.57it/s]
2|971|Loss: 0.8984168171882629:  15%|█▍        | 970/6470 [11:06<58:18,  1.57it/s]
2|971|Loss: 0.8984168171882629:  15%|█▌        | 971/6470 [11:06<56:00,  1.64it/s]
2|972|Loss: 0.9568005204200745:  15%|█▌        | 971/6470 [11:07<56:00,  1.64it/s]
2|972|Loss: 0.9568005204200745:  15%|█▌        | 972/6470 [11:07<57:35,  1.59it/s]
2|973|Loss: 0.8508702516555786:  15%|█▌        | 972/6470 [11:07<57:35,  1.59it/s]
2|973|Loss: 0.8508702516555786:  15%|█▌        | 973/6470 [11:08<56:53,  1.61it/s]
2|974|Loss: 0.8288754820823669:  15%|█▌        | 973/6470 [11:08<56:53,  1.61it/s]
2|974|Loss: 0.8288754820823669:  15%|█▌        | 974/6470 [11:08<54:21,  1.69it/s]
2|975|Loss: 0.7399004101753235:  15%|█▌        | 974/6470 [11:08<54:21,  1.69it/s]
2|975|Loss: 0.7399004101753235:  15%|█▌        | 975/6470 [11:09<50:53,  1.80it/s]
2|976|Loss: 0.7874304056167603:  15%|█▌        | 975/6470 [11:09<50:53,  1.80it/s]
2|976|Loss: 0.7874304056167603:  15%|█▌        | 976/6470 [11:09<56:13,  1.63it/s]
2|977|Loss: 0.8881211876869202:  15%|█▌        | 976/6470 [11:10<56:13,  1.63it/s]
2|977|Loss: 0.8881211876869202:  15%|█▌        | 977/6470 [11:10<1:04:04,  1.43it/s]
2|978|Loss: 0.9016094207763672:  15%|█▌        | 977/6470 [11:11<1:04:04,  1.43it/s]
2|978|Loss: 0.9016094207763672:  15%|█▌        | 978/6470 [11:11<1:08:17,  1.34it/s]
2|979|Loss: 0.8901857137680054:  15%|█▌        | 978/6470 [11:11<1:08:17,  1.34it/s]
2|979|Loss: 0.8901857137680054:  15%|█▌        | 979/6470 [11:12<1:06:15,  1.38it/s]
2|980|Loss: 0.9334219098091125:  15%|█▌        | 979/6470 [11:12<1:06:15,  1.38it/s]
2|980|Loss: 0.9334219098091125:  15%|█▌        | 980/6470 [11:12<1:06:01,  1.39it/s]
2|981|Loss: 0.8624491095542908:  15%|█▌        | 980/6470 [11:13<1:06:01,  1.39it/s]
2|981|Loss: 0.8624491095542908:  15%|█▌        | 981/6470 [11:13<1:07:55,  1.35it/s]
2|982|Loss: 0.8049437403678894:  15%|█▌        | 981/6470 [11:14<1:07:55,  1.35it/s]
2|982|Loss: 0.8049437403678894:  15%|█▌        | 982/6470 [11:14<1:11:21,  1.28it/s]
2|983|Loss: 0.9151397347450256:  15%|█▌        | 982/6470 [11:15<1:11:21,  1.28it/s]
2|983|Loss: 0.9151397347450256:  15%|█▌        | 983/6470 [11:15<1:10:16,  1.30it/s]
2|984|Loss: 0.7882909774780273:  15%|█▌        | 983/6470 [11:15<1:10:16,  1.30it/s]
2|984|Loss: 0.7882909774780273:  15%|█▌        | 984/6470 [11:16<1:06:56,  1.37it/s]
2|985|Loss: 0.7947599291801453:  15%|█▌        | 984/6470 [11:16<1:06:56,  1.37it/s]
2|985|Loss: 0.7947599291801453:  15%|█▌        | 985/6470 [11:16<1:01:33,  1.48it/s]
2|986|Loss: 0.8211725354194641:  15%|█▌        | 985/6470 [11:16<1:01:33,  1.48it/s]
2|986|Loss: 0.8211725354194641:  15%|█▌        | 986/6470 [11:17<1:02:20,  1.47it/s]
2|987|Loss: 0.8835464715957642:  15%|█▌        | 986/6470 [11:17<1:02:20,  1.47it/s]
2|987|Loss: 0.8835464715957642:  15%|█▌        | 987/6470 [11:18<1:03:55,  1.43it/s]
2|988|Loss: 0.8659191727638245:  15%|█▌        | 987/6470 [11:18<1:03:55,  1.43it/s]
2|988|Loss: 0.8659191727638245:  15%|█▌        | 988/6470 [11:18<1:04:49,  1.41it/s]
2|989|Loss: 0.9031358361244202:  15%|█▌        | 988/6470 [11:19<1:04:49,  1.41it/s]
2|989|Loss: 0.9031358361244202:  15%|█▌        | 989/6470 [11:19<1:08:02,  1.34it/s]
2|990|Loss: 0.9882447123527527:  15%|█▌        | 989/6470 [11:19<1:08:02,  1.34it/s]
2|990|Loss: 0.9882447123527527:  15%|█▌        | 990/6470 [11:20<1:06:25,  1.37it/s]
2|991|Loss: 0.8300654292106628:  15%|█▌        | 990/6470 [11:20<1:06:25,  1.37it/s]
2|991|Loss: 0.8300654292106628:  15%|█▌        | 991/6470 [11:21<1:08:06,  1.34it/s]
2|992|Loss: 1.033119559288025:  15%|█▌        | 991/6470 [11:21<1:08:06,  1.34it/s] 
2|992|Loss: 1.033119559288025:  15%|█▌        | 992/6470 [11:21<1:09:41,  1.31it/s]
2|993|Loss: 0.8160026669502258:  15%|█▌        | 992/6470 [11:22<1:09:41,  1.31it/s]
2|993|Loss: 0.8160026669502258:  15%|█▌        | 993/6470 [11:22<1:08:23,  1.33it/s]
2|994|Loss: 0.9419479966163635:  15%|█▌        | 993/6470 [11:22<1:08:23,  1.33it/s]
2|994|Loss: 0.9419479966163635:  15%|█▌        | 994/6470 [11:23<1:07:10,  1.36it/s]
2|995|Loss: 0.8084484934806824:  15%|█▌        | 994/6470 [11:23<1:07:10,  1.36it/s]
2|995|Loss: 0.8084484934806824:  15%|█▌        | 995/6470 [11:23<1:05:00,  1.40it/s]
2|996|Loss: 0.7907291054725647:  15%|█▌        | 995/6470 [11:24<1:05:00,  1.40it/s]
2|996|Loss: 0.7907291054725647:  15%|█▌        | 996/6470 [11:24<1:07:40,  1.35it/s]
2|997|Loss: 0.8664026260375977:  15%|█▌        | 996/6470 [11:25<1:07:40,  1.35it/s]
2|997|Loss: 0.8664026260375977:  15%|█▌        | 997/6470 [11:25<1:08:44,  1.33it/s]
2|998|Loss: 0.8850764632225037:  15%|█▌        | 997/6470 [11:25<1:08:44,  1.33it/s]
2|998|Loss: 0.8850764632225037:  15%|█▌        | 998/6470 [11:26<1:05:57,  1.38it/s]
2|999|Loss: 1.0051521062850952:  15%|█▌        | 998/6470 [11:26<1:05:57,  1.38it/s]
2|999|Loss: 1.0051521062850952:  15%|█▌        | 999/6470 [11:26<1:02:08,  1.47it/s]
2|1000|Loss: 0.8940077424049377:  15%|█▌        | 999/6470 [11:27<1:02:08,  1.47it/s]
2|1000|Loss: 0.8940077424049377:  15%|█▌        | 1000/6470 [11:27<1:00:49,  1.50it/s]
2|1001|Loss: 0.7300077676773071:  15%|█▌        | 1000/6470 [11:27<1:00:49,  1.50it/s]
2|1001|Loss: 0.7300077676773071:  15%|█▌        | 1001/6470 [11:27<57:18,  1.59it/s]  
2|1002|Loss: 0.9143060445785522:  15%|█▌        | 1001/6470 [11:28<57:18,  1.59it/s]
2|1002|Loss: 0.9143060445785522:  15%|█▌        | 1002/6470 [11:28<1:00:28,  1.51it/s]
2|1003|Loss: 0.7695780396461487:  15%|█▌        | 1002/6470 [11:29<1:00:28,  1.51it/s]
2|1003|Loss: 0.7695780396461487:  16%|█▌        | 1003/6470 [11:29<1:05:33,  1.39it/s]
2|1004|Loss: 0.810868501663208:  16%|█▌        | 1003/6470 [11:29<1:05:33,  1.39it/s] 
2|1004|Loss: 0.810868501663208:  16%|█▌        | 1004/6470 [11:30<1:04:51,  1.40it/s]
2|1005|Loss: 0.837812066078186:  16%|█▌        | 1004/6470 [11:30<1:04:51,  1.40it/s]
2|1005|Loss: 0.837812066078186:  16%|█▌        | 1005/6470 [11:30<1:01:34,  1.48it/s]
2|1006|Loss: 0.8489231467247009:  16%|█▌        | 1005/6470 [11:31<1:01:34,  1.48it/s]
2|1006|Loss: 0.8489231467247009:  16%|█▌        | 1006/6470 [11:31<1:02:50,  1.45it/s]
2|1007|Loss: 1.0063289403915405:  16%|█▌        | 1006/6470 [11:31<1:02:50,  1.45it/s]
2|1007|Loss: 1.0063289403915405:  16%|█▌        | 1007/6470 [11:32<1:03:31,  1.43it/s]
2|1008|Loss: 0.7173125147819519:  16%|█▌        | 1007/6470 [11:32<1:03:31,  1.43it/s]
2|1008|Loss: 0.7173125147819519:  16%|█▌        | 1008/6470 [11:33<1:09:24,  1.31it/s]
2|1009|Loss: 0.853240966796875:  16%|█▌        | 1008/6470 [11:33<1:09:24,  1.31it/s] 
2|1009|Loss: 0.853240966796875:  16%|█▌        | 1009/6470 [11:33<1:09:21,  1.31it/s]
2|1010|Loss: 0.9413650631904602:  16%|█▌        | 1009/6470 [11:34<1:09:21,  1.31it/s]
2|1010|Loss: 0.9413650631904602:  16%|█▌        | 1010/6470 [11:34<1:07:40,  1.34it/s]
2|1011|Loss: 0.8363437652587891:  16%|█▌        | 1010/6470 [11:35<1:07:40,  1.34it/s]
2|1011|Loss: 0.8363437652587891:  16%|█▌        | 1011/6470 [11:35<1:09:05,  1.32it/s]
2|1012|Loss: 0.7154591679573059:  16%|█▌        | 1011/6470 [11:35<1:09:05,  1.32it/s]
2|1012|Loss: 0.7154591679573059:  16%|█▌        | 1012/6470 [11:36<1:09:20,  1.31it/s]
2|1013|Loss: 0.8992063999176025:  16%|█▌        | 1012/6470 [11:36<1:09:20,  1.31it/s]
2|1013|Loss: 0.8992063999176025:  16%|█▌        | 1013/6470 [11:36<1:08:02,  1.34it/s]
2|1014|Loss: 0.8847254514694214:  16%|█▌        | 1013/6470 [11:37<1:08:02,  1.34it/s]
2|1014|Loss: 0.8847254514694214:  16%|█▌        | 1014/6470 [11:37<1:05:29,  1.39it/s]
2|1015|Loss: 0.8629357218742371:  16%|█▌        | 1014/6470 [11:37<1:05:29,  1.39it/s]
2|1015|Loss: 0.8629357218742371:  16%|█▌        | 1015/6470 [11:38<1:07:21,  1.35it/s]
2|1016|Loss: 0.7140786051750183:  16%|█▌        | 1015/6470 [11:38<1:07:21,  1.35it/s]
2|1016|Loss: 0.7140786051750183:  16%|█▌        | 1016/6470 [11:39<1:05:44,  1.38it/s]
2|1017|Loss: 0.8386821746826172:  16%|█▌        | 1016/6470 [11:39<1:05:44,  1.38it/s]
2|1017|Loss: 0.8386821746826172:  16%|█▌        | 1017/6470 [11:39<1:03:36,  1.43it/s]
2|1018|Loss: 0.8800653219223022:  16%|█▌        | 1017/6470 [11:40<1:03:36,  1.43it/s]
2|1018|Loss: 0.8800653219223022:  16%|█▌        | 1018/6470 [11:40<1:06:20,  1.37it/s]
2|1019|Loss: 0.781810462474823:  16%|█▌        | 1018/6470 [11:40<1:06:20,  1.37it/s] 
2|1019|Loss: 0.781810462474823:  16%|█▌        | 1019/6470 [11:41<1:04:44,  1.40it/s]
2|1020|Loss: 0.7269324660301208:  16%|█▌        | 1019/6470 [11:41<1:04:44,  1.40it/s]
2|1020|Loss: 0.7269324660301208:  16%|█▌        | 1020/6470 [11:41<1:04:41,  1.40it/s]
2|1021|Loss: 0.7971667647361755:  16%|█▌        | 1020/6470 [11:42<1:04:41,  1.40it/s]
2|1021|Loss: 0.7971667647361755:  16%|█▌        | 1021/6470 [11:42<1:04:32,  1.41it/s]
2|1022|Loss: 0.8053759336471558:  16%|█▌        | 1021/6470 [11:42<1:04:32,  1.41it/s]
2|1022|Loss: 0.8053759336471558:  16%|█▌        | 1022/6470 [11:43<1:00:08,  1.51it/s]
2|1023|Loss: 0.8073131442070007:  16%|█▌        | 1022/6470 [11:43<1:00:08,  1.51it/s]
2|1023|Loss: 0.8073131442070007:  16%|█▌        | 1023/6470 [11:43<57:30,  1.58it/s]  
2|1024|Loss: 1.0188595056533813:  16%|█▌        | 1023/6470 [11:44<57:30,  1.58it/s]
2|1024|Loss: 1.0188595056533813:  16%|█▌        | 1024/6470 [11:44<1:01:29,  1.48it/s]
2|1025|Loss: 1.0999054908752441:  16%|█▌        | 1024/6470 [11:44<1:01:29,  1.48it/s]
2|1025|Loss: 1.0999054908752441:  16%|█▌        | 1025/6470 [11:45<1:03:29,  1.43it/s]
2|1026|Loss: 0.8832594752311707:  16%|█▌        | 1025/6470 [11:45<1:03:29,  1.43it/s]
2|1026|Loss: 0.8832594752311707:  16%|█▌        | 1026/6470 [11:45<1:03:50,  1.42it/s]
2|1027|Loss: 0.693555474281311:  16%|█▌        | 1026/6470 [11:46<1:03:50,  1.42it/s] 
2|1027|Loss: 0.693555474281311:  16%|█▌        | 1027/6470 [11:46<1:05:56,  1.38it/s]
2|1028|Loss: 0.8429746627807617:  16%|█▌        | 1027/6470 [11:47<1:05:56,  1.38it/s]
2|1028|Loss: 0.8429746627807617:  16%|█▌        | 1028/6470 [11:47<1:05:43,  1.38it/s]
2|1029|Loss: 0.7203544974327087:  16%|█▌        | 1028/6470 [11:47<1:05:43,  1.38it/s]
2|1029|Loss: 0.7203544974327087:  16%|█▌        | 1029/6470 [11:48<1:03:21,  1.43it/s]
2|1030|Loss: 0.8806440234184265:  16%|█▌        | 1029/6470 [11:48<1:03:21,  1.43it/s]
2|1030|Loss: 0.8806440234184265:  16%|█▌        | 1030/6470 [11:48<1:05:37,  1.38it/s]
2|1031|Loss: 0.7961913347244263:  16%|█▌        | 1030/6470 [11:49<1:05:37,  1.38it/s]
2|1031|Loss: 0.7961913347244263:  16%|█▌        | 1031/6470 [11:49<1:06:55,  1.35it/s]
2|1032|Loss: 0.8331108689308167:  16%|█▌        | 1031/6470 [11:50<1:06:55,  1.35it/s]
2|1032|Loss: 0.8331108689308167:  16%|█▌        | 1032/6470 [11:50<1:04:39,  1.40it/s]
2|1033|Loss: 0.7802692651748657:  16%|█▌        | 1032/6470 [11:50<1:04:39,  1.40it/s]
2|1033|Loss: 0.7802692651748657:  16%|█▌        | 1033/6470 [11:50<1:03:54,  1.42it/s]
2|1034|Loss: 0.9943200945854187:  16%|█▌        | 1033/6470 [11:51<1:03:54,  1.42it/s]
2|1034|Loss: 0.9943200945854187:  16%|█▌        | 1034/6470 [11:51<1:03:56,  1.42it/s]
2|1035|Loss: 0.8642841577529907:  16%|█▌        | 1034/6470 [11:52<1:03:56,  1.42it/s]
2|1035|Loss: 0.8642841577529907:  16%|█▌        | 1035/6470 [11:52<1:07:21,  1.34it/s]
2|1036|Loss: 0.9580783843994141:  16%|█▌        | 1035/6470 [11:52<1:07:21,  1.34it/s]
2|1036|Loss: 0.9580783843994141:  16%|█▌        | 1036/6470 [11:53<1:05:33,  1.38it/s]
2|1037|Loss: 1.0042917728424072:  16%|█▌        | 1036/6470 [11:53<1:05:33,  1.38it/s]
2|1037|Loss: 1.0042917728424072:  16%|█▌        | 1037/6470 [11:53<1:05:23,  1.38it/s]
2|1038|Loss: 0.948391318321228:  16%|█▌        | 1037/6470 [11:54<1:05:23,  1.38it/s] 
2|1038|Loss: 0.948391318321228:  16%|█▌        | 1038/6470 [11:54<1:05:40,  1.38it/s]
2|1039|Loss: 0.705696702003479:  16%|█▌        | 1038/6470 [11:55<1:05:40,  1.38it/s]
2|1039|Loss: 0.705696702003479:  16%|█▌        | 1039/6470 [11:55<1:04:00,  1.41it/s]
2|1040|Loss: 0.9998123645782471:  16%|█▌        | 1039/6470 [11:55<1:04:00,  1.41it/s]
2|1040|Loss: 0.9998123645782471:  16%|█▌        | 1040/6470 [11:56<1:06:53,  1.35it/s]
2|1041|Loss: 0.8218669295310974:  16%|█▌        | 1040/6470 [11:56<1:06:53,  1.35it/s]
2|1041|Loss: 0.8218669295310974:  16%|█▌        | 1041/6470 [11:56<1:07:41,  1.34it/s]
2|1042|Loss: 0.7008392214775085:  16%|█▌        | 1041/6470 [11:57<1:07:41,  1.34it/s]
2|1042|Loss: 0.7008392214775085:  16%|█▌        | 1042/6470 [11:57<1:11:13,  1.27it/s]
2|1043|Loss: 0.7968025803565979:  16%|█▌        | 1042/6470 [11:58<1:11:13,  1.27it/s]
2|1043|Loss: 0.7968025803565979:  16%|█▌        | 1043/6470 [11:58<1:12:30,  1.25it/s]
2|1044|Loss: 0.75994473695755:  16%|█▌        | 1043/6470 [11:59<1:12:30,  1.25it/s]  
2|1044|Loss: 0.75994473695755:  16%|█▌        | 1044/6470 [11:59<1:14:13,  1.22it/s]
2|1045|Loss: 0.8932800889015198:  16%|█▌        | 1044/6470 [11:59<1:14:13,  1.22it/s]
2|1045|Loss: 0.8932800889015198:  16%|█▌        | 1045/6470 [12:00<1:15:06,  1.20it/s]
2|1046|Loss: 0.9375123381614685:  16%|█▌        | 1045/6470 [12:00<1:15:06,  1.20it/s]
2|1046|Loss: 0.9375123381614685:  16%|█▌        | 1046/6470 [12:01<1:13:17,  1.23it/s]
2|1047|Loss: 0.8872778415679932:  16%|█▌        | 1046/6470 [12:01<1:13:17,  1.23it/s]
2|1047|Loss: 0.8872778415679932:  16%|█▌        | 1047/6470 [12:01<1:11:45,  1.26it/s]
2|1048|Loss: 0.9026851058006287:  16%|█▌        | 1047/6470 [12:02<1:11:45,  1.26it/s]
2|1048|Loss: 0.9026851058006287:  16%|█▌        | 1048/6470 [12:02<1:07:17,  1.34it/s]
2|1049|Loss: 0.8746651411056519:  16%|█▌        | 1048/6470 [12:02<1:07:17,  1.34it/s]
2|1049|Loss: 0.8746651411056519:  16%|█▌        | 1049/6470 [12:03<1:04:16,  1.41it/s]
2|1050|Loss: 0.78522789478302:  16%|█▌        | 1049/6470 [12:03<1:04:16,  1.41it/s]  
2|1050|Loss: 0.78522789478302:  16%|█▌        | 1050/6470 [12:03<58:32,  1.54it/s]  
2|1051|Loss: 0.9429894089698792:  16%|█▌        | 1050/6470 [12:03<58:32,  1.54it/s]
2|1051|Loss: 0.9429894089698792:  16%|█▌        | 1051/6470 [12:04<55:41,  1.62it/s]
2|1052|Loss: 0.7993295192718506:  16%|█▌        | 1051/6470 [12:04<55:41,  1.62it/s]
2|1052|Loss: 0.7993295192718506:  16%|█▋        | 1052/6470 [12:04<53:10,  1.70it/s]
2|1053|Loss: 0.8872669339179993:  16%|█▋        | 1052/6470 [12:05<53:10,  1.70it/s]
2|1053|Loss: 0.8872669339179993:  16%|█▋        | 1053/6470 [12:05<58:28,  1.54it/s]
2|1054|Loss: 0.941317081451416:  16%|█▋        | 1053/6470 [12:05<58:28,  1.54it/s] 
2|1054|Loss: 0.941317081451416:  16%|█▋        | 1054/6470 [12:06<1:01:41,  1.46it/s]
2|1055|Loss: 0.7507093548774719:  16%|█▋        | 1054/6470 [12:06<1:01:41,  1.46it/s]
2|1055|Loss: 0.7507093548774719:  16%|█▋        | 1055/6470 [12:07<1:05:28,  1.38it/s]
2|1056|Loss: 0.8567716479301453:  16%|█▋        | 1055/6470 [12:07<1:05:28,  1.38it/s]
2|1056|Loss: 0.8567716479301453:  16%|█▋        | 1056/6470 [12:07<1:04:31,  1.40it/s]
2|1057|Loss: 0.8931211829185486:  16%|█▋        | 1056/6470 [12:08<1:04:31,  1.40it/s]
2|1057|Loss: 0.8931211829185486:  16%|█▋        | 1057/6470 [12:08<1:04:29,  1.40it/s]
2|1058|Loss: 0.737484335899353:  16%|█▋        | 1057/6470 [12:08<1:04:29,  1.40it/s] 
2|1058|Loss: 0.737484335899353:  16%|█▋        | 1058/6470 [12:09<1:08:04,  1.32it/s]
2|1059|Loss: 0.7759838104248047:  16%|█▋        | 1058/6470 [12:09<1:08:04,  1.32it/s]
2|1059|Loss: 0.7759838104248047:  16%|█▋        | 1059/6470 [12:10<1:07:46,  1.33it/s]
2|1060|Loss: 1.107100248336792:  16%|█▋        | 1059/6470 [12:10<1:07:46,  1.33it/s] 
2|1060|Loss: 1.107100248336792:  16%|█▋        | 1060/6470 [12:10<1:02:15,  1.45it/s]
2|1061|Loss: 1.0665733814239502:  16%|█▋        | 1060/6470 [12:10<1:02:15,  1.45it/s]
2|1061|Loss: 1.0665733814239502:  16%|█▋        | 1061/6470 [12:11<58:18,  1.55it/s]  
2|1062|Loss: 0.7509452700614929:  16%|█▋        | 1061/6470 [12:11<58:18,  1.55it/s]
2|1062|Loss: 0.7509452700614929:  16%|█▋        | 1062/6470 [12:11<59:05,  1.53it/s]
2|1063|Loss: 0.7642917633056641:  16%|█▋        | 1062/6470 [12:12<59:05,  1.53it/s]
2|1063|Loss: 0.7642917633056641:  16%|█▋        | 1063/6470 [12:12<1:01:03,  1.48it/s]
2|1064|Loss: 0.9119210243225098:  16%|█▋        | 1063/6470 [12:12<1:01:03,  1.48it/s]
2|1064|Loss: 0.9119210243225098:  16%|█▋        | 1064/6470 [12:13<1:00:18,  1.49it/s]
2|1065|Loss: 0.9125325679779053:  16%|█▋        | 1064/6470 [12:13<1:00:18,  1.49it/s]
2|1065|Loss: 0.9125325679779053:  16%|█▋        | 1065/6470 [12:13<58:16,  1.55it/s]  
2|1066|Loss: 0.8660483956336975:  16%|█▋        | 1065/6470 [12:14<58:16,  1.55it/s]
2|1066|Loss: 0.8660483956336975:  16%|█▋        | 1066/6470 [12:14<1:01:10,  1.47it/s]
2|1067|Loss: 0.796471893787384:  16%|█▋        | 1066/6470 [12:14<1:01:10,  1.47it/s] 
2|1067|Loss: 0.796471893787384:  16%|█▋        | 1067/6470 [12:15<1:06:14,  1.36it/s]
2|1068|Loss: 0.8379932641983032:  16%|█▋        | 1067/6470 [12:15<1:06:14,  1.36it/s]
2|1068|Loss: 0.8379932641983032:  17%|█▋        | 1068/6470 [12:16<1:04:45,  1.39it/s]
2|1069|Loss: 0.8813865184783936:  17%|█▋        | 1068/6470 [12:16<1:04:45,  1.39it/s]
2|1069|Loss: 0.8813865184783936:  17%|█▋        | 1069/6470 [12:16<1:05:48,  1.37it/s]
2|1070|Loss: 0.8635314106941223:  17%|█▋        | 1069/6470 [12:17<1:05:48,  1.37it/s]
2|1070|Loss: 0.8635314106941223:  17%|█▋        | 1070/6470 [12:17<1:08:19,  1.32it/s]
2|1071|Loss: 0.8668950200080872:  17%|█▋        | 1070/6470 [12:18<1:08:19,  1.32it/s]
2|1071|Loss: 0.8668950200080872:  17%|█▋        | 1071/6470 [12:18<1:09:42,  1.29it/s]
2|1072|Loss: 0.8815847039222717:  17%|█▋        | 1071/6470 [12:18<1:09:42,  1.29it/s]
2|1072|Loss: 0.8815847039222717:  17%|█▋        | 1072/6470 [12:19<1:09:32,  1.29it/s]
2|1073|Loss: 0.8356354832649231:  17%|█▋        | 1072/6470 [12:19<1:09:32,  1.29it/s]
2|1073|Loss: 0.8356354832649231:  17%|█▋        | 1073/6470 [12:20<1:11:18,  1.26it/s]
2|1074|Loss: 0.9795655608177185:  17%|█▋        | 1073/6470 [12:20<1:11:18,  1.26it/s]
2|1074|Loss: 0.9795655608177185:  17%|█▋        | 1074/6470 [12:20<1:08:01,  1.32it/s]
2|1075|Loss: 0.9322965741157532:  17%|█▋        | 1074/6470 [12:21<1:08:01,  1.32it/s]
2|1075|Loss: 0.9322965741157532:  17%|█▋        | 1075/6470 [12:21<1:06:09,  1.36it/s]
2|1076|Loss: 0.9305950403213501:  17%|█▋        | 1075/6470 [12:21<1:06:09,  1.36it/s]
2|1076|Loss: 0.9305950403213501:  17%|█▋        | 1076/6470 [12:22<1:03:30,  1.42it/s]
2|1077|Loss: 0.8840424418449402:  17%|█▋        | 1076/6470 [12:22<1:03:30,  1.42it/s]
2|1077|Loss: 0.8840424418449402:  17%|█▋        | 1077/6470 [12:22<1:05:04,  1.38it/s]
2|1078|Loss: 0.7786638736724854:  17%|█▋        | 1077/6470 [12:23<1:05:04,  1.38it/s]
2|1078|Loss: 0.7786638736724854:  17%|█▋        | 1078/6470 [12:23<1:06:08,  1.36it/s]
2|1079|Loss: 0.6509726643562317:  17%|█▋        | 1078/6470 [12:24<1:06:08,  1.36it/s]
2|1079|Loss: 0.6509726643562317:  17%|█▋        | 1079/6470 [12:24<1:07:45,  1.33it/s]
2|1080|Loss: 0.9055106043815613:  17%|█▋        | 1079/6470 [12:24<1:07:45,  1.33it/s]
2|1080|Loss: 0.9055106043815613:  17%|█▋        | 1080/6470 [12:25<1:07:40,  1.33it/s]
2|1081|Loss: 0.9651656150817871:  17%|█▋        | 1080/6470 [12:25<1:07:40,  1.33it/s]
2|1081|Loss: 0.9651656150817871:  17%|█▋        | 1081/6470 [12:25<1:01:44,  1.45it/s]
2|1082|Loss: 0.9100634455680847:  17%|█▋        | 1081/6470 [12:26<1:01:44,  1.45it/s]
2|1082|Loss: 0.9100634455680847:  17%|█▋        | 1082/6470 [12:26<1:00:46,  1.48it/s]
2|1083|Loss: 0.9531391263008118:  17%|█▋        | 1082/6470 [12:26<1:00:46,  1.48it/s]
2|1083|Loss: 0.9531391263008118:  17%|█▋        | 1083/6470 [12:27<1:02:08,  1.44it/s]
2|1084|Loss: 0.84494948387146:  17%|█▋        | 1083/6470 [12:27<1:02:08,  1.44it/s]  
2|1084|Loss: 0.84494948387146:  17%|█▋        | 1084/6470 [12:27<1:03:28,  1.41it/s]
2|1085|Loss: 0.7908526659011841:  17%|█▋        | 1084/6470 [12:28<1:03:28,  1.41it/s]
2|1085|Loss: 0.7908526659011841:  17%|█▋        | 1085/6470 [12:28<1:02:16,  1.44it/s]
2|1086|Loss: 0.929724395275116:  17%|█▋        | 1085/6470 [12:28<1:02:16,  1.44it/s] 
2|1086|Loss: 0.929724395275116:  17%|█▋        | 1086/6470 [12:29<1:02:48,  1.43it/s]
2|1087|Loss: 0.8851770758628845:  17%|█▋        | 1086/6470 [12:29<1:02:48,  1.43it/s]
2|1087|Loss: 0.8851770758628845:  17%|█▋        | 1087/6470 [12:30<1:06:17,  1.35it/s]
2|1088|Loss: 0.7738739848136902:  17%|█▋        | 1087/6470 [12:30<1:06:17,  1.35it/s]
2|1088|Loss: 0.7738739848136902:  17%|█▋        | 1088/6470 [12:30<1:06:59,  1.34it/s]
2|1089|Loss: 0.8673465847969055:  17%|█▋        | 1088/6470 [12:31<1:06:59,  1.34it/s]
2|1089|Loss: 0.8673465847969055:  17%|█▋        | 1089/6470 [12:31<1:02:11,  1.44it/s]
2|1090|Loss: 0.9061955213546753:  17%|█▋        | 1089/6470 [12:31<1:02:11,  1.44it/s]
2|1090|Loss: 0.9061955213546753:  17%|█▋        | 1090/6470 [12:32<1:02:30,  1.43it/s]
2|1091|Loss: 0.7835689187049866:  17%|█▋        | 1090/6470 [12:32<1:02:30,  1.43it/s]
2|1091|Loss: 0.7835689187049866:  17%|█▋        | 1091/6470 [12:32<1:03:58,  1.40it/s]
2|1092|Loss: 0.9358444809913635:  17%|█▋        | 1091/6470 [12:33<1:03:58,  1.40it/s]
2|1092|Loss: 0.9358444809913635:  17%|█▋        | 1092/6470 [12:33<1:04:57,  1.38it/s]
2|1093|Loss: 0.8431116938591003:  17%|█▋        | 1092/6470 [12:33<1:04:57,  1.38it/s]
2|1093|Loss: 0.8431116938591003:  17%|█▋        | 1093/6470 [12:34<1:06:32,  1.35it/s]
2|1094|Loss: 0.7914628982543945:  17%|█▋        | 1093/6470 [12:34<1:06:32,  1.35it/s]
2|1094|Loss: 0.7914628982543945:  17%|█▋        | 1094/6470 [12:34<1:04:02,  1.40it/s]
2|1095|Loss: 0.9773410558700562:  17%|█▋        | 1094/6470 [12:35<1:04:02,  1.40it/s]
2|1095|Loss: 0.9773410558700562:  17%|█▋        | 1095/6470 [12:35<1:01:58,  1.45it/s]
2|1096|Loss: 0.8756964802742004:  17%|█▋        | 1095/6470 [12:35<1:01:58,  1.45it/s]
2|1096|Loss: 0.8756964802742004:  17%|█▋        | 1096/6470 [12:36<1:00:29,  1.48it/s]
2|1097|Loss: 0.689707338809967:  17%|█▋        | 1096/6470 [12:36<1:00:29,  1.48it/s] 
2|1097|Loss: 0.689707338809967:  17%|█▋        | 1097/6470 [12:37<1:06:23,  1.35it/s]
2|1098|Loss: 0.8531181216239929:  17%|█▋        | 1097/6470 [12:37<1:06:23,  1.35it/s]
2|1098|Loss: 0.8531181216239929:  17%|█▋        | 1098/6470 [12:37<1:07:37,  1.32it/s]
2|1099|Loss: 0.7780345678329468:  17%|█▋        | 1098/6470 [12:38<1:07:37,  1.32it/s]
2|1099|Loss: 0.7780345678329468:  17%|█▋        | 1099/6470 [12:38<1:02:21,  1.44it/s]
2|1100|Loss: 0.8685113787651062:  17%|█▋        | 1099/6470 [12:38<1:02:21,  1.44it/s]
2|1100|Loss: 0.8685113787651062:  17%|█▋        | 1100/6470 [12:39<1:03:26,  1.41it/s]
2|1101|Loss: 0.8595718741416931:  17%|█▋        | 1100/6470 [12:39<1:03:26,  1.41it/s]
2|1101|Loss: 0.8595718741416931:  17%|█▋        | 1101/6470 [12:39<59:13,  1.51it/s]  
2|1102|Loss: 0.8478148579597473:  17%|█▋        | 1101/6470 [12:40<59:13,  1.51it/s]
2|1102|Loss: 0.8478148579597473:  17%|█▋        | 1102/6470 [12:40<54:24,  1.64it/s]
2|1103|Loss: 0.9931192398071289:  17%|█▋        | 1102/6470 [12:40<54:24,  1.64it/s]
2|1103|Loss: 0.9931192398071289:  17%|█▋        | 1103/6470 [12:40<56:16,  1.59it/s]
2|1104|Loss: 0.8441643118858337:  17%|█▋        | 1103/6470 [12:41<56:16,  1.59it/s]
2|1104|Loss: 0.8441643118858337:  17%|█▋        | 1104/6470 [12:41<58:43,  1.52it/s]
2|1105|Loss: 0.937604546546936:  17%|█▋        | 1104/6470 [12:42<58:43,  1.52it/s] 
2|1105|Loss: 0.937604546546936:  17%|█▋        | 1105/6470 [12:42<58:42,  1.52it/s]
2|1106|Loss: 0.8656495809555054:  17%|█▋        | 1105/6470 [12:42<58:42,  1.52it/s]
2|1106|Loss: 0.8656495809555054:  17%|█▋        | 1106/6470 [12:42<57:07,  1.57it/s]
2|1107|Loss: 0.9639009833335876:  17%|█▋        | 1106/6470 [12:43<57:07,  1.57it/s]
2|1107|Loss: 0.9639009833335876:  17%|█▋        | 1107/6470 [12:43<56:55,  1.57it/s]
2|1108|Loss: 0.8472130298614502:  17%|█▋        | 1107/6470 [12:43<56:55,  1.57it/s]
2|1108|Loss: 0.8472130298614502:  17%|█▋        | 1108/6470 [12:44<59:01,  1.51it/s]
2|1109|Loss: 0.9233671426773071:  17%|█▋        | 1108/6470 [12:44<59:01,  1.51it/s]
2|1109|Loss: 0.9233671426773071:  17%|█▋        | 1109/6470 [12:44<58:29,  1.53it/s]
2|1110|Loss: 0.949682891368866:  17%|█▋        | 1109/6470 [12:45<58:29,  1.53it/s] 
2|1110|Loss: 0.949682891368866:  17%|█▋        | 1110/6470 [12:45<56:35,  1.58it/s]
2|1111|Loss: 0.8884332180023193:  17%|█▋        | 1110/6470 [12:45<56:35,  1.58it/s]
2|1111|Loss: 0.8884332180023193:  17%|█▋        | 1111/6470 [12:46<1:00:34,  1.47it/s]
2|1112|Loss: 0.8847627639770508:  17%|█▋        | 1111/6470 [12:46<1:00:34,  1.47it/s]
2|1112|Loss: 0.8847627639770508:  17%|█▋        | 1112/6470 [12:47<1:05:36,  1.36it/s]
2|1113|Loss: 0.9255720973014832:  17%|█▋        | 1112/6470 [12:47<1:05:36,  1.36it/s]
2|1113|Loss: 0.9255720973014832:  17%|█▋        | 1113/6470 [12:48<1:08:36,  1.30it/s]
2|1114|Loss: 0.9718083143234253:  17%|█▋        | 1113/6470 [12:48<1:08:36,  1.30it/s]
2|1114|Loss: 0.9718083143234253:  17%|█▋        | 1114/6470 [12:48<1:06:02,  1.35it/s]
2|1115|Loss: 0.8964316248893738:  17%|█▋        | 1114/6470 [12:49<1:06:02,  1.35it/s]
2|1115|Loss: 0.8964316248893738:  17%|█▋        | 1115/6470 [12:49<1:02:52,  1.42it/s]
2|1116|Loss: 0.9948145747184753:  17%|█▋        | 1115/6470 [12:49<1:02:52,  1.42it/s]
2|1116|Loss: 0.9948145747184753:  17%|█▋        | 1116/6470 [12:49<1:01:04,  1.46it/s]
2|1117|Loss: 0.7090097069740295:  17%|█▋        | 1116/6470 [12:50<1:01:04,  1.46it/s]
2|1117|Loss: 0.7090097069740295:  17%|█▋        | 1117/6470 [12:50<55:52,  1.60it/s]  
2|1118|Loss: 0.846376359462738:  17%|█▋        | 1117/6470 [12:50<55:52,  1.60it/s] 
2|1118|Loss: 0.846376359462738:  17%|█▋        | 1118/6470 [12:51<56:25,  1.58it/s]
2|1119|Loss: 0.8570247888565063:  17%|█▋        | 1118/6470 [12:51<56:25,  1.58it/s]
2|1119|Loss: 0.8570247888565063:  17%|█▋        | 1119/6470 [12:51<53:05,  1.68it/s]
2|1120|Loss: 0.9650647640228271:  17%|█▋        | 1119/6470 [12:51<53:05,  1.68it/s]
2|1120|Loss: 0.9650647640228271:  17%|█▋        | 1120/6470 [12:52<54:21,  1.64it/s]
2|1121|Loss: 0.8986299633979797:  17%|█▋        | 1120/6470 [12:52<54:21,  1.64it/s]
2|1121|Loss: 0.8986299633979797:  17%|█▋        | 1121/6470 [12:52<57:52,  1.54it/s]
2|1122|Loss: 0.9507534503936768:  17%|█▋        | 1121/6470 [12:53<57:52,  1.54it/s]
2|1122|Loss: 0.9507534503936768:  17%|█▋        | 1122/6470 [12:53<59:42,  1.49it/s]
2|1123|Loss: 0.836151123046875:  17%|█▋        | 1122/6470 [12:54<59:42,  1.49it/s] 
2|1123|Loss: 0.836151123046875:  17%|█▋        | 1123/6470 [12:54<1:00:33,  1.47it/s]
2|1124|Loss: 0.7676224112510681:  17%|█▋        | 1123/6470 [12:54<1:00:33,  1.47it/s]
2|1124|Loss: 0.7676224112510681:  17%|█▋        | 1124/6470 [12:55<59:53,  1.49it/s]  
2|1125|Loss: 0.9760799407958984:  17%|█▋        | 1124/6470 [12:55<59:53,  1.49it/s]
2|1125|Loss: 0.9760799407958984:  17%|█▋        | 1125/6470 [12:55<1:04:02,  1.39it/s]
2|1126|Loss: 0.8078709840774536:  17%|█▋        | 1125/6470 [12:56<1:04:02,  1.39it/s]
2|1126|Loss: 0.8078709840774536:  17%|█▋        | 1126/6470 [12:56<1:03:01,  1.41it/s]
2|1127|Loss: 0.9489056468009949:  17%|█▋        | 1126/6470 [12:56<1:03:01,  1.41it/s]
2|1127|Loss: 0.9489056468009949:  17%|█▋        | 1127/6470 [12:57<57:03,  1.56it/s]  
2|1128|Loss: 0.7755289077758789:  17%|█▋        | 1127/6470 [12:57<57:03,  1.56it/s]
2|1128|Loss: 0.7755289077758789:  17%|█▋        | 1128/6470 [12:57<57:01,  1.56it/s]
2|1129|Loss: 1.0529812574386597:  17%|█▋        | 1128/6470 [12:58<57:01,  1.56it/s]
2|1129|Loss: 1.0529812574386597:  17%|█▋        | 1129/6470 [12:58<58:47,  1.51it/s]
2|1130|Loss: 0.9183956384658813:  17%|█▋        | 1129/6470 [12:58<58:47,  1.51it/s]
2|1130|Loss: 0.9183956384658813:  17%|█▋        | 1130/6470 [12:59<1:00:34,  1.47it/s]
2|1131|Loss: 0.7967167496681213:  17%|█▋        | 1130/6470 [12:59<1:00:34,  1.47it/s]
2|1131|Loss: 0.7967167496681213:  17%|█▋        | 1131/6470 [12:59<1:02:20,  1.43it/s]
2|1132|Loss: 0.945921778678894:  17%|█▋        | 1131/6470 [13:00<1:02:20,  1.43it/s] 
2|1132|Loss: 0.945921778678894:  17%|█▋        | 1132/6470 [13:00<55:26,  1.60it/s]  
2|1133|Loss: 0.7421301603317261:  17%|█▋        | 1132/6470 [13:00<55:26,  1.60it/s]
2|1133|Loss: 0.7421301603317261:  18%|█▊        | 1133/6470 [13:00<56:57,  1.56it/s]
2|1134|Loss: 0.880304217338562:  18%|█▊        | 1133/6470 [13:01<56:57,  1.56it/s] 
2|1134|Loss: 0.880304217338562:  18%|█▊        | 1134/6470 [13:01<57:44,  1.54it/s]
2|1135|Loss: 0.8390710949897766:  18%|█▊        | 1134/6470 [13:02<57:44,  1.54it/s]
2|1135|Loss: 0.8390710949897766:  18%|█▊        | 1135/6470 [13:02<58:45,  1.51it/s]
2|1136|Loss: 0.8274927139282227:  18%|█▊        | 1135/6470 [13:02<58:45,  1.51it/s]
2|1136|Loss: 0.8274927139282227:  18%|█▊        | 1136/6470 [13:03<1:03:50,  1.39it/s]
2|1137|Loss: 0.9462809562683105:  18%|█▊        | 1136/6470 [13:03<1:03:50,  1.39it/s]
2|1137|Loss: 0.9462809562683105:  18%|█▊        | 1137/6470 [13:03<1:05:20,  1.36it/s]
2|1138|Loss: 0.8187898993492126:  18%|█▊        | 1137/6470 [13:04<1:05:20,  1.36it/s]
2|1138|Loss: 0.8187898993492126:  18%|█▊        | 1138/6470 [13:04<59:12,  1.50it/s]  
2|1139|Loss: 0.8812156319618225:  18%|█▊        | 1138/6470 [13:04<59:12,  1.50it/s]
2|1139|Loss: 0.8812156319618225:  18%|█▊        | 1139/6470 [13:05<56:16,  1.58it/s]
2|1140|Loss: 0.772268533706665:  18%|█▊        | 1139/6470 [13:05<56:16,  1.58it/s] 
2|1140|Loss: 0.772268533706665:  18%|█▊        | 1140/6470 [13:05<53:18,  1.67it/s]
2|1141|Loss: 0.9303589463233948:  18%|█▊        | 1140/6470 [13:05<53:18,  1.67it/s]
2|1141|Loss: 0.9303589463233948:  18%|█▊        | 1141/6470 [13:06<51:45,  1.72it/s]
2|1142|Loss: 0.9880027770996094:  18%|█▊        | 1141/6470 [13:06<51:45,  1.72it/s]
2|1142|Loss: 0.9880027770996094:  18%|█▊        | 1142/6470 [13:06<50:29,  1.76it/s]
2|1143|Loss: 1.0465567111968994:  18%|█▊        | 1142/6470 [13:06<50:29,  1.76it/s]
2|1143|Loss: 1.0465567111968994:  18%|█▊        | 1143/6470 [13:07<51:13,  1.73it/s]
2|1144|Loss: 0.8997184038162231:  18%|█▊        | 1143/6470 [13:07<51:13,  1.73it/s]
2|1144|Loss: 0.8997184038162231:  18%|█▊        | 1144/6470 [13:07<54:56,  1.62it/s]
2|1145|Loss: 0.8154709339141846:  18%|█▊        | 1144/6470 [13:08<54:56,  1.62it/s]
2|1145|Loss: 0.8154709339141846:  18%|█▊        | 1145/6470 [13:08<1:00:03,  1.48it/s]
2|1146|Loss: 0.9705091118812561:  18%|█▊        | 1145/6470 [13:09<1:00:03,  1.48it/s]
2|1146|Loss: 0.9705091118812561:  18%|█▊        | 1146/6470 [13:09<58:56,  1.51it/s]  
2|1147|Loss: 0.9247342944145203:  18%|█▊        | 1146/6470 [13:09<58:56,  1.51it/s]
2|1147|Loss: 0.9247342944145203:  18%|█▊        | 1147/6470 [13:09<56:56,  1.56it/s]
2|1148|Loss: 0.7985102534294128:  18%|█▊        | 1147/6470 [13:10<56:56,  1.56it/s]
2|1148|Loss: 0.7985102534294128:  18%|█▊        | 1148/6470 [13:10<57:59,  1.53it/s]
2|1149|Loss: 0.8018839359283447:  18%|█▊        | 1148/6470 [13:11<57:59,  1.53it/s]
2|1149|Loss: 0.8018839359283447:  18%|█▊        | 1149/6470 [13:11<56:50,  1.56it/s]
2|1150|Loss: 0.8771836757659912:  18%|█▊        | 1149/6470 [13:11<56:50,  1.56it/s]
2|1150|Loss: 0.8771836757659912:  18%|█▊        | 1150/6470 [13:11<56:03,  1.58it/s]
2|1151|Loss: 0.7955152988433838:  18%|█▊        | 1150/6470 [13:12<56:03,  1.58it/s]
2|1151|Loss: 0.7955152988433838:  18%|█▊        | 1151/6470 [13:12<1:00:50,  1.46it/s]
2|1152|Loss: 0.8582340478897095:  18%|█▊        | 1151/6470 [13:13<1:00:50,  1.46it/s]
2|1152|Loss: 0.8582340478897095:  18%|█▊        | 1152/6470 [13:13<55:02,  1.61it/s]  
2|1153|Loss: 0.9547041058540344:  18%|█▊        | 1152/6470 [13:13<55:02,  1.61it/s]
2|1153|Loss: 0.9547041058540344:  18%|█▊        | 1153/6470 [13:13<56:38,  1.56it/s]
2|1154|Loss: 0.8660866618156433:  18%|█▊        | 1153/6470 [13:14<56:38,  1.56it/s]
2|1154|Loss: 0.8660866618156433:  18%|█▊        | 1154/6470 [13:14<1:00:43,  1.46it/s]
2|1155|Loss: 0.9427567720413208:  18%|█▊        | 1154/6470 [13:14<1:00:43,  1.46it/s]
2|1155|Loss: 0.9427567720413208:  18%|█▊        | 1155/6470 [13:15<56:20,  1.57it/s]  
2|1156|Loss: 0.801909327507019:  18%|█▊        | 1155/6470 [13:15<56:20,  1.57it/s] 
2|1156|Loss: 0.801909327507019:  18%|█▊        | 1156/6470 [13:15<51:54,  1.71it/s]
2|1157|Loss: 0.9895724654197693:  18%|█▊        | 1156/6470 [13:16<51:54,  1.71it/s]
2|1157|Loss: 0.9895724654197693:  18%|█▊        | 1157/6470 [13:16<55:32,  1.59it/s]
2|1158|Loss: 0.92486572265625:  18%|█▊        | 1157/6470 [13:16<55:32,  1.59it/s]  
2|1158|Loss: 0.92486572265625:  18%|█▊        | 1158/6470 [13:17<1:01:23,  1.44it/s]
2|1159|Loss: 0.8560986518859863:  18%|█▊        | 1158/6470 [13:17<1:01:23,  1.44it/s]
2|1159|Loss: 0.8560986518859863:  18%|█▊        | 1159/6470 [13:18<1:04:44,  1.37it/s]
2|1160|Loss: 0.7602778673171997:  18%|█▊        | 1159/6470 [13:18<1:04:44,  1.37it/s]
2|1160|Loss: 0.7602778673171997:  18%|█▊        | 1160/6470 [13:18<1:07:07,  1.32it/s]
2|1161|Loss: 0.8135506510734558:  18%|█▊        | 1160/6470 [13:19<1:07:07,  1.32it/s]
2|1161|Loss: 0.8135506510734558:  18%|█▊        | 1161/6470 [13:19<1:06:48,  1.32it/s]
2|1162|Loss: 0.8409737944602966:  18%|█▊        | 1161/6470 [13:20<1:06:48,  1.32it/s]
2|1162|Loss: 0.8409737944602966:  18%|█▊        | 1162/6470 [13:20<1:08:03,  1.30it/s]
2|1163|Loss: 0.8550588488578796:  18%|█▊        | 1162/6470 [13:20<1:08:03,  1.30it/s]
2|1163|Loss: 0.8550588488578796:  18%|█▊        | 1163/6470 [13:21<1:05:14,  1.36it/s]
2|1164|Loss: 0.8054133057594299:  18%|█▊        | 1163/6470 [13:21<1:05:14,  1.36it/s]
2|1164|Loss: 0.8054133057594299:  18%|█▊        | 1164/6470 [13:21<1:00:36,  1.46it/s]
2|1165|Loss: 0.803767204284668:  18%|█▊        | 1164/6470 [13:21<1:00:36,  1.46it/s] 
2|1165|Loss: 0.803767204284668:  18%|█▊        | 1165/6470 [13:22<56:21,  1.57it/s]  
2|1166|Loss: 1.0092132091522217:  18%|█▊        | 1165/6470 [13:22<56:21,  1.57it/s]
2|1166|Loss: 1.0092132091522217:  18%|█▊        | 1166/6470 [13:22<1:00:45,  1.46it/s]
2|1167|Loss: 0.770215630531311:  18%|█▊        | 1166/6470 [13:23<1:00:45,  1.46it/s] 
2|1167|Loss: 0.770215630531311:  18%|█▊        | 1167/6470 [13:23<1:03:01,  1.40it/s]
2|1168|Loss: 1.1292165517807007:  18%|█▊        | 1167/6470 [13:24<1:03:01,  1.40it/s]
2|1168|Loss: 1.1292165517807007:  18%|█▊        | 1168/6470 [13:24<1:07:38,  1.31it/s]
2|1169|Loss: 0.7816831469535828:  18%|█▊        | 1168/6470 [13:25<1:07:38,  1.31it/s]
2|1169|Loss: 0.7816831469535828:  18%|█▊        | 1169/6470 [13:25<1:07:13,  1.31it/s]
2|1170|Loss: 0.7735763192176819:  18%|█▊        | 1169/6470 [13:25<1:07:13,  1.31it/s]
2|1170|Loss: 0.7735763192176819:  18%|█▊        | 1170/6470 [13:26<1:06:18,  1.33it/s]
2|1171|Loss: 0.7504509687423706:  18%|█▊        | 1170/6470 [13:26<1:06:18,  1.33it/s]
2|1171|Loss: 0.7504509687423706:  18%|█▊        | 1171/6470 [13:26<1:06:16,  1.33it/s]
2|1172|Loss: 0.8434745669364929:  18%|█▊        | 1171/6470 [13:27<1:06:16,  1.33it/s]
2|1172|Loss: 0.8434745669364929:  18%|█▊        | 1172/6470 [13:27<1:05:21,  1.35it/s]
2|1173|Loss: 1.0278018712997437:  18%|█▊        | 1172/6470 [13:27<1:05:21,  1.35it/s]
2|1173|Loss: 1.0278018712997437:  18%|█▊        | 1173/6470 [13:28<1:02:40,  1.41it/s]
2|1174|Loss: 0.625789225101471:  18%|█▊        | 1173/6470 [13:28<1:02:40,  1.41it/s] 
2|1174|Loss: 0.625789225101471:  18%|█▊        | 1174/6470 [13:28<1:01:55,  1.43it/s]
2|1175|Loss: 0.8527412414550781:  18%|█▊        | 1174/6470 [13:29<1:01:55,  1.43it/s]
2|1175|Loss: 0.8527412414550781:  18%|█▊        | 1175/6470 [13:29<59:26,  1.48it/s]  
2|1176|Loss: 0.6539012789726257:  18%|█▊        | 1175/6470 [13:29<59:26,  1.48it/s]
2|1176|Loss: 0.6539012789726257:  18%|█▊        | 1176/6470 [13:30<57:42,  1.53it/s]
2|1177|Loss: 0.8194881081581116:  18%|█▊        | 1176/6470 [13:30<57:42,  1.53it/s]
2|1177|Loss: 0.8194881081581116:  18%|█▊        | 1177/6470 [13:30<1:00:38,  1.45it/s]
2|1178|Loss: 0.8262083530426025:  18%|█▊        | 1177/6470 [13:31<1:00:38,  1.45it/s]
2|1178|Loss: 0.8262083530426025:  18%|█▊        | 1178/6470 [13:31<1:02:01,  1.42it/s]
2|1179|Loss: 0.9960300922393799:  18%|█▊        | 1178/6470 [13:31<1:02:01,  1.42it/s]
2|1179|Loss: 0.9960300922393799:  18%|█▊        | 1179/6470 [13:32<1:02:07,  1.42it/s]
2|1180|Loss: 0.8967676162719727:  18%|█▊        | 1179/6470 [13:32<1:02:07,  1.42it/s]
2|1180|Loss: 0.8967676162719727:  18%|█▊        | 1180/6470 [13:33<1:05:12,  1.35it/s]
2|1181|Loss: 0.7839308381080627:  18%|█▊        | 1180/6470 [13:33<1:05:12,  1.35it/s]
2|1181|Loss: 0.7839308381080627:  18%|█▊        | 1181/6470 [13:33<1:01:10,  1.44it/s]
2|1182|Loss: 0.8625384569168091:  18%|█▊        | 1181/6470 [13:34<1:01:10,  1.44it/s]
2|1182|Loss: 0.8625384569168091:  18%|█▊        | 1182/6470 [13:34<1:04:17,  1.37it/s]
2|1183|Loss: 0.7154653668403625:  18%|█▊        | 1182/6470 [13:34<1:04:17,  1.37it/s]
2|1183|Loss: 0.7154653668403625:  18%|█▊        | 1183/6470 [13:35<1:01:41,  1.43it/s]
2|1184|Loss: 0.9042551517486572:  18%|█▊        | 1183/6470 [13:35<1:01:41,  1.43it/s]
2|1184|Loss: 0.9042551517486572:  18%|█▊        | 1184/6470 [13:35<56:55,  1.55it/s]  
2|1185|Loss: 1.0528370141983032:  18%|█▊        | 1184/6470 [13:36<56:55,  1.55it/s]
2|1185|Loss: 1.0528370141983032:  18%|█▊        | 1185/6470 [13:36<57:24,  1.53it/s]
2|1186|Loss: 0.7853726148605347:  18%|█▊        | 1185/6470 [13:36<57:24,  1.53it/s]
2|1186|Loss: 0.7853726148605347:  18%|█▊        | 1186/6470 [13:37<1:03:10,  1.39it/s]
2|1187|Loss: 0.974155843257904:  18%|█▊        | 1186/6470 [13:37<1:03:10,  1.39it/s] 
2|1187|Loss: 0.974155843257904:  18%|█▊        | 1187/6470 [13:37<1:02:41,  1.40it/s]
2|1188|Loss: 0.8283041715621948:  18%|█▊        | 1187/6470 [13:38<1:02:41,  1.40it/s]
2|1188|Loss: 0.8283041715621948:  18%|█▊        | 1188/6470 [13:38<57:59,  1.52it/s]  
2|1189|Loss: 0.905421257019043:  18%|█▊        | 1188/6470 [13:38<57:59,  1.52it/s] 
2|1189|Loss: 0.905421257019043:  18%|█▊        | 1189/6470 [13:39<1:00:38,  1.45it/s]
2|1190|Loss: 0.8897011876106262:  18%|█▊        | 1189/6470 [13:39<1:00:38,  1.45it/s]
2|1190|Loss: 0.8897011876106262:  18%|█▊        | 1190/6470 [13:40<1:03:45,  1.38it/s]
2|1191|Loss: 0.8334341049194336:  18%|█▊        | 1190/6470 [13:40<1:03:45,  1.38it/s]
2|1191|Loss: 0.8334341049194336:  18%|█▊        | 1191/6470 [13:40<1:06:26,  1.32it/s]
2|1192|Loss: 0.8479543924331665:  18%|█▊        | 1191/6470 [13:41<1:06:26,  1.32it/s]
2|1192|Loss: 0.8479543924331665:  18%|█▊        | 1192/6470 [13:41<1:05:53,  1.33it/s]
2|1193|Loss: 0.7575702667236328:  18%|█▊        | 1192/6470 [13:41<1:05:53,  1.33it/s]
2|1193|Loss: 0.7575702667236328:  18%|█▊        | 1193/6470 [13:42<1:01:58,  1.42it/s]
2|1194|Loss: 0.8152989149093628:  18%|█▊        | 1193/6470 [13:42<1:01:58,  1.42it/s]
2|1194|Loss: 0.8152989149093628:  18%|█▊        | 1194/6470 [13:42<1:01:51,  1.42it/s]
2|1195|Loss: 0.7505435347557068:  18%|█▊        | 1194/6470 [13:43<1:01:51,  1.42it/s]
2|1195|Loss: 0.7505435347557068:  18%|█▊        | 1195/6470 [13:43<59:04,  1.49it/s]  
2|1196|Loss: 0.7655441761016846:  18%|█▊        | 1195/6470 [13:43<59:04,  1.49it/s]
2|1196|Loss: 0.7655441761016846:  18%|█▊        | 1196/6470 [13:44<1:02:27,  1.41it/s]
2|1197|Loss: 0.912273645401001:  18%|█▊        | 1196/6470 [13:44<1:02:27,  1.41it/s] 
2|1197|Loss: 0.912273645401001:  19%|█▊        | 1197/6470 [13:44<1:01:49,  1.42it/s]
2|1198|Loss: 1.0147061347961426:  19%|█▊        | 1197/6470 [13:45<1:01:49,  1.42it/s]
2|1198|Loss: 1.0147061347961426:  19%|█▊        | 1198/6470 [13:45<59:43,  1.47it/s]  
2|1199|Loss: 0.8721476793289185:  19%|█▊        | 1198/6470 [13:45<59:43,  1.47it/s]
2|1199|Loss: 0.8721476793289185:  19%|█▊        | 1199/6470 [13:46<58:33,  1.50it/s]
2|1200|Loss: 0.916183590888977:  19%|█▊        | 1199/6470 [13:46<58:33,  1.50it/s] 
2|1200|Loss: 0.916183590888977:  19%|█▊        | 1200/6470 [13:46<55:15,  1.59it/s]
2|1201|Loss: 0.8601344227790833:  19%|█▊        | 1200/6470 [13:47<55:15,  1.59it/s]
2|1201|Loss: 0.8601344227790833:  19%|█▊        | 1201/6470 [13:47<56:17,  1.56it/s]
2|1202|Loss: 0.782484233379364:  19%|█▊        | 1201/6470 [13:47<56:17,  1.56it/s] 
2|1202|Loss: 0.782484233379364:  19%|█▊        | 1202/6470 [13:48<1:00:19,  1.46it/s]
2|1203|Loss: 0.7851460576057434:  19%|█▊        | 1202/6470 [13:48<1:00:19,  1.46it/s]
2|1203|Loss: 0.7851460576057434:  19%|█▊        | 1203/6470 [13:48<1:00:09,  1.46it/s]
2|1204|Loss: 0.7844941020011902:  19%|█▊        | 1203/6470 [13:49<1:00:09,  1.46it/s]
2|1204|Loss: 0.7844941020011902:  19%|█▊        | 1204/6470 [13:49<58:33,  1.50it/s]  
2|1205|Loss: 0.8408812880516052:  19%|█▊        | 1204/6470 [13:49<58:33,  1.50it/s]
2|1205|Loss: 0.8408812880516052:  19%|█▊        | 1205/6470 [13:50<55:01,  1.59it/s]
2|1206|Loss: 0.9534670114517212:  19%|█▊        | 1205/6470 [13:50<55:01,  1.59it/s]
2|1206|Loss: 0.9534670114517212:  19%|█▊        | 1206/6470 [13:50<58:06,  1.51it/s]
2|1207|Loss: 0.7174627780914307:  19%|█▊        | 1206/6470 [13:51<58:06,  1.51it/s]
2|1207|Loss: 0.7174627780914307:  19%|█▊        | 1207/6470 [13:51<58:04,  1.51it/s]
2|1208|Loss: 0.9021298289299011:  19%|█▊        | 1207/6470 [13:51<58:04,  1.51it/s]
2|1208|Loss: 0.9021298289299011:  19%|█▊        | 1208/6470 [13:51<54:04,  1.62it/s]
2|1209|Loss: 1.0098974704742432:  19%|█▊        | 1208/6470 [13:52<54:04,  1.62it/s]
2|1209|Loss: 1.0098974704742432:  19%|█▊        | 1209/6470 [13:52<53:55,  1.63it/s]
2|1210|Loss: 0.5903149843215942:  19%|█▊        | 1209/6470 [13:52<53:55,  1.63it/s]
2|1210|Loss: 0.5903149843215942:  19%|█▊        | 1210/6470 [13:53<54:05,  1.62it/s]
2|1211|Loss: 0.869361937046051:  19%|█▊        | 1210/6470 [13:53<54:05,  1.62it/s] 
2|1211|Loss: 0.869361937046051:  19%|█▊        | 1211/6470 [13:53<53:24,  1.64it/s]
2|1212|Loss: 0.8981267809867859:  19%|█▊        | 1211/6470 [13:54<53:24,  1.64it/s]
2|1212|Loss: 0.8981267809867859:  19%|█▊        | 1212/6470 [13:54<55:51,  1.57it/s]
2|1213|Loss: 0.8846530318260193:  19%|█▊        | 1212/6470 [13:54<55:51,  1.57it/s]
2|1213|Loss: 0.8846530318260193:  19%|█▊        | 1213/6470 [13:55<1:01:21,  1.43it/s]
2|1214|Loss: 0.8078554272651672:  19%|█▊        | 1213/6470 [13:55<1:01:21,  1.43it/s]
2|1214|Loss: 0.8078554272651672:  19%|█▉        | 1214/6470 [13:56<1:00:45,  1.44it/s]
2|1215|Loss: 0.8090539574623108:  19%|█▉        | 1214/6470 [13:56<1:00:45,  1.44it/s]
2|1215|Loss: 0.8090539574623108:  19%|█▉        | 1215/6470 [13:56<1:01:32,  1.42it/s]
2|1216|Loss: 0.7877475619316101:  19%|█▉        | 1215/6470 [13:57<1:01:32,  1.42it/s]
2|1216|Loss: 0.7877475619316101:  19%|█▉        | 1216/6470 [13:57<54:18,  1.61it/s]  
2|1217|Loss: 0.8503903150558472:  19%|█▉        | 1216/6470 [13:57<54:18,  1.61it/s]
2|1217|Loss: 0.8503903150558472:  19%|█▉        | 1217/6470 [13:57<53:11,  1.65it/s]
2|1218|Loss: 0.8525686264038086:  19%|█▉        | 1217/6470 [13:58<53:11,  1.65it/s]
2|1218|Loss: 0.8525686264038086:  19%|█▉        | 1218/6470 [13:58<59:58,  1.46it/s]
2|1219|Loss: 1.016409158706665:  19%|█▉        | 1218/6470 [13:59<59:58,  1.46it/s] 
2|1219|Loss: 1.016409158706665:  19%|█▉        | 1219/6470 [13:59<1:04:55,  1.35it/s]
2|1220|Loss: 0.8451744914054871:  19%|█▉        | 1219/6470 [13:59<1:04:55,  1.35it/s]
2|1220|Loss: 0.8451744914054871:  19%|█▉        | 1220/6470 [14:00<1:03:03,  1.39it/s]
2|1221|Loss: 0.7697639465332031:  19%|█▉        | 1220/6470 [14:00<1:03:03,  1.39it/s]
2|1221|Loss: 0.7697639465332031:  19%|█▉        | 1221/6470 [14:00<1:04:45,  1.35it/s]
2|1222|Loss: 1.0373222827911377:  19%|█▉        | 1221/6470 [14:01<1:04:45,  1.35it/s]
2|1222|Loss: 1.0373222827911377:  19%|█▉        | 1222/6470 [14:01<1:05:33,  1.33it/s]
2|1223|Loss: 0.6193526983261108:  19%|█▉        | 1222/6470 [14:02<1:05:33,  1.33it/s]
2|1223|Loss: 0.6193526983261108:  19%|█▉        | 1223/6470 [14:02<1:05:11,  1.34it/s]
2|1224|Loss: 0.9096217751502991:  19%|█▉        | 1223/6470 [14:02<1:05:11,  1.34it/s]
2|1224|Loss: 0.9096217751502991:  19%|█▉        | 1224/6470 [14:03<1:04:51,  1.35it/s]
2|1225|Loss: 0.7500377893447876:  19%|█▉        | 1224/6470 [14:03<1:04:51,  1.35it/s]
2|1225|Loss: 0.7500377893447876:  19%|█▉        | 1225/6470 [14:03<1:00:43,  1.44it/s]
2|1226|Loss: 0.9436948895454407:  19%|█▉        | 1225/6470 [14:04<1:00:43,  1.44it/s]
2|1226|Loss: 0.9436948895454407:  19%|█▉        | 1226/6470 [14:04<1:00:06,  1.45it/s]
2|1227|Loss: 0.8100278973579407:  19%|█▉        | 1226/6470 [14:04<1:00:06,  1.45it/s]
2|1227|Loss: 0.8100278973579407:  19%|█▉        | 1227/6470 [14:05<59:25,  1.47it/s]  
2|1228|Loss: 0.9481480717658997:  19%|█▉        | 1227/6470 [14:05<59:25,  1.47it/s]
2|1228|Loss: 0.9481480717658997:  19%|█▉        | 1228/6470 [14:05<55:45,  1.57it/s]
2|1229|Loss: 0.8335569500923157:  19%|█▉        | 1228/6470 [14:05<55:45,  1.57it/s]
2|1229|Loss: 0.8335569500923157:  19%|█▉        | 1229/6470 [14:06<53:44,  1.63it/s]
2|1230|Loss: 0.7996829152107239:  19%|█▉        | 1229/6470 [14:06<53:44,  1.63it/s]
2|1230|Loss: 0.7996829152107239:  19%|█▉        | 1230/6470 [14:07<58:10,  1.50it/s]
2|1231|Loss: 0.8618231415748596:  19%|█▉        | 1230/6470 [14:07<58:10,  1.50it/s]
2|1231|Loss: 0.8618231415748596:  19%|█▉        | 1231/6470 [14:07<1:03:03,  1.38it/s]
2|1232|Loss: 0.8498767614364624:  19%|█▉        | 1231/6470 [14:08<1:03:03,  1.38it/s]
2|1232|Loss: 0.8498767614364624:  19%|█▉        | 1232/6470 [14:08<1:06:14,  1.32it/s]
2|1233|Loss: 0.8124937415122986:  19%|█▉        | 1232/6470 [14:09<1:06:14,  1.32it/s]
2|1233|Loss: 0.8124937415122986:  19%|█▉        | 1233/6470 [14:09<1:02:16,  1.40it/s]
2|1234|Loss: 0.8086007237434387:  19%|█▉        | 1233/6470 [14:09<1:02:16,  1.40it/s]
2|1234|Loss: 0.8086007237434387:  19%|█▉        | 1234/6470 [14:10<1:01:49,  1.41it/s]
2|1235|Loss: 0.9105175733566284:  19%|█▉        | 1234/6470 [14:10<1:01:49,  1.41it/s]
2|1235|Loss: 0.9105175733566284:  19%|█▉        | 1235/6470 [14:10<1:00:31,  1.44it/s]
2|1236|Loss: 1.0208622217178345:  19%|█▉        | 1235/6470 [14:11<1:00:31,  1.44it/s]
2|1236|Loss: 1.0208622217178345:  19%|█▉        | 1236/6470 [14:11<1:03:36,  1.37it/s]
2|1237|Loss: 0.893976092338562:  19%|█▉        | 1236/6470 [14:11<1:03:36,  1.37it/s] 
2|1237|Loss: 0.893976092338562:  19%|█▉        | 1237/6470 [14:12<1:06:30,  1.31it/s]
2|1238|Loss: 0.9330499172210693:  19%|█▉        | 1237/6470 [14:12<1:06:30,  1.31it/s]
2|1238|Loss: 0.9330499172210693:  19%|█▉        | 1238/6470 [14:13<1:06:46,  1.31it/s]
2|1239|Loss: 0.962408185005188:  19%|█▉        | 1238/6470 [14:13<1:06:46,  1.31it/s] 
2|1239|Loss: 0.962408185005188:  19%|█▉        | 1239/6470 [14:13<1:08:42,  1.27it/s]
2|1240|Loss: 0.8989389538764954:  19%|█▉        | 1239/6470 [14:14<1:08:42,  1.27it/s]
2|1240|Loss: 0.8989389538764954:  19%|█▉        | 1240/6470 [14:14<1:08:17,  1.28it/s]
2|1241|Loss: 0.7692696452140808:  19%|█▉        | 1240/6470 [14:15<1:08:17,  1.28it/s]
2|1241|Loss: 0.7692696452140808:  19%|█▉        | 1241/6470 [14:15<1:10:55,  1.23it/s]
2|1242|Loss: 0.7735480070114136:  19%|█▉        | 1241/6470 [14:16<1:10:55,  1.23it/s]
2|1242|Loss: 0.7735480070114136:  19%|█▉        | 1242/6470 [14:16<1:07:34,  1.29it/s]
2|1243|Loss: 0.9310060739517212:  19%|█▉        | 1242/6470 [14:16<1:07:34,  1.29it/s]
2|1243|Loss: 0.9310060739517212:  19%|█▉        | 1243/6470 [14:16<1:03:28,  1.37it/s]
2|1244|Loss: 0.7325844764709473:  19%|█▉        | 1243/6470 [14:17<1:03:28,  1.37it/s]
2|1244|Loss: 0.7325844764709473:  19%|█▉        | 1244/6470 [14:17<59:46,  1.46it/s]  
2|1245|Loss: 0.8725809454917908:  19%|█▉        | 1244/6470 [14:17<59:46,  1.46it/s]
2|1245|Loss: 0.8725809454917908:  19%|█▉        | 1245/6470 [14:18<58:16,  1.49it/s]
2|1246|Loss: 1.0636868476867676:  19%|█▉        | 1245/6470 [14:18<58:16,  1.49it/s]
2|1246|Loss: 1.0636868476867676:  19%|█▉        | 1246/6470 [14:18<55:12,  1.58it/s]
2|1247|Loss: 0.869517982006073:  19%|█▉        | 1246/6470 [14:18<55:12,  1.58it/s] 
2|1247|Loss: 0.869517982006073:  19%|█▉        | 1247/6470 [14:19<53:18,  1.63it/s]
2|1248|Loss: 0.861527681350708:  19%|█▉        | 1247/6470 [14:19<53:18,  1.63it/s]
2|1248|Loss: 0.861527681350708:  19%|█▉        | 1248/6470 [14:19<53:09,  1.64it/s]
2|1249|Loss: 0.8530963659286499:  19%|█▉        | 1248/6470 [14:20<53:09,  1.64it/s]
2|1249|Loss: 0.8530963659286499:  19%|█▉        | 1249/6470 [14:20<55:13,  1.58it/s]
2|1250|Loss: 0.8362672924995422:  19%|█▉        | 1249/6470 [14:20<55:13,  1.58it/s]
2|1250|Loss: 0.8362672924995422:  19%|█▉        | 1250/6470 [14:21<59:06,  1.47it/s]
2|1251|Loss: 0.9353014826774597:  19%|█▉        | 1250/6470 [14:21<59:06,  1.47it/s]
2|1251|Loss: 0.9353014826774597:  19%|█▉        | 1251/6470 [14:22<1:04:56,  1.34it/s]
2|1252|Loss: 0.8994641900062561:  19%|█▉        | 1251/6470 [14:22<1:04:56,  1.34it/s]
2|1252|Loss: 0.8994641900062561:  19%|█▉        | 1252/6470 [14:22<1:05:47,  1.32it/s]
2|1253|Loss: 0.8960344195365906:  19%|█▉        | 1252/6470 [14:23<1:05:47,  1.32it/s]
2|1253|Loss: 0.8960344195365906:  19%|█▉        | 1253/6470 [14:23<1:04:34,  1.35it/s]
2|1254|Loss: 0.8870345950126648:  19%|█▉        | 1253/6470 [14:24<1:04:34,  1.35it/s]
2|1254|Loss: 0.8870345950126648:  19%|█▉        | 1254/6470 [14:24<1:01:40,  1.41it/s]
2|1255|Loss: 0.8261123895645142:  19%|█▉        | 1254/6470 [14:24<1:01:40,  1.41it/s]
2|1255|Loss: 0.8261123895645142:  19%|█▉        | 1255/6470 [14:24<59:49,  1.45it/s]  
2|1256|Loss: 0.8312835693359375:  19%|█▉        | 1255/6470 [14:25<59:49,  1.45it/s]
2|1256|Loss: 0.8312835693359375:  19%|█▉        | 1256/6470 [14:25<1:02:39,  1.39it/s]
2|1257|Loss: 0.9528513550758362:  19%|█▉        | 1256/6470 [14:26<1:02:39,  1.39it/s]
2|1257|Loss: 0.9528513550758362:  19%|█▉        | 1257/6470 [14:26<1:02:55,  1.38it/s]
2|1258|Loss: 0.845059871673584:  19%|█▉        | 1257/6470 [14:26<1:02:55,  1.38it/s] 
2|1258|Loss: 0.845059871673584:  19%|█▉        | 1258/6470 [14:27<1:04:17,  1.35it/s]
2|1259|Loss: 0.7438886165618896:  19%|█▉        | 1258/6470 [14:27<1:04:17,  1.35it/s]
2|1259|Loss: 0.7438886165618896:  19%|█▉        | 1259/6470 [14:27<1:01:12,  1.42it/s]
2|1260|Loss: 0.9595024585723877:  19%|█▉        | 1259/6470 [14:28<1:01:12,  1.42it/s]
2|1260|Loss: 0.9595024585723877:  19%|█▉        | 1260/6470 [14:28<57:51,  1.50it/s]  
2|1261|Loss: 0.8211771845817566:  19%|█▉        | 1260/6470 [14:28<57:51,  1.50it/s]
2|1261|Loss: 0.8211771845817566:  19%|█▉        | 1261/6470 [14:29<55:39,  1.56it/s]
2|1262|Loss: 0.8995090126991272:  19%|█▉        | 1261/6470 [14:29<55:39,  1.56it/s]
2|1262|Loss: 0.8995090126991272:  20%|█▉        | 1262/6470 [14:29<56:50,  1.53it/s]
2|1263|Loss: 0.8154401183128357:  20%|█▉        | 1262/6470 [14:30<56:50,  1.53it/s]
2|1263|Loss: 0.8154401183128357:  20%|█▉        | 1263/6470 [14:30<58:06,  1.49it/s]
2|1264|Loss: 0.8121528029441833:  20%|█▉        | 1263/6470 [14:30<58:06,  1.49it/s]
2|1264|Loss: 0.8121528029441833:  20%|█▉        | 1264/6470 [14:30<51:44,  1.68it/s]
2|1265|Loss: 0.9951106905937195:  20%|█▉        | 1264/6470 [14:31<51:44,  1.68it/s]
2|1265|Loss: 0.9951106905937195:  20%|█▉        | 1265/6470 [14:31<51:38,  1.68it/s]
2|1266|Loss: 0.7953587770462036:  20%|█▉        | 1265/6470 [14:31<51:38,  1.68it/s]
2|1266|Loss: 0.7953587770462036:  20%|█▉        | 1266/6470 [14:32<52:56,  1.64it/s]
2|1267|Loss: 0.8496106266975403:  20%|█▉        | 1266/6470 [14:32<52:56,  1.64it/s]
2|1267|Loss: 0.8496106266975403:  20%|█▉        | 1267/6470 [14:32<53:17,  1.63it/s]
2|1268|Loss: 0.9162099957466125:  20%|█▉        | 1267/6470 [14:33<53:17,  1.63it/s]
2|1268|Loss: 0.9162099957466125:  20%|█▉        | 1268/6470 [14:33<55:55,  1.55it/s]
2|1269|Loss: 0.9690887928009033:  20%|█▉        | 1268/6470 [14:33<55:55,  1.55it/s]
2|1269|Loss: 0.9690887928009033:  20%|█▉        | 1269/6470 [14:34<59:25,  1.46it/s]
2|1270|Loss: 0.8311166763305664:  20%|█▉        | 1269/6470 [14:34<59:25,  1.46it/s]
2|1270|Loss: 0.8311166763305664:  20%|█▉        | 1270/6470 [14:34<58:45,  1.48it/s]
2|1271|Loss: 0.8440976738929749:  20%|█▉        | 1270/6470 [14:35<58:45,  1.48it/s]
2|1271|Loss: 0.8440976738929749:  20%|█▉        | 1271/6470 [14:35<57:36,  1.50it/s]
2|1272|Loss: 0.8999496102333069:  20%|█▉        | 1271/6470 [14:35<57:36,  1.50it/s]
2|1272|Loss: 0.8999496102333069:  20%|█▉        | 1272/6470 [14:36<58:54,  1.47it/s]
2|1273|Loss: 1.022676944732666:  20%|█▉        | 1272/6470 [14:36<58:54,  1.47it/s] 
2|1273|Loss: 1.022676944732666:  20%|█▉        | 1273/6470 [14:36<57:03,  1.52it/s]
2|1274|Loss: 0.783549964427948:  20%|█▉        | 1273/6470 [14:37<57:03,  1.52it/s]
2|1274|Loss: 0.783549964427948:  20%|█▉        | 1274/6470 [14:37<57:33,  1.50it/s]
2|1275|Loss: 0.735196590423584:  20%|█▉        | 1274/6470 [14:37<57:33,  1.50it/s]
2|1275|Loss: 0.735196590423584:  20%|█▉        | 1275/6470 [14:38<57:29,  1.51it/s]
2|1276|Loss: 0.8016965389251709:  20%|█▉        | 1275/6470 [14:38<57:29,  1.51it/s]
2|1276|Loss: 0.8016965389251709:  20%|█▉        | 1276/6470 [14:38<1:00:52,  1.42it/s]
2|1277|Loss: 0.8842475414276123:  20%|█▉        | 1276/6470 [14:39<1:00:52,  1.42it/s]
2|1277|Loss: 0.8842475414276123:  20%|█▉        | 1277/6470 [14:39<1:02:46,  1.38it/s]
2|1278|Loss: 0.8164142966270447:  20%|█▉        | 1277/6470 [14:40<1:02:46,  1.38it/s]
2|1278|Loss: 0.8164142966270447:  20%|█▉        | 1278/6470 [14:40<1:05:42,  1.32it/s]
2|1279|Loss: 0.9929091334342957:  20%|█▉        | 1278/6470 [14:41<1:05:42,  1.32it/s]
2|1279|Loss: 0.9929091334342957:  20%|█▉        | 1279/6470 [14:41<1:03:27,  1.36it/s]
2|1280|Loss: 0.8515760898590088:  20%|█▉        | 1279/6470 [14:41<1:03:27,  1.36it/s]
2|1280|Loss: 0.8515760898590088:  20%|█▉        | 1280/6470 [14:42<1:03:09,  1.37it/s]
2|1281|Loss: 0.7788310647010803:  20%|█▉        | 1280/6470 [14:42<1:03:09,  1.37it/s]
2|1281|Loss: 0.7788310647010803:  20%|█▉        | 1281/6470 [14:42<1:01:07,  1.42it/s]
2|1282|Loss: 0.9312613010406494:  20%|█▉        | 1281/6470 [14:43<1:01:07,  1.42it/s]
2|1282|Loss: 0.9312613010406494:  20%|█▉        | 1282/6470 [14:43<1:01:34,  1.40it/s]
2|1283|Loss: 0.9927502274513245:  20%|█▉        | 1282/6470 [14:43<1:01:34,  1.40it/s]
2|1283|Loss: 0.9927502274513245:  20%|█▉        | 1283/6470 [14:44<1:00:30,  1.43it/s]
2|1284|Loss: 0.8867840766906738:  20%|█▉        | 1283/6470 [14:44<1:00:30,  1.43it/s]
2|1284|Loss: 0.8867840766906738:  20%|█▉        | 1284/6470 [14:44<58:27,  1.48it/s]  
2|1285|Loss: 0.8893100023269653:  20%|█▉        | 1284/6470 [14:45<58:27,  1.48it/s]
2|1285|Loss: 0.8893100023269653:  20%|█▉        | 1285/6470 [14:45<59:14,  1.46it/s]
2|1286|Loss: 0.9106302261352539:  20%|█▉        | 1285/6470 [14:45<59:14,  1.46it/s]
2|1286|Loss: 0.9106302261352539:  20%|█▉        | 1286/6470 [14:46<59:54,  1.44it/s]
2|1287|Loss: 0.7527395486831665:  20%|█▉        | 1286/6470 [14:46<59:54,  1.44it/s]
2|1287|Loss: 0.7527395486831665:  20%|█▉        | 1287/6470 [14:46<56:05,  1.54it/s]
2|1288|Loss: 0.9075397849082947:  20%|█▉        | 1287/6470 [14:47<56:05,  1.54it/s]
2|1288|Loss: 0.9075397849082947:  20%|█▉        | 1288/6470 [14:47<57:10,  1.51it/s]
2|1289|Loss: 0.732746958732605:  20%|█▉        | 1288/6470 [14:47<57:10,  1.51it/s] 
2|1289|Loss: 0.732746958732605:  20%|█▉        | 1289/6470 [14:48<1:01:36,  1.40it/s]
2|1290|Loss: 0.9215857982635498:  20%|█▉        | 1289/6470 [14:48<1:01:36,  1.40it/s]
2|1290|Loss: 0.9215857982635498:  20%|█▉        | 1290/6470 [14:48<58:39,  1.47it/s]  
2|1291|Loss: 0.82823246717453:  20%|█▉        | 1290/6470 [14:49<58:39,  1.47it/s]  
2|1291|Loss: 0.82823246717453:  20%|█▉        | 1291/6470 [14:49<56:38,  1.52it/s]
2|1292|Loss: 0.7849974632263184:  20%|█▉        | 1291/6470 [14:49<56:38,  1.52it/s]
2|1292|Loss: 0.7849974632263184:  20%|█▉        | 1292/6470 [14:50<1:02:43,  1.38it/s]
2|1293|Loss: 0.7799782752990723:  20%|█▉        | 1292/6470 [14:50<1:02:43,  1.38it/s]
2|1293|Loss: 0.7799782752990723:  20%|█▉        | 1293/6470 [14:51<1:07:29,  1.28it/s]
2|1294|Loss: 0.9666112065315247:  20%|█▉        | 1293/6470 [14:51<1:07:29,  1.28it/s]
2|1294|Loss: 0.9666112065315247:  20%|██        | 1294/6470 [14:51<1:07:28,  1.28it/s]
2|1295|Loss: 0.912634551525116:  20%|██        | 1294/6470 [14:52<1:07:28,  1.28it/s] 
2|1295|Loss: 0.912634551525116:  20%|██        | 1295/6470 [14:52<1:07:50,  1.27it/s]
2|1296|Loss: 0.836202085018158:  20%|██        | 1295/6470 [14:53<1:07:50,  1.27it/s]
2|1296|Loss: 0.836202085018158:  20%|██        | 1296/6470 [14:53<1:06:51,  1.29it/s]
2|1297|Loss: 0.908814013004303:  20%|██        | 1296/6470 [14:53<1:06:51,  1.29it/s]
2|1297|Loss: 0.908814013004303:  20%|██        | 1297/6470 [14:54<1:07:23,  1.28it/s]
2|1298|Loss: 0.7799484133720398:  20%|██        | 1297/6470 [14:54<1:07:23,  1.28it/s]
2|1298|Loss: 0.7799484133720398:  20%|██        | 1298/6470 [14:55<1:06:06,  1.30it/s]
2|1299|Loss: 0.952986478805542:  20%|██        | 1298/6470 [14:55<1:06:06,  1.30it/s] 
2|1299|Loss: 0.952986478805542:  20%|██        | 1299/6470 [14:55<1:06:18,  1.30it/s]
2|1300|Loss: 0.9132676720619202:  20%|██        | 1299/6470 [14:56<1:06:18,  1.30it/s]
2|1300|Loss: 0.9132676720619202:  20%|██        | 1300/6470 [14:56<1:03:21,  1.36it/s]
2|1301|Loss: 0.7456368207931519:  20%|██        | 1300/6470 [14:56<1:03:21,  1.36it/s]
2|1301|Loss: 0.7456368207931519:  20%|██        | 1301/6470 [14:57<59:24,  1.45it/s]  
2|1302|Loss: 0.9780927896499634:  20%|██        | 1301/6470 [14:57<59:24,  1.45it/s]
2|1302|Loss: 0.9780927896499634:  20%|██        | 1302/6470 [14:57<59:09,  1.46it/s]
2|1303|Loss: 0.8022074103355408:  20%|██        | 1302/6470 [14:58<59:09,  1.46it/s]
2|1303|Loss: 0.8022074103355408:  20%|██        | 1303/6470 [14:58<59:44,  1.44it/s]
2|1304|Loss: 0.9942692518234253:  20%|██        | 1303/6470 [14:58<59:44,  1.44it/s]
2|1304|Loss: 0.9942692518234253:  20%|██        | 1304/6470 [14:59<57:39,  1.49it/s]
2|1305|Loss: 1.007289171218872:  20%|██        | 1304/6470 [14:59<57:39,  1.49it/s] 
2|1305|Loss: 1.007289171218872:  20%|██        | 1305/6470 [14:59<58:52,  1.46it/s]
2|1306|Loss: 0.7431394457817078:  20%|██        | 1305/6470 [15:00<58:52,  1.46it/s]
2|1306|Loss: 0.7431394457817078:  20%|██        | 1306/6470 [15:00<57:58,  1.48it/s]
2|1307|Loss: 0.8071656823158264:  20%|██        | 1306/6470 [15:00<57:58,  1.48it/s]
2|1307|Loss: 0.8071656823158264:  20%|██        | 1307/6470 [15:01<58:50,  1.46it/s]
2|1308|Loss: 0.9092627763748169:  20%|██        | 1307/6470 [15:01<58:50,  1.46it/s]
2|1308|Loss: 0.9092627763748169:  20%|██        | 1308/6470 [15:01<1:02:23,  1.38it/s]
2|1309|Loss: 0.7801741361618042:  20%|██        | 1308/6470 [15:02<1:02:23,  1.38it/s]
2|1309|Loss: 0.7801741361618042:  20%|██        | 1309/6470 [15:02<1:01:35,  1.40it/s]
2|1310|Loss: 0.8502472043037415:  20%|██        | 1309/6470 [15:03<1:01:35,  1.40it/s]
2|1310|Loss: 0.8502472043037415:  20%|██        | 1310/6470 [15:03<1:01:29,  1.40it/s]
2|1311|Loss: 0.8181049227714539:  20%|██        | 1310/6470 [15:03<1:01:29,  1.40it/s]
2|1311|Loss: 0.8181049227714539:  20%|██        | 1311/6470 [15:04<1:01:34,  1.40it/s]
2|1312|Loss: 0.8514443635940552:  20%|██        | 1311/6470 [15:04<1:01:34,  1.40it/s]
2|1312|Loss: 0.8514443635940552:  20%|██        | 1312/6470 [15:04<1:01:10,  1.41it/s]
2|1313|Loss: 0.6344515681266785:  20%|██        | 1312/6470 [15:05<1:01:10,  1.41it/s]
2|1313|Loss: 0.6344515681266785:  20%|██        | 1313/6470 [15:05<1:01:25,  1.40it/s]
2|1314|Loss: 0.9165977835655212:  20%|██        | 1313/6470 [15:05<1:01:25,  1.40it/s]
2|1314|Loss: 0.9165977835655212:  20%|██        | 1314/6470 [15:05<56:04,  1.53it/s]  
2|1315|Loss: 0.8971662521362305:  20%|██        | 1314/6470 [15:06<56:04,  1.53it/s]
2|1315|Loss: 0.8971662521362305:  20%|██        | 1315/6470 [15:06<54:16,  1.58it/s]
2|1316|Loss: 0.9305046200752258:  20%|██        | 1315/6470 [15:06<54:16,  1.58it/s]
2|1316|Loss: 0.9305046200752258:  20%|██        | 1316/6470 [15:07<55:41,  1.54it/s]
2|1317|Loss: 0.8324300646781921:  20%|██        | 1316/6470 [15:07<55:41,  1.54it/s]
2|1317|Loss: 0.8324300646781921:  20%|██        | 1317/6470 [15:08<58:13,  1.48it/s]
2|1318|Loss: 0.9869435429573059:  20%|██        | 1317/6470 [15:08<58:13,  1.48it/s]
2|1318|Loss: 0.9869435429573059:  20%|██        | 1318/6470 [15:08<1:00:01,  1.43it/s]
2|1319|Loss: 0.7561704516410828:  20%|██        | 1318/6470 [15:09<1:00:01,  1.43it/s]
2|1319|Loss: 0.7561704516410828:  20%|██        | 1319/6470 [15:09<1:03:37,  1.35it/s]
2|1320|Loss: 0.7038911581039429:  20%|██        | 1319/6470 [15:10<1:03:37,  1.35it/s]
2|1320|Loss: 0.7038911581039429:  20%|██        | 1320/6470 [15:10<1:07:07,  1.28it/s]
2|1321|Loss: 0.817902684211731:  20%|██        | 1320/6470 [15:10<1:07:07,  1.28it/s] 
2|1321|Loss: 0.817902684211731:  20%|██        | 1321/6470 [15:11<1:04:59,  1.32it/s]
2|1322|Loss: 0.8581002354621887:  20%|██        | 1321/6470 [15:11<1:04:59,  1.32it/s]
2|1322|Loss: 0.8581002354621887:  20%|██        | 1322/6470 [15:11<1:01:51,  1.39it/s]
2|1323|Loss: 0.7657912373542786:  20%|██        | 1322/6470 [15:12<1:01:51,  1.39it/s]
2|1323|Loss: 0.7657912373542786:  20%|██        | 1323/6470 [15:12<1:04:13,  1.34it/s]
2|1324|Loss: 0.9530988335609436:  20%|██        | 1323/6470 [15:13<1:04:13,  1.34it/s]
2|1324|Loss: 0.9530988335609436:  20%|██        | 1324/6470 [15:13<1:04:02,  1.34it/s]
2|1325|Loss: 0.8294222950935364:  20%|██        | 1324/6470 [15:13<1:04:02,  1.34it/s]
2|1325|Loss: 0.8294222950935364:  20%|██        | 1325/6470 [15:14<1:03:28,  1.35it/s]
2|1326|Loss: 0.8568744659423828:  20%|██        | 1325/6470 [15:14<1:03:28,  1.35it/s]
2|1326|Loss: 0.8568744659423828:  20%|██        | 1326/6470 [15:14<1:02:41,  1.37it/s]
2|1327|Loss: 0.9276712536811829:  20%|██        | 1326/6470 [15:15<1:02:41,  1.37it/s]
2|1327|Loss: 0.9276712536811829:  21%|██        | 1327/6470 [15:15<1:02:38,  1.37it/s]
2|1328|Loss: 0.8285669684410095:  21%|██        | 1327/6470 [15:15<1:02:38,  1.37it/s]
2|1328|Loss: 0.8285669684410095:  21%|██        | 1328/6470 [15:16<1:00:24,  1.42it/s]
2|1329|Loss: 0.9289446473121643:  21%|██        | 1328/6470 [15:16<1:00:24,  1.42it/s]
2|1329|Loss: 0.9289446473121643:  21%|██        | 1329/6470 [15:16<59:51,  1.43it/s]  
2|1330|Loss: 0.9151625037193298:  21%|██        | 1329/6470 [15:17<59:51,  1.43it/s]
2|1330|Loss: 0.9151625037193298:  21%|██        | 1330/6470 [15:17<53:25,  1.60it/s]
2|1331|Loss: 1.116986870765686:  21%|██        | 1330/6470 [15:17<53:25,  1.60it/s] 
2|1331|Loss: 1.116986870765686:  21%|██        | 1331/6470 [15:18<1:00:37,  1.41it/s]
2|1332|Loss: 0.9653062224388123:  21%|██        | 1331/6470 [15:18<1:00:37,  1.41it/s]
2|1332|Loss: 0.9653062224388123:  21%|██        | 1332/6470 [15:19<1:05:11,  1.31it/s]
2|1333|Loss: 0.9358920454978943:  21%|██        | 1332/6470 [15:19<1:05:11,  1.31it/s]
2|1333|Loss: 0.9358920454978943:  21%|██        | 1333/6470 [15:19<1:02:03,  1.38it/s]
2|1334|Loss: 0.7890970706939697:  21%|██        | 1333/6470 [15:20<1:02:03,  1.38it/s]
2|1334|Loss: 0.7890970706939697:  21%|██        | 1334/6470 [15:20<58:09,  1.47it/s]  
2|1335|Loss: 0.7275903820991516:  21%|██        | 1334/6470 [15:20<58:09,  1.47it/s]
2|1335|Loss: 0.7275903820991516:  21%|██        | 1335/6470 [15:21<58:49,  1.45it/s]
2|1336|Loss: 0.8842517137527466:  21%|██        | 1335/6470 [15:21<58:49,  1.45it/s]
2|1336|Loss: 0.8842517137527466:  21%|██        | 1336/6470 [15:21<58:05,  1.47it/s]
2|1337|Loss: 0.7390272617340088:  21%|██        | 1336/6470 [15:22<58:05,  1.47it/s]
2|1337|Loss: 0.7390272617340088:  21%|██        | 1337/6470 [15:22<57:01,  1.50it/s]
2|1338|Loss: 0.840833842754364:  21%|██        | 1337/6470 [15:22<57:01,  1.50it/s] 
2|1338|Loss: 0.840833842754364:  21%|██        | 1338/6470 [15:22<56:12,  1.52it/s]
2|1339|Loss: 0.8134310245513916:  21%|██        | 1338/6470 [15:23<56:12,  1.52it/s]
2|1339|Loss: 0.8134310245513916:  21%|██        | 1339/6470 [15:23<58:40,  1.46it/s]
2|1340|Loss: 0.7205370664596558:  21%|██        | 1339/6470 [15:24<58:40,  1.46it/s]
2|1340|Loss: 0.7205370664596558:  21%|██        | 1340/6470 [15:24<59:31,  1.44it/s]
2|1341|Loss: 0.8702389597892761:  21%|██        | 1340/6470 [15:24<59:31,  1.44it/s]
2|1341|Loss: 0.8702389597892761:  21%|██        | 1341/6470 [15:25<59:44,  1.43it/s]
2|1342|Loss: 0.9140537977218628:  21%|██        | 1341/6470 [15:25<59:44,  1.43it/s]
2|1342|Loss: 0.9140537977218628:  21%|██        | 1342/6470 [15:25<58:05,  1.47it/s]
2|1343|Loss: 0.8291286826133728:  21%|██        | 1342/6470 [15:26<58:05,  1.47it/s]
2|1343|Loss: 0.8291286826133728:  21%|██        | 1343/6470 [15:26<53:07,  1.61it/s]
2|1344|Loss: 0.8659422397613525:  21%|██        | 1343/6470 [15:26<53:07,  1.61it/s]
2|1344|Loss: 0.8659422397613525:  21%|██        | 1344/6470 [15:26<54:17,  1.57it/s]
2|1345|Loss: 0.8690324425697327:  21%|██        | 1344/6470 [15:27<54:17,  1.57it/s]
2|1345|Loss: 0.8690324425697327:  21%|██        | 1345/6470 [15:27<55:02,  1.55it/s]
2|1346|Loss: 0.839878499507904:  21%|██        | 1345/6470 [15:27<55:02,  1.55it/s] 
2|1346|Loss: 0.839878499507904:  21%|██        | 1346/6470 [15:28<51:44,  1.65it/s]
2|1347|Loss: 0.8921694159507751:  21%|██        | 1346/6470 [15:28<51:44,  1.65it/s]
2|1347|Loss: 0.8921694159507751:  21%|██        | 1347/6470 [15:28<51:21,  1.66it/s]
2|1348|Loss: 0.8635938763618469:  21%|██        | 1347/6470 [15:29<51:21,  1.66it/s]
2|1348|Loss: 0.8635938763618469:  21%|██        | 1348/6470 [15:29<57:05,  1.50it/s]
2|1349|Loss: 0.851603627204895:  21%|██        | 1348/6470 [15:29<57:05,  1.50it/s] 
2|1349|Loss: 0.851603627204895:  21%|██        | 1349/6470 [15:30<57:24,  1.49it/s]
2|1350|Loss: 1.0256621837615967:  21%|██        | 1349/6470 [15:30<57:24,  1.49it/s]
2|1350|Loss: 1.0256621837615967:  21%|██        | 1350/6470 [15:31<1:00:38,  1.41it/s]
2|1351|Loss: 0.8897889256477356:  21%|██        | 1350/6470 [15:31<1:00:38,  1.41it/s]
2|1351|Loss: 0.8897889256477356:  21%|██        | 1351/6470 [15:31<1:01:22,  1.39it/s]
2|1352|Loss: 0.9061601758003235:  21%|██        | 1351/6470 [15:32<1:01:22,  1.39it/s]
2|1352|Loss: 0.9061601758003235:  21%|██        | 1352/6470 [15:32<1:01:46,  1.38it/s]
2|1353|Loss: 1.082523226737976:  21%|██        | 1352/6470 [15:32<1:01:46,  1.38it/s] 
2|1353|Loss: 1.082523226737976:  21%|██        | 1353/6470 [15:33<57:05,  1.49it/s]  
2|1354|Loss: 0.762597918510437:  21%|██        | 1353/6470 [15:33<57:05,  1.49it/s]
2|1354|Loss: 0.762597918510437:  21%|██        | 1354/6470 [15:33<53:36,  1.59it/s]
2|1355|Loss: 1.04899263381958:  21%|██        | 1354/6470 [15:33<53:36,  1.59it/s] 
2|1355|Loss: 1.04899263381958:  21%|██        | 1355/6470 [15:34<51:33,  1.65it/s]
2|1356|Loss: 0.663538932800293:  21%|██        | 1355/6470 [15:34<51:33,  1.65it/s]
2|1356|Loss: 0.663538932800293:  21%|██        | 1356/6470 [15:34<54:20,  1.57it/s]
2|1357|Loss: 0.7585691213607788:  21%|██        | 1356/6470 [15:35<54:20,  1.57it/s]
2|1357|Loss: 0.7585691213607788:  21%|██        | 1357/6470 [15:35<56:32,  1.51it/s]
2|1358|Loss: 0.8295624256134033:  21%|██        | 1357/6470 [15:35<56:32,  1.51it/s]
2|1358|Loss: 0.8295624256134033:  21%|██        | 1358/6470 [15:36<57:43,  1.48it/s]
2|1359|Loss: 0.942027747631073:  21%|██        | 1358/6470 [15:36<57:43,  1.48it/s] 
2|1359|Loss: 0.942027747631073:  21%|██        | 1359/6470 [15:36<55:13,  1.54it/s]
2|1360|Loss: 0.9057315587997437:  21%|██        | 1359/6470 [15:37<55:13,  1.54it/s]
2|1360|Loss: 0.9057315587997437:  21%|██        | 1360/6470 [15:37<54:35,  1.56it/s]
2|1361|Loss: 0.8586270213127136:  21%|██        | 1360/6470 [15:37<54:35,  1.56it/s]
2|1361|Loss: 0.8586270213127136:  21%|██        | 1361/6470 [15:38<58:56,  1.44it/s]
2|1362|Loss: 0.7838650345802307:  21%|██        | 1361/6470 [15:38<58:56,  1.44it/s]
2|1362|Loss: 0.7838650345802307:  21%|██        | 1362/6470 [15:38<56:15,  1.51it/s]
2|1363|Loss: 0.9499266147613525:  21%|██        | 1362/6470 [15:39<56:15,  1.51it/s]
2|1363|Loss: 0.9499266147613525:  21%|██        | 1363/6470 [15:39<53:43,  1.58it/s]
2|1364|Loss: 0.8484375476837158:  21%|██        | 1363/6470 [15:39<53:43,  1.58it/s]
2|1364|Loss: 0.8484375476837158:  21%|██        | 1364/6470 [15:40<56:58,  1.49it/s]
2|1365|Loss: 0.8185878396034241:  21%|██        | 1364/6470 [15:40<56:58,  1.49it/s]
2|1365|Loss: 0.8185878396034241:  21%|██        | 1365/6470 [15:40<57:19,  1.48it/s]
2|1366|Loss: 0.7142997980117798:  21%|██        | 1365/6470 [15:41<57:19,  1.48it/s]
2|1366|Loss: 0.7142997980117798:  21%|██        | 1366/6470 [15:41<58:29,  1.45it/s]
2|1367|Loss: 0.8192505836486816:  21%|██        | 1366/6470 [15:41<58:29,  1.45it/s]
2|1367|Loss: 0.8192505836486816:  21%|██        | 1367/6470 [15:42<54:03,  1.57it/s]
2|1368|Loss: 0.9697538018226624:  21%|██        | 1367/6470 [15:42<54:03,  1.57it/s]
2|1368|Loss: 0.9697538018226624:  21%|██        | 1368/6470 [15:42<49:12,  1.73it/s]
2|1369|Loss: 0.7555593848228455:  21%|██        | 1368/6470 [15:42<49:12,  1.73it/s]
2|1369|Loss: 0.7555593848228455:  21%|██        | 1369/6470 [15:43<52:56,  1.61it/s]
2|1370|Loss: 1.0563724040985107:  21%|██        | 1369/6470 [15:43<52:56,  1.61it/s]
2|1370|Loss: 1.0563724040985107:  21%|██        | 1370/6470 [15:44<58:01,  1.46it/s]
2|1371|Loss: 0.9252256751060486:  21%|██        | 1370/6470 [15:44<58:01,  1.46it/s]
2|1371|Loss: 0.9252256751060486:  21%|██        | 1371/6470 [15:44<59:48,  1.42it/s]
2|1372|Loss: 0.8001068830490112:  21%|██        | 1371/6470 [15:45<59:48,  1.42it/s]
2|1372|Loss: 0.8001068830490112:  21%|██        | 1372/6470 [15:45<59:50,  1.42it/s]
2|1373|Loss: 0.7286385297775269:  21%|██        | 1372/6470 [15:45<59:50,  1.42it/s]
2|1373|Loss: 0.7286385297775269:  21%|██        | 1373/6470 [15:46<56:34,  1.50it/s]
2|1374|Loss: 0.954002320766449:  21%|██        | 1373/6470 [15:46<56:34,  1.50it/s] 
2|1374|Loss: 0.954002320766449:  21%|██        | 1374/6470 [15:46<52:52,  1.61it/s]
2|1375|Loss: 0.6791483759880066:  21%|██        | 1374/6470 [15:47<52:52,  1.61it/s]
2|1375|Loss: 0.6791483759880066:  21%|██▏       | 1375/6470 [15:47<54:46,  1.55it/s]
2|1376|Loss: 0.9015393853187561:  21%|██▏       | 1375/6470 [15:47<54:46,  1.55it/s]
2|1376|Loss: 0.9015393853187561:  21%|██▏       | 1376/6470 [15:47<52:55,  1.60it/s]
2|1377|Loss: 0.6976857781410217:  21%|██▏       | 1376/6470 [15:48<52:55,  1.60it/s]
2|1377|Loss: 0.6976857781410217:  21%|██▏       | 1377/6470 [15:48<56:52,  1.49it/s]
2|1378|Loss: 0.8509992957115173:  21%|██▏       | 1377/6470 [15:49<56:52,  1.49it/s]
2|1378|Loss: 0.8509992957115173:  21%|██▏       | 1378/6470 [15:49<57:15,  1.48it/s]
2|1379|Loss: 0.8654179573059082:  21%|██▏       | 1378/6470 [15:49<57:15,  1.48it/s]
2|1379|Loss: 0.8654179573059082:  21%|██▏       | 1379/6470 [15:50<58:06,  1.46it/s]
2|1380|Loss: 0.7930092215538025:  21%|██▏       | 1379/6470 [15:50<58:06,  1.46it/s]
2|1380|Loss: 0.7930092215538025:  21%|██▏       | 1380/6470 [15:50<57:03,  1.49it/s]
2|1381|Loss: 0.8736404180526733:  21%|██▏       | 1380/6470 [15:51<57:03,  1.49it/s]
2|1381|Loss: 0.8736404180526733:  21%|██▏       | 1381/6470 [15:51<52:16,  1.62it/s]
2|1382|Loss: 0.9197710156440735:  21%|██▏       | 1381/6470 [15:51<52:16,  1.62it/s]
2|1382|Loss: 0.9197710156440735:  21%|██▏       | 1382/6470 [15:51<51:39,  1.64it/s]
2|1383|Loss: 0.9267149567604065:  21%|██▏       | 1382/6470 [15:52<51:39,  1.64it/s]
2|1383|Loss: 0.9267149567604065:  21%|██▏       | 1383/6470 [15:52<55:36,  1.52it/s]
2|1384|Loss: 0.9425836205482483:  21%|██▏       | 1383/6470 [15:53<55:36,  1.52it/s]
2|1384|Loss: 0.9425836205482483:  21%|██▏       | 1384/6470 [15:53<59:55,  1.41it/s]
2|1385|Loss: 1.0081498622894287:  21%|██▏       | 1384/6470 [15:53<59:55,  1.41it/s]
2|1385|Loss: 1.0081498622894287:  21%|██▏       | 1385/6470 [15:54<57:59,  1.46it/s]
2|1386|Loss: 0.8252115249633789:  21%|██▏       | 1385/6470 [15:54<57:59,  1.46it/s]
2|1386|Loss: 0.8252115249633789:  21%|██▏       | 1386/6470 [15:54<53:25,  1.59it/s]
2|1387|Loss: 0.9067190885543823:  21%|██▏       | 1386/6470 [15:54<53:25,  1.59it/s]
2|1387|Loss: 0.9067190885543823:  21%|██▏       | 1387/6470 [15:55<54:36,  1.55it/s]
2|1388|Loss: 0.9613754749298096:  21%|██▏       | 1387/6470 [15:55<54:36,  1.55it/s]
2|1388|Loss: 0.9613754749298096:  21%|██▏       | 1388/6470 [15:55<57:08,  1.48it/s]
2|1389|Loss: 0.9479409456253052:  21%|██▏       | 1388/6470 [15:56<57:08,  1.48it/s]
2|1389|Loss: 0.9479409456253052:  21%|██▏       | 1389/6470 [15:56<58:18,  1.45it/s]
2|1390|Loss: 0.7899225950241089:  21%|██▏       | 1389/6470 [15:57<58:18,  1.45it/s]
2|1390|Loss: 0.7899225950241089:  21%|██▏       | 1390/6470 [15:57<57:17,  1.48it/s]
2|1391|Loss: 0.931669294834137:  21%|██▏       | 1390/6470 [15:57<57:17,  1.48it/s] 
2|1391|Loss: 0.931669294834137:  21%|██▏       | 1391/6470 [15:57<55:06,  1.54it/s]
2|1392|Loss: 0.9481152892112732:  21%|██▏       | 1391/6470 [15:58<55:06,  1.54it/s]
2|1392|Loss: 0.9481152892112732:  22%|██▏       | 1392/6470 [15:58<54:33,  1.55it/s]
2|1393|Loss: 0.8407345414161682:  22%|██▏       | 1392/6470 [15:58<54:33,  1.55it/s]
2|1393|Loss: 0.8407345414161682:  22%|██▏       | 1393/6470 [15:59<51:46,  1.63it/s]
2|1394|Loss: 0.9248384833335876:  22%|██▏       | 1393/6470 [15:59<51:46,  1.63it/s]
2|1394|Loss: 0.9248384833335876:  22%|██▏       | 1394/6470 [15:59<54:44,  1.55it/s]
2|1395|Loss: 0.9456425905227661:  22%|██▏       | 1394/6470 [16:00<54:44,  1.55it/s]
2|1395|Loss: 0.9456425905227661:  22%|██▏       | 1395/6470 [16:00<57:50,  1.46it/s]
2|1396|Loss: 0.8079585433006287:  22%|██▏       | 1395/6470 [16:00<57:50,  1.46it/s]
2|1396|Loss: 0.8079585433006287:  22%|██▏       | 1396/6470 [16:01<56:54,  1.49it/s]
2|1397|Loss: 0.7825362086296082:  22%|██▏       | 1396/6470 [16:01<56:54,  1.49it/s]
2|1397|Loss: 0.7825362086296082:  22%|██▏       | 1397/6470 [16:01<53:35,  1.58it/s]
2|1398|Loss: 0.793529748916626:  22%|██▏       | 1397/6470 [16:02<53:35,  1.58it/s] 
2|1398|Loss: 0.793529748916626:  22%|██▏       | 1398/6470 [16:02<56:04,  1.51it/s]
2|1399|Loss: 0.7403020858764648:  22%|██▏       | 1398/6470 [16:02<56:04,  1.51it/s]
2|1399|Loss: 0.7403020858764648:  22%|██▏       | 1399/6470 [16:03<53:37,  1.58it/s]
2|1400|Loss: 0.8260950446128845:  22%|██▏       | 1399/6470 [16:03<53:37,  1.58it/s]
2|1400|Loss: 0.8260950446128845:  22%|██▏       | 1400/6470 [16:03<49:15,  1.72it/s]
2|1401|Loss: 0.8766159415245056:  22%|██▏       | 1400/6470 [16:03<49:15,  1.72it/s]
2|1401|Loss: 0.8766159415245056:  22%|██▏       | 1401/6470 [16:04<47:54,  1.76it/s]
2|1402|Loss: 0.8495243191719055:  22%|██▏       | 1401/6470 [16:04<47:54,  1.76it/s]
2|1402|Loss: 0.8495243191719055:  22%|██▏       | 1402/6470 [16:04<49:25,  1.71it/s]
2|1403|Loss: 0.9985154271125793:  22%|██▏       | 1402/6470 [16:05<49:25,  1.71it/s]
2|1403|Loss: 0.9985154271125793:  22%|██▏       | 1403/6470 [16:05<52:17,  1.61it/s]
2|1404|Loss: 0.9420766830444336:  22%|██▏       | 1403/6470 [16:05<52:17,  1.61it/s]
2|1404|Loss: 0.9420766830444336:  22%|██▏       | 1404/6470 [16:06<55:14,  1.53it/s]
2|1405|Loss: 0.9149883985519409:  22%|██▏       | 1404/6470 [16:06<55:14,  1.53it/s]
2|1405|Loss: 0.9149883985519409:  22%|██▏       | 1405/6470 [16:06<52:44,  1.60it/s]
2|1406|Loss: 1.039610743522644:  22%|██▏       | 1405/6470 [16:06<52:44,  1.60it/s] 
2|1406|Loss: 1.039610743522644:  22%|██▏       | 1406/6470 [16:07<48:42,  1.73it/s]
2|1407|Loss: 0.9403259754180908:  22%|██▏       | 1406/6470 [16:07<48:42,  1.73it/s]
2|1407|Loss: 0.9403259754180908:  22%|██▏       | 1407/6470 [16:07<50:29,  1.67it/s]
2|1408|Loss: 0.8719550967216492:  22%|██▏       | 1407/6470 [16:08<50:29,  1.67it/s]
2|1408|Loss: 0.8719550967216492:  22%|██▏       | 1408/6470 [16:08<50:34,  1.67it/s]
2|1409|Loss: 0.8009118437767029:  22%|██▏       | 1408/6470 [16:08<50:34,  1.67it/s]
2|1409|Loss: 0.8009118437767029:  22%|██▏       | 1409/6470 [16:09<51:16,  1.65it/s]
2|1410|Loss: 0.9720434546470642:  22%|██▏       | 1409/6470 [16:09<51:16,  1.65it/s]
2|1410|Loss: 0.9720434546470642:  22%|██▏       | 1410/6470 [16:09<49:06,  1.72it/s]
2|1411|Loss: 0.887805163860321:  22%|██▏       | 1410/6470 [16:09<49:06,  1.72it/s] 
2|1411|Loss: 0.887805163860321:  22%|██▏       | 1411/6470 [16:10<54:05,  1.56it/s]
2|1412|Loss: 1.015594720840454:  22%|██▏       | 1411/6470 [16:10<54:05,  1.56it/s]
2|1412|Loss: 1.015594720840454:  22%|██▏       | 1412/6470 [16:10<52:40,  1.60it/s]
2|1413|Loss: 0.8875636458396912:  22%|██▏       | 1412/6470 [16:11<52:40,  1.60it/s]
2|1413|Loss: 0.8875636458396912:  22%|██▏       | 1413/6470 [16:11<51:07,  1.65it/s]
2|1414|Loss: 0.6952369213104248:  22%|██▏       | 1413/6470 [16:11<51:07,  1.65it/s]
2|1414|Loss: 0.6952369213104248:  22%|██▏       | 1414/6470 [16:12<51:42,  1.63it/s]
2|1415|Loss: 0.973200798034668:  22%|██▏       | 1414/6470 [16:12<51:42,  1.63it/s] 
2|1415|Loss: 0.973200798034668:  22%|██▏       | 1415/6470 [16:12<54:33,  1.54it/s]
2|1416|Loss: 0.7957406044006348:  22%|██▏       | 1415/6470 [16:13<54:33,  1.54it/s]
2|1416|Loss: 0.7957406044006348:  22%|██▏       | 1416/6470 [16:13<51:42,  1.63it/s]
2|1417|Loss: 0.9463805556297302:  22%|██▏       | 1416/6470 [16:13<51:42,  1.63it/s]
2|1417|Loss: 0.9463805556297302:  22%|██▏       | 1417/6470 [16:14<53:11,  1.58it/s]
2|1418|Loss: 0.8507779240608215:  22%|██▏       | 1417/6470 [16:14<53:11,  1.58it/s]
2|1418|Loss: 0.8507779240608215:  22%|██▏       | 1418/6470 [16:14<56:54,  1.48it/s]
2|1419|Loss: 0.8554149866104126:  22%|██▏       | 1418/6470 [16:15<56:54,  1.48it/s]
2|1419|Loss: 0.8554149866104126:  22%|██▏       | 1419/6470 [16:15<58:06,  1.45it/s]
2|1420|Loss: 0.8323581218719482:  22%|██▏       | 1419/6470 [16:15<58:06,  1.45it/s]
2|1420|Loss: 0.8323581218719482:  22%|██▏       | 1420/6470 [16:16<58:59,  1.43it/s]
2|1421|Loss: 0.7318992614746094:  22%|██▏       | 1420/6470 [16:16<58:59,  1.43it/s]
2|1421|Loss: 0.7318992614746094:  22%|██▏       | 1421/6470 [16:16<57:39,  1.46it/s]
2|1422|Loss: 0.907379150390625:  22%|██▏       | 1421/6470 [16:17<57:39,  1.46it/s] 
2|1422|Loss: 0.907379150390625:  22%|██▏       | 1422/6470 [16:17<1:00:16,  1.40it/s]
2|1423|Loss: 0.9225612878799438:  22%|██▏       | 1422/6470 [16:18<1:00:16,  1.40it/s]
2|1423|Loss: 0.9225612878799438:  22%|██▏       | 1423/6470 [16:18<1:00:42,  1.39it/s]
2|1424|Loss: 0.8866190314292908:  22%|██▏       | 1423/6470 [16:18<1:00:42,  1.39it/s]
2|1424|Loss: 0.8866190314292908:  22%|██▏       | 1424/6470 [16:19<57:46,  1.46it/s]  
2|1425|Loss: 0.8798664808273315:  22%|██▏       | 1424/6470 [16:19<57:46,  1.46it/s]
2|1425|Loss: 0.8798664808273315:  22%|██▏       | 1425/6470 [16:19<56:26,  1.49it/s]
2|1426|Loss: 0.849898099899292:  22%|██▏       | 1425/6470 [16:20<56:26,  1.49it/s] 
2|1426|Loss: 0.849898099899292:  22%|██▏       | 1426/6470 [16:20<1:00:08,  1.40it/s]
2|1427|Loss: 0.8469255566596985:  22%|██▏       | 1426/6470 [16:20<1:00:08,  1.40it/s]
2|1427|Loss: 0.8469255566596985:  22%|██▏       | 1427/6470 [16:21<59:29,  1.41it/s]  
2|1428|Loss: 0.9405432343482971:  22%|██▏       | 1427/6470 [16:21<59:29,  1.41it/s]
2|1428|Loss: 0.9405432343482971:  22%|██▏       | 1428/6470 [16:21<57:21,  1.46it/s]
2|1429|Loss: 1.0030680894851685:  22%|██▏       | 1428/6470 [16:22<57:21,  1.46it/s]
2|1429|Loss: 1.0030680894851685:  22%|██▏       | 1429/6470 [16:22<52:28,  1.60it/s]
2|1430|Loss: 0.9775509834289551:  22%|██▏       | 1429/6470 [16:22<52:28,  1.60it/s]
2|1430|Loss: 0.9775509834289551:  22%|██▏       | 1430/6470 [16:22<53:58,  1.56it/s]
2|1431|Loss: 0.8696699142456055:  22%|██▏       | 1430/6470 [16:23<53:58,  1.56it/s]
2|1431|Loss: 0.8696699142456055:  22%|██▏       | 1431/6470 [16:23<58:55,  1.43it/s]
2|1432|Loss: 0.9470987319946289:  22%|██▏       | 1431/6470 [16:24<58:55,  1.43it/s]
2|1432|Loss: 0.9470987319946289:  22%|██▏       | 1432/6470 [16:24<59:52,  1.40it/s]
2|1433|Loss: 0.9005120396614075:  22%|██▏       | 1432/6470 [16:24<59:52,  1.40it/s]
2|1433|Loss: 0.9005120396614075:  22%|██▏       | 1433/6470 [16:25<57:47,  1.45it/s]
2|1434|Loss: 0.929569661617279:  22%|██▏       | 1433/6470 [16:25<57:47,  1.45it/s] 
2|1434|Loss: 0.929569661617279:  22%|██▏       | 1434/6470 [16:25<55:30,  1.51it/s]
2|1435|Loss: 0.8594250082969666:  22%|██▏       | 1434/6470 [16:26<55:30,  1.51it/s]
2|1435|Loss: 0.8594250082969666:  22%|██▏       | 1435/6470 [16:26<57:48,  1.45it/s]
2|1436|Loss: 0.8695008158683777:  22%|██▏       | 1435/6470 [16:26<57:48,  1.45it/s]
2|1436|Loss: 0.8695008158683777:  22%|██▏       | 1436/6470 [16:27<59:23,  1.41it/s]
2|1437|Loss: 0.8112561702728271:  22%|██▏       | 1436/6470 [16:27<59:23,  1.41it/s]
2|1437|Loss: 0.8112561702728271:  22%|██▏       | 1437/6470 [16:28<1:03:51,  1.31it/s]
2|1438|Loss: 0.9303104877471924:  22%|██▏       | 1437/6470 [16:28<1:03:51,  1.31it/s]
2|1438|Loss: 0.9303104877471924:  22%|██▏       | 1438/6470 [16:28<1:03:35,  1.32it/s]
2|1439|Loss: 0.7832910418510437:  22%|██▏       | 1438/6470 [16:29<1:03:35,  1.32it/s]
2|1439|Loss: 0.7832910418510437:  22%|██▏       | 1439/6470 [16:29<1:01:14,  1.37it/s]
2|1440|Loss: 0.9308593273162842:  22%|██▏       | 1439/6470 [16:29<1:01:14,  1.37it/s]
2|1440|Loss: 0.9308593273162842:  22%|██▏       | 1440/6470 [16:30<1:00:19,  1.39it/s]
2|1441|Loss: 0.7703341245651245:  22%|██▏       | 1440/6470 [16:30<1:00:19,  1.39it/s]
2|1441|Loss: 0.7703341245651245:  22%|██▏       | 1441/6470 [16:31<59:51,  1.40it/s]  
2|1442|Loss: 0.78877192735672:  22%|██▏       | 1441/6470 [16:31<59:51,  1.40it/s]  
2|1442|Loss: 0.78877192735672:  22%|██▏       | 1442/6470 [16:31<1:01:59,  1.35it/s]
2|1443|Loss: 0.9656168818473816:  22%|██▏       | 1442/6470 [16:32<1:01:59,  1.35it/s]
2|1443|Loss: 0.9656168818473816:  22%|██▏       | 1443/6470 [16:32<1:03:25,  1.32it/s]
2|1444|Loss: 0.8018739819526672:  22%|██▏       | 1443/6470 [16:33<1:03:25,  1.32it/s]
2|1444|Loss: 0.8018739819526672:  22%|██▏       | 1444/6470 [16:33<1:01:30,  1.36it/s]
2|1445|Loss: 0.8781847357749939:  22%|██▏       | 1444/6470 [16:33<1:01:30,  1.36it/s]
2|1445|Loss: 0.8781847357749939:  22%|██▏       | 1445/6470 [16:33<55:21,  1.51it/s]  
2|1446|Loss: 0.8293779492378235:  22%|██▏       | 1445/6470 [16:34<55:21,  1.51it/s]
2|1446|Loss: 0.8293779492378235:  22%|██▏       | 1446/6470 [16:34<53:41,  1.56it/s]
2|1447|Loss: 0.8656842112541199:  22%|██▏       | 1446/6470 [16:34<53:41,  1.56it/s]
2|1447|Loss: 0.8656842112541199:  22%|██▏       | 1447/6470 [16:35<54:53,  1.53it/s]
2|1448|Loss: 0.8496900200843811:  22%|██▏       | 1447/6470 [16:35<54:53,  1.53it/s]
2|1448|Loss: 0.8496900200843811:  22%|██▏       | 1448/6470 [16:35<59:00,  1.42it/s]
2|1449|Loss: 0.7052363753318787:  22%|██▏       | 1448/6470 [16:36<59:00,  1.42it/s]
2|1449|Loss: 0.7052363753318787:  22%|██▏       | 1449/6470 [16:36<58:13,  1.44it/s]
2|1450|Loss: 0.9409529566764832:  22%|██▏       | 1449/6470 [16:36<58:13,  1.44it/s]
2|1450|Loss: 0.9409529566764832:  22%|██▏       | 1450/6470 [16:37<54:05,  1.55it/s]
2|1451|Loss: 0.7927823662757874:  22%|██▏       | 1450/6470 [16:37<54:05,  1.55it/s]
2|1451|Loss: 0.7927823662757874:  22%|██▏       | 1451/6470 [16:37<55:08,  1.52it/s]
2|1452|Loss: 0.9589868783950806:  22%|██▏       | 1451/6470 [16:38<55:08,  1.52it/s]
2|1452|Loss: 0.9589868783950806:  22%|██▏       | 1452/6470 [16:38<55:07,  1.52it/s]
2|1453|Loss: 0.919226884841919:  22%|██▏       | 1452/6470 [16:38<55:07,  1.52it/s] 
2|1453|Loss: 0.919226884841919:  22%|██▏       | 1453/6470 [16:39<59:02,  1.42it/s]
2|1454|Loss: 0.9108254909515381:  22%|██▏       | 1453/6470 [16:39<59:02,  1.42it/s]
2|1454|Loss: 0.9108254909515381:  22%|██▏       | 1454/6470 [16:40<1:00:51,  1.37it/s]
2|1455|Loss: 0.8511784076690674:  22%|██▏       | 1454/6470 [16:40<1:00:51,  1.37it/s]
2|1455|Loss: 0.8511784076690674:  22%|██▏       | 1455/6470 [16:40<58:46,  1.42it/s]  
2|1456|Loss: 0.8743848204612732:  22%|██▏       | 1455/6470 [16:41<58:46,  1.42it/s]
2|1456|Loss: 0.8743848204612732:  23%|██▎       | 1456/6470 [16:41<58:08,  1.44it/s]
2|1457|Loss: 1.0086747407913208:  23%|██▎       | 1456/6470 [16:41<58:08,  1.44it/s]
2|1457|Loss: 1.0086747407913208:  23%|██▎       | 1457/6470 [16:42<59:18,  1.41it/s]
2|1458|Loss: 0.9234315752983093:  23%|██▎       | 1457/6470 [16:42<59:18,  1.41it/s]
2|1458|Loss: 0.9234315752983093:  23%|██▎       | 1458/6470 [16:42<1:00:35,  1.38it/s]
2|1459|Loss: 0.9658468961715698:  23%|██▎       | 1458/6470 [16:43<1:00:35,  1.38it/s]
2|1459|Loss: 0.9658468961715698:  23%|██▎       | 1459/6470 [16:43<55:58,  1.49it/s]  
2|1460|Loss: 0.7531928420066833:  23%|██▎       | 1459/6470 [16:43<55:58,  1.49it/s]
2|1460|Loss: 0.7531928420066833:  23%|██▎       | 1460/6470 [16:44<1:00:00,  1.39it/s]
2|1461|Loss: 0.8083944320678711:  23%|██▎       | 1460/6470 [16:44<1:00:00,  1.39it/s]
2|1461|Loss: 0.8083944320678711:  23%|██▎       | 1461/6470 [16:44<56:27,  1.48it/s]  
2|1462|Loss: 0.9185827374458313:  23%|██▎       | 1461/6470 [16:45<56:27,  1.48it/s]
2|1462|Loss: 0.9185827374458313:  23%|██▎       | 1462/6470 [16:45<56:03,  1.49it/s]
2|1463|Loss: 0.8253480792045593:  23%|██▎       | 1462/6470 [16:45<56:03,  1.49it/s]
2|1463|Loss: 0.8253480792045593:  23%|██▎       | 1463/6470 [16:46<59:31,  1.40it/s]
2|1464|Loss: 0.9622375965118408:  23%|██▎       | 1463/6470 [16:46<59:31,  1.40it/s]
2|1464|Loss: 0.9622375965118408:  23%|██▎       | 1464/6470 [16:47<1:00:31,  1.38it/s]
2|1465|Loss: 0.9714292883872986:  23%|██▎       | 1464/6470 [16:47<1:00:31,  1.38it/s]
2|1465|Loss: 0.9714292883872986:  23%|██▎       | 1465/6470 [16:47<1:00:47,  1.37it/s]
2|1466|Loss: 0.9193310737609863:  23%|██▎       | 1465/6470 [16:48<1:00:47,  1.37it/s]
2|1466|Loss: 0.9193310737609863:  23%|██▎       | 1466/6470 [16:48<1:00:59,  1.37it/s]
2|1467|Loss: 0.9729449152946472:  23%|██▎       | 1466/6470 [16:48<1:00:59,  1.37it/s]
2|1467|Loss: 0.9729449152946472:  23%|██▎       | 1467/6470 [16:49<59:13,  1.41it/s]  
2|1468|Loss: 0.830714225769043:  23%|██▎       | 1467/6470 [16:49<59:13,  1.41it/s] 
2|1468|Loss: 0.830714225769043:  23%|██▎       | 1468/6470 [16:49<56:17,  1.48it/s]
2|1469|Loss: 0.7586193680763245:  23%|██▎       | 1468/6470 [16:50<56:17,  1.48it/s]
2|1469|Loss: 0.7586193680763245:  23%|██▎       | 1469/6470 [16:50<54:55,  1.52it/s]
2|1470|Loss: 0.7771719694137573:  23%|██▎       | 1469/6470 [16:50<54:55,  1.52it/s]
2|1470|Loss: 0.7771719694137573:  23%|██▎       | 1470/6470 [16:51<54:28,  1.53it/s]
2|1471|Loss: 0.9832279086112976:  23%|██▎       | 1470/6470 [16:51<54:28,  1.53it/s]
2|1471|Loss: 0.9832279086112976:  23%|██▎       | 1471/6470 [16:51<56:17,  1.48it/s]
2|1472|Loss: 0.8757711052894592:  23%|██▎       | 1471/6470 [16:52<56:17,  1.48it/s]
2|1472|Loss: 0.8757711052894592:  23%|██▎       | 1472/6470 [16:52<1:00:14,  1.38it/s]
2|1473|Loss: 0.7587071061134338:  23%|██▎       | 1472/6470 [16:53<1:00:14,  1.38it/s]
2|1473|Loss: 0.7587071061134338:  23%|██▎       | 1473/6470 [16:53<1:00:33,  1.38it/s]
2|1474|Loss: 0.8900187611579895:  23%|██▎       | 1473/6470 [16:53<1:00:33,  1.38it/s]
2|1474|Loss: 0.8900187611579895:  23%|██▎       | 1474/6470 [16:53<54:56,  1.52it/s]  
2|1475|Loss: 0.9692261815071106:  23%|██▎       | 1474/6470 [16:54<54:56,  1.52it/s]
2|1475|Loss: 0.9692261815071106:  23%|██▎       | 1475/6470 [16:54<51:10,  1.63it/s]
2|1476|Loss: 0.9221097826957703:  23%|██▎       | 1475/6470 [16:54<51:10,  1.63it/s]
2|1476|Loss: 0.9221097826957703:  23%|██▎       | 1476/6470 [16:54<50:52,  1.64it/s]
2|1477|Loss: 0.8874399065971375:  23%|██▎       | 1476/6470 [16:55<50:52,  1.64it/s]
2|1477|Loss: 0.8874399065971375:  23%|██▎       | 1477/6470 [16:55<55:20,  1.50it/s]
2|1478|Loss: 0.8927768468856812:  23%|██▎       | 1477/6470 [16:56<55:20,  1.50it/s]
2|1478|Loss: 0.8927768468856812:  23%|██▎       | 1478/6470 [16:56<56:56,  1.46it/s]
2|1479|Loss: 0.8889310359954834:  23%|██▎       | 1478/6470 [16:56<56:56,  1.46it/s]
2|1479|Loss: 0.8889310359954834:  23%|██▎       | 1479/6470 [16:57<1:02:26,  1.33it/s]
2|1480|Loss: 0.9529657959938049:  23%|██▎       | 1479/6470 [16:57<1:02:26,  1.33it/s]
2|1480|Loss: 0.9529657959938049:  23%|██▎       | 1480/6470 [16:57<57:37,  1.44it/s]  
2|1481|Loss: 0.8415431380271912:  23%|██▎       | 1480/6470 [16:58<57:37,  1.44it/s]
2|1481|Loss: 0.8415431380271912:  23%|██▎       | 1481/6470 [16:58<59:03,  1.41it/s]
2|1482|Loss: 0.839163601398468:  23%|██▎       | 1481/6470 [16:59<59:03,  1.41it/s] 
2|1482|Loss: 0.839163601398468:  23%|██▎       | 1482/6470 [16:59<59:40,  1.39it/s]
2|1483|Loss: 0.9341831207275391:  23%|██▎       | 1482/6470 [16:59<59:40,  1.39it/s]
2|1483|Loss: 0.9341831207275391:  23%|██▎       | 1483/6470 [17:00<1:00:10,  1.38it/s]
2|1484|Loss: 0.8478602766990662:  23%|██▎       | 1483/6470 [17:00<1:00:10,  1.38it/s]
2|1484|Loss: 0.8478602766990662:  23%|██▎       | 1484/6470 [17:00<59:54,  1.39it/s]  
2|1485|Loss: 0.7602415680885315:  23%|██▎       | 1484/6470 [17:01<59:54,  1.39it/s]
2|1485|Loss: 0.7602415680885315:  23%|██▎       | 1485/6470 [17:01<58:18,  1.42it/s]
2|1486|Loss: 0.826148509979248:  23%|██▎       | 1485/6470 [17:01<58:18,  1.42it/s] 
2|1486|Loss: 0.826148509979248:  23%|██▎       | 1486/6470 [17:02<59:50,  1.39it/s]
2|1487|Loss: 0.8679093718528748:  23%|██▎       | 1486/6470 [17:02<59:50,  1.39it/s]
2|1487|Loss: 0.8679093718528748:  23%|██▎       | 1487/6470 [17:02<56:28,  1.47it/s]
2|1488|Loss: 1.0245959758758545:  23%|██▎       | 1487/6470 [17:03<56:28,  1.47it/s]
2|1488|Loss: 1.0245959758758545:  23%|██▎       | 1488/6470 [17:03<54:54,  1.51it/s]
2|1489|Loss: 0.8993715643882751:  23%|██▎       | 1488/6470 [17:03<54:54,  1.51it/s]
2|1489|Loss: 0.8993715643882751:  23%|██▎       | 1489/6470 [17:04<58:45,  1.41it/s]
2|1490|Loss: 0.7760974168777466:  23%|██▎       | 1489/6470 [17:04<58:45,  1.41it/s]
2|1490|Loss: 0.7760974168777466:  23%|██▎       | 1490/6470 [17:04<57:42,  1.44it/s]
2|1491|Loss: 0.8823657035827637:  23%|██▎       | 1490/6470 [17:05<57:42,  1.44it/s]
2|1491|Loss: 0.8823657035827637:  23%|██▎       | 1491/6470 [17:05<1:00:24,  1.37it/s]
2|1492|Loss: 0.8969772458076477:  23%|██▎       | 1491/6470 [17:06<1:00:24,  1.37it/s]
2|1492|Loss: 0.8969772458076477:  23%|██▎       | 1492/6470 [17:06<1:01:22,  1.35it/s]
2|1493|Loss: 0.7771173715591431:  23%|██▎       | 1492/6470 [17:06<1:01:22,  1.35it/s]
2|1493|Loss: 0.7771173715591431:  23%|██▎       | 1493/6470 [17:07<59:32,  1.39it/s]  
2|1494|Loss: 0.7706362009048462:  23%|██▎       | 1493/6470 [17:07<59:32,  1.39it/s]
2|1494|Loss: 0.7706362009048462:  23%|██▎       | 1494/6470 [17:07<58:43,  1.41it/s]
2|1495|Loss: 1.339463710784912:  23%|██▎       | 1494/6470 [17:08<58:43,  1.41it/s] 
2|1495|Loss: 1.339463710784912:  23%|██▎       | 1495/6470 [17:08<59:29,  1.39it/s]
2|1496|Loss: 0.8649984002113342:  23%|██▎       | 1495/6470 [17:08<59:29,  1.39it/s]
2|1496|Loss: 0.8649984002113342:  23%|██▎       | 1496/6470 [17:09<55:40,  1.49it/s]
2|1497|Loss: 0.570553183555603:  23%|██▎       | 1496/6470 [17:09<55:40,  1.49it/s] 
2|1497|Loss: 0.570553183555603:  23%|██▎       | 1497/6470 [17:10<1:13:05,  1.13it/s]
2|1498|Loss: 0.80165034532547:  23%|██▎       | 1497/6470 [17:11<1:13:05,  1.13it/s] 
2|1498|Loss: 0.80165034532547:  23%|██▎       | 1498/6470 [17:11<1:12:10,  1.15it/s]
2|1499|Loss: 0.8713055849075317:  23%|██▎       | 1498/6470 [17:11<1:12:10,  1.15it/s]
2|1499|Loss: 0.8713055849075317:  23%|██▎       | 1499/6470 [17:12<1:10:48,  1.17it/s]
2|1500|Loss: 0.8637328743934631:  23%|██▎       | 1499/6470 [17:12<1:10:48,  1.17it/s]
2|1500|Loss: 0.8637328743934631:  23%|██▎       | 1500/6470 [17:12<1:08:40,  1.21it/s]
2|1501|Loss: 0.8691998720169067:  23%|██▎       | 1500/6470 [17:13<1:08:40,  1.21it/s]
2|1501|Loss: 0.8691998720169067:  23%|██▎       | 1501/6470 [17:13<1:04:09,  1.29it/s]
2|1502|Loss: 0.8104680776596069:  23%|██▎       | 1501/6470 [17:14<1:04:09,  1.29it/s]
2|1502|Loss: 0.8104680776596069:  23%|██▎       | 1502/6470 [17:14<1:06:14,  1.25it/s]
2|1503|Loss: 0.9456264972686768:  23%|██▎       | 1502/6470 [17:14<1:06:14,  1.25it/s]
2|1503|Loss: 0.9456264972686768:  23%|██▎       | 1503/6470 [17:15<1:07:38,  1.22it/s]
2|1504|Loss: 0.8649819493293762:  23%|██▎       | 1503/6470 [17:15<1:07:38,  1.22it/s]
2|1504|Loss: 0.8649819493293762:  23%|██▎       | 1504/6470 [17:16<1:08:07,  1.21it/s]
2|1505|Loss: 0.7120015621185303:  23%|██▎       | 1504/6470 [17:16<1:08:07,  1.21it/s]
2|1505|Loss: 0.7120015621185303:  23%|██▎       | 1505/6470 [17:17<1:08:23,  1.21it/s]
2|1506|Loss: 0.879971981048584:  23%|██▎       | 1505/6470 [17:17<1:08:23,  1.21it/s] 
2|1506|Loss: 0.879971981048584:  23%|██▎       | 1506/6470 [17:17<1:05:48,  1.26it/s]
2|1507|Loss: 0.8181672692298889:  23%|██▎       | 1506/6470 [17:18<1:05:48,  1.26it/s]
2|1507|Loss: 0.8181672692298889:  23%|██▎       | 1507/6470 [17:18<1:01:45,  1.34it/s]
2|1508|Loss: 0.8684209585189819:  23%|██▎       | 1507/6470 [17:18<1:01:45,  1.34it/s]
2|1508|Loss: 0.8684209585189819:  23%|██▎       | 1508/6470 [17:19<1:03:07,  1.31it/s]
2|1509|Loss: 0.8315702676773071:  23%|██▎       | 1508/6470 [17:19<1:03:07,  1.31it/s]
2|1509|Loss: 0.8315702676773071:  23%|██▎       | 1509/6470 [17:20<1:05:26,  1.26it/s]
2|1510|Loss: 0.9568219780921936:  23%|██▎       | 1509/6470 [17:20<1:05:26,  1.26it/s]
2|1510|Loss: 0.9568219780921936:  23%|██▎       | 1510/6470 [17:20<1:04:02,  1.29it/s]
2|1511|Loss: 0.846555769443512:  23%|██▎       | 1510/6470 [17:21<1:04:02,  1.29it/s] 
2|1511|Loss: 0.846555769443512:  23%|██▎       | 1511/6470 [17:21<1:03:04,  1.31it/s]
2|1512|Loss: 0.9725856184959412:  23%|██▎       | 1511/6470 [17:21<1:03:04,  1.31it/s]
2|1512|Loss: 0.9725856184959412:  23%|██▎       | 1512/6470 [17:22<1:04:18,  1.29it/s]
2|1513|Loss: 0.8628155589103699:  23%|██▎       | 1512/6470 [17:22<1:04:18,  1.29it/s]
2|1513|Loss: 0.8628155589103699:  23%|██▎       | 1513/6470 [17:23<1:08:07,  1.21it/s]
2|1514|Loss: 0.8628339171409607:  23%|██▎       | 1513/6470 [17:23<1:08:07,  1.21it/s]
2|1514|Loss: 0.8628339171409607:  23%|██▎       | 1514/6470 [17:24<1:10:04,  1.18it/s]
2|1515|Loss: 0.9359538555145264:  23%|██▎       | 1514/6470 [17:24<1:10:04,  1.18it/s]
2|1515|Loss: 0.9359538555145264:  23%|██▎       | 1515/6470 [17:24<1:09:31,  1.19it/s]
2|1516|Loss: 0.919757068157196:  23%|██▎       | 1515/6470 [17:25<1:09:31,  1.19it/s] 
2|1516|Loss: 0.919757068157196:  23%|██▎       | 1516/6470 [17:25<1:06:33,  1.24it/s]
2|1517|Loss: 0.8449095487594604:  23%|██▎       | 1516/6470 [17:26<1:06:33,  1.24it/s]
2|1517|Loss: 0.8449095487594604:  23%|██▎       | 1517/6470 [17:26<1:00:53,  1.36it/s]
2|1518|Loss: 0.9305006265640259:  23%|██▎       | 1517/6470 [17:26<1:00:53,  1.36it/s]
2|1518|Loss: 0.9305006265640259:  23%|██▎       | 1518/6470 [17:26<59:14,  1.39it/s]  
2|1519|Loss: 0.9348528385162354:  23%|██▎       | 1518/6470 [17:27<59:14,  1.39it/s]
2|1519|Loss: 0.9348528385162354:  23%|██▎       | 1519/6470 [17:27<59:51,  1.38it/s]
2|1520|Loss: 0.9446929693222046:  23%|██▎       | 1519/6470 [17:28<59:51,  1.38it/s]
2|1520|Loss: 0.9446929693222046:  23%|██▎       | 1520/6470 [17:28<55:42,  1.48it/s]
2|1521|Loss: 1.0685557126998901:  23%|██▎       | 1520/6470 [17:28<55:42,  1.48it/s]
2|1521|Loss: 1.0685557126998901:  24%|██▎       | 1521/6470 [17:28<54:25,  1.52it/s]
2|1522|Loss: 0.9974507093429565:  24%|██▎       | 1521/6470 [17:29<54:25,  1.52it/s]
2|1522|Loss: 0.9974507093429565:  24%|██▎       | 1522/6470 [17:29<56:17,  1.46it/s]
2|1523|Loss: 0.8984858393669128:  24%|██▎       | 1522/6470 [17:29<56:17,  1.46it/s]
2|1523|Loss: 0.8984858393669128:  24%|██▎       | 1523/6470 [17:30<53:12,  1.55it/s]
2|1524|Loss: 0.7435966730117798:  24%|██▎       | 1523/6470 [17:30<53:12,  1.55it/s]
2|1524|Loss: 0.7435966730117798:  24%|██▎       | 1524/6470 [17:30<56:31,  1.46it/s]
2|1525|Loss: 0.9249866604804993:  24%|██▎       | 1524/6470 [17:31<56:31,  1.46it/s]
2|1525|Loss: 0.9249866604804993:  24%|██▎       | 1525/6470 [17:31<58:20,  1.41it/s]
2|1526|Loss: 0.9235740303993225:  24%|██▎       | 1525/6470 [17:32<58:20,  1.41it/s]
2|1526|Loss: 0.9235740303993225:  24%|██▎       | 1526/6470 [17:32<56:03,  1.47it/s]
2|1527|Loss: 0.9064180254936218:  24%|██▎       | 1526/6470 [17:32<56:03,  1.47it/s]
2|1527|Loss: 0.9064180254936218:  24%|██▎       | 1527/6470 [17:33<56:53,  1.45it/s]
2|1528|Loss: 0.7584852576255798:  24%|██▎       | 1527/6470 [17:33<56:53,  1.45it/s]
2|1528|Loss: 0.7584852576255798:  24%|██▎       | 1528/6470 [17:33<56:14,  1.46it/s]
2|1529|Loss: 0.8124469518661499:  24%|██▎       | 1528/6470 [17:34<56:14,  1.46it/s]
2|1529|Loss: 0.8124469518661499:  24%|██▎       | 1529/6470 [17:34<57:14,  1.44it/s]
2|1530|Loss: 0.8215740323066711:  24%|██▎       | 1529/6470 [17:34<57:14,  1.44it/s]
2|1530|Loss: 0.8215740323066711:  24%|██▎       | 1530/6470 [17:34<53:49,  1.53it/s]
2|1531|Loss: 0.8733763098716736:  24%|██▎       | 1530/6470 [17:35<53:49,  1.53it/s]
2|1531|Loss: 0.8733763098716736:  24%|██▎       | 1531/6470 [17:35<51:33,  1.60it/s]
2|1532|Loss: 0.9226671457290649:  24%|██▎       | 1531/6470 [17:35<51:33,  1.60it/s]
2|1532|Loss: 0.9226671457290649:  24%|██▎       | 1532/6470 [17:36<51:34,  1.60it/s]
2|1533|Loss: 0.8826578259468079:  24%|██▎       | 1532/6470 [17:36<51:34,  1.60it/s]
2|1533|Loss: 0.8826578259468079:  24%|██▎       | 1533/6470 [17:37<56:14,  1.46it/s]
2|1534|Loss: 0.9182551503181458:  24%|██▎       | 1533/6470 [17:37<56:14,  1.46it/s]
2|1534|Loss: 0.9182551503181458:  24%|██▎       | 1534/6470 [17:37<59:59,  1.37it/s]
2|1535|Loss: 1.0133877992630005:  24%|██▎       | 1534/6470 [17:38<59:59,  1.37it/s]
2|1535|Loss: 1.0133877992630005:  24%|██▎       | 1535/6470 [17:38<59:57,  1.37it/s]
2|1536|Loss: 0.9946964383125305:  24%|██▎       | 1535/6470 [17:38<59:57,  1.37it/s]
2|1536|Loss: 0.9946964383125305:  24%|██▎       | 1536/6470 [17:39<58:09,  1.41it/s]
2|1537|Loss: 0.7931358814239502:  24%|██▎       | 1536/6470 [17:39<58:09,  1.41it/s]
2|1537|Loss: 0.7931358814239502:  24%|██▍       | 1537/6470 [17:39<57:35,  1.43it/s]
2|1538|Loss: 0.9405643939971924:  24%|██▍       | 1537/6470 [17:40<57:35,  1.43it/s]
2|1538|Loss: 0.9405643939971924:  24%|██▍       | 1538/6470 [17:40<56:31,  1.45it/s]
2|1539|Loss: 0.7499700784683228:  24%|██▍       | 1538/6470 [17:40<56:31,  1.45it/s]
2|1539|Loss: 0.7499700784683228:  24%|██▍       | 1539/6470 [17:41<55:27,  1.48it/s]
2|1540|Loss: 0.7089075446128845:  24%|██▍       | 1539/6470 [17:41<55:27,  1.48it/s]
2|1540|Loss: 0.7089075446128845:  24%|██▍       | 1540/6470 [17:42<1:01:26,  1.34it/s]
2|1541|Loss: 0.8034257888793945:  24%|██▍       | 1540/6470 [17:42<1:01:26,  1.34it/s]
2|1541|Loss: 0.8034257888793945:  24%|██▍       | 1541/6470 [17:43<1:04:52,  1.27it/s]
2|1542|Loss: 0.8309835195541382:  24%|██▍       | 1541/6470 [17:43<1:04:52,  1.27it/s]
2|1542|Loss: 0.8309835195541382:  24%|██▍       | 1542/6470 [17:43<1:01:52,  1.33it/s]
2|1543|Loss: 0.735012948513031:  24%|██▍       | 1542/6470 [17:44<1:01:52,  1.33it/s] 
2|1543|Loss: 0.735012948513031:  24%|██▍       | 1543/6470 [17:44<1:01:09,  1.34it/s]
2|1544|Loss: 0.8422808647155762:  24%|██▍       | 1543/6470 [17:44<1:01:09,  1.34it/s]
2|1544|Loss: 0.8422808647155762:  24%|██▍       | 1544/6470 [17:45<1:01:07,  1.34it/s]
2|1545|Loss: 0.818607747554779:  24%|██▍       | 1544/6470 [17:45<1:01:07,  1.34it/s] 
2|1545|Loss: 0.818607747554779:  24%|██▍       | 1545/6470 [17:45<58:31,  1.40it/s]  
2|1546|Loss: 0.9717308282852173:  24%|██▍       | 1545/6470 [17:46<58:31,  1.40it/s]
2|1546|Loss: 0.9717308282852173:  24%|██▍       | 1546/6470 [17:46<57:50,  1.42it/s]
2|1547|Loss: 0.6515008807182312:  24%|██▍       | 1546/6470 [17:46<57:50,  1.42it/s]
2|1547|Loss: 0.6515008807182312:  24%|██▍       | 1547/6470 [17:47<55:22,  1.48it/s]
2|1548|Loss: 0.7777704000473022:  24%|██▍       | 1547/6470 [17:47<55:22,  1.48it/s]
2|1548|Loss: 0.7777704000473022:  24%|██▍       | 1548/6470 [17:47<53:47,  1.53it/s]
2|1549|Loss: 0.8825958967208862:  24%|██▍       | 1548/6470 [17:48<53:47,  1.53it/s]
2|1549|Loss: 0.8825958967208862:  24%|██▍       | 1549/6470 [17:48<50:17,  1.63it/s]
2|1550|Loss: 0.9125272631645203:  24%|██▍       | 1549/6470 [17:48<50:17,  1.63it/s]
2|1550|Loss: 0.9125272631645203:  24%|██▍       | 1550/6470 [17:48<48:19,  1.70it/s]
2|1551|Loss: 0.8862332105636597:  24%|██▍       | 1550/6470 [17:49<48:19,  1.70it/s]
2|1551|Loss: 0.8862332105636597:  24%|██▍       | 1551/6470 [17:49<47:20,  1.73it/s]
2|1552|Loss: 0.7468885183334351:  24%|██▍       | 1551/6470 [17:49<47:20,  1.73it/s]
2|1552|Loss: 0.7468885183334351:  24%|██▍       | 1552/6470 [17:49<47:51,  1.71it/s]
2|1553|Loss: 0.8580788373947144:  24%|██▍       | 1552/6470 [17:50<47:51,  1.71it/s]
2|1553|Loss: 0.8580788373947144:  24%|██▍       | 1553/6470 [17:50<47:51,  1.71it/s]
2|1554|Loss: 0.9391823410987854:  24%|██▍       | 1553/6470 [17:50<47:51,  1.71it/s]
2|1554|Loss: 0.9391823410987854:  24%|██▍       | 1554/6470 [17:51<53:17,  1.54it/s]
2|1555|Loss: 0.9443696141242981:  24%|██▍       | 1554/6470 [17:51<53:17,  1.54it/s]
2|1555|Loss: 0.9443696141242981:  24%|██▍       | 1555/6470 [17:51<54:14,  1.51it/s]
2|1556|Loss: 0.9590346813201904:  24%|██▍       | 1555/6470 [17:52<54:14,  1.51it/s]
2|1556|Loss: 0.9590346813201904:  24%|██▍       | 1556/6470 [17:52<54:49,  1.49it/s]
2|1557|Loss: 0.8885207176208496:  24%|██▍       | 1556/6470 [17:53<54:49,  1.49it/s]
2|1557|Loss: 0.8885207176208496:  24%|██▍       | 1557/6470 [17:53<52:08,  1.57it/s]
2|1558|Loss: 0.9302965998649597:  24%|██▍       | 1557/6470 [17:53<52:08,  1.57it/s]
2|1558|Loss: 0.9302965998649597:  24%|██▍       | 1558/6470 [17:53<48:35,  1.68it/s]
2|1559|Loss: 0.9176344275474548:  24%|██▍       | 1558/6470 [17:54<48:35,  1.68it/s]
2|1559|Loss: 0.9176344275474548:  24%|██▍       | 1559/6470 [17:54<47:09,  1.74it/s]
2|1560|Loss: 0.8206849098205566:  24%|██▍       | 1559/6470 [17:54<47:09,  1.74it/s]
2|1560|Loss: 0.8206849098205566:  24%|██▍       | 1560/6470 [17:54<47:12,  1.73it/s]
2|1561|Loss: 0.6273446679115295:  24%|██▍       | 1560/6470 [17:55<47:12,  1.73it/s]
2|1561|Loss: 0.6273446679115295:  24%|██▍       | 1561/6470 [17:55<54:49,  1.49it/s]
2|1562|Loss: 0.7346681356430054:  24%|██▍       | 1561/6470 [17:56<54:49,  1.49it/s]
2|1562|Loss: 0.7346681356430054:  24%|██▍       | 1562/6470 [17:56<54:28,  1.50it/s]
2|1563|Loss: 0.8408735990524292:  24%|██▍       | 1562/6470 [17:56<54:28,  1.50it/s]
2|1563|Loss: 0.8408735990524292:  24%|██▍       | 1563/6470 [17:56<51:47,  1.58it/s]
2|1564|Loss: 0.898219883441925:  24%|██▍       | 1563/6470 [17:57<51:47,  1.58it/s] 
2|1564|Loss: 0.898219883441925:  24%|██▍       | 1564/6470 [17:57<53:41,  1.52it/s]
2|1565|Loss: 0.8510789275169373:  24%|██▍       | 1564/6470 [17:58<53:41,  1.52it/s]
2|1565|Loss: 0.8510789275169373:  24%|██▍       | 1565/6470 [17:58<55:35,  1.47it/s]
2|1566|Loss: 0.7741266489028931:  24%|██▍       | 1565/6470 [17:58<55:35,  1.47it/s]
2|1566|Loss: 0.7741266489028931:  24%|██▍       | 1566/6470 [17:58<50:00,  1.63it/s]
2|1567|Loss: 0.8065528869628906:  24%|██▍       | 1566/6470 [17:59<50:00,  1.63it/s]
2|1567|Loss: 0.8065528869628906:  24%|██▍       | 1567/6470 [17:59<50:47,  1.61it/s]
2|1568|Loss: 1.097254753112793:  24%|██▍       | 1567/6470 [17:59<50:47,  1.61it/s] 
2|1568|Loss: 1.097254753112793:  24%|██▍       | 1568/6470 [18:00<52:51,  1.55it/s]
2|1569|Loss: 0.9017952084541321:  24%|██▍       | 1568/6470 [18:00<52:51,  1.55it/s]
2|1569|Loss: 0.9017952084541321:  24%|██▍       | 1569/6470 [18:00<51:14,  1.59it/s]
2|1570|Loss: 0.8406296968460083:  24%|██▍       | 1569/6470 [18:01<51:14,  1.59it/s]
2|1570|Loss: 0.8406296968460083:  24%|██▍       | 1570/6470 [18:01<51:18,  1.59it/s]
2|1571|Loss: 0.7528306841850281:  24%|██▍       | 1570/6470 [18:01<51:18,  1.59it/s]
2|1571|Loss: 0.7528306841850281:  24%|██▍       | 1571/6470 [18:02<52:52,  1.54it/s]
2|1572|Loss: 0.9209325909614563:  24%|██▍       | 1571/6470 [18:02<52:52,  1.54it/s]
2|1572|Loss: 0.9209325909614563:  24%|██▍       | 1572/6470 [18:02<54:43,  1.49it/s]
2|1573|Loss: 0.9259839653968811:  24%|██▍       | 1572/6470 [18:03<54:43,  1.49it/s]
2|1573|Loss: 0.9259839653968811:  24%|██▍       | 1573/6470 [18:03<58:37,  1.39it/s]
2|1574|Loss: 0.905498743057251:  24%|██▍       | 1573/6470 [18:04<58:37,  1.39it/s] 
2|1574|Loss: 0.905498743057251:  24%|██▍       | 1574/6470 [18:04<1:00:10,  1.36it/s]
2|1575|Loss: 0.7845438718795776:  24%|██▍       | 1574/6470 [18:04<1:00:10,  1.36it/s]
2|1575|Loss: 0.7845438718795776:  24%|██▍       | 1575/6470 [18:05<1:00:14,  1.35it/s]
2|1576|Loss: 0.7991395592689514:  24%|██▍       | 1575/6470 [18:05<1:00:14,  1.35it/s]
2|1576|Loss: 0.7991395592689514:  24%|██▍       | 1576/6470 [18:05<59:36,  1.37it/s]  
2|1577|Loss: 1.0065239667892456:  24%|██▍       | 1576/6470 [18:06<59:36,  1.37it/s]
2|1577|Loss: 1.0065239667892456:  24%|██▍       | 1577/6470 [18:06<57:12,  1.43it/s]
2|1578|Loss: 0.9054169058799744:  24%|██▍       | 1577/6470 [18:06<57:12,  1.43it/s]
2|1578|Loss: 0.9054169058799744:  24%|██▍       | 1578/6470 [18:07<55:04,  1.48it/s]
2|1579|Loss: 1.0079185962677002:  24%|██▍       | 1578/6470 [18:07<55:04,  1.48it/s]
2|1579|Loss: 1.0079185962677002:  24%|██▍       | 1579/6470 [18:07<51:55,  1.57it/s]
2|1580|Loss: 0.9268884658813477:  24%|██▍       | 1579/6470 [18:07<51:55,  1.57it/s]
2|1580|Loss: 0.9268884658813477:  24%|██▍       | 1580/6470 [18:08<50:19,  1.62it/s]
2|1581|Loss: 0.8662062883377075:  24%|██▍       | 1580/6470 [18:08<50:19,  1.62it/s]
2|1581|Loss: 0.8662062883377075:  24%|██▍       | 1581/6470 [18:08<52:52,  1.54it/s]
2|1582|Loss: 0.9266058206558228:  24%|██▍       | 1581/6470 [18:09<52:52,  1.54it/s]
2|1582|Loss: 0.9266058206558228:  24%|██▍       | 1582/6470 [18:09<55:04,  1.48it/s]
2|1583|Loss: 0.7868478298187256:  24%|██▍       | 1582/6470 [18:10<55:04,  1.48it/s]
2|1583|Loss: 0.7868478298187256:  24%|██▍       | 1583/6470 [18:10<1:00:44,  1.34it/s]
2|1584|Loss: 0.8018652200698853:  24%|██▍       | 1583/6470 [18:11<1:00:44,  1.34it/s]
2|1584|Loss: 0.8018652200698853:  24%|██▍       | 1584/6470 [18:11<1:01:43,  1.32it/s]
2|1585|Loss: 0.9681780338287354:  24%|██▍       | 1584/6470 [18:11<1:01:43,  1.32it/s]
2|1585|Loss: 0.9681780338287354:  24%|██▍       | 1585/6470 [18:11<55:37,  1.46it/s]  
2|1586|Loss: 1.0324714183807373:  24%|██▍       | 1585/6470 [18:12<55:37,  1.46it/s]
2|1586|Loss: 1.0324714183807373:  25%|██▍       | 1586/6470 [18:12<52:06,  1.56it/s]
2|1587|Loss: 0.9666783809661865:  25%|██▍       | 1586/6470 [18:12<52:06,  1.56it/s]
2|1587|Loss: 0.9666783809661865:  25%|██▍       | 1587/6470 [18:13<51:25,  1.58it/s]
2|1588|Loss: 0.9315651059150696:  25%|██▍       | 1587/6470 [18:13<51:25,  1.58it/s]
2|1588|Loss: 0.9315651059150696:  25%|██▍       | 1588/6470 [18:13<53:33,  1.52it/s]
2|1589|Loss: 0.9798939824104309:  25%|██▍       | 1588/6470 [18:14<53:33,  1.52it/s]
2|1589|Loss: 0.9798939824104309:  25%|██▍       | 1589/6470 [18:14<57:45,  1.41it/s]
2|1590|Loss: 1.0062029361724854:  25%|██▍       | 1589/6470 [18:15<57:45,  1.41it/s]
2|1590|Loss: 1.0062029361724854:  25%|██▍       | 1590/6470 [18:15<57:09,  1.42it/s]
2|1591|Loss: 0.8097586035728455:  25%|██▍       | 1590/6470 [18:15<57:09,  1.42it/s]
2|1591|Loss: 0.8097586035728455:  25%|██▍       | 1591/6470 [18:16<1:00:51,  1.34it/s]
2|1592|Loss: 0.797060489654541:  25%|██▍       | 1591/6470 [18:16<1:00:51,  1.34it/s] 
2|1592|Loss: 0.797060489654541:  25%|██▍       | 1592/6470 [18:16<1:01:32,  1.32it/s]
2|1593|Loss: 0.84674471616745:  25%|██▍       | 1592/6470 [18:17<1:01:32,  1.32it/s] 
2|1593|Loss: 0.84674471616745:  25%|██▍       | 1593/6470 [18:17<1:01:21,  1.32it/s]
2|1594|Loss: 1.0071001052856445:  25%|██▍       | 1593/6470 [18:18<1:01:21,  1.32it/s]
2|1594|Loss: 1.0071001052856445:  25%|██▍       | 1594/6470 [18:18<59:17,  1.37it/s]  
2|1595|Loss: 0.936284065246582:  25%|██▍       | 1594/6470 [18:18<59:17,  1.37it/s] 
2|1595|Loss: 0.936284065246582:  25%|██▍       | 1595/6470 [18:19<1:00:03,  1.35it/s]
2|1596|Loss: 0.9612950086593628:  25%|██▍       | 1595/6470 [18:19<1:00:03,  1.35it/s]
2|1596|Loss: 0.9612950086593628:  25%|██▍       | 1596/6470 [18:19<56:51,  1.43it/s]  
2|1597|Loss: 0.8685206174850464:  25%|██▍       | 1596/6470 [18:20<56:51,  1.43it/s]
2|1597|Loss: 0.8685206174850464:  25%|██▍       | 1597/6470 [18:20<56:10,  1.45it/s]
2|1598|Loss: 0.9143015742301941:  25%|██▍       | 1597/6470 [18:20<56:10,  1.45it/s]
2|1598|Loss: 0.9143015742301941:  25%|██▍       | 1598/6470 [18:21<55:16,  1.47it/s]
2|1599|Loss: 0.7761286497116089:  25%|██▍       | 1598/6470 [18:21<55:16,  1.47it/s]
2|1599|Loss: 0.7761286497116089:  25%|██▍       | 1599/6470 [18:21<1:00:34,  1.34it/s]
2|1600|Loss: 0.9949029088020325:  25%|██▍       | 1599/6470 [18:22<1:00:34,  1.34it/s]
2|1600|Loss: 0.9949029088020325:  25%|██▍       | 1600/6470 [18:22<59:18,  1.37it/s]  
2|1601|Loss: 0.7881694436073303:  25%|██▍       | 1600/6470 [18:23<59:18,  1.37it/s]
2|1601|Loss: 0.7881694436073303:  25%|██▍       | 1601/6470 [18:23<58:49,  1.38it/s]
2|1602|Loss: 0.7984743714332581:  25%|██▍       | 1601/6470 [18:23<58:49,  1.38it/s]
2|1602|Loss: 0.7984743714332581:  25%|██▍       | 1602/6470 [18:24<1:06:18,  1.22it/s]
2|1603|Loss: 0.870535671710968:  25%|██▍       | 1602/6470 [18:24<1:06:18,  1.22it/s] 
2|1603|Loss: 0.870535671710968:  25%|██▍       | 1603/6470 [18:25<1:09:00,  1.18it/s]
2|1604|Loss: 0.750140368938446:  25%|██▍       | 1603/6470 [18:25<1:09:00,  1.18it/s]
2|1604|Loss: 0.750140368938446:  25%|██▍       | 1604/6470 [18:26<1:06:31,  1.22it/s]
2|1605|Loss: 0.8540017008781433:  25%|██▍       | 1604/6470 [18:26<1:06:31,  1.22it/s]
2|1605|Loss: 0.8540017008781433:  25%|██▍       | 1605/6470 [18:26<1:06:04,  1.23it/s]
2|1606|Loss: 0.883962094783783:  25%|██▍       | 1605/6470 [18:27<1:06:04,  1.23it/s] 
2|1606|Loss: 0.883962094783783:  25%|██▍       | 1606/6470 [18:27<1:02:47,  1.29it/s]
2|1607|Loss: 1.0234051942825317:  25%|██▍       | 1606/6470 [18:27<1:02:47,  1.29it/s]
2|1607|Loss: 1.0234051942825317:  25%|██▍       | 1607/6470 [18:28<59:26,  1.36it/s]  
2|1608|Loss: 0.7891058921813965:  25%|██▍       | 1607/6470 [18:28<59:26,  1.36it/s]
2|1608|Loss: 0.7891058921813965:  25%|██▍       | 1608/6470 [18:28<1:01:17,  1.32it/s]
2|1609|Loss: 0.8804105520248413:  25%|██▍       | 1608/6470 [18:29<1:01:17,  1.32it/s]
2|1609|Loss: 0.8804105520248413:  25%|██▍       | 1609/6470 [18:29<58:17,  1.39it/s]  
2|1610|Loss: 0.7726536393165588:  25%|██▍       | 1609/6470 [18:29<58:17,  1.39it/s]
2|1610|Loss: 0.7726536393165588:  25%|██▍       | 1610/6470 [18:30<55:58,  1.45it/s]
2|1611|Loss: 0.7654998898506165:  25%|██▍       | 1610/6470 [18:30<55:58,  1.45it/s]
2|1611|Loss: 0.7654998898506165:  25%|██▍       | 1611/6470 [18:30<54:17,  1.49it/s]
2|1612|Loss: 0.8424176573753357:  25%|██▍       | 1611/6470 [18:31<54:17,  1.49it/s]
2|1612|Loss: 0.8424176573753357:  25%|██▍       | 1612/6470 [18:31<51:35,  1.57it/s]
2|1613|Loss: 0.8410045504570007:  25%|██▍       | 1612/6470 [18:31<51:35,  1.57it/s]
2|1613|Loss: 0.8410045504570007:  25%|██▍       | 1613/6470 [18:31<47:31,  1.70it/s]
2|1614|Loss: 1.0074292421340942:  25%|██▍       | 1613/6470 [18:32<47:31,  1.70it/s]
2|1614|Loss: 1.0074292421340942:  25%|██▍       | 1614/6470 [18:32<47:32,  1.70it/s]
2|1615|Loss: 0.865332841873169:  25%|██▍       | 1614/6470 [18:32<47:32,  1.70it/s] 
2|1615|Loss: 0.865332841873169:  25%|██▍       | 1615/6470 [18:33<48:57,  1.65it/s]
2|1616|Loss: 0.7554599642753601:  25%|██▍       | 1615/6470 [18:33<48:57,  1.65it/s]
2|1616|Loss: 0.7554599642753601:  25%|██▍       | 1616/6470 [18:33<47:41,  1.70it/s]
2|1617|Loss: 0.702211856842041:  25%|██▍       | 1616/6470 [18:34<47:41,  1.70it/s] 
2|1617|Loss: 0.702211856842041:  25%|██▍       | 1617/6470 [18:34<52:20,  1.55it/s]
2|1618|Loss: 0.8209171891212463:  25%|██▍       | 1617/6470 [18:34<52:20,  1.55it/s]
2|1618|Loss: 0.8209171891212463:  25%|██▌       | 1618/6470 [18:35<55:39,  1.45it/s]
2|1619|Loss: 0.8311169743537903:  25%|██▌       | 1618/6470 [18:35<55:39,  1.45it/s]
2|1619|Loss: 0.8311169743537903:  25%|██▌       | 1619/6470 [18:35<49:50,  1.62it/s]
2|1620|Loss: 0.954667329788208:  25%|██▌       | 1619/6470 [18:36<49:50,  1.62it/s] 
2|1620|Loss: 0.954667329788208:  25%|██▌       | 1620/6470 [18:36<50:21,  1.61it/s]
2|1621|Loss: 0.9089088439941406:  25%|██▌       | 1620/6470 [18:36<50:21,  1.61it/s]
2|1621|Loss: 0.9089088439941406:  25%|██▌       | 1621/6470 [18:37<53:07,  1.52it/s]
2|1622|Loss: 0.7992665767669678:  25%|██▌       | 1621/6470 [18:37<53:07,  1.52it/s]
2|1622|Loss: 0.7992665767669678:  25%|██▌       | 1622/6470 [18:37<51:59,  1.55it/s]
2|1623|Loss: 0.7691761255264282:  25%|██▌       | 1622/6470 [18:38<51:59,  1.55it/s]
2|1623|Loss: 0.7691761255264282:  25%|██▌       | 1623/6470 [18:38<49:07,  1.64it/s]
2|1624|Loss: 0.7356697916984558:  25%|██▌       | 1623/6470 [18:38<49:07,  1.64it/s]
2|1624|Loss: 0.7356697916984558:  25%|██▌       | 1624/6470 [18:39<54:33,  1.48it/s]
2|1625|Loss: 0.9418472051620483:  25%|██▌       | 1624/6470 [18:39<54:33,  1.48it/s]
2|1625|Loss: 0.9418472051620483:  25%|██▌       | 1625/6470 [18:39<58:10,  1.39it/s]
2|1626|Loss: 0.7559250593185425:  25%|██▌       | 1625/6470 [18:40<58:10,  1.39it/s]
2|1626|Loss: 0.7559250593185425:  25%|██▌       | 1626/6470 [18:40<59:11,  1.36it/s]
2|1627|Loss: 0.9407552480697632:  25%|██▌       | 1626/6470 [18:41<59:11,  1.36it/s]
2|1627|Loss: 0.9407552480697632:  25%|██▌       | 1627/6470 [18:41<1:00:40,  1.33it/s]
2|1628|Loss: 0.9425938129425049:  25%|██▌       | 1627/6470 [18:41<1:00:40,  1.33it/s]
2|1628|Loss: 0.9425938129425049:  25%|██▌       | 1628/6470 [18:42<1:01:10,  1.32it/s]
2|1629|Loss: 0.8920960426330566:  25%|██▌       | 1628/6470 [18:42<1:01:10,  1.32it/s]
2|1629|Loss: 0.8920960426330566:  25%|██▌       | 1629/6470 [18:43<1:05:12,  1.24it/s]
2|1630|Loss: 0.884423553943634:  25%|██▌       | 1629/6470 [18:43<1:05:12,  1.24it/s] 
2|1630|Loss: 0.884423553943634:  25%|██▌       | 1630/6470 [18:43<1:05:05,  1.24it/s]
2|1631|Loss: 0.9401628375053406:  25%|██▌       | 1630/6470 [18:44<1:05:05,  1.24it/s]
2|1631|Loss: 0.9401628375053406:  25%|██▌       | 1631/6470 [18:44<1:03:03,  1.28it/s]
2|1632|Loss: 0.7763869762420654:  25%|██▌       | 1631/6470 [18:45<1:03:03,  1.28it/s]
2|1632|Loss: 0.7763869762420654:  25%|██▌       | 1632/6470 [18:45<1:03:38,  1.27it/s]
2|1633|Loss: 0.912151038646698:  25%|██▌       | 1632/6470 [18:45<1:03:38,  1.27it/s] 
2|1633|Loss: 0.912151038646698:  25%|██▌       | 1633/6470 [18:46<1:06:19,  1.22it/s]
2|1634|Loss: 0.8750141263008118:  25%|██▌       | 1633/6470 [18:46<1:06:19,  1.22it/s]
2|1634|Loss: 0.8750141263008118:  25%|██▌       | 1634/6470 [18:47<1:02:17,  1.29it/s]
2|1635|Loss: 0.8833456039428711:  25%|██▌       | 1634/6470 [18:47<1:02:17,  1.29it/s]
2|1635|Loss: 0.8833456039428711:  25%|██▌       | 1635/6470 [18:47<58:18,  1.38it/s]  
2|1636|Loss: 0.7776247262954712:  25%|██▌       | 1635/6470 [18:47<58:18,  1.38it/s]
2|1636|Loss: 0.7776247262954712:  25%|██▌       | 1636/6470 [18:48<54:03,  1.49it/s]
2|1637|Loss: 0.8510364294052124:  25%|██▌       | 1636/6470 [18:48<54:03,  1.49it/s]
2|1637|Loss: 0.8510364294052124:  25%|██▌       | 1637/6470 [18:48<54:23,  1.48it/s]
2|1638|Loss: 0.8177833557128906:  25%|██▌       | 1637/6470 [18:49<54:23,  1.48it/s]
2|1638|Loss: 0.8177833557128906:  25%|██▌       | 1638/6470 [18:49<54:08,  1.49it/s]
2|1639|Loss: 0.8269565105438232:  25%|██▌       | 1638/6470 [18:49<54:08,  1.49it/s]
2|1639|Loss: 0.8269565105438232:  25%|██▌       | 1639/6470 [18:50<52:47,  1.53it/s]
2|1640|Loss: 0.8231269717216492:  25%|██▌       | 1639/6470 [18:50<52:47,  1.53it/s]
2|1640|Loss: 0.8231269717216492:  25%|██▌       | 1640/6470 [18:50<53:27,  1.51it/s]
2|1641|Loss: 0.9037588238716125:  25%|██▌       | 1640/6470 [18:51<53:27,  1.51it/s]
2|1641|Loss: 0.9037588238716125:  25%|██▌       | 1641/6470 [18:51<55:32,  1.45it/s]
2|1642|Loss: 0.8297854065895081:  25%|██▌       | 1641/6470 [18:51<55:32,  1.45it/s]
2|1642|Loss: 0.8297854065895081:  25%|██▌       | 1642/6470 [18:52<54:39,  1.47it/s]
2|1643|Loss: 0.8828068375587463:  25%|██▌       | 1642/6470 [18:52<54:39,  1.47it/s]
2|1643|Loss: 0.8828068375587463:  25%|██▌       | 1643/6470 [18:52<54:48,  1.47it/s]
2|1644|Loss: 0.7690914869308472:  25%|██▌       | 1643/6470 [18:53<54:48,  1.47it/s]
2|1644|Loss: 0.7690914869308472:  25%|██▌       | 1644/6470 [18:53<53:37,  1.50it/s]
2|1645|Loss: 0.9126936793327332:  25%|██▌       | 1644/6470 [18:53<53:37,  1.50it/s]
2|1645|Loss: 0.9126936793327332:  25%|██▌       | 1645/6470 [18:54<56:21,  1.43it/s]
2|1646|Loss: 0.6597551703453064:  25%|██▌       | 1645/6470 [18:54<56:21,  1.43it/s]
2|1646|Loss: 0.6597551703453064:  25%|██▌       | 1646/6470 [18:54<54:37,  1.47it/s]
2|1647|Loss: 0.7642890810966492:  25%|██▌       | 1646/6470 [18:55<54:37,  1.47it/s]
2|1647|Loss: 0.7642890810966492:  25%|██▌       | 1647/6470 [18:55<53:27,  1.50it/s]
2|1648|Loss: 0.848898708820343:  25%|██▌       | 1647/6470 [18:55<53:27,  1.50it/s] 
2|1648|Loss: 0.848898708820343:  25%|██▌       | 1648/6470 [18:56<47:40,  1.69it/s]
2|1649|Loss: 0.7722334861755371:  25%|██▌       | 1648/6470 [18:56<47:40,  1.69it/s]
2|1649|Loss: 0.7722334861755371:  25%|██▌       | 1649/6470 [18:56<51:04,  1.57it/s]
2|1650|Loss: 0.8327193260192871:  25%|██▌       | 1649/6470 [18:57<51:04,  1.57it/s]
2|1650|Loss: 0.8327193260192871:  26%|██▌       | 1650/6470 [18:57<49:46,  1.61it/s]
2|1651|Loss: 0.8769063353538513:  26%|██▌       | 1650/6470 [18:57<49:46,  1.61it/s]
2|1651|Loss: 0.8769063353538513:  26%|██▌       | 1651/6470 [18:58<51:29,  1.56it/s]
2|1652|Loss: 0.6164031624794006:  26%|██▌       | 1651/6470 [18:58<51:29,  1.56it/s]
2|1652|Loss: 0.6164031624794006:  26%|██▌       | 1652/6470 [18:58<56:05,  1.43it/s]
2|1653|Loss: 0.8764380812644958:  26%|██▌       | 1652/6470 [18:59<56:05,  1.43it/s]
2|1653|Loss: 0.8764380812644958:  26%|██▌       | 1653/6470 [18:59<53:30,  1.50it/s]
2|1654|Loss: 0.9382756352424622:  26%|██▌       | 1653/6470 [18:59<53:30,  1.50it/s]
2|1654|Loss: 0.9382756352424622:  26%|██▌       | 1654/6470 [19:00<54:09,  1.48it/s]
2|1655|Loss: 1.1042035818099976:  26%|██▌       | 1654/6470 [19:00<54:09,  1.48it/s]
2|1655|Loss: 1.1042035818099976:  26%|██▌       | 1655/6470 [19:00<55:07,  1.46it/s]
2|1656|Loss: 0.9464076161384583:  26%|██▌       | 1655/6470 [19:01<55:07,  1.46it/s]
2|1656|Loss: 0.9464076161384583:  26%|██▌       | 1656/6470 [19:01<56:21,  1.42it/s]
2|1657|Loss: 0.9552725553512573:  26%|██▌       | 1656/6470 [19:01<56:21,  1.42it/s]
2|1657|Loss: 0.9552725553512573:  26%|██▌       | 1657/6470 [19:02<54:16,  1.48it/s]
2|1658|Loss: 0.7777406573295593:  26%|██▌       | 1657/6470 [19:02<54:16,  1.48it/s]
2|1658|Loss: 0.7777406573295593:  26%|██▌       | 1658/6470 [19:02<54:49,  1.46it/s]
2|1659|Loss: 0.6955928802490234:  26%|██▌       | 1658/6470 [19:03<54:49,  1.46it/s]
2|1659|Loss: 0.6955928802490234:  26%|██▌       | 1659/6470 [19:03<55:22,  1.45it/s]
2|1660|Loss: 0.9203100800514221:  26%|██▌       | 1659/6470 [19:03<55:22,  1.45it/s]
2|1660|Loss: 0.9203100800514221:  26%|██▌       | 1660/6470 [19:04<53:23,  1.50it/s]
2|1661|Loss: 0.7983505725860596:  26%|██▌       | 1660/6470 [19:04<53:23,  1.50it/s]
2|1661|Loss: 0.7983505725860596:  26%|██▌       | 1661/6470 [19:04<52:37,  1.52it/s]
2|1662|Loss: 0.7555540204048157:  26%|██▌       | 1661/6470 [19:05<52:37,  1.52it/s]
2|1662|Loss: 0.7555540204048157:  26%|██▌       | 1662/6470 [19:05<52:18,  1.53it/s]
2|1663|Loss: 0.807870090007782:  26%|██▌       | 1662/6470 [19:05<52:18,  1.53it/s] 
2|1663|Loss: 0.807870090007782:  26%|██▌       | 1663/6470 [19:06<51:32,  1.55it/s]
2|1664|Loss: 0.6770288944244385:  26%|██▌       | 1663/6470 [19:06<51:32,  1.55it/s]
2|1664|Loss: 0.6770288944244385:  26%|██▌       | 1664/6470 [19:06<56:31,  1.42it/s]
2|1665|Loss: 0.771878719329834:  26%|██▌       | 1664/6470 [19:07<56:31,  1.42it/s] 
2|1665|Loss: 0.771878719329834:  26%|██▌       | 1665/6470 [19:07<56:28,  1.42it/s]
2|1666|Loss: 0.7161482572555542:  26%|██▌       | 1665/6470 [19:07<56:28,  1.42it/s]
2|1666|Loss: 0.7161482572555542:  26%|██▌       | 1666/6470 [19:08<48:53,  1.64it/s]
2|1667|Loss: 0.9159272313117981:  26%|██▌       | 1666/6470 [19:08<48:53,  1.64it/s]
2|1667|Loss: 0.9159272313117981:  26%|██▌       | 1667/6470 [19:08<52:00,  1.54it/s]
2|1668|Loss: 0.8412101864814758:  26%|██▌       | 1667/6470 [19:09<52:00,  1.54it/s]
2|1668|Loss: 0.8412101864814758:  26%|██▌       | 1668/6470 [19:09<53:08,  1.51it/s]
2|1669|Loss: 0.8178385496139526:  26%|██▌       | 1668/6470 [19:09<53:08,  1.51it/s]
2|1669|Loss: 0.8178385496139526:  26%|██▌       | 1669/6470 [19:10<52:06,  1.54it/s]
2|1670|Loss: 0.8027860522270203:  26%|██▌       | 1669/6470 [19:10<52:06,  1.54it/s]
2|1670|Loss: 0.8027860522270203:  26%|██▌       | 1670/6470 [19:10<55:13,  1.45it/s]
2|1671|Loss: 0.7558317184448242:  26%|██▌       | 1670/6470 [19:11<55:13,  1.45it/s]
2|1671|Loss: 0.7558317184448242:  26%|██▌       | 1671/6470 [19:11<55:09,  1.45it/s]
2|1672|Loss: 0.8462637066841125:  26%|██▌       | 1671/6470 [19:11<55:09,  1.45it/s]
2|1672|Loss: 0.8462637066841125:  26%|██▌       | 1672/6470 [19:12<53:27,  1.50it/s]
2|1673|Loss: 0.858897864818573:  26%|██▌       | 1672/6470 [19:12<53:27,  1.50it/s] 
2|1673|Loss: 0.858897864818573:  26%|██▌       | 1673/6470 [19:12<52:06,  1.53it/s]
2|1674|Loss: 0.9071210026741028:  26%|██▌       | 1673/6470 [19:13<52:06,  1.53it/s]
2|1674|Loss: 0.9071210026741028:  26%|██▌       | 1674/6470 [19:13<49:57,  1.60it/s]
2|1675|Loss: 0.8305380940437317:  26%|██▌       | 1674/6470 [19:13<49:57,  1.60it/s]
2|1675|Loss: 0.8305380940437317:  26%|██▌       | 1675/6470 [19:14<52:45,  1.51it/s]
2|1676|Loss: 1.0691392421722412:  26%|██▌       | 1675/6470 [19:14<52:45,  1.51it/s]
2|1676|Loss: 1.0691392421722412:  26%|██▌       | 1676/6470 [19:14<57:10,  1.40it/s]
2|1677|Loss: 0.9485287666320801:  26%|██▌       | 1676/6470 [19:15<57:10,  1.40it/s]
2|1677|Loss: 0.9485287666320801:  26%|██▌       | 1677/6470 [19:15<56:13,  1.42it/s]
2|1678|Loss: 1.0338469743728638:  26%|██▌       | 1677/6470 [19:16<56:13,  1.42it/s]
2|1678|Loss: 1.0338469743728638:  26%|██▌       | 1678/6470 [19:16<56:31,  1.41it/s]
2|1679|Loss: 0.7435340285301208:  26%|██▌       | 1678/6470 [19:16<56:31,  1.41it/s]
2|1679|Loss: 0.7435340285301208:  26%|██▌       | 1679/6470 [19:17<55:28,  1.44it/s]
2|1680|Loss: 0.9781543612480164:  26%|██▌       | 1679/6470 [19:17<55:28,  1.44it/s]
2|1680|Loss: 0.9781543612480164:  26%|██▌       | 1680/6470 [19:17<57:39,  1.38it/s]
2|1681|Loss: 0.828478991985321:  26%|██▌       | 1680/6470 [19:18<57:39,  1.38it/s] 
2|1681|Loss: 0.828478991985321:  26%|██▌       | 1681/6470 [19:18<55:23,  1.44it/s]
2|1682|Loss: 0.7437726259231567:  26%|██▌       | 1681/6470 [19:18<55:23,  1.44it/s]
2|1682|Loss: 0.7437726259231567:  26%|██▌       | 1682/6470 [19:19<52:39,  1.52it/s]
2|1683|Loss: 0.9424126744270325:  26%|██▌       | 1682/6470 [19:19<52:39,  1.52it/s]
2|1683|Loss: 0.9424126744270325:  26%|██▌       | 1683/6470 [19:19<55:00,  1.45it/s]
2|1684|Loss: 0.9022418856620789:  26%|██▌       | 1683/6470 [19:20<55:00,  1.45it/s]
2|1684|Loss: 0.9022418856620789:  26%|██▌       | 1684/6470 [19:20<56:30,  1.41it/s]
2|1685|Loss: 0.8872526288032532:  26%|██▌       | 1684/6470 [19:20<56:30,  1.41it/s]
2|1685|Loss: 0.8872526288032532:  26%|██▌       | 1685/6470 [19:21<56:52,  1.40it/s]
2|1686|Loss: 0.8619183301925659:  26%|██▌       | 1685/6470 [19:21<56:52,  1.40it/s]
2|1686|Loss: 0.8619183301925659:  26%|██▌       | 1686/6470 [19:21<56:45,  1.40it/s]
2|1687|Loss: 0.7988340258598328:  26%|██▌       | 1686/6470 [19:22<56:45,  1.40it/s]
2|1687|Loss: 0.7988340258598328:  26%|██▌       | 1687/6470 [19:22<55:17,  1.44it/s]
2|1688|Loss: 0.8170689940452576:  26%|██▌       | 1687/6470 [19:23<55:17,  1.44it/s]
2|1688|Loss: 0.8170689940452576:  26%|██▌       | 1688/6470 [19:23<55:47,  1.43it/s]
2|1689|Loss: 0.8717911243438721:  26%|██▌       | 1688/6470 [19:23<55:47,  1.43it/s]
2|1689|Loss: 0.8717911243438721:  26%|██▌       | 1689/6470 [19:23<54:23,  1.47it/s]
2|1690|Loss: 0.8603699803352356:  26%|██▌       | 1689/6470 [19:24<54:23,  1.47it/s]
2|1690|Loss: 0.8603699803352356:  26%|██▌       | 1690/6470 [19:24<55:21,  1.44it/s]
2|1691|Loss: 0.851378858089447:  26%|██▌       | 1690/6470 [19:25<55:21,  1.44it/s] 
2|1691|Loss: 0.851378858089447:  26%|██▌       | 1691/6470 [19:25<56:04,  1.42it/s]
2|1692|Loss: 0.6777306199073792:  26%|██▌       | 1691/6470 [19:25<56:04,  1.42it/s]
2|1692|Loss: 0.6777306199073792:  26%|██▌       | 1692/6470 [19:26<1:01:43,  1.29it/s]
2|1693|Loss: 0.880466878414154:  26%|██▌       | 1692/6470 [19:26<1:01:43,  1.29it/s] 
2|1693|Loss: 0.880466878414154:  26%|██▌       | 1693/6470 [19:27<1:01:44,  1.29it/s]
2|1694|Loss: 0.930679440498352:  26%|██▌       | 1693/6470 [19:27<1:01:44,  1.29it/s]
2|1694|Loss: 0.930679440498352:  26%|██▌       | 1694/6470 [19:28<1:03:23,  1.26it/s]
2|1695|Loss: 0.707204282283783:  26%|██▌       | 1694/6470 [19:28<1:03:23,  1.26it/s]
2|1695|Loss: 0.707204282283783:  26%|██▌       | 1695/6470 [19:28<1:03:53,  1.25it/s]
2|1696|Loss: 0.8881130814552307:  26%|██▌       | 1695/6470 [19:29<1:03:53,  1.25it/s]
2|1696|Loss: 0.8881130814552307:  26%|██▌       | 1696/6470 [19:29<1:00:31,  1.31it/s]
2|1697|Loss: 0.7381898164749146:  26%|██▌       | 1696/6470 [19:29<1:00:31,  1.31it/s]
2|1697|Loss: 0.7381898164749146:  26%|██▌       | 1697/6470 [19:30<1:04:16,  1.24it/s]
2|1698|Loss: 0.9656015634536743:  26%|██▌       | 1697/6470 [19:30<1:04:16,  1.24it/s]
2|1698|Loss: 0.9656015634536743:  26%|██▌       | 1698/6470 [19:31<1:03:21,  1.26it/s]
2|1699|Loss: 0.8475449681282043:  26%|██▌       | 1698/6470 [19:31<1:03:21,  1.26it/s]
2|1699|Loss: 0.8475449681282043:  26%|██▋       | 1699/6470 [19:31<1:00:06,  1.32it/s]
2|1700|Loss: 0.8172438144683838:  26%|██▋       | 1699/6470 [19:32<1:00:06,  1.32it/s]
2|1700|Loss: 0.8172438144683838:  26%|██▋       | 1700/6470 [19:32<56:08,  1.42it/s]  
2|1701|Loss: 0.9423509836196899:  26%|██▋       | 1700/6470 [19:32<56:08,  1.42it/s]
2|1701|Loss: 0.9423509836196899:  26%|██▋       | 1701/6470 [19:33<53:10,  1.49it/s]
2|1702|Loss: 0.9100819230079651:  26%|██▋       | 1701/6470 [19:33<53:10,  1.49it/s]
2|1702|Loss: 0.9100819230079651:  26%|██▋       | 1702/6470 [19:33<55:17,  1.44it/s]
2|1703|Loss: 0.8463709354400635:  26%|██▋       | 1702/6470 [19:34<55:17,  1.44it/s]
2|1703|Loss: 0.8463709354400635:  26%|██▋       | 1703/6470 [19:34<53:46,  1.48it/s]
2|1704|Loss: 0.9346012473106384:  26%|██▋       | 1703/6470 [19:34<53:46,  1.48it/s]
2|1704|Loss: 0.9346012473106384:  26%|██▋       | 1704/6470 [19:35<54:23,  1.46it/s]
2|1705|Loss: 1.250161051750183:  26%|██▋       | 1704/6470 [19:35<54:23,  1.46it/s] 
2|1705|Loss: 1.250161051750183:  26%|██▋       | 1705/6470 [19:35<51:40,  1.54it/s]
2|1706|Loss: 0.7602797150611877:  26%|██▋       | 1705/6470 [19:35<51:40,  1.54it/s]
2|1706|Loss: 0.7602797150611877:  26%|██▋       | 1706/6470 [19:36<48:43,  1.63it/s]
2|1707|Loss: 1.0243618488311768:  26%|██▋       | 1706/6470 [19:36<48:43,  1.63it/s]
2|1707|Loss: 1.0243618488311768:  26%|██▋       | 1707/6470 [19:36<48:47,  1.63it/s]
2|1708|Loss: 0.8478094935417175:  26%|██▋       | 1707/6470 [19:37<48:47,  1.63it/s]
2|1708|Loss: 0.8478094935417175:  26%|██▋       | 1708/6470 [19:37<51:15,  1.55it/s]
2|1709|Loss: 0.8805791735649109:  26%|██▋       | 1708/6470 [19:37<51:15,  1.55it/s]
2|1709|Loss: 0.8805791735649109:  26%|██▋       | 1709/6470 [19:38<54:22,  1.46it/s]
2|1710|Loss: 0.9123123288154602:  26%|██▋       | 1709/6470 [19:38<54:22,  1.46it/s]
2|1710|Loss: 0.9123123288154602:  26%|██▋       | 1710/6470 [19:39<58:01,  1.37it/s]
2|1711|Loss: 0.7702955603599548:  26%|██▋       | 1710/6470 [19:39<58:01,  1.37it/s]
2|1711|Loss: 0.7702955603599548:  26%|██▋       | 1711/6470 [19:39<58:54,  1.35it/s]
2|1712|Loss: 0.85594242811203:  26%|██▋       | 1711/6470 [19:40<58:54,  1.35it/s]  
2|1712|Loss: 0.85594242811203:  26%|██▋       | 1712/6470 [19:40<58:16,  1.36it/s]
2|1713|Loss: 0.9829665422439575:  26%|██▋       | 1712/6470 [19:40<58:16,  1.36it/s]
2|1713|Loss: 0.9829665422439575:  26%|██▋       | 1713/6470 [19:41<55:13,  1.44it/s]
2|1714|Loss: 0.9442753195762634:  26%|██▋       | 1713/6470 [19:41<55:13,  1.44it/s]
2|1714|Loss: 0.9442753195762634:  26%|██▋       | 1714/6470 [19:41<53:22,  1.49it/s]
2|1715|Loss: 0.8003697991371155:  26%|██▋       | 1714/6470 [19:42<53:22,  1.49it/s]
2|1715|Loss: 0.8003697991371155:  27%|██▋       | 1715/6470 [19:42<50:53,  1.56it/s]
2|1716|Loss: 0.7540782690048218:  27%|██▋       | 1715/6470 [19:42<50:53,  1.56it/s]
2|1716|Loss: 0.7540782690048218:  27%|██▋       | 1716/6470 [19:43<50:03,  1.58it/s]
2|1717|Loss: 0.7743566632270813:  27%|██▋       | 1716/6470 [19:43<50:03,  1.58it/s]
2|1717|Loss: 0.7743566632270813:  27%|██▋       | 1717/6470 [19:43<47:40,  1.66it/s]
2|1718|Loss: 0.8126742839813232:  27%|██▋       | 1717/6470 [19:43<47:40,  1.66it/s]
2|1718|Loss: 0.8126742839813232:  27%|██▋       | 1718/6470 [19:44<51:52,  1.53it/s]
2|1719|Loss: 0.6738576889038086:  27%|██▋       | 1718/6470 [19:44<51:52,  1.53it/s]
2|1719|Loss: 0.6738576889038086:  27%|██▋       | 1719/6470 [19:45<59:23,  1.33it/s]
2|1720|Loss: 0.9455996155738831:  27%|██▋       | 1719/6470 [19:45<59:23,  1.33it/s]
2|1720|Loss: 0.9455996155738831:  27%|██▋       | 1720/6470 [19:46<1:00:55,  1.30it/s]
2|1721|Loss: 0.8927392959594727:  27%|██▋       | 1720/6470 [19:46<1:00:55,  1.30it/s]
2|1721|Loss: 0.8927392959594727:  27%|██▋       | 1721/6470 [19:46<58:09,  1.36it/s]  
2|1722|Loss: 1.0049666166305542:  27%|██▋       | 1721/6470 [19:47<58:09,  1.36it/s]
2|1722|Loss: 1.0049666166305542:  27%|██▋       | 1722/6470 [19:47<55:15,  1.43it/s]
2|1723|Loss: 0.870493471622467:  27%|██▋       | 1722/6470 [19:47<55:15,  1.43it/s] 
2|1723|Loss: 0.870493471622467:  27%|██▋       | 1723/6470 [19:48<57:22,  1.38it/s]
2|1724|Loss: 0.7561954855918884:  27%|██▋       | 1723/6470 [19:48<57:22,  1.38it/s]
2|1724|Loss: 0.7561954855918884:  27%|██▋       | 1724/6470 [19:48<55:43,  1.42it/s]
2|1725|Loss: 0.9458490610122681:  27%|██▋       | 1724/6470 [19:49<55:43,  1.42it/s]
2|1725|Loss: 0.9458490610122681:  27%|██▋       | 1725/6470 [19:49<55:57,  1.41it/s]
2|1726|Loss: 0.840431272983551:  27%|██▋       | 1725/6470 [19:50<55:57,  1.41it/s] 
2|1726|Loss: 0.840431272983551:  27%|██▋       | 1726/6470 [19:50<58:29,  1.35it/s]
2|1727|Loss: 0.6914034485816956:  27%|██▋       | 1726/6470 [19:50<58:29,  1.35it/s]
2|1727|Loss: 0.6914034485816956:  27%|██▋       | 1727/6470 [19:50<52:17,  1.51it/s]
2|1728|Loss: 0.8736437559127808:  27%|██▋       | 1727/6470 [19:51<52:17,  1.51it/s]
2|1728|Loss: 0.8736437559127808:  27%|██▋       | 1728/6470 [19:51<53:42,  1.47it/s]
2|1729|Loss: 0.8699348568916321:  27%|██▋       | 1728/6470 [19:52<53:42,  1.47it/s]
2|1729|Loss: 0.8699348568916321:  27%|██▋       | 1729/6470 [19:52<56:36,  1.40it/s]
2|1730|Loss: 0.7895297408103943:  27%|██▋       | 1729/6470 [19:52<56:36,  1.40it/s]
2|1730|Loss: 0.7895297408103943:  27%|██▋       | 1730/6470 [19:53<55:47,  1.42it/s]
2|1731|Loss: 1.1228159666061401:  27%|██▋       | 1730/6470 [19:53<55:47,  1.42it/s]
2|1731|Loss: 1.1228159666061401:  27%|██▋       | 1731/6470 [19:53<58:44,  1.34it/s]
2|1732|Loss: 0.7826098799705505:  27%|██▋       | 1731/6470 [19:54<58:44,  1.34it/s]
2|1732|Loss: 0.7826098799705505:  27%|██▋       | 1732/6470 [19:54<57:37,  1.37it/s]
2|1733|Loss: 1.0029042959213257:  27%|██▋       | 1732/6470 [19:54<57:37,  1.37it/s]
2|1733|Loss: 1.0029042959213257:  27%|██▋       | 1733/6470 [19:55<53:08,  1.49it/s]
2|1734|Loss: 0.7785689234733582:  27%|██▋       | 1733/6470 [19:55<53:08,  1.49it/s]
2|1734|Loss: 0.7785689234733582:  27%|██▋       | 1734/6470 [19:55<53:34,  1.47it/s]
2|1735|Loss: 0.7059670686721802:  27%|██▋       | 1734/6470 [19:56<53:34,  1.47it/s]
2|1735|Loss: 0.7059670686721802:  27%|██▋       | 1735/6470 [19:56<53:10,  1.48it/s]
2|1736|Loss: 0.9156854152679443:  27%|██▋       | 1735/6470 [19:56<53:10,  1.48it/s]
2|1736|Loss: 0.9156854152679443:  27%|██▋       | 1736/6470 [19:57<52:21,  1.51it/s]
2|1737|Loss: 0.9678387641906738:  27%|██▋       | 1736/6470 [19:57<52:21,  1.51it/s]
2|1737|Loss: 0.9678387641906738:  27%|██▋       | 1737/6470 [19:57<54:44,  1.44it/s]
2|1738|Loss: 0.8042243123054504:  27%|██▋       | 1737/6470 [19:58<54:44,  1.44it/s]
2|1738|Loss: 0.8042243123054504:  27%|██▋       | 1738/6470 [19:58<55:29,  1.42it/s]
2|1739|Loss: 0.9710779190063477:  27%|██▋       | 1738/6470 [19:58<55:29,  1.42it/s]
2|1739|Loss: 0.9710779190063477:  27%|██▋       | 1739/6470 [19:59<51:50,  1.52it/s]
2|1740|Loss: 0.9132823348045349:  27%|██▋       | 1739/6470 [19:59<51:50,  1.52it/s]
2|1740|Loss: 0.9132823348045349:  27%|██▋       | 1740/6470 [19:59<52:28,  1.50it/s]
2|1741|Loss: 0.8234910368919373:  27%|██▋       | 1740/6470 [20:00<52:28,  1.50it/s]
2|1741|Loss: 0.8234910368919373:  27%|██▋       | 1741/6470 [20:00<54:05,  1.46it/s]
2|1742|Loss: 0.8761899471282959:  27%|██▋       | 1741/6470 [20:00<54:05,  1.46it/s]
2|1742|Loss: 0.8761899471282959:  27%|██▋       | 1742/6470 [20:01<52:03,  1.51it/s]
2|1743|Loss: 0.8276594877243042:  27%|██▋       | 1742/6470 [20:01<52:03,  1.51it/s]
2|1743|Loss: 0.8276594877243042:  27%|██▋       | 1743/6470 [20:01<52:56,  1.49it/s]
2|1744|Loss: 0.9288832545280457:  27%|██▋       | 1743/6470 [20:02<52:56,  1.49it/s]
2|1744|Loss: 0.9288832545280457:  27%|██▋       | 1744/6470 [20:02<54:37,  1.44it/s]
2|1745|Loss: 0.9135190844535828:  27%|██▋       | 1744/6470 [20:02<54:37,  1.44it/s]
2|1745|Loss: 0.9135190844535828:  27%|██▋       | 1745/6470 [20:03<52:36,  1.50it/s]
2|1746|Loss: 0.9066729545593262:  27%|██▋       | 1745/6470 [20:03<52:36,  1.50it/s]
2|1746|Loss: 0.9066729545593262:  27%|██▋       | 1746/6470 [20:03<53:34,  1.47it/s]
2|1747|Loss: 0.891344428062439:  27%|██▋       | 1746/6470 [20:04<53:34,  1.47it/s] 
2|1747|Loss: 0.891344428062439:  27%|██▋       | 1747/6470 [20:04<54:18,  1.45it/s]
2|1748|Loss: 0.9563544988632202:  27%|██▋       | 1747/6470 [20:05<54:18,  1.45it/s]
2|1748|Loss: 0.9563544988632202:  27%|██▋       | 1748/6470 [20:05<57:27,  1.37it/s]
2|1749|Loss: 0.7282047867774963:  27%|██▋       | 1748/6470 [20:05<57:27,  1.37it/s]
2|1749|Loss: 0.7282047867774963:  27%|██▋       | 1749/6470 [20:06<58:14,  1.35it/s]
2|1750|Loss: 0.6363412141799927:  27%|██▋       | 1749/6470 [20:06<58:14,  1.35it/s]
2|1750|Loss: 0.6363412141799927:  27%|██▋       | 1750/6470 [20:07<59:24,  1.32it/s]
2|1751|Loss: 0.7809215784072876:  27%|██▋       | 1750/6470 [20:07<59:24,  1.32it/s]
2|1751|Loss: 0.7809215784072876:  27%|██▋       | 1751/6470 [20:07<1:00:58,  1.29it/s]
2|1752|Loss: 0.9448081254959106:  27%|██▋       | 1751/6470 [20:08<1:00:58,  1.29it/s]
2|1752|Loss: 0.9448081254959106:  27%|██▋       | 1752/6470 [20:08<1:00:07,  1.31it/s]
2|1753|Loss: 1.008448600769043:  27%|██▋       | 1752/6470 [20:09<1:00:07,  1.31it/s] 
2|1753|Loss: 1.008448600769043:  27%|██▋       | 1753/6470 [20:09<1:01:39,  1.28it/s]
2|1754|Loss: 0.8510517477989197:  27%|██▋       | 1753/6470 [20:09<1:01:39,  1.28it/s]
2|1754|Loss: 0.8510517477989197:  27%|██▋       | 1754/6470 [20:10<1:00:43,  1.29it/s]
2|1755|Loss: 0.9930830597877502:  27%|██▋       | 1754/6470 [20:10<1:00:43,  1.29it/s]
2|1755|Loss: 0.9930830597877502:  27%|██▋       | 1755/6470 [20:10<57:57,  1.36it/s]  
2|1756|Loss: 0.8747317790985107:  27%|██▋       | 1755/6470 [20:11<57:57,  1.36it/s]
2|1756|Loss: 0.8747317790985107:  27%|██▋       | 1756/6470 [20:11<58:35,  1.34it/s]
2|1757|Loss: 0.8749351501464844:  27%|██▋       | 1756/6470 [20:11<58:35,  1.34it/s]
2|1757|Loss: 0.8749351501464844:  27%|██▋       | 1757/6470 [20:12<55:44,  1.41it/s]
2|1758|Loss: 0.7912124395370483:  27%|██▋       | 1757/6470 [20:12<55:44,  1.41it/s]
2|1758|Loss: 0.7912124395370483:  27%|██▋       | 1758/6470 [20:12<51:17,  1.53it/s]
2|1759|Loss: 0.9949818253517151:  27%|██▋       | 1758/6470 [20:13<51:17,  1.53it/s]
2|1759|Loss: 0.9949818253517151:  27%|██▋       | 1759/6470 [20:13<52:15,  1.50it/s]
2|1760|Loss: 0.8432494401931763:  27%|██▋       | 1759/6470 [20:13<52:15,  1.50it/s]
2|1760|Loss: 0.8432494401931763:  27%|██▋       | 1760/6470 [20:14<54:04,  1.45it/s]
2|1761|Loss: 0.8381505012512207:  27%|██▋       | 1760/6470 [20:14<54:04,  1.45it/s]
2|1761|Loss: 0.8381505012512207:  27%|██▋       | 1761/6470 [20:14<54:32,  1.44it/s]
2|1762|Loss: 0.9344097971916199:  27%|██▋       | 1761/6470 [20:15<54:32,  1.44it/s]
2|1762|Loss: 0.9344097971916199:  27%|██▋       | 1762/6470 [20:15<51:19,  1.53it/s]
2|1763|Loss: 0.7291768193244934:  27%|██▋       | 1762/6470 [20:15<51:19,  1.53it/s]
2|1763|Loss: 0.7291768193244934:  27%|██▋       | 1763/6470 [20:16<53:56,  1.45it/s]
2|1764|Loss: 0.9466450214385986:  27%|██▋       | 1763/6470 [20:16<53:56,  1.45it/s]
2|1764|Loss: 0.9466450214385986:  27%|██▋       | 1764/6470 [20:17<56:46,  1.38it/s]
2|1765|Loss: 0.8385633826255798:  27%|██▋       | 1764/6470 [20:17<56:46,  1.38it/s]
2|1765|Loss: 0.8385633826255798:  27%|██▋       | 1765/6470 [20:17<57:44,  1.36it/s]
2|1766|Loss: 0.881453812122345:  27%|██▋       | 1765/6470 [20:18<57:44,  1.36it/s] 
2|1766|Loss: 0.881453812122345:  27%|██▋       | 1766/6470 [20:18<55:41,  1.41it/s]
2|1767|Loss: 0.8089724779129028:  27%|██▋       | 1766/6470 [20:18<55:41,  1.41it/s]
2|1767|Loss: 0.8089724779129028:  27%|██▋       | 1767/6470 [20:19<55:59,  1.40it/s]
2|1768|Loss: 0.9044157862663269:  27%|██▋       | 1767/6470 [20:19<55:59,  1.40it/s]
2|1768|Loss: 0.9044157862663269:  27%|██▋       | 1768/6470 [20:19<53:34,  1.46it/s]
2|1769|Loss: 0.9707946181297302:  27%|██▋       | 1768/6470 [20:20<53:34,  1.46it/s]
2|1769|Loss: 0.9707946181297302:  27%|██▋       | 1769/6470 [20:20<51:06,  1.53it/s]
2|1770|Loss: 0.756453812122345:  27%|██▋       | 1769/6470 [20:20<51:06,  1.53it/s] 
2|1770|Loss: 0.756453812122345:  27%|██▋       | 1770/6470 [20:21<58:20,  1.34it/s]
2|1771|Loss: 0.9425615668296814:  27%|██▋       | 1770/6470 [20:21<58:20,  1.34it/s]
2|1771|Loss: 0.9425615668296814:  27%|██▋       | 1771/6470 [20:21<55:29,  1.41it/s]
2|1772|Loss: 0.9811729192733765:  27%|██▋       | 1771/6470 [20:22<55:29,  1.41it/s]
2|1772|Loss: 0.9811729192733765:  27%|██▋       | 1772/6470 [20:22<52:54,  1.48it/s]
2|1773|Loss: 0.7429785132408142:  27%|██▋       | 1772/6470 [20:22<52:54,  1.48it/s]
2|1773|Loss: 0.7429785132408142:  27%|██▋       | 1773/6470 [20:23<49:38,  1.58it/s]
2|1774|Loss: 0.8525055050849915:  27%|██▋       | 1773/6470 [20:23<49:38,  1.58it/s]
2|1774|Loss: 0.8525055050849915:  27%|██▋       | 1774/6470 [20:23<49:11,  1.59it/s]
2|1775|Loss: 0.8538903594017029:  27%|██▋       | 1774/6470 [20:24<49:11,  1.59it/s]
2|1775|Loss: 0.8538903594017029:  27%|██▋       | 1775/6470 [20:24<52:35,  1.49it/s]
2|1776|Loss: 0.7825193405151367:  27%|██▋       | 1775/6470 [20:24<52:35,  1.49it/s]
2|1776|Loss: 0.7825193405151367:  27%|██▋       | 1776/6470 [20:25<49:46,  1.57it/s]
2|1777|Loss: 0.8470340371131897:  27%|██▋       | 1776/6470 [20:25<49:46,  1.57it/s]
2|1777|Loss: 0.8470340371131897:  27%|██▋       | 1777/6470 [20:25<51:59,  1.50it/s]
2|1778|Loss: 0.9877790212631226:  27%|██▋       | 1777/6470 [20:26<51:59,  1.50it/s]
2|1778|Loss: 0.9877790212631226:  27%|██▋       | 1778/6470 [20:26<53:57,  1.45it/s]
2|1779|Loss: 0.9313327670097351:  27%|██▋       | 1778/6470 [20:26<53:57,  1.45it/s]
2|1779|Loss: 0.9313327670097351:  27%|██▋       | 1779/6470 [20:27<52:20,  1.49it/s]
2|1780|Loss: 0.9056620597839355:  27%|██▋       | 1779/6470 [20:27<52:20,  1.49it/s]
2|1780|Loss: 0.9056620597839355:  28%|██▊       | 1780/6470 [20:27<53:03,  1.47it/s]
2|1781|Loss: 0.8346461057662964:  28%|██▊       | 1780/6470 [20:28<53:03,  1.47it/s]
2|1781|Loss: 0.8346461057662964:  28%|██▊       | 1781/6470 [20:28<55:22,  1.41it/s]
2|1782|Loss: 0.9040156602859497:  28%|██▊       | 1781/6470 [20:28<55:22,  1.41it/s]
2|1782|Loss: 0.9040156602859497:  28%|██▊       | 1782/6470 [20:29<54:31,  1.43it/s]
2|1783|Loss: 0.8762888312339783:  28%|██▊       | 1782/6470 [20:29<54:31,  1.43it/s]
2|1783|Loss: 0.8762888312339783:  28%|██▊       | 1783/6470 [20:29<53:02,  1.47it/s]
2|1784|Loss: 1.0132641792297363:  28%|██▊       | 1783/6470 [20:30<53:02,  1.47it/s]
2|1784|Loss: 1.0132641792297363:  28%|██▊       | 1784/6470 [20:30<51:50,  1.51it/s]
2|1785|Loss: 0.8229597806930542:  28%|██▊       | 1784/6470 [20:30<51:50,  1.51it/s]
2|1785|Loss: 0.8229597806930542:  28%|██▊       | 1785/6470 [20:31<51:19,  1.52it/s]
2|1786|Loss: 0.8268973231315613:  28%|██▊       | 1785/6470 [20:31<51:19,  1.52it/s]
2|1786|Loss: 0.8268973231315613:  28%|██▊       | 1786/6470 [20:31<49:54,  1.56it/s]
2|1787|Loss: 0.9438472986221313:  28%|██▊       | 1786/6470 [20:32<49:54,  1.56it/s]
2|1787|Loss: 0.9438472986221313:  28%|██▊       | 1787/6470 [20:32<47:31,  1.64it/s]
2|1788|Loss: 0.7819609642028809:  28%|██▊       | 1787/6470 [20:32<47:31,  1.64it/s]
2|1788|Loss: 0.7819609642028809:  28%|██▊       | 1788/6470 [20:32<42:35,  1.83it/s]
2|1789|Loss: 0.9479999542236328:  28%|██▊       | 1788/6470 [20:33<42:35,  1.83it/s]
2|1789|Loss: 0.9479999542236328:  28%|██▊       | 1789/6470 [20:33<44:33,  1.75it/s]
2|1790|Loss: 1.0195103883743286:  28%|██▊       | 1789/6470 [20:33<44:33,  1.75it/s]
2|1790|Loss: 1.0195103883743286:  28%|██▊       | 1790/6470 [20:34<49:10,  1.59it/s]
2|1791|Loss: 0.9121794104576111:  28%|██▊       | 1790/6470 [20:34<49:10,  1.59it/s]
2|1791|Loss: 0.9121794104576111:  28%|██▊       | 1791/6470 [20:34<49:50,  1.56it/s]
2|1792|Loss: 0.7770942449569702:  28%|██▊       | 1791/6470 [20:35<49:50,  1.56it/s]
2|1792|Loss: 0.7770942449569702:  28%|██▊       | 1792/6470 [20:35<51:24,  1.52it/s]
2|1793|Loss: 0.8166788220405579:  28%|██▊       | 1792/6470 [20:35<51:24,  1.52it/s]
2|1793|Loss: 0.8166788220405579:  28%|██▊       | 1793/6470 [20:36<53:03,  1.47it/s]
2|1794|Loss: 0.8530881404876709:  28%|██▊       | 1793/6470 [20:36<53:03,  1.47it/s]
2|1794|Loss: 0.8530881404876709:  28%|██▊       | 1794/6470 [20:36<52:34,  1.48it/s]
2|1795|Loss: 0.8254829049110413:  28%|██▊       | 1794/6470 [20:37<52:34,  1.48it/s]
2|1795|Loss: 0.8254829049110413:  28%|██▊       | 1795/6470 [20:37<51:44,  1.51it/s]
2|1796|Loss: 0.7955109477043152:  28%|██▊       | 1795/6470 [20:37<51:44,  1.51it/s]
2|1796|Loss: 0.7955109477043152:  28%|██▊       | 1796/6470 [20:38<50:47,  1.53it/s]
2|1797|Loss: 0.7569353580474854:  28%|██▊       | 1796/6470 [20:38<50:47,  1.53it/s]
2|1797|Loss: 0.7569353580474854:  28%|██▊       | 1797/6470 [20:38<51:37,  1.51it/s]
2|1798|Loss: 0.8762996792793274:  28%|██▊       | 1797/6470 [20:39<51:37,  1.51it/s]
2|1798|Loss: 0.8762996792793274:  28%|██▊       | 1798/6470 [20:39<55:20,  1.41it/s]
2|1799|Loss: 1.0360655784606934:  28%|██▊       | 1798/6470 [20:40<55:20,  1.41it/s]
2|1799|Loss: 1.0360655784606934:  28%|██▊       | 1799/6470 [20:40<53:51,  1.45it/s]
2|1800|Loss: 0.9231054782867432:  28%|██▊       | 1799/6470 [20:40<53:51,  1.45it/s]
2|1800|Loss: 0.9231054782867432:  28%|██▊       | 1800/6470 [20:40<52:18,  1.49it/s]
2|1801|Loss: 0.8551185131072998:  28%|██▊       | 1800/6470 [20:41<52:18,  1.49it/s]
2|1801|Loss: 0.8551185131072998:  28%|██▊       | 1801/6470 [20:41<53:33,  1.45it/s]
2|1802|Loss: 0.8588130474090576:  28%|██▊       | 1801/6470 [20:41<53:33,  1.45it/s]
2|1802|Loss: 0.8588130474090576:  28%|██▊       | 1802/6470 [20:42<50:53,  1.53it/s]
2|1803|Loss: 0.7861197590827942:  28%|██▊       | 1802/6470 [20:42<50:53,  1.53it/s]
2|1803|Loss: 0.7861197590827942:  28%|██▊       | 1803/6470 [20:42<49:56,  1.56it/s]
2|1804|Loss: 0.6967955827713013:  28%|██▊       | 1803/6470 [20:43<49:56,  1.56it/s]
2|1804|Loss: 0.6967955827713013:  28%|██▊       | 1804/6470 [20:43<51:48,  1.50it/s]
2|1805|Loss: 0.9426461458206177:  28%|██▊       | 1804/6470 [20:43<51:48,  1.50it/s]
2|1805|Loss: 0.9426461458206177:  28%|██▊       | 1805/6470 [20:44<53:30,  1.45it/s]
2|1806|Loss: 0.8291847705841064:  28%|██▊       | 1805/6470 [20:44<53:30,  1.45it/s]
2|1806|Loss: 0.8291847705841064:  28%|██▊       | 1806/6470 [20:44<52:40,  1.48it/s]
2|1807|Loss: 0.8205827474594116:  28%|██▊       | 1806/6470 [20:45<52:40,  1.48it/s]
2|1807|Loss: 0.8205827474594116:  28%|██▊       | 1807/6470 [20:45<56:05,  1.39it/s]
2|1808|Loss: 0.9385168552398682:  28%|██▊       | 1807/6470 [20:46<56:05,  1.39it/s]
2|1808|Loss: 0.9385168552398682:  28%|██▊       | 1808/6470 [20:46<53:50,  1.44it/s]
2|1809|Loss: 0.7226164937019348:  28%|██▊       | 1808/6470 [20:46<53:50,  1.44it/s]
2|1809|Loss: 0.7226164937019348:  28%|██▊       | 1809/6470 [20:47<52:14,  1.49it/s]
2|1810|Loss: 0.8797105550765991:  28%|██▊       | 1809/6470 [20:47<52:14,  1.49it/s]
2|1810|Loss: 0.8797105550765991:  28%|██▊       | 1810/6470 [20:47<55:11,  1.41it/s]
2|1811|Loss: 0.8604561686515808:  28%|██▊       | 1810/6470 [20:48<55:11,  1.41it/s]
2|1811|Loss: 0.8604561686515808:  28%|██▊       | 1811/6470 [20:48<54:34,  1.42it/s]
2|1812|Loss: 0.8637447953224182:  28%|██▊       | 1811/6470 [20:48<54:34,  1.42it/s]
2|1812|Loss: 0.8637447953224182:  28%|██▊       | 1812/6470 [20:49<53:19,  1.46it/s]
2|1813|Loss: 0.8659687042236328:  28%|██▊       | 1812/6470 [20:49<53:19,  1.46it/s]
2|1813|Loss: 0.8659687042236328:  28%|██▊       | 1813/6470 [20:49<50:40,  1.53it/s]
2|1814|Loss: 0.9675818681716919:  28%|██▊       | 1813/6470 [20:50<50:40,  1.53it/s]
2|1814|Loss: 0.9675818681716919:  28%|██▊       | 1814/6470 [20:50<51:35,  1.50it/s]
2|1815|Loss: 1.0111695528030396:  28%|██▊       | 1814/6470 [20:50<51:35,  1.50it/s]
2|1815|Loss: 1.0111695528030396:  28%|██▊       | 1815/6470 [20:51<55:27,  1.40it/s]
2|1816|Loss: 0.851517915725708:  28%|██▊       | 1815/6470 [20:51<55:27,  1.40it/s] 
2|1816|Loss: 0.851517915725708:  28%|██▊       | 1816/6470 [20:51<52:54,  1.47it/s]
2|1817|Loss: 0.7021730542182922:  28%|██▊       | 1816/6470 [20:52<52:54,  1.47it/s]
2|1817|Loss: 0.7021730542182922:  28%|██▊       | 1817/6470 [20:52<52:29,  1.48it/s]
2|1818|Loss: 0.8884863257408142:  28%|██▊       | 1817/6470 [20:52<52:29,  1.48it/s]
2|1818|Loss: 0.8884863257408142:  28%|██▊       | 1818/6470 [20:53<51:57,  1.49it/s]
2|1819|Loss: 0.9438068270683289:  28%|██▊       | 1818/6470 [20:53<51:57,  1.49it/s]
2|1819|Loss: 0.9438068270683289:  28%|██▊       | 1819/6470 [20:53<54:37,  1.42it/s]
2|1820|Loss: 0.887588381767273:  28%|██▊       | 1819/6470 [20:54<54:37,  1.42it/s] 
2|1820|Loss: 0.887588381767273:  28%|██▊       | 1820/6470 [20:54<55:11,  1.40it/s]
2|1821|Loss: 0.9010012149810791:  28%|██▊       | 1820/6470 [20:55<55:11,  1.40it/s]
2|1821|Loss: 0.9010012149810791:  28%|██▊       | 1821/6470 [20:55<57:20,  1.35it/s]
2|1822|Loss: 0.7737450003623962:  28%|██▊       | 1821/6470 [20:55<57:20,  1.35it/s]
2|1822|Loss: 0.7737450003623962:  28%|██▊       | 1822/6470 [20:56<57:35,  1.35it/s]
2|1823|Loss: 0.8594321012496948:  28%|██▊       | 1822/6470 [20:56<57:35,  1.35it/s]
2|1823|Loss: 0.8594321012496948:  28%|██▊       | 1823/6470 [20:56<55:40,  1.39it/s]
2|1824|Loss: 0.8980018496513367:  28%|██▊       | 1823/6470 [20:57<55:40,  1.39it/s]
2|1824|Loss: 0.8980018496513367:  28%|██▊       | 1824/6470 [20:57<53:27,  1.45it/s]
2|1825|Loss: 0.9541550874710083:  28%|██▊       | 1824/6470 [20:57<53:27,  1.45it/s]
2|1825|Loss: 0.9541550874710083:  28%|██▊       | 1825/6470 [20:58<55:34,  1.39it/s]
2|1826|Loss: 0.8419989347457886:  28%|██▊       | 1825/6470 [20:58<55:34,  1.39it/s]
2|1826|Loss: 0.8419989347457886:  28%|██▊       | 1826/6470 [20:59<58:08,  1.33it/s]
2|1827|Loss: 0.9344026446342468:  28%|██▊       | 1826/6470 [20:59<58:08,  1.33it/s]
2|1827|Loss: 0.9344026446342468:  28%|██▊       | 1827/6470 [20:59<56:05,  1.38it/s]
2|1828|Loss: 0.9226018190383911:  28%|██▊       | 1827/6470 [21:00<56:05,  1.38it/s]
2|1828|Loss: 0.9226018190383911:  28%|██▊       | 1828/6470 [21:00<55:54,  1.38it/s]
2|1829|Loss: 0.8667179942131042:  28%|██▊       | 1828/6470 [21:00<55:54,  1.38it/s]
2|1829|Loss: 0.8667179942131042:  28%|██▊       | 1829/6470 [21:01<57:16,  1.35it/s]
2|1830|Loss: 0.7822473049163818:  28%|██▊       | 1829/6470 [21:01<57:16,  1.35it/s]
2|1830|Loss: 0.7822473049163818:  28%|██▊       | 1830/6470 [21:02<57:25,  1.35it/s]
2|1831|Loss: 0.8434872627258301:  28%|██▊       | 1830/6470 [21:02<57:25,  1.35it/s]
2|1831|Loss: 0.8434872627258301:  28%|██▊       | 1831/6470 [21:02<55:33,  1.39it/s]
2|1832|Loss: 0.754037618637085:  28%|██▊       | 1831/6470 [21:03<55:33,  1.39it/s] 
2|1832|Loss: 0.754037618637085:  28%|██▊       | 1832/6470 [21:03<53:45,  1.44it/s]
2|1833|Loss: 0.8368223309516907:  28%|██▊       | 1832/6470 [21:03<53:45,  1.44it/s]
2|1833|Loss: 0.8368223309516907:  28%|██▊       | 1833/6470 [21:03<52:29,  1.47it/s]
2|1834|Loss: 0.8059543967247009:  28%|██▊       | 1833/6470 [21:04<52:29,  1.47it/s]
2|1834|Loss: 0.8059543967247009:  28%|██▊       | 1834/6470 [21:04<53:27,  1.45it/s]
2|1835|Loss: 0.9651570916175842:  28%|██▊       | 1834/6470 [21:05<53:27,  1.45it/s]
2|1835|Loss: 0.9651570916175842:  28%|██▊       | 1835/6470 [21:05<52:13,  1.48it/s]
2|1836|Loss: 0.8584720492362976:  28%|██▊       | 1835/6470 [21:05<52:13,  1.48it/s]
2|1836|Loss: 0.8584720492362976:  28%|██▊       | 1836/6470 [21:06<52:50,  1.46it/s]
2|1837|Loss: 0.5029646754264832:  28%|██▊       | 1836/6470 [21:06<52:50,  1.46it/s]
2|1837|Loss: 0.5029646754264832:  28%|██▊       | 1837/6470 [21:07<59:48,  1.29it/s]
2|1838|Loss: 0.8891660571098328:  28%|██▊       | 1837/6470 [21:07<59:48,  1.29it/s]
2|1838|Loss: 0.8891660571098328:  28%|██▊       | 1838/6470 [21:07<58:11,  1.33it/s]
2|1839|Loss: 0.8484104871749878:  28%|██▊       | 1838/6470 [21:08<58:11,  1.33it/s]
2|1839|Loss: 0.8484104871749878:  28%|██▊       | 1839/6470 [21:08<54:11,  1.42it/s]
2|1840|Loss: 0.8958527445793152:  28%|██▊       | 1839/6470 [21:08<54:11,  1.42it/s]
2|1840|Loss: 0.8958527445793152:  28%|██▊       | 1840/6470 [21:08<52:06,  1.48it/s]
2|1841|Loss: 0.8635444045066833:  28%|██▊       | 1840/6470 [21:09<52:06,  1.48it/s]
2|1841|Loss: 0.8635444045066833:  28%|██▊       | 1841/6470 [21:09<49:08,  1.57it/s]
2|1842|Loss: 0.9476193785667419:  28%|██▊       | 1841/6470 [21:09<49:08,  1.57it/s]
2|1842|Loss: 0.9476193785667419:  28%|██▊       | 1842/6470 [21:10<52:09,  1.48it/s]
2|1843|Loss: 0.9652429819107056:  28%|██▊       | 1842/6470 [21:10<52:09,  1.48it/s]
2|1843|Loss: 0.9652429819107056:  28%|██▊       | 1843/6470 [21:11<55:17,  1.39it/s]
2|1844|Loss: 0.7098122835159302:  28%|██▊       | 1843/6470 [21:11<55:17,  1.39it/s]
2|1844|Loss: 0.7098122835159302:  29%|██▊       | 1844/6470 [21:11<56:07,  1.37it/s]
2|1845|Loss: 0.8487805128097534:  29%|██▊       | 1844/6470 [21:12<56:07,  1.37it/s]
2|1845|Loss: 0.8487805128097534:  29%|██▊       | 1845/6470 [21:12<50:55,  1.51it/s]
2|1846|Loss: 0.893984317779541:  29%|██▊       | 1845/6470 [21:12<50:55,  1.51it/s] 
2|1846|Loss: 0.893984317779541:  29%|██▊       | 1846/6470 [21:12<50:31,  1.53it/s]
2|1847|Loss: 0.8325158953666687:  29%|██▊       | 1846/6470 [21:13<50:31,  1.53it/s]
2|1847|Loss: 0.8325158953666687:  29%|██▊       | 1847/6470 [21:13<52:13,  1.48it/s]
2|1848|Loss: 0.9381506443023682:  29%|██▊       | 1847/6470 [21:14<52:13,  1.48it/s]
2|1848|Loss: 0.9381506443023682:  29%|██▊       | 1848/6470 [21:14<50:43,  1.52it/s]
2|1849|Loss: 0.9111548066139221:  29%|██▊       | 1848/6470 [21:14<50:43,  1.52it/s]
2|1849|Loss: 0.9111548066139221:  29%|██▊       | 1849/6470 [21:14<47:43,  1.61it/s]
2|1850|Loss: 0.8318917155265808:  29%|██▊       | 1849/6470 [21:15<47:43,  1.61it/s]
2|1850|Loss: 0.8318917155265808:  29%|██▊       | 1850/6470 [21:15<49:22,  1.56it/s]
2|1851|Loss: 0.8594924807548523:  29%|██▊       | 1850/6470 [21:15<49:22,  1.56it/s]
2|1851|Loss: 0.8594924807548523:  29%|██▊       | 1851/6470 [21:16<48:42,  1.58it/s]
2|1852|Loss: 1.0113788843154907:  29%|██▊       | 1851/6470 [21:16<48:42,  1.58it/s]
2|1852|Loss: 1.0113788843154907:  29%|██▊       | 1852/6470 [21:16<51:43,  1.49it/s]
2|1853|Loss: 0.7617541551589966:  29%|██▊       | 1852/6470 [21:17<51:43,  1.49it/s]
2|1853|Loss: 0.7617541551589966:  29%|██▊       | 1853/6470 [21:17<52:48,  1.46it/s]
2|1854|Loss: 0.8508363366127014:  29%|██▊       | 1853/6470 [21:18<52:48,  1.46it/s]
2|1854|Loss: 0.8508363366127014:  29%|██▊       | 1854/6470 [21:18<53:51,  1.43it/s]
2|1855|Loss: 0.6910795569419861:  29%|██▊       | 1854/6470 [21:18<53:51,  1.43it/s]
2|1855|Loss: 0.6910795569419861:  29%|██▊       | 1855/6470 [21:19<57:01,  1.35it/s]
2|1856|Loss: 0.8276156783103943:  29%|██▊       | 1855/6470 [21:19<57:01,  1.35it/s]
2|1856|Loss: 0.8276156783103943:  29%|██▊       | 1856/6470 [21:19<54:06,  1.42it/s]
2|1857|Loss: 0.7731417417526245:  29%|██▊       | 1856/6470 [21:20<54:06,  1.42it/s]
2|1857|Loss: 0.7731417417526245:  29%|██▊       | 1857/6470 [21:20<52:07,  1.47it/s]
2|1858|Loss: 0.7966911792755127:  29%|██▊       | 1857/6470 [21:20<52:07,  1.47it/s]
2|1858|Loss: 0.7966911792755127:  29%|██▊       | 1858/6470 [21:21<53:04,  1.45it/s]
2|1859|Loss: 0.9633942246437073:  29%|██▊       | 1858/6470 [21:21<53:04,  1.45it/s]
2|1859|Loss: 0.9633942246437073:  29%|██▊       | 1859/6470 [21:21<55:33,  1.38it/s]
2|1860|Loss: 0.7836207747459412:  29%|██▊       | 1859/6470 [21:22<55:33,  1.38it/s]
2|1860|Loss: 0.7836207747459412:  29%|██▊       | 1860/6470 [21:22<52:07,  1.47it/s]
2|1861|Loss: 0.891451358795166:  29%|██▊       | 1860/6470 [21:22<52:07,  1.47it/s] 
2|1861|Loss: 0.891451358795166:  29%|██▉       | 1861/6470 [21:23<50:39,  1.52it/s]
2|1862|Loss: 0.7907273173332214:  29%|██▉       | 1861/6470 [21:23<50:39,  1.52it/s]
2|1862|Loss: 0.7907273173332214:  29%|██▉       | 1862/6470 [21:23<50:16,  1.53it/s]
2|1863|Loss: 0.9555721282958984:  29%|██▉       | 1862/6470 [21:24<50:16,  1.53it/s]
2|1863|Loss: 0.9555721282958984:  29%|██▉       | 1863/6470 [21:24<51:52,  1.48it/s]
2|1864|Loss: 0.9017617702484131:  29%|██▉       | 1863/6470 [21:24<51:52,  1.48it/s]
2|1864|Loss: 0.9017617702484131:  29%|██▉       | 1864/6470 [21:25<50:40,  1.51it/s]
2|1865|Loss: 0.8837250471115112:  29%|██▉       | 1864/6470 [21:25<50:40,  1.51it/s]
2|1865|Loss: 0.8837250471115112:  29%|██▉       | 1865/6470 [21:25<46:29,  1.65it/s]
2|1866|Loss: 0.9693586230278015:  29%|██▉       | 1865/6470 [21:25<46:29,  1.65it/s]
2|1866|Loss: 0.9693586230278015:  29%|██▉       | 1866/6470 [21:26<50:39,  1.51it/s]
2|1867|Loss: 0.899107813835144:  29%|██▉       | 1866/6470 [21:26<50:39,  1.51it/s] 
2|1867|Loss: 0.899107813835144:  29%|██▉       | 1867/6470 [21:26<49:05,  1.56it/s]
2|1868|Loss: 0.7024344801902771:  29%|██▉       | 1867/6470 [21:27<49:05,  1.56it/s]
2|1868|Loss: 0.7024344801902771:  29%|██▉       | 1868/6470 [21:27<45:42,  1.68it/s]
2|1869|Loss: 0.7899857759475708:  29%|██▉       | 1868/6470 [21:27<45:42,  1.68it/s]
2|1869|Loss: 0.7899857759475708:  29%|██▉       | 1869/6470 [21:28<43:52,  1.75it/s]
2|1870|Loss: 1.051552414894104:  29%|██▉       | 1869/6470 [21:28<43:52,  1.75it/s] 
2|1870|Loss: 1.051552414894104:  29%|██▉       | 1870/6470 [21:28<47:02,  1.63it/s]
2|1871|Loss: 0.8108332753181458:  29%|██▉       | 1870/6470 [21:29<47:02,  1.63it/s]
2|1871|Loss: 0.8108332753181458:  29%|██▉       | 1871/6470 [21:29<51:09,  1.50it/s]
2|1872|Loss: 0.8044283390045166:  29%|██▉       | 1871/6470 [21:29<51:09,  1.50it/s]
2|1872|Loss: 0.8044283390045166:  29%|██▉       | 1872/6470 [21:30<53:32,  1.43it/s]
2|1873|Loss: 1.0064454078674316:  29%|██▉       | 1872/6470 [21:30<53:32,  1.43it/s]
2|1873|Loss: 1.0064454078674316:  29%|██▉       | 1873/6470 [21:31<54:53,  1.40it/s]
2|1874|Loss: 0.970075786113739:  29%|██▉       | 1873/6470 [21:31<54:53,  1.40it/s] 
2|1874|Loss: 0.970075786113739:  29%|██▉       | 1874/6470 [21:31<53:28,  1.43it/s]
2|1875|Loss: 0.8052495718002319:  29%|██▉       | 1874/6470 [21:32<53:28,  1.43it/s]
2|1875|Loss: 0.8052495718002319:  29%|██▉       | 1875/6470 [21:32<51:44,  1.48it/s]
2|1876|Loss: 0.966319739818573:  29%|██▉       | 1875/6470 [21:32<51:44,  1.48it/s] 
2|1876|Loss: 0.966319739818573:  29%|██▉       | 1876/6470 [21:32<50:59,  1.50it/s]
2|1877|Loss: 0.7337795495986938:  29%|██▉       | 1876/6470 [21:33<50:59,  1.50it/s]
2|1877|Loss: 0.7337795495986938:  29%|██▉       | 1877/6470 [21:33<53:55,  1.42it/s]
2|1878|Loss: 0.9992868900299072:  29%|██▉       | 1877/6470 [21:34<53:55,  1.42it/s]
2|1878|Loss: 0.9992868900299072:  29%|██▉       | 1878/6470 [21:34<52:11,  1.47it/s]
2|1879|Loss: 0.8001073002815247:  29%|██▉       | 1878/6470 [21:34<52:11,  1.47it/s]
2|1879|Loss: 0.8001073002815247:  29%|██▉       | 1879/6470 [21:34<50:05,  1.53it/s]
2|1880|Loss: 0.8981536626815796:  29%|██▉       | 1879/6470 [21:35<50:05,  1.53it/s]
2|1880|Loss: 0.8981536626815796:  29%|██▉       | 1880/6470 [21:35<49:07,  1.56it/s]
2|1881|Loss: 0.9997143149375916:  29%|██▉       | 1880/6470 [21:35<49:07,  1.56it/s]
2|1881|Loss: 0.9997143149375916:  29%|██▉       | 1881/6470 [21:36<50:58,  1.50it/s]
2|1882|Loss: 0.8934253454208374:  29%|██▉       | 1881/6470 [21:36<50:58,  1.50it/s]
2|1882|Loss: 0.8934253454208374:  29%|██▉       | 1882/6470 [21:36<48:51,  1.57it/s]
2|1883|Loss: 0.8310346007347107:  29%|██▉       | 1882/6470 [21:37<48:51,  1.57it/s]
2|1883|Loss: 0.8310346007347107:  29%|██▉       | 1883/6470 [21:37<47:03,  1.62it/s]
2|1884|Loss: 0.756300687789917:  29%|██▉       | 1883/6470 [21:37<47:03,  1.62it/s] 
2|1884|Loss: 0.756300687789917:  29%|██▉       | 1884/6470 [21:37<45:20,  1.69it/s]
2|1885|Loss: 0.7656065821647644:  29%|██▉       | 1884/6470 [21:38<45:20,  1.69it/s]
2|1885|Loss: 0.7656065821647644:  29%|██▉       | 1885/6470 [21:38<45:28,  1.68it/s]
2|1886|Loss: 0.8995066285133362:  29%|██▉       | 1885/6470 [21:38<45:28,  1.68it/s]
2|1886|Loss: 0.8995066285133362:  29%|██▉       | 1886/6470 [21:39<48:10,  1.59it/s]
2|1887|Loss: 0.5968562364578247:  29%|██▉       | 1886/6470 [21:39<48:10,  1.59it/s]
2|1887|Loss: 0.5968562364578247:  29%|██▉       | 1887/6470 [21:40<54:36,  1.40it/s]
2|1888|Loss: 0.910923182964325:  29%|██▉       | 1887/6470 [21:40<54:36,  1.40it/s] 
2|1888|Loss: 0.910923182964325:  29%|██▉       | 1888/6470 [21:40<55:46,  1.37it/s]
2|1889|Loss: 0.7775114178657532:  29%|██▉       | 1888/6470 [21:41<55:46,  1.37it/s]
2|1889|Loss: 0.7775114178657532:  29%|██▉       | 1889/6470 [21:41<55:37,  1.37it/s]
2|1890|Loss: 0.9016171097755432:  29%|██▉       | 1889/6470 [21:42<55:37,  1.37it/s]
2|1890|Loss: 0.9016171097755432:  29%|██▉       | 1890/6470 [21:42<52:48,  1.45it/s]
2|1891|Loss: 0.8511499166488647:  29%|██▉       | 1890/6470 [21:42<52:48,  1.45it/s]
2|1891|Loss: 0.8511499166488647:  29%|██▉       | 1891/6470 [21:42<52:20,  1.46it/s]
2|1892|Loss: 0.8255074620246887:  29%|██▉       | 1891/6470 [21:43<52:20,  1.46it/s]
2|1892|Loss: 0.8255074620246887:  29%|██▉       | 1892/6470 [21:43<51:28,  1.48it/s]
2|1893|Loss: 0.942916750907898:  29%|██▉       | 1892/6470 [21:43<51:28,  1.48it/s] 
2|1893|Loss: 0.942916750907898:  29%|██▉       | 1893/6470 [21:44<49:03,  1.56it/s]
2|1894|Loss: 0.9266462922096252:  29%|██▉       | 1893/6470 [21:44<49:03,  1.56it/s]
2|1894|Loss: 0.9266462922096252:  29%|██▉       | 1894/6470 [21:44<48:31,  1.57it/s]
2|1895|Loss: 0.8406866788864136:  29%|██▉       | 1894/6470 [21:45<48:31,  1.57it/s]
2|1895|Loss: 0.8406866788864136:  29%|██▉       | 1895/6470 [21:45<51:21,  1.48it/s]
2|1896|Loss: 0.8556533455848694:  29%|██▉       | 1895/6470 [21:45<51:21,  1.48it/s]
2|1896|Loss: 0.8556533455848694:  29%|██▉       | 1896/6470 [21:46<51:21,  1.48it/s]
2|1897|Loss: 0.8164761066436768:  29%|██▉       | 1896/6470 [21:46<51:21,  1.48it/s]
2|1897|Loss: 0.8164761066436768:  29%|██▉       | 1897/6470 [21:46<47:08,  1.62it/s]
2|1898|Loss: 0.9007667303085327:  29%|██▉       | 1897/6470 [21:47<47:08,  1.62it/s]
2|1898|Loss: 0.9007667303085327:  29%|██▉       | 1898/6470 [21:47<47:46,  1.60it/s]
2|1899|Loss: 0.8682305216789246:  29%|██▉       | 1898/6470 [21:47<47:46,  1.60it/s]
2|1899|Loss: 0.8682305216789246:  29%|██▉       | 1899/6470 [21:48<50:26,  1.51it/s]
2|1900|Loss: 0.9346485733985901:  29%|██▉       | 1899/6470 [21:48<50:26,  1.51it/s]
2|1900|Loss: 0.9346485733985901:  29%|██▉       | 1900/6470 [21:48<50:09,  1.52it/s]
2|1901|Loss: 0.7077186703681946:  29%|██▉       | 1900/6470 [21:49<50:09,  1.52it/s]
2|1901|Loss: 0.7077186703681946:  29%|██▉       | 1901/6470 [21:49<51:45,  1.47it/s]
2|1902|Loss: 0.8717659711837769:  29%|██▉       | 1901/6470 [21:49<51:45,  1.47it/s]
2|1902|Loss: 0.8717659711837769:  29%|██▉       | 1902/6470 [21:50<51:31,  1.48it/s]
2|1903|Loss: 0.7457354068756104:  29%|██▉       | 1902/6470 [21:50<51:31,  1.48it/s]
2|1903|Loss: 0.7457354068756104:  29%|██▉       | 1903/6470 [21:50<52:19,  1.45it/s]
2|1904|Loss: 0.847568690776825:  29%|██▉       | 1903/6470 [21:51<52:19,  1.45it/s] 
2|1904|Loss: 0.847568690776825:  29%|██▉       | 1904/6470 [21:51<50:35,  1.50it/s]
2|1905|Loss: 0.7528876662254333:  29%|██▉       | 1904/6470 [21:51<50:35,  1.50it/s]
2|1905|Loss: 0.7528876662254333:  29%|██▉       | 1905/6470 [21:52<51:30,  1.48it/s]
2|1906|Loss: 0.9195588827133179:  29%|██▉       | 1905/6470 [21:52<51:30,  1.48it/s]
2|1906|Loss: 0.9195588827133179:  29%|██▉       | 1906/6470 [21:52<48:48,  1.56it/s]
2|1907|Loss: 0.9340728521347046:  29%|██▉       | 1906/6470 [21:53<48:48,  1.56it/s]
2|1907|Loss: 0.9340728521347046:  29%|██▉       | 1907/6470 [21:53<46:57,  1.62it/s]
2|1908|Loss: 0.7351406216621399:  29%|██▉       | 1907/6470 [21:53<46:57,  1.62it/s]
2|1908|Loss: 0.7351406216621399:  29%|██▉       | 1908/6470 [21:53<45:08,  1.68it/s]
2|1909|Loss: 0.9859983921051025:  29%|██▉       | 1908/6470 [21:54<45:08,  1.68it/s]
2|1909|Loss: 0.9859983921051025:  30%|██▉       | 1909/6470 [21:54<40:42,  1.87it/s]
2|1910|Loss: 0.9754874110221863:  30%|██▉       | 1909/6470 [21:54<40:42,  1.87it/s]
2|1910|Loss: 0.9754874110221863:  30%|██▉       | 1910/6470 [21:54<43:38,  1.74it/s]
2|1911|Loss: 1.0216702222824097:  30%|██▉       | 1910/6470 [21:55<43:38,  1.74it/s]
2|1911|Loss: 1.0216702222824097:  30%|██▉       | 1911/6470 [21:55<47:15,  1.61it/s]
2|1912|Loss: 0.8124127984046936:  30%|██▉       | 1911/6470 [21:56<47:15,  1.61it/s]
2|1912|Loss: 0.8124127984046936:  30%|██▉       | 1912/6470 [21:56<49:48,  1.53it/s]
2|1913|Loss: 0.8118997812271118:  30%|██▉       | 1912/6470 [21:56<49:48,  1.53it/s]
2|1913|Loss: 0.8118997812271118:  30%|██▉       | 1913/6470 [21:57<49:40,  1.53it/s]
2|1914|Loss: 0.8703025579452515:  30%|██▉       | 1913/6470 [21:57<49:40,  1.53it/s]
2|1914|Loss: 0.8703025579452515:  30%|██▉       | 1914/6470 [21:57<49:18,  1.54it/s]
2|1915|Loss: 0.7594925165176392:  30%|██▉       | 1914/6470 [21:58<49:18,  1.54it/s]
2|1915|Loss: 0.7594925165176392:  30%|██▉       | 1915/6470 [21:58<50:43,  1.50it/s]
2|1916|Loss: 0.8466846346855164:  30%|██▉       | 1915/6470 [21:58<50:43,  1.50it/s]
2|1916|Loss: 0.8466846346855164:  30%|██▉       | 1916/6470 [21:59<51:37,  1.47it/s]
2|1917|Loss: 0.8018475770950317:  30%|██▉       | 1916/6470 [21:59<51:37,  1.47it/s]
2|1917|Loss: 0.8018475770950317:  30%|██▉       | 1917/6470 [21:59<54:28,  1.39it/s]
2|1918|Loss: 1.073922872543335:  30%|██▉       | 1917/6470 [22:00<54:28,  1.39it/s] 
2|1918|Loss: 1.073922872543335:  30%|██▉       | 1918/6470 [22:00<52:42,  1.44it/s]
2|1919|Loss: 0.7585129141807556:  30%|██▉       | 1918/6470 [22:00<52:42,  1.44it/s]
2|1919|Loss: 0.7585129141807556:  30%|██▉       | 1919/6470 [22:01<55:00,  1.38it/s]
2|1920|Loss: 1.0584285259246826:  30%|██▉       | 1919/6470 [22:01<55:00,  1.38it/s]
2|1920|Loss: 1.0584285259246826:  30%|██▉       | 1920/6470 [22:02<57:38,  1.32it/s]
2|1921|Loss: 0.7390846014022827:  30%|██▉       | 1920/6470 [22:02<57:38,  1.32it/s]
2|1921|Loss: 0.7390846014022827:  30%|██▉       | 1921/6470 [22:02<55:23,  1.37it/s]
2|1922|Loss: 0.7387044429779053:  30%|██▉       | 1921/6470 [22:03<55:23,  1.37it/s]
2|1922|Loss: 0.7387044429779053:  30%|██▉       | 1922/6470 [22:03<56:38,  1.34it/s]
2|1923|Loss: 1.1869688034057617:  30%|██▉       | 1922/6470 [22:04<56:38,  1.34it/s]
2|1923|Loss: 1.1869688034057617:  30%|██▉       | 1923/6470 [22:04<56:52,  1.33it/s]
2|1924|Loss: 0.9617794156074524:  30%|██▉       | 1923/6470 [22:04<56:52,  1.33it/s]
2|1924|Loss: 0.9617794156074524:  30%|██▉       | 1924/6470 [22:05<56:09,  1.35it/s]
2|1925|Loss: 0.778871476650238:  30%|██▉       | 1924/6470 [22:05<56:09,  1.35it/s] 
2|1925|Loss: 0.778871476650238:  30%|██▉       | 1925/6470 [22:05<58:03,  1.30it/s]
2|1926|Loss: 0.9369223117828369:  30%|██▉       | 1925/6470 [22:06<58:03,  1.30it/s]
2|1926|Loss: 0.9369223117828369:  30%|██▉       | 1926/6470 [22:06<58:07,  1.30it/s]
2|1927|Loss: 0.8322269916534424:  30%|██▉       | 1926/6470 [22:07<58:07,  1.30it/s]
2|1927|Loss: 0.8322269916534424:  30%|██▉       | 1927/6470 [22:07<56:58,  1.33it/s]
2|1928|Loss: 0.8563055396080017:  30%|██▉       | 1927/6470 [22:07<56:58,  1.33it/s]
2|1928|Loss: 0.8563055396080017:  30%|██▉       | 1928/6470 [22:08<57:31,  1.32it/s]
2|1929|Loss: 0.6926750540733337:  30%|██▉       | 1928/6470 [22:08<57:31,  1.32it/s]
2|1929|Loss: 0.6926750540733337:  30%|██▉       | 1929/6470 [22:09<59:01,  1.28it/s]
2|1930|Loss: 1.0255253314971924:  30%|██▉       | 1929/6470 [22:09<59:01,  1.28it/s]
2|1930|Loss: 1.0255253314971924:  30%|██▉       | 1930/6470 [22:09<56:20,  1.34it/s]
2|1931|Loss: 0.8666549921035767:  30%|██▉       | 1930/6470 [22:10<56:20,  1.34it/s]
2|1931|Loss: 0.8666549921035767:  30%|██▉       | 1931/6470 [22:10<55:36,  1.36it/s]
2|1932|Loss: 0.7569565773010254:  30%|██▉       | 1931/6470 [22:10<55:36,  1.36it/s]
2|1932|Loss: 0.7569565773010254:  30%|██▉       | 1932/6470 [22:11<52:42,  1.43it/s]
2|1933|Loss: 0.7501051425933838:  30%|██▉       | 1932/6470 [22:11<52:42,  1.43it/s]
2|1933|Loss: 0.7501051425933838:  30%|██▉       | 1933/6470 [22:11<48:19,  1.56it/s]
2|1934|Loss: 0.8201443552970886:  30%|██▉       | 1933/6470 [22:11<48:19,  1.56it/s]
2|1934|Loss: 0.8201443552970886:  30%|██▉       | 1934/6470 [22:12<51:57,  1.46it/s]
2|1935|Loss: 0.8800214529037476:  30%|██▉       | 1934/6470 [22:12<51:57,  1.46it/s]
2|1935|Loss: 0.8800214529037476:  30%|██▉       | 1935/6470 [22:12<50:54,  1.48it/s]
2|1936|Loss: 0.9198371767997742:  30%|██▉       | 1935/6470 [22:13<50:54,  1.48it/s]
2|1936|Loss: 0.9198371767997742:  30%|██▉       | 1936/6470 [22:13<50:59,  1.48it/s]
2|1937|Loss: 0.7705122232437134:  30%|██▉       | 1936/6470 [22:13<50:59,  1.48it/s]
2|1937|Loss: 0.7705122232437134:  30%|██▉       | 1937/6470 [22:14<47:37,  1.59it/s]
2|1938|Loss: 0.9270176291465759:  30%|██▉       | 1937/6470 [22:14<47:37,  1.59it/s]
2|1938|Loss: 0.9270176291465759:  30%|██▉       | 1938/6470 [22:14<46:01,  1.64it/s]
2|1939|Loss: 0.9183402061462402:  30%|██▉       | 1938/6470 [22:15<46:01,  1.64it/s]
2|1939|Loss: 0.9183402061462402:  30%|██▉       | 1939/6470 [22:15<44:07,  1.71it/s]
2|1940|Loss: 0.9116760492324829:  30%|██▉       | 1939/6470 [22:15<44:07,  1.71it/s]
2|1940|Loss: 0.9116760492324829:  30%|██▉       | 1940/6470 [22:15<46:21,  1.63it/s]
2|1941|Loss: 0.8710392713546753:  30%|██▉       | 1940/6470 [22:16<46:21,  1.63it/s]
2|1941|Loss: 0.8710392713546753:  30%|███       | 1941/6470 [22:16<51:12,  1.47it/s]
2|1942|Loss: 1.0648235082626343:  30%|███       | 1941/6470 [22:17<51:12,  1.47it/s]
2|1942|Loss: 1.0648235082626343:  30%|███       | 1942/6470 [22:17<48:38,  1.55it/s]
2|1943|Loss: 0.907107949256897:  30%|███       | 1942/6470 [22:17<48:38,  1.55it/s] 
2|1943|Loss: 0.907107949256897:  30%|███       | 1943/6470 [22:18<49:24,  1.53it/s]
2|1944|Loss: 0.9101642966270447:  30%|███       | 1943/6470 [22:18<49:24,  1.53it/s]
2|1944|Loss: 0.9101642966270447:  30%|███       | 1944/6470 [22:18<50:38,  1.49it/s]
2|1945|Loss: 0.801054060459137:  30%|███       | 1944/6470 [22:19<50:38,  1.49it/s] 
2|1945|Loss: 0.801054060459137:  30%|███       | 1945/6470 [22:19<49:02,  1.54it/s]
2|1946|Loss: 0.9644708037376404:  30%|███       | 1945/6470 [22:19<49:02,  1.54it/s]
2|1946|Loss: 0.9644708037376404:  30%|███       | 1946/6470 [22:19<47:40,  1.58it/s]
2|1947|Loss: 0.9898570775985718:  30%|███       | 1946/6470 [22:20<47:40,  1.58it/s]
2|1947|Loss: 0.9898570775985718:  30%|███       | 1947/6470 [22:20<50:22,  1.50it/s]
2|1948|Loss: 0.9270473122596741:  30%|███       | 1947/6470 [22:21<50:22,  1.50it/s]
2|1948|Loss: 0.9270473122596741:  30%|███       | 1948/6470 [22:21<52:05,  1.45it/s]
2|1949|Loss: 0.785961389541626:  30%|███       | 1948/6470 [22:21<52:05,  1.45it/s] 
2|1949|Loss: 0.785961389541626:  30%|███       | 1949/6470 [22:22<51:21,  1.47it/s]
2|1950|Loss: 0.9863992929458618:  30%|███       | 1949/6470 [22:22<51:21,  1.47it/s]
2|1950|Loss: 0.9863992929458618:  30%|███       | 1950/6470 [22:22<45:27,  1.66it/s]
2|1951|Loss: 0.7744970321655273:  30%|███       | 1950/6470 [22:22<45:27,  1.66it/s]
2|1951|Loss: 0.7744970321655273:  30%|███       | 1951/6470 [22:23<47:03,  1.60it/s]
2|1952|Loss: 0.9002951979637146:  30%|███       | 1951/6470 [22:23<47:03,  1.60it/s]
2|1952|Loss: 0.9002951979637146:  30%|███       | 1952/6470 [22:23<45:39,  1.65it/s]
2|1953|Loss: 0.8139099478721619:  30%|███       | 1952/6470 [22:24<45:39,  1.65it/s]
2|1953|Loss: 0.8139099478721619:  30%|███       | 1953/6470 [22:24<43:48,  1.72it/s]
2|1954|Loss: 0.8551037907600403:  30%|███       | 1953/6470 [22:24<43:48,  1.72it/s]
2|1954|Loss: 0.8551037907600403:  30%|███       | 1954/6470 [22:24<44:13,  1.70it/s]
2|1955|Loss: 1.0294777154922485:  30%|███       | 1954/6470 [22:25<44:13,  1.70it/s]
2|1955|Loss: 1.0294777154922485:  30%|███       | 1955/6470 [22:25<46:59,  1.60it/s]
2|1956|Loss: 0.9169186949729919:  30%|███       | 1955/6470 [22:25<46:59,  1.60it/s]
2|1956|Loss: 0.9169186949729919:  30%|███       | 1956/6470 [22:26<47:56,  1.57it/s]
2|1957|Loss: 0.8689095377922058:  30%|███       | 1956/6470 [22:26<47:56,  1.57it/s]
2|1957|Loss: 0.8689095377922058:  30%|███       | 1957/6470 [22:26<45:39,  1.65it/s]
2|1958|Loss: 0.8706678748130798:  30%|███       | 1957/6470 [22:27<45:39,  1.65it/s]
2|1958|Loss: 0.8706678748130798:  30%|███       | 1958/6470 [22:27<45:47,  1.64it/s]
2|1959|Loss: 0.9005094170570374:  30%|███       | 1958/6470 [22:27<45:47,  1.64it/s]
2|1959|Loss: 0.9005094170570374:  30%|███       | 1959/6470 [22:27<44:33,  1.69it/s]
2|1960|Loss: 0.8455321788787842:  30%|███       | 1959/6470 [22:28<44:33,  1.69it/s]
2|1960|Loss: 0.8455321788787842:  30%|███       | 1960/6470 [22:28<47:16,  1.59it/s]
2|1961|Loss: 0.7486464977264404:  30%|███       | 1960/6470 [22:29<47:16,  1.59it/s]
2|1961|Loss: 0.7486464977264404:  30%|███       | 1961/6470 [22:29<52:15,  1.44it/s]
2|1962|Loss: 0.8757988214492798:  30%|███       | 1961/6470 [22:29<52:15,  1.44it/s]
2|1962|Loss: 0.8757988214492798:  30%|███       | 1962/6470 [22:30<51:06,  1.47it/s]
2|1963|Loss: 0.7951528429985046:  30%|███       | 1962/6470 [22:30<51:06,  1.47it/s]
2|1963|Loss: 0.7951528429985046:  30%|███       | 1963/6470 [22:30<51:47,  1.45it/s]
2|1964|Loss: 0.9048790335655212:  30%|███       | 1963/6470 [22:31<51:47,  1.45it/s]
2|1964|Loss: 0.9048790335655212:  30%|███       | 1964/6470 [22:31<55:20,  1.36it/s]
2|1965|Loss: 0.9876313209533691:  30%|███       | 1964/6470 [22:32<55:20,  1.36it/s]
2|1965|Loss: 0.9876313209533691:  30%|███       | 1965/6470 [22:32<55:32,  1.35it/s]
2|1966|Loss: 0.9634227156639099:  30%|███       | 1965/6470 [22:32<55:32,  1.35it/s]
2|1966|Loss: 0.9634227156639099:  30%|███       | 1966/6470 [22:33<53:13,  1.41it/s]
2|1967|Loss: 0.8900671005249023:  30%|███       | 1966/6470 [22:33<53:13,  1.41it/s]
2|1967|Loss: 0.8900671005249023:  30%|███       | 1967/6470 [22:33<51:39,  1.45it/s]
2|1968|Loss: 0.8517552018165588:  30%|███       | 1967/6470 [22:34<51:39,  1.45it/s]
2|1968|Loss: 0.8517552018165588:  30%|███       | 1968/6470 [22:34<50:29,  1.49it/s]
2|1969|Loss: 0.8962388634681702:  30%|███       | 1968/6470 [22:34<50:29,  1.49it/s]
2|1969|Loss: 0.8962388634681702:  30%|███       | 1969/6470 [22:35<50:54,  1.47it/s]
2|1970|Loss: 0.9557828903198242:  30%|███       | 1969/6470 [22:35<50:54,  1.47it/s]
2|1970|Loss: 0.9557828903198242:  30%|███       | 1970/6470 [22:35<51:24,  1.46it/s]
2|1971|Loss: 0.9394205212593079:  30%|███       | 1970/6470 [22:36<51:24,  1.46it/s]
2|1971|Loss: 0.9394205212593079:  30%|███       | 1971/6470 [22:36<49:32,  1.51it/s]
2|1972|Loss: 0.8071426153182983:  30%|███       | 1971/6470 [22:36<49:32,  1.51it/s]
2|1972|Loss: 0.8071426153182983:  30%|███       | 1972/6470 [22:37<50:42,  1.48it/s]
2|1973|Loss: 0.8033880591392517:  30%|███       | 1972/6470 [22:37<50:42,  1.48it/s]
2|1973|Loss: 0.8033880591392517:  30%|███       | 1973/6470 [22:37<50:33,  1.48it/s]
2|1974|Loss: 0.8686664700508118:  30%|███       | 1973/6470 [22:38<50:33,  1.48it/s]
2|1974|Loss: 0.8686664700508118:  31%|███       | 1974/6470 [22:38<48:42,  1.54it/s]
2|1975|Loss: 0.9277867078781128:  31%|███       | 1974/6470 [22:38<48:42,  1.54it/s]
2|1975|Loss: 0.9277867078781128:  31%|███       | 1975/6470 [22:39<49:16,  1.52it/s]
2|1976|Loss: 0.8217251300811768:  31%|███       | 1975/6470 [22:39<49:16,  1.52it/s]
2|1976|Loss: 0.8217251300811768:  31%|███       | 1976/6470 [22:39<46:44,  1.60it/s]
2|1977|Loss: 0.8544077277183533:  31%|███       | 1976/6470 [22:39<46:44,  1.60it/s]
2|1977|Loss: 0.8544077277183533:  31%|███       | 1977/6470 [22:40<50:18,  1.49it/s]
2|1978|Loss: 0.7686192393302917:  31%|███       | 1977/6470 [22:40<50:18,  1.49it/s]
2|1978|Loss: 0.7686192393302917:  31%|███       | 1978/6470 [22:40<45:47,  1.63it/s]
2|1979|Loss: 0.9069895148277283:  31%|███       | 1978/6470 [22:41<45:47,  1.63it/s]
2|1979|Loss: 0.9069895148277283:  31%|███       | 1979/6470 [22:41<42:54,  1.74it/s]
2|1980|Loss: 0.9696503281593323:  31%|███       | 1979/6470 [22:41<42:54,  1.74it/s]
2|1980|Loss: 0.9696503281593323:  31%|███       | 1980/6470 [22:41<44:49,  1.67it/s]
2|1981|Loss: 0.8515653610229492:  31%|███       | 1980/6470 [22:42<44:49,  1.67it/s]
2|1981|Loss: 0.8515653610229492:  31%|███       | 1981/6470 [22:42<49:25,  1.51it/s]
2|1982|Loss: 0.8645238876342773:  31%|███       | 1981/6470 [22:43<49:25,  1.51it/s]
2|1982|Loss: 0.8645238876342773:  31%|███       | 1982/6470 [22:43<47:48,  1.56it/s]
2|1983|Loss: 0.7460997700691223:  31%|███       | 1982/6470 [22:43<47:48,  1.56it/s]
2|1983|Loss: 0.7460997700691223:  31%|███       | 1983/6470 [22:43<46:52,  1.60it/s]
2|1984|Loss: 0.8612516522407532:  31%|███       | 1983/6470 [22:44<46:52,  1.60it/s]
2|1984|Loss: 0.8612516522407532:  31%|███       | 1984/6470 [22:44<46:47,  1.60it/s]
2|1985|Loss: 0.9089495539665222:  31%|███       | 1984/6470 [22:44<46:47,  1.60it/s]
2|1985|Loss: 0.9089495539665222:  31%|███       | 1985/6470 [22:45<46:57,  1.59it/s]
2|1986|Loss: 0.832065761089325:  31%|███       | 1985/6470 [22:45<46:57,  1.59it/s] 
2|1986|Loss: 0.832065761089325:  31%|███       | 1986/6470 [22:45<43:47,  1.71it/s]
2|1987|Loss: 0.9851133823394775:  31%|███       | 1986/6470 [22:46<43:47,  1.71it/s]
2|1987|Loss: 0.9851133823394775:  31%|███       | 1987/6470 [22:46<47:29,  1.57it/s]
2|1988|Loss: 0.8922402262687683:  31%|███       | 1987/6470 [22:46<47:29,  1.57it/s]
2|1988|Loss: 0.8922402262687683:  31%|███       | 1988/6470 [22:47<49:44,  1.50it/s]
2|1989|Loss: 0.9677700996398926:  31%|███       | 1988/6470 [22:47<49:44,  1.50it/s]
2|1989|Loss: 0.9677700996398926:  31%|███       | 1989/6470 [22:47<52:21,  1.43it/s]
2|1990|Loss: 0.8249215483665466:  31%|███       | 1989/6470 [22:48<52:21,  1.43it/s]
2|1990|Loss: 0.8249215483665466:  31%|███       | 1990/6470 [22:48<48:20,  1.54it/s]
2|1991|Loss: 0.8428688645362854:  31%|███       | 1990/6470 [22:48<48:20,  1.54it/s]
2|1991|Loss: 0.8428688645362854:  31%|███       | 1991/6470 [22:49<48:22,  1.54it/s]
2|1992|Loss: 0.9121236801147461:  31%|███       | 1991/6470 [22:49<48:22,  1.54it/s]
2|1992|Loss: 0.9121236801147461:  31%|███       | 1992/6470 [22:49<49:36,  1.50it/s]
2|1993|Loss: 0.8593904972076416:  31%|███       | 1992/6470 [22:50<49:36,  1.50it/s]
2|1993|Loss: 0.8593904972076416:  31%|███       | 1993/6470 [22:50<53:36,  1.39it/s]
2|1994|Loss: 0.8866384625434875:  31%|███       | 1993/6470 [22:51<53:36,  1.39it/s]
2|1994|Loss: 0.8866384625434875:  31%|███       | 1994/6470 [22:51<54:06,  1.38it/s]
2|1995|Loss: 0.8941091299057007:  31%|███       | 1994/6470 [22:51<54:06,  1.38it/s]
2|1995|Loss: 0.8941091299057007:  31%|███       | 1995/6470 [22:52<52:33,  1.42it/s]
2|1996|Loss: 0.7999306321144104:  31%|███       | 1995/6470 [22:52<52:33,  1.42it/s]
2|1996|Loss: 0.7999306321144104:  31%|███       | 1996/6470 [22:52<55:00,  1.36it/s]
2|1997|Loss: 0.9239707589149475:  31%|███       | 1996/6470 [22:53<55:00,  1.36it/s]
2|1997|Loss: 0.9239707589149475:  31%|███       | 1997/6470 [22:53<56:40,  1.32it/s]
2|1998|Loss: 0.8730161190032959:  31%|███       | 1997/6470 [22:54<56:40,  1.32it/s]
2|1998|Loss: 0.8730161190032959:  31%|███       | 1998/6470 [22:54<58:07,  1.28it/s]
2|1999|Loss: 0.8886910676956177:  31%|███       | 1998/6470 [22:54<58:07,  1.28it/s]
2|1999|Loss: 0.8886910676956177:  31%|███       | 1999/6470 [22:55<55:29,  1.34it/s]
2|2000|Loss: 0.8518002033233643:  31%|███       | 1999/6470 [22:55<55:29,  1.34it/s]
2|2000|Loss: 0.8518002033233643:  31%|███       | 2000/6470 [22:55<52:34,  1.42it/s]
2|2001|Loss: 0.6967374682426453:  31%|███       | 2000/6470 [22:56<52:34,  1.42it/s]
2|2001|Loss: 0.6967374682426453:  31%|███       | 2001/6470 [22:56<55:54,  1.33it/s]
2|2002|Loss: 0.8070642948150635:  31%|███       | 2001/6470 [22:57<55:54,  1.33it/s]
2|2002|Loss: 0.8070642948150635:  31%|███       | 2002/6470 [22:57<54:42,  1.36it/s]
2|2003|Loss: 0.9001073837280273:  31%|███       | 2002/6470 [22:57<54:42,  1.36it/s]
2|2003|Loss: 0.9001073837280273:  31%|███       | 2003/6470 [22:58<54:14,  1.37it/s]
2|2004|Loss: 0.984653651714325:  31%|███       | 2003/6470 [22:58<54:14,  1.37it/s] 
2|2004|Loss: 0.984653651714325:  31%|███       | 2004/6470 [22:58<54:25,  1.37it/s]
2|2005|Loss: 0.9128488302230835:  31%|███       | 2004/6470 [22:59<54:25,  1.37it/s]
2|2005|Loss: 0.9128488302230835:  31%|███       | 2005/6470 [22:59<52:51,  1.41it/s]
2|2006|Loss: 0.7981622219085693:  31%|███       | 2005/6470 [22:59<52:51,  1.41it/s]
2|2006|Loss: 0.7981622219085693:  31%|███       | 2006/6470 [23:00<53:01,  1.40it/s]
2|2007|Loss: 0.8013124465942383:  31%|███       | 2006/6470 [23:00<53:01,  1.40it/s]
2|2007|Loss: 0.8013124465942383:  31%|███       | 2007/6470 [23:00<50:54,  1.46it/s]
2|2008|Loss: 0.7534816265106201:  31%|███       | 2007/6470 [23:01<50:54,  1.46it/s]
2|2008|Loss: 0.7534816265106201:  31%|███       | 2008/6470 [23:01<44:45,  1.66it/s]
2|2009|Loss: 0.9127906560897827:  31%|███       | 2008/6470 [23:01<44:45,  1.66it/s]
2|2009|Loss: 0.9127906560897827:  31%|███       | 2009/6470 [23:01<42:14,  1.76it/s]
2|2010|Loss: 0.7748214602470398:  31%|███       | 2009/6470 [23:02<42:14,  1.76it/s]
2|2010|Loss: 0.7748214602470398:  31%|███       | 2010/6470 [23:02<44:16,  1.68it/s]
2|2011|Loss: 0.8964208960533142:  31%|███       | 2010/6470 [23:02<44:16,  1.68it/s]
2|2011|Loss: 0.8964208960533142:  31%|███       | 2011/6470 [23:02<42:11,  1.76it/s]
2|2012|Loss: 0.7970195412635803:  31%|███       | 2011/6470 [23:03<42:11,  1.76it/s]
2|2012|Loss: 0.7970195412635803:  31%|███       | 2012/6470 [23:03<42:24,  1.75it/s]
2|2013|Loss: 0.8907464146614075:  31%|███       | 2012/6470 [23:03<42:24,  1.75it/s]
2|2013|Loss: 0.8907464146614075:  31%|███       | 2013/6470 [23:04<42:42,  1.74it/s]
2|2014|Loss: 0.8194277286529541:  31%|███       | 2013/6470 [23:04<42:42,  1.74it/s]
2|2014|Loss: 0.8194277286529541:  31%|███       | 2014/6470 [23:04<42:51,  1.73it/s]
2|2015|Loss: 0.7042039036750793:  31%|███       | 2014/6470 [23:05<42:51,  1.73it/s]
2|2015|Loss: 0.7042039036750793:  31%|███       | 2015/6470 [23:05<48:15,  1.54it/s]
2|2016|Loss: 0.8439595103263855:  31%|███       | 2015/6470 [23:05<48:15,  1.54it/s]
2|2016|Loss: 0.8439595103263855:  31%|███       | 2016/6470 [23:06<50:50,  1.46it/s]
2|2017|Loss: 0.929437518119812:  31%|███       | 2016/6470 [23:06<50:50,  1.46it/s] 
2|2017|Loss: 0.929437518119812:  31%|███       | 2017/6470 [23:06<51:55,  1.43it/s]
2|2018|Loss: 0.8971923589706421:  31%|███       | 2017/6470 [23:07<51:55,  1.43it/s]
2|2018|Loss: 0.8971923589706421:  31%|███       | 2018/6470 [23:07<52:06,  1.42it/s]
2|2019|Loss: 0.7354881167411804:  31%|███       | 2018/6470 [23:08<52:06,  1.42it/s]
2|2019|Loss: 0.7354881167411804:  31%|███       | 2019/6470 [23:08<52:35,  1.41it/s]
2|2020|Loss: 0.9341585636138916:  31%|███       | 2019/6470 [23:08<52:35,  1.41it/s]
2|2020|Loss: 0.9341585636138916:  31%|███       | 2020/6470 [23:09<51:05,  1.45it/s]
2|2021|Loss: 0.7912663817405701:  31%|███       | 2020/6470 [23:09<51:05,  1.45it/s]
2|2021|Loss: 0.7912663817405701:  31%|███       | 2021/6470 [23:09<52:33,  1.41it/s]
2|2022|Loss: 0.8711304664611816:  31%|███       | 2021/6470 [23:10<52:33,  1.41it/s]
2|2022|Loss: 0.8711304664611816:  31%|███▏      | 2022/6470 [23:10<53:32,  1.38it/s]
2|2023|Loss: 0.8307167291641235:  31%|███▏      | 2022/6470 [23:10<53:32,  1.38it/s]
2|2023|Loss: 0.8307167291641235:  31%|███▏      | 2023/6470 [23:11<52:13,  1.42it/s]
2|2024|Loss: 0.6798298358917236:  31%|███▏      | 2023/6470 [23:11<52:13,  1.42it/s]
2|2024|Loss: 0.6798298358917236:  31%|███▏      | 2024/6470 [23:11<48:27,  1.53it/s]
2|2025|Loss: 0.8244138956069946:  31%|███▏      | 2024/6470 [23:12<48:27,  1.53it/s]
2|2025|Loss: 0.8244138956069946:  31%|███▏      | 2025/6470 [23:12<48:34,  1.52it/s]
2|2026|Loss: 0.7146074771881104:  31%|███▏      | 2025/6470 [23:12<48:34,  1.52it/s]
2|2026|Loss: 0.7146074771881104:  31%|███▏      | 2026/6470 [23:12<45:43,  1.62it/s]
2|2027|Loss: 0.8433788418769836:  31%|███▏      | 2026/6470 [23:13<45:43,  1.62it/s]
2|2027|Loss: 0.8433788418769836:  31%|███▏      | 2027/6470 [23:13<46:33,  1.59it/s]
2|2028|Loss: 0.7621685266494751:  31%|███▏      | 2027/6470 [23:13<46:33,  1.59it/s]
2|2028|Loss: 0.7621685266494751:  31%|███▏      | 2028/6470 [23:14<50:34,  1.46it/s]
2|2029|Loss: 0.9260885715484619:  31%|███▏      | 2028/6470 [23:14<50:34,  1.46it/s]
2|2029|Loss: 0.9260885715484619:  31%|███▏      | 2029/6470 [23:14<45:35,  1.62it/s]
2|2030|Loss: 0.7033129930496216:  31%|███▏      | 2029/6470 [23:15<45:35,  1.62it/s]
2|2030|Loss: 0.7033129930496216:  31%|███▏      | 2030/6470 [23:15<46:39,  1.59it/s]
2|2031|Loss: 0.6677239537239075:  31%|███▏      | 2030/6470 [23:15<46:39,  1.59it/s]
2|2031|Loss: 0.6677239537239075:  31%|███▏      | 2031/6470 [23:16<46:22,  1.60it/s]
2|2032|Loss: 1.0489654541015625:  31%|███▏      | 2031/6470 [23:16<46:22,  1.60it/s]
2|2032|Loss: 1.0489654541015625:  31%|███▏      | 2032/6470 [23:16<49:06,  1.51it/s]
2|2033|Loss: 0.893835186958313:  31%|███▏      | 2032/6470 [23:17<49:06,  1.51it/s] 
2|2033|Loss: 0.893835186958313:  31%|███▏      | 2033/6470 [23:17<49:19,  1.50it/s]
2|2034|Loss: 0.8280469179153442:  31%|███▏      | 2033/6470 [23:17<49:19,  1.50it/s]
2|2034|Loss: 0.8280469179153442:  31%|███▏      | 2034/6470 [23:18<51:18,  1.44it/s]
2|2035|Loss: 0.7811184525489807:  31%|███▏      | 2034/6470 [23:18<51:18,  1.44it/s]
2|2035|Loss: 0.7811184525489807:  31%|███▏      | 2035/6470 [23:18<46:00,  1.61it/s]
2|2036|Loss: 0.9510993957519531:  31%|███▏      | 2035/6470 [23:19<46:00,  1.61it/s]
2|2036|Loss: 0.9510993957519531:  31%|███▏      | 2036/6470 [23:19<47:58,  1.54it/s]
2|2037|Loss: 0.863274872303009:  31%|███▏      | 2036/6470 [23:19<47:58,  1.54it/s] 
2|2037|Loss: 0.863274872303009:  31%|███▏      | 2037/6470 [23:20<52:29,  1.41it/s]
2|2038|Loss: 0.8398549556732178:  31%|███▏      | 2037/6470 [23:20<52:29,  1.41it/s]
2|2038|Loss: 0.8398549556732178:  31%|███▏      | 2038/6470 [23:21<53:02,  1.39it/s]
2|2039|Loss: 0.7536806464195251:  31%|███▏      | 2038/6470 [23:21<53:02,  1.39it/s]
2|2039|Loss: 0.7536806464195251:  32%|███▏      | 2039/6470 [23:21<54:28,  1.36it/s]
2|2040|Loss: 0.9438837170600891:  32%|███▏      | 2039/6470 [23:22<54:28,  1.36it/s]
2|2040|Loss: 0.9438837170600891:  32%|███▏      | 2040/6470 [23:22<50:55,  1.45it/s]
2|2041|Loss: 0.9203765392303467:  32%|███▏      | 2040/6470 [23:22<50:55,  1.45it/s]
2|2041|Loss: 0.9203765392303467:  32%|███▏      | 2041/6470 [23:23<48:47,  1.51it/s]
2|2042|Loss: 0.8540545701980591:  32%|███▏      | 2041/6470 [23:23<48:47,  1.51it/s]
2|2042|Loss: 0.8540545701980591:  32%|███▏      | 2042/6470 [23:23<51:11,  1.44it/s]
2|2043|Loss: 0.8720713257789612:  32%|███▏      | 2042/6470 [23:24<51:11,  1.44it/s]
2|2043|Loss: 0.8720713257789612:  32%|███▏      | 2043/6470 [23:24<45:54,  1.61it/s]
2|2044|Loss: 0.7848178148269653:  32%|███▏      | 2043/6470 [23:24<45:54,  1.61it/s]
2|2044|Loss: 0.7848178148269653:  32%|███▏      | 2044/6470 [23:24<47:51,  1.54it/s]
2|2045|Loss: 0.9552369713783264:  32%|███▏      | 2044/6470 [23:25<47:51,  1.54it/s]
2|2045|Loss: 0.9552369713783264:  32%|███▏      | 2045/6470 [23:25<49:38,  1.49it/s]
2|2046|Loss: 0.6951366662979126:  32%|███▏      | 2045/6470 [23:26<49:38,  1.49it/s]
2|2046|Loss: 0.6951366662979126:  32%|███▏      | 2046/6470 [23:26<50:54,  1.45it/s]
2|2047|Loss: 0.8545209765434265:  32%|███▏      | 2046/6470 [23:26<50:54,  1.45it/s]
2|2047|Loss: 0.8545209765434265:  32%|███▏      | 2047/6470 [23:27<49:05,  1.50it/s]
2|2048|Loss: 0.8768569231033325:  32%|███▏      | 2047/6470 [23:27<49:05,  1.50it/s]
2|2048|Loss: 0.8768569231033325:  32%|███▏      | 2048/6470 [23:27<49:59,  1.47it/s]
2|2049|Loss: 0.948999285697937:  32%|███▏      | 2048/6470 [23:28<49:59,  1.47it/s] 
2|2049|Loss: 0.948999285697937:  32%|███▏      | 2049/6470 [23:28<47:34,  1.55it/s]
2|2050|Loss: 0.8511942625045776:  32%|███▏      | 2049/6470 [23:28<47:34,  1.55it/s]
2|2050|Loss: 0.8511942625045776:  32%|███▏      | 2050/6470 [23:28<46:33,  1.58it/s]
2|2051|Loss: 0.9046642184257507:  32%|███▏      | 2050/6470 [23:29<46:33,  1.58it/s]
2|2051|Loss: 0.9046642184257507:  32%|███▏      | 2051/6470 [23:29<47:52,  1.54it/s]
2|2052|Loss: 0.881318986415863:  32%|███▏      | 2051/6470 [23:29<47:52,  1.54it/s] 
2|2052|Loss: 0.881318986415863:  32%|███▏      | 2052/6470 [23:30<48:05,  1.53it/s]
2|2053|Loss: 0.8227018117904663:  32%|███▏      | 2052/6470 [23:30<48:05,  1.53it/s]
2|2053|Loss: 0.8227018117904663:  32%|███▏      | 2053/6470 [23:30<45:33,  1.62it/s]
2|2054|Loss: 0.8283966183662415:  32%|███▏      | 2053/6470 [23:31<45:33,  1.62it/s]
2|2054|Loss: 0.8283966183662415:  32%|███▏      | 2054/6470 [23:31<50:23,  1.46it/s]
2|2055|Loss: 0.8078073859214783:  32%|███▏      | 2054/6470 [23:32<50:23,  1.46it/s]
2|2055|Loss: 0.8078073859214783:  32%|███▏      | 2055/6470 [23:32<54:27,  1.35it/s]
2|2056|Loss: 0.9595649242401123:  32%|███▏      | 2055/6470 [23:32<54:27,  1.35it/s]
2|2056|Loss: 0.9595649242401123:  32%|███▏      | 2056/6470 [23:33<51:49,  1.42it/s]
2|2057|Loss: 0.8601094484329224:  32%|███▏      | 2056/6470 [23:33<51:49,  1.42it/s]
2|2057|Loss: 0.8601094484329224:  32%|███▏      | 2057/6470 [23:33<52:01,  1.41it/s]
2|2058|Loss: 0.7177709937095642:  32%|███▏      | 2057/6470 [23:34<52:01,  1.41it/s]
2|2058|Loss: 0.7177709937095642:  32%|███▏      | 2058/6470 [23:34<52:47,  1.39it/s]
2|2059|Loss: 0.8396424055099487:  32%|███▏      | 2058/6470 [23:34<52:47,  1.39it/s]
2|2059|Loss: 0.8396424055099487:  32%|███▏      | 2059/6470 [23:35<50:24,  1.46it/s]
2|2060|Loss: 0.95015949010849:  32%|███▏      | 2059/6470 [23:35<50:24,  1.46it/s]  
2|2060|Loss: 0.95015949010849:  32%|███▏      | 2060/6470 [23:35<50:08,  1.47it/s]
2|2061|Loss: 0.8129370212554932:  32%|███▏      | 2060/6470 [23:36<50:08,  1.47it/s]
2|2061|Loss: 0.8129370212554932:  32%|███▏      | 2061/6470 [23:36<50:31,  1.45it/s]
2|2062|Loss: 0.8252468109130859:  32%|███▏      | 2061/6470 [23:36<50:31,  1.45it/s]
2|2062|Loss: 0.8252468109130859:  32%|███▏      | 2062/6470 [23:37<49:39,  1.48it/s]
2|2063|Loss: 0.8959511518478394:  32%|███▏      | 2062/6470 [23:37<49:39,  1.48it/s]
2|2063|Loss: 0.8959511518478394:  32%|███▏      | 2063/6470 [23:37<51:39,  1.42it/s]
2|2064|Loss: 0.8741432428359985:  32%|███▏      | 2063/6470 [23:38<51:39,  1.42it/s]
2|2064|Loss: 0.8741432428359985:  32%|███▏      | 2064/6470 [23:38<55:22,  1.33it/s]
2|2065|Loss: 0.7577800154685974:  32%|███▏      | 2064/6470 [23:39<55:22,  1.33it/s]
2|2065|Loss: 0.7577800154685974:  32%|███▏      | 2065/6470 [23:39<55:45,  1.32it/s]
2|2066|Loss: 0.886786699295044:  32%|███▏      | 2065/6470 [23:39<55:45,  1.32it/s] 
2|2066|Loss: 0.886786699295044:  32%|███▏      | 2066/6470 [23:40<51:13,  1.43it/s]
2|2067|Loss: 1.0666086673736572:  32%|███▏      | 2066/6470 [23:40<51:13,  1.43it/s]
2|2067|Loss: 1.0666086673736572:  32%|███▏      | 2067/6470 [23:40<51:57,  1.41it/s]
2|2068|Loss: 0.9160544872283936:  32%|███▏      | 2067/6470 [23:41<51:57,  1.41it/s]
2|2068|Loss: 0.9160544872283936:  32%|███▏      | 2068/6470 [23:41<50:54,  1.44it/s]
2|2069|Loss: 0.891653299331665:  32%|███▏      | 2068/6470 [23:41<50:54,  1.44it/s] 
2|2069|Loss: 0.891653299331665:  32%|███▏      | 2069/6470 [23:42<48:30,  1.51it/s]
2|2070|Loss: 0.8688076734542847:  32%|███▏      | 2069/6470 [23:42<48:30,  1.51it/s]
2|2070|Loss: 0.8688076734542847:  32%|███▏      | 2070/6470 [23:42<51:15,  1.43it/s]
2|2071|Loss: 0.7894994616508484:  32%|███▏      | 2070/6470 [23:43<51:15,  1.43it/s]
2|2071|Loss: 0.7894994616508484:  32%|███▏      | 2071/6470 [23:43<54:39,  1.34it/s]
2|2072|Loss: 0.8708518147468567:  32%|███▏      | 2071/6470 [23:44<54:39,  1.34it/s]
2|2072|Loss: 0.8708518147468567:  32%|███▏      | 2072/6470 [23:44<53:24,  1.37it/s]
2|2073|Loss: 0.9497761726379395:  32%|███▏      | 2072/6470 [23:44<53:24,  1.37it/s]
2|2073|Loss: 0.9497761726379395:  32%|███▏      | 2073/6470 [23:45<52:34,  1.39it/s]
2|2074|Loss: 0.7495552897453308:  32%|███▏      | 2073/6470 [23:45<52:34,  1.39it/s]
2|2074|Loss: 0.7495552897453308:  32%|███▏      | 2074/6470 [23:45<53:49,  1.36it/s]
2|2075|Loss: 0.8542431592941284:  32%|███▏      | 2074/6470 [23:46<53:49,  1.36it/s]
2|2075|Loss: 0.8542431592941284:  32%|███▏      | 2075/6470 [23:46<53:47,  1.36it/s]
2|2076|Loss: 0.9083322286605835:  32%|███▏      | 2075/6470 [23:47<53:47,  1.36it/s]
2|2076|Loss: 0.9083322286605835:  32%|███▏      | 2076/6470 [23:47<52:04,  1.41it/s]
2|2077|Loss: 0.8776820302009583:  32%|███▏      | 2076/6470 [23:47<52:04,  1.41it/s]
2|2077|Loss: 0.8776820302009583:  32%|███▏      | 2077/6470 [23:47<48:58,  1.49it/s]
2|2078|Loss: 0.8107513785362244:  32%|███▏      | 2077/6470 [23:48<48:58,  1.49it/s]
2|2078|Loss: 0.8107513785362244:  32%|███▏      | 2078/6470 [23:48<48:00,  1.52it/s]
2|2079|Loss: 0.8115386366844177:  32%|███▏      | 2078/6470 [23:48<48:00,  1.52it/s]
2|2079|Loss: 0.8115386366844177:  32%|███▏      | 2079/6470 [23:49<47:34,  1.54it/s]
2|2080|Loss: 0.7773816585540771:  32%|███▏      | 2079/6470 [23:49<47:34,  1.54it/s]
2|2080|Loss: 0.7773816585540771:  32%|███▏      | 2080/6470 [23:49<46:13,  1.58it/s]
2|2081|Loss: 0.8848583698272705:  32%|███▏      | 2080/6470 [23:50<46:13,  1.58it/s]
2|2081|Loss: 0.8848583698272705:  32%|███▏      | 2081/6470 [23:50<50:03,  1.46it/s]
2|2082|Loss: 0.9822354316711426:  32%|███▏      | 2081/6470 [23:51<50:03,  1.46it/s]
2|2082|Loss: 0.9822354316711426:  32%|███▏      | 2082/6470 [23:51<54:03,  1.35it/s]
2|2083|Loss: 0.9229605197906494:  32%|███▏      | 2082/6470 [23:51<54:03,  1.35it/s]
2|2083|Loss: 0.9229605197906494:  32%|███▏      | 2083/6470 [23:52<51:25,  1.42it/s]
2|2084|Loss: 1.02107572555542:  32%|███▏      | 2083/6470 [23:52<51:25,  1.42it/s]  
2|2084|Loss: 1.02107572555542:  32%|███▏      | 2084/6470 [23:52<49:17,  1.48it/s]
2|2085|Loss: 0.9203934073448181:  32%|███▏      | 2084/6470 [23:53<49:17,  1.48it/s]
2|2085|Loss: 0.9203934073448181:  32%|███▏      | 2085/6470 [23:53<49:34,  1.47it/s]
2|2086|Loss: 0.8970454335212708:  32%|███▏      | 2085/6470 [23:53<49:34,  1.47it/s]
2|2086|Loss: 0.8970454335212708:  32%|███▏      | 2086/6470 [23:53<49:03,  1.49it/s]
2|2087|Loss: 0.8761259913444519:  32%|███▏      | 2086/6470 [23:54<49:03,  1.49it/s]
2|2087|Loss: 0.8761259913444519:  32%|███▏      | 2087/6470 [23:54<47:15,  1.55it/s]
2|2088|Loss: 0.8067502379417419:  32%|███▏      | 2087/6470 [23:54<47:15,  1.55it/s]
2|2088|Loss: 0.8067502379417419:  32%|███▏      | 2088/6470 [23:55<46:38,  1.57it/s]
2|2089|Loss: 0.7965532541275024:  32%|███▏      | 2088/6470 [23:55<46:38,  1.57it/s]
2|2089|Loss: 0.7965532541275024:  32%|███▏      | 2089/6470 [23:56<50:41,  1.44it/s]
2|2090|Loss: 0.814765989780426:  32%|███▏      | 2089/6470 [23:56<50:41,  1.44it/s] 
2|2090|Loss: 0.814765989780426:  32%|███▏      | 2090/6470 [23:56<51:48,  1.41it/s]
2|2091|Loss: 0.8892954587936401:  32%|███▏      | 2090/6470 [23:57<51:48,  1.41it/s]
2|2091|Loss: 0.8892954587936401:  32%|███▏      | 2091/6470 [23:57<53:56,  1.35it/s]
2|2092|Loss: 0.8549201488494873:  32%|███▏      | 2091/6470 [23:58<53:56,  1.35it/s]
2|2092|Loss: 0.8549201488494873:  32%|███▏      | 2092/6470 [23:58<54:32,  1.34it/s]
2|2093|Loss: 0.7982658743858337:  32%|███▏      | 2092/6470 [23:58<54:32,  1.34it/s]
2|2093|Loss: 0.7982658743858337:  32%|███▏      | 2093/6470 [23:58<48:07,  1.52it/s]
2|2094|Loss: 0.922153115272522:  32%|███▏      | 2093/6470 [23:59<48:07,  1.52it/s] 
2|2094|Loss: 0.922153115272522:  32%|███▏      | 2094/6470 [23:59<48:46,  1.50it/s]
2|2095|Loss: 0.8347349762916565:  32%|███▏      | 2094/6470 [23:59<48:46,  1.50it/s]
2|2095|Loss: 0.8347349762916565:  32%|███▏      | 2095/6470 [24:00<52:40,  1.38it/s]
2|2096|Loss: 0.9386669993400574:  32%|███▏      | 2095/6470 [24:00<52:40,  1.38it/s]
2|2096|Loss: 0.9386669993400574:  32%|███▏      | 2096/6470 [24:01<52:02,  1.40it/s]
2|2097|Loss: 0.8556780219078064:  32%|███▏      | 2096/6470 [24:01<52:02,  1.40it/s]
2|2097|Loss: 0.8556780219078064:  32%|███▏      | 2097/6470 [24:01<53:02,  1.37it/s]
2|2098|Loss: 0.7922969460487366:  32%|███▏      | 2097/6470 [24:02<53:02,  1.37it/s]
2|2098|Loss: 0.7922969460487366:  32%|███▏      | 2098/6470 [24:02<52:55,  1.38it/s]
2|2099|Loss: 0.9673166275024414:  32%|███▏      | 2098/6470 [24:02<52:55,  1.38it/s]
2|2099|Loss: 0.9673166275024414:  32%|███▏      | 2099/6470 [24:03<50:58,  1.43it/s]
2|2100|Loss: 0.8378156423568726:  32%|███▏      | 2099/6470 [24:03<50:58,  1.43it/s]
2|2100|Loss: 0.8378156423568726:  32%|███▏      | 2100/6470 [24:03<50:32,  1.44it/s]
2|2101|Loss: 0.8772953152656555:  32%|███▏      | 2100/6470 [24:04<50:32,  1.44it/s]
2|2101|Loss: 0.8772953152656555:  32%|███▏      | 2101/6470 [24:04<52:17,  1.39it/s]
2|2102|Loss: 0.8660211563110352:  32%|███▏      | 2101/6470 [24:04<52:17,  1.39it/s]
2|2102|Loss: 0.8660211563110352:  32%|███▏      | 2102/6470 [24:05<50:24,  1.44it/s]
2|2103|Loss: 0.8364534378051758:  32%|███▏      | 2102/6470 [24:05<50:24,  1.44it/s]
2|2103|Loss: 0.8364534378051758:  33%|███▎      | 2103/6470 [24:05<49:59,  1.46it/s]
2|2104|Loss: 0.8251618146896362:  33%|███▎      | 2103/6470 [24:06<49:59,  1.46it/s]
2|2104|Loss: 0.8251618146896362:  33%|███▎      | 2104/6470 [24:06<49:13,  1.48it/s]
2|2105|Loss: 0.7823768258094788:  33%|███▎      | 2104/6470 [24:06<49:13,  1.48it/s]
2|2105|Loss: 0.7823768258094788:  33%|███▎      | 2105/6470 [24:07<52:11,  1.39it/s]
2|2106|Loss: 0.9049001336097717:  33%|███▎      | 2105/6470 [24:07<52:11,  1.39it/s]
2|2106|Loss: 0.9049001336097717:  33%|███▎      | 2106/6470 [24:08<51:32,  1.41it/s]
2|2107|Loss: 0.8791208267211914:  33%|███▎      | 2106/6470 [24:08<51:32,  1.41it/s]
2|2107|Loss: 0.8791208267211914:  33%|███▎      | 2107/6470 [24:08<47:44,  1.52it/s]
2|2108|Loss: 0.8471285700798035:  33%|███▎      | 2107/6470 [24:08<47:44,  1.52it/s]
2|2108|Loss: 0.8471285700798035:  33%|███▎      | 2108/6470 [24:09<48:30,  1.50it/s]
2|2109|Loss: 0.9573771953582764:  33%|███▎      | 2108/6470 [24:09<48:30,  1.50it/s]
2|2109|Loss: 0.9573771953582764:  33%|███▎      | 2109/6470 [24:09<48:22,  1.50it/s]
2|2110|Loss: 0.8668414950370789:  33%|███▎      | 2109/6470 [24:10<48:22,  1.50it/s]
2|2110|Loss: 0.8668414950370789:  33%|███▎      | 2110/6470 [24:10<46:09,  1.57it/s]
2|2111|Loss: 0.9672027230262756:  33%|███▎      | 2110/6470 [24:10<46:09,  1.57it/s]
2|2111|Loss: 0.9672027230262756:  33%|███▎      | 2111/6470 [24:11<44:50,  1.62it/s]
2|2112|Loss: 0.785098135471344:  33%|███▎      | 2111/6470 [24:11<44:50,  1.62it/s] 
2|2112|Loss: 0.785098135471344:  33%|███▎      | 2112/6470 [24:11<44:41,  1.63it/s]
2|2113|Loss: 0.6670706868171692:  33%|███▎      | 2112/6470 [24:12<44:41,  1.63it/s]
2|2113|Loss: 0.6670706868171692:  33%|███▎      | 2113/6470 [24:12<49:14,  1.47it/s]
2|2114|Loss: 0.8271083235740662:  33%|███▎      | 2113/6470 [24:13<49:14,  1.47it/s]
2|2114|Loss: 0.8271083235740662:  33%|███▎      | 2114/6470 [24:13<53:14,  1.36it/s]
2|2115|Loss: 0.8733594417572021:  33%|███▎      | 2114/6470 [24:13<53:14,  1.36it/s]
2|2115|Loss: 0.8733594417572021:  33%|███▎      | 2115/6470 [24:14<54:15,  1.34it/s]
2|2116|Loss: 1.0064854621887207:  33%|███▎      | 2115/6470 [24:14<54:15,  1.34it/s]
2|2116|Loss: 1.0064854621887207:  33%|███▎      | 2116/6470 [24:14<54:12,  1.34it/s]
2|2117|Loss: 0.9082784652709961:  33%|███▎      | 2116/6470 [24:15<54:12,  1.34it/s]
2|2117|Loss: 0.9082784652709961:  33%|███▎      | 2117/6470 [24:15<50:49,  1.43it/s]
2|2118|Loss: 0.929239809513092:  33%|███▎      | 2117/6470 [24:15<50:49,  1.43it/s] 
2|2118|Loss: 0.929239809513092:  33%|███▎      | 2118/6470 [24:16<46:26,  1.56it/s]
2|2119|Loss: 0.9131724834442139:  33%|███▎      | 2118/6470 [24:16<46:26,  1.56it/s]
2|2119|Loss: 0.9131724834442139:  33%|███▎      | 2119/6470 [24:16<48:29,  1.50it/s]
2|2120|Loss: 0.8487825989723206:  33%|███▎      | 2119/6470 [24:17<48:29,  1.50it/s]
2|2120|Loss: 0.8487825989723206:  33%|███▎      | 2120/6470 [24:17<46:46,  1.55it/s]
2|2121|Loss: 0.8467113375663757:  33%|███▎      | 2120/6470 [24:17<46:46,  1.55it/s]
2|2121|Loss: 0.8467113375663757:  33%|███▎      | 2121/6470 [24:18<50:19,  1.44it/s]
2|2122|Loss: 0.803449809551239:  33%|███▎      | 2121/6470 [24:18<50:19,  1.44it/s] 
2|2122|Loss: 0.803449809551239:  33%|███▎      | 2122/6470 [24:18<52:16,  1.39it/s]
2|2123|Loss: 0.7315996885299683:  33%|███▎      | 2122/6470 [24:19<52:16,  1.39it/s]
2|2123|Loss: 0.7315996885299683:  33%|███▎      | 2123/6470 [24:19<51:13,  1.41it/s]
2|2124|Loss: 0.800318717956543:  33%|███▎      | 2123/6470 [24:20<51:13,  1.41it/s] 
2|2124|Loss: 0.800318717956543:  33%|███▎      | 2124/6470 [24:20<53:35,  1.35it/s]
2|2125|Loss: 0.8863615393638611:  33%|███▎      | 2124/6470 [24:20<53:35,  1.35it/s]
2|2125|Loss: 0.8863615393638611:  33%|███▎      | 2125/6470 [24:21<52:07,  1.39it/s]
2|2126|Loss: 0.8627119064331055:  33%|███▎      | 2125/6470 [24:21<52:07,  1.39it/s]
2|2126|Loss: 0.8627119064331055:  33%|███▎      | 2126/6470 [24:21<50:14,  1.44it/s]
2|2127|Loss: 0.8049361109733582:  33%|███▎      | 2126/6470 [24:22<50:14,  1.44it/s]
2|2127|Loss: 0.8049361109733582:  33%|███▎      | 2127/6470 [24:22<52:47,  1.37it/s]
2|2128|Loss: 0.9483246803283691:  33%|███▎      | 2127/6470 [24:22<52:47,  1.37it/s]
2|2128|Loss: 0.9483246803283691:  33%|███▎      | 2128/6470 [24:23<53:04,  1.36it/s]
2|2129|Loss: 0.9648549556732178:  33%|███▎      | 2128/6470 [24:23<53:04,  1.36it/s]
2|2129|Loss: 0.9648549556732178:  33%|███▎      | 2129/6470 [24:24<53:59,  1.34it/s]
2|2130|Loss: 0.9748203158378601:  33%|███▎      | 2129/6470 [24:24<53:59,  1.34it/s]
2|2130|Loss: 0.9748203158378601:  33%|███▎      | 2130/6470 [24:24<55:05,  1.31it/s]
2|2131|Loss: 0.7564507126808167:  33%|███▎      | 2130/6470 [24:25<55:05,  1.31it/s]
2|2131|Loss: 0.7564507126808167:  33%|███▎      | 2131/6470 [24:25<51:19,  1.41it/s]
2|2132|Loss: 0.9515432715415955:  33%|███▎      | 2131/6470 [24:25<51:19,  1.41it/s]
2|2132|Loss: 0.9515432715415955:  33%|███▎      | 2132/6470 [24:26<50:47,  1.42it/s]
2|2133|Loss: 0.8410024642944336:  33%|███▎      | 2132/6470 [24:26<50:47,  1.42it/s]
2|2133|Loss: 0.8410024642944336:  33%|███▎      | 2133/6470 [24:26<48:47,  1.48it/s]
2|2134|Loss: 0.690479576587677:  33%|███▎      | 2133/6470 [24:27<48:47,  1.48it/s] 
2|2134|Loss: 0.690479576587677:  33%|███▎      | 2134/6470 [24:27<49:37,  1.46it/s]
2|2135|Loss: 0.8313573598861694:  33%|███▎      | 2134/6470 [24:27<49:37,  1.46it/s]
2|2135|Loss: 0.8313573598861694:  33%|███▎      | 2135/6470 [24:28<50:47,  1.42it/s]
2|2136|Loss: 0.8334491848945618:  33%|███▎      | 2135/6470 [24:28<50:47,  1.42it/s]
2|2136|Loss: 0.8334491848945618:  33%|███▎      | 2136/6470 [24:28<51:28,  1.40it/s]
2|2137|Loss: 0.866354763507843:  33%|███▎      | 2136/6470 [24:29<51:28,  1.40it/s] 
2|2137|Loss: 0.866354763507843:  33%|███▎      | 2137/6470 [24:29<51:34,  1.40it/s]
2|2138|Loss: 0.7813365459442139:  33%|███▎      | 2137/6470 [24:30<51:34,  1.40it/s]
2|2138|Loss: 0.7813365459442139:  33%|███▎      | 2138/6470 [24:30<51:57,  1.39it/s]
2|2139|Loss: 0.8882729411125183:  33%|███▎      | 2138/6470 [24:30<51:57,  1.39it/s]
2|2139|Loss: 0.8882729411125183:  33%|███▎      | 2139/6470 [24:31<52:41,  1.37it/s]
2|2140|Loss: 0.8918144106864929:  33%|███▎      | 2139/6470 [24:31<52:41,  1.37it/s]
2|2140|Loss: 0.8918144106864929:  33%|███▎      | 2140/6470 [24:31<52:51,  1.37it/s]
2|2141|Loss: 0.8008235096931458:  33%|███▎      | 2140/6470 [24:32<52:51,  1.37it/s]
2|2141|Loss: 0.8008235096931458:  33%|███▎      | 2141/6470 [24:32<49:42,  1.45it/s]
2|2142|Loss: 0.7483905553817749:  33%|███▎      | 2141/6470 [24:32<49:42,  1.45it/s]
2|2142|Loss: 0.7483905553817749:  33%|███▎      | 2142/6470 [24:33<51:28,  1.40it/s]
2|2143|Loss: 0.9209125638008118:  33%|███▎      | 2142/6470 [24:33<51:28,  1.40it/s]
2|2143|Loss: 0.9209125638008118:  33%|███▎      | 2143/6470 [24:34<52:32,  1.37it/s]
2|2144|Loss: 0.9601280689239502:  33%|███▎      | 2143/6470 [24:34<52:32,  1.37it/s]
2|2144|Loss: 0.9601280689239502:  33%|███▎      | 2144/6470 [24:34<51:07,  1.41it/s]
2|2145|Loss: 0.9475606679916382:  33%|███▎      | 2144/6470 [24:34<51:07,  1.41it/s]
2|2145|Loss: 0.9475606679916382:  33%|███▎      | 2145/6470 [24:35<48:00,  1.50it/s]
2|2146|Loss: 0.9075618386268616:  33%|███▎      | 2145/6470 [24:35<48:00,  1.50it/s]
2|2146|Loss: 0.9075618386268616:  33%|███▎      | 2146/6470 [24:35<48:18,  1.49it/s]
2|2147|Loss: 0.9686180353164673:  33%|███▎      | 2146/6470 [24:36<48:18,  1.49it/s]
2|2147|Loss: 0.9686180353164673:  33%|███▎      | 2147/6470 [24:36<44:51,  1.61it/s]
2|2148|Loss: 0.9223000407218933:  33%|███▎      | 2147/6470 [24:36<44:51,  1.61it/s]
2|2148|Loss: 0.9223000407218933:  33%|███▎      | 2148/6470 [24:37<45:53,  1.57it/s]
2|2149|Loss: 0.9560228586196899:  33%|███▎      | 2148/6470 [24:37<45:53,  1.57it/s]
2|2149|Loss: 0.9560228586196899:  33%|███▎      | 2149/6470 [24:37<41:41,  1.73it/s]
2|2150|Loss: 0.8026368618011475:  33%|███▎      | 2149/6470 [24:37<41:41,  1.73it/s]
2|2150|Loss: 0.8026368618011475:  33%|███▎      | 2150/6470 [24:37<38:42,  1.86it/s]
2|2151|Loss: 0.9298107028007507:  33%|███▎      | 2150/6470 [24:38<38:42,  1.86it/s]
2|2151|Loss: 0.9298107028007507:  33%|███▎      | 2151/6470 [24:38<41:01,  1.75it/s]
2|2152|Loss: 0.7083244919776917:  33%|███▎      | 2151/6470 [24:38<41:01,  1.75it/s]
2|2152|Loss: 0.7083244919776917:  33%|███▎      | 2152/6470 [24:39<40:32,  1.78it/s]
2|2153|Loss: 0.9772560596466064:  33%|███▎      | 2152/6470 [24:39<40:32,  1.78it/s]
2|2153|Loss: 0.9772560596466064:  33%|███▎      | 2153/6470 [24:39<38:57,  1.85it/s]
2|2154|Loss: 0.9023947715759277:  33%|███▎      | 2153/6470 [24:40<38:57,  1.85it/s]
2|2154|Loss: 0.9023947715759277:  33%|███▎      | 2154/6470 [24:40<41:58,  1.71it/s]
2|2155|Loss: 0.8454878330230713:  33%|███▎      | 2154/6470 [24:40<41:58,  1.71it/s]
2|2155|Loss: 0.8454878330230713:  33%|███▎      | 2155/6470 [24:41<46:27,  1.55it/s]
2|2156|Loss: 0.8999282121658325:  33%|███▎      | 2155/6470 [24:41<46:27,  1.55it/s]
2|2156|Loss: 0.8999282121658325:  33%|███▎      | 2156/6470 [24:41<43:48,  1.64it/s]
2|2157|Loss: 0.8234518766403198:  33%|███▎      | 2156/6470 [24:41<43:48,  1.64it/s]
2|2157|Loss: 0.8234518766403198:  33%|███▎      | 2157/6470 [24:42<43:36,  1.65it/s]
2|2158|Loss: 0.9223390817642212:  33%|███▎      | 2157/6470 [24:42<43:36,  1.65it/s]
2|2158|Loss: 0.9223390817642212:  33%|███▎      | 2158/6470 [24:42<42:45,  1.68it/s]
2|2159|Loss: 0.9484624266624451:  33%|███▎      | 2158/6470 [24:43<42:45,  1.68it/s]
2|2159|Loss: 0.9484624266624451:  33%|███▎      | 2159/6470 [24:43<46:39,  1.54it/s]
2|2160|Loss: 0.8747453689575195:  33%|███▎      | 2159/6470 [24:44<46:39,  1.54it/s]
2|2160|Loss: 0.8747453689575195:  33%|███▎      | 2160/6470 [24:44<57:35,  1.25it/s]
2|2161|Loss: 0.7307606339454651:  33%|███▎      | 2160/6470 [24:45<57:35,  1.25it/s]
2|2161|Loss: 0.7307606339454651:  33%|███▎      | 2161/6470 [24:45<56:41,  1.27it/s]
2|2162|Loss: 0.7852095365524292:  33%|███▎      | 2161/6470 [24:45<56:41,  1.27it/s]
2|2162|Loss: 0.7852095365524292:  33%|███▎      | 2162/6470 [24:46<55:17,  1.30it/s]
2|2163|Loss: 0.7749305963516235:  33%|███▎      | 2162/6470 [24:46<55:17,  1.30it/s]
2|2163|Loss: 0.7749305963516235:  33%|███▎      | 2163/6470 [24:46<52:10,  1.38it/s]
2|2164|Loss: 0.9128521084785461:  33%|███▎      | 2163/6470 [24:47<52:10,  1.38it/s]
2|2164|Loss: 0.9128521084785461:  33%|███▎      | 2164/6470 [24:47<48:53,  1.47it/s]
2|2165|Loss: 0.9238417744636536:  33%|███▎      | 2164/6470 [24:47<48:53,  1.47it/s]
2|2165|Loss: 0.9238417744636536:  33%|███▎      | 2165/6470 [24:48<47:24,  1.51it/s]
2|2166|Loss: 1.0086009502410889:  33%|███▎      | 2165/6470 [24:48<47:24,  1.51it/s]
2|2166|Loss: 1.0086009502410889:  33%|███▎      | 2166/6470 [24:48<46:51,  1.53it/s]
2|2167|Loss: 0.8114091157913208:  33%|███▎      | 2166/6470 [24:49<46:51,  1.53it/s]
2|2167|Loss: 0.8114091157913208:  33%|███▎      | 2167/6470 [24:49<50:47,  1.41it/s]
2|2168|Loss: 0.7804127335548401:  33%|███▎      | 2167/6470 [24:49<50:47,  1.41it/s]
2|2168|Loss: 0.7804127335548401:  34%|███▎      | 2168/6470 [24:50<47:17,  1.52it/s]
2|2169|Loss: 0.7922740578651428:  34%|███▎      | 2168/6470 [24:50<47:17,  1.52it/s]
2|2169|Loss: 0.7922740578651428:  34%|███▎      | 2169/6470 [24:50<47:32,  1.51it/s]
2|2170|Loss: 0.8439835906028748:  34%|███▎      | 2169/6470 [24:51<47:32,  1.51it/s]
2|2170|Loss: 0.8439835906028748:  34%|███▎      | 2170/6470 [24:51<44:17,  1.62it/s]
2|2171|Loss: 1.0120583772659302:  34%|███▎      | 2170/6470 [24:51<44:17,  1.62it/s]
2|2171|Loss: 1.0120583772659302:  34%|███▎      | 2171/6470 [24:51<42:42,  1.68it/s]
2|2172|Loss: 1.152047872543335:  34%|███▎      | 2171/6470 [24:52<42:42,  1.68it/s] 
2|2172|Loss: 1.152047872543335:  34%|███▎      | 2172/6470 [24:52<41:06,  1.74it/s]
2|2173|Loss: 0.8421317338943481:  34%|███▎      | 2172/6470 [24:52<41:06,  1.74it/s]
2|2173|Loss: 0.8421317338943481:  34%|███▎      | 2173/6470 [24:52<41:55,  1.71it/s]
2|2174|Loss: 0.7057653069496155:  34%|███▎      | 2173/6470 [24:53<41:55,  1.71it/s]
2|2174|Loss: 0.7057653069496155:  34%|███▎      | 2174/6470 [24:53<44:16,  1.62it/s]
2|2175|Loss: 0.8085687160491943:  34%|███▎      | 2174/6470 [24:54<44:16,  1.62it/s]
2|2175|Loss: 0.8085687160491943:  34%|███▎      | 2175/6470 [24:54<46:04,  1.55it/s]
2|2176|Loss: 0.9181138277053833:  34%|███▎      | 2175/6470 [24:54<46:04,  1.55it/s]
2|2176|Loss: 0.9181138277053833:  34%|███▎      | 2176/6470 [24:55<48:04,  1.49it/s]
2|2177|Loss: 1.064422845840454:  34%|███▎      | 2176/6470 [24:55<48:04,  1.49it/s] 
2|2177|Loss: 1.064422845840454:  34%|███▎      | 2177/6470 [24:55<49:26,  1.45it/s]
2|2178|Loss: 0.9348879456520081:  34%|███▎      | 2177/6470 [24:56<49:26,  1.45it/s]
2|2178|Loss: 0.9348879456520081:  34%|███▎      | 2178/6470 [24:56<50:19,  1.42it/s]
2|2179|Loss: 0.8557620048522949:  34%|███▎      | 2178/6470 [24:56<50:19,  1.42it/s]
2|2179|Loss: 0.8557620048522949:  34%|███▎      | 2179/6470 [24:57<48:52,  1.46it/s]
2|2180|Loss: 0.7657245397567749:  34%|███▎      | 2179/6470 [24:57<48:52,  1.46it/s]
2|2180|Loss: 0.7657245397567749:  34%|███▎      | 2180/6470 [24:57<46:11,  1.55it/s]
2|2181|Loss: 0.8153841495513916:  34%|███▎      | 2180/6470 [24:58<46:11,  1.55it/s]
2|2181|Loss: 0.8153841495513916:  34%|███▎      | 2181/6470 [24:58<49:56,  1.43it/s]
2|2182|Loss: 0.8979909420013428:  34%|███▎      | 2181/6470 [24:59<49:56,  1.43it/s]
2|2182|Loss: 0.8979909420013428:  34%|███▎      | 2182/6470 [24:59<53:15,  1.34it/s]
2|2183|Loss: 0.8087902069091797:  34%|███▎      | 2182/6470 [24:59<53:15,  1.34it/s]
2|2183|Loss: 0.8087902069091797:  34%|███▎      | 2183/6470 [25:00<52:01,  1.37it/s]
2|2184|Loss: 0.9189385175704956:  34%|███▎      | 2183/6470 [25:00<52:01,  1.37it/s]
2|2184|Loss: 0.9189385175704956:  34%|███▍      | 2184/6470 [25:00<51:41,  1.38it/s]
2|2185|Loss: 0.9470641613006592:  34%|███▍      | 2184/6470 [25:01<51:41,  1.38it/s]
2|2185|Loss: 0.9470641613006592:  34%|███▍      | 2185/6470 [25:01<48:49,  1.46it/s]
2|2186|Loss: 0.8509800434112549:  34%|███▍      | 2185/6470 [25:01<48:49,  1.46it/s]
2|2186|Loss: 0.8509800434112549:  34%|███▍      | 2186/6470 [25:02<47:32,  1.50it/s]
2|2187|Loss: 0.8227736949920654:  34%|███▍      | 2186/6470 [25:02<47:32,  1.50it/s]
2|2187|Loss: 0.8227736949920654:  34%|███▍      | 2187/6470 [25:02<48:56,  1.46it/s]
2|2188|Loss: 0.8610230088233948:  34%|███▍      | 2187/6470 [25:03<48:56,  1.46it/s]
2|2188|Loss: 0.8610230088233948:  34%|███▍      | 2188/6470 [25:03<48:07,  1.48it/s]
2|2189|Loss: 0.8542531728744507:  34%|███▍      | 2188/6470 [25:03<48:07,  1.48it/s]
2|2189|Loss: 0.8542531728744507:  34%|███▍      | 2189/6470 [25:04<48:24,  1.47it/s]
2|2190|Loss: 0.7943781018257141:  34%|███▍      | 2189/6470 [25:04<48:24,  1.47it/s]
2|2190|Loss: 0.7943781018257141:  34%|███▍      | 2190/6470 [25:04<47:24,  1.50it/s]
2|2191|Loss: 0.7605159282684326:  34%|███▍      | 2190/6470 [25:05<47:24,  1.50it/s]
2|2191|Loss: 0.7605159282684326:  34%|███▍      | 2191/6470 [25:05<48:23,  1.47it/s]
2|2192|Loss: 0.7983108758926392:  34%|███▍      | 2191/6470 [25:05<48:23,  1.47it/s]
2|2192|Loss: 0.7983108758926392:  34%|███▍      | 2192/6470 [25:06<49:34,  1.44it/s]
2|2193|Loss: 0.8595231175422668:  34%|███▍      | 2192/6470 [25:06<49:34,  1.44it/s]
2|2193|Loss: 0.8595231175422668:  34%|███▍      | 2193/6470 [25:06<49:57,  1.43it/s]
2|2194|Loss: 0.9463589191436768:  34%|███▍      | 2193/6470 [25:07<49:57,  1.43it/s]
2|2194|Loss: 0.9463589191436768:  34%|███▍      | 2194/6470 [25:07<50:26,  1.41it/s]
2|2195|Loss: 0.8710059523582458:  34%|███▍      | 2194/6470 [25:07<50:26,  1.41it/s]
2|2195|Loss: 0.8710059523582458:  34%|███▍      | 2195/6470 [25:08<49:01,  1.45it/s]
2|2196|Loss: 0.8201954364776611:  34%|███▍      | 2195/6470 [25:08<49:01,  1.45it/s]
2|2196|Loss: 0.8201954364776611:  34%|███▍      | 2196/6470 [25:08<48:57,  1.45it/s]
2|2197|Loss: 0.8508586883544922:  34%|███▍      | 2196/6470 [25:09<48:57,  1.45it/s]
2|2197|Loss: 0.8508586883544922:  34%|███▍      | 2197/6470 [25:09<46:17,  1.54it/s]
2|2198|Loss: 0.8614116311073303:  34%|███▍      | 2197/6470 [25:09<46:17,  1.54it/s]
2|2198|Loss: 0.8614116311073303:  34%|███▍      | 2198/6470 [25:10<46:37,  1.53it/s]
2|2199|Loss: 0.8976170420646667:  34%|███▍      | 2198/6470 [25:10<46:37,  1.53it/s]
2|2199|Loss: 0.8976170420646667:  34%|███▍      | 2199/6470 [25:10<46:21,  1.54it/s]
2|2200|Loss: 0.84133380651474:  34%|███▍      | 2199/6470 [25:11<46:21,  1.54it/s]  
2|2200|Loss: 0.84133380651474:  34%|███▍      | 2200/6470 [25:11<49:41,  1.43it/s]
2|2201|Loss: 0.9337602257728577:  34%|███▍      | 2200/6470 [25:12<49:41,  1.43it/s]
2|2201|Loss: 0.9337602257728577:  34%|███▍      | 2201/6470 [25:12<52:45,  1.35it/s]
2|2202|Loss: 0.7004321813583374:  34%|███▍      | 2201/6470 [25:12<52:45,  1.35it/s]
2|2202|Loss: 0.7004321813583374:  34%|███▍      | 2202/6470 [25:13<50:36,  1.41it/s]
2|2203|Loss: 0.8460936546325684:  34%|███▍      | 2202/6470 [25:13<50:36,  1.41it/s]
2|2203|Loss: 0.8460936546325684:  34%|███▍      | 2203/6470 [25:13<48:21,  1.47it/s]
2|2204|Loss: 0.7878321409225464:  34%|███▍      | 2203/6470 [25:14<48:21,  1.47it/s]
2|2204|Loss: 0.7878321409225464:  34%|███▍      | 2204/6470 [25:14<51:03,  1.39it/s]
2|2205|Loss: 1.0974349975585938:  34%|███▍      | 2204/6470 [25:14<51:03,  1.39it/s]
2|2205|Loss: 1.0974349975585938:  34%|███▍      | 2205/6470 [25:15<50:07,  1.42it/s]
2|2206|Loss: 0.8360452055931091:  34%|███▍      | 2205/6470 [25:15<50:07,  1.42it/s]
2|2206|Loss: 0.8360452055931091:  34%|███▍      | 2206/6470 [25:15<44:16,  1.61it/s]
2|2207|Loss: 0.805758535861969:  34%|███▍      | 2206/6470 [25:16<44:16,  1.61it/s] 
2|2207|Loss: 0.805758535861969:  34%|███▍      | 2207/6470 [25:16<48:25,  1.47it/s]
2|2208|Loss: 0.7148342132568359:  34%|███▍      | 2207/6470 [25:16<48:25,  1.47it/s]
2|2208|Loss: 0.7148342132568359:  34%|███▍      | 2208/6470 [25:17<51:16,  1.39it/s]
2|2209|Loss: 0.8288213610649109:  34%|███▍      | 2208/6470 [25:17<51:16,  1.39it/s]
2|2209|Loss: 0.8288213610649109:  34%|███▍      | 2209/6470 [25:18<51:49,  1.37it/s]
2|2210|Loss: 0.8413241505622864:  34%|███▍      | 2209/6470 [25:18<51:49,  1.37it/s]
2|2210|Loss: 0.8413241505622864:  34%|███▍      | 2210/6470 [25:18<49:49,  1.42it/s]
2|2211|Loss: 0.7741103172302246:  34%|███▍      | 2210/6470 [25:19<49:49,  1.42it/s]
2|2211|Loss: 0.7741103172302246:  34%|███▍      | 2211/6470 [25:19<52:04,  1.36it/s]
2|2212|Loss: 0.9393284320831299:  34%|███▍      | 2211/6470 [25:19<52:04,  1.36it/s]
2|2212|Loss: 0.9393284320831299:  34%|███▍      | 2212/6470 [25:20<53:37,  1.32it/s]
2|2213|Loss: 1.1016340255737305:  34%|███▍      | 2212/6470 [25:20<53:37,  1.32it/s]
2|2213|Loss: 1.1016340255737305:  34%|███▍      | 2213/6470 [25:20<50:51,  1.40it/s]
2|2214|Loss: 0.8121682405471802:  34%|███▍      | 2213/6470 [25:21<50:51,  1.40it/s]
2|2214|Loss: 0.8121682405471802:  34%|███▍      | 2214/6470 [25:21<48:34,  1.46it/s]
2|2215|Loss: 0.8192908763885498:  34%|███▍      | 2214/6470 [25:21<48:34,  1.46it/s]
2|2215|Loss: 0.8192908763885498:  34%|███▍      | 2215/6470 [25:22<48:35,  1.46it/s]
2|2216|Loss: 0.7512413263320923:  34%|███▍      | 2215/6470 [25:22<48:35,  1.46it/s]
2|2216|Loss: 0.7512413263320923:  34%|███▍      | 2216/6470 [25:22<47:53,  1.48it/s]
2|2217|Loss: 0.900488555431366:  34%|███▍      | 2216/6470 [25:23<47:53,  1.48it/s] 
2|2217|Loss: 0.900488555431366:  34%|███▍      | 2217/6470 [25:23<48:18,  1.47it/s]
2|2218|Loss: 0.890702486038208:  34%|███▍      | 2217/6470 [25:23<48:18,  1.47it/s]
2|2218|Loss: 0.890702486038208:  34%|███▍      | 2218/6470 [25:24<47:51,  1.48it/s]
2|2219|Loss: 0.8941633105278015:  34%|███▍      | 2218/6470 [25:24<47:51,  1.48it/s]
2|2219|Loss: 0.8941633105278015:  34%|███▍      | 2219/6470 [25:24<43:58,  1.61it/s]
2|2220|Loss: 0.7938728332519531:  34%|███▍      | 2219/6470 [25:25<43:58,  1.61it/s]
2|2220|Loss: 0.7938728332519531:  34%|███▍      | 2220/6470 [25:25<43:22,  1.63it/s]
2|2221|Loss: 0.9417179822921753:  34%|███▍      | 2220/6470 [25:25<43:22,  1.63it/s]
2|2221|Loss: 0.9417179822921753:  34%|███▍      | 2221/6470 [25:25<44:04,  1.61it/s]
2|2222|Loss: 0.8295475244522095:  34%|███▍      | 2221/6470 [25:26<44:04,  1.61it/s]
2|2222|Loss: 0.8295475244522095:  34%|███▍      | 2222/6470 [25:26<46:02,  1.54it/s]
2|2223|Loss: 0.9084358215332031:  34%|███▍      | 2222/6470 [25:27<46:02,  1.54it/s]
2|2223|Loss: 0.9084358215332031:  34%|███▍      | 2223/6470 [25:27<47:46,  1.48it/s]
2|2224|Loss: 0.8130344152450562:  34%|███▍      | 2223/6470 [25:27<47:46,  1.48it/s]
2|2224|Loss: 0.8130344152450562:  34%|███▍      | 2224/6470 [25:28<50:59,  1.39it/s]
2|2225|Loss: 0.8242660760879517:  34%|███▍      | 2224/6470 [25:28<50:59,  1.39it/s]
2|2225|Loss: 0.8242660760879517:  34%|███▍      | 2225/6470 [25:28<48:21,  1.46it/s]
2|2226|Loss: 0.8547768592834473:  34%|███▍      | 2225/6470 [25:29<48:21,  1.46it/s]
2|2226|Loss: 0.8547768592834473:  34%|███▍      | 2226/6470 [25:29<49:28,  1.43it/s]
2|2227|Loss: 1.0016028881072998:  34%|███▍      | 2226/6470 [25:29<49:28,  1.43it/s]
2|2227|Loss: 1.0016028881072998:  34%|███▍      | 2227/6470 [25:30<52:10,  1.36it/s]
2|2228|Loss: 0.8964621424674988:  34%|███▍      | 2227/6470 [25:30<52:10,  1.36it/s]
2|2228|Loss: 0.8964621424674988:  34%|███▍      | 2228/6470 [25:31<54:11,  1.30it/s]
2|2229|Loss: 0.8955405354499817:  34%|███▍      | 2228/6470 [25:31<54:11,  1.30it/s]
2|2229|Loss: 0.8955405354499817:  34%|███▍      | 2229/6470 [25:31<53:23,  1.32it/s]
2|2230|Loss: 0.9217703938484192:  34%|███▍      | 2229/6470 [25:32<53:23,  1.32it/s]
2|2230|Loss: 0.9217703938484192:  34%|███▍      | 2230/6470 [25:32<50:20,  1.40it/s]
2|2231|Loss: 0.8136571049690247:  34%|███▍      | 2230/6470 [25:32<50:20,  1.40it/s]
2|2231|Loss: 0.8136571049690247:  34%|███▍      | 2231/6470 [25:33<46:22,  1.52it/s]
2|2232|Loss: 0.8420424461364746:  34%|███▍      | 2231/6470 [25:33<46:22,  1.52it/s]
2|2232|Loss: 0.8420424461364746:  34%|███▍      | 2232/6470 [25:33<43:33,  1.62it/s]
2|2233|Loss: 0.8889252543449402:  34%|███▍      | 2232/6470 [25:33<43:33,  1.62it/s]
2|2233|Loss: 0.8889252543449402:  35%|███▍      | 2233/6470 [25:34<47:09,  1.50it/s]
2|2234|Loss: 0.8163899183273315:  35%|███▍      | 2233/6470 [25:34<47:09,  1.50it/s]
2|2234|Loss: 0.8163899183273315:  35%|███▍      | 2234/6470 [25:35<47:39,  1.48it/s]
2|2235|Loss: 0.8590551614761353:  35%|███▍      | 2234/6470 [25:35<47:39,  1.48it/s]
2|2235|Loss: 0.8590551614761353:  35%|███▍      | 2235/6470 [25:35<48:29,  1.46it/s]
2|2236|Loss: 1.0203711986541748:  35%|███▍      | 2235/6470 [25:36<48:29,  1.46it/s]
2|2236|Loss: 1.0203711986541748:  35%|███▍      | 2236/6470 [25:36<48:01,  1.47it/s]
2|2237|Loss: 0.9593068361282349:  35%|███▍      | 2236/6470 [25:36<48:01,  1.47it/s]
2|2237|Loss: 0.9593068361282349:  35%|███▍      | 2237/6470 [25:37<49:05,  1.44it/s]
2|2238|Loss: 0.7627842426300049:  35%|███▍      | 2237/6470 [25:37<49:05,  1.44it/s]
2|2238|Loss: 0.7627842426300049:  35%|███▍      | 2238/6470 [25:37<47:23,  1.49it/s]
2|2239|Loss: 0.8548827767372131:  35%|███▍      | 2238/6470 [25:38<47:23,  1.49it/s]
2|2239|Loss: 0.8548827767372131:  35%|███▍      | 2239/6470 [25:38<49:58,  1.41it/s]
2|2240|Loss: 0.8446314930915833:  35%|███▍      | 2239/6470 [25:39<49:58,  1.41it/s]
2|2240|Loss: 0.8446314930915833:  35%|███▍      | 2240/6470 [25:39<51:14,  1.38it/s]
2|2241|Loss: 0.7484008073806763:  35%|███▍      | 2240/6470 [25:39<51:14,  1.38it/s]
2|2241|Loss: 0.7484008073806763:  35%|███▍      | 2241/6470 [25:40<54:36,  1.29it/s]
2|2242|Loss: 0.6473197937011719:  35%|███▍      | 2241/6470 [25:40<54:36,  1.29it/s]
2|2242|Loss: 0.6473197937011719:  35%|███▍      | 2242/6470 [25:40<52:43,  1.34it/s]
2|2243|Loss: 0.9278983473777771:  35%|███▍      | 2242/6470 [25:41<52:43,  1.34it/s]
2|2243|Loss: 0.9278983473777771:  35%|███▍      | 2243/6470 [25:41<53:27,  1.32it/s]
2|2244|Loss: 0.900865375995636:  35%|███▍      | 2243/6470 [25:42<53:27,  1.32it/s] 
2|2244|Loss: 0.900865375995636:  35%|███▍      | 2244/6470 [25:42<51:47,  1.36it/s]
2|2245|Loss: 0.9080794453620911:  35%|███▍      | 2244/6470 [25:42<51:47,  1.36it/s]
2|2245|Loss: 0.9080794453620911:  35%|███▍      | 2245/6470 [25:43<50:48,  1.39it/s]
2|2246|Loss: 0.9027985334396362:  35%|███▍      | 2245/6470 [25:43<50:48,  1.39it/s]
2|2246|Loss: 0.9027985334396362:  35%|███▍      | 2246/6470 [25:43<48:29,  1.45it/s]
2|2247|Loss: 0.7401958703994751:  35%|███▍      | 2246/6470 [25:44<48:29,  1.45it/s]
2|2247|Loss: 0.7401958703994751:  35%|███▍      | 2247/6470 [25:44<52:30,  1.34it/s]
2|2248|Loss: 0.9224299192428589:  35%|███▍      | 2247/6470 [25:45<52:30,  1.34it/s]
2|2248|Loss: 0.9224299192428589:  35%|███▍      | 2248/6470 [25:45<53:20,  1.32it/s]
2|2249|Loss: 0.6781248450279236:  35%|███▍      | 2248/6470 [25:45<53:20,  1.32it/s]
2|2249|Loss: 0.6781248450279236:  35%|███▍      | 2249/6470 [25:46<53:47,  1.31it/s]
2|2250|Loss: 0.7086819410324097:  35%|███▍      | 2249/6470 [25:46<53:47,  1.31it/s]
2|2250|Loss: 0.7086819410324097:  35%|███▍      | 2250/6470 [25:46<53:14,  1.32it/s]
2|2251|Loss: 0.8205519318580627:  35%|███▍      | 2250/6470 [25:47<53:14,  1.32it/s]
2|2251|Loss: 0.8205519318580627:  35%|███▍      | 2251/6470 [25:47<52:12,  1.35it/s]
2|2252|Loss: 0.8979455232620239:  35%|███▍      | 2251/6470 [25:47<52:12,  1.35it/s]
2|2252|Loss: 0.8979455232620239:  35%|███▍      | 2252/6470 [25:48<49:58,  1.41it/s]
2|2253|Loss: 0.9089809060096741:  35%|███▍      | 2252/6470 [25:48<49:58,  1.41it/s]
2|2253|Loss: 0.9089809060096741:  35%|███▍      | 2253/6470 [25:48<50:56,  1.38it/s]
2|2254|Loss: 0.8102689385414124:  35%|███▍      | 2253/6470 [25:49<50:56,  1.38it/s]
2|2254|Loss: 0.8102689385414124:  35%|███▍      | 2254/6470 [25:49<51:26,  1.37it/s]
2|2255|Loss: 0.7413915991783142:  35%|███▍      | 2254/6470 [25:50<51:26,  1.37it/s]
2|2255|Loss: 0.7413915991783142:  35%|███▍      | 2255/6470 [25:50<46:39,  1.51it/s]
2|2256|Loss: 0.984428346157074:  35%|███▍      | 2255/6470 [25:50<46:39,  1.51it/s] 
2|2256|Loss: 0.984428346157074:  35%|███▍      | 2256/6470 [25:50<44:51,  1.57it/s]
2|2257|Loss: 0.8574733138084412:  35%|███▍      | 2256/6470 [25:51<44:51,  1.57it/s]
2|2257|Loss: 0.8574733138084412:  35%|███▍      | 2257/6470 [25:51<44:49,  1.57it/s]
2|2258|Loss: 0.8985039591789246:  35%|███▍      | 2257/6470 [25:51<44:49,  1.57it/s]
2|2258|Loss: 0.8985039591789246:  35%|███▍      | 2258/6470 [25:51<41:42,  1.68it/s]
2|2259|Loss: 0.9802345633506775:  35%|███▍      | 2258/6470 [25:52<41:42,  1.68it/s]
2|2259|Loss: 0.9802345633506775:  35%|███▍      | 2259/6470 [25:52<39:53,  1.76it/s]
2|2260|Loss: 0.9086253046989441:  35%|███▍      | 2259/6470 [25:52<39:53,  1.76it/s]
2|2260|Loss: 0.9086253046989441:  35%|███▍      | 2260/6470 [25:53<39:48,  1.76it/s]
2|2261|Loss: 0.9304420948028564:  35%|███▍      | 2260/6470 [25:53<39:48,  1.76it/s]
2|2261|Loss: 0.9304420948028564:  35%|███▍      | 2261/6470 [25:53<38:54,  1.80it/s]
2|2262|Loss: 0.8691902160644531:  35%|███▍      | 2261/6470 [25:53<38:54,  1.80it/s]
2|2262|Loss: 0.8691902160644531:  35%|███▍      | 2262/6470 [25:54<41:57,  1.67it/s]
2|2263|Loss: 0.8427169919013977:  35%|███▍      | 2262/6470 [25:54<41:57,  1.67it/s]
2|2263|Loss: 0.8427169919013977:  35%|███▍      | 2263/6470 [25:54<44:29,  1.58it/s]
2|2264|Loss: 0.9949905872344971:  35%|███▍      | 2263/6470 [25:55<44:29,  1.58it/s]
2|2264|Loss: 0.9949905872344971:  35%|███▍      | 2264/6470 [25:55<49:35,  1.41it/s]
2|2265|Loss: 0.7753314971923828:  35%|███▍      | 2264/6470 [25:56<49:35,  1.41it/s]
2|2265|Loss: 0.7753314971923828:  35%|███▌      | 2265/6470 [25:56<47:14,  1.48it/s]
2|2266|Loss: 0.8834162950515747:  35%|███▌      | 2265/6470 [25:56<47:14,  1.48it/s]
2|2266|Loss: 0.8834162950515747:  35%|███▌      | 2266/6470 [25:57<47:21,  1.48it/s]
2|2267|Loss: 0.9190524220466614:  35%|███▌      | 2266/6470 [25:57<47:21,  1.48it/s]
2|2267|Loss: 0.9190524220466614:  35%|███▌      | 2267/6470 [25:57<42:25,  1.65it/s]
2|2268|Loss: 0.9514126777648926:  35%|███▌      | 2267/6470 [25:57<42:25,  1.65it/s]
2|2268|Loss: 0.9514126777648926:  35%|███▌      | 2268/6470 [25:58<42:06,  1.66it/s]
2|2269|Loss: 0.7478539347648621:  35%|███▌      | 2268/6470 [25:58<42:06,  1.66it/s]
2|2269|Loss: 0.7478539347648621:  35%|███▌      | 2269/6470 [25:58<45:19,  1.54it/s]
2|2270|Loss: 0.8185389041900635:  35%|███▌      | 2269/6470 [25:59<45:19,  1.54it/s]
2|2270|Loss: 0.8185389041900635:  35%|███▌      | 2270/6470 [25:59<49:06,  1.43it/s]
2|2271|Loss: 1.050676941871643:  35%|███▌      | 2270/6470 [26:00<49:06,  1.43it/s] 
2|2271|Loss: 1.050676941871643:  35%|███▌      | 2271/6470 [26:00<46:49,  1.49it/s]
2|2272|Loss: 0.7358924746513367:  35%|███▌      | 2271/6470 [26:00<46:49,  1.49it/s]
2|2272|Loss: 0.7358924746513367:  35%|███▌      | 2272/6470 [26:01<49:22,  1.42it/s]
2|2273|Loss: 0.9019296765327454:  35%|███▌      | 2272/6470 [26:01<49:22,  1.42it/s]
2|2273|Loss: 0.9019296765327454:  35%|███▌      | 2273/6470 [26:01<48:56,  1.43it/s]
2|2274|Loss: 0.7728606462478638:  35%|███▌      | 2273/6470 [26:02<48:56,  1.43it/s]
2|2274|Loss: 0.7728606462478638:  35%|███▌      | 2274/6470 [26:02<55:42,  1.26it/s]
2|2275|Loss: 0.8871492743492126:  35%|███▌      | 2274/6470 [26:03<55:42,  1.26it/s]
2|2275|Loss: 0.8871492743492126:  35%|███▌      | 2275/6470 [26:03<54:25,  1.28it/s]
2|2276|Loss: 0.8894214034080505:  35%|███▌      | 2275/6470 [26:03<54:25,  1.28it/s]
2|2276|Loss: 0.8894214034080505:  35%|███▌      | 2276/6470 [26:04<54:36,  1.28it/s]
2|2277|Loss: 0.7908359169960022:  35%|███▌      | 2276/6470 [26:04<54:36,  1.28it/s]
2|2277|Loss: 0.7908359169960022:  35%|███▌      | 2277/6470 [26:05<52:29,  1.33it/s]
2|2278|Loss: 0.7397948503494263:  35%|███▌      | 2277/6470 [26:05<52:29,  1.33it/s]
2|2278|Loss: 0.7397948503494263:  35%|███▌      | 2278/6470 [26:05<51:22,  1.36it/s]
2|2279|Loss: 0.9486054182052612:  35%|███▌      | 2278/6470 [26:06<51:22,  1.36it/s]
2|2279|Loss: 0.9486054182052612:  35%|███▌      | 2279/6470 [26:06<49:22,  1.41it/s]
2|2280|Loss: 0.7352747917175293:  35%|███▌      | 2279/6470 [26:06<49:22,  1.41it/s]
2|2280|Loss: 0.7352747917175293:  35%|███▌      | 2280/6470 [26:07<51:29,  1.36it/s]
2|2281|Loss: 0.9356131553649902:  35%|███▌      | 2280/6470 [26:07<51:29,  1.36it/s]
2|2281|Loss: 0.9356131553649902:  35%|███▌      | 2281/6470 [26:07<51:27,  1.36it/s]
2|2282|Loss: 0.8497070074081421:  35%|███▌      | 2281/6470 [26:08<51:27,  1.36it/s]
2|2282|Loss: 0.8497070074081421:  35%|███▌      | 2282/6470 [26:08<47:26,  1.47it/s]
2|2283|Loss: 0.9868856072425842:  35%|███▌      | 2282/6470 [26:08<47:26,  1.47it/s]
2|2283|Loss: 0.9868856072425842:  35%|███▌      | 2283/6470 [26:09<49:17,  1.42it/s]
2|2284|Loss: 0.8676810264587402:  35%|███▌      | 2283/6470 [26:09<49:17,  1.42it/s]
2|2284|Loss: 0.8676810264587402:  35%|███▌      | 2284/6470 [26:10<51:51,  1.35it/s]
2|2285|Loss: 0.8366372585296631:  35%|███▌      | 2284/6470 [26:10<51:51,  1.35it/s]
2|2285|Loss: 0.8366372585296631:  35%|███▌      | 2285/6470 [26:10<48:02,  1.45it/s]
2|2286|Loss: 0.9460904598236084:  35%|███▌      | 2285/6470 [26:10<48:02,  1.45it/s]
2|2286|Loss: 0.9460904598236084:  35%|███▌      | 2286/6470 [26:11<48:51,  1.43it/s]
2|2287|Loss: 0.938606321811676:  35%|███▌      | 2286/6470 [26:11<48:51,  1.43it/s] 
2|2287|Loss: 0.938606321811676:  35%|███▌      | 2287/6470 [26:11<45:13,  1.54it/s]
2|2288|Loss: 0.9039782285690308:  35%|███▌      | 2287/6470 [26:12<45:13,  1.54it/s]
2|2288|Loss: 0.9039782285690308:  35%|███▌      | 2288/6470 [26:12<47:19,  1.47it/s]
2|2289|Loss: 0.8847237825393677:  35%|███▌      | 2288/6470 [26:13<47:19,  1.47it/s]
2|2289|Loss: 0.8847237825393677:  35%|███▌      | 2289/6470 [26:13<50:58,  1.37it/s]
2|2290|Loss: 0.8880114555358887:  35%|███▌      | 2289/6470 [26:13<50:58,  1.37it/s]
2|2290|Loss: 0.8880114555358887:  35%|███▌      | 2290/6470 [26:14<49:48,  1.40it/s]
2|2291|Loss: 0.7678155899047852:  35%|███▌      | 2290/6470 [26:14<49:48,  1.40it/s]
2|2291|Loss: 0.7678155899047852:  35%|███▌      | 2291/6470 [26:14<49:41,  1.40it/s]
2|2292|Loss: 0.823449969291687:  35%|███▌      | 2291/6470 [26:15<49:41,  1.40it/s] 
2|2292|Loss: 0.823449969291687:  35%|███▌      | 2292/6470 [26:15<50:23,  1.38it/s]
2|2293|Loss: 0.7891622185707092:  35%|███▌      | 2292/6470 [26:16<50:23,  1.38it/s]
2|2293|Loss: 0.7891622185707092:  35%|███▌      | 2293/6470 [26:16<52:53,  1.32it/s]
2|2294|Loss: 0.749725341796875:  35%|███▌      | 2293/6470 [26:16<52:53,  1.32it/s] 
2|2294|Loss: 0.749725341796875:  35%|███▌      | 2294/6470 [26:17<52:53,  1.32it/s]
2|2295|Loss: 0.8315796256065369:  35%|███▌      | 2294/6470 [26:17<52:53,  1.32it/s]
2|2295|Loss: 0.8315796256065369:  35%|███▌      | 2295/6470 [26:17<47:40,  1.46it/s]
2|2296|Loss: 1.043360948562622:  35%|███▌      | 2295/6470 [26:18<47:40,  1.46it/s] 
2|2296|Loss: 1.043360948562622:  35%|███▌      | 2296/6470 [26:18<47:20,  1.47it/s]
2|2297|Loss: 0.8731935024261475:  35%|███▌      | 2296/6470 [26:18<47:20,  1.47it/s]
2|2297|Loss: 0.8731935024261475:  36%|███▌      | 2297/6470 [26:19<48:43,  1.43it/s]
2|2298|Loss: 0.9690673351287842:  36%|███▌      | 2297/6470 [26:19<48:43,  1.43it/s]
2|2298|Loss: 0.9690673351287842:  36%|███▌      | 2298/6470 [26:19<49:06,  1.42it/s]
2|2299|Loss: 0.8148394823074341:  36%|███▌      | 2298/6470 [26:20<49:06,  1.42it/s]
2|2299|Loss: 0.8148394823074341:  36%|███▌      | 2299/6470 [26:20<49:28,  1.40it/s]
2|2300|Loss: 0.8666972517967224:  36%|███▌      | 2299/6470 [26:20<49:28,  1.40it/s]
2|2300|Loss: 0.8666972517967224:  36%|███▌      | 2300/6470 [26:21<48:14,  1.44it/s]
2|2301|Loss: 0.827911913394928:  36%|███▌      | 2300/6470 [26:21<48:14,  1.44it/s] 
2|2301|Loss: 0.827911913394928:  36%|███▌      | 2301/6470 [26:21<47:08,  1.47it/s]
2|2302|Loss: 0.959149956703186:  36%|███▌      | 2301/6470 [26:22<47:08,  1.47it/s]
2|2302|Loss: 0.959149956703186:  36%|███▌      | 2302/6470 [26:22<50:16,  1.38it/s]
2|2303|Loss: 0.868101179599762:  36%|███▌      | 2302/6470 [26:23<50:16,  1.38it/s]
2|2303|Loss: 0.868101179599762:  36%|███▌      | 2303/6470 [26:23<47:52,  1.45it/s]
2|2304|Loss: 0.8726695775985718:  36%|███▌      | 2303/6470 [26:23<47:52,  1.45it/s]
2|2304|Loss: 0.8726695775985718:  36%|███▌      | 2304/6470 [26:24<48:57,  1.42it/s]
2|2305|Loss: 0.8166489601135254:  36%|███▌      | 2304/6470 [26:24<48:57,  1.42it/s]
2|2305|Loss: 0.8166489601135254:  36%|███▌      | 2305/6470 [26:24<52:16,  1.33it/s]
2|2306|Loss: 0.8985136151313782:  36%|███▌      | 2305/6470 [26:25<52:16,  1.33it/s]
2|2306|Loss: 0.8985136151313782:  36%|███▌      | 2306/6470 [26:25<53:50,  1.29it/s]
2|2307|Loss: 0.8116989731788635:  36%|███▌      | 2306/6470 [26:26<53:50,  1.29it/s]
2|2307|Loss: 0.8116989731788635:  36%|███▌      | 2307/6470 [26:26<54:13,  1.28it/s]
2|2308|Loss: 0.9202209711074829:  36%|███▌      | 2307/6470 [26:26<54:13,  1.28it/s]
2|2308|Loss: 0.9202209711074829:  36%|███▌      | 2308/6470 [26:27<53:25,  1.30it/s]
2|2309|Loss: 0.9215508699417114:  36%|███▌      | 2308/6470 [26:27<53:25,  1.30it/s]
2|2309|Loss: 0.9215508699417114:  36%|███▌      | 2309/6470 [26:28<52:48,  1.31it/s]
2|2310|Loss: 0.7753393054008484:  36%|███▌      | 2309/6470 [26:28<52:48,  1.31it/s]
2|2310|Loss: 0.7753393054008484:  36%|███▌      | 2310/6470 [26:28<53:10,  1.30it/s]
2|2311|Loss: 0.7744450569152832:  36%|███▌      | 2310/6470 [26:29<53:10,  1.30it/s]
2|2311|Loss: 0.7744450569152832:  36%|███▌      | 2311/6470 [26:29<50:20,  1.38it/s]
2|2312|Loss: 0.9266652464866638:  36%|███▌      | 2311/6470 [26:29<50:20,  1.38it/s]
2|2312|Loss: 0.9266652464866638:  36%|███▌      | 2312/6470 [26:30<46:54,  1.48it/s]
2|2313|Loss: 0.7302719950675964:  36%|███▌      | 2312/6470 [26:30<46:54,  1.48it/s]
2|2313|Loss: 0.7302719950675964:  36%|███▌      | 2313/6470 [26:30<47:34,  1.46it/s]
2|2314|Loss: 0.7843068242073059:  36%|███▌      | 2313/6470 [26:31<47:34,  1.46it/s]
2|2314|Loss: 0.7843068242073059:  36%|███▌      | 2314/6470 [26:31<45:57,  1.51it/s]
2|2315|Loss: 0.8526975512504578:  36%|███▌      | 2314/6470 [26:31<45:57,  1.51it/s]
2|2315|Loss: 0.8526975512504578:  36%|███▌      | 2315/6470 [26:31<44:19,  1.56it/s]
2|2316|Loss: 0.8930384516716003:  36%|███▌      | 2315/6470 [26:32<44:19,  1.56it/s]
2|2316|Loss: 0.8930384516716003:  36%|███▌      | 2316/6470 [26:32<45:37,  1.52it/s]
2|2317|Loss: 0.8776389360427856:  36%|███▌      | 2316/6470 [26:33<45:37,  1.52it/s]
2|2317|Loss: 0.8776389360427856:  36%|███▌      | 2317/6470 [26:33<49:02,  1.41it/s]
2|2318|Loss: 0.933229923248291:  36%|███▌      | 2317/6470 [26:33<49:02,  1.41it/s] 
2|2318|Loss: 0.933229923248291:  36%|███▌      | 2318/6470 [26:34<48:35,  1.42it/s]
2|2319|Loss: 0.9552076458930969:  36%|███▌      | 2318/6470 [26:34<48:35,  1.42it/s]
2|2319|Loss: 0.9552076458930969:  36%|███▌      | 2319/6470 [26:34<49:18,  1.40it/s]
2|2320|Loss: 0.9288881421089172:  36%|███▌      | 2319/6470 [26:35<49:18,  1.40it/s]
2|2320|Loss: 0.9288881421089172:  36%|███▌      | 2320/6470 [26:35<49:26,  1.40it/s]
2|2321|Loss: 1.1183890104293823:  36%|███▌      | 2320/6470 [26:35<49:26,  1.40it/s]
2|2321|Loss: 1.1183890104293823:  36%|███▌      | 2321/6470 [26:36<49:57,  1.38it/s]
2|2322|Loss: 0.9286617636680603:  36%|███▌      | 2321/6470 [26:36<49:57,  1.38it/s]
2|2322|Loss: 0.9286617636680603:  36%|███▌      | 2322/6470 [26:36<48:23,  1.43it/s]
2|2323|Loss: 0.9026527404785156:  36%|███▌      | 2322/6470 [26:37<48:23,  1.43it/s]
2|2323|Loss: 0.9026527404785156:  36%|███▌      | 2323/6470 [26:37<44:27,  1.55it/s]
2|2324|Loss: 0.8561198115348816:  36%|███▌      | 2323/6470 [26:37<44:27,  1.55it/s]
2|2324|Loss: 0.8561198115348816:  36%|███▌      | 2324/6470 [26:38<42:14,  1.64it/s]
2|2325|Loss: 1.1336674690246582:  36%|███▌      | 2324/6470 [26:38<42:14,  1.64it/s]
2|2325|Loss: 1.1336674690246582:  36%|███▌      | 2325/6470 [26:38<43:35,  1.58it/s]
2|2326|Loss: 0.9149908423423767:  36%|███▌      | 2325/6470 [26:39<43:35,  1.58it/s]
2|2326|Loss: 0.9149908423423767:  36%|███▌      | 2326/6470 [26:39<46:47,  1.48it/s]
2|2327|Loss: 0.8291447758674622:  36%|███▌      | 2326/6470 [26:39<46:47,  1.48it/s]
2|2327|Loss: 0.8291447758674622:  36%|███▌      | 2327/6470 [26:40<50:23,  1.37it/s]
2|2328|Loss: 0.7994590997695923:  36%|███▌      | 2327/6470 [26:40<50:23,  1.37it/s]
2|2328|Loss: 0.7994590997695923:  36%|███▌      | 2328/6470 [26:40<48:05,  1.44it/s]
2|2329|Loss: 1.0168036222457886:  36%|███▌      | 2328/6470 [26:41<48:05,  1.44it/s]
2|2329|Loss: 1.0168036222457886:  36%|███▌      | 2329/6470 [26:41<48:21,  1.43it/s]
2|2330|Loss: 0.834400475025177:  36%|███▌      | 2329/6470 [26:42<48:21,  1.43it/s] 
2|2330|Loss: 0.834400475025177:  36%|███▌      | 2330/6470 [26:42<49:11,  1.40it/s]
2|2331|Loss: 0.8993883728981018:  36%|███▌      | 2330/6470 [26:42<49:11,  1.40it/s]
2|2331|Loss: 0.8993883728981018:  36%|███▌      | 2331/6470 [26:43<50:37,  1.36it/s]
2|2332|Loss: 0.8496406674385071:  36%|███▌      | 2331/6470 [26:43<50:37,  1.36it/s]
2|2332|Loss: 0.8496406674385071:  36%|███▌      | 2332/6470 [26:43<49:08,  1.40it/s]
2|2333|Loss: 0.8414806127548218:  36%|███▌      | 2332/6470 [26:44<49:08,  1.40it/s]
2|2333|Loss: 0.8414806127548218:  36%|███▌      | 2333/6470 [26:44<51:00,  1.35it/s]
2|2334|Loss: 0.7795920372009277:  36%|███▌      | 2333/6470 [26:45<51:00,  1.35it/s]
2|2334|Loss: 0.7795920372009277:  36%|███▌      | 2334/6470 [26:45<54:29,  1.27it/s]
2|2335|Loss: 0.8816875219345093:  36%|███▌      | 2334/6470 [26:46<54:29,  1.27it/s]
2|2335|Loss: 0.8816875219345093:  36%|███▌      | 2335/6470 [26:46<53:47,  1.28it/s]
2|2336|Loss: 0.8922815322875977:  36%|███▌      | 2335/6470 [26:46<53:47,  1.28it/s]
2|2336|Loss: 0.8922815322875977:  36%|███▌      | 2336/6470 [26:46<49:00,  1.41it/s]
2|2337|Loss: 0.9064722061157227:  36%|███▌      | 2336/6470 [26:47<49:00,  1.41it/s]
2|2337|Loss: 0.9064722061157227:  36%|███▌      | 2337/6470 [26:47<46:55,  1.47it/s]
2|2338|Loss: 0.7979332804679871:  36%|███▌      | 2337/6470 [26:47<46:55,  1.47it/s]
2|2338|Loss: 0.7979332804679871:  36%|███▌      | 2338/6470 [26:47<42:58,  1.60it/s]
2|2339|Loss: 0.6595891714096069:  36%|███▌      | 2338/6470 [26:48<42:58,  1.60it/s]
2|2339|Loss: 0.6595891714096069:  36%|███▌      | 2339/6470 [26:48<41:57,  1.64it/s]
2|2340|Loss: 0.9248384237289429:  36%|███▌      | 2339/6470 [26:48<41:57,  1.64it/s]
2|2340|Loss: 0.9248384237289429:  36%|███▌      | 2340/6470 [26:49<43:36,  1.58it/s]
2|2341|Loss: 0.8702532649040222:  36%|███▌      | 2340/6470 [26:49<43:36,  1.58it/s]
2|2341|Loss: 0.8702532649040222:  36%|███▌      | 2341/6470 [26:50<47:55,  1.44it/s]
2|2342|Loss: 0.8189688920974731:  36%|███▌      | 2341/6470 [26:50<47:55,  1.44it/s]
2|2342|Loss: 0.8189688920974731:  36%|███▌      | 2342/6470 [26:50<51:39,  1.33it/s]
2|2343|Loss: 0.8834328055381775:  36%|███▌      | 2342/6470 [26:51<51:39,  1.33it/s]
2|2343|Loss: 0.8834328055381775:  36%|███▌      | 2343/6470 [26:51<52:01,  1.32it/s]
2|2344|Loss: 0.8767827749252319:  36%|███▌      | 2343/6470 [26:52<52:01,  1.32it/s]
2|2344|Loss: 0.8767827749252319:  36%|███▌      | 2344/6470 [26:52<50:57,  1.35it/s]
2|2345|Loss: 0.6972252726554871:  36%|███▌      | 2344/6470 [26:52<50:57,  1.35it/s]
2|2345|Loss: 0.6972252726554871:  36%|███▌      | 2345/6470 [26:53<47:37,  1.44it/s]
2|2346|Loss: 0.9249445796012878:  36%|███▌      | 2345/6470 [26:53<47:37,  1.44it/s]
2|2346|Loss: 0.9249445796012878:  36%|███▋      | 2346/6470 [26:53<49:20,  1.39it/s]
2|2347|Loss: 0.770094096660614:  36%|███▋      | 2346/6470 [26:54<49:20,  1.39it/s] 
2|2347|Loss: 0.770094096660614:  36%|███▋      | 2347/6470 [26:54<49:26,  1.39it/s]
2|2348|Loss: 0.9802244305610657:  36%|███▋      | 2347/6470 [26:54<49:26,  1.39it/s]
2|2348|Loss: 0.9802244305610657:  36%|███▋      | 2348/6470 [26:55<45:52,  1.50it/s]
2|2349|Loss: 0.9529356360435486:  36%|███▋      | 2348/6470 [26:55<45:52,  1.50it/s]
2|2349|Loss: 0.9529356360435486:  36%|███▋      | 2349/6470 [26:55<45:49,  1.50it/s]
2|2350|Loss: 0.8560127019882202:  36%|███▋      | 2349/6470 [26:56<45:49,  1.50it/s]
2|2350|Loss: 0.8560127019882202:  36%|███▋      | 2350/6470 [26:56<46:36,  1.47it/s]
2|2351|Loss: 0.8490341901779175:  36%|███▋      | 2350/6470 [26:56<46:36,  1.47it/s]
2|2351|Loss: 0.8490341901779175:  36%|███▋      | 2351/6470 [26:57<44:33,  1.54it/s]
2|2352|Loss: 0.979498028755188:  36%|███▋      | 2351/6470 [26:57<44:33,  1.54it/s] 
2|2352|Loss: 0.979498028755188:  36%|███▋      | 2352/6470 [26:57<43:45,  1.57it/s]
2|2353|Loss: 0.7707024216651917:  36%|███▋      | 2352/6470 [26:57<43:45,  1.57it/s]
2|2353|Loss: 0.7707024216651917:  36%|███▋      | 2353/6470 [26:58<43:20,  1.58it/s]
2|2354|Loss: 0.9290043711662292:  36%|███▋      | 2353/6470 [26:58<43:20,  1.58it/s]
2|2354|Loss: 0.9290043711662292:  36%|███▋      | 2354/6470 [26:58<43:03,  1.59it/s]
2|2355|Loss: 0.7774681448936462:  36%|███▋      | 2354/6470 [26:59<43:03,  1.59it/s]
2|2355|Loss: 0.7774681448936462:  36%|███▋      | 2355/6470 [26:59<42:48,  1.60it/s]
2|2356|Loss: 0.8705567121505737:  36%|███▋      | 2355/6470 [26:59<42:48,  1.60it/s]
2|2356|Loss: 0.8705567121505737:  36%|███▋      | 2356/6470 [27:00<44:30,  1.54it/s]
2|2357|Loss: 0.7828917503356934:  36%|███▋      | 2356/6470 [27:00<44:30,  1.54it/s]
2|2357|Loss: 0.7828917503356934:  36%|███▋      | 2357/6470 [27:00<44:29,  1.54it/s]
2|2358|Loss: 0.8371654748916626:  36%|███▋      | 2357/6470 [27:01<44:29,  1.54it/s]
2|2358|Loss: 0.8371654748916626:  36%|███▋      | 2358/6470 [27:01<44:07,  1.55it/s]
2|2359|Loss: 0.8647872805595398:  36%|███▋      | 2358/6470 [27:01<44:07,  1.55it/s]
2|2359|Loss: 0.8647872805595398:  36%|███▋      | 2359/6470 [27:02<44:02,  1.56it/s]
2|2360|Loss: 0.7603241801261902:  36%|███▋      | 2359/6470 [27:02<44:02,  1.56it/s]
2|2360|Loss: 0.7603241801261902:  36%|███▋      | 2360/6470 [27:02<47:57,  1.43it/s]
2|2361|Loss: 0.8329797387123108:  36%|███▋      | 2360/6470 [27:03<47:57,  1.43it/s]
2|2361|Loss: 0.8329797387123108:  36%|███▋      | 2361/6470 [27:03<53:08,  1.29it/s]
2|2362|Loss: 0.7871003746986389:  36%|███▋      | 2361/6470 [27:04<53:08,  1.29it/s]
2|2362|Loss: 0.7871003746986389:  37%|███▋      | 2362/6470 [27:04<51:23,  1.33it/s]
2|2363|Loss: 0.8985669016838074:  37%|███▋      | 2362/6470 [27:04<51:23,  1.33it/s]
2|2363|Loss: 0.8985669016838074:  37%|███▋      | 2363/6470 [27:05<51:31,  1.33it/s]
2|2364|Loss: 0.8902457356452942:  37%|███▋      | 2363/6470 [27:05<51:31,  1.33it/s]
2|2364|Loss: 0.8902457356452942:  37%|███▋      | 2364/6470 [27:06<52:59,  1.29it/s]
2|2365|Loss: 0.8298830389976501:  37%|███▋      | 2364/6470 [27:06<52:59,  1.29it/s]
2|2365|Loss: 0.8298830389976501:  37%|███▋      | 2365/6470 [27:06<50:45,  1.35it/s]
2|2366|Loss: 0.9072631597518921:  37%|███▋      | 2365/6470 [27:07<50:45,  1.35it/s]
2|2366|Loss: 0.9072631597518921:  37%|███▋      | 2366/6470 [27:07<51:45,  1.32it/s]
2|2367|Loss: 0.8453400731086731:  37%|███▋      | 2366/6470 [27:08<51:45,  1.32it/s]
2|2367|Loss: 0.8453400731086731:  37%|███▋      | 2367/6470 [27:08<54:12,  1.26it/s]
2|2368|Loss: 0.8873753547668457:  37%|███▋      | 2367/6470 [27:08<54:12,  1.26it/s]
2|2368|Loss: 0.8873753547668457:  37%|███▋      | 2368/6470 [27:09<53:48,  1.27it/s]
2|2369|Loss: 0.8094860315322876:  37%|███▋      | 2368/6470 [27:09<53:48,  1.27it/s]
2|2369|Loss: 0.8094860315322876:  37%|███▋      | 2369/6470 [27:09<51:11,  1.34it/s]
2|2370|Loss: 0.709831178188324:  37%|███▋      | 2369/6470 [27:10<51:11,  1.34it/s] 
2|2370|Loss: 0.709831178188324:  37%|███▋      | 2370/6470 [27:10<48:52,  1.40it/s]
2|2371|Loss: 0.8206440806388855:  37%|███▋      | 2370/6470 [27:10<48:52,  1.40it/s]
2|2371|Loss: 0.8206440806388855:  37%|███▋      | 2371/6470 [27:11<49:13,  1.39it/s]
2|2372|Loss: 0.951111912727356:  37%|███▋      | 2371/6470 [27:11<49:13,  1.39it/s] 
2|2372|Loss: 0.951111912727356:  37%|███▋      | 2372/6470 [27:12<49:03,  1.39it/s]
2|2373|Loss: 0.9479026198387146:  37%|███▋      | 2372/6470 [27:12<49:03,  1.39it/s]
2|2373|Loss: 0.9479026198387146:  37%|███▋      | 2373/6470 [27:12<45:28,  1.50it/s]
2|2374|Loss: 0.7185639142990112:  37%|███▋      | 2373/6470 [27:12<45:28,  1.50it/s]
2|2374|Loss: 0.7185639142990112:  37%|███▋      | 2374/6470 [27:13<46:17,  1.47it/s]
2|2375|Loss: 0.8483402729034424:  37%|███▋      | 2374/6470 [27:13<46:17,  1.47it/s]
2|2375|Loss: 0.8483402729034424:  37%|███▋      | 2375/6470 [27:13<47:02,  1.45it/s]
2|2376|Loss: 0.850583553314209:  37%|███▋      | 2375/6470 [27:14<47:02,  1.45it/s] 
2|2376|Loss: 0.850583553314209:  37%|███▋      | 2376/6470 [27:14<47:49,  1.43it/s]
2|2377|Loss: 0.9338021278381348:  37%|███▋      | 2376/6470 [27:15<47:49,  1.43it/s]
2|2377|Loss: 0.9338021278381348:  37%|███▋      | 2377/6470 [27:15<48:32,  1.41it/s]
2|2378|Loss: 0.7790942192077637:  37%|███▋      | 2377/6470 [27:15<48:32,  1.41it/s]
2|2378|Loss: 0.7790942192077637:  37%|███▋      | 2378/6470 [27:16<47:16,  1.44it/s]
2|2379|Loss: 0.9146824479103088:  37%|███▋      | 2378/6470 [27:16<47:16,  1.44it/s]
2|2379|Loss: 0.9146824479103088:  37%|███▋      | 2379/6470 [27:16<49:33,  1.38it/s]
2|2380|Loss: 0.9923335909843445:  37%|███▋      | 2379/6470 [27:17<49:33,  1.38it/s]
2|2380|Loss: 0.9923335909843445:  37%|███▋      | 2380/6470 [27:17<46:16,  1.47it/s]
2|2381|Loss: 0.9471969604492188:  37%|███▋      | 2380/6470 [27:17<46:16,  1.47it/s]
2|2381|Loss: 0.9471969604492188:  37%|███▋      | 2381/6470 [27:17<43:17,  1.57it/s]
2|2382|Loss: 0.7846153974533081:  37%|███▋      | 2381/6470 [27:18<43:17,  1.57it/s]
2|2382|Loss: 0.7846153974533081:  37%|███▋      | 2382/6470 [27:18<44:34,  1.53it/s]
2|2383|Loss: 0.9224136471748352:  37%|███▋      | 2382/6470 [27:19<44:34,  1.53it/s]
2|2383|Loss: 0.9224136471748352:  37%|███▋      | 2383/6470 [27:19<41:37,  1.64it/s]
2|2384|Loss: 0.860505998134613:  37%|███▋      | 2383/6470 [27:19<41:37,  1.64it/s] 
2|2384|Loss: 0.860505998134613:  37%|███▋      | 2384/6470 [27:19<43:03,  1.58it/s]
2|2385|Loss: 0.8447116613388062:  37%|███▋      | 2384/6470 [27:20<43:03,  1.58it/s]
2|2385|Loss: 0.8447116613388062:  37%|███▋      | 2385/6470 [27:20<44:44,  1.52it/s]
2|2386|Loss: 0.6616572141647339:  37%|███▋      | 2385/6470 [27:20<44:44,  1.52it/s]
2|2386|Loss: 0.6616572141647339:  37%|███▋      | 2386/6470 [27:21<43:29,  1.56it/s]
2|2387|Loss: 0.8003895878791809:  37%|███▋      | 2386/6470 [27:21<43:29,  1.56it/s]
2|2387|Loss: 0.8003895878791809:  37%|███▋      | 2387/6470 [27:21<45:11,  1.51it/s]
2|2388|Loss: 0.9154808521270752:  37%|███▋      | 2387/6470 [27:22<45:11,  1.51it/s]
2|2388|Loss: 0.9154808521270752:  37%|███▋      | 2388/6470 [27:22<49:11,  1.38it/s]
2|2389|Loss: 1.012905240058899:  37%|███▋      | 2388/6470 [27:23<49:11,  1.38it/s] 
2|2389|Loss: 1.012905240058899:  37%|███▋      | 2389/6470 [27:23<48:45,  1.39it/s]
2|2390|Loss: 0.8985966444015503:  37%|███▋      | 2389/6470 [27:23<48:45,  1.39it/s]
2|2390|Loss: 0.8985966444015503:  37%|███▋      | 2390/6470 [27:24<48:36,  1.40it/s]
2|2391|Loss: 0.7369318604469299:  37%|███▋      | 2390/6470 [27:24<48:36,  1.40it/s]
2|2391|Loss: 0.7369318604469299:  37%|███▋      | 2391/6470 [27:24<50:05,  1.36it/s]
2|2392|Loss: 0.7637932896614075:  37%|███▋      | 2391/6470 [27:25<50:05,  1.36it/s]
2|2392|Loss: 0.7637932896614075:  37%|███▋      | 2392/6470 [27:25<48:56,  1.39it/s]
2|2393|Loss: 0.8516547083854675:  37%|███▋      | 2392/6470 [27:26<48:56,  1.39it/s]
2|2393|Loss: 0.8516547083854675:  37%|███▋      | 2393/6470 [27:26<50:51,  1.34it/s]
2|2394|Loss: 1.0568807125091553:  37%|███▋      | 2393/6470 [27:26<50:51,  1.34it/s]
2|2394|Loss: 1.0568807125091553:  37%|███▋      | 2394/6470 [27:27<49:19,  1.38it/s]
2|2395|Loss: 0.957572877407074:  37%|███▋      | 2394/6470 [27:27<49:19,  1.38it/s] 
2|2395|Loss: 0.957572877407074:  37%|███▋      | 2395/6470 [27:27<46:58,  1.45it/s]
2|2396|Loss: 0.7318344712257385:  37%|███▋      | 2395/6470 [27:28<46:58,  1.45it/s]
2|2396|Loss: 0.7318344712257385:  37%|███▋      | 2396/6470 [27:28<44:24,  1.53it/s]
2|2397|Loss: 0.817862331867218:  37%|███▋      | 2396/6470 [27:28<44:24,  1.53it/s] 
2|2397|Loss: 0.817862331867218:  37%|███▋      | 2397/6470 [27:29<46:31,  1.46it/s]
2|2398|Loss: 0.8329244256019592:  37%|███▋      | 2397/6470 [27:29<46:31,  1.46it/s]
2|2398|Loss: 0.8329244256019592:  37%|███▋      | 2398/6470 [27:29<47:44,  1.42it/s]
2|2399|Loss: 0.8038150668144226:  37%|███▋      | 2398/6470 [27:30<47:44,  1.42it/s]
2|2399|Loss: 0.8038150668144226:  37%|███▋      | 2399/6470 [27:30<49:57,  1.36it/s]
2|2400|Loss: 0.9021567106246948:  37%|███▋      | 2399/6470 [27:31<49:57,  1.36it/s]
2|2400|Loss: 0.9021567106246948:  37%|███▋      | 2400/6470 [27:31<49:46,  1.36it/s]
2|2401|Loss: 0.9257679581642151:  37%|███▋      | 2400/6470 [27:31<49:46,  1.36it/s]
2|2401|Loss: 0.9257679581642151:  37%|███▋      | 2401/6470 [27:32<47:42,  1.42it/s]
2|2402|Loss: 0.9434292316436768:  37%|███▋      | 2401/6470 [27:32<47:42,  1.42it/s]
2|2402|Loss: 0.9434292316436768:  37%|███▋      | 2402/6470 [27:32<49:44,  1.36it/s]
2|2403|Loss: 0.8910667896270752:  37%|███▋      | 2402/6470 [27:33<49:44,  1.36it/s]
2|2403|Loss: 0.8910667896270752:  37%|███▋      | 2403/6470 [27:33<48:29,  1.40it/s]
2|2404|Loss: 0.9247745871543884:  37%|███▋      | 2403/6470 [27:33<48:29,  1.40it/s]
2|2404|Loss: 0.9247745871543884:  37%|███▋      | 2404/6470 [27:34<46:51,  1.45it/s]
2|2405|Loss: 0.7399421334266663:  37%|███▋      | 2404/6470 [27:34<46:51,  1.45it/s]
2|2405|Loss: 0.7399421334266663:  37%|███▋      | 2405/6470 [27:34<49:27,  1.37it/s]
2|2406|Loss: 0.9380744695663452:  37%|███▋      | 2405/6470 [27:35<49:27,  1.37it/s]
2|2406|Loss: 0.9380744695663452:  37%|███▋      | 2406/6470 [27:35<51:44,  1.31it/s]
2|2407|Loss: 0.8985717296600342:  37%|███▋      | 2406/6470 [27:36<51:44,  1.31it/s]
2|2407|Loss: 0.8985717296600342:  37%|███▋      | 2407/6470 [27:36<51:29,  1.32it/s]
2|2408|Loss: 0.9638334512710571:  37%|███▋      | 2407/6470 [27:36<51:29,  1.32it/s]
2|2408|Loss: 0.9638334512710571:  37%|███▋      | 2408/6470 [27:37<50:20,  1.34it/s]
2|2409|Loss: 0.8979502320289612:  37%|███▋      | 2408/6470 [27:37<50:20,  1.34it/s]
2|2409|Loss: 0.8979502320289612:  37%|███▋      | 2409/6470 [27:37<50:08,  1.35it/s]
2|2410|Loss: 0.8000127673149109:  37%|███▋      | 2409/6470 [27:38<50:08,  1.35it/s]
2|2410|Loss: 0.8000127673149109:  37%|███▋      | 2410/6470 [27:38<50:23,  1.34it/s]
2|2411|Loss: 0.8671818971633911:  37%|███▋      | 2410/6470 [27:39<50:23,  1.34it/s]
2|2411|Loss: 0.8671818971633911:  37%|███▋      | 2411/6470 [27:39<48:51,  1.38it/s]
2|2412|Loss: 0.8917120695114136:  37%|███▋      | 2411/6470 [27:39<48:51,  1.38it/s]
2|2412|Loss: 0.8917120695114136:  37%|███▋      | 2412/6470 [27:40<49:36,  1.36it/s]
2|2413|Loss: 0.8177637457847595:  37%|███▋      | 2412/6470 [27:40<49:36,  1.36it/s]
2|2413|Loss: 0.8177637457847595:  37%|███▋      | 2413/6470 [27:41<53:26,  1.27it/s]
2|2414|Loss: 0.856099009513855:  37%|███▋      | 2413/6470 [27:41<53:26,  1.27it/s] 
2|2414|Loss: 0.856099009513855:  37%|███▋      | 2414/6470 [27:41<49:40,  1.36it/s]
2|2415|Loss: 0.8537616729736328:  37%|███▋      | 2414/6470 [27:42<49:40,  1.36it/s]
2|2415|Loss: 0.8537616729736328:  37%|███▋      | 2415/6470 [27:42<50:40,  1.33it/s]
2|2416|Loss: 0.9157673716545105:  37%|███▋      | 2415/6470 [27:42<50:40,  1.33it/s]
2|2416|Loss: 0.9157673716545105:  37%|███▋      | 2416/6470 [27:43<50:50,  1.33it/s]
2|2417|Loss: 0.9458160996437073:  37%|███▋      | 2416/6470 [27:43<50:50,  1.33it/s]
2|2417|Loss: 0.9458160996437073:  37%|███▋      | 2417/6470 [27:43<45:53,  1.47it/s]
2|2418|Loss: 0.8027924299240112:  37%|███▋      | 2417/6470 [27:44<45:53,  1.47it/s]
2|2418|Loss: 0.8027924299240112:  37%|███▋      | 2418/6470 [27:44<45:43,  1.48it/s]
2|2419|Loss: 0.965161144733429:  37%|███▋      | 2418/6470 [27:44<45:43,  1.48it/s] 
2|2419|Loss: 0.965161144733429:  37%|███▋      | 2419/6470 [27:45<46:27,  1.45it/s]
2|2420|Loss: 0.7622331380844116:  37%|███▋      | 2419/6470 [27:45<46:27,  1.45it/s]
2|2420|Loss: 0.7622331380844116:  37%|███▋      | 2420/6470 [27:45<47:14,  1.43it/s]
2|2421|Loss: 0.875561535358429:  37%|███▋      | 2420/6470 [27:46<47:14,  1.43it/s] 
2|2421|Loss: 0.875561535358429:  37%|███▋      | 2421/6470 [27:46<47:15,  1.43it/s]
2|2422|Loss: 0.7508763074874878:  37%|███▋      | 2421/6470 [27:46<47:15,  1.43it/s]
2|2422|Loss: 0.7508763074874878:  37%|███▋      | 2422/6470 [27:47<43:13,  1.56it/s]
2|2423|Loss: 0.7997111082077026:  37%|███▋      | 2422/6470 [27:47<43:13,  1.56it/s]
2|2423|Loss: 0.7997111082077026:  37%|███▋      | 2423/6470 [27:47<43:20,  1.56it/s]
2|2424|Loss: 0.8624238967895508:  37%|███▋      | 2423/6470 [27:48<43:20,  1.56it/s]
2|2424|Loss: 0.8624238967895508:  37%|███▋      | 2424/6470 [27:48<46:41,  1.44it/s]
2|2425|Loss: 0.8838030099868774:  37%|███▋      | 2424/6470 [27:48<46:41,  1.44it/s]
2|2425|Loss: 0.8838030099868774:  37%|███▋      | 2425/6470 [27:49<45:42,  1.48it/s]
2|2426|Loss: 0.7766667604446411:  37%|███▋      | 2425/6470 [27:49<45:42,  1.48it/s]
2|2426|Loss: 0.7766667604446411:  37%|███▋      | 2426/6470 [27:49<45:48,  1.47it/s]
2|2427|Loss: 0.9392978549003601:  37%|███▋      | 2426/6470 [27:50<45:48,  1.47it/s]
2|2427|Loss: 0.9392978549003601:  38%|███▊      | 2427/6470 [27:50<43:23,  1.55it/s]
2|2428|Loss: 0.9618320465087891:  38%|███▊      | 2427/6470 [27:50<43:23,  1.55it/s]
2|2428|Loss: 0.9618320465087891:  38%|███▊      | 2428/6470 [27:51<42:45,  1.58it/s]
2|2429|Loss: 0.7771056294441223:  38%|███▊      | 2428/6470 [27:51<42:45,  1.58it/s]
2|2429|Loss: 0.7771056294441223:  38%|███▊      | 2429/6470 [27:51<43:51,  1.54it/s]
2|2430|Loss: 0.8667322993278503:  38%|███▊      | 2429/6470 [27:52<43:51,  1.54it/s]
2|2430|Loss: 0.8667322993278503:  38%|███▊      | 2430/6470 [27:52<45:34,  1.48it/s]
2|2431|Loss: 0.7852001190185547:  38%|███▊      | 2430/6470 [27:52<45:34,  1.48it/s]
2|2431|Loss: 0.7852001190185547:  38%|███▊      | 2431/6470 [27:53<48:45,  1.38it/s]
2|2432|Loss: 0.7375365495681763:  38%|███▊      | 2431/6470 [27:53<48:45,  1.38it/s]
2|2432|Loss: 0.7375365495681763:  38%|███▊      | 2432/6470 [27:53<47:47,  1.41it/s]
2|2433|Loss: 0.7772750854492188:  38%|███▊      | 2432/6470 [27:54<47:47,  1.41it/s]
2|2433|Loss: 0.7772750854492188:  38%|███▊      | 2433/6470 [27:54<44:52,  1.50it/s]
2|2434|Loss: 0.860603392124176:  38%|███▊      | 2433/6470 [27:54<44:52,  1.50it/s] 
2|2434|Loss: 0.860603392124176:  38%|███▊      | 2434/6470 [27:55<45:44,  1.47it/s]
2|2435|Loss: 0.8504915833473206:  38%|███▊      | 2434/6470 [27:55<45:44,  1.47it/s]
2|2435|Loss: 0.8504915833473206:  38%|███▊      | 2435/6470 [27:55<45:08,  1.49it/s]
2|2436|Loss: 0.9875836372375488:  38%|███▊      | 2435/6470 [27:56<45:08,  1.49it/s]
2|2436|Loss: 0.9875836372375488:  38%|███▊      | 2436/6470 [27:56<44:25,  1.51it/s]
2|2437|Loss: 0.8278021216392517:  38%|███▊      | 2436/6470 [27:56<44:25,  1.51it/s]
2|2437|Loss: 0.8278021216392517:  38%|███▊      | 2437/6470 [27:57<43:43,  1.54it/s]
2|2438|Loss: 0.9482002854347229:  38%|███▊      | 2437/6470 [27:57<43:43,  1.54it/s]
2|2438|Loss: 0.9482002854347229:  38%|███▊      | 2438/6470 [27:57<45:07,  1.49it/s]
2|2439|Loss: 1.0015699863433838:  38%|███▊      | 2438/6470 [27:58<45:07,  1.49it/s]
2|2439|Loss: 1.0015699863433838:  38%|███▊      | 2439/6470 [27:58<45:59,  1.46it/s]
2|2440|Loss: 0.864198625087738:  38%|███▊      | 2439/6470 [27:58<45:59,  1.46it/s] 
2|2440|Loss: 0.864198625087738:  38%|███▊      | 2440/6470 [27:59<44:15,  1.52it/s]
2|2441|Loss: 0.8839778900146484:  38%|███▊      | 2440/6470 [27:59<44:15,  1.52it/s]
2|2441|Loss: 0.8839778900146484:  38%|███▊      | 2441/6470 [27:59<47:02,  1.43it/s]
2|2442|Loss: 0.8972073793411255:  38%|███▊      | 2441/6470 [28:00<47:02,  1.43it/s]
2|2442|Loss: 0.8972073793411255:  38%|███▊      | 2442/6470 [28:00<46:29,  1.44it/s]
2|2443|Loss: 0.823805570602417:  38%|███▊      | 2442/6470 [28:01<46:29,  1.44it/s] 
2|2443|Loss: 0.823805570602417:  38%|███▊      | 2443/6470 [28:01<48:48,  1.38it/s]
2|2444|Loss: 0.9280688762664795:  38%|███▊      | 2443/6470 [28:01<48:48,  1.38it/s]
2|2444|Loss: 0.9280688762664795:  38%|███▊      | 2444/6470 [28:02<49:04,  1.37it/s]
2|2445|Loss: 0.9948914647102356:  38%|███▊      | 2444/6470 [28:02<49:04,  1.37it/s]
2|2445|Loss: 0.9948914647102356:  38%|███▊      | 2445/6470 [28:02<47:14,  1.42it/s]
2|2446|Loss: 0.8017902970314026:  38%|███▊      | 2445/6470 [28:03<47:14,  1.42it/s]
2|2446|Loss: 0.8017902970314026:  38%|███▊      | 2446/6470 [28:03<44:54,  1.49it/s]
2|2447|Loss: 0.8731750845909119:  38%|███▊      | 2446/6470 [28:03<44:54,  1.49it/s]
2|2447|Loss: 0.8731750845909119:  38%|███▊      | 2447/6470 [28:04<46:31,  1.44it/s]
2|2448|Loss: 0.7880560755729675:  38%|███▊      | 2447/6470 [28:04<46:31,  1.44it/s]
2|2448|Loss: 0.7880560755729675:  38%|███▊      | 2448/6470 [28:04<47:08,  1.42it/s]
2|2449|Loss: 0.8694674372673035:  38%|███▊      | 2448/6470 [28:05<47:08,  1.42it/s]
2|2449|Loss: 0.8694674372673035:  38%|███▊      | 2449/6470 [28:05<47:44,  1.40it/s]
2|2450|Loss: 0.823907732963562:  38%|███▊      | 2449/6470 [28:05<47:44,  1.40it/s] 
2|2450|Loss: 0.823907732963562:  38%|███▊      | 2450/6470 [28:06<44:33,  1.50it/s]
2|2451|Loss: 0.9036859273910522:  38%|███▊      | 2450/6470 [28:06<44:33,  1.50it/s]
2|2451|Loss: 0.9036859273910522:  38%|███▊      | 2451/6470 [28:06<43:17,  1.55it/s]
2|2452|Loss: 0.9487614035606384:  38%|███▊      | 2451/6470 [28:07<43:17,  1.55it/s]
2|2452|Loss: 0.9487614035606384:  38%|███▊      | 2452/6470 [28:07<40:07,  1.67it/s]
2|2453|Loss: 0.9530486464500427:  38%|███▊      | 2452/6470 [28:07<40:07,  1.67it/s]
2|2453|Loss: 0.9530486464500427:  38%|███▊      | 2453/6470 [28:07<37:33,  1.78it/s]
2|2454|Loss: 0.9421069622039795:  38%|███▊      | 2453/6470 [28:08<37:33,  1.78it/s]
2|2454|Loss: 0.9421069622039795:  38%|███▊      | 2454/6470 [28:08<39:30,  1.69it/s]
2|2455|Loss: 0.8461358547210693:  38%|███▊      | 2454/6470 [28:08<39:30,  1.69it/s]
2|2455|Loss: 0.8461358547210693:  38%|███▊      | 2455/6470 [28:09<42:36,  1.57it/s]
2|2456|Loss: 0.7698890566825867:  38%|███▊      | 2455/6470 [28:09<42:36,  1.57it/s]
2|2456|Loss: 0.7698890566825867:  38%|███▊      | 2456/6470 [28:09<41:53,  1.60it/s]
2|2457|Loss: 0.8743627667427063:  38%|███▊      | 2456/6470 [28:10<41:53,  1.60it/s]
2|2457|Loss: 0.8743627667427063:  38%|███▊      | 2457/6470 [28:10<42:52,  1.56it/s]
2|2458|Loss: 0.8132610321044922:  38%|███▊      | 2457/6470 [28:10<42:52,  1.56it/s]
2|2458|Loss: 0.8132610321044922:  38%|███▊      | 2458/6470 [28:10<41:04,  1.63it/s]
2|2459|Loss: 0.9136062860488892:  38%|███▊      | 2458/6470 [28:11<41:04,  1.63it/s]
2|2459|Loss: 0.9136062860488892:  38%|███▊      | 2459/6470 [28:11<41:59,  1.59it/s]
2|2460|Loss: 0.825468897819519:  38%|███▊      | 2459/6470 [28:12<41:59,  1.59it/s] 
2|2460|Loss: 0.825468897819519:  38%|███▊      | 2460/6470 [28:12<42:08,  1.59it/s]
2|2461|Loss: 0.862073540687561:  38%|███▊      | 2460/6470 [28:12<42:08,  1.59it/s]
2|2461|Loss: 0.862073540687561:  38%|███▊      | 2461/6470 [28:12<42:13,  1.58it/s]
2|2462|Loss: 0.8586409091949463:  38%|███▊      | 2461/6470 [28:13<42:13,  1.58it/s]
2|2462|Loss: 0.8586409091949463:  38%|███▊      | 2462/6470 [28:13<41:38,  1.60it/s]
2|2463|Loss: 0.877295196056366:  38%|███▊      | 2462/6470 [28:13<41:38,  1.60it/s] 
2|2463|Loss: 0.877295196056366:  38%|███▊      | 2463/6470 [28:14<43:28,  1.54it/s]
2|2464|Loss: 1.0262411832809448:  38%|███▊      | 2463/6470 [28:14<43:28,  1.54it/s]
2|2464|Loss: 1.0262411832809448:  38%|███▊      | 2464/6470 [28:14<44:47,  1.49it/s]
2|2465|Loss: 0.7895703911781311:  38%|███▊      | 2464/6470 [28:15<44:47,  1.49it/s]
2|2465|Loss: 0.7895703911781311:  38%|███▊      | 2465/6470 [28:15<44:10,  1.51it/s]
2|2466|Loss: 0.7534610033035278:  38%|███▊      | 2465/6470 [28:16<44:10,  1.51it/s]
2|2466|Loss: 0.7534610033035278:  38%|███▊      | 2466/6470 [28:16<46:31,  1.43it/s]
2|2467|Loss: 0.9310095906257629:  38%|███▊      | 2466/6470 [28:16<46:31,  1.43it/s]
2|2467|Loss: 0.9310095906257629:  38%|███▊      | 2467/6470 [28:17<49:03,  1.36it/s]
2|2468|Loss: 0.8359014987945557:  38%|███▊      | 2467/6470 [28:17<49:03,  1.36it/s]
2|2468|Loss: 0.8359014987945557:  38%|███▊      | 2468/6470 [28:17<50:18,  1.33it/s]
2|2469|Loss: 0.8087621927261353:  38%|███▊      | 2468/6470 [28:18<50:18,  1.33it/s]
2|2469|Loss: 0.8087621927261353:  38%|███▊      | 2469/6470 [28:18<52:07,  1.28it/s]
2|2470|Loss: 0.9276240468025208:  38%|███▊      | 2469/6470 [28:19<52:07,  1.28it/s]
2|2470|Loss: 0.9276240468025208:  38%|███▊      | 2470/6470 [28:19<48:29,  1.37it/s]
2|2471|Loss: 0.8396924138069153:  38%|███▊      | 2470/6470 [28:19<48:29,  1.37it/s]
2|2471|Loss: 0.8396924138069153:  38%|███▊      | 2471/6470 [28:20<47:17,  1.41it/s]
2|2472|Loss: 0.8179007768630981:  38%|███▊      | 2471/6470 [28:20<47:17,  1.41it/s]
2|2472|Loss: 0.8179007768630981:  38%|███▊      | 2472/6470 [28:20<48:06,  1.38it/s]
2|2473|Loss: 0.8159045577049255:  38%|███▊      | 2472/6470 [28:21<48:06,  1.38it/s]
2|2473|Loss: 0.8159045577049255:  38%|███▊      | 2473/6470 [28:21<50:25,  1.32it/s]
2|2474|Loss: 0.7774753570556641:  38%|███▊      | 2473/6470 [28:22<50:25,  1.32it/s]
2|2474|Loss: 0.7774753570556641:  38%|███▊      | 2474/6470 [28:22<48:10,  1.38it/s]
2|2475|Loss: 0.8624584674835205:  38%|███▊      | 2474/6470 [28:22<48:10,  1.38it/s]
2|2475|Loss: 0.8624584674835205:  38%|███▊      | 2475/6470 [28:23<48:53,  1.36it/s]
2|2476|Loss: 0.811934769153595:  38%|███▊      | 2475/6470 [28:23<48:53,  1.36it/s] 
2|2476|Loss: 0.811934769153595:  38%|███▊      | 2476/6470 [28:23<49:12,  1.35it/s]
2|2477|Loss: 0.7596213817596436:  38%|███▊      | 2476/6470 [28:24<49:12,  1.35it/s]
2|2477|Loss: 0.7596213817596436:  38%|███▊      | 2477/6470 [28:24<48:29,  1.37it/s]
2|2478|Loss: 0.7766582369804382:  38%|███▊      | 2477/6470 [28:24<48:29,  1.37it/s]
2|2478|Loss: 0.7766582369804382:  38%|███▊      | 2478/6470 [28:25<48:18,  1.38it/s]
2|2479|Loss: 0.790144681930542:  38%|███▊      | 2478/6470 [28:25<48:18,  1.38it/s] 
2|2479|Loss: 0.790144681930542:  38%|███▊      | 2479/6470 [28:25<45:58,  1.45it/s]
2|2480|Loss: 0.7096306681632996:  38%|███▊      | 2479/6470 [28:26<45:58,  1.45it/s]
2|2480|Loss: 0.7096306681632996:  38%|███▊      | 2480/6470 [28:26<46:07,  1.44it/s]
2|2481|Loss: 0.8812446594238281:  38%|███▊      | 2480/6470 [28:26<46:07,  1.44it/s]
2|2481|Loss: 0.8812446594238281:  38%|███▊      | 2481/6470 [28:27<45:03,  1.48it/s]
2|2482|Loss: 0.9629273414611816:  38%|███▊      | 2481/6470 [28:27<45:03,  1.48it/s]
2|2482|Loss: 0.9629273414611816:  38%|███▊      | 2482/6470 [28:27<45:08,  1.47it/s]
2|2483|Loss: 0.7889232039451599:  38%|███▊      | 2482/6470 [28:28<45:08,  1.47it/s]
2|2483|Loss: 0.7889232039451599:  38%|███▊      | 2483/6470 [28:28<41:35,  1.60it/s]
2|2484|Loss: 0.8917743563652039:  38%|███▊      | 2483/6470 [28:28<41:35,  1.60it/s]
2|2484|Loss: 0.8917743563652039:  38%|███▊      | 2484/6470 [28:29<42:05,  1.58it/s]
2|2485|Loss: 0.9540430903434753:  38%|███▊      | 2484/6470 [28:29<42:05,  1.58it/s]
2|2485|Loss: 0.9540430903434753:  38%|███▊      | 2485/6470 [28:29<44:15,  1.50it/s]
2|2486|Loss: 0.7834512591362:  38%|███▊      | 2485/6470 [28:30<44:15,  1.50it/s]   
2|2486|Loss: 0.7834512591362:  38%|███▊      | 2486/6470 [28:30<45:12,  1.47it/s]
2|2487|Loss: 0.9476404190063477:  38%|███▊      | 2486/6470 [28:30<45:12,  1.47it/s]
2|2487|Loss: 0.9476404190063477:  38%|███▊      | 2487/6470 [28:31<47:01,  1.41it/s]
2|2488|Loss: 0.8072571158409119:  38%|███▊      | 2487/6470 [28:31<47:01,  1.41it/s]
2|2488|Loss: 0.8072571158409119:  38%|███▊      | 2488/6470 [28:32<47:52,  1.39it/s]
2|2489|Loss: 1.0521692037582397:  38%|███▊      | 2488/6470 [28:32<47:52,  1.39it/s]
2|2489|Loss: 1.0521692037582397:  38%|███▊      | 2489/6470 [28:32<48:23,  1.37it/s]
2|2490|Loss: 0.7837433815002441:  38%|███▊      | 2489/6470 [28:33<48:23,  1.37it/s]
2|2490|Loss: 0.7837433815002441:  38%|███▊      | 2490/6470 [28:33<49:04,  1.35it/s]
2|2491|Loss: 0.8317299485206604:  38%|███▊      | 2490/6470 [28:33<49:04,  1.35it/s]
2|2491|Loss: 0.8317299485206604:  39%|███▊      | 2491/6470 [28:34<45:28,  1.46it/s]
2|2492|Loss: 0.8596450090408325:  39%|███▊      | 2491/6470 [28:34<45:28,  1.46it/s]
2|2492|Loss: 0.8596450090408325:  39%|███▊      | 2492/6470 [28:34<43:59,  1.51it/s]
2|2493|Loss: 0.8059192299842834:  39%|███▊      | 2492/6470 [28:35<43:59,  1.51it/s]
2|2493|Loss: 0.8059192299842834:  39%|███▊      | 2493/6470 [28:35<43:34,  1.52it/s]
2|2494|Loss: 0.925832211971283:  39%|███▊      | 2493/6470 [28:35<43:34,  1.52it/s] 
2|2494|Loss: 0.925832211971283:  39%|███▊      | 2494/6470 [28:36<44:15,  1.50it/s]
2|2495|Loss: 0.8523272275924683:  39%|███▊      | 2494/6470 [28:36<44:15,  1.50it/s]
2|2495|Loss: 0.8523272275924683:  39%|███▊      | 2495/6470 [28:36<42:05,  1.57it/s]
2|2496|Loss: 0.6807611584663391:  39%|███▊      | 2495/6470 [28:36<42:05,  1.57it/s]
2|2496|Loss: 0.6807611584663391:  39%|███▊      | 2496/6470 [28:37<42:44,  1.55it/s]
2|2497|Loss: 0.848924458026886:  39%|███▊      | 2496/6470 [28:37<42:44,  1.55it/s] 
2|2497|Loss: 0.848924458026886:  39%|███▊      | 2497/6470 [28:38<46:43,  1.42it/s]
2|2498|Loss: 0.8087506294250488:  39%|███▊      | 2497/6470 [28:38<46:43,  1.42it/s]
2|2498|Loss: 0.8087506294250488:  39%|███▊      | 2498/6470 [28:39<50:02,  1.32it/s]
2|2499|Loss: 0.7974239587783813:  39%|███▊      | 2498/6470 [28:39<50:02,  1.32it/s]
2|2499|Loss: 0.7974239587783813:  39%|███▊      | 2499/6470 [28:39<49:49,  1.33it/s]
2|2500|Loss: 0.9378826022148132:  39%|███▊      | 2499/6470 [28:40<49:49,  1.33it/s]
2|2500|Loss: 0.9378826022148132:  39%|███▊      | 2500/6470 [28:40<50:46,  1.30it/s]
2|2501|Loss: 0.7580762505531311:  39%|███▊      | 2500/6470 [28:40<50:46,  1.30it/s]
2|2501|Loss: 0.7580762505531311:  39%|███▊      | 2501/6470 [28:41<49:56,  1.32it/s]
2|2502|Loss: 0.9115530848503113:  39%|███▊      | 2501/6470 [28:41<49:56,  1.32it/s]
2|2502|Loss: 0.9115530848503113:  39%|███▊      | 2502/6470 [28:41<48:53,  1.35it/s]
2|2503|Loss: 0.8362529277801514:  39%|███▊      | 2502/6470 [28:42<48:53,  1.35it/s]
2|2503|Loss: 0.8362529277801514:  39%|███▊      | 2503/6470 [28:42<49:04,  1.35it/s]
2|2504|Loss: 0.7654845714569092:  39%|███▊      | 2503/6470 [28:43<49:04,  1.35it/s]
2|2504|Loss: 0.7654845714569092:  39%|███▊      | 2504/6470 [28:43<50:56,  1.30it/s]
2|2505|Loss: 0.8975399732589722:  39%|███▊      | 2504/6470 [28:44<50:56,  1.30it/s]
2|2505|Loss: 0.8975399732589722:  39%|███▊      | 2505/6470 [28:44<51:17,  1.29it/s]
2|2506|Loss: 0.7671435475349426:  39%|███▊      | 2505/6470 [28:44<51:17,  1.29it/s]
2|2506|Loss: 0.7671435475349426:  39%|███▊      | 2506/6470 [28:44<48:25,  1.36it/s]
2|2507|Loss: 0.7128628492355347:  39%|███▊      | 2506/6470 [28:45<48:25,  1.36it/s]
2|2507|Loss: 0.7128628492355347:  39%|███▊      | 2507/6470 [28:45<45:54,  1.44it/s]
2|2508|Loss: 0.8974846005439758:  39%|███▊      | 2507/6470 [28:45<45:54,  1.44it/s]
2|2508|Loss: 0.8974846005439758:  39%|███▉      | 2508/6470 [28:46<43:22,  1.52it/s]
2|2509|Loss: 0.8927057981491089:  39%|███▉      | 2508/6470 [28:46<43:22,  1.52it/s]
2|2509|Loss: 0.8927057981491089:  39%|███▉      | 2509/6470 [28:46<39:48,  1.66it/s]
2|2510|Loss: 0.9712855815887451:  39%|███▉      | 2509/6470 [28:46<39:48,  1.66it/s]
2|2510|Loss: 0.9712855815887451:  39%|███▉      | 2510/6470 [28:47<41:12,  1.60it/s]
2|2511|Loss: 0.8135169744491577:  39%|███▉      | 2510/6470 [28:47<41:12,  1.60it/s]
2|2511|Loss: 0.8135169744491577:  39%|███▉      | 2511/6470 [28:47<41:59,  1.57it/s]
2|2512|Loss: 0.871568500995636:  39%|███▉      | 2511/6470 [28:48<41:59,  1.57it/s] 
2|2512|Loss: 0.871568500995636:  39%|███▉      | 2512/6470 [28:48<41:38,  1.58it/s]
2|2513|Loss: 0.8394676446914673:  39%|███▉      | 2512/6470 [28:48<41:38,  1.58it/s]
2|2513|Loss: 0.8394676446914673:  39%|███▉      | 2513/6470 [28:49<42:37,  1.55it/s]
2|2514|Loss: 0.9229325652122498:  39%|███▉      | 2513/6470 [28:49<42:37,  1.55it/s]
2|2514|Loss: 0.9229325652122498:  39%|███▉      | 2514/6470 [28:50<46:06,  1.43it/s]
2|2515|Loss: 0.8384804725646973:  39%|███▉      | 2514/6470 [28:50<46:06,  1.43it/s]
2|2515|Loss: 0.8384804725646973:  39%|███▉      | 2515/6470 [28:50<48:20,  1.36it/s]
2|2516|Loss: 0.858811616897583:  39%|███▉      | 2515/6470 [28:51<48:20,  1.36it/s] 
2|2516|Loss: 0.858811616897583:  39%|███▉      | 2516/6470 [28:51<44:39,  1.48it/s]
2|2517|Loss: 0.9460083842277527:  39%|███▉      | 2516/6470 [28:51<44:39,  1.48it/s]
2|2517|Loss: 0.9460083842277527:  39%|███▉      | 2517/6470 [28:51<41:28,  1.59it/s]
2|2518|Loss: 0.9162753820419312:  39%|███▉      | 2517/6470 [28:52<41:28,  1.59it/s]
2|2518|Loss: 0.9162753820419312:  39%|███▉      | 2518/6470 [28:52<46:18,  1.42it/s]
2|2519|Loss: 0.9502630829811096:  39%|███▉      | 2518/6470 [28:53<46:18,  1.42it/s]
2|2519|Loss: 0.9502630829811096:  39%|███▉      | 2519/6470 [28:53<46:26,  1.42it/s]
2|2520|Loss: 0.905497133731842:  39%|███▉      | 2519/6470 [28:53<46:26,  1.42it/s] 
2|2520|Loss: 0.905497133731842:  39%|███▉      | 2520/6470 [28:54<46:15,  1.42it/s]
2|2521|Loss: 0.817126989364624:  39%|███▉      | 2520/6470 [28:54<46:15,  1.42it/s]
2|2521|Loss: 0.817126989364624:  39%|███▉      | 2521/6470 [28:54<42:46,  1.54it/s]
2|2522|Loss: 0.7780404686927795:  39%|███▉      | 2521/6470 [28:55<42:46,  1.54it/s]
2|2522|Loss: 0.7780404686927795:  39%|███▉      | 2522/6470 [28:55<41:30,  1.59it/s]
2|2523|Loss: 0.7178557515144348:  39%|███▉      | 2522/6470 [28:55<41:30,  1.59it/s]
2|2523|Loss: 0.7178557515144348:  39%|███▉      | 2523/6470 [28:56<45:05,  1.46it/s]
2|2524|Loss: 0.7444677352905273:  39%|███▉      | 2523/6470 [28:56<45:05,  1.46it/s]
2|2524|Loss: 0.7444677352905273:  39%|███▉      | 2524/6470 [28:56<44:54,  1.46it/s]
2|2525|Loss: 0.9480399489402771:  39%|███▉      | 2524/6470 [28:57<44:54,  1.46it/s]
2|2525|Loss: 0.9480399489402771:  39%|███▉      | 2525/6470 [28:57<43:53,  1.50it/s]
2|2526|Loss: 0.8462694883346558:  39%|███▉      | 2525/6470 [28:57<43:53,  1.50it/s]
2|2526|Loss: 0.8462694883346558:  39%|███▉      | 2526/6470 [28:58<41:16,  1.59it/s]
2|2527|Loss: 1.0276988744735718:  39%|███▉      | 2526/6470 [28:58<41:16,  1.59it/s]
2|2527|Loss: 1.0276988744735718:  39%|███▉      | 2527/6470 [28:58<43:15,  1.52it/s]
2|2528|Loss: 0.8690944910049438:  39%|███▉      | 2527/6470 [28:59<43:15,  1.52it/s]
2|2528|Loss: 0.8690944910049438:  39%|███▉      | 2528/6470 [28:59<43:35,  1.51it/s]
2|2529|Loss: 0.8031722903251648:  39%|███▉      | 2528/6470 [28:59<43:35,  1.51it/s]
2|2529|Loss: 0.8031722903251648:  39%|███▉      | 2529/6470 [29:00<43:11,  1.52it/s]
2|2530|Loss: 1.0451337099075317:  39%|███▉      | 2529/6470 [29:00<43:11,  1.52it/s]
2|2530|Loss: 1.0451337099075317:  39%|███▉      | 2530/6470 [29:00<45:12,  1.45it/s]
2|2531|Loss: 0.8274393081665039:  39%|███▉      | 2530/6470 [29:01<45:12,  1.45it/s]
2|2531|Loss: 0.8274393081665039:  39%|███▉      | 2531/6470 [29:01<47:47,  1.37it/s]
2|2532|Loss: 0.766671895980835:  39%|███▉      | 2531/6470 [29:02<47:47,  1.37it/s] 
2|2532|Loss: 0.766671895980835:  39%|███▉      | 2532/6470 [29:02<48:34,  1.35it/s]
2|2533|Loss: 1.0158635377883911:  39%|███▉      | 2532/6470 [29:02<48:34,  1.35it/s]
2|2533|Loss: 1.0158635377883911:  39%|███▉      | 2533/6470 [29:03<47:11,  1.39it/s]
2|2534|Loss: 0.9748863577842712:  39%|███▉      | 2533/6470 [29:03<47:11,  1.39it/s]
2|2534|Loss: 0.9748863577842712:  39%|███▉      | 2534/6470 [29:03<47:23,  1.38it/s]
2|2535|Loss: 1.01404869556427:  39%|███▉      | 2534/6470 [29:04<47:23,  1.38it/s]  
2|2535|Loss: 1.01404869556427:  39%|███▉      | 2535/6470 [29:04<45:58,  1.43it/s]
2|2536|Loss: 0.9894858002662659:  39%|███▉      | 2535/6470 [29:04<45:58,  1.43it/s]
2|2536|Loss: 0.9894858002662659:  39%|███▉      | 2536/6470 [29:04<41:46,  1.57it/s]
2|2537|Loss: 0.8067460060119629:  39%|███▉      | 2536/6470 [29:05<41:46,  1.57it/s]
2|2537|Loss: 0.8067460060119629:  39%|███▉      | 2537/6470 [29:05<42:20,  1.55it/s]
2|2538|Loss: 0.7513176202774048:  39%|███▉      | 2537/6470 [29:06<42:20,  1.55it/s]
2|2538|Loss: 0.7513176202774048:  39%|███▉      | 2538/6470 [29:06<41:47,  1.57it/s]
2|2539|Loss: 0.8774015307426453:  39%|███▉      | 2538/6470 [29:06<41:47,  1.57it/s]
2|2539|Loss: 0.8774015307426453:  39%|███▉      | 2539/6470 [29:06<43:24,  1.51it/s]
2|2540|Loss: 0.7850188612937927:  39%|███▉      | 2539/6470 [29:07<43:24,  1.51it/s]
2|2540|Loss: 0.7850188612937927:  39%|███▉      | 2540/6470 [29:07<43:30,  1.51it/s]
2|2541|Loss: 0.689875602722168:  39%|███▉      | 2540/6470 [29:08<43:30,  1.51it/s] 
2|2541|Loss: 0.689875602722168:  39%|███▉      | 2541/6470 [29:08<45:52,  1.43it/s]
2|2542|Loss: 0.8897514343261719:  39%|███▉      | 2541/6470 [29:08<45:52,  1.43it/s]
2|2542|Loss: 0.8897514343261719:  39%|███▉      | 2542/6470 [29:09<45:31,  1.44it/s]
2|2543|Loss: 0.7942142486572266:  39%|███▉      | 2542/6470 [29:09<45:31,  1.44it/s]
2|2543|Loss: 0.7942142486572266:  39%|███▉      | 2543/6470 [29:09<43:22,  1.51it/s]
2|2544|Loss: 0.7686445713043213:  39%|███▉      | 2543/6470 [29:10<43:22,  1.51it/s]
2|2544|Loss: 0.7686445713043213:  39%|███▉      | 2544/6470 [29:10<42:38,  1.53it/s]
2|2545|Loss: 0.8142244219779968:  39%|███▉      | 2544/6470 [29:10<42:38,  1.53it/s]
2|2545|Loss: 0.8142244219779968:  39%|███▉      | 2545/6470 [29:10<42:33,  1.54it/s]
2|2546|Loss: 0.9445130228996277:  39%|███▉      | 2545/6470 [29:11<42:33,  1.54it/s]
2|2546|Loss: 0.9445130228996277:  39%|███▉      | 2546/6470 [29:11<43:49,  1.49it/s]
2|2547|Loss: 0.9523104429244995:  39%|███▉      | 2546/6470 [29:12<43:49,  1.49it/s]
2|2547|Loss: 0.9523104429244995:  39%|███▉      | 2547/6470 [29:12<42:13,  1.55it/s]
2|2548|Loss: 0.8491954803466797:  39%|███▉      | 2547/6470 [29:12<42:13,  1.55it/s]
2|2548|Loss: 0.8491954803466797:  39%|███▉      | 2548/6470 [29:12<41:28,  1.58it/s]
2|2549|Loss: 0.8269274830818176:  39%|███▉      | 2548/6470 [29:13<41:28,  1.58it/s]
2|2549|Loss: 0.8269274830818176:  39%|███▉      | 2549/6470 [29:13<42:28,  1.54it/s]
2|2550|Loss: 0.8444251418113708:  39%|███▉      | 2549/6470 [29:13<42:28,  1.54it/s]
2|2550|Loss: 0.8444251418113708:  39%|███▉      | 2550/6470 [29:14<42:06,  1.55it/s]
2|2551|Loss: 0.838038980960846:  39%|███▉      | 2550/6470 [29:14<42:06,  1.55it/s] 
2|2551|Loss: 0.838038980960846:  39%|███▉      | 2551/6470 [29:14<43:30,  1.50it/s]
2|2552|Loss: 0.7084386944770813:  39%|███▉      | 2551/6470 [29:15<43:30,  1.50it/s]
2|2552|Loss: 0.7084386944770813:  39%|███▉      | 2552/6470 [29:15<42:03,  1.55it/s]
2|2553|Loss: 0.8223496079444885:  39%|███▉      | 2552/6470 [29:15<42:03,  1.55it/s]
2|2553|Loss: 0.8223496079444885:  39%|███▉      | 2553/6470 [29:16<41:12,  1.58it/s]
2|2554|Loss: 0.7958711385726929:  39%|███▉      | 2553/6470 [29:16<41:12,  1.58it/s]
2|2554|Loss: 0.7958711385726929:  39%|███▉      | 2554/6470 [29:16<40:16,  1.62it/s]
2|2555|Loss: 0.8029759526252747:  39%|███▉      | 2554/6470 [29:17<40:16,  1.62it/s]
2|2555|Loss: 0.8029759526252747:  39%|███▉      | 2555/6470 [29:17<40:06,  1.63it/s]
2|2556|Loss: 0.9029319882392883:  39%|███▉      | 2555/6470 [29:17<40:06,  1.63it/s]
2|2556|Loss: 0.9029319882392883:  40%|███▉      | 2556/6470 [29:17<38:30,  1.69it/s]
2|2557|Loss: 0.7886573672294617:  40%|███▉      | 2556/6470 [29:18<38:30,  1.69it/s]
2|2557|Loss: 0.7886573672294617:  40%|███▉      | 2557/6470 [29:18<40:42,  1.60it/s]
2|2558|Loss: 0.705211341381073:  40%|███▉      | 2557/6470 [29:18<40:42,  1.60it/s] 
2|2558|Loss: 0.705211341381073:  40%|███▉      | 2558/6470 [29:19<42:58,  1.52it/s]
2|2559|Loss: 0.8317872881889343:  40%|███▉      | 2558/6470 [29:19<42:58,  1.52it/s]
2|2559|Loss: 0.8317872881889343:  40%|███▉      | 2559/6470 [29:19<42:35,  1.53it/s]
2|2560|Loss: 0.704944908618927:  40%|███▉      | 2559/6470 [29:20<42:35,  1.53it/s] 
2|2560|Loss: 0.704944908618927:  40%|███▉      | 2560/6470 [29:20<44:25,  1.47it/s]
2|2561|Loss: 0.8383229374885559:  40%|███▉      | 2560/6470 [29:21<44:25,  1.47it/s]
2|2561|Loss: 0.8383229374885559:  40%|███▉      | 2561/6470 [29:21<43:58,  1.48it/s]
2|2562|Loss: 0.8417633175849915:  40%|███▉      | 2561/6470 [29:21<43:58,  1.48it/s]
2|2562|Loss: 0.8417633175849915:  40%|███▉      | 2562/6470 [29:22<44:42,  1.46it/s]
2|2563|Loss: 0.740268349647522:  40%|███▉      | 2562/6470 [29:22<44:42,  1.46it/s] 
2|2563|Loss: 0.740268349647522:  40%|███▉      | 2563/6470 [29:22<44:10,  1.47it/s]
2|2564|Loss: 0.791032612323761:  40%|███▉      | 2563/6470 [29:23<44:10,  1.47it/s]
2|2564|Loss: 0.791032612323761:  40%|███▉      | 2564/6470 [29:23<41:22,  1.57it/s]
2|2565|Loss: 0.9134352803230286:  40%|███▉      | 2564/6470 [29:23<41:22,  1.57it/s]
2|2565|Loss: 0.9134352803230286:  40%|███▉      | 2565/6470 [29:23<36:56,  1.76it/s]
2|2566|Loss: 0.8457021713256836:  40%|███▉      | 2565/6470 [29:24<36:56,  1.76it/s]
2|2566|Loss: 0.8457021713256836:  40%|███▉      | 2566/6470 [29:24<41:14,  1.58it/s]
2|2567|Loss: 0.7896791696548462:  40%|███▉      | 2566/6470 [29:24<41:14,  1.58it/s]
2|2567|Loss: 0.7896791696548462:  40%|███▉      | 2567/6470 [29:25<45:25,  1.43it/s]
2|2568|Loss: 0.8970420956611633:  40%|███▉      | 2567/6470 [29:25<45:25,  1.43it/s]
2|2568|Loss: 0.8970420956611633:  40%|███▉      | 2568/6470 [29:26<46:34,  1.40it/s]
2|2569|Loss: 0.8265174031257629:  40%|███▉      | 2568/6470 [29:26<46:34,  1.40it/s]
2|2569|Loss: 0.8265174031257629:  40%|███▉      | 2569/6470 [29:26<43:21,  1.50it/s]
2|2570|Loss: 0.8535673022270203:  40%|███▉      | 2569/6470 [29:26<43:21,  1.50it/s]
2|2570|Loss: 0.8535673022270203:  40%|███▉      | 2570/6470 [29:27<44:36,  1.46it/s]
2|2571|Loss: 0.8858192563056946:  40%|███▉      | 2570/6470 [29:27<44:36,  1.46it/s]
2|2571|Loss: 0.8858192563056946:  40%|███▉      | 2571/6470 [29:27<44:05,  1.47it/s]
2|2572|Loss: 0.8453759551048279:  40%|███▉      | 2571/6470 [29:28<44:05,  1.47it/s]
2|2572|Loss: 0.8453759551048279:  40%|███▉      | 2572/6470 [29:28<45:00,  1.44it/s]
2|2573|Loss: 0.886934220790863:  40%|███▉      | 2572/6470 [29:29<45:00,  1.44it/s] 
2|2573|Loss: 0.886934220790863:  40%|███▉      | 2573/6470 [29:29<47:02,  1.38it/s]
2|2574|Loss: 0.8702326416969299:  40%|███▉      | 2573/6470 [29:29<47:02,  1.38it/s]
2|2574|Loss: 0.8702326416969299:  40%|███▉      | 2574/6470 [29:30<47:39,  1.36it/s]
2|2575|Loss: 1.0085959434509277:  40%|███▉      | 2574/6470 [29:30<47:39,  1.36it/s]
2|2575|Loss: 1.0085959434509277:  40%|███▉      | 2575/6470 [29:31<49:18,  1.32it/s]
2|2576|Loss: 0.6807594895362854:  40%|███▉      | 2575/6470 [29:31<49:18,  1.32it/s]
2|2576|Loss: 0.6807594895362854:  40%|███▉      | 2576/6470 [29:31<47:47,  1.36it/s]
2|2577|Loss: 1.0172793865203857:  40%|███▉      | 2576/6470 [29:32<47:47,  1.36it/s]
2|2577|Loss: 1.0172793865203857:  40%|███▉      | 2577/6470 [29:32<46:48,  1.39it/s]
2|2578|Loss: 0.7150756120681763:  40%|███▉      | 2577/6470 [29:32<46:48,  1.39it/s]
2|2578|Loss: 0.7150756120681763:  40%|███▉      | 2578/6470 [29:33<47:03,  1.38it/s]
2|2579|Loss: 0.8587458729743958:  40%|███▉      | 2578/6470 [29:33<47:03,  1.38it/s]
2|2579|Loss: 0.8587458729743958:  40%|███▉      | 2579/6470 [29:33<48:26,  1.34it/s]
2|2580|Loss: 1.0383998155593872:  40%|███▉      | 2579/6470 [29:34<48:26,  1.34it/s]
2|2580|Loss: 1.0383998155593872:  40%|███▉      | 2580/6470 [29:34<45:03,  1.44it/s]
2|2581|Loss: 0.8205376267433167:  40%|███▉      | 2580/6470 [29:34<45:03,  1.44it/s]
2|2581|Loss: 0.8205376267433167:  40%|███▉      | 2581/6470 [29:35<44:48,  1.45it/s]
2|2582|Loss: 0.8314070701599121:  40%|███▉      | 2581/6470 [29:35<44:48,  1.45it/s]
2|2582|Loss: 0.8314070701599121:  40%|███▉      | 2582/6470 [29:36<47:03,  1.38it/s]
2|2583|Loss: 0.7432960867881775:  40%|███▉      | 2582/6470 [29:36<47:03,  1.38it/s]
2|2583|Loss: 0.7432960867881775:  40%|███▉      | 2583/6470 [29:36<48:01,  1.35it/s]
2|2584|Loss: 0.9572023749351501:  40%|███▉      | 2583/6470 [29:37<48:01,  1.35it/s]
2|2584|Loss: 0.9572023749351501:  40%|███▉      | 2584/6470 [29:37<48:52,  1.33it/s]
2|2585|Loss: 0.9071729779243469:  40%|███▉      | 2584/6470 [29:38<48:52,  1.33it/s]
2|2585|Loss: 0.9071729779243469:  40%|███▉      | 2585/6470 [29:38<48:27,  1.34it/s]
2|2586|Loss: 0.9415714740753174:  40%|███▉      | 2585/6470 [29:38<48:27,  1.34it/s]
2|2586|Loss: 0.9415714740753174:  40%|███▉      | 2586/6470 [29:39<50:35,  1.28it/s]
2|2587|Loss: 0.9395337700843811:  40%|███▉      | 2586/6470 [29:39<50:35,  1.28it/s]
2|2587|Loss: 0.9395337700843811:  40%|███▉      | 2587/6470 [29:39<47:15,  1.37it/s]
2|2588|Loss: 0.8128858208656311:  40%|███▉      | 2587/6470 [29:40<47:15,  1.37it/s]
2|2588|Loss: 0.8128858208656311:  40%|████      | 2588/6470 [29:40<46:37,  1.39it/s]
2|2589|Loss: 0.8390870094299316:  40%|████      | 2588/6470 [29:40<46:37,  1.39it/s]
2|2589|Loss: 0.8390870094299316:  40%|████      | 2589/6470 [29:41<46:59,  1.38it/s]
2|2590|Loss: 1.0607194900512695:  40%|████      | 2589/6470 [29:41<46:59,  1.38it/s]
2|2590|Loss: 1.0607194900512695:  40%|████      | 2590/6470 [29:42<47:21,  1.37it/s]
2|2591|Loss: 0.8631724715232849:  40%|████      | 2590/6470 [29:42<47:21,  1.37it/s]
2|2591|Loss: 0.8631724715232849:  40%|████      | 2591/6470 [29:42<45:35,  1.42it/s]
2|2592|Loss: 1.0242207050323486:  40%|████      | 2591/6470 [29:43<45:35,  1.42it/s]
2|2592|Loss: 1.0242207050323486:  40%|████      | 2592/6470 [29:43<45:14,  1.43it/s]
2|2593|Loss: 0.9713694453239441:  40%|████      | 2592/6470 [29:43<45:14,  1.43it/s]
2|2593|Loss: 0.9713694453239441:  40%|████      | 2593/6470 [29:44<46:37,  1.39it/s]
2|2594|Loss: 0.761038064956665:  40%|████      | 2593/6470 [29:44<46:37,  1.39it/s] 
2|2594|Loss: 0.761038064956665:  40%|████      | 2594/6470 [29:44<46:40,  1.38it/s]
2|2595|Loss: 0.9241341948509216:  40%|████      | 2594/6470 [29:45<46:40,  1.38it/s]
2|2595|Loss: 0.9241341948509216:  40%|████      | 2595/6470 [29:45<46:16,  1.40it/s]
2|2596|Loss: 0.8228204846382141:  40%|████      | 2595/6470 [29:45<46:16,  1.40it/s]
2|2596|Loss: 0.8228204846382141:  40%|████      | 2596/6470 [29:46<45:59,  1.40it/s]
2|2597|Loss: 0.8521605134010315:  40%|████      | 2596/6470 [29:46<45:59,  1.40it/s]
2|2597|Loss: 0.8521605134010315:  40%|████      | 2597/6470 [29:46<46:16,  1.39it/s]
2|2598|Loss: 0.803813636302948:  40%|████      | 2597/6470 [29:47<46:16,  1.39it/s] 
2|2598|Loss: 0.803813636302948:  40%|████      | 2598/6470 [29:47<45:09,  1.43it/s]
2|2599|Loss: 0.8677701950073242:  40%|████      | 2598/6470 [29:47<45:09,  1.43it/s]
2|2599|Loss: 0.8677701950073242:  40%|████      | 2599/6470 [29:48<43:45,  1.47it/s]
2|2600|Loss: 0.7377334833145142:  40%|████      | 2599/6470 [29:48<43:45,  1.47it/s]
2|2600|Loss: 0.7377334833145142:  40%|████      | 2600/6470 [29:48<44:00,  1.47it/s]
2|2601|Loss: 0.8046015501022339:  40%|████      | 2600/6470 [29:49<44:00,  1.47it/s]
2|2601|Loss: 0.8046015501022339:  40%|████      | 2601/6470 [29:49<45:14,  1.43it/s]
2|2602|Loss: 0.9288989901542664:  40%|████      | 2601/6470 [29:50<45:14,  1.43it/s]
2|2602|Loss: 0.9288989901542664:  40%|████      | 2602/6470 [29:50<44:34,  1.45it/s]
2|2603|Loss: 0.823824405670166:  40%|████      | 2602/6470 [29:50<44:34,  1.45it/s] 
2|2603|Loss: 0.823824405670166:  40%|████      | 2603/6470 [29:51<45:56,  1.40it/s]
2|2604|Loss: 0.930867612361908:  40%|████      | 2603/6470 [29:51<45:56,  1.40it/s]
2|2604|Loss: 0.930867612361908:  40%|████      | 2604/6470 [29:51<48:11,  1.34it/s]
2|2605|Loss: 0.9030036330223083:  40%|████      | 2604/6470 [29:52<48:11,  1.34it/s]
2|2605|Loss: 0.9030036330223083:  40%|████      | 2605/6470 [29:52<46:56,  1.37it/s]
2|2606|Loss: 0.9047784209251404:  40%|████      | 2605/6470 [29:52<46:56,  1.37it/s]
2|2606|Loss: 0.9047784209251404:  40%|████      | 2606/6470 [29:53<42:18,  1.52it/s]
2|2607|Loss: 0.795512318611145:  40%|████      | 2606/6470 [29:53<42:18,  1.52it/s] 
2|2607|Loss: 0.795512318611145:  40%|████      | 2607/6470 [29:53<42:32,  1.51it/s]
2|2608|Loss: 0.7606156468391418:  40%|████      | 2607/6470 [29:54<42:32,  1.51it/s]
2|2608|Loss: 0.7606156468391418:  40%|████      | 2608/6470 [29:54<42:13,  1.52it/s]
2|2609|Loss: 0.9309636950492859:  40%|████      | 2608/6470 [29:54<42:13,  1.52it/s]
2|2609|Loss: 0.9309636950492859:  40%|████      | 2609/6470 [29:55<45:01,  1.43it/s]
2|2610|Loss: 0.7994788885116577:  40%|████      | 2609/6470 [29:55<45:01,  1.43it/s]
2|2610|Loss: 0.7994788885116577:  40%|████      | 2610/6470 [29:55<44:20,  1.45it/s]
2|2611|Loss: 0.7888928055763245:  40%|████      | 2610/6470 [29:56<44:20,  1.45it/s]
2|2611|Loss: 0.7888928055763245:  40%|████      | 2611/6470 [29:56<44:37,  1.44it/s]
2|2612|Loss: 1.019046425819397:  40%|████      | 2611/6470 [29:56<44:37,  1.44it/s] 
2|2612|Loss: 1.019046425819397:  40%|████      | 2612/6470 [29:57<43:59,  1.46it/s]
2|2613|Loss: 0.8468726277351379:  40%|████      | 2612/6470 [29:57<43:59,  1.46it/s]
2|2613|Loss: 0.8468726277351379:  40%|████      | 2613/6470 [29:58<44:58,  1.43it/s]
2|2614|Loss: 0.82958984375:  40%|████      | 2613/6470 [29:58<44:58,  1.43it/s]     
2|2614|Loss: 0.82958984375:  40%|████      | 2614/6470 [29:58<43:18,  1.48it/s]
2|2615|Loss: 0.894946813583374:  40%|████      | 2614/6470 [29:58<43:18,  1.48it/s]
2|2615|Loss: 0.894946813583374:  40%|████      | 2615/6470 [29:59<41:59,  1.53it/s]
2|2616|Loss: 0.9407051205635071:  40%|████      | 2615/6470 [29:59<41:59,  1.53it/s]
2|2616|Loss: 0.9407051205635071:  40%|████      | 2616/6470 [30:00<44:52,  1.43it/s]
2|2617|Loss: 0.9419962167739868:  40%|████      | 2616/6470 [30:00<44:52,  1.43it/s]
2|2617|Loss: 0.9419962167739868:  40%|████      | 2617/6470 [30:00<45:44,  1.40it/s]
2|2618|Loss: 0.9562207460403442:  40%|████      | 2617/6470 [30:01<45:44,  1.40it/s]
2|2618|Loss: 0.9562207460403442:  40%|████      | 2618/6470 [30:01<42:35,  1.51it/s]
2|2619|Loss: 0.7731088995933533:  40%|████      | 2618/6470 [30:01<42:35,  1.51it/s]
2|2619|Loss: 0.7731088995933533:  40%|████      | 2619/6470 [30:01<40:36,  1.58it/s]
2|2620|Loss: 0.6786335706710815:  40%|████      | 2619/6470 [30:02<40:36,  1.58it/s]
2|2620|Loss: 0.6786335706710815:  40%|████      | 2620/6470 [30:02<43:49,  1.46it/s]
2|2621|Loss: 0.7740122079849243:  40%|████      | 2620/6470 [30:03<43:49,  1.46it/s]
2|2621|Loss: 0.7740122079849243:  41%|████      | 2621/6470 [30:03<43:53,  1.46it/s]
2|2622|Loss: 0.8323327898979187:  41%|████      | 2621/6470 [30:03<43:53,  1.46it/s]
2|2622|Loss: 0.8323327898979187:  41%|████      | 2622/6470 [30:03<41:04,  1.56it/s]
2|2623|Loss: 0.8292465806007385:  41%|████      | 2622/6470 [30:04<41:04,  1.56it/s]
2|2623|Loss: 0.8292465806007385:  41%|████      | 2623/6470 [30:04<38:53,  1.65it/s]
2|2624|Loss: 0.7964563965797424:  41%|████      | 2623/6470 [30:04<38:53,  1.65it/s]
2|2624|Loss: 0.7964563965797424:  41%|████      | 2624/6470 [30:04<37:07,  1.73it/s]
2|2625|Loss: 0.8379245400428772:  41%|████      | 2624/6470 [30:05<37:07,  1.73it/s]
2|2625|Loss: 0.8379245400428772:  41%|████      | 2625/6470 [30:05<38:39,  1.66it/s]
2|2626|Loss: 0.9285877346992493:  41%|████      | 2625/6470 [30:05<38:39,  1.66it/s]
2|2626|Loss: 0.9285877346992493:  41%|████      | 2626/6470 [30:06<39:24,  1.63it/s]
2|2627|Loss: 1.0264828205108643:  41%|████      | 2626/6470 [30:06<39:24,  1.63it/s]
2|2627|Loss: 1.0264828205108643:  41%|████      | 2627/6470 [30:06<39:24,  1.63it/s]
2|2628|Loss: 0.9590327739715576:  41%|████      | 2627/6470 [30:07<39:24,  1.63it/s]
2|2628|Loss: 0.9590327739715576:  41%|████      | 2628/6470 [30:07<41:12,  1.55it/s]
2|2629|Loss: 0.8144059777259827:  41%|████      | 2628/6470 [30:08<41:12,  1.55it/s]
2|2629|Loss: 0.8144059777259827:  41%|████      | 2629/6470 [30:08<45:06,  1.42it/s]
2|2630|Loss: 0.7588286399841309:  41%|████      | 2629/6470 [30:08<45:06,  1.42it/s]
2|2630|Loss: 0.7588286399841309:  41%|████      | 2630/6470 [30:09<44:32,  1.44it/s]
2|2631|Loss: 0.8824611902236938:  41%|████      | 2630/6470 [30:09<44:32,  1.44it/s]
2|2631|Loss: 0.8824611902236938:  41%|████      | 2631/6470 [30:09<43:05,  1.48it/s]
2|2632|Loss: 0.7904794812202454:  41%|████      | 2631/6470 [30:10<43:05,  1.48it/s]
2|2632|Loss: 0.7904794812202454:  41%|████      | 2632/6470 [30:10<43:23,  1.47it/s]
2|2633|Loss: 0.9032881259918213:  41%|████      | 2632/6470 [30:10<43:23,  1.47it/s]
2|2633|Loss: 0.9032881259918213:  41%|████      | 2633/6470 [30:11<44:04,  1.45it/s]
2|2634|Loss: 0.7318232655525208:  41%|████      | 2633/6470 [30:11<44:04,  1.45it/s]
2|2634|Loss: 0.7318232655525208:  41%|████      | 2634/6470 [30:11<45:39,  1.40it/s]
2|2635|Loss: 0.8400404453277588:  41%|████      | 2634/6470 [30:12<45:39,  1.40it/s]
2|2635|Loss: 0.8400404453277588:  41%|████      | 2635/6470 [30:12<46:29,  1.37it/s]
2|2636|Loss: 0.8884229063987732:  41%|████      | 2635/6470 [30:13<46:29,  1.37it/s]
2|2636|Loss: 0.8884229063987732:  41%|████      | 2636/6470 [30:13<42:52,  1.49it/s]
2|2637|Loss: 0.8931871652603149:  41%|████      | 2636/6470 [30:13<42:52,  1.49it/s]
2|2637|Loss: 0.8931871652603149:  41%|████      | 2637/6470 [30:13<42:33,  1.50it/s]
2|2638|Loss: 0.9771136045455933:  41%|████      | 2637/6470 [30:14<42:33,  1.50it/s]
2|2638|Loss: 0.9771136045455933:  41%|████      | 2638/6470 [30:14<41:31,  1.54it/s]
2|2639|Loss: 0.8672274947166443:  41%|████      | 2638/6470 [30:14<41:31,  1.54it/s]
2|2639|Loss: 0.8672274947166443:  41%|████      | 2639/6470 [30:15<42:06,  1.52it/s]
2|2640|Loss: 0.8756464123725891:  41%|████      | 2639/6470 [30:15<42:06,  1.52it/s]
2|2640|Loss: 0.8756464123725891:  41%|████      | 2640/6470 [30:15<40:33,  1.57it/s]
2|2641|Loss: 0.9140730500221252:  41%|████      | 2640/6470 [30:16<40:33,  1.57it/s]
2|2641|Loss: 0.9140730500221252:  41%|████      | 2641/6470 [30:16<43:19,  1.47it/s]
2|2642|Loss: 0.9836762547492981:  41%|████      | 2641/6470 [30:16<43:19,  1.47it/s]
2|2642|Loss: 0.9836762547492981:  41%|████      | 2642/6470 [30:17<43:24,  1.47it/s]
2|2643|Loss: 0.8692546486854553:  41%|████      | 2642/6470 [30:17<43:24,  1.47it/s]
2|2643|Loss: 0.8692546486854553:  41%|████      | 2643/6470 [30:17<42:21,  1.51it/s]
2|2644|Loss: 0.7311339974403381:  41%|████      | 2643/6470 [30:18<42:21,  1.51it/s]
2|2644|Loss: 0.7311339974403381:  41%|████      | 2644/6470 [30:18<38:58,  1.64it/s]
2|2645|Loss: 0.8143091201782227:  41%|████      | 2644/6470 [30:18<38:58,  1.64it/s]
2|2645|Loss: 0.8143091201782227:  41%|████      | 2645/6470 [30:19<41:11,  1.55it/s]
2|2646|Loss: 0.679661214351654:  41%|████      | 2645/6470 [30:19<41:11,  1.55it/s] 
2|2646|Loss: 0.679661214351654:  41%|████      | 2646/6470 [30:19<41:34,  1.53it/s]
2|2647|Loss: 0.9652188420295715:  41%|████      | 2646/6470 [30:20<41:34,  1.53it/s]
2|2647|Loss: 0.9652188420295715:  41%|████      | 2647/6470 [30:20<41:15,  1.54it/s]
2|2648|Loss: 0.7563086152076721:  41%|████      | 2647/6470 [30:20<41:15,  1.54it/s]
2|2648|Loss: 0.7563086152076721:  41%|████      | 2648/6470 [30:21<42:05,  1.51it/s]
2|2649|Loss: 0.8104780912399292:  41%|████      | 2648/6470 [30:21<42:05,  1.51it/s]
2|2649|Loss: 0.8104780912399292:  41%|████      | 2649/6470 [30:21<43:20,  1.47it/s]
2|2650|Loss: 0.688417375087738:  41%|████      | 2649/6470 [30:22<43:20,  1.47it/s] 
2|2650|Loss: 0.688417375087738:  41%|████      | 2650/6470 [30:22<41:50,  1.52it/s]
2|2651|Loss: 0.7852956056594849:  41%|████      | 2650/6470 [30:22<41:50,  1.52it/s]
2|2651|Loss: 0.7852956056594849:  41%|████      | 2651/6470 [30:22<40:25,  1.57it/s]
2|2652|Loss: 0.807420015335083:  41%|████      | 2651/6470 [30:23<40:25,  1.57it/s] 
2|2652|Loss: 0.807420015335083:  41%|████      | 2652/6470 [30:23<39:51,  1.60it/s]
2|2653|Loss: 0.7074828743934631:  41%|████      | 2652/6470 [30:23<39:51,  1.60it/s]
2|2653|Loss: 0.7074828743934631:  41%|████      | 2653/6470 [30:24<39:42,  1.60it/s]
2|2654|Loss: 0.9167386293411255:  41%|████      | 2653/6470 [30:24<39:42,  1.60it/s]
2|2654|Loss: 0.9167386293411255:  41%|████      | 2654/6470 [30:24<41:43,  1.52it/s]
2|2655|Loss: 0.8797919750213623:  41%|████      | 2654/6470 [30:25<41:43,  1.52it/s]
2|2655|Loss: 0.8797919750213623:  41%|████      | 2655/6470 [30:25<42:01,  1.51it/s]
2|2656|Loss: 0.9210231304168701:  41%|████      | 2655/6470 [30:25<42:01,  1.51it/s]
2|2656|Loss: 0.9210231304168701:  41%|████      | 2656/6470 [30:26<44:27,  1.43it/s]
2|2657|Loss: 0.7683965563774109:  41%|████      | 2656/6470 [30:26<44:27,  1.43it/s]
2|2657|Loss: 0.7683965563774109:  41%|████      | 2657/6470 [30:27<45:39,  1.39it/s]
2|2658|Loss: 0.9975408911705017:  41%|████      | 2657/6470 [30:27<45:39,  1.39it/s]
2|2658|Loss: 0.9975408911705017:  41%|████      | 2658/6470 [30:27<46:10,  1.38it/s]
2|2659|Loss: 0.8442164063453674:  41%|████      | 2658/6470 [30:28<46:10,  1.38it/s]
2|2659|Loss: 0.8442164063453674:  41%|████      | 2659/6470 [30:28<44:37,  1.42it/s]
2|2660|Loss: 0.7269023656845093:  41%|████      | 2659/6470 [30:28<44:37,  1.42it/s]
2|2660|Loss: 0.7269023656845093:  41%|████      | 2660/6470 [30:29<42:58,  1.48it/s]
2|2661|Loss: 0.6065399050712585:  41%|████      | 2660/6470 [30:29<42:58,  1.48it/s]
2|2661|Loss: 0.6065399050712585:  41%|████      | 2661/6470 [30:30<48:30,  1.31it/s]
2|2662|Loss: 0.9043223261833191:  41%|████      | 2661/6470 [30:30<48:30,  1.31it/s]
2|2662|Loss: 0.9043223261833191:  41%|████      | 2662/6470 [30:30<49:07,  1.29it/s]
2|2663|Loss: 0.739137589931488:  41%|████      | 2662/6470 [30:31<49:07,  1.29it/s] 
2|2663|Loss: 0.739137589931488:  41%|████      | 2663/6470 [30:31<48:29,  1.31it/s]
2|2664|Loss: 0.6959038972854614:  41%|████      | 2663/6470 [30:32<48:29,  1.31it/s]
2|2664|Loss: 0.6959038972854614:  41%|████      | 2664/6470 [30:32<49:43,  1.28it/s]
2|2665|Loss: 0.7599279284477234:  41%|████      | 2664/6470 [30:32<49:43,  1.28it/s]
2|2665|Loss: 0.7599279284477234:  41%|████      | 2665/6470 [30:33<49:34,  1.28it/s]
2|2666|Loss: 0.9578205347061157:  41%|████      | 2665/6470 [30:33<49:34,  1.28it/s]
2|2666|Loss: 0.9578205347061157:  41%|████      | 2666/6470 [30:33<45:59,  1.38it/s]
2|2667|Loss: 0.7546912431716919:  41%|████      | 2666/6470 [30:34<45:59,  1.38it/s]
2|2667|Loss: 0.7546912431716919:  41%|████      | 2667/6470 [30:34<42:59,  1.47it/s]
2|2668|Loss: 0.8519908785820007:  41%|████      | 2667/6470 [30:34<42:59,  1.47it/s]
2|2668|Loss: 0.8519908785820007:  41%|████      | 2668/6470 [30:35<41:40,  1.52it/s]
2|2669|Loss: 0.8100705146789551:  41%|████      | 2668/6470 [30:35<41:40,  1.52it/s]
2|2669|Loss: 0.8100705146789551:  41%|████▏     | 2669/6470 [30:35<44:59,  1.41it/s]
2|2670|Loss: 0.7897354960441589:  41%|████▏     | 2669/6470 [30:36<44:59,  1.41it/s]
2|2670|Loss: 0.7897354960441589:  41%|████▏     | 2670/6470 [30:36<46:28,  1.36it/s]
2|2671|Loss: 0.8954209685325623:  41%|████▏     | 2670/6470 [30:37<46:28,  1.36it/s]
2|2671|Loss: 0.8954209685325623:  41%|████▏     | 2671/6470 [30:37<48:18,  1.31it/s]
2|2672|Loss: 0.9568578004837036:  41%|████▏     | 2671/6470 [30:37<48:18,  1.31it/s]
2|2672|Loss: 0.9568578004837036:  41%|████▏     | 2672/6470 [30:38<46:29,  1.36it/s]
2|2673|Loss: 0.9127174019813538:  41%|████▏     | 2672/6470 [30:38<46:29,  1.36it/s]
2|2673|Loss: 0.9127174019813538:  41%|████▏     | 2673/6470 [30:38<46:10,  1.37it/s]
2|2674|Loss: 0.996001124382019:  41%|████▏     | 2673/6470 [30:39<46:10,  1.37it/s] 
2|2674|Loss: 0.996001124382019:  41%|████▏     | 2674/6470 [30:39<40:39,  1.56it/s]
2|2675|Loss: 0.8629006147384644:  41%|████▏     | 2674/6470 [30:39<40:39,  1.56it/s]
2|2675|Loss: 0.8629006147384644:  41%|████▏     | 2675/6470 [30:39<40:29,  1.56it/s]
2|2676|Loss: 0.9157031178474426:  41%|████▏     | 2675/6470 [30:40<40:29,  1.56it/s]
2|2676|Loss: 0.9157031178474426:  41%|████▏     | 2676/6470 [30:40<40:18,  1.57it/s]
2|2677|Loss: 0.8377575278282166:  41%|████▏     | 2676/6470 [30:40<40:18,  1.57it/s]
2|2677|Loss: 0.8377575278282166:  41%|████▏     | 2677/6470 [30:41<37:49,  1.67it/s]
2|2678|Loss: 0.7647051215171814:  41%|████▏     | 2677/6470 [30:41<37:49,  1.67it/s]
2|2678|Loss: 0.7647051215171814:  41%|████▏     | 2678/6470 [30:41<41:26,  1.52it/s]
2|2679|Loss: 0.8856496214866638:  41%|████▏     | 2678/6470 [30:42<41:26,  1.52it/s]
2|2679|Loss: 0.8856496214866638:  41%|████▏     | 2679/6470 [30:42<45:00,  1.40it/s]
2|2680|Loss: 0.8400758504867554:  41%|████▏     | 2679/6470 [30:43<45:00,  1.40it/s]
2|2680|Loss: 0.8400758504867554:  41%|████▏     | 2680/6470 [30:43<47:17,  1.34it/s]
2|2681|Loss: 1.0203797817230225:  41%|████▏     | 2680/6470 [30:43<47:17,  1.34it/s]
2|2681|Loss: 1.0203797817230225:  41%|████▏     | 2681/6470 [30:44<47:50,  1.32it/s]
2|2682|Loss: 0.860537588596344:  41%|████▏     | 2681/6470 [30:44<47:50,  1.32it/s] 
2|2682|Loss: 0.860537588596344:  41%|████▏     | 2682/6470 [30:45<47:12,  1.34it/s]
2|2683|Loss: 0.883659839630127:  41%|████▏     | 2682/6470 [30:45<47:12,  1.34it/s]
2|2683|Loss: 0.883659839630127:  41%|████▏     | 2683/6470 [30:45<44:40,  1.41it/s]
2|2684|Loss: 0.8844465017318726:  41%|████▏     | 2683/6470 [30:46<44:40,  1.41it/s]
2|2684|Loss: 0.8844465017318726:  41%|████▏     | 2684/6470 [30:46<44:30,  1.42it/s]
2|2685|Loss: 0.8192093968391418:  41%|████▏     | 2684/6470 [30:46<44:30,  1.42it/s]
2|2685|Loss: 0.8192093968391418:  41%|████▏     | 2685/6470 [30:46<41:42,  1.51it/s]
2|2686|Loss: 0.8340378999710083:  41%|████▏     | 2685/6470 [30:47<41:42,  1.51it/s]
2|2686|Loss: 0.8340378999710083:  42%|████▏     | 2686/6470 [30:47<39:25,  1.60it/s]
2|2687|Loss: 0.8661385178565979:  42%|████▏     | 2686/6470 [30:47<39:25,  1.60it/s]
2|2687|Loss: 0.8661385178565979:  42%|████▏     | 2687/6470 [30:48<40:11,  1.57it/s]
2|2688|Loss: 0.8409112095832825:  42%|████▏     | 2687/6470 [30:48<40:11,  1.57it/s]
2|2688|Loss: 0.8409112095832825:  42%|████▏     | 2688/6470 [30:48<42:44,  1.47it/s]
2|2689|Loss: 0.9129102826118469:  42%|████▏     | 2688/6470 [30:49<42:44,  1.47it/s]
2|2689|Loss: 0.9129102826118469:  42%|████▏     | 2689/6470 [30:49<44:05,  1.43it/s]
2|2690|Loss: 0.9875965714454651:  42%|████▏     | 2689/6470 [30:50<44:05,  1.43it/s]
2|2690|Loss: 0.9875965714454651:  42%|████▏     | 2690/6470 [30:50<44:54,  1.40it/s]
2|2691|Loss: 0.8161055445671082:  42%|████▏     | 2690/6470 [30:50<44:54,  1.40it/s]
2|2691|Loss: 0.8161055445671082:  42%|████▏     | 2691/6470 [30:51<44:04,  1.43it/s]
2|2692|Loss: 0.9364556670188904:  42%|████▏     | 2691/6470 [30:51<44:04,  1.43it/s]
2|2692|Loss: 0.9364556670188904:  42%|████▏     | 2692/6470 [30:51<45:13,  1.39it/s]
2|2693|Loss: 0.9179325699806213:  42%|████▏     | 2692/6470 [30:52<45:13,  1.39it/s]
2|2693|Loss: 0.9179325699806213:  42%|████▏     | 2693/6470 [30:52<41:38,  1.51it/s]
2|2694|Loss: 0.8094277381896973:  42%|████▏     | 2693/6470 [30:52<41:38,  1.51it/s]
2|2694|Loss: 0.8094277381896973:  42%|████▏     | 2694/6470 [30:52<39:08,  1.61it/s]
2|2695|Loss: 0.8773632645606995:  42%|████▏     | 2694/6470 [30:53<39:08,  1.61it/s]
2|2695|Loss: 0.8773632645606995:  42%|████▏     | 2695/6470 [30:53<40:07,  1.57it/s]
2|2696|Loss: 0.80124431848526:  42%|████▏     | 2695/6470 [30:53<40:07,  1.57it/s]  
2|2696|Loss: 0.80124431848526:  42%|████▏     | 2696/6470 [30:54<41:30,  1.52it/s]
2|2697|Loss: 0.9493572115898132:  42%|████▏     | 2696/6470 [30:54<41:30,  1.52it/s]
2|2697|Loss: 0.9493572115898132:  42%|████▏     | 2697/6470 [30:55<43:10,  1.46it/s]
2|2698|Loss: 0.6173776984214783:  42%|████▏     | 2697/6470 [30:55<43:10,  1.46it/s]
2|2698|Loss: 0.6173776984214783:  42%|████▏     | 2698/6470 [30:55<44:31,  1.41it/s]
2|2699|Loss: 0.7979137897491455:  42%|████▏     | 2698/6470 [30:56<44:31,  1.41it/s]
2|2699|Loss: 0.7979137897491455:  42%|████▏     | 2699/6470 [30:56<45:09,  1.39it/s]
2|2700|Loss: 0.7410613894462585:  42%|████▏     | 2699/6470 [30:56<45:09,  1.39it/s]
2|2700|Loss: 0.7410613894462585:  42%|████▏     | 2700/6470 [30:57<44:06,  1.42it/s]
2|2701|Loss: 0.9900527000427246:  42%|████▏     | 2700/6470 [30:57<44:06,  1.42it/s]
2|2701|Loss: 0.9900527000427246:  42%|████▏     | 2701/6470 [30:57<42:56,  1.46it/s]
2|2702|Loss: 0.8588075637817383:  42%|████▏     | 2701/6470 [30:58<42:56,  1.46it/s]
2|2702|Loss: 0.8588075637817383:  42%|████▏     | 2702/6470 [30:58<43:03,  1.46it/s]
2|2703|Loss: 0.8090782761573792:  42%|████▏     | 2702/6470 [30:58<43:03,  1.46it/s]
2|2703|Loss: 0.8090782761573792:  42%|████▏     | 2703/6470 [30:58<38:15,  1.64it/s]
2|2704|Loss: 0.6850695610046387:  42%|████▏     | 2703/6470 [30:59<38:15,  1.64it/s]
2|2704|Loss: 0.6850695610046387:  42%|████▏     | 2704/6470 [30:59<44:16,  1.42it/s]
2|2705|Loss: 0.7824339866638184:  42%|████▏     | 2704/6470 [31:00<44:16,  1.42it/s]
2|2705|Loss: 0.7824339866638184:  42%|████▏     | 2705/6470 [31:00<45:41,  1.37it/s]
2|2706|Loss: 0.8936958312988281:  42%|████▏     | 2705/6470 [31:00<45:41,  1.37it/s]
2|2706|Loss: 0.8936958312988281:  42%|████▏     | 2706/6470 [31:01<42:32,  1.47it/s]
2|2707|Loss: 0.8737044334411621:  42%|████▏     | 2706/6470 [31:01<42:32,  1.47it/s]
2|2707|Loss: 0.8737044334411621:  42%|████▏     | 2707/6470 [31:01<44:21,  1.41it/s]
2|2708|Loss: 0.9015995264053345:  42%|████▏     | 2707/6470 [31:02<44:21,  1.41it/s]
2|2708|Loss: 0.9015995264053345:  42%|████▏     | 2708/6470 [31:02<47:04,  1.33it/s]
2|2709|Loss: 0.9238976240158081:  42%|████▏     | 2708/6470 [31:03<47:04,  1.33it/s]
2|2709|Loss: 0.9238976240158081:  42%|████▏     | 2709/6470 [31:03<45:48,  1.37it/s]
2|2710|Loss: 1.042359471321106:  42%|████▏     | 2709/6470 [31:03<45:48,  1.37it/s] 
2|2710|Loss: 1.042359471321106:  42%|████▏     | 2710/6470 [31:04<42:03,  1.49it/s]
2|2711|Loss: 1.114802360534668:  42%|████▏     | 2710/6470 [31:04<42:03,  1.49it/s]
2|2711|Loss: 1.114802360534668:  42%|████▏     | 2711/6470 [31:04<41:55,  1.49it/s]
2|2712|Loss: 0.9202682971954346:  42%|████▏     | 2711/6470 [31:05<41:55,  1.49it/s]
2|2712|Loss: 0.9202682971954346:  42%|████▏     | 2712/6470 [31:05<40:42,  1.54it/s]
2|2713|Loss: 0.7816578149795532:  42%|████▏     | 2712/6470 [31:05<40:42,  1.54it/s]
2|2713|Loss: 0.7816578149795532:  42%|████▏     | 2713/6470 [31:05<40:19,  1.55it/s]
2|2714|Loss: 0.8187939524650574:  42%|████▏     | 2713/6470 [31:06<40:19,  1.55it/s]
2|2714|Loss: 0.8187939524650574:  42%|████▏     | 2714/6470 [31:06<41:14,  1.52it/s]
2|2715|Loss: 0.9759564995765686:  42%|████▏     | 2714/6470 [31:07<41:14,  1.52it/s]
2|2715|Loss: 0.9759564995765686:  42%|████▏     | 2715/6470 [31:07<42:30,  1.47it/s]
2|2716|Loss: 0.7481339573860168:  42%|████▏     | 2715/6470 [31:07<42:30,  1.47it/s]
2|2716|Loss: 0.7481339573860168:  42%|████▏     | 2716/6470 [31:08<45:19,  1.38it/s]
2|2717|Loss: 0.9777097702026367:  42%|████▏     | 2716/6470 [31:08<45:19,  1.38it/s]
2|2717|Loss: 0.9777097702026367:  42%|████▏     | 2717/6470 [31:08<43:00,  1.45it/s]
2|2718|Loss: 0.8927099108695984:  42%|████▏     | 2717/6470 [31:09<43:00,  1.45it/s]
2|2718|Loss: 0.8927099108695984:  42%|████▏     | 2718/6470 [31:09<43:16,  1.45it/s]
2|2719|Loss: 0.8402758240699768:  42%|████▏     | 2718/6470 [31:09<43:16,  1.45it/s]
2|2719|Loss: 0.8402758240699768:  42%|████▏     | 2719/6470 [31:10<42:04,  1.49it/s]
2|2720|Loss: 0.8758440613746643:  42%|████▏     | 2719/6470 [31:10<42:04,  1.49it/s]
2|2720|Loss: 0.8758440613746643:  42%|████▏     | 2720/6470 [31:10<39:57,  1.56it/s]
2|2721|Loss: 0.8620761036872864:  42%|████▏     | 2720/6470 [31:11<39:57,  1.56it/s]
2|2721|Loss: 0.8620761036872864:  42%|████▏     | 2721/6470 [31:11<41:11,  1.52it/s]
2|2722|Loss: 0.7642415165901184:  42%|████▏     | 2721/6470 [31:11<41:11,  1.52it/s]
2|2722|Loss: 0.7642415165901184:  42%|████▏     | 2722/6470 [31:12<43:34,  1.43it/s]
2|2723|Loss: 0.8029152154922485:  42%|████▏     | 2722/6470 [31:12<43:34,  1.43it/s]
2|2723|Loss: 0.8029152154922485:  42%|████▏     | 2723/6470 [31:12<44:50,  1.39it/s]
2|2724|Loss: 0.9886891841888428:  42%|████▏     | 2723/6470 [31:13<44:50,  1.39it/s]
2|2724|Loss: 0.9886891841888428:  42%|████▏     | 2724/6470 [31:13<44:45,  1.39it/s]
2|2725|Loss: 0.9153426289558411:  42%|████▏     | 2724/6470 [31:14<44:45,  1.39it/s]
2|2725|Loss: 0.9153426289558411:  42%|████▏     | 2725/6470 [31:14<42:29,  1.47it/s]
2|2726|Loss: 0.7562206983566284:  42%|████▏     | 2725/6470 [31:14<42:29,  1.47it/s]
2|2726|Loss: 0.7562206983566284:  42%|████▏     | 2726/6470 [31:14<42:53,  1.45it/s]
2|2727|Loss: 0.870464026927948:  42%|████▏     | 2726/6470 [31:15<42:53,  1.45it/s] 
2|2727|Loss: 0.870464026927948:  42%|████▏     | 2727/6470 [31:15<41:57,  1.49it/s]
2|2728|Loss: 0.9745247960090637:  42%|████▏     | 2727/6470 [31:15<41:57,  1.49it/s]
2|2728|Loss: 0.9745247960090637:  42%|████▏     | 2728/6470 [31:16<39:13,  1.59it/s]
2|2729|Loss: 0.6752168536186218:  42%|████▏     | 2728/6470 [31:16<39:13,  1.59it/s]
2|2729|Loss: 0.6752168536186218:  42%|████▏     | 2729/6470 [31:16<39:50,  1.57it/s]
2|2730|Loss: 0.7690509557723999:  42%|████▏     | 2729/6470 [31:17<39:50,  1.57it/s]
2|2730|Loss: 0.7690509557723999:  42%|████▏     | 2730/6470 [31:17<40:56,  1.52it/s]
2|2731|Loss: 0.9917240738868713:  42%|████▏     | 2730/6470 [31:17<40:56,  1.52it/s]
2|2731|Loss: 0.9917240738868713:  42%|████▏     | 2731/6470 [31:18<43:17,  1.44it/s]
2|2732|Loss: 0.7203966975212097:  42%|████▏     | 2731/6470 [31:18<43:17,  1.44it/s]
2|2732|Loss: 0.7203966975212097:  42%|████▏     | 2732/6470 [31:19<45:13,  1.38it/s]
2|2733|Loss: 0.8314230442047119:  42%|████▏     | 2732/6470 [31:19<45:13,  1.38it/s]
2|2733|Loss: 0.8314230442047119:  42%|████▏     | 2733/6470 [31:19<45:07,  1.38it/s]
2|2734|Loss: 0.719814658164978:  42%|████▏     | 2733/6470 [31:20<45:07,  1.38it/s] 
2|2734|Loss: 0.719814658164978:  42%|████▏     | 2734/6470 [31:20<48:32,  1.28it/s]
2|2735|Loss: 0.8263677358627319:  42%|████▏     | 2734/6470 [31:21<48:32,  1.28it/s]
2|2735|Loss: 0.8263677358627319:  42%|████▏     | 2735/6470 [31:21<47:11,  1.32it/s]
2|2736|Loss: 0.8109530806541443:  42%|████▏     | 2735/6470 [31:21<47:11,  1.32it/s]
2|2736|Loss: 0.8109530806541443:  42%|████▏     | 2736/6470 [31:22<47:30,  1.31it/s]
2|2737|Loss: 0.8048266172409058:  42%|████▏     | 2736/6470 [31:22<47:30,  1.31it/s]
2|2737|Loss: 0.8048266172409058:  42%|████▏     | 2737/6470 [31:22<45:12,  1.38it/s]
2|2738|Loss: 0.8421556949615479:  42%|████▏     | 2737/6470 [31:23<45:12,  1.38it/s]
2|2738|Loss: 0.8421556949615479:  42%|████▏     | 2738/6470 [31:23<44:27,  1.40it/s]
2|2739|Loss: 0.896637499332428:  42%|████▏     | 2738/6470 [31:23<44:27,  1.40it/s] 
2|2739|Loss: 0.896637499332428:  42%|████▏     | 2739/6470 [31:24<45:02,  1.38it/s]
2|2740|Loss: 0.8976954221725464:  42%|████▏     | 2739/6470 [31:24<45:02,  1.38it/s]
2|2740|Loss: 0.8976954221725464:  42%|████▏     | 2740/6470 [31:25<45:23,  1.37it/s]
2|2741|Loss: 0.8690681457519531:  42%|████▏     | 2740/6470 [31:25<45:23,  1.37it/s]
2|2741|Loss: 0.8690681457519531:  42%|████▏     | 2741/6470 [31:25<43:52,  1.42it/s]
2|2742|Loss: 0.8011562824249268:  42%|████▏     | 2741/6470 [31:25<43:52,  1.42it/s]
2|2742|Loss: 0.8011562824249268:  42%|████▏     | 2742/6470 [31:26<41:14,  1.51it/s]
2|2743|Loss: 0.7514299154281616:  42%|████▏     | 2742/6470 [31:26<41:14,  1.51it/s]
2|2743|Loss: 0.7514299154281616:  42%|████▏     | 2743/6470 [31:26<41:21,  1.50it/s]
2|2744|Loss: 0.8687773942947388:  42%|████▏     | 2743/6470 [31:27<41:21,  1.50it/s]
2|2744|Loss: 0.8687773942947388:  42%|████▏     | 2744/6470 [31:27<40:53,  1.52it/s]
2|2745|Loss: 0.8839669823646545:  42%|████▏     | 2744/6470 [31:27<40:53,  1.52it/s]
2|2745|Loss: 0.8839669823646545:  42%|████▏     | 2745/6470 [31:28<38:47,  1.60it/s]
2|2746|Loss: 0.8395370244979858:  42%|████▏     | 2745/6470 [31:28<38:47,  1.60it/s]
2|2746|Loss: 0.8395370244979858:  42%|████▏     | 2746/6470 [31:28<39:56,  1.55it/s]
2|2747|Loss: 0.8454213738441467:  42%|████▏     | 2746/6470 [31:29<39:56,  1.55it/s]
2|2747|Loss: 0.8454213738441467:  42%|████▏     | 2747/6470 [31:29<40:21,  1.54it/s]
2|2748|Loss: 0.7658795714378357:  42%|████▏     | 2747/6470 [31:29<40:21,  1.54it/s]
2|2748|Loss: 0.7658795714378357:  42%|████▏     | 2748/6470 [31:30<43:48,  1.42it/s]
2|2749|Loss: 0.8468961119651794:  42%|████▏     | 2748/6470 [31:30<43:48,  1.42it/s]
2|2749|Loss: 0.8468961119651794:  42%|████▏     | 2749/6470 [31:31<47:07,  1.32it/s]
2|2750|Loss: 1.0057342052459717:  42%|████▏     | 2749/6470 [31:31<47:07,  1.32it/s]
2|2750|Loss: 1.0057342052459717:  43%|████▎     | 2750/6470 [31:32<48:45,  1.27it/s]
2|2751|Loss: 0.795428991317749:  43%|████▎     | 2750/6470 [31:32<48:45,  1.27it/s] 
2|2751|Loss: 0.795428991317749:  43%|████▎     | 2751/6470 [31:32<46:30,  1.33it/s]
2|2752|Loss: 0.8008862733840942:  43%|████▎     | 2751/6470 [31:33<46:30,  1.33it/s]
2|2752|Loss: 0.8008862733840942:  43%|████▎     | 2752/6470 [31:33<45:59,  1.35it/s]
2|2753|Loss: 0.9595803022384644:  43%|████▎     | 2752/6470 [31:33<45:59,  1.35it/s]
2|2753|Loss: 0.9595803022384644:  43%|████▎     | 2753/6470 [31:34<45:52,  1.35it/s]
2|2754|Loss: 0.9072615504264832:  43%|████▎     | 2753/6470 [31:34<45:52,  1.35it/s]
2|2754|Loss: 0.9072615504264832:  43%|████▎     | 2754/6470 [31:34<42:25,  1.46it/s]
2|2755|Loss: 0.8551982045173645:  43%|████▎     | 2754/6470 [31:35<42:25,  1.46it/s]
2|2755|Loss: 0.8551982045173645:  43%|████▎     | 2755/6470 [31:35<39:48,  1.56it/s]
2|2756|Loss: 0.9174212217330933:  43%|████▎     | 2755/6470 [31:35<39:48,  1.56it/s]
2|2756|Loss: 0.9174212217330933:  43%|████▎     | 2756/6470 [31:35<40:49,  1.52it/s]
2|2757|Loss: 0.9704943299293518:  43%|████▎     | 2756/6470 [31:36<40:49,  1.52it/s]
2|2757|Loss: 0.9704943299293518:  43%|████▎     | 2757/6470 [31:36<41:45,  1.48it/s]
2|2758|Loss: 0.8838840126991272:  43%|████▎     | 2757/6470 [31:37<41:45,  1.48it/s]
2|2758|Loss: 0.8838840126991272:  43%|████▎     | 2758/6470 [31:37<42:15,  1.46it/s]
2|2759|Loss: 0.9153897166252136:  43%|████▎     | 2758/6470 [31:37<42:15,  1.46it/s]
2|2759|Loss: 0.9153897166252136:  43%|████▎     | 2759/6470 [31:38<44:46,  1.38it/s]
2|2760|Loss: 0.8098896741867065:  43%|████▎     | 2759/6470 [31:38<44:46,  1.38it/s]
2|2760|Loss: 0.8098896741867065:  43%|████▎     | 2760/6470 [31:39<47:13,  1.31it/s]
2|2761|Loss: 0.7895625829696655:  43%|████▎     | 2760/6470 [31:39<47:13,  1.31it/s]
2|2761|Loss: 0.7895625829696655:  43%|████▎     | 2761/6470 [31:39<45:53,  1.35it/s]
2|2762|Loss: 0.8522152900695801:  43%|████▎     | 2761/6470 [31:40<45:53,  1.35it/s]
2|2762|Loss: 0.8522152900695801:  43%|████▎     | 2762/6470 [31:40<41:37,  1.48it/s]
2|2763|Loss: 0.918789803981781:  43%|████▎     | 2762/6470 [31:40<41:37,  1.48it/s] 
2|2763|Loss: 0.918789803981781:  43%|████▎     | 2763/6470 [31:40<39:59,  1.55it/s]
2|2764|Loss: 0.9380853772163391:  43%|████▎     | 2763/6470 [31:41<39:59,  1.55it/s]
2|2764|Loss: 0.9380853772163391:  43%|████▎     | 2764/6470 [31:41<37:04,  1.67it/s]
2|2765|Loss: 0.8269053101539612:  43%|████▎     | 2764/6470 [31:41<37:04,  1.67it/s]
2|2765|Loss: 0.8269053101539612:  43%|████▎     | 2765/6470 [31:41<38:38,  1.60it/s]
2|2766|Loss: 0.9628351330757141:  43%|████▎     | 2765/6470 [31:42<38:38,  1.60it/s]
2|2766|Loss: 0.9628351330757141:  43%|████▎     | 2766/6470 [31:42<40:20,  1.53it/s]
2|2767|Loss: 0.9482219815254211:  43%|████▎     | 2766/6470 [31:43<40:20,  1.53it/s]
2|2767|Loss: 0.9482219815254211:  43%|████▎     | 2767/6470 [31:43<40:27,  1.53it/s]
2|2768|Loss: 0.8862190842628479:  43%|████▎     | 2767/6470 [31:43<40:27,  1.53it/s]
2|2768|Loss: 0.8862190842628479:  43%|████▎     | 2768/6470 [31:44<42:37,  1.45it/s]
2|2769|Loss: 0.7474366426467896:  43%|████▎     | 2768/6470 [31:44<42:37,  1.45it/s]
2|2769|Loss: 0.7474366426467896:  43%|████▎     | 2769/6470 [31:44<44:43,  1.38it/s]
2|2770|Loss: 0.9497978687286377:  43%|████▎     | 2769/6470 [31:45<44:43,  1.38it/s]
2|2770|Loss: 0.9497978687286377:  43%|████▎     | 2770/6470 [31:45<44:41,  1.38it/s]
2|2771|Loss: 1.0622649192810059:  43%|████▎     | 2770/6470 [31:46<44:41,  1.38it/s]
2|2771|Loss: 1.0622649192810059:  43%|████▎     | 2771/6470 [31:46<45:45,  1.35it/s]
2|2772|Loss: 0.9425604343414307:  43%|████▎     | 2771/6470 [31:46<45:45,  1.35it/s]
2|2772|Loss: 0.9425604343414307:  43%|████▎     | 2772/6470 [31:47<47:40,  1.29it/s]
2|2773|Loss: 0.9759806394577026:  43%|████▎     | 2772/6470 [31:47<47:40,  1.29it/s]
2|2773|Loss: 0.9759806394577026:  43%|████▎     | 2773/6470 [31:47<44:13,  1.39it/s]
2|2774|Loss: 0.8452473878860474:  43%|████▎     | 2773/6470 [31:48<44:13,  1.39it/s]
2|2774|Loss: 0.8452473878860474:  43%|████▎     | 2774/6470 [31:48<43:57,  1.40it/s]
2|2775|Loss: 0.8861073851585388:  43%|████▎     | 2774/6470 [31:48<43:57,  1.40it/s]
2|2775|Loss: 0.8861073851585388:  43%|████▎     | 2775/6470 [31:49<39:05,  1.58it/s]
2|2776|Loss: 0.6823804974555969:  43%|████▎     | 2775/6470 [31:49<39:05,  1.58it/s]
2|2776|Loss: 0.6823804974555969:  43%|████▎     | 2776/6470 [31:49<39:39,  1.55it/s]
2|2777|Loss: 0.8239275217056274:  43%|████▎     | 2776/6470 [31:50<39:39,  1.55it/s]
2|2777|Loss: 0.8239275217056274:  43%|████▎     | 2777/6470 [31:50<39:46,  1.55it/s]
2|2778|Loss: 0.9176074266433716:  43%|████▎     | 2777/6470 [31:50<39:46,  1.55it/s]
2|2778|Loss: 0.9176074266433716:  43%|████▎     | 2778/6470 [31:50<38:14,  1.61it/s]
2|2779|Loss: 0.887209415435791:  43%|████▎     | 2778/6470 [31:51<38:14,  1.61it/s] 
2|2779|Loss: 0.887209415435791:  43%|████▎     | 2779/6470 [31:51<40:33,  1.52it/s]
2|2780|Loss: 0.8626471757888794:  43%|████▎     | 2779/6470 [31:52<40:33,  1.52it/s]
2|2780|Loss: 0.8626471757888794:  43%|████▎     | 2780/6470 [31:52<42:19,  1.45it/s]
2|2781|Loss: 0.9337574243545532:  43%|████▎     | 2780/6470 [31:52<42:19,  1.45it/s]
2|2781|Loss: 0.9337574243545532:  43%|████▎     | 2781/6470 [31:53<43:30,  1.41it/s]
2|2782|Loss: 0.8235809803009033:  43%|████▎     | 2781/6470 [31:53<43:30,  1.41it/s]
2|2782|Loss: 0.8235809803009033:  43%|████▎     | 2782/6470 [31:53<44:19,  1.39it/s]
2|2783|Loss: 0.8839402198791504:  43%|████▎     | 2782/6470 [31:54<44:19,  1.39it/s]
2|2783|Loss: 0.8839402198791504:  43%|████▎     | 2783/6470 [31:54<43:21,  1.42it/s]
2|2784|Loss: 0.8565916419029236:  43%|████▎     | 2783/6470 [31:54<43:21,  1.42it/s]
2|2784|Loss: 0.8565916419029236:  43%|████▎     | 2784/6470 [31:55<43:12,  1.42it/s]
2|2785|Loss: 0.8601009845733643:  43%|████▎     | 2784/6470 [31:55<43:12,  1.42it/s]
2|2785|Loss: 0.8601009845733643:  43%|████▎     | 2785/6470 [31:56<45:23,  1.35it/s]
2|2786|Loss: 0.7630084753036499:  43%|████▎     | 2785/6470 [31:56<45:23,  1.35it/s]
2|2786|Loss: 0.7630084753036499:  43%|████▎     | 2786/6470 [31:56<47:09,  1.30it/s]
2|2787|Loss: 0.8355291485786438:  43%|████▎     | 2786/6470 [31:57<47:09,  1.30it/s]
2|2787|Loss: 0.8355291485786438:  43%|████▎     | 2787/6470 [31:57<44:08,  1.39it/s]
2|2788|Loss: 0.839866042137146:  43%|████▎     | 2787/6470 [31:57<44:08,  1.39it/s] 
2|2788|Loss: 0.839866042137146:  43%|████▎     | 2788/6470 [31:58<43:40,  1.40it/s]
2|2789|Loss: 0.8304453492164612:  43%|████▎     | 2788/6470 [31:58<43:40,  1.40it/s]
2|2789|Loss: 0.8304453492164612:  43%|████▎     | 2789/6470 [31:58<44:06,  1.39it/s]
2|2790|Loss: 0.8438097238540649:  43%|████▎     | 2789/6470 [31:59<44:06,  1.39it/s]
2|2790|Loss: 0.8438097238540649:  43%|████▎     | 2790/6470 [31:59<43:54,  1.40it/s]
2|2791|Loss: 0.9448453187942505:  43%|████▎     | 2790/6470 [32:00<43:54,  1.40it/s]
2|2791|Loss: 0.9448453187942505:  43%|████▎     | 2791/6470 [32:00<46:08,  1.33it/s]
2|2792|Loss: 0.9227920770645142:  43%|████▎     | 2791/6470 [32:00<46:08,  1.33it/s]
2|2792|Loss: 0.9227920770645142:  43%|████▎     | 2792/6470 [32:01<45:53,  1.34it/s]
2|2793|Loss: 0.9626491069793701:  43%|████▎     | 2792/6470 [32:01<45:53,  1.34it/s]
2|2793|Loss: 0.9626491069793701:  43%|████▎     | 2793/6470 [32:01<45:02,  1.36it/s]
2|2794|Loss: 0.9039453864097595:  43%|████▎     | 2793/6470 [32:02<45:02,  1.36it/s]
2|2794|Loss: 0.9039453864097595:  43%|████▎     | 2794/6470 [32:02<45:13,  1.35it/s]
2|2795|Loss: 0.847495973110199:  43%|████▎     | 2794/6470 [32:03<45:13,  1.35it/s] 
2|2795|Loss: 0.847495973110199:  43%|████▎     | 2795/6470 [32:03<46:08,  1.33it/s]
2|2796|Loss: 0.8248453140258789:  43%|████▎     | 2795/6470 [32:03<46:08,  1.33it/s]
2|2796|Loss: 0.8248453140258789:  43%|████▎     | 2796/6470 [32:04<44:44,  1.37it/s]
2|2797|Loss: 0.716715395450592:  43%|████▎     | 2796/6470 [32:04<44:44,  1.37it/s] 
2|2797|Loss: 0.716715395450592:  43%|████▎     | 2797/6470 [32:04<42:19,  1.45it/s]
2|2798|Loss: 0.8478788733482361:  43%|████▎     | 2797/6470 [32:05<42:19,  1.45it/s]
2|2798|Loss: 0.8478788733482361:  43%|████▎     | 2798/6470 [32:05<41:12,  1.48it/s]
2|2799|Loss: 0.7336636781692505:  43%|████▎     | 2798/6470 [32:05<41:12,  1.48it/s]
2|2799|Loss: 0.7336636781692505:  43%|████▎     | 2799/6470 [32:06<40:41,  1.50it/s]
2|2800|Loss: 0.8277064561843872:  43%|████▎     | 2799/6470 [32:06<40:41,  1.50it/s]
2|2800|Loss: 0.8277064561843872:  43%|████▎     | 2800/6470 [32:06<42:59,  1.42it/s]
2|2801|Loss: 0.7228889465332031:  43%|████▎     | 2800/6470 [32:07<42:59,  1.42it/s]
2|2801|Loss: 0.7228889465332031:  43%|████▎     | 2801/6470 [32:07<39:46,  1.54it/s]
2|2802|Loss: 1.0084712505340576:  43%|████▎     | 2801/6470 [32:07<39:46,  1.54it/s]
2|2802|Loss: 1.0084712505340576:  43%|████▎     | 2802/6470 [32:08<41:35,  1.47it/s]
2|2803|Loss: 0.8427248001098633:  43%|████▎     | 2802/6470 [32:08<41:35,  1.47it/s]
2|2803|Loss: 0.8427248001098633:  43%|████▎     | 2803/6470 [32:08<44:43,  1.37it/s]
2|2804|Loss: 0.8971025943756104:  43%|████▎     | 2803/6470 [32:09<44:43,  1.37it/s]
2|2804|Loss: 0.8971025943756104:  43%|████▎     | 2804/6470 [32:09<42:57,  1.42it/s]
2|2805|Loss: 0.7139174342155457:  43%|████▎     | 2804/6470 [32:09<42:57,  1.42it/s]
2|2805|Loss: 0.7139174342155457:  43%|████▎     | 2805/6470 [32:10<40:39,  1.50it/s]
2|2806|Loss: 0.8998890519142151:  43%|████▎     | 2805/6470 [32:10<40:39,  1.50it/s]
2|2806|Loss: 0.8998890519142151:  43%|████▎     | 2806/6470 [32:10<40:50,  1.50it/s]
2|2807|Loss: 0.9179311990737915:  43%|████▎     | 2806/6470 [32:11<40:50,  1.50it/s]
2|2807|Loss: 0.9179311990737915:  43%|████▎     | 2807/6470 [32:11<40:34,  1.50it/s]
2|2808|Loss: 0.796028196811676:  43%|████▎     | 2807/6470 [32:11<40:34,  1.50it/s] 
2|2808|Loss: 0.796028196811676:  43%|████▎     | 2808/6470 [32:12<39:41,  1.54it/s]
2|2809|Loss: 0.7742137908935547:  43%|████▎     | 2808/6470 [32:12<39:41,  1.54it/s]
2|2809|Loss: 0.7742137908935547:  43%|████▎     | 2809/6470 [32:12<42:03,  1.45it/s]
2|2810|Loss: 0.9195334911346436:  43%|████▎     | 2809/6470 [32:13<42:03,  1.45it/s]
2|2810|Loss: 0.9195334911346436:  43%|████▎     | 2810/6470 [32:13<39:08,  1.56it/s]
2|2811|Loss: 0.8384900093078613:  43%|████▎     | 2810/6470 [32:13<39:08,  1.56it/s]
2|2811|Loss: 0.8384900093078613:  43%|████▎     | 2811/6470 [32:14<39:52,  1.53it/s]
2|2812|Loss: 0.8325054049491882:  43%|████▎     | 2811/6470 [32:14<39:52,  1.53it/s]
2|2812|Loss: 0.8325054049491882:  43%|████▎     | 2812/6470 [32:15<44:11,  1.38it/s]
2|2813|Loss: 0.856453537940979:  43%|████▎     | 2812/6470 [32:15<44:11,  1.38it/s] 
2|2813|Loss: 0.856453537940979:  43%|████▎     | 2813/6470 [32:15<41:27,  1.47it/s]
2|2814|Loss: 1.0644395351409912:  43%|████▎     | 2813/6470 [32:15<41:27,  1.47it/s]
2|2814|Loss: 1.0644395351409912:  43%|████▎     | 2814/6470 [32:16<41:37,  1.46it/s]
2|2815|Loss: 0.8524425029754639:  43%|████▎     | 2814/6470 [32:16<41:37,  1.46it/s]
2|2815|Loss: 0.8524425029754639:  44%|████▎     | 2815/6470 [32:17<42:40,  1.43it/s]
2|2816|Loss: 0.805576503276825:  44%|████▎     | 2815/6470 [32:17<42:40,  1.43it/s] 
2|2816|Loss: 0.805576503276825:  44%|████▎     | 2816/6470 [32:17<42:02,  1.45it/s]
2|2817|Loss: 0.8081802129745483:  44%|████▎     | 2816/6470 [32:18<42:02,  1.45it/s]
2|2817|Loss: 0.8081802129745483:  44%|████▎     | 2817/6470 [32:18<39:43,  1.53it/s]
2|2818|Loss: 0.6926641464233398:  44%|████▎     | 2817/6470 [32:18<39:43,  1.53it/s]
2|2818|Loss: 0.6926641464233398:  44%|████▎     | 2818/6470 [32:18<40:31,  1.50it/s]
2|2819|Loss: 0.8886572122573853:  44%|████▎     | 2818/6470 [32:19<40:31,  1.50it/s]
2|2819|Loss: 0.8886572122573853:  44%|████▎     | 2819/6470 [32:19<42:38,  1.43it/s]
2|2820|Loss: 0.7808999419212341:  44%|████▎     | 2819/6470 [32:20<42:38,  1.43it/s]
2|2820|Loss: 0.7808999419212341:  44%|████▎     | 2820/6470 [32:20<42:15,  1.44it/s]
2|2821|Loss: 1.0322445631027222:  44%|████▎     | 2820/6470 [32:20<42:15,  1.44it/s]
2|2821|Loss: 1.0322445631027222:  44%|████▎     | 2821/6470 [32:21<42:46,  1.42it/s]
2|2822|Loss: 0.758018970489502:  44%|████▎     | 2821/6470 [32:21<42:46,  1.42it/s] 
2|2822|Loss: 0.758018970489502:  44%|████▎     | 2822/6470 [32:21<41:51,  1.45it/s]
2|2823|Loss: 0.9074558615684509:  44%|████▎     | 2822/6470 [32:22<41:51,  1.45it/s]
2|2823|Loss: 0.9074558615684509:  44%|████▎     | 2823/6470 [32:22<44:00,  1.38it/s]
2|2824|Loss: 1.0262718200683594:  44%|████▎     | 2823/6470 [32:23<44:00,  1.38it/s]
2|2824|Loss: 1.0262718200683594:  44%|████▎     | 2824/6470 [32:23<44:18,  1.37it/s]
2|2825|Loss: 0.9162677526473999:  44%|████▎     | 2824/6470 [32:23<44:18,  1.37it/s]
2|2825|Loss: 0.9162677526473999:  44%|████▎     | 2825/6470 [32:24<44:15,  1.37it/s]
2|2826|Loss: 0.9972465634346008:  44%|████▎     | 2825/6470 [32:24<44:15,  1.37it/s]
2|2826|Loss: 0.9972465634346008:  44%|████▎     | 2826/6470 [32:24<45:13,  1.34it/s]
2|2827|Loss: 0.8493372797966003:  44%|████▎     | 2826/6470 [32:25<45:13,  1.34it/s]
2|2827|Loss: 0.8493372797966003:  44%|████▎     | 2827/6470 [32:25<46:19,  1.31it/s]
2|2828|Loss: 0.8897908926010132:  44%|████▎     | 2827/6470 [32:26<46:19,  1.31it/s]
2|2828|Loss: 0.8897908926010132:  44%|████▎     | 2828/6470 [32:26<44:46,  1.36it/s]
2|2829|Loss: 0.9044455289840698:  44%|████▎     | 2828/6470 [32:26<44:46,  1.36it/s]
2|2829|Loss: 0.9044455289840698:  44%|████▎     | 2829/6470 [32:27<45:14,  1.34it/s]
2|2830|Loss: 0.8353201746940613:  44%|████▎     | 2829/6470 [32:27<45:14,  1.34it/s]
2|2830|Loss: 0.8353201746940613:  44%|████▎     | 2830/6470 [32:27<43:55,  1.38it/s]
2|2831|Loss: 0.6386452913284302:  44%|████▎     | 2830/6470 [32:28<43:55,  1.38it/s]
2|2831|Loss: 0.6386452913284302:  44%|████▍     | 2831/6470 [32:28<43:11,  1.40it/s]
2|2832|Loss: 0.7897910475730896:  44%|████▍     | 2831/6470 [32:28<43:11,  1.40it/s]
2|2832|Loss: 0.7897910475730896:  44%|████▍     | 2832/6470 [32:29<45:10,  1.34it/s]
2|2833|Loss: 0.8356364369392395:  44%|████▍     | 2832/6470 [32:29<45:10,  1.34it/s]
2|2833|Loss: 0.8356364369392395:  44%|████▍     | 2833/6470 [32:29<42:36,  1.42it/s]
2|2834|Loss: 0.7185691595077515:  44%|████▍     | 2833/6470 [32:30<42:36,  1.42it/s]
2|2834|Loss: 0.7185691595077515:  44%|████▍     | 2834/6470 [32:30<40:36,  1.49it/s]
2|2835|Loss: 0.895556628704071:  44%|████▍     | 2834/6470 [32:30<40:36,  1.49it/s] 
2|2835|Loss: 0.895556628704071:  44%|████▍     | 2835/6470 [32:31<38:55,  1.56it/s]
2|2836|Loss: 0.7895275950431824:  44%|████▍     | 2835/6470 [32:31<38:55,  1.56it/s]
2|2836|Loss: 0.7895275950431824:  44%|████▍     | 2836/6470 [32:31<34:48,  1.74it/s]
2|2837|Loss: 0.8949095010757446:  44%|████▍     | 2836/6470 [32:31<34:48,  1.74it/s]
2|2837|Loss: 0.8949095010757446:  44%|████▍     | 2837/6470 [32:32<34:35,  1.75it/s]
2|2838|Loss: 0.9409350156784058:  44%|████▍     | 2837/6470 [32:32<34:35,  1.75it/s]
2|2838|Loss: 0.9409350156784058:  44%|████▍     | 2838/6470 [32:32<35:24,  1.71it/s]
2|2839|Loss: 0.8551104068756104:  44%|████▍     | 2838/6470 [32:33<35:24,  1.71it/s]
2|2839|Loss: 0.8551104068756104:  44%|████▍     | 2839/6470 [32:33<37:43,  1.60it/s]
2|2840|Loss: 0.8577889204025269:  44%|████▍     | 2839/6470 [32:33<37:43,  1.60it/s]
2|2840|Loss: 0.8577889204025269:  44%|████▍     | 2840/6470 [32:34<40:30,  1.49it/s]
2|2841|Loss: 0.8851869702339172:  44%|████▍     | 2840/6470 [32:34<40:30,  1.49it/s]
2|2841|Loss: 0.8851869702339172:  44%|████▍     | 2841/6470 [32:34<41:58,  1.44it/s]
2|2842|Loss: 0.9248358011245728:  44%|████▍     | 2841/6470 [32:35<41:58,  1.44it/s]
2|2842|Loss: 0.9248358011245728:  44%|████▍     | 2842/6470 [32:35<39:31,  1.53it/s]
2|2843|Loss: 0.7638059854507446:  44%|████▍     | 2842/6470 [32:35<39:31,  1.53it/s]
2|2843|Loss: 0.7638059854507446:  44%|████▍     | 2843/6470 [32:35<36:30,  1.66it/s]
2|2844|Loss: 0.8402963876724243:  44%|████▍     | 2843/6470 [32:36<36:30,  1.66it/s]
2|2844|Loss: 0.8402963876724243:  44%|████▍     | 2844/6470 [32:36<39:15,  1.54it/s]
2|2845|Loss: 0.9614951610565186:  44%|████▍     | 2844/6470 [32:37<39:15,  1.54it/s]
2|2845|Loss: 0.9614951610565186:  44%|████▍     | 2845/6470 [32:37<41:07,  1.47it/s]
2|2846|Loss: 0.9907014966011047:  44%|████▍     | 2845/6470 [32:37<41:07,  1.47it/s]
2|2846|Loss: 0.9907014966011047:  44%|████▍     | 2846/6470 [32:38<41:36,  1.45it/s]
2|2847|Loss: 0.7909122109413147:  44%|████▍     | 2846/6470 [32:38<41:36,  1.45it/s]
2|2847|Loss: 0.7909122109413147:  44%|████▍     | 2847/6470 [32:38<43:14,  1.40it/s]
2|2848|Loss: 0.8556387424468994:  44%|████▍     | 2847/6470 [32:39<43:14,  1.40it/s]
2|2848|Loss: 0.8556387424468994:  44%|████▍     | 2848/6470 [32:39<45:45,  1.32it/s]
2|2849|Loss: 0.7193141579627991:  44%|████▍     | 2848/6470 [32:40<45:45,  1.32it/s]
2|2849|Loss: 0.7193141579627991:  44%|████▍     | 2849/6470 [32:40<40:24,  1.49it/s]
2|2850|Loss: 0.8166853785514832:  44%|████▍     | 2849/6470 [32:40<40:24,  1.49it/s]
2|2850|Loss: 0.8166853785514832:  44%|████▍     | 2850/6470 [32:40<40:20,  1.50it/s]
2|2851|Loss: 0.7795246839523315:  44%|████▍     | 2850/6470 [32:41<40:20,  1.50it/s]
2|2851|Loss: 0.7795246839523315:  44%|████▍     | 2851/6470 [32:41<38:37,  1.56it/s]
2|2852|Loss: 0.8683856725692749:  44%|████▍     | 2851/6470 [32:41<38:37,  1.56it/s]
2|2852|Loss: 0.8683856725692749:  44%|████▍     | 2852/6470 [32:42<36:55,  1.63it/s]
2|2853|Loss: 1.0099499225616455:  44%|████▍     | 2852/6470 [32:42<36:55,  1.63it/s]
2|2853|Loss: 1.0099499225616455:  44%|████▍     | 2853/6470 [32:42<36:21,  1.66it/s]
2|2854|Loss: 0.8512158989906311:  44%|████▍     | 2853/6470 [32:42<36:21,  1.66it/s]
2|2854|Loss: 0.8512158989906311:  44%|████▍     | 2854/6470 [32:43<36:44,  1.64it/s]
2|2855|Loss: 0.8549923896789551:  44%|████▍     | 2854/6470 [32:43<36:44,  1.64it/s]
2|2855|Loss: 0.8549923896789551:  44%|████▍     | 2855/6470 [32:44<40:08,  1.50it/s]
2|2856|Loss: 0.9061757922172546:  44%|████▍     | 2855/6470 [32:44<40:08,  1.50it/s]
2|2856|Loss: 0.9061757922172546:  44%|████▍     | 2856/6470 [32:44<40:02,  1.50it/s]
2|2857|Loss: 1.0046508312225342:  44%|████▍     | 2856/6470 [32:45<40:02,  1.50it/s]
2|2857|Loss: 1.0046508312225342:  44%|████▍     | 2857/6470 [32:45<40:24,  1.49it/s]
2|2858|Loss: 0.8507733941078186:  44%|████▍     | 2857/6470 [32:45<40:24,  1.49it/s]
2|2858|Loss: 0.8507733941078186:  44%|████▍     | 2858/6470 [32:46<40:56,  1.47it/s]
2|2859|Loss: 0.9496426582336426:  44%|████▍     | 2858/6470 [32:46<40:56,  1.47it/s]
2|2859|Loss: 0.9496426582336426:  44%|████▍     | 2859/6470 [32:46<41:58,  1.43it/s]
2|2860|Loss: 0.8280661106109619:  44%|████▍     | 2859/6470 [32:47<41:58,  1.43it/s]
2|2860|Loss: 0.8280661106109619:  44%|████▍     | 2860/6470 [32:47<44:30,  1.35it/s]
2|2861|Loss: 0.8616259694099426:  44%|████▍     | 2860/6470 [32:48<44:30,  1.35it/s]
2|2861|Loss: 0.8616259694099426:  44%|████▍     | 2861/6470 [32:48<42:36,  1.41it/s]
2|2862|Loss: 0.7194184064865112:  44%|████▍     | 2861/6470 [32:48<42:36,  1.41it/s]
2|2862|Loss: 0.7194184064865112:  44%|████▍     | 2862/6470 [32:49<44:15,  1.36it/s]
2|2863|Loss: 1.0381897687911987:  44%|████▍     | 2862/6470 [32:49<44:15,  1.36it/s]
2|2863|Loss: 1.0381897687911987:  44%|████▍     | 2863/6470 [32:49<43:11,  1.39it/s]
2|2864|Loss: 0.7819593548774719:  44%|████▍     | 2863/6470 [32:50<43:11,  1.39it/s]
2|2864|Loss: 0.7819593548774719:  44%|████▍     | 2864/6470 [32:50<42:31,  1.41it/s]
2|2865|Loss: 1.0009868144989014:  44%|████▍     | 2864/6470 [32:50<42:31,  1.41it/s]
2|2865|Loss: 1.0009868144989014:  44%|████▍     | 2865/6470 [32:51<42:21,  1.42it/s]
2|2866|Loss: 0.8064713478088379:  44%|████▍     | 2865/6470 [32:51<42:21,  1.42it/s]
2|2866|Loss: 0.8064713478088379:  44%|████▍     | 2866/6470 [32:51<42:50,  1.40it/s]
2|2867|Loss: 0.8076058626174927:  44%|████▍     | 2866/6470 [32:52<42:50,  1.40it/s]
2|2867|Loss: 0.8076058626174927:  44%|████▍     | 2867/6470 [32:52<43:16,  1.39it/s]
2|2868|Loss: 0.9876803159713745:  44%|████▍     | 2867/6470 [32:53<43:16,  1.39it/s]
2|2868|Loss: 0.9876803159713745:  44%|████▍     | 2868/6470 [32:53<43:39,  1.37it/s]
2|2869|Loss: 0.859017014503479:  44%|████▍     | 2868/6470 [32:53<43:39,  1.37it/s] 
2|2869|Loss: 0.859017014503479:  44%|████▍     | 2869/6470 [32:54<42:31,  1.41it/s]
2|2870|Loss: 0.8074820637702942:  44%|████▍     | 2869/6470 [32:54<42:31,  1.41it/s]
2|2870|Loss: 0.8074820637702942:  44%|████▍     | 2870/6470 [32:54<40:24,  1.48it/s]
2|2871|Loss: 1.0151572227478027:  44%|████▍     | 2870/6470 [32:55<40:24,  1.48it/s]
2|2871|Loss: 1.0151572227478027:  44%|████▍     | 2871/6470 [32:55<42:24,  1.41it/s]
2|2872|Loss: 0.8463525176048279:  44%|████▍     | 2871/6470 [32:55<42:24,  1.41it/s]
2|2872|Loss: 0.8463525176048279:  44%|████▍     | 2872/6470 [32:56<42:47,  1.40it/s]
2|2873|Loss: 0.7507273554801941:  44%|████▍     | 2872/6470 [32:56<42:47,  1.40it/s]
2|2873|Loss: 0.7507273554801941:  44%|████▍     | 2873/6470 [32:56<44:21,  1.35it/s]
2|2874|Loss: 0.8340550661087036:  44%|████▍     | 2873/6470 [32:57<44:21,  1.35it/s]
2|2874|Loss: 0.8340550661087036:  44%|████▍     | 2874/6470 [32:57<44:53,  1.33it/s]
2|2875|Loss: 0.7292396426200867:  44%|████▍     | 2874/6470 [32:58<44:53,  1.33it/s]
2|2875|Loss: 0.7292396426200867:  44%|████▍     | 2875/6470 [32:58<43:18,  1.38it/s]
2|2876|Loss: 0.8136548399925232:  44%|████▍     | 2875/6470 [32:58<43:18,  1.38it/s]
2|2876|Loss: 0.8136548399925232:  44%|████▍     | 2876/6470 [32:59<41:26,  1.45it/s]
2|2877|Loss: 0.8063313364982605:  44%|████▍     | 2876/6470 [32:59<41:26,  1.45it/s]
2|2877|Loss: 0.8063313364982605:  44%|████▍     | 2877/6470 [32:59<40:03,  1.50it/s]
2|2878|Loss: 0.7845324873924255:  44%|████▍     | 2877/6470 [33:00<40:03,  1.50it/s]
2|2878|Loss: 0.7845324873924255:  44%|████▍     | 2878/6470 [33:00<42:21,  1.41it/s]
2|2879|Loss: 0.9159602522850037:  44%|████▍     | 2878/6470 [33:00<42:21,  1.41it/s]
2|2879|Loss: 0.9159602522850037:  44%|████▍     | 2879/6470 [33:01<44:03,  1.36it/s]
2|2880|Loss: 0.8529912829399109:  44%|████▍     | 2879/6470 [33:01<44:03,  1.36it/s]
2|2880|Loss: 0.8529912829399109:  45%|████▍     | 2880/6470 [33:01<42:37,  1.40it/s]
2|2881|Loss: 0.9184333086013794:  45%|████▍     | 2880/6470 [33:02<42:37,  1.40it/s]
2|2881|Loss: 0.9184333086013794:  45%|████▍     | 2881/6470 [33:02<40:21,  1.48it/s]
2|2882|Loss: 0.9093562364578247:  45%|████▍     | 2881/6470 [33:02<40:21,  1.48it/s]
2|2882|Loss: 0.9093562364578247:  45%|████▍     | 2882/6470 [33:03<39:45,  1.50it/s]
2|2883|Loss: 0.9640128016471863:  45%|████▍     | 2882/6470 [33:03<39:45,  1.50it/s]
2|2883|Loss: 0.9640128016471863:  45%|████▍     | 2883/6470 [33:03<40:16,  1.48it/s]
2|2884|Loss: 0.7060717344284058:  45%|████▍     | 2883/6470 [33:04<40:16,  1.48it/s]
2|2884|Loss: 0.7060717344284058:  45%|████▍     | 2884/6470 [33:04<43:19,  1.38it/s]
2|2885|Loss: 0.9823983311653137:  45%|████▍     | 2884/6470 [33:05<43:19,  1.38it/s]
2|2885|Loss: 0.9823983311653137:  45%|████▍     | 2885/6470 [33:05<41:52,  1.43it/s]
2|2886|Loss: 0.8673321604728699:  45%|████▍     | 2885/6470 [33:05<41:52,  1.43it/s]
2|2886|Loss: 0.8673321604728699:  45%|████▍     | 2886/6470 [33:05<40:36,  1.47it/s]
2|2887|Loss: 0.8152328729629517:  45%|████▍     | 2886/6470 [33:06<40:36,  1.47it/s]
2|2887|Loss: 0.8152328729629517:  45%|████▍     | 2887/6470 [33:06<42:03,  1.42it/s]
2|2888|Loss: 0.7279656529426575:  45%|████▍     | 2887/6470 [33:07<42:03,  1.42it/s]
2|2888|Loss: 0.7279656529426575:  45%|████▍     | 2888/6470 [33:07<39:41,  1.50it/s]
2|2889|Loss: 0.7745262384414673:  45%|████▍     | 2888/6470 [33:07<39:41,  1.50it/s]
2|2889|Loss: 0.7745262384414673:  45%|████▍     | 2889/6470 [33:07<40:04,  1.49it/s]
2|2890|Loss: 0.886854887008667:  45%|████▍     | 2889/6470 [33:08<40:04,  1.49it/s] 
2|2890|Loss: 0.886854887008667:  45%|████▍     | 2890/6470 [33:08<42:57,  1.39it/s]
2|2891|Loss: 0.8370269536972046:  45%|████▍     | 2890/6470 [33:09<42:57,  1.39it/s]
2|2891|Loss: 0.8370269536972046:  45%|████▍     | 2891/6470 [33:09<45:43,  1.30it/s]
2|2892|Loss: 0.8096831440925598:  45%|████▍     | 2891/6470 [33:10<45:43,  1.30it/s]
2|2892|Loss: 0.8096831440925598:  45%|████▍     | 2892/6470 [33:10<47:06,  1.27it/s]
2|2893|Loss: 0.8861439228057861:  45%|████▍     | 2892/6470 [33:10<47:06,  1.27it/s]
2|2893|Loss: 0.8861439228057861:  45%|████▍     | 2893/6470 [33:11<47:19,  1.26it/s]
2|2894|Loss: 0.6335687637329102:  45%|████▍     | 2893/6470 [33:11<47:19,  1.26it/s]
2|2894|Loss: 0.6335687637329102:  45%|████▍     | 2894/6470 [33:12<46:44,  1.28it/s]
2|2895|Loss: 0.7961530089378357:  45%|████▍     | 2894/6470 [33:12<46:44,  1.28it/s]
2|2895|Loss: 0.7961530089378357:  45%|████▍     | 2895/6470 [33:12<45:54,  1.30it/s]
2|2896|Loss: 0.9553467631340027:  45%|████▍     | 2895/6470 [33:13<45:54,  1.30it/s]
2|2896|Loss: 0.9553467631340027:  45%|████▍     | 2896/6470 [33:13<42:57,  1.39it/s]
2|2897|Loss: 0.8339334726333618:  45%|████▍     | 2896/6470 [33:13<42:57,  1.39it/s]
2|2897|Loss: 0.8339334726333618:  45%|████▍     | 2897/6470 [33:14<42:52,  1.39it/s]
2|2898|Loss: 0.7830079197883606:  45%|████▍     | 2897/6470 [33:14<42:52,  1.39it/s]
2|2898|Loss: 0.7830079197883606:  45%|████▍     | 2898/6470 [33:14<39:11,  1.52it/s]
2|2899|Loss: 0.9151163101196289:  45%|████▍     | 2898/6470 [33:14<39:11,  1.52it/s]
2|2899|Loss: 0.9151163101196289:  45%|████▍     | 2899/6470 [33:15<37:39,  1.58it/s]
2|2900|Loss: 0.8970260620117188:  45%|████▍     | 2899/6470 [33:15<37:39,  1.58it/s]
2|2900|Loss: 0.8970260620117188:  45%|████▍     | 2900/6470 [33:16<40:27,  1.47it/s]
2|2901|Loss: 0.8742332458496094:  45%|████▍     | 2900/6470 [33:16<40:27,  1.47it/s]
2|2901|Loss: 0.8742332458496094:  45%|████▍     | 2901/6470 [33:16<39:37,  1.50it/s]
2|2902|Loss: 0.7648192644119263:  45%|████▍     | 2901/6470 [33:17<39:37,  1.50it/s]
2|2902|Loss: 0.7648192644119263:  45%|████▍     | 2902/6470 [33:17<40:23,  1.47it/s]
2|2903|Loss: 0.9291645884513855:  45%|████▍     | 2902/6470 [33:17<40:23,  1.47it/s]
2|2903|Loss: 0.9291645884513855:  45%|████▍     | 2903/6470 [33:18<41:19,  1.44it/s]
2|2904|Loss: 0.6494637727737427:  45%|████▍     | 2903/6470 [33:18<41:19,  1.44it/s]
2|2904|Loss: 0.6494637727737427:  45%|████▍     | 2904/6470 [33:18<41:56,  1.42it/s]
2|2905|Loss: 0.9102098941802979:  45%|████▍     | 2904/6470 [33:19<41:56,  1.42it/s]
2|2905|Loss: 0.9102098941802979:  45%|████▍     | 2905/6470 [33:19<42:02,  1.41it/s]
2|2906|Loss: 0.9223275780677795:  45%|████▍     | 2905/6470 [33:19<42:02,  1.41it/s]
2|2906|Loss: 0.9223275780677795:  45%|████▍     | 2906/6470 [33:20<44:27,  1.34it/s]
2|2907|Loss: 0.8716521859169006:  45%|████▍     | 2906/6470 [33:20<44:27,  1.34it/s]
2|2907|Loss: 0.8716521859169006:  45%|████▍     | 2907/6470 [33:21<45:40,  1.30it/s]
2|2908|Loss: 0.8754319548606873:  45%|████▍     | 2907/6470 [33:21<45:40,  1.30it/s]
2|2908|Loss: 0.8754319548606873:  45%|████▍     | 2908/6470 [33:22<47:22,  1.25it/s]
2|2909|Loss: 1.0062347650527954:  45%|████▍     | 2908/6470 [33:22<47:22,  1.25it/s]
2|2909|Loss: 1.0062347650527954:  45%|████▍     | 2909/6470 [33:22<48:20,  1.23it/s]
2|2910|Loss: 0.9245194792747498:  45%|████▍     | 2909/6470 [33:23<48:20,  1.23it/s]
2|2910|Loss: 0.9245194792747498:  45%|████▍     | 2910/6470 [33:23<45:35,  1.30it/s]
2|2911|Loss: 0.7593980431556702:  45%|████▍     | 2910/6470 [33:23<45:35,  1.30it/s]
2|2911|Loss: 0.7593980431556702:  45%|████▍     | 2911/6470 [33:24<45:19,  1.31it/s]
2|2912|Loss: 0.8110731840133667:  45%|████▍     | 2911/6470 [33:24<45:19,  1.31it/s]
2|2912|Loss: 0.8110731840133667:  45%|████▌     | 2912/6470 [33:25<45:53,  1.29it/s]
2|2913|Loss: 0.7739070057868958:  45%|████▌     | 2912/6470 [33:25<45:53,  1.29it/s]
2|2913|Loss: 0.7739070057868958:  45%|████▌     | 2913/6470 [33:25<43:52,  1.35it/s]
2|2914|Loss: 0.9166863560676575:  45%|████▌     | 2913/6470 [33:26<43:52,  1.35it/s]
2|2914|Loss: 0.9166863560676575:  45%|████▌     | 2914/6470 [33:26<43:21,  1.37it/s]
2|2915|Loss: 0.9058347344398499:  45%|████▌     | 2914/6470 [33:26<43:21,  1.37it/s]
2|2915|Loss: 0.9058347344398499:  45%|████▌     | 2915/6470 [33:27<44:13,  1.34it/s]
2|2916|Loss: 0.8273342847824097:  45%|████▌     | 2915/6470 [33:27<44:13,  1.34it/s]
2|2916|Loss: 0.8273342847824097:  45%|████▌     | 2916/6470 [33:27<42:07,  1.41it/s]
2|2917|Loss: 0.9388412833213806:  45%|████▌     | 2916/6470 [33:28<42:07,  1.41it/s]
2|2917|Loss: 0.9388412833213806:  45%|████▌     | 2917/6470 [33:28<41:29,  1.43it/s]
2|2918|Loss: 0.9358711838722229:  45%|████▌     | 2917/6470 [33:29<41:29,  1.43it/s]
2|2918|Loss: 0.9358711838722229:  45%|████▌     | 2918/6470 [33:29<43:43,  1.35it/s]
2|2919|Loss: 0.9059931039810181:  45%|████▌     | 2918/6470 [33:29<43:43,  1.35it/s]
2|2919|Loss: 0.9059931039810181:  45%|████▌     | 2919/6470 [33:30<41:55,  1.41it/s]
2|2920|Loss: 0.8206831812858582:  45%|████▌     | 2919/6470 [33:30<41:55,  1.41it/s]
2|2920|Loss: 0.8206831812858582:  45%|████▌     | 2920/6470 [33:30<41:48,  1.42it/s]
2|2921|Loss: 0.791800856590271:  45%|████▌     | 2920/6470 [33:31<41:48,  1.42it/s] 
2|2921|Loss: 0.791800856590271:  45%|████▌     | 2921/6470 [33:31<43:15,  1.37it/s]
2|2922|Loss: 1.025132417678833:  45%|████▌     | 2921/6470 [33:31<43:15,  1.37it/s]
2|2922|Loss: 1.025132417678833:  45%|████▌     | 2922/6470 [33:32<43:50,  1.35it/s]
2|2923|Loss: 0.6984193921089172:  45%|████▌     | 2922/6470 [33:32<43:50,  1.35it/s]
2|2923|Loss: 0.6984193921089172:  45%|████▌     | 2923/6470 [33:32<42:32,  1.39it/s]
2|2924|Loss: 0.7494181990623474:  45%|████▌     | 2923/6470 [33:33<42:32,  1.39it/s]
2|2924|Loss: 0.7494181990623474:  45%|████▌     | 2924/6470 [33:33<42:33,  1.39it/s]
2|2925|Loss: 0.9500545263290405:  45%|████▌     | 2924/6470 [33:34<42:33,  1.39it/s]
2|2925|Loss: 0.9500545263290405:  45%|████▌     | 2925/6470 [33:34<41:08,  1.44it/s]
2|2926|Loss: 0.9001798033714294:  45%|████▌     | 2925/6470 [33:34<41:08,  1.44it/s]
2|2926|Loss: 0.9001798033714294:  45%|████▌     | 2926/6470 [33:34<39:37,  1.49it/s]
2|2927|Loss: 0.8348307609558105:  45%|████▌     | 2926/6470 [33:35<39:37,  1.49it/s]
2|2927|Loss: 0.8348307609558105:  45%|████▌     | 2927/6470 [33:35<37:09,  1.59it/s]
2|2928|Loss: 0.8691132068634033:  45%|████▌     | 2927/6470 [33:35<37:09,  1.59it/s]
2|2928|Loss: 0.8691132068634033:  45%|████▌     | 2928/6470 [33:36<36:30,  1.62it/s]
2|2929|Loss: 0.7199257612228394:  45%|████▌     | 2928/6470 [33:36<36:30,  1.62it/s]
2|2929|Loss: 0.7199257612228394:  45%|████▌     | 2929/6470 [33:36<38:00,  1.55it/s]
2|2930|Loss: 0.8382782936096191:  45%|████▌     | 2929/6470 [33:37<38:00,  1.55it/s]
2|2930|Loss: 0.8382782936096191:  45%|████▌     | 2930/6470 [33:37<36:20,  1.62it/s]
2|2931|Loss: 0.7896175980567932:  45%|████▌     | 2930/6470 [33:37<36:20,  1.62it/s]
2|2931|Loss: 0.7896175980567932:  45%|████▌     | 2931/6470 [33:37<36:20,  1.62it/s]
2|2932|Loss: 0.8445101976394653:  45%|████▌     | 2931/6470 [33:38<36:20,  1.62it/s]
2|2932|Loss: 0.8445101976394653:  45%|████▌     | 2932/6470 [33:38<40:01,  1.47it/s]
2|2933|Loss: 0.8701125383377075:  45%|████▌     | 2932/6470 [33:39<40:01,  1.47it/s]
2|2933|Loss: 0.8701125383377075:  45%|████▌     | 2933/6470 [33:39<41:07,  1.43it/s]
2|2934|Loss: 0.7931240797042847:  45%|████▌     | 2933/6470 [33:39<41:07,  1.43it/s]
2|2934|Loss: 0.7931240797042847:  45%|████▌     | 2934/6470 [33:40<41:34,  1.42it/s]
2|2935|Loss: 0.8166863918304443:  45%|████▌     | 2934/6470 [33:40<41:34,  1.42it/s]
2|2935|Loss: 0.8166863918304443:  45%|████▌     | 2935/6470 [33:41<43:26,  1.36it/s]
2|2936|Loss: 0.8835335969924927:  45%|████▌     | 2935/6470 [33:41<43:26,  1.36it/s]
2|2936|Loss: 0.8835335969924927:  45%|████▌     | 2936/6470 [33:41<43:24,  1.36it/s]
2|2937|Loss: 0.8397373557090759:  45%|████▌     | 2936/6470 [33:42<43:24,  1.36it/s]
2|2937|Loss: 0.8397373557090759:  45%|████▌     | 2937/6470 [33:42<41:55,  1.40it/s]
2|2938|Loss: 0.9061399698257446:  45%|████▌     | 2937/6470 [33:42<41:55,  1.40it/s]
2|2938|Loss: 0.9061399698257446:  45%|████▌     | 2938/6470 [33:43<42:07,  1.40it/s]
2|2939|Loss: 0.859870970249176:  45%|████▌     | 2938/6470 [33:43<42:07,  1.40it/s] 
2|2939|Loss: 0.859870970249176:  45%|████▌     | 2939/6470 [33:43<43:52,  1.34it/s]
2|2940|Loss: 0.9202128648757935:  45%|████▌     | 2939/6470 [33:44<43:52,  1.34it/s]
2|2940|Loss: 0.9202128648757935:  45%|████▌     | 2940/6470 [33:44<43:31,  1.35it/s]
2|2941|Loss: 0.8111826181411743:  45%|████▌     | 2940/6470 [33:45<43:31,  1.35it/s]
2|2941|Loss: 0.8111826181411743:  45%|████▌     | 2941/6470 [33:45<40:59,  1.43it/s]
2|2942|Loss: 0.7384402751922607:  45%|████▌     | 2941/6470 [33:45<40:59,  1.43it/s]
2|2942|Loss: 0.7384402751922607:  45%|████▌     | 2942/6470 [33:46<42:00,  1.40it/s]
2|2943|Loss: 0.8207780122756958:  45%|████▌     | 2942/6470 [33:46<42:00,  1.40it/s]
2|2943|Loss: 0.8207780122756958:  45%|████▌     | 2943/6470 [33:46<40:20,  1.46it/s]
2|2944|Loss: 1.005717396736145:  45%|████▌     | 2943/6470 [33:47<40:20,  1.46it/s] 
2|2944|Loss: 1.005717396736145:  46%|████▌     | 2944/6470 [33:47<41:53,  1.40it/s]
2|2945|Loss: 0.9141271710395813:  46%|████▌     | 2944/6470 [33:47<41:53,  1.40it/s]
2|2945|Loss: 0.9141271710395813:  46%|████▌     | 2945/6470 [33:48<43:51,  1.34it/s]
2|2946|Loss: 0.8284837007522583:  46%|████▌     | 2945/6470 [33:48<43:51,  1.34it/s]
2|2946|Loss: 0.8284837007522583:  46%|████▌     | 2946/6470 [33:48<40:52,  1.44it/s]
2|2947|Loss: 0.887894332408905:  46%|████▌     | 2946/6470 [33:49<40:52,  1.44it/s] 
2|2947|Loss: 0.887894332408905:  46%|████▌     | 2947/6470 [33:49<39:04,  1.50it/s]
2|2948|Loss: 0.9079124331474304:  46%|████▌     | 2947/6470 [33:49<39:04,  1.50it/s]
2|2948|Loss: 0.9079124331474304:  46%|████▌     | 2948/6470 [33:50<38:27,  1.53it/s]
2|2949|Loss: 0.7450752258300781:  46%|████▌     | 2948/6470 [33:50<38:27,  1.53it/s]
2|2949|Loss: 0.7450752258300781:  46%|████▌     | 2949/6470 [33:50<36:31,  1.61it/s]
2|2950|Loss: 0.7791420221328735:  46%|████▌     | 2949/6470 [33:51<36:31,  1.61it/s]
2|2950|Loss: 0.7791420221328735:  46%|████▌     | 2950/6470 [33:51<38:39,  1.52it/s]
2|2951|Loss: 0.8743156790733337:  46%|████▌     | 2950/6470 [33:51<38:39,  1.52it/s]
2|2951|Loss: 0.8743156790733337:  46%|████▌     | 2951/6470 [33:52<41:02,  1.43it/s]
2|2952|Loss: 0.9428529143333435:  46%|████▌     | 2951/6470 [33:52<41:02,  1.43it/s]
2|2952|Loss: 0.9428529143333435:  46%|████▌     | 2952/6470 [33:52<41:29,  1.41it/s]
2|2953|Loss: 0.8798776865005493:  46%|████▌     | 2952/6470 [33:53<41:29,  1.41it/s]
2|2953|Loss: 0.8798776865005493:  46%|████▌     | 2953/6470 [33:53<40:34,  1.44it/s]
2|2954|Loss: 0.8791096806526184:  46%|████▌     | 2953/6470 [33:53<40:34,  1.44it/s]
2|2954|Loss: 0.8791096806526184:  46%|████▌     | 2954/6470 [33:54<38:49,  1.51it/s]
2|2955|Loss: 0.8135366439819336:  46%|████▌     | 2954/6470 [33:54<38:49,  1.51it/s]
2|2955|Loss: 0.8135366439819336:  46%|████▌     | 2955/6470 [33:54<41:20,  1.42it/s]
2|2956|Loss: 0.9338163137435913:  46%|████▌     | 2955/6470 [33:55<41:20,  1.42it/s]
2|2956|Loss: 0.9338163137435913:  46%|████▌     | 2956/6470 [33:55<39:16,  1.49it/s]
2|2957|Loss: 0.9035335779190063:  46%|████▌     | 2956/6470 [33:55<39:16,  1.49it/s]
2|2957|Loss: 0.9035335779190063:  46%|████▌     | 2957/6470 [33:56<39:26,  1.48it/s]
2|2958|Loss: 0.7389540076255798:  46%|████▌     | 2957/6470 [33:56<39:26,  1.48it/s]
2|2958|Loss: 0.7389540076255798:  46%|████▌     | 2958/6470 [33:56<40:50,  1.43it/s]
2|2959|Loss: 0.7849401831626892:  46%|████▌     | 2958/6470 [33:57<40:50,  1.43it/s]
2|2959|Loss: 0.7849401831626892:  46%|████▌     | 2959/6470 [33:57<40:06,  1.46it/s]
2|2960|Loss: 1.0044207572937012:  46%|████▌     | 2959/6470 [33:57<40:06,  1.46it/s]
2|2960|Loss: 1.0044207572937012:  46%|████▌     | 2960/6470 [33:58<37:41,  1.55it/s]
2|2961|Loss: 0.831360399723053:  46%|████▌     | 2960/6470 [33:58<37:41,  1.55it/s] 
2|2961|Loss: 0.831360399723053:  46%|████▌     | 2961/6470 [33:58<36:56,  1.58it/s]
2|2962|Loss: 0.7549526691436768:  46%|████▌     | 2961/6470 [33:59<36:56,  1.58it/s]
2|2962|Loss: 0.7549526691436768:  46%|████▌     | 2962/6470 [33:59<36:36,  1.60it/s]
2|2963|Loss: 0.8771184086799622:  46%|████▌     | 2962/6470 [33:59<36:36,  1.60it/s]
2|2963|Loss: 0.8771184086799622:  46%|████▌     | 2963/6470 [34:00<38:20,  1.52it/s]
2|2964|Loss: 0.8928730487823486:  46%|████▌     | 2963/6470 [34:00<38:20,  1.52it/s]
2|2964|Loss: 0.8928730487823486:  46%|████▌     | 2964/6470 [34:00<38:16,  1.53it/s]
2|2965|Loss: 0.8659977316856384:  46%|████▌     | 2964/6470 [34:01<38:16,  1.53it/s]
2|2965|Loss: 0.8659977316856384:  46%|████▌     | 2965/6470 [34:01<36:57,  1.58it/s]
2|2966|Loss: 0.8615469336509705:  46%|████▌     | 2965/6470 [34:01<36:57,  1.58it/s]
2|2966|Loss: 0.8615469336509705:  46%|████▌     | 2966/6470 [34:02<38:22,  1.52it/s]
2|2967|Loss: 0.8466573357582092:  46%|████▌     | 2966/6470 [34:02<38:22,  1.52it/s]
2|2967|Loss: 0.8466573357582092:  46%|████▌     | 2967/6470 [34:02<39:24,  1.48it/s]
2|2968|Loss: 0.9573312401771545:  46%|████▌     | 2967/6470 [34:03<39:24,  1.48it/s]
2|2968|Loss: 0.9573312401771545:  46%|████▌     | 2968/6470 [34:03<40:33,  1.44it/s]
2|2969|Loss: 0.6733839511871338:  46%|████▌     | 2968/6470 [34:03<40:33,  1.44it/s]
2|2969|Loss: 0.6733839511871338:  46%|████▌     | 2969/6470 [34:04<39:11,  1.49it/s]
2|2970|Loss: 0.7722553014755249:  46%|████▌     | 2969/6470 [34:04<39:11,  1.49it/s]
2|2970|Loss: 0.7722553014755249:  46%|████▌     | 2970/6470 [34:04<34:34,  1.69it/s]
2|2971|Loss: 0.8087525963783264:  46%|████▌     | 2970/6470 [34:04<34:34,  1.69it/s]
2|2971|Loss: 0.8087525963783264:  46%|████▌     | 2971/6470 [34:05<36:45,  1.59it/s]
2|2972|Loss: 0.9429881572723389:  46%|████▌     | 2971/6470 [34:05<36:45,  1.59it/s]
2|2972|Loss: 0.9429881572723389:  46%|████▌     | 2972/6470 [34:06<42:31,  1.37it/s]
2|2973|Loss: 0.882897138595581:  46%|████▌     | 2972/6470 [34:06<42:31,  1.37it/s] 
2|2973|Loss: 0.882897138595581:  46%|████▌     | 2973/6470 [34:07<45:15,  1.29it/s]
2|2974|Loss: 0.9723256230354309:  46%|████▌     | 2973/6470 [34:07<45:15,  1.29it/s]
2|2974|Loss: 0.9723256230354309:  46%|████▌     | 2974/6470 [34:07<43:35,  1.34it/s]
2|2975|Loss: 0.8597050309181213:  46%|████▌     | 2974/6470 [34:08<43:35,  1.34it/s]
2|2975|Loss: 0.8597050309181213:  46%|████▌     | 2975/6470 [34:08<41:42,  1.40it/s]
2|2976|Loss: 1.0351091623306274:  46%|████▌     | 2975/6470 [34:08<41:42,  1.40it/s]
2|2976|Loss: 1.0351091623306274:  46%|████▌     | 2976/6470 [34:09<42:29,  1.37it/s]
2|2977|Loss: 0.7688989043235779:  46%|████▌     | 2976/6470 [34:09<42:29,  1.37it/s]
2|2977|Loss: 0.7688989043235779:  46%|████▌     | 2977/6470 [34:09<39:15,  1.48it/s]
2|2978|Loss: 0.7848482728004456:  46%|████▌     | 2977/6470 [34:10<39:15,  1.48it/s]
2|2978|Loss: 0.7848482728004456:  46%|████▌     | 2978/6470 [34:10<45:28,  1.28it/s]
2|2979|Loss: 0.8865481615066528:  46%|████▌     | 2978/6470 [34:11<45:28,  1.28it/s]
2|2979|Loss: 0.8865481615066528:  46%|████▌     | 2979/6470 [34:11<43:58,  1.32it/s]
2|2980|Loss: 0.8193626999855042:  46%|████▌     | 2979/6470 [34:11<43:58,  1.32it/s]
2|2980|Loss: 0.8193626999855042:  46%|████▌     | 2980/6470 [34:12<41:51,  1.39it/s]
2|2981|Loss: 0.9349809885025024:  46%|████▌     | 2980/6470 [34:12<41:51,  1.39it/s]
2|2981|Loss: 0.9349809885025024:  46%|████▌     | 2981/6470 [34:12<42:10,  1.38it/s]
2|2982|Loss: 0.886417806148529:  46%|████▌     | 2981/6470 [34:13<42:10,  1.38it/s] 
2|2982|Loss: 0.886417806148529:  46%|████▌     | 2982/6470 [34:13<42:24,  1.37it/s]
2|2983|Loss: 0.8524577617645264:  46%|████▌     | 2982/6470 [34:13<42:24,  1.37it/s]
2|2983|Loss: 0.8524577617645264:  46%|████▌     | 2983/6470 [34:14<39:53,  1.46it/s]
2|2984|Loss: 0.8796411752700806:  46%|████▌     | 2983/6470 [34:14<39:53,  1.46it/s]
2|2984|Loss: 0.8796411752700806:  46%|████▌     | 2984/6470 [34:14<39:51,  1.46it/s]
2|2985|Loss: 0.9536864757537842:  46%|████▌     | 2984/6470 [34:15<39:51,  1.46it/s]
2|2985|Loss: 0.9536864757537842:  46%|████▌     | 2985/6470 [34:15<40:09,  1.45it/s]
2|2986|Loss: 0.7637151479721069:  46%|████▌     | 2985/6470 [34:15<40:09,  1.45it/s]
2|2986|Loss: 0.7637151479721069:  46%|████▌     | 2986/6470 [34:16<40:13,  1.44it/s]
2|2987|Loss: 0.7873594760894775:  46%|████▌     | 2986/6470 [34:16<40:13,  1.44it/s]
2|2987|Loss: 0.7873594760894775:  46%|████▌     | 2987/6470 [34:16<37:54,  1.53it/s]
2|2988|Loss: 0.8847052454948425:  46%|████▌     | 2987/6470 [34:17<37:54,  1.53it/s]
2|2988|Loss: 0.8847052454948425:  46%|████▌     | 2988/6470 [34:17<36:57,  1.57it/s]
2|2989|Loss: 0.6096981167793274:  46%|████▌     | 2988/6470 [34:17<36:57,  1.57it/s]
2|2989|Loss: 0.6096981167793274:  46%|████▌     | 2989/6470 [34:18<40:07,  1.45it/s]
2|2990|Loss: 0.7854438424110413:  46%|████▌     | 2989/6470 [34:18<40:07,  1.45it/s]
2|2990|Loss: 0.7854438424110413:  46%|████▌     | 2990/6470 [34:18<39:51,  1.46it/s]
2|2991|Loss: 0.794829249382019:  46%|████▌     | 2990/6470 [34:19<39:51,  1.46it/s] 
2|2991|Loss: 0.794829249382019:  46%|████▌     | 2991/6470 [34:19<38:34,  1.50it/s]
2|2992|Loss: 0.7907984256744385:  46%|████▌     | 2991/6470 [34:19<38:34,  1.50it/s]
2|2992|Loss: 0.7907984256744385:  46%|████▌     | 2992/6470 [34:20<39:30,  1.47it/s]
2|2993|Loss: 0.7861685752868652:  46%|████▌     | 2992/6470 [34:20<39:30,  1.47it/s]
2|2993|Loss: 0.7861685752868652:  46%|████▋     | 2993/6470 [34:21<40:47,  1.42it/s]
2|2994|Loss: 0.6752110719680786:  46%|████▋     | 2993/6470 [34:21<40:47,  1.42it/s]
2|2994|Loss: 0.6752110719680786:  46%|████▋     | 2994/6470 [34:21<43:04,  1.34it/s]
2|2995|Loss: 0.9165283441543579:  46%|████▋     | 2994/6470 [34:22<43:04,  1.34it/s]
2|2995|Loss: 0.9165283441543579:  46%|████▋     | 2995/6470 [34:22<43:01,  1.35it/s]
2|2996|Loss: 0.8787147998809814:  46%|████▋     | 2995/6470 [34:22<43:01,  1.35it/s]
2|2996|Loss: 0.8787147998809814:  46%|████▋     | 2996/6470 [34:23<39:38,  1.46it/s]
2|2997|Loss: 0.8205932974815369:  46%|████▋     | 2996/6470 [34:23<39:38,  1.46it/s]
2|2997|Loss: 0.8205932974815369:  46%|████▋     | 2997/6470 [34:23<38:10,  1.52it/s]
2|2998|Loss: 0.8856626152992249:  46%|████▋     | 2997/6470 [34:24<38:10,  1.52it/s]
2|2998|Loss: 0.8856626152992249:  46%|████▋     | 2998/6470 [34:24<36:09,  1.60it/s]
2|2999|Loss: 0.8286121487617493:  46%|████▋     | 2998/6470 [34:24<36:09,  1.60it/s]
2|2999|Loss: 0.8286121487617493:  46%|████▋     | 2999/6470 [34:25<38:10,  1.52it/s]
2|3000|Loss: 0.8584571480751038:  46%|████▋     | 2999/6470 [34:25<38:10,  1.52it/s]
2|3000|Loss: 0.8584571480751038:  46%|████▋     | 3000/6470 [34:25<39:18,  1.47it/s]
2|3001|Loss: 0.7687407732009888:  46%|████▋     | 3000/6470 [34:26<39:18,  1.47it/s]
2|3001|Loss: 0.7687407732009888:  46%|████▋     | 3001/6470 [34:26<46:32,  1.24it/s]
2|3002|Loss: 0.9432716965675354:  46%|████▋     | 3001/6470 [34:27<46:32,  1.24it/s]
2|3002|Loss: 0.9432716965675354:  46%|████▋     | 3002/6470 [34:27<46:40,  1.24it/s]
2|3003|Loss: 0.924888014793396:  46%|████▋     | 3002/6470 [34:28<46:40,  1.24it/s] 
2|3003|Loss: 0.924888014793396:  46%|████▋     | 3003/6470 [34:28<43:46,  1.32it/s]
2|3004|Loss: 0.800801694393158:  46%|████▋     | 3003/6470 [34:28<43:46,  1.32it/s]
2|3004|Loss: 0.800801694393158:  46%|████▋     | 3004/6470 [34:28<42:51,  1.35it/s]
2|3005|Loss: 0.808570921421051:  46%|████▋     | 3004/6470 [34:29<42:51,  1.35it/s]
2|3005|Loss: 0.808570921421051:  46%|████▋     | 3005/6470 [34:29<42:51,  1.35it/s]
2|3006|Loss: 0.8988369107246399:  46%|████▋     | 3005/6470 [34:30<42:51,  1.35it/s]
2|3006|Loss: 0.8988369107246399:  46%|████▋     | 3006/6470 [34:30<41:36,  1.39it/s]
2|3007|Loss: 0.733075737953186:  46%|████▋     | 3006/6470 [34:30<41:36,  1.39it/s] 
2|3007|Loss: 0.733075737953186:  46%|████▋     | 3007/6470 [34:31<42:18,  1.36it/s]
2|3008|Loss: 0.9066857099533081:  46%|████▋     | 3007/6470 [34:31<42:18,  1.36it/s]
2|3008|Loss: 0.9066857099533081:  46%|████▋     | 3008/6470 [34:31<41:19,  1.40it/s]
2|3009|Loss: 0.7254767417907715:  46%|████▋     | 3008/6470 [34:32<41:19,  1.40it/s]
2|3009|Loss: 0.7254767417907715:  47%|████▋     | 3009/6470 [34:32<39:39,  1.45it/s]
2|3010|Loss: 0.977312445640564:  47%|████▋     | 3009/6470 [34:32<39:39,  1.45it/s] 
2|3010|Loss: 0.977312445640564:  47%|████▋     | 3010/6470 [34:33<37:11,  1.55it/s]
2|3011|Loss: 0.6473312377929688:  47%|████▋     | 3010/6470 [34:33<37:11,  1.55it/s]
2|3011|Loss: 0.6473312377929688:  47%|████▋     | 3011/6470 [34:33<33:10,  1.74it/s]
2|3012|Loss: 0.7588306069374084:  47%|████▋     | 3011/6470 [34:33<33:10,  1.74it/s]
2|3012|Loss: 0.7588306069374084:  47%|████▋     | 3012/6470 [34:34<33:11,  1.74it/s]
2|3013|Loss: 0.9254162907600403:  47%|████▋     | 3012/6470 [34:34<33:11,  1.74it/s]
2|3013|Loss: 0.9254162907600403:  47%|████▋     | 3013/6470 [34:34<36:14,  1.59it/s]
2|3014|Loss: 0.9663258790969849:  47%|████▋     | 3013/6470 [34:35<36:14,  1.59it/s]
2|3014|Loss: 0.9663258790969849:  47%|████▋     | 3014/6470 [34:35<35:47,  1.61it/s]
2|3015|Loss: 0.7945215106010437:  47%|████▋     | 3014/6470 [34:35<35:47,  1.61it/s]
2|3015|Loss: 0.7945215106010437:  47%|████▋     | 3015/6470 [34:36<36:40,  1.57it/s]
2|3016|Loss: 0.8587350845336914:  47%|████▋     | 3015/6470 [34:36<36:40,  1.57it/s]
2|3016|Loss: 0.8587350845336914:  47%|████▋     | 3016/6470 [34:36<37:00,  1.56it/s]
2|3017|Loss: 0.9688517451286316:  47%|████▋     | 3016/6470 [34:37<37:00,  1.56it/s]
2|3017|Loss: 0.9688517451286316:  47%|████▋     | 3017/6470 [34:37<37:47,  1.52it/s]
2|3018|Loss: 0.874375581741333:  47%|████▋     | 3017/6470 [34:37<37:47,  1.52it/s] 
2|3018|Loss: 0.874375581741333:  47%|████▋     | 3018/6470 [34:37<35:59,  1.60it/s]
2|3019|Loss: 0.8951722979545593:  47%|████▋     | 3018/6470 [34:38<35:59,  1.60it/s]
2|3019|Loss: 0.8951722979545593:  47%|████▋     | 3019/6470 [34:38<37:55,  1.52it/s]
2|3020|Loss: 0.7494929432868958:  47%|████▋     | 3019/6470 [34:39<37:55,  1.52it/s]
2|3020|Loss: 0.7494929432868958:  47%|████▋     | 3020/6470 [34:39<37:56,  1.52it/s]
2|3021|Loss: 0.8126347064971924:  47%|████▋     | 3020/6470 [34:39<37:56,  1.52it/s]
2|3021|Loss: 0.8126347064971924:  47%|████▋     | 3021/6470 [34:40<38:26,  1.50it/s]
2|3022|Loss: 0.8548638820648193:  47%|████▋     | 3021/6470 [34:40<38:26,  1.50it/s]
2|3022|Loss: 0.8548638820648193:  47%|████▋     | 3022/6470 [34:40<39:27,  1.46it/s]
2|3023|Loss: 1.0730434656143188:  47%|████▋     | 3022/6470 [34:41<39:27,  1.46it/s]
2|3023|Loss: 1.0730434656143188:  47%|████▋     | 3023/6470 [34:41<40:26,  1.42it/s]
2|3024|Loss: 0.7681857943534851:  47%|████▋     | 3023/6470 [34:41<40:26,  1.42it/s]
2|3024|Loss: 0.7681857943534851:  47%|████▋     | 3024/6470 [34:42<39:45,  1.44it/s]
2|3025|Loss: 0.8614028692245483:  47%|████▋     | 3024/6470 [34:42<39:45,  1.44it/s]
2|3025|Loss: 0.8614028692245483:  47%|████▋     | 3025/6470 [34:42<38:01,  1.51it/s]
2|3026|Loss: 0.8106695413589478:  47%|████▋     | 3025/6470 [34:43<38:01,  1.51it/s]
2|3026|Loss: 0.8106695413589478:  47%|████▋     | 3026/6470 [34:43<37:07,  1.55it/s]
2|3027|Loss: 0.8485016822814941:  47%|████▋     | 3026/6470 [34:43<37:07,  1.55it/s]
2|3027|Loss: 0.8485016822814941:  47%|████▋     | 3027/6470 [34:44<40:37,  1.41it/s]
2|3028|Loss: 0.853922963142395:  47%|████▋     | 3027/6470 [34:44<40:37,  1.41it/s] 
2|3028|Loss: 0.853922963142395:  47%|████▋     | 3028/6470 [34:45<43:42,  1.31it/s]
2|3029|Loss: 0.930967390537262:  47%|████▋     | 3028/6470 [34:45<43:42,  1.31it/s]
2|3029|Loss: 0.930967390537262:  47%|████▋     | 3029/6470 [34:45<41:38,  1.38it/s]
2|3030|Loss: 0.8781222701072693:  47%|████▋     | 3029/6470 [34:46<41:38,  1.38it/s]
2|3030|Loss: 0.8781222701072693:  47%|████▋     | 3030/6470 [34:46<39:52,  1.44it/s]
2|3031|Loss: 0.8254429697990417:  47%|████▋     | 3030/6470 [34:46<39:52,  1.44it/s]
2|3031|Loss: 0.8254429697990417:  47%|████▋     | 3031/6470 [34:47<40:09,  1.43it/s]
2|3032|Loss: 0.8359687328338623:  47%|████▋     | 3031/6470 [34:47<40:09,  1.43it/s]
2|3032|Loss: 0.8359687328338623:  47%|████▋     | 3032/6470 [34:47<41:32,  1.38it/s]
2|3033|Loss: 0.9715184569358826:  47%|████▋     | 3032/6470 [34:48<41:32,  1.38it/s]
2|3033|Loss: 0.9715184569358826:  47%|████▋     | 3033/6470 [34:48<39:28,  1.45it/s]
2|3034|Loss: 0.9598472118377686:  47%|████▋     | 3033/6470 [34:48<39:28,  1.45it/s]
2|3034|Loss: 0.9598472118377686:  47%|████▋     | 3034/6470 [34:49<39:42,  1.44it/s]
2|3035|Loss: 0.74624103307724:  47%|████▋     | 3034/6470 [34:49<39:42,  1.44it/s]  
2|3035|Loss: 0.74624103307724:  47%|████▋     | 3035/6470 [34:49<38:22,  1.49it/s]
2|3036|Loss: 0.9767642617225647:  47%|████▋     | 3035/6470 [34:50<38:22,  1.49it/s]
2|3036|Loss: 0.9767642617225647:  47%|████▋     | 3036/6470 [34:50<39:54,  1.43it/s]
2|3037|Loss: 0.8580962419509888:  47%|████▋     | 3036/6470 [34:51<39:54,  1.43it/s]
2|3037|Loss: 0.8580962419509888:  47%|████▋     | 3037/6470 [34:51<42:25,  1.35it/s]
2|3038|Loss: 0.9447805881500244:  47%|████▋     | 3037/6470 [34:51<42:25,  1.35it/s]
2|3038|Loss: 0.9447805881500244:  47%|████▋     | 3038/6470 [34:52<41:33,  1.38it/s]
2|3039|Loss: 0.7563446164131165:  47%|████▋     | 3038/6470 [34:52<41:33,  1.38it/s]
2|3039|Loss: 0.7563446164131165:  47%|████▋     | 3039/6470 [34:52<42:16,  1.35it/s]
2|3040|Loss: 0.7631999254226685:  47%|████▋     | 3039/6470 [34:53<42:16,  1.35it/s]
2|3040|Loss: 0.7631999254226685:  47%|████▋     | 3040/6470 [34:53<36:51,  1.55it/s]
2|3041|Loss: 0.8503379821777344:  47%|████▋     | 3040/6470 [34:53<36:51,  1.55it/s]
2|3041|Loss: 0.8503379821777344:  47%|████▋     | 3041/6470 [34:53<36:21,  1.57it/s]
2|3042|Loss: 0.9578460454940796:  47%|████▋     | 3041/6470 [34:54<36:21,  1.57it/s]
2|3042|Loss: 0.9578460454940796:  47%|████▋     | 3042/6470 [34:54<35:05,  1.63it/s]
2|3043|Loss: 0.7175610065460205:  47%|████▋     | 3042/6470 [34:54<35:05,  1.63it/s]
2|3043|Loss: 0.7175610065460205:  47%|████▋     | 3043/6470 [34:55<37:15,  1.53it/s]
2|3044|Loss: 0.8012440204620361:  47%|████▋     | 3043/6470 [34:55<37:15,  1.53it/s]
2|3044|Loss: 0.8012440204620361:  47%|████▋     | 3044/6470 [34:55<39:43,  1.44it/s]
2|3045|Loss: 0.8844432234764099:  47%|████▋     | 3044/6470 [34:56<39:43,  1.44it/s]
2|3045|Loss: 0.8844432234764099:  47%|████▋     | 3045/6470 [34:56<39:20,  1.45it/s]
2|3046|Loss: 0.8425244688987732:  47%|████▋     | 3045/6470 [34:56<39:20,  1.45it/s]
2|3046|Loss: 0.8425244688987732:  47%|████▋     | 3046/6470 [34:57<35:51,  1.59it/s]
2|3047|Loss: 0.8476497530937195:  47%|████▋     | 3046/6470 [34:57<35:51,  1.59it/s]
2|3047|Loss: 0.8476497530937195:  47%|████▋     | 3047/6470 [34:57<38:12,  1.49it/s]
2|3048|Loss: 0.8588703274726868:  47%|████▋     | 3047/6470 [34:58<38:12,  1.49it/s]
2|3048|Loss: 0.8588703274726868:  47%|████▋     | 3048/6470 [34:58<36:45,  1.55it/s]
2|3049|Loss: 0.8608796000480652:  47%|████▋     | 3048/6470 [34:58<36:45,  1.55it/s]
2|3049|Loss: 0.8608796000480652:  47%|████▋     | 3049/6470 [34:59<38:17,  1.49it/s]
2|3050|Loss: 0.8101497292518616:  47%|████▋     | 3049/6470 [34:59<38:17,  1.49it/s]
2|3050|Loss: 0.8101497292518616:  47%|████▋     | 3050/6470 [34:59<39:44,  1.43it/s]
2|3051|Loss: 0.8617955446243286:  47%|████▋     | 3050/6470 [35:00<39:44,  1.43it/s]
2|3051|Loss: 0.8617955446243286:  47%|████▋     | 3051/6470 [35:00<38:51,  1.47it/s]
2|3052|Loss: 0.7990298271179199:  47%|████▋     | 3051/6470 [35:00<38:51,  1.47it/s]
2|3052|Loss: 0.7990298271179199:  47%|████▋     | 3052/6470 [35:01<36:27,  1.56it/s]
2|3053|Loss: 0.8406005501747131:  47%|████▋     | 3052/6470 [35:01<36:27,  1.56it/s]
2|3053|Loss: 0.8406005501747131:  47%|████▋     | 3053/6470 [35:01<36:58,  1.54it/s]
2|3054|Loss: 0.8389253616333008:  47%|████▋     | 3053/6470 [35:02<36:58,  1.54it/s]
2|3054|Loss: 0.8389253616333008:  47%|████▋     | 3054/6470 [35:02<37:08,  1.53it/s]
2|3055|Loss: 0.741324782371521:  47%|████▋     | 3054/6470 [35:02<37:08,  1.53it/s] 
2|3055|Loss: 0.741324782371521:  47%|████▋     | 3055/6470 [35:03<36:46,  1.55it/s]
2|3056|Loss: 0.7801012396812439:  47%|████▋     | 3055/6470 [35:03<36:46,  1.55it/s]
2|3056|Loss: 0.7801012396812439:  47%|████▋     | 3056/6470 [35:03<36:23,  1.56it/s]
2|3057|Loss: 0.8731812238693237:  47%|████▋     | 3056/6470 [35:04<36:23,  1.56it/s]
2|3057|Loss: 0.8731812238693237:  47%|████▋     | 3057/6470 [35:04<36:22,  1.56it/s]
2|3058|Loss: 0.7322832942008972:  47%|████▋     | 3057/6470 [35:04<36:22,  1.56it/s]
2|3058|Loss: 0.7322832942008972:  47%|████▋     | 3058/6470 [35:05<37:48,  1.50it/s]
2|3059|Loss: 0.9321532249450684:  47%|████▋     | 3058/6470 [35:05<37:48,  1.50it/s]
2|3059|Loss: 0.9321532249450684:  47%|████▋     | 3059/6470 [35:05<37:41,  1.51it/s]
2|3060|Loss: 0.9020268321037292:  47%|████▋     | 3059/6470 [35:06<37:41,  1.51it/s]
2|3060|Loss: 0.9020268321037292:  47%|████▋     | 3060/6470 [35:06<37:17,  1.52it/s]
2|3061|Loss: 0.8551232218742371:  47%|████▋     | 3060/6470 [35:06<37:17,  1.52it/s]
2|3061|Loss: 0.8551232218742371:  47%|████▋     | 3061/6470 [35:07<36:43,  1.55it/s]
2|3062|Loss: 1.016825795173645:  47%|████▋     | 3061/6470 [35:07<36:43,  1.55it/s] 
2|3062|Loss: 1.016825795173645:  47%|████▋     | 3062/6470 [35:07<38:02,  1.49it/s]
2|3063|Loss: 0.8100231885910034:  47%|████▋     | 3062/6470 [35:08<38:02,  1.49it/s]
2|3063|Loss: 0.8100231885910034:  47%|████▋     | 3063/6470 [35:08<40:17,  1.41it/s]
2|3064|Loss: 0.8897193670272827:  47%|████▋     | 3063/6470 [35:09<40:17,  1.41it/s]
2|3064|Loss: 0.8897193670272827:  47%|████▋     | 3064/6470 [35:09<41:17,  1.37it/s]
2|3065|Loss: 0.8898241519927979:  47%|████▋     | 3064/6470 [35:09<41:17,  1.37it/s]
2|3065|Loss: 0.8898241519927979:  47%|████▋     | 3065/6470 [35:10<41:02,  1.38it/s]
2|3066|Loss: 1.0148141384124756:  47%|████▋     | 3065/6470 [35:10<41:02,  1.38it/s]
2|3066|Loss: 1.0148141384124756:  47%|████▋     | 3066/6470 [35:10<40:42,  1.39it/s]
2|3067|Loss: 0.9483921527862549:  47%|████▋     | 3066/6470 [35:11<40:42,  1.39it/s]
2|3067|Loss: 0.9483921527862549:  47%|████▋     | 3067/6470 [35:11<38:53,  1.46it/s]
2|3068|Loss: 0.940839409828186:  47%|████▋     | 3067/6470 [35:11<38:53,  1.46it/s] 
2|3068|Loss: 0.940839409828186:  47%|████▋     | 3068/6470 [35:12<41:11,  1.38it/s]
2|3069|Loss: 0.8851892352104187:  47%|████▋     | 3068/6470 [35:12<41:11,  1.38it/s]
2|3069|Loss: 0.8851892352104187:  47%|████▋     | 3069/6470 [35:12<42:03,  1.35it/s]
2|3070|Loss: 0.9053214192390442:  47%|████▋     | 3069/6470 [35:13<42:03,  1.35it/s]
2|3070|Loss: 0.9053214192390442:  47%|████▋     | 3070/6470 [35:13<42:59,  1.32it/s]
2|3071|Loss: 0.8431295156478882:  47%|████▋     | 3070/6470 [35:14<42:59,  1.32it/s]
2|3071|Loss: 0.8431295156478882:  47%|████▋     | 3071/6470 [35:14<41:36,  1.36it/s]
2|3072|Loss: 0.7976946830749512:  47%|████▋     | 3071/6470 [35:14<41:36,  1.36it/s]
2|3072|Loss: 0.7976946830749512:  47%|████▋     | 3072/6470 [35:15<42:06,  1.35it/s]
2|3073|Loss: 0.7909641861915588:  47%|████▋     | 3072/6470 [35:15<42:06,  1.35it/s]
2|3073|Loss: 0.7909641861915588:  47%|████▋     | 3073/6470 [35:15<42:14,  1.34it/s]
2|3074|Loss: 0.6994370818138123:  47%|████▋     | 3073/6470 [35:16<42:14,  1.34it/s]
2|3074|Loss: 0.6994370818138123:  48%|████▊     | 3074/6470 [35:16<43:46,  1.29it/s]
2|3075|Loss: 0.8839992880821228:  48%|████▊     | 3074/6470 [35:17<43:46,  1.29it/s]
2|3075|Loss: 0.8839992880821228:  48%|████▊     | 3075/6470 [35:17<41:28,  1.36it/s]
2|3076|Loss: 0.8125247955322266:  48%|████▊     | 3075/6470 [35:17<41:28,  1.36it/s]
2|3076|Loss: 0.8125247955322266:  48%|████▊     | 3076/6470 [35:17<37:09,  1.52it/s]
2|3077|Loss: 1.0066556930541992:  48%|████▊     | 3076/6470 [35:18<37:09,  1.52it/s]
2|3077|Loss: 1.0066556930541992:  48%|████▊     | 3077/6470 [35:18<37:35,  1.50it/s]
2|3078|Loss: 0.9564573764801025:  48%|████▊     | 3077/6470 [35:18<37:35,  1.50it/s]
2|3078|Loss: 0.9564573764801025:  48%|████▊     | 3078/6470 [35:19<36:58,  1.53it/s]
2|3079|Loss: 0.9370405077934265:  48%|████▊     | 3078/6470 [35:19<36:58,  1.53it/s]
2|3079|Loss: 0.9370405077934265:  48%|████▊     | 3079/6470 [35:19<36:11,  1.56it/s]
2|3080|Loss: 0.9501632452011108:  48%|████▊     | 3079/6470 [35:20<36:11,  1.56it/s]
2|3080|Loss: 0.9501632452011108:  48%|████▊     | 3080/6470 [35:20<34:22,  1.64it/s]
2|3081|Loss: 0.8539242148399353:  48%|████▊     | 3080/6470 [35:20<34:22,  1.64it/s]
2|3081|Loss: 0.8539242148399353:  48%|████▊     | 3081/6470 [35:21<37:04,  1.52it/s]
2|3082|Loss: 0.9628279209136963:  48%|████▊     | 3081/6470 [35:21<37:04,  1.52it/s]
2|3082|Loss: 0.9628279209136963:  48%|████▊     | 3082/6470 [35:22<40:24,  1.40it/s]
2|3083|Loss: 0.711640477180481:  48%|████▊     | 3082/6470 [35:22<40:24,  1.40it/s] 
2|3083|Loss: 0.711640477180481:  48%|████▊     | 3083/6470 [35:22<39:49,  1.42it/s]
2|3084|Loss: 0.949321985244751:  48%|████▊     | 3083/6470 [35:23<39:49,  1.42it/s]
2|3084|Loss: 0.949321985244751:  48%|████▊     | 3084/6470 [35:23<38:36,  1.46it/s]
2|3085|Loss: 0.8559408783912659:  48%|████▊     | 3084/6470 [35:23<38:36,  1.46it/s]
2|3085|Loss: 0.8559408783912659:  48%|████▊     | 3085/6470 [35:23<36:21,  1.55it/s]
2|3086|Loss: 0.9556788206100464:  48%|████▊     | 3085/6470 [35:24<36:21,  1.55it/s]
2|3086|Loss: 0.9556788206100464:  48%|████▊     | 3086/6470 [35:24<35:52,  1.57it/s]
2|3087|Loss: 1.0218535661697388:  48%|████▊     | 3086/6470 [35:24<35:52,  1.57it/s]
2|3087|Loss: 1.0218535661697388:  48%|████▊     | 3087/6470 [35:25<36:51,  1.53it/s]
2|3088|Loss: 0.8185907602310181:  48%|████▊     | 3087/6470 [35:25<36:51,  1.53it/s]
2|3088|Loss: 0.8185907602310181:  48%|████▊     | 3088/6470 [35:25<38:29,  1.46it/s]
2|3089|Loss: 0.8834199905395508:  48%|████▊     | 3088/6470 [35:26<38:29,  1.46it/s]
2|3089|Loss: 0.8834199905395508:  48%|████▊     | 3089/6470 [35:26<39:22,  1.43it/s]
2|3090|Loss: 0.8107895851135254:  48%|████▊     | 3089/6470 [35:27<39:22,  1.43it/s]
2|3090|Loss: 0.8107895851135254:  48%|████▊     | 3090/6470 [35:27<39:26,  1.43it/s]
2|3091|Loss: 0.7970202565193176:  48%|████▊     | 3090/6470 [35:27<39:26,  1.43it/s]
2|3091|Loss: 0.7970202565193176:  48%|████▊     | 3091/6470 [35:28<38:40,  1.46it/s]
2|3092|Loss: 0.7650347948074341:  48%|████▊     | 3091/6470 [35:28<38:40,  1.46it/s]
2|3092|Loss: 0.7650347948074341:  48%|████▊     | 3092/6470 [35:28<40:24,  1.39it/s]
2|3093|Loss: 1.0679281949996948:  48%|████▊     | 3092/6470 [35:29<40:24,  1.39it/s]
2|3093|Loss: 1.0679281949996948:  48%|████▊     | 3093/6470 [35:29<39:02,  1.44it/s]
2|3094|Loss: 0.8901978135108948:  48%|████▊     | 3093/6470 [35:29<39:02,  1.44it/s]
2|3094|Loss: 0.8901978135108948:  48%|████▊     | 3094/6470 [35:30<37:37,  1.50it/s]
2|3095|Loss: 0.740717887878418:  48%|████▊     | 3094/6470 [35:30<37:37,  1.50it/s] 
2|3095|Loss: 0.740717887878418:  48%|████▊     | 3095/6470 [35:30<40:14,  1.40it/s]
2|3096|Loss: 0.9161788821220398:  48%|████▊     | 3095/6470 [35:31<40:14,  1.40it/s]
2|3096|Loss: 0.9161788821220398:  48%|████▊     | 3096/6470 [35:31<39:55,  1.41it/s]
2|3097|Loss: 0.8169293403625488:  48%|████▊     | 3096/6470 [35:31<39:55,  1.41it/s]
2|3097|Loss: 0.8169293403625488:  48%|████▊     | 3097/6470 [35:32<39:56,  1.41it/s]
2|3098|Loss: 0.7120983600616455:  48%|████▊     | 3097/6470 [35:32<39:56,  1.41it/s]
2|3098|Loss: 0.7120983600616455:  48%|████▊     | 3098/6470 [35:33<42:11,  1.33it/s]
2|3099|Loss: 0.8895296454429626:  48%|████▊     | 3098/6470 [35:33<42:11,  1.33it/s]
2|3099|Loss: 0.8895296454429626:  48%|████▊     | 3099/6470 [35:33<40:53,  1.37it/s]
2|3100|Loss: 0.8278006911277771:  48%|████▊     | 3099/6470 [35:34<40:53,  1.37it/s]
2|3100|Loss: 0.8278006911277771:  48%|████▊     | 3100/6470 [35:34<40:07,  1.40it/s]
2|3101|Loss: 0.9489957094192505:  48%|████▊     | 3100/6470 [35:34<40:07,  1.40it/s]
2|3101|Loss: 0.9489957094192505:  48%|████▊     | 3101/6470 [35:35<38:11,  1.47it/s]
2|3102|Loss: 0.7846935987472534:  48%|████▊     | 3101/6470 [35:35<38:11,  1.47it/s]
2|3102|Loss: 0.7846935987472534:  48%|████▊     | 3102/6470 [35:35<38:07,  1.47it/s]
2|3103|Loss: 0.8867383003234863:  48%|████▊     | 3102/6470 [35:36<38:07,  1.47it/s]
2|3103|Loss: 0.8867383003234863:  48%|████▊     | 3103/6470 [35:36<35:07,  1.60it/s]
2|3104|Loss: 0.9496077299118042:  48%|████▊     | 3103/6470 [35:36<35:07,  1.60it/s]
2|3104|Loss: 0.9496077299118042:  48%|████▊     | 3104/6470 [35:36<33:27,  1.68it/s]
2|3105|Loss: 0.7868301272392273:  48%|████▊     | 3104/6470 [35:37<33:27,  1.68it/s]
2|3105|Loss: 0.7868301272392273:  48%|████▊     | 3105/6470 [35:37<36:59,  1.52it/s]
2|3106|Loss: 0.9281041026115417:  48%|████▊     | 3105/6470 [35:38<36:59,  1.52it/s]
2|3106|Loss: 0.9281041026115417:  48%|████▊     | 3106/6470 [35:38<37:15,  1.50it/s]
2|3107|Loss: 0.7542401552200317:  48%|████▊     | 3106/6470 [35:38<37:15,  1.50it/s]
2|3107|Loss: 0.7542401552200317:  48%|████▊     | 3107/6470 [35:38<37:42,  1.49it/s]
2|3108|Loss: 0.9542272090911865:  48%|████▊     | 3107/6470 [35:39<37:42,  1.49it/s]
2|3108|Loss: 0.9542272090911865:  48%|████▊     | 3108/6470 [35:39<38:48,  1.44it/s]
2|3109|Loss: 0.9017704725265503:  48%|████▊     | 3108/6470 [35:40<38:48,  1.44it/s]
2|3109|Loss: 0.9017704725265503:  48%|████▊     | 3109/6470 [35:40<38:08,  1.47it/s]
2|3110|Loss: 0.7813553810119629:  48%|████▊     | 3109/6470 [35:40<38:08,  1.47it/s]
2|3110|Loss: 0.7813553810119629:  48%|████▊     | 3110/6470 [35:41<38:56,  1.44it/s]
2|3111|Loss: 0.8541916608810425:  48%|████▊     | 3110/6470 [35:41<38:56,  1.44it/s]
2|3111|Loss: 0.8541916608810425:  48%|████▊     | 3111/6470 [35:41<38:16,  1.46it/s]
2|3112|Loss: 0.7768144011497498:  48%|████▊     | 3111/6470 [35:42<38:16,  1.46it/s]
2|3112|Loss: 0.7768144011497498:  48%|████▊     | 3112/6470 [35:42<36:43,  1.52it/s]
2|3113|Loss: 1.1018579006195068:  48%|████▊     | 3112/6470 [35:42<36:43,  1.52it/s]
2|3113|Loss: 1.1018579006195068:  48%|████▊     | 3113/6470 [35:43<39:12,  1.43it/s]
2|3114|Loss: 0.8731475472450256:  48%|████▊     | 3113/6470 [35:43<39:12,  1.43it/s]
2|3114|Loss: 0.8731475472450256:  48%|████▊     | 3114/6470 [35:43<41:03,  1.36it/s]
2|3115|Loss: 0.8305570483207703:  48%|████▊     | 3114/6470 [35:44<41:03,  1.36it/s]
2|3115|Loss: 0.8305570483207703:  48%|████▊     | 3115/6470 [35:44<39:45,  1.41it/s]
2|3116|Loss: 0.9173356294631958:  48%|████▊     | 3115/6470 [35:45<39:45,  1.41it/s]
2|3116|Loss: 0.9173356294631958:  48%|████▊     | 3116/6470 [35:45<39:16,  1.42it/s]
2|3117|Loss: 1.0043892860412598:  48%|████▊     | 3116/6470 [35:45<39:16,  1.42it/s]
2|3117|Loss: 1.0043892860412598:  48%|████▊     | 3117/6470 [35:45<38:14,  1.46it/s]
2|3118|Loss: 0.8679099678993225:  48%|████▊     | 3117/6470 [35:46<38:14,  1.46it/s]
2|3118|Loss: 0.8679099678993225:  48%|████▊     | 3118/6470 [35:46<37:06,  1.51it/s]
2|3119|Loss: 0.8588942289352417:  48%|████▊     | 3118/6470 [35:46<37:06,  1.51it/s]
2|3119|Loss: 0.8588942289352417:  48%|████▊     | 3119/6470 [35:47<37:37,  1.48it/s]
2|3120|Loss: 0.8626833558082581:  48%|████▊     | 3119/6470 [35:47<37:37,  1.48it/s]
2|3120|Loss: 0.8626833558082581:  48%|████▊     | 3120/6470 [35:47<38:33,  1.45it/s]
2|3121|Loss: 0.8267821669578552:  48%|████▊     | 3120/6470 [35:48<38:33,  1.45it/s]
2|3121|Loss: 0.8267821669578552:  48%|████▊     | 3121/6470 [35:48<39:24,  1.42it/s]
2|3122|Loss: 0.9652276635169983:  48%|████▊     | 3121/6470 [35:49<39:24,  1.42it/s]
2|3122|Loss: 0.9652276635169983:  48%|████▊     | 3122/6470 [35:49<37:34,  1.49it/s]
2|3123|Loss: 0.7924076318740845:  48%|████▊     | 3122/6470 [35:49<37:34,  1.49it/s]
2|3123|Loss: 0.7924076318740845:  48%|████▊     | 3123/6470 [35:49<36:47,  1.52it/s]
2|3124|Loss: 0.9445043206214905:  48%|████▊     | 3123/6470 [35:50<36:47,  1.52it/s]
2|3124|Loss: 0.9445043206214905:  48%|████▊     | 3124/6470 [35:50<37:58,  1.47it/s]
2|3125|Loss: 0.7865497469902039:  48%|████▊     | 3124/6470 [35:51<37:58,  1.47it/s]
2|3125|Loss: 0.7865497469902039:  48%|████▊     | 3125/6470 [35:51<40:38,  1.37it/s]
2|3126|Loss: 0.793866753578186:  48%|████▊     | 3125/6470 [35:51<40:38,  1.37it/s] 
2|3126|Loss: 0.793866753578186:  48%|████▊     | 3126/6470 [35:52<41:32,  1.34it/s]
2|3127|Loss: 0.8921999931335449:  48%|████▊     | 3126/6470 [35:52<41:32,  1.34it/s]
2|3127|Loss: 0.8921999931335449:  48%|████▊     | 3127/6470 [35:53<41:27,  1.34it/s]
2|3128|Loss: 1.008009672164917:  48%|████▊     | 3127/6470 [35:53<41:27,  1.34it/s] 
2|3128|Loss: 1.008009672164917:  48%|████▊     | 3128/6470 [35:53<40:52,  1.36it/s]
2|3129|Loss: 0.8150594234466553:  48%|████▊     | 3128/6470 [35:54<40:52,  1.36it/s]
2|3129|Loss: 0.8150594234466553:  48%|████▊     | 3129/6470 [35:54<40:27,  1.38it/s]
2|3130|Loss: 0.8817377686500549:  48%|████▊     | 3129/6470 [35:54<40:27,  1.38it/s]
2|3130|Loss: 0.8817377686500549:  48%|████▊     | 3130/6470 [35:55<39:59,  1.39it/s]
2|3131|Loss: 0.8559480309486389:  48%|████▊     | 3130/6470 [35:55<39:59,  1.39it/s]
2|3131|Loss: 0.8559480309486389:  48%|████▊     | 3131/6470 [35:55<40:49,  1.36it/s]
2|3132|Loss: 0.8177201747894287:  48%|████▊     | 3131/6470 [35:56<40:49,  1.36it/s]
2|3132|Loss: 0.8177201747894287:  48%|████▊     | 3132/6470 [35:56<39:30,  1.41it/s]
2|3133|Loss: 0.9305665493011475:  48%|████▊     | 3132/6470 [35:56<39:30,  1.41it/s]
2|3133|Loss: 0.9305665493011475:  48%|████▊     | 3133/6470 [35:57<39:04,  1.42it/s]
2|3134|Loss: 0.8384732604026794:  48%|████▊     | 3133/6470 [35:57<39:04,  1.42it/s]
2|3134|Loss: 0.8384732604026794:  48%|████▊     | 3134/6470 [35:57<36:28,  1.52it/s]
2|3135|Loss: 0.8436468839645386:  48%|████▊     | 3134/6470 [35:58<36:28,  1.52it/s]
2|3135|Loss: 0.8436468839645386:  48%|████▊     | 3135/6470 [35:58<37:00,  1.50it/s]
2|3136|Loss: 0.841285765171051:  48%|████▊     | 3135/6470 [35:58<37:00,  1.50it/s] 
2|3136|Loss: 0.841285765171051:  48%|████▊     | 3136/6470 [35:59<38:24,  1.45it/s]
2|3137|Loss: 0.7858932614326477:  48%|████▊     | 3136/6470 [35:59<38:24,  1.45it/s]
2|3137|Loss: 0.7858932614326477:  48%|████▊     | 3137/6470 [35:59<37:14,  1.49it/s]
2|3138|Loss: 0.8781188130378723:  48%|████▊     | 3137/6470 [36:00<37:14,  1.49it/s]
2|3138|Loss: 0.8781188130378723:  49%|████▊     | 3138/6470 [36:00<37:24,  1.48it/s]
2|3139|Loss: 0.8228098750114441:  49%|████▊     | 3138/6470 [36:01<37:24,  1.48it/s]
2|3139|Loss: 0.8228098750114441:  49%|████▊     | 3139/6470 [36:01<39:58,  1.39it/s]
2|3140|Loss: 0.9301272630691528:  49%|████▊     | 3139/6470 [36:01<39:58,  1.39it/s]
2|3140|Loss: 0.9301272630691528:  49%|████▊     | 3140/6470 [36:02<40:42,  1.36it/s]
2|3141|Loss: 0.8811300992965698:  49%|████▊     | 3140/6470 [36:02<40:42,  1.36it/s]
2|3141|Loss: 0.8811300992965698:  49%|████▊     | 3141/6470 [36:02<40:43,  1.36it/s]
2|3142|Loss: 0.8492691516876221:  49%|████▊     | 3141/6470 [36:03<40:43,  1.36it/s]
2|3142|Loss: 0.8492691516876221:  49%|████▊     | 3142/6470 [36:03<38:37,  1.44it/s]
2|3143|Loss: 0.9740859866142273:  49%|████▊     | 3142/6470 [36:03<38:37,  1.44it/s]
2|3143|Loss: 0.9740859866142273:  49%|████▊     | 3143/6470 [36:04<36:45,  1.51it/s]
2|3144|Loss: 0.8772597908973694:  49%|████▊     | 3143/6470 [36:04<36:45,  1.51it/s]
2|3144|Loss: 0.8772597908973694:  49%|████▊     | 3144/6470 [36:04<35:00,  1.58it/s]
2|3145|Loss: 0.8237603902816772:  49%|████▊     | 3144/6470 [36:05<35:00,  1.58it/s]
2|3145|Loss: 0.8237603902816772:  49%|████▊     | 3145/6470 [36:05<37:21,  1.48it/s]
2|3146|Loss: 0.8400951027870178:  49%|████▊     | 3145/6470 [36:05<37:21,  1.48it/s]
2|3146|Loss: 0.8400951027870178:  49%|████▊     | 3146/6470 [36:06<35:46,  1.55it/s]
2|3147|Loss: 0.8002318739891052:  49%|████▊     | 3146/6470 [36:06<35:46,  1.55it/s]
2|3147|Loss: 0.8002318739891052:  49%|████▊     | 3147/6470 [36:06<36:28,  1.52it/s]
2|3148|Loss: 0.8591598868370056:  49%|████▊     | 3147/6470 [36:07<36:28,  1.52it/s]
2|3148|Loss: 0.8591598868370056:  49%|████▊     | 3148/6470 [36:07<36:40,  1.51it/s]
2|3149|Loss: 0.8028262853622437:  49%|████▊     | 3148/6470 [36:07<36:40,  1.51it/s]
2|3149|Loss: 0.8028262853622437:  49%|████▊     | 3149/6470 [36:08<37:05,  1.49it/s]
2|3150|Loss: 0.9635524153709412:  49%|████▊     | 3149/6470 [36:08<37:05,  1.49it/s]
2|3150|Loss: 0.9635524153709412:  49%|████▊     | 3150/6470 [36:08<39:57,  1.39it/s]
2|3151|Loss: 0.7910283207893372:  49%|████▊     | 3150/6470 [36:09<39:57,  1.39it/s]
2|3151|Loss: 0.7910283207893372:  49%|████▊     | 3151/6470 [36:09<40:20,  1.37it/s]
2|3152|Loss: 0.8021340370178223:  49%|████▊     | 3151/6470 [36:10<40:20,  1.37it/s]
2|3152|Loss: 0.8021340370178223:  49%|████▊     | 3152/6470 [36:10<39:06,  1.41it/s]
2|3153|Loss: 0.8667358756065369:  49%|████▊     | 3152/6470 [36:10<39:06,  1.41it/s]
2|3153|Loss: 0.8667358756065369:  49%|████▊     | 3153/6470 [36:11<39:13,  1.41it/s]
2|3154|Loss: 0.7814605236053467:  49%|████▊     | 3153/6470 [36:11<39:13,  1.41it/s]
2|3154|Loss: 0.7814605236053467:  49%|████▊     | 3154/6470 [36:11<38:26,  1.44it/s]
2|3155|Loss: 0.706565797328949:  49%|████▊     | 3154/6470 [36:11<38:26,  1.44it/s] 
2|3155|Loss: 0.706565797328949:  49%|████▉     | 3155/6470 [36:12<35:02,  1.58it/s]
2|3156|Loss: 0.927663266658783:  49%|████▉     | 3155/6470 [36:12<35:02,  1.58it/s]
2|3156|Loss: 0.927663266658783:  49%|████▉     | 3156/6470 [36:12<34:10,  1.62it/s]
2|3157|Loss: 0.8775803446769714:  49%|████▉     | 3156/6470 [36:13<34:10,  1.62it/s]
2|3157|Loss: 0.8775803446769714:  49%|████▉     | 3157/6470 [36:13<36:34,  1.51it/s]
2|3158|Loss: 0.8722648620605469:  49%|████▉     | 3157/6470 [36:13<36:34,  1.51it/s]
2|3158|Loss: 0.8722648620605469:  49%|████▉     | 3158/6470 [36:14<34:18,  1.61it/s]
2|3159|Loss: 0.7987570762634277:  49%|████▉     | 3158/6470 [36:14<34:18,  1.61it/s]
2|3159|Loss: 0.7987570762634277:  49%|████▉     | 3159/6470 [36:14<35:59,  1.53it/s]
2|3160|Loss: 0.9439657330513:  49%|████▉     | 3159/6470 [36:15<35:59,  1.53it/s]   
2|3160|Loss: 0.9439657330513:  49%|████▉     | 3160/6470 [36:15<37:11,  1.48it/s]
2|3161|Loss: 0.769895613193512:  49%|████▉     | 3160/6470 [36:15<37:11,  1.48it/s]
2|3161|Loss: 0.769895613193512:  49%|████▉     | 3161/6470 [36:16<38:20,  1.44it/s]
2|3162|Loss: 0.8320100903511047:  49%|████▉     | 3161/6470 [36:16<38:20,  1.44it/s]
2|3162|Loss: 0.8320100903511047:  49%|████▉     | 3162/6470 [36:17<39:23,  1.40it/s]
2|3163|Loss: 0.9179337620735168:  49%|████▉     | 3162/6470 [36:17<39:23,  1.40it/s]
2|3163|Loss: 0.9179337620735168:  49%|████▉     | 3163/6470 [36:17<42:19,  1.30it/s]
2|3164|Loss: 0.8520483374595642:  49%|████▉     | 3163/6470 [36:18<42:19,  1.30it/s]
2|3164|Loss: 0.8520483374595642:  49%|████▉     | 3164/6470 [36:18<44:03,  1.25it/s]
2|3165|Loss: 0.6970046162605286:  49%|████▉     | 3164/6470 [36:19<44:03,  1.25it/s]
2|3165|Loss: 0.6970046162605286:  49%|████▉     | 3165/6470 [36:19<44:13,  1.25it/s]
2|3166|Loss: 0.6868017911911011:  49%|████▉     | 3165/6470 [36:20<44:13,  1.25it/s]
2|3166|Loss: 0.6868017911911011:  49%|████▉     | 3166/6470 [36:20<44:06,  1.25it/s]
2|3167|Loss: 0.9241331815719604:  49%|████▉     | 3166/6470 [36:20<44:06,  1.25it/s]
2|3167|Loss: 0.9241331815719604:  49%|████▉     | 3167/6470 [36:21<41:47,  1.32it/s]
2|3168|Loss: 0.7541252374649048:  49%|████▉     | 3167/6470 [36:21<41:47,  1.32it/s]
2|3168|Loss: 0.7541252374649048:  49%|████▉     | 3168/6470 [36:21<41:10,  1.34it/s]
2|3169|Loss: 0.9053606390953064:  49%|████▉     | 3168/6470 [36:22<41:10,  1.34it/s]
2|3169|Loss: 0.9053606390953064:  49%|████▉     | 3169/6470 [36:22<42:37,  1.29it/s]
2|3170|Loss: 0.9646210670471191:  49%|████▉     | 3169/6470 [36:23<42:37,  1.29it/s]
2|3170|Loss: 0.9646210670471191:  49%|████▉     | 3170/6470 [36:23<42:28,  1.30it/s]
2|3171|Loss: 0.7782942652702332:  49%|████▉     | 3170/6470 [36:23<42:28,  1.30it/s]
2|3171|Loss: 0.7782942652702332:  49%|████▉     | 3171/6470 [36:23<38:06,  1.44it/s]
2|3172|Loss: 0.943295955657959:  49%|████▉     | 3171/6470 [36:24<38:06,  1.44it/s] 
2|3172|Loss: 0.943295955657959:  49%|████▉     | 3172/6470 [36:24<35:09,  1.56it/s]
2|3173|Loss: 0.7486146092414856:  49%|████▉     | 3172/6470 [36:24<35:09,  1.56it/s]
2|3173|Loss: 0.7486146092414856:  49%|████▉     | 3173/6470 [36:25<35:41,  1.54it/s]
2|3174|Loss: 0.9571285843849182:  49%|████▉     | 3173/6470 [36:25<35:41,  1.54it/s]
2|3174|Loss: 0.9571285843849182:  49%|████▉     | 3174/6470 [36:25<37:03,  1.48it/s]
2|3175|Loss: 0.8887721300125122:  49%|████▉     | 3174/6470 [36:26<37:03,  1.48it/s]
2|3175|Loss: 0.8887721300125122:  49%|████▉     | 3175/6470 [36:26<36:33,  1.50it/s]
2|3176|Loss: 1.0309473276138306:  49%|████▉     | 3175/6470 [36:26<36:33,  1.50it/s]
2|3176|Loss: 1.0309473276138306:  49%|████▉     | 3176/6470 [36:27<37:15,  1.47it/s]
2|3177|Loss: 0.9799098372459412:  49%|████▉     | 3176/6470 [36:27<37:15,  1.47it/s]
2|3177|Loss: 0.9799098372459412:  49%|████▉     | 3177/6470 [36:27<36:04,  1.52it/s]
2|3178|Loss: 0.8687717914581299:  49%|████▉     | 3177/6470 [36:28<36:04,  1.52it/s]
2|3178|Loss: 0.8687717914581299:  49%|████▉     | 3178/6470 [36:28<37:05,  1.48it/s]
2|3179|Loss: 0.714920163154602:  49%|████▉     | 3178/6470 [36:28<37:05,  1.48it/s] 
2|3179|Loss: 0.714920163154602:  49%|████▉     | 3179/6470 [36:29<41:03,  1.34it/s]
2|3180|Loss: 0.8467661738395691:  49%|████▉     | 3179/6470 [36:29<41:03,  1.34it/s]
2|3180|Loss: 0.8467661738395691:  49%|████▉     | 3180/6470 [36:30<38:56,  1.41it/s]
2|3181|Loss: 0.8323163986206055:  49%|████▉     | 3180/6470 [36:30<38:56,  1.41it/s]
2|3181|Loss: 0.8323163986206055:  49%|████▉     | 3181/6470 [36:30<40:10,  1.36it/s]
2|3182|Loss: 0.8076473474502563:  49%|████▉     | 3181/6470 [36:31<40:10,  1.36it/s]
2|3182|Loss: 0.8076473474502563:  49%|████▉     | 3182/6470 [36:31<38:16,  1.43it/s]
2|3183|Loss: 0.6793837547302246:  49%|████▉     | 3182/6470 [36:31<38:16,  1.43it/s]
2|3183|Loss: 0.6793837547302246:  49%|████▉     | 3183/6470 [36:31<36:08,  1.52it/s]
2|3184|Loss: 0.8273082375526428:  49%|████▉     | 3183/6470 [36:32<36:08,  1.52it/s]
2|3184|Loss: 0.8273082375526428:  49%|████▉     | 3184/6470 [36:32<36:25,  1.50it/s]
2|3185|Loss: 0.838051974773407:  49%|████▉     | 3184/6470 [36:33<36:25,  1.50it/s] 
2|3185|Loss: 0.838051974773407:  49%|████▉     | 3185/6470 [36:33<38:39,  1.42it/s]
2|3186|Loss: 0.7691657543182373:  49%|████▉     | 3185/6470 [36:33<38:39,  1.42it/s]
2|3186|Loss: 0.7691657543182373:  49%|████▉     | 3186/6470 [36:34<37:55,  1.44it/s]
2|3187|Loss: 0.9086995720863342:  49%|████▉     | 3186/6470 [36:34<37:55,  1.44it/s]
2|3187|Loss: 0.9086995720863342:  49%|████▉     | 3187/6470 [36:34<39:04,  1.40it/s]
2|3188|Loss: 0.706134021282196:  49%|████▉     | 3187/6470 [36:35<39:04,  1.40it/s] 
2|3188|Loss: 0.706134021282196:  49%|████▉     | 3188/6470 [36:35<42:29,  1.29it/s]
2|3189|Loss: 0.9274751543998718:  49%|████▉     | 3188/6470 [36:36<42:29,  1.29it/s]
2|3189|Loss: 0.9274751543998718:  49%|████▉     | 3189/6470 [36:36<40:33,  1.35it/s]
2|3190|Loss: 0.9212620854377747:  49%|████▉     | 3189/6470 [36:36<40:33,  1.35it/s]
2|3190|Loss: 0.9212620854377747:  49%|████▉     | 3190/6470 [36:37<40:03,  1.36it/s]
2|3191|Loss: 0.7693008184432983:  49%|████▉     | 3190/6470 [36:37<40:03,  1.36it/s]
2|3191|Loss: 0.7693008184432983:  49%|████▉     | 3191/6470 [36:37<40:28,  1.35it/s]
2|3192|Loss: 0.7782087326049805:  49%|████▉     | 3191/6470 [36:38<40:28,  1.35it/s]
2|3192|Loss: 0.7782087326049805:  49%|████▉     | 3192/6470 [36:38<40:26,  1.35it/s]
2|3193|Loss: 0.856987714767456:  49%|████▉     | 3192/6470 [36:39<40:26,  1.35it/s] 
2|3193|Loss: 0.856987714767456:  49%|████▉     | 3193/6470 [36:39<39:06,  1.40it/s]
2|3194|Loss: 0.9529396891593933:  49%|████▉     | 3193/6470 [36:39<39:06,  1.40it/s]
2|3194|Loss: 0.9529396891593933:  49%|████▉     | 3194/6470 [36:39<37:35,  1.45it/s]
2|3195|Loss: 0.8730731010437012:  49%|████▉     | 3194/6470 [36:40<37:35,  1.45it/s]
2|3195|Loss: 0.8730731010437012:  49%|████▉     | 3195/6470 [36:40<38:55,  1.40it/s]
2|3196|Loss: 0.8972704410552979:  49%|████▉     | 3195/6470 [36:41<38:55,  1.40it/s]
2|3196|Loss: 0.8972704410552979:  49%|████▉     | 3196/6470 [36:41<38:01,  1.44it/s]
2|3197|Loss: 0.9397186636924744:  49%|████▉     | 3196/6470 [36:41<38:01,  1.44it/s]
2|3197|Loss: 0.9397186636924744:  49%|████▉     | 3197/6470 [36:42<38:50,  1.40it/s]
2|3198|Loss: 0.769180178642273:  49%|████▉     | 3197/6470 [36:42<38:50,  1.40it/s] 
2|3198|Loss: 0.769180178642273:  49%|████▉     | 3198/6470 [36:42<38:57,  1.40it/s]
2|3199|Loss: 0.984268307685852:  49%|████▉     | 3198/6470 [36:43<38:57,  1.40it/s]
2|3199|Loss: 0.984268307685852:  49%|████▉     | 3199/6470 [36:43<37:53,  1.44it/s]
2|3200|Loss: 0.9561546444892883:  49%|████▉     | 3199/6470 [36:43<37:53,  1.44it/s]
2|3200|Loss: 0.9561546444892883:  49%|████▉     | 3200/6470 [36:44<38:03,  1.43it/s]
2|3201|Loss: 0.9450663924217224:  49%|████▉     | 3200/6470 [36:44<38:03,  1.43it/s]
2|3201|Loss: 0.9450663924217224:  49%|████▉     | 3201/6470 [36:45<40:11,  1.36it/s]
2|3202|Loss: 0.9723573923110962:  49%|████▉     | 3201/6470 [36:45<40:11,  1.36it/s]
2|3202|Loss: 0.9723573923110962:  49%|████▉     | 3202/6470 [36:45<42:13,  1.29it/s]
2|3203|Loss: 0.7684260010719299:  49%|████▉     | 3202/6470 [36:46<42:13,  1.29it/s]
2|3203|Loss: 0.7684260010719299:  50%|████▉     | 3203/6470 [36:46<38:16,  1.42it/s]
2|3204|Loss: 0.8283303380012512:  50%|████▉     | 3203/6470 [36:46<38:16,  1.42it/s]
2|3204|Loss: 0.8283303380012512:  50%|████▉     | 3204/6470 [36:46<34:50,  1.56it/s]
2|3205|Loss: 0.806701123714447:  50%|████▉     | 3204/6470 [36:47<34:50,  1.56it/s] 
2|3205|Loss: 0.806701123714447:  50%|████▉     | 3205/6470 [36:47<34:26,  1.58it/s]
2|3206|Loss: 0.6932699084281921:  50%|████▉     | 3205/6470 [36:47<34:26,  1.58it/s]
2|3206|Loss: 0.6932699084281921:  50%|████▉     | 3206/6470 [36:48<35:23,  1.54it/s]
2|3207|Loss: 0.8123435974121094:  50%|████▉     | 3206/6470 [36:48<35:23,  1.54it/s]
2|3207|Loss: 0.8123435974121094:  50%|████▉     | 3207/6470 [36:48<36:39,  1.48it/s]
2|3208|Loss: 0.7458487153053284:  50%|████▉     | 3207/6470 [36:49<36:39,  1.48it/s]
2|3208|Loss: 0.7458487153053284:  50%|████▉     | 3208/6470 [36:49<39:33,  1.37it/s]
2|3209|Loss: 0.7021624445915222:  50%|████▉     | 3208/6470 [36:50<39:33,  1.37it/s]
2|3209|Loss: 0.7021624445915222:  50%|████▉     | 3209/6470 [36:50<40:13,  1.35it/s]
2|3210|Loss: 0.8515811562538147:  50%|████▉     | 3209/6470 [36:50<40:13,  1.35it/s]
2|3210|Loss: 0.8515811562538147:  50%|████▉     | 3210/6470 [36:51<36:27,  1.49it/s]
2|3211|Loss: 0.8357223272323608:  50%|████▉     | 3210/6470 [36:51<36:27,  1.49it/s]
2|3211|Loss: 0.8357223272323608:  50%|████▉     | 3211/6470 [36:51<35:15,  1.54it/s]
2|3212|Loss: 0.9062048196792603:  50%|████▉     | 3211/6470 [36:52<35:15,  1.54it/s]
2|3212|Loss: 0.9062048196792603:  50%|████▉     | 3212/6470 [36:52<36:17,  1.50it/s]
2|3213|Loss: 0.8640219569206238:  50%|████▉     | 3212/6470 [36:52<36:17,  1.50it/s]
2|3213|Loss: 0.8640219569206238:  50%|████▉     | 3213/6470 [36:53<35:29,  1.53it/s]
2|3214|Loss: 0.9758490324020386:  50%|████▉     | 3213/6470 [36:53<35:29,  1.53it/s]
2|3214|Loss: 0.9758490324020386:  50%|████▉     | 3214/6470 [36:53<36:18,  1.49it/s]
2|3215|Loss: 0.8697804808616638:  50%|████▉     | 3214/6470 [36:54<36:18,  1.49it/s]
2|3215|Loss: 0.8697804808616638:  50%|████▉     | 3215/6470 [36:54<35:48,  1.51it/s]
2|3216|Loss: 0.7546309232711792:  50%|████▉     | 3215/6470 [36:54<35:48,  1.51it/s]
2|3216|Loss: 0.7546309232711792:  50%|████▉     | 3216/6470 [36:55<36:46,  1.47it/s]
2|3217|Loss: 0.9097086787223816:  50%|████▉     | 3216/6470 [36:55<36:46,  1.47it/s]
2|3217|Loss: 0.9097086787223816:  50%|████▉     | 3217/6470 [36:55<39:38,  1.37it/s]
2|3218|Loss: 0.856863796710968:  50%|████▉     | 3217/6470 [36:56<39:38,  1.37it/s] 
2|3218|Loss: 0.856863796710968:  50%|████▉     | 3218/6470 [36:56<39:02,  1.39it/s]
2|3219|Loss: 0.9017866253852844:  50%|████▉     | 3218/6470 [36:57<39:02,  1.39it/s]
2|3219|Loss: 0.9017866253852844:  50%|████▉     | 3219/6470 [36:57<37:27,  1.45it/s]
2|3220|Loss: 0.8943135738372803:  50%|████▉     | 3219/6470 [36:57<37:27,  1.45it/s]
2|3220|Loss: 0.8943135738372803:  50%|████▉     | 3220/6470 [36:57<36:31,  1.48it/s]
2|3221|Loss: 0.951392650604248:  50%|████▉     | 3220/6470 [36:58<36:31,  1.48it/s] 
2|3221|Loss: 0.951392650604248:  50%|████▉     | 3221/6470 [36:58<34:15,  1.58it/s]
2|3222|Loss: 0.7376968860626221:  50%|████▉     | 3221/6470 [36:58<34:15,  1.58it/s]
2|3222|Loss: 0.7376968860626221:  50%|████▉     | 3222/6470 [36:58<31:31,  1.72it/s]
2|3223|Loss: 0.66485995054245:  50%|████▉     | 3222/6470 [36:59<31:31,  1.72it/s]  
2|3223|Loss: 0.66485995054245:  50%|████▉     | 3223/6470 [36:59<30:01,  1.80it/s]
2|3224|Loss: 0.8180364370346069:  50%|████▉     | 3223/6470 [36:59<30:01,  1.80it/s]
2|3224|Loss: 0.8180364370346069:  50%|████▉     | 3224/6470 [37:00<32:08,  1.68it/s]
2|3225|Loss: 0.9327517747879028:  50%|████▉     | 3224/6470 [37:00<32:08,  1.68it/s]
2|3225|Loss: 0.9327517747879028:  50%|████▉     | 3225/6470 [37:00<32:35,  1.66it/s]
2|3226|Loss: 0.851640522480011:  50%|████▉     | 3225/6470 [37:01<32:35,  1.66it/s] 
2|3226|Loss: 0.851640522480011:  50%|████▉     | 3226/6470 [37:01<32:14,  1.68it/s]
2|3227|Loss: 0.8252490758895874:  50%|████▉     | 3226/6470 [37:01<32:14,  1.68it/s]
2|3227|Loss: 0.8252490758895874:  50%|████▉     | 3227/6470 [37:01<33:35,  1.61it/s]
2|3228|Loss: 0.9065935611724854:  50%|████▉     | 3227/6470 [37:02<33:35,  1.61it/s]
2|3228|Loss: 0.9065935611724854:  50%|████▉     | 3228/6470 [37:02<35:32,  1.52it/s]
2|3229|Loss: 0.76175856590271:  50%|████▉     | 3228/6470 [37:03<35:32,  1.52it/s]  
2|3229|Loss: 0.76175856590271:  50%|████▉     | 3229/6470 [37:03<35:19,  1.53it/s]
2|3230|Loss: 0.7328470349311829:  50%|████▉     | 3229/6470 [37:03<35:19,  1.53it/s]
2|3230|Loss: 0.7328470349311829:  50%|████▉     | 3230/6470 [37:04<37:05,  1.46it/s]
2|3231|Loss: 0.9111084342002869:  50%|████▉     | 3230/6470 [37:04<37:05,  1.46it/s]
2|3231|Loss: 0.9111084342002869:  50%|████▉     | 3231/6470 [37:04<34:28,  1.57it/s]
2|3232|Loss: 0.9900720119476318:  50%|████▉     | 3231/6470 [37:04<34:28,  1.57it/s]
2|3232|Loss: 0.9900720119476318:  50%|████▉     | 3232/6470 [37:05<33:33,  1.61it/s]
2|3233|Loss: 0.8185039162635803:  50%|████▉     | 3232/6470 [37:05<33:33,  1.61it/s]
2|3233|Loss: 0.8185039162635803:  50%|████▉     | 3233/6470 [37:05<34:57,  1.54it/s]
2|3234|Loss: 0.8738771677017212:  50%|████▉     | 3233/6470 [37:06<34:57,  1.54it/s]
2|3234|Loss: 0.8738771677017212:  50%|████▉     | 3234/6470 [37:06<35:51,  1.50it/s]
2|3235|Loss: 0.8515321612358093:  50%|████▉     | 3234/6470 [37:06<35:51,  1.50it/s]
2|3235|Loss: 0.8515321612358093:  50%|█████     | 3235/6470 [37:07<32:15,  1.67it/s]
2|3236|Loss: 0.967594563961029:  50%|█████     | 3235/6470 [37:07<32:15,  1.67it/s] 
2|3236|Loss: 0.967594563961029:  50%|█████     | 3236/6470 [37:07<33:24,  1.61it/s]
2|3237|Loss: 0.8959505558013916:  50%|█████     | 3236/6470 [37:08<33:24,  1.61it/s]
2|3237|Loss: 0.8959505558013916:  50%|█████     | 3237/6470 [37:08<34:12,  1.58it/s]
2|3238|Loss: 0.8852331638336182:  50%|█████     | 3237/6470 [37:08<34:12,  1.58it/s]
2|3238|Loss: 0.8852331638336182:  50%|█████     | 3238/6470 [37:09<35:12,  1.53it/s]
2|3239|Loss: 0.8761643171310425:  50%|█████     | 3238/6470 [37:09<35:12,  1.53it/s]
2|3239|Loss: 0.8761643171310425:  50%|█████     | 3239/6470 [37:09<35:09,  1.53it/s]
2|3240|Loss: 0.9727213978767395:  50%|█████     | 3239/6470 [37:10<35:09,  1.53it/s]
2|3240|Loss: 0.9727213978767395:  50%|█████     | 3240/6470 [37:10<32:54,  1.64it/s]
2|3241|Loss: 0.8686780333518982:  50%|█████     | 3240/6470 [37:10<32:54,  1.64it/s]
2|3241|Loss: 0.8686780333518982:  50%|█████     | 3241/6470 [37:10<34:09,  1.58it/s]
2|3242|Loss: 0.844092845916748:  50%|█████     | 3241/6470 [37:11<34:09,  1.58it/s] 
2|3242|Loss: 0.844092845916748:  50%|█████     | 3242/6470 [37:11<33:13,  1.62it/s]
2|3243|Loss: 0.9240936636924744:  50%|█████     | 3242/6470 [37:11<33:13,  1.62it/s]
2|3243|Loss: 0.9240936636924744:  50%|█████     | 3243/6470 [37:12<32:08,  1.67it/s]
2|3244|Loss: 0.992297887802124:  50%|█████     | 3243/6470 [37:12<32:08,  1.67it/s] 
2|3244|Loss: 0.992297887802124:  50%|█████     | 3244/6470 [37:12<33:46,  1.59it/s]
2|3245|Loss: 0.7854859232902527:  50%|█████     | 3244/6470 [37:13<33:46,  1.59it/s]
2|3245|Loss: 0.7854859232902527:  50%|█████     | 3245/6470 [37:13<35:28,  1.52it/s]
2|3246|Loss: 0.8530011177062988:  50%|█████     | 3245/6470 [37:13<35:28,  1.52it/s]
2|3246|Loss: 0.8530011177062988:  50%|█████     | 3246/6470 [37:14<34:13,  1.57it/s]
2|3247|Loss: 0.8894670009613037:  50%|█████     | 3246/6470 [37:14<34:13,  1.57it/s]
2|3247|Loss: 0.8894670009613037:  50%|█████     | 3247/6470 [37:14<35:10,  1.53it/s]
2|3248|Loss: 0.8297793865203857:  50%|█████     | 3247/6470 [37:15<35:10,  1.53it/s]
2|3248|Loss: 0.8297793865203857:  50%|█████     | 3248/6470 [37:15<36:27,  1.47it/s]
2|3249|Loss: 0.8316671848297119:  50%|█████     | 3248/6470 [37:15<36:27,  1.47it/s]
2|3249|Loss: 0.8316671848297119:  50%|█████     | 3249/6470 [37:16<36:53,  1.45it/s]
2|3250|Loss: 0.5279161334037781:  50%|█████     | 3249/6470 [37:17<36:53,  1.45it/s]
2|3250|Loss: 0.5279161334037781:  50%|█████     | 3250/6470 [37:17<53:00,  1.01it/s]
2|3251|Loss: 0.8324148654937744:  50%|█████     | 3250/6470 [37:18<53:00,  1.01it/s]
2|3251|Loss: 0.8324148654937744:  50%|█████     | 3251/6470 [37:18<52:49,  1.02it/s]
2|3252|Loss: 0.8929868340492249:  50%|█████     | 3251/6470 [37:19<52:49,  1.02it/s]
2|3252|Loss: 0.8929868340492249:  50%|█████     | 3252/6470 [37:19<49:49,  1.08it/s]
2|3253|Loss: 0.9105023741722107:  50%|█████     | 3252/6470 [37:20<49:49,  1.08it/s]
2|3253|Loss: 0.9105023741722107:  50%|█████     | 3253/6470 [37:20<48:13,  1.11it/s]
2|3254|Loss: 0.8627765774726868:  50%|█████     | 3253/6470 [37:20<48:13,  1.11it/s]
2|3254|Loss: 0.8627765774726868:  50%|█████     | 3254/6470 [37:21<45:58,  1.17it/s]
2|3255|Loss: 0.867561936378479:  50%|█████     | 3254/6470 [37:21<45:58,  1.17it/s] 
2|3255|Loss: 0.867561936378479:  50%|█████     | 3255/6470 [37:21<42:00,  1.28it/s]
2|3256|Loss: 0.862359881401062:  50%|█████     | 3255/6470 [37:22<42:00,  1.28it/s]
2|3256|Loss: 0.862359881401062:  50%|█████     | 3256/6470 [37:22<40:40,  1.32it/s]
2|3257|Loss: 0.832002580165863:  50%|█████     | 3256/6470 [37:23<40:40,  1.32it/s]
2|3257|Loss: 0.832002580165863:  50%|█████     | 3257/6470 [37:23<38:45,  1.38it/s]
2|3258|Loss: 0.8972286581993103:  50%|█████     | 3257/6470 [37:23<38:45,  1.38it/s]
2|3258|Loss: 0.8972286581993103:  50%|█████     | 3258/6470 [37:23<38:12,  1.40it/s]
2|3259|Loss: 0.8034461736679077:  50%|█████     | 3258/6470 [37:24<38:12,  1.40it/s]
2|3259|Loss: 0.8034461736679077:  50%|█████     | 3259/6470 [37:24<38:05,  1.41it/s]
2|3260|Loss: 0.9409858584403992:  50%|█████     | 3259/6470 [37:25<38:05,  1.41it/s]
2|3260|Loss: 0.9409858584403992:  50%|█████     | 3260/6470 [37:25<37:53,  1.41it/s]
2|3261|Loss: 0.8334139585494995:  50%|█████     | 3260/6470 [37:25<37:53,  1.41it/s]
2|3261|Loss: 0.8334139585494995:  50%|█████     | 3261/6470 [37:26<36:59,  1.45it/s]
2|3262|Loss: 0.8660886883735657:  50%|█████     | 3261/6470 [37:26<36:59,  1.45it/s]
2|3262|Loss: 0.8660886883735657:  50%|█████     | 3262/6470 [37:26<39:09,  1.37it/s]
2|3263|Loss: 0.9420583248138428:  50%|█████     | 3262/6470 [37:27<39:09,  1.37it/s]
2|3263|Loss: 0.9420583248138428:  50%|█████     | 3263/6470 [37:27<37:28,  1.43it/s]
2|3264|Loss: 0.987312376499176:  50%|█████     | 3263/6470 [37:27<37:28,  1.43it/s] 
2|3264|Loss: 0.987312376499176:  50%|█████     | 3264/6470 [37:28<39:09,  1.36it/s]
2|3265|Loss: 0.8086804747581482:  50%|█████     | 3264/6470 [37:28<39:09,  1.36it/s]
2|3265|Loss: 0.8086804747581482:  50%|█████     | 3265/6470 [37:29<40:25,  1.32it/s]
2|3266|Loss: 0.900543749332428:  50%|█████     | 3265/6470 [37:29<40:25,  1.32it/s] 
2|3266|Loss: 0.900543749332428:  50%|█████     | 3266/6470 [37:29<38:53,  1.37it/s]
2|3267|Loss: 0.7762783765792847:  50%|█████     | 3266/6470 [37:30<38:53,  1.37it/s]
2|3267|Loss: 0.7762783765792847:  50%|█████     | 3267/6470 [37:30<38:32,  1.39it/s]
2|3268|Loss: 0.9017488956451416:  50%|█████     | 3267/6470 [37:30<38:32,  1.39it/s]
2|3268|Loss: 0.9017488956451416:  51%|█████     | 3268/6470 [37:31<38:39,  1.38it/s]
2|3269|Loss: 0.8691373467445374:  51%|█████     | 3268/6470 [37:31<38:39,  1.38it/s]
2|3269|Loss: 0.8691373467445374:  51%|█████     | 3269/6470 [37:32<40:41,  1.31it/s]
2|3270|Loss: 1.0034770965576172:  51%|█████     | 3269/6470 [37:32<40:41,  1.31it/s]
2|3270|Loss: 1.0034770965576172:  51%|█████     | 3270/6470 [37:32<42:24,  1.26it/s]
2|3271|Loss: 0.7688923478126526:  51%|█████     | 3270/6470 [37:33<42:24,  1.26it/s]
2|3271|Loss: 0.7688923478126526:  51%|█████     | 3271/6470 [37:33<44:10,  1.21it/s]
2|3272|Loss: 0.8946705460548401:  51%|█████     | 3271/6470 [37:34<44:10,  1.21it/s]
2|3272|Loss: 0.8946705460548401:  51%|█████     | 3272/6470 [37:34<40:27,  1.32it/s]
2|3273|Loss: 0.9348574876785278:  51%|█████     | 3272/6470 [37:34<40:27,  1.32it/s]
2|3273|Loss: 0.9348574876785278:  51%|█████     | 3273/6470 [37:35<40:51,  1.30it/s]
2|3274|Loss: 0.7600650787353516:  51%|█████     | 3273/6470 [37:35<40:51,  1.30it/s]
2|3274|Loss: 0.7600650787353516:  51%|█████     | 3274/6470 [37:35<40:44,  1.31it/s]
2|3275|Loss: 0.7702801823616028:  51%|█████     | 3274/6470 [37:36<40:44,  1.31it/s]
2|3275|Loss: 0.7702801823616028:  51%|█████     | 3275/6470 [37:36<40:15,  1.32it/s]
2|3276|Loss: 0.7832129001617432:  51%|█████     | 3275/6470 [37:37<40:15,  1.32it/s]
2|3276|Loss: 0.7832129001617432:  51%|█████     | 3276/6470 [37:37<41:39,  1.28it/s]
2|3277|Loss: 0.6455748081207275:  51%|█████     | 3276/6470 [37:37<41:39,  1.28it/s]
2|3277|Loss: 0.6455748081207275:  51%|█████     | 3277/6470 [37:38<38:31,  1.38it/s]
2|3278|Loss: 0.8867235779762268:  51%|█████     | 3277/6470 [37:38<38:31,  1.38it/s]
2|3278|Loss: 0.8867235779762268:  51%|█████     | 3278/6470 [37:38<36:41,  1.45it/s]
2|3279|Loss: 0.7885918617248535:  51%|█████     | 3278/6470 [37:39<36:41,  1.45it/s]
2|3279|Loss: 0.7885918617248535:  51%|█████     | 3279/6470 [37:39<35:10,  1.51it/s]
2|3280|Loss: 0.831977367401123:  51%|█████     | 3279/6470 [37:39<35:10,  1.51it/s] 
2|3280|Loss: 0.831977367401123:  51%|█████     | 3280/6470 [37:39<33:13,  1.60it/s]
2|3281|Loss: 0.9779895544052124:  51%|█████     | 3280/6470 [37:40<33:13,  1.60it/s]
2|3281|Loss: 0.9779895544052124:  51%|█████     | 3281/6470 [37:40<36:11,  1.47it/s]
2|3282|Loss: 0.9537613391876221:  51%|█████     | 3281/6470 [37:41<36:11,  1.47it/s]
2|3282|Loss: 0.9537613391876221:  51%|█████     | 3282/6470 [37:41<33:12,  1.60it/s]
2|3283|Loss: 0.9197030663490295:  51%|█████     | 3282/6470 [37:41<33:12,  1.60it/s]
2|3283|Loss: 0.9197030663490295:  51%|█████     | 3283/6470 [37:41<33:43,  1.57it/s]
2|3284|Loss: 0.8611157536506653:  51%|█████     | 3283/6470 [37:42<33:43,  1.57it/s]
2|3284|Loss: 0.8611157536506653:  51%|█████     | 3284/6470 [37:42<36:17,  1.46it/s]
2|3285|Loss: 0.8106088042259216:  51%|█████     | 3284/6470 [37:43<36:17,  1.46it/s]
2|3285|Loss: 0.8106088042259216:  51%|█████     | 3285/6470 [37:43<36:20,  1.46it/s]
2|3286|Loss: 0.9152778387069702:  51%|█████     | 3285/6470 [37:43<36:20,  1.46it/s]
2|3286|Loss: 0.9152778387069702:  51%|█████     | 3286/6470 [37:43<35:37,  1.49it/s]
2|3287|Loss: 0.9559523463249207:  51%|█████     | 3286/6470 [37:44<35:37,  1.49it/s]
2|3287|Loss: 0.9559523463249207:  51%|█████     | 3287/6470 [37:44<37:11,  1.43it/s]
2|3288|Loss: 0.8965479135513306:  51%|█████     | 3287/6470 [37:45<37:11,  1.43it/s]
2|3288|Loss: 0.8965479135513306:  51%|█████     | 3288/6470 [37:45<38:37,  1.37it/s]
2|3289|Loss: 0.827619194984436:  51%|█████     | 3288/6470 [37:45<38:37,  1.37it/s] 
2|3289|Loss: 0.827619194984436:  51%|█████     | 3289/6470 [37:46<39:01,  1.36it/s]
2|3290|Loss: 0.8153391480445862:  51%|█████     | 3289/6470 [37:46<39:01,  1.36it/s]
2|3290|Loss: 0.8153391480445862:  51%|█████     | 3290/6470 [37:46<37:49,  1.40it/s]
2|3291|Loss: 1.0029960870742798:  51%|█████     | 3290/6470 [37:47<37:49,  1.40it/s]
2|3291|Loss: 1.0029960870742798:  51%|█████     | 3291/6470 [37:47<37:58,  1.40it/s]
2|3292|Loss: 0.8532211780548096:  51%|█████     | 3291/6470 [37:48<37:58,  1.40it/s]
2|3292|Loss: 0.8532211780548096:  51%|█████     | 3292/6470 [37:48<38:16,  1.38it/s]
2|3293|Loss: 0.8883852362632751:  51%|█████     | 3292/6470 [37:48<38:16,  1.38it/s]
2|3293|Loss: 0.8883852362632751:  51%|█████     | 3293/6470 [37:49<37:14,  1.42it/s]
2|3294|Loss: 0.8714650869369507:  51%|█████     | 3293/6470 [37:49<37:14,  1.42it/s]
2|3294|Loss: 0.8714650869369507:  51%|█████     | 3294/6470 [37:49<37:03,  1.43it/s]
2|3295|Loss: 0.8328924179077148:  51%|█████     | 3294/6470 [37:50<37:03,  1.43it/s]
2|3295|Loss: 0.8328924179077148:  51%|█████     | 3295/6470 [37:50<36:20,  1.46it/s]
2|3296|Loss: 0.8616487979888916:  51%|█████     | 3295/6470 [37:50<36:20,  1.46it/s]
2|3296|Loss: 0.8616487979888916:  51%|█████     | 3296/6470 [37:51<35:20,  1.50it/s]
2|3297|Loss: 0.8119920492172241:  51%|█████     | 3296/6470 [37:51<35:20,  1.50it/s]
2|3297|Loss: 0.8119920492172241:  51%|█████     | 3297/6470 [37:51<35:28,  1.49it/s]
2|3298|Loss: 0.7560986876487732:  51%|█████     | 3297/6470 [37:52<35:28,  1.49it/s]
2|3298|Loss: 0.7560986876487732:  51%|█████     | 3298/6470 [37:52<32:47,  1.61it/s]
2|3299|Loss: 1.0979750156402588:  51%|█████     | 3298/6470 [37:52<32:47,  1.61it/s]
2|3299|Loss: 1.0979750156402588:  51%|█████     | 3299/6470 [37:52<33:32,  1.58it/s]
2|3300|Loss: 0.9047138094902039:  51%|█████     | 3299/6470 [37:53<33:32,  1.58it/s]
2|3300|Loss: 0.9047138094902039:  51%|█████     | 3300/6470 [37:53<35:08,  1.50it/s]
2|3301|Loss: 0.9186068773269653:  51%|█████     | 3300/6470 [37:54<35:08,  1.50it/s]
2|3301|Loss: 0.9186068773269653:  51%|█████     | 3301/6470 [37:54<36:13,  1.46it/s]
2|3302|Loss: 0.852207601070404:  51%|█████     | 3301/6470 [37:54<36:13,  1.46it/s] 
2|3302|Loss: 0.852207601070404:  51%|█████     | 3302/6470 [37:55<37:49,  1.40it/s]
2|3303|Loss: 0.7720550298690796:  51%|█████     | 3302/6470 [37:55<37:49,  1.40it/s]
2|3303|Loss: 0.7720550298690796:  51%|█████     | 3303/6470 [37:55<35:08,  1.50it/s]
2|3304|Loss: 0.6818680167198181:  51%|█████     | 3303/6470 [37:56<35:08,  1.50it/s]
2|3304|Loss: 0.6818680167198181:  51%|█████     | 3304/6470 [37:56<37:41,  1.40it/s]
2|3305|Loss: 0.897853672504425:  51%|█████     | 3304/6470 [37:56<37:41,  1.40it/s] 
2|3305|Loss: 0.897853672504425:  51%|█████     | 3305/6470 [37:57<38:31,  1.37it/s]
2|3306|Loss: 0.8496521711349487:  51%|█████     | 3305/6470 [37:57<38:31,  1.37it/s]
2|3306|Loss: 0.8496521711349487:  51%|█████     | 3306/6470 [37:58<38:14,  1.38it/s]
2|3307|Loss: 0.9503080248832703:  51%|█████     | 3306/6470 [37:58<38:14,  1.38it/s]
2|3307|Loss: 0.9503080248832703:  51%|█████     | 3307/6470 [37:58<36:28,  1.45it/s]
2|3308|Loss: 1.0071828365325928:  51%|█████     | 3307/6470 [37:58<36:28,  1.45it/s]
2|3308|Loss: 1.0071828365325928:  51%|█████     | 3308/6470 [37:59<34:21,  1.53it/s]
2|3309|Loss: 0.8182806968688965:  51%|█████     | 3308/6470 [37:59<34:21,  1.53it/s]
2|3309|Loss: 0.8182806968688965:  51%|█████     | 3309/6470 [37:59<35:02,  1.50it/s]
2|3310|Loss: 0.872810423374176:  51%|█████     | 3309/6470 [38:00<35:02,  1.50it/s] 
2|3310|Loss: 0.872810423374176:  51%|█████     | 3310/6470 [38:00<34:11,  1.54it/s]
2|3311|Loss: 1.0121480226516724:  51%|█████     | 3310/6470 [38:00<34:11,  1.54it/s]
2|3311|Loss: 1.0121480226516724:  51%|█████     | 3311/6470 [38:01<35:57,  1.46it/s]
2|3312|Loss: 0.7564101815223694:  51%|█████     | 3311/6470 [38:01<35:57,  1.46it/s]
2|3312|Loss: 0.7564101815223694:  51%|█████     | 3312/6470 [38:02<37:10,  1.42it/s]
2|3313|Loss: 0.8873695135116577:  51%|█████     | 3312/6470 [38:02<37:10,  1.42it/s]
2|3313|Loss: 0.8873695135116577:  51%|█████     | 3313/6470 [38:02<37:54,  1.39it/s]
2|3314|Loss: 0.8805347681045532:  51%|█████     | 3313/6470 [38:03<37:54,  1.39it/s]
2|3314|Loss: 0.8805347681045532:  51%|█████     | 3314/6470 [38:03<39:04,  1.35it/s]
2|3315|Loss: 0.976421594619751:  51%|█████     | 3314/6470 [38:03<39:04,  1.35it/s] 
2|3315|Loss: 0.976421594619751:  51%|█████     | 3315/6470 [38:04<36:01,  1.46it/s]
2|3316|Loss: 0.9900698065757751:  51%|█████     | 3315/6470 [38:04<36:01,  1.46it/s]
2|3316|Loss: 0.9900698065757751:  51%|█████▏    | 3316/6470 [38:04<36:37,  1.44it/s]
2|3317|Loss: 0.8417131304740906:  51%|█████▏    | 3316/6470 [38:05<36:37,  1.44it/s]
2|3317|Loss: 0.8417131304740906:  51%|█████▏    | 3317/6470 [38:05<35:05,  1.50it/s]
2|3318|Loss: 0.8461251854896545:  51%|█████▏    | 3317/6470 [38:05<35:05,  1.50it/s]
2|3318|Loss: 0.8461251854896545:  51%|█████▏    | 3318/6470 [38:06<35:34,  1.48it/s]
2|3319|Loss: 0.7726500630378723:  51%|█████▏    | 3318/6470 [38:06<35:34,  1.48it/s]
2|3319|Loss: 0.7726500630378723:  51%|█████▏    | 3319/6470 [38:06<36:45,  1.43it/s]
2|3320|Loss: 0.7899804711341858:  51%|█████▏    | 3319/6470 [38:07<36:45,  1.43it/s]
2|3320|Loss: 0.7899804711341858:  51%|█████▏    | 3320/6470 [38:07<36:12,  1.45it/s]
2|3321|Loss: 0.957481324672699:  51%|█████▏    | 3320/6470 [38:07<36:12,  1.45it/s] 
2|3321|Loss: 0.957481324672699:  51%|█████▏    | 3321/6470 [38:08<36:45,  1.43it/s]
2|3322|Loss: 0.7883579134941101:  51%|█████▏    | 3321/6470 [38:08<36:45,  1.43it/s]
2|3322|Loss: 0.7883579134941101:  51%|█████▏    | 3322/6470 [38:08<36:10,  1.45it/s]
2|3323|Loss: 0.8867267370223999:  51%|█████▏    | 3322/6470 [38:09<36:10,  1.45it/s]
2|3323|Loss: 0.8867267370223999:  51%|█████▏    | 3323/6470 [38:09<36:11,  1.45it/s]
2|3324|Loss: 0.9837170839309692:  51%|█████▏    | 3323/6470 [38:09<36:11,  1.45it/s]
2|3324|Loss: 0.9837170839309692:  51%|█████▏    | 3324/6470 [38:10<34:55,  1.50it/s]
2|3325|Loss: 0.8391897678375244:  51%|█████▏    | 3324/6470 [38:10<34:55,  1.50it/s]
2|3325|Loss: 0.8391897678375244:  51%|█████▏    | 3325/6470 [38:10<32:53,  1.59it/s]
2|3326|Loss: 0.9035108089447021:  51%|█████▏    | 3325/6470 [38:11<32:53,  1.59it/s]
2|3326|Loss: 0.9035108089447021:  51%|█████▏    | 3326/6470 [38:11<33:47,  1.55it/s]
2|3327|Loss: 0.8809999227523804:  51%|█████▏    | 3326/6470 [38:11<33:47,  1.55it/s]
2|3327|Loss: 0.8809999227523804:  51%|█████▏    | 3327/6470 [38:12<36:43,  1.43it/s]
2|3328|Loss: 0.8955823183059692:  51%|█████▏    | 3327/6470 [38:12<36:43,  1.43it/s]
2|3328|Loss: 0.8955823183059692:  51%|█████▏    | 3328/6470 [38:12<36:15,  1.44it/s]
2|3329|Loss: 0.7631446719169617:  51%|█████▏    | 3328/6470 [38:13<36:15,  1.44it/s]
2|3329|Loss: 0.7631446719169617:  51%|█████▏    | 3329/6470 [38:13<37:36,  1.39it/s]
2|3330|Loss: 0.8359144330024719:  51%|█████▏    | 3329/6470 [38:14<37:36,  1.39it/s]
2|3330|Loss: 0.8359144330024719:  51%|█████▏    | 3330/6470 [38:14<36:59,  1.41it/s]
2|3331|Loss: 0.9386134743690491:  51%|█████▏    | 3330/6470 [38:14<36:59,  1.41it/s]
2|3331|Loss: 0.9386134743690491:  51%|█████▏    | 3331/6470 [38:15<37:03,  1.41it/s]
2|3332|Loss: 0.8389531373977661:  51%|█████▏    | 3331/6470 [38:15<37:03,  1.41it/s]
2|3332|Loss: 0.8389531373977661:  51%|█████▏    | 3332/6470 [38:15<38:32,  1.36it/s]
2|3333|Loss: 1.05071222782135:  51%|█████▏    | 3332/6470 [38:16<38:32,  1.36it/s]  
2|3333|Loss: 1.05071222782135:  52%|█████▏    | 3333/6470 [38:16<37:19,  1.40it/s]
2|3334|Loss: 0.9451178908348083:  52%|█████▏    | 3333/6470 [38:16<37:19,  1.40it/s]
2|3334|Loss: 0.9451178908348083:  52%|█████▏    | 3334/6470 [38:17<37:11,  1.41it/s]
2|3335|Loss: 0.8741447925567627:  52%|█████▏    | 3334/6470 [38:17<37:11,  1.41it/s]
2|3335|Loss: 0.8741447925567627:  52%|█████▏    | 3335/6470 [38:18<38:43,  1.35it/s]
2|3336|Loss: 0.9466775059700012:  52%|█████▏    | 3335/6470 [38:18<38:43,  1.35it/s]
2|3336|Loss: 0.9466775059700012:  52%|█████▏    | 3336/6470 [38:18<40:29,  1.29it/s]
2|3337|Loss: 0.8277533054351807:  52%|█████▏    | 3336/6470 [38:19<40:29,  1.29it/s]
2|3337|Loss: 0.8277533054351807:  52%|█████▏    | 3337/6470 [38:19<40:33,  1.29it/s]
2|3338|Loss: 0.8526709675788879:  52%|█████▏    | 3337/6470 [38:20<40:33,  1.29it/s]
2|3338|Loss: 0.8526709675788879:  52%|█████▏    | 3338/6470 [38:20<37:41,  1.39it/s]
2|3339|Loss: 0.8305927515029907:  52%|█████▏    | 3338/6470 [38:20<37:41,  1.39it/s]
2|3339|Loss: 0.8305927515029907:  52%|█████▏    | 3339/6470 [38:21<37:13,  1.40it/s]
2|3340|Loss: 0.9934858083724976:  52%|█████▏    | 3339/6470 [38:21<37:13,  1.40it/s]
2|3340|Loss: 0.9934858083724976:  52%|█████▏    | 3340/6470 [38:21<34:01,  1.53it/s]
2|3341|Loss: 0.9144772887229919:  52%|█████▏    | 3340/6470 [38:21<34:01,  1.53it/s]
2|3341|Loss: 0.9144772887229919:  52%|█████▏    | 3341/6470 [38:22<32:06,  1.62it/s]
2|3342|Loss: 0.8767017722129822:  52%|█████▏    | 3341/6470 [38:22<32:06,  1.62it/s]
2|3342|Loss: 0.8767017722129822:  52%|█████▏    | 3342/6470 [38:22<35:35,  1.46it/s]
2|3343|Loss: 0.8497300148010254:  52%|█████▏    | 3342/6470 [38:23<35:35,  1.46it/s]
2|3343|Loss: 0.8497300148010254:  52%|█████▏    | 3343/6470 [38:23<35:49,  1.45it/s]
2|3344|Loss: 0.9891709685325623:  52%|█████▏    | 3343/6470 [38:23<35:49,  1.45it/s]
2|3344|Loss: 0.9891709685325623:  52%|█████▏    | 3344/6470 [38:24<35:05,  1.48it/s]
2|3345|Loss: 0.9680919647216797:  52%|█████▏    | 3344/6470 [38:24<35:05,  1.48it/s]
2|3345|Loss: 0.9680919647216797:  52%|█████▏    | 3345/6470 [38:25<37:17,  1.40it/s]
2|3346|Loss: 0.8309205770492554:  52%|█████▏    | 3345/6470 [38:25<37:17,  1.40it/s]
2|3346|Loss: 0.8309205770492554:  52%|█████▏    | 3346/6470 [38:25<37:59,  1.37it/s]
2|3347|Loss: 0.7423696517944336:  52%|█████▏    | 3346/6470 [38:26<37:59,  1.37it/s]
2|3347|Loss: 0.7423696517944336:  52%|█████▏    | 3347/6470 [38:26<37:56,  1.37it/s]
2|3348|Loss: 0.9357289671897888:  52%|█████▏    | 3347/6470 [38:26<37:56,  1.37it/s]
2|3348|Loss: 0.9357289671897888:  52%|█████▏    | 3348/6470 [38:27<38:01,  1.37it/s]
2|3349|Loss: 0.7720759510993958:  52%|█████▏    | 3348/6470 [38:27<38:01,  1.37it/s]
2|3349|Loss: 0.7720759510993958:  52%|█████▏    | 3349/6470 [38:28<39:49,  1.31it/s]
2|3350|Loss: 0.8009682297706604:  52%|█████▏    | 3349/6470 [38:28<39:49,  1.31it/s]
2|3350|Loss: 0.8009682297706604:  52%|█████▏    | 3350/6470 [38:29<41:36,  1.25it/s]
2|3351|Loss: 0.8221186995506287:  52%|█████▏    | 3350/6470 [38:29<41:36,  1.25it/s]
2|3351|Loss: 0.8221186995506287:  52%|█████▏    | 3351/6470 [38:29<39:42,  1.31it/s]
2|3352|Loss: 0.8793481588363647:  52%|█████▏    | 3351/6470 [38:30<39:42,  1.31it/s]
2|3352|Loss: 0.8793481588363647:  52%|█████▏    | 3352/6470 [38:30<38:51,  1.34it/s]
2|3353|Loss: 0.9191902875900269:  52%|█████▏    | 3352/6470 [38:30<38:51,  1.34it/s]
2|3353|Loss: 0.9191902875900269:  52%|█████▏    | 3353/6470 [38:31<36:39,  1.42it/s]
2|3354|Loss: 0.8977686762809753:  52%|█████▏    | 3353/6470 [38:31<36:39,  1.42it/s]
2|3354|Loss: 0.8977686762809753:  52%|█████▏    | 3354/6470 [38:31<36:16,  1.43it/s]
2|3355|Loss: 0.8166500926017761:  52%|█████▏    | 3354/6470 [38:32<36:16,  1.43it/s]
2|3355|Loss: 0.8166500926017761:  52%|█████▏    | 3355/6470 [38:32<36:49,  1.41it/s]
2|3356|Loss: 0.8392497897148132:  52%|█████▏    | 3355/6470 [38:32<36:49,  1.41it/s]
2|3356|Loss: 0.8392497897148132:  52%|█████▏    | 3356/6470 [38:33<38:25,  1.35it/s]
2|3357|Loss: 0.8443578481674194:  52%|█████▏    | 3356/6470 [38:33<38:25,  1.35it/s]
2|3357|Loss: 0.8443578481674194:  52%|█████▏    | 3357/6470 [38:33<38:18,  1.35it/s]
2|3358|Loss: 0.7410304546356201:  52%|█████▏    | 3357/6470 [38:34<38:18,  1.35it/s]
2|3358|Loss: 0.7410304546356201:  52%|█████▏    | 3358/6470 [38:34<39:14,  1.32it/s]
2|3359|Loss: 0.8748794198036194:  52%|█████▏    | 3358/6470 [38:35<39:14,  1.32it/s]
2|3359|Loss: 0.8748794198036194:  52%|█████▏    | 3359/6470 [38:35<36:37,  1.42it/s]
2|3360|Loss: 0.9571493864059448:  52%|█████▏    | 3359/6470 [38:35<36:37,  1.42it/s]
2|3360|Loss: 0.9571493864059448:  52%|█████▏    | 3360/6470 [38:36<36:02,  1.44it/s]
2|3361|Loss: 0.713546097278595:  52%|█████▏    | 3360/6470 [38:36<36:02,  1.44it/s] 
2|3361|Loss: 0.713546097278595:  52%|█████▏    | 3361/6470 [38:36<34:16,  1.51it/s]
2|3362|Loss: 0.8709295988082886:  52%|█████▏    | 3361/6470 [38:36<34:16,  1.51it/s]
2|3362|Loss: 0.8709295988082886:  52%|█████▏    | 3362/6470 [38:37<34:43,  1.49it/s]
2|3363|Loss: 0.8704970479011536:  52%|█████▏    | 3362/6470 [38:37<34:43,  1.49it/s]
2|3363|Loss: 0.8704970479011536:  52%|█████▏    | 3363/6470 [38:38<35:37,  1.45it/s]
2|3364|Loss: 0.791386604309082:  52%|█████▏    | 3363/6470 [38:38<35:37,  1.45it/s] 
2|3364|Loss: 0.791386604309082:  52%|█████▏    | 3364/6470 [38:38<37:51,  1.37it/s]
2|3365|Loss: 0.8058260083198547:  52%|█████▏    | 3364/6470 [38:39<37:51,  1.37it/s]
2|3365|Loss: 0.8058260083198547:  52%|█████▏    | 3365/6470 [38:39<37:13,  1.39it/s]
2|3366|Loss: 0.9000564813613892:  52%|█████▏    | 3365/6470 [38:39<37:13,  1.39it/s]
2|3366|Loss: 0.9000564813613892:  52%|█████▏    | 3366/6470 [38:40<35:43,  1.45it/s]
2|3367|Loss: 0.8933192491531372:  52%|█████▏    | 3366/6470 [38:40<35:43,  1.45it/s]
2|3367|Loss: 0.8933192491531372:  52%|█████▏    | 3367/6470 [38:40<34:55,  1.48it/s]
2|3368|Loss: 0.8774880170822144:  52%|█████▏    | 3367/6470 [38:41<34:55,  1.48it/s]
2|3368|Loss: 0.8774880170822144:  52%|█████▏    | 3368/6470 [38:41<33:05,  1.56it/s]
2|3369|Loss: 0.8865693807601929:  52%|█████▏    | 3368/6470 [38:41<33:05,  1.56it/s]
2|3369|Loss: 0.8865693807601929:  52%|█████▏    | 3369/6470 [38:42<35:25,  1.46it/s]
2|3370|Loss: 0.7779024839401245:  52%|█████▏    | 3369/6470 [38:42<35:25,  1.46it/s]
2|3370|Loss: 0.7779024839401245:  52%|█████▏    | 3370/6470 [38:42<35:14,  1.47it/s]
2|3371|Loss: 0.9222699999809265:  52%|█████▏    | 3370/6470 [38:43<35:14,  1.47it/s]
2|3371|Loss: 0.9222699999809265:  52%|█████▏    | 3371/6470 [38:43<33:40,  1.53it/s]
2|3372|Loss: 0.7551920413970947:  52%|█████▏    | 3371/6470 [38:43<33:40,  1.53it/s]
2|3372|Loss: 0.7551920413970947:  52%|█████▏    | 3372/6470 [38:44<35:16,  1.46it/s]
2|3373|Loss: 0.993416428565979:  52%|█████▏    | 3372/6470 [38:44<35:16,  1.46it/s] 
2|3373|Loss: 0.993416428565979:  52%|█████▏    | 3373/6470 [38:45<37:44,  1.37it/s]
2|3374|Loss: 0.7176221013069153:  52%|█████▏    | 3373/6470 [38:45<37:44,  1.37it/s]
2|3374|Loss: 0.7176221013069153:  52%|█████▏    | 3374/6470 [38:45<36:16,  1.42it/s]
2|3375|Loss: 0.87596595287323:  52%|█████▏    | 3374/6470 [38:46<36:16,  1.42it/s]  
2|3375|Loss: 0.87596595287323:  52%|█████▏    | 3375/6470 [38:46<35:05,  1.47it/s]
2|3376|Loss: 0.7884593605995178:  52%|█████▏    | 3375/6470 [38:46<35:05,  1.47it/s]
2|3376|Loss: 0.7884593605995178:  52%|█████▏    | 3376/6470 [38:47<36:36,  1.41it/s]
2|3377|Loss: 0.8722717761993408:  52%|█████▏    | 3376/6470 [38:47<36:36,  1.41it/s]
2|3377|Loss: 0.8722717761993408:  52%|█████▏    | 3377/6470 [38:47<38:55,  1.32it/s]
2|3378|Loss: 0.7907782196998596:  52%|█████▏    | 3377/6470 [38:48<38:55,  1.32it/s]
2|3378|Loss: 0.7907782196998596:  52%|█████▏    | 3378/6470 [38:48<39:21,  1.31it/s]
2|3379|Loss: 0.9231500029563904:  52%|█████▏    | 3378/6470 [38:49<39:21,  1.31it/s]
2|3379|Loss: 0.9231500029563904:  52%|█████▏    | 3379/6470 [38:49<40:25,  1.27it/s]
2|3380|Loss: 0.8390969634056091:  52%|█████▏    | 3379/6470 [38:50<40:25,  1.27it/s]
2|3380|Loss: 0.8390969634056091:  52%|█████▏    | 3380/6470 [38:50<40:46,  1.26it/s]
2|3381|Loss: 0.840087890625:  52%|█████▏    | 3380/6470 [38:50<40:46,  1.26it/s]    
2|3381|Loss: 0.840087890625:  52%|█████▏    | 3381/6470 [38:51<41:37,  1.24it/s]
2|3382|Loss: 0.8347982168197632:  52%|█████▏    | 3381/6470 [38:51<41:37,  1.24it/s]
2|3382|Loss: 0.8347982168197632:  52%|█████▏    | 3382/6470 [38:51<41:05,  1.25it/s]
2|3383|Loss: 0.8539116382598877:  52%|█████▏    | 3382/6470 [38:52<41:05,  1.25it/s]
2|3383|Loss: 0.8539116382598877:  52%|█████▏    | 3383/6470 [38:52<41:02,  1.25it/s]
2|3384|Loss: 0.9708830118179321:  52%|█████▏    | 3383/6470 [38:53<41:02,  1.25it/s]
2|3384|Loss: 0.9708830118179321:  52%|█████▏    | 3384/6470 [38:53<40:25,  1.27it/s]
2|3385|Loss: 0.7413915395736694:  52%|█████▏    | 3384/6470 [38:53<40:25,  1.27it/s]
2|3385|Loss: 0.7413915395736694:  52%|█████▏    | 3385/6470 [38:54<39:56,  1.29it/s]
2|3386|Loss: 0.912699818611145:  52%|█████▏    | 3385/6470 [38:54<39:56,  1.29it/s] 
2|3386|Loss: 0.912699818611145:  52%|█████▏    | 3386/6470 [38:55<40:08,  1.28it/s]
2|3387|Loss: 0.956596314907074:  52%|█████▏    | 3386/6470 [38:55<40:08,  1.28it/s]
2|3387|Loss: 0.956596314907074:  52%|█████▏    | 3387/6470 [38:55<39:43,  1.29it/s]
2|3388|Loss: 0.7895670533180237:  52%|█████▏    | 3387/6470 [38:56<39:43,  1.29it/s]
2|3388|Loss: 0.7895670533180237:  52%|█████▏    | 3388/6470 [38:56<39:15,  1.31it/s]
2|3389|Loss: 0.8376514911651611:  52%|█████▏    | 3388/6470 [38:56<39:15,  1.31it/s]
2|3389|Loss: 0.8376514911651611:  52%|█████▏    | 3389/6470 [38:57<37:41,  1.36it/s]
2|3390|Loss: 0.9340283274650574:  52%|█████▏    | 3389/6470 [38:57<37:41,  1.36it/s]
2|3390|Loss: 0.9340283274650574:  52%|█████▏    | 3390/6470 [38:57<35:50,  1.43it/s]
2|3391|Loss: 0.8695867657661438:  52%|█████▏    | 3390/6470 [38:58<35:50,  1.43it/s]
2|3391|Loss: 0.8695867657661438:  52%|█████▏    | 3391/6470 [38:58<34:59,  1.47it/s]
2|3392|Loss: 0.9272260665893555:  52%|█████▏    | 3391/6470 [38:58<34:59,  1.47it/s]
2|3392|Loss: 0.9272260665893555:  52%|█████▏    | 3392/6470 [38:59<34:25,  1.49it/s]
2|3393|Loss: 0.7764770984649658:  52%|█████▏    | 3392/6470 [38:59<34:25,  1.49it/s]
2|3393|Loss: 0.7764770984649658:  52%|█████▏    | 3393/6470 [38:59<33:10,  1.55it/s]
2|3394|Loss: 0.8108134269714355:  52%|█████▏    | 3393/6470 [39:00<33:10,  1.55it/s]
2|3394|Loss: 0.8108134269714355:  52%|█████▏    | 3394/6470 [39:00<34:43,  1.48it/s]
2|3395|Loss: 1.036961555480957:  52%|█████▏    | 3394/6470 [39:00<34:43,  1.48it/s] 
2|3395|Loss: 1.036961555480957:  52%|█████▏    | 3395/6470 [39:01<36:29,  1.40it/s]
2|3396|Loss: 0.9064066410064697:  52%|█████▏    | 3395/6470 [39:01<36:29,  1.40it/s]
2|3396|Loss: 0.9064066410064697:  52%|█████▏    | 3396/6470 [39:02<37:46,  1.36it/s]
2|3397|Loss: 0.8773140907287598:  52%|█████▏    | 3396/6470 [39:02<37:46,  1.36it/s]
2|3397|Loss: 0.8773140907287598:  53%|█████▎    | 3397/6470 [39:02<37:34,  1.36it/s]
2|3398|Loss: 0.8890506625175476:  53%|█████▎    | 3397/6470 [39:03<37:34,  1.36it/s]
2|3398|Loss: 0.8890506625175476:  53%|█████▎    | 3398/6470 [39:03<37:30,  1.37it/s]
2|3399|Loss: 0.8858191967010498:  53%|█████▎    | 3398/6470 [39:03<37:30,  1.37it/s]
2|3399|Loss: 0.8858191967010498:  53%|█████▎    | 3399/6470 [39:04<37:00,  1.38it/s]
2|3400|Loss: 1.033351182937622:  53%|█████▎    | 3399/6470 [39:04<37:00,  1.38it/s] 
2|3400|Loss: 1.033351182937622:  53%|█████▎    | 3400/6470 [39:05<39:18,  1.30it/s]
2|3401|Loss: 1.008934497833252:  53%|█████▎    | 3400/6470 [39:05<39:18,  1.30it/s]
2|3401|Loss: 1.008934497833252:  53%|█████▎    | 3401/6470 [39:05<38:01,  1.34it/s]
2|3402|Loss: 0.8770984411239624:  53%|█████▎    | 3401/6470 [39:06<38:01,  1.34it/s]
2|3402|Loss: 0.8770984411239624:  53%|█████▎    | 3402/6470 [39:06<38:41,  1.32it/s]
2|3403|Loss: 0.8942766189575195:  53%|█████▎    | 3402/6470 [39:07<38:41,  1.32it/s]
2|3403|Loss: 0.8942766189575195:  53%|█████▎    | 3403/6470 [39:07<39:52,  1.28it/s]
2|3404|Loss: 0.9120615720748901:  53%|█████▎    | 3403/6470 [39:07<39:52,  1.28it/s]
2|3404|Loss: 0.9120615720748901:  53%|█████▎    | 3404/6470 [39:08<39:02,  1.31it/s]
2|3405|Loss: 0.8494910597801208:  53%|█████▎    | 3404/6470 [39:08<39:02,  1.31it/s]
2|3405|Loss: 0.8494910597801208:  53%|█████▎    | 3405/6470 [39:08<38:04,  1.34it/s]
2|3406|Loss: 0.9580308198928833:  53%|█████▎    | 3405/6470 [39:09<38:04,  1.34it/s]
2|3406|Loss: 0.9580308198928833:  53%|█████▎    | 3406/6470 [39:09<35:34,  1.44it/s]
2|3407|Loss: 0.9513207674026489:  53%|█████▎    | 3406/6470 [39:09<35:34,  1.44it/s]
2|3407|Loss: 0.9513207674026489:  53%|█████▎    | 3407/6470 [39:10<36:29,  1.40it/s]
2|3408|Loss: 0.8616339564323425:  53%|█████▎    | 3407/6470 [39:10<36:29,  1.40it/s]
2|3408|Loss: 0.8616339564323425:  53%|█████▎    | 3408/6470 [39:10<34:51,  1.46it/s]
2|3409|Loss: 0.8639922738075256:  53%|█████▎    | 3408/6470 [39:11<34:51,  1.46it/s]
2|3409|Loss: 0.8639922738075256:  53%|█████▎    | 3409/6470 [39:11<35:02,  1.46it/s]
2|3410|Loss: 0.9117573499679565:  53%|█████▎    | 3409/6470 [39:11<35:02,  1.46it/s]
2|3410|Loss: 0.9117573499679565:  53%|█████▎    | 3410/6470 [39:12<33:59,  1.50it/s]
2|3411|Loss: 0.8127683401107788:  53%|█████▎    | 3410/6470 [39:12<33:59,  1.50it/s]
2|3411|Loss: 0.8127683401107788:  53%|█████▎    | 3411/6470 [39:12<36:08,  1.41it/s]
2|3412|Loss: 0.8825985193252563:  53%|█████▎    | 3411/6470 [39:13<36:08,  1.41it/s]
2|3412|Loss: 0.8825985193252563:  53%|█████▎    | 3412/6470 [39:13<35:12,  1.45it/s]
2|3413|Loss: 0.8385888934135437:  53%|█████▎    | 3412/6470 [39:13<35:12,  1.45it/s]
2|3413|Loss: 0.8385888934135437:  53%|█████▎    | 3413/6470 [39:14<34:16,  1.49it/s]
2|3414|Loss: 0.9382327795028687:  53%|█████▎    | 3413/6470 [39:14<34:16,  1.49it/s]
2|3414|Loss: 0.9382327795028687:  53%|█████▎    | 3414/6470 [39:14<32:13,  1.58it/s]
2|3415|Loss: 0.9139055013656616:  53%|█████▎    | 3414/6470 [39:15<32:13,  1.58it/s]
2|3415|Loss: 0.9139055013656616:  53%|█████▎    | 3415/6470 [39:15<30:31,  1.67it/s]
2|3416|Loss: 0.8800176382064819:  53%|█████▎    | 3415/6470 [39:15<30:31,  1.67it/s]
2|3416|Loss: 0.8800176382064819:  53%|█████▎    | 3416/6470 [39:15<31:37,  1.61it/s]
2|3417|Loss: 0.9195391535758972:  53%|█████▎    | 3416/6470 [39:16<31:37,  1.61it/s]
2|3417|Loss: 0.9195391535758972:  53%|█████▎    | 3417/6470 [39:16<35:09,  1.45it/s]
2|3418|Loss: 0.8863257169723511:  53%|█████▎    | 3417/6470 [39:17<35:09,  1.45it/s]
2|3418|Loss: 0.8863257169723511:  53%|█████▎    | 3418/6470 [39:17<35:22,  1.44it/s]
2|3419|Loss: 1.0826820135116577:  53%|█████▎    | 3418/6470 [39:17<35:22,  1.44it/s]
2|3419|Loss: 1.0826820135116577:  53%|█████▎    | 3419/6470 [39:18<35:17,  1.44it/s]
2|3420|Loss: 0.8221139907836914:  53%|█████▎    | 3419/6470 [39:18<35:17,  1.44it/s]
2|3420|Loss: 0.8221139907836914:  53%|█████▎    | 3420/6470 [39:18<33:00,  1.54it/s]
2|3421|Loss: 0.7992846369743347:  53%|█████▎    | 3420/6470 [39:19<33:00,  1.54it/s]
2|3421|Loss: 0.7992846369743347:  53%|█████▎    | 3421/6470 [39:19<31:41,  1.60it/s]
2|3422|Loss: 0.7984579801559448:  53%|█████▎    | 3421/6470 [39:19<31:41,  1.60it/s]
2|3422|Loss: 0.7984579801559448:  53%|█████▎    | 3422/6470 [39:19<32:32,  1.56it/s]
2|3423|Loss: 0.8969415426254272:  53%|█████▎    | 3422/6470 [39:20<32:32,  1.56it/s]
2|3423|Loss: 0.8969415426254272:  53%|█████▎    | 3423/6470 [39:20<32:49,  1.55it/s]
2|3424|Loss: 0.8948149085044861:  53%|█████▎    | 3423/6470 [39:21<32:49,  1.55it/s]
2|3424|Loss: 0.8948149085044861:  53%|█████▎    | 3424/6470 [39:21<35:43,  1.42it/s]
2|3425|Loss: 0.8667106032371521:  53%|█████▎    | 3424/6470 [39:21<35:43,  1.42it/s]
2|3425|Loss: 0.8667106032371521:  53%|█████▎    | 3425/6470 [39:22<35:40,  1.42it/s]
2|3426|Loss: 0.7879346013069153:  53%|█████▎    | 3425/6470 [39:22<35:40,  1.42it/s]
2|3426|Loss: 0.7879346013069153:  53%|█████▎    | 3426/6470 [39:22<32:24,  1.57it/s]
2|3427|Loss: 0.8189615607261658:  53%|█████▎    | 3426/6470 [39:23<32:24,  1.57it/s]
2|3427|Loss: 0.8189615607261658:  53%|█████▎    | 3427/6470 [39:23<35:05,  1.45it/s]
2|3428|Loss: 0.8413495421409607:  53%|█████▎    | 3427/6470 [39:23<35:05,  1.45it/s]
2|3428|Loss: 0.8413495421409607:  53%|█████▎    | 3428/6470 [39:24<37:30,  1.35it/s]
2|3429|Loss: 0.8032661080360413:  53%|█████▎    | 3428/6470 [39:24<37:30,  1.35it/s]
2|3429|Loss: 0.8032661080360413:  53%|█████▎    | 3429/6470 [39:25<40:30,  1.25it/s]
2|3430|Loss: 0.7946223020553589:  53%|█████▎    | 3429/6470 [39:25<40:30,  1.25it/s]
2|3430|Loss: 0.7946223020553589:  53%|█████▎    | 3430/6470 [39:26<41:30,  1.22it/s]
2|3431|Loss: 0.8776419758796692:  53%|█████▎    | 3430/6470 [39:26<41:30,  1.22it/s]
2|3431|Loss: 0.8776419758796692:  53%|█████▎    | 3431/6470 [39:26<41:26,  1.22it/s]
2|3432|Loss: 0.8461397290229797:  53%|█████▎    | 3431/6470 [39:27<41:26,  1.22it/s]
2|3432|Loss: 0.8461397290229797:  53%|█████▎    | 3432/6470 [39:27<37:24,  1.35it/s]
2|3433|Loss: 0.8849998116493225:  53%|█████▎    | 3432/6470 [39:27<37:24,  1.35it/s]
2|3433|Loss: 0.8849998116493225:  53%|█████▎    | 3433/6470 [39:28<35:27,  1.43it/s]
2|3434|Loss: 0.8633203506469727:  53%|█████▎    | 3433/6470 [39:28<35:27,  1.43it/s]
2|3434|Loss: 0.8633203506469727:  53%|█████▎    | 3434/6470 [39:28<33:41,  1.50it/s]
2|3435|Loss: 0.8967593312263489:  53%|█████▎    | 3434/6470 [39:29<33:41,  1.50it/s]
2|3435|Loss: 0.8967593312263489:  53%|█████▎    | 3435/6470 [39:29<34:25,  1.47it/s]
2|3436|Loss: 0.748453676700592:  53%|█████▎    | 3435/6470 [39:29<34:25,  1.47it/s] 
2|3436|Loss: 0.748453676700592:  53%|█████▎    | 3436/6470 [39:30<33:25,  1.51it/s]
2|3437|Loss: 0.8819616436958313:  53%|█████▎    | 3436/6470 [39:30<33:25,  1.51it/s]
2|3437|Loss: 0.8819616436958313:  53%|█████▎    | 3437/6470 [39:30<32:39,  1.55it/s]
2|3438|Loss: 0.8680286407470703:  53%|█████▎    | 3437/6470 [39:30<32:39,  1.55it/s]
2|3438|Loss: 0.8680286407470703:  53%|█████▎    | 3438/6470 [39:31<32:33,  1.55it/s]
2|3439|Loss: 0.8026538491249084:  53%|█████▎    | 3438/6470 [39:31<32:33,  1.55it/s]
2|3439|Loss: 0.8026538491249084:  53%|█████▎    | 3439/6470 [39:31<33:17,  1.52it/s]
2|3440|Loss: 0.8771118521690369:  53%|█████▎    | 3439/6470 [39:32<33:17,  1.52it/s]
2|3440|Loss: 0.8771118521690369:  53%|█████▎    | 3440/6470 [39:32<33:01,  1.53it/s]
2|3441|Loss: 0.8101303577423096:  53%|█████▎    | 3440/6470 [39:33<33:01,  1.53it/s]
2|3441|Loss: 0.8101303577423096:  53%|█████▎    | 3441/6470 [39:33<36:05,  1.40it/s]
2|3442|Loss: 0.8833734393119812:  53%|█████▎    | 3441/6470 [39:33<36:05,  1.40it/s]
2|3442|Loss: 0.8833734393119812:  53%|█████▎    | 3442/6470 [39:34<33:59,  1.48it/s]
2|3443|Loss: 0.8308822512626648:  53%|█████▎    | 3442/6470 [39:34<33:59,  1.48it/s]
2|3443|Loss: 0.8308822512626648:  53%|█████▎    | 3443/6470 [39:34<32:54,  1.53it/s]
2|3444|Loss: 0.7958673238754272:  53%|█████▎    | 3443/6470 [39:35<32:54,  1.53it/s]
2|3444|Loss: 0.7958673238754272:  53%|█████▎    | 3444/6470 [39:35<32:27,  1.55it/s]
2|3445|Loss: 0.932845950126648:  53%|█████▎    | 3444/6470 [39:35<32:27,  1.55it/s] 
2|3445|Loss: 0.932845950126648:  53%|█████▎    | 3445/6470 [39:35<33:07,  1.52it/s]
2|3446|Loss: 0.8824623227119446:  53%|█████▎    | 3445/6470 [39:36<33:07,  1.52it/s]
2|3446|Loss: 0.8824623227119446:  53%|█████▎    | 3446/6470 [39:36<33:50,  1.49it/s]
2|3447|Loss: 0.7847368121147156:  53%|█████▎    | 3446/6470 [39:37<33:50,  1.49it/s]
2|3447|Loss: 0.7847368121147156:  53%|█████▎    | 3447/6470 [39:37<33:15,  1.51it/s]
2|3448|Loss: 0.8804509043693542:  53%|█████▎    | 3447/6470 [39:37<33:15,  1.51it/s]
2|3448|Loss: 0.8804509043693542:  53%|█████▎    | 3448/6470 [39:37<33:37,  1.50it/s]
2|3449|Loss: 0.7330395579338074:  53%|█████▎    | 3448/6470 [39:38<33:37,  1.50it/s]
2|3449|Loss: 0.7330395579338074:  53%|█████▎    | 3449/6470 [39:38<33:02,  1.52it/s]
2|3450|Loss: 0.8292021751403809:  53%|█████▎    | 3449/6470 [39:38<33:02,  1.52it/s]
2|3450|Loss: 0.8292021751403809:  53%|█████▎    | 3450/6470 [39:39<33:28,  1.50it/s]
2|3451|Loss: 0.8029633164405823:  53%|█████▎    | 3450/6470 [39:39<33:28,  1.50it/s]
2|3451|Loss: 0.8029633164405823:  53%|█████▎    | 3451/6470 [39:40<36:31,  1.38it/s]
2|3452|Loss: 0.7505190372467041:  53%|█████▎    | 3451/6470 [39:40<36:31,  1.38it/s]
2|3452|Loss: 0.7505190372467041:  53%|█████▎    | 3452/6470 [39:40<36:03,  1.40it/s]
2|3453|Loss: 1.069270372390747:  53%|█████▎    | 3452/6470 [39:41<36:03,  1.40it/s] 
2|3453|Loss: 1.069270372390747:  53%|█████▎    | 3453/6470 [39:41<36:08,  1.39it/s]
2|3454|Loss: 0.7836035490036011:  53%|█████▎    | 3453/6470 [39:42<36:08,  1.39it/s]
2|3454|Loss: 0.7836035490036011:  53%|█████▎    | 3454/6470 [39:42<37:35,  1.34it/s]
2|3455|Loss: 1.019798755645752:  53%|█████▎    | 3454/6470 [39:42<37:35,  1.34it/s] 
2|3455|Loss: 1.019798755645752:  53%|█████▎    | 3455/6470 [39:43<37:14,  1.35it/s]
2|3456|Loss: 0.9000204801559448:  53%|█████▎    | 3455/6470 [39:43<37:14,  1.35it/s]
2|3456|Loss: 0.9000204801559448:  53%|█████▎    | 3456/6470 [39:43<36:37,  1.37it/s]
2|3457|Loss: 0.8913961052894592:  53%|█████▎    | 3456/6470 [39:44<36:37,  1.37it/s]
2|3457|Loss: 0.8913961052894592:  53%|█████▎    | 3457/6470 [39:44<37:43,  1.33it/s]
2|3458|Loss: 0.9879636168479919:  53%|█████▎    | 3457/6470 [39:44<37:43,  1.33it/s]
2|3458|Loss: 0.9879636168479919:  53%|█████▎    | 3458/6470 [39:45<34:44,  1.44it/s]
2|3459|Loss: 0.9384191036224365:  53%|█████▎    | 3458/6470 [39:45<34:44,  1.44it/s]
2|3459|Loss: 0.9384191036224365:  53%|█████▎    | 3459/6470 [39:45<31:12,  1.61it/s]
2|3460|Loss: 1.0746341943740845:  53%|█████▎    | 3459/6470 [39:46<31:12,  1.61it/s]
2|3460|Loss: 1.0746341943740845:  53%|█████▎    | 3460/6470 [39:46<33:25,  1.50it/s]
2|3461|Loss: 0.8689118027687073:  53%|█████▎    | 3460/6470 [39:46<33:25,  1.50it/s]
2|3461|Loss: 0.8689118027687073:  53%|█████▎    | 3461/6470 [39:47<34:28,  1.46it/s]
2|3462|Loss: 0.9156626462936401:  53%|█████▎    | 3461/6470 [39:47<34:28,  1.46it/s]
2|3462|Loss: 0.9156626462936401:  54%|█████▎    | 3462/6470 [39:47<35:08,  1.43it/s]
2|3463|Loss: 0.8363698720932007:  54%|█████▎    | 3462/6470 [39:48<35:08,  1.43it/s]
2|3463|Loss: 0.8363698720932007:  54%|█████▎    | 3463/6470 [39:48<35:56,  1.39it/s]
2|3464|Loss: 0.7467460632324219:  54%|█████▎    | 3463/6470 [39:49<35:56,  1.39it/s]
2|3464|Loss: 0.7467460632324219:  54%|█████▎    | 3464/6470 [39:49<36:18,  1.38it/s]
2|3465|Loss: 0.9933159351348877:  54%|█████▎    | 3464/6470 [39:49<36:18,  1.38it/s]
2|3465|Loss: 0.9933159351348877:  54%|█████▎    | 3465/6470 [39:49<33:06,  1.51it/s]
2|3466|Loss: 0.8127991557121277:  54%|█████▎    | 3465/6470 [39:50<33:06,  1.51it/s]
2|3466|Loss: 0.8127991557121277:  54%|█████▎    | 3466/6470 [39:50<30:35,  1.64it/s]
2|3467|Loss: 0.8222014307975769:  54%|█████▎    | 3466/6470 [39:50<30:35,  1.64it/s]
2|3467|Loss: 0.8222014307975769:  54%|█████▎    | 3467/6470 [39:51<32:55,  1.52it/s]
2|3468|Loss: 0.8003562092781067:  54%|█████▎    | 3467/6470 [39:51<32:55,  1.52it/s]
2|3468|Loss: 0.8003562092781067:  54%|█████▎    | 3468/6470 [39:51<33:55,  1.47it/s]
2|3469|Loss: 0.8797935843467712:  54%|█████▎    | 3468/6470 [39:52<33:55,  1.47it/s]
2|3469|Loss: 0.8797935843467712:  54%|█████▎    | 3469/6470 [39:52<34:41,  1.44it/s]
2|3470|Loss: 0.8756235837936401:  54%|█████▎    | 3469/6470 [39:53<34:41,  1.44it/s]
2|3470|Loss: 0.8756235837936401:  54%|█████▎    | 3470/6470 [39:53<35:58,  1.39it/s]
2|3471|Loss: 1.0229295492172241:  54%|█████▎    | 3470/6470 [39:53<35:58,  1.39it/s]
2|3471|Loss: 1.0229295492172241:  54%|█████▎    | 3471/6470 [39:54<37:52,  1.32it/s]
2|3472|Loss: 0.7318648099899292:  54%|█████▎    | 3471/6470 [39:54<37:52,  1.32it/s]
2|3472|Loss: 0.7318648099899292:  54%|█████▎    | 3472/6470 [39:55<38:05,  1.31it/s]
2|3473|Loss: 0.9418370723724365:  54%|█████▎    | 3472/6470 [39:55<38:05,  1.31it/s]
2|3473|Loss: 0.9418370723724365:  54%|█████▎    | 3473/6470 [39:55<37:16,  1.34it/s]
2|3474|Loss: 0.8081809878349304:  54%|█████▎    | 3473/6470 [39:56<37:16,  1.34it/s]
2|3474|Loss: 0.8081809878349304:  54%|█████▎    | 3474/6470 [39:56<38:42,  1.29it/s]
2|3475|Loss: 0.976661741733551:  54%|█████▎    | 3474/6470 [39:56<38:42,  1.29it/s] 
2|3475|Loss: 0.976661741733551:  54%|█████▎    | 3475/6470 [39:57<36:45,  1.36it/s]
2|3476|Loss: 0.9925932884216309:  54%|█████▎    | 3475/6470 [39:57<36:45,  1.36it/s]
2|3476|Loss: 0.9925932884216309:  54%|█████▎    | 3476/6470 [39:57<34:54,  1.43it/s]
2|3477|Loss: 0.839924693107605:  54%|█████▎    | 3476/6470 [39:58<34:54,  1.43it/s] 
2|3477|Loss: 0.839924693107605:  54%|█████▎    | 3477/6470 [39:58<36:59,  1.35it/s]
2|3478|Loss: 0.8469381928443909:  54%|█████▎    | 3477/6470 [39:59<36:59,  1.35it/s]
2|3478|Loss: 0.8469381928443909:  54%|█████▍    | 3478/6470 [39:59<35:42,  1.40it/s]
2|3479|Loss: 0.7973467707633972:  54%|█████▍    | 3478/6470 [39:59<35:42,  1.40it/s]
2|3479|Loss: 0.7973467707633972:  54%|█████▍    | 3479/6470 [39:59<32:57,  1.51it/s]
2|3480|Loss: 0.8983457684516907:  54%|█████▍    | 3479/6470 [40:00<32:57,  1.51it/s]
2|3480|Loss: 0.8983457684516907:  54%|█████▍    | 3480/6470 [40:00<32:17,  1.54it/s]
2|3481|Loss: 0.969739556312561:  54%|█████▍    | 3480/6470 [40:00<32:17,  1.54it/s] 
2|3481|Loss: 0.969739556312561:  54%|█████▍    | 3481/6470 [40:01<32:14,  1.54it/s]
2|3482|Loss: 0.9944682121276855:  54%|█████▍    | 3481/6470 [40:01<32:14,  1.54it/s]
2|3482|Loss: 0.9944682121276855:  54%|█████▍    | 3482/6470 [40:01<32:59,  1.51it/s]
2|3483|Loss: 0.9345133304595947:  54%|█████▍    | 3482/6470 [40:02<32:59,  1.51it/s]
2|3483|Loss: 0.9345133304595947:  54%|█████▍    | 3483/6470 [40:02<33:52,  1.47it/s]
2|3484|Loss: 0.9367543458938599:  54%|█████▍    | 3483/6470 [40:02<33:52,  1.47it/s]
2|3484|Loss: 0.9367543458938599:  54%|█████▍    | 3484/6470 [40:03<33:14,  1.50it/s]
2|3485|Loss: 0.905021071434021:  54%|█████▍    | 3484/6470 [40:03<33:14,  1.50it/s] 
2|3485|Loss: 0.905021071434021:  54%|█████▍    | 3485/6470 [40:03<32:45,  1.52it/s]
2|3486|Loss: 1.0012927055358887:  54%|█████▍    | 3485/6470 [40:04<32:45,  1.52it/s]
2|3486|Loss: 1.0012927055358887:  54%|█████▍    | 3486/6470 [40:04<32:15,  1.54it/s]
2|3487|Loss: 0.8224743008613586:  54%|█████▍    | 3486/6470 [40:04<32:15,  1.54it/s]
2|3487|Loss: 0.8224743008613586:  54%|█████▍    | 3487/6470 [40:05<31:12,  1.59it/s]
2|3488|Loss: 0.8438265919685364:  54%|█████▍    | 3487/6470 [40:05<31:12,  1.59it/s]
2|3488|Loss: 0.8438265919685364:  54%|█████▍    | 3488/6470 [40:05<33:53,  1.47it/s]
2|3489|Loss: 0.8160873055458069:  54%|█████▍    | 3488/6470 [40:06<33:53,  1.47it/s]
2|3489|Loss: 0.8160873055458069:  54%|█████▍    | 3489/6470 [40:06<32:56,  1.51it/s]
2|3490|Loss: 0.9252467155456543:  54%|█████▍    | 3489/6470 [40:06<32:56,  1.51it/s]
2|3490|Loss: 0.9252467155456543:  54%|█████▍    | 3490/6470 [40:07<31:34,  1.57it/s]
2|3491|Loss: 0.9710348844528198:  54%|█████▍    | 3490/6470 [40:07<31:34,  1.57it/s]
2|3491|Loss: 0.9710348844528198:  54%|█████▍    | 3491/6470 [40:07<31:26,  1.58it/s]
2|3492|Loss: 0.8129246830940247:  54%|█████▍    | 3491/6470 [40:08<31:26,  1.58it/s]
2|3492|Loss: 0.8129246830940247:  54%|█████▍    | 3492/6470 [40:08<31:31,  1.57it/s]
2|3493|Loss: 0.7891570925712585:  54%|█████▍    | 3492/6470 [40:08<31:31,  1.57it/s]
2|3493|Loss: 0.7891570925712585:  54%|█████▍    | 3493/6470 [40:08<30:02,  1.65it/s]
2|3494|Loss: 0.888990581035614:  54%|█████▍    | 3493/6470 [40:09<30:02,  1.65it/s] 
2|3494|Loss: 0.888990581035614:  54%|█████▍    | 3494/6470 [40:09<29:52,  1.66it/s]
2|3495|Loss: 0.9583532214164734:  54%|█████▍    | 3494/6470 [40:09<29:52,  1.66it/s]
2|3495|Loss: 0.9583532214164734:  54%|█████▍    | 3495/6470 [40:09<29:36,  1.68it/s]
2|3496|Loss: 0.8754284977912903:  54%|█████▍    | 3495/6470 [40:10<29:36,  1.68it/s]
2|3496|Loss: 0.8754284977912903:  54%|█████▍    | 3496/6470 [40:10<32:36,  1.52it/s]
2|3497|Loss: 0.7497882843017578:  54%|█████▍    | 3496/6470 [40:11<32:36,  1.52it/s]
2|3497|Loss: 0.7497882843017578:  54%|█████▍    | 3497/6470 [40:11<34:04,  1.45it/s]
2|3498|Loss: 0.8039405941963196:  54%|█████▍    | 3497/6470 [40:11<34:04,  1.45it/s]
2|3498|Loss: 0.8039405941963196:  54%|█████▍    | 3498/6470 [40:12<32:53,  1.51it/s]
2|3499|Loss: 0.921086311340332:  54%|█████▍    | 3498/6470 [40:12<32:53,  1.51it/s] 
2|3499|Loss: 0.921086311340332:  54%|█████▍    | 3499/6470 [40:12<33:09,  1.49it/s]
2|3500|Loss: 0.8361383676528931:  54%|█████▍    | 3499/6470 [40:13<33:09,  1.49it/s]
2|3500|Loss: 0.8361383676528931:  54%|█████▍    | 3500/6470 [40:13<32:42,  1.51it/s]
2|3501|Loss: 1.0361446142196655:  54%|█████▍    | 3500/6470 [40:13<32:42,  1.51it/s]
2|3501|Loss: 1.0361446142196655:  54%|█████▍    | 3501/6470 [40:14<31:32,  1.57it/s]
2|3502|Loss: 0.8234603404998779:  54%|█████▍    | 3501/6470 [40:14<31:32,  1.57it/s]
2|3502|Loss: 0.8234603404998779:  54%|█████▍    | 3502/6470 [40:14<31:27,  1.57it/s]
2|3503|Loss: 0.8208373188972473:  54%|█████▍    | 3502/6470 [40:15<31:27,  1.57it/s]
2|3503|Loss: 0.8208373188972473:  54%|█████▍    | 3503/6470 [40:15<32:52,  1.50it/s]
2|3504|Loss: 0.8427577614784241:  54%|█████▍    | 3503/6470 [40:15<32:52,  1.50it/s]
2|3504|Loss: 0.8427577614784241:  54%|█████▍    | 3504/6470 [40:16<33:37,  1.47it/s]
2|3505|Loss: 0.8784691691398621:  54%|█████▍    | 3504/6470 [40:16<33:37,  1.47it/s]
2|3505|Loss: 0.8784691691398621:  54%|█████▍    | 3505/6470 [40:16<32:55,  1.50it/s]
2|3506|Loss: 0.7827835083007812:  54%|█████▍    | 3505/6470 [40:17<32:55,  1.50it/s]
2|3506|Loss: 0.7827835083007812:  54%|█████▍    | 3506/6470 [40:17<34:43,  1.42it/s]
2|3507|Loss: 0.7310340404510498:  54%|█████▍    | 3506/6470 [40:17<34:43,  1.42it/s]
2|3507|Loss: 0.7310340404510498:  54%|█████▍    | 3507/6470 [40:18<35:09,  1.40it/s]
2|3508|Loss: 0.6679410934448242:  54%|█████▍    | 3507/6470 [40:18<35:09,  1.40it/s]
2|3508|Loss: 0.6679410934448242:  54%|█████▍    | 3508/6470 [40:19<36:06,  1.37it/s]
2|3509|Loss: 0.7057963013648987:  54%|█████▍    | 3508/6470 [40:19<36:06,  1.37it/s]
2|3509|Loss: 0.7057963013648987:  54%|█████▍    | 3509/6470 [40:19<37:46,  1.31it/s]
2|3510|Loss: 0.8527243137359619:  54%|█████▍    | 3509/6470 [40:20<37:46,  1.31it/s]
2|3510|Loss: 0.8527243137359619:  54%|█████▍    | 3510/6470 [40:20<33:15,  1.48it/s]
2|3511|Loss: 0.8361653685569763:  54%|█████▍    | 3510/6470 [40:20<33:15,  1.48it/s]
2|3511|Loss: 0.8361653685569763:  54%|█████▍    | 3511/6470 [40:21<34:10,  1.44it/s]
2|3512|Loss: 0.7811302542686462:  54%|█████▍    | 3511/6470 [40:21<34:10,  1.44it/s]
2|3512|Loss: 0.7811302542686462:  54%|█████▍    | 3512/6470 [40:21<31:53,  1.55it/s]
2|3513|Loss: 0.920157790184021:  54%|█████▍    | 3512/6470 [40:21<31:53,  1.55it/s] 
2|3513|Loss: 0.920157790184021:  54%|█████▍    | 3513/6470 [40:22<31:09,  1.58it/s]
2|3514|Loss: 0.8770073652267456:  54%|█████▍    | 3513/6470 [40:22<31:09,  1.58it/s]
2|3514|Loss: 0.8770073652267456:  54%|█████▍    | 3514/6470 [40:22<31:18,  1.57it/s]
2|3515|Loss: 0.8680455088615417:  54%|█████▍    | 3514/6470 [40:23<31:18,  1.57it/s]
2|3515|Loss: 0.8680455088615417:  54%|█████▍    | 3515/6470 [40:23<32:43,  1.50it/s]
2|3516|Loss: 0.8193017840385437:  54%|█████▍    | 3515/6470 [40:23<32:43,  1.50it/s]
2|3516|Loss: 0.8193017840385437:  54%|█████▍    | 3516/6470 [40:24<31:13,  1.58it/s]
2|3517|Loss: 0.9762217998504639:  54%|█████▍    | 3516/6470 [40:24<31:13,  1.58it/s]
2|3517|Loss: 0.9762217998504639:  54%|█████▍    | 3517/6470 [40:24<33:11,  1.48it/s]
2|3518|Loss: 0.8982692956924438:  54%|█████▍    | 3517/6470 [40:25<33:11,  1.48it/s]
2|3518|Loss: 0.8982692956924438:  54%|█████▍    | 3518/6470 [40:25<36:29,  1.35it/s]
2|3519|Loss: 0.8319029808044434:  54%|█████▍    | 3518/6470 [40:26<36:29,  1.35it/s]
2|3519|Loss: 0.8319029808044434:  54%|█████▍    | 3519/6470 [40:26<36:34,  1.34it/s]
2|3520|Loss: 0.8563058376312256:  54%|█████▍    | 3519/6470 [40:26<36:34,  1.34it/s]
2|3520|Loss: 0.8563058376312256:  54%|█████▍    | 3520/6470 [40:27<34:09,  1.44it/s]
2|3521|Loss: 0.7698949575424194:  54%|█████▍    | 3520/6470 [40:27<34:09,  1.44it/s]
2|3521|Loss: 0.7698949575424194:  54%|█████▍    | 3521/6470 [40:27<35:22,  1.39it/s]
2|3522|Loss: 1.1346724033355713:  54%|█████▍    | 3521/6470 [40:28<35:22,  1.39it/s]
2|3522|Loss: 1.1346724033355713:  54%|█████▍    | 3522/6470 [40:28<37:21,  1.32it/s]
2|3523|Loss: 0.8019059300422668:  54%|█████▍    | 3522/6470 [40:29<37:21,  1.32it/s]
2|3523|Loss: 0.8019059300422668:  54%|█████▍    | 3523/6470 [40:29<35:06,  1.40it/s]
2|3524|Loss: 0.7442947030067444:  54%|█████▍    | 3523/6470 [40:29<35:06,  1.40it/s]
2|3524|Loss: 0.7442947030067444:  54%|█████▍    | 3524/6470 [40:30<36:29,  1.35it/s]
2|3525|Loss: 0.8267126083374023:  54%|█████▍    | 3524/6470 [40:30<36:29,  1.35it/s]
2|3525|Loss: 0.8267126083374023:  54%|█████▍    | 3525/6470 [40:30<36:31,  1.34it/s]
2|3526|Loss: 0.9109929800033569:  54%|█████▍    | 3525/6470 [40:31<36:31,  1.34it/s]
2|3526|Loss: 0.9109929800033569:  54%|█████▍    | 3526/6470 [40:31<37:31,  1.31it/s]
2|3527|Loss: 1.036370873451233:  54%|█████▍    | 3526/6470 [40:32<37:31,  1.31it/s] 
2|3527|Loss: 1.036370873451233:  55%|█████▍    | 3527/6470 [40:32<35:29,  1.38it/s]
2|3528|Loss: 0.816402018070221:  55%|█████▍    | 3527/6470 [40:32<35:29,  1.38it/s]
2|3528|Loss: 0.816402018070221:  55%|█████▍    | 3528/6470 [40:33<33:47,  1.45it/s]
2|3529|Loss: 0.8288929462432861:  55%|█████▍    | 3528/6470 [40:33<33:47,  1.45it/s]
2|3529|Loss: 0.8288929462432861:  55%|█████▍    | 3529/6470 [40:33<31:09,  1.57it/s]
2|3530|Loss: 0.9993060231208801:  55%|█████▍    | 3529/6470 [40:33<31:09,  1.57it/s]
2|3530|Loss: 0.9993060231208801:  55%|█████▍    | 3530/6470 [40:34<29:12,  1.68it/s]
2|3531|Loss: 0.9039146304130554:  55%|█████▍    | 3530/6470 [40:34<29:12,  1.68it/s]
2|3531|Loss: 0.9039146304130554:  55%|█████▍    | 3531/6470 [40:34<27:54,  1.76it/s]
2|3532|Loss: 1.0268412828445435:  55%|█████▍    | 3531/6470 [40:34<27:54,  1.76it/s]
2|3532|Loss: 1.0268412828445435:  55%|█████▍    | 3532/6470 [40:35<29:10,  1.68it/s]
2|3533|Loss: 0.8806681632995605:  55%|█████▍    | 3532/6470 [40:35<29:10,  1.68it/s]
2|3533|Loss: 0.8806681632995605:  55%|█████▍    | 3533/6470 [40:35<29:43,  1.65it/s]
2|3534|Loss: 0.8423743844032288:  55%|█████▍    | 3533/6470 [40:36<29:43,  1.65it/s]
2|3534|Loss: 0.8423743844032288:  55%|█████▍    | 3534/6470 [40:36<29:21,  1.67it/s]
2|3535|Loss: 0.8766180872917175:  55%|█████▍    | 3534/6470 [40:36<29:21,  1.67it/s]
2|3535|Loss: 0.8766180872917175:  55%|█████▍    | 3535/6470 [40:37<30:49,  1.59it/s]
2|3536|Loss: 0.8709376454353333:  55%|█████▍    | 3535/6470 [40:37<30:49,  1.59it/s]
2|3536|Loss: 0.8709376454353333:  55%|█████▍    | 3536/6470 [40:37<33:41,  1.45it/s]
2|3537|Loss: 0.8132362961769104:  55%|█████▍    | 3536/6470 [40:38<33:41,  1.45it/s]
2|3537|Loss: 0.8132362961769104:  55%|█████▍    | 3537/6470 [40:38<32:58,  1.48it/s]
2|3538|Loss: 1.0611042976379395:  55%|█████▍    | 3537/6470 [40:38<32:58,  1.48it/s]
2|3538|Loss: 1.0611042976379395:  55%|█████▍    | 3538/6470 [40:39<33:19,  1.47it/s]
2|3539|Loss: 0.8550894260406494:  55%|█████▍    | 3538/6470 [40:39<33:19,  1.47it/s]
2|3539|Loss: 0.8550894260406494:  55%|█████▍    | 3539/6470 [40:39<32:44,  1.49it/s]
2|3540|Loss: 0.805152177810669:  55%|█████▍    | 3539/6470 [40:40<32:44,  1.49it/s] 
2|3540|Loss: 0.805152177810669:  55%|█████▍    | 3540/6470 [40:40<32:16,  1.51it/s]
2|3541|Loss: 0.8434281945228577:  55%|█████▍    | 3540/6470 [40:41<32:16,  1.51it/s]
2|3541|Loss: 0.8434281945228577:  55%|█████▍    | 3541/6470 [40:41<34:31,  1.41it/s]
2|3542|Loss: 0.8568673729896545:  55%|█████▍    | 3541/6470 [40:41<34:31,  1.41it/s]
2|3542|Loss: 0.8568673729896545:  55%|█████▍    | 3542/6470 [40:42<33:27,  1.46it/s]
2|3543|Loss: 0.8520179986953735:  55%|█████▍    | 3542/6470 [40:42<33:27,  1.46it/s]
2|3543|Loss: 0.8520179986953735:  55%|█████▍    | 3543/6470 [40:42<32:36,  1.50it/s]
2|3544|Loss: 0.8677853345870972:  55%|█████▍    | 3543/6470 [40:43<32:36,  1.50it/s]
2|3544|Loss: 0.8677853345870972:  55%|█████▍    | 3544/6470 [40:43<32:16,  1.51it/s]
2|3545|Loss: 0.6892008185386658:  55%|█████▍    | 3544/6470 [40:43<32:16,  1.51it/s]
2|3545|Loss: 0.6892008185386658:  55%|█████▍    | 3545/6470 [40:44<33:37,  1.45it/s]
2|3546|Loss: 0.9153910875320435:  55%|█████▍    | 3545/6470 [40:44<33:37,  1.45it/s]
2|3546|Loss: 0.9153910875320435:  55%|█████▍    | 3546/6470 [40:44<31:52,  1.53it/s]
2|3547|Loss: 0.6401851177215576:  55%|█████▍    | 3546/6470 [40:44<31:52,  1.53it/s]
2|3547|Loss: 0.6401851177215576:  55%|█████▍    | 3547/6470 [40:45<32:16,  1.51it/s]
2|3548|Loss: 0.8372043371200562:  55%|█████▍    | 3547/6470 [40:45<32:16,  1.51it/s]
2|3548|Loss: 0.8372043371200562:  55%|█████▍    | 3548/6470 [40:45<29:14,  1.67it/s]
2|3549|Loss: 0.9291389584541321:  55%|█████▍    | 3548/6470 [40:46<29:14,  1.67it/s]
2|3549|Loss: 0.9291389584541321:  55%|█████▍    | 3549/6470 [40:46<30:28,  1.60it/s]
2|3550|Loss: 0.8310181498527527:  55%|█████▍    | 3549/6470 [40:46<30:28,  1.60it/s]
2|3550|Loss: 0.8310181498527527:  55%|█████▍    | 3550/6470 [40:47<31:51,  1.53it/s]
2|3551|Loss: 0.9062412977218628:  55%|█████▍    | 3550/6470 [40:47<31:51,  1.53it/s]
2|3551|Loss: 0.9062412977218628:  55%|█████▍    | 3551/6470 [40:47<30:00,  1.62it/s]
2|3552|Loss: 0.8498472571372986:  55%|█████▍    | 3551/6470 [40:48<30:00,  1.62it/s]
2|3552|Loss: 0.8498472571372986:  55%|█████▍    | 3552/6470 [40:48<31:35,  1.54it/s]
2|3553|Loss: 0.6851226091384888:  55%|█████▍    | 3552/6470 [40:48<31:35,  1.54it/s]
2|3553|Loss: 0.6851226091384888:  55%|█████▍    | 3553/6470 [40:49<31:56,  1.52it/s]
2|3554|Loss: 0.6612924337387085:  55%|█████▍    | 3553/6470 [40:49<31:56,  1.52it/s]
2|3554|Loss: 0.6612924337387085:  55%|█████▍    | 3554/6470 [40:49<31:00,  1.57it/s]
2|3555|Loss: 0.9172140955924988:  55%|█████▍    | 3554/6470 [40:50<31:00,  1.57it/s]
2|3555|Loss: 0.9172140955924988:  55%|█████▍    | 3555/6470 [40:50<33:04,  1.47it/s]
2|3556|Loss: 0.8588528037071228:  55%|█████▍    | 3555/6470 [40:50<33:04,  1.47it/s]
2|3556|Loss: 0.8588528037071228:  55%|█████▍    | 3556/6470 [40:51<34:04,  1.43it/s]
2|3557|Loss: 0.7091629505157471:  55%|█████▍    | 3556/6470 [40:51<34:04,  1.43it/s]
2|3557|Loss: 0.7091629505157471:  55%|█████▍    | 3557/6470 [40:51<33:25,  1.45it/s]
2|3558|Loss: 0.8405771255493164:  55%|█████▍    | 3557/6470 [40:52<33:25,  1.45it/s]
2|3558|Loss: 0.8405771255493164:  55%|█████▍    | 3558/6470 [40:52<32:41,  1.48it/s]
2|3559|Loss: 0.7355137467384338:  55%|█████▍    | 3558/6470 [40:52<32:41,  1.48it/s]
2|3559|Loss: 0.7355137467384338:  55%|█████▌    | 3559/6470 [40:53<33:00,  1.47it/s]
2|3560|Loss: 0.722477376461029:  55%|█████▌    | 3559/6470 [40:53<33:00,  1.47it/s] 
2|3560|Loss: 0.722477376461029:  55%|█████▌    | 3560/6470 [40:54<34:47,  1.39it/s]
2|3561|Loss: 0.8396490812301636:  55%|█████▌    | 3560/6470 [40:54<34:47,  1.39it/s]
2|3561|Loss: 0.8396490812301636:  55%|█████▌    | 3561/6470 [40:54<34:04,  1.42it/s]
2|3562|Loss: 0.9885168075561523:  55%|█████▌    | 3561/6470 [40:55<34:04,  1.42it/s]
2|3562|Loss: 0.9885168075561523:  55%|█████▌    | 3562/6470 [40:55<32:45,  1.48it/s]
2|3563|Loss: 1.048929214477539:  55%|█████▌    | 3562/6470 [40:55<32:45,  1.48it/s] 
2|3563|Loss: 1.048929214477539:  55%|█████▌    | 3563/6470 [40:55<32:51,  1.47it/s]
2|3564|Loss: 0.9197337031364441:  55%|█████▌    | 3563/6470 [40:56<32:51,  1.47it/s]
2|3564|Loss: 0.9197337031364441:  55%|█████▌    | 3564/6470 [40:56<33:50,  1.43it/s]
2|3565|Loss: 0.9397392868995667:  55%|█████▌    | 3564/6470 [40:57<33:50,  1.43it/s]
2|3565|Loss: 0.9397392868995667:  55%|█████▌    | 3565/6470 [40:57<34:45,  1.39it/s]
2|3566|Loss: 0.8608899712562561:  55%|█████▌    | 3565/6470 [40:57<34:45,  1.39it/s]
2|3566|Loss: 0.8608899712562561:  55%|█████▌    | 3566/6470 [40:58<33:50,  1.43it/s]
2|3567|Loss: 0.870360791683197:  55%|█████▌    | 3566/6470 [40:58<33:50,  1.43it/s] 
2|3567|Loss: 0.870360791683197:  55%|█████▌    | 3567/6470 [40:58<34:04,  1.42it/s]
2|3568|Loss: 0.8678004741668701:  55%|█████▌    | 3567/6470 [40:59<34:04,  1.42it/s]
2|3568|Loss: 0.8678004741668701:  55%|█████▌    | 3568/6470 [40:59<33:12,  1.46it/s]
2|3569|Loss: 0.9380576014518738:  55%|█████▌    | 3568/6470 [40:59<33:12,  1.46it/s]
2|3569|Loss: 0.9380576014518738:  55%|█████▌    | 3569/6470 [41:00<34:37,  1.40it/s]
2|3570|Loss: 0.9131599068641663:  55%|█████▌    | 3569/6470 [41:00<34:37,  1.40it/s]
2|3570|Loss: 0.9131599068641663:  55%|█████▌    | 3570/6470 [41:00<34:06,  1.42it/s]
2|3571|Loss: 0.9672437310218811:  55%|█████▌    | 3570/6470 [41:01<34:06,  1.42it/s]
2|3571|Loss: 0.9672437310218811:  55%|█████▌    | 3571/6470 [41:01<31:17,  1.54it/s]
2|3572|Loss: 0.770160973072052:  55%|█████▌    | 3571/6470 [41:01<31:17,  1.54it/s] 
2|3572|Loss: 0.770160973072052:  55%|█████▌    | 3572/6470 [41:02<30:26,  1.59it/s]
2|3573|Loss: 0.9051871299743652:  55%|█████▌    | 3572/6470 [41:02<30:26,  1.59it/s]
2|3573|Loss: 0.9051871299743652:  55%|█████▌    | 3573/6470 [41:02<33:10,  1.46it/s]
2|3574|Loss: 0.9143277406692505:  55%|█████▌    | 3573/6470 [41:03<33:10,  1.46it/s]
2|3574|Loss: 0.9143277406692505:  55%|█████▌    | 3574/6470 [41:03<33:54,  1.42it/s]
2|3575|Loss: 0.8448522686958313:  55%|█████▌    | 3574/6470 [41:04<33:54,  1.42it/s]
2|3575|Loss: 0.8448522686958313:  55%|█████▌    | 3575/6470 [41:04<33:08,  1.46it/s]
2|3576|Loss: 0.8768680691719055:  55%|█████▌    | 3575/6470 [41:04<33:08,  1.46it/s]
2|3576|Loss: 0.8768680691719055:  55%|█████▌    | 3576/6470 [41:05<33:42,  1.43it/s]
2|3577|Loss: 0.8090516328811646:  55%|█████▌    | 3576/6470 [41:05<33:42,  1.43it/s]
2|3577|Loss: 0.8090516328811646:  55%|█████▌    | 3577/6470 [41:05<35:03,  1.38it/s]
2|3578|Loss: 0.8436752557754517:  55%|█████▌    | 3577/6470 [41:06<35:03,  1.38it/s]
2|3578|Loss: 0.8436752557754517:  55%|█████▌    | 3578/6470 [41:06<35:18,  1.36it/s]
2|3579|Loss: 0.8312543630599976:  55%|█████▌    | 3578/6470 [41:07<35:18,  1.36it/s]
2|3579|Loss: 0.8312543630599976:  55%|█████▌    | 3579/6470 [41:07<36:38,  1.32it/s]
2|3580|Loss: 0.7588556408882141:  55%|█████▌    | 3579/6470 [41:07<36:38,  1.32it/s]
2|3580|Loss: 0.7588556408882141:  55%|█████▌    | 3580/6470 [41:08<39:53,  1.21it/s]
2|3581|Loss: 0.8708391785621643:  55%|█████▌    | 3580/6470 [41:08<39:53,  1.21it/s]
2|3581|Loss: 0.8708391785621643:  55%|█████▌    | 3581/6470 [41:09<41:14,  1.17it/s]
2|3582|Loss: 0.8983845114707947:  55%|█████▌    | 3581/6470 [41:09<41:14,  1.17it/s]
2|3582|Loss: 0.8983845114707947:  55%|█████▌    | 3582/6470 [41:10<39:53,  1.21it/s]
2|3583|Loss: 0.8021647334098816:  55%|█████▌    | 3582/6470 [41:10<39:53,  1.21it/s]
2|3583|Loss: 0.8021647334098816:  55%|█████▌    | 3583/6470 [41:10<39:05,  1.23it/s]
2|3584|Loss: 0.9628726840019226:  55%|█████▌    | 3583/6470 [41:11<39:05,  1.23it/s]
2|3584|Loss: 0.9628726840019226:  55%|█████▌    | 3584/6470 [41:11<36:52,  1.30it/s]
2|3585|Loss: 0.9859839677810669:  55%|█████▌    | 3584/6470 [41:11<36:52,  1.30it/s]
2|3585|Loss: 0.9859839677810669:  55%|█████▌    | 3585/6470 [41:12<35:59,  1.34it/s]
2|3586|Loss: 0.9290381669998169:  55%|█████▌    | 3585/6470 [41:12<35:59,  1.34it/s]
2|3586|Loss: 0.9290381669998169:  55%|█████▌    | 3586/6470 [41:12<32:55,  1.46it/s]
2|3587|Loss: 0.9190577864646912:  55%|█████▌    | 3586/6470 [41:13<32:55,  1.46it/s]
2|3587|Loss: 0.9190577864646912:  55%|█████▌    | 3587/6470 [41:13<32:52,  1.46it/s]
2|3588|Loss: 0.6192675828933716:  55%|█████▌    | 3587/6470 [41:13<32:52,  1.46it/s]
2|3588|Loss: 0.6192675828933716:  55%|█████▌    | 3588/6470 [41:14<34:14,  1.40it/s]
2|3589|Loss: 0.9579151272773743:  55%|█████▌    | 3588/6470 [41:14<34:14,  1.40it/s]
2|3589|Loss: 0.9579151272773743:  55%|█████▌    | 3589/6470 [41:14<33:40,  1.43it/s]
2|3590|Loss: 0.769135594367981:  55%|█████▌    | 3589/6470 [41:15<33:40,  1.43it/s] 
2|3590|Loss: 0.769135594367981:  55%|█████▌    | 3590/6470 [41:15<35:52,  1.34it/s]
2|3591|Loss: 0.9077602028846741:  55%|█████▌    | 3590/6470 [41:16<35:52,  1.34it/s]
2|3591|Loss: 0.9077602028846741:  56%|█████▌    | 3591/6470 [41:16<37:34,  1.28it/s]
2|3592|Loss: 1.118041753768921:  56%|█████▌    | 3591/6470 [41:17<37:34,  1.28it/s] 
2|3592|Loss: 1.118041753768921:  56%|█████▌    | 3592/6470 [41:17<38:39,  1.24it/s]
2|3593|Loss: 0.863864004611969:  56%|█████▌    | 3592/6470 [41:17<38:39,  1.24it/s]
2|3593|Loss: 0.863864004611969:  56%|█████▌    | 3593/6470 [41:18<37:01,  1.30it/s]
2|3594|Loss: 0.8271030187606812:  56%|█████▌    | 3593/6470 [41:18<37:01,  1.30it/s]
2|3594|Loss: 0.8271030187606812:  56%|█████▌    | 3594/6470 [41:18<37:17,  1.29it/s]
2|3595|Loss: 0.9340630173683167:  56%|█████▌    | 3594/6470 [41:19<37:17,  1.29it/s]
2|3595|Loss: 0.9340630173683167:  56%|█████▌    | 3595/6470 [41:19<37:35,  1.27it/s]
2|3596|Loss: 0.8691256046295166:  56%|█████▌    | 3595/6470 [41:20<37:35,  1.27it/s]
2|3596|Loss: 0.8691256046295166:  56%|█████▌    | 3596/6470 [41:20<37:15,  1.29it/s]
2|3597|Loss: 0.8681612610816956:  56%|█████▌    | 3596/6470 [41:20<37:15,  1.29it/s]
2|3597|Loss: 0.8681612610816956:  56%|█████▌    | 3597/6470 [41:21<35:26,  1.35it/s]
2|3598|Loss: 0.8565073013305664:  56%|█████▌    | 3597/6470 [41:21<35:26,  1.35it/s]
2|3598|Loss: 0.8565073013305664:  56%|█████▌    | 3598/6470 [41:21<32:40,  1.47it/s]
2|3599|Loss: 0.7782848477363586:  56%|█████▌    | 3598/6470 [41:22<32:40,  1.47it/s]
2|3599|Loss: 0.7782848477363586:  56%|█████▌    | 3599/6470 [41:22<32:33,  1.47it/s]
2|3600|Loss: 0.8911757469177246:  56%|█████▌    | 3599/6470 [41:22<32:33,  1.47it/s]
2|3600|Loss: 0.8911757469177246:  56%|█████▌    | 3600/6470 [41:23<31:59,  1.49it/s]
2|3601|Loss: 0.8800347447395325:  56%|█████▌    | 3600/6470 [41:23<31:59,  1.49it/s]
2|3601|Loss: 0.8800347447395325:  56%|█████▌    | 3601/6470 [41:23<31:10,  1.53it/s]
2|3602|Loss: 0.8638778328895569:  56%|█████▌    | 3601/6470 [41:24<31:10,  1.53it/s]
2|3602|Loss: 0.8638778328895569:  56%|█████▌    | 3602/6470 [41:24<31:57,  1.50it/s]
2|3603|Loss: 0.8697082996368408:  56%|█████▌    | 3602/6470 [41:24<31:57,  1.50it/s]
2|3603|Loss: 0.8697082996368408:  56%|█████▌    | 3603/6470 [41:25<32:51,  1.45it/s]
2|3604|Loss: 0.7704916000366211:  56%|█████▌    | 3603/6470 [41:25<32:51,  1.45it/s]
2|3604|Loss: 0.7704916000366211:  56%|█████▌    | 3604/6470 [41:25<33:10,  1.44it/s]
2|3605|Loss: 0.975251317024231:  56%|█████▌    | 3604/6470 [41:26<33:10,  1.44it/s] 
2|3605|Loss: 0.975251317024231:  56%|█████▌    | 3605/6470 [41:26<31:26,  1.52it/s]
2|3606|Loss: 0.8579073548316956:  56%|█████▌    | 3605/6470 [41:26<31:26,  1.52it/s]
2|3606|Loss: 0.8579073548316956:  56%|█████▌    | 3606/6470 [41:26<29:24,  1.62it/s]
2|3607|Loss: 0.8187657594680786:  56%|█████▌    | 3606/6470 [41:27<29:24,  1.62it/s]
2|3607|Loss: 0.8187657594680786:  56%|█████▌    | 3607/6470 [41:27<28:03,  1.70it/s]
2|3608|Loss: 0.8805556893348694:  56%|█████▌    | 3607/6470 [41:27<28:03,  1.70it/s]
2|3608|Loss: 0.8805556893348694:  56%|█████▌    | 3608/6470 [41:27<28:09,  1.69it/s]
2|3609|Loss: 0.8740721940994263:  56%|█████▌    | 3608/6470 [41:28<28:09,  1.69it/s]
2|3609|Loss: 0.8740721940994263:  56%|█████▌    | 3609/6470 [41:28<30:54,  1.54it/s]
2|3610|Loss: 0.8289369344711304:  56%|█████▌    | 3609/6470 [41:29<30:54,  1.54it/s]
2|3610|Loss: 0.8289369344711304:  56%|█████▌    | 3610/6470 [41:29<32:06,  1.48it/s]
2|3611|Loss: 0.8933323621749878:  56%|█████▌    | 3610/6470 [41:29<32:06,  1.48it/s]
2|3611|Loss: 0.8933323621749878:  56%|█████▌    | 3611/6470 [41:30<34:00,  1.40it/s]
2|3612|Loss: 0.7832733988761902:  56%|█████▌    | 3611/6470 [41:30<34:00,  1.40it/s]
2|3612|Loss: 0.7832733988761902:  56%|█████▌    | 3612/6470 [41:30<31:44,  1.50it/s]
2|3613|Loss: 0.891830325126648:  56%|█████▌    | 3612/6470 [41:31<31:44,  1.50it/s] 
2|3613|Loss: 0.891830325126648:  56%|█████▌    | 3613/6470 [41:31<32:09,  1.48it/s]
2|3614|Loss: 0.6883625388145447:  56%|█████▌    | 3613/6470 [41:31<32:09,  1.48it/s]
2|3614|Loss: 0.6883625388145447:  56%|█████▌    | 3614/6470 [41:32<31:46,  1.50it/s]
2|3615|Loss: 0.678408145904541:  56%|█████▌    | 3614/6470 [41:32<31:46,  1.50it/s] 
2|3615|Loss: 0.678408145904541:  56%|█████▌    | 3615/6470 [41:33<34:28,  1.38it/s]
2|3616|Loss: 0.7131783962249756:  56%|█████▌    | 3615/6470 [41:33<34:28,  1.38it/s]
2|3616|Loss: 0.7131783962249756:  56%|█████▌    | 3616/6470 [41:33<34:49,  1.37it/s]
2|3617|Loss: 0.8523328304290771:  56%|█████▌    | 3616/6470 [41:34<34:49,  1.37it/s]
2|3617|Loss: 0.8523328304290771:  56%|█████▌    | 3617/6470 [41:34<33:24,  1.42it/s]
2|3618|Loss: 0.8531603813171387:  56%|█████▌    | 3617/6470 [41:34<33:24,  1.42it/s]
2|3618|Loss: 0.8531603813171387:  56%|█████▌    | 3618/6470 [41:34<29:13,  1.63it/s]
2|3619|Loss: 0.7228140234947205:  56%|█████▌    | 3618/6470 [41:35<29:13,  1.63it/s]
2|3619|Loss: 0.7228140234947205:  56%|█████▌    | 3619/6470 [41:35<29:57,  1.59it/s]
2|3620|Loss: 0.8823696374893188:  56%|█████▌    | 3619/6470 [41:35<29:57,  1.59it/s]
2|3620|Loss: 0.8823696374893188:  56%|█████▌    | 3620/6470 [41:36<32:21,  1.47it/s]
2|3621|Loss: 0.9588280916213989:  56%|█████▌    | 3620/6470 [41:36<32:21,  1.47it/s]
2|3621|Loss: 0.9588280916213989:  56%|█████▌    | 3621/6470 [41:37<34:39,  1.37it/s]
2|3622|Loss: 0.9462252259254456:  56%|█████▌    | 3621/6470 [41:37<34:39,  1.37it/s]
2|3622|Loss: 0.9462252259254456:  56%|█████▌    | 3622/6470 [41:37<35:13,  1.35it/s]
2|3623|Loss: 0.8389377593994141:  56%|█████▌    | 3622/6470 [41:38<35:13,  1.35it/s]
2|3623|Loss: 0.8389377593994141:  56%|█████▌    | 3623/6470 [41:38<34:46,  1.36it/s]
2|3624|Loss: 0.938274085521698:  56%|█████▌    | 3623/6470 [41:39<34:46,  1.36it/s] 
2|3624|Loss: 0.938274085521698:  56%|█████▌    | 3624/6470 [41:39<33:40,  1.41it/s]
2|3625|Loss: 0.9904683828353882:  56%|█████▌    | 3624/6470 [41:39<33:40,  1.41it/s]
2|3625|Loss: 0.9904683828353882:  56%|█████▌    | 3625/6470 [41:40<34:51,  1.36it/s]
2|3626|Loss: 0.774383544921875:  56%|█████▌    | 3625/6470 [41:40<34:51,  1.36it/s] 
2|3626|Loss: 0.774383544921875:  56%|█████▌    | 3626/6470 [41:40<36:36,  1.29it/s]
2|3627|Loss: 0.7697139382362366:  56%|█████▌    | 3626/6470 [41:41<36:36,  1.29it/s]
2|3627|Loss: 0.7697139382362366:  56%|█████▌    | 3627/6470 [41:41<34:52,  1.36it/s]
2|3628|Loss: 0.7404835820198059:  56%|█████▌    | 3627/6470 [41:42<34:52,  1.36it/s]
2|3628|Loss: 0.7404835820198059:  56%|█████▌    | 3628/6470 [41:42<36:32,  1.30it/s]
2|3629|Loss: 0.9184956550598145:  56%|█████▌    | 3628/6470 [41:42<36:32,  1.30it/s]
2|3629|Loss: 0.9184956550598145:  56%|█████▌    | 3629/6470 [41:43<36:31,  1.30it/s]
2|3630|Loss: 0.8728046417236328:  56%|█████▌    | 3629/6470 [41:43<36:31,  1.30it/s]
2|3630|Loss: 0.8728046417236328:  56%|█████▌    | 3630/6470 [41:44<37:21,  1.27it/s]
2|3631|Loss: 0.7978151440620422:  56%|█████▌    | 3630/6470 [41:44<37:21,  1.27it/s]
2|3631|Loss: 0.7978151440620422:  56%|█████▌    | 3631/6470 [41:44<39:00,  1.21it/s]
2|3632|Loss: 0.9164777994155884:  56%|█████▌    | 3631/6470 [41:45<39:00,  1.21it/s]
2|3632|Loss: 0.9164777994155884:  56%|█████▌    | 3632/6470 [41:45<37:10,  1.27it/s]
2|3633|Loss: 0.8930367231369019:  56%|█████▌    | 3632/6470 [41:46<37:10,  1.27it/s]
2|3633|Loss: 0.8930367231369019:  56%|█████▌    | 3633/6470 [41:46<37:31,  1.26it/s]
2|3634|Loss: 0.686407744884491:  56%|█████▌    | 3633/6470 [41:46<37:31,  1.26it/s] 
2|3634|Loss: 0.686407744884491:  56%|█████▌    | 3634/6470 [41:46<32:57,  1.43it/s]
2|3635|Loss: 0.7735986709594727:  56%|█████▌    | 3634/6470 [41:47<32:57,  1.43it/s]
2|3635|Loss: 0.7735986709594727:  56%|█████▌    | 3635/6470 [41:47<33:22,  1.42it/s]
2|3636|Loss: 0.829852819442749:  56%|█████▌    | 3635/6470 [41:48<33:22,  1.42it/s] 
2|3636|Loss: 0.829852819442749:  56%|█████▌    | 3636/6470 [41:48<33:42,  1.40it/s]
2|3637|Loss: 0.8307557106018066:  56%|█████▌    | 3636/6470 [41:48<33:42,  1.40it/s]
2|3637|Loss: 0.8307557106018066:  56%|█████▌    | 3637/6470 [41:49<35:38,  1.32it/s]
2|3638|Loss: 0.7539746761322021:  56%|█████▌    | 3637/6470 [41:49<35:38,  1.32it/s]
2|3638|Loss: 0.7539746761322021:  56%|█████▌    | 3638/6470 [41:50<36:01,  1.31it/s]
2|3639|Loss: 0.7487437725067139:  56%|█████▌    | 3638/6470 [41:50<36:01,  1.31it/s]
2|3639|Loss: 0.7487437725067139:  56%|█████▌    | 3639/6470 [41:50<34:25,  1.37it/s]
2|3640|Loss: 0.8789400458335876:  56%|█████▌    | 3639/6470 [41:51<34:25,  1.37it/s]
2|3640|Loss: 0.8789400458335876:  56%|█████▋    | 3640/6470 [41:51<34:58,  1.35it/s]
2|3641|Loss: 0.7416009306907654:  56%|█████▋    | 3640/6470 [41:51<34:58,  1.35it/s]
2|3641|Loss: 0.7416009306907654:  56%|█████▋    | 3641/6470 [41:52<34:39,  1.36it/s]
2|3642|Loss: 0.7792658805847168:  56%|█████▋    | 3641/6470 [41:52<34:39,  1.36it/s]
2|3642|Loss: 0.7792658805847168:  56%|█████▋    | 3642/6470 [41:53<36:02,  1.31it/s]
2|3643|Loss: 0.9164202213287354:  56%|█████▋    | 3642/6470 [41:53<36:02,  1.31it/s]
2|3643|Loss: 0.9164202213287354:  56%|█████▋    | 3643/6470 [41:53<34:19,  1.37it/s]
2|3644|Loss: 0.855957567691803:  56%|█████▋    | 3643/6470 [41:54<34:19,  1.37it/s] 
2|3644|Loss: 0.855957567691803:  56%|█████▋    | 3644/6470 [41:54<36:03,  1.31it/s]
2|3645|Loss: 0.958638072013855:  56%|█████▋    | 3644/6470 [41:54<36:03,  1.31it/s]
2|3645|Loss: 0.958638072013855:  56%|█████▋    | 3645/6470 [41:55<36:06,  1.30it/s]
2|3646|Loss: 0.7356640696525574:  56%|█████▋    | 3645/6470 [41:55<36:06,  1.30it/s]
2|3646|Loss: 0.7356640696525574:  56%|█████▋    | 3646/6470 [41:55<33:54,  1.39it/s]
2|3647|Loss: 0.6308618783950806:  56%|█████▋    | 3646/6470 [41:56<33:54,  1.39it/s]
2|3647|Loss: 0.6308618783950806:  56%|█████▋    | 3647/6470 [41:56<33:53,  1.39it/s]
2|3648|Loss: 0.9799855947494507:  56%|█████▋    | 3647/6470 [41:57<33:53,  1.39it/s]
2|3648|Loss: 0.9799855947494507:  56%|█████▋    | 3648/6470 [41:57<34:47,  1.35it/s]
2|3649|Loss: 0.7896873354911804:  56%|█████▋    | 3648/6470 [41:57<34:47,  1.35it/s]
2|3649|Loss: 0.7896873354911804:  56%|█████▋    | 3649/6470 [41:58<35:02,  1.34it/s]
2|3650|Loss: 0.8745014667510986:  56%|█████▋    | 3649/6470 [41:58<35:02,  1.34it/s]
2|3650|Loss: 0.8745014667510986:  56%|█████▋    | 3650/6470 [41:59<37:00,  1.27it/s]
2|3651|Loss: 0.8101509809494019:  56%|█████▋    | 3650/6470 [41:59<37:00,  1.27it/s]
2|3651|Loss: 0.8101509809494019:  56%|█████▋    | 3651/6470 [41:59<37:32,  1.25it/s]
2|3652|Loss: 0.838852047920227:  56%|█████▋    | 3651/6470 [42:00<37:32,  1.25it/s] 
2|3652|Loss: 0.838852047920227:  56%|█████▋    | 3652/6470 [42:00<36:50,  1.27it/s]
2|3653|Loss: 0.9624207019805908:  56%|█████▋    | 3652/6470 [42:01<36:50,  1.27it/s]
2|3653|Loss: 0.9624207019805908:  56%|█████▋    | 3653/6470 [42:01<36:15,  1.29it/s]
2|3654|Loss: 0.8572084307670593:  56%|█████▋    | 3653/6470 [42:01<36:15,  1.29it/s]
2|3654|Loss: 0.8572084307670593:  56%|█████▋    | 3654/6470 [42:02<34:32,  1.36it/s]
2|3655|Loss: 0.8881315588951111:  56%|█████▋    | 3654/6470 [42:02<34:32,  1.36it/s]
2|3655|Loss: 0.8881315588951111:  56%|█████▋    | 3655/6470 [42:02<33:05,  1.42it/s]
2|3656|Loss: 0.7868797183036804:  56%|█████▋    | 3655/6470 [42:02<33:05,  1.42it/s]
2|3656|Loss: 0.7868797183036804:  57%|█████▋    | 3656/6470 [42:03<30:41,  1.53it/s]
2|3657|Loss: 0.8502581119537354:  57%|█████▋    | 3656/6470 [42:03<30:41,  1.53it/s]
2|3657|Loss: 0.8502581119537354:  57%|█████▋    | 3657/6470 [42:03<28:23,  1.65it/s]
2|3658|Loss: 1.0018162727355957:  57%|█████▋    | 3657/6470 [42:04<28:23,  1.65it/s]
2|3658|Loss: 1.0018162727355957:  57%|█████▋    | 3658/6470 [42:04<29:07,  1.61it/s]
2|3659|Loss: 0.8323163390159607:  57%|█████▋    | 3658/6470 [42:04<29:07,  1.61it/s]
2|3659|Loss: 0.8323163390159607:  57%|█████▋    | 3659/6470 [42:05<30:45,  1.52it/s]
2|3660|Loss: 0.7953072786331177:  57%|█████▋    | 3659/6470 [42:05<30:45,  1.52it/s]
2|3660|Loss: 0.7953072786331177:  57%|█████▋    | 3660/6470 [42:05<31:58,  1.46it/s]
2|3661|Loss: 0.9422284364700317:  57%|█████▋    | 3660/6470 [42:06<31:58,  1.46it/s]
2|3661|Loss: 0.9422284364700317:  57%|█████▋    | 3661/6470 [42:06<30:15,  1.55it/s]
2|3662|Loss: 0.8256058692932129:  57%|█████▋    | 3661/6470 [42:06<30:15,  1.55it/s]
2|3662|Loss: 0.8256058692932129:  57%|█████▋    | 3662/6470 [42:06<29:14,  1.60it/s]
2|3663|Loss: 0.748840868473053:  57%|█████▋    | 3662/6470 [42:07<29:14,  1.60it/s] 
2|3663|Loss: 0.748840868473053:  57%|█████▋    | 3663/6470 [42:07<32:50,  1.42it/s]
2|3664|Loss: 0.7428068518638611:  57%|█████▋    | 3663/6470 [42:08<32:50,  1.42it/s]
2|3664|Loss: 0.7428068518638611:  57%|█████▋    | 3664/6470 [42:08<32:08,  1.46it/s]
2|3665|Loss: 0.8660674095153809:  57%|█████▋    | 3664/6470 [42:08<32:08,  1.46it/s]
2|3665|Loss: 0.8660674095153809:  57%|█████▋    | 3665/6470 [42:09<31:04,  1.50it/s]
2|3666|Loss: 1.0768003463745117:  57%|█████▋    | 3665/6470 [42:09<31:04,  1.50it/s]
2|3666|Loss: 1.0768003463745117:  57%|█████▋    | 3666/6470 [42:09<32:48,  1.42it/s]
2|3667|Loss: 0.7999898195266724:  57%|█████▋    | 3666/6470 [42:10<32:48,  1.42it/s]
2|3667|Loss: 0.7999898195266724:  57%|█████▋    | 3667/6470 [42:10<34:14,  1.36it/s]
2|3668|Loss: 0.9071381688117981:  57%|█████▋    | 3667/6470 [42:11<34:14,  1.36it/s]
2|3668|Loss: 0.9071381688117981:  57%|█████▋    | 3668/6470 [42:11<34:36,  1.35it/s]
2|3669|Loss: 0.8289957642555237:  57%|█████▋    | 3668/6470 [42:11<34:36,  1.35it/s]
2|3669|Loss: 0.8289957642555237:  57%|█████▋    | 3669/6470 [42:12<32:32,  1.43it/s]
2|3670|Loss: 0.8440732955932617:  57%|█████▋    | 3669/6470 [42:12<32:32,  1.43it/s]
2|3670|Loss: 0.8440732955932617:  57%|█████▋    | 3670/6470 [42:12<32:30,  1.44it/s]
2|3671|Loss: 1.0680346488952637:  57%|█████▋    | 3670/6470 [42:13<32:30,  1.44it/s]
2|3671|Loss: 1.0680346488952637:  57%|█████▋    | 3671/6470 [42:13<29:51,  1.56it/s]
2|3672|Loss: 1.1306697130203247:  57%|█████▋    | 3671/6470 [42:13<29:51,  1.56it/s]
2|3672|Loss: 1.1306697130203247:  57%|█████▋    | 3672/6470 [42:14<31:21,  1.49it/s]
2|3673|Loss: 0.9413171410560608:  57%|█████▋    | 3672/6470 [42:14<31:21,  1.49it/s]
2|3673|Loss: 0.9413171410560608:  57%|█████▋    | 3673/6470 [42:14<31:15,  1.49it/s]
2|3674|Loss: 0.794150710105896:  57%|█████▋    | 3673/6470 [42:15<31:15,  1.49it/s] 
2|3674|Loss: 0.794150710105896:  57%|█████▋    | 3674/6470 [42:15<32:49,  1.42it/s]
2|3675|Loss: 0.8211492896080017:  57%|█████▋    | 3674/6470 [42:15<32:49,  1.42it/s]
2|3675|Loss: 0.8211492896080017:  57%|█████▋    | 3675/6470 [42:16<33:10,  1.40it/s]
2|3676|Loss: 0.7619308233261108:  57%|█████▋    | 3675/6470 [42:16<33:10,  1.40it/s]
2|3676|Loss: 0.7619308233261108:  57%|█████▋    | 3676/6470 [42:16<31:00,  1.50it/s]
2|3677|Loss: 0.8642400503158569:  57%|█████▋    | 3676/6470 [42:17<31:00,  1.50it/s]
2|3677|Loss: 0.8642400503158569:  57%|█████▋    | 3677/6470 [42:17<31:19,  1.49it/s]
2|3678|Loss: 0.8963287472724915:  57%|█████▋    | 3677/6470 [42:17<31:19,  1.49it/s]
2|3678|Loss: 0.8963287472724915:  57%|█████▋    | 3678/6470 [42:18<31:11,  1.49it/s]
2|3679|Loss: 0.7123178839683533:  57%|█████▋    | 3678/6470 [42:18<31:11,  1.49it/s]
2|3679|Loss: 0.7123178839683533:  57%|█████▋    | 3679/6470 [42:18<28:43,  1.62it/s]
2|3680|Loss: 0.8532638549804688:  57%|█████▋    | 3679/6470 [42:18<28:43,  1.62it/s]
2|3680|Loss: 0.8532638549804688:  57%|█████▋    | 3680/6470 [42:19<27:54,  1.67it/s]
2|3681|Loss: 0.7004221677780151:  57%|█████▋    | 3680/6470 [42:19<27:54,  1.67it/s]
2|3681|Loss: 0.7004221677780151:  57%|█████▋    | 3681/6470 [42:19<28:15,  1.65it/s]
2|3682|Loss: 0.9315667152404785:  57%|█████▋    | 3681/6470 [42:20<28:15,  1.65it/s]
2|3682|Loss: 0.9315667152404785:  57%|█████▋    | 3682/6470 [42:20<29:53,  1.55it/s]
2|3683|Loss: 0.8365132808685303:  57%|█████▋    | 3682/6470 [42:20<29:53,  1.55it/s]
2|3683|Loss: 0.8365132808685303:  57%|█████▋    | 3683/6470 [42:21<28:27,  1.63it/s]
2|3684|Loss: 0.9711698889732361:  57%|█████▋    | 3683/6470 [42:21<28:27,  1.63it/s]
2|3684|Loss: 0.9711698889732361:  57%|█████▋    | 3684/6470 [42:21<29:03,  1.60it/s]
2|3685|Loss: 0.8205373883247375:  57%|█████▋    | 3684/6470 [42:22<29:03,  1.60it/s]
2|3685|Loss: 0.8205373883247375:  57%|█████▋    | 3685/6470 [42:22<28:56,  1.60it/s]
2|3686|Loss: 0.8955515027046204:  57%|█████▋    | 3685/6470 [42:22<28:56,  1.60it/s]
2|3686|Loss: 0.8955515027046204:  57%|█████▋    | 3686/6470 [42:22<27:15,  1.70it/s]
2|3687|Loss: 0.899451732635498:  57%|█████▋    | 3686/6470 [42:23<27:15,  1.70it/s] 
2|3687|Loss: 0.899451732635498:  57%|█████▋    | 3687/6470 [42:23<28:38,  1.62it/s]
2|3688|Loss: 0.7749636769294739:  57%|█████▋    | 3687/6470 [42:23<28:38,  1.62it/s]
2|3688|Loss: 0.7749636769294739:  57%|█████▋    | 3688/6470 [42:24<27:52,  1.66it/s]
2|3689|Loss: 0.8361968994140625:  57%|█████▋    | 3688/6470 [42:24<27:52,  1.66it/s]
2|3689|Loss: 0.8361968994140625:  57%|█████▋    | 3689/6470 [42:24<28:01,  1.65it/s]
2|3690|Loss: 0.733693540096283:  57%|█████▋    | 3689/6470 [42:25<28:01,  1.65it/s] 
2|3690|Loss: 0.733693540096283:  57%|█████▋    | 3690/6470 [42:25<28:15,  1.64it/s]
2|3691|Loss: 0.8445821404457092:  57%|█████▋    | 3690/6470 [42:25<28:15,  1.64it/s]
2|3691|Loss: 0.8445821404457092:  57%|█████▋    | 3691/6470 [42:26<29:19,  1.58it/s]
2|3692|Loss: 0.8467269539833069:  57%|█████▋    | 3691/6470 [42:26<29:19,  1.58it/s]
2|3692|Loss: 0.8467269539833069:  57%|█████▋    | 3692/6470 [42:26<28:38,  1.62it/s]
2|3693|Loss: 0.8656756281852722:  57%|█████▋    | 3692/6470 [42:26<28:38,  1.62it/s]
2|3693|Loss: 0.8656756281852722:  57%|█████▋    | 3693/6470 [42:27<28:03,  1.65it/s]
2|3694|Loss: 1.0200504064559937:  57%|█████▋    | 3693/6470 [42:27<28:03,  1.65it/s]
2|3694|Loss: 1.0200504064559937:  57%|█████▋    | 3694/6470 [42:27<30:22,  1.52it/s]
2|3695|Loss: 0.8574168682098389:  57%|█████▋    | 3694/6470 [42:28<30:22,  1.52it/s]
2|3695|Loss: 0.8574168682098389:  57%|█████▋    | 3695/6470 [42:28<29:38,  1.56it/s]
2|3696|Loss: 0.7822870016098022:  57%|█████▋    | 3695/6470 [42:28<29:38,  1.56it/s]
2|3696|Loss: 0.7822870016098022:  57%|█████▋    | 3696/6470 [42:29<31:45,  1.46it/s]
2|3697|Loss: 0.8379942178726196:  57%|█████▋    | 3696/6470 [42:29<31:45,  1.46it/s]
2|3697|Loss: 0.8379942178726196:  57%|█████▋    | 3697/6470 [42:30<34:05,  1.36it/s]
2|3698|Loss: 0.8929644227027893:  57%|█████▋    | 3697/6470 [42:30<34:05,  1.36it/s]
2|3698|Loss: 0.8929644227027893:  57%|█████▋    | 3698/6470 [42:30<34:08,  1.35it/s]
2|3699|Loss: 0.7550055980682373:  57%|█████▋    | 3698/6470 [42:31<34:08,  1.35it/s]
2|3699|Loss: 0.7550055980682373:  57%|█████▋    | 3699/6470 [42:31<35:42,  1.29it/s]
2|3700|Loss: 0.909197211265564:  57%|█████▋    | 3699/6470 [42:32<35:42,  1.29it/s] 
2|3700|Loss: 0.909197211265564:  57%|█████▋    | 3700/6470 [42:32<35:35,  1.30it/s]
2|3701|Loss: 0.6612756848335266:  57%|█████▋    | 3700/6470 [42:33<35:35,  1.30it/s]
2|3701|Loss: 0.6612756848335266:  57%|█████▋    | 3701/6470 [42:33<37:50,  1.22it/s]
2|3702|Loss: 0.7637220025062561:  57%|█████▋    | 3701/6470 [42:33<37:50,  1.22it/s]
2|3702|Loss: 0.7637220025062561:  57%|█████▋    | 3702/6470 [42:34<37:10,  1.24it/s]
2|3703|Loss: 0.8077065348625183:  57%|█████▋    | 3702/6470 [42:34<37:10,  1.24it/s]
2|3703|Loss: 0.8077065348625183:  57%|█████▋    | 3703/6470 [42:35<36:42,  1.26it/s]
2|3704|Loss: 0.8055795431137085:  57%|█████▋    | 3703/6470 [42:35<36:42,  1.26it/s]
2|3704|Loss: 0.8055795431137085:  57%|█████▋    | 3704/6470 [42:35<37:19,  1.24it/s]
2|3705|Loss: 0.8487357497215271:  57%|█████▋    | 3704/6470 [42:36<37:19,  1.24it/s]
2|3705|Loss: 0.8487357497215271:  57%|█████▋    | 3705/6470 [42:36<36:36,  1.26it/s]
2|3706|Loss: 0.8230075836181641:  57%|█████▋    | 3705/6470 [42:37<36:36,  1.26it/s]
2|3706|Loss: 0.8230075836181641:  57%|█████▋    | 3706/6470 [42:37<34:25,  1.34it/s]
2|3707|Loss: 0.8292404413223267:  57%|█████▋    | 3706/6470 [42:37<34:25,  1.34it/s]
2|3707|Loss: 0.8292404413223267:  57%|█████▋    | 3707/6470 [42:37<33:58,  1.36it/s]
2|3708|Loss: 0.7495836615562439:  57%|█████▋    | 3707/6470 [42:38<33:58,  1.36it/s]
2|3708|Loss: 0.7495836615562439:  57%|█████▋    | 3708/6470 [42:38<33:00,  1.39it/s]
2|3709|Loss: 0.8802912831306458:  57%|█████▋    | 3708/6470 [42:39<33:00,  1.39it/s]
2|3709|Loss: 0.8802912831306458:  57%|█████▋    | 3709/6470 [42:39<32:45,  1.40it/s]
2|3710|Loss: 0.9005827903747559:  57%|█████▋    | 3709/6470 [42:39<32:45,  1.40it/s]
2|3710|Loss: 0.9005827903747559:  57%|█████▋    | 3710/6470 [42:40<32:40,  1.41it/s]
2|3711|Loss: 0.9460770487785339:  57%|█████▋    | 3710/6470 [42:40<32:40,  1.41it/s]
2|3711|Loss: 0.9460770487785339:  57%|█████▋    | 3711/6470 [42:40<33:36,  1.37it/s]
2|3712|Loss: 0.9669420123100281:  57%|█████▋    | 3711/6470 [42:41<33:36,  1.37it/s]
2|3712|Loss: 0.9669420123100281:  57%|█████▋    | 3712/6470 [42:41<34:04,  1.35it/s]
2|3713|Loss: 0.7006463408470154:  57%|█████▋    | 3712/6470 [42:41<34:04,  1.35it/s]
2|3713|Loss: 0.7006463408470154:  57%|█████▋    | 3713/6470 [42:42<30:02,  1.53it/s]
2|3714|Loss: 0.7200905084609985:  57%|█████▋    | 3713/6470 [42:42<30:02,  1.53it/s]
2|3714|Loss: 0.7200905084609985:  57%|█████▋    | 3714/6470 [42:42<27:52,  1.65it/s]
2|3715|Loss: 0.8175968527793884:  57%|█████▋    | 3714/6470 [42:42<27:52,  1.65it/s]
2|3715|Loss: 0.8175968527793884:  57%|█████▋    | 3715/6470 [42:43<27:58,  1.64it/s]
2|3716|Loss: 0.8172569274902344:  57%|█████▋    | 3715/6470 [42:43<27:58,  1.64it/s]
2|3716|Loss: 0.8172569274902344:  57%|█████▋    | 3716/6470 [42:43<29:08,  1.58it/s]
2|3717|Loss: 1.004995584487915:  57%|█████▋    | 3716/6470 [42:44<29:08,  1.58it/s] 
2|3717|Loss: 1.004995584487915:  57%|█████▋    | 3717/6470 [42:44<31:06,  1.47it/s]
2|3718|Loss: 0.856671929359436:  57%|█████▋    | 3717/6470 [42:45<31:06,  1.47it/s]
2|3718|Loss: 0.856671929359436:  57%|█████▋    | 3718/6470 [42:45<30:48,  1.49it/s]
2|3719|Loss: 0.8346785306930542:  57%|█████▋    | 3718/6470 [42:45<30:48,  1.49it/s]
2|3719|Loss: 0.8346785306930542:  57%|█████▋    | 3719/6470 [42:45<30:14,  1.52it/s]
2|3720|Loss: 0.8433129787445068:  57%|█████▋    | 3719/6470 [42:46<30:14,  1.52it/s]
2|3720|Loss: 0.8433129787445068:  57%|█████▋    | 3720/6470 [42:46<31:14,  1.47it/s]
2|3721|Loss: 0.9115431308746338:  57%|█████▋    | 3720/6470 [42:47<31:14,  1.47it/s]
2|3721|Loss: 0.9115431308746338:  58%|█████▊    | 3721/6470 [42:47<31:44,  1.44it/s]
2|3722|Loss: 0.8215359449386597:  58%|█████▊    | 3721/6470 [42:47<31:44,  1.44it/s]
2|3722|Loss: 0.8215359449386597:  58%|█████▊    | 3722/6470 [42:48<31:53,  1.44it/s]
2|3723|Loss: 0.8187763690948486:  58%|█████▊    | 3722/6470 [42:48<31:53,  1.44it/s]
2|3723|Loss: 0.8187763690948486:  58%|█████▊    | 3723/6470 [42:48<33:22,  1.37it/s]
2|3724|Loss: 0.9344757199287415:  58%|█████▊    | 3723/6470 [42:49<33:22,  1.37it/s]
2|3724|Loss: 0.9344757199287415:  58%|█████▊    | 3724/6470 [42:49<30:36,  1.50it/s]
2|3725|Loss: 0.6561011075973511:  58%|█████▊    | 3724/6470 [42:49<30:36,  1.50it/s]
2|3725|Loss: 0.6561011075973511:  58%|█████▊    | 3725/6470 [42:50<30:21,  1.51it/s]
2|3726|Loss: 0.8957447409629822:  58%|█████▊    | 3725/6470 [42:50<30:21,  1.51it/s]
2|3726|Loss: 0.8957447409629822:  58%|█████▊    | 3726/6470 [42:50<31:15,  1.46it/s]
2|3727|Loss: 0.7776533961296082:  58%|█████▊    | 3726/6470 [42:51<31:15,  1.46it/s]
2|3727|Loss: 0.7776533961296082:  58%|█████▊    | 3727/6470 [42:51<32:04,  1.43it/s]
2|3728|Loss: 0.7607612609863281:  58%|█████▊    | 3727/6470 [42:51<32:04,  1.43it/s]
2|3728|Loss: 0.7607612609863281:  58%|█████▊    | 3728/6470 [42:52<31:20,  1.46it/s]
2|3729|Loss: 0.6847952604293823:  58%|█████▊    | 3728/6470 [42:52<31:20,  1.46it/s]
2|3729|Loss: 0.6847952604293823:  58%|█████▊    | 3729/6470 [42:52<30:21,  1.51it/s]
2|3730|Loss: 0.895061731338501:  58%|█████▊    | 3729/6470 [42:53<30:21,  1.51it/s] 
2|3730|Loss: 0.895061731338501:  58%|█████▊    | 3730/6470 [42:53<30:37,  1.49it/s]
2|3731|Loss: 0.9248350262641907:  58%|█████▊    | 3730/6470 [42:53<30:37,  1.49it/s]
2|3731|Loss: 0.9248350262641907:  58%|█████▊    | 3731/6470 [42:54<31:27,  1.45it/s]
2|3732|Loss: 0.8268547058105469:  58%|█████▊    | 3731/6470 [42:54<31:27,  1.45it/s]
2|3732|Loss: 0.8268547058105469:  58%|█████▊    | 3732/6470 [42:54<30:53,  1.48it/s]
2|3733|Loss: 0.8698821663856506:  58%|█████▊    | 3732/6470 [42:55<30:53,  1.48it/s]
2|3733|Loss: 0.8698821663856506:  58%|█████▊    | 3733/6470 [42:55<30:19,  1.50it/s]
2|3734|Loss: 0.8174446225166321:  58%|█████▊    | 3733/6470 [42:55<30:19,  1.50it/s]
2|3734|Loss: 0.8174446225166321:  58%|█████▊    | 3734/6470 [42:56<31:00,  1.47it/s]
2|3735|Loss: 0.8988105058670044:  58%|█████▊    | 3734/6470 [42:56<31:00,  1.47it/s]
2|3735|Loss: 0.8988105058670044:  58%|█████▊    | 3735/6470 [42:56<29:17,  1.56it/s]
2|3736|Loss: 0.9006720185279846:  58%|█████▊    | 3735/6470 [42:57<29:17,  1.56it/s]
2|3736|Loss: 0.9006720185279846:  58%|█████▊    | 3736/6470 [42:57<30:01,  1.52it/s]
2|3737|Loss: 0.9201399087905884:  58%|█████▊    | 3736/6470 [42:57<30:01,  1.52it/s]
2|3737|Loss: 0.9201399087905884:  58%|█████▊    | 3737/6470 [42:58<31:15,  1.46it/s]
2|3738|Loss: 0.8780565857887268:  58%|█████▊    | 3737/6470 [42:58<31:15,  1.46it/s]
2|3738|Loss: 0.8780565857887268:  58%|█████▊    | 3738/6470 [42:58<30:43,  1.48it/s]
2|3739|Loss: 0.927013099193573:  58%|█████▊    | 3738/6470 [42:59<30:43,  1.48it/s] 
2|3739|Loss: 0.927013099193573:  58%|█████▊    | 3739/6470 [42:59<29:28,  1.54it/s]
2|3740|Loss: 0.8078222274780273:  58%|█████▊    | 3739/6470 [42:59<29:28,  1.54it/s]
2|3740|Loss: 0.8078222274780273:  58%|█████▊    | 3740/6470 [43:00<29:11,  1.56it/s]
2|3741|Loss: 0.8756146430969238:  58%|█████▊    | 3740/6470 [43:00<29:11,  1.56it/s]
2|3741|Loss: 0.8756146430969238:  58%|█████▊    | 3741/6470 [43:00<28:25,  1.60it/s]
2|3742|Loss: 0.9253172278404236:  58%|█████▊    | 3741/6470 [43:01<28:25,  1.60it/s]
2|3742|Loss: 0.9253172278404236:  58%|█████▊    | 3742/6470 [43:01<29:25,  1.55it/s]
2|3743|Loss: 0.7945092916488647:  58%|█████▊    | 3742/6470 [43:01<29:25,  1.55it/s]
2|3743|Loss: 0.7945092916488647:  58%|█████▊    | 3743/6470 [43:02<31:54,  1.42it/s]
2|3744|Loss: 0.874474823474884:  58%|█████▊    | 3743/6470 [43:02<31:54,  1.42it/s] 
2|3744|Loss: 0.874474823474884:  58%|█████▊    | 3744/6470 [43:03<33:55,  1.34it/s]
2|3745|Loss: 0.9917913675308228:  58%|█████▊    | 3744/6470 [43:03<33:55,  1.34it/s]
2|3745|Loss: 0.9917913675308228:  58%|█████▊    | 3745/6470 [43:03<34:44,  1.31it/s]
2|3746|Loss: 0.7646281719207764:  58%|█████▊    | 3745/6470 [43:04<34:44,  1.31it/s]
2|3746|Loss: 0.7646281719207764:  58%|█████▊    | 3746/6470 [43:04<35:29,  1.28it/s]
2|3747|Loss: 0.865227997303009:  58%|█████▊    | 3746/6470 [43:05<35:29,  1.28it/s] 
2|3747|Loss: 0.865227997303009:  58%|█████▊    | 3747/6470 [43:05<33:51,  1.34it/s]
2|3748|Loss: 0.7076889276504517:  58%|█████▊    | 3747/6470 [43:05<33:51,  1.34it/s]
2|3748|Loss: 0.7076889276504517:  58%|█████▊    | 3748/6470 [43:05<32:03,  1.42it/s]
2|3749|Loss: 0.6527084112167358:  58%|█████▊    | 3748/6470 [43:06<32:03,  1.42it/s]
2|3749|Loss: 0.6527084112167358:  58%|█████▊    | 3749/6470 [43:06<30:52,  1.47it/s]
2|3750|Loss: 0.7498607039451599:  58%|█████▊    | 3749/6470 [43:06<30:52,  1.47it/s]
2|3750|Loss: 0.7498607039451599:  58%|█████▊    | 3750/6470 [43:07<30:15,  1.50it/s]
2|3751|Loss: 0.9157795310020447:  58%|█████▊    | 3750/6470 [43:07<30:15,  1.50it/s]
2|3751|Loss: 0.9157795310020447:  58%|█████▊    | 3751/6470 [43:07<29:12,  1.55it/s]
2|3752|Loss: 0.8728344440460205:  58%|█████▊    | 3751/6470 [43:08<29:12,  1.55it/s]
2|3752|Loss: 0.8728344440460205:  58%|█████▊    | 3752/6470 [43:08<28:42,  1.58it/s]
2|3753|Loss: 0.7960575819015503:  58%|█████▊    | 3752/6470 [43:08<28:42,  1.58it/s]
2|3753|Loss: 0.7960575819015503:  58%|█████▊    | 3753/6470 [43:09<30:18,  1.49it/s]
2|3754|Loss: 0.792992889881134:  58%|█████▊    | 3753/6470 [43:09<30:18,  1.49it/s] 
2|3754|Loss: 0.792992889881134:  58%|█████▊    | 3754/6470 [43:10<33:18,  1.36it/s]
2|3755|Loss: 0.9187214374542236:  58%|█████▊    | 3754/6470 [43:10<33:18,  1.36it/s]
2|3755|Loss: 0.9187214374542236:  58%|█████▊    | 3755/6470 [43:10<35:18,  1.28it/s]
2|3756|Loss: 0.7683263421058655:  58%|█████▊    | 3755/6470 [43:11<35:18,  1.28it/s]
2|3756|Loss: 0.7683263421058655:  58%|█████▊    | 3756/6470 [43:11<35:42,  1.27it/s]
2|3757|Loss: 1.017655611038208:  58%|█████▊    | 3756/6470 [43:12<35:42,  1.27it/s] 
2|3757|Loss: 1.017655611038208:  58%|█████▊    | 3757/6470 [43:12<35:06,  1.29it/s]
2|3758|Loss: 0.8390781879425049:  58%|█████▊    | 3757/6470 [43:12<35:06,  1.29it/s]
2|3758|Loss: 0.8390781879425049:  58%|█████▊    | 3758/6470 [43:13<32:03,  1.41it/s]
2|3759|Loss: 0.868363618850708:  58%|█████▊    | 3758/6470 [43:13<32:03,  1.41it/s] 
2|3759|Loss: 0.868363618850708:  58%|█████▊    | 3759/6470 [43:13<31:56,  1.41it/s]
2|3760|Loss: 1.3642860651016235:  58%|█████▊    | 3759/6470 [43:14<31:56,  1.41it/s]
2|3760|Loss: 1.3642860651016235:  58%|█████▊    | 3760/6470 [43:14<38:37,  1.17it/s]
2|3761|Loss: 0.8035223484039307:  58%|█████▊    | 3760/6470 [43:15<38:37,  1.17it/s]
2|3761|Loss: 0.8035223484039307:  58%|█████▊    | 3761/6470 [43:15<37:36,  1.20it/s]
2|3762|Loss: 0.7152743339538574:  58%|█████▊    | 3761/6470 [43:16<37:36,  1.20it/s]
2|3762|Loss: 0.7152743339538574:  58%|█████▊    | 3762/6470 [43:16<31:28,  1.43it/s]
2|3763|Loss: 0.8609713912010193:  58%|█████▊    | 3762/6470 [43:16<31:28,  1.43it/s]
2|3763|Loss: 0.8609713912010193:  58%|█████▊    | 3763/6470 [43:16<31:57,  1.41it/s]
2|3764|Loss: 0.6880863308906555:  58%|█████▊    | 3763/6470 [43:17<31:57,  1.41it/s]
2|3764|Loss: 0.6880863308906555:  58%|█████▊    | 3764/6470 [43:17<31:43,  1.42it/s]
2|3765|Loss: 0.9661254286766052:  58%|█████▊    | 3764/6470 [43:17<31:43,  1.42it/s]
2|3765|Loss: 0.9661254286766052:  58%|█████▊    | 3765/6470 [43:18<29:06,  1.55it/s]
2|3766|Loss: 0.8709716796875:  58%|█████▊    | 3765/6470 [43:18<29:06,  1.55it/s]   
2|3766|Loss: 0.8709716796875:  58%|█████▊    | 3766/6470 [43:18<29:13,  1.54it/s]
2|3767|Loss: 0.946847677230835:  58%|█████▊    | 3766/6470 [43:19<29:13,  1.54it/s]
2|3767|Loss: 0.946847677230835:  58%|█████▊    | 3767/6470 [43:19<29:06,  1.55it/s]
2|3768|Loss: 0.9762049317359924:  58%|█████▊    | 3767/6470 [43:19<29:06,  1.55it/s]
2|3768|Loss: 0.9762049317359924:  58%|█████▊    | 3768/6470 [43:19<28:13,  1.60it/s]
2|3769|Loss: 0.8916245698928833:  58%|█████▊    | 3768/6470 [43:20<28:13,  1.60it/s]
2|3769|Loss: 0.8916245698928833:  58%|█████▊    | 3769/6470 [43:20<29:21,  1.53it/s]
2|3770|Loss: 0.8864929676055908:  58%|█████▊    | 3769/6470 [43:21<29:21,  1.53it/s]
2|3770|Loss: 0.8864929676055908:  58%|█████▊    | 3770/6470 [43:21<30:11,  1.49it/s]
2|3771|Loss: 0.9449373483657837:  58%|█████▊    | 3770/6470 [43:21<30:11,  1.49it/s]
2|3771|Loss: 0.9449373483657837:  58%|█████▊    | 3771/6470 [43:22<32:21,  1.39it/s]
2|3772|Loss: 0.9340897798538208:  58%|█████▊    | 3771/6470 [43:22<32:21,  1.39it/s]
2|3772|Loss: 0.9340897798538208:  58%|█████▊    | 3772/6470 [43:23<34:17,  1.31it/s]
2|3773|Loss: 0.8142388463020325:  58%|█████▊    | 3772/6470 [43:23<34:17,  1.31it/s]
2|3773|Loss: 0.8142388463020325:  58%|█████▊    | 3773/6470 [43:23<30:25,  1.48it/s]
2|3774|Loss: 0.8109219074249268:  58%|█████▊    | 3773/6470 [43:23<30:25,  1.48it/s]
2|3774|Loss: 0.8109219074249268:  58%|█████▊    | 3774/6470 [43:24<30:55,  1.45it/s]
2|3775|Loss: 0.8384901881217957:  58%|█████▊    | 3774/6470 [43:24<30:55,  1.45it/s]
2|3775|Loss: 0.8384901881217957:  58%|█████▊    | 3775/6470 [43:24<30:07,  1.49it/s]
2|3776|Loss: 0.9322074055671692:  58%|█████▊    | 3775/6470 [43:25<30:07,  1.49it/s]
2|3776|Loss: 0.9322074055671692:  58%|█████▊    | 3776/6470 [43:25<30:39,  1.46it/s]
2|3777|Loss: 0.9145082235336304:  58%|█████▊    | 3776/6470 [43:26<30:39,  1.46it/s]
2|3777|Loss: 0.9145082235336304:  58%|█████▊    | 3777/6470 [43:26<33:18,  1.35it/s]
2|3778|Loss: 0.8474182486534119:  58%|█████▊    | 3777/6470 [43:26<33:18,  1.35it/s]
2|3778|Loss: 0.8474182486534119:  58%|█████▊    | 3778/6470 [43:27<32:26,  1.38it/s]
2|3779|Loss: 1.0229413509368896:  58%|█████▊    | 3778/6470 [43:27<32:26,  1.38it/s]
2|3779|Loss: 1.0229413509368896:  58%|█████▊    | 3779/6470 [43:27<32:25,  1.38it/s]
2|3780|Loss: 0.8567771315574646:  58%|█████▊    | 3779/6470 [43:28<32:25,  1.38it/s]
2|3780|Loss: 0.8567771315574646:  58%|█████▊    | 3780/6470 [43:28<32:50,  1.36it/s]
2|3781|Loss: 0.8053745031356812:  58%|█████▊    | 3780/6470 [43:29<32:50,  1.36it/s]
2|3781|Loss: 0.8053745031356812:  58%|█████▊    | 3781/6470 [43:29<34:59,  1.28it/s]
2|3782|Loss: 0.7396900653839111:  58%|█████▊    | 3781/6470 [43:30<34:59,  1.28it/s]
2|3782|Loss: 0.7396900653839111:  58%|█████▊    | 3782/6470 [43:30<36:25,  1.23it/s]
2|3783|Loss: 0.9908198714256287:  58%|█████▊    | 3782/6470 [43:30<36:25,  1.23it/s]
2|3783|Loss: 0.9908198714256287:  58%|█████▊    | 3783/6470 [43:30<33:07,  1.35it/s]
2|3784|Loss: 0.8319697380065918:  58%|█████▊    | 3783/6470 [43:31<33:07,  1.35it/s]
2|3784|Loss: 0.8319697380065918:  58%|█████▊    | 3784/6470 [43:31<31:18,  1.43it/s]
2|3785|Loss: 0.8051548004150391:  58%|█████▊    | 3784/6470 [43:31<31:18,  1.43it/s]
2|3785|Loss: 0.8051548004150391:  59%|█████▊    | 3785/6470 [43:32<31:05,  1.44it/s]
2|3786|Loss: 0.8261577486991882:  59%|█████▊    | 3785/6470 [43:32<31:05,  1.44it/s]
2|3786|Loss: 0.8261577486991882:  59%|█████▊    | 3786/6470 [43:32<30:30,  1.47it/s]
2|3787|Loss: 0.8524609208106995:  59%|█████▊    | 3786/6470 [43:33<30:30,  1.47it/s]
2|3787|Loss: 0.8524609208106995:  59%|█████▊    | 3787/6470 [43:33<29:37,  1.51it/s]
2|3788|Loss: 0.9806025624275208:  59%|█████▊    | 3787/6470 [43:33<29:37,  1.51it/s]
2|3788|Loss: 0.9806025624275208:  59%|█████▊    | 3788/6470 [43:34<29:15,  1.53it/s]
2|3789|Loss: 0.7945055961608887:  59%|█████▊    | 3788/6470 [43:34<29:15,  1.53it/s]
2|3789|Loss: 0.7945055961608887:  59%|█████▊    | 3789/6470 [43:34<27:03,  1.65it/s]
2|3790|Loss: 0.8079516887664795:  59%|█████▊    | 3789/6470 [43:34<27:03,  1.65it/s]
2|3790|Loss: 0.8079516887664795:  59%|█████▊    | 3790/6470 [43:35<25:16,  1.77it/s]
2|3791|Loss: 0.8925855159759521:  59%|█████▊    | 3790/6470 [43:35<25:16,  1.77it/s]
2|3791|Loss: 0.8925855159759521:  59%|█████▊    | 3791/6470 [43:35<26:47,  1.67it/s]
2|3792|Loss: 0.8965024948120117:  59%|█████▊    | 3791/6470 [43:36<26:47,  1.67it/s]
2|3792|Loss: 0.8965024948120117:  59%|█████▊    | 3792/6470 [43:36<29:38,  1.51it/s]
2|3793|Loss: 0.8778415322303772:  59%|█████▊    | 3792/6470 [43:37<29:38,  1.51it/s]
2|3793|Loss: 0.8778415322303772:  59%|█████▊    | 3793/6470 [43:37<29:08,  1.53it/s]
2|3794|Loss: 0.6746049523353577:  59%|█████▊    | 3793/6470 [43:37<29:08,  1.53it/s]
2|3794|Loss: 0.6746049523353577:  59%|█████▊    | 3794/6470 [43:37<29:45,  1.50it/s]
2|3795|Loss: 0.8983210921287537:  59%|█████▊    | 3794/6470 [43:38<29:45,  1.50it/s]
2|3795|Loss: 0.8983210921287537:  59%|█████▊    | 3795/6470 [43:38<30:19,  1.47it/s]
2|3796|Loss: 1.0149577856063843:  59%|█████▊    | 3795/6470 [43:39<30:19,  1.47it/s]
2|3796|Loss: 1.0149577856063843:  59%|█████▊    | 3796/6470 [43:39<28:31,  1.56it/s]
2|3797|Loss: 0.915418803691864:  59%|█████▊    | 3796/6470 [43:39<28:31,  1.56it/s] 
2|3797|Loss: 0.915418803691864:  59%|█████▊    | 3797/6470 [43:39<29:05,  1.53it/s]
2|3798|Loss: 1.1924543380737305:  59%|█████▊    | 3797/6470 [43:40<29:05,  1.53it/s]
2|3798|Loss: 1.1924543380737305:  59%|█████▊    | 3798/6470 [43:40<31:23,  1.42it/s]
2|3799|Loss: 1.0144463777542114:  59%|█████▊    | 3798/6470 [43:41<31:23,  1.42it/s]
2|3799|Loss: 1.0144463777542114:  59%|█████▊    | 3799/6470 [43:41<33:29,  1.33it/s]
2|3800|Loss: 0.9257960915565491:  59%|█████▊    | 3799/6470 [43:42<33:29,  1.33it/s]
2|3800|Loss: 0.9257960915565491:  59%|█████▊    | 3800/6470 [43:42<33:15,  1.34it/s]
2|3801|Loss: 0.8478695750236511:  59%|█████▊    | 3800/6470 [43:42<33:15,  1.34it/s]
2|3801|Loss: 0.8478695750236511:  59%|█████▊    | 3801/6470 [43:42<31:44,  1.40it/s]
2|3802|Loss: 0.5610276460647583:  59%|█████▊    | 3801/6470 [43:43<31:44,  1.40it/s]
2|3802|Loss: 0.5610276460647583:  59%|█████▉    | 3802/6470 [43:43<30:40,  1.45it/s]
2|3803|Loss: 0.9199168086051941:  59%|█████▉    | 3802/6470 [43:44<30:40,  1.45it/s]
2|3803|Loss: 0.9199168086051941:  59%|█████▉    | 3803/6470 [43:44<32:16,  1.38it/s]
2|3804|Loss: 0.6732438206672668:  59%|█████▉    | 3803/6470 [43:44<32:16,  1.38it/s]
2|3804|Loss: 0.6732438206672668:  59%|█████▉    | 3804/6470 [43:45<32:45,  1.36it/s]
2|3805|Loss: 0.865119218826294:  59%|█████▉    | 3804/6470 [43:45<32:45,  1.36it/s] 
2|3805|Loss: 0.865119218826294:  59%|█████▉    | 3805/6470 [43:45<33:45,  1.32it/s]
2|3806|Loss: 0.8293723464012146:  59%|█████▉    | 3805/6470 [43:46<33:45,  1.32it/s]
2|3806|Loss: 0.8293723464012146:  59%|█████▉    | 3806/6470 [43:46<32:29,  1.37it/s]
2|3807|Loss: 0.975864052772522:  59%|█████▉    | 3806/6470 [43:46<32:29,  1.37it/s] 
2|3807|Loss: 0.975864052772522:  59%|█████▉    | 3807/6470 [43:47<30:55,  1.43it/s]
2|3808|Loss: 0.8933029174804688:  59%|█████▉    | 3807/6470 [43:47<30:55,  1.43it/s]
2|3808|Loss: 0.8933029174804688:  59%|█████▉    | 3808/6470 [43:47<31:05,  1.43it/s]
2|3809|Loss: 0.8375394940376282:  59%|█████▉    | 3808/6470 [43:48<31:05,  1.43it/s]
2|3809|Loss: 0.8375394940376282:  59%|█████▉    | 3809/6470 [43:48<28:31,  1.55it/s]
2|3810|Loss: 0.7374113202095032:  59%|█████▉    | 3809/6470 [43:48<28:31,  1.55it/s]
2|3810|Loss: 0.7374113202095032:  59%|█████▉    | 3810/6470 [43:49<27:49,  1.59it/s]
2|3811|Loss: 0.8027363419532776:  59%|█████▉    | 3810/6470 [43:49<27:49,  1.59it/s]
2|3811|Loss: 0.8027363419532776:  59%|█████▉    | 3811/6470 [43:49<29:30,  1.50it/s]
2|3812|Loss: 0.7271673083305359:  59%|█████▉    | 3811/6470 [43:50<29:30,  1.50it/s]
2|3812|Loss: 0.7271673083305359:  59%|█████▉    | 3812/6470 [43:50<29:34,  1.50it/s]
2|3813|Loss: 0.6877098083496094:  59%|█████▉    | 3812/6470 [43:50<29:34,  1.50it/s]
2|3813|Loss: 0.6877098083496094:  59%|█████▉    | 3813/6470 [43:51<30:11,  1.47it/s]
2|3814|Loss: 0.72834712266922:  59%|█████▉    | 3813/6470 [43:51<30:11,  1.47it/s]  
2|3814|Loss: 0.72834712266922:  59%|█████▉    | 3814/6470 [43:51<30:37,  1.45it/s]
2|3815|Loss: 0.8256775140762329:  59%|█████▉    | 3814/6470 [43:52<30:37,  1.45it/s]
2|3815|Loss: 0.8256775140762329:  59%|█████▉    | 3815/6470 [43:52<27:23,  1.62it/s]
2|3816|Loss: 0.9038591384887695:  59%|█████▉    | 3815/6470 [43:52<27:23,  1.62it/s]
2|3816|Loss: 0.9038591384887695:  59%|█████▉    | 3816/6470 [43:53<29:07,  1.52it/s]
2|3817|Loss: 0.8958823084831238:  59%|█████▉    | 3816/6470 [43:53<29:07,  1.52it/s]
2|3817|Loss: 0.8958823084831238:  59%|█████▉    | 3817/6470 [43:53<28:46,  1.54it/s]
2|3818|Loss: 0.8377134203910828:  59%|█████▉    | 3817/6470 [43:54<28:46,  1.54it/s]
2|3818|Loss: 0.8377134203910828:  59%|█████▉    | 3818/6470 [43:54<29:30,  1.50it/s]
2|3819|Loss: 0.9357705116271973:  59%|█████▉    | 3818/6470 [43:54<29:30,  1.50it/s]
2|3819|Loss: 0.9357705116271973:  59%|█████▉    | 3819/6470 [43:55<29:25,  1.50it/s]
2|3820|Loss: 0.8095235824584961:  59%|█████▉    | 3819/6470 [43:55<29:25,  1.50it/s]
2|3820|Loss: 0.8095235824584961:  59%|█████▉    | 3820/6470 [43:55<28:28,  1.55it/s]
2|3821|Loss: 0.7667250037193298:  59%|█████▉    | 3820/6470 [43:56<28:28,  1.55it/s]
2|3821|Loss: 0.7667250037193298:  59%|█████▉    | 3821/6470 [43:56<29:18,  1.51it/s]
2|3822|Loss: 0.9117558598518372:  59%|█████▉    | 3821/6470 [43:56<29:18,  1.51it/s]
2|3822|Loss: 0.9117558598518372:  59%|█████▉    | 3822/6470 [43:57<28:02,  1.57it/s]
2|3823|Loss: 0.7190157175064087:  59%|█████▉    | 3822/6470 [43:57<28:02,  1.57it/s]
2|3823|Loss: 0.7190157175064087:  59%|█████▉    | 3823/6470 [43:57<27:08,  1.63it/s]
2|3824|Loss: 0.7246658802032471:  59%|█████▉    | 3823/6470 [43:57<27:08,  1.63it/s]
2|3824|Loss: 0.7246658802032471:  59%|█████▉    | 3824/6470 [43:58<27:59,  1.58it/s]
2|3825|Loss: 0.8270284533500671:  59%|█████▉    | 3824/6470 [43:58<27:59,  1.58it/s]
2|3825|Loss: 0.8270284533500671:  59%|█████▉    | 3825/6470 [43:59<30:38,  1.44it/s]
2|3826|Loss: 0.7352086901664734:  59%|█████▉    | 3825/6470 [43:59<30:38,  1.44it/s]
2|3826|Loss: 0.7352086901664734:  59%|█████▉    | 3826/6470 [43:59<28:38,  1.54it/s]
2|3827|Loss: 0.8644060492515564:  59%|█████▉    | 3826/6470 [43:59<28:38,  1.54it/s]
2|3827|Loss: 0.8644060492515564:  59%|█████▉    | 3827/6470 [44:00<27:38,  1.59it/s]
2|3828|Loss: 0.8742997646331787:  59%|█████▉    | 3827/6470 [44:00<27:38,  1.59it/s]
2|3828|Loss: 0.8742997646331787:  59%|█████▉    | 3828/6470 [44:00<29:16,  1.50it/s]
2|3829|Loss: 0.8402388691902161:  59%|█████▉    | 3828/6470 [44:01<29:16,  1.50it/s]
2|3829|Loss: 0.8402388691902161:  59%|█████▉    | 3829/6470 [44:01<29:19,  1.50it/s]
2|3830|Loss: 0.794016420841217:  59%|█████▉    | 3829/6470 [44:01<29:19,  1.50it/s] 
2|3830|Loss: 0.794016420841217:  59%|█████▉    | 3830/6470 [44:02<28:43,  1.53it/s]
2|3831|Loss: 0.8271358609199524:  59%|█████▉    | 3830/6470 [44:02<28:43,  1.53it/s]
2|3831|Loss: 0.8271358609199524:  59%|█████▉    | 3831/6470 [44:03<31:07,  1.41it/s]
2|3832|Loss: 0.9365331530570984:  59%|█████▉    | 3831/6470 [44:03<31:07,  1.41it/s]
2|3832|Loss: 0.9365331530570984:  59%|█████▉    | 3832/6470 [44:03<30:47,  1.43it/s]
2|3833|Loss: 0.8797847032546997:  59%|█████▉    | 3832/6470 [44:04<30:47,  1.43it/s]
2|3833|Loss: 0.8797847032546997:  59%|█████▉    | 3833/6470 [44:04<28:55,  1.52it/s]
2|3834|Loss: 0.8766475319862366:  59%|█████▉    | 3833/6470 [44:04<28:55,  1.52it/s]
2|3834|Loss: 0.8766475319862366:  59%|█████▉    | 3834/6470 [44:04<28:15,  1.55it/s]
2|3835|Loss: 0.8788209557533264:  59%|█████▉    | 3834/6470 [44:05<28:15,  1.55it/s]
2|3835|Loss: 0.8788209557533264:  59%|█████▉    | 3835/6470 [44:05<29:20,  1.50it/s]
2|3836|Loss: 0.7265546321868896:  59%|█████▉    | 3835/6470 [44:06<29:20,  1.50it/s]
2|3836|Loss: 0.7265546321868896:  59%|█████▉    | 3836/6470 [44:06<28:39,  1.53it/s]
2|3837|Loss: 0.9791359901428223:  59%|█████▉    | 3836/6470 [44:06<28:39,  1.53it/s]
2|3837|Loss: 0.9791359901428223:  59%|█████▉    | 3837/6470 [44:06<29:18,  1.50it/s]
2|3838|Loss: 1.0331140756607056:  59%|█████▉    | 3837/6470 [44:07<29:18,  1.50it/s]
2|3838|Loss: 1.0331140756607056:  59%|█████▉    | 3838/6470 [44:07<28:15,  1.55it/s]
2|3839|Loss: 0.948693037033081:  59%|█████▉    | 3838/6470 [44:07<28:15,  1.55it/s] 
2|3839|Loss: 0.948693037033081:  59%|█████▉    | 3839/6470 [44:08<28:43,  1.53it/s]
2|3840|Loss: 0.8627746105194092:  59%|█████▉    | 3839/6470 [44:08<28:43,  1.53it/s]
2|3840|Loss: 0.8627746105194092:  59%|█████▉    | 3840/6470 [44:09<30:43,  1.43it/s]
2|3841|Loss: 0.8065048456192017:  59%|█████▉    | 3840/6470 [44:09<30:43,  1.43it/s]
2|3841|Loss: 0.8065048456192017:  59%|█████▉    | 3841/6470 [44:09<31:27,  1.39it/s]
2|3842|Loss: 0.8004041314125061:  59%|█████▉    | 3841/6470 [44:10<31:27,  1.39it/s]
2|3842|Loss: 0.8004041314125061:  59%|█████▉    | 3842/6470 [44:10<30:41,  1.43it/s]
2|3843|Loss: 0.8591770529747009:  59%|█████▉    | 3842/6470 [44:10<30:41,  1.43it/s]
2|3843|Loss: 0.8591770529747009:  59%|█████▉    | 3843/6470 [44:11<29:37,  1.48it/s]
2|3844|Loss: 0.7774670124053955:  59%|█████▉    | 3843/6470 [44:11<29:37,  1.48it/s]
2|3844|Loss: 0.7774670124053955:  59%|█████▉    | 3844/6470 [44:11<31:03,  1.41it/s]
2|3845|Loss: 0.723868191242218:  59%|█████▉    | 3844/6470 [44:12<31:03,  1.41it/s] 
2|3845|Loss: 0.723868191242218:  59%|█████▉    | 3845/6470 [44:12<31:12,  1.40it/s]
2|3846|Loss: 0.8335159420967102:  59%|█████▉    | 3845/6470 [44:12<31:12,  1.40it/s]
2|3846|Loss: 0.8335159420967102:  59%|█████▉    | 3846/6470 [44:13<29:44,  1.47it/s]
2|3847|Loss: 0.9365463256835938:  59%|█████▉    | 3846/6470 [44:13<29:44,  1.47it/s]
2|3847|Loss: 0.9365463256835938:  59%|█████▉    | 3847/6470 [44:13<27:54,  1.57it/s]
2|3848|Loss: 0.7837068438529968:  59%|█████▉    | 3847/6470 [44:14<27:54,  1.57it/s]
2|3848|Loss: 0.7837068438529968:  59%|█████▉    | 3848/6470 [44:14<28:46,  1.52it/s]
2|3849|Loss: 0.8037669658660889:  59%|█████▉    | 3848/6470 [44:14<28:46,  1.52it/s]
2|3849|Loss: 0.8037669658660889:  59%|█████▉    | 3849/6470 [44:15<28:21,  1.54it/s]
2|3850|Loss: 0.8020511269569397:  59%|█████▉    | 3849/6470 [44:15<28:21,  1.54it/s]
2|3850|Loss: 0.8020511269569397:  60%|█████▉    | 3850/6470 [44:15<29:06,  1.50it/s]
2|3851|Loss: 0.7750252485275269:  60%|█████▉    | 3850/6470 [44:16<29:06,  1.50it/s]
2|3851|Loss: 0.7750252485275269:  60%|█████▉    | 3851/6470 [44:16<30:37,  1.43it/s]
2|3852|Loss: 0.7493964433670044:  60%|█████▉    | 3851/6470 [44:16<30:37,  1.43it/s]
2|3852|Loss: 0.7493964433670044:  60%|█████▉    | 3852/6470 [44:17<30:03,  1.45it/s]
2|3853|Loss: 0.9068328738212585:  60%|█████▉    | 3852/6470 [44:17<30:03,  1.45it/s]
2|3853|Loss: 0.9068328738212585:  60%|█████▉    | 3853/6470 [44:18<31:22,  1.39it/s]
2|3854|Loss: 0.8033887147903442:  60%|█████▉    | 3853/6470 [44:18<31:22,  1.39it/s]
2|3854|Loss: 0.8033887147903442:  60%|█████▉    | 3854/6470 [44:18<32:48,  1.33it/s]
2|3855|Loss: 0.8845698237419128:  60%|█████▉    | 3854/6470 [44:19<32:48,  1.33it/s]
2|3855|Loss: 0.8845698237419128:  60%|█████▉    | 3855/6470 [44:19<29:52,  1.46it/s]
2|3856|Loss: 0.8881844878196716:  60%|█████▉    | 3855/6470 [44:19<29:52,  1.46it/s]
2|3856|Loss: 0.8881844878196716:  60%|█████▉    | 3856/6470 [44:19<28:32,  1.53it/s]
2|3857|Loss: 0.7302664518356323:  60%|█████▉    | 3856/6470 [44:20<28:32,  1.53it/s]
2|3857|Loss: 0.7302664518356323:  60%|█████▉    | 3857/6470 [44:20<28:17,  1.54it/s]
2|3858|Loss: 0.6835691332817078:  60%|█████▉    | 3857/6470 [44:20<28:17,  1.54it/s]
2|3858|Loss: 0.6835691332817078:  60%|█████▉    | 3858/6470 [44:21<26:13,  1.66it/s]
2|3859|Loss: 0.8308407068252563:  60%|█████▉    | 3858/6470 [44:21<26:13,  1.66it/s]
2|3859|Loss: 0.8308407068252563:  60%|█████▉    | 3859/6470 [44:21<25:01,  1.74it/s]
2|3860|Loss: 1.0253649950027466:  60%|█████▉    | 3859/6470 [44:21<25:01,  1.74it/s]
2|3860|Loss: 1.0253649950027466:  60%|█████▉    | 3860/6470 [44:22<26:07,  1.66it/s]
2|3861|Loss: 0.6931585669517517:  60%|█████▉    | 3860/6470 [44:22<26:07,  1.66it/s]
2|3861|Loss: 0.6931585669517517:  60%|█████▉    | 3861/6470 [44:23<30:09,  1.44it/s]
2|3862|Loss: 0.7782780528068542:  60%|█████▉    | 3861/6470 [44:23<30:09,  1.44it/s]
2|3862|Loss: 0.7782780528068542:  60%|█████▉    | 3862/6470 [44:23<29:23,  1.48it/s]
2|3863|Loss: 0.8909145593643188:  60%|█████▉    | 3862/6470 [44:24<29:23,  1.48it/s]
2|3863|Loss: 0.8909145593643188:  60%|█████▉    | 3863/6470 [44:24<28:40,  1.51it/s]
2|3864|Loss: 0.9098539352416992:  60%|█████▉    | 3863/6470 [44:24<28:40,  1.51it/s]
2|3864|Loss: 0.9098539352416992:  60%|█████▉    | 3864/6470 [44:25<28:33,  1.52it/s]
2|3865|Loss: 0.8989614248275757:  60%|█████▉    | 3864/6470 [44:25<28:33,  1.52it/s]
2|3865|Loss: 0.8989614248275757:  60%|█████▉    | 3865/6470 [44:25<27:49,  1.56it/s]
2|3866|Loss: 0.735017716884613:  60%|█████▉    | 3865/6470 [44:26<27:49,  1.56it/s] 
2|3866|Loss: 0.735017716884613:  60%|█████▉    | 3866/6470 [44:26<27:41,  1.57it/s]
2|3867|Loss: 0.8560245037078857:  60%|█████▉    | 3866/6470 [44:26<27:41,  1.57it/s]
2|3867|Loss: 0.8560245037078857:  60%|█████▉    | 3867/6470 [44:26<27:46,  1.56it/s]
2|3868|Loss: 0.8495686054229736:  60%|█████▉    | 3867/6470 [44:27<27:46,  1.56it/s]
2|3868|Loss: 0.8495686054229736:  60%|█████▉    | 3868/6470 [44:27<27:11,  1.59it/s]
2|3869|Loss: 0.899819016456604:  60%|█████▉    | 3868/6470 [44:27<27:11,  1.59it/s] 
2|3869|Loss: 0.899819016456604:  60%|█████▉    | 3869/6470 [44:28<29:43,  1.46it/s]
2|3870|Loss: 0.7895838618278503:  60%|█████▉    | 3869/6470 [44:28<29:43,  1.46it/s]
2|3870|Loss: 0.7895838618278503:  60%|█████▉    | 3870/6470 [44:29<30:44,  1.41it/s]
2|3871|Loss: 0.9373003840446472:  60%|█████▉    | 3870/6470 [44:29<30:44,  1.41it/s]
2|3871|Loss: 0.9373003840446472:  60%|█████▉    | 3871/6470 [44:29<30:05,  1.44it/s]
2|3872|Loss: 0.8884915113449097:  60%|█████▉    | 3871/6470 [44:30<30:05,  1.44it/s]
2|3872|Loss: 0.8884915113449097:  60%|█████▉    | 3872/6470 [44:30<29:04,  1.49it/s]
2|3873|Loss: 0.8737378120422363:  60%|█████▉    | 3872/6470 [44:30<29:04,  1.49it/s]
2|3873|Loss: 0.8737378120422363:  60%|█████▉    | 3873/6470 [44:31<28:38,  1.51it/s]
2|3874|Loss: 0.979351282119751:  60%|█████▉    | 3873/6470 [44:31<28:38,  1.51it/s] 
2|3874|Loss: 0.979351282119751:  60%|█████▉    | 3874/6470 [44:31<30:52,  1.40it/s]
2|3875|Loss: 0.8503687977790833:  60%|█████▉    | 3874/6470 [44:32<30:52,  1.40it/s]
2|3875|Loss: 0.8503687977790833:  60%|█████▉    | 3875/6470 [44:32<29:52,  1.45it/s]
2|3876|Loss: 0.9079186916351318:  60%|█████▉    | 3875/6470 [44:32<29:52,  1.45it/s]
2|3876|Loss: 0.9079186916351318:  60%|█████▉    | 3876/6470 [44:33<31:32,  1.37it/s]
2|3877|Loss: 0.9333828091621399:  60%|█████▉    | 3876/6470 [44:33<31:32,  1.37it/s]
2|3877|Loss: 0.9333828091621399:  60%|█████▉    | 3877/6470 [44:34<32:03,  1.35it/s]
2|3878|Loss: 0.8931836485862732:  60%|█████▉    | 3877/6470 [44:34<32:03,  1.35it/s]
2|3878|Loss: 0.8931836485862732:  60%|█████▉    | 3878/6470 [44:34<32:36,  1.32it/s]
2|3879|Loss: 0.95047926902771:  60%|█████▉    | 3878/6470 [44:35<32:36,  1.32it/s]  
2|3879|Loss: 0.95047926902771:  60%|█████▉    | 3879/6470 [44:35<32:44,  1.32it/s]
2|3880|Loss: 0.9177836775779724:  60%|█████▉    | 3879/6470 [44:36<32:44,  1.32it/s]
2|3880|Loss: 0.9177836775779724:  60%|█████▉    | 3880/6470 [44:36<33:11,  1.30it/s]
2|3881|Loss: 0.7936522960662842:  60%|█████▉    | 3880/6470 [44:36<33:11,  1.30it/s]
2|3881|Loss: 0.7936522960662842:  60%|█████▉    | 3881/6470 [44:37<30:18,  1.42it/s]
2|3882|Loss: 0.824970543384552:  60%|█████▉    | 3881/6470 [44:37<30:18,  1.42it/s] 
2|3882|Loss: 0.824970543384552:  60%|██████    | 3882/6470 [44:37<29:56,  1.44it/s]
2|3883|Loss: 0.8539678454399109:  60%|██████    | 3882/6470 [44:38<29:56,  1.44it/s]
2|3883|Loss: 0.8539678454399109:  60%|██████    | 3883/6470 [44:38<31:55,  1.35it/s]
2|3884|Loss: 0.7624323964118958:  60%|██████    | 3883/6470 [44:38<31:55,  1.35it/s]
2|3884|Loss: 0.7624323964118958:  60%|██████    | 3884/6470 [44:39<29:22,  1.47it/s]
2|3885|Loss: 0.7998735904693604:  60%|██████    | 3884/6470 [44:39<29:22,  1.47it/s]
2|3885|Loss: 0.7998735904693604:  60%|██████    | 3885/6470 [44:39<28:14,  1.53it/s]
2|3886|Loss: 0.7809317111968994:  60%|██████    | 3885/6470 [44:40<28:14,  1.53it/s]
2|3886|Loss: 0.7809317111968994:  60%|██████    | 3886/6470 [44:40<27:48,  1.55it/s]
2|3887|Loss: 1.0456160306930542:  60%|██████    | 3886/6470 [44:40<27:48,  1.55it/s]
2|3887|Loss: 1.0456160306930542:  60%|██████    | 3887/6470 [44:40<26:46,  1.61it/s]
2|3888|Loss: 0.8067419528961182:  60%|██████    | 3887/6470 [44:41<26:46,  1.61it/s]
2|3888|Loss: 0.8067419528961182:  60%|██████    | 3888/6470 [44:41<27:52,  1.54it/s]
2|3889|Loss: 0.8654518723487854:  60%|██████    | 3888/6470 [44:41<27:52,  1.54it/s]
2|3889|Loss: 0.8654518723487854:  60%|██████    | 3889/6470 [44:42<28:07,  1.53it/s]
2|3890|Loss: 0.7927926182746887:  60%|██████    | 3889/6470 [44:42<28:07,  1.53it/s]
2|3890|Loss: 0.7927926182746887:  60%|██████    | 3890/6470 [44:42<28:39,  1.50it/s]
2|3891|Loss: 0.9262107610702515:  60%|██████    | 3890/6470 [44:43<28:39,  1.50it/s]
2|3891|Loss: 0.9262107610702515:  60%|██████    | 3891/6470 [44:43<28:12,  1.52it/s]
2|3892|Loss: 0.9164084792137146:  60%|██████    | 3891/6470 [44:43<28:12,  1.52it/s]
2|3892|Loss: 0.9164084792137146:  60%|██████    | 3892/6470 [44:44<27:36,  1.56it/s]
2|3893|Loss: 0.8818376660346985:  60%|██████    | 3892/6470 [44:44<27:36,  1.56it/s]
2|3893|Loss: 0.8818376660346985:  60%|██████    | 3893/6470 [44:44<28:36,  1.50it/s]
2|3894|Loss: 0.7567543983459473:  60%|██████    | 3893/6470 [44:45<28:36,  1.50it/s]
2|3894|Loss: 0.7567543983459473:  60%|██████    | 3894/6470 [44:45<28:21,  1.51it/s]
2|3895|Loss: 0.8248727917671204:  60%|██████    | 3894/6470 [44:45<28:21,  1.51it/s]
2|3895|Loss: 0.8248727917671204:  60%|██████    | 3895/6470 [44:46<29:54,  1.43it/s]
2|3896|Loss: 0.7693631052970886:  60%|██████    | 3895/6470 [44:46<29:54,  1.43it/s]
2|3896|Loss: 0.7693631052970886:  60%|██████    | 3896/6470 [44:47<30:21,  1.41it/s]
2|3897|Loss: 0.7682559490203857:  60%|██████    | 3896/6470 [44:47<30:21,  1.41it/s]
2|3897|Loss: 0.7682559490203857:  60%|██████    | 3897/6470 [44:47<31:51,  1.35it/s]
2|3898|Loss: 0.9413674473762512:  60%|██████    | 3897/6470 [44:48<31:51,  1.35it/s]
2|3898|Loss: 0.9413674473762512:  60%|██████    | 3898/6470 [44:48<32:43,  1.31it/s]
2|3899|Loss: 0.8538419008255005:  60%|██████    | 3898/6470 [44:49<32:43,  1.31it/s]
2|3899|Loss: 0.8538419008255005:  60%|██████    | 3899/6470 [44:49<31:35,  1.36it/s]
2|3900|Loss: 0.7649226784706116:  60%|██████    | 3899/6470 [44:49<31:35,  1.36it/s]
2|3900|Loss: 0.7649226784706116:  60%|██████    | 3900/6470 [44:49<29:38,  1.45it/s]
2|3901|Loss: 0.6724809408187866:  60%|██████    | 3900/6470 [44:50<29:38,  1.45it/s]
2|3901|Loss: 0.6724809408187866:  60%|██████    | 3901/6470 [44:51<37:30,  1.14it/s]
2|3902|Loss: 0.8472555875778198:  60%|██████    | 3901/6470 [44:51<37:30,  1.14it/s]
2|3902|Loss: 0.8472555875778198:  60%|██████    | 3902/6470 [44:52<37:48,  1.13it/s]
2|3903|Loss: 0.8312690258026123:  60%|██████    | 3902/6470 [44:52<37:48,  1.13it/s]
2|3903|Loss: 0.8312690258026123:  60%|██████    | 3903/6470 [44:52<36:32,  1.17it/s]
2|3904|Loss: 0.8007053732872009:  60%|██████    | 3903/6470 [44:53<36:32,  1.17it/s]
2|3904|Loss: 0.8007053732872009:  60%|██████    | 3904/6470 [44:53<35:06,  1.22it/s]
2|3905|Loss: 0.8947436809539795:  60%|██████    | 3904/6470 [44:54<35:06,  1.22it/s]
2|3905|Loss: 0.8947436809539795:  60%|██████    | 3905/6470 [44:54<32:46,  1.30it/s]
2|3906|Loss: 1.0778937339782715:  60%|██████    | 3905/6470 [44:54<32:46,  1.30it/s]
2|3906|Loss: 1.0778937339782715:  60%|██████    | 3906/6470 [44:55<32:58,  1.30it/s]
2|3907|Loss: 0.8153124451637268:  60%|██████    | 3906/6470 [44:55<32:58,  1.30it/s]
2|3907|Loss: 0.8153124451637268:  60%|██████    | 3907/6470 [44:55<32:25,  1.32it/s]
2|3908|Loss: 0.9265996813774109:  60%|██████    | 3907/6470 [44:56<32:25,  1.32it/s]
2|3908|Loss: 0.9265996813774109:  60%|██████    | 3908/6470 [44:56<31:46,  1.34it/s]
2|3909|Loss: 0.9262059330940247:  60%|██████    | 3908/6470 [44:56<31:46,  1.34it/s]
2|3909|Loss: 0.9262059330940247:  60%|██████    | 3909/6470 [44:57<30:24,  1.40it/s]
2|3910|Loss: 0.8508859872817993:  60%|██████    | 3909/6470 [44:57<30:24,  1.40it/s]
2|3910|Loss: 0.8508859872817993:  60%|██████    | 3910/6470 [44:57<30:03,  1.42it/s]
2|3911|Loss: 0.804164707660675:  60%|██████    | 3910/6470 [44:58<30:03,  1.42it/s] 
2|3911|Loss: 0.804164707660675:  60%|██████    | 3911/6470 [44:58<29:55,  1.43it/s]
2|3912|Loss: 0.9085078835487366:  60%|██████    | 3911/6470 [44:58<29:55,  1.43it/s]
2|3912|Loss: 0.9085078835487366:  60%|██████    | 3912/6470 [44:59<29:59,  1.42it/s]
2|3913|Loss: 0.8790181279182434:  60%|██████    | 3912/6470 [44:59<29:59,  1.42it/s]
2|3913|Loss: 0.8790181279182434:  60%|██████    | 3913/6470 [44:59<29:23,  1.45it/s]
2|3914|Loss: 0.9661286473274231:  60%|██████    | 3913/6470 [45:00<29:23,  1.45it/s]
2|3914|Loss: 0.9661286473274231:  60%|██████    | 3914/6470 [45:00<27:32,  1.55it/s]
2|3915|Loss: 0.7146677374839783:  60%|██████    | 3914/6470 [45:00<27:32,  1.55it/s]
2|3915|Loss: 0.7146677374839783:  61%|██████    | 3915/6470 [45:01<29:16,  1.45it/s]
2|3916|Loss: 0.8940954804420471:  61%|██████    | 3915/6470 [45:01<29:16,  1.45it/s]
2|3916|Loss: 0.8940954804420471:  61%|██████    | 3916/6470 [45:02<31:08,  1.37it/s]
2|3917|Loss: 0.8890713453292847:  61%|██████    | 3916/6470 [45:02<31:08,  1.37it/s]
2|3917|Loss: 0.8890713453292847:  61%|██████    | 3917/6470 [45:02<31:14,  1.36it/s]
2|3918|Loss: 0.8534464240074158:  61%|██████    | 3917/6470 [45:03<31:14,  1.36it/s]
2|3918|Loss: 0.8534464240074158:  61%|██████    | 3918/6470 [45:03<33:27,  1.27it/s]
2|3919|Loss: 0.936726450920105:  61%|██████    | 3918/6470 [45:04<33:27,  1.27it/s] 
2|3919|Loss: 0.936726450920105:  61%|██████    | 3919/6470 [45:04<34:39,  1.23it/s]
2|3920|Loss: 0.8367353081703186:  61%|██████    | 3919/6470 [45:05<34:39,  1.23it/s]
2|3920|Loss: 0.8367353081703186:  61%|██████    | 3920/6470 [45:05<32:55,  1.29it/s]
2|3921|Loss: 0.9463805556297302:  61%|██████    | 3920/6470 [45:05<32:55,  1.29it/s]
2|3921|Loss: 0.9463805556297302:  61%|██████    | 3921/6470 [45:05<30:27,  1.39it/s]
2|3922|Loss: 0.8906030058860779:  61%|██████    | 3921/6470 [45:06<30:27,  1.39it/s]
2|3922|Loss: 0.8906030058860779:  61%|██████    | 3922/6470 [45:06<30:28,  1.39it/s]
2|3923|Loss: 0.8419522643089294:  61%|██████    | 3922/6470 [45:07<30:28,  1.39it/s]
2|3923|Loss: 0.8419522643089294:  61%|██████    | 3923/6470 [45:07<30:43,  1.38it/s]
2|3924|Loss: 0.6848875880241394:  61%|██████    | 3923/6470 [45:07<30:43,  1.38it/s]
2|3924|Loss: 0.6848875880241394:  61%|██████    | 3924/6470 [45:08<29:41,  1.43it/s]
2|3925|Loss: 0.7965928912162781:  61%|██████    | 3924/6470 [45:08<29:41,  1.43it/s]
2|3925|Loss: 0.7965928912162781:  61%|██████    | 3925/6470 [45:08<31:16,  1.36it/s]
2|3926|Loss: 0.8311704397201538:  61%|██████    | 3925/6470 [45:09<31:16,  1.36it/s]
2|3926|Loss: 0.8311704397201538:  61%|██████    | 3926/6470 [45:09<32:46,  1.29it/s]
2|3927|Loss: 0.8814611434936523:  61%|██████    | 3926/6470 [45:10<32:46,  1.29it/s]
2|3927|Loss: 0.8814611434936523:  61%|██████    | 3927/6470 [45:10<32:21,  1.31it/s]
2|3928|Loss: 0.8269627690315247:  61%|██████    | 3927/6470 [45:10<32:21,  1.31it/s]
2|3928|Loss: 0.8269627690315247:  61%|██████    | 3928/6470 [45:11<31:53,  1.33it/s]
2|3929|Loss: 0.7199817299842834:  61%|██████    | 3928/6470 [45:11<31:53,  1.33it/s]
2|3929|Loss: 0.7199817299842834:  61%|██████    | 3929/6470 [45:11<30:04,  1.41it/s]
2|3930|Loss: 0.9415128827095032:  61%|██████    | 3929/6470 [45:12<30:04,  1.41it/s]
2|3930|Loss: 0.9415128827095032:  61%|██████    | 3930/6470 [45:12<28:28,  1.49it/s]
2|3931|Loss: 0.9607042670249939:  61%|██████    | 3930/6470 [45:12<28:28,  1.49it/s]
2|3931|Loss: 0.9607042670249939:  61%|██████    | 3931/6470 [45:12<27:55,  1.52it/s]
2|3932|Loss: 0.8694402575492859:  61%|██████    | 3931/6470 [45:13<27:55,  1.52it/s]
2|3932|Loss: 0.8694402575492859:  61%|██████    | 3932/6470 [45:13<28:37,  1.48it/s]
2|3933|Loss: 0.9096891283988953:  61%|██████    | 3932/6470 [45:14<28:37,  1.48it/s]
2|3933|Loss: 0.9096891283988953:  61%|██████    | 3933/6470 [45:14<28:07,  1.50it/s]
2|3934|Loss: 0.8293004631996155:  61%|██████    | 3933/6470 [45:14<28:07,  1.50it/s]
2|3934|Loss: 0.8293004631996155:  61%|██████    | 3934/6470 [45:15<29:15,  1.44it/s]
2|3935|Loss: 0.7714806199073792:  61%|██████    | 3934/6470 [45:15<29:15,  1.44it/s]
2|3935|Loss: 0.7714806199073792:  61%|██████    | 3935/6470 [45:15<31:15,  1.35it/s]
2|3936|Loss: 0.8111810684204102:  61%|██████    | 3935/6470 [45:16<31:15,  1.35it/s]
2|3936|Loss: 0.8111810684204102:  61%|██████    | 3936/6470 [45:16<30:29,  1.39it/s]
2|3937|Loss: 0.8809525966644287:  61%|██████    | 3936/6470 [45:16<30:29,  1.39it/s]
2|3937|Loss: 0.8809525966644287:  61%|██████    | 3937/6470 [45:17<28:27,  1.48it/s]
2|3938|Loss: 0.8807874917984009:  61%|██████    | 3937/6470 [45:17<28:27,  1.48it/s]
2|3938|Loss: 0.8807874917984009:  61%|██████    | 3938/6470 [45:17<29:30,  1.43it/s]
2|3939|Loss: 0.9089086651802063:  61%|██████    | 3938/6470 [45:18<29:30,  1.43it/s]
2|3939|Loss: 0.9089086651802063:  61%|██████    | 3939/6470 [45:18<29:48,  1.41it/s]
2|3940|Loss: 1.0190610885620117:  61%|██████    | 3939/6470 [45:19<29:48,  1.41it/s]
2|3940|Loss: 1.0190610885620117:  61%|██████    | 3940/6470 [45:19<30:34,  1.38it/s]
2|3941|Loss: 0.6126630902290344:  61%|██████    | 3940/6470 [45:19<30:34,  1.38it/s]
2|3941|Loss: 0.6126630902290344:  61%|██████    | 3941/6470 [45:20<31:26,  1.34it/s]
2|3942|Loss: 0.8513801097869873:  61%|██████    | 3941/6470 [45:20<31:26,  1.34it/s]
2|3942|Loss: 0.8513801097869873:  61%|██████    | 3942/6470 [45:21<32:57,  1.28it/s]
2|3943|Loss: 0.8997535109519958:  61%|██████    | 3942/6470 [45:21<32:57,  1.28it/s]
2|3943|Loss: 0.8997535109519958:  61%|██████    | 3943/6470 [45:21<32:26,  1.30it/s]
2|3944|Loss: 0.7530612945556641:  61%|██████    | 3943/6470 [45:22<32:26,  1.30it/s]
2|3944|Loss: 0.7530612945556641:  61%|██████    | 3944/6470 [45:22<31:36,  1.33it/s]
2|3945|Loss: 1.0640004873275757:  61%|██████    | 3944/6470 [45:22<31:36,  1.33it/s]
2|3945|Loss: 1.0640004873275757:  61%|██████    | 3945/6470 [45:23<30:19,  1.39it/s]
2|3946|Loss: 0.6700660586357117:  61%|██████    | 3945/6470 [45:23<30:19,  1.39it/s]
2|3946|Loss: 0.6700660586357117:  61%|██████    | 3946/6470 [45:24<31:23,  1.34it/s]
2|3947|Loss: 0.736081063747406:  61%|██████    | 3946/6470 [45:24<31:23,  1.34it/s] 
2|3947|Loss: 0.736081063747406:  61%|██████    | 3947/6470 [45:24<29:16,  1.44it/s]
2|3948|Loss: 0.9957216382026672:  61%|██████    | 3947/6470 [45:24<29:16,  1.44it/s]
2|3948|Loss: 0.9957216382026672:  61%|██████    | 3948/6470 [45:25<28:13,  1.49it/s]
2|3949|Loss: 0.8451404571533203:  61%|██████    | 3948/6470 [45:25<28:13,  1.49it/s]
2|3949|Loss: 0.8451404571533203:  61%|██████    | 3949/6470 [45:25<26:54,  1.56it/s]
2|3950|Loss: 0.8996820449829102:  61%|██████    | 3949/6470 [45:26<26:54,  1.56it/s]
2|3950|Loss: 0.8996820449829102:  61%|██████    | 3950/6470 [45:26<24:48,  1.69it/s]
2|3951|Loss: 0.770719587802887:  61%|██████    | 3950/6470 [45:26<24:48,  1.69it/s] 
2|3951|Loss: 0.770719587802887:  61%|██████    | 3951/6470 [45:26<25:54,  1.62it/s]
2|3952|Loss: 0.7372967004776001:  61%|██████    | 3951/6470 [45:27<25:54,  1.62it/s]
2|3952|Loss: 0.7372967004776001:  61%|██████    | 3952/6470 [45:27<27:25,  1.53it/s]
2|3953|Loss: 0.94312983751297:  61%|██████    | 3952/6470 [45:28<27:25,  1.53it/s]  
2|3953|Loss: 0.94312983751297:  61%|██████    | 3953/6470 [45:28<26:07,  1.61it/s]
2|3954|Loss: 0.908150315284729:  61%|██████    | 3953/6470 [45:28<26:07,  1.61it/s]
2|3954|Loss: 0.908150315284729:  61%|██████    | 3954/6470 [45:28<26:00,  1.61it/s]
2|3955|Loss: 0.8060737252235413:  61%|██████    | 3954/6470 [45:29<26:00,  1.61it/s]
2|3955|Loss: 0.8060737252235413:  61%|██████    | 3955/6470 [45:29<25:21,  1.65it/s]
2|3956|Loss: 1.0213356018066406:  61%|██████    | 3955/6470 [45:29<25:21,  1.65it/s]
2|3956|Loss: 1.0213356018066406:  61%|██████    | 3956/6470 [45:29<24:19,  1.72it/s]
2|3957|Loss: 1.0261509418487549:  61%|██████    | 3956/6470 [45:30<24:19,  1.72it/s]
2|3957|Loss: 1.0261509418487549:  61%|██████    | 3957/6470 [45:30<23:50,  1.76it/s]
2|3958|Loss: 0.8738560080528259:  61%|██████    | 3957/6470 [45:30<23:50,  1.76it/s]
2|3958|Loss: 0.8738560080528259:  61%|██████    | 3958/6470 [45:31<24:32,  1.71it/s]
2|3959|Loss: 0.9354971647262573:  61%|██████    | 3958/6470 [45:31<24:32,  1.71it/s]
2|3959|Loss: 0.9354971647262573:  61%|██████    | 3959/6470 [45:31<24:56,  1.68it/s]
2|3960|Loss: 0.8405134677886963:  61%|██████    | 3959/6470 [45:32<24:56,  1.68it/s]
2|3960|Loss: 0.8405134677886963:  61%|██████    | 3960/6470 [45:32<26:15,  1.59it/s]
2|3961|Loss: 0.9828092455863953:  61%|██████    | 3960/6470 [45:32<26:15,  1.59it/s]
2|3961|Loss: 0.9828092455863953:  61%|██████    | 3961/6470 [45:32<25:18,  1.65it/s]
2|3962|Loss: 0.9609129428863525:  61%|██████    | 3961/6470 [45:33<25:18,  1.65it/s]
2|3962|Loss: 0.9609129428863525:  61%|██████    | 3962/6470 [45:33<26:22,  1.59it/s]
2|3963|Loss: 0.7185731530189514:  61%|██████    | 3962/6470 [45:34<26:22,  1.59it/s]
2|3963|Loss: 0.7185731530189514:  61%|██████▏   | 3963/6470 [45:34<29:29,  1.42it/s]
2|3964|Loss: 0.7386154532432556:  61%|██████▏   | 3963/6470 [45:34<29:29,  1.42it/s]
2|3964|Loss: 0.7386154532432556:  61%|██████▏   | 3964/6470 [45:35<30:52,  1.35it/s]
2|3965|Loss: 0.9577711820602417:  61%|██████▏   | 3964/6470 [45:35<30:52,  1.35it/s]
2|3965|Loss: 0.9577711820602417:  61%|██████▏   | 3965/6470 [45:36<30:40,  1.36it/s]
2|3966|Loss: 0.8852159380912781:  61%|██████▏   | 3965/6470 [45:36<30:40,  1.36it/s]
2|3966|Loss: 0.8852159380912781:  61%|██████▏   | 3966/6470 [45:36<29:01,  1.44it/s]
2|3967|Loss: 0.8688992261886597:  61%|██████▏   | 3966/6470 [45:37<29:01,  1.44it/s]
2|3967|Loss: 0.8688992261886597:  61%|██████▏   | 3967/6470 [45:37<30:04,  1.39it/s]
2|3968|Loss: 0.8546454906463623:  61%|██████▏   | 3967/6470 [45:37<30:04,  1.39it/s]
2|3968|Loss: 0.8546454906463623:  61%|██████▏   | 3968/6470 [45:38<29:35,  1.41it/s]
2|3969|Loss: 0.8192477226257324:  61%|██████▏   | 3968/6470 [45:38<29:35,  1.41it/s]
2|3969|Loss: 0.8192477226257324:  61%|██████▏   | 3969/6470 [45:38<28:04,  1.48it/s]
2|3970|Loss: 0.8988329172134399:  61%|██████▏   | 3969/6470 [45:39<28:04,  1.48it/s]
2|3970|Loss: 0.8988329172134399:  61%|██████▏   | 3970/6470 [45:39<27:31,  1.51it/s]
2|3971|Loss: 0.9642972350120544:  61%|██████▏   | 3970/6470 [45:39<27:31,  1.51it/s]
2|3971|Loss: 0.9642972350120544:  61%|██████▏   | 3971/6470 [45:39<26:25,  1.58it/s]
2|3972|Loss: 0.7768896222114563:  61%|██████▏   | 3971/6470 [45:40<26:25,  1.58it/s]
2|3972|Loss: 0.7768896222114563:  61%|██████▏   | 3972/6470 [45:40<26:57,  1.54it/s]
2|3973|Loss: 0.7334120869636536:  61%|██████▏   | 3972/6470 [45:40<26:57,  1.54it/s]
2|3973|Loss: 0.7334120869636536:  61%|██████▏   | 3973/6470 [45:41<25:41,  1.62it/s]
2|3974|Loss: 0.8962962031364441:  61%|██████▏   | 3973/6470 [45:41<25:41,  1.62it/s]
2|3974|Loss: 0.8962962031364441:  61%|██████▏   | 3974/6470 [45:41<27:14,  1.53it/s]
2|3975|Loss: 1.0080280303955078:  61%|██████▏   | 3974/6470 [45:42<27:14,  1.53it/s]
2|3975|Loss: 1.0080280303955078:  61%|██████▏   | 3975/6470 [45:42<28:33,  1.46it/s]
2|3976|Loss: 0.97532719373703:  61%|██████▏   | 3975/6470 [45:43<28:33,  1.46it/s]  
2|3976|Loss: 0.97532719373703:  61%|██████▏   | 3976/6470 [45:43<28:02,  1.48it/s]
2|3977|Loss: 0.674155592918396:  61%|██████▏   | 3976/6470 [45:43<28:02,  1.48it/s]
2|3977|Loss: 0.674155592918396:  61%|██████▏   | 3977/6470 [45:43<27:15,  1.52it/s]
2|3978|Loss: 0.7903850674629211:  61%|██████▏   | 3977/6470 [45:44<27:15,  1.52it/s]
2|3978|Loss: 0.7903850674629211:  61%|██████▏   | 3978/6470 [45:44<28:02,  1.48it/s]
2|3979|Loss: 0.7525482177734375:  61%|██████▏   | 3978/6470 [45:45<28:02,  1.48it/s]
2|3979|Loss: 0.7525482177734375:  61%|██████▏   | 3979/6470 [45:45<29:00,  1.43it/s]
2|3980|Loss: 0.8806759715080261:  61%|██████▏   | 3979/6470 [45:45<29:00,  1.43it/s]
2|3980|Loss: 0.8806759715080261:  62%|██████▏   | 3980/6470 [45:45<27:16,  1.52it/s]
2|3981|Loss: 0.825421929359436:  62%|██████▏   | 3980/6470 [45:46<27:16,  1.52it/s] 
2|3981|Loss: 0.825421929359436:  62%|██████▏   | 3981/6470 [45:46<26:13,  1.58it/s]
2|3982|Loss: 0.8811697959899902:  62%|██████▏   | 3981/6470 [45:46<26:13,  1.58it/s]
2|3982|Loss: 0.8811697959899902:  62%|██████▏   | 3982/6470 [45:47<26:42,  1.55it/s]
2|3983|Loss: 0.9522954225540161:  62%|██████▏   | 3982/6470 [45:47<26:42,  1.55it/s]
2|3983|Loss: 0.9522954225540161:  62%|██████▏   | 3983/6470 [45:47<26:37,  1.56it/s]
2|3984|Loss: 0.8081799149513245:  62%|██████▏   | 3983/6470 [45:48<26:37,  1.56it/s]
2|3984|Loss: 0.8081799149513245:  62%|██████▏   | 3984/6470 [45:48<28:39,  1.45it/s]
2|3985|Loss: 0.9652202129364014:  62%|██████▏   | 3984/6470 [45:49<28:39,  1.45it/s]
2|3985|Loss: 0.9652202129364014:  62%|██████▏   | 3985/6470 [45:49<30:54,  1.34it/s]
2|3986|Loss: 1.1988118886947632:  62%|██████▏   | 3985/6470 [45:50<30:54,  1.34it/s]
2|3986|Loss: 1.1988118886947632:  62%|██████▏   | 3986/6470 [45:50<35:02,  1.18it/s]
2|3987|Loss: 0.916517436504364:  62%|██████▏   | 3986/6470 [45:51<35:02,  1.18it/s] 
2|3987|Loss: 0.916517436504364:  62%|██████▏   | 3987/6470 [45:51<35:11,  1.18it/s]
2|3988|Loss: 1.0071632862091064:  62%|██████▏   | 3987/6470 [45:51<35:11,  1.18it/s]
2|3988|Loss: 1.0071632862091064:  62%|██████▏   | 3988/6470 [45:52<32:58,  1.25it/s]
2|3989|Loss: 1.0657824277877808:  62%|██████▏   | 3988/6470 [45:52<32:58,  1.25it/s]
2|3989|Loss: 1.0657824277877808:  62%|██████▏   | 3989/6470 [45:52<32:03,  1.29it/s]
2|3990|Loss: 0.8545016050338745:  62%|██████▏   | 3989/6470 [45:53<32:03,  1.29it/s]
2|3990|Loss: 0.8545016050338745:  62%|██████▏   | 3990/6470 [45:53<30:06,  1.37it/s]
2|3991|Loss: 1.0266574621200562:  62%|██████▏   | 3990/6470 [45:53<30:06,  1.37it/s]
2|3991|Loss: 1.0266574621200562:  62%|██████▏   | 3991/6470 [45:54<31:01,  1.33it/s]
2|3992|Loss: 0.8573946356773376:  62%|██████▏   | 3991/6470 [45:54<31:01,  1.33it/s]
2|3992|Loss: 0.8573946356773376:  62%|██████▏   | 3992/6470 [45:54<29:47,  1.39it/s]
2|3993|Loss: 0.9316739439964294:  62%|██████▏   | 3992/6470 [45:55<29:47,  1.39it/s]
2|3993|Loss: 0.9316739439964294:  62%|██████▏   | 3993/6470 [45:55<30:06,  1.37it/s]
2|3994|Loss: 0.9452139735221863:  62%|██████▏   | 3993/6470 [45:56<30:06,  1.37it/s]
2|3994|Loss: 0.9452139735221863:  62%|██████▏   | 3994/6470 [45:56<26:54,  1.53it/s]
2|3995|Loss: 0.7615705132484436:  62%|██████▏   | 3994/6470 [45:56<26:54,  1.53it/s]
2|3995|Loss: 0.7615705132484436:  62%|██████▏   | 3995/6470 [45:56<26:00,  1.59it/s]
2|3996|Loss: 0.9397925734519958:  62%|██████▏   | 3995/6470 [45:57<26:00,  1.59it/s]
2|3996|Loss: 0.9397925734519958:  62%|██████▏   | 3996/6470 [45:57<27:30,  1.50it/s]
2|3997|Loss: 0.8275042176246643:  62%|██████▏   | 3996/6470 [45:57<27:30,  1.50it/s]
2|3997|Loss: 0.8275042176246643:  62%|██████▏   | 3997/6470 [45:58<27:24,  1.50it/s]
2|3998|Loss: 0.8706664443016052:  62%|██████▏   | 3997/6470 [45:58<27:24,  1.50it/s]
2|3998|Loss: 0.8706664443016052:  62%|██████▏   | 3998/6470 [45:58<26:09,  1.58it/s]
2|3999|Loss: 0.9702563285827637:  62%|██████▏   | 3998/6470 [45:59<26:09,  1.58it/s]
2|3999|Loss: 0.9702563285827637:  62%|██████▏   | 3999/6470 [45:59<25:24,  1.62it/s]
2|4000|Loss: 0.737837553024292:  62%|██████▏   | 3999/6470 [45:59<25:24,  1.62it/s] 
2|4000|Loss: 0.737837553024292:  62%|██████▏   | 4000/6470 [45:59<25:30,  1.61it/s]
2|4001|Loss: 0.9157877564430237:  62%|██████▏   | 4000/6470 [46:00<25:30,  1.61it/s]
2|4001|Loss: 0.9157877564430237:  62%|██████▏   | 4001/6470 [46:00<27:24,  1.50it/s]
2|4002|Loss: 0.881297767162323:  62%|██████▏   | 4001/6470 [46:01<27:24,  1.50it/s] 
2|4002|Loss: 0.881297767162323:  62%|██████▏   | 4002/6470 [46:01<26:12,  1.57it/s]
2|4003|Loss: 0.7759916186332703:  62%|██████▏   | 4002/6470 [46:01<26:12,  1.57it/s]
2|4003|Loss: 0.7759916186332703:  62%|██████▏   | 4003/6470 [46:01<26:51,  1.53it/s]
2|4004|Loss: 0.8917878866195679:  62%|██████▏   | 4003/6470 [46:02<26:51,  1.53it/s]
2|4004|Loss: 0.8917878866195679:  62%|██████▏   | 4004/6470 [46:02<26:46,  1.54it/s]
2|4005|Loss: 0.9330887198448181:  62%|██████▏   | 4004/6470 [46:02<26:46,  1.54it/s]
2|4005|Loss: 0.9330887198448181:  62%|██████▏   | 4005/6470 [46:03<27:37,  1.49it/s]
2|4006|Loss: 0.7930610775947571:  62%|██████▏   | 4005/6470 [46:03<27:37,  1.49it/s]
2|4006|Loss: 0.7930610775947571:  62%|██████▏   | 4006/6470 [46:04<29:06,  1.41it/s]
2|4007|Loss: 0.7746990323066711:  62%|██████▏   | 4006/6470 [46:04<29:06,  1.41it/s]
2|4007|Loss: 0.7746990323066711:  62%|██████▏   | 4007/6470 [46:04<28:26,  1.44it/s]
2|4008|Loss: 0.850991427898407:  62%|██████▏   | 4007/6470 [46:05<28:26,  1.44it/s] 
2|4008|Loss: 0.850991427898407:  62%|██████▏   | 4008/6470 [46:05<28:49,  1.42it/s]
2|4009|Loss: 0.9626822471618652:  62%|██████▏   | 4008/6470 [46:05<28:49,  1.42it/s]
2|4009|Loss: 0.9626822471618652:  62%|██████▏   | 4009/6470 [46:06<30:46,  1.33it/s]
2|4010|Loss: 0.8415517210960388:  62%|██████▏   | 4009/6470 [46:06<30:46,  1.33it/s]
2|4010|Loss: 0.8415517210960388:  62%|██████▏   | 4010/6470 [46:07<31:44,  1.29it/s]
2|4011|Loss: 1.0809085369110107:  62%|██████▏   | 4010/6470 [46:07<31:44,  1.29it/s]
2|4011|Loss: 1.0809085369110107:  62%|██████▏   | 4011/6470 [46:07<31:32,  1.30it/s]
2|4012|Loss: 0.8286892175674438:  62%|██████▏   | 4011/6470 [46:08<31:32,  1.30it/s]
2|4012|Loss: 0.8286892175674438:  62%|██████▏   | 4012/6470 [46:08<32:03,  1.28it/s]
2|4013|Loss: 0.8336455821990967:  62%|██████▏   | 4012/6470 [46:09<32:03,  1.28it/s]
2|4013|Loss: 0.8336455821990967:  62%|██████▏   | 4013/6470 [46:09<30:47,  1.33it/s]
2|4014|Loss: 0.9271352887153625:  62%|██████▏   | 4013/6470 [46:09<30:47,  1.33it/s]
2|4014|Loss: 0.9271352887153625:  62%|██████▏   | 4014/6470 [46:10<31:48,  1.29it/s]
2|4015|Loss: 0.6923942565917969:  62%|██████▏   | 4014/6470 [46:10<31:48,  1.29it/s]
2|4015|Loss: 0.6923942565917969:  62%|██████▏   | 4015/6470 [46:10<30:30,  1.34it/s]
2|4016|Loss: 0.8411379456520081:  62%|██████▏   | 4015/6470 [46:11<30:30,  1.34it/s]
2|4016|Loss: 0.8411379456520081:  62%|██████▏   | 4016/6470 [46:11<29:39,  1.38it/s]
2|4017|Loss: 0.9456359148025513:  62%|██████▏   | 4016/6470 [46:11<29:39,  1.38it/s]
2|4017|Loss: 0.9456359148025513:  62%|██████▏   | 4017/6470 [46:12<28:32,  1.43it/s]
2|4018|Loss: 0.8350467085838318:  62%|██████▏   | 4017/6470 [46:12<28:32,  1.43it/s]
2|4018|Loss: 0.8350467085838318:  62%|██████▏   | 4018/6470 [46:13<29:51,  1.37it/s]
2|4019|Loss: 1.0021780729293823:  62%|██████▏   | 4018/6470 [46:13<29:51,  1.37it/s]
2|4019|Loss: 1.0021780729293823:  62%|██████▏   | 4019/6470 [46:13<31:24,  1.30it/s]
2|4020|Loss: 0.9870365858078003:  62%|██████▏   | 4019/6470 [46:14<31:24,  1.30it/s]
2|4020|Loss: 0.9870365858078003:  62%|██████▏   | 4020/6470 [46:14<31:31,  1.29it/s]
2|4021|Loss: 0.9196243286132812:  62%|██████▏   | 4020/6470 [46:15<31:31,  1.29it/s]
2|4021|Loss: 0.9196243286132812:  62%|██████▏   | 4021/6470 [46:15<29:10,  1.40it/s]
2|4022|Loss: 0.8348280787467957:  62%|██████▏   | 4021/6470 [46:15<29:10,  1.40it/s]
2|4022|Loss: 0.8348280787467957:  62%|██████▏   | 4022/6470 [46:15<28:03,  1.45it/s]
2|4023|Loss: 0.8544182181358337:  62%|██████▏   | 4022/6470 [46:16<28:03,  1.45it/s]
2|4023|Loss: 0.8544182181358337:  62%|██████▏   | 4023/6470 [46:16<27:14,  1.50it/s]
2|4024|Loss: 0.5941339731216431:  62%|██████▏   | 4023/6470 [46:16<27:14,  1.50it/s]
2|4024|Loss: 0.5941339731216431:  62%|██████▏   | 4024/6470 [46:17<30:33,  1.33it/s]
2|4025|Loss: 0.9497791528701782:  62%|██████▏   | 4024/6470 [46:17<30:33,  1.33it/s]
2|4025|Loss: 0.9497791528701782:  62%|██████▏   | 4025/6470 [46:18<30:15,  1.35it/s]
2|4026|Loss: 0.8654696345329285:  62%|██████▏   | 4025/6470 [46:18<30:15,  1.35it/s]
2|4026|Loss: 0.8654696345329285:  62%|██████▏   | 4026/6470 [46:18<29:54,  1.36it/s]
2|4027|Loss: 0.810312032699585:  62%|██████▏   | 4026/6470 [46:19<29:54,  1.36it/s] 
2|4027|Loss: 0.810312032699585:  62%|██████▏   | 4027/6470 [46:19<29:56,  1.36it/s]
2|4028|Loss: 0.8303319215774536:  62%|██████▏   | 4027/6470 [46:20<29:56,  1.36it/s]
2|4028|Loss: 0.8303319215774536:  62%|██████▏   | 4028/6470 [46:20<28:32,  1.43it/s]
2|4029|Loss: 0.818149745464325:  62%|██████▏   | 4028/6470 [46:20<28:32,  1.43it/s] 
2|4029|Loss: 0.818149745464325:  62%|██████▏   | 4029/6470 [46:20<27:29,  1.48it/s]
2|4030|Loss: 0.9551496505737305:  62%|██████▏   | 4029/6470 [46:21<27:29,  1.48it/s]
2|4030|Loss: 0.9551496505737305:  62%|██████▏   | 4030/6470 [46:21<26:24,  1.54it/s]
2|4031|Loss: 0.875609278678894:  62%|██████▏   | 4030/6470 [46:21<26:24,  1.54it/s] 
2|4031|Loss: 0.875609278678894:  62%|██████▏   | 4031/6470 [46:22<27:02,  1.50it/s]
2|4032|Loss: 0.919368326663971:  62%|██████▏   | 4031/6470 [46:22<27:02,  1.50it/s]
2|4032|Loss: 0.919368326663971:  62%|██████▏   | 4032/6470 [46:22<26:16,  1.55it/s]
2|4033|Loss: 0.8792388439178467:  62%|██████▏   | 4032/6470 [46:23<26:16,  1.55it/s]
2|4033|Loss: 0.8792388439178467:  62%|██████▏   | 4033/6470 [46:23<26:56,  1.51it/s]
2|4034|Loss: 0.8093895316123962:  62%|██████▏   | 4033/6470 [46:23<26:56,  1.51it/s]
2|4034|Loss: 0.8093895316123962:  62%|██████▏   | 4034/6470 [46:24<29:34,  1.37it/s]
2|4035|Loss: 0.9073111414909363:  62%|██████▏   | 4034/6470 [46:24<29:34,  1.37it/s]
2|4035|Loss: 0.9073111414909363:  62%|██████▏   | 4035/6470 [46:25<30:49,  1.32it/s]
2|4036|Loss: 0.8945335149765015:  62%|██████▏   | 4035/6470 [46:25<30:49,  1.32it/s]
2|4036|Loss: 0.8945335149765015:  62%|██████▏   | 4036/6470 [46:25<28:00,  1.45it/s]
2|4037|Loss: 0.9265653491020203:  62%|██████▏   | 4036/6470 [46:26<28:00,  1.45it/s]
2|4037|Loss: 0.9265653491020203:  62%|██████▏   | 4037/6470 [46:26<28:26,  1.43it/s]
2|4038|Loss: 0.8969966173171997:  62%|██████▏   | 4037/6470 [46:26<28:26,  1.43it/s]
2|4038|Loss: 0.8969966173171997:  62%|██████▏   | 4038/6470 [46:27<28:46,  1.41it/s]
2|4039|Loss: 0.8853570818901062:  62%|██████▏   | 4038/6470 [46:27<28:46,  1.41it/s]
2|4039|Loss: 0.8853570818901062:  62%|██████▏   | 4039/6470 [46:27<28:41,  1.41it/s]
2|4040|Loss: 0.856342077255249:  62%|██████▏   | 4039/6470 [46:28<28:41,  1.41it/s] 
2|4040|Loss: 0.856342077255249:  62%|██████▏   | 4040/6470 [46:28<28:57,  1.40it/s]
2|4041|Loss: 0.8369393944740295:  62%|██████▏   | 4040/6470 [46:29<28:57,  1.40it/s]
2|4041|Loss: 0.8369393944740295:  62%|██████▏   | 4041/6470 [46:29<29:10,  1.39it/s]
2|4042|Loss: 0.9560549259185791:  62%|██████▏   | 4041/6470 [46:29<29:10,  1.39it/s]
2|4042|Loss: 0.9560549259185791:  62%|██████▏   | 4042/6470 [46:29<27:50,  1.45it/s]
2|4043|Loss: 0.8977792859077454:  62%|██████▏   | 4042/6470 [46:30<27:50,  1.45it/s]
2|4043|Loss: 0.8977792859077454:  62%|██████▏   | 4043/6470 [46:30<25:52,  1.56it/s]
2|4044|Loss: 0.808753252029419:  62%|██████▏   | 4043/6470 [46:30<25:52,  1.56it/s] 
2|4044|Loss: 0.808753252029419:  63%|██████▎   | 4044/6470 [46:31<26:22,  1.53it/s]
2|4045|Loss: 0.8595129251480103:  63%|██████▎   | 4044/6470 [46:31<26:22,  1.53it/s]
2|4045|Loss: 0.8595129251480103:  63%|██████▎   | 4045/6470 [46:31<26:31,  1.52it/s]
2|4046|Loss: 0.8948544859886169:  63%|██████▎   | 4045/6470 [46:32<26:31,  1.52it/s]
2|4046|Loss: 0.8948544859886169:  63%|██████▎   | 4046/6470 [46:32<25:46,  1.57it/s]
2|4047|Loss: 0.9058647751808167:  63%|██████▎   | 4046/6470 [46:32<25:46,  1.57it/s]
2|4047|Loss: 0.9058647751808167:  63%|██████▎   | 4047/6470 [46:33<25:27,  1.59it/s]
2|4048|Loss: 0.9263916015625:  63%|██████▎   | 4047/6470 [46:33<25:27,  1.59it/s]   
2|4048|Loss: 0.9263916015625:  63%|██████▎   | 4048/6470 [46:33<25:15,  1.60it/s]
2|4049|Loss: 0.7899778485298157:  63%|██████▎   | 4048/6470 [46:33<25:15,  1.60it/s]
2|4049|Loss: 0.7899778485298157:  63%|██████▎   | 4049/6470 [46:34<25:10,  1.60it/s]
2|4050|Loss: 0.8958325982093811:  63%|██████▎   | 4049/6470 [46:34<25:10,  1.60it/s]
2|4050|Loss: 0.8958325982093811:  63%|██████▎   | 4050/6470 [46:35<26:57,  1.50it/s]
2|4051|Loss: 0.9701181054115295:  63%|██████▎   | 4050/6470 [46:35<26:57,  1.50it/s]
2|4051|Loss: 0.9701181054115295:  63%|██████▎   | 4051/6470 [46:35<29:15,  1.38it/s]
2|4052|Loss: 1.014156699180603:  63%|██████▎   | 4051/6470 [46:36<29:15,  1.38it/s] 
2|4052|Loss: 1.014156699180603:  63%|██████▎   | 4052/6470 [46:36<31:13,  1.29it/s]
2|4053|Loss: 0.7579753398895264:  63%|██████▎   | 4052/6470 [46:37<31:13,  1.29it/s]
2|4053|Loss: 0.7579753398895264:  63%|██████▎   | 4053/6470 [46:37<28:27,  1.42it/s]
2|4054|Loss: 0.8896111845970154:  63%|██████▎   | 4053/6470 [46:37<28:27,  1.42it/s]
2|4054|Loss: 0.8896111845970154:  63%|██████▎   | 4054/6470 [46:38<28:13,  1.43it/s]
2|4055|Loss: 0.8156676292419434:  63%|██████▎   | 4054/6470 [46:38<28:13,  1.43it/s]
2|4055|Loss: 0.8156676292419434:  63%|██████▎   | 4055/6470 [46:38<29:17,  1.37it/s]
2|4056|Loss: 0.7586784958839417:  63%|██████▎   | 4055/6470 [46:39<29:17,  1.37it/s]
2|4056|Loss: 0.7586784958839417:  63%|██████▎   | 4056/6470 [46:39<29:45,  1.35it/s]
2|4057|Loss: 0.7083070874214172:  63%|██████▎   | 4056/6470 [46:40<29:45,  1.35it/s]
2|4057|Loss: 0.7083070874214172:  63%|██████▎   | 4057/6470 [46:40<29:58,  1.34it/s]
2|4058|Loss: 0.8780829310417175:  63%|██████▎   | 4057/6470 [46:40<29:58,  1.34it/s]
2|4058|Loss: 0.8780829310417175:  63%|██████▎   | 4058/6470 [46:41<28:59,  1.39it/s]
2|4059|Loss: 0.8427053093910217:  63%|██████▎   | 4058/6470 [46:41<28:59,  1.39it/s]
2|4059|Loss: 0.8427053093910217:  63%|██████▎   | 4059/6470 [46:41<27:08,  1.48it/s]
2|4060|Loss: 0.8400196433067322:  63%|██████▎   | 4059/6470 [46:41<27:08,  1.48it/s]
2|4060|Loss: 0.8400196433067322:  63%|██████▎   | 4060/6470 [46:42<28:53,  1.39it/s]
2|4061|Loss: 0.828921377658844:  63%|██████▎   | 4060/6470 [46:42<28:53,  1.39it/s] 
2|4061|Loss: 0.828921377658844:  63%|██████▎   | 4061/6470 [46:42<25:56,  1.55it/s]
2|4062|Loss: 0.8065379858016968:  63%|██████▎   | 4061/6470 [46:43<25:56,  1.55it/s]
2|4062|Loss: 0.8065379858016968:  63%|██████▎   | 4062/6470 [46:43<23:31,  1.71it/s]
2|4063|Loss: 0.9115456938743591:  63%|██████▎   | 4062/6470 [46:43<23:31,  1.71it/s]
2|4063|Loss: 0.9115456938743591:  63%|██████▎   | 4063/6470 [46:43<24:37,  1.63it/s]
2|4064|Loss: 0.8616667985916138:  63%|██████▎   | 4063/6470 [46:44<24:37,  1.63it/s]
2|4064|Loss: 0.8616667985916138:  63%|██████▎   | 4064/6470 [46:44<25:02,  1.60it/s]
2|4065|Loss: 0.9132384061813354:  63%|██████▎   | 4064/6470 [46:45<25:02,  1.60it/s]
2|4065|Loss: 0.9132384061813354:  63%|██████▎   | 4065/6470 [46:45<26:14,  1.53it/s]
2|4066|Loss: 0.850867509841919:  63%|██████▎   | 4065/6470 [46:45<26:14,  1.53it/s] 
2|4066|Loss: 0.850867509841919:  63%|██████▎   | 4066/6470 [46:46<26:24,  1.52it/s]
2|4067|Loss: 0.8885375261306763:  63%|██████▎   | 4066/6470 [46:46<26:24,  1.52it/s]
2|4067|Loss: 0.8885375261306763:  63%|██████▎   | 4067/6470 [46:46<28:31,  1.40it/s]
2|4068|Loss: 0.9781466722488403:  63%|██████▎   | 4067/6470 [46:47<28:31,  1.40it/s]
2|4068|Loss: 0.9781466722488403:  63%|██████▎   | 4068/6470 [46:47<31:20,  1.28it/s]
2|4069|Loss: 0.9735172390937805:  63%|██████▎   | 4068/6470 [46:48<31:20,  1.28it/s]
2|4069|Loss: 0.9735172390937805:  63%|██████▎   | 4069/6470 [46:48<29:12,  1.37it/s]
2|4070|Loss: 0.7578508853912354:  63%|██████▎   | 4069/6470 [46:48<29:12,  1.37it/s]
2|4070|Loss: 0.7578508853912354:  63%|██████▎   | 4070/6470 [46:49<28:22,  1.41it/s]
2|4071|Loss: 0.7099018692970276:  63%|██████▎   | 4070/6470 [46:49<28:22,  1.41it/s]
2|4071|Loss: 0.7099018692970276:  63%|██████▎   | 4071/6470 [46:49<28:44,  1.39it/s]
2|4072|Loss: 0.8152828216552734:  63%|██████▎   | 4071/6470 [46:50<28:44,  1.39it/s]
2|4072|Loss: 0.8152828216552734:  63%|██████▎   | 4072/6470 [46:50<26:51,  1.49it/s]
2|4073|Loss: 0.7577477693557739:  63%|██████▎   | 4072/6470 [46:50<26:51,  1.49it/s]
2|4073|Loss: 0.7577477693557739:  63%|██████▎   | 4073/6470 [46:51<26:10,  1.53it/s]
2|4074|Loss: 0.8471364378929138:  63%|██████▎   | 4073/6470 [46:51<26:10,  1.53it/s]
2|4074|Loss: 0.8471364378929138:  63%|██████▎   | 4074/6470 [46:51<28:07,  1.42it/s]
2|4075|Loss: 0.7437689900398254:  63%|██████▎   | 4074/6470 [46:52<28:07,  1.42it/s]
2|4075|Loss: 0.7437689900398254:  63%|██████▎   | 4075/6470 [46:52<30:16,  1.32it/s]
2|4076|Loss: 0.9083463549613953:  63%|██████▎   | 4075/6470 [46:53<30:16,  1.32it/s]
2|4076|Loss: 0.9083463549613953:  63%|██████▎   | 4076/6470 [46:53<30:35,  1.30it/s]
2|4077|Loss: 0.828141450881958:  63%|██████▎   | 4076/6470 [46:53<30:35,  1.30it/s] 
2|4077|Loss: 0.828141450881958:  63%|██████▎   | 4077/6470 [46:54<30:05,  1.33it/s]
2|4078|Loss: 0.7971200942993164:  63%|██████▎   | 4077/6470 [46:54<30:05,  1.33it/s]
2|4078|Loss: 0.7971200942993164:  63%|██████▎   | 4078/6470 [46:54<29:26,  1.35it/s]
2|4079|Loss: 0.8619800209999084:  63%|██████▎   | 4078/6470 [46:55<29:26,  1.35it/s]
2|4079|Loss: 0.8619800209999084:  63%|██████▎   | 4079/6470 [46:55<28:22,  1.40it/s]
2|4080|Loss: 0.7366253137588501:  63%|██████▎   | 4079/6470 [46:55<28:22,  1.40it/s]
2|4080|Loss: 0.7366253137588501:  63%|██████▎   | 4080/6470 [46:56<28:29,  1.40it/s]
2|4081|Loss: 0.9495236277580261:  63%|██████▎   | 4080/6470 [46:56<28:29,  1.40it/s]
2|4081|Loss: 0.9495236277580261:  63%|██████▎   | 4081/6470 [46:56<26:36,  1.50it/s]
2|4082|Loss: 0.9139278531074524:  63%|██████▎   | 4081/6470 [46:57<26:36,  1.50it/s]
2|4082|Loss: 0.9139278531074524:  63%|██████▎   | 4082/6470 [46:57<24:43,  1.61it/s]
2|4083|Loss: 0.9826833605766296:  63%|██████▎   | 4082/6470 [46:57<24:43,  1.61it/s]
2|4083|Loss: 0.9826833605766296:  63%|██████▎   | 4083/6470 [46:58<25:31,  1.56it/s]
2|4084|Loss: 0.7174587845802307:  63%|██████▎   | 4083/6470 [46:58<25:31,  1.56it/s]
2|4084|Loss: 0.7174587845802307:  63%|██████▎   | 4084/6470 [46:58<26:19,  1.51it/s]
2|4085|Loss: 0.9722684025764465:  63%|██████▎   | 4084/6470 [46:59<26:19,  1.51it/s]
2|4085|Loss: 0.9722684025764465:  63%|██████▎   | 4085/6470 [46:59<27:59,  1.42it/s]
2|4086|Loss: 0.7463778853416443:  63%|██████▎   | 4085/6470 [47:00<27:59,  1.42it/s]
2|4086|Loss: 0.7463778853416443:  63%|██████▎   | 4086/6470 [47:00<28:46,  1.38it/s]
2|4087|Loss: 0.8554825186729431:  63%|██████▎   | 4086/6470 [47:00<28:46,  1.38it/s]
2|4087|Loss: 0.8554825186729431:  63%|██████▎   | 4087/6470 [47:01<30:16,  1.31it/s]
2|4088|Loss: 0.8917869925498962:  63%|██████▎   | 4087/6470 [47:01<30:16,  1.31it/s]
2|4088|Loss: 0.8917869925498962:  63%|██████▎   | 4088/6470 [47:02<30:53,  1.29it/s]
2|4089|Loss: 0.9475997686386108:  63%|██████▎   | 4088/6470 [47:02<30:53,  1.29it/s]
2|4089|Loss: 0.9475997686386108:  63%|██████▎   | 4089/6470 [47:02<30:28,  1.30it/s]
2|4090|Loss: 0.9106026887893677:  63%|██████▎   | 4089/6470 [47:03<30:28,  1.30it/s]
2|4090|Loss: 0.9106026887893677:  63%|██████▎   | 4090/6470 [47:03<30:02,  1.32it/s]
2|4091|Loss: 0.8971886038780212:  63%|██████▎   | 4090/6470 [47:03<30:02,  1.32it/s]
2|4091|Loss: 0.8971886038780212:  63%|██████▎   | 4091/6470 [47:04<31:05,  1.28it/s]
2|4092|Loss: 0.920876681804657:  63%|██████▎   | 4091/6470 [47:04<31:05,  1.28it/s] 
2|4092|Loss: 0.920876681804657:  63%|██████▎   | 4092/6470 [47:04<29:24,  1.35it/s]
2|4093|Loss: 0.8245187997817993:  63%|██████▎   | 4092/6470 [47:05<29:24,  1.35it/s]
2|4093|Loss: 0.8245187997817993:  63%|██████▎   | 4093/6470 [47:05<29:25,  1.35it/s]
2|4094|Loss: 0.733843207359314:  63%|██████▎   | 4093/6470 [47:06<29:25,  1.35it/s] 
2|4094|Loss: 0.733843207359314:  63%|██████▎   | 4094/6470 [47:06<28:37,  1.38it/s]
2|4095|Loss: 0.8346207737922668:  63%|██████▎   | 4094/6470 [47:06<28:37,  1.38it/s]
2|4095|Loss: 0.8346207737922668:  63%|██████▎   | 4095/6470 [47:06<26:07,  1.52it/s]
2|4096|Loss: 0.7906679511070251:  63%|██████▎   | 4095/6470 [47:07<26:07,  1.52it/s]
2|4096|Loss: 0.7906679511070251:  63%|██████▎   | 4096/6470 [47:07<26:04,  1.52it/s]
2|4097|Loss: 0.7689119577407837:  63%|██████▎   | 4096/6470 [47:08<26:04,  1.52it/s]
2|4097|Loss: 0.7689119577407837:  63%|██████▎   | 4097/6470 [47:08<28:08,  1.41it/s]
2|4098|Loss: 0.8663704991340637:  63%|██████▎   | 4097/6470 [47:08<28:08,  1.41it/s]
2|4098|Loss: 0.8663704991340637:  63%|██████▎   | 4098/6470 [47:09<28:46,  1.37it/s]
2|4099|Loss: 1.007108449935913:  63%|██████▎   | 4098/6470 [47:09<28:46,  1.37it/s] 
2|4099|Loss: 1.007108449935913:  63%|██████▎   | 4099/6470 [47:09<29:42,  1.33it/s]
2|4100|Loss: 0.9131401777267456:  63%|██████▎   | 4099/6470 [47:10<29:42,  1.33it/s]
2|4100|Loss: 0.9131401777267456:  63%|██████▎   | 4100/6470 [47:10<28:38,  1.38it/s]
2|4101|Loss: 0.7180013060569763:  63%|██████▎   | 4100/6470 [47:10<28:38,  1.38it/s]
2|4101|Loss: 0.7180013060569763:  63%|██████▎   | 4101/6470 [47:11<27:17,  1.45it/s]
2|4102|Loss: 0.8669298887252808:  63%|██████▎   | 4101/6470 [47:11<27:17,  1.45it/s]
2|4102|Loss: 0.8669298887252808:  63%|██████▎   | 4102/6470 [47:11<27:38,  1.43it/s]
2|4103|Loss: 0.7595502734184265:  63%|██████▎   | 4102/6470 [47:12<27:38,  1.43it/s]
2|4103|Loss: 0.7595502734184265:  63%|██████▎   | 4103/6470 [47:12<25:58,  1.52it/s]
2|4104|Loss: 0.9104162454605103:  63%|██████▎   | 4103/6470 [47:12<25:58,  1.52it/s]
2|4104|Loss: 0.9104162454605103:  63%|██████▎   | 4104/6470 [47:13<25:12,  1.56it/s]
2|4105|Loss: 0.9152651429176331:  63%|██████▎   | 4104/6470 [47:13<25:12,  1.56it/s]
2|4105|Loss: 0.9152651429176331:  63%|██████▎   | 4105/6470 [47:13<25:07,  1.57it/s]
2|4106|Loss: 1.1257998943328857:  63%|██████▎   | 4105/6470 [47:14<25:07,  1.57it/s]
2|4106|Loss: 1.1257998943328857:  63%|██████▎   | 4106/6470 [47:14<25:45,  1.53it/s]
2|4107|Loss: 0.8056558966636658:  63%|██████▎   | 4106/6470 [47:14<25:45,  1.53it/s]
2|4107|Loss: 0.8056558966636658:  63%|██████▎   | 4107/6470 [47:15<26:22,  1.49it/s]
2|4108|Loss: 0.9786725640296936:  63%|██████▎   | 4107/6470 [47:15<26:22,  1.49it/s]
2|4108|Loss: 0.9786725640296936:  63%|██████▎   | 4108/6470 [47:15<26:13,  1.50it/s]
2|4109|Loss: 0.8390726447105408:  63%|██████▎   | 4108/6470 [47:16<26:13,  1.50it/s]
2|4109|Loss: 0.8390726447105408:  64%|██████▎   | 4109/6470 [47:16<25:43,  1.53it/s]
2|4110|Loss: 0.9033676981925964:  64%|██████▎   | 4109/6470 [47:16<25:43,  1.53it/s]
2|4110|Loss: 0.9033676981925964:  64%|██████▎   | 4110/6470 [47:17<26:07,  1.51it/s]
2|4111|Loss: 0.9292628765106201:  64%|██████▎   | 4110/6470 [47:17<26:07,  1.51it/s]
2|4111|Loss: 0.9292628765106201:  64%|██████▎   | 4111/6470 [47:17<27:46,  1.42it/s]
2|4112|Loss: 0.8045982122421265:  64%|██████▎   | 4111/6470 [47:18<27:46,  1.42it/s]
2|4112|Loss: 0.8045982122421265:  64%|██████▎   | 4112/6470 [47:18<26:11,  1.50it/s]
2|4113|Loss: 0.7882092595100403:  64%|██████▎   | 4112/6470 [47:18<26:11,  1.50it/s]
2|4113|Loss: 0.7882092595100403:  64%|██████▎   | 4113/6470 [47:19<25:02,  1.57it/s]
2|4114|Loss: 0.9041368961334229:  64%|██████▎   | 4113/6470 [47:19<25:02,  1.57it/s]
2|4114|Loss: 0.9041368961334229:  64%|██████▎   | 4114/6470 [47:19<26:57,  1.46it/s]
2|4115|Loss: 0.8157703876495361:  64%|██████▎   | 4114/6470 [47:20<26:57,  1.46it/s]
2|4115|Loss: 0.8157703876495361:  64%|██████▎   | 4115/6470 [47:20<27:32,  1.43it/s]
2|4116|Loss: 0.5504224300384521:  64%|██████▎   | 4115/6470 [47:21<27:32,  1.43it/s]
2|4116|Loss: 0.5504224300384521:  64%|██████▎   | 4116/6470 [47:21<30:15,  1.30it/s]
2|4117|Loss: 0.6714086532592773:  64%|██████▎   | 4116/6470 [47:21<30:15,  1.30it/s]
2|4117|Loss: 0.6714086532592773:  64%|██████▎   | 4117/6470 [47:22<27:03,  1.45it/s]
2|4118|Loss: 0.7828073501586914:  64%|██████▎   | 4117/6470 [47:22<27:03,  1.45it/s]
2|4118|Loss: 0.7828073501586914:  64%|██████▎   | 4118/6470 [47:22<25:55,  1.51it/s]
2|4119|Loss: 0.8826755881309509:  64%|██████▎   | 4118/6470 [47:22<25:55,  1.51it/s]
2|4119|Loss: 0.8826755881309509:  64%|██████▎   | 4119/6470 [47:23<25:30,  1.54it/s]
2|4120|Loss: 0.9702355861663818:  64%|██████▎   | 4119/6470 [47:23<25:30,  1.54it/s]
2|4120|Loss: 0.9702355861663818:  64%|██████▎   | 4120/6470 [47:24<27:21,  1.43it/s]
2|4121|Loss: 0.7704010009765625:  64%|██████▎   | 4120/6470 [47:24<27:21,  1.43it/s]
2|4121|Loss: 0.7704010009765625:  64%|██████▎   | 4121/6470 [47:24<27:50,  1.41it/s]
2|4122|Loss: 0.9283466935157776:  64%|██████▎   | 4121/6470 [47:25<27:50,  1.41it/s]
2|4122|Loss: 0.9283466935157776:  64%|██████▎   | 4122/6470 [47:25<26:32,  1.47it/s]
2|4123|Loss: 0.9776224493980408:  64%|██████▎   | 4122/6470 [47:25<26:32,  1.47it/s]
2|4123|Loss: 0.9776224493980408:  64%|██████▎   | 4123/6470 [47:25<24:45,  1.58it/s]
2|4124|Loss: 0.7577072978019714:  64%|██████▎   | 4123/6470 [47:26<24:45,  1.58it/s]
2|4124|Loss: 0.7577072978019714:  64%|██████▎   | 4124/6470 [47:26<24:17,  1.61it/s]
2|4125|Loss: 0.886525571346283:  64%|██████▎   | 4124/6470 [47:26<24:17,  1.61it/s] 
2|4125|Loss: 0.886525571346283:  64%|██████▍   | 4125/6470 [47:27<26:47,  1.46it/s]
2|4126|Loss: 0.6840317845344543:  64%|██████▍   | 4125/6470 [47:27<26:47,  1.46it/s]
2|4126|Loss: 0.6840317845344543:  64%|██████▍   | 4126/6470 [47:27<25:42,  1.52it/s]
2|4127|Loss: 0.8556405305862427:  64%|██████▍   | 4126/6470 [47:28<25:42,  1.52it/s]
2|4127|Loss: 0.8556405305862427:  64%|██████▍   | 4127/6470 [47:28<24:55,  1.57it/s]
2|4128|Loss: 0.7159408330917358:  64%|██████▍   | 4127/6470 [47:28<24:55,  1.57it/s]
2|4128|Loss: 0.7159408330917358:  64%|██████▍   | 4128/6470 [47:29<24:48,  1.57it/s]
2|4129|Loss: 0.8517273664474487:  64%|██████▍   | 4128/6470 [47:29<24:48,  1.57it/s]
2|4129|Loss: 0.8517273664474487:  64%|██████▍   | 4129/6470 [47:29<25:41,  1.52it/s]
2|4130|Loss: 0.7543976902961731:  64%|██████▍   | 4129/6470 [47:30<25:41,  1.52it/s]
2|4130|Loss: 0.7543976902961731:  64%|██████▍   | 4130/6470 [47:30<26:33,  1.47it/s]
2|4131|Loss: 0.8005407452583313:  64%|██████▍   | 4130/6470 [47:30<26:33,  1.47it/s]
2|4131|Loss: 0.8005407452583313:  64%|██████▍   | 4131/6470 [47:31<25:13,  1.55it/s]
2|4132|Loss: 0.842514157295227:  64%|██████▍   | 4131/6470 [47:31<25:13,  1.55it/s] 
2|4132|Loss: 0.842514157295227:  64%|██████▍   | 4132/6470 [47:31<24:01,  1.62it/s]
2|4133|Loss: 0.7594616413116455:  64%|██████▍   | 4132/6470 [47:32<24:01,  1.62it/s]
2|4133|Loss: 0.7594616413116455:  64%|██████▍   | 4133/6470 [47:32<26:23,  1.48it/s]
2|4134|Loss: 0.9841707944869995:  64%|██████▍   | 4133/6470 [47:32<26:23,  1.48it/s]
2|4134|Loss: 0.9841707944869995:  64%|██████▍   | 4134/6470 [47:33<25:58,  1.50it/s]
2|4135|Loss: 0.7105308771133423:  64%|██████▍   | 4134/6470 [47:33<25:58,  1.50it/s]
2|4135|Loss: 0.7105308771133423:  64%|██████▍   | 4135/6470 [47:34<27:45,  1.40it/s]
2|4136|Loss: 0.7321525812149048:  64%|██████▍   | 4135/6470 [47:34<27:45,  1.40it/s]
2|4136|Loss: 0.7321525812149048:  64%|██████▍   | 4136/6470 [47:34<28:05,  1.38it/s]
2|4137|Loss: 0.946139395236969:  64%|██████▍   | 4136/6470 [47:35<28:05,  1.38it/s] 
2|4137|Loss: 0.946139395236969:  64%|██████▍   | 4137/6470 [47:35<28:20,  1.37it/s]
2|4138|Loss: 0.9252305626869202:  64%|██████▍   | 4137/6470 [47:35<28:20,  1.37it/s]
2|4138|Loss: 0.9252305626869202:  64%|██████▍   | 4138/6470 [47:36<29:20,  1.32it/s]
2|4139|Loss: 0.7409031987190247:  64%|██████▍   | 4138/6470 [47:36<29:20,  1.32it/s]
2|4139|Loss: 0.7409031987190247:  64%|██████▍   | 4139/6470 [47:37<29:28,  1.32it/s]
2|4140|Loss: 0.9418612122535706:  64%|██████▍   | 4139/6470 [47:37<29:28,  1.32it/s]
2|4140|Loss: 0.9418612122535706:  64%|██████▍   | 4140/6470 [47:37<26:55,  1.44it/s]
2|4141|Loss: 0.8878380060195923:  64%|██████▍   | 4140/6470 [47:37<26:55,  1.44it/s]
2|4141|Loss: 0.8878380060195923:  64%|██████▍   | 4141/6470 [47:38<25:15,  1.54it/s]
2|4142|Loss: 0.8043605089187622:  64%|██████▍   | 4141/6470 [47:38<25:15,  1.54it/s]
2|4142|Loss: 0.8043605089187622:  64%|██████▍   | 4142/6470 [47:38<23:14,  1.67it/s]
2|4143|Loss: 0.921455979347229:  64%|██████▍   | 4142/6470 [47:39<23:14,  1.67it/s] 
2|4143|Loss: 0.921455979347229:  64%|██████▍   | 4143/6470 [47:39<25:13,  1.54it/s]
2|4144|Loss: 0.9039309620857239:  64%|██████▍   | 4143/6470 [47:39<25:13,  1.54it/s]
2|4144|Loss: 0.9039309620857239:  64%|██████▍   | 4144/6470 [47:40<28:20,  1.37it/s]
2|4145|Loss: 0.929938554763794:  64%|██████▍   | 4144/6470 [47:40<28:20,  1.37it/s] 
2|4145|Loss: 0.929938554763794:  64%|██████▍   | 4145/6470 [47:41<33:28,  1.16it/s]
2|4146|Loss: 0.8029080033302307:  64%|██████▍   | 4145/6470 [47:42<33:28,  1.16it/s]
2|4146|Loss: 0.8029080033302307:  64%|██████▍   | 4146/6470 [47:42<33:38,  1.15it/s]
2|4147|Loss: 0.7309006452560425:  64%|██████▍   | 4146/6470 [47:42<33:38,  1.15it/s]
2|4147|Loss: 0.7309006452560425:  64%|██████▍   | 4147/6470 [47:43<32:47,  1.18it/s]
2|4148|Loss: 0.8156583309173584:  64%|██████▍   | 4147/6470 [47:43<32:47,  1.18it/s]
2|4148|Loss: 0.8156583309173584:  64%|██████▍   | 4148/6470 [47:43<30:33,  1.27it/s]
2|4149|Loss: 0.8347732424736023:  64%|██████▍   | 4148/6470 [47:44<30:33,  1.27it/s]
2|4149|Loss: 0.8347732424736023:  64%|██████▍   | 4149/6470 [47:44<28:32,  1.36it/s]
2|4150|Loss: 0.7343422174453735:  64%|██████▍   | 4149/6470 [47:44<28:32,  1.36it/s]
2|4150|Loss: 0.7343422174453735:  64%|██████▍   | 4150/6470 [47:45<29:56,  1.29it/s]
2|4151|Loss: 0.7395452260971069:  64%|██████▍   | 4150/6470 [47:45<29:56,  1.29it/s]
2|4151|Loss: 0.7395452260971069:  64%|██████▍   | 4151/6470 [47:46<29:59,  1.29it/s]
2|4152|Loss: 0.8604246973991394:  64%|██████▍   | 4151/6470 [47:46<29:59,  1.29it/s]
2|4152|Loss: 0.8604246973991394:  64%|██████▍   | 4152/6470 [47:46<29:40,  1.30it/s]
2|4153|Loss: 0.8779769539833069:  64%|██████▍   | 4152/6470 [47:47<29:40,  1.30it/s]
2|4153|Loss: 0.8779769539833069:  64%|██████▍   | 4153/6470 [47:47<29:20,  1.32it/s]
2|4154|Loss: 0.8921223878860474:  64%|██████▍   | 4153/6470 [47:48<29:20,  1.32it/s]
2|4154|Loss: 0.8921223878860474:  64%|██████▍   | 4154/6470 [47:48<29:58,  1.29it/s]
2|4155|Loss: 1.0072884559631348:  64%|██████▍   | 4154/6470 [47:48<29:58,  1.29it/s]
2|4155|Loss: 1.0072884559631348:  64%|██████▍   | 4155/6470 [47:49<30:47,  1.25it/s]
2|4156|Loss: 0.7227521538734436:  64%|██████▍   | 4155/6470 [47:49<30:47,  1.25it/s]
2|4156|Loss: 0.7227521538734436:  64%|██████▍   | 4156/6470 [47:49<28:40,  1.35it/s]
2|4157|Loss: 0.713135302066803:  64%|██████▍   | 4156/6470 [47:50<28:40,  1.35it/s] 
2|4157|Loss: 0.713135302066803:  64%|██████▍   | 4157/6470 [47:50<24:12,  1.59it/s]
2|4158|Loss: 0.9061580896377563:  64%|██████▍   | 4157/6470 [47:50<24:12,  1.59it/s]
2|4158|Loss: 0.9061580896377563:  64%|██████▍   | 4158/6470 [47:50<23:12,  1.66it/s]
2|4159|Loss: 0.7814985513687134:  64%|██████▍   | 4158/6470 [47:51<23:12,  1.66it/s]
2|4159|Loss: 0.7814985513687134:  64%|██████▍   | 4159/6470 [47:51<25:00,  1.54it/s]
2|4160|Loss: 0.8193041086196899:  64%|██████▍   | 4159/6470 [47:51<25:00,  1.54it/s]
2|4160|Loss: 0.8193041086196899:  64%|██████▍   | 4160/6470 [47:52<26:04,  1.48it/s]
2|4161|Loss: 0.7658501863479614:  64%|██████▍   | 4160/6470 [47:52<26:04,  1.48it/s]
2|4161|Loss: 0.7658501863479614:  64%|██████▍   | 4161/6470 [47:53<27:19,  1.41it/s]
2|4162|Loss: 0.9208828210830688:  64%|██████▍   | 4161/6470 [47:53<27:19,  1.41it/s]
2|4162|Loss: 0.9208828210830688:  64%|██████▍   | 4162/6470 [47:53<28:20,  1.36it/s]
2|4163|Loss: 0.8146699070930481:  64%|██████▍   | 4162/6470 [47:54<28:20,  1.36it/s]
2|4163|Loss: 0.8146699070930481:  64%|██████▍   | 4163/6470 [47:54<28:59,  1.33it/s]
2|4164|Loss: 0.762901782989502:  64%|██████▍   | 4163/6470 [47:55<28:59,  1.33it/s] 
2|4164|Loss: 0.762901782989502:  64%|██████▍   | 4164/6470 [47:55<28:48,  1.33it/s]
2|4165|Loss: 0.9132795929908752:  64%|██████▍   | 4164/6470 [47:55<28:48,  1.33it/s]
2|4165|Loss: 0.9132795929908752:  64%|██████▍   | 4165/6470 [47:56<29:06,  1.32it/s]
2|4166|Loss: 0.9166198372840881:  64%|██████▍   | 4165/6470 [47:56<29:06,  1.32it/s]
2|4166|Loss: 0.9166198372840881:  64%|██████▍   | 4166/6470 [47:57<30:44,  1.25it/s]
2|4167|Loss: 0.8505825400352478:  64%|██████▍   | 4166/6470 [47:57<30:44,  1.25it/s]
2|4167|Loss: 0.8505825400352478:  64%|██████▍   | 4167/6470 [47:57<29:34,  1.30it/s]
2|4168|Loss: 0.7993332147598267:  64%|██████▍   | 4167/6470 [47:58<29:34,  1.30it/s]
2|4168|Loss: 0.7993332147598267:  64%|██████▍   | 4168/6470 [47:58<28:40,  1.34it/s]
2|4169|Loss: 0.8782885074615479:  64%|██████▍   | 4168/6470 [47:58<28:40,  1.34it/s]
2|4169|Loss: 0.8782885074615479:  64%|██████▍   | 4169/6470 [47:59<27:06,  1.41it/s]
2|4170|Loss: 1.0635924339294434:  64%|██████▍   | 4169/6470 [47:59<27:06,  1.41it/s]
2|4170|Loss: 1.0635924339294434:  64%|██████▍   | 4170/6470 [47:59<26:10,  1.46it/s]
2|4171|Loss: 0.8533098101615906:  64%|██████▍   | 4170/6470 [48:00<26:10,  1.46it/s]
2|4171|Loss: 0.8533098101615906:  64%|██████▍   | 4171/6470 [48:00<26:34,  1.44it/s]
2|4172|Loss: 0.8091372847557068:  64%|██████▍   | 4171/6470 [48:00<26:34,  1.44it/s]
2|4172|Loss: 0.8091372847557068:  64%|██████▍   | 4172/6470 [48:01<27:12,  1.41it/s]
2|4173|Loss: 0.8332626819610596:  64%|██████▍   | 4172/6470 [48:01<27:12,  1.41it/s]
2|4173|Loss: 0.8332626819610596:  64%|██████▍   | 4173/6470 [48:01<27:39,  1.38it/s]
2|4174|Loss: 0.8484651446342468:  64%|██████▍   | 4173/6470 [48:02<27:39,  1.38it/s]
2|4174|Loss: 0.8484651446342468:  65%|██████▍   | 4174/6470 [48:02<27:45,  1.38it/s]
2|4175|Loss: 0.7962351441383362:  65%|██████▍   | 4174/6470 [48:03<27:45,  1.38it/s]
2|4175|Loss: 0.7962351441383362:  65%|██████▍   | 4175/6470 [48:03<29:08,  1.31it/s]
2|4176|Loss: 0.8202211260795593:  65%|██████▍   | 4175/6470 [48:03<29:08,  1.31it/s]
2|4176|Loss: 0.8202211260795593:  65%|██████▍   | 4176/6470 [48:04<28:07,  1.36it/s]
2|4177|Loss: 0.9230769276618958:  65%|██████▍   | 4176/6470 [48:04<28:07,  1.36it/s]
2|4177|Loss: 0.9230769276618958:  65%|██████▍   | 4177/6470 [48:04<28:19,  1.35it/s]
2|4178|Loss: 0.7646257877349854:  65%|██████▍   | 4177/6470 [48:05<28:19,  1.35it/s]
2|4178|Loss: 0.7646257877349854:  65%|██████▍   | 4178/6470 [48:05<28:32,  1.34it/s]
2|4179|Loss: 0.8296186327934265:  65%|██████▍   | 4178/6470 [48:06<28:32,  1.34it/s]
2|4179|Loss: 0.8296186327934265:  65%|██████▍   | 4179/6470 [48:06<28:56,  1.32it/s]
2|4180|Loss: 0.7410551905632019:  65%|██████▍   | 4179/6470 [48:06<28:56,  1.32it/s]
2|4180|Loss: 0.7410551905632019:  65%|██████▍   | 4180/6470 [48:07<28:32,  1.34it/s]
2|4181|Loss: 0.8198840022087097:  65%|██████▍   | 4180/6470 [48:07<28:32,  1.34it/s]
2|4181|Loss: 0.8198840022087097:  65%|██████▍   | 4181/6470 [48:07<27:15,  1.40it/s]
2|4182|Loss: 0.7410494089126587:  65%|██████▍   | 4181/6470 [48:08<27:15,  1.40it/s]
2|4182|Loss: 0.7410494089126587:  65%|██████▍   | 4182/6470 [48:08<25:31,  1.49it/s]
2|4183|Loss: 0.7490004301071167:  65%|██████▍   | 4182/6470 [48:08<25:31,  1.49it/s]
2|4183|Loss: 0.7490004301071167:  65%|██████▍   | 4183/6470 [48:09<25:53,  1.47it/s]
2|4184|Loss: 0.9611880779266357:  65%|██████▍   | 4183/6470 [48:09<25:53,  1.47it/s]
2|4184|Loss: 0.9611880779266357:  65%|██████▍   | 4184/6470 [48:09<25:43,  1.48it/s]
2|4185|Loss: 1.0193382501602173:  65%|██████▍   | 4184/6470 [48:10<25:43,  1.48it/s]
2|4185|Loss: 1.0193382501602173:  65%|██████▍   | 4185/6470 [48:10<25:54,  1.47it/s]
2|4186|Loss: 0.7332448363304138:  65%|██████▍   | 4185/6470 [48:10<25:54,  1.47it/s]
2|4186|Loss: 0.7332448363304138:  65%|██████▍   | 4186/6470 [48:11<25:34,  1.49it/s]
2|4187|Loss: 0.943338930606842:  65%|██████▍   | 4186/6470 [48:11<25:34,  1.49it/s] 
2|4187|Loss: 0.943338930606842:  65%|██████▍   | 4187/6470 [48:11<25:58,  1.46it/s]
2|4188|Loss: 0.8372840285301208:  65%|██████▍   | 4187/6470 [48:12<25:58,  1.46it/s]
2|4188|Loss: 0.8372840285301208:  65%|██████▍   | 4188/6470 [48:12<25:27,  1.49it/s]
2|4189|Loss: 0.6990777850151062:  65%|██████▍   | 4188/6470 [48:12<25:27,  1.49it/s]
2|4189|Loss: 0.6990777850151062:  65%|██████▍   | 4189/6470 [48:13<24:17,  1.57it/s]
2|4190|Loss: 0.887126088142395:  65%|██████▍   | 4189/6470 [48:13<24:17,  1.57it/s] 
2|4190|Loss: 0.887126088142395:  65%|██████▍   | 4190/6470 [48:13<23:35,  1.61it/s]
2|4191|Loss: 0.8536882996559143:  65%|██████▍   | 4190/6470 [48:14<23:35,  1.61it/s]
2|4191|Loss: 0.8536882996559143:  65%|██████▍   | 4191/6470 [48:14<25:39,  1.48it/s]
2|4192|Loss: 0.878095805644989:  65%|██████▍   | 4191/6470 [48:14<25:39,  1.48it/s] 
2|4192|Loss: 0.878095805644989:  65%|██████▍   | 4192/6470 [48:15<26:23,  1.44it/s]
2|4193|Loss: 0.7702606320381165:  65%|██████▍   | 4192/6470 [48:15<26:23,  1.44it/s]
2|4193|Loss: 0.7702606320381165:  65%|██████▍   | 4193/6470 [48:16<28:03,  1.35it/s]
2|4194|Loss: 0.9629576206207275:  65%|██████▍   | 4193/6470 [48:16<28:03,  1.35it/s]
2|4194|Loss: 0.9629576206207275:  65%|██████▍   | 4194/6470 [48:16<26:42,  1.42it/s]
2|4195|Loss: 0.8419456481933594:  65%|██████▍   | 4194/6470 [48:16<26:42,  1.42it/s]
2|4195|Loss: 0.8419456481933594:  65%|██████▍   | 4195/6470 [48:17<24:55,  1.52it/s]
2|4196|Loss: 0.8629209995269775:  65%|██████▍   | 4195/6470 [48:17<24:55,  1.52it/s]
2|4196|Loss: 0.8629209995269775:  65%|██████▍   | 4196/6470 [48:17<26:30,  1.43it/s]
2|4197|Loss: 1.1357097625732422:  65%|██████▍   | 4196/6470 [48:18<26:30,  1.43it/s]
2|4197|Loss: 1.1357097625732422:  65%|██████▍   | 4197/6470 [48:18<28:08,  1.35it/s]
2|4198|Loss: 0.9145737290382385:  65%|██████▍   | 4197/6470 [48:19<28:08,  1.35it/s]
2|4198|Loss: 0.9145737290382385:  65%|██████▍   | 4198/6470 [48:19<29:24,  1.29it/s]
2|4199|Loss: 0.8229181170463562:  65%|██████▍   | 4198/6470 [48:20<29:24,  1.29it/s]
2|4199|Loss: 0.8229181170463562:  65%|██████▍   | 4199/6470 [48:20<27:13,  1.39it/s]
2|4200|Loss: 0.8626803755760193:  65%|██████▍   | 4199/6470 [48:20<27:13,  1.39it/s]
2|4200|Loss: 0.8626803755760193:  65%|██████▍   | 4200/6470 [48:20<27:02,  1.40it/s]
2|4201|Loss: 0.7431536316871643:  65%|██████▍   | 4200/6470 [48:21<27:02,  1.40it/s]
2|4201|Loss: 0.7431536316871643:  65%|██████▍   | 4201/6470 [48:21<27:30,  1.37it/s]
2|4202|Loss: 0.9180915355682373:  65%|██████▍   | 4201/6470 [48:22<27:30,  1.37it/s]
2|4202|Loss: 0.9180915355682373:  65%|██████▍   | 4202/6470 [48:22<26:33,  1.42it/s]
2|4203|Loss: 0.8189188838005066:  65%|██████▍   | 4202/6470 [48:22<26:33,  1.42it/s]
2|4203|Loss: 0.8189188838005066:  65%|██████▍   | 4203/6470 [48:23<26:17,  1.44it/s]
2|4204|Loss: 0.8902283906936646:  65%|██████▍   | 4203/6470 [48:23<26:17,  1.44it/s]
2|4204|Loss: 0.8902283906936646:  65%|██████▍   | 4204/6470 [48:23<26:36,  1.42it/s]
2|4205|Loss: 0.8835603594779968:  65%|██████▍   | 4204/6470 [48:24<26:36,  1.42it/s]
2|4205|Loss: 0.8835603594779968:  65%|██████▍   | 4205/6470 [48:24<27:03,  1.39it/s]
2|4206|Loss: 0.9068341255187988:  65%|██████▍   | 4205/6470 [48:24<27:03,  1.39it/s]
2|4206|Loss: 0.9068341255187988:  65%|██████▌   | 4206/6470 [48:25<25:55,  1.46it/s]
2|4207|Loss: 0.8431478142738342:  65%|██████▌   | 4206/6470 [48:25<25:55,  1.46it/s]
2|4207|Loss: 0.8431478142738342:  65%|██████▌   | 4207/6470 [48:25<25:49,  1.46it/s]
2|4208|Loss: 0.9729197025299072:  65%|██████▌   | 4207/6470 [48:26<25:49,  1.46it/s]
2|4208|Loss: 0.9729197025299072:  65%|██████▌   | 4208/6470 [48:26<25:16,  1.49it/s]
2|4209|Loss: 0.8731124401092529:  65%|██████▌   | 4208/6470 [48:26<25:16,  1.49it/s]
2|4209|Loss: 0.8731124401092529:  65%|██████▌   | 4209/6470 [48:27<26:58,  1.40it/s]
2|4210|Loss: 0.778874397277832:  65%|██████▌   | 4209/6470 [48:27<26:58,  1.40it/s] 
2|4210|Loss: 0.778874397277832:  65%|██████▌   | 4210/6470 [48:27<25:18,  1.49it/s]
2|4211|Loss: 0.8192170858383179:  65%|██████▌   | 4210/6470 [48:28<25:18,  1.49it/s]
2|4211|Loss: 0.8192170858383179:  65%|██████▌   | 4211/6470 [48:28<23:08,  1.63it/s]
2|4212|Loss: 0.9247610569000244:  65%|██████▌   | 4211/6470 [48:28<23:08,  1.63it/s]
2|4212|Loss: 0.9247610569000244:  65%|██████▌   | 4212/6470 [48:28<23:31,  1.60it/s]
2|4213|Loss: 0.8477236032485962:  65%|██████▌   | 4212/6470 [48:29<23:31,  1.60it/s]
2|4213|Loss: 0.8477236032485962:  65%|██████▌   | 4213/6470 [48:29<22:40,  1.66it/s]
2|4214|Loss: 0.7257301211357117:  65%|██████▌   | 4213/6470 [48:29<22:40,  1.66it/s]
2|4214|Loss: 0.7257301211357117:  65%|██████▌   | 4214/6470 [48:30<22:14,  1.69it/s]
2|4215|Loss: 0.9177668690681458:  65%|██████▌   | 4214/6470 [48:30<22:14,  1.69it/s]
2|4215|Loss: 0.9177668690681458:  65%|██████▌   | 4215/6470 [48:30<23:11,  1.62it/s]
2|4216|Loss: 0.8192145228385925:  65%|██████▌   | 4215/6470 [48:31<23:11,  1.62it/s]
2|4216|Loss: 0.8192145228385925:  65%|██████▌   | 4216/6470 [48:31<23:24,  1.60it/s]
2|4217|Loss: 0.8702649474143982:  65%|██████▌   | 4216/6470 [48:31<23:24,  1.60it/s]
2|4217|Loss: 0.8702649474143982:  65%|██████▌   | 4217/6470 [48:32<25:32,  1.47it/s]
2|4218|Loss: 0.7868183851242065:  65%|██████▌   | 4217/6470 [48:32<25:32,  1.47it/s]
2|4218|Loss: 0.7868183851242065:  65%|██████▌   | 4218/6470 [48:33<27:23,  1.37it/s]
2|4219|Loss: 0.921630322933197:  65%|██████▌   | 4218/6470 [48:33<27:23,  1.37it/s] 
2|4219|Loss: 0.921630322933197:  65%|██████▌   | 4219/6470 [48:33<26:15,  1.43it/s]
2|4220|Loss: 0.80418461561203:  65%|██████▌   | 4219/6470 [48:34<26:15,  1.43it/s] 
2|4220|Loss: 0.80418461561203:  65%|██████▌   | 4220/6470 [48:34<26:23,  1.42it/s]
2|4221|Loss: 0.9017040133476257:  65%|██████▌   | 4220/6470 [48:34<26:23,  1.42it/s]
2|4221|Loss: 0.9017040133476257:  65%|██████▌   | 4221/6470 [48:35<28:03,  1.34it/s]
2|4222|Loss: 0.866153359413147:  65%|██████▌   | 4221/6470 [48:35<28:03,  1.34it/s] 
2|4222|Loss: 0.866153359413147:  65%|██████▌   | 4222/6470 [48:36<28:14,  1.33it/s]
2|4223|Loss: 0.8553745150566101:  65%|██████▌   | 4222/6470 [48:36<28:14,  1.33it/s]
2|4223|Loss: 0.8553745150566101:  65%|██████▌   | 4223/6470 [48:36<27:10,  1.38it/s]
2|4224|Loss: 0.87129145860672:  65%|██████▌   | 4223/6470 [48:37<27:10,  1.38it/s]  
2|4224|Loss: 0.87129145860672:  65%|██████▌   | 4224/6470 [48:37<27:52,  1.34it/s]
2|4225|Loss: 0.8869886994361877:  65%|██████▌   | 4224/6470 [48:37<27:52,  1.34it/s]
2|4225|Loss: 0.8869886994361877:  65%|██████▌   | 4225/6470 [48:38<27:07,  1.38it/s]
2|4226|Loss: 0.9759631752967834:  65%|██████▌   | 4225/6470 [48:38<27:07,  1.38it/s]
2|4226|Loss: 0.9759631752967834:  65%|██████▌   | 4226/6470 [48:38<26:47,  1.40it/s]
2|4227|Loss: 1.0051215887069702:  65%|██████▌   | 4226/6470 [48:39<26:47,  1.40it/s]
2|4227|Loss: 1.0051215887069702:  65%|██████▌   | 4227/6470 [48:39<26:54,  1.39it/s]
2|4228|Loss: 0.8586416840553284:  65%|██████▌   | 4227/6470 [48:40<26:54,  1.39it/s]
2|4228|Loss: 0.8586416840553284:  65%|██████▌   | 4228/6470 [48:40<27:36,  1.35it/s]
2|4229|Loss: 0.9462348818778992:  65%|██████▌   | 4228/6470 [48:40<27:36,  1.35it/s]
2|4229|Loss: 0.9462348818778992:  65%|██████▌   | 4229/6470 [48:41<26:40,  1.40it/s]
2|4230|Loss: 0.7237527966499329:  65%|██████▌   | 4229/6470 [48:41<26:40,  1.40it/s]
2|4230|Loss: 0.7237527966499329:  65%|██████▌   | 4230/6470 [48:41<22:57,  1.63it/s]
2|4231|Loss: 1.0159945487976074:  65%|██████▌   | 4230/6470 [48:41<22:57,  1.63it/s]
2|4231|Loss: 1.0159945487976074:  65%|██████▌   | 4231/6470 [48:42<24:15,  1.54it/s]
2|4232|Loss: 0.9008429050445557:  65%|██████▌   | 4231/6470 [48:42<24:15,  1.54it/s]
2|4232|Loss: 0.9008429050445557:  65%|██████▌   | 4232/6470 [48:43<26:46,  1.39it/s]
2|4233|Loss: 0.9595068097114563:  65%|██████▌   | 4232/6470 [48:43<26:46,  1.39it/s]
2|4233|Loss: 0.9595068097114563:  65%|██████▌   | 4233/6470 [48:43<27:22,  1.36it/s]
2|4234|Loss: 0.7708490490913391:  65%|██████▌   | 4233/6470 [48:44<27:22,  1.36it/s]
2|4234|Loss: 0.7708490490913391:  65%|██████▌   | 4234/6470 [48:44<28:28,  1.31it/s]
2|4235|Loss: 1.0709171295166016:  65%|██████▌   | 4234/6470 [48:45<28:28,  1.31it/s]
2|4235|Loss: 1.0709171295166016:  65%|██████▌   | 4235/6470 [48:45<29:33,  1.26it/s]
2|4236|Loss: 0.8106157183647156:  65%|██████▌   | 4235/6470 [48:45<29:33,  1.26it/s]
2|4236|Loss: 0.8106157183647156:  65%|██████▌   | 4236/6470 [48:46<28:23,  1.31it/s]
2|4237|Loss: 0.9677554368972778:  65%|██████▌   | 4236/6470 [48:46<28:23,  1.31it/s]
2|4237|Loss: 0.9677554368972778:  65%|██████▌   | 4237/6470 [48:46<28:28,  1.31it/s]
2|4238|Loss: 0.8621475100517273:  65%|██████▌   | 4237/6470 [48:47<28:28,  1.31it/s]
2|4238|Loss: 0.8621475100517273:  66%|██████▌   | 4238/6470 [48:47<28:00,  1.33it/s]
2|4239|Loss: 0.9289300441741943:  66%|██████▌   | 4238/6470 [48:48<28:00,  1.33it/s]
2|4239|Loss: 0.9289300441741943:  66%|██████▌   | 4239/6470 [48:48<26:52,  1.38it/s]
2|4240|Loss: 0.7766276597976685:  66%|██████▌   | 4239/6470 [48:48<26:52,  1.38it/s]
2|4240|Loss: 0.7766276597976685:  66%|██████▌   | 4240/6470 [48:48<26:00,  1.43it/s]
2|4241|Loss: 0.8726979494094849:  66%|██████▌   | 4240/6470 [48:49<26:00,  1.43it/s]
2|4241|Loss: 0.8726979494094849:  66%|██████▌   | 4241/6470 [48:49<27:15,  1.36it/s]
2|4242|Loss: 0.7607181668281555:  66%|██████▌   | 4241/6470 [48:50<27:15,  1.36it/s]
2|4242|Loss: 0.7607181668281555:  66%|██████▌   | 4242/6470 [48:50<27:44,  1.34it/s]
2|4243|Loss: 0.8233169913291931:  66%|██████▌   | 4242/6470 [48:51<27:44,  1.34it/s]
2|4243|Loss: 0.8233169913291931:  66%|██████▌   | 4243/6470 [48:51<28:44,  1.29it/s]
2|4244|Loss: 0.9941021800041199:  66%|██████▌   | 4243/6470 [48:51<28:44,  1.29it/s]
2|4244|Loss: 0.9941021800041199:  66%|██████▌   | 4244/6470 [48:52<27:35,  1.34it/s]
2|4245|Loss: 0.8211287260055542:  66%|██████▌   | 4244/6470 [48:52<27:35,  1.34it/s]
2|4245|Loss: 0.8211287260055542:  66%|██████▌   | 4245/6470 [48:52<26:08,  1.42it/s]
2|4246|Loss: 0.6758197546005249:  66%|██████▌   | 4245/6470 [48:53<26:08,  1.42it/s]
2|4246|Loss: 0.6758197546005249:  66%|██████▌   | 4246/6470 [48:53<27:21,  1.35it/s]
2|4247|Loss: 0.9672012329101562:  66%|██████▌   | 4246/6470 [48:53<27:21,  1.35it/s]
2|4247|Loss: 0.9672012329101562:  66%|██████▌   | 4247/6470 [48:54<26:54,  1.38it/s]
2|4248|Loss: 0.8448886275291443:  66%|██████▌   | 4247/6470 [48:54<26:54,  1.38it/s]
2|4248|Loss: 0.8448886275291443:  66%|██████▌   | 4248/6470 [48:54<25:42,  1.44it/s]
2|4249|Loss: 0.8052613139152527:  66%|██████▌   | 4248/6470 [48:55<25:42,  1.44it/s]
2|4249|Loss: 0.8052613139152527:  66%|██████▌   | 4249/6470 [48:55<25:35,  1.45it/s]
2|4250|Loss: 0.8877162337303162:  66%|██████▌   | 4249/6470 [48:55<25:35,  1.45it/s]
2|4250|Loss: 0.8877162337303162:  66%|██████▌   | 4250/6470 [48:56<24:21,  1.52it/s]
2|4251|Loss: 0.8093670010566711:  66%|██████▌   | 4250/6470 [48:56<24:21,  1.52it/s]
2|4251|Loss: 0.8093670010566711:  66%|██████▌   | 4251/6470 [48:56<24:57,  1.48it/s]
2|4252|Loss: 0.7667767405509949:  66%|██████▌   | 4251/6470 [48:57<24:57,  1.48it/s]
2|4252|Loss: 0.7667767405509949:  66%|██████▌   | 4252/6470 [48:57<26:12,  1.41it/s]
2|4253|Loss: 1.0826432704925537:  66%|██████▌   | 4252/6470 [48:58<26:12,  1.41it/s]
2|4253|Loss: 1.0826432704925537:  66%|██████▌   | 4253/6470 [48:58<26:18,  1.40it/s]
2|4254|Loss: 0.9122866988182068:  66%|██████▌   | 4253/6470 [48:58<26:18,  1.40it/s]
2|4254|Loss: 0.9122866988182068:  66%|██████▌   | 4254/6470 [48:59<26:10,  1.41it/s]
2|4255|Loss: 0.8732960224151611:  66%|██████▌   | 4254/6470 [48:59<26:10,  1.41it/s]
2|4255|Loss: 0.8732960224151611:  66%|██████▌   | 4255/6470 [48:59<26:21,  1.40it/s]
2|4256|Loss: 0.8480970859527588:  66%|██████▌   | 4255/6470 [49:00<26:21,  1.40it/s]
2|4256|Loss: 0.8480970859527588:  66%|██████▌   | 4256/6470 [49:00<25:49,  1.43it/s]
2|4257|Loss: 0.7852499485015869:  66%|██████▌   | 4256/6470 [49:00<25:49,  1.43it/s]
2|4257|Loss: 0.7852499485015869:  66%|██████▌   | 4257/6470 [49:01<26:05,  1.41it/s]
2|4258|Loss: 0.8006224036216736:  66%|██████▌   | 4257/6470 [49:01<26:05,  1.41it/s]
2|4258|Loss: 0.8006224036216736:  66%|██████▌   | 4258/6470 [49:01<23:56,  1.54it/s]
2|4259|Loss: 0.7778632044792175:  66%|██████▌   | 4258/6470 [49:01<23:56,  1.54it/s]
2|4259|Loss: 0.7778632044792175:  66%|██████▌   | 4259/6470 [49:02<24:08,  1.53it/s]
2|4260|Loss: 0.9628562331199646:  66%|██████▌   | 4259/6470 [49:02<24:08,  1.53it/s]
2|4260|Loss: 0.9628562331199646:  66%|██████▌   | 4260/6470 [49:02<23:11,  1.59it/s]
2|4261|Loss: 0.7779508233070374:  66%|██████▌   | 4260/6470 [49:03<23:11,  1.59it/s]
2|4261|Loss: 0.7779508233070374:  66%|██████▌   | 4261/6470 [49:03<21:53,  1.68it/s]
2|4262|Loss: 0.8166177272796631:  66%|██████▌   | 4261/6470 [49:03<21:53,  1.68it/s]
2|4262|Loss: 0.8166177272796631:  66%|██████▌   | 4262/6470 [49:04<23:44,  1.55it/s]
2|4263|Loss: 0.856718122959137:  66%|██████▌   | 4262/6470 [49:04<23:44,  1.55it/s] 
2|4263|Loss: 0.856718122959137:  66%|██████▌   | 4263/6470 [49:04<23:39,  1.55it/s]
2|4264|Loss: 0.8334959149360657:  66%|██████▌   | 4263/6470 [49:05<23:39,  1.55it/s]
2|4264|Loss: 0.8334959149360657:  66%|██████▌   | 4264/6470 [49:05<22:18,  1.65it/s]
2|4265|Loss: 0.9910542368888855:  66%|██████▌   | 4264/6470 [49:05<22:18,  1.65it/s]
2|4265|Loss: 0.9910542368888855:  66%|██████▌   | 4265/6470 [49:05<22:55,  1.60it/s]
2|4266|Loss: 0.7093727588653564:  66%|██████▌   | 4265/6470 [49:06<22:55,  1.60it/s]
2|4266|Loss: 0.7093727588653564:  66%|██████▌   | 4266/6470 [49:06<23:48,  1.54it/s]
2|4267|Loss: 0.9015549421310425:  66%|██████▌   | 4266/6470 [49:07<23:48,  1.54it/s]
2|4267|Loss: 0.9015549421310425:  66%|██████▌   | 4267/6470 [49:07<23:06,  1.59it/s]
2|4268|Loss: 1.0012363195419312:  66%|██████▌   | 4267/6470 [49:07<23:06,  1.59it/s]
2|4268|Loss: 1.0012363195419312:  66%|██████▌   | 4268/6470 [49:08<24:53,  1.47it/s]
2|4269|Loss: 0.9386441707611084:  66%|██████▌   | 4268/6470 [49:08<24:53,  1.47it/s]
2|4269|Loss: 0.9386441707611084:  66%|██████▌   | 4269/6470 [49:08<26:57,  1.36it/s]
2|4270|Loss: 0.8821594715118408:  66%|██████▌   | 4269/6470 [49:09<26:57,  1.36it/s]
2|4270|Loss: 0.8821594715118408:  66%|██████▌   | 4270/6470 [49:09<27:20,  1.34it/s]
2|4271|Loss: 0.8040248155593872:  66%|██████▌   | 4270/6470 [49:10<27:20,  1.34it/s]
2|4271|Loss: 0.8040248155593872:  66%|██████▌   | 4271/6470 [49:10<28:09,  1.30it/s]
2|4272|Loss: 0.9388394951820374:  66%|██████▌   | 4271/6470 [49:10<28:09,  1.30it/s]
2|4272|Loss: 0.9388394951820374:  66%|██████▌   | 4272/6470 [49:11<26:59,  1.36it/s]
2|4273|Loss: 0.8449297547340393:  66%|██████▌   | 4272/6470 [49:11<26:59,  1.36it/s]
2|4273|Loss: 0.8449297547340393:  66%|██████▌   | 4273/6470 [49:11<27:33,  1.33it/s]
2|4274|Loss: 0.7799452543258667:  66%|██████▌   | 4273/6470 [49:12<27:33,  1.33it/s]
2|4274|Loss: 0.7799452543258667:  66%|██████▌   | 4274/6470 [49:12<27:48,  1.32it/s]
2|4275|Loss: 0.758629560470581:  66%|██████▌   | 4274/6470 [49:13<27:48,  1.32it/s] 
2|4275|Loss: 0.758629560470581:  66%|██████▌   | 4275/6470 [49:13<25:49,  1.42it/s]
2|4276|Loss: 0.8804861307144165:  66%|██████▌   | 4275/6470 [49:13<25:49,  1.42it/s]
2|4276|Loss: 0.8804861307144165:  66%|██████▌   | 4276/6470 [49:14<25:49,  1.42it/s]
2|4277|Loss: 0.8488597869873047:  66%|██████▌   | 4276/6470 [49:14<25:49,  1.42it/s]
2|4277|Loss: 0.8488597869873047:  66%|██████▌   | 4277/6470 [49:14<25:57,  1.41it/s]
2|4278|Loss: 1.04733145236969:  66%|██████▌   | 4277/6470 [49:15<25:57,  1.41it/s]  
2|4278|Loss: 1.04733145236969:  66%|██████▌   | 4278/6470 [49:15<26:41,  1.37it/s]
2|4279|Loss: 0.7938530445098877:  66%|██████▌   | 4278/6470 [49:15<26:41,  1.37it/s]
2|4279|Loss: 0.7938530445098877:  66%|██████▌   | 4279/6470 [49:15<23:38,  1.54it/s]
2|4280|Loss: 0.7758297920227051:  66%|██████▌   | 4279/6470 [49:16<23:38,  1.54it/s]
2|4280|Loss: 0.7758297920227051:  66%|██████▌   | 4280/6470 [49:16<22:13,  1.64it/s]
2|4281|Loss: 0.9932904839515686:  66%|██████▌   | 4280/6470 [49:16<22:13,  1.64it/s]
2|4281|Loss: 0.9932904839515686:  66%|██████▌   | 4281/6470 [49:17<21:49,  1.67it/s]
2|4282|Loss: 0.8818313479423523:  66%|██████▌   | 4281/6470 [49:17<21:49,  1.67it/s]
2|4282|Loss: 0.8818313479423523:  66%|██████▌   | 4282/6470 [49:17<23:50,  1.53it/s]
2|4283|Loss: 0.8818361163139343:  66%|██████▌   | 4282/6470 [49:18<23:50,  1.53it/s]
2|4283|Loss: 0.8818361163139343:  66%|██████▌   | 4283/6470 [49:18<23:58,  1.52it/s]
2|4284|Loss: 0.7033347487449646:  66%|██████▌   | 4283/6470 [49:18<23:58,  1.52it/s]
2|4284|Loss: 0.7033347487449646:  66%|██████▌   | 4284/6470 [49:19<25:03,  1.45it/s]
2|4285|Loss: 0.8070762157440186:  66%|██████▌   | 4284/6470 [49:19<25:03,  1.45it/s]
2|4285|Loss: 0.8070762157440186:  66%|██████▌   | 4285/6470 [49:19<24:56,  1.46it/s]
2|4286|Loss: 0.9107186198234558:  66%|██████▌   | 4285/6470 [49:20<24:56,  1.46it/s]
2|4286|Loss: 0.9107186198234558:  66%|██████▌   | 4286/6470 [49:20<23:58,  1.52it/s]
2|4287|Loss: 1.0076346397399902:  66%|██████▌   | 4286/6470 [49:20<23:58,  1.52it/s]
2|4287|Loss: 1.0076346397399902:  66%|██████▋   | 4287/6470 [49:21<24:53,  1.46it/s]
2|4288|Loss: 0.89934903383255:  66%|██████▋   | 4287/6470 [49:21<24:53,  1.46it/s]  
2|4288|Loss: 0.89934903383255:  66%|██████▋   | 4288/6470 [49:22<26:39,  1.36it/s]
2|4289|Loss: 0.7247658371925354:  66%|██████▋   | 4288/6470 [49:22<26:39,  1.36it/s]
2|4289|Loss: 0.7247658371925354:  66%|██████▋   | 4289/6470 [49:22<26:10,  1.39it/s]
2|4290|Loss: 0.738001823425293:  66%|██████▋   | 4289/6470 [49:23<26:10,  1.39it/s] 
2|4290|Loss: 0.738001823425293:  66%|██████▋   | 4290/6470 [49:23<27:24,  1.33it/s]
2|4291|Loss: 0.89649498462677:  66%|██████▋   | 4290/6470 [49:24<27:24,  1.33it/s] 
2|4291|Loss: 0.89649498462677:  66%|██████▋   | 4291/6470 [49:24<28:34,  1.27it/s]
2|4292|Loss: 0.8626854419708252:  66%|██████▋   | 4291/6470 [49:24<28:34,  1.27it/s]
2|4292|Loss: 0.8626854419708252:  66%|██████▋   | 4292/6470 [49:25<27:34,  1.32it/s]
2|4293|Loss: 0.7649307250976562:  66%|██████▋   | 4292/6470 [49:25<27:34,  1.32it/s]
2|4293|Loss: 0.7649307250976562:  66%|██████▋   | 4293/6470 [49:25<27:07,  1.34it/s]
2|4294|Loss: 0.8603301644325256:  66%|██████▋   | 4293/6470 [49:26<27:07,  1.34it/s]
2|4294|Loss: 0.8603301644325256:  66%|██████▋   | 4294/6470 [49:26<26:59,  1.34it/s]
2|4295|Loss: 0.9629215002059937:  66%|██████▋   | 4294/6470 [49:27<26:59,  1.34it/s]
2|4295|Loss: 0.9629215002059937:  66%|██████▋   | 4295/6470 [49:27<25:51,  1.40it/s]
2|4296|Loss: 0.8904882669448853:  66%|██████▋   | 4295/6470 [49:27<25:51,  1.40it/s]
2|4296|Loss: 0.8904882669448853:  66%|██████▋   | 4296/6470 [49:27<24:31,  1.48it/s]
2|4297|Loss: 0.9392138719558716:  66%|██████▋   | 4296/6470 [49:28<24:31,  1.48it/s]
2|4297|Loss: 0.9392138719558716:  66%|██████▋   | 4297/6470 [49:28<24:46,  1.46it/s]
2|4298|Loss: 0.8232846260070801:  66%|██████▋   | 4297/6470 [49:29<24:46,  1.46it/s]
2|4298|Loss: 0.8232846260070801:  66%|██████▋   | 4298/6470 [49:29<25:18,  1.43it/s]
2|4299|Loss: 1.0026801824569702:  66%|██████▋   | 4298/6470 [49:29<25:18,  1.43it/s]
2|4299|Loss: 1.0026801824569702:  66%|██████▋   | 4299/6470 [49:30<24:55,  1.45it/s]
2|4300|Loss: 0.6324228048324585:  66%|██████▋   | 4299/6470 [49:30<24:55,  1.45it/s]
2|4300|Loss: 0.6324228048324585:  66%|██████▋   | 4300/6470 [49:30<24:25,  1.48it/s]
2|4301|Loss: 0.7915170192718506:  66%|██████▋   | 4300/6470 [49:31<24:25,  1.48it/s]
2|4301|Loss: 0.7915170192718506:  66%|██████▋   | 4301/6470 [49:31<24:01,  1.50it/s]
2|4302|Loss: 0.7992476224899292:  66%|██████▋   | 4301/6470 [49:31<24:01,  1.50it/s]
2|4302|Loss: 0.7992476224899292:  66%|██████▋   | 4302/6470 [49:31<22:44,  1.59it/s]
2|4303|Loss: 0.8317290544509888:  66%|██████▋   | 4302/6470 [49:32<22:44,  1.59it/s]
2|4303|Loss: 0.8317290544509888:  67%|██████▋   | 4303/6470 [49:32<22:09,  1.63it/s]
2|4304|Loss: 0.8067204356193542:  67%|██████▋   | 4303/6470 [49:32<22:09,  1.63it/s]
2|4304|Loss: 0.8067204356193542:  67%|██████▋   | 4304/6470 [49:33<24:09,  1.49it/s]
2|4305|Loss: 0.7690120339393616:  67%|██████▋   | 4304/6470 [49:33<24:09,  1.49it/s]
2|4305|Loss: 0.7690120339393616:  67%|██████▋   | 4305/6470 [49:33<25:06,  1.44it/s]
2|4306|Loss: 0.7912207841873169:  67%|██████▋   | 4305/6470 [49:34<25:06,  1.44it/s]
2|4306|Loss: 0.7912207841873169:  67%|██████▋   | 4306/6470 [49:34<25:11,  1.43it/s]
2|4307|Loss: 0.783645749092102:  67%|██████▋   | 4306/6470 [49:35<25:11,  1.43it/s] 
2|4307|Loss: 0.783645749092102:  67%|██████▋   | 4307/6470 [49:35<26:07,  1.38it/s]
2|4308|Loss: 0.8036456108093262:  67%|██████▋   | 4307/6470 [49:35<26:07,  1.38it/s]
2|4308|Loss: 0.8036456108093262:  67%|██████▋   | 4308/6470 [49:36<25:05,  1.44it/s]
2|4309|Loss: 0.9144692420959473:  67%|██████▋   | 4308/6470 [49:36<25:05,  1.44it/s]
2|4309|Loss: 0.9144692420959473:  67%|██████▋   | 4309/6470 [49:36<24:20,  1.48it/s]
2|4310|Loss: 0.9580087065696716:  67%|██████▋   | 4309/6470 [49:37<24:20,  1.48it/s]
2|4310|Loss: 0.9580087065696716:  67%|██████▋   | 4310/6470 [49:37<24:48,  1.45it/s]
2|4311|Loss: 0.6806365251541138:  67%|██████▋   | 4310/6470 [49:37<24:48,  1.45it/s]
2|4311|Loss: 0.6806365251541138:  67%|██████▋   | 4311/6470 [49:38<26:06,  1.38it/s]
2|4312|Loss: 0.8719664812088013:  67%|██████▋   | 4311/6470 [49:38<26:06,  1.38it/s]
2|4312|Loss: 0.8719664812088013:  67%|██████▋   | 4312/6470 [49:38<25:38,  1.40it/s]
2|4313|Loss: 0.8626071810722351:  67%|██████▋   | 4312/6470 [49:39<25:38,  1.40it/s]
2|4313|Loss: 0.8626071810722351:  67%|██████▋   | 4313/6470 [49:39<24:16,  1.48it/s]
2|4314|Loss: 0.8689528107643127:  67%|██████▋   | 4313/6470 [49:39<24:16,  1.48it/s]
2|4314|Loss: 0.8689528107643127:  67%|██████▋   | 4314/6470 [49:40<24:18,  1.48it/s]
2|4315|Loss: 0.9043121933937073:  67%|██████▋   | 4314/6470 [49:40<24:18,  1.48it/s]
2|4315|Loss: 0.9043121933937073:  67%|██████▋   | 4315/6470 [49:40<24:53,  1.44it/s]
2|4316|Loss: 0.8149579167366028:  67%|██████▋   | 4315/6470 [49:41<24:53,  1.44it/s]
2|4316|Loss: 0.8149579167366028:  67%|██████▋   | 4316/6470 [49:41<25:27,  1.41it/s]
2|4317|Loss: 0.6047497987747192:  67%|██████▋   | 4316/6470 [49:42<25:27,  1.41it/s]
2|4317|Loss: 0.6047497987747192:  67%|██████▋   | 4317/6470 [49:42<26:49,  1.34it/s]
2|4318|Loss: 1.0396807193756104:  67%|██████▋   | 4317/6470 [49:42<26:49,  1.34it/s]
2|4318|Loss: 1.0396807193756104:  67%|██████▋   | 4318/6470 [49:43<24:48,  1.45it/s]
2|4319|Loss: 0.8851737976074219:  67%|██████▋   | 4318/6470 [49:43<24:48,  1.45it/s]
2|4319|Loss: 0.8851737976074219:  67%|██████▋   | 4319/6470 [49:43<24:52,  1.44it/s]
2|4320|Loss: 0.9068361520767212:  67%|██████▋   | 4319/6470 [49:44<24:52,  1.44it/s]
2|4320|Loss: 0.9068361520767212:  67%|██████▋   | 4320/6470 [49:44<26:38,  1.35it/s]
2|4321|Loss: 0.9976662993431091:  67%|██████▋   | 4320/6470 [49:45<26:38,  1.35it/s]
2|4321|Loss: 0.9976662993431091:  67%|██████▋   | 4321/6470 [49:45<27:43,  1.29it/s]
2|4322|Loss: 0.8600333333015442:  67%|██████▋   | 4321/6470 [49:45<27:43,  1.29it/s]
2|4322|Loss: 0.8600333333015442:  67%|██████▋   | 4322/6470 [49:46<28:29,  1.26it/s]
2|4323|Loss: 0.9325928092002869:  67%|██████▋   | 4322/6470 [49:46<28:29,  1.26it/s]
2|4323|Loss: 0.9325928092002869:  67%|██████▋   | 4323/6470 [49:46<25:46,  1.39it/s]
2|4324|Loss: 0.9578952789306641:  67%|██████▋   | 4323/6470 [49:47<25:46,  1.39it/s]
2|4324|Loss: 0.9578952789306641:  67%|██████▋   | 4324/6470 [49:47<25:15,  1.42it/s]
2|4325|Loss: 0.9908847808837891:  67%|██████▋   | 4324/6470 [49:47<25:15,  1.42it/s]
2|4325|Loss: 0.9908847808837891:  67%|██████▋   | 4325/6470 [49:48<24:16,  1.47it/s]
2|4326|Loss: 0.8202186226844788:  67%|██████▋   | 4325/6470 [49:48<24:16,  1.47it/s]
2|4326|Loss: 0.8202186226844788:  67%|██████▋   | 4326/6470 [49:48<23:33,  1.52it/s]
2|4327|Loss: 0.879518449306488:  67%|██████▋   | 4326/6470 [49:49<23:33,  1.52it/s] 
2|4327|Loss: 0.879518449306488:  67%|██████▋   | 4327/6470 [49:49<22:34,  1.58it/s]
2|4328|Loss: 0.9245242476463318:  67%|██████▋   | 4327/6470 [49:49<22:34,  1.58it/s]
2|4328|Loss: 0.9245242476463318:  67%|██████▋   | 4328/6470 [49:49<22:14,  1.61it/s]
2|4329|Loss: 0.7236583828926086:  67%|██████▋   | 4328/6470 [49:50<22:14,  1.61it/s]
2|4329|Loss: 0.7236583828926086:  67%|██████▋   | 4329/6470 [49:50<22:21,  1.60it/s]
2|4330|Loss: 0.8189728856086731:  67%|██████▋   | 4329/6470 [49:50<22:21,  1.60it/s]
2|4330|Loss: 0.8189728856086731:  67%|██████▋   | 4330/6470 [49:51<22:26,  1.59it/s]
2|4331|Loss: 0.8681973218917847:  67%|██████▋   | 4330/6470 [49:51<22:26,  1.59it/s]
2|4331|Loss: 0.8681973218917847:  67%|██████▋   | 4331/6470 [49:51<23:29,  1.52it/s]
2|4332|Loss: 0.8428144454956055:  67%|██████▋   | 4331/6470 [49:52<23:29,  1.52it/s]
2|4332|Loss: 0.8428144454956055:  67%|██████▋   | 4332/6470 [49:52<24:29,  1.45it/s]
2|4333|Loss: 0.8201321363449097:  67%|██████▋   | 4332/6470 [49:53<24:29,  1.45it/s]
2|4333|Loss: 0.8201321363449097:  67%|██████▋   | 4333/6470 [49:53<24:44,  1.44it/s]
2|4334|Loss: 0.8535056114196777:  67%|██████▋   | 4333/6470 [49:53<24:44,  1.44it/s]
2|4334|Loss: 0.8535056114196777:  67%|██████▋   | 4334/6470 [49:54<25:54,  1.37it/s]
2|4335|Loss: 0.891367495059967:  67%|██████▋   | 4334/6470 [49:54<25:54,  1.37it/s] 
2|4335|Loss: 0.891367495059967:  67%|██████▋   | 4335/6470 [49:54<24:05,  1.48it/s]
2|4336|Loss: 0.9398266673088074:  67%|██████▋   | 4335/6470 [49:55<24:05,  1.48it/s]
2|4336|Loss: 0.9398266673088074:  67%|██████▋   | 4336/6470 [49:55<24:09,  1.47it/s]
2|4337|Loss: 0.8155031204223633:  67%|██████▋   | 4336/6470 [49:55<24:09,  1.47it/s]
2|4337|Loss: 0.8155031204223633:  67%|██████▋   | 4337/6470 [49:56<23:28,  1.51it/s]
2|4338|Loss: 0.7721008658409119:  67%|██████▋   | 4337/6470 [49:56<23:28,  1.51it/s]
2|4338|Loss: 0.7721008658409119:  67%|██████▋   | 4338/6470 [49:56<23:06,  1.54it/s]
2|4339|Loss: 0.8701122403144836:  67%|██████▋   | 4338/6470 [49:57<23:06,  1.54it/s]
2|4339|Loss: 0.8701122403144836:  67%|██████▋   | 4339/6470 [49:57<22:24,  1.59it/s]
2|4340|Loss: 0.9616529941558838:  67%|██████▋   | 4339/6470 [49:57<22:24,  1.59it/s]
2|4340|Loss: 0.9616529941558838:  67%|██████▋   | 4340/6470 [49:57<22:55,  1.55it/s]
2|4341|Loss: 0.845696747303009:  67%|██████▋   | 4340/6470 [49:58<22:55,  1.55it/s] 
2|4341|Loss: 0.845696747303009:  67%|██████▋   | 4341/6470 [49:58<21:25,  1.66it/s]
2|4342|Loss: 0.804368793964386:  67%|██████▋   | 4341/6470 [49:58<21:25,  1.66it/s]
2|4342|Loss: 0.804368793964386:  67%|██████▋   | 4342/6470 [49:59<21:50,  1.62it/s]
2|4343|Loss: 0.8695739507675171:  67%|██████▋   | 4342/6470 [49:59<21:50,  1.62it/s]
2|4343|Loss: 0.8695739507675171:  67%|██████▋   | 4343/6470 [49:59<23:01,  1.54it/s]
2|4344|Loss: 0.9721552729606628:  67%|██████▋   | 4343/6470 [50:00<23:01,  1.54it/s]
2|4344|Loss: 0.9721552729606628:  67%|██████▋   | 4344/6470 [50:00<24:04,  1.47it/s]
2|4345|Loss: 0.6745073199272156:  67%|██████▋   | 4344/6470 [50:01<24:04,  1.47it/s]
2|4345|Loss: 0.6745073199272156:  67%|██████▋   | 4345/6470 [50:01<25:31,  1.39it/s]
2|4346|Loss: 0.9206128716468811:  67%|██████▋   | 4345/6470 [50:01<25:31,  1.39it/s]
2|4346|Loss: 0.9206128716468811:  67%|██████▋   | 4346/6470 [50:02<24:55,  1.42it/s]
2|4347|Loss: 0.9080571532249451:  67%|██████▋   | 4346/6470 [50:02<24:55,  1.42it/s]
2|4347|Loss: 0.9080571532249451:  67%|██████▋   | 4347/6470 [50:02<24:01,  1.47it/s]
2|4348|Loss: 0.8136783242225647:  67%|██████▋   | 4347/6470 [50:03<24:01,  1.47it/s]
2|4348|Loss: 0.8136783242225647:  67%|██████▋   | 4348/6470 [50:03<23:37,  1.50it/s]
2|4349|Loss: 0.8564964532852173:  67%|██████▋   | 4348/6470 [50:03<23:37,  1.50it/s]
2|4349|Loss: 0.8564964532852173:  67%|██████▋   | 4349/6470 [50:04<23:19,  1.52it/s]
2|4350|Loss: 0.8875322341918945:  67%|██████▋   | 4349/6470 [50:04<23:19,  1.52it/s]
2|4350|Loss: 0.8875322341918945:  67%|██████▋   | 4350/6470 [50:04<22:57,  1.54it/s]
2|4351|Loss: 0.9611378312110901:  67%|██████▋   | 4350/6470 [50:04<22:57,  1.54it/s]
2|4351|Loss: 0.9611378312110901:  67%|██████▋   | 4351/6470 [50:05<22:46,  1.55it/s]
2|4352|Loss: 0.8500846028327942:  67%|██████▋   | 4351/6470 [50:05<22:46,  1.55it/s]
2|4352|Loss: 0.8500846028327942:  67%|██████▋   | 4352/6470 [50:05<22:33,  1.57it/s]
2|4353|Loss: 0.9600551128387451:  67%|██████▋   | 4352/6470 [50:06<22:33,  1.57it/s]
2|4353|Loss: 0.9600551128387451:  67%|██████▋   | 4353/6470 [50:06<23:04,  1.53it/s]
2|4354|Loss: 0.8377892374992371:  67%|██████▋   | 4353/6470 [50:06<23:04,  1.53it/s]
2|4354|Loss: 0.8377892374992371:  67%|██████▋   | 4354/6470 [50:07<20:52,  1.69it/s]
2|4355|Loss: 0.7044460773468018:  67%|██████▋   | 4354/6470 [50:07<20:52,  1.69it/s]
2|4355|Loss: 0.7044460773468018:  67%|██████▋   | 4355/6470 [50:07<19:21,  1.82it/s]
2|4356|Loss: 0.9898572564125061:  67%|██████▋   | 4355/6470 [50:07<19:21,  1.82it/s]
2|4356|Loss: 0.9898572564125061:  67%|██████▋   | 4356/6470 [50:08<21:29,  1.64it/s]
2|4357|Loss: 0.9131739735603333:  67%|██████▋   | 4356/6470 [50:08<21:29,  1.64it/s]
2|4357|Loss: 0.9131739735603333:  67%|██████▋   | 4357/6470 [50:08<21:39,  1.63it/s]
2|4358|Loss: 0.9462404847145081:  67%|██████▋   | 4357/6470 [50:09<21:39,  1.63it/s]
2|4358|Loss: 0.9462404847145081:  67%|██████▋   | 4358/6470 [50:09<21:15,  1.66it/s]
2|4359|Loss: 0.8338030576705933:  67%|██████▋   | 4358/6470 [50:09<21:15,  1.66it/s]
2|4359|Loss: 0.8338030576705933:  67%|██████▋   | 4359/6470 [50:10<22:58,  1.53it/s]
2|4360|Loss: 0.8739164471626282:  67%|██████▋   | 4359/6470 [50:10<22:58,  1.53it/s]
2|4360|Loss: 0.8739164471626282:  67%|██████▋   | 4360/6470 [50:10<22:40,  1.55it/s]
2|4361|Loss: 0.8192247748374939:  67%|██████▋   | 4360/6470 [50:11<22:40,  1.55it/s]
2|4361|Loss: 0.8192247748374939:  67%|██████▋   | 4361/6470 [50:11<23:23,  1.50it/s]
2|4362|Loss: 0.686370313167572:  67%|██████▋   | 4361/6470 [50:11<23:23,  1.50it/s] 
2|4362|Loss: 0.686370313167572:  67%|██████▋   | 4362/6470 [50:12<24:08,  1.46it/s]
2|4363|Loss: 0.8809319138526917:  67%|██████▋   | 4362/6470 [50:12<24:08,  1.46it/s]
2|4363|Loss: 0.8809319138526917:  67%|██████▋   | 4363/6470 [50:12<23:42,  1.48it/s]
2|4364|Loss: 0.8652063012123108:  67%|██████▋   | 4363/6470 [50:13<23:42,  1.48it/s]
2|4364|Loss: 0.8652063012123108:  67%|██████▋   | 4364/6470 [50:13<24:01,  1.46it/s]
2|4365|Loss: 0.8223142623901367:  67%|██████▋   | 4364/6470 [50:13<24:01,  1.46it/s]
2|4365|Loss: 0.8223142623901367:  67%|██████▋   | 4365/6470 [50:14<22:27,  1.56it/s]
2|4366|Loss: 0.8883034586906433:  67%|██████▋   | 4365/6470 [50:14<22:27,  1.56it/s]
2|4366|Loss: 0.8883034586906433:  67%|██████▋   | 4366/6470 [50:14<21:58,  1.60it/s]
2|4367|Loss: 0.9861201047897339:  67%|██████▋   | 4366/6470 [50:15<21:58,  1.60it/s]
2|4367|Loss: 0.9861201047897339:  67%|██████▋   | 4367/6470 [50:15<21:54,  1.60it/s]
2|4368|Loss: 0.879085123538971:  67%|██████▋   | 4367/6470 [50:15<21:54,  1.60it/s] 
2|4368|Loss: 0.879085123538971:  68%|██████▊   | 4368/6470 [50:16<22:58,  1.53it/s]
2|4369|Loss: 0.6844615340232849:  68%|██████▊   | 4368/6470 [50:16<22:58,  1.53it/s]
2|4369|Loss: 0.6844615340232849:  68%|██████▊   | 4369/6470 [50:16<23:36,  1.48it/s]
2|4370|Loss: 0.7165992259979248:  68%|██████▊   | 4369/6470 [50:17<23:36,  1.48it/s]
2|4370|Loss: 0.7165992259979248:  68%|██████▊   | 4370/6470 [50:17<23:09,  1.51it/s]
2|4371|Loss: 0.9267916083335876:  68%|██████▊   | 4370/6470 [50:17<23:09,  1.51it/s]
2|4371|Loss: 0.9267916083335876:  68%|██████▊   | 4371/6470 [50:18<24:40,  1.42it/s]
2|4372|Loss: 0.7723591327667236:  68%|██████▊   | 4371/6470 [50:18<24:40,  1.42it/s]
2|4372|Loss: 0.7723591327667236:  68%|██████▊   | 4372/6470 [50:19<25:02,  1.40it/s]
2|4373|Loss: 0.9934222102165222:  68%|██████▊   | 4372/6470 [50:19<25:02,  1.40it/s]
2|4373|Loss: 0.9934222102165222:  68%|██████▊   | 4373/6470 [50:19<23:37,  1.48it/s]
2|4374|Loss: 0.9424573183059692:  68%|██████▊   | 4373/6470 [50:19<23:37,  1.48it/s]
2|4374|Loss: 0.9424573183059692:  68%|██████▊   | 4374/6470 [50:20<23:49,  1.47it/s]
2|4375|Loss: 0.7004165053367615:  68%|██████▊   | 4374/6470 [50:20<23:49,  1.47it/s]
2|4375|Loss: 0.7004165053367615:  68%|██████▊   | 4375/6470 [50:21<24:49,  1.41it/s]
2|4376|Loss: 0.9992761611938477:  68%|██████▊   | 4375/6470 [50:21<24:49,  1.41it/s]
2|4376|Loss: 0.9992761611938477:  68%|██████▊   | 4376/6470 [50:21<24:27,  1.43it/s]
2|4377|Loss: 0.9147784113883972:  68%|██████▊   | 4376/6470 [50:22<24:27,  1.43it/s]
2|4377|Loss: 0.9147784113883972:  68%|██████▊   | 4377/6470 [50:22<23:21,  1.49it/s]
2|4378|Loss: 0.8184329271316528:  68%|██████▊   | 4377/6470 [50:22<23:21,  1.49it/s]
2|4378|Loss: 0.8184329271316528:  68%|██████▊   | 4378/6470 [50:22<22:21,  1.56it/s]
2|4379|Loss: 0.86236572265625:  68%|██████▊   | 4378/6470 [50:23<22:21,  1.56it/s]  
2|4379|Loss: 0.86236572265625:  68%|██████▊   | 4379/6470 [50:23<22:10,  1.57it/s]
2|4380|Loss: 0.8654570579528809:  68%|██████▊   | 4379/6470 [50:23<22:10,  1.57it/s]
2|4380|Loss: 0.8654570579528809:  68%|██████▊   | 4380/6470 [50:24<22:47,  1.53it/s]
2|4381|Loss: 0.7993823289871216:  68%|██████▊   | 4380/6470 [50:24<22:47,  1.53it/s]
2|4381|Loss: 0.7993823289871216:  68%|██████▊   | 4381/6470 [50:24<22:37,  1.54it/s]
2|4382|Loss: 0.8607906103134155:  68%|██████▊   | 4381/6470 [50:25<22:37,  1.54it/s]
2|4382|Loss: 0.8607906103134155:  68%|██████▊   | 4382/6470 [50:25<23:14,  1.50it/s]
2|4383|Loss: 0.7410924434661865:  68%|██████▊   | 4382/6470 [50:25<23:14,  1.50it/s]
2|4383|Loss: 0.7410924434661865:  68%|██████▊   | 4383/6470 [50:26<24:03,  1.45it/s]
2|4384|Loss: 0.9043322205543518:  68%|██████▊   | 4383/6470 [50:26<24:03,  1.45it/s]
2|4384|Loss: 0.9043322205543518:  68%|██████▊   | 4384/6470 [50:26<23:41,  1.47it/s]
2|4385|Loss: 1.0335698127746582:  68%|██████▊   | 4384/6470 [50:27<23:41,  1.47it/s]
2|4385|Loss: 1.0335698127746582:  68%|██████▊   | 4385/6470 [50:27<24:08,  1.44it/s]
2|4386|Loss: 0.8517221808433533:  68%|██████▊   | 4385/6470 [50:28<24:08,  1.44it/s]
2|4386|Loss: 0.8517221808433533:  68%|██████▊   | 4386/6470 [50:28<24:48,  1.40it/s]
2|4387|Loss: 0.9344780445098877:  68%|██████▊   | 4386/6470 [50:28<24:48,  1.40it/s]
2|4387|Loss: 0.9344780445098877:  68%|██████▊   | 4387/6470 [50:29<25:53,  1.34it/s]
2|4388|Loss: 0.7426632046699524:  68%|██████▊   | 4387/6470 [50:29<25:53,  1.34it/s]
2|4388|Loss: 0.7426632046699524:  68%|██████▊   | 4388/6470 [50:30<27:30,  1.26it/s]
2|4389|Loss: 0.7920680642127991:  68%|██████▊   | 4388/6470 [50:30<27:30,  1.26it/s]
2|4389|Loss: 0.7920680642127991:  68%|██████▊   | 4389/6470 [50:30<27:19,  1.27it/s]
2|4390|Loss: 0.7973054051399231:  68%|██████▊   | 4389/6470 [50:31<27:19,  1.27it/s]
2|4390|Loss: 0.7973054051399231:  68%|██████▊   | 4390/6470 [50:31<26:46,  1.29it/s]
2|4391|Loss: 0.8009483218193054:  68%|██████▊   | 4390/6470 [50:32<26:46,  1.29it/s]
2|4391|Loss: 0.8009483218193054:  68%|██████▊   | 4391/6470 [50:32<26:22,  1.31it/s]
2|4392|Loss: 0.7679688930511475:  68%|██████▊   | 4391/6470 [50:32<26:22,  1.31it/s]
2|4392|Loss: 0.7679688930511475:  68%|██████▊   | 4392/6470 [50:33<25:06,  1.38it/s]
2|4393|Loss: 0.9172394275665283:  68%|██████▊   | 4392/6470 [50:33<25:06,  1.38it/s]
2|4393|Loss: 0.9172394275665283:  68%|██████▊   | 4393/6470 [50:33<23:38,  1.46it/s]
2|4394|Loss: 0.8311446905136108:  68%|██████▊   | 4393/6470 [50:33<23:38,  1.46it/s]
2|4394|Loss: 0.8311446905136108:  68%|██████▊   | 4394/6470 [50:34<22:13,  1.56it/s]
2|4395|Loss: 0.9875631928443909:  68%|██████▊   | 4394/6470 [50:34<22:13,  1.56it/s]
2|4395|Loss: 0.9875631928443909:  68%|██████▊   | 4395/6470 [50:34<20:26,  1.69it/s]
2|4396|Loss: 0.7266527414321899:  68%|██████▊   | 4395/6470 [50:35<20:26,  1.69it/s]
2|4396|Loss: 0.7266527414321899:  68%|██████▊   | 4396/6470 [50:35<21:48,  1.58it/s]
2|4397|Loss: 0.9913964867591858:  68%|██████▊   | 4396/6470 [50:35<21:48,  1.58it/s]
2|4397|Loss: 0.9913964867591858:  68%|██████▊   | 4397/6470 [50:36<23:09,  1.49it/s]
2|4398|Loss: 0.7126349806785583:  68%|██████▊   | 4397/6470 [50:36<23:09,  1.49it/s]
2|4398|Loss: 0.7126349806785583:  68%|██████▊   | 4398/6470 [50:36<24:22,  1.42it/s]
2|4399|Loss: 0.633773684501648:  68%|██████▊   | 4398/6470 [50:37<24:22,  1.42it/s] 
2|4399|Loss: 0.633773684501648:  68%|██████▊   | 4399/6470 [50:37<21:44,  1.59it/s]
2|4400|Loss: 0.8566428422927856:  68%|██████▊   | 4399/6470 [50:37<21:44,  1.59it/s]
2|4400|Loss: 0.8566428422927856:  68%|██████▊   | 4400/6470 [50:37<21:11,  1.63it/s]
2|4401|Loss: 0.8793553709983826:  68%|██████▊   | 4400/6470 [50:38<21:11,  1.63it/s]
2|4401|Loss: 0.8793553709983826:  68%|██████▊   | 4401/6470 [50:38<20:49,  1.66it/s]
2|4402|Loss: 0.9489100575447083:  68%|██████▊   | 4401/6470 [50:38<20:49,  1.66it/s]
2|4402|Loss: 0.9489100575447083:  68%|██████▊   | 4402/6470 [50:39<22:46,  1.51it/s]
2|4403|Loss: 0.9672884941101074:  68%|██████▊   | 4402/6470 [50:39<22:46,  1.51it/s]
2|4403|Loss: 0.9672884941101074:  68%|██████▊   | 4403/6470 [50:40<23:06,  1.49it/s]
2|4404|Loss: 0.8715488910675049:  68%|██████▊   | 4403/6470 [50:40<23:06,  1.49it/s]
2|4404|Loss: 0.8715488910675049:  68%|██████▊   | 4404/6470 [50:40<22:37,  1.52it/s]
2|4405|Loss: 0.7114809155464172:  68%|██████▊   | 4404/6470 [50:41<22:37,  1.52it/s]
2|4405|Loss: 0.7114809155464172:  68%|██████▊   | 4405/6470 [50:41<25:04,  1.37it/s]
2|4406|Loss: 0.8081970810890198:  68%|██████▊   | 4405/6470 [50:42<25:04,  1.37it/s]
2|4406|Loss: 0.8081970810890198:  68%|██████▊   | 4406/6470 [50:42<25:23,  1.35it/s]
2|4407|Loss: 0.8768417835235596:  68%|██████▊   | 4406/6470 [50:42<25:23,  1.35it/s]
2|4407|Loss: 0.8768417835235596:  68%|██████▊   | 4407/6470 [50:42<23:25,  1.47it/s]
2|4408|Loss: 1.0135183334350586:  68%|██████▊   | 4407/6470 [50:43<23:25,  1.47it/s]
2|4408|Loss: 1.0135183334350586:  68%|██████▊   | 4408/6470 [50:43<23:29,  1.46it/s]
2|4409|Loss: 0.9041262865066528:  68%|██████▊   | 4408/6470 [50:43<23:29,  1.46it/s]
2|4409|Loss: 0.9041262865066528:  68%|██████▊   | 4409/6470 [50:44<22:47,  1.51it/s]
2|4410|Loss: 0.8903584480285645:  68%|██████▊   | 4409/6470 [50:44<22:47,  1.51it/s]
2|4410|Loss: 0.8903584480285645:  68%|██████▊   | 4410/6470 [50:45<24:21,  1.41it/s]
2|4411|Loss: 0.9236260056495667:  68%|██████▊   | 4410/6470 [50:45<24:21,  1.41it/s]
2|4411|Loss: 0.9236260056495667:  68%|██████▊   | 4411/6470 [50:45<24:59,  1.37it/s]
2|4412|Loss: 0.7379237413406372:  68%|██████▊   | 4411/6470 [50:46<24:59,  1.37it/s]
2|4412|Loss: 0.7379237413406372:  68%|██████▊   | 4412/6470 [50:46<24:47,  1.38it/s]
2|4413|Loss: 0.8970481157302856:  68%|██████▊   | 4412/6470 [50:46<24:47,  1.38it/s]
2|4413|Loss: 0.8970481157302856:  68%|██████▊   | 4413/6470 [50:47<24:58,  1.37it/s]
2|4414|Loss: 1.0317171812057495:  68%|██████▊   | 4413/6470 [50:47<24:58,  1.37it/s]
2|4414|Loss: 1.0317171812057495:  68%|██████▊   | 4414/6470 [50:47<24:07,  1.42it/s]
2|4415|Loss: 0.8732537627220154:  68%|██████▊   | 4414/6470 [50:48<24:07,  1.42it/s]
2|4415|Loss: 0.8732537627220154:  68%|██████▊   | 4415/6470 [50:48<25:09,  1.36it/s]
2|4416|Loss: 0.7800031900405884:  68%|██████▊   | 4415/6470 [50:49<25:09,  1.36it/s]
2|4416|Loss: 0.7800031900405884:  68%|██████▊   | 4416/6470 [50:49<24:13,  1.41it/s]
2|4417|Loss: 0.7975648045539856:  68%|██████▊   | 4416/6470 [50:49<24:13,  1.41it/s]
2|4417|Loss: 0.7975648045539856:  68%|██████▊   | 4417/6470 [50:50<24:08,  1.42it/s]
2|4418|Loss: 0.9339547157287598:  68%|██████▊   | 4417/6470 [50:50<24:08,  1.42it/s]
2|4418|Loss: 0.9339547157287598:  68%|██████▊   | 4418/6470 [50:50<23:09,  1.48it/s]
2|4419|Loss: 0.7808923125267029:  68%|██████▊   | 4418/6470 [50:51<23:09,  1.48it/s]
2|4419|Loss: 0.7808923125267029:  68%|██████▊   | 4419/6470 [50:51<23:28,  1.46it/s]
2|4420|Loss: 0.8640473484992981:  68%|██████▊   | 4419/6470 [50:51<23:28,  1.46it/s]
2|4420|Loss: 0.8640473484992981:  68%|██████▊   | 4420/6470 [50:52<24:08,  1.42it/s]
2|4421|Loss: 0.9684392213821411:  68%|██████▊   | 4420/6470 [50:52<24:08,  1.42it/s]
2|4421|Loss: 0.9684392213821411:  68%|██████▊   | 4421/6470 [50:52<25:02,  1.36it/s]
2|4422|Loss: 0.8410716652870178:  68%|██████▊   | 4421/6470 [50:53<25:02,  1.36it/s]
2|4422|Loss: 0.8410716652870178:  68%|██████▊   | 4422/6470 [50:53<23:54,  1.43it/s]
2|4423|Loss: 0.8728893995285034:  68%|██████▊   | 4422/6470 [50:53<23:54,  1.43it/s]
2|4423|Loss: 0.8728893995285034:  68%|██████▊   | 4423/6470 [50:54<24:57,  1.37it/s]
2|4424|Loss: 0.9611959457397461:  68%|██████▊   | 4423/6470 [50:54<24:57,  1.37it/s]
2|4424|Loss: 0.9611959457397461:  68%|██████▊   | 4424/6470 [50:54<23:58,  1.42it/s]
2|4425|Loss: 0.7432056069374084:  68%|██████▊   | 4424/6470 [50:55<23:58,  1.42it/s]
2|4425|Loss: 0.7432056069374084:  68%|██████▊   | 4425/6470 [50:55<23:03,  1.48it/s]
2|4426|Loss: 0.8897112011909485:  68%|██████▊   | 4425/6470 [50:55<23:03,  1.48it/s]
2|4426|Loss: 0.8897112011909485:  68%|██████▊   | 4426/6470 [50:56<22:39,  1.50it/s]
2|4427|Loss: 0.819740355014801:  68%|██████▊   | 4426/6470 [50:56<22:39,  1.50it/s] 
2|4427|Loss: 0.819740355014801:  68%|██████▊   | 4427/6470 [50:56<22:10,  1.54it/s]
2|4428|Loss: 0.9128469228744507:  68%|██████▊   | 4427/6470 [50:57<22:10,  1.54it/s]
2|4428|Loss: 0.9128469228744507:  68%|██████▊   | 4428/6470 [50:57<23:58,  1.42it/s]
2|4429|Loss: 0.9201298356056213:  68%|██████▊   | 4428/6470 [50:58<23:58,  1.42it/s]
2|4429|Loss: 0.9201298356056213:  68%|██████▊   | 4429/6470 [50:58<24:38,  1.38it/s]
2|4430|Loss: 0.9816423654556274:  68%|██████▊   | 4429/6470 [50:58<24:38,  1.38it/s]
2|4430|Loss: 0.9816423654556274:  68%|██████▊   | 4430/6470 [50:59<24:42,  1.38it/s]
2|4431|Loss: 0.9014538526535034:  68%|██████▊   | 4430/6470 [50:59<24:42,  1.38it/s]
2|4431|Loss: 0.9014538526535034:  68%|██████▊   | 4431/6470 [50:59<23:50,  1.43it/s]
2|4432|Loss: 0.965502917766571:  68%|██████▊   | 4431/6470 [51:00<23:50,  1.43it/s] 
2|4432|Loss: 0.965502917766571:  69%|██████▊   | 4432/6470 [51:00<22:14,  1.53it/s]
2|4433|Loss: 0.907964825630188:  69%|██████▊   | 4432/6470 [51:00<22:14,  1.53it/s]
2|4433|Loss: 0.907964825630188:  69%|██████▊   | 4433/6470 [51:00<21:37,  1.57it/s]
2|4434|Loss: 0.8504290580749512:  69%|██████▊   | 4433/6470 [51:01<21:37,  1.57it/s]
2|4434|Loss: 0.8504290580749512:  69%|██████▊   | 4434/6470 [51:01<21:38,  1.57it/s]
2|4435|Loss: 0.8495667576789856:  69%|██████▊   | 4434/6470 [51:02<21:38,  1.57it/s]
2|4435|Loss: 0.8495667576789856:  69%|██████▊   | 4435/6470 [51:02<23:10,  1.46it/s]
2|4436|Loss: 0.9289097189903259:  69%|██████▊   | 4435/6470 [51:02<23:10,  1.46it/s]
2|4436|Loss: 0.9289097189903259:  69%|██████▊   | 4436/6470 [51:03<23:38,  1.43it/s]
2|4437|Loss: 0.7751646637916565:  69%|██████▊   | 4436/6470 [51:03<23:38,  1.43it/s]
2|4437|Loss: 0.7751646637916565:  69%|██████▊   | 4437/6470 [51:03<24:52,  1.36it/s]
2|4438|Loss: 0.8281400799751282:  69%|██████▊   | 4437/6470 [51:04<24:52,  1.36it/s]
2|4438|Loss: 0.8281400799751282:  69%|██████▊   | 4438/6470 [51:04<25:13,  1.34it/s]
2|4439|Loss: 0.7580074071884155:  69%|██████▊   | 4438/6470 [51:05<25:13,  1.34it/s]
2|4439|Loss: 0.7580074071884155:  69%|██████▊   | 4439/6470 [51:05<24:12,  1.40it/s]
2|4440|Loss: 0.785244345664978:  69%|██████▊   | 4439/6470 [51:05<24:12,  1.40it/s] 
2|4440|Loss: 0.785244345664978:  69%|██████▊   | 4440/6470 [51:05<22:52,  1.48it/s]
2|4441|Loss: 0.9707396626472473:  69%|██████▊   | 4440/6470 [51:06<22:52,  1.48it/s]
2|4441|Loss: 0.9707396626472473:  69%|██████▊   | 4441/6470 [51:06<24:42,  1.37it/s]
2|4442|Loss: 0.8831430673599243:  69%|██████▊   | 4441/6470 [51:07<24:42,  1.37it/s]
2|4442|Loss: 0.8831430673599243:  69%|██████▊   | 4442/6470 [51:07<25:17,  1.34it/s]
2|4443|Loss: 0.8592824339866638:  69%|██████▊   | 4442/6470 [51:07<25:17,  1.34it/s]
2|4443|Loss: 0.8592824339866638:  69%|██████▊   | 4443/6470 [51:08<24:28,  1.38it/s]
2|4444|Loss: 0.9078614115715027:  69%|██████▊   | 4443/6470 [51:08<24:28,  1.38it/s]
2|4444|Loss: 0.9078614115715027:  69%|██████▊   | 4444/6470 [51:09<25:18,  1.33it/s]
2|4445|Loss: 0.7725878357887268:  69%|██████▊   | 4444/6470 [51:09<25:18,  1.33it/s]
2|4445|Loss: 0.7725878357887268:  69%|██████▊   | 4445/6470 [51:09<22:35,  1.49it/s]
2|4446|Loss: 0.90739506483078:  69%|██████▊   | 4445/6470 [51:09<22:35,  1.49it/s]  
2|4446|Loss: 0.90739506483078:  69%|██████▊   | 4446/6470 [51:10<21:41,  1.55it/s]
2|4447|Loss: 0.9124895334243774:  69%|██████▊   | 4446/6470 [51:10<21:41,  1.55it/s]
2|4447|Loss: 0.9124895334243774:  69%|██████▊   | 4447/6470 [51:10<22:32,  1.50it/s]
2|4448|Loss: 0.6974819302558899:  69%|██████▊   | 4447/6470 [51:11<22:32,  1.50it/s]
2|4448|Loss: 0.6974819302558899:  69%|██████▊   | 4448/6470 [51:11<23:01,  1.46it/s]
2|4449|Loss: 0.8527235984802246:  69%|██████▊   | 4448/6470 [51:11<23:01,  1.46it/s]
2|4449|Loss: 0.8527235984802246:  69%|██████▉   | 4449/6470 [51:12<23:31,  1.43it/s]
2|4450|Loss: 0.958389163017273:  69%|██████▉   | 4449/6470 [51:12<23:31,  1.43it/s] 
2|4450|Loss: 0.958389163017273:  69%|██████▉   | 4450/6470 [51:13<24:59,  1.35it/s]
2|4451|Loss: 0.9256225228309631:  69%|██████▉   | 4450/6470 [51:13<24:59,  1.35it/s]
2|4451|Loss: 0.9256225228309631:  69%|██████▉   | 4451/6470 [51:13<23:37,  1.42it/s]
2|4452|Loss: 0.9027245044708252:  69%|██████▉   | 4451/6470 [51:14<23:37,  1.42it/s]
2|4452|Loss: 0.9027245044708252:  69%|██████▉   | 4452/6470 [51:14<22:36,  1.49it/s]
2|4453|Loss: 0.8020544052124023:  69%|██████▉   | 4452/6470 [51:14<22:36,  1.49it/s]
2|4453|Loss: 0.8020544052124023:  69%|██████▉   | 4453/6470 [51:14<20:42,  1.62it/s]
2|4454|Loss: 0.8290128707885742:  69%|██████▉   | 4453/6470 [51:15<20:42,  1.62it/s]
2|4454|Loss: 0.8290128707885742:  69%|██████▉   | 4454/6470 [51:15<20:05,  1.67it/s]
2|4455|Loss: 0.6712085008621216:  69%|██████▉   | 4454/6470 [51:15<20:05,  1.67it/s]
2|4455|Loss: 0.6712085008621216:  69%|██████▉   | 4455/6470 [51:15<17:52,  1.88it/s]
2|4456|Loss: 0.930293619632721:  69%|██████▉   | 4455/6470 [51:16<17:52,  1.88it/s] 
2|4456|Loss: 0.930293619632721:  69%|██████▉   | 4456/6470 [51:16<18:02,  1.86it/s]
2|4457|Loss: 1.0053002834320068:  69%|██████▉   | 4456/6470 [51:16<18:02,  1.86it/s]
2|4457|Loss: 1.0053002834320068:  69%|██████▉   | 4457/6470 [51:16<18:50,  1.78it/s]
2|4458|Loss: 0.7872424125671387:  69%|██████▉   | 4457/6470 [51:17<18:50,  1.78it/s]
2|4458|Loss: 0.7872424125671387:  69%|██████▉   | 4458/6470 [51:17<21:08,  1.59it/s]
2|4459|Loss: 0.8218035697937012:  69%|██████▉   | 4458/6470 [51:18<21:08,  1.59it/s]
2|4459|Loss: 0.8218035697937012:  69%|██████▉   | 4459/6470 [51:18<22:24,  1.50it/s]
2|4460|Loss: 0.7921938300132751:  69%|██████▉   | 4459/6470 [51:18<22:24,  1.50it/s]
2|4460|Loss: 0.7921938300132751:  69%|██████▉   | 4460/6470 [51:19<24:34,  1.36it/s]
2|4461|Loss: 0.8445510268211365:  69%|██████▉   | 4460/6470 [51:19<24:34,  1.36it/s]
2|4461|Loss: 0.8445510268211365:  69%|██████▉   | 4461/6470 [51:19<23:19,  1.44it/s]
2|4462|Loss: 0.8578122854232788:  69%|██████▉   | 4461/6470 [51:20<23:19,  1.44it/s]
2|4462|Loss: 0.8578122854232788:  69%|██████▉   | 4462/6470 [51:20<21:46,  1.54it/s]
2|4463|Loss: 0.9565538763999939:  69%|██████▉   | 4462/6470 [51:20<21:46,  1.54it/s]
2|4463|Loss: 0.9565538763999939:  69%|██████▉   | 4463/6470 [51:21<23:39,  1.41it/s]
2|4464|Loss: 0.9535881876945496:  69%|██████▉   | 4463/6470 [51:21<23:39,  1.41it/s]
2|4464|Loss: 0.9535881876945496:  69%|██████▉   | 4464/6470 [51:21<22:47,  1.47it/s]
2|4465|Loss: 0.9515863656997681:  69%|██████▉   | 4464/6470 [51:22<22:47,  1.47it/s]
2|4465|Loss: 0.9515863656997681:  69%|██████▉   | 4465/6470 [51:22<21:07,  1.58it/s]
2|4466|Loss: 0.7830885648727417:  69%|██████▉   | 4465/6470 [51:22<21:07,  1.58it/s]
2|4466|Loss: 0.7830885648727417:  69%|██████▉   | 4466/6470 [51:23<21:23,  1.56it/s]
2|4467|Loss: 0.916277289390564:  69%|██████▉   | 4466/6470 [51:23<21:23,  1.56it/s] 
2|4467|Loss: 0.916277289390564:  69%|██████▉   | 4467/6470 [51:24<23:41,  1.41it/s]
2|4468|Loss: 0.807817280292511:  69%|██████▉   | 4467/6470 [51:24<23:41,  1.41it/s]
2|4468|Loss: 0.807817280292511:  69%|██████▉   | 4468/6470 [51:24<24:19,  1.37it/s]
2|4469|Loss: 0.7995774149894714:  69%|██████▉   | 4468/6470 [51:25<24:19,  1.37it/s]
2|4469|Loss: 0.7995774149894714:  69%|██████▉   | 4469/6470 [51:25<24:26,  1.36it/s]
2|4470|Loss: 0.8075532913208008:  69%|██████▉   | 4469/6470 [51:25<24:26,  1.36it/s]
2|4470|Loss: 0.8075532913208008:  69%|██████▉   | 4470/6470 [51:26<24:24,  1.37it/s]
2|4471|Loss: 0.8155843615531921:  69%|██████▉   | 4470/6470 [51:26<24:24,  1.37it/s]
2|4471|Loss: 0.8155843615531921:  69%|██████▉   | 4471/6470 [51:27<24:24,  1.37it/s]
2|4472|Loss: 0.7179545760154724:  69%|██████▉   | 4471/6470 [51:27<24:24,  1.37it/s]
2|4472|Loss: 0.7179545760154724:  69%|██████▉   | 4472/6470 [51:27<24:50,  1.34it/s]
2|4473|Loss: 0.7537112832069397:  69%|██████▉   | 4472/6470 [51:28<24:50,  1.34it/s]
2|4473|Loss: 0.7537112832069397:  69%|██████▉   | 4473/6470 [51:28<24:46,  1.34it/s]
2|4474|Loss: 0.72496098279953:  69%|██████▉   | 4473/6470 [51:28<24:46,  1.34it/s]  
2|4474|Loss: 0.72496098279953:  69%|██████▉   | 4474/6470 [51:29<24:23,  1.36it/s]
2|4475|Loss: 0.8086525201797485:  69%|██████▉   | 4474/6470 [51:29<24:23,  1.36it/s]
2|4475|Loss: 0.8086525201797485:  69%|██████▉   | 4475/6470 [51:29<24:17,  1.37it/s]
2|4476|Loss: 0.8154591917991638:  69%|██████▉   | 4475/6470 [51:30<24:17,  1.37it/s]
2|4476|Loss: 0.8154591917991638:  69%|██████▉   | 4476/6470 [51:30<24:18,  1.37it/s]
2|4477|Loss: 0.8400816321372986:  69%|██████▉   | 4476/6470 [51:31<24:18,  1.37it/s]
2|4477|Loss: 0.8400816321372986:  69%|██████▉   | 4477/6470 [51:31<25:03,  1.33it/s]
2|4478|Loss: 0.8183208107948303:  69%|██████▉   | 4477/6470 [51:31<25:03,  1.33it/s]
2|4478|Loss: 0.8183208107948303:  69%|██████▉   | 4478/6470 [51:32<26:12,  1.27it/s]
2|4479|Loss: 1.002514362335205:  69%|██████▉   | 4478/6470 [51:32<26:12,  1.27it/s] 
2|4479|Loss: 1.002514362335205:  69%|██████▉   | 4479/6470 [51:33<25:59,  1.28it/s]
2|4480|Loss: 0.9579640030860901:  69%|██████▉   | 4479/6470 [51:33<25:59,  1.28it/s]
2|4480|Loss: 0.9579640030860901:  69%|██████▉   | 4480/6470 [51:33<24:31,  1.35it/s]
2|4481|Loss: 0.8020698428153992:  69%|██████▉   | 4480/6470 [51:34<24:31,  1.35it/s]
2|4481|Loss: 0.8020698428153992:  69%|██████▉   | 4481/6470 [51:34<24:06,  1.38it/s]
2|4482|Loss: 0.6389017701148987:  69%|██████▉   | 4481/6470 [51:34<24:06,  1.38it/s]
2|4482|Loss: 0.6389017701148987:  69%|██████▉   | 4482/6470 [51:35<22:20,  1.48it/s]
2|4483|Loss: 1.0370010137557983:  69%|██████▉   | 4482/6470 [51:35<22:20,  1.48it/s]
2|4483|Loss: 1.0370010137557983:  69%|██████▉   | 4483/6470 [51:35<21:30,  1.54it/s]
2|4484|Loss: 0.823215126991272:  69%|██████▉   | 4483/6470 [51:36<21:30,  1.54it/s] 
2|4484|Loss: 0.823215126991272:  69%|██████▉   | 4484/6470 [51:36<23:34,  1.40it/s]
2|4485|Loss: 0.836503803730011:  69%|██████▉   | 4484/6470 [51:36<23:34,  1.40it/s]
2|4485|Loss: 0.836503803730011:  69%|██████▉   | 4485/6470 [51:37<24:42,  1.34it/s]
2|4486|Loss: 0.8233226537704468:  69%|██████▉   | 4485/6470 [51:37<24:42,  1.34it/s]
2|4486|Loss: 0.8233226537704468:  69%|██████▉   | 4486/6470 [51:38<24:24,  1.35it/s]
2|4487|Loss: 0.7829582095146179:  69%|██████▉   | 4486/6470 [51:38<24:24,  1.35it/s]
2|4487|Loss: 0.7829582095146179:  69%|██████▉   | 4487/6470 [51:38<22:04,  1.50it/s]
2|4488|Loss: 0.8571767210960388:  69%|██████▉   | 4487/6470 [51:38<22:04,  1.50it/s]
2|4488|Loss: 0.8571767210960388:  69%|██████▉   | 4488/6470 [51:39<21:48,  1.51it/s]
2|4489|Loss: 0.8260560631752014:  69%|██████▉   | 4488/6470 [51:39<21:48,  1.51it/s]
2|4489|Loss: 0.8260560631752014:  69%|██████▉   | 4489/6470 [51:40<23:39,  1.40it/s]
2|4490|Loss: 0.898033082485199:  69%|██████▉   | 4489/6470 [51:40<23:39,  1.40it/s] 
2|4490|Loss: 0.898033082485199:  69%|██████▉   | 4490/6470 [51:40<24:07,  1.37it/s]
2|4491|Loss: 0.8741281628608704:  69%|██████▉   | 4490/6470 [51:41<24:07,  1.37it/s]
2|4491|Loss: 0.8741281628608704:  69%|██████▉   | 4491/6470 [51:41<24:08,  1.37it/s]
2|4492|Loss: 0.8339381814002991:  69%|██████▉   | 4491/6470 [51:41<24:08,  1.37it/s]
2|4492|Loss: 0.8339381814002991:  69%|██████▉   | 4492/6470 [51:42<24:34,  1.34it/s]
2|4493|Loss: 0.8940476775169373:  69%|██████▉   | 4492/6470 [51:42<24:34,  1.34it/s]
2|4493|Loss: 0.8940476775169373:  69%|██████▉   | 4493/6470 [51:43<27:15,  1.21it/s]
2|4494|Loss: 0.9592511653900146:  69%|██████▉   | 4493/6470 [51:43<27:15,  1.21it/s]
2|4494|Loss: 0.9592511653900146:  69%|██████▉   | 4494/6470 [51:44<27:45,  1.19it/s]
2|4495|Loss: 0.9651566743850708:  69%|██████▉   | 4494/6470 [51:44<27:45,  1.19it/s]
2|4495|Loss: 0.9651566743850708:  69%|██████▉   | 4495/6470 [51:44<27:01,  1.22it/s]
2|4496|Loss: 0.9223619699478149:  69%|██████▉   | 4495/6470 [51:45<27:01,  1.22it/s]
2|4496|Loss: 0.9223619699478149:  69%|██████▉   | 4496/6470 [51:45<27:08,  1.21it/s]
2|4497|Loss: 0.9621779322624207:  69%|██████▉   | 4496/6470 [51:46<27:08,  1.21it/s]
2|4497|Loss: 0.9621779322624207:  70%|██████▉   | 4497/6470 [51:46<25:36,  1.28it/s]
2|4498|Loss: 0.9114910960197449:  70%|██████▉   | 4497/6470 [51:46<25:36,  1.28it/s]
2|4498|Loss: 0.9114910960197449:  70%|██████▉   | 4498/6470 [51:47<23:09,  1.42it/s]
2|4499|Loss: 0.8949989676475525:  70%|██████▉   | 4498/6470 [51:47<23:09,  1.42it/s]
2|4499|Loss: 0.8949989676475525:  70%|██████▉   | 4499/6470 [51:47<22:41,  1.45it/s]
2|4500|Loss: 0.8809069991111755:  70%|██████▉   | 4499/6470 [51:48<22:41,  1.45it/s]
2|4500|Loss: 0.8809069991111755:  70%|██████▉   | 4500/6470 [51:48<23:37,  1.39it/s]
2|4501|Loss: 0.8714872598648071:  70%|██████▉   | 4500/6470 [51:48<23:37,  1.39it/s]
2|4501|Loss: 0.8714872598648071:  70%|██████▉   | 4501/6470 [51:49<24:03,  1.36it/s]
2|4502|Loss: 0.8320630788803101:  70%|██████▉   | 4501/6470 [51:49<24:03,  1.36it/s]
2|4502|Loss: 0.8320630788803101:  70%|██████▉   | 4502/6470 [51:49<22:30,  1.46it/s]
2|4503|Loss: 0.7755228281021118:  70%|██████▉   | 4502/6470 [51:50<22:30,  1.46it/s]
2|4503|Loss: 0.7755228281021118:  70%|██████▉   | 4503/6470 [51:50<22:38,  1.45it/s]
2|4504|Loss: 0.7818753719329834:  70%|██████▉   | 4503/6470 [51:50<22:38,  1.45it/s]
2|4504|Loss: 0.7818753719329834:  70%|██████▉   | 4504/6470 [51:51<23:43,  1.38it/s]
2|4505|Loss: 0.8246386051177979:  70%|██████▉   | 4504/6470 [51:51<23:43,  1.38it/s]
2|4505|Loss: 0.8246386051177979:  70%|██████▉   | 4505/6470 [51:52<25:04,  1.31it/s]
2|4506|Loss: 0.7879859209060669:  70%|██████▉   | 4505/6470 [51:52<25:04,  1.31it/s]
2|4506|Loss: 0.7879859209060669:  70%|██████▉   | 4506/6470 [51:52<24:55,  1.31it/s]
2|4507|Loss: 0.9050571322441101:  70%|██████▉   | 4506/6470 [51:53<24:55,  1.31it/s]
2|4507|Loss: 0.9050571322441101:  70%|██████▉   | 4507/6470 [51:53<23:50,  1.37it/s]
2|4508|Loss: 0.9011217355728149:  70%|██████▉   | 4507/6470 [51:53<23:50,  1.37it/s]
2|4508|Loss: 0.9011217355728149:  70%|██████▉   | 4508/6470 [51:54<22:47,  1.43it/s]
2|4509|Loss: 0.9526317715644836:  70%|██████▉   | 4508/6470 [51:54<22:47,  1.43it/s]
2|4509|Loss: 0.9526317715644836:  70%|██████▉   | 4509/6470 [51:54<22:43,  1.44it/s]
2|4510|Loss: 0.9167038202285767:  70%|██████▉   | 4509/6470 [51:55<22:43,  1.44it/s]
2|4510|Loss: 0.9167038202285767:  70%|██████▉   | 4510/6470 [51:55<23:10,  1.41it/s]
2|4511|Loss: 0.8570745587348938:  70%|██████▉   | 4510/6470 [51:56<23:10,  1.41it/s]
2|4511|Loss: 0.8570745587348938:  70%|██████▉   | 4511/6470 [51:56<23:26,  1.39it/s]
2|4512|Loss: 0.9372504949569702:  70%|██████▉   | 4511/6470 [51:56<23:26,  1.39it/s]
2|4512|Loss: 0.9372504949569702:  70%|██████▉   | 4512/6470 [51:57<24:24,  1.34it/s]
2|4513|Loss: 0.9641760587692261:  70%|██████▉   | 4512/6470 [51:57<24:24,  1.34it/s]
2|4513|Loss: 0.9641760587692261:  70%|██████▉   | 4513/6470 [51:57<23:47,  1.37it/s]
2|4514|Loss: 0.8994103074073792:  70%|██████▉   | 4513/6470 [51:58<23:47,  1.37it/s]
2|4514|Loss: 0.8994103074073792:  70%|██████▉   | 4514/6470 [51:58<24:24,  1.34it/s]
2|4515|Loss: 0.9108706712722778:  70%|██████▉   | 4514/6470 [51:59<24:24,  1.34it/s]
2|4515|Loss: 0.9108706712722778:  70%|██████▉   | 4515/6470 [51:59<25:21,  1.28it/s]
2|4516|Loss: 0.922730028629303:  70%|██████▉   | 4515/6470 [51:59<25:21,  1.28it/s] 
2|4516|Loss: 0.922730028629303:  70%|██████▉   | 4516/6470 [52:00<24:00,  1.36it/s]
2|4517|Loss: 0.7579097747802734:  70%|██████▉   | 4516/6470 [52:00<24:00,  1.36it/s]
2|4517|Loss: 0.7579097747802734:  70%|██████▉   | 4517/6470 [52:00<24:08,  1.35it/s]
2|4518|Loss: 0.882383406162262:  70%|██████▉   | 4517/6470 [52:01<24:08,  1.35it/s] 
2|4518|Loss: 0.882383406162262:  70%|██████▉   | 4518/6470 [52:01<24:14,  1.34it/s]
2|4519|Loss: 0.5781996250152588:  70%|██████▉   | 4518/6470 [52:02<24:14,  1.34it/s]
2|4519|Loss: 0.5781996250152588:  70%|██████▉   | 4519/6470 [52:02<24:52,  1.31it/s]
2|4520|Loss: 0.7340556979179382:  70%|██████▉   | 4519/6470 [52:02<24:52,  1.31it/s]
2|4520|Loss: 0.7340556979179382:  70%|██████▉   | 4520/6470 [52:03<22:46,  1.43it/s]
2|4521|Loss: 0.9778133034706116:  70%|██████▉   | 4520/6470 [52:03<22:46,  1.43it/s]
2|4521|Loss: 0.9778133034706116:  70%|██████▉   | 4521/6470 [52:03<21:38,  1.50it/s]
2|4522|Loss: 0.8124191164970398:  70%|██████▉   | 4521/6470 [52:03<21:38,  1.50it/s]
2|4522|Loss: 0.8124191164970398:  70%|██████▉   | 4522/6470 [52:04<22:02,  1.47it/s]
2|4523|Loss: 1.1179653406143188:  70%|██████▉   | 4522/6470 [52:04<22:02,  1.47it/s]
2|4523|Loss: 1.1179653406143188:  70%|██████▉   | 4523/6470 [52:05<22:32,  1.44it/s]
2|4524|Loss: 0.763687789440155:  70%|██████▉   | 4523/6470 [52:05<22:32,  1.44it/s] 
2|4524|Loss: 0.763687789440155:  70%|██████▉   | 4524/6470 [52:05<23:01,  1.41it/s]
2|4525|Loss: 0.9007225632667542:  70%|██████▉   | 4524/6470 [52:06<23:01,  1.41it/s]
2|4525|Loss: 0.9007225632667542:  70%|██████▉   | 4525/6470 [52:06<23:06,  1.40it/s]
2|4526|Loss: 0.8427788615226746:  70%|██████▉   | 4525/6470 [52:06<23:06,  1.40it/s]
2|4526|Loss: 0.8427788615226746:  70%|██████▉   | 4526/6470 [52:07<22:06,  1.47it/s]
2|4527|Loss: 0.8478653430938721:  70%|██████▉   | 4526/6470 [52:07<22:06,  1.47it/s]
2|4527|Loss: 0.8478653430938721:  70%|██████▉   | 4527/6470 [52:07<22:00,  1.47it/s]
2|4528|Loss: 0.7550392746925354:  70%|██████▉   | 4527/6470 [52:08<22:00,  1.47it/s]
2|4528|Loss: 0.7550392746925354:  70%|██████▉   | 4528/6470 [52:08<21:31,  1.50it/s]
2|4529|Loss: 0.827286958694458:  70%|██████▉   | 4528/6470 [52:08<21:31,  1.50it/s] 
2|4529|Loss: 0.827286958694458:  70%|███████   | 4529/6470 [52:08<19:44,  1.64it/s]
2|4530|Loss: 0.9419421553611755:  70%|███████   | 4529/6470 [52:09<19:44,  1.64it/s]
2|4530|Loss: 0.9419421553611755:  70%|███████   | 4530/6470 [52:09<19:38,  1.65it/s]
2|4531|Loss: 0.8817051649093628:  70%|███████   | 4530/6470 [52:09<19:38,  1.65it/s]
2|4531|Loss: 0.8817051649093628:  70%|███████   | 4531/6470 [52:10<22:04,  1.46it/s]
2|4532|Loss: 0.9164486527442932:  70%|███████   | 4531/6470 [52:10<22:04,  1.46it/s]
2|4532|Loss: 0.9164486527442932:  70%|███████   | 4532/6470 [52:10<20:48,  1.55it/s]
2|4533|Loss: 0.9876844882965088:  70%|███████   | 4532/6470 [52:11<20:48,  1.55it/s]
2|4533|Loss: 0.9876844882965088:  70%|███████   | 4533/6470 [52:11<20:51,  1.55it/s]
2|4534|Loss: 0.7706957459449768:  70%|███████   | 4533/6470 [52:11<20:51,  1.55it/s]
2|4534|Loss: 0.7706957459449768:  70%|███████   | 4534/6470 [52:12<20:25,  1.58it/s]
2|4535|Loss: 0.8914487361907959:  70%|███████   | 4534/6470 [52:12<20:25,  1.58it/s]
2|4535|Loss: 0.8914487361907959:  70%|███████   | 4535/6470 [52:12<21:11,  1.52it/s]
2|4536|Loss: 0.8579296469688416:  70%|███████   | 4535/6470 [52:13<21:11,  1.52it/s]
2|4536|Loss: 0.8579296469688416:  70%|███████   | 4536/6470 [52:13<22:11,  1.45it/s]
2|4537|Loss: 0.7434409856796265:  70%|███████   | 4536/6470 [52:14<22:11,  1.45it/s]
2|4537|Loss: 0.7434409856796265:  70%|███████   | 4537/6470 [52:14<22:24,  1.44it/s]
2|4538|Loss: 0.9175947308540344:  70%|███████   | 4537/6470 [52:14<22:24,  1.44it/s]
2|4538|Loss: 0.9175947308540344:  70%|███████   | 4538/6470 [52:14<20:57,  1.54it/s]
2|4539|Loss: 0.9717050790786743:  70%|███████   | 4538/6470 [52:15<20:57,  1.54it/s]
2|4539|Loss: 0.9717050790786743:  70%|███████   | 4539/6470 [52:15<20:04,  1.60it/s]
2|4540|Loss: 0.8266530632972717:  70%|███████   | 4539/6470 [52:15<20:04,  1.60it/s]
2|4540|Loss: 0.8266530632972717:  70%|███████   | 4540/6470 [52:16<20:50,  1.54it/s]
2|4541|Loss: 0.7816010117530823:  70%|███████   | 4540/6470 [52:16<20:50,  1.54it/s]
2|4541|Loss: 0.7816010117530823:  70%|███████   | 4541/6470 [52:16<20:49,  1.54it/s]
2|4542|Loss: 1.0110825300216675:  70%|███████   | 4541/6470 [52:17<20:49,  1.54it/s]
2|4542|Loss: 1.0110825300216675:  70%|███████   | 4542/6470 [52:17<22:19,  1.44it/s]
2|4543|Loss: 0.8520732522010803:  70%|███████   | 4542/6470 [52:18<22:19,  1.44it/s]
2|4543|Loss: 0.8520732522010803:  70%|███████   | 4543/6470 [52:18<21:21,  1.50it/s]
2|4544|Loss: 0.7452335357666016:  70%|███████   | 4543/6470 [52:18<21:21,  1.50it/s]
2|4544|Loss: 0.7452335357666016:  70%|███████   | 4544/6470 [52:18<20:51,  1.54it/s]
2|4545|Loss: 0.8312339186668396:  70%|███████   | 4544/6470 [52:19<20:51,  1.54it/s]
2|4545|Loss: 0.8312339186668396:  70%|███████   | 4545/6470 [52:19<20:47,  1.54it/s]
2|4546|Loss: 0.7877410650253296:  70%|███████   | 4545/6470 [52:19<20:47,  1.54it/s]
2|4546|Loss: 0.7877410650253296:  70%|███████   | 4546/6470 [52:20<21:23,  1.50it/s]
2|4547|Loss: 0.9222368597984314:  70%|███████   | 4546/6470 [52:20<21:23,  1.50it/s]
2|4547|Loss: 0.9222368597984314:  70%|███████   | 4547/6470 [52:21<22:56,  1.40it/s]
2|4548|Loss: 0.8851833343505859:  70%|███████   | 4547/6470 [52:21<22:56,  1.40it/s]
2|4548|Loss: 0.8851833343505859:  70%|███████   | 4548/6470 [52:21<22:43,  1.41it/s]
2|4549|Loss: 0.7486466765403748:  70%|███████   | 4548/6470 [52:22<22:43,  1.41it/s]
2|4549|Loss: 0.7486466765403748:  70%|███████   | 4549/6470 [52:22<21:50,  1.47it/s]
2|4550|Loss: 0.7730733156204224:  70%|███████   | 4549/6470 [52:22<21:50,  1.47it/s]
2|4550|Loss: 0.7730733156204224:  70%|███████   | 4550/6470 [52:22<21:14,  1.51it/s]
2|4551|Loss: 0.7576082944869995:  70%|███████   | 4550/6470 [52:23<21:14,  1.51it/s]
2|4551|Loss: 0.7576082944869995:  70%|███████   | 4551/6470 [52:23<21:48,  1.47it/s]
2|4552|Loss: 0.8304789662361145:  70%|███████   | 4551/6470 [52:24<21:48,  1.47it/s]
2|4552|Loss: 0.8304789662361145:  70%|███████   | 4552/6470 [52:24<21:37,  1.48it/s]
2|4553|Loss: 0.8725813031196594:  70%|███████   | 4552/6470 [52:24<21:37,  1.48it/s]
2|4553|Loss: 0.8725813031196594:  70%|███████   | 4553/6470 [52:25<23:20,  1.37it/s]
2|4554|Loss: 0.9458529949188232:  70%|███████   | 4553/6470 [52:25<23:20,  1.37it/s]
2|4554|Loss: 0.9458529949188232:  70%|███████   | 4554/6470 [52:25<23:01,  1.39it/s]
2|4555|Loss: 0.8176384568214417:  70%|███████   | 4554/6470 [52:26<23:01,  1.39it/s]
2|4555|Loss: 0.8176384568214417:  70%|███████   | 4555/6470 [52:26<22:17,  1.43it/s]
2|4556|Loss: 0.7663256525993347:  70%|███████   | 4555/6470 [52:26<22:17,  1.43it/s]
2|4556|Loss: 0.7663256525993347:  70%|███████   | 4556/6470 [52:27<21:35,  1.48it/s]
2|4557|Loss: 0.8714560270309448:  70%|███████   | 4556/6470 [52:27<21:35,  1.48it/s]
2|4557|Loss: 0.8714560270309448:  70%|███████   | 4557/6470 [52:27<20:46,  1.54it/s]
2|4558|Loss: 0.7896071076393127:  70%|███████   | 4557/6470 [52:28<20:46,  1.54it/s]
2|4558|Loss: 0.7896071076393127:  70%|███████   | 4558/6470 [52:28<20:30,  1.55it/s]
2|4559|Loss: 0.8712131381034851:  70%|███████   | 4558/6470 [52:28<20:30,  1.55it/s]
2|4559|Loss: 0.8712131381034851:  70%|███████   | 4559/6470 [52:29<20:57,  1.52it/s]
2|4560|Loss: 0.9568139910697937:  70%|███████   | 4559/6470 [52:29<20:57,  1.52it/s]
2|4560|Loss: 0.9568139910697937:  70%|███████   | 4560/6470 [52:29<20:53,  1.52it/s]
2|4561|Loss: 0.8364246487617493:  70%|███████   | 4560/6470 [52:30<20:53,  1.52it/s]
2|4561|Loss: 0.8364246487617493:  70%|███████   | 4561/6470 [52:30<21:12,  1.50it/s]
2|4562|Loss: 0.7753690481185913:  70%|███████   | 4561/6470 [52:30<21:12,  1.50it/s]
2|4562|Loss: 0.7753690481185913:  71%|███████   | 4562/6470 [52:31<21:31,  1.48it/s]
2|4563|Loss: 0.9061766266822815:  71%|███████   | 4562/6470 [52:31<21:31,  1.48it/s]
2|4563|Loss: 0.9061766266822815:  71%|███████   | 4563/6470 [52:31<21:48,  1.46it/s]
2|4564|Loss: 0.9108691215515137:  71%|███████   | 4563/6470 [52:32<21:48,  1.46it/s]
2|4564|Loss: 0.9108691215515137:  71%|███████   | 4564/6470 [52:32<21:20,  1.49it/s]
2|4565|Loss: 0.9477897882461548:  71%|███████   | 4564/6470 [52:32<21:20,  1.49it/s]
2|4565|Loss: 0.9477897882461548:  71%|███████   | 4565/6470 [52:33<21:00,  1.51it/s]
2|4566|Loss: 0.9076547026634216:  71%|███████   | 4565/6470 [52:33<21:00,  1.51it/s]
2|4566|Loss: 0.9076547026634216:  71%|███████   | 4566/6470 [52:33<20:34,  1.54it/s]
2|4567|Loss: 0.9299535155296326:  71%|███████   | 4566/6470 [52:34<20:34,  1.54it/s]
2|4567|Loss: 0.9299535155296326:  71%|███████   | 4567/6470 [52:34<19:49,  1.60it/s]
2|4568|Loss: 0.9596127271652222:  71%|███████   | 4567/6470 [52:34<19:49,  1.60it/s]
2|4568|Loss: 0.9596127271652222:  71%|███████   | 4568/6470 [52:34<19:19,  1.64it/s]
2|4569|Loss: 0.7707670331001282:  71%|███████   | 4568/6470 [52:35<19:19,  1.64it/s]
2|4569|Loss: 0.7707670331001282:  71%|███████   | 4569/6470 [52:35<20:16,  1.56it/s]
2|4570|Loss: 0.9046775102615356:  71%|███████   | 4569/6470 [52:35<20:16,  1.56it/s]
2|4570|Loss: 0.9046775102615356:  71%|███████   | 4570/6470 [52:36<20:06,  1.57it/s]
2|4571|Loss: 0.906821608543396:  71%|███████   | 4570/6470 [52:36<20:06,  1.57it/s] 
2|4571|Loss: 0.906821608543396:  71%|███████   | 4571/6470 [52:36<19:11,  1.65it/s]
2|4572|Loss: 0.9256207942962646:  71%|███████   | 4571/6470 [52:37<19:11,  1.65it/s]
2|4572|Loss: 0.9256207942962646:  71%|███████   | 4572/6470 [52:37<20:43,  1.53it/s]
2|4573|Loss: 0.7984597086906433:  71%|███████   | 4572/6470 [52:37<20:43,  1.53it/s]
2|4573|Loss: 0.7984597086906433:  71%|███████   | 4573/6470 [52:38<21:35,  1.46it/s]
2|4574|Loss: 0.8677728772163391:  71%|███████   | 4573/6470 [52:38<21:35,  1.46it/s]
2|4574|Loss: 0.8677728772163391:  71%|███████   | 4574/6470 [52:38<22:03,  1.43it/s]
2|4575|Loss: 0.887192964553833:  71%|███████   | 4574/6470 [52:39<22:03,  1.43it/s] 
2|4575|Loss: 0.887192964553833:  71%|███████   | 4575/6470 [52:39<23:15,  1.36it/s]
2|4576|Loss: 0.9267712831497192:  71%|███████   | 4575/6470 [52:40<23:15,  1.36it/s]
2|4576|Loss: 0.9267712831497192:  71%|███████   | 4576/6470 [52:40<23:18,  1.35it/s]
2|4577|Loss: 0.9032886624336243:  71%|███████   | 4576/6470 [52:40<23:18,  1.35it/s]
2|4577|Loss: 0.9032886624336243:  71%|███████   | 4577/6470 [52:41<23:45,  1.33it/s]
2|4578|Loss: 0.9333277344703674:  71%|███████   | 4577/6470 [52:41<23:45,  1.33it/s]
2|4578|Loss: 0.9333277344703674:  71%|███████   | 4578/6470 [52:42<22:56,  1.37it/s]
2|4579|Loss: 0.8193334937095642:  71%|███████   | 4578/6470 [52:42<22:56,  1.37it/s]
2|4579|Loss: 0.8193334937095642:  71%|███████   | 4579/6470 [52:42<21:40,  1.45it/s]
2|4580|Loss: 0.8680270314216614:  71%|███████   | 4579/6470 [52:42<21:40,  1.45it/s]
2|4580|Loss: 0.8680270314216614:  71%|███████   | 4580/6470 [52:43<20:57,  1.50it/s]
2|4581|Loss: 0.7476573586463928:  71%|███████   | 4580/6470 [52:43<20:57,  1.50it/s]
2|4581|Loss: 0.7476573586463928:  71%|███████   | 4581/6470 [52:43<20:37,  1.53it/s]
2|4582|Loss: 0.8949517607688904:  71%|███████   | 4581/6470 [52:44<20:37,  1.53it/s]
2|4582|Loss: 0.8949517607688904:  71%|███████   | 4582/6470 [52:44<19:56,  1.58it/s]
2|4583|Loss: 0.8900400996208191:  71%|███████   | 4582/6470 [52:44<19:56,  1.58it/s]
2|4583|Loss: 0.8900400996208191:  71%|███████   | 4583/6470 [52:45<20:21,  1.55it/s]
2|4584|Loss: 0.8435965180397034:  71%|███████   | 4583/6470 [52:45<20:21,  1.55it/s]
2|4584|Loss: 0.8435965180397034:  71%|███████   | 4584/6470 [52:45<21:15,  1.48it/s]
2|4585|Loss: 0.8491881489753723:  71%|███████   | 4584/6470 [52:46<21:15,  1.48it/s]
2|4585|Loss: 0.8491881489753723:  71%|███████   | 4585/6470 [52:46<21:37,  1.45it/s]
2|4586|Loss: 0.9258586168289185:  71%|███████   | 4585/6470 [52:46<21:37,  1.45it/s]
2|4586|Loss: 0.9258586168289185:  71%|███████   | 4586/6470 [52:47<21:46,  1.44it/s]
2|4587|Loss: 0.7578246593475342:  71%|███████   | 4586/6470 [52:47<21:46,  1.44it/s]
2|4587|Loss: 0.7578246593475342:  71%|███████   | 4587/6470 [52:47<20:23,  1.54it/s]
2|4588|Loss: 0.8563897609710693:  71%|███████   | 4587/6470 [52:48<20:23,  1.54it/s]
2|4588|Loss: 0.8563897609710693:  71%|███████   | 4588/6470 [52:48<21:07,  1.49it/s]
2|4589|Loss: 0.8168469667434692:  71%|███████   | 4588/6470 [52:48<21:07,  1.49it/s]
2|4589|Loss: 0.8168469667434692:  71%|███████   | 4589/6470 [52:49<21:11,  1.48it/s]
2|4590|Loss: 0.8540839552879333:  71%|███████   | 4589/6470 [52:49<21:11,  1.48it/s]
2|4590|Loss: 0.8540839552879333:  71%|███████   | 4590/6470 [52:49<20:20,  1.54it/s]
2|4591|Loss: 0.9733259081840515:  71%|███████   | 4590/6470 [52:50<20:20,  1.54it/s]
2|4591|Loss: 0.9733259081840515:  71%|███████   | 4591/6470 [52:50<19:30,  1.61it/s]
2|4592|Loss: 0.8570055365562439:  71%|███████   | 4591/6470 [52:50<19:30,  1.61it/s]
2|4592|Loss: 0.8570055365562439:  71%|███████   | 4592/6470 [52:51<19:55,  1.57it/s]
2|4593|Loss: 0.932105302810669:  71%|███████   | 4592/6470 [52:51<19:55,  1.57it/s] 
2|4593|Loss: 0.932105302810669:  71%|███████   | 4593/6470 [52:51<20:35,  1.52it/s]
2|4594|Loss: 0.9780558347702026:  71%|███████   | 4593/6470 [52:52<20:35,  1.52it/s]
2|4594|Loss: 0.9780558347702026:  71%|███████   | 4594/6470 [52:52<20:30,  1.52it/s]
2|4595|Loss: 1.2399333715438843:  71%|███████   | 4594/6470 [52:52<20:30,  1.52it/s]
2|4595|Loss: 1.2399333715438843:  71%|███████   | 4595/6470 [52:53<20:47,  1.50it/s]
2|4596|Loss: 0.8311845660209656:  71%|███████   | 4595/6470 [52:53<20:47,  1.50it/s]
2|4596|Loss: 0.8311845660209656:  71%|███████   | 4596/6470 [52:53<19:58,  1.56it/s]
2|4597|Loss: 1.0044413805007935:  71%|███████   | 4596/6470 [52:54<19:58,  1.56it/s]
2|4597|Loss: 1.0044413805007935:  71%|███████   | 4597/6470 [52:54<20:37,  1.51it/s]
2|4598|Loss: 0.8929610252380371:  71%|███████   | 4597/6470 [52:54<20:37,  1.51it/s]
2|4598|Loss: 0.8929610252380371:  71%|███████   | 4598/6470 [52:55<21:09,  1.47it/s]
2|4599|Loss: 0.8421897292137146:  71%|███████   | 4598/6470 [52:55<21:09,  1.47it/s]
2|4599|Loss: 0.8421897292137146:  71%|███████   | 4599/6470 [52:55<21:41,  1.44it/s]
2|4600|Loss: 0.8285521864891052:  71%|███████   | 4599/6470 [52:56<21:41,  1.44it/s]
2|4600|Loss: 0.8285521864891052:  71%|███████   | 4600/6470 [52:56<21:49,  1.43it/s]
2|4601|Loss: 0.8584112524986267:  71%|███████   | 4600/6470 [52:56<21:49,  1.43it/s]
2|4601|Loss: 0.8584112524986267:  71%|███████   | 4601/6470 [52:57<22:04,  1.41it/s]
2|4602|Loss: 1.0260838270187378:  71%|███████   | 4601/6470 [52:57<22:04,  1.41it/s]
2|4602|Loss: 1.0260838270187378:  71%|███████   | 4602/6470 [52:57<21:28,  1.45it/s]
2|4603|Loss: 0.7640621662139893:  71%|███████   | 4602/6470 [52:58<21:28,  1.45it/s]
2|4603|Loss: 0.7640621662139893:  71%|███████   | 4603/6470 [52:58<21:46,  1.43it/s]
2|4604|Loss: 0.929571270942688:  71%|███████   | 4603/6470 [52:59<21:46,  1.43it/s] 
2|4604|Loss: 0.929571270942688:  71%|███████   | 4604/6470 [52:59<23:03,  1.35it/s]
2|4605|Loss: 0.831773042678833:  71%|███████   | 4604/6470 [52:59<23:03,  1.35it/s]
2|4605|Loss: 0.831773042678833:  71%|███████   | 4605/6470 [53:00<23:16,  1.34it/s]
2|4606|Loss: 0.8069663643836975:  71%|███████   | 4605/6470 [53:00<23:16,  1.34it/s]
2|4606|Loss: 0.8069663643836975:  71%|███████   | 4606/6470 [53:01<24:29,  1.27it/s]
2|4607|Loss: 0.8000112175941467:  71%|███████   | 4606/6470 [53:01<24:29,  1.27it/s]
2|4607|Loss: 0.8000112175941467:  71%|███████   | 4607/6470 [53:01<23:09,  1.34it/s]
2|4608|Loss: 0.7214818000793457:  71%|███████   | 4607/6470 [53:02<23:09,  1.34it/s]
2|4608|Loss: 0.7214818000793457:  71%|███████   | 4608/6470 [53:02<23:09,  1.34it/s]
2|4609|Loss: 0.7952180504798889:  71%|███████   | 4608/6470 [53:02<23:09,  1.34it/s]
2|4609|Loss: 0.7952180504798889:  71%|███████   | 4609/6470 [53:03<22:00,  1.41it/s]
2|4610|Loss: 0.7375422120094299:  71%|███████   | 4609/6470 [53:03<22:00,  1.41it/s]
2|4610|Loss: 0.7375422120094299:  71%|███████▏  | 4610/6470 [53:03<21:55,  1.41it/s]
2|4611|Loss: 0.8820759654045105:  71%|███████▏  | 4610/6470 [53:04<21:55,  1.41it/s]
2|4611|Loss: 0.8820759654045105:  71%|███████▏  | 4611/6470 [53:04<22:35,  1.37it/s]
2|4612|Loss: 0.8640171885490417:  71%|███████▏  | 4611/6470 [53:05<22:35,  1.37it/s]
2|4612|Loss: 0.8640171885490417:  71%|███████▏  | 4612/6470 [53:05<21:53,  1.41it/s]
2|4613|Loss: 0.8781399726867676:  71%|███████▏  | 4612/6470 [53:05<21:53,  1.41it/s]
2|4613|Loss: 0.8781399726867676:  71%|███████▏  | 4613/6470 [53:05<20:41,  1.50it/s]
2|4614|Loss: 1.0801676511764526:  71%|███████▏  | 4613/6470 [53:06<20:41,  1.50it/s]
2|4614|Loss: 1.0801676511764526:  71%|███████▏  | 4614/6470 [53:06<20:19,  1.52it/s]
2|4615|Loss: 0.901209831237793:  71%|███████▏  | 4614/6470 [53:06<20:19,  1.52it/s] 
2|4615|Loss: 0.901209831237793:  71%|███████▏  | 4615/6470 [53:07<20:37,  1.50it/s]
2|4616|Loss: 0.8402058482170105:  71%|███████▏  | 4615/6470 [53:07<20:37,  1.50it/s]
2|4616|Loss: 0.8402058482170105:  71%|███████▏  | 4616/6470 [53:07<19:29,  1.59it/s]
2|4617|Loss: 0.8010533452033997:  71%|███████▏  | 4616/6470 [53:08<19:29,  1.59it/s]
2|4617|Loss: 0.8010533452033997:  71%|███████▏  | 4617/6470 [53:08<20:04,  1.54it/s]
2|4618|Loss: 0.8957239985466003:  71%|███████▏  | 4617/6470 [53:08<20:04,  1.54it/s]
2|4618|Loss: 0.8957239985466003:  71%|███████▏  | 4618/6470 [53:09<20:43,  1.49it/s]
2|4619|Loss: 0.6900657415390015:  71%|███████▏  | 4618/6470 [53:09<20:43,  1.49it/s]
2|4619|Loss: 0.6900657415390015:  71%|███████▏  | 4619/6470 [53:09<21:25,  1.44it/s]
2|4620|Loss: 0.8145268559455872:  71%|███████▏  | 4619/6470 [53:10<21:25,  1.44it/s]
2|4620|Loss: 0.8145268559455872:  71%|███████▏  | 4620/6470 [53:10<20:44,  1.49it/s]
2|4621|Loss: 0.7610372304916382:  71%|███████▏  | 4620/6470 [53:10<20:44,  1.49it/s]
2|4621|Loss: 0.7610372304916382:  71%|███████▏  | 4621/6470 [53:11<21:28,  1.44it/s]
2|4622|Loss: 0.9490373134613037:  71%|███████▏  | 4621/6470 [53:11<21:28,  1.44it/s]
2|4622|Loss: 0.9490373134613037:  71%|███████▏  | 4622/6470 [53:11<21:17,  1.45it/s]
2|4623|Loss: 0.7654632329940796:  71%|███████▏  | 4622/6470 [53:12<21:17,  1.45it/s]
2|4623|Loss: 0.7654632329940796:  71%|███████▏  | 4623/6470 [53:12<20:18,  1.52it/s]
2|4624|Loss: 0.8176469206809998:  71%|███████▏  | 4623/6470 [53:12<20:18,  1.52it/s]
2|4624|Loss: 0.8176469206809998:  71%|███████▏  | 4624/6470 [53:13<21:59,  1.40it/s]
2|4625|Loss: 0.7632583379745483:  71%|███████▏  | 4624/6470 [53:13<21:59,  1.40it/s]
2|4625|Loss: 0.7632583379745483:  71%|███████▏  | 4625/6470 [53:14<21:50,  1.41it/s]
2|4626|Loss: 0.7899704575538635:  71%|███████▏  | 4625/6470 [53:14<21:50,  1.41it/s]
2|4626|Loss: 0.7899704575538635:  71%|███████▏  | 4626/6470 [53:14<22:38,  1.36it/s]
2|4627|Loss: 0.7876328825950623:  71%|███████▏  | 4626/6470 [53:15<22:38,  1.36it/s]
2|4627|Loss: 0.7876328825950623:  72%|███████▏  | 4627/6470 [53:15<22:36,  1.36it/s]
2|4628|Loss: 0.9326160550117493:  72%|███████▏  | 4627/6470 [53:16<22:36,  1.36it/s]
2|4628|Loss: 0.9326160550117493:  72%|███████▏  | 4628/6470 [53:16<22:37,  1.36it/s]
2|4629|Loss: 0.6648158431053162:  72%|███████▏  | 4628/6470 [53:16<22:37,  1.36it/s]
2|4629|Loss: 0.6648158431053162:  72%|███████▏  | 4629/6470 [53:17<23:02,  1.33it/s]
2|4630|Loss: 0.7188684344291687:  72%|███████▏  | 4629/6470 [53:17<23:02,  1.33it/s]
2|4630|Loss: 0.7188684344291687:  72%|███████▏  | 4630/6470 [53:17<23:38,  1.30it/s]
2|4631|Loss: 0.8165987730026245:  72%|███████▏  | 4630/6470 [53:18<23:38,  1.30it/s]
2|4631|Loss: 0.8165987730026245:  72%|███████▏  | 4631/6470 [53:18<21:55,  1.40it/s]
2|4632|Loss: 0.8080256581306458:  72%|███████▏  | 4631/6470 [53:18<21:55,  1.40it/s]
2|4632|Loss: 0.8080256581306458:  72%|███████▏  | 4632/6470 [53:19<21:46,  1.41it/s]
2|4633|Loss: 0.9245255589485168:  72%|███████▏  | 4632/6470 [53:19<21:46,  1.41it/s]
2|4633|Loss: 0.9245255589485168:  72%|███████▏  | 4633/6470 [53:19<20:24,  1.50it/s]
2|4634|Loss: 0.705608606338501:  72%|███████▏  | 4633/6470 [53:20<20:24,  1.50it/s] 
2|4634|Loss: 0.705608606338501:  72%|███████▏  | 4634/6470 [53:20<21:57,  1.39it/s]
2|4635|Loss: 0.884609580039978:  72%|███████▏  | 4634/6470 [53:21<21:57,  1.39it/s]
2|4635|Loss: 0.884609580039978:  72%|███████▏  | 4635/6470 [53:21<22:25,  1.36it/s]
2|4636|Loss: 0.7905377149581909:  72%|███████▏  | 4635/6470 [53:21<22:25,  1.36it/s]
2|4636|Loss: 0.7905377149581909:  72%|███████▏  | 4636/6470 [53:22<23:17,  1.31it/s]
2|4637|Loss: 0.8446038961410522:  72%|███████▏  | 4636/6470 [53:22<23:17,  1.31it/s]
2|4637|Loss: 0.8446038961410522:  72%|███████▏  | 4637/6470 [53:22<22:26,  1.36it/s]
2|4638|Loss: 0.776069164276123:  72%|███████▏  | 4637/6470 [53:23<22:26,  1.36it/s] 
2|4638|Loss: 0.776069164276123:  72%|███████▏  | 4638/6470 [53:23<23:05,  1.32it/s]
2|4639|Loss: 0.8219150304794312:  72%|███████▏  | 4638/6470 [53:24<23:05,  1.32it/s]
2|4639|Loss: 0.8219150304794312:  72%|███████▏  | 4639/6470 [53:24<21:39,  1.41it/s]
2|4640|Loss: 0.8380024433135986:  72%|███████▏  | 4639/6470 [53:24<21:39,  1.41it/s]
2|4640|Loss: 0.8380024433135986:  72%|███████▏  | 4640/6470 [53:24<19:25,  1.57it/s]
2|4641|Loss: 0.9068378210067749:  72%|███████▏  | 4640/6470 [53:25<19:25,  1.57it/s]
2|4641|Loss: 0.9068378210067749:  72%|███████▏  | 4641/6470 [53:25<18:54,  1.61it/s]
2|4642|Loss: 1.018431305885315:  72%|███████▏  | 4641/6470 [53:25<18:54,  1.61it/s] 
2|4642|Loss: 1.018431305885315:  72%|███████▏  | 4642/6470 [53:26<19:41,  1.55it/s]
2|4643|Loss: 0.7873260974884033:  72%|███████▏  | 4642/6470 [53:26<19:41,  1.55it/s]
2|4643|Loss: 0.7873260974884033:  72%|███████▏  | 4643/6470 [53:26<19:47,  1.54it/s]
2|4644|Loss: 0.7825716733932495:  72%|███████▏  | 4643/6470 [53:27<19:47,  1.54it/s]
2|4644|Loss: 0.7825716733932495:  72%|███████▏  | 4644/6470 [53:27<20:13,  1.50it/s]
2|4645|Loss: 0.7842452526092529:  72%|███████▏  | 4644/6470 [53:27<20:13,  1.50it/s]
2|4645|Loss: 0.7842452526092529:  72%|███████▏  | 4645/6470 [53:28<20:47,  1.46it/s]
2|4646|Loss: 0.7625468373298645:  72%|███████▏  | 4645/6470 [53:28<20:47,  1.46it/s]
2|4646|Loss: 0.7625468373298645:  72%|███████▏  | 4646/6470 [53:28<21:01,  1.45it/s]
2|4647|Loss: 0.8916040658950806:  72%|███████▏  | 4646/6470 [53:29<21:01,  1.45it/s]
2|4647|Loss: 0.8916040658950806:  72%|███████▏  | 4647/6470 [53:29<22:00,  1.38it/s]
2|4648|Loss: 0.7958882451057434:  72%|███████▏  | 4647/6470 [53:30<22:00,  1.38it/s]
2|4648|Loss: 0.7958882451057434:  72%|███████▏  | 4648/6470 [53:30<21:24,  1.42it/s]
2|4649|Loss: 0.7280552387237549:  72%|███████▏  | 4648/6470 [53:30<21:24,  1.42it/s]
2|4649|Loss: 0.7280552387237549:  72%|███████▏  | 4649/6470 [53:31<21:13,  1.43it/s]
2|4650|Loss: 0.6975602507591248:  72%|███████▏  | 4649/6470 [53:31<21:13,  1.43it/s]
2|4650|Loss: 0.6975602507591248:  72%|███████▏  | 4650/6470 [53:31<19:25,  1.56it/s]
2|4651|Loss: 0.8027878999710083:  72%|███████▏  | 4650/6470 [53:31<19:25,  1.56it/s]
2|4651|Loss: 0.8027878999710083:  72%|███████▏  | 4651/6470 [53:32<18:54,  1.60it/s]
2|4652|Loss: 0.865551769733429:  72%|███████▏  | 4651/6470 [53:32<18:54,  1.60it/s] 
2|4652|Loss: 0.865551769733429:  72%|███████▏  | 4652/6470 [53:32<18:32,  1.63it/s]
2|4653|Loss: 0.8137220740318298:  72%|███████▏  | 4652/6470 [53:33<18:32,  1.63it/s]
2|4653|Loss: 0.8137220740318298:  72%|███████▏  | 4653/6470 [53:33<19:46,  1.53it/s]
2|4654|Loss: 0.6868770718574524:  72%|███████▏  | 4653/6470 [53:33<19:46,  1.53it/s]
2|4654|Loss: 0.6868770718574524:  72%|███████▏  | 4654/6470 [53:34<21:55,  1.38it/s]
2|4655|Loss: 0.932837963104248:  72%|███████▏  | 4654/6470 [53:34<21:55,  1.38it/s] 
2|4655|Loss: 0.932837963104248:  72%|███████▏  | 4655/6470 [53:35<21:43,  1.39it/s]
2|4656|Loss: 0.8264811635017395:  72%|███████▏  | 4655/6470 [53:35<21:43,  1.39it/s]
2|4656|Loss: 0.8264811635017395:  72%|███████▏  | 4656/6470 [53:35<22:23,  1.35it/s]
2|4657|Loss: 0.8668031096458435:  72%|███████▏  | 4656/6470 [53:36<22:23,  1.35it/s]
2|4657|Loss: 0.8668031096458435:  72%|███████▏  | 4657/6470 [53:36<21:02,  1.44it/s]
2|4658|Loss: 1.0337553024291992:  72%|███████▏  | 4657/6470 [53:36<21:02,  1.44it/s]
2|4658|Loss: 1.0337553024291992:  72%|███████▏  | 4658/6470 [53:36<19:23,  1.56it/s]
2|4659|Loss: 0.9592357277870178:  72%|███████▏  | 4658/6470 [53:37<19:23,  1.56it/s]
2|4659|Loss: 0.9592357277870178:  72%|███████▏  | 4659/6470 [53:37<19:06,  1.58it/s]
2|4660|Loss: 1.036380648612976:  72%|███████▏  | 4659/6470 [53:37<19:06,  1.58it/s] 
2|4660|Loss: 1.036380648612976:  72%|███████▏  | 4660/6470 [53:38<19:52,  1.52it/s]
2|4661|Loss: 0.8097279071807861:  72%|███████▏  | 4660/6470 [53:38<19:52,  1.52it/s]
2|4661|Loss: 0.8097279071807861:  72%|███████▏  | 4661/6470 [53:38<19:52,  1.52it/s]
2|4662|Loss: 0.9307084083557129:  72%|███████▏  | 4661/6470 [53:39<19:52,  1.52it/s]
2|4662|Loss: 0.9307084083557129:  72%|███████▏  | 4662/6470 [53:39<20:06,  1.50it/s]
2|4663|Loss: 0.9704022407531738:  72%|███████▏  | 4662/6470 [53:40<20:06,  1.50it/s]
2|4663|Loss: 0.9704022407531738:  72%|███████▏  | 4663/6470 [53:40<20:37,  1.46it/s]
2|4664|Loss: 0.8649045825004578:  72%|███████▏  | 4663/6470 [53:40<20:37,  1.46it/s]
2|4664|Loss: 0.8649045825004578:  72%|███████▏  | 4664/6470 [53:41<21:49,  1.38it/s]
2|4665|Loss: 1.091963291168213:  72%|███████▏  | 4664/6470 [53:41<21:49,  1.38it/s] 
2|4665|Loss: 1.091963291168213:  72%|███████▏  | 4665/6470 [53:41<22:04,  1.36it/s]
2|4666|Loss: 0.8167270421981812:  72%|███████▏  | 4665/6470 [53:42<22:04,  1.36it/s]
2|4666|Loss: 0.8167270421981812:  72%|███████▏  | 4666/6470 [53:42<21:49,  1.38it/s]
2|4667|Loss: 0.9687855839729309:  72%|███████▏  | 4666/6470 [53:43<21:49,  1.38it/s]
2|4667|Loss: 0.9687855839729309:  72%|███████▏  | 4667/6470 [53:43<22:28,  1.34it/s]
2|4668|Loss: 1.025031566619873:  72%|███████▏  | 4667/6470 [53:43<22:28,  1.34it/s] 
2|4668|Loss: 1.025031566619873:  72%|███████▏  | 4668/6470 [53:44<22:17,  1.35it/s]
2|4669|Loss: 0.7771432995796204:  72%|███████▏  | 4668/6470 [53:44<22:17,  1.35it/s]
2|4669|Loss: 0.7771432995796204:  72%|███████▏  | 4669/6470 [53:44<20:58,  1.43it/s]
2|4670|Loss: 0.7773160934448242:  72%|███████▏  | 4669/6470 [53:45<20:58,  1.43it/s]
2|4670|Loss: 0.7773160934448242:  72%|███████▏  | 4670/6470 [53:45<20:08,  1.49it/s]
2|4671|Loss: 0.7880719304084778:  72%|███████▏  | 4670/6470 [53:45<20:08,  1.49it/s]
2|4671|Loss: 0.7880719304084778:  72%|███████▏  | 4671/6470 [53:45<19:23,  1.55it/s]
2|4672|Loss: 0.8826449513435364:  72%|███████▏  | 4671/6470 [53:46<19:23,  1.55it/s]
2|4672|Loss: 0.8826449513435364:  72%|███████▏  | 4672/6470 [53:46<21:01,  1.43it/s]
2|4673|Loss: 0.8493922352790833:  72%|███████▏  | 4672/6470 [53:47<21:01,  1.43it/s]
2|4673|Loss: 0.8493922352790833:  72%|███████▏  | 4673/6470 [53:47<21:35,  1.39it/s]
2|4674|Loss: 0.9057989716529846:  72%|███████▏  | 4673/6470 [53:47<21:35,  1.39it/s]
2|4674|Loss: 0.9057989716529846:  72%|███████▏  | 4674/6470 [53:48<20:06,  1.49it/s]
2|4675|Loss: 0.8076064586639404:  72%|███████▏  | 4674/6470 [53:48<20:06,  1.49it/s]
2|4675|Loss: 0.8076064586639404:  72%|███████▏  | 4675/6470 [53:48<20:15,  1.48it/s]
2|4676|Loss: 0.9969061017036438:  72%|███████▏  | 4675/6470 [53:49<20:15,  1.48it/s]
2|4676|Loss: 0.9969061017036438:  72%|███████▏  | 4676/6470 [53:49<20:46,  1.44it/s]
2|4677|Loss: 0.8633944988250732:  72%|███████▏  | 4676/6470 [53:49<20:46,  1.44it/s]
2|4677|Loss: 0.8633944988250732:  72%|███████▏  | 4677/6470 [53:50<21:59,  1.36it/s]
2|4678|Loss: 0.7799848914146423:  72%|███████▏  | 4677/6470 [53:50<21:59,  1.36it/s]
2|4678|Loss: 0.7799848914146423:  72%|███████▏  | 4678/6470 [53:51<22:22,  1.33it/s]
2|4679|Loss: 0.9176594018936157:  72%|███████▏  | 4678/6470 [53:51<22:22,  1.33it/s]
2|4679|Loss: 0.9176594018936157:  72%|███████▏  | 4679/6470 [53:51<21:25,  1.39it/s]
2|4680|Loss: 0.8311554789543152:  72%|███████▏  | 4679/6470 [53:52<21:25,  1.39it/s]
2|4680|Loss: 0.8311554789543152:  72%|███████▏  | 4680/6470 [53:52<19:49,  1.50it/s]
2|4681|Loss: 0.9752601981163025:  72%|███████▏  | 4680/6470 [53:52<19:49,  1.50it/s]
2|4681|Loss: 0.9752601981163025:  72%|███████▏  | 4681/6470 [53:53<20:42,  1.44it/s]
2|4682|Loss: 0.8279736042022705:  72%|███████▏  | 4681/6470 [53:53<20:42,  1.44it/s]
2|4682|Loss: 0.8279736042022705:  72%|███████▏  | 4682/6470 [53:53<21:13,  1.40it/s]
2|4683|Loss: 1.0307021141052246:  72%|███████▏  | 4682/6470 [53:54<21:13,  1.40it/s]
2|4683|Loss: 1.0307021141052246:  72%|███████▏  | 4683/6470 [53:54<20:33,  1.45it/s]
2|4684|Loss: 0.6670594215393066:  72%|███████▏  | 4683/6470 [53:54<20:33,  1.45it/s]
2|4684|Loss: 0.6670594215393066:  72%|███████▏  | 4684/6470 [53:55<20:49,  1.43it/s]
2|4685|Loss: 0.8219435811042786:  72%|███████▏  | 4684/6470 [53:55<20:49,  1.43it/s]
2|4685|Loss: 0.8219435811042786:  72%|███████▏  | 4685/6470 [53:55<21:16,  1.40it/s]
2|4686|Loss: 0.935974657535553:  72%|███████▏  | 4685/6470 [53:56<21:16,  1.40it/s] 
2|4686|Loss: 0.935974657535553:  72%|███████▏  | 4686/6470 [53:56<20:50,  1.43it/s]
2|4687|Loss: 0.8384016156196594:  72%|███████▏  | 4686/6470 [53:56<20:50,  1.43it/s]
2|4687|Loss: 0.8384016156196594:  72%|███████▏  | 4687/6470 [53:57<20:03,  1.48it/s]
2|4688|Loss: 0.8685111999511719:  72%|███████▏  | 4687/6470 [53:57<20:03,  1.48it/s]
2|4688|Loss: 0.8685111999511719:  72%|███████▏  | 4688/6470 [53:57<20:20,  1.46it/s]
2|4689|Loss: 0.9673154950141907:  72%|███████▏  | 4688/6470 [53:58<20:20,  1.46it/s]
2|4689|Loss: 0.9673154950141907:  72%|███████▏  | 4689/6470 [53:58<19:28,  1.52it/s]
2|4690|Loss: 0.8869531750679016:  72%|███████▏  | 4689/6470 [53:58<19:28,  1.52it/s]
2|4690|Loss: 0.8869531750679016:  72%|███████▏  | 4690/6470 [53:59<20:16,  1.46it/s]
2|4691|Loss: 0.956049382686615:  72%|███████▏  | 4690/6470 [53:59<20:16,  1.46it/s] 
2|4691|Loss: 0.956049382686615:  73%|███████▎  | 4691/6470 [53:59<20:03,  1.48it/s]
2|4692|Loss: 0.862882137298584:  73%|███████▎  | 4691/6470 [54:00<20:03,  1.48it/s]
2|4692|Loss: 0.862882137298584:  73%|███████▎  | 4692/6470 [54:00<20:26,  1.45it/s]
2|4693|Loss: 0.7652140259742737:  73%|███████▎  | 4692/6470 [54:01<20:26,  1.45it/s]
2|4693|Loss: 0.7652140259742737:  73%|███████▎  | 4693/6470 [54:01<20:04,  1.48it/s]
2|4694|Loss: 0.8248345255851746:  73%|███████▎  | 4693/6470 [54:01<20:04,  1.48it/s]
2|4694|Loss: 0.8248345255851746:  73%|███████▎  | 4694/6470 [54:02<21:07,  1.40it/s]
2|4695|Loss: 0.6772879958152771:  73%|███████▎  | 4694/6470 [54:02<21:07,  1.40it/s]
2|4695|Loss: 0.6772879958152771:  73%|███████▎  | 4695/6470 [54:02<22:10,  1.33it/s]
2|4696|Loss: 0.7979785203933716:  73%|███████▎  | 4695/6470 [54:03<22:10,  1.33it/s]
2|4696|Loss: 0.7979785203933716:  73%|███████▎  | 4696/6470 [54:03<22:52,  1.29it/s]
2|4697|Loss: 0.9226217269897461:  73%|███████▎  | 4696/6470 [54:04<22:52,  1.29it/s]
2|4697|Loss: 0.9226217269897461:  73%|███████▎  | 4697/6470 [54:04<21:52,  1.35it/s]
2|4698|Loss: 0.8640430569648743:  73%|███████▎  | 4697/6470 [54:04<21:52,  1.35it/s]
2|4698|Loss: 0.8640430569648743:  73%|███████▎  | 4698/6470 [54:04<19:59,  1.48it/s]
2|4699|Loss: 0.8863388895988464:  73%|███████▎  | 4698/6470 [54:05<19:59,  1.48it/s]
2|4699|Loss: 0.8863388895988464:  73%|███████▎  | 4699/6470 [54:05<18:47,  1.57it/s]
2|4700|Loss: 0.8033742308616638:  73%|███████▎  | 4699/6470 [54:05<18:47,  1.57it/s]
2|4700|Loss: 0.8033742308616638:  73%|███████▎  | 4700/6470 [54:06<18:32,  1.59it/s]
2|4701|Loss: 0.9330343008041382:  73%|███████▎  | 4700/6470 [54:06<18:32,  1.59it/s]
2|4701|Loss: 0.9330343008041382:  73%|███████▎  | 4701/6470 [54:06<20:04,  1.47it/s]
2|4702|Loss: 0.8482981324195862:  73%|███████▎  | 4701/6470 [54:07<20:04,  1.47it/s]
2|4702|Loss: 0.8482981324195862:  73%|███████▎  | 4702/6470 [54:07<21:38,  1.36it/s]
2|4703|Loss: 0.7646080851554871:  73%|███████▎  | 4702/6470 [54:08<21:38,  1.36it/s]
2|4703|Loss: 0.7646080851554871:  73%|███████▎  | 4703/6470 [54:08<21:59,  1.34it/s]
2|4704|Loss: 0.8394035696983337:  73%|███████▎  | 4703/6470 [54:08<21:59,  1.34it/s]
2|4704|Loss: 0.8394035696983337:  73%|███████▎  | 4704/6470 [54:09<19:54,  1.48it/s]
2|4705|Loss: 0.8139946460723877:  73%|███████▎  | 4704/6470 [54:09<19:54,  1.48it/s]
2|4705|Loss: 0.8139946460723877:  73%|███████▎  | 4705/6470 [54:09<18:58,  1.55it/s]
2|4706|Loss: 0.7575938701629639:  73%|███████▎  | 4705/6470 [54:09<18:58,  1.55it/s]
2|4706|Loss: 0.7575938701629639:  73%|███████▎  | 4706/6470 [54:10<18:51,  1.56it/s]
2|4707|Loss: 0.7747277021408081:  73%|███████▎  | 4706/6470 [54:10<18:51,  1.56it/s]
2|4707|Loss: 0.7747277021408081:  73%|███████▎  | 4707/6470 [54:11<20:25,  1.44it/s]
2|4708|Loss: 0.8216289281845093:  73%|███████▎  | 4707/6470 [54:11<20:25,  1.44it/s]
2|4708|Loss: 0.8216289281845093:  73%|███████▎  | 4708/6470 [54:11<21:30,  1.37it/s]
2|4709|Loss: 0.8247939944267273:  73%|███████▎  | 4708/6470 [54:12<21:30,  1.37it/s]
2|4709|Loss: 0.8247939944267273:  73%|███████▎  | 4709/6470 [54:12<20:53,  1.41it/s]
2|4710|Loss: 0.9001534581184387:  73%|███████▎  | 4709/6470 [54:12<20:53,  1.41it/s]
2|4710|Loss: 0.9001534581184387:  73%|███████▎  | 4710/6470 [54:13<21:48,  1.34it/s]
2|4711|Loss: 0.9353829026222229:  73%|███████▎  | 4710/6470 [54:13<21:48,  1.34it/s]
2|4711|Loss: 0.9353829026222229:  73%|███████▎  | 4711/6470 [54:14<22:24,  1.31it/s]
2|4712|Loss: 0.6512370705604553:  73%|███████▎  | 4711/6470 [54:14<22:24,  1.31it/s]
2|4712|Loss: 0.6512370705604553:  73%|███████▎  | 4712/6470 [54:14<21:59,  1.33it/s]
2|4713|Loss: 0.768227756023407:  73%|███████▎  | 4712/6470 [54:15<21:59,  1.33it/s] 
2|4713|Loss: 0.768227756023407:  73%|███████▎  | 4713/6470 [54:15<21:51,  1.34it/s]
2|4714|Loss: 1.1643344163894653:  73%|███████▎  | 4713/6470 [54:16<21:51,  1.34it/s]
2|4714|Loss: 1.1643344163894653:  73%|███████▎  | 4714/6470 [54:16<23:31,  1.24it/s]
2|4715|Loss: 0.8671970963478088:  73%|███████▎  | 4714/6470 [54:17<23:31,  1.24it/s]
2|4715|Loss: 0.8671970963478088:  73%|███████▎  | 4715/6470 [54:17<24:06,  1.21it/s]
2|4716|Loss: 0.7397110462188721:  73%|███████▎  | 4715/6470 [54:17<24:06,  1.21it/s]
2|4716|Loss: 0.7397110462188721:  73%|███████▎  | 4716/6470 [54:18<22:55,  1.28it/s]
2|4717|Loss: 0.9537011384963989:  73%|███████▎  | 4716/6470 [54:18<22:55,  1.28it/s]
2|4717|Loss: 0.9537011384963989:  73%|███████▎  | 4717/6470 [54:18<21:12,  1.38it/s]
2|4718|Loss: 0.8665353059768677:  73%|███████▎  | 4717/6470 [54:19<21:12,  1.38it/s]
2|4718|Loss: 0.8665353059768677:  73%|███████▎  | 4718/6470 [54:19<20:11,  1.45it/s]
2|4719|Loss: 0.8986337780952454:  73%|███████▎  | 4718/6470 [54:19<20:11,  1.45it/s]
2|4719|Loss: 0.8986337780952454:  73%|███████▎  | 4719/6470 [54:20<20:07,  1.45it/s]
2|4720|Loss: 0.9197618961334229:  73%|███████▎  | 4719/6470 [54:20<20:07,  1.45it/s]
2|4720|Loss: 0.9197618961334229:  73%|███████▎  | 4720/6470 [54:20<20:11,  1.44it/s]
2|4721|Loss: 0.7649939656257629:  73%|███████▎  | 4720/6470 [54:21<20:11,  1.44it/s]
2|4721|Loss: 0.7649939656257629:  73%|███████▎  | 4721/6470 [54:21<20:14,  1.44it/s]
2|4722|Loss: 0.9213675260543823:  73%|███████▎  | 4721/6470 [54:21<20:14,  1.44it/s]
2|4722|Loss: 0.9213675260543823:  73%|███████▎  | 4722/6470 [54:22<20:20,  1.43it/s]
2|4723|Loss: 0.8155173659324646:  73%|███████▎  | 4722/6470 [54:22<20:20,  1.43it/s]
2|4723|Loss: 0.8155173659324646:  73%|███████▎  | 4723/6470 [54:22<20:24,  1.43it/s]
2|4724|Loss: 0.8189689517021179:  73%|███████▎  | 4723/6470 [54:23<20:24,  1.43it/s]
2|4724|Loss: 0.8189689517021179:  73%|███████▎  | 4724/6470 [54:23<20:22,  1.43it/s]
2|4725|Loss: 0.9091174006462097:  73%|███████▎  | 4724/6470 [54:23<20:22,  1.43it/s]
2|4725|Loss: 0.9091174006462097:  73%|███████▎  | 4725/6470 [54:24<20:26,  1.42it/s]
2|4726|Loss: 0.8165419101715088:  73%|███████▎  | 4725/6470 [54:24<20:26,  1.42it/s]
2|4726|Loss: 0.8165419101715088:  73%|███████▎  | 4726/6470 [54:25<21:18,  1.36it/s]
2|4727|Loss: 0.7691362500190735:  73%|███████▎  | 4726/6470 [54:25<21:18,  1.36it/s]
2|4727|Loss: 0.7691362500190735:  73%|███████▎  | 4727/6470 [54:25<21:27,  1.35it/s]
2|4728|Loss: 0.737184464931488:  73%|███████▎  | 4727/6470 [54:26<21:27,  1.35it/s] 
2|4728|Loss: 0.737184464931488:  73%|███████▎  | 4728/6470 [54:26<21:52,  1.33it/s]
2|4729|Loss: 0.8770293593406677:  73%|███████▎  | 4728/6470 [54:27<21:52,  1.33it/s]
2|4729|Loss: 0.8770293593406677:  73%|███████▎  | 4729/6470 [54:27<21:48,  1.33it/s]
2|4730|Loss: 0.7874548435211182:  73%|███████▎  | 4729/6470 [54:27<21:48,  1.33it/s]
2|4730|Loss: 0.7874548435211182:  73%|███████▎  | 4730/6470 [54:28<22:34,  1.28it/s]
2|4731|Loss: 0.9055159687995911:  73%|███████▎  | 4730/6470 [54:28<22:34,  1.28it/s]
2|4731|Loss: 0.9055159687995911:  73%|███████▎  | 4731/6470 [54:28<21:23,  1.35it/s]
2|4732|Loss: 0.8828161954879761:  73%|███████▎  | 4731/6470 [54:29<21:23,  1.35it/s]
2|4732|Loss: 0.8828161954879761:  73%|███████▎  | 4732/6470 [54:29<20:48,  1.39it/s]
2|4733|Loss: 0.788620114326477:  73%|███████▎  | 4732/6470 [54:29<20:48,  1.39it/s] 
2|4733|Loss: 0.788620114326477:  73%|███████▎  | 4733/6470 [54:30<20:51,  1.39it/s]
2|4734|Loss: 0.887107253074646:  73%|███████▎  | 4733/6470 [54:30<20:51,  1.39it/s]
2|4734|Loss: 0.887107253074646:  73%|███████▎  | 4734/6470 [54:30<20:10,  1.43it/s]
2|4735|Loss: 0.7290019392967224:  73%|███████▎  | 4734/6470 [54:31<20:10,  1.43it/s]
2|4735|Loss: 0.7290019392967224:  73%|███████▎  | 4735/6470 [54:31<22:04,  1.31it/s]
2|4736|Loss: 0.9336123466491699:  73%|███████▎  | 4735/6470 [54:32<22:04,  1.31it/s]
2|4736|Loss: 0.9336123466491699:  73%|███████▎  | 4736/6470 [54:32<21:11,  1.36it/s]
2|4737|Loss: 1.0407451391220093:  73%|███████▎  | 4736/6470 [54:32<21:11,  1.36it/s]
2|4737|Loss: 1.0407451391220093:  73%|███████▎  | 4737/6470 [54:33<20:52,  1.38it/s]
2|4738|Loss: 0.9855679869651794:  73%|███████▎  | 4737/6470 [54:33<20:52,  1.38it/s]
2|4738|Loss: 0.9855679869651794:  73%|███████▎  | 4738/6470 [54:33<20:21,  1.42it/s]
2|4739|Loss: 0.8008857369422913:  73%|███████▎  | 4738/6470 [54:34<20:21,  1.42it/s]
2|4739|Loss: 0.8008857369422913:  73%|███████▎  | 4739/6470 [54:34<18:02,  1.60it/s]
2|4740|Loss: 0.8017700910568237:  73%|███████▎  | 4739/6470 [54:34<18:02,  1.60it/s]
2|4740|Loss: 0.8017700910568237:  73%|███████▎  | 4740/6470 [54:34<18:31,  1.56it/s]
2|4741|Loss: 0.7868794798851013:  73%|███████▎  | 4740/6470 [54:35<18:31,  1.56it/s]
2|4741|Loss: 0.7868794798851013:  73%|███████▎  | 4741/6470 [54:35<17:26,  1.65it/s]
2|4742|Loss: 0.9120417237281799:  73%|███████▎  | 4741/6470 [54:35<17:26,  1.65it/s]
2|4742|Loss: 0.9120417237281799:  73%|███████▎  | 4742/6470 [54:36<18:59,  1.52it/s]
2|4743|Loss: 0.8590572476387024:  73%|███████▎  | 4742/6470 [54:36<18:59,  1.52it/s]
2|4743|Loss: 0.8590572476387024:  73%|███████▎  | 4743/6470 [54:36<18:41,  1.54it/s]
2|4744|Loss: 0.7929225564002991:  73%|███████▎  | 4743/6470 [54:37<18:41,  1.54it/s]
2|4744|Loss: 0.7929225564002991:  73%|███████▎  | 4744/6470 [54:37<18:26,  1.56it/s]
2|4745|Loss: 0.9301936626434326:  73%|███████▎  | 4744/6470 [54:37<18:26,  1.56it/s]
2|4745|Loss: 0.9301936626434326:  73%|███████▎  | 4745/6470 [54:38<20:20,  1.41it/s]
2|4746|Loss: 0.9410814642906189:  73%|███████▎  | 4745/6470 [54:38<20:20,  1.41it/s]
2|4746|Loss: 0.9410814642906189:  73%|███████▎  | 4746/6470 [54:39<20:05,  1.43it/s]
2|4747|Loss: 0.830969512462616:  73%|███████▎  | 4746/6470 [54:39<20:05,  1.43it/s] 
2|4747|Loss: 0.830969512462616:  73%|███████▎  | 4747/6470 [54:39<19:26,  1.48it/s]
2|4748|Loss: 0.8068446516990662:  73%|███████▎  | 4747/6470 [54:40<19:26,  1.48it/s]
2|4748|Loss: 0.8068446516990662:  73%|███████▎  | 4748/6470 [54:40<19:34,  1.47it/s]
2|4749|Loss: 0.7748851180076599:  73%|███████▎  | 4748/6470 [54:40<19:34,  1.47it/s]
2|4749|Loss: 0.7748851180076599:  73%|███████▎  | 4749/6470 [54:41<19:45,  1.45it/s]
2|4750|Loss: 0.8083539009094238:  73%|███████▎  | 4749/6470 [54:41<19:45,  1.45it/s]
2|4750|Loss: 0.8083539009094238:  73%|███████▎  | 4750/6470 [54:41<20:44,  1.38it/s]
2|4751|Loss: 0.9044710993766785:  73%|███████▎  | 4750/6470 [54:42<20:44,  1.38it/s]
2|4751|Loss: 0.9044710993766785:  73%|███████▎  | 4751/6470 [54:42<20:57,  1.37it/s]
2|4752|Loss: 1.1821019649505615:  73%|███████▎  | 4751/6470 [54:43<20:57,  1.37it/s]
2|4752|Loss: 1.1821019649505615:  73%|███████▎  | 4752/6470 [54:43<21:02,  1.36it/s]
2|4753|Loss: 0.7670916318893433:  73%|███████▎  | 4752/6470 [54:43<21:02,  1.36it/s]
2|4753|Loss: 0.7670916318893433:  73%|███████▎  | 4753/6470 [54:43<19:51,  1.44it/s]
2|4754|Loss: 0.8271034359931946:  73%|███████▎  | 4753/6470 [54:44<19:51,  1.44it/s]
2|4754|Loss: 0.8271034359931946:  73%|███████▎  | 4754/6470 [54:44<20:23,  1.40it/s]
2|4755|Loss: 1.2112581729888916:  73%|███████▎  | 4754/6470 [54:45<20:23,  1.40it/s]
2|4755|Loss: 1.2112581729888916:  73%|███████▎  | 4755/6470 [54:45<19:55,  1.43it/s]
2|4756|Loss: 0.9370770454406738:  73%|███████▎  | 4755/6470 [54:45<19:55,  1.43it/s]
2|4756|Loss: 0.9370770454406738:  74%|███████▎  | 4756/6470 [54:45<19:14,  1.48it/s]
2|4757|Loss: 0.9101006984710693:  74%|███████▎  | 4756/6470 [54:46<19:14,  1.48it/s]
2|4757|Loss: 0.9101006984710693:  74%|███████▎  | 4757/6470 [54:46<19:37,  1.45it/s]
2|4758|Loss: 0.9565449953079224:  74%|███████▎  | 4757/6470 [54:47<19:37,  1.45it/s]
2|4758|Loss: 0.9565449953079224:  74%|███████▎  | 4758/6470 [54:47<19:25,  1.47it/s]
2|4759|Loss: 0.8182396292686462:  74%|███████▎  | 4758/6470 [54:47<19:25,  1.47it/s]
2|4759|Loss: 0.8182396292686462:  74%|███████▎  | 4759/6470 [54:47<18:16,  1.56it/s]
2|4760|Loss: 0.9588256478309631:  74%|███████▎  | 4759/6470 [54:48<18:16,  1.56it/s]
2|4760|Loss: 0.9588256478309631:  74%|███████▎  | 4760/6470 [54:48<18:48,  1.52it/s]
2|4761|Loss: 0.9861589670181274:  74%|███████▎  | 4760/6470 [54:49<18:48,  1.52it/s]
2|4761|Loss: 0.9861589670181274:  74%|███████▎  | 4761/6470 [54:49<19:18,  1.47it/s]
2|4762|Loss: 0.7881782054901123:  74%|███████▎  | 4761/6470 [54:49<19:18,  1.47it/s]
2|4762|Loss: 0.7881782054901123:  74%|███████▎  | 4762/6470 [54:50<21:01,  1.35it/s]
2|4763|Loss: 0.8267970681190491:  74%|███████▎  | 4762/6470 [54:50<21:01,  1.35it/s]
2|4763|Loss: 0.8267970681190491:  74%|███████▎  | 4763/6470 [54:51<21:41,  1.31it/s]
2|4764|Loss: 0.8033753633499146:  74%|███████▎  | 4763/6470 [54:51<21:41,  1.31it/s]
2|4764|Loss: 0.8033753633499146:  74%|███████▎  | 4764/6470 [54:51<20:30,  1.39it/s]
2|4765|Loss: 0.9109373092651367:  74%|███████▎  | 4764/6470 [54:52<20:30,  1.39it/s]
2|4765|Loss: 0.9109373092651367:  74%|███████▎  | 4765/6470 [54:52<20:21,  1.40it/s]
2|4766|Loss: 0.7185167670249939:  74%|███████▎  | 4765/6470 [54:52<20:21,  1.40it/s]
2|4766|Loss: 0.7185167670249939:  74%|███████▎  | 4766/6470 [54:53<19:44,  1.44it/s]
2|4767|Loss: 0.7993497848510742:  74%|███████▎  | 4766/6470 [54:53<19:44,  1.44it/s]
2|4767|Loss: 0.7993497848510742:  74%|███████▎  | 4767/6470 [54:53<19:35,  1.45it/s]
2|4768|Loss: 0.8264197707176208:  74%|███████▎  | 4767/6470 [54:54<19:35,  1.45it/s]
2|4768|Loss: 0.8264197707176208:  74%|███████▎  | 4768/6470 [54:54<19:14,  1.47it/s]
2|4769|Loss: 0.8384848833084106:  74%|███████▎  | 4768/6470 [54:54<19:14,  1.47it/s]
2|4769|Loss: 0.8384848833084106:  74%|███████▎  | 4769/6470 [54:55<20:36,  1.38it/s]
2|4770|Loss: 0.6924614310264587:  74%|███████▎  | 4769/6470 [54:55<20:36,  1.38it/s]
2|4770|Loss: 0.6924614310264587:  74%|███████▎  | 4770/6470 [54:56<21:34,  1.31it/s]
2|4771|Loss: 0.823710560798645:  74%|███████▎  | 4770/6470 [54:56<21:34,  1.31it/s] 
2|4771|Loss: 0.823710560798645:  74%|███████▎  | 4771/6470 [54:56<22:08,  1.28it/s]
2|4772|Loss: 0.8086604475975037:  74%|███████▎  | 4771/6470 [54:57<22:08,  1.28it/s]
2|4772|Loss: 0.8086604475975037:  74%|███████▍  | 4772/6470 [54:57<22:17,  1.27it/s]
2|4773|Loss: 0.9214791059494019:  74%|███████▍  | 4772/6470 [54:58<22:17,  1.27it/s]
2|4773|Loss: 0.9214791059494019:  74%|███████▍  | 4773/6470 [54:58<21:11,  1.33it/s]
2|4774|Loss: 0.7391059398651123:  74%|███████▍  | 4773/6470 [54:58<21:11,  1.33it/s]
2|4774|Loss: 0.7391059398651123:  74%|███████▍  | 4774/6470 [54:59<20:50,  1.36it/s]
2|4775|Loss: 0.7594665288925171:  74%|███████▍  | 4774/6470 [54:59<20:50,  1.36it/s]
2|4775|Loss: 0.7594665288925171:  74%|███████▍  | 4775/6470 [54:59<20:34,  1.37it/s]
2|4776|Loss: 0.9242775440216064:  74%|███████▍  | 4775/6470 [55:00<20:34,  1.37it/s]
2|4776|Loss: 0.9242775440216064:  74%|███████▍  | 4776/6470 [55:00<20:18,  1.39it/s]
2|4777|Loss: 0.7746363282203674:  74%|███████▍  | 4776/6470 [55:00<20:18,  1.39it/s]
2|4777|Loss: 0.7746363282203674:  74%|███████▍  | 4777/6470 [55:01<21:18,  1.32it/s]
2|4778|Loss: 0.8766098618507385:  74%|███████▍  | 4777/6470 [55:01<21:18,  1.32it/s]
2|4778|Loss: 0.8766098618507385:  74%|███████▍  | 4778/6470 [55:01<19:55,  1.42it/s]
2|4779|Loss: 0.8878208994865417:  74%|███████▍  | 4778/6470 [55:02<19:55,  1.42it/s]
2|4779|Loss: 0.8878208994865417:  74%|███████▍  | 4779/6470 [55:02<20:08,  1.40it/s]
2|4780|Loss: 0.8998327851295471:  74%|███████▍  | 4779/6470 [55:03<20:08,  1.40it/s]
2|4780|Loss: 0.8998327851295471:  74%|███████▍  | 4780/6470 [55:03<20:31,  1.37it/s]
2|4781|Loss: 0.7540576457977295:  74%|███████▍  | 4780/6470 [55:03<20:31,  1.37it/s]
2|4781|Loss: 0.7540576457977295:  74%|███████▍  | 4781/6470 [55:04<20:23,  1.38it/s]
2|4782|Loss: 0.8788139224052429:  74%|███████▍  | 4781/6470 [55:04<20:23,  1.38it/s]
2|4782|Loss: 0.8788139224052429:  74%|███████▍  | 4782/6470 [55:04<19:37,  1.43it/s]
2|4783|Loss: 1.0120292901992798:  74%|███████▍  | 4782/6470 [55:05<19:37,  1.43it/s]
2|4783|Loss: 1.0120292901992798:  74%|███████▍  | 4783/6470 [55:05<19:32,  1.44it/s]
2|4784|Loss: 0.9728567004203796:  74%|███████▍  | 4783/6470 [55:05<19:32,  1.44it/s]
2|4784|Loss: 0.9728567004203796:  74%|███████▍  | 4784/6470 [55:06<19:49,  1.42it/s]
2|4785|Loss: 0.8017930388450623:  74%|███████▍  | 4784/6470 [55:06<19:49,  1.42it/s]
2|4785|Loss: 0.8017930388450623:  74%|███████▍  | 4785/6470 [55:06<20:02,  1.40it/s]
2|4786|Loss: 0.8827545046806335:  74%|███████▍  | 4785/6470 [55:07<20:02,  1.40it/s]
2|4786|Loss: 0.8827545046806335:  74%|███████▍  | 4786/6470 [55:07<20:10,  1.39it/s]
2|4787|Loss: 0.6239181160926819:  74%|███████▍  | 4786/6470 [55:08<20:10,  1.39it/s]
2|4787|Loss: 0.6239181160926819:  74%|███████▍  | 4787/6470 [55:08<21:50,  1.28it/s]
2|4788|Loss: 0.9404214024543762:  74%|███████▍  | 4787/6470 [55:08<21:50,  1.28it/s]
2|4788|Loss: 0.9404214024543762:  74%|███████▍  | 4788/6470 [55:09<21:05,  1.33it/s]
2|4789|Loss: 0.8376380801200867:  74%|███████▍  | 4788/6470 [55:09<21:05,  1.33it/s]
2|4789|Loss: 0.8376380801200867:  74%|███████▍  | 4789/6470 [55:09<21:18,  1.31it/s]
2|4790|Loss: 0.9971874356269836:  74%|███████▍  | 4789/6470 [55:10<21:18,  1.31it/s]
2|4790|Loss: 0.9971874356269836:  74%|███████▍  | 4790/6470 [55:10<21:33,  1.30it/s]
2|4791|Loss: 0.9194990396499634:  74%|███████▍  | 4790/6470 [55:11<21:33,  1.30it/s]
2|4791|Loss: 0.9194990396499634:  74%|███████▍  | 4791/6470 [55:11<20:09,  1.39it/s]
2|4792|Loss: 0.8653432726860046:  74%|███████▍  | 4791/6470 [55:11<20:09,  1.39it/s]
2|4792|Loss: 0.8653432726860046:  74%|███████▍  | 4792/6470 [55:12<20:44,  1.35it/s]
2|4793|Loss: 0.8012887239456177:  74%|███████▍  | 4792/6470 [55:12<20:44,  1.35it/s]
2|4793|Loss: 0.8012887239456177:  74%|███████▍  | 4793/6470 [55:12<20:39,  1.35it/s]
2|4794|Loss: 0.924935519695282:  74%|███████▍  | 4793/6470 [55:13<20:39,  1.35it/s] 
2|4794|Loss: 0.924935519695282:  74%|███████▍  | 4794/6470 [55:13<20:38,  1.35it/s]
2|4795|Loss: 0.9318568110466003:  74%|███████▍  | 4794/6470 [55:14<20:38,  1.35it/s]
2|4795|Loss: 0.9318568110466003:  74%|███████▍  | 4795/6470 [55:14<20:42,  1.35it/s]
2|4796|Loss: 0.8172208070755005:  74%|███████▍  | 4795/6470 [55:14<20:42,  1.35it/s]
2|4796|Loss: 0.8172208070755005:  74%|███████▍  | 4796/6470 [55:15<20:37,  1.35it/s]
2|4797|Loss: 0.7368606925010681:  74%|███████▍  | 4796/6470 [55:15<20:37,  1.35it/s]
2|4797|Loss: 0.7368606925010681:  74%|███████▍  | 4797/6470 [55:15<21:29,  1.30it/s]
2|4798|Loss: 0.6923319101333618:  74%|███████▍  | 4797/6470 [55:16<21:29,  1.30it/s]
2|4798|Loss: 0.6923319101333618:  74%|███████▍  | 4798/6470 [55:16<22:06,  1.26it/s]
2|4799|Loss: 0.7963393926620483:  74%|███████▍  | 4798/6470 [55:17<22:06,  1.26it/s]
2|4799|Loss: 0.7963393926620483:  74%|███████▍  | 4799/6470 [55:17<21:29,  1.30it/s]
2|4800|Loss: 1.0231659412384033:  74%|███████▍  | 4799/6470 [55:17<21:29,  1.30it/s]
2|4800|Loss: 1.0231659412384033:  74%|███████▍  | 4800/6470 [55:18<20:28,  1.36it/s]
2|4801|Loss: 0.9318381547927856:  74%|███████▍  | 4800/6470 [55:18<20:28,  1.36it/s]
2|4801|Loss: 0.9318381547927856:  74%|███████▍  | 4801/6470 [55:18<20:20,  1.37it/s]
2|4802|Loss: 0.7472324371337891:  74%|███████▍  | 4801/6470 [55:19<20:20,  1.37it/s]
2|4802|Loss: 0.7472324371337891:  74%|███████▍  | 4802/6470 [55:19<20:59,  1.32it/s]
2|4803|Loss: 0.7858880162239075:  74%|███████▍  | 4802/6470 [55:20<20:59,  1.32it/s]
2|4803|Loss: 0.7858880162239075:  74%|███████▍  | 4803/6470 [55:20<20:58,  1.32it/s]
2|4804|Loss: 0.7708530426025391:  74%|███████▍  | 4803/6470 [55:20<20:58,  1.32it/s]
2|4804|Loss: 0.7708530426025391:  74%|███████▍  | 4804/6470 [55:21<21:11,  1.31it/s]
2|4805|Loss: 1.0026642084121704:  74%|███████▍  | 4804/6470 [55:21<21:11,  1.31it/s]
2|4805|Loss: 1.0026642084121704:  74%|███████▍  | 4805/6470 [55:22<21:28,  1.29it/s]
2|4806|Loss: 0.955010175704956:  74%|███████▍  | 4805/6470 [55:22<21:28,  1.29it/s] 
2|4806|Loss: 0.955010175704956:  74%|███████▍  | 4806/6470 [55:22<21:03,  1.32it/s]
2|4807|Loss: 0.7798666954040527:  74%|███████▍  | 4806/6470 [55:23<21:03,  1.32it/s]
2|4807|Loss: 0.7798666954040527:  74%|███████▍  | 4807/6470 [55:23<19:59,  1.39it/s]
2|4808|Loss: 0.8498241305351257:  74%|███████▍  | 4807/6470 [55:23<19:59,  1.39it/s]
2|4808|Loss: 0.8498241305351257:  74%|███████▍  | 4808/6470 [55:24<19:41,  1.41it/s]
2|4809|Loss: 1.0204188823699951:  74%|███████▍  | 4808/6470 [55:24<19:41,  1.41it/s]
2|4809|Loss: 1.0204188823699951:  74%|███████▍  | 4809/6470 [55:24<19:16,  1.44it/s]
2|4810|Loss: 1.0074419975280762:  74%|███████▍  | 4809/6470 [55:25<19:16,  1.44it/s]
2|4810|Loss: 1.0074419975280762:  74%|███████▍  | 4810/6470 [55:25<18:48,  1.47it/s]
2|4811|Loss: 0.9091245532035828:  74%|███████▍  | 4810/6470 [55:25<18:48,  1.47it/s]
2|4811|Loss: 0.9091245532035828:  74%|███████▍  | 4811/6470 [55:26<19:04,  1.45it/s]
2|4812|Loss: 1.1140706539154053:  74%|███████▍  | 4811/6470 [55:26<19:04,  1.45it/s]
2|4812|Loss: 1.1140706539154053:  74%|███████▍  | 4812/6470 [55:26<18:50,  1.47it/s]
2|4813|Loss: 0.9324879050254822:  74%|███████▍  | 4812/6470 [55:27<18:50,  1.47it/s]
2|4813|Loss: 0.9324879050254822:  74%|███████▍  | 4813/6470 [55:27<18:30,  1.49it/s]
2|4814|Loss: 0.8654412627220154:  74%|███████▍  | 4813/6470 [55:27<18:30,  1.49it/s]
2|4814|Loss: 0.8654412627220154:  74%|███████▍  | 4814/6470 [55:28<19:32,  1.41it/s]
2|4815|Loss: 0.8623544573783875:  74%|███████▍  | 4814/6470 [55:28<19:32,  1.41it/s]
2|4815|Loss: 0.8623544573783875:  74%|███████▍  | 4815/6470 [55:28<19:39,  1.40it/s]
2|4816|Loss: 0.8438959717750549:  74%|███████▍  | 4815/6470 [55:29<19:39,  1.40it/s]
2|4816|Loss: 0.8438959717750549:  74%|███████▍  | 4816/6470 [55:29<18:23,  1.50it/s]
2|4817|Loss: 0.8453173041343689:  74%|███████▍  | 4816/6470 [55:29<18:23,  1.50it/s]
2|4817|Loss: 0.8453173041343689:  74%|███████▍  | 4817/6470 [55:30<18:35,  1.48it/s]
2|4818|Loss: 0.9009290337562561:  74%|███████▍  | 4817/6470 [55:30<18:35,  1.48it/s]
2|4818|Loss: 0.9009290337562561:  74%|███████▍  | 4818/6470 [55:30<18:03,  1.52it/s]
2|4819|Loss: 0.780210018157959:  74%|███████▍  | 4818/6470 [55:31<18:03,  1.52it/s] 
2|4819|Loss: 0.780210018157959:  74%|███████▍  | 4819/6470 [55:31<18:16,  1.51it/s]
2|4820|Loss: 0.8353200554847717:  74%|███████▍  | 4819/6470 [55:31<18:16,  1.51it/s]
2|4820|Loss: 0.8353200554847717:  74%|███████▍  | 4820/6470 [55:32<16:58,  1.62it/s]
2|4821|Loss: 0.8585273623466492:  74%|███████▍  | 4820/6470 [55:32<16:58,  1.62it/s]
2|4821|Loss: 0.8585273623466492:  75%|███████▍  | 4821/6470 [55:32<17:22,  1.58it/s]
2|4822|Loss: 0.8951352834701538:  75%|███████▍  | 4821/6470 [55:33<17:22,  1.58it/s]
2|4822|Loss: 0.8951352834701538:  75%|███████▍  | 4822/6470 [55:33<17:26,  1.57it/s]
2|4823|Loss: 0.7644526958465576:  75%|███████▍  | 4822/6470 [55:33<17:26,  1.57it/s]
2|4823|Loss: 0.7644526958465576:  75%|███████▍  | 4823/6470 [55:33<17:00,  1.61it/s]
2|4824|Loss: 0.7823140621185303:  75%|███████▍  | 4823/6470 [55:34<17:00,  1.61it/s]
2|4824|Loss: 0.7823140621185303:  75%|███████▍  | 4824/6470 [55:34<16:55,  1.62it/s]
2|4825|Loss: 0.8891295790672302:  75%|███████▍  | 4824/6470 [55:34<16:55,  1.62it/s]
2|4825|Loss: 0.8891295790672302:  75%|███████▍  | 4825/6470 [55:35<16:58,  1.62it/s]
2|4826|Loss: 0.9344334006309509:  75%|███████▍  | 4825/6470 [55:35<16:58,  1.62it/s]
2|4826|Loss: 0.9344334006309509:  75%|███████▍  | 4826/6470 [55:35<17:32,  1.56it/s]
2|4827|Loss: 0.9168003797531128:  75%|███████▍  | 4826/6470 [55:36<17:32,  1.56it/s]
2|4827|Loss: 0.9168003797531128:  75%|███████▍  | 4827/6470 [55:36<18:15,  1.50it/s]
2|4828|Loss: 0.8153400421142578:  75%|███████▍  | 4827/6470 [55:36<18:15,  1.50it/s]
2|4828|Loss: 0.8153400421142578:  75%|███████▍  | 4828/6470 [55:37<17:51,  1.53it/s]
2|4829|Loss: 0.8371313214302063:  75%|███████▍  | 4828/6470 [55:37<17:51,  1.53it/s]
2|4829|Loss: 0.8371313214302063:  75%|███████▍  | 4829/6470 [55:37<19:10,  1.43it/s]
2|4830|Loss: 1.0743279457092285:  75%|███████▍  | 4829/6470 [55:38<19:10,  1.43it/s]
2|4830|Loss: 1.0743279457092285:  75%|███████▍  | 4830/6470 [55:38<18:41,  1.46it/s]
2|4831|Loss: 0.771670937538147:  75%|███████▍  | 4830/6470 [55:39<18:41,  1.46it/s] 
2|4831|Loss: 0.771670937538147:  75%|███████▍  | 4831/6470 [55:39<19:16,  1.42it/s]
2|4832|Loss: 0.8523382544517517:  75%|███████▍  | 4831/6470 [55:39<19:16,  1.42it/s]
2|4832|Loss: 0.8523382544517517:  75%|███████▍  | 4832/6470 [55:39<17:59,  1.52it/s]
2|4833|Loss: 0.8630802035331726:  75%|███████▍  | 4832/6470 [55:40<17:59,  1.52it/s]
2|4833|Loss: 0.8630802035331726:  75%|███████▍  | 4833/6470 [55:40<17:34,  1.55it/s]
2|4834|Loss: 0.9094386696815491:  75%|███████▍  | 4833/6470 [55:40<17:34,  1.55it/s]
2|4834|Loss: 0.9094386696815491:  75%|███████▍  | 4834/6470 [55:41<18:12,  1.50it/s]
2|4835|Loss: 0.7909104228019714:  75%|███████▍  | 4834/6470 [55:41<18:12,  1.50it/s]
2|4835|Loss: 0.7909104228019714:  75%|███████▍  | 4835/6470 [55:41<17:33,  1.55it/s]
2|4836|Loss: 0.8834629654884338:  75%|███████▍  | 4835/6470 [55:42<17:33,  1.55it/s]
2|4836|Loss: 0.8834629654884338:  75%|███████▍  | 4836/6470 [55:42<16:58,  1.60it/s]
2|4837|Loss: 0.8273635506629944:  75%|███████▍  | 4836/6470 [55:42<16:58,  1.60it/s]
2|4837|Loss: 0.8273635506629944:  75%|███████▍  | 4837/6470 [55:43<16:35,  1.64it/s]
2|4838|Loss: 0.775631844997406:  75%|███████▍  | 4837/6470 [55:43<16:35,  1.64it/s] 
2|4838|Loss: 0.775631844997406:  75%|███████▍  | 4838/6470 [55:43<17:21,  1.57it/s]
2|4839|Loss: 1.0079153776168823:  75%|███████▍  | 4838/6470 [55:44<17:21,  1.57it/s]
2|4839|Loss: 1.0079153776168823:  75%|███████▍  | 4839/6470 [55:44<18:06,  1.50it/s]
2|4840|Loss: 0.8757439255714417:  75%|███████▍  | 4839/6470 [55:44<18:06,  1.50it/s]
2|4840|Loss: 0.8757439255714417:  75%|███████▍  | 4840/6470 [55:45<17:58,  1.51it/s]
2|4841|Loss: 0.8265455365180969:  75%|███████▍  | 4840/6470 [55:45<17:58,  1.51it/s]
2|4841|Loss: 0.8265455365180969:  75%|███████▍  | 4841/6470 [55:45<18:18,  1.48it/s]
2|4842|Loss: 0.8473458290100098:  75%|███████▍  | 4841/6470 [55:46<18:18,  1.48it/s]
2|4842|Loss: 0.8473458290100098:  75%|███████▍  | 4842/6470 [55:46<18:32,  1.46it/s]
2|4843|Loss: 0.7980931997299194:  75%|███████▍  | 4842/6470 [55:46<18:32,  1.46it/s]
2|4843|Loss: 0.7980931997299194:  75%|███████▍  | 4843/6470 [55:47<17:26,  1.56it/s]
2|4844|Loss: 0.7976014018058777:  75%|███████▍  | 4843/6470 [55:47<17:26,  1.56it/s]
2|4844|Loss: 0.7976014018058777:  75%|███████▍  | 4844/6470 [55:47<17:45,  1.53it/s]
2|4845|Loss: 0.7518770098686218:  75%|███████▍  | 4844/6470 [55:48<17:45,  1.53it/s]
2|4845|Loss: 0.7518770098686218:  75%|███████▍  | 4845/6470 [55:48<17:13,  1.57it/s]
2|4846|Loss: 0.6202436685562134:  75%|███████▍  | 4845/6470 [55:48<17:13,  1.57it/s]
2|4846|Loss: 0.6202436685562134:  75%|███████▍  | 4846/6470 [55:49<18:58,  1.43it/s]
2|4847|Loss: 0.8015407919883728:  75%|███████▍  | 4846/6470 [55:49<18:58,  1.43it/s]
2|4847|Loss: 0.8015407919883728:  75%|███████▍  | 4847/6470 [55:49<19:35,  1.38it/s]
2|4848|Loss: 0.8901164531707764:  75%|███████▍  | 4847/6470 [55:50<19:35,  1.38it/s]
2|4848|Loss: 0.8901164531707764:  75%|███████▍  | 4848/6470 [55:50<19:45,  1.37it/s]
2|4849|Loss: 0.9347519278526306:  75%|███████▍  | 4848/6470 [55:51<19:45,  1.37it/s]
2|4849|Loss: 0.9347519278526306:  75%|███████▍  | 4849/6470 [55:51<19:14,  1.40it/s]
2|4850|Loss: 0.798072338104248:  75%|███████▍  | 4849/6470 [55:51<19:14,  1.40it/s] 
2|4850|Loss: 0.798072338104248:  75%|███████▍  | 4850/6470 [55:51<18:30,  1.46it/s]
2|4851|Loss: 0.7277129888534546:  75%|███████▍  | 4850/6470 [55:52<18:30,  1.46it/s]
2|4851|Loss: 0.7277129888534546:  75%|███████▍  | 4851/6470 [55:52<17:23,  1.55it/s]
2|4852|Loss: 0.8469353318214417:  75%|███████▍  | 4851/6470 [55:52<17:23,  1.55it/s]
2|4852|Loss: 0.8469353318214417:  75%|███████▍  | 4852/6470 [55:53<17:09,  1.57it/s]
2|4853|Loss: 0.9510196447372437:  75%|███████▍  | 4852/6470 [55:53<17:09,  1.57it/s]
2|4853|Loss: 0.9510196447372437:  75%|███████▌  | 4853/6470 [55:53<17:05,  1.58it/s]
2|4854|Loss: 0.902226448059082:  75%|███████▌  | 4853/6470 [55:54<17:05,  1.58it/s] 
2|4854|Loss: 0.902226448059082:  75%|███████▌  | 4854/6470 [55:54<17:41,  1.52it/s]
2|4855|Loss: 0.9719645977020264:  75%|███████▌  | 4854/6470 [55:54<17:41,  1.52it/s]
2|4855|Loss: 0.9719645977020264:  75%|███████▌  | 4855/6470 [55:55<18:06,  1.49it/s]
2|4856|Loss: 0.7893455028533936:  75%|███████▌  | 4855/6470 [55:55<18:06,  1.49it/s]
2|4856|Loss: 0.7893455028533936:  75%|███████▌  | 4856/6470 [55:55<16:55,  1.59it/s]
2|4857|Loss: 0.8872874975204468:  75%|███████▌  | 4856/6470 [55:56<16:55,  1.59it/s]
2|4857|Loss: 0.8872874975204468:  75%|███████▌  | 4857/6470 [55:56<16:34,  1.62it/s]
2|4858|Loss: 0.8660979270935059:  75%|███████▌  | 4857/6470 [55:56<16:34,  1.62it/s]
2|4858|Loss: 0.8660979270935059:  75%|███████▌  | 4858/6470 [55:56<16:45,  1.60it/s]
2|4859|Loss: 0.8840962052345276:  75%|███████▌  | 4858/6470 [55:57<16:45,  1.60it/s]
2|4859|Loss: 0.8840962052345276:  75%|███████▌  | 4859/6470 [55:57<17:55,  1.50it/s]
2|4860|Loss: 0.8908659219741821:  75%|███████▌  | 4859/6470 [55:58<17:55,  1.50it/s]
2|4860|Loss: 0.8908659219741821:  75%|███████▌  | 4860/6470 [55:58<18:20,  1.46it/s]
2|4861|Loss: 1.0695372819900513:  75%|███████▌  | 4860/6470 [55:58<18:20,  1.46it/s]
2|4861|Loss: 1.0695372819900513:  75%|███████▌  | 4861/6470 [55:59<18:49,  1.43it/s]
2|4862|Loss: 0.8778044581413269:  75%|███████▌  | 4861/6470 [55:59<18:49,  1.43it/s]
2|4862|Loss: 0.8778044581413269:  75%|███████▌  | 4862/6470 [55:59<19:04,  1.40it/s]
2|4863|Loss: 0.8436102867126465:  75%|███████▌  | 4862/6470 [56:00<19:04,  1.40it/s]
2|4863|Loss: 0.8436102867126465:  75%|███████▌  | 4863/6470 [56:00<19:48,  1.35it/s]
2|4864|Loss: 0.7882874011993408:  75%|███████▌  | 4863/6470 [56:01<19:48,  1.35it/s]
2|4864|Loss: 0.7882874011993408:  75%|███████▌  | 4864/6470 [56:01<20:34,  1.30it/s]
2|4865|Loss: 0.9092613458633423:  75%|███████▌  | 4864/6470 [56:02<20:34,  1.30it/s]
2|4865|Loss: 0.9092613458633423:  75%|███████▌  | 4865/6470 [56:02<20:28,  1.31it/s]
2|4866|Loss: 0.9358893036842346:  75%|███████▌  | 4865/6470 [56:02<20:28,  1.31it/s]
2|4866|Loss: 0.9358893036842346:  75%|███████▌  | 4866/6470 [56:03<20:11,  1.32it/s]
2|4867|Loss: 0.7291989922523499:  75%|███████▌  | 4866/6470 [56:03<20:11,  1.32it/s]
2|4867|Loss: 0.7291989922523499:  75%|███████▌  | 4867/6470 [56:03<20:58,  1.27it/s]
2|4868|Loss: 0.9592648148536682:  75%|███████▌  | 4867/6470 [56:04<20:58,  1.27it/s]
2|4868|Loss: 0.9592648148536682:  75%|███████▌  | 4868/6470 [56:04<20:49,  1.28it/s]
2|4869|Loss: 0.8783361315727234:  75%|███████▌  | 4868/6470 [56:05<20:49,  1.28it/s]
2|4869|Loss: 0.8783361315727234:  75%|███████▌  | 4869/6470 [56:05<20:29,  1.30it/s]
2|4870|Loss: 0.8227853178977966:  75%|███████▌  | 4869/6470 [56:05<20:29,  1.30it/s]
2|4870|Loss: 0.8227853178977966:  75%|███████▌  | 4870/6470 [56:06<20:40,  1.29it/s]
2|4871|Loss: 0.9783170819282532:  75%|███████▌  | 4870/6470 [56:06<20:40,  1.29it/s]
2|4871|Loss: 0.9783170819282532:  75%|███████▌  | 4871/6470 [56:06<20:16,  1.31it/s]
2|4872|Loss: 0.9568146467208862:  75%|███████▌  | 4871/6470 [56:07<20:16,  1.31it/s]
2|4872|Loss: 0.9568146467208862:  75%|███████▌  | 4872/6470 [56:07<20:07,  1.32it/s]
2|4873|Loss: 0.8206952810287476:  75%|███████▌  | 4872/6470 [56:08<20:07,  1.32it/s]
2|4873|Loss: 0.8206952810287476:  75%|███████▌  | 4873/6470 [56:08<19:15,  1.38it/s]
2|4874|Loss: 0.7861117720603943:  75%|███████▌  | 4873/6470 [56:08<19:15,  1.38it/s]
2|4874|Loss: 0.7861117720603943:  75%|███████▌  | 4874/6470 [56:08<17:41,  1.50it/s]
2|4875|Loss: 0.9275516867637634:  75%|███████▌  | 4874/6470 [56:09<17:41,  1.50it/s]
2|4875|Loss: 0.9275516867637634:  75%|███████▌  | 4875/6470 [56:09<17:59,  1.48it/s]
2|4876|Loss: 0.8021004796028137:  75%|███████▌  | 4875/6470 [56:09<17:59,  1.48it/s]
2|4876|Loss: 0.8021004796028137:  75%|███████▌  | 4876/6470 [56:10<18:33,  1.43it/s]
2|4877|Loss: 0.8442484140396118:  75%|███████▌  | 4876/6470 [56:10<18:33,  1.43it/s]
2|4877|Loss: 0.8442484140396118:  75%|███████▌  | 4877/6470 [56:10<17:06,  1.55it/s]
2|4878|Loss: 0.8647083044052124:  75%|███████▌  | 4877/6470 [56:11<17:06,  1.55it/s]
2|4878|Loss: 0.8647083044052124:  75%|███████▌  | 4878/6470 [56:11<16:18,  1.63it/s]
2|4879|Loss: 0.8352990746498108:  75%|███████▌  | 4878/6470 [56:11<16:18,  1.63it/s]
2|4879|Loss: 0.8352990746498108:  75%|███████▌  | 4879/6470 [56:12<16:47,  1.58it/s]
2|4880|Loss: 0.9676221013069153:  75%|███████▌  | 4879/6470 [56:12<16:47,  1.58it/s]
2|4880|Loss: 0.9676221013069153:  75%|███████▌  | 4880/6470 [56:12<16:24,  1.62it/s]
2|4881|Loss: 0.9721695184707642:  75%|███████▌  | 4880/6470 [56:12<16:24,  1.62it/s]
2|4881|Loss: 0.9721695184707642:  75%|███████▌  | 4881/6470 [56:13<15:38,  1.69it/s]
2|4882|Loss: 0.8895238637924194:  75%|███████▌  | 4881/6470 [56:13<15:38,  1.69it/s]
2|4882|Loss: 0.8895238637924194:  75%|███████▌  | 4882/6470 [56:13<16:24,  1.61it/s]
2|4883|Loss: 0.8992111086845398:  75%|███████▌  | 4882/6470 [56:14<16:24,  1.61it/s]
2|4883|Loss: 0.8992111086845398:  75%|███████▌  | 4883/6470 [56:14<17:28,  1.51it/s]
2|4884|Loss: 0.9137616157531738:  75%|███████▌  | 4883/6470 [56:14<17:28,  1.51it/s]
2|4884|Loss: 0.9137616157531738:  75%|███████▌  | 4884/6470 [56:15<17:07,  1.54it/s]
2|4885|Loss: 0.8032525777816772:  75%|███████▌  | 4884/6470 [56:15<17:07,  1.54it/s]
2|4885|Loss: 0.8032525777816772:  76%|███████▌  | 4885/6470 [56:15<16:32,  1.60it/s]
2|4886|Loss: 0.8168166875839233:  76%|███████▌  | 4885/6470 [56:16<16:32,  1.60it/s]
2|4886|Loss: 0.8168166875839233:  76%|███████▌  | 4886/6470 [56:16<17:07,  1.54it/s]
2|4887|Loss: 0.8070996403694153:  76%|███████▌  | 4886/6470 [56:16<17:07,  1.54it/s]
2|4887|Loss: 0.8070996403694153:  76%|███████▌  | 4887/6470 [56:17<18:13,  1.45it/s]
2|4888|Loss: 0.7771823406219482:  76%|███████▌  | 4887/6470 [56:17<18:13,  1.45it/s]
2|4888|Loss: 0.7771823406219482:  76%|███████▌  | 4888/6470 [56:17<17:41,  1.49it/s]
2|4889|Loss: 0.9186199903488159:  76%|███████▌  | 4888/6470 [56:18<17:41,  1.49it/s]
2|4889|Loss: 0.9186199903488159:  76%|███████▌  | 4889/6470 [56:18<17:53,  1.47it/s]
2|4890|Loss: 0.7264499068260193:  76%|███████▌  | 4889/6470 [56:19<17:53,  1.47it/s]
2|4890|Loss: 0.7264499068260193:  76%|███████▌  | 4890/6470 [56:19<18:07,  1.45it/s]
2|4891|Loss: 0.7781829237937927:  76%|███████▌  | 4890/6470 [56:19<18:07,  1.45it/s]
2|4891|Loss: 0.7781829237937927:  76%|███████▌  | 4891/6470 [56:20<18:27,  1.43it/s]
2|4892|Loss: 0.8991909027099609:  76%|███████▌  | 4891/6470 [56:20<18:27,  1.43it/s]
2|4892|Loss: 0.8991909027099609:  76%|███████▌  | 4892/6470 [56:20<19:28,  1.35it/s]
2|4893|Loss: 0.9194138646125793:  76%|███████▌  | 4892/6470 [56:21<19:28,  1.35it/s]
2|4893|Loss: 0.9194138646125793:  76%|███████▌  | 4893/6470 [56:21<19:02,  1.38it/s]
2|4894|Loss: 0.8959246873855591:  76%|███████▌  | 4893/6470 [56:22<19:02,  1.38it/s]
2|4894|Loss: 0.8959246873855591:  76%|███████▌  | 4894/6470 [56:22<19:42,  1.33it/s]
2|4895|Loss: 1.0350980758666992:  76%|███████▌  | 4894/6470 [56:22<19:42,  1.33it/s]
2|4895|Loss: 1.0350980758666992:  76%|███████▌  | 4895/6470 [56:23<19:12,  1.37it/s]
2|4896|Loss: 0.7313078045845032:  76%|███████▌  | 4895/6470 [56:23<19:12,  1.37it/s]
2|4896|Loss: 0.7313078045845032:  76%|███████▌  | 4896/6470 [56:23<18:51,  1.39it/s]
2|4897|Loss: 0.6609278917312622:  76%|███████▌  | 4896/6470 [56:24<18:51,  1.39it/s]
2|4897|Loss: 0.6609278917312622:  76%|███████▌  | 4897/6470 [56:24<17:07,  1.53it/s]
2|4898|Loss: 0.780906617641449:  76%|███████▌  | 4897/6470 [56:24<17:07,  1.53it/s] 
2|4898|Loss: 0.780906617641449:  76%|███████▌  | 4898/6470 [56:25<18:12,  1.44it/s]
2|4899|Loss: 0.9123004674911499:  76%|███████▌  | 4898/6470 [56:25<18:12,  1.44it/s]
2|4899|Loss: 0.9123004674911499:  76%|███████▌  | 4899/6470 [56:25<18:05,  1.45it/s]
2|4900|Loss: 0.9433576464653015:  76%|███████▌  | 4899/6470 [56:26<18:05,  1.45it/s]
2|4900|Loss: 0.9433576464653015:  76%|███████▌  | 4900/6470 [56:26<18:21,  1.43it/s]
2|4901|Loss: 1.0040515661239624:  76%|███████▌  | 4900/6470 [56:26<18:21,  1.43it/s]
2|4901|Loss: 1.0040515661239624:  76%|███████▌  | 4901/6470 [56:27<18:31,  1.41it/s]
2|4902|Loss: 0.7363108396530151:  76%|███████▌  | 4901/6470 [56:27<18:31,  1.41it/s]
2|4902|Loss: 0.7363108396530151:  76%|███████▌  | 4902/6470 [56:27<17:57,  1.46it/s]
2|4903|Loss: 0.9495598077774048:  76%|███████▌  | 4902/6470 [56:28<17:57,  1.46it/s]
2|4903|Loss: 0.9495598077774048:  76%|███████▌  | 4903/6470 [56:28<17:08,  1.52it/s]
2|4904|Loss: 0.8746127486228943:  76%|███████▌  | 4903/6470 [56:28<17:08,  1.52it/s]
2|4904|Loss: 0.8746127486228943:  76%|███████▌  | 4904/6470 [56:29<16:48,  1.55it/s]
2|4905|Loss: 0.9669583439826965:  76%|███████▌  | 4904/6470 [56:29<16:48,  1.55it/s]
2|4905|Loss: 0.9669583439826965:  76%|███████▌  | 4905/6470 [56:29<17:42,  1.47it/s]
2|4906|Loss: 0.8985490202903748:  76%|███████▌  | 4905/6470 [56:30<17:42,  1.47it/s]
2|4906|Loss: 0.8985490202903748:  76%|███████▌  | 4906/6470 [56:30<16:30,  1.58it/s]
2|4907|Loss: 0.9048604965209961:  76%|███████▌  | 4906/6470 [56:30<16:30,  1.58it/s]
2|4907|Loss: 0.9048604965209961:  76%|███████▌  | 4907/6470 [56:30<15:04,  1.73it/s]
2|4908|Loss: 0.8325692415237427:  76%|███████▌  | 4907/6470 [56:31<15:04,  1.73it/s]
2|4908|Loss: 0.8325692415237427:  76%|███████▌  | 4908/6470 [56:31<15:46,  1.65it/s]
2|4909|Loss: 0.8582843542098999:  76%|███████▌  | 4908/6470 [56:31<15:46,  1.65it/s]
2|4909|Loss: 0.8582843542098999:  76%|███████▌  | 4909/6470 [56:32<17:39,  1.47it/s]
2|4910|Loss: 0.8837969303131104:  76%|███████▌  | 4909/6470 [56:32<17:39,  1.47it/s]
2|4910|Loss: 0.8837969303131104:  76%|███████▌  | 4910/6470 [56:33<18:18,  1.42it/s]
2|4911|Loss: 0.7903718948364258:  76%|███████▌  | 4910/6470 [56:33<18:18,  1.42it/s]
2|4911|Loss: 0.7903718948364258:  76%|███████▌  | 4911/6470 [56:33<17:49,  1.46it/s]
2|4912|Loss: 0.94413822889328:  76%|███████▌  | 4911/6470 [56:34<17:49,  1.46it/s]  
2|4912|Loss: 0.94413822889328:  76%|███████▌  | 4912/6470 [56:34<18:28,  1.41it/s]
2|4913|Loss: 0.8750710487365723:  76%|███████▌  | 4912/6470 [56:34<18:28,  1.41it/s]
2|4913|Loss: 0.8750710487365723:  76%|███████▌  | 4913/6470 [56:35<17:30,  1.48it/s]
2|4914|Loss: 0.8344709873199463:  76%|███████▌  | 4913/6470 [56:35<17:30,  1.48it/s]
2|4914|Loss: 0.8344709873199463:  76%|███████▌  | 4914/6470 [56:35<17:24,  1.49it/s]
2|4915|Loss: 0.8703886270523071:  76%|███████▌  | 4914/6470 [56:36<17:24,  1.49it/s]
2|4915|Loss: 0.8703886270523071:  76%|███████▌  | 4915/6470 [56:36<17:15,  1.50it/s]
2|4916|Loss: 1.017521858215332:  76%|███████▌  | 4915/6470 [56:36<17:15,  1.50it/s] 
2|4916|Loss: 1.017521858215332:  76%|███████▌  | 4916/6470 [56:37<17:36,  1.47it/s]
2|4917|Loss: 0.8017314672470093:  76%|███████▌  | 4916/6470 [56:37<17:36,  1.47it/s]
2|4917|Loss: 0.8017314672470093:  76%|███████▌  | 4917/6470 [56:37<17:52,  1.45it/s]
2|4918|Loss: 0.792361855506897:  76%|███████▌  | 4917/6470 [56:38<17:52,  1.45it/s] 
2|4918|Loss: 0.792361855506897:  76%|███████▌  | 4918/6470 [56:38<16:51,  1.53it/s]
2|4919|Loss: 0.9521534442901611:  76%|███████▌  | 4918/6470 [56:38<16:51,  1.53it/s]
2|4919|Loss: 0.9521534442901611:  76%|███████▌  | 4919/6470 [56:39<16:54,  1.53it/s]
2|4920|Loss: 0.9010655879974365:  76%|███████▌  | 4919/6470 [56:39<16:54,  1.53it/s]
2|4920|Loss: 0.9010655879974365:  76%|███████▌  | 4920/6470 [56:39<16:55,  1.53it/s]
2|4921|Loss: 0.8441661596298218:  76%|███████▌  | 4920/6470 [56:40<16:55,  1.53it/s]
2|4921|Loss: 0.8441661596298218:  76%|███████▌  | 4921/6470 [56:40<16:51,  1.53it/s]
2|4922|Loss: 0.9974014163017273:  76%|███████▌  | 4921/6470 [56:40<16:51,  1.53it/s]
2|4922|Loss: 0.9974014163017273:  76%|███████▌  | 4922/6470 [56:40<16:47,  1.54it/s]
2|4923|Loss: 0.9620707035064697:  76%|███████▌  | 4922/6470 [56:41<16:47,  1.54it/s]
2|4923|Loss: 0.9620707035064697:  76%|███████▌  | 4923/6470 [56:41<17:37,  1.46it/s]
2|4924|Loss: 0.99018394947052:  76%|███████▌  | 4923/6470 [56:42<17:37,  1.46it/s]  
2|4924|Loss: 0.99018394947052:  76%|███████▌  | 4924/6470 [56:42<17:23,  1.48it/s]
2|4925|Loss: 0.8572606444358826:  76%|███████▌  | 4924/6470 [56:42<17:23,  1.48it/s]
2|4925|Loss: 0.8572606444358826:  76%|███████▌  | 4925/6470 [56:43<17:43,  1.45it/s]
2|4926|Loss: 0.7980666756629944:  76%|███████▌  | 4925/6470 [56:43<17:43,  1.45it/s]
2|4926|Loss: 0.7980666756629944:  76%|███████▌  | 4926/6470 [56:43<17:55,  1.44it/s]
2|4927|Loss: 1.0046272277832031:  76%|███████▌  | 4926/6470 [56:44<17:55,  1.44it/s]
2|4927|Loss: 1.0046272277832031:  76%|███████▌  | 4927/6470 [56:44<18:18,  1.40it/s]
2|4928|Loss: 0.820536732673645:  76%|███████▌  | 4927/6470 [56:44<18:18,  1.40it/s] 
2|4928|Loss: 0.820536732673645:  76%|███████▌  | 4928/6470 [56:45<18:31,  1.39it/s]
2|4929|Loss: 0.8914769291877747:  76%|███████▌  | 4928/6470 [56:45<18:31,  1.39it/s]
2|4929|Loss: 0.8914769291877747:  76%|███████▌  | 4929/6470 [56:46<18:24,  1.39it/s]
2|4930|Loss: 0.8006613254547119:  76%|███████▌  | 4929/6470 [56:46<18:24,  1.39it/s]
2|4930|Loss: 0.8006613254547119:  76%|███████▌  | 4930/6470 [56:46<18:27,  1.39it/s]
2|4931|Loss: 0.837700605392456:  76%|███████▌  | 4930/6470 [56:47<18:27,  1.39it/s] 
2|4931|Loss: 0.837700605392456:  76%|███████▌  | 4931/6470 [56:47<18:53,  1.36it/s]
2|4932|Loss: 0.88523268699646:  76%|███████▌  | 4931/6470 [56:47<18:53,  1.36it/s] 
2|4932|Loss: 0.88523268699646:  76%|███████▌  | 4932/6470 [56:48<19:04,  1.34it/s]
2|4933|Loss: 0.95380038022995:  76%|███████▌  | 4932/6470 [56:48<19:04,  1.34it/s]
2|4933|Loss: 0.95380038022995:  76%|███████▌  | 4933/6470 [56:48<17:43,  1.45it/s]
2|4934|Loss: 0.9950773119926453:  76%|███████▌  | 4933/6470 [56:49<17:43,  1.45it/s]
2|4934|Loss: 0.9950773119926453:  76%|███████▋  | 4934/6470 [56:49<18:03,  1.42it/s]
2|4935|Loss: 0.8586753010749817:  76%|███████▋  | 4934/6470 [56:50<18:03,  1.42it/s]
2|4935|Loss: 0.8586753010749817:  76%|███████▋  | 4935/6470 [56:50<18:11,  1.41it/s]
2|4936|Loss: 0.7669593691825867:  76%|███████▋  | 4935/6470 [56:50<18:11,  1.41it/s]
2|4936|Loss: 0.7669593691825867:  76%|███████▋  | 4936/6470 [56:50<17:09,  1.49it/s]
2|4937|Loss: 0.8785129189491272:  76%|███████▋  | 4936/6470 [56:51<17:09,  1.49it/s]
2|4937|Loss: 0.8785129189491272:  76%|███████▋  | 4937/6470 [56:51<16:49,  1.52it/s]
2|4938|Loss: 0.9236498475074768:  76%|███████▋  | 4937/6470 [56:51<16:49,  1.52it/s]
2|4938|Loss: 0.9236498475074768:  76%|███████▋  | 4938/6470 [56:52<16:14,  1.57it/s]
2|4939|Loss: 0.9623825550079346:  76%|███████▋  | 4938/6470 [56:52<16:14,  1.57it/s]
2|4939|Loss: 0.9623825550079346:  76%|███████▋  | 4939/6470 [56:52<17:19,  1.47it/s]
2|4940|Loss: 0.915490448474884:  76%|███████▋  | 4939/6470 [56:53<17:19,  1.47it/s] 
2|4940|Loss: 0.915490448474884:  76%|███████▋  | 4940/6470 [56:53<18:28,  1.38it/s]
2|4941|Loss: 1.005248785018921:  76%|███████▋  | 4940/6470 [56:54<18:28,  1.38it/s]
2|4941|Loss: 1.005248785018921:  76%|███████▋  | 4941/6470 [56:54<18:42,  1.36it/s]
2|4942|Loss: 0.8420383334159851:  76%|███████▋  | 4941/6470 [56:54<18:42,  1.36it/s]
2|4942|Loss: 0.8420383334159851:  76%|███████▋  | 4942/6470 [56:55<18:48,  1.35it/s]
2|4943|Loss: 0.8937072157859802:  76%|███████▋  | 4942/6470 [56:55<18:48,  1.35it/s]
2|4943|Loss: 0.8937072157859802:  76%|███████▋  | 4943/6470 [56:55<18:14,  1.40it/s]
2|4944|Loss: 0.9302838444709778:  76%|███████▋  | 4943/6470 [56:56<18:14,  1.40it/s]
2|4944|Loss: 0.9302838444709778:  76%|███████▋  | 4944/6470 [56:56<17:38,  1.44it/s]
2|4945|Loss: 0.8473043441772461:  76%|███████▋  | 4944/6470 [56:56<17:38,  1.44it/s]
2|4945|Loss: 0.8473043441772461:  76%|███████▋  | 4945/6470 [56:57<17:46,  1.43it/s]
2|4946|Loss: 0.874439537525177:  76%|███████▋  | 4945/6470 [56:57<17:46,  1.43it/s] 
2|4946|Loss: 0.874439537525177:  76%|███████▋  | 4946/6470 [56:57<18:04,  1.41it/s]
2|4947|Loss: 0.7152448296546936:  76%|███████▋  | 4946/6470 [56:58<18:04,  1.41it/s]
2|4947|Loss: 0.7152448296546936:  76%|███████▋  | 4947/6470 [56:58<18:03,  1.41it/s]
2|4948|Loss: 0.9005577564239502:  76%|███████▋  | 4947/6470 [56:59<18:03,  1.41it/s]
2|4948|Loss: 0.9005577564239502:  76%|███████▋  | 4948/6470 [56:59<17:27,  1.45it/s]
2|4949|Loss: 0.9214455485343933:  76%|███████▋  | 4948/6470 [56:59<17:27,  1.45it/s]
2|4949|Loss: 0.9214455485343933:  76%|███████▋  | 4949/6470 [56:59<16:16,  1.56it/s]
2|4950|Loss: 0.8386520743370056:  76%|███████▋  | 4949/6470 [57:00<16:16,  1.56it/s]
2|4950|Loss: 0.8386520743370056:  77%|███████▋  | 4950/6470 [57:00<17:04,  1.48it/s]
2|4951|Loss: 0.9376440048217773:  77%|███████▋  | 4950/6470 [57:01<17:04,  1.48it/s]
2|4951|Loss: 0.9376440048217773:  77%|███████▋  | 4951/6470 [57:01<17:04,  1.48it/s]
2|4952|Loss: 0.8639459609985352:  77%|███████▋  | 4951/6470 [57:01<17:04,  1.48it/s]
2|4952|Loss: 0.8639459609985352:  77%|███████▋  | 4952/6470 [57:02<17:43,  1.43it/s]
2|4953|Loss: 0.7282686829566956:  77%|███████▋  | 4952/6470 [57:02<17:43,  1.43it/s]
2|4953|Loss: 0.7282686829566956:  77%|███████▋  | 4953/6470 [57:02<18:40,  1.35it/s]
2|4954|Loss: 0.8565858602523804:  77%|███████▋  | 4953/6470 [57:03<18:40,  1.35it/s]
2|4954|Loss: 0.8565858602523804:  77%|███████▋  | 4954/6470 [57:03<19:09,  1.32it/s]
2|4955|Loss: 0.9031548500061035:  77%|███████▋  | 4954/6470 [57:04<19:09,  1.32it/s]
2|4955|Loss: 0.9031548500061035:  77%|███████▋  | 4955/6470 [57:04<19:10,  1.32it/s]
2|4956|Loss: 0.854991614818573:  77%|███████▋  | 4955/6470 [57:04<19:10,  1.32it/s] 
2|4956|Loss: 0.854991614818573:  77%|███████▋  | 4956/6470 [57:05<18:08,  1.39it/s]
2|4957|Loss: 0.8084690570831299:  77%|███████▋  | 4956/6470 [57:05<18:08,  1.39it/s]
2|4957|Loss: 0.8084690570831299:  77%|███████▋  | 4957/6470 [57:05<17:59,  1.40it/s]
2|4958|Loss: 0.8160348534584045:  77%|███████▋  | 4957/6470 [57:06<17:59,  1.40it/s]
2|4958|Loss: 0.8160348534584045:  77%|███████▋  | 4958/6470 [57:06<17:16,  1.46it/s]
2|4959|Loss: 0.9265912771224976:  77%|███████▋  | 4958/6470 [57:06<17:16,  1.46it/s]
2|4959|Loss: 0.9265912771224976:  77%|███████▋  | 4959/6470 [57:07<17:25,  1.45it/s]
2|4960|Loss: 0.9211627244949341:  77%|███████▋  | 4959/6470 [57:07<17:25,  1.45it/s]
2|4960|Loss: 0.9211627244949341:  77%|███████▋  | 4960/6470 [57:07<17:49,  1.41it/s]
2|4961|Loss: 0.7954230904579163:  77%|███████▋  | 4960/6470 [57:08<17:49,  1.41it/s]
2|4961|Loss: 0.7954230904579163:  77%|███████▋  | 4961/6470 [57:08<18:35,  1.35it/s]
2|4962|Loss: 0.8637149930000305:  77%|███████▋  | 4961/6470 [57:09<18:35,  1.35it/s]
2|4962|Loss: 0.8637149930000305:  77%|███████▋  | 4962/6470 [57:09<18:36,  1.35it/s]
2|4963|Loss: 0.8011236190795898:  77%|███████▋  | 4962/6470 [57:09<18:36,  1.35it/s]
2|4963|Loss: 0.8011236190795898:  77%|███████▋  | 4963/6470 [57:10<18:20,  1.37it/s]
2|4964|Loss: 0.8282594680786133:  77%|███████▋  | 4963/6470 [57:10<18:20,  1.37it/s]
2|4964|Loss: 0.8282594680786133:  77%|███████▋  | 4964/6470 [57:10<16:47,  1.50it/s]
2|4965|Loss: 0.9322987794876099:  77%|███████▋  | 4964/6470 [57:10<16:47,  1.50it/s]
2|4965|Loss: 0.9322987794876099:  77%|███████▋  | 4965/6470 [57:11<17:11,  1.46it/s]
2|4966|Loss: 0.9518781304359436:  77%|███████▋  | 4965/6470 [57:11<17:11,  1.46it/s]
2|4966|Loss: 0.9518781304359436:  77%|███████▋  | 4966/6470 [57:12<17:00,  1.47it/s]
2|4967|Loss: 0.9829261898994446:  77%|███████▋  | 4966/6470 [57:12<17:00,  1.47it/s]
2|4967|Loss: 0.9829261898994446:  77%|███████▋  | 4967/6470 [57:12<16:17,  1.54it/s]
2|4968|Loss: 0.8491206765174866:  77%|███████▋  | 4967/6470 [57:12<16:17,  1.54it/s]
2|4968|Loss: 0.8491206765174866:  77%|███████▋  | 4968/6470 [57:13<16:28,  1.52it/s]
2|4969|Loss: 0.918577253818512:  77%|███████▋  | 4968/6470 [57:13<16:28,  1.52it/s] 
2|4969|Loss: 0.918577253818512:  77%|███████▋  | 4969/6470 [57:14<17:36,  1.42it/s]
2|4970|Loss: 0.9360447525978088:  77%|███████▋  | 4969/6470 [57:14<17:36,  1.42it/s]
2|4970|Loss: 0.9360447525978088:  77%|███████▋  | 4970/6470 [57:14<17:58,  1.39it/s]
2|4971|Loss: 0.7746943831443787:  77%|███████▋  | 4970/6470 [57:15<17:58,  1.39it/s]
2|4971|Loss: 0.7746943831443787:  77%|███████▋  | 4971/6470 [57:15<18:02,  1.39it/s]
2|4972|Loss: 0.8726616501808167:  77%|███████▋  | 4971/6470 [57:15<18:02,  1.39it/s]
2|4972|Loss: 0.8726616501808167:  77%|███████▋  | 4972/6470 [57:16<17:24,  1.43it/s]
2|4973|Loss: 0.8899585604667664:  77%|███████▋  | 4972/6470 [57:16<17:24,  1.43it/s]
2|4973|Loss: 0.8899585604667664:  77%|███████▋  | 4973/6470 [57:16<16:24,  1.52it/s]
2|4974|Loss: 0.895101010799408:  77%|███████▋  | 4973/6470 [57:17<16:24,  1.52it/s] 
2|4974|Loss: 0.895101010799408:  77%|███████▋  | 4974/6470 [57:17<16:10,  1.54it/s]
2|4975|Loss: 0.9143809676170349:  77%|███████▋  | 4974/6470 [57:17<16:10,  1.54it/s]
2|4975|Loss: 0.9143809676170349:  77%|███████▋  | 4975/6470 [57:18<16:07,  1.54it/s]
2|4976|Loss: 0.8577592968940735:  77%|███████▋  | 4975/6470 [57:18<16:07,  1.54it/s]
2|4976|Loss: 0.8577592968940735:  77%|███████▋  | 4976/6470 [57:18<15:55,  1.56it/s]
2|4977|Loss: 0.8685203790664673:  77%|███████▋  | 4976/6470 [57:19<15:55,  1.56it/s]
2|4977|Loss: 0.8685203790664673:  77%|███████▋  | 4977/6470 [57:19<15:46,  1.58it/s]
2|4978|Loss: 0.7498764395713806:  77%|███████▋  | 4977/6470 [57:19<15:46,  1.58it/s]
2|4978|Loss: 0.7498764395713806:  77%|███████▋  | 4978/6470 [57:19<16:06,  1.54it/s]
2|4979|Loss: 0.8974955081939697:  77%|███████▋  | 4978/6470 [57:20<16:06,  1.54it/s]
2|4979|Loss: 0.8974955081939697:  77%|███████▋  | 4979/6470 [57:20<16:10,  1.54it/s]
2|4980|Loss: 0.9195096492767334:  77%|███████▋  | 4979/6470 [57:21<16:10,  1.54it/s]
2|4980|Loss: 0.9195096492767334:  77%|███████▋  | 4980/6470 [57:21<16:47,  1.48it/s]
2|4981|Loss: 0.7832391262054443:  77%|███████▋  | 4980/6470 [57:21<16:47,  1.48it/s]
2|4981|Loss: 0.7832391262054443:  77%|███████▋  | 4981/6470 [57:22<17:06,  1.45it/s]
2|4982|Loss: 0.7643795013427734:  77%|███████▋  | 4981/6470 [57:22<17:06,  1.45it/s]
2|4982|Loss: 0.7643795013427734:  77%|███████▋  | 4982/6470 [57:22<16:26,  1.51it/s]
2|4983|Loss: 1.04482102394104:  77%|███████▋  | 4982/6470 [57:23<16:26,  1.51it/s]  
2|4983|Loss: 1.04482102394104:  77%|███████▋  | 4983/6470 [57:23<17:28,  1.42it/s]
2|4984|Loss: 0.8321691751480103:  77%|███████▋  | 4983/6470 [57:23<17:28,  1.42it/s]
2|4984|Loss: 0.8321691751480103:  77%|███████▋  | 4984/6470 [57:24<17:54,  1.38it/s]
2|4985|Loss: 0.9243085384368896:  77%|███████▋  | 4984/6470 [57:24<17:54,  1.38it/s]
2|4985|Loss: 0.9243085384368896:  77%|███████▋  | 4985/6470 [57:24<17:25,  1.42it/s]
2|4986|Loss: 0.9863278865814209:  77%|███████▋  | 4985/6470 [57:25<17:25,  1.42it/s]
2|4986|Loss: 0.9863278865814209:  77%|███████▋  | 4986/6470 [57:25<17:29,  1.41it/s]
2|4987|Loss: 0.8339967727661133:  77%|███████▋  | 4986/6470 [57:25<17:29,  1.41it/s]
2|4987|Loss: 0.8339967727661133:  77%|███████▋  | 4987/6470 [57:26<15:30,  1.59it/s]
2|4988|Loss: 0.8116015195846558:  77%|███████▋  | 4987/6470 [57:26<15:30,  1.59it/s]
2|4988|Loss: 0.8116015195846558:  77%|███████▋  | 4988/6470 [57:26<15:11,  1.63it/s]
2|4989|Loss: 0.8573071956634521:  77%|███████▋  | 4988/6470 [57:27<15:11,  1.63it/s]
2|4989|Loss: 0.8573071956634521:  77%|███████▋  | 4989/6470 [57:27<15:13,  1.62it/s]
2|4990|Loss: 0.709470808506012:  77%|███████▋  | 4989/6470 [57:27<15:13,  1.62it/s] 
2|4990|Loss: 0.709470808506012:  77%|███████▋  | 4990/6470 [57:28<16:12,  1.52it/s]
2|4991|Loss: 0.6690183877944946:  77%|███████▋  | 4990/6470 [57:28<16:12,  1.52it/s]
2|4991|Loss: 0.6690183877944946:  77%|███████▋  | 4991/6470 [57:28<16:58,  1.45it/s]
2|4992|Loss: 0.9006091356277466:  77%|███████▋  | 4991/6470 [57:29<16:58,  1.45it/s]
2|4992|Loss: 0.9006091356277466:  77%|███████▋  | 4992/6470 [57:29<17:26,  1.41it/s]
2|4993|Loss: 0.7141117453575134:  77%|███████▋  | 4992/6470 [57:29<17:26,  1.41it/s]
2|4993|Loss: 0.7141117453575134:  77%|███████▋  | 4993/6470 [57:30<17:37,  1.40it/s]
2|4994|Loss: 0.7585974931716919:  77%|███████▋  | 4993/6470 [57:30<17:37,  1.40it/s]
2|4994|Loss: 0.7585974931716919:  77%|███████▋  | 4994/6470 [57:30<17:07,  1.44it/s]
2|4995|Loss: 0.9022139310836792:  77%|███████▋  | 4994/6470 [57:31<17:07,  1.44it/s]
2|4995|Loss: 0.9022139310836792:  77%|███████▋  | 4995/6470 [57:31<17:21,  1.42it/s]
2|4996|Loss: 0.7519868612289429:  77%|███████▋  | 4995/6470 [57:32<17:21,  1.42it/s]
2|4996|Loss: 0.7519868612289429:  77%|███████▋  | 4996/6470 [57:32<18:29,  1.33it/s]
2|4997|Loss: 0.7701953053474426:  77%|███████▋  | 4996/6470 [57:32<18:29,  1.33it/s]
2|4997|Loss: 0.7701953053474426:  77%|███████▋  | 4997/6470 [57:33<17:54,  1.37it/s]
2|4998|Loss: 0.8155631422996521:  77%|███████▋  | 4997/6470 [57:33<17:54,  1.37it/s]
2|4998|Loss: 0.8155631422996521:  77%|███████▋  | 4998/6470 [57:33<17:32,  1.40it/s]
2|4999|Loss: 0.9118379354476929:  77%|███████▋  | 4998/6470 [57:34<17:32,  1.40it/s]
2|4999|Loss: 0.9118379354476929:  77%|███████▋  | 4999/6470 [57:34<16:24,  1.49it/s]
2|5000|Loss: 0.8818015456199646:  77%|███████▋  | 4999/6470 [57:34<16:24,  1.49it/s]
2|5000|Loss: 0.8818015456199646:  77%|███████▋  | 5000/6470 [57:35<16:22,  1.50it/s]
2|5001|Loss: 0.7945470809936523:  77%|███████▋  | 5000/6470 [57:35<16:22,  1.50it/s]
2|5001|Loss: 0.7945470809936523:  77%|███████▋  | 5001/6470 [57:35<16:40,  1.47it/s]
2|5002|Loss: 0.8221444487571716:  77%|███████▋  | 5001/6470 [57:36<16:40,  1.47it/s]
2|5002|Loss: 0.8221444487571716:  77%|███████▋  | 5002/6470 [57:36<16:29,  1.48it/s]
2|5003|Loss: 0.9510266780853271:  77%|███████▋  | 5002/6470 [57:36<16:29,  1.48it/s]
2|5003|Loss: 0.9510266780853271:  77%|███████▋  | 5003/6470 [57:37<16:13,  1.51it/s]
2|5004|Loss: 0.8791706562042236:  77%|███████▋  | 5003/6470 [57:37<16:13,  1.51it/s]
2|5004|Loss: 0.8791706562042236:  77%|███████▋  | 5004/6470 [57:37<16:03,  1.52it/s]
2|5005|Loss: 0.8794379234313965:  77%|███████▋  | 5004/6470 [57:38<16:03,  1.52it/s]
2|5005|Loss: 0.8794379234313965:  77%|███████▋  | 5005/6470 [57:38<15:01,  1.62it/s]
2|5006|Loss: 0.7192591428756714:  77%|███████▋  | 5005/6470 [57:38<15:01,  1.62it/s]
2|5006|Loss: 0.7192591428756714:  77%|███████▋  | 5006/6470 [57:38<14:10,  1.72it/s]
2|5007|Loss: 0.8410652279853821:  77%|███████▋  | 5006/6470 [57:39<14:10,  1.72it/s]
2|5007|Loss: 0.8410652279853821:  77%|███████▋  | 5007/6470 [57:39<15:22,  1.59it/s]
2|5008|Loss: 1.0648167133331299:  77%|███████▋  | 5007/6470 [57:39<15:22,  1.59it/s]
2|5008|Loss: 1.0648167133331299:  77%|███████▋  | 5008/6470 [57:40<16:11,  1.50it/s]
2|5009|Loss: 0.722195029258728:  77%|███████▋  | 5008/6470 [57:40<16:11,  1.50it/s] 
2|5009|Loss: 0.722195029258728:  77%|███████▋  | 5009/6470 [57:40<15:59,  1.52it/s]
2|5010|Loss: 0.8606474995613098:  77%|███████▋  | 5009/6470 [57:41<15:59,  1.52it/s]
2|5010|Loss: 0.8606474995613098:  77%|███████▋  | 5010/6470 [57:41<16:08,  1.51it/s]
2|5011|Loss: 0.8043093681335449:  77%|███████▋  | 5010/6470 [57:41<16:08,  1.51it/s]
2|5011|Loss: 0.8043093681335449:  77%|███████▋  | 5011/6470 [57:42<16:51,  1.44it/s]
2|5012|Loss: 0.9250077605247498:  77%|███████▋  | 5011/6470 [57:42<16:51,  1.44it/s]
2|5012|Loss: 0.9250077605247498:  77%|███████▋  | 5012/6470 [57:42<16:20,  1.49it/s]
2|5013|Loss: 0.8643024563789368:  77%|███████▋  | 5012/6470 [57:43<16:20,  1.49it/s]
2|5013|Loss: 0.8643024563789368:  77%|███████▋  | 5013/6470 [57:43<16:38,  1.46it/s]
2|5014|Loss: 1.0153613090515137:  77%|███████▋  | 5013/6470 [57:44<16:38,  1.46it/s]
2|5014|Loss: 1.0153613090515137:  77%|███████▋  | 5014/6470 [57:44<17:22,  1.40it/s]
2|5015|Loss: 0.9293287396430969:  77%|███████▋  | 5014/6470 [57:44<17:22,  1.40it/s]
2|5015|Loss: 0.9293287396430969:  78%|███████▊  | 5015/6470 [57:45<17:07,  1.42it/s]
2|5016|Loss: 0.9519315361976624:  78%|███████▊  | 5015/6470 [57:45<17:07,  1.42it/s]
2|5016|Loss: 0.9519315361976624:  78%|███████▊  | 5016/6470 [57:45<16:07,  1.50it/s]
2|5017|Loss: 0.7880361676216125:  78%|███████▊  | 5016/6470 [57:46<16:07,  1.50it/s]
2|5017|Loss: 0.7880361676216125:  78%|███████▊  | 5017/6470 [57:46<16:27,  1.47it/s]
2|5018|Loss: 0.7400555610656738:  78%|███████▊  | 5017/6470 [57:46<16:27,  1.47it/s]
2|5018|Loss: 0.7400555610656738:  78%|███████▊  | 5018/6470 [57:47<16:42,  1.45it/s]
2|5019|Loss: 0.9135894775390625:  78%|███████▊  | 5018/6470 [57:47<16:42,  1.45it/s]
2|5019|Loss: 0.9135894775390625:  78%|███████▊  | 5019/6470 [57:48<18:02,  1.34it/s]
2|5020|Loss: 0.8588201403617859:  78%|███████▊  | 5019/6470 [57:48<18:02,  1.34it/s]
2|5020|Loss: 0.8588201403617859:  78%|███████▊  | 5020/6470 [57:48<17:23,  1.39it/s]
2|5021|Loss: 0.9973818063735962:  78%|███████▊  | 5020/6470 [57:49<17:23,  1.39it/s]
2|5021|Loss: 0.9973818063735962:  78%|███████▊  | 5021/6470 [57:49<17:48,  1.36it/s]
2|5022|Loss: 0.7729530334472656:  78%|███████▊  | 5021/6470 [57:49<17:48,  1.36it/s]
2|5022|Loss: 0.7729530334472656:  78%|███████▊  | 5022/6470 [57:50<17:26,  1.38it/s]
2|5023|Loss: 0.8854259252548218:  78%|███████▊  | 5022/6470 [57:50<17:26,  1.38it/s]
2|5023|Loss: 0.8854259252548218:  78%|███████▊  | 5023/6470 [57:50<17:55,  1.35it/s]
2|5024|Loss: 0.8430621027946472:  78%|███████▊  | 5023/6470 [57:51<17:55,  1.35it/s]
2|5024|Loss: 0.8430621027946472:  78%|███████▊  | 5024/6470 [57:51<18:46,  1.28it/s]
2|5025|Loss: 0.8200623393058777:  78%|███████▊  | 5024/6470 [57:52<18:46,  1.28it/s]
2|5025|Loss: 0.8200623393058777:  78%|███████▊  | 5025/6470 [57:52<18:42,  1.29it/s]
2|5026|Loss: 0.8511343002319336:  78%|███████▊  | 5025/6470 [57:52<18:42,  1.29it/s]
2|5026|Loss: 0.8511343002319336:  78%|███████▊  | 5026/6470 [57:53<17:44,  1.36it/s]
2|5027|Loss: 0.7859752178192139:  78%|███████▊  | 5026/6470 [57:53<17:44,  1.36it/s]
2|5027|Loss: 0.7859752178192139:  78%|███████▊  | 5027/6470 [57:53<16:41,  1.44it/s]
2|5028|Loss: 0.8139010071754456:  78%|███████▊  | 5027/6470 [57:54<16:41,  1.44it/s]
2|5028|Loss: 0.8139010071754456:  78%|███████▊  | 5028/6470 [57:54<18:16,  1.31it/s]
2|5029|Loss: 0.8373205065727234:  78%|███████▊  | 5028/6470 [57:55<18:16,  1.31it/s]
2|5029|Loss: 0.8373205065727234:  78%|███████▊  | 5029/6470 [57:55<18:37,  1.29it/s]
2|5030|Loss: 0.7186647057533264:  78%|███████▊  | 5029/6470 [57:55<18:37,  1.29it/s]
2|5030|Loss: 0.7186647057533264:  78%|███████▊  | 5030/6470 [57:56<18:28,  1.30it/s]
2|5031|Loss: 0.8328303694725037:  78%|███████▊  | 5030/6470 [57:56<18:28,  1.30it/s]
2|5031|Loss: 0.8328303694725037:  78%|███████▊  | 5031/6470 [57:57<19:06,  1.26it/s]
2|5032|Loss: 0.8942157626152039:  78%|███████▊  | 5031/6470 [57:57<19:06,  1.26it/s]
2|5032|Loss: 0.8942157626152039:  78%|███████▊  | 5032/6470 [57:57<19:15,  1.24it/s]
2|5033|Loss: 0.8123303055763245:  78%|███████▊  | 5032/6470 [57:58<19:15,  1.24it/s]
2|5033|Loss: 0.8123303055763245:  78%|███████▊  | 5033/6470 [57:58<17:32,  1.37it/s]
2|5034|Loss: 1.0212122201919556:  78%|███████▊  | 5033/6470 [57:58<17:32,  1.37it/s]
2|5034|Loss: 1.0212122201919556:  78%|███████▊  | 5034/6470 [57:59<16:40,  1.44it/s]
2|5035|Loss: 0.9421769976615906:  78%|███████▊  | 5034/6470 [57:59<16:40,  1.44it/s]
2|5035|Loss: 0.9421769976615906:  78%|███████▊  | 5035/6470 [57:59<15:35,  1.53it/s]
2|5036|Loss: 0.7914767265319824:  78%|███████▊  | 5035/6470 [58:00<15:35,  1.53it/s]
2|5036|Loss: 0.7914767265319824:  78%|███████▊  | 5036/6470 [58:00<16:36,  1.44it/s]
2|5037|Loss: 0.7840282320976257:  78%|███████▊  | 5036/6470 [58:00<16:36,  1.44it/s]
2|5037|Loss: 0.7840282320976257:  78%|███████▊  | 5037/6470 [58:01<17:10,  1.39it/s]
2|5038|Loss: 0.9027549624443054:  78%|███████▊  | 5037/6470 [58:01<17:10,  1.39it/s]
2|5038|Loss: 0.9027549624443054:  78%|███████▊  | 5038/6470 [58:01<16:17,  1.47it/s]
2|5039|Loss: 0.7233497500419617:  78%|███████▊  | 5038/6470 [58:02<16:17,  1.47it/s]
2|5039|Loss: 0.7233497500419617:  78%|███████▊  | 5039/6470 [58:02<17:07,  1.39it/s]
2|5040|Loss: 0.7407078146934509:  78%|███████▊  | 5039/6470 [58:03<17:07,  1.39it/s]
2|5040|Loss: 0.7407078146934509:  78%|███████▊  | 5040/6470 [58:03<17:43,  1.34it/s]
2|5041|Loss: 0.8784310221672058:  78%|███████▊  | 5040/6470 [58:03<17:43,  1.34it/s]
2|5041|Loss: 0.8784310221672058:  78%|███████▊  | 5041/6470 [58:04<17:11,  1.39it/s]
2|5042|Loss: 0.8365632891654968:  78%|███████▊  | 5041/6470 [58:04<17:11,  1.39it/s]
2|5042|Loss: 0.8365632891654968:  78%|███████▊  | 5042/6470 [58:04<16:25,  1.45it/s]
2|5043|Loss: 1.0586861371994019:  78%|███████▊  | 5042/6470 [58:05<16:25,  1.45it/s]
2|5043|Loss: 1.0586861371994019:  78%|███████▊  | 5043/6470 [58:05<16:02,  1.48it/s]
2|5044|Loss: 0.7762638926506042:  78%|███████▊  | 5043/6470 [58:05<16:02,  1.48it/s]
2|5044|Loss: 0.7762638926506042:  78%|███████▊  | 5044/6470 [58:06<16:37,  1.43it/s]
2|5045|Loss: 0.8551583290100098:  78%|███████▊  | 5044/6470 [58:06<16:37,  1.43it/s]
2|5045|Loss: 0.8551583290100098:  78%|███████▊  | 5045/6470 [58:06<16:55,  1.40it/s]
2|5046|Loss: 0.9815111756324768:  78%|███████▊  | 5045/6470 [58:07<16:55,  1.40it/s]
2|5046|Loss: 0.9815111756324768:  78%|███████▊  | 5046/6470 [58:07<17:01,  1.39it/s]
2|5047|Loss: 0.9661002159118652:  78%|███████▊  | 5046/6470 [58:08<17:01,  1.39it/s]
2|5047|Loss: 0.9661002159118652:  78%|███████▊  | 5047/6470 [58:08<16:56,  1.40it/s]
2|5048|Loss: 0.9295430779457092:  78%|███████▊  | 5047/6470 [58:08<16:56,  1.40it/s]
2|5048|Loss: 0.9295430779457092:  78%|███████▊  | 5048/6470 [58:09<17:18,  1.37it/s]
2|5049|Loss: 0.7818078398704529:  78%|███████▊  | 5048/6470 [58:09<17:18,  1.37it/s]
2|5049|Loss: 0.7818078398704529:  78%|███████▊  | 5049/6470 [58:09<16:47,  1.41it/s]
2|5050|Loss: 0.6607162952423096:  78%|███████▊  | 5049/6470 [58:10<16:47,  1.41it/s]
2|5050|Loss: 0.6607162952423096:  78%|███████▊  | 5050/6470 [58:10<17:20,  1.37it/s]
2|5051|Loss: 0.8129257559776306:  78%|███████▊  | 5050/6470 [58:10<17:20,  1.37it/s]
2|5051|Loss: 0.8129257559776306:  78%|███████▊  | 5051/6470 [58:11<16:37,  1.42it/s]
2|5052|Loss: 0.6622628569602966:  78%|███████▊  | 5051/6470 [58:11<16:37,  1.42it/s]
2|5052|Loss: 0.6622628569602966:  78%|███████▊  | 5052/6470 [58:11<16:05,  1.47it/s]
2|5053|Loss: 0.7722356915473938:  78%|███████▊  | 5052/6470 [58:12<16:05,  1.47it/s]
2|5053|Loss: 0.7722356915473938:  78%|███████▊  | 5053/6470 [58:12<16:08,  1.46it/s]
2|5054|Loss: 0.852578341960907:  78%|███████▊  | 5053/6470 [58:12<16:08,  1.46it/s] 
2|5054|Loss: 0.852578341960907:  78%|███████▊  | 5054/6470 [58:13<15:33,  1.52it/s]
2|5055|Loss: 0.9016686677932739:  78%|███████▊  | 5054/6470 [58:13<15:33,  1.52it/s]
2|5055|Loss: 0.9016686677932739:  78%|███████▊  | 5055/6470 [58:13<15:51,  1.49it/s]
2|5056|Loss: 0.9490880370140076:  78%|███████▊  | 5055/6470 [58:14<15:51,  1.49it/s]
2|5056|Loss: 0.9490880370140076:  78%|███████▊  | 5056/6470 [58:14<15:47,  1.49it/s]
2|5057|Loss: 0.8983115553855896:  78%|███████▊  | 5056/6470 [58:14<15:47,  1.49it/s]
2|5057|Loss: 0.8983115553855896:  78%|███████▊  | 5057/6470 [58:15<15:28,  1.52it/s]
2|5058|Loss: 0.895878255367279:  78%|███████▊  | 5057/6470 [58:15<15:28,  1.52it/s] 
2|5058|Loss: 0.895878255367279:  78%|███████▊  | 5058/6470 [58:15<15:39,  1.50it/s]
2|5059|Loss: 0.8056792616844177:  78%|███████▊  | 5058/6470 [58:16<15:39,  1.50it/s]
2|5059|Loss: 0.8056792616844177:  78%|███████▊  | 5059/6470 [58:16<14:05,  1.67it/s]
2|5060|Loss: 0.9246955513954163:  78%|███████▊  | 5059/6470 [58:16<14:05,  1.67it/s]
2|5060|Loss: 0.9246955513954163:  78%|███████▊  | 5060/6470 [58:16<14:04,  1.67it/s]
2|5061|Loss: 0.8009189963340759:  78%|███████▊  | 5060/6470 [58:17<14:04,  1.67it/s]
2|5061|Loss: 0.8009189963340759:  78%|███████▊  | 5061/6470 [58:17<14:51,  1.58it/s]
2|5062|Loss: 0.8436662554740906:  78%|███████▊  | 5061/6470 [58:17<14:51,  1.58it/s]
2|5062|Loss: 0.8436662554740906:  78%|███████▊  | 5062/6470 [58:18<16:22,  1.43it/s]
2|5063|Loss: 0.8830097913742065:  78%|███████▊  | 5062/6470 [58:18<16:22,  1.43it/s]
2|5063|Loss: 0.8830097913742065:  78%|███████▊  | 5063/6470 [58:18<15:43,  1.49it/s]
2|5064|Loss: 0.861260712146759:  78%|███████▊  | 5063/6470 [58:19<15:43,  1.49it/s] 
2|5064|Loss: 0.861260712146759:  78%|███████▊  | 5064/6470 [58:19<15:11,  1.54it/s]
2|5065|Loss: 0.8207272887229919:  78%|███████▊  | 5064/6470 [58:19<15:11,  1.54it/s]
2|5065|Loss: 0.8207272887229919:  78%|███████▊  | 5065/6470 [58:20<15:06,  1.55it/s]
2|5066|Loss: 0.9450942277908325:  78%|███████▊  | 5065/6470 [58:20<15:06,  1.55it/s]
2|5066|Loss: 0.9450942277908325:  78%|███████▊  | 5066/6470 [58:20<14:56,  1.57it/s]
2|5067|Loss: 0.9256424307823181:  78%|███████▊  | 5066/6470 [58:21<14:56,  1.57it/s]
2|5067|Loss: 0.9256424307823181:  78%|███████▊  | 5067/6470 [58:21<15:50,  1.48it/s]
2|5068|Loss: 0.8000489473342896:  78%|███████▊  | 5067/6470 [58:21<15:50,  1.48it/s]
2|5068|Loss: 0.8000489473342896:  78%|███████▊  | 5068/6470 [58:22<15:04,  1.55it/s]
2|5069|Loss: 0.8862029314041138:  78%|███████▊  | 5068/6470 [58:22<15:04,  1.55it/s]
2|5069|Loss: 0.8862029314041138:  78%|███████▊  | 5069/6470 [58:22<16:03,  1.45it/s]
2|5070|Loss: 0.9929081797599792:  78%|███████▊  | 5069/6470 [58:23<16:03,  1.45it/s]
2|5070|Loss: 0.9929081797599792:  78%|███████▊  | 5070/6470 [58:23<15:57,  1.46it/s]
2|5071|Loss: 0.9070430994033813:  78%|███████▊  | 5070/6470 [58:24<15:57,  1.46it/s]
2|5071|Loss: 0.9070430994033813:  78%|███████▊  | 5071/6470 [58:24<15:57,  1.46it/s]
2|5072|Loss: 0.925110936164856:  78%|███████▊  | 5071/6470 [58:24<15:57,  1.46it/s] 
2|5072|Loss: 0.925110936164856:  78%|███████▊  | 5072/6470 [58:25<16:13,  1.44it/s]
2|5073|Loss: 0.8918209075927734:  78%|███████▊  | 5072/6470 [58:25<16:13,  1.44it/s]
2|5073|Loss: 0.8918209075927734:  78%|███████▊  | 5073/6470 [58:25<15:56,  1.46it/s]
2|5074|Loss: 0.8125380873680115:  78%|███████▊  | 5073/6470 [58:26<15:56,  1.46it/s]
2|5074|Loss: 0.8125380873680115:  78%|███████▊  | 5074/6470 [58:26<14:51,  1.57it/s]
2|5075|Loss: 0.9787819385528564:  78%|███████▊  | 5074/6470 [58:26<14:51,  1.57it/s]
2|5075|Loss: 0.9787819385528564:  78%|███████▊  | 5075/6470 [58:26<13:56,  1.67it/s]
2|5076|Loss: 0.8720870018005371:  78%|███████▊  | 5075/6470 [58:27<13:56,  1.67it/s]
2|5076|Loss: 0.8720870018005371:  78%|███████▊  | 5076/6470 [58:27<14:37,  1.59it/s]
2|5077|Loss: 1.0353024005889893:  78%|███████▊  | 5076/6470 [58:27<14:37,  1.59it/s]
2|5077|Loss: 1.0353024005889893:  78%|███████▊  | 5077/6470 [58:28<14:30,  1.60it/s]
2|5078|Loss: 0.8559777736663818:  78%|███████▊  | 5077/6470 [58:28<14:30,  1.60it/s]
2|5078|Loss: 0.8559777736663818:  78%|███████▊  | 5078/6470 [58:28<14:24,  1.61it/s]
2|5079|Loss: 0.8789812922477722:  78%|███████▊  | 5078/6470 [58:29<14:24,  1.61it/s]
2|5079|Loss: 0.8789812922477722:  79%|███████▊  | 5079/6470 [58:29<14:54,  1.56it/s]
2|5080|Loss: 0.7966324090957642:  79%|███████▊  | 5079/6470 [58:29<14:54,  1.56it/s]
2|5080|Loss: 0.7966324090957642:  79%|███████▊  | 5080/6470 [58:30<15:01,  1.54it/s]
2|5081|Loss: 0.9570067524909973:  79%|███████▊  | 5080/6470 [58:30<15:01,  1.54it/s]
2|5081|Loss: 0.9570067524909973:  79%|███████▊  | 5081/6470 [58:30<15:34,  1.49it/s]
2|5082|Loss: 0.7259835004806519:  79%|███████▊  | 5081/6470 [58:31<15:34,  1.49it/s]
2|5082|Loss: 0.7259835004806519:  79%|███████▊  | 5082/6470 [58:31<14:53,  1.55it/s]
2|5083|Loss: 0.8898675441741943:  79%|███████▊  | 5082/6470 [58:31<14:53,  1.55it/s]
2|5083|Loss: 0.8898675441741943:  79%|███████▊  | 5083/6470 [58:32<15:17,  1.51it/s]
2|5084|Loss: 0.8357201814651489:  79%|███████▊  | 5083/6470 [58:32<15:17,  1.51it/s]
2|5084|Loss: 0.8357201814651489:  79%|███████▊  | 5084/6470 [58:32<16:20,  1.41it/s]
2|5085|Loss: 0.8805736303329468:  79%|███████▊  | 5084/6470 [58:33<16:20,  1.41it/s]
2|5085|Loss: 0.8805736303329468:  79%|███████▊  | 5085/6470 [58:33<16:02,  1.44it/s]
2|5086|Loss: 0.7916737198829651:  79%|███████▊  | 5085/6470 [58:33<16:02,  1.44it/s]
2|5086|Loss: 0.7916737198829651:  79%|███████▊  | 5086/6470 [58:34<15:39,  1.47it/s]
2|5087|Loss: 0.7015567421913147:  79%|███████▊  | 5086/6470 [58:34<15:39,  1.47it/s]
2|5087|Loss: 0.7015567421913147:  79%|███████▊  | 5087/6470 [58:34<16:43,  1.38it/s]
2|5088|Loss: 0.8235599994659424:  79%|███████▊  | 5087/6470 [58:35<16:43,  1.38it/s]
2|5088|Loss: 0.8235599994659424:  79%|███████▊  | 5088/6470 [58:35<17:41,  1.30it/s]
2|5089|Loss: 0.828671395778656:  79%|███████▊  | 5088/6470 [58:36<17:41,  1.30it/s] 
2|5089|Loss: 0.828671395778656:  79%|███████▊  | 5089/6470 [58:36<17:08,  1.34it/s]
2|5090|Loss: 0.8739651441574097:  79%|███████▊  | 5089/6470 [58:36<17:08,  1.34it/s]
2|5090|Loss: 0.8739651441574097:  79%|███████▊  | 5090/6470 [58:37<17:13,  1.33it/s]
2|5091|Loss: 0.8231774568557739:  79%|███████▊  | 5090/6470 [58:37<17:13,  1.33it/s]
2|5091|Loss: 0.8231774568557739:  79%|███████▊  | 5091/6470 [58:38<17:01,  1.35it/s]
2|5092|Loss: 0.9924997687339783:  79%|███████▊  | 5091/6470 [58:38<17:01,  1.35it/s]
2|5092|Loss: 0.9924997687339783:  79%|███████▊  | 5092/6470 [58:38<16:17,  1.41it/s]
2|5093|Loss: 0.8247100710868835:  79%|███████▊  | 5092/6470 [58:39<16:17,  1.41it/s]
2|5093|Loss: 0.8247100710868835:  79%|███████▊  | 5093/6470 [58:39<16:38,  1.38it/s]
2|5094|Loss: 0.9055453538894653:  79%|███████▊  | 5093/6470 [58:39<16:38,  1.38it/s]
2|5094|Loss: 0.9055453538894653:  79%|███████▊  | 5094/6470 [58:40<16:21,  1.40it/s]
2|5095|Loss: 0.89373379945755:  79%|███████▊  | 5094/6470 [58:40<16:21,  1.40it/s]  
2|5095|Loss: 0.89373379945755:  79%|███████▊  | 5095/6470 [58:40<16:23,  1.40it/s]
2|5096|Loss: 0.9060544371604919:  79%|███████▊  | 5095/6470 [58:41<16:23,  1.40it/s]
2|5096|Loss: 0.9060544371604919:  79%|███████▉  | 5096/6470 [58:41<16:35,  1.38it/s]
2|5097|Loss: 0.8272817730903625:  79%|███████▉  | 5096/6470 [58:41<16:35,  1.38it/s]
2|5097|Loss: 0.8272817730903625:  79%|███████▉  | 5097/6470 [58:42<15:28,  1.48it/s]
2|5098|Loss: 0.7734712362289429:  79%|███████▉  | 5097/6470 [58:42<15:28,  1.48it/s]
2|5098|Loss: 0.7734712362289429:  79%|███████▉  | 5098/6470 [58:42<15:30,  1.47it/s]
2|5099|Loss: 1.0171805620193481:  79%|███████▉  | 5098/6470 [58:43<15:30,  1.47it/s]
2|5099|Loss: 1.0171805620193481:  79%|███████▉  | 5099/6470 [58:43<15:22,  1.49it/s]
2|5100|Loss: 0.6749429702758789:  79%|███████▉  | 5099/6470 [58:43<15:22,  1.49it/s]
2|5100|Loss: 0.6749429702758789:  79%|███████▉  | 5100/6470 [58:43<14:06,  1.62it/s]
2|5101|Loss: 0.9408839344978333:  79%|███████▉  | 5100/6470 [58:44<14:06,  1.62it/s]
2|5101|Loss: 0.9408839344978333:  79%|███████▉  | 5101/6470 [58:44<14:27,  1.58it/s]
2|5102|Loss: 0.9054873585700989:  79%|███████▉  | 5101/6470 [58:45<14:27,  1.58it/s]
2|5102|Loss: 0.9054873585700989:  79%|███████▉  | 5102/6470 [58:45<14:31,  1.57it/s]
2|5103|Loss: 0.9563375115394592:  79%|███████▉  | 5102/6470 [58:45<14:31,  1.57it/s]
2|5103|Loss: 0.9563375115394592:  79%|███████▉  | 5103/6470 [58:46<15:46,  1.44it/s]
2|5104|Loss: 0.7800824046134949:  79%|███████▉  | 5103/6470 [58:46<15:46,  1.44it/s]
2|5104|Loss: 0.7800824046134949:  79%|███████▉  | 5104/6470 [58:46<16:04,  1.42it/s]
2|5105|Loss: 0.7777913212776184:  79%|███████▉  | 5104/6470 [58:47<16:04,  1.42it/s]
2|5105|Loss: 0.7777913212776184:  79%|███████▉  | 5105/6470 [58:47<15:43,  1.45it/s]
2|5106|Loss: 0.9610049724578857:  79%|███████▉  | 5105/6470 [58:47<15:43,  1.45it/s]
2|5106|Loss: 0.9610049724578857:  79%|███████▉  | 5106/6470 [58:48<15:02,  1.51it/s]
2|5107|Loss: 0.9366433024406433:  79%|███████▉  | 5106/6470 [58:48<15:02,  1.51it/s]
2|5107|Loss: 0.9366433024406433:  79%|███████▉  | 5107/6470 [58:48<15:52,  1.43it/s]
2|5108|Loss: 0.9895812273025513:  79%|███████▉  | 5107/6470 [58:49<15:52,  1.43it/s]
2|5108|Loss: 0.9895812273025513:  79%|███████▉  | 5108/6470 [58:49<14:14,  1.59it/s]
2|5109|Loss: 0.8425534963607788:  79%|███████▉  | 5108/6470 [58:49<14:14,  1.59it/s]
2|5109|Loss: 0.8425534963607788:  79%|███████▉  | 5109/6470 [58:49<13:18,  1.70it/s]
2|5110|Loss: 0.919918417930603:  79%|███████▉  | 5109/6470 [58:50<13:18,  1.70it/s] 
2|5110|Loss: 0.919918417930603:  79%|███████▉  | 5110/6470 [58:50<12:58,  1.75it/s]
2|5111|Loss: 0.8846149444580078:  79%|███████▉  | 5110/6470 [58:50<12:58,  1.75it/s]
2|5111|Loss: 0.8846149444580078:  79%|███████▉  | 5111/6470 [58:51<13:40,  1.66it/s]
2|5112|Loss: 0.8244547843933105:  79%|███████▉  | 5111/6470 [58:51<13:40,  1.66it/s]
2|5112|Loss: 0.8244547843933105:  79%|███████▉  | 5112/6470 [58:51<13:08,  1.72it/s]
2|5113|Loss: 0.9817123413085938:  79%|███████▉  | 5112/6470 [58:51<13:08,  1.72it/s]
2|5113|Loss: 0.9817123413085938:  79%|███████▉  | 5113/6470 [58:52<13:57,  1.62it/s]
2|5114|Loss: 0.8611587882041931:  79%|███████▉  | 5113/6470 [58:52<13:57,  1.62it/s]
2|5114|Loss: 0.8611587882041931:  79%|███████▉  | 5114/6470 [58:52<14:21,  1.57it/s]
2|5115|Loss: 0.8168598413467407:  79%|███████▉  | 5114/6470 [58:53<14:21,  1.57it/s]
2|5115|Loss: 0.8168598413467407:  79%|███████▉  | 5115/6470 [58:53<15:12,  1.49it/s]
2|5116|Loss: 0.9011380076408386:  79%|███████▉  | 5115/6470 [58:54<15:12,  1.49it/s]
2|5116|Loss: 0.9011380076408386:  79%|███████▉  | 5116/6470 [58:54<15:13,  1.48it/s]
2|5117|Loss: 0.8388991951942444:  79%|███████▉  | 5116/6470 [58:54<15:13,  1.48it/s]
2|5117|Loss: 0.8388991951942444:  79%|███████▉  | 5117/6470 [58:54<14:37,  1.54it/s]
2|5118|Loss: 0.7731406092643738:  79%|███████▉  | 5117/6470 [58:55<14:37,  1.54it/s]
2|5118|Loss: 0.7731406092643738:  79%|███████▉  | 5118/6470 [58:55<14:13,  1.58it/s]
2|5119|Loss: 0.8147127628326416:  79%|███████▉  | 5118/6470 [58:55<14:13,  1.58it/s]
2|5119|Loss: 0.8147127628326416:  79%|███████▉  | 5119/6470 [58:56<14:41,  1.53it/s]
2|5120|Loss: 0.8811843991279602:  79%|███████▉  | 5119/6470 [58:56<14:41,  1.53it/s]
2|5120|Loss: 0.8811843991279602:  79%|███████▉  | 5120/6470 [58:57<15:19,  1.47it/s]
2|5121|Loss: 0.8894180655479431:  79%|███████▉  | 5120/6470 [58:57<15:19,  1.47it/s]
2|5121|Loss: 0.8894180655479431:  79%|███████▉  | 5121/6470 [58:57<14:54,  1.51it/s]
2|5122|Loss: 1.147852897644043:  79%|███████▉  | 5121/6470 [58:58<14:54,  1.51it/s] 
2|5122|Loss: 1.147852897644043:  79%|███████▉  | 5122/6470 [58:58<14:39,  1.53it/s]
2|5123|Loss: 0.9958992600440979:  79%|███████▉  | 5122/6470 [58:58<14:39,  1.53it/s]
2|5123|Loss: 0.9958992600440979:  79%|███████▉  | 5123/6470 [58:58<14:33,  1.54it/s]
2|5124|Loss: 1.0255628824234009:  79%|███████▉  | 5123/6470 [58:59<14:33,  1.54it/s]
2|5124|Loss: 1.0255628824234009:  79%|███████▉  | 5124/6470 [58:59<14:49,  1.51it/s]
2|5125|Loss: 0.8780422806739807:  79%|███████▉  | 5124/6470 [59:00<14:49,  1.51it/s]
2|5125|Loss: 0.8780422806739807:  79%|███████▉  | 5125/6470 [59:00<15:08,  1.48it/s]
2|5126|Loss: 0.8446754217147827:  79%|███████▉  | 5125/6470 [59:00<15:08,  1.48it/s]
2|5126|Loss: 0.8446754217147827:  79%|███████▉  | 5126/6470 [59:01<15:18,  1.46it/s]
2|5127|Loss: 0.8252252340316772:  79%|███████▉  | 5126/6470 [59:01<15:18,  1.46it/s]
2|5127|Loss: 0.8252252340316772:  79%|███████▉  | 5127/6470 [59:01<15:06,  1.48it/s]
2|5128|Loss: 0.7476034760475159:  79%|███████▉  | 5127/6470 [59:02<15:06,  1.48it/s]
2|5128|Loss: 0.7476034760475159:  79%|███████▉  | 5128/6470 [59:02<15:13,  1.47it/s]
2|5129|Loss: 0.8867372274398804:  79%|███████▉  | 5128/6470 [59:02<15:13,  1.47it/s]
2|5129|Loss: 0.8867372274398804:  79%|███████▉  | 5129/6470 [59:03<16:09,  1.38it/s]
2|5130|Loss: 0.6858570575714111:  79%|███████▉  | 5129/6470 [59:03<16:09,  1.38it/s]
2|5130|Loss: 0.6858570575714111:  79%|███████▉  | 5130/6470 [59:03<16:17,  1.37it/s]
2|5131|Loss: 0.8131020665168762:  79%|███████▉  | 5130/6470 [59:04<16:17,  1.37it/s]
2|5131|Loss: 0.8131020665168762:  79%|███████▉  | 5131/6470 [59:04<16:36,  1.34it/s]
2|5132|Loss: 0.9256717562675476:  79%|███████▉  | 5131/6470 [59:05<16:36,  1.34it/s]
2|5132|Loss: 0.9256717562675476:  79%|███████▉  | 5132/6470 [59:05<16:34,  1.35it/s]
2|5133|Loss: 0.7459291219711304:  79%|███████▉  | 5132/6470 [59:05<16:34,  1.35it/s]
2|5133|Loss: 0.7459291219711304:  79%|███████▉  | 5133/6470 [59:06<15:59,  1.39it/s]
2|5134|Loss: 0.8620986342430115:  79%|███████▉  | 5133/6470 [59:06<15:59,  1.39it/s]
2|5134|Loss: 0.8620986342430115:  79%|███████▉  | 5134/6470 [59:06<15:54,  1.40it/s]
2|5135|Loss: 0.9479131102561951:  79%|███████▉  | 5134/6470 [59:07<15:54,  1.40it/s]
2|5135|Loss: 0.9479131102561951:  79%|███████▉  | 5135/6470 [59:07<16:22,  1.36it/s]
2|5136|Loss: 0.7609314918518066:  79%|███████▉  | 5135/6470 [59:08<16:22,  1.36it/s]
2|5136|Loss: 0.7609314918518066:  79%|███████▉  | 5136/6470 [59:08<16:31,  1.34it/s]
2|5137|Loss: 0.9683547019958496:  79%|███████▉  | 5136/6470 [59:08<16:31,  1.34it/s]
2|5137|Loss: 0.9683547019958496:  79%|███████▉  | 5137/6470 [59:09<16:17,  1.36it/s]
2|5138|Loss: 0.7933099269866943:  79%|███████▉  | 5137/6470 [59:09<16:17,  1.36it/s]
2|5138|Loss: 0.7933099269866943:  79%|███████▉  | 5138/6470 [59:09<16:02,  1.38it/s]
2|5139|Loss: 0.5643933415412903:  79%|███████▉  | 5138/6470 [59:10<16:02,  1.38it/s]
2|5139|Loss: 0.5643933415412903:  79%|███████▉  | 5139/6470 [59:10<16:51,  1.32it/s]
2|5140|Loss: 0.7003085613250732:  79%|███████▉  | 5139/6470 [59:11<16:51,  1.32it/s]
2|5140|Loss: 0.7003085613250732:  79%|███████▉  | 5140/6470 [59:11<16:53,  1.31it/s]
2|5141|Loss: 1.1003656387329102:  79%|███████▉  | 5140/6470 [59:11<16:53,  1.31it/s]
2|5141|Loss: 1.1003656387329102:  79%|███████▉  | 5141/6470 [59:12<16:05,  1.38it/s]
2|5142|Loss: 0.7187324166297913:  79%|███████▉  | 5141/6470 [59:12<16:05,  1.38it/s]
2|5142|Loss: 0.7187324166297913:  79%|███████▉  | 5142/6470 [59:12<15:57,  1.39it/s]
2|5143|Loss: 0.902515172958374:  79%|███████▉  | 5142/6470 [59:13<15:57,  1.39it/s] 
2|5143|Loss: 0.902515172958374:  79%|███████▉  | 5143/6470 [59:13<16:31,  1.34it/s]
2|5144|Loss: 0.8247209191322327:  79%|███████▉  | 5143/6470 [59:13<16:31,  1.34it/s]
2|5144|Loss: 0.8247209191322327:  80%|███████▉  | 5144/6470 [59:14<16:26,  1.34it/s]
2|5145|Loss: 0.7181006669998169:  80%|███████▉  | 5144/6470 [59:14<16:26,  1.34it/s]
2|5145|Loss: 0.7181006669998169:  80%|███████▉  | 5145/6470 [59:15<16:19,  1.35it/s]
2|5146|Loss: 0.868892252445221:  80%|███████▉  | 5145/6470 [59:15<16:19,  1.35it/s] 
2|5146|Loss: 0.868892252445221:  80%|███████▉  | 5146/6470 [59:15<15:19,  1.44it/s]
2|5147|Loss: 0.7738311886787415:  80%|███████▉  | 5146/6470 [59:16<15:19,  1.44it/s]
2|5147|Loss: 0.7738311886787415:  80%|███████▉  | 5147/6470 [59:16<16:44,  1.32it/s]
2|5148|Loss: 0.7113028168678284:  80%|███████▉  | 5147/6470 [59:16<16:44,  1.32it/s]
2|5148|Loss: 0.7113028168678284:  80%|███████▉  | 5148/6470 [59:17<16:30,  1.33it/s]
2|5149|Loss: 1.1497000455856323:  80%|███████▉  | 5148/6470 [59:17<16:30,  1.33it/s]
2|5149|Loss: 1.1497000455856323:  80%|███████▉  | 5149/6470 [59:17<14:46,  1.49it/s]
2|5150|Loss: 0.8954361081123352:  80%|███████▉  | 5149/6470 [59:18<14:46,  1.49it/s]
2|5150|Loss: 0.8954361081123352:  80%|███████▉  | 5150/6470 [59:18<14:07,  1.56it/s]
2|5151|Loss: 0.9133603572845459:  80%|███████▉  | 5150/6470 [59:18<14:07,  1.56it/s]
2|5151|Loss: 0.9133603572845459:  80%|███████▉  | 5151/6470 [59:18<14:25,  1.52it/s]
2|5152|Loss: 0.8635261654853821:  80%|███████▉  | 5151/6470 [59:19<14:25,  1.52it/s]
2|5152|Loss: 0.8635261654853821:  80%|███████▉  | 5152/6470 [59:19<14:57,  1.47it/s]
2|5153|Loss: 0.8816691637039185:  80%|███████▉  | 5152/6470 [59:20<14:57,  1.47it/s]
2|5153|Loss: 0.8816691637039185:  80%|███████▉  | 5153/6470 [59:20<14:22,  1.53it/s]
2|5154|Loss: 0.9318639039993286:  80%|███████▉  | 5153/6470 [59:20<14:22,  1.53it/s]
2|5154|Loss: 0.9318639039993286:  80%|███████▉  | 5154/6470 [59:20<13:29,  1.63it/s]
2|5155|Loss: 0.831976592540741:  80%|███████▉  | 5154/6470 [59:21<13:29,  1.63it/s] 
2|5155|Loss: 0.831976592540741:  80%|███████▉  | 5155/6470 [59:21<13:07,  1.67it/s]
2|5156|Loss: 0.9908720850944519:  80%|███████▉  | 5155/6470 [59:21<13:07,  1.67it/s]
2|5156|Loss: 0.9908720850944519:  80%|███████▉  | 5156/6470 [59:22<14:21,  1.53it/s]
2|5157|Loss: 0.8355753421783447:  80%|███████▉  | 5156/6470 [59:22<14:21,  1.53it/s]
2|5157|Loss: 0.8355753421783447:  80%|███████▉  | 5157/6470 [59:22<14:33,  1.50it/s]
2|5158|Loss: 0.9755622744560242:  80%|███████▉  | 5157/6470 [59:23<14:33,  1.50it/s]
2|5158|Loss: 0.9755622744560242:  80%|███████▉  | 5158/6470 [59:23<14:44,  1.48it/s]
2|5159|Loss: 0.8240675330162048:  80%|███████▉  | 5158/6470 [59:23<14:44,  1.48it/s]
2|5159|Loss: 0.8240675330162048:  80%|███████▉  | 5159/6470 [59:24<15:12,  1.44it/s]
2|5160|Loss: 0.8618874549865723:  80%|███████▉  | 5159/6470 [59:24<15:12,  1.44it/s]
2|5160|Loss: 0.8618874549865723:  80%|███████▉  | 5160/6470 [59:25<15:01,  1.45it/s]
2|5161|Loss: 0.8979446887969971:  80%|███████▉  | 5160/6470 [59:25<15:01,  1.45it/s]
2|5161|Loss: 0.8979446887969971:  80%|███████▉  | 5161/6470 [59:25<16:06,  1.35it/s]
2|5162|Loss: 0.795729398727417:  80%|███████▉  | 5161/6470 [59:26<16:06,  1.35it/s] 
2|5162|Loss: 0.795729398727417:  80%|███████▉  | 5162/6470 [59:26<16:26,  1.33it/s]
2|5163|Loss: 0.7695238590240479:  80%|███████▉  | 5162/6470 [59:27<16:26,  1.33it/s]
2|5163|Loss: 0.7695238590240479:  80%|███████▉  | 5163/6470 [59:27<16:21,  1.33it/s]
2|5164|Loss: 0.9365198612213135:  80%|███████▉  | 5163/6470 [59:27<16:21,  1.33it/s]
2|5164|Loss: 0.9365198612213135:  80%|███████▉  | 5164/6470 [59:28<15:46,  1.38it/s]
2|5165|Loss: 0.8648930788040161:  80%|███████▉  | 5164/6470 [59:28<15:46,  1.38it/s]
2|5165|Loss: 0.8648930788040161:  80%|███████▉  | 5165/6470 [59:28<15:45,  1.38it/s]
2|5166|Loss: 0.8696458339691162:  80%|███████▉  | 5165/6470 [59:29<15:45,  1.38it/s]
2|5166|Loss: 0.8696458339691162:  80%|███████▉  | 5166/6470 [59:29<15:51,  1.37it/s]
2|5167|Loss: 1.0789700746536255:  80%|███████▉  | 5166/6470 [59:29<15:51,  1.37it/s]
2|5167|Loss: 1.0789700746536255:  80%|███████▉  | 5167/6470 [59:30<15:16,  1.42it/s]
2|5168|Loss: 0.8655458092689514:  80%|███████▉  | 5167/6470 [59:30<15:16,  1.42it/s]
2|5168|Loss: 0.8655458092689514:  80%|███████▉  | 5168/6470 [59:30<14:08,  1.53it/s]
2|5169|Loss: 0.9293575286865234:  80%|███████▉  | 5168/6470 [59:31<14:08,  1.53it/s]
2|5169|Loss: 0.9293575286865234:  80%|███████▉  | 5169/6470 [59:31<14:14,  1.52it/s]
2|5170|Loss: 0.9836928248405457:  80%|███████▉  | 5169/6470 [59:31<14:14,  1.52it/s]
2|5170|Loss: 0.9836928248405457:  80%|███████▉  | 5170/6470 [59:32<14:42,  1.47it/s]
2|5171|Loss: 0.9670851826667786:  80%|███████▉  | 5170/6470 [59:32<14:42,  1.47it/s]
2|5171|Loss: 0.9670851826667786:  80%|███████▉  | 5171/6470 [59:32<14:55,  1.45it/s]
2|5172|Loss: 0.885872483253479:  80%|███████▉  | 5171/6470 [59:33<14:55,  1.45it/s] 
2|5172|Loss: 0.885872483253479:  80%|███████▉  | 5172/6470 [59:33<15:27,  1.40it/s]
2|5173|Loss: 0.8768978714942932:  80%|███████▉  | 5172/6470 [59:34<15:27,  1.40it/s]
2|5173|Loss: 0.8768978714942932:  80%|███████▉  | 5173/6470 [59:34<15:58,  1.35it/s]
2|5174|Loss: 0.80866938829422:  80%|███████▉  | 5173/6470 [59:34<15:58,  1.35it/s]  
2|5174|Loss: 0.80866938829422:  80%|███████▉  | 5174/6470 [59:35<16:38,  1.30it/s]
2|5175|Loss: 0.9141719937324524:  80%|███████▉  | 5174/6470 [59:35<16:38,  1.30it/s]
2|5175|Loss: 0.9141719937324524:  80%|███████▉  | 5175/6470 [59:35<15:58,  1.35it/s]
2|5176|Loss: 0.897376298904419:  80%|███████▉  | 5175/6470 [59:36<15:58,  1.35it/s] 
2|5176|Loss: 0.897376298904419:  80%|████████  | 5176/6470 [59:36<14:20,  1.50it/s]
2|5177|Loss: 0.986640214920044:  80%|████████  | 5176/6470 [59:36<14:20,  1.50it/s]
2|5177|Loss: 0.986640214920044:  80%|████████  | 5177/6470 [59:36<13:34,  1.59it/s]
2|5178|Loss: 0.7897961735725403:  80%|████████  | 5177/6470 [59:37<13:34,  1.59it/s]
2|5178|Loss: 0.7897961735725403:  80%|████████  | 5178/6470 [59:37<14:03,  1.53it/s]
2|5179|Loss: 0.8891230821609497:  80%|████████  | 5178/6470 [59:38<14:03,  1.53it/s]
2|5179|Loss: 0.8891230821609497:  80%|████████  | 5179/6470 [59:38<15:28,  1.39it/s]
2|5180|Loss: 0.9785918593406677:  80%|████████  | 5179/6470 [59:38<15:28,  1.39it/s]
2|5180|Loss: 0.9785918593406677:  80%|████████  | 5180/6470 [59:39<15:13,  1.41it/s]
2|5181|Loss: 0.7165284156799316:  80%|████████  | 5180/6470 [59:39<15:13,  1.41it/s]
2|5181|Loss: 0.7165284156799316:  80%|████████  | 5181/6470 [59:40<15:59,  1.34it/s]
2|5182|Loss: 0.8171250820159912:  80%|████████  | 5181/6470 [59:40<15:59,  1.34it/s]
2|5182|Loss: 0.8171250820159912:  80%|████████  | 5182/6470 [59:40<16:08,  1.33it/s]
2|5183|Loss: 0.7905785441398621:  80%|████████  | 5182/6470 [59:41<16:08,  1.33it/s]
2|5183|Loss: 0.7905785441398621:  80%|████████  | 5183/6470 [59:41<16:04,  1.33it/s]
2|5184|Loss: 0.8144756555557251:  80%|████████  | 5183/6470 [59:41<16:04,  1.33it/s]
2|5184|Loss: 0.8144756555557251:  80%|████████  | 5184/6470 [59:42<15:14,  1.41it/s]
2|5185|Loss: 0.7722259163856506:  80%|████████  | 5184/6470 [59:42<15:14,  1.41it/s]
2|5185|Loss: 0.7722259163856506:  80%|████████  | 5185/6470 [59:42<15:14,  1.41it/s]
2|5186|Loss: 0.8328989148139954:  80%|████████  | 5185/6470 [59:43<15:14,  1.41it/s]
2|5186|Loss: 0.8328989148139954:  80%|████████  | 5186/6470 [59:43<14:20,  1.49it/s]
2|5187|Loss: 0.7207142114639282:  80%|████████  | 5186/6470 [59:43<14:20,  1.49it/s]
2|5187|Loss: 0.7207142114639282:  80%|████████  | 5187/6470 [59:44<15:07,  1.41it/s]
2|5188|Loss: 0.9124497771263123:  80%|████████  | 5187/6470 [59:44<15:07,  1.41it/s]
2|5188|Loss: 0.9124497771263123:  80%|████████  | 5188/6470 [59:44<14:53,  1.43it/s]
2|5189|Loss: 0.8356145620346069:  80%|████████  | 5188/6470 [59:45<14:53,  1.43it/s]
2|5189|Loss: 0.8356145620346069:  80%|████████  | 5189/6470 [59:45<14:57,  1.43it/s]
2|5190|Loss: 0.8346273303031921:  80%|████████  | 5189/6470 [59:46<14:57,  1.43it/s]
2|5190|Loss: 0.8346273303031921:  80%|████████  | 5190/6470 [59:46<15:09,  1.41it/s]
2|5191|Loss: 0.8186458349227905:  80%|████████  | 5190/6470 [59:46<15:09,  1.41it/s]
2|5191|Loss: 0.8186458349227905:  80%|████████  | 5191/6470 [59:46<14:29,  1.47it/s]
2|5192|Loss: 0.7171046733856201:  80%|████████  | 5191/6470 [59:47<14:29,  1.47it/s]
2|5192|Loss: 0.7171046733856201:  80%|████████  | 5192/6470 [59:47<15:48,  1.35it/s]
2|5193|Loss: 0.9577401876449585:  80%|████████  | 5192/6470 [59:48<15:48,  1.35it/s]
2|5193|Loss: 0.9577401876449585:  80%|████████  | 5193/6470 [59:48<16:06,  1.32it/s]
2|5194|Loss: 0.6868106126785278:  80%|████████  | 5193/6470 [59:49<16:06,  1.32it/s]
2|5194|Loss: 0.6868106126785278:  80%|████████  | 5194/6470 [59:49<15:24,  1.38it/s]
2|5195|Loss: 0.8010611534118652:  80%|████████  | 5194/6470 [59:49<15:24,  1.38it/s]
2|5195|Loss: 0.8010611534118652:  80%|████████  | 5195/6470 [59:50<16:13,  1.31it/s]
2|5196|Loss: 1.0065245628356934:  80%|████████  | 5195/6470 [59:50<16:13,  1.31it/s]
2|5196|Loss: 1.0065245628356934:  80%|████████  | 5196/6470 [59:50<15:50,  1.34it/s]
2|5197|Loss: 0.8361423015594482:  80%|████████  | 5196/6470 [59:51<15:50,  1.34it/s]
2|5197|Loss: 0.8361423015594482:  80%|████████  | 5197/6470 [59:51<15:29,  1.37it/s]
2|5198|Loss: 0.9079641103744507:  80%|████████  | 5197/6470 [59:51<15:29,  1.37it/s]
2|5198|Loss: 0.9079641103744507:  80%|████████  | 5198/6470 [59:52<15:17,  1.39it/s]
2|5199|Loss: 0.8573055863380432:  80%|████████  | 5198/6470 [59:52<15:17,  1.39it/s]
2|5199|Loss: 0.8573055863380432:  80%|████████  | 5199/6470 [59:52<15:09,  1.40it/s]
2|5200|Loss: 0.7221003770828247:  80%|████████  | 5199/6470 [59:53<15:09,  1.40it/s]
2|5200|Loss: 0.7221003770828247:  80%|████████  | 5200/6470 [59:53<14:37,  1.45it/s]
2|5201|Loss: 0.9091063737869263:  80%|████████  | 5200/6470 [59:53<14:37,  1.45it/s]
2|5201|Loss: 0.9091063737869263:  80%|████████  | 5201/6470 [59:54<14:44,  1.43it/s]
2|5202|Loss: 0.85874342918396:  80%|████████  | 5201/6470 [59:54<14:44,  1.43it/s]  
2|5202|Loss: 0.85874342918396:  80%|████████  | 5202/6470 [59:54<14:22,  1.47it/s]
2|5203|Loss: 0.8693279027938843:  80%|████████  | 5202/6470 [59:55<14:22,  1.47it/s]
2|5203|Loss: 0.8693279027938843:  80%|████████  | 5203/6470 [59:55<13:59,  1.51it/s]
2|5204|Loss: 0.8300684690475464:  80%|████████  | 5203/6470 [59:55<13:59,  1.51it/s]
2|5204|Loss: 0.8300684690475464:  80%|████████  | 5204/6470 [59:56<13:53,  1.52it/s]
2|5205|Loss: 0.882684588432312:  80%|████████  | 5204/6470 [59:56<13:53,  1.52it/s] 
2|5205|Loss: 0.882684588432312:  80%|████████  | 5205/6470 [59:56<12:51,  1.64it/s]
2|5206|Loss: 0.906301736831665:  80%|████████  | 5205/6470 [59:56<12:51,  1.64it/s]
2|5206|Loss: 0.906301736831665:  80%|████████  | 5206/6470 [59:57<12:07,  1.74it/s]
2|5207|Loss: 0.7329190969467163:  80%|████████  | 5206/6470 [59:57<12:07,  1.74it/s]
2|5207|Loss: 0.7329190969467163:  80%|████████  | 5207/6470 [59:57<12:24,  1.70it/s]
2|5208|Loss: 0.8292599320411682:  80%|████████  | 5207/6470 [59:58<12:24,  1.70it/s]
2|5208|Loss: 0.8292599320411682:  80%|████████  | 5208/6470 [59:58<13:03,  1.61it/s]
2|5209|Loss: 0.8583440184593201:  80%|████████  | 5208/6470 [59:58<13:03,  1.61it/s]
2|5209|Loss: 0.8583440184593201:  81%|████████  | 5209/6470 [59:59<13:46,  1.53it/s]
2|5210|Loss: 1.025604248046875:  81%|████████  | 5209/6470 [59:59<13:46,  1.53it/s] 
2|5210|Loss: 1.025604248046875:  81%|████████  | 5210/6470 [59:59<14:14,  1.47it/s]
2|5211|Loss: 0.9900030493736267:  81%|████████  | 5210/6470 [1:00:00<14:14,  1.47it/s]
2|5211|Loss: 0.9900030493736267:  81%|████████  | 5211/6470 [1:00:00<14:00,  1.50it/s]
2|5212|Loss: 0.7860460877418518:  81%|████████  | 5211/6470 [1:00:00<14:00,  1.50it/s]
2|5212|Loss: 0.7860460877418518:  81%|████████  | 5212/6470 [1:00:01<13:30,  1.55it/s]
2|5213|Loss: 0.8291374444961548:  81%|████████  | 5212/6470 [1:00:01<13:30,  1.55it/s]
2|5213|Loss: 0.8291374444961548:  81%|████████  | 5213/6470 [1:00:02<14:35,  1.44it/s]
2|5214|Loss: 0.7752684950828552:  81%|████████  | 5213/6470 [1:00:02<14:35,  1.44it/s]
2|5214|Loss: 0.7752684950828552:  81%|████████  | 5214/6470 [1:00:02<14:52,  1.41it/s]
2|5215|Loss: 0.9056386947631836:  81%|████████  | 5214/6470 [1:00:03<14:52,  1.41it/s]
2|5215|Loss: 0.9056386947631836:  81%|████████  | 5215/6470 [1:00:03<13:55,  1.50it/s]
2|5216|Loss: 0.9809499979019165:  81%|████████  | 5215/6470 [1:00:03<13:55,  1.50it/s]
2|5216|Loss: 0.9809499979019165:  81%|████████  | 5216/6470 [1:00:03<13:16,  1.57it/s]
2|5217|Loss: 0.9487449526786804:  81%|████████  | 5216/6470 [1:00:04<13:16,  1.57it/s]
2|5217|Loss: 0.9487449526786804:  81%|████████  | 5217/6470 [1:00:04<13:15,  1.57it/s]
2|5218|Loss: 0.9196493625640869:  81%|████████  | 5217/6470 [1:00:04<13:15,  1.57it/s]
2|5218|Loss: 0.9196493625640869:  81%|████████  | 5218/6470 [1:00:05<13:11,  1.58it/s]
2|5219|Loss: 0.8298817873001099:  81%|████████  | 5218/6470 [1:00:05<13:11,  1.58it/s]
2|5219|Loss: 0.8298817873001099:  81%|████████  | 5219/6470 [1:00:05<12:38,  1.65it/s]
2|5220|Loss: 0.7964686155319214:  81%|████████  | 5219/6470 [1:00:06<12:38,  1.65it/s]
2|5220|Loss: 0.7964686155319214:  81%|████████  | 5220/6470 [1:00:06<13:15,  1.57it/s]
2|5221|Loss: 0.8269524574279785:  81%|████████  | 5220/6470 [1:00:06<13:15,  1.57it/s]
2|5221|Loss: 0.8269524574279785:  81%|████████  | 5221/6470 [1:00:07<13:30,  1.54it/s]
2|5222|Loss: 0.7951385378837585:  81%|████████  | 5221/6470 [1:00:07<13:30,  1.54it/s]
2|5222|Loss: 0.7951385378837585:  81%|████████  | 5222/6470 [1:00:07<13:58,  1.49it/s]
2|5223|Loss: 0.8245918154716492:  81%|████████  | 5222/6470 [1:00:08<13:58,  1.49it/s]
2|5223|Loss: 0.8245918154716492:  81%|████████  | 5223/6470 [1:00:08<13:21,  1.56it/s]
2|5224|Loss: 0.9716002941131592:  81%|████████  | 5223/6470 [1:00:08<13:21,  1.56it/s]
2|5224|Loss: 0.9716002941131592:  81%|████████  | 5224/6470 [1:00:09<13:32,  1.53it/s]
2|5225|Loss: 0.9666328430175781:  81%|████████  | 5224/6470 [1:00:09<13:32,  1.53it/s]
2|5225|Loss: 0.9666328430175781:  81%|████████  | 5225/6470 [1:00:09<12:39,  1.64it/s]
2|5226|Loss: 0.8700402975082397:  81%|████████  | 5225/6470 [1:00:09<12:39,  1.64it/s]
2|5226|Loss: 0.8700402975082397:  81%|████████  | 5226/6470 [1:00:10<12:53,  1.61it/s]
2|5227|Loss: 0.8408905267715454:  81%|████████  | 5226/6470 [1:00:10<12:53,  1.61it/s]
2|5227|Loss: 0.8408905267715454:  81%|████████  | 5227/6470 [1:00:10<11:36,  1.78it/s]
2|5228|Loss: 0.8198536038398743:  81%|████████  | 5227/6470 [1:00:10<11:36,  1.78it/s]
2|5228|Loss: 0.8198536038398743:  81%|████████  | 5228/6470 [1:00:11<10:53,  1.90it/s]
2|5229|Loss: 0.9281004071235657:  81%|████████  | 5228/6470 [1:00:11<10:53,  1.90it/s]
2|5229|Loss: 0.9281004071235657:  81%|████████  | 5229/6470 [1:00:11<11:20,  1.82it/s]
2|5230|Loss: 0.8187122941017151:  81%|████████  | 5229/6470 [1:00:12<11:20,  1.82it/s]
2|5230|Loss: 0.8187122941017151:  81%|████████  | 5230/6470 [1:00:12<11:28,  1.80it/s]
2|5231|Loss: 0.8764441609382629:  81%|████████  | 5230/6470 [1:00:12<11:28,  1.80it/s]
2|5231|Loss: 0.8764441609382629:  81%|████████  | 5231/6470 [1:00:13<13:01,  1.59it/s]
2|5232|Loss: 0.7673563957214355:  81%|████████  | 5231/6470 [1:00:13<13:01,  1.59it/s]
2|5232|Loss: 0.7673563957214355:  81%|████████  | 5232/6470 [1:00:13<14:42,  1.40it/s]
2|5233|Loss: 0.9170362949371338:  81%|████████  | 5232/6470 [1:00:14<14:42,  1.40it/s]
2|5233|Loss: 0.9170362949371338:  81%|████████  | 5233/6470 [1:00:14<15:40,  1.32it/s]
2|5234|Loss: 0.9382641315460205:  81%|████████  | 5233/6470 [1:00:15<15:40,  1.32it/s]
2|5234|Loss: 0.9382641315460205:  81%|████████  | 5234/6470 [1:00:15<16:33,  1.24it/s]
2|5235|Loss: 0.9868784546852112:  81%|████████  | 5234/6470 [1:00:16<16:33,  1.24it/s]
2|5235|Loss: 0.9868784546852112:  81%|████████  | 5235/6470 [1:00:16<16:25,  1.25it/s]
2|5236|Loss: 0.8972635865211487:  81%|████████  | 5235/6470 [1:00:16<16:25,  1.25it/s]
2|5236|Loss: 0.8972635865211487:  81%|████████  | 5236/6470 [1:00:17<16:39,  1.23it/s]
2|5237|Loss: 0.8349756598472595:  81%|████████  | 5236/6470 [1:00:17<16:39,  1.23it/s]
2|5237|Loss: 0.8349756598472595:  81%|████████  | 5237/6470 [1:00:18<16:21,  1.26it/s]
2|5238|Loss: 0.9461846947669983:  81%|████████  | 5237/6470 [1:00:18<16:21,  1.26it/s]
2|5238|Loss: 0.9461846947669983:  81%|████████  | 5238/6470 [1:00:18<15:03,  1.36it/s]
2|5239|Loss: 0.8415675759315491:  81%|████████  | 5238/6470 [1:00:19<15:03,  1.36it/s]
2|5239|Loss: 0.8415675759315491:  81%|████████  | 5239/6470 [1:00:19<14:52,  1.38it/s]
2|5240|Loss: 0.8464542627334595:  81%|████████  | 5239/6470 [1:00:19<14:52,  1.38it/s]
2|5240|Loss: 0.8464542627334595:  81%|████████  | 5240/6470 [1:00:20<14:23,  1.42it/s]
2|5241|Loss: 0.8635542988777161:  81%|████████  | 5240/6470 [1:00:20<14:23,  1.42it/s]
2|5241|Loss: 0.8635542988777161:  81%|████████  | 5241/6470 [1:00:20<14:53,  1.38it/s]
2|5242|Loss: 0.8096125721931458:  81%|████████  | 5241/6470 [1:00:21<14:53,  1.38it/s]
2|5242|Loss: 0.8096125721931458:  81%|████████  | 5242/6470 [1:00:21<15:00,  1.36it/s]
2|5243|Loss: 0.8728352785110474:  81%|████████  | 5242/6470 [1:00:21<15:00,  1.36it/s]
2|5243|Loss: 0.8728352785110474:  81%|████████  | 5243/6470 [1:00:22<13:53,  1.47it/s]
2|5244|Loss: 0.9757133722305298:  81%|████████  | 5243/6470 [1:00:22<13:53,  1.47it/s]
2|5244|Loss: 0.9757133722305298:  81%|████████  | 5244/6470 [1:00:22<14:23,  1.42it/s]
2|5245|Loss: 0.8502233624458313:  81%|████████  | 5244/6470 [1:00:23<14:23,  1.42it/s]
2|5245|Loss: 0.8502233624458313:  81%|████████  | 5245/6470 [1:00:23<14:45,  1.38it/s]
2|5246|Loss: 0.7650408148765564:  81%|████████  | 5245/6470 [1:00:24<14:45,  1.38it/s]
2|5246|Loss: 0.7650408148765564:  81%|████████  | 5246/6470 [1:00:24<14:22,  1.42it/s]
2|5247|Loss: 0.9132354855537415:  81%|████████  | 5246/6470 [1:00:24<14:22,  1.42it/s]
2|5247|Loss: 0.9132354855537415:  81%|████████  | 5247/6470 [1:00:25<14:42,  1.39it/s]
2|5248|Loss: 0.8580573797225952:  81%|████████  | 5247/6470 [1:00:25<14:42,  1.39it/s]
2|5248|Loss: 0.8580573797225952:  81%|████████  | 5248/6470 [1:00:25<14:55,  1.36it/s]
2|5249|Loss: 0.9114683866500854:  81%|████████  | 5248/6470 [1:00:26<14:55,  1.36it/s]
2|5249|Loss: 0.9114683866500854:  81%|████████  | 5249/6470 [1:00:26<15:32,  1.31it/s]
2|5250|Loss: 0.8733800053596497:  81%|████████  | 5249/6470 [1:00:27<15:32,  1.31it/s]
2|5250|Loss: 0.8733800053596497:  81%|████████  | 5250/6470 [1:00:27<14:59,  1.36it/s]
2|5251|Loss: 0.8826971650123596:  81%|████████  | 5250/6470 [1:00:27<14:59,  1.36it/s]
2|5251|Loss: 0.8826971650123596:  81%|████████  | 5251/6470 [1:00:27<14:01,  1.45it/s]
2|5252|Loss: 0.9485821723937988:  81%|████████  | 5251/6470 [1:00:28<14:01,  1.45it/s]
2|5252|Loss: 0.9485821723937988:  81%|████████  | 5252/6470 [1:00:28<14:11,  1.43it/s]
2|5253|Loss: 0.912437379360199:  81%|████████  | 5252/6470 [1:00:29<14:11,  1.43it/s] 
2|5253|Loss: 0.912437379360199:  81%|████████  | 5253/6470 [1:00:29<14:17,  1.42it/s]
2|5254|Loss: 0.8071454167366028:  81%|████████  | 5253/6470 [1:00:29<14:17,  1.42it/s]
2|5254|Loss: 0.8071454167366028:  81%|████████  | 5254/6470 [1:00:30<14:28,  1.40it/s]
2|5255|Loss: 0.7451444864273071:  81%|████████  | 5254/6470 [1:00:30<14:28,  1.40it/s]
2|5255|Loss: 0.7451444864273071:  81%|████████  | 5255/6470 [1:00:30<14:35,  1.39it/s]
2|5256|Loss: 0.928457498550415:  81%|████████  | 5255/6470 [1:00:31<14:35,  1.39it/s] 
2|5256|Loss: 0.928457498550415:  81%|████████  | 5256/6470 [1:00:31<14:43,  1.37it/s]
2|5257|Loss: 0.9990707635879517:  81%|████████  | 5256/6470 [1:00:31<14:43,  1.37it/s]
2|5257|Loss: 0.9990707635879517:  81%|████████▏ | 5257/6470 [1:00:32<14:13,  1.42it/s]
2|5258|Loss: 0.8571359515190125:  81%|████████▏ | 5257/6470 [1:00:32<14:13,  1.42it/s]
2|5258|Loss: 0.8571359515190125:  81%|████████▏ | 5258/6470 [1:00:32<14:08,  1.43it/s]
2|5259|Loss: 0.8899511098861694:  81%|████████▏ | 5258/6470 [1:00:33<14:08,  1.43it/s]
2|5259|Loss: 0.8899511098861694:  81%|████████▏ | 5259/6470 [1:00:33<12:56,  1.56it/s]
2|5260|Loss: 0.8763033747673035:  81%|████████▏ | 5259/6470 [1:00:33<12:56,  1.56it/s]
2|5260|Loss: 0.8763033747673035:  81%|████████▏ | 5260/6470 [1:00:33<11:57,  1.69it/s]
2|5261|Loss: 0.7579917907714844:  81%|████████▏ | 5260/6470 [1:00:34<11:57,  1.69it/s]
2|5261|Loss: 0.7579917907714844:  81%|████████▏ | 5261/6470 [1:00:34<11:56,  1.69it/s]
2|5262|Loss: 0.8344410061836243:  81%|████████▏ | 5261/6470 [1:00:34<11:56,  1.69it/s]
2|5262|Loss: 0.8344410061836243:  81%|████████▏ | 5262/6470 [1:00:35<13:06,  1.54it/s]
2|5263|Loss: 0.9710595011711121:  81%|████████▏ | 5262/6470 [1:00:35<13:06,  1.54it/s]
2|5263|Loss: 0.9710595011711121:  81%|████████▏ | 5263/6470 [1:00:35<12:58,  1.55it/s]
2|5264|Loss: 0.8736052513122559:  81%|████████▏ | 5263/6470 [1:00:36<12:58,  1.55it/s]
2|5264|Loss: 0.8736052513122559:  81%|████████▏ | 5264/6470 [1:00:36<13:53,  1.45it/s]
2|5265|Loss: 0.8868311047554016:  81%|████████▏ | 5264/6470 [1:00:37<13:53,  1.45it/s]
2|5265|Loss: 0.8868311047554016:  81%|████████▏ | 5265/6470 [1:00:37<13:51,  1.45it/s]
2|5266|Loss: 0.7320167422294617:  81%|████████▏ | 5265/6470 [1:00:37<13:51,  1.45it/s]
2|5266|Loss: 0.7320167422294617:  81%|████████▏ | 5266/6470 [1:00:38<13:24,  1.50it/s]
2|5267|Loss: 0.8870916366577148:  81%|████████▏ | 5266/6470 [1:00:38<13:24,  1.50it/s]
2|5267|Loss: 0.8870916366577148:  81%|████████▏ | 5267/6470 [1:00:38<13:57,  1.44it/s]
2|5268|Loss: 0.7888619899749756:  81%|████████▏ | 5267/6470 [1:00:39<13:57,  1.44it/s]
2|5268|Loss: 0.7888619899749756:  81%|████████▏ | 5268/6470 [1:00:39<13:15,  1.51it/s]
2|5269|Loss: 0.7833316326141357:  81%|████████▏ | 5268/6470 [1:00:39<13:15,  1.51it/s]
2|5269|Loss: 0.7833316326141357:  81%|████████▏ | 5269/6470 [1:00:39<12:40,  1.58it/s]
2|5270|Loss: 0.8936043977737427:  81%|████████▏ | 5269/6470 [1:00:40<12:40,  1.58it/s]
2|5270|Loss: 0.8936043977737427:  81%|████████▏ | 5270/6470 [1:00:40<13:04,  1.53it/s]
2|5271|Loss: 0.9265077710151672:  81%|████████▏ | 5270/6470 [1:00:41<13:04,  1.53it/s]
2|5271|Loss: 0.9265077710151672:  81%|████████▏ | 5271/6470 [1:00:41<13:20,  1.50it/s]
2|5272|Loss: 0.9199833869934082:  81%|████████▏ | 5271/6470 [1:00:41<13:20,  1.50it/s]
2|5272|Loss: 0.9199833869934082:  81%|████████▏ | 5272/6470 [1:00:42<13:29,  1.48it/s]
2|5273|Loss: 0.8162819147109985:  81%|████████▏ | 5272/6470 [1:00:42<13:29,  1.48it/s]
2|5273|Loss: 0.8162819147109985:  81%|████████▏ | 5273/6470 [1:00:42<14:05,  1.42it/s]
2|5274|Loss: 0.8414177894592285:  81%|████████▏ | 5273/6470 [1:00:43<14:05,  1.42it/s]
2|5274|Loss: 0.8414177894592285:  82%|████████▏ | 5274/6470 [1:00:43<14:40,  1.36it/s]
2|5275|Loss: 0.8942744135856628:  82%|████████▏ | 5274/6470 [1:00:44<14:40,  1.36it/s]
2|5275|Loss: 0.8942744135856628:  82%|████████▏ | 5275/6470 [1:00:44<15:10,  1.31it/s]
2|5276|Loss: 0.72447270154953:  82%|████████▏ | 5275/6470 [1:00:44<15:10,  1.31it/s]  
2|5276|Loss: 0.72447270154953:  82%|████████▏ | 5276/6470 [1:00:44<13:22,  1.49it/s]
2|5277|Loss: 0.9568561911582947:  82%|████████▏ | 5276/6470 [1:00:45<13:22,  1.49it/s]
2|5277|Loss: 0.9568561911582947:  82%|████████▏ | 5277/6470 [1:00:45<12:12,  1.63it/s]
2|5278|Loss: 0.7232434153556824:  82%|████████▏ | 5277/6470 [1:00:45<12:12,  1.63it/s]
2|5278|Loss: 0.7232434153556824:  82%|████████▏ | 5278/6470 [1:00:45<11:34,  1.72it/s]
2|5279|Loss: 1.0506895780563354:  82%|████████▏ | 5278/6470 [1:00:46<11:34,  1.72it/s]
2|5279|Loss: 1.0506895780563354:  82%|████████▏ | 5279/6470 [1:00:46<11:36,  1.71it/s]
2|5280|Loss: 0.8144105672836304:  82%|████████▏ | 5279/6470 [1:00:46<11:36,  1.71it/s]
2|5280|Loss: 0.8144105672836304:  82%|████████▏ | 5280/6470 [1:00:46<10:33,  1.88it/s]
2|5281|Loss: 0.9259656667709351:  82%|████████▏ | 5280/6470 [1:00:47<10:33,  1.88it/s]
2|5281|Loss: 0.9259656667709351:  82%|████████▏ | 5281/6470 [1:00:47<10:15,  1.93it/s]
2|5282|Loss: 0.8188583850860596:  82%|████████▏ | 5281/6470 [1:00:47<10:15,  1.93it/s]
2|5282|Loss: 0.8188583850860596:  82%|████████▏ | 5282/6470 [1:00:48<11:41,  1.69it/s]
2|5283|Loss: 0.823896586894989:  82%|████████▏ | 5282/6470 [1:00:48<11:41,  1.69it/s] 
2|5283|Loss: 0.823896586894989:  82%|████████▏ | 5283/6470 [1:00:48<12:05,  1.64it/s]
2|5284|Loss: 0.8460009098052979:  82%|████████▏ | 5283/6470 [1:00:49<12:05,  1.64it/s]
2|5284|Loss: 0.8460009098052979:  82%|████████▏ | 5284/6470 [1:00:49<11:46,  1.68it/s]
2|5285|Loss: 0.7351577877998352:  82%|████████▏ | 5284/6470 [1:00:49<11:46,  1.68it/s]
2|5285|Loss: 0.7351577877998352:  82%|████████▏ | 5285/6470 [1:00:49<10:28,  1.88it/s]
2|5286|Loss: 0.8979054093360901:  82%|████████▏ | 5285/6470 [1:00:50<10:28,  1.88it/s]
2|5286|Loss: 0.8979054093360901:  82%|████████▏ | 5286/6470 [1:00:50<11:13,  1.76it/s]
2|5287|Loss: 0.8350785374641418:  82%|████████▏ | 5286/6470 [1:00:50<11:13,  1.76it/s]
2|5287|Loss: 0.8350785374641418:  82%|████████▏ | 5287/6470 [1:00:51<12:08,  1.62it/s]
2|5288|Loss: 0.9575527906417847:  82%|████████▏ | 5287/6470 [1:00:51<12:08,  1.62it/s]
2|5288|Loss: 0.9575527906417847:  82%|████████▏ | 5288/6470 [1:00:51<12:52,  1.53it/s]
2|5289|Loss: 0.8039939403533936:  82%|████████▏ | 5288/6470 [1:00:52<12:52,  1.53it/s]
2|5289|Loss: 0.8039939403533936:  82%|████████▏ | 5289/6470 [1:00:52<12:37,  1.56it/s]
2|5290|Loss: 0.7346702814102173:  82%|████████▏ | 5289/6470 [1:00:52<12:37,  1.56it/s]
2|5290|Loss: 0.7346702814102173:  82%|████████▏ | 5290/6470 [1:00:53<12:05,  1.63it/s]
2|5291|Loss: 0.7390539646148682:  82%|████████▏ | 5290/6470 [1:00:53<12:05,  1.63it/s]
2|5291|Loss: 0.7390539646148682:  82%|████████▏ | 5291/6470 [1:00:53<11:50,  1.66it/s]
2|5292|Loss: 0.643860936164856:  82%|████████▏ | 5291/6470 [1:00:54<11:50,  1.66it/s] 
2|5292|Loss: 0.643860936164856:  82%|████████▏ | 5292/6470 [1:00:54<12:58,  1.51it/s]
2|5293|Loss: 0.9216045141220093:  82%|████████▏ | 5292/6470 [1:00:54<12:58,  1.51it/s]
2|5293|Loss: 0.9216045141220093:  82%|████████▏ | 5293/6470 [1:00:55<14:09,  1.38it/s]
2|5294|Loss: 0.8902086019515991:  82%|████████▏ | 5293/6470 [1:00:55<14:09,  1.38it/s]
2|5294|Loss: 0.8902086019515991:  82%|████████▏ | 5294/6470 [1:00:55<13:23,  1.46it/s]
2|5295|Loss: 0.8009689450263977:  82%|████████▏ | 5294/6470 [1:00:56<13:23,  1.46it/s]
2|5295|Loss: 0.8009689450263977:  82%|████████▏ | 5295/6470 [1:00:56<12:39,  1.55it/s]
2|5296|Loss: 0.89043790102005:  82%|████████▏ | 5295/6470 [1:00:56<12:39,  1.55it/s]  
2|5296|Loss: 0.89043790102005:  82%|████████▏ | 5296/6470 [1:00:57<13:33,  1.44it/s]
2|5297|Loss: 0.8694406151771545:  82%|████████▏ | 5296/6470 [1:00:57<13:33,  1.44it/s]
2|5297|Loss: 0.8694406151771545:  82%|████████▏ | 5297/6470 [1:00:57<13:49,  1.41it/s]
2|5298|Loss: 0.818751335144043:  82%|████████▏ | 5297/6470 [1:00:58<13:49,  1.41it/s] 
2|5298|Loss: 0.818751335144043:  82%|████████▏ | 5298/6470 [1:00:58<14:28,  1.35it/s]
2|5299|Loss: 0.7982874512672424:  82%|████████▏ | 5298/6470 [1:00:59<14:28,  1.35it/s]
2|5299|Loss: 0.7982874512672424:  82%|████████▏ | 5299/6470 [1:00:59<14:27,  1.35it/s]
2|5300|Loss: 0.8865101337432861:  82%|████████▏ | 5299/6470 [1:00:59<14:27,  1.35it/s]
2|5300|Loss: 0.8865101337432861:  82%|████████▏ | 5300/6470 [1:01:00<13:38,  1.43it/s]
2|5301|Loss: 0.9495904445648193:  82%|████████▏ | 5300/6470 [1:01:00<13:38,  1.43it/s]
2|5301|Loss: 0.9495904445648193:  82%|████████▏ | 5301/6470 [1:01:00<13:42,  1.42it/s]
2|5302|Loss: 0.8625720143318176:  82%|████████▏ | 5301/6470 [1:01:01<13:42,  1.42it/s]
2|5302|Loss: 0.8625720143318176:  82%|████████▏ | 5302/6470 [1:01:01<13:56,  1.40it/s]
2|5303|Loss: 0.9635614156723022:  82%|████████▏ | 5302/6470 [1:01:01<13:56,  1.40it/s]
2|5303|Loss: 0.9635614156723022:  82%|████████▏ | 5303/6470 [1:01:02<13:32,  1.44it/s]
2|5304|Loss: 0.9048012495040894:  82%|████████▏ | 5303/6470 [1:01:02<13:32,  1.44it/s]
2|5304|Loss: 0.9048012495040894:  82%|████████▏ | 5304/6470 [1:01:02<13:41,  1.42it/s]
2|5305|Loss: 0.7452855110168457:  82%|████████▏ | 5304/6470 [1:01:03<13:41,  1.42it/s]
2|5305|Loss: 0.7452855110168457:  82%|████████▏ | 5305/6470 [1:01:03<14:34,  1.33it/s]
2|5306|Loss: 1.0971180200576782:  82%|████████▏ | 5305/6470 [1:01:04<14:34,  1.33it/s]
2|5306|Loss: 1.0971180200576782:  82%|████████▏ | 5306/6470 [1:01:04<14:30,  1.34it/s]
2|5307|Loss: 0.579379141330719:  82%|████████▏ | 5306/6470 [1:01:05<14:30,  1.34it/s] 
2|5307|Loss: 0.579379141330719:  82%|████████▏ | 5307/6470 [1:01:05<17:40,  1.10it/s]
2|5308|Loss: 0.7625756859779358:  82%|████████▏ | 5307/6470 [1:01:06<17:40,  1.10it/s]
2|5308|Loss: 0.7625756859779358:  82%|████████▏ | 5308/6470 [1:01:06<17:21,  1.12it/s]
2|5309|Loss: 0.8348978161811829:  82%|████████▏ | 5308/6470 [1:01:07<17:21,  1.12it/s]
2|5309|Loss: 0.8348978161811829:  82%|████████▏ | 5309/6470 [1:01:07<16:27,  1.18it/s]
2|5310|Loss: 0.8514019250869751:  82%|████████▏ | 5309/6470 [1:01:07<16:27,  1.18it/s]
2|5310|Loss: 0.8514019250869751:  82%|████████▏ | 5310/6470 [1:01:08<15:39,  1.23it/s]
2|5311|Loss: 0.8549360632896423:  82%|████████▏ | 5310/6470 [1:01:08<15:39,  1.23it/s]
2|5311|Loss: 0.8549360632896423:  82%|████████▏ | 5311/6470 [1:01:08<14:45,  1.31it/s]
2|5312|Loss: 0.742685854434967:  82%|████████▏ | 5311/6470 [1:01:09<14:45,  1.31it/s] 
2|5312|Loss: 0.742685854434967:  82%|████████▏ | 5312/6470 [1:01:09<13:30,  1.43it/s]
2|5313|Loss: 0.8737428188323975:  82%|████████▏ | 5312/6470 [1:01:09<13:30,  1.43it/s]
2|5313|Loss: 0.8737428188323975:  82%|████████▏ | 5313/6470 [1:01:10<12:59,  1.48it/s]
2|5314|Loss: 0.8749427199363708:  82%|████████▏ | 5313/6470 [1:01:10<12:59,  1.48it/s]
2|5314|Loss: 0.8749427199363708:  82%|████████▏ | 5314/6470 [1:01:10<12:48,  1.50it/s]
2|5315|Loss: 0.726643443107605:  82%|████████▏ | 5314/6470 [1:01:11<12:48,  1.50it/s] 
2|5315|Loss: 0.726643443107605:  82%|████████▏ | 5315/6470 [1:01:11<13:48,  1.39it/s]
2|5316|Loss: 0.9893030524253845:  82%|████████▏ | 5315/6470 [1:01:11<13:48,  1.39it/s]
2|5316|Loss: 0.9893030524253845:  82%|████████▏ | 5316/6470 [1:01:12<14:05,  1.37it/s]
2|5317|Loss: 0.7010707259178162:  82%|████████▏ | 5316/6470 [1:01:12<14:05,  1.37it/s]
2|5317|Loss: 0.7010707259178162:  82%|████████▏ | 5317/6470 [1:01:12<13:22,  1.44it/s]
2|5318|Loss: 0.7233720421791077:  82%|████████▏ | 5317/6470 [1:01:13<13:22,  1.44it/s]
2|5318|Loss: 0.7233720421791077:  82%|████████▏ | 5318/6470 [1:01:13<13:16,  1.45it/s]
2|5319|Loss: 0.8292789459228516:  82%|████████▏ | 5318/6470 [1:01:13<13:16,  1.45it/s]
2|5319|Loss: 0.8292789459228516:  82%|████████▏ | 5319/6470 [1:01:14<13:28,  1.42it/s]
2|5320|Loss: 0.9055227637290955:  82%|████████▏ | 5319/6470 [1:01:14<13:28,  1.42it/s]
2|5320|Loss: 0.9055227637290955:  82%|████████▏ | 5320/6470 [1:01:14<13:07,  1.46it/s]
2|5321|Loss: 0.6950698494911194:  82%|████████▏ | 5320/6470 [1:01:15<13:07,  1.46it/s]
2|5321|Loss: 0.6950698494911194:  82%|████████▏ | 5321/6470 [1:01:15<12:31,  1.53it/s]
2|5322|Loss: 0.920693576335907:  82%|████████▏ | 5321/6470 [1:01:15<12:31,  1.53it/s] 
2|5322|Loss: 0.920693576335907:  82%|████████▏ | 5322/6470 [1:01:16<12:25,  1.54it/s]
2|5323|Loss: 0.979110598564148:  82%|████████▏ | 5322/6470 [1:01:16<12:25,  1.54it/s]
2|5323|Loss: 0.979110598564148:  82%|████████▏ | 5323/6470 [1:01:16<12:48,  1.49it/s]
2|5324|Loss: 0.8975017666816711:  82%|████████▏ | 5323/6470 [1:01:17<12:48,  1.49it/s]
2|5324|Loss: 0.8975017666816711:  82%|████████▏ | 5324/6470 [1:01:17<12:08,  1.57it/s]
2|5325|Loss: 0.8149558305740356:  82%|████████▏ | 5324/6470 [1:01:17<12:08,  1.57it/s]
2|5325|Loss: 0.8149558305740356:  82%|████████▏ | 5325/6470 [1:01:18<12:59,  1.47it/s]
2|5326|Loss: 0.8332351446151733:  82%|████████▏ | 5325/6470 [1:01:18<12:59,  1.47it/s]
2|5326|Loss: 0.8332351446151733:  82%|████████▏ | 5326/6470 [1:01:18<12:37,  1.51it/s]
2|5327|Loss: 0.8828510642051697:  82%|████████▏ | 5326/6470 [1:01:19<12:37,  1.51it/s]
2|5327|Loss: 0.8828510642051697:  82%|████████▏ | 5327/6470 [1:01:19<12:54,  1.47it/s]
2|5328|Loss: 0.7967578768730164:  82%|████████▏ | 5327/6470 [1:01:19<12:54,  1.47it/s]
2|5328|Loss: 0.7967578768730164:  82%|████████▏ | 5328/6470 [1:01:20<12:45,  1.49it/s]
2|5329|Loss: 0.8275130987167358:  82%|████████▏ | 5328/6470 [1:01:20<12:45,  1.49it/s]
2|5329|Loss: 0.8275130987167358:  82%|████████▏ | 5329/6470 [1:01:20<12:51,  1.48it/s]
2|5330|Loss: 1.0228549242019653:  82%|████████▏ | 5329/6470 [1:01:21<12:51,  1.48it/s]
2|5330|Loss: 1.0228549242019653:  82%|████████▏ | 5330/6470 [1:01:21<13:00,  1.46it/s]
2|5331|Loss: 0.8738053441047668:  82%|████████▏ | 5330/6470 [1:01:21<13:00,  1.46it/s]
2|5331|Loss: 0.8738053441047668:  82%|████████▏ | 5331/6470 [1:01:22<12:43,  1.49it/s]
2|5332|Loss: 0.8658382296562195:  82%|████████▏ | 5331/6470 [1:01:22<12:43,  1.49it/s]
2|5332|Loss: 0.8658382296562195:  82%|████████▏ | 5332/6470 [1:01:22<12:47,  1.48it/s]
2|5333|Loss: 0.9029870629310608:  82%|████████▏ | 5332/6470 [1:01:23<12:47,  1.48it/s]
2|5333|Loss: 0.9029870629310608:  82%|████████▏ | 5333/6470 [1:01:23<13:19,  1.42it/s]
2|5334|Loss: 0.8971354961395264:  82%|████████▏ | 5333/6470 [1:01:24<13:19,  1.42it/s]
2|5334|Loss: 0.8971354961395264:  82%|████████▏ | 5334/6470 [1:01:24<12:39,  1.50it/s]
2|5335|Loss: 0.8059713244438171:  82%|████████▏ | 5334/6470 [1:01:24<12:39,  1.50it/s]
2|5335|Loss: 0.8059713244438171:  82%|████████▏ | 5335/6470 [1:01:25<13:12,  1.43it/s]
2|5336|Loss: 0.7646492719650269:  82%|████████▏ | 5335/6470 [1:01:25<13:12,  1.43it/s]
2|5336|Loss: 0.7646492719650269:  82%|████████▏ | 5336/6470 [1:01:25<13:22,  1.41it/s]
2|5337|Loss: 1.04400634765625:  82%|████████▏ | 5336/6470 [1:01:26<13:22,  1.41it/s]  
2|5337|Loss: 1.04400634765625:  82%|████████▏ | 5337/6470 [1:01:26<13:27,  1.40it/s]
2|5338|Loss: 0.9119659066200256:  82%|████████▏ | 5337/6470 [1:01:26<13:27,  1.40it/s]
2|5338|Loss: 0.9119659066200256:  83%|████████▎ | 5338/6470 [1:01:27<13:00,  1.45it/s]
2|5339|Loss: 0.8773908019065857:  83%|████████▎ | 5338/6470 [1:01:27<13:00,  1.45it/s]
2|5339|Loss: 0.8773908019065857:  83%|████████▎ | 5339/6470 [1:01:27<12:34,  1.50it/s]
2|5340|Loss: 0.9478163123130798:  83%|████████▎ | 5339/6470 [1:01:28<12:34,  1.50it/s]
2|5340|Loss: 0.9478163123130798:  83%|████████▎ | 5340/6470 [1:01:28<11:15,  1.67it/s]
2|5341|Loss: 0.8939374089241028:  83%|████████▎ | 5340/6470 [1:01:28<11:15,  1.67it/s]
2|5341|Loss: 0.8939374089241028:  83%|████████▎ | 5341/6470 [1:01:28<11:15,  1.67it/s]
2|5342|Loss: 0.8317071795463562:  83%|████████▎ | 5341/6470 [1:01:29<11:15,  1.67it/s]
2|5342|Loss: 0.8317071795463562:  83%|████████▎ | 5342/6470 [1:01:29<11:00,  1.71it/s]
2|5343|Loss: 1.008700966835022:  83%|████████▎ | 5342/6470 [1:01:29<11:00,  1.71it/s] 
2|5343|Loss: 1.008700966835022:  83%|████████▎ | 5343/6470 [1:01:30<11:43,  1.60it/s]
2|5344|Loss: 0.8745934367179871:  83%|████████▎ | 5343/6470 [1:01:30<11:43,  1.60it/s]
2|5344|Loss: 0.8745934367179871:  83%|████████▎ | 5344/6470 [1:01:30<12:56,  1.45it/s]
2|5345|Loss: 0.7522872090339661:  83%|████████▎ | 5344/6470 [1:01:31<12:56,  1.45it/s]
2|5345|Loss: 0.7522872090339661:  83%|████████▎ | 5345/6470 [1:01:31<13:27,  1.39it/s]
2|5346|Loss: 0.7797566056251526:  83%|████████▎ | 5345/6470 [1:01:32<13:27,  1.39it/s]
2|5346|Loss: 0.7797566056251526:  83%|████████▎ | 5346/6470 [1:01:32<13:48,  1.36it/s]
2|5347|Loss: 0.8520107269287109:  83%|████████▎ | 5346/6470 [1:01:32<13:48,  1.36it/s]
2|5347|Loss: 0.8520107269287109:  83%|████████▎ | 5347/6470 [1:01:33<13:52,  1.35it/s]
2|5348|Loss: 0.8744319081306458:  83%|████████▎ | 5347/6470 [1:01:33<13:52,  1.35it/s]
2|5348|Loss: 0.8744319081306458:  83%|████████▎ | 5348/6470 [1:01:33<13:16,  1.41it/s]
2|5349|Loss: 0.8729475736618042:  83%|████████▎ | 5348/6470 [1:01:34<13:16,  1.41it/s]
2|5349|Loss: 0.8729475736618042:  83%|████████▎ | 5349/6470 [1:01:34<13:34,  1.38it/s]
2|5350|Loss: 0.8492706418037415:  83%|████████▎ | 5349/6470 [1:01:34<13:34,  1.38it/s]
2|5350|Loss: 0.8492706418037415:  83%|████████▎ | 5350/6470 [1:01:35<13:10,  1.42it/s]
2|5351|Loss: 0.9250673055648804:  83%|████████▎ | 5350/6470 [1:01:35<13:10,  1.42it/s]
2|5351|Loss: 0.9250673055648804:  83%|████████▎ | 5351/6470 [1:01:35<13:08,  1.42it/s]
2|5352|Loss: 0.8131972551345825:  83%|████████▎ | 5351/6470 [1:01:36<13:08,  1.42it/s]
2|5352|Loss: 0.8131972551345825:  83%|████████▎ | 5352/6470 [1:01:36<12:53,  1.45it/s]
2|5353|Loss: 0.8062067627906799:  83%|████████▎ | 5352/6470 [1:01:37<12:53,  1.45it/s]
2|5353|Loss: 0.8062067627906799:  83%|████████▎ | 5353/6470 [1:01:37<13:39,  1.36it/s]
2|5354|Loss: 0.8011997938156128:  83%|████████▎ | 5353/6470 [1:01:37<13:39,  1.36it/s]
2|5354|Loss: 0.8011997938156128:  83%|████████▎ | 5354/6470 [1:01:38<13:20,  1.39it/s]
2|5355|Loss: 0.7890843749046326:  83%|████████▎ | 5354/6470 [1:01:38<13:20,  1.39it/s]
2|5355|Loss: 0.7890843749046326:  83%|████████▎ | 5355/6470 [1:01:38<12:23,  1.50it/s]
2|5356|Loss: 0.9222118258476257:  83%|████████▎ | 5355/6470 [1:01:39<12:23,  1.50it/s]
2|5356|Loss: 0.9222118258476257:  83%|████████▎ | 5356/6470 [1:01:39<12:07,  1.53it/s]
2|5357|Loss: 0.7777702808380127:  83%|████████▎ | 5356/6470 [1:01:39<12:07,  1.53it/s]
2|5357|Loss: 0.7777702808380127:  83%|████████▎ | 5357/6470 [1:01:39<12:19,  1.51it/s]
2|5358|Loss: 0.9708009362220764:  83%|████████▎ | 5357/6470 [1:01:40<12:19,  1.51it/s]
2|5358|Loss: 0.9708009362220764:  83%|████████▎ | 5358/6470 [1:01:40<11:49,  1.57it/s]
2|5359|Loss: 0.900604248046875:  83%|████████▎ | 5358/6470 [1:01:40<11:49,  1.57it/s] 
2|5359|Loss: 0.900604248046875:  83%|████████▎ | 5359/6470 [1:01:41<11:27,  1.62it/s]
2|5360|Loss: 0.8206448554992676:  83%|████████▎ | 5359/6470 [1:01:41<11:27,  1.62it/s]
2|5360|Loss: 0.8206448554992676:  83%|████████▎ | 5360/6470 [1:01:41<11:11,  1.65it/s]
2|5361|Loss: 0.7239097356796265:  83%|████████▎ | 5360/6470 [1:01:42<11:11,  1.65it/s]
2|5361|Loss: 0.7239097356796265:  83%|████████▎ | 5361/6470 [1:01:42<12:15,  1.51it/s]
2|5362|Loss: 0.767123281955719:  83%|████████▎ | 5361/6470 [1:01:42<12:15,  1.51it/s] 
2|5362|Loss: 0.767123281955719:  83%|████████▎ | 5362/6470 [1:01:43<12:47,  1.44it/s]
2|5363|Loss: 0.8589760661125183:  83%|████████▎ | 5362/6470 [1:01:43<12:47,  1.44it/s]
2|5363|Loss: 0.8589760661125183:  83%|████████▎ | 5363/6470 [1:01:44<13:33,  1.36it/s]
2|5364|Loss: 0.8514153361320496:  83%|████████▎ | 5363/6470 [1:01:44<13:33,  1.36it/s]
2|5364|Loss: 0.8514153361320496:  83%|████████▎ | 5364/6470 [1:01:44<13:41,  1.35it/s]
2|5365|Loss: 0.8365509510040283:  83%|████████▎ | 5364/6470 [1:01:45<13:41,  1.35it/s]
2|5365|Loss: 0.8365509510040283:  83%|████████▎ | 5365/6470 [1:01:45<13:27,  1.37it/s]
2|5366|Loss: 0.8633459210395813:  83%|████████▎ | 5365/6470 [1:01:45<13:27,  1.37it/s]
2|5366|Loss: 0.8633459210395813:  83%|████████▎ | 5366/6470 [1:01:46<13:27,  1.37it/s]
2|5367|Loss: 0.88386470079422:  83%|████████▎ | 5366/6470 [1:01:46<13:27,  1.37it/s]  
2|5367|Loss: 0.88386470079422:  83%|████████▎ | 5367/6470 [1:01:46<12:58,  1.42it/s]
2|5368|Loss: 0.8852013349533081:  83%|████████▎ | 5367/6470 [1:01:47<12:58,  1.42it/s]
2|5368|Loss: 0.8852013349533081:  83%|████████▎ | 5368/6470 [1:01:47<13:13,  1.39it/s]
2|5369|Loss: 0.9175769686698914:  83%|████████▎ | 5368/6470 [1:01:48<13:13,  1.39it/s]
2|5369|Loss: 0.9175769686698914:  83%|████████▎ | 5369/6470 [1:01:48<13:20,  1.38it/s]
2|5370|Loss: 0.6944650411605835:  83%|████████▎ | 5369/6470 [1:01:48<13:20,  1.38it/s]
2|5370|Loss: 0.6944650411605835:  83%|████████▎ | 5370/6470 [1:01:49<13:22,  1.37it/s]
2|5371|Loss: 0.9460179209709167:  83%|████████▎ | 5370/6470 [1:01:49<13:22,  1.37it/s]
2|5371|Loss: 0.9460179209709167:  83%|████████▎ | 5371/6470 [1:01:49<13:01,  1.41it/s]
2|5372|Loss: 0.8062241673469543:  83%|████████▎ | 5371/6470 [1:01:50<13:01,  1.41it/s]
2|5372|Loss: 0.8062241673469543:  83%|████████▎ | 5372/6470 [1:01:50<12:32,  1.46it/s]
2|5373|Loss: 0.8823995590209961:  83%|████████▎ | 5372/6470 [1:01:50<12:32,  1.46it/s]
2|5373|Loss: 0.8823995590209961:  83%|████████▎ | 5373/6470 [1:01:51<12:09,  1.50it/s]
2|5374|Loss: 0.8489779233932495:  83%|████████▎ | 5373/6470 [1:01:51<12:09,  1.50it/s]
2|5374|Loss: 0.8489779233932495:  83%|████████▎ | 5374/6470 [1:01:51<11:54,  1.53it/s]
2|5375|Loss: 0.9198529124259949:  83%|████████▎ | 5374/6470 [1:01:52<11:54,  1.53it/s]
2|5375|Loss: 0.9198529124259949:  83%|████████▎ | 5375/6470 [1:01:52<11:44,  1.55it/s]
2|5376|Loss: 0.7982677221298218:  83%|████████▎ | 5375/6470 [1:01:52<11:44,  1.55it/s]
2|5376|Loss: 0.7982677221298218:  83%|████████▎ | 5376/6470 [1:01:52<11:36,  1.57it/s]
2|5377|Loss: 0.8448279500007629:  83%|████████▎ | 5376/6470 [1:01:53<11:36,  1.57it/s]
2|5377|Loss: 0.8448279500007629:  83%|████████▎ | 5377/6470 [1:01:53<12:25,  1.47it/s]
2|5378|Loss: 0.8908026814460754:  83%|████████▎ | 5377/6470 [1:01:54<12:25,  1.47it/s]
2|5378|Loss: 0.8908026814460754:  83%|████████▎ | 5378/6470 [1:01:54<12:02,  1.51it/s]
2|5379|Loss: 0.9459284543991089:  83%|████████▎ | 5378/6470 [1:01:54<12:02,  1.51it/s]
2|5379|Loss: 0.9459284543991089:  83%|████████▎ | 5379/6470 [1:01:55<12:04,  1.51it/s]
2|5380|Loss: 0.8484399914741516:  83%|████████▎ | 5379/6470 [1:01:55<12:04,  1.51it/s]
2|5380|Loss: 0.8484399914741516:  83%|████████▎ | 5380/6470 [1:01:55<12:26,  1.46it/s]
2|5381|Loss: 0.963637113571167:  83%|████████▎ | 5380/6470 [1:01:56<12:26,  1.46it/s] 
2|5381|Loss: 0.963637113571167:  83%|████████▎ | 5381/6470 [1:01:56<11:42,  1.55it/s]
2|5382|Loss: 0.8869826197624207:  83%|████████▎ | 5381/6470 [1:01:56<11:42,  1.55it/s]
2|5382|Loss: 0.8869826197624207:  83%|████████▎ | 5382/6470 [1:01:56<11:46,  1.54it/s]
2|5383|Loss: 0.7911219000816345:  83%|████████▎ | 5382/6470 [1:01:57<11:46,  1.54it/s]
2|5383|Loss: 0.7911219000816345:  83%|████████▎ | 5383/6470 [1:01:57<11:41,  1.55it/s]
2|5384|Loss: 0.724371075630188:  83%|████████▎ | 5383/6470 [1:01:57<11:41,  1.55it/s] 
2|5384|Loss: 0.724371075630188:  83%|████████▎ | 5384/6470 [1:01:58<11:54,  1.52it/s]
2|5385|Loss: 0.8664408922195435:  83%|████████▎ | 5384/6470 [1:01:58<11:54,  1.52it/s]
2|5385|Loss: 0.8664408922195435:  83%|████████▎ | 5385/6470 [1:01:59<12:16,  1.47it/s]
2|5386|Loss: 0.8658875823020935:  83%|████████▎ | 5385/6470 [1:01:59<12:16,  1.47it/s]
2|5386|Loss: 0.8658875823020935:  83%|████████▎ | 5386/6470 [1:01:59<12:34,  1.44it/s]
2|5387|Loss: 0.899837851524353:  83%|████████▎ | 5386/6470 [1:02:00<12:34,  1.44it/s] 
2|5387|Loss: 0.899837851524353:  83%|████████▎ | 5387/6470 [1:02:00<12:39,  1.43it/s]
2|5388|Loss: 0.8221802115440369:  83%|████████▎ | 5387/6470 [1:02:00<12:39,  1.43it/s]
2|5388|Loss: 0.8221802115440369:  83%|████████▎ | 5388/6470 [1:02:01<12:23,  1.45it/s]
2|5389|Loss: 1.0056920051574707:  83%|████████▎ | 5388/6470 [1:02:01<12:23,  1.45it/s]
2|5389|Loss: 1.0056920051574707:  83%|████████▎ | 5389/6470 [1:02:01<12:08,  1.48it/s]
2|5390|Loss: 0.8479956984519958:  83%|████████▎ | 5389/6470 [1:02:02<12:08,  1.48it/s]
2|5390|Loss: 0.8479956984519958:  83%|████████▎ | 5390/6470 [1:02:02<12:13,  1.47it/s]
2|5391|Loss: 0.7874118089675903:  83%|████████▎ | 5390/6470 [1:02:02<12:13,  1.47it/s]
2|5391|Loss: 0.7874118089675903:  83%|████████▎ | 5391/6470 [1:02:03<12:45,  1.41it/s]
2|5392|Loss: 0.7235576510429382:  83%|████████▎ | 5391/6470 [1:02:03<12:45,  1.41it/s]
2|5392|Loss: 0.7235576510429382:  83%|████████▎ | 5392/6470 [1:02:04<13:00,  1.38it/s]
2|5393|Loss: 0.8439871072769165:  83%|████████▎ | 5392/6470 [1:02:04<13:00,  1.38it/s]
2|5393|Loss: 0.8439871072769165:  83%|████████▎ | 5393/6470 [1:02:04<12:43,  1.41it/s]
2|5394|Loss: 0.9017611742019653:  83%|████████▎ | 5393/6470 [1:02:05<12:43,  1.41it/s]
2|5394|Loss: 0.9017611742019653:  83%|████████▎ | 5394/6470 [1:02:05<13:18,  1.35it/s]
2|5395|Loss: 0.7367478609085083:  83%|████████▎ | 5394/6470 [1:02:05<13:18,  1.35it/s]
2|5395|Loss: 0.7367478609085083:  83%|████████▎ | 5395/6470 [1:02:05<11:54,  1.50it/s]
2|5396|Loss: 0.7732064723968506:  83%|████████▎ | 5395/6470 [1:02:06<11:54,  1.50it/s]
2|5396|Loss: 0.7732064723968506:  83%|████████▎ | 5396/6470 [1:02:06<11:09,  1.60it/s]
2|5397|Loss: 0.8489061594009399:  83%|████████▎ | 5396/6470 [1:02:06<11:09,  1.60it/s]
2|5397|Loss: 0.8489061594009399:  83%|████████▎ | 5397/6470 [1:02:07<12:12,  1.47it/s]
2|5398|Loss: 0.8565927147865295:  83%|████████▎ | 5397/6470 [1:02:07<12:12,  1.47it/s]
2|5398|Loss: 0.8565927147865295:  83%|████████▎ | 5398/6470 [1:02:08<12:43,  1.40it/s]
2|5399|Loss: 0.8652880191802979:  83%|████████▎ | 5398/6470 [1:02:08<12:43,  1.40it/s]
2|5399|Loss: 0.8652880191802979:  83%|████████▎ | 5399/6470 [1:02:08<12:11,  1.46it/s]
2|5400|Loss: 0.838897705078125:  83%|████████▎ | 5399/6470 [1:02:09<12:11,  1.46it/s] 
2|5400|Loss: 0.838897705078125:  83%|████████▎ | 5400/6470 [1:02:09<11:35,  1.54it/s]
2|5401|Loss: 0.8200517296791077:  83%|████████▎ | 5400/6470 [1:02:09<11:35,  1.54it/s]
2|5401|Loss: 0.8200517296791077:  83%|████████▎ | 5401/6470 [1:02:09<10:40,  1.67it/s]
2|5402|Loss: 0.9031295776367188:  83%|████████▎ | 5401/6470 [1:02:10<10:40,  1.67it/s]
2|5402|Loss: 0.9031295776367188:  83%|████████▎ | 5402/6470 [1:02:10<10:17,  1.73it/s]
2|5403|Loss: 0.8823341727256775:  83%|████████▎ | 5402/6470 [1:02:10<10:17,  1.73it/s]
2|5403|Loss: 0.8823341727256775:  84%|████████▎ | 5403/6470 [1:02:11<10:59,  1.62it/s]
2|5404|Loss: 0.9079988598823547:  84%|████████▎ | 5403/6470 [1:02:11<10:59,  1.62it/s]
2|5404|Loss: 0.9079988598823547:  84%|████████▎ | 5404/6470 [1:02:11<11:02,  1.61it/s]
2|5405|Loss: 0.9521006941795349:  84%|████████▎ | 5404/6470 [1:02:12<11:02,  1.61it/s]
2|5405|Loss: 0.9521006941795349:  84%|████████▎ | 5405/6470 [1:02:12<11:28,  1.55it/s]
2|5406|Loss: 0.9841682314872742:  84%|████████▎ | 5405/6470 [1:02:12<11:28,  1.55it/s]
2|5406|Loss: 0.9841682314872742:  84%|████████▎ | 5406/6470 [1:02:13<11:54,  1.49it/s]
2|5407|Loss: 0.7856675982475281:  84%|████████▎ | 5406/6470 [1:02:13<11:54,  1.49it/s]
2|5407|Loss: 0.7856675982475281:  84%|████████▎ | 5407/6470 [1:02:13<11:52,  1.49it/s]
2|5408|Loss: 0.8943483829498291:  84%|████████▎ | 5407/6470 [1:02:14<11:52,  1.49it/s]
2|5408|Loss: 0.8943483829498291:  84%|████████▎ | 5408/6470 [1:02:14<11:58,  1.48it/s]
2|5409|Loss: 0.8539579510688782:  84%|████████▎ | 5408/6470 [1:02:14<11:58,  1.48it/s]
2|5409|Loss: 0.8539579510688782:  84%|████████▎ | 5409/6470 [1:02:15<12:05,  1.46it/s]
2|5410|Loss: 0.9081981778144836:  84%|████████▎ | 5409/6470 [1:02:15<12:05,  1.46it/s]
2|5410|Loss: 0.9081981778144836:  84%|████████▎ | 5410/6470 [1:02:15<11:31,  1.53it/s]
2|5411|Loss: 0.9266713857650757:  84%|████████▎ | 5410/6470 [1:02:16<11:31,  1.53it/s]
2|5411|Loss: 0.9266713857650757:  84%|████████▎ | 5411/6470 [1:02:16<11:45,  1.50it/s]
2|5412|Loss: 0.8743799328804016:  84%|████████▎ | 5411/6470 [1:02:16<11:45,  1.50it/s]
2|5412|Loss: 0.8743799328804016:  84%|████████▎ | 5412/6470 [1:02:17<12:37,  1.40it/s]
2|5413|Loss: 0.8139612078666687:  84%|████████▎ | 5412/6470 [1:02:17<12:37,  1.40it/s]
2|5413|Loss: 0.8139612078666687:  84%|████████▎ | 5413/6470 [1:02:17<11:39,  1.51it/s]
2|5414|Loss: 0.8332250714302063:  84%|████████▎ | 5413/6470 [1:02:18<11:39,  1.51it/s]
2|5414|Loss: 0.8332250714302063:  84%|████████▎ | 5414/6470 [1:02:18<11:13,  1.57it/s]
2|5415|Loss: 0.7490809559822083:  84%|████████▎ | 5414/6470 [1:02:18<11:13,  1.57it/s]
2|5415|Loss: 0.7490809559822083:  84%|████████▎ | 5415/6470 [1:02:19<11:48,  1.49it/s]
2|5416|Loss: 0.8619334697723389:  84%|████████▎ | 5415/6470 [1:02:19<11:48,  1.49it/s]
2|5416|Loss: 0.8619334697723389:  84%|████████▎ | 5416/6470 [1:02:19<12:28,  1.41it/s]
2|5417|Loss: 0.7359958291053772:  84%|████████▎ | 5416/6470 [1:02:20<12:28,  1.41it/s]
2|5417|Loss: 0.7359958291053772:  84%|████████▎ | 5417/6470 [1:02:20<12:30,  1.40it/s]
2|5418|Loss: 0.7293906807899475:  84%|████████▎ | 5417/6470 [1:02:21<12:30,  1.40it/s]
2|5418|Loss: 0.7293906807899475:  84%|████████▎ | 5418/6470 [1:02:21<13:37,  1.29it/s]
2|5419|Loss: 0.8138631582260132:  84%|████████▎ | 5418/6470 [1:02:22<13:37,  1.29it/s]
2|5419|Loss: 0.8138631582260132:  84%|████████▍ | 5419/6470 [1:02:22<13:42,  1.28it/s]
2|5420|Loss: 0.8128208518028259:  84%|████████▍ | 5419/6470 [1:02:22<13:42,  1.28it/s]
2|5420|Loss: 0.8128208518028259:  84%|████████▍ | 5420/6470 [1:02:23<13:06,  1.33it/s]
2|5421|Loss: 0.8947596549987793:  84%|████████▍ | 5420/6470 [1:02:23<13:06,  1.33it/s]
2|5421|Loss: 0.8947596549987793:  84%|████████▍ | 5421/6470 [1:02:23<12:58,  1.35it/s]
2|5422|Loss: 0.7856227159500122:  84%|████████▍ | 5421/6470 [1:02:24<12:58,  1.35it/s]
2|5422|Loss: 0.7856227159500122:  84%|████████▍ | 5422/6470 [1:02:24<12:46,  1.37it/s]
2|5423|Loss: 0.8269372582435608:  84%|████████▍ | 5422/6470 [1:02:24<12:46,  1.37it/s]
2|5423|Loss: 0.8269372582435608:  84%|████████▍ | 5423/6470 [1:02:25<13:27,  1.30it/s]
2|5424|Loss: 0.8524280190467834:  84%|████████▍ | 5423/6470 [1:02:25<13:27,  1.30it/s]
2|5424|Loss: 0.8524280190467834:  84%|████████▍ | 5424/6470 [1:02:26<13:03,  1.34it/s]
2|5425|Loss: 0.7143486738204956:  84%|████████▍ | 5424/6470 [1:02:26<13:03,  1.34it/s]
2|5425|Loss: 0.7143486738204956:  84%|████████▍ | 5425/6470 [1:02:26<13:13,  1.32it/s]
2|5426|Loss: 0.7859835028648376:  84%|████████▍ | 5425/6470 [1:02:27<13:13,  1.32it/s]
2|5426|Loss: 0.7859835028648376:  84%|████████▍ | 5426/6470 [1:02:27<13:02,  1.33it/s]
2|5427|Loss: 1.0168688297271729:  84%|████████▍ | 5426/6470 [1:02:27<13:02,  1.33it/s]
2|5427|Loss: 1.0168688297271729:  84%|████████▍ | 5427/6470 [1:02:28<13:02,  1.33it/s]
2|5428|Loss: 0.7863962650299072:  84%|████████▍ | 5427/6470 [1:02:28<13:02,  1.33it/s]
2|5428|Loss: 0.7863962650299072:  84%|████████▍ | 5428/6470 [1:02:29<12:51,  1.35it/s]
2|5429|Loss: 0.8949260711669922:  84%|████████▍ | 5428/6470 [1:02:29<12:51,  1.35it/s]
2|5429|Loss: 0.8949260711669922:  84%|████████▍ | 5429/6470 [1:02:29<12:47,  1.36it/s]
2|5430|Loss: 0.8112840056419373:  84%|████████▍ | 5429/6470 [1:02:30<12:47,  1.36it/s]
2|5430|Loss: 0.8112840056419373:  84%|████████▍ | 5430/6470 [1:02:30<12:47,  1.36it/s]
2|5431|Loss: 0.8666318655014038:  84%|████████▍ | 5430/6470 [1:02:30<12:47,  1.36it/s]
2|5431|Loss: 0.8666318655014038:  84%|████████▍ | 5431/6470 [1:02:31<12:18,  1.41it/s]
2|5432|Loss: 0.8242771625518799:  84%|████████▍ | 5431/6470 [1:02:31<12:18,  1.41it/s]
2|5432|Loss: 0.8242771625518799:  84%|████████▍ | 5432/6470 [1:02:31<12:09,  1.42it/s]
2|5433|Loss: 0.8637038469314575:  84%|████████▍ | 5432/6470 [1:02:32<12:09,  1.42it/s]
2|5433|Loss: 0.8637038469314575:  84%|████████▍ | 5433/6470 [1:02:32<11:19,  1.53it/s]
2|5434|Loss: 0.7861436009407043:  84%|████████▍ | 5433/6470 [1:02:32<11:19,  1.53it/s]
2|5434|Loss: 0.7861436009407043:  84%|████████▍ | 5434/6470 [1:02:32<10:58,  1.57it/s]
2|5435|Loss: 0.8807922601699829:  84%|████████▍ | 5434/6470 [1:02:33<10:58,  1.57it/s]
2|5435|Loss: 0.8807922601699829:  84%|████████▍ | 5435/6470 [1:02:33<10:41,  1.61it/s]
2|5436|Loss: 0.7417685389518738:  84%|████████▍ | 5435/6470 [1:02:33<10:41,  1.61it/s]
2|5436|Loss: 0.7417685389518738:  84%|████████▍ | 5436/6470 [1:02:34<11:12,  1.54it/s]
2|5437|Loss: 0.846095085144043:  84%|████████▍ | 5436/6470 [1:02:34<11:12,  1.54it/s] 
2|5437|Loss: 0.846095085144043:  84%|████████▍ | 5437/6470 [1:02:34<11:17,  1.52it/s]
2|5438|Loss: 0.874481737613678:  84%|████████▍ | 5437/6470 [1:02:35<11:17,  1.52it/s]
2|5438|Loss: 0.874481737613678:  84%|████████▍ | 5438/6470 [1:02:35<11:11,  1.54it/s]
2|5439|Loss: 1.0329927206039429:  84%|████████▍ | 5438/6470 [1:02:35<11:11,  1.54it/s]
2|5439|Loss: 1.0329927206039429:  84%|████████▍ | 5439/6470 [1:02:36<10:46,  1.60it/s]
2|5440|Loss: 0.7617524266242981:  84%|████████▍ | 5439/6470 [1:02:36<10:46,  1.60it/s]
2|5440|Loss: 0.7617524266242981:  84%|████████▍ | 5440/6470 [1:02:36<10:11,  1.68it/s]
2|5441|Loss: 0.8321593403816223:  84%|████████▍ | 5440/6470 [1:02:36<10:11,  1.68it/s]
2|5441|Loss: 0.8321593403816223:  84%|████████▍ | 5441/6470 [1:02:37<10:39,  1.61it/s]
2|5442|Loss: 0.784004807472229:  84%|████████▍ | 5441/6470 [1:02:37<10:39,  1.61it/s] 
2|5442|Loss: 0.784004807472229:  84%|████████▍ | 5442/6470 [1:02:37<10:45,  1.59it/s]
2|5443|Loss: 0.8921416401863098:  84%|████████▍ | 5442/6470 [1:02:38<10:45,  1.59it/s]
2|5443|Loss: 0.8921416401863098:  84%|████████▍ | 5443/6470 [1:02:38<10:41,  1.60it/s]
2|5444|Loss: 0.8000665307044983:  84%|████████▍ | 5443/6470 [1:02:38<10:41,  1.60it/s]
2|5444|Loss: 0.8000665307044983:  84%|████████▍ | 5444/6470 [1:02:39<11:19,  1.51it/s]
2|5445|Loss: 0.9624029397964478:  84%|████████▍ | 5444/6470 [1:02:39<11:19,  1.51it/s]
2|5445|Loss: 0.9624029397964478:  84%|████████▍ | 5445/6470 [1:02:40<11:22,  1.50it/s]
2|5446|Loss: 0.8903658390045166:  84%|████████▍ | 5445/6470 [1:02:40<11:22,  1.50it/s]
2|5446|Loss: 0.8903658390045166:  84%|████████▍ | 5446/6470 [1:02:40<12:01,  1.42it/s]
2|5447|Loss: 0.9547404050827026:  84%|████████▍ | 5446/6470 [1:02:41<12:01,  1.42it/s]
2|5447|Loss: 0.9547404050827026:  84%|████████▍ | 5447/6470 [1:02:41<11:26,  1.49it/s]
2|5448|Loss: 0.9087631702423096:  84%|████████▍ | 5447/6470 [1:02:41<11:26,  1.49it/s]
2|5448|Loss: 0.9087631702423096:  84%|████████▍ | 5448/6470 [1:02:41<10:55,  1.56it/s]
2|5449|Loss: 0.7333328127861023:  84%|████████▍ | 5448/6470 [1:02:42<10:55,  1.56it/s]
2|5449|Loss: 0.7333328127861023:  84%|████████▍ | 5449/6470 [1:02:42<12:10,  1.40it/s]
2|5450|Loss: 0.7229998111724854:  84%|████████▍ | 5449/6470 [1:02:43<12:10,  1.40it/s]
2|5450|Loss: 0.7229998111724854:  84%|████████▍ | 5450/6470 [1:02:43<12:22,  1.37it/s]
2|5451|Loss: 0.8038191795349121:  84%|████████▍ | 5450/6470 [1:02:43<12:22,  1.37it/s]
2|5451|Loss: 0.8038191795349121:  84%|████████▍ | 5451/6470 [1:02:44<11:37,  1.46it/s]
2|5452|Loss: 0.7890146970748901:  84%|████████▍ | 5451/6470 [1:02:44<11:37,  1.46it/s]
2|5452|Loss: 0.7890146970748901:  84%|████████▍ | 5452/6470 [1:02:44<11:20,  1.50it/s]
2|5453|Loss: 0.8594585061073303:  84%|████████▍ | 5452/6470 [1:02:45<11:20,  1.50it/s]
2|5453|Loss: 0.8594585061073303:  84%|████████▍ | 5453/6470 [1:02:45<11:39,  1.45it/s]
2|5454|Loss: 0.7920648455619812:  84%|████████▍ | 5453/6470 [1:02:46<11:39,  1.45it/s]
2|5454|Loss: 0.7920648455619812:  84%|████████▍ | 5454/6470 [1:02:46<12:18,  1.37it/s]
2|5455|Loss: 0.941902220249176:  84%|████████▍ | 5454/6470 [1:02:46<12:18,  1.37it/s] 
2|5455|Loss: 0.941902220249176:  84%|████████▍ | 5455/6470 [1:02:46<11:41,  1.45it/s]
2|5456|Loss: 0.8881272673606873:  84%|████████▍ | 5455/6470 [1:02:47<11:41,  1.45it/s]
2|5456|Loss: 0.8881272673606873:  84%|████████▍ | 5456/6470 [1:02:47<11:36,  1.46it/s]
2|5457|Loss: 0.9956265091896057:  84%|████████▍ | 5456/6470 [1:02:48<11:36,  1.46it/s]
2|5457|Loss: 0.9956265091896057:  84%|████████▍ | 5457/6470 [1:02:48<12:19,  1.37it/s]
2|5458|Loss: 0.6577370166778564:  84%|████████▍ | 5457/6470 [1:02:48<12:19,  1.37it/s]
2|5458|Loss: 0.6577370166778564:  84%|████████▍ | 5458/6470 [1:02:48<10:59,  1.53it/s]
2|5459|Loss: 0.921431839466095:  84%|████████▍ | 5458/6470 [1:02:49<10:59,  1.53it/s] 
2|5459|Loss: 0.921431839466095:  84%|████████▍ | 5459/6470 [1:02:49<10:04,  1.67it/s]
2|5460|Loss: 0.7204766869544983:  84%|████████▍ | 5459/6470 [1:02:49<10:04,  1.67it/s]
2|5460|Loss: 0.7204766869544983:  84%|████████▍ | 5460/6470 [1:02:50<10:33,  1.59it/s]
2|5461|Loss: 0.8017967343330383:  84%|████████▍ | 5460/6470 [1:02:50<10:33,  1.59it/s]
2|5461|Loss: 0.8017967343330383:  84%|████████▍ | 5461/6470 [1:02:50<11:35,  1.45it/s]
2|5462|Loss: 0.8192816376686096:  84%|████████▍ | 5461/6470 [1:02:51<11:35,  1.45it/s]
2|5462|Loss: 0.8192816376686096:  84%|████████▍ | 5462/6470 [1:02:51<12:13,  1.37it/s]
2|5463|Loss: 0.789697527885437:  84%|████████▍ | 5462/6470 [1:02:52<12:13,  1.37it/s] 
2|5463|Loss: 0.789697527885437:  84%|████████▍ | 5463/6470 [1:02:52<10:54,  1.54it/s]
2|5464|Loss: 0.914465069770813:  84%|████████▍ | 5463/6470 [1:02:52<10:54,  1.54it/s]
2|5464|Loss: 0.914465069770813:  84%|████████▍ | 5464/6470 [1:02:52<10:54,  1.54it/s]
2|5465|Loss: 0.7618790864944458:  84%|████████▍ | 5464/6470 [1:02:53<10:54,  1.54it/s]
2|5465|Loss: 0.7618790864944458:  84%|████████▍ | 5465/6470 [1:02:53<11:37,  1.44it/s]
2|5466|Loss: 0.8301302194595337:  84%|████████▍ | 5465/6470 [1:02:54<11:37,  1.44it/s]
2|5466|Loss: 0.8301302194595337:  84%|████████▍ | 5466/6470 [1:02:54<11:20,  1.48it/s]
2|5467|Loss: 0.7667081952095032:  84%|████████▍ | 5466/6470 [1:02:54<11:20,  1.48it/s]
2|5467|Loss: 0.7667081952095032:  84%|████████▍ | 5467/6470 [1:02:55<12:00,  1.39it/s]
2|5468|Loss: 0.917884886264801:  84%|████████▍ | 5467/6470 [1:02:55<12:00,  1.39it/s] 
2|5468|Loss: 0.917884886264801:  85%|████████▍ | 5468/6470 [1:02:55<12:12,  1.37it/s]
2|5469|Loss: 0.788735032081604:  85%|████████▍ | 5468/6470 [1:02:56<12:12,  1.37it/s]
2|5469|Loss: 0.788735032081604:  85%|████████▍ | 5469/6470 [1:02:56<12:10,  1.37it/s]
2|5470|Loss: 0.7574654221534729:  85%|████████▍ | 5469/6470 [1:02:57<12:10,  1.37it/s]
2|5470|Loss: 0.7574654221534729:  85%|████████▍ | 5470/6470 [1:02:57<11:51,  1.41it/s]
2|5471|Loss: 0.7665205001831055:  85%|████████▍ | 5470/6470 [1:02:57<11:51,  1.41it/s]
2|5471|Loss: 0.7665205001831055:  85%|████████▍ | 5471/6470 [1:02:57<11:28,  1.45it/s]
2|5472|Loss: 0.8530679941177368:  85%|████████▍ | 5471/6470 [1:02:58<11:28,  1.45it/s]
2|5472|Loss: 0.8530679941177368:  85%|████████▍ | 5472/6470 [1:02:58<11:52,  1.40it/s]
2|5473|Loss: 0.7275194525718689:  85%|████████▍ | 5472/6470 [1:02:59<11:52,  1.40it/s]
2|5473|Loss: 0.7275194525718689:  85%|████████▍ | 5473/6470 [1:02:59<13:00,  1.28it/s]
2|5474|Loss: 0.795278787612915:  85%|████████▍ | 5473/6470 [1:03:00<13:00,  1.28it/s] 
2|5474|Loss: 0.795278787612915:  85%|████████▍ | 5474/6470 [1:03:00<12:40,  1.31it/s]
2|5475|Loss: 0.8927240967750549:  85%|████████▍ | 5474/6470 [1:03:00<12:40,  1.31it/s]
2|5475|Loss: 0.8927240967750549:  85%|████████▍ | 5475/6470 [1:03:01<12:26,  1.33it/s]
2|5476|Loss: 0.8413178324699402:  85%|████████▍ | 5475/6470 [1:03:01<12:26,  1.33it/s]
2|5476|Loss: 0.8413178324699402:  85%|████████▍ | 5476/6470 [1:03:01<12:23,  1.34it/s]
2|5477|Loss: 0.829599142074585:  85%|████████▍ | 5476/6470 [1:03:02<12:23,  1.34it/s] 
2|5477|Loss: 0.829599142074585:  85%|████████▍ | 5477/6470 [1:03:02<12:19,  1.34it/s]
2|5478|Loss: 0.8394721746444702:  85%|████████▍ | 5477/6470 [1:03:02<12:19,  1.34it/s]
2|5478|Loss: 0.8394721746444702:  85%|████████▍ | 5478/6470 [1:03:03<12:17,  1.34it/s]
2|5479|Loss: 0.8651401996612549:  85%|████████▍ | 5478/6470 [1:03:03<12:17,  1.34it/s]
2|5479|Loss: 0.8651401996612549:  85%|████████▍ | 5479/6470 [1:03:03<11:53,  1.39it/s]
2|5480|Loss: 0.8509731292724609:  85%|████████▍ | 5479/6470 [1:03:04<11:53,  1.39it/s]
2|5480|Loss: 0.8509731292724609:  85%|████████▍ | 5480/6470 [1:03:04<12:35,  1.31it/s]
2|5481|Loss: 0.9079092144966125:  85%|████████▍ | 5480/6470 [1:03:05<12:35,  1.31it/s]
2|5481|Loss: 0.9079092144966125:  85%|████████▍ | 5481/6470 [1:03:05<12:16,  1.34it/s]
2|5482|Loss: 0.9293504357337952:  85%|████████▍ | 5481/6470 [1:03:05<12:16,  1.34it/s]
2|5482|Loss: 0.9293504357337952:  85%|████████▍ | 5482/6470 [1:03:06<12:35,  1.31it/s]
2|5483|Loss: 0.8685007095336914:  85%|████████▍ | 5482/6470 [1:03:06<12:35,  1.31it/s]
2|5483|Loss: 0.8685007095336914:  85%|████████▍ | 5483/6470 [1:03:07<12:40,  1.30it/s]
2|5484|Loss: 0.9166023135185242:  85%|████████▍ | 5483/6470 [1:03:07<12:40,  1.30it/s]
2|5484|Loss: 0.9166023135185242:  85%|████████▍ | 5484/6470 [1:03:07<12:26,  1.32it/s]
2|5485|Loss: 0.9301040768623352:  85%|████████▍ | 5484/6470 [1:03:08<12:26,  1.32it/s]
2|5485|Loss: 0.9301040768623352:  85%|████████▍ | 5485/6470 [1:03:08<11:58,  1.37it/s]
2|5486|Loss: 0.9056652188301086:  85%|████████▍ | 5485/6470 [1:03:08<11:58,  1.37it/s]
2|5486|Loss: 0.9056652188301086:  85%|████████▍ | 5486/6470 [1:03:09<11:15,  1.46it/s]
2|5487|Loss: 0.6955874562263489:  85%|████████▍ | 5486/6470 [1:03:09<11:15,  1.46it/s]
2|5487|Loss: 0.6955874562263489:  85%|████████▍ | 5487/6470 [1:03:09<11:42,  1.40it/s]
2|5488|Loss: 0.8796573281288147:  85%|████████▍ | 5487/6470 [1:03:10<11:42,  1.40it/s]
2|5488|Loss: 0.8796573281288147:  85%|████████▍ | 5488/6470 [1:03:10<11:01,  1.48it/s]
2|5489|Loss: 0.7508516311645508:  85%|████████▍ | 5488/6470 [1:03:10<11:01,  1.48it/s]
2|5489|Loss: 0.7508516311645508:  85%|████████▍ | 5489/6470 [1:03:10<10:06,  1.62it/s]
2|5490|Loss: 0.679031252861023:  85%|████████▍ | 5489/6470 [1:03:11<10:06,  1.62it/s] 
2|5490|Loss: 0.679031252861023:  85%|████████▍ | 5490/6470 [1:03:11<10:37,  1.54it/s]
2|5491|Loss: 0.8464432954788208:  85%|████████▍ | 5490/6470 [1:03:12<10:37,  1.54it/s]
2|5491|Loss: 0.8464432954788208:  85%|████████▍ | 5491/6470 [1:03:12<10:40,  1.53it/s]
2|5492|Loss: 0.7977167963981628:  85%|████████▍ | 5491/6470 [1:03:12<10:40,  1.53it/s]
2|5492|Loss: 0.7977167963981628:  85%|████████▍ | 5492/6470 [1:03:12<09:56,  1.64it/s]
2|5493|Loss: 0.7948848009109497:  85%|████████▍ | 5492/6470 [1:03:13<09:56,  1.64it/s]
2|5493|Loss: 0.7948848009109497:  85%|████████▍ | 5493/6470 [1:03:13<09:48,  1.66it/s]
2|5494|Loss: 0.7979901432991028:  85%|████████▍ | 5493/6470 [1:03:13<09:48,  1.66it/s]
2|5494|Loss: 0.7979901432991028:  85%|████████▍ | 5494/6470 [1:03:14<10:14,  1.59it/s]
2|5495|Loss: 0.8437482714653015:  85%|████████▍ | 5494/6470 [1:03:14<10:14,  1.59it/s]
2|5495|Loss: 0.8437482714653015:  85%|████████▍ | 5495/6470 [1:03:14<09:37,  1.69it/s]
2|5496|Loss: 0.5746607780456543:  85%|████████▍ | 5495/6470 [1:03:14<09:37,  1.69it/s]
2|5496|Loss: 0.5746607780456543:  85%|████████▍ | 5496/6470 [1:03:15<09:12,  1.76it/s]
2|5497|Loss: 0.8678428530693054:  85%|████████▍ | 5496/6470 [1:03:15<09:12,  1.76it/s]
2|5497|Loss: 0.8678428530693054:  85%|████████▍ | 5497/6470 [1:03:15<09:27,  1.71it/s]
2|5498|Loss: 0.7109965085983276:  85%|████████▍ | 5497/6470 [1:03:16<09:27,  1.71it/s]
2|5498|Loss: 0.7109965085983276:  85%|████████▍ | 5498/6470 [1:03:16<09:44,  1.66it/s]
2|5499|Loss: 0.9370387196540833:  85%|████████▍ | 5498/6470 [1:03:16<09:44,  1.66it/s]
2|5499|Loss: 0.9370387196540833:  85%|████████▍ | 5499/6470 [1:03:16<09:19,  1.73it/s]
2|5500|Loss: 0.7713608145713806:  85%|████████▍ | 5499/6470 [1:03:17<09:19,  1.73it/s]
2|5500|Loss: 0.7713608145713806:  85%|████████▌ | 5500/6470 [1:03:17<09:43,  1.66it/s]
2|5501|Loss: 0.9363470077514648:  85%|████████▌ | 5500/6470 [1:03:17<09:43,  1.66it/s]
2|5501|Loss: 0.9363470077514648:  85%|████████▌ | 5501/6470 [1:03:18<09:44,  1.66it/s]
2|5502|Loss: 0.893648624420166:  85%|████████▌ | 5501/6470 [1:03:18<09:44,  1.66it/s] 
2|5502|Loss: 0.893648624420166:  85%|████████▌ | 5502/6470 [1:03:18<10:13,  1.58it/s]
2|5503|Loss: 0.8726136684417725:  85%|████████▌ | 5502/6470 [1:03:19<10:13,  1.58it/s]
2|5503|Loss: 0.8726136684417725:  85%|████████▌ | 5503/6470 [1:03:19<10:14,  1.57it/s]
2|5504|Loss: 1.0373026132583618:  85%|████████▌ | 5503/6470 [1:03:19<10:14,  1.57it/s]
2|5504|Loss: 1.0373026132583618:  85%|████████▌ | 5504/6470 [1:03:20<10:27,  1.54it/s]
2|5505|Loss: 0.903698742389679:  85%|████████▌ | 5504/6470 [1:03:20<10:27,  1.54it/s] 
2|5505|Loss: 0.903698742389679:  85%|████████▌ | 5505/6470 [1:03:20<10:24,  1.54it/s]
2|5506|Loss: 0.9618281126022339:  85%|████████▌ | 5505/6470 [1:03:21<10:24,  1.54it/s]
2|5506|Loss: 0.9618281126022339:  85%|████████▌ | 5506/6470 [1:03:21<10:21,  1.55it/s]
2|5507|Loss: 0.9392973780632019:  85%|████████▌ | 5506/6470 [1:03:21<10:21,  1.55it/s]
2|5507|Loss: 0.9392973780632019:  85%|████████▌ | 5507/6470 [1:03:22<10:13,  1.57it/s]
2|5508|Loss: 0.9247733950614929:  85%|████████▌ | 5507/6470 [1:03:22<10:13,  1.57it/s]
2|5508|Loss: 0.9247733950614929:  85%|████████▌ | 5508/6470 [1:03:22<09:52,  1.62it/s]
2|5509|Loss: 0.9660179615020752:  85%|████████▌ | 5508/6470 [1:03:23<09:52,  1.62it/s]
2|5509|Loss: 0.9660179615020752:  85%|████████▌ | 5509/6470 [1:03:23<09:54,  1.62it/s]
2|5510|Loss: 0.9537214636802673:  85%|████████▌ | 5509/6470 [1:03:23<09:54,  1.62it/s]
2|5510|Loss: 0.9537214636802673:  85%|████████▌ | 5510/6470 [1:03:24<10:22,  1.54it/s]
2|5511|Loss: 0.8742684125900269:  85%|████████▌ | 5510/6470 [1:03:24<10:22,  1.54it/s]
2|5511|Loss: 0.8742684125900269:  85%|████████▌ | 5511/6470 [1:03:24<10:22,  1.54it/s]
2|5512|Loss: 0.8225414156913757:  85%|████████▌ | 5511/6470 [1:03:25<10:22,  1.54it/s]
2|5512|Loss: 0.8225414156913757:  85%|████████▌ | 5512/6470 [1:03:25<09:41,  1.65it/s]
2|5513|Loss: 0.8228313326835632:  85%|████████▌ | 5512/6470 [1:03:25<09:41,  1.65it/s]
2|5513|Loss: 0.8228313326835632:  85%|████████▌ | 5513/6470 [1:03:25<10:01,  1.59it/s]
2|5514|Loss: 0.8754509091377258:  85%|████████▌ | 5513/6470 [1:03:26<10:01,  1.59it/s]
2|5514|Loss: 0.8754509091377258:  85%|████████▌ | 5514/6470 [1:03:26<09:56,  1.60it/s]
2|5515|Loss: 0.8630016446113586:  85%|████████▌ | 5514/6470 [1:03:26<09:56,  1.60it/s]
2|5515|Loss: 0.8630016446113586:  85%|████████▌ | 5515/6470 [1:03:27<10:40,  1.49it/s]
2|5516|Loss: 0.8027273416519165:  85%|████████▌ | 5515/6470 [1:03:27<10:40,  1.49it/s]
2|5516|Loss: 0.8027273416519165:  85%|████████▌ | 5516/6470 [1:03:27<10:38,  1.49it/s]
2|5517|Loss: 0.765957236289978:  85%|████████▌ | 5516/6470 [1:03:28<10:38,  1.49it/s] 
2|5517|Loss: 0.765957236289978:  85%|████████▌ | 5517/6470 [1:03:28<10:22,  1.53it/s]
2|5518|Loss: 0.8161142468452454:  85%|████████▌ | 5517/6470 [1:03:28<10:22,  1.53it/s]
2|5518|Loss: 0.8161142468452454:  85%|████████▌ | 5518/6470 [1:03:29<10:55,  1.45it/s]
2|5519|Loss: 0.8377693295478821:  85%|████████▌ | 5518/6470 [1:03:29<10:55,  1.45it/s]
2|5519|Loss: 0.8377693295478821:  85%|████████▌ | 5519/6470 [1:03:30<10:52,  1.46it/s]
2|5520|Loss: 0.8133144974708557:  85%|████████▌ | 5519/6470 [1:03:30<10:52,  1.46it/s]
2|5520|Loss: 0.8133144974708557:  85%|████████▌ | 5520/6470 [1:03:30<10:38,  1.49it/s]
2|5521|Loss: 0.928162693977356:  85%|████████▌ | 5520/6470 [1:03:31<10:38,  1.49it/s] 
2|5521|Loss: 0.928162693977356:  85%|████████▌ | 5521/6470 [1:03:31<11:23,  1.39it/s]
2|5522|Loss: 0.8307908177375793:  85%|████████▌ | 5521/6470 [1:03:31<11:23,  1.39it/s]
2|5522|Loss: 0.8307908177375793:  85%|████████▌ | 5522/6470 [1:03:32<10:51,  1.45it/s]
2|5523|Loss: 0.8030180335044861:  85%|████████▌ | 5522/6470 [1:03:32<10:51,  1.45it/s]
2|5523|Loss: 0.8030180335044861:  85%|████████▌ | 5523/6470 [1:03:32<10:56,  1.44it/s]
2|5524|Loss: 0.8965604901313782:  85%|████████▌ | 5523/6470 [1:03:33<10:56,  1.44it/s]
2|5524|Loss: 0.8965604901313782:  85%|████████▌ | 5524/6470 [1:03:33<11:44,  1.34it/s]
2|5525|Loss: 0.7563838958740234:  85%|████████▌ | 5524/6470 [1:03:34<11:44,  1.34it/s]
2|5525|Loss: 0.7563838958740234:  85%|████████▌ | 5525/6470 [1:03:34<12:42,  1.24it/s]
2|5526|Loss: 0.7787662148475647:  85%|████████▌ | 5525/6470 [1:03:35<12:42,  1.24it/s]
2|5526|Loss: 0.7787662148475647:  85%|████████▌ | 5526/6470 [1:03:35<13:01,  1.21it/s]
2|5527|Loss: 0.7822064757347107:  85%|████████▌ | 5526/6470 [1:03:35<13:01,  1.21it/s]
2|5527|Loss: 0.7822064757347107:  85%|████████▌ | 5527/6470 [1:03:36<12:16,  1.28it/s]
2|5528|Loss: 1.2218297719955444:  85%|████████▌ | 5527/6470 [1:03:36<12:16,  1.28it/s]
2|5528|Loss: 1.2218297719955444:  85%|████████▌ | 5528/6470 [1:03:36<12:20,  1.27it/s]
2|5529|Loss: 0.8081614971160889:  85%|████████▌ | 5528/6470 [1:03:37<12:20,  1.27it/s]
2|5529|Loss: 0.8081614971160889:  85%|████████▌ | 5529/6470 [1:03:37<12:07,  1.29it/s]
2|5530|Loss: 1.0354925394058228:  85%|████████▌ | 5529/6470 [1:03:38<12:07,  1.29it/s]
2|5530|Loss: 1.0354925394058228:  85%|████████▌ | 5530/6470 [1:03:38<11:28,  1.37it/s]
2|5531|Loss: 0.8282477855682373:  85%|████████▌ | 5530/6470 [1:03:38<11:28,  1.37it/s]
2|5531|Loss: 0.8282477855682373:  85%|████████▌ | 5531/6470 [1:03:39<11:21,  1.38it/s]
2|5532|Loss: 0.9937632083892822:  85%|████████▌ | 5531/6470 [1:03:39<11:21,  1.38it/s]
2|5532|Loss: 0.9937632083892822:  86%|████████▌ | 5532/6470 [1:03:39<11:59,  1.30it/s]
2|5533|Loss: 0.8451195955276489:  86%|████████▌ | 5532/6470 [1:03:40<11:59,  1.30it/s]
2|5533|Loss: 0.8451195955276489:  86%|████████▌ | 5533/6470 [1:03:40<12:13,  1.28it/s]
2|5534|Loss: 0.7518916726112366:  86%|████████▌ | 5533/6470 [1:03:41<12:13,  1.28it/s]
2|5534|Loss: 0.7518916726112366:  86%|████████▌ | 5534/6470 [1:03:41<11:39,  1.34it/s]
2|5535|Loss: 0.8958266377449036:  86%|████████▌ | 5534/6470 [1:03:41<11:39,  1.34it/s]
2|5535|Loss: 0.8958266377449036:  86%|████████▌ | 5535/6470 [1:03:42<11:34,  1.35it/s]
2|5536|Loss: 1.0803829431533813:  86%|████████▌ | 5535/6470 [1:03:42<11:34,  1.35it/s]
2|5536|Loss: 1.0803829431533813:  86%|████████▌ | 5536/6470 [1:03:42<11:59,  1.30it/s]
2|5537|Loss: 0.8548926115036011:  86%|████████▌ | 5536/6470 [1:03:43<11:59,  1.30it/s]
2|5537|Loss: 0.8548926115036011:  86%|████████▌ | 5537/6470 [1:03:43<12:00,  1.29it/s]
2|5538|Loss: 0.7523097395896912:  86%|████████▌ | 5537/6470 [1:03:44<12:00,  1.29it/s]
2|5538|Loss: 0.7523097395896912:  86%|████████▌ | 5538/6470 [1:03:44<11:25,  1.36it/s]
2|5539|Loss: 0.8222934603691101:  86%|████████▌ | 5538/6470 [1:03:44<11:25,  1.36it/s]
2|5539|Loss: 0.8222934603691101:  86%|████████▌ | 5539/6470 [1:03:45<10:49,  1.43it/s]
2|5540|Loss: 0.7741430401802063:  86%|████████▌ | 5539/6470 [1:03:45<10:49,  1.43it/s]
2|5540|Loss: 0.7741430401802063:  86%|████████▌ | 5540/6470 [1:03:45<11:24,  1.36it/s]
2|5541|Loss: 0.7444721460342407:  86%|████████▌ | 5540/6470 [1:03:46<11:24,  1.36it/s]
2|5541|Loss: 0.7444721460342407:  86%|████████▌ | 5541/6470 [1:03:46<11:06,  1.39it/s]
2|5542|Loss: 0.7721437215805054:  86%|████████▌ | 5541/6470 [1:03:46<11:06,  1.39it/s]
2|5542|Loss: 0.7721437215805054:  86%|████████▌ | 5542/6470 [1:03:47<10:37,  1.46it/s]
2|5543|Loss: 0.890677273273468:  86%|████████▌ | 5542/6470 [1:03:47<10:37,  1.46it/s] 
2|5543|Loss: 0.890677273273468:  86%|████████▌ | 5543/6470 [1:03:47<10:47,  1.43it/s]
2|5544|Loss: 0.8572196960449219:  86%|████████▌ | 5543/6470 [1:03:48<10:47,  1.43it/s]
2|5544|Loss: 0.8572196960449219:  86%|████████▌ | 5544/6470 [1:03:48<11:20,  1.36it/s]
2|5545|Loss: 0.7470716238021851:  86%|████████▌ | 5544/6470 [1:03:49<11:20,  1.36it/s]
2|5545|Loss: 0.7470716238021851:  86%|████████▌ | 5545/6470 [1:03:49<10:23,  1.48it/s]
2|5546|Loss: 0.8607630729675293:  86%|████████▌ | 5545/6470 [1:03:49<10:23,  1.48it/s]
2|5546|Loss: 0.8607630729675293:  86%|████████▌ | 5546/6470 [1:03:49<10:23,  1.48it/s]
2|5547|Loss: 0.8739601373672485:  86%|████████▌ | 5546/6470 [1:03:50<10:23,  1.48it/s]
2|5547|Loss: 0.8739601373672485:  86%|████████▌ | 5547/6470 [1:03:50<10:43,  1.43it/s]
2|5548|Loss: 0.8540005683898926:  86%|████████▌ | 5547/6470 [1:03:51<10:43,  1.43it/s]
2|5548|Loss: 0.8540005683898926:  86%|████████▌ | 5548/6470 [1:03:51<10:35,  1.45it/s]
2|5549|Loss: 0.9711350798606873:  86%|████████▌ | 5548/6470 [1:03:51<10:35,  1.45it/s]
2|5549|Loss: 0.9711350798606873:  86%|████████▌ | 5549/6470 [1:03:51<10:21,  1.48it/s]
2|5550|Loss: 0.7328464388847351:  86%|████████▌ | 5549/6470 [1:03:52<10:21,  1.48it/s]
2|5550|Loss: 0.7328464388847351:  86%|████████▌ | 5550/6470 [1:03:52<10:24,  1.47it/s]
2|5551|Loss: 0.8759896755218506:  86%|████████▌ | 5550/6470 [1:03:53<10:24,  1.47it/s]
2|5551|Loss: 0.8759896755218506:  86%|████████▌ | 5551/6470 [1:03:53<10:41,  1.43it/s]
2|5552|Loss: 0.812275230884552:  86%|████████▌ | 5551/6470 [1:03:53<10:41,  1.43it/s] 
2|5552|Loss: 0.812275230884552:  86%|████████▌ | 5552/6470 [1:03:54<11:15,  1.36it/s]
2|5553|Loss: 0.8202295303344727:  86%|████████▌ | 5552/6470 [1:03:54<11:15,  1.36it/s]
2|5553|Loss: 0.8202295303344727:  86%|████████▌ | 5553/6470 [1:03:54<11:15,  1.36it/s]
2|5554|Loss: 1.0519399642944336:  86%|████████▌ | 5553/6470 [1:03:55<11:15,  1.36it/s]
2|5554|Loss: 1.0519399642944336:  86%|████████▌ | 5554/6470 [1:03:55<11:04,  1.38it/s]
2|5555|Loss: 0.7907003164291382:  86%|████████▌ | 5554/6470 [1:03:55<11:04,  1.38it/s]
2|5555|Loss: 0.7907003164291382:  86%|████████▌ | 5555/6470 [1:03:56<10:22,  1.47it/s]
2|5556|Loss: 0.8954747915267944:  86%|████████▌ | 5555/6470 [1:03:56<10:22,  1.47it/s]
2|5556|Loss: 0.8954747915267944:  86%|████████▌ | 5556/6470 [1:03:56<10:29,  1.45it/s]
2|5557|Loss: 0.8675605058670044:  86%|████████▌ | 5556/6470 [1:03:57<10:29,  1.45it/s]
2|5557|Loss: 0.8675605058670044:  86%|████████▌ | 5557/6470 [1:03:57<11:04,  1.37it/s]
2|5558|Loss: 0.9391257166862488:  86%|████████▌ | 5557/6470 [1:03:58<11:04,  1.37it/s]
2|5558|Loss: 0.9391257166862488:  86%|████████▌ | 5558/6470 [1:03:58<10:52,  1.40it/s]
2|5559|Loss: 0.9470787644386292:  86%|████████▌ | 5558/6470 [1:03:58<10:52,  1.40it/s]
2|5559|Loss: 0.9470787644386292:  86%|████████▌ | 5559/6470 [1:03:59<11:23,  1.33it/s]
2|5560|Loss: 0.7850272059440613:  86%|████████▌ | 5559/6470 [1:03:59<11:23,  1.33it/s]
2|5560|Loss: 0.7850272059440613:  86%|████████▌ | 5560/6470 [1:03:59<11:07,  1.36it/s]
2|5561|Loss: 0.9121657013893127:  86%|████████▌ | 5560/6470 [1:04:00<11:07,  1.36it/s]
2|5561|Loss: 0.9121657013893127:  86%|████████▌ | 5561/6470 [1:04:00<11:06,  1.36it/s]
2|5562|Loss: 0.7883468866348267:  86%|████████▌ | 5561/6470 [1:04:01<11:06,  1.36it/s]
2|5562|Loss: 0.7883468866348267:  86%|████████▌ | 5562/6470 [1:04:01<10:43,  1.41it/s]
2|5563|Loss: 0.8673391938209534:  86%|████████▌ | 5562/6470 [1:04:01<10:43,  1.41it/s]
2|5563|Loss: 0.8673391938209534:  86%|████████▌ | 5563/6470 [1:04:02<10:38,  1.42it/s]
2|5564|Loss: 0.8520849347114563:  86%|████████▌ | 5563/6470 [1:04:02<10:38,  1.42it/s]
2|5564|Loss: 0.8520849347114563:  86%|████████▌ | 5564/6470 [1:04:02<10:46,  1.40it/s]
2|5565|Loss: 0.7231133580207825:  86%|████████▌ | 5564/6470 [1:04:03<10:46,  1.40it/s]
2|5565|Loss: 0.7231133580207825:  86%|████████▌ | 5565/6470 [1:04:03<10:52,  1.39it/s]
2|5566|Loss: 0.7897903323173523:  86%|████████▌ | 5565/6470 [1:04:03<10:52,  1.39it/s]
2|5566|Loss: 0.7897903323173523:  86%|████████▌ | 5566/6470 [1:04:04<11:09,  1.35it/s]
2|5567|Loss: 0.8450492024421692:  86%|████████▌ | 5566/6470 [1:04:04<11:09,  1.35it/s]
2|5567|Loss: 0.8450492024421692:  86%|████████▌ | 5567/6470 [1:04:04<10:17,  1.46it/s]
2|5568|Loss: 0.8688540458679199:  86%|████████▌ | 5567/6470 [1:04:05<10:17,  1.46it/s]
2|5568|Loss: 0.8688540458679199:  86%|████████▌ | 5568/6470 [1:04:05<09:43,  1.55it/s]
2|5569|Loss: 0.898512065410614:  86%|████████▌ | 5568/6470 [1:04:05<09:43,  1.55it/s] 
2|5569|Loss: 0.898512065410614:  86%|████████▌ | 5569/6470 [1:04:05<09:33,  1.57it/s]
2|5570|Loss: 0.7870478630065918:  86%|████████▌ | 5569/6470 [1:04:06<09:33,  1.57it/s]
2|5570|Loss: 0.7870478630065918:  86%|████████▌ | 5570/6470 [1:04:06<09:34,  1.57it/s]
2|5571|Loss: 0.7879921793937683:  86%|████████▌ | 5570/6470 [1:04:07<09:34,  1.57it/s]
2|5571|Loss: 0.7879921793937683:  86%|████████▌ | 5571/6470 [1:04:07<10:16,  1.46it/s]
2|5572|Loss: 0.9155194163322449:  86%|████████▌ | 5571/6470 [1:04:07<10:16,  1.46it/s]
2|5572|Loss: 0.9155194163322449:  86%|████████▌ | 5572/6470 [1:04:08<11:00,  1.36it/s]
2|5573|Loss: 0.8694784641265869:  86%|████████▌ | 5572/6470 [1:04:08<11:00,  1.36it/s]
2|5573|Loss: 0.8694784641265869:  86%|████████▌ | 5573/6470 [1:04:09<11:19,  1.32it/s]
2|5574|Loss: 0.8175442814826965:  86%|████████▌ | 5573/6470 [1:04:09<11:19,  1.32it/s]
2|5574|Loss: 0.8175442814826965:  86%|████████▌ | 5574/6470 [1:04:09<10:51,  1.38it/s]
2|5575|Loss: 0.8223522305488586:  86%|████████▌ | 5574/6470 [1:04:10<10:51,  1.38it/s]
2|5575|Loss: 0.8223522305488586:  86%|████████▌ | 5575/6470 [1:04:10<10:25,  1.43it/s]
2|5576|Loss: 0.9164561629295349:  86%|████████▌ | 5575/6470 [1:04:10<10:25,  1.43it/s]
2|5576|Loss: 0.9164561629295349:  86%|████████▌ | 5576/6470 [1:04:10<09:28,  1.57it/s]
2|5577|Loss: 0.6940905451774597:  86%|████████▌ | 5576/6470 [1:04:11<09:28,  1.57it/s]
2|5577|Loss: 0.6940905451774597:  86%|████████▌ | 5577/6470 [1:04:11<08:50,  1.68it/s]
2|5578|Loss: 0.9029515385627747:  86%|████████▌ | 5577/6470 [1:04:11<08:50,  1.68it/s]
2|5578|Loss: 0.9029515385627747:  86%|████████▌ | 5578/6470 [1:04:12<09:09,  1.62it/s]
2|5579|Loss: 0.9258826971054077:  86%|████████▌ | 5578/6470 [1:04:12<09:09,  1.62it/s]
2|5579|Loss: 0.9258826971054077:  86%|████████▌ | 5579/6470 [1:04:12<09:14,  1.61it/s]
2|5580|Loss: 0.8431500792503357:  86%|████████▌ | 5579/6470 [1:04:13<09:14,  1.61it/s]
2|5580|Loss: 0.8431500792503357:  86%|████████▌ | 5580/6470 [1:04:13<09:39,  1.54it/s]
2|5581|Loss: 0.8712019324302673:  86%|████████▌ | 5580/6470 [1:04:13<09:39,  1.54it/s]
2|5581|Loss: 0.8712019324302673:  86%|████████▋ | 5581/6470 [1:04:14<09:44,  1.52it/s]
2|5582|Loss: 1.1215763092041016:  86%|████████▋ | 5581/6470 [1:04:14<09:44,  1.52it/s]
2|5582|Loss: 1.1215763092041016:  86%|████████▋ | 5582/6470 [1:04:14<09:33,  1.55it/s]
2|5583|Loss: 0.7222482562065125:  86%|████████▋ | 5582/6470 [1:04:15<09:33,  1.55it/s]
2|5583|Loss: 0.7222482562065125:  86%|████████▋ | 5583/6470 [1:04:15<10:44,  1.38it/s]
2|5584|Loss: 0.610833466053009:  86%|████████▋ | 5583/6470 [1:04:16<10:44,  1.38it/s] 
2|5584|Loss: 0.610833466053009:  86%|████████▋ | 5584/6470 [1:04:16<11:13,  1.32it/s]
2|5585|Loss: 0.896477460861206:  86%|████████▋ | 5584/6470 [1:04:16<11:13,  1.32it/s]
2|5585|Loss: 0.896477460861206:  86%|████████▋ | 5585/6470 [1:04:17<10:49,  1.36it/s]
2|5586|Loss: 0.8559116125106812:  86%|████████▋ | 5585/6470 [1:04:17<10:49,  1.36it/s]
2|5586|Loss: 0.8559116125106812:  86%|████████▋ | 5586/6470 [1:04:17<10:42,  1.38it/s]
2|5587|Loss: 0.9050644636154175:  86%|████████▋ | 5586/6470 [1:04:18<10:42,  1.38it/s]
2|5587|Loss: 0.9050644636154175:  86%|████████▋ | 5587/6470 [1:04:18<09:43,  1.51it/s]
2|5588|Loss: 0.7406381964683533:  86%|████████▋ | 5587/6470 [1:04:18<09:43,  1.51it/s]
2|5588|Loss: 0.7406381964683533:  86%|████████▋ | 5588/6470 [1:04:18<09:22,  1.57it/s]
2|5589|Loss: 0.7707579731941223:  86%|████████▋ | 5588/6470 [1:04:19<09:22,  1.57it/s]
2|5589|Loss: 0.7707579731941223:  86%|████████▋ | 5589/6470 [1:04:19<09:55,  1.48it/s]
2|5590|Loss: 0.6525301933288574:  86%|████████▋ | 5589/6470 [1:04:20<09:55,  1.48it/s]
2|5590|Loss: 0.6525301933288574:  86%|████████▋ | 5590/6470 [1:04:20<10:34,  1.39it/s]
2|5591|Loss: 0.7699320316314697:  86%|████████▋ | 5590/6470 [1:04:20<10:34,  1.39it/s]
2|5591|Loss: 0.7699320316314697:  86%|████████▋ | 5591/6470 [1:04:21<10:24,  1.41it/s]
2|5592|Loss: 0.71551913022995:  86%|████████▋ | 5591/6470 [1:04:21<10:24,  1.41it/s]  
2|5592|Loss: 0.71551913022995:  86%|████████▋ | 5592/6470 [1:04:21<10:17,  1.42it/s]
2|5593|Loss: 0.7869563698768616:  86%|████████▋ | 5592/6470 [1:04:22<10:17,  1.42it/s]
2|5593|Loss: 0.7869563698768616:  86%|████████▋ | 5593/6470 [1:04:22<09:52,  1.48it/s]
2|5594|Loss: 0.8495978713035583:  86%|████████▋ | 5593/6470 [1:04:22<09:52,  1.48it/s]
2|5594|Loss: 0.8495978713035583:  86%|████████▋ | 5594/6470 [1:04:23<10:19,  1.41it/s]
2|5595|Loss: 1.1992788314819336:  86%|████████▋ | 5594/6470 [1:04:23<10:19,  1.41it/s]
2|5595|Loss: 1.1992788314819336:  86%|████████▋ | 5595/6470 [1:04:23<10:07,  1.44it/s]
2|5596|Loss: 0.9775765538215637:  86%|████████▋ | 5595/6470 [1:04:24<10:07,  1.44it/s]
2|5596|Loss: 0.9775765538215637:  86%|████████▋ | 5596/6470 [1:04:24<09:36,  1.52it/s]
2|5597|Loss: 0.8690366744995117:  86%|████████▋ | 5596/6470 [1:04:24<09:36,  1.52it/s]
2|5597|Loss: 0.8690366744995117:  87%|████████▋ | 5597/6470 [1:04:25<09:00,  1.61it/s]
2|5598|Loss: 0.6944844722747803:  87%|████████▋ | 5597/6470 [1:04:25<09:00,  1.61it/s]
2|5598|Loss: 0.6944844722747803:  87%|████████▋ | 5598/6470 [1:04:25<09:19,  1.56it/s]
2|5599|Loss: 0.8754251599311829:  87%|████████▋ | 5598/6470 [1:04:26<09:19,  1.56it/s]
2|5599|Loss: 0.8754251599311829:  87%|████████▋ | 5599/6470 [1:04:26<09:25,  1.54it/s]
2|5600|Loss: 0.8359248042106628:  87%|████████▋ | 5599/6470 [1:04:26<09:25,  1.54it/s]
2|5600|Loss: 0.8359248042106628:  87%|████████▋ | 5600/6470 [1:04:27<09:41,  1.50it/s]
2|5601|Loss: 0.9042612910270691:  87%|████████▋ | 5600/6470 [1:04:27<09:41,  1.50it/s]
2|5601|Loss: 0.9042612910270691:  87%|████████▋ | 5601/6470 [1:04:27<09:50,  1.47it/s]
2|5602|Loss: 0.8328280448913574:  87%|████████▋ | 5601/6470 [1:04:28<09:50,  1.47it/s]
2|5602|Loss: 0.8328280448913574:  87%|████████▋ | 5602/6470 [1:04:28<10:24,  1.39it/s]
2|5603|Loss: 0.9665919542312622:  87%|████████▋ | 5602/6470 [1:04:29<10:24,  1.39it/s]
2|5603|Loss: 0.9665919542312622:  87%|████████▋ | 5603/6470 [1:04:29<10:32,  1.37it/s]
2|5604|Loss: 0.8483253121376038:  87%|████████▋ | 5603/6470 [1:04:29<10:32,  1.37it/s]
2|5604|Loss: 0.8483253121376038:  87%|████████▋ | 5604/6470 [1:04:30<10:03,  1.43it/s]
2|5605|Loss: 0.873716413974762:  87%|████████▋ | 5604/6470 [1:04:30<10:03,  1.43it/s] 
2|5605|Loss: 0.873716413974762:  87%|████████▋ | 5605/6470 [1:04:30<10:29,  1.37it/s]
2|5606|Loss: 0.7672588229179382:  87%|████████▋ | 5605/6470 [1:04:31<10:29,  1.37it/s]
2|5606|Loss: 0.7672588229179382:  87%|████████▋ | 5606/6470 [1:04:31<11:01,  1.31it/s]
2|5607|Loss: 0.7372820973396301:  87%|████████▋ | 5606/6470 [1:04:32<11:01,  1.31it/s]
2|5607|Loss: 0.7372820973396301:  87%|████████▋ | 5607/6470 [1:04:32<10:41,  1.34it/s]
2|5608|Loss: 0.8737137317657471:  87%|████████▋ | 5607/6470 [1:04:32<10:41,  1.34it/s]
2|5608|Loss: 0.8737137317657471:  87%|████████▋ | 5608/6470 [1:04:32<10:14,  1.40it/s]
2|5609|Loss: 0.9844076037406921:  87%|████████▋ | 5608/6470 [1:04:33<10:14,  1.40it/s]
2|5609|Loss: 0.9844076037406921:  87%|████████▋ | 5609/6470 [1:04:33<10:13,  1.40it/s]
2|5610|Loss: 0.9013636708259583:  87%|████████▋ | 5609/6470 [1:04:34<10:13,  1.40it/s]
2|5610|Loss: 0.9013636708259583:  87%|████████▋ | 5610/6470 [1:04:34<09:44,  1.47it/s]
2|5611|Loss: 0.8535168170928955:  87%|████████▋ | 5610/6470 [1:04:34<09:44,  1.47it/s]
2|5611|Loss: 0.8535168170928955:  87%|████████▋ | 5611/6470 [1:04:34<09:29,  1.51it/s]
2|5612|Loss: 0.9251901507377625:  87%|████████▋ | 5611/6470 [1:04:35<09:29,  1.51it/s]
2|5612|Loss: 0.9251901507377625:  87%|████████▋ | 5612/6470 [1:04:35<09:57,  1.44it/s]
2|5613|Loss: 1.0023092031478882:  87%|████████▋ | 5612/6470 [1:04:36<09:57,  1.44it/s]
2|5613|Loss: 1.0023092031478882:  87%|████████▋ | 5613/6470 [1:04:36<09:38,  1.48it/s]
2|5614|Loss: 0.8424273729324341:  87%|████████▋ | 5613/6470 [1:04:36<09:38,  1.48it/s]
2|5614|Loss: 0.8424273729324341:  87%|████████▋ | 5614/6470 [1:04:36<09:27,  1.51it/s]
2|5615|Loss: 0.7640835642814636:  87%|████████▋ | 5614/6470 [1:04:37<09:27,  1.51it/s]
2|5615|Loss: 0.7640835642814636:  87%|████████▋ | 5615/6470 [1:04:37<08:58,  1.59it/s]
2|5616|Loss: 0.7883039116859436:  87%|████████▋ | 5615/6470 [1:04:37<08:58,  1.59it/s]
2|5616|Loss: 0.7883039116859436:  87%|████████▋ | 5616/6470 [1:04:38<09:39,  1.47it/s]
2|5617|Loss: 0.6553460955619812:  87%|████████▋ | 5616/6470 [1:04:38<09:39,  1.47it/s]
2|5617|Loss: 0.6553460955619812:  87%|████████▋ | 5617/6470 [1:04:38<08:39,  1.64it/s]
2|5618|Loss: 0.8057763576507568:  87%|████████▋ | 5617/6470 [1:04:38<08:39,  1.64it/s]
2|5618|Loss: 0.8057763576507568:  87%|████████▋ | 5618/6470 [1:04:39<07:50,  1.81it/s]
2|5619|Loss: 0.9147480130195618:  87%|████████▋ | 5618/6470 [1:04:39<07:50,  1.81it/s]
2|5619|Loss: 0.9147480130195618:  87%|████████▋ | 5619/6470 [1:04:39<08:29,  1.67it/s]
2|5620|Loss: 0.7400073409080505:  87%|████████▋ | 5619/6470 [1:04:40<08:29,  1.67it/s]
2|5620|Loss: 0.7400073409080505:  87%|████████▋ | 5620/6470 [1:04:40<09:25,  1.50it/s]
2|5621|Loss: 1.0357348918914795:  87%|████████▋ | 5620/6470 [1:04:41<09:25,  1.50it/s]
2|5621|Loss: 1.0357348918914795:  87%|████████▋ | 5621/6470 [1:04:41<09:25,  1.50it/s]
2|5622|Loss: 0.6866810917854309:  87%|████████▋ | 5621/6470 [1:04:41<09:25,  1.50it/s]
2|5622|Loss: 0.6866810917854309:  87%|████████▋ | 5622/6470 [1:04:42<09:30,  1.49it/s]
2|5623|Loss: 1.0965683460235596:  87%|████████▋ | 5622/6470 [1:04:42<09:30,  1.49it/s]
2|5623|Loss: 1.0965683460235596:  87%|████████▋ | 5623/6470 [1:04:42<09:24,  1.50it/s]
2|5624|Loss: 0.8490090370178223:  87%|████████▋ | 5623/6470 [1:04:43<09:24,  1.50it/s]
2|5624|Loss: 0.8490090370178223:  87%|████████▋ | 5624/6470 [1:04:43<08:58,  1.57it/s]
2|5625|Loss: 1.0415126085281372:  87%|████████▋ | 5624/6470 [1:04:43<08:58,  1.57it/s]
2|5625|Loss: 1.0415126085281372:  87%|████████▋ | 5625/6470 [1:04:43<09:17,  1.52it/s]
2|5626|Loss: 0.9220183491706848:  87%|████████▋ | 5625/6470 [1:04:44<09:17,  1.52it/s]
2|5626|Loss: 0.9220183491706848:  87%|████████▋ | 5626/6470 [1:04:44<09:38,  1.46it/s]
2|5627|Loss: 0.8114979267120361:  87%|████████▋ | 5626/6470 [1:04:45<09:38,  1.46it/s]
2|5627|Loss: 0.8114979267120361:  87%|████████▋ | 5627/6470 [1:04:45<09:49,  1.43it/s]
2|5628|Loss: 0.9078156352043152:  87%|████████▋ | 5627/6470 [1:04:45<09:49,  1.43it/s]
2|5628|Loss: 0.9078156352043152:  87%|████████▋ | 5628/6470 [1:04:46<09:19,  1.50it/s]
2|5629|Loss: 0.8884027004241943:  87%|████████▋ | 5628/6470 [1:04:46<09:19,  1.50it/s]
2|5629|Loss: 0.8884027004241943:  87%|████████▋ | 5629/6470 [1:04:46<09:22,  1.50it/s]
2|5630|Loss: 0.6804958581924438:  87%|████████▋ | 5629/6470 [1:04:47<09:22,  1.50it/s]
2|5630|Loss: 0.6804958581924438:  87%|████████▋ | 5630/6470 [1:04:47<08:47,  1.59it/s]
2|5631|Loss: 0.8683838248252869:  87%|████████▋ | 5630/6470 [1:04:47<08:47,  1.59it/s]
2|5631|Loss: 0.8683838248252869:  87%|████████▋ | 5631/6470 [1:04:47<08:41,  1.61it/s]
2|5632|Loss: 0.983920693397522:  87%|████████▋ | 5631/6470 [1:04:48<08:41,  1.61it/s] 
2|5632|Loss: 0.983920693397522:  87%|████████▋ | 5632/6470 [1:04:48<08:59,  1.55it/s]
2|5633|Loss: 0.8877797722816467:  87%|████████▋ | 5632/6470 [1:04:49<08:59,  1.55it/s]
2|5633|Loss: 0.8877797722816467:  87%|████████▋ | 5633/6470 [1:04:49<09:48,  1.42it/s]
2|5634|Loss: 0.970931351184845:  87%|████████▋ | 5633/6470 [1:04:49<09:48,  1.42it/s] 
2|5634|Loss: 0.970931351184845:  87%|████████▋ | 5634/6470 [1:04:50<09:58,  1.40it/s]
2|5635|Loss: 0.7061505317687988:  87%|████████▋ | 5634/6470 [1:04:50<09:58,  1.40it/s]
2|5635|Loss: 0.7061505317687988:  87%|████████▋ | 5635/6470 [1:04:50<09:55,  1.40it/s]
2|5636|Loss: 0.8420352935791016:  87%|████████▋ | 5635/6470 [1:04:51<09:55,  1.40it/s]
2|5636|Loss: 0.8420352935791016:  87%|████████▋ | 5636/6470 [1:04:51<09:28,  1.47it/s]
2|5637|Loss: 0.88329017162323:  87%|████████▋ | 5636/6470 [1:04:51<09:28,  1.47it/s]  
2|5637|Loss: 0.88329017162323:  87%|████████▋ | 5637/6470 [1:04:52<09:27,  1.47it/s]
2|5638|Loss: 0.8832099437713623:  87%|████████▋ | 5637/6470 [1:04:52<09:27,  1.47it/s]
2|5638|Loss: 0.8832099437713623:  87%|████████▋ | 5638/6470 [1:04:52<09:33,  1.45it/s]
2|5639|Loss: 0.7548157572746277:  87%|████████▋ | 5638/6470 [1:04:53<09:33,  1.45it/s]
2|5639|Loss: 0.7548157572746277:  87%|████████▋ | 5639/6470 [1:04:53<09:35,  1.44it/s]
2|5640|Loss: 0.8765833973884583:  87%|████████▋ | 5639/6470 [1:04:53<09:35,  1.44it/s]
2|5640|Loss: 0.8765833973884583:  87%|████████▋ | 5640/6470 [1:04:54<09:24,  1.47it/s]
2|5641|Loss: 0.8677715063095093:  87%|████████▋ | 5640/6470 [1:04:54<09:24,  1.47it/s]
2|5641|Loss: 0.8677715063095093:  87%|████████▋ | 5641/6470 [1:04:54<09:32,  1.45it/s]
2|5642|Loss: 0.8942207098007202:  87%|████████▋ | 5641/6470 [1:04:55<09:32,  1.45it/s]
2|5642|Loss: 0.8942207098007202:  87%|████████▋ | 5642/6470 [1:04:55<10:21,  1.33it/s]
2|5643|Loss: 0.8359122276306152:  87%|████████▋ | 5642/6470 [1:04:56<10:21,  1.33it/s]
2|5643|Loss: 0.8359122276306152:  87%|████████▋ | 5643/6470 [1:04:56<10:21,  1.33it/s]
2|5644|Loss: 0.9745283126831055:  87%|████████▋ | 5643/6470 [1:04:56<10:21,  1.33it/s]
2|5644|Loss: 0.9745283126831055:  87%|████████▋ | 5644/6470 [1:04:57<09:53,  1.39it/s]
2|5645|Loss: 0.8146023750305176:  87%|████████▋ | 5644/6470 [1:04:57<09:53,  1.39it/s]
2|5645|Loss: 0.8146023750305176:  87%|████████▋ | 5645/6470 [1:04:57<10:10,  1.35it/s]
2|5646|Loss: 0.7621311545372009:  87%|████████▋ | 5645/6470 [1:04:58<10:10,  1.35it/s]
2|5646|Loss: 0.7621311545372009:  87%|████████▋ | 5646/6470 [1:04:58<10:13,  1.34it/s]
2|5647|Loss: 0.8239048719406128:  87%|████████▋ | 5646/6470 [1:04:59<10:13,  1.34it/s]
2|5647|Loss: 0.8239048719406128:  87%|████████▋ | 5647/6470 [1:04:59<09:25,  1.45it/s]
2|5648|Loss: 0.7570332288742065:  87%|████████▋ | 5647/6470 [1:04:59<09:25,  1.45it/s]
2|5648|Loss: 0.7570332288742065:  87%|████████▋ | 5648/6470 [1:04:59<09:18,  1.47it/s]
2|5649|Loss: 0.888870358467102:  87%|████████▋ | 5648/6470 [1:05:00<09:18,  1.47it/s] 
2|5649|Loss: 0.888870358467102:  87%|████████▋ | 5649/6470 [1:05:00<09:13,  1.48it/s]
2|5650|Loss: 0.8087722659111023:  87%|████████▋ | 5649/6470 [1:05:00<09:13,  1.48it/s]
2|5650|Loss: 0.8087722659111023:  87%|████████▋ | 5650/6470 [1:05:01<08:59,  1.52it/s]
2|5651|Loss: 0.9760240316390991:  87%|████████▋ | 5650/6470 [1:05:01<08:59,  1.52it/s]
2|5651|Loss: 0.9760240316390991:  87%|████████▋ | 5651/6470 [1:05:02<09:22,  1.45it/s]
2|5652|Loss: 0.84745854139328:  87%|████████▋ | 5651/6470 [1:05:02<09:22,  1.45it/s]  
2|5652|Loss: 0.84745854139328:  87%|████████▋ | 5652/6470 [1:05:02<09:37,  1.42it/s]
2|5653|Loss: 0.8288485407829285:  87%|████████▋ | 5652/6470 [1:05:03<09:37,  1.42it/s]
2|5653|Loss: 0.8288485407829285:  87%|████████▋ | 5653/6470 [1:05:03<09:13,  1.48it/s]
2|5654|Loss: 0.9114564061164856:  87%|████████▋ | 5653/6470 [1:05:03<09:13,  1.48it/s]
2|5654|Loss: 0.9114564061164856:  87%|████████▋ | 5654/6470 [1:05:04<09:16,  1.47it/s]
2|5655|Loss: 0.885612428188324:  87%|████████▋ | 5654/6470 [1:05:04<09:16,  1.47it/s] 
2|5655|Loss: 0.885612428188324:  87%|████████▋ | 5655/6470 [1:05:04<09:09,  1.48it/s]
2|5656|Loss: 0.9574824571609497:  87%|████████▋ | 5655/6470 [1:05:05<09:09,  1.48it/s]
2|5656|Loss: 0.9574824571609497:  87%|████████▋ | 5656/6470 [1:05:05<09:01,  1.50it/s]
2|5657|Loss: 0.7410132884979248:  87%|████████▋ | 5656/6470 [1:05:05<09:01,  1.50it/s]
2|5657|Loss: 0.7410132884979248:  87%|████████▋ | 5657/6470 [1:05:06<09:31,  1.42it/s]
2|5658|Loss: 0.864706814289093:  87%|████████▋ | 5657/6470 [1:05:06<09:31,  1.42it/s] 
2|5658|Loss: 0.864706814289093:  87%|████████▋ | 5658/6470 [1:05:06<09:38,  1.40it/s]
2|5659|Loss: 0.8180203437805176:  87%|████████▋ | 5658/6470 [1:05:07<09:38,  1.40it/s]
2|5659|Loss: 0.8180203437805176:  87%|████████▋ | 5659/6470 [1:05:07<09:18,  1.45it/s]
2|5660|Loss: 0.920013964176178:  87%|████████▋ | 5659/6470 [1:05:07<09:18,  1.45it/s] 
2|5660|Loss: 0.920013964176178:  87%|████████▋ | 5660/6470 [1:05:08<08:59,  1.50it/s]
2|5661|Loss: 0.7927793860435486:  87%|████████▋ | 5660/6470 [1:05:08<08:59,  1.50it/s]
2|5661|Loss: 0.7927793860435486:  87%|████████▋ | 5661/6470 [1:05:08<09:28,  1.42it/s]
2|5662|Loss: 0.9128792881965637:  87%|████████▋ | 5661/6470 [1:05:09<09:28,  1.42it/s]
2|5662|Loss: 0.9128792881965637:  88%|████████▊ | 5662/6470 [1:05:09<10:04,  1.34it/s]
2|5663|Loss: 0.718411922454834:  88%|████████▊ | 5662/6470 [1:05:10<10:04,  1.34it/s] 
2|5663|Loss: 0.718411922454834:  88%|████████▊ | 5663/6470 [1:05:10<10:24,  1.29it/s]
2|5664|Loss: 0.9405090808868408:  88%|████████▊ | 5663/6470 [1:05:11<10:24,  1.29it/s]
2|5664|Loss: 0.9405090808868408:  88%|████████▊ | 5664/6470 [1:05:11<10:22,  1.30it/s]
2|5665|Loss: 0.6916870474815369:  88%|████████▊ | 5664/6470 [1:05:11<10:22,  1.30it/s]
2|5665|Loss: 0.6916870474815369:  88%|████████▊ | 5665/6470 [1:05:12<10:12,  1.31it/s]
2|5666|Loss: 0.726473331451416:  88%|████████▊ | 5665/6470 [1:05:12<10:12,  1.31it/s] 
2|5666|Loss: 0.726473331451416:  88%|████████▊ | 5666/6470 [1:05:13<10:59,  1.22it/s]
2|5667|Loss: 0.9012454748153687:  88%|████████▊ | 5666/6470 [1:05:13<10:59,  1.22it/s]
2|5667|Loss: 0.9012454748153687:  88%|████████▊ | 5667/6470 [1:05:13<11:23,  1.17it/s]
2|5668|Loss: 0.9230424761772156:  88%|████████▊ | 5667/6470 [1:05:14<11:23,  1.17it/s]
2|5668|Loss: 0.9230424761772156:  88%|████████▊ | 5668/6470 [1:05:14<11:27,  1.17it/s]
2|5669|Loss: 0.8273683190345764:  88%|████████▊ | 5668/6470 [1:05:15<11:27,  1.17it/s]
2|5669|Loss: 0.8273683190345764:  88%|████████▊ | 5669/6470 [1:05:15<10:42,  1.25it/s]
2|5670|Loss: 1.0240728855133057:  88%|████████▊ | 5669/6470 [1:05:15<10:42,  1.25it/s]
2|5670|Loss: 1.0240728855133057:  88%|████████▊ | 5670/6470 [1:05:16<10:45,  1.24it/s]
2|5671|Loss: 0.7526516318321228:  88%|████████▊ | 5670/6470 [1:05:16<10:45,  1.24it/s]
2|5671|Loss: 0.7526516318321228:  88%|████████▊ | 5671/6470 [1:05:16<09:29,  1.40it/s]
2|5672|Loss: 0.8342193961143494:  88%|████████▊ | 5671/6470 [1:05:17<09:29,  1.40it/s]
2|5672|Loss: 0.8342193961143494:  88%|████████▊ | 5672/6470 [1:05:17<09:12,  1.44it/s]
2|5673|Loss: 0.8193632960319519:  88%|████████▊ | 5672/6470 [1:05:17<09:12,  1.44it/s]
2|5673|Loss: 0.8193632960319519:  88%|████████▊ | 5673/6470 [1:05:18<09:24,  1.41it/s]
2|5674|Loss: 0.7139831781387329:  88%|████████▊ | 5673/6470 [1:05:18<09:24,  1.41it/s]
2|5674|Loss: 0.7139831781387329:  88%|████████▊ | 5674/6470 [1:05:18<09:34,  1.39it/s]
2|5675|Loss: 0.861734926700592:  88%|████████▊ | 5674/6470 [1:05:19<09:34,  1.39it/s] 
2|5675|Loss: 0.861734926700592:  88%|████████▊ | 5675/6470 [1:05:19<09:31,  1.39it/s]
2|5676|Loss: 0.9701372385025024:  88%|████████▊ | 5675/6470 [1:05:20<09:31,  1.39it/s]
2|5676|Loss: 0.9701372385025024:  88%|████████▊ | 5676/6470 [1:05:20<09:14,  1.43it/s]
2|5677|Loss: 0.9031169414520264:  88%|████████▊ | 5676/6470 [1:05:20<09:14,  1.43it/s]
2|5677|Loss: 0.9031169414520264:  88%|████████▊ | 5677/6470 [1:05:20<09:00,  1.47it/s]
2|5678|Loss: 0.9820899963378906:  88%|████████▊ | 5677/6470 [1:05:21<09:00,  1.47it/s]
2|5678|Loss: 0.9820899963378906:  88%|████████▊ | 5678/6470 [1:05:21<09:09,  1.44it/s]
2|5679|Loss: 0.9452453255653381:  88%|████████▊ | 5678/6470 [1:05:22<09:09,  1.44it/s]
2|5679|Loss: 0.9452453255653381:  88%|████████▊ | 5679/6470 [1:05:22<08:40,  1.52it/s]
2|5680|Loss: 0.8880308270454407:  88%|████████▊ | 5679/6470 [1:05:22<08:40,  1.52it/s]
2|5680|Loss: 0.8880308270454407:  88%|████████▊ | 5680/6470 [1:05:22<08:49,  1.49it/s]
2|5681|Loss: 0.7675424218177795:  88%|████████▊ | 5680/6470 [1:05:23<08:49,  1.49it/s]
2|5681|Loss: 0.7675424218177795:  88%|████████▊ | 5681/6470 [1:05:23<09:00,  1.46it/s]
2|5682|Loss: 0.8320156335830688:  88%|████████▊ | 5681/6470 [1:05:24<09:00,  1.46it/s]
2|5682|Loss: 0.8320156335830688:  88%|████████▊ | 5682/6470 [1:05:24<09:10,  1.43it/s]
2|5683|Loss: 0.823496401309967:  88%|████████▊ | 5682/6470 [1:05:24<09:10,  1.43it/s] 
2|5683|Loss: 0.823496401309967:  88%|████████▊ | 5683/6470 [1:05:25<09:17,  1.41it/s]
2|5684|Loss: 0.8640298247337341:  88%|████████▊ | 5683/6470 [1:05:25<09:17,  1.41it/s]
2|5684|Loss: 0.8640298247337341:  88%|████████▊ | 5684/6470 [1:05:25<09:22,  1.40it/s]
2|5685|Loss: 0.825763463973999:  88%|████████▊ | 5684/6470 [1:05:26<09:22,  1.40it/s] 
2|5685|Loss: 0.825763463973999:  88%|████████▊ | 5685/6470 [1:05:26<09:28,  1.38it/s]
2|5686|Loss: 0.8056372404098511:  88%|████████▊ | 5685/6470 [1:05:27<09:28,  1.38it/s]
2|5686|Loss: 0.8056372404098511:  88%|████████▊ | 5686/6470 [1:05:27<09:14,  1.41it/s]
2|5687|Loss: 0.88349848985672:  88%|████████▊ | 5686/6470 [1:05:27<09:14,  1.41it/s]  
2|5687|Loss: 0.88349848985672:  88%|████████▊ | 5687/6470 [1:05:28<09:39,  1.35it/s]
2|5688|Loss: 0.9582092761993408:  88%|████████▊ | 5687/6470 [1:05:28<09:39,  1.35it/s]
2|5688|Loss: 0.9582092761993408:  88%|████████▊ | 5688/6470 [1:05:28<09:24,  1.39it/s]
2|5689|Loss: 0.8443299531936646:  88%|████████▊ | 5688/6470 [1:05:29<09:24,  1.39it/s]
2|5689|Loss: 0.8443299531936646:  88%|████████▊ | 5689/6470 [1:05:29<08:42,  1.49it/s]
2|5690|Loss: 0.7295834422111511:  88%|████████▊ | 5689/6470 [1:05:29<08:42,  1.49it/s]
2|5690|Loss: 0.7295834422111511:  88%|████████▊ | 5690/6470 [1:05:30<09:00,  1.44it/s]
2|5691|Loss: 0.866323709487915:  88%|████████▊ | 5690/6470 [1:05:30<09:00,  1.44it/s] 
2|5691|Loss: 0.866323709487915:  88%|████████▊ | 5691/6470 [1:05:30<08:20,  1.56it/s]
2|5692|Loss: 0.8496965169906616:  88%|████████▊ | 5691/6470 [1:05:30<08:20,  1.56it/s]
2|5692|Loss: 0.8496965169906616:  88%|████████▊ | 5692/6470 [1:05:31<08:05,  1.60it/s]
2|5693|Loss: 0.8898760676383972:  88%|████████▊ | 5692/6470 [1:05:31<08:05,  1.60it/s]
2|5693|Loss: 0.8898760676383972:  88%|████████▊ | 5693/6470 [1:05:31<08:04,  1.60it/s]
2|5694|Loss: 0.9437342286109924:  88%|████████▊ | 5693/6470 [1:05:32<08:04,  1.60it/s]
2|5694|Loss: 0.9437342286109924:  88%|████████▊ | 5694/6470 [1:05:32<08:08,  1.59it/s]
2|5695|Loss: 0.9131374955177307:  88%|████████▊ | 5694/6470 [1:05:32<08:08,  1.59it/s]
2|5695|Loss: 0.9131374955177307:  88%|████████▊ | 5695/6470 [1:05:33<07:46,  1.66it/s]
2|5696|Loss: 0.8630899786949158:  88%|████████▊ | 5695/6470 [1:05:33<07:46,  1.66it/s]
2|5696|Loss: 0.8630899786949158:  88%|████████▊ | 5696/6470 [1:05:33<07:44,  1.67it/s]
2|5697|Loss: 0.8578906059265137:  88%|████████▊ | 5696/6470 [1:05:33<07:44,  1.67it/s]
2|5697|Loss: 0.8578906059265137:  88%|████████▊ | 5697/6470 [1:05:34<07:35,  1.70it/s]
2|5698|Loss: 1.059030532836914:  88%|████████▊ | 5697/6470 [1:05:34<07:35,  1.70it/s] 
2|5698|Loss: 1.059030532836914:  88%|████████▊ | 5698/6470 [1:05:34<07:56,  1.62it/s]
2|5699|Loss: 0.8991303443908691:  88%|████████▊ | 5698/6470 [1:05:35<07:56,  1.62it/s]
2|5699|Loss: 0.8991303443908691:  88%|████████▊ | 5699/6470 [1:05:35<07:55,  1.62it/s]
2|5700|Loss: 1.0704320669174194:  88%|████████▊ | 5699/6470 [1:05:35<07:55,  1.62it/s]
2|5700|Loss: 1.0704320669174194:  88%|████████▊ | 5700/6470 [1:05:36<07:58,  1.61it/s]
2|5701|Loss: 0.8552386164665222:  88%|████████▊ | 5700/6470 [1:05:36<07:58,  1.61it/s]
2|5701|Loss: 0.8552386164665222:  88%|████████▊ | 5701/6470 [1:05:36<08:45,  1.46it/s]
2|5702|Loss: 0.9985554218292236:  88%|████████▊ | 5701/6470 [1:05:37<08:45,  1.46it/s]
2|5702|Loss: 0.9985554218292236:  88%|████████▊ | 5702/6470 [1:05:37<08:47,  1.45it/s]
2|5703|Loss: 0.8817954063415527:  88%|████████▊ | 5702/6470 [1:05:37<08:47,  1.45it/s]
2|5703|Loss: 0.8817954063415527:  88%|████████▊ | 5703/6470 [1:05:38<08:25,  1.52it/s]
2|5704|Loss: 0.960868775844574:  88%|████████▊ | 5703/6470 [1:05:38<08:25,  1.52it/s] 
2|5704|Loss: 0.960868775844574:  88%|████████▊ | 5704/6470 [1:05:39<08:55,  1.43it/s]
2|5705|Loss: 0.8968455195426941:  88%|████████▊ | 5704/6470 [1:05:39<08:55,  1.43it/s]
2|5705|Loss: 0.8968455195426941:  88%|████████▊ | 5705/6470 [1:05:39<09:01,  1.41it/s]
2|5706|Loss: 0.8224512338638306:  88%|████████▊ | 5705/6470 [1:05:40<09:01,  1.41it/s]
2|5706|Loss: 0.8224512338638306:  88%|████████▊ | 5706/6470 [1:05:40<09:09,  1.39it/s]
2|5707|Loss: 0.8831987977027893:  88%|████████▊ | 5706/6470 [1:05:40<09:09,  1.39it/s]
2|5707|Loss: 0.8831987977027893:  88%|████████▊ | 5707/6470 [1:05:41<08:56,  1.42it/s]
2|5708|Loss: 0.7598931789398193:  88%|████████▊ | 5707/6470 [1:05:41<08:56,  1.42it/s]
2|5708|Loss: 0.7598931789398193:  88%|████████▊ | 5708/6470 [1:05:41<08:30,  1.49it/s]
2|5709|Loss: 1.0003721714019775:  88%|████████▊ | 5708/6470 [1:05:42<08:30,  1.49it/s]
2|5709|Loss: 1.0003721714019775:  88%|████████▊ | 5709/6470 [1:05:42<07:57,  1.59it/s]
2|5710|Loss: 0.8390973210334778:  88%|████████▊ | 5709/6470 [1:05:42<07:57,  1.59it/s]
2|5710|Loss: 0.8390973210334778:  88%|████████▊ | 5710/6470 [1:05:43<08:29,  1.49it/s]
2|5711|Loss: 0.8407480716705322:  88%|████████▊ | 5710/6470 [1:05:43<08:29,  1.49it/s]
2|5711|Loss: 0.8407480716705322:  88%|████████▊ | 5711/6470 [1:05:43<08:28,  1.49it/s]
2|5712|Loss: 0.8466613292694092:  88%|████████▊ | 5711/6470 [1:05:44<08:28,  1.49it/s]
2|5712|Loss: 0.8466613292694092:  88%|████████▊ | 5712/6470 [1:05:44<08:31,  1.48it/s]
2|5713|Loss: 0.7631078362464905:  88%|████████▊ | 5712/6470 [1:05:44<08:31,  1.48it/s]
2|5713|Loss: 0.7631078362464905:  88%|████████▊ | 5713/6470 [1:05:45<09:10,  1.38it/s]
2|5714|Loss: 0.8724969625473022:  88%|████████▊ | 5713/6470 [1:05:45<09:10,  1.38it/s]
2|5714|Loss: 0.8724969625473022:  88%|████████▊ | 5714/6470 [1:05:45<08:47,  1.43it/s]
2|5715|Loss: 0.8896707892417908:  88%|████████▊ | 5714/6470 [1:05:46<08:47,  1.43it/s]
2|5715|Loss: 0.8896707892417908:  88%|████████▊ | 5715/6470 [1:05:46<08:29,  1.48it/s]
2|5716|Loss: 0.9235121011734009:  88%|████████▊ | 5715/6470 [1:05:46<08:29,  1.48it/s]
2|5716|Loss: 0.9235121011734009:  88%|████████▊ | 5716/6470 [1:05:47<08:52,  1.42it/s]
2|5717|Loss: 0.842893660068512:  88%|████████▊ | 5716/6470 [1:05:47<08:52,  1.42it/s] 
2|5717|Loss: 0.842893660068512:  88%|████████▊ | 5717/6470 [1:05:47<08:36,  1.46it/s]
2|5718|Loss: 0.8591698408126831:  88%|████████▊ | 5717/6470 [1:05:48<08:36,  1.46it/s]
2|5718|Loss: 0.8591698408126831:  88%|████████▊ | 5718/6470 [1:05:48<08:42,  1.44it/s]
2|5719|Loss: 0.89799565076828:  88%|████████▊ | 5718/6470 [1:05:48<08:42,  1.44it/s]  
2|5719|Loss: 0.89799565076828:  88%|████████▊ | 5719/6470 [1:05:49<08:14,  1.52it/s]
2|5720|Loss: 0.8389678001403809:  88%|████████▊ | 5719/6470 [1:05:49<08:14,  1.52it/s]
2|5720|Loss: 0.8389678001403809:  88%|████████▊ | 5720/6470 [1:05:49<08:20,  1.50it/s]
2|5721|Loss: 0.8949819803237915:  88%|████████▊ | 5720/6470 [1:05:50<08:20,  1.50it/s]
2|5721|Loss: 0.8949819803237915:  88%|████████▊ | 5721/6470 [1:05:50<08:20,  1.50it/s]
2|5722|Loss: 0.8453283309936523:  88%|████████▊ | 5721/6470 [1:05:50<08:20,  1.50it/s]
2|5722|Loss: 0.8453283309936523:  88%|████████▊ | 5722/6470 [1:05:51<07:50,  1.59it/s]
2|5723|Loss: 0.683897078037262:  88%|████████▊ | 5722/6470 [1:05:51<07:50,  1.59it/s] 
2|5723|Loss: 0.683897078037262:  88%|████████▊ | 5723/6470 [1:05:51<08:08,  1.53it/s]
2|5724|Loss: 0.7321643233299255:  88%|████████▊ | 5723/6470 [1:05:52<08:08,  1.53it/s]
2|5724|Loss: 0.7321643233299255:  88%|████████▊ | 5724/6470 [1:05:52<07:58,  1.56it/s]
2|5725|Loss: 0.8407599925994873:  88%|████████▊ | 5724/6470 [1:05:52<07:58,  1.56it/s]
2|5725|Loss: 0.8407599925994873:  88%|████████▊ | 5725/6470 [1:05:53<08:05,  1.54it/s]
2|5726|Loss: 0.9053324460983276:  88%|████████▊ | 5725/6470 [1:05:53<08:05,  1.54it/s]
2|5726|Loss: 0.9053324460983276:  89%|████████▊ | 5726/6470 [1:05:53<08:05,  1.53it/s]
2|5727|Loss: 0.8656991720199585:  89%|████████▊ | 5726/6470 [1:05:54<08:05,  1.53it/s]
2|5727|Loss: 0.8656991720199585:  89%|████████▊ | 5727/6470 [1:05:54<08:19,  1.49it/s]
2|5728|Loss: 0.9022335410118103:  89%|████████▊ | 5727/6470 [1:05:54<08:19,  1.49it/s]
2|5728|Loss: 0.9022335410118103:  89%|████████▊ | 5728/6470 [1:05:55<08:49,  1.40it/s]
2|5729|Loss: 0.9221231937408447:  89%|████████▊ | 5728/6470 [1:05:55<08:49,  1.40it/s]
2|5729|Loss: 0.9221231937408447:  89%|████████▊ | 5729/6470 [1:05:56<09:15,  1.33it/s]
2|5730|Loss: 0.9410338401794434:  89%|████████▊ | 5729/6470 [1:05:56<09:15,  1.33it/s]
2|5730|Loss: 0.9410338401794434:  89%|████████▊ | 5730/6470 [1:05:56<09:16,  1.33it/s]
2|5731|Loss: 0.7453948855400085:  89%|████████▊ | 5730/6470 [1:05:57<09:16,  1.33it/s]
2|5731|Loss: 0.7453948855400085:  89%|████████▊ | 5731/6470 [1:05:57<08:57,  1.38it/s]
2|5732|Loss: 0.8270420432090759:  89%|████████▊ | 5731/6470 [1:05:57<08:57,  1.38it/s]
2|5732|Loss: 0.8270420432090759:  89%|████████▊ | 5732/6470 [1:05:58<08:48,  1.40it/s]
2|5733|Loss: 0.8007842302322388:  89%|████████▊ | 5732/6470 [1:05:58<08:48,  1.40it/s]
2|5733|Loss: 0.8007842302322388:  89%|████████▊ | 5733/6470 [1:05:58<08:13,  1.49it/s]
2|5734|Loss: 0.9971562623977661:  89%|████████▊ | 5733/6470 [1:05:59<08:13,  1.49it/s]
2|5734|Loss: 0.9971562623977661:  89%|████████▊ | 5734/6470 [1:05:59<07:58,  1.54it/s]
2|5735|Loss: 0.8504785895347595:  89%|████████▊ | 5734/6470 [1:05:59<07:58,  1.54it/s]
2|5735|Loss: 0.8504785895347595:  89%|████████▊ | 5735/6470 [1:05:59<07:32,  1.62it/s]
2|5736|Loss: 0.9186724424362183:  89%|████████▊ | 5735/6470 [1:06:00<07:32,  1.62it/s]
2|5736|Loss: 0.9186724424362183:  89%|████████▊ | 5736/6470 [1:06:00<07:50,  1.56it/s]
2|5737|Loss: 0.8916308283805847:  89%|████████▊ | 5736/6470 [1:06:01<07:50,  1.56it/s]
2|5737|Loss: 0.8916308283805847:  89%|████████▊ | 5737/6470 [1:06:01<07:51,  1.55it/s]
2|5738|Loss: 0.8932845592498779:  89%|████████▊ | 5737/6470 [1:06:01<07:51,  1.55it/s]
2|5738|Loss: 0.8932845592498779:  89%|████████▊ | 5738/6470 [1:06:02<08:05,  1.51it/s]
2|5739|Loss: 0.9152712225914001:  89%|████████▊ | 5738/6470 [1:06:02<08:05,  1.51it/s]
2|5739|Loss: 0.9152712225914001:  89%|████████▊ | 5739/6470 [1:06:02<08:04,  1.51it/s]
2|5740|Loss: 0.7894498705863953:  89%|████████▊ | 5739/6470 [1:06:03<08:04,  1.51it/s]
2|5740|Loss: 0.7894498705863953:  89%|████████▊ | 5740/6470 [1:06:03<08:15,  1.47it/s]
2|5741|Loss: 0.9720515012741089:  89%|████████▊ | 5740/6470 [1:06:03<08:15,  1.47it/s]
2|5741|Loss: 0.9720515012741089:  89%|████████▊ | 5741/6470 [1:06:04<08:21,  1.45it/s]
2|5742|Loss: 0.8495301008224487:  89%|████████▊ | 5741/6470 [1:06:04<08:21,  1.45it/s]
2|5742|Loss: 0.8495301008224487:  89%|████████▊ | 5742/6470 [1:06:04<08:02,  1.51it/s]
2|5743|Loss: 0.9770098328590393:  89%|████████▊ | 5742/6470 [1:06:05<08:02,  1.51it/s]
2|5743|Loss: 0.9770098328590393:  89%|████████▉ | 5743/6470 [1:06:05<08:24,  1.44it/s]
2|5744|Loss: 0.7633498311042786:  89%|████████▉ | 5743/6470 [1:06:05<08:24,  1.44it/s]
2|5744|Loss: 0.7633498311042786:  89%|████████▉ | 5744/6470 [1:06:06<08:08,  1.49it/s]
2|5745|Loss: 0.8298892974853516:  89%|████████▉ | 5744/6470 [1:06:06<08:08,  1.49it/s]
2|5745|Loss: 0.8298892974853516:  89%|████████▉ | 5745/6470 [1:06:06<07:53,  1.53it/s]
2|5746|Loss: 0.9116846323013306:  89%|████████▉ | 5745/6470 [1:06:07<07:53,  1.53it/s]
2|5746|Loss: 0.9116846323013306:  89%|████████▉ | 5746/6470 [1:06:07<08:00,  1.51it/s]
2|5747|Loss: 0.8700340986251831:  89%|████████▉ | 5746/6470 [1:06:07<08:00,  1.51it/s]
2|5747|Loss: 0.8700340986251831:  89%|████████▉ | 5747/6470 [1:06:07<07:25,  1.62it/s]
2|5748|Loss: 0.7827591896057129:  89%|████████▉ | 5747/6470 [1:06:08<07:25,  1.62it/s]
2|5748|Loss: 0.7827591896057129:  89%|████████▉ | 5748/6470 [1:06:08<07:31,  1.60it/s]
2|5749|Loss: 0.9269144535064697:  89%|████████▉ | 5748/6470 [1:06:08<07:31,  1.60it/s]
2|5749|Loss: 0.9269144535064697:  89%|████████▉ | 5749/6470 [1:06:09<07:57,  1.51it/s]
2|5750|Loss: 0.798923671245575:  89%|████████▉ | 5749/6470 [1:06:09<07:57,  1.51it/s] 
2|5750|Loss: 0.798923671245575:  89%|████████▉ | 5750/6470 [1:06:10<08:16,  1.45it/s]
2|5751|Loss: 1.0722168684005737:  89%|████████▉ | 5750/6470 [1:06:10<08:16,  1.45it/s]
2|5751|Loss: 1.0722168684005737:  89%|████████▉ | 5751/6470 [1:06:10<07:50,  1.53it/s]
2|5752|Loss: 0.8233086466789246:  89%|████████▉ | 5751/6470 [1:06:10<07:50,  1.53it/s]
2|5752|Loss: 0.8233086466789246:  89%|████████▉ | 5752/6470 [1:06:11<08:00,  1.50it/s]
2|5753|Loss: 0.8204329609870911:  89%|████████▉ | 5752/6470 [1:06:11<08:00,  1.50it/s]
2|5753|Loss: 0.8204329609870911:  89%|████████▉ | 5753/6470 [1:06:11<07:55,  1.51it/s]
2|5754|Loss: 0.9265409111976624:  89%|████████▉ | 5753/6470 [1:06:12<07:55,  1.51it/s]
2|5754|Loss: 0.9265409111976624:  89%|████████▉ | 5754/6470 [1:06:12<08:13,  1.45it/s]
2|5755|Loss: 1.0023224353790283:  89%|████████▉ | 5754/6470 [1:06:13<08:13,  1.45it/s]
2|5755|Loss: 1.0023224353790283:  89%|████████▉ | 5755/6470 [1:06:13<08:46,  1.36it/s]
2|5756|Loss: 0.7424556016921997:  89%|████████▉ | 5755/6470 [1:06:13<08:46,  1.36it/s]
2|5756|Loss: 0.7424556016921997:  89%|████████▉ | 5756/6470 [1:06:14<08:12,  1.45it/s]
2|5757|Loss: 0.7492395043373108:  89%|████████▉ | 5756/6470 [1:06:14<08:12,  1.45it/s]
2|5757|Loss: 0.7492395043373108:  89%|████████▉ | 5757/6470 [1:06:14<08:28,  1.40it/s]
2|5758|Loss: 0.8334079384803772:  89%|████████▉ | 5757/6470 [1:06:15<08:28,  1.40it/s]
2|5758|Loss: 0.8334079384803772:  89%|████████▉ | 5758/6470 [1:06:15<08:22,  1.42it/s]
2|5759|Loss: 0.8019749522209167:  89%|████████▉ | 5758/6470 [1:06:15<08:22,  1.42it/s]
2|5759|Loss: 0.8019749522209167:  89%|████████▉ | 5759/6470 [1:06:16<07:45,  1.53it/s]
2|5760|Loss: 0.7749776840209961:  89%|████████▉ | 5759/6470 [1:06:16<07:45,  1.53it/s]
2|5760|Loss: 0.7749776840209961:  89%|████████▉ | 5760/6470 [1:06:16<07:31,  1.57it/s]
2|5761|Loss: 0.857772946357727:  89%|████████▉ | 5760/6470 [1:06:17<07:31,  1.57it/s] 
2|5761|Loss: 0.857772946357727:  89%|████████▉ | 5761/6470 [1:06:17<08:09,  1.45it/s]
2|5762|Loss: 0.7699571847915649:  89%|████████▉ | 5761/6470 [1:06:17<08:09,  1.45it/s]
2|5762|Loss: 0.7699571847915649:  89%|████████▉ | 5762/6470 [1:06:18<08:26,  1.40it/s]
2|5763|Loss: 0.8325173854827881:  89%|████████▉ | 5762/6470 [1:06:18<08:26,  1.40it/s]
2|5763|Loss: 0.8325173854827881:  89%|████████▉ | 5763/6470 [1:06:18<08:11,  1.44it/s]
2|5764|Loss: 0.9623570442199707:  89%|████████▉ | 5763/6470 [1:06:19<08:11,  1.44it/s]
2|5764|Loss: 0.9623570442199707:  89%|████████▉ | 5764/6470 [1:06:19<08:15,  1.42it/s]
2|5765|Loss: 0.918167769908905:  89%|████████▉ | 5764/6470 [1:06:20<08:15,  1.42it/s] 
2|5765|Loss: 0.918167769908905:  89%|████████▉ | 5765/6470 [1:06:20<08:03,  1.46it/s]
2|5766|Loss: 0.7325453758239746:  89%|████████▉ | 5765/6470 [1:06:20<08:03,  1.46it/s]
2|5766|Loss: 0.7325453758239746:  89%|████████▉ | 5766/6470 [1:06:21<08:10,  1.44it/s]
2|5767|Loss: 0.723757803440094:  89%|████████▉ | 5766/6470 [1:06:21<08:10,  1.44it/s] 
2|5767|Loss: 0.723757803440094:  89%|████████▉ | 5767/6470 [1:06:21<08:13,  1.42it/s]
2|5768|Loss: 1.088874101638794:  89%|████████▉ | 5767/6470 [1:06:22<08:13,  1.42it/s]
2|5768|Loss: 1.088874101638794:  89%|████████▉ | 5768/6470 [1:06:22<08:49,  1.33it/s]
2|5769|Loss: 0.7988365888595581:  89%|████████▉ | 5768/6470 [1:06:23<08:49,  1.33it/s]
2|5769|Loss: 0.7988365888595581:  89%|████████▉ | 5769/6470 [1:06:23<08:52,  1.32it/s]
2|5770|Loss: 0.9780561327934265:  89%|████████▉ | 5769/6470 [1:06:23<08:52,  1.32it/s]
2|5770|Loss: 0.9780561327934265:  89%|████████▉ | 5770/6470 [1:06:24<08:19,  1.40it/s]
2|5771|Loss: 0.9568345546722412:  89%|████████▉ | 5770/6470 [1:06:24<08:19,  1.40it/s]
2|5771|Loss: 0.9568345546722412:  89%|████████▉ | 5771/6470 [1:06:24<08:27,  1.38it/s]
2|5772|Loss: 0.7737334370613098:  89%|████████▉ | 5771/6470 [1:06:25<08:27,  1.38it/s]
2|5772|Loss: 0.7737334370613098:  89%|████████▉ | 5772/6470 [1:06:25<07:44,  1.50it/s]
2|5773|Loss: 0.9426785111427307:  89%|████████▉ | 5772/6470 [1:06:25<07:44,  1.50it/s]
2|5773|Loss: 0.9426785111427307:  89%|████████▉ | 5773/6470 [1:06:25<07:30,  1.55it/s]
2|5774|Loss: 1.0966131687164307:  89%|████████▉ | 5773/6470 [1:06:26<07:30,  1.55it/s]
2|5774|Loss: 1.0966131687164307:  89%|████████▉ | 5774/6470 [1:06:26<07:46,  1.49it/s]
2|5775|Loss: 0.8030180335044861:  89%|████████▉ | 5774/6470 [1:06:27<07:46,  1.49it/s]
2|5775|Loss: 0.8030180335044861:  89%|████████▉ | 5775/6470 [1:06:27<07:44,  1.50it/s]
2|5776|Loss: 0.9551227688789368:  89%|████████▉ | 5775/6470 [1:06:27<07:44,  1.50it/s]
2|5776|Loss: 0.9551227688789368:  89%|████████▉ | 5776/6470 [1:06:27<07:27,  1.55it/s]
2|5777|Loss: 0.8893347382545471:  89%|████████▉ | 5776/6470 [1:06:28<07:27,  1.55it/s]
2|5777|Loss: 0.8893347382545471:  89%|████████▉ | 5777/6470 [1:06:28<07:39,  1.51it/s]
2|5778|Loss: 0.8514834642410278:  89%|████████▉ | 5777/6470 [1:06:28<07:39,  1.51it/s]
2|5778|Loss: 0.8514834642410278:  89%|████████▉ | 5778/6470 [1:06:29<07:16,  1.58it/s]
2|5779|Loss: 0.7644426822662354:  89%|████████▉ | 5778/6470 [1:06:29<07:16,  1.58it/s]
2|5779|Loss: 0.7644426822662354:  89%|████████▉ | 5779/6470 [1:06:29<07:08,  1.61it/s]
2|5780|Loss: 0.7136894464492798:  89%|████████▉ | 5779/6470 [1:06:30<07:08,  1.61it/s]
2|5780|Loss: 0.7136894464492798:  89%|████████▉ | 5780/6470 [1:06:30<07:21,  1.56it/s]
2|5781|Loss: 0.8634580373764038:  89%|████████▉ | 5780/6470 [1:06:30<07:21,  1.56it/s]
2|5781|Loss: 0.8634580373764038:  89%|████████▉ | 5781/6470 [1:06:31<08:01,  1.43it/s]
2|5782|Loss: 0.8494085073471069:  89%|████████▉ | 5781/6470 [1:06:31<08:01,  1.43it/s]
2|5782|Loss: 0.8494085073471069:  89%|████████▉ | 5782/6470 [1:06:31<07:38,  1.50it/s]
2|5783|Loss: 0.8931109309196472:  89%|████████▉ | 5782/6470 [1:06:32<07:38,  1.50it/s]
2|5783|Loss: 0.8931109309196472:  89%|████████▉ | 5783/6470 [1:06:32<07:41,  1.49it/s]
2|5784|Loss: 0.8215239644050598:  89%|████████▉ | 5783/6470 [1:06:32<07:41,  1.49it/s]
2|5784|Loss: 0.8215239644050598:  89%|████████▉ | 5784/6470 [1:06:33<08:09,  1.40it/s]
2|5785|Loss: 0.6866248250007629:  89%|████████▉ | 5784/6470 [1:06:33<08:09,  1.40it/s]
2|5785|Loss: 0.6866248250007629:  89%|████████▉ | 5785/6470 [1:06:33<07:19,  1.56it/s]
2|5786|Loss: 1.0355262756347656:  89%|████████▉ | 5785/6470 [1:06:34<07:19,  1.56it/s]
2|5786|Loss: 1.0355262756347656:  89%|████████▉ | 5786/6470 [1:06:34<07:09,  1.59it/s]
2|5787|Loss: 0.9173040986061096:  89%|████████▉ | 5786/6470 [1:06:34<07:09,  1.59it/s]
2|5787|Loss: 0.9173040986061096:  89%|████████▉ | 5787/6470 [1:06:35<07:10,  1.59it/s]
2|5788|Loss: 0.7842196226119995:  89%|████████▉ | 5787/6470 [1:06:35<07:10,  1.59it/s]
2|5788|Loss: 0.7842196226119995:  89%|████████▉ | 5788/6470 [1:06:35<07:28,  1.52it/s]
2|5789|Loss: 0.9225983023643494:  89%|████████▉ | 5788/6470 [1:06:36<07:28,  1.52it/s]
2|5789|Loss: 0.9225983023643494:  89%|████████▉ | 5789/6470 [1:06:36<07:39,  1.48it/s]
2|5790|Loss: 0.9146869778633118:  89%|████████▉ | 5789/6470 [1:06:36<07:39,  1.48it/s]
2|5790|Loss: 0.9146869778633118:  89%|████████▉ | 5790/6470 [1:06:37<07:45,  1.46it/s]
2|5791|Loss: 0.781597375869751:  89%|████████▉ | 5790/6470 [1:06:37<07:45,  1.46it/s] 
2|5791|Loss: 0.781597375869751:  90%|████████▉ | 5791/6470 [1:06:37<07:28,  1.51it/s]
2|5792|Loss: 0.883021891117096:  90%|████████▉ | 5791/6470 [1:06:38<07:28,  1.51it/s]
2|5792|Loss: 0.883021891117096:  90%|████████▉ | 5792/6470 [1:06:38<07:37,  1.48it/s]
2|5793|Loss: 0.7290952205657959:  90%|████████▉ | 5792/6470 [1:06:38<07:37,  1.48it/s]
2|5793|Loss: 0.7290952205657959:  90%|████████▉ | 5793/6470 [1:06:39<07:44,  1.46it/s]
2|5794|Loss: 0.9002773761749268:  90%|████████▉ | 5793/6470 [1:06:39<07:44,  1.46it/s]
2|5794|Loss: 0.9002773761749268:  90%|████████▉ | 5794/6470 [1:06:39<07:21,  1.53it/s]
2|5795|Loss: 0.8272377252578735:  90%|████████▉ | 5794/6470 [1:06:40<07:21,  1.53it/s]
2|5795|Loss: 0.8272377252578735:  90%|████████▉ | 5795/6470 [1:06:40<06:55,  1.63it/s]
2|5796|Loss: 0.8909348249435425:  90%|████████▉ | 5795/6470 [1:06:40<06:55,  1.63it/s]
2|5796|Loss: 0.8909348249435425:  90%|████████▉ | 5796/6470 [1:06:40<06:50,  1.64it/s]
2|5797|Loss: 0.8753686547279358:  90%|████████▉ | 5796/6470 [1:06:41<06:50,  1.64it/s]
2|5797|Loss: 0.8753686547279358:  90%|████████▉ | 5797/6470 [1:06:41<06:34,  1.71it/s]
2|5798|Loss: 0.9483585953712463:  90%|████████▉ | 5797/6470 [1:06:41<06:34,  1.71it/s]
2|5798|Loss: 0.9483585953712463:  90%|████████▉ | 5798/6470 [1:06:42<06:38,  1.68it/s]
2|5799|Loss: 1.0238618850708008:  90%|████████▉ | 5798/6470 [1:06:42<06:38,  1.68it/s]
2|5799|Loss: 1.0238618850708008:  90%|████████▉ | 5799/6470 [1:06:42<07:11,  1.55it/s]
2|5800|Loss: 0.7831145524978638:  90%|████████▉ | 5799/6470 [1:06:43<07:11,  1.55it/s]
2|5800|Loss: 0.7831145524978638:  90%|████████▉ | 5800/6470 [1:06:43<07:52,  1.42it/s]
2|5801|Loss: 0.8242453932762146:  90%|████████▉ | 5800/6470 [1:06:44<07:52,  1.42it/s]
2|5801|Loss: 0.8242453932762146:  90%|████████▉ | 5801/6470 [1:06:44<07:48,  1.43it/s]
2|5802|Loss: 0.8999337553977966:  90%|████████▉ | 5801/6470 [1:06:44<07:48,  1.43it/s]
2|5802|Loss: 0.8999337553977966:  90%|████████▉ | 5802/6470 [1:06:45<07:51,  1.42it/s]
2|5803|Loss: 0.9413397312164307:  90%|████████▉ | 5802/6470 [1:06:45<07:51,  1.42it/s]
2|5803|Loss: 0.9413397312164307:  90%|████████▉ | 5803/6470 [1:06:45<08:09,  1.36it/s]
2|5804|Loss: 0.9635618925094604:  90%|████████▉ | 5803/6470 [1:06:46<08:09,  1.36it/s]
2|5804|Loss: 0.9635618925094604:  90%|████████▉ | 5804/6470 [1:06:46<08:14,  1.35it/s]
2|5805|Loss: 0.8663293719291687:  90%|████████▉ | 5804/6470 [1:06:47<08:14,  1.35it/s]
2|5805|Loss: 0.8663293719291687:  90%|████████▉ | 5805/6470 [1:06:47<07:55,  1.40it/s]
2|5806|Loss: 0.7530933618545532:  90%|████████▉ | 5805/6470 [1:06:47<07:55,  1.40it/s]
2|5806|Loss: 0.7530933618545532:  90%|████████▉ | 5806/6470 [1:06:47<07:53,  1.40it/s]
2|5807|Loss: 0.875117301940918:  90%|████████▉ | 5806/6470 [1:06:48<07:53,  1.40it/s] 
2|5807|Loss: 0.875117301940918:  90%|████████▉ | 5807/6470 [1:06:48<08:25,  1.31it/s]
2|5808|Loss: 0.9073039293289185:  90%|████████▉ | 5807/6470 [1:06:49<08:25,  1.31it/s]
2|5808|Loss: 0.9073039293289185:  90%|████████▉ | 5808/6470 [1:06:49<08:09,  1.35it/s]
2|5809|Loss: 0.8857742547988892:  90%|████████▉ | 5808/6470 [1:06:49<08:09,  1.35it/s]
2|5809|Loss: 0.8857742547988892:  90%|████████▉ | 5809/6470 [1:06:50<08:24,  1.31it/s]
2|5810|Loss: 0.8950215578079224:  90%|████████▉ | 5809/6470 [1:06:50<08:24,  1.31it/s]
2|5810|Loss: 0.8950215578079224:  90%|████████▉ | 5810/6470 [1:06:51<08:10,  1.35it/s]
2|5811|Loss: 0.7728436589241028:  90%|████████▉ | 5810/6470 [1:06:51<08:10,  1.35it/s]
2|5811|Loss: 0.7728436589241028:  90%|████████▉ | 5811/6470 [1:06:51<07:19,  1.50it/s]
2|5812|Loss: 0.8341202735900879:  90%|████████▉ | 5811/6470 [1:06:51<07:19,  1.50it/s]
2|5812|Loss: 0.8341202735900879:  90%|████████▉ | 5812/6470 [1:06:52<07:20,  1.49it/s]
2|5813|Loss: 0.9342296719551086:  90%|████████▉ | 5812/6470 [1:06:52<07:20,  1.49it/s]
2|5813|Loss: 0.9342296719551086:  90%|████████▉ | 5813/6470 [1:06:52<07:33,  1.45it/s]
2|5814|Loss: 0.9196629524230957:  90%|████████▉ | 5813/6470 [1:06:53<07:33,  1.45it/s]
2|5814|Loss: 0.9196629524230957:  90%|████████▉ | 5814/6470 [1:06:53<07:43,  1.41it/s]
2|5815|Loss: 0.7518600821495056:  90%|████████▉ | 5814/6470 [1:06:54<07:43,  1.41it/s]
2|5815|Loss: 0.7518600821495056:  90%|████████▉ | 5815/6470 [1:06:54<07:44,  1.41it/s]
2|5816|Loss: 0.7481538653373718:  90%|████████▉ | 5815/6470 [1:06:54<07:44,  1.41it/s]
2|5816|Loss: 0.7481538653373718:  90%|████████▉ | 5816/6470 [1:06:54<07:12,  1.51it/s]
2|5817|Loss: 0.8155869245529175:  90%|████████▉ | 5816/6470 [1:06:55<07:12,  1.51it/s]
2|5817|Loss: 0.8155869245529175:  90%|████████▉ | 5817/6470 [1:06:55<07:11,  1.51it/s]
2|5818|Loss: 0.9533868432044983:  90%|████████▉ | 5817/6470 [1:06:56<07:11,  1.51it/s]
2|5818|Loss: 0.9533868432044983:  90%|████████▉ | 5818/6470 [1:06:56<07:38,  1.42it/s]
2|5819|Loss: 0.9454377889633179:  90%|████████▉ | 5818/6470 [1:06:56<07:38,  1.42it/s]
2|5819|Loss: 0.9454377889633179:  90%|████████▉ | 5819/6470 [1:06:57<07:34,  1.43it/s]
2|5820|Loss: 0.8654773235321045:  90%|████████▉ | 5819/6470 [1:06:57<07:34,  1.43it/s]
2|5820|Loss: 0.8654773235321045:  90%|████████▉ | 5820/6470 [1:06:57<07:36,  1.42it/s]
2|5821|Loss: 1.0014697313308716:  90%|████████▉ | 5820/6470 [1:06:58<07:36,  1.42it/s]
2|5821|Loss: 1.0014697313308716:  90%|████████▉ | 5821/6470 [1:06:58<07:41,  1.41it/s]
2|5822|Loss: 0.7361050248146057:  90%|████████▉ | 5821/6470 [1:06:58<07:41,  1.41it/s]
2|5822|Loss: 0.7361050248146057:  90%|████████▉ | 5822/6470 [1:06:59<07:39,  1.41it/s]
2|5823|Loss: 0.756895899772644:  90%|████████▉ | 5822/6470 [1:06:59<07:39,  1.41it/s] 
2|5823|Loss: 0.756895899772644:  90%|█████████ | 5823/6470 [1:06:59<07:25,  1.45it/s]
2|5824|Loss: 0.8260294198989868:  90%|█████████ | 5823/6470 [1:07:00<07:25,  1.45it/s]
2|5824|Loss: 0.8260294198989868:  90%|█████████ | 5824/6470 [1:07:00<07:23,  1.46it/s]
2|5825|Loss: 0.8680406808853149:  90%|█████████ | 5824/6470 [1:07:00<07:23,  1.46it/s]
2|5825|Loss: 0.8680406808853149:  90%|█████████ | 5825/6470 [1:07:01<07:12,  1.49it/s]
2|5826|Loss: 0.8987379670143127:  90%|█████████ | 5825/6470 [1:07:01<07:12,  1.49it/s]
2|5826|Loss: 0.8987379670143127:  90%|█████████ | 5826/6470 [1:07:01<07:14,  1.48it/s]
2|5827|Loss: 0.9193029999732971:  90%|█████████ | 5826/6470 [1:07:02<07:14,  1.48it/s]
2|5827|Loss: 0.9193029999732971:  90%|█████████ | 5827/6470 [1:07:02<07:42,  1.39it/s]
2|5828|Loss: 0.8697850108146667:  90%|█████████ | 5827/6470 [1:07:03<07:42,  1.39it/s]
2|5828|Loss: 0.8697850108146667:  90%|█████████ | 5828/6470 [1:07:03<07:46,  1.37it/s]
2|5829|Loss: 0.7995072603225708:  90%|█████████ | 5828/6470 [1:07:03<07:46,  1.37it/s]
2|5829|Loss: 0.7995072603225708:  90%|█████████ | 5829/6470 [1:07:04<07:27,  1.43it/s]
2|5830|Loss: 0.8670907616615295:  90%|█████████ | 5829/6470 [1:07:04<07:27,  1.43it/s]
2|5830|Loss: 0.8670907616615295:  90%|█████████ | 5830/6470 [1:07:04<07:24,  1.44it/s]
2|5831|Loss: 0.8626874089241028:  90%|█████████ | 5830/6470 [1:07:05<07:24,  1.44it/s]
2|5831|Loss: 0.8626874089241028:  90%|█████████ | 5831/6470 [1:07:05<07:12,  1.48it/s]
2|5832|Loss: 0.9185519814491272:  90%|█████████ | 5831/6470 [1:07:05<07:12,  1.48it/s]
2|5832|Loss: 0.9185519814491272:  90%|█████████ | 5832/6470 [1:07:06<07:02,  1.51it/s]
2|5833|Loss: 0.8542177677154541:  90%|█████████ | 5832/6470 [1:07:06<07:02,  1.51it/s]
2|5833|Loss: 0.8542177677154541:  90%|█████████ | 5833/6470 [1:07:06<07:26,  1.43it/s]
2|5834|Loss: 1.144169807434082:  90%|█████████ | 5833/6470 [1:07:07<07:26,  1.43it/s] 
2|5834|Loss: 1.144169807434082:  90%|█████████ | 5834/6470 [1:07:07<07:57,  1.33it/s]
2|5835|Loss: 0.8962317109107971:  90%|█████████ | 5834/6470 [1:07:08<07:57,  1.33it/s]
2|5835|Loss: 0.8962317109107971:  90%|█████████ | 5835/6470 [1:07:08<08:00,  1.32it/s]
2|5836|Loss: 0.8825109004974365:  90%|█████████ | 5835/6470 [1:07:08<08:00,  1.32it/s]
2|5836|Loss: 0.8825109004974365:  90%|█████████ | 5836/6470 [1:07:09<07:38,  1.38it/s]
2|5837|Loss: 0.8242875933647156:  90%|█████████ | 5836/6470 [1:07:09<07:38,  1.38it/s]
2|5837|Loss: 0.8242875933647156:  90%|█████████ | 5837/6470 [1:07:09<07:37,  1.38it/s]
2|5838|Loss: 0.7858689427375793:  90%|█████████ | 5837/6470 [1:07:10<07:37,  1.38it/s]
2|5838|Loss: 0.7858689427375793:  90%|█████████ | 5838/6470 [1:07:10<07:17,  1.45it/s]
2|5839|Loss: 0.8541263937950134:  90%|█████████ | 5838/6470 [1:07:10<07:17,  1.45it/s]
2|5839|Loss: 0.8541263937950134:  90%|█████████ | 5839/6470 [1:07:11<06:55,  1.52it/s]
2|5840|Loss: 0.8779522776603699:  90%|█████████ | 5839/6470 [1:07:11<06:55,  1.52it/s]
2|5840|Loss: 0.8779522776603699:  90%|█████████ | 5840/6470 [1:07:11<07:00,  1.50it/s]
2|5841|Loss: 0.9235466718673706:  90%|█████████ | 5840/6470 [1:07:12<07:00,  1.50it/s]
2|5841|Loss: 0.9235466718673706:  90%|█████████ | 5841/6470 [1:07:12<07:26,  1.41it/s]
2|5842|Loss: 0.7977960705757141:  90%|█████████ | 5841/6470 [1:07:12<07:26,  1.41it/s]
2|5842|Loss: 0.7977960705757141:  90%|█████████ | 5842/6470 [1:07:13<07:28,  1.40it/s]
2|5843|Loss: 0.8032163977622986:  90%|█████████ | 5842/6470 [1:07:13<07:28,  1.40it/s]
2|5843|Loss: 0.8032163977622986:  90%|█████████ | 5843/6470 [1:07:13<07:24,  1.41it/s]
2|5844|Loss: 0.763645589351654:  90%|█████████ | 5843/6470 [1:07:14<07:24,  1.41it/s] 
2|5844|Loss: 0.763645589351654:  90%|█████████ | 5844/6470 [1:07:14<07:12,  1.45it/s]
2|5845|Loss: 0.9829090237617493:  90%|█████████ | 5844/6470 [1:07:15<07:12,  1.45it/s]
2|5845|Loss: 0.9829090237617493:  90%|█████████ | 5845/6470 [1:07:15<07:25,  1.40it/s]
2|5846|Loss: 0.886364758014679:  90%|█████████ | 5845/6470 [1:07:15<07:25,  1.40it/s] 
2|5846|Loss: 0.886364758014679:  90%|█████████ | 5846/6470 [1:07:16<07:49,  1.33it/s]
2|5847|Loss: 0.7766119241714478:  90%|█████████ | 5846/6470 [1:07:16<07:49,  1.33it/s]
2|5847|Loss: 0.7766119241714478:  90%|█████████ | 5847/6470 [1:07:16<07:47,  1.33it/s]
2|5848|Loss: 0.8591736555099487:  90%|█████████ | 5847/6470 [1:07:17<07:47,  1.33it/s]
2|5848|Loss: 0.8591736555099487:  90%|█████████ | 5848/6470 [1:07:17<07:28,  1.39it/s]
2|5849|Loss: 1.0402945280075073:  90%|█████████ | 5848/6470 [1:07:17<07:28,  1.39it/s]
2|5849|Loss: 1.0402945280075073:  90%|█████████ | 5849/6470 [1:07:18<06:45,  1.53it/s]
2|5850|Loss: 0.7340925931930542:  90%|█████████ | 5849/6470 [1:07:18<06:45,  1.53it/s]
2|5850|Loss: 0.7340925931930542:  90%|█████████ | 5850/6470 [1:07:18<07:07,  1.45it/s]
2|5851|Loss: 0.9211283922195435:  90%|█████████ | 5850/6470 [1:07:19<07:07,  1.45it/s]
2|5851|Loss: 0.9211283922195435:  90%|█████████ | 5851/6470 [1:07:19<07:38,  1.35it/s]
2|5852|Loss: 1.0720213651657104:  90%|█████████ | 5851/6470 [1:07:20<07:38,  1.35it/s]
2|5852|Loss: 1.0720213651657104:  90%|█████████ | 5852/6470 [1:07:20<07:28,  1.38it/s]
2|5853|Loss: 0.8150662183761597:  90%|█████████ | 5852/6470 [1:07:20<07:28,  1.38it/s]
2|5853|Loss: 0.8150662183761597:  90%|█████████ | 5853/6470 [1:07:21<07:40,  1.34it/s]
2|5854|Loss: 0.8086544275283813:  90%|█████████ | 5853/6470 [1:07:21<07:40,  1.34it/s]
2|5854|Loss: 0.8086544275283813:  90%|█████████ | 5854/6470 [1:07:22<07:41,  1.33it/s]
2|5855|Loss: 0.815323531627655:  90%|█████████ | 5854/6470 [1:07:22<07:41,  1.33it/s] 
2|5855|Loss: 0.815323531627655:  90%|█████████ | 5855/6470 [1:07:22<08:15,  1.24it/s]
2|5856|Loss: 0.7938168048858643:  90%|█████████ | 5855/6470 [1:07:23<08:15,  1.24it/s]
2|5856|Loss: 0.7938168048858643:  91%|█████████ | 5856/6470 [1:07:23<07:54,  1.29it/s]
2|5857|Loss: 0.850473165512085:  91%|█████████ | 5856/6470 [1:07:23<07:54,  1.29it/s] 
2|5857|Loss: 0.850473165512085:  91%|█████████ | 5857/6470 [1:07:24<07:11,  1.42it/s]
2|5858|Loss: 0.8609293103218079:  91%|█████████ | 5857/6470 [1:07:24<07:11,  1.42it/s]
2|5858|Loss: 0.8609293103218079:  91%|█████████ | 5858/6470 [1:07:24<06:36,  1.54it/s]
2|5859|Loss: 0.8084149360656738:  91%|█████████ | 5858/6470 [1:07:25<06:36,  1.54it/s]
2|5859|Loss: 0.8084149360656738:  91%|█████████ | 5859/6470 [1:07:25<06:37,  1.54it/s]
2|5860|Loss: 0.9174978733062744:  91%|█████████ | 5859/6470 [1:07:25<06:37,  1.54it/s]
2|5860|Loss: 0.9174978733062744:  91%|█████████ | 5860/6470 [1:07:26<06:46,  1.50it/s]
2|5861|Loss: 0.8958631157875061:  91%|█████████ | 5860/6470 [1:07:26<06:46,  1.50it/s]
2|5861|Loss: 0.8958631157875061:  91%|█████████ | 5861/6470 [1:07:26<06:59,  1.45it/s]
2|5862|Loss: 0.7012632489204407:  91%|█████████ | 5861/6470 [1:07:27<06:59,  1.45it/s]
2|5862|Loss: 0.7012632489204407:  91%|█████████ | 5862/6470 [1:07:27<07:21,  1.38it/s]
2|5863|Loss: 0.9175978302955627:  91%|█████████ | 5862/6470 [1:07:28<07:21,  1.38it/s]
2|5863|Loss: 0.9175978302955627:  91%|█████████ | 5863/6470 [1:07:28<07:34,  1.34it/s]
2|5864|Loss: 0.7043974995613098:  91%|█████████ | 5863/6470 [1:07:28<07:34,  1.34it/s]
2|5864|Loss: 0.7043974995613098:  91%|█████████ | 5864/6470 [1:07:29<07:32,  1.34it/s]
2|5865|Loss: 0.9619988203048706:  91%|█████████ | 5864/6470 [1:07:29<07:32,  1.34it/s]
2|5865|Loss: 0.9619988203048706:  91%|█████████ | 5865/6470 [1:07:29<07:46,  1.30it/s]
2|5866|Loss: 0.9947562217712402:  91%|█████████ | 5865/6470 [1:07:30<07:46,  1.30it/s]
2|5866|Loss: 0.9947562217712402:  91%|█████████ | 5866/6470 [1:07:30<07:30,  1.34it/s]
2|5867|Loss: 0.9234297871589661:  91%|█████████ | 5866/6470 [1:07:31<07:30,  1.34it/s]
2|5867|Loss: 0.9234297871589661:  91%|█████████ | 5867/6470 [1:07:31<07:20,  1.37it/s]
2|5868|Loss: 0.8488993048667908:  91%|█████████ | 5867/6470 [1:07:31<07:20,  1.37it/s]
2|5868|Loss: 0.8488993048667908:  91%|█████████ | 5868/6470 [1:07:32<07:06,  1.41it/s]
2|5869|Loss: 0.8223944902420044:  91%|█████████ | 5868/6470 [1:07:32<07:06,  1.41it/s]
2|5869|Loss: 0.8223944902420044:  91%|█████████ | 5869/6470 [1:07:32<06:43,  1.49it/s]
2|5870|Loss: 0.8240576386451721:  91%|█████████ | 5869/6470 [1:07:32<06:43,  1.49it/s]
2|5870|Loss: 0.8240576386451721:  91%|█████████ | 5870/6470 [1:07:33<06:48,  1.47it/s]
2|5871|Loss: 1.1070860624313354:  91%|█████████ | 5870/6470 [1:07:33<06:48,  1.47it/s]
2|5871|Loss: 1.1070860624313354:  91%|█████████ | 5871/6470 [1:07:34<06:57,  1.43it/s]
2|5872|Loss: 0.8873140811920166:  91%|█████████ | 5871/6470 [1:07:34<06:57,  1.43it/s]
2|5872|Loss: 0.8873140811920166:  91%|█████████ | 5872/6470 [1:07:34<07:03,  1.41it/s]
2|5873|Loss: 0.9017010927200317:  91%|█████████ | 5872/6470 [1:07:35<07:03,  1.41it/s]
2|5873|Loss: 0.9017010927200317:  91%|█████████ | 5873/6470 [1:07:35<07:07,  1.40it/s]
2|5874|Loss: 0.9990455508232117:  91%|█████████ | 5873/6470 [1:07:35<07:07,  1.40it/s]
2|5874|Loss: 0.9990455508232117:  91%|█████████ | 5874/6470 [1:07:35<06:16,  1.58it/s]
2|5875|Loss: 1.0061215162277222:  91%|█████████ | 5874/6470 [1:07:36<06:16,  1.58it/s]
2|5875|Loss: 1.0061215162277222:  91%|█████████ | 5875/6470 [1:07:36<06:16,  1.58it/s]
2|5876|Loss: 1.0012061595916748:  91%|█████████ | 5875/6470 [1:07:36<06:16,  1.58it/s]
2|5876|Loss: 1.0012061595916748:  91%|█████████ | 5876/6470 [1:07:37<06:14,  1.59it/s]
2|5877|Loss: 0.7192491292953491:  91%|█████████ | 5876/6470 [1:07:37<06:14,  1.59it/s]
2|5877|Loss: 0.7192491292953491:  91%|█████████ | 5877/6470 [1:07:38<06:41,  1.48it/s]
2|5878|Loss: 0.8171344995498657:  91%|█████████ | 5877/6470 [1:07:38<06:41,  1.48it/s]
2|5878|Loss: 0.8171344995498657:  91%|█████████ | 5878/6470 [1:07:38<06:55,  1.42it/s]
2|5879|Loss: 0.9245577454566956:  91%|█████████ | 5878/6470 [1:07:39<06:55,  1.42it/s]
2|5879|Loss: 0.9245577454566956:  91%|█████████ | 5879/6470 [1:07:39<06:49,  1.44it/s]
2|5880|Loss: 0.9701297879219055:  91%|█████████ | 5879/6470 [1:07:39<06:49,  1.44it/s]
2|5880|Loss: 0.9701297879219055:  91%|█████████ | 5880/6470 [1:07:40<06:59,  1.41it/s]
2|5881|Loss: 0.8435536026954651:  91%|█████████ | 5880/6470 [1:07:40<06:59,  1.41it/s]
2|5881|Loss: 0.8435536026954651:  91%|█████████ | 5881/6470 [1:07:40<07:01,  1.40it/s]
2|5882|Loss: 0.9091688394546509:  91%|█████████ | 5881/6470 [1:07:41<07:01,  1.40it/s]
2|5882|Loss: 0.9091688394546509:  91%|█████████ | 5882/6470 [1:07:41<06:49,  1.43it/s]
2|5883|Loss: 0.8156943321228027:  91%|█████████ | 5882/6470 [1:07:41<06:49,  1.43it/s]
2|5883|Loss: 0.8156943321228027:  91%|█████████ | 5883/6470 [1:07:42<06:54,  1.42it/s]
2|5884|Loss: 0.9976699948310852:  91%|█████████ | 5883/6470 [1:07:42<06:54,  1.42it/s]
2|5884|Loss: 0.9976699948310852:  91%|█████████ | 5884/6470 [1:07:43<07:12,  1.35it/s]
2|5885|Loss: 0.8822254538536072:  91%|█████████ | 5884/6470 [1:07:43<07:12,  1.35it/s]
2|5885|Loss: 0.8822254538536072:  91%|█████████ | 5885/6470 [1:07:43<07:14,  1.35it/s]
2|5886|Loss: 0.8339030742645264:  91%|█████████ | 5885/6470 [1:07:44<07:14,  1.35it/s]
2|5886|Loss: 0.8339030742645264:  91%|█████████ | 5886/6470 [1:07:44<06:49,  1.43it/s]
2|5887|Loss: 0.7419211864471436:  91%|█████████ | 5886/6470 [1:07:44<06:49,  1.43it/s]
2|5887|Loss: 0.7419211864471436:  91%|█████████ | 5887/6470 [1:07:45<06:31,  1.49it/s]
2|5888|Loss: 0.8906880021095276:  91%|█████████ | 5887/6470 [1:07:45<06:31,  1.49it/s]
2|5888|Loss: 0.8906880021095276:  91%|█████████ | 5888/6470 [1:07:45<06:14,  1.55it/s]
2|5889|Loss: 0.9697144031524658:  91%|█████████ | 5888/6470 [1:07:45<06:14,  1.55it/s]
2|5889|Loss: 0.9697144031524658:  91%|█████████ | 5889/6470 [1:07:46<05:45,  1.68it/s]
2|5890|Loss: 0.726851761341095:  91%|█████████ | 5889/6470 [1:07:46<05:45,  1.68it/s] 
2|5890|Loss: 0.726851761341095:  91%|█████████ | 5890/6470 [1:07:46<05:45,  1.68it/s]
2|5891|Loss: 0.9000467658042908:  91%|█████████ | 5890/6470 [1:07:47<05:45,  1.68it/s]
2|5891|Loss: 0.9000467658042908:  91%|█████████ | 5891/6470 [1:07:47<06:00,  1.61it/s]
2|5892|Loss: 0.7939425110816956:  91%|█████████ | 5891/6470 [1:07:47<06:00,  1.61it/s]
2|5892|Loss: 0.7939425110816956:  91%|█████████ | 5892/6470 [1:07:48<06:11,  1.55it/s]
2|5893|Loss: 0.8525089621543884:  91%|█████████ | 5892/6470 [1:07:48<06:11,  1.55it/s]
2|5893|Loss: 0.8525089621543884:  91%|█████████ | 5893/6470 [1:07:48<06:21,  1.51it/s]
2|5894|Loss: 1.0542620420455933:  91%|█████████ | 5893/6470 [1:07:49<06:21,  1.51it/s]
2|5894|Loss: 1.0542620420455933:  91%|█████████ | 5894/6470 [1:07:49<06:27,  1.49it/s]
2|5895|Loss: 0.9542796015739441:  91%|█████████ | 5894/6470 [1:07:49<06:27,  1.49it/s]
2|5895|Loss: 0.9542796015739441:  91%|█████████ | 5895/6470 [1:07:50<06:32,  1.47it/s]
2|5896|Loss: 0.6755011677742004:  91%|█████████ | 5895/6470 [1:07:50<06:32,  1.47it/s]
2|5896|Loss: 0.6755011677742004:  91%|█████████ | 5896/6470 [1:07:50<06:39,  1.44it/s]
2|5897|Loss: 0.911354660987854:  91%|█████████ | 5896/6470 [1:07:51<06:39,  1.44it/s] 
2|5897|Loss: 0.911354660987854:  91%|█████████ | 5897/6470 [1:07:51<06:10,  1.55it/s]
2|5898|Loss: 0.9065675139427185:  91%|█████████ | 5897/6470 [1:07:51<06:10,  1.55it/s]
2|5898|Loss: 0.9065675139427185:  91%|█████████ | 5898/6470 [1:07:52<06:00,  1.59it/s]
2|5899|Loss: 0.8852935433387756:  91%|█████████ | 5898/6470 [1:07:52<06:00,  1.59it/s]
2|5899|Loss: 0.8852935433387756:  91%|█████████ | 5899/6470 [1:07:52<05:56,  1.60it/s]
2|5900|Loss: 1.1046767234802246:  91%|█████████ | 5899/6470 [1:07:53<05:56,  1.60it/s]
2|5900|Loss: 1.1046767234802246:  91%|█████████ | 5900/6470 [1:07:53<06:06,  1.55it/s]
2|5901|Loss: 0.7592819333076477:  91%|█████████ | 5900/6470 [1:07:53<06:06,  1.55it/s]
2|5901|Loss: 0.7592819333076477:  91%|█████████ | 5901/6470 [1:07:54<06:20,  1.50it/s]
2|5902|Loss: 0.8515218496322632:  91%|█████████ | 5901/6470 [1:07:54<06:20,  1.50it/s]
2|5902|Loss: 0.8515218496322632:  91%|█████████ | 5902/6470 [1:07:54<06:25,  1.47it/s]
2|5903|Loss: 0.7043390274047852:  91%|█████████ | 5902/6470 [1:07:55<06:25,  1.47it/s]
2|5903|Loss: 0.7043390274047852:  91%|█████████ | 5903/6470 [1:07:55<06:53,  1.37it/s]
2|5904|Loss: 0.8387301564216614:  91%|█████████ | 5903/6470 [1:07:56<06:53,  1.37it/s]
2|5904|Loss: 0.8387301564216614:  91%|█████████▏| 5904/6470 [1:07:56<07:02,  1.34it/s]
2|5905|Loss: 0.6782087683677673:  91%|█████████▏| 5904/6470 [1:07:56<07:02,  1.34it/s]
2|5905|Loss: 0.6782087683677673:  91%|█████████▏| 5905/6470 [1:07:57<06:45,  1.39it/s]
2|5906|Loss: 0.8324146270751953:  91%|█████████▏| 5905/6470 [1:07:57<06:45,  1.39it/s]
2|5906|Loss: 0.8324146270751953:  91%|█████████▏| 5906/6470 [1:07:57<06:49,  1.38it/s]
2|5907|Loss: 0.804895281791687:  91%|█████████▏| 5906/6470 [1:07:58<06:49,  1.38it/s] 
2|5907|Loss: 0.804895281791687:  91%|█████████▏| 5907/6470 [1:07:58<06:28,  1.45it/s]
2|5908|Loss: 1.0496931076049805:  91%|█████████▏| 5907/6470 [1:07:58<06:28,  1.45it/s]
2|5908|Loss: 1.0496931076049805:  91%|█████████▏| 5908/6470 [1:07:59<06:30,  1.44it/s]
2|5909|Loss: 0.7596964836120605:  91%|█████████▏| 5908/6470 [1:07:59<06:30,  1.44it/s]
2|5909|Loss: 0.7596964836120605:  91%|█████████▏| 5909/6470 [1:07:59<06:23,  1.46it/s]
2|5910|Loss: 0.9988676905632019:  91%|█████████▏| 5909/6470 [1:08:00<06:23,  1.46it/s]
2|5910|Loss: 0.9988676905632019:  91%|█████████▏| 5910/6470 [1:08:00<05:59,  1.56it/s]
2|5911|Loss: 0.8220570683479309:  91%|█████████▏| 5910/6470 [1:08:00<05:59,  1.56it/s]
2|5911|Loss: 0.8220570683479309:  91%|█████████▏| 5911/6470 [1:08:01<06:05,  1.53it/s]
2|5912|Loss: 0.7340090274810791:  91%|█████████▏| 5911/6470 [1:08:01<06:05,  1.53it/s]
2|5912|Loss: 0.7340090274810791:  91%|█████████▏| 5912/6470 [1:08:01<06:38,  1.40it/s]
2|5913|Loss: 0.8821031451225281:  91%|█████████▏| 5912/6470 [1:08:02<06:38,  1.40it/s]
2|5913|Loss: 0.8821031451225281:  91%|█████████▏| 5913/6470 [1:08:02<07:11,  1.29it/s]
2|5914|Loss: 0.901005208492279:  91%|█████████▏| 5913/6470 [1:08:03<07:11,  1.29it/s] 
2|5914|Loss: 0.901005208492279:  91%|█████████▏| 5914/6470 [1:08:03<06:50,  1.35it/s]
2|5915|Loss: 0.8261606097221375:  91%|█████████▏| 5914/6470 [1:08:03<06:50,  1.35it/s]
2|5915|Loss: 0.8261606097221375:  91%|█████████▏| 5915/6470 [1:08:04<06:27,  1.43it/s]
2|5916|Loss: 0.8309673070907593:  91%|█████████▏| 5915/6470 [1:08:04<06:27,  1.43it/s]
2|5916|Loss: 0.8309673070907593:  91%|█████████▏| 5916/6470 [1:08:04<06:25,  1.44it/s]
2|5917|Loss: 0.8039356470108032:  91%|█████████▏| 5916/6470 [1:08:05<06:25,  1.44it/s]
2|5917|Loss: 0.8039356470108032:  91%|█████████▏| 5917/6470 [1:08:05<06:11,  1.49it/s]
2|5918|Loss: 0.824249804019928:  91%|█████████▏| 5917/6470 [1:08:05<06:11,  1.49it/s] 
2|5918|Loss: 0.824249804019928:  91%|█████████▏| 5918/6470 [1:08:05<06:03,  1.52it/s]
2|5919|Loss: 0.9842001795768738:  91%|█████████▏| 5918/6470 [1:08:06<06:03,  1.52it/s]
2|5919|Loss: 0.9842001795768738:  91%|█████████▏| 5919/6470 [1:08:06<06:00,  1.53it/s]
2|5920|Loss: 0.9020509719848633:  91%|█████████▏| 5919/6470 [1:08:07<06:00,  1.53it/s]
2|5920|Loss: 0.9020509719848633:  91%|█████████▏| 5920/6470 [1:08:07<06:09,  1.49it/s]
2|5921|Loss: 0.773942768573761:  91%|█████████▏| 5920/6470 [1:08:07<06:09,  1.49it/s] 
2|5921|Loss: 0.773942768573761:  92%|█████████▏| 5921/6470 [1:08:08<06:15,  1.46it/s]
2|5922|Loss: 0.7107378840446472:  92%|█████████▏| 5921/6470 [1:08:08<06:15,  1.46it/s]
2|5922|Loss: 0.7107378840446472:  92%|█████████▏| 5922/6470 [1:08:08<06:29,  1.41it/s]
2|5923|Loss: 0.8765829801559448:  92%|█████████▏| 5922/6470 [1:08:09<06:29,  1.41it/s]
2|5923|Loss: 0.8765829801559448:  92%|█████████▏| 5923/6470 [1:08:09<06:14,  1.46it/s]
2|5924|Loss: 0.8709895014762878:  92%|█████████▏| 5923/6470 [1:08:09<06:14,  1.46it/s]
2|5924|Loss: 0.8709895014762878:  92%|█████████▏| 5924/6470 [1:08:10<06:04,  1.50it/s]
2|5925|Loss: 1.0042474269866943:  92%|█████████▏| 5924/6470 [1:08:10<06:04,  1.50it/s]
2|5925|Loss: 1.0042474269866943:  92%|█████████▏| 5925/6470 [1:08:10<06:28,  1.40it/s]
2|5926|Loss: 0.8428101539611816:  92%|█████████▏| 5925/6470 [1:08:11<06:28,  1.40it/s]
2|5926|Loss: 0.8428101539611816:  92%|█████████▏| 5926/6470 [1:08:11<06:32,  1.39it/s]
2|5927|Loss: 0.7587350606918335:  92%|█████████▏| 5926/6470 [1:08:12<06:32,  1.39it/s]
2|5927|Loss: 0.7587350606918335:  92%|█████████▏| 5927/6470 [1:08:12<06:35,  1.37it/s]
2|5928|Loss: 0.7264963388442993:  92%|█████████▏| 5927/6470 [1:08:12<06:35,  1.37it/s]
2|5928|Loss: 0.7264963388442993:  92%|█████████▏| 5928/6470 [1:08:13<06:45,  1.34it/s]
2|5929|Loss: 0.7472584843635559:  92%|█████████▏| 5928/6470 [1:08:13<06:45,  1.34it/s]
2|5929|Loss: 0.7472584843635559:  92%|█████████▏| 5929/6470 [1:08:13<06:44,  1.34it/s]
2|5930|Loss: 0.786985456943512:  92%|█████████▏| 5929/6470 [1:08:14<06:44,  1.34it/s] 
2|5930|Loss: 0.786985456943512:  92%|█████████▏| 5930/6470 [1:08:14<06:50,  1.32it/s]
2|5931|Loss: 0.7951504588127136:  92%|█████████▏| 5930/6470 [1:08:15<06:50,  1.32it/s]
2|5931|Loss: 0.7951504588127136:  92%|█████████▏| 5931/6470 [1:08:15<06:27,  1.39it/s]
2|5932|Loss: 0.8927188515663147:  92%|█████████▏| 5931/6470 [1:08:15<06:27,  1.39it/s]
2|5932|Loss: 0.8927188515663147:  92%|█████████▏| 5932/6470 [1:08:16<06:21,  1.41it/s]
2|5933|Loss: 0.8299881815910339:  92%|█████████▏| 5932/6470 [1:08:16<06:21,  1.41it/s]
2|5933|Loss: 0.8299881815910339:  92%|█████████▏| 5933/6470 [1:08:16<06:35,  1.36it/s]
2|5934|Loss: 1.0644010305404663:  92%|█████████▏| 5933/6470 [1:08:17<06:35,  1.36it/s]
2|5934|Loss: 1.0644010305404663:  92%|█████████▏| 5934/6470 [1:08:17<06:39,  1.34it/s]
2|5935|Loss: 0.8664320707321167:  92%|█████████▏| 5934/6470 [1:08:17<06:39,  1.34it/s]
2|5935|Loss: 0.8664320707321167:  92%|█████████▏| 5935/6470 [1:08:18<06:37,  1.35it/s]
2|5936|Loss: 0.874623715877533:  92%|█████████▏| 5935/6470 [1:08:18<06:37,  1.35it/s] 
2|5936|Loss: 0.874623715877533:  92%|█████████▏| 5936/6470 [1:08:19<06:37,  1.34it/s]
2|5937|Loss: 0.9961999654769897:  92%|█████████▏| 5936/6470 [1:08:19<06:37,  1.34it/s]
2|5937|Loss: 0.9961999654769897:  92%|█████████▏| 5937/6470 [1:08:19<06:12,  1.43it/s]
2|5938|Loss: 0.8814180493354797:  92%|█████████▏| 5937/6470 [1:08:20<06:12,  1.43it/s]
2|5938|Loss: 0.8814180493354797:  92%|█████████▏| 5938/6470 [1:08:20<05:56,  1.49it/s]
2|5939|Loss: 0.7124845385551453:  92%|█████████▏| 5938/6470 [1:08:20<05:56,  1.49it/s]
2|5939|Loss: 0.7124845385551453:  92%|█████████▏| 5939/6470 [1:08:20<05:59,  1.48it/s]
2|5940|Loss: 0.8133692741394043:  92%|█████████▏| 5939/6470 [1:08:21<05:59,  1.48it/s]
2|5940|Loss: 0.8133692741394043:  92%|█████████▏| 5940/6470 [1:08:21<05:52,  1.51it/s]
2|5941|Loss: 0.9980130195617676:  92%|█████████▏| 5940/6470 [1:08:21<05:52,  1.51it/s]
2|5941|Loss: 0.9980130195617676:  92%|█████████▏| 5941/6470 [1:08:22<05:36,  1.57it/s]
2|5942|Loss: 0.8852048516273499:  92%|█████████▏| 5941/6470 [1:08:22<05:36,  1.57it/s]
2|5942|Loss: 0.8852048516273499:  92%|█████████▏| 5942/6470 [1:08:22<05:31,  1.59it/s]
2|5943|Loss: 0.8130648136138916:  92%|█████████▏| 5942/6470 [1:08:23<05:31,  1.59it/s]
2|5943|Loss: 0.8130648136138916:  92%|█████████▏| 5943/6470 [1:08:23<05:40,  1.55it/s]
2|5944|Loss: 1.0031635761260986:  92%|█████████▏| 5943/6470 [1:08:23<05:40,  1.55it/s]
2|5944|Loss: 1.0031635761260986:  92%|█████████▏| 5944/6470 [1:08:24<05:52,  1.49it/s]
2|5945|Loss: 0.8048694729804993:  92%|█████████▏| 5944/6470 [1:08:24<05:52,  1.49it/s]
2|5945|Loss: 0.8048694729804993:  92%|█████████▏| 5945/6470 [1:08:24<06:09,  1.42it/s]
2|5946|Loss: 0.8102565407752991:  92%|█████████▏| 5945/6470 [1:08:25<06:09,  1.42it/s]
2|5946|Loss: 0.8102565407752991:  92%|█████████▏| 5946/6470 [1:08:25<06:04,  1.44it/s]
2|5947|Loss: 0.832757830619812:  92%|█████████▏| 5946/6470 [1:08:26<06:04,  1.44it/s] 
2|5947|Loss: 0.832757830619812:  92%|█████████▏| 5947/6470 [1:08:26<06:07,  1.42it/s]
2|5948|Loss: 0.7772579789161682:  92%|█████████▏| 5947/6470 [1:08:26<06:07,  1.42it/s]
2|5948|Loss: 0.7772579789161682:  92%|█████████▏| 5948/6470 [1:08:27<05:58,  1.46it/s]
2|5949|Loss: 0.7821073532104492:  92%|█████████▏| 5948/6470 [1:08:27<05:58,  1.46it/s]
2|5949|Loss: 0.7821073532104492:  92%|█████████▏| 5949/6470 [1:08:27<06:24,  1.35it/s]
2|5950|Loss: 0.8280306458473206:  92%|█████████▏| 5949/6470 [1:08:28<06:24,  1.35it/s]
2|5950|Loss: 0.8280306458473206:  92%|█████████▏| 5950/6470 [1:08:28<07:05,  1.22it/s]
2|5951|Loss: 0.8701523542404175:  92%|█████████▏| 5950/6470 [1:08:29<07:05,  1.22it/s]
2|5951|Loss: 0.8701523542404175:  92%|█████████▏| 5951/6470 [1:08:29<06:50,  1.26it/s]
2|5952|Loss: 0.7356551885604858:  92%|█████████▏| 5951/6470 [1:08:29<06:50,  1.26it/s]
2|5952|Loss: 0.7356551885604858:  92%|█████████▏| 5952/6470 [1:08:30<06:11,  1.39it/s]
2|5953|Loss: 0.9443905353546143:  92%|█████████▏| 5952/6470 [1:08:30<06:11,  1.39it/s]
2|5953|Loss: 0.9443905353546143:  92%|█████████▏| 5953/6470 [1:08:30<05:51,  1.47it/s]
2|5954|Loss: 0.8568049669265747:  92%|█████████▏| 5953/6470 [1:08:31<05:51,  1.47it/s]
2|5954|Loss: 0.8568049669265747:  92%|█████████▏| 5954/6470 [1:08:31<06:04,  1.42it/s]
2|5955|Loss: 0.7291925549507141:  92%|█████████▏| 5954/6470 [1:08:31<06:04,  1.42it/s]
2|5955|Loss: 0.7291925549507141:  92%|█████████▏| 5955/6470 [1:08:32<05:34,  1.54it/s]
2|5956|Loss: 1.0522794723510742:  92%|█████████▏| 5955/6470 [1:08:32<05:34,  1.54it/s]
2|5956|Loss: 1.0522794723510742:  92%|█████████▏| 5956/6470 [1:08:32<05:23,  1.59it/s]
2|5957|Loss: 0.6971855163574219:  92%|█████████▏| 5956/6470 [1:08:32<05:23,  1.59it/s]
2|5957|Loss: 0.6971855163574219:  92%|█████████▏| 5957/6470 [1:08:33<05:04,  1.68it/s]
2|5958|Loss: 0.8910446166992188:  92%|█████████▏| 5957/6470 [1:08:33<05:04,  1.68it/s]
2|5958|Loss: 0.8910446166992188:  92%|█████████▏| 5958/6470 [1:08:33<05:13,  1.63it/s]
2|5959|Loss: 0.7672825455665588:  92%|█████████▏| 5958/6470 [1:08:34<05:13,  1.63it/s]
2|5959|Loss: 0.7672825455665588:  92%|█████████▏| 5959/6470 [1:08:34<05:45,  1.48it/s]
2|5960|Loss: 0.8888590335845947:  92%|█████████▏| 5959/6470 [1:08:35<05:45,  1.48it/s]
2|5960|Loss: 0.8888590335845947:  92%|█████████▏| 5960/6470 [1:08:35<06:10,  1.38it/s]
2|5961|Loss: 0.809149444103241:  92%|█████████▏| 5960/6470 [1:08:35<06:10,  1.38it/s] 
2|5961|Loss: 0.809149444103241:  92%|█████████▏| 5961/6470 [1:08:36<06:09,  1.38it/s]
2|5962|Loss: 0.9211868047714233:  92%|█████████▏| 5961/6470 [1:08:36<06:09,  1.38it/s]
2|5962|Loss: 0.9211868047714233:  92%|█████████▏| 5962/6470 [1:08:36<06:09,  1.38it/s]
2|5963|Loss: 0.8079388737678528:  92%|█████████▏| 5962/6470 [1:08:37<06:09,  1.38it/s]
2|5963|Loss: 0.8079388737678528:  92%|█████████▏| 5963/6470 [1:08:37<06:09,  1.37it/s]
2|5964|Loss: 0.8562654256820679:  92%|█████████▏| 5963/6470 [1:08:38<06:09,  1.37it/s]
2|5964|Loss: 0.8562654256820679:  92%|█████████▏| 5964/6470 [1:08:38<05:58,  1.41it/s]
2|5965|Loss: 0.8900846242904663:  92%|█████████▏| 5964/6470 [1:08:38<05:58,  1.41it/s]
2|5965|Loss: 0.8900846242904663:  92%|█████████▏| 5965/6470 [1:08:38<05:47,  1.45it/s]
2|5966|Loss: 0.9130601286888123:  92%|█████████▏| 5965/6470 [1:08:39<05:47,  1.45it/s]
2|5966|Loss: 0.9130601286888123:  92%|█████████▏| 5966/6470 [1:08:39<05:39,  1.49it/s]
2|5967|Loss: 0.8113528490066528:  92%|█████████▏| 5966/6470 [1:08:39<05:39,  1.49it/s]
2|5967|Loss: 0.8113528490066528:  92%|█████████▏| 5967/6470 [1:08:40<05:22,  1.56it/s]
2|5968|Loss: 0.8329200744628906:  92%|█████████▏| 5967/6470 [1:08:40<05:22,  1.56it/s]
2|5968|Loss: 0.8329200744628906:  92%|█████████▏| 5968/6470 [1:08:40<05:27,  1.53it/s]
2|5969|Loss: 0.6614564657211304:  92%|█████████▏| 5968/6470 [1:08:41<05:27,  1.53it/s]
2|5969|Loss: 0.6614564657211304:  92%|█████████▏| 5969/6470 [1:08:41<05:34,  1.50it/s]
2|5970|Loss: 0.8796143531799316:  92%|█████████▏| 5969/6470 [1:08:41<05:34,  1.50it/s]
2|5970|Loss: 0.8796143531799316:  92%|█████████▏| 5970/6470 [1:08:42<05:45,  1.45it/s]
2|5971|Loss: 0.8619496822357178:  92%|█████████▏| 5970/6470 [1:08:42<05:45,  1.45it/s]
2|5971|Loss: 0.8619496822357178:  92%|█████████▏| 5971/6470 [1:08:42<05:24,  1.54it/s]
2|5972|Loss: 0.8345897793769836:  92%|█████████▏| 5971/6470 [1:08:43<05:24,  1.54it/s]
2|5972|Loss: 0.8345897793769836:  92%|█████████▏| 5972/6470 [1:08:43<05:41,  1.46it/s]
2|5973|Loss: 1.0780458450317383:  92%|█████████▏| 5972/6470 [1:08:43<05:41,  1.46it/s]
2|5973|Loss: 1.0780458450317383:  92%|█████████▏| 5973/6470 [1:08:44<05:37,  1.47it/s]
2|5974|Loss: 0.9727374911308289:  92%|█████████▏| 5973/6470 [1:08:44<05:37,  1.47it/s]
2|5974|Loss: 0.9727374911308289:  92%|█████████▏| 5974/6470 [1:08:44<05:14,  1.58it/s]
2|5975|Loss: 0.6107653379440308:  92%|█████████▏| 5974/6470 [1:08:45<05:14,  1.58it/s]
2|5975|Loss: 0.6107653379440308:  92%|█████████▏| 5975/6470 [1:08:45<05:48,  1.42it/s]
2|5976|Loss: 0.8505201935768127:  92%|█████████▏| 5975/6470 [1:08:46<05:48,  1.42it/s]
2|5976|Loss: 0.8505201935768127:  92%|█████████▏| 5976/6470 [1:08:46<06:18,  1.31it/s]
2|5977|Loss: 0.8915310502052307:  92%|█████████▏| 5976/6470 [1:08:46<06:18,  1.31it/s]
2|5977|Loss: 0.8915310502052307:  92%|█████████▏| 5977/6470 [1:08:47<05:52,  1.40it/s]
2|5978|Loss: 0.8064706921577454:  92%|█████████▏| 5977/6470 [1:08:47<05:52,  1.40it/s]
2|5978|Loss: 0.8064706921577454:  92%|█████████▏| 5978/6470 [1:08:47<06:08,  1.34it/s]
2|5979|Loss: 0.9361342191696167:  92%|█████████▏| 5978/6470 [1:08:48<06:08,  1.34it/s]
2|5979|Loss: 0.9361342191696167:  92%|█████████▏| 5979/6470 [1:08:48<06:07,  1.33it/s]
2|5980|Loss: 0.7405628561973572:  92%|█████████▏| 5979/6470 [1:08:49<06:07,  1.33it/s]
2|5980|Loss: 0.7405628561973572:  92%|█████████▏| 5980/6470 [1:08:49<06:13,  1.31it/s]
2|5981|Loss: 0.6967013478279114:  92%|█████████▏| 5980/6470 [1:08:49<06:13,  1.31it/s]
2|5981|Loss: 0.6967013478279114:  92%|█████████▏| 5981/6470 [1:08:50<05:54,  1.38it/s]
2|5982|Loss: 0.6876150369644165:  92%|█████████▏| 5981/6470 [1:08:50<05:54,  1.38it/s]
2|5982|Loss: 0.6876150369644165:  92%|█████████▏| 5982/6470 [1:08:50<05:39,  1.44it/s]
2|5983|Loss: 0.7191184163093567:  92%|█████████▏| 5982/6470 [1:08:51<05:39,  1.44it/s]
2|5983|Loss: 0.7191184163093567:  92%|█████████▏| 5983/6470 [1:08:51<05:41,  1.43it/s]
2|5984|Loss: 0.7657359838485718:  92%|█████████▏| 5983/6470 [1:08:51<05:41,  1.43it/s]
2|5984|Loss: 0.7657359838485718:  92%|█████████▏| 5984/6470 [1:08:52<05:19,  1.52it/s]
2|5985|Loss: 0.8430867791175842:  92%|█████████▏| 5984/6470 [1:08:52<05:19,  1.52it/s]
2|5985|Loss: 0.8430867791175842:  93%|█████████▎| 5985/6470 [1:08:52<05:04,  1.59it/s]
2|5986|Loss: 0.8317094445228577:  93%|█████████▎| 5985/6470 [1:08:52<05:04,  1.59it/s]
2|5986|Loss: 0.8317094445228577:  93%|█████████▎| 5986/6470 [1:08:53<05:15,  1.54it/s]
2|5987|Loss: 0.8867287635803223:  93%|█████████▎| 5986/6470 [1:08:53<05:15,  1.54it/s]
2|5987|Loss: 0.8867287635803223:  93%|█████████▎| 5987/6470 [1:08:54<05:34,  1.44it/s]
2|5988|Loss: 0.8122119307518005:  93%|█████████▎| 5987/6470 [1:08:54<05:34,  1.44it/s]
2|5988|Loss: 0.8122119307518005:  93%|█████████▎| 5988/6470 [1:08:54<05:25,  1.48it/s]
2|5989|Loss: 0.9190274477005005:  93%|█████████▎| 5988/6470 [1:08:55<05:25,  1.48it/s]
2|5989|Loss: 0.9190274477005005:  93%|█████████▎| 5989/6470 [1:08:55<05:07,  1.56it/s]
2|5990|Loss: 0.7944535613059998:  93%|█████████▎| 5989/6470 [1:08:55<05:07,  1.56it/s]
2|5990|Loss: 0.7944535613059998:  93%|█████████▎| 5990/6470 [1:08:55<05:04,  1.58it/s]
2|5991|Loss: 0.7143503427505493:  93%|█████████▎| 5990/6470 [1:08:56<05:04,  1.58it/s]
2|5991|Loss: 0.7143503427505493:  93%|█████████▎| 5991/6470 [1:08:56<05:16,  1.51it/s]
2|5992|Loss: 0.93994140625:  93%|█████████▎| 5991/6470 [1:08:57<05:16,  1.51it/s]     
2|5992|Loss: 0.93994140625:  93%|█████████▎| 5992/6470 [1:08:57<05:37,  1.42it/s]
2|5993|Loss: 0.8843832612037659:  93%|█████████▎| 5992/6470 [1:08:57<05:37,  1.42it/s]
2|5993|Loss: 0.8843832612037659:  93%|█████████▎| 5993/6470 [1:08:58<05:34,  1.43it/s]
2|5994|Loss: 0.9946828484535217:  93%|█████████▎| 5993/6470 [1:08:58<05:34,  1.43it/s]
2|5994|Loss: 0.9946828484535217:  93%|█████████▎| 5994/6470 [1:08:58<05:51,  1.35it/s]
2|5995|Loss: 0.8700764775276184:  93%|█████████▎| 5994/6470 [1:08:59<05:51,  1.35it/s]
2|5995|Loss: 0.8700764775276184:  93%|█████████▎| 5995/6470 [1:08:59<05:23,  1.47it/s]
2|5996|Loss: 0.9498919248580933:  93%|█████████▎| 5995/6470 [1:08:59<05:23,  1.47it/s]
2|5996|Loss: 0.9498919248580933:  93%|█████████▎| 5996/6470 [1:09:00<05:22,  1.47it/s]
2|5997|Loss: 0.818644106388092:  93%|█████████▎| 5996/6470 [1:09:00<05:22,  1.47it/s] 
2|5997|Loss: 0.818644106388092:  93%|█████████▎| 5997/6470 [1:09:00<05:32,  1.42it/s]
2|5998|Loss: 0.8182855248451233:  93%|█████████▎| 5997/6470 [1:09:01<05:32,  1.42it/s]
2|5998|Loss: 0.8182855248451233:  93%|█████████▎| 5998/6470 [1:09:01<05:21,  1.47it/s]
2|5999|Loss: 0.8343127369880676:  93%|█████████▎| 5998/6470 [1:09:01<05:21,  1.47it/s]
2|5999|Loss: 0.8343127369880676:  93%|█████████▎| 5999/6470 [1:09:02<05:12,  1.51it/s]
2|6000|Loss: 0.8080202341079712:  93%|█████████▎| 5999/6470 [1:09:02<05:12,  1.51it/s]
2|6000|Loss: 0.8080202341079712:  93%|█████████▎| 6000/6470 [1:09:02<05:05,  1.54it/s]
2|6001|Loss: 0.9053202867507935:  93%|█████████▎| 6000/6470 [1:09:03<05:05,  1.54it/s]
2|6001|Loss: 0.9053202867507935:  93%|█████████▎| 6001/6470 [1:09:03<05:11,  1.51it/s]
2|6002|Loss: 1.2023992538452148:  93%|█████████▎| 6001/6470 [1:09:03<05:11,  1.51it/s]
2|6002|Loss: 1.2023992538452148:  93%|█████████▎| 6002/6470 [1:09:04<05:03,  1.54it/s]
2|6003|Loss: 0.9149366617202759:  93%|█████████▎| 6002/6470 [1:09:04<05:03,  1.54it/s]
2|6003|Loss: 0.9149366617202759:  93%|█████████▎| 6003/6470 [1:09:04<05:22,  1.45it/s]
2|6004|Loss: 0.7920218706130981:  93%|█████████▎| 6003/6470 [1:09:05<05:22,  1.45it/s]
2|6004|Loss: 0.7920218706130981:  93%|█████████▎| 6004/6470 [1:09:05<05:43,  1.36it/s]
2|6005|Loss: 0.8840495347976685:  93%|█████████▎| 6004/6470 [1:09:06<05:43,  1.36it/s]
2|6005|Loss: 0.8840495347976685:  93%|█████████▎| 6005/6470 [1:09:06<05:35,  1.38it/s]
2|6006|Loss: 0.9550386667251587:  93%|█████████▎| 6005/6470 [1:09:06<05:35,  1.38it/s]
2|6006|Loss: 0.9550386667251587:  93%|█████████▎| 6006/6470 [1:09:07<05:20,  1.45it/s]
2|6007|Loss: 0.8755016326904297:  93%|█████████▎| 6006/6470 [1:09:07<05:20,  1.45it/s]
2|6007|Loss: 0.8755016326904297:  93%|█████████▎| 6007/6470 [1:09:07<04:59,  1.55it/s]
2|6008|Loss: 0.7599275708198547:  93%|█████████▎| 6007/6470 [1:09:07<04:59,  1.55it/s]
2|6008|Loss: 0.7599275708198547:  93%|█████████▎| 6008/6470 [1:09:08<04:33,  1.69it/s]
2|6009|Loss: 0.8597453236579895:  93%|█████████▎| 6008/6470 [1:09:08<04:33,  1.69it/s]
2|6009|Loss: 0.8597453236579895:  93%|█████████▎| 6009/6470 [1:09:08<04:53,  1.57it/s]
2|6010|Loss: 0.8852407932281494:  93%|█████████▎| 6009/6470 [1:09:09<04:53,  1.57it/s]
2|6010|Loss: 0.8852407932281494:  93%|█████████▎| 6010/6470 [1:09:09<04:53,  1.57it/s]
2|6011|Loss: 0.7126863598823547:  93%|█████████▎| 6010/6470 [1:09:09<04:53,  1.57it/s]
2|6011|Loss: 0.7126863598823547:  93%|█████████▎| 6011/6470 [1:09:10<05:12,  1.47it/s]
2|6012|Loss: 0.8820192217826843:  93%|█████████▎| 6011/6470 [1:09:10<05:12,  1.47it/s]
2|6012|Loss: 0.8820192217826843:  93%|█████████▎| 6012/6470 [1:09:11<05:32,  1.38it/s]
2|6013|Loss: 0.8619831204414368:  93%|█████████▎| 6012/6470 [1:09:11<05:32,  1.38it/s]
2|6013|Loss: 0.8619831204414368:  93%|█████████▎| 6013/6470 [1:09:11<05:42,  1.33it/s]
2|6014|Loss: 0.8440731167793274:  93%|█████████▎| 6013/6470 [1:09:12<05:42,  1.33it/s]
2|6014|Loss: 0.8440731167793274:  93%|█████████▎| 6014/6470 [1:09:12<05:29,  1.38it/s]
2|6015|Loss: 0.9294840693473816:  93%|█████████▎| 6014/6470 [1:09:12<05:29,  1.38it/s]
2|6015|Loss: 0.9294840693473816:  93%|█████████▎| 6015/6470 [1:09:13<05:24,  1.40it/s]
2|6016|Loss: 0.7913562655448914:  93%|█████████▎| 6015/6470 [1:09:13<05:24,  1.40it/s]
2|6016|Loss: 0.7913562655448914:  93%|█████████▎| 6016/6470 [1:09:13<05:22,  1.41it/s]
2|6017|Loss: 0.7720323801040649:  93%|█████████▎| 6016/6470 [1:09:14<05:22,  1.41it/s]
2|6017|Loss: 0.7720323801040649:  93%|█████████▎| 6017/6470 [1:09:14<04:59,  1.51it/s]
2|6018|Loss: 0.9113235473632812:  93%|█████████▎| 6017/6470 [1:09:14<04:59,  1.51it/s]
2|6018|Loss: 0.9113235473632812:  93%|█████████▎| 6018/6470 [1:09:15<05:02,  1.49it/s]
2|6019|Loss: 0.6886248588562012:  93%|█████████▎| 6018/6470 [1:09:15<05:02,  1.49it/s]
2|6019|Loss: 0.6886248588562012:  93%|█████████▎| 6019/6470 [1:09:16<05:26,  1.38it/s]
2|6020|Loss: 0.8657718896865845:  93%|█████████▎| 6019/6470 [1:09:16<05:26,  1.38it/s]
2|6020|Loss: 0.8657718896865845:  93%|█████████▎| 6020/6470 [1:09:16<05:21,  1.40it/s]
2|6021|Loss: 0.6508557200431824:  93%|█████████▎| 6020/6470 [1:09:17<05:21,  1.40it/s]
2|6021|Loss: 0.6508557200431824:  93%|█████████▎| 6021/6470 [1:09:17<05:20,  1.40it/s]
2|6022|Loss: 0.9073790907859802:  93%|█████████▎| 6021/6470 [1:09:17<05:20,  1.40it/s]
2|6022|Loss: 0.9073790907859802:  93%|█████████▎| 6022/6470 [1:09:18<05:32,  1.35it/s]
2|6023|Loss: 0.8972569704055786:  93%|█████████▎| 6022/6470 [1:09:18<05:32,  1.35it/s]
2|6023|Loss: 0.8972569704055786:  93%|█████████▎| 6023/6470 [1:09:18<05:16,  1.41it/s]
2|6024|Loss: 0.8451322913169861:  93%|█████████▎| 6023/6470 [1:09:19<05:16,  1.41it/s]
2|6024|Loss: 0.8451322913169861:  93%|█████████▎| 6024/6470 [1:09:19<05:15,  1.41it/s]
2|6025|Loss: 0.9924226999282837:  93%|█████████▎| 6024/6470 [1:09:19<05:15,  1.41it/s]
2|6025|Loss: 0.9924226999282837:  93%|█████████▎| 6025/6470 [1:09:20<04:58,  1.49it/s]
2|6026|Loss: 1.0460113286972046:  93%|█████████▎| 6025/6470 [1:09:20<04:58,  1.49it/s]
2|6026|Loss: 1.0460113286972046:  93%|█████████▎| 6026/6470 [1:09:20<04:51,  1.52it/s]
2|6027|Loss: 0.9890682101249695:  93%|█████████▎| 6026/6470 [1:09:21<04:51,  1.52it/s]
2|6027|Loss: 0.9890682101249695:  93%|█████████▎| 6027/6470 [1:09:21<04:43,  1.56it/s]
2|6028|Loss: 0.9323307871818542:  93%|█████████▎| 6027/6470 [1:09:21<04:43,  1.56it/s]
2|6028|Loss: 0.9323307871818542:  93%|█████████▎| 6028/6470 [1:09:21<04:36,  1.60it/s]
2|6029|Loss: 0.925308883190155:  93%|█████████▎| 6028/6470 [1:09:22<04:36,  1.60it/s] 
2|6029|Loss: 0.925308883190155:  93%|█████████▎| 6029/6470 [1:09:22<04:29,  1.64it/s]
2|6030|Loss: 0.9246017932891846:  93%|█████████▎| 6029/6470 [1:09:22<04:29,  1.64it/s]
2|6030|Loss: 0.9246017932891846:  93%|█████████▎| 6030/6470 [1:09:23<04:28,  1.64it/s]
2|6031|Loss: 0.9035685062408447:  93%|█████████▎| 6030/6470 [1:09:23<04:28,  1.64it/s]
2|6031|Loss: 0.9035685062408447:  93%|█████████▎| 6031/6470 [1:09:23<04:51,  1.51it/s]
2|6032|Loss: 0.9789499044418335:  93%|█████████▎| 6031/6470 [1:09:24<04:51,  1.51it/s]
2|6032|Loss: 0.9789499044418335:  93%|█████████▎| 6032/6470 [1:09:24<05:18,  1.38it/s]
2|6033|Loss: 0.9272263646125793:  93%|█████████▎| 6032/6470 [1:09:25<05:18,  1.38it/s]
2|6033|Loss: 0.9272263646125793:  93%|█████████▎| 6033/6470 [1:09:25<05:24,  1.35it/s]
2|6034|Loss: 0.6963021159172058:  93%|█████████▎| 6033/6470 [1:09:26<05:24,  1.35it/s]
2|6034|Loss: 0.6963021159172058:  93%|█████████▎| 6034/6470 [1:09:26<05:31,  1.32it/s]
2|6035|Loss: 0.8123696446418762:  93%|█████████▎| 6034/6470 [1:09:26<05:31,  1.32it/s]
2|6035|Loss: 0.8123696446418762:  93%|█████████▎| 6035/6470 [1:09:27<05:37,  1.29it/s]
2|6036|Loss: 0.8658944964408875:  93%|█████████▎| 6035/6470 [1:09:27<05:37,  1.29it/s]
2|6036|Loss: 0.8658944964408875:  93%|█████████▎| 6036/6470 [1:09:27<05:23,  1.34it/s]
2|6037|Loss: 0.8839026689529419:  93%|█████████▎| 6036/6470 [1:09:28<05:23,  1.34it/s]
2|6037|Loss: 0.8839026689529419:  93%|█████████▎| 6037/6470 [1:09:28<05:09,  1.40it/s]
2|6038|Loss: 0.7713502645492554:  93%|█████████▎| 6037/6470 [1:09:28<05:09,  1.40it/s]
2|6038|Loss: 0.7713502645492554:  93%|█████████▎| 6038/6470 [1:09:29<05:24,  1.33it/s]
2|6039|Loss: 1.054641842842102:  93%|█████████▎| 6038/6470 [1:09:29<05:24,  1.33it/s] 
2|6039|Loss: 1.054641842842102:  93%|█████████▎| 6039/6470 [1:09:30<05:17,  1.36it/s]
2|6040|Loss: 0.9273437261581421:  93%|█████████▎| 6039/6470 [1:09:30<05:17,  1.36it/s]
2|6040|Loss: 0.9273437261581421:  93%|█████████▎| 6040/6470 [1:09:30<05:26,  1.32it/s]
2|6041|Loss: 0.9309229254722595:  93%|█████████▎| 6040/6470 [1:09:31<05:26,  1.32it/s]
2|6041|Loss: 0.9309229254722595:  93%|█████████▎| 6041/6470 [1:09:31<05:27,  1.31it/s]
2|6042|Loss: 0.9216049909591675:  93%|█████████▎| 6041/6470 [1:09:32<05:27,  1.31it/s]
2|6042|Loss: 0.9216049909591675:  93%|█████████▎| 6042/6470 [1:09:32<05:40,  1.26it/s]
2|6043|Loss: 0.8537712097167969:  93%|█████████▎| 6042/6470 [1:09:32<05:40,  1.26it/s]
2|6043|Loss: 0.8537712097167969:  93%|█████████▎| 6043/6470 [1:09:33<05:37,  1.26it/s]
2|6044|Loss: 1.0053234100341797:  93%|█████████▎| 6043/6470 [1:09:33<05:37,  1.26it/s]
2|6044|Loss: 1.0053234100341797:  93%|█████████▎| 6044/6470 [1:09:34<05:32,  1.28it/s]
2|6045|Loss: 0.8561816811561584:  93%|█████████▎| 6044/6470 [1:09:34<05:32,  1.28it/s]
2|6045|Loss: 0.8561816811561584:  93%|█████████▎| 6045/6470 [1:09:34<05:24,  1.31it/s]
2|6046|Loss: 0.8402726650238037:  93%|█████████▎| 6045/6470 [1:09:35<05:24,  1.31it/s]
2|6046|Loss: 0.8402726650238037:  93%|█████████▎| 6046/6470 [1:09:35<05:20,  1.32it/s]
2|6047|Loss: 0.8479331135749817:  93%|█████████▎| 6046/6470 [1:09:35<05:20,  1.32it/s]
2|6047|Loss: 0.8479331135749817:  93%|█████████▎| 6047/6470 [1:09:36<05:26,  1.30it/s]
2|6048|Loss: 0.7849153280258179:  93%|█████████▎| 6047/6470 [1:09:36<05:26,  1.30it/s]
2|6048|Loss: 0.7849153280258179:  93%|█████████▎| 6048/6470 [1:09:37<05:24,  1.30it/s]
2|6049|Loss: 0.725104570388794:  93%|█████████▎| 6048/6470 [1:09:37<05:24,  1.30it/s] 
2|6049|Loss: 0.725104570388794:  93%|█████████▎| 6049/6470 [1:09:37<05:20,  1.32it/s]
2|6050|Loss: 0.7757316827774048:  93%|█████████▎| 6049/6470 [1:09:38<05:20,  1.32it/s]
2|6050|Loss: 0.7757316827774048:  94%|█████████▎| 6050/6470 [1:09:38<05:15,  1.33it/s]
2|6051|Loss: 0.6732367873191833:  94%|█████████▎| 6050/6470 [1:09:38<05:15,  1.33it/s]
2|6051|Loss: 0.6732367873191833:  94%|█████████▎| 6051/6470 [1:09:39<05:13,  1.34it/s]
2|6052|Loss: 0.9809630513191223:  94%|█████████▎| 6051/6470 [1:09:39<05:13,  1.34it/s]
2|6052|Loss: 0.9809630513191223:  94%|█████████▎| 6052/6470 [1:09:39<04:53,  1.42it/s]
2|6053|Loss: 0.8493690490722656:  94%|█████████▎| 6052/6470 [1:09:40<04:53,  1.42it/s]
2|6053|Loss: 0.8493690490722656:  94%|█████████▎| 6053/6470 [1:09:40<04:53,  1.42it/s]
2|6054|Loss: 0.8450331091880798:  94%|█████████▎| 6053/6470 [1:09:40<04:53,  1.42it/s]
2|6054|Loss: 0.8450331091880798:  94%|█████████▎| 6054/6470 [1:09:41<04:41,  1.48it/s]
2|6055|Loss: 0.95827716588974:  94%|█████████▎| 6054/6470 [1:09:41<04:41,  1.48it/s]  
2|6055|Loss: 0.95827716588974:  94%|█████████▎| 6055/6470 [1:09:41<04:30,  1.54it/s]
2|6056|Loss: 0.8013570308685303:  94%|█████████▎| 6055/6470 [1:09:42<04:30,  1.54it/s]
2|6056|Loss: 0.8013570308685303:  94%|█████████▎| 6056/6470 [1:09:42<04:33,  1.51it/s]
2|6057|Loss: 0.760018527507782:  94%|█████████▎| 6056/6470 [1:09:42<04:33,  1.51it/s] 
2|6057|Loss: 0.760018527507782:  94%|█████████▎| 6057/6470 [1:09:43<04:42,  1.46it/s]
2|6058|Loss: 0.8815745115280151:  94%|█████████▎| 6057/6470 [1:09:43<04:42,  1.46it/s]
2|6058|Loss: 0.8815745115280151:  94%|█████████▎| 6058/6470 [1:09:43<04:49,  1.43it/s]
2|6059|Loss: 0.8526607155799866:  94%|█████████▎| 6058/6470 [1:09:44<04:49,  1.43it/s]
2|6059|Loss: 0.8526607155799866:  94%|█████████▎| 6059/6470 [1:09:44<04:43,  1.45it/s]
2|6060|Loss: 0.9169812202453613:  94%|█████████▎| 6059/6470 [1:09:44<04:43,  1.45it/s]
2|6060|Loss: 0.9169812202453613:  94%|█████████▎| 6060/6470 [1:09:45<04:18,  1.59it/s]
2|6061|Loss: 0.850701093673706:  94%|█████████▎| 6060/6470 [1:09:45<04:18,  1.59it/s] 
2|6061|Loss: 0.850701093673706:  94%|█████████▎| 6061/6470 [1:09:45<04:22,  1.56it/s]
2|6062|Loss: 0.7911685109138489:  94%|█████████▎| 6061/6470 [1:09:46<04:22,  1.56it/s]
2|6062|Loss: 0.7911685109138489:  94%|█████████▎| 6062/6470 [1:09:46<04:22,  1.55it/s]
2|6063|Loss: 0.8655825257301331:  94%|█████████▎| 6062/6470 [1:09:46<04:22,  1.55it/s]
2|6063|Loss: 0.8655825257301331:  94%|█████████▎| 6063/6470 [1:09:47<04:39,  1.45it/s]
2|6064|Loss: 0.7693740725517273:  94%|█████████▎| 6063/6470 [1:09:47<04:39,  1.45it/s]
2|6064|Loss: 0.7693740725517273:  94%|█████████▎| 6064/6470 [1:09:47<04:45,  1.42it/s]
2|6065|Loss: 0.844046950340271:  94%|█████████▎| 6064/6470 [1:09:48<04:45,  1.42it/s] 
2|6065|Loss: 0.844046950340271:  94%|█████████▎| 6065/6470 [1:09:48<04:37,  1.46it/s]
2|6066|Loss: 0.8939424157142639:  94%|█████████▎| 6065/6470 [1:09:48<04:37,  1.46it/s]
2|6066|Loss: 0.8939424157142639:  94%|█████████▍| 6066/6470 [1:09:49<04:30,  1.49it/s]
2|6067|Loss: 0.796291708946228:  94%|█████████▍| 6066/6470 [1:09:49<04:30,  1.49it/s] 
2|6067|Loss: 0.796291708946228:  94%|█████████▍| 6067/6470 [1:09:49<04:02,  1.66it/s]
2|6068|Loss: 0.8654844760894775:  94%|█████████▍| 6067/6470 [1:09:50<04:02,  1.66it/s]
2|6068|Loss: 0.8654844760894775:  94%|█████████▍| 6068/6470 [1:09:50<03:59,  1.68it/s]
2|6069|Loss: 0.8040380477905273:  94%|█████████▍| 6068/6470 [1:09:50<03:59,  1.68it/s]
2|6069|Loss: 0.8040380477905273:  94%|█████████▍| 6069/6470 [1:09:51<04:21,  1.53it/s]
2|6070|Loss: 0.9745211005210876:  94%|█████████▍| 6069/6470 [1:09:51<04:21,  1.53it/s]
2|6070|Loss: 0.9745211005210876:  94%|█████████▍| 6070/6470 [1:09:51<04:44,  1.41it/s]
2|6071|Loss: 0.8117719888687134:  94%|█████████▍| 6070/6470 [1:09:52<04:44,  1.41it/s]
2|6071|Loss: 0.8117719888687134:  94%|█████████▍| 6071/6470 [1:09:52<04:51,  1.37it/s]
2|6072|Loss: 0.9570868015289307:  94%|█████████▍| 6071/6470 [1:09:53<04:51,  1.37it/s]
2|6072|Loss: 0.9570868015289307:  94%|█████████▍| 6072/6470 [1:09:53<04:52,  1.36it/s]
2|6073|Loss: 0.8102838397026062:  94%|█████████▍| 6072/6470 [1:09:53<04:52,  1.36it/s]
2|6073|Loss: 0.8102838397026062:  94%|█████████▍| 6073/6470 [1:09:54<05:06,  1.29it/s]
2|6074|Loss: 0.7729052901268005:  94%|█████████▍| 6073/6470 [1:09:54<05:06,  1.29it/s]
2|6074|Loss: 0.7729052901268005:  94%|█████████▍| 6074/6470 [1:09:55<05:17,  1.25it/s]
2|6075|Loss: 0.8992329835891724:  94%|█████████▍| 6074/6470 [1:09:55<05:17,  1.25it/s]
2|6075|Loss: 0.8992329835891724:  94%|█████████▍| 6075/6470 [1:09:56<05:21,  1.23it/s]
2|6076|Loss: 0.8917073607444763:  94%|█████████▍| 6075/6470 [1:09:56<05:21,  1.23it/s]
2|6076|Loss: 0.8917073607444763:  94%|█████████▍| 6076/6470 [1:09:56<05:14,  1.25it/s]
2|6077|Loss: 1.024739384651184:  94%|█████████▍| 6076/6470 [1:09:57<05:14,  1.25it/s] 
2|6077|Loss: 1.024739384651184:  94%|█████████▍| 6077/6470 [1:09:57<04:55,  1.33it/s]
2|6078|Loss: 0.9297393560409546:  94%|█████████▍| 6077/6470 [1:09:57<04:55,  1.33it/s]
2|6078|Loss: 0.9297393560409546:  94%|█████████▍| 6078/6470 [1:09:58<04:38,  1.41it/s]
2|6079|Loss: 0.8582183718681335:  94%|█████████▍| 6078/6470 [1:09:58<04:38,  1.41it/s]
2|6079|Loss: 0.8582183718681335:  94%|█████████▍| 6079/6470 [1:09:58<04:38,  1.41it/s]
2|6080|Loss: 0.6796820163726807:  94%|█████████▍| 6079/6470 [1:09:59<04:38,  1.41it/s]
2|6080|Loss: 0.6796820163726807:  94%|█████████▍| 6080/6470 [1:09:59<04:29,  1.45it/s]
2|6081|Loss: 0.8761981129646301:  94%|█████████▍| 6080/6470 [1:09:59<04:29,  1.45it/s]
2|6081|Loss: 0.8761981129646301:  94%|█████████▍| 6081/6470 [1:10:00<04:21,  1.49it/s]
2|6082|Loss: 0.8082396984100342:  94%|█████████▍| 6081/6470 [1:10:00<04:21,  1.49it/s]
2|6082|Loss: 0.8082396984100342:  94%|█████████▍| 6082/6470 [1:10:00<04:02,  1.60it/s]
2|6083|Loss: 0.6924936175346375:  94%|█████████▍| 6082/6470 [1:10:00<04:02,  1.60it/s]
2|6083|Loss: 0.6924936175346375:  94%|█████████▍| 6083/6470 [1:10:00<03:34,  1.80it/s]
2|6084|Loss: 0.8855868577957153:  94%|█████████▍| 6083/6470 [1:10:01<03:34,  1.80it/s]
2|6084|Loss: 0.8855868577957153:  94%|█████████▍| 6084/6470 [1:10:01<03:36,  1.79it/s]
2|6085|Loss: 0.9176397323608398:  94%|█████████▍| 6084/6470 [1:10:01<03:36,  1.79it/s]
2|6085|Loss: 0.9176397323608398:  94%|█████████▍| 6085/6470 [1:10:02<03:52,  1.65it/s]
2|6086|Loss: 0.7842398881912231:  94%|█████████▍| 6085/6470 [1:10:02<03:52,  1.65it/s]
2|6086|Loss: 0.7842398881912231:  94%|█████████▍| 6086/6470 [1:10:02<03:52,  1.65it/s]
2|6087|Loss: 0.9073112607002258:  94%|█████████▍| 6086/6470 [1:10:03<03:52,  1.65it/s]
2|6087|Loss: 0.9073112607002258:  94%|█████████▍| 6087/6470 [1:10:03<03:37,  1.76it/s]
2|6088|Loss: 1.029693841934204:  94%|█████████▍| 6087/6470 [1:10:03<03:37,  1.76it/s] 
2|6088|Loss: 1.029693841934204:  94%|█████████▍| 6088/6470 [1:10:03<03:48,  1.67it/s]
2|6089|Loss: 0.8995826244354248:  94%|█████████▍| 6088/6470 [1:10:04<03:48,  1.67it/s]
2|6089|Loss: 0.8995826244354248:  94%|█████████▍| 6089/6470 [1:10:04<03:52,  1.64it/s]
2|6090|Loss: 0.8685463070869446:  94%|█████████▍| 6089/6470 [1:10:04<03:52,  1.64it/s]
2|6090|Loss: 0.8685463070869446:  94%|█████████▍| 6090/6470 [1:10:05<04:00,  1.58it/s]
2|6091|Loss: 0.9062564969062805:  94%|█████████▍| 6090/6470 [1:10:05<04:00,  1.58it/s]
2|6091|Loss: 0.9062564969062805:  94%|█████████▍| 6091/6470 [1:10:05<03:51,  1.64it/s]
2|6092|Loss: 0.8144153356552124:  94%|█████████▍| 6091/6470 [1:10:06<03:51,  1.64it/s]
2|6092|Loss: 0.8144153356552124:  94%|█████████▍| 6092/6470 [1:10:06<03:56,  1.60it/s]
2|6093|Loss: 0.9164869785308838:  94%|█████████▍| 6092/6470 [1:10:06<03:56,  1.60it/s]
2|6093|Loss: 0.9164869785308838:  94%|█████████▍| 6093/6470 [1:10:07<03:41,  1.70it/s]
2|6094|Loss: 1.0336062908172607:  94%|█████████▍| 6093/6470 [1:10:07<03:41,  1.70it/s]
2|6094|Loss: 1.0336062908172607:  94%|█████████▍| 6094/6470 [1:10:07<03:56,  1.59it/s]
2|6095|Loss: 1.0496904850006104:  94%|█████████▍| 6094/6470 [1:10:08<03:56,  1.59it/s]
2|6095|Loss: 1.0496904850006104:  94%|█████████▍| 6095/6470 [1:10:08<03:51,  1.62it/s]
2|6096|Loss: 0.8031728267669678:  94%|█████████▍| 6095/6470 [1:10:08<03:51,  1.62it/s]
2|6096|Loss: 0.8031728267669678:  94%|█████████▍| 6096/6470 [1:10:08<03:56,  1.58it/s]
2|6097|Loss: 0.8067263960838318:  94%|█████████▍| 6096/6470 [1:10:09<03:56,  1.58it/s]
2|6097|Loss: 0.8067263960838318:  94%|█████████▍| 6097/6470 [1:10:09<04:19,  1.44it/s]
2|6098|Loss: 0.9342668652534485:  94%|█████████▍| 6097/6470 [1:10:10<04:19,  1.44it/s]
2|6098|Loss: 0.9342668652534485:  94%|█████████▍| 6098/6470 [1:10:10<04:13,  1.47it/s]
2|6099|Loss: 0.7174439430236816:  94%|█████████▍| 6098/6470 [1:10:10<04:13,  1.47it/s]
2|6099|Loss: 0.7174439430236816:  94%|█████████▍| 6099/6470 [1:10:11<04:12,  1.47it/s]
2|6100|Loss: 0.8467333912849426:  94%|█████████▍| 6099/6470 [1:10:11<04:12,  1.47it/s]
2|6100|Loss: 0.8467333912849426:  94%|█████████▍| 6100/6470 [1:10:11<04:05,  1.50it/s]
2|6101|Loss: 0.8282253742218018:  94%|█████████▍| 6100/6470 [1:10:12<04:05,  1.50it/s]
2|6101|Loss: 0.8282253742218018:  94%|█████████▍| 6101/6470 [1:10:12<03:45,  1.64it/s]
2|6102|Loss: 0.857880711555481:  94%|█████████▍| 6101/6470 [1:10:12<03:45,  1.64it/s] 
2|6102|Loss: 0.857880711555481:  94%|█████████▍| 6102/6470 [1:10:12<03:48,  1.61it/s]
2|6103|Loss: 0.8734174370765686:  94%|█████████▍| 6102/6470 [1:10:13<03:48,  1.61it/s]
2|6103|Loss: 0.8734174370765686:  94%|█████████▍| 6103/6470 [1:10:13<03:59,  1.53it/s]
2|6104|Loss: 0.796374499797821:  94%|█████████▍| 6103/6470 [1:10:14<03:59,  1.53it/s] 
2|6104|Loss: 0.796374499797821:  94%|█████████▍| 6104/6470 [1:10:14<03:54,  1.56it/s]
2|6105|Loss: 1.015411376953125:  94%|█████████▍| 6104/6470 [1:10:14<03:54,  1.56it/s]
2|6105|Loss: 1.015411376953125:  94%|█████████▍| 6105/6470 [1:10:15<04:17,  1.42it/s]
2|6106|Loss: 0.9228284358978271:  94%|█████████▍| 6105/6470 [1:10:15<04:17,  1.42it/s]
2|6106|Loss: 0.9228284358978271:  94%|█████████▍| 6106/6470 [1:10:15<04:13,  1.43it/s]
2|6107|Loss: 0.8659087419509888:  94%|█████████▍| 6106/6470 [1:10:16<04:13,  1.43it/s]
2|6107|Loss: 0.8659087419509888:  94%|█████████▍| 6107/6470 [1:10:16<04:04,  1.49it/s]
2|6108|Loss: 0.9530439972877502:  94%|█████████▍| 6107/6470 [1:10:16<04:04,  1.49it/s]
2|6108|Loss: 0.9530439972877502:  94%|█████████▍| 6108/6470 [1:10:17<04:07,  1.46it/s]
2|6109|Loss: 0.8990936279296875:  94%|█████████▍| 6108/6470 [1:10:17<04:07,  1.46it/s]
2|6109|Loss: 0.8990936279296875:  94%|█████████▍| 6109/6470 [1:10:17<04:29,  1.34it/s]
2|6110|Loss: 0.8653246760368347:  94%|█████████▍| 6109/6470 [1:10:18<04:29,  1.34it/s]
2|6110|Loss: 0.8653246760368347:  94%|█████████▍| 6110/6470 [1:10:18<04:42,  1.27it/s]
2|6111|Loss: 0.7567399144172668:  94%|█████████▍| 6110/6470 [1:10:19<04:42,  1.27it/s]
2|6111|Loss: 0.7567399144172668:  94%|█████████▍| 6111/6470 [1:10:19<04:52,  1.23it/s]
2|6112|Loss: 1.0020098686218262:  94%|█████████▍| 6111/6470 [1:10:20<04:52,  1.23it/s]
2|6112|Loss: 1.0020098686218262:  94%|█████████▍| 6112/6470 [1:10:20<04:45,  1.26it/s]
2|6113|Loss: 0.8254290223121643:  94%|█████████▍| 6112/6470 [1:10:20<04:45,  1.26it/s]
2|6113|Loss: 0.8254290223121643:  94%|█████████▍| 6113/6470 [1:10:21<04:47,  1.24it/s]
2|6114|Loss: 0.9319949150085449:  94%|█████████▍| 6113/6470 [1:10:21<04:47,  1.24it/s]
2|6114|Loss: 0.9319949150085449:  94%|█████████▍| 6114/6470 [1:10:22<04:35,  1.29it/s]
2|6115|Loss: 0.9459221959114075:  94%|█████████▍| 6114/6470 [1:10:22<04:35,  1.29it/s]
2|6115|Loss: 0.9459221959114075:  95%|█████████▍| 6115/6470 [1:10:22<04:26,  1.33it/s]
2|6116|Loss: 0.8468617796897888:  95%|█████████▍| 6115/6470 [1:10:23<04:26,  1.33it/s]
2|6116|Loss: 0.8468617796897888:  95%|█████████▍| 6116/6470 [1:10:23<04:07,  1.43it/s]
2|6117|Loss: 0.8247711062431335:  95%|█████████▍| 6116/6470 [1:10:23<04:07,  1.43it/s]
2|6117|Loss: 0.8247711062431335:  95%|█████████▍| 6117/6470 [1:10:24<04:07,  1.43it/s]
2|6118|Loss: 1.0135663747787476:  95%|█████████▍| 6117/6470 [1:10:24<04:07,  1.43it/s]
2|6118|Loss: 1.0135663747787476:  95%|█████████▍| 6118/6470 [1:10:24<04:19,  1.36it/s]
2|6119|Loss: 0.8402660489082336:  95%|█████████▍| 6118/6470 [1:10:25<04:19,  1.36it/s]
2|6119|Loss: 0.8402660489082336:  95%|█████████▍| 6119/6470 [1:10:25<04:01,  1.45it/s]
2|6120|Loss: 0.9157389998435974:  95%|█████████▍| 6119/6470 [1:10:25<04:01,  1.45it/s]
2|6120|Loss: 0.9157389998435974:  95%|█████████▍| 6120/6470 [1:10:25<03:43,  1.56it/s]
2|6121|Loss: 0.8437619805335999:  95%|█████████▍| 6120/6470 [1:10:26<03:43,  1.56it/s]
2|6121|Loss: 0.8437619805335999:  95%|█████████▍| 6121/6470 [1:10:26<03:50,  1.52it/s]
2|6122|Loss: 0.9796464443206787:  95%|█████████▍| 6121/6470 [1:10:27<03:50,  1.52it/s]
2|6122|Loss: 0.9796464443206787:  95%|█████████▍| 6122/6470 [1:10:27<03:50,  1.51it/s]
2|6123|Loss: 0.9265627861022949:  95%|█████████▍| 6122/6470 [1:10:27<03:50,  1.51it/s]
2|6123|Loss: 0.9265627861022949:  95%|█████████▍| 6123/6470 [1:10:28<04:00,  1.44it/s]
2|6124|Loss: 0.8948764801025391:  95%|█████████▍| 6123/6470 [1:10:28<04:00,  1.44it/s]
2|6124|Loss: 0.8948764801025391:  95%|█████████▍| 6124/6470 [1:10:28<04:03,  1.42it/s]
2|6125|Loss: 0.840272843837738:  95%|█████████▍| 6124/6470 [1:10:29<04:03,  1.42it/s] 
2|6125|Loss: 0.840272843837738:  95%|█████████▍| 6125/6470 [1:10:29<03:41,  1.56it/s]
2|6126|Loss: 0.9962071776390076:  95%|█████████▍| 6125/6470 [1:10:29<03:41,  1.56it/s]
2|6126|Loss: 0.9962071776390076:  95%|█████████▍| 6126/6470 [1:10:29<03:34,  1.60it/s]
2|6127|Loss: 0.8508254289627075:  95%|█████████▍| 6126/6470 [1:10:30<03:34,  1.60it/s]
2|6127|Loss: 0.8508254289627075:  95%|█████████▍| 6127/6470 [1:10:30<03:34,  1.60it/s]
2|6128|Loss: 0.8616678714752197:  95%|█████████▍| 6127/6470 [1:10:30<03:34,  1.60it/s]
2|6128|Loss: 0.8616678714752197:  95%|█████████▍| 6128/6470 [1:10:31<03:47,  1.50it/s]
2|6129|Loss: 0.9418651461601257:  95%|█████████▍| 6128/6470 [1:10:31<03:47,  1.50it/s]
2|6129|Loss: 0.9418651461601257:  95%|█████████▍| 6129/6470 [1:10:32<03:57,  1.44it/s]
2|6130|Loss: 0.8398658633232117:  95%|█████████▍| 6129/6470 [1:10:32<03:57,  1.44it/s]
2|6130|Loss: 0.8398658633232117:  95%|█████████▍| 6130/6470 [1:10:32<03:58,  1.42it/s]
2|6131|Loss: 0.9468361139297485:  95%|█████████▍| 6130/6470 [1:10:33<03:58,  1.42it/s]
2|6131|Loss: 0.9468361139297485:  95%|█████████▍| 6131/6470 [1:10:33<04:05,  1.38it/s]
2|6132|Loss: 0.8004176020622253:  95%|█████████▍| 6131/6470 [1:10:33<04:05,  1.38it/s]
2|6132|Loss: 0.8004176020622253:  95%|█████████▍| 6132/6470 [1:10:34<04:08,  1.36it/s]
2|6133|Loss: 0.8477135300636292:  95%|█████████▍| 6132/6470 [1:10:34<04:08,  1.36it/s]
2|6133|Loss: 0.8477135300636292:  95%|█████████▍| 6133/6470 [1:10:34<03:59,  1.41it/s]
2|6134|Loss: 0.8117060661315918:  95%|█████████▍| 6133/6470 [1:10:35<03:59,  1.41it/s]
2|6134|Loss: 0.8117060661315918:  95%|█████████▍| 6134/6470 [1:10:35<03:58,  1.41it/s]
2|6135|Loss: 0.7384775876998901:  95%|█████████▍| 6134/6470 [1:10:36<03:58,  1.41it/s]
2|6135|Loss: 0.7384775876998901:  95%|█████████▍| 6135/6470 [1:10:36<03:56,  1.41it/s]
2|6136|Loss: 0.8079219460487366:  95%|█████████▍| 6135/6470 [1:10:36<03:56,  1.41it/s]
2|6136|Loss: 0.8079219460487366:  95%|█████████▍| 6136/6470 [1:10:36<03:49,  1.46it/s]
2|6137|Loss: 0.8746176958084106:  95%|█████████▍| 6136/6470 [1:10:37<03:49,  1.46it/s]
2|6137|Loss: 0.8746176958084106:  95%|█████████▍| 6137/6470 [1:10:37<03:32,  1.56it/s]
2|6138|Loss: 0.8996448516845703:  95%|█████████▍| 6137/6470 [1:10:37<03:32,  1.56it/s]
2|6138|Loss: 0.8996448516845703:  95%|█████████▍| 6138/6470 [1:10:38<03:19,  1.67it/s]
2|6139|Loss: 0.833674430847168:  95%|█████████▍| 6138/6470 [1:10:38<03:19,  1.67it/s] 
2|6139|Loss: 0.833674430847168:  95%|█████████▍| 6139/6470 [1:10:38<03:34,  1.55it/s]
2|6140|Loss: 0.8928514122962952:  95%|█████████▍| 6139/6470 [1:10:39<03:34,  1.55it/s]
2|6140|Loss: 0.8928514122962952:  95%|█████████▍| 6140/6470 [1:10:39<03:36,  1.52it/s]
2|6141|Loss: 0.8270554542541504:  95%|█████████▍| 6140/6470 [1:10:39<03:36,  1.52it/s]
2|6141|Loss: 0.8270554542541504:  95%|█████████▍| 6141/6470 [1:10:40<03:39,  1.50it/s]
2|6142|Loss: 0.7666683793067932:  95%|█████████▍| 6141/6470 [1:10:40<03:39,  1.50it/s]
2|6142|Loss: 0.7666683793067932:  95%|█████████▍| 6142/6470 [1:10:40<03:32,  1.54it/s]
2|6143|Loss: 0.9684593677520752:  95%|█████████▍| 6142/6470 [1:10:41<03:32,  1.54it/s]
2|6143|Loss: 0.9684593677520752:  95%|█████████▍| 6143/6470 [1:10:41<03:42,  1.47it/s]
2|6144|Loss: 0.9244118928909302:  95%|█████████▍| 6143/6470 [1:10:41<03:42,  1.47it/s]
2|6144|Loss: 0.9244118928909302:  95%|█████████▍| 6144/6470 [1:10:42<03:50,  1.42it/s]
2|6145|Loss: 0.7888540625572205:  95%|█████████▍| 6144/6470 [1:10:42<03:50,  1.42it/s]
2|6145|Loss: 0.7888540625572205:  95%|█████████▍| 6145/6470 [1:10:43<03:58,  1.36it/s]
2|6146|Loss: 0.7539469003677368:  95%|█████████▍| 6145/6470 [1:10:43<03:58,  1.36it/s]
2|6146|Loss: 0.7539469003677368:  95%|█████████▍| 6146/6470 [1:10:43<03:36,  1.49it/s]
2|6147|Loss: 0.9413543343544006:  95%|█████████▍| 6146/6470 [1:10:43<03:36,  1.49it/s]
2|6147|Loss: 0.9413543343544006:  95%|█████████▌| 6147/6470 [1:10:44<03:40,  1.46it/s]
2|6148|Loss: 0.8861480951309204:  95%|█████████▌| 6147/6470 [1:10:44<03:40,  1.46it/s]
2|6148|Loss: 0.8861480951309204:  95%|█████████▌| 6148/6470 [1:10:45<03:46,  1.42it/s]
2|6149|Loss: 0.804502546787262:  95%|█████████▌| 6148/6470 [1:10:45<03:46,  1.42it/s] 
2|6149|Loss: 0.804502546787262:  95%|█████████▌| 6149/6470 [1:10:45<03:49,  1.40it/s]
2|6150|Loss: 0.8603052496910095:  95%|█████████▌| 6149/6470 [1:10:46<03:49,  1.40it/s]
2|6150|Loss: 0.8603052496910095:  95%|█████████▌| 6150/6470 [1:10:46<03:38,  1.47it/s]
2|6151|Loss: 0.9148353338241577:  95%|█████████▌| 6150/6470 [1:10:46<03:38,  1.47it/s]
2|6151|Loss: 0.9148353338241577:  95%|█████████▌| 6151/6470 [1:10:47<03:32,  1.50it/s]
2|6152|Loss: 0.8708182573318481:  95%|█████████▌| 6151/6470 [1:10:47<03:32,  1.50it/s]
2|6152|Loss: 0.8708182573318481:  95%|█████████▌| 6152/6470 [1:10:47<03:18,  1.61it/s]
2|6153|Loss: 0.9038119316101074:  95%|█████████▌| 6152/6470 [1:10:47<03:18,  1.61it/s]
2|6153|Loss: 0.9038119316101074:  95%|█████████▌| 6153/6470 [1:10:48<03:22,  1.56it/s]
2|6154|Loss: 0.9322562217712402:  95%|█████████▌| 6153/6470 [1:10:48<03:22,  1.56it/s]
2|6154|Loss: 0.9322562217712402:  95%|█████████▌| 6154/6470 [1:10:49<03:42,  1.42it/s]
2|6155|Loss: 0.790032684803009:  95%|█████████▌| 6154/6470 [1:10:49<03:42,  1.42it/s] 
2|6155|Loss: 0.790032684803009:  95%|█████████▌| 6155/6470 [1:10:49<03:39,  1.43it/s]
2|6156|Loss: 0.7781760096549988:  95%|█████████▌| 6155/6470 [1:10:50<03:39,  1.43it/s]
2|6156|Loss: 0.7781760096549988:  95%|█████████▌| 6156/6470 [1:10:50<03:39,  1.43it/s]
2|6157|Loss: 0.8282677531242371:  95%|█████████▌| 6156/6470 [1:10:50<03:39,  1.43it/s]
2|6157|Loss: 0.8282677531242371:  95%|█████████▌| 6157/6470 [1:10:51<03:36,  1.45it/s]
2|6158|Loss: 1.0598803758621216:  95%|█████████▌| 6157/6470 [1:10:51<03:36,  1.45it/s]
2|6158|Loss: 1.0598803758621216:  95%|█████████▌| 6158/6470 [1:10:51<03:43,  1.40it/s]
2|6159|Loss: 0.9418751001358032:  95%|█████████▌| 6158/6470 [1:10:52<03:43,  1.40it/s]
2|6159|Loss: 0.9418751001358032:  95%|█████████▌| 6159/6470 [1:10:52<03:55,  1.32it/s]
2|6160|Loss: 0.8658784627914429:  95%|█████████▌| 6159/6470 [1:10:53<03:55,  1.32it/s]
2|6160|Loss: 0.8658784627914429:  95%|█████████▌| 6160/6470 [1:10:53<04:04,  1.27it/s]
2|6161|Loss: 0.8524947762489319:  95%|█████████▌| 6160/6470 [1:10:54<04:04,  1.27it/s]
2|6161|Loss: 0.8524947762489319:  95%|█████████▌| 6161/6470 [1:10:54<04:02,  1.28it/s]
2|6162|Loss: 0.9900112152099609:  95%|█████████▌| 6161/6470 [1:10:54<04:02,  1.28it/s]
2|6162|Loss: 0.9900112152099609:  95%|█████████▌| 6162/6470 [1:10:54<03:40,  1.40it/s]
2|6163|Loss: 0.7202823162078857:  95%|█████████▌| 6162/6470 [1:10:55<03:40,  1.40it/s]
2|6163|Loss: 0.7202823162078857:  95%|█████████▌| 6163/6470 [1:10:55<03:16,  1.56it/s]
2|6164|Loss: 0.7326158881187439:  95%|█████████▌| 6163/6470 [1:10:55<03:16,  1.56it/s]
2|6164|Loss: 0.7326158881187439:  95%|█████████▌| 6164/6470 [1:10:56<03:25,  1.49it/s]
2|6165|Loss: 0.7416338324546814:  95%|█████████▌| 6164/6470 [1:10:56<03:25,  1.49it/s]
2|6165|Loss: 0.7416338324546814:  95%|█████████▌| 6165/6470 [1:10:57<03:49,  1.33it/s]
2|6166|Loss: 1.016405701637268:  95%|█████████▌| 6165/6470 [1:10:57<03:49,  1.33it/s] 
2|6166|Loss: 1.016405701637268:  95%|█████████▌| 6166/6470 [1:10:57<03:59,  1.27it/s]
2|6167|Loss: 0.8616037964820862:  95%|█████████▌| 6166/6470 [1:10:58<03:59,  1.27it/s]
2|6167|Loss: 0.8616037964820862:  95%|█████████▌| 6167/6470 [1:10:58<03:39,  1.38it/s]
2|6168|Loss: 0.9376164078712463:  95%|█████████▌| 6167/6470 [1:10:58<03:39,  1.38it/s]
2|6168|Loss: 0.9376164078712463:  95%|█████████▌| 6168/6470 [1:10:59<03:35,  1.40it/s]
2|6169|Loss: 1.0198791027069092:  95%|█████████▌| 6168/6470 [1:10:59<03:35,  1.40it/s]
2|6169|Loss: 1.0198791027069092:  95%|█████████▌| 6169/6470 [1:10:59<03:33,  1.41it/s]
2|6170|Loss: 0.7927513718605042:  95%|█████████▌| 6169/6470 [1:11:00<03:33,  1.41it/s]
2|6170|Loss: 0.7927513718605042:  95%|█████████▌| 6170/6470 [1:11:00<03:34,  1.40it/s]
2|6171|Loss: 0.8046396374702454:  95%|█████████▌| 6170/6470 [1:11:01<03:34,  1.40it/s]
2|6171|Loss: 0.8046396374702454:  95%|█████████▌| 6171/6470 [1:11:01<03:35,  1.39it/s]
2|6172|Loss: 0.9834700226783752:  95%|█████████▌| 6171/6470 [1:11:01<03:35,  1.39it/s]
2|6172|Loss: 0.9834700226783752:  95%|█████████▌| 6172/6470 [1:11:02<03:24,  1.46it/s]
2|6173|Loss: 1.006515622138977:  95%|█████████▌| 6172/6470 [1:11:02<03:24,  1.46it/s] 
2|6173|Loss: 1.006515622138977:  95%|█████████▌| 6173/6470 [1:11:02<03:26,  1.44it/s]
2|6174|Loss: 0.7633753418922424:  95%|█████████▌| 6173/6470 [1:11:03<03:26,  1.44it/s]
2|6174|Loss: 0.7633753418922424:  95%|█████████▌| 6174/6470 [1:11:03<03:29,  1.41it/s]
2|6175|Loss: 0.7602192759513855:  95%|█████████▌| 6174/6470 [1:11:03<03:29,  1.41it/s]
2|6175|Loss: 0.7602192759513855:  95%|█████████▌| 6175/6470 [1:11:04<03:35,  1.37it/s]
2|6176|Loss: 0.9297407269477844:  95%|█████████▌| 6175/6470 [1:11:04<03:35,  1.37it/s]
2|6176|Loss: 0.9297407269477844:  95%|█████████▌| 6176/6470 [1:11:05<03:39,  1.34it/s]
2|6177|Loss: 0.7374225854873657:  95%|█████████▌| 6176/6470 [1:11:05<03:39,  1.34it/s]
2|6177|Loss: 0.7374225854873657:  95%|█████████▌| 6177/6470 [1:11:05<03:26,  1.42it/s]
2|6178|Loss: 0.9429295063018799:  95%|█████████▌| 6177/6470 [1:11:06<03:26,  1.42it/s]
2|6178|Loss: 0.9429295063018799:  95%|█████████▌| 6178/6470 [1:11:06<03:25,  1.42it/s]
2|6179|Loss: 0.9594689011573792:  95%|█████████▌| 6178/6470 [1:11:06<03:25,  1.42it/s]
2|6179|Loss: 0.9594689011573792:  96%|█████████▌| 6179/6470 [1:11:07<03:25,  1.41it/s]
2|6180|Loss: 0.8732566237449646:  96%|█████████▌| 6179/6470 [1:11:07<03:25,  1.41it/s]
2|6180|Loss: 0.8732566237449646:  96%|█████████▌| 6180/6470 [1:11:07<03:39,  1.32it/s]
2|6181|Loss: 1.0668915510177612:  96%|█████████▌| 6180/6470 [1:11:08<03:39,  1.32it/s]
2|6181|Loss: 1.0668915510177612:  96%|█████████▌| 6181/6470 [1:11:08<03:50,  1.26it/s]
2|6182|Loss: 0.775294303894043:  96%|█████████▌| 6181/6470 [1:11:09<03:50,  1.26it/s] 
2|6182|Loss: 0.775294303894043:  96%|█████████▌| 6182/6470 [1:11:09<03:38,  1.32it/s]
2|6183|Loss: 0.9142999053001404:  96%|█████████▌| 6182/6470 [1:11:09<03:38,  1.32it/s]
2|6183|Loss: 0.9142999053001404:  96%|█████████▌| 6183/6470 [1:11:10<03:27,  1.38it/s]
2|6184|Loss: 0.7963774800300598:  96%|█████████▌| 6183/6470 [1:11:10<03:27,  1.38it/s]
2|6184|Loss: 0.7963774800300598:  96%|█████████▌| 6184/6470 [1:11:10<03:09,  1.51it/s]
2|6185|Loss: 0.8734928965568542:  96%|█████████▌| 6184/6470 [1:11:11<03:09,  1.51it/s]
2|6185|Loss: 0.8734928965568542:  96%|█████████▌| 6185/6470 [1:11:11<03:14,  1.47it/s]
2|6186|Loss: 0.8316885828971863:  96%|█████████▌| 6185/6470 [1:11:11<03:14,  1.47it/s]
2|6186|Loss: 0.8316885828971863:  96%|█████████▌| 6186/6470 [1:11:12<03:11,  1.48it/s]
2|6187|Loss: 0.9650349020957947:  96%|█████████▌| 6186/6470 [1:11:12<03:11,  1.48it/s]
2|6187|Loss: 0.9650349020957947:  96%|█████████▌| 6187/6470 [1:11:12<03:14,  1.45it/s]
2|6188|Loss: 0.7311307191848755:  96%|█████████▌| 6187/6470 [1:11:13<03:14,  1.45it/s]
2|6188|Loss: 0.7311307191848755:  96%|█████████▌| 6188/6470 [1:11:13<02:59,  1.58it/s]
2|6189|Loss: 0.8351417183876038:  96%|█████████▌| 6188/6470 [1:11:13<02:59,  1.58it/s]
2|6189|Loss: 0.8351417183876038:  96%|█████████▌| 6189/6470 [1:11:13<02:55,  1.60it/s]
2|6190|Loss: 0.7421047687530518:  96%|█████████▌| 6189/6470 [1:11:14<02:55,  1.60it/s]
2|6190|Loss: 0.7421047687530518:  96%|█████████▌| 6190/6470 [1:11:14<03:05,  1.51it/s]
2|6191|Loss: 0.8820061087608337:  96%|█████████▌| 6190/6470 [1:11:14<03:05,  1.51it/s]
2|6191|Loss: 0.8820061087608337:  96%|█████████▌| 6191/6470 [1:11:15<02:55,  1.59it/s]
2|6192|Loss: 0.8239338994026184:  96%|█████████▌| 6191/6470 [1:11:15<02:55,  1.59it/s]
2|6192|Loss: 0.8239338994026184:  96%|█████████▌| 6192/6470 [1:11:15<03:07,  1.48it/s]
2|6193|Loss: 0.8323874473571777:  96%|█████████▌| 6192/6470 [1:11:16<03:07,  1.48it/s]
2|6193|Loss: 0.8323874473571777:  96%|█████████▌| 6193/6470 [1:11:16<03:18,  1.40it/s]
2|6194|Loss: 0.8819456100463867:  96%|█████████▌| 6193/6470 [1:11:17<03:18,  1.40it/s]
2|6194|Loss: 0.8819456100463867:  96%|█████████▌| 6194/6470 [1:11:17<03:24,  1.35it/s]
2|6195|Loss: 0.8212618231773376:  96%|█████████▌| 6194/6470 [1:11:18<03:24,  1.35it/s]
2|6195|Loss: 0.8212618231773376:  96%|█████████▌| 6195/6470 [1:11:18<03:31,  1.30it/s]
2|6196|Loss: 0.7793806195259094:  96%|█████████▌| 6195/6470 [1:11:18<03:31,  1.30it/s]
2|6196|Loss: 0.7793806195259094:  96%|█████████▌| 6196/6470 [1:11:19<03:27,  1.32it/s]
2|6197|Loss: 0.899364709854126:  96%|█████████▌| 6196/6470 [1:11:19<03:27,  1.32it/s] 
2|6197|Loss: 0.899364709854126:  96%|█████████▌| 6197/6470 [1:11:19<03:17,  1.38it/s]
2|6198|Loss: 0.9673914313316345:  96%|█████████▌| 6197/6470 [1:11:20<03:17,  1.38it/s]
2|6198|Loss: 0.9673914313316345:  96%|█████████▌| 6198/6470 [1:11:20<03:07,  1.45it/s]
2|6199|Loss: 1.0145539045333862:  96%|█████████▌| 6198/6470 [1:11:20<03:07,  1.45it/s]
2|6199|Loss: 1.0145539045333862:  96%|█████████▌| 6199/6470 [1:11:21<03:14,  1.39it/s]
2|6200|Loss: 0.8449721932411194:  96%|█████████▌| 6199/6470 [1:11:21<03:14,  1.39it/s]
2|6200|Loss: 0.8449721932411194:  96%|█████████▌| 6200/6470 [1:11:21<03:14,  1.39it/s]
2|6201|Loss: 0.8067160248756409:  96%|█████████▌| 6200/6470 [1:11:22<03:14,  1.39it/s]
2|6201|Loss: 0.8067160248756409:  96%|█████████▌| 6201/6470 [1:11:22<03:15,  1.37it/s]
2|6202|Loss: 0.8569220304489136:  96%|█████████▌| 6201/6470 [1:11:23<03:15,  1.37it/s]
2|6202|Loss: 0.8569220304489136:  96%|█████████▌| 6202/6470 [1:11:23<03:08,  1.42it/s]
2|6203|Loss: 0.8213103413581848:  96%|█████████▌| 6202/6470 [1:11:23<03:08,  1.42it/s]
2|6203|Loss: 0.8213103413581848:  96%|█████████▌| 6203/6470 [1:11:24<03:14,  1.37it/s]
2|6204|Loss: 0.8036778569221497:  96%|█████████▌| 6203/6470 [1:11:24<03:14,  1.37it/s]
2|6204|Loss: 0.8036778569221497:  96%|█████████▌| 6204/6470 [1:11:24<03:05,  1.43it/s]
2|6205|Loss: 1.0157670974731445:  96%|█████████▌| 6204/6470 [1:11:25<03:05,  1.43it/s]
2|6205|Loss: 1.0157670974731445:  96%|█████████▌| 6205/6470 [1:11:25<02:59,  1.48it/s]
2|6206|Loss: 0.7262340188026428:  96%|█████████▌| 6205/6470 [1:11:25<02:59,  1.48it/s]
2|6206|Loss: 0.7262340188026428:  96%|█████████▌| 6206/6470 [1:11:25<02:53,  1.53it/s]
2|6207|Loss: 0.8898452520370483:  96%|█████████▌| 6206/6470 [1:11:26<02:53,  1.53it/s]
2|6207|Loss: 0.8898452520370483:  96%|█████████▌| 6207/6470 [1:11:26<02:46,  1.58it/s]
2|6208|Loss: 1.0138429403305054:  96%|█████████▌| 6207/6470 [1:11:26<02:46,  1.58it/s]
2|6208|Loss: 1.0138429403305054:  96%|█████████▌| 6208/6470 [1:11:27<02:45,  1.58it/s]
2|6209|Loss: 0.9560397863388062:  96%|█████████▌| 6208/6470 [1:11:27<02:45,  1.58it/s]
2|6209|Loss: 0.9560397863388062:  96%|█████████▌| 6209/6470 [1:11:27<02:44,  1.59it/s]
2|6210|Loss: 0.8210716247558594:  96%|█████████▌| 6209/6470 [1:11:28<02:44,  1.59it/s]
2|6210|Loss: 0.8210716247558594:  96%|█████████▌| 6210/6470 [1:11:28<02:57,  1.47it/s]
2|6211|Loss: 0.9058736562728882:  96%|█████████▌| 6210/6470 [1:11:28<02:57,  1.47it/s]
2|6211|Loss: 0.9058736562728882:  96%|█████████▌| 6211/6470 [1:11:29<02:45,  1.57it/s]
2|6212|Loss: 0.9976361393928528:  96%|█████████▌| 6211/6470 [1:11:29<02:45,  1.57it/s]
2|6212|Loss: 0.9976361393928528:  96%|█████████▌| 6212/6470 [1:11:29<02:46,  1.55it/s]
2|6213|Loss: 1.0113435983657837:  96%|█████████▌| 6212/6470 [1:11:30<02:46,  1.55it/s]
2|6213|Loss: 1.0113435983657837:  96%|█████████▌| 6213/6470 [1:11:30<02:52,  1.49it/s]
2|6214|Loss: 0.756261944770813:  96%|█████████▌| 6213/6470 [1:11:30<02:52,  1.49it/s] 
2|6214|Loss: 0.756261944770813:  96%|█████████▌| 6214/6470 [1:11:31<02:57,  1.45it/s]
2|6215|Loss: 0.8345505595207214:  96%|█████████▌| 6214/6470 [1:11:31<02:57,  1.45it/s]
2|6215|Loss: 0.8345505595207214:  96%|█████████▌| 6215/6470 [1:11:32<03:05,  1.37it/s]
2|6216|Loss: 0.9050931930541992:  96%|█████████▌| 6215/6470 [1:11:32<03:05,  1.37it/s]
2|6216|Loss: 0.9050931930541992:  96%|█████████▌| 6216/6470 [1:11:32<03:07,  1.35it/s]
2|6217|Loss: 1.0058549642562866:  96%|█████████▌| 6216/6470 [1:11:33<03:07,  1.35it/s]
2|6217|Loss: 1.0058549642562866:  96%|█████████▌| 6217/6470 [1:11:33<02:57,  1.43it/s]
2|6218|Loss: 0.8354438543319702:  96%|█████████▌| 6217/6470 [1:11:33<02:57,  1.43it/s]
2|6218|Loss: 0.8354438543319702:  96%|█████████▌| 6218/6470 [1:11:34<02:57,  1.42it/s]
2|6219|Loss: 0.7860307693481445:  96%|█████████▌| 6218/6470 [1:11:34<02:57,  1.42it/s]
2|6219|Loss: 0.7860307693481445:  96%|█████████▌| 6219/6470 [1:11:34<03:00,  1.39it/s]
2|6220|Loss: 0.9346418976783752:  96%|█████████▌| 6219/6470 [1:11:35<03:00,  1.39it/s]
2|6220|Loss: 0.9346418976783752:  96%|█████████▌| 6220/6470 [1:11:35<02:51,  1.45it/s]
2|6221|Loss: 0.7401962876319885:  96%|█████████▌| 6220/6470 [1:11:35<02:51,  1.45it/s]
2|6221|Loss: 0.7401962876319885:  96%|█████████▌| 6221/6470 [1:11:36<02:46,  1.49it/s]
2|6222|Loss: 0.8334164023399353:  96%|█████████▌| 6221/6470 [1:11:36<02:46,  1.49it/s]
2|6222|Loss: 0.8334164023399353:  96%|█████████▌| 6222/6470 [1:11:36<02:48,  1.47it/s]
2|6223|Loss: 0.8885740041732788:  96%|█████████▌| 6222/6470 [1:11:37<02:48,  1.47it/s]
2|6223|Loss: 0.8885740041732788:  96%|█████████▌| 6223/6470 [1:11:37<02:59,  1.38it/s]
2|6224|Loss: 0.9643133282661438:  96%|█████████▌| 6223/6470 [1:11:38<02:59,  1.38it/s]
2|6224|Loss: 0.9643133282661438:  96%|█████████▌| 6224/6470 [1:11:38<02:48,  1.46it/s]
2|6225|Loss: 0.7561758756637573:  96%|█████████▌| 6224/6470 [1:11:38<02:48,  1.46it/s]
2|6225|Loss: 0.7561758756637573:  96%|█████████▌| 6225/6470 [1:11:38<02:46,  1.47it/s]
2|6226|Loss: 0.9362037181854248:  96%|█████████▌| 6225/6470 [1:11:39<02:46,  1.47it/s]
2|6226|Loss: 0.9362037181854248:  96%|█████████▌| 6226/6470 [1:11:39<02:48,  1.45it/s]
2|6227|Loss: 0.8879138231277466:  96%|█████████▌| 6226/6470 [1:11:40<02:48,  1.45it/s]
2|6227|Loss: 0.8879138231277466:  96%|█████████▌| 6227/6470 [1:11:40<02:50,  1.42it/s]
2|6228|Loss: 0.8258199095726013:  96%|█████████▌| 6227/6470 [1:11:40<02:50,  1.42it/s]
2|6228|Loss: 0.8258199095726013:  96%|█████████▋| 6228/6470 [1:11:41<02:52,  1.40it/s]
2|6229|Loss: 0.9265369772911072:  96%|█████████▋| 6228/6470 [1:11:41<02:52,  1.40it/s]
2|6229|Loss: 0.9265369772911072:  96%|█████████▋| 6229/6470 [1:11:41<03:00,  1.34it/s]
2|6230|Loss: 0.8680201172828674:  96%|█████████▋| 6229/6470 [1:11:42<03:00,  1.34it/s]
2|6230|Loss: 0.8680201172828674:  96%|█████████▋| 6230/6470 [1:11:42<02:54,  1.38it/s]
2|6231|Loss: 0.937471330165863:  96%|█████████▋| 6230/6470 [1:11:42<02:54,  1.38it/s] 
2|6231|Loss: 0.937471330165863:  96%|█████████▋| 6231/6470 [1:11:43<02:43,  1.46it/s]
2|6232|Loss: 0.7710868716239929:  96%|█████████▋| 6231/6470 [1:11:43<02:43,  1.46it/s]
2|6232|Loss: 0.7710868716239929:  96%|█████████▋| 6232/6470 [1:11:43<02:36,  1.52it/s]
2|6233|Loss: 0.6782092452049255:  96%|█████████▋| 6232/6470 [1:11:44<02:36,  1.52it/s]
2|6233|Loss: 0.6782092452049255:  96%|█████████▋| 6233/6470 [1:11:44<02:22,  1.66it/s]
2|6234|Loss: 0.8898656368255615:  96%|█████████▋| 6233/6470 [1:11:44<02:22,  1.66it/s]
2|6234|Loss: 0.8898656368255615:  96%|█████████▋| 6234/6470 [1:11:44<02:21,  1.67it/s]
2|6235|Loss: 0.8144374489784241:  96%|█████████▋| 6234/6470 [1:11:45<02:21,  1.67it/s]
2|6235|Loss: 0.8144374489784241:  96%|█████████▋| 6235/6470 [1:11:45<02:23,  1.64it/s]
2|6236|Loss: 1.0498974323272705:  96%|█████████▋| 6235/6470 [1:11:45<02:23,  1.64it/s]
2|6236|Loss: 1.0498974323272705:  96%|█████████▋| 6236/6470 [1:11:46<02:37,  1.49it/s]
2|6237|Loss: 0.8269703388214111:  96%|█████████▋| 6236/6470 [1:11:46<02:37,  1.49it/s]
2|6237|Loss: 0.8269703388214111:  96%|█████████▋| 6237/6470 [1:11:47<02:48,  1.38it/s]
2|6238|Loss: 0.8530403971672058:  96%|█████████▋| 6237/6470 [1:11:47<02:48,  1.38it/s]
2|6238|Loss: 0.8530403971672058:  96%|█████████▋| 6238/6470 [1:11:47<02:50,  1.36it/s]
2|6239|Loss: 0.808732271194458:  96%|█████████▋| 6238/6470 [1:11:48<02:50,  1.36it/s] 
2|6239|Loss: 0.808732271194458:  96%|█████████▋| 6239/6470 [1:11:48<02:40,  1.44it/s]
2|6240|Loss: 0.8550379276275635:  96%|█████████▋| 6239/6470 [1:11:48<02:40,  1.44it/s]
2|6240|Loss: 0.8550379276275635:  96%|█████████▋| 6240/6470 [1:11:49<02:44,  1.40it/s]
2|6241|Loss: 0.8422328233718872:  96%|█████████▋| 6240/6470 [1:11:49<02:44,  1.40it/s]
2|6241|Loss: 0.8422328233718872:  96%|█████████▋| 6241/6470 [1:11:49<02:41,  1.42it/s]
2|6242|Loss: 0.9812912344932556:  96%|█████████▋| 6241/6470 [1:11:50<02:41,  1.42it/s]
2|6242|Loss: 0.9812912344932556:  96%|█████████▋| 6242/6470 [1:11:50<02:39,  1.43it/s]
2|6243|Loss: 0.8598299026489258:  96%|█████████▋| 6242/6470 [1:11:51<02:39,  1.43it/s]
2|6243|Loss: 0.8598299026489258:  96%|█████████▋| 6243/6470 [1:11:51<02:39,  1.42it/s]
2|6244|Loss: 0.7384448051452637:  96%|█████████▋| 6243/6470 [1:11:51<02:39,  1.42it/s]
2|6244|Loss: 0.7384448051452637:  97%|█████████▋| 6244/6470 [1:11:51<02:32,  1.48it/s]
2|6245|Loss: 0.8264685869216919:  97%|█████████▋| 6244/6470 [1:11:52<02:32,  1.48it/s]
2|6245|Loss: 0.8264685869216919:  97%|█████████▋| 6245/6470 [1:11:52<02:33,  1.47it/s]
2|6246|Loss: 0.6919780373573303:  97%|█████████▋| 6245/6470 [1:11:53<02:33,  1.47it/s]
2|6246|Loss: 0.6919780373573303:  97%|█████████▋| 6246/6470 [1:11:53<02:37,  1.42it/s]
2|6247|Loss: 0.9810929894447327:  97%|█████████▋| 6246/6470 [1:11:53<02:37,  1.42it/s]
2|6247|Loss: 0.9810929894447327:  97%|█████████▋| 6247/6470 [1:11:54<02:29,  1.49it/s]
2|6248|Loss: 0.8701410293579102:  97%|█████████▋| 6247/6470 [1:11:54<02:29,  1.49it/s]
2|6248|Loss: 0.8701410293579102:  97%|█████████▋| 6248/6470 [1:11:54<02:19,  1.60it/s]
2|6249|Loss: 0.9775238037109375:  97%|█████████▋| 6248/6470 [1:11:54<02:19,  1.60it/s]
2|6249|Loss: 0.9775238037109375:  97%|█████████▋| 6249/6470 [1:11:55<02:18,  1.60it/s]
2|6250|Loss: 0.764358401298523:  97%|█████████▋| 6249/6470 [1:11:55<02:18,  1.60it/s] 
2|6250|Loss: 0.764358401298523:  97%|█████████▋| 6250/6470 [1:11:55<02:22,  1.55it/s]
2|6251|Loss: 0.8149765729904175:  97%|█████████▋| 6250/6470 [1:11:56<02:22,  1.55it/s]
2|6251|Loss: 0.8149765729904175:  97%|█████████▋| 6251/6470 [1:11:56<02:21,  1.55it/s]
2|6252|Loss: 0.8385604619979858:  97%|█████████▋| 6251/6470 [1:11:56<02:21,  1.55it/s]
2|6252|Loss: 0.8385604619979858:  97%|█████████▋| 6252/6470 [1:11:57<02:25,  1.50it/s]
2|6253|Loss: 0.8358909487724304:  97%|█████████▋| 6252/6470 [1:11:57<02:25,  1.50it/s]
2|6253|Loss: 0.8358909487724304:  97%|█████████▋| 6253/6470 [1:11:57<02:23,  1.51it/s]
2|6254|Loss: 0.8613473773002625:  97%|█████████▋| 6253/6470 [1:11:58<02:23,  1.51it/s]
2|6254|Loss: 0.8613473773002625:  97%|█████████▋| 6254/6470 [1:11:58<02:29,  1.45it/s]
2|6255|Loss: 0.723747730255127:  97%|█████████▋| 6254/6470 [1:11:59<02:29,  1.45it/s] 
2|6255|Loss: 0.723747730255127:  97%|█████████▋| 6255/6470 [1:11:59<02:31,  1.42it/s]
2|6256|Loss: 0.8478964567184448:  97%|█████████▋| 6255/6470 [1:11:59<02:31,  1.42it/s]
2|6256|Loss: 0.8478964567184448:  97%|█████████▋| 6256/6470 [1:11:59<02:23,  1.49it/s]
2|6257|Loss: 0.8550968766212463:  97%|█████████▋| 6256/6470 [1:12:00<02:23,  1.49it/s]
2|6257|Loss: 0.8550968766212463:  97%|█████████▋| 6257/6470 [1:12:00<02:28,  1.43it/s]
2|6258|Loss: 0.7912915349006653:  97%|█████████▋| 6257/6470 [1:12:01<02:28,  1.43it/s]
2|6258|Loss: 0.7912915349006653:  97%|█████████▋| 6258/6470 [1:12:01<02:39,  1.33it/s]
2|6259|Loss: 0.8617140054702759:  97%|█████████▋| 6258/6470 [1:12:02<02:39,  1.33it/s]
2|6259|Loss: 0.8617140054702759:  97%|█████████▋| 6259/6470 [1:12:02<02:38,  1.33it/s]
2|6260|Loss: 0.794084370136261:  97%|█████████▋| 6259/6470 [1:12:02<02:38,  1.33it/s] 
2|6260|Loss: 0.794084370136261:  97%|█████████▋| 6260/6470 [1:12:02<02:22,  1.48it/s]
2|6261|Loss: 0.7910795211791992:  97%|█████████▋| 6260/6470 [1:12:03<02:22,  1.48it/s]
2|6261|Loss: 0.7910795211791992:  97%|█████████▋| 6261/6470 [1:12:03<02:19,  1.50it/s]
2|6262|Loss: 0.6702440977096558:  97%|█████████▋| 6261/6470 [1:12:03<02:19,  1.50it/s]
2|6262|Loss: 0.6702440977096558:  97%|█████████▋| 6262/6470 [1:12:04<02:17,  1.52it/s]
2|6263|Loss: 0.7774473428726196:  97%|█████████▋| 6262/6470 [1:12:04<02:17,  1.52it/s]
2|6263|Loss: 0.7774473428726196:  97%|█████████▋| 6263/6470 [1:12:04<02:25,  1.42it/s]
2|6264|Loss: 1.030924677848816:  97%|█████████▋| 6263/6470 [1:12:05<02:25,  1.42it/s] 
2|6264|Loss: 1.030924677848816:  97%|█████████▋| 6264/6470 [1:12:05<02:31,  1.36it/s]
2|6265|Loss: 0.7837918996810913:  97%|█████████▋| 6264/6470 [1:12:06<02:31,  1.36it/s]
2|6265|Loss: 0.7837918996810913:  97%|█████████▋| 6265/6470 [1:12:06<02:21,  1.45it/s]
2|6266|Loss: 0.7868067622184753:  97%|█████████▋| 6265/6470 [1:12:06<02:21,  1.45it/s]
2|6266|Loss: 0.7868067622184753:  97%|█████████▋| 6266/6470 [1:12:07<02:22,  1.43it/s]
2|6267|Loss: 0.8293249607086182:  97%|█████████▋| 6266/6470 [1:12:07<02:22,  1.43it/s]
2|6267|Loss: 0.8293249607086182:  97%|█████████▋| 6267/6470 [1:12:07<02:23,  1.42it/s]
2|6268|Loss: 0.9769173264503479:  97%|█████████▋| 6267/6470 [1:12:08<02:23,  1.42it/s]
2|6268|Loss: 0.9769173264503479:  97%|█████████▋| 6268/6470 [1:12:08<02:17,  1.47it/s]
2|6269|Loss: 0.8573272824287415:  97%|█████████▋| 6268/6470 [1:12:08<02:17,  1.47it/s]
2|6269|Loss: 0.8573272824287415:  97%|█████████▋| 6269/6470 [1:12:08<02:06,  1.58it/s]
2|6270|Loss: 0.9479888677597046:  97%|█████████▋| 6269/6470 [1:12:09<02:06,  1.58it/s]
2|6270|Loss: 0.9479888677597046:  97%|█████████▋| 6270/6470 [1:12:09<02:03,  1.61it/s]
2|6271|Loss: 1.0507270097732544:  97%|█████████▋| 6270/6470 [1:12:09<02:03,  1.61it/s]
2|6271|Loss: 1.0507270097732544:  97%|█████████▋| 6271/6470 [1:12:10<02:12,  1.51it/s]
2|6272|Loss: 1.031831979751587:  97%|█████████▋| 6271/6470 [1:12:10<02:12,  1.51it/s] 
2|6272|Loss: 1.031831979751587:  97%|█████████▋| 6272/6470 [1:12:11<02:17,  1.44it/s]
2|6273|Loss: 0.8865164518356323:  97%|█████████▋| 6272/6470 [1:12:11<02:17,  1.44it/s]
2|6273|Loss: 0.8865164518356323:  97%|█████████▋| 6273/6470 [1:12:11<02:20,  1.40it/s]
2|6274|Loss: 0.8003543019294739:  97%|█████████▋| 6273/6470 [1:12:12<02:20,  1.40it/s]
2|6274|Loss: 0.8003543019294739:  97%|█████████▋| 6274/6470 [1:12:12<02:20,  1.40it/s]
2|6275|Loss: 0.7819660305976868:  97%|█████████▋| 6274/6470 [1:12:12<02:20,  1.40it/s]
2|6275|Loss: 0.7819660305976868:  97%|█████████▋| 6275/6470 [1:12:13<02:08,  1.52it/s]
2|6276|Loss: 0.9237727522850037:  97%|█████████▋| 6275/6470 [1:12:13<02:08,  1.52it/s]
2|6276|Loss: 0.9237727522850037:  97%|█████████▋| 6276/6470 [1:12:13<02:07,  1.52it/s]
2|6277|Loss: 0.7909780144691467:  97%|█████████▋| 6276/6470 [1:12:14<02:07,  1.52it/s]
2|6277|Loss: 0.7909780144691467:  97%|█████████▋| 6277/6470 [1:12:14<02:14,  1.43it/s]
2|6278|Loss: 0.8609262108802795:  97%|█████████▋| 6277/6470 [1:12:14<02:14,  1.43it/s]
2|6278|Loss: 0.8609262108802795:  97%|█████████▋| 6278/6470 [1:12:15<02:11,  1.46it/s]
2|6279|Loss: 0.8068729639053345:  97%|█████████▋| 6278/6470 [1:12:15<02:11,  1.46it/s]
2|6279|Loss: 0.8068729639053345:  97%|█████████▋| 6279/6470 [1:12:15<02:12,  1.44it/s]
2|6280|Loss: 0.8527144193649292:  97%|█████████▋| 6279/6470 [1:12:16<02:12,  1.44it/s]
2|6280|Loss: 0.8527144193649292:  97%|█████████▋| 6280/6470 [1:12:16<02:06,  1.50it/s]
2|6281|Loss: 0.9385493397712708:  97%|█████████▋| 6280/6470 [1:12:16<02:06,  1.50it/s]
2|6281|Loss: 0.9385493397712708:  97%|█████████▋| 6281/6470 [1:12:16<01:56,  1.62it/s]
2|6282|Loss: 0.8966261148452759:  97%|█████████▋| 6281/6470 [1:12:17<01:56,  1.62it/s]
2|6282|Loss: 0.8966261148452759:  97%|█████████▋| 6282/6470 [1:12:17<01:59,  1.57it/s]
2|6283|Loss: 0.8378124833106995:  97%|█████████▋| 6282/6470 [1:12:17<01:59,  1.57it/s]
2|6283|Loss: 0.8378124833106995:  97%|█████████▋| 6283/6470 [1:12:18<01:47,  1.73it/s]
2|6284|Loss: 0.8742192983627319:  97%|█████████▋| 6283/6470 [1:12:18<01:47,  1.73it/s]
2|6284|Loss: 0.8742192983627319:  97%|█████████▋| 6284/6470 [1:12:18<01:52,  1.65it/s]
2|6285|Loss: 0.775881290435791:  97%|█████████▋| 6284/6470 [1:12:19<01:52,  1.65it/s] 
2|6285|Loss: 0.775881290435791:  97%|█████████▋| 6285/6470 [1:12:19<01:54,  1.61it/s]
2|6286|Loss: 0.972424328327179:  97%|█████████▋| 6285/6470 [1:12:19<01:54,  1.61it/s]
2|6286|Loss: 0.972424328327179:  97%|█████████▋| 6286/6470 [1:12:20<01:58,  1.55it/s]
2|6287|Loss: 0.8214929699897766:  97%|█████████▋| 6286/6470 [1:12:20<01:58,  1.55it/s]
2|6287|Loss: 0.8214929699897766:  97%|█████████▋| 6287/6470 [1:12:20<02:07,  1.43it/s]
2|6288|Loss: 0.7850510478019714:  97%|█████████▋| 6287/6470 [1:12:21<02:07,  1.43it/s]
2|6288|Loss: 0.7850510478019714:  97%|█████████▋| 6288/6470 [1:12:21<02:08,  1.41it/s]
2|6289|Loss: 0.959080159664154:  97%|█████████▋| 6288/6470 [1:12:22<02:08,  1.41it/s] 
2|6289|Loss: 0.959080159664154:  97%|█████████▋| 6289/6470 [1:12:22<02:07,  1.42it/s]
2|6290|Loss: 1.0205442905426025:  97%|█████████▋| 6289/6470 [1:12:22<02:07,  1.42it/s]
2|6290|Loss: 1.0205442905426025:  97%|█████████▋| 6290/6470 [1:12:23<02:10,  1.38it/s]
2|6291|Loss: 0.8139699697494507:  97%|█████████▋| 6290/6470 [1:12:23<02:10,  1.38it/s]
2|6291|Loss: 0.8139699697494507:  97%|█████████▋| 6291/6470 [1:12:23<02:12,  1.35it/s]
2|6292|Loss: 0.8369443416595459:  97%|█████████▋| 6291/6470 [1:12:24<02:12,  1.35it/s]
2|6292|Loss: 0.8369443416595459:  97%|█████████▋| 6292/6470 [1:12:24<02:16,  1.30it/s]
2|6293|Loss: 0.861775279045105:  97%|█████████▋| 6292/6470 [1:12:25<02:16,  1.30it/s] 
2|6293|Loss: 0.861775279045105:  97%|█████████▋| 6293/6470 [1:12:25<02:11,  1.34it/s]
2|6294|Loss: 0.8066403865814209:  97%|█████████▋| 6293/6470 [1:12:25<02:11,  1.34it/s]
2|6294|Loss: 0.8066403865814209:  97%|█████████▋| 6294/6470 [1:12:26<02:07,  1.38it/s]
2|6295|Loss: 0.9384454488754272:  97%|█████████▋| 6294/6470 [1:12:26<02:07,  1.38it/s]
2|6295|Loss: 0.9384454488754272:  97%|█████████▋| 6295/6470 [1:12:26<02:09,  1.35it/s]
2|6296|Loss: 0.8103502988815308:  97%|█████████▋| 6295/6470 [1:12:27<02:09,  1.35it/s]
2|6296|Loss: 0.8103502988815308:  97%|█████████▋| 6296/6470 [1:12:27<02:08,  1.35it/s]
2|6297|Loss: 0.7434309124946594:  97%|█████████▋| 6296/6470 [1:12:28<02:08,  1.35it/s]
2|6297|Loss: 0.7434309124946594:  97%|█████████▋| 6297/6470 [1:12:28<02:17,  1.25it/s]
2|6298|Loss: 0.8118402361869812:  97%|█████████▋| 6297/6470 [1:12:29<02:17,  1.25it/s]
2|6298|Loss: 0.8118402361869812:  97%|█████████▋| 6298/6470 [1:12:29<02:12,  1.30it/s]
2|6299|Loss: 0.661212146282196:  97%|█████████▋| 6298/6470 [1:12:29<02:12,  1.30it/s] 
2|6299|Loss: 0.661212146282196:  97%|█████████▋| 6299/6470 [1:12:30<02:16,  1.26it/s]
2|6300|Loss: 0.6844174861907959:  97%|█████████▋| 6299/6470 [1:12:30<02:16,  1.26it/s]
2|6300|Loss: 0.6844174861907959:  97%|█████████▋| 6300/6470 [1:12:31<02:24,  1.18it/s]
2|6301|Loss: 0.7743574380874634:  97%|█████████▋| 6300/6470 [1:12:31<02:24,  1.18it/s]
2|6301|Loss: 0.7743574380874634:  97%|█████████▋| 6301/6470 [1:12:31<02:16,  1.24it/s]
2|6302|Loss: 0.8190906047821045:  97%|█████████▋| 6301/6470 [1:12:32<02:16,  1.24it/s]
2|6302|Loss: 0.8190906047821045:  97%|█████████▋| 6302/6470 [1:12:32<02:10,  1.28it/s]
2|6303|Loss: 0.8934342265129089:  97%|█████████▋| 6302/6470 [1:12:32<02:10,  1.28it/s]
2|6303|Loss: 0.8934342265129089:  97%|█████████▋| 6303/6470 [1:12:33<02:08,  1.30it/s]
2|6304|Loss: 0.894690990447998:  97%|█████████▋| 6303/6470 [1:12:33<02:08,  1.30it/s] 
2|6304|Loss: 0.894690990447998:  97%|█████████▋| 6304/6470 [1:12:33<02:01,  1.37it/s]
2|6305|Loss: 0.9463527798652649:  97%|█████████▋| 6304/6470 [1:12:34<02:01,  1.37it/s]
2|6305|Loss: 0.9463527798652649:  97%|█████████▋| 6305/6470 [1:12:34<01:53,  1.46it/s]
2|6306|Loss: 0.7416913509368896:  97%|█████████▋| 6305/6470 [1:12:34<01:53,  1.46it/s]
2|6306|Loss: 0.7416913509368896:  97%|█████████▋| 6306/6470 [1:12:35<01:58,  1.38it/s]
2|6307|Loss: 0.8242160081863403:  97%|█████████▋| 6306/6470 [1:12:35<01:58,  1.38it/s]
2|6307|Loss: 0.8242160081863403:  97%|█████████▋| 6307/6470 [1:12:35<01:51,  1.46it/s]
2|6308|Loss: 0.8612983226776123:  97%|█████████▋| 6307/6470 [1:12:36<01:51,  1.46it/s]
2|6308|Loss: 0.8612983226776123:  97%|█████████▋| 6308/6470 [1:12:36<01:47,  1.50it/s]
2|6309|Loss: 0.7629485130310059:  97%|█████████▋| 6308/6470 [1:12:36<01:47,  1.50it/s]
2|6309|Loss: 0.7629485130310059:  98%|█████████▊| 6309/6470 [1:12:37<01:55,  1.39it/s]
2|6310|Loss: 0.7690621018409729:  98%|█████████▊| 6309/6470 [1:12:37<01:55,  1.39it/s]
2|6310|Loss: 0.7690621018409729:  98%|█████████▊| 6310/6470 [1:12:38<01:54,  1.40it/s]
2|6311|Loss: 0.8902978301048279:  98%|█████████▊| 6310/6470 [1:12:38<01:54,  1.40it/s]
2|6311|Loss: 0.8902978301048279:  98%|█████████▊| 6311/6470 [1:12:38<02:00,  1.32it/s]
2|6312|Loss: 0.8101360201835632:  98%|█████████▊| 6311/6470 [1:12:39<02:00,  1.32it/s]
2|6312|Loss: 0.8101360201835632:  98%|█████████▊| 6312/6470 [1:12:39<02:04,  1.27it/s]
2|6313|Loss: 0.899226725101471:  98%|█████████▊| 6312/6470 [1:12:40<02:04,  1.27it/s] 
2|6313|Loss: 0.899226725101471:  98%|█████████▊| 6313/6470 [1:12:40<01:54,  1.37it/s]
2|6314|Loss: 0.731798529624939:  98%|█████████▊| 6313/6470 [1:12:40<01:54,  1.37it/s]
2|6314|Loss: 0.731798529624939:  98%|█████████▊| 6314/6470 [1:12:40<01:47,  1.45it/s]
2|6315|Loss: 0.8342519402503967:  98%|█████████▊| 6314/6470 [1:12:41<01:47,  1.45it/s]
2|6315|Loss: 0.8342519402503967:  98%|█████████▊| 6315/6470 [1:12:41<01:51,  1.40it/s]
2|6316|Loss: 0.7433543801307678:  98%|█████████▊| 6315/6470 [1:12:42<01:51,  1.40it/s]
2|6316|Loss: 0.7433543801307678:  98%|█████████▊| 6316/6470 [1:12:42<01:52,  1.37it/s]
2|6317|Loss: 0.9610083699226379:  98%|█████████▊| 6316/6470 [1:12:42<01:52,  1.37it/s]
2|6317|Loss: 0.9610083699226379:  98%|█████████▊| 6317/6470 [1:12:43<01:56,  1.32it/s]
2|6318|Loss: 0.9034081697463989:  98%|█████████▊| 6317/6470 [1:12:43<01:56,  1.32it/s]
2|6318|Loss: 0.9034081697463989:  98%|█████████▊| 6318/6470 [1:12:44<01:59,  1.27it/s]
2|6319|Loss: 0.898045003414154:  98%|█████████▊| 6318/6470 [1:12:44<01:59,  1.27it/s] 
2|6319|Loss: 0.898045003414154:  98%|█████████▊| 6319/6470 [1:12:44<01:57,  1.28it/s]
2|6320|Loss: 0.6079434156417847:  98%|█████████▊| 6319/6470 [1:12:45<01:57,  1.28it/s]
2|6320|Loss: 0.6079434156417847:  98%|█████████▊| 6320/6470 [1:12:45<02:04,  1.20it/s]
2|6321|Loss: 0.8125686049461365:  98%|█████████▊| 6320/6470 [1:12:46<02:04,  1.20it/s]
2|6321|Loss: 0.8125686049461365:  98%|█████████▊| 6321/6470 [1:12:46<01:58,  1.26it/s]
2|6322|Loss: 0.8755654692649841:  98%|█████████▊| 6321/6470 [1:12:47<01:58,  1.26it/s]
2|6322|Loss: 0.8755654692649841:  98%|█████████▊| 6322/6470 [1:12:47<01:56,  1.27it/s]
2|6323|Loss: 0.8669988512992859:  98%|█████████▊| 6322/6470 [1:12:47<01:56,  1.27it/s]
2|6323|Loss: 0.8669988512992859:  98%|█████████▊| 6323/6470 [1:12:47<01:44,  1.41it/s]
2|6324|Loss: 0.774009108543396:  98%|█████████▊| 6323/6470 [1:12:48<01:44,  1.41it/s] 
2|6324|Loss: 0.774009108543396:  98%|█████████▊| 6324/6470 [1:12:48<01:41,  1.43it/s]
2|6325|Loss: 0.8315186500549316:  98%|█████████▊| 6324/6470 [1:12:49<01:41,  1.43it/s]
2|6325|Loss: 0.8315186500549316:  98%|█████████▊| 6325/6470 [1:12:49<01:41,  1.42it/s]
2|6326|Loss: 0.7880785465240479:  98%|█████████▊| 6325/6470 [1:12:49<01:41,  1.42it/s]
2|6326|Loss: 0.7880785465240479:  98%|█████████▊| 6326/6470 [1:12:49<01:38,  1.47it/s]
2|6327|Loss: 0.9589091539382935:  98%|█████████▊| 6326/6470 [1:12:50<01:38,  1.47it/s]
2|6327|Loss: 0.9589091539382935:  98%|█████████▊| 6327/6470 [1:12:50<01:30,  1.58it/s]
2|6328|Loss: 0.7989503741264343:  98%|█████████▊| 6327/6470 [1:12:50<01:30,  1.58it/s]
2|6328|Loss: 0.7989503741264343:  98%|█████████▊| 6328/6470 [1:12:51<01:28,  1.60it/s]
2|6329|Loss: 0.8978175520896912:  98%|█████████▊| 6328/6470 [1:12:51<01:28,  1.60it/s]
2|6329|Loss: 0.8978175520896912:  98%|█████████▊| 6329/6470 [1:12:51<01:34,  1.50it/s]
2|6330|Loss: 0.9149937629699707:  98%|█████████▊| 6329/6470 [1:12:52<01:34,  1.50it/s]
2|6330|Loss: 0.9149937629699707:  98%|█████████▊| 6330/6470 [1:12:52<01:36,  1.44it/s]
2|6331|Loss: 0.9216583371162415:  98%|█████████▊| 6330/6470 [1:12:52<01:36,  1.44it/s]
2|6331|Loss: 0.9216583371162415:  98%|█████████▊| 6331/6470 [1:12:53<01:34,  1.47it/s]
2|6332|Loss: 0.8547952771186829:  98%|█████████▊| 6331/6470 [1:12:53<01:34,  1.47it/s]
2|6332|Loss: 0.8547952771186829:  98%|█████████▊| 6332/6470 [1:12:53<01:36,  1.43it/s]
2|6333|Loss: 0.9223146438598633:  98%|█████████▊| 6332/6470 [1:12:54<01:36,  1.43it/s]
2|6333|Loss: 0.9223146438598633:  98%|█████████▊| 6333/6470 [1:12:54<01:38,  1.39it/s]
2|6334|Loss: 0.8172767758369446:  98%|█████████▊| 6333/6470 [1:12:55<01:38,  1.39it/s]
2|6334|Loss: 0.8172767758369446:  98%|█████████▊| 6334/6470 [1:12:55<01:31,  1.49it/s]
2|6335|Loss: 0.7724738121032715:  98%|█████████▊| 6334/6470 [1:12:55<01:31,  1.49it/s]
2|6335|Loss: 0.7724738121032715:  98%|█████████▊| 6335/6470 [1:12:56<01:37,  1.39it/s]
2|6336|Loss: 0.7592065930366516:  98%|█████████▊| 6335/6470 [1:12:56<01:37,  1.39it/s]
2|6336|Loss: 0.7592065930366516:  98%|█████████▊| 6336/6470 [1:12:57<01:43,  1.30it/s]
2|6337|Loss: 0.9091812968254089:  98%|█████████▊| 6336/6470 [1:12:57<01:43,  1.30it/s]
2|6337|Loss: 0.9091812968254089:  98%|█████████▊| 6337/6470 [1:12:57<01:45,  1.26it/s]
2|6338|Loss: 0.8573036789894104:  98%|█████████▊| 6337/6470 [1:12:58<01:45,  1.26it/s]
2|6338|Loss: 0.8573036789894104:  98%|█████████▊| 6338/6470 [1:12:58<01:43,  1.28it/s]
2|6339|Loss: 0.853659451007843:  98%|█████████▊| 6338/6470 [1:12:59<01:43,  1.28it/s] 
2|6339|Loss: 0.853659451007843:  98%|█████████▊| 6339/6470 [1:12:59<01:35,  1.37it/s]
2|6340|Loss: 0.6617804765701294:  98%|█████████▊| 6339/6470 [1:12:59<01:35,  1.37it/s]
2|6340|Loss: 0.6617804765701294:  98%|█████████▊| 6340/6470 [1:12:59<01:28,  1.47it/s]
2|6341|Loss: 0.9423606991767883:  98%|█████████▊| 6340/6470 [1:13:00<01:28,  1.47it/s]
2|6341|Loss: 0.9423606991767883:  98%|█████████▊| 6341/6470 [1:13:00<01:25,  1.51it/s]
2|6342|Loss: 0.7216024994850159:  98%|█████████▊| 6341/6470 [1:13:00<01:25,  1.51it/s]
2|6342|Loss: 0.7216024994850159:  98%|█████████▊| 6342/6470 [1:13:00<01:20,  1.60it/s]
2|6343|Loss: 0.7583393454551697:  98%|█████████▊| 6342/6470 [1:13:01<01:20,  1.60it/s]
2|6343|Loss: 0.7583393454551697:  98%|█████████▊| 6343/6470 [1:13:01<01:24,  1.50it/s]
2|6344|Loss: 1.0847420692443848:  98%|█████████▊| 6343/6470 [1:13:02<01:24,  1.50it/s]
2|6344|Loss: 1.0847420692443848:  98%|█████████▊| 6344/6470 [1:13:02<01:23,  1.50it/s]
2|6345|Loss: 0.8501639366149902:  98%|█████████▊| 6344/6470 [1:13:02<01:23,  1.50it/s]
2|6345|Loss: 0.8501639366149902:  98%|█████████▊| 6345/6470 [1:13:03<01:21,  1.54it/s]
2|6346|Loss: 0.8890465497970581:  98%|█████████▊| 6345/6470 [1:13:03<01:21,  1.54it/s]
2|6346|Loss: 0.8890465497970581:  98%|█████████▊| 6346/6470 [1:13:03<01:18,  1.58it/s]
2|6347|Loss: 0.8721648454666138:  98%|█████████▊| 6346/6470 [1:13:04<01:18,  1.58it/s]
2|6347|Loss: 0.8721648454666138:  98%|█████████▊| 6347/6470 [1:13:04<01:23,  1.48it/s]
2|6348|Loss: 0.8565523624420166:  98%|█████████▊| 6347/6470 [1:13:04<01:23,  1.48it/s]
2|6348|Loss: 0.8565523624420166:  98%|█████████▊| 6348/6470 [1:13:05<01:25,  1.42it/s]
2|6349|Loss: 0.8848878741264343:  98%|█████████▊| 6348/6470 [1:13:05<01:25,  1.42it/s]
2|6349|Loss: 0.8848878741264343:  98%|█████████▊| 6349/6470 [1:13:05<01:26,  1.40it/s]
2|6350|Loss: 0.9068366289138794:  98%|█████████▊| 6349/6470 [1:13:06<01:26,  1.40it/s]
2|6350|Loss: 0.9068366289138794:  98%|█████████▊| 6350/6470 [1:13:06<01:28,  1.36it/s]
2|6351|Loss: 0.7145634293556213:  98%|█████████▊| 6350/6470 [1:13:07<01:28,  1.36it/s]
2|6351|Loss: 0.7145634293556213:  98%|█████████▊| 6351/6470 [1:13:07<01:19,  1.49it/s]
2|6352|Loss: 0.8916372060775757:  98%|█████████▊| 6351/6470 [1:13:07<01:19,  1.49it/s]
2|6352|Loss: 0.8916372060775757:  98%|█████████▊| 6352/6470 [1:13:07<01:16,  1.54it/s]
2|6353|Loss: 0.7899292707443237:  98%|█████████▊| 6352/6470 [1:13:08<01:16,  1.54it/s]
2|6353|Loss: 0.7899292707443237:  98%|█████████▊| 6353/6470 [1:13:08<01:23,  1.40it/s]
2|6354|Loss: 0.8130102157592773:  98%|█████████▊| 6353/6470 [1:13:09<01:23,  1.40it/s]
2|6354|Loss: 0.8130102157592773:  98%|█████████▊| 6354/6470 [1:13:09<01:24,  1.38it/s]
2|6355|Loss: 0.8952379822731018:  98%|█████████▊| 6354/6470 [1:13:09<01:24,  1.38it/s]
2|6355|Loss: 0.8952379822731018:  98%|█████████▊| 6355/6470 [1:13:10<01:22,  1.39it/s]
2|6356|Loss: 0.7827837467193604:  98%|█████████▊| 6355/6470 [1:13:10<01:22,  1.39it/s]
2|6356|Loss: 0.7827837467193604:  98%|█████████▊| 6356/6470 [1:13:10<01:14,  1.53it/s]
2|6357|Loss: 0.6460399031639099:  98%|█████████▊| 6356/6470 [1:13:10<01:14,  1.53it/s]
2|6357|Loss: 0.6460399031639099:  98%|█████████▊| 6357/6470 [1:13:11<01:13,  1.54it/s]
2|6358|Loss: 1.0419466495513916:  98%|█████████▊| 6357/6470 [1:13:11<01:13,  1.54it/s]
2|6358|Loss: 1.0419466495513916:  98%|█████████▊| 6358/6470 [1:13:11<01:15,  1.48it/s]
2|6359|Loss: 0.9011580348014832:  98%|█████████▊| 6358/6470 [1:13:12<01:15,  1.48it/s]
2|6359|Loss: 0.9011580348014832:  98%|█████████▊| 6359/6470 [1:13:12<01:19,  1.39it/s]
2|6360|Loss: 0.8793028593063354:  98%|█████████▊| 6359/6470 [1:13:13<01:19,  1.39it/s]
2|6360|Loss: 0.8793028593063354:  98%|█████████▊| 6360/6470 [1:13:13<01:23,  1.31it/s]
2|6361|Loss: 0.7634036540985107:  98%|█████████▊| 6360/6470 [1:13:14<01:23,  1.31it/s]
2|6361|Loss: 0.7634036540985107:  98%|█████████▊| 6361/6470 [1:13:14<01:13,  1.48it/s]
2|6362|Loss: 0.7753180861473083:  98%|█████████▊| 6361/6470 [1:13:14<01:13,  1.48it/s]
2|6362|Loss: 0.7753180861473083:  98%|█████████▊| 6362/6470 [1:13:14<01:09,  1.55it/s]
2|6363|Loss: 0.7682226896286011:  98%|█████████▊| 6362/6470 [1:13:15<01:09,  1.55it/s]
2|6363|Loss: 0.7682226896286011:  98%|█████████▊| 6363/6470 [1:13:15<01:10,  1.52it/s]
2|6364|Loss: 0.790662944316864:  98%|█████████▊| 6363/6470 [1:13:15<01:10,  1.52it/s] 
2|6364|Loss: 0.790662944316864:  98%|█████████▊| 6364/6470 [1:13:16<01:11,  1.49it/s]
2|6365|Loss: 0.9336207509040833:  98%|█████████▊| 6364/6470 [1:13:16<01:11,  1.49it/s]
2|6365|Loss: 0.9336207509040833:  98%|█████████▊| 6365/6470 [1:13:16<01:09,  1.51it/s]
2|6366|Loss: 0.8317641615867615:  98%|█████████▊| 6365/6470 [1:13:17<01:09,  1.51it/s]
2|6366|Loss: 0.8317641615867615:  98%|█████████▊| 6366/6470 [1:13:17<01:10,  1.47it/s]
2|6367|Loss: 0.8857101798057556:  98%|█████████▊| 6366/6470 [1:13:17<01:10,  1.47it/s]
2|6367|Loss: 0.8857101798057556:  98%|█████████▊| 6367/6470 [1:13:18<01:08,  1.51it/s]
2|6368|Loss: 0.8052031993865967:  98%|█████████▊| 6367/6470 [1:13:18<01:08,  1.51it/s]
2|6368|Loss: 0.8052031993865967:  98%|█████████▊| 6368/6470 [1:13:18<01:05,  1.56it/s]
2|6369|Loss: 0.7734692096710205:  98%|█████████▊| 6368/6470 [1:13:18<01:05,  1.56it/s]
2|6369|Loss: 0.7734692096710205:  98%|█████████▊| 6369/6470 [1:13:19<00:59,  1.69it/s]
2|6370|Loss: 0.902904212474823:  98%|█████████▊| 6369/6470 [1:13:19<00:59,  1.69it/s] 
2|6370|Loss: 0.902904212474823:  98%|█████████▊| 6370/6470 [1:13:19<01:05,  1.53it/s]
2|6371|Loss: 0.8181862235069275:  98%|█████████▊| 6370/6470 [1:13:20<01:05,  1.53it/s]
2|6371|Loss: 0.8181862235069275:  98%|█████████▊| 6371/6470 [1:13:20<01:05,  1.51it/s]
2|6372|Loss: 0.8697492480278015:  98%|█████████▊| 6371/6470 [1:13:21<01:05,  1.51it/s]
2|6372|Loss: 0.8697492480278015:  98%|█████████▊| 6372/6470 [1:13:21<01:08,  1.44it/s]
2|6373|Loss: 0.8299587368965149:  98%|█████████▊| 6372/6470 [1:13:21<01:08,  1.44it/s]
2|6373|Loss: 0.8299587368965149:  99%|█████████▊| 6373/6470 [1:13:22<01:08,  1.42it/s]
2|6374|Loss: 1.0106498003005981:  99%|█████████▊| 6373/6470 [1:13:22<01:08,  1.42it/s]
2|6374|Loss: 1.0106498003005981:  99%|█████████▊| 6374/6470 [1:13:22<01:04,  1.48it/s]
2|6375|Loss: 0.7487004995346069:  99%|█████████▊| 6374/6470 [1:13:23<01:04,  1.48it/s]
2|6375|Loss: 0.7487004995346069:  99%|█████████▊| 6375/6470 [1:13:23<01:02,  1.52it/s]
2|6376|Loss: 0.840853214263916:  99%|█████████▊| 6375/6470 [1:13:23<01:02,  1.52it/s] 
2|6376|Loss: 0.840853214263916:  99%|█████████▊| 6376/6470 [1:13:23<01:00,  1.56it/s]
2|6377|Loss: 0.8405143618583679:  99%|█████████▊| 6376/6470 [1:13:24<01:00,  1.56it/s]
2|6377|Loss: 0.8405143618583679:  99%|█████████▊| 6377/6470 [1:13:24<00:58,  1.58it/s]
2|6378|Loss: 0.8289594650268555:  99%|█████████▊| 6377/6470 [1:13:24<00:58,  1.58it/s]
2|6378|Loss: 0.8289594650268555:  99%|█████████▊| 6378/6470 [1:13:25<01:00,  1.51it/s]
2|6379|Loss: 1.01778244972229:  99%|█████████▊| 6378/6470 [1:13:25<01:00,  1.51it/s]  
2|6379|Loss: 1.01778244972229:  99%|█████████▊| 6379/6470 [1:13:25<00:57,  1.58it/s]
2|6380|Loss: 0.8457683324813843:  99%|█████████▊| 6379/6470 [1:13:26<00:57,  1.58it/s]
2|6380|Loss: 0.8457683324813843:  99%|█████████▊| 6380/6470 [1:13:26<00:56,  1.60it/s]
2|6381|Loss: 0.9891428351402283:  99%|█████████▊| 6380/6470 [1:13:26<00:56,  1.60it/s]
2|6381|Loss: 0.9891428351402283:  99%|█████████▊| 6381/6470 [1:13:27<00:55,  1.62it/s]
2|6382|Loss: 0.8831998109817505:  99%|█████████▊| 6381/6470 [1:13:27<00:55,  1.62it/s]
2|6382|Loss: 0.8831998109817505:  99%|█████████▊| 6382/6470 [1:13:27<00:56,  1.55it/s]
2|6383|Loss: 0.9399693012237549:  99%|█████████▊| 6382/6470 [1:13:28<00:56,  1.55it/s]
2|6383|Loss: 0.9399693012237549:  99%|█████████▊| 6383/6470 [1:13:28<00:56,  1.55it/s]
2|6384|Loss: 0.7684940099716187:  99%|█████████▊| 6383/6470 [1:13:28<00:56,  1.55it/s]
2|6384|Loss: 0.7684940099716187:  99%|█████████▊| 6384/6470 [1:13:28<00:52,  1.63it/s]
2|6385|Loss: 0.817484438419342:  99%|█████████▊| 6384/6470 [1:13:29<00:52,  1.63it/s] 
2|6385|Loss: 0.817484438419342:  99%|█████████▊| 6385/6470 [1:13:29<00:52,  1.63it/s]
2|6386|Loss: 1.066131830215454:  99%|█████████▊| 6385/6470 [1:13:30<00:52,  1.63it/s]
2|6386|Loss: 1.066131830215454:  99%|█████████▊| 6386/6470 [1:13:30<00:57,  1.45it/s]
2|6387|Loss: 0.8259163498878479:  99%|█████████▊| 6386/6470 [1:13:30<00:57,  1.45it/s]
2|6387|Loss: 0.8259163498878479:  99%|█████████▊| 6387/6470 [1:13:31<00:57,  1.45it/s]
2|6388|Loss: 0.6036517024040222:  99%|█████████▊| 6387/6470 [1:13:31<00:57,  1.45it/s]
2|6388|Loss: 0.6036517024040222:  99%|█████████▊| 6388/6470 [1:13:31<00:54,  1.50it/s]
2|6389|Loss: 0.8491379618644714:  99%|█████████▊| 6388/6470 [1:13:32<00:54,  1.50it/s]
2|6389|Loss: 0.8491379618644714:  99%|█████████▊| 6389/6470 [1:13:32<00:53,  1.52it/s]
2|6390|Loss: 1.0196242332458496:  99%|█████████▊| 6389/6470 [1:13:32<00:53,  1.52it/s]
2|6390|Loss: 1.0196242332458496:  99%|█████████▉| 6390/6470 [1:13:33<00:55,  1.45it/s]
2|6391|Loss: 0.8509121537208557:  99%|█████████▉| 6390/6470 [1:13:33<00:55,  1.45it/s]
2|6391|Loss: 0.8509121537208557:  99%|█████████▉| 6391/6470 [1:13:33<00:55,  1.41it/s]
2|6392|Loss: 0.8071825504302979:  99%|█████████▉| 6391/6470 [1:13:34<00:55,  1.41it/s]
2|6392|Loss: 0.8071825504302979:  99%|█████████▉| 6392/6470 [1:13:34<00:51,  1.53it/s]
2|6393|Loss: 0.7695637345314026:  99%|█████████▉| 6392/6470 [1:13:34<00:51,  1.53it/s]
2|6393|Loss: 0.7695637345314026:  99%|█████████▉| 6393/6470 [1:13:35<00:50,  1.53it/s]
2|6394|Loss: 0.7653874158859253:  99%|█████████▉| 6393/6470 [1:13:35<00:50,  1.53it/s]
2|6394|Loss: 0.7653874158859253:  99%|█████████▉| 6394/6470 [1:13:35<00:49,  1.53it/s]
2|6395|Loss: 0.9408978819847107:  99%|█████████▉| 6394/6470 [1:13:36<00:49,  1.53it/s]
2|6395|Loss: 0.9408978819847107:  99%|█████████▉| 6395/6470 [1:13:36<00:48,  1.55it/s]
2|6396|Loss: 0.8988052010536194:  99%|█████████▉| 6395/6470 [1:13:36<00:48,  1.55it/s]
2|6396|Loss: 0.8988052010536194:  99%|█████████▉| 6396/6470 [1:13:36<00:46,  1.59it/s]
2|6397|Loss: 0.8173491358757019:  99%|█████████▉| 6396/6470 [1:13:37<00:46,  1.59it/s]
2|6397|Loss: 0.8173491358757019:  99%|█████████▉| 6397/6470 [1:13:37<00:46,  1.56it/s]
2|6398|Loss: 0.7984494566917419:  99%|█████████▉| 6397/6470 [1:13:38<00:46,  1.56it/s]
2|6398|Loss: 0.7984494566917419:  99%|█████████▉| 6398/6470 [1:13:38<00:50,  1.42it/s]
2|6399|Loss: 0.8920841217041016:  99%|█████████▉| 6398/6470 [1:13:38<00:50,  1.42it/s]
2|6399|Loss: 0.8920841217041016:  99%|█████████▉| 6399/6470 [1:13:39<00:48,  1.47it/s]
2|6400|Loss: 0.8504558205604553:  99%|█████████▉| 6399/6470 [1:13:39<00:48,  1.47it/s]
2|6400|Loss: 0.8504558205604553:  99%|█████████▉| 6400/6470 [1:13:39<00:46,  1.51it/s]
2|6401|Loss: 0.810302197933197:  99%|█████████▉| 6400/6470 [1:13:40<00:46,  1.51it/s] 
2|6401|Loss: 0.810302197933197:  99%|█████████▉| 6401/6470 [1:13:40<00:48,  1.41it/s]
2|6402|Loss: 0.8031486868858337:  99%|█████████▉| 6401/6470 [1:13:40<00:48,  1.41it/s]
2|6402|Loss: 0.8031486868858337:  99%|█████████▉| 6402/6470 [1:13:41<00:47,  1.42it/s]
2|6403|Loss: 0.8127009868621826:  99%|█████████▉| 6402/6470 [1:13:41<00:47,  1.42it/s]
2|6403|Loss: 0.8127009868621826:  99%|█████████▉| 6403/6470 [1:13:41<00:42,  1.56it/s]
2|6404|Loss: 1.0638915300369263:  99%|█████████▉| 6403/6470 [1:13:42<00:42,  1.56it/s]
2|6404|Loss: 1.0638915300369263:  99%|█████████▉| 6404/6470 [1:13:42<00:40,  1.64it/s]
2|6405|Loss: 0.8590457439422607:  99%|█████████▉| 6404/6470 [1:13:42<00:40,  1.64it/s]
2|6405|Loss: 0.8590457439422607:  99%|█████████▉| 6405/6470 [1:13:42<00:41,  1.56it/s]
2|6406|Loss: 0.9076052904129028:  99%|█████████▉| 6405/6470 [1:13:43<00:41,  1.56it/s]
2|6406|Loss: 0.9076052904129028:  99%|█████████▉| 6406/6470 [1:13:43<00:40,  1.59it/s]
2|6407|Loss: 0.8526079058647156:  99%|█████████▉| 6406/6470 [1:13:43<00:40,  1.59it/s]
2|6407|Loss: 0.8526079058647156:  99%|█████████▉| 6407/6470 [1:13:44<00:40,  1.55it/s]
2|6408|Loss: 0.905427098274231:  99%|█████████▉| 6407/6470 [1:13:44<00:40,  1.55it/s] 
2|6408|Loss: 0.905427098274231:  99%|█████████▉| 6408/6470 [1:13:44<00:39,  1.55it/s]
2|6409|Loss: 0.8367152810096741:  99%|█████████▉| 6408/6470 [1:13:45<00:39,  1.55it/s]
2|6409|Loss: 0.8367152810096741:  99%|█████████▉| 6409/6470 [1:13:45<00:38,  1.60it/s]
2|6410|Loss: 0.7757003307342529:  99%|█████████▉| 6409/6470 [1:13:45<00:38,  1.60it/s]
2|6410|Loss: 0.7757003307342529:  99%|█████████▉| 6410/6470 [1:13:46<00:36,  1.64it/s]
2|6411|Loss: 0.8893022537231445:  99%|█████████▉| 6410/6470 [1:13:46<00:36,  1.64it/s]
2|6411|Loss: 0.8893022537231445:  99%|█████████▉| 6411/6470 [1:13:46<00:37,  1.59it/s]
2|6412|Loss: 0.8064598441123962:  99%|█████████▉| 6411/6470 [1:13:47<00:37,  1.59it/s]
2|6412|Loss: 0.8064598441123962:  99%|█████████▉| 6412/6470 [1:13:47<00:38,  1.51it/s]
2|6413|Loss: 0.9951397776603699:  99%|█████████▉| 6412/6470 [1:13:47<00:38,  1.51it/s]
2|6413|Loss: 0.9951397776603699:  99%|█████████▉| 6413/6470 [1:13:48<00:38,  1.48it/s]
2|6414|Loss: 0.8549380302429199:  99%|█████████▉| 6413/6470 [1:13:48<00:38,  1.48it/s]
2|6414|Loss: 0.8549380302429199:  99%|█████████▉| 6414/6470 [1:13:48<00:39,  1.41it/s]
2|6415|Loss: 0.9861307740211487:  99%|█████████▉| 6414/6470 [1:13:49<00:39,  1.41it/s]
2|6415|Loss: 0.9861307740211487:  99%|█████████▉| 6415/6470 [1:13:49<00:37,  1.48it/s]
2|6416|Loss: 0.9151324033737183:  99%|█████████▉| 6415/6470 [1:13:49<00:37,  1.48it/s]
2|6416|Loss: 0.9151324033737183:  99%|█████████▉| 6416/6470 [1:13:50<00:36,  1.48it/s]
2|6417|Loss: 0.8772107362747192:  99%|█████████▉| 6416/6470 [1:13:50<00:36,  1.48it/s]
2|6417|Loss: 0.8772107362747192:  99%|█████████▉| 6417/6470 [1:13:50<00:35,  1.50it/s]
2|6418|Loss: 0.9431943297386169:  99%|█████████▉| 6417/6470 [1:13:51<00:35,  1.50it/s]
2|6418|Loss: 0.9431943297386169:  99%|█████████▉| 6418/6470 [1:13:51<00:36,  1.44it/s]
2|6419|Loss: 0.8320912718772888:  99%|█████████▉| 6418/6470 [1:13:52<00:36,  1.44it/s]
2|6419|Loss: 0.8320912718772888:  99%|█████████▉| 6419/6470 [1:13:52<00:36,  1.39it/s]
2|6420|Loss: 0.884599506855011:  99%|█████████▉| 6419/6470 [1:13:52<00:36,  1.39it/s] 
2|6420|Loss: 0.884599506855011:  99%|█████████▉| 6420/6470 [1:13:53<00:35,  1.42it/s]
2|6421|Loss: 0.9571815133094788:  99%|█████████▉| 6420/6470 [1:13:53<00:35,  1.42it/s]
2|6421|Loss: 0.9571815133094788:  99%|█████████▉| 6421/6470 [1:13:53<00:34,  1.43it/s]
2|6422|Loss: 0.8833566904067993:  99%|█████████▉| 6421/6470 [1:13:54<00:34,  1.43it/s]
2|6422|Loss: 0.8833566904067993:  99%|█████████▉| 6422/6470 [1:13:54<00:33,  1.43it/s]
2|6423|Loss: 0.8152651190757751:  99%|█████████▉| 6422/6470 [1:13:54<00:33,  1.43it/s]
2|6423|Loss: 0.8152651190757751:  99%|█████████▉| 6423/6470 [1:13:55<00:34,  1.35it/s]
2|6424|Loss: 0.8408586978912354:  99%|█████████▉| 6423/6470 [1:13:55<00:34,  1.35it/s]
2|6424|Loss: 0.8408586978912354:  99%|█████████▉| 6424/6470 [1:13:55<00:32,  1.42it/s]
2|6425|Loss: 0.8276510834693909:  99%|█████████▉| 6424/6470 [1:13:56<00:32,  1.42it/s]
2|6425|Loss: 0.8276510834693909:  99%|█████████▉| 6425/6470 [1:13:56<00:32,  1.38it/s]
2|6426|Loss: 0.8548231720924377:  99%|█████████▉| 6425/6470 [1:13:57<00:32,  1.38it/s]
2|6426|Loss: 0.8548231720924377:  99%|█████████▉| 6426/6470 [1:13:57<00:32,  1.35it/s]
2|6427|Loss: 0.8342987298965454:  99%|█████████▉| 6426/6470 [1:13:57<00:32,  1.35it/s]
2|6427|Loss: 0.8342987298965454:  99%|█████████▉| 6427/6470 [1:13:58<00:30,  1.39it/s]
2|6428|Loss: 0.9876839518547058:  99%|█████████▉| 6427/6470 [1:13:58<00:30,  1.39it/s]
2|6428|Loss: 0.9876839518547058:  99%|█████████▉| 6428/6470 [1:13:58<00:30,  1.40it/s]
2|6429|Loss: 0.9380698204040527:  99%|█████████▉| 6428/6470 [1:13:59<00:30,  1.40it/s]
2|6429|Loss: 0.9380698204040527:  99%|█████████▉| 6429/6470 [1:13:59<00:28,  1.44it/s]
2|6430|Loss: 0.9086444973945618:  99%|█████████▉| 6429/6470 [1:13:59<00:28,  1.44it/s]
2|6430|Loss: 0.9086444973945618:  99%|█████████▉| 6430/6470 [1:14:00<00:26,  1.49it/s]
2|6431|Loss: 0.8699305057525635:  99%|█████████▉| 6430/6470 [1:14:00<00:26,  1.49it/s]
2|6431|Loss: 0.8699305057525635:  99%|█████████▉| 6431/6470 [1:14:00<00:26,  1.48it/s]
2|6432|Loss: 0.8693887591362:  99%|█████████▉| 6431/6470 [1:14:01<00:26,  1.48it/s]   
2|6432|Loss: 0.8693887591362:  99%|█████████▉| 6432/6470 [1:14:01<00:24,  1.54it/s]
2|6433|Loss: 0.7438454627990723:  99%|█████████▉| 6432/6470 [1:14:01<00:24,  1.54it/s]
2|6433|Loss: 0.7438454627990723:  99%|█████████▉| 6433/6470 [1:14:02<00:23,  1.56it/s]
2|6434|Loss: 0.8917336463928223:  99%|█████████▉| 6433/6470 [1:14:02<00:23,  1.56it/s]
2|6434|Loss: 0.8917336463928223:  99%|█████████▉| 6434/6470 [1:14:02<00:24,  1.47it/s]
2|6435|Loss: 0.8357610702514648:  99%|█████████▉| 6434/6470 [1:14:03<00:24,  1.47it/s]
2|6435|Loss: 0.8357610702514648:  99%|█████████▉| 6435/6470 [1:14:03<00:25,  1.36it/s]
2|6436|Loss: 0.7544050216674805:  99%|█████████▉| 6435/6470 [1:14:04<00:25,  1.36it/s]
2|6436|Loss: 0.7544050216674805:  99%|█████████▉| 6436/6470 [1:14:04<00:24,  1.41it/s]
2|6437|Loss: 0.7907702922821045:  99%|█████████▉| 6436/6470 [1:14:04<00:24,  1.41it/s]
2|6437|Loss: 0.7907702922821045:  99%|█████████▉| 6437/6470 [1:14:04<00:22,  1.47it/s]
2|6438|Loss: 0.9322279095649719:  99%|█████████▉| 6437/6470 [1:14:05<00:22,  1.47it/s]
2|6438|Loss: 0.9322279095649719: 100%|█████████▉| 6438/6470 [1:14:05<00:22,  1.40it/s]
2|6439|Loss: 0.9852831959724426: 100%|█████████▉| 6438/6470 [1:14:06<00:22,  1.40it/s]
2|6439|Loss: 0.9852831959724426: 100%|█████████▉| 6439/6470 [1:14:06<00:23,  1.32it/s]
2|6440|Loss: 0.7157201766967773: 100%|█████████▉| 6439/6470 [1:14:06<00:23,  1.32it/s]
2|6440|Loss: 0.7157201766967773: 100%|█████████▉| 6440/6470 [1:14:07<00:22,  1.36it/s]
2|6441|Loss: 0.9875970482826233: 100%|█████████▉| 6440/6470 [1:14:07<00:22,  1.36it/s]
2|6441|Loss: 0.9875970482826233: 100%|█████████▉| 6441/6470 [1:14:08<00:22,  1.31it/s]
2|6442|Loss: 0.7748650908470154: 100%|█████████▉| 6441/6470 [1:14:08<00:22,  1.31it/s]
2|6442|Loss: 0.7748650908470154: 100%|█████████▉| 6442/6470 [1:14:08<00:20,  1.36it/s]
2|6443|Loss: 0.8863523006439209: 100%|█████████▉| 6442/6470 [1:14:09<00:20,  1.36it/s]
2|6443|Loss: 0.8863523006439209: 100%|█████████▉| 6443/6470 [1:14:09<00:19,  1.37it/s]
2|6444|Loss: 0.8069586753845215: 100%|█████████▉| 6443/6470 [1:14:09<00:19,  1.37it/s]
2|6444|Loss: 0.8069586753845215: 100%|█████████▉| 6444/6470 [1:14:10<00:18,  1.41it/s]
2|6445|Loss: 0.947068989276886: 100%|█████████▉| 6444/6470 [1:14:10<00:18,  1.41it/s] 
2|6445|Loss: 0.947068989276886: 100%|█████████▉| 6445/6470 [1:14:10<00:17,  1.47it/s]
2|6446|Loss: 0.9826352000236511: 100%|█████████▉| 6445/6470 [1:14:11<00:17,  1.47it/s]
2|6446|Loss: 0.9826352000236511: 100%|█████████▉| 6446/6470 [1:14:11<00:16,  1.44it/s]
2|6447|Loss: 0.7678338885307312: 100%|█████████▉| 6446/6470 [1:14:11<00:16,  1.44it/s]
2|6447|Loss: 0.7678338885307312: 100%|█████████▉| 6447/6470 [1:14:12<00:15,  1.49it/s]
2|6448|Loss: 0.9138094186782837: 100%|█████████▉| 6447/6470 [1:14:12<00:15,  1.49it/s]
2|6448|Loss: 0.9138094186782837: 100%|█████████▉| 6448/6470 [1:14:12<00:15,  1.38it/s]
2|6449|Loss: 0.817288339138031: 100%|█████████▉| 6448/6470 [1:14:13<00:15,  1.38it/s] 
2|6449|Loss: 0.817288339138031: 100%|█████████▉| 6449/6470 [1:14:13<00:15,  1.39it/s]
2|6450|Loss: 0.9396461844444275: 100%|█████████▉| 6449/6470 [1:14:14<00:15,  1.39it/s]
2|6450|Loss: 0.9396461844444275: 100%|█████████▉| 6450/6470 [1:14:14<00:14,  1.39it/s]
2|6451|Loss: 0.7648702263832092: 100%|█████████▉| 6450/6470 [1:14:14<00:14,  1.39it/s]
2|6451|Loss: 0.7648702263832092: 100%|█████████▉| 6451/6470 [1:14:15<00:14,  1.32it/s]
2|6452|Loss: 0.7761071920394897: 100%|█████████▉| 6451/6470 [1:14:15<00:14,  1.32it/s]
2|6452|Loss: 0.7761071920394897: 100%|█████████▉| 6452/6470 [1:14:15<00:13,  1.35it/s]
2|6453|Loss: 0.6691471934318542: 100%|█████████▉| 6452/6470 [1:14:16<00:13,  1.35it/s]
2|6453|Loss: 0.6691471934318542: 100%|█████████▉| 6453/6470 [1:14:16<00:12,  1.41it/s]
2|6454|Loss: 1.0263279676437378: 100%|█████████▉| 6453/6470 [1:14:16<00:12,  1.41it/s]
2|6454|Loss: 1.0263279676437378: 100%|█████████▉| 6454/6470 [1:14:17<00:10,  1.51it/s]
2|6455|Loss: 0.8128522038459778: 100%|█████████▉| 6454/6470 [1:14:17<00:10,  1.51it/s]
2|6455|Loss: 0.8128522038459778: 100%|█████████▉| 6455/6470 [1:14:17<00:10,  1.42it/s]
2|6456|Loss: 0.8206058740615845: 100%|█████████▉| 6455/6470 [1:14:18<00:10,  1.42it/s]
2|6456|Loss: 0.8206058740615845: 100%|█████████▉| 6456/6470 [1:14:18<00:10,  1.39it/s]
2|6457|Loss: 0.806088387966156: 100%|█████████▉| 6456/6470 [1:14:18<00:10,  1.39it/s] 
2|6457|Loss: 0.806088387966156: 100%|█████████▉| 6457/6470 [1:14:19<00:08,  1.57it/s]
2|6458|Loss: 0.9044293165206909: 100%|█████████▉| 6457/6470 [1:14:19<00:08,  1.57it/s]
2|6458|Loss: 0.9044293165206909: 100%|█████████▉| 6458/6470 [1:14:19<00:07,  1.67it/s]
2|6459|Loss: 0.7963762283325195: 100%|█████████▉| 6458/6470 [1:14:19<00:07,  1.67it/s]
2|6459|Loss: 0.7963762283325195: 100%|█████████▉| 6459/6470 [1:14:20<00:07,  1.55it/s]
2|6460|Loss: 0.9967835545539856: 100%|█████████▉| 6459/6470 [1:14:20<00:07,  1.55it/s]
2|6460|Loss: 0.9967835545539856: 100%|█████████▉| 6460/6470 [1:14:21<00:06,  1.53it/s]
2|6461|Loss: 0.7564560770988464: 100%|█████████▉| 6460/6470 [1:14:21<00:06,  1.53it/s]
2|6461|Loss: 0.7564560770988464: 100%|█████████▉| 6461/6470 [1:14:21<00:05,  1.53it/s]
2|6462|Loss: 0.9337231516838074: 100%|█████████▉| 6461/6470 [1:14:22<00:05,  1.53it/s]
2|6462|Loss: 0.9337231516838074: 100%|█████████▉| 6462/6470 [1:14:22<00:05,  1.44it/s]
2|6463|Loss: 0.8584589958190918: 100%|█████████▉| 6462/6470 [1:14:22<00:05,  1.44it/s]
2|6463|Loss: 0.8584589958190918: 100%|█████████▉| 6463/6470 [1:14:23<00:05,  1.38it/s]
2|6464|Loss: 0.748441755771637: 100%|█████████▉| 6463/6470 [1:14:23<00:05,  1.38it/s] 
2|6464|Loss: 0.748441755771637: 100%|█████████▉| 6464/6470 [1:14:24<00:04,  1.36it/s]
2|6465|Loss: 0.866754412651062: 100%|█████████▉| 6464/6470 [1:14:24<00:04,  1.36it/s]
2|6465|Loss: 0.866754412651062: 100%|█████████▉| 6465/6470 [1:14:24<00:03,  1.40it/s]
2|6466|Loss: 0.9412911534309387: 100%|█████████▉| 6465/6470 [1:14:25<00:03,  1.40it/s]
2|6466|Loss: 0.9412911534309387: 100%|█████████▉| 6466/6470 [1:14:25<00:02,  1.36it/s]
2|6467|Loss: 0.8394758701324463: 100%|█████████▉| 6466/6470 [1:14:25<00:02,  1.36it/s]
2|6467|Loss: 0.8394758701324463: 100%|█████████▉| 6467/6470 [1:14:26<00:02,  1.29it/s]
2|6468|Loss: 0.9869176745414734: 100%|█████████▉| 6467/6470 [1:14:26<00:02,  1.29it/s]
2|6468|Loss: 0.9869176745414734: 100%|█████████▉| 6468/6470 [1:14:27<00:01,  1.31it/s]
2|6469|Loss: 1.0414140224456787: 100%|█████████▉| 6468/6470 [1:14:27<00:01,  1.31it/s]
2|6469|Loss: 1.0414140224456787: 100%|█████████▉| 6469/6470 [1:14:27<00:00,  1.27it/s]
2|6470|Loss: 0.8677141070365906: 100%|█████████▉| 6469/6470 [1:14:28<00:00,  1.27it/s]
2|6470|Loss: 0.8677141070365906: 100%|██████████| 6470/6470 [1:14:28<00:00,  1.30it/s]
2|6470|Loss: 0.8677141070365906: 100%|██████████| 6470/6470 [1:14:28<00:00,  1.45it/s]
Model checkpoint of size 25705 MB saved to output/alpaca-llama2-baseline/model_1_12940.ckpt

  0%|          | 0/6470 [00:00<?, ?it/s]
3|1|Loss: 0.8025644421577454:   0%|          | 0/6470 [00:00<?, ?it/s]
3|1|Loss: 0.8025644421577454:   0%|          | 1/6470 [00:00<1:01:57,  1.74it/s]
3|2|Loss: 0.8867295384407043:   0%|          | 1/6470 [00:00<1:01:57,  1.74it/s]
3|2|Loss: 0.8867295384407043:   0%|          | 2/6470 [00:01<1:07:06,  1.61it/s]
3|3|Loss: 0.7995667457580566:   0%|          | 2/6470 [00:01<1:07:06,  1.61it/s]
3|3|Loss: 0.7995667457580566:   0%|          | 3/6470 [00:02<1:23:45,  1.29it/s]
3|4|Loss: 0.7501549124717712:   0%|          | 3/6470 [00:02<1:23:45,  1.29it/s]
3|4|Loss: 0.7501549124717712:   0%|          | 4/6470 [00:02<1:17:03,  1.40it/s]
3|5|Loss: 0.8112979531288147:   0%|          | 4/6470 [00:03<1:17:03,  1.40it/s]
3|5|Loss: 0.8112979531288147:   0%|          | 5/6470 [00:03<1:21:19,  1.32it/s]
3|6|Loss: 0.7725906372070312:   0%|          | 5/6470 [00:04<1:21:19,  1.32it/s]
3|6|Loss: 0.7725906372070312:   0%|          | 6/6470 [00:04<1:23:26,  1.29it/s]
3|7|Loss: 0.6957234144210815:   0%|          | 6/6470 [00:04<1:23:26,  1.29it/s]
3|7|Loss: 0.6957234144210815:   0%|          | 7/6470 [00:05<1:15:58,  1.42it/s]
3|8|Loss: 0.9015733599662781:   0%|          | 7/6470 [00:05<1:15:58,  1.42it/s]
3|8|Loss: 0.9015733599662781:   0%|          | 8/6470 [00:05<1:15:06,  1.43it/s]
3|9|Loss: 0.8708434700965881:   0%|          | 8/6470 [00:06<1:15:06,  1.43it/s]
3|9|Loss: 0.8708434700965881:   0%|          | 9/6470 [00:06<1:15:26,  1.43it/s]
3|10|Loss: 0.7941644191741943:   0%|          | 9/6470 [00:06<1:15:26,  1.43it/s]
3|10|Loss: 0.7941644191741943:   0%|          | 10/6470 [00:07<1:16:27,  1.41it/s]
3|11|Loss: 0.7807738184928894:   0%|          | 10/6470 [00:07<1:16:27,  1.41it/s]
3|11|Loss: 0.7807738184928894:   0%|          | 11/6470 [00:07<1:13:00,  1.47it/s]
3|12|Loss: 1.0330477952957153:   0%|          | 11/6470 [00:08<1:13:00,  1.47it/s]
3|12|Loss: 1.0330477952957153:   0%|          | 12/6470 [00:08<1:08:47,  1.56it/s]
3|13|Loss: 0.7629852890968323:   0%|          | 12/6470 [00:08<1:08:47,  1.56it/s]
3|13|Loss: 0.7629852890968323:   0%|          | 13/6470 [00:08<1:03:13,  1.70it/s]
3|14|Loss: 0.6541407704353333:   0%|          | 13/6470 [00:09<1:03:13,  1.70it/s]
3|14|Loss: 0.6541407704353333:   0%|          | 14/6470 [00:09<1:04:44,  1.66it/s]
3|15|Loss: 0.7278057932853699:   0%|          | 14/6470 [00:09<1:04:44,  1.66it/s]
3|15|Loss: 0.7278057932853699:   0%|          | 15/6470 [00:10<1:14:58,  1.43it/s]
3|16|Loss: 0.8674522638320923:   0%|          | 15/6470 [00:10<1:14:58,  1.43it/s]
3|16|Loss: 0.8674522638320923:   0%|          | 16/6470 [00:11<1:15:35,  1.42it/s]
3|17|Loss: 0.8643494844436646:   0%|          | 16/6470 [00:11<1:15:35,  1.42it/s]
3|17|Loss: 0.8643494844436646:   0%|          | 17/6470 [00:11<1:11:39,  1.50it/s]
3|18|Loss: 0.9158749580383301:   0%|          | 17/6470 [00:11<1:11:39,  1.50it/s]
3|18|Loss: 0.9158749580383301:   0%|          | 18/6470 [00:12<1:11:46,  1.50it/s]
3|19|Loss: 0.745514452457428:   0%|          | 18/6470 [00:12<1:11:46,  1.50it/s] 
3|19|Loss: 0.745514452457428:   0%|          | 19/6470 [00:12<1:13:31,  1.46it/s]
3|20|Loss: 0.8155548572540283:   0%|          | 19/6470 [00:13<1:13:31,  1.46it/s]
3|20|Loss: 0.8155548572540283:   0%|          | 20/6470 [00:13<1:14:38,  1.44it/s]
3|21|Loss: 0.7632969617843628:   0%|          | 20/6470 [00:14<1:14:38,  1.44it/s]
3|21|Loss: 0.7632969617843628:   0%|          | 21/6470 [00:14<1:11:32,  1.50it/s]
3|22|Loss: 0.8129178285598755:   0%|          | 21/6470 [00:14<1:11:32,  1.50it/s]
3|22|Loss: 0.8129178285598755:   0%|          | 22/6470 [00:15<1:13:07,  1.47it/s]
3|23|Loss: 0.8636367321014404:   0%|          | 22/6470 [00:15<1:13:07,  1.47it/s]
3|23|Loss: 0.8636367321014404:   0%|          | 23/6470 [00:15<1:11:50,  1.50it/s]
3|24|Loss: 0.9155308604240417:   0%|          | 23/6470 [00:15<1:11:50,  1.50it/s]
3|24|Loss: 0.9155308604240417:   0%|          | 24/6470 [00:16<1:07:11,  1.60it/s]
3|25|Loss: 0.8999413251876831:   0%|          | 24/6470 [00:16<1:07:11,  1.60it/s]
3|25|Loss: 0.8999413251876831:   0%|          | 25/6470 [00:16<1:07:57,  1.58it/s]
3|26|Loss: 0.7893534302711487:   0%|          | 25/6470 [00:17<1:07:57,  1.58it/s]
3|26|Loss: 0.7893534302711487:   0%|          | 26/6470 [00:17<1:14:36,  1.44it/s]
3|27|Loss: 0.801353394985199:   0%|          | 26/6470 [00:18<1:14:36,  1.44it/s] 
3|27|Loss: 0.801353394985199:   0%|          | 27/6470 [00:18<1:11:06,  1.51it/s]
3|28|Loss: 0.7917906641960144:   0%|          | 27/6470 [00:18<1:11:06,  1.51it/s]
3|28|Loss: 0.7917906641960144:   0%|          | 28/6470 [00:18<1:11:05,  1.51it/s]
3|29|Loss: 0.9126035571098328:   0%|          | 28/6470 [00:19<1:11:05,  1.51it/s]
3|29|Loss: 0.9126035571098328:   0%|          | 29/6470 [00:19<1:14:43,  1.44it/s]
3|30|Loss: 0.9330308437347412:   0%|          | 29/6470 [00:20<1:14:43,  1.44it/s]
3|30|Loss: 0.9330308437347412:   0%|          | 30/6470 [00:20<1:15:31,  1.42it/s]
3|31|Loss: 0.8162426948547363:   0%|          | 30/6470 [00:20<1:15:31,  1.42it/s]
3|31|Loss: 0.8162426948547363:   0%|          | 31/6470 [00:21<1:16:06,  1.41it/s]
3|32|Loss: 1.117655873298645:   0%|          | 31/6470 [00:21<1:16:06,  1.41it/s] 
3|32|Loss: 1.117655873298645:   0%|          | 32/6470 [00:21<1:18:24,  1.37it/s]
3|33|Loss: 0.8561909198760986:   0%|          | 32/6470 [00:22<1:18:24,  1.37it/s]
3|33|Loss: 0.8561909198760986:   1%|          | 33/6470 [00:22<1:18:01,  1.37it/s]
3|34|Loss: 0.7228312492370605:   1%|          | 33/6470 [00:23<1:18:01,  1.37it/s]
3|34|Loss: 0.7228312492370605:   1%|          | 34/6470 [00:23<1:14:50,  1.43it/s]
3|35|Loss: 0.7787443995475769:   1%|          | 34/6470 [00:23<1:14:50,  1.43it/s]
3|35|Loss: 0.7787443995475769:   1%|          | 35/6470 [00:24<1:16:34,  1.40it/s]
3|36|Loss: 0.7010077238082886:   1%|          | 35/6470 [00:24<1:16:34,  1.40it/s]
3|36|Loss: 0.7010077238082886:   1%|          | 36/6470 [00:24<1:19:40,  1.35it/s]
3|37|Loss: 0.8920784592628479:   1%|          | 36/6470 [00:25<1:19:40,  1.35it/s]
3|37|Loss: 0.8920784592628479:   1%|          | 37/6470 [00:25<1:22:04,  1.31it/s]
3|38|Loss: 1.0997602939605713:   1%|          | 37/6470 [00:26<1:22:04,  1.31it/s]
3|38|Loss: 1.0997602939605713:   1%|          | 38/6470 [00:26<1:28:45,  1.21it/s]
3|39|Loss: 0.8654353618621826:   1%|          | 38/6470 [00:27<1:28:45,  1.21it/s]
3|39|Loss: 0.8654353618621826:   1%|          | 39/6470 [00:27<1:24:57,  1.26it/s]
3|40|Loss: 0.9613155126571655:   1%|          | 39/6470 [00:27<1:24:57,  1.26it/s]
3|40|Loss: 0.9613155126571655:   1%|          | 40/6470 [00:27<1:19:16,  1.35it/s]
3|41|Loss: 1.0067085027694702:   1%|          | 40/6470 [00:28<1:19:16,  1.35it/s]
3|41|Loss: 1.0067085027694702:   1%|          | 41/6470 [00:28<1:15:59,  1.41it/s]
3|42|Loss: 1.0695992708206177:   1%|          | 41/6470 [00:28<1:15:59,  1.41it/s]
3|42|Loss: 1.0695992708206177:   1%|          | 42/6470 [00:29<1:11:28,  1.50it/s]
3|43|Loss: 0.8315622210502625:   1%|          | 42/6470 [00:29<1:11:28,  1.50it/s]
3|43|Loss: 0.8315622210502625:   1%|          | 43/6470 [00:29<1:08:22,  1.57it/s]
3|44|Loss: 0.9281262159347534:   1%|          | 43/6470 [00:30<1:08:22,  1.57it/s]
3|44|Loss: 0.9281262159347534:   1%|          | 44/6470 [00:30<1:12:53,  1.47it/s]
3|45|Loss: 0.8400195240974426:   1%|          | 44/6470 [00:30<1:12:53,  1.47it/s]
3|45|Loss: 0.8400195240974426:   1%|          | 45/6470 [00:31<1:15:30,  1.42it/s]
3|46|Loss: 1.020271897315979:   1%|          | 45/6470 [00:31<1:15:30,  1.42it/s] 
3|46|Loss: 1.020271897315979:   1%|          | 46/6470 [00:31<1:13:51,  1.45it/s]
3|47|Loss: 0.8276044130325317:   1%|          | 46/6470 [00:32<1:13:51,  1.45it/s]
3|47|Loss: 0.8276044130325317:   1%|          | 47/6470 [00:32<1:13:40,  1.45it/s]
3|48|Loss: 0.8462290167808533:   1%|          | 47/6470 [00:33<1:13:40,  1.45it/s]
3|48|Loss: 0.8462290167808533:   1%|          | 48/6470 [00:33<1:15:03,  1.43it/s]
3|49|Loss: 0.9149571061134338:   1%|          | 48/6470 [00:33<1:15:03,  1.43it/s]
3|49|Loss: 0.9149571061134338:   1%|          | 49/6470 [00:33<1:11:23,  1.50it/s]
3|50|Loss: 1.090711236000061:   1%|          | 49/6470 [00:34<1:11:23,  1.50it/s] 
3|50|Loss: 1.090711236000061:   1%|          | 50/6470 [00:34<1:09:56,  1.53it/s]
3|51|Loss: 0.7717053294181824:   1%|          | 50/6470 [00:34<1:09:56,  1.53it/s]
3|51|Loss: 0.7717053294181824:   1%|          | 51/6470 [00:35<1:11:53,  1.49it/s]
3|52|Loss: 0.789639949798584:   1%|          | 51/6470 [00:35<1:11:53,  1.49it/s] 
3|52|Loss: 0.789639949798584:   1%|          | 52/6470 [00:35<1:11:02,  1.51it/s]
3|53|Loss: 0.8997575640678406:   1%|          | 52/6470 [00:36<1:11:02,  1.51it/s]
3|53|Loss: 0.8997575640678406:   1%|          | 53/6470 [00:36<1:09:35,  1.54it/s]
3|54|Loss: 0.8739726543426514:   1%|          | 53/6470 [00:36<1:09:35,  1.54it/s]
3|54|Loss: 0.8739726543426514:   1%|          | 54/6470 [00:37<1:12:48,  1.47it/s]
3|55|Loss: 0.8408629298210144:   1%|          | 54/6470 [00:37<1:12:48,  1.47it/s]
3|55|Loss: 0.8408629298210144:   1%|          | 55/6470 [00:38<1:15:09,  1.42it/s]
3|56|Loss: 0.6246089339256287:   1%|          | 55/6470 [00:38<1:15:09,  1.42it/s]
3|56|Loss: 0.6246089339256287:   1%|          | 56/6470 [00:38<1:22:01,  1.30it/s]
3|57|Loss: 0.958889901638031:   1%|          | 56/6470 [00:39<1:22:01,  1.30it/s] 
3|57|Loss: 0.958889901638031:   1%|          | 57/6470 [00:39<1:19:25,  1.35it/s]
3|58|Loss: 0.8364598751068115:   1%|          | 57/6470 [00:40<1:19:25,  1.35it/s]
3|58|Loss: 0.8364598751068115:   1%|          | 58/6470 [00:40<1:17:44,  1.37it/s]
3|59|Loss: 0.831338107585907:   1%|          | 58/6470 [00:40<1:17:44,  1.37it/s] 
3|59|Loss: 0.831338107585907:   1%|          | 59/6470 [00:41<1:19:32,  1.34it/s]
3|60|Loss: 0.7864399552345276:   1%|          | 59/6470 [00:41<1:19:32,  1.34it/s]
3|60|Loss: 0.7864399552345276:   1%|          | 60/6470 [00:41<1:17:15,  1.38it/s]
3|61|Loss: 0.9859371781349182:   1%|          | 60/6470 [00:42<1:17:15,  1.38it/s]
3|61|Loss: 0.9859371781349182:   1%|          | 61/6470 [00:42<1:14:30,  1.43it/s]
3|62|Loss: 0.7333967089653015:   1%|          | 61/6470 [00:42<1:14:30,  1.43it/s]
3|62|Loss: 0.7333967089653015:   1%|          | 62/6470 [00:42<1:09:48,  1.53it/s]
3|63|Loss: 0.8948119282722473:   1%|          | 62/6470 [00:43<1:09:48,  1.53it/s]
3|63|Loss: 0.8948119282722473:   1%|          | 63/6470 [00:43<1:08:03,  1.57it/s]
3|64|Loss: 0.875274121761322:   1%|          | 63/6470 [00:43<1:08:03,  1.57it/s] 
3|64|Loss: 0.875274121761322:   1%|          | 64/6470 [00:44<1:10:39,  1.51it/s]
3|65|Loss: 0.7589163780212402:   1%|          | 64/6470 [00:44<1:10:39,  1.51it/s]
3|65|Loss: 0.7589163780212402:   1%|          | 65/6470 [00:45<1:15:59,  1.40it/s]
3|66|Loss: 0.8762858510017395:   1%|          | 65/6470 [00:45<1:15:59,  1.40it/s]
3|66|Loss: 0.8762858510017395:   1%|          | 66/6470 [00:45<1:13:45,  1.45it/s]
3|67|Loss: 0.8609301447868347:   1%|          | 66/6470 [00:46<1:13:45,  1.45it/s]
3|67|Loss: 0.8609301447868347:   1%|          | 67/6470 [00:46<1:10:19,  1.52it/s]
3|68|Loss: 0.8714693188667297:   1%|          | 67/6470 [00:46<1:10:19,  1.52it/s]
3|68|Loss: 0.8714693188667297:   1%|          | 68/6470 [00:47<1:11:37,  1.49it/s]
3|69|Loss: 0.8467450737953186:   1%|          | 68/6470 [00:47<1:11:37,  1.49it/s]
3|69|Loss: 0.8467450737953186:   1%|          | 69/6470 [00:47<1:13:58,  1.44it/s]
3|70|Loss: 0.8889750838279724:   1%|          | 69/6470 [00:48<1:13:58,  1.44it/s]
3|70|Loss: 0.8889750838279724:   1%|          | 70/6470 [00:48<1:15:54,  1.41it/s]
3|71|Loss: 0.7284873127937317:   1%|          | 70/6470 [00:49<1:15:54,  1.41it/s]
3|71|Loss: 0.7284873127937317:   1%|          | 71/6470 [00:49<1:23:39,  1.27it/s]
3|72|Loss: 1.0065412521362305:   1%|          | 71/6470 [00:49<1:23:39,  1.27it/s]
3|72|Loss: 1.0065412521362305:   1%|          | 72/6470 [00:50<1:24:16,  1.27it/s]
3|73|Loss: 0.721939206123352:   1%|          | 72/6470 [00:50<1:24:16,  1.27it/s] 
3|73|Loss: 0.721939206123352:   1%|          | 73/6470 [00:51<1:24:56,  1.26it/s]
3|74|Loss: 0.9224231243133545:   1%|          | 73/6470 [00:51<1:24:56,  1.26it/s]
3|74|Loss: 0.9224231243133545:   1%|          | 74/6470 [00:51<1:25:12,  1.25it/s]
3|75|Loss: 0.8478580117225647:   1%|          | 74/6470 [00:52<1:25:12,  1.25it/s]
3|75|Loss: 0.8478580117225647:   1%|          | 75/6470 [00:52<1:18:56,  1.35it/s]
3|76|Loss: 0.826469361782074:   1%|          | 75/6470 [00:52<1:18:56,  1.35it/s] 
3|76|Loss: 0.826469361782074:   1%|          | 76/6470 [00:53<1:17:59,  1.37it/s]
3|77|Loss: 0.8045516014099121:   1%|          | 76/6470 [00:53<1:17:59,  1.37it/s]
3|77|Loss: 0.8045516014099121:   1%|          | 77/6470 [00:53<1:16:08,  1.40it/s]
3|78|Loss: 0.9833650588989258:   1%|          | 77/6470 [00:54<1:16:08,  1.40it/s]
3|78|Loss: 0.9833650588989258:   1%|          | 78/6470 [00:54<1:16:25,  1.39it/s]
3|79|Loss: 0.7635901570320129:   1%|          | 78/6470 [00:54<1:16:25,  1.39it/s]
3|79|Loss: 0.7635901570320129:   1%|          | 79/6470 [00:55<1:09:48,  1.53it/s]
3|80|Loss: 0.8027730584144592:   1%|          | 79/6470 [00:55<1:09:48,  1.53it/s]
3|80|Loss: 0.8027730584144592:   1%|          | 80/6470 [00:55<1:09:27,  1.53it/s]
3|81|Loss: 0.9079174995422363:   1%|          | 80/6470 [00:56<1:09:27,  1.53it/s]
3|81|Loss: 0.9079174995422363:   1%|▏         | 81/6470 [00:56<1:09:09,  1.54it/s]
3|82|Loss: 1.0393779277801514:   1%|▏         | 81/6470 [00:56<1:09:09,  1.54it/s]
3|82|Loss: 1.0393779277801514:   1%|▏         | 82/6470 [00:57<1:12:29,  1.47it/s]
3|83|Loss: 0.9395301342010498:   1%|▏         | 82/6470 [00:57<1:12:29,  1.47it/s]
3|83|Loss: 0.9395301342010498:   1%|▏         | 83/6470 [00:58<1:17:35,  1.37it/s]
3|84|Loss: 0.8852839469909668:   1%|▏         | 83/6470 [00:58<1:17:35,  1.37it/s]
3|84|Loss: 0.8852839469909668:   1%|▏         | 84/6470 [00:58<1:16:35,  1.39it/s]
3|85|Loss: 0.8733837604522705:   1%|▏         | 84/6470 [00:59<1:16:35,  1.39it/s]
3|85|Loss: 0.8733837604522705:   1%|▏         | 85/6470 [00:59<1:13:17,  1.45it/s]
3|86|Loss: 0.8094688653945923:   1%|▏         | 85/6470 [00:59<1:13:17,  1.45it/s]
3|86|Loss: 0.8094688653945923:   1%|▏         | 86/6470 [01:00<1:15:46,  1.40it/s]
3|87|Loss: 0.8280637860298157:   1%|▏         | 86/6470 [01:00<1:15:46,  1.40it/s]
3|87|Loss: 0.8280637860298157:   1%|▏         | 87/6470 [01:00<1:14:02,  1.44it/s]
3|88|Loss: 0.7610127925872803:   1%|▏         | 87/6470 [01:01<1:14:02,  1.44it/s]
3|88|Loss: 0.7610127925872803:   1%|▏         | 88/6470 [01:01<1:11:30,  1.49it/s]
3|89|Loss: 0.7653381824493408:   1%|▏         | 88/6470 [01:01<1:11:30,  1.49it/s]
3|89|Loss: 0.7653381824493408:   1%|▏         | 89/6470 [01:02<1:11:48,  1.48it/s]
3|90|Loss: 0.8280721306800842:   1%|▏         | 89/6470 [01:02<1:11:48,  1.48it/s]
3|90|Loss: 0.8280721306800842:   1%|▏         | 90/6470 [01:02<1:15:23,  1.41it/s]
3|91|Loss: 0.9597785472869873:   1%|▏         | 90/6470 [01:03<1:15:23,  1.41it/s]
3|91|Loss: 0.9597785472869873:   1%|▏         | 91/6470 [01:03<1:09:41,  1.53it/s]
3|92|Loss: 0.95533686876297:   1%|▏         | 91/6470 [01:03<1:09:41,  1.53it/s]  
3|92|Loss: 0.95533686876297:   1%|▏         | 92/6470 [01:03<1:07:22,  1.58it/s]
3|93|Loss: 0.9867274761199951:   1%|▏         | 92/6470 [01:04<1:07:22,  1.58it/s]
3|93|Loss: 0.9867274761199951:   1%|▏         | 93/6470 [01:04<1:08:57,  1.54it/s]
3|94|Loss: 0.9322819113731384:   1%|▏         | 93/6470 [01:05<1:08:57,  1.54it/s]
3|94|Loss: 0.9322819113731384:   1%|▏         | 94/6470 [01:05<1:10:31,  1.51it/s]
3|95|Loss: 0.8262209296226501:   1%|▏         | 94/6470 [01:05<1:10:31,  1.51it/s]
3|95|Loss: 0.8262209296226501:   1%|▏         | 95/6470 [01:06<1:14:27,  1.43it/s]
3|96|Loss: 0.8142127394676208:   1%|▏         | 95/6470 [01:06<1:14:27,  1.43it/s]
3|96|Loss: 0.8142127394676208:   1%|▏         | 96/6470 [01:06<1:16:34,  1.39it/s]
3|97|Loss: 0.8670098185539246:   1%|▏         | 96/6470 [01:07<1:16:34,  1.39it/s]
3|97|Loss: 0.8670098185539246:   1%|▏         | 97/6470 [01:07<1:14:22,  1.43it/s]
3|98|Loss: 0.9453912973403931:   1%|▏         | 97/6470 [01:07<1:14:22,  1.43it/s]
3|98|Loss: 0.9453912973403931:   2%|▏         | 98/6470 [01:08<1:16:24,  1.39it/s]
3|99|Loss: 0.7967919707298279:   2%|▏         | 98/6470 [01:08<1:16:24,  1.39it/s]
3|99|Loss: 0.7967919707298279:   2%|▏         | 99/6470 [01:09<1:18:45,  1.35it/s]
3|100|Loss: 0.8947305679321289:   2%|▏         | 99/6470 [01:09<1:18:45,  1.35it/s]
3|100|Loss: 0.8947305679321289:   2%|▏         | 100/6470 [01:09<1:18:09,  1.36it/s]
3|101|Loss: 0.9484151601791382:   2%|▏         | 100/6470 [01:10<1:18:09,  1.36it/s]
3|101|Loss: 0.9484151601791382:   2%|▏         | 101/6470 [01:10<1:13:55,  1.44it/s]
3|102|Loss: 0.8335058689117432:   2%|▏         | 101/6470 [01:10<1:13:55,  1.44it/s]
3|102|Loss: 0.8335058689117432:   2%|▏         | 102/6470 [01:11<1:10:08,  1.51it/s]
3|103|Loss: 0.8479660153388977:   2%|▏         | 102/6470 [01:11<1:10:08,  1.51it/s]
3|103|Loss: 0.8479660153388977:   2%|▏         | 103/6470 [01:11<1:13:00,  1.45it/s]
3|104|Loss: 0.887139081954956:   2%|▏         | 103/6470 [01:12<1:13:00,  1.45it/s] 
3|104|Loss: 0.887139081954956:   2%|▏         | 104/6470 [01:12<1:15:14,  1.41it/s]
3|105|Loss: 0.8180887699127197:   2%|▏         | 104/6470 [01:12<1:15:14,  1.41it/s]
3|105|Loss: 0.8180887699127197:   2%|▏         | 105/6470 [01:13<1:12:34,  1.46it/s]
3|106|Loss: 0.9616226553916931:   2%|▏         | 105/6470 [01:13<1:12:34,  1.46it/s]
3|106|Loss: 0.9616226553916931:   2%|▏         | 106/6470 [01:13<1:07:57,  1.56it/s]
3|107|Loss: 0.952282190322876:   2%|▏         | 106/6470 [01:14<1:07:57,  1.56it/s] 
3|107|Loss: 0.952282190322876:   2%|▏         | 107/6470 [01:14<1:05:37,  1.62it/s]
3|108|Loss: 0.8329039216041565:   2%|▏         | 107/6470 [01:14<1:05:37,  1.62it/s]
3|108|Loss: 0.8329039216041565:   2%|▏         | 108/6470 [01:14<1:08:36,  1.55it/s]
3|109|Loss: 0.9947512149810791:   2%|▏         | 108/6470 [01:15<1:08:36,  1.55it/s]
3|109|Loss: 0.9947512149810791:   2%|▏         | 109/6470 [01:15<1:10:45,  1.50it/s]
3|110|Loss: 0.8888758420944214:   2%|▏         | 109/6470 [01:16<1:10:45,  1.50it/s]
3|110|Loss: 0.8888758420944214:   2%|▏         | 110/6470 [01:16<1:06:54,  1.58it/s]
3|111|Loss: 0.895944356918335:   2%|▏         | 110/6470 [01:16<1:06:54,  1.58it/s] 
3|111|Loss: 0.895944356918335:   2%|▏         | 111/6470 [01:16<1:05:50,  1.61it/s]
3|112|Loss: 0.8478407263755798:   2%|▏         | 111/6470 [01:17<1:05:50,  1.61it/s]
3|112|Loss: 0.8478407263755798:   2%|▏         | 112/6470 [01:17<1:09:01,  1.54it/s]
3|113|Loss: 0.8621708154678345:   2%|▏         | 112/6470 [01:17<1:09:01,  1.54it/s]
3|113|Loss: 0.8621708154678345:   2%|▏         | 113/6470 [01:18<1:08:54,  1.54it/s]
3|114|Loss: 0.8199256062507629:   2%|▏         | 113/6470 [01:18<1:08:54,  1.54it/s]
3|114|Loss: 0.8199256062507629:   2%|▏         | 114/6470 [01:18<1:10:40,  1.50it/s]
3|115|Loss: 0.6897166967391968:   2%|▏         | 114/6470 [01:19<1:10:40,  1.50it/s]
3|115|Loss: 0.6897166967391968:   2%|▏         | 115/6470 [01:19<1:10:33,  1.50it/s]
3|116|Loss: 0.9784963726997375:   2%|▏         | 115/6470 [01:19<1:10:33,  1.50it/s]
3|116|Loss: 0.9784963726997375:   2%|▏         | 116/6470 [01:20<1:09:44,  1.52it/s]
3|117|Loss: 0.8701257705688477:   2%|▏         | 116/6470 [01:20<1:09:44,  1.52it/s]
3|117|Loss: 0.8701257705688477:   2%|▏         | 117/6470 [01:20<1:07:47,  1.56it/s]
3|118|Loss: 0.9182336926460266:   2%|▏         | 117/6470 [01:21<1:07:47,  1.56it/s]
3|118|Loss: 0.9182336926460266:   2%|▏         | 118/6470 [01:21<1:06:43,  1.59it/s]
3|119|Loss: 0.8856481909751892:   2%|▏         | 118/6470 [01:21<1:06:43,  1.59it/s]
3|119|Loss: 0.8856481909751892:   2%|▏         | 119/6470 [01:22<1:12:16,  1.46it/s]
3|120|Loss: 0.8825321793556213:   2%|▏         | 119/6470 [01:22<1:12:16,  1.46it/s]
3|120|Loss: 0.8825321793556213:   2%|▏         | 120/6470 [01:23<1:15:09,  1.41it/s]
3|121|Loss: 0.7755618095397949:   2%|▏         | 120/6470 [01:23<1:15:09,  1.41it/s]
3|121|Loss: 0.7755618095397949:   2%|▏         | 121/6470 [01:23<1:15:53,  1.39it/s]
3|122|Loss: 0.8151716589927673:   2%|▏         | 121/6470 [01:24<1:15:53,  1.39it/s]
3|122|Loss: 0.8151716589927673:   2%|▏         | 122/6470 [01:24<1:15:25,  1.40it/s]
3|123|Loss: 0.9000324606895447:   2%|▏         | 122/6470 [01:24<1:15:25,  1.40it/s]
3|123|Loss: 0.9000324606895447:   2%|▏         | 123/6470 [01:25<1:13:24,  1.44it/s]
3|124|Loss: 0.8576400876045227:   2%|▏         | 123/6470 [01:25<1:13:24,  1.44it/s]
3|124|Loss: 0.8576400876045227:   2%|▏         | 124/6470 [01:25<1:06:53,  1.58it/s]
3|125|Loss: 1.0725969076156616:   2%|▏         | 124/6470 [01:25<1:06:53,  1.58it/s]
3|125|Loss: 1.0725969076156616:   2%|▏         | 125/6470 [01:26<1:04:02,  1.65it/s]
3|126|Loss: 1.149056315422058:   2%|▏         | 125/6470 [01:26<1:04:02,  1.65it/s] 
3|126|Loss: 1.149056315422058:   2%|▏         | 126/6470 [01:26<1:07:44,  1.56it/s]
3|127|Loss: 0.8331239223480225:   2%|▏         | 126/6470 [01:27<1:07:44,  1.56it/s]
3|127|Loss: 0.8331239223480225:   2%|▏         | 127/6470 [01:27<1:05:36,  1.61it/s]
3|128|Loss: 0.855252742767334:   2%|▏         | 127/6470 [01:27<1:05:36,  1.61it/s] 
3|128|Loss: 0.855252742767334:   2%|▏         | 128/6470 [01:28<1:12:32,  1.46it/s]
3|129|Loss: 0.9050754308700562:   2%|▏         | 128/6470 [01:28<1:12:32,  1.46it/s]
3|129|Loss: 0.9050754308700562:   2%|▏         | 129/6470 [01:29<1:14:26,  1.42it/s]
3|130|Loss: 0.7969111800193787:   2%|▏         | 129/6470 [01:29<1:14:26,  1.42it/s]
3|130|Loss: 0.7969111800193787:   2%|▏         | 130/6470 [01:29<1:19:47,  1.32it/s]
3|131|Loss: 0.9575899839401245:   2%|▏         | 130/6470 [01:30<1:19:47,  1.32it/s]
3|131|Loss: 0.9575899839401245:   2%|▏         | 131/6470 [01:30<1:17:30,  1.36it/s]
3|132|Loss: 0.8682932257652283:   2%|▏         | 131/6470 [01:30<1:17:30,  1.36it/s]
3|132|Loss: 0.8682932257652283:   2%|▏         | 132/6470 [01:31<1:13:59,  1.43it/s]
3|133|Loss: 0.9330494403839111:   2%|▏         | 132/6470 [01:31<1:13:59,  1.43it/s]
3|133|Loss: 0.9330494403839111:   2%|▏         | 133/6470 [01:31<1:07:20,  1.57it/s]
3|134|Loss: 0.7847461700439453:   2%|▏         | 133/6470 [01:31<1:07:20,  1.57it/s]
3|134|Loss: 0.7847461700439453:   2%|▏         | 134/6470 [01:32<1:02:56,  1.68it/s]
3|135|Loss: 0.9112280607223511:   2%|▏         | 134/6470 [01:32<1:02:56,  1.68it/s]
3|135|Loss: 0.9112280607223511:   2%|▏         | 135/6470 [01:32<1:03:33,  1.66it/s]
3|136|Loss: 0.700175940990448:   2%|▏         | 135/6470 [01:33<1:03:33,  1.66it/s] 
3|136|Loss: 0.700175940990448:   2%|▏         | 136/6470 [01:33<1:04:55,  1.63it/s]
3|137|Loss: 0.7730169892311096:   2%|▏         | 136/6470 [01:33<1:04:55,  1.63it/s]
3|137|Loss: 0.7730169892311096:   2%|▏         | 137/6470 [01:34<1:11:17,  1.48it/s]
3|138|Loss: 0.7579233646392822:   2%|▏         | 137/6470 [01:34<1:11:17,  1.48it/s]
3|138|Loss: 0.7579233646392822:   2%|▏         | 138/6470 [01:35<1:15:37,  1.40it/s]
3|139|Loss: 0.8639974594116211:   2%|▏         | 138/6470 [01:35<1:15:37,  1.40it/s]
3|139|Loss: 0.8639974594116211:   2%|▏         | 139/6470 [01:35<1:14:16,  1.42it/s]
3|140|Loss: 0.8924257755279541:   2%|▏         | 139/6470 [01:36<1:14:16,  1.42it/s]
3|140|Loss: 0.8924257755279541:   2%|▏         | 140/6470 [01:36<1:10:30,  1.50it/s]
3|141|Loss: 0.721642017364502:   2%|▏         | 140/6470 [01:36<1:10:30,  1.50it/s] 
3|141|Loss: 0.721642017364502:   2%|▏         | 141/6470 [01:37<1:13:55,  1.43it/s]
3|142|Loss: 1.0891557931900024:   2%|▏         | 141/6470 [01:37<1:13:55,  1.43it/s]
3|142|Loss: 1.0891557931900024:   2%|▏         | 142/6470 [01:37<1:14:53,  1.41it/s]
3|143|Loss: 0.7806432247161865:   2%|▏         | 142/6470 [01:38<1:14:53,  1.41it/s]
3|143|Loss: 0.7806432247161865:   2%|▏         | 143/6470 [01:38<1:13:05,  1.44it/s]
3|144|Loss: 0.8383587002754211:   2%|▏         | 143/6470 [01:38<1:13:05,  1.44it/s]
3|144|Loss: 0.8383587002754211:   2%|▏         | 144/6470 [01:39<1:13:41,  1.43it/s]
3|145|Loss: 0.8726257681846619:   2%|▏         | 144/6470 [01:39<1:13:41,  1.43it/s]
3|145|Loss: 0.8726257681846619:   2%|▏         | 145/6470 [01:39<1:10:49,  1.49it/s]
3|146|Loss: 0.8642048239707947:   2%|▏         | 145/6470 [01:40<1:10:49,  1.49it/s]
3|146|Loss: 0.8642048239707947:   2%|▏         | 146/6470 [01:40<1:04:46,  1.63it/s]
3|147|Loss: 0.8369536995887756:   2%|▏         | 146/6470 [01:40<1:04:46,  1.63it/s]
3|147|Loss: 0.8369536995887756:   2%|▏         | 147/6470 [01:40<1:01:02,  1.73it/s]
3|148|Loss: 0.8379501104354858:   2%|▏         | 147/6470 [01:41<1:01:02,  1.73it/s]
3|148|Loss: 0.8379501104354858:   2%|▏         | 148/6470 [01:41<1:04:48,  1.63it/s]
3|149|Loss: 0.7153027057647705:   2%|▏         | 148/6470 [01:41<1:04:48,  1.63it/s]
3|149|Loss: 0.7153027057647705:   2%|▏         | 149/6470 [01:42<1:05:50,  1.60it/s]
3|150|Loss: 0.744930624961853:   2%|▏         | 149/6470 [01:42<1:05:50,  1.60it/s] 
3|150|Loss: 0.744930624961853:   2%|▏         | 150/6470 [01:42<1:04:28,  1.63it/s]
3|151|Loss: 0.9001057744026184:   2%|▏         | 150/6470 [01:43<1:04:28,  1.63it/s]
3|151|Loss: 0.9001057744026184:   2%|▏         | 151/6470 [01:43<1:06:32,  1.58it/s]
3|152|Loss: 0.8492603302001953:   2%|▏         | 151/6470 [01:43<1:06:32,  1.58it/s]
3|152|Loss: 0.8492603302001953:   2%|▏         | 152/6470 [01:44<1:09:27,  1.52it/s]
3|153|Loss: 0.8472480773925781:   2%|▏         | 152/6470 [01:44<1:09:27,  1.52it/s]
3|153|Loss: 0.8472480773925781:   2%|▏         | 153/6470 [01:44<1:05:27,  1.61it/s]
3|154|Loss: 0.7901546359062195:   2%|▏         | 153/6470 [01:45<1:05:27,  1.61it/s]
3|154|Loss: 0.7901546359062195:   2%|▏         | 154/6470 [01:45<1:07:44,  1.55it/s]
3|155|Loss: 0.7699115872383118:   2%|▏         | 154/6470 [01:45<1:07:44,  1.55it/s]
3|155|Loss: 0.7699115872383118:   2%|▏         | 155/6470 [01:45<1:03:49,  1.65it/s]
3|156|Loss: 0.6446107029914856:   2%|▏         | 155/6470 [01:46<1:03:49,  1.65it/s]
3|156|Loss: 0.6446107029914856:   2%|▏         | 156/6470 [01:46<1:08:48,  1.53it/s]
3|157|Loss: 0.8109052777290344:   2%|▏         | 156/6470 [01:47<1:08:48,  1.53it/s]
3|157|Loss: 0.8109052777290344:   2%|▏         | 157/6470 [01:47<1:05:19,  1.61it/s]
3|158|Loss: 0.7049790024757385:   2%|▏         | 157/6470 [01:47<1:05:19,  1.61it/s]
3|158|Loss: 0.7049790024757385:   2%|▏         | 158/6470 [01:47<1:03:02,  1.67it/s]
3|159|Loss: 0.8461975455284119:   2%|▏         | 158/6470 [01:48<1:03:02,  1.67it/s]
3|159|Loss: 0.8461975455284119:   2%|▏         | 159/6470 [01:48<1:06:28,  1.58it/s]
3|160|Loss: 0.8749237656593323:   2%|▏         | 159/6470 [01:48<1:06:28,  1.58it/s]
3|160|Loss: 0.8749237656593323:   2%|▏         | 160/6470 [01:49<1:10:04,  1.50it/s]
3|161|Loss: 0.923115074634552:   2%|▏         | 160/6470 [01:49<1:10:04,  1.50it/s] 
3|161|Loss: 0.923115074634552:   2%|▏         | 161/6470 [01:49<1:14:13,  1.42it/s]
3|162|Loss: 0.7230648994445801:   2%|▏         | 161/6470 [01:50<1:14:13,  1.42it/s]
3|162|Loss: 0.7230648994445801:   3%|▎         | 162/6470 [01:50<1:21:18,  1.29it/s]
3|163|Loss: 0.955030083656311:   3%|▎         | 162/6470 [01:51<1:21:18,  1.29it/s] 
3|163|Loss: 0.955030083656311:   3%|▎         | 163/6470 [01:51<1:17:38,  1.35it/s]
3|164|Loss: 0.858593225479126:   3%|▎         | 163/6470 [01:51<1:17:38,  1.35it/s]
3|164|Loss: 0.858593225479126:   3%|▎         | 164/6470 [01:52<1:13:49,  1.42it/s]
3|165|Loss: 0.9121506214141846:   3%|▎         | 164/6470 [01:52<1:13:49,  1.42it/s]
3|165|Loss: 0.9121506214141846:   3%|▎         | 165/6470 [01:52<1:10:20,  1.49it/s]
3|166|Loss: 0.9578447341918945:   3%|▎         | 165/6470 [01:53<1:10:20,  1.49it/s]
3|166|Loss: 0.9578447341918945:   3%|▎         | 166/6470 [01:53<1:04:26,  1.63it/s]
3|167|Loss: 0.7668138146400452:   3%|▎         | 166/6470 [01:53<1:04:26,  1.63it/s]
3|167|Loss: 0.7668138146400452:   3%|▎         | 167/6470 [01:53<1:00:37,  1.73it/s]
3|168|Loss: 0.9731550216674805:   3%|▎         | 167/6470 [01:54<1:00:37,  1.73it/s]
3|168|Loss: 0.9731550216674805:   3%|▎         | 168/6470 [01:54<1:04:25,  1.63it/s]
3|169|Loss: 1.0145493745803833:   3%|▎         | 168/6470 [01:54<1:04:25,  1.63it/s]
3|169|Loss: 1.0145493745803833:   3%|▎         | 169/6470 [01:55<1:10:43,  1.48it/s]
3|170|Loss: 0.7734414339065552:   3%|▎         | 169/6470 [01:55<1:10:43,  1.48it/s]
3|170|Loss: 0.7734414339065552:   3%|▎         | 170/6470 [01:56<1:12:30,  1.45it/s]
3|171|Loss: 0.810823917388916:   3%|▎         | 170/6470 [01:56<1:12:30,  1.45it/s] 
3|171|Loss: 0.810823917388916:   3%|▎         | 171/6470 [01:56<1:12:58,  1.44it/s]
3|172|Loss: 0.8321633338928223:   3%|▎         | 171/6470 [01:57<1:12:58,  1.44it/s]
3|172|Loss: 0.8321633338928223:   3%|▎         | 172/6470 [01:57<1:13:13,  1.43it/s]
3|173|Loss: 0.9099122285842896:   3%|▎         | 172/6470 [01:57<1:13:13,  1.43it/s]
3|173|Loss: 0.9099122285842896:   3%|▎         | 173/6470 [01:58<1:15:54,  1.38it/s]
3|174|Loss: 0.8489805459976196:   3%|▎         | 173/6470 [01:58<1:15:54,  1.38it/s]
3|174|Loss: 0.8489805459976196:   3%|▎         | 174/6470 [01:58<1:15:52,  1.38it/s]
3|175|Loss: 0.8380398154258728:   3%|▎         | 174/6470 [01:59<1:15:52,  1.38it/s]
3|175|Loss: 0.8380398154258728:   3%|▎         | 175/6470 [01:59<1:12:15,  1.45it/s]
3|176|Loss: 0.7629998922348022:   3%|▎         | 175/6470 [01:59<1:12:15,  1.45it/s]
3|176|Loss: 0.7629998922348022:   3%|▎         | 176/6470 [02:00<1:09:47,  1.50it/s]
3|177|Loss: 0.85439133644104:   3%|▎         | 176/6470 [02:00<1:09:47,  1.50it/s]  
3|177|Loss: 0.85439133644104:   3%|▎         | 177/6470 [02:00<1:15:37,  1.39it/s]
3|178|Loss: 0.8631573915481567:   3%|▎         | 177/6470 [02:01<1:15:37,  1.39it/s]
3|178|Loss: 0.8631573915481567:   3%|▎         | 178/6470 [02:01<1:16:12,  1.38it/s]
3|179|Loss: 0.9015297889709473:   3%|▎         | 178/6470 [02:02<1:16:12,  1.38it/s]
3|179|Loss: 0.9015297889709473:   3%|▎         | 179/6470 [02:02<1:12:11,  1.45it/s]
3|180|Loss: 0.9334485530853271:   3%|▎         | 179/6470 [02:02<1:12:11,  1.45it/s]
3|180|Loss: 0.9334485530853271:   3%|▎         | 180/6470 [02:03<1:15:32,  1.39it/s]
3|181|Loss: 0.7932233810424805:   3%|▎         | 180/6470 [02:03<1:15:32,  1.39it/s]
3|181|Loss: 0.7932233810424805:   3%|▎         | 181/6470 [02:03<1:14:38,  1.40it/s]
3|182|Loss: 1.0273081064224243:   3%|▎         | 181/6470 [02:04<1:14:38,  1.40it/s]
3|182|Loss: 1.0273081064224243:   3%|▎         | 182/6470 [02:04<1:17:09,  1.36it/s]
3|183|Loss: 0.7204064130783081:   3%|▎         | 182/6470 [02:04<1:17:09,  1.36it/s]
3|183|Loss: 0.7204064130783081:   3%|▎         | 183/6470 [02:05<1:13:02,  1.43it/s]
3|184|Loss: 0.9265615344047546:   3%|▎         | 183/6470 [02:05<1:13:02,  1.43it/s]
3|184|Loss: 0.9265615344047546:   3%|▎         | 184/6470 [02:06<1:16:27,  1.37it/s]
3|185|Loss: 0.8733946084976196:   3%|▎         | 184/6470 [02:06<1:16:27,  1.37it/s]
3|185|Loss: 0.8733946084976196:   3%|▎         | 185/6470 [02:06<1:17:29,  1.35it/s]
3|186|Loss: 0.8638542294502258:   3%|▎         | 185/6470 [02:07<1:17:29,  1.35it/s]
3|186|Loss: 0.8638542294502258:   3%|▎         | 186/6470 [02:07<1:17:26,  1.35it/s]
3|187|Loss: 0.8442813158035278:   3%|▎         | 186/6470 [02:07<1:17:26,  1.35it/s]
3|187|Loss: 0.8442813158035278:   3%|▎         | 187/6470 [02:08<1:14:53,  1.40it/s]
3|188|Loss: 0.8463060259819031:   3%|▎         | 187/6470 [02:08<1:14:53,  1.40it/s]
3|188|Loss: 0.8463060259819031:   3%|▎         | 188/6470 [02:08<1:14:02,  1.41it/s]
3|189|Loss: 0.7603330612182617:   3%|▎         | 188/6470 [02:09<1:14:02,  1.41it/s]
3|189|Loss: 0.7603330612182617:   3%|▎         | 189/6470 [02:09<1:16:07,  1.38it/s]
3|190|Loss: 0.8225072026252747:   3%|▎         | 189/6470 [02:10<1:16:07,  1.38it/s]
3|190|Loss: 0.8225072026252747:   3%|▎         | 190/6470 [02:10<1:17:07,  1.36it/s]
3|191|Loss: 0.7694495916366577:   3%|▎         | 190/6470 [02:10<1:17:07,  1.36it/s]
3|191|Loss: 0.7694495916366577:   3%|▎         | 191/6470 [02:11<1:19:28,  1.32it/s]
3|192|Loss: 0.7886473536491394:   3%|▎         | 191/6470 [02:11<1:19:28,  1.32it/s]
3|192|Loss: 0.7886473536491394:   3%|▎         | 192/6470 [02:11<1:14:42,  1.40it/s]
3|193|Loss: 0.8514128923416138:   3%|▎         | 192/6470 [02:12<1:14:42,  1.40it/s]
3|193|Loss: 0.8514128923416138:   3%|▎         | 193/6470 [02:12<1:14:03,  1.41it/s]
3|194|Loss: 0.642105221748352:   3%|▎         | 193/6470 [02:12<1:14:03,  1.41it/s] 
3|194|Loss: 0.642105221748352:   3%|▎         | 194/6470 [02:12<1:05:24,  1.60it/s]
3|195|Loss: 0.8849442601203918:   3%|▎         | 194/6470 [02:13<1:05:24,  1.60it/s]
3|195|Loss: 0.8849442601203918:   3%|▎         | 195/6470 [02:13<1:08:11,  1.53it/s]
3|196|Loss: 0.8849326968193054:   3%|▎         | 195/6470 [02:14<1:08:11,  1.53it/s]
3|196|Loss: 0.8849326968193054:   3%|▎         | 196/6470 [02:14<1:09:24,  1.51it/s]
3|197|Loss: 0.8732026815414429:   3%|▎         | 196/6470 [02:14<1:09:24,  1.51it/s]
3|197|Loss: 0.8732026815414429:   3%|▎         | 197/6470 [02:14<1:08:11,  1.53it/s]
3|198|Loss: 0.8300651907920837:   3%|▎         | 197/6470 [02:15<1:08:11,  1.53it/s]
3|198|Loss: 0.8300651907920837:   3%|▎         | 198/6470 [02:15<1:13:23,  1.42it/s]
3|199|Loss: 0.9766692519187927:   3%|▎         | 198/6470 [02:16<1:13:23,  1.42it/s]
3|199|Loss: 0.9766692519187927:   3%|▎         | 199/6470 [02:16<1:18:37,  1.33it/s]
3|200|Loss: 0.9046062231063843:   3%|▎         | 199/6470 [02:17<1:18:37,  1.33it/s]
3|200|Loss: 0.9046062231063843:   3%|▎         | 200/6470 [02:17<1:18:18,  1.33it/s]
3|201|Loss: 0.7646253108978271:   3%|▎         | 200/6470 [02:17<1:18:18,  1.33it/s]
3|201|Loss: 0.7646253108978271:   3%|▎         | 201/6470 [02:18<1:15:14,  1.39it/s]
3|202|Loss: 0.9960184097290039:   3%|▎         | 201/6470 [02:18<1:15:14,  1.39it/s]
3|202|Loss: 0.9960184097290039:   3%|▎         | 202/6470 [02:18<1:12:32,  1.44it/s]
3|203|Loss: 0.8674038052558899:   3%|▎         | 202/6470 [02:19<1:12:32,  1.44it/s]
3|203|Loss: 0.8674038052558899:   3%|▎         | 203/6470 [02:19<1:07:30,  1.55it/s]
3|204|Loss: 0.8639494776725769:   3%|▎         | 203/6470 [02:19<1:07:30,  1.55it/s]
3|204|Loss: 0.8639494776725769:   3%|▎         | 204/6470 [02:19<1:08:44,  1.52it/s]
3|205|Loss: 1.0201548337936401:   3%|▎         | 204/6470 [02:20<1:08:44,  1.52it/s]
3|205|Loss: 1.0201548337936401:   3%|▎         | 205/6470 [02:20<1:11:07,  1.47it/s]
3|206|Loss: 0.8735259175300598:   3%|▎         | 205/6470 [02:21<1:11:07,  1.47it/s]
3|206|Loss: 0.8735259175300598:   3%|▎         | 206/6470 [02:21<1:09:57,  1.49it/s]
3|207|Loss: 0.9298552870750427:   3%|▎         | 206/6470 [02:21<1:09:57,  1.49it/s]
3|207|Loss: 0.9298552870750427:   3%|▎         | 207/6470 [02:22<1:14:14,  1.41it/s]
3|208|Loss: 0.8678603172302246:   3%|▎         | 207/6470 [02:22<1:14:14,  1.41it/s]
3|208|Loss: 0.8678603172302246:   3%|▎         | 208/6470 [02:22<1:18:47,  1.32it/s]
3|209|Loss: 0.8493521213531494:   3%|▎         | 208/6470 [02:23<1:18:47,  1.32it/s]
3|209|Loss: 0.8493521213531494:   3%|▎         | 209/6470 [02:23<1:16:11,  1.37it/s]
3|210|Loss: 0.8661205172538757:   3%|▎         | 209/6470 [02:24<1:16:11,  1.37it/s]
3|210|Loss: 0.8661205172538757:   3%|▎         | 210/6470 [02:24<1:15:37,  1.38it/s]
3|211|Loss: 0.757642924785614:   3%|▎         | 210/6470 [02:24<1:15:37,  1.38it/s] 
3|211|Loss: 0.757642924785614:   3%|▎         | 211/6470 [02:24<1:10:44,  1.47it/s]
3|212|Loss: 0.9734699130058289:   3%|▎         | 211/6470 [02:25<1:10:44,  1.47it/s]
3|212|Loss: 0.9734699130058289:   3%|▎         | 212/6470 [02:25<1:08:49,  1.52it/s]
3|213|Loss: 0.9676357507705688:   3%|▎         | 212/6470 [02:25<1:08:49,  1.52it/s]
3|213|Loss: 0.9676357507705688:   3%|▎         | 213/6470 [02:26<1:12:21,  1.44it/s]
3|214|Loss: 0.9474179148674011:   3%|▎         | 213/6470 [02:26<1:12:21,  1.44it/s]
3|214|Loss: 0.9474179148674011:   3%|▎         | 214/6470 [02:27<1:13:22,  1.42it/s]
3|215|Loss: 0.9179324507713318:   3%|▎         | 214/6470 [02:27<1:13:22,  1.42it/s]
3|215|Loss: 0.9179324507713318:   3%|▎         | 215/6470 [02:27<1:08:33,  1.52it/s]
3|216|Loss: 0.8264678120613098:   3%|▎         | 215/6470 [02:27<1:08:33,  1.52it/s]
3|216|Loss: 0.8264678120613098:   3%|▎         | 216/6470 [02:28<1:06:38,  1.56it/s]
3|217|Loss: 0.7571268677711487:   3%|▎         | 216/6470 [02:28<1:06:38,  1.56it/s]
3|217|Loss: 0.7571268677711487:   3%|▎         | 217/6470 [02:28<1:05:03,  1.60it/s]
3|218|Loss: 0.8482828736305237:   3%|▎         | 217/6470 [02:29<1:05:03,  1.60it/s]
3|218|Loss: 0.8482828736305237:   3%|▎         | 218/6470 [02:29<1:07:27,  1.54it/s]
3|219|Loss: 0.8758920431137085:   3%|▎         | 218/6470 [02:29<1:07:27,  1.54it/s]
3|219|Loss: 0.8758920431137085:   3%|▎         | 219/6470 [02:30<1:12:23,  1.44it/s]
3|220|Loss: 0.9100200533866882:   3%|▎         | 219/6470 [02:30<1:12:23,  1.44it/s]
3|220|Loss: 0.9100200533866882:   3%|▎         | 220/6470 [02:31<1:13:45,  1.41it/s]
3|221|Loss: 0.7552951574325562:   3%|▎         | 220/6470 [02:31<1:13:45,  1.41it/s]
3|221|Loss: 0.7552951574325562:   3%|▎         | 221/6470 [02:31<1:14:22,  1.40it/s]
3|222|Loss: 1.0446828603744507:   3%|▎         | 221/6470 [02:32<1:14:22,  1.40it/s]
3|222|Loss: 1.0446828603744507:   3%|▎         | 222/6470 [02:32<1:12:05,  1.44it/s]
3|223|Loss: 0.8555960655212402:   3%|▎         | 222/6470 [02:32<1:12:05,  1.44it/s]
3|223|Loss: 0.8555960655212402:   3%|▎         | 223/6470 [02:33<1:14:55,  1.39it/s]
3|224|Loss: 0.8559087514877319:   3%|▎         | 223/6470 [02:33<1:14:55,  1.39it/s]
3|224|Loss: 0.8559087514877319:   3%|▎         | 224/6470 [02:34<1:18:55,  1.32it/s]
3|225|Loss: 0.8059391975402832:   3%|▎         | 224/6470 [02:34<1:18:55,  1.32it/s]
3|225|Loss: 0.8059391975402832:   3%|▎         | 225/6470 [02:34<1:16:48,  1.36it/s]
3|226|Loss: 0.7454943060874939:   3%|▎         | 225/6470 [02:35<1:16:48,  1.36it/s]
3|226|Loss: 0.7454943060874939:   3%|▎         | 226/6470 [02:35<1:10:27,  1.48it/s]
3|227|Loss: 0.8263156414031982:   3%|▎         | 226/6470 [02:35<1:10:27,  1.48it/s]
3|227|Loss: 0.8263156414031982:   4%|▎         | 227/6470 [02:36<1:19:37,  1.31it/s]
3|228|Loss: 0.8003271222114563:   4%|▎         | 227/6470 [02:36<1:19:37,  1.31it/s]
3|228|Loss: 0.8003271222114563:   4%|▎         | 228/6470 [02:37<1:21:28,  1.28it/s]
3|229|Loss: 0.9302462935447693:   4%|▎         | 228/6470 [02:37<1:21:28,  1.28it/s]
3|229|Loss: 0.9302462935447693:   4%|▎         | 229/6470 [02:37<1:20:12,  1.30it/s]
3|230|Loss: 0.9145032167434692:   4%|▎         | 229/6470 [02:38<1:20:12,  1.30it/s]
3|230|Loss: 0.9145032167434692:   4%|▎         | 230/6470 [02:38<1:18:13,  1.33it/s]
3|231|Loss: 0.8488834500312805:   4%|▎         | 230/6470 [02:38<1:18:13,  1.33it/s]
3|231|Loss: 0.8488834500312805:   4%|▎         | 231/6470 [02:39<1:11:47,  1.45it/s]
3|232|Loss: 0.8080335855484009:   4%|▎         | 231/6470 [02:39<1:11:47,  1.45it/s]
3|232|Loss: 0.8080335855484009:   4%|▎         | 232/6470 [02:39<1:13:13,  1.42it/s]
3|233|Loss: 0.8698418140411377:   4%|▎         | 232/6470 [02:40<1:13:13,  1.42it/s]
3|233|Loss: 0.8698418140411377:   4%|▎         | 233/6470 [02:40<1:14:37,  1.39it/s]
3|234|Loss: 0.8169558048248291:   4%|▎         | 233/6470 [02:40<1:14:37,  1.39it/s]
3|234|Loss: 0.8169558048248291:   4%|▎         | 234/6470 [02:41<1:15:09,  1.38it/s]
3|235|Loss: 0.7986258268356323:   4%|▎         | 234/6470 [02:41<1:15:09,  1.38it/s]
3|235|Loss: 0.7986258268356323:   4%|▎         | 235/6470 [02:41<1:15:31,  1.38it/s]
3|236|Loss: 0.9774287939071655:   4%|▎         | 235/6470 [02:42<1:15:31,  1.38it/s]
3|236|Loss: 0.9774287939071655:   4%|▎         | 236/6470 [02:42<1:07:05,  1.55it/s]
3|237|Loss: 0.7760695815086365:   4%|▎         | 236/6470 [02:42<1:07:05,  1.55it/s]
3|237|Loss: 0.7760695815086365:   4%|▎         | 237/6470 [02:43<1:06:56,  1.55it/s]
3|238|Loss: 0.9054073691368103:   4%|▎         | 237/6470 [02:43<1:06:56,  1.55it/s]
3|238|Loss: 0.9054073691368103:   4%|▎         | 238/6470 [02:43<1:06:45,  1.56it/s]
3|239|Loss: 0.9470463991165161:   4%|▎         | 238/6470 [02:44<1:06:45,  1.56it/s]
3|239|Loss: 0.9470463991165161:   4%|▎         | 239/6470 [02:44<1:09:24,  1.50it/s]
3|240|Loss: 0.7711628079414368:   4%|▎         | 239/6470 [02:44<1:09:24,  1.50it/s]
3|240|Loss: 0.7711628079414368:   4%|▎         | 240/6470 [02:45<1:09:05,  1.50it/s]
3|241|Loss: 0.869541347026825:   4%|▎         | 240/6470 [02:45<1:09:05,  1.50it/s] 
3|241|Loss: 0.869541347026825:   4%|▎         | 241/6470 [02:45<1:10:49,  1.47it/s]
3|242|Loss: 0.9093416333198547:   4%|▎         | 241/6470 [02:46<1:10:49,  1.47it/s]
3|242|Loss: 0.9093416333198547:   4%|▎         | 242/6470 [02:46<1:12:25,  1.43it/s]
3|243|Loss: 0.9221065640449524:   4%|▎         | 242/6470 [02:46<1:12:25,  1.43it/s]
3|243|Loss: 0.9221065640449524:   4%|▍         | 243/6470 [02:47<1:12:53,  1.42it/s]
3|244|Loss: 0.9555412530899048:   4%|▍         | 243/6470 [02:47<1:12:53,  1.42it/s]
3|244|Loss: 0.9555412530899048:   4%|▍         | 244/6470 [02:47<1:12:55,  1.42it/s]
3|245|Loss: 0.6828320622444153:   4%|▍         | 244/6470 [02:48<1:12:55,  1.42it/s]
3|245|Loss: 0.6828320622444153:   4%|▍         | 245/6470 [02:48<1:15:01,  1.38it/s]
3|246|Loss: 0.7883201837539673:   4%|▍         | 245/6470 [02:49<1:15:01,  1.38it/s]
3|246|Loss: 0.7883201837539673:   4%|▍         | 246/6470 [02:49<1:16:11,  1.36it/s]
3|247|Loss: 0.7547393441200256:   4%|▍         | 246/6470 [02:49<1:16:11,  1.36it/s]
3|247|Loss: 0.7547393441200256:   4%|▍         | 247/6470 [02:50<1:19:27,  1.31it/s]
3|248|Loss: 0.8130921721458435:   4%|▍         | 247/6470 [02:50<1:19:27,  1.31it/s]
3|248|Loss: 0.8130921721458435:   4%|▍         | 248/6470 [02:51<1:18:47,  1.32it/s]
3|249|Loss: 0.7316297888755798:   4%|▍         | 248/6470 [02:51<1:18:47,  1.32it/s]
3|249|Loss: 0.7316297888755798:   4%|▍         | 249/6470 [02:51<1:15:00,  1.38it/s]
3|250|Loss: 0.9747985601425171:   4%|▍         | 249/6470 [02:52<1:15:00,  1.38it/s]
3|250|Loss: 0.9747985601425171:   4%|▍         | 250/6470 [02:52<1:15:03,  1.38it/s]
3|251|Loss: 0.7602479457855225:   4%|▍         | 250/6470 [02:52<1:15:03,  1.38it/s]
3|251|Loss: 0.7602479457855225:   4%|▍         | 251/6470 [02:53<1:16:15,  1.36it/s]
3|252|Loss: 0.9658448696136475:   4%|▍         | 251/6470 [02:53<1:16:15,  1.36it/s]
3|252|Loss: 0.9658448696136475:   4%|▍         | 252/6470 [02:53<1:10:53,  1.46it/s]
3|253|Loss: 0.9553463459014893:   4%|▍         | 252/6470 [02:54<1:10:53,  1.46it/s]
3|253|Loss: 0.9553463459014893:   4%|▍         | 253/6470 [02:54<1:05:00,  1.59it/s]
3|254|Loss: 0.8689509630203247:   4%|▍         | 253/6470 [02:54<1:05:00,  1.59it/s]
3|254|Loss: 0.8689509630203247:   4%|▍         | 254/6470 [02:54<1:07:05,  1.54it/s]
3|255|Loss: 0.7348840236663818:   4%|▍         | 254/6470 [02:55<1:07:05,  1.54it/s]
3|255|Loss: 0.7348840236663818:   4%|▍         | 255/6470 [02:55<1:07:51,  1.53it/s]
3|256|Loss: 0.9654371738433838:   4%|▍         | 255/6470 [02:56<1:07:51,  1.53it/s]
3|256|Loss: 0.9654371738433838:   4%|▍         | 256/6470 [02:56<1:09:08,  1.50it/s]
3|257|Loss: 0.8048111796379089:   4%|▍         | 256/6470 [02:56<1:09:08,  1.50it/s]
3|257|Loss: 0.8048111796379089:   4%|▍         | 257/6470 [02:57<1:08:43,  1.51it/s]
3|258|Loss: 0.9822081327438354:   4%|▍         | 257/6470 [02:57<1:08:43,  1.51it/s]
3|258|Loss: 0.9822081327438354:   4%|▍         | 258/6470 [02:57<1:13:32,  1.41it/s]
3|259|Loss: 0.8714895248413086:   4%|▍         | 258/6470 [02:58<1:13:32,  1.41it/s]
3|259|Loss: 0.8714895248413086:   4%|▍         | 259/6470 [02:58<1:11:24,  1.45it/s]
3|260|Loss: 0.8060883283615112:   4%|▍         | 259/6470 [02:58<1:11:24,  1.45it/s]
3|260|Loss: 0.8060883283615112:   4%|▍         | 260/6470 [02:59<1:12:21,  1.43it/s]
3|261|Loss: 0.6837733387947083:   4%|▍         | 260/6470 [02:59<1:12:21,  1.43it/s]
3|261|Loss: 0.6837733387947083:   4%|▍         | 261/6470 [03:00<1:16:09,  1.36it/s]
3|262|Loss: 0.8566804528236389:   4%|▍         | 261/6470 [03:00<1:16:09,  1.36it/s]
3|262|Loss: 0.8566804528236389:   4%|▍         | 262/6470 [03:00<1:11:13,  1.45it/s]
3|263|Loss: 0.878058910369873:   4%|▍         | 262/6470 [03:00<1:11:13,  1.45it/s] 
3|263|Loss: 0.878058910369873:   4%|▍         | 263/6470 [03:01<1:10:16,  1.47it/s]
3|264|Loss: 0.9463605284690857:   4%|▍         | 263/6470 [03:01<1:10:16,  1.47it/s]
3|264|Loss: 0.9463605284690857:   4%|▍         | 264/6470 [03:01<1:11:31,  1.45it/s]
3|265|Loss: 1.0187493562698364:   4%|▍         | 264/6470 [03:02<1:11:31,  1.45it/s]
3|265|Loss: 1.0187493562698364:   4%|▍         | 265/6470 [03:02<1:12:46,  1.42it/s]
3|266|Loss: 0.8103174567222595:   4%|▍         | 265/6470 [03:03<1:12:46,  1.42it/s]
3|266|Loss: 0.8103174567222595:   4%|▍         | 266/6470 [03:03<1:14:10,  1.39it/s]
3|267|Loss: 0.871900737285614:   4%|▍         | 266/6470 [03:03<1:14:10,  1.39it/s] 
3|267|Loss: 0.871900737285614:   4%|▍         | 267/6470 [03:04<1:12:00,  1.44it/s]
3|268|Loss: 0.8732038736343384:   4%|▍         | 267/6470 [03:04<1:12:00,  1.44it/s]
3|268|Loss: 0.8732038736343384:   4%|▍         | 268/6470 [03:04<1:13:49,  1.40it/s]
3|269|Loss: 0.8151087164878845:   4%|▍         | 268/6470 [03:05<1:13:49,  1.40it/s]
3|269|Loss: 0.8151087164878845:   4%|▍         | 269/6470 [03:05<1:11:51,  1.44it/s]
3|270|Loss: 0.861437976360321:   4%|▍         | 269/6470 [03:05<1:11:51,  1.44it/s] 
3|270|Loss: 0.861437976360321:   4%|▍         | 270/6470 [03:06<1:09:46,  1.48it/s]
3|271|Loss: 0.7736282348632812:   4%|▍         | 270/6470 [03:06<1:09:46,  1.48it/s]
3|271|Loss: 0.7736282348632812:   4%|▍         | 271/6470 [03:06<1:10:57,  1.46it/s]
3|272|Loss: 0.8954359889030457:   4%|▍         | 271/6470 [03:07<1:10:57,  1.46it/s]
3|272|Loss: 0.8954359889030457:   4%|▍         | 272/6470 [03:07<1:11:38,  1.44it/s]
3|273|Loss: 0.7521461844444275:   4%|▍         | 272/6470 [03:07<1:11:38,  1.44it/s]
3|273|Loss: 0.7521461844444275:   4%|▍         | 273/6470 [03:08<1:10:36,  1.46it/s]
3|274|Loss: 0.9332347512245178:   4%|▍         | 273/6470 [03:08<1:10:36,  1.46it/s]
3|274|Loss: 0.9332347512245178:   4%|▍         | 274/6470 [03:08<1:09:15,  1.49it/s]
3|275|Loss: 0.8704636096954346:   4%|▍         | 274/6470 [03:09<1:09:15,  1.49it/s]
3|275|Loss: 0.8704636096954346:   4%|▍         | 275/6470 [03:09<1:08:23,  1.51it/s]
3|276|Loss: 0.8237176537513733:   4%|▍         | 275/6470 [03:09<1:08:23,  1.51it/s]
3|276|Loss: 0.8237176537513733:   4%|▍         | 276/6470 [03:10<1:06:23,  1.56it/s]
3|277|Loss: 0.9353325366973877:   4%|▍         | 276/6470 [03:10<1:06:23,  1.56it/s]
3|277|Loss: 0.9353325366973877:   4%|▍         | 277/6470 [03:10<1:05:21,  1.58it/s]
3|278|Loss: 0.9080893993377686:   4%|▍         | 277/6470 [03:11<1:05:21,  1.58it/s]
3|278|Loss: 0.9080893993377686:   4%|▍         | 278/6470 [03:11<1:04:48,  1.59it/s]
3|279|Loss: 0.8006142973899841:   4%|▍         | 278/6470 [03:11<1:04:48,  1.59it/s]
3|279|Loss: 0.8006142973899841:   4%|▍         | 279/6470 [03:11<1:01:48,  1.67it/s]
3|280|Loss: 0.971389889717102:   4%|▍         | 279/6470 [03:12<1:01:48,  1.67it/s] 
3|280|Loss: 0.971389889717102:   4%|▍         | 280/6470 [03:12<1:04:41,  1.59it/s]
3|281|Loss: 0.9538220763206482:   4%|▍         | 280/6470 [03:12<1:04:41,  1.59it/s]
3|281|Loss: 0.9538220763206482:   4%|▍         | 281/6470 [03:13<1:07:28,  1.53it/s]
3|282|Loss: 0.7968989610671997:   4%|▍         | 281/6470 [03:13<1:07:28,  1.53it/s]
3|282|Loss: 0.7968989610671997:   4%|▍         | 282/6470 [03:13<1:04:15,  1.60it/s]
3|283|Loss: 0.9800841212272644:   4%|▍         | 282/6470 [03:14<1:04:15,  1.60it/s]
3|283|Loss: 0.9800841212272644:   4%|▍         | 283/6470 [03:14<1:02:25,  1.65it/s]
3|284|Loss: 0.8864821791648865:   4%|▍         | 283/6470 [03:14<1:02:25,  1.65it/s]
3|284|Loss: 0.8864821791648865:   4%|▍         | 284/6470 [03:15<1:03:03,  1.63it/s]
3|285|Loss: 0.914794385433197:   4%|▍         | 284/6470 [03:15<1:03:03,  1.63it/s] 
3|285|Loss: 0.914794385433197:   4%|▍         | 285/6470 [03:15<1:10:18,  1.47it/s]
3|286|Loss: 0.5640669465065002:   4%|▍         | 285/6470 [03:16<1:10:18,  1.47it/s]
3|286|Loss: 0.5640669465065002:   4%|▍         | 286/6470 [03:16<1:18:57,  1.31it/s]
3|287|Loss: 0.9094802141189575:   4%|▍         | 286/6470 [03:17<1:18:57,  1.31it/s]
3|287|Loss: 0.9094802141189575:   4%|▍         | 287/6470 [03:17<1:20:02,  1.29it/s]
3|288|Loss: 0.7917692065238953:   4%|▍         | 287/6470 [03:18<1:20:02,  1.29it/s]
3|288|Loss: 0.7917692065238953:   4%|▍         | 288/6470 [03:18<1:18:19,  1.32it/s]
3|289|Loss: 1.0107053518295288:   4%|▍         | 288/6470 [03:18<1:18:19,  1.32it/s]
3|289|Loss: 1.0107053518295288:   4%|▍         | 289/6470 [03:18<1:14:57,  1.37it/s]
3|290|Loss: 0.8879120349884033:   4%|▍         | 289/6470 [03:19<1:14:57,  1.37it/s]
3|290|Loss: 0.8879120349884033:   4%|▍         | 290/6470 [03:19<1:10:53,  1.45it/s]
3|291|Loss: 0.7365846037864685:   4%|▍         | 290/6470 [03:19<1:10:53,  1.45it/s]
3|291|Loss: 0.7365846037864685:   4%|▍         | 291/6470 [03:20<1:04:26,  1.60it/s]
3|292|Loss: 0.9146444201469421:   4%|▍         | 291/6470 [03:20<1:04:26,  1.60it/s]
3|292|Loss: 0.9146444201469421:   5%|▍         | 292/6470 [03:20<1:00:54,  1.69it/s]
3|293|Loss: 0.9027358889579773:   5%|▍         | 292/6470 [03:20<1:00:54,  1.69it/s]
3|293|Loss: 0.9027358889579773:   5%|▍         | 293/6470 [03:21<1:03:59,  1.61it/s]
3|294|Loss: 0.8662711977958679:   5%|▍         | 293/6470 [03:21<1:03:59,  1.61it/s]
3|294|Loss: 0.8662711977958679:   5%|▍         | 294/6470 [03:22<1:10:32,  1.46it/s]
3|295|Loss: 0.8823490142822266:   5%|▍         | 294/6470 [03:22<1:10:32,  1.46it/s]
3|295|Loss: 0.8823490142822266:   5%|▍         | 295/6470 [03:22<1:07:28,  1.53it/s]
3|296|Loss: 0.9236472249031067:   5%|▍         | 295/6470 [03:23<1:07:28,  1.53it/s]
3|296|Loss: 0.9236472249031067:   5%|▍         | 296/6470 [03:23<1:08:38,  1.50it/s]
3|297|Loss: 0.8250625729560852:   5%|▍         | 296/6470 [03:23<1:08:38,  1.50it/s]
3|297|Loss: 0.8250625729560852:   5%|▍         | 297/6470 [03:24<1:08:15,  1.51it/s]
3|298|Loss: 0.868024468421936:   5%|▍         | 297/6470 [03:24<1:08:15,  1.51it/s] 
3|298|Loss: 0.868024468421936:   5%|▍         | 298/6470 [03:24<1:14:26,  1.38it/s]
3|299|Loss: 0.8958678841590881:   5%|▍         | 298/6470 [03:25<1:14:26,  1.38it/s]
3|299|Loss: 0.8958678841590881:   5%|▍         | 299/6470 [03:25<1:19:07,  1.30it/s]
3|300|Loss: 0.8857624530792236:   5%|▍         | 299/6470 [03:26<1:19:07,  1.30it/s]
3|300|Loss: 0.8857624530792236:   5%|▍         | 300/6470 [03:26<1:15:12,  1.37it/s]
3|301|Loss: 0.9996389746665955:   5%|▍         | 300/6470 [03:26<1:15:12,  1.37it/s]
3|301|Loss: 0.9996389746665955:   5%|▍         | 301/6470 [03:27<1:11:17,  1.44it/s]
3|302|Loss: 0.8535995483398438:   5%|▍         | 301/6470 [03:27<1:11:17,  1.44it/s]
3|302|Loss: 0.8535995483398438:   5%|▍         | 302/6470 [03:27<1:11:48,  1.43it/s]
3|303|Loss: 0.9747299551963806:   5%|▍         | 302/6470 [03:28<1:11:48,  1.43it/s]
3|303|Loss: 0.9747299551963806:   5%|▍         | 303/6470 [03:28<1:13:10,  1.40it/s]
3|304|Loss: 0.871782124042511:   5%|▍         | 303/6470 [03:28<1:13:10,  1.40it/s] 
3|304|Loss: 0.871782124042511:   5%|▍         | 304/6470 [03:29<1:11:08,  1.44it/s]
3|305|Loss: 0.8406854867935181:   5%|▍         | 304/6470 [03:29<1:11:08,  1.44it/s]
3|305|Loss: 0.8406854867935181:   5%|▍         | 305/6470 [03:29<1:06:03,  1.56it/s]
3|306|Loss: 0.7404324412345886:   5%|▍         | 305/6470 [03:29<1:06:03,  1.56it/s]
3|306|Loss: 0.7404324412345886:   5%|▍         | 306/6470 [03:30<1:05:09,  1.58it/s]
3|307|Loss: 0.7644345760345459:   5%|▍         | 306/6470 [03:30<1:05:09,  1.58it/s]
3|307|Loss: 0.7644345760345459:   5%|▍         | 307/6470 [03:30<1:08:01,  1.51it/s]
3|308|Loss: 1.00033438205719:   5%|▍         | 307/6470 [03:31<1:08:01,  1.51it/s]  
3|308|Loss: 1.00033438205719:   5%|▍         | 308/6470 [03:31<1:06:35,  1.54it/s]
3|309|Loss: 0.7574107646942139:   5%|▍         | 308/6470 [03:31<1:06:35,  1.54it/s]
3|309|Loss: 0.7574107646942139:   5%|▍         | 309/6470 [03:32<1:07:26,  1.52it/s]
3|310|Loss: 1.0024923086166382:   5%|▍         | 309/6470 [03:32<1:07:26,  1.52it/s]
3|310|Loss: 1.0024923086166382:   5%|▍         | 310/6470 [03:32<1:05:38,  1.56it/s]
3|311|Loss: 0.6441293954849243:   5%|▍         | 310/6470 [03:33<1:05:38,  1.56it/s]
3|311|Loss: 0.6441293954849243:   5%|▍         | 311/6470 [03:33<1:13:15,  1.40it/s]
3|312|Loss: 0.8700981140136719:   5%|▍         | 311/6470 [03:34<1:13:15,  1.40it/s]
3|312|Loss: 0.8700981140136719:   5%|▍         | 312/6470 [03:34<1:14:36,  1.38it/s]
3|313|Loss: 0.9777651429176331:   5%|▍         | 312/6470 [03:34<1:14:36,  1.38it/s]
3|313|Loss: 0.9777651429176331:   5%|▍         | 313/6470 [03:35<1:11:44,  1.43it/s]
3|314|Loss: 0.7968536019325256:   5%|▍         | 313/6470 [03:35<1:11:44,  1.43it/s]
3|314|Loss: 0.7968536019325256:   5%|▍         | 314/6470 [03:35<1:08:10,  1.51it/s]
3|315|Loss: 0.8319797515869141:   5%|▍         | 314/6470 [03:36<1:08:10,  1.51it/s]
3|315|Loss: 0.8319797515869141:   5%|▍         | 315/6470 [03:36<1:11:19,  1.44it/s]
3|316|Loss: 0.7683606743812561:   5%|▍         | 315/6470 [03:36<1:11:19,  1.44it/s]
3|316|Loss: 0.7683606743812561:   5%|▍         | 316/6470 [03:37<1:12:56,  1.41it/s]
3|317|Loss: 0.927232027053833:   5%|▍         | 316/6470 [03:37<1:12:56,  1.41it/s] 
3|317|Loss: 0.927232027053833:   5%|▍         | 317/6470 [03:38<1:17:07,  1.33it/s]
3|318|Loss: 0.8195313215255737:   5%|▍         | 317/6470 [03:38<1:17:07,  1.33it/s]
3|318|Loss: 0.8195313215255737:   5%|▍         | 318/6470 [03:38<1:17:58,  1.31it/s]
3|319|Loss: 0.9435911774635315:   5%|▍         | 318/6470 [03:39<1:17:58,  1.31it/s]
3|319|Loss: 0.9435911774635315:   5%|▍         | 319/6470 [03:39<1:13:24,  1.40it/s]
3|320|Loss: 0.8810974359512329:   5%|▍         | 319/6470 [03:39<1:13:24,  1.40it/s]
3|320|Loss: 0.8810974359512329:   5%|▍         | 320/6470 [03:40<1:09:03,  1.48it/s]
3|321|Loss: 0.7560892105102539:   5%|▍         | 320/6470 [03:40<1:09:03,  1.48it/s]
3|321|Loss: 0.7560892105102539:   5%|▍         | 321/6470 [03:40<1:12:21,  1.42it/s]
3|322|Loss: 0.8177131414413452:   5%|▍         | 321/6470 [03:41<1:12:21,  1.42it/s]
3|322|Loss: 0.8177131414413452:   5%|▍         | 322/6470 [03:41<1:13:37,  1.39it/s]
3|323|Loss: 0.8368629217147827:   5%|▍         | 322/6470 [03:41<1:13:37,  1.39it/s]
3|323|Loss: 0.8368629217147827:   5%|▍         | 323/6470 [03:42<1:11:14,  1.44it/s]
3|324|Loss: 0.7416557669639587:   5%|▍         | 323/6470 [03:42<1:11:14,  1.44it/s]
3|324|Loss: 0.7416557669639587:   5%|▌         | 324/6470 [03:42<1:08:51,  1.49it/s]
3|325|Loss: 0.8581913709640503:   5%|▌         | 324/6470 [03:43<1:08:51,  1.49it/s]
3|325|Loss: 0.8581913709640503:   5%|▌         | 325/6470 [03:43<1:12:27,  1.41it/s]
3|326|Loss: 0.772430956363678:   5%|▌         | 325/6470 [03:44<1:12:27,  1.41it/s] 
3|326|Loss: 0.772430956363678:   5%|▌         | 326/6470 [03:44<1:13:10,  1.40it/s]
3|327|Loss: 0.7421010136604309:   5%|▌         | 326/6470 [03:44<1:13:10,  1.40it/s]
3|327|Loss: 0.7421010136604309:   5%|▌         | 327/6470 [03:45<1:13:24,  1.39it/s]
3|328|Loss: 0.7508907318115234:   5%|▌         | 327/6470 [03:45<1:13:24,  1.39it/s]
3|328|Loss: 0.7508907318115234:   5%|▌         | 328/6470 [03:45<1:10:59,  1.44it/s]
3|329|Loss: 0.9113070964813232:   5%|▌         | 328/6470 [03:46<1:10:59,  1.44it/s]
3|329|Loss: 0.9113070964813232:   5%|▌         | 329/6470 [03:46<1:07:30,  1.52it/s]
3|330|Loss: 0.9872968792915344:   5%|▌         | 329/6470 [03:46<1:07:30,  1.52it/s]
3|330|Loss: 0.9872968792915344:   5%|▌         | 330/6470 [03:46<1:05:05,  1.57it/s]
3|331|Loss: 0.8370344638824463:   5%|▌         | 330/6470 [03:47<1:05:05,  1.57it/s]
3|331|Loss: 0.8370344638824463:   5%|▌         | 331/6470 [03:47<1:04:50,  1.58it/s]
3|332|Loss: 0.8661092519760132:   5%|▌         | 331/6470 [03:47<1:04:50,  1.58it/s]
3|332|Loss: 0.8661092519760132:   5%|▌         | 332/6470 [03:48<1:07:25,  1.52it/s]
3|333|Loss: 0.7936899065971375:   5%|▌         | 332/6470 [03:48<1:07:25,  1.52it/s]
3|333|Loss: 0.7936899065971375:   5%|▌         | 333/6470 [03:48<1:00:41,  1.69it/s]
3|334|Loss: 0.957931637763977:   5%|▌         | 333/6470 [03:49<1:00:41,  1.69it/s] 
3|334|Loss: 0.957931637763977:   5%|▌         | 334/6470 [03:49<1:03:04,  1.62it/s]
3|335|Loss: 0.7690235376358032:   5%|▌         | 334/6470 [03:49<1:03:04,  1.62it/s]
3|335|Loss: 0.7690235376358032:   5%|▌         | 335/6470 [03:50<1:08:23,  1.50it/s]
3|336|Loss: 0.7477759122848511:   5%|▌         | 335/6470 [03:50<1:08:23,  1.50it/s]
3|336|Loss: 0.7477759122848511:   5%|▌         | 336/6470 [03:50<1:08:38,  1.49it/s]
3|337|Loss: 0.8040043711662292:   5%|▌         | 336/6470 [03:51<1:08:38,  1.49it/s]
3|337|Loss: 0.8040043711662292:   5%|▌         | 337/6470 [03:51<1:11:20,  1.43it/s]
3|338|Loss: 0.7793408632278442:   5%|▌         | 337/6470 [03:51<1:11:20,  1.43it/s]
3|338|Loss: 0.7793408632278442:   5%|▌         | 338/6470 [03:52<1:06:04,  1.55it/s]
3|339|Loss: 0.6588653922080994:   5%|▌         | 338/6470 [03:52<1:06:04,  1.55it/s]
3|339|Loss: 0.6588653922080994:   5%|▌         | 339/6470 [03:52<58:17,  1.75it/s]  
3|340|Loss: 0.9339260458946228:   5%|▌         | 339/6470 [03:52<58:17,  1.75it/s]
3|340|Loss: 0.9339260458946228:   5%|▌         | 340/6470 [03:53<57:32,  1.78it/s]
3|341|Loss: 0.7647345662117004:   5%|▌         | 340/6470 [03:53<57:32,  1.78it/s]
3|341|Loss: 0.7647345662117004:   5%|▌         | 341/6470 [03:53<58:20,  1.75it/s]
3|342|Loss: 0.7768259644508362:   5%|▌         | 341/6470 [03:54<58:20,  1.75it/s]
3|342|Loss: 0.7768259644508362:   5%|▌         | 342/6470 [03:54<1:05:20,  1.56it/s]
3|343|Loss: 0.8828428983688354:   5%|▌         | 342/6470 [03:54<1:05:20,  1.56it/s]
3|343|Loss: 0.8828428983688354:   5%|▌         | 343/6470 [03:55<1:09:22,  1.47it/s]
3|344|Loss: 0.9661685824394226:   5%|▌         | 343/6470 [03:55<1:09:22,  1.47it/s]
3|344|Loss: 0.9661685824394226:   5%|▌         | 344/6470 [03:55<1:11:10,  1.43it/s]
3|345|Loss: 0.875836193561554:   5%|▌         | 344/6470 [03:56<1:11:10,  1.43it/s] 
3|345|Loss: 0.875836193561554:   5%|▌         | 345/6470 [03:56<1:11:28,  1.43it/s]
3|346|Loss: 0.8632096648216248:   5%|▌         | 345/6470 [03:57<1:11:28,  1.43it/s]
3|346|Loss: 0.8632096648216248:   5%|▌         | 346/6470 [03:57<1:08:38,  1.49it/s]
3|347|Loss: 0.9325672388076782:   5%|▌         | 346/6470 [03:57<1:08:38,  1.49it/s]
3|347|Loss: 0.9325672388076782:   5%|▌         | 347/6470 [03:57<1:04:34,  1.58it/s]
3|348|Loss: 0.8354230523109436:   5%|▌         | 347/6470 [03:58<1:04:34,  1.58it/s]
3|348|Loss: 0.8354230523109436:   5%|▌         | 348/6470 [03:58<1:06:12,  1.54it/s]
3|349|Loss: 0.6883471608161926:   5%|▌         | 348/6470 [03:58<1:06:12,  1.54it/s]
3|349|Loss: 0.6883471608161926:   5%|▌         | 349/6470 [03:59<1:05:04,  1.57it/s]
3|350|Loss: 0.8722769618034363:   5%|▌         | 349/6470 [03:59<1:05:04,  1.57it/s]
3|350|Loss: 0.8722769618034363:   5%|▌         | 350/6470 [03:59<1:07:07,  1.52it/s]
3|351|Loss: 0.810511589050293:   5%|▌         | 350/6470 [04:00<1:07:07,  1.52it/s] 
3|351|Loss: 0.810511589050293:   5%|▌         | 351/6470 [04:00<1:13:02,  1.40it/s]
3|352|Loss: 0.8537311553955078:   5%|▌         | 351/6470 [04:01<1:13:02,  1.40it/s]
3|352|Loss: 0.8537311553955078:   5%|▌         | 352/6470 [04:01<1:17:05,  1.32it/s]
3|353|Loss: 0.8781119585037231:   5%|▌         | 352/6470 [04:01<1:17:05,  1.32it/s]
3|353|Loss: 0.8781119585037231:   5%|▌         | 353/6470 [04:02<1:20:39,  1.26it/s]
3|354|Loss: 1.0974055528640747:   5%|▌         | 353/6470 [04:02<1:20:39,  1.26it/s]
3|354|Loss: 1.0974055528640747:   5%|▌         | 354/6470 [04:03<1:17:41,  1.31it/s]
3|355|Loss: 1.0109894275665283:   5%|▌         | 354/6470 [04:03<1:17:41,  1.31it/s]
3|355|Loss: 1.0109894275665283:   5%|▌         | 355/6470 [04:03<1:19:55,  1.28it/s]
3|356|Loss: 0.8553063273429871:   5%|▌         | 355/6470 [04:04<1:19:55,  1.28it/s]
3|356|Loss: 0.8553063273429871:   6%|▌         | 356/6470 [04:04<1:17:08,  1.32it/s]
3|357|Loss: 0.6015745997428894:   6%|▌         | 356/6470 [04:05<1:17:08,  1.32it/s]
3|357|Loss: 0.6015745997428894:   6%|▌         | 357/6470 [04:05<1:17:18,  1.32it/s]
3|358|Loss: 0.7063030004501343:   6%|▌         | 357/6470 [04:05<1:17:18,  1.32it/s]
3|358|Loss: 0.7063030004501343:   6%|▌         | 358/6470 [04:06<1:16:27,  1.33it/s]
3|359|Loss: 0.7623550295829773:   6%|▌         | 358/6470 [04:06<1:16:27,  1.33it/s]
3|359|Loss: 0.7623550295829773:   6%|▌         | 359/6470 [04:06<1:19:21,  1.28it/s]
3|360|Loss: 0.858898937702179:   6%|▌         | 359/6470 [04:07<1:19:21,  1.28it/s] 
3|360|Loss: 0.858898937702179:   6%|▌         | 360/6470 [04:07<1:15:14,  1.35it/s]
3|361|Loss: 0.8735377192497253:   6%|▌         | 360/6470 [04:07<1:15:14,  1.35it/s]
3|361|Loss: 0.8735377192497253:   6%|▌         | 361/6470 [04:08<1:13:19,  1.39it/s]
3|362|Loss: 1.072118878364563:   6%|▌         | 361/6470 [04:08<1:13:19,  1.39it/s] 
3|362|Loss: 1.072118878364563:   6%|▌         | 362/6470 [04:09<1:15:29,  1.35it/s]
3|363|Loss: 0.7959190011024475:   6%|▌         | 362/6470 [04:09<1:15:29,  1.35it/s]
3|363|Loss: 0.7959190011024475:   6%|▌         | 363/6470 [04:09<1:17:18,  1.32it/s]
3|364|Loss: 0.8508473634719849:   6%|▌         | 363/6470 [04:10<1:17:18,  1.32it/s]
3|364|Loss: 0.8508473634719849:   6%|▌         | 364/6470 [04:10<1:19:54,  1.27it/s]
3|365|Loss: 0.837163507938385:   6%|▌         | 364/6470 [04:11<1:19:54,  1.27it/s] 
3|365|Loss: 0.837163507938385:   6%|▌         | 365/6470 [04:11<1:23:41,  1.22it/s]
3|366|Loss: 0.9010207056999207:   6%|▌         | 365/6470 [04:12<1:23:41,  1.22it/s]
3|366|Loss: 0.9010207056999207:   6%|▌         | 366/6470 [04:12<1:22:09,  1.24it/s]
3|367|Loss: 0.821956992149353:   6%|▌         | 366/6470 [04:12<1:22:09,  1.24it/s] 
3|367|Loss: 0.821956992149353:   6%|▌         | 367/6470 [04:13<1:19:17,  1.28it/s]
3|368|Loss: 0.7742157578468323:   6%|▌         | 367/6470 [04:13<1:19:17,  1.28it/s]
3|368|Loss: 0.7742157578468323:   6%|▌         | 368/6470 [04:13<1:11:00,  1.43it/s]
3|369|Loss: 0.850824236869812:   6%|▌         | 368/6470 [04:13<1:11:00,  1.43it/s] 
3|369|Loss: 0.850824236869812:   6%|▌         | 369/6470 [04:14<1:10:27,  1.44it/s]
3|370|Loss: 0.8835456967353821:   6%|▌         | 369/6470 [04:14<1:10:27,  1.44it/s]
3|370|Loss: 0.8835456967353821:   6%|▌         | 370/6470 [04:15<1:11:25,  1.42it/s]
3|371|Loss: 1.1652706861495972:   6%|▌         | 370/6470 [04:15<1:11:25,  1.42it/s]
3|371|Loss: 1.1652706861495972:   6%|▌         | 371/6470 [04:15<1:08:43,  1.48it/s]
3|372|Loss: 0.7815265655517578:   6%|▌         | 371/6470 [04:16<1:08:43,  1.48it/s]
3|372|Loss: 0.7815265655517578:   6%|▌         | 372/6470 [04:16<1:10:01,  1.45it/s]
3|373|Loss: 0.8318374752998352:   6%|▌         | 372/6470 [04:16<1:10:01,  1.45it/s]
3|373|Loss: 0.8318374752998352:   6%|▌         | 373/6470 [04:16<1:07:08,  1.51it/s]
3|374|Loss: 0.6654788255691528:   6%|▌         | 373/6470 [04:17<1:07:08,  1.51it/s]
3|374|Loss: 0.6654788255691528:   6%|▌         | 374/6470 [04:17<1:08:14,  1.49it/s]
3|375|Loss: 0.6700925230979919:   6%|▌         | 374/6470 [04:18<1:08:14,  1.49it/s]
3|375|Loss: 0.6700925230979919:   6%|▌         | 375/6470 [04:18<1:13:18,  1.39it/s]
3|376|Loss: 0.8041730523109436:   6%|▌         | 375/6470 [04:18<1:13:18,  1.39it/s]
3|376|Loss: 0.8041730523109436:   6%|▌         | 376/6470 [04:19<1:07:53,  1.50it/s]
3|377|Loss: 0.8217034339904785:   6%|▌         | 376/6470 [04:19<1:07:53,  1.50it/s]
3|377|Loss: 0.8217034339904785:   6%|▌         | 377/6470 [04:19<1:07:14,  1.51it/s]
3|378|Loss: 0.8282922506332397:   6%|▌         | 377/6470 [04:20<1:07:14,  1.51it/s]
3|378|Loss: 0.8282922506332397:   6%|▌         | 378/6470 [04:20<1:11:42,  1.42it/s]
3|379|Loss: 0.9168544411659241:   6%|▌         | 378/6470 [04:20<1:11:42,  1.42it/s]
3|379|Loss: 0.9168544411659241:   6%|▌         | 379/6470 [04:21<1:07:45,  1.50it/s]
3|380|Loss: 0.9147270917892456:   6%|▌         | 379/6470 [04:21<1:07:45,  1.50it/s]
3|380|Loss: 0.9147270917892456:   6%|▌         | 380/6470 [04:21<1:05:35,  1.55it/s]
3|381|Loss: 0.9393942356109619:   6%|▌         | 380/6470 [04:22<1:05:35,  1.55it/s]
3|381|Loss: 0.9393942356109619:   6%|▌         | 381/6470 [04:22<1:03:40,  1.59it/s]
3|382|Loss: 0.8061501979827881:   6%|▌         | 381/6470 [04:22<1:03:40,  1.59it/s]
3|382|Loss: 0.8061501979827881:   6%|▌         | 382/6470 [04:22<59:16,  1.71it/s]  
3|383|Loss: 0.8744007349014282:   6%|▌         | 382/6470 [04:23<59:16,  1.71it/s]
3|383|Loss: 0.8744007349014282:   6%|▌         | 383/6470 [04:23<1:02:15,  1.63it/s]
3|384|Loss: 0.890852689743042:   6%|▌         | 383/6470 [04:23<1:02:15,  1.63it/s] 
3|384|Loss: 0.890852689743042:   6%|▌         | 384/6470 [04:23<1:00:49,  1.67it/s]
3|385|Loss: 0.8852031230926514:   6%|▌         | 384/6470 [04:24<1:00:49,  1.67it/s]
3|385|Loss: 0.8852031230926514:   6%|▌         | 385/6470 [04:24<1:00:38,  1.67it/s]
3|386|Loss: 0.8009584546089172:   6%|▌         | 385/6470 [04:24<1:00:38,  1.67it/s]
3|386|Loss: 0.8009584546089172:   6%|▌         | 386/6470 [04:25<1:01:38,  1.65it/s]
3|387|Loss: 0.8467909097671509:   6%|▌         | 386/6470 [04:25<1:01:38,  1.65it/s]
3|387|Loss: 0.8467909097671509:   6%|▌         | 387/6470 [04:25<58:14,  1.74it/s]  
3|388|Loss: 1.0072336196899414:   6%|▌         | 387/6470 [04:26<58:14,  1.74it/s]
3|388|Loss: 1.0072336196899414:   6%|▌         | 388/6470 [04:26<57:29,  1.76it/s]
3|389|Loss: 0.8414338827133179:   6%|▌         | 388/6470 [04:26<57:29,  1.76it/s]
3|389|Loss: 0.8414338827133179:   6%|▌         | 389/6470 [04:26<58:56,  1.72it/s]
3|390|Loss: 0.8229561448097229:   6%|▌         | 389/6470 [04:27<58:56,  1.72it/s]
3|390|Loss: 0.8229561448097229:   6%|▌         | 390/6470 [04:27<1:00:15,  1.68it/s]
3|391|Loss: 0.7600451111793518:   6%|▌         | 390/6470 [04:27<1:00:15,  1.68it/s]
3|391|Loss: 0.7600451111793518:   6%|▌         | 391/6470 [04:28<1:04:17,  1.58it/s]
3|392|Loss: 0.9390788078308105:   6%|▌         | 391/6470 [04:28<1:04:17,  1.58it/s]
3|392|Loss: 0.9390788078308105:   6%|▌         | 392/6470 [04:28<1:07:45,  1.50it/s]
3|393|Loss: 0.8514161705970764:   6%|▌         | 392/6470 [04:29<1:07:45,  1.50it/s]
3|393|Loss: 0.8514161705970764:   6%|▌         | 393/6470 [04:29<1:09:27,  1.46it/s]
3|394|Loss: 0.8078902363777161:   6%|▌         | 393/6470 [04:30<1:09:27,  1.46it/s]
3|394|Loss: 0.8078902363777161:   6%|▌         | 394/6470 [04:30<1:10:59,  1.43it/s]
3|395|Loss: 0.9107376337051392:   6%|▌         | 394/6470 [04:30<1:10:59,  1.43it/s]
3|395|Loss: 0.9107376337051392:   6%|▌         | 395/6470 [04:31<1:14:57,  1.35it/s]
3|396|Loss: 0.7818624377250671:   6%|▌         | 395/6470 [04:31<1:14:57,  1.35it/s]
3|396|Loss: 0.7818624377250671:   6%|▌         | 396/6470 [04:32<1:19:07,  1.28it/s]
3|397|Loss: 0.8798372745513916:   6%|▌         | 396/6470 [04:32<1:19:07,  1.28it/s]
3|397|Loss: 0.8798372745513916:   6%|▌         | 397/6470 [04:33<1:21:41,  1.24it/s]
3|398|Loss: 0.7266626954078674:   6%|▌         | 397/6470 [04:33<1:21:41,  1.24it/s]
3|398|Loss: 0.7266626954078674:   6%|▌         | 398/6470 [04:33<1:22:35,  1.23it/s]
3|399|Loss: 0.7812327742576599:   6%|▌         | 398/6470 [04:34<1:22:35,  1.23it/s]
3|399|Loss: 0.7812327742576599:   6%|▌         | 399/6470 [04:34<1:23:33,  1.21it/s]
3|400|Loss: 0.7854824662208557:   6%|▌         | 399/6470 [04:35<1:23:33,  1.21it/s]
3|400|Loss: 0.7854824662208557:   6%|▌         | 400/6470 [04:35<1:12:44,  1.39it/s]
3|401|Loss: 0.8647289276123047:   6%|▌         | 400/6470 [04:35<1:12:44,  1.39it/s]
3|401|Loss: 0.8647289276123047:   6%|▌         | 401/6470 [04:35<1:10:37,  1.43it/s]
3|402|Loss: 0.8121594190597534:   6%|▌         | 401/6470 [04:36<1:10:37,  1.43it/s]
3|402|Loss: 0.8121594190597534:   6%|▌         | 402/6470 [04:36<1:12:08,  1.40it/s]
3|403|Loss: 0.8252537846565247:   6%|▌         | 402/6470 [04:36<1:12:08,  1.40it/s]
3|403|Loss: 0.8252537846565247:   6%|▌         | 403/6470 [04:37<1:13:32,  1.37it/s]
3|404|Loss: 0.8127715587615967:   6%|▌         | 403/6470 [04:37<1:13:32,  1.37it/s]
3|404|Loss: 0.8127715587615967:   6%|▌         | 404/6470 [04:38<1:15:35,  1.34it/s]
3|405|Loss: 0.781406044960022:   6%|▌         | 404/6470 [04:38<1:15:35,  1.34it/s] 
3|405|Loss: 0.781406044960022:   6%|▋         | 405/6470 [04:38<1:09:34,  1.45it/s]
3|406|Loss: 0.9022513628005981:   6%|▋         | 405/6470 [04:38<1:09:34,  1.45it/s]
3|406|Loss: 0.9022513628005981:   6%|▋         | 406/6470 [04:39<1:04:34,  1.56it/s]
3|407|Loss: 0.9311989545822144:   6%|▋         | 406/6470 [04:39<1:04:34,  1.56it/s]
3|407|Loss: 0.9311989545822144:   6%|▋         | 407/6470 [04:39<1:00:48,  1.66it/s]
3|408|Loss: 0.670600175857544:   6%|▋         | 407/6470 [04:40<1:00:48,  1.66it/s] 
3|408|Loss: 0.670600175857544:   6%|▋         | 408/6470 [04:40<1:05:08,  1.55it/s]
3|409|Loss: 0.815680742263794:   6%|▋         | 408/6470 [04:40<1:05:08,  1.55it/s]
3|409|Loss: 0.815680742263794:   6%|▋         | 409/6470 [04:41<1:06:04,  1.53it/s]
3|410|Loss: 0.8223496079444885:   6%|▋         | 409/6470 [04:41<1:06:04,  1.53it/s]
3|410|Loss: 0.8223496079444885:   6%|▋         | 410/6470 [04:41<1:03:22,  1.59it/s]
3|411|Loss: 0.9252628684043884:   6%|▋         | 410/6470 [04:42<1:03:22,  1.59it/s]
3|411|Loss: 0.9252628684043884:   6%|▋         | 411/6470 [04:42<1:09:04,  1.46it/s]
3|412|Loss: 0.9295427203178406:   6%|▋         | 411/6470 [04:42<1:09:04,  1.46it/s]
3|412|Loss: 0.9295427203178406:   6%|▋         | 412/6470 [04:43<1:11:41,  1.41it/s]
3|413|Loss: 0.7736861705780029:   6%|▋         | 412/6470 [04:43<1:11:41,  1.41it/s]
3|413|Loss: 0.7736861705780029:   6%|▋         | 413/6470 [04:43<1:11:52,  1.40it/s]
3|414|Loss: 1.0041658878326416:   6%|▋         | 413/6470 [04:44<1:11:52,  1.40it/s]
3|414|Loss: 1.0041658878326416:   6%|▋         | 414/6470 [04:44<1:07:50,  1.49it/s]
3|415|Loss: 0.8636760711669922:   6%|▋         | 414/6470 [04:44<1:07:50,  1.49it/s]
3|415|Loss: 0.8636760711669922:   6%|▋         | 415/6470 [04:45<1:07:56,  1.49it/s]
3|416|Loss: 0.8873665928840637:   6%|▋         | 415/6470 [04:45<1:07:56,  1.49it/s]
3|416|Loss: 0.8873665928840637:   6%|▋         | 416/6470 [04:46<1:14:06,  1.36it/s]
3|417|Loss: 0.874074399471283:   6%|▋         | 416/6470 [04:46<1:14:06,  1.36it/s] 
3|417|Loss: 0.874074399471283:   6%|▋         | 417/6470 [04:46<1:16:41,  1.32it/s]
3|418|Loss: 0.7590236067771912:   6%|▋         | 417/6470 [04:47<1:16:41,  1.32it/s]
3|418|Loss: 0.7590236067771912:   6%|▋         | 418/6470 [04:47<1:15:21,  1.34it/s]
3|419|Loss: 0.624140739440918:   6%|▋         | 418/6470 [04:48<1:15:21,  1.34it/s] 
3|419|Loss: 0.624140739440918:   6%|▋         | 419/6470 [04:48<1:13:49,  1.37it/s]
3|420|Loss: 0.8195375800132751:   6%|▋         | 419/6470 [04:48<1:13:49,  1.37it/s]
3|420|Loss: 0.8195375800132751:   6%|▋         | 420/6470 [04:48<1:06:52,  1.51it/s]
3|421|Loss: 0.898630678653717:   6%|▋         | 420/6470 [04:49<1:06:52,  1.51it/s] 
3|421|Loss: 0.898630678653717:   7%|▋         | 421/6470 [04:49<1:04:13,  1.57it/s]
3|422|Loss: 0.9320271611213684:   7%|▋         | 421/6470 [04:49<1:04:13,  1.57it/s]
3|422|Loss: 0.9320271611213684:   7%|▋         | 422/6470 [04:50<1:06:22,  1.52it/s]
3|423|Loss: 0.8811360597610474:   7%|▋         | 422/6470 [04:50<1:06:22,  1.52it/s]
3|423|Loss: 0.8811360597610474:   7%|▋         | 423/6470 [04:50<1:07:52,  1.48it/s]
3|424|Loss: 0.914776086807251:   7%|▋         | 423/6470 [04:51<1:07:52,  1.48it/s] 
3|424|Loss: 0.914776086807251:   7%|▋         | 424/6470 [04:51<1:03:59,  1.57it/s]
3|425|Loss: 0.799824059009552:   7%|▋         | 424/6470 [04:51<1:03:59,  1.57it/s]
3|425|Loss: 0.799824059009552:   7%|▋         | 425/6470 [04:52<1:07:55,  1.48it/s]
3|426|Loss: 0.9508580565452576:   7%|▋         | 425/6470 [04:52<1:07:55,  1.48it/s]
3|426|Loss: 0.9508580565452576:   7%|▋         | 426/6470 [04:52<1:03:31,  1.59it/s]
3|427|Loss: 0.8925696015357971:   7%|▋         | 426/6470 [04:53<1:03:31,  1.59it/s]
3|427|Loss: 0.8925696015357971:   7%|▋         | 427/6470 [04:53<1:01:03,  1.65it/s]
3|428|Loss: 0.9960368275642395:   7%|▋         | 427/6470 [04:53<1:01:03,  1.65it/s]
3|428|Loss: 0.9960368275642395:   7%|▋         | 428/6470 [04:53<1:00:11,  1.67it/s]
3|429|Loss: 0.8959667682647705:   7%|▋         | 428/6470 [04:54<1:00:11,  1.67it/s]
3|429|Loss: 0.8959667682647705:   7%|▋         | 429/6470 [04:54<1:01:17,  1.64it/s]
3|430|Loss: 0.8179900646209717:   7%|▋         | 429/6470 [04:54<1:01:17,  1.64it/s]
3|430|Loss: 0.8179900646209717:   7%|▋         | 430/6470 [04:55<1:02:12,  1.62it/s]
3|431|Loss: 0.9118926525115967:   7%|▋         | 430/6470 [04:55<1:02:12,  1.62it/s]
3|431|Loss: 0.9118926525115967:   7%|▋         | 431/6470 [04:55<1:05:05,  1.55it/s]
3|432|Loss: 0.8095237612724304:   7%|▋         | 431/6470 [04:56<1:05:05,  1.55it/s]
3|432|Loss: 0.8095237612724304:   7%|▋         | 432/6470 [04:56<1:06:51,  1.51it/s]
3|433|Loss: 0.8911543488502502:   7%|▋         | 432/6470 [04:56<1:06:51,  1.51it/s]
3|433|Loss: 0.8911543488502502:   7%|▋         | 433/6470 [04:57<1:09:00,  1.46it/s]
3|434|Loss: 0.8454731106758118:   7%|▋         | 433/6470 [04:57<1:09:00,  1.46it/s]
3|434|Loss: 0.8454731106758118:   7%|▋         | 434/6470 [04:58<1:11:03,  1.42it/s]
3|435|Loss: 0.9293904900550842:   7%|▋         | 434/6470 [04:58<1:11:03,  1.42it/s]
3|435|Loss: 0.9293904900550842:   7%|▋         | 435/6470 [04:58<1:12:38,  1.38it/s]
3|436|Loss: 0.7010550498962402:   7%|▋         | 435/6470 [04:59<1:12:38,  1.38it/s]
3|436|Loss: 0.7010550498962402:   7%|▋         | 436/6470 [04:59<1:07:52,  1.48it/s]
3|437|Loss: 0.8420668840408325:   7%|▋         | 436/6470 [04:59<1:07:52,  1.48it/s]
3|437|Loss: 0.8420668840408325:   7%|▋         | 437/6470 [05:00<1:09:52,  1.44it/s]
3|438|Loss: 0.7141578197479248:   7%|▋         | 437/6470 [05:00<1:09:52,  1.44it/s]
3|438|Loss: 0.7141578197479248:   7%|▋         | 438/6470 [05:00<1:14:28,  1.35it/s]
3|439|Loss: 0.7012898325920105:   7%|▋         | 438/6470 [05:01<1:14:28,  1.35it/s]
3|439|Loss: 0.7012898325920105:   7%|▋         | 439/6470 [05:01<1:21:27,  1.23it/s]
3|440|Loss: 0.9701799154281616:   7%|▋         | 439/6470 [05:02<1:21:27,  1.23it/s]
3|440|Loss: 0.9701799154281616:   7%|▋         | 440/6470 [05:02<1:21:13,  1.24it/s]
3|441|Loss: 0.8421924114227295:   7%|▋         | 440/6470 [05:03<1:21:13,  1.24it/s]
3|441|Loss: 0.8421924114227295:   7%|▋         | 441/6470 [05:03<1:21:20,  1.24it/s]
3|442|Loss: 0.7542774081230164:   7%|▋         | 441/6470 [05:03<1:21:20,  1.24it/s]
3|442|Loss: 0.7542774081230164:   7%|▋         | 442/6470 [05:04<1:19:37,  1.26it/s]
3|443|Loss: 0.9716799855232239:   7%|▋         | 442/6470 [05:04<1:19:37,  1.26it/s]
3|443|Loss: 0.9716799855232239:   7%|▋         | 443/6470 [05:04<1:15:00,  1.34it/s]
3|444|Loss: 0.7985885143280029:   7%|▋         | 443/6470 [05:05<1:15:00,  1.34it/s]
3|444|Loss: 0.7985885143280029:   7%|▋         | 444/6470 [05:05<1:08:56,  1.46it/s]
3|445|Loss: 0.8033150434494019:   7%|▋         | 444/6470 [05:05<1:08:56,  1.46it/s]
3|445|Loss: 0.8033150434494019:   7%|▋         | 445/6470 [05:05<1:02:26,  1.61it/s]
3|446|Loss: 0.8862403035163879:   7%|▋         | 445/6470 [05:06<1:02:26,  1.61it/s]
3|446|Loss: 0.8862403035163879:   7%|▋         | 446/6470 [05:06<1:05:59,  1.52it/s]
3|447|Loss: 0.8597877621650696:   7%|▋         | 446/6470 [05:07<1:05:59,  1.52it/s]
3|447|Loss: 0.8597877621650696:   7%|▋         | 447/6470 [05:07<1:06:09,  1.52it/s]
3|448|Loss: 0.9279947876930237:   7%|▋         | 447/6470 [05:07<1:06:09,  1.52it/s]
3|448|Loss: 0.9279947876930237:   7%|▋         | 448/6470 [05:07<1:03:52,  1.57it/s]
3|449|Loss: 0.82049959897995:   7%|▋         | 448/6470 [05:08<1:03:52,  1.57it/s]  
3|449|Loss: 0.82049959897995:   7%|▋         | 449/6470 [05:08<1:01:40,  1.63it/s]
3|450|Loss: 0.7647032737731934:   7%|▋         | 449/6470 [05:08<1:01:40,  1.63it/s]
3|450|Loss: 0.7647032737731934:   7%|▋         | 450/6470 [05:09<1:06:28,  1.51it/s]
3|451|Loss: 0.8178578019142151:   7%|▋         | 450/6470 [05:09<1:06:28,  1.51it/s]
3|451|Loss: 0.8178578019142151:   7%|▋         | 451/6470 [05:09<1:06:22,  1.51it/s]
3|452|Loss: 0.9040875434875488:   7%|▋         | 451/6470 [05:10<1:06:22,  1.51it/s]
3|452|Loss: 0.9040875434875488:   7%|▋         | 452/6470 [05:10<1:05:28,  1.53it/s]
3|453|Loss: 0.8682751059532166:   7%|▋         | 452/6470 [05:10<1:05:28,  1.53it/s]
3|453|Loss: 0.8682751059532166:   7%|▋         | 453/6470 [05:11<1:03:39,  1.58it/s]
3|454|Loss: 0.7837480306625366:   7%|▋         | 453/6470 [05:11<1:03:39,  1.58it/s]
3|454|Loss: 0.7837480306625366:   7%|▋         | 454/6470 [05:11<1:05:37,  1.53it/s]
3|455|Loss: 0.8867166638374329:   7%|▋         | 454/6470 [05:12<1:05:37,  1.53it/s]
3|455|Loss: 0.8867166638374329:   7%|▋         | 455/6470 [05:12<1:01:16,  1.64it/s]
3|456|Loss: 0.887902557849884:   7%|▋         | 455/6470 [05:12<1:01:16,  1.64it/s] 
3|456|Loss: 0.887902557849884:   7%|▋         | 456/6470 [05:12<57:50,  1.73it/s]  
3|457|Loss: 0.8853698968887329:   7%|▋         | 456/6470 [05:13<57:50,  1.73it/s]
3|457|Loss: 0.8853698968887329:   7%|▋         | 457/6470 [05:13<52:20,  1.91it/s]
3|458|Loss: 0.7767297625541687:   7%|▋         | 457/6470 [05:13<52:20,  1.91it/s]
3|458|Loss: 0.7767297625541687:   7%|▋         | 458/6470 [05:13<58:21,  1.72it/s]
3|459|Loss: 0.9036898016929626:   7%|▋         | 458/6470 [05:14<58:21,  1.72it/s]
3|459|Loss: 0.9036898016929626:   7%|▋         | 459/6470 [05:14<1:03:29,  1.58it/s]
3|460|Loss: 0.849270224571228:   7%|▋         | 459/6470 [05:15<1:03:29,  1.58it/s] 
3|460|Loss: 0.849270224571228:   7%|▋         | 460/6470 [05:15<1:04:21,  1.56it/s]
3|461|Loss: 0.803863525390625:   7%|▋         | 460/6470 [05:15<1:04:21,  1.56it/s]
3|461|Loss: 0.803863525390625:   7%|▋         | 461/6470 [05:15<1:01:09,  1.64it/s]
3|462|Loss: 0.7143614292144775:   7%|▋         | 461/6470 [05:16<1:01:09,  1.64it/s]
3|462|Loss: 0.7143614292144775:   7%|▋         | 462/6470 [05:16<1:00:45,  1.65it/s]
3|463|Loss: 0.834527313709259:   7%|▋         | 462/6470 [05:16<1:00:45,  1.65it/s] 
3|463|Loss: 0.834527313709259:   7%|▋         | 463/6470 [05:17<1:01:08,  1.64it/s]
3|464|Loss: 1.007869005203247:   7%|▋         | 463/6470 [05:17<1:01:08,  1.64it/s]
3|464|Loss: 1.007869005203247:   7%|▋         | 464/6470 [05:17<1:01:58,  1.62it/s]
3|465|Loss: 0.8927080631256104:   7%|▋         | 464/6470 [05:18<1:01:58,  1.62it/s]
3|465|Loss: 0.8927080631256104:   7%|▋         | 465/6470 [05:18<1:02:37,  1.60it/s]
3|466|Loss: 0.731705904006958:   7%|▋         | 465/6470 [05:18<1:02:37,  1.60it/s] 
3|466|Loss: 0.731705904006958:   7%|▋         | 466/6470 [05:19<1:01:44,  1.62it/s]
3|467|Loss: 0.8830520510673523:   7%|▋         | 466/6470 [05:19<1:01:44,  1.62it/s]
3|467|Loss: 0.8830520510673523:   7%|▋         | 467/6470 [05:19<1:02:13,  1.61it/s]
3|468|Loss: 1.0496459007263184:   7%|▋         | 467/6470 [05:19<1:02:13,  1.61it/s]
3|468|Loss: 1.0496459007263184:   7%|▋         | 468/6470 [05:20<56:57,  1.76it/s]  
3|469|Loss: 0.8413833379745483:   7%|▋         | 468/6470 [05:20<56:57,  1.76it/s]
3|469|Loss: 0.8413833379745483:   7%|▋         | 469/6470 [05:20<1:03:01,  1.59it/s]
3|470|Loss: 1.1065304279327393:   7%|▋         | 469/6470 [05:21<1:03:01,  1.59it/s]
3|470|Loss: 1.1065304279327393:   7%|▋         | 470/6470 [05:21<1:06:22,  1.51it/s]
3|471|Loss: 0.941656231880188:   7%|▋         | 470/6470 [05:22<1:06:22,  1.51it/s] 
3|471|Loss: 0.941656231880188:   7%|▋         | 471/6470 [05:22<1:11:21,  1.40it/s]
3|472|Loss: 0.8584616780281067:   7%|▋         | 471/6470 [05:22<1:11:21,  1.40it/s]
3|472|Loss: 0.8584616780281067:   7%|▋         | 472/6470 [05:23<1:10:15,  1.42it/s]
3|473|Loss: 0.7411774396896362:   7%|▋         | 472/6470 [05:23<1:10:15,  1.42it/s]
3|473|Loss: 0.7411774396896362:   7%|▋         | 473/6470 [05:23<1:07:35,  1.48it/s]
3|474|Loss: 1.0626492500305176:   7%|▋         | 473/6470 [05:24<1:07:35,  1.48it/s]
3|474|Loss: 1.0626492500305176:   7%|▋         | 474/6470 [05:24<1:08:03,  1.47it/s]
3|475|Loss: 0.8137498497962952:   7%|▋         | 474/6470 [05:24<1:08:03,  1.47it/s]
3|475|Loss: 0.8137498497962952:   7%|▋         | 475/6470 [05:25<1:06:47,  1.50it/s]
3|476|Loss: 0.9664934277534485:   7%|▋         | 475/6470 [05:25<1:06:47,  1.50it/s]
3|476|Loss: 0.9664934277534485:   7%|▋         | 476/6470 [05:25<1:07:12,  1.49it/s]
3|477|Loss: 0.8216273188591003:   7%|▋         | 476/6470 [05:26<1:07:12,  1.49it/s]
3|477|Loss: 0.8216273188591003:   7%|▋         | 477/6470 [05:26<1:08:54,  1.45it/s]
3|478|Loss: 0.8825142979621887:   7%|▋         | 477/6470 [05:26<1:08:54,  1.45it/s]
3|478|Loss: 0.8825142979621887:   7%|▋         | 478/6470 [05:27<1:10:15,  1.42it/s]
3|479|Loss: 0.7823975682258606:   7%|▋         | 478/6470 [05:27<1:10:15,  1.42it/s]
3|479|Loss: 0.7823975682258606:   7%|▋         | 479/6470 [05:27<1:11:25,  1.40it/s]
3|480|Loss: 0.8499476909637451:   7%|▋         | 479/6470 [05:28<1:11:25,  1.40it/s]
3|480|Loss: 0.8499476909637451:   7%|▋         | 480/6470 [05:28<1:08:41,  1.45it/s]
3|481|Loss: 0.7446801066398621:   7%|▋         | 480/6470 [05:28<1:08:41,  1.45it/s]
3|481|Loss: 0.7446801066398621:   7%|▋         | 481/6470 [05:29<1:09:11,  1.44it/s]
3|482|Loss: 0.7791319489479065:   7%|▋         | 481/6470 [05:29<1:09:11,  1.44it/s]
3|482|Loss: 0.7791319489479065:   7%|▋         | 482/6470 [05:30<1:13:24,  1.36it/s]
3|483|Loss: 0.863709032535553:   7%|▋         | 482/6470 [05:30<1:13:24,  1.36it/s] 
3|483|Loss: 0.863709032535553:   7%|▋         | 483/6470 [05:30<1:13:40,  1.35it/s]
3|484|Loss: 0.8417018055915833:   7%|▋         | 483/6470 [05:31<1:13:40,  1.35it/s]
3|484|Loss: 0.8417018055915833:   7%|▋         | 484/6470 [05:31<1:14:51,  1.33it/s]
3|485|Loss: 0.7723287343978882:   7%|▋         | 484/6470 [05:32<1:14:51,  1.33it/s]
3|485|Loss: 0.7723287343978882:   7%|▋         | 485/6470 [05:32<1:12:00,  1.39it/s]
3|486|Loss: 0.8842329978942871:   7%|▋         | 485/6470 [05:32<1:12:00,  1.39it/s]
3|486|Loss: 0.8842329978942871:   8%|▊         | 486/6470 [05:32<1:10:53,  1.41it/s]
3|487|Loss: 0.7326567769050598:   8%|▊         | 486/6470 [05:33<1:10:53,  1.41it/s]
3|487|Loss: 0.7326567769050598:   8%|▊         | 487/6470 [05:33<1:08:45,  1.45it/s]
3|488|Loss: 0.8800028562545776:   8%|▊         | 487/6470 [05:33<1:08:45,  1.45it/s]
3|488|Loss: 0.8800028562545776:   8%|▊         | 488/6470 [05:34<1:09:49,  1.43it/s]
3|489|Loss: 0.969391405582428:   8%|▊         | 488/6470 [05:34<1:09:49,  1.43it/s] 
3|489|Loss: 0.969391405582428:   8%|▊         | 489/6470 [05:35<1:10:51,  1.41it/s]
3|490|Loss: 0.8947610855102539:   8%|▊         | 489/6470 [05:35<1:10:51,  1.41it/s]
3|490|Loss: 0.8947610855102539:   8%|▊         | 490/6470 [05:35<1:11:31,  1.39it/s]
3|491|Loss: 0.9152471423149109:   8%|▊         | 490/6470 [05:36<1:11:31,  1.39it/s]
3|491|Loss: 0.9152471423149109:   8%|▊         | 491/6470 [05:36<1:08:27,  1.46it/s]
3|492|Loss: 0.8313219547271729:   8%|▊         | 491/6470 [05:36<1:08:27,  1.46it/s]
3|492|Loss: 0.8313219547271729:   8%|▊         | 492/6470 [05:37<1:13:14,  1.36it/s]
3|493|Loss: 0.8625786304473877:   8%|▊         | 492/6470 [05:37<1:13:14,  1.36it/s]
3|493|Loss: 0.8625786304473877:   8%|▊         | 493/6470 [05:38<1:14:27,  1.34it/s]
3|494|Loss: 0.8646658062934875:   8%|▊         | 493/6470 [05:38<1:14:27,  1.34it/s]
3|494|Loss: 0.8646658062934875:   8%|▊         | 494/6470 [05:38<1:11:32,  1.39it/s]
3|495|Loss: 0.8492782115936279:   8%|▊         | 494/6470 [05:38<1:11:32,  1.39it/s]
3|495|Loss: 0.8492782115936279:   8%|▊         | 495/6470 [05:39<1:02:15,  1.60it/s]
3|496|Loss: 0.9132599234580994:   8%|▊         | 495/6470 [05:39<1:02:15,  1.60it/s]
3|496|Loss: 0.9132599234580994:   8%|▊         | 496/6470 [05:39<1:00:24,  1.65it/s]
3|497|Loss: 0.9330468773841858:   8%|▊         | 496/6470 [05:40<1:00:24,  1.65it/s]
3|497|Loss: 0.9330468773841858:   8%|▊         | 497/6470 [05:40<1:06:54,  1.49it/s]
3|498|Loss: 0.8399592638015747:   8%|▊         | 497/6470 [05:40<1:06:54,  1.49it/s]
3|498|Loss: 0.8399592638015747:   8%|▊         | 498/6470 [05:41<1:09:44,  1.43it/s]
3|499|Loss: 0.7478266954421997:   8%|▊         | 498/6470 [05:41<1:09:44,  1.43it/s]
3|499|Loss: 0.7478266954421997:   8%|▊         | 499/6470 [05:42<1:13:14,  1.36it/s]
3|500|Loss: 0.8346728086471558:   8%|▊         | 499/6470 [05:42<1:13:14,  1.36it/s]
3|500|Loss: 0.8346728086471558:   8%|▊         | 500/6470 [05:42<1:09:57,  1.42it/s]
3|501|Loss: 0.8473414778709412:   8%|▊         | 500/6470 [05:43<1:09:57,  1.42it/s]
3|501|Loss: 0.8473414778709412:   8%|▊         | 501/6470 [05:43<1:07:39,  1.47it/s]
3|502|Loss: 0.9738657474517822:   8%|▊         | 501/6470 [05:43<1:07:39,  1.47it/s]
3|502|Loss: 0.9738657474517822:   8%|▊         | 502/6470 [05:44<1:11:15,  1.40it/s]
3|503|Loss: 0.8682639002799988:   8%|▊         | 502/6470 [05:44<1:11:15,  1.40it/s]
3|503|Loss: 0.8682639002799988:   8%|▊         | 503/6470 [05:44<1:09:34,  1.43it/s]
3|504|Loss: 0.8344094753265381:   8%|▊         | 503/6470 [05:45<1:09:34,  1.43it/s]
3|504|Loss: 0.8344094753265381:   8%|▊         | 504/6470 [05:45<1:06:02,  1.51it/s]
3|505|Loss: 0.9384928345680237:   8%|▊         | 504/6470 [05:45<1:06:02,  1.51it/s]
3|505|Loss: 0.9384928345680237:   8%|▊         | 505/6470 [05:46<1:10:07,  1.42it/s]
3|506|Loss: 0.8121776580810547:   8%|▊         | 505/6470 [05:46<1:10:07,  1.42it/s]
3|506|Loss: 0.8121776580810547:   8%|▊         | 506/6470 [05:46<1:13:36,  1.35it/s]
3|507|Loss: 0.9068548083305359:   8%|▊         | 506/6470 [05:47<1:13:36,  1.35it/s]
3|507|Loss: 0.9068548083305359:   8%|▊         | 507/6470 [05:47<1:11:45,  1.38it/s]
3|508|Loss: 0.9114431738853455:   8%|▊         | 507/6470 [05:48<1:11:45,  1.38it/s]
3|508|Loss: 0.9114431738853455:   8%|▊         | 508/6470 [05:48<1:12:52,  1.36it/s]
3|509|Loss: 0.924784779548645:   8%|▊         | 508/6470 [05:48<1:12:52,  1.36it/s] 
3|509|Loss: 0.924784779548645:   8%|▊         | 509/6470 [05:49<1:08:26,  1.45it/s]
3|510|Loss: 0.8417297005653381:   8%|▊         | 509/6470 [05:49<1:08:26,  1.45it/s]
3|510|Loss: 0.8417297005653381:   8%|▊         | 510/6470 [05:49<1:06:16,  1.50it/s]
3|511|Loss: 0.8222954273223877:   8%|▊         | 510/6470 [05:50<1:06:16,  1.50it/s]
3|511|Loss: 0.8222954273223877:   8%|▊         | 511/6470 [05:50<1:05:40,  1.51it/s]
3|512|Loss: 0.7679402232170105:   8%|▊         | 511/6470 [05:50<1:05:40,  1.51it/s]
3|512|Loss: 0.7679402232170105:   8%|▊         | 512/6470 [05:50<1:03:27,  1.56it/s]
3|513|Loss: 0.9133824110031128:   8%|▊         | 512/6470 [05:51<1:03:27,  1.56it/s]
3|513|Loss: 0.9133824110031128:   8%|▊         | 513/6470 [05:51<1:05:39,  1.51it/s]
3|514|Loss: 0.9645679593086243:   8%|▊         | 513/6470 [05:51<1:05:39,  1.51it/s]
3|514|Loss: 0.9645679593086243:   8%|▊         | 514/6470 [05:52<1:04:17,  1.54it/s]
3|515|Loss: 0.8960580825805664:   8%|▊         | 514/6470 [05:52<1:04:17,  1.54it/s]
3|515|Loss: 0.8960580825805664:   8%|▊         | 515/6470 [05:52<1:02:13,  1.60it/s]
3|516|Loss: 0.871554434299469:   8%|▊         | 515/6470 [05:53<1:02:13,  1.60it/s] 
3|516|Loss: 0.871554434299469:   8%|▊         | 516/6470 [05:53<1:01:46,  1.61it/s]
3|517|Loss: 0.7872878313064575:   8%|▊         | 516/6470 [05:53<1:01:46,  1.61it/s]
3|517|Loss: 0.7872878313064575:   8%|▊         | 517/6470 [05:54<1:02:03,  1.60it/s]
3|518|Loss: 0.7853968739509583:   8%|▊         | 517/6470 [05:54<1:02:03,  1.60it/s]
3|518|Loss: 0.7853968739509583:   8%|▊         | 518/6470 [05:54<1:06:02,  1.50it/s]
3|519|Loss: 0.8183852434158325:   8%|▊         | 518/6470 [05:55<1:06:02,  1.50it/s]
3|519|Loss: 0.8183852434158325:   8%|▊         | 519/6470 [05:55<1:11:35,  1.39it/s]
3|520|Loss: 0.866868793964386:   8%|▊         | 519/6470 [05:56<1:11:35,  1.39it/s] 
3|520|Loss: 0.866868793964386:   8%|▊         | 520/6470 [05:56<1:15:11,  1.32it/s]
3|521|Loss: 0.9003412127494812:   8%|▊         | 520/6470 [05:56<1:15:11,  1.32it/s]
3|521|Loss: 0.9003412127494812:   8%|▊         | 521/6470 [05:57<1:10:19,  1.41it/s]
3|522|Loss: 0.8574023246765137:   8%|▊         | 521/6470 [05:57<1:10:19,  1.41it/s]
3|522|Loss: 0.8574023246765137:   8%|▊         | 522/6470 [05:57<1:07:32,  1.47it/s]
3|523|Loss: 0.886483907699585:   8%|▊         | 522/6470 [05:58<1:07:32,  1.47it/s] 
3|523|Loss: 0.886483907699585:   8%|▊         | 523/6470 [05:58<1:11:40,  1.38it/s]
3|524|Loss: 0.9057767987251282:   8%|▊         | 523/6470 [05:58<1:11:40,  1.38it/s]
3|524|Loss: 0.9057767987251282:   8%|▊         | 524/6470 [05:59<1:15:17,  1.32it/s]
3|525|Loss: 0.9631253480911255:   8%|▊         | 524/6470 [05:59<1:15:17,  1.32it/s]
3|525|Loss: 0.9631253480911255:   8%|▊         | 525/6470 [05:59<1:11:38,  1.38it/s]
3|526|Loss: 0.9263019561767578:   8%|▊         | 525/6470 [06:00<1:11:38,  1.38it/s]
3|526|Loss: 0.9263019561767578:   8%|▊         | 526/6470 [06:00<1:11:02,  1.39it/s]
3|527|Loss: 0.719009280204773:   8%|▊         | 526/6470 [06:01<1:11:02,  1.39it/s] 
3|527|Loss: 0.719009280204773:   8%|▊         | 527/6470 [06:01<1:12:49,  1.36it/s]
3|528|Loss: 0.7450994849205017:   8%|▊         | 527/6470 [06:01<1:12:49,  1.36it/s]
3|528|Loss: 0.7450994849205017:   8%|▊         | 528/6470 [06:02<1:13:06,  1.35it/s]
3|529|Loss: 0.8257790803909302:   8%|▊         | 528/6470 [06:02<1:13:06,  1.35it/s]
3|529|Loss: 0.8257790803909302:   8%|▊         | 529/6470 [06:02<1:12:19,  1.37it/s]
3|530|Loss: 0.7761492729187012:   8%|▊         | 529/6470 [06:03<1:12:19,  1.37it/s]
3|530|Loss: 0.7761492729187012:   8%|▊         | 530/6470 [06:03<1:11:19,  1.39it/s]
3|531|Loss: 0.9472872614860535:   8%|▊         | 530/6470 [06:03<1:11:19,  1.39it/s]
3|531|Loss: 0.9472872614860535:   8%|▊         | 531/6470 [06:04<1:07:07,  1.47it/s]
3|532|Loss: 0.8762199878692627:   8%|▊         | 531/6470 [06:04<1:07:07,  1.47it/s]
3|532|Loss: 0.8762199878692627:   8%|▊         | 532/6470 [06:05<1:11:03,  1.39it/s]
3|533|Loss: 0.6813810467720032:   8%|▊         | 532/6470 [06:05<1:11:03,  1.39it/s]
3|533|Loss: 0.6813810467720032:   8%|▊         | 533/6470 [06:05<1:16:44,  1.29it/s]
3|534|Loss: 0.8082644939422607:   8%|▊         | 533/6470 [06:06<1:16:44,  1.29it/s]
3|534|Loss: 0.8082644939422607:   8%|▊         | 534/6470 [06:06<1:14:24,  1.33it/s]
3|535|Loss: 1.0286270380020142:   8%|▊         | 534/6470 [06:07<1:14:24,  1.33it/s]
3|535|Loss: 1.0286270380020142:   8%|▊         | 535/6470 [06:07<1:12:41,  1.36it/s]
3|536|Loss: 0.8360951542854309:   8%|▊         | 535/6470 [06:07<1:12:41,  1.36it/s]
3|536|Loss: 0.8360951542854309:   8%|▊         | 536/6470 [06:07<1:06:32,  1.49it/s]
3|537|Loss: 0.8472989201545715:   8%|▊         | 536/6470 [06:08<1:06:32,  1.49it/s]
3|537|Loss: 0.8472989201545715:   8%|▊         | 537/6470 [06:08<1:11:11,  1.39it/s]
3|538|Loss: 0.8251233696937561:   8%|▊         | 537/6470 [06:09<1:11:11,  1.39it/s]
3|538|Loss: 0.8251233696937561:   8%|▊         | 538/6470 [06:09<1:13:05,  1.35it/s]
3|539|Loss: 0.8622859120368958:   8%|▊         | 538/6470 [06:09<1:13:05,  1.35it/s]
3|539|Loss: 0.8622859120368958:   8%|▊         | 539/6470 [06:10<1:13:27,  1.35it/s]
3|540|Loss: 0.7888336777687073:   8%|▊         | 539/6470 [06:10<1:13:27,  1.35it/s]
3|540|Loss: 0.7888336777687073:   8%|▊         | 540/6470 [06:10<1:11:10,  1.39it/s]
3|541|Loss: 0.7894598245620728:   8%|▊         | 540/6470 [06:11<1:11:10,  1.39it/s]
3|541|Loss: 0.7894598245620728:   8%|▊         | 541/6470 [06:11<1:10:58,  1.39it/s]
3|542|Loss: 0.9057499766349792:   8%|▊         | 541/6470 [06:12<1:10:58,  1.39it/s]
3|542|Loss: 0.9057499766349792:   8%|▊         | 542/6470 [06:12<1:11:10,  1.39it/s]
3|543|Loss: 0.8609279990196228:   8%|▊         | 542/6470 [06:12<1:11:10,  1.39it/s]
3|543|Loss: 0.8609279990196228:   8%|▊         | 543/6470 [06:12<1:06:07,  1.49it/s]
3|544|Loss: 0.8193185925483704:   8%|▊         | 543/6470 [06:13<1:06:07,  1.49it/s]
3|544|Loss: 0.8193185925483704:   8%|▊         | 544/6470 [06:13<1:06:52,  1.48it/s]
3|545|Loss: 0.8825575113296509:   8%|▊         | 544/6470 [06:13<1:06:52,  1.48it/s]
3|545|Loss: 0.8825575113296509:   8%|▊         | 545/6470 [06:14<1:04:20,  1.53it/s]
3|546|Loss: 0.8890438675880432:   8%|▊         | 545/6470 [06:14<1:04:20,  1.53it/s]
3|546|Loss: 0.8890438675880432:   8%|▊         | 546/6470 [06:15<1:10:03,  1.41it/s]
3|547|Loss: 0.8652203679084778:   8%|▊         | 546/6470 [06:15<1:10:03,  1.41it/s]
3|547|Loss: 0.8652203679084778:   8%|▊         | 547/6470 [06:15<1:13:14,  1.35it/s]
3|548|Loss: 0.9940358400344849:   8%|▊         | 547/6470 [06:16<1:13:14,  1.35it/s]
3|548|Loss: 0.9940358400344849:   8%|▊         | 548/6470 [06:16<1:12:47,  1.36it/s]
3|549|Loss: 0.9180943369865417:   8%|▊         | 548/6470 [06:16<1:12:47,  1.36it/s]
3|549|Loss: 0.9180943369865417:   8%|▊         | 549/6470 [06:17<1:12:28,  1.36it/s]
3|550|Loss: 0.7775648832321167:   8%|▊         | 549/6470 [06:17<1:12:28,  1.36it/s]
3|550|Loss: 0.7775648832321167:   9%|▊         | 550/6470 [06:18<1:17:54,  1.27it/s]
3|551|Loss: 0.8217750787734985:   9%|▊         | 550/6470 [06:18<1:17:54,  1.27it/s]
3|551|Loss: 0.8217750787734985:   9%|▊         | 551/6470 [06:18<1:18:25,  1.26it/s]
3|552|Loss: 0.8780907392501831:   9%|▊         | 551/6470 [06:19<1:18:25,  1.26it/s]
3|552|Loss: 0.8780907392501831:   9%|▊         | 552/6470 [06:19<1:14:05,  1.33it/s]
3|553|Loss: 0.7305482029914856:   9%|▊         | 552/6470 [06:19<1:14:05,  1.33it/s]
3|553|Loss: 0.7305482029914856:   9%|▊         | 553/6470 [06:20<1:09:53,  1.41it/s]
3|554|Loss: 0.8259639143943787:   9%|▊         | 553/6470 [06:20<1:09:53,  1.41it/s]
3|554|Loss: 0.8259639143943787:   9%|▊         | 554/6470 [06:20<1:07:45,  1.46it/s]
3|555|Loss: 0.8769844770431519:   9%|▊         | 554/6470 [06:21<1:07:45,  1.46it/s]
3|555|Loss: 0.8769844770431519:   9%|▊         | 555/6470 [06:21<1:06:19,  1.49it/s]
3|556|Loss: 0.7991821765899658:   9%|▊         | 555/6470 [06:21<1:06:19,  1.49it/s]
3|556|Loss: 0.7991821765899658:   9%|▊         | 556/6470 [06:22<1:07:21,  1.46it/s]
3|557|Loss: 0.8403463363647461:   9%|▊         | 556/6470 [06:22<1:07:21,  1.46it/s]
3|557|Loss: 0.8403463363647461:   9%|▊         | 557/6470 [06:22<1:06:04,  1.49it/s]
3|558|Loss: 0.8324236869812012:   9%|▊         | 557/6470 [06:23<1:06:04,  1.49it/s]
3|558|Loss: 0.8324236869812012:   9%|▊         | 558/6470 [06:23<1:12:21,  1.36it/s]
3|559|Loss: 1.1086294651031494:   9%|▊         | 558/6470 [06:24<1:12:21,  1.36it/s]
3|559|Loss: 1.1086294651031494:   9%|▊         | 559/6470 [06:24<1:10:02,  1.41it/s]
3|560|Loss: 0.8073006272315979:   9%|▊         | 559/6470 [06:24<1:10:02,  1.41it/s]
3|560|Loss: 0.8073006272315979:   9%|▊         | 560/6470 [06:24<1:05:28,  1.50it/s]
3|561|Loss: 0.8702059984207153:   9%|▊         | 560/6470 [06:25<1:05:28,  1.50it/s]
3|561|Loss: 0.8702059984207153:   9%|▊         | 561/6470 [06:25<1:05:45,  1.50it/s]
3|562|Loss: 0.9578555822372437:   9%|▊         | 561/6470 [06:26<1:05:45,  1.50it/s]
3|562|Loss: 0.9578555822372437:   9%|▊         | 562/6470 [06:26<1:03:57,  1.54it/s]
3|563|Loss: 0.8417794704437256:   9%|▊         | 562/6470 [06:26<1:03:57,  1.54it/s]
3|563|Loss: 0.8417794704437256:   9%|▊         | 563/6470 [06:26<1:06:00,  1.49it/s]
3|564|Loss: 0.8533405065536499:   9%|▊         | 563/6470 [06:27<1:06:00,  1.49it/s]
3|564|Loss: 0.8533405065536499:   9%|▊         | 564/6470 [06:27<1:05:25,  1.50it/s]
3|565|Loss: 0.9122471809387207:   9%|▊         | 564/6470 [06:27<1:05:25,  1.50it/s]
3|565|Loss: 0.9122471809387207:   9%|▊         | 565/6470 [06:28<1:02:59,  1.56it/s]
3|566|Loss: 0.8896287083625793:   9%|▊         | 565/6470 [06:28<1:02:59,  1.56it/s]
3|566|Loss: 0.8896287083625793:   9%|▊         | 566/6470 [06:28<1:02:02,  1.59it/s]
3|567|Loss: 0.7356534004211426:   9%|▊         | 566/6470 [06:29<1:02:02,  1.59it/s]
3|567|Loss: 0.7356534004211426:   9%|▉         | 567/6470 [06:29<1:04:11,  1.53it/s]
3|568|Loss: 0.9506083726882935:   9%|▉         | 567/6470 [06:29<1:04:11,  1.53it/s]
3|568|Loss: 0.9506083726882935:   9%|▉         | 568/6470 [06:30<1:04:10,  1.53it/s]
3|569|Loss: 0.8390833735466003:   9%|▉         | 568/6470 [06:30<1:04:10,  1.53it/s]
3|569|Loss: 0.8390833735466003:   9%|▉         | 569/6470 [06:30<1:06:35,  1.48it/s]
3|570|Loss: 0.7534211874008179:   9%|▉         | 569/6470 [06:31<1:06:35,  1.48it/s]
3|570|Loss: 0.7534211874008179:   9%|▉         | 570/6470 [06:31<1:07:50,  1.45it/s]
3|571|Loss: 0.6901229619979858:   9%|▉         | 570/6470 [06:32<1:07:50,  1.45it/s]
3|571|Loss: 0.6901229619979858:   9%|▉         | 571/6470 [06:32<1:11:54,  1.37it/s]
3|572|Loss: 0.9027739763259888:   9%|▉         | 571/6470 [06:32<1:11:54,  1.37it/s]
3|572|Loss: 0.9027739763259888:   9%|▉         | 572/6470 [06:33<1:14:18,  1.32it/s]
3|573|Loss: 0.771460771560669:   9%|▉         | 572/6470 [06:33<1:14:18,  1.32it/s] 
3|573|Loss: 0.771460771560669:   9%|▉         | 573/6470 [06:34<1:18:13,  1.26it/s]
3|574|Loss: 0.8700003623962402:   9%|▉         | 573/6470 [06:34<1:18:13,  1.26it/s]
3|574|Loss: 0.8700003623962402:   9%|▉         | 574/6470 [06:34<1:13:54,  1.33it/s]
3|575|Loss: 0.8952977061271667:   9%|▉         | 574/6470 [06:35<1:13:54,  1.33it/s]
3|575|Loss: 0.8952977061271667:   9%|▉         | 575/6470 [06:35<1:15:25,  1.30it/s]
3|576|Loss: 0.8431397080421448:   9%|▉         | 575/6470 [06:36<1:15:25,  1.30it/s]
3|576|Loss: 0.8431397080421448:   9%|▉         | 576/6470 [06:36<1:16:36,  1.28it/s]
3|577|Loss: 0.8456221222877502:   9%|▉         | 576/6470 [06:36<1:16:36,  1.28it/s]
3|577|Loss: 0.8456221222877502:   9%|▉         | 577/6470 [06:37<1:14:56,  1.31it/s]
3|578|Loss: 0.8288562297821045:   9%|▉         | 577/6470 [06:37<1:14:56,  1.31it/s]
3|578|Loss: 0.8288562297821045:   9%|▉         | 578/6470 [06:37<1:11:20,  1.38it/s]
3|579|Loss: 0.761282742023468:   9%|▉         | 578/6470 [06:38<1:11:20,  1.38it/s] 
3|579|Loss: 0.761282742023468:   9%|▉         | 579/6470 [06:38<1:11:14,  1.38it/s]
3|580|Loss: 0.8915620446205139:   9%|▉         | 579/6470 [06:38<1:11:14,  1.38it/s]
3|580|Loss: 0.8915620446205139:   9%|▉         | 580/6470 [06:39<1:13:08,  1.34it/s]
3|581|Loss: 0.8874731659889221:   9%|▉         | 580/6470 [06:39<1:13:08,  1.34it/s]
3|581|Loss: 0.8874731659889221:   9%|▉         | 581/6470 [06:40<1:12:35,  1.35it/s]
3|582|Loss: 0.8729162216186523:   9%|▉         | 581/6470 [06:40<1:12:35,  1.35it/s]
3|582|Loss: 0.8729162216186523:   9%|▉         | 582/6470 [06:40<1:09:33,  1.41it/s]
3|583|Loss: 0.949241578578949:   9%|▉         | 582/6470 [06:41<1:09:33,  1.41it/s] 
3|583|Loss: 0.949241578578949:   9%|▉         | 583/6470 [06:41<1:09:29,  1.41it/s]
3|584|Loss: 0.7847610712051392:   9%|▉         | 583/6470 [06:41<1:09:29,  1.41it/s]
3|584|Loss: 0.7847610712051392:   9%|▉         | 584/6470 [06:41<1:04:48,  1.51it/s]
3|585|Loss: 1.0109925270080566:   9%|▉         | 584/6470 [06:42<1:04:48,  1.51it/s]
3|585|Loss: 1.0109925270080566:   9%|▉         | 585/6470 [06:42<1:04:42,  1.52it/s]
3|586|Loss: 0.8973239064216614:   9%|▉         | 585/6470 [06:42<1:04:42,  1.52it/s]
3|586|Loss: 0.8973239064216614:   9%|▉         | 586/6470 [06:43<1:06:09,  1.48it/s]
3|587|Loss: 0.7963172197341919:   9%|▉         | 586/6470 [06:43<1:06:09,  1.48it/s]
3|587|Loss: 0.7963172197341919:   9%|▉         | 587/6470 [06:44<1:08:05,  1.44it/s]
3|588|Loss: 0.9547374248504639:   9%|▉         | 587/6470 [06:44<1:08:05,  1.44it/s]
3|588|Loss: 0.9547374248504639:   9%|▉         | 588/6470 [06:44<1:09:53,  1.40it/s]
3|589|Loss: 0.9443961977958679:   9%|▉         | 588/6470 [06:45<1:09:53,  1.40it/s]
3|589|Loss: 0.9443961977958679:   9%|▉         | 589/6470 [06:45<1:11:06,  1.38it/s]
3|590|Loss: 1.0017603635787964:   9%|▉         | 589/6470 [06:45<1:11:06,  1.38it/s]
3|590|Loss: 1.0017603635787964:   9%|▉         | 590/6470 [06:46<1:11:47,  1.37it/s]
3|591|Loss: 0.8423658013343811:   9%|▉         | 590/6470 [06:46<1:11:47,  1.37it/s]
3|591|Loss: 0.8423658013343811:   9%|▉         | 591/6470 [06:47<1:12:03,  1.36it/s]
3|592|Loss: 0.7801231741905212:   9%|▉         | 591/6470 [06:47<1:12:03,  1.36it/s]
3|592|Loss: 0.7801231741905212:   9%|▉         | 592/6470 [06:47<1:14:57,  1.31it/s]
3|593|Loss: 0.8239200711250305:   9%|▉         | 592/6470 [06:48<1:14:57,  1.31it/s]
3|593|Loss: 0.8239200711250305:   9%|▉         | 593/6470 [06:48<1:12:13,  1.36it/s]
3|594|Loss: 0.9378395676612854:   9%|▉         | 593/6470 [06:48<1:12:13,  1.36it/s]
3|594|Loss: 0.9378395676612854:   9%|▉         | 594/6470 [06:49<1:13:38,  1.33it/s]
3|595|Loss: 0.817720890045166:   9%|▉         | 594/6470 [06:49<1:13:38,  1.33it/s] 
3|595|Loss: 0.817720890045166:   9%|▉         | 595/6470 [06:49<1:08:33,  1.43it/s]
3|596|Loss: 0.8297885656356812:   9%|▉         | 595/6470 [06:50<1:08:33,  1.43it/s]
3|596|Loss: 0.8297885656356812:   9%|▉         | 596/6470 [06:50<1:03:54,  1.53it/s]
3|597|Loss: 0.7315740585327148:   9%|▉         | 596/6470 [06:50<1:03:54,  1.53it/s]
3|597|Loss: 0.7315740585327148:   9%|▉         | 597/6470 [06:51<1:01:45,  1.59it/s]
3|598|Loss: 1.0224058628082275:   9%|▉         | 597/6470 [06:51<1:01:45,  1.59it/s]
3|598|Loss: 1.0224058628082275:   9%|▉         | 598/6470 [06:51<1:01:40,  1.59it/s]
3|599|Loss: 0.8480312824249268:   9%|▉         | 598/6470 [06:52<1:01:40,  1.59it/s]
3|599|Loss: 0.8480312824249268:   9%|▉         | 599/6470 [06:52<1:04:07,  1.53it/s]
3|600|Loss: 0.8387867212295532:   9%|▉         | 599/6470 [06:52<1:04:07,  1.53it/s]
3|600|Loss: 0.8387867212295532:   9%|▉         | 600/6470 [06:53<1:03:48,  1.53it/s]
3|601|Loss: 0.7871684432029724:   9%|▉         | 600/6470 [06:53<1:03:48,  1.53it/s]
3|601|Loss: 0.7871684432029724:   9%|▉         | 601/6470 [06:53<1:01:42,  1.59it/s]
3|602|Loss: 0.816408097743988:   9%|▉         | 601/6470 [06:53<1:01:42,  1.59it/s] 
3|602|Loss: 0.816408097743988:   9%|▉         | 602/6470 [06:54<1:05:14,  1.50it/s]
3|603|Loss: 0.9391854405403137:   9%|▉         | 602/6470 [06:54<1:05:14,  1.50it/s]
3|603|Loss: 0.9391854405403137:   9%|▉         | 603/6470 [06:54<1:03:24,  1.54it/s]
3|604|Loss: 0.95137619972229:   9%|▉         | 603/6470 [06:55<1:03:24,  1.54it/s]  
3|604|Loss: 0.95137619972229:   9%|▉         | 604/6470 [06:55<1:04:51,  1.51it/s]
3|605|Loss: 0.8474839329719543:   9%|▉         | 604/6470 [06:56<1:04:51,  1.51it/s]
3|605|Loss: 0.8474839329719543:   9%|▉         | 605/6470 [06:56<1:07:12,  1.45it/s]
3|606|Loss: 0.8150806427001953:   9%|▉         | 605/6470 [06:56<1:07:12,  1.45it/s]
3|606|Loss: 0.8150806427001953:   9%|▉         | 606/6470 [06:57<1:09:05,  1.41it/s]
3|607|Loss: 0.8736886382102966:   9%|▉         | 606/6470 [06:57<1:09:05,  1.41it/s]
3|607|Loss: 0.8736886382102966:   9%|▉         | 607/6470 [06:57<1:09:56,  1.40it/s]
3|608|Loss: 0.9155718684196472:   9%|▉         | 607/6470 [06:58<1:09:56,  1.40it/s]
3|608|Loss: 0.9155718684196472:   9%|▉         | 608/6470 [06:58<1:10:32,  1.38it/s]
3|609|Loss: 0.8255487680435181:   9%|▉         | 608/6470 [06:58<1:10:32,  1.38it/s]
3|609|Loss: 0.8255487680435181:   9%|▉         | 609/6470 [06:59<1:05:39,  1.49it/s]
3|610|Loss: 0.8251726627349854:   9%|▉         | 609/6470 [06:59<1:05:39,  1.49it/s]
3|610|Loss: 0.8251726627349854:   9%|▉         | 610/6470 [06:59<1:05:39,  1.49it/s]
3|611|Loss: 0.6789649128913879:   9%|▉         | 610/6470 [07:00<1:05:39,  1.49it/s]
3|611|Loss: 0.6789649128913879:   9%|▉         | 611/6470 [07:00<1:07:15,  1.45it/s]
3|612|Loss: 0.9389522671699524:   9%|▉         | 611/6470 [07:01<1:07:15,  1.45it/s]
3|612|Loss: 0.9389522671699524:   9%|▉         | 612/6470 [07:01<1:12:03,  1.35it/s]
3|613|Loss: 0.7588596940040588:   9%|▉         | 612/6470 [07:01<1:12:03,  1.35it/s]
3|613|Loss: 0.7588596940040588:   9%|▉         | 613/6470 [07:02<1:13:11,  1.33it/s]
3|614|Loss: 0.8409066200256348:   9%|▉         | 613/6470 [07:02<1:13:11,  1.33it/s]
3|614|Loss: 0.8409066200256348:   9%|▉         | 614/6470 [07:02<1:10:36,  1.38it/s]
3|615|Loss: 0.7898264527320862:   9%|▉         | 614/6470 [07:03<1:10:36,  1.38it/s]
3|615|Loss: 0.7898264527320862:  10%|▉         | 615/6470 [07:03<1:09:29,  1.40it/s]
3|616|Loss: 0.9422829151153564:  10%|▉         | 615/6470 [07:03<1:09:29,  1.40it/s]
3|616|Loss: 0.9422829151153564:  10%|▉         | 616/6470 [07:04<1:12:01,  1.35it/s]
3|617|Loss: 0.85467928647995:  10%|▉         | 616/6470 [07:04<1:12:01,  1.35it/s]  
3|617|Loss: 0.85467928647995:  10%|▉         | 617/6470 [07:04<1:07:20,  1.45it/s]
3|618|Loss: 0.890800952911377:  10%|▉         | 617/6470 [07:05<1:07:20,  1.45it/s]
3|618|Loss: 0.890800952911377:  10%|▉         | 618/6470 [07:05<1:05:06,  1.50it/s]
3|619|Loss: 0.8021606206893921:  10%|▉         | 618/6470 [07:06<1:05:06,  1.50it/s]
3|619|Loss: 0.8021606206893921:  10%|▉         | 619/6470 [07:06<1:15:30,  1.29it/s]
3|620|Loss: 0.8292388319969177:  10%|▉         | 619/6470 [07:07<1:15:30,  1.29it/s]
3|620|Loss: 0.8292388319969177:  10%|▉         | 620/6470 [07:07<1:14:41,  1.31it/s]
3|621|Loss: 0.948161244392395:  10%|▉         | 620/6470 [07:07<1:14:41,  1.31it/s] 
3|621|Loss: 0.948161244392395:  10%|▉         | 621/6470 [07:08<1:12:29,  1.34it/s]
3|622|Loss: 0.7977648973464966:  10%|▉         | 621/6470 [07:08<1:12:29,  1.34it/s]
3|622|Loss: 0.7977648973464966:  10%|▉         | 622/6470 [07:08<1:16:07,  1.28it/s]
3|623|Loss: 0.8555946350097656:  10%|▉         | 622/6470 [07:09<1:16:07,  1.28it/s]
3|623|Loss: 0.8555946350097656:  10%|▉         | 623/6470 [07:09<1:15:54,  1.28it/s]
3|624|Loss: 0.8814004063606262:  10%|▉         | 623/6470 [07:10<1:15:54,  1.28it/s]
3|624|Loss: 0.8814004063606262:  10%|▉         | 624/6470 [07:10<1:14:53,  1.30it/s]
3|625|Loss: 0.723583459854126:  10%|▉         | 624/6470 [07:10<1:14:53,  1.30it/s] 
3|625|Loss: 0.723583459854126:  10%|▉         | 625/6470 [07:11<1:16:43,  1.27it/s]
3|626|Loss: 0.7405055165290833:  10%|▉         | 625/6470 [07:11<1:16:43,  1.27it/s]
3|626|Loss: 0.7405055165290833:  10%|▉         | 626/6470 [07:11<1:13:58,  1.32it/s]
3|627|Loss: 0.8909727334976196:  10%|▉         | 626/6470 [07:12<1:13:58,  1.32it/s]
3|627|Loss: 0.8909727334976196:  10%|▉         | 627/6470 [07:12<1:11:56,  1.35it/s]
3|628|Loss: 0.7391266822814941:  10%|▉         | 627/6470 [07:12<1:11:56,  1.35it/s]
3|628|Loss: 0.7391266822814941:  10%|▉         | 628/6470 [07:13<1:07:07,  1.45it/s]
3|629|Loss: 0.9303786158561707:  10%|▉         | 628/6470 [07:13<1:07:07,  1.45it/s]
3|629|Loss: 0.9303786158561707:  10%|▉         | 629/6470 [07:13<1:08:45,  1.42it/s]
3|630|Loss: 0.8090436458587646:  10%|▉         | 629/6470 [07:14<1:08:45,  1.42it/s]
3|630|Loss: 0.8090436458587646:  10%|▉         | 630/6470 [07:14<1:09:26,  1.40it/s]
3|631|Loss: 0.8503614664077759:  10%|▉         | 630/6470 [07:15<1:09:26,  1.40it/s]
3|631|Loss: 0.8503614664077759:  10%|▉         | 631/6470 [07:15<1:07:15,  1.45it/s]
3|632|Loss: 0.7807258367538452:  10%|▉         | 631/6470 [07:15<1:07:15,  1.45it/s]
3|632|Loss: 0.7807258367538452:  10%|▉         | 632/6470 [07:15<1:01:09,  1.59it/s]
3|633|Loss: 0.8446615934371948:  10%|▉         | 632/6470 [07:16<1:01:09,  1.59it/s]
3|633|Loss: 0.8446615934371948:  10%|▉         | 633/6470 [07:16<1:00:06,  1.62it/s]
3|634|Loss: 0.846545934677124:  10%|▉         | 633/6470 [07:16<1:00:06,  1.62it/s] 
3|634|Loss: 0.846545934677124:  10%|▉         | 634/6470 [07:17<1:00:47,  1.60it/s]
3|635|Loss: 0.7020496129989624:  10%|▉         | 634/6470 [07:17<1:00:47,  1.60it/s]
3|635|Loss: 0.7020496129989624:  10%|▉         | 635/6470 [07:17<1:00:43,  1.60it/s]
3|636|Loss: 0.6391738057136536:  10%|▉         | 635/6470 [07:18<1:00:43,  1.60it/s]
3|636|Loss: 0.6391738057136536:  10%|▉         | 636/6470 [07:18<1:06:05,  1.47it/s]
3|637|Loss: 0.862205445766449:  10%|▉         | 636/6470 [07:18<1:06:05,  1.47it/s] 
3|637|Loss: 0.862205445766449:  10%|▉         | 637/6470 [07:19<1:10:01,  1.39it/s]
3|638|Loss: 0.9137789011001587:  10%|▉         | 637/6470 [07:19<1:10:01,  1.39it/s]
3|638|Loss: 0.9137789011001587:  10%|▉         | 638/6470 [07:19<1:08:34,  1.42it/s]
3|639|Loss: 0.8693317174911499:  10%|▉         | 638/6470 [07:20<1:08:34,  1.42it/s]
3|639|Loss: 0.8693317174911499:  10%|▉         | 639/6470 [07:20<1:06:02,  1.47it/s]
3|640|Loss: 0.9396121501922607:  10%|▉         | 639/6470 [07:20<1:06:02,  1.47it/s]
3|640|Loss: 0.9396121501922607:  10%|▉         | 640/6470 [07:21<1:08:34,  1.42it/s]
3|641|Loss: 0.860758364200592:  10%|▉         | 640/6470 [07:21<1:08:34,  1.42it/s] 
3|641|Loss: 0.860758364200592:  10%|▉         | 641/6470 [07:21<1:05:18,  1.49it/s]
3|642|Loss: 0.8580188751220703:  10%|▉         | 641/6470 [07:22<1:05:18,  1.49it/s]
3|642|Loss: 0.8580188751220703:  10%|▉         | 642/6470 [07:22<1:05:48,  1.48it/s]
3|643|Loss: 0.759584903717041:  10%|▉         | 642/6470 [07:23<1:05:48,  1.48it/s] 
3|643|Loss: 0.759584903717041:  10%|▉         | 643/6470 [07:23<1:06:41,  1.46it/s]
3|644|Loss: 0.8125154972076416:  10%|▉         | 643/6470 [07:23<1:06:41,  1.46it/s]
3|644|Loss: 0.8125154972076416:  10%|▉         | 644/6470 [07:24<1:07:44,  1.43it/s]
3|645|Loss: 0.8677665591239929:  10%|▉         | 644/6470 [07:24<1:07:44,  1.43it/s]
3|645|Loss: 0.8677665591239929:  10%|▉         | 645/6470 [07:24<1:08:05,  1.43it/s]
3|646|Loss: 0.9415516257286072:  10%|▉         | 645/6470 [07:25<1:08:05,  1.43it/s]
3|646|Loss: 0.9415516257286072:  10%|▉         | 646/6470 [07:25<1:06:53,  1.45it/s]
3|647|Loss: 0.862077534198761:  10%|▉         | 646/6470 [07:25<1:06:53,  1.45it/s] 
3|647|Loss: 0.862077534198761:  10%|█         | 647/6470 [07:26<1:04:56,  1.49it/s]
3|648|Loss: 0.6764406561851501:  10%|█         | 647/6470 [07:26<1:04:56,  1.49it/s]
3|648|Loss: 0.6764406561851501:  10%|█         | 648/6470 [07:26<59:40,  1.63it/s]  
3|649|Loss: 0.9037491679191589:  10%|█         | 648/6470 [07:26<59:40,  1.63it/s]
3|649|Loss: 0.9037491679191589:  10%|█         | 649/6470 [07:27<1:06:40,  1.46it/s]
3|650|Loss: 0.8003734350204468:  10%|█         | 649/6470 [07:27<1:06:40,  1.46it/s]
3|650|Loss: 0.8003734350204468:  10%|█         | 650/6470 [07:28<1:11:00,  1.37it/s]
3|651|Loss: 0.9050953984260559:  10%|█         | 650/6470 [07:28<1:11:00,  1.37it/s]
3|651|Loss: 0.9050953984260559:  10%|█         | 651/6470 [07:29<1:13:05,  1.33it/s]
3|652|Loss: 0.8824089169502258:  10%|█         | 651/6470 [07:29<1:13:05,  1.33it/s]
3|652|Loss: 0.8824089169502258:  10%|█         | 652/6470 [07:29<1:15:08,  1.29it/s]
3|653|Loss: 0.8191450238227844:  10%|█         | 652/6470 [07:30<1:15:08,  1.29it/s]
3|653|Loss: 0.8191450238227844:  10%|█         | 653/6470 [07:30<1:13:48,  1.31it/s]
3|654|Loss: 0.8554269671440125:  10%|█         | 653/6470 [07:30<1:13:48,  1.31it/s]
3|654|Loss: 0.8554269671440125:  10%|█         | 654/6470 [07:31<1:14:03,  1.31it/s]
3|655|Loss: 0.7893235087394714:  10%|█         | 654/6470 [07:31<1:14:03,  1.31it/s]
3|655|Loss: 0.7893235087394714:  10%|█         | 655/6470 [07:32<1:17:54,  1.24it/s]
3|656|Loss: 0.8281589150428772:  10%|█         | 655/6470 [07:32<1:17:54,  1.24it/s]
3|656|Loss: 0.8281589150428772:  10%|█         | 656/6470 [07:32<1:14:59,  1.29it/s]
3|657|Loss: 0.8891476392745972:  10%|█         | 656/6470 [07:33<1:14:59,  1.29it/s]
3|657|Loss: 0.8891476392745972:  10%|█         | 657/6470 [07:33<1:13:38,  1.32it/s]
3|658|Loss: 0.8543043732643127:  10%|█         | 657/6470 [07:34<1:13:38,  1.32it/s]
3|658|Loss: 0.8543043732643127:  10%|█         | 658/6470 [07:34<1:09:35,  1.39it/s]
3|659|Loss: 0.836706817150116:  10%|█         | 658/6470 [07:34<1:09:35,  1.39it/s] 
3|659|Loss: 0.836706817150116:  10%|█         | 659/6470 [07:34<1:07:07,  1.44it/s]
3|660|Loss: 0.8250141143798828:  10%|█         | 659/6470 [07:35<1:07:07,  1.44it/s]
3|660|Loss: 0.8250141143798828:  10%|█         | 660/6470 [07:35<1:08:09,  1.42it/s]
3|661|Loss: 0.9457192420959473:  10%|█         | 660/6470 [07:36<1:08:09,  1.42it/s]
3|661|Loss: 0.9457192420959473:  10%|█         | 661/6470 [07:36<1:10:57,  1.36it/s]
3|662|Loss: 0.7440887093544006:  10%|█         | 661/6470 [07:36<1:10:57,  1.36it/s]
3|662|Loss: 0.7440887093544006:  10%|█         | 662/6470 [07:37<1:05:50,  1.47it/s]
3|663|Loss: 0.846847653388977:  10%|█         | 662/6470 [07:37<1:05:50,  1.47it/s] 
3|663|Loss: 0.846847653388977:  10%|█         | 663/6470 [07:37<1:10:06,  1.38it/s]
3|664|Loss: 0.8251544833183289:  10%|█         | 663/6470 [07:38<1:10:06,  1.38it/s]
3|664|Loss: 0.8251544833183289:  10%|█         | 664/6470 [07:38<1:13:10,  1.32it/s]
3|665|Loss: 0.8849447965621948:  10%|█         | 664/6470 [07:39<1:13:10,  1.32it/s]
3|665|Loss: 0.8849447965621948:  10%|█         | 665/6470 [07:39<1:15:43,  1.28it/s]
3|666|Loss: 0.8224693536758423:  10%|█         | 665/6470 [07:39<1:15:43,  1.28it/s]
3|666|Loss: 0.8224693536758423:  10%|█         | 666/6470 [07:40<1:14:30,  1.30it/s]
3|667|Loss: 0.6508649587631226:  10%|█         | 666/6470 [07:40<1:14:30,  1.30it/s]
3|667|Loss: 0.6508649587631226:  10%|█         | 667/6470 [07:41<1:16:01,  1.27it/s]
3|668|Loss: 0.856109082698822:  10%|█         | 667/6470 [07:41<1:16:01,  1.27it/s] 
3|668|Loss: 0.856109082698822:  10%|█         | 668/6470 [07:41<1:12:45,  1.33it/s]
3|669|Loss: 0.7254130840301514:  10%|█         | 668/6470 [07:42<1:12:45,  1.33it/s]
3|669|Loss: 0.7254130840301514:  10%|█         | 669/6470 [07:42<1:04:58,  1.49it/s]
3|670|Loss: 1.0069290399551392:  10%|█         | 669/6470 [07:42<1:04:58,  1.49it/s]
3|670|Loss: 1.0069290399551392:  10%|█         | 670/6470 [07:42<1:01:16,  1.58it/s]
3|671|Loss: 0.9268692135810852:  10%|█         | 670/6470 [07:43<1:01:16,  1.58it/s]
3|671|Loss: 0.9268692135810852:  10%|█         | 671/6470 [07:43<1:03:48,  1.51it/s]
3|672|Loss: 0.8957903385162354:  10%|█         | 671/6470 [07:43<1:03:48,  1.51it/s]
3|672|Loss: 0.8957903385162354:  10%|█         | 672/6470 [07:44<1:01:05,  1.58it/s]
3|673|Loss: 0.8401978611946106:  10%|█         | 672/6470 [07:44<1:01:05,  1.58it/s]
3|673|Loss: 0.8401978611946106:  10%|█         | 673/6470 [07:44<57:40,  1.68it/s]  
3|674|Loss: 0.847886860370636:  10%|█         | 673/6470 [07:44<57:40,  1.68it/s] 
3|674|Loss: 0.847886860370636:  10%|█         | 674/6470 [07:45<58:13,  1.66it/s]
3|675|Loss: 0.8639010190963745:  10%|█         | 674/6470 [07:45<58:13,  1.66it/s]
3|675|Loss: 0.8639010190963745:  10%|█         | 675/6470 [07:45<1:01:53,  1.56it/s]
3|676|Loss: 0.7270607352256775:  10%|█         | 675/6470 [07:46<1:01:53,  1.56it/s]
3|676|Loss: 0.7270607352256775:  10%|█         | 676/6470 [07:46<1:04:14,  1.50it/s]
3|677|Loss: 0.8346324563026428:  10%|█         | 676/6470 [07:47<1:04:14,  1.50it/s]
3|677|Loss: 0.8346324563026428:  10%|█         | 677/6470 [07:47<1:03:52,  1.51it/s]
3|678|Loss: 0.9173700213432312:  10%|█         | 677/6470 [07:47<1:03:52,  1.51it/s]
3|678|Loss: 0.9173700213432312:  10%|█         | 678/6470 [07:47<1:04:46,  1.49it/s]
3|679|Loss: 0.8562072515487671:  10%|█         | 678/6470 [07:48<1:04:46,  1.49it/s]
3|679|Loss: 0.8562072515487671:  10%|█         | 679/6470 [07:48<1:02:50,  1.54it/s]
3|680|Loss: 0.8128505349159241:  10%|█         | 679/6470 [07:49<1:02:50,  1.54it/s]
3|680|Loss: 0.8128505349159241:  11%|█         | 680/6470 [07:49<1:07:18,  1.43it/s]
3|681|Loss: 0.7992923259735107:  11%|█         | 680/6470 [07:49<1:07:18,  1.43it/s]
3|681|Loss: 0.7992923259735107:  11%|█         | 681/6470 [07:50<1:09:39,  1.39it/s]
3|682|Loss: 0.8317980766296387:  11%|█         | 681/6470 [07:50<1:09:39,  1.39it/s]
3|682|Loss: 0.8317980766296387:  11%|█         | 682/6470 [07:50<1:06:38,  1.45it/s]
3|683|Loss: 0.8264924883842468:  11%|█         | 682/6470 [07:51<1:06:38,  1.45it/s]
3|683|Loss: 0.8264924883842468:  11%|█         | 683/6470 [07:51<1:04:49,  1.49it/s]
3|684|Loss: 1.1539251804351807:  11%|█         | 683/6470 [07:51<1:04:49,  1.49it/s]
3|684|Loss: 1.1539251804351807:  11%|█         | 684/6470 [07:52<1:02:48,  1.54it/s]
3|685|Loss: 0.8574686646461487:  11%|█         | 684/6470 [07:52<1:02:48,  1.54it/s]
3|685|Loss: 0.8574686646461487:  11%|█         | 685/6470 [07:52<1:04:38,  1.49it/s]
3|686|Loss: 0.6190696954727173:  11%|█         | 685/6470 [07:53<1:04:38,  1.49it/s]
3|686|Loss: 0.6190696954727173:  11%|█         | 686/6470 [07:53<1:10:28,  1.37it/s]
3|687|Loss: 1.0388416051864624:  11%|█         | 686/6470 [07:54<1:10:28,  1.37it/s]
3|687|Loss: 1.0388416051864624:  11%|█         | 687/6470 [07:54<1:11:35,  1.35it/s]
3|688|Loss: 0.8228809237480164:  11%|█         | 687/6470 [07:54<1:11:35,  1.35it/s]
3|688|Loss: 0.8228809237480164:  11%|█         | 688/6470 [07:55<1:11:20,  1.35it/s]
3|689|Loss: 0.8467465043067932:  11%|█         | 688/6470 [07:55<1:11:20,  1.35it/s]
3|689|Loss: 0.8467465043067932:  11%|█         | 689/6470 [07:55<1:02:40,  1.54it/s]
3|690|Loss: 0.9357479810714722:  11%|█         | 689/6470 [07:55<1:02:40,  1.54it/s]
3|690|Loss: 0.9357479810714722:  11%|█         | 690/6470 [07:56<58:53,  1.64it/s]  
3|691|Loss: 0.9119830131530762:  11%|█         | 690/6470 [07:56<58:53,  1.64it/s]
3|691|Loss: 0.9119830131530762:  11%|█         | 691/6470 [07:56<1:05:09,  1.48it/s]
3|692|Loss: 0.8829043507575989:  11%|█         | 691/6470 [07:57<1:05:09,  1.48it/s]
3|692|Loss: 0.8829043507575989:  11%|█         | 692/6470 [07:57<1:02:48,  1.53it/s]
3|693|Loss: 0.8096897006034851:  11%|█         | 692/6470 [07:57<1:02:48,  1.53it/s]
3|693|Loss: 0.8096897006034851:  11%|█         | 693/6470 [07:58<1:05:27,  1.47it/s]
3|694|Loss: 0.9635112881660461:  11%|█         | 693/6470 [07:58<1:05:27,  1.47it/s]
3|694|Loss: 0.9635112881660461:  11%|█         | 694/6470 [07:58<1:05:26,  1.47it/s]
3|695|Loss: 1.0002261400222778:  11%|█         | 694/6470 [07:59<1:05:26,  1.47it/s]
3|695|Loss: 1.0002261400222778:  11%|█         | 695/6470 [07:59<1:05:42,  1.46it/s]
3|696|Loss: 0.8848634362220764:  11%|█         | 695/6470 [07:59<1:05:42,  1.46it/s]
3|696|Loss: 0.8848634362220764:  11%|█         | 696/6470 [08:00<1:03:23,  1.52it/s]
3|697|Loss: 0.9104260802268982:  11%|█         | 696/6470 [08:00<1:03:23,  1.52it/s]
3|697|Loss: 0.9104260802268982:  11%|█         | 697/6470 [08:01<1:08:16,  1.41it/s]
3|698|Loss: 0.9238818287849426:  11%|█         | 697/6470 [08:01<1:08:16,  1.41it/s]
3|698|Loss: 0.9238818287849426:  11%|█         | 698/6470 [08:01<1:09:26,  1.39it/s]
3|699|Loss: 0.8378522396087646:  11%|█         | 698/6470 [08:02<1:09:26,  1.39it/s]
3|699|Loss: 0.8378522396087646:  11%|█         | 699/6470 [08:02<1:06:55,  1.44it/s]
3|700|Loss: 0.8251718282699585:  11%|█         | 699/6470 [08:02<1:06:55,  1.44it/s]
3|700|Loss: 0.8251718282699585:  11%|█         | 700/6470 [08:03<1:05:03,  1.48it/s]
3|701|Loss: 0.9026289582252502:  11%|█         | 700/6470 [08:03<1:05:03,  1.48it/s]
3|701|Loss: 0.9026289582252502:  11%|█         | 701/6470 [08:03<1:04:00,  1.50it/s]
3|702|Loss: 1.0889865159988403:  11%|█         | 701/6470 [08:04<1:04:00,  1.50it/s]
3|702|Loss: 1.0889865159988403:  11%|█         | 702/6470 [08:04<1:07:11,  1.43it/s]
3|703|Loss: 0.8187708258628845:  11%|█         | 702/6470 [08:04<1:07:11,  1.43it/s]
3|703|Loss: 0.8187708258628845:  11%|█         | 703/6470 [08:05<1:08:00,  1.41it/s]
3|704|Loss: 0.9608035683631897:  11%|█         | 703/6470 [08:05<1:08:00,  1.41it/s]
3|704|Loss: 0.9608035683631897:  11%|█         | 704/6470 [08:05<1:08:30,  1.40it/s]
3|705|Loss: 0.910642147064209:  11%|█         | 704/6470 [08:06<1:08:30,  1.40it/s] 
3|705|Loss: 0.910642147064209:  11%|█         | 705/6470 [08:06<1:04:04,  1.50it/s]
3|706|Loss: 0.7964445352554321:  11%|█         | 705/6470 [08:06<1:04:04,  1.50it/s]
3|706|Loss: 0.7964445352554321:  11%|█         | 706/6470 [08:07<1:04:07,  1.50it/s]
3|707|Loss: 0.7542160749435425:  11%|█         | 706/6470 [08:07<1:04:07,  1.50it/s]
3|707|Loss: 0.7542160749435425:  11%|█         | 707/6470 [08:07<1:07:32,  1.42it/s]
3|708|Loss: 0.7933154106140137:  11%|█         | 707/6470 [08:08<1:07:32,  1.42it/s]
3|708|Loss: 0.7933154106140137:  11%|█         | 708/6470 [08:08<1:03:07,  1.52it/s]
3|709|Loss: 0.8845441341400146:  11%|█         | 708/6470 [08:08<1:03:07,  1.52it/s]
3|709|Loss: 0.8845441341400146:  11%|█         | 709/6470 [08:09<1:03:12,  1.52it/s]
3|710|Loss: 0.8254852294921875:  11%|█         | 709/6470 [08:09<1:03:12,  1.52it/s]
3|710|Loss: 0.8254852294921875:  11%|█         | 710/6470 [08:09<1:05:24,  1.47it/s]
3|711|Loss: 1.062986969947815:  11%|█         | 710/6470 [08:10<1:05:24,  1.47it/s] 
3|711|Loss: 1.062986969947815:  11%|█         | 711/6470 [08:10<1:04:51,  1.48it/s]
3|712|Loss: 0.9787973761558533:  11%|█         | 711/6470 [08:10<1:04:51,  1.48it/s]
3|712|Loss: 0.9787973761558533:  11%|█         | 712/6470 [08:11<1:05:18,  1.47it/s]
3|713|Loss: 0.942225456237793:  11%|█         | 712/6470 [08:11<1:05:18,  1.47it/s] 
3|713|Loss: 0.942225456237793:  11%|█         | 713/6470 [08:11<1:04:12,  1.49it/s]
3|714|Loss: 0.8736911416053772:  11%|█         | 713/6470 [08:12<1:04:12,  1.49it/s]
3|714|Loss: 0.8736911416053772:  11%|█         | 714/6470 [08:12<1:06:36,  1.44it/s]
3|715|Loss: 0.8026210069656372:  11%|█         | 714/6470 [08:13<1:06:36,  1.44it/s]
3|715|Loss: 0.8026210069656372:  11%|█         | 715/6470 [08:13<1:10:26,  1.36it/s]
3|716|Loss: 0.8370845913887024:  11%|█         | 715/6470 [08:13<1:10:26,  1.36it/s]
3|716|Loss: 0.8370845913887024:  11%|█         | 716/6470 [08:14<1:08:49,  1.39it/s]
3|717|Loss: 0.997954249382019:  11%|█         | 716/6470 [08:14<1:08:49,  1.39it/s] 
3|717|Loss: 0.997954249382019:  11%|█         | 717/6470 [08:14<1:04:36,  1.48it/s]
3|718|Loss: 0.9024472832679749:  11%|█         | 717/6470 [08:15<1:04:36,  1.48it/s]
3|718|Loss: 0.9024472832679749:  11%|█         | 718/6470 [08:15<1:05:06,  1.47it/s]
3|719|Loss: 0.7943102121353149:  11%|█         | 718/6470 [08:15<1:05:06,  1.47it/s]
3|719|Loss: 0.7943102121353149:  11%|█         | 719/6470 [08:15<1:01:50,  1.55it/s]
3|720|Loss: 0.954017698764801:  11%|█         | 719/6470 [08:16<1:01:50,  1.55it/s] 
3|720|Loss: 0.954017698764801:  11%|█         | 720/6470 [08:16<1:04:26,  1.49it/s]
3|721|Loss: 0.7252993583679199:  11%|█         | 720/6470 [08:17<1:04:26,  1.49it/s]
3|721|Loss: 0.7252993583679199:  11%|█         | 721/6470 [08:17<1:10:36,  1.36it/s]
3|722|Loss: 0.9153662323951721:  11%|█         | 721/6470 [08:18<1:10:36,  1.36it/s]
3|722|Loss: 0.9153662323951721:  11%|█         | 722/6470 [08:18<1:11:48,  1.33it/s]
3|723|Loss: 0.8668877482414246:  11%|█         | 722/6470 [08:18<1:11:48,  1.33it/s]
3|723|Loss: 0.8668877482414246:  11%|█         | 723/6470 [08:19<1:11:33,  1.34it/s]
3|724|Loss: 0.7251203656196594:  11%|█         | 723/6470 [08:19<1:11:33,  1.34it/s]
3|724|Loss: 0.7251203656196594:  11%|█         | 724/6470 [08:19<1:12:28,  1.32it/s]
3|725|Loss: 0.7932107448577881:  11%|█         | 724/6470 [08:20<1:12:28,  1.32it/s]
3|725|Loss: 0.7932107448577881:  11%|█         | 725/6470 [08:20<1:11:22,  1.34it/s]
3|726|Loss: 0.9153911471366882:  11%|█         | 725/6470 [08:21<1:11:22,  1.34it/s]
3|726|Loss: 0.9153911471366882:  11%|█         | 726/6470 [08:21<1:11:14,  1.34it/s]
3|727|Loss: 0.8163737654685974:  11%|█         | 726/6470 [08:21<1:11:14,  1.34it/s]
3|727|Loss: 0.8163737654685974:  11%|█         | 727/6470 [08:22<1:13:45,  1.30it/s]
3|728|Loss: 0.8154714703559875:  11%|█         | 727/6470 [08:22<1:13:45,  1.30it/s]
3|728|Loss: 0.8154714703559875:  11%|█▏        | 728/6470 [08:22<1:13:24,  1.30it/s]
3|729|Loss: 0.8479903340339661:  11%|█▏        | 728/6470 [08:23<1:13:24,  1.30it/s]
3|729|Loss: 0.8479903340339661:  11%|█▏        | 729/6470 [08:23<1:05:54,  1.45it/s]
3|730|Loss: 0.7582510709762573:  11%|█▏        | 729/6470 [08:23<1:05:54,  1.45it/s]
3|730|Loss: 0.7582510709762573:  11%|█▏        | 730/6470 [08:24<1:05:20,  1.46it/s]
3|731|Loss: 0.8101226687431335:  11%|█▏        | 730/6470 [08:24<1:05:20,  1.46it/s]
3|731|Loss: 0.8101226687431335:  11%|█▏        | 731/6470 [08:24<1:02:45,  1.52it/s]
3|732|Loss: 0.8658406734466553:  11%|█▏        | 731/6470 [08:25<1:02:45,  1.52it/s]
3|732|Loss: 0.8658406734466553:  11%|█▏        | 732/6470 [08:25<58:49,  1.63it/s]  
3|733|Loss: 0.9018353819847107:  11%|█▏        | 732/6470 [08:25<58:49,  1.63it/s]
3|733|Loss: 0.9018353819847107:  11%|█▏        | 733/6470 [08:25<1:01:18,  1.56it/s]
3|734|Loss: 0.8687542080879211:  11%|█▏        | 733/6470 [08:26<1:01:18,  1.56it/s]
3|734|Loss: 0.8687542080879211:  11%|█▏        | 734/6470 [08:26<1:04:43,  1.48it/s]
3|735|Loss: 0.981019139289856:  11%|█▏        | 734/6470 [08:27<1:04:43,  1.48it/s] 
3|735|Loss: 0.981019139289856:  11%|█▏        | 735/6470 [08:27<1:08:19,  1.40it/s]
3|736|Loss: 0.7363754510879517:  11%|█▏        | 735/6470 [08:27<1:08:19,  1.40it/s]
3|736|Loss: 0.7363754510879517:  11%|█▏        | 736/6470 [08:28<1:06:47,  1.43it/s]
3|737|Loss: 0.9411028623580933:  11%|█▏        | 736/6470 [08:28<1:06:47,  1.43it/s]
3|737|Loss: 0.9411028623580933:  11%|█▏        | 737/6470 [08:28<1:06:21,  1.44it/s]
3|738|Loss: 0.9334763288497925:  11%|█▏        | 737/6470 [08:29<1:06:21,  1.44it/s]
3|738|Loss: 0.9334763288497925:  11%|█▏        | 738/6470 [08:29<1:07:23,  1.42it/s]
3|739|Loss: 0.7823364734649658:  11%|█▏        | 738/6470 [08:30<1:07:23,  1.42it/s]
3|739|Loss: 0.7823364734649658:  11%|█▏        | 739/6470 [08:30<1:13:33,  1.30it/s]
3|740|Loss: 0.8444001078605652:  11%|█▏        | 739/6470 [08:30<1:13:33,  1.30it/s]
3|740|Loss: 0.8444001078605652:  11%|█▏        | 740/6470 [08:31<1:14:05,  1.29it/s]
3|741|Loss: 0.6638492345809937:  11%|█▏        | 740/6470 [08:31<1:14:05,  1.29it/s]
3|741|Loss: 0.6638492345809937:  11%|█▏        | 741/6470 [08:31<1:09:39,  1.37it/s]
3|742|Loss: 0.9218763113021851:  11%|█▏        | 741/6470 [08:32<1:09:39,  1.37it/s]
3|742|Loss: 0.9218763113021851:  11%|█▏        | 742/6470 [08:32<1:06:45,  1.43it/s]
3|743|Loss: 0.8684629797935486:  11%|█▏        | 742/6470 [08:32<1:06:45,  1.43it/s]
3|743|Loss: 0.8684629797935486:  11%|█▏        | 743/6470 [08:33<1:06:34,  1.43it/s]
3|744|Loss: 0.8776081204414368:  11%|█▏        | 743/6470 [08:33<1:06:34,  1.43it/s]
3|744|Loss: 0.8776081204414368:  11%|█▏        | 744/6470 [08:33<1:07:26,  1.42it/s]
3|745|Loss: 0.7725813388824463:  11%|█▏        | 744/6470 [08:34<1:07:26,  1.42it/s]
3|745|Loss: 0.7725813388824463:  12%|█▏        | 745/6470 [08:34<1:11:10,  1.34it/s]
3|746|Loss: 0.7966834902763367:  12%|█▏        | 745/6470 [08:35<1:11:10,  1.34it/s]
3|746|Loss: 0.7966834902763367:  12%|█▏        | 746/6470 [08:35<1:09:09,  1.38it/s]
3|747|Loss: 0.7299047708511353:  12%|█▏        | 746/6470 [08:35<1:09:09,  1.38it/s]
3|747|Loss: 0.7299047708511353:  12%|█▏        | 747/6470 [08:36<1:08:32,  1.39it/s]
3|748|Loss: 0.8261451721191406:  12%|█▏        | 747/6470 [08:36<1:08:32,  1.39it/s]
3|748|Loss: 0.8261451721191406:  12%|█▏        | 748/6470 [08:36<1:06:22,  1.44it/s]
3|749|Loss: 0.8596287965774536:  12%|█▏        | 748/6470 [08:37<1:06:22,  1.44it/s]
3|749|Loss: 0.8596287965774536:  12%|█▏        | 749/6470 [08:37<1:01:35,  1.55it/s]
3|750|Loss: 0.6736376285552979:  12%|█▏        | 749/6470 [08:37<1:01:35,  1.55it/s]
3|750|Loss: 0.6736376285552979:  12%|█▏        | 750/6470 [08:37<59:07,  1.61it/s]  
3|751|Loss: 0.9436854720115662:  12%|█▏        | 750/6470 [08:38<59:07,  1.61it/s]
3|751|Loss: 0.9436854720115662:  12%|█▏        | 751/6470 [08:38<58:46,  1.62it/s]
3|752|Loss: 0.9394764304161072:  12%|█▏        | 751/6470 [08:38<58:46,  1.62it/s]
3|752|Loss: 0.9394764304161072:  12%|█▏        | 752/6470 [08:39<1:04:37,  1.47it/s]
3|753|Loss: 0.9751895070075989:  12%|█▏        | 752/6470 [08:39<1:04:37,  1.47it/s]
3|753|Loss: 0.9751895070075989:  12%|█▏        | 753/6470 [08:40<1:03:57,  1.49it/s]
3|754|Loss: 0.7763641476631165:  12%|█▏        | 753/6470 [08:40<1:03:57,  1.49it/s]
3|754|Loss: 0.7763641476631165:  12%|█▏        | 754/6470 [08:40<1:02:37,  1.52it/s]
3|755|Loss: 0.8396170735359192:  12%|█▏        | 754/6470 [08:40<1:02:37,  1.52it/s]
3|755|Loss: 0.8396170735359192:  12%|█▏        | 755/6470 [08:41<1:00:07,  1.58it/s]
3|756|Loss: 0.8531622290611267:  12%|█▏        | 755/6470 [08:41<1:00:07,  1.58it/s]
3|756|Loss: 0.8531622290611267:  12%|█▏        | 756/6470 [08:41<59:24,  1.60it/s]  
3|757|Loss: 0.9031278491020203:  12%|█▏        | 756/6470 [08:42<59:24,  1.60it/s]
3|757|Loss: 0.9031278491020203:  12%|█▏        | 757/6470 [08:42<1:03:14,  1.51it/s]
3|758|Loss: 0.8178061842918396:  12%|█▏        | 757/6470 [08:42<1:03:14,  1.51it/s]
3|758|Loss: 0.8178061842918396:  12%|█▏        | 758/6470 [08:43<1:03:43,  1.49it/s]
3|759|Loss: 0.8216474056243896:  12%|█▏        | 758/6470 [08:43<1:03:43,  1.49it/s]
3|759|Loss: 0.8216474056243896:  12%|█▏        | 759/6470 [08:43<56:33,  1.68it/s]  
3|760|Loss: 0.8362322449684143:  12%|█▏        | 759/6470 [08:43<56:33,  1.68it/s]
3|760|Loss: 0.8362322449684143:  12%|█▏        | 760/6470 [08:44<55:43,  1.71it/s]
3|761|Loss: 0.7015348672866821:  12%|█▏        | 760/6470 [08:44<55:43,  1.71it/s]
3|761|Loss: 0.7015348672866821:  12%|█▏        | 761/6470 [08:44<57:18,  1.66it/s]
3|762|Loss: 0.7574688792228699:  12%|█▏        | 761/6470 [08:45<57:18,  1.66it/s]
3|762|Loss: 0.7574688792228699:  12%|█▏        | 762/6470 [08:45<1:02:32,  1.52it/s]
3|763|Loss: 0.8093841075897217:  12%|█▏        | 762/6470 [08:46<1:02:32,  1.52it/s]
3|763|Loss: 0.8093841075897217:  12%|█▏        | 763/6470 [08:46<1:02:30,  1.52it/s]
3|764|Loss: 0.8960601091384888:  12%|█▏        | 763/6470 [08:46<1:02:30,  1.52it/s]
3|764|Loss: 0.8960601091384888:  12%|█▏        | 764/6470 [08:46<1:01:53,  1.54it/s]
3|765|Loss: 0.8815998435020447:  12%|█▏        | 764/6470 [08:47<1:01:53,  1.54it/s]
3|765|Loss: 0.8815998435020447:  12%|█▏        | 765/6470 [08:47<1:05:06,  1.46it/s]
3|766|Loss: 0.8284195065498352:  12%|█▏        | 765/6470 [08:48<1:05:06,  1.46it/s]
3|766|Loss: 0.8284195065498352:  12%|█▏        | 766/6470 [08:48<1:06:33,  1.43it/s]
3|767|Loss: 0.855180025100708:  12%|█▏        | 766/6470 [08:48<1:06:33,  1.43it/s] 
3|767|Loss: 0.855180025100708:  12%|█▏        | 767/6470 [08:49<1:06:49,  1.42it/s]
3|768|Loss: 0.8081071376800537:  12%|█▏        | 767/6470 [08:49<1:06:49,  1.42it/s]
3|768|Loss: 0.8081071376800537:  12%|█▏        | 768/6470 [08:49<1:06:49,  1.42it/s]
3|769|Loss: 0.7234600782394409:  12%|█▏        | 768/6470 [08:50<1:06:49,  1.42it/s]
3|769|Loss: 0.7234600782394409:  12%|█▏        | 769/6470 [08:50<1:10:37,  1.35it/s]
3|770|Loss: 0.9058855772018433:  12%|█▏        | 769/6470 [08:51<1:10:37,  1.35it/s]
3|770|Loss: 0.9058855772018433:  12%|█▏        | 770/6470 [08:51<1:07:10,  1.41it/s]
3|771|Loss: 0.8286872506141663:  12%|█▏        | 770/6470 [08:51<1:07:10,  1.41it/s]
3|771|Loss: 0.8286872506141663:  12%|█▏        | 771/6470 [08:52<1:06:14,  1.43it/s]
3|772|Loss: 0.8371040225028992:  12%|█▏        | 771/6470 [08:52<1:06:14,  1.43it/s]
3|772|Loss: 0.8371040225028992:  12%|█▏        | 772/6470 [08:52<1:08:23,  1.39it/s]
3|773|Loss: 0.8884574174880981:  12%|█▏        | 772/6470 [08:53<1:08:23,  1.39it/s]
3|773|Loss: 0.8884574174880981:  12%|█▏        | 773/6470 [08:53<1:04:59,  1.46it/s]
3|774|Loss: 0.822832465171814:  12%|█▏        | 773/6470 [08:53<1:04:59,  1.46it/s] 
3|774|Loss: 0.822832465171814:  12%|█▏        | 774/6470 [08:54<1:04:39,  1.47it/s]
3|775|Loss: 0.7848058938980103:  12%|█▏        | 774/6470 [08:54<1:04:39,  1.47it/s]
3|775|Loss: 0.7848058938980103:  12%|█▏        | 775/6470 [08:54<1:05:12,  1.46it/s]
3|776|Loss: 0.8708520531654358:  12%|█▏        | 775/6470 [08:55<1:05:12,  1.46it/s]
3|776|Loss: 0.8708520531654358:  12%|█▏        | 776/6470 [08:55<1:08:20,  1.39it/s]
3|777|Loss: 0.8376595377922058:  12%|█▏        | 776/6470 [08:55<1:08:20,  1.39it/s]
3|777|Loss: 0.8376595377922058:  12%|█▏        | 777/6470 [08:56<1:05:14,  1.45it/s]
3|778|Loss: 0.8866097927093506:  12%|█▏        | 777/6470 [08:56<1:05:14,  1.45it/s]
3|778|Loss: 0.8866097927093506:  12%|█▏        | 778/6470 [08:56<1:05:53,  1.44it/s]
3|779|Loss: 0.7217180728912354:  12%|█▏        | 778/6470 [08:57<1:05:53,  1.44it/s]
3|779|Loss: 0.7217180728912354:  12%|█▏        | 779/6470 [08:57<1:09:16,  1.37it/s]
3|780|Loss: 0.8518783450126648:  12%|█▏        | 779/6470 [08:58<1:09:16,  1.37it/s]
3|780|Loss: 0.8518783450126648:  12%|█▏        | 780/6470 [08:58<1:05:43,  1.44it/s]
3|781|Loss: 0.7477031350135803:  12%|█▏        | 780/6470 [08:58<1:05:43,  1.44it/s]
3|781|Loss: 0.7477031350135803:  12%|█▏        | 781/6470 [08:59<1:08:36,  1.38it/s]
3|782|Loss: 0.9511250853538513:  12%|█▏        | 781/6470 [08:59<1:08:36,  1.38it/s]
3|782|Loss: 0.9511250853538513:  12%|█▏        | 782/6470 [08:59<1:06:19,  1.43it/s]
3|783|Loss: 0.745851993560791:  12%|█▏        | 782/6470 [09:00<1:06:19,  1.43it/s] 
3|783|Loss: 0.745851993560791:  12%|█▏        | 783/6470 [09:00<1:02:16,  1.52it/s]
3|784|Loss: 0.875484049320221:  12%|█▏        | 783/6470 [09:00<1:02:16,  1.52it/s]
3|784|Loss: 0.875484049320221:  12%|█▏        | 784/6470 [09:00<59:59,  1.58it/s]  
3|785|Loss: 0.8841559886932373:  12%|█▏        | 784/6470 [09:01<59:59,  1.58it/s]
3|785|Loss: 0.8841559886932373:  12%|█▏        | 785/6470 [09:01<59:25,  1.59it/s]
3|786|Loss: 0.9341539144515991:  12%|█▏        | 785/6470 [09:01<59:25,  1.59it/s]
3|786|Loss: 0.9341539144515991:  12%|█▏        | 786/6470 [09:02<59:08,  1.60it/s]
3|787|Loss: 0.833447277545929:  12%|█▏        | 786/6470 [09:02<59:08,  1.60it/s] 
3|787|Loss: 0.833447277545929:  12%|█▏        | 787/6470 [09:02<1:01:23,  1.54it/s]
3|788|Loss: 0.9204527735710144:  12%|█▏        | 787/6470 [09:03<1:01:23,  1.54it/s]
3|788|Loss: 0.9204527735710144:  12%|█▏        | 788/6470 [09:03<1:01:34,  1.54it/s]
3|789|Loss: 0.8998051881790161:  12%|█▏        | 788/6470 [09:03<1:01:34,  1.54it/s]
3|789|Loss: 0.8998051881790161:  12%|█▏        | 789/6470 [09:04<59:09,  1.60it/s]  
3|790|Loss: 0.830777108669281:  12%|█▏        | 789/6470 [09:04<59:09,  1.60it/s] 
3|790|Loss: 0.830777108669281:  12%|█▏        | 790/6470 [09:04<1:00:43,  1.56it/s]
3|791|Loss: 0.9500100016593933:  12%|█▏        | 790/6470 [09:05<1:00:43,  1.56it/s]
3|791|Loss: 0.9500100016593933:  12%|█▏        | 791/6470 [09:05<58:08,  1.63it/s]  
3|792|Loss: 0.7766405940055847:  12%|█▏        | 791/6470 [09:05<58:08,  1.63it/s]
3|792|Loss: 0.7766405940055847:  12%|█▏        | 792/6470 [09:05<1:00:27,  1.57it/s]
3|793|Loss: 0.9084407091140747:  12%|█▏        | 792/6470 [09:06<1:00:27,  1.57it/s]
3|793|Loss: 0.9084407091140747:  12%|█▏        | 793/6470 [09:06<58:08,  1.63it/s]  
3|794|Loss: 0.8705480098724365:  12%|█▏        | 793/6470 [09:06<58:08,  1.63it/s]
3|794|Loss: 0.8705480098724365:  12%|█▏        | 794/6470 [09:07<58:05,  1.63it/s]
3|795|Loss: 0.763372004032135:  12%|█▏        | 794/6470 [09:07<58:05,  1.63it/s] 
3|795|Loss: 0.763372004032135:  12%|█▏        | 795/6470 [09:07<58:25,  1.62it/s]
3|796|Loss: 0.8634313344955444:  12%|█▏        | 795/6470 [09:08<58:25,  1.62it/s]
3|796|Loss: 0.8634313344955444:  12%|█▏        | 796/6470 [09:08<1:01:08,  1.55it/s]
3|797|Loss: 0.7739351987838745:  12%|█▏        | 796/6470 [09:08<1:01:08,  1.55it/s]
3|797|Loss: 0.7739351987838745:  12%|█▏        | 797/6470 [09:09<1:01:33,  1.54it/s]
3|798|Loss: 0.7922194600105286:  12%|█▏        | 797/6470 [09:09<1:01:33,  1.54it/s]
3|798|Loss: 0.7922194600105286:  12%|█▏        | 798/6470 [09:09<59:06,  1.60it/s]  
3|799|Loss: 0.7393066883087158:  12%|█▏        | 798/6470 [09:10<59:06,  1.60it/s]
3|799|Loss: 0.7393066883087158:  12%|█▏        | 799/6470 [09:10<57:41,  1.64it/s]
3|800|Loss: 0.8435006141662598:  12%|█▏        | 799/6470 [09:10<57:41,  1.64it/s]
3|800|Loss: 0.8435006141662598:  12%|█▏        | 800/6470 [09:10<58:04,  1.63it/s]
3|801|Loss: 0.8789715766906738:  12%|█▏        | 800/6470 [09:11<58:04,  1.63it/s]
3|801|Loss: 0.8789715766906738:  12%|█▏        | 801/6470 [09:11<1:00:17,  1.57it/s]
3|802|Loss: 0.8757624626159668:  12%|█▏        | 801/6470 [09:11<1:00:17,  1.57it/s]
3|802|Loss: 0.8757624626159668:  12%|█▏        | 802/6470 [09:12<1:00:54,  1.55it/s]
3|803|Loss: 0.8952500224113464:  12%|█▏        | 802/6470 [09:12<1:00:54,  1.55it/s]
3|803|Loss: 0.8952500224113464:  12%|█▏        | 803/6470 [09:12<1:00:14,  1.57it/s]
3|804|Loss: 0.7651143074035645:  12%|█▏        | 803/6470 [09:13<1:00:14,  1.57it/s]
3|804|Loss: 0.7651143074035645:  12%|█▏        | 804/6470 [09:13<1:07:07,  1.41it/s]
3|805|Loss: 0.8888089656829834:  12%|█▏        | 804/6470 [09:14<1:07:07,  1.41it/s]
3|805|Loss: 0.8888089656829834:  12%|█▏        | 805/6470 [09:14<1:06:24,  1.42it/s]
3|806|Loss: 0.922039806842804:  12%|█▏        | 805/6470 [09:14<1:06:24,  1.42it/s] 
3|806|Loss: 0.922039806842804:  12%|█▏        | 806/6470 [09:14<58:45,  1.61it/s]  
3|807|Loss: 0.842601478099823:  12%|█▏        | 806/6470 [09:15<58:45,  1.61it/s]
3|807|Loss: 0.842601478099823:  12%|█▏        | 807/6470 [09:15<1:01:25,  1.54it/s]
3|808|Loss: 0.8391458988189697:  12%|█▏        | 807/6470 [09:15<1:01:25,  1.54it/s]
3|808|Loss: 0.8391458988189697:  12%|█▏        | 808/6470 [09:16<1:01:17,  1.54it/s]
3|809|Loss: 0.7646065950393677:  12%|█▏        | 808/6470 [09:16<1:01:17,  1.54it/s]
3|809|Loss: 0.7646065950393677:  13%|█▎        | 809/6470 [09:16<54:20,  1.74it/s]  
3|810|Loss: 0.7943284511566162:  13%|█▎        | 809/6470 [09:17<54:20,  1.74it/s]
3|810|Loss: 0.7943284511566162:  13%|█▎        | 810/6470 [09:17<59:11,  1.59it/s]
3|811|Loss: 0.9595094919204712:  13%|█▎        | 810/6470 [09:17<59:11,  1.59it/s]
3|811|Loss: 0.9595094919204712:  13%|█▎        | 811/6470 [09:18<1:03:21,  1.49it/s]
3|812|Loss: 0.7994379997253418:  13%|█▎        | 811/6470 [09:18<1:03:21,  1.49it/s]
3|812|Loss: 0.7994379997253418:  13%|█▎        | 812/6470 [09:18<1:04:42,  1.46it/s]
3|813|Loss: 0.7832348942756653:  13%|█▎        | 812/6470 [09:19<1:04:42,  1.46it/s]
3|813|Loss: 0.7832348942756653:  13%|█▎        | 813/6470 [09:19<1:05:46,  1.43it/s]
3|814|Loss: 0.7531721591949463:  13%|█▎        | 813/6470 [09:19<1:05:46,  1.43it/s]
3|814|Loss: 0.7531721591949463:  13%|█▎        | 814/6470 [09:20<1:00:21,  1.56it/s]
3|815|Loss: 0.7250788807868958:  13%|█▎        | 814/6470 [09:20<1:00:21,  1.56it/s]
3|815|Loss: 0.7250788807868958:  13%|█▎        | 815/6470 [09:20<1:01:21,  1.54it/s]
3|816|Loss: 0.833975076675415:  13%|█▎        | 815/6470 [09:21<1:01:21,  1.54it/s] 
3|816|Loss: 0.833975076675415:  13%|█▎        | 816/6470 [09:21<1:07:01,  1.41it/s]
3|817|Loss: 0.961083710193634:  13%|█▎        | 816/6470 [09:21<1:07:01,  1.41it/s]
3|817|Loss: 0.961083710193634:  13%|█▎        | 817/6470 [09:22<1:02:28,  1.51it/s]
3|818|Loss: 0.9831919074058533:  13%|█▎        | 817/6470 [09:22<1:02:28,  1.51it/s]
3|818|Loss: 0.9831919074058533:  13%|█▎        | 818/6470 [09:22<1:03:03,  1.49it/s]
3|819|Loss: 1.0028274059295654:  13%|█▎        | 818/6470 [09:23<1:03:03,  1.49it/s]
3|819|Loss: 1.0028274059295654:  13%|█▎        | 819/6470 [09:23<58:52,  1.60it/s]  
3|820|Loss: 0.8498356342315674:  13%|█▎        | 819/6470 [09:23<58:52,  1.60it/s]
3|820|Loss: 0.8498356342315674:  13%|█▎        | 820/6470 [09:24<1:00:22,  1.56it/s]
3|821|Loss: 0.7311246395111084:  13%|█▎        | 820/6470 [09:24<1:00:22,  1.56it/s]
3|821|Loss: 0.7311246395111084:  13%|█▎        | 821/6470 [09:24<1:00:43,  1.55it/s]
3|822|Loss: 0.8793941140174866:  13%|█▎        | 821/6470 [09:25<1:00:43,  1.55it/s]
3|822|Loss: 0.8793941140174866:  13%|█▎        | 822/6470 [09:25<1:00:19,  1.56it/s]
3|823|Loss: 0.8056551218032837:  13%|█▎        | 822/6470 [09:25<1:00:19,  1.56it/s]
3|823|Loss: 0.8056551218032837:  13%|█▎        | 823/6470 [09:26<1:05:29,  1.44it/s]
3|824|Loss: 0.9428754448890686:  13%|█▎        | 823/6470 [09:26<1:05:29,  1.44it/s]
3|824|Loss: 0.9428754448890686:  13%|█▎        | 824/6470 [09:26<1:04:54,  1.45it/s]
3|825|Loss: 0.8413335680961609:  13%|█▎        | 824/6470 [09:27<1:04:54,  1.45it/s]
3|825|Loss: 0.8413335680961609:  13%|█▎        | 825/6470 [09:27<1:04:30,  1.46it/s]
3|826|Loss: 0.8006697297096252:  13%|█▎        | 825/6470 [09:27<1:04:30,  1.46it/s]
3|826|Loss: 0.8006697297096252:  13%|█▎        | 826/6470 [09:28<1:03:34,  1.48it/s]
3|827|Loss: 0.7207469940185547:  13%|█▎        | 826/6470 [09:28<1:03:34,  1.48it/s]
3|827|Loss: 0.7207469940185547:  13%|█▎        | 827/6470 [09:28<1:04:33,  1.46it/s]
3|828|Loss: 0.8494929671287537:  13%|█▎        | 827/6470 [09:29<1:04:33,  1.46it/s]
3|828|Loss: 0.8494929671287537:  13%|█▎        | 828/6470 [09:29<1:02:19,  1.51it/s]
3|829|Loss: 0.7591204047203064:  13%|█▎        | 828/6470 [09:29<1:02:19,  1.51it/s]
3|829|Loss: 0.7591204047203064:  13%|█▎        | 829/6470 [09:30<1:03:16,  1.49it/s]
3|830|Loss: 0.7149060964584351:  13%|█▎        | 829/6470 [09:30<1:03:16,  1.49it/s]
3|830|Loss: 0.7149060964584351:  13%|█▎        | 830/6470 [09:30<1:06:21,  1.42it/s]
3|831|Loss: 0.752395510673523:  13%|█▎        | 830/6470 [09:31<1:06:21,  1.42it/s] 
3|831|Loss: 0.752395510673523:  13%|█▎        | 831/6470 [09:31<1:08:53,  1.36it/s]
3|832|Loss: 0.9380084276199341:  13%|█▎        | 831/6470 [09:32<1:08:53,  1.36it/s]
3|832|Loss: 0.9380084276199341:  13%|█▎        | 832/6470 [09:32<1:05:05,  1.44it/s]
3|833|Loss: 0.8515543937683105:  13%|█▎        | 832/6470 [09:32<1:05:05,  1.44it/s]
3|833|Loss: 0.8515543937683105:  13%|█▎        | 833/6470 [09:33<1:07:32,  1.39it/s]
3|834|Loss: 0.965690553188324:  13%|█▎        | 833/6470 [09:33<1:07:32,  1.39it/s] 
3|834|Loss: 0.965690553188324:  13%|█▎        | 834/6470 [09:33<1:05:02,  1.44it/s]
3|835|Loss: 0.9701809287071228:  13%|█▎        | 834/6470 [09:34<1:05:02,  1.44it/s]
3|835|Loss: 0.9701809287071228:  13%|█▎        | 835/6470 [09:34<1:05:46,  1.43it/s]
3|836|Loss: 0.6970372200012207:  13%|█▎        | 835/6470 [09:34<1:05:46,  1.43it/s]
3|836|Loss: 0.6970372200012207:  13%|█▎        | 836/6470 [09:35<1:03:24,  1.48it/s]
3|837|Loss: 0.9410319328308105:  13%|█▎        | 836/6470 [09:35<1:03:24,  1.48it/s]
3|837|Loss: 0.9410319328308105:  13%|█▎        | 837/6470 [09:35<1:04:35,  1.45it/s]
3|838|Loss: 0.7940893769264221:  13%|█▎        | 837/6470 [09:36<1:04:35,  1.45it/s]
3|838|Loss: 0.7940893769264221:  13%|█▎        | 838/6470 [09:36<1:06:32,  1.41it/s]
3|839|Loss: 0.838420569896698:  13%|█▎        | 838/6470 [09:37<1:06:32,  1.41it/s] 
3|839|Loss: 0.838420569896698:  13%|█▎        | 839/6470 [09:37<1:07:28,  1.39it/s]
3|840|Loss: 1.0319775342941284:  13%|█▎        | 839/6470 [09:37<1:07:28,  1.39it/s]
3|840|Loss: 1.0319775342941284:  13%|█▎        | 840/6470 [09:37<1:02:46,  1.49it/s]
3|841|Loss: 0.9099444150924683:  13%|█▎        | 840/6470 [09:38<1:02:46,  1.49it/s]
3|841|Loss: 0.9099444150924683:  13%|█▎        | 841/6470 [09:38<1:03:51,  1.47it/s]
3|842|Loss: 0.8407267928123474:  13%|█▎        | 841/6470 [09:39<1:03:51,  1.47it/s]
3|842|Loss: 0.8407267928123474:  13%|█▎        | 842/6470 [09:39<1:08:24,  1.37it/s]
3|843|Loss: 0.7975725531578064:  13%|█▎        | 842/6470 [09:39<1:08:24,  1.37it/s]
3|843|Loss: 0.7975725531578064:  13%|█▎        | 843/6470 [09:40<1:07:04,  1.40it/s]
3|844|Loss: 0.9667150974273682:  13%|█▎        | 843/6470 [09:40<1:07:04,  1.40it/s]
3|844|Loss: 0.9667150974273682:  13%|█▎        | 844/6470 [09:40<1:06:06,  1.42it/s]
3|845|Loss: 0.861116349697113:  13%|█▎        | 844/6470 [09:41<1:06:06,  1.42it/s] 
3|845|Loss: 0.861116349697113:  13%|█▎        | 845/6470 [09:41<1:02:38,  1.50it/s]
3|846|Loss: 0.9348101615905762:  13%|█▎        | 845/6470 [09:41<1:02:38,  1.50it/s]
3|846|Loss: 0.9348101615905762:  13%|█▎        | 846/6470 [09:41<58:42,  1.60it/s]  
3|847|Loss: 0.9249489307403564:  13%|█▎        | 846/6470 [09:42<58:42,  1.60it/s]
3|847|Loss: 0.9249489307403564:  13%|█▎        | 847/6470 [09:42<57:47,  1.62it/s]
3|848|Loss: 0.8699718713760376:  13%|█▎        | 847/6470 [09:42<57:47,  1.62it/s]
3|848|Loss: 0.8699718713760376:  13%|█▎        | 848/6470 [09:43<1:00:40,  1.54it/s]
3|849|Loss: 0.8096885085105896:  13%|█▎        | 848/6470 [09:43<1:00:40,  1.54it/s]
3|849|Loss: 0.8096885085105896:  13%|█▎        | 849/6470 [09:43<1:03:21,  1.48it/s]
3|850|Loss: 0.7818633317947388:  13%|█▎        | 849/6470 [09:44<1:03:21,  1.48it/s]
3|850|Loss: 0.7818633317947388:  13%|█▎        | 850/6470 [09:44<1:06:23,  1.41it/s]
3|851|Loss: 0.7373208999633789:  13%|█▎        | 850/6470 [09:45<1:06:23,  1.41it/s]
3|851|Loss: 0.7373208999633789:  13%|█▎        | 851/6470 [09:45<1:07:50,  1.38it/s]
3|852|Loss: 0.8970217108726501:  13%|█▎        | 851/6470 [09:45<1:07:50,  1.38it/s]
3|852|Loss: 0.8970217108726501:  13%|█▎        | 852/6470 [09:46<1:07:32,  1.39it/s]
3|853|Loss: 0.7422971725463867:  13%|█▎        | 852/6470 [09:46<1:07:32,  1.39it/s]
3|853|Loss: 0.7422971725463867:  13%|█▎        | 853/6470 [09:46<1:04:01,  1.46it/s]
3|854|Loss: 0.8504422307014465:  13%|█▎        | 853/6470 [09:47<1:04:01,  1.46it/s]
3|854|Loss: 0.8504422307014465:  13%|█▎        | 854/6470 [09:47<1:04:32,  1.45it/s]
3|855|Loss: 0.9108409881591797:  13%|█▎        | 854/6470 [09:47<1:04:32,  1.45it/s]
3|855|Loss: 0.9108409881591797:  13%|█▎        | 855/6470 [09:48<1:05:17,  1.43it/s]
3|856|Loss: 0.9568232893943787:  13%|█▎        | 855/6470 [09:48<1:05:17,  1.43it/s]
3|856|Loss: 0.9568232893943787:  13%|█▎        | 856/6470 [09:48<1:06:11,  1.41it/s]
3|857|Loss: 0.935505211353302:  13%|█▎        | 856/6470 [09:49<1:06:11,  1.41it/s] 
3|857|Loss: 0.935505211353302:  13%|█▎        | 857/6470 [09:49<1:04:36,  1.45it/s]
3|858|Loss: 0.8126912713050842:  13%|█▎        | 857/6470 [09:49<1:04:36,  1.45it/s]
3|858|Loss: 0.8126912713050842:  13%|█▎        | 858/6470 [09:50<1:01:50,  1.51it/s]
3|859|Loss: 0.8222883939743042:  13%|█▎        | 858/6470 [09:50<1:01:50,  1.51it/s]
3|859|Loss: 0.8222883939743042:  13%|█▎        | 859/6470 [09:51<1:06:02,  1.42it/s]
3|860|Loss: 0.8078537583351135:  13%|█▎        | 859/6470 [09:51<1:06:02,  1.42it/s]
3|860|Loss: 0.8078537583351135:  13%|█▎        | 860/6470 [09:51<1:04:06,  1.46it/s]
3|861|Loss: 0.8408461809158325:  13%|█▎        | 860/6470 [09:52<1:04:06,  1.46it/s]
3|861|Loss: 0.8408461809158325:  13%|█▎        | 861/6470 [09:52<1:05:50,  1.42it/s]
3|862|Loss: 0.9639785289764404:  13%|█▎        | 861/6470 [09:52<1:05:50,  1.42it/s]
3|862|Loss: 0.9639785289764404:  13%|█▎        | 862/6470 [09:53<1:04:35,  1.45it/s]
3|863|Loss: 0.6536580920219421:  13%|█▎        | 862/6470 [09:53<1:04:35,  1.45it/s]
3|863|Loss: 0.6536580920219421:  13%|█▎        | 863/6470 [09:53<1:02:22,  1.50it/s]
3|864|Loss: 0.8733024001121521:  13%|█▎        | 863/6470 [09:54<1:02:22,  1.50it/s]
3|864|Loss: 0.8733024001121521:  13%|█▎        | 864/6470 [09:54<1:03:02,  1.48it/s]
3|865|Loss: 0.8253796100616455:  13%|█▎        | 864/6470 [09:54<1:03:02,  1.48it/s]
3|865|Loss: 0.8253796100616455:  13%|█▎        | 865/6470 [09:55<1:01:58,  1.51it/s]
3|866|Loss: 0.8679901361465454:  13%|█▎        | 865/6470 [09:55<1:01:58,  1.51it/s]
3|866|Loss: 0.8679901361465454:  13%|█▎        | 866/6470 [09:55<1:06:24,  1.41it/s]
3|867|Loss: 0.787919819355011:  13%|█▎        | 866/6470 [09:56<1:06:24,  1.41it/s] 
3|867|Loss: 0.787919819355011:  13%|█▎        | 867/6470 [09:56<1:07:33,  1.38it/s]
3|868|Loss: 0.8383612036705017:  13%|█▎        | 867/6470 [09:56<1:07:33,  1.38it/s]
3|868|Loss: 0.8383612036705017:  13%|█▎        | 868/6470 [09:57<1:07:39,  1.38it/s]
3|869|Loss: 0.8177411556243896:  13%|█▎        | 868/6470 [09:57<1:07:39,  1.38it/s]
3|869|Loss: 0.8177411556243896:  13%|█▎        | 869/6470 [09:57<1:05:17,  1.43it/s]
3|870|Loss: 0.8230534195899963:  13%|█▎        | 869/6470 [09:58<1:05:17,  1.43it/s]
3|870|Loss: 0.8230534195899963:  13%|█▎        | 870/6470 [09:58<1:01:27,  1.52it/s]
3|871|Loss: 0.6783946752548218:  13%|█▎        | 870/6470 [09:58<1:01:27,  1.52it/s]
3|871|Loss: 0.6783946752548218:  13%|█▎        | 871/6470 [09:59<57:35,  1.62it/s]  
3|872|Loss: 0.8582126498222351:  13%|█▎        | 871/6470 [09:59<57:35,  1.62it/s]
3|872|Loss: 0.8582126498222351:  13%|█▎        | 872/6470 [09:59<59:19,  1.57it/s]
3|873|Loss: 1.0471676588058472:  13%|█▎        | 872/6470 [10:00<59:19,  1.57it/s]
3|873|Loss: 1.0471676588058472:  13%|█▎        | 873/6470 [10:00<1:04:36,  1.44it/s]
3|874|Loss: 0.8029953837394714:  13%|█▎        | 873/6470 [10:00<1:04:36,  1.44it/s]
3|874|Loss: 0.8029953837394714:  14%|█▎        | 874/6470 [10:01<1:04:35,  1.44it/s]
3|875|Loss: 0.8072132468223572:  14%|█▎        | 874/6470 [10:01<1:04:35,  1.44it/s]
3|875|Loss: 0.8072132468223572:  14%|█▎        | 875/6470 [10:01<1:03:08,  1.48it/s]
3|876|Loss: 0.8141997456550598:  14%|█▎        | 875/6470 [10:02<1:03:08,  1.48it/s]
3|876|Loss: 0.8141997456550598:  14%|█▎        | 876/6470 [10:02<59:35,  1.56it/s]  
3|877|Loss: 0.8676955103874207:  14%|█▎        | 876/6470 [10:02<59:35,  1.56it/s]
3|877|Loss: 0.8676955103874207:  14%|█▎        | 877/6470 [10:02<57:06,  1.63it/s]
3|878|Loss: 0.8041321039199829:  14%|█▎        | 877/6470 [10:03<57:06,  1.63it/s]
3|878|Loss: 0.8041321039199829:  14%|█▎        | 878/6470 [10:03<1:02:23,  1.49it/s]
3|879|Loss: 0.8980564475059509:  14%|█▎        | 878/6470 [10:04<1:02:23,  1.49it/s]
3|879|Loss: 0.8980564475059509:  14%|█▎        | 879/6470 [10:04<1:02:25,  1.49it/s]
3|880|Loss: 0.8929786086082458:  14%|█▎        | 879/6470 [10:04<1:02:25,  1.49it/s]
3|880|Loss: 0.8929786086082458:  14%|█▎        | 880/6470 [10:05<1:04:38,  1.44it/s]
3|881|Loss: 0.7629924416542053:  14%|█▎        | 880/6470 [10:05<1:04:38,  1.44it/s]
3|881|Loss: 0.7629924416542053:  14%|█▎        | 881/6470 [10:05<1:05:36,  1.42it/s]
3|882|Loss: 0.9338517189025879:  14%|█▎        | 881/6470 [10:06<1:05:36,  1.42it/s]
3|882|Loss: 0.9338517189025879:  14%|█▎        | 882/6470 [10:06<59:54,  1.55it/s]  
3|883|Loss: 0.9273344874382019:  14%|█▎        | 882/6470 [10:06<59:54,  1.55it/s]
3|883|Loss: 0.9273344874382019:  14%|█▎        | 883/6470 [10:07<1:02:12,  1.50it/s]
3|884|Loss: 0.9643517732620239:  14%|█▎        | 883/6470 [10:07<1:02:12,  1.50it/s]
3|884|Loss: 0.9643517732620239:  14%|█▎        | 884/6470 [10:07<1:01:25,  1.52it/s]
3|885|Loss: 0.8447067141532898:  14%|█▎        | 884/6470 [10:08<1:01:25,  1.52it/s]
3|885|Loss: 0.8447067141532898:  14%|█▎        | 885/6470 [10:08<1:02:39,  1.49it/s]
3|886|Loss: 0.7735751867294312:  14%|█▎        | 885/6470 [10:08<1:02:39,  1.49it/s]
3|886|Loss: 0.7735751867294312:  14%|█▎        | 886/6470 [10:09<1:01:40,  1.51it/s]
3|887|Loss: 0.8644248843193054:  14%|█▎        | 886/6470 [10:09<1:01:40,  1.51it/s]
3|887|Loss: 0.8644248843193054:  14%|█▎        | 887/6470 [10:09<1:00:56,  1.53it/s]
3|888|Loss: 0.8467966914176941:  14%|█▎        | 887/6470 [10:10<1:00:56,  1.53it/s]
3|888|Loss: 0.8467966914176941:  14%|█▎        | 888/6470 [10:10<1:03:05,  1.47it/s]
3|889|Loss: 0.8068410158157349:  14%|█▎        | 888/6470 [10:10<1:03:05,  1.47it/s]
3|889|Loss: 0.8068410158157349:  14%|█▎        | 889/6470 [10:11<59:16,  1.57it/s]  
3|890|Loss: 0.7815227508544922:  14%|█▎        | 889/6470 [10:11<59:16,  1.57it/s]
3|890|Loss: 0.7815227508544922:  14%|█▍        | 890/6470 [10:11<58:36,  1.59it/s]
3|891|Loss: 0.9535974860191345:  14%|█▍        | 890/6470 [10:12<58:36,  1.59it/s]
3|891|Loss: 0.9535974860191345:  14%|█▍        | 891/6470 [10:12<1:01:24,  1.51it/s]
3|892|Loss: 0.9328484535217285:  14%|█▍        | 891/6470 [10:12<1:01:24,  1.51it/s]
3|892|Loss: 0.9328484535217285:  14%|█▍        | 892/6470 [10:13<1:05:09,  1.43it/s]
3|893|Loss: 0.7590524554252625:  14%|█▍        | 892/6470 [10:13<1:05:09,  1.43it/s]
3|893|Loss: 0.7590524554252625:  14%|█▍        | 893/6470 [10:13<1:04:05,  1.45it/s]
3|894|Loss: 0.9251068234443665:  14%|█▍        | 893/6470 [10:14<1:04:05,  1.45it/s]
3|894|Loss: 0.9251068234443665:  14%|█▍        | 894/6470 [10:14<1:02:01,  1.50it/s]
3|895|Loss: 0.8449618816375732:  14%|█▍        | 894/6470 [10:14<1:02:01,  1.50it/s]
3|895|Loss: 0.8449618816375732:  14%|█▍        | 895/6470 [10:15<1:06:29,  1.40it/s]
3|896|Loss: 0.9745516180992126:  14%|█▍        | 895/6470 [10:15<1:06:29,  1.40it/s]
3|896|Loss: 0.9745516180992126:  14%|█▍        | 896/6470 [10:15<1:04:42,  1.44it/s]
3|897|Loss: 0.7464358806610107:  14%|█▍        | 896/6470 [10:16<1:04:42,  1.44it/s]
3|897|Loss: 0.7464358806610107:  14%|█▍        | 897/6470 [10:16<1:07:38,  1.37it/s]
3|898|Loss: 0.9070538282394409:  14%|█▍        | 897/6470 [10:17<1:07:38,  1.37it/s]
3|898|Loss: 0.9070538282394409:  14%|█▍        | 898/6470 [10:17<1:00:52,  1.53it/s]
3|899|Loss: 0.8233659267425537:  14%|█▍        | 898/6470 [10:17<1:00:52,  1.53it/s]
3|899|Loss: 0.8233659267425537:  14%|█▍        | 899/6470 [10:17<59:20,  1.56it/s]  
3|900|Loss: 0.9017605781555176:  14%|█▍        | 899/6470 [10:18<59:20,  1.56it/s]
3|900|Loss: 0.9017605781555176:  14%|█▍        | 900/6470 [10:18<1:00:41,  1.53it/s]
3|901|Loss: 0.8890225291252136:  14%|█▍        | 900/6470 [10:18<1:00:41,  1.53it/s]
3|901|Loss: 0.8890225291252136:  14%|█▍        | 901/6470 [10:19<1:05:09,  1.42it/s]
3|902|Loss: 0.8262516260147095:  14%|█▍        | 901/6470 [10:19<1:05:09,  1.42it/s]
3|902|Loss: 0.8262516260147095:  14%|█▍        | 902/6470 [10:19<1:03:13,  1.47it/s]
3|903|Loss: 0.779817521572113:  14%|█▍        | 902/6470 [10:20<1:03:13,  1.47it/s] 
3|903|Loss: 0.779817521572113:  14%|█▍        | 903/6470 [10:20<1:05:05,  1.43it/s]
3|904|Loss: 0.789665162563324:  14%|█▍        | 903/6470 [10:21<1:05:05,  1.43it/s]
3|904|Loss: 0.789665162563324:  14%|█▍        | 904/6470 [10:21<1:03:14,  1.47it/s]
3|905|Loss: 0.8086659908294678:  14%|█▍        | 904/6470 [10:21<1:03:14,  1.47it/s]
3|905|Loss: 0.8086659908294678:  14%|█▍        | 905/6470 [10:21<1:00:12,  1.54it/s]
3|906|Loss: 0.7849833965301514:  14%|█▍        | 905/6470 [10:22<1:00:12,  1.54it/s]
3|906|Loss: 0.7849833965301514:  14%|█▍        | 906/6470 [10:22<1:04:59,  1.43it/s]
3|907|Loss: 0.8503893613815308:  14%|█▍        | 906/6470 [10:23<1:04:59,  1.43it/s]
3|907|Loss: 0.8503893613815308:  14%|█▍        | 907/6470 [10:23<1:06:07,  1.40it/s]
3|908|Loss: 0.8413451313972473:  14%|█▍        | 907/6470 [10:23<1:06:07,  1.40it/s]
3|908|Loss: 0.8413451313972473:  14%|█▍        | 908/6470 [10:24<1:07:00,  1.38it/s]
3|909|Loss: 0.8625426888465881:  14%|█▍        | 908/6470 [10:24<1:07:00,  1.38it/s]
3|909|Loss: 0.8625426888465881:  14%|█▍        | 909/6470 [10:25<1:08:34,  1.35it/s]
3|910|Loss: 0.8903422951698303:  14%|█▍        | 909/6470 [10:25<1:08:34,  1.35it/s]
3|910|Loss: 0.8903422951698303:  14%|█▍        | 910/6470 [10:25<1:05:23,  1.42it/s]
3|911|Loss: 0.9368969202041626:  14%|█▍        | 910/6470 [10:26<1:05:23,  1.42it/s]
3|911|Loss: 0.9368969202041626:  14%|█▍        | 911/6470 [10:26<1:07:36,  1.37it/s]
3|912|Loss: 0.8890957832336426:  14%|█▍        | 911/6470 [10:26<1:07:36,  1.37it/s]
3|912|Loss: 0.8890957832336426:  14%|█▍        | 912/6470 [10:26<1:00:12,  1.54it/s]
3|913|Loss: 0.9440026879310608:  14%|█▍        | 912/6470 [10:27<1:00:12,  1.54it/s]
3|913|Loss: 0.9440026879310608:  14%|█▍        | 913/6470 [10:27<1:00:51,  1.52it/s]
3|914|Loss: 0.6851557493209839:  14%|█▍        | 913/6470 [10:27<1:00:51,  1.52it/s]
3|914|Loss: 0.6851557493209839:  14%|█▍        | 914/6470 [10:28<1:02:24,  1.48it/s]
3|915|Loss: 0.8869856595993042:  14%|█▍        | 914/6470 [10:28<1:02:24,  1.48it/s]
3|915|Loss: 0.8869856595993042:  14%|█▍        | 915/6470 [10:28<1:00:23,  1.53it/s]
3|916|Loss: 0.6139568090438843:  14%|█▍        | 915/6470 [10:29<1:00:23,  1.53it/s]
3|916|Loss: 0.6139568090438843:  14%|█▍        | 916/6470 [10:29<56:56,  1.63it/s]  
3|917|Loss: 0.9056418538093567:  14%|█▍        | 916/6470 [10:29<56:56,  1.63it/s]
3|917|Loss: 0.9056418538093567:  14%|█▍        | 917/6470 [10:30<56:47,  1.63it/s]
3|918|Loss: 0.8941778540611267:  14%|█▍        | 917/6470 [10:30<56:47,  1.63it/s]
3|918|Loss: 0.8941778540611267:  14%|█▍        | 918/6470 [10:30<57:42,  1.60it/s]
3|919|Loss: 0.9884077310562134:  14%|█▍        | 918/6470 [10:31<57:42,  1.60it/s]
3|919|Loss: 0.9884077310562134:  14%|█▍        | 919/6470 [10:31<57:50,  1.60it/s]
3|920|Loss: 1.0151535272598267:  14%|█▍        | 919/6470 [10:31<57:50,  1.60it/s]
3|920|Loss: 1.0151535272598267:  14%|█▍        | 920/6470 [10:32<1:01:42,  1.50it/s]
3|921|Loss: 0.6920645236968994:  14%|█▍        | 920/6470 [10:32<1:01:42,  1.50it/s]
3|921|Loss: 0.6920645236968994:  14%|█▍        | 921/6470 [10:32<1:06:34,  1.39it/s]
3|922|Loss: 0.8291198015213013:  14%|█▍        | 921/6470 [10:33<1:06:34,  1.39it/s]
3|922|Loss: 0.8291198015213013:  14%|█▍        | 922/6470 [10:33<1:05:09,  1.42it/s]
3|923|Loss: 0.8426903486251831:  14%|█▍        | 922/6470 [10:33<1:05:09,  1.42it/s]
3|923|Loss: 0.8426903486251831:  14%|█▍        | 923/6470 [10:34<1:00:39,  1.52it/s]
3|924|Loss: 0.8286096453666687:  14%|█▍        | 923/6470 [10:34<1:00:39,  1.52it/s]
3|924|Loss: 0.8286096453666687:  14%|█▍        | 924/6470 [10:34<1:01:02,  1.51it/s]
3|925|Loss: 0.7717470526695251:  14%|█▍        | 924/6470 [10:35<1:01:02,  1.51it/s]
3|925|Loss: 0.7717470526695251:  14%|█▍        | 925/6470 [10:35<56:45,  1.63it/s]  
3|926|Loss: 0.9463673233985901:  14%|█▍        | 925/6470 [10:35<56:45,  1.63it/s]
3|926|Loss: 0.9463673233985901:  14%|█▍        | 926/6470 [10:36<1:02:31,  1.48it/s]
3|927|Loss: 0.8796647787094116:  14%|█▍        | 926/6470 [10:36<1:02:31,  1.48it/s]
3|927|Loss: 0.8796647787094116:  14%|█▍        | 927/6470 [10:36<1:00:47,  1.52it/s]
3|928|Loss: 0.7825047969818115:  14%|█▍        | 927/6470 [10:37<1:00:47,  1.52it/s]
3|928|Loss: 0.7825047969818115:  14%|█▍        | 928/6470 [10:37<1:01:36,  1.50it/s]
3|929|Loss: 0.9193664789199829:  14%|█▍        | 928/6470 [10:37<1:01:36,  1.50it/s]
3|929|Loss: 0.9193664789199829:  14%|█▍        | 929/6470 [10:38<1:02:51,  1.47it/s]
3|930|Loss: 0.8354138731956482:  14%|█▍        | 929/6470 [10:38<1:02:51,  1.47it/s]
3|930|Loss: 0.8354138731956482:  14%|█▍        | 930/6470 [10:38<1:03:28,  1.45it/s]
3|931|Loss: 0.80989009141922:  14%|█▍        | 930/6470 [10:39<1:03:28,  1.45it/s]  
3|931|Loss: 0.80989009141922:  14%|█▍        | 931/6470 [10:39<1:01:24,  1.50it/s]
3|932|Loss: 0.9244685769081116:  14%|█▍        | 931/6470 [10:39<1:01:24,  1.50it/s]
3|932|Loss: 0.9244685769081116:  14%|█▍        | 932/6470 [10:40<1:04:00,  1.44it/s]
3|933|Loss: 0.8238339424133301:  14%|█▍        | 932/6470 [10:40<1:04:00,  1.44it/s]
3|933|Loss: 0.8238339424133301:  14%|█▍        | 933/6470 [10:40<1:04:47,  1.42it/s]
3|934|Loss: 0.8385763764381409:  14%|█▍        | 933/6470 [10:41<1:04:47,  1.42it/s]
3|934|Loss: 0.8385763764381409:  14%|█▍        | 934/6470 [10:41<1:05:36,  1.41it/s]
3|935|Loss: 0.8072051405906677:  14%|█▍        | 934/6470 [10:42<1:05:36,  1.41it/s]
3|935|Loss: 0.8072051405906677:  14%|█▍        | 935/6470 [10:42<1:04:16,  1.44it/s]
3|936|Loss: 0.8326416611671448:  14%|█▍        | 935/6470 [10:42<1:04:16,  1.44it/s]
3|936|Loss: 0.8326416611671448:  14%|█▍        | 936/6470 [10:42<1:01:16,  1.51it/s]
3|937|Loss: 0.8804276585578918:  14%|█▍        | 936/6470 [10:43<1:01:16,  1.51it/s]
3|937|Loss: 0.8804276585578918:  14%|█▍        | 937/6470 [10:43<58:22,  1.58it/s]  
3|938|Loss: 0.8762214183807373:  14%|█▍        | 937/6470 [10:43<58:22,  1.58it/s]
3|938|Loss: 0.8762214183807373:  14%|█▍        | 938/6470 [10:44<1:00:11,  1.53it/s]
3|939|Loss: 0.7282111048698425:  14%|█▍        | 938/6470 [10:44<1:00:11,  1.53it/s]
3|939|Loss: 0.7282111048698425:  15%|█▍        | 939/6470 [10:45<1:08:25,  1.35it/s]
3|940|Loss: 0.9903864860534668:  15%|█▍        | 939/6470 [10:45<1:08:25,  1.35it/s]
3|940|Loss: 0.9903864860534668:  15%|█▍        | 940/6470 [10:45<1:05:01,  1.42it/s]
3|941|Loss: 0.7338272929191589:  15%|█▍        | 940/6470 [10:46<1:05:01,  1.42it/s]
3|941|Loss: 0.7338272929191589:  15%|█▍        | 941/6470 [10:46<1:01:49,  1.49it/s]
3|942|Loss: 0.9034706950187683:  15%|█▍        | 941/6470 [10:46<1:01:49,  1.49it/s]
3|942|Loss: 0.9034706950187683:  15%|█▍        | 942/6470 [10:47<1:03:18,  1.46it/s]
3|943|Loss: 0.7301156520843506:  15%|█▍        | 942/6470 [10:47<1:03:18,  1.46it/s]
3|943|Loss: 0.7301156520843506:  15%|█▍        | 943/6470 [10:47<1:04:08,  1.44it/s]
3|944|Loss: 0.7581488490104675:  15%|█▍        | 943/6470 [10:48<1:04:08,  1.44it/s]
3|944|Loss: 0.7581488490104675:  15%|█▍        | 944/6470 [10:48<1:02:28,  1.47it/s]
3|945|Loss: 0.8111400008201599:  15%|█▍        | 944/6470 [10:48<1:02:28,  1.47it/s]
3|945|Loss: 0.8111400008201599:  15%|█▍        | 945/6470 [10:48<58:42,  1.57it/s]  
3|946|Loss: 0.8162148594856262:  15%|█▍        | 945/6470 [10:49<58:42,  1.57it/s]
3|946|Loss: 0.8162148594856262:  15%|█▍        | 946/6470 [10:49<1:00:14,  1.53it/s]
3|947|Loss: 0.9188547730445862:  15%|█▍        | 946/6470 [10:50<1:00:14,  1.53it/s]
3|947|Loss: 0.9188547730445862:  15%|█▍        | 947/6470 [10:50<1:00:06,  1.53it/s]
3|948|Loss: 0.969514787197113:  15%|█▍        | 947/6470 [10:50<1:00:06,  1.53it/s] 
3|948|Loss: 0.969514787197113:  15%|█▍        | 948/6470 [10:51<1:04:09,  1.43it/s]
3|949|Loss: 0.8726800680160522:  15%|█▍        | 948/6470 [10:51<1:04:09,  1.43it/s]
3|949|Loss: 0.8726800680160522:  15%|█▍        | 949/6470 [10:51<1:03:57,  1.44it/s]
3|950|Loss: 0.753183126449585:  15%|█▍        | 949/6470 [10:52<1:03:57,  1.44it/s] 
3|950|Loss: 0.753183126449585:  15%|█▍        | 950/6470 [10:52<1:01:56,  1.49it/s]
3|951|Loss: 0.7553460001945496:  15%|█▍        | 950/6470 [10:52<1:01:56,  1.49it/s]
3|951|Loss: 0.7553460001945496:  15%|█▍        | 951/6470 [10:53<1:00:32,  1.52it/s]
3|952|Loss: 0.8057736754417419:  15%|█▍        | 951/6470 [10:53<1:00:32,  1.52it/s]
3|952|Loss: 0.8057736754417419:  15%|█▍        | 952/6470 [10:53<59:57,  1.53it/s]  
3|953|Loss: 0.8723161220550537:  15%|█▍        | 952/6470 [10:54<59:57,  1.53it/s]
3|953|Loss: 0.8723161220550537:  15%|█▍        | 953/6470 [10:54<1:03:09,  1.46it/s]
3|954|Loss: 1.0099256038665771:  15%|█▍        | 953/6470 [10:54<1:03:09,  1.46it/s]
3|954|Loss: 1.0099256038665771:  15%|█▍        | 954/6470 [10:55<1:05:06,  1.41it/s]
3|955|Loss: 0.8402812480926514:  15%|█▍        | 954/6470 [10:55<1:05:06,  1.41it/s]
3|955|Loss: 0.8402812480926514:  15%|█▍        | 955/6470 [10:55<1:04:02,  1.44it/s]
3|956|Loss: 0.7324817776679993:  15%|█▍        | 955/6470 [10:56<1:04:02,  1.44it/s]
3|956|Loss: 0.7324817776679993:  15%|█▍        | 956/6470 [10:56<1:02:35,  1.47it/s]
3|957|Loss: 0.866710364818573:  15%|█▍        | 956/6470 [10:56<1:02:35,  1.47it/s] 
3|957|Loss: 0.866710364818573:  15%|█▍        | 957/6470 [10:57<1:01:32,  1.49it/s]
3|958|Loss: 0.8479994535446167:  15%|█▍        | 957/6470 [10:57<1:01:32,  1.49it/s]
3|958|Loss: 0.8479994535446167:  15%|█▍        | 958/6470 [10:57<1:03:58,  1.44it/s]
3|959|Loss: 0.8924762606620789:  15%|█▍        | 958/6470 [10:58<1:03:58,  1.44it/s]
3|959|Loss: 0.8924762606620789:  15%|█▍        | 959/6470 [10:58<1:01:18,  1.50it/s]
3|960|Loss: 0.7925556898117065:  15%|█▍        | 959/6470 [10:58<1:01:18,  1.50it/s]
3|960|Loss: 0.7925556898117065:  15%|█▍        | 960/6470 [10:59<1:04:13,  1.43it/s]
3|961|Loss: 0.8386886119842529:  15%|█▍        | 960/6470 [10:59<1:04:13,  1.43it/s]
3|961|Loss: 0.8386886119842529:  15%|█▍        | 961/6470 [11:00<1:05:09,  1.41it/s]
3|962|Loss: 0.9459363222122192:  15%|█▍        | 961/6470 [11:00<1:05:09,  1.41it/s]
3|962|Loss: 0.9459363222122192:  15%|█▍        | 962/6470 [11:00<1:03:34,  1.44it/s]
3|963|Loss: 0.8199604749679565:  15%|█▍        | 962/6470 [11:00<1:03:34,  1.44it/s]
3|963|Loss: 0.8199604749679565:  15%|█▍        | 963/6470 [11:01<58:05,  1.58it/s]  
3|964|Loss: 0.9886467456817627:  15%|█▍        | 963/6470 [11:01<58:05,  1.58it/s]
3|964|Loss: 0.9886467456817627:  15%|█▍        | 964/6470 [11:01<57:00,  1.61it/s]
3|965|Loss: 0.9145822525024414:  15%|█▍        | 964/6470 [11:02<57:00,  1.61it/s]
3|965|Loss: 0.9145822525024414:  15%|█▍        | 965/6470 [11:02<58:55,  1.56it/s]
3|966|Loss: 0.8350861668586731:  15%|█▍        | 965/6470 [11:02<58:55,  1.56it/s]
3|966|Loss: 0.8350861668586731:  15%|█▍        | 966/6470 [11:03<56:17,  1.63it/s]
3|967|Loss: 0.8233741521835327:  15%|█▍        | 966/6470 [11:03<56:17,  1.63it/s]
3|967|Loss: 0.8233741521835327:  15%|█▍        | 967/6470 [11:03<56:11,  1.63it/s]
3|968|Loss: 0.6473571062088013:  15%|█▍        | 967/6470 [11:03<56:11,  1.63it/s]
3|968|Loss: 0.6473571062088013:  15%|█▍        | 968/6470 [11:04<58:16,  1.57it/s]
3|969|Loss: 0.9026529788970947:  15%|█▍        | 968/6470 [11:04<58:16,  1.57it/s]
3|969|Loss: 0.9026529788970947:  15%|█▍        | 969/6470 [11:04<57:05,  1.61it/s]
3|970|Loss: 0.7608523368835449:  15%|█▍        | 969/6470 [11:05<57:05,  1.61it/s]
3|970|Loss: 0.7608523368835449:  15%|█▍        | 970/6470 [11:05<56:41,  1.62it/s]
3|971|Loss: 0.9016751646995544:  15%|█▍        | 970/6470 [11:05<56:41,  1.62it/s]
3|971|Loss: 0.9016751646995544:  15%|█▌        | 971/6470 [11:06<1:01:50,  1.48it/s]
3|972|Loss: 0.9293880462646484:  15%|█▌        | 971/6470 [11:06<1:01:50,  1.48it/s]
3|972|Loss: 0.9293880462646484:  15%|█▌        | 972/6470 [11:07<1:04:17,  1.43it/s]
3|973|Loss: 0.8934758901596069:  15%|█▌        | 972/6470 [11:07<1:04:17,  1.43it/s]
3|973|Loss: 0.8934758901596069:  15%|█▌        | 973/6470 [11:07<57:02,  1.61it/s]  
3|974|Loss: 0.8317466378211975:  15%|█▌        | 973/6470 [11:07<57:02,  1.61it/s]
3|974|Loss: 0.8317466378211975:  15%|█▌        | 974/6470 [11:08<55:48,  1.64it/s]
3|975|Loss: 0.8168748021125793:  15%|█▌        | 974/6470 [11:08<55:48,  1.64it/s]
3|975|Loss: 0.8168748021125793:  15%|█▌        | 975/6470 [11:08<1:00:02,  1.53it/s]
3|976|Loss: 0.8396410346031189:  15%|█▌        | 975/6470 [11:09<1:00:02,  1.53it/s]
3|976|Loss: 0.8396410346031189:  15%|█▌        | 976/6470 [11:09<1:02:49,  1.46it/s]
3|977|Loss: 0.8832002282142639:  15%|█▌        | 976/6470 [11:10<1:02:49,  1.46it/s]
3|977|Loss: 0.8832002282142639:  15%|█▌        | 977/6470 [11:10<1:07:36,  1.35it/s]
3|978|Loss: 0.8434663414955139:  15%|█▌        | 977/6470 [11:10<1:07:36,  1.35it/s]
3|978|Loss: 0.8434663414955139:  15%|█▌        | 978/6470 [11:11<1:05:22,  1.40it/s]
3|979|Loss: 0.7210291028022766:  15%|█▌        | 978/6470 [11:11<1:05:22,  1.40it/s]
3|979|Loss: 0.7210291028022766:  15%|█▌        | 979/6470 [11:11<1:01:35,  1.49it/s]
3|980|Loss: 0.7082808613777161:  15%|█▌        | 979/6470 [11:12<1:01:35,  1.49it/s]
3|980|Loss: 0.7082808613777161:  15%|█▌        | 980/6470 [11:12<1:04:55,  1.41it/s]
3|981|Loss: 0.8882905840873718:  15%|█▌        | 980/6470 [11:12<1:04:55,  1.41it/s]
3|981|Loss: 0.8882905840873718:  15%|█▌        | 981/6470 [11:13<1:06:09,  1.38it/s]
3|982|Loss: 0.854423463344574:  15%|█▌        | 981/6470 [11:13<1:06:09,  1.38it/s] 
3|982|Loss: 0.854423463344574:  15%|█▌        | 982/6470 [11:13<1:06:25,  1.38it/s]
3|983|Loss: 0.9119110703468323:  15%|█▌        | 982/6470 [11:14<1:06:25,  1.38it/s]
3|983|Loss: 0.9119110703468323:  15%|█▌        | 983/6470 [11:14<1:06:08,  1.38it/s]
3|984|Loss: 0.8265472054481506:  15%|█▌        | 983/6470 [11:15<1:06:08,  1.38it/s]
3|984|Loss: 0.8265472054481506:  15%|█▌        | 984/6470 [11:15<1:03:51,  1.43it/s]
3|985|Loss: 0.9124355912208557:  15%|█▌        | 984/6470 [11:15<1:03:51,  1.43it/s]
3|985|Loss: 0.9124355912208557:  15%|█▌        | 985/6470 [11:16<1:06:52,  1.37it/s]
3|986|Loss: 0.8738130927085876:  15%|█▌        | 985/6470 [11:16<1:06:52,  1.37it/s]
3|986|Loss: 0.8738130927085876:  15%|█▌        | 986/6470 [11:17<1:11:39,  1.28it/s]
3|987|Loss: 0.8219345211982727:  15%|█▌        | 986/6470 [11:17<1:11:39,  1.28it/s]
3|987|Loss: 0.8219345211982727:  15%|█▌        | 987/6470 [11:17<1:11:14,  1.28it/s]
3|988|Loss: 0.9538512229919434:  15%|█▌        | 987/6470 [11:18<1:11:14,  1.28it/s]
3|988|Loss: 0.9538512229919434:  15%|█▌        | 988/6470 [11:18<1:09:25,  1.32it/s]
3|989|Loss: 0.8626707196235657:  15%|█▌        | 988/6470 [11:18<1:09:25,  1.32it/s]
3|989|Loss: 0.8626707196235657:  15%|█▌        | 989/6470 [11:19<1:05:59,  1.38it/s]
3|990|Loss: 0.9206476807594299:  15%|█▌        | 989/6470 [11:19<1:05:59,  1.38it/s]
3|990|Loss: 0.9206476807594299:  15%|█▌        | 990/6470 [11:19<1:03:34,  1.44it/s]
3|991|Loss: 0.832267701625824:  15%|█▌        | 990/6470 [11:20<1:03:34,  1.44it/s] 
3|991|Loss: 0.832267701625824:  15%|█▌        | 991/6470 [11:20<1:06:57,  1.36it/s]
3|992|Loss: 0.9584425687789917:  15%|█▌        | 991/6470 [11:21<1:06:57,  1.36it/s]
3|992|Loss: 0.9584425687789917:  15%|█▌        | 992/6470 [11:21<1:05:13,  1.40it/s]
3|993|Loss: 0.8182992339134216:  15%|█▌        | 992/6470 [11:21<1:05:13,  1.40it/s]
3|993|Loss: 0.8182992339134216:  15%|█▌        | 993/6470 [11:21<1:01:31,  1.48it/s]
3|994|Loss: 0.9448118209838867:  15%|█▌        | 993/6470 [11:22<1:01:31,  1.48it/s]
3|994|Loss: 0.9448118209838867:  15%|█▌        | 994/6470 [11:22<1:02:38,  1.46it/s]
3|995|Loss: 0.8951222896575928:  15%|█▌        | 994/6470 [11:23<1:02:38,  1.46it/s]
3|995|Loss: 0.8951222896575928:  15%|█▌        | 995/6470 [11:23<1:03:31,  1.44it/s]
3|996|Loss: 0.7089405059814453:  15%|█▌        | 995/6470 [11:23<1:03:31,  1.44it/s]
3|996|Loss: 0.7089405059814453:  15%|█▌        | 996/6470 [11:23<1:02:24,  1.46it/s]
3|997|Loss: 0.865013062953949:  15%|█▌        | 996/6470 [11:24<1:02:24,  1.46it/s] 
3|997|Loss: 0.865013062953949:  15%|█▌        | 997/6470 [11:24<1:02:29,  1.46it/s]
3|998|Loss: 0.9035319089889526:  15%|█▌        | 997/6470 [11:25<1:02:29,  1.46it/s]
3|998|Loss: 0.9035319089889526:  15%|█▌        | 998/6470 [11:25<1:01:11,  1.49it/s]
3|999|Loss: 0.7810530066490173:  15%|█▌        | 998/6470 [11:25<1:01:11,  1.49it/s]
3|999|Loss: 0.7810530066490173:  15%|█▌        | 999/6470 [11:26<1:02:23,  1.46it/s]
3|1000|Loss: 0.8580191135406494:  15%|█▌        | 999/6470 [11:26<1:02:23,  1.46it/s]
3|1000|Loss: 0.8580191135406494:  15%|█▌        | 1000/6470 [11:26<1:03:03,  1.45it/s]
3|1001|Loss: 0.80379718542099:  15%|█▌        | 1000/6470 [11:27<1:03:03,  1.45it/s]  
3|1001|Loss: 0.80379718542099:  15%|█▌        | 1001/6470 [11:27<1:02:01,  1.47it/s]
3|1002|Loss: 0.7736545205116272:  15%|█▌        | 1001/6470 [11:27<1:02:01,  1.47it/s]
3|1002|Loss: 0.7736545205116272:  15%|█▌        | 1002/6470 [11:27<53:48,  1.69it/s]  
3|1003|Loss: 0.8516629338264465:  15%|█▌        | 1002/6470 [11:28<53:48,  1.69it/s]
3|1003|Loss: 0.8516629338264465:  16%|█▌        | 1003/6470 [11:28<54:24,  1.67it/s]
3|1004|Loss: 0.8437478542327881:  16%|█▌        | 1003/6470 [11:28<54:24,  1.67it/s]
3|1004|Loss: 0.8437478542327881:  16%|█▌        | 1004/6470 [11:29<1:00:39,  1.50it/s]
3|1005|Loss: 0.8893647193908691:  16%|█▌        | 1004/6470 [11:29<1:00:39,  1.50it/s]
3|1005|Loss: 0.8893647193908691:  16%|█▌        | 1005/6470 [11:30<1:05:57,  1.38it/s]
3|1006|Loss: 0.7746319770812988:  16%|█▌        | 1005/6470 [11:30<1:05:57,  1.38it/s]
3|1006|Loss: 0.7746319770812988:  16%|█▌        | 1006/6470 [11:30<1:09:56,  1.30it/s]
3|1007|Loss: 0.8385075330734253:  16%|█▌        | 1006/6470 [11:31<1:09:56,  1.30it/s]
3|1007|Loss: 0.8385075330734253:  16%|█▌        | 1007/6470 [11:31<1:09:11,  1.32it/s]
3|1008|Loss: 0.8254878520965576:  16%|█▌        | 1007/6470 [11:32<1:09:11,  1.32it/s]
3|1008|Loss: 0.8254878520965576:  16%|█▌        | 1008/6470 [11:32<1:08:34,  1.33it/s]
3|1009|Loss: 0.9357433319091797:  16%|█▌        | 1008/6470 [11:32<1:08:34,  1.33it/s]
3|1009|Loss: 0.9357433319091797:  16%|█▌        | 1009/6470 [11:33<1:07:45,  1.34it/s]
3|1010|Loss: 0.8565617203712463:  16%|█▌        | 1009/6470 [11:33<1:07:45,  1.34it/s]
3|1010|Loss: 0.8565617203712463:  16%|█▌        | 1010/6470 [11:33<1:07:11,  1.35it/s]
3|1011|Loss: 0.9251630306243896:  16%|█▌        | 1010/6470 [11:34<1:07:11,  1.35it/s]
3|1011|Loss: 0.9251630306243896:  16%|█▌        | 1011/6470 [11:34<1:06:55,  1.36it/s]
3|1012|Loss: 0.8099601864814758:  16%|█▌        | 1011/6470 [11:34<1:06:55,  1.36it/s]
3|1012|Loss: 0.8099601864814758:  16%|█▌        | 1012/6470 [11:35<1:04:55,  1.40it/s]
3|1013|Loss: 1.0197360515594482:  16%|█▌        | 1012/6470 [11:35<1:04:55,  1.40it/s]
3|1013|Loss: 1.0197360515594482:  16%|█▌        | 1013/6470 [11:35<1:02:24,  1.46it/s]
3|1014|Loss: 0.794605016708374:  16%|█▌        | 1013/6470 [11:36<1:02:24,  1.46it/s] 
3|1014|Loss: 0.794605016708374:  16%|█▌        | 1014/6470 [11:36<57:04,  1.59it/s]  
3|1015|Loss: 0.9556783437728882:  16%|█▌        | 1014/6470 [11:36<57:04,  1.59it/s]
3|1015|Loss: 0.9556783437728882:  16%|█▌        | 1015/6470 [11:37<58:09,  1.56it/s]
3|1016|Loss: 0.9749690890312195:  16%|█▌        | 1015/6470 [11:37<58:09,  1.56it/s]
3|1016|Loss: 0.9749690890312195:  16%|█▌        | 1016/6470 [11:37<1:01:09,  1.49it/s]
3|1017|Loss: 0.7592642903327942:  16%|█▌        | 1016/6470 [11:38<1:01:09,  1.49it/s]
3|1017|Loss: 0.7592642903327942:  16%|█▌        | 1017/6470 [11:38<1:00:58,  1.49it/s]
3|1018|Loss: 0.8680427074432373:  16%|█▌        | 1017/6470 [11:38<1:00:58,  1.49it/s]
3|1018|Loss: 0.8680427074432373:  16%|█▌        | 1018/6470 [11:38<56:05,  1.62it/s]  
3|1019|Loss: 0.7225472331047058:  16%|█▌        | 1018/6470 [11:39<56:05,  1.62it/s]
3|1019|Loss: 0.7225472331047058:  16%|█▌        | 1019/6470 [11:39<55:30,  1.64it/s]
3|1020|Loss: 0.8996614813804626:  16%|█▌        | 1019/6470 [11:39<55:30,  1.64it/s]
3|1020|Loss: 0.8996614813804626:  16%|█▌        | 1020/6470 [11:40<58:20,  1.56it/s]
3|1021|Loss: 0.7873632907867432:  16%|█▌        | 1020/6470 [11:40<58:20,  1.56it/s]
3|1021|Loss: 0.7873632907867432:  16%|█▌        | 1021/6470 [11:41<1:02:57,  1.44it/s]
3|1022|Loss: 0.9409821629524231:  16%|█▌        | 1021/6470 [11:41<1:02:57,  1.44it/s]
3|1022|Loss: 0.9409821629524231:  16%|█▌        | 1022/6470 [11:41<1:03:58,  1.42it/s]
3|1023|Loss: 0.9057222604751587:  16%|█▌        | 1022/6470 [11:42<1:03:58,  1.42it/s]
3|1023|Loss: 0.9057222604751587:  16%|█▌        | 1023/6470 [11:42<1:06:07,  1.37it/s]
3|1024|Loss: 0.7700603604316711:  16%|█▌        | 1023/6470 [11:42<1:06:07,  1.37it/s]
3|1024|Loss: 0.7700603604316711:  16%|█▌        | 1024/6470 [11:43<1:00:34,  1.50it/s]
3|1025|Loss: 0.7772649526596069:  16%|█▌        | 1024/6470 [11:43<1:00:34,  1.50it/s]
3|1025|Loss: 0.7772649526596069:  16%|█▌        | 1025/6470 [11:43<58:42,  1.55it/s]  
3|1026|Loss: 0.7090628743171692:  16%|█▌        | 1025/6470 [11:44<58:42,  1.55it/s]
3|1026|Loss: 0.7090628743171692:  16%|█▌        | 1026/6470 [11:44<57:04,  1.59it/s]
3|1027|Loss: 0.6809553503990173:  16%|█▌        | 1026/6470 [11:44<57:04,  1.59it/s]
3|1027|Loss: 0.6809553503990173:  16%|█▌        | 1027/6470 [11:45<1:01:49,  1.47it/s]
3|1028|Loss: 0.90438312292099:  16%|█▌        | 1027/6470 [11:45<1:01:49,  1.47it/s]  
3|1028|Loss: 0.90438312292099:  16%|█▌        | 1028/6470 [11:45<1:04:13,  1.41it/s]
3|1029|Loss: 0.8703252673149109:  16%|█▌        | 1028/6470 [11:46<1:04:13,  1.41it/s]
3|1029|Loss: 0.8703252673149109:  16%|█▌        | 1029/6470 [11:46<1:06:17,  1.37it/s]
3|1030|Loss: 0.7359382510185242:  16%|█▌        | 1029/6470 [11:47<1:06:17,  1.37it/s]
3|1030|Loss: 0.7359382510185242:  16%|█▌        | 1030/6470 [11:47<1:06:52,  1.36it/s]
3|1031|Loss: 0.8971384167671204:  16%|█▌        | 1030/6470 [11:47<1:06:52,  1.36it/s]
3|1031|Loss: 0.8971384167671204:  16%|█▌        | 1031/6470 [11:47<1:01:17,  1.48it/s]
3|1032|Loss: 0.8010683059692383:  16%|█▌        | 1031/6470 [11:48<1:01:17,  1.48it/s]
3|1032|Loss: 0.8010683059692383:  16%|█▌        | 1032/6470 [11:48<1:01:11,  1.48it/s]
3|1033|Loss: 0.9607900977134705:  16%|█▌        | 1032/6470 [11:49<1:01:11,  1.48it/s]
3|1033|Loss: 0.9607900977134705:  16%|█▌        | 1033/6470 [11:49<1:02:14,  1.46it/s]
3|1034|Loss: 0.8757085204124451:  16%|█▌        | 1033/6470 [11:49<1:02:14,  1.46it/s]
3|1034|Loss: 0.8757085204124451:  16%|█▌        | 1034/6470 [11:49<1:01:25,  1.47it/s]
3|1035|Loss: 0.8006487488746643:  16%|█▌        | 1034/6470 [11:50<1:01:25,  1.47it/s]
3|1035|Loss: 0.8006487488746643:  16%|█▌        | 1035/6470 [11:50<57:48,  1.57it/s]  
3|1036|Loss: 0.8480366468429565:  16%|█▌        | 1035/6470 [11:50<57:48,  1.57it/s]
3|1036|Loss: 0.8480366468429565:  16%|█▌        | 1036/6470 [11:51<1:01:05,  1.48it/s]
3|1037|Loss: 0.908405065536499:  16%|█▌        | 1036/6470 [11:51<1:01:05,  1.48it/s] 
3|1037|Loss: 0.908405065536499:  16%|█▌        | 1037/6470 [11:51<1:00:35,  1.49it/s]
3|1038|Loss: 0.7802716493606567:  16%|█▌        | 1037/6470 [11:52<1:00:35,  1.49it/s]
3|1038|Loss: 0.7802716493606567:  16%|█▌        | 1038/6470 [11:52<59:27,  1.52it/s]  
3|1039|Loss: 0.8108828067779541:  16%|█▌        | 1038/6470 [11:52<59:27,  1.52it/s]
3|1039|Loss: 0.8108828067779541:  16%|█▌        | 1039/6470 [11:53<58:34,  1.55it/s]
3|1040|Loss: 0.7742003202438354:  16%|█▌        | 1039/6470 [11:53<58:34,  1.55it/s]
3|1040|Loss: 0.7742003202438354:  16%|█▌        | 1040/6470 [11:53<1:00:29,  1.50it/s]
3|1041|Loss: 0.8942747712135315:  16%|█▌        | 1040/6470 [11:54<1:00:29,  1.50it/s]
3|1041|Loss: 0.8942747712135315:  16%|█▌        | 1041/6470 [11:54<1:00:21,  1.50it/s]
3|1042|Loss: 0.7945342063903809:  16%|█▌        | 1041/6470 [11:54<1:00:21,  1.50it/s]
3|1042|Loss: 0.7945342063903809:  16%|█▌        | 1042/6470 [11:55<1:02:13,  1.45it/s]
3|1043|Loss: 0.9091922640800476:  16%|█▌        | 1042/6470 [11:55<1:02:13,  1.45it/s]
3|1043|Loss: 0.9091922640800476:  16%|█▌        | 1043/6470 [11:55<1:01:43,  1.47it/s]
3|1044|Loss: 0.87970370054245:  16%|█▌        | 1043/6470 [11:56<1:01:43,  1.47it/s]  
3|1044|Loss: 0.87970370054245:  16%|█▌        | 1044/6470 [11:56<1:05:10,  1.39it/s]
3|1045|Loss: 0.9181697964668274:  16%|█▌        | 1044/6470 [11:57<1:05:10,  1.39it/s]
3|1045|Loss: 0.9181697964668274:  16%|█▌        | 1045/6470 [11:57<1:04:14,  1.41it/s]
3|1046|Loss: 0.9382541179656982:  16%|█▌        | 1045/6470 [11:57<1:04:14,  1.41it/s]
3|1046|Loss: 0.9382541179656982:  16%|█▌        | 1046/6470 [11:58<1:02:17,  1.45it/s]
3|1047|Loss: 0.8722683787345886:  16%|█▌        | 1046/6470 [11:58<1:02:17,  1.45it/s]
3|1047|Loss: 0.8722683787345886:  16%|█▌        | 1047/6470 [11:58<1:03:22,  1.43it/s]
3|1048|Loss: 0.6767088174819946:  16%|█▌        | 1047/6470 [11:59<1:03:22,  1.43it/s]
3|1048|Loss: 0.6767088174819946:  16%|█▌        | 1048/6470 [11:59<1:02:24,  1.45it/s]
3|1049|Loss: 0.8554917573928833:  16%|█▌        | 1048/6470 [11:59<1:02:24,  1.45it/s]
3|1049|Loss: 0.8554917573928833:  16%|█▌        | 1049/6470 [12:00<1:03:03,  1.43it/s]
3|1050|Loss: 0.8857287168502808:  16%|█▌        | 1049/6470 [12:00<1:03:03,  1.43it/s]
3|1050|Loss: 0.8857287168502808:  16%|█▌        | 1050/6470 [12:00<1:04:21,  1.40it/s]
3|1051|Loss: 0.8426635265350342:  16%|█▌        | 1050/6470 [12:01<1:04:21,  1.40it/s]
3|1051|Loss: 0.8426635265350342:  16%|█▌        | 1051/6470 [12:01<1:05:42,  1.37it/s]
3|1052|Loss: 0.728660523891449:  16%|█▌        | 1051/6470 [12:02<1:05:42,  1.37it/s] 
3|1052|Loss: 0.728660523891449:  16%|█▋        | 1052/6470 [12:02<1:09:48,  1.29it/s]
3|1053|Loss: 0.8675063252449036:  16%|█▋        | 1052/6470 [12:03<1:09:48,  1.29it/s]
3|1053|Loss: 0.8675063252449036:  16%|█▋        | 1053/6470 [12:03<1:11:43,  1.26it/s]
3|1054|Loss: 0.8952451944351196:  16%|█▋        | 1053/6470 [12:03<1:11:43,  1.26it/s]
3|1054|Loss: 0.8952451944351196:  16%|█▋        | 1054/6470 [12:04<1:09:53,  1.29it/s]
3|1055|Loss: 0.8466223478317261:  16%|█▋        | 1054/6470 [12:04<1:09:53,  1.29it/s]
3|1055|Loss: 0.8466223478317261:  16%|█▋        | 1055/6470 [12:04<1:07:59,  1.33it/s]
3|1056|Loss: 0.8853737115859985:  16%|█▋        | 1055/6470 [12:05<1:07:59,  1.33it/s]
3|1056|Loss: 0.8853737115859985:  16%|█▋        | 1056/6470 [12:05<1:07:21,  1.34it/s]
3|1057|Loss: 1.0903700590133667:  16%|█▋        | 1056/6470 [12:06<1:07:21,  1.34it/s]
3|1057|Loss: 1.0903700590133667:  16%|█▋        | 1057/6470 [12:06<1:09:05,  1.31it/s]
3|1058|Loss: 0.8433220982551575:  16%|█▋        | 1057/6470 [12:06<1:09:05,  1.31it/s]
3|1058|Loss: 0.8433220982551575:  16%|█▋        | 1058/6470 [12:07<1:12:02,  1.25it/s]
3|1059|Loss: 0.8167093992233276:  16%|█▋        | 1058/6470 [12:07<1:12:02,  1.25it/s]
3|1059|Loss: 0.8167093992233276:  16%|█▋        | 1059/6470 [12:08<1:11:38,  1.26it/s]
3|1060|Loss: 0.7108450531959534:  16%|█▋        | 1059/6470 [12:08<1:11:38,  1.26it/s]
3|1060|Loss: 0.7108450531959534:  16%|█▋        | 1060/6470 [12:08<1:09:23,  1.30it/s]
3|1061|Loss: 0.8195595741271973:  16%|█▋        | 1060/6470 [12:09<1:09:23,  1.30it/s]
3|1061|Loss: 0.8195595741271973:  16%|█▋        | 1061/6470 [12:09<1:05:46,  1.37it/s]
3|1062|Loss: 0.748570442199707:  16%|█▋        | 1061/6470 [12:09<1:05:46,  1.37it/s] 
3|1062|Loss: 0.748570442199707:  16%|█▋        | 1062/6470 [12:10<1:05:10,  1.38it/s]
3|1063|Loss: 0.785250186920166:  16%|█▋        | 1062/6470 [12:10<1:05:10,  1.38it/s]
3|1063|Loss: 0.785250186920166:  16%|█▋        | 1063/6470 [12:10<1:05:52,  1.37it/s]
3|1064|Loss: 0.7769026160240173:  16%|█▋        | 1063/6470 [12:11<1:05:52,  1.37it/s]
3|1064|Loss: 0.7769026160240173:  16%|█▋        | 1064/6470 [12:11<1:04:07,  1.40it/s]
3|1065|Loss: 0.8836018443107605:  16%|█▋        | 1064/6470 [12:11<1:04:07,  1.40it/s]
3|1065|Loss: 0.8836018443107605:  16%|█▋        | 1065/6470 [12:12<1:03:37,  1.42it/s]
3|1066|Loss: 0.8269198536872864:  16%|█▋        | 1065/6470 [12:12<1:03:37,  1.42it/s]
3|1066|Loss: 0.8269198536872864:  16%|█▋        | 1066/6470 [12:12<1:03:32,  1.42it/s]
3|1067|Loss: 0.7563071846961975:  16%|█▋        | 1066/6470 [12:13<1:03:32,  1.42it/s]
3|1067|Loss: 0.7563071846961975:  16%|█▋        | 1067/6470 [12:13<1:03:18,  1.42it/s]
3|1068|Loss: 0.7702884674072266:  16%|█▋        | 1067/6470 [12:14<1:03:18,  1.42it/s]
3|1068|Loss: 0.7702884674072266:  17%|█▋        | 1068/6470 [12:14<1:01:14,  1.47it/s]
3|1069|Loss: 1.0062562227249146:  17%|█▋        | 1068/6470 [12:14<1:01:14,  1.47it/s]
3|1069|Loss: 1.0062562227249146:  17%|█▋        | 1069/6470 [12:14<58:46,  1.53it/s]  
3|1070|Loss: 0.8720088005065918:  17%|█▋        | 1069/6470 [12:15<58:46,  1.53it/s]
3|1070|Loss: 0.8720088005065918:  17%|█▋        | 1070/6470 [12:15<57:32,  1.56it/s]
3|1071|Loss: 0.9137837290763855:  17%|█▋        | 1070/6470 [12:15<57:32,  1.56it/s]
3|1071|Loss: 0.9137837290763855:  17%|█▋        | 1071/6470 [12:16<59:15,  1.52it/s]
3|1072|Loss: 0.7311863899230957:  17%|█▋        | 1071/6470 [12:16<59:15,  1.52it/s]
3|1072|Loss: 0.7311863899230957:  17%|█▋        | 1072/6470 [12:16<1:02:47,  1.43it/s]
3|1073|Loss: 0.8674328923225403:  17%|█▋        | 1072/6470 [12:17<1:02:47,  1.43it/s]
3|1073|Loss: 0.8674328923225403:  17%|█▋        | 1073/6470 [12:17<59:47,  1.50it/s]  
3|1074|Loss: 0.9436268210411072:  17%|█▋        | 1073/6470 [12:17<59:47,  1.50it/s]
3|1074|Loss: 0.9436268210411072:  17%|█▋        | 1074/6470 [12:18<57:04,  1.58it/s]
3|1075|Loss: 0.839783251285553:  17%|█▋        | 1074/6470 [12:18<57:04,  1.58it/s] 
3|1075|Loss: 0.839783251285553:  17%|█▋        | 1075/6470 [12:18<54:26,  1.65it/s]
3|1076|Loss: 0.839045524597168:  17%|█▋        | 1075/6470 [12:19<54:26,  1.65it/s]
3|1076|Loss: 0.839045524597168:  17%|█▋        | 1076/6470 [12:19<57:25,  1.57it/s]
3|1077|Loss: 0.8782204389572144:  17%|█▋        | 1076/6470 [12:19<57:25,  1.57it/s]
3|1077|Loss: 0.8782204389572144:  17%|█▋        | 1077/6470 [12:20<56:59,  1.58it/s]
3|1078|Loss: 0.7546432614326477:  17%|█▋        | 1077/6470 [12:20<56:59,  1.58it/s]
3|1078|Loss: 0.7546432614326477:  17%|█▋        | 1078/6470 [12:20<58:53,  1.53it/s]
3|1079|Loss: 0.8563315272331238:  17%|█▋        | 1078/6470 [12:21<58:53,  1.53it/s]
3|1079|Loss: 0.8563315272331238:  17%|█▋        | 1079/6470 [12:21<1:01:00,  1.47it/s]
3|1080|Loss: 0.8809714913368225:  17%|█▋        | 1079/6470 [12:21<1:01:00,  1.47it/s]
3|1080|Loss: 0.8809714913368225:  17%|█▋        | 1080/6470 [12:22<1:00:01,  1.50it/s]
3|1081|Loss: 0.951120138168335:  17%|█▋        | 1080/6470 [12:22<1:00:01,  1.50it/s] 
3|1081|Loss: 0.951120138168335:  17%|█▋        | 1081/6470 [12:22<57:39,  1.56it/s]  
3|1082|Loss: 0.787347674369812:  17%|█▋        | 1081/6470 [12:23<57:39,  1.56it/s]
3|1082|Loss: 0.787347674369812:  17%|█▋        | 1082/6470 [12:23<58:38,  1.53it/s]
3|1083|Loss: 0.8238927125930786:  17%|█▋        | 1082/6470 [12:23<58:38,  1.53it/s]
3|1083|Loss: 0.8238927125930786:  17%|█▋        | 1083/6470 [12:24<1:01:54,  1.45it/s]
3|1084|Loss: 0.9366858005523682:  17%|█▋        | 1083/6470 [12:24<1:01:54,  1.45it/s]
3|1084|Loss: 0.9366858005523682:  17%|█▋        | 1084/6470 [12:24<1:04:54,  1.38it/s]
3|1085|Loss: 0.7702775597572327:  17%|█▋        | 1084/6470 [12:25<1:04:54,  1.38it/s]
3|1085|Loss: 0.7702775597572327:  17%|█▋        | 1085/6470 [12:25<1:06:05,  1.36it/s]
3|1086|Loss: 0.7675797939300537:  17%|█▋        | 1085/6470 [12:26<1:06:05,  1.36it/s]
3|1086|Loss: 0.7675797939300537:  17%|█▋        | 1086/6470 [12:26<1:07:45,  1.32it/s]
3|1087|Loss: 1.029619812965393:  17%|█▋        | 1086/6470 [12:26<1:07:45,  1.32it/s] 
3|1087|Loss: 1.029619812965393:  17%|█▋        | 1087/6470 [12:27<1:07:34,  1.33it/s]
3|1088|Loss: 0.8060809969902039:  17%|█▋        | 1087/6470 [12:27<1:07:34,  1.33it/s]
3|1088|Loss: 0.8060809969902039:  17%|█▋        | 1088/6470 [12:27<1:07:20,  1.33it/s]
3|1089|Loss: 0.9128662347793579:  17%|█▋        | 1088/6470 [12:28<1:07:20,  1.33it/s]
3|1089|Loss: 0.9128662347793579:  17%|█▋        | 1089/6470 [12:28<1:04:28,  1.39it/s]
3|1090|Loss: 0.7902991771697998:  17%|█▋        | 1089/6470 [12:28<1:04:28,  1.39it/s]
3|1090|Loss: 0.7902991771697998:  17%|█▋        | 1090/6470 [12:29<1:00:40,  1.48it/s]
3|1091|Loss: 0.8523852229118347:  17%|█▋        | 1090/6470 [12:29<1:00:40,  1.48it/s]
3|1091|Loss: 0.8523852229118347:  17%|█▋        | 1091/6470 [12:29<1:00:53,  1.47it/s]
3|1092|Loss: 0.6967492699623108:  17%|█▋        | 1091/6470 [12:30<1:00:53,  1.47it/s]
3|1092|Loss: 0.6967492699623108:  17%|█▋        | 1092/6470 [12:30<59:11,  1.51it/s]  
3|1093|Loss: 0.6854665279388428:  17%|█▋        | 1092/6470 [12:30<59:11,  1.51it/s]
3|1093|Loss: 0.6854665279388428:  17%|█▋        | 1093/6470 [12:31<58:14,  1.54it/s]
3|1094|Loss: 0.9212191700935364:  17%|█▋        | 1093/6470 [12:31<58:14,  1.54it/s]
3|1094|Loss: 0.9212191700935364:  17%|█▋        | 1094/6470 [12:31<59:56,  1.49it/s]
3|1095|Loss: 0.887945830821991:  17%|█▋        | 1094/6470 [12:32<59:56,  1.49it/s] 
3|1095|Loss: 0.887945830821991:  17%|█▋        | 1095/6470 [12:32<1:01:42,  1.45it/s]
3|1096|Loss: 0.8983917236328125:  17%|█▋        | 1095/6470 [12:32<1:01:42,  1.45it/s]
3|1096|Loss: 0.8983917236328125:  17%|█▋        | 1096/6470 [12:33<55:05,  1.63it/s]  
3|1097|Loss: 0.7240228056907654:  17%|█▋        | 1096/6470 [12:33<55:05,  1.63it/s]
3|1097|Loss: 0.7240228056907654:  17%|█▋        | 1097/6470 [12:33<54:20,  1.65it/s]
3|1098|Loss: 0.7566147446632385:  17%|█▋        | 1097/6470 [12:33<54:20,  1.65it/s]
3|1098|Loss: 0.7566147446632385:  17%|█▋        | 1098/6470 [12:34<55:22,  1.62it/s]
3|1099|Loss: 0.7173712253570557:  17%|█▋        | 1098/6470 [12:34<55:22,  1.62it/s]
3|1099|Loss: 0.7173712253570557:  17%|█▋        | 1099/6470 [12:34<51:58,  1.72it/s]
3|1100|Loss: 0.9624509811401367:  17%|█▋        | 1099/6470 [12:35<51:58,  1.72it/s]
3|1100|Loss: 0.9624509811401367:  17%|█▋        | 1100/6470 [12:35<51:20,  1.74it/s]
3|1101|Loss: 0.8759560585021973:  17%|█▋        | 1100/6470 [12:35<51:20,  1.74it/s]
3|1101|Loss: 0.8759560585021973:  17%|█▋        | 1101/6470 [12:36<58:44,  1.52it/s]
3|1102|Loss: 0.7230027318000793:  17%|█▋        | 1101/6470 [12:36<58:44,  1.52it/s]
3|1102|Loss: 0.7230027318000793:  17%|█▋        | 1102/6470 [12:36<1:02:18,  1.44it/s]
3|1103|Loss: 0.91585773229599:  17%|█▋        | 1102/6470 [12:37<1:02:18,  1.44it/s]  
3|1103|Loss: 0.91585773229599:  17%|█▋        | 1103/6470 [12:37<1:05:29,  1.37it/s]
3|1104|Loss: 0.6348444223403931:  17%|█▋        | 1103/6470 [12:38<1:05:29,  1.37it/s]
3|1104|Loss: 0.6348444223403931:  17%|█▋        | 1104/6470 [12:38<1:03:27,  1.41it/s]
3|1105|Loss: 0.9268995523452759:  17%|█▋        | 1104/6470 [12:38<1:03:27,  1.41it/s]
3|1105|Loss: 0.9268995523452759:  17%|█▋        | 1105/6470 [12:38<59:28,  1.50it/s]  
3|1106|Loss: 0.7519583702087402:  17%|█▋        | 1105/6470 [12:39<59:28,  1.50it/s]
3|1106|Loss: 0.7519583702087402:  17%|█▋        | 1106/6470 [12:39<1:04:21,  1.39it/s]
3|1107|Loss: 0.8500227928161621:  17%|█▋        | 1106/6470 [12:40<1:04:21,  1.39it/s]
3|1107|Loss: 0.8500227928161621:  17%|█▋        | 1107/6470 [12:40<1:07:05,  1.33it/s]
3|1108|Loss: 0.7742922902107239:  17%|█▋        | 1107/6470 [12:41<1:07:05,  1.33it/s]
3|1108|Loss: 0.7742922902107239:  17%|█▋        | 1108/6470 [12:41<1:09:37,  1.28it/s]
3|1109|Loss: 0.7901442646980286:  17%|█▋        | 1108/6470 [12:41<1:09:37,  1.28it/s]
3|1109|Loss: 0.7901442646980286:  17%|█▋        | 1109/6470 [12:42<1:09:14,  1.29it/s]
3|1110|Loss: 0.8745774030685425:  17%|█▋        | 1109/6470 [12:42<1:09:14,  1.29it/s]
3|1110|Loss: 0.8745774030685425:  17%|█▋        | 1110/6470 [12:43<1:11:50,  1.24it/s]
3|1111|Loss: 0.8849943280220032:  17%|█▋        | 1110/6470 [12:43<1:11:50,  1.24it/s]
3|1111|Loss: 0.8849943280220032:  17%|█▋        | 1111/6470 [12:43<1:08:33,  1.30it/s]
3|1112|Loss: 0.8833135366439819:  17%|█▋        | 1111/6470 [12:44<1:08:33,  1.30it/s]
3|1112|Loss: 0.8833135366439819:  17%|█▋        | 1112/6470 [12:44<1:06:50,  1.34it/s]
3|1113|Loss: 0.8105754256248474:  17%|█▋        | 1112/6470 [12:44<1:06:50,  1.34it/s]
3|1113|Loss: 0.8105754256248474:  17%|█▋        | 1113/6470 [12:45<1:08:35,  1.30it/s]
3|1114|Loss: 0.9207190275192261:  17%|█▋        | 1113/6470 [12:45<1:08:35,  1.30it/s]
3|1114|Loss: 0.9207190275192261:  17%|█▋        | 1114/6470 [12:46<1:08:35,  1.30it/s]
3|1115|Loss: 0.8875518441200256:  17%|█▋        | 1114/6470 [12:46<1:08:35,  1.30it/s]
3|1115|Loss: 0.8875518441200256:  17%|█▋        | 1115/6470 [12:46<1:10:29,  1.27it/s]
3|1116|Loss: 0.8675450086593628:  17%|█▋        | 1115/6470 [12:47<1:10:29,  1.27it/s]
3|1116|Loss: 0.8675450086593628:  17%|█▋        | 1116/6470 [12:47<1:07:53,  1.31it/s]
3|1117|Loss: 0.8322404026985168:  17%|█▋        | 1116/6470 [12:48<1:07:53,  1.31it/s]
3|1117|Loss: 0.8322404026985168:  17%|█▋        | 1117/6470 [12:48<1:06:59,  1.33it/s]
3|1118|Loss: 0.7977321147918701:  17%|█▋        | 1117/6470 [12:48<1:06:59,  1.33it/s]
3|1118|Loss: 0.7977321147918701:  17%|█▋        | 1118/6470 [12:49<1:04:18,  1.39it/s]
3|1119|Loss: 0.917680025100708:  17%|█▋        | 1118/6470 [12:49<1:04:18,  1.39it/s] 
3|1119|Loss: 0.917680025100708:  17%|█▋        | 1119/6470 [12:49<59:56,  1.49it/s]  
3|1120|Loss: 0.5618255734443665:  17%|█▋        | 1119/6470 [12:50<59:56,  1.49it/s]
3|1120|Loss: 0.5618255734443665:  17%|█▋        | 1120/6470 [12:50<1:07:39,  1.32it/s]
3|1121|Loss: 0.8137293457984924:  17%|█▋        | 1120/6470 [12:50<1:07:39,  1.32it/s]
3|1121|Loss: 0.8137293457984924:  17%|█▋        | 1121/6470 [12:51<1:05:26,  1.36it/s]
3|1122|Loss: 0.8010022640228271:  17%|█▋        | 1121/6470 [12:51<1:05:26,  1.36it/s]
3|1122|Loss: 0.8010022640228271:  17%|█▋        | 1122/6470 [12:51<1:04:54,  1.37it/s]
3|1123|Loss: 0.8937456607818604:  17%|█▋        | 1122/6470 [12:52<1:04:54,  1.37it/s]
3|1123|Loss: 0.8937456607818604:  17%|█▋        | 1123/6470 [12:52<1:00:23,  1.48it/s]
3|1124|Loss: 0.83807772397995:  17%|█▋        | 1123/6470 [12:52<1:00:23,  1.48it/s]  
3|1124|Loss: 0.83807772397995:  17%|█▋        | 1124/6470 [12:53<56:41,  1.57it/s]  
3|1125|Loss: 0.9401890635490417:  17%|█▋        | 1124/6470 [12:53<56:41,  1.57it/s]
3|1125|Loss: 0.9401890635490417:  17%|█▋        | 1125/6470 [12:53<58:20,  1.53it/s]
3|1126|Loss: 0.7894212007522583:  17%|█▋        | 1125/6470 [12:54<58:20,  1.53it/s]
3|1126|Loss: 0.7894212007522583:  17%|█▋        | 1126/6470 [12:54<1:00:20,  1.48it/s]
3|1127|Loss: 0.8794996738433838:  17%|█▋        | 1126/6470 [12:54<1:00:20,  1.48it/s]
3|1127|Loss: 0.8794996738433838:  17%|█▋        | 1127/6470 [12:55<59:53,  1.49it/s]  
3|1128|Loss: 0.9413851499557495:  17%|█▋        | 1127/6470 [12:55<59:53,  1.49it/s]
3|1128|Loss: 0.9413851499557495:  17%|█▋        | 1128/6470 [12:55<1:00:26,  1.47it/s]
3|1129|Loss: 0.6974326968193054:  17%|█▋        | 1128/6470 [12:56<1:00:26,  1.47it/s]
3|1129|Loss: 0.6974326968193054:  17%|█▋        | 1129/6470 [12:56<58:21,  1.53it/s]  
3|1130|Loss: 0.9531961679458618:  17%|█▋        | 1129/6470 [12:56<58:21,  1.53it/s]
3|1130|Loss: 0.9531961679458618:  17%|█▋        | 1130/6470 [12:56<54:20,  1.64it/s]
3|1131|Loss: 0.7440043091773987:  17%|█▋        | 1130/6470 [12:57<54:20,  1.64it/s]
3|1131|Loss: 0.7440043091773987:  17%|█▋        | 1131/6470 [12:57<56:05,  1.59it/s]
3|1132|Loss: 0.7272948026657104:  17%|█▋        | 1131/6470 [12:58<56:05,  1.59it/s]
3|1132|Loss: 0.7272948026657104:  17%|█▋        | 1132/6470 [12:58<1:01:47,  1.44it/s]
3|1133|Loss: 0.8744661211967468:  17%|█▋        | 1132/6470 [12:58<1:01:47,  1.44it/s]
3|1133|Loss: 0.8744661211967468:  18%|█▊        | 1133/6470 [12:59<1:03:05,  1.41it/s]
3|1134|Loss: 0.7989777326583862:  18%|█▊        | 1133/6470 [12:59<1:03:05,  1.41it/s]
3|1134|Loss: 0.7989777326583862:  18%|█▊        | 1134/6470 [12:59<1:03:21,  1.40it/s]
3|1135|Loss: 0.7340894937515259:  18%|█▊        | 1134/6470 [13:00<1:03:21,  1.40it/s]
3|1135|Loss: 0.7340894937515259:  18%|█▊        | 1135/6470 [13:00<1:03:20,  1.40it/s]
3|1136|Loss: 0.7278034687042236:  18%|█▊        | 1135/6470 [13:00<1:03:20,  1.40it/s]
3|1136|Loss: 0.7278034687042236:  18%|█▊        | 1136/6470 [13:01<1:01:14,  1.45it/s]
3|1137|Loss: 0.840598464012146:  18%|█▊        | 1136/6470 [13:01<1:01:14,  1.45it/s] 
3|1137|Loss: 0.840598464012146:  18%|█▊        | 1137/6470 [13:01<59:43,  1.49it/s]  
3|1138|Loss: 0.8834773898124695:  18%|█▊        | 1137/6470 [13:02<59:43,  1.49it/s]
3|1138|Loss: 0.8834773898124695:  18%|█▊        | 1138/6470 [13:02<1:03:49,  1.39it/s]
3|1139|Loss: 0.9039292335510254:  18%|█▊        | 1138/6470 [13:03<1:03:49,  1.39it/s]
3|1139|Loss: 0.9039292335510254:  18%|█▊        | 1139/6470 [13:03<1:03:15,  1.40it/s]
3|1140|Loss: 0.8006975650787354:  18%|█▊        | 1139/6470 [13:03<1:03:15,  1.40it/s]
3|1140|Loss: 0.8006975650787354:  18%|█▊        | 1140/6470 [13:04<1:01:22,  1.45it/s]
3|1141|Loss: 0.9385669231414795:  18%|█▊        | 1140/6470 [13:04<1:01:22,  1.45it/s]
3|1141|Loss: 0.9385669231414795:  18%|█▊        | 1141/6470 [13:04<1:01:28,  1.44it/s]
3|1142|Loss: 0.9319277405738831:  18%|█▊        | 1141/6470 [13:05<1:01:28,  1.44it/s]
3|1142|Loss: 0.9319277405738831:  18%|█▊        | 1142/6470 [13:05<1:02:23,  1.42it/s]
3|1143|Loss: 0.9225428104400635:  18%|█▊        | 1142/6470 [13:05<1:02:23,  1.42it/s]
3|1143|Loss: 0.9225428104400635:  18%|█▊        | 1143/6470 [13:06<1:00:45,  1.46it/s]
3|1144|Loss: 0.8920931816101074:  18%|█▊        | 1143/6470 [13:06<1:00:45,  1.46it/s]
3|1144|Loss: 0.8920931816101074:  18%|█▊        | 1144/6470 [13:06<1:02:34,  1.42it/s]
3|1145|Loss: 0.8253636956214905:  18%|█▊        | 1144/6470 [13:07<1:02:34,  1.42it/s]
3|1145|Loss: 0.8253636956214905:  18%|█▊        | 1145/6470 [13:07<1:00:45,  1.46it/s]
3|1146|Loss: 0.8297840356826782:  18%|█▊        | 1145/6470 [13:07<1:00:45,  1.46it/s]
3|1146|Loss: 0.8297840356826782:  18%|█▊        | 1146/6470 [13:08<59:11,  1.50it/s]  
3|1147|Loss: 0.8318729996681213:  18%|█▊        | 1146/6470 [13:08<59:11,  1.50it/s]
3|1147|Loss: 0.8318729996681213:  18%|█▊        | 1147/6470 [13:08<55:58,  1.58it/s]
3|1148|Loss: 0.7867229580879211:  18%|█▊        | 1147/6470 [13:09<55:58,  1.58it/s]
3|1148|Loss: 0.7867229580879211:  18%|█▊        | 1148/6470 [13:09<55:26,  1.60it/s]
3|1149|Loss: 0.7897807955741882:  18%|█▊        | 1148/6470 [13:09<55:26,  1.60it/s]
3|1149|Loss: 0.7897807955741882:  18%|█▊        | 1149/6470 [13:10<58:07,  1.53it/s]
3|1150|Loss: 0.8154200911521912:  18%|█▊        | 1149/6470 [13:10<58:07,  1.53it/s]
3|1150|Loss: 0.8154200911521912:  18%|█▊        | 1150/6470 [13:10<54:23,  1.63it/s]
3|1151|Loss: 0.930968165397644:  18%|█▊        | 1150/6470 [13:10<54:23,  1.63it/s] 
3|1151|Loss: 0.930968165397644:  18%|█▊        | 1151/6470 [13:11<57:27,  1.54it/s]
3|1152|Loss: 0.7785512804985046:  18%|█▊        | 1151/6470 [13:11<57:27,  1.54it/s]
3|1152|Loss: 0.7785512804985046:  18%|█▊        | 1152/6470 [13:12<1:03:20,  1.40it/s]
3|1153|Loss: 0.8082284927368164:  18%|█▊        | 1152/6470 [13:12<1:03:20,  1.40it/s]
3|1153|Loss: 0.8082284927368164:  18%|█▊        | 1153/6470 [13:12<1:04:04,  1.38it/s]
3|1154|Loss: 0.9494235515594482:  18%|█▊        | 1153/6470 [13:13<1:04:04,  1.38it/s]
3|1154|Loss: 0.9494235515594482:  18%|█▊        | 1154/6470 [13:13<58:10,  1.52it/s]  
3|1155|Loss: 0.9214565753936768:  18%|█▊        | 1154/6470 [13:13<58:10,  1.52it/s]
3|1155|Loss: 0.9214565753936768:  18%|█▊        | 1155/6470 [13:13<53:47,  1.65it/s]
3|1156|Loss: 1.0240490436553955:  18%|█▊        | 1155/6470 [13:14<53:47,  1.65it/s]
3|1156|Loss: 1.0240490436553955:  18%|█▊        | 1156/6470 [13:14<55:59,  1.58it/s]
3|1157|Loss: 0.7945796847343445:  18%|█▊        | 1156/6470 [13:14<55:59,  1.58it/s]
3|1157|Loss: 0.7945796847343445:  18%|█▊        | 1157/6470 [13:15<1:00:41,  1.46it/s]
3|1158|Loss: 0.7968188524246216:  18%|█▊        | 1157/6470 [13:15<1:00:41,  1.46it/s]
3|1158|Loss: 0.7968188524246216:  18%|█▊        | 1158/6470 [13:16<1:00:34,  1.46it/s]
3|1159|Loss: 0.8872310519218445:  18%|█▊        | 1158/6470 [13:16<1:00:34,  1.46it/s]
3|1159|Loss: 0.8872310519218445:  18%|█▊        | 1159/6470 [13:16<59:27,  1.49it/s]  
3|1160|Loss: 0.8157889246940613:  18%|█▊        | 1159/6470 [13:17<59:27,  1.49it/s]
3|1160|Loss: 0.8157889246940613:  18%|█▊        | 1160/6470 [13:17<59:58,  1.48it/s]
3|1161|Loss: 1.0198464393615723:  18%|█▊        | 1160/6470 [13:17<59:58,  1.48it/s]
3|1161|Loss: 1.0198464393615723:  18%|█▊        | 1161/6470 [13:18<1:01:28,  1.44it/s]
3|1162|Loss: 0.9180780649185181:  18%|█▊        | 1161/6470 [13:18<1:01:28,  1.44it/s]
3|1162|Loss: 0.9180780649185181:  18%|█▊        | 1162/6470 [13:18<1:02:36,  1.41it/s]
3|1163|Loss: 0.6785749793052673:  18%|█▊        | 1162/6470 [13:19<1:02:36,  1.41it/s]
3|1163|Loss: 0.6785749793052673:  18%|█▊        | 1163/6470 [13:19<59:17,  1.49it/s]  
3|1164|Loss: 0.8504183292388916:  18%|█▊        | 1163/6470 [13:19<59:17,  1.49it/s]
3|1164|Loss: 0.8504183292388916:  18%|█▊        | 1164/6470 [13:20<1:00:17,  1.47it/s]
3|1165|Loss: 0.8102067112922668:  18%|█▊        | 1164/6470 [13:20<1:00:17,  1.47it/s]
3|1165|Loss: 0.8102067112922668:  18%|█▊        | 1165/6470 [13:21<1:05:25,  1.35it/s]
3|1166|Loss: 0.9144912362098694:  18%|█▊        | 1165/6470 [13:21<1:05:25,  1.35it/s]
3|1166|Loss: 0.9144912362098694:  18%|█▊        | 1166/6470 [13:21<1:06:19,  1.33it/s]
3|1167|Loss: 0.8282816410064697:  18%|█▊        | 1166/6470 [13:22<1:06:19,  1.33it/s]
3|1167|Loss: 0.8282816410064697:  18%|█▊        | 1167/6470 [13:22<1:07:51,  1.30it/s]
3|1168|Loss: 0.7657210826873779:  18%|█▊        | 1167/6470 [13:23<1:07:51,  1.30it/s]
3|1168|Loss: 0.7657210826873779:  18%|█▊        | 1168/6470 [13:23<1:15:39,  1.17it/s]
3|1169|Loss: 0.7121750116348267:  18%|█▊        | 1168/6470 [13:24<1:15:39,  1.17it/s]
3|1169|Loss: 0.7121750116348267:  18%|█▊        | 1169/6470 [13:24<1:10:59,  1.24it/s]
3|1170|Loss: 0.7172228097915649:  18%|█▊        | 1169/6470 [13:24<1:10:59,  1.24it/s]
3|1170|Loss: 0.7172228097915649:  18%|█▊        | 1170/6470 [13:24<1:05:44,  1.34it/s]
3|1171|Loss: 0.8337964415550232:  18%|█▊        | 1170/6470 [13:25<1:05:44,  1.34it/s]
3|1171|Loss: 0.8337964415550232:  18%|█▊        | 1171/6470 [13:25<1:02:42,  1.41it/s]
3|1172|Loss: 0.6812520623207092:  18%|█▊        | 1171/6470 [13:26<1:02:42,  1.41it/s]
3|1172|Loss: 0.6812520623207092:  18%|█▊        | 1172/6470 [13:26<1:05:27,  1.35it/s]
3|1173|Loss: 0.931315004825592:  18%|█▊        | 1172/6470 [13:26<1:05:27,  1.35it/s] 
3|1173|Loss: 0.931315004825592:  18%|█▊        | 1173/6470 [13:27<1:08:05,  1.30it/s]
3|1174|Loss: 0.836707592010498:  18%|█▊        | 1173/6470 [13:27<1:08:05,  1.30it/s]
3|1174|Loss: 0.836707592010498:  18%|█▊        | 1174/6470 [13:27<1:02:52,  1.40it/s]
3|1175|Loss: 0.8795455098152161:  18%|█▊        | 1174/6470 [13:28<1:02:52,  1.40it/s]
3|1175|Loss: 0.8795455098152161:  18%|█▊        | 1175/6470 [13:28<54:55,  1.61it/s]  
3|1176|Loss: 0.7233070731163025:  18%|█▊        | 1175/6470 [13:28<54:55,  1.61it/s]
3|1176|Loss: 0.7233070731163025:  18%|█▊        | 1176/6470 [13:28<48:02,  1.84it/s]
3|1177|Loss: 0.9078070521354675:  18%|█▊        | 1176/6470 [13:28<48:02,  1.84it/s]
3|1177|Loss: 0.9078070521354675:  18%|█▊        | 1177/6470 [13:29<50:34,  1.74it/s]
3|1178|Loss: 0.7350059151649475:  18%|█▊        | 1177/6470 [13:29<50:34,  1.74it/s]
3|1178|Loss: 0.7350059151649475:  18%|█▊        | 1178/6470 [13:29<55:13,  1.60it/s]
3|1179|Loss: 0.806917667388916:  18%|█▊        | 1178/6470 [13:30<55:13,  1.60it/s] 
3|1179|Loss: 0.806917667388916:  18%|█▊        | 1179/6470 [13:30<55:57,  1.58it/s]
3|1180|Loss: 0.9262176156044006:  18%|█▊        | 1179/6470 [13:31<55:57,  1.58it/s]
3|1180|Loss: 0.9262176156044006:  18%|█▊        | 1180/6470 [13:31<57:27,  1.53it/s]
3|1181|Loss: 0.9782721996307373:  18%|█▊        | 1180/6470 [13:31<57:27,  1.53it/s]
3|1181|Loss: 0.9782721996307373:  18%|█▊        | 1181/6470 [13:32<1:01:28,  1.43it/s]
3|1182|Loss: 0.8943306803703308:  18%|█▊        | 1181/6470 [13:32<1:01:28,  1.43it/s]
3|1182|Loss: 0.8943306803703308:  18%|█▊        | 1182/6470 [13:32<59:05,  1.49it/s]  
3|1183|Loss: 0.9025780558586121:  18%|█▊        | 1182/6470 [13:33<59:05,  1.49it/s]
3|1183|Loss: 0.9025780558586121:  18%|█▊        | 1183/6470 [13:33<1:01:57,  1.42it/s]
3|1184|Loss: 0.8800408840179443:  18%|█▊        | 1183/6470 [13:33<1:01:57,  1.42it/s]
3|1184|Loss: 0.8800408840179443:  18%|█▊        | 1184/6470 [13:34<57:27,  1.53it/s]  
3|1185|Loss: 0.8923547863960266:  18%|█▊        | 1184/6470 [13:34<57:27,  1.53it/s]
3|1185|Loss: 0.8923547863960266:  18%|█▊        | 1185/6470 [13:34<57:46,  1.52it/s]
3|1186|Loss: 1.007134199142456:  18%|█▊        | 1185/6470 [13:35<57:46,  1.52it/s] 
3|1186|Loss: 1.007134199142456:  18%|█▊        | 1186/6470 [13:35<57:25,  1.53it/s]
3|1187|Loss: 0.8724387884140015:  18%|█▊        | 1186/6470 [13:35<57:25,  1.53it/s]
3|1187|Loss: 0.8724387884140015:  18%|█▊        | 1187/6470 [13:35<53:39,  1.64it/s]
3|1188|Loss: 0.9283384084701538:  18%|█▊        | 1187/6470 [13:36<53:39,  1.64it/s]
3|1188|Loss: 0.9283384084701538:  18%|█▊        | 1188/6470 [13:36<55:32,  1.59it/s]
3|1189|Loss: 0.9386605024337769:  18%|█▊        | 1188/6470 [13:36<55:32,  1.59it/s]
3|1189|Loss: 0.9386605024337769:  18%|█▊        | 1189/6470 [13:37<53:46,  1.64it/s]
3|1190|Loss: 0.8579765558242798:  18%|█▊        | 1189/6470 [13:37<53:46,  1.64it/s]
3|1190|Loss: 0.8579765558242798:  18%|█▊        | 1190/6470 [13:37<52:32,  1.67it/s]
3|1191|Loss: 0.749342143535614:  18%|█▊        | 1190/6470 [13:38<52:32,  1.67it/s] 
3|1191|Loss: 0.749342143535614:  18%|█▊        | 1191/6470 [13:38<52:52,  1.66it/s]
3|1192|Loss: 0.8697026371955872:  18%|█▊        | 1191/6470 [13:38<52:52,  1.66it/s]
3|1192|Loss: 0.8697026371955872:  18%|█▊        | 1192/6470 [13:39<58:28,  1.50it/s]
3|1193|Loss: 1.0665634870529175:  18%|█▊        | 1192/6470 [13:39<58:28,  1.50it/s]
3|1193|Loss: 1.0665634870529175:  18%|█▊        | 1193/6470 [13:39<57:11,  1.54it/s]
3|1194|Loss: 0.832429051399231:  18%|█▊        | 1193/6470 [13:40<57:11,  1.54it/s] 
3|1194|Loss: 0.832429051399231:  18%|█▊        | 1194/6470 [13:40<55:08,  1.59it/s]
3|1195|Loss: 0.7165924906730652:  18%|█▊        | 1194/6470 [13:40<55:08,  1.59it/s]
3|1195|Loss: 0.7165924906730652:  18%|█▊        | 1195/6470 [13:41<59:06,  1.49it/s]
3|1196|Loss: 0.8351679444313049:  18%|█▊        | 1195/6470 [13:41<59:06,  1.49it/s]
3|1196|Loss: 0.8351679444313049:  18%|█▊        | 1196/6470 [13:41<1:02:33,  1.41it/s]
3|1197|Loss: 0.9184556007385254:  18%|█▊        | 1196/6470 [13:42<1:02:33,  1.41it/s]
3|1197|Loss: 0.9184556007385254:  19%|█▊        | 1197/6470 [13:42<1:02:51,  1.40it/s]
3|1198|Loss: 0.6036155819892883:  19%|█▊        | 1197/6470 [13:42<1:02:51,  1.40it/s]
3|1198|Loss: 0.6036155819892883:  19%|█▊        | 1198/6470 [13:43<59:19,  1.48it/s]  
3|1199|Loss: 0.7854434251785278:  19%|█▊        | 1198/6470 [13:43<59:19,  1.48it/s]
3|1199|Loss: 0.7854434251785278:  19%|█▊        | 1199/6470 [13:43<1:01:20,  1.43it/s]
3|1200|Loss: 0.9851961135864258:  19%|█▊        | 1199/6470 [13:44<1:01:20,  1.43it/s]
3|1200|Loss: 0.9851961135864258:  19%|█▊        | 1200/6470 [13:44<1:02:32,  1.40it/s]
3|1201|Loss: 0.8403565883636475:  19%|█▊        | 1200/6470 [13:45<1:02:32,  1.40it/s]
3|1201|Loss: 0.8403565883636475:  19%|█▊        | 1201/6470 [13:45<1:01:10,  1.44it/s]
3|1202|Loss: 0.8685113191604614:  19%|█▊        | 1201/6470 [13:45<1:01:10,  1.44it/s]
3|1202|Loss: 0.8685113191604614:  19%|█▊        | 1202/6470 [13:46<1:07:13,  1.31it/s]
3|1203|Loss: 0.882060706615448:  19%|█▊        | 1202/6470 [13:46<1:07:13,  1.31it/s] 
3|1203|Loss: 0.882060706615448:  19%|█▊        | 1203/6470 [13:46<1:06:05,  1.33it/s]
3|1204|Loss: 1.0089205503463745:  19%|█▊        | 1203/6470 [13:47<1:06:05,  1.33it/s]
3|1204|Loss: 1.0089205503463745:  19%|█▊        | 1204/6470 [13:47<1:06:13,  1.33it/s]
3|1205|Loss: 0.9122391939163208:  19%|█▊        | 1204/6470 [13:48<1:06:13,  1.33it/s]
3|1205|Loss: 0.9122391939163208:  19%|█▊        | 1205/6470 [13:48<1:03:41,  1.38it/s]
3|1206|Loss: 0.8291030526161194:  19%|█▊        | 1205/6470 [13:48<1:03:41,  1.38it/s]
3|1206|Loss: 0.8291030526161194:  19%|█▊        | 1206/6470 [13:49<1:02:29,  1.40it/s]
3|1207|Loss: 1.004227638244629:  19%|█▊        | 1206/6470 [13:49<1:02:29,  1.40it/s] 
3|1207|Loss: 1.004227638244629:  19%|█▊        | 1207/6470 [13:49<1:03:58,  1.37it/s]
3|1208|Loss: 0.8617410063743591:  19%|█▊        | 1207/6470 [13:50<1:03:58,  1.37it/s]
3|1208|Loss: 0.8617410063743591:  19%|█▊        | 1208/6470 [13:50<1:04:21,  1.36it/s]
3|1209|Loss: 0.8298223614692688:  19%|█▊        | 1208/6470 [13:50<1:04:21,  1.36it/s]
3|1209|Loss: 0.8298223614692688:  19%|█▊        | 1209/6470 [13:51<59:41,  1.47it/s]  
3|1210|Loss: 0.8177107572555542:  19%|█▊        | 1209/6470 [13:51<59:41,  1.47it/s]
3|1210|Loss: 0.8177107572555542:  19%|█▊        | 1210/6470 [13:51<56:36,  1.55it/s]
3|1211|Loss: 0.9919503927230835:  19%|█▊        | 1210/6470 [13:52<56:36,  1.55it/s]
3|1211|Loss: 0.9919503927230835:  19%|█▊        | 1211/6470 [13:52<52:16,  1.68it/s]
3|1212|Loss: 0.7829843163490295:  19%|█▊        | 1211/6470 [13:52<52:16,  1.68it/s]
3|1212|Loss: 0.7829843163490295:  19%|█▊        | 1212/6470 [13:52<50:04,  1.75it/s]
3|1213|Loss: 0.8617285490036011:  19%|█▊        | 1212/6470 [13:53<50:04,  1.75it/s]
3|1213|Loss: 0.8617285490036011:  19%|█▊        | 1213/6470 [13:53<52:39,  1.66it/s]
3|1214|Loss: 0.8810108304023743:  19%|█▊        | 1213/6470 [13:53<52:39,  1.66it/s]
3|1214|Loss: 0.8810108304023743:  19%|█▉        | 1214/6470 [13:54<55:59,  1.56it/s]
3|1215|Loss: 0.8728121519088745:  19%|█▉        | 1214/6470 [13:54<55:59,  1.56it/s]
3|1215|Loss: 0.8728121519088745:  19%|█▉        | 1215/6470 [13:54<57:50,  1.51it/s]
3|1216|Loss: 0.7204654216766357:  19%|█▉        | 1215/6470 [13:55<57:50,  1.51it/s]
3|1216|Loss: 0.7204654216766357:  19%|█▉        | 1216/6470 [13:55<53:43,  1.63it/s]
3|1217|Loss: 0.9355737566947937:  19%|█▉        | 1216/6470 [13:55<53:43,  1.63it/s]
3|1217|Loss: 0.9355737566947937:  19%|█▉        | 1217/6470 [13:55<54:36,  1.60it/s]
3|1218|Loss: 0.8558322191238403:  19%|█▉        | 1217/6470 [13:56<54:36,  1.60it/s]
3|1218|Loss: 0.8558322191238403:  19%|█▉        | 1218/6470 [13:56<55:26,  1.58it/s]
3|1219|Loss: 1.0662529468536377:  19%|█▉        | 1218/6470 [13:57<55:26,  1.58it/s]
3|1219|Loss: 1.0662529468536377:  19%|█▉        | 1219/6470 [13:57<58:49,  1.49it/s]
3|1220|Loss: 0.9177009463310242:  19%|█▉        | 1219/6470 [13:57<58:49,  1.49it/s]
3|1220|Loss: 0.9177009463310242:  19%|█▉        | 1220/6470 [13:58<1:00:47,  1.44it/s]
3|1221|Loss: 0.8960567712783813:  19%|█▉        | 1220/6470 [13:58<1:00:47,  1.44it/s]
3|1221|Loss: 0.8960567712783813:  19%|█▉        | 1221/6470 [13:58<58:37,  1.49it/s]  
3|1222|Loss: 0.8663432002067566:  19%|█▉        | 1221/6470 [13:59<58:37,  1.49it/s]
3|1222|Loss: 0.8663432002067566:  19%|█▉        | 1222/6470 [13:59<59:46,  1.46it/s]
3|1223|Loss: 0.8381810784339905:  19%|█▉        | 1222/6470 [13:59<59:46,  1.46it/s]
3|1223|Loss: 0.8381810784339905:  19%|█▉        | 1223/6470 [14:00<1:01:38,  1.42it/s]
3|1224|Loss: 0.8253650069236755:  19%|█▉        | 1223/6470 [14:00<1:01:38,  1.42it/s]
3|1224|Loss: 0.8253650069236755:  19%|█▉        | 1224/6470 [14:00<1:00:12,  1.45it/s]
3|1225|Loss: 1.0466347932815552:  19%|█▉        | 1224/6470 [14:01<1:00:12,  1.45it/s]
3|1225|Loss: 1.0466347932815552:  19%|█▉        | 1225/6470 [14:01<58:20,  1.50it/s]  
3|1226|Loss: 0.9064982533454895:  19%|█▉        | 1225/6470 [14:01<58:20,  1.50it/s]
3|1226|Loss: 0.9064982533454895:  19%|█▉        | 1226/6470 [14:02<56:23,  1.55it/s]
3|1227|Loss: 0.9005985856056213:  19%|█▉        | 1226/6470 [14:02<56:23,  1.55it/s]
3|1227|Loss: 0.9005985856056213:  19%|█▉        | 1227/6470 [14:02<59:06,  1.48it/s]
3|1228|Loss: 0.8466435074806213:  19%|█▉        | 1227/6470 [14:03<59:06,  1.48it/s]
3|1228|Loss: 0.8466435074806213:  19%|█▉        | 1228/6470 [14:03<59:10,  1.48it/s]
3|1229|Loss: 0.9627363681793213:  19%|█▉        | 1228/6470 [14:03<59:10,  1.48it/s]
3|1229|Loss: 0.9627363681793213:  19%|█▉        | 1229/6470 [14:04<55:28,  1.57it/s]
3|1230|Loss: 0.8361779451370239:  19%|█▉        | 1229/6470 [14:04<55:28,  1.57it/s]
3|1230|Loss: 0.8361779451370239:  19%|█▉        | 1230/6470 [14:04<56:00,  1.56it/s]
3|1231|Loss: 0.8213850259780884:  19%|█▉        | 1230/6470 [14:05<56:00,  1.56it/s]
3|1231|Loss: 0.8213850259780884:  19%|█▉        | 1231/6470 [14:05<56:34,  1.54it/s]
3|1232|Loss: 0.899411678314209:  19%|█▉        | 1231/6470 [14:05<56:34,  1.54it/s] 
3|1232|Loss: 0.899411678314209:  19%|█▉        | 1232/6470 [14:06<1:00:37,  1.44it/s]
3|1233|Loss: 0.8875713348388672:  19%|█▉        | 1232/6470 [14:06<1:00:37,  1.44it/s]
3|1233|Loss: 0.8875713348388672:  19%|█▉        | 1233/6470 [14:06<1:02:37,  1.39it/s]
3|1234|Loss: 0.8966099619865417:  19%|█▉        | 1233/6470 [14:07<1:02:37,  1.39it/s]
3|1234|Loss: 0.8966099619865417:  19%|█▉        | 1234/6470 [14:07<1:03:38,  1.37it/s]
3|1235|Loss: 0.9015601277351379:  19%|█▉        | 1234/6470 [14:08<1:03:38,  1.37it/s]
3|1235|Loss: 0.9015601277351379:  19%|█▉        | 1235/6470 [14:08<1:03:19,  1.38it/s]
3|1236|Loss: 0.8954044580459595:  19%|█▉        | 1235/6470 [14:08<1:03:19,  1.38it/s]
3|1236|Loss: 0.8954044580459595:  19%|█▉        | 1236/6470 [14:09<1:02:35,  1.39it/s]
3|1237|Loss: 0.7571002840995789:  19%|█▉        | 1236/6470 [14:09<1:02:35,  1.39it/s]
3|1237|Loss: 0.7571002840995789:  19%|█▉        | 1237/6470 [14:09<58:07,  1.50it/s]  
3|1238|Loss: 1.002947211265564:  19%|█▉        | 1237/6470 [14:10<58:07,  1.50it/s] 
3|1238|Loss: 1.002947211265564:  19%|█▉        | 1238/6470 [14:10<1:01:51,  1.41it/s]
3|1239|Loss: 0.8293623328208923:  19%|█▉        | 1238/6470 [14:10<1:01:51,  1.41it/s]
3|1239|Loss: 0.8293623328208923:  19%|█▉        | 1239/6470 [14:11<1:04:44,  1.35it/s]
3|1240|Loss: 0.8156231045722961:  19%|█▉        | 1239/6470 [14:11<1:04:44,  1.35it/s]
3|1240|Loss: 0.8156231045722961:  19%|█▉        | 1240/6470 [14:12<1:04:56,  1.34it/s]
3|1241|Loss: 0.9740689992904663:  19%|█▉        | 1240/6470 [14:12<1:04:56,  1.34it/s]
3|1241|Loss: 0.9740689992904663:  19%|█▉        | 1241/6470 [14:12<1:08:30,  1.27it/s]
3|1242|Loss: 1.0041126012802124:  19%|█▉        | 1241/6470 [14:13<1:08:30,  1.27it/s]
3|1242|Loss: 1.0041126012802124:  19%|█▉        | 1242/6470 [14:13<1:07:32,  1.29it/s]
3|1243|Loss: 0.7408050894737244:  19%|█▉        | 1242/6470 [14:14<1:07:32,  1.29it/s]
3|1243|Loss: 0.7408050894737244:  19%|█▉        | 1243/6470 [14:14<1:02:24,  1.40it/s]
3|1244|Loss: 0.9210546016693115:  19%|█▉        | 1243/6470 [14:14<1:02:24,  1.40it/s]
3|1244|Loss: 0.9210546016693115:  19%|█▉        | 1244/6470 [14:15<1:03:26,  1.37it/s]
3|1245|Loss: 0.8692333102226257:  19%|█▉        | 1244/6470 [14:15<1:03:26,  1.37it/s]
3|1245|Loss: 0.8692333102226257:  19%|█▉        | 1245/6470 [14:15<1:04:23,  1.35it/s]
3|1246|Loss: 0.686972975730896:  19%|█▉        | 1245/6470 [14:16<1:04:23,  1.35it/s] 
3|1246|Loss: 0.686972975730896:  19%|█▉        | 1246/6470 [14:16<1:05:47,  1.32it/s]
3|1247|Loss: 0.7579892873764038:  19%|█▉        | 1246/6470 [14:16<1:05:47,  1.32it/s]
3|1247|Loss: 0.7579892873764038:  19%|█▉        | 1247/6470 [14:17<1:02:14,  1.40it/s]
3|1248|Loss: 0.766468346118927:  19%|█▉        | 1247/6470 [14:17<1:02:14,  1.40it/s] 
3|1248|Loss: 0.766468346118927:  19%|█▉        | 1248/6470 [14:17<1:01:52,  1.41it/s]
3|1249|Loss: 1.0296063423156738:  19%|█▉        | 1248/6470 [14:18<1:01:52,  1.41it/s]
3|1249|Loss: 1.0296063423156738:  19%|█▉        | 1249/6470 [14:18<1:02:47,  1.39it/s]
3|1250|Loss: 0.7173550128936768:  19%|█▉        | 1249/6470 [14:18<1:02:47,  1.39it/s]
3|1250|Loss: 0.7173550128936768:  19%|█▉        | 1250/6470 [14:19<55:24,  1.57it/s]  
3|1251|Loss: 0.8168466091156006:  19%|█▉        | 1250/6470 [14:19<55:24,  1.57it/s]
3|1251|Loss: 0.8168466091156006:  19%|█▉        | 1251/6470 [14:19<52:42,  1.65it/s]
3|1252|Loss: 0.8144696354866028:  19%|█▉        | 1251/6470 [14:19<52:42,  1.65it/s]
3|1252|Loss: 0.8144696354866028:  19%|█▉        | 1252/6470 [14:20<55:25,  1.57it/s]
3|1253|Loss: 0.9974958300590515:  19%|█▉        | 1252/6470 [14:20<55:25,  1.57it/s]
3|1253|Loss: 0.9974958300590515:  19%|█▉        | 1253/6470 [14:20<53:41,  1.62it/s]
3|1254|Loss: 0.8235817551612854:  19%|█▉        | 1253/6470 [14:21<53:41,  1.62it/s]
3|1254|Loss: 0.8235817551612854:  19%|█▉        | 1254/6470 [14:21<53:44,  1.62it/s]
3|1255|Loss: 0.9652740955352783:  19%|█▉        | 1254/6470 [14:21<53:44,  1.62it/s]
3|1255|Loss: 0.9652740955352783:  19%|█▉        | 1255/6470 [14:22<56:48,  1.53it/s]
3|1256|Loss: 0.8417583703994751:  19%|█▉        | 1255/6470 [14:22<56:48,  1.53it/s]
3|1256|Loss: 0.8417583703994751:  19%|█▉        | 1256/6470 [14:22<58:23,  1.49it/s]
3|1257|Loss: 0.73465895652771:  19%|█▉        | 1256/6470 [14:23<58:23,  1.49it/s]  
3|1257|Loss: 0.73465895652771:  19%|█▉        | 1257/6470 [14:23<59:16,  1.47it/s]
3|1258|Loss: 0.7713434100151062:  19%|█▉        | 1257/6470 [14:24<59:16,  1.47it/s]
3|1258|Loss: 0.7713434100151062:  19%|█▉        | 1258/6470 [14:24<1:03:01,  1.38it/s]
3|1259|Loss: 0.8561365008354187:  19%|█▉        | 1258/6470 [14:24<1:03:01,  1.38it/s]
3|1259|Loss: 0.8561365008354187:  19%|█▉        | 1259/6470 [14:25<1:04:03,  1.36it/s]
3|1260|Loss: 0.7041481137275696:  19%|█▉        | 1259/6470 [14:25<1:04:03,  1.36it/s]
3|1260|Loss: 0.7041481137275696:  19%|█▉        | 1260/6470 [14:25<1:03:16,  1.37it/s]
3|1261|Loss: 0.8987756967544556:  19%|█▉        | 1260/6470 [14:26<1:03:16,  1.37it/s]
3|1261|Loss: 0.8987756967544556:  19%|█▉        | 1261/6470 [14:26<59:54,  1.45it/s]  
3|1262|Loss: 0.8922383189201355:  19%|█▉        | 1261/6470 [14:26<59:54,  1.45it/s]
3|1262|Loss: 0.8922383189201355:  20%|█▉        | 1262/6470 [14:27<56:26,  1.54it/s]
3|1263|Loss: 0.8375176191329956:  20%|█▉        | 1262/6470 [14:27<56:26,  1.54it/s]
3|1263|Loss: 0.8375176191329956:  20%|█▉        | 1263/6470 [14:27<56:59,  1.52it/s]
3|1264|Loss: 0.8216143846511841:  20%|█▉        | 1263/6470 [14:28<56:59,  1.52it/s]
3|1264|Loss: 0.8216143846511841:  20%|█▉        | 1264/6470 [14:28<56:36,  1.53it/s]
3|1265|Loss: 0.8513531684875488:  20%|█▉        | 1264/6470 [14:28<56:36,  1.53it/s]
3|1265|Loss: 0.8513531684875488:  20%|█▉        | 1265/6470 [14:29<59:27,  1.46it/s]
3|1266|Loss: 1.0702674388885498:  20%|█▉        | 1265/6470 [14:29<59:27,  1.46it/s]
3|1266|Loss: 1.0702674388885498:  20%|█▉        | 1266/6470 [14:30<1:06:09,  1.31it/s]
3|1267|Loss: 0.8439083695411682:  20%|█▉        | 1266/6470 [14:30<1:06:09,  1.31it/s]
3|1267|Loss: 0.8439083695411682:  20%|█▉        | 1267/6470 [14:30<1:06:25,  1.31it/s]
3|1268|Loss: 0.866612434387207:  20%|█▉        | 1267/6470 [14:31<1:06:25,  1.31it/s] 
3|1268|Loss: 0.866612434387207:  20%|█▉        | 1268/6470 [14:31<1:00:40,  1.43it/s]
3|1269|Loss: 1.1049684286117554:  20%|█▉        | 1268/6470 [14:31<1:00:40,  1.43it/s]
3|1269|Loss: 1.1049684286117554:  20%|█▉        | 1269/6470 [14:32<57:48,  1.50it/s]  
3|1270|Loss: 0.8547587990760803:  20%|█▉        | 1269/6470 [14:32<57:48,  1.50it/s]
3|1270|Loss: 0.8547587990760803:  20%|█▉        | 1270/6470 [14:32<59:21,  1.46it/s]
3|1271|Loss: 0.9590672254562378:  20%|█▉        | 1270/6470 [14:33<59:21,  1.46it/s]
3|1271|Loss: 0.9590672254562378:  20%|█▉        | 1271/6470 [14:33<58:33,  1.48it/s]
3|1272|Loss: 0.8538722991943359:  20%|█▉        | 1271/6470 [14:33<58:33,  1.48it/s]
3|1272|Loss: 0.8538722991943359:  20%|█▉        | 1272/6470 [14:34<1:00:44,  1.43it/s]
3|1273|Loss: 0.7864603400230408:  20%|█▉        | 1272/6470 [14:34<1:00:44,  1.43it/s]
3|1273|Loss: 0.7864603400230408:  20%|█▉        | 1273/6470 [14:34<56:10,  1.54it/s]  
3|1274|Loss: 0.7622052431106567:  20%|█▉        | 1273/6470 [14:35<56:10,  1.54it/s]
3|1274|Loss: 0.7622052431106567:  20%|█▉        | 1274/6470 [14:35<54:51,  1.58it/s]
3|1275|Loss: 0.9818453788757324:  20%|█▉        | 1274/6470 [14:35<54:51,  1.58it/s]
3|1275|Loss: 0.9818453788757324:  20%|█▉        | 1275/6470 [14:35<53:57,  1.60it/s]
3|1276|Loss: 0.842660129070282:  20%|█▉        | 1275/6470 [14:36<53:57,  1.60it/s] 
3|1276|Loss: 0.842660129070282:  20%|█▉        | 1276/6470 [14:36<56:01,  1.55it/s]
3|1277|Loss: 0.8441203832626343:  20%|█▉        | 1276/6470 [14:37<56:01,  1.55it/s]
3|1277|Loss: 0.8441203832626343:  20%|█▉        | 1277/6470 [14:37<1:01:20,  1.41it/s]
3|1278|Loss: 1.0855032205581665:  20%|█▉        | 1277/6470 [14:37<1:01:20,  1.41it/s]
3|1278|Loss: 1.0855032205581665:  20%|█▉        | 1278/6470 [14:38<1:05:25,  1.32it/s]
3|1279|Loss: 0.8291981816291809:  20%|█▉        | 1278/6470 [14:38<1:05:25,  1.32it/s]
3|1279|Loss: 0.8291981816291809:  20%|█▉        | 1279/6470 [14:38<1:00:52,  1.42it/s]
3|1280|Loss: 0.9078978300094604:  20%|█▉        | 1279/6470 [14:39<1:00:52,  1.42it/s]
3|1280|Loss: 0.9078978300094604:  20%|█▉        | 1280/6470 [14:39<57:12,  1.51it/s]  
3|1281|Loss: 0.7829623222351074:  20%|█▉        | 1280/6470 [14:39<57:12,  1.51it/s]
3|1281|Loss: 0.7829623222351074:  20%|█▉        | 1281/6470 [14:40<57:46,  1.50it/s]
3|1282|Loss: 0.8869338631629944:  20%|█▉        | 1281/6470 [14:40<57:46,  1.50it/s]
3|1282|Loss: 0.8869338631629944:  20%|█▉        | 1282/6470 [14:40<56:51,  1.52it/s]
3|1283|Loss: 0.9319120645523071:  20%|█▉        | 1282/6470 [14:41<56:51,  1.52it/s]
3|1283|Loss: 0.9319120645523071:  20%|█▉        | 1283/6470 [14:41<55:43,  1.55it/s]
3|1284|Loss: 0.8423905372619629:  20%|█▉        | 1283/6470 [14:41<55:43,  1.55it/s]
3|1284|Loss: 0.8423905372619629:  20%|█▉        | 1284/6470 [14:42<57:22,  1.51it/s]
3|1285|Loss: 0.913824737071991:  20%|█▉        | 1284/6470 [14:42<57:22,  1.51it/s] 
3|1285|Loss: 0.913824737071991:  20%|█▉        | 1285/6470 [14:42<1:00:39,  1.42it/s]
3|1286|Loss: 0.8010495901107788:  20%|█▉        | 1285/6470 [14:43<1:00:39,  1.42it/s]
3|1286|Loss: 0.8010495901107788:  20%|█▉        | 1286/6470 [14:43<59:54,  1.44it/s]  
3|1287|Loss: 0.8174887299537659:  20%|█▉        | 1286/6470 [14:43<59:54,  1.44it/s]
3|1287|Loss: 0.8174887299537659:  20%|█▉        | 1287/6470 [14:44<59:51,  1.44it/s]
3|1288|Loss: 0.7879186868667603:  20%|█▉        | 1287/6470 [14:44<59:51,  1.44it/s]
3|1288|Loss: 0.7879186868667603:  20%|█▉        | 1288/6470 [14:45<1:04:15,  1.34it/s]
3|1289|Loss: 0.7544454336166382:  20%|█▉        | 1288/6470 [14:45<1:04:15,  1.34it/s]
3|1289|Loss: 0.7544454336166382:  20%|█▉        | 1289/6470 [14:45<1:01:54,  1.39it/s]
3|1290|Loss: 0.9372480511665344:  20%|█▉        | 1289/6470 [14:46<1:01:54,  1.39it/s]
3|1290|Loss: 0.9372480511665344:  20%|█▉        | 1290/6470 [14:46<58:32,  1.47it/s]  
3|1291|Loss: 0.7598835229873657:  20%|█▉        | 1290/6470 [14:46<58:32,  1.47it/s]
3|1291|Loss: 0.7598835229873657:  20%|█▉        | 1291/6470 [14:46<56:42,  1.52it/s]
3|1292|Loss: 0.8179275393486023:  20%|█▉        | 1291/6470 [14:47<56:42,  1.52it/s]
3|1292|Loss: 0.8179275393486023:  20%|█▉        | 1292/6470 [14:47<1:00:54,  1.42it/s]
3|1293|Loss: 0.8590819835662842:  20%|█▉        | 1292/6470 [14:48<1:00:54,  1.42it/s]
3|1293|Loss: 0.8590819835662842:  20%|█▉        | 1293/6470 [14:48<1:00:43,  1.42it/s]
3|1294|Loss: 0.9712599515914917:  20%|█▉        | 1293/6470 [14:48<1:00:43,  1.42it/s]
3|1294|Loss: 0.9712599515914917:  20%|██        | 1294/6470 [14:49<1:01:20,  1.41it/s]
3|1295|Loss: 0.7283490896224976:  20%|██        | 1294/6470 [14:49<1:01:20,  1.41it/s]
3|1295|Loss: 0.7283490896224976:  20%|██        | 1295/6470 [14:49<58:46,  1.47it/s]  
3|1296|Loss: 0.9070839285850525:  20%|██        | 1295/6470 [14:50<58:46,  1.47it/s]
3|1296|Loss: 0.9070839285850525:  20%|██        | 1296/6470 [14:50<59:31,  1.45it/s]
3|1297|Loss: 1.022000789642334:  20%|██        | 1296/6470 [14:50<59:31,  1.45it/s] 
3|1297|Loss: 1.022000789642334:  20%|██        | 1297/6470 [14:51<55:05,  1.56it/s]
3|1298|Loss: 0.9476917386054993:  20%|██        | 1297/6470 [14:51<55:05,  1.56it/s]
3|1298|Loss: 0.9476917386054993:  20%|██        | 1298/6470 [14:51<55:49,  1.54it/s]
3|1299|Loss: 0.7145712971687317:  20%|██        | 1298/6470 [14:52<55:49,  1.54it/s]
3|1299|Loss: 0.7145712971687317:  20%|██        | 1299/6470 [14:52<57:24,  1.50it/s]
3|1300|Loss: 0.9033957719802856:  20%|██        | 1299/6470 [14:52<57:24,  1.50it/s]
3|1300|Loss: 0.9033957719802856:  20%|██        | 1300/6470 [14:53<1:02:28,  1.38it/s]
3|1301|Loss: 0.7515667080879211:  20%|██        | 1300/6470 [14:53<1:02:28,  1.38it/s]
3|1301|Loss: 0.7515667080879211:  20%|██        | 1301/6470 [14:53<57:51,  1.49it/s]  
3|1302|Loss: 0.9471871256828308:  20%|██        | 1301/6470 [14:54<57:51,  1.49it/s]
3|1302|Loss: 0.9471871256828308:  20%|██        | 1302/6470 [14:54<57:11,  1.51it/s]
3|1303|Loss: 0.9720969796180725:  20%|██        | 1302/6470 [14:54<57:11,  1.51it/s]
3|1303|Loss: 0.9720969796180725:  20%|██        | 1303/6470 [14:55<59:06,  1.46it/s]
3|1304|Loss: 1.0385000705718994:  20%|██        | 1303/6470 [14:55<59:06,  1.46it/s]
3|1304|Loss: 1.0385000705718994:  20%|██        | 1304/6470 [14:55<58:35,  1.47it/s]
3|1305|Loss: 0.8131083846092224:  20%|██        | 1304/6470 [14:56<58:35,  1.47it/s]
3|1305|Loss: 0.8131083846092224:  20%|██        | 1305/6470 [14:56<1:03:12,  1.36it/s]
3|1306|Loss: 0.9334602355957031:  20%|██        | 1305/6470 [14:57<1:03:12,  1.36it/s]
3|1306|Loss: 0.9334602355957031:  20%|██        | 1306/6470 [14:57<1:06:05,  1.30it/s]
3|1307|Loss: 0.7558505535125732:  20%|██        | 1306/6470 [14:58<1:06:05,  1.30it/s]
3|1307|Loss: 0.7558505535125732:  20%|██        | 1307/6470 [14:58<1:03:26,  1.36it/s]
3|1308|Loss: 0.7997680306434631:  20%|██        | 1307/6470 [14:58<1:03:26,  1.36it/s]
3|1308|Loss: 0.7997680306434631:  20%|██        | 1308/6470 [14:58<57:39,  1.49it/s]  
3|1309|Loss: 0.6237415671348572:  20%|██        | 1308/6470 [14:59<57:39,  1.49it/s]
3|1309|Loss: 0.6237415671348572:  20%|██        | 1309/6470 [14:59<1:01:43,  1.39it/s]
3|1310|Loss: 0.8415198922157288:  20%|██        | 1309/6470 [15:00<1:01:43,  1.39it/s]
3|1310|Loss: 0.8415198922157288:  20%|██        | 1310/6470 [15:00<1:00:46,  1.42it/s]
3|1311|Loss: 0.8241000175476074:  20%|██        | 1310/6470 [15:00<1:00:46,  1.42it/s]
3|1311|Loss: 0.8241000175476074:  20%|██        | 1311/6470 [15:01<1:02:05,  1.38it/s]
3|1312|Loss: 0.7260864973068237:  20%|██        | 1311/6470 [15:01<1:02:05,  1.38it/s]
3|1312|Loss: 0.7260864973068237:  20%|██        | 1312/6470 [15:01<1:03:55,  1.34it/s]
3|1313|Loss: 0.7140482068061829:  20%|██        | 1312/6470 [15:02<1:03:55,  1.34it/s]
3|1313|Loss: 0.7140482068061829:  20%|██        | 1313/6470 [15:02<1:06:35,  1.29it/s]
3|1314|Loss: 0.7648602724075317:  20%|██        | 1313/6470 [15:03<1:06:35,  1.29it/s]
3|1314|Loss: 0.7648602724075317:  20%|██        | 1314/6470 [15:03<1:02:57,  1.36it/s]
3|1315|Loss: 0.8590183854103088:  20%|██        | 1314/6470 [15:03<1:02:57,  1.36it/s]
3|1315|Loss: 0.8590183854103088:  20%|██        | 1315/6470 [15:04<1:02:18,  1.38it/s]
3|1316|Loss: 0.8570465445518494:  20%|██        | 1315/6470 [15:04<1:02:18,  1.38it/s]
3|1316|Loss: 0.8570465445518494:  20%|██        | 1316/6470 [15:04<1:00:44,  1.41it/s]
3|1317|Loss: 0.7615188360214233:  20%|██        | 1316/6470 [15:05<1:00:44,  1.41it/s]
3|1317|Loss: 0.7615188360214233:  20%|██        | 1317/6470 [15:05<1:02:24,  1.38it/s]
3|1318|Loss: 0.691153883934021:  20%|██        | 1317/6470 [15:05<1:02:24,  1.38it/s] 
3|1318|Loss: 0.691153883934021:  20%|██        | 1318/6470 [15:06<1:02:54,  1.36it/s]
3|1319|Loss: 0.987244188785553:  20%|██        | 1318/6470 [15:06<1:02:54,  1.36it/s]
3|1319|Loss: 0.987244188785553:  20%|██        | 1319/6470 [15:06<59:10,  1.45it/s]  
3|1320|Loss: 0.7987647652626038:  20%|██        | 1319/6470 [15:07<59:10,  1.45it/s]
3|1320|Loss: 0.7987647652626038:  20%|██        | 1320/6470 [15:07<57:35,  1.49it/s]
3|1321|Loss: 0.7903789281845093:  20%|██        | 1320/6470 [15:07<57:35,  1.49it/s]
3|1321|Loss: 0.7903789281845093:  20%|██        | 1321/6470 [15:08<56:33,  1.52it/s]
3|1322|Loss: 0.8545008897781372:  20%|██        | 1321/6470 [15:08<56:33,  1.52it/s]
3|1322|Loss: 0.8545008897781372:  20%|██        | 1322/6470 [15:08<56:07,  1.53it/s]
3|1323|Loss: 0.8230485320091248:  20%|██        | 1322/6470 [15:09<56:07,  1.53it/s]
3|1323|Loss: 0.8230485320091248:  20%|██        | 1323/6470 [15:09<54:44,  1.57it/s]
3|1324|Loss: 0.8652053475379944:  20%|██        | 1323/6470 [15:09<54:44,  1.57it/s]
3|1324|Loss: 0.8652053475379944:  20%|██        | 1324/6470 [15:10<56:33,  1.52it/s]
3|1325|Loss: 0.7930508255958557:  20%|██        | 1324/6470 [15:10<56:33,  1.52it/s]
3|1325|Loss: 0.7930508255958557:  20%|██        | 1325/6470 [15:10<56:39,  1.51it/s]
3|1326|Loss: 0.7316434979438782:  20%|██        | 1325/6470 [15:11<56:39,  1.51it/s]
3|1326|Loss: 0.7316434979438782:  20%|██        | 1326/6470 [15:11<58:26,  1.47it/s]
3|1327|Loss: 0.9241202473640442:  20%|██        | 1326/6470 [15:11<58:26,  1.47it/s]
3|1327|Loss: 0.9241202473640442:  21%|██        | 1327/6470 [15:12<59:21,  1.44it/s]
3|1328|Loss: 0.8868082761764526:  21%|██        | 1327/6470 [15:12<59:21,  1.44it/s]
3|1328|Loss: 0.8868082761764526:  21%|██        | 1328/6470 [15:12<54:26,  1.57it/s]
3|1329|Loss: 0.8226111531257629:  21%|██        | 1328/6470 [15:12<54:26,  1.57it/s]
3|1329|Loss: 0.8226111531257629:  21%|██        | 1329/6470 [15:13<53:18,  1.61it/s]
3|1330|Loss: 1.1149688959121704:  21%|██        | 1329/6470 [15:13<53:18,  1.61it/s]
3|1330|Loss: 1.1149688959121704:  21%|██        | 1330/6470 [15:14<59:16,  1.45it/s]
3|1331|Loss: 0.6912990212440491:  21%|██        | 1330/6470 [15:14<59:16,  1.45it/s]
3|1331|Loss: 0.6912990212440491:  21%|██        | 1331/6470 [15:14<1:01:28,  1.39it/s]
3|1332|Loss: 0.9215613007545471:  21%|██        | 1331/6470 [15:15<1:01:28,  1.39it/s]
3|1332|Loss: 0.9215613007545471:  21%|██        | 1332/6470 [15:15<1:04:42,  1.32it/s]
3|1333|Loss: 0.9571661949157715:  21%|██        | 1332/6470 [15:16<1:04:42,  1.32it/s]
3|1333|Loss: 0.9571661949157715:  21%|██        | 1333/6470 [15:16<1:05:21,  1.31it/s]
3|1334|Loss: 0.8720707297325134:  21%|██        | 1333/6470 [15:16<1:05:21,  1.31it/s]
3|1334|Loss: 0.8720707297325134:  21%|██        | 1334/6470 [15:17<1:00:11,  1.42it/s]
3|1335|Loss: 0.9851162433624268:  21%|██        | 1334/6470 [15:17<1:00:11,  1.42it/s]
3|1335|Loss: 0.9851162433624268:  21%|██        | 1335/6470 [15:17<55:08,  1.55it/s]  
3|1336|Loss: 0.7826212048530579:  21%|██        | 1335/6470 [15:17<55:08,  1.55it/s]
3|1336|Loss: 0.7826212048530579:  21%|██        | 1336/6470 [15:18<54:13,  1.58it/s]
3|1337|Loss: 0.8129797577857971:  21%|██        | 1336/6470 [15:18<54:13,  1.58it/s]
3|1337|Loss: 0.8129797577857971:  21%|██        | 1337/6470 [15:18<52:34,  1.63it/s]
3|1338|Loss: 0.7663798332214355:  21%|██        | 1337/6470 [15:19<52:34,  1.63it/s]
3|1338|Loss: 0.7663798332214355:  21%|██        | 1338/6470 [15:19<56:35,  1.51it/s]
3|1339|Loss: 0.9859236478805542:  21%|██        | 1338/6470 [15:19<56:35,  1.51it/s]
3|1339|Loss: 0.9859236478805542:  21%|██        | 1339/6470 [15:20<56:31,  1.51it/s]
3|1340|Loss: 0.8191356062889099:  21%|██        | 1339/6470 [15:20<56:31,  1.51it/s]
3|1340|Loss: 0.8191356062889099:  21%|██        | 1340/6470 [15:20<57:50,  1.48it/s]
3|1341|Loss: 0.7614104747772217:  21%|██        | 1340/6470 [15:21<57:50,  1.48it/s]
3|1341|Loss: 0.7614104747772217:  21%|██        | 1341/6470 [15:21<59:27,  1.44it/s]
3|1342|Loss: 0.8478876352310181:  21%|██        | 1341/6470 [15:22<59:27,  1.44it/s]
3|1342|Loss: 0.8478876352310181:  21%|██        | 1342/6470 [15:22<58:42,  1.46it/s]
3|1343|Loss: 0.7313414812088013:  21%|██        | 1342/6470 [15:22<58:42,  1.46it/s]
3|1343|Loss: 0.7313414812088013:  21%|██        | 1343/6470 [15:23<59:26,  1.44it/s]
3|1344|Loss: 1.0466768741607666:  21%|██        | 1343/6470 [15:23<59:26,  1.44it/s]
3|1344|Loss: 1.0466768741607666:  21%|██        | 1344/6470 [15:23<1:04:12,  1.33it/s]
3|1345|Loss: 0.8065432906150818:  21%|██        | 1344/6470 [15:24<1:04:12,  1.33it/s]
3|1345|Loss: 0.8065432906150818:  21%|██        | 1345/6470 [15:24<59:49,  1.43it/s]  
3|1346|Loss: 0.8535953164100647:  21%|██        | 1345/6470 [15:24<59:49,  1.43it/s]
3|1346|Loss: 0.8535953164100647:  21%|██        | 1346/6470 [15:25<57:32,  1.48it/s]
3|1347|Loss: 0.8337360620498657:  21%|██        | 1346/6470 [15:25<57:32,  1.48it/s]
3|1347|Loss: 0.8337360620498657:  21%|██        | 1347/6470 [15:25<56:49,  1.50it/s]
3|1348|Loss: 0.7460258603096008:  21%|██        | 1347/6470 [15:26<56:49,  1.50it/s]
3|1348|Loss: 0.7460258603096008:  21%|██        | 1348/6470 [15:26<58:06,  1.47it/s]
3|1349|Loss: 0.816616952419281:  21%|██        | 1348/6470 [15:26<58:06,  1.47it/s] 
3|1349|Loss: 0.816616952419281:  21%|██        | 1349/6470 [15:27<58:44,  1.45it/s]
3|1350|Loss: 0.9704449772834778:  21%|██        | 1349/6470 [15:27<58:44,  1.45it/s]
3|1350|Loss: 0.9704449772834778:  21%|██        | 1350/6470 [15:27<57:24,  1.49it/s]
3|1351|Loss: 0.9486892819404602:  21%|██        | 1350/6470 [15:28<57:24,  1.49it/s]
3|1351|Loss: 0.9486892819404602:  21%|██        | 1351/6470 [15:28<55:22,  1.54it/s]
3|1352|Loss: 0.9129199981689453:  21%|██        | 1351/6470 [15:28<55:22,  1.54it/s]
3|1352|Loss: 0.9129199981689453:  21%|██        | 1352/6470 [15:29<56:49,  1.50it/s]
3|1353|Loss: 0.8774897456169128:  21%|██        | 1352/6470 [15:29<56:49,  1.50it/s]
3|1353|Loss: 0.8774897456169128:  21%|██        | 1353/6470 [15:29<58:42,  1.45it/s]
3|1354|Loss: 0.8758169412612915:  21%|██        | 1353/6470 [15:30<58:42,  1.45it/s]
3|1354|Loss: 0.8758169412612915:  21%|██        | 1354/6470 [15:30<1:01:56,  1.38it/s]
3|1355|Loss: 0.8706028461456299:  21%|██        | 1354/6470 [15:31<1:01:56,  1.38it/s]
3|1355|Loss: 0.8706028461456299:  21%|██        | 1355/6470 [15:31<1:00:16,  1.41it/s]
3|1356|Loss: 0.9600441455841064:  21%|██        | 1355/6470 [15:31<1:00:16,  1.41it/s]
3|1356|Loss: 0.9600441455841064:  21%|██        | 1356/6470 [15:31<58:18,  1.46it/s]  
3|1357|Loss: 0.7787889838218689:  21%|██        | 1356/6470 [15:32<58:18,  1.46it/s]
3|1357|Loss: 0.7787889838218689:  21%|██        | 1357/6470 [15:32<55:52,  1.53it/s]
3|1358|Loss: 0.7364792227745056:  21%|██        | 1357/6470 [15:32<55:52,  1.53it/s]
3|1358|Loss: 0.7364792227745056:  21%|██        | 1358/6470 [15:33<52:28,  1.62it/s]
3|1359|Loss: 1.1443243026733398:  21%|██        | 1358/6470 [15:33<52:28,  1.62it/s]
3|1359|Loss: 1.1443243026733398:  21%|██        | 1359/6470 [15:34<1:02:07,  1.37it/s]
3|1360|Loss: 0.8992953300476074:  21%|██        | 1359/6470 [15:34<1:02:07,  1.37it/s]
3|1360|Loss: 0.8992953300476074:  21%|██        | 1360/6470 [15:34<1:04:44,  1.32it/s]
3|1361|Loss: 0.7555379867553711:  21%|██        | 1360/6470 [15:35<1:04:44,  1.32it/s]
3|1361|Loss: 0.7555379867553711:  21%|██        | 1361/6470 [15:35<1:03:40,  1.34it/s]
3|1362|Loss: 0.9317674040794373:  21%|██        | 1361/6470 [15:35<1:03:40,  1.34it/s]
3|1362|Loss: 0.9317674040794373:  21%|██        | 1362/6470 [15:36<1:00:43,  1.40it/s]
3|1363|Loss: 0.7993144989013672:  21%|██        | 1362/6470 [15:36<1:00:43,  1.40it/s]
3|1363|Loss: 0.7993144989013672:  21%|██        | 1363/6470 [15:37<1:03:01,  1.35it/s]
3|1364|Loss: 0.844017505645752:  21%|██        | 1363/6470 [15:37<1:03:01,  1.35it/s] 
3|1364|Loss: 0.844017505645752:  21%|██        | 1364/6470 [15:37<1:03:45,  1.33it/s]
3|1365|Loss: 0.8584334850311279:  21%|██        | 1364/6470 [15:38<1:03:45,  1.33it/s]
3|1365|Loss: 0.8584334850311279:  21%|██        | 1365/6470 [15:38<1:06:36,  1.28it/s]
3|1366|Loss: 0.834860622882843:  21%|██        | 1365/6470 [15:39<1:06:36,  1.28it/s] 
3|1366|Loss: 0.834860622882843:  21%|██        | 1366/6470 [15:39<1:04:09,  1.33it/s]
3|1367|Loss: 0.706415593624115:  21%|██        | 1366/6470 [15:39<1:04:09,  1.33it/s]
3|1367|Loss: 0.706415593624115:  21%|██        | 1367/6470 [15:40<1:07:33,  1.26it/s]
3|1368|Loss: 0.9474954009056091:  21%|██        | 1367/6470 [15:40<1:07:33,  1.26it/s]
3|1368|Loss: 0.9474954009056091:  21%|██        | 1368/6470 [15:40<1:02:42,  1.36it/s]
3|1369|Loss: 0.7848439812660217:  21%|██        | 1368/6470 [15:41<1:02:42,  1.36it/s]
3|1369|Loss: 0.7848439812660217:  21%|██        | 1369/6470 [15:41<59:33,  1.43it/s]  
3|1370|Loss: 0.8987497091293335:  21%|██        | 1369/6470 [15:41<59:33,  1.43it/s]
3|1370|Loss: 0.8987497091293335:  21%|██        | 1370/6470 [15:42<58:08,  1.46it/s]
3|1371|Loss: 1.0071698427200317:  21%|██        | 1370/6470 [15:42<58:08,  1.46it/s]
3|1371|Loss: 1.0071698427200317:  21%|██        | 1371/6470 [15:42<59:01,  1.44it/s]
3|1372|Loss: 0.7726403474807739:  21%|██        | 1371/6470 [15:43<59:01,  1.44it/s]
3|1372|Loss: 0.7726403474807739:  21%|██        | 1372/6470 [15:43<57:38,  1.47it/s]
3|1373|Loss: 0.8497072458267212:  21%|██        | 1372/6470 [15:43<57:38,  1.47it/s]
3|1373|Loss: 0.8497072458267212:  21%|██        | 1373/6470 [15:44<56:00,  1.52it/s]
3|1374|Loss: 0.8320311903953552:  21%|██        | 1373/6470 [15:44<56:00,  1.52it/s]
3|1374|Loss: 0.8320311903953552:  21%|██        | 1374/6470 [15:44<54:22,  1.56it/s]
3|1375|Loss: 0.8482632040977478:  21%|██        | 1374/6470 [15:45<54:22,  1.56it/s]
3|1375|Loss: 0.8482632040977478:  21%|██▏       | 1375/6470 [15:45<54:06,  1.57it/s]
3|1376|Loss: 0.8143154978752136:  21%|██▏       | 1375/6470 [15:45<54:06,  1.57it/s]
3|1376|Loss: 0.8143154978752136:  21%|██▏       | 1376/6470 [15:45<54:16,  1.56it/s]
3|1377|Loss: 0.9264904856681824:  21%|██▏       | 1376/6470 [15:46<54:16,  1.56it/s]
3|1377|Loss: 0.9264904856681824:  21%|██▏       | 1377/6470 [15:46<55:24,  1.53it/s]
3|1378|Loss: 0.7935388088226318:  21%|██▏       | 1377/6470 [15:47<55:24,  1.53it/s]
3|1378|Loss: 0.7935388088226318:  21%|██▏       | 1378/6470 [15:47<55:20,  1.53it/s]
3|1379|Loss: 0.7296144366264343:  21%|██▏       | 1378/6470 [15:47<55:20,  1.53it/s]
3|1379|Loss: 0.7296144366264343:  21%|██▏       | 1379/6470 [15:48<57:12,  1.48it/s]
3|1380|Loss: 0.8341935276985168:  21%|██▏       | 1379/6470 [15:48<57:12,  1.48it/s]
3|1380|Loss: 0.8341935276985168:  21%|██▏       | 1380/6470 [15:48<58:15,  1.46it/s]
3|1381|Loss: 0.7466427683830261:  21%|██▏       | 1380/6470 [15:49<58:15,  1.46it/s]
3|1381|Loss: 0.7466427683830261:  21%|██▏       | 1381/6470 [15:49<1:01:09,  1.39it/s]
3|1382|Loss: 0.9873526692390442:  21%|██▏       | 1381/6470 [15:49<1:01:09,  1.39it/s]
3|1382|Loss: 0.9873526692390442:  21%|██▏       | 1382/6470 [15:50<1:01:56,  1.37it/s]
3|1383|Loss: 0.821454644203186:  21%|██▏       | 1382/6470 [15:50<1:01:56,  1.37it/s] 
3|1383|Loss: 0.821454644203186:  21%|██▏       | 1383/6470 [15:51<1:03:55,  1.33it/s]
3|1384|Loss: 0.7793881893157959:  21%|██▏       | 1383/6470 [15:51<1:03:55,  1.33it/s]
3|1384|Loss: 0.7793881893157959:  21%|██▏       | 1384/6470 [15:51<1:01:25,  1.38it/s]
3|1385|Loss: 0.7849311828613281:  21%|██▏       | 1384/6470 [15:52<1:01:25,  1.38it/s]
3|1385|Loss: 0.7849311828613281:  21%|██▏       | 1385/6470 [15:52<1:00:29,  1.40it/s]
3|1386|Loss: 0.899914562702179:  21%|██▏       | 1385/6470 [15:52<1:00:29,  1.40it/s] 
3|1386|Loss: 0.899914562702179:  21%|██▏       | 1386/6470 [15:53<1:00:46,  1.39it/s]
3|1387|Loss: 0.8947423696517944:  21%|██▏       | 1386/6470 [15:53<1:00:46,  1.39it/s]
3|1387|Loss: 0.8947423696517944:  21%|██▏       | 1387/6470 [15:53<1:01:29,  1.38it/s]
3|1388|Loss: 0.8339555859565735:  21%|██▏       | 1387/6470 [15:54<1:01:29,  1.38it/s]
3|1388|Loss: 0.8339555859565735:  21%|██▏       | 1388/6470 [15:54<1:02:03,  1.36it/s]
3|1389|Loss: 0.7556814551353455:  21%|██▏       | 1388/6470 [15:55<1:02:03,  1.36it/s]
3|1389|Loss: 0.7556814551353455:  21%|██▏       | 1389/6470 [15:55<58:59,  1.44it/s]  
3|1390|Loss: 0.9365798830986023:  21%|██▏       | 1389/6470 [15:55<58:59,  1.44it/s]
3|1390|Loss: 0.9365798830986023:  21%|██▏       | 1390/6470 [15:55<53:21,  1.59it/s]
3|1391|Loss: 0.9749997854232788:  21%|██▏       | 1390/6470 [15:56<53:21,  1.59it/s]
3|1391|Loss: 0.9749997854232788:  21%|██▏       | 1391/6470 [15:56<54:38,  1.55it/s]
3|1392|Loss: 0.7670912742614746:  21%|██▏       | 1391/6470 [15:56<54:38,  1.55it/s]
3|1392|Loss: 0.7670912742614746:  22%|██▏       | 1392/6470 [15:56<49:10,  1.72it/s]
3|1393|Loss: 0.8351088166236877:  22%|██▏       | 1392/6470 [15:57<49:10,  1.72it/s]
3|1393|Loss: 0.8351088166236877:  22%|██▏       | 1393/6470 [15:57<51:08,  1.65it/s]
3|1394|Loss: 0.852712094783783:  22%|██▏       | 1393/6470 [15:57<51:08,  1.65it/s] 
3|1394|Loss: 0.852712094783783:  22%|██▏       | 1394/6470 [15:58<52:33,  1.61it/s]
3|1395|Loss: 0.8507259488105774:  22%|██▏       | 1394/6470 [15:58<52:33,  1.61it/s]
3|1395|Loss: 0.8507259488105774:  22%|██▏       | 1395/6470 [15:58<55:14,  1.53it/s]
3|1396|Loss: 0.9442518353462219:  22%|██▏       | 1395/6470 [15:59<55:14,  1.53it/s]
3|1396|Loss: 0.9442518353462219:  22%|██▏       | 1396/6470 [15:59<55:40,  1.52it/s]
3|1397|Loss: 0.9338454604148865:  22%|██▏       | 1396/6470 [15:59<55:40,  1.52it/s]
3|1397|Loss: 0.9338454604148865:  22%|██▏       | 1397/6470 [16:00<56:31,  1.50it/s]
3|1398|Loss: 0.8125612139701843:  22%|██▏       | 1397/6470 [16:00<56:31,  1.50it/s]
3|1398|Loss: 0.8125612139701843:  22%|██▏       | 1398/6470 [16:00<55:31,  1.52it/s]
3|1399|Loss: 0.8859478235244751:  22%|██▏       | 1398/6470 [16:01<55:31,  1.52it/s]
3|1399|Loss: 0.8859478235244751:  22%|██▏       | 1399/6470 [16:01<54:17,  1.56it/s]
3|1400|Loss: 0.7450792193412781:  22%|██▏       | 1399/6470 [16:01<54:17,  1.56it/s]
3|1400|Loss: 0.7450792193412781:  22%|██▏       | 1400/6470 [16:02<52:47,  1.60it/s]
3|1401|Loss: 0.7935512661933899:  22%|██▏       | 1400/6470 [16:02<52:47,  1.60it/s]
3|1401|Loss: 0.7935512661933899:  22%|██▏       | 1401/6470 [16:02<52:34,  1.61it/s]
3|1402|Loss: 0.7073416113853455:  22%|██▏       | 1401/6470 [16:03<52:34,  1.61it/s]
3|1402|Loss: 0.7073416113853455:  22%|██▏       | 1402/6470 [16:03<55:56,  1.51it/s]
3|1403|Loss: 0.8754057288169861:  22%|██▏       | 1402/6470 [16:03<55:56,  1.51it/s]
3|1403|Loss: 0.8754057288169861:  22%|██▏       | 1403/6470 [16:04<58:08,  1.45it/s]
3|1404|Loss: 0.8904142379760742:  22%|██▏       | 1403/6470 [16:04<58:08,  1.45it/s]
3|1404|Loss: 0.8904142379760742:  22%|██▏       | 1404/6470 [16:04<59:13,  1.43it/s]
3|1405|Loss: 0.9293326735496521:  22%|██▏       | 1404/6470 [16:05<59:13,  1.43it/s]
3|1405|Loss: 0.9293326735496521:  22%|██▏       | 1405/6470 [16:05<59:59,  1.41it/s]
3|1406|Loss: 0.8293389081954956:  22%|██▏       | 1405/6470 [16:06<59:59,  1.41it/s]
3|1406|Loss: 0.8293389081954956:  22%|██▏       | 1406/6470 [16:06<59:59,  1.41it/s]
3|1407|Loss: 0.8624290227890015:  22%|██▏       | 1406/6470 [16:06<59:59,  1.41it/s]
3|1407|Loss: 0.8624290227890015:  22%|██▏       | 1407/6470 [16:06<57:22,  1.47it/s]
3|1408|Loss: 0.8412447571754456:  22%|██▏       | 1407/6470 [16:07<57:22,  1.47it/s]
3|1408|Loss: 0.8412447571754456:  22%|██▏       | 1408/6470 [16:07<55:42,  1.51it/s]
3|1409|Loss: 0.8552448749542236:  22%|██▏       | 1408/6470 [16:08<55:42,  1.51it/s]
3|1409|Loss: 0.8552448749542236:  22%|██▏       | 1409/6470 [16:08<1:00:02,  1.40it/s]
3|1410|Loss: 0.889294445514679:  22%|██▏       | 1409/6470 [16:08<1:00:02,  1.40it/s] 
3|1410|Loss: 0.889294445514679:  22%|██▏       | 1410/6470 [16:09<59:42,  1.41it/s]  
3|1411|Loss: 0.9768943190574646:  22%|██▏       | 1410/6470 [16:09<59:42,  1.41it/s]
3|1411|Loss: 0.9768943190574646:  22%|██▏       | 1411/6470 [16:09<1:01:01,  1.38it/s]
3|1412|Loss: 0.8165717720985413:  22%|██▏       | 1411/6470 [16:10<1:01:01,  1.38it/s]
3|1412|Loss: 0.8165717720985413:  22%|██▏       | 1412/6470 [16:10<58:32,  1.44it/s]  
3|1413|Loss: 0.7633709907531738:  22%|██▏       | 1412/6470 [16:10<58:32,  1.44it/s]
3|1413|Loss: 0.7633709907531738:  22%|██▏       | 1413/6470 [16:11<1:00:14,  1.40it/s]
3|1414|Loss: 0.8719521164894104:  22%|██▏       | 1413/6470 [16:11<1:00:14,  1.40it/s]
3|1414|Loss: 0.8719521164894104:  22%|██▏       | 1414/6470 [16:11<58:10,  1.45it/s]  
3|1415|Loss: 0.7790018916130066:  22%|██▏       | 1414/6470 [16:12<58:10,  1.45it/s]
3|1415|Loss: 0.7790018916130066:  22%|██▏       | 1415/6470 [16:12<53:55,  1.56it/s]
3|1416|Loss: 0.7890363335609436:  22%|██▏       | 1415/6470 [16:12<53:55,  1.56it/s]
3|1416|Loss: 0.7890363335609436:  22%|██▏       | 1416/6470 [16:13<55:00,  1.53it/s]
3|1417|Loss: 0.9193812012672424:  22%|██▏       | 1416/6470 [16:13<55:00,  1.53it/s]
3|1417|Loss: 0.9193812012672424:  22%|██▏       | 1417/6470 [16:13<57:16,  1.47it/s]
3|1418|Loss: 0.9690096974372864:  22%|██▏       | 1417/6470 [16:14<57:16,  1.47it/s]
3|1418|Loss: 0.9690096974372864:  22%|██▏       | 1418/6470 [16:14<58:10,  1.45it/s]
3|1419|Loss: 0.7895603775978088:  22%|██▏       | 1418/6470 [16:14<58:10,  1.45it/s]
3|1419|Loss: 0.7895603775978088:  22%|██▏       | 1419/6470 [16:15<55:51,  1.51it/s]
3|1420|Loss: 0.8407033681869507:  22%|██▏       | 1419/6470 [16:15<55:51,  1.51it/s]
3|1420|Loss: 0.8407033681869507:  22%|██▏       | 1420/6470 [16:15<53:39,  1.57it/s]
3|1421|Loss: 0.786823570728302:  22%|██▏       | 1420/6470 [16:16<53:39,  1.57it/s] 
3|1421|Loss: 0.786823570728302:  22%|██▏       | 1421/6470 [16:16<55:18,  1.52it/s]
3|1422|Loss: 0.8723056316375732:  22%|██▏       | 1421/6470 [16:16<55:18,  1.52it/s]
3|1422|Loss: 0.8723056316375732:  22%|██▏       | 1422/6470 [16:17<58:27,  1.44it/s]
3|1423|Loss: 0.8022568225860596:  22%|██▏       | 1422/6470 [16:17<58:27,  1.44it/s]
3|1423|Loss: 0.8022568225860596:  22%|██▏       | 1423/6470 [16:17<55:14,  1.52it/s]
3|1424|Loss: 0.898499608039856:  22%|██▏       | 1423/6470 [16:18<55:14,  1.52it/s] 
3|1424|Loss: 0.898499608039856:  22%|██▏       | 1424/6470 [16:18<55:55,  1.50it/s]
3|1425|Loss: 0.9487019777297974:  22%|██▏       | 1424/6470 [16:18<55:55,  1.50it/s]
3|1425|Loss: 0.9487019777297974:  22%|██▏       | 1425/6470 [16:19<57:03,  1.47it/s]
3|1426|Loss: 0.8140241503715515:  22%|██▏       | 1425/6470 [16:19<57:03,  1.47it/s]
3|1426|Loss: 0.8140241503715515:  22%|██▏       | 1426/6470 [16:19<55:00,  1.53it/s]
3|1427|Loss: 0.7585952281951904:  22%|██▏       | 1426/6470 [16:20<55:00,  1.53it/s]
3|1427|Loss: 0.7585952281951904:  22%|██▏       | 1427/6470 [16:20<56:35,  1.49it/s]
3|1428|Loss: 0.8604674935340881:  22%|██▏       | 1427/6470 [16:20<56:35,  1.49it/s]
3|1428|Loss: 0.8604674935340881:  22%|██▏       | 1428/6470 [16:21<1:00:44,  1.38it/s]
3|1429|Loss: 0.840309739112854:  22%|██▏       | 1428/6470 [16:21<1:00:44,  1.38it/s] 
3|1429|Loss: 0.840309739112854:  22%|██▏       | 1429/6470 [16:21<57:26,  1.46it/s]  
3|1430|Loss: 0.8262413740158081:  22%|██▏       | 1429/6470 [16:22<57:26,  1.46it/s]
3|1430|Loss: 0.8262413740158081:  22%|██▏       | 1430/6470 [16:22<55:39,  1.51it/s]
3|1431|Loss: 0.8310315608978271:  22%|██▏       | 1430/6470 [16:22<55:39,  1.51it/s]
3|1431|Loss: 0.8310315608978271:  22%|██▏       | 1431/6470 [16:23<58:29,  1.44it/s]
3|1432|Loss: 0.9964023232460022:  22%|██▏       | 1431/6470 [16:23<58:29,  1.44it/s]
3|1432|Loss: 0.9964023232460022:  22%|██▏       | 1432/6470 [16:24<59:48,  1.40it/s]
3|1433|Loss: 0.8780186772346497:  22%|██▏       | 1432/6470 [16:24<59:48,  1.40it/s]
3|1433|Loss: 0.8780186772346497:  22%|██▏       | 1433/6470 [16:24<58:10,  1.44it/s]
3|1434|Loss: 0.8230915665626526:  22%|██▏       | 1433/6470 [16:25<58:10,  1.44it/s]
3|1434|Loss: 0.8230915665626526:  22%|██▏       | 1434/6470 [16:25<57:52,  1.45it/s]
3|1435|Loss: 0.8254661560058594:  22%|██▏       | 1434/6470 [16:25<57:52,  1.45it/s]
3|1435|Loss: 0.8254661560058594:  22%|██▏       | 1435/6470 [16:26<55:00,  1.53it/s]
3|1436|Loss: 0.7269154787063599:  22%|██▏       | 1435/6470 [16:26<55:00,  1.53it/s]
3|1436|Loss: 0.7269154787063599:  22%|██▏       | 1436/6470 [16:26<53:37,  1.56it/s]
3|1437|Loss: 0.8493931889533997:  22%|██▏       | 1436/6470 [16:26<53:37,  1.56it/s]
3|1437|Loss: 0.8493931889533997:  22%|██▏       | 1437/6470 [16:27<53:35,  1.57it/s]
3|1438|Loss: 0.8330999612808228:  22%|██▏       | 1437/6470 [16:27<53:35,  1.57it/s]
3|1438|Loss: 0.8330999612808228:  22%|██▏       | 1438/6470 [16:28<58:35,  1.43it/s]
3|1439|Loss: 0.8679409623146057:  22%|██▏       | 1438/6470 [16:28<58:35,  1.43it/s]
3|1439|Loss: 0.8679409623146057:  22%|██▏       | 1439/6470 [16:28<1:00:26,  1.39it/s]
3|1440|Loss: 0.866641104221344:  22%|██▏       | 1439/6470 [16:29<1:00:26,  1.39it/s] 
3|1440|Loss: 0.866641104221344:  22%|██▏       | 1440/6470 [16:29<1:03:09,  1.33it/s]
3|1441|Loss: 0.8617056012153625:  22%|██▏       | 1440/6470 [16:30<1:03:09,  1.33it/s]
3|1441|Loss: 0.8617056012153625:  22%|██▏       | 1441/6470 [16:30<1:05:23,  1.28it/s]
3|1442|Loss: 0.7945462465286255:  22%|██▏       | 1441/6470 [16:30<1:05:23,  1.28it/s]
3|1442|Loss: 0.7945462465286255:  22%|██▏       | 1442/6470 [16:31<1:07:16,  1.25it/s]
3|1443|Loss: 1.0174548625946045:  22%|██▏       | 1442/6470 [16:31<1:07:16,  1.25it/s]
3|1443|Loss: 1.0174548625946045:  22%|██▏       | 1443/6470 [16:32<1:04:27,  1.30it/s]
3|1444|Loss: 0.9339293241500854:  22%|██▏       | 1443/6470 [16:32<1:04:27,  1.30it/s]
3|1444|Loss: 0.9339293241500854:  22%|██▏       | 1444/6470 [16:32<57:21,  1.46it/s]  
3|1445|Loss: 0.8508343696594238:  22%|██▏       | 1444/6470 [16:32<57:21,  1.46it/s]
3|1445|Loss: 0.8508343696594238:  22%|██▏       | 1445/6470 [16:33<56:15,  1.49it/s]
3|1446|Loss: 0.9195061326026917:  22%|██▏       | 1445/6470 [16:33<56:15,  1.49it/s]
3|1446|Loss: 0.9195061326026917:  22%|██▏       | 1446/6470 [16:33<58:48,  1.42it/s]
3|1447|Loss: 0.809770405292511:  22%|██▏       | 1446/6470 [16:34<58:48,  1.42it/s] 
3|1447|Loss: 0.809770405292511:  22%|██▏       | 1447/6470 [16:34<1:01:54,  1.35it/s]
3|1448|Loss: 0.9462207555770874:  22%|██▏       | 1447/6470 [16:35<1:01:54,  1.35it/s]
3|1448|Loss: 0.9462207555770874:  22%|██▏       | 1448/6470 [16:35<59:18,  1.41it/s]  
3|1449|Loss: 0.8493540287017822:  22%|██▏       | 1448/6470 [16:35<59:18,  1.41it/s]
3|1449|Loss: 0.8493540287017822:  22%|██▏       | 1449/6470 [16:36<1:00:22,  1.39it/s]
3|1450|Loss: 0.7097432017326355:  22%|██▏       | 1449/6470 [16:36<1:00:22,  1.39it/s]
3|1450|Loss: 0.7097432017326355:  22%|██▏       | 1450/6470 [16:36<56:39,  1.48it/s]  
3|1451|Loss: 0.8354742527008057:  22%|██▏       | 1450/6470 [16:37<56:39,  1.48it/s]
3|1451|Loss: 0.8354742527008057:  22%|██▏       | 1451/6470 [16:37<56:16,  1.49it/s]
3|1452|Loss: 0.9040545225143433:  22%|██▏       | 1451/6470 [16:37<56:16,  1.49it/s]
3|1452|Loss: 0.9040545225143433:  22%|██▏       | 1452/6470 [16:38<58:52,  1.42it/s]
3|1453|Loss: 0.764982283115387:  22%|██▏       | 1452/6470 [16:38<58:52,  1.42it/s] 
3|1453|Loss: 0.764982283115387:  22%|██▏       | 1453/6470 [16:38<1:00:02,  1.39it/s]
3|1454|Loss: 0.7085843086242676:  22%|██▏       | 1453/6470 [16:39<1:00:02,  1.39it/s]
3|1454|Loss: 0.7085843086242676:  22%|██▏       | 1454/6470 [16:39<59:42,  1.40it/s]  
3|1455|Loss: 0.8731429576873779:  22%|██▏       | 1454/6470 [16:40<59:42,  1.40it/s]
3|1455|Loss: 0.8731429576873779:  22%|██▏       | 1455/6470 [16:40<1:00:25,  1.38it/s]
3|1456|Loss: 0.9209374785423279:  22%|██▏       | 1455/6470 [16:40<1:00:25,  1.38it/s]
3|1456|Loss: 0.9209374785423279:  23%|██▎       | 1456/6470 [16:40<57:11,  1.46it/s]  
3|1457|Loss: 0.7622687220573425:  23%|██▎       | 1456/6470 [16:41<57:11,  1.46it/s]
3|1457|Loss: 0.7622687220573425:  23%|██▎       | 1457/6470 [16:41<51:51,  1.61it/s]
3|1458|Loss: 0.8598291277885437:  23%|██▎       | 1457/6470 [16:41<51:51,  1.61it/s]
3|1458|Loss: 0.8598291277885437:  23%|██▎       | 1458/6470 [16:42<52:59,  1.58it/s]
3|1459|Loss: 0.7817807793617249:  23%|██▎       | 1458/6470 [16:42<52:59,  1.58it/s]
3|1459|Loss: 0.7817807793617249:  23%|██▎       | 1459/6470 [16:42<51:05,  1.63it/s]
3|1460|Loss: 0.8772117495536804:  23%|██▎       | 1459/6470 [16:43<51:05,  1.63it/s]
3|1460|Loss: 0.8772117495536804:  23%|██▎       | 1460/6470 [16:43<50:38,  1.65it/s]
3|1461|Loss: 0.8270022869110107:  23%|██▎       | 1460/6470 [16:43<50:38,  1.65it/s]
3|1461|Loss: 0.8270022869110107:  23%|██▎       | 1461/6470 [16:43<53:11,  1.57it/s]
3|1462|Loss: 0.803283154964447:  23%|██▎       | 1461/6470 [16:44<53:11,  1.57it/s] 
3|1462|Loss: 0.803283154964447:  23%|██▎       | 1462/6470 [16:44<53:18,  1.57it/s]
3|1463|Loss: 0.8931427597999573:  23%|██▎       | 1462/6470 [16:44<53:18,  1.57it/s]
3|1463|Loss: 0.8931427597999573:  23%|██▎       | 1463/6470 [16:45<53:02,  1.57it/s]
3|1464|Loss: 0.8961179852485657:  23%|██▎       | 1463/6470 [16:45<53:02,  1.57it/s]
3|1464|Loss: 0.8961179852485657:  23%|██▎       | 1464/6470 [16:45<54:15,  1.54it/s]
3|1465|Loss: 0.7672713994979858:  23%|██▎       | 1464/6470 [16:46<54:15,  1.54it/s]
3|1465|Loss: 0.7672713994979858:  23%|██▎       | 1465/6470 [16:46<55:19,  1.51it/s]
3|1466|Loss: 0.835269033908844:  23%|██▎       | 1465/6470 [16:46<55:19,  1.51it/s] 
3|1466|Loss: 0.835269033908844:  23%|██▎       | 1466/6470 [16:47<53:08,  1.57it/s]
3|1467|Loss: 0.8696808218955994:  23%|██▎       | 1466/6470 [16:47<53:08,  1.57it/s]
3|1467|Loss: 0.8696808218955994:  23%|██▎       | 1467/6470 [16:47<56:41,  1.47it/s]
3|1468|Loss: 0.8908607959747314:  23%|██▎       | 1467/6470 [16:48<56:41,  1.47it/s]
3|1468|Loss: 0.8908607959747314:  23%|██▎       | 1468/6470 [16:48<56:43,  1.47it/s]
3|1469|Loss: 0.9003010988235474:  23%|██▎       | 1468/6470 [16:49<56:43,  1.47it/s]
3|1469|Loss: 0.9003010988235474:  23%|██▎       | 1469/6470 [16:49<55:05,  1.51it/s]
3|1470|Loss: 0.7518143057823181:  23%|██▎       | 1469/6470 [16:49<55:05,  1.51it/s]
3|1470|Loss: 0.7518143057823181:  23%|██▎       | 1470/6470 [16:49<54:10,  1.54it/s]
3|1471|Loss: 0.9512332081794739:  23%|██▎       | 1470/6470 [16:50<54:10,  1.54it/s]
3|1471|Loss: 0.9512332081794739:  23%|██▎       | 1471/6470 [16:50<53:51,  1.55it/s]
3|1472|Loss: 0.8474348783493042:  23%|██▎       | 1471/6470 [16:50<53:51,  1.55it/s]
3|1472|Loss: 0.8474348783493042:  23%|██▎       | 1472/6470 [16:51<56:40,  1.47it/s]
3|1473|Loss: 0.7335394024848938:  23%|██▎       | 1472/6470 [16:51<56:40,  1.47it/s]
3|1473|Loss: 0.7335394024848938:  23%|██▎       | 1473/6470 [16:52<58:34,  1.42it/s]
3|1474|Loss: 0.7965162992477417:  23%|██▎       | 1473/6470 [16:52<58:34,  1.42it/s]
3|1474|Loss: 0.7965162992477417:  23%|██▎       | 1474/6470 [16:52<55:25,  1.50it/s]
3|1475|Loss: 0.7387126088142395:  23%|██▎       | 1474/6470 [16:52<55:25,  1.50it/s]
3|1475|Loss: 0.7387126088142395:  23%|██▎       | 1475/6470 [16:53<53:48,  1.55it/s]
3|1476|Loss: 0.9370813965797424:  23%|██▎       | 1475/6470 [16:53<53:48,  1.55it/s]
3|1476|Loss: 0.9370813965797424:  23%|██▎       | 1476/6470 [16:53<53:26,  1.56it/s]
3|1477|Loss: 0.7837175726890564:  23%|██▎       | 1476/6470 [16:54<53:26,  1.56it/s]
3|1477|Loss: 0.7837175726890564:  23%|██▎       | 1477/6470 [16:54<49:38,  1.68it/s]
3|1478|Loss: 0.8741523027420044:  23%|██▎       | 1477/6470 [16:54<49:38,  1.68it/s]
3|1478|Loss: 0.8741523027420044:  23%|██▎       | 1478/6470 [16:55<51:26,  1.62it/s]
3|1479|Loss: 0.7853808403015137:  23%|██▎       | 1478/6470 [16:55<51:26,  1.62it/s]
3|1479|Loss: 0.7853808403015137:  23%|██▎       | 1479/6470 [16:55<53:54,  1.54it/s]
3|1480|Loss: 0.8346673250198364:  23%|██▎       | 1479/6470 [16:56<53:54,  1.54it/s]
3|1480|Loss: 0.8346673250198364:  23%|██▎       | 1480/6470 [16:56<55:13,  1.51it/s]
3|1481|Loss: 0.9082351922988892:  23%|██▎       | 1480/6470 [16:56<55:13,  1.51it/s]
3|1481|Loss: 0.9082351922988892:  23%|██▎       | 1481/6470 [16:57<56:53,  1.46it/s]
3|1482|Loss: 0.9643221497535706:  23%|██▎       | 1481/6470 [16:57<56:53,  1.46it/s]
3|1482|Loss: 0.9643221497535706:  23%|██▎       | 1482/6470 [16:57<58:18,  1.43it/s]
3|1483|Loss: 0.9322473406791687:  23%|██▎       | 1482/6470 [16:58<58:18,  1.43it/s]
3|1483|Loss: 0.9322473406791687:  23%|██▎       | 1483/6470 [16:58<59:08,  1.41it/s]
3|1484|Loss: 0.8104357123374939:  23%|██▎       | 1483/6470 [16:59<59:08,  1.41it/s]
3|1484|Loss: 0.8104357123374939:  23%|██▎       | 1484/6470 [16:59<57:52,  1.44it/s]
3|1485|Loss: 0.8881751298904419:  23%|██▎       | 1484/6470 [16:59<57:52,  1.44it/s]
3|1485|Loss: 0.8881751298904419:  23%|██▎       | 1485/6470 [16:59<56:23,  1.47it/s]
3|1486|Loss: 0.7760425209999084:  23%|██▎       | 1485/6470 [17:00<56:23,  1.47it/s]
3|1486|Loss: 0.7760425209999084:  23%|██▎       | 1486/6470 [17:00<56:37,  1.47it/s]
3|1487|Loss: 0.9913555383682251:  23%|██▎       | 1486/6470 [17:01<56:37,  1.47it/s]
3|1487|Loss: 0.9913555383682251:  23%|██▎       | 1487/6470 [17:01<57:05,  1.45it/s]
3|1488|Loss: 0.9795058965682983:  23%|██▎       | 1487/6470 [17:01<57:05,  1.45it/s]
3|1488|Loss: 0.9795058965682983:  23%|██▎       | 1488/6470 [17:02<57:33,  1.44it/s]
3|1489|Loss: 0.7432137131690979:  23%|██▎       | 1488/6470 [17:02<57:33,  1.44it/s]
3|1489|Loss: 0.7432137131690979:  23%|██▎       | 1489/6470 [17:02<57:57,  1.43it/s]
3|1490|Loss: 0.6483778357505798:  23%|██▎       | 1489/6470 [17:03<57:57,  1.43it/s]
3|1490|Loss: 0.6483778357505798:  23%|██▎       | 1490/6470 [17:03<1:03:07,  1.31it/s]
3|1491|Loss: 0.7141780257225037:  23%|██▎       | 1490/6470 [17:04<1:03:07,  1.31it/s]
3|1491|Loss: 0.7141780257225037:  23%|██▎       | 1491/6470 [17:04<1:01:19,  1.35it/s]
3|1492|Loss: 0.9250668287277222:  23%|██▎       | 1491/6470 [17:04<1:01:19,  1.35it/s]
3|1492|Loss: 0.9250668287277222:  23%|██▎       | 1492/6470 [17:04<53:27,  1.55it/s]  
3|1493|Loss: 0.9782759547233582:  23%|██▎       | 1492/6470 [17:05<53:27,  1.55it/s]
3|1493|Loss: 0.9782759547233582:  23%|██▎       | 1493/6470 [17:05<49:40,  1.67it/s]
3|1494|Loss: 0.8495795726776123:  23%|██▎       | 1493/6470 [17:05<49:40,  1.67it/s]
3|1494|Loss: 0.8495795726776123:  23%|██▎       | 1494/6470 [17:06<55:36,  1.49it/s]
3|1495|Loss: 0.8618703484535217:  23%|██▎       | 1494/6470 [17:06<55:36,  1.49it/s]
3|1495|Loss: 0.8618703484535217:  23%|██▎       | 1495/6470 [17:06<59:33,  1.39it/s]
3|1496|Loss: 0.9130462408065796:  23%|██▎       | 1495/6470 [17:07<59:33,  1.39it/s]
3|1496|Loss: 0.9130462408065796:  23%|██▎       | 1496/6470 [17:07<59:32,  1.39it/s]
3|1497|Loss: 0.9625910520553589:  23%|██▎       | 1496/6470 [17:08<59:32,  1.39it/s]
3|1497|Loss: 0.9625910520553589:  23%|██▎       | 1497/6470 [17:08<1:01:37,  1.34it/s]
3|1498|Loss: 0.8797568678855896:  23%|██▎       | 1497/6470 [17:08<1:01:37,  1.34it/s]
3|1498|Loss: 0.8797568678855896:  23%|██▎       | 1498/6470 [17:09<56:19,  1.47it/s]  
3|1499|Loss: 0.8188003897666931:  23%|██▎       | 1498/6470 [17:09<56:19,  1.47it/s]
3|1499|Loss: 0.8188003897666931:  23%|██▎       | 1499/6470 [17:09<53:45,  1.54it/s]
3|1500|Loss: 0.7402598857879639:  23%|██▎       | 1499/6470 [17:09<53:45,  1.54it/s]
3|1500|Loss: 0.7402598857879639:  23%|██▎       | 1500/6470 [17:10<55:14,  1.50it/s]
3|1501|Loss: 0.8486840128898621:  23%|██▎       | 1500/6470 [17:10<55:14,  1.50it/s]
3|1501|Loss: 0.8486840128898621:  23%|██▎       | 1501/6470 [17:10<55:05,  1.50it/s]
3|1502|Loss: 0.7767952680587769:  23%|██▎       | 1501/6470 [17:11<55:05,  1.50it/s]
3|1502|Loss: 0.7767952680587769:  23%|██▎       | 1502/6470 [17:11<56:40,  1.46it/s]
3|1503|Loss: 0.7786356210708618:  23%|██▎       | 1502/6470 [17:12<56:40,  1.46it/s]
3|1503|Loss: 0.7786356210708618:  23%|██▎       | 1503/6470 [17:12<58:02,  1.43it/s]
3|1504|Loss: 0.7866490483283997:  23%|██▎       | 1503/6470 [17:12<58:02,  1.43it/s]
3|1504|Loss: 0.7866490483283997:  23%|██▎       | 1504/6470 [17:13<58:19,  1.42it/s]
3|1505|Loss: 0.895673930644989:  23%|██▎       | 1504/6470 [17:13<58:19,  1.42it/s] 
3|1505|Loss: 0.895673930644989:  23%|██▎       | 1505/6470 [17:13<57:06,  1.45it/s]
3|1506|Loss: 0.892481803894043:  23%|██▎       | 1505/6470 [17:14<57:06,  1.45it/s]
3|1506|Loss: 0.892481803894043:  23%|██▎       | 1506/6470 [17:14<52:44,  1.57it/s]
3|1507|Loss: 0.8885015249252319:  23%|██▎       | 1506/6470 [17:14<52:44,  1.57it/s]
3|1507|Loss: 0.8885015249252319:  23%|██▎       | 1507/6470 [17:15<57:14,  1.44it/s]
3|1508|Loss: 0.9250707030296326:  23%|██▎       | 1507/6470 [17:15<57:14,  1.44it/s]
3|1508|Loss: 0.9250707030296326:  23%|██▎       | 1508/6470 [17:15<1:00:13,  1.37it/s]
3|1509|Loss: 0.8568567633628845:  23%|██▎       | 1508/6470 [17:16<1:00:13,  1.37it/s]
3|1509|Loss: 0.8568567633628845:  23%|██▎       | 1509/6470 [17:16<1:00:17,  1.37it/s]
3|1510|Loss: 0.8100296854972839:  23%|██▎       | 1509/6470 [17:17<1:00:17,  1.37it/s]
3|1510|Loss: 0.8100296854972839:  23%|██▎       | 1510/6470 [17:17<58:22,  1.42it/s]  
3|1511|Loss: 0.869170069694519:  23%|██▎       | 1510/6470 [17:17<58:22,  1.42it/s] 
3|1511|Loss: 0.869170069694519:  23%|██▎       | 1511/6470 [17:18<57:53,  1.43it/s]
3|1512|Loss: 0.9142467975616455:  23%|██▎       | 1511/6470 [17:18<57:53,  1.43it/s]
3|1512|Loss: 0.9142467975616455:  23%|██▎       | 1512/6470 [17:18<55:20,  1.49it/s]
3|1513|Loss: 0.8374905586242676:  23%|██▎       | 1512/6470 [17:18<55:20,  1.49it/s]
3|1513|Loss: 0.8374905586242676:  23%|██▎       | 1513/6470 [17:19<50:30,  1.64it/s]
3|1514|Loss: 0.8358903527259827:  23%|██▎       | 1513/6470 [17:19<50:30,  1.64it/s]
3|1514|Loss: 0.8358903527259827:  23%|██▎       | 1514/6470 [17:19<51:59,  1.59it/s]
3|1515|Loss: 0.8723469972610474:  23%|██▎       | 1514/6470 [17:20<51:59,  1.59it/s]
3|1515|Loss: 0.8723469972610474:  23%|██▎       | 1515/6470 [17:20<52:19,  1.58it/s]
3|1516|Loss: 0.839885413646698:  23%|██▎       | 1515/6470 [17:20<52:19,  1.58it/s] 
3|1516|Loss: 0.839885413646698:  23%|██▎       | 1516/6470 [17:21<51:51,  1.59it/s]
3|1517|Loss: 0.8920921087265015:  23%|██▎       | 1516/6470 [17:21<51:51,  1.59it/s]
3|1517|Loss: 0.8920921087265015:  23%|██▎       | 1517/6470 [17:21<53:47,  1.53it/s]
3|1518|Loss: 0.7952435612678528:  23%|██▎       | 1517/6470 [17:22<53:47,  1.53it/s]
3|1518|Loss: 0.7952435612678528:  23%|██▎       | 1518/6470 [17:22<55:16,  1.49it/s]
3|1519|Loss: 0.841793417930603:  23%|██▎       | 1518/6470 [17:22<55:16,  1.49it/s] 
3|1519|Loss: 0.841793417930603:  23%|██▎       | 1519/6470 [17:23<54:50,  1.50it/s]
3|1520|Loss: 0.8715921640396118:  23%|██▎       | 1519/6470 [17:23<54:50,  1.50it/s]
3|1520|Loss: 0.8715921640396118:  23%|██▎       | 1520/6470 [17:23<57:02,  1.45it/s]
3|1521|Loss: 0.6692837476730347:  23%|██▎       | 1520/6470 [17:24<57:02,  1.45it/s]
3|1521|Loss: 0.6692837476730347:  24%|██▎       | 1521/6470 [17:24<54:17,  1.52it/s]
3|1522|Loss: 0.7528093457221985:  24%|██▎       | 1521/6470 [17:24<54:17,  1.52it/s]
3|1522|Loss: 0.7528093457221985:  24%|██▎       | 1522/6470 [17:24<51:27,  1.60it/s]
3|1523|Loss: 0.9759798049926758:  24%|██▎       | 1522/6470 [17:25<51:27,  1.60it/s]
3|1523|Loss: 0.9759798049926758:  24%|██▎       | 1523/6470 [17:25<55:11,  1.49it/s]
3|1524|Loss: 0.9796602725982666:  24%|██▎       | 1523/6470 [17:26<55:11,  1.49it/s]
3|1524|Loss: 0.9796602725982666:  24%|██▎       | 1524/6470 [17:26<55:25,  1.49it/s]
3|1525|Loss: 0.8662503957748413:  24%|██▎       | 1524/6470 [17:26<55:25,  1.49it/s]
3|1525|Loss: 0.8662503957748413:  24%|██▎       | 1525/6470 [17:27<54:43,  1.51it/s]
3|1526|Loss: 0.8853098154067993:  24%|██▎       | 1525/6470 [17:27<54:43,  1.51it/s]
3|1526|Loss: 0.8853098154067993:  24%|██▎       | 1526/6470 [17:27<51:54,  1.59it/s]
3|1527|Loss: 0.8073557019233704:  24%|██▎       | 1526/6470 [17:27<51:54,  1.59it/s]
3|1527|Loss: 0.8073557019233704:  24%|██▎       | 1527/6470 [17:28<52:49,  1.56it/s]
3|1528|Loss: 1.0336004495620728:  24%|██▎       | 1527/6470 [17:28<52:49,  1.56it/s]
3|1528|Loss: 1.0336004495620728:  24%|██▎       | 1528/6470 [17:28<50:57,  1.62it/s]
3|1529|Loss: 0.8666806221008301:  24%|██▎       | 1528/6470 [17:29<50:57,  1.62it/s]
3|1529|Loss: 0.8666806221008301:  24%|██▎       | 1529/6470 [17:29<52:12,  1.58it/s]
3|1530|Loss: 0.8292617797851562:  24%|██▎       | 1529/6470 [17:29<52:12,  1.58it/s]
3|1530|Loss: 0.8292617797851562:  24%|██▎       | 1530/6470 [17:30<52:46,  1.56it/s]
3|1531|Loss: 0.8415968418121338:  24%|██▎       | 1530/6470 [17:30<52:46,  1.56it/s]
3|1531|Loss: 0.8415968418121338:  24%|██▎       | 1531/6470 [17:30<52:45,  1.56it/s]
3|1532|Loss: 0.8964681029319763:  24%|██▎       | 1531/6470 [17:31<52:45,  1.56it/s]
3|1532|Loss: 0.8964681029319763:  24%|██▎       | 1532/6470 [17:31<57:02,  1.44it/s]
3|1533|Loss: 0.8911423683166504:  24%|██▎       | 1532/6470 [17:32<57:02,  1.44it/s]
3|1533|Loss: 0.8911423683166504:  24%|██▎       | 1533/6470 [17:32<58:38,  1.40it/s]
3|1534|Loss: 0.8677868247032166:  24%|██▎       | 1533/6470 [17:32<58:38,  1.40it/s]
3|1534|Loss: 0.8677868247032166:  24%|██▎       | 1534/6470 [17:33<59:05,  1.39it/s]
3|1535|Loss: 0.8160277009010315:  24%|██▎       | 1534/6470 [17:33<59:05,  1.39it/s]
3|1535|Loss: 0.8160277009010315:  24%|██▎       | 1535/6470 [17:33<57:39,  1.43it/s]
3|1536|Loss: 0.8885395526885986:  24%|██▎       | 1535/6470 [17:34<57:39,  1.43it/s]
3|1536|Loss: 0.8885395526885986:  24%|██▎       | 1536/6470 [17:34<57:14,  1.44it/s]
3|1537|Loss: 0.8556797504425049:  24%|██▎       | 1536/6470 [17:34<57:14,  1.44it/s]
3|1537|Loss: 0.8556797504425049:  24%|██▍       | 1537/6470 [17:35<59:04,  1.39it/s]
3|1538|Loss: 0.9741077423095703:  24%|██▍       | 1537/6470 [17:35<59:04,  1.39it/s]
3|1538|Loss: 0.9741077423095703:  24%|██▍       | 1538/6470 [17:35<59:36,  1.38it/s]
3|1539|Loss: 0.8010117411613464:  24%|██▍       | 1538/6470 [17:36<59:36,  1.38it/s]
3|1539|Loss: 0.8010117411613464:  24%|██▍       | 1539/6470 [17:36<57:56,  1.42it/s]
3|1540|Loss: 1.0469913482666016:  24%|██▍       | 1539/6470 [17:37<57:56,  1.42it/s]
3|1540|Loss: 1.0469913482666016:  24%|██▍       | 1540/6470 [17:37<59:17,  1.39it/s]
3|1541|Loss: 0.8025479912757874:  24%|██▍       | 1540/6470 [17:37<59:17,  1.39it/s]
3|1541|Loss: 0.8025479912757874:  24%|██▍       | 1541/6470 [17:38<1:00:19,  1.36it/s]
3|1542|Loss: 1.0105664730072021:  24%|██▍       | 1541/6470 [17:38<1:00:19,  1.36it/s]
3|1542|Loss: 1.0105664730072021:  24%|██▍       | 1542/6470 [17:38<59:47,  1.37it/s]  
3|1543|Loss: 0.8516201972961426:  24%|██▍       | 1542/6470 [17:39<59:47,  1.37it/s]
3|1543|Loss: 0.8516201972961426:  24%|██▍       | 1543/6470 [17:39<58:02,  1.41it/s]
3|1544|Loss: 0.7889097929000854:  24%|██▍       | 1543/6470 [17:39<58:02,  1.41it/s]
3|1544|Loss: 0.7889097929000854:  24%|██▍       | 1544/6470 [17:40<57:40,  1.42it/s]
3|1545|Loss: 0.7572394013404846:  24%|██▍       | 1544/6470 [17:40<57:40,  1.42it/s]
3|1545|Loss: 0.7572394013404846:  24%|██▍       | 1545/6470 [17:40<58:32,  1.40it/s]
3|1546|Loss: 0.7473111748695374:  24%|██▍       | 1545/6470 [17:41<58:32,  1.40it/s]
3|1546|Loss: 0.7473111748695374:  24%|██▍       | 1546/6470 [17:41<53:35,  1.53it/s]
3|1547|Loss: 1.0462677478790283:  24%|██▍       | 1546/6470 [17:41<53:35,  1.53it/s]
3|1547|Loss: 1.0462677478790283:  24%|██▍       | 1547/6470 [17:42<56:24,  1.45it/s]
3|1548|Loss: 0.7730542421340942:  24%|██▍       | 1547/6470 [17:42<56:24,  1.45it/s]
3|1548|Loss: 0.7730542421340942:  24%|██▍       | 1548/6470 [17:42<58:09,  1.41it/s]
3|1549|Loss: 0.8694072365760803:  24%|██▍       | 1548/6470 [17:43<58:09,  1.41it/s]
3|1549|Loss: 0.8694072365760803:  24%|██▍       | 1549/6470 [17:43<59:55,  1.37it/s]
3|1550|Loss: 0.862770676612854:  24%|██▍       | 1549/6470 [17:44<59:55,  1.37it/s] 
3|1550|Loss: 0.862770676612854:  24%|██▍       | 1550/6470 [17:44<1:01:43,  1.33it/s]
3|1551|Loss: 0.9428859949111938:  24%|██▍       | 1550/6470 [17:45<1:01:43,  1.33it/s]
3|1551|Loss: 0.9428859949111938:  24%|██▍       | 1551/6470 [17:45<1:01:38,  1.33it/s]
3|1552|Loss: 0.8328866362571716:  24%|██▍       | 1551/6470 [17:45<1:01:38,  1.33it/s]
3|1552|Loss: 0.8328866362571716:  24%|██▍       | 1552/6470 [17:45<58:46,  1.39it/s]  
3|1553|Loss: 0.8709091544151306:  24%|██▍       | 1552/6470 [17:46<58:46,  1.39it/s]
3|1553|Loss: 0.8709091544151306:  24%|██▍       | 1553/6470 [17:46<57:49,  1.42it/s]
3|1554|Loss: 0.9060784578323364:  24%|██▍       | 1553/6470 [17:47<57:49,  1.42it/s]
3|1554|Loss: 0.9060784578323364:  24%|██▍       | 1554/6470 [17:47<57:44,  1.42it/s]
3|1555|Loss: 0.8191041350364685:  24%|██▍       | 1554/6470 [17:47<57:44,  1.42it/s]
3|1555|Loss: 0.8191041350364685:  24%|██▍       | 1555/6470 [17:48<1:00:26,  1.36it/s]
3|1556|Loss: 0.8871508240699768:  24%|██▍       | 1555/6470 [17:48<1:00:26,  1.36it/s]
3|1556|Loss: 0.8871508240699768:  24%|██▍       | 1556/6470 [17:48<1:00:16,  1.36it/s]
3|1557|Loss: 0.9678744673728943:  24%|██▍       | 1556/6470 [17:49<1:00:16,  1.36it/s]
3|1557|Loss: 0.9678744673728943:  24%|██▍       | 1557/6470 [17:49<58:10,  1.41it/s]  
3|1558|Loss: 0.8371617197990417:  24%|██▍       | 1557/6470 [17:49<58:10,  1.41it/s]
3|1558|Loss: 0.8371617197990417:  24%|██▍       | 1558/6470 [17:50<53:47,  1.52it/s]
3|1559|Loss: 0.941431999206543:  24%|██▍       | 1558/6470 [17:50<53:47,  1.52it/s] 
3|1559|Loss: 0.941431999206543:  24%|██▍       | 1559/6470 [17:50<53:46,  1.52it/s]
3|1560|Loss: 0.7579156756401062:  24%|██▍       | 1559/6470 [17:51<53:46,  1.52it/s]
3|1560|Loss: 0.7579156756401062:  24%|██▍       | 1560/6470 [17:51<54:49,  1.49it/s]
3|1561|Loss: 0.8846526741981506:  24%|██▍       | 1560/6470 [17:51<54:49,  1.49it/s]
3|1561|Loss: 0.8846526741981506:  24%|██▍       | 1561/6470 [17:52<56:19,  1.45it/s]
3|1562|Loss: 0.8678281903266907:  24%|██▍       | 1561/6470 [17:52<56:19,  1.45it/s]
3|1562|Loss: 0.8678281903266907:  24%|██▍       | 1562/6470 [17:52<52:01,  1.57it/s]
3|1563|Loss: 0.8462221622467041:  24%|██▍       | 1562/6470 [17:53<52:01,  1.57it/s]
3|1563|Loss: 0.8462221622467041:  24%|██▍       | 1563/6470 [17:53<53:57,  1.52it/s]
3|1564|Loss: 0.859524130821228:  24%|██▍       | 1563/6470 [17:53<53:57,  1.52it/s] 
3|1564|Loss: 0.859524130821228:  24%|██▍       | 1564/6470 [17:54<54:26,  1.50it/s]
3|1565|Loss: 0.6895714998245239:  24%|██▍       | 1564/6470 [17:54<54:26,  1.50it/s]
3|1565|Loss: 0.6895714998245239:  24%|██▍       | 1565/6470 [17:54<53:17,  1.53it/s]
3|1566|Loss: 0.8330100178718567:  24%|██▍       | 1565/6470 [17:55<53:17,  1.53it/s]
3|1566|Loss: 0.8330100178718567:  24%|██▍       | 1566/6470 [17:55<54:02,  1.51it/s]
3|1567|Loss: 0.7969459891319275:  24%|██▍       | 1566/6470 [17:55<54:02,  1.51it/s]
3|1567|Loss: 0.7969459891319275:  24%|██▍       | 1567/6470 [17:56<55:32,  1.47it/s]
3|1568|Loss: 0.9624809622764587:  24%|██▍       | 1567/6470 [17:56<55:32,  1.47it/s]
3|1568|Loss: 0.9624809622764587:  24%|██▍       | 1568/6470 [17:56<57:09,  1.43it/s]
3|1569|Loss: 0.8998140692710876:  24%|██▍       | 1568/6470 [17:57<57:09,  1.43it/s]
3|1569|Loss: 0.8998140692710876:  24%|██▍       | 1569/6470 [17:57<58:02,  1.41it/s]
3|1570|Loss: 0.8378871083259583:  24%|██▍       | 1569/6470 [17:58<58:02,  1.41it/s]
3|1570|Loss: 0.8378871083259583:  24%|██▍       | 1570/6470 [17:58<59:56,  1.36it/s]
3|1571|Loss: 0.4659956991672516:  24%|██▍       | 1570/6470 [17:58<59:56,  1.36it/s]
3|1571|Loss: 0.4659956991672516:  24%|██▍       | 1571/6470 [17:59<1:00:07,  1.36it/s]
3|1572|Loss: 0.9771708250045776:  24%|██▍       | 1571/6470 [17:59<1:00:07,  1.36it/s]
3|1572|Loss: 0.9771708250045776:  24%|██▍       | 1572/6470 [17:59<56:17,  1.45it/s]  
3|1573|Loss: 0.7143997550010681:  24%|██▍       | 1572/6470 [18:00<56:17,  1.45it/s]
3|1573|Loss: 0.7143997550010681:  24%|██▍       | 1573/6470 [18:00<55:49,  1.46it/s]
3|1574|Loss: 0.9366047978401184:  24%|██▍       | 1573/6470 [18:00<55:49,  1.46it/s]
3|1574|Loss: 0.9366047978401184:  24%|██▍       | 1574/6470 [18:01<55:14,  1.48it/s]
3|1575|Loss: 0.8525678515434265:  24%|██▍       | 1574/6470 [18:01<55:14,  1.48it/s]
3|1575|Loss: 0.8525678515434265:  24%|██▍       | 1575/6470 [18:01<56:35,  1.44it/s]
3|1576|Loss: 0.7994966506958008:  24%|██▍       | 1575/6470 [18:02<56:35,  1.44it/s]
3|1576|Loss: 0.7994966506958008:  24%|██▍       | 1576/6470 [18:02<57:43,  1.41it/s]
3|1577|Loss: 0.9887750148773193:  24%|██▍       | 1576/6470 [18:02<57:43,  1.41it/s]
3|1577|Loss: 0.9887750148773193:  24%|██▍       | 1577/6470 [18:03<54:15,  1.50it/s]
3|1578|Loss: 1.0140291452407837:  24%|██▍       | 1577/6470 [18:03<54:15,  1.50it/s]
3|1578|Loss: 1.0140291452407837:  24%|██▍       | 1578/6470 [18:03<54:59,  1.48it/s]
3|1579|Loss: 0.7375230193138123:  24%|██▍       | 1578/6470 [18:04<54:59,  1.48it/s]
3|1579|Loss: 0.7375230193138123:  24%|██▍       | 1579/6470 [18:04<51:35,  1.58it/s]
3|1580|Loss: 0.7812398672103882:  24%|██▍       | 1579/6470 [18:04<51:35,  1.58it/s]
3|1580|Loss: 0.7812398672103882:  24%|██▍       | 1580/6470 [18:04<49:59,  1.63it/s]
3|1581|Loss: 0.7677765488624573:  24%|██▍       | 1580/6470 [18:05<49:59,  1.63it/s]
3|1581|Loss: 0.7677765488624573:  24%|██▍       | 1581/6470 [18:05<52:41,  1.55it/s]
3|1582|Loss: 0.9001972079277039:  24%|██▍       | 1581/6470 [18:05<52:41,  1.55it/s]
3|1582|Loss: 0.9001972079277039:  24%|██▍       | 1582/6470 [18:06<50:38,  1.61it/s]
3|1583|Loss: 0.7714722752571106:  24%|██▍       | 1582/6470 [18:06<50:38,  1.61it/s]
3|1583|Loss: 0.7714722752571106:  24%|██▍       | 1583/6470 [18:06<50:02,  1.63it/s]
3|1584|Loss: 0.6853851079940796:  24%|██▍       | 1583/6470 [18:07<50:02,  1.63it/s]
3|1584|Loss: 0.6853851079940796:  24%|██▍       | 1584/6470 [18:07<54:42,  1.49it/s]
3|1585|Loss: 0.8438388109207153:  24%|██▍       | 1584/6470 [18:08<54:42,  1.49it/s]
3|1585|Loss: 0.8438388109207153:  24%|██▍       | 1585/6470 [18:08<57:59,  1.40it/s]
3|1586|Loss: 0.8935121893882751:  24%|██▍       | 1585/6470 [18:08<57:59,  1.40it/s]
3|1586|Loss: 0.8935121893882751:  25%|██▍       | 1586/6470 [18:09<56:38,  1.44it/s]
3|1587|Loss: 0.8461886644363403:  25%|██▍       | 1586/6470 [18:09<56:38,  1.44it/s]
3|1587|Loss: 0.8461886644363403:  25%|██▍       | 1587/6470 [18:09<54:02,  1.51it/s]
3|1588|Loss: 0.7826582193374634:  25%|██▍       | 1587/6470 [18:09<54:02,  1.51it/s]
3|1588|Loss: 0.7826582193374634:  25%|██▍       | 1588/6470 [18:10<53:13,  1.53it/s]
3|1589|Loss: 0.8451492786407471:  25%|██▍       | 1588/6470 [18:10<53:13,  1.53it/s]
3|1589|Loss: 0.8451492786407471:  25%|██▍       | 1589/6470 [18:10<52:52,  1.54it/s]
3|1590|Loss: 0.9334190487861633:  25%|██▍       | 1589/6470 [18:11<52:52,  1.54it/s]
3|1590|Loss: 0.9334190487861633:  25%|██▍       | 1590/6470 [18:11<54:13,  1.50it/s]
3|1591|Loss: 0.9010958671569824:  25%|██▍       | 1590/6470 [18:11<54:13,  1.50it/s]
3|1591|Loss: 0.9010958671569824:  25%|██▍       | 1591/6470 [18:12<52:39,  1.54it/s]
3|1592|Loss: 0.6853833794593811:  25%|██▍       | 1591/6470 [18:12<52:39,  1.54it/s]
3|1592|Loss: 0.6853833794593811:  25%|██▍       | 1592/6470 [18:12<53:16,  1.53it/s]
3|1593|Loss: 0.8970875144004822:  25%|██▍       | 1592/6470 [18:13<53:16,  1.53it/s]
3|1593|Loss: 0.8970875144004822:  25%|██▍       | 1593/6470 [18:13<51:23,  1.58it/s]
3|1594|Loss: 0.611285924911499:  25%|██▍       | 1593/6470 [18:13<51:23,  1.58it/s] 
3|1594|Loss: 0.611285924911499:  25%|██▍       | 1594/6470 [18:14<50:40,  1.60it/s]
3|1595|Loss: 0.8222699165344238:  25%|██▍       | 1594/6470 [18:14<50:40,  1.60it/s]
3|1595|Loss: 0.8222699165344238:  25%|██▍       | 1595/6470 [18:14<52:52,  1.54it/s]
3|1596|Loss: 0.9837453961372375:  25%|██▍       | 1595/6470 [18:15<52:52,  1.54it/s]
3|1596|Loss: 0.9837453961372375:  25%|██▍       | 1596/6470 [18:15<52:00,  1.56it/s]
3|1597|Loss: 0.836637556552887:  25%|██▍       | 1596/6470 [18:15<52:00,  1.56it/s] 
3|1597|Loss: 0.836637556552887:  25%|██▍       | 1597/6470 [18:16<53:56,  1.51it/s]
3|1598|Loss: 0.917780876159668:  25%|██▍       | 1597/6470 [18:16<53:56,  1.51it/s]
3|1598|Loss: 0.917780876159668:  25%|██▍       | 1598/6470 [18:16<55:17,  1.47it/s]
3|1599|Loss: 0.7924934029579163:  25%|██▍       | 1598/6470 [18:17<55:17,  1.47it/s]
3|1599|Loss: 0.7924934029579163:  25%|██▍       | 1599/6470 [18:17<56:36,  1.43it/s]
3|1600|Loss: 0.7070629000663757:  25%|██▍       | 1599/6470 [18:17<56:36,  1.43it/s]
3|1600|Loss: 0.7070629000663757:  25%|██▍       | 1600/6470 [18:18<55:13,  1.47it/s]
3|1601|Loss: 0.940075159072876:  25%|██▍       | 1600/6470 [18:18<55:13,  1.47it/s] 
3|1601|Loss: 0.940075159072876:  25%|██▍       | 1601/6470 [18:18<55:51,  1.45it/s]
3|1602|Loss: 0.9122602343559265:  25%|██▍       | 1601/6470 [18:19<55:51,  1.45it/s]
3|1602|Loss: 0.9122602343559265:  25%|██▍       | 1602/6470 [18:19<56:12,  1.44it/s]
3|1603|Loss: 0.7671857476234436:  25%|██▍       | 1602/6470 [18:20<56:12,  1.44it/s]
3|1603|Loss: 0.7671857476234436:  25%|██▍       | 1603/6470 [18:20<59:49,  1.36it/s]
3|1604|Loss: 0.8824855089187622:  25%|██▍       | 1603/6470 [18:20<59:49,  1.36it/s]
3|1604|Loss: 0.8824855089187622:  25%|██▍       | 1604/6470 [18:21<57:45,  1.40it/s]
3|1605|Loss: 0.8150436878204346:  25%|██▍       | 1604/6470 [18:21<57:45,  1.40it/s]
3|1605|Loss: 0.8150436878204346:  25%|██▍       | 1605/6470 [18:21<57:27,  1.41it/s]
3|1606|Loss: 0.9354138374328613:  25%|██▍       | 1605/6470 [18:22<57:27,  1.41it/s]
3|1606|Loss: 0.9354138374328613:  25%|██▍       | 1606/6470 [18:22<57:33,  1.41it/s]
3|1607|Loss: 0.7573196291923523:  25%|██▍       | 1606/6470 [18:22<57:33,  1.41it/s]
3|1607|Loss: 0.7573196291923523:  25%|██▍       | 1607/6470 [18:23<59:59,  1.35it/s]
3|1608|Loss: 0.8857025504112244:  25%|██▍       | 1607/6470 [18:23<59:59,  1.35it/s]
3|1608|Loss: 0.8857025504112244:  25%|██▍       | 1608/6470 [18:24<1:02:40,  1.29it/s]
3|1609|Loss: 0.7681447267532349:  25%|██▍       | 1608/6470 [18:24<1:02:40,  1.29it/s]
3|1609|Loss: 0.7681447267532349:  25%|██▍       | 1609/6470 [18:25<1:05:03,  1.25it/s]
3|1610|Loss: 0.755038321018219:  25%|██▍       | 1609/6470 [18:25<1:05:03,  1.25it/s] 
3|1610|Loss: 0.755038321018219:  25%|██▍       | 1610/6470 [18:25<1:04:21,  1.26it/s]
3|1611|Loss: 0.9084658026695251:  25%|██▍       | 1610/6470 [18:26<1:04:21,  1.26it/s]
3|1611|Loss: 0.9084658026695251:  25%|██▍       | 1611/6470 [18:26<1:03:26,  1.28it/s]
3|1612|Loss: 0.8070577383041382:  25%|██▍       | 1611/6470 [18:26<1:03:26,  1.28it/s]
3|1612|Loss: 0.8070577383041382:  25%|██▍       | 1612/6470 [18:27<1:00:41,  1.33it/s]
3|1613|Loss: 0.8899142742156982:  25%|██▍       | 1612/6470 [18:27<1:00:41,  1.33it/s]
3|1613|Loss: 0.8899142742156982:  25%|██▍       | 1613/6470 [18:28<1:02:33,  1.29it/s]
3|1614|Loss: 0.858047366142273:  25%|██▍       | 1613/6470 [18:28<1:02:33,  1.29it/s] 
3|1614|Loss: 0.858047366142273:  25%|██▍       | 1614/6470 [18:28<1:00:40,  1.33it/s]
3|1615|Loss: 0.9560785889625549:  25%|██▍       | 1614/6470 [18:29<1:00:40,  1.33it/s]
3|1615|Loss: 0.9560785889625549:  25%|██▍       | 1615/6470 [18:29<54:19,  1.49it/s]  
3|1616|Loss: 0.8356853127479553:  25%|██▍       | 1615/6470 [18:29<54:19,  1.49it/s]
3|1616|Loss: 0.8356853127479553:  25%|██▍       | 1616/6470 [18:29<54:10,  1.49it/s]
3|1617|Loss: 0.8622190356254578:  25%|██▍       | 1616/6470 [18:30<54:10,  1.49it/s]
3|1617|Loss: 0.8622190356254578:  25%|██▍       | 1617/6470 [18:30<55:10,  1.47it/s]
3|1618|Loss: 0.7618780732154846:  25%|██▍       | 1617/6470 [18:31<55:10,  1.47it/s]
3|1618|Loss: 0.7618780732154846:  25%|██▌       | 1618/6470 [18:31<58:05,  1.39it/s]
3|1619|Loss: 0.7999053001403809:  25%|██▌       | 1618/6470 [18:31<58:05,  1.39it/s]
3|1619|Loss: 0.7999053001403809:  25%|██▌       | 1619/6470 [18:32<1:01:03,  1.32it/s]
3|1620|Loss: 0.8071960806846619:  25%|██▌       | 1619/6470 [18:32<1:01:03,  1.32it/s]
3|1620|Loss: 0.8071960806846619:  25%|██▌       | 1620/6470 [18:32<59:22,  1.36it/s]  
3|1621|Loss: 0.9267858266830444:  25%|██▌       | 1620/6470 [18:33<59:22,  1.36it/s]
3|1621|Loss: 0.9267858266830444:  25%|██▌       | 1621/6470 [18:33<58:25,  1.38it/s]
3|1622|Loss: 0.906538188457489:  25%|██▌       | 1621/6470 [18:34<58:25,  1.38it/s] 
3|1622|Loss: 0.906538188457489:  25%|██▌       | 1622/6470 [18:34<1:00:56,  1.33it/s]
3|1623|Loss: 0.9055989980697632:  25%|██▌       | 1622/6470 [18:34<1:00:56,  1.33it/s]
3|1623|Loss: 0.9055989980697632:  25%|██▌       | 1623/6470 [18:35<59:41,  1.35it/s]  
3|1624|Loss: 0.801686704158783:  25%|██▌       | 1623/6470 [18:35<59:41,  1.35it/s] 
3|1624|Loss: 0.801686704158783:  25%|██▌       | 1624/6470 [18:36<1:02:10,  1.30it/s]
3|1625|Loss: 0.7632297277450562:  25%|██▌       | 1624/6470 [18:36<1:02:10,  1.30it/s]
3|1625|Loss: 0.7632297277450562:  25%|██▌       | 1625/6470 [18:36<1:00:23,  1.34it/s]
3|1626|Loss: 0.8369564414024353:  25%|██▌       | 1625/6470 [18:37<1:00:23,  1.34it/s]
3|1626|Loss: 0.8369564414024353:  25%|██▌       | 1626/6470 [18:37<57:47,  1.40it/s]  
3|1627|Loss: 0.9362139105796814:  25%|██▌       | 1626/6470 [18:37<57:47,  1.40it/s]
3|1627|Loss: 0.9362139105796814:  25%|██▌       | 1627/6470 [18:37<52:57,  1.52it/s]
3|1628|Loss: 0.875268280506134:  25%|██▌       | 1627/6470 [18:38<52:57,  1.52it/s] 
3|1628|Loss: 0.875268280506134:  25%|██▌       | 1628/6470 [18:38<54:34,  1.48it/s]
3|1629|Loss: 0.985065758228302:  25%|██▌       | 1628/6470 [18:39<54:34,  1.48it/s]
3|1629|Loss: 0.985065758228302:  25%|██▌       | 1629/6470 [18:39<54:34,  1.48it/s]
3|1630|Loss: 0.8665851950645447:  25%|██▌       | 1629/6470 [18:39<54:34,  1.48it/s]
3|1630|Loss: 0.8665851950645447:  25%|██▌       | 1630/6470 [18:40<55:27,  1.45it/s]
3|1631|Loss: 0.8959169387817383:  25%|██▌       | 1630/6470 [18:40<55:27,  1.45it/s]
3|1631|Loss: 0.8959169387817383:  25%|██▌       | 1631/6470 [18:40<54:48,  1.47it/s]
3|1632|Loss: 0.6272220015525818:  25%|██▌       | 1631/6470 [18:41<54:48,  1.47it/s]
3|1632|Loss: 0.6272220015525818:  25%|██▌       | 1632/6470 [18:41<56:59,  1.41it/s]
3|1633|Loss: 0.9014759063720703:  25%|██▌       | 1632/6470 [18:41<56:59,  1.41it/s]
3|1633|Loss: 0.9014759063720703:  25%|██▌       | 1633/6470 [18:42<57:16,  1.41it/s]
3|1634|Loss: 0.8496724963188171:  25%|██▌       | 1633/6470 [18:42<57:16,  1.41it/s]
3|1634|Loss: 0.8496724963188171:  25%|██▌       | 1634/6470 [18:42<57:34,  1.40it/s]
3|1635|Loss: 0.9518612623214722:  25%|██▌       | 1634/6470 [18:43<57:34,  1.40it/s]
3|1635|Loss: 0.9518612623214722:  25%|██▌       | 1635/6470 [18:43<57:28,  1.40it/s]
3|1636|Loss: 0.8690281510353088:  25%|██▌       | 1635/6470 [18:43<57:28,  1.40it/s]
3|1636|Loss: 0.8690281510353088:  25%|██▌       | 1636/6470 [18:44<57:11,  1.41it/s]
3|1637|Loss: 0.8388861417770386:  25%|██▌       | 1636/6470 [18:44<57:11,  1.41it/s]
3|1637|Loss: 0.8388861417770386:  25%|██▌       | 1637/6470 [18:44<55:47,  1.44it/s]
3|1638|Loss: 0.8982594013214111:  25%|██▌       | 1637/6470 [18:45<55:47,  1.44it/s]
3|1638|Loss: 0.8982594013214111:  25%|██▌       | 1638/6470 [18:45<54:35,  1.48it/s]
3|1639|Loss: 0.9340901970863342:  25%|██▌       | 1638/6470 [18:45<54:35,  1.48it/s]
3|1639|Loss: 0.9340901970863342:  25%|██▌       | 1639/6470 [18:46<54:59,  1.46it/s]
3|1640|Loss: 0.8313499689102173:  25%|██▌       | 1639/6470 [18:46<54:59,  1.46it/s]
3|1640|Loss: 0.8313499689102173:  25%|██▌       | 1640/6470 [18:47<57:50,  1.39it/s]
3|1641|Loss: 0.8233113884925842:  25%|██▌       | 1640/6470 [18:47<57:50,  1.39it/s]
3|1641|Loss: 0.8233113884925842:  25%|██▌       | 1641/6470 [18:47<56:29,  1.42it/s]
3|1642|Loss: 0.7536097764968872:  25%|██▌       | 1641/6470 [18:48<56:29,  1.42it/s]
3|1642|Loss: 0.7536097764968872:  25%|██▌       | 1642/6470 [18:48<56:36,  1.42it/s]
3|1643|Loss: 0.7827579975128174:  25%|██▌       | 1642/6470 [18:48<56:36,  1.42it/s]
3|1643|Loss: 0.7827579975128174:  25%|██▌       | 1643/6470 [18:49<59:13,  1.36it/s]
3|1644|Loss: 0.8360599279403687:  25%|██▌       | 1643/6470 [18:49<59:13,  1.36it/s]
3|1644|Loss: 0.8360599279403687:  25%|██▌       | 1644/6470 [18:50<1:00:03,  1.34it/s]
3|1645|Loss: 0.7700856924057007:  25%|██▌       | 1644/6470 [18:50<1:00:03,  1.34it/s]
3|1645|Loss: 0.7700856924057007:  25%|██▌       | 1645/6470 [18:50<1:00:18,  1.33it/s]
3|1646|Loss: 0.8690524697303772:  25%|██▌       | 1645/6470 [18:51<1:00:18,  1.33it/s]
3|1646|Loss: 0.8690524697303772:  25%|██▌       | 1646/6470 [18:51<1:00:03,  1.34it/s]
3|1647|Loss: 1.0099462270736694:  25%|██▌       | 1646/6470 [18:51<1:00:03,  1.34it/s]
3|1647|Loss: 1.0099462270736694:  25%|██▌       | 1647/6470 [18:52<1:00:09,  1.34it/s]
3|1648|Loss: 0.8751097917556763:  25%|██▌       | 1647/6470 [18:52<1:00:09,  1.34it/s]
3|1648|Loss: 0.8751097917556763:  25%|██▌       | 1648/6470 [18:52<57:47,  1.39it/s]  
3|1649|Loss: 0.8236448764801025:  25%|██▌       | 1648/6470 [18:53<57:47,  1.39it/s]
3|1649|Loss: 0.8236448764801025:  25%|██▌       | 1649/6470 [18:53<56:59,  1.41it/s]
3|1650|Loss: 0.7753885388374329:  25%|██▌       | 1649/6470 [18:53<56:59,  1.41it/s]
3|1650|Loss: 0.7753885388374329:  26%|██▌       | 1650/6470 [18:54<54:32,  1.47it/s]
3|1651|Loss: 0.7776816487312317:  26%|██▌       | 1650/6470 [18:54<54:32,  1.47it/s]
3|1651|Loss: 0.7776816487312317:  26%|██▌       | 1651/6470 [18:54<52:07,  1.54it/s]
3|1652|Loss: 0.8145065903663635:  26%|██▌       | 1651/6470 [18:55<52:07,  1.54it/s]
3|1652|Loss: 0.8145065903663635:  26%|██▌       | 1652/6470 [18:55<55:20,  1.45it/s]
3|1653|Loss: 0.8379964232444763:  26%|██▌       | 1652/6470 [18:55<55:20,  1.45it/s]
3|1653|Loss: 0.8379964232444763:  26%|██▌       | 1653/6470 [18:56<55:04,  1.46it/s]
3|1654|Loss: 0.7770282626152039:  26%|██▌       | 1653/6470 [18:56<55:04,  1.46it/s]
3|1654|Loss: 0.7770282626152039:  26%|██▌       | 1654/6470 [18:57<58:16,  1.38it/s]
3|1655|Loss: 0.874254584312439:  26%|██▌       | 1654/6470 [18:57<58:16,  1.38it/s] 
3|1655|Loss: 0.874254584312439:  26%|██▌       | 1655/6470 [18:57<58:39,  1.37it/s]
3|1656|Loss: 0.79217529296875:  26%|██▌       | 1655/6470 [18:58<58:39,  1.37it/s] 
3|1656|Loss: 0.79217529296875:  26%|██▌       | 1656/6470 [18:58<1:01:11,  1.31it/s]
3|1657|Loss: 0.7540168762207031:  26%|██▌       | 1656/6470 [18:59<1:01:11,  1.31it/s]
3|1657|Loss: 0.7540168762207031:  26%|██▌       | 1657/6470 [18:59<1:00:53,  1.32it/s]
3|1658|Loss: 0.9768801927566528:  26%|██▌       | 1657/6470 [18:59<1:00:53,  1.32it/s]
3|1658|Loss: 0.9768801927566528:  26%|██▌       | 1658/6470 [19:00<57:59,  1.38it/s]  
3|1659|Loss: 0.8709267377853394:  26%|██▌       | 1658/6470 [19:00<57:59,  1.38it/s]
3|1659|Loss: 0.8709267377853394:  26%|██▌       | 1659/6470 [19:00<58:03,  1.38it/s]
3|1660|Loss: 1.0339678525924683:  26%|██▌       | 1659/6470 [19:01<58:03,  1.38it/s]
3|1660|Loss: 1.0339678525924683:  26%|██▌       | 1660/6470 [19:01<54:08,  1.48it/s]
3|1661|Loss: 1.098605751991272:  26%|██▌       | 1660/6470 [19:01<54:08,  1.48it/s] 
3|1661|Loss: 1.098605751991272:  26%|██▌       | 1661/6470 [19:02<54:47,  1.46it/s]
3|1662|Loss: 0.7898659706115723:  26%|██▌       | 1661/6470 [19:02<54:47,  1.46it/s]
3|1662|Loss: 0.7898659706115723:  26%|██▌       | 1662/6470 [19:02<51:54,  1.54it/s]
3|1663|Loss: 0.8641093969345093:  26%|██▌       | 1662/6470 [19:02<51:54,  1.54it/s]
3|1663|Loss: 0.8641093969345093:  26%|██▌       | 1663/6470 [19:03<47:23,  1.69it/s]
3|1664|Loss: 0.7640233635902405:  26%|██▌       | 1663/6470 [19:03<47:23,  1.69it/s]
3|1664|Loss: 0.7640233635902405:  26%|██▌       | 1664/6470 [19:03<49:12,  1.63it/s]
3|1665|Loss: 0.8666476607322693:  26%|██▌       | 1664/6470 [19:04<49:12,  1.63it/s]
3|1665|Loss: 0.8666476607322693:  26%|██▌       | 1665/6470 [19:04<48:11,  1.66it/s]
3|1666|Loss: 0.7915943264961243:  26%|██▌       | 1665/6470 [19:04<48:11,  1.66it/s]
3|1666|Loss: 0.7915943264961243:  26%|██▌       | 1666/6470 [19:05<50:40,  1.58it/s]
3|1667|Loss: 0.8734748959541321:  26%|██▌       | 1666/6470 [19:05<50:40,  1.58it/s]
3|1667|Loss: 0.8734748959541321:  26%|██▌       | 1667/6470 [19:05<52:41,  1.52it/s]
3|1668|Loss: 0.9148937463760376:  26%|██▌       | 1667/6470 [19:06<52:41,  1.52it/s]
3|1668|Loss: 0.9148937463760376:  26%|██▌       | 1668/6470 [19:06<52:31,  1.52it/s]
3|1669|Loss: 0.9004722833633423:  26%|██▌       | 1668/6470 [19:06<52:31,  1.52it/s]
3|1669|Loss: 0.9004722833633423:  26%|██▌       | 1669/6470 [19:07<54:15,  1.47it/s]
3|1670|Loss: 1.0888479948043823:  26%|██▌       | 1669/6470 [19:07<54:15,  1.47it/s]
3|1670|Loss: 1.0888479948043823:  26%|██▌       | 1670/6470 [19:07<54:09,  1.48it/s]
3|1671|Loss: 0.9973528385162354:  26%|██▌       | 1670/6470 [19:08<54:09,  1.48it/s]
3|1671|Loss: 0.9973528385162354:  26%|██▌       | 1671/6470 [19:08<54:29,  1.47it/s]
3|1672|Loss: 0.8116247057914734:  26%|██▌       | 1671/6470 [19:08<54:29,  1.47it/s]
3|1672|Loss: 0.8116247057914734:  26%|██▌       | 1672/6470 [19:09<53:17,  1.50it/s]
3|1673|Loss: 0.9585578441619873:  26%|██▌       | 1672/6470 [19:09<53:17,  1.50it/s]
3|1673|Loss: 0.9585578441619873:  26%|██▌       | 1673/6470 [19:09<48:52,  1.64it/s]
3|1674|Loss: 0.8581181764602661:  26%|██▌       | 1673/6470 [19:09<48:52,  1.64it/s]
3|1674|Loss: 0.8581181764602661:  26%|██▌       | 1674/6470 [19:10<48:36,  1.64it/s]
3|1675|Loss: 0.868007242679596:  26%|██▌       | 1674/6470 [19:10<48:36,  1.64it/s] 
3|1675|Loss: 0.868007242679596:  26%|██▌       | 1675/6470 [19:10<49:30,  1.61it/s]
3|1676|Loss: 0.9464333057403564:  26%|██▌       | 1675/6470 [19:11<49:30,  1.61it/s]
3|1676|Loss: 0.9464333057403564:  26%|██▌       | 1676/6470 [19:11<51:10,  1.56it/s]
3|1677|Loss: 0.9018517732620239:  26%|██▌       | 1676/6470 [19:11<51:10,  1.56it/s]
3|1677|Loss: 0.9018517732620239:  26%|██▌       | 1677/6470 [19:12<55:42,  1.43it/s]
3|1678|Loss: 0.8837890625:  26%|██▌       | 1677/6470 [19:12<55:42,  1.43it/s]      
3|1678|Loss: 0.8837890625:  26%|██▌       | 1678/6470 [19:13<57:02,  1.40it/s]
3|1679|Loss: 0.9298389554023743:  26%|██▌       | 1678/6470 [19:13<57:02,  1.40it/s]
3|1679|Loss: 0.9298389554023743:  26%|██▌       | 1679/6470 [19:13<54:26,  1.47it/s]
3|1680|Loss: 0.9428847432136536:  26%|██▌       | 1679/6470 [19:14<54:26,  1.47it/s]
3|1680|Loss: 0.9428847432136536:  26%|██▌       | 1680/6470 [19:14<51:05,  1.56it/s]
3|1681|Loss: 0.9457733035087585:  26%|██▌       | 1680/6470 [19:14<51:05,  1.56it/s]
3|1681|Loss: 0.9457733035087585:  26%|██▌       | 1681/6470 [19:14<50:33,  1.58it/s]
3|1682|Loss: 0.8171479105949402:  26%|██▌       | 1681/6470 [19:15<50:33,  1.58it/s]
3|1682|Loss: 0.8171479105949402:  26%|██▌       | 1682/6470 [19:15<53:35,  1.49it/s]
3|1683|Loss: 0.8957675099372864:  26%|██▌       | 1682/6470 [19:15<53:35,  1.49it/s]
3|1683|Loss: 0.8957675099372864:  26%|██▌       | 1683/6470 [19:16<50:36,  1.58it/s]
3|1684|Loss: 0.7701195478439331:  26%|██▌       | 1683/6470 [19:16<50:36,  1.58it/s]
3|1684|Loss: 0.7701195478439331:  26%|██▌       | 1684/6470 [19:16<49:31,  1.61it/s]
3|1685|Loss: 0.7824549078941345:  26%|██▌       | 1684/6470 [19:17<49:31,  1.61it/s]
3|1685|Loss: 0.7824549078941345:  26%|██▌       | 1685/6470 [19:17<46:53,  1.70it/s]
3|1686|Loss: 0.8200822472572327:  26%|██▌       | 1685/6470 [19:17<46:53,  1.70it/s]
3|1686|Loss: 0.8200822472572327:  26%|██▌       | 1686/6470 [19:17<48:58,  1.63it/s]
3|1687|Loss: 0.9890724420547485:  26%|██▌       | 1686/6470 [19:18<48:58,  1.63it/s]
3|1687|Loss: 0.9890724420547485:  26%|██▌       | 1687/6470 [19:18<51:58,  1.53it/s]
3|1688|Loss: 0.797177791595459:  26%|██▌       | 1687/6470 [19:19<51:58,  1.53it/s] 
3|1688|Loss: 0.797177791595459:  26%|██▌       | 1688/6470 [19:19<47:17,  1.69it/s]
3|1689|Loss: 0.8751225471496582:  26%|██▌       | 1688/6470 [19:19<47:17,  1.69it/s]
3|1689|Loss: 0.8751225471496582:  26%|██▌       | 1689/6470 [19:19<47:29,  1.68it/s]
3|1690|Loss: 0.7585991621017456:  26%|██▌       | 1689/6470 [19:20<47:29,  1.68it/s]
3|1690|Loss: 0.7585991621017456:  26%|██▌       | 1690/6470 [19:20<48:01,  1.66it/s]
3|1691|Loss: 0.5548161864280701:  26%|██▌       | 1690/6470 [19:20<48:01,  1.66it/s]
3|1691|Loss: 0.5548161864280701:  26%|██▌       | 1691/6470 [19:21<53:03,  1.50it/s]
3|1692|Loss: 0.9094327092170715:  26%|██▌       | 1691/6470 [19:21<53:03,  1.50it/s]
3|1692|Loss: 0.9094327092170715:  26%|██▌       | 1692/6470 [19:21<53:43,  1.48it/s]
3|1693|Loss: 0.8683109283447266:  26%|██▌       | 1692/6470 [19:22<53:43,  1.48it/s]
3|1693|Loss: 0.8683109283447266:  26%|██▌       | 1693/6470 [19:22<55:03,  1.45it/s]
3|1694|Loss: 0.8152201771736145:  26%|██▌       | 1693/6470 [19:23<55:03,  1.45it/s]
3|1694|Loss: 0.8152201771736145:  26%|██▌       | 1694/6470 [19:23<59:34,  1.34it/s]
3|1695|Loss: 0.7193019390106201:  26%|██▌       | 1694/6470 [19:23<59:34,  1.34it/s]
3|1695|Loss: 0.7193019390106201:  26%|██▌       | 1695/6470 [19:24<58:28,  1.36it/s]
3|1696|Loss: 0.8929003477096558:  26%|██▌       | 1695/6470 [19:24<58:28,  1.36it/s]
3|1696|Loss: 0.8929003477096558:  26%|██▌       | 1696/6470 [19:24<55:50,  1.43it/s]
3|1697|Loss: 0.8767105937004089:  26%|██▌       | 1696/6470 [19:25<55:50,  1.43it/s]
3|1697|Loss: 0.8767105937004089:  26%|██▌       | 1697/6470 [19:25<55:58,  1.42it/s]
3|1698|Loss: 0.8490378260612488:  26%|██▌       | 1697/6470 [19:25<55:58,  1.42it/s]
3|1698|Loss: 0.8490378260612488:  26%|██▌       | 1698/6470 [19:26<51:44,  1.54it/s]
3|1699|Loss: 0.778528094291687:  26%|██▌       | 1698/6470 [19:26<51:44,  1.54it/s] 
3|1699|Loss: 0.778528094291687:  26%|██▋       | 1699/6470 [19:26<51:51,  1.53it/s]
3|1700|Loss: 0.9284597635269165:  26%|██▋       | 1699/6470 [19:27<51:51,  1.53it/s]
3|1700|Loss: 0.9284597635269165:  26%|██▋       | 1700/6470 [19:27<54:00,  1.47it/s]
3|1701|Loss: 0.8471137881278992:  26%|██▋       | 1700/6470 [19:27<54:00,  1.47it/s]
3|1701|Loss: 0.8471137881278992:  26%|██▋       | 1701/6470 [19:28<56:40,  1.40it/s]
3|1702|Loss: 0.7929095029830933:  26%|██▋       | 1701/6470 [19:28<56:40,  1.40it/s]
3|1702|Loss: 0.7929095029830933:  26%|██▋       | 1702/6470 [19:29<1:01:50,  1.29it/s]
3|1703|Loss: 0.9729126691818237:  26%|██▋       | 1702/6470 [19:29<1:01:50,  1.29it/s]
3|1703|Loss: 0.9729126691818237:  26%|██▋       | 1703/6470 [19:29<1:01:31,  1.29it/s]
3|1704|Loss: 0.7263533473014832:  26%|██▋       | 1703/6470 [19:30<1:01:31,  1.29it/s]
3|1704|Loss: 0.7263533473014832:  26%|██▋       | 1704/6470 [19:30<59:55,  1.33it/s]  
3|1705|Loss: 0.7412073612213135:  26%|██▋       | 1704/6470 [19:31<59:55,  1.33it/s]
3|1705|Loss: 0.7412073612213135:  26%|██▋       | 1705/6470 [19:31<59:23,  1.34it/s]
3|1706|Loss: 0.8499382138252258:  26%|██▋       | 1705/6470 [19:31<59:23,  1.34it/s]
3|1706|Loss: 0.8499382138252258:  26%|██▋       | 1706/6470 [19:31<55:11,  1.44it/s]
3|1707|Loss: 0.8477384448051453:  26%|██▋       | 1706/6470 [19:32<55:11,  1.44it/s]
3|1707|Loss: 0.8477384448051453:  26%|██▋       | 1707/6470 [19:32<54:38,  1.45it/s]
3|1708|Loss: 0.7740799188613892:  26%|██▋       | 1707/6470 [19:33<54:38,  1.45it/s]
3|1708|Loss: 0.7740799188613892:  26%|██▋       | 1708/6470 [19:33<1:03:54,  1.24it/s]
3|1709|Loss: 0.97932368516922:  26%|██▋       | 1708/6470 [19:34<1:03:54,  1.24it/s]  
3|1709|Loss: 0.97932368516922:  26%|██▋       | 1709/6470 [19:34<1:04:31,  1.23it/s]
3|1710|Loss: 0.749854564666748:  26%|██▋       | 1709/6470 [19:34<1:04:31,  1.23it/s]
3|1710|Loss: 0.749854564666748:  26%|██▋       | 1710/6470 [19:35<1:05:01,  1.22it/s]
3|1711|Loss: 0.800414502620697:  26%|██▋       | 1710/6470 [19:35<1:05:01,  1.22it/s]
3|1711|Loss: 0.800414502620697:  26%|██▋       | 1711/6470 [19:36<1:00:43,  1.31it/s]
3|1712|Loss: 0.862491250038147:  26%|██▋       | 1711/6470 [19:36<1:00:43,  1.31it/s]
3|1712|Loss: 0.862491250038147:  26%|██▋       | 1712/6470 [19:36<59:14,  1.34it/s]  
3|1713|Loss: 0.7690606117248535:  26%|██▋       | 1712/6470 [19:37<59:14,  1.34it/s]
3|1713|Loss: 0.7690606117248535:  26%|██▋       | 1713/6470 [19:37<56:01,  1.42it/s]
3|1714|Loss: 0.928761899471283:  26%|██▋       | 1713/6470 [19:37<56:01,  1.42it/s] 
3|1714|Loss: 0.928761899471283:  26%|██▋       | 1714/6470 [19:37<51:38,  1.54it/s]
3|1715|Loss: 0.8686190843582153:  26%|██▋       | 1714/6470 [19:38<51:38,  1.54it/s]
3|1715|Loss: 0.8686190843582153:  27%|██▋       | 1715/6470 [19:38<57:39,  1.37it/s]
3|1716|Loss: 0.8354877829551697:  27%|██▋       | 1715/6470 [19:39<57:39,  1.37it/s]
3|1716|Loss: 0.8354877829551697:  27%|██▋       | 1716/6470 [19:39<57:18,  1.38it/s]
3|1717|Loss: 0.8612293601036072:  27%|██▋       | 1716/6470 [19:39<57:18,  1.38it/s]
3|1717|Loss: 0.8612293601036072:  27%|██▋       | 1717/6470 [19:40<56:15,  1.41it/s]
3|1718|Loss: 0.8407683372497559:  27%|██▋       | 1717/6470 [19:40<56:15,  1.41it/s]
3|1718|Loss: 0.8407683372497559:  27%|██▋       | 1718/6470 [19:40<58:50,  1.35it/s]
3|1719|Loss: 0.720641553401947:  27%|██▋       | 1718/6470 [19:41<58:50,  1.35it/s] 
3|1719|Loss: 0.720641553401947:  27%|██▋       | 1719/6470 [19:41<59:54,  1.32it/s]
3|1720|Loss: 0.7990788221359253:  27%|██▋       | 1719/6470 [19:42<59:54,  1.32it/s]
3|1720|Loss: 0.7990788221359253:  27%|██▋       | 1720/6470 [19:42<1:03:10,  1.25it/s]
3|1721|Loss: 0.9074083566665649:  27%|██▋       | 1720/6470 [19:43<1:03:10,  1.25it/s]
3|1721|Loss: 0.9074083566665649:  27%|██▋       | 1721/6470 [19:43<1:04:35,  1.23it/s]
3|1722|Loss: 0.8708536624908447:  27%|██▋       | 1721/6470 [19:43<1:04:35,  1.23it/s]
3|1722|Loss: 0.8708536624908447:  27%|██▋       | 1722/6470 [19:44<1:03:52,  1.24it/s]
3|1723|Loss: 0.894478440284729:  27%|██▋       | 1722/6470 [19:44<1:03:52,  1.24it/s] 
3|1723|Loss: 0.894478440284729:  27%|██▋       | 1723/6470 [19:45<1:03:46,  1.24it/s]
3|1724|Loss: 0.8522635698318481:  27%|██▋       | 1723/6470 [19:45<1:03:46,  1.24it/s]
3|1724|Loss: 0.8522635698318481:  27%|██▋       | 1724/6470 [19:45<1:00:13,  1.31it/s]
3|1725|Loss: 0.7460117936134338:  27%|██▋       | 1724/6470 [19:46<1:00:13,  1.31it/s]
3|1725|Loss: 0.7460117936134338:  27%|██▋       | 1725/6470 [19:46<1:01:16,  1.29it/s]
3|1726|Loss: 0.8894485831260681:  27%|██▋       | 1725/6470 [19:47<1:01:16,  1.29it/s]
3|1726|Loss: 0.8894485831260681:  27%|██▋       | 1726/6470 [19:47<1:00:26,  1.31it/s]
3|1727|Loss: 0.7647141218185425:  27%|██▋       | 1726/6470 [19:47<1:00:26,  1.31it/s]
3|1727|Loss: 0.7647141218185425:  27%|██▋       | 1727/6470 [19:47<52:33,  1.50it/s]  
3|1728|Loss: 0.6467812657356262:  27%|██▋       | 1727/6470 [19:48<52:33,  1.50it/s]
3|1728|Loss: 0.6467812657356262:  27%|██▋       | 1728/6470 [19:48<54:11,  1.46it/s]
3|1729|Loss: 0.8495799899101257:  27%|██▋       | 1728/6470 [19:48<54:11,  1.46it/s]
3|1729|Loss: 0.8495799899101257:  27%|██▋       | 1729/6470 [19:49<55:21,  1.43it/s]
3|1730|Loss: 0.9509859681129456:  27%|██▋       | 1729/6470 [19:49<55:21,  1.43it/s]
3|1730|Loss: 0.9509859681129456:  27%|██▋       | 1730/6470 [19:49<55:24,  1.43it/s]
3|1731|Loss: 0.9460458755493164:  27%|██▋       | 1730/6470 [19:50<55:24,  1.43it/s]
3|1731|Loss: 0.9460458755493164:  27%|██▋       | 1731/6470 [19:50<54:11,  1.46it/s]
3|1732|Loss: 0.8608333468437195:  27%|██▋       | 1731/6470 [19:50<54:11,  1.46it/s]
3|1732|Loss: 0.8608333468437195:  27%|██▋       | 1732/6470 [19:51<54:09,  1.46it/s]
3|1733|Loss: 0.8126879930496216:  27%|██▋       | 1732/6470 [19:51<54:09,  1.46it/s]
3|1733|Loss: 0.8126879930496216:  27%|██▋       | 1733/6470 [19:51<53:04,  1.49it/s]
3|1734|Loss: 0.9003427624702454:  27%|██▋       | 1733/6470 [19:52<53:04,  1.49it/s]
3|1734|Loss: 0.9003427624702454:  27%|██▋       | 1734/6470 [19:52<51:47,  1.52it/s]
3|1735|Loss: 0.7873452305793762:  27%|██▋       | 1734/6470 [19:52<51:47,  1.52it/s]
3|1735|Loss: 0.7873452305793762:  27%|██▋       | 1735/6470 [19:53<49:59,  1.58it/s]
3|1736|Loss: 0.7540298104286194:  27%|██▋       | 1735/6470 [19:53<49:59,  1.58it/s]
3|1736|Loss: 0.7540298104286194:  27%|██▋       | 1736/6470 [19:53<49:50,  1.58it/s]
3|1737|Loss: 0.8230293393135071:  27%|██▋       | 1736/6470 [19:54<49:50,  1.58it/s]
3|1737|Loss: 0.8230293393135071:  27%|██▋       | 1737/6470 [19:54<49:37,  1.59it/s]
3|1738|Loss: 0.7548704743385315:  27%|██▋       | 1737/6470 [19:54<49:37,  1.59it/s]
3|1738|Loss: 0.7548704743385315:  27%|██▋       | 1738/6470 [19:54<49:44,  1.59it/s]
3|1739|Loss: 1.0183424949645996:  27%|██▋       | 1738/6470 [19:55<49:44,  1.59it/s]
3|1739|Loss: 1.0183424949645996:  27%|██▋       | 1739/6470 [19:55<51:01,  1.55it/s]
3|1740|Loss: 0.8978525996208191:  27%|██▋       | 1739/6470 [19:56<51:01,  1.55it/s]
3|1740|Loss: 0.8978525996208191:  27%|██▋       | 1740/6470 [19:56<48:51,  1.61it/s]
3|1741|Loss: 1.017648458480835:  27%|██▋       | 1740/6470 [19:56<48:51,  1.61it/s] 
3|1741|Loss: 1.017648458480835:  27%|██▋       | 1741/6470 [19:56<47:34,  1.66it/s]
3|1742|Loss: 1.016096830368042:  27%|██▋       | 1741/6470 [19:57<47:34,  1.66it/s]
3|1742|Loss: 1.016096830368042:  27%|██▋       | 1742/6470 [19:57<48:07,  1.64it/s]
3|1743|Loss: 0.6787433624267578:  27%|██▋       | 1742/6470 [19:57<48:07,  1.64it/s]
3|1743|Loss: 0.6787433624267578:  27%|██▋       | 1743/6470 [19:58<56:03,  1.41it/s]
3|1744|Loss: 1.0082722902297974:  27%|██▋       | 1743/6470 [19:58<56:03,  1.41it/s]
3|1744|Loss: 1.0082722902297974:  27%|██▋       | 1744/6470 [19:59<56:38,  1.39it/s]
3|1745|Loss: 0.7869007587432861:  27%|██▋       | 1744/6470 [19:59<56:38,  1.39it/s]
3|1745|Loss: 0.7869007587432861:  27%|██▋       | 1745/6470 [19:59<54:18,  1.45it/s]
3|1746|Loss: 0.8704807758331299:  27%|██▋       | 1745/6470 [20:00<54:18,  1.45it/s]
3|1746|Loss: 0.8704807758331299:  27%|██▋       | 1746/6470 [20:00<51:21,  1.53it/s]
3|1747|Loss: 1.0179511308670044:  27%|██▋       | 1746/6470 [20:00<51:21,  1.53it/s]
3|1747|Loss: 1.0179511308670044:  27%|██▋       | 1747/6470 [20:00<50:12,  1.57it/s]
3|1748|Loss: 0.8685097694396973:  27%|██▋       | 1747/6470 [20:01<50:12,  1.57it/s]
3|1748|Loss: 0.8685097694396973:  27%|██▋       | 1748/6470 [20:01<50:06,  1.57it/s]
3|1749|Loss: 0.9806650280952454:  27%|██▋       | 1748/6470 [20:01<50:06,  1.57it/s]
3|1749|Loss: 0.9806650280952454:  27%|██▋       | 1749/6470 [20:02<50:14,  1.57it/s]
3|1750|Loss: 0.8303356766700745:  27%|██▋       | 1749/6470 [20:02<50:14,  1.57it/s]
3|1750|Loss: 0.8303356766700745:  27%|██▋       | 1750/6470 [20:02<51:31,  1.53it/s]
3|1751|Loss: 0.8983387351036072:  27%|██▋       | 1750/6470 [20:03<51:31,  1.53it/s]
3|1751|Loss: 0.8983387351036072:  27%|██▋       | 1751/6470 [20:03<50:56,  1.54it/s]
3|1752|Loss: 1.0121190547943115:  27%|██▋       | 1751/6470 [20:03<50:56,  1.54it/s]
3|1752|Loss: 1.0121190547943115:  27%|██▋       | 1752/6470 [20:04<50:20,  1.56it/s]
3|1753|Loss: 0.868037223815918:  27%|██▋       | 1752/6470 [20:04<50:20,  1.56it/s] 
3|1753|Loss: 0.868037223815918:  27%|██▋       | 1753/6470 [20:04<52:01,  1.51it/s]
3|1754|Loss: 0.6643499732017517:  27%|██▋       | 1753/6470 [20:05<52:01,  1.51it/s]
3|1754|Loss: 0.6643499732017517:  27%|██▋       | 1754/6470 [20:05<1:02:39,  1.25it/s]
3|1755|Loss: 0.7742633819580078:  27%|██▋       | 1754/6470 [20:06<1:02:39,  1.25it/s]
3|1755|Loss: 0.7742633819580078:  27%|██▋       | 1755/6470 [20:06<1:02:50,  1.25it/s]
3|1756|Loss: 0.884380042552948:  27%|██▋       | 1755/6470 [20:07<1:02:50,  1.25it/s] 
3|1756|Loss: 0.884380042552948:  27%|██▋       | 1756/6470 [20:07<55:47,  1.41it/s]  
3|1757|Loss: 0.9643248915672302:  27%|██▋       | 1756/6470 [20:07<55:47,  1.41it/s]
3|1757|Loss: 0.9643248915672302:  27%|██▋       | 1757/6470 [20:07<54:20,  1.45it/s]
3|1758|Loss: 0.8012418150901794:  27%|██▋       | 1757/6470 [20:08<54:20,  1.45it/s]
3|1758|Loss: 0.8012418150901794:  27%|██▋       | 1758/6470 [20:08<57:29,  1.37it/s]
3|1759|Loss: 0.7783046960830688:  27%|██▋       | 1758/6470 [20:09<57:29,  1.37it/s]
3|1759|Loss: 0.7783046960830688:  27%|██▋       | 1759/6470 [20:09<1:00:34,  1.30it/s]
3|1760|Loss: 0.6983628869056702:  27%|██▋       | 1759/6470 [20:10<1:00:34,  1.30it/s]
3|1760|Loss: 0.6983628869056702:  27%|██▋       | 1760/6470 [20:10<1:00:45,  1.29it/s]
3|1761|Loss: 0.744054913520813:  27%|██▋       | 1760/6470 [20:10<1:00:45,  1.29it/s] 
3|1761|Loss: 0.744054913520813:  27%|██▋       | 1761/6470 [20:11<57:49,  1.36it/s]  
3|1762|Loss: 0.8474422693252563:  27%|██▋       | 1761/6470 [20:11<57:49,  1.36it/s]
3|1762|Loss: 0.8474422693252563:  27%|██▋       | 1762/6470 [20:11<57:27,  1.37it/s]
3|1763|Loss: 0.925796627998352:  27%|██▋       | 1762/6470 [20:12<57:27,  1.37it/s] 
3|1763|Loss: 0.925796627998352:  27%|██▋       | 1763/6470 [20:12<57:07,  1.37it/s]
3|1764|Loss: 0.7746365666389465:  27%|██▋       | 1763/6470 [20:12<57:07,  1.37it/s]
3|1764|Loss: 0.7746365666389465:  27%|██▋       | 1764/6470 [20:13<58:12,  1.35it/s]
3|1765|Loss: 0.7458230257034302:  27%|██▋       | 1764/6470 [20:13<58:12,  1.35it/s]
3|1765|Loss: 0.7458230257034302:  27%|██▋       | 1765/6470 [20:13<56:40,  1.38it/s]
3|1766|Loss: 0.8206114768981934:  27%|██▋       | 1765/6470 [20:14<56:40,  1.38it/s]
3|1766|Loss: 0.8206114768981934:  27%|██▋       | 1766/6470 [20:14<54:14,  1.45it/s]
3|1767|Loss: 0.8216153383255005:  27%|██▋       | 1766/6470 [20:14<54:14,  1.45it/s]
3|1767|Loss: 0.8216153383255005:  27%|██▋       | 1767/6470 [20:15<55:51,  1.40it/s]
3|1768|Loss: 0.9337162971496582:  27%|██▋       | 1767/6470 [20:15<55:51,  1.40it/s]
3|1768|Loss: 0.9337162971496582:  27%|██▋       | 1768/6470 [20:16<58:09,  1.35it/s]
3|1769|Loss: 0.8257580399513245:  27%|██▋       | 1768/6470 [20:16<58:09,  1.35it/s]
3|1769|Loss: 0.8257580399513245:  27%|██▋       | 1769/6470 [20:16<55:31,  1.41it/s]
3|1770|Loss: 0.8796239495277405:  27%|██▋       | 1769/6470 [20:17<55:31,  1.41it/s]
3|1770|Loss: 0.8796239495277405:  27%|██▋       | 1770/6470 [20:17<53:38,  1.46it/s]
3|1771|Loss: 0.6925290822982788:  27%|██▋       | 1770/6470 [20:17<53:38,  1.46it/s]
3|1771|Loss: 0.6925290822982788:  27%|██▋       | 1771/6470 [20:18<54:22,  1.44it/s]
3|1772|Loss: 0.7512940764427185:  27%|██▋       | 1771/6470 [20:18<54:22,  1.44it/s]
3|1772|Loss: 0.7512940764427185:  27%|██▋       | 1772/6470 [20:18<55:42,  1.41it/s]
3|1773|Loss: 0.8107010126113892:  27%|██▋       | 1772/6470 [20:19<55:42,  1.41it/s]
3|1773|Loss: 0.8107010126113892:  27%|██▋       | 1773/6470 [20:19<54:24,  1.44it/s]
3|1774|Loss: 0.8538414239883423:  27%|██▋       | 1773/6470 [20:19<54:24,  1.44it/s]
3|1774|Loss: 0.8538414239883423:  27%|██▋       | 1774/6470 [20:20<51:43,  1.51it/s]
3|1775|Loss: 0.7697169780731201:  27%|██▋       | 1774/6470 [20:20<51:43,  1.51it/s]
3|1775|Loss: 0.7697169780731201:  27%|██▋       | 1775/6470 [20:20<55:25,  1.41it/s]
3|1776|Loss: 0.9286315441131592:  27%|██▋       | 1775/6470 [20:21<55:25,  1.41it/s]
3|1776|Loss: 0.9286315441131592:  27%|██▋       | 1776/6470 [20:21<56:12,  1.39it/s]
3|1777|Loss: 0.7061794996261597:  27%|██▋       | 1776/6470 [20:22<56:12,  1.39it/s]
3|1777|Loss: 0.7061794996261597:  27%|██▋       | 1777/6470 [20:22<56:24,  1.39it/s]
3|1778|Loss: 0.774740993976593:  27%|██▋       | 1777/6470 [20:22<56:24,  1.39it/s] 
3|1778|Loss: 0.774740993976593:  27%|██▋       | 1778/6470 [20:23<54:52,  1.42it/s]
3|1779|Loss: 0.8577561974525452:  27%|██▋       | 1778/6470 [20:23<54:52,  1.42it/s]
3|1779|Loss: 0.8577561974525452:  27%|██▋       | 1779/6470 [20:23<56:28,  1.38it/s]
3|1780|Loss: 0.8760339021682739:  27%|██▋       | 1779/6470 [20:24<56:28,  1.38it/s]
3|1780|Loss: 0.8760339021682739:  28%|██▊       | 1780/6470 [20:24<56:55,  1.37it/s]
3|1781|Loss: 0.7432367205619812:  28%|██▊       | 1780/6470 [20:24<56:55,  1.37it/s]
3|1781|Loss: 0.7432367205619812:  28%|██▊       | 1781/6470 [20:25<56:27,  1.38it/s]
3|1782|Loss: 0.7957081198692322:  28%|██▊       | 1781/6470 [20:25<56:27,  1.38it/s]
3|1782|Loss: 0.7957081198692322:  28%|██▊       | 1782/6470 [20:25<54:54,  1.42it/s]
3|1783|Loss: 0.8759793639183044:  28%|██▊       | 1782/6470 [20:26<54:54,  1.42it/s]
3|1783|Loss: 0.8759793639183044:  28%|██▊       | 1783/6470 [20:26<53:02,  1.47it/s]
3|1784|Loss: 0.8020355105400085:  28%|██▊       | 1783/6470 [20:26<53:02,  1.47it/s]
3|1784|Loss: 0.8020355105400085:  28%|██▊       | 1784/6470 [20:27<55:30,  1.41it/s]
3|1785|Loss: 0.8414415121078491:  28%|██▊       | 1784/6470 [20:27<55:30,  1.41it/s]
3|1785|Loss: 0.8414415121078491:  28%|██▊       | 1785/6470 [20:28<55:58,  1.39it/s]
3|1786|Loss: 0.8045856952667236:  28%|██▊       | 1785/6470 [20:28<55:58,  1.39it/s]
3|1786|Loss: 0.8045856952667236:  28%|██▊       | 1786/6470 [20:28<56:13,  1.39it/s]
3|1787|Loss: 0.979222297668457:  28%|██▊       | 1786/6470 [20:29<56:13,  1.39it/s] 
3|1787|Loss: 0.979222297668457:  28%|██▊       | 1787/6470 [20:29<58:10,  1.34it/s]
3|1788|Loss: 0.8545576333999634:  28%|██▊       | 1787/6470 [20:29<58:10,  1.34it/s]
3|1788|Loss: 0.8545576333999634:  28%|██▊       | 1788/6470 [20:30<56:42,  1.38it/s]
3|1789|Loss: 0.8972962498664856:  28%|██▊       | 1788/6470 [20:30<56:42,  1.38it/s]
3|1789|Loss: 0.8972962498664856:  28%|██▊       | 1789/6470 [20:30<52:11,  1.49it/s]
3|1790|Loss: 0.8849231600761414:  28%|██▊       | 1789/6470 [20:31<52:11,  1.49it/s]
3|1790|Loss: 0.8849231600761414:  28%|██▊       | 1790/6470 [20:31<50:29,  1.54it/s]
3|1791|Loss: 0.8741027116775513:  28%|██▊       | 1790/6470 [20:31<50:29,  1.54it/s]
3|1791|Loss: 0.8741027116775513:  28%|██▊       | 1791/6470 [20:31<48:55,  1.59it/s]
3|1792|Loss: 0.9114882946014404:  28%|██▊       | 1791/6470 [20:32<48:55,  1.59it/s]
3|1792|Loss: 0.9114882946014404:  28%|██▊       | 1792/6470 [20:32<45:28,  1.71it/s]
3|1793|Loss: 0.5962294340133667:  28%|██▊       | 1792/6470 [20:32<45:28,  1.71it/s]
3|1793|Loss: 0.5962294340133667:  28%|██▊       | 1793/6470 [20:33<52:18,  1.49it/s]
3|1794|Loss: 0.8706116676330566:  28%|██▊       | 1793/6470 [20:33<52:18,  1.49it/s]
3|1794|Loss: 0.8706116676330566:  28%|██▊       | 1794/6470 [20:34<1:04:21,  1.21it/s]
3|1795|Loss: 0.8286072611808777:  28%|██▊       | 1794/6470 [20:34<1:04:21,  1.21it/s]
3|1795|Loss: 0.8286072611808777:  28%|██▊       | 1795/6470 [20:35<1:01:26,  1.27it/s]
3|1796|Loss: 0.8924726843833923:  28%|██▊       | 1795/6470 [20:35<1:01:26,  1.27it/s]
3|1796|Loss: 0.8924726843833923:  28%|██▊       | 1796/6470 [20:36<1:01:46,  1.26it/s]
3|1797|Loss: 0.8088774085044861:  28%|██▊       | 1796/6470 [20:36<1:01:46,  1.26it/s]
3|1797|Loss: 0.8088774085044861:  28%|██▊       | 1797/6470 [20:36<59:03,  1.32it/s]  
3|1798|Loss: 0.9023243188858032:  28%|██▊       | 1797/6470 [20:37<59:03,  1.32it/s]
3|1798|Loss: 0.9023243188858032:  28%|██▊       | 1798/6470 [20:37<1:00:15,  1.29it/s]
3|1799|Loss: 0.709880530834198:  28%|██▊       | 1798/6470 [20:37<1:00:15,  1.29it/s] 
3|1799|Loss: 0.709880530834198:  28%|██▊       | 1799/6470 [20:38<58:23,  1.33it/s]  
3|1800|Loss: 0.8246567845344543:  28%|██▊       | 1799/6470 [20:38<58:23,  1.33it/s]
3|1800|Loss: 0.8246567845344543:  28%|██▊       | 1800/6470 [20:38<57:33,  1.35it/s]
3|1801|Loss: 0.5576186776161194:  28%|██▊       | 1800/6470 [20:39<57:33,  1.35it/s]
3|1801|Loss: 0.5576186776161194:  28%|██▊       | 1801/6470 [20:39<1:02:30,  1.24it/s]
3|1802|Loss: 0.835940957069397:  28%|██▊       | 1801/6470 [20:40<1:02:30,  1.24it/s] 
3|1802|Loss: 0.835940957069397:  28%|██▊       | 1802/6470 [20:40<58:20,  1.33it/s]  
3|1803|Loss: 0.7848607897758484:  28%|██▊       | 1802/6470 [20:40<58:20,  1.33it/s]
3|1803|Loss: 0.7848607897758484:  28%|██▊       | 1803/6470 [20:40<50:12,  1.55it/s]
3|1804|Loss: 0.7206190228462219:  28%|██▊       | 1803/6470 [20:41<50:12,  1.55it/s]
3|1804|Loss: 0.7206190228462219:  28%|██▊       | 1804/6470 [20:41<50:26,  1.54it/s]
3|1805|Loss: 0.8596188426017761:  28%|██▊       | 1804/6470 [20:41<50:26,  1.54it/s]
3|1805|Loss: 0.8596188426017761:  28%|██▊       | 1805/6470 [20:42<52:57,  1.47it/s]
3|1806|Loss: 0.9496404528617859:  28%|██▊       | 1805/6470 [20:42<52:57,  1.47it/s]
3|1806|Loss: 0.9496404528617859:  28%|██▊       | 1806/6470 [20:42<52:10,  1.49it/s]
3|1807|Loss: 0.9600861668586731:  28%|██▊       | 1806/6470 [20:43<52:10,  1.49it/s]
3|1807|Loss: 0.9600861668586731:  28%|██▊       | 1807/6470 [20:43<51:11,  1.52it/s]
3|1808|Loss: 0.8691676259040833:  28%|██▊       | 1807/6470 [20:43<51:11,  1.52it/s]
3|1808|Loss: 0.8691676259040833:  28%|██▊       | 1808/6470 [20:44<50:46,  1.53it/s]
3|1809|Loss: 0.7349193692207336:  28%|██▊       | 1808/6470 [20:44<50:46,  1.53it/s]
3|1809|Loss: 0.7349193692207336:  28%|██▊       | 1809/6470 [20:45<54:32,  1.42it/s]
3|1810|Loss: 1.0513392686843872:  28%|██▊       | 1809/6470 [20:45<54:32,  1.42it/s]
3|1810|Loss: 1.0513392686843872:  28%|██▊       | 1810/6470 [20:45<54:12,  1.43it/s]
3|1811|Loss: 0.766364574432373:  28%|██▊       | 1810/6470 [20:46<54:12,  1.43it/s] 
3|1811|Loss: 0.766364574432373:  28%|██▊       | 1811/6470 [20:46<54:40,  1.42it/s]
3|1812|Loss: 0.9066463112831116:  28%|██▊       | 1811/6470 [20:46<54:40,  1.42it/s]
3|1812|Loss: 0.9066463112831116:  28%|██▊       | 1812/6470 [20:47<52:50,  1.47it/s]
3|1813|Loss: 0.8474878668785095:  28%|██▊       | 1812/6470 [20:47<52:50,  1.47it/s]
3|1813|Loss: 0.8474878668785095:  28%|██▊       | 1813/6470 [20:47<52:48,  1.47it/s]
3|1814|Loss: 0.8687387108802795:  28%|██▊       | 1813/6470 [20:48<52:48,  1.47it/s]
3|1814|Loss: 0.8687387108802795:  28%|██▊       | 1814/6470 [20:48<52:10,  1.49it/s]
3|1815|Loss: 0.7660748958587646:  28%|██▊       | 1814/6470 [20:48<52:10,  1.49it/s]
3|1815|Loss: 0.7660748958587646:  28%|██▊       | 1815/6470 [20:49<53:27,  1.45it/s]
3|1816|Loss: 0.9168105721473694:  28%|██▊       | 1815/6470 [20:49<53:27,  1.45it/s]
3|1816|Loss: 0.9168105721473694:  28%|██▊       | 1816/6470 [20:49<56:24,  1.38it/s]
3|1817|Loss: 0.9071753621101379:  28%|██▊       | 1816/6470 [20:50<56:24,  1.38it/s]
3|1817|Loss: 0.9071753621101379:  28%|██▊       | 1817/6470 [20:50<58:42,  1.32it/s]
3|1818|Loss: 0.9345443248748779:  28%|██▊       | 1817/6470 [20:51<58:42,  1.32it/s]
3|1818|Loss: 0.9345443248748779:  28%|██▊       | 1818/6470 [20:51<58:38,  1.32it/s]
3|1819|Loss: 0.8074546456336975:  28%|██▊       | 1818/6470 [20:51<58:38,  1.32it/s]
3|1819|Loss: 0.8074546456336975:  28%|██▊       | 1819/6470 [20:52<54:37,  1.42it/s]
3|1820|Loss: 0.8480391502380371:  28%|██▊       | 1819/6470 [20:52<54:37,  1.42it/s]
3|1820|Loss: 0.8480391502380371:  28%|██▊       | 1820/6470 [20:52<53:55,  1.44it/s]
3|1821|Loss: 0.8901965618133545:  28%|██▊       | 1820/6470 [20:53<53:55,  1.44it/s]
3|1821|Loss: 0.8901965618133545:  28%|██▊       | 1821/6470 [20:53<53:01,  1.46it/s]
3|1822|Loss: 0.9286235570907593:  28%|██▊       | 1821/6470 [20:53<53:01,  1.46it/s]
3|1822|Loss: 0.9286235570907593:  28%|██▊       | 1822/6470 [20:54<54:52,  1.41it/s]
3|1823|Loss: 0.796168863773346:  28%|██▊       | 1822/6470 [20:54<54:52,  1.41it/s] 
3|1823|Loss: 0.796168863773346:  28%|██▊       | 1823/6470 [20:54<55:12,  1.40it/s]
3|1824|Loss: 0.9120535850524902:  28%|██▊       | 1823/6470 [20:55<55:12,  1.40it/s]
3|1824|Loss: 0.9120535850524902:  28%|██▊       | 1824/6470 [20:55<57:23,  1.35it/s]
3|1825|Loss: 0.7557684779167175:  28%|██▊       | 1824/6470 [20:56<57:23,  1.35it/s]
3|1825|Loss: 0.7557684779167175:  28%|██▊       | 1825/6470 [20:56<57:09,  1.35it/s]
3|1826|Loss: 1.008725643157959:  28%|██▊       | 1825/6470 [20:56<57:09,  1.35it/s] 
3|1826|Loss: 1.008725643157959:  28%|██▊       | 1826/6470 [20:57<57:07,  1.35it/s]
3|1827|Loss: 0.987456738948822:  28%|██▊       | 1826/6470 [20:57<57:07,  1.35it/s]
3|1827|Loss: 0.987456738948822:  28%|██▊       | 1827/6470 [20:57<52:01,  1.49it/s]
3|1828|Loss: 0.7756604552268982:  28%|██▊       | 1827/6470 [20:58<52:01,  1.49it/s]
3|1828|Loss: 0.7756604552268982:  28%|██▊       | 1828/6470 [20:58<52:12,  1.48it/s]
3|1829|Loss: 0.7805338501930237:  28%|██▊       | 1828/6470 [20:58<52:12,  1.48it/s]
3|1829|Loss: 0.7805338501930237:  28%|██▊       | 1829/6470 [20:59<53:52,  1.44it/s]
3|1830|Loss: 0.8361050486564636:  28%|██▊       | 1829/6470 [20:59<53:52,  1.44it/s]
3|1830|Loss: 0.8361050486564636:  28%|██▊       | 1830/6470 [20:59<50:57,  1.52it/s]
3|1831|Loss: 0.8245013952255249:  28%|██▊       | 1830/6470 [21:00<50:57,  1.52it/s]
3|1831|Loss: 0.8245013952255249:  28%|██▊       | 1831/6470 [21:00<52:58,  1.46it/s]
3|1832|Loss: 1.0170773267745972:  28%|██▊       | 1831/6470 [21:00<52:58,  1.46it/s]
3|1832|Loss: 1.0170773267745972:  28%|██▊       | 1832/6470 [21:01<52:23,  1.48it/s]
3|1833|Loss: 0.8140045404434204:  28%|██▊       | 1832/6470 [21:01<52:23,  1.48it/s]
3|1833|Loss: 0.8140045404434204:  28%|██▊       | 1833/6470 [21:01<50:18,  1.54it/s]
3|1834|Loss: 0.9053390622138977:  28%|██▊       | 1833/6470 [21:02<50:18,  1.54it/s]
3|1834|Loss: 0.9053390622138977:  28%|██▊       | 1834/6470 [21:02<49:30,  1.56it/s]
3|1835|Loss: 0.9595195055007935:  28%|██▊       | 1834/6470 [21:02<49:30,  1.56it/s]
3|1835|Loss: 0.9595195055007935:  28%|██▊       | 1835/6470 [21:02<49:00,  1.58it/s]
3|1836|Loss: 1.0456401109695435:  28%|██▊       | 1835/6470 [21:03<49:00,  1.58it/s]
3|1836|Loss: 1.0456401109695435:  28%|██▊       | 1836/6470 [21:03<49:07,  1.57it/s]
3|1837|Loss: 0.9223621487617493:  28%|██▊       | 1836/6470 [21:03<49:07,  1.57it/s]
3|1837|Loss: 0.9223621487617493:  28%|██▊       | 1837/6470 [21:04<50:24,  1.53it/s]
3|1838|Loss: 0.932837188243866:  28%|██▊       | 1837/6470 [21:04<50:24,  1.53it/s] 
3|1838|Loss: 0.932837188243866:  28%|██▊       | 1838/6470 [21:04<49:19,  1.57it/s]
3|1839|Loss: 0.6746095418930054:  28%|██▊       | 1838/6470 [21:05<49:19,  1.57it/s]
3|1839|Loss: 0.6746095418930054:  28%|██▊       | 1839/6470 [21:05<47:22,  1.63it/s]
3|1840|Loss: 0.8032346367835999:  28%|██▊       | 1839/6470 [21:05<47:22,  1.63it/s]
3|1840|Loss: 0.8032346367835999:  28%|██▊       | 1840/6470 [21:06<46:15,  1.67it/s]
3|1841|Loss: 0.8658890128135681:  28%|██▊       | 1840/6470 [21:06<46:15,  1.67it/s]
3|1841|Loss: 0.8658890128135681:  28%|██▊       | 1841/6470 [21:06<48:46,  1.58it/s]
3|1842|Loss: 0.8181332945823669:  28%|██▊       | 1841/6470 [21:07<48:46,  1.58it/s]
3|1842|Loss: 0.8181332945823669:  28%|██▊       | 1842/6470 [21:07<50:40,  1.52it/s]
3|1843|Loss: 1.058112621307373:  28%|██▊       | 1842/6470 [21:07<50:40,  1.52it/s] 
3|1843|Loss: 1.058112621307373:  28%|██▊       | 1843/6470 [21:08<51:53,  1.49it/s]
3|1844|Loss: 0.8980007171630859:  28%|██▊       | 1843/6470 [21:08<51:53,  1.49it/s]
3|1844|Loss: 0.8980007171630859:  29%|██▊       | 1844/6470 [21:08<52:32,  1.47it/s]
3|1845|Loss: 0.9227173328399658:  29%|██▊       | 1844/6470 [21:09<52:32,  1.47it/s]
3|1845|Loss: 0.9227173328399658:  29%|██▊       | 1845/6470 [21:09<52:56,  1.46it/s]
3|1846|Loss: 0.8848817348480225:  29%|██▊       | 1845/6470 [21:09<52:56,  1.46it/s]
3|1846|Loss: 0.8848817348480225:  29%|██▊       | 1846/6470 [21:10<49:14,  1.57it/s]
3|1847|Loss: 0.8998995423316956:  29%|██▊       | 1846/6470 [21:10<49:14,  1.57it/s]
3|1847|Loss: 0.8998995423316956:  29%|██▊       | 1847/6470 [21:10<46:53,  1.64it/s]
3|1848|Loss: 0.8442189693450928:  29%|██▊       | 1847/6470 [21:10<46:53,  1.64it/s]
3|1848|Loss: 0.8442189693450928:  29%|██▊       | 1848/6470 [21:11<48:59,  1.57it/s]
3|1849|Loss: 0.9651837348937988:  29%|██▊       | 1848/6470 [21:11<48:59,  1.57it/s]
3|1849|Loss: 0.9651837348937988:  29%|██▊       | 1849/6470 [21:11<47:56,  1.61it/s]
3|1850|Loss: 0.7248210906982422:  29%|██▊       | 1849/6470 [21:12<47:56,  1.61it/s]
3|1850|Loss: 0.7248210906982422:  29%|██▊       | 1850/6470 [21:12<45:37,  1.69it/s]
3|1851|Loss: 0.8689310550689697:  29%|██▊       | 1850/6470 [21:12<45:37,  1.69it/s]
3|1851|Loss: 0.8689310550689697:  29%|██▊       | 1851/6470 [21:13<47:59,  1.60it/s]
3|1852|Loss: 0.8138154149055481:  29%|██▊       | 1851/6470 [21:13<47:59,  1.60it/s]
3|1852|Loss: 0.8138154149055481:  29%|██▊       | 1852/6470 [21:13<48:45,  1.58it/s]
3|1853|Loss: 1.0596873760223389:  29%|██▊       | 1852/6470 [21:14<48:45,  1.58it/s]
3|1853|Loss: 1.0596873760223389:  29%|██▊       | 1853/6470 [21:14<52:46,  1.46it/s]
3|1854|Loss: 0.857517659664154:  29%|██▊       | 1853/6470 [21:15<52:46,  1.46it/s] 
3|1854|Loss: 0.857517659664154:  29%|██▊       | 1854/6470 [21:15<53:58,  1.43it/s]
3|1855|Loss: 0.7897226214408875:  29%|██▊       | 1854/6470 [21:15<53:58,  1.43it/s]
3|1855|Loss: 0.7897226214408875:  29%|██▊       | 1855/6470 [21:15<47:32,  1.62it/s]
3|1856|Loss: 0.784051775932312:  29%|██▊       | 1855/6470 [21:16<47:32,  1.62it/s] 
3|1856|Loss: 0.784051775932312:  29%|██▊       | 1856/6470 [21:16<48:06,  1.60it/s]
3|1857|Loss: 0.7626641988754272:  29%|██▊       | 1856/6470 [21:16<48:06,  1.60it/s]
3|1857|Loss: 0.7626641988754272:  29%|██▊       | 1857/6470 [21:17<50:22,  1.53it/s]
3|1858|Loss: 0.8904579877853394:  29%|██▊       | 1857/6470 [21:17<50:22,  1.53it/s]
3|1858|Loss: 0.8904579877853394:  29%|██▊       | 1858/6470 [21:17<50:08,  1.53it/s]
3|1859|Loss: 0.9013306498527527:  29%|██▊       | 1858/6470 [21:18<50:08,  1.53it/s]
3|1859|Loss: 0.9013306498527527:  29%|██▊       | 1859/6470 [21:18<49:47,  1.54it/s]
3|1860|Loss: 0.768159031867981:  29%|██▊       | 1859/6470 [21:18<49:47,  1.54it/s] 
3|1860|Loss: 0.768159031867981:  29%|██▊       | 1860/6470 [21:18<48:24,  1.59it/s]
3|1861|Loss: 0.657377302646637:  29%|██▊       | 1860/6470 [21:19<48:24,  1.59it/s]
3|1861|Loss: 0.657377302646637:  29%|██▉       | 1861/6470 [21:19<46:50,  1.64it/s]
3|1862|Loss: 0.7807034850120544:  29%|██▉       | 1861/6470 [21:19<46:50,  1.64it/s]
3|1862|Loss: 0.7807034850120544:  29%|██▉       | 1862/6470 [21:20<51:46,  1.48it/s]
3|1863|Loss: 1.0224218368530273:  29%|██▉       | 1862/6470 [21:20<51:46,  1.48it/s]
3|1863|Loss: 1.0224218368530273:  29%|██▉       | 1863/6470 [21:21<51:25,  1.49it/s]
3|1864|Loss: 0.7625795006752014:  29%|██▉       | 1863/6470 [21:21<51:25,  1.49it/s]
3|1864|Loss: 0.7625795006752014:  29%|██▉       | 1864/6470 [21:21<54:29,  1.41it/s]
3|1865|Loss: 0.8684319853782654:  29%|██▉       | 1864/6470 [21:22<54:29,  1.41it/s]
3|1865|Loss: 0.8684319853782654:  29%|██▉       | 1865/6470 [21:22<53:00,  1.45it/s]
3|1866|Loss: 0.8459853529930115:  29%|██▉       | 1865/6470 [21:22<53:00,  1.45it/s]
3|1866|Loss: 0.8459853529930115:  29%|██▉       | 1866/6470 [21:23<51:35,  1.49it/s]
3|1867|Loss: 0.8580163717269897:  29%|██▉       | 1866/6470 [21:23<51:35,  1.49it/s]
3|1867|Loss: 0.8580163717269897:  29%|██▉       | 1867/6470 [21:23<50:56,  1.51it/s]
3|1868|Loss: 0.7947227358818054:  29%|██▉       | 1867/6470 [21:24<50:56,  1.51it/s]
3|1868|Loss: 0.7947227358818054:  29%|██▉       | 1868/6470 [21:24<52:04,  1.47it/s]
3|1869|Loss: 0.8516348004341125:  29%|██▉       | 1868/6470 [21:24<52:04,  1.47it/s]
3|1869|Loss: 0.8516348004341125:  29%|██▉       | 1869/6470 [21:25<52:39,  1.46it/s]
3|1870|Loss: 0.8329170346260071:  29%|██▉       | 1869/6470 [21:25<52:39,  1.46it/s]
3|1870|Loss: 0.8329170346260071:  29%|██▉       | 1870/6470 [21:25<48:59,  1.57it/s]
3|1871|Loss: 0.8904085755348206:  29%|██▉       | 1870/6470 [21:26<48:59,  1.57it/s]
3|1871|Loss: 0.8904085755348206:  29%|██▉       | 1871/6470 [21:26<52:13,  1.47it/s]
3|1872|Loss: 0.996910810470581:  29%|██▉       | 1871/6470 [21:26<52:13,  1.47it/s] 
3|1872|Loss: 0.996910810470581:  29%|██▉       | 1872/6470 [21:27<49:38,  1.54it/s]
3|1873|Loss: 0.7482033967971802:  29%|██▉       | 1872/6470 [21:27<49:38,  1.54it/s]
3|1873|Loss: 0.7482033967971802:  29%|██▉       | 1873/6470 [21:27<48:13,  1.59it/s]
3|1874|Loss: 0.8745170831680298:  29%|██▉       | 1873/6470 [21:28<48:13,  1.59it/s]
3|1874|Loss: 0.8745170831680298:  29%|██▉       | 1874/6470 [21:28<50:25,  1.52it/s]
3|1875|Loss: 0.8326689004898071:  29%|██▉       | 1874/6470 [21:28<50:25,  1.52it/s]
3|1875|Loss: 0.8326689004898071:  29%|██▉       | 1875/6470 [21:29<53:34,  1.43it/s]
3|1876|Loss: 0.8070904612541199:  29%|██▉       | 1875/6470 [21:29<53:34,  1.43it/s]
3|1876|Loss: 0.8070904612541199:  29%|██▉       | 1876/6470 [21:29<54:59,  1.39it/s]
3|1877|Loss: 0.8937349915504456:  29%|██▉       | 1876/6470 [21:30<54:59,  1.39it/s]
3|1877|Loss: 0.8937349915504456:  29%|██▉       | 1877/6470 [21:30<55:38,  1.38it/s]
3|1878|Loss: 0.853492021560669:  29%|██▉       | 1877/6470 [21:31<55:38,  1.38it/s] 
3|1878|Loss: 0.853492021560669:  29%|██▉       | 1878/6470 [21:31<55:20,  1.38it/s]
3|1879|Loss: 0.8880175948143005:  29%|██▉       | 1878/6470 [21:31<55:20,  1.38it/s]
3|1879|Loss: 0.8880175948143005:  29%|██▉       | 1879/6470 [21:31<51:16,  1.49it/s]
3|1880|Loss: 0.7558183670043945:  29%|██▉       | 1879/6470 [21:32<51:16,  1.49it/s]
3|1880|Loss: 0.7558183670043945:  29%|██▉       | 1880/6470 [21:32<51:05,  1.50it/s]
3|1881|Loss: 0.7873360514640808:  29%|██▉       | 1880/6470 [21:33<51:05,  1.50it/s]
3|1881|Loss: 0.7873360514640808:  29%|██▉       | 1881/6470 [21:33<54:50,  1.39it/s]
3|1882|Loss: 0.9034140110015869:  29%|██▉       | 1881/6470 [21:33<54:50,  1.39it/s]
3|1882|Loss: 0.9034140110015869:  29%|██▉       | 1882/6470 [21:34<55:56,  1.37it/s]
3|1883|Loss: 0.8126136064529419:  29%|██▉       | 1882/6470 [21:34<55:56,  1.37it/s]
3|1883|Loss: 0.8126136064529419:  29%|██▉       | 1883/6470 [21:34<53:52,  1.42it/s]
3|1884|Loss: 0.9481614232063293:  29%|██▉       | 1883/6470 [21:35<53:52,  1.42it/s]
3|1884|Loss: 0.9481614232063293:  29%|██▉       | 1884/6470 [21:35<56:18,  1.36it/s]
3|1885|Loss: 0.9249975681304932:  29%|██▉       | 1884/6470 [21:36<56:18,  1.36it/s]
3|1885|Loss: 0.9249975681304932:  29%|██▉       | 1885/6470 [21:36<56:24,  1.35it/s]
3|1886|Loss: 1.0528336763381958:  29%|██▉       | 1885/6470 [21:36<56:24,  1.35it/s]
3|1886|Loss: 1.0528336763381958:  29%|██▉       | 1886/6470 [21:37<56:25,  1.35it/s]
3|1887|Loss: 0.841756284236908:  29%|██▉       | 1886/6470 [21:37<56:25,  1.35it/s] 
3|1887|Loss: 0.841756284236908:  29%|██▉       | 1887/6470 [21:37<58:07,  1.31it/s]
3|1888|Loss: 0.7686874270439148:  29%|██▉       | 1887/6470 [21:38<58:07,  1.31it/s]
3|1888|Loss: 0.7686874270439148:  29%|██▉       | 1888/6470 [21:38<55:14,  1.38it/s]
3|1889|Loss: 0.7077242732048035:  29%|██▉       | 1888/6470 [21:38<55:14,  1.38it/s]
3|1889|Loss: 0.7077242732048035:  29%|██▉       | 1889/6470 [21:39<55:54,  1.37it/s]
3|1890|Loss: 0.8535749316215515:  29%|██▉       | 1889/6470 [21:39<55:54,  1.37it/s]
3|1890|Loss: 0.8535749316215515:  29%|██▉       | 1890/6470 [21:39<54:33,  1.40it/s]
3|1891|Loss: 1.0237675905227661:  29%|██▉       | 1890/6470 [21:40<54:33,  1.40it/s]
3|1891|Loss: 1.0237675905227661:  29%|██▉       | 1891/6470 [21:40<54:01,  1.41it/s]
3|1892|Loss: 0.9466374516487122:  29%|██▉       | 1891/6470 [21:41<54:01,  1.41it/s]
3|1892|Loss: 0.9466374516487122:  29%|██▉       | 1892/6470 [21:41<54:49,  1.39it/s]
3|1893|Loss: 0.7942089438438416:  29%|██▉       | 1892/6470 [21:41<54:49,  1.39it/s]
3|1893|Loss: 0.7942089438438416:  29%|██▉       | 1893/6470 [21:42<54:51,  1.39it/s]
3|1894|Loss: 0.9568297863006592:  29%|██▉       | 1893/6470 [21:42<54:51,  1.39it/s]
3|1894|Loss: 0.9568297863006592:  29%|██▉       | 1894/6470 [21:42<55:00,  1.39it/s]
3|1895|Loss: 0.8506125807762146:  29%|██▉       | 1894/6470 [21:43<55:00,  1.39it/s]
3|1895|Loss: 0.8506125807762146:  29%|██▉       | 1895/6470 [21:43<57:29,  1.33it/s]
3|1896|Loss: 0.8508909344673157:  29%|██▉       | 1895/6470 [21:44<57:29,  1.33it/s]
3|1896|Loss: 0.8508909344673157:  29%|██▉       | 1896/6470 [21:44<1:00:20,  1.26it/s]
3|1897|Loss: 0.803816556930542:  29%|██▉       | 1896/6470 [21:45<1:00:20,  1.26it/s] 
3|1897|Loss: 0.803816556930542:  29%|██▉       | 1897/6470 [21:45<59:20,  1.28it/s]  
3|1898|Loss: 0.9542297720909119:  29%|██▉       | 1897/6470 [21:45<59:20,  1.28it/s]
3|1898|Loss: 0.9542297720909119:  29%|██▉       | 1898/6470 [21:45<56:06,  1.36it/s]
3|1899|Loss: 0.8179835081100464:  29%|██▉       | 1898/6470 [21:46<56:06,  1.36it/s]
3|1899|Loss: 0.8179835081100464:  29%|██▉       | 1899/6470 [21:46<54:50,  1.39it/s]
3|1900|Loss: 0.7114158272743225:  29%|██▉       | 1899/6470 [21:47<54:50,  1.39it/s]
3|1900|Loss: 0.7114158272743225:  29%|██▉       | 1900/6470 [21:47<54:29,  1.40it/s]
3|1901|Loss: 0.9070780277252197:  29%|██▉       | 1900/6470 [21:47<54:29,  1.40it/s]
3|1901|Loss: 0.9070780277252197:  29%|██▉       | 1901/6470 [21:48<53:03,  1.44it/s]
3|1902|Loss: 0.7836223840713501:  29%|██▉       | 1901/6470 [21:48<53:03,  1.44it/s]
3|1902|Loss: 0.7836223840713501:  29%|██▉       | 1902/6470 [21:48<51:46,  1.47it/s]
3|1903|Loss: 0.8326345682144165:  29%|██▉       | 1902/6470 [21:49<51:46,  1.47it/s]
3|1903|Loss: 0.8326345682144165:  29%|██▉       | 1903/6470 [21:49<50:57,  1.49it/s]
3|1904|Loss: 0.9355916380882263:  29%|██▉       | 1903/6470 [21:49<50:57,  1.49it/s]
3|1904|Loss: 0.9355916380882263:  29%|██▉       | 1904/6470 [21:49<50:22,  1.51it/s]
3|1905|Loss: 0.9373844265937805:  29%|██▉       | 1904/6470 [21:50<50:22,  1.51it/s]
3|1905|Loss: 0.9373844265937805:  29%|██▉       | 1905/6470 [21:50<48:49,  1.56it/s]
3|1906|Loss: 0.8331494927406311:  29%|██▉       | 1905/6470 [21:50<48:49,  1.56it/s]
3|1906|Loss: 0.8331494927406311:  29%|██▉       | 1906/6470 [21:51<47:24,  1.60it/s]
3|1907|Loss: 0.9421928524971008:  29%|██▉       | 1906/6470 [21:51<47:24,  1.60it/s]
3|1907|Loss: 0.9421928524971008:  29%|██▉       | 1907/6470 [21:51<50:32,  1.50it/s]
3|1908|Loss: 0.8010630011558533:  29%|██▉       | 1907/6470 [21:52<50:32,  1.50it/s]
3|1908|Loss: 0.8010630011558533:  29%|██▉       | 1908/6470 [21:52<48:42,  1.56it/s]
3|1909|Loss: 0.9143792986869812:  29%|██▉       | 1908/6470 [21:52<48:42,  1.56it/s]
3|1909|Loss: 0.9143792986869812:  30%|██▉       | 1909/6470 [21:53<47:36,  1.60it/s]
3|1910|Loss: 0.8154166340827942:  30%|██▉       | 1909/6470 [21:53<47:36,  1.60it/s]
3|1910|Loss: 0.8154166340827942:  30%|██▉       | 1910/6470 [21:53<46:46,  1.62it/s]
3|1911|Loss: 0.8993358612060547:  30%|██▉       | 1910/6470 [21:54<46:46,  1.62it/s]
3|1911|Loss: 0.8993358612060547:  30%|██▉       | 1911/6470 [21:54<49:50,  1.52it/s]
3|1912|Loss: 0.8713081479072571:  30%|██▉       | 1911/6470 [21:54<49:50,  1.52it/s]
3|1912|Loss: 0.8713081479072571:  30%|██▉       | 1912/6470 [21:54<46:56,  1.62it/s]
3|1913|Loss: 0.8232590556144714:  30%|██▉       | 1912/6470 [21:55<46:56,  1.62it/s]
3|1913|Loss: 0.8232590556144714:  30%|██▉       | 1913/6470 [21:55<46:39,  1.63it/s]
3|1914|Loss: 0.9002378582954407:  30%|██▉       | 1913/6470 [21:55<46:39,  1.63it/s]
3|1914|Loss: 0.9002378582954407:  30%|██▉       | 1914/6470 [21:56<44:59,  1.69it/s]
3|1915|Loss: 0.7888267636299133:  30%|██▉       | 1914/6470 [21:56<44:59,  1.69it/s]
3|1915|Loss: 0.7888267636299133:  30%|██▉       | 1915/6470 [21:56<46:35,  1.63it/s]
3|1916|Loss: 0.863150417804718:  30%|██▉       | 1915/6470 [21:57<46:35,  1.63it/s] 
3|1916|Loss: 0.863150417804718:  30%|██▉       | 1916/6470 [21:57<47:09,  1.61it/s]
3|1917|Loss: 0.716583788394928:  30%|██▉       | 1916/6470 [21:57<47:09,  1.61it/s]
3|1917|Loss: 0.716583788394928:  30%|██▉       | 1917/6470 [21:58<51:26,  1.48it/s]
3|1918|Loss: 0.973513126373291:  30%|██▉       | 1917/6470 [21:58<51:26,  1.48it/s]
3|1918|Loss: 0.973513126373291:  30%|██▉       | 1918/6470 [21:59<54:41,  1.39it/s]
3|1919|Loss: 0.9591848254203796:  30%|██▉       | 1918/6470 [21:59<54:41,  1.39it/s]
3|1919|Loss: 0.9591848254203796:  30%|██▉       | 1919/6470 [21:59<55:34,  1.36it/s]
3|1920|Loss: 0.9108215570449829:  30%|██▉       | 1919/6470 [22:00<55:34,  1.36it/s]
3|1920|Loss: 0.9108215570449829:  30%|██▉       | 1920/6470 [22:00<53:38,  1.41it/s]
3|1921|Loss: 0.888973593711853:  30%|██▉       | 1920/6470 [22:00<53:38,  1.41it/s] 
3|1921|Loss: 0.888973593711853:  30%|██▉       | 1921/6470 [22:01<54:02,  1.40it/s]
3|1922|Loss: 0.832144558429718:  30%|██▉       | 1921/6470 [22:01<54:02,  1.40it/s]
3|1922|Loss: 0.832144558429718:  30%|██▉       | 1922/6470 [22:01<51:52,  1.46it/s]
3|1923|Loss: 0.875575602054596:  30%|██▉       | 1922/6470 [22:02<51:52,  1.46it/s]
3|1923|Loss: 0.875575602054596:  30%|██▉       | 1923/6470 [22:02<52:11,  1.45it/s]
3|1924|Loss: 0.8002001047134399:  30%|██▉       | 1923/6470 [22:02<52:11,  1.45it/s]
3|1924|Loss: 0.8002001047134399:  30%|██▉       | 1924/6470 [22:03<53:17,  1.42it/s]
3|1925|Loss: 0.7319434881210327:  30%|██▉       | 1924/6470 [22:03<53:17,  1.42it/s]
3|1925|Loss: 0.7319434881210327:  30%|██▉       | 1925/6470 [22:04<58:39,  1.29it/s]
3|1926|Loss: 0.8630001544952393:  30%|██▉       | 1925/6470 [22:04<58:39,  1.29it/s]
3|1926|Loss: 0.8630001544952393:  30%|██▉       | 1926/6470 [22:04<56:54,  1.33it/s]
3|1927|Loss: 0.726676881313324:  30%|██▉       | 1926/6470 [22:05<56:54,  1.33it/s] 
3|1927|Loss: 0.726676881313324:  30%|██▉       | 1927/6470 [22:05<58:49,  1.29it/s]
3|1928|Loss: 0.8719991445541382:  30%|██▉       | 1927/6470 [22:06<58:49,  1.29it/s]
3|1928|Loss: 0.8719991445541382:  30%|██▉       | 1928/6470 [22:06<55:03,  1.38it/s]
3|1929|Loss: 0.7019340395927429:  30%|██▉       | 1928/6470 [22:06<55:03,  1.38it/s]
3|1929|Loss: 0.7019340395927429:  30%|██▉       | 1929/6470 [22:06<52:03,  1.45it/s]
3|1930|Loss: 0.7211223244667053:  30%|██▉       | 1929/6470 [22:07<52:03,  1.45it/s]
3|1930|Loss: 0.7211223244667053:  30%|██▉       | 1930/6470 [22:07<55:45,  1.36it/s]
3|1931|Loss: 0.9909497499465942:  30%|██▉       | 1930/6470 [22:08<55:45,  1.36it/s]
3|1931|Loss: 0.9909497499465942:  30%|██▉       | 1931/6470 [22:08<56:51,  1.33it/s]
3|1932|Loss: 0.9097225069999695:  30%|██▉       | 1931/6470 [22:08<56:51,  1.33it/s]
3|1932|Loss: 0.9097225069999695:  30%|██▉       | 1932/6470 [22:09<56:50,  1.33it/s]
3|1933|Loss: 0.7229918241500854:  30%|██▉       | 1932/6470 [22:09<56:50,  1.33it/s]
3|1933|Loss: 0.7229918241500854:  30%|██▉       | 1933/6470 [22:09<51:22,  1.47it/s]
3|1934|Loss: 0.9120455384254456:  30%|██▉       | 1933/6470 [22:10<51:22,  1.47it/s]
3|1934|Loss: 0.9120455384254456:  30%|██▉       | 1934/6470 [22:10<49:31,  1.53it/s]
3|1935|Loss: 0.7926326394081116:  30%|██▉       | 1934/6470 [22:10<49:31,  1.53it/s]
3|1935|Loss: 0.7926326394081116:  30%|██▉       | 1935/6470 [22:11<48:51,  1.55it/s]
3|1936|Loss: 0.9136239886283875:  30%|██▉       | 1935/6470 [22:11<48:51,  1.55it/s]
3|1936|Loss: 0.9136239886283875:  30%|██▉       | 1936/6470 [22:11<47:41,  1.58it/s]
3|1937|Loss: 0.9574733972549438:  30%|██▉       | 1936/6470 [22:11<47:41,  1.58it/s]
3|1937|Loss: 0.9574733972549438:  30%|██▉       | 1937/6470 [22:12<45:14,  1.67it/s]
3|1938|Loss: 0.8150909543037415:  30%|██▉       | 1937/6470 [22:12<45:14,  1.67it/s]
3|1938|Loss: 0.8150909543037415:  30%|██▉       | 1938/6470 [22:12<44:38,  1.69it/s]
3|1939|Loss: 0.7954761981964111:  30%|██▉       | 1938/6470 [22:13<44:38,  1.69it/s]
3|1939|Loss: 0.7954761981964111:  30%|██▉       | 1939/6470 [22:13<47:20,  1.60it/s]
3|1940|Loss: 0.8573968410491943:  30%|██▉       | 1939/6470 [22:13<47:20,  1.60it/s]
3|1940|Loss: 0.8573968410491943:  30%|██▉       | 1940/6470 [22:14<46:57,  1.61it/s]
3|1941|Loss: 0.7553361654281616:  30%|██▉       | 1940/6470 [22:14<46:57,  1.61it/s]
3|1941|Loss: 0.7553361654281616:  30%|███       | 1941/6470 [22:14<48:58,  1.54it/s]
3|1942|Loss: 0.7975507974624634:  30%|███       | 1941/6470 [22:15<48:58,  1.54it/s]
3|1942|Loss: 0.7975507974624634:  30%|███       | 1942/6470 [22:15<48:57,  1.54it/s]
3|1943|Loss: 0.8053049445152283:  30%|███       | 1942/6470 [22:15<48:57,  1.54it/s]
3|1943|Loss: 0.8053049445152283:  30%|███       | 1943/6470 [22:16<50:38,  1.49it/s]
3|1944|Loss: 0.8445833325386047:  30%|███       | 1943/6470 [22:16<50:38,  1.49it/s]
3|1944|Loss: 0.8445833325386047:  30%|███       | 1944/6470 [22:16<48:07,  1.57it/s]
3|1945|Loss: 0.7800830602645874:  30%|███       | 1944/6470 [22:16<48:07,  1.57it/s]
3|1945|Loss: 0.7800830602645874:  30%|███       | 1945/6470 [22:17<47:01,  1.60it/s]
3|1946|Loss: 0.7805051207542419:  30%|███       | 1945/6470 [22:17<47:01,  1.60it/s]
3|1946|Loss: 0.7805051207542419:  30%|███       | 1946/6470 [22:17<48:20,  1.56it/s]
3|1947|Loss: 0.952172577381134:  30%|███       | 1946/6470 [22:18<48:20,  1.56it/s] 
3|1947|Loss: 0.952172577381134:  30%|███       | 1947/6470 [22:18<51:19,  1.47it/s]
3|1948|Loss: 0.6658545136451721:  30%|███       | 1947/6470 [22:19<51:19,  1.47it/s]
3|1948|Loss: 0.6658545136451721:  30%|███       | 1948/6470 [22:19<52:49,  1.43it/s]
3|1949|Loss: 0.7642991542816162:  30%|███       | 1948/6470 [22:19<52:49,  1.43it/s]
3|1949|Loss: 0.7642991542816162:  30%|███       | 1949/6470 [22:20<58:06,  1.30it/s]
3|1950|Loss: 0.7918533086776733:  30%|███       | 1949/6470 [22:20<58:06,  1.30it/s]
3|1950|Loss: 0.7918533086776733:  30%|███       | 1950/6470 [22:21<55:52,  1.35it/s]
3|1951|Loss: 1.008089542388916:  30%|███       | 1950/6470 [22:21<55:52,  1.35it/s] 
3|1951|Loss: 1.008089542388916:  30%|███       | 1951/6470 [22:21<54:34,  1.38it/s]
3|1952|Loss: 0.9511409401893616:  30%|███       | 1951/6470 [22:22<54:34,  1.38it/s]
3|1952|Loss: 0.9511409401893616:  30%|███       | 1952/6470 [22:22<52:36,  1.43it/s]
3|1953|Loss: 0.8000100255012512:  30%|███       | 1952/6470 [22:22<52:36,  1.43it/s]
3|1953|Loss: 0.8000100255012512:  30%|███       | 1953/6470 [22:22<48:53,  1.54it/s]
3|1954|Loss: 0.9584996700286865:  30%|███       | 1953/6470 [22:23<48:53,  1.54it/s]
3|1954|Loss: 0.9584996700286865:  30%|███       | 1954/6470 [22:23<52:59,  1.42it/s]
3|1955|Loss: 0.8111440539360046:  30%|███       | 1954/6470 [22:24<52:59,  1.42it/s]
3|1955|Loss: 0.8111440539360046:  30%|███       | 1955/6470 [22:24<52:34,  1.43it/s]
3|1956|Loss: 0.8718663454055786:  30%|███       | 1955/6470 [22:24<52:34,  1.43it/s]
3|1956|Loss: 0.8718663454055786:  30%|███       | 1956/6470 [22:25<52:23,  1.44it/s]
3|1957|Loss: 0.7557751536369324:  30%|███       | 1956/6470 [22:25<52:23,  1.44it/s]
3|1957|Loss: 0.7557751536369324:  30%|███       | 1957/6470 [22:25<51:27,  1.46it/s]
3|1958|Loss: 0.9349831342697144:  30%|███       | 1957/6470 [22:26<51:27,  1.46it/s]
3|1958|Loss: 0.9349831342697144:  30%|███       | 1958/6470 [22:26<47:00,  1.60it/s]
3|1959|Loss: 0.897594153881073:  30%|███       | 1958/6470 [22:26<47:00,  1.60it/s] 
3|1959|Loss: 0.897594153881073:  30%|███       | 1959/6470 [22:26<45:57,  1.64it/s]
3|1960|Loss: 0.8841701745986938:  30%|███       | 1959/6470 [22:27<45:57,  1.64it/s]
3|1960|Loss: 0.8841701745986938:  30%|███       | 1960/6470 [22:27<48:07,  1.56it/s]
3|1961|Loss: 0.85279381275177:  30%|███       | 1960/6470 [22:27<48:07,  1.56it/s]  
3|1961|Loss: 0.85279381275177:  30%|███       | 1961/6470 [22:28<48:11,  1.56it/s]
3|1962|Loss: 0.8971278071403503:  30%|███       | 1961/6470 [22:28<48:11,  1.56it/s]
3|1962|Loss: 0.8971278071403503:  30%|███       | 1962/6470 [22:29<52:53,  1.42it/s]
3|1963|Loss: 0.810562014579773:  30%|███       | 1962/6470 [22:29<52:53,  1.42it/s] 
3|1963|Loss: 0.810562014579773:  30%|███       | 1963/6470 [22:29<50:30,  1.49it/s]
3|1964|Loss: 0.8518553376197815:  30%|███       | 1963/6470 [22:29<50:30,  1.49it/s]
3|1964|Loss: 0.8518553376197815:  30%|███       | 1964/6470 [22:30<49:11,  1.53it/s]
3|1965|Loss: 0.9779112935066223:  30%|███       | 1964/6470 [22:30<49:11,  1.53it/s]
3|1965|Loss: 0.9779112935066223:  30%|███       | 1965/6470 [22:30<50:33,  1.49it/s]
3|1966|Loss: 0.8037826418876648:  30%|███       | 1965/6470 [22:31<50:33,  1.49it/s]
3|1966|Loss: 0.8037826418876648:  30%|███       | 1966/6470 [22:31<47:55,  1.57it/s]
3|1967|Loss: 0.9877710938453674:  30%|███       | 1966/6470 [22:31<47:55,  1.57it/s]
3|1967|Loss: 0.9877710938453674:  30%|███       | 1967/6470 [22:32<51:07,  1.47it/s]
3|1968|Loss: 0.9039578437805176:  30%|███       | 1967/6470 [22:32<51:07,  1.47it/s]
3|1968|Loss: 0.9039578437805176:  30%|███       | 1968/6470 [22:33<54:29,  1.38it/s]
3|1969|Loss: 0.8351601958274841:  30%|███       | 1968/6470 [22:33<54:29,  1.38it/s]
3|1969|Loss: 0.8351601958274841:  30%|███       | 1969/6470 [22:33<54:36,  1.37it/s]
3|1970|Loss: 0.8840938210487366:  30%|███       | 1969/6470 [22:34<54:36,  1.37it/s]
3|1970|Loss: 0.8840938210487366:  30%|███       | 1970/6470 [22:34<56:39,  1.32it/s]
3|1971|Loss: 0.8080193400382996:  30%|███       | 1970/6470 [22:35<56:39,  1.32it/s]
3|1971|Loss: 0.8080193400382996:  30%|███       | 1971/6470 [22:35<55:10,  1.36it/s]
3|1972|Loss: 0.8584715723991394:  30%|███       | 1971/6470 [22:35<55:10,  1.36it/s]
3|1972|Loss: 0.8584715723991394:  30%|███       | 1972/6470 [22:36<53:02,  1.41it/s]
3|1973|Loss: 0.769519031047821:  30%|███       | 1972/6470 [22:36<53:02,  1.41it/s] 
3|1973|Loss: 0.769519031047821:  30%|███       | 1973/6470 [22:36<55:31,  1.35it/s]
3|1974|Loss: 0.8017699718475342:  30%|███       | 1973/6470 [22:37<55:31,  1.35it/s]
3|1974|Loss: 0.8017699718475342:  31%|███       | 1974/6470 [22:37<56:17,  1.33it/s]
3|1975|Loss: 0.8963689804077148:  31%|███       | 1974/6470 [22:38<56:17,  1.33it/s]
3|1975|Loss: 0.8963689804077148:  31%|███       | 1975/6470 [22:38<57:19,  1.31it/s]
3|1976|Loss: 0.8561241626739502:  31%|███       | 1975/6470 [22:38<57:19,  1.31it/s]
3|1976|Loss: 0.8561241626739502:  31%|███       | 1976/6470 [22:39<56:59,  1.31it/s]
3|1977|Loss: 0.9263527989387512:  31%|███       | 1976/6470 [22:39<56:59,  1.31it/s]
3|1977|Loss: 0.9263527989387512:  31%|███       | 1977/6470 [22:39<56:24,  1.33it/s]
3|1978|Loss: 0.7842941284179688:  31%|███       | 1977/6470 [22:40<56:24,  1.33it/s]
3|1978|Loss: 0.7842941284179688:  31%|███       | 1978/6470 [22:40<57:16,  1.31it/s]
3|1979|Loss: 0.8122437596321106:  31%|███       | 1978/6470 [22:41<57:16,  1.31it/s]
3|1979|Loss: 0.8122437596321106:  31%|███       | 1979/6470 [22:41<53:40,  1.39it/s]
3|1980|Loss: 0.9256103038787842:  31%|███       | 1979/6470 [22:41<53:40,  1.39it/s]
3|1980|Loss: 0.9256103038787842:  31%|███       | 1980/6470 [22:41<52:41,  1.42it/s]
3|1981|Loss: 0.923994779586792:  31%|███       | 1980/6470 [22:42<52:41,  1.42it/s] 
3|1981|Loss: 0.923994779586792:  31%|███       | 1981/6470 [22:42<53:31,  1.40it/s]
3|1982|Loss: 0.9517765641212463:  31%|███       | 1981/6470 [22:43<53:31,  1.40it/s]
3|1982|Loss: 0.9517765641212463:  31%|███       | 1982/6470 [22:43<53:34,  1.40it/s]
3|1983|Loss: 0.8844324946403503:  31%|███       | 1982/6470 [22:43<53:34,  1.40it/s]
3|1983|Loss: 0.8844324946403503:  31%|███       | 1983/6470 [22:43<49:45,  1.50it/s]
3|1984|Loss: 0.8390429019927979:  31%|███       | 1983/6470 [22:44<49:45,  1.50it/s]
3|1984|Loss: 0.8390429019927979:  31%|███       | 1984/6470 [22:44<49:42,  1.50it/s]
3|1985|Loss: 0.7606050372123718:  31%|███       | 1984/6470 [22:45<49:42,  1.50it/s]
3|1985|Loss: 0.7606050372123718:  31%|███       | 1985/6470 [22:45<52:56,  1.41it/s]
3|1986|Loss: 0.8972266316413879:  31%|███       | 1985/6470 [22:45<52:56,  1.41it/s]
3|1986|Loss: 0.8972266316413879:  31%|███       | 1986/6470 [22:46<54:01,  1.38it/s]
3|1987|Loss: 0.9994113445281982:  31%|███       | 1986/6470 [22:46<54:01,  1.38it/s]
3|1987|Loss: 0.9994113445281982:  31%|███       | 1987/6470 [22:46<52:36,  1.42it/s]
3|1988|Loss: 0.7879675626754761:  31%|███       | 1987/6470 [22:47<52:36,  1.42it/s]
3|1988|Loss: 0.7879675626754761:  31%|███       | 1988/6470 [22:47<48:55,  1.53it/s]
3|1989|Loss: 0.9314700961112976:  31%|███       | 1988/6470 [22:47<48:55,  1.53it/s]
3|1989|Loss: 0.9314700961112976:  31%|███       | 1989/6470 [22:48<49:49,  1.50it/s]
3|1990|Loss: 0.9743776917457581:  31%|███       | 1989/6470 [22:48<49:49,  1.50it/s]
3|1990|Loss: 0.9743776917457581:  31%|███       | 1990/6470 [22:48<51:24,  1.45it/s]
3|1991|Loss: 0.9248667359352112:  31%|███       | 1990/6470 [22:49<51:24,  1.45it/s]
3|1991|Loss: 0.9248667359352112:  31%|███       | 1991/6470 [22:49<54:33,  1.37it/s]
3|1992|Loss: 0.9120965003967285:  31%|███       | 1991/6470 [22:50<54:33,  1.37it/s]
3|1992|Loss: 0.9120965003967285:  31%|███       | 1992/6470 [22:50<56:40,  1.32it/s]
3|1993|Loss: 0.7181673645973206:  31%|███       | 1992/6470 [22:50<56:40,  1.32it/s]
3|1993|Loss: 0.7181673645973206:  31%|███       | 1993/6470 [22:50<49:50,  1.50it/s]
3|1994|Loss: 0.8051507472991943:  31%|███       | 1993/6470 [22:51<49:50,  1.50it/s]
3|1994|Loss: 0.8051507472991943:  31%|███       | 1994/6470 [22:51<49:41,  1.50it/s]
3|1995|Loss: 0.9251722693443298:  31%|███       | 1994/6470 [22:52<49:41,  1.50it/s]
3|1995|Loss: 0.9251722693443298:  31%|███       | 1995/6470 [22:52<51:19,  1.45it/s]
3|1996|Loss: 0.9340383410453796:  31%|███       | 1995/6470 [22:52<51:19,  1.45it/s]
3|1996|Loss: 0.9340383410453796:  31%|███       | 1996/6470 [22:53<50:23,  1.48it/s]
3|1997|Loss: 0.8427401781082153:  31%|███       | 1996/6470 [22:53<50:23,  1.48it/s]
3|1997|Loss: 0.8427401781082153:  31%|███       | 1997/6470 [22:53<52:52,  1.41it/s]
3|1998|Loss: 0.7646126747131348:  31%|███       | 1997/6470 [22:54<52:52,  1.41it/s]
3|1998|Loss: 0.7646126747131348:  31%|███       | 1998/6470 [22:54<51:51,  1.44it/s]
3|1999|Loss: 0.9302911162376404:  31%|███       | 1998/6470 [22:54<51:51,  1.44it/s]
3|1999|Loss: 0.9302911162376404:  31%|███       | 1999/6470 [22:55<51:32,  1.45it/s]
3|2000|Loss: 0.9880408048629761:  31%|███       | 1999/6470 [22:55<51:32,  1.45it/s]
3|2000|Loss: 0.9880408048629761:  31%|███       | 2000/6470 [22:55<47:55,  1.55it/s]
3|2001|Loss: 0.8187904953956604:  31%|███       | 2000/6470 [22:56<47:55,  1.55it/s]
3|2001|Loss: 0.8187904953956604:  31%|███       | 2001/6470 [22:56<50:38,  1.47it/s]
3|2002|Loss: 0.8483371734619141:  31%|███       | 2001/6470 [22:56<50:38,  1.47it/s]
3|2002|Loss: 0.8483371734619141:  31%|███       | 2002/6470 [22:57<51:41,  1.44it/s]
3|2003|Loss: 0.7432946562767029:  31%|███       | 2002/6470 [22:57<51:41,  1.44it/s]
3|2003|Loss: 0.7432946562767029:  31%|███       | 2003/6470 [22:57<51:00,  1.46it/s]
3|2004|Loss: 0.8311602473258972:  31%|███       | 2003/6470 [22:58<51:00,  1.46it/s]
3|2004|Loss: 0.8311602473258972:  31%|███       | 2004/6470 [22:58<49:32,  1.50it/s]
3|2005|Loss: 0.8451380729675293:  31%|███       | 2004/6470 [22:58<49:32,  1.50it/s]
3|2005|Loss: 0.8451380729675293:  31%|███       | 2005/6470 [22:59<47:48,  1.56it/s]
3|2006|Loss: 0.8459123969078064:  31%|███       | 2005/6470 [22:59<47:48,  1.56it/s]
3|2006|Loss: 0.8459123969078064:  31%|███       | 2006/6470 [22:59<48:35,  1.53it/s]
3|2007|Loss: 0.7788041830062866:  31%|███       | 2006/6470 [23:00<48:35,  1.53it/s]
3|2007|Loss: 0.7788041830062866:  31%|███       | 2007/6470 [23:00<47:29,  1.57it/s]
3|2008|Loss: 0.8060346245765686:  31%|███       | 2007/6470 [23:00<47:29,  1.57it/s]
3|2008|Loss: 0.8060346245765686:  31%|███       | 2008/6470 [23:01<48:50,  1.52it/s]
3|2009|Loss: 0.9128775596618652:  31%|███       | 2008/6470 [23:01<48:50,  1.52it/s]
3|2009|Loss: 0.9128775596618652:  31%|███       | 2009/6470 [23:01<52:42,  1.41it/s]
3|2010|Loss: 0.8947191834449768:  31%|███       | 2009/6470 [23:02<52:42,  1.41it/s]
3|2010|Loss: 0.8947191834449768:  31%|███       | 2010/6470 [23:02<52:00,  1.43it/s]
3|2011|Loss: 0.7544269561767578:  31%|███       | 2010/6470 [23:02<52:00,  1.43it/s]
3|2011|Loss: 0.7544269561767578:  31%|███       | 2011/6470 [23:03<51:44,  1.44it/s]
3|2012|Loss: 0.8242109417915344:  31%|███       | 2011/6470 [23:03<51:44,  1.44it/s]
3|2012|Loss: 0.8242109417915344:  31%|███       | 2012/6470 [23:03<52:46,  1.41it/s]
3|2013|Loss: 0.8683747053146362:  31%|███       | 2012/6470 [23:04<52:46,  1.41it/s]
3|2013|Loss: 0.8683747053146362:  31%|███       | 2013/6470 [23:04<53:27,  1.39it/s]
3|2014|Loss: 0.8857998251914978:  31%|███       | 2013/6470 [23:05<53:27,  1.39it/s]
3|2014|Loss: 0.8857998251914978:  31%|███       | 2014/6470 [23:05<53:18,  1.39it/s]
3|2015|Loss: 0.9119745492935181:  31%|███       | 2014/6470 [23:05<53:18,  1.39it/s]
3|2015|Loss: 0.9119745492935181:  31%|███       | 2015/6470 [23:06<50:43,  1.46it/s]
3|2016|Loss: 0.7002629041671753:  31%|███       | 2015/6470 [23:06<50:43,  1.46it/s]
3|2016|Loss: 0.7002629041671753:  31%|███       | 2016/6470 [23:06<47:31,  1.56it/s]
3|2017|Loss: 0.8884928226470947:  31%|███       | 2016/6470 [23:06<47:31,  1.56it/s]
3|2017|Loss: 0.8884928226470947:  31%|███       | 2017/6470 [23:07<46:58,  1.58it/s]
3|2018|Loss: 0.9279294610023499:  31%|███       | 2017/6470 [23:07<46:58,  1.58it/s]
3|2018|Loss: 0.9279294610023499:  31%|███       | 2018/6470 [23:07<47:05,  1.58it/s]
3|2019|Loss: 0.8964794278144836:  31%|███       | 2018/6470 [23:08<47:05,  1.58it/s]
3|2019|Loss: 0.8964794278144836:  31%|███       | 2019/6470 [23:08<49:14,  1.51it/s]
3|2020|Loss: 0.912388265132904:  31%|███       | 2019/6470 [23:08<49:14,  1.51it/s] 
3|2020|Loss: 0.912388265132904:  31%|███       | 2020/6470 [23:09<51:00,  1.45it/s]
3|2021|Loss: 0.7378202080726624:  31%|███       | 2020/6470 [23:09<51:00,  1.45it/s]
3|2021|Loss: 0.7378202080726624:  31%|███       | 2021/6470 [23:09<50:36,  1.47it/s]
3|2022|Loss: 0.9376275539398193:  31%|███       | 2021/6470 [23:10<50:36,  1.47it/s]
3|2022|Loss: 0.9376275539398193:  31%|███▏      | 2022/6470 [23:10<49:51,  1.49it/s]
3|2023|Loss: 0.9766172766685486:  31%|███▏      | 2022/6470 [23:10<49:51,  1.49it/s]
3|2023|Loss: 0.9766172766685486:  31%|███▏      | 2023/6470 [23:11<49:13,  1.51it/s]
3|2024|Loss: 0.9367640614509583:  31%|███▏      | 2023/6470 [23:11<49:13,  1.51it/s]
3|2024|Loss: 0.9367640614509583:  31%|███▏      | 2024/6470 [23:11<49:48,  1.49it/s]
3|2025|Loss: 0.8957308530807495:  31%|███▏      | 2024/6470 [23:12<49:48,  1.49it/s]
3|2025|Loss: 0.8957308530807495:  31%|███▏      | 2025/6470 [23:12<49:29,  1.50it/s]
3|2026|Loss: 1.0859777927398682:  31%|███▏      | 2025/6470 [23:12<49:29,  1.50it/s]
3|2026|Loss: 1.0859777927398682:  31%|███▏      | 2026/6470 [23:13<50:36,  1.46it/s]
3|2027|Loss: 0.9063289165496826:  31%|███▏      | 2026/6470 [23:13<50:36,  1.46it/s]
3|2027|Loss: 0.9063289165496826:  31%|███▏      | 2027/6470 [23:14<51:39,  1.43it/s]
3|2028|Loss: 0.7530518174171448:  31%|███▏      | 2027/6470 [23:14<51:39,  1.43it/s]
3|2028|Loss: 0.7530518174171448:  31%|███▏      | 2028/6470 [23:14<49:48,  1.49it/s]
3|2029|Loss: 0.8847470879554749:  31%|███▏      | 2028/6470 [23:15<49:48,  1.49it/s]
3|2029|Loss: 0.8847470879554749:  31%|███▏      | 2029/6470 [23:15<49:47,  1.49it/s]
3|2030|Loss: 0.8473187685012817:  31%|███▏      | 2029/6470 [23:15<49:47,  1.49it/s]
3|2030|Loss: 0.8473187685012817:  31%|███▏      | 2030/6470 [23:16<51:05,  1.45it/s]
3|2031|Loss: 1.085269570350647:  31%|███▏      | 2030/6470 [23:16<51:05,  1.45it/s] 
3|2031|Loss: 1.085269570350647:  31%|███▏      | 2031/6470 [23:16<50:06,  1.48it/s]
3|2032|Loss: 0.7332873940467834:  31%|███▏      | 2031/6470 [23:17<50:06,  1.48it/s]
3|2032|Loss: 0.7332873940467834:  31%|███▏      | 2032/6470 [23:17<52:56,  1.40it/s]
3|2033|Loss: 0.8009963035583496:  31%|███▏      | 2032/6470 [23:17<52:56,  1.40it/s]
3|2033|Loss: 0.8009963035583496:  31%|███▏      | 2033/6470 [23:18<54:18,  1.36it/s]
3|2034|Loss: 0.9064613580703735:  31%|███▏      | 2033/6470 [23:18<54:18,  1.36it/s]
3|2034|Loss: 0.9064613580703735:  31%|███▏      | 2034/6470 [23:19<54:38,  1.35it/s]
3|2035|Loss: 0.76756751537323:  31%|███▏      | 2034/6470 [23:19<54:38,  1.35it/s]  
3|2035|Loss: 0.76756751537323:  31%|███▏      | 2035/6470 [23:19<54:49,  1.35it/s]
3|2036|Loss: 0.8292962312698364:  31%|███▏      | 2035/6470 [23:20<54:49,  1.35it/s]
3|2036|Loss: 0.8292962312698364:  31%|███▏      | 2036/6470 [23:20<52:50,  1.40it/s]
3|2037|Loss: 0.7160335183143616:  31%|███▏      | 2036/6470 [23:20<52:50,  1.40it/s]
3|2037|Loss: 0.7160335183143616:  31%|███▏      | 2037/6470 [23:21<50:43,  1.46it/s]
3|2038|Loss: 0.7338494658470154:  31%|███▏      | 2037/6470 [23:21<50:43,  1.46it/s]
3|2038|Loss: 0.7338494658470154:  31%|███▏      | 2038/6470 [23:21<48:31,  1.52it/s]
3|2039|Loss: 0.8911969661712646:  31%|███▏      | 2038/6470 [23:22<48:31,  1.52it/s]
3|2039|Loss: 0.8911969661712646:  32%|███▏      | 2039/6470 [23:22<49:03,  1.51it/s]
3|2040|Loss: 0.8994396924972534:  32%|███▏      | 2039/6470 [23:22<49:03,  1.51it/s]
3|2040|Loss: 0.8994396924972534:  32%|███▏      | 2040/6470 [23:23<52:53,  1.40it/s]
3|2041|Loss: 0.9109441637992859:  32%|███▏      | 2040/6470 [23:23<52:53,  1.40it/s]
3|2041|Loss: 0.9109441637992859:  32%|███▏      | 2041/6470 [23:24<56:11,  1.31it/s]
3|2042|Loss: 0.7584782838821411:  32%|███▏      | 2041/6470 [23:24<56:11,  1.31it/s]
3|2042|Loss: 0.7584782838821411:  32%|███▏      | 2042/6470 [23:24<57:13,  1.29it/s]
3|2043|Loss: 0.815866231918335:  32%|███▏      | 2042/6470 [23:25<57:13,  1.29it/s] 
3|2043|Loss: 0.815866231918335:  32%|███▏      | 2043/6470 [23:25<59:06,  1.25it/s]
3|2044|Loss: 0.9632320404052734:  32%|███▏      | 2043/6470 [23:26<59:06,  1.25it/s]
3|2044|Loss: 0.9632320404052734:  32%|███▏      | 2044/6470 [23:26<1:00:12,  1.23it/s]
3|2045|Loss: 0.731589674949646:  32%|███▏      | 2044/6470 [23:27<1:00:12,  1.23it/s] 
3|2045|Loss: 0.731589674949646:  32%|███▏      | 2045/6470 [23:27<1:01:19,  1.20it/s]
3|2046|Loss: 0.7748045325279236:  32%|███▏      | 2045/6470 [23:27<1:01:19,  1.20it/s]
3|2046|Loss: 0.7748045325279236:  32%|███▏      | 2046/6470 [23:28<59:58,  1.23it/s]  
3|2047|Loss: 0.9469329714775085:  32%|███▏      | 2046/6470 [23:28<59:58,  1.23it/s]
3|2047|Loss: 0.9469329714775085:  32%|███▏      | 2047/6470 [23:29<1:00:05,  1.23it/s]
3|2048|Loss: 0.973132312297821:  32%|███▏      | 2047/6470 [23:29<1:00:05,  1.23it/s] 
3|2048|Loss: 0.973132312297821:  32%|███▏      | 2048/6470 [23:29<58:07,  1.27it/s]  
3|2049|Loss: 0.7600172758102417:  32%|███▏      | 2048/6470 [23:30<58:07,  1.27it/s]
3|2049|Loss: 0.7600172758102417:  32%|███▏      | 2049/6470 [23:30<53:55,  1.37it/s]
3|2050|Loss: 0.9945390820503235:  32%|███▏      | 2049/6470 [23:30<53:55,  1.37it/s]
3|2050|Loss: 0.9945390820503235:  32%|███▏      | 2050/6470 [23:31<56:11,  1.31it/s]
3|2051|Loss: 0.7336224317550659:  32%|███▏      | 2050/6470 [23:31<56:11,  1.31it/s]
3|2051|Loss: 0.7336224317550659:  32%|███▏      | 2051/6470 [23:31<55:54,  1.32it/s]
3|2052|Loss: 0.8355029821395874:  32%|███▏      | 2051/6470 [23:32<55:54,  1.32it/s]
3|2052|Loss: 0.8355029821395874:  32%|███▏      | 2052/6470 [23:32<55:35,  1.32it/s]
3|2053|Loss: 0.8040816187858582:  32%|███▏      | 2052/6470 [23:33<55:35,  1.32it/s]
3|2053|Loss: 0.8040816187858582:  32%|███▏      | 2053/6470 [23:33<57:54,  1.27it/s]
3|2054|Loss: 1.1191554069519043:  32%|███▏      | 2053/6470 [23:33<57:54,  1.27it/s]
3|2054|Loss: 1.1191554069519043:  32%|███▏      | 2054/6470 [23:34<53:58,  1.36it/s]
3|2055|Loss: 0.756268322467804:  32%|███▏      | 2054/6470 [23:34<53:58,  1.36it/s] 
3|2055|Loss: 0.756268322467804:  32%|███▏      | 2055/6470 [23:34<48:09,  1.53it/s]
3|2056|Loss: 0.8145251870155334:  32%|███▏      | 2055/6470 [23:34<48:09,  1.53it/s]
3|2056|Loss: 0.8145251870155334:  32%|███▏      | 2056/6470 [23:35<48:34,  1.51it/s]
3|2057|Loss: 0.9266151785850525:  32%|███▏      | 2056/6470 [23:35<48:34,  1.51it/s]
3|2057|Loss: 0.9266151785850525:  32%|███▏      | 2057/6470 [23:36<50:29,  1.46it/s]
3|2058|Loss: 0.6949179172515869:  32%|███▏      | 2057/6470 [23:36<50:29,  1.46it/s]
3|2058|Loss: 0.6949179172515869:  32%|███▏      | 2058/6470 [23:36<51:38,  1.42it/s]
3|2059|Loss: 0.7034964561462402:  32%|███▏      | 2058/6470 [23:37<51:38,  1.42it/s]
3|2059|Loss: 0.7034964561462402:  32%|███▏      | 2059/6470 [23:37<50:44,  1.45it/s]
3|2060|Loss: 0.7581468224525452:  32%|███▏      | 2059/6470 [23:37<50:44,  1.45it/s]
3|2060|Loss: 0.7581468224525452:  32%|███▏      | 2060/6470 [23:38<50:57,  1.44it/s]
3|2061|Loss: 0.7027924656867981:  32%|███▏      | 2060/6470 [23:38<50:57,  1.44it/s]
3|2061|Loss: 0.7027924656867981:  32%|███▏      | 2061/6470 [23:38<51:45,  1.42it/s]
3|2062|Loss: 0.944716215133667:  32%|███▏      | 2061/6470 [23:39<51:45,  1.42it/s] 
3|2062|Loss: 0.944716215133667:  32%|███▏      | 2062/6470 [23:39<52:40,  1.39it/s]
3|2063|Loss: 0.7845475673675537:  32%|███▏      | 2062/6470 [23:40<52:40,  1.39it/s]
3|2063|Loss: 0.7845475673675537:  32%|███▏      | 2063/6470 [23:40<51:08,  1.44it/s]
3|2064|Loss: 1.0084943771362305:  32%|███▏      | 2063/6470 [23:40<51:08,  1.44it/s]
3|2064|Loss: 1.0084943771362305:  32%|███▏      | 2064/6470 [23:41<52:41,  1.39it/s]
3|2065|Loss: 0.8826400637626648:  32%|███▏      | 2064/6470 [23:41<52:41,  1.39it/s]
3|2065|Loss: 0.8826400637626648:  32%|███▏      | 2065/6470 [23:41<53:25,  1.37it/s]
3|2066|Loss: 0.8817151188850403:  32%|███▏      | 2065/6470 [23:42<53:25,  1.37it/s]
3|2066|Loss: 0.8817151188850403:  32%|███▏      | 2066/6470 [23:42<54:42,  1.34it/s]
3|2067|Loss: 0.8786669969558716:  32%|███▏      | 2066/6470 [23:42<54:42,  1.34it/s]
3|2067|Loss: 0.8786669969558716:  32%|███▏      | 2067/6470 [23:43<52:53,  1.39it/s]
3|2068|Loss: 0.9291660189628601:  32%|███▏      | 2067/6470 [23:43<52:53,  1.39it/s]
3|2068|Loss: 0.9291660189628601:  32%|███▏      | 2068/6470 [23:44<54:43,  1.34it/s]
3|2069|Loss: 0.9094139337539673:  32%|███▏      | 2068/6470 [23:44<54:43,  1.34it/s]
3|2069|Loss: 0.9094139337539673:  32%|███▏      | 2069/6470 [23:44<53:06,  1.38it/s]
3|2070|Loss: 0.8280567526817322:  32%|███▏      | 2069/6470 [23:45<53:06,  1.38it/s]
3|2070|Loss: 0.8280567526817322:  32%|███▏      | 2070/6470 [23:45<55:02,  1.33it/s]
3|2071|Loss: 0.8857924938201904:  32%|███▏      | 2070/6470 [23:45<55:02,  1.33it/s]
3|2071|Loss: 0.8857924938201904:  32%|███▏      | 2071/6470 [23:46<52:51,  1.39it/s]
3|2072|Loss: 0.9291493892669678:  32%|███▏      | 2071/6470 [23:46<52:51,  1.39it/s]
3|2072|Loss: 0.9291493892669678:  32%|███▏      | 2072/6470 [23:46<54:33,  1.34it/s]
3|2073|Loss: 0.8466814160346985:  32%|███▏      | 2072/6470 [23:47<54:33,  1.34it/s]
3|2073|Loss: 0.8466814160346985:  32%|███▏      | 2073/6470 [23:47<52:07,  1.41it/s]
3|2074|Loss: 0.6629576683044434:  32%|███▏      | 2073/6470 [23:48<52:07,  1.41it/s]
3|2074|Loss: 0.6629576683044434:  32%|███▏      | 2074/6470 [23:48<54:28,  1.35it/s]
3|2075|Loss: 0.9996476173400879:  32%|███▏      | 2074/6470 [23:48<54:28,  1.35it/s]
3|2075|Loss: 0.9996476173400879:  32%|███▏      | 2075/6470 [23:49<53:31,  1.37it/s]
3|2076|Loss: 0.8550925850868225:  32%|███▏      | 2075/6470 [23:49<53:31,  1.37it/s]
3|2076|Loss: 0.8550925850868225:  32%|███▏      | 2076/6470 [23:49<51:12,  1.43it/s]
3|2077|Loss: 0.9021643996238708:  32%|███▏      | 2076/6470 [23:50<51:12,  1.43it/s]
3|2077|Loss: 0.9021643996238708:  32%|███▏      | 2077/6470 [23:50<48:11,  1.52it/s]
3|2078|Loss: 0.8409314751625061:  32%|███▏      | 2077/6470 [23:50<48:11,  1.52it/s]
3|2078|Loss: 0.8409314751625061:  32%|███▏      | 2078/6470 [23:51<50:57,  1.44it/s]
3|2079|Loss: 0.8801586031913757:  32%|███▏      | 2078/6470 [23:51<50:57,  1.44it/s]
3|2079|Loss: 0.8801586031913757:  32%|███▏      | 2079/6470 [23:51<51:36,  1.42it/s]
3|2080|Loss: 0.8000272512435913:  32%|███▏      | 2079/6470 [23:52<51:36,  1.42it/s]
3|2080|Loss: 0.8000272512435913:  32%|███▏      | 2080/6470 [23:52<49:29,  1.48it/s]
3|2081|Loss: 0.8950625061988831:  32%|███▏      | 2080/6470 [23:52<49:29,  1.48it/s]
3|2081|Loss: 0.8950625061988831:  32%|███▏      | 2081/6470 [23:53<50:05,  1.46it/s]
3|2082|Loss: 0.9174785614013672:  32%|███▏      | 2081/6470 [23:53<50:05,  1.46it/s]
3|2082|Loss: 0.9174785614013672:  32%|███▏      | 2082/6470 [23:53<53:14,  1.37it/s]
3|2083|Loss: 0.8623034954071045:  32%|███▏      | 2082/6470 [23:54<53:14,  1.37it/s]
3|2083|Loss: 0.8623034954071045:  32%|███▏      | 2083/6470 [23:54<54:14,  1.35it/s]
3|2084|Loss: 1.0060291290283203:  32%|███▏      | 2083/6470 [23:55<54:14,  1.35it/s]
3|2084|Loss: 1.0060291290283203:  32%|███▏      | 2084/6470 [23:55<55:06,  1.33it/s]
3|2085|Loss: 0.9246192574501038:  32%|███▏      | 2084/6470 [23:55<55:06,  1.33it/s]
3|2085|Loss: 0.9246192574501038:  32%|███▏      | 2085/6470 [23:56<51:05,  1.43it/s]
3|2086|Loss: 0.9482953548431396:  32%|███▏      | 2085/6470 [23:56<51:05,  1.43it/s]
3|2086|Loss: 0.9482953548431396:  32%|███▏      | 2086/6470 [23:56<50:22,  1.45it/s]
3|2087|Loss: 0.8876296281814575:  32%|███▏      | 2086/6470 [23:57<50:22,  1.45it/s]
3|2087|Loss: 0.8876296281814575:  32%|███▏      | 2087/6470 [23:57<48:24,  1.51it/s]
3|2088|Loss: 0.7452894449234009:  32%|███▏      | 2087/6470 [23:57<48:24,  1.51it/s]
3|2088|Loss: 0.7452894449234009:  32%|███▏      | 2088/6470 [23:58<51:02,  1.43it/s]
3|2089|Loss: 0.9558809995651245:  32%|███▏      | 2088/6470 [23:58<51:02,  1.43it/s]
3|2089|Loss: 0.9558809995651245:  32%|███▏      | 2089/6470 [23:58<51:43,  1.41it/s]
3|2090|Loss: 0.7434888482093811:  32%|███▏      | 2089/6470 [23:59<51:43,  1.41it/s]
3|2090|Loss: 0.7434888482093811:  32%|███▏      | 2090/6470 [23:59<52:13,  1.40it/s]
3|2091|Loss: 0.775713324546814:  32%|███▏      | 2090/6470 [24:00<52:13,  1.40it/s] 
3|2091|Loss: 0.775713324546814:  32%|███▏      | 2091/6470 [24:00<56:37,  1.29it/s]
3|2092|Loss: 0.6459735631942749:  32%|███▏      | 2091/6470 [24:00<56:37,  1.29it/s]
3|2092|Loss: 0.6459735631942749:  32%|███▏      | 2092/6470 [24:01<55:09,  1.32it/s]
3|2093|Loss: 0.8178049325942993:  32%|███▏      | 2092/6470 [24:01<55:09,  1.32it/s]
3|2093|Loss: 0.8178049325942993:  32%|███▏      | 2093/6470 [24:01<51:29,  1.42it/s]
3|2094|Loss: 0.9277591705322266:  32%|███▏      | 2093/6470 [24:02<51:29,  1.42it/s]
3|2094|Loss: 0.9277591705322266:  32%|███▏      | 2094/6470 [24:02<48:45,  1.50it/s]
3|2095|Loss: 0.8729874491691589:  32%|███▏      | 2094/6470 [24:02<48:45,  1.50it/s]
3|2095|Loss: 0.8729874491691589:  32%|███▏      | 2095/6470 [24:03<49:45,  1.47it/s]
3|2096|Loss: 0.9508547782897949:  32%|███▏      | 2095/6470 [24:03<49:45,  1.47it/s]
3|2096|Loss: 0.9508547782897949:  32%|███▏      | 2096/6470 [24:03<50:58,  1.43it/s]
3|2097|Loss: 0.7861555814743042:  32%|███▏      | 2096/6470 [24:04<50:58,  1.43it/s]
3|2097|Loss: 0.7861555814743042:  32%|███▏      | 2097/6470 [24:04<49:42,  1.47it/s]
3|2098|Loss: 1.0002292394638062:  32%|███▏      | 2097/6470 [24:04<49:42,  1.47it/s]
3|2098|Loss: 1.0002292394638062:  32%|███▏      | 2098/6470 [24:05<48:13,  1.51it/s]
3|2099|Loss: 0.8199377059936523:  32%|███▏      | 2098/6470 [24:05<48:13,  1.51it/s]
3|2099|Loss: 0.8199377059936523:  32%|███▏      | 2099/6470 [24:05<47:14,  1.54it/s]
3|2100|Loss: 0.8581939935684204:  32%|███▏      | 2099/6470 [24:06<47:14,  1.54it/s]
3|2100|Loss: 0.8581939935684204:  32%|███▏      | 2100/6470 [24:06<43:50,  1.66it/s]
3|2101|Loss: 0.9322090148925781:  32%|███▏      | 2100/6470 [24:06<43:50,  1.66it/s]
3|2101|Loss: 0.9322090148925781:  32%|███▏      | 2101/6470 [24:06<46:20,  1.57it/s]
3|2102|Loss: 0.6964845061302185:  32%|███▏      | 2101/6470 [24:07<46:20,  1.57it/s]
3|2102|Loss: 0.6964845061302185:  32%|███▏      | 2102/6470 [24:07<43:55,  1.66it/s]
3|2103|Loss: 0.8410330414772034:  32%|███▏      | 2102/6470 [24:07<43:55,  1.66it/s]
3|2103|Loss: 0.8410330414772034:  33%|███▎      | 2103/6470 [24:08<42:51,  1.70it/s]
3|2104|Loss: 0.8576459884643555:  33%|███▎      | 2103/6470 [24:08<42:51,  1.70it/s]
3|2104|Loss: 0.8576459884643555:  33%|███▎      | 2104/6470 [24:08<43:45,  1.66it/s]
3|2105|Loss: 0.8313605785369873:  33%|███▎      | 2104/6470 [24:09<43:45,  1.66it/s]
3|2105|Loss: 0.8313605785369873:  33%|███▎      | 2105/6470 [24:09<46:35,  1.56it/s]
3|2106|Loss: 0.8773059248924255:  33%|███▎      | 2105/6470 [24:09<46:35,  1.56it/s]
3|2106|Loss: 0.8773059248924255:  33%|███▎      | 2106/6470 [24:10<49:02,  1.48it/s]
3|2107|Loss: 0.8937627673149109:  33%|███▎      | 2106/6470 [24:10<49:02,  1.48it/s]
3|2107|Loss: 0.8937627673149109:  33%|███▎      | 2107/6470 [24:10<48:54,  1.49it/s]
3|2108|Loss: 0.6199559569358826:  33%|███▎      | 2107/6470 [24:11<48:54,  1.49it/s]
3|2108|Loss: 0.6199559569358826:  33%|███▎      | 2108/6470 [24:11<47:48,  1.52it/s]
3|2109|Loss: 0.9451749920845032:  33%|███▎      | 2108/6470 [24:11<47:48,  1.52it/s]
3|2109|Loss: 0.9451749920845032:  33%|███▎      | 2109/6470 [24:12<48:26,  1.50it/s]
3|2110|Loss: 0.964981198310852:  33%|███▎      | 2109/6470 [24:12<48:26,  1.50it/s] 
3|2110|Loss: 0.964981198310852:  33%|███▎      | 2110/6470 [24:12<49:07,  1.48it/s]
3|2111|Loss: 0.794066309928894:  33%|███▎      | 2110/6470 [24:13<49:07,  1.48it/s]
3|2111|Loss: 0.794066309928894:  33%|███▎      | 2111/6470 [24:13<51:02,  1.42it/s]
3|2112|Loss: 0.9576678276062012:  33%|███▎      | 2111/6470 [24:14<51:02,  1.42it/s]
3|2112|Loss: 0.9576678276062012:  33%|███▎      | 2112/6470 [24:14<52:15,  1.39it/s]
3|2113|Loss: 0.9134559035301208:  33%|███▎      | 2112/6470 [24:14<52:15,  1.39it/s]
3|2113|Loss: 0.9134559035301208:  33%|███▎      | 2113/6470 [24:15<51:10,  1.42it/s]
3|2114|Loss: 0.8617468476295471:  33%|███▎      | 2113/6470 [24:15<51:10,  1.42it/s]
3|2114|Loss: 0.8617468476295471:  33%|███▎      | 2114/6470 [24:15<51:24,  1.41it/s]
3|2115|Loss: 0.8889309763908386:  33%|███▎      | 2114/6470 [24:16<51:24,  1.41it/s]
3|2115|Loss: 0.8889309763908386:  33%|███▎      | 2115/6470 [24:16<49:27,  1.47it/s]
3|2116|Loss: 0.7766743898391724:  33%|███▎      | 2115/6470 [24:16<49:27,  1.47it/s]
3|2116|Loss: 0.7766743898391724:  33%|███▎      | 2116/6470 [24:17<50:10,  1.45it/s]
3|2117|Loss: 0.9671791195869446:  33%|███▎      | 2116/6470 [24:17<50:10,  1.45it/s]
3|2117|Loss: 0.9671791195869446:  33%|███▎      | 2117/6470 [24:17<50:32,  1.44it/s]
3|2118|Loss: 0.9338836669921875:  33%|███▎      | 2117/6470 [24:18<50:32,  1.44it/s]
3|2118|Loss: 0.9338836669921875:  33%|███▎      | 2118/6470 [24:18<52:59,  1.37it/s]
3|2119|Loss: 0.7451193928718567:  33%|███▎      | 2118/6470 [24:18<52:59,  1.37it/s]
3|2119|Loss: 0.7451193928718567:  33%|███▎      | 2119/6470 [24:19<51:34,  1.41it/s]
3|2120|Loss: 0.7690566778182983:  33%|███▎      | 2119/6470 [24:19<51:34,  1.41it/s]
3|2120|Loss: 0.7690566778182983:  33%|███▎      | 2120/6470 [24:19<49:23,  1.47it/s]
3|2121|Loss: 0.8257519602775574:  33%|███▎      | 2120/6470 [24:20<49:23,  1.47it/s]
3|2121|Loss: 0.8257519602775574:  33%|███▎      | 2121/6470 [24:20<51:43,  1.40it/s]
3|2122|Loss: 0.8017140626907349:  33%|███▎      | 2121/6470 [24:21<51:43,  1.40it/s]
3|2122|Loss: 0.8017140626907349:  33%|███▎      | 2122/6470 [24:21<52:08,  1.39it/s]
3|2123|Loss: 0.8098496198654175:  33%|███▎      | 2122/6470 [24:21<52:08,  1.39it/s]
3|2123|Loss: 0.8098496198654175:  33%|███▎      | 2123/6470 [24:21<48:43,  1.49it/s]
3|2124|Loss: 0.8511629700660706:  33%|███▎      | 2123/6470 [24:22<48:43,  1.49it/s]
3|2124|Loss: 0.8511629700660706:  33%|███▎      | 2124/6470 [24:22<49:26,  1.47it/s]
3|2125|Loss: 0.8819325566291809:  33%|███▎      | 2124/6470 [24:23<49:26,  1.47it/s]
3|2125|Loss: 0.8819325566291809:  33%|███▎      | 2125/6470 [24:23<50:11,  1.44it/s]
3|2126|Loss: 0.7600932121276855:  33%|███▎      | 2125/6470 [24:23<50:11,  1.44it/s]
3|2126|Loss: 0.7600932121276855:  33%|███▎      | 2126/6470 [24:24<50:34,  1.43it/s]
3|2127|Loss: 0.739161491394043:  33%|███▎      | 2126/6470 [24:24<50:34,  1.43it/s] 
3|2127|Loss: 0.739161491394043:  33%|███▎      | 2127/6470 [24:24<53:17,  1.36it/s]
3|2128|Loss: 0.8106340169906616:  33%|███▎      | 2127/6470 [24:25<53:17,  1.36it/s]
3|2128|Loss: 0.8106340169906616:  33%|███▎      | 2128/6470 [24:25<53:55,  1.34it/s]
3|2129|Loss: 0.8848636150360107:  33%|███▎      | 2128/6470 [24:26<53:55,  1.34it/s]
3|2129|Loss: 0.8848636150360107:  33%|███▎      | 2129/6470 [24:26<50:55,  1.42it/s]
3|2130|Loss: 0.9129648804664612:  33%|███▎      | 2129/6470 [24:26<50:55,  1.42it/s]
3|2130|Loss: 0.9129648804664612:  33%|███▎      | 2130/6470 [24:26<50:24,  1.44it/s]
3|2131|Loss: 0.8219708800315857:  33%|███▎      | 2130/6470 [24:27<50:24,  1.44it/s]
3|2131|Loss: 0.8219708800315857:  33%|███▎      | 2131/6470 [24:27<49:08,  1.47it/s]
3|2132|Loss: 0.730678915977478:  33%|███▎      | 2131/6470 [24:28<49:08,  1.47it/s] 
3|2132|Loss: 0.730678915977478:  33%|███▎      | 2132/6470 [24:28<50:47,  1.42it/s]
3|2133|Loss: 0.7688202261924744:  33%|███▎      | 2132/6470 [24:28<50:47,  1.42it/s]
3|2133|Loss: 0.7688202261924744:  33%|███▎      | 2133/6470 [24:29<55:52,  1.29it/s]
3|2134|Loss: 0.8561793565750122:  33%|███▎      | 2133/6470 [24:29<55:52,  1.29it/s]
3|2134|Loss: 0.8561793565750122:  33%|███▎      | 2134/6470 [24:30<54:34,  1.32it/s]
3|2135|Loss: 0.765963077545166:  33%|███▎      | 2134/6470 [24:30<54:34,  1.32it/s] 
3|2135|Loss: 0.765963077545166:  33%|███▎      | 2135/6470 [24:30<53:09,  1.36it/s]
3|2136|Loss: 0.9230732321739197:  33%|███▎      | 2135/6470 [24:31<53:09,  1.36it/s]
3|2136|Loss: 0.9230732321739197:  33%|███▎      | 2136/6470 [24:31<54:02,  1.34it/s]
3|2137|Loss: 0.8408174514770508:  33%|███▎      | 2136/6470 [24:31<54:02,  1.34it/s]
3|2137|Loss: 0.8408174514770508:  33%|███▎      | 2137/6470 [24:32<52:12,  1.38it/s]
3|2138|Loss: 0.6586340069770813:  33%|███▎      | 2137/6470 [24:32<52:12,  1.38it/s]
3|2138|Loss: 0.6586340069770813:  33%|███▎      | 2138/6470 [24:32<45:12,  1.60it/s]
3|2139|Loss: 0.7334362268447876:  33%|███▎      | 2138/6470 [24:32<45:12,  1.60it/s]
3|2139|Loss: 0.7334362268447876:  33%|███▎      | 2139/6470 [24:33<48:12,  1.50it/s]
3|2140|Loss: 0.916187584400177:  33%|███▎      | 2139/6470 [24:33<48:12,  1.50it/s] 
3|2140|Loss: 0.916187584400177:  33%|███▎      | 2140/6470 [24:33<48:32,  1.49it/s]
3|2141|Loss: 0.8878846764564514:  33%|███▎      | 2140/6470 [24:34<48:32,  1.49it/s]
3|2141|Loss: 0.8878846764564514:  33%|███▎      | 2141/6470 [24:34<48:54,  1.47it/s]
3|2142|Loss: 0.835892915725708:  33%|███▎      | 2141/6470 [24:35<48:54,  1.47it/s] 
3|2142|Loss: 0.835892915725708:  33%|███▎      | 2142/6470 [24:35<49:32,  1.46it/s]
3|2143|Loss: 0.8253201246261597:  33%|███▎      | 2142/6470 [24:35<49:32,  1.46it/s]
3|2143|Loss: 0.8253201246261597:  33%|███▎      | 2143/6470 [24:36<49:52,  1.45it/s]
3|2144|Loss: 0.7333559989929199:  33%|███▎      | 2143/6470 [24:36<49:52,  1.45it/s]
3|2144|Loss: 0.7333559989929199:  33%|███▎      | 2144/6470 [24:36<49:59,  1.44it/s]
3|2145|Loss: 0.7964840531349182:  33%|███▎      | 2144/6470 [24:37<49:59,  1.44it/s]
3|2145|Loss: 0.7964840531349182:  33%|███▎      | 2145/6470 [24:37<50:11,  1.44it/s]
3|2146|Loss: 0.9655665755271912:  33%|███▎      | 2145/6470 [24:37<50:11,  1.44it/s]
3|2146|Loss: 0.9655665755271912:  33%|███▎      | 2146/6470 [24:38<51:03,  1.41it/s]
3|2147|Loss: 0.7653221487998962:  33%|███▎      | 2146/6470 [24:38<51:03,  1.41it/s]
3|2147|Loss: 0.7653221487998962:  33%|███▎      | 2147/6470 [24:38<51:13,  1.41it/s]
3|2148|Loss: 0.6679893732070923:  33%|███▎      | 2147/6470 [24:39<51:13,  1.41it/s]
3|2148|Loss: 0.6679893732070923:  33%|███▎      | 2148/6470 [24:39<54:43,  1.32it/s]
3|2149|Loss: 0.794792115688324:  33%|███▎      | 2148/6470 [24:40<54:43,  1.32it/s] 
3|2149|Loss: 0.794792115688324:  33%|███▎      | 2149/6470 [24:40<51:55,  1.39it/s]
3|2150|Loss: 0.9907297492027283:  33%|███▎      | 2149/6470 [24:40<51:55,  1.39it/s]
3|2150|Loss: 0.9907297492027283:  33%|███▎      | 2150/6470 [24:40<46:31,  1.55it/s]
3|2151|Loss: 0.8996211290359497:  33%|███▎      | 2150/6470 [24:41<46:31,  1.55it/s]
3|2151|Loss: 0.8996211290359497:  33%|███▎      | 2151/6470 [24:41<46:39,  1.54it/s]
3|2152|Loss: 0.716191828250885:  33%|███▎      | 2151/6470 [24:42<46:39,  1.54it/s] 
3|2152|Loss: 0.716191828250885:  33%|███▎      | 2152/6470 [24:42<50:29,  1.43it/s]
3|2153|Loss: 0.7286540269851685:  33%|███▎      | 2152/6470 [24:42<50:29,  1.43it/s]
3|2153|Loss: 0.7286540269851685:  33%|███▎      | 2153/6470 [24:43<51:48,  1.39it/s]
3|2154|Loss: 0.8705625534057617:  33%|███▎      | 2153/6470 [24:43<51:48,  1.39it/s]
3|2154|Loss: 0.8705625534057617:  33%|███▎      | 2154/6470 [24:43<50:21,  1.43it/s]
3|2155|Loss: 0.8266860246658325:  33%|███▎      | 2154/6470 [24:44<50:21,  1.43it/s]
3|2155|Loss: 0.8266860246658325:  33%|███▎      | 2155/6470 [24:44<53:49,  1.34it/s]
3|2156|Loss: 0.7986927032470703:  33%|███▎      | 2155/6470 [24:45<53:49,  1.34it/s]
3|2156|Loss: 0.7986927032470703:  33%|███▎      | 2156/6470 [24:45<52:26,  1.37it/s]
3|2157|Loss: 0.8178485631942749:  33%|███▎      | 2156/6470 [24:45<52:26,  1.37it/s]
3|2157|Loss: 0.8178485631942749:  33%|███▎      | 2157/6470 [24:45<49:57,  1.44it/s]
3|2158|Loss: 0.7733123302459717:  33%|███▎      | 2157/6470 [24:46<49:57,  1.44it/s]
3|2158|Loss: 0.7733123302459717:  33%|███▎      | 2158/6470 [24:46<49:42,  1.45it/s]
3|2159|Loss: 0.8413035869598389:  33%|███▎      | 2158/6470 [24:47<49:42,  1.45it/s]
3|2159|Loss: 0.8413035869598389:  33%|███▎      | 2159/6470 [24:47<50:22,  1.43it/s]
3|2160|Loss: 0.7764911651611328:  33%|███▎      | 2159/6470 [24:47<50:22,  1.43it/s]
3|2160|Loss: 0.7764911651611328:  33%|███▎      | 2160/6470 [24:48<51:02,  1.41it/s]
3|2161|Loss: 0.8999661803245544:  33%|███▎      | 2160/6470 [24:48<51:02,  1.41it/s]
3|2161|Loss: 0.8999661803245544:  33%|███▎      | 2161/6470 [24:48<52:24,  1.37it/s]
3|2162|Loss: 0.804857611656189:  33%|███▎      | 2161/6470 [24:49<52:24,  1.37it/s] 
3|2162|Loss: 0.804857611656189:  33%|███▎      | 2162/6470 [24:49<46:23,  1.55it/s]
3|2163|Loss: 0.8727006912231445:  33%|███▎      | 2162/6470 [24:49<46:23,  1.55it/s]
3|2163|Loss: 0.8727006912231445:  33%|███▎      | 2163/6470 [24:49<43:42,  1.64it/s]
3|2164|Loss: 0.8423541188240051:  33%|███▎      | 2163/6470 [24:50<43:42,  1.64it/s]
3|2164|Loss: 0.8423541188240051:  33%|███▎      | 2164/6470 [24:50<43:43,  1.64it/s]
3|2165|Loss: 0.8797904849052429:  33%|███▎      | 2164/6470 [24:50<43:43,  1.64it/s]
3|2165|Loss: 0.8797904849052429:  33%|███▎      | 2165/6470 [24:51<46:18,  1.55it/s]
3|2166|Loss: 0.7913883924484253:  33%|███▎      | 2165/6470 [24:51<46:18,  1.55it/s]
3|2166|Loss: 0.7913883924484253:  33%|███▎      | 2166/6470 [24:51<46:24,  1.55it/s]
3|2167|Loss: 0.7284658551216125:  33%|███▎      | 2166/6470 [24:52<46:24,  1.55it/s]
3|2167|Loss: 0.7284658551216125:  33%|███▎      | 2167/6470 [24:52<44:07,  1.63it/s]
3|2168|Loss: 0.9007745385169983:  33%|███▎      | 2167/6470 [24:52<44:07,  1.63it/s]
3|2168|Loss: 0.9007745385169983:  34%|███▎      | 2168/6470 [24:53<45:36,  1.57it/s]
3|2169|Loss: 0.7719148397445679:  34%|███▎      | 2168/6470 [24:53<45:36,  1.57it/s]
3|2169|Loss: 0.7719148397445679:  34%|███▎      | 2169/6470 [24:53<49:17,  1.45it/s]
3|2170|Loss: 0.745869517326355:  34%|███▎      | 2169/6470 [24:54<49:17,  1.45it/s] 
3|2170|Loss: 0.745869517326355:  34%|███▎      | 2170/6470 [24:54<52:29,  1.37it/s]
3|2171|Loss: 0.9207085967063904:  34%|███▎      | 2170/6470 [24:55<52:29,  1.37it/s]
3|2171|Loss: 0.9207085967063904:  34%|███▎      | 2171/6470 [24:55<52:40,  1.36it/s]
3|2172|Loss: 0.8628965020179749:  34%|███▎      | 2171/6470 [24:55<52:40,  1.36it/s]
3|2172|Loss: 0.8628965020179749:  34%|███▎      | 2172/6470 [24:56<54:41,  1.31it/s]
3|2173|Loss: 0.8456490635871887:  34%|███▎      | 2172/6470 [24:56<54:41,  1.31it/s]
3|2173|Loss: 0.8456490635871887:  34%|███▎      | 2173/6470 [24:57<55:00,  1.30it/s]
3|2174|Loss: 0.7759398221969604:  34%|███▎      | 2173/6470 [24:57<55:00,  1.30it/s]
3|2174|Loss: 0.7759398221969604:  34%|███▎      | 2174/6470 [24:57<52:42,  1.36it/s]
3|2175|Loss: 0.9009619355201721:  34%|███▎      | 2174/6470 [24:58<52:42,  1.36it/s]
3|2175|Loss: 0.9009619355201721:  34%|███▎      | 2175/6470 [24:58<52:24,  1.37it/s]
3|2176|Loss: 0.894709587097168:  34%|███▎      | 2175/6470 [24:58<52:24,  1.37it/s] 
3|2176|Loss: 0.894709587097168:  34%|███▎      | 2176/6470 [24:59<50:51,  1.41it/s]
3|2177|Loss: 1.1482536792755127:  34%|███▎      | 2176/6470 [24:59<50:51,  1.41it/s]
3|2177|Loss: 1.1482536792755127:  34%|███▎      | 2177/6470 [24:59<52:46,  1.36it/s]
3|2178|Loss: 0.9475048780441284:  34%|███▎      | 2177/6470 [25:00<52:46,  1.36it/s]
3|2178|Loss: 0.9475048780441284:  34%|███▎      | 2178/6470 [25:00<52:44,  1.36it/s]
3|2179|Loss: 0.8354507088661194:  34%|███▎      | 2178/6470 [25:01<52:44,  1.36it/s]
3|2179|Loss: 0.8354507088661194:  34%|███▎      | 2179/6470 [25:01<50:59,  1.40it/s]
3|2180|Loss: 0.8437069058418274:  34%|███▎      | 2179/6470 [25:01<50:59,  1.40it/s]
3|2180|Loss: 0.8437069058418274:  34%|███▎      | 2180/6470 [25:01<49:00,  1.46it/s]
3|2181|Loss: 0.962783932685852:  34%|███▎      | 2180/6470 [25:02<49:00,  1.46it/s] 
3|2181|Loss: 0.962783932685852:  34%|███▎      | 2181/6470 [25:02<49:27,  1.45it/s]
3|2182|Loss: 0.9165081977844238:  34%|███▎      | 2181/6470 [25:03<49:27,  1.45it/s]
3|2182|Loss: 0.9165081977844238:  34%|███▎      | 2182/6470 [25:03<48:43,  1.47it/s]
3|2183|Loss: 0.8664193153381348:  34%|███▎      | 2182/6470 [25:03<48:43,  1.47it/s]
3|2183|Loss: 0.8664193153381348:  34%|███▎      | 2183/6470 [25:04<51:30,  1.39it/s]
3|2184|Loss: 0.9428369402885437:  34%|███▎      | 2183/6470 [25:04<51:30,  1.39it/s]
3|2184|Loss: 0.9428369402885437:  34%|███▍      | 2184/6470 [25:04<54:04,  1.32it/s]
3|2185|Loss: 0.87428218126297:  34%|███▍      | 2184/6470 [25:05<54:04,  1.32it/s]  
3|2185|Loss: 0.87428218126297:  34%|███▍      | 2185/6470 [25:05<53:58,  1.32it/s]
3|2186|Loss: 0.7130826115608215:  34%|███▍      | 2185/6470 [25:06<53:58,  1.32it/s]
3|2186|Loss: 0.7130826115608215:  34%|███▍      | 2186/6470 [25:06<49:37,  1.44it/s]
3|2187|Loss: 0.8722031712532043:  34%|███▍      | 2186/6470 [25:06<49:37,  1.44it/s]
3|2187|Loss: 0.8722031712532043:  34%|███▍      | 2187/6470 [25:06<45:19,  1.57it/s]
3|2188|Loss: 0.9516885876655579:  34%|███▍      | 2187/6470 [25:07<45:19,  1.57it/s]
3|2188|Loss: 0.9516885876655579:  34%|███▍      | 2188/6470 [25:07<48:41,  1.47it/s]
3|2189|Loss: 0.9423730373382568:  34%|███▍      | 2188/6470 [25:07<48:41,  1.47it/s]
3|2189|Loss: 0.9423730373382568:  34%|███▍      | 2189/6470 [25:08<46:45,  1.53it/s]
3|2190|Loss: 0.8767867088317871:  34%|███▍      | 2189/6470 [25:08<46:45,  1.53it/s]
3|2190|Loss: 0.8767867088317871:  34%|███▍      | 2190/6470 [25:08<48:36,  1.47it/s]
3|2191|Loss: 0.9419206380844116:  34%|███▍      | 2190/6470 [25:09<48:36,  1.47it/s]
3|2191|Loss: 0.9419206380844116:  34%|███▍      | 2191/6470 [25:09<48:04,  1.48it/s]
3|2192|Loss: 0.8413743376731873:  34%|███▍      | 2191/6470 [25:09<48:04,  1.48it/s]
3|2192|Loss: 0.8413743376731873:  34%|███▍      | 2192/6470 [25:10<48:43,  1.46it/s]
3|2193|Loss: 0.9221056699752808:  34%|███▍      | 2192/6470 [25:10<48:43,  1.46it/s]
3|2193|Loss: 0.9221056699752808:  34%|███▍      | 2193/6470 [25:11<50:47,  1.40it/s]
3|2194|Loss: 0.8694364428520203:  34%|███▍      | 2193/6470 [25:11<50:47,  1.40it/s]
3|2194|Loss: 0.8694364428520203:  34%|███▍      | 2194/6470 [25:11<51:45,  1.38it/s]
3|2195|Loss: 0.7729530930519104:  34%|███▍      | 2194/6470 [25:12<51:45,  1.38it/s]
3|2195|Loss: 0.7729530930519104:  34%|███▍      | 2195/6470 [25:12<45:33,  1.56it/s]
3|2196|Loss: 0.8353480696678162:  34%|███▍      | 2195/6470 [25:12<45:33,  1.56it/s]
3|2196|Loss: 0.8353480696678162:  34%|███▍      | 2196/6470 [25:12<46:15,  1.54it/s]
3|2197|Loss: 0.8723753690719604:  34%|███▍      | 2196/6470 [25:13<46:15,  1.54it/s]
3|2197|Loss: 0.8723753690719604:  34%|███▍      | 2197/6470 [25:13<41:54,  1.70it/s]
3|2198|Loss: 0.8241354823112488:  34%|███▍      | 2197/6470 [25:13<41:54,  1.70it/s]
3|2198|Loss: 0.8241354823112488:  34%|███▍      | 2198/6470 [25:13<43:01,  1.65it/s]
3|2199|Loss: 0.8445011377334595:  34%|███▍      | 2198/6470 [25:14<43:01,  1.65it/s]
3|2199|Loss: 0.8445011377334595:  34%|███▍      | 2199/6470 [25:14<45:43,  1.56it/s]
3|2200|Loss: 0.8978350758552551:  34%|███▍      | 2199/6470 [25:15<45:43,  1.56it/s]
3|2200|Loss: 0.8978350758552551:  34%|███▍      | 2200/6470 [25:15<49:29,  1.44it/s]
3|2201|Loss: 1.020776629447937:  34%|███▍      | 2200/6470 [25:15<49:29,  1.44it/s] 
3|2201|Loss: 1.020776629447937:  34%|███▍      | 2201/6470 [25:16<48:52,  1.46it/s]
3|2202|Loss: 0.8403509855270386:  34%|███▍      | 2201/6470 [25:16<48:52,  1.46it/s]
3|2202|Loss: 0.8403509855270386:  34%|███▍      | 2202/6470 [25:16<48:46,  1.46it/s]
3|2203|Loss: 0.7846596837043762:  34%|███▍      | 2202/6470 [25:17<48:46,  1.46it/s]
3|2203|Loss: 0.7846596837043762:  34%|███▍      | 2203/6470 [25:17<48:05,  1.48it/s]
3|2204|Loss: 0.788557231426239:  34%|███▍      | 2203/6470 [25:17<48:05,  1.48it/s] 
3|2204|Loss: 0.788557231426239:  34%|███▍      | 2204/6470 [25:18<49:52,  1.43it/s]
3|2205|Loss: 0.8924325108528137:  34%|███▍      | 2204/6470 [25:18<49:52,  1.43it/s]
3|2205|Loss: 0.8924325108528137:  34%|███▍      | 2205/6470 [25:19<53:24,  1.33it/s]
3|2206|Loss: 0.8609926104545593:  34%|███▍      | 2205/6470 [25:19<53:24,  1.33it/s]
3|2206|Loss: 0.8609926104545593:  34%|███▍      | 2206/6470 [25:19<51:56,  1.37it/s]
3|2207|Loss: 0.7694344520568848:  34%|███▍      | 2206/6470 [25:20<51:56,  1.37it/s]
3|2207|Loss: 0.7694344520568848:  34%|███▍      | 2207/6470 [25:20<53:34,  1.33it/s]
3|2208|Loss: 0.8171942830085754:  34%|███▍      | 2207/6470 [25:21<53:34,  1.33it/s]
3|2208|Loss: 0.8171942830085754:  34%|███▍      | 2208/6470 [25:21<53:45,  1.32it/s]
3|2209|Loss: 0.926112711429596:  34%|███▍      | 2208/6470 [25:21<53:45,  1.32it/s] 
3|2209|Loss: 0.926112711429596:  34%|███▍      | 2209/6470 [25:22<52:42,  1.35it/s]
3|2210|Loss: 0.8901672959327698:  34%|███▍      | 2209/6470 [25:22<52:42,  1.35it/s]
3|2210|Loss: 0.8901672959327698:  34%|███▍      | 2210/6470 [25:22<54:00,  1.31it/s]
3|2211|Loss: 0.7763769626617432:  34%|███▍      | 2210/6470 [25:23<54:00,  1.31it/s]
3|2211|Loss: 0.7763769626617432:  34%|███▍      | 2211/6470 [25:23<53:30,  1.33it/s]
3|2212|Loss: 0.9224851131439209:  34%|███▍      | 2211/6470 [25:24<53:30,  1.33it/s]
3|2212|Loss: 0.9224851131439209:  34%|███▍      | 2212/6470 [25:24<51:29,  1.38it/s]
3|2213|Loss: 0.810833752155304:  34%|███▍      | 2212/6470 [25:24<51:29,  1.38it/s] 
3|2213|Loss: 0.810833752155304:  34%|███▍      | 2213/6470 [25:25<51:27,  1.38it/s]
3|2214|Loss: 0.9077578783035278:  34%|███▍      | 2213/6470 [25:25<51:27,  1.38it/s]
3|2214|Loss: 0.9077578783035278:  34%|███▍      | 2214/6470 [25:25<49:42,  1.43it/s]
3|2215|Loss: 0.8892058730125427:  34%|███▍      | 2214/6470 [25:26<49:42,  1.43it/s]
3|2215|Loss: 0.8892058730125427:  34%|███▍      | 2215/6470 [25:26<49:19,  1.44it/s]
3|2216|Loss: 0.785564124584198:  34%|███▍      | 2215/6470 [25:26<49:19,  1.44it/s] 
3|2216|Loss: 0.785564124584198:  34%|███▍      | 2216/6470 [25:27<50:52,  1.39it/s]
3|2217|Loss: 1.1039706468582153:  34%|███▍      | 2216/6470 [25:27<50:52,  1.39it/s]
3|2217|Loss: 1.1039706468582153:  34%|███▍      | 2217/6470 [25:27<50:00,  1.42it/s]
3|2218|Loss: 0.8659762740135193:  34%|███▍      | 2217/6470 [25:28<50:00,  1.42it/s]
3|2218|Loss: 0.8659762740135193:  34%|███▍      | 2218/6470 [25:28<51:25,  1.38it/s]
3|2219|Loss: 0.8799510598182678:  34%|███▍      | 2218/6470 [25:28<51:25,  1.38it/s]
3|2219|Loss: 0.8799510598182678:  34%|███▍      | 2219/6470 [25:29<50:01,  1.42it/s]
3|2220|Loss: 0.7743700742721558:  34%|███▍      | 2219/6470 [25:29<50:01,  1.42it/s]
3|2220|Loss: 0.7743700742721558:  34%|███▍      | 2220/6470 [25:29<45:49,  1.55it/s]
3|2221|Loss: 0.9489081501960754:  34%|███▍      | 2220/6470 [25:30<45:49,  1.55it/s]
3|2221|Loss: 0.9489081501960754:  34%|███▍      | 2221/6470 [25:30<47:29,  1.49it/s]
3|2222|Loss: 0.9306021928787231:  34%|███▍      | 2221/6470 [25:30<47:29,  1.49it/s]
3|2222|Loss: 0.9306021928787231:  34%|███▍      | 2222/6470 [25:31<48:40,  1.45it/s]
3|2223|Loss: 0.9756243228912354:  34%|███▍      | 2222/6470 [25:31<48:40,  1.45it/s]
3|2223|Loss: 0.9756243228912354:  34%|███▍      | 2223/6470 [25:31<49:00,  1.44it/s]
3|2224|Loss: 0.6653133630752563:  34%|███▍      | 2223/6470 [25:32<49:00,  1.44it/s]
3|2224|Loss: 0.6653133630752563:  34%|███▍      | 2224/6470 [25:32<51:13,  1.38it/s]
3|2225|Loss: 0.8998196721076965:  34%|███▍      | 2224/6470 [25:33<51:13,  1.38it/s]
3|2225|Loss: 0.8998196721076965:  34%|███▍      | 2225/6470 [25:33<51:56,  1.36it/s]
3|2226|Loss: 0.8837795853614807:  34%|███▍      | 2225/6470 [25:33<51:56,  1.36it/s]
3|2226|Loss: 0.8837795853614807:  34%|███▍      | 2226/6470 [25:34<52:13,  1.35it/s]
3|2227|Loss: 0.8222190141677856:  34%|███▍      | 2226/6470 [25:34<52:13,  1.35it/s]
3|2227|Loss: 0.8222190141677856:  34%|███▍      | 2227/6470 [25:34<47:30,  1.49it/s]
3|2228|Loss: 0.9729653000831604:  34%|███▍      | 2227/6470 [25:35<47:30,  1.49it/s]
3|2228|Loss: 0.9729653000831604:  34%|███▍      | 2228/6470 [25:35<44:47,  1.58it/s]
3|2229|Loss: 0.9342103600502014:  34%|███▍      | 2228/6470 [25:35<44:47,  1.58it/s]
3|2229|Loss: 0.9342103600502014:  34%|███▍      | 2229/6470 [25:35<43:39,  1.62it/s]
3|2230|Loss: 0.8500418066978455:  34%|███▍      | 2229/6470 [25:36<43:39,  1.62it/s]
3|2230|Loss: 0.8500418066978455:  34%|███▍      | 2230/6470 [25:36<46:30,  1.52it/s]
3|2231|Loss: 0.9043793082237244:  34%|███▍      | 2230/6470 [25:36<46:30,  1.52it/s]
3|2231|Loss: 0.9043793082237244:  34%|███▍      | 2231/6470 [25:37<45:23,  1.56it/s]
3|2232|Loss: 0.7519297003746033:  34%|███▍      | 2231/6470 [25:37<45:23,  1.56it/s]
3|2232|Loss: 0.7519297003746033:  34%|███▍      | 2232/6470 [25:37<46:58,  1.50it/s]
3|2233|Loss: 0.9230455160140991:  34%|███▍      | 2232/6470 [25:38<46:58,  1.50it/s]
3|2233|Loss: 0.9230455160140991:  35%|███▍      | 2233/6470 [25:38<48:30,  1.46it/s]
3|2234|Loss: 0.8686140179634094:  35%|███▍      | 2233/6470 [25:39<48:30,  1.46it/s]
3|2234|Loss: 0.8686140179634094:  35%|███▍      | 2234/6470 [25:39<52:09,  1.35it/s]
3|2235|Loss: 0.9704312086105347:  35%|███▍      | 2234/6470 [25:39<52:09,  1.35it/s]
3|2235|Loss: 0.9704312086105347:  35%|███▍      | 2235/6470 [25:40<50:54,  1.39it/s]
3|2236|Loss: 0.9358370304107666:  35%|███▍      | 2235/6470 [25:40<50:54,  1.39it/s]
3|2236|Loss: 0.9358370304107666:  35%|███▍      | 2236/6470 [25:41<52:55,  1.33it/s]
3|2237|Loss: 0.8330744504928589:  35%|███▍      | 2236/6470 [25:41<52:55,  1.33it/s]
3|2237|Loss: 0.8330744504928589:  35%|███▍      | 2237/6470 [25:41<54:09,  1.30it/s]
3|2238|Loss: 0.9549527764320374:  35%|███▍      | 2237/6470 [25:42<54:09,  1.30it/s]
3|2238|Loss: 0.9549527764320374:  35%|███▍      | 2238/6470 [25:42<51:52,  1.36it/s]
3|2239|Loss: 0.7643378376960754:  35%|███▍      | 2238/6470 [25:42<51:52,  1.36it/s]
3|2239|Loss: 0.7643378376960754:  35%|███▍      | 2239/6470 [25:43<49:18,  1.43it/s]
3|2240|Loss: 0.9398239850997925:  35%|███▍      | 2239/6470 [25:43<49:18,  1.43it/s]
3|2240|Loss: 0.9398239850997925:  35%|███▍      | 2240/6470 [25:43<51:38,  1.37it/s]
3|2241|Loss: 0.9953311681747437:  35%|███▍      | 2240/6470 [25:44<51:38,  1.37it/s]
3|2241|Loss: 0.9953311681747437:  35%|███▍      | 2241/6470 [25:44<53:24,  1.32it/s]
3|2242|Loss: 1.0134283304214478:  35%|███▍      | 2241/6470 [25:45<53:24,  1.32it/s]
3|2242|Loss: 1.0134283304214478:  35%|███▍      | 2242/6470 [25:45<48:27,  1.45it/s]
3|2243|Loss: 0.7935264706611633:  35%|███▍      | 2242/6470 [25:45<48:27,  1.45it/s]
3|2243|Loss: 0.7935264706611633:  35%|███▍      | 2243/6470 [25:45<45:41,  1.54it/s]
3|2244|Loss: 0.8736588358879089:  35%|███▍      | 2243/6470 [25:46<45:41,  1.54it/s]
3|2244|Loss: 0.8736588358879089:  35%|███▍      | 2244/6470 [25:46<44:23,  1.59it/s]
3|2245|Loss: 1.0096430778503418:  35%|███▍      | 2244/6470 [25:46<44:23,  1.59it/s]
3|2245|Loss: 1.0096430778503418:  35%|███▍      | 2245/6470 [25:47<46:06,  1.53it/s]
3|2246|Loss: 0.8058150410652161:  35%|███▍      | 2245/6470 [25:47<46:06,  1.53it/s]
3|2246|Loss: 0.8058150410652161:  35%|███▍      | 2246/6470 [25:47<46:13,  1.52it/s]
3|2247|Loss: 0.855334460735321:  35%|███▍      | 2246/6470 [25:48<46:13,  1.52it/s] 
3|2247|Loss: 0.855334460735321:  35%|███▍      | 2247/6470 [25:48<49:26,  1.42it/s]
3|2248|Loss: 0.7883255481719971:  35%|███▍      | 2247/6470 [25:49<49:26,  1.42it/s]
3|2248|Loss: 0.7883255481719971:  35%|███▍      | 2248/6470 [25:49<53:03,  1.33it/s]
3|2249|Loss: 0.9322744011878967:  35%|███▍      | 2248/6470 [25:49<53:03,  1.33it/s]
3|2249|Loss: 0.9322744011878967:  35%|███▍      | 2249/6470 [25:50<52:51,  1.33it/s]
3|2250|Loss: 0.7849566340446472:  35%|███▍      | 2249/6470 [25:50<52:51,  1.33it/s]
3|2250|Loss: 0.7849566340446472:  35%|███▍      | 2250/6470 [25:50<52:23,  1.34it/s]
3|2251|Loss: 0.7960115671157837:  35%|███▍      | 2250/6470 [25:51<52:23,  1.34it/s]
3|2251|Loss: 0.7960115671157837:  35%|███▍      | 2251/6470 [25:51<50:13,  1.40it/s]
3|2252|Loss: 0.8241088390350342:  35%|███▍      | 2251/6470 [25:51<50:13,  1.40it/s]
3|2252|Loss: 0.8241088390350342:  35%|███▍      | 2252/6470 [25:52<46:36,  1.51it/s]
3|2253|Loss: 0.764377772808075:  35%|███▍      | 2252/6470 [25:52<46:36,  1.51it/s] 
3|2253|Loss: 0.764377772808075:  35%|███▍      | 2253/6470 [25:52<48:07,  1.46it/s]
3|2254|Loss: 0.647415816783905:  35%|███▍      | 2253/6470 [25:53<48:07,  1.46it/s]
3|2254|Loss: 0.647415816783905:  35%|███▍      | 2254/6470 [25:54<1:02:10,  1.13it/s]
3|2255|Loss: 0.7691206932067871:  35%|███▍      | 2254/6470 [25:54<1:02:10,  1.13it/s]
3|2255|Loss: 0.7691206932067871:  35%|███▍      | 2255/6470 [25:54<58:37,  1.20it/s]  
3|2256|Loss: 0.8895994424819946:  35%|███▍      | 2255/6470 [25:55<58:37,  1.20it/s]
3|2256|Loss: 0.8895994424819946:  35%|███▍      | 2256/6470 [25:55<57:09,  1.23it/s]
3|2257|Loss: 0.8896077275276184:  35%|███▍      | 2256/6470 [25:56<57:09,  1.23it/s]
3|2257|Loss: 0.8896077275276184:  35%|███▍      | 2257/6470 [25:56<54:02,  1.30it/s]
3|2258|Loss: 0.8734900951385498:  35%|███▍      | 2257/6470 [25:56<54:02,  1.30it/s]
3|2258|Loss: 0.8734900951385498:  35%|███▍      | 2258/6470 [25:56<50:08,  1.40it/s]
3|2259|Loss: 0.8896088004112244:  35%|███▍      | 2258/6470 [25:57<50:08,  1.40it/s]
3|2259|Loss: 0.8896088004112244:  35%|███▍      | 2259/6470 [25:57<48:01,  1.46it/s]
3|2260|Loss: 0.8084236979484558:  35%|███▍      | 2259/6470 [25:57<48:01,  1.46it/s]
3|2260|Loss: 0.8084236979484558:  35%|███▍      | 2260/6470 [25:58<51:00,  1.38it/s]
3|2261|Loss: 0.7327215075492859:  35%|███▍      | 2260/6470 [25:58<51:00,  1.38it/s]
3|2261|Loss: 0.7327215075492859:  35%|███▍      | 2261/6470 [25:59<51:49,  1.35it/s]
3|2262|Loss: 0.7901983857154846:  35%|███▍      | 2261/6470 [25:59<51:49,  1.35it/s]
3|2262|Loss: 0.7901983857154846:  35%|███▍      | 2262/6470 [25:59<50:15,  1.40it/s]
3|2263|Loss: 0.864388644695282:  35%|███▍      | 2262/6470 [26:00<50:15,  1.40it/s] 
3|2263|Loss: 0.864388644695282:  35%|███▍      | 2263/6470 [26:00<49:44,  1.41it/s]
3|2264|Loss: 0.8186966180801392:  35%|███▍      | 2263/6470 [26:00<49:44,  1.41it/s]
3|2264|Loss: 0.8186966180801392:  35%|███▍      | 2264/6470 [26:01<51:45,  1.35it/s]
3|2265|Loss: 0.8826315999031067:  35%|███▍      | 2264/6470 [26:01<51:45,  1.35it/s]
3|2265|Loss: 0.8826315999031067:  35%|███▌      | 2265/6470 [26:02<53:38,  1.31it/s]
3|2266|Loss: 0.8071159720420837:  35%|███▌      | 2265/6470 [26:02<53:38,  1.31it/s]
3|2266|Loss: 0.8071159720420837:  35%|███▌      | 2266/6470 [26:02<51:48,  1.35it/s]
3|2267|Loss: 1.0292197465896606:  35%|███▌      | 2266/6470 [26:03<51:48,  1.35it/s]
3|2267|Loss: 1.0292197465896606:  35%|███▌      | 2267/6470 [26:03<50:49,  1.38it/s]
3|2268|Loss: 0.9528145790100098:  35%|███▌      | 2267/6470 [26:03<50:49,  1.38it/s]
3|2268|Loss: 0.9528145790100098:  35%|███▌      | 2268/6470 [26:04<53:59,  1.30it/s]
3|2269|Loss: 1.0350496768951416:  35%|███▌      | 2268/6470 [26:04<53:59,  1.30it/s]
3|2269|Loss: 1.0350496768951416:  35%|███▌      | 2269/6470 [26:05<56:08,  1.25it/s]
3|2270|Loss: 1.0337485074996948:  35%|███▌      | 2269/6470 [26:05<56:08,  1.25it/s]
3|2270|Loss: 1.0337485074996948:  35%|███▌      | 2270/6470 [26:05<52:49,  1.33it/s]
3|2271|Loss: 0.9002591371536255:  35%|███▌      | 2270/6470 [26:06<52:49,  1.33it/s]
3|2271|Loss: 0.9002591371536255:  35%|███▌      | 2271/6470 [26:06<48:45,  1.44it/s]
3|2272|Loss: 0.9160212874412537:  35%|███▌      | 2271/6470 [26:06<48:45,  1.44it/s]
3|2272|Loss: 0.9160212874412537:  35%|███▌      | 2272/6470 [26:07<48:42,  1.44it/s]
3|2273|Loss: 0.7763097286224365:  35%|███▌      | 2272/6470 [26:07<48:42,  1.44it/s]
3|2273|Loss: 0.7763097286224365:  35%|███▌      | 2273/6470 [26:07<48:53,  1.43it/s]
3|2274|Loss: 0.8710197806358337:  35%|███▌      | 2273/6470 [26:08<48:53,  1.43it/s]
3|2274|Loss: 0.8710197806358337:  35%|███▌      | 2274/6470 [26:08<47:35,  1.47it/s]
3|2275|Loss: 0.7957485914230347:  35%|███▌      | 2274/6470 [26:08<47:35,  1.47it/s]
3|2275|Loss: 0.7957485914230347:  35%|███▌      | 2275/6470 [26:09<46:12,  1.51it/s]
3|2276|Loss: 0.850179135799408:  35%|███▌      | 2275/6470 [26:09<46:12,  1.51it/s] 
3|2276|Loss: 0.850179135799408:  35%|███▌      | 2276/6470 [26:09<50:16,  1.39it/s]
3|2277|Loss: 0.8936181664466858:  35%|███▌      | 2276/6470 [26:10<50:16,  1.39it/s]
3|2277|Loss: 0.8936181664466858:  35%|███▌      | 2277/6470 [26:10<53:04,  1.32it/s]
3|2278|Loss: 1.0020910501480103:  35%|███▌      | 2277/6470 [26:11<53:04,  1.32it/s]
3|2278|Loss: 1.0020910501480103:  35%|███▌      | 2278/6470 [26:11<52:26,  1.33it/s]
3|2279|Loss: 0.7599352598190308:  35%|███▌      | 2278/6470 [26:11<52:26,  1.33it/s]
3|2279|Loss: 0.7599352598190308:  35%|███▌      | 2279/6470 [26:12<50:03,  1.40it/s]
3|2280|Loss: 0.9492712616920471:  35%|███▌      | 2279/6470 [26:12<50:03,  1.40it/s]
3|2280|Loss: 0.9492712616920471:  35%|███▌      | 2280/6470 [26:12<49:20,  1.42it/s]
3|2281|Loss: 0.8135302066802979:  35%|███▌      | 2280/6470 [26:13<49:20,  1.42it/s]
3|2281|Loss: 0.8135302066802979:  35%|███▌      | 2281/6470 [26:13<52:48,  1.32it/s]
3|2282|Loss: 0.8909655213356018:  35%|███▌      | 2281/6470 [26:14<52:48,  1.32it/s]
3|2282|Loss: 0.8909655213356018:  35%|███▌      | 2282/6470 [26:14<50:50,  1.37it/s]
3|2283|Loss: 0.7184644937515259:  35%|███▌      | 2282/6470 [26:14<50:50,  1.37it/s]
3|2283|Loss: 0.7184644937515259:  35%|███▌      | 2283/6470 [26:15<50:18,  1.39it/s]
3|2284|Loss: 1.015019416809082:  35%|███▌      | 2283/6470 [26:15<50:18,  1.39it/s] 
3|2284|Loss: 1.015019416809082:  35%|███▌      | 2284/6470 [26:15<52:24,  1.33it/s]
3|2285|Loss: 0.8113382458686829:  35%|███▌      | 2284/6470 [26:16<52:24,  1.33it/s]
3|2285|Loss: 0.8113382458686829:  35%|███▌      | 2285/6470 [26:16<51:13,  1.36it/s]
3|2286|Loss: 0.7783196568489075:  35%|███▌      | 2285/6470 [26:16<51:13,  1.36it/s]
3|2286|Loss: 0.7783196568489075:  35%|███▌      | 2286/6470 [26:17<47:22,  1.47it/s]
3|2287|Loss: 0.8555340766906738:  35%|███▌      | 2286/6470 [26:17<47:22,  1.47it/s]
3|2287|Loss: 0.8555340766906738:  35%|███▌      | 2287/6470 [26:17<45:37,  1.53it/s]
3|2288|Loss: 0.9625257849693298:  35%|███▌      | 2287/6470 [26:18<45:37,  1.53it/s]
3|2288|Loss: 0.9625257849693298:  35%|███▌      | 2288/6470 [26:18<46:51,  1.49it/s]
3|2289|Loss: 0.7889941334724426:  35%|███▌      | 2288/6470 [26:18<46:51,  1.49it/s]
3|2289|Loss: 0.7889941334724426:  35%|███▌      | 2289/6470 [26:19<48:18,  1.44it/s]
3|2290|Loss: 0.829133927822113:  35%|███▌      | 2289/6470 [26:19<48:18,  1.44it/s] 
3|2290|Loss: 0.829133927822113:  35%|███▌      | 2290/6470 [26:19<47:39,  1.46it/s]
3|2291|Loss: 0.8109465837478638:  35%|███▌      | 2290/6470 [26:20<47:39,  1.46it/s]
3|2291|Loss: 0.8109465837478638:  35%|███▌      | 2291/6470 [26:20<44:45,  1.56it/s]
3|2292|Loss: 0.9343255758285522:  35%|███▌      | 2291/6470 [26:20<44:45,  1.56it/s]
3|2292|Loss: 0.9343255758285522:  35%|███▌      | 2292/6470 [26:21<43:03,  1.62it/s]
3|2293|Loss: 0.9915981888771057:  35%|███▌      | 2292/6470 [26:21<43:03,  1.62it/s]
3|2293|Loss: 0.9915981888771057:  35%|███▌      | 2293/6470 [26:21<42:14,  1.65it/s]
3|2294|Loss: 0.9814618825912476:  35%|███▌      | 2293/6470 [26:21<42:14,  1.65it/s]
3|2294|Loss: 0.9814618825912476:  35%|███▌      | 2294/6470 [26:22<41:46,  1.67it/s]
3|2295|Loss: 1.0614030361175537:  35%|███▌      | 2294/6470 [26:22<41:46,  1.67it/s]
3|2295|Loss: 1.0614030361175537:  35%|███▌      | 2295/6470 [26:22<38:10,  1.82it/s]
3|2296|Loss: 0.9498035311698914:  35%|███▌      | 2295/6470 [26:22<38:10,  1.82it/s]
3|2296|Loss: 0.9498035311698914:  35%|███▌      | 2296/6470 [26:23<41:03,  1.69it/s]
3|2297|Loss: 0.7411403656005859:  35%|███▌      | 2296/6470 [26:23<41:03,  1.69it/s]
3|2297|Loss: 0.7411403656005859:  36%|███▌      | 2297/6470 [26:23<42:24,  1.64it/s]
3|2298|Loss: 0.8162574768066406:  36%|███▌      | 2297/6470 [26:24<42:24,  1.64it/s]
3|2298|Loss: 0.8162574768066406:  36%|███▌      | 2298/6470 [26:24<43:59,  1.58it/s]
3|2299|Loss: 0.7498562335968018:  36%|███▌      | 2298/6470 [26:25<43:59,  1.58it/s]
3|2299|Loss: 0.7498562335968018:  36%|███▌      | 2299/6470 [26:25<43:34,  1.60it/s]
3|2300|Loss: 0.7706157565116882:  36%|███▌      | 2299/6470 [26:25<43:34,  1.60it/s]
3|2300|Loss: 0.7706157565116882:  36%|███▌      | 2300/6470 [26:26<47:14,  1.47it/s]
3|2301|Loss: 0.7764390110969543:  36%|███▌      | 2300/6470 [26:26<47:14,  1.47it/s]
3|2301|Loss: 0.7764390110969543:  36%|███▌      | 2301/6470 [26:26<45:55,  1.51it/s]
3|2302|Loss: 0.7505797147750854:  36%|███▌      | 2301/6470 [26:27<45:55,  1.51it/s]
3|2302|Loss: 0.7505797147750854:  36%|███▌      | 2302/6470 [26:27<46:05,  1.51it/s]
3|2303|Loss: 0.7614943981170654:  36%|███▌      | 2302/6470 [26:27<46:05,  1.51it/s]
3|2303|Loss: 0.7614943981170654:  36%|███▌      | 2303/6470 [26:28<46:54,  1.48it/s]
3|2304|Loss: 0.8467477560043335:  36%|███▌      | 2303/6470 [26:28<46:54,  1.48it/s]
3|2304|Loss: 0.8467477560043335:  36%|███▌      | 2304/6470 [26:28<47:27,  1.46it/s]
3|2305|Loss: 0.882762610912323:  36%|███▌      | 2304/6470 [26:29<47:27,  1.46it/s] 
3|2305|Loss: 0.882762610912323:  36%|███▌      | 2305/6470 [26:29<46:29,  1.49it/s]
3|2306|Loss: 0.9940575957298279:  36%|███▌      | 2305/6470 [26:29<46:29,  1.49it/s]
3|2306|Loss: 0.9940575957298279:  36%|███▌      | 2306/6470 [26:29<43:39,  1.59it/s]
3|2307|Loss: 0.8411198854446411:  36%|███▌      | 2306/6470 [26:30<43:39,  1.59it/s]
3|2307|Loss: 0.8411198854446411:  36%|███▌      | 2307/6470 [26:30<43:27,  1.60it/s]
3|2308|Loss: 0.8510792255401611:  36%|███▌      | 2307/6470 [26:30<43:27,  1.60it/s]
3|2308|Loss: 0.8510792255401611:  36%|███▌      | 2308/6470 [26:31<45:22,  1.53it/s]
3|2309|Loss: 0.7660808563232422:  36%|███▌      | 2308/6470 [26:31<45:22,  1.53it/s]
3|2309|Loss: 0.7660808563232422:  36%|███▌      | 2309/6470 [26:32<50:02,  1.39it/s]
3|2310|Loss: 0.7705650329589844:  36%|███▌      | 2309/6470 [26:32<50:02,  1.39it/s]
3|2310|Loss: 0.7705650329589844:  36%|███▌      | 2310/6470 [26:32<49:03,  1.41it/s]
3|2311|Loss: 0.7300960421562195:  36%|███▌      | 2310/6470 [26:33<49:03,  1.41it/s]
3|2311|Loss: 0.7300960421562195:  36%|███▌      | 2311/6470 [26:33<48:30,  1.43it/s]
3|2312|Loss: 0.9445740580558777:  36%|███▌      | 2311/6470 [26:33<48:30,  1.43it/s]
3|2312|Loss: 0.9445740580558777:  36%|███▌      | 2312/6470 [26:34<47:16,  1.47it/s]
3|2313|Loss: 0.7541975378990173:  36%|███▌      | 2312/6470 [26:34<47:16,  1.47it/s]
3|2313|Loss: 0.7541975378990173:  36%|███▌      | 2313/6470 [26:34<43:16,  1.60it/s]
3|2314|Loss: 0.9118891954421997:  36%|███▌      | 2313/6470 [26:34<43:16,  1.60it/s]
3|2314|Loss: 0.9118891954421997:  36%|███▌      | 2314/6470 [26:35<45:15,  1.53it/s]
3|2315|Loss: 0.784708559513092:  36%|███▌      | 2314/6470 [26:35<45:15,  1.53it/s] 
3|2315|Loss: 0.784708559513092:  36%|███▌      | 2315/6470 [26:35<43:59,  1.57it/s]
3|2316|Loss: 0.9658553600311279:  36%|███▌      | 2315/6470 [26:36<43:59,  1.57it/s]
3|2316|Loss: 0.9658553600311279:  36%|███▌      | 2316/6470 [26:36<46:11,  1.50it/s]
3|2317|Loss: 0.7410117983818054:  36%|███▌      | 2316/6470 [26:37<46:11,  1.50it/s]
3|2317|Loss: 0.7410117983818054:  36%|███▌      | 2317/6470 [26:37<47:23,  1.46it/s]
3|2318|Loss: 0.7742655873298645:  36%|███▌      | 2317/6470 [26:37<47:23,  1.46it/s]
3|2318|Loss: 0.7742655873298645:  36%|███▌      | 2318/6470 [26:38<49:15,  1.40it/s]
3|2319|Loss: 0.99619060754776:  36%|███▌      | 2318/6470 [26:38<49:15,  1.40it/s]  
3|2319|Loss: 0.99619060754776:  36%|███▌      | 2319/6470 [26:38<49:57,  1.38it/s]
3|2320|Loss: 0.8316704630851746:  36%|███▌      | 2319/6470 [26:39<49:57,  1.38it/s]
3|2320|Loss: 0.8316704630851746:  36%|███▌      | 2320/6470 [26:39<52:26,  1.32it/s]
3|2321|Loss: 0.7745029330253601:  36%|███▌      | 2320/6470 [26:40<52:26,  1.32it/s]
3|2321|Loss: 0.7745029330253601:  36%|███▌      | 2321/6470 [26:40<50:16,  1.38it/s]
3|2322|Loss: 0.8495396971702576:  36%|███▌      | 2321/6470 [26:40<50:16,  1.38it/s]
3|2322|Loss: 0.8495396971702576:  36%|███▌      | 2322/6470 [26:41<48:11,  1.43it/s]
3|2323|Loss: 0.8120586276054382:  36%|███▌      | 2322/6470 [26:41<48:11,  1.43it/s]
3|2323|Loss: 0.8120586276054382:  36%|███▌      | 2323/6470 [26:41<46:06,  1.50it/s]
3|2324|Loss: 0.7916290760040283:  36%|███▌      | 2323/6470 [26:42<46:06,  1.50it/s]
3|2324|Loss: 0.7916290760040283:  36%|███▌      | 2324/6470 [26:42<45:23,  1.52it/s]
3|2325|Loss: 0.9392667412757874:  36%|███▌      | 2324/6470 [26:42<45:23,  1.52it/s]
3|2325|Loss: 0.9392667412757874:  36%|███▌      | 2325/6470 [26:42<44:10,  1.56it/s]
3|2326|Loss: 0.8644800782203674:  36%|███▌      | 2325/6470 [26:43<44:10,  1.56it/s]
3|2326|Loss: 0.8644800782203674:  36%|███▌      | 2326/6470 [26:43<45:30,  1.52it/s]
3|2327|Loss: 0.7720662951469421:  36%|███▌      | 2326/6470 [26:43<45:30,  1.52it/s]
3|2327|Loss: 0.7720662951469421:  36%|███▌      | 2327/6470 [26:44<43:41,  1.58it/s]
3|2328|Loss: 0.8351151943206787:  36%|███▌      | 2327/6470 [26:44<43:41,  1.58it/s]
3|2328|Loss: 0.8351151943206787:  36%|███▌      | 2328/6470 [26:44<41:15,  1.67it/s]
3|2329|Loss: 0.92134690284729:  36%|███▌      | 2328/6470 [26:45<41:15,  1.67it/s]  
3|2329|Loss: 0.92134690284729:  36%|███▌      | 2329/6470 [26:45<43:03,  1.60it/s]
3|2330|Loss: 0.8534983396530151:  36%|███▌      | 2329/6470 [26:45<43:03,  1.60it/s]
3|2330|Loss: 0.8534983396530151:  36%|███▌      | 2330/6470 [26:46<43:56,  1.57it/s]
3|2331|Loss: 0.8257591128349304:  36%|███▌      | 2330/6470 [26:46<43:56,  1.57it/s]
3|2331|Loss: 0.8257591128349304:  36%|███▌      | 2331/6470 [26:46<43:59,  1.57it/s]
3|2332|Loss: 0.7642393708229065:  36%|███▌      | 2331/6470 [26:47<43:59,  1.57it/s]
3|2332|Loss: 0.7642393708229065:  36%|███▌      | 2332/6470 [26:47<47:28,  1.45it/s]
3|2333|Loss: 0.9202969074249268:  36%|███▌      | 2332/6470 [26:47<47:28,  1.45it/s]
3|2333|Loss: 0.9202969074249268:  36%|███▌      | 2333/6470 [26:48<47:31,  1.45it/s]
3|2334|Loss: 0.6926174163818359:  36%|███▌      | 2333/6470 [26:48<47:31,  1.45it/s]
3|2334|Loss: 0.6926174163818359:  36%|███▌      | 2334/6470 [26:48<45:22,  1.52it/s]
3|2335|Loss: 0.8179954886436462:  36%|███▌      | 2334/6470 [26:49<45:22,  1.52it/s]
3|2335|Loss: 0.8179954886436462:  36%|███▌      | 2335/6470 [26:49<48:23,  1.42it/s]
3|2336|Loss: 0.8309665322303772:  36%|███▌      | 2335/6470 [26:49<48:23,  1.42it/s]
3|2336|Loss: 0.8309665322303772:  36%|███▌      | 2336/6470 [26:50<48:03,  1.43it/s]
3|2337|Loss: 0.9081887602806091:  36%|███▌      | 2336/6470 [26:50<48:03,  1.43it/s]
3|2337|Loss: 0.9081887602806091:  36%|███▌      | 2337/6470 [26:50<48:01,  1.43it/s]
3|2338|Loss: 0.871023952960968:  36%|███▌      | 2337/6470 [26:51<48:01,  1.43it/s] 
3|2338|Loss: 0.871023952960968:  36%|███▌      | 2338/6470 [26:51<46:41,  1.48it/s]
3|2339|Loss: 0.7745935320854187:  36%|███▌      | 2338/6470 [26:51<46:41,  1.48it/s]
3|2339|Loss: 0.7745935320854187:  36%|███▌      | 2339/6470 [26:52<48:54,  1.41it/s]
3|2340|Loss: 0.99347984790802:  36%|███▌      | 2339/6470 [26:52<48:54,  1.41it/s]  
3|2340|Loss: 0.99347984790802:  36%|███▌      | 2340/6470 [26:53<49:18,  1.40it/s]
3|2341|Loss: 0.8601689338684082:  36%|███▌      | 2340/6470 [26:53<49:18,  1.40it/s]
3|2341|Loss: 0.8601689338684082:  36%|███▌      | 2341/6470 [26:53<49:38,  1.39it/s]
3|2342|Loss: 0.9563971161842346:  36%|███▌      | 2341/6470 [26:54<49:38,  1.39it/s]
3|2342|Loss: 0.9563971161842346:  36%|███▌      | 2342/6470 [26:54<49:53,  1.38it/s]
3|2343|Loss: 0.9105236530303955:  36%|███▌      | 2342/6470 [26:55<49:53,  1.38it/s]
3|2343|Loss: 0.9105236530303955:  36%|███▌      | 2343/6470 [26:55<51:59,  1.32it/s]
3|2344|Loss: 0.9022316932678223:  36%|███▌      | 2343/6470 [26:55<51:59,  1.32it/s]
3|2344|Loss: 0.9022316932678223:  36%|███▌      | 2344/6470 [26:56<52:24,  1.31it/s]
3|2345|Loss: 0.8871018290519714:  36%|███▌      | 2344/6470 [26:56<52:24,  1.31it/s]
3|2345|Loss: 0.8871018290519714:  36%|███▌      | 2345/6470 [26:56<49:23,  1.39it/s]
3|2346|Loss: 0.8364208936691284:  36%|███▌      | 2345/6470 [26:57<49:23,  1.39it/s]
3|2346|Loss: 0.8364208936691284:  36%|███▋      | 2346/6470 [26:57<48:38,  1.41it/s]
3|2347|Loss: 0.7977603077888489:  36%|███▋      | 2346/6470 [26:57<48:38,  1.41it/s]
3|2347|Loss: 0.7977603077888489:  36%|███▋      | 2347/6470 [26:58<48:38,  1.41it/s]
3|2348|Loss: 0.8619317412376404:  36%|███▋      | 2347/6470 [26:58<48:38,  1.41it/s]
3|2348|Loss: 0.8619317412376404:  36%|███▋      | 2348/6470 [26:58<47:20,  1.45it/s]
3|2349|Loss: 0.7548696994781494:  36%|███▋      | 2348/6470 [26:59<47:20,  1.45it/s]
3|2349|Loss: 0.7548696994781494:  36%|███▋      | 2349/6470 [26:59<46:11,  1.49it/s]
3|2350|Loss: 0.8534957766532898:  36%|███▋      | 2349/6470 [26:59<46:11,  1.49it/s]
3|2350|Loss: 0.8534957766532898:  36%|███▋      | 2350/6470 [27:00<48:05,  1.43it/s]
3|2351|Loss: 0.8046798706054688:  36%|███▋      | 2350/6470 [27:00<48:05,  1.43it/s]
3|2351|Loss: 0.8046798706054688:  36%|███▋      | 2351/6470 [27:00<48:32,  1.41it/s]
3|2352|Loss: 0.6718270778656006:  36%|███▋      | 2351/6470 [27:01<48:32,  1.41it/s]
3|2352|Loss: 0.6718270778656006:  36%|███▋      | 2352/6470 [27:01<48:52,  1.40it/s]
3|2353|Loss: 0.8316864967346191:  36%|███▋      | 2352/6470 [27:02<48:52,  1.40it/s]
3|2353|Loss: 0.8316864967346191:  36%|███▋      | 2353/6470 [27:02<47:19,  1.45it/s]
3|2354|Loss: 0.8484683036804199:  36%|███▋      | 2353/6470 [27:02<47:19,  1.45it/s]
3|2354|Loss: 0.8484683036804199:  36%|███▋      | 2354/6470 [27:02<44:37,  1.54it/s]
3|2355|Loss: 0.8764179944992065:  36%|███▋      | 2354/6470 [27:03<44:37,  1.54it/s]
3|2355|Loss: 0.8764179944992065:  36%|███▋      | 2355/6470 [27:03<45:48,  1.50it/s]
3|2356|Loss: 0.9215790629386902:  36%|███▋      | 2355/6470 [27:04<45:48,  1.50it/s]
3|2356|Loss: 0.9215790629386902:  36%|███▋      | 2356/6470 [27:04<49:08,  1.40it/s]
3|2357|Loss: 0.8770482540130615:  36%|███▋      | 2356/6470 [27:04<49:08,  1.40it/s]
3|2357|Loss: 0.8770482540130615:  36%|███▋      | 2357/6470 [27:05<51:45,  1.32it/s]
3|2358|Loss: 0.9479207396507263:  36%|███▋      | 2357/6470 [27:05<51:45,  1.32it/s]
3|2358|Loss: 0.9479207396507263:  36%|███▋      | 2358/6470 [27:05<49:55,  1.37it/s]
3|2359|Loss: 0.8629340529441833:  36%|███▋      | 2358/6470 [27:06<49:55,  1.37it/s]
3|2359|Loss: 0.8629340529441833:  36%|███▋      | 2359/6470 [27:06<47:55,  1.43it/s]
3|2360|Loss: 0.95883709192276:  36%|███▋      | 2359/6470 [27:06<47:55,  1.43it/s]  
3|2360|Loss: 0.95883709192276:  36%|███▋      | 2360/6470 [27:07<46:46,  1.46it/s]
3|2361|Loss: 0.8567434549331665:  36%|███▋      | 2360/6470 [27:07<46:46,  1.46it/s]
3|2361|Loss: 0.8567434549331665:  36%|███▋      | 2361/6470 [27:07<44:22,  1.54it/s]
3|2362|Loss: 0.7876556515693665:  36%|███▋      | 2361/6470 [27:08<44:22,  1.54it/s]
3|2362|Loss: 0.7876556515693665:  37%|███▋      | 2362/6470 [27:08<47:03,  1.45it/s]
3|2363|Loss: 0.8007792830467224:  37%|███▋      | 2362/6470 [27:08<47:03,  1.45it/s]
3|2363|Loss: 0.8007792830467224:  37%|███▋      | 2363/6470 [27:09<50:06,  1.37it/s]
3|2364|Loss: 0.8690575957298279:  37%|███▋      | 2363/6470 [27:09<50:06,  1.37it/s]
3|2364|Loss: 0.8690575957298279:  37%|███▋      | 2364/6470 [27:09<44:57,  1.52it/s]
3|2365|Loss: 0.7975290417671204:  37%|███▋      | 2364/6470 [27:10<44:57,  1.52it/s]
3|2365|Loss: 0.7975290417671204:  37%|███▋      | 2365/6470 [27:10<45:36,  1.50it/s]
3|2366|Loss: 0.8735909461975098:  37%|███▋      | 2365/6470 [27:10<45:36,  1.50it/s]
3|2366|Loss: 0.8735909461975098:  37%|███▋      | 2366/6470 [27:11<49:15,  1.39it/s]
3|2367|Loss: 0.8924596905708313:  37%|███▋      | 2366/6470 [27:11<49:15,  1.39it/s]
3|2367|Loss: 0.8924596905708313:  37%|███▋      | 2367/6470 [27:11<43:59,  1.55it/s]
3|2368|Loss: 0.9428069591522217:  37%|███▋      | 2367/6470 [27:12<43:59,  1.55it/s]
3|2368|Loss: 0.9428069591522217:  37%|███▋      | 2368/6470 [27:12<43:54,  1.56it/s]
3|2369|Loss: 0.8156840205192566:  37%|███▋      | 2368/6470 [27:12<43:54,  1.56it/s]
3|2369|Loss: 0.8156840205192566:  37%|███▋      | 2369/6470 [27:13<43:47,  1.56it/s]
3|2370|Loss: 0.9095484018325806:  37%|███▋      | 2369/6470 [27:13<43:47,  1.56it/s]
3|2370|Loss: 0.9095484018325806:  37%|███▋      | 2370/6470 [27:13<43:49,  1.56it/s]
3|2371|Loss: 0.8505974411964417:  37%|███▋      | 2370/6470 [27:14<43:49,  1.56it/s]
3|2371|Loss: 0.8505974411964417:  37%|███▋      | 2371/6470 [27:14<43:55,  1.56it/s]
3|2372|Loss: 0.7875744104385376:  37%|███▋      | 2371/6470 [27:14<43:55,  1.56it/s]
3|2372|Loss: 0.7875744104385376:  37%|███▋      | 2372/6470 [27:15<42:48,  1.60it/s]
3|2373|Loss: 0.8877948522567749:  37%|███▋      | 2372/6470 [27:15<42:48,  1.60it/s]
3|2373|Loss: 0.8877948522567749:  37%|███▋      | 2373/6470 [27:15<41:57,  1.63it/s]
3|2374|Loss: 0.8192970752716064:  37%|███▋      | 2373/6470 [27:15<41:57,  1.63it/s]
3|2374|Loss: 0.8192970752716064:  37%|███▋      | 2374/6470 [27:16<43:15,  1.58it/s]
3|2375|Loss: 0.9809838533401489:  37%|███▋      | 2374/6470 [27:16<43:15,  1.58it/s]
3|2375|Loss: 0.9809838533401489:  37%|███▋      | 2375/6470 [27:17<46:06,  1.48it/s]
3|2376|Loss: 0.75629061460495:  37%|███▋      | 2375/6470 [27:17<46:06,  1.48it/s]  
3|2376|Loss: 0.75629061460495:  37%|███▋      | 2376/6470 [27:17<45:48,  1.49it/s]
3|2377|Loss: 0.8265482783317566:  37%|███▋      | 2376/6470 [27:18<45:48,  1.49it/s]
3|2377|Loss: 0.8265482783317566:  37%|███▋      | 2377/6470 [27:18<46:37,  1.46it/s]
3|2378|Loss: 0.9488183856010437:  37%|███▋      | 2377/6470 [27:18<46:37,  1.46it/s]
3|2378|Loss: 0.9488183856010437:  37%|███▋      | 2378/6470 [27:19<48:25,  1.41it/s]
3|2379|Loss: 0.8586799502372742:  37%|███▋      | 2378/6470 [27:19<48:25,  1.41it/s]
3|2379|Loss: 0.8586799502372742:  37%|███▋      | 2379/6470 [27:20<51:36,  1.32it/s]
3|2380|Loss: 0.8641987442970276:  37%|███▋      | 2379/6470 [27:20<51:36,  1.32it/s]
3|2380|Loss: 0.8641987442970276:  37%|███▋      | 2380/6470 [27:20<54:22,  1.25it/s]
3|2381|Loss: 0.8003661036491394:  37%|███▋      | 2380/6470 [27:21<54:22,  1.25it/s]
3|2381|Loss: 0.8003661036491394:  37%|███▋      | 2381/6470 [27:21<55:48,  1.22it/s]
3|2382|Loss: 0.752108633518219:  37%|███▋      | 2381/6470 [27:22<55:48,  1.22it/s] 
3|2382|Loss: 0.752108633518219:  37%|███▋      | 2382/6470 [27:22<54:36,  1.25it/s]
3|2383|Loss: 0.7831494808197021:  37%|███▋      | 2382/6470 [27:23<54:36,  1.25it/s]
3|2383|Loss: 0.7831494808197021:  37%|███▋      | 2383/6470 [27:23<55:04,  1.24it/s]
3|2384|Loss: 0.9804189205169678:  37%|███▋      | 2383/6470 [27:23<55:04,  1.24it/s]
3|2384|Loss: 0.9804189205169678:  37%|███▋      | 2384/6470 [27:24<54:04,  1.26it/s]
3|2385|Loss: 0.8994954228401184:  37%|███▋      | 2384/6470 [27:24<54:04,  1.26it/s]
3|2385|Loss: 0.8994954228401184:  37%|███▋      | 2385/6470 [27:24<52:59,  1.28it/s]
3|2386|Loss: 0.8612220883369446:  37%|███▋      | 2385/6470 [27:25<52:59,  1.28it/s]
3|2386|Loss: 0.8612220883369446:  37%|███▋      | 2386/6470 [27:25<48:41,  1.40it/s]
3|2387|Loss: 0.8603261709213257:  37%|███▋      | 2386/6470 [27:25<48:41,  1.40it/s]
3|2387|Loss: 0.8603261709213257:  37%|███▋      | 2387/6470 [27:26<48:26,  1.40it/s]
3|2388|Loss: 0.8421589732170105:  37%|███▋      | 2387/6470 [27:26<48:26,  1.40it/s]
3|2388|Loss: 0.8421589732170105:  37%|███▋      | 2388/6470 [27:26<50:42,  1.34it/s]
3|2389|Loss: 0.9389176964759827:  37%|███▋      | 2388/6470 [27:27<50:42,  1.34it/s]
3|2389|Loss: 0.9389176964759827:  37%|███▋      | 2389/6470 [27:27<51:03,  1.33it/s]
3|2390|Loss: 0.8141443729400635:  37%|███▋      | 2389/6470 [27:28<51:03,  1.33it/s]
3|2390|Loss: 0.8141443729400635:  37%|███▋      | 2390/6470 [27:28<48:46,  1.39it/s]
3|2391|Loss: 0.8763602375984192:  37%|███▋      | 2390/6470 [27:28<48:46,  1.39it/s]
3|2391|Loss: 0.8763602375984192:  37%|███▋      | 2391/6470 [27:29<48:36,  1.40it/s]
3|2392|Loss: 0.7497485876083374:  37%|███▋      | 2391/6470 [27:29<48:36,  1.40it/s]
3|2392|Loss: 0.7497485876083374:  37%|███▋      | 2392/6470 [27:29<47:36,  1.43it/s]
3|2393|Loss: 0.9252033829689026:  37%|███▋      | 2392/6470 [27:30<47:36,  1.43it/s]
3|2393|Loss: 0.9252033829689026:  37%|███▋      | 2393/6470 [27:30<46:22,  1.47it/s]
3|2394|Loss: 0.8829063177108765:  37%|███▋      | 2393/6470 [27:30<46:22,  1.47it/s]
3|2394|Loss: 0.8829063177108765:  37%|███▋      | 2394/6470 [27:31<47:49,  1.42it/s]
3|2395|Loss: 0.8289328813552856:  37%|███▋      | 2394/6470 [27:31<47:49,  1.42it/s]
3|2395|Loss: 0.8289328813552856:  37%|███▋      | 2395/6470 [27:31<49:35,  1.37it/s]
3|2396|Loss: 0.9282276034355164:  37%|███▋      | 2395/6470 [27:32<49:35,  1.37it/s]
3|2396|Loss: 0.9282276034355164:  37%|███▋      | 2396/6470 [27:32<51:33,  1.32it/s]
3|2397|Loss: 0.766491174697876:  37%|███▋      | 2396/6470 [27:33<51:33,  1.32it/s] 
3|2397|Loss: 0.766491174697876:  37%|███▋      | 2397/6470 [27:33<50:03,  1.36it/s]
3|2398|Loss: 0.7777132987976074:  37%|███▋      | 2397/6470 [27:33<50:03,  1.36it/s]
3|2398|Loss: 0.7777132987976074:  37%|███▋      | 2398/6470 [27:34<49:47,  1.36it/s]
3|2399|Loss: 0.9533438086509705:  37%|███▋      | 2398/6470 [27:34<49:47,  1.36it/s]
3|2399|Loss: 0.9533438086509705:  37%|███▋      | 2399/6470 [27:34<49:53,  1.36it/s]
3|2400|Loss: 0.928808867931366:  37%|███▋      | 2399/6470 [27:35<49:53,  1.36it/s] 
3|2400|Loss: 0.928808867931366:  37%|███▋      | 2400/6470 [27:35<50:55,  1.33it/s]
3|2401|Loss: 0.9713107943534851:  37%|███▋      | 2400/6470 [27:36<50:55,  1.33it/s]
3|2401|Loss: 0.9713107943534851:  37%|███▋      | 2401/6470 [27:36<51:55,  1.31it/s]
3|2402|Loss: 0.7826066017150879:  37%|███▋      | 2401/6470 [27:36<51:55,  1.31it/s]
3|2402|Loss: 0.7826066017150879:  37%|███▋      | 2402/6470 [27:37<52:25,  1.29it/s]
3|2403|Loss: 0.7690563201904297:  37%|███▋      | 2402/6470 [27:37<52:25,  1.29it/s]
3|2403|Loss: 0.7690563201904297:  37%|███▋      | 2403/6470 [27:38<51:31,  1.32it/s]
3|2404|Loss: 0.9539283514022827:  37%|███▋      | 2403/6470 [27:38<51:31,  1.32it/s]
3|2404|Loss: 0.9539283514022827:  37%|███▋      | 2404/6470 [27:38<51:05,  1.33it/s]
3|2405|Loss: 1.0402427911758423:  37%|███▋      | 2404/6470 [27:39<51:05,  1.33it/s]
3|2405|Loss: 1.0402427911758423:  37%|███▋      | 2405/6470 [27:39<49:22,  1.37it/s]
3|2406|Loss: 0.8498359322547913:  37%|███▋      | 2405/6470 [27:39<49:22,  1.37it/s]
3|2406|Loss: 0.8498359322547913:  37%|███▋      | 2406/6470 [27:40<50:37,  1.34it/s]
3|2407|Loss: 0.7023524045944214:  37%|███▋      | 2406/6470 [27:40<50:37,  1.34it/s]
3|2407|Loss: 0.7023524045944214:  37%|███▋      | 2407/6470 [27:41<50:46,  1.33it/s]
3|2408|Loss: 0.9771463871002197:  37%|███▋      | 2407/6470 [27:41<50:46,  1.33it/s]
3|2408|Loss: 0.9771463871002197:  37%|███▋      | 2408/6470 [27:41<50:28,  1.34it/s]
3|2409|Loss: 0.8316168189048767:  37%|███▋      | 2408/6470 [27:42<50:28,  1.34it/s]
3|2409|Loss: 0.8316168189048767:  37%|███▋      | 2409/6470 [27:42<51:03,  1.33it/s]
3|2410|Loss: 0.8918049931526184:  37%|███▋      | 2409/6470 [27:42<51:03,  1.33it/s]
3|2410|Loss: 0.8918049931526184:  37%|███▋      | 2410/6470 [27:43<54:00,  1.25it/s]
3|2411|Loss: 0.9310534000396729:  37%|███▋      | 2410/6470 [27:43<54:00,  1.25it/s]
3|2411|Loss: 0.9310534000396729:  37%|███▋      | 2411/6470 [27:44<53:04,  1.27it/s]
3|2412|Loss: 1.021033525466919:  37%|███▋      | 2411/6470 [27:44<53:04,  1.27it/s] 
3|2412|Loss: 1.021033525466919:  37%|███▋      | 2412/6470 [27:44<52:00,  1.30it/s]
3|2413|Loss: 0.8160117864608765:  37%|███▋      | 2412/6470 [27:45<52:00,  1.30it/s]
3|2413|Loss: 0.8160117864608765:  37%|███▋      | 2413/6470 [27:45<51:30,  1.31it/s]
3|2414|Loss: 0.9221246838569641:  37%|███▋      | 2413/6470 [27:46<51:30,  1.31it/s]
3|2414|Loss: 0.9221246838569641:  37%|███▋      | 2414/6470 [27:46<49:17,  1.37it/s]
3|2415|Loss: 0.8484589457511902:  37%|███▋      | 2414/6470 [27:46<49:17,  1.37it/s]
3|2415|Loss: 0.8484589457511902:  37%|███▋      | 2415/6470 [27:47<51:01,  1.32it/s]
3|2416|Loss: 0.7654343843460083:  37%|███▋      | 2415/6470 [27:47<51:01,  1.32it/s]
3|2416|Loss: 0.7654343843460083:  37%|███▋      | 2416/6470 [27:47<52:50,  1.28it/s]
3|2417|Loss: 0.9155424237251282:  37%|███▋      | 2416/6470 [27:48<52:50,  1.28it/s]
3|2417|Loss: 0.9155424237251282:  37%|███▋      | 2417/6470 [27:48<53:08,  1.27it/s]
3|2418|Loss: 0.9228878021240234:  37%|███▋      | 2417/6470 [27:49<53:08,  1.27it/s]
3|2418|Loss: 0.9228878021240234:  37%|███▋      | 2418/6470 [27:49<52:18,  1.29it/s]
3|2419|Loss: 0.917610228061676:  37%|███▋      | 2418/6470 [27:49<52:18,  1.29it/s] 
3|2419|Loss: 0.917610228061676:  37%|███▋      | 2419/6470 [27:50<49:01,  1.38it/s]
3|2420|Loss: 0.8400087356567383:  37%|███▋      | 2419/6470 [27:50<49:01,  1.38it/s]
3|2420|Loss: 0.8400087356567383:  37%|███▋      | 2420/6470 [27:50<48:17,  1.40it/s]
3|2421|Loss: 0.9511798620223999:  37%|███▋      | 2420/6470 [27:51<48:17,  1.40it/s]
3|2421|Loss: 0.9511798620223999:  37%|███▋      | 2421/6470 [27:51<48:37,  1.39it/s]
3|2422|Loss: 0.8904830813407898:  37%|███▋      | 2421/6470 [27:51<48:37,  1.39it/s]
3|2422|Loss: 0.8904830813407898:  37%|███▋      | 2422/6470 [27:52<50:55,  1.32it/s]
3|2423|Loss: 0.7829316258430481:  37%|███▋      | 2422/6470 [27:52<50:55,  1.32it/s]
3|2423|Loss: 0.7829316258430481:  37%|███▋      | 2423/6470 [27:53<49:09,  1.37it/s]
3|2424|Loss: 0.8514693379402161:  37%|███▋      | 2423/6470 [27:53<49:09,  1.37it/s]
3|2424|Loss: 0.8514693379402161:  37%|███▋      | 2424/6470 [27:53<48:37,  1.39it/s]
3|2425|Loss: 0.9421108961105347:  37%|███▋      | 2424/6470 [27:54<48:37,  1.39it/s]
3|2425|Loss: 0.9421108961105347:  37%|███▋      | 2425/6470 [27:54<48:15,  1.40it/s]
3|2426|Loss: 0.9230977892875671:  37%|███▋      | 2425/6470 [27:54<48:15,  1.40it/s]
3|2426|Loss: 0.9230977892875671:  37%|███▋      | 2426/6470 [27:55<46:59,  1.43it/s]
3|2427|Loss: 0.956813395023346:  37%|███▋      | 2426/6470 [27:55<46:59,  1.43it/s] 
3|2427|Loss: 0.956813395023346:  38%|███▊      | 2427/6470 [27:55<44:58,  1.50it/s]
3|2428|Loss: 0.8672646880149841:  38%|███▊      | 2427/6470 [27:56<44:58,  1.50it/s]
3|2428|Loss: 0.8672646880149841:  38%|███▊      | 2428/6470 [27:56<49:20,  1.37it/s]
3|2429|Loss: 0.8386948108673096:  38%|███▊      | 2428/6470 [27:57<49:20,  1.37it/s]
3|2429|Loss: 0.8386948108673096:  38%|███▊      | 2429/6470 [27:57<50:45,  1.33it/s]
3|2430|Loss: 0.7711214423179626:  38%|███▊      | 2429/6470 [27:57<50:45,  1.33it/s]
3|2430|Loss: 0.7711214423179626:  38%|███▊      | 2430/6470 [27:58<48:40,  1.38it/s]
3|2431|Loss: 0.8037315011024475:  38%|███▊      | 2430/6470 [27:58<48:40,  1.38it/s]
3|2431|Loss: 0.8037315011024475:  38%|███▊      | 2431/6470 [27:58<52:03,  1.29it/s]
3|2432|Loss: 0.6838141679763794:  38%|███▊      | 2431/6470 [27:59<52:03,  1.29it/s]
3|2432|Loss: 0.6838141679763794:  38%|███▊      | 2432/6470 [27:59<53:55,  1.25it/s]
3|2433|Loss: 0.708276629447937:  38%|███▊      | 2432/6470 [28:00<53:55,  1.25it/s] 
3|2433|Loss: 0.708276629447937:  38%|███▊      | 2433/6470 [28:00<52:37,  1.28it/s]
3|2434|Loss: 0.8479673266410828:  38%|███▊      | 2433/6470 [28:00<52:37,  1.28it/s]
3|2434|Loss: 0.8479673266410828:  38%|███▊      | 2434/6470 [28:01<51:27,  1.31it/s]
3|2435|Loss: 0.8623641729354858:  38%|███▊      | 2434/6470 [28:01<51:27,  1.31it/s]
3|2435|Loss: 0.8623641729354858:  38%|███▊      | 2435/6470 [28:02<51:06,  1.32it/s]
3|2436|Loss: 0.8477029204368591:  38%|███▊      | 2435/6470 [28:02<51:06,  1.32it/s]
3|2436|Loss: 0.8477029204368591:  38%|███▊      | 2436/6470 [28:02<50:54,  1.32it/s]
3|2437|Loss: 0.884093701839447:  38%|███▊      | 2436/6470 [28:03<50:54,  1.32it/s] 
3|2437|Loss: 0.884093701839447:  38%|███▊      | 2437/6470 [28:03<49:58,  1.34it/s]
3|2438|Loss: 0.707899808883667:  38%|███▊      | 2437/6470 [28:03<49:58,  1.34it/s]
3|2438|Loss: 0.707899808883667:  38%|███▊      | 2438/6470 [28:04<46:36,  1.44it/s]
3|2439|Loss: 0.8412432074546814:  38%|███▊      | 2438/6470 [28:04<46:36,  1.44it/s]
3|2439|Loss: 0.8412432074546814:  38%|███▊      | 2439/6470 [28:04<47:33,  1.41it/s]
3|2440|Loss: 0.9322414994239807:  38%|███▊      | 2439/6470 [28:05<47:33,  1.41it/s]
3|2440|Loss: 0.9322414994239807:  38%|███▊      | 2440/6470 [28:05<46:54,  1.43it/s]
3|2441|Loss: 0.8524150848388672:  38%|███▊      | 2440/6470 [28:05<46:54,  1.43it/s]
3|2441|Loss: 0.8524150848388672:  38%|███▊      | 2441/6470 [28:06<46:49,  1.43it/s]
3|2442|Loss: 0.8752470016479492:  38%|███▊      | 2441/6470 [28:06<46:49,  1.43it/s]
3|2442|Loss: 0.8752470016479492:  38%|███▊      | 2442/6470 [28:06<45:56,  1.46it/s]
3|2443|Loss: 0.7692330479621887:  38%|███▊      | 2442/6470 [28:07<45:56,  1.46it/s]
3|2443|Loss: 0.7692330479621887:  38%|███▊      | 2443/6470 [28:07<45:03,  1.49it/s]
3|2444|Loss: 0.8478724360466003:  38%|███▊      | 2443/6470 [28:07<45:03,  1.49it/s]
3|2444|Loss: 0.8478724360466003:  38%|███▊      | 2444/6470 [28:08<47:30,  1.41it/s]
3|2445|Loss: 0.8773837685585022:  38%|███▊      | 2444/6470 [28:08<47:30,  1.41it/s]
3|2445|Loss: 0.8773837685585022:  38%|███▊      | 2445/6470 [28:09<49:21,  1.36it/s]
3|2446|Loss: 0.8776825070381165:  38%|███▊      | 2445/6470 [28:09<49:21,  1.36it/s]
3|2446|Loss: 0.8776825070381165:  38%|███▊      | 2446/6470 [28:09<49:54,  1.34it/s]
3|2447|Loss: 0.8251445889472961:  38%|███▊      | 2446/6470 [28:10<49:54,  1.34it/s]
3|2447|Loss: 0.8251445889472961:  38%|███▊      | 2447/6470 [28:10<46:13,  1.45it/s]
3|2448|Loss: 0.8410131931304932:  38%|███▊      | 2447/6470 [28:10<46:13,  1.45it/s]
3|2448|Loss: 0.8410131931304932:  38%|███▊      | 2448/6470 [28:11<45:43,  1.47it/s]
3|2449|Loss: 0.8463800549507141:  38%|███▊      | 2448/6470 [28:11<45:43,  1.47it/s]
3|2449|Loss: 0.8463800549507141:  38%|███▊      | 2449/6470 [28:11<46:08,  1.45it/s]
3|2450|Loss: 0.8219295144081116:  38%|███▊      | 2449/6470 [28:12<46:08,  1.45it/s]
3|2450|Loss: 0.8219295144081116:  38%|███▊      | 2450/6470 [28:12<47:50,  1.40it/s]
3|2451|Loss: 0.7446526885032654:  38%|███▊      | 2450/6470 [28:12<47:50,  1.40it/s]
3|2451|Loss: 0.7446526885032654:  38%|███▊      | 2451/6470 [28:13<46:45,  1.43it/s]
3|2452|Loss: 0.8870972394943237:  38%|███▊      | 2451/6470 [28:13<46:45,  1.43it/s]
3|2452|Loss: 0.8870972394943237:  38%|███▊      | 2452/6470 [28:13<45:28,  1.47it/s]
3|2453|Loss: 0.7418612837791443:  38%|███▊      | 2452/6470 [28:14<45:28,  1.47it/s]
3|2453|Loss: 0.7418612837791443:  38%|███▊      | 2453/6470 [28:14<45:48,  1.46it/s]
3|2454|Loss: 0.8553897142410278:  38%|███▊      | 2453/6470 [28:14<45:48,  1.46it/s]
3|2454|Loss: 0.8553897142410278:  38%|███▊      | 2454/6470 [28:15<44:15,  1.51it/s]
3|2455|Loss: 0.7382543087005615:  38%|███▊      | 2454/6470 [28:15<44:15,  1.51it/s]
3|2455|Loss: 0.7382543087005615:  38%|███▊      | 2455/6470 [28:15<43:34,  1.54it/s]
3|2456|Loss: 0.9377691745758057:  38%|███▊      | 2455/6470 [28:16<43:34,  1.54it/s]
3|2456|Loss: 0.9377691745758057:  38%|███▊      | 2456/6470 [28:16<43:06,  1.55it/s]
3|2457|Loss: 0.937461793422699:  38%|███▊      | 2456/6470 [28:16<43:06,  1.55it/s] 
3|2457|Loss: 0.937461793422699:  38%|███▊      | 2457/6470 [28:16<41:08,  1.63it/s]
3|2458|Loss: 0.8380179405212402:  38%|███▊      | 2457/6470 [28:17<41:08,  1.63it/s]
3|2458|Loss: 0.8380179405212402:  38%|███▊      | 2458/6470 [28:17<39:06,  1.71it/s]
3|2459|Loss: 0.6797385215759277:  38%|███▊      | 2458/6470 [28:17<39:06,  1.71it/s]
3|2459|Loss: 0.6797385215759277:  38%|███▊      | 2459/6470 [28:18<40:35,  1.65it/s]
3|2460|Loss: 0.7833177447319031:  38%|███▊      | 2459/6470 [28:18<40:35,  1.65it/s]
3|2460|Loss: 0.7833177447319031:  38%|███▊      | 2460/6470 [28:18<44:32,  1.50it/s]
3|2461|Loss: 0.9019538164138794:  38%|███▊      | 2460/6470 [28:19<44:32,  1.50it/s]
3|2461|Loss: 0.9019538164138794:  38%|███▊      | 2461/6470 [28:19<46:32,  1.44it/s]
3|2462|Loss: 0.9967860579490662:  38%|███▊      | 2461/6470 [28:20<46:32,  1.44it/s]
3|2462|Loss: 0.9967860579490662:  38%|███▊      | 2462/6470 [28:20<45:38,  1.46it/s]
3|2463|Loss: 0.9045429229736328:  38%|███▊      | 2462/6470 [28:20<45:38,  1.46it/s]
3|2463|Loss: 0.9045429229736328:  38%|███▊      | 2463/6470 [28:20<45:37,  1.46it/s]
3|2464|Loss: 0.8713859915733337:  38%|███▊      | 2463/6470 [28:21<45:37,  1.46it/s]
3|2464|Loss: 0.8713859915733337:  38%|███▊      | 2464/6470 [28:21<46:04,  1.45it/s]
3|2465|Loss: 0.884451687335968:  38%|███▊      | 2464/6470 [28:22<46:04,  1.45it/s] 
3|2465|Loss: 0.884451687335968:  38%|███▊      | 2465/6470 [28:22<46:55,  1.42it/s]
3|2466|Loss: 0.7597355246543884:  38%|███▊      | 2465/6470 [28:22<46:55,  1.42it/s]
3|2466|Loss: 0.7597355246543884:  38%|███▊      | 2466/6470 [28:23<47:38,  1.40it/s]
3|2467|Loss: 0.9812225103378296:  38%|███▊      | 2466/6470 [28:23<47:38,  1.40it/s]
3|2467|Loss: 0.9812225103378296:  38%|███▊      | 2467/6470 [28:23<48:59,  1.36it/s]
3|2468|Loss: 0.852459192276001:  38%|███▊      | 2467/6470 [28:24<48:59,  1.36it/s] 
3|2468|Loss: 0.852459192276001:  38%|███▊      | 2468/6470 [28:24<46:26,  1.44it/s]
3|2469|Loss: 0.9066556692123413:  38%|███▊      | 2468/6470 [28:24<46:26,  1.44it/s]
3|2469|Loss: 0.9066556692123413:  38%|███▊      | 2469/6470 [28:25<44:11,  1.51it/s]
3|2470|Loss: 0.6066542863845825:  38%|███▊      | 2469/6470 [28:25<44:11,  1.51it/s]
3|2470|Loss: 0.6066542863845825:  38%|███▊      | 2470/6470 [28:26<48:40,  1.37it/s]
3|2471|Loss: 0.8589985370635986:  38%|███▊      | 2470/6470 [28:26<48:40,  1.37it/s]
3|2471|Loss: 0.8589985370635986:  38%|███▊      | 2471/6470 [28:26<49:59,  1.33it/s]
3|2472|Loss: 0.742444634437561:  38%|███▊      | 2471/6470 [28:27<49:59,  1.33it/s] 
3|2472|Loss: 0.742444634437561:  38%|███▊      | 2472/6470 [28:27<48:18,  1.38it/s]
3|2473|Loss: 0.9138646125793457:  38%|███▊      | 2472/6470 [28:27<48:18,  1.38it/s]
3|2473|Loss: 0.9138646125793457:  38%|███▊      | 2473/6470 [28:28<46:06,  1.44it/s]
3|2474|Loss: 0.8384313583374023:  38%|███▊      | 2473/6470 [28:28<46:06,  1.44it/s]
3|2474|Loss: 0.8384313583374023:  38%|███▊      | 2474/6470 [28:28<44:37,  1.49it/s]
3|2475|Loss: 0.8371461033821106:  38%|███▊      | 2474/6470 [28:29<44:37,  1.49it/s]
3|2475|Loss: 0.8371461033821106:  38%|███▊      | 2475/6470 [28:29<44:58,  1.48it/s]
3|2476|Loss: 0.7980656027793884:  38%|███▊      | 2475/6470 [28:29<44:58,  1.48it/s]
3|2476|Loss: 0.7980656027793884:  38%|███▊      | 2476/6470 [28:30<48:58,  1.36it/s]
3|2477|Loss: 0.9684513807296753:  38%|███▊      | 2476/6470 [28:30<48:58,  1.36it/s]
3|2477|Loss: 0.9684513807296753:  38%|███▊      | 2477/6470 [28:31<49:12,  1.35it/s]
3|2478|Loss: 0.870458722114563:  38%|███▊      | 2477/6470 [28:31<49:12,  1.35it/s] 
3|2478|Loss: 0.870458722114563:  38%|███▊      | 2478/6470 [28:31<48:29,  1.37it/s]
3|2479|Loss: 1.0151292085647583:  38%|███▊      | 2478/6470 [28:32<48:29,  1.37it/s]
3|2479|Loss: 1.0151292085647583:  38%|███▊      | 2479/6470 [28:32<46:08,  1.44it/s]
3|2480|Loss: 0.7946184873580933:  38%|███▊      | 2479/6470 [28:32<46:08,  1.44it/s]
3|2480|Loss: 0.7946184873580933:  38%|███▊      | 2480/6470 [28:32<44:53,  1.48it/s]
3|2481|Loss: 0.7912415266036987:  38%|███▊      | 2480/6470 [28:33<44:53,  1.48it/s]
3|2481|Loss: 0.7912415266036987:  38%|███▊      | 2481/6470 [28:33<45:57,  1.45it/s]
3|2482|Loss: 0.8737693428993225:  38%|███▊      | 2481/6470 [28:34<45:57,  1.45it/s]
3|2482|Loss: 0.8737693428993225:  38%|███▊      | 2482/6470 [28:34<46:39,  1.42it/s]
3|2483|Loss: 0.7037853002548218:  38%|███▊      | 2482/6470 [28:34<46:39,  1.42it/s]
3|2483|Loss: 0.7037853002548218:  38%|███▊      | 2483/6470 [28:35<50:44,  1.31it/s]
3|2484|Loss: 0.9640315771102905:  38%|███▊      | 2483/6470 [28:35<50:44,  1.31it/s]
3|2484|Loss: 0.9640315771102905:  38%|███▊      | 2484/6470 [28:36<51:10,  1.30it/s]
3|2485|Loss: 0.9178540706634521:  38%|███▊      | 2484/6470 [28:36<51:10,  1.30it/s]
3|2485|Loss: 0.9178540706634521:  38%|███▊      | 2485/6470 [28:36<50:29,  1.32it/s]
3|2486|Loss: 0.8424047231674194:  38%|███▊      | 2485/6470 [28:37<50:29,  1.32it/s]
3|2486|Loss: 0.8424047231674194:  38%|███▊      | 2486/6470 [28:37<49:25,  1.34it/s]
3|2487|Loss: 0.8301968574523926:  38%|███▊      | 2486/6470 [28:37<49:25,  1.34it/s]
3|2487|Loss: 0.8301968574523926:  38%|███▊      | 2487/6470 [28:38<46:36,  1.42it/s]
3|2488|Loss: 0.7635676264762878:  38%|███▊      | 2487/6470 [28:38<46:36,  1.42it/s]
3|2488|Loss: 0.7635676264762878:  38%|███▊      | 2488/6470 [28:38<44:51,  1.48it/s]
3|2489|Loss: 1.034529447555542:  38%|███▊      | 2488/6470 [28:39<44:51,  1.48it/s] 
3|2489|Loss: 1.034529447555542:  38%|███▊      | 2489/6470 [28:39<43:55,  1.51it/s]
3|2490|Loss: 0.7543081045150757:  38%|███▊      | 2489/6470 [28:39<43:55,  1.51it/s]
3|2490|Loss: 0.7543081045150757:  38%|███▊      | 2490/6470 [28:40<47:23,  1.40it/s]
3|2491|Loss: 0.8380059003829956:  38%|███▊      | 2490/6470 [28:40<47:23,  1.40it/s]
3|2491|Loss: 0.8380059003829956:  39%|███▊      | 2491/6470 [28:40<46:47,  1.42it/s]
3|2492|Loss: 0.9763273596763611:  39%|███▊      | 2491/6470 [28:41<46:47,  1.42it/s]
3|2492|Loss: 0.9763273596763611:  39%|███▊      | 2492/6470 [28:41<45:00,  1.47it/s]
3|2493|Loss: 0.8067101240158081:  39%|███▊      | 2492/6470 [28:41<45:00,  1.47it/s]
3|2493|Loss: 0.8067101240158081:  39%|███▊      | 2493/6470 [28:42<46:44,  1.42it/s]
3|2494|Loss: 0.7917415499687195:  39%|███▊      | 2493/6470 [28:42<46:44,  1.42it/s]
3|2494|Loss: 0.7917415499687195:  39%|███▊      | 2494/6470 [28:43<47:14,  1.40it/s]
3|2495|Loss: 0.883640706539154:  39%|███▊      | 2494/6470 [28:43<47:14,  1.40it/s] 
3|2495|Loss: 0.883640706539154:  39%|███▊      | 2495/6470 [28:43<47:12,  1.40it/s]
3|2496|Loss: 0.8626747727394104:  39%|███▊      | 2495/6470 [28:44<47:12,  1.40it/s]
3|2496|Loss: 0.8626747727394104:  39%|███▊      | 2496/6470 [28:44<47:45,  1.39it/s]
3|2497|Loss: 0.719886064529419:  39%|███▊      | 2496/6470 [28:44<47:45,  1.39it/s] 
3|2497|Loss: 0.719886064529419:  39%|███▊      | 2497/6470 [28:45<45:11,  1.47it/s]
3|2498|Loss: 1.0097668170928955:  39%|███▊      | 2497/6470 [28:45<45:11,  1.47it/s]
3|2498|Loss: 1.0097668170928955:  39%|███▊      | 2498/6470 [28:45<46:58,  1.41it/s]
3|2499|Loss: 0.6727858185768127:  39%|███▊      | 2498/6470 [28:46<46:58,  1.41it/s]
3|2499|Loss: 0.6727858185768127:  39%|███▊      | 2499/6470 [28:46<50:48,  1.30it/s]
3|2500|Loss: 0.7323217988014221:  39%|███▊      | 2499/6470 [28:47<50:48,  1.30it/s]
3|2500|Loss: 0.7323217988014221:  39%|███▊      | 2500/6470 [28:47<47:27,  1.39it/s]
3|2501|Loss: 0.7273627519607544:  39%|███▊      | 2500/6470 [28:47<47:27,  1.39it/s]
3|2501|Loss: 0.7273627519607544:  39%|███▊      | 2501/6470 [28:47<44:58,  1.47it/s]
3|2502|Loss: 0.8091948628425598:  39%|███▊      | 2501/6470 [28:48<44:58,  1.47it/s]
3|2502|Loss: 0.8091948628425598:  39%|███▊      | 2502/6470 [28:48<45:22,  1.46it/s]
3|2503|Loss: 0.8862888216972351:  39%|███▊      | 2502/6470 [28:49<45:22,  1.46it/s]
3|2503|Loss: 0.8862888216972351:  39%|███▊      | 2503/6470 [28:49<44:53,  1.47it/s]
3|2504|Loss: 0.9697219133377075:  39%|███▊      | 2503/6470 [28:49<44:53,  1.47it/s]
3|2504|Loss: 0.9697219133377075:  39%|███▊      | 2504/6470 [28:49<41:39,  1.59it/s]
3|2505|Loss: 0.7864708304405212:  39%|███▊      | 2504/6470 [28:50<41:39,  1.59it/s]
3|2505|Loss: 0.7864708304405212:  39%|███▊      | 2505/6470 [28:50<42:11,  1.57it/s]
3|2506|Loss: 0.9075140953063965:  39%|███▊      | 2505/6470 [28:50<42:11,  1.57it/s]
3|2506|Loss: 0.9075140953063965:  39%|███▊      | 2506/6470 [28:51<42:35,  1.55it/s]
3|2507|Loss: 0.7363503575325012:  39%|███▊      | 2506/6470 [28:51<42:35,  1.55it/s]
3|2507|Loss: 0.7363503575325012:  39%|███▊      | 2507/6470 [28:51<45:03,  1.47it/s]
3|2508|Loss: 0.750850260257721:  39%|███▊      | 2507/6470 [28:52<45:03,  1.47it/s] 
3|2508|Loss: 0.750850260257721:  39%|███▉      | 2508/6470 [28:52<47:52,  1.38it/s]
3|2509|Loss: 0.82771235704422:  39%|███▉      | 2508/6470 [28:53<47:52,  1.38it/s] 
3|2509|Loss: 0.82771235704422:  39%|███▉      | 2509/6470 [28:53<44:01,  1.50it/s]
3|2510|Loss: 0.9251165986061096:  39%|███▉      | 2509/6470 [28:53<44:01,  1.50it/s]
3|2510|Loss: 0.9251165986061096:  39%|███▉      | 2510/6470 [28:53<41:39,  1.58it/s]
3|2511|Loss: 0.942138671875:  39%|███▉      | 2510/6470 [28:54<41:39,  1.58it/s]    
3|2511|Loss: 0.942138671875:  39%|███▉      | 2511/6470 [28:54<40:33,  1.63it/s]
3|2512|Loss: 1.0210801362991333:  39%|███▉      | 2511/6470 [28:54<40:33,  1.63it/s]
3|2512|Loss: 1.0210801362991333:  39%|███▉      | 2512/6470 [28:55<43:26,  1.52it/s]
3|2513|Loss: 0.8724129796028137:  39%|███▉      | 2512/6470 [28:55<43:26,  1.52it/s]
3|2513|Loss: 0.8724129796028137:  39%|███▉      | 2513/6470 [28:55<44:44,  1.47it/s]
3|2514|Loss: 0.8639413714408875:  39%|███▉      | 2513/6470 [28:56<44:44,  1.47it/s]
3|2514|Loss: 0.8639413714408875:  39%|███▉      | 2514/6470 [28:56<45:25,  1.45it/s]
3|2515|Loss: 0.9165546298027039:  39%|███▉      | 2514/6470 [28:57<45:25,  1.45it/s]
3|2515|Loss: 0.9165546298027039:  39%|███▉      | 2515/6470 [28:57<46:13,  1.43it/s]
3|2516|Loss: 0.9200525879859924:  39%|███▉      | 2515/6470 [28:57<46:13,  1.43it/s]
3|2516|Loss: 0.9200525879859924:  39%|███▉      | 2516/6470 [28:58<49:45,  1.32it/s]
3|2517|Loss: 1.0513955354690552:  39%|███▉      | 2516/6470 [28:58<49:45,  1.32it/s]
3|2517|Loss: 1.0513955354690552:  39%|███▉      | 2517/6470 [28:59<52:17,  1.26it/s]
3|2518|Loss: 0.8970619440078735:  39%|███▉      | 2517/6470 [28:59<52:17,  1.26it/s]
3|2518|Loss: 0.8970619440078735:  39%|███▉      | 2518/6470 [28:59<53:37,  1.23it/s]
3|2519|Loss: 0.9969909191131592:  39%|███▉      | 2518/6470 [29:00<53:37,  1.23it/s]
3|2519|Loss: 0.9969909191131592:  39%|███▉      | 2519/6470 [29:00<51:12,  1.29it/s]
3|2520|Loss: 0.8489386439323425:  39%|███▉      | 2519/6470 [29:01<51:12,  1.29it/s]
3|2520|Loss: 0.8489386439323425:  39%|███▉      | 2520/6470 [29:01<46:44,  1.41it/s]
3|2521|Loss: 0.9581733345985413:  39%|███▉      | 2520/6470 [29:01<46:44,  1.41it/s]
3|2521|Loss: 0.9581733345985413:  39%|███▉      | 2521/6470 [29:01<47:14,  1.39it/s]
3|2522|Loss: 0.8437657952308655:  39%|███▉      | 2521/6470 [29:02<47:14,  1.39it/s]
3|2522|Loss: 0.8437657952308655:  39%|███▉      | 2522/6470 [29:02<48:44,  1.35it/s]
3|2523|Loss: 0.806212306022644:  39%|███▉      | 2522/6470 [29:03<48:44,  1.35it/s] 
3|2523|Loss: 0.806212306022644:  39%|███▉      | 2523/6470 [29:03<50:25,  1.30it/s]
3|2524|Loss: 0.9914875030517578:  39%|███▉      | 2523/6470 [29:03<50:25,  1.30it/s]
3|2524|Loss: 0.9914875030517578:  39%|███▉      | 2524/6470 [29:04<48:48,  1.35it/s]
3|2525|Loss: 0.8410829901695251:  39%|███▉      | 2524/6470 [29:04<48:48,  1.35it/s]
3|2525|Loss: 0.8410829901695251:  39%|███▉      | 2525/6470 [29:04<46:23,  1.42it/s]
3|2526|Loss: 0.8949798941612244:  39%|███▉      | 2525/6470 [29:05<46:23,  1.42it/s]
3|2526|Loss: 0.8949798941612244:  39%|███▉      | 2526/6470 [29:05<47:19,  1.39it/s]
3|2527|Loss: 1.0561259984970093:  39%|███▉      | 2526/6470 [29:06<47:19,  1.39it/s]
3|2527|Loss: 1.0561259984970093:  39%|███▉      | 2527/6470 [29:06<46:17,  1.42it/s]
3|2528|Loss: 0.8318483829498291:  39%|███▉      | 2527/6470 [29:06<46:17,  1.42it/s]
3|2528|Loss: 0.8318483829498291:  39%|███▉      | 2528/6470 [29:06<43:33,  1.51it/s]
3|2529|Loss: 0.821804404258728:  39%|███▉      | 2528/6470 [29:07<43:33,  1.51it/s] 
3|2529|Loss: 0.821804404258728:  39%|███▉      | 2529/6470 [29:07<46:28,  1.41it/s]
3|2530|Loss: 0.8223586678504944:  39%|███▉      | 2529/6470 [29:08<46:28,  1.41it/s]
3|2530|Loss: 0.8223586678504944:  39%|███▉      | 2530/6470 [29:08<48:03,  1.37it/s]
3|2531|Loss: 0.7374219298362732:  39%|███▉      | 2530/6470 [29:08<48:03,  1.37it/s]
3|2531|Loss: 0.7374219298362732:  39%|███▉      | 2531/6470 [29:09<47:53,  1.37it/s]
3|2532|Loss: 0.8489186763763428:  39%|███▉      | 2531/6470 [29:09<47:53,  1.37it/s]
3|2532|Loss: 0.8489186763763428:  39%|███▉      | 2532/6470 [29:09<44:17,  1.48it/s]
3|2533|Loss: 0.8710270524024963:  39%|███▉      | 2532/6470 [29:10<44:17,  1.48it/s]
3|2533|Loss: 0.8710270524024963:  39%|███▉      | 2533/6470 [29:10<46:11,  1.42it/s]
3|2534|Loss: 0.8874228000640869:  39%|███▉      | 2533/6470 [29:10<46:11,  1.42it/s]
3|2534|Loss: 0.8874228000640869:  39%|███▉      | 2534/6470 [29:11<43:40,  1.50it/s]
3|2535|Loss: 0.8370703458786011:  39%|███▉      | 2534/6470 [29:11<43:40,  1.50it/s]
3|2535|Loss: 0.8370703458786011:  39%|███▉      | 2535/6470 [29:11<39:45,  1.65it/s]
3|2536|Loss: 0.8524993658065796:  39%|███▉      | 2535/6470 [29:11<39:45,  1.65it/s]
3|2536|Loss: 0.8524993658065796:  39%|███▉      | 2536/6470 [29:12<39:18,  1.67it/s]
3|2537|Loss: 0.8167747855186462:  39%|███▉      | 2536/6470 [29:12<39:18,  1.67it/s]
3|2537|Loss: 0.8167747855186462:  39%|███▉      | 2537/6470 [29:12<41:37,  1.57it/s]
3|2538|Loss: 1.0485508441925049:  39%|███▉      | 2537/6470 [29:13<41:37,  1.57it/s]
3|2538|Loss: 1.0485508441925049:  39%|███▉      | 2538/6470 [29:13<43:14,  1.52it/s]
3|2539|Loss: 0.7445672154426575:  39%|███▉      | 2538/6470 [29:14<43:14,  1.52it/s]
3|2539|Loss: 0.7445672154426575:  39%|███▉      | 2539/6470 [29:14<45:26,  1.44it/s]
3|2540|Loss: 0.8414013981819153:  39%|███▉      | 2539/6470 [29:14<45:26,  1.44it/s]
3|2540|Loss: 0.8414013981819153:  39%|███▉      | 2540/6470 [29:15<46:34,  1.41it/s]
3|2541|Loss: 0.8269542455673218:  39%|███▉      | 2540/6470 [29:15<46:34,  1.41it/s]
3|2541|Loss: 0.8269542455673218:  39%|███▉      | 2541/6470 [29:15<45:41,  1.43it/s]
3|2542|Loss: 0.8328383564949036:  39%|███▉      | 2541/6470 [29:16<45:41,  1.43it/s]
3|2542|Loss: 0.8328383564949036:  39%|███▉      | 2542/6470 [29:16<46:07,  1.42it/s]
3|2543|Loss: 0.8137867450714111:  39%|███▉      | 2542/6470 [29:16<46:07,  1.42it/s]
3|2543|Loss: 0.8137867450714111:  39%|███▉      | 2543/6470 [29:17<47:41,  1.37it/s]
3|2544|Loss: 1.0380431413650513:  39%|███▉      | 2543/6470 [29:17<47:41,  1.37it/s]
3|2544|Loss: 1.0380431413650513:  39%|███▉      | 2544/6470 [29:17<46:19,  1.41it/s]
3|2545|Loss: 0.838681697845459:  39%|███▉      | 2544/6470 [29:18<46:19,  1.41it/s] 
3|2545|Loss: 0.838681697845459:  39%|███▉      | 2545/6470 [29:18<47:51,  1.37it/s]
3|2546|Loss: 0.8219636678695679:  39%|███▉      | 2545/6470 [29:19<47:51,  1.37it/s]
3|2546|Loss: 0.8219636678695679:  39%|███▉      | 2546/6470 [29:19<47:48,  1.37it/s]
3|2547|Loss: 0.7040058374404907:  39%|███▉      | 2546/6470 [29:19<47:48,  1.37it/s]
3|2547|Loss: 0.7040058374404907:  39%|███▉      | 2547/6470 [29:20<44:24,  1.47it/s]
3|2548|Loss: 0.9739668965339661:  39%|███▉      | 2547/6470 [29:20<44:24,  1.47it/s]
3|2548|Loss: 0.9739668965339661:  39%|███▉      | 2548/6470 [29:20<44:02,  1.48it/s]
3|2549|Loss: 0.7288216948509216:  39%|███▉      | 2548/6470 [29:21<44:02,  1.48it/s]
3|2549|Loss: 0.7288216948509216:  39%|███▉      | 2549/6470 [29:21<41:53,  1.56it/s]
3|2550|Loss: 0.9038017392158508:  39%|███▉      | 2549/6470 [29:21<41:53,  1.56it/s]
3|2550|Loss: 0.9038017392158508:  39%|███▉      | 2550/6470 [29:21<41:22,  1.58it/s]
3|2551|Loss: 0.899475634098053:  39%|███▉      | 2550/6470 [29:22<41:22,  1.58it/s] 
3|2551|Loss: 0.899475634098053:  39%|███▉      | 2551/6470 [29:22<40:26,  1.62it/s]
3|2552|Loss: 0.7034612894058228:  39%|███▉      | 2551/6470 [29:22<40:26,  1.62it/s]
3|2552|Loss: 0.7034612894058228:  39%|███▉      | 2552/6470 [29:23<43:07,  1.51it/s]
3|2553|Loss: 0.9690107107162476:  39%|███▉      | 2552/6470 [29:23<43:07,  1.51it/s]
3|2553|Loss: 0.9690107107162476:  39%|███▉      | 2553/6470 [29:23<41:57,  1.56it/s]
3|2554|Loss: 0.8493533730506897:  39%|███▉      | 2553/6470 [29:24<41:57,  1.56it/s]
3|2554|Loss: 0.8493533730506897:  39%|███▉      | 2554/6470 [29:24<42:57,  1.52it/s]
3|2555|Loss: 0.9320111870765686:  39%|███▉      | 2554/6470 [29:24<42:57,  1.52it/s]
3|2555|Loss: 0.9320111870765686:  39%|███▉      | 2555/6470 [29:25<40:59,  1.59it/s]
3|2556|Loss: 1.0200130939483643:  39%|███▉      | 2555/6470 [29:25<40:59,  1.59it/s]
3|2556|Loss: 1.0200130939483643:  40%|███▉      | 2556/6470 [29:25<38:23,  1.70it/s]
3|2557|Loss: 0.9248717427253723:  40%|███▉      | 2556/6470 [29:25<38:23,  1.70it/s]
3|2557|Loss: 0.9248717427253723:  40%|███▉      | 2557/6470 [29:26<41:39,  1.57it/s]
3|2558|Loss: 0.923391580581665:  40%|███▉      | 2557/6470 [29:26<41:39,  1.57it/s] 
3|2558|Loss: 0.923391580581665:  40%|███▉      | 2558/6470 [29:27<43:26,  1.50it/s]
3|2559|Loss: 0.8766050338745117:  40%|███▉      | 2558/6470 [29:27<43:26,  1.50it/s]
3|2559|Loss: 0.8766050338745117:  40%|███▉      | 2559/6470 [29:27<40:38,  1.60it/s]
3|2560|Loss: 0.8506746888160706:  40%|███▉      | 2559/6470 [29:27<40:38,  1.60it/s]
3|2560|Loss: 0.8506746888160706:  40%|███▉      | 2560/6470 [29:28<41:39,  1.56it/s]
3|2561|Loss: 0.7730175256729126:  40%|███▉      | 2560/6470 [29:28<41:39,  1.56it/s]
3|2561|Loss: 0.7730175256729126:  40%|███▉      | 2561/6470 [29:28<40:14,  1.62it/s]
3|2562|Loss: 0.749250054359436:  40%|███▉      | 2561/6470 [29:29<40:14,  1.62it/s] 
3|2562|Loss: 0.749250054359436:  40%|███▉      | 2562/6470 [29:29<40:15,  1.62it/s]
3|2563|Loss: 0.7549049854278564:  40%|███▉      | 2562/6470 [29:29<40:15,  1.62it/s]
3|2563|Loss: 0.7549049854278564:  40%|███▉      | 2563/6470 [29:30<41:47,  1.56it/s]
3|2564|Loss: 1.008862018585205:  40%|███▉      | 2563/6470 [29:30<41:47,  1.56it/s] 
3|2564|Loss: 1.008862018585205:  40%|███▉      | 2564/6470 [29:30<40:35,  1.60it/s]
3|2565|Loss: 0.930673360824585:  40%|███▉      | 2564/6470 [29:31<40:35,  1.60it/s]
3|2565|Loss: 0.930673360824585:  40%|███▉      | 2565/6470 [29:31<39:37,  1.64it/s]
3|2566|Loss: 0.9189964532852173:  40%|███▉      | 2565/6470 [29:31<39:37,  1.64it/s]
3|2566|Loss: 0.9189964532852173:  40%|███▉      | 2566/6470 [29:32<42:35,  1.53it/s]
3|2567|Loss: 0.8697157502174377:  40%|███▉      | 2566/6470 [29:32<42:35,  1.53it/s]
3|2567|Loss: 0.8697157502174377:  40%|███▉      | 2567/6470 [29:32<42:48,  1.52it/s]
3|2568|Loss: 0.7396968007087708:  40%|███▉      | 2567/6470 [29:33<42:48,  1.52it/s]
3|2568|Loss: 0.7396968007087708:  40%|███▉      | 2568/6470 [29:33<42:21,  1.54it/s]
3|2569|Loss: 0.7617154121398926:  40%|███▉      | 2568/6470 [29:33<42:21,  1.54it/s]
3|2569|Loss: 0.7617154121398926:  40%|███▉      | 2569/6470 [29:34<45:55,  1.42it/s]
3|2570|Loss: 0.891665518283844:  40%|███▉      | 2569/6470 [29:34<45:55,  1.42it/s] 
3|2570|Loss: 0.891665518283844:  40%|███▉      | 2570/6470 [29:34<47:02,  1.38it/s]
3|2571|Loss: 0.9652625322341919:  40%|███▉      | 2570/6470 [29:35<47:02,  1.38it/s]
3|2571|Loss: 0.9652625322341919:  40%|███▉      | 2571/6470 [29:35<47:14,  1.38it/s]
3|2572|Loss: 0.9023657441139221:  40%|███▉      | 2571/6470 [29:36<47:14,  1.38it/s]
3|2572|Loss: 0.9023657441139221:  40%|███▉      | 2572/6470 [29:36<46:52,  1.39it/s]
3|2573|Loss: 0.7042660117149353:  40%|███▉      | 2572/6470 [29:36<46:52,  1.39it/s]
3|2573|Loss: 0.7042660117149353:  40%|███▉      | 2573/6470 [29:37<45:33,  1.43it/s]
3|2574|Loss: 1.0758131742477417:  40%|███▉      | 2573/6470 [29:37<45:33,  1.43it/s]
3|2574|Loss: 1.0758131742477417:  40%|███▉      | 2574/6470 [29:37<45:58,  1.41it/s]
3|2575|Loss: 0.7993313074111938:  40%|███▉      | 2574/6470 [29:38<45:58,  1.41it/s]
3|2575|Loss: 0.7993313074111938:  40%|███▉      | 2575/6470 [29:38<46:06,  1.41it/s]
3|2576|Loss: 0.7633360028266907:  40%|███▉      | 2575/6470 [29:38<46:06,  1.41it/s]
3|2576|Loss: 0.7633360028266907:  40%|███▉      | 2576/6470 [29:39<44:40,  1.45it/s]
3|2577|Loss: 0.7706626653671265:  40%|███▉      | 2576/6470 [29:39<44:40,  1.45it/s]
3|2577|Loss: 0.7706626653671265:  40%|███▉      | 2577/6470 [29:39<42:52,  1.51it/s]
3|2578|Loss: 0.8256736993789673:  40%|███▉      | 2577/6470 [29:40<42:52,  1.51it/s]
3|2578|Loss: 0.8256736993789673:  40%|███▉      | 2578/6470 [29:40<42:14,  1.54it/s]
3|2579|Loss: 0.7966297268867493:  40%|███▉      | 2578/6470 [29:40<42:14,  1.54it/s]
3|2579|Loss: 0.7966297268867493:  40%|███▉      | 2579/6470 [29:41<44:56,  1.44it/s]
3|2580|Loss: 0.9279767870903015:  40%|███▉      | 2579/6470 [29:41<44:56,  1.44it/s]
3|2580|Loss: 0.9279767870903015:  40%|███▉      | 2580/6470 [29:41<44:17,  1.46it/s]
3|2581|Loss: 0.9040238261222839:  40%|███▉      | 2580/6470 [29:42<44:17,  1.46it/s]
3|2581|Loss: 0.9040238261222839:  40%|███▉      | 2581/6470 [29:42<46:44,  1.39it/s]
3|2582|Loss: 0.9201149940490723:  40%|███▉      | 2581/6470 [29:43<46:44,  1.39it/s]
3|2582|Loss: 0.9201149940490723:  40%|███▉      | 2582/6470 [29:43<45:53,  1.41it/s]
3|2583|Loss: 0.8732688426971436:  40%|███▉      | 2582/6470 [29:43<45:53,  1.41it/s]
3|2583|Loss: 0.8732688426971436:  40%|███▉      | 2583/6470 [29:44<47:21,  1.37it/s]
3|2584|Loss: 0.8328428864479065:  40%|███▉      | 2583/6470 [29:44<47:21,  1.37it/s]
3|2584|Loss: 0.8328428864479065:  40%|███▉      | 2584/6470 [29:44<46:24,  1.40it/s]
3|2585|Loss: 0.8746702075004578:  40%|███▉      | 2584/6470 [29:45<46:24,  1.40it/s]
3|2585|Loss: 0.8746702075004578:  40%|███▉      | 2585/6470 [29:45<45:01,  1.44it/s]
3|2586|Loss: 0.9520906209945679:  40%|███▉      | 2585/6470 [29:45<45:01,  1.44it/s]
3|2586|Loss: 0.9520906209945679:  40%|███▉      | 2586/6470 [29:45<42:02,  1.54it/s]
3|2587|Loss: 0.9242023229598999:  40%|███▉      | 2586/6470 [29:46<42:02,  1.54it/s]
3|2587|Loss: 0.9242023229598999:  40%|███▉      | 2587/6470 [29:46<41:21,  1.56it/s]
3|2588|Loss: 0.8528938889503479:  40%|███▉      | 2587/6470 [29:46<41:21,  1.56it/s]
3|2588|Loss: 0.8528938889503479:  40%|████      | 2588/6470 [29:47<41:30,  1.56it/s]
3|2589|Loss: 0.8460003137588501:  40%|████      | 2588/6470 [29:47<41:30,  1.56it/s]
3|2589|Loss: 0.8460003137588501:  40%|████      | 2589/6470 [29:47<39:04,  1.66it/s]
3|2590|Loss: 1.0083723068237305:  40%|████      | 2589/6470 [29:48<39:04,  1.66it/s]
3|2590|Loss: 1.0083723068237305:  40%|████      | 2590/6470 [29:48<37:20,  1.73it/s]
3|2591|Loss: 0.8729583024978638:  40%|████      | 2590/6470 [29:48<37:20,  1.73it/s]
3|2591|Loss: 0.8729583024978638:  40%|████      | 2591/6470 [29:48<36:20,  1.78it/s]
3|2592|Loss: 0.8207709193229675:  40%|████      | 2591/6470 [29:49<36:20,  1.78it/s]
3|2592|Loss: 0.8207709193229675:  40%|████      | 2592/6470 [29:49<38:27,  1.68it/s]
3|2593|Loss: 0.7633697986602783:  40%|████      | 2592/6470 [29:49<38:27,  1.68it/s]
3|2593|Loss: 0.7633697986602783:  40%|████      | 2593/6470 [29:50<38:50,  1.66it/s]
3|2594|Loss: 0.8067899346351624:  40%|████      | 2593/6470 [29:50<38:50,  1.66it/s]
3|2594|Loss: 0.8067899346351624:  40%|████      | 2594/6470 [29:50<38:53,  1.66it/s]
3|2595|Loss: 0.9787408113479614:  40%|████      | 2594/6470 [29:51<38:53,  1.66it/s]
3|2595|Loss: 0.9787408113479614:  40%|████      | 2595/6470 [29:51<40:29,  1.60it/s]
3|2596|Loss: 0.7115792632102966:  40%|████      | 2595/6470 [29:51<40:29,  1.60it/s]
3|2596|Loss: 0.7115792632102966:  40%|████      | 2596/6470 [29:52<42:20,  1.53it/s]
3|2597|Loss: 0.8545560240745544:  40%|████      | 2596/6470 [29:52<42:20,  1.53it/s]
3|2597|Loss: 0.8545560240745544:  40%|████      | 2597/6470 [29:52<41:22,  1.56it/s]
3|2598|Loss: 0.7946150302886963:  40%|████      | 2597/6470 [29:53<41:22,  1.56it/s]
3|2598|Loss: 0.7946150302886963:  40%|████      | 2598/6470 [29:53<40:49,  1.58it/s]
3|2599|Loss: 0.749417781829834:  40%|████      | 2598/6470 [29:53<40:49,  1.58it/s] 
3|2599|Loss: 0.749417781829834:  40%|████      | 2599/6470 [29:53<40:42,  1.59it/s]
3|2600|Loss: 1.0863186120986938:  40%|████      | 2599/6470 [29:54<40:42,  1.59it/s]
3|2600|Loss: 1.0863186120986938:  40%|████      | 2600/6470 [29:54<40:49,  1.58it/s]
3|2601|Loss: 0.8158025145530701:  40%|████      | 2600/6470 [29:54<40:49,  1.58it/s]
3|2601|Loss: 0.8158025145530701:  40%|████      | 2601/6470 [29:55<41:56,  1.54it/s]
3|2602|Loss: 0.8058815002441406:  40%|████      | 2601/6470 [29:55<41:56,  1.54it/s]
3|2602|Loss: 0.8058815002441406:  40%|████      | 2602/6470 [29:55<38:00,  1.70it/s]
3|2603|Loss: 0.8539988398551941:  40%|████      | 2602/6470 [29:56<38:00,  1.70it/s]
3|2603|Loss: 0.8539988398551941:  40%|████      | 2603/6470 [29:56<39:48,  1.62it/s]
3|2604|Loss: 0.970964252948761:  40%|████      | 2603/6470 [29:56<39:48,  1.62it/s] 
3|2604|Loss: 0.970964252948761:  40%|████      | 2604/6470 [29:57<43:04,  1.50it/s]
3|2605|Loss: 0.890764594078064:  40%|████      | 2604/6470 [29:57<43:04,  1.50it/s]
3|2605|Loss: 0.890764594078064:  40%|████      | 2605/6470 [29:57<41:04,  1.57it/s]
3|2606|Loss: 0.8996583223342896:  40%|████      | 2605/6470 [29:58<41:04,  1.57it/s]
3|2606|Loss: 0.8996583223342896:  40%|████      | 2606/6470 [29:58<43:31,  1.48it/s]
3|2607|Loss: 0.771695613861084:  40%|████      | 2606/6470 [29:58<43:31,  1.48it/s] 
3|2607|Loss: 0.771695613861084:  40%|████      | 2607/6470 [29:59<43:10,  1.49it/s]
3|2608|Loss: 0.750830888748169:  40%|████      | 2607/6470 [29:59<43:10,  1.49it/s]
3|2608|Loss: 0.750830888748169:  40%|████      | 2608/6470 [29:59<42:23,  1.52it/s]
3|2609|Loss: 0.9711296558380127:  40%|████      | 2608/6470 [30:00<42:23,  1.52it/s]
3|2609|Loss: 0.9711296558380127:  40%|████      | 2609/6470 [30:00<42:05,  1.53it/s]
3|2610|Loss: 0.6861515045166016:  40%|████      | 2609/6470 [30:00<42:05,  1.53it/s]
3|2610|Loss: 0.6861515045166016:  40%|████      | 2610/6470 [30:01<46:57,  1.37it/s]
3|2611|Loss: 0.8542752861976624:  40%|████      | 2610/6470 [30:01<46:57,  1.37it/s]
3|2611|Loss: 0.8542752861976624:  40%|████      | 2611/6470 [30:02<47:58,  1.34it/s]
3|2612|Loss: 0.826319694519043:  40%|████      | 2611/6470 [30:02<47:58,  1.34it/s] 
3|2612|Loss: 0.826319694519043:  40%|████      | 2612/6470 [30:02<45:24,  1.42it/s]
3|2613|Loss: 0.8490990996360779:  40%|████      | 2612/6470 [30:03<45:24,  1.42it/s]
3|2613|Loss: 0.8490990996360779:  40%|████      | 2613/6470 [30:03<47:12,  1.36it/s]
3|2614|Loss: 0.7315184473991394:  40%|████      | 2613/6470 [30:03<47:12,  1.36it/s]
3|2614|Loss: 0.7315184473991394:  40%|████      | 2614/6470 [30:04<47:52,  1.34it/s]
3|2615|Loss: 0.8856494426727295:  40%|████      | 2614/6470 [30:04<47:52,  1.34it/s]
3|2615|Loss: 0.8856494426727295:  40%|████      | 2615/6470 [30:04<47:14,  1.36it/s]
3|2616|Loss: 0.9418409466743469:  40%|████      | 2615/6470 [30:05<47:14,  1.36it/s]
3|2616|Loss: 0.9418409466743469:  40%|████      | 2616/6470 [30:05<47:22,  1.36it/s]
3|2617|Loss: 0.8040088415145874:  40%|████      | 2616/6470 [30:06<47:22,  1.36it/s]
3|2617|Loss: 0.8040088415145874:  40%|████      | 2617/6470 [30:06<47:25,  1.35it/s]
3|2618|Loss: 0.7340238690376282:  40%|████      | 2617/6470 [30:06<47:25,  1.35it/s]
3|2618|Loss: 0.7340238690376282:  40%|████      | 2618/6470 [30:07<44:06,  1.46it/s]
3|2619|Loss: 0.8273664712905884:  40%|████      | 2618/6470 [30:07<44:06,  1.46it/s]
3|2619|Loss: 0.8273664712905884:  40%|████      | 2619/6470 [30:07<42:24,  1.51it/s]
3|2620|Loss: 0.9947684407234192:  40%|████      | 2619/6470 [30:08<42:24,  1.51it/s]
3|2620|Loss: 0.9947684407234192:  40%|████      | 2620/6470 [30:08<43:23,  1.48it/s]
3|2621|Loss: 0.7668178081512451:  40%|████      | 2620/6470 [30:08<43:23,  1.48it/s]
3|2621|Loss: 0.7668178081512451:  41%|████      | 2621/6470 [30:09<42:58,  1.49it/s]
3|2622|Loss: 0.9112323522567749:  41%|████      | 2621/6470 [30:09<42:58,  1.49it/s]
3|2622|Loss: 0.9112323522567749:  41%|████      | 2622/6470 [30:09<43:42,  1.47it/s]
3|2623|Loss: 1.0898056030273438:  41%|████      | 2622/6470 [30:10<43:42,  1.47it/s]
3|2623|Loss: 1.0898056030273438:  41%|████      | 2623/6470 [30:10<44:11,  1.45it/s]
3|2624|Loss: 0.7803722620010376:  41%|████      | 2623/6470 [30:10<44:11,  1.45it/s]
3|2624|Loss: 0.7803722620010376:  41%|████      | 2624/6470 [30:11<48:55,  1.31it/s]
3|2625|Loss: 1.045487880706787:  41%|████      | 2624/6470 [30:11<48:55,  1.31it/s] 
3|2625|Loss: 1.045487880706787:  41%|████      | 2625/6470 [30:12<47:42,  1.34it/s]
3|2626|Loss: 0.8621553182601929:  41%|████      | 2625/6470 [30:12<47:42,  1.34it/s]
3|2626|Loss: 0.8621553182601929:  41%|████      | 2626/6470 [30:12<45:32,  1.41it/s]
3|2627|Loss: 0.994811475276947:  41%|████      | 2626/6470 [30:13<45:32,  1.41it/s] 
3|2627|Loss: 0.994811475276947:  41%|████      | 2627/6470 [30:13<46:26,  1.38it/s]
3|2628|Loss: 0.8579745292663574:  41%|████      | 2627/6470 [30:13<46:26,  1.38it/s]
3|2628|Loss: 0.8579745292663574:  41%|████      | 2628/6470 [30:14<47:53,  1.34it/s]
3|2629|Loss: 0.9617449641227722:  41%|████      | 2628/6470 [30:14<47:53,  1.34it/s]
3|2629|Loss: 0.9617449641227722:  41%|████      | 2629/6470 [30:14<46:16,  1.38it/s]
3|2630|Loss: 0.7875350713729858:  41%|████      | 2629/6470 [30:15<46:16,  1.38it/s]
3|2630|Loss: 0.7875350713729858:  41%|████      | 2630/6470 [30:15<46:09,  1.39it/s]
3|2631|Loss: 0.8413733243942261:  41%|████      | 2630/6470 [30:16<46:09,  1.39it/s]
3|2631|Loss: 0.8413733243942261:  41%|████      | 2631/6470 [30:16<44:18,  1.44it/s]
3|2632|Loss: 0.7743958234786987:  41%|████      | 2631/6470 [30:16<44:18,  1.44it/s]
3|2632|Loss: 0.7743958234786987:  41%|████      | 2632/6470 [30:17<46:57,  1.36it/s]
3|2633|Loss: 0.821492612361908:  41%|████      | 2632/6470 [30:17<46:57,  1.36it/s] 
3|2633|Loss: 0.821492612361908:  41%|████      | 2633/6470 [30:17<48:46,  1.31it/s]
3|2634|Loss: 0.7467151284217834:  41%|████      | 2633/6470 [30:18<48:46,  1.31it/s]
3|2634|Loss: 0.7467151284217834:  41%|████      | 2634/6470 [30:18<50:30,  1.27it/s]
3|2635|Loss: 0.8277692198753357:  41%|████      | 2634/6470 [30:19<50:30,  1.27it/s]
3|2635|Loss: 0.8277692198753357:  41%|████      | 2635/6470 [30:19<48:44,  1.31it/s]
3|2636|Loss: 0.8532527089118958:  41%|████      | 2635/6470 [30:19<48:44,  1.31it/s]
3|2636|Loss: 0.8532527089118958:  41%|████      | 2636/6470 [30:20<49:53,  1.28it/s]
3|2637|Loss: 0.8415182828903198:  41%|████      | 2636/6470 [30:20<49:53,  1.28it/s]
3|2637|Loss: 0.8415182828903198:  41%|████      | 2637/6470 [30:20<47:09,  1.35it/s]
3|2638|Loss: 0.8173930644989014:  41%|████      | 2637/6470 [30:21<47:09,  1.35it/s]
3|2638|Loss: 0.8173930644989014:  41%|████      | 2638/6470 [30:21<44:41,  1.43it/s]
3|2639|Loss: 0.9200224280357361:  41%|████      | 2638/6470 [30:21<44:41,  1.43it/s]
3|2639|Loss: 0.9200224280357361:  41%|████      | 2639/6470 [30:22<44:57,  1.42it/s]
3|2640|Loss: 0.902179479598999:  41%|████      | 2639/6470 [30:22<44:57,  1.42it/s] 
3|2640|Loss: 0.902179479598999:  41%|████      | 2640/6470 [30:23<46:41,  1.37it/s]
3|2641|Loss: 0.8668138384819031:  41%|████      | 2640/6470 [30:23<46:41,  1.37it/s]
3|2641|Loss: 0.8668138384819031:  41%|████      | 2641/6470 [30:23<45:36,  1.40it/s]
3|2642|Loss: 0.7635140419006348:  41%|████      | 2641/6470 [30:24<45:36,  1.40it/s]
3|2642|Loss: 0.7635140419006348:  41%|████      | 2642/6470 [30:24<41:15,  1.55it/s]
3|2643|Loss: 0.8515521287918091:  41%|████      | 2642/6470 [30:24<41:15,  1.55it/s]
3|2643|Loss: 0.8515521287918091:  41%|████      | 2643/6470 [30:24<41:54,  1.52it/s]
3|2644|Loss: 0.8507218360900879:  41%|████      | 2643/6470 [30:25<41:54,  1.52it/s]
3|2644|Loss: 0.8507218360900879:  41%|████      | 2644/6470 [30:25<45:26,  1.40it/s]
3|2645|Loss: 0.8112156391143799:  41%|████      | 2644/6470 [30:26<45:26,  1.40it/s]
3|2645|Loss: 0.8112156391143799:  41%|████      | 2645/6470 [30:26<49:19,  1.29it/s]
3|2646|Loss: 0.7784814238548279:  41%|████      | 2645/6470 [30:27<49:19,  1.29it/s]
3|2646|Loss: 0.7784814238548279:  41%|████      | 2646/6470 [30:27<51:24,  1.24it/s]
3|2647|Loss: 0.8914067149162292:  41%|████      | 2646/6470 [30:27<51:24,  1.24it/s]
3|2647|Loss: 0.8914067149162292:  41%|████      | 2647/6470 [30:28<48:27,  1.32it/s]
3|2648|Loss: 0.7188497185707092:  41%|████      | 2647/6470 [30:28<48:27,  1.32it/s]
3|2648|Loss: 0.7188497185707092:  41%|████      | 2648/6470 [30:28<48:06,  1.32it/s]
3|2649|Loss: 0.8102075457572937:  41%|████      | 2648/6470 [30:29<48:06,  1.32it/s]
3|2649|Loss: 0.8102075457572937:  41%|████      | 2649/6470 [30:29<45:10,  1.41it/s]
3|2650|Loss: 0.8657256960868835:  41%|████      | 2649/6470 [30:29<45:10,  1.41it/s]
3|2650|Loss: 0.8657256960868835:  41%|████      | 2650/6470 [30:30<47:20,  1.34it/s]
3|2651|Loss: 0.822037935256958:  41%|████      | 2650/6470 [30:30<47:20,  1.34it/s] 
3|2651|Loss: 0.822037935256958:  41%|████      | 2651/6470 [30:31<48:00,  1.33it/s]
3|2652|Loss: 0.8395144939422607:  41%|████      | 2651/6470 [30:31<48:00,  1.33it/s]
3|2652|Loss: 0.8395144939422607:  41%|████      | 2652/6470 [30:31<47:17,  1.35it/s]
3|2653|Loss: 0.8603755235671997:  41%|████      | 2652/6470 [30:32<47:17,  1.35it/s]
3|2653|Loss: 0.8603755235671997:  41%|████      | 2653/6470 [30:32<44:08,  1.44it/s]
3|2654|Loss: 0.7845390439033508:  41%|████      | 2653/6470 [30:32<44:08,  1.44it/s]
3|2654|Loss: 0.7845390439033508:  41%|████      | 2654/6470 [30:33<42:28,  1.50it/s]
3|2655|Loss: 0.7298552989959717:  41%|████      | 2654/6470 [30:33<42:28,  1.50it/s]
3|2655|Loss: 0.7298552989959717:  41%|████      | 2655/6470 [30:33<40:58,  1.55it/s]
3|2656|Loss: 0.8019499182701111:  41%|████      | 2655/6470 [30:34<40:58,  1.55it/s]
3|2656|Loss: 0.8019499182701111:  41%|████      | 2656/6470 [30:34<42:14,  1.50it/s]
3|2657|Loss: 0.8222930431365967:  41%|████      | 2656/6470 [30:34<42:14,  1.50it/s]
3|2657|Loss: 0.8222930431365967:  41%|████      | 2657/6470 [30:35<45:44,  1.39it/s]
3|2658|Loss: 0.8759230971336365:  41%|████      | 2657/6470 [30:35<45:44,  1.39it/s]
3|2658|Loss: 0.8759230971336365:  41%|████      | 2658/6470 [30:36<48:45,  1.30it/s]
3|2659|Loss: 0.98308265209198:  41%|████      | 2658/6470 [30:36<48:45,  1.30it/s]  
3|2659|Loss: 0.98308265209198:  41%|████      | 2659/6470 [30:36<45:17,  1.40it/s]
3|2660|Loss: 0.9832350611686707:  41%|████      | 2659/6470 [30:36<45:17,  1.40it/s]
3|2660|Loss: 0.9832350611686707:  41%|████      | 2660/6470 [30:37<43:01,  1.48it/s]
3|2661|Loss: 0.7626569867134094:  41%|████      | 2660/6470 [30:37<43:01,  1.48it/s]
3|2661|Loss: 0.7626569867134094:  41%|████      | 2661/6470 [30:37<41:52,  1.52it/s]
3|2662|Loss: 0.7693083882331848:  41%|████      | 2661/6470 [30:38<41:52,  1.52it/s]
3|2662|Loss: 0.7693083882331848:  41%|████      | 2662/6470 [30:38<44:22,  1.43it/s]
3|2663|Loss: 0.8954496383666992:  41%|████      | 2662/6470 [30:39<44:22,  1.43it/s]
3|2663|Loss: 0.8954496383666992:  41%|████      | 2663/6470 [30:39<41:06,  1.54it/s]
3|2664|Loss: 0.842211902141571:  41%|████      | 2663/6470 [30:39<41:06,  1.54it/s] 
3|2664|Loss: 0.842211902141571:  41%|████      | 2664/6470 [30:39<41:04,  1.54it/s]
3|2665|Loss: 0.7622601985931396:  41%|████      | 2664/6470 [30:40<41:04,  1.54it/s]
3|2665|Loss: 0.7622601985931396:  41%|████      | 2665/6470 [30:40<45:00,  1.41it/s]
3|2666|Loss: 0.8561477065086365:  41%|████      | 2665/6470 [30:41<45:00,  1.41it/s]
3|2666|Loss: 0.8561477065086365:  41%|████      | 2666/6470 [30:41<48:40,  1.30it/s]
3|2667|Loss: 0.9713835716247559:  41%|████      | 2666/6470 [30:42<48:40,  1.30it/s]
3|2667|Loss: 0.9713835716247559:  41%|████      | 2667/6470 [30:42<50:37,  1.25it/s]
3|2668|Loss: 0.8597114086151123:  41%|████      | 2667/6470 [30:42<50:37,  1.25it/s]
3|2668|Loss: 0.8597114086151123:  41%|████      | 2668/6470 [30:43<50:50,  1.25it/s]
3|2669|Loss: 0.796977698802948:  41%|████      | 2668/6470 [30:43<50:50,  1.25it/s] 
3|2669|Loss: 0.796977698802948:  41%|████▏     | 2669/6470 [30:44<49:19,  1.28it/s]
3|2670|Loss: 0.8597016334533691:  41%|████▏     | 2669/6470 [30:44<49:19,  1.28it/s]
3|2670|Loss: 0.8597016334533691:  41%|████▏     | 2670/6470 [30:44<46:57,  1.35it/s]
3|2671|Loss: 0.9989805817604065:  41%|████▏     | 2670/6470 [30:45<46:57,  1.35it/s]
3|2671|Loss: 0.9989805817604065:  41%|████▏     | 2671/6470 [30:45<45:57,  1.38it/s]
3|2672|Loss: 0.9344192743301392:  41%|████▏     | 2671/6470 [30:45<45:57,  1.38it/s]
3|2672|Loss: 0.9344192743301392:  41%|████▏     | 2672/6470 [30:46<44:34,  1.42it/s]
3|2673|Loss: 0.8145490288734436:  41%|████▏     | 2672/6470 [30:46<44:34,  1.42it/s]
3|2673|Loss: 0.8145490288734436:  41%|████▏     | 2673/6470 [30:46<42:55,  1.47it/s]
3|2674|Loss: 0.9038386344909668:  41%|████▏     | 2673/6470 [30:46<42:55,  1.47it/s]
3|2674|Loss: 0.9038386344909668:  41%|████▏     | 2674/6470 [30:47<41:48,  1.51it/s]
3|2675|Loss: 0.8121346831321716:  41%|████▏     | 2674/6470 [30:47<41:48,  1.51it/s]
3|2675|Loss: 0.8121346831321716:  41%|████▏     | 2675/6470 [30:47<43:08,  1.47it/s]
3|2676|Loss: 0.8114821314811707:  41%|████▏     | 2675/6470 [30:48<43:08,  1.47it/s]
3|2676|Loss: 0.8114821314811707:  41%|████▏     | 2676/6470 [30:48<42:46,  1.48it/s]
3|2677|Loss: 1.002751350402832:  41%|████▏     | 2676/6470 [30:48<42:46,  1.48it/s] 
3|2677|Loss: 1.002751350402832:  41%|████▏     | 2677/6470 [30:49<40:57,  1.54it/s]
3|2678|Loss: 0.8969849944114685:  41%|████▏     | 2677/6470 [30:49<40:57,  1.54it/s]
3|2678|Loss: 0.8969849944114685:  41%|████▏     | 2678/6470 [30:49<40:10,  1.57it/s]
3|2679|Loss: 0.8338403701782227:  41%|████▏     | 2678/6470 [30:50<40:10,  1.57it/s]
3|2679|Loss: 0.8338403701782227:  41%|████▏     | 2679/6470 [30:50<40:11,  1.57it/s]
3|2680|Loss: 0.7531282901763916:  41%|████▏     | 2679/6470 [30:50<40:11,  1.57it/s]
3|2680|Loss: 0.7531282901763916:  41%|████▏     | 2680/6470 [30:51<43:11,  1.46it/s]
3|2681|Loss: 0.8745347261428833:  41%|████▏     | 2680/6470 [30:51<43:11,  1.46it/s]
3|2681|Loss: 0.8745347261428833:  41%|████▏     | 2681/6470 [30:51<42:08,  1.50it/s]
3|2682|Loss: 0.8392057418823242:  41%|████▏     | 2681/6470 [30:52<42:08,  1.50it/s]
3|2682|Loss: 0.8392057418823242:  41%|████▏     | 2682/6470 [30:52<40:18,  1.57it/s]
3|2683|Loss: 0.9265222549438477:  41%|████▏     | 2682/6470 [30:52<40:18,  1.57it/s]
3|2683|Loss: 0.9265222549438477:  41%|████▏     | 2683/6470 [30:53<40:05,  1.57it/s]
3|2684|Loss: 1.0149555206298828:  41%|████▏     | 2683/6470 [30:53<40:05,  1.57it/s]
3|2684|Loss: 1.0149555206298828:  41%|████▏     | 2684/6470 [30:53<40:18,  1.57it/s]
3|2685|Loss: 0.8211474418640137:  41%|████▏     | 2684/6470 [30:54<40:18,  1.57it/s]
3|2685|Loss: 0.8211474418640137:  41%|████▏     | 2685/6470 [30:54<43:38,  1.45it/s]
3|2686|Loss: 0.8669679164886475:  41%|████▏     | 2685/6470 [30:54<43:38,  1.45it/s]
3|2686|Loss: 0.8669679164886475:  42%|████▏     | 2686/6470 [30:55<42:35,  1.48it/s]
3|2687|Loss: 0.7368154525756836:  42%|████▏     | 2686/6470 [30:55<42:35,  1.48it/s]
3|2687|Loss: 0.7368154525756836:  42%|████▏     | 2687/6470 [30:55<40:53,  1.54it/s]
3|2688|Loss: 0.6853329539299011:  42%|████▏     | 2687/6470 [30:56<40:53,  1.54it/s]
3|2688|Loss: 0.6853329539299011:  42%|████▏     | 2688/6470 [30:56<39:04,  1.61it/s]
3|2689|Loss: 0.9632796049118042:  42%|████▏     | 2688/6470 [30:56<39:04,  1.61it/s]
3|2689|Loss: 0.9632796049118042:  42%|████▏     | 2689/6470 [30:56<38:18,  1.65it/s]
3|2690|Loss: 0.7652430534362793:  42%|████▏     | 2689/6470 [30:57<38:18,  1.65it/s]
3|2690|Loss: 0.7652430534362793:  42%|████▏     | 2690/6470 [30:57<41:41,  1.51it/s]
3|2691|Loss: 0.8296682834625244:  42%|████▏     | 2690/6470 [30:58<41:41,  1.51it/s]
3|2691|Loss: 0.8296682834625244:  42%|████▏     | 2691/6470 [30:58<41:46,  1.51it/s]
3|2692|Loss: 0.7638676762580872:  42%|████▏     | 2691/6470 [30:58<41:46,  1.51it/s]
3|2692|Loss: 0.7638676762580872:  42%|████▏     | 2692/6470 [30:58<39:30,  1.59it/s]
3|2693|Loss: 0.6833947896957397:  42%|████▏     | 2692/6470 [30:59<39:30,  1.59it/s]
3|2693|Loss: 0.6833947896957397:  42%|████▏     | 2693/6470 [30:59<39:00,  1.61it/s]
3|2694|Loss: 0.873671293258667:  42%|████▏     | 2693/6470 [30:59<39:00,  1.61it/s] 
3|2694|Loss: 0.873671293258667:  42%|████▏     | 2694/6470 [31:00<41:02,  1.53it/s]
3|2695|Loss: 0.7650606036186218:  42%|████▏     | 2694/6470 [31:00<41:02,  1.53it/s]
3|2695|Loss: 0.7650606036186218:  42%|████▏     | 2695/6470 [31:00<41:02,  1.53it/s]
3|2696|Loss: 0.9649816751480103:  42%|████▏     | 2695/6470 [31:01<41:02,  1.53it/s]
3|2696|Loss: 0.9649816751480103:  42%|████▏     | 2696/6470 [31:01<42:19,  1.49it/s]
3|2697|Loss: 0.7087927460670471:  42%|████▏     | 2696/6470 [31:02<42:19,  1.49it/s]
3|2697|Loss: 0.7087927460670471:  42%|████▏     | 2697/6470 [31:02<43:31,  1.44it/s]
3|2698|Loss: 0.845670223236084:  42%|████▏     | 2697/6470 [31:02<43:31,  1.44it/s] 
3|2698|Loss: 0.845670223236084:  42%|████▏     | 2698/6470 [31:03<45:15,  1.39it/s]
3|2699|Loss: 0.744513750076294:  42%|████▏     | 2698/6470 [31:03<45:15,  1.39it/s]
3|2699|Loss: 0.744513750076294:  42%|████▏     | 2699/6470 [31:03<42:26,  1.48it/s]
3|2700|Loss: 0.7527229189872742:  42%|████▏     | 2699/6470 [31:04<42:26,  1.48it/s]
3|2700|Loss: 0.7527229189872742:  42%|████▏     | 2700/6470 [31:04<42:34,  1.48it/s]
3|2701|Loss: 0.91492760181427:  42%|████▏     | 2700/6470 [31:04<42:34,  1.48it/s]  
3|2701|Loss: 0.91492760181427:  42%|████▏     | 2701/6470 [31:04<40:16,  1.56it/s]
3|2702|Loss: 0.8222801685333252:  42%|████▏     | 2701/6470 [31:05<40:16,  1.56it/s]
3|2702|Loss: 0.8222801685333252:  42%|████▏     | 2702/6470 [31:05<39:25,  1.59it/s]
3|2703|Loss: 0.8928752541542053:  42%|████▏     | 2702/6470 [31:05<39:25,  1.59it/s]
3|2703|Loss: 0.8928752541542053:  42%|████▏     | 2703/6470 [31:06<40:32,  1.55it/s]
3|2704|Loss: 0.8521880507469177:  42%|████▏     | 2703/6470 [31:06<40:32,  1.55it/s]
3|2704|Loss: 0.8521880507469177:  42%|████▏     | 2704/6470 [31:06<40:48,  1.54it/s]
3|2705|Loss: 0.9093493223190308:  42%|████▏     | 2704/6470 [31:07<40:48,  1.54it/s]
3|2705|Loss: 0.9093493223190308:  42%|████▏     | 2705/6470 [31:07<42:21,  1.48it/s]
3|2706|Loss: 0.9086198210716248:  42%|████▏     | 2705/6470 [31:08<42:21,  1.48it/s]
3|2706|Loss: 0.9086198210716248:  42%|████▏     | 2706/6470 [31:08<45:02,  1.39it/s]
3|2707|Loss: 0.8847970366477966:  42%|████▏     | 2706/6470 [31:08<45:02,  1.39it/s]
3|2707|Loss: 0.8847970366477966:  42%|████▏     | 2707/6470 [31:09<44:08,  1.42it/s]
3|2708|Loss: 0.7418512105941772:  42%|████▏     | 2707/6470 [31:09<44:08,  1.42it/s]
3|2708|Loss: 0.7418512105941772:  42%|████▏     | 2708/6470 [31:09<45:17,  1.38it/s]
3|2709|Loss: 0.9394128322601318:  42%|████▏     | 2708/6470 [31:10<45:17,  1.38it/s]
3|2709|Loss: 0.9394128322601318:  42%|████▏     | 2709/6470 [31:10<47:38,  1.32it/s]
3|2710|Loss: 0.8237295746803284:  42%|████▏     | 2709/6470 [31:11<47:38,  1.32it/s]
3|2710|Loss: 0.8237295746803284:  42%|████▏     | 2710/6470 [31:11<44:33,  1.41it/s]
3|2711|Loss: 0.8173666596412659:  42%|████▏     | 2710/6470 [31:11<44:33,  1.41it/s]
3|2711|Loss: 0.8173666596412659:  42%|████▏     | 2711/6470 [31:12<46:17,  1.35it/s]
3|2712|Loss: 0.9367362260818481:  42%|████▏     | 2711/6470 [31:12<46:17,  1.35it/s]
3|2712|Loss: 0.9367362260818481:  42%|████▏     | 2712/6470 [31:12<46:40,  1.34it/s]
3|2713|Loss: 0.8408027291297913:  42%|████▏     | 2712/6470 [31:13<46:40,  1.34it/s]
3|2713|Loss: 0.8408027291297913:  42%|████▏     | 2713/6470 [31:13<48:45,  1.28it/s]
3|2714|Loss: 0.9239731431007385:  42%|████▏     | 2713/6470 [31:14<48:45,  1.28it/s]
3|2714|Loss: 0.9239731431007385:  42%|████▏     | 2714/6470 [31:14<47:21,  1.32it/s]
3|2715|Loss: 0.79637610912323:  42%|████▏     | 2714/6470 [31:14<47:21,  1.32it/s]  
3|2715|Loss: 0.79637610912323:  42%|████▏     | 2715/6470 [31:15<46:51,  1.34it/s]
3|2716|Loss: 0.7089344263076782:  42%|████▏     | 2715/6470 [31:15<46:51,  1.34it/s]
3|2716|Loss: 0.7089344263076782:  42%|████▏     | 2716/6470 [31:15<46:06,  1.36it/s]
3|2717|Loss: 0.9767215847969055:  42%|████▏     | 2716/6470 [31:16<46:06,  1.36it/s]
3|2717|Loss: 0.9767215847969055:  42%|████▏     | 2717/6470 [31:16<45:51,  1.36it/s]
3|2718|Loss: 0.9593766927719116:  42%|████▏     | 2717/6470 [31:17<45:51,  1.36it/s]
3|2718|Loss: 0.9593766927719116:  42%|████▏     | 2718/6470 [31:17<45:58,  1.36it/s]
3|2719|Loss: 0.9090859293937683:  42%|████▏     | 2718/6470 [31:17<45:58,  1.36it/s]
3|2719|Loss: 0.9090859293937683:  42%|████▏     | 2719/6470 [31:18<46:24,  1.35it/s]
3|2720|Loss: 0.7451092600822449:  42%|████▏     | 2719/6470 [31:18<46:24,  1.35it/s]
3|2720|Loss: 0.7451092600822449:  42%|████▏     | 2720/6470 [31:18<47:25,  1.32it/s]
3|2721|Loss: 0.905669629573822:  42%|████▏     | 2720/6470 [31:19<47:25,  1.32it/s] 
3|2721|Loss: 0.905669629573822:  42%|████▏     | 2721/6470 [31:19<47:59,  1.30it/s]
3|2722|Loss: 0.8977243900299072:  42%|████▏     | 2721/6470 [31:20<47:59,  1.30it/s]
3|2722|Loss: 0.8977243900299072:  42%|████▏     | 2722/6470 [31:20<46:11,  1.35it/s]
3|2723|Loss: 0.8367142081260681:  42%|████▏     | 2722/6470 [31:20<46:11,  1.35it/s]
3|2723|Loss: 0.8367142081260681:  42%|████▏     | 2723/6470 [31:20<43:18,  1.44it/s]
3|2724|Loss: 0.9287067651748657:  42%|████▏     | 2723/6470 [31:21<43:18,  1.44it/s]
3|2724|Loss: 0.9287067651748657:  42%|████▏     | 2724/6470 [31:21<41:19,  1.51it/s]
3|2725|Loss: 0.8437113165855408:  42%|████▏     | 2724/6470 [31:21<41:19,  1.51it/s]
3|2725|Loss: 0.8437113165855408:  42%|████▏     | 2725/6470 [31:22<41:38,  1.50it/s]
3|2726|Loss: 0.8490817546844482:  42%|████▏     | 2725/6470 [31:22<41:38,  1.50it/s]
3|2726|Loss: 0.8490817546844482:  42%|████▏     | 2726/6470 [31:22<42:46,  1.46it/s]
3|2727|Loss: 0.78178870677948:  42%|████▏     | 2726/6470 [31:23<42:46,  1.46it/s]  
3|2727|Loss: 0.78178870677948:  42%|████▏     | 2727/6470 [31:23<44:40,  1.40it/s]
3|2728|Loss: 0.8131445646286011:  42%|████▏     | 2727/6470 [31:24<44:40,  1.40it/s]
3|2728|Loss: 0.8131445646286011:  42%|████▏     | 2728/6470 [31:24<41:55,  1.49it/s]
3|2729|Loss: 0.9600399732589722:  42%|████▏     | 2728/6470 [31:24<41:55,  1.49it/s]
3|2729|Loss: 0.9600399732589722:  42%|████▏     | 2729/6470 [31:25<44:04,  1.41it/s]
3|2730|Loss: 0.8044484853744507:  42%|████▏     | 2729/6470 [31:25<44:04,  1.41it/s]
3|2730|Loss: 0.8044484853744507:  42%|████▏     | 2730/6470 [31:25<44:39,  1.40it/s]
3|2731|Loss: 0.9134063124656677:  42%|████▏     | 2730/6470 [31:26<44:39,  1.40it/s]
3|2731|Loss: 0.9134063124656677:  42%|████▏     | 2731/6470 [31:26<43:30,  1.43it/s]
3|2732|Loss: 0.8920819163322449:  42%|████▏     | 2731/6470 [31:26<43:30,  1.43it/s]
3|2732|Loss: 0.8920819163322449:  42%|████▏     | 2732/6470 [31:27<43:22,  1.44it/s]
3|2733|Loss: 0.8694155812263489:  42%|████▏     | 2732/6470 [31:27<43:22,  1.44it/s]
3|2733|Loss: 0.8694155812263489:  42%|████▏     | 2733/6470 [31:27<44:09,  1.41it/s]
3|2734|Loss: 0.9231244921684265:  42%|████▏     | 2733/6470 [31:28<44:09,  1.41it/s]
3|2734|Loss: 0.9231244921684265:  42%|████▏     | 2734/6470 [31:28<42:23,  1.47it/s]
3|2735|Loss: 0.9241661429405212:  42%|████▏     | 2734/6470 [31:28<42:23,  1.47it/s]
3|2735|Loss: 0.9241661429405212:  42%|████▏     | 2735/6470 [31:29<39:28,  1.58it/s]
3|2736|Loss: 0.8770577311515808:  42%|████▏     | 2735/6470 [31:29<39:28,  1.58it/s]
3|2736|Loss: 0.8770577311515808:  42%|████▏     | 2736/6470 [31:29<39:00,  1.60it/s]
3|2737|Loss: 0.9022122025489807:  42%|████▏     | 2736/6470 [31:30<39:00,  1.60it/s]
3|2737|Loss: 0.9022122025489807:  42%|████▏     | 2737/6470 [31:30<40:17,  1.54it/s]
3|2738|Loss: 0.6997191309928894:  42%|████▏     | 2737/6470 [31:30<40:17,  1.54it/s]
3|2738|Loss: 0.6997191309928894:  42%|████▏     | 2738/6470 [31:31<43:43,  1.42it/s]
3|2739|Loss: 0.8553779125213623:  42%|████▏     | 2738/6470 [31:31<43:43,  1.42it/s]
3|2739|Loss: 0.8553779125213623:  42%|████▏     | 2739/6470 [31:31<43:26,  1.43it/s]
3|2740|Loss: 0.9145573973655701:  42%|████▏     | 2739/6470 [31:32<43:26,  1.43it/s]
3|2740|Loss: 0.9145573973655701:  42%|████▏     | 2740/6470 [31:32<44:26,  1.40it/s]
3|2741|Loss: 0.7528573274612427:  42%|████▏     | 2740/6470 [31:33<44:26,  1.40it/s]
3|2741|Loss: 0.7528573274612427:  42%|████▏     | 2741/6470 [31:33<46:28,  1.34it/s]
3|2742|Loss: 0.7457537055015564:  42%|████▏     | 2741/6470 [31:33<46:28,  1.34it/s]
3|2742|Loss: 0.7457537055015564:  42%|████▏     | 2742/6470 [31:34<45:14,  1.37it/s]
3|2743|Loss: 0.920996904373169:  42%|████▏     | 2742/6470 [31:34<45:14,  1.37it/s] 
3|2743|Loss: 0.920996904373169:  42%|████▏     | 2743/6470 [31:34<44:31,  1.40it/s]
3|2744|Loss: 0.9384096264839172:  42%|████▏     | 2743/6470 [31:35<44:31,  1.40it/s]
3|2744|Loss: 0.9384096264839172:  42%|████▏     | 2744/6470 [31:35<44:14,  1.40it/s]
3|2745|Loss: 0.7177059650421143:  42%|████▏     | 2744/6470 [31:35<44:14,  1.40it/s]
3|2745|Loss: 0.7177059650421143:  42%|████▏     | 2745/6470 [31:36<44:43,  1.39it/s]
3|2746|Loss: 0.9055300951004028:  42%|████▏     | 2745/6470 [31:36<44:43,  1.39it/s]
3|2746|Loss: 0.9055300951004028:  42%|████▏     | 2746/6470 [31:37<46:47,  1.33it/s]
3|2747|Loss: 1.0585711002349854:  42%|████▏     | 2746/6470 [31:37<46:47,  1.33it/s]
3|2747|Loss: 1.0585711002349854:  42%|████▏     | 2747/6470 [31:37<44:33,  1.39it/s]
3|2748|Loss: 0.9999145269393921:  42%|████▏     | 2747/6470 [31:38<44:33,  1.39it/s]
3|2748|Loss: 0.9999145269393921:  42%|████▏     | 2748/6470 [31:38<41:57,  1.48it/s]
3|2749|Loss: 0.9772325158119202:  42%|████▏     | 2748/6470 [31:38<41:57,  1.48it/s]
3|2749|Loss: 0.9772325158119202:  42%|████▏     | 2749/6470 [31:39<44:16,  1.40it/s]
3|2750|Loss: 0.7580913305282593:  42%|████▏     | 2749/6470 [31:39<44:16,  1.40it/s]
3|2750|Loss: 0.7580913305282593:  43%|████▎     | 2750/6470 [31:39<44:34,  1.39it/s]
3|2751|Loss: 0.8597218990325928:  43%|████▎     | 2750/6470 [31:40<44:34,  1.39it/s]
3|2751|Loss: 0.8597218990325928:  43%|████▎     | 2751/6470 [31:40<46:01,  1.35it/s]
3|2752|Loss: 0.9821590781211853:  43%|████▎     | 2751/6470 [31:41<46:01,  1.35it/s]
3|2752|Loss: 0.9821590781211853:  43%|████▎     | 2752/6470 [31:41<44:33,  1.39it/s]
3|2753|Loss: 0.7323806285858154:  43%|████▎     | 2752/6470 [31:41<44:33,  1.39it/s]
3|2753|Loss: 0.7323806285858154:  43%|████▎     | 2753/6470 [31:42<47:50,  1.29it/s]
3|2754|Loss: 0.7174654603004456:  43%|████▎     | 2753/6470 [31:42<47:50,  1.29it/s]
3|2754|Loss: 0.7174654603004456:  43%|████▎     | 2754/6470 [31:42<48:05,  1.29it/s]
3|2755|Loss: 0.7487466335296631:  43%|████▎     | 2754/6470 [31:43<48:05,  1.29it/s]
3|2755|Loss: 0.7487466335296631:  43%|████▎     | 2755/6470 [31:43<47:41,  1.30it/s]
3|2756|Loss: 0.881531834602356:  43%|████▎     | 2755/6470 [31:44<47:41,  1.30it/s] 
3|2756|Loss: 0.881531834602356:  43%|████▎     | 2756/6470 [31:44<46:42,  1.33it/s]
3|2757|Loss: 0.890548586845398:  43%|████▎     | 2756/6470 [31:44<46:42,  1.33it/s]
3|2757|Loss: 0.890548586845398:  43%|████▎     | 2757/6470 [31:45<46:11,  1.34it/s]
3|2758|Loss: 0.7113341093063354:  43%|████▎     | 2757/6470 [31:45<46:11,  1.34it/s]
3|2758|Loss: 0.7113341093063354:  43%|████▎     | 2758/6470 [31:45<43:14,  1.43it/s]
3|2759|Loss: 0.9973469376564026:  43%|████▎     | 2758/6470 [31:46<43:14,  1.43it/s]
3|2759|Loss: 0.9973469376564026:  43%|████▎     | 2759/6470 [31:46<43:23,  1.43it/s]
3|2760|Loss: 0.867416262626648:  43%|████▎     | 2759/6470 [31:46<43:23,  1.43it/s] 
3|2760|Loss: 0.867416262626648:  43%|████▎     | 2760/6470 [31:47<44:25,  1.39it/s]
3|2761|Loss: 0.887547492980957:  43%|████▎     | 2760/6470 [31:47<44:25,  1.39it/s]
3|2761|Loss: 0.887547492980957:  43%|████▎     | 2761/6470 [31:47<43:31,  1.42it/s]
3|2762|Loss: 0.9285770654678345:  43%|████▎     | 2761/6470 [31:48<43:31,  1.42it/s]
3|2762|Loss: 0.9285770654678345:  43%|████▎     | 2762/6470 [31:48<43:10,  1.43it/s]
3|2763|Loss: 0.8115169405937195:  43%|████▎     | 2762/6470 [31:49<43:10,  1.43it/s]
3|2763|Loss: 0.8115169405937195:  43%|████▎     | 2763/6470 [31:49<44:41,  1.38it/s]
3|2764|Loss: 0.8062233924865723:  43%|████▎     | 2763/6470 [31:49<44:41,  1.38it/s]
3|2764|Loss: 0.8062233924865723:  43%|████▎     | 2764/6470 [31:49<42:29,  1.45it/s]
3|2765|Loss: 0.9172683954238892:  43%|████▎     | 2764/6470 [31:50<42:29,  1.45it/s]
3|2765|Loss: 0.9172683954238892:  43%|████▎     | 2765/6470 [31:50<39:19,  1.57it/s]
3|2766|Loss: 0.8132193088531494:  43%|████▎     | 2765/6470 [31:50<39:19,  1.57it/s]
3|2766|Loss: 0.8132193088531494:  43%|████▎     | 2766/6470 [31:51<38:27,  1.61it/s]
3|2767|Loss: 0.9905941486358643:  43%|████▎     | 2766/6470 [31:51<38:27,  1.61it/s]
3|2767|Loss: 0.9905941486358643:  43%|████▎     | 2767/6470 [31:51<39:36,  1.56it/s]
3|2768|Loss: 0.8903836607933044:  43%|████▎     | 2767/6470 [31:52<39:36,  1.56it/s]
3|2768|Loss: 0.8903836607933044:  43%|████▎     | 2768/6470 [31:52<42:37,  1.45it/s]
3|2769|Loss: 0.9731732606887817:  43%|████▎     | 2768/6470 [31:53<42:37,  1.45it/s]
3|2769|Loss: 0.9731732606887817:  43%|████▎     | 2769/6470 [31:53<44:48,  1.38it/s]
3|2770|Loss: 0.9758060574531555:  43%|████▎     | 2769/6470 [31:53<44:48,  1.38it/s]
3|2770|Loss: 0.9758060574531555:  43%|████▎     | 2770/6470 [31:53<41:55,  1.47it/s]
3|2771|Loss: 0.8784082531929016:  43%|████▎     | 2770/6470 [31:54<41:55,  1.47it/s]
3|2771|Loss: 0.8784082531929016:  43%|████▎     | 2771/6470 [31:54<42:17,  1.46it/s]
3|2772|Loss: 0.6210166215896606:  43%|████▎     | 2771/6470 [31:55<42:17,  1.46it/s]
3|2772|Loss: 0.6210166215896606:  43%|████▎     | 2772/6470 [31:55<41:10,  1.50it/s]
3|2773|Loss: 0.8084203004837036:  43%|████▎     | 2772/6470 [31:55<41:10,  1.50it/s]
3|2773|Loss: 0.8084203004837036:  43%|████▎     | 2773/6470 [31:55<40:29,  1.52it/s]
3|2774|Loss: 0.8975988030433655:  43%|████▎     | 2773/6470 [31:56<40:29,  1.52it/s]
3|2774|Loss: 0.8975988030433655:  43%|████▎     | 2774/6470 [31:56<41:36,  1.48it/s]
3|2775|Loss: 0.900884747505188:  43%|████▎     | 2774/6470 [31:57<41:36,  1.48it/s] 
3|2775|Loss: 0.900884747505188:  43%|████▎     | 2775/6470 [31:57<42:49,  1.44it/s]
3|2776|Loss: 0.8159018754959106:  43%|████▎     | 2775/6470 [31:57<42:49,  1.44it/s]
3|2776|Loss: 0.8159018754959106:  43%|████▎     | 2776/6470 [31:58<41:57,  1.47it/s]
3|2777|Loss: 0.8159744143486023:  43%|████▎     | 2776/6470 [31:58<41:57,  1.47it/s]
3|2777|Loss: 0.8159744143486023:  43%|████▎     | 2777/6470 [31:58<40:44,  1.51it/s]
3|2778|Loss: 0.8156868815422058:  43%|████▎     | 2777/6470 [31:59<40:44,  1.51it/s]
3|2778|Loss: 0.8156868815422058:  43%|████▎     | 2778/6470 [31:59<43:26,  1.42it/s]
3|2779|Loss: 0.8499248027801514:  43%|████▎     | 2778/6470 [31:59<43:26,  1.42it/s]
3|2779|Loss: 0.8499248027801514:  43%|████▎     | 2779/6470 [32:00<44:36,  1.38it/s]
3|2780|Loss: 0.920502245426178:  43%|████▎     | 2779/6470 [32:00<44:36,  1.38it/s] 
3|2780|Loss: 0.920502245426178:  43%|████▎     | 2780/6470 [32:00<44:45,  1.37it/s]
3|2781|Loss: 0.9491607546806335:  43%|████▎     | 2780/6470 [32:01<44:45,  1.37it/s]
3|2781|Loss: 0.9491607546806335:  43%|████▎     | 2781/6470 [32:01<42:30,  1.45it/s]
3|2782|Loss: 0.837133526802063:  43%|████▎     | 2781/6470 [32:01<42:30,  1.45it/s] 
3|2782|Loss: 0.837133526802063:  43%|████▎     | 2782/6470 [32:02<38:36,  1.59it/s]
3|2783|Loss: 0.7439185380935669:  43%|████▎     | 2782/6470 [32:02<38:36,  1.59it/s]
3|2783|Loss: 0.7439185380935669:  43%|████▎     | 2783/6470 [32:02<37:46,  1.63it/s]
3|2784|Loss: 0.8344225287437439:  43%|████▎     | 2783/6470 [32:03<37:46,  1.63it/s]
3|2784|Loss: 0.8344225287437439:  43%|████▎     | 2784/6470 [32:03<41:40,  1.47it/s]
3|2785|Loss: 0.9099779725074768:  43%|████▎     | 2784/6470 [32:03<41:40,  1.47it/s]
3|2785|Loss: 0.9099779725074768:  43%|████▎     | 2785/6470 [32:03<37:52,  1.62it/s]
3|2786|Loss: 0.8415892124176025:  43%|████▎     | 2785/6470 [32:04<37:52,  1.62it/s]
3|2786|Loss: 0.8415892124176025:  43%|████▎     | 2786/6470 [32:04<39:35,  1.55it/s]
3|2787|Loss: 0.8108540177345276:  43%|████▎     | 2786/6470 [32:05<39:35,  1.55it/s]
3|2787|Loss: 0.8108540177345276:  43%|████▎     | 2787/6470 [32:05<41:16,  1.49it/s]
3|2788|Loss: 0.7310287356376648:  43%|████▎     | 2787/6470 [32:05<41:16,  1.49it/s]
3|2788|Loss: 0.7310287356376648:  43%|████▎     | 2788/6470 [32:06<42:17,  1.45it/s]
3|2789|Loss: 0.9026026725769043:  43%|████▎     | 2788/6470 [32:06<42:17,  1.45it/s]
3|2789|Loss: 0.9026026725769043:  43%|████▎     | 2789/6470 [32:06<39:41,  1.55it/s]
3|2790|Loss: 0.7823639512062073:  43%|████▎     | 2789/6470 [32:07<39:41,  1.55it/s]
3|2790|Loss: 0.7823639512062073:  43%|████▎     | 2790/6470 [32:07<42:19,  1.45it/s]
3|2791|Loss: 0.8598151206970215:  43%|████▎     | 2790/6470 [32:07<42:19,  1.45it/s]
3|2791|Loss: 0.8598151206970215:  43%|████▎     | 2791/6470 [32:08<43:53,  1.40it/s]
3|2792|Loss: 0.8934707641601562:  43%|████▎     | 2791/6470 [32:08<43:53,  1.40it/s]
3|2792|Loss: 0.8934707641601562:  43%|████▎     | 2792/6470 [32:08<42:58,  1.43it/s]
3|2793|Loss: 0.8525732159614563:  43%|████▎     | 2792/6470 [32:09<42:58,  1.43it/s]
3|2793|Loss: 0.8525732159614563:  43%|████▎     | 2793/6470 [32:09<40:12,  1.52it/s]
3|2794|Loss: 0.8542609214782715:  43%|████▎     | 2793/6470 [32:09<40:12,  1.52it/s]
3|2794|Loss: 0.8542609214782715:  43%|████▎     | 2794/6470 [32:10<40:58,  1.50it/s]
3|2795|Loss: 0.9518008232116699:  43%|████▎     | 2794/6470 [32:10<40:58,  1.50it/s]
3|2795|Loss: 0.9518008232116699:  43%|████▎     | 2795/6470 [32:10<44:01,  1.39it/s]
3|2796|Loss: 0.8675068020820618:  43%|████▎     | 2795/6470 [32:11<44:01,  1.39it/s]
3|2796|Loss: 0.8675068020820618:  43%|████▎     | 2796/6470 [32:11<46:38,  1.31it/s]
3|2797|Loss: 0.9749074578285217:  43%|████▎     | 2796/6470 [32:12<46:38,  1.31it/s]
3|2797|Loss: 0.9749074578285217:  43%|████▎     | 2797/6470 [32:12<40:55,  1.50it/s]
3|2798|Loss: 0.7788341045379639:  43%|████▎     | 2797/6470 [32:12<40:55,  1.50it/s]
3|2798|Loss: 0.7788341045379639:  43%|████▎     | 2798/6470 [32:13<41:51,  1.46it/s]
3|2799|Loss: 0.848384439945221:  43%|████▎     | 2798/6470 [32:13<41:51,  1.46it/s] 
3|2799|Loss: 0.848384439945221:  43%|████▎     | 2799/6470 [32:13<41:51,  1.46it/s]
3|2800|Loss: 0.749360978603363:  43%|████▎     | 2799/6470 [32:14<41:51,  1.46it/s]
3|2800|Loss: 0.749360978603363:  43%|████▎     | 2800/6470 [32:14<41:02,  1.49it/s]
3|2801|Loss: 0.9123660326004028:  43%|████▎     | 2800/6470 [32:14<41:02,  1.49it/s]
3|2801|Loss: 0.9123660326004028:  43%|████▎     | 2801/6470 [32:15<42:13,  1.45it/s]
3|2802|Loss: 0.7915847897529602:  43%|████▎     | 2801/6470 [32:15<42:13,  1.45it/s]
3|2802|Loss: 0.7915847897529602:  43%|████▎     | 2802/6470 [32:15<43:19,  1.41it/s]
3|2803|Loss: 0.9703695774078369:  43%|████▎     | 2802/6470 [32:16<43:19,  1.41it/s]
3|2803|Loss: 0.9703695774078369:  43%|████▎     | 2803/6470 [32:16<41:15,  1.48it/s]
3|2804|Loss: 0.7604220509529114:  43%|████▎     | 2803/6470 [32:16<41:15,  1.48it/s]
3|2804|Loss: 0.7604220509529114:  43%|████▎     | 2804/6470 [32:17<44:57,  1.36it/s]
3|2805|Loss: 0.8870370984077454:  43%|████▎     | 2804/6470 [32:17<44:57,  1.36it/s]
3|2805|Loss: 0.8870370984077454:  43%|████▎     | 2805/6470 [32:18<48:10,  1.27it/s]
3|2806|Loss: 0.9315456748008728:  43%|████▎     | 2805/6470 [32:18<48:10,  1.27it/s]
3|2806|Loss: 0.9315456748008728:  43%|████▎     | 2806/6470 [32:18<44:55,  1.36it/s]
3|2807|Loss: 0.8389363288879395:  43%|████▎     | 2806/6470 [32:19<44:55,  1.36it/s]
3|2807|Loss: 0.8389363288879395:  43%|████▎     | 2807/6470 [32:19<43:32,  1.40it/s]
3|2808|Loss: 0.8126603364944458:  43%|████▎     | 2807/6470 [32:19<43:32,  1.40it/s]
3|2808|Loss: 0.8126603364944458:  43%|████▎     | 2808/6470 [32:20<44:44,  1.36it/s]
3|2809|Loss: 1.0876433849334717:  43%|████▎     | 2808/6470 [32:20<44:44,  1.36it/s]
3|2809|Loss: 1.0876433849334717:  43%|████▎     | 2809/6470 [32:21<46:46,  1.30it/s]
3|2810|Loss: 0.8688430190086365:  43%|████▎     | 2809/6470 [32:21<46:46,  1.30it/s]
3|2810|Loss: 0.8688430190086365:  43%|████▎     | 2810/6470 [32:21<42:34,  1.43it/s]
3|2811|Loss: 0.8472414612770081:  43%|████▎     | 2810/6470 [32:21<42:34,  1.43it/s]
3|2811|Loss: 0.8472414612770081:  43%|████▎     | 2811/6470 [32:22<38:31,  1.58it/s]
3|2812|Loss: 0.8872718214988708:  43%|████▎     | 2811/6470 [32:22<38:31,  1.58it/s]
3|2812|Loss: 0.8872718214988708:  43%|████▎     | 2812/6470 [32:22<39:40,  1.54it/s]
3|2813|Loss: 0.9178447127342224:  43%|████▎     | 2812/6470 [32:23<39:40,  1.54it/s]
3|2813|Loss: 0.9178447127342224:  43%|████▎     | 2813/6470 [32:23<39:45,  1.53it/s]
3|2814|Loss: 0.8555209636688232:  43%|████▎     | 2813/6470 [32:23<39:45,  1.53it/s]
3|2814|Loss: 0.8555209636688232:  43%|████▎     | 2814/6470 [32:24<42:29,  1.43it/s]
3|2815|Loss: 0.7375848889350891:  43%|████▎     | 2814/6470 [32:24<42:29,  1.43it/s]
3|2815|Loss: 0.7375848889350891:  44%|████▎     | 2815/6470 [32:24<39:30,  1.54it/s]
3|2816|Loss: 0.8736605644226074:  44%|████▎     | 2815/6470 [32:25<39:30,  1.54it/s]
3|2816|Loss: 0.8736605644226074:  44%|████▎     | 2816/6470 [32:25<39:35,  1.54it/s]
3|2817|Loss: 0.6939305663108826:  44%|████▎     | 2816/6470 [32:25<39:35,  1.54it/s]
3|2817|Loss: 0.6939305663108826:  44%|████▎     | 2817/6470 [32:26<38:19,  1.59it/s]
3|2818|Loss: 1.0062096118927002:  44%|████▎     | 2817/6470 [32:26<38:19,  1.59it/s]
3|2818|Loss: 1.0062096118927002:  44%|████▎     | 2818/6470 [32:26<35:30,  1.71it/s]
3|2819|Loss: 0.9164073467254639:  44%|████▎     | 2818/6470 [32:26<35:30,  1.71it/s]
3|2819|Loss: 0.9164073467254639:  44%|████▎     | 2819/6470 [32:27<37:06,  1.64it/s]
3|2820|Loss: 0.8218281865119934:  44%|████▎     | 2819/6470 [32:27<37:06,  1.64it/s]
3|2820|Loss: 0.8218281865119934:  44%|████▎     | 2820/6470 [32:27<38:07,  1.60it/s]
3|2821|Loss: 0.8641251921653748:  44%|████▎     | 2820/6470 [32:28<38:07,  1.60it/s]
3|2821|Loss: 0.8641251921653748:  44%|████▎     | 2821/6470 [32:28<34:19,  1.77it/s]
3|2822|Loss: 0.7729917764663696:  44%|████▎     | 2821/6470 [32:28<34:19,  1.77it/s]
3|2822|Loss: 0.7729917764663696:  44%|████▎     | 2822/6470 [32:28<34:21,  1.77it/s]
3|2823|Loss: 0.8317299485206604:  44%|████▎     | 2822/6470 [32:29<34:21,  1.77it/s]
3|2823|Loss: 0.8317299485206604:  44%|████▎     | 2823/6470 [32:29<35:20,  1.72it/s]
3|2824|Loss: 0.9682406783103943:  44%|████▎     | 2823/6470 [32:29<35:20,  1.72it/s]
3|2824|Loss: 0.9682406783103943:  44%|████▎     | 2824/6470 [32:30<38:43,  1.57it/s]
3|2825|Loss: 0.9593793153762817:  44%|████▎     | 2824/6470 [32:30<38:43,  1.57it/s]
3|2825|Loss: 0.9593793153762817:  44%|████▎     | 2825/6470 [32:30<39:19,  1.54it/s]
3|2826|Loss: 0.798224925994873:  44%|████▎     | 2825/6470 [32:31<39:19,  1.54it/s] 
3|2826|Loss: 0.798224925994873:  44%|████▎     | 2826/6470 [32:31<39:58,  1.52it/s]
3|2827|Loss: 0.8969842195510864:  44%|████▎     | 2826/6470 [32:32<39:58,  1.52it/s]
3|2827|Loss: 0.8969842195510864:  44%|████▎     | 2827/6470 [32:32<42:53,  1.42it/s]
3|2828|Loss: 0.9550265073776245:  44%|████▎     | 2827/6470 [32:32<42:53,  1.42it/s]
3|2828|Loss: 0.9550265073776245:  44%|████▎     | 2828/6470 [32:33<45:01,  1.35it/s]
3|2829|Loss: 0.6748385429382324:  44%|████▎     | 2828/6470 [32:33<45:01,  1.35it/s]
3|2829|Loss: 0.6748385429382324:  44%|████▎     | 2829/6470 [32:34<45:59,  1.32it/s]
3|2830|Loss: 0.8562151789665222:  44%|████▎     | 2829/6470 [32:34<45:59,  1.32it/s]
3|2830|Loss: 0.8562151789665222:  44%|████▎     | 2830/6470 [32:34<44:07,  1.37it/s]
3|2831|Loss: 0.7907199859619141:  44%|████▎     | 2830/6470 [32:35<44:07,  1.37it/s]
3|2831|Loss: 0.7907199859619141:  44%|████▍     | 2831/6470 [32:35<43:50,  1.38it/s]
3|2832|Loss: 0.8543151617050171:  44%|████▍     | 2831/6470 [32:35<43:50,  1.38it/s]
3|2832|Loss: 0.8543151617050171:  44%|████▍     | 2832/6470 [32:36<43:35,  1.39it/s]
3|2833|Loss: 0.7488248944282532:  44%|████▍     | 2832/6470 [32:36<43:35,  1.39it/s]
3|2833|Loss: 0.7488248944282532:  44%|████▍     | 2833/6470 [32:36<41:20,  1.47it/s]
3|2834|Loss: 0.7973222136497498:  44%|████▍     | 2833/6470 [32:37<41:20,  1.47it/s]
3|2834|Loss: 0.7973222136497498:  44%|████▍     | 2834/6470 [32:37<40:27,  1.50it/s]
3|2835|Loss: 0.7667399644851685:  44%|████▍     | 2834/6470 [32:37<40:27,  1.50it/s]
3|2835|Loss: 0.7667399644851685:  44%|████▍     | 2835/6470 [32:38<41:36,  1.46it/s]
3|2836|Loss: 0.9293671250343323:  44%|████▍     | 2835/6470 [32:38<41:36,  1.46it/s]
3|2836|Loss: 0.9293671250343323:  44%|████▍     | 2836/6470 [32:38<39:55,  1.52it/s]
3|2837|Loss: 0.7872173190116882:  44%|████▍     | 2836/6470 [32:39<39:55,  1.52it/s]
3|2837|Loss: 0.7872173190116882:  44%|████▍     | 2837/6470 [32:39<38:54,  1.56it/s]
3|2838|Loss: 0.7892866134643555:  44%|████▍     | 2837/6470 [32:39<38:54,  1.56it/s]
3|2838|Loss: 0.7892866134643555:  44%|████▍     | 2838/6470 [32:39<39:43,  1.52it/s]
3|2839|Loss: 0.7723677754402161:  44%|████▍     | 2838/6470 [32:40<39:43,  1.52it/s]
3|2839|Loss: 0.7723677754402161:  44%|████▍     | 2839/6470 [32:40<42:56,  1.41it/s]
3|2840|Loss: 0.8655960559844971:  44%|████▍     | 2839/6470 [32:41<42:56,  1.41it/s]
3|2840|Loss: 0.8655960559844971:  44%|████▍     | 2840/6470 [32:41<42:41,  1.42it/s]
3|2841|Loss: 0.8644196391105652:  44%|████▍     | 2840/6470 [32:41<42:41,  1.42it/s]
3|2841|Loss: 0.8644196391105652:  44%|████▍     | 2841/6470 [32:42<43:53,  1.38it/s]
3|2842|Loss: 0.7960228323936462:  44%|████▍     | 2841/6470 [32:42<43:53,  1.38it/s]
3|2842|Loss: 0.7960228323936462:  44%|████▍     | 2842/6470 [32:42<40:18,  1.50it/s]
3|2843|Loss: 0.8745623826980591:  44%|████▍     | 2842/6470 [32:43<40:18,  1.50it/s]
3|2843|Loss: 0.8745623826980591:  44%|████▍     | 2843/6470 [32:43<37:25,  1.62it/s]
3|2844|Loss: 0.8472493290901184:  44%|████▍     | 2843/6470 [32:43<37:25,  1.62it/s]
3|2844|Loss: 0.8472493290901184:  44%|████▍     | 2844/6470 [32:43<37:08,  1.63it/s]
3|2845|Loss: 0.8897625803947449:  44%|████▍     | 2844/6470 [32:44<37:08,  1.63it/s]
3|2845|Loss: 0.8897625803947449:  44%|████▍     | 2845/6470 [32:44<36:36,  1.65it/s]
3|2846|Loss: 0.8820732235908508:  44%|████▍     | 2845/6470 [32:44<36:36,  1.65it/s]
3|2846|Loss: 0.8820732235908508:  44%|████▍     | 2846/6470 [32:45<36:38,  1.65it/s]
3|2847|Loss: 0.7312451004981995:  44%|████▍     | 2846/6470 [32:45<36:38,  1.65it/s]
3|2847|Loss: 0.7312451004981995:  44%|████▍     | 2847/6470 [32:45<35:11,  1.72it/s]
3|2848|Loss: 0.9582115411758423:  44%|████▍     | 2847/6470 [32:45<35:11,  1.72it/s]
3|2848|Loss: 0.9582115411758423:  44%|████▍     | 2848/6470 [32:46<36:28,  1.65it/s]
3|2849|Loss: 0.8206067681312561:  44%|████▍     | 2848/6470 [32:46<36:28,  1.65it/s]
3|2849|Loss: 0.8206067681312561:  44%|████▍     | 2849/6470 [32:47<40:45,  1.48it/s]
3|2850|Loss: 0.9752605557441711:  44%|████▍     | 2849/6470 [32:47<40:45,  1.48it/s]
3|2850|Loss: 0.9752605557441711:  44%|████▍     | 2850/6470 [32:47<40:41,  1.48it/s]
3|2851|Loss: 0.7601072192192078:  44%|████▍     | 2850/6470 [32:48<40:41,  1.48it/s]
3|2851|Loss: 0.7601072192192078:  44%|████▍     | 2851/6470 [32:48<39:10,  1.54it/s]
3|2852|Loss: 0.8457521200180054:  44%|████▍     | 2851/6470 [32:48<39:10,  1.54it/s]
3|2852|Loss: 0.8457521200180054:  44%|████▍     | 2852/6470 [32:48<37:52,  1.59it/s]
3|2853|Loss: 0.9646261930465698:  44%|████▍     | 2852/6470 [32:49<37:52,  1.59it/s]
3|2853|Loss: 0.9646261930465698:  44%|████▍     | 2853/6470 [32:49<37:08,  1.62it/s]
3|2854|Loss: 0.9261004328727722:  44%|████▍     | 2853/6470 [32:49<37:08,  1.62it/s]
3|2854|Loss: 0.9261004328727722:  44%|████▍     | 2854/6470 [32:50<40:03,  1.50it/s]
3|2855|Loss: 0.7885612845420837:  44%|████▍     | 2854/6470 [32:50<40:03,  1.50it/s]
3|2855|Loss: 0.7885612845420837:  44%|████▍     | 2855/6470 [32:51<42:21,  1.42it/s]
3|2856|Loss: 0.9594240188598633:  44%|████▍     | 2855/6470 [32:51<42:21,  1.42it/s]
3|2856|Loss: 0.9594240188598633:  44%|████▍     | 2856/6470 [32:51<43:26,  1.39it/s]
3|2857|Loss: 0.7930282950401306:  44%|████▍     | 2856/6470 [32:52<43:26,  1.39it/s]
3|2857|Loss: 0.7930282950401306:  44%|████▍     | 2857/6470 [32:52<45:58,  1.31it/s]
3|2858|Loss: 0.7513250112533569:  44%|████▍     | 2857/6470 [32:53<45:58,  1.31it/s]
3|2858|Loss: 0.7513250112533569:  44%|████▍     | 2858/6470 [32:53<44:26,  1.35it/s]
3|2859|Loss: 0.8420314192771912:  44%|████▍     | 2858/6470 [32:53<44:26,  1.35it/s]
3|2859|Loss: 0.8420314192771912:  44%|████▍     | 2859/6470 [32:54<45:19,  1.33it/s]
3|2860|Loss: 0.8598567843437195:  44%|████▍     | 2859/6470 [32:54<45:19,  1.33it/s]
3|2860|Loss: 0.8598567843437195:  44%|████▍     | 2860/6470 [32:55<47:49,  1.26it/s]
3|2861|Loss: 0.9253575205802917:  44%|████▍     | 2860/6470 [32:55<47:49,  1.26it/s]
3|2861|Loss: 0.9253575205802917:  44%|████▍     | 2861/6470 [32:55<46:55,  1.28it/s]
3|2862|Loss: 0.890616774559021:  44%|████▍     | 2861/6470 [32:56<46:55,  1.28it/s] 
3|2862|Loss: 0.890616774559021:  44%|████▍     | 2862/6470 [32:56<41:51,  1.44it/s]
3|2863|Loss: 0.7983285188674927:  44%|████▍     | 2862/6470 [32:56<41:51,  1.44it/s]
3|2863|Loss: 0.7983285188674927:  44%|████▍     | 2863/6470 [32:56<40:58,  1.47it/s]
3|2864|Loss: 0.8776989579200745:  44%|████▍     | 2863/6470 [32:57<40:58,  1.47it/s]
3|2864|Loss: 0.8776989579200745:  44%|████▍     | 2864/6470 [32:57<42:02,  1.43it/s]
3|2865|Loss: 0.8285220861434937:  44%|████▍     | 2864/6470 [32:58<42:02,  1.43it/s]
3|2865|Loss: 0.8285220861434937:  44%|████▍     | 2865/6470 [32:58<42:18,  1.42it/s]
3|2866|Loss: 1.0461983680725098:  44%|████▍     | 2865/6470 [32:58<42:18,  1.42it/s]
3|2866|Loss: 1.0461983680725098:  44%|████▍     | 2866/6470 [32:59<41:05,  1.46it/s]
3|2867|Loss: 0.9227744936943054:  44%|████▍     | 2866/6470 [32:59<41:05,  1.46it/s]
3|2867|Loss: 0.9227744936943054:  44%|████▍     | 2867/6470 [32:59<40:08,  1.50it/s]
3|2868|Loss: 0.9565660357475281:  44%|████▍     | 2867/6470 [33:00<40:08,  1.50it/s]
3|2868|Loss: 0.9565660357475281:  44%|████▍     | 2868/6470 [33:00<43:01,  1.40it/s]
3|2869|Loss: 0.8190921545028687:  44%|████▍     | 2868/6470 [33:01<43:01,  1.40it/s]
3|2869|Loss: 0.8190921545028687:  44%|████▍     | 2869/6470 [33:01<45:44,  1.31it/s]
3|2870|Loss: 0.8266854882240295:  44%|████▍     | 2869/6470 [33:01<45:44,  1.31it/s]
3|2870|Loss: 0.8266854882240295:  44%|████▍     | 2870/6470 [33:02<45:55,  1.31it/s]
3|2871|Loss: 0.8542670607566833:  44%|████▍     | 2870/6470 [33:02<45:55,  1.31it/s]
3|2871|Loss: 0.8542670607566833:  44%|████▍     | 2871/6470 [33:03<47:37,  1.26it/s]
3|2872|Loss: 0.9281609654426575:  44%|████▍     | 2871/6470 [33:03<47:37,  1.26it/s]
3|2872|Loss: 0.9281609654426575:  44%|████▍     | 2872/6470 [33:03<44:53,  1.34it/s]
3|2873|Loss: 1.122671127319336:  44%|████▍     | 2872/6470 [33:04<44:53,  1.34it/s] 
3|2873|Loss: 1.122671127319336:  44%|████▍     | 2873/6470 [33:04<43:31,  1.38it/s]
3|2874|Loss: 0.8647089004516602:  44%|████▍     | 2873/6470 [33:04<43:31,  1.38it/s]
3|2874|Loss: 0.8647089004516602:  44%|████▍     | 2874/6470 [33:05<44:19,  1.35it/s]
3|2875|Loss: 0.7869260311126709:  44%|████▍     | 2874/6470 [33:05<44:19,  1.35it/s]
3|2875|Loss: 0.7869260311126709:  44%|████▍     | 2875/6470 [33:05<43:58,  1.36it/s]
3|2876|Loss: 0.8163791298866272:  44%|████▍     | 2875/6470 [33:06<43:58,  1.36it/s]
3|2876|Loss: 0.8163791298866272:  44%|████▍     | 2876/6470 [33:06<43:52,  1.37it/s]
3|2877|Loss: 0.9660543203353882:  44%|████▍     | 2876/6470 [33:06<43:52,  1.37it/s]
3|2877|Loss: 0.9660543203353882:  44%|████▍     | 2877/6470 [33:07<42:31,  1.41it/s]
3|2878|Loss: 0.8391427993774414:  44%|████▍     | 2877/6470 [33:07<42:31,  1.41it/s]
3|2878|Loss: 0.8391427993774414:  44%|████▍     | 2878/6470 [33:07<42:36,  1.40it/s]
3|2879|Loss: 0.8226786851882935:  44%|████▍     | 2878/6470 [33:08<42:36,  1.40it/s]
3|2879|Loss: 0.8226786851882935:  44%|████▍     | 2879/6470 [33:08<40:20,  1.48it/s]
3|2880|Loss: 1.0473418235778809:  44%|████▍     | 2879/6470 [33:08<40:20,  1.48it/s]
3|2880|Loss: 1.0473418235778809:  45%|████▍     | 2880/6470 [33:09<42:52,  1.40it/s]
3|2881|Loss: 0.9376446008682251:  45%|████▍     | 2880/6470 [33:09<42:52,  1.40it/s]
3|2881|Loss: 0.9376446008682251:  45%|████▍     | 2881/6470 [33:10<42:28,  1.41it/s]
3|2882|Loss: 0.8469510078430176:  45%|████▍     | 2881/6470 [33:10<42:28,  1.41it/s]
3|2882|Loss: 0.8469510078430176:  45%|████▍     | 2882/6470 [33:10<42:03,  1.42it/s]
3|2883|Loss: 0.9363290667533875:  45%|████▍     | 2882/6470 [33:11<42:03,  1.42it/s]
3|2883|Loss: 0.9363290667533875:  45%|████▍     | 2883/6470 [33:11<40:53,  1.46it/s]
3|2884|Loss: 0.8084354996681213:  45%|████▍     | 2883/6470 [33:11<40:53,  1.46it/s]
3|2884|Loss: 0.8084354996681213:  45%|████▍     | 2884/6470 [33:12<40:47,  1.46it/s]
3|2885|Loss: 0.9936867356300354:  45%|████▍     | 2884/6470 [33:12<40:47,  1.46it/s]
3|2885|Loss: 0.9936867356300354:  45%|████▍     | 2885/6470 [33:12<41:05,  1.45it/s]
3|2886|Loss: 0.9924103617668152:  45%|████▍     | 2885/6470 [33:13<41:05,  1.45it/s]
3|2886|Loss: 0.9924103617668152:  45%|████▍     | 2886/6470 [33:13<39:07,  1.53it/s]
3|2887|Loss: 0.8525446057319641:  45%|████▍     | 2886/6470 [33:13<39:07,  1.53it/s]
3|2887|Loss: 0.8525446057319641:  45%|████▍     | 2887/6470 [33:14<40:10,  1.49it/s]
3|2888|Loss: 0.7971373796463013:  45%|████▍     | 2887/6470 [33:14<40:10,  1.49it/s]
3|2888|Loss: 0.7971373796463013:  45%|████▍     | 2888/6470 [33:14<40:01,  1.49it/s]
3|2889|Loss: 0.8742678761482239:  45%|████▍     | 2888/6470 [33:15<40:01,  1.49it/s]
3|2889|Loss: 0.8742678761482239:  45%|████▍     | 2889/6470 [33:15<44:33,  1.34it/s]
3|2890|Loss: 0.9683650135993958:  45%|████▍     | 2889/6470 [33:16<44:33,  1.34it/s]
3|2890|Loss: 0.9683650135993958:  45%|████▍     | 2890/6470 [33:16<43:43,  1.36it/s]
3|2891|Loss: 0.898342490196228:  45%|████▍     | 2890/6470 [33:16<43:43,  1.36it/s] 
3|2891|Loss: 0.898342490196228:  45%|████▍     | 2891/6470 [33:17<44:23,  1.34it/s]
3|2892|Loss: 0.9281541109085083:  45%|████▍     | 2891/6470 [33:17<44:23,  1.34it/s]
3|2892|Loss: 0.9281541109085083:  45%|████▍     | 2892/6470 [33:17<45:55,  1.30it/s]
3|2893|Loss: 0.8337526321411133:  45%|████▍     | 2892/6470 [33:18<45:55,  1.30it/s]
3|2893|Loss: 0.8337526321411133:  45%|████▍     | 2893/6470 [33:18<45:16,  1.32it/s]
3|2894|Loss: 0.7903352379798889:  45%|████▍     | 2893/6470 [33:19<45:16,  1.32it/s]
3|2894|Loss: 0.7903352379798889:  45%|████▍     | 2894/6470 [33:19<42:01,  1.42it/s]
3|2895|Loss: 0.844247579574585:  45%|████▍     | 2894/6470 [33:19<42:01,  1.42it/s] 
3|2895|Loss: 0.844247579574585:  45%|████▍     | 2895/6470 [33:19<39:46,  1.50it/s]
3|2896|Loss: 0.9098861217498779:  45%|████▍     | 2895/6470 [33:20<39:46,  1.50it/s]
3|2896|Loss: 0.9098861217498779:  45%|████▍     | 2896/6470 [33:20<40:49,  1.46it/s]
3|2897|Loss: 0.927291989326477:  45%|████▍     | 2896/6470 [33:20<40:49,  1.46it/s] 
3|2897|Loss: 0.927291989326477:  45%|████▍     | 2897/6470 [33:21<38:36,  1.54it/s]
3|2898|Loss: 0.8626977801322937:  45%|████▍     | 2897/6470 [33:21<38:36,  1.54it/s]
3|2898|Loss: 0.8626977801322937:  45%|████▍     | 2898/6470 [33:21<37:39,  1.58it/s]
3|2899|Loss: 0.9586166739463806:  45%|████▍     | 2898/6470 [33:22<37:39,  1.58it/s]
3|2899|Loss: 0.9586166739463806:  45%|████▍     | 2899/6470 [33:22<37:21,  1.59it/s]
3|2900|Loss: 0.889557421207428:  45%|████▍     | 2899/6470 [33:22<37:21,  1.59it/s] 
3|2900|Loss: 0.889557421207428:  45%|████▍     | 2900/6470 [33:22<37:16,  1.60it/s]
3|2901|Loss: 0.7917531728744507:  45%|████▍     | 2900/6470 [33:23<37:16,  1.60it/s]
3|2901|Loss: 0.7917531728744507:  45%|████▍     | 2901/6470 [33:23<33:38,  1.77it/s]
3|2902|Loss: 0.812856137752533:  45%|████▍     | 2901/6470 [33:23<33:38,  1.77it/s] 
3|2902|Loss: 0.812856137752533:  45%|████▍     | 2902/6470 [33:24<35:10,  1.69it/s]
3|2903|Loss: 0.8759084343910217:  45%|████▍     | 2902/6470 [33:24<35:10,  1.69it/s]
3|2903|Loss: 0.8759084343910217:  45%|████▍     | 2903/6470 [33:24<37:57,  1.57it/s]
3|2904|Loss: 0.8419428467750549:  45%|████▍     | 2903/6470 [33:25<37:57,  1.57it/s]
3|2904|Loss: 0.8419428467750549:  45%|████▍     | 2904/6470 [33:25<39:51,  1.49it/s]
3|2905|Loss: 0.884791374206543:  45%|████▍     | 2904/6470 [33:25<39:51,  1.49it/s] 
3|2905|Loss: 0.884791374206543:  45%|████▍     | 2905/6470 [33:26<40:34,  1.46it/s]
3|2906|Loss: 0.7770673036575317:  45%|████▍     | 2905/6470 [33:26<40:34,  1.46it/s]
3|2906|Loss: 0.7770673036575317:  45%|████▍     | 2906/6470 [33:27<43:30,  1.37it/s]
3|2907|Loss: 0.9159072637557983:  45%|████▍     | 2906/6470 [33:27<43:30,  1.37it/s]
3|2907|Loss: 0.9159072637557983:  45%|████▍     | 2907/6470 [33:27<44:22,  1.34it/s]
3|2908|Loss: 0.8648954033851624:  45%|████▍     | 2907/6470 [33:28<44:22,  1.34it/s]
3|2908|Loss: 0.8648954033851624:  45%|████▍     | 2908/6470 [33:28<41:03,  1.45it/s]
3|2909|Loss: 0.8431006073951721:  45%|████▍     | 2908/6470 [33:28<41:03,  1.45it/s]
3|2909|Loss: 0.8431006073951721:  45%|████▍     | 2909/6470 [33:29<39:38,  1.50it/s]
3|2910|Loss: 0.7485558390617371:  45%|████▍     | 2909/6470 [33:29<39:38,  1.50it/s]
3|2910|Loss: 0.7485558390617371:  45%|████▍     | 2910/6470 [33:29<40:32,  1.46it/s]
3|2911|Loss: 0.8557828664779663:  45%|████▍     | 2910/6470 [33:30<40:32,  1.46it/s]
3|2911|Loss: 0.8557828664779663:  45%|████▍     | 2911/6470 [33:30<43:08,  1.37it/s]
3|2912|Loss: 0.777521014213562:  45%|████▍     | 2911/6470 [33:31<43:08,  1.37it/s] 
3|2912|Loss: 0.777521014213562:  45%|████▌     | 2912/6470 [33:31<42:06,  1.41it/s]
3|2913|Loss: 0.9041087627410889:  45%|████▌     | 2912/6470 [33:31<42:06,  1.41it/s]
3|2913|Loss: 0.9041087627410889:  45%|████▌     | 2913/6470 [33:31<39:23,  1.50it/s]
3|2914|Loss: 0.9725551605224609:  45%|████▌     | 2913/6470 [33:32<39:23,  1.50it/s]
3|2914|Loss: 0.9725551605224609:  45%|████▌     | 2914/6470 [33:32<38:40,  1.53it/s]
3|2915|Loss: 1.0113215446472168:  45%|████▌     | 2914/6470 [33:32<38:40,  1.53it/s]
3|2915|Loss: 1.0113215446472168:  45%|████▌     | 2915/6470 [33:33<43:26,  1.36it/s]
3|2916|Loss: 0.6442250609397888:  45%|████▌     | 2915/6470 [33:33<43:26,  1.36it/s]
3|2916|Loss: 0.6442250609397888:  45%|████▌     | 2916/6470 [33:34<47:49,  1.24it/s]
3|2917|Loss: 0.8821330666542053:  45%|████▌     | 2916/6470 [33:34<47:49,  1.24it/s]
3|2917|Loss: 0.8821330666542053:  45%|████▌     | 2917/6470 [33:35<48:22,  1.22it/s]
3|2918|Loss: 0.7542725205421448:  45%|████▌     | 2917/6470 [33:35<48:22,  1.22it/s]
3|2918|Loss: 0.7542725205421448:  45%|████▌     | 2918/6470 [33:36<48:21,  1.22it/s]
3|2919|Loss: 0.9435160756111145:  45%|████▌     | 2918/6470 [33:36<48:21,  1.22it/s]
3|2919|Loss: 0.9435160756111145:  45%|████▌     | 2919/6470 [33:36<45:59,  1.29it/s]
3|2920|Loss: 0.7767220139503479:  45%|████▌     | 2919/6470 [33:37<45:59,  1.29it/s]
3|2920|Loss: 0.7767220139503479:  45%|████▌     | 2920/6470 [33:37<44:51,  1.32it/s]
3|2921|Loss: 0.8797860145568848:  45%|████▌     | 2920/6470 [33:37<44:51,  1.32it/s]
3|2921|Loss: 0.8797860145568848:  45%|████▌     | 2921/6470 [33:38<44:02,  1.34it/s]
3|2922|Loss: 0.9419002532958984:  45%|████▌     | 2921/6470 [33:38<44:02,  1.34it/s]
3|2922|Loss: 0.9419002532958984:  45%|████▌     | 2922/6470 [33:38<42:27,  1.39it/s]
3|2923|Loss: 0.8259556889533997:  45%|████▌     | 2922/6470 [33:39<42:27,  1.39it/s]
3|2923|Loss: 0.8259556889533997:  45%|████▌     | 2923/6470 [33:39<41:53,  1.41it/s]
3|2924|Loss: 0.8224682211875916:  45%|████▌     | 2923/6470 [33:39<41:53,  1.41it/s]
3|2924|Loss: 0.8224682211875916:  45%|████▌     | 2924/6470 [33:39<36:56,  1.60it/s]
3|2925|Loss: 0.7763967514038086:  45%|████▌     | 2924/6470 [33:40<36:56,  1.60it/s]
3|2925|Loss: 0.7763967514038086:  45%|████▌     | 2925/6470 [33:40<36:02,  1.64it/s]
3|2926|Loss: 0.7341484427452087:  45%|████▌     | 2925/6470 [33:40<36:02,  1.64it/s]
3|2926|Loss: 0.7341484427452087:  45%|████▌     | 2926/6470 [33:41<38:49,  1.52it/s]
3|2927|Loss: 0.9597045183181763:  45%|████▌     | 2926/6470 [33:41<38:49,  1.52it/s]
3|2927|Loss: 0.9597045183181763:  45%|████▌     | 2927/6470 [33:41<39:57,  1.48it/s]
3|2928|Loss: 0.7600235342979431:  45%|████▌     | 2927/6470 [33:42<39:57,  1.48it/s]
3|2928|Loss: 0.7600235342979431:  45%|████▌     | 2928/6470 [33:42<38:34,  1.53it/s]
3|2929|Loss: 0.814603865146637:  45%|████▌     | 2928/6470 [33:42<38:34,  1.53it/s] 
3|2929|Loss: 0.814603865146637:  45%|████▌     | 2929/6470 [33:43<39:47,  1.48it/s]
3|2930|Loss: 0.9299100041389465:  45%|████▌     | 2929/6470 [33:43<39:47,  1.48it/s]
3|2930|Loss: 0.9299100041389465:  45%|████▌     | 2930/6470 [33:44<42:19,  1.39it/s]
3|2931|Loss: 0.9804348945617676:  45%|████▌     | 2930/6470 [33:44<42:19,  1.39it/s]
3|2931|Loss: 0.9804348945617676:  45%|████▌     | 2931/6470 [33:44<42:30,  1.39it/s]
3|2932|Loss: 0.8668850660324097:  45%|████▌     | 2931/6470 [33:45<42:30,  1.39it/s]
3|2932|Loss: 0.8668850660324097:  45%|████▌     | 2932/6470 [33:45<39:56,  1.48it/s]
3|2933|Loss: 0.9731418490409851:  45%|████▌     | 2932/6470 [33:45<39:56,  1.48it/s]
3|2933|Loss: 0.9731418490409851:  45%|████▌     | 2933/6470 [33:46<39:58,  1.47it/s]
3|2934|Loss: 0.9003469347953796:  45%|████▌     | 2933/6470 [33:46<39:58,  1.47it/s]
3|2934|Loss: 0.9003469347953796:  45%|████▌     | 2934/6470 [33:46<40:49,  1.44it/s]
3|2935|Loss: 0.8935950994491577:  45%|████▌     | 2934/6470 [33:47<40:49,  1.44it/s]
3|2935|Loss: 0.8935950994491577:  45%|████▌     | 2935/6470 [33:47<41:33,  1.42it/s]
3|2936|Loss: 0.8388234972953796:  45%|████▌     | 2935/6470 [33:47<41:33,  1.42it/s]
3|2936|Loss: 0.8388234972953796:  45%|████▌     | 2936/6470 [33:48<38:50,  1.52it/s]
3|2937|Loss: 0.9110958576202393:  45%|████▌     | 2936/6470 [33:48<38:50,  1.52it/s]
3|2937|Loss: 0.9110958576202393:  45%|████▌     | 2937/6470 [33:48<40:39,  1.45it/s]
3|2938|Loss: 0.9622653126716614:  45%|████▌     | 2937/6470 [33:49<40:39,  1.45it/s]
3|2938|Loss: 0.9622653126716614:  45%|████▌     | 2938/6470 [33:49<41:13,  1.43it/s]
3|2939|Loss: 0.9033352732658386:  45%|████▌     | 2938/6470 [33:50<41:13,  1.43it/s]
3|2939|Loss: 0.9033352732658386:  45%|████▌     | 2939/6470 [33:50<43:29,  1.35it/s]
3|2940|Loss: 0.9451794028282166:  45%|████▌     | 2939/6470 [33:50<43:29,  1.35it/s]
3|2940|Loss: 0.9451794028282166:  45%|████▌     | 2940/6470 [33:51<44:55,  1.31it/s]
3|2941|Loss: 0.8462644815444946:  45%|████▌     | 2940/6470 [33:51<44:55,  1.31it/s]
3|2941|Loss: 0.8462644815444946:  45%|████▌     | 2941/6470 [33:52<44:59,  1.31it/s]
3|2942|Loss: 0.8579679727554321:  45%|████▌     | 2941/6470 [33:52<44:59,  1.31it/s]
3|2942|Loss: 0.8579679727554321:  45%|████▌     | 2942/6470 [33:52<44:15,  1.33it/s]
3|2943|Loss: 0.8641120791435242:  45%|████▌     | 2942/6470 [33:53<44:15,  1.33it/s]
3|2943|Loss: 0.8641120791435242:  45%|████▌     | 2943/6470 [33:53<44:32,  1.32it/s]
3|2944|Loss: 0.8471123576164246:  45%|████▌     | 2943/6470 [33:53<44:32,  1.32it/s]
3|2944|Loss: 0.8471123576164246:  46%|████▌     | 2944/6470 [33:54<42:53,  1.37it/s]
3|2945|Loss: 0.7736192345619202:  46%|████▌     | 2944/6470 [33:54<42:53,  1.37it/s]
3|2945|Loss: 0.7736192345619202:  46%|████▌     | 2945/6470 [33:54<40:53,  1.44it/s]
3|2946|Loss: 0.8401625752449036:  46%|████▌     | 2945/6470 [33:55<40:53,  1.44it/s]
3|2946|Loss: 0.8401625752449036:  46%|████▌     | 2946/6470 [33:55<40:44,  1.44it/s]
3|2947|Loss: 0.8677654266357422:  46%|████▌     | 2946/6470 [33:55<40:44,  1.44it/s]
3|2947|Loss: 0.8677654266357422:  46%|████▌     | 2947/6470 [33:56<38:07,  1.54it/s]
3|2948|Loss: 0.9464696645736694:  46%|████▌     | 2947/6470 [33:56<38:07,  1.54it/s]
3|2948|Loss: 0.9464696645736694:  46%|████▌     | 2948/6470 [33:56<39:46,  1.48it/s]
3|2949|Loss: 0.7731738090515137:  46%|████▌     | 2948/6470 [33:57<39:46,  1.48it/s]
3|2949|Loss: 0.7731738090515137:  46%|████▌     | 2949/6470 [33:57<41:17,  1.42it/s]
3|2950|Loss: 0.8951804637908936:  46%|████▌     | 2949/6470 [33:57<41:17,  1.42it/s]
3|2950|Loss: 0.8951804637908936:  46%|████▌     | 2950/6470 [33:58<40:41,  1.44it/s]
3|2951|Loss: 0.8537556529045105:  46%|████▌     | 2950/6470 [33:58<40:41,  1.44it/s]
3|2951|Loss: 0.8537556529045105:  46%|████▌     | 2951/6470 [33:58<40:32,  1.45it/s]
3|2952|Loss: 0.7549788951873779:  46%|████▌     | 2951/6470 [33:59<40:32,  1.45it/s]
3|2952|Loss: 0.7549788951873779:  46%|████▌     | 2952/6470 [33:59<40:42,  1.44it/s]
3|2953|Loss: 0.7789428234100342:  46%|████▌     | 2952/6470 [33:59<40:42,  1.44it/s]
3|2953|Loss: 0.7789428234100342:  46%|████▌     | 2953/6470 [34:00<41:21,  1.42it/s]
3|2954|Loss: 0.7508912086486816:  46%|████▌     | 2953/6470 [34:00<41:21,  1.42it/s]
3|2954|Loss: 0.7508912086486816:  46%|████▌     | 2954/6470 [34:01<44:26,  1.32it/s]
3|2955|Loss: 1.0614551305770874:  46%|████▌     | 2954/6470 [34:01<44:26,  1.32it/s]
3|2955|Loss: 1.0614551305770874:  46%|████▌     | 2955/6470 [34:02<45:42,  1.28it/s]
3|2956|Loss: 0.9023354649543762:  46%|████▌     | 2955/6470 [34:02<45:42,  1.28it/s]
3|2956|Loss: 0.9023354649543762:  46%|████▌     | 2956/6470 [34:02<44:46,  1.31it/s]
3|2957|Loss: 0.8955931663513184:  46%|████▌     | 2956/6470 [34:03<44:46,  1.31it/s]
3|2957|Loss: 0.8955931663513184:  46%|████▌     | 2957/6470 [34:03<45:49,  1.28it/s]
3|2958|Loss: 0.9349531531333923:  46%|████▌     | 2957/6470 [34:04<45:49,  1.28it/s]
3|2958|Loss: 0.9349531531333923:  46%|████▌     | 2958/6470 [34:04<45:00,  1.30it/s]
3|2959|Loss: 0.7123480439186096:  46%|████▌     | 2958/6470 [34:04<45:00,  1.30it/s]
3|2959|Loss: 0.7123480439186096:  46%|████▌     | 2959/6470 [34:05<44:28,  1.32it/s]
3|2960|Loss: 0.8541079759597778:  46%|████▌     | 2959/6470 [34:05<44:28,  1.32it/s]
3|2960|Loss: 0.8541079759597778:  46%|████▌     | 2960/6470 [34:05<44:05,  1.33it/s]
3|2961|Loss: 0.793632984161377:  46%|████▌     | 2960/6470 [34:06<44:05,  1.33it/s] 
3|2961|Loss: 0.793632984161377:  46%|████▌     | 2961/6470 [34:06<39:47,  1.47it/s]
3|2962|Loss: 0.8142575621604919:  46%|████▌     | 2961/6470 [34:06<39:47,  1.47it/s]
3|2962|Loss: 0.8142575621604919:  46%|████▌     | 2962/6470 [34:07<40:29,  1.44it/s]
3|2963|Loss: 0.8270503878593445:  46%|████▌     | 2962/6470 [34:07<40:29,  1.44it/s]
3|2963|Loss: 0.8270503878593445:  46%|████▌     | 2963/6470 [34:07<40:17,  1.45it/s]
3|2964|Loss: 0.8724398016929626:  46%|████▌     | 2963/6470 [34:08<40:17,  1.45it/s]
3|2964|Loss: 0.8724398016929626:  46%|████▌     | 2964/6470 [34:08<39:29,  1.48it/s]
3|2965|Loss: 0.9104306101799011:  46%|████▌     | 2964/6470 [34:08<39:29,  1.48it/s]
3|2965|Loss: 0.9104306101799011:  46%|████▌     | 2965/6470 [34:08<38:47,  1.51it/s]
3|2966|Loss: 0.8552872538566589:  46%|████▌     | 2965/6470 [34:09<38:47,  1.51it/s]
3|2966|Loss: 0.8552872538566589:  46%|████▌     | 2966/6470 [34:09<39:18,  1.49it/s]
3|2967|Loss: 0.7740065455436707:  46%|████▌     | 2966/6470 [34:10<39:18,  1.49it/s]
3|2967|Loss: 0.7740065455436707:  46%|████▌     | 2967/6470 [34:10<39:02,  1.50it/s]
3|2968|Loss: 0.726374626159668:  46%|████▌     | 2967/6470 [34:10<39:02,  1.50it/s] 
3|2968|Loss: 0.726374626159668:  46%|████▌     | 2968/6470 [34:10<36:40,  1.59it/s]
3|2969|Loss: 1.0056192874908447:  46%|████▌     | 2968/6470 [34:11<36:40,  1.59it/s]
3|2969|Loss: 1.0056192874908447:  46%|████▌     | 2969/6470 [34:11<38:42,  1.51it/s]
3|2970|Loss: 0.7673203349113464:  46%|████▌     | 2969/6470 [34:12<38:42,  1.51it/s]
3|2970|Loss: 0.7673203349113464:  46%|████▌     | 2970/6470 [34:12<40:13,  1.45it/s]
3|2971|Loss: 0.7648879289627075:  46%|████▌     | 2970/6470 [34:12<40:13,  1.45it/s]
3|2971|Loss: 0.7648879289627075:  46%|████▌     | 2971/6470 [34:12<38:04,  1.53it/s]
3|2972|Loss: 0.7876851558685303:  46%|████▌     | 2971/6470 [34:13<38:04,  1.53it/s]
3|2972|Loss: 0.7876851558685303:  46%|████▌     | 2972/6470 [34:13<35:38,  1.64it/s]
3|2973|Loss: 0.9333875179290771:  46%|████▌     | 2972/6470 [34:13<35:38,  1.64it/s]
3|2973|Loss: 0.9333875179290771:  46%|████▌     | 2973/6470 [34:14<36:30,  1.60it/s]
3|2974|Loss: 0.8702770471572876:  46%|████▌     | 2973/6470 [34:14<36:30,  1.60it/s]
3|2974|Loss: 0.8702770471572876:  46%|████▌     | 2974/6470 [34:14<35:08,  1.66it/s]
3|2975|Loss: 0.9248163104057312:  46%|████▌     | 2974/6470 [34:15<35:08,  1.66it/s]
3|2975|Loss: 0.9248163104057312:  46%|████▌     | 2975/6470 [34:15<36:05,  1.61it/s]
3|2976|Loss: 0.7712212800979614:  46%|████▌     | 2975/6470 [34:15<36:05,  1.61it/s]
3|2976|Loss: 0.7712212800979614:  46%|████▌     | 2976/6470 [34:15<34:56,  1.67it/s]
3|2977|Loss: 0.7207625508308411:  46%|████▌     | 2976/6470 [34:16<34:56,  1.67it/s]
3|2977|Loss: 0.7207625508308411:  46%|████▌     | 2977/6470 [34:16<34:24,  1.69it/s]
3|2978|Loss: 0.8930297493934631:  46%|████▌     | 2977/6470 [34:16<34:24,  1.69it/s]
3|2978|Loss: 0.8930297493934631:  46%|████▌     | 2978/6470 [34:17<36:02,  1.62it/s]
3|2979|Loss: 0.9965410828590393:  46%|████▌     | 2978/6470 [34:17<36:02,  1.62it/s]
3|2979|Loss: 0.9965410828590393:  46%|████▌     | 2979/6470 [34:17<35:52,  1.62it/s]
3|2980|Loss: 0.9781465530395508:  46%|████▌     | 2979/6470 [34:18<35:52,  1.62it/s]
3|2980|Loss: 0.9781465530395508:  46%|████▌     | 2980/6470 [34:18<33:24,  1.74it/s]
3|2981|Loss: 0.7734205722808838:  46%|████▌     | 2980/6470 [34:18<33:24,  1.74it/s]
3|2981|Loss: 0.7734205722808838:  46%|████▌     | 2981/6470 [34:18<34:35,  1.68it/s]
3|2982|Loss: 0.8372793793678284:  46%|████▌     | 2981/6470 [34:19<34:35,  1.68it/s]
3|2982|Loss: 0.8372793793678284:  46%|████▌     | 2982/6470 [34:19<36:54,  1.58it/s]
3|2983|Loss: 0.692768931388855:  46%|████▌     | 2982/6470 [34:19<36:54,  1.58it/s] 
3|2983|Loss: 0.692768931388855:  46%|████▌     | 2983/6470 [34:20<38:42,  1.50it/s]
3|2984|Loss: 0.8829907178878784:  46%|████▌     | 2983/6470 [34:20<38:42,  1.50it/s]
3|2984|Loss: 0.8829907178878784:  46%|████▌     | 2984/6470 [34:21<39:27,  1.47it/s]
3|2985|Loss: 0.8364666700363159:  46%|████▌     | 2984/6470 [34:21<39:27,  1.47it/s]
3|2985|Loss: 0.8364666700363159:  46%|████▌     | 2985/6470 [34:21<41:57,  1.38it/s]
3|2986|Loss: 0.9134799242019653:  46%|████▌     | 2985/6470 [34:22<41:57,  1.38it/s]
3|2986|Loss: 0.9134799242019653:  46%|████▌     | 2986/6470 [34:22<44:19,  1.31it/s]
3|2987|Loss: 0.986356794834137:  46%|████▌     | 2986/6470 [34:23<44:19,  1.31it/s] 
3|2987|Loss: 0.986356794834137:  46%|████▌     | 2987/6470 [34:23<40:28,  1.43it/s]
3|2988|Loss: 0.9062347412109375:  46%|████▌     | 2987/6470 [34:23<40:28,  1.43it/s]
3|2988|Loss: 0.9062347412109375:  46%|████▌     | 2988/6470 [34:23<39:43,  1.46it/s]
3|2989|Loss: 0.742920458316803:  46%|████▌     | 2988/6470 [34:24<39:43,  1.46it/s] 
3|2989|Loss: 0.742920458316803:  46%|████▌     | 2989/6470 [34:24<39:16,  1.48it/s]
3|2990|Loss: 0.715364396572113:  46%|████▌     | 2989/6470 [34:24<39:16,  1.48it/s]
3|2990|Loss: 0.715364396572113:  46%|████▌     | 2990/6470 [34:25<37:42,  1.54it/s]
3|2991|Loss: 0.9230495095252991:  46%|████▌     | 2990/6470 [34:25<37:42,  1.54it/s]
3|2991|Loss: 0.9230495095252991:  46%|████▌     | 2991/6470 [34:25<38:38,  1.50it/s]
3|2992|Loss: 0.7090430855751038:  46%|████▌     | 2991/6470 [34:26<38:38,  1.50it/s]
3|2992|Loss: 0.7090430855751038:  46%|████▌     | 2992/6470 [34:26<36:41,  1.58it/s]
3|2993|Loss: 1.0331673622131348:  46%|████▌     | 2992/6470 [34:26<36:41,  1.58it/s]
3|2993|Loss: 1.0331673622131348:  46%|████▋     | 2993/6470 [34:26<35:39,  1.63it/s]
3|2994|Loss: 0.6796098947525024:  46%|████▋     | 2993/6470 [34:27<35:39,  1.63it/s]
3|2994|Loss: 0.6796098947525024:  46%|████▋     | 2994/6470 [34:27<35:52,  1.61it/s]
3|2995|Loss: 0.7736173868179321:  46%|████▋     | 2994/6470 [34:28<35:52,  1.61it/s]
3|2995|Loss: 0.7736173868179321:  46%|████▋     | 2995/6470 [34:28<37:11,  1.56it/s]
3|2996|Loss: 0.9989365935325623:  46%|████▋     | 2995/6470 [34:28<37:11,  1.56it/s]
3|2996|Loss: 0.9989365935325623:  46%|████▋     | 2996/6470 [34:29<40:48,  1.42it/s]
3|2997|Loss: 0.7928171157836914:  46%|████▋     | 2996/6470 [34:29<40:48,  1.42it/s]
3|2997|Loss: 0.7928171157836914:  46%|████▋     | 2997/6470 [34:29<42:01,  1.38it/s]
3|2998|Loss: 0.8299424648284912:  46%|████▋     | 2997/6470 [34:30<42:01,  1.38it/s]
3|2998|Loss: 0.8299424648284912:  46%|████▋     | 2998/6470 [34:30<44:36,  1.30it/s]
3|2999|Loss: 0.7753098011016846:  46%|████▋     | 2998/6470 [34:31<44:36,  1.30it/s]
3|2999|Loss: 0.7753098011016846:  46%|████▋     | 2999/6470 [34:31<46:30,  1.24it/s]
3|3000|Loss: 0.5680808424949646:  46%|████▋     | 2999/6470 [34:32<46:30,  1.24it/s]
3|3000|Loss: 0.5680808424949646:  46%|████▋     | 3000/6470 [34:32<45:44,  1.26it/s]
3|3001|Loss: 0.9062720537185669:  46%|████▋     | 3000/6470 [34:32<45:44,  1.26it/s]
3|3001|Loss: 0.9062720537185669:  46%|████▋     | 3001/6470 [34:33<45:01,  1.28it/s]
3|3002|Loss: 0.8578260540962219:  46%|████▋     | 3001/6470 [34:33<45:01,  1.28it/s]
3|3002|Loss: 0.8578260540962219:  46%|████▋     | 3002/6470 [34:33<42:49,  1.35it/s]
3|3003|Loss: 0.9101721048355103:  46%|████▋     | 3002/6470 [34:34<42:49,  1.35it/s]
3|3003|Loss: 0.9101721048355103:  46%|████▋     | 3003/6470 [34:34<43:15,  1.34it/s]
3|3004|Loss: 1.0470534563064575:  46%|████▋     | 3003/6470 [34:35<43:15,  1.34it/s]
3|3004|Loss: 1.0470534563064575:  46%|████▋     | 3004/6470 [34:35<44:54,  1.29it/s]
3|3005|Loss: 0.8281137943267822:  46%|████▋     | 3004/6470 [34:35<44:54,  1.29it/s]
3|3005|Loss: 0.8281137943267822:  46%|████▋     | 3005/6470 [34:36<41:52,  1.38it/s]
3|3006|Loss: 0.8045838475227356:  46%|████▋     | 3005/6470 [34:36<41:52,  1.38it/s]
3|3006|Loss: 0.8045838475227356:  46%|████▋     | 3006/6470 [34:36<42:09,  1.37it/s]
3|3007|Loss: 0.7488406300544739:  46%|████▋     | 3006/6470 [34:37<42:09,  1.37it/s]
3|3007|Loss: 0.7488406300544739:  46%|████▋     | 3007/6470 [34:37<43:15,  1.33it/s]
3|3008|Loss: 0.906668484210968:  46%|████▋     | 3007/6470 [34:38<43:15,  1.33it/s] 
3|3008|Loss: 0.906668484210968:  46%|████▋     | 3008/6470 [34:38<43:34,  1.32it/s]
3|3009|Loss: 0.9191778898239136:  46%|████▋     | 3008/6470 [34:38<43:34,  1.32it/s]
3|3009|Loss: 0.9191778898239136:  47%|████▋     | 3009/6470 [34:39<41:46,  1.38it/s]
3|3010|Loss: 0.6781768202781677:  47%|████▋     | 3009/6470 [34:39<41:46,  1.38it/s]
3|3010|Loss: 0.6781768202781677:  47%|████▋     | 3010/6470 [34:39<41:43,  1.38it/s]
3|3011|Loss: 0.9354141354560852:  47%|████▋     | 3010/6470 [34:40<41:43,  1.38it/s]
3|3011|Loss: 0.9354141354560852:  47%|████▋     | 3011/6470 [34:40<42:06,  1.37it/s]
3|3012|Loss: 0.8289097547531128:  47%|████▋     | 3011/6470 [34:40<42:06,  1.37it/s]
3|3012|Loss: 0.8289097547531128:  47%|████▋     | 3012/6470 [34:41<40:07,  1.44it/s]
3|3013|Loss: 0.8827027082443237:  47%|████▋     | 3012/6470 [34:41<40:07,  1.44it/s]
3|3013|Loss: 0.8827027082443237:  47%|████▋     | 3013/6470 [34:41<41:26,  1.39it/s]
3|3014|Loss: 0.7289531230926514:  47%|████▋     | 3013/6470 [34:42<41:26,  1.39it/s]
3|3014|Loss: 0.7289531230926514:  47%|████▋     | 3014/6470 [34:42<40:40,  1.42it/s]
3|3015|Loss: 0.7635464668273926:  47%|████▋     | 3014/6470 [34:42<40:40,  1.42it/s]
3|3015|Loss: 0.7635464668273926:  47%|████▋     | 3015/6470 [34:43<39:13,  1.47it/s]
3|3016|Loss: 1.0148732662200928:  47%|████▋     | 3015/6470 [34:43<39:13,  1.47it/s]
3|3016|Loss: 1.0148732662200928:  47%|████▋     | 3016/6470 [34:43<36:41,  1.57it/s]
3|3017|Loss: 0.8116222620010376:  47%|████▋     | 3016/6470 [34:44<36:41,  1.57it/s]
3|3017|Loss: 0.8116222620010376:  47%|████▋     | 3017/6470 [34:44<37:43,  1.53it/s]
3|3018|Loss: 0.9505084156990051:  47%|████▋     | 3017/6470 [34:44<37:43,  1.53it/s]
3|3018|Loss: 0.9505084156990051:  47%|████▋     | 3018/6470 [34:45<37:42,  1.53it/s]
3|3019|Loss: 0.8117003440856934:  47%|████▋     | 3018/6470 [34:45<37:42,  1.53it/s]
3|3019|Loss: 0.8117003440856934:  47%|████▋     | 3019/6470 [34:45<37:01,  1.55it/s]
3|3020|Loss: 0.7673231959342957:  47%|████▋     | 3019/6470 [34:46<37:01,  1.55it/s]
3|3020|Loss: 0.7673231959342957:  47%|████▋     | 3020/6470 [34:46<34:19,  1.67it/s]
3|3021|Loss: 0.9376987218856812:  47%|████▋     | 3020/6470 [34:46<34:19,  1.67it/s]
3|3021|Loss: 0.9376987218856812:  47%|████▋     | 3021/6470 [34:46<33:01,  1.74it/s]
3|3022|Loss: 0.7182071805000305:  47%|████▋     | 3021/6470 [34:47<33:01,  1.74it/s]
3|3022|Loss: 0.7182071805000305:  47%|████▋     | 3022/6470 [34:47<33:49,  1.70it/s]
3|3023|Loss: 0.8483508825302124:  47%|████▋     | 3022/6470 [34:47<33:49,  1.70it/s]
3|3023|Loss: 0.8483508825302124:  47%|████▋     | 3023/6470 [34:47<34:02,  1.69it/s]
3|3024|Loss: 1.2413506507873535:  47%|████▋     | 3023/6470 [34:48<34:02,  1.69it/s]
3|3024|Loss: 1.2413506507873535:  47%|████▋     | 3024/6470 [34:48<37:35,  1.53it/s]
3|3025|Loss: 0.8663982152938843:  47%|████▋     | 3024/6470 [34:49<37:35,  1.53it/s]
3|3025|Loss: 0.8663982152938843:  47%|████▋     | 3025/6470 [34:49<37:13,  1.54it/s]
3|3026|Loss: 0.7889713644981384:  47%|████▋     | 3025/6470 [34:49<37:13,  1.54it/s]
3|3026|Loss: 0.7889713644981384:  47%|████▋     | 3026/6470 [34:50<39:10,  1.47it/s]
3|3027|Loss: 0.8034995794296265:  47%|████▋     | 3026/6470 [34:50<39:10,  1.47it/s]
3|3027|Loss: 0.8034995794296265:  47%|████▋     | 3027/6470 [34:50<38:11,  1.50it/s]
3|3028|Loss: 0.749180257320404:  47%|████▋     | 3027/6470 [34:51<38:11,  1.50it/s] 
3|3028|Loss: 0.749180257320404:  47%|████▋     | 3028/6470 [34:51<39:02,  1.47it/s]
3|3029|Loss: 0.9587975740432739:  47%|████▋     | 3028/6470 [34:51<39:02,  1.47it/s]
3|3029|Loss: 0.9587975740432739:  47%|████▋     | 3029/6470 [34:52<39:44,  1.44it/s]
3|3030|Loss: 0.8965593576431274:  47%|████▋     | 3029/6470 [34:52<39:44,  1.44it/s]
3|3030|Loss: 0.8965593576431274:  47%|████▋     | 3030/6470 [34:52<41:13,  1.39it/s]
3|3031|Loss: 0.8903886079788208:  47%|████▋     | 3030/6470 [34:53<41:13,  1.39it/s]
3|3031|Loss: 0.8903886079788208:  47%|████▋     | 3031/6470 [34:53<43:01,  1.33it/s]
3|3032|Loss: 0.7773599624633789:  47%|████▋     | 3031/6470 [34:54<43:01,  1.33it/s]
3|3032|Loss: 0.7773599624633789:  47%|████▋     | 3032/6470 [34:54<43:08,  1.33it/s]
3|3033|Loss: 1.0320308208465576:  47%|████▋     | 3032/6470 [34:54<43:08,  1.33it/s]
3|3033|Loss: 1.0320308208465576:  47%|████▋     | 3033/6470 [34:55<44:05,  1.30it/s]
3|3034|Loss: 0.9160419702529907:  47%|████▋     | 3033/6470 [34:55<44:05,  1.30it/s]
3|3034|Loss: 0.9160419702529907:  47%|████▋     | 3034/6470 [34:56<43:26,  1.32it/s]
3|3035|Loss: 0.7885375022888184:  47%|████▋     | 3034/6470 [34:56<43:26,  1.32it/s]
3|3035|Loss: 0.7885375022888184:  47%|████▋     | 3035/6470 [34:56<42:49,  1.34it/s]
3|3036|Loss: 0.9936913847923279:  47%|████▋     | 3035/6470 [34:57<42:49,  1.34it/s]
3|3036|Loss: 0.9936913847923279:  47%|████▋     | 3036/6470 [34:57<41:26,  1.38it/s]
3|3037|Loss: 0.9655362367630005:  47%|████▋     | 3036/6470 [34:57<41:26,  1.38it/s]
3|3037|Loss: 0.9655362367630005:  47%|████▋     | 3037/6470 [34:58<40:50,  1.40it/s]
3|3038|Loss: 0.973775327205658:  47%|████▋     | 3037/6470 [34:58<40:50,  1.40it/s] 
3|3038|Loss: 0.973775327205658:  47%|████▋     | 3038/6470 [34:58<41:53,  1.37it/s]
3|3039|Loss: 0.7895958423614502:  47%|████▋     | 3038/6470 [34:59<41:53,  1.37it/s]
3|3039|Loss: 0.7895958423614502:  47%|████▋     | 3039/6470 [34:59<41:47,  1.37it/s]
3|3040|Loss: 0.791428804397583:  47%|████▋     | 3039/6470 [35:00<41:47,  1.37it/s] 
3|3040|Loss: 0.791428804397583:  47%|████▋     | 3040/6470 [35:00<41:23,  1.38it/s]
3|3041|Loss: 0.8893521428108215:  47%|████▋     | 3040/6470 [35:00<41:23,  1.38it/s]
3|3041|Loss: 0.8893521428108215:  47%|████▋     | 3041/6470 [35:01<39:50,  1.43it/s]
3|3042|Loss: 0.8090468049049377:  47%|████▋     | 3041/6470 [35:01<39:50,  1.43it/s]
3|3042|Loss: 0.8090468049049377:  47%|████▋     | 3042/6470 [35:01<38:24,  1.49it/s]
3|3043|Loss: 0.8057902455329895:  47%|████▋     | 3042/6470 [35:01<38:24,  1.49it/s]
3|3043|Loss: 0.8057902455329895:  47%|████▋     | 3043/6470 [35:02<35:59,  1.59it/s]
3|3044|Loss: 0.9354790449142456:  47%|████▋     | 3043/6470 [35:02<35:59,  1.59it/s]
3|3044|Loss: 0.9354790449142456:  47%|████▋     | 3044/6470 [35:02<36:31,  1.56it/s]
3|3045|Loss: 0.8007290363311768:  47%|████▋     | 3044/6470 [35:03<36:31,  1.56it/s]
3|3045|Loss: 0.8007290363311768:  47%|████▋     | 3045/6470 [35:03<36:02,  1.58it/s]
3|3046|Loss: 1.2380199432373047:  47%|████▋     | 3045/6470 [35:03<36:02,  1.58it/s]
3|3046|Loss: 1.2380199432373047:  47%|████▋     | 3046/6470 [35:03<34:11,  1.67it/s]
3|3047|Loss: 0.9387821555137634:  47%|████▋     | 3046/6470 [35:04<34:11,  1.67it/s]
3|3047|Loss: 0.9387821555137634:  47%|████▋     | 3047/6470 [35:04<35:15,  1.62it/s]
3|3048|Loss: 0.9335724711418152:  47%|████▋     | 3047/6470 [35:05<35:15,  1.62it/s]
3|3048|Loss: 0.9335724711418152:  47%|████▋     | 3048/6470 [35:05<39:16,  1.45it/s]
3|3049|Loss: 0.8354218006134033:  47%|████▋     | 3048/6470 [35:05<39:16,  1.45it/s]
3|3049|Loss: 0.8354218006134033:  47%|████▋     | 3049/6470 [35:06<39:13,  1.45it/s]
3|3050|Loss: 0.6262648105621338:  47%|████▋     | 3049/6470 [35:06<39:13,  1.45it/s]
3|3050|Loss: 0.6262648105621338:  47%|████▋     | 3050/6470 [35:06<40:51,  1.39it/s]
3|3051|Loss: 1.0216773748397827:  47%|████▋     | 3050/6470 [35:07<40:51,  1.39it/s]
3|3051|Loss: 1.0216773748397827:  47%|████▋     | 3051/6470 [35:07<38:26,  1.48it/s]
3|3052|Loss: 0.7189499735832214:  47%|████▋     | 3051/6470 [35:07<38:26,  1.48it/s]
3|3052|Loss: 0.7189499735832214:  47%|████▋     | 3052/6470 [35:08<40:16,  1.41it/s]
3|3053|Loss: 0.794815719127655:  47%|████▋     | 3052/6470 [35:08<40:16,  1.41it/s] 
3|3053|Loss: 0.794815719127655:  47%|████▋     | 3053/6470 [35:09<42:59,  1.32it/s]
3|3054|Loss: 0.7470136880874634:  47%|████▋     | 3053/6470 [35:09<42:59,  1.32it/s]
3|3054|Loss: 0.7470136880874634:  47%|████▋     | 3054/6470 [35:09<41:04,  1.39it/s]
3|3055|Loss: 0.8829060792922974:  47%|████▋     | 3054/6470 [35:10<41:04,  1.39it/s]
3|3055|Loss: 0.8829060792922974:  47%|████▋     | 3055/6470 [35:10<40:55,  1.39it/s]
3|3056|Loss: 0.9160705208778381:  47%|████▋     | 3055/6470 [35:10<40:55,  1.39it/s]
3|3056|Loss: 0.9160705208778381:  47%|████▋     | 3056/6470 [35:11<39:05,  1.46it/s]
3|3057|Loss: 0.9343807697296143:  47%|████▋     | 3056/6470 [35:11<39:05,  1.46it/s]
3|3057|Loss: 0.9343807697296143:  47%|████▋     | 3057/6470 [35:11<38:49,  1.47it/s]
3|3058|Loss: 1.0864276885986328:  47%|████▋     | 3057/6470 [35:12<38:49,  1.47it/s]
3|3058|Loss: 1.0864276885986328:  47%|████▋     | 3058/6470 [35:12<37:24,  1.52it/s]
3|3059|Loss: 0.8715566992759705:  47%|████▋     | 3058/6470 [35:12<37:24,  1.52it/s]
3|3059|Loss: 0.8715566992759705:  47%|████▋     | 3059/6470 [35:13<40:14,  1.41it/s]
3|3060|Loss: 0.8150107264518738:  47%|████▋     | 3059/6470 [35:13<40:14,  1.41it/s]
3|3060|Loss: 0.8150107264518738:  47%|████▋     | 3060/6470 [35:14<41:34,  1.37it/s]
3|3061|Loss: 0.9413614273071289:  47%|████▋     | 3060/6470 [35:14<41:34,  1.37it/s]
3|3061|Loss: 0.9413614273071289:  47%|████▋     | 3061/6470 [35:14<41:44,  1.36it/s]
3|3062|Loss: 0.8970057368278503:  47%|████▋     | 3061/6470 [35:15<41:44,  1.36it/s]
3|3062|Loss: 0.8970057368278503:  47%|████▋     | 3062/6470 [35:15<43:17,  1.31it/s]
3|3063|Loss: 0.8493629693984985:  47%|████▋     | 3062/6470 [35:16<43:17,  1.31it/s]
3|3063|Loss: 0.8493629693984985:  47%|████▋     | 3063/6470 [35:16<44:31,  1.28it/s]
3|3064|Loss: 0.8190325498580933:  47%|████▋     | 3063/6470 [35:16<44:31,  1.28it/s]
3|3064|Loss: 0.8190325498580933:  47%|████▋     | 3064/6470 [35:17<45:15,  1.25it/s]
3|3065|Loss: 0.8649197220802307:  47%|████▋     | 3064/6470 [35:17<45:15,  1.25it/s]
3|3065|Loss: 0.8649197220802307:  47%|████▋     | 3065/6470 [35:17<43:18,  1.31it/s]
3|3066|Loss: 0.810476541519165:  47%|████▋     | 3065/6470 [35:18<43:18,  1.31it/s] 
3|3066|Loss: 0.810476541519165:  47%|████▋     | 3066/6470 [35:18<44:07,  1.29it/s]
3|3067|Loss: 0.8333310484886169:  47%|████▋     | 3066/6470 [35:19<44:07,  1.29it/s]
3|3067|Loss: 0.8333310484886169:  47%|████▋     | 3067/6470 [35:19<43:49,  1.29it/s]
3|3068|Loss: 0.8183352947235107:  47%|████▋     | 3067/6470 [35:19<43:49,  1.29it/s]
3|3068|Loss: 0.8183352947235107:  47%|████▋     | 3068/6470 [35:20<44:46,  1.27it/s]
3|3069|Loss: 1.026822805404663:  47%|████▋     | 3068/6470 [35:20<44:46,  1.27it/s] 
3|3069|Loss: 1.026822805404663:  47%|████▋     | 3069/6470 [35:21<42:59,  1.32it/s]
3|3070|Loss: 0.8328171968460083:  47%|████▋     | 3069/6470 [35:21<42:59,  1.32it/s]
3|3070|Loss: 0.8328171968460083:  47%|████▋     | 3070/6470 [35:21<41:43,  1.36it/s]
3|3071|Loss: 0.8341094255447388:  47%|████▋     | 3070/6470 [35:22<41:43,  1.36it/s]
3|3071|Loss: 0.8341094255447388:  47%|████▋     | 3071/6470 [35:22<43:15,  1.31it/s]
3|3072|Loss: 0.6680120229721069:  47%|████▋     | 3071/6470 [35:22<43:15,  1.31it/s]
3|3072|Loss: 0.6680120229721069:  47%|████▋     | 3072/6470 [35:23<43:25,  1.30it/s]
3|3073|Loss: 0.9123927354812622:  47%|████▋     | 3072/6470 [35:23<43:25,  1.30it/s]
3|3073|Loss: 0.9123927354812622:  47%|████▋     | 3073/6470 [35:24<43:09,  1.31it/s]
3|3074|Loss: 0.9028985500335693:  47%|████▋     | 3073/6470 [35:24<43:09,  1.31it/s]
3|3074|Loss: 0.9028985500335693:  48%|████▊     | 3074/6470 [35:24<42:17,  1.34it/s]
3|3075|Loss: 0.9988689422607422:  48%|████▊     | 3074/6470 [35:25<42:17,  1.34it/s]
3|3075|Loss: 0.9988689422607422:  48%|████▊     | 3075/6470 [35:25<39:08,  1.45it/s]
3|3076|Loss: 0.847106397151947:  48%|████▊     | 3075/6470 [35:25<39:08,  1.45it/s] 
3|3076|Loss: 0.847106397151947:  48%|████▊     | 3076/6470 [35:26<41:05,  1.38it/s]
3|3077|Loss: 0.9794092178344727:  48%|████▊     | 3076/6470 [35:26<41:05,  1.38it/s]
3|3077|Loss: 0.9794092178344727:  48%|████▊     | 3077/6470 [35:26<39:05,  1.45it/s]
3|3078|Loss: 0.596482515335083:  48%|████▊     | 3077/6470 [35:27<39:05,  1.45it/s] 
3|3078|Loss: 0.596482515335083:  48%|████▊     | 3078/6470 [35:27<40:51,  1.38it/s]
3|3079|Loss: 0.9686345458030701:  48%|████▊     | 3078/6470 [35:27<40:51,  1.38it/s]
3|3079|Loss: 0.9686345458030701:  48%|████▊     | 3079/6470 [35:28<41:40,  1.36it/s]
3|3080|Loss: 0.8219645619392395:  48%|████▊     | 3079/6470 [35:28<41:40,  1.36it/s]
3|3080|Loss: 0.8219645619392395:  48%|████▊     | 3080/6470 [35:28<40:33,  1.39it/s]
3|3081|Loss: 0.9183330535888672:  48%|████▊     | 3080/6470 [35:29<40:33,  1.39it/s]
3|3081|Loss: 0.9183330535888672:  48%|████▊     | 3081/6470 [35:29<38:00,  1.49it/s]
3|3082|Loss: 0.8650503158569336:  48%|████▊     | 3081/6470 [35:29<38:00,  1.49it/s]
3|3082|Loss: 0.8650503158569336:  48%|████▊     | 3082/6470 [35:30<39:50,  1.42it/s]
3|3083|Loss: 0.6493242383003235:  48%|████▊     | 3082/6470 [35:30<39:50,  1.42it/s]
3|3083|Loss: 0.6493242383003235:  48%|████▊     | 3083/6470 [35:31<39:25,  1.43it/s]
3|3084|Loss: 0.9092275500297546:  48%|████▊     | 3083/6470 [35:31<39:25,  1.43it/s]
3|3084|Loss: 0.9092275500297546:  48%|████▊     | 3084/6470 [35:31<37:10,  1.52it/s]
3|3085|Loss: 0.9142180681228638:  48%|████▊     | 3084/6470 [35:31<37:10,  1.52it/s]
3|3085|Loss: 0.9142180681228638:  48%|████▊     | 3085/6470 [35:32<37:27,  1.51it/s]
3|3086|Loss: 0.696478545665741:  48%|████▊     | 3085/6470 [35:32<37:27,  1.51it/s] 
3|3086|Loss: 0.696478545665741:  48%|████▊     | 3086/6470 [35:32<37:18,  1.51it/s]
3|3087|Loss: 0.8431189656257629:  48%|████▊     | 3086/6470 [35:33<37:18,  1.51it/s]
3|3087|Loss: 0.8431189656257629:  48%|████▊     | 3087/6470 [35:33<38:14,  1.47it/s]
3|3088|Loss: 0.908891499042511:  48%|████▊     | 3087/6470 [35:34<38:14,  1.47it/s] 
3|3088|Loss: 0.908891499042511:  48%|████▊     | 3088/6470 [35:34<40:25,  1.39it/s]
3|3089|Loss: 0.8186722993850708:  48%|████▊     | 3088/6470 [35:34<40:25,  1.39it/s]
3|3089|Loss: 0.8186722993850708:  48%|████▊     | 3089/6470 [35:35<41:15,  1.37it/s]
3|3090|Loss: 0.8298512697219849:  48%|████▊     | 3089/6470 [35:35<41:15,  1.37it/s]
3|3090|Loss: 0.8298512697219849:  48%|████▊     | 3090/6470 [35:35<39:47,  1.42it/s]
3|3091|Loss: 0.9563175439834595:  48%|████▊     | 3090/6470 [35:36<39:47,  1.42it/s]
3|3091|Loss: 0.9563175439834595:  48%|████▊     | 3091/6470 [35:36<40:32,  1.39it/s]
3|3092|Loss: 0.9109489321708679:  48%|████▊     | 3091/6470 [35:37<40:32,  1.39it/s]
3|3092|Loss: 0.9109489321708679:  48%|████▊     | 3092/6470 [35:37<40:54,  1.38it/s]
3|3093|Loss: 0.6975999474525452:  48%|████▊     | 3092/6470 [35:37<40:54,  1.38it/s]
3|3093|Loss: 0.6975999474525452:  48%|████▊     | 3093/6470 [35:38<42:37,  1.32it/s]
3|3094|Loss: 0.7883931994438171:  48%|████▊     | 3093/6470 [35:38<42:37,  1.32it/s]
3|3094|Loss: 0.7883931994438171:  48%|████▊     | 3094/6470 [35:39<44:00,  1.28it/s]
3|3095|Loss: 0.923276424407959:  48%|████▊     | 3094/6470 [35:39<44:00,  1.28it/s] 
3|3095|Loss: 0.923276424407959:  48%|████▊     | 3095/6470 [35:39<43:33,  1.29it/s]
3|3096|Loss: 0.6949280500411987:  48%|████▊     | 3095/6470 [35:40<43:33,  1.29it/s]
3|3096|Loss: 0.6949280500411987:  48%|████▊     | 3096/6470 [35:40<40:27,  1.39it/s]
3|3097|Loss: 0.7890408635139465:  48%|████▊     | 3096/6470 [35:40<40:27,  1.39it/s]
3|3097|Loss: 0.7890408635139465:  48%|████▊     | 3097/6470 [35:40<34:49,  1.61it/s]
3|3098|Loss: 0.7454801201820374:  48%|████▊     | 3097/6470 [35:41<34:49,  1.61it/s]
3|3098|Loss: 0.7454801201820374:  48%|████▊     | 3098/6470 [35:41<32:50,  1.71it/s]
3|3099|Loss: 0.8899216055870056:  48%|████▊     | 3098/6470 [35:41<32:50,  1.71it/s]
3|3099|Loss: 0.8899216055870056:  48%|████▊     | 3099/6470 [35:42<36:04,  1.56it/s]
3|3100|Loss: 0.7864204049110413:  48%|████▊     | 3099/6470 [35:42<36:04,  1.56it/s]
3|3100|Loss: 0.7864204049110413:  48%|████▊     | 3100/6470 [35:42<34:55,  1.61it/s]
3|3101|Loss: 0.9151044487953186:  48%|████▊     | 3100/6470 [35:42<34:55,  1.61it/s]
3|3101|Loss: 0.9151044487953186:  48%|████▊     | 3101/6470 [35:43<33:13,  1.69it/s]
3|3102|Loss: 0.8756182193756104:  48%|████▊     | 3101/6470 [35:43<33:13,  1.69it/s]
3|3102|Loss: 0.8756182193756104:  48%|████▊     | 3102/6470 [35:43<35:10,  1.60it/s]
3|3103|Loss: 0.8284689784049988:  48%|████▊     | 3102/6470 [35:44<35:10,  1.60it/s]
3|3103|Loss: 0.8284689784049988:  48%|████▊     | 3103/6470 [35:44<37:08,  1.51it/s]
3|3104|Loss: 0.8535308837890625:  48%|████▊     | 3103/6470 [35:44<37:08,  1.51it/s]
3|3104|Loss: 0.8535308837890625:  48%|████▊     | 3104/6470 [35:45<34:29,  1.63it/s]
3|3105|Loss: 0.7712504863739014:  48%|████▊     | 3104/6470 [35:45<34:29,  1.63it/s]
3|3105|Loss: 0.7712504863739014:  48%|████▊     | 3105/6470 [35:45<36:07,  1.55it/s]
3|3106|Loss: 0.7776767015457153:  48%|████▊     | 3105/6470 [35:46<36:07,  1.55it/s]
3|3106|Loss: 0.7776767015457153:  48%|████▊     | 3106/6470 [35:46<39:39,  1.41it/s]
3|3107|Loss: 0.8812794089317322:  48%|████▊     | 3106/6470 [35:47<39:39,  1.41it/s]
3|3107|Loss: 0.8812794089317322:  48%|████▊     | 3107/6470 [35:47<40:37,  1.38it/s]
3|3108|Loss: 0.7840597033500671:  48%|████▊     | 3107/6470 [35:47<40:37,  1.38it/s]
3|3108|Loss: 0.7840597033500671:  48%|████▊     | 3108/6470 [35:48<39:11,  1.43it/s]
3|3109|Loss: 0.9510113000869751:  48%|████▊     | 3108/6470 [35:48<39:11,  1.43it/s]
3|3109|Loss: 0.9510113000869751:  48%|████▊     | 3109/6470 [35:48<38:55,  1.44it/s]
3|3110|Loss: 0.9017333388328552:  48%|████▊     | 3109/6470 [35:49<38:55,  1.44it/s]
3|3110|Loss: 0.9017333388328552:  48%|████▊     | 3110/6470 [35:49<39:26,  1.42it/s]
3|3111|Loss: 0.9725289940834045:  48%|████▊     | 3110/6470 [35:49<39:26,  1.42it/s]
3|3111|Loss: 0.9725289940834045:  48%|████▊     | 3111/6470 [35:50<39:35,  1.41it/s]
3|3112|Loss: 0.769462525844574:  48%|████▊     | 3111/6470 [35:50<39:35,  1.41it/s] 
3|3112|Loss: 0.769462525844574:  48%|████▊     | 3112/6470 [35:50<37:45,  1.48it/s]
3|3113|Loss: 0.8557957410812378:  48%|████▊     | 3112/6470 [35:51<37:45,  1.48it/s]
3|3113|Loss: 0.8557957410812378:  48%|████▊     | 3113/6470 [35:51<37:44,  1.48it/s]
3|3114|Loss: 0.7601239085197449:  48%|████▊     | 3113/6470 [35:51<37:44,  1.48it/s]
3|3114|Loss: 0.7601239085197449:  48%|████▊     | 3114/6470 [35:52<38:34,  1.45it/s]
3|3115|Loss: 0.8349854946136475:  48%|████▊     | 3114/6470 [35:52<38:34,  1.45it/s]
3|3115|Loss: 0.8349854946136475:  48%|████▊     | 3115/6470 [35:52<35:49,  1.56it/s]
3|3116|Loss: 0.7485036849975586:  48%|████▊     | 3115/6470 [35:53<35:49,  1.56it/s]
3|3116|Loss: 0.7485036849975586:  48%|████▊     | 3116/6470 [35:53<34:54,  1.60it/s]
3|3117|Loss: 0.8337036967277527:  48%|████▊     | 3116/6470 [35:53<34:54,  1.60it/s]
3|3117|Loss: 0.8337036967277527:  48%|████▊     | 3117/6470 [35:53<33:52,  1.65it/s]
3|3118|Loss: 1.010801911354065:  48%|████▊     | 3117/6470 [35:54<33:52,  1.65it/s] 
3|3118|Loss: 1.010801911354065:  48%|████▊     | 3118/6470 [35:54<35:45,  1.56it/s]
3|3119|Loss: 0.8836497664451599:  48%|████▊     | 3118/6470 [35:55<35:45,  1.56it/s]
3|3119|Loss: 0.8836497664451599:  48%|████▊     | 3119/6470 [35:55<38:44,  1.44it/s]
3|3120|Loss: 0.7526281476020813:  48%|████▊     | 3119/6470 [35:55<38:44,  1.44it/s]
3|3120|Loss: 0.7526281476020813:  48%|████▊     | 3120/6470 [35:56<40:00,  1.40it/s]
3|3121|Loss: 0.7203413248062134:  48%|████▊     | 3120/6470 [35:56<40:00,  1.40it/s]
3|3121|Loss: 0.7203413248062134:  48%|████▊     | 3121/6470 [35:56<39:03,  1.43it/s]
3|3122|Loss: 0.9489113688468933:  48%|████▊     | 3121/6470 [35:57<39:03,  1.43it/s]
3|3122|Loss: 0.9489113688468933:  48%|████▊     | 3122/6470 [35:57<40:38,  1.37it/s]
3|3123|Loss: 0.9676462411880493:  48%|████▊     | 3122/6470 [35:58<40:38,  1.37it/s]
3|3123|Loss: 0.9676462411880493:  48%|████▊     | 3123/6470 [35:58<38:36,  1.44it/s]
3|3124|Loss: 1.0000228881835938:  48%|████▊     | 3123/6470 [35:58<38:36,  1.44it/s]
3|3124|Loss: 1.0000228881835938:  48%|████▊     | 3124/6470 [35:58<36:43,  1.52it/s]
3|3125|Loss: 1.01813805103302:  48%|████▊     | 3124/6470 [35:59<36:43,  1.52it/s]  
3|3125|Loss: 1.01813805103302:  48%|████▊     | 3125/6470 [35:59<35:27,  1.57it/s]
3|3126|Loss: 1.0001360177993774:  48%|████▊     | 3125/6470 [35:59<35:27,  1.57it/s]
3|3126|Loss: 1.0001360177993774:  48%|████▊     | 3126/6470 [35:59<32:53,  1.69it/s]
3|3127|Loss: 0.7302360534667969:  48%|████▊     | 3126/6470 [36:00<32:53,  1.69it/s]
3|3127|Loss: 0.7302360534667969:  48%|████▊     | 3127/6470 [36:00<34:25,  1.62it/s]
3|3128|Loss: 0.7590215802192688:  48%|████▊     | 3127/6470 [36:00<34:25,  1.62it/s]
3|3128|Loss: 0.7590215802192688:  48%|████▊     | 3128/6470 [36:01<34:06,  1.63it/s]
3|3129|Loss: 0.7938475012779236:  48%|████▊     | 3128/6470 [36:01<34:06,  1.63it/s]
3|3129|Loss: 0.7938475012779236:  48%|████▊     | 3129/6470 [36:01<35:02,  1.59it/s]
3|3130|Loss: 0.8300547003746033:  48%|████▊     | 3129/6470 [36:02<35:02,  1.59it/s]
3|3130|Loss: 0.8300547003746033:  48%|████▊     | 3130/6470 [36:02<35:25,  1.57it/s]
3|3131|Loss: 0.9376870393753052:  48%|████▊     | 3130/6470 [36:02<35:25,  1.57it/s]
3|3131|Loss: 0.9376870393753052:  48%|████▊     | 3131/6470 [36:03<37:30,  1.48it/s]
3|3132|Loss: 0.9140872955322266:  48%|████▊     | 3131/6470 [36:03<37:30,  1.48it/s]
3|3132|Loss: 0.9140872955322266:  48%|████▊     | 3132/6470 [36:04<39:02,  1.42it/s]
3|3133|Loss: 0.6919993758201599:  48%|████▊     | 3132/6470 [36:04<39:02,  1.42it/s]
3|3133|Loss: 0.6919993758201599:  48%|████▊     | 3133/6470 [36:04<36:57,  1.50it/s]
3|3134|Loss: 0.9706937074661255:  48%|████▊     | 3133/6470 [36:04<36:57,  1.50it/s]
3|3134|Loss: 0.9706937074661255:  48%|████▊     | 3134/6470 [36:05<38:37,  1.44it/s]
3|3135|Loss: 0.7822626829147339:  48%|████▊     | 3134/6470 [36:05<38:37,  1.44it/s]
3|3135|Loss: 0.7822626829147339:  48%|████▊     | 3135/6470 [36:05<34:37,  1.61it/s]
3|3136|Loss: 0.9421255588531494:  48%|████▊     | 3135/6470 [36:06<34:37,  1.61it/s]
3|3136|Loss: 0.9421255588531494:  48%|████▊     | 3136/6470 [36:06<35:34,  1.56it/s]
3|3137|Loss: 0.9549202919006348:  48%|████▊     | 3136/6470 [36:06<35:34,  1.56it/s]
3|3137|Loss: 0.9549202919006348:  48%|████▊     | 3137/6470 [36:07<38:38,  1.44it/s]
3|3138|Loss: 0.8379469513893127:  48%|████▊     | 3137/6470 [36:07<38:38,  1.44it/s]
3|3138|Loss: 0.8379469513893127:  49%|████▊     | 3138/6470 [36:07<37:32,  1.48it/s]
3|3139|Loss: 0.8748616576194763:  49%|████▊     | 3138/6470 [36:08<37:32,  1.48it/s]
3|3139|Loss: 0.8748616576194763:  49%|████▊     | 3139/6470 [36:08<39:33,  1.40it/s]
3|3140|Loss: 0.9093361496925354:  49%|████▊     | 3139/6470 [36:09<39:33,  1.40it/s]
3|3140|Loss: 0.9093361496925354:  49%|████▊     | 3140/6470 [36:09<35:25,  1.57it/s]
3|3141|Loss: 0.649642288684845:  49%|████▊     | 3140/6470 [36:09<35:25,  1.57it/s] 
3|3141|Loss: 0.649642288684845:  49%|████▊     | 3141/6470 [36:09<34:42,  1.60it/s]
3|3142|Loss: 0.904506266117096:  49%|████▊     | 3141/6470 [36:10<34:42,  1.60it/s]
3|3142|Loss: 0.904506266117096:  49%|████▊     | 3142/6470 [36:10<36:12,  1.53it/s]
3|3143|Loss: 0.964634895324707:  49%|████▊     | 3142/6470 [36:10<36:12,  1.53it/s]
3|3143|Loss: 0.964634895324707:  49%|████▊     | 3143/6470 [36:11<35:56,  1.54it/s]
3|3144|Loss: 0.9224346280097961:  49%|████▊     | 3143/6470 [36:11<35:56,  1.54it/s]
3|3144|Loss: 0.9224346280097961:  49%|████▊     | 3144/6470 [36:11<36:58,  1.50it/s]
3|3145|Loss: 1.0267990827560425:  49%|████▊     | 3144/6470 [36:12<36:58,  1.50it/s]
3|3145|Loss: 1.0267990827560425:  49%|████▊     | 3145/6470 [36:12<38:07,  1.45it/s]
3|3146|Loss: 0.9565821290016174:  49%|████▊     | 3145/6470 [36:13<38:07,  1.45it/s]
3|3146|Loss: 0.9565821290016174:  49%|████▊     | 3146/6470 [36:13<40:06,  1.38it/s]
3|3147|Loss: 0.8858937621116638:  49%|████▊     | 3146/6470 [36:13<40:06,  1.38it/s]
3|3147|Loss: 0.8858937621116638:  49%|████▊     | 3147/6470 [36:14<42:27,  1.30it/s]
3|3148|Loss: 0.9854304790496826:  49%|████▊     | 3147/6470 [36:14<42:27,  1.30it/s]
3|3148|Loss: 0.9854304790496826:  49%|████▊     | 3148/6470 [36:14<40:33,  1.36it/s]
3|3149|Loss: 0.8997884392738342:  49%|████▊     | 3148/6470 [36:15<40:33,  1.36it/s]
3|3149|Loss: 0.8997884392738342:  49%|████▊     | 3149/6470 [36:15<38:02,  1.45it/s]
3|3150|Loss: 0.6963900923728943:  49%|████▊     | 3149/6470 [36:15<38:02,  1.45it/s]
3|3150|Loss: 0.6963900923728943:  49%|████▊     | 3150/6470 [36:16<40:02,  1.38it/s]
3|3151|Loss: 0.9097188115119934:  49%|████▊     | 3150/6470 [36:16<40:02,  1.38it/s]
3|3151|Loss: 0.9097188115119934:  49%|████▊     | 3151/6470 [36:17<42:37,  1.30it/s]
3|3152|Loss: 0.9787476062774658:  49%|████▊     | 3151/6470 [36:17<42:37,  1.30it/s]
3|3152|Loss: 0.9787476062774658:  49%|████▊     | 3152/6470 [36:17<40:10,  1.38it/s]
3|3153|Loss: 0.7366382479667664:  49%|████▊     | 3152/6470 [36:18<40:10,  1.38it/s]
3|3153|Loss: 0.7366382479667664:  49%|████▊     | 3153/6470 [36:18<35:58,  1.54it/s]
3|3154|Loss: 0.7634907960891724:  49%|████▊     | 3153/6470 [36:18<35:58,  1.54it/s]
3|3154|Loss: 0.7634907960891724:  49%|████▊     | 3154/6470 [36:18<35:03,  1.58it/s]
3|3155|Loss: 0.8035957217216492:  49%|████▊     | 3154/6470 [36:19<35:03,  1.58it/s]
3|3155|Loss: 0.8035957217216492:  49%|████▉     | 3155/6470 [36:19<32:39,  1.69it/s]
3|3156|Loss: 0.8469956517219543:  49%|████▉     | 3155/6470 [36:19<32:39,  1.69it/s]
3|3156|Loss: 0.8469956517219543:  49%|████▉     | 3156/6470 [36:20<33:40,  1.64it/s]
3|3157|Loss: 0.8522918224334717:  49%|████▉     | 3156/6470 [36:20<33:40,  1.64it/s]
3|3157|Loss: 0.8522918224334717:  49%|████▉     | 3157/6470 [36:20<37:19,  1.48it/s]
3|3158|Loss: 1.0062236785888672:  49%|████▉     | 3157/6470 [36:21<37:19,  1.48it/s]
3|3158|Loss: 1.0062236785888672:  49%|████▉     | 3158/6470 [36:21<37:17,  1.48it/s]
3|3159|Loss: 0.9621714353561401:  49%|████▉     | 3158/6470 [36:21<37:17,  1.48it/s]
3|3159|Loss: 0.9621714353561401:  49%|████▉     | 3159/6470 [36:22<37:47,  1.46it/s]
3|3160|Loss: 0.8298168182373047:  49%|████▉     | 3159/6470 [36:22<37:47,  1.46it/s]
3|3160|Loss: 0.8298168182373047:  49%|████▉     | 3160/6470 [36:22<37:05,  1.49it/s]
3|3161|Loss: 0.8685294985771179:  49%|████▉     | 3160/6470 [36:23<37:05,  1.49it/s]
3|3161|Loss: 0.8685294985771179:  49%|████▉     | 3161/6470 [36:23<36:24,  1.51it/s]
3|3162|Loss: 1.0260823965072632:  49%|████▉     | 3161/6470 [36:23<36:24,  1.51it/s]
3|3162|Loss: 1.0260823965072632:  49%|████▉     | 3162/6470 [36:24<37:32,  1.47it/s]
3|3163|Loss: 0.828794538974762:  49%|████▉     | 3162/6470 [36:24<37:32,  1.47it/s] 
3|3163|Loss: 0.828794538974762:  49%|████▉     | 3163/6470 [36:25<39:20,  1.40it/s]
3|3164|Loss: 0.9292829632759094:  49%|████▉     | 3163/6470 [36:25<39:20,  1.40it/s]
3|3164|Loss: 0.9292829632759094:  49%|████▉     | 3164/6470 [36:25<41:05,  1.34it/s]
3|3165|Loss: 0.884618878364563:  49%|████▉     | 3164/6470 [36:26<41:05,  1.34it/s] 
3|3165|Loss: 0.884618878364563:  49%|████▉     | 3165/6470 [36:26<43:03,  1.28it/s]
3|3166|Loss: 0.7699130177497864:  49%|████▉     | 3165/6470 [36:27<43:03,  1.28it/s]
3|3166|Loss: 0.7699130177497864:  49%|████▉     | 3166/6470 [36:27<42:53,  1.28it/s]
3|3167|Loss: 0.83443683385849:  49%|████▉     | 3166/6470 [36:27<42:53,  1.28it/s]  
3|3167|Loss: 0.83443683385849:  49%|████▉     | 3167/6470 [36:28<42:29,  1.30it/s]
3|3168|Loss: 0.9735674262046814:  49%|████▉     | 3167/6470 [36:28<42:29,  1.30it/s]
3|3168|Loss: 0.9735674262046814:  49%|████▉     | 3168/6470 [36:29<42:06,  1.31it/s]
3|3169|Loss: 0.9284617900848389:  49%|████▉     | 3168/6470 [36:29<42:06,  1.31it/s]
3|3169|Loss: 0.9284617900848389:  49%|████▉     | 3169/6470 [36:29<43:15,  1.27it/s]
3|3170|Loss: 0.8001188635826111:  49%|████▉     | 3169/6470 [36:30<43:15,  1.27it/s]
3|3170|Loss: 0.8001188635826111:  49%|████▉     | 3170/6470 [36:30<40:45,  1.35it/s]
3|3171|Loss: 0.8957226276397705:  49%|████▉     | 3170/6470 [36:30<40:45,  1.35it/s]
3|3171|Loss: 0.8957226276397705:  49%|████▉     | 3171/6470 [36:31<38:37,  1.42it/s]
3|3172|Loss: 0.6737200617790222:  49%|████▉     | 3171/6470 [36:31<38:37,  1.42it/s]
3|3172|Loss: 0.6737200617790222:  49%|████▉     | 3172/6470 [36:31<33:14,  1.65it/s]
3|3173|Loss: 0.8655325770378113:  49%|████▉     | 3172/6470 [36:31<33:14,  1.65it/s]
3|3173|Loss: 0.8655325770378113:  49%|████▉     | 3173/6470 [36:31<30:07,  1.82it/s]
3|3174|Loss: 0.7492116689682007:  49%|████▉     | 3173/6470 [36:32<30:07,  1.82it/s]
3|3174|Loss: 0.7492116689682007:  49%|████▉     | 3174/6470 [36:32<30:52,  1.78it/s]
3|3175|Loss: 0.9274084568023682:  49%|████▉     | 3174/6470 [36:32<30:52,  1.78it/s]
3|3175|Loss: 0.9274084568023682:  49%|████▉     | 3175/6470 [36:33<33:41,  1.63it/s]
3|3176|Loss: 0.8331735134124756:  49%|████▉     | 3175/6470 [36:33<33:41,  1.63it/s]
3|3176|Loss: 0.8331735134124756:  49%|████▉     | 3176/6470 [36:33<34:18,  1.60it/s]
3|3177|Loss: 0.9129645824432373:  49%|████▉     | 3176/6470 [36:34<34:18,  1.60it/s]
3|3177|Loss: 0.9129645824432373:  49%|████▉     | 3177/6470 [36:34<36:26,  1.51it/s]
3|3178|Loss: 0.8365009427070618:  49%|████▉     | 3177/6470 [36:35<36:26,  1.51it/s]
3|3178|Loss: 0.8365009427070618:  49%|████▉     | 3178/6470 [36:35<38:33,  1.42it/s]
3|3179|Loss: 0.6463005542755127:  49%|████▉     | 3178/6470 [36:35<38:33,  1.42it/s]
3|3179|Loss: 0.6463005542755127:  49%|████▉     | 3179/6470 [36:36<37:44,  1.45it/s]
3|3180|Loss: 0.819818377494812:  49%|████▉     | 3179/6470 [36:36<37:44,  1.45it/s] 
3|3180|Loss: 0.819818377494812:  49%|████▉     | 3180/6470 [36:36<38:20,  1.43it/s]
3|3181|Loss: 0.8571652770042419:  49%|████▉     | 3180/6470 [36:37<38:20,  1.43it/s]
3|3181|Loss: 0.8571652770042419:  49%|████▉     | 3181/6470 [36:37<38:35,  1.42it/s]
3|3182|Loss: 0.9369932413101196:  49%|████▉     | 3181/6470 [36:37<38:35,  1.42it/s]
3|3182|Loss: 0.9369932413101196:  49%|████▉     | 3182/6470 [36:38<37:41,  1.45it/s]
3|3183|Loss: 0.8241134881973267:  49%|████▉     | 3182/6470 [36:38<37:41,  1.45it/s]
3|3183|Loss: 0.8241134881973267:  49%|████▉     | 3183/6470 [36:38<36:14,  1.51it/s]
3|3184|Loss: 1.009304404258728:  49%|████▉     | 3183/6470 [36:39<36:14,  1.51it/s] 
3|3184|Loss: 1.009304404258728:  49%|████▉     | 3184/6470 [36:39<35:38,  1.54it/s]
3|3185|Loss: 0.9514407515525818:  49%|████▉     | 3184/6470 [36:39<35:38,  1.54it/s]
3|3185|Loss: 0.9514407515525818:  49%|████▉     | 3185/6470 [36:40<36:53,  1.48it/s]
3|3186|Loss: 0.8578922748565674:  49%|████▉     | 3185/6470 [36:40<36:53,  1.48it/s]
3|3186|Loss: 0.8578922748565674:  49%|████▉     | 3186/6470 [36:40<38:00,  1.44it/s]
3|3187|Loss: 0.7704381346702576:  49%|████▉     | 3186/6470 [36:41<38:00,  1.44it/s]
3|3187|Loss: 0.7704381346702576:  49%|████▉     | 3187/6470 [36:41<40:25,  1.35it/s]
3|3188|Loss: 0.7679460048675537:  49%|████▉     | 3187/6470 [36:42<40:25,  1.35it/s]
3|3188|Loss: 0.7679460048675537:  49%|████▉     | 3188/6470 [36:42<40:51,  1.34it/s]
3|3189|Loss: 0.8177044987678528:  49%|████▉     | 3188/6470 [36:42<40:51,  1.34it/s]
3|3189|Loss: 0.8177044987678528:  49%|████▉     | 3189/6470 [36:43<40:12,  1.36it/s]
3|3190|Loss: 0.8481126427650452:  49%|████▉     | 3189/6470 [36:43<40:12,  1.36it/s]
3|3190|Loss: 0.8481126427650452:  49%|████▉     | 3190/6470 [36:43<40:06,  1.36it/s]
3|3191|Loss: 0.9992901086807251:  49%|████▉     | 3190/6470 [36:44<40:06,  1.36it/s]
3|3191|Loss: 0.9992901086807251:  49%|████▉     | 3191/6470 [36:44<38:04,  1.44it/s]
3|3192|Loss: 0.8171707391738892:  49%|████▉     | 3191/6470 [36:44<38:04,  1.44it/s]
3|3192|Loss: 0.8171707391738892:  49%|████▉     | 3192/6470 [36:45<36:39,  1.49it/s]
3|3193|Loss: 0.8498448729515076:  49%|████▉     | 3192/6470 [36:45<36:39,  1.49it/s]
3|3193|Loss: 0.8498448729515076:  49%|████▉     | 3193/6470 [36:45<38:32,  1.42it/s]
3|3194|Loss: 0.8002445101737976:  49%|████▉     | 3193/6470 [36:46<38:32,  1.42it/s]
3|3194|Loss: 0.8002445101737976:  49%|████▉     | 3194/6470 [36:46<37:26,  1.46it/s]
3|3195|Loss: 0.8362473845481873:  49%|████▉     | 3194/6470 [36:46<37:26,  1.46it/s]
3|3195|Loss: 0.8362473845481873:  49%|████▉     | 3195/6470 [36:47<35:47,  1.52it/s]
3|3196|Loss: 0.7989559769630432:  49%|████▉     | 3195/6470 [36:47<35:47,  1.52it/s]
3|3196|Loss: 0.7989559769630432:  49%|████▉     | 3196/6470 [36:47<35:18,  1.55it/s]
3|3197|Loss: 0.9793789386749268:  49%|████▉     | 3196/6470 [36:48<35:18,  1.55it/s]
3|3197|Loss: 0.9793789386749268:  49%|████▉     | 3197/6470 [36:48<35:16,  1.55it/s]
3|3198|Loss: 0.9430636167526245:  49%|████▉     | 3197/6470 [36:48<35:16,  1.55it/s]
3|3198|Loss: 0.9430636167526245:  49%|████▉     | 3198/6470 [36:49<34:21,  1.59it/s]
3|3199|Loss: 0.8874689936637878:  49%|████▉     | 3198/6470 [36:49<34:21,  1.59it/s]
3|3199|Loss: 0.8874689936637878:  49%|████▉     | 3199/6470 [36:49<33:32,  1.63it/s]
3|3200|Loss: 0.8631603121757507:  49%|████▉     | 3199/6470 [36:49<33:32,  1.63it/s]
3|3200|Loss: 0.8631603121757507:  49%|████▉     | 3200/6470 [36:50<33:46,  1.61it/s]
3|3201|Loss: 0.9927833080291748:  49%|████▉     | 3200/6470 [36:50<33:46,  1.61it/s]
3|3201|Loss: 0.9927833080291748:  49%|████▉     | 3201/6470 [36:50<35:17,  1.54it/s]
3|3202|Loss: 0.7283499240875244:  49%|████▉     | 3201/6470 [36:51<35:17,  1.54it/s]
3|3202|Loss: 0.7283499240875244:  49%|████▉     | 3202/6470 [36:51<35:11,  1.55it/s]
3|3203|Loss: 0.8244174718856812:  49%|████▉     | 3202/6470 [36:51<35:11,  1.55it/s]
3|3203|Loss: 0.8244174718856812:  50%|████▉     | 3203/6470 [36:52<36:10,  1.51it/s]
3|3204|Loss: 0.9447420239448547:  50%|████▉     | 3203/6470 [36:52<36:10,  1.51it/s]
3|3204|Loss: 0.9447420239448547:  50%|████▉     | 3204/6470 [36:52<36:11,  1.50it/s]
3|3205|Loss: 0.8578276038169861:  50%|████▉     | 3204/6470 [36:53<36:11,  1.50it/s]
3|3205|Loss: 0.8578276038169861:  50%|████▉     | 3205/6470 [36:53<36:33,  1.49it/s]
3|3206|Loss: 0.8746151328086853:  50%|████▉     | 3205/6470 [36:54<36:33,  1.49it/s]
3|3206|Loss: 0.8746151328086853:  50%|████▉     | 3206/6470 [36:54<37:24,  1.45it/s]
3|3207|Loss: 0.8126007318496704:  50%|████▉     | 3206/6470 [36:54<37:24,  1.45it/s]
3|3207|Loss: 0.8126007318496704:  50%|████▉     | 3207/6470 [36:55<38:10,  1.42it/s]
3|3208|Loss: 0.8537086844444275:  50%|████▉     | 3207/6470 [36:55<38:10,  1.42it/s]
3|3208|Loss: 0.8537086844444275:  50%|████▉     | 3208/6470 [36:55<37:30,  1.45it/s]
3|3209|Loss: 0.8632429242134094:  50%|████▉     | 3208/6470 [36:56<37:30,  1.45it/s]
3|3209|Loss: 0.8632429242134094:  50%|████▉     | 3209/6470 [36:56<38:41,  1.40it/s]
3|3210|Loss: 0.6625876426696777:  50%|████▉     | 3209/6470 [36:56<38:41,  1.40it/s]
3|3210|Loss: 0.6625876426696777:  50%|████▉     | 3210/6470 [36:57<40:25,  1.34it/s]
3|3211|Loss: 0.9816742539405823:  50%|████▉     | 3210/6470 [36:57<40:25,  1.34it/s]
3|3211|Loss: 0.9816742539405823:  50%|████▉     | 3211/6470 [36:57<39:02,  1.39it/s]
3|3212|Loss: 0.973612368106842:  50%|████▉     | 3211/6470 [36:58<39:02,  1.39it/s] 
3|3212|Loss: 0.973612368106842:  50%|████▉     | 3212/6470 [36:58<38:53,  1.40it/s]
3|3213|Loss: 0.8911380767822266:  50%|████▉     | 3212/6470 [36:59<38:53,  1.40it/s]
3|3213|Loss: 0.8911380767822266:  50%|████▉     | 3213/6470 [36:59<39:08,  1.39it/s]
3|3214|Loss: 0.9600741267204285:  50%|████▉     | 3213/6470 [36:59<39:08,  1.39it/s]
3|3214|Loss: 0.9600741267204285:  50%|████▉     | 3214/6470 [37:00<38:12,  1.42it/s]
3|3215|Loss: 0.7502260804176331:  50%|████▉     | 3214/6470 [37:00<38:12,  1.42it/s]
3|3215|Loss: 0.7502260804176331:  50%|████▉     | 3215/6470 [37:00<36:27,  1.49it/s]
3|3216|Loss: 0.8447403311729431:  50%|████▉     | 3215/6470 [37:01<36:27,  1.49it/s]
3|3216|Loss: 0.8447403311729431:  50%|████▉     | 3216/6470 [37:01<35:24,  1.53it/s]
3|3217|Loss: 0.7981708645820618:  50%|████▉     | 3216/6470 [37:01<35:24,  1.53it/s]
3|3217|Loss: 0.7981708645820618:  50%|████▉     | 3217/6470 [37:02<36:27,  1.49it/s]
3|3218|Loss: 0.8678626418113708:  50%|████▉     | 3217/6470 [37:02<36:27,  1.49it/s]
3|3218|Loss: 0.8678626418113708:  50%|████▉     | 3218/6470 [37:02<34:53,  1.55it/s]
3|3219|Loss: 0.8216235637664795:  50%|████▉     | 3218/6470 [37:02<34:53,  1.55it/s]
3|3219|Loss: 0.8216235637664795:  50%|████▉     | 3219/6470 [37:03<33:01,  1.64it/s]
3|3220|Loss: 1.0716753005981445:  50%|████▉     | 3219/6470 [37:03<33:01,  1.64it/s]
3|3220|Loss: 1.0716753005981445:  50%|████▉     | 3220/6470 [37:03<32:14,  1.68it/s]
3|3221|Loss: 1.2830901145935059:  50%|████▉     | 3220/6470 [37:04<32:14,  1.68it/s]
3|3221|Loss: 1.2830901145935059:  50%|████▉     | 3221/6470 [37:04<41:21,  1.31it/s]
3|3222|Loss: 0.7727205157279968:  50%|████▉     | 3221/6470 [37:05<41:21,  1.31it/s]
3|3222|Loss: 0.7727205157279968:  50%|████▉     | 3222/6470 [37:05<41:19,  1.31it/s]
3|3223|Loss: 0.9100215435028076:  50%|████▉     | 3222/6470 [37:06<41:19,  1.31it/s]
3|3223|Loss: 0.9100215435028076:  50%|████▉     | 3223/6470 [37:06<42:15,  1.28it/s]
3|3224|Loss: 0.8204728364944458:  50%|████▉     | 3223/6470 [37:06<42:15,  1.28it/s]
3|3224|Loss: 0.8204728364944458:  50%|████▉     | 3224/6470 [37:07<40:24,  1.34it/s]
3|3225|Loss: 0.9163491725921631:  50%|████▉     | 3224/6470 [37:07<40:24,  1.34it/s]
3|3225|Loss: 0.9163491725921631:  50%|████▉     | 3225/6470 [37:07<39:55,  1.35it/s]
3|3226|Loss: 0.8516077399253845:  50%|████▉     | 3225/6470 [37:08<39:55,  1.35it/s]
3|3226|Loss: 0.8516077399253845:  50%|████▉     | 3226/6470 [37:08<40:38,  1.33it/s]
3|3227|Loss: 0.8024817109107971:  50%|████▉     | 3226/6470 [37:09<40:38,  1.33it/s]
3|3227|Loss: 0.8024817109107971:  50%|████▉     | 3227/6470 [37:09<41:32,  1.30it/s]
3|3228|Loss: 0.691921591758728:  50%|████▉     | 3227/6470 [37:09<41:32,  1.30it/s] 
3|3228|Loss: 0.691921591758728:  50%|████▉     | 3228/6470 [37:10<44:32,  1.21it/s]
3|3229|Loss: 0.8145039081573486:  50%|████▉     | 3228/6470 [37:10<44:32,  1.21it/s]
3|3229|Loss: 0.8145039081573486:  50%|████▉     | 3229/6470 [37:11<45:08,  1.20it/s]
3|3230|Loss: 0.8633269667625427:  50%|████▉     | 3229/6470 [37:11<45:08,  1.20it/s]
3|3230|Loss: 0.8633269667625427:  50%|████▉     | 3230/6470 [37:12<45:37,  1.18it/s]
3|3231|Loss: 0.8597314953804016:  50%|████▉     | 3230/6470 [37:12<45:37,  1.18it/s]
3|3231|Loss: 0.8597314953804016:  50%|████▉     | 3231/6470 [37:13<46:41,  1.16it/s]
3|3232|Loss: 0.8982132077217102:  50%|████▉     | 3231/6470 [37:13<46:41,  1.16it/s]
3|3232|Loss: 0.8982132077217102:  50%|████▉     | 3232/6470 [37:13<45:29,  1.19it/s]
3|3233|Loss: 0.8024535775184631:  50%|████▉     | 3232/6470 [37:14<45:29,  1.19it/s]
3|3233|Loss: 0.8024535775184631:  50%|████▉     | 3233/6470 [37:14<44:02,  1.22it/s]
3|3234|Loss: 0.8405976891517639:  50%|████▉     | 3233/6470 [37:14<44:02,  1.22it/s]
3|3234|Loss: 0.8405976891517639:  50%|████▉     | 3234/6470 [37:15<44:19,  1.22it/s]
3|3235|Loss: 0.7851617932319641:  50%|████▉     | 3234/6470 [37:15<44:19,  1.22it/s]
3|3235|Loss: 0.7851617932319641:  50%|█████     | 3235/6470 [37:16<47:44,  1.13it/s]
3|3236|Loss: 0.7042028903961182:  50%|█████     | 3235/6470 [37:16<47:44,  1.13it/s]
3|3236|Loss: 0.7042028903961182:  50%|█████     | 3236/6470 [37:17<48:48,  1.10it/s]
3|3237|Loss: 0.8984336853027344:  50%|█████     | 3236/6470 [37:17<48:48,  1.10it/s]
3|3237|Loss: 0.8984336853027344:  50%|█████     | 3237/6470 [37:18<46:48,  1.15it/s]
3|3238|Loss: 0.8179753422737122:  50%|█████     | 3237/6470 [37:18<46:48,  1.15it/s]
3|3238|Loss: 0.8179753422737122:  50%|█████     | 3238/6470 [37:18<42:01,  1.28it/s]
3|3239|Loss: 0.9196501970291138:  50%|█████     | 3238/6470 [37:19<42:01,  1.28it/s]
3|3239|Loss: 0.9196501970291138:  50%|█████     | 3239/6470 [37:19<40:34,  1.33it/s]
3|3240|Loss: 0.8939208984375:  50%|█████     | 3239/6470 [37:19<40:34,  1.33it/s]   
3|3240|Loss: 0.8939208984375:  50%|█████     | 3240/6470 [37:20<39:05,  1.38it/s]
3|3241|Loss: 0.802783191204071:  50%|█████     | 3240/6470 [37:20<39:05,  1.38it/s]
3|3241|Loss: 0.802783191204071:  50%|█████     | 3241/6470 [37:20<37:37,  1.43it/s]
3|3242|Loss: 0.8368074893951416:  50%|█████     | 3241/6470 [37:21<37:37,  1.43it/s]
3|3242|Loss: 0.8368074893951416:  50%|█████     | 3242/6470 [37:21<37:58,  1.42it/s]
3|3243|Loss: 0.8748220801353455:  50%|█████     | 3242/6470 [37:21<37:58,  1.42it/s]
3|3243|Loss: 0.8748220801353455:  50%|█████     | 3243/6470 [37:22<37:18,  1.44it/s]
3|3244|Loss: 0.8263459801673889:  50%|█████     | 3243/6470 [37:22<37:18,  1.44it/s]
3|3244|Loss: 0.8263459801673889:  50%|█████     | 3244/6470 [37:22<35:48,  1.50it/s]
3|3245|Loss: 0.8591384291648865:  50%|█████     | 3244/6470 [37:23<35:48,  1.50it/s]
3|3245|Loss: 0.8591384291648865:  50%|█████     | 3245/6470 [37:23<35:58,  1.49it/s]
3|3246|Loss: 0.8676026463508606:  50%|█████     | 3245/6470 [37:23<35:58,  1.49it/s]
3|3246|Loss: 0.8676026463508606:  50%|█████     | 3246/6470 [37:24<36:32,  1.47it/s]
3|3247|Loss: 0.7986848950386047:  50%|█████     | 3246/6470 [37:24<36:32,  1.47it/s]
3|3247|Loss: 0.7986848950386047:  50%|█████     | 3247/6470 [37:24<38:34,  1.39it/s]
3|3248|Loss: 0.8856044411659241:  50%|█████     | 3247/6470 [37:25<38:34,  1.39it/s]
3|3248|Loss: 0.8856044411659241:  50%|█████     | 3248/6470 [37:25<37:17,  1.44it/s]
3|3249|Loss: 0.8077020645141602:  50%|█████     | 3248/6470 [37:25<37:17,  1.44it/s]
3|3249|Loss: 0.8077020645141602:  50%|█████     | 3249/6470 [37:26<37:25,  1.43it/s]
3|3250|Loss: 0.8807369470596313:  50%|█████     | 3249/6470 [37:26<37:25,  1.43it/s]
3|3250|Loss: 0.8807369470596313:  50%|█████     | 3250/6470 [37:26<36:49,  1.46it/s]
3|3251|Loss: 0.8107410073280334:  50%|█████     | 3250/6470 [37:27<36:49,  1.46it/s]
3|3251|Loss: 0.8107410073280334:  50%|█████     | 3251/6470 [37:27<38:39,  1.39it/s]
3|3252|Loss: 0.805212140083313:  50%|█████     | 3251/6470 [37:28<38:39,  1.39it/s] 
3|3252|Loss: 0.805212140083313:  50%|█████     | 3252/6470 [37:28<39:33,  1.36it/s]
3|3253|Loss: 0.9071275591850281:  50%|█████     | 3252/6470 [37:28<39:33,  1.36it/s]
3|3253|Loss: 0.9071275591850281:  50%|█████     | 3253/6470 [37:29<39:20,  1.36it/s]
3|3254|Loss: 0.7805184721946716:  50%|█████     | 3253/6470 [37:29<39:20,  1.36it/s]
3|3254|Loss: 0.7805184721946716:  50%|█████     | 3254/6470 [37:29<38:44,  1.38it/s]
3|3255|Loss: 0.8141152858734131:  50%|█████     | 3254/6470 [37:30<38:44,  1.38it/s]
3|3255|Loss: 0.8141152858734131:  50%|█████     | 3255/6470 [37:30<36:05,  1.48it/s]
3|3256|Loss: 0.7927050590515137:  50%|█████     | 3255/6470 [37:30<36:05,  1.48it/s]
3|3256|Loss: 0.7927050590515137:  50%|█████     | 3256/6470 [37:30<33:31,  1.60it/s]
3|3257|Loss: 0.8287705183029175:  50%|█████     | 3256/6470 [37:31<33:31,  1.60it/s]
3|3257|Loss: 0.8287705183029175:  50%|█████     | 3257/6470 [37:31<33:19,  1.61it/s]
3|3258|Loss: 0.9298950433731079:  50%|█████     | 3257/6470 [37:31<33:19,  1.61it/s]
3|3258|Loss: 0.9298950433731079:  50%|█████     | 3258/6470 [37:32<32:52,  1.63it/s]
3|3259|Loss: 0.7917183637619019:  50%|█████     | 3258/6470 [37:32<32:52,  1.63it/s]
3|3259|Loss: 0.7917183637619019:  50%|█████     | 3259/6470 [37:32<34:36,  1.55it/s]
3|3260|Loss: 0.8606610298156738:  50%|█████     | 3259/6470 [37:33<34:36,  1.55it/s]
3|3260|Loss: 0.8606610298156738:  50%|█████     | 3260/6470 [37:33<35:49,  1.49it/s]
3|3261|Loss: 0.8211787939071655:  50%|█████     | 3260/6470 [37:34<35:49,  1.49it/s]
3|3261|Loss: 0.8211787939071655:  50%|█████     | 3261/6470 [37:34<38:13,  1.40it/s]
3|3262|Loss: 0.7912905812263489:  50%|█████     | 3261/6470 [37:34<38:13,  1.40it/s]
3|3262|Loss: 0.7912905812263489:  50%|█████     | 3262/6470 [37:35<35:49,  1.49it/s]
3|3263|Loss: 0.8826189637184143:  50%|█████     | 3262/6470 [37:35<35:49,  1.49it/s]
3|3263|Loss: 0.8826189637184143:  50%|█████     | 3263/6470 [37:35<37:32,  1.42it/s]
3|3264|Loss: 0.8995662927627563:  50%|█████     | 3263/6470 [37:36<37:32,  1.42it/s]
3|3264|Loss: 0.8995662927627563:  50%|█████     | 3264/6470 [37:36<35:57,  1.49it/s]
3|3265|Loss: 0.7907961010932922:  50%|█████     | 3264/6470 [37:36<35:57,  1.49it/s]
3|3265|Loss: 0.7907961010932922:  50%|█████     | 3265/6470 [37:37<37:04,  1.44it/s]
3|3266|Loss: 0.8070773482322693:  50%|█████     | 3265/6470 [37:37<37:04,  1.44it/s]
3|3266|Loss: 0.8070773482322693:  50%|█████     | 3266/6470 [37:37<36:46,  1.45it/s]
3|3267|Loss: 0.8550069332122803:  50%|█████     | 3266/6470 [37:38<36:46,  1.45it/s]
3|3267|Loss: 0.8550069332122803:  50%|█████     | 3267/6470 [37:38<37:07,  1.44it/s]
3|3268|Loss: 0.8179747462272644:  50%|█████     | 3267/6470 [37:38<37:07,  1.44it/s]
3|3268|Loss: 0.8179747462272644:  51%|█████     | 3268/6470 [37:39<38:54,  1.37it/s]
3|3269|Loss: 0.9977741241455078:  51%|█████     | 3268/6470 [37:39<38:54,  1.37it/s]
3|3269|Loss: 0.9977741241455078:  51%|█████     | 3269/6470 [37:40<39:17,  1.36it/s]
3|3270|Loss: 0.9676058888435364:  51%|█████     | 3269/6470 [37:40<39:17,  1.36it/s]
3|3270|Loss: 0.9676058888435364:  51%|█████     | 3270/6470 [37:40<40:44,  1.31it/s]
3|3271|Loss: 0.9117298722267151:  51%|█████     | 3270/6470 [37:41<40:44,  1.31it/s]
3|3271|Loss: 0.9117298722267151:  51%|█████     | 3271/6470 [37:41<40:45,  1.31it/s]
3|3272|Loss: 1.0066428184509277:  51%|█████     | 3271/6470 [37:42<40:45,  1.31it/s]
3|3272|Loss: 1.0066428184509277:  51%|█████     | 3272/6470 [37:42<40:24,  1.32it/s]
3|3273|Loss: 0.99913489818573:  51%|█████     | 3272/6470 [37:42<40:24,  1.32it/s]  
3|3273|Loss: 0.99913489818573:  51%|█████     | 3273/6470 [37:43<40:06,  1.33it/s]
3|3274|Loss: 0.8475900292396545:  51%|█████     | 3273/6470 [37:43<40:06,  1.33it/s]
3|3274|Loss: 0.8475900292396545:  51%|█████     | 3274/6470 [37:43<38:41,  1.38it/s]
3|3275|Loss: 0.8357236385345459:  51%|█████     | 3274/6470 [37:44<38:41,  1.38it/s]
3|3275|Loss: 0.8357236385345459:  51%|█████     | 3275/6470 [37:44<38:41,  1.38it/s]
3|3276|Loss: 0.7700717449188232:  51%|█████     | 3275/6470 [37:44<38:41,  1.38it/s]
3|3276|Loss: 0.7700717449188232:  51%|█████     | 3276/6470 [37:45<36:46,  1.45it/s]
3|3277|Loss: 0.7718866467475891:  51%|█████     | 3276/6470 [37:45<36:46,  1.45it/s]
3|3277|Loss: 0.7718866467475891:  51%|█████     | 3277/6470 [37:45<36:30,  1.46it/s]
3|3278|Loss: 0.907891571521759:  51%|█████     | 3277/6470 [37:46<36:30,  1.46it/s] 
3|3278|Loss: 0.907891571521759:  51%|█████     | 3278/6470 [37:46<38:04,  1.40it/s]
3|3279|Loss: 0.7952589392662048:  51%|█████     | 3278/6470 [37:47<38:04,  1.40it/s]
3|3279|Loss: 0.7952589392662048:  51%|█████     | 3279/6470 [37:47<38:45,  1.37it/s]
3|3280|Loss: 0.8998539447784424:  51%|█████     | 3279/6470 [37:47<38:45,  1.37it/s]
3|3280|Loss: 0.8998539447784424:  51%|█████     | 3280/6470 [37:48<39:34,  1.34it/s]
3|3281|Loss: 0.9704055786132812:  51%|█████     | 3280/6470 [37:48<39:34,  1.34it/s]
3|3281|Loss: 0.9704055786132812:  51%|█████     | 3281/6470 [37:48<38:25,  1.38it/s]
3|3282|Loss: 1.097779393196106:  51%|█████     | 3281/6470 [37:49<38:25,  1.38it/s] 
3|3282|Loss: 1.097779393196106:  51%|█████     | 3282/6470 [37:49<37:06,  1.43it/s]
3|3283|Loss: 0.8955574035644531:  51%|█████     | 3282/6470 [37:49<37:06,  1.43it/s]
3|3283|Loss: 0.8955574035644531:  51%|█████     | 3283/6470 [37:50<34:31,  1.54it/s]
3|3284|Loss: 1.0637844800949097:  51%|█████     | 3283/6470 [37:50<34:31,  1.54it/s]
3|3284|Loss: 1.0637844800949097:  51%|█████     | 3284/6470 [37:50<36:18,  1.46it/s]
3|3285|Loss: 0.8387082815170288:  51%|█████     | 3284/6470 [37:51<36:18,  1.46it/s]
3|3285|Loss: 0.8387082815170288:  51%|█████     | 3285/6470 [37:51<37:18,  1.42it/s]
3|3286|Loss: 0.8855244517326355:  51%|█████     | 3285/6470 [37:51<37:18,  1.42it/s]
3|3286|Loss: 0.8855244517326355:  51%|█████     | 3286/6470 [37:52<38:04,  1.39it/s]
3|3287|Loss: 0.9503238797187805:  51%|█████     | 3286/6470 [37:52<38:04,  1.39it/s]
3|3287|Loss: 0.9503238797187805:  51%|█████     | 3287/6470 [37:53<39:42,  1.34it/s]
3|3288|Loss: 0.8034507632255554:  51%|█████     | 3287/6470 [37:53<39:42,  1.34it/s]
3|3288|Loss: 0.8034507632255554:  51%|█████     | 3288/6470 [37:53<41:29,  1.28it/s]
3|3289|Loss: 0.9280387163162231:  51%|█████     | 3288/6470 [37:54<41:29,  1.28it/s]
3|3289|Loss: 0.9280387163162231:  51%|█████     | 3289/6470 [37:54<39:43,  1.33it/s]
3|3290|Loss: 0.757400631904602:  51%|█████     | 3289/6470 [37:54<39:43,  1.33it/s] 
3|3290|Loss: 0.757400631904602:  51%|█████     | 3290/6470 [37:55<38:34,  1.37it/s]
3|3291|Loss: 0.8330581784248352:  51%|█████     | 3290/6470 [37:55<38:34,  1.37it/s]
3|3291|Loss: 0.8330581784248352:  51%|█████     | 3291/6470 [37:55<36:13,  1.46it/s]
3|3292|Loss: 1.030779242515564:  51%|█████     | 3291/6470 [37:56<36:13,  1.46it/s] 
3|3292|Loss: 1.030779242515564:  51%|█████     | 3292/6470 [37:56<34:55,  1.52it/s]
3|3293|Loss: 0.699137270450592:  51%|█████     | 3292/6470 [37:56<34:55,  1.52it/s]
3|3293|Loss: 0.699137270450592:  51%|█████     | 3293/6470 [37:57<35:18,  1.50it/s]
3|3294|Loss: 0.8657939434051514:  51%|█████     | 3293/6470 [37:57<35:18,  1.50it/s]
3|3294|Loss: 0.8657939434051514:  51%|█████     | 3294/6470 [37:57<35:54,  1.47it/s]
3|3295|Loss: 0.8322046995162964:  51%|█████     | 3294/6470 [37:58<35:54,  1.47it/s]
3|3295|Loss: 0.8322046995162964:  51%|█████     | 3295/6470 [37:58<35:15,  1.50it/s]
3|3296|Loss: 0.9177901744842529:  51%|█████     | 3295/6470 [37:58<35:15,  1.50it/s]
3|3296|Loss: 0.9177901744842529:  51%|█████     | 3296/6470 [37:59<33:57,  1.56it/s]
3|3297|Loss: 0.7843878269195557:  51%|█████     | 3296/6470 [37:59<33:57,  1.56it/s]
3|3297|Loss: 0.7843878269195557:  51%|█████     | 3297/6470 [37:59<33:47,  1.56it/s]
3|3298|Loss: 0.9128453731536865:  51%|█████     | 3297/6470 [38:00<33:47,  1.56it/s]
3|3298|Loss: 0.9128453731536865:  51%|█████     | 3298/6470 [38:00<33:57,  1.56it/s]
3|3299|Loss: 0.8728964328765869:  51%|█████     | 3298/6470 [38:00<33:57,  1.56it/s]
3|3299|Loss: 0.8728964328765869:  51%|█████     | 3299/6470 [38:00<32:32,  1.62it/s]
3|3300|Loss: 0.8545829653739929:  51%|█████     | 3299/6470 [38:01<32:32,  1.62it/s]
3|3300|Loss: 0.8545829653739929:  51%|█████     | 3300/6470 [38:01<29:23,  1.80it/s]
3|3301|Loss: 0.691622793674469:  51%|█████     | 3300/6470 [38:01<29:23,  1.80it/s] 
3|3301|Loss: 0.691622793674469:  51%|█████     | 3301/6470 [38:01<28:56,  1.83it/s]
3|3302|Loss: 0.8811960816383362:  51%|█████     | 3301/6470 [38:02<28:56,  1.83it/s]
3|3302|Loss: 0.8811960816383362:  51%|█████     | 3302/6470 [38:02<30:08,  1.75it/s]
3|3303|Loss: 0.9626826047897339:  51%|█████     | 3302/6470 [38:02<30:08,  1.75it/s]
3|3303|Loss: 0.9626826047897339:  51%|█████     | 3303/6470 [38:03<32:15,  1.64it/s]
3|3304|Loss: 0.9343474507331848:  51%|█████     | 3303/6470 [38:03<32:15,  1.64it/s]
3|3304|Loss: 0.9343474507331848:  51%|█████     | 3304/6470 [38:03<30:31,  1.73it/s]
3|3305|Loss: 0.8074671626091003:  51%|█████     | 3304/6470 [38:04<30:31,  1.73it/s]
3|3305|Loss: 0.8074671626091003:  51%|█████     | 3305/6470 [38:04<31:54,  1.65it/s]
3|3306|Loss: 0.8440508842468262:  51%|█████     | 3305/6470 [38:04<31:54,  1.65it/s]
3|3306|Loss: 0.8440508842468262:  51%|█████     | 3306/6470 [38:04<31:57,  1.65it/s]
3|3307|Loss: 0.9149823188781738:  51%|█████     | 3306/6470 [38:05<31:57,  1.65it/s]
3|3307|Loss: 0.9149823188781738:  51%|█████     | 3307/6470 [38:05<31:59,  1.65it/s]
3|3308|Loss: 0.9341598153114319:  51%|█████     | 3307/6470 [38:05<31:59,  1.65it/s]
3|3308|Loss: 0.9341598153114319:  51%|█████     | 3308/6470 [38:06<33:31,  1.57it/s]
3|3309|Loss: 0.8846666216850281:  51%|█████     | 3308/6470 [38:06<33:31,  1.57it/s]
3|3309|Loss: 0.8846666216850281:  51%|█████     | 3309/6470 [38:07<35:10,  1.50it/s]
3|3310|Loss: 1.1069999933242798:  51%|█████     | 3309/6470 [38:07<35:10,  1.50it/s]
3|3310|Loss: 1.1069999933242798:  51%|█████     | 3310/6470 [38:07<36:17,  1.45it/s]
3|3311|Loss: 0.924848735332489:  51%|█████     | 3310/6470 [38:08<36:17,  1.45it/s] 
3|3311|Loss: 0.924848735332489:  51%|█████     | 3311/6470 [38:08<37:13,  1.41it/s]
3|3312|Loss: 0.6455246210098267:  51%|█████     | 3311/6470 [38:09<37:13,  1.41it/s]
3|3312|Loss: 0.6455246210098267:  51%|█████     | 3312/6470 [38:09<40:06,  1.31it/s]
3|3313|Loss: 0.9197207689285278:  51%|█████     | 3312/6470 [38:09<40:06,  1.31it/s]
3|3313|Loss: 0.9197207689285278:  51%|█████     | 3313/6470 [38:10<40:28,  1.30it/s]
3|3314|Loss: 0.8302642703056335:  51%|█████     | 3313/6470 [38:10<40:28,  1.30it/s]
3|3314|Loss: 0.8302642703056335:  51%|█████     | 3314/6470 [38:10<39:41,  1.33it/s]
3|3315|Loss: 0.9078548550605774:  51%|█████     | 3314/6470 [38:11<39:41,  1.33it/s]
3|3315|Loss: 0.9078548550605774:  51%|█████     | 3315/6470 [38:11<37:15,  1.41it/s]
3|3316|Loss: 0.5598336458206177:  51%|█████     | 3315/6470 [38:11<37:15,  1.41it/s]
3|3316|Loss: 0.5598336458206177:  51%|█████▏    | 3316/6470 [38:12<38:25,  1.37it/s]
3|3317|Loss: 0.8358249068260193:  51%|█████▏    | 3316/6470 [38:12<38:25,  1.37it/s]
3|3317|Loss: 0.8358249068260193:  51%|█████▏    | 3317/6470 [38:12<37:39,  1.40it/s]
3|3318|Loss: 0.8311342597007751:  51%|█████▏    | 3317/6470 [38:13<37:39,  1.40it/s]
3|3318|Loss: 0.8311342597007751:  51%|█████▏    | 3318/6470 [38:13<37:18,  1.41it/s]
3|3319|Loss: 0.9021700024604797:  51%|█████▏    | 3318/6470 [38:14<37:18,  1.41it/s]
3|3319|Loss: 0.9021700024604797:  51%|█████▏    | 3319/6470 [38:14<38:44,  1.36it/s]
3|3320|Loss: 0.8881967067718506:  51%|█████▏    | 3319/6470 [38:14<38:44,  1.36it/s]
3|3320|Loss: 0.8881967067718506:  51%|█████▏    | 3320/6470 [38:15<39:00,  1.35it/s]
3|3321|Loss: 0.6568261384963989:  51%|█████▏    | 3320/6470 [38:15<39:00,  1.35it/s]
3|3321|Loss: 0.6568261384963989:  51%|█████▏    | 3321/6470 [38:15<34:15,  1.53it/s]
3|3322|Loss: 0.8576774597167969:  51%|█████▏    | 3321/6470 [38:16<34:15,  1.53it/s]
3|3322|Loss: 0.8576774597167969:  51%|█████▏    | 3322/6470 [38:16<35:12,  1.49it/s]
3|3323|Loss: 0.5275880098342896:  51%|█████▏    | 3322/6470 [38:16<35:12,  1.49it/s]
3|3323|Loss: 0.5275880098342896:  51%|█████▏    | 3323/6470 [38:17<36:39,  1.43it/s]
3|3324|Loss: 0.7847515344619751:  51%|█████▏    | 3323/6470 [38:17<36:39,  1.43it/s]
3|3324|Loss: 0.7847515344619751:  51%|█████▏    | 3324/6470 [38:17<38:33,  1.36it/s]
3|3325|Loss: 0.826058566570282:  51%|█████▏    | 3324/6470 [38:18<38:33,  1.36it/s] 
3|3325|Loss: 0.826058566570282:  51%|█████▏    | 3325/6470 [38:18<37:40,  1.39it/s]
3|3326|Loss: 0.8970491886138916:  51%|█████▏    | 3325/6470 [38:19<37:40,  1.39it/s]
3|3326|Loss: 0.8970491886138916:  51%|█████▏    | 3326/6470 [38:19<36:27,  1.44it/s]
3|3327|Loss: 0.8441221117973328:  51%|█████▏    | 3326/6470 [38:19<36:27,  1.44it/s]
3|3327|Loss: 0.8441221117973328:  51%|█████▏    | 3327/6470 [38:19<35:18,  1.48it/s]
3|3328|Loss: 0.8752694725990295:  51%|█████▏    | 3327/6470 [38:20<35:18,  1.48it/s]
3|3328|Loss: 0.8752694725990295:  51%|█████▏    | 3328/6470 [38:20<34:48,  1.50it/s]
3|3329|Loss: 0.9066113233566284:  51%|█████▏    | 3328/6470 [38:20<34:48,  1.50it/s]
3|3329|Loss: 0.9066113233566284:  51%|█████▏    | 3329/6470 [38:21<35:48,  1.46it/s]
3|3330|Loss: 0.8025351166725159:  51%|█████▏    | 3329/6470 [38:21<35:48,  1.46it/s]
3|3330|Loss: 0.8025351166725159:  51%|█████▏    | 3330/6470 [38:22<36:17,  1.44it/s]
3|3331|Loss: 0.833933413028717:  51%|█████▏    | 3330/6470 [38:22<36:17,  1.44it/s] 
3|3331|Loss: 0.833933413028717:  51%|█████▏    | 3331/6470 [38:22<35:42,  1.47it/s]
3|3332|Loss: 0.850016176700592:  51%|█████▏    | 3331/6470 [38:23<35:42,  1.47it/s]
3|3332|Loss: 0.850016176700592:  51%|█████▏    | 3332/6470 [38:23<36:53,  1.42it/s]
3|3333|Loss: 0.9200519323348999:  51%|█████▏    | 3332/6470 [38:23<36:53,  1.42it/s]
3|3333|Loss: 0.9200519323348999:  52%|█████▏    | 3333/6470 [38:24<36:31,  1.43it/s]
3|3334|Loss: 0.7105871438980103:  52%|█████▏    | 3333/6470 [38:24<36:31,  1.43it/s]
3|3334|Loss: 0.7105871438980103:  52%|█████▏    | 3334/6470 [38:24<36:46,  1.42it/s]
3|3335|Loss: 0.8709375262260437:  52%|█████▏    | 3334/6470 [38:25<36:46,  1.42it/s]
3|3335|Loss: 0.8709375262260437:  52%|█████▏    | 3335/6470 [38:25<37:22,  1.40it/s]
3|3336|Loss: 0.7822874784469604:  52%|█████▏    | 3335/6470 [38:25<37:22,  1.40it/s]
3|3336|Loss: 0.7822874784469604:  52%|█████▏    | 3336/6470 [38:26<37:22,  1.40it/s]
3|3337|Loss: 0.7503575682640076:  52%|█████▏    | 3336/6470 [38:26<37:22,  1.40it/s]
3|3337|Loss: 0.7503575682640076:  52%|█████▏    | 3337/6470 [38:27<37:34,  1.39it/s]
3|3338|Loss: 0.9664180278778076:  52%|█████▏    | 3337/6470 [38:27<37:34,  1.39it/s]
3|3338|Loss: 0.9664180278778076:  52%|█████▏    | 3338/6470 [38:27<37:21,  1.40it/s]
3|3339|Loss: 0.9109019637107849:  52%|█████▏    | 3338/6470 [38:28<37:21,  1.40it/s]
3|3339|Loss: 0.9109019637107849:  52%|█████▏    | 3339/6470 [38:28<37:31,  1.39it/s]
3|3340|Loss: 0.9890992641448975:  52%|█████▏    | 3339/6470 [38:28<37:31,  1.39it/s]
3|3340|Loss: 0.9890992641448975:  52%|█████▏    | 3340/6470 [38:29<36:21,  1.43it/s]
3|3341|Loss: 0.8182401061058044:  52%|█████▏    | 3340/6470 [38:29<36:21,  1.43it/s]
3|3341|Loss: 0.8182401061058044:  52%|█████▏    | 3341/6470 [38:29<36:31,  1.43it/s]
3|3342|Loss: 0.8381206393241882:  52%|█████▏    | 3341/6470 [38:30<36:31,  1.43it/s]
3|3342|Loss: 0.8381206393241882:  52%|█████▏    | 3342/6470 [38:30<36:59,  1.41it/s]
3|3343|Loss: 1.0027806758880615:  52%|█████▏    | 3342/6470 [38:30<36:59,  1.41it/s]
3|3343|Loss: 1.0027806758880615:  52%|█████▏    | 3343/6470 [38:31<37:32,  1.39it/s]
3|3344|Loss: 0.9960163831710815:  52%|█████▏    | 3343/6470 [38:31<37:32,  1.39it/s]
3|3344|Loss: 0.9960163831710815:  52%|█████▏    | 3344/6470 [38:31<35:37,  1.46it/s]
3|3345|Loss: 0.7965596914291382:  52%|█████▏    | 3344/6470 [38:32<35:37,  1.46it/s]
3|3345|Loss: 0.7965596914291382:  52%|█████▏    | 3345/6470 [38:32<35:56,  1.45it/s]
3|3346|Loss: 0.80865079164505:  52%|█████▏    | 3345/6470 [38:33<35:56,  1.45it/s]  
3|3346|Loss: 0.80865079164505:  52%|█████▏    | 3346/6470 [38:33<37:29,  1.39it/s]
3|3347|Loss: 0.8532540798187256:  52%|█████▏    | 3346/6470 [38:33<37:29,  1.39it/s]
3|3347|Loss: 0.8532540798187256:  52%|█████▏    | 3347/6470 [38:33<35:08,  1.48it/s]
3|3348|Loss: 0.6617656946182251:  52%|█████▏    | 3347/6470 [38:34<35:08,  1.48it/s]
3|3348|Loss: 0.6617656946182251:  52%|█████▏    | 3348/6470 [38:34<33:46,  1.54it/s]
3|3349|Loss: 0.7527914643287659:  52%|█████▏    | 3348/6470 [38:34<33:46,  1.54it/s]
3|3349|Loss: 0.7527914643287659:  52%|█████▏    | 3349/6470 [38:35<34:36,  1.50it/s]
3|3350|Loss: 0.7771206498146057:  52%|█████▏    | 3349/6470 [38:35<34:36,  1.50it/s]
3|3350|Loss: 0.7771206498146057:  52%|█████▏    | 3350/6470 [38:35<33:19,  1.56it/s]
3|3351|Loss: 0.8003365397453308:  52%|█████▏    | 3350/6470 [38:36<33:19,  1.56it/s]
3|3351|Loss: 0.8003365397453308:  52%|█████▏    | 3351/6470 [38:36<33:05,  1.57it/s]
3|3352|Loss: 0.7030436396598816:  52%|█████▏    | 3351/6470 [38:36<33:05,  1.57it/s]
3|3352|Loss: 0.7030436396598816:  52%|█████▏    | 3352/6470 [38:37<37:16,  1.39it/s]
3|3353|Loss: 0.9802460670471191:  52%|█████▏    | 3352/6470 [38:37<37:16,  1.39it/s]
3|3353|Loss: 0.9802460670471191:  52%|█████▏    | 3353/6470 [38:38<39:03,  1.33it/s]
3|3354|Loss: 0.5789986848831177:  52%|█████▏    | 3353/6470 [38:38<39:03,  1.33it/s]
3|3354|Loss: 0.5789986848831177:  52%|█████▏    | 3354/6470 [38:38<35:50,  1.45it/s]
3|3355|Loss: 0.8940703868865967:  52%|█████▏    | 3354/6470 [38:39<35:50,  1.45it/s]
3|3355|Loss: 0.8940703868865967:  52%|█████▏    | 3355/6470 [38:39<35:20,  1.47it/s]
3|3356|Loss: 0.8370205760002136:  52%|█████▏    | 3355/6470 [38:39<35:20,  1.47it/s]
3|3356|Loss: 0.8370205760002136:  52%|█████▏    | 3356/6470 [38:40<35:03,  1.48it/s]
3|3357|Loss: 0.8714675307273865:  52%|█████▏    | 3356/6470 [38:40<35:03,  1.48it/s]
3|3357|Loss: 0.8714675307273865:  52%|█████▏    | 3357/6470 [38:40<34:26,  1.51it/s]
3|3358|Loss: 0.8951736688613892:  52%|█████▏    | 3357/6470 [38:41<34:26,  1.51it/s]
3|3358|Loss: 0.8951736688613892:  52%|█████▏    | 3358/6470 [38:41<32:58,  1.57it/s]
3|3359|Loss: 0.6547642946243286:  52%|█████▏    | 3358/6470 [38:41<32:58,  1.57it/s]
3|3359|Loss: 0.6547642946243286:  52%|█████▏    | 3359/6470 [38:41<31:12,  1.66it/s]
3|3360|Loss: 0.8518868684768677:  52%|█████▏    | 3359/6470 [38:42<31:12,  1.66it/s]
3|3360|Loss: 0.8518868684768677:  52%|█████▏    | 3360/6470 [38:42<30:19,  1.71it/s]
3|3361|Loss: 0.8971534371376038:  52%|█████▏    | 3360/6470 [38:42<30:19,  1.71it/s]
3|3361|Loss: 0.8971534371376038:  52%|█████▏    | 3361/6470 [38:42<30:38,  1.69it/s]
3|3362|Loss: 0.7792645692825317:  52%|█████▏    | 3361/6470 [38:43<30:38,  1.69it/s]
3|3362|Loss: 0.7792645692825317:  52%|█████▏    | 3362/6470 [38:43<29:22,  1.76it/s]
3|3363|Loss: 0.9241482615470886:  52%|█████▏    | 3362/6470 [38:43<29:22,  1.76it/s]
3|3363|Loss: 0.9241482615470886:  52%|█████▏    | 3363/6470 [38:44<32:51,  1.58it/s]
3|3364|Loss: 0.874083936214447:  52%|█████▏    | 3363/6470 [38:44<32:51,  1.58it/s] 
3|3364|Loss: 0.874083936214447:  52%|█████▏    | 3364/6470 [38:45<36:08,  1.43it/s]
3|3365|Loss: 0.6852566003799438:  52%|█████▏    | 3364/6470 [38:45<36:08,  1.43it/s]
3|3365|Loss: 0.6852566003799438:  52%|█████▏    | 3365/6470 [38:45<34:17,  1.51it/s]
3|3366|Loss: 0.9428444504737854:  52%|█████▏    | 3365/6470 [38:46<34:17,  1.51it/s]
3|3366|Loss: 0.9428444504737854:  52%|█████▏    | 3366/6470 [38:46<34:51,  1.48it/s]
3|3367|Loss: 0.6689758896827698:  52%|█████▏    | 3366/6470 [38:46<34:51,  1.48it/s]
3|3367|Loss: 0.6689758896827698:  52%|█████▏    | 3367/6470 [38:47<34:46,  1.49it/s]
3|3368|Loss: 0.8341641426086426:  52%|█████▏    | 3367/6470 [38:47<34:46,  1.49it/s]
3|3368|Loss: 0.8341641426086426:  52%|█████▏    | 3368/6470 [38:47<37:17,  1.39it/s]
3|3369|Loss: 0.9093557596206665:  52%|█████▏    | 3368/6470 [38:48<37:17,  1.39it/s]
3|3369|Loss: 0.9093557596206665:  52%|█████▏    | 3369/6470 [38:48<38:04,  1.36it/s]
3|3370|Loss: 0.8846217393875122:  52%|█████▏    | 3369/6470 [38:49<38:04,  1.36it/s]
3|3370|Loss: 0.8846217393875122:  52%|█████▏    | 3370/6470 [38:49<38:14,  1.35it/s]
3|3371|Loss: 0.9517473578453064:  52%|█████▏    | 3370/6470 [38:49<38:14,  1.35it/s]
3|3371|Loss: 0.9517473578453064:  52%|█████▏    | 3371/6470 [38:50<38:30,  1.34it/s]
3|3372|Loss: 0.9050970077514648:  52%|█████▏    | 3371/6470 [38:50<38:30,  1.34it/s]
3|3372|Loss: 0.9050970077514648:  52%|█████▏    | 3372/6470 [38:50<37:16,  1.39it/s]
3|3373|Loss: 1.0012180805206299:  52%|█████▏    | 3372/6470 [38:51<37:16,  1.39it/s]
3|3373|Loss: 1.0012180805206299:  52%|█████▏    | 3373/6470 [38:51<37:06,  1.39it/s]
3|3374|Loss: 0.8031773567199707:  52%|█████▏    | 3373/6470 [38:51<37:06,  1.39it/s]
3|3374|Loss: 0.8031773567199707:  52%|█████▏    | 3374/6470 [38:52<35:22,  1.46it/s]
3|3375|Loss: 0.7277795672416687:  52%|█████▏    | 3374/6470 [38:52<35:22,  1.46it/s]
3|3375|Loss: 0.7277795672416687:  52%|█████▏    | 3375/6470 [38:52<34:04,  1.51it/s]
3|3376|Loss: 0.8400464057922363:  52%|█████▏    | 3375/6470 [38:53<34:04,  1.51it/s]
3|3376|Loss: 0.8400464057922363:  52%|█████▏    | 3376/6470 [38:53<34:29,  1.50it/s]
3|3377|Loss: 0.9875109195709229:  52%|█████▏    | 3376/6470 [38:53<34:29,  1.50it/s]
3|3377|Loss: 0.9875109195709229:  52%|█████▏    | 3377/6470 [38:54<35:03,  1.47it/s]
3|3378|Loss: 0.7511062622070312:  52%|█████▏    | 3377/6470 [38:54<35:03,  1.47it/s]
3|3378|Loss: 0.7511062622070312:  52%|█████▏    | 3378/6470 [38:54<35:53,  1.44it/s]
3|3379|Loss: 0.9638104438781738:  52%|█████▏    | 3378/6470 [38:55<35:53,  1.44it/s]
3|3379|Loss: 0.9638104438781738:  52%|█████▏    | 3379/6470 [38:55<34:42,  1.48it/s]
3|3380|Loss: 0.9057032465934753:  52%|█████▏    | 3379/6470 [38:55<34:42,  1.48it/s]
3|3380|Loss: 0.9057032465934753:  52%|█████▏    | 3380/6470 [38:56<33:45,  1.53it/s]
3|3381|Loss: 0.9485002160072327:  52%|█████▏    | 3380/6470 [38:56<33:45,  1.53it/s]
3|3381|Loss: 0.9485002160072327:  52%|█████▏    | 3381/6470 [38:56<33:14,  1.55it/s]
3|3382|Loss: 0.783391535282135:  52%|█████▏    | 3381/6470 [38:57<33:14,  1.55it/s] 
3|3382|Loss: 0.783391535282135:  52%|█████▏    | 3382/6470 [38:57<34:13,  1.50it/s]
3|3383|Loss: 0.6513203382492065:  52%|█████▏    | 3382/6470 [38:57<34:13,  1.50it/s]
3|3383|Loss: 0.6513203382492065:  52%|█████▏    | 3383/6470 [38:58<34:56,  1.47it/s]
3|3384|Loss: 0.9991000890731812:  52%|█████▏    | 3383/6470 [38:58<34:56,  1.47it/s]
3|3384|Loss: 0.9991000890731812:  52%|█████▏    | 3384/6470 [38:58<36:54,  1.39it/s]
3|3385|Loss: 0.9862759113311768:  52%|█████▏    | 3384/6470 [38:59<36:54,  1.39it/s]
3|3385|Loss: 0.9862759113311768:  52%|█████▏    | 3385/6470 [38:59<36:05,  1.42it/s]
3|3386|Loss: 0.8564659953117371:  52%|█████▏    | 3385/6470 [38:59<36:05,  1.42it/s]
3|3386|Loss: 0.8564659953117371:  52%|█████▏    | 3386/6470 [39:00<33:04,  1.55it/s]
3|3387|Loss: 0.8346467614173889:  52%|█████▏    | 3386/6470 [39:00<33:04,  1.55it/s]
3|3387|Loss: 0.8346467614173889:  52%|█████▏    | 3387/6470 [39:00<31:50,  1.61it/s]
3|3388|Loss: 0.8719545602798462:  52%|█████▏    | 3387/6470 [39:01<31:50,  1.61it/s]
3|3388|Loss: 0.8719545602798462:  52%|█████▏    | 3388/6470 [39:01<34:36,  1.48it/s]
3|3389|Loss: 0.9001201391220093:  52%|█████▏    | 3388/6470 [39:01<34:36,  1.48it/s]
3|3389|Loss: 0.9001201391220093:  52%|█████▏    | 3389/6470 [39:02<36:16,  1.42it/s]
3|3390|Loss: 0.936435878276825:  52%|█████▏    | 3389/6470 [39:02<36:16,  1.42it/s] 
3|3390|Loss: 0.936435878276825:  52%|█████▏    | 3390/6470 [39:02<35:25,  1.45it/s]
3|3391|Loss: 0.9445802569389343:  52%|█████▏    | 3390/6470 [39:03<35:25,  1.45it/s]
3|3391|Loss: 0.9445802569389343:  52%|█████▏    | 3391/6470 [39:03<32:52,  1.56it/s]
3|3392|Loss: 0.8951887488365173:  52%|█████▏    | 3391/6470 [39:03<32:52,  1.56it/s]
3|3392|Loss: 0.8951887488365173:  52%|█████▏    | 3392/6470 [39:04<32:06,  1.60it/s]
3|3393|Loss: 0.9224565625190735:  52%|█████▏    | 3392/6470 [39:04<32:06,  1.60it/s]
3|3393|Loss: 0.9224565625190735:  52%|█████▏    | 3393/6470 [39:04<33:08,  1.55it/s]
3|3394|Loss: 0.9860019087791443:  52%|█████▏    | 3393/6470 [39:05<33:08,  1.55it/s]
3|3394|Loss: 0.9860019087791443:  52%|█████▏    | 3394/6470 [39:05<34:40,  1.48it/s]
3|3395|Loss: 0.8540439009666443:  52%|█████▏    | 3394/6470 [39:05<34:40,  1.48it/s]
3|3395|Loss: 0.8540439009666443:  52%|█████▏    | 3395/6470 [39:06<33:50,  1.51it/s]
3|3396|Loss: 0.8895999193191528:  52%|█████▏    | 3395/6470 [39:06<33:50,  1.51it/s]
3|3396|Loss: 0.8895999193191528:  52%|█████▏    | 3396/6470 [39:06<34:27,  1.49it/s]
3|3397|Loss: 0.9569860696792603:  52%|█████▏    | 3396/6470 [39:07<34:27,  1.49it/s]
3|3397|Loss: 0.9569860696792603:  53%|█████▎    | 3397/6470 [39:07<36:05,  1.42it/s]
3|3398|Loss: 1.0398201942443848:  53%|█████▎    | 3397/6470 [39:08<36:05,  1.42it/s]
3|3398|Loss: 1.0398201942443848:  53%|█████▎    | 3398/6470 [39:08<36:56,  1.39it/s]
3|3399|Loss: 0.9284822940826416:  53%|█████▎    | 3398/6470 [39:08<36:56,  1.39it/s]
3|3399|Loss: 0.9284822940826416:  53%|█████▎    | 3399/6470 [39:08<33:41,  1.52it/s]
3|3400|Loss: 0.7689946293830872:  53%|█████▎    | 3399/6470 [39:09<33:41,  1.52it/s]
3|3400|Loss: 0.7689946293830872:  53%|█████▎    | 3400/6470 [39:09<33:54,  1.51it/s]
3|3401|Loss: 0.8160778880119324:  53%|█████▎    | 3400/6470 [39:09<33:54,  1.51it/s]
3|3401|Loss: 0.8160778880119324:  53%|█████▎    | 3401/6470 [39:10<33:56,  1.51it/s]
3|3402|Loss: 0.8134236335754395:  53%|█████▎    | 3401/6470 [39:10<33:56,  1.51it/s]
3|3402|Loss: 0.8134236335754395:  53%|█████▎    | 3402/6470 [39:11<36:06,  1.42it/s]
3|3403|Loss: 0.7661237120628357:  53%|█████▎    | 3402/6470 [39:11<36:06,  1.42it/s]
3|3403|Loss: 0.7661237120628357:  53%|█████▎    | 3403/6470 [39:11<37:35,  1.36it/s]
3|3404|Loss: 0.7918980121612549:  53%|█████▎    | 3403/6470 [39:12<37:35,  1.36it/s]
3|3404|Loss: 0.7918980121612549:  53%|█████▎    | 3404/6470 [39:12<37:19,  1.37it/s]
3|3405|Loss: 0.8790780901908875:  53%|█████▎    | 3404/6470 [39:12<37:19,  1.37it/s]
3|3405|Loss: 0.8790780901908875:  53%|█████▎    | 3405/6470 [39:13<37:59,  1.34it/s]
3|3406|Loss: 0.7813224792480469:  53%|█████▎    | 3405/6470 [39:13<37:59,  1.34it/s]
3|3406|Loss: 0.7813224792480469:  53%|█████▎    | 3406/6470 [39:14<37:40,  1.36it/s]
3|3407|Loss: 0.969316303730011:  53%|█████▎    | 3406/6470 [39:14<37:40,  1.36it/s] 
3|3407|Loss: 0.969316303730011:  53%|█████▎    | 3407/6470 [39:14<37:28,  1.36it/s]
3|3408|Loss: 0.860582709312439:  53%|█████▎    | 3407/6470 [39:15<37:28,  1.36it/s]
3|3408|Loss: 0.860582709312439:  53%|█████▎    | 3408/6470 [39:15<36:06,  1.41it/s]
3|3409|Loss: 0.7490618228912354:  53%|█████▎    | 3408/6470 [39:15<36:06,  1.41it/s]
3|3409|Loss: 0.7490618228912354:  53%|█████▎    | 3409/6470 [39:16<34:57,  1.46it/s]
3|3410|Loss: 0.9830387830734253:  53%|█████▎    | 3409/6470 [39:16<34:57,  1.46it/s]
3|3410|Loss: 0.9830387830734253:  53%|█████▎    | 3410/6470 [39:16<33:08,  1.54it/s]
3|3411|Loss: 0.8770743012428284:  53%|█████▎    | 3410/6470 [39:16<33:08,  1.54it/s]
3|3411|Loss: 0.8770743012428284:  53%|█████▎    | 3411/6470 [39:17<33:54,  1.50it/s]
3|3412|Loss: 0.6996309161186218:  53%|█████▎    | 3411/6470 [39:17<33:54,  1.50it/s]
3|3412|Loss: 0.6996309161186218:  53%|█████▎    | 3412/6470 [39:17<33:38,  1.52it/s]
3|3413|Loss: 0.8992444276809692:  53%|█████▎    | 3412/6470 [39:18<33:38,  1.52it/s]
3|3413|Loss: 0.8992444276809692:  53%|█████▎    | 3413/6470 [39:18<33:16,  1.53it/s]
3|3414|Loss: 1.0426859855651855:  53%|█████▎    | 3413/6470 [39:19<33:16,  1.53it/s]
3|3414|Loss: 1.0426859855651855:  53%|█████▎    | 3414/6470 [39:19<35:54,  1.42it/s]
3|3415|Loss: 1.0331674814224243:  53%|█████▎    | 3414/6470 [39:19<35:54,  1.42it/s]
3|3415|Loss: 1.0331674814224243:  53%|█████▎    | 3415/6470 [39:20<36:49,  1.38it/s]
3|3416|Loss: 0.8402122259140015:  53%|█████▎    | 3415/6470 [39:20<36:49,  1.38it/s]
3|3416|Loss: 0.8402122259140015:  53%|█████▎    | 3416/6470 [39:20<35:54,  1.42it/s]
3|3417|Loss: 0.7952502369880676:  53%|█████▎    | 3416/6470 [39:21<35:54,  1.42it/s]
3|3417|Loss: 0.7952502369880676:  53%|█████▎    | 3417/6470 [39:21<34:36,  1.47it/s]
3|3418|Loss: 0.8901790976524353:  53%|█████▎    | 3417/6470 [39:21<34:36,  1.47it/s]
3|3418|Loss: 0.8901790976524353:  53%|█████▎    | 3418/6470 [39:22<36:40,  1.39it/s]
3|3419|Loss: 0.9023271799087524:  53%|█████▎    | 3418/6470 [39:22<36:40,  1.39it/s]
3|3419|Loss: 0.9023271799087524:  53%|█████▎    | 3419/6470 [39:23<37:03,  1.37it/s]
3|3420|Loss: 0.9051059484481812:  53%|█████▎    | 3419/6470 [39:23<37:03,  1.37it/s]
3|3420|Loss: 0.9051059484481812:  53%|█████▎    | 3420/6470 [39:23<38:32,  1.32it/s]
3|3421|Loss: 0.8276506066322327:  53%|█████▎    | 3420/6470 [39:24<38:32,  1.32it/s]
3|3421|Loss: 0.8276506066322327:  53%|█████▎    | 3421/6470 [39:24<39:31,  1.29it/s]
3|3422|Loss: 1.0688693523406982:  53%|█████▎    | 3421/6470 [39:25<39:31,  1.29it/s]
3|3422|Loss: 1.0688693523406982:  53%|█████▎    | 3422/6470 [39:25<38:04,  1.33it/s]
3|3423|Loss: 0.8433167934417725:  53%|█████▎    | 3422/6470 [39:25<38:04,  1.33it/s]
3|3423|Loss: 0.8433167934417725:  53%|█████▎    | 3423/6470 [39:25<35:17,  1.44it/s]
3|3424|Loss: 0.8877439498901367:  53%|█████▎    | 3423/6470 [39:26<35:17,  1.44it/s]
3|3424|Loss: 0.8877439498901367:  53%|█████▎    | 3424/6470 [39:26<36:07,  1.41it/s]
3|3425|Loss: 0.9829213619232178:  53%|█████▎    | 3424/6470 [39:27<36:07,  1.41it/s]
3|3425|Loss: 0.9829213619232178:  53%|█████▎    | 3425/6470 [39:27<36:16,  1.40it/s]
3|3426|Loss: 0.8534762859344482:  53%|█████▎    | 3425/6470 [39:27<36:16,  1.40it/s]
3|3426|Loss: 0.8534762859344482:  53%|█████▎    | 3426/6470 [39:28<37:56,  1.34it/s]
3|3427|Loss: 0.8472155332565308:  53%|█████▎    | 3426/6470 [39:28<37:56,  1.34it/s]
3|3427|Loss: 0.8472155332565308:  53%|█████▎    | 3427/6470 [39:28<36:13,  1.40it/s]
3|3428|Loss: 0.7592411637306213:  53%|█████▎    | 3427/6470 [39:29<36:13,  1.40it/s]
3|3428|Loss: 0.7592411637306213:  53%|█████▎    | 3428/6470 [39:29<37:49,  1.34it/s]
3|3429|Loss: 0.795749306678772:  53%|█████▎    | 3428/6470 [39:30<37:49,  1.34it/s] 
3|3429|Loss: 0.795749306678772:  53%|█████▎    | 3429/6470 [39:30<39:55,  1.27it/s]
3|3430|Loss: 0.7362539172172546:  53%|█████▎    | 3429/6470 [39:31<39:55,  1.27it/s]
3|3430|Loss: 0.7362539172172546:  53%|█████▎    | 3430/6470 [39:31<39:38,  1.28it/s]
3|3431|Loss: 0.8860832452774048:  53%|█████▎    | 3430/6470 [39:31<39:38,  1.28it/s]
3|3431|Loss: 0.8860832452774048:  53%|█████▎    | 3431/6470 [39:32<40:03,  1.26it/s]
3|3432|Loss: 0.9838108420372009:  53%|█████▎    | 3431/6470 [39:32<40:03,  1.26it/s]
3|3432|Loss: 0.9838108420372009:  53%|█████▎    | 3432/6470 [39:32<40:07,  1.26it/s]
3|3433|Loss: 0.8696194887161255:  53%|█████▎    | 3432/6470 [39:33<40:07,  1.26it/s]
3|3433|Loss: 0.8696194887161255:  53%|█████▎    | 3433/6470 [39:33<40:39,  1.25it/s]
3|3434|Loss: 0.9261694550514221:  53%|█████▎    | 3433/6470 [39:34<40:39,  1.25it/s]
3|3434|Loss: 0.9261694550514221:  53%|█████▎    | 3434/6470 [39:34<39:32,  1.28it/s]
3|3435|Loss: 0.9100687503814697:  53%|█████▎    | 3434/6470 [39:34<39:32,  1.28it/s]
3|3435|Loss: 0.9100687503814697:  53%|█████▎    | 3435/6470 [39:34<33:56,  1.49it/s]
3|3436|Loss: 0.8807289004325867:  53%|█████▎    | 3435/6470 [39:35<33:56,  1.49it/s]
3|3436|Loss: 0.8807289004325867:  53%|█████▎    | 3436/6470 [39:35<33:34,  1.51it/s]
3|3437|Loss: 0.9859541058540344:  53%|█████▎    | 3436/6470 [39:35<33:34,  1.51it/s]
3|3437|Loss: 0.9859541058540344:  53%|█████▎    | 3437/6470 [39:36<35:16,  1.43it/s]
3|3438|Loss: 0.9136415123939514:  53%|█████▎    | 3437/6470 [39:36<35:16,  1.43it/s]
3|3438|Loss: 0.9136415123939514:  53%|█████▎    | 3438/6470 [39:37<35:43,  1.41it/s]
3|3439|Loss: 1.1754977703094482:  53%|█████▎    | 3438/6470 [39:37<35:43,  1.41it/s]
3|3439|Loss: 1.1754977703094482:  53%|█████▎    | 3439/6470 [39:37<33:01,  1.53it/s]
3|3440|Loss: 0.8745834827423096:  53%|█████▎    | 3439/6470 [39:37<33:01,  1.53it/s]
3|3440|Loss: 0.8745834827423096:  53%|█████▎    | 3440/6470 [39:38<34:03,  1.48it/s]
3|3441|Loss: 0.7672586441040039:  53%|█████▎    | 3440/6470 [39:38<34:03,  1.48it/s]
3|3441|Loss: 0.7672586441040039:  53%|█████▎    | 3441/6470 [39:39<35:54,  1.41it/s]
3|3442|Loss: 0.6978520154953003:  53%|█████▎    | 3441/6470 [39:39<35:54,  1.41it/s]
3|3442|Loss: 0.6978520154953003:  53%|█████▎    | 3442/6470 [39:40<38:34,  1.31it/s]
3|3443|Loss: 0.9702535271644592:  53%|█████▎    | 3442/6470 [39:40<38:34,  1.31it/s]
3|3443|Loss: 0.9702535271644592:  53%|█████▎    | 3443/6470 [39:40<40:22,  1.25it/s]
3|3444|Loss: 0.8612377047538757:  53%|█████▎    | 3443/6470 [39:41<40:22,  1.25it/s]
3|3444|Loss: 0.8612377047538757:  53%|█████▎    | 3444/6470 [39:41<37:26,  1.35it/s]
3|3445|Loss: 0.9097595810890198:  53%|█████▎    | 3444/6470 [39:41<37:26,  1.35it/s]
3|3445|Loss: 0.9097595810890198:  53%|█████▎    | 3445/6470 [39:42<35:33,  1.42it/s]
3|3446|Loss: 0.8380140662193298:  53%|█████▎    | 3445/6470 [39:42<35:33,  1.42it/s]
3|3446|Loss: 0.8380140662193298:  53%|█████▎    | 3446/6470 [39:42<35:43,  1.41it/s]
3|3447|Loss: 0.8769270777702332:  53%|█████▎    | 3446/6470 [39:43<35:43,  1.41it/s]
3|3447|Loss: 0.8769270777702332:  53%|█████▎    | 3447/6470 [39:43<36:03,  1.40it/s]
3|3448|Loss: 0.8069905042648315:  53%|█████▎    | 3447/6470 [39:43<36:03,  1.40it/s]
3|3448|Loss: 0.8069905042648315:  53%|█████▎    | 3448/6470 [39:44<32:02,  1.57it/s]
3|3449|Loss: 0.8249569535255432:  53%|█████▎    | 3448/6470 [39:44<32:02,  1.57it/s]
3|3449|Loss: 0.8249569535255432:  53%|█████▎    | 3449/6470 [39:44<30:18,  1.66it/s]
3|3450|Loss: 0.836039125919342:  53%|█████▎    | 3449/6470 [39:44<30:18,  1.66it/s] 
3|3450|Loss: 0.836039125919342:  53%|█████▎    | 3450/6470 [39:45<29:02,  1.73it/s]
3|3451|Loss: 0.9935420155525208:  53%|█████▎    | 3450/6470 [39:45<29:02,  1.73it/s]
3|3451|Loss: 0.9935420155525208:  53%|█████▎    | 3451/6470 [39:45<28:48,  1.75it/s]
3|3452|Loss: 0.7412488460540771:  53%|█████▎    | 3451/6470 [39:45<28:48,  1.75it/s]
3|3452|Loss: 0.7412488460540771:  53%|█████▎    | 3452/6470 [39:46<26:05,  1.93it/s]
3|3453|Loss: 0.9354494214057922:  53%|█████▎    | 3452/6470 [39:46<26:05,  1.93it/s]
3|3453|Loss: 0.9354494214057922:  53%|█████▎    | 3453/6470 [39:46<27:57,  1.80it/s]
3|3454|Loss: 0.8735441565513611:  53%|█████▎    | 3453/6470 [39:47<27:57,  1.80it/s]
3|3454|Loss: 0.8735441565513611:  53%|█████▎    | 3454/6470 [39:47<30:21,  1.66it/s]
3|3455|Loss: 0.8368034362792969:  53%|█████▎    | 3454/6470 [39:47<30:21,  1.66it/s]
3|3455|Loss: 0.8368034362792969:  53%|█████▎    | 3455/6470 [39:48<30:46,  1.63it/s]
3|3456|Loss: 1.1668939590454102:  53%|█████▎    | 3455/6470 [39:48<30:46,  1.63it/s]
3|3456|Loss: 1.1668939590454102:  53%|█████▎    | 3456/6470 [39:48<32:10,  1.56it/s]
3|3457|Loss: 0.8971319198608398:  53%|█████▎    | 3456/6470 [39:49<32:10,  1.56it/s]
3|3457|Loss: 0.8971319198608398:  53%|█████▎    | 3457/6470 [39:49<33:41,  1.49it/s]
3|3458|Loss: 0.8238789439201355:  53%|█████▎    | 3457/6470 [39:49<33:41,  1.49it/s]
3|3458|Loss: 0.8238789439201355:  53%|█████▎    | 3458/6470 [39:50<34:46,  1.44it/s]
3|3459|Loss: 0.8715338706970215:  53%|█████▎    | 3458/6470 [39:50<34:46,  1.44it/s]
3|3459|Loss: 0.8715338706970215:  53%|█████▎    | 3459/6470 [39:51<37:37,  1.33it/s]
3|3460|Loss: 0.8837926387786865:  53%|█████▎    | 3459/6470 [39:51<37:37,  1.33it/s]
3|3460|Loss: 0.8837926387786865:  53%|█████▎    | 3460/6470 [39:51<37:53,  1.32it/s]
3|3461|Loss: 0.7700921297073364:  53%|█████▎    | 3460/6470 [39:52<37:53,  1.32it/s]
3|3461|Loss: 0.7700921297073364:  53%|█████▎    | 3461/6470 [39:52<36:20,  1.38it/s]
3|3462|Loss: 0.787326991558075:  53%|█████▎    | 3461/6470 [39:52<36:20,  1.38it/s] 
3|3462|Loss: 0.787326991558075:  54%|█████▎    | 3462/6470 [39:52<31:14,  1.60it/s]
3|3463|Loss: 0.9108631014823914:  54%|█████▎    | 3462/6470 [39:53<31:14,  1.60it/s]
3|3463|Loss: 0.9108631014823914:  54%|█████▎    | 3463/6470 [39:53<29:46,  1.68it/s]
3|3464|Loss: 0.8407723307609558:  54%|█████▎    | 3463/6470 [39:53<29:46,  1.68it/s]
3|3464|Loss: 0.8407723307609558:  54%|█████▎    | 3464/6470 [39:54<32:57,  1.52it/s]
3|3465|Loss: 0.8057774305343628:  54%|█████▎    | 3464/6470 [39:54<32:57,  1.52it/s]
3|3465|Loss: 0.8057774305343628:  54%|█████▎    | 3465/6470 [39:54<34:39,  1.45it/s]
3|3466|Loss: 0.7178503274917603:  54%|█████▎    | 3465/6470 [39:55<34:39,  1.45it/s]
3|3466|Loss: 0.7178503274917603:  54%|█████▎    | 3466/6470 [39:55<32:37,  1.53it/s]
3|3467|Loss: 0.7848338484764099:  54%|█████▎    | 3466/6470 [39:55<32:37,  1.53it/s]
3|3467|Loss: 0.7848338484764099:  54%|█████▎    | 3467/6470 [39:56<31:12,  1.60it/s]
3|3468|Loss: 0.8045482039451599:  54%|█████▎    | 3467/6470 [39:56<31:12,  1.60it/s]
3|3468|Loss: 0.8045482039451599:  54%|█████▎    | 3468/6470 [39:56<29:29,  1.70it/s]
3|3469|Loss: 0.9290680289268494:  54%|█████▎    | 3468/6470 [39:56<29:29,  1.70it/s]
3|3469|Loss: 0.9290680289268494:  54%|█████▎    | 3469/6470 [39:57<29:27,  1.70it/s]
3|3470|Loss: 0.8301011323928833:  54%|█████▎    | 3469/6470 [39:57<29:27,  1.70it/s]
3|3470|Loss: 0.8301011323928833:  54%|█████▎    | 3470/6470 [39:57<30:54,  1.62it/s]
3|3471|Loss: 0.9305033087730408:  54%|█████▎    | 3470/6470 [39:58<30:54,  1.62it/s]
3|3471|Loss: 0.9305033087730408:  54%|█████▎    | 3471/6470 [39:58<31:22,  1.59it/s]
3|3472|Loss: 0.8409935832023621:  54%|█████▎    | 3471/6470 [39:58<31:22,  1.59it/s]
3|3472|Loss: 0.8409935832023621:  54%|█████▎    | 3472/6470 [39:59<32:45,  1.53it/s]
3|3473|Loss: 0.9055166244506836:  54%|█████▎    | 3472/6470 [39:59<32:45,  1.53it/s]
3|3473|Loss: 0.9055166244506836:  54%|█████▎    | 3473/6470 [39:59<32:13,  1.55it/s]
3|3474|Loss: 0.7260448932647705:  54%|█████▎    | 3473/6470 [40:00<32:13,  1.55it/s]
3|3474|Loss: 0.7260448932647705:  54%|█████▎    | 3474/6470 [40:00<31:42,  1.58it/s]
3|3475|Loss: 0.8421762585639954:  54%|█████▎    | 3474/6470 [40:00<31:42,  1.58it/s]
3|3475|Loss: 0.8421762585639954:  54%|█████▎    | 3475/6470 [40:01<31:27,  1.59it/s]
3|3476|Loss: 0.8478549718856812:  54%|█████▎    | 3475/6470 [40:01<31:27,  1.59it/s]
3|3476|Loss: 0.8478549718856812:  54%|█████▎    | 3476/6470 [40:01<31:29,  1.58it/s]
3|3477|Loss: 0.7638432383537292:  54%|█████▎    | 3476/6470 [40:02<31:29,  1.58it/s]
3|3477|Loss: 0.7638432383537292:  54%|█████▎    | 3477/6470 [40:02<33:58,  1.47it/s]
3|3478|Loss: 0.933647871017456:  54%|█████▎    | 3477/6470 [40:03<33:58,  1.47it/s] 
3|3478|Loss: 0.933647871017456:  54%|█████▍    | 3478/6470 [40:03<36:28,  1.37it/s]
3|3479|Loss: 0.9201403856277466:  54%|█████▍    | 3478/6470 [40:03<36:28,  1.37it/s]
3|3479|Loss: 0.9201403856277466:  54%|█████▍    | 3479/6470 [40:04<36:36,  1.36it/s]
3|3480|Loss: 0.9154060482978821:  54%|█████▍    | 3479/6470 [40:04<36:36,  1.36it/s]
3|3480|Loss: 0.9154060482978821:  54%|█████▍    | 3480/6470 [40:04<35:04,  1.42it/s]
3|3481|Loss: 0.6878881454467773:  54%|█████▍    | 3480/6470 [40:05<35:04,  1.42it/s]
3|3481|Loss: 0.6878881454467773:  54%|█████▍    | 3481/6470 [40:05<32:55,  1.51it/s]
3|3482|Loss: 0.7618594765663147:  54%|█████▍    | 3481/6470 [40:05<32:55,  1.51it/s]
3|3482|Loss: 0.7618594765663147:  54%|█████▍    | 3482/6470 [40:06<35:08,  1.42it/s]
3|3483|Loss: 0.7877549529075623:  54%|█████▍    | 3482/6470 [40:06<35:08,  1.42it/s]
3|3483|Loss: 0.7877549529075623:  54%|█████▍    | 3483/6470 [40:06<33:51,  1.47it/s]
3|3484|Loss: 0.7158029079437256:  54%|█████▍    | 3483/6470 [40:07<33:51,  1.47it/s]
3|3484|Loss: 0.7158029079437256:  54%|█████▍    | 3484/6470 [40:07<32:44,  1.52it/s]
3|3485|Loss: 0.878100574016571:  54%|█████▍    | 3484/6470 [40:07<32:44,  1.52it/s] 
3|3485|Loss: 0.878100574016571:  54%|█████▍    | 3485/6470 [40:08<33:13,  1.50it/s]
3|3486|Loss: 0.7511565089225769:  54%|█████▍    | 3485/6470 [40:08<33:13,  1.50it/s]
3|3486|Loss: 0.7511565089225769:  54%|█████▍    | 3486/6470 [40:08<31:06,  1.60it/s]
3|3487|Loss: 0.861345112323761:  54%|█████▍    | 3486/6470 [40:08<31:06,  1.60it/s] 
3|3487|Loss: 0.861345112323761:  54%|█████▍    | 3487/6470 [40:09<31:39,  1.57it/s]
3|3488|Loss: 0.9030723571777344:  54%|█████▍    | 3487/6470 [40:09<31:39,  1.57it/s]
3|3488|Loss: 0.9030723571777344:  54%|█████▍    | 3488/6470 [40:09<31:56,  1.56it/s]
3|3489|Loss: 0.7817858457565308:  54%|█████▍    | 3488/6470 [40:10<31:56,  1.56it/s]
3|3489|Loss: 0.7817858457565308:  54%|█████▍    | 3489/6470 [40:10<32:38,  1.52it/s]
3|3490|Loss: 0.7841544151306152:  54%|█████▍    | 3489/6470 [40:10<32:38,  1.52it/s]
3|3490|Loss: 0.7841544151306152:  54%|█████▍    | 3490/6470 [40:11<33:21,  1.49it/s]
3|3491|Loss: 0.8355631828308105:  54%|█████▍    | 3490/6470 [40:11<33:21,  1.49it/s]
3|3491|Loss: 0.8355631828308105:  54%|█████▍    | 3491/6470 [40:11<32:22,  1.53it/s]
3|3492|Loss: 0.8226327300071716:  54%|█████▍    | 3491/6470 [40:12<32:22,  1.53it/s]
3|3492|Loss: 0.8226327300071716:  54%|█████▍    | 3492/6470 [40:12<32:42,  1.52it/s]
3|3493|Loss: 0.9640317559242249:  54%|█████▍    | 3492/6470 [40:12<32:42,  1.52it/s]
3|3493|Loss: 0.9640317559242249:  54%|█████▍    | 3493/6470 [40:13<32:18,  1.54it/s]
3|3494|Loss: 0.8388043642044067:  54%|█████▍    | 3493/6470 [40:13<32:18,  1.54it/s]
3|3494|Loss: 0.8388043642044067:  54%|█████▍    | 3494/6470 [40:13<32:48,  1.51it/s]
3|3495|Loss: 0.6994636058807373:  54%|█████▍    | 3494/6470 [40:14<32:48,  1.51it/s]
3|3495|Loss: 0.6994636058807373:  54%|█████▍    | 3495/6470 [40:14<35:36,  1.39it/s]
3|3496|Loss: 0.9237382411956787:  54%|█████▍    | 3495/6470 [40:15<35:36,  1.39it/s]
3|3496|Loss: 0.9237382411956787:  54%|█████▍    | 3496/6470 [40:15<38:10,  1.30it/s]
3|3497|Loss: 0.7960272431373596:  54%|█████▍    | 3496/6470 [40:16<38:10,  1.30it/s]
3|3497|Loss: 0.7960272431373596:  54%|█████▍    | 3497/6470 [40:16<36:44,  1.35it/s]
3|3498|Loss: 0.8809794783592224:  54%|█████▍    | 3497/6470 [40:16<36:44,  1.35it/s]
3|3498|Loss: 0.8809794783592224:  54%|█████▍    | 3498/6470 [40:17<36:30,  1.36it/s]
3|3499|Loss: 0.760024905204773:  54%|█████▍    | 3498/6470 [40:17<36:30,  1.36it/s] 
3|3499|Loss: 0.760024905204773:  54%|█████▍    | 3499/6470 [40:17<35:12,  1.41it/s]
3|3500|Loss: 0.9086977243423462:  54%|█████▍    | 3499/6470 [40:18<35:12,  1.41it/s]
3|3500|Loss: 0.9086977243423462:  54%|█████▍    | 3500/6470 [40:18<34:48,  1.42it/s]
3|3501|Loss: 0.8524712324142456:  54%|█████▍    | 3500/6470 [40:18<34:48,  1.42it/s]
3|3501|Loss: 0.8524712324142456:  54%|█████▍    | 3501/6470 [40:18<32:58,  1.50it/s]
3|3502|Loss: 0.8295998573303223:  54%|█████▍    | 3501/6470 [40:19<32:58,  1.50it/s]
3|3502|Loss: 0.8295998573303223:  54%|█████▍    | 3502/6470 [40:19<33:27,  1.48it/s]
3|3503|Loss: 0.9028812646865845:  54%|█████▍    | 3502/6470 [40:20<33:27,  1.48it/s]
3|3503|Loss: 0.9028812646865845:  54%|█████▍    | 3503/6470 [40:20<35:41,  1.39it/s]
3|3504|Loss: 0.859721302986145:  54%|█████▍    | 3503/6470 [40:20<35:41,  1.39it/s] 
3|3504|Loss: 0.859721302986145:  54%|█████▍    | 3504/6470 [40:21<33:17,  1.48it/s]
3|3505|Loss: 0.9606480002403259:  54%|█████▍    | 3504/6470 [40:21<33:17,  1.48it/s]
3|3505|Loss: 0.9606480002403259:  54%|█████▍    | 3505/6470 [40:21<32:13,  1.53it/s]
3|3506|Loss: 0.8012507557868958:  54%|█████▍    | 3505/6470 [40:22<32:13,  1.53it/s]
3|3506|Loss: 0.8012507557868958:  54%|█████▍    | 3506/6470 [40:22<33:18,  1.48it/s]
3|3507|Loss: 0.7481352090835571:  54%|█████▍    | 3506/6470 [40:22<33:18,  1.48it/s]
3|3507|Loss: 0.7481352090835571:  54%|█████▍    | 3507/6470 [40:22<31:13,  1.58it/s]
3|3508|Loss: 0.8807234168052673:  54%|█████▍    | 3507/6470 [40:23<31:13,  1.58it/s]
3|3508|Loss: 0.8807234168052673:  54%|█████▍    | 3508/6470 [40:23<29:01,  1.70it/s]
3|3509|Loss: 0.8963325023651123:  54%|█████▍    | 3508/6470 [40:23<29:01,  1.70it/s]
3|3509|Loss: 0.8963325023651123:  54%|█████▍    | 3509/6470 [40:23<28:29,  1.73it/s]
3|3510|Loss: 0.759669303894043:  54%|█████▍    | 3509/6470 [40:24<28:29,  1.73it/s] 
3|3510|Loss: 0.759669303894043:  54%|█████▍    | 3510/6470 [40:24<29:13,  1.69it/s]
3|3511|Loss: 0.7844089269638062:  54%|█████▍    | 3510/6470 [40:24<29:13,  1.69it/s]
3|3511|Loss: 0.7844089269638062:  54%|█████▍    | 3511/6470 [40:25<30:44,  1.60it/s]
3|3512|Loss: 0.8401048183441162:  54%|█████▍    | 3511/6470 [40:25<30:44,  1.60it/s]
3|3512|Loss: 0.8401048183441162:  54%|█████▍    | 3512/6470 [40:26<33:21,  1.48it/s]
3|3513|Loss: 0.7153348922729492:  54%|█████▍    | 3512/6470 [40:26<33:21,  1.48it/s]
3|3513|Loss: 0.7153348922729492:  54%|█████▍    | 3513/6470 [40:26<36:32,  1.35it/s]
3|3514|Loss: 0.9444725513458252:  54%|█████▍    | 3513/6470 [40:27<36:32,  1.35it/s]
3|3514|Loss: 0.9444725513458252:  54%|█████▍    | 3514/6470 [40:27<35:39,  1.38it/s]
3|3515|Loss: 0.8925244212150574:  54%|█████▍    | 3514/6470 [40:28<35:39,  1.38it/s]
3|3515|Loss: 0.8925244212150574:  54%|█████▍    | 3515/6470 [40:28<35:20,  1.39it/s]
3|3516|Loss: 0.8181859850883484:  54%|█████▍    | 3515/6470 [40:28<35:20,  1.39it/s]
3|3516|Loss: 0.8181859850883484:  54%|█████▍    | 3516/6470 [40:28<33:40,  1.46it/s]
3|3517|Loss: 0.8643853068351746:  54%|█████▍    | 3516/6470 [40:29<33:40,  1.46it/s]
3|3517|Loss: 0.8643853068351746:  54%|█████▍    | 3517/6470 [40:29<34:06,  1.44it/s]
3|3518|Loss: 0.8193082809448242:  54%|█████▍    | 3517/6470 [40:30<34:06,  1.44it/s]
3|3518|Loss: 0.8193082809448242:  54%|█████▍    | 3518/6470 [40:30<33:21,  1.47it/s]
3|3519|Loss: 0.703973114490509:  54%|█████▍    | 3518/6470 [40:30<33:21,  1.47it/s] 
3|3519|Loss: 0.703973114490509:  54%|█████▍    | 3519/6470 [40:30<32:23,  1.52it/s]
3|3520|Loss: 0.817660927772522:  54%|█████▍    | 3519/6470 [40:31<32:23,  1.52it/s]
3|3520|Loss: 0.817660927772522:  54%|█████▍    | 3520/6470 [40:31<32:06,  1.53it/s]
3|3521|Loss: 0.9857491254806519:  54%|█████▍    | 3520/6470 [40:31<32:06,  1.53it/s]
3|3521|Loss: 0.9857491254806519:  54%|█████▍    | 3521/6470 [40:32<33:11,  1.48it/s]
3|3522|Loss: 0.9155179262161255:  54%|█████▍    | 3521/6470 [40:32<33:11,  1.48it/s]
3|3522|Loss: 0.9155179262161255:  54%|█████▍    | 3522/6470 [40:33<34:46,  1.41it/s]
3|3523|Loss: 1.0064826011657715:  54%|█████▍    | 3522/6470 [40:33<34:46,  1.41it/s]
3|3523|Loss: 1.0064826011657715:  54%|█████▍    | 3523/6470 [40:33<35:28,  1.38it/s]
3|3524|Loss: 0.79721999168396:  54%|█████▍    | 3523/6470 [40:34<35:28,  1.38it/s]  
3|3524|Loss: 0.79721999168396:  54%|█████▍    | 3524/6470 [40:34<35:40,  1.38it/s]
3|3525|Loss: 0.8280458450317383:  54%|█████▍    | 3524/6470 [40:34<35:40,  1.38it/s]
3|3525|Loss: 0.8280458450317383:  54%|█████▍    | 3525/6470 [40:35<36:24,  1.35it/s]
3|3526|Loss: 0.8989880084991455:  54%|█████▍    | 3525/6470 [40:35<36:24,  1.35it/s]
3|3526|Loss: 0.8989880084991455:  54%|█████▍    | 3526/6470 [40:36<37:22,  1.31it/s]
3|3527|Loss: 0.8339927196502686:  54%|█████▍    | 3526/6470 [40:36<37:22,  1.31it/s]
3|3527|Loss: 0.8339927196502686:  55%|█████▍    | 3527/6470 [40:36<38:18,  1.28it/s]
3|3528|Loss: 0.9095377922058105:  55%|█████▍    | 3527/6470 [40:37<38:18,  1.28it/s]
3|3528|Loss: 0.9095377922058105:  55%|█████▍    | 3528/6470 [40:37<36:53,  1.33it/s]
3|3529|Loss: 0.9025195240974426:  55%|█████▍    | 3528/6470 [40:38<36:53,  1.33it/s]
3|3529|Loss: 0.9025195240974426:  55%|█████▍    | 3529/6470 [40:38<34:59,  1.40it/s]
3|3530|Loss: 0.8558183312416077:  55%|█████▍    | 3529/6470 [40:38<34:59,  1.40it/s]
3|3530|Loss: 0.8558183312416077:  55%|█████▍    | 3530/6470 [40:38<33:17,  1.47it/s]
3|3531|Loss: 0.9255357384681702:  55%|█████▍    | 3530/6470 [40:39<33:17,  1.47it/s]
3|3531|Loss: 0.9255357384681702:  55%|█████▍    | 3531/6470 [40:39<35:10,  1.39it/s]
3|3532|Loss: 0.9136607646942139:  55%|█████▍    | 3531/6470 [40:40<35:10,  1.39it/s]
3|3532|Loss: 0.9136607646942139:  55%|█████▍    | 3532/6470 [40:40<33:14,  1.47it/s]
3|3533|Loss: 0.8930861353874207:  55%|█████▍    | 3532/6470 [40:40<33:14,  1.47it/s]
3|3533|Loss: 0.8930861353874207:  55%|█████▍    | 3533/6470 [40:40<31:17,  1.56it/s]
3|3534|Loss: 0.938608705997467:  55%|█████▍    | 3533/6470 [40:41<31:17,  1.56it/s] 
3|3534|Loss: 0.938608705997467:  55%|█████▍    | 3534/6470 [40:41<30:46,  1.59it/s]
3|3535|Loss: 0.8472993969917297:  55%|█████▍    | 3534/6470 [40:41<30:46,  1.59it/s]
3|3535|Loss: 0.8472993969917297:  55%|█████▍    | 3535/6470 [40:42<30:51,  1.59it/s]
3|3536|Loss: 0.8452273607254028:  55%|█████▍    | 3535/6470 [40:42<30:51,  1.59it/s]
3|3536|Loss: 0.8452273607254028:  55%|█████▍    | 3536/6470 [40:42<31:04,  1.57it/s]
3|3537|Loss: 0.8122745752334595:  55%|█████▍    | 3536/6470 [40:43<31:04,  1.57it/s]
3|3537|Loss: 0.8122745752334595:  55%|█████▍    | 3537/6470 [40:43<33:25,  1.46it/s]
3|3538|Loss: 0.7331638336181641:  55%|█████▍    | 3537/6470 [40:43<33:25,  1.46it/s]
3|3538|Loss: 0.7331638336181641:  55%|█████▍    | 3538/6470 [40:44<33:05,  1.48it/s]
3|3539|Loss: 1.030869722366333:  55%|█████▍    | 3538/6470 [40:44<33:05,  1.48it/s] 
3|3539|Loss: 1.030869722366333:  55%|█████▍    | 3539/6470 [40:44<34:28,  1.42it/s]
3|3540|Loss: 0.7924909591674805:  55%|█████▍    | 3539/6470 [40:45<34:28,  1.42it/s]
3|3540|Loss: 0.7924909591674805:  55%|█████▍    | 3540/6470 [40:45<34:03,  1.43it/s]
3|3541|Loss: 0.7716583013534546:  55%|█████▍    | 3540/6470 [40:46<34:03,  1.43it/s]
3|3541|Loss: 0.7716583013534546:  55%|█████▍    | 3541/6470 [40:46<35:51,  1.36it/s]
3|3542|Loss: 0.6923502087593079:  55%|█████▍    | 3541/6470 [40:46<35:51,  1.36it/s]
3|3542|Loss: 0.6923502087593079:  55%|█████▍    | 3542/6470 [40:47<33:57,  1.44it/s]
3|3543|Loss: 0.6856397986412048:  55%|█████▍    | 3542/6470 [40:47<33:57,  1.44it/s]
3|3543|Loss: 0.6856397986412048:  55%|█████▍    | 3543/6470 [40:47<33:32,  1.45it/s]
3|3544|Loss: 0.9213840961456299:  55%|█████▍    | 3543/6470 [40:48<33:32,  1.45it/s]
3|3544|Loss: 0.9213840961456299:  55%|█████▍    | 3544/6470 [40:48<36:43,  1.33it/s]
3|3545|Loss: 0.781417191028595:  55%|█████▍    | 3544/6470 [40:49<36:43,  1.33it/s] 
3|3545|Loss: 0.781417191028595:  55%|█████▍    | 3545/6470 [40:49<36:52,  1.32it/s]
3|3546|Loss: 0.8908513784408569:  55%|█████▍    | 3545/6470 [40:49<36:52,  1.32it/s]
3|3546|Loss: 0.8908513784408569:  55%|█████▍    | 3546/6470 [40:50<37:11,  1.31it/s]
3|3547|Loss: 0.7928847074508667:  55%|█████▍    | 3546/6470 [40:50<37:11,  1.31it/s]
3|3547|Loss: 0.7928847074508667:  55%|█████▍    | 3547/6470 [40:50<36:35,  1.33it/s]
3|3548|Loss: 0.7946929931640625:  55%|█████▍    | 3547/6470 [40:51<36:35,  1.33it/s]
3|3548|Loss: 0.7946929931640625:  55%|█████▍    | 3548/6470 [40:51<34:28,  1.41it/s]
3|3549|Loss: 0.8293137550354004:  55%|█████▍    | 3548/6470 [40:51<34:28,  1.41it/s]
3|3549|Loss: 0.8293137550354004:  55%|█████▍    | 3549/6470 [40:52<31:46,  1.53it/s]
3|3550|Loss: 0.8184617161750793:  55%|█████▍    | 3549/6470 [40:52<31:46,  1.53it/s]
3|3550|Loss: 0.8184617161750793:  55%|█████▍    | 3550/6470 [40:52<33:38,  1.45it/s]
3|3551|Loss: 0.7978705763816833:  55%|█████▍    | 3550/6470 [40:53<33:38,  1.45it/s]
3|3551|Loss: 0.7978705763816833:  55%|█████▍    | 3551/6470 [40:53<34:45,  1.40it/s]
3|3552|Loss: 0.9657334685325623:  55%|█████▍    | 3551/6470 [40:54<34:45,  1.40it/s]
3|3552|Loss: 0.9657334685325623:  55%|█████▍    | 3552/6470 [40:54<36:35,  1.33it/s]
3|3553|Loss: 0.8570317029953003:  55%|█████▍    | 3552/6470 [40:54<36:35,  1.33it/s]
3|3553|Loss: 0.8570317029953003:  55%|█████▍    | 3553/6470 [40:55<37:58,  1.28it/s]
3|3554|Loss: 1.0536168813705444:  55%|█████▍    | 3553/6470 [40:55<37:58,  1.28it/s]
3|3554|Loss: 1.0536168813705444:  55%|█████▍    | 3554/6470 [40:55<36:16,  1.34it/s]
3|3555|Loss: 0.8902103304862976:  55%|█████▍    | 3554/6470 [40:56<36:16,  1.34it/s]
3|3555|Loss: 0.8902103304862976:  55%|█████▍    | 3555/6470 [40:56<31:39,  1.53it/s]
3|3556|Loss: 0.8971112370491028:  55%|█████▍    | 3555/6470 [40:56<31:39,  1.53it/s]
3|3556|Loss: 0.8971112370491028:  55%|█████▍    | 3556/6470 [40:56<30:56,  1.57it/s]
3|3557|Loss: 0.7444746494293213:  55%|█████▍    | 3556/6470 [40:57<30:56,  1.57it/s]
3|3557|Loss: 0.7444746494293213:  55%|█████▍    | 3557/6470 [40:57<28:51,  1.68it/s]
3|3558|Loss: 0.9365943670272827:  55%|█████▍    | 3557/6470 [40:57<28:51,  1.68it/s]
3|3558|Loss: 0.9365943670272827:  55%|█████▍    | 3558/6470 [40:58<29:31,  1.64it/s]
3|3559|Loss: 0.9146713614463806:  55%|█████▍    | 3558/6470 [40:58<29:31,  1.64it/s]
3|3559|Loss: 0.9146713614463806:  55%|█████▌    | 3559/6470 [40:58<29:24,  1.65it/s]
3|3560|Loss: 0.7881242632865906:  55%|█████▌    | 3559/6470 [40:59<29:24,  1.65it/s]
3|3560|Loss: 0.7881242632865906:  55%|█████▌    | 3560/6470 [40:59<31:57,  1.52it/s]
3|3561|Loss: 0.8401248455047607:  55%|█████▌    | 3560/6470 [40:59<31:57,  1.52it/s]
3|3561|Loss: 0.8401248455047607:  55%|█████▌    | 3561/6470 [41:00<32:03,  1.51it/s]
3|3562|Loss: 0.7835875153541565:  55%|█████▌    | 3561/6470 [41:00<32:03,  1.51it/s]
3|3562|Loss: 0.7835875153541565:  55%|█████▌    | 3562/6470 [41:00<31:37,  1.53it/s]
3|3563|Loss: 0.9729718565940857:  55%|█████▌    | 3562/6470 [41:01<31:37,  1.53it/s]
3|3563|Loss: 0.9729718565940857:  55%|█████▌    | 3563/6470 [41:01<29:14,  1.66it/s]
3|3564|Loss: 0.832364022731781:  55%|█████▌    | 3563/6470 [41:01<29:14,  1.66it/s] 
3|3564|Loss: 0.832364022731781:  55%|█████▌    | 3564/6470 [41:01<29:53,  1.62it/s]
3|3565|Loss: 0.9178373217582703:  55%|█████▌    | 3564/6470 [41:02<29:53,  1.62it/s]
3|3565|Loss: 0.9178373217582703:  55%|█████▌    | 3565/6470 [41:02<33:15,  1.46it/s]
3|3566|Loss: 0.8000078201293945:  55%|█████▌    | 3565/6470 [41:03<33:15,  1.46it/s]
3|3566|Loss: 0.8000078201293945:  55%|█████▌    | 3566/6470 [41:03<34:44,  1.39it/s]
3|3567|Loss: 0.7256886959075928:  55%|█████▌    | 3566/6470 [41:03<34:44,  1.39it/s]
3|3567|Loss: 0.7256886959075928:  55%|█████▌    | 3567/6470 [41:04<31:45,  1.52it/s]
3|3568|Loss: 0.7687512040138245:  55%|█████▌    | 3567/6470 [41:04<31:45,  1.52it/s]
3|3568|Loss: 0.7687512040138245:  55%|█████▌    | 3568/6470 [41:04<31:33,  1.53it/s]
3|3569|Loss: 0.9447479844093323:  55%|█████▌    | 3568/6470 [41:05<31:33,  1.53it/s]
3|3569|Loss: 0.9447479844093323:  55%|█████▌    | 3569/6470 [41:05<31:16,  1.55it/s]
3|3570|Loss: 0.815947413444519:  55%|█████▌    | 3569/6470 [41:05<31:16,  1.55it/s] 
3|3570|Loss: 0.815947413444519:  55%|█████▌    | 3570/6470 [41:06<32:25,  1.49it/s]
3|3571|Loss: 0.820314347743988:  55%|█████▌    | 3570/6470 [41:06<32:25,  1.49it/s]
3|3571|Loss: 0.820314347743988:  55%|█████▌    | 3571/6470 [41:06<34:46,  1.39it/s]
3|3572|Loss: 0.8343889713287354:  55%|█████▌    | 3571/6470 [41:07<34:46,  1.39it/s]
3|3572|Loss: 0.8343889713287354:  55%|█████▌    | 3572/6470 [41:07<35:33,  1.36it/s]
3|3573|Loss: 0.8368316292762756:  55%|█████▌    | 3572/6470 [41:08<35:33,  1.36it/s]
3|3573|Loss: 0.8368316292762756:  55%|█████▌    | 3573/6470 [41:08<36:13,  1.33it/s]
3|3574|Loss: 0.77216637134552:  55%|█████▌    | 3573/6470 [41:08<36:13,  1.33it/s]  
3|3574|Loss: 0.77216637134552:  55%|█████▌    | 3574/6470 [41:09<34:21,  1.40it/s]
3|3575|Loss: 1.016477346420288:  55%|█████▌    | 3574/6470 [41:09<34:21,  1.40it/s]
3|3575|Loss: 1.016477346420288:  55%|█████▌    | 3575/6470 [41:09<33:05,  1.46it/s]
3|3576|Loss: 0.8845834732055664:  55%|█████▌    | 3575/6470 [41:10<33:05,  1.46it/s]
3|3576|Loss: 0.8845834732055664:  55%|█████▌    | 3576/6470 [41:10<33:34,  1.44it/s]
3|3577|Loss: 0.7669561505317688:  55%|█████▌    | 3576/6470 [41:10<33:34,  1.44it/s]
3|3577|Loss: 0.7669561505317688:  55%|█████▌    | 3577/6470 [41:11<32:22,  1.49it/s]
3|3578|Loss: 0.9209479689598083:  55%|█████▌    | 3577/6470 [41:11<32:22,  1.49it/s]
3|3578|Loss: 0.9209479689598083:  55%|█████▌    | 3578/6470 [41:11<32:28,  1.48it/s]
3|3579|Loss: 1.0191622972488403:  55%|█████▌    | 3578/6470 [41:12<32:28,  1.48it/s]
3|3579|Loss: 1.0191622972488403:  55%|█████▌    | 3579/6470 [41:12<31:37,  1.52it/s]
3|3580|Loss: 0.7516387104988098:  55%|█████▌    | 3579/6470 [41:12<31:37,  1.52it/s]
3|3580|Loss: 0.7516387104988098:  55%|█████▌    | 3580/6470 [41:12<28:00,  1.72it/s]
3|3581|Loss: 0.7989336252212524:  55%|█████▌    | 3580/6470 [41:13<28:00,  1.72it/s]
3|3581|Loss: 0.7989336252212524:  55%|█████▌    | 3581/6470 [41:13<29:14,  1.65it/s]
3|3582|Loss: 0.7798352837562561:  55%|█████▌    | 3581/6470 [41:13<29:14,  1.65it/s]
3|3582|Loss: 0.7798352837562561:  55%|█████▌    | 3582/6470 [41:14<31:14,  1.54it/s]
3|3583|Loss: 0.8170360326766968:  55%|█████▌    | 3582/6470 [41:14<31:14,  1.54it/s]
3|3583|Loss: 0.8170360326766968:  55%|█████▌    | 3583/6470 [41:14<33:42,  1.43it/s]
3|3584|Loss: 0.8129526972770691:  55%|█████▌    | 3583/6470 [41:15<33:42,  1.43it/s]
3|3584|Loss: 0.8129526972770691:  55%|█████▌    | 3584/6470 [41:15<35:15,  1.36it/s]
3|3585|Loss: 0.8475826382637024:  55%|█████▌    | 3584/6470 [41:16<35:15,  1.36it/s]
3|3585|Loss: 0.8475826382637024:  55%|█████▌    | 3585/6470 [41:16<34:13,  1.40it/s]
3|3586|Loss: 0.9534235596656799:  55%|█████▌    | 3585/6470 [41:16<34:13,  1.40it/s]
3|3586|Loss: 0.9534235596656799:  55%|█████▌    | 3586/6470 [41:17<34:10,  1.41it/s]
3|3587|Loss: 0.8820851445198059:  55%|█████▌    | 3586/6470 [41:17<34:10,  1.41it/s]
3|3587|Loss: 0.8820851445198059:  55%|█████▌    | 3587/6470 [41:17<34:38,  1.39it/s]
3|3588|Loss: 0.8395063877105713:  55%|█████▌    | 3587/6470 [41:18<34:38,  1.39it/s]
3|3588|Loss: 0.8395063877105713:  55%|█████▌    | 3588/6470 [41:18<33:37,  1.43it/s]
3|3589|Loss: 1.0000237226486206:  55%|█████▌    | 3588/6470 [41:18<33:37,  1.43it/s]
3|3589|Loss: 1.0000237226486206:  55%|█████▌    | 3589/6470 [41:19<31:58,  1.50it/s]
3|3590|Loss: 0.8560095429420471:  55%|█████▌    | 3589/6470 [41:19<31:58,  1.50it/s]
3|3590|Loss: 0.8560095429420471:  55%|█████▌    | 3590/6470 [41:19<30:43,  1.56it/s]
3|3591|Loss: 0.9099211692810059:  55%|█████▌    | 3590/6470 [41:20<30:43,  1.56it/s]
3|3591|Loss: 0.9099211692810059:  56%|█████▌    | 3591/6470 [41:20<31:19,  1.53it/s]
3|3592|Loss: 0.7612582445144653:  56%|█████▌    | 3591/6470 [41:20<31:19,  1.53it/s]
3|3592|Loss: 0.7612582445144653:  56%|█████▌    | 3592/6470 [41:21<31:10,  1.54it/s]
3|3593|Loss: 0.8312370181083679:  56%|█████▌    | 3592/6470 [41:21<31:10,  1.54it/s]
3|3593|Loss: 0.8312370181083679:  56%|█████▌    | 3593/6470 [41:21<31:38,  1.52it/s]
3|3594|Loss: 1.2274346351623535:  56%|█████▌    | 3593/6470 [41:22<31:38,  1.52it/s]
3|3594|Loss: 1.2274346351623535:  56%|█████▌    | 3594/6470 [41:22<29:59,  1.60it/s]
3|3595|Loss: 0.8220208883285522:  56%|█████▌    | 3594/6470 [41:22<29:59,  1.60it/s]
3|3595|Loss: 0.8220208883285522:  56%|█████▌    | 3595/6470 [41:22<28:16,  1.69it/s]
3|3596|Loss: 0.9517767429351807:  56%|█████▌    | 3595/6470 [41:23<28:16,  1.69it/s]
3|3596|Loss: 0.9517767429351807:  56%|█████▌    | 3596/6470 [41:23<29:49,  1.61it/s]
3|3597|Loss: 0.8131917715072632:  56%|█████▌    | 3596/6470 [41:23<29:49,  1.61it/s]
3|3597|Loss: 0.8131917715072632:  56%|█████▌    | 3597/6470 [41:24<31:31,  1.52it/s]
3|3598|Loss: 0.7177308201789856:  56%|█████▌    | 3597/6470 [41:24<31:31,  1.52it/s]
3|3598|Loss: 0.7177308201789856:  56%|█████▌    | 3598/6470 [41:25<33:42,  1.42it/s]
3|3599|Loss: 0.8643311262130737:  56%|█████▌    | 3598/6470 [41:25<33:42,  1.42it/s]
3|3599|Loss: 0.8643311262130737:  56%|█████▌    | 3599/6470 [41:25<33:25,  1.43it/s]
3|3600|Loss: 0.7338603138923645:  56%|█████▌    | 3599/6470 [41:26<33:25,  1.43it/s]
3|3600|Loss: 0.7338603138923645:  56%|█████▌    | 3600/6470 [41:26<32:18,  1.48it/s]
3|3601|Loss: 0.931901752948761:  56%|█████▌    | 3600/6470 [41:26<32:18,  1.48it/s] 
3|3601|Loss: 0.931901752948761:  56%|█████▌    | 3601/6470 [41:27<32:30,  1.47it/s]
3|3602|Loss: 0.9598456025123596:  56%|█████▌    | 3601/6470 [41:27<32:30,  1.47it/s]
3|3602|Loss: 0.9598456025123596:  56%|█████▌    | 3602/6470 [41:27<31:31,  1.52it/s]
3|3603|Loss: 0.8207991719245911:  56%|█████▌    | 3602/6470 [41:28<31:31,  1.52it/s]
3|3603|Loss: 0.8207991719245911:  56%|█████▌    | 3603/6470 [41:28<33:12,  1.44it/s]
3|3604|Loss: 0.6814207434654236:  56%|█████▌    | 3603/6470 [41:28<33:12,  1.44it/s]
3|3604|Loss: 0.6814207434654236:  56%|█████▌    | 3604/6470 [41:29<32:59,  1.45it/s]
3|3605|Loss: 0.9559321403503418:  56%|█████▌    | 3604/6470 [41:29<32:59,  1.45it/s]
3|3605|Loss: 0.9559321403503418:  56%|█████▌    | 3605/6470 [41:29<30:06,  1.59it/s]
3|3606|Loss: 0.8690997958183289:  56%|█████▌    | 3605/6470 [41:29<30:06,  1.59it/s]
3|3606|Loss: 0.8690997958183289:  56%|█████▌    | 3606/6470 [41:30<29:22,  1.62it/s]
3|3607|Loss: 0.8081640005111694:  56%|█████▌    | 3606/6470 [41:30<29:22,  1.62it/s]
3|3607|Loss: 0.8081640005111694:  56%|█████▌    | 3607/6470 [41:30<30:59,  1.54it/s]
3|3608|Loss: 0.8319470882415771:  56%|█████▌    | 3607/6470 [41:31<30:59,  1.54it/s]
3|3608|Loss: 0.8319470882415771:  56%|█████▌    | 3608/6470 [41:31<31:00,  1.54it/s]
3|3609|Loss: 1.0034844875335693:  56%|█████▌    | 3608/6470 [41:31<31:00,  1.54it/s]
3|3609|Loss: 1.0034844875335693:  56%|█████▌    | 3609/6470 [41:32<31:52,  1.50it/s]
3|3610|Loss: 0.8506553173065186:  56%|█████▌    | 3609/6470 [41:32<31:52,  1.50it/s]
3|3610|Loss: 0.8506553173065186:  56%|█████▌    | 3610/6470 [41:33<34:06,  1.40it/s]
3|3611|Loss: 0.8291409611701965:  56%|█████▌    | 3610/6470 [41:33<34:06,  1.40it/s]
3|3611|Loss: 0.8291409611701965:  56%|█████▌    | 3611/6470 [41:33<33:35,  1.42it/s]
3|3612|Loss: 0.9256563186645508:  56%|█████▌    | 3611/6470 [41:34<33:35,  1.42it/s]
3|3612|Loss: 0.9256563186645508:  56%|█████▌    | 3612/6470 [41:34<32:32,  1.46it/s]
3|3613|Loss: 0.8275375366210938:  56%|█████▌    | 3612/6470 [41:34<32:32,  1.46it/s]
3|3613|Loss: 0.8275375366210938:  56%|█████▌    | 3613/6470 [41:35<33:40,  1.41it/s]
3|3614|Loss: 0.8203793168067932:  56%|█████▌    | 3613/6470 [41:35<33:40,  1.41it/s]
3|3614|Loss: 0.8203793168067932:  56%|█████▌    | 3614/6470 [41:35<33:55,  1.40it/s]
3|3615|Loss: 0.7606512308120728:  56%|█████▌    | 3614/6470 [41:36<33:55,  1.40it/s]
3|3615|Loss: 0.7606512308120728:  56%|█████▌    | 3615/6470 [41:36<30:55,  1.54it/s]
3|3616|Loss: 0.9557119607925415:  56%|█████▌    | 3615/6470 [41:36<30:55,  1.54it/s]
3|3616|Loss: 0.9557119607925415:  56%|█████▌    | 3616/6470 [41:37<30:53,  1.54it/s]
3|3617|Loss: 0.8584838509559631:  56%|█████▌    | 3616/6470 [41:37<30:53,  1.54it/s]
3|3617|Loss: 0.8584838509559631:  56%|█████▌    | 3617/6470 [41:37<31:46,  1.50it/s]
3|3618|Loss: 0.9397489428520203:  56%|█████▌    | 3617/6470 [41:38<31:46,  1.50it/s]
3|3618|Loss: 0.9397489428520203:  56%|█████▌    | 3618/6470 [41:38<32:48,  1.45it/s]
3|3619|Loss: 0.7458434104919434:  56%|█████▌    | 3618/6470 [41:38<32:48,  1.45it/s]
3|3619|Loss: 0.7458434104919434:  56%|█████▌    | 3619/6470 [41:38<28:51,  1.65it/s]
3|3620|Loss: 0.8881860971450806:  56%|█████▌    | 3619/6470 [41:39<28:51,  1.65it/s]
3|3620|Loss: 0.8881860971450806:  56%|█████▌    | 3620/6470 [41:39<26:50,  1.77it/s]
3|3621|Loss: 1.0195832252502441:  56%|█████▌    | 3620/6470 [41:39<26:50,  1.77it/s]
3|3621|Loss: 1.0195832252502441:  56%|█████▌    | 3621/6470 [41:39<26:02,  1.82it/s]
3|3622|Loss: 0.8799811601638794:  56%|█████▌    | 3621/6470 [41:40<26:02,  1.82it/s]
3|3622|Loss: 0.8799811601638794:  56%|█████▌    | 3622/6470 [41:40<27:42,  1.71it/s]
3|3623|Loss: 0.8684746026992798:  56%|█████▌    | 3622/6470 [41:40<27:42,  1.71it/s]
3|3623|Loss: 0.8684746026992798:  56%|█████▌    | 3623/6470 [41:41<29:47,  1.59it/s]
3|3624|Loss: 0.7975977063179016:  56%|█████▌    | 3623/6470 [41:41<29:47,  1.59it/s]
3|3624|Loss: 0.7975977063179016:  56%|█████▌    | 3624/6470 [41:41<28:40,  1.65it/s]
3|3625|Loss: 0.7794696092605591:  56%|█████▌    | 3624/6470 [41:42<28:40,  1.65it/s]
3|3625|Loss: 0.7794696092605591:  56%|█████▌    | 3625/6470 [41:42<30:58,  1.53it/s]
3|3626|Loss: 0.838886559009552:  56%|█████▌    | 3625/6470 [41:43<30:58,  1.53it/s] 
3|3626|Loss: 0.838886559009552:  56%|█████▌    | 3626/6470 [41:43<32:29,  1.46it/s]
3|3627|Loss: 0.7941758036613464:  56%|█████▌    | 3626/6470 [41:43<32:29,  1.46it/s]
3|3627|Loss: 0.7941758036613464:  56%|█████▌    | 3627/6470 [41:44<32:01,  1.48it/s]
3|3628|Loss: 0.9525807499885559:  56%|█████▌    | 3627/6470 [41:44<32:01,  1.48it/s]
3|3628|Loss: 0.9525807499885559:  56%|█████▌    | 3628/6470 [41:44<30:48,  1.54it/s]
3|3629|Loss: 0.7575247883796692:  56%|█████▌    | 3628/6470 [41:44<30:48,  1.54it/s]
3|3629|Loss: 0.7575247883796692:  56%|█████▌    | 3629/6470 [41:45<31:29,  1.50it/s]
3|3630|Loss: 0.7370399832725525:  56%|█████▌    | 3629/6470 [41:45<31:29,  1.50it/s]
3|3630|Loss: 0.7370399832725525:  56%|█████▌    | 3630/6470 [41:46<32:27,  1.46it/s]
3|3631|Loss: 0.9186800122261047:  56%|█████▌    | 3630/6470 [41:46<32:27,  1.46it/s]
3|3631|Loss: 0.9186800122261047:  56%|█████▌    | 3631/6470 [41:46<31:27,  1.50it/s]
3|3632|Loss: 0.9001185894012451:  56%|█████▌    | 3631/6470 [41:47<31:27,  1.50it/s]
3|3632|Loss: 0.9001185894012451:  56%|█████▌    | 3632/6470 [41:47<30:55,  1.53it/s]
3|3633|Loss: 1.1458450555801392:  56%|█████▌    | 3632/6470 [41:47<30:55,  1.53it/s]
3|3633|Loss: 1.1458450555801392:  56%|█████▌    | 3633/6470 [41:48<31:49,  1.49it/s]
3|3634|Loss: 0.775920569896698:  56%|█████▌    | 3633/6470 [41:48<31:49,  1.49it/s] 
3|3634|Loss: 0.775920569896698:  56%|█████▌    | 3634/6470 [41:48<35:30,  1.33it/s]
3|3635|Loss: 0.9395900964736938:  56%|█████▌    | 3634/6470 [41:49<35:30,  1.33it/s]
3|3635|Loss: 0.9395900964736938:  56%|█████▌    | 3635/6470 [41:49<35:48,  1.32it/s]
3|3636|Loss: 0.7893664240837097:  56%|█████▌    | 3635/6470 [41:50<35:48,  1.32it/s]
3|3636|Loss: 0.7893664240837097:  56%|█████▌    | 3636/6470 [41:50<35:29,  1.33it/s]
3|3637|Loss: 0.9875367879867554:  56%|█████▌    | 3636/6470 [41:50<35:29,  1.33it/s]
3|3637|Loss: 0.9875367879867554:  56%|█████▌    | 3637/6470 [41:51<32:58,  1.43it/s]
3|3638|Loss: 0.8402254581451416:  56%|█████▌    | 3637/6470 [41:51<32:58,  1.43it/s]
3|3638|Loss: 0.8402254581451416:  56%|█████▌    | 3638/6470 [41:51<30:23,  1.55it/s]
3|3639|Loss: 0.8693731427192688:  56%|█████▌    | 3638/6470 [41:51<30:23,  1.55it/s]
3|3639|Loss: 0.8693731427192688:  56%|█████▌    | 3639/6470 [41:52<30:54,  1.53it/s]
3|3640|Loss: 0.8396292924880981:  56%|█████▌    | 3639/6470 [41:52<30:54,  1.53it/s]
3|3640|Loss: 0.8396292924880981:  56%|█████▋    | 3640/6470 [41:52<31:48,  1.48it/s]
3|3641|Loss: 0.884450376033783:  56%|█████▋    | 3640/6470 [41:53<31:48,  1.48it/s] 
3|3641|Loss: 0.884450376033783:  56%|█████▋    | 3641/6470 [41:53<30:02,  1.57it/s]
3|3642|Loss: 1.003373146057129:  56%|█████▋    | 3641/6470 [41:53<30:02,  1.57it/s]
3|3642|Loss: 1.003373146057129:  56%|█████▋    | 3642/6470 [41:54<30:28,  1.55it/s]
3|3643|Loss: 0.8470334410667419:  56%|█████▋    | 3642/6470 [41:54<30:28,  1.55it/s]
3|3643|Loss: 0.8470334410667419:  56%|█████▋    | 3643/6470 [41:54<31:48,  1.48it/s]
3|3644|Loss: 0.9985239505767822:  56%|█████▋    | 3643/6470 [41:55<31:48,  1.48it/s]
3|3644|Loss: 0.9985239505767822:  56%|█████▋    | 3644/6470 [41:55<30:10,  1.56it/s]
3|3645|Loss: 0.8211366534233093:  56%|█████▋    | 3644/6470 [41:55<30:10,  1.56it/s]
3|3645|Loss: 0.8211366534233093:  56%|█████▋    | 3645/6470 [41:56<30:25,  1.55it/s]
3|3646|Loss: 0.6428708434104919:  56%|█████▋    | 3645/6470 [41:56<30:25,  1.55it/s]
3|3646|Loss: 0.6428708434104919:  56%|█████▋    | 3646/6470 [41:56<31:41,  1.49it/s]
3|3647|Loss: 0.8138003945350647:  56%|█████▋    | 3646/6470 [41:57<31:41,  1.49it/s]
3|3647|Loss: 0.8138003945350647:  56%|█████▋    | 3647/6470 [41:57<30:02,  1.57it/s]
3|3648|Loss: 0.6311266422271729:  56%|█████▋    | 3647/6470 [41:57<30:02,  1.57it/s]
3|3648|Loss: 0.6311266422271729:  56%|█████▋    | 3648/6470 [41:57<28:40,  1.64it/s]
3|3649|Loss: 1.0312702655792236:  56%|█████▋    | 3648/6470 [41:58<28:40,  1.64it/s]
3|3649|Loss: 1.0312702655792236:  56%|█████▋    | 3649/6470 [41:58<28:50,  1.63it/s]
3|3650|Loss: 0.8806853294372559:  56%|█████▋    | 3649/6470 [41:58<28:50,  1.63it/s]
3|3650|Loss: 0.8806853294372559:  56%|█████▋    | 3650/6470 [41:59<29:01,  1.62it/s]
3|3651|Loss: 0.7873191833496094:  56%|█████▋    | 3650/6470 [41:59<29:01,  1.62it/s]
3|3651|Loss: 0.7873191833496094:  56%|█████▋    | 3651/6470 [41:59<29:18,  1.60it/s]
3|3652|Loss: 0.836946427822113:  56%|█████▋    | 3651/6470 [42:00<29:18,  1.60it/s] 
3|3652|Loss: 0.836946427822113:  56%|█████▋    | 3652/6470 [42:00<30:32,  1.54it/s]
3|3653|Loss: 0.9829010963439941:  56%|█████▋    | 3652/6470 [42:00<30:32,  1.54it/s]
3|3653|Loss: 0.9829010963439941:  56%|█████▋    | 3653/6470 [42:01<31:16,  1.50it/s]
3|3654|Loss: 0.8230684995651245:  56%|█████▋    | 3653/6470 [42:01<31:16,  1.50it/s]
3|3654|Loss: 0.8230684995651245:  56%|█████▋    | 3654/6470 [42:01<32:18,  1.45it/s]
3|3655|Loss: 0.9223378300666809:  56%|█████▋    | 3654/6470 [42:02<32:18,  1.45it/s]
3|3655|Loss: 0.9223378300666809:  56%|█████▋    | 3655/6470 [42:02<32:00,  1.47it/s]
3|3656|Loss: 0.8652292490005493:  56%|█████▋    | 3655/6470 [42:02<32:00,  1.47it/s]
3|3656|Loss: 0.8652292490005493:  57%|█████▋    | 3656/6470 [42:03<30:10,  1.55it/s]
3|3657|Loss: 0.8175458312034607:  57%|█████▋    | 3656/6470 [42:03<30:10,  1.55it/s]
3|3657|Loss: 0.8175458312034607:  57%|█████▋    | 3657/6470 [42:03<32:02,  1.46it/s]
3|3658|Loss: 0.8539843559265137:  57%|█████▋    | 3657/6470 [42:04<32:02,  1.46it/s]
3|3658|Loss: 0.8539843559265137:  57%|█████▋    | 3658/6470 [42:04<32:37,  1.44it/s]
3|3659|Loss: 0.7964362502098083:  57%|█████▋    | 3658/6470 [42:05<32:37,  1.44it/s]
3|3659|Loss: 0.7964362502098083:  57%|█████▋    | 3659/6470 [42:05<29:07,  1.61it/s]
3|3660|Loss: 0.8407014608383179:  57%|█████▋    | 3659/6470 [42:05<29:07,  1.61it/s]
3|3660|Loss: 0.8407014608383179:  57%|█████▋    | 3660/6470 [42:05<29:49,  1.57it/s]
3|3661|Loss: 0.8285167217254639:  57%|█████▋    | 3660/6470 [42:06<29:49,  1.57it/s]
3|3661|Loss: 0.8285167217254639:  57%|█████▋    | 3661/6470 [42:06<30:09,  1.55it/s]
3|3662|Loss: 0.8429736495018005:  57%|█████▋    | 3661/6470 [42:06<30:09,  1.55it/s]
3|3662|Loss: 0.8429736495018005:  57%|█████▋    | 3662/6470 [42:07<30:40,  1.53it/s]
3|3663|Loss: 0.8652672171592712:  57%|█████▋    | 3662/6470 [42:07<30:40,  1.53it/s]
3|3663|Loss: 0.8652672171592712:  57%|█████▋    | 3663/6470 [42:07<31:54,  1.47it/s]
3|3664|Loss: 0.8567894697189331:  57%|█████▋    | 3663/6470 [42:08<31:54,  1.47it/s]
3|3664|Loss: 0.8567894697189331:  57%|█████▋    | 3664/6470 [42:08<33:31,  1.40it/s]
3|3665|Loss: 0.946509599685669:  57%|█████▋    | 3664/6470 [42:09<33:31,  1.40it/s] 
3|3665|Loss: 0.946509599685669:  57%|█████▋    | 3665/6470 [42:09<33:42,  1.39it/s]
3|3666|Loss: 0.9623326063156128:  57%|█████▋    | 3665/6470 [42:09<33:42,  1.39it/s]
3|3666|Loss: 0.9623326063156128:  57%|█████▋    | 3666/6470 [42:10<32:46,  1.43it/s]
3|3667|Loss: 0.8474531769752502:  57%|█████▋    | 3666/6470 [42:10<32:46,  1.43it/s]
3|3667|Loss: 0.8474531769752502:  57%|█████▋    | 3667/6470 [42:10<33:04,  1.41it/s]
3|3668|Loss: 0.9225786924362183:  57%|█████▋    | 3667/6470 [42:11<33:04,  1.41it/s]
3|3668|Loss: 0.9225786924362183:  57%|█████▋    | 3668/6470 [42:11<32:24,  1.44it/s]
3|3669|Loss: 0.7287634611129761:  57%|█████▋    | 3668/6470 [42:11<32:24,  1.44it/s]
3|3669|Loss: 0.7287634611129761:  57%|█████▋    | 3669/6470 [42:12<34:02,  1.37it/s]
3|3670|Loss: 0.7079638242721558:  57%|█████▋    | 3669/6470 [42:12<34:02,  1.37it/s]
3|3670|Loss: 0.7079638242721558:  57%|█████▋    | 3670/6470 [42:12<33:08,  1.41it/s]
3|3671|Loss: 0.934942901134491:  57%|█████▋    | 3670/6470 [42:13<33:08,  1.41it/s] 
3|3671|Loss: 0.934942901134491:  57%|█████▋    | 3671/6470 [42:13<33:15,  1.40it/s]
3|3672|Loss: 0.8049414753913879:  57%|█████▋    | 3671/6470 [42:14<33:15,  1.40it/s]
3|3672|Loss: 0.8049414753913879:  57%|█████▋    | 3672/6470 [42:14<34:14,  1.36it/s]
3|3673|Loss: 0.7890995740890503:  57%|█████▋    | 3672/6470 [42:14<34:14,  1.36it/s]
3|3673|Loss: 0.7890995740890503:  57%|█████▋    | 3673/6470 [42:15<32:49,  1.42it/s]
3|3674|Loss: 0.9158363342285156:  57%|█████▋    | 3673/6470 [42:15<32:49,  1.42it/s]
3|3674|Loss: 0.9158363342285156:  57%|█████▋    | 3674/6470 [42:15<32:48,  1.42it/s]
3|3675|Loss: 0.9498128890991211:  57%|█████▋    | 3674/6470 [42:16<32:48,  1.42it/s]
3|3675|Loss: 0.9498128890991211:  57%|█████▋    | 3675/6470 [42:16<32:14,  1.44it/s]
3|3676|Loss: 0.8775331377983093:  57%|█████▋    | 3675/6470 [42:16<32:14,  1.44it/s]
3|3676|Loss: 0.8775331377983093:  57%|█████▋    | 3676/6470 [42:17<31:23,  1.48it/s]
3|3677|Loss: 0.7107027173042297:  57%|█████▋    | 3676/6470 [42:17<31:23,  1.48it/s]
3|3677|Loss: 0.7107027173042297:  57%|█████▋    | 3677/6470 [42:18<34:24,  1.35it/s]
3|3678|Loss: 0.9201951622962952:  57%|█████▋    | 3677/6470 [42:18<34:24,  1.35it/s]
3|3678|Loss: 0.9201951622962952:  57%|█████▋    | 3678/6470 [42:18<33:13,  1.40it/s]
3|3679|Loss: 0.8007714748382568:  57%|█████▋    | 3678/6470 [42:19<33:13,  1.40it/s]
3|3679|Loss: 0.8007714748382568:  57%|█████▋    | 3679/6470 [42:19<33:02,  1.41it/s]
3|3680|Loss: 0.8287568092346191:  57%|█████▋    | 3679/6470 [42:19<33:02,  1.41it/s]
3|3680|Loss: 0.8287568092346191:  57%|█████▋    | 3680/6470 [42:20<34:23,  1.35it/s]
3|3681|Loss: 0.9296354055404663:  57%|█████▋    | 3680/6470 [42:20<34:23,  1.35it/s]
3|3681|Loss: 0.9296354055404663:  57%|█████▋    | 3681/6470 [42:20<34:34,  1.34it/s]
3|3682|Loss: 0.6968168020248413:  57%|█████▋    | 3681/6470 [42:21<34:34,  1.34it/s]
3|3682|Loss: 0.6968168020248413:  57%|█████▋    | 3682/6470 [42:22<44:09,  1.05it/s]
3|3683|Loss: 0.9126212000846863:  57%|█████▋    | 3682/6470 [42:22<44:09,  1.05it/s]
3|3683|Loss: 0.9126212000846863:  57%|█████▋    | 3683/6470 [42:23<44:56,  1.03it/s]
3|3684|Loss: 0.8461547493934631:  57%|█████▋    | 3683/6470 [42:23<44:56,  1.03it/s]
3|3684|Loss: 0.8461547493934631:  57%|█████▋    | 3684/6470 [42:23<39:59,  1.16it/s]
3|3685|Loss: 0.7817291021347046:  57%|█████▋    | 3684/6470 [42:24<39:59,  1.16it/s]
3|3685|Loss: 0.7817291021347046:  57%|█████▋    | 3685/6470 [42:24<37:51,  1.23it/s]
3|3686|Loss: 0.7574766874313354:  57%|█████▋    | 3685/6470 [42:25<37:51,  1.23it/s]
3|3686|Loss: 0.7574766874313354:  57%|█████▋    | 3686/6470 [42:25<38:19,  1.21it/s]
3|3687|Loss: 0.8149765729904175:  57%|█████▋    | 3686/6470 [42:25<38:19,  1.21it/s]
3|3687|Loss: 0.8149765729904175:  57%|█████▋    | 3687/6470 [42:26<36:14,  1.28it/s]
3|3688|Loss: 0.8000094294548035:  57%|█████▋    | 3687/6470 [42:26<36:14,  1.28it/s]
3|3688|Loss: 0.8000094294548035:  57%|█████▋    | 3688/6470 [42:26<34:54,  1.33it/s]
3|3689|Loss: 0.9734618663787842:  57%|█████▋    | 3688/6470 [42:27<34:54,  1.33it/s]
3|3689|Loss: 0.9734618663787842:  57%|█████▋    | 3689/6470 [42:27<34:12,  1.35it/s]
3|3690|Loss: 0.9295012950897217:  57%|█████▋    | 3689/6470 [42:28<34:12,  1.35it/s]
3|3690|Loss: 0.9295012950897217:  57%|█████▋    | 3690/6470 [42:28<35:07,  1.32it/s]
3|3691|Loss: 0.8151530027389526:  57%|█████▋    | 3690/6470 [42:28<35:07,  1.32it/s]
3|3691|Loss: 0.8151530027389526:  57%|█████▋    | 3691/6470 [42:29<35:11,  1.32it/s]
3|3692|Loss: 0.8603625893592834:  57%|█████▋    | 3691/6470 [42:29<35:11,  1.32it/s]
3|3692|Loss: 0.8603625893592834:  57%|█████▋    | 3692/6470 [42:29<34:53,  1.33it/s]
3|3693|Loss: 0.8958689570426941:  57%|█████▋    | 3692/6470 [42:30<34:53,  1.33it/s]
3|3693|Loss: 0.8958689570426941:  57%|█████▋    | 3693/6470 [42:30<34:14,  1.35it/s]
3|3694|Loss: 0.8634829521179199:  57%|█████▋    | 3693/6470 [42:31<34:14,  1.35it/s]
3|3694|Loss: 0.8634829521179199:  57%|█████▋    | 3694/6470 [42:31<34:14,  1.35it/s]
3|3695|Loss: 0.9163903594017029:  57%|█████▋    | 3694/6470 [42:31<34:14,  1.35it/s]
3|3695|Loss: 0.9163903594017029:  57%|█████▋    | 3695/6470 [42:32<33:59,  1.36it/s]
3|3696|Loss: 0.8114109635353088:  57%|█████▋    | 3695/6470 [42:32<33:59,  1.36it/s]
3|3696|Loss: 0.8114109635353088:  57%|█████▋    | 3696/6470 [42:32<33:36,  1.38it/s]
3|3697|Loss: 0.7918742299079895:  57%|█████▋    | 3696/6470 [42:33<33:36,  1.38it/s]
3|3697|Loss: 0.7918742299079895:  57%|█████▋    | 3697/6470 [42:33<34:34,  1.34it/s]
3|3698|Loss: 0.858167827129364:  57%|█████▋    | 3697/6470 [42:33<34:34,  1.34it/s] 
3|3698|Loss: 0.858167827129364:  57%|█████▋    | 3698/6470 [42:34<32:47,  1.41it/s]
3|3699|Loss: 0.7266312837600708:  57%|█████▋    | 3698/6470 [42:34<32:47,  1.41it/s]
3|3699|Loss: 0.7266312837600708:  57%|█████▋    | 3699/6470 [42:34<31:39,  1.46it/s]
3|3700|Loss: 0.7654139399528503:  57%|█████▋    | 3699/6470 [42:35<31:39,  1.46it/s]
3|3700|Loss: 0.7654139399528503:  57%|█████▋    | 3700/6470 [42:35<33:11,  1.39it/s]
3|3701|Loss: 0.8568549156188965:  57%|█████▋    | 3700/6470 [42:36<33:11,  1.39it/s]
3|3701|Loss: 0.8568549156188965:  57%|█████▋    | 3701/6470 [42:36<33:55,  1.36it/s]
3|3702|Loss: 0.9175412654876709:  57%|█████▋    | 3701/6470 [42:36<33:55,  1.36it/s]
3|3702|Loss: 0.9175412654876709:  57%|█████▋    | 3702/6470 [42:37<34:13,  1.35it/s]
3|3703|Loss: 0.8722527623176575:  57%|█████▋    | 3702/6470 [42:37<34:13,  1.35it/s]
3|3703|Loss: 0.8722527623176575:  57%|█████▋    | 3703/6470 [42:37<35:10,  1.31it/s]
3|3704|Loss: 0.8454247713088989:  57%|█████▋    | 3703/6470 [42:38<35:10,  1.31it/s]
3|3704|Loss: 0.8454247713088989:  57%|█████▋    | 3704/6470 [42:38<37:04,  1.24it/s]
3|3705|Loss: 0.9164170026779175:  57%|█████▋    | 3704/6470 [42:39<37:04,  1.24it/s]
3|3705|Loss: 0.9164170026779175:  57%|█████▋    | 3705/6470 [42:39<35:20,  1.30it/s]
3|3706|Loss: 0.8136353492736816:  57%|█████▋    | 3705/6470 [42:39<35:20,  1.30it/s]
3|3706|Loss: 0.8136353492736816:  57%|█████▋    | 3706/6470 [42:40<32:29,  1.42it/s]
3|3707|Loss: 0.8476290106773376:  57%|█████▋    | 3706/6470 [42:40<32:29,  1.42it/s]
3|3707|Loss: 0.8476290106773376:  57%|█████▋    | 3707/6470 [42:40<32:25,  1.42it/s]
3|3708|Loss: 0.8954489231109619:  57%|█████▋    | 3707/6470 [42:41<32:25,  1.42it/s]
3|3708|Loss: 0.8954489231109619:  57%|█████▋    | 3708/6470 [42:41<33:07,  1.39it/s]
3|3709|Loss: 0.8262510895729065:  57%|█████▋    | 3708/6470 [42:42<33:07,  1.39it/s]
3|3709|Loss: 0.8262510895729065:  57%|█████▋    | 3709/6470 [42:42<34:30,  1.33it/s]
3|3710|Loss: 0.8914082050323486:  57%|█████▋    | 3709/6470 [42:42<34:30,  1.33it/s]
3|3710|Loss: 0.8914082050323486:  57%|█████▋    | 3710/6470 [42:43<34:44,  1.32it/s]
3|3711|Loss: 0.8397259712219238:  57%|█████▋    | 3710/6470 [42:43<34:44,  1.32it/s]
3|3711|Loss: 0.8397259712219238:  57%|█████▋    | 3711/6470 [42:43<33:15,  1.38it/s]
3|3712|Loss: 0.7521242499351501:  57%|█████▋    | 3711/6470 [42:44<33:15,  1.38it/s]
3|3712|Loss: 0.7521242499351501:  57%|█████▋    | 3712/6470 [42:44<32:46,  1.40it/s]
3|3713|Loss: 0.7427543997764587:  57%|█████▋    | 3712/6470 [42:44<32:46,  1.40it/s]
3|3713|Loss: 0.7427543997764587:  57%|█████▋    | 3713/6470 [42:45<31:55,  1.44it/s]
3|3714|Loss: 0.8440821170806885:  57%|█████▋    | 3713/6470 [42:45<31:55,  1.44it/s]
3|3714|Loss: 0.8440821170806885:  57%|█████▋    | 3714/6470 [42:45<30:55,  1.49it/s]
3|3715|Loss: 0.7583821415901184:  57%|█████▋    | 3714/6470 [42:46<30:55,  1.49it/s]
3|3715|Loss: 0.7583821415901184:  57%|█████▋    | 3715/6470 [42:46<32:59,  1.39it/s]
3|3716|Loss: 1.00125253200531:  57%|█████▋    | 3715/6470 [42:47<32:59,  1.39it/s]  
3|3716|Loss: 1.00125253200531:  57%|█████▋    | 3716/6470 [42:47<33:40,  1.36it/s]
3|3717|Loss: 0.9429406523704529:  57%|█████▋    | 3716/6470 [42:47<33:40,  1.36it/s]
3|3717|Loss: 0.9429406523704529:  57%|█████▋    | 3717/6470 [42:47<31:29,  1.46it/s]
3|3718|Loss: 0.7777286767959595:  57%|█████▋    | 3717/6470 [42:48<31:29,  1.46it/s]
3|3718|Loss: 0.7777286767959595:  57%|█████▋    | 3718/6470 [42:48<30:20,  1.51it/s]
3|3719|Loss: 0.9863286018371582:  57%|█████▋    | 3718/6470 [42:48<30:20,  1.51it/s]
3|3719|Loss: 0.9863286018371582:  57%|█████▋    | 3719/6470 [42:49<29:47,  1.54it/s]
3|3720|Loss: 0.8249236345291138:  57%|█████▋    | 3719/6470 [42:49<29:47,  1.54it/s]
3|3720|Loss: 0.8249236345291138:  57%|█████▋    | 3720/6470 [42:49<29:04,  1.58it/s]
3|3721|Loss: 0.8537341356277466:  57%|█████▋    | 3720/6470 [42:50<29:04,  1.58it/s]
3|3721|Loss: 0.8537341356277466:  58%|█████▊    | 3721/6470 [42:50<31:33,  1.45it/s]
3|3722|Loss: 0.8406780362129211:  58%|█████▊    | 3721/6470 [42:51<31:33,  1.45it/s]
3|3722|Loss: 0.8406780362129211:  58%|█████▊    | 3722/6470 [42:51<32:14,  1.42it/s]
3|3723|Loss: 0.71504145860672:  58%|█████▊    | 3722/6470 [42:51<32:14,  1.42it/s]  
3|3723|Loss: 0.71504145860672:  58%|█████▊    | 3723/6470 [42:52<35:23,  1.29it/s]
3|3724|Loss: 0.8911706805229187:  58%|█████▊    | 3723/6470 [42:52<35:23,  1.29it/s]
3|3724|Loss: 0.8911706805229187:  58%|█████▊    | 3724/6470 [42:52<33:55,  1.35it/s]
3|3725|Loss: 0.843212902545929:  58%|█████▊    | 3724/6470 [42:53<33:55,  1.35it/s] 
3|3725|Loss: 0.843212902545929:  58%|█████▊    | 3725/6470 [42:53<32:02,  1.43it/s]
3|3726|Loss: 0.8472337126731873:  58%|█████▊    | 3725/6470 [42:53<32:02,  1.43it/s]
3|3726|Loss: 0.8472337126731873:  58%|█████▊    | 3726/6470 [42:54<32:53,  1.39it/s]
3|3727|Loss: 0.9772907495498657:  58%|█████▊    | 3726/6470 [42:54<32:53,  1.39it/s]
3|3727|Loss: 0.9772907495498657:  58%|█████▊    | 3727/6470 [42:54<32:03,  1.43it/s]
3|3728|Loss: 0.8390892744064331:  58%|█████▊    | 3727/6470 [42:55<32:03,  1.43it/s]
3|3728|Loss: 0.8390892744064331:  58%|█████▊    | 3728/6470 [42:55<30:22,  1.50it/s]
3|3729|Loss: 0.9307429194450378:  58%|█████▊    | 3728/6470 [42:55<30:22,  1.50it/s]
3|3729|Loss: 0.9307429194450378:  58%|█████▊    | 3729/6470 [42:56<32:25,  1.41it/s]
3|3730|Loss: 0.8951156735420227:  58%|█████▊    | 3729/6470 [42:56<32:25,  1.41it/s]
3|3730|Loss: 0.8951156735420227:  58%|█████▊    | 3730/6470 [42:57<32:54,  1.39it/s]
3|3731|Loss: 0.9045676589012146:  58%|█████▊    | 3730/6470 [42:57<32:54,  1.39it/s]
3|3731|Loss: 0.9045676589012146:  58%|█████▊    | 3731/6470 [42:57<31:19,  1.46it/s]
3|3732|Loss: 0.8723958730697632:  58%|█████▊    | 3731/6470 [42:58<31:19,  1.46it/s]
3|3732|Loss: 0.8723958730697632:  58%|█████▊    | 3732/6470 [42:58<31:13,  1.46it/s]
3|3733|Loss: 1.0634161233901978:  58%|█████▊    | 3732/6470 [42:58<31:13,  1.46it/s]
3|3733|Loss: 1.0634161233901978:  58%|█████▊    | 3733/6470 [42:59<31:26,  1.45it/s]
3|3734|Loss: 0.8893327116966248:  58%|█████▊    | 3733/6470 [42:59<31:26,  1.45it/s]
3|3734|Loss: 0.8893327116966248:  58%|█████▊    | 3734/6470 [42:59<30:38,  1.49it/s]
3|3735|Loss: 0.8033002018928528:  58%|█████▊    | 3734/6470 [43:00<30:38,  1.49it/s]
3|3735|Loss: 0.8033002018928528:  58%|█████▊    | 3735/6470 [43:00<31:02,  1.47it/s]
3|3736|Loss: 0.7877395749092102:  58%|█████▊    | 3735/6470 [43:00<31:02,  1.47it/s]
3|3736|Loss: 0.7877395749092102:  58%|█████▊    | 3736/6470 [43:01<33:00,  1.38it/s]
3|3737|Loss: 0.8210278153419495:  58%|█████▊    | 3736/6470 [43:01<33:00,  1.38it/s]
3|3737|Loss: 0.8210278153419495:  58%|█████▊    | 3737/6470 [43:01<32:21,  1.41it/s]
3|3738|Loss: 0.9141902923583984:  58%|█████▊    | 3737/6470 [43:02<32:21,  1.41it/s]
3|3738|Loss: 0.9141902923583984:  58%|█████▊    | 3738/6470 [43:02<31:06,  1.46it/s]
3|3739|Loss: 0.8269464373588562:  58%|█████▊    | 3738/6470 [43:02<31:06,  1.46it/s]
3|3739|Loss: 0.8269464373588562:  58%|█████▊    | 3739/6470 [43:03<31:32,  1.44it/s]
3|3740|Loss: 0.8223518133163452:  58%|█████▊    | 3739/6470 [43:03<31:32,  1.44it/s]
3|3740|Loss: 0.8223518133163452:  58%|█████▊    | 3740/6470 [43:03<30:19,  1.50it/s]
3|3741|Loss: 0.8936541080474854:  58%|█████▊    | 3740/6470 [43:04<30:19,  1.50it/s]
3|3741|Loss: 0.8936541080474854:  58%|█████▊    | 3741/6470 [43:04<31:03,  1.46it/s]
3|3742|Loss: 1.00399649143219:  58%|█████▊    | 3741/6470 [43:05<31:03,  1.46it/s]  
3|3742|Loss: 1.00399649143219:  58%|█████▊    | 3742/6470 [43:05<32:27,  1.40it/s]
3|3743|Loss: 0.7768122553825378:  58%|█████▊    | 3742/6470 [43:05<32:27,  1.40it/s]
3|3743|Loss: 0.7768122553825378:  58%|█████▊    | 3743/6470 [43:05<30:40,  1.48it/s]
3|3744|Loss: 0.9731806516647339:  58%|█████▊    | 3743/6470 [43:06<30:40,  1.48it/s]
3|3744|Loss: 0.9731806516647339:  58%|█████▊    | 3744/6470 [43:06<30:25,  1.49it/s]
3|3745|Loss: 0.8990683555603027:  58%|█████▊    | 3744/6470 [43:07<30:25,  1.49it/s]
3|3745|Loss: 0.8990683555603027:  58%|█████▊    | 3745/6470 [43:07<31:50,  1.43it/s]
3|3746|Loss: 0.7604002952575684:  58%|█████▊    | 3745/6470 [43:07<31:50,  1.43it/s]
3|3746|Loss: 0.7604002952575684:  58%|█████▊    | 3746/6470 [43:08<32:28,  1.40it/s]
3|3747|Loss: 0.9573873281478882:  58%|█████▊    | 3746/6470 [43:08<32:28,  1.40it/s]
3|3747|Loss: 0.9573873281478882:  58%|█████▊    | 3747/6470 [43:08<30:57,  1.47it/s]
3|3748|Loss: 0.858389675617218:  58%|█████▊    | 3747/6470 [43:09<30:57,  1.47it/s] 
3|3748|Loss: 0.858389675617218:  58%|█████▊    | 3748/6470 [43:09<30:59,  1.46it/s]
3|3749|Loss: 0.8597472310066223:  58%|█████▊    | 3748/6470 [43:09<30:59,  1.46it/s]
3|3749|Loss: 0.8597472310066223:  58%|█████▊    | 3749/6470 [43:10<30:02,  1.51it/s]
3|3750|Loss: 0.9166615009307861:  58%|█████▊    | 3749/6470 [43:10<30:02,  1.51it/s]
3|3750|Loss: 0.9166615009307861:  58%|█████▊    | 3750/6470 [43:10<28:13,  1.61it/s]
3|3751|Loss: 0.8682711124420166:  58%|█████▊    | 3750/6470 [43:10<28:13,  1.61it/s]
3|3751|Loss: 0.8682711124420166:  58%|█████▊    | 3751/6470 [43:11<28:50,  1.57it/s]
3|3752|Loss: 0.9123191833496094:  58%|█████▊    | 3751/6470 [43:11<28:50,  1.57it/s]
3|3752|Loss: 0.9123191833496094:  58%|█████▊    | 3752/6470 [43:11<28:52,  1.57it/s]
3|3753|Loss: 0.9287768602371216:  58%|█████▊    | 3752/6470 [43:12<28:52,  1.57it/s]
3|3753|Loss: 0.9287768602371216:  58%|█████▊    | 3753/6470 [43:12<28:33,  1.59it/s]
3|3754|Loss: 0.8865602612495422:  58%|█████▊    | 3753/6470 [43:12<28:33,  1.59it/s]
3|3754|Loss: 0.8865602612495422:  58%|█████▊    | 3754/6470 [43:13<29:24,  1.54it/s]
3|3755|Loss: 0.7831616997718811:  58%|█████▊    | 3754/6470 [43:13<29:24,  1.54it/s]
3|3755|Loss: 0.7831616997718811:  58%|█████▊    | 3755/6470 [43:13<29:29,  1.53it/s]
3|3756|Loss: 0.9529829621315002:  58%|█████▊    | 3755/6470 [43:14<29:29,  1.53it/s]
3|3756|Loss: 0.9529829621315002:  58%|█████▊    | 3756/6470 [43:14<27:54,  1.62it/s]
3|3757|Loss: 0.9057949781417847:  58%|█████▊    | 3756/6470 [43:14<27:54,  1.62it/s]
3|3757|Loss: 0.9057949781417847:  58%|█████▊    | 3757/6470 [43:15<29:53,  1.51it/s]
3|3758|Loss: 0.9420085549354553:  58%|█████▊    | 3757/6470 [43:15<29:53,  1.51it/s]
3|3758|Loss: 0.9420085549354553:  58%|█████▊    | 3758/6470 [43:15<28:43,  1.57it/s]
3|3759|Loss: 0.8062219023704529:  58%|█████▊    | 3758/6470 [43:16<28:43,  1.57it/s]
3|3759|Loss: 0.8062219023704529:  58%|█████▊    | 3759/6470 [43:16<30:56,  1.46it/s]
3|3760|Loss: 0.7432337403297424:  58%|█████▊    | 3759/6470 [43:16<30:56,  1.46it/s]
3|3760|Loss: 0.7432337403297424:  58%|█████▊    | 3760/6470 [43:17<33:11,  1.36it/s]
3|3761|Loss: 0.8765034675598145:  58%|█████▊    | 3760/6470 [43:17<33:11,  1.36it/s]
3|3761|Loss: 0.8765034675598145:  58%|█████▊    | 3761/6470 [43:17<31:18,  1.44it/s]
3|3762|Loss: 0.867031991481781:  58%|█████▊    | 3761/6470 [43:18<31:18,  1.44it/s] 
3|3762|Loss: 0.867031991481781:  58%|█████▊    | 3762/6470 [43:18<31:24,  1.44it/s]
3|3763|Loss: 0.8514276146888733:  58%|█████▊    | 3762/6470 [43:19<31:24,  1.44it/s]
3|3763|Loss: 0.8514276146888733:  58%|█████▊    | 3763/6470 [43:19<30:17,  1.49it/s]
3|3764|Loss: 0.952690839767456:  58%|█████▊    | 3763/6470 [43:19<30:17,  1.49it/s] 
3|3764|Loss: 0.952690839767456:  58%|█████▊    | 3764/6470 [43:19<30:56,  1.46it/s]
3|3765|Loss: 0.9348570108413696:  58%|█████▊    | 3764/6470 [43:20<30:56,  1.46it/s]
3|3765|Loss: 0.9348570108413696:  58%|█████▊    | 3765/6470 [43:20<30:22,  1.48it/s]
3|3766|Loss: 0.938787043094635:  58%|█████▊    | 3765/6470 [43:21<30:22,  1.48it/s] 
3|3766|Loss: 0.938787043094635:  58%|█████▊    | 3766/6470 [43:21<31:47,  1.42it/s]
3|3767|Loss: 0.7348909378051758:  58%|█████▊    | 3766/6470 [43:21<31:47,  1.42it/s]
3|3767|Loss: 0.7348909378051758:  58%|█████▊    | 3767/6470 [43:21<28:14,  1.60it/s]
3|3768|Loss: 0.8607898354530334:  58%|█████▊    | 3767/6470 [43:22<28:14,  1.60it/s]
3|3768|Loss: 0.8607898354530334:  58%|█████▊    | 3768/6470 [43:22<27:17,  1.65it/s]
3|3769|Loss: 0.9081882834434509:  58%|█████▊    | 3768/6470 [43:22<27:17,  1.65it/s]
3|3769|Loss: 0.9081882834434509:  58%|█████▊    | 3769/6470 [43:22<26:29,  1.70it/s]
3|3770|Loss: 0.9866230487823486:  58%|█████▊    | 3769/6470 [43:23<26:29,  1.70it/s]
3|3770|Loss: 0.9866230487823486:  58%|█████▊    | 3770/6470 [43:23<26:53,  1.67it/s]
3|3771|Loss: 0.9579213857650757:  58%|█████▊    | 3770/6470 [43:23<26:53,  1.67it/s]
3|3771|Loss: 0.9579213857650757:  58%|█████▊    | 3771/6470 [43:24<28:34,  1.57it/s]
3|3772|Loss: 0.7866289615631104:  58%|█████▊    | 3771/6470 [43:24<28:34,  1.57it/s]
3|3772|Loss: 0.7866289615631104:  58%|█████▊    | 3772/6470 [43:24<26:54,  1.67it/s]
3|3773|Loss: 0.800815761089325:  58%|█████▊    | 3772/6470 [43:25<26:54,  1.67it/s] 
3|3773|Loss: 0.800815761089325:  58%|█████▊    | 3773/6470 [43:25<25:59,  1.73it/s]
3|3774|Loss: 0.876675546169281:  58%|█████▊    | 3773/6470 [43:25<25:59,  1.73it/s]
3|3774|Loss: 0.876675546169281:  58%|█████▊    | 3774/6470 [43:25<26:23,  1.70it/s]
3|3775|Loss: 0.6200593709945679:  58%|█████▊    | 3774/6470 [43:26<26:23,  1.70it/s]
3|3775|Loss: 0.6200593709945679:  58%|█████▊    | 3775/6470 [43:26<28:01,  1.60it/s]
3|3776|Loss: 0.8997125029563904:  58%|█████▊    | 3775/6470 [43:27<28:01,  1.60it/s]
3|3776|Loss: 0.8997125029563904:  58%|█████▊    | 3776/6470 [43:27<30:49,  1.46it/s]
3|3777|Loss: 0.5986971259117126:  58%|█████▊    | 3776/6470 [43:28<30:49,  1.46it/s]
3|3777|Loss: 0.5986971259117126:  58%|█████▊    | 3777/6470 [43:28<34:38,  1.30it/s]
3|3778|Loss: 0.8656131625175476:  58%|█████▊    | 3777/6470 [43:28<34:38,  1.30it/s]
3|3778|Loss: 0.8656131625175476:  58%|█████▊    | 3778/6470 [43:29<33:41,  1.33it/s]
3|3779|Loss: 0.8906621336936951:  58%|█████▊    | 3778/6470 [43:29<33:41,  1.33it/s]
3|3779|Loss: 0.8906621336936951:  58%|█████▊    | 3779/6470 [43:29<33:09,  1.35it/s]
3|3780|Loss: 0.7627967596054077:  58%|█████▊    | 3779/6470 [43:30<33:09,  1.35it/s]
3|3780|Loss: 0.7627967596054077:  58%|█████▊    | 3780/6470 [43:30<33:14,  1.35it/s]
3|3781|Loss: 0.874927818775177:  58%|█████▊    | 3780/6470 [43:31<33:14,  1.35it/s] 
3|3781|Loss: 0.874927818775177:  58%|█████▊    | 3781/6470 [43:31<33:00,  1.36it/s]
3|3782|Loss: 0.7600919604301453:  58%|█████▊    | 3781/6470 [43:31<33:00,  1.36it/s]
3|3782|Loss: 0.7600919604301453:  58%|█████▊    | 3782/6470 [43:32<33:50,  1.32it/s]
3|3783|Loss: 0.8714686036109924:  58%|█████▊    | 3782/6470 [43:32<33:50,  1.32it/s]
3|3783|Loss: 0.8714686036109924:  58%|█████▊    | 3783/6470 [43:32<34:54,  1.28it/s]
3|3784|Loss: 1.0537055730819702:  58%|█████▊    | 3783/6470 [43:33<34:54,  1.28it/s]
3|3784|Loss: 1.0537055730819702:  58%|█████▊    | 3784/6470 [43:33<30:40,  1.46it/s]
3|3785|Loss: 0.7843495607376099:  58%|█████▊    | 3784/6470 [43:33<30:40,  1.46it/s]
3|3785|Loss: 0.7843495607376099:  59%|█████▊    | 3785/6470 [43:34<29:03,  1.54it/s]
3|3786|Loss: 0.7960074543952942:  59%|█████▊    | 3785/6470 [43:34<29:03,  1.54it/s]
3|3786|Loss: 0.7960074543952942:  59%|█████▊    | 3786/6470 [43:34<28:42,  1.56it/s]
3|3787|Loss: 0.8340846300125122:  59%|█████▊    | 3786/6470 [43:35<28:42,  1.56it/s]
3|3787|Loss: 0.8340846300125122:  59%|█████▊    | 3787/6470 [43:35<29:55,  1.49it/s]
3|3788|Loss: 0.7856385707855225:  59%|█████▊    | 3787/6470 [43:35<29:55,  1.49it/s]
3|3788|Loss: 0.7856385707855225:  59%|█████▊    | 3788/6470 [43:35<26:51,  1.66it/s]
3|3789|Loss: 0.7467170357704163:  59%|█████▊    | 3788/6470 [43:36<26:51,  1.66it/s]
3|3789|Loss: 0.7467170357704163:  59%|█████▊    | 3789/6470 [43:36<27:47,  1.61it/s]
3|3790|Loss: 0.9281063079833984:  59%|█████▊    | 3789/6470 [43:36<27:47,  1.61it/s]
3|3790|Loss: 0.9281063079833984:  59%|█████▊    | 3790/6470 [43:37<28:09,  1.59it/s]
3|3791|Loss: 0.8544487953186035:  59%|█████▊    | 3790/6470 [43:37<28:09,  1.59it/s]
3|3791|Loss: 0.8544487953186035:  59%|█████▊    | 3791/6470 [43:37<28:10,  1.58it/s]
3|3792|Loss: 0.9426531195640564:  59%|█████▊    | 3791/6470 [43:38<28:10,  1.58it/s]
3|3792|Loss: 0.9426531195640564:  59%|█████▊    | 3792/6470 [43:38<29:25,  1.52it/s]
3|3793|Loss: 0.8601157069206238:  59%|█████▊    | 3792/6470 [43:38<29:25,  1.52it/s]
3|3793|Loss: 0.8601157069206238:  59%|█████▊    | 3793/6470 [43:39<28:07,  1.59it/s]
3|3794|Loss: 0.8734098076820374:  59%|█████▊    | 3793/6470 [43:39<28:07,  1.59it/s]
3|3794|Loss: 0.8734098076820374:  59%|█████▊    | 3794/6470 [43:39<31:59,  1.39it/s]
3|3795|Loss: 0.9759177565574646:  59%|█████▊    | 3794/6470 [43:40<31:59,  1.39it/s]
3|3795|Loss: 0.9759177565574646:  59%|█████▊    | 3795/6470 [43:40<33:46,  1.32it/s]
3|3796|Loss: 0.8828596472740173:  59%|█████▊    | 3795/6470 [43:41<33:46,  1.32it/s]
3|3796|Loss: 0.8828596472740173:  59%|█████▊    | 3796/6470 [43:41<30:57,  1.44it/s]
3|3797|Loss: 0.8788794875144958:  59%|█████▊    | 3796/6470 [43:41<30:57,  1.44it/s]
3|3797|Loss: 0.8788794875144958:  59%|█████▊    | 3797/6470 [43:41<29:43,  1.50it/s]
3|3798|Loss: 0.7935025095939636:  59%|█████▊    | 3797/6470 [43:42<29:43,  1.50it/s]
3|3798|Loss: 0.7935025095939636:  59%|█████▊    | 3798/6470 [43:42<30:01,  1.48it/s]
3|3799|Loss: 0.9551536440849304:  59%|█████▊    | 3798/6470 [43:43<30:01,  1.48it/s]
3|3799|Loss: 0.9551536440849304:  59%|█████▊    | 3799/6470 [43:43<31:46,  1.40it/s]
3|3800|Loss: 0.6134228110313416:  59%|█████▊    | 3799/6470 [43:43<31:46,  1.40it/s]
3|3800|Loss: 0.6134228110313416:  59%|█████▊    | 3800/6470 [43:44<31:16,  1.42it/s]
3|3801|Loss: 0.9672151207923889:  59%|█████▊    | 3800/6470 [43:44<31:16,  1.42it/s]
3|3801|Loss: 0.9672151207923889:  59%|█████▊    | 3801/6470 [43:44<29:01,  1.53it/s]
3|3802|Loss: 0.7615412473678589:  59%|█████▊    | 3801/6470 [43:45<29:01,  1.53it/s]
3|3802|Loss: 0.7615412473678589:  59%|█████▉    | 3802/6470 [43:45<27:43,  1.60it/s]
3|3803|Loss: 0.7673628330230713:  59%|█████▉    | 3802/6470 [43:45<27:43,  1.60it/s]
3|3803|Loss: 0.7673628330230713:  59%|█████▉    | 3803/6470 [43:46<29:53,  1.49it/s]
3|3804|Loss: 0.9424660801887512:  59%|█████▉    | 3803/6470 [43:46<29:53,  1.49it/s]
3|3804|Loss: 0.9424660801887512:  59%|█████▉    | 3804/6470 [43:46<28:16,  1.57it/s]
3|3805|Loss: 0.7510044574737549:  59%|█████▉    | 3804/6470 [43:46<28:16,  1.57it/s]
3|3805|Loss: 0.7510044574737549:  59%|█████▉    | 3805/6470 [43:47<29:51,  1.49it/s]
3|3806|Loss: 0.6947661638259888:  59%|█████▉    | 3805/6470 [43:47<29:51,  1.49it/s]
3|3806|Loss: 0.6947661638259888:  59%|█████▉    | 3806/6470 [43:48<31:24,  1.41it/s]
3|3807|Loss: 0.9716848731040955:  59%|█████▉    | 3806/6470 [43:48<31:24,  1.41it/s]
3|3807|Loss: 0.9716848731040955:  59%|█████▉    | 3807/6470 [43:48<30:57,  1.43it/s]
3|3808|Loss: 0.8398456573486328:  59%|█████▉    | 3807/6470 [43:49<30:57,  1.43it/s]
3|3808|Loss: 0.8398456573486328:  59%|█████▉    | 3808/6470 [43:49<29:36,  1.50it/s]
3|3809|Loss: 0.9563573598861694:  59%|█████▉    | 3808/6470 [43:49<29:36,  1.50it/s]
3|3809|Loss: 0.9563573598861694:  59%|█████▉    | 3809/6470 [43:50<29:04,  1.53it/s]
3|3810|Loss: 0.7408782839775085:  59%|█████▉    | 3809/6470 [43:50<29:04,  1.53it/s]
3|3810|Loss: 0.7408782839775085:  59%|█████▉    | 3810/6470 [43:50<29:33,  1.50it/s]
3|3811|Loss: 0.8791846036911011:  59%|█████▉    | 3810/6470 [43:51<29:33,  1.50it/s]
3|3811|Loss: 0.8791846036911011:  59%|█████▉    | 3811/6470 [43:51<30:19,  1.46it/s]
3|3812|Loss: 0.8639571070671082:  59%|█████▉    | 3811/6470 [43:51<30:19,  1.46it/s]
3|3812|Loss: 0.8639571070671082:  59%|█████▉    | 3812/6470 [43:52<30:05,  1.47it/s]
3|3813|Loss: 0.7803047299385071:  59%|█████▉    | 3812/6470 [43:52<30:05,  1.47it/s]
3|3813|Loss: 0.7803047299385071:  59%|█████▉    | 3813/6470 [43:52<30:28,  1.45it/s]
3|3814|Loss: 0.7399718165397644:  59%|█████▉    | 3813/6470 [43:53<30:28,  1.45it/s]
3|3814|Loss: 0.7399718165397644:  59%|█████▉    | 3814/6470 [43:53<29:51,  1.48it/s]
3|3815|Loss: 0.8866435885429382:  59%|█████▉    | 3814/6470 [43:53<29:51,  1.48it/s]
3|3815|Loss: 0.8866435885429382:  59%|█████▉    | 3815/6470 [43:54<30:29,  1.45it/s]
3|3816|Loss: 0.721181333065033:  59%|█████▉    | 3815/6470 [43:54<30:29,  1.45it/s] 
3|3816|Loss: 0.721181333065033:  59%|█████▉    | 3816/6470 [43:54<29:31,  1.50it/s]
3|3817|Loss: 0.7270869612693787:  59%|█████▉    | 3816/6470 [43:55<29:31,  1.50it/s]
3|3817|Loss: 0.7270869612693787:  59%|█████▉    | 3817/6470 [43:55<29:38,  1.49it/s]
3|3818|Loss: 0.8043335676193237:  59%|█████▉    | 3817/6470 [43:55<29:38,  1.49it/s]
3|3818|Loss: 0.8043335676193237:  59%|█████▉    | 3818/6470 [43:56<31:21,  1.41it/s]
3|3819|Loss: 0.915101945400238:  59%|█████▉    | 3818/6470 [43:56<31:21,  1.41it/s] 
3|3819|Loss: 0.915101945400238:  59%|█████▉    | 3819/6470 [43:56<30:22,  1.45it/s]
3|3820|Loss: 0.8836369514465332:  59%|█████▉    | 3819/6470 [43:57<30:22,  1.45it/s]
3|3820|Loss: 0.8836369514465332:  59%|█████▉    | 3820/6470 [43:57<31:41,  1.39it/s]
3|3821|Loss: 0.8535963296890259:  59%|█████▉    | 3820/6470 [43:58<31:41,  1.39it/s]
3|3821|Loss: 0.8535963296890259:  59%|█████▉    | 3821/6470 [43:58<32:07,  1.37it/s]
3|3822|Loss: 0.7396592497825623:  59%|█████▉    | 3821/6470 [43:58<32:07,  1.37it/s]
3|3822|Loss: 0.7396592497825623:  59%|█████▉    | 3822/6470 [43:59<32:10,  1.37it/s]
3|3823|Loss: 0.7715590596199036:  59%|█████▉    | 3822/6470 [43:59<32:10,  1.37it/s]
3|3823|Loss: 0.7715590596199036:  59%|█████▉    | 3823/6470 [43:59<32:24,  1.36it/s]
3|3824|Loss: 0.6909452676773071:  59%|█████▉    | 3823/6470 [44:00<32:24,  1.36it/s]
3|3824|Loss: 0.6909452676773071:  59%|█████▉    | 3824/6470 [44:00<33:27,  1.32it/s]
3|3825|Loss: 0.8908286690711975:  59%|█████▉    | 3824/6470 [44:01<33:27,  1.32it/s]
3|3825|Loss: 0.8908286690711975:  59%|█████▉    | 3825/6470 [44:01<33:22,  1.32it/s]
3|3826|Loss: 0.8947152495384216:  59%|█████▉    | 3825/6470 [44:01<33:22,  1.32it/s]
3|3826|Loss: 0.8947152495384216:  59%|█████▉    | 3826/6470 [44:02<32:06,  1.37it/s]
3|3827|Loss: 0.9125433564186096:  59%|█████▉    | 3826/6470 [44:02<32:06,  1.37it/s]
3|3827|Loss: 0.9125433564186096:  59%|█████▉    | 3827/6470 [44:02<32:03,  1.37it/s]
3|3828|Loss: 0.7604549527168274:  59%|█████▉    | 3827/6470 [44:03<32:03,  1.37it/s]
3|3828|Loss: 0.7604549527168274:  59%|█████▉    | 3828/6470 [44:03<31:13,  1.41it/s]
3|3829|Loss: 0.6750817894935608:  59%|█████▉    | 3828/6470 [44:03<31:13,  1.41it/s]
3|3829|Loss: 0.6750817894935608:  59%|█████▉    | 3829/6470 [44:04<31:27,  1.40it/s]
3|3830|Loss: 0.8212451934814453:  59%|█████▉    | 3829/6470 [44:04<31:27,  1.40it/s]
3|3830|Loss: 0.8212451934814453:  59%|█████▉    | 3830/6470 [44:04<30:48,  1.43it/s]
3|3831|Loss: 0.7925997972488403:  59%|█████▉    | 3830/6470 [44:05<30:48,  1.43it/s]
3|3831|Loss: 0.7925997972488403:  59%|█████▉    | 3831/6470 [44:05<27:59,  1.57it/s]
3|3832|Loss: 0.8541433215141296:  59%|█████▉    | 3831/6470 [44:05<27:59,  1.57it/s]
3|3832|Loss: 0.8541433215141296:  59%|█████▉    | 3832/6470 [44:06<28:30,  1.54it/s]
3|3833|Loss: 0.9133110642433167:  59%|█████▉    | 3832/6470 [44:06<28:30,  1.54it/s]
3|3833|Loss: 0.9133110642433167:  59%|█████▉    | 3833/6470 [44:06<28:49,  1.52it/s]
3|3834|Loss: 0.9384123086929321:  59%|█████▉    | 3833/6470 [44:07<28:49,  1.52it/s]
3|3834|Loss: 0.9384123086929321:  59%|█████▉    | 3834/6470 [44:07<27:24,  1.60it/s]
3|3835|Loss: 0.8694419860839844:  59%|█████▉    | 3834/6470 [44:07<27:24,  1.60it/s]
3|3835|Loss: 0.8694419860839844:  59%|█████▉    | 3835/6470 [44:07<27:00,  1.63it/s]
3|3836|Loss: 0.9399184584617615:  59%|█████▉    | 3835/6470 [44:08<27:00,  1.63it/s]
3|3836|Loss: 0.9399184584617615:  59%|█████▉    | 3836/6470 [44:08<27:00,  1.63it/s]
3|3837|Loss: 0.8717044591903687:  59%|█████▉    | 3836/6470 [44:08<27:00,  1.63it/s]
3|3837|Loss: 0.8717044591903687:  59%|█████▉    | 3837/6470 [44:09<28:20,  1.55it/s]
3|3838|Loss: 0.8538879752159119:  59%|█████▉    | 3837/6470 [44:09<28:20,  1.55it/s]
3|3838|Loss: 0.8538879752159119:  59%|█████▉    | 3838/6470 [44:09<28:24,  1.54it/s]
3|3839|Loss: 0.8402464389801025:  59%|█████▉    | 3838/6470 [44:10<28:24,  1.54it/s]
3|3839|Loss: 0.8402464389801025:  59%|█████▉    | 3839/6470 [44:10<30:39,  1.43it/s]
3|3840|Loss: 0.8902868628501892:  59%|█████▉    | 3839/6470 [44:11<30:39,  1.43it/s]
3|3840|Loss: 0.8902868628501892:  59%|█████▉    | 3840/6470 [44:11<31:29,  1.39it/s]
3|3841|Loss: 1.0152133703231812:  59%|█████▉    | 3840/6470 [44:11<31:29,  1.39it/s]
3|3841|Loss: 1.0152133703231812:  59%|█████▉    | 3841/6470 [44:12<31:44,  1.38it/s]
3|3842|Loss: 0.785385251045227:  59%|█████▉    | 3841/6470 [44:12<31:44,  1.38it/s] 
3|3842|Loss: 0.785385251045227:  59%|█████▉    | 3842/6470 [44:13<32:36,  1.34it/s]
3|3843|Loss: 0.7956448197364807:  59%|█████▉    | 3842/6470 [44:13<32:36,  1.34it/s]
3|3843|Loss: 0.7956448197364807:  59%|█████▉    | 3843/6470 [44:13<32:43,  1.34it/s]
3|3844|Loss: 0.7970111966133118:  59%|█████▉    | 3843/6470 [44:14<32:43,  1.34it/s]
3|3844|Loss: 0.7970111966133118:  59%|█████▉    | 3844/6470 [44:14<32:34,  1.34it/s]
3|3845|Loss: 0.9217405319213867:  59%|█████▉    | 3844/6470 [44:14<32:34,  1.34it/s]
3|3845|Loss: 0.9217405319213867:  59%|█████▉    | 3845/6470 [44:15<32:37,  1.34it/s]
3|3846|Loss: 0.895911455154419:  59%|█████▉    | 3845/6470 [44:15<32:37,  1.34it/s] 
3|3846|Loss: 0.895911455154419:  59%|█████▉    | 3846/6470 [44:15<30:50,  1.42it/s]
3|3847|Loss: 0.8115543723106384:  59%|█████▉    | 3846/6470 [44:16<30:50,  1.42it/s]
3|3847|Loss: 0.8115543723106384:  59%|█████▉    | 3847/6470 [44:16<29:08,  1.50it/s]
3|3848|Loss: 0.7787855863571167:  59%|█████▉    | 3847/6470 [44:16<29:08,  1.50it/s]
3|3848|Loss: 0.7787855863571167:  59%|█████▉    | 3848/6470 [44:17<30:13,  1.45it/s]
3|3849|Loss: 0.7443335652351379:  59%|█████▉    | 3848/6470 [44:17<30:13,  1.45it/s]
3|3849|Loss: 0.7443335652351379:  59%|█████▉    | 3849/6470 [44:17<28:38,  1.53it/s]
3|3850|Loss: 0.849378764629364:  59%|█████▉    | 3849/6470 [44:18<28:38,  1.53it/s] 
3|3850|Loss: 0.849378764629364:  60%|█████▉    | 3850/6470 [44:18<27:31,  1.59it/s]
3|3851|Loss: 0.8894514441490173:  60%|█████▉    | 3850/6470 [44:18<27:31,  1.59it/s]
3|3851|Loss: 0.8894514441490173:  60%|█████▉    | 3851/6470 [44:19<28:02,  1.56it/s]
3|3852|Loss: 0.6911731362342834:  60%|█████▉    | 3851/6470 [44:19<28:02,  1.56it/s]
3|3852|Loss: 0.6911731362342834:  60%|█████▉    | 3852/6470 [44:19<27:55,  1.56it/s]
3|3853|Loss: 0.8523644804954529:  60%|█████▉    | 3852/6470 [44:20<27:55,  1.56it/s]
3|3853|Loss: 0.8523644804954529:  60%|█████▉    | 3853/6470 [44:20<28:31,  1.53it/s]
3|3854|Loss: 0.8185628056526184:  60%|█████▉    | 3853/6470 [44:20<28:31,  1.53it/s]
3|3854|Loss: 0.8185628056526184:  60%|█████▉    | 3854/6470 [44:20<28:11,  1.55it/s]
3|3855|Loss: 0.9031678438186646:  60%|█████▉    | 3854/6470 [44:21<28:11,  1.55it/s]
3|3855|Loss: 0.9031678438186646:  60%|█████▉    | 3855/6470 [44:21<29:33,  1.47it/s]
3|3856|Loss: 0.9180389046669006:  60%|█████▉    | 3855/6470 [44:22<29:33,  1.47it/s]
3|3856|Loss: 0.9180389046669006:  60%|█████▉    | 3856/6470 [44:22<28:20,  1.54it/s]
3|3857|Loss: 0.6372914910316467:  60%|█████▉    | 3856/6470 [44:22<28:20,  1.54it/s]
3|3857|Loss: 0.6372914910316467:  60%|█████▉    | 3857/6470 [44:23<29:27,  1.48it/s]
3|3858|Loss: 0.849353551864624:  60%|█████▉    | 3857/6470 [44:23<29:27,  1.48it/s] 
3|3858|Loss: 0.849353551864624:  60%|█████▉    | 3858/6470 [44:23<30:21,  1.43it/s]
3|3859|Loss: 0.8503704071044922:  60%|█████▉    | 3858/6470 [44:24<30:21,  1.43it/s]
3|3859|Loss: 0.8503704071044922:  60%|█████▉    | 3859/6470 [44:24<31:54,  1.36it/s]
3|3860|Loss: 1.0376416444778442:  60%|█████▉    | 3859/6470 [44:24<31:54,  1.36it/s]
3|3860|Loss: 1.0376416444778442:  60%|█████▉    | 3860/6470 [44:25<29:59,  1.45it/s]
3|3861|Loss: 0.7889202833175659:  60%|█████▉    | 3860/6470 [44:25<29:59,  1.45it/s]
3|3861|Loss: 0.7889202833175659:  60%|█████▉    | 3861/6470 [44:25<30:09,  1.44it/s]
3|3862|Loss: 0.9075638651847839:  60%|█████▉    | 3861/6470 [44:26<30:09,  1.44it/s]
3|3862|Loss: 0.9075638651847839:  60%|█████▉    | 3862/6470 [44:26<32:13,  1.35it/s]
3|3863|Loss: 0.8690149188041687:  60%|█████▉    | 3862/6470 [44:27<32:13,  1.35it/s]
3|3863|Loss: 0.8690149188041687:  60%|█████▉    | 3863/6470 [44:27<34:01,  1.28it/s]
3|3864|Loss: 0.9003602862358093:  60%|█████▉    | 3863/6470 [44:28<34:01,  1.28it/s]
3|3864|Loss: 0.9003602862358093:  60%|█████▉    | 3864/6470 [44:28<32:48,  1.32it/s]
3|3865|Loss: 0.6346033215522766:  60%|█████▉    | 3864/6470 [44:28<32:48,  1.32it/s]
3|3865|Loss: 0.6346033215522766:  60%|█████▉    | 3865/6470 [44:29<34:33,  1.26it/s]
3|3866|Loss: 0.8938263654708862:  60%|█████▉    | 3865/6470 [44:29<34:33,  1.26it/s]
3|3866|Loss: 0.8938263654708862:  60%|█████▉    | 3866/6470 [44:29<33:06,  1.31it/s]
3|3867|Loss: 0.8597294688224792:  60%|█████▉    | 3866/6470 [44:30<33:06,  1.31it/s]
3|3867|Loss: 0.8597294688224792:  60%|█████▉    | 3867/6470 [44:30<31:24,  1.38it/s]
3|3868|Loss: 0.7901409268379211:  60%|█████▉    | 3867/6470 [44:30<31:24,  1.38it/s]
3|3868|Loss: 0.7901409268379211:  60%|█████▉    | 3868/6470 [44:31<31:14,  1.39it/s]
3|3869|Loss: 0.9637073278427124:  60%|█████▉    | 3868/6470 [44:31<31:14,  1.39it/s]
3|3869|Loss: 0.9637073278427124:  60%|█████▉    | 3869/6470 [44:32<32:02,  1.35it/s]
3|3870|Loss: 0.8272867798805237:  60%|█████▉    | 3869/6470 [44:32<32:02,  1.35it/s]
3|3870|Loss: 0.8272867798805237:  60%|█████▉    | 3870/6470 [44:32<30:57,  1.40it/s]
3|3871|Loss: 0.8217414021492004:  60%|█████▉    | 3870/6470 [44:33<30:57,  1.40it/s]
3|3871|Loss: 0.8217414021492004:  60%|█████▉    | 3871/6470 [44:33<31:38,  1.37it/s]
3|3872|Loss: 0.7994347214698792:  60%|█████▉    | 3871/6470 [44:33<31:38,  1.37it/s]
3|3872|Loss: 0.7994347214698792:  60%|█████▉    | 3872/6470 [44:34<31:38,  1.37it/s]
3|3873|Loss: 0.8212693929672241:  60%|█████▉    | 3872/6470 [44:34<31:38,  1.37it/s]
3|3873|Loss: 0.8212693929672241:  60%|█████▉    | 3873/6470 [44:34<29:56,  1.45it/s]
3|3874|Loss: 0.8363370299339294:  60%|█████▉    | 3873/6470 [44:35<29:56,  1.45it/s]
3|3874|Loss: 0.8363370299339294:  60%|█████▉    | 3874/6470 [44:35<29:08,  1.49it/s]
3|3875|Loss: 0.8407276272773743:  60%|█████▉    | 3874/6470 [44:35<29:08,  1.49it/s]
3|3875|Loss: 0.8407276272773743:  60%|█████▉    | 3875/6470 [44:36<30:56,  1.40it/s]
3|3876|Loss: 0.8824267983436584:  60%|█████▉    | 3875/6470 [44:36<30:56,  1.40it/s]
3|3876|Loss: 0.8824267983436584:  60%|█████▉    | 3876/6470 [44:36<30:18,  1.43it/s]
3|3877|Loss: 0.7519664168357849:  60%|█████▉    | 3876/6470 [44:37<30:18,  1.43it/s]
3|3877|Loss: 0.7519664168357849:  60%|█████▉    | 3877/6470 [44:37<27:29,  1.57it/s]
3|3878|Loss: 0.7924834489822388:  60%|█████▉    | 3877/6470 [44:37<27:29,  1.57it/s]
3|3878|Loss: 0.7924834489822388:  60%|█████▉    | 3878/6470 [44:37<26:29,  1.63it/s]
3|3879|Loss: 0.7185643315315247:  60%|█████▉    | 3878/6470 [44:38<26:29,  1.63it/s]
3|3879|Loss: 0.7185643315315247:  60%|█████▉    | 3879/6470 [44:38<27:20,  1.58it/s]
3|3880|Loss: 0.8145082592964172:  60%|█████▉    | 3879/6470 [44:39<27:20,  1.58it/s]
3|3880|Loss: 0.8145082592964172:  60%|█████▉    | 3880/6470 [44:39<30:25,  1.42it/s]
3|3881|Loss: 0.7770251035690308:  60%|█████▉    | 3880/6470 [44:39<30:25,  1.42it/s]
3|3881|Loss: 0.7770251035690308:  60%|█████▉    | 3881/6470 [44:40<29:43,  1.45it/s]
3|3882|Loss: 0.875296413898468:  60%|█████▉    | 3881/6470 [44:40<29:43,  1.45it/s] 
3|3882|Loss: 0.875296413898468:  60%|██████    | 3882/6470 [44:40<31:11,  1.38it/s]
3|3883|Loss: 0.8270328044891357:  60%|██████    | 3882/6470 [44:41<31:11,  1.38it/s]
3|3883|Loss: 0.8270328044891357:  60%|██████    | 3883/6470 [44:41<31:54,  1.35it/s]
3|3884|Loss: 0.807314395904541:  60%|██████    | 3883/6470 [44:42<31:54,  1.35it/s] 
3|3884|Loss: 0.807314395904541:  60%|██████    | 3884/6470 [44:42<30:18,  1.42it/s]
3|3885|Loss: 0.7651353478431702:  60%|██████    | 3884/6470 [44:42<30:18,  1.42it/s]
3|3885|Loss: 0.7651353478431702:  60%|██████    | 3885/6470 [44:43<32:08,  1.34it/s]
3|3886|Loss: 0.8877370953559875:  60%|██████    | 3885/6470 [44:43<32:08,  1.34it/s]
3|3886|Loss: 0.8877370953559875:  60%|██████    | 3886/6470 [44:44<33:08,  1.30it/s]
3|3887|Loss: 0.8263089060783386:  60%|██████    | 3886/6470 [44:44<33:08,  1.30it/s]
3|3887|Loss: 0.8263089060783386:  60%|██████    | 3887/6470 [44:44<31:58,  1.35it/s]
3|3888|Loss: 0.822494626045227:  60%|██████    | 3887/6470 [44:44<31:58,  1.35it/s] 
3|3888|Loss: 0.822494626045227:  60%|██████    | 3888/6470 [44:45<27:47,  1.55it/s]
3|3889|Loss: 0.7659040093421936:  60%|██████    | 3888/6470 [44:45<27:47,  1.55it/s]
3|3889|Loss: 0.7659040093421936:  60%|██████    | 3889/6470 [44:45<27:47,  1.55it/s]
3|3890|Loss: 0.8502218723297119:  60%|██████    | 3889/6470 [44:46<27:47,  1.55it/s]
3|3890|Loss: 0.8502218723297119:  60%|██████    | 3890/6470 [44:46<26:27,  1.63it/s]
3|3891|Loss: 0.8684170842170715:  60%|██████    | 3890/6470 [44:46<26:27,  1.63it/s]
3|3891|Loss: 0.8684170842170715:  60%|██████    | 3891/6470 [44:47<27:55,  1.54it/s]
3|3892|Loss: 0.9294187426567078:  60%|██████    | 3891/6470 [44:47<27:55,  1.54it/s]
3|3892|Loss: 0.9294187426567078:  60%|██████    | 3892/6470 [44:47<27:21,  1.57it/s]
3|3893|Loss: 0.8820387721061707:  60%|██████    | 3892/6470 [44:48<27:21,  1.57it/s]
3|3893|Loss: 0.8820387721061707:  60%|██████    | 3893/6470 [44:48<28:07,  1.53it/s]
3|3894|Loss: 0.9239517450332642:  60%|██████    | 3893/6470 [44:48<28:07,  1.53it/s]
3|3894|Loss: 0.9239517450332642:  60%|██████    | 3894/6470 [44:49<29:09,  1.47it/s]
3|3895|Loss: 0.781798243522644:  60%|██████    | 3894/6470 [44:49<29:09,  1.47it/s] 
3|3895|Loss: 0.781798243522644:  60%|██████    | 3895/6470 [44:49<30:50,  1.39it/s]
3|3896|Loss: 0.7036309242248535:  60%|██████    | 3895/6470 [44:50<30:50,  1.39it/s]
3|3896|Loss: 0.7036309242248535:  60%|██████    | 3896/6470 [44:50<31:28,  1.36it/s]
3|3897|Loss: 0.8546326160430908:  60%|██████    | 3896/6470 [44:51<31:28,  1.36it/s]
3|3897|Loss: 0.8546326160430908:  60%|██████    | 3897/6470 [44:51<31:13,  1.37it/s]
3|3898|Loss: 0.9646115303039551:  60%|██████    | 3897/6470 [44:51<31:13,  1.37it/s]
3|3898|Loss: 0.9646115303039551:  60%|██████    | 3898/6470 [44:51<29:16,  1.46it/s]
3|3899|Loss: 0.8440274000167847:  60%|██████    | 3898/6470 [44:52<29:16,  1.46it/s]
3|3899|Loss: 0.8440274000167847:  60%|██████    | 3899/6470 [44:52<27:47,  1.54it/s]
3|3900|Loss: 0.8968667984008789:  60%|██████    | 3899/6470 [44:52<27:47,  1.54it/s]
3|3900|Loss: 0.8968667984008789:  60%|██████    | 3900/6470 [44:53<30:01,  1.43it/s]
3|3901|Loss: 0.5768556594848633:  60%|██████    | 3900/6470 [44:53<30:01,  1.43it/s]
3|3901|Loss: 0.5768556594848633:  60%|██████    | 3901/6470 [44:54<30:34,  1.40it/s]
3|3902|Loss: 0.9231860637664795:  60%|██████    | 3901/6470 [44:54<30:34,  1.40it/s]
3|3902|Loss: 0.9231860637664795:  60%|██████    | 3902/6470 [44:54<29:53,  1.43it/s]
3|3903|Loss: 0.8369351029396057:  60%|██████    | 3902/6470 [44:55<29:53,  1.43it/s]
3|3903|Loss: 0.8369351029396057:  60%|██████    | 3903/6470 [44:55<28:26,  1.50it/s]
3|3904|Loss: 0.9151990413665771:  60%|██████    | 3903/6470 [44:55<28:26,  1.50it/s]
3|3904|Loss: 0.9151990413665771:  60%|██████    | 3904/6470 [44:56<30:16,  1.41it/s]
3|3905|Loss: 0.8805093765258789:  60%|██████    | 3904/6470 [44:56<30:16,  1.41it/s]
3|3905|Loss: 0.8805093765258789:  60%|██████    | 3905/6470 [44:56<29:52,  1.43it/s]
3|3906|Loss: 0.8305903673171997:  60%|██████    | 3905/6470 [44:57<29:52,  1.43it/s]
3|3906|Loss: 0.8305903673171997:  60%|██████    | 3906/6470 [44:57<31:14,  1.37it/s]
3|3907|Loss: 0.8881432414054871:  60%|██████    | 3906/6470 [44:58<31:14,  1.37it/s]
3|3907|Loss: 0.8881432414054871:  60%|██████    | 3907/6470 [44:58<33:36,  1.27it/s]
3|3908|Loss: 0.8098292350769043:  60%|██████    | 3907/6470 [44:59<33:36,  1.27it/s]
3|3908|Loss: 0.8098292350769043:  60%|██████    | 3908/6470 [44:59<34:44,  1.23it/s]
3|3909|Loss: 0.6609002351760864:  60%|██████    | 3908/6470 [44:59<34:44,  1.23it/s]
3|3909|Loss: 0.6609002351760864:  60%|██████    | 3909/6470 [45:00<34:11,  1.25it/s]
3|3910|Loss: 0.8245132565498352:  60%|██████    | 3909/6470 [45:00<34:11,  1.25it/s]
3|3910|Loss: 0.8245132565498352:  60%|██████    | 3910/6470 [45:00<31:45,  1.34it/s]
3|3911|Loss: 0.5714316964149475:  60%|██████    | 3910/6470 [45:01<31:45,  1.34it/s]
3|3911|Loss: 0.5714316964149475:  60%|██████    | 3911/6470 [45:01<34:33,  1.23it/s]
3|3912|Loss: 0.9015088677406311:  60%|██████    | 3911/6470 [45:02<34:33,  1.23it/s]
3|3912|Loss: 0.9015088677406311:  60%|██████    | 3912/6470 [45:02<34:36,  1.23it/s]
3|3913|Loss: 0.9034945964813232:  60%|██████    | 3912/6470 [45:02<34:36,  1.23it/s]
3|3913|Loss: 0.9034945964813232:  60%|██████    | 3913/6470 [45:03<32:31,  1.31it/s]
3|3914|Loss: 0.8660874962806702:  60%|██████    | 3913/6470 [45:03<32:31,  1.31it/s]
3|3914|Loss: 0.8660874962806702:  60%|██████    | 3914/6470 [45:03<30:33,  1.39it/s]
3|3915|Loss: 0.8898646831512451:  60%|██████    | 3914/6470 [45:04<30:33,  1.39it/s]
3|3915|Loss: 0.8898646831512451:  61%|██████    | 3915/6470 [45:04<32:16,  1.32it/s]
3|3916|Loss: 0.7991681098937988:  61%|██████    | 3915/6470 [45:05<32:16,  1.32it/s]
3|3916|Loss: 0.7991681098937988:  61%|██████    | 3916/6470 [45:05<30:07,  1.41it/s]
3|3917|Loss: 1.0536309480667114:  61%|██████    | 3916/6470 [45:05<30:07,  1.41it/s]
3|3917|Loss: 1.0536309480667114:  61%|██████    | 3917/6470 [45:05<27:31,  1.55it/s]
3|3918|Loss: 0.9484959840774536:  61%|██████    | 3917/6470 [45:06<27:31,  1.55it/s]
3|3918|Loss: 0.9484959840774536:  61%|██████    | 3918/6470 [45:06<29:02,  1.46it/s]
3|3919|Loss: 0.8520460724830627:  61%|██████    | 3918/6470 [45:06<29:02,  1.46it/s]
3|3919|Loss: 0.8520460724830627:  61%|██████    | 3919/6470 [45:07<29:38,  1.43it/s]
3|3920|Loss: 0.8976172208786011:  61%|██████    | 3919/6470 [45:07<29:38,  1.43it/s]
3|3920|Loss: 0.8976172208786011:  61%|██████    | 3920/6470 [45:08<30:01,  1.42it/s]
3|3921|Loss: 0.828303337097168:  61%|██████    | 3920/6470 [45:08<30:01,  1.42it/s] 
3|3921|Loss: 0.828303337097168:  61%|██████    | 3921/6470 [45:08<30:18,  1.40it/s]
3|3922|Loss: 0.8012319207191467:  61%|██████    | 3921/6470 [45:09<30:18,  1.40it/s]
3|3922|Loss: 0.8012319207191467:  61%|██████    | 3922/6470 [45:09<29:17,  1.45it/s]
3|3923|Loss: 0.7929660081863403:  61%|██████    | 3922/6470 [45:09<29:17,  1.45it/s]
3|3923|Loss: 0.7929660081863403:  61%|██████    | 3923/6470 [45:10<28:34,  1.49it/s]
3|3924|Loss: 0.8839152455329895:  61%|██████    | 3923/6470 [45:10<28:34,  1.49it/s]
3|3924|Loss: 0.8839152455329895:  61%|██████    | 3924/6470 [45:10<30:39,  1.38it/s]
3|3925|Loss: 0.8520697951316833:  61%|██████    | 3924/6470 [45:11<30:39,  1.38it/s]
3|3925|Loss: 0.8520697951316833:  61%|██████    | 3925/6470 [45:11<31:13,  1.36it/s]
3|3926|Loss: 0.8721882104873657:  61%|██████    | 3925/6470 [45:12<31:13,  1.36it/s]
3|3926|Loss: 0.8721882104873657:  61%|██████    | 3926/6470 [45:12<30:12,  1.40it/s]
3|3927|Loss: 0.895137369632721:  61%|██████    | 3926/6470 [45:12<30:12,  1.40it/s] 
3|3927|Loss: 0.895137369632721:  61%|██████    | 3927/6470 [45:12<29:04,  1.46it/s]
3|3928|Loss: 0.9901048541069031:  61%|██████    | 3927/6470 [45:13<29:04,  1.46it/s]
3|3928|Loss: 0.9901048541069031:  61%|██████    | 3928/6470 [45:13<27:08,  1.56it/s]
3|3929|Loss: 0.8780860304832458:  61%|██████    | 3928/6470 [45:13<27:08,  1.56it/s]
3|3929|Loss: 0.8780860304832458:  61%|██████    | 3929/6470 [45:14<27:23,  1.55it/s]
3|3930|Loss: 0.6975154876708984:  61%|██████    | 3929/6470 [45:14<27:23,  1.55it/s]
3|3930|Loss: 0.6975154876708984:  61%|██████    | 3930/6470 [45:14<27:13,  1.55it/s]
3|3931|Loss: 0.9854500889778137:  61%|██████    | 3930/6470 [45:15<27:13,  1.55it/s]
3|3931|Loss: 0.9854500889778137:  61%|██████    | 3931/6470 [45:15<28:02,  1.51it/s]
3|3932|Loss: 0.8646253943443298:  61%|██████    | 3931/6470 [45:15<28:02,  1.51it/s]
3|3932|Loss: 0.8646253943443298:  61%|██████    | 3932/6470 [45:16<27:50,  1.52it/s]
3|3933|Loss: 0.8281177282333374:  61%|██████    | 3932/6470 [45:16<27:50,  1.52it/s]
3|3933|Loss: 0.8281177282333374:  61%|██████    | 3933/6470 [45:16<26:54,  1.57it/s]
3|3934|Loss: 0.8730171918869019:  61%|██████    | 3933/6470 [45:17<26:54,  1.57it/s]
3|3934|Loss: 0.8730171918869019:  61%|██████    | 3934/6470 [45:17<27:49,  1.52it/s]
3|3935|Loss: 0.6599968671798706:  61%|██████    | 3934/6470 [45:17<27:49,  1.52it/s]
3|3935|Loss: 0.6599968671798706:  61%|██████    | 3935/6470 [45:18<27:13,  1.55it/s]
3|3936|Loss: 0.9733440279960632:  61%|██████    | 3935/6470 [45:18<27:13,  1.55it/s]
3|3936|Loss: 0.9733440279960632:  61%|██████    | 3936/6470 [45:18<29:16,  1.44it/s]
3|3937|Loss: 0.9089627861976624:  61%|██████    | 3936/6470 [45:19<29:16,  1.44it/s]
3|3937|Loss: 0.9089627861976624:  61%|██████    | 3937/6470 [45:19<28:44,  1.47it/s]
3|3938|Loss: 0.852835476398468:  61%|██████    | 3937/6470 [45:19<28:44,  1.47it/s] 
3|3938|Loss: 0.852835476398468:  61%|██████    | 3938/6470 [45:20<29:14,  1.44it/s]
3|3939|Loss: 0.9128152132034302:  61%|██████    | 3938/6470 [45:20<29:14,  1.44it/s]
3|3939|Loss: 0.9128152132034302:  61%|██████    | 3939/6470 [45:20<30:46,  1.37it/s]
3|3940|Loss: 0.993694543838501:  61%|██████    | 3939/6470 [45:21<30:46,  1.37it/s] 
3|3940|Loss: 0.993694543838501:  61%|██████    | 3940/6470 [45:21<31:14,  1.35it/s]
3|3941|Loss: 0.9922366738319397:  61%|██████    | 3940/6470 [45:22<31:14,  1.35it/s]
3|3941|Loss: 0.9922366738319397:  61%|██████    | 3941/6470 [45:22<28:20,  1.49it/s]
3|3942|Loss: 0.9600377082824707:  61%|██████    | 3941/6470 [45:22<28:20,  1.49it/s]
3|3942|Loss: 0.9600377082824707:  61%|██████    | 3942/6470 [45:22<28:19,  1.49it/s]
3|3943|Loss: 0.7529928684234619:  61%|██████    | 3942/6470 [45:23<28:19,  1.49it/s]
3|3943|Loss: 0.7529928684234619:  61%|██████    | 3943/6470 [45:23<28:14,  1.49it/s]
3|3944|Loss: 0.9442320466041565:  61%|██████    | 3943/6470 [45:23<28:14,  1.49it/s]
3|3944|Loss: 0.9442320466041565:  61%|██████    | 3944/6470 [45:24<27:12,  1.55it/s]
3|3945|Loss: 1.0311983823776245:  61%|██████    | 3944/6470 [45:24<27:12,  1.55it/s]
3|3945|Loss: 1.0311983823776245:  61%|██████    | 3945/6470 [45:25<29:10,  1.44it/s]
3|3946|Loss: 0.8948519229888916:  61%|██████    | 3945/6470 [45:25<29:10,  1.44it/s]
3|3946|Loss: 0.8948519229888916:  61%|██████    | 3946/6470 [45:25<28:54,  1.45it/s]
3|3947|Loss: 0.8626811504364014:  61%|██████    | 3946/6470 [45:25<28:54,  1.45it/s]
3|3947|Loss: 0.8626811504364014:  61%|██████    | 3947/6470 [45:26<26:21,  1.59it/s]
3|3948|Loss: 0.793914258480072:  61%|██████    | 3947/6470 [45:26<26:21,  1.59it/s] 
3|3948|Loss: 0.793914258480072:  61%|██████    | 3948/6470 [45:26<26:34,  1.58it/s]
3|3949|Loss: 0.8912487030029297:  61%|██████    | 3948/6470 [45:27<26:34,  1.58it/s]
3|3949|Loss: 0.8912487030029297:  61%|██████    | 3949/6470 [45:27<26:54,  1.56it/s]
3|3950|Loss: 0.7684974670410156:  61%|██████    | 3949/6470 [45:27<26:54,  1.56it/s]
3|3950|Loss: 0.7684974670410156:  61%|██████    | 3950/6470 [45:28<26:53,  1.56it/s]
3|3951|Loss: 0.7214065790176392:  61%|██████    | 3950/6470 [45:28<26:53,  1.56it/s]
3|3951|Loss: 0.7214065790176392:  61%|██████    | 3951/6470 [45:28<28:21,  1.48it/s]
3|3952|Loss: 0.8164559006690979:  61%|██████    | 3951/6470 [45:29<28:21,  1.48it/s]
3|3952|Loss: 0.8164559006690979:  61%|██████    | 3952/6470 [45:29<27:48,  1.51it/s]
3|3953|Loss: 0.8398510217666626:  61%|██████    | 3952/6470 [45:29<27:48,  1.51it/s]
3|3953|Loss: 0.8398510217666626:  61%|██████    | 3953/6470 [45:30<28:17,  1.48it/s]
3|3954|Loss: 0.8423876166343689:  61%|██████    | 3953/6470 [45:30<28:17,  1.48it/s]
3|3954|Loss: 0.8423876166343689:  61%|██████    | 3954/6470 [45:30<29:01,  1.45it/s]
3|3955|Loss: 0.9316868185997009:  61%|██████    | 3954/6470 [45:31<29:01,  1.45it/s]
3|3955|Loss: 0.9316868185997009:  61%|██████    | 3955/6470 [45:31<29:10,  1.44it/s]
3|3956|Loss: 0.7359088659286499:  61%|██████    | 3955/6470 [45:32<29:10,  1.44it/s]
3|3956|Loss: 0.7359088659286499:  61%|██████    | 3956/6470 [45:32<28:20,  1.48it/s]
3|3957|Loss: 0.9169043898582458:  61%|██████    | 3956/6470 [45:32<28:20,  1.48it/s]
3|3957|Loss: 0.9169043898582458:  61%|██████    | 3957/6470 [45:33<30:32,  1.37it/s]
3|3958|Loss: 1.068379282951355:  61%|██████    | 3957/6470 [45:33<30:32,  1.37it/s] 
3|3958|Loss: 1.068379282951355:  61%|██████    | 3958/6470 [45:33<28:22,  1.48it/s]
3|3959|Loss: 0.7730280756950378:  61%|██████    | 3958/6470 [45:34<28:22,  1.48it/s]
3|3959|Loss: 0.7730280756950378:  61%|██████    | 3959/6470 [45:34<28:23,  1.47it/s]
3|3960|Loss: 0.8917115926742554:  61%|██████    | 3959/6470 [45:34<28:23,  1.47it/s]
3|3960|Loss: 0.8917115926742554:  61%|██████    | 3960/6470 [45:35<30:22,  1.38it/s]
3|3961|Loss: 0.8574398159980774:  61%|██████    | 3960/6470 [45:35<30:22,  1.38it/s]
3|3961|Loss: 0.8574398159980774:  61%|██████    | 3961/6470 [45:35<28:59,  1.44it/s]
3|3962|Loss: 0.8555089831352234:  61%|██████    | 3961/6470 [45:36<28:59,  1.44it/s]
3|3962|Loss: 0.8555089831352234:  61%|██████    | 3962/6470 [45:36<28:40,  1.46it/s]
3|3963|Loss: 0.8231344223022461:  61%|██████    | 3962/6470 [45:36<28:40,  1.46it/s]
3|3963|Loss: 0.8231344223022461:  61%|██████▏   | 3963/6470 [45:37<30:30,  1.37it/s]
3|3964|Loss: 0.7192106246948242:  61%|██████▏   | 3963/6470 [45:37<30:30,  1.37it/s]
3|3964|Loss: 0.7192106246948242:  61%|██████▏   | 3964/6470 [45:38<31:54,  1.31it/s]
3|3965|Loss: 0.7908578515052795:  61%|██████▏   | 3964/6470 [45:38<31:54,  1.31it/s]
3|3965|Loss: 0.7908578515052795:  61%|██████▏   | 3965/6470 [45:38<30:38,  1.36it/s]
3|3966|Loss: 0.8245260119438171:  61%|██████▏   | 3965/6470 [45:39<30:38,  1.36it/s]
3|3966|Loss: 0.8245260119438171:  61%|██████▏   | 3966/6470 [45:39<30:18,  1.38it/s]
3|3967|Loss: 0.8618977069854736:  61%|██████▏   | 3966/6470 [45:39<30:18,  1.38it/s]
3|3967|Loss: 0.8618977069854736:  61%|██████▏   | 3967/6470 [45:40<30:04,  1.39it/s]
3|3968|Loss: 0.913240373134613:  61%|██████▏   | 3967/6470 [45:40<30:04,  1.39it/s] 
3|3968|Loss: 0.913240373134613:  61%|██████▏   | 3968/6470 [45:41<30:43,  1.36it/s]
3|3969|Loss: 0.9163494110107422:  61%|██████▏   | 3968/6470 [45:41<30:43,  1.36it/s]
3|3969|Loss: 0.9163494110107422:  61%|██████▏   | 3969/6470 [45:41<31:38,  1.32it/s]
3|3970|Loss: 0.6292738318443298:  61%|██████▏   | 3969/6470 [45:42<31:38,  1.32it/s]
3|3970|Loss: 0.6292738318443298:  61%|██████▏   | 3970/6470 [45:42<29:01,  1.44it/s]
3|3971|Loss: 0.8963356018066406:  61%|██████▏   | 3970/6470 [45:42<29:01,  1.44it/s]
3|3971|Loss: 0.8963356018066406:  61%|██████▏   | 3971/6470 [45:42<27:09,  1.53it/s]
3|3972|Loss: 0.7799844741821289:  61%|██████▏   | 3971/6470 [45:43<27:09,  1.53it/s]
3|3972|Loss: 0.7799844741821289:  61%|██████▏   | 3972/6470 [45:43<29:36,  1.41it/s]
3|3973|Loss: 0.7395759224891663:  61%|██████▏   | 3972/6470 [45:44<29:36,  1.41it/s]
3|3973|Loss: 0.7395759224891663:  61%|██████▏   | 3973/6470 [45:44<31:01,  1.34it/s]
3|3974|Loss: 0.7304175496101379:  61%|██████▏   | 3973/6470 [45:45<31:01,  1.34it/s]
3|3974|Loss: 0.7304175496101379:  61%|██████▏   | 3974/6470 [45:45<30:11,  1.38it/s]
3|3975|Loss: 0.9279473423957825:  61%|██████▏   | 3974/6470 [45:45<30:11,  1.38it/s]
3|3975|Loss: 0.9279473423957825:  61%|██████▏   | 3975/6470 [45:46<30:12,  1.38it/s]
3|3976|Loss: 0.7792805433273315:  61%|██████▏   | 3975/6470 [45:46<30:12,  1.38it/s]
3|3976|Loss: 0.7792805433273315:  61%|██████▏   | 3976/6470 [45:46<29:11,  1.42it/s]
3|3977|Loss: 0.8170952200889587:  61%|██████▏   | 3976/6470 [45:47<29:11,  1.42it/s]
3|3977|Loss: 0.8170952200889587:  61%|██████▏   | 3977/6470 [45:47<28:09,  1.48it/s]
3|3978|Loss: 0.8676621913909912:  61%|██████▏   | 3977/6470 [45:47<28:09,  1.48it/s]
3|3978|Loss: 0.8676621913909912:  61%|██████▏   | 3978/6470 [45:47<26:32,  1.56it/s]
3|3979|Loss: 0.819838285446167:  61%|██████▏   | 3978/6470 [45:48<26:32,  1.56it/s] 
3|3979|Loss: 0.819838285446167:  61%|██████▏   | 3979/6470 [45:48<26:48,  1.55it/s]
3|3980|Loss: 0.7382317781448364:  61%|██████▏   | 3979/6470 [45:48<26:48,  1.55it/s]
3|3980|Loss: 0.7382317781448364:  62%|██████▏   | 3980/6470 [45:49<27:32,  1.51it/s]
3|3981|Loss: 0.7690984606742859:  62%|██████▏   | 3980/6470 [45:49<27:32,  1.51it/s]
3|3981|Loss: 0.7690984606742859:  62%|██████▏   | 3981/6470 [45:49<27:27,  1.51it/s]
3|3982|Loss: 1.059994101524353:  62%|██████▏   | 3981/6470 [45:50<27:27,  1.51it/s] 
3|3982|Loss: 1.059994101524353:  62%|██████▏   | 3982/6470 [45:50<27:08,  1.53it/s]
3|3983|Loss: 0.8074200749397278:  62%|██████▏   | 3982/6470 [45:50<27:08,  1.53it/s]
3|3983|Loss: 0.8074200749397278:  62%|██████▏   | 3983/6470 [45:51<25:48,  1.61it/s]
3|3984|Loss: 0.8757200837135315:  62%|██████▏   | 3983/6470 [45:51<25:48,  1.61it/s]
3|3984|Loss: 0.8757200837135315:  62%|██████▏   | 3984/6470 [45:51<26:46,  1.55it/s]
3|3985|Loss: 0.8938469290733337:  62%|██████▏   | 3984/6470 [45:52<26:46,  1.55it/s]
3|3985|Loss: 0.8938469290733337:  62%|██████▏   | 3985/6470 [45:52<28:06,  1.47it/s]
3|3986|Loss: 0.834382951259613:  62%|██████▏   | 3985/6470 [45:52<28:06,  1.47it/s] 
3|3986|Loss: 0.834382951259613:  62%|██████▏   | 3986/6470 [45:53<29:43,  1.39it/s]
3|3987|Loss: 0.8781518340110779:  62%|██████▏   | 3986/6470 [45:53<29:43,  1.39it/s]
3|3987|Loss: 0.8781518340110779:  62%|██████▏   | 3987/6470 [45:54<30:09,  1.37it/s]
3|3988|Loss: 0.6940165758132935:  62%|██████▏   | 3987/6470 [45:54<30:09,  1.37it/s]
3|3988|Loss: 0.6940165758132935:  62%|██████▏   | 3988/6470 [45:54<28:39,  1.44it/s]
3|3989|Loss: 0.8595574498176575:  62%|██████▏   | 3988/6470 [45:55<28:39,  1.44it/s]
3|3989|Loss: 0.8595574498176575:  62%|██████▏   | 3989/6470 [45:55<29:01,  1.42it/s]
3|3990|Loss: 0.9327504634857178:  62%|██████▏   | 3989/6470 [45:55<29:01,  1.42it/s]
3|3990|Loss: 0.9327504634857178:  62%|██████▏   | 3990/6470 [45:56<29:14,  1.41it/s]
3|3991|Loss: 0.8654527068138123:  62%|██████▏   | 3990/6470 [45:56<29:14,  1.41it/s]
3|3991|Loss: 0.8654527068138123:  62%|██████▏   | 3991/6470 [45:56<29:30,  1.40it/s]
3|3992|Loss: 0.8355526328086853:  62%|██████▏   | 3991/6470 [45:57<29:30,  1.40it/s]
3|3992|Loss: 0.8355526328086853:  62%|██████▏   | 3992/6470 [45:57<30:57,  1.33it/s]
3|3993|Loss: 0.9576338529586792:  62%|██████▏   | 3992/6470 [45:58<30:57,  1.33it/s]
3|3993|Loss: 0.9576338529586792:  62%|██████▏   | 3993/6470 [45:58<30:48,  1.34it/s]
3|3994|Loss: 0.9168179631233215:  62%|██████▏   | 3993/6470 [45:58<30:48,  1.34it/s]
3|3994|Loss: 0.9168179631233215:  62%|██████▏   | 3994/6470 [45:59<30:19,  1.36it/s]
3|3995|Loss: 0.8643700480461121:  62%|██████▏   | 3994/6470 [45:59<30:19,  1.36it/s]
3|3995|Loss: 0.8643700480461121:  62%|██████▏   | 3995/6470 [45:59<30:15,  1.36it/s]
3|3996|Loss: 0.9186826348304749:  62%|██████▏   | 3995/6470 [46:00<30:15,  1.36it/s]
3|3996|Loss: 0.9186826348304749:  62%|██████▏   | 3996/6470 [46:00<29:20,  1.41it/s]
3|3997|Loss: 0.8001300096511841:  62%|██████▏   | 3996/6470 [46:00<29:20,  1.41it/s]
3|3997|Loss: 0.8001300096511841:  62%|██████▏   | 3997/6470 [46:00<26:34,  1.55it/s]
3|3998|Loss: 0.9928523302078247:  62%|██████▏   | 3997/6470 [46:01<26:34,  1.55it/s]
3|3998|Loss: 0.9928523302078247:  62%|██████▏   | 3998/6470 [46:01<25:18,  1.63it/s]
3|3999|Loss: 0.7014315724372864:  62%|██████▏   | 3998/6470 [46:01<25:18,  1.63it/s]
3|3999|Loss: 0.7014315724372864:  62%|██████▏   | 3999/6470 [46:02<25:26,  1.62it/s]
3|4000|Loss: 0.9732639193534851:  62%|██████▏   | 3999/6470 [46:02<25:26,  1.62it/s]
3|4000|Loss: 0.9732639193534851:  62%|██████▏   | 4000/6470 [46:02<25:27,  1.62it/s]
3|4001|Loss: 0.8834244608879089:  62%|██████▏   | 4000/6470 [46:03<25:27,  1.62it/s]
3|4001|Loss: 0.8834244608879089:  62%|██████▏   | 4001/6470 [46:03<26:39,  1.54it/s]
3|4002|Loss: 0.9613620638847351:  62%|██████▏   | 4001/6470 [46:03<26:39,  1.54it/s]
3|4002|Loss: 0.9613620638847351:  62%|██████▏   | 4002/6470 [46:04<27:56,  1.47it/s]
3|4003|Loss: 0.8262553811073303:  62%|██████▏   | 4002/6470 [46:04<27:56,  1.47it/s]
3|4003|Loss: 0.8262553811073303:  62%|██████▏   | 4003/6470 [46:05<29:56,  1.37it/s]
3|4004|Loss: 1.0673545598983765:  62%|██████▏   | 4003/6470 [46:05<29:56,  1.37it/s]
3|4004|Loss: 1.0673545598983765:  62%|██████▏   | 4004/6470 [46:05<28:49,  1.43it/s]
3|4005|Loss: 0.895205020904541:  62%|██████▏   | 4004/6470 [46:06<28:49,  1.43it/s] 
3|4005|Loss: 0.895205020904541:  62%|██████▏   | 4005/6470 [46:06<27:37,  1.49it/s]
3|4006|Loss: 0.9140456914901733:  62%|██████▏   | 4005/6470 [46:06<27:37,  1.49it/s]
3|4006|Loss: 0.9140456914901733:  62%|██████▏   | 4006/6470 [46:06<26:53,  1.53it/s]
3|4007|Loss: 0.9120776057243347:  62%|██████▏   | 4006/6470 [46:07<26:53,  1.53it/s]
3|4007|Loss: 0.9120776057243347:  62%|██████▏   | 4007/6470 [46:07<25:45,  1.59it/s]
3|4008|Loss: 0.7709037065505981:  62%|██████▏   | 4007/6470 [46:07<25:45,  1.59it/s]
3|4008|Loss: 0.7709037065505981:  62%|██████▏   | 4008/6470 [46:07<23:00,  1.78it/s]
3|4009|Loss: 0.8812031149864197:  62%|██████▏   | 4008/6470 [46:08<23:00,  1.78it/s]
3|4009|Loss: 0.8812031149864197:  62%|██████▏   | 4009/6470 [46:08<22:18,  1.84it/s]
3|4010|Loss: 0.8883625268936157:  62%|██████▏   | 4009/6470 [46:08<22:18,  1.84it/s]
3|4010|Loss: 0.8883625268936157:  62%|██████▏   | 4010/6470 [46:09<25:51,  1.59it/s]
3|4011|Loss: 0.8962286710739136:  62%|██████▏   | 4010/6470 [46:09<25:51,  1.59it/s]
3|4011|Loss: 0.8962286710739136:  62%|██████▏   | 4011/6470 [46:10<27:38,  1.48it/s]
3|4012|Loss: 0.9357563257217407:  62%|██████▏   | 4011/6470 [46:10<27:38,  1.48it/s]
3|4012|Loss: 0.9357563257217407:  62%|██████▏   | 4012/6470 [46:10<27:22,  1.50it/s]
3|4013|Loss: 0.8641382455825806:  62%|██████▏   | 4012/6470 [46:11<27:22,  1.50it/s]
3|4013|Loss: 0.8641382455825806:  62%|██████▏   | 4013/6470 [46:11<28:57,  1.41it/s]
3|4014|Loss: 0.8287103176116943:  62%|██████▏   | 4013/6470 [46:11<28:57,  1.41it/s]
3|4014|Loss: 0.8287103176116943:  62%|██████▏   | 4014/6470 [46:12<30:53,  1.33it/s]
3|4015|Loss: 0.7748616337776184:  62%|██████▏   | 4014/6470 [46:12<30:53,  1.33it/s]
3|4015|Loss: 0.7748616337776184:  62%|██████▏   | 4015/6470 [46:13<30:43,  1.33it/s]
3|4016|Loss: 1.0519126653671265:  62%|██████▏   | 4015/6470 [46:13<30:43,  1.33it/s]
3|4016|Loss: 1.0519126653671265:  62%|██████▏   | 4016/6470 [46:13<30:06,  1.36it/s]
3|4017|Loss: 0.8645257949829102:  62%|██████▏   | 4016/6470 [46:14<30:06,  1.36it/s]
3|4017|Loss: 0.8645257949829102:  62%|██████▏   | 4017/6470 [46:14<29:04,  1.41it/s]
3|4018|Loss: 0.5736554861068726:  62%|██████▏   | 4017/6470 [46:15<29:04,  1.41it/s]
3|4018|Loss: 0.5736554861068726:  62%|██████▏   | 4018/6470 [46:16<40:55,  1.00s/it]
3|4019|Loss: 0.7416535019874573:  62%|██████▏   | 4018/6470 [46:16<40:55,  1.00s/it]
3|4019|Loss: 0.7416535019874573:  62%|██████▏   | 4019/6470 [46:16<38:14,  1.07it/s]
3|4020|Loss: 0.8479819893836975:  62%|██████▏   | 4019/6470 [46:17<38:14,  1.07it/s]
3|4020|Loss: 0.8479819893836975:  62%|██████▏   | 4020/6470 [46:17<33:58,  1.20it/s]
3|4021|Loss: 0.7765097618103027:  62%|██████▏   | 4020/6470 [46:17<33:58,  1.20it/s]
3|4021|Loss: 0.7765097618103027:  62%|██████▏   | 4021/6470 [46:18<34:13,  1.19it/s]
3|4022|Loss: 0.8609510660171509:  62%|██████▏   | 4021/6470 [46:18<34:13,  1.19it/s]
3|4022|Loss: 0.8609510660171509:  62%|██████▏   | 4022/6470 [46:19<34:03,  1.20it/s]
3|4023|Loss: 0.9847782850265503:  62%|██████▏   | 4022/6470 [46:19<34:03,  1.20it/s]
3|4023|Loss: 0.9847782850265503:  62%|██████▏   | 4023/6470 [46:20<34:21,  1.19it/s]
3|4024|Loss: 0.807974100112915:  62%|██████▏   | 4023/6470 [46:20<34:21,  1.19it/s] 
3|4024|Loss: 0.807974100112915:  62%|██████▏   | 4024/6470 [46:20<32:31,  1.25it/s]
3|4025|Loss: 0.8784807324409485:  62%|██████▏   | 4024/6470 [46:21<32:31,  1.25it/s]
3|4025|Loss: 0.8784807324409485:  62%|██████▏   | 4025/6470 [46:21<32:23,  1.26it/s]
3|4026|Loss: 0.7909128069877625:  62%|██████▏   | 4025/6470 [46:21<32:23,  1.26it/s]
3|4026|Loss: 0.7909128069877625:  62%|██████▏   | 4026/6470 [46:22<30:42,  1.33it/s]
3|4027|Loss: 0.8152803778648376:  62%|██████▏   | 4026/6470 [46:22<30:42,  1.33it/s]
3|4027|Loss: 0.8152803778648376:  62%|██████▏   | 4027/6470 [46:22<27:23,  1.49it/s]
3|4028|Loss: 0.9072082042694092:  62%|██████▏   | 4027/6470 [46:23<27:23,  1.49it/s]
3|4028|Loss: 0.9072082042694092:  62%|██████▏   | 4028/6470 [46:23<28:37,  1.42it/s]
3|4029|Loss: 1.0513601303100586:  62%|██████▏   | 4028/6470 [46:23<28:37,  1.42it/s]
3|4029|Loss: 1.0513601303100586:  62%|██████▏   | 4029/6470 [46:24<28:32,  1.43it/s]
3|4030|Loss: 0.7364749908447266:  62%|██████▏   | 4029/6470 [46:24<28:32,  1.43it/s]
3|4030|Loss: 0.7364749908447266:  62%|██████▏   | 4030/6470 [46:24<27:32,  1.48it/s]
3|4031|Loss: 0.7543375492095947:  62%|██████▏   | 4030/6470 [46:25<27:32,  1.48it/s]
3|4031|Loss: 0.7543375492095947:  62%|██████▏   | 4031/6470 [46:25<27:39,  1.47it/s]
3|4032|Loss: 0.8083102107048035:  62%|██████▏   | 4031/6470 [46:25<27:39,  1.47it/s]
3|4032|Loss: 0.8083102107048035:  62%|██████▏   | 4032/6470 [46:26<28:24,  1.43it/s]
3|4033|Loss: 0.8416156768798828:  62%|██████▏   | 4032/6470 [46:26<28:24,  1.43it/s]
3|4033|Loss: 0.8416156768798828:  62%|██████▏   | 4033/6470 [46:26<28:03,  1.45it/s]
3|4034|Loss: 0.6718420386314392:  62%|██████▏   | 4033/6470 [46:27<28:03,  1.45it/s]
3|4034|Loss: 0.6718420386314392:  62%|██████▏   | 4034/6470 [46:27<25:33,  1.59it/s]
3|4035|Loss: 0.7990053296089172:  62%|██████▏   | 4034/6470 [46:27<25:33,  1.59it/s]
3|4035|Loss: 0.7990053296089172:  62%|██████▏   | 4035/6470 [46:27<24:26,  1.66it/s]
3|4036|Loss: 0.910067081451416:  62%|██████▏   | 4035/6470 [46:28<24:26,  1.66it/s] 
3|4036|Loss: 0.910067081451416:  62%|██████▏   | 4036/6470 [46:28<25:19,  1.60it/s]
3|4037|Loss: 0.8504781723022461:  62%|██████▏   | 4036/6470 [46:28<25:19,  1.60it/s]
3|4037|Loss: 0.8504781723022461:  62%|██████▏   | 4037/6470 [46:29<24:45,  1.64it/s]
3|4038|Loss: 0.9792052507400513:  62%|██████▏   | 4037/6470 [46:29<24:45,  1.64it/s]
3|4038|Loss: 0.9792052507400513:  62%|██████▏   | 4038/6470 [46:29<24:41,  1.64it/s]
3|4039|Loss: 0.8931235074996948:  62%|██████▏   | 4038/6470 [46:30<24:41,  1.64it/s]
3|4039|Loss: 0.8931235074996948:  62%|██████▏   | 4039/6470 [46:30<23:57,  1.69it/s]
3|4040|Loss: 0.7294830679893494:  62%|██████▏   | 4039/6470 [46:30<23:57,  1.69it/s]
3|4040|Loss: 0.7294830679893494:  62%|██████▏   | 4040/6470 [46:30<24:03,  1.68it/s]
3|4041|Loss: 0.8800562620162964:  62%|██████▏   | 4040/6470 [46:31<24:03,  1.68it/s]
3|4041|Loss: 0.8800562620162964:  62%|██████▏   | 4041/6470 [46:31<23:27,  1.73it/s]
3|4042|Loss: 0.8748083114624023:  62%|██████▏   | 4041/6470 [46:31<23:27,  1.73it/s]
3|4042|Loss: 0.8748083114624023:  62%|██████▏   | 4042/6470 [46:32<23:55,  1.69it/s]
3|4043|Loss: 0.5769374966621399:  62%|██████▏   | 4042/6470 [46:32<23:55,  1.69it/s]
3|4043|Loss: 0.5769374966621399:  62%|██████▏   | 4043/6470 [46:32<24:34,  1.65it/s]
3|4044|Loss: 0.6803317070007324:  62%|██████▏   | 4043/6470 [46:33<24:34,  1.65it/s]
3|4044|Loss: 0.6803317070007324:  63%|██████▎   | 4044/6470 [46:33<25:34,  1.58it/s]
3|4045|Loss: 0.7906522750854492:  63%|██████▎   | 4044/6470 [46:33<25:34,  1.58it/s]
3|4045|Loss: 0.7906522750854492:  63%|██████▎   | 4045/6470 [46:33<24:31,  1.65it/s]
3|4046|Loss: 0.8798459768295288:  63%|██████▎   | 4045/6470 [46:34<24:31,  1.65it/s]
3|4046|Loss: 0.8798459768295288:  63%|██████▎   | 4046/6470 [46:34<25:10,  1.60it/s]
3|4047|Loss: 0.9042825102806091:  63%|██████▎   | 4046/6470 [46:34<25:10,  1.60it/s]
3|4047|Loss: 0.9042825102806091:  63%|██████▎   | 4047/6470 [46:35<25:25,  1.59it/s]
3|4048|Loss: 0.9363980889320374:  63%|██████▎   | 4047/6470 [46:35<25:25,  1.59it/s]
3|4048|Loss: 0.9363980889320374:  63%|██████▎   | 4048/6470 [46:35<24:48,  1.63it/s]
3|4049|Loss: 0.894634485244751:  63%|██████▎   | 4048/6470 [46:36<24:48,  1.63it/s] 
3|4049|Loss: 0.894634485244751:  63%|██████▎   | 4049/6470 [46:36<26:27,  1.53it/s]
3|4050|Loss: 0.8482940793037415:  63%|██████▎   | 4049/6470 [46:37<26:27,  1.53it/s]
3|4050|Loss: 0.8482940793037415:  63%|██████▎   | 4050/6470 [46:37<27:33,  1.46it/s]
3|4051|Loss: 0.8925115466117859:  63%|██████▎   | 4050/6470 [46:37<27:33,  1.46it/s]
3|4051|Loss: 0.8925115466117859:  63%|██████▎   | 4051/6470 [46:37<25:26,  1.58it/s]
3|4052|Loss: 0.8806069493293762:  63%|██████▎   | 4051/6470 [46:38<25:26,  1.58it/s]
3|4052|Loss: 0.8806069493293762:  63%|██████▎   | 4052/6470 [46:38<24:31,  1.64it/s]
3|4053|Loss: 0.9071330428123474:  63%|██████▎   | 4052/6470 [46:38<24:31,  1.64it/s]
3|4053|Loss: 0.9071330428123474:  63%|██████▎   | 4053/6470 [46:39<26:49,  1.50it/s]
3|4054|Loss: 0.8562328815460205:  63%|██████▎   | 4053/6470 [46:39<26:49,  1.50it/s]
3|4054|Loss: 0.8562328815460205:  63%|██████▎   | 4054/6470 [46:39<25:47,  1.56it/s]
3|4055|Loss: 0.9418436288833618:  63%|██████▎   | 4054/6470 [46:40<25:47,  1.56it/s]
3|4055|Loss: 0.9418436288833618:  63%|██████▎   | 4055/6470 [46:40<26:20,  1.53it/s]
3|4056|Loss: 0.7192156910896301:  63%|██████▎   | 4055/6470 [46:40<26:20,  1.53it/s]
3|4056|Loss: 0.7192156910896301:  63%|██████▎   | 4056/6470 [46:41<26:14,  1.53it/s]
3|4057|Loss: 0.8097177743911743:  63%|██████▎   | 4056/6470 [46:41<26:14,  1.53it/s]
3|4057|Loss: 0.8097177743911743:  63%|██████▎   | 4057/6470 [46:41<25:11,  1.60it/s]
3|4058|Loss: 0.9020435214042664:  63%|██████▎   | 4057/6470 [46:42<25:11,  1.60it/s]
3|4058|Loss: 0.9020435214042664:  63%|██████▎   | 4058/6470 [46:42<27:01,  1.49it/s]
3|4059|Loss: 0.9356938004493713:  63%|██████▎   | 4058/6470 [46:42<27:01,  1.49it/s]
3|4059|Loss: 0.9356938004493713:  63%|██████▎   | 4059/6470 [46:43<28:12,  1.42it/s]
3|4060|Loss: 1.211132287979126:  63%|██████▎   | 4059/6470 [46:43<28:12,  1.42it/s] 
3|4060|Loss: 1.211132287979126:  63%|██████▎   | 4060/6470 [46:43<28:49,  1.39it/s]
3|4061|Loss: 0.8040515184402466:  63%|██████▎   | 4060/6470 [46:44<28:49,  1.39it/s]
3|4061|Loss: 0.8040515184402466:  63%|██████▎   | 4061/6470 [46:44<27:57,  1.44it/s]
3|4062|Loss: 0.8472556471824646:  63%|██████▎   | 4061/6470 [46:44<27:57,  1.44it/s]
3|4062|Loss: 0.8472556471824646:  63%|██████▎   | 4062/6470 [46:45<26:32,  1.51it/s]
3|4063|Loss: 0.7777734398841858:  63%|██████▎   | 4062/6470 [46:45<26:32,  1.51it/s]
3|4063|Loss: 0.7777734398841858:  63%|██████▎   | 4063/6470 [46:45<25:35,  1.57it/s]
3|4064|Loss: 0.879233717918396:  63%|██████▎   | 4063/6470 [46:46<25:35,  1.57it/s] 
3|4064|Loss: 0.879233717918396:  63%|██████▎   | 4064/6470 [46:46<26:25,  1.52it/s]
3|4065|Loss: 0.9126083850860596:  63%|██████▎   | 4064/6470 [46:46<26:25,  1.52it/s]
3|4065|Loss: 0.9126083850860596:  63%|██████▎   | 4065/6470 [46:47<27:18,  1.47it/s]
3|4066|Loss: 0.8053794503211975:  63%|██████▎   | 4065/6470 [46:47<27:18,  1.47it/s]
3|4066|Loss: 0.8053794503211975:  63%|██████▎   | 4066/6470 [46:47<26:56,  1.49it/s]
3|4067|Loss: 0.7705281972885132:  63%|██████▎   | 4066/6470 [46:48<26:56,  1.49it/s]
3|4067|Loss: 0.7705281972885132:  63%|██████▎   | 4067/6470 [46:48<28:00,  1.43it/s]
3|4068|Loss: 0.8882670998573303:  63%|██████▎   | 4067/6470 [46:49<28:00,  1.43it/s]
3|4068|Loss: 0.8882670998573303:  63%|██████▎   | 4068/6470 [46:49<29:07,  1.37it/s]
3|4069|Loss: 0.8091369867324829:  63%|██████▎   | 4068/6470 [46:49<29:07,  1.37it/s]
3|4069|Loss: 0.8091369867324829:  63%|██████▎   | 4069/6470 [46:50<29:09,  1.37it/s]
3|4070|Loss: 0.7342458963394165:  63%|██████▎   | 4069/6470 [46:50<29:09,  1.37it/s]
3|4070|Loss: 0.7342458963394165:  63%|██████▎   | 4070/6470 [46:50<28:07,  1.42it/s]
3|4071|Loss: 0.8534870743751526:  63%|██████▎   | 4070/6470 [46:51<28:07,  1.42it/s]
3|4071|Loss: 0.8534870743751526:  63%|██████▎   | 4071/6470 [46:51<29:09,  1.37it/s]
3|4072|Loss: 0.7905993461608887:  63%|██████▎   | 4071/6470 [46:52<29:09,  1.37it/s]
3|4072|Loss: 0.7905993461608887:  63%|██████▎   | 4072/6470 [46:52<30:36,  1.31it/s]
3|4073|Loss: 0.8973979353904724:  63%|██████▎   | 4072/6470 [46:52<30:36,  1.31it/s]
3|4073|Loss: 0.8973979353904724:  63%|██████▎   | 4073/6470 [46:53<29:28,  1.36it/s]
3|4074|Loss: 0.9680524468421936:  63%|██████▎   | 4073/6470 [46:53<29:28,  1.36it/s]
3|4074|Loss: 0.9680524468421936:  63%|██████▎   | 4074/6470 [46:53<27:43,  1.44it/s]
3|4075|Loss: 0.7395385503768921:  63%|██████▎   | 4074/6470 [46:54<27:43,  1.44it/s]
3|4075|Loss: 0.7395385503768921:  63%|██████▎   | 4075/6470 [46:54<25:42,  1.55it/s]
3|4076|Loss: 0.906078577041626:  63%|██████▎   | 4075/6470 [46:54<25:42,  1.55it/s] 
3|4076|Loss: 0.906078577041626:  63%|██████▎   | 4076/6470 [46:55<28:01,  1.42it/s]
3|4077|Loss: 0.8792040348052979:  63%|██████▎   | 4076/6470 [46:55<28:01,  1.42it/s]
3|4077|Loss: 0.8792040348052979:  63%|██████▎   | 4077/6470 [46:55<26:54,  1.48it/s]
3|4078|Loss: 0.8974314332008362:  63%|██████▎   | 4077/6470 [46:56<26:54,  1.48it/s]
3|4078|Loss: 0.8974314332008362:  63%|██████▎   | 4078/6470 [46:56<25:10,  1.58it/s]
3|4079|Loss: 0.8208196759223938:  63%|██████▎   | 4078/6470 [46:56<25:10,  1.58it/s]
3|4079|Loss: 0.8208196759223938:  63%|██████▎   | 4079/6470 [46:57<27:19,  1.46it/s]
3|4080|Loss: 0.9058224558830261:  63%|██████▎   | 4079/6470 [46:57<27:19,  1.46it/s]
3|4080|Loss: 0.9058224558830261:  63%|██████▎   | 4080/6470 [46:57<28:18,  1.41it/s]
3|4081|Loss: 0.9864954352378845:  63%|██████▎   | 4080/6470 [46:58<28:18,  1.41it/s]
3|4081|Loss: 0.9864954352378845:  63%|██████▎   | 4081/6470 [46:58<29:07,  1.37it/s]
3|4082|Loss: 0.8147809505462646:  63%|██████▎   | 4081/6470 [46:59<29:07,  1.37it/s]
3|4082|Loss: 0.8147809505462646:  63%|██████▎   | 4082/6470 [46:59<29:16,  1.36it/s]
3|4083|Loss: 0.6750466227531433:  63%|██████▎   | 4082/6470 [46:59<29:16,  1.36it/s]
3|4083|Loss: 0.6750466227531433:  63%|██████▎   | 4083/6470 [47:00<30:10,  1.32it/s]
3|4084|Loss: 0.8481358885765076:  63%|██████▎   | 4083/6470 [47:00<30:10,  1.32it/s]
3|4084|Loss: 0.8481358885765076:  63%|██████▎   | 4084/6470 [47:01<31:26,  1.27it/s]
3|4085|Loss: 0.863243579864502:  63%|██████▎   | 4084/6470 [47:01<31:26,  1.27it/s] 
3|4085|Loss: 0.863243579864502:  63%|██████▎   | 4085/6470 [47:01<31:42,  1.25it/s]
3|4086|Loss: 0.7621645331382751:  63%|██████▎   | 4085/6470 [47:02<31:42,  1.25it/s]
3|4086|Loss: 0.7621645331382751:  63%|██████▎   | 4086/6470 [47:02<31:16,  1.27it/s]
3|4087|Loss: 0.6764335036277771:  63%|██████▎   | 4086/6470 [47:02<31:16,  1.27it/s]
3|4087|Loss: 0.6764335036277771:  63%|██████▎   | 4087/6470 [47:03<29:51,  1.33it/s]
3|4088|Loss: 0.8335779905319214:  63%|██████▎   | 4087/6470 [47:03<29:51,  1.33it/s]
3|4088|Loss: 0.8335779905319214:  63%|██████▎   | 4088/6470 [47:03<25:58,  1.53it/s]
3|4089|Loss: 0.794032871723175:  63%|██████▎   | 4088/6470 [47:04<25:58,  1.53it/s] 
3|4089|Loss: 0.794032871723175:  63%|██████▎   | 4089/6470 [47:04<25:57,  1.53it/s]
3|4090|Loss: 0.852670431137085:  63%|██████▎   | 4089/6470 [47:04<25:57,  1.53it/s]
3|4090|Loss: 0.852670431137085:  63%|██████▎   | 4090/6470 [47:05<27:05,  1.46it/s]
3|4091|Loss: 0.7532605528831482:  63%|██████▎   | 4090/6470 [47:05<27:05,  1.46it/s]
3|4091|Loss: 0.7532605528831482:  63%|██████▎   | 4091/6470 [47:05<26:14,  1.51it/s]
3|4092|Loss: 0.8446216583251953:  63%|██████▎   | 4091/6470 [47:06<26:14,  1.51it/s]
3|4092|Loss: 0.8446216583251953:  63%|██████▎   | 4092/6470 [47:06<26:51,  1.48it/s]
3|4093|Loss: 1.0019294023513794:  63%|██████▎   | 4092/6470 [47:06<26:51,  1.48it/s]
3|4093|Loss: 1.0019294023513794:  63%|██████▎   | 4093/6470 [47:07<28:43,  1.38it/s]
3|4094|Loss: 0.8262530565261841:  63%|██████▎   | 4093/6470 [47:07<28:43,  1.38it/s]
3|4094|Loss: 0.8262530565261841:  63%|██████▎   | 4094/6470 [47:07<28:08,  1.41it/s]
3|4095|Loss: 0.8208113312721252:  63%|██████▎   | 4094/6470 [47:08<28:08,  1.41it/s]
3|4095|Loss: 0.8208113312721252:  63%|██████▎   | 4095/6470 [47:08<28:04,  1.41it/s]
3|4096|Loss: 0.9617534279823303:  63%|██████▎   | 4095/6470 [47:08<28:04,  1.41it/s]
3|4096|Loss: 0.9617534279823303:  63%|██████▎   | 4096/6470 [47:09<26:23,  1.50it/s]
3|4097|Loss: 0.9534527063369751:  63%|██████▎   | 4096/6470 [47:09<26:23,  1.50it/s]
3|4097|Loss: 0.9534527063369751:  63%|██████▎   | 4097/6470 [47:09<26:45,  1.48it/s]
3|4098|Loss: 0.8277902007102966:  63%|██████▎   | 4097/6470 [47:10<26:45,  1.48it/s]
3|4098|Loss: 0.8277902007102966:  63%|██████▎   | 4098/6470 [47:10<24:14,  1.63it/s]
3|4099|Loss: 0.9299476742744446:  63%|██████▎   | 4098/6470 [47:10<24:14,  1.63it/s]
3|4099|Loss: 0.9299476742744446:  63%|██████▎   | 4099/6470 [47:11<25:00,  1.58it/s]
3|4100|Loss: 0.8343124389648438:  63%|██████▎   | 4099/6470 [47:11<25:00,  1.58it/s]
3|4100|Loss: 0.8343124389648438:  63%|██████▎   | 4100/6470 [47:11<27:59,  1.41it/s]
3|4101|Loss: 0.9857669472694397:  63%|██████▎   | 4100/6470 [47:12<27:59,  1.41it/s]
3|4101|Loss: 0.9857669472694397:  63%|██████▎   | 4101/6470 [47:12<26:08,  1.51it/s]
3|4102|Loss: 0.7387367486953735:  63%|██████▎   | 4101/6470 [47:12<26:08,  1.51it/s]
3|4102|Loss: 0.7387367486953735:  63%|██████▎   | 4102/6470 [47:13<26:43,  1.48it/s]
3|4103|Loss: 0.7472476959228516:  63%|██████▎   | 4102/6470 [47:13<26:43,  1.48it/s]
3|4103|Loss: 0.7472476959228516:  63%|██████▎   | 4103/6470 [47:13<27:41,  1.42it/s]
3|4104|Loss: 0.8351944088935852:  63%|██████▎   | 4103/6470 [47:14<27:41,  1.42it/s]
3|4104|Loss: 0.8351944088935852:  63%|██████▎   | 4104/6470 [47:14<29:01,  1.36it/s]
3|4105|Loss: 0.8659577965736389:  63%|██████▎   | 4104/6470 [47:15<29:01,  1.36it/s]
3|4105|Loss: 0.8659577965736389:  63%|██████▎   | 4105/6470 [47:15<28:21,  1.39it/s]
3|4106|Loss: 0.683786928653717:  63%|██████▎   | 4105/6470 [47:15<28:21,  1.39it/s] 
3|4106|Loss: 0.683786928653717:  63%|██████▎   | 4106/6470 [47:16<27:14,  1.45it/s]
3|4107|Loss: 0.920158326625824:  63%|██████▎   | 4106/6470 [47:16<27:14,  1.45it/s]
3|4107|Loss: 0.920158326625824:  63%|██████▎   | 4107/6470 [47:16<27:28,  1.43it/s]
3|4108|Loss: 0.938921332359314:  63%|██████▎   | 4107/6470 [47:17<27:28,  1.43it/s]
3|4108|Loss: 0.938921332359314:  63%|██████▎   | 4108/6470 [47:17<26:50,  1.47it/s]
3|4109|Loss: 0.8558058142662048:  63%|██████▎   | 4108/6470 [47:17<26:50,  1.47it/s]
3|4109|Loss: 0.8558058142662048:  64%|██████▎   | 4109/6470 [47:18<27:19,  1.44it/s]
3|4110|Loss: 0.9723339676856995:  64%|██████▎   | 4109/6470 [47:18<27:19,  1.44it/s]
3|4110|Loss: 0.9723339676856995:  64%|██████▎   | 4110/6470 [47:18<27:34,  1.43it/s]
3|4111|Loss: 0.9218801856040955:  64%|██████▎   | 4110/6470 [47:19<27:34,  1.43it/s]
3|4111|Loss: 0.9218801856040955:  64%|██████▎   | 4111/6470 [47:19<27:38,  1.42it/s]
3|4112|Loss: 0.8746964335441589:  64%|██████▎   | 4111/6470 [47:19<27:38,  1.42it/s]
3|4112|Loss: 0.8746964335441589:  64%|██████▎   | 4112/6470 [47:20<26:22,  1.49it/s]
3|4113|Loss: 0.7895814180374146:  64%|██████▎   | 4112/6470 [47:20<26:22,  1.49it/s]
3|4113|Loss: 0.7895814180374146:  64%|██████▎   | 4113/6470 [47:20<25:13,  1.56it/s]
3|4114|Loss: 0.8333039283752441:  64%|██████▎   | 4113/6470 [47:21<25:13,  1.56it/s]
3|4114|Loss: 0.8333039283752441:  64%|██████▎   | 4114/6470 [47:21<24:35,  1.60it/s]
3|4115|Loss: 0.7455552816390991:  64%|██████▎   | 4114/6470 [47:21<24:35,  1.60it/s]
3|4115|Loss: 0.7455552816390991:  64%|██████▎   | 4115/6470 [47:21<24:40,  1.59it/s]
3|4116|Loss: 0.94697505235672:  64%|██████▎   | 4115/6470 [47:22<24:40,  1.59it/s]  
3|4116|Loss: 0.94697505235672:  64%|██████▎   | 4116/6470 [47:22<24:36,  1.59it/s]
3|4117|Loss: 1.0050621032714844:  64%|██████▎   | 4116/6470 [47:23<24:36,  1.59it/s]
3|4117|Loss: 1.0050621032714844:  64%|██████▎   | 4117/6470 [47:23<26:57,  1.45it/s]
3|4118|Loss: 0.992965817451477:  64%|██████▎   | 4117/6470 [47:23<26:57,  1.45it/s] 
3|4118|Loss: 0.992965817451477:  64%|██████▎   | 4118/6470 [47:24<27:01,  1.45it/s]
3|4119|Loss: 0.7662760615348816:  64%|██████▎   | 4118/6470 [47:24<27:01,  1.45it/s]
3|4119|Loss: 0.7662760615348816:  64%|██████▎   | 4119/6470 [47:24<26:27,  1.48it/s]
3|4120|Loss: 0.9064733386039734:  64%|██████▎   | 4119/6470 [47:25<26:27,  1.48it/s]
3|4120|Loss: 0.9064733386039734:  64%|██████▎   | 4120/6470 [47:25<27:10,  1.44it/s]
3|4121|Loss: 0.8871771097183228:  64%|██████▎   | 4120/6470 [47:25<27:10,  1.44it/s]
3|4121|Loss: 0.8871771097183228:  64%|██████▎   | 4121/6470 [47:26<26:16,  1.49it/s]
3|4122|Loss: 0.8353009819984436:  64%|██████▎   | 4121/6470 [47:26<26:16,  1.49it/s]
3|4122|Loss: 0.8353009819984436:  64%|██████▎   | 4122/6470 [47:26<23:22,  1.67it/s]
3|4123|Loss: 0.7617306113243103:  64%|██████▎   | 4122/6470 [47:26<23:22,  1.67it/s]
3|4123|Loss: 0.7617306113243103:  64%|██████▎   | 4123/6470 [47:27<24:26,  1.60it/s]
3|4124|Loss: 0.9648737907409668:  64%|██████▎   | 4123/6470 [47:27<24:26,  1.60it/s]
3|4124|Loss: 0.9648737907409668:  64%|██████▎   | 4124/6470 [47:27<24:30,  1.60it/s]
3|4125|Loss: 0.933995246887207:  64%|██████▎   | 4124/6470 [47:28<24:30,  1.60it/s] 
3|4125|Loss: 0.933995246887207:  64%|██████▍   | 4125/6470 [47:28<26:53,  1.45it/s]
3|4126|Loss: 0.9668277502059937:  64%|██████▍   | 4125/6470 [47:29<26:53,  1.45it/s]
3|4126|Loss: 0.9668277502059937:  64%|██████▍   | 4126/6470 [47:29<27:49,  1.40it/s]
3|4127|Loss: 0.9187262654304504:  64%|██████▍   | 4126/6470 [47:29<27:49,  1.40it/s]
3|4127|Loss: 0.9187262654304504:  64%|██████▍   | 4127/6470 [47:30<27:51,  1.40it/s]
3|4128|Loss: 0.7634254693984985:  64%|██████▍   | 4127/6470 [47:30<27:51,  1.40it/s]
3|4128|Loss: 0.7634254693984985:  64%|██████▍   | 4128/6470 [47:30<28:55,  1.35it/s]
3|4129|Loss: 0.8046882748603821:  64%|██████▍   | 4128/6470 [47:31<28:55,  1.35it/s]
3|4129|Loss: 0.8046882748603821:  64%|██████▍   | 4129/6470 [47:31<30:46,  1.27it/s]
3|4130|Loss: 0.9603057503700256:  64%|██████▍   | 4129/6470 [47:32<30:46,  1.27it/s]
3|4130|Loss: 0.9603057503700256:  64%|██████▍   | 4130/6470 [47:32<30:42,  1.27it/s]
3|4131|Loss: 0.9376872181892395:  64%|██████▍   | 4130/6470 [47:33<30:42,  1.27it/s]
3|4131|Loss: 0.9376872181892395:  64%|██████▍   | 4131/6470 [47:33<29:07,  1.34it/s]
3|4132|Loss: 0.8083811402320862:  64%|██████▍   | 4131/6470 [47:33<29:07,  1.34it/s]
3|4132|Loss: 0.8083811402320862:  64%|██████▍   | 4132/6470 [47:33<27:12,  1.43it/s]
3|4133|Loss: 0.7257398962974548:  64%|██████▍   | 4132/6470 [47:34<27:12,  1.43it/s]
3|4133|Loss: 0.7257398962974548:  64%|██████▍   | 4133/6470 [47:34<27:22,  1.42it/s]
3|4134|Loss: 0.7616540789604187:  64%|██████▍   | 4133/6470 [47:35<27:22,  1.42it/s]
3|4134|Loss: 0.7616540789604187:  64%|██████▍   | 4134/6470 [47:35<26:45,  1.46it/s]
3|4135|Loss: 0.8178836107254028:  64%|██████▍   | 4134/6470 [47:35<26:45,  1.46it/s]
3|4135|Loss: 0.8178836107254028:  64%|██████▍   | 4135/6470 [47:36<27:56,  1.39it/s]
3|4136|Loss: 0.860417366027832:  64%|██████▍   | 4135/6470 [47:36<27:56,  1.39it/s] 
3|4136|Loss: 0.860417366027832:  64%|██████▍   | 4136/6470 [47:36<27:31,  1.41it/s]
3|4137|Loss: 0.8794114589691162:  64%|██████▍   | 4136/6470 [47:37<27:31,  1.41it/s]
3|4137|Loss: 0.8794114589691162:  64%|██████▍   | 4137/6470 [47:37<28:57,  1.34it/s]
3|4138|Loss: 0.9303853511810303:  64%|██████▍   | 4137/6470 [47:38<28:57,  1.34it/s]
3|4138|Loss: 0.9303853511810303:  64%|██████▍   | 4138/6470 [47:38<30:16,  1.28it/s]
3|4139|Loss: 0.8759669065475464:  64%|██████▍   | 4138/6470 [47:38<30:16,  1.28it/s]
3|4139|Loss: 0.8759669065475464:  64%|██████▍   | 4139/6470 [47:39<27:57,  1.39it/s]
3|4140|Loss: 0.7498353123664856:  64%|██████▍   | 4139/6470 [47:39<27:57,  1.39it/s]
3|4140|Loss: 0.7498353123664856:  64%|██████▍   | 4140/6470 [47:39<26:32,  1.46it/s]
3|4141|Loss: 0.9492455124855042:  64%|██████▍   | 4140/6470 [47:40<26:32,  1.46it/s]
3|4141|Loss: 0.9492455124855042:  64%|██████▍   | 4141/6470 [47:40<28:21,  1.37it/s]
3|4142|Loss: 0.7681400179862976:  64%|██████▍   | 4141/6470 [47:40<28:21,  1.37it/s]
3|4142|Loss: 0.7681400179862976:  64%|██████▍   | 4142/6470 [47:41<29:40,  1.31it/s]
3|4143|Loss: 0.8773558735847473:  64%|██████▍   | 4142/6470 [47:41<29:40,  1.31it/s]
3|4143|Loss: 0.8773558735847473:  64%|██████▍   | 4143/6470 [47:41<28:46,  1.35it/s]
3|4144|Loss: 0.7912867665290833:  64%|██████▍   | 4143/6470 [47:42<28:46,  1.35it/s]
3|4144|Loss: 0.7912867665290833:  64%|██████▍   | 4144/6470 [47:42<29:43,  1.30it/s]
3|4145|Loss: 0.8793197274208069:  64%|██████▍   | 4144/6470 [47:43<29:43,  1.30it/s]
3|4145|Loss: 0.8793197274208069:  64%|██████▍   | 4145/6470 [47:43<28:19,  1.37it/s]
3|4146|Loss: 0.9237403869628906:  64%|██████▍   | 4145/6470 [47:43<28:19,  1.37it/s]
3|4146|Loss: 0.9237403869628906:  64%|██████▍   | 4146/6470 [47:44<28:11,  1.37it/s]
3|4147|Loss: 0.8221700191497803:  64%|██████▍   | 4146/6470 [47:44<28:11,  1.37it/s]
3|4147|Loss: 0.8221700191497803:  64%|██████▍   | 4147/6470 [47:44<28:17,  1.37it/s]
3|4148|Loss: 0.9341634511947632:  64%|██████▍   | 4147/6470 [47:45<28:17,  1.37it/s]
3|4148|Loss: 0.9341634511947632:  64%|██████▍   | 4148/6470 [47:45<27:26,  1.41it/s]
3|4149|Loss: 0.78066086769104:  64%|██████▍   | 4148/6470 [47:46<27:26,  1.41it/s]  
3|4149|Loss: 0.78066086769104:  64%|██████▍   | 4149/6470 [47:46<28:49,  1.34it/s]
3|4150|Loss: 0.7803488373756409:  64%|██████▍   | 4149/6470 [47:46<28:49,  1.34it/s]
3|4150|Loss: 0.7803488373756409:  64%|██████▍   | 4150/6470 [47:46<26:47,  1.44it/s]
3|4151|Loss: 0.8772189021110535:  64%|██████▍   | 4150/6470 [47:47<26:47,  1.44it/s]
3|4151|Loss: 0.8772189021110535:  64%|██████▍   | 4151/6470 [47:47<25:07,  1.54it/s]
3|4152|Loss: 0.7928729057312012:  64%|██████▍   | 4151/6470 [47:47<25:07,  1.54it/s]
3|4152|Loss: 0.7928729057312012:  64%|██████▍   | 4152/6470 [47:48<26:51,  1.44it/s]
3|4153|Loss: 0.8485298752784729:  64%|██████▍   | 4152/6470 [47:48<26:51,  1.44it/s]
3|4153|Loss: 0.8485298752784729:  64%|██████▍   | 4153/6470 [47:49<27:18,  1.41it/s]
3|4154|Loss: 0.6870179772377014:  64%|██████▍   | 4153/6470 [47:49<27:18,  1.41it/s]
3|4154|Loss: 0.6870179772377014:  64%|██████▍   | 4154/6470 [47:49<25:35,  1.51it/s]
3|4155|Loss: 0.9538962244987488:  64%|██████▍   | 4154/6470 [47:49<25:35,  1.51it/s]
3|4155|Loss: 0.9538962244987488:  64%|██████▍   | 4155/6470 [47:50<25:08,  1.53it/s]
3|4156|Loss: 0.8177053332328796:  64%|██████▍   | 4155/6470 [47:50<25:08,  1.53it/s]
3|4156|Loss: 0.8177053332328796:  64%|██████▍   | 4156/6470 [47:50<25:50,  1.49it/s]
3|4157|Loss: 0.8867446184158325:  64%|██████▍   | 4156/6470 [47:51<25:50,  1.49it/s]
3|4157|Loss: 0.8867446184158325:  64%|██████▍   | 4157/6470 [47:51<26:16,  1.47it/s]
3|4158|Loss: 0.8333694934844971:  64%|██████▍   | 4157/6470 [47:52<26:16,  1.47it/s]
3|4158|Loss: 0.8333694934844971:  64%|██████▍   | 4158/6470 [47:52<26:37,  1.45it/s]
3|4159|Loss: 0.8802335262298584:  64%|██████▍   | 4158/6470 [47:52<26:37,  1.45it/s]
3|4159|Loss: 0.8802335262298584:  64%|██████▍   | 4159/6470 [47:53<27:42,  1.39it/s]
3|4160|Loss: 0.7382221817970276:  64%|██████▍   | 4159/6470 [47:53<27:42,  1.39it/s]
3|4160|Loss: 0.7382221817970276:  64%|██████▍   | 4160/6470 [47:54<29:13,  1.32it/s]
3|4161|Loss: 0.7620488405227661:  64%|██████▍   | 4160/6470 [47:54<29:13,  1.32it/s]
3|4161|Loss: 0.7620488405227661:  64%|██████▍   | 4161/6470 [47:54<29:18,  1.31it/s]
3|4162|Loss: 1.0312570333480835:  64%|██████▍   | 4161/6470 [47:55<29:18,  1.31it/s]
3|4162|Loss: 1.0312570333480835:  64%|██████▍   | 4162/6470 [47:55<27:15,  1.41it/s]
3|4163|Loss: 0.9012312889099121:  64%|██████▍   | 4162/6470 [47:55<27:15,  1.41it/s]
3|4163|Loss: 0.9012312889099121:  64%|██████▍   | 4163/6470 [47:55<26:04,  1.47it/s]
3|4164|Loss: 0.8755578398704529:  64%|██████▍   | 4163/6470 [47:56<26:04,  1.47it/s]
3|4164|Loss: 0.8755578398704529:  64%|██████▍   | 4164/6470 [47:56<26:22,  1.46it/s]
3|4165|Loss: 1.0361814498901367:  64%|██████▍   | 4164/6470 [47:57<26:22,  1.46it/s]
3|4165|Loss: 1.0361814498901367:  64%|██████▍   | 4165/6470 [47:57<27:00,  1.42it/s]
3|4166|Loss: 0.7151798009872437:  64%|██████▍   | 4165/6470 [47:57<27:00,  1.42it/s]
3|4166|Loss: 0.7151798009872437:  64%|██████▍   | 4166/6470 [47:58<26:32,  1.45it/s]
3|4167|Loss: 0.8989031910896301:  64%|██████▍   | 4166/6470 [47:58<26:32,  1.45it/s]
3|4167|Loss: 0.8989031910896301:  64%|██████▍   | 4167/6470 [47:58<25:52,  1.48it/s]
3|4168|Loss: 0.8881474733352661:  64%|██████▍   | 4167/6470 [47:59<25:52,  1.48it/s]
3|4168|Loss: 0.8881474733352661:  64%|██████▍   | 4168/6470 [47:59<24:53,  1.54it/s]
3|4169|Loss: 0.8986833095550537:  64%|██████▍   | 4168/6470 [47:59<24:53,  1.54it/s]
3|4169|Loss: 0.8986833095550537:  64%|██████▍   | 4169/6470 [47:59<24:29,  1.57it/s]
3|4170|Loss: 0.8353731036186218:  64%|██████▍   | 4169/6470 [48:00<24:29,  1.57it/s]
3|4170|Loss: 0.8353731036186218:  64%|██████▍   | 4170/6470 [48:00<26:25,  1.45it/s]
3|4171|Loss: 0.8414186239242554:  64%|██████▍   | 4170/6470 [48:01<26:25,  1.45it/s]
3|4171|Loss: 0.8414186239242554:  64%|██████▍   | 4171/6470 [48:01<27:20,  1.40it/s]
3|4172|Loss: 0.9338329434394836:  64%|██████▍   | 4171/6470 [48:01<27:20,  1.40it/s]
3|4172|Loss: 0.9338329434394836:  64%|██████▍   | 4172/6470 [48:02<28:27,  1.35it/s]
3|4173|Loss: 0.8007230162620544:  64%|██████▍   | 4172/6470 [48:02<28:27,  1.35it/s]
3|4173|Loss: 0.8007230162620544:  64%|██████▍   | 4173/6470 [48:03<28:32,  1.34it/s]
3|4174|Loss: 0.8764635920524597:  64%|██████▍   | 4173/6470 [48:03<28:32,  1.34it/s]
3|4174|Loss: 0.8764635920524597:  65%|██████▍   | 4174/6470 [48:03<26:42,  1.43it/s]
3|4175|Loss: 0.8274762034416199:  65%|██████▍   | 4174/6470 [48:04<26:42,  1.43it/s]
3|4175|Loss: 0.8274762034416199:  65%|██████▍   | 4175/6470 [48:04<28:51,  1.33it/s]
3|4176|Loss: 0.804450273513794:  65%|██████▍   | 4175/6470 [48:04<28:51,  1.33it/s] 
3|4176|Loss: 0.804450273513794:  65%|██████▍   | 4176/6470 [48:05<28:04,  1.36it/s]
3|4177|Loss: 0.8440483212471008:  65%|██████▍   | 4176/6470 [48:05<28:04,  1.36it/s]
3|4177|Loss: 0.8440483212471008:  65%|██████▍   | 4177/6470 [48:05<27:27,  1.39it/s]
3|4178|Loss: 0.8306224346160889:  65%|██████▍   | 4177/6470 [48:06<27:27,  1.39it/s]
3|4178|Loss: 0.8306224346160889:  65%|██████▍   | 4178/6470 [48:06<26:30,  1.44it/s]
3|4179|Loss: 0.7091326713562012:  65%|██████▍   | 4178/6470 [48:06<26:30,  1.44it/s]
3|4179|Loss: 0.7091326713562012:  65%|██████▍   | 4179/6470 [48:07<25:40,  1.49it/s]
3|4180|Loss: 0.9109010696411133:  65%|██████▍   | 4179/6470 [48:07<25:40,  1.49it/s]
3|4180|Loss: 0.9109010696411133:  65%|██████▍   | 4180/6470 [48:07<26:14,  1.45it/s]
3|4181|Loss: 0.9859986305236816:  65%|██████▍   | 4180/6470 [48:08<26:14,  1.45it/s]
3|4181|Loss: 0.9859986305236816:  65%|██████▍   | 4181/6470 [48:08<26:53,  1.42it/s]
3|4182|Loss: 0.9242603778839111:  65%|██████▍   | 4181/6470 [48:09<26:53,  1.42it/s]
3|4182|Loss: 0.9242603778839111:  65%|██████▍   | 4182/6470 [48:09<25:46,  1.48it/s]
3|4183|Loss: 0.7301314473152161:  65%|██████▍   | 4182/6470 [48:09<25:46,  1.48it/s]
3|4183|Loss: 0.7301314473152161:  65%|██████▍   | 4183/6470 [48:09<24:10,  1.58it/s]
3|4184|Loss: 0.789169192314148:  65%|██████▍   | 4183/6470 [48:10<24:10,  1.58it/s] 
3|4184|Loss: 0.789169192314148:  65%|██████▍   | 4184/6470 [48:10<23:58,  1.59it/s]
3|4185|Loss: 0.7493741512298584:  65%|██████▍   | 4184/6470 [48:10<23:58,  1.59it/s]
3|4185|Loss: 0.7493741512298584:  65%|██████▍   | 4185/6470 [48:11<24:39,  1.54it/s]
3|4186|Loss: 1.0763766765594482:  65%|██████▍   | 4185/6470 [48:11<24:39,  1.54it/s]
3|4186|Loss: 1.0763766765594482:  65%|██████▍   | 4186/6470 [48:11<26:37,  1.43it/s]
3|4187|Loss: 0.9343491792678833:  65%|██████▍   | 4186/6470 [48:12<26:37,  1.43it/s]
3|4187|Loss: 0.9343491792678833:  65%|██████▍   | 4187/6470 [48:12<28:14,  1.35it/s]
3|4188|Loss: 0.9042229652404785:  65%|██████▍   | 4187/6470 [48:13<28:14,  1.35it/s]
3|4188|Loss: 0.9042229652404785:  65%|██████▍   | 4188/6470 [48:13<26:20,  1.44it/s]
3|4189|Loss: 0.958109438419342:  65%|██████▍   | 4188/6470 [48:13<26:20,  1.44it/s] 
3|4189|Loss: 0.958109438419342:  65%|██████▍   | 4189/6470 [48:13<25:13,  1.51it/s]
3|4190|Loss: 1.0524193048477173:  65%|██████▍   | 4189/6470 [48:14<25:13,  1.51it/s]
3|4190|Loss: 1.0524193048477173:  65%|██████▍   | 4190/6470 [48:14<23:16,  1.63it/s]
3|4191|Loss: 0.8765337467193604:  65%|██████▍   | 4190/6470 [48:14<23:16,  1.63it/s]
3|4191|Loss: 0.8765337467193604:  65%|██████▍   | 4191/6470 [48:14<21:54,  1.73it/s]
3|4192|Loss: 0.7428185343742371:  65%|██████▍   | 4191/6470 [48:15<21:54,  1.73it/s]
3|4192|Loss: 0.7428185343742371:  65%|██████▍   | 4192/6470 [48:15<19:37,  1.94it/s]
3|4193|Loss: 0.7231716513633728:  65%|██████▍   | 4192/6470 [48:15<19:37,  1.94it/s]
3|4193|Loss: 0.7231716513633728:  65%|██████▍   | 4193/6470 [48:15<19:04,  1.99it/s]
3|4194|Loss: 0.7907559275627136:  65%|██████▍   | 4193/6470 [48:16<19:04,  1.99it/s]
3|4194|Loss: 0.7907559275627136:  65%|██████▍   | 4194/6470 [48:16<21:10,  1.79it/s]
3|4195|Loss: 0.7828896641731262:  65%|██████▍   | 4194/6470 [48:16<21:10,  1.79it/s]
3|4195|Loss: 0.7828896641731262:  65%|██████▍   | 4195/6470 [48:17<21:20,  1.78it/s]
3|4196|Loss: 1.0046111345291138:  65%|██████▍   | 4195/6470 [48:17<21:20,  1.78it/s]
3|4196|Loss: 1.0046111345291138:  65%|██████▍   | 4196/6470 [48:17<21:08,  1.79it/s]
3|4197|Loss: 0.8375858664512634:  65%|██████▍   | 4196/6470 [48:17<21:08,  1.79it/s]
3|4197|Loss: 0.8375858664512634:  65%|██████▍   | 4197/6470 [48:18<20:13,  1.87it/s]
3|4198|Loss: 0.8195798397064209:  65%|██████▍   | 4197/6470 [48:18<20:13,  1.87it/s]
3|4198|Loss: 0.8195798397064209:  65%|██████▍   | 4198/6470 [48:18<20:56,  1.81it/s]
3|4199|Loss: 0.7693520188331604:  65%|██████▍   | 4198/6470 [48:18<20:56,  1.81it/s]
3|4199|Loss: 0.7693520188331604:  65%|██████▍   | 4199/6470 [48:19<20:44,  1.82it/s]
3|4200|Loss: 0.9016183614730835:  65%|██████▍   | 4199/6470 [48:19<20:44,  1.82it/s]
3|4200|Loss: 0.9016183614730835:  65%|██████▍   | 4200/6470 [48:19<21:13,  1.78it/s]
3|4201|Loss: 0.8086779117584229:  65%|██████▍   | 4200/6470 [48:20<21:13,  1.78it/s]
3|4201|Loss: 0.8086779117584229:  65%|██████▍   | 4201/6470 [48:20<23:34,  1.60it/s]
3|4202|Loss: 1.0319933891296387:  65%|██████▍   | 4201/6470 [48:20<23:34,  1.60it/s]
3|4202|Loss: 1.0319933891296387:  65%|██████▍   | 4202/6470 [48:21<23:08,  1.63it/s]
3|4203|Loss: 0.9354686737060547:  65%|██████▍   | 4202/6470 [48:21<23:08,  1.63it/s]
3|4203|Loss: 0.9354686737060547:  65%|██████▍   | 4203/6470 [48:21<23:13,  1.63it/s]
3|4204|Loss: 0.8529040813446045:  65%|██████▍   | 4203/6470 [48:22<23:13,  1.63it/s]
3|4204|Loss: 0.8529040813446045:  65%|██████▍   | 4204/6470 [48:22<22:59,  1.64it/s]
3|4205|Loss: 0.9124716520309448:  65%|██████▍   | 4204/6470 [48:22<22:59,  1.64it/s]
3|4205|Loss: 0.9124716520309448:  65%|██████▍   | 4205/6470 [48:22<22:26,  1.68it/s]
3|4206|Loss: 0.8180010914802551:  65%|██████▍   | 4205/6470 [48:23<22:26,  1.68it/s]
3|4206|Loss: 0.8180010914802551:  65%|██████▌   | 4206/6470 [48:23<24:32,  1.54it/s]
3|4207|Loss: 0.8260031938552856:  65%|██████▌   | 4206/6470 [48:24<24:32,  1.54it/s]
3|4207|Loss: 0.8260031938552856:  65%|██████▌   | 4207/6470 [48:24<24:23,  1.55it/s]
3|4208|Loss: 0.7382034659385681:  65%|██████▌   | 4207/6470 [48:24<24:23,  1.55it/s]
3|4208|Loss: 0.7382034659385681:  65%|██████▌   | 4208/6470 [48:24<24:08,  1.56it/s]
3|4209|Loss: 0.798480212688446:  65%|██████▌   | 4208/6470 [48:25<24:08,  1.56it/s] 
3|4209|Loss: 0.798480212688446:  65%|██████▌   | 4209/6470 [48:25<25:10,  1.50it/s]
3|4210|Loss: 0.7407510876655579:  65%|██████▌   | 4209/6470 [48:26<25:10,  1.50it/s]
3|4210|Loss: 0.7407510876655579:  65%|██████▌   | 4210/6470 [48:26<27:07,  1.39it/s]
3|4211|Loss: 0.9212055206298828:  65%|██████▌   | 4210/6470 [48:26<27:07,  1.39it/s]
3|4211|Loss: 0.9212055206298828:  65%|██████▌   | 4211/6470 [48:27<27:24,  1.37it/s]
3|4212|Loss: 0.8636752367019653:  65%|██████▌   | 4211/6470 [48:27<27:24,  1.37it/s]
3|4212|Loss: 0.8636752367019653:  65%|██████▌   | 4212/6470 [48:28<27:30,  1.37it/s]
3|4213|Loss: 0.8014007806777954:  65%|██████▌   | 4212/6470 [48:28<27:30,  1.37it/s]
3|4213|Loss: 0.8014007806777954:  65%|██████▌   | 4213/6470 [48:28<25:03,  1.50it/s]
3|4214|Loss: 0.810634434223175:  65%|██████▌   | 4213/6470 [48:28<25:03,  1.50it/s] 
3|4214|Loss: 0.810634434223175:  65%|██████▌   | 4214/6470 [48:29<23:38,  1.59it/s]
3|4215|Loss: 0.8318970799446106:  65%|██████▌   | 4214/6470 [48:29<23:38,  1.59it/s]
3|4215|Loss: 0.8318970799446106:  65%|██████▌   | 4215/6470 [48:29<23:01,  1.63it/s]
3|4216|Loss: 0.7238187789916992:  65%|██████▌   | 4215/6470 [48:29<23:01,  1.63it/s]
3|4216|Loss: 0.7238187789916992:  65%|██████▌   | 4216/6470 [48:30<22:56,  1.64it/s]
3|4217|Loss: 0.876923680305481:  65%|██████▌   | 4216/6470 [48:30<22:56,  1.64it/s] 
3|4217|Loss: 0.876923680305481:  65%|██████▌   | 4217/6470 [48:30<23:59,  1.56it/s]
3|4218|Loss: 0.8422626852989197:  65%|██████▌   | 4217/6470 [48:31<23:59,  1.56it/s]
3|4218|Loss: 0.8422626852989197:  65%|██████▌   | 4218/6470 [48:31<25:56,  1.45it/s]
3|4219|Loss: 0.9728097319602966:  65%|██████▌   | 4218/6470 [48:32<25:56,  1.45it/s]
3|4219|Loss: 0.9728097319602966:  65%|██████▌   | 4219/6470 [48:32<25:38,  1.46it/s]
3|4220|Loss: 0.8027770519256592:  65%|██████▌   | 4219/6470 [48:32<25:38,  1.46it/s]
3|4220|Loss: 0.8027770519256592:  65%|██████▌   | 4220/6470 [48:33<25:55,  1.45it/s]
3|4221|Loss: 0.8524362444877625:  65%|██████▌   | 4220/6470 [48:33<25:55,  1.45it/s]
3|4221|Loss: 0.8524362444877625:  65%|██████▌   | 4221/6470 [48:33<27:18,  1.37it/s]
3|4222|Loss: 0.9577473402023315:  65%|██████▌   | 4221/6470 [48:34<27:18,  1.37it/s]
3|4222|Loss: 0.9577473402023315:  65%|██████▌   | 4222/6470 [48:34<26:33,  1.41it/s]
3|4223|Loss: 0.9929896593093872:  65%|██████▌   | 4222/6470 [48:34<26:33,  1.41it/s]
3|4223|Loss: 0.9929896593093872:  65%|██████▌   | 4223/6470 [48:35<25:41,  1.46it/s]
3|4224|Loss: 0.864051103591919:  65%|██████▌   | 4223/6470 [48:35<25:41,  1.46it/s] 
3|4224|Loss: 0.864051103591919:  65%|██████▌   | 4224/6470 [48:35<24:53,  1.50it/s]
3|4225|Loss: 0.9008962512016296:  65%|██████▌   | 4224/6470 [48:36<24:53,  1.50it/s]
3|4225|Loss: 0.9008962512016296:  65%|██████▌   | 4225/6470 [48:36<25:10,  1.49it/s]
3|4226|Loss: 0.7829601168632507:  65%|██████▌   | 4225/6470 [48:36<25:10,  1.49it/s]
3|4226|Loss: 0.7829601168632507:  65%|██████▌   | 4226/6470 [48:37<25:46,  1.45it/s]
3|4227|Loss: 0.814409613609314:  65%|██████▌   | 4226/6470 [48:37<25:46,  1.45it/s] 
3|4227|Loss: 0.814409613609314:  65%|██████▌   | 4227/6470 [48:37<24:21,  1.53it/s]
3|4228|Loss: 0.6833057403564453:  65%|██████▌   | 4227/6470 [48:38<24:21,  1.53it/s]
3|4228|Loss: 0.6833057403564453:  65%|██████▌   | 4228/6470 [48:38<24:29,  1.53it/s]
3|4229|Loss: 0.8094322085380554:  65%|██████▌   | 4228/6470 [48:38<24:29,  1.53it/s]
3|4229|Loss: 0.8094322085380554:  65%|██████▌   | 4229/6470 [48:39<23:55,  1.56it/s]
3|4230|Loss: 0.769610583782196:  65%|██████▌   | 4229/6470 [48:39<23:55,  1.56it/s] 
3|4230|Loss: 0.769610583782196:  65%|██████▌   | 4230/6470 [48:39<24:44,  1.51it/s]
3|4231|Loss: 0.9488383531570435:  65%|██████▌   | 4230/6470 [48:40<24:44,  1.51it/s]
3|4231|Loss: 0.9488383531570435:  65%|██████▌   | 4231/6470 [48:40<24:49,  1.50it/s]
3|4232|Loss: 0.892310619354248:  65%|██████▌   | 4231/6470 [48:40<24:49,  1.50it/s] 
3|4232|Loss: 0.892310619354248:  65%|██████▌   | 4232/6470 [48:41<27:57,  1.33it/s]
3|4233|Loss: 0.9461076259613037:  65%|██████▌   | 4232/6470 [48:41<27:57,  1.33it/s]
3|4233|Loss: 0.9461076259613037:  65%|██████▌   | 4233/6470 [48:42<28:17,  1.32it/s]
3|4234|Loss: 0.7941969633102417:  65%|██████▌   | 4233/6470 [48:42<28:17,  1.32it/s]
3|4234|Loss: 0.7941969633102417:  65%|██████▌   | 4234/6470 [48:42<25:22,  1.47it/s]
3|4235|Loss: 0.7943887710571289:  65%|██████▌   | 4234/6470 [48:43<25:22,  1.47it/s]
3|4235|Loss: 0.7943887710571289:  65%|██████▌   | 4235/6470 [48:43<23:51,  1.56it/s]
3|4236|Loss: 0.840125322341919:  65%|██████▌   | 4235/6470 [48:43<23:51,  1.56it/s] 
3|4236|Loss: 0.840125322341919:  65%|██████▌   | 4236/6470 [48:43<22:32,  1.65it/s]
3|4237|Loss: 0.7218461036682129:  65%|██████▌   | 4236/6470 [48:44<22:32,  1.65it/s]
3|4237|Loss: 0.7218461036682129:  65%|██████▌   | 4237/6470 [48:44<24:19,  1.53it/s]
3|4238|Loss: 0.9306104183197021:  65%|██████▌   | 4237/6470 [48:45<24:19,  1.53it/s]
3|4238|Loss: 0.9306104183197021:  66%|██████▌   | 4238/6470 [48:45<26:33,  1.40it/s]
3|4239|Loss: 0.7631075978279114:  66%|██████▌   | 4238/6470 [48:45<26:33,  1.40it/s]
3|4239|Loss: 0.7631075978279114:  66%|██████▌   | 4239/6470 [48:46<27:19,  1.36it/s]
3|4240|Loss: 0.7508546113967896:  66%|██████▌   | 4239/6470 [48:46<27:19,  1.36it/s]
3|4240|Loss: 0.7508546113967896:  66%|██████▌   | 4240/6470 [48:46<27:54,  1.33it/s]
3|4241|Loss: 0.7712063789367676:  66%|██████▌   | 4240/6470 [48:47<27:54,  1.33it/s]
3|4241|Loss: 0.7712063789367676:  66%|██████▌   | 4241/6470 [48:47<27:32,  1.35it/s]
3|4242|Loss: 0.8478525876998901:  66%|██████▌   | 4241/6470 [48:48<27:32,  1.35it/s]
3|4242|Loss: 0.8478525876998901:  66%|██████▌   | 4242/6470 [48:48<27:55,  1.33it/s]
3|4243|Loss: 1.016309142112732:  66%|██████▌   | 4242/6470 [48:48<27:55,  1.33it/s] 
3|4243|Loss: 1.016309142112732:  66%|██████▌   | 4243/6470 [48:49<26:19,  1.41it/s]
3|4244|Loss: 0.824077308177948:  66%|██████▌   | 4243/6470 [48:49<26:19,  1.41it/s]
3|4244|Loss: 0.824077308177948:  66%|██████▌   | 4244/6470 [48:49<26:11,  1.42it/s]
3|4245|Loss: 0.9293318390846252:  66%|██████▌   | 4244/6470 [48:50<26:11,  1.42it/s]
3|4245|Loss: 0.9293318390846252:  66%|██████▌   | 4245/6470 [48:50<26:34,  1.40it/s]
3|4246|Loss: 0.8894238471984863:  66%|██████▌   | 4245/6470 [48:50<26:34,  1.40it/s]
3|4246|Loss: 0.8894238471984863:  66%|██████▌   | 4246/6470 [48:51<26:28,  1.40it/s]
3|4247|Loss: 0.7594668865203857:  66%|██████▌   | 4246/6470 [48:51<26:28,  1.40it/s]
3|4247|Loss: 0.7594668865203857:  66%|██████▌   | 4247/6470 [48:51<26:22,  1.41it/s]
3|4248|Loss: 0.8318538069725037:  66%|██████▌   | 4247/6470 [48:52<26:22,  1.41it/s]
3|4248|Loss: 0.8318538069725037:  66%|██████▌   | 4248/6470 [48:52<25:34,  1.45it/s]
3|4249|Loss: 0.8949910402297974:  66%|██████▌   | 4248/6470 [48:52<25:34,  1.45it/s]
3|4249|Loss: 0.8949910402297974:  66%|██████▌   | 4249/6470 [48:53<25:48,  1.43it/s]
3|4250|Loss: 0.8433656096458435:  66%|██████▌   | 4249/6470 [48:53<25:48,  1.43it/s]
3|4250|Loss: 0.8433656096458435:  66%|██████▌   | 4250/6470 [48:53<24:37,  1.50it/s]
3|4251|Loss: 0.708003044128418:  66%|██████▌   | 4250/6470 [48:54<24:37,  1.50it/s] 
3|4251|Loss: 0.708003044128418:  66%|██████▌   | 4251/6470 [48:54<23:59,  1.54it/s]
3|4252|Loss: 0.7359627485275269:  66%|██████▌   | 4251/6470 [48:54<23:59,  1.54it/s]
3|4252|Loss: 0.7359627485275269:  66%|██████▌   | 4252/6470 [48:55<24:36,  1.50it/s]
3|4253|Loss: 0.6853464245796204:  66%|██████▌   | 4252/6470 [48:55<24:36,  1.50it/s]
3|4253|Loss: 0.6853464245796204:  66%|██████▌   | 4253/6470 [48:55<24:16,  1.52it/s]
3|4254|Loss: 0.8951209783554077:  66%|██████▌   | 4253/6470 [48:56<24:16,  1.52it/s]
3|4254|Loss: 0.8951209783554077:  66%|██████▌   | 4254/6470 [48:56<22:39,  1.63it/s]
3|4255|Loss: 0.8697012662887573:  66%|██████▌   | 4254/6470 [48:56<22:39,  1.63it/s]
3|4255|Loss: 0.8697012662887573:  66%|██████▌   | 4255/6470 [48:56<22:35,  1.63it/s]
3|4256|Loss: 0.7506212592124939:  66%|██████▌   | 4255/6470 [48:57<22:35,  1.63it/s]
3|4256|Loss: 0.7506212592124939:  66%|██████▌   | 4256/6470 [48:57<23:46,  1.55it/s]
3|4257|Loss: 0.9756509065628052:  66%|██████▌   | 4256/6470 [48:58<23:46,  1.55it/s]
3|4257|Loss: 0.9756509065628052:  66%|██████▌   | 4257/6470 [48:58<24:33,  1.50it/s]
3|4258|Loss: 0.9114567041397095:  66%|██████▌   | 4257/6470 [48:58<24:33,  1.50it/s]
3|4258|Loss: 0.9114567041397095:  66%|██████▌   | 4258/6470 [48:59<25:02,  1.47it/s]
3|4259|Loss: 0.8660328984260559:  66%|██████▌   | 4258/6470 [48:59<25:02,  1.47it/s]
3|4259|Loss: 0.8660328984260559:  66%|██████▌   | 4259/6470 [48:59<25:45,  1.43it/s]
3|4260|Loss: 0.8181567192077637:  66%|██████▌   | 4259/6470 [49:00<25:45,  1.43it/s]
3|4260|Loss: 0.8181567192077637:  66%|██████▌   | 4260/6470 [49:00<27:30,  1.34it/s]
3|4261|Loss: 0.7485014200210571:  66%|██████▌   | 4260/6470 [49:01<27:30,  1.34it/s]
3|4261|Loss: 0.7485014200210571:  66%|██████▌   | 4261/6470 [49:01<28:34,  1.29it/s]
3|4262|Loss: 0.852325439453125:  66%|██████▌   | 4261/6470 [49:01<28:34,  1.29it/s] 
3|4262|Loss: 0.852325439453125:  66%|██████▌   | 4262/6470 [49:02<26:45,  1.38it/s]
3|4263|Loss: 0.8345401883125305:  66%|██████▌   | 4262/6470 [49:02<26:45,  1.38it/s]
3|4263|Loss: 0.8345401883125305:  66%|██████▌   | 4263/6470 [49:02<27:20,  1.35it/s]
3|4264|Loss: 0.8352822661399841:  66%|██████▌   | 4263/6470 [49:03<27:20,  1.35it/s]
3|4264|Loss: 0.8352822661399841:  66%|██████▌   | 4264/6470 [49:03<27:25,  1.34it/s]
3|4265|Loss: 0.857020914554596:  66%|██████▌   | 4264/6470 [49:04<27:25,  1.34it/s] 
3|4265|Loss: 0.857020914554596:  66%|██████▌   | 4265/6470 [49:04<25:56,  1.42it/s]
3|4266|Loss: 0.9734939932823181:  66%|██████▌   | 4265/6470 [49:04<25:56,  1.42it/s]
3|4266|Loss: 0.9734939932823181:  66%|██████▌   | 4266/6470 [49:04<24:07,  1.52it/s]
3|4267|Loss: 0.7806150317192078:  66%|██████▌   | 4266/6470 [49:05<24:07,  1.52it/s]
3|4267|Loss: 0.7806150317192078:  66%|██████▌   | 4267/6470 [49:05<22:38,  1.62it/s]
3|4268|Loss: 0.7916054725646973:  66%|██████▌   | 4267/6470 [49:05<22:38,  1.62it/s]
3|4268|Loss: 0.7916054725646973:  66%|██████▌   | 4268/6470 [49:06<23:32,  1.56it/s]
3|4269|Loss: 1.0007414817810059:  66%|██████▌   | 4268/6470 [49:06<23:32,  1.56it/s]
3|4269|Loss: 1.0007414817810059:  66%|██████▌   | 4269/6470 [49:06<22:07,  1.66it/s]
3|4270|Loss: 0.848447859287262:  66%|██████▌   | 4269/6470 [49:06<22:07,  1.66it/s] 
3|4270|Loss: 0.848447859287262:  66%|██████▌   | 4270/6470 [49:07<22:52,  1.60it/s]
3|4271|Loss: 0.9728425741195679:  66%|██████▌   | 4270/6470 [49:07<22:52,  1.60it/s]
3|4271|Loss: 0.9728425741195679:  66%|██████▌   | 4271/6470 [49:08<25:04,  1.46it/s]
3|4272|Loss: 0.9265515804290771:  66%|██████▌   | 4271/6470 [49:08<25:04,  1.46it/s]
3|4272|Loss: 0.9265515804290771:  66%|██████▌   | 4272/6470 [49:08<25:59,  1.41it/s]
3|4273|Loss: 0.8624732494354248:  66%|██████▌   | 4272/6470 [49:09<25:59,  1.41it/s]
3|4273|Loss: 0.8624732494354248:  66%|██████▌   | 4273/6470 [49:09<26:31,  1.38it/s]
3|4274|Loss: 0.7984223961830139:  66%|██████▌   | 4273/6470 [49:10<26:31,  1.38it/s]
3|4274|Loss: 0.7984223961830139:  66%|██████▌   | 4274/6470 [49:10<25:52,  1.41it/s]
3|4275|Loss: 0.8788523077964783:  66%|██████▌   | 4274/6470 [49:10<25:52,  1.41it/s]
3|4275|Loss: 0.8788523077964783:  66%|██████▌   | 4275/6470 [49:11<26:50,  1.36it/s]
3|4276|Loss: 0.8675994873046875:  66%|██████▌   | 4275/6470 [49:11<26:50,  1.36it/s]
3|4276|Loss: 0.8675994873046875:  66%|██████▌   | 4276/6470 [49:11<27:05,  1.35it/s]
3|4277|Loss: 0.8719868063926697:  66%|██████▌   | 4276/6470 [49:12<27:05,  1.35it/s]
3|4277|Loss: 0.8719868063926697:  66%|██████▌   | 4277/6470 [49:12<27:03,  1.35it/s]
3|4278|Loss: 0.7174988985061646:  66%|██████▌   | 4277/6470 [49:13<27:03,  1.35it/s]
3|4278|Loss: 0.7174988985061646:  66%|██████▌   | 4278/6470 [49:13<27:36,  1.32it/s]
3|4279|Loss: 0.8171895146369934:  66%|██████▌   | 4278/6470 [49:13<27:36,  1.32it/s]
3|4279|Loss: 0.8171895146369934:  66%|██████▌   | 4279/6470 [49:14<27:38,  1.32it/s]
3|4280|Loss: 0.8953747153282166:  66%|██████▌   | 4279/6470 [49:14<27:38,  1.32it/s]
3|4280|Loss: 0.8953747153282166:  66%|██████▌   | 4280/6470 [49:14<27:34,  1.32it/s]
3|4281|Loss: 0.7762015461921692:  66%|██████▌   | 4280/6470 [49:15<27:34,  1.32it/s]
3|4281|Loss: 0.7762015461921692:  66%|██████▌   | 4281/6470 [49:15<27:08,  1.34it/s]
3|4282|Loss: 0.9219771027565002:  66%|██████▌   | 4281/6470 [49:15<27:08,  1.34it/s]
3|4282|Loss: 0.9219771027565002:  66%|██████▌   | 4282/6470 [49:16<26:01,  1.40it/s]
3|4283|Loss: 1.210561752319336:  66%|██████▌   | 4282/6470 [49:16<26:01,  1.40it/s] 
3|4283|Loss: 1.210561752319336:  66%|██████▌   | 4283/6470 [49:16<25:58,  1.40it/s]
3|4284|Loss: 0.9227006435394287:  66%|██████▌   | 4283/6470 [49:17<25:58,  1.40it/s]
3|4284|Loss: 0.9227006435394287:  66%|██████▌   | 4284/6470 [49:17<26:14,  1.39it/s]
3|4285|Loss: 0.7783271074295044:  66%|██████▌   | 4284/6470 [49:18<26:14,  1.39it/s]
3|4285|Loss: 0.7783271074295044:  66%|██████▌   | 4285/6470 [49:18<25:04,  1.45it/s]
3|4286|Loss: 1.0158371925354004:  66%|██████▌   | 4285/6470 [49:18<25:04,  1.45it/s]
3|4286|Loss: 1.0158371925354004:  66%|██████▌   | 4286/6470 [49:18<24:12,  1.50it/s]
3|4287|Loss: 0.8127279281616211:  66%|██████▌   | 4286/6470 [49:19<24:12,  1.50it/s]
3|4287|Loss: 0.8127279281616211:  66%|██████▋   | 4287/6470 [49:19<23:39,  1.54it/s]
3|4288|Loss: 0.6755099892616272:  66%|██████▋   | 4287/6470 [49:19<23:39,  1.54it/s]
3|4288|Loss: 0.6755099892616272:  66%|██████▋   | 4288/6470 [49:20<25:50,  1.41it/s]
3|4289|Loss: 0.8674789667129517:  66%|██████▋   | 4288/6470 [49:20<25:50,  1.41it/s]
3|4289|Loss: 0.8674789667129517:  66%|██████▋   | 4289/6470 [49:21<25:41,  1.41it/s]
3|4290|Loss: 1.1579452753067017:  66%|██████▋   | 4289/6470 [49:21<25:41,  1.41it/s]
3|4290|Loss: 1.1579452753067017:  66%|██████▋   | 4290/6470 [49:21<26:22,  1.38it/s]
3|4291|Loss: 0.8143476843833923:  66%|██████▋   | 4290/6470 [49:22<26:22,  1.38it/s]
3|4291|Loss: 0.8143476843833923:  66%|██████▋   | 4291/6470 [49:22<27:10,  1.34it/s]
3|4292|Loss: 0.8701115250587463:  66%|██████▋   | 4291/6470 [49:23<27:10,  1.34it/s]
3|4292|Loss: 0.8701115250587463:  66%|██████▋   | 4292/6470 [49:23<24:59,  1.45it/s]
3|4293|Loss: 0.8303143382072449:  66%|██████▋   | 4292/6470 [49:23<24:59,  1.45it/s]
3|4293|Loss: 0.8303143382072449:  66%|██████▋   | 4293/6470 [49:23<22:27,  1.62it/s]
3|4294|Loss: 0.6822763681411743:  66%|██████▋   | 4293/6470 [49:23<22:27,  1.62it/s]
3|4294|Loss: 0.6822763681411743:  66%|██████▋   | 4294/6470 [49:24<22:14,  1.63it/s]
3|4295|Loss: 0.8735689520835876:  66%|██████▋   | 4294/6470 [49:24<22:14,  1.63it/s]
3|4295|Loss: 0.8735689520835876:  66%|██████▋   | 4295/6470 [49:24<23:05,  1.57it/s]
3|4296|Loss: 0.9033992886543274:  66%|██████▋   | 4295/6470 [49:25<23:05,  1.57it/s]
3|4296|Loss: 0.9033992886543274:  66%|██████▋   | 4296/6470 [49:25<24:10,  1.50it/s]
3|4297|Loss: 0.9173545241355896:  66%|██████▋   | 4296/6470 [49:26<24:10,  1.50it/s]
3|4297|Loss: 0.9173545241355896:  66%|██████▋   | 4297/6470 [49:26<24:38,  1.47it/s]
3|4298|Loss: 0.7018571496009827:  66%|██████▋   | 4297/6470 [49:26<24:38,  1.47it/s]
3|4298|Loss: 0.7018571496009827:  66%|██████▋   | 4298/6470 [49:26<22:53,  1.58it/s]
3|4299|Loss: 0.9073534607887268:  66%|██████▋   | 4298/6470 [49:27<22:53,  1.58it/s]
3|4299|Loss: 0.9073534607887268:  66%|██████▋   | 4299/6470 [49:27<21:47,  1.66it/s]
3|4300|Loss: 0.926893949508667:  66%|██████▋   | 4299/6470 [49:27<21:47,  1.66it/s] 
3|4300|Loss: 0.926893949508667:  66%|██████▋   | 4300/6470 [49:27<20:53,  1.73it/s]
3|4301|Loss: 0.8072099685668945:  66%|██████▋   | 4300/6470 [49:28<20:53,  1.73it/s]
3|4301|Loss: 0.8072099685668945:  66%|██████▋   | 4301/6470 [49:28<22:05,  1.64it/s]
3|4302|Loss: 0.945766806602478:  66%|██████▋   | 4301/6470 [49:29<22:05,  1.64it/s] 
3|4302|Loss: 0.945766806602478:  66%|██████▋   | 4302/6470 [49:29<24:27,  1.48it/s]
3|4303|Loss: 0.9309098720550537:  66%|██████▋   | 4302/6470 [49:29<24:27,  1.48it/s]
3|4303|Loss: 0.9309098720550537:  67%|██████▋   | 4303/6470 [49:30<25:04,  1.44it/s]
3|4304|Loss: 0.6793826818466187:  67%|██████▋   | 4303/6470 [49:30<25:04,  1.44it/s]
3|4304|Loss: 0.6793826818466187:  67%|██████▋   | 4304/6470 [49:30<25:51,  1.40it/s]
3|4305|Loss: 0.8908534049987793:  67%|██████▋   | 4304/6470 [49:31<25:51,  1.40it/s]
3|4305|Loss: 0.8908534049987793:  67%|██████▋   | 4305/6470 [49:31<26:18,  1.37it/s]
3|4306|Loss: 0.8225371241569519:  67%|██████▋   | 4305/6470 [49:32<26:18,  1.37it/s]
3|4306|Loss: 0.8225371241569519:  67%|██████▋   | 4306/6470 [49:32<25:04,  1.44it/s]
3|4307|Loss: 0.7656723856925964:  67%|██████▋   | 4306/6470 [49:32<25:04,  1.44it/s]
3|4307|Loss: 0.7656723856925964:  67%|██████▋   | 4307/6470 [49:32<24:08,  1.49it/s]
3|4308|Loss: 0.816699743270874:  67%|██████▋   | 4307/6470 [49:33<24:08,  1.49it/s] 
3|4308|Loss: 0.816699743270874:  67%|██████▋   | 4308/6470 [49:33<24:19,  1.48it/s]
3|4309|Loss: 0.9164993762969971:  67%|██████▋   | 4308/6470 [49:34<24:19,  1.48it/s]
3|4309|Loss: 0.9164993762969971:  67%|██████▋   | 4309/6470 [49:34<22:42,  1.59it/s]
3|4310|Loss: 0.7543666362762451:  67%|██████▋   | 4309/6470 [49:34<22:42,  1.59it/s]
3|4310|Loss: 0.7543666362762451:  67%|██████▋   | 4310/6470 [49:34<24:18,  1.48it/s]
3|4311|Loss: 0.7966083884239197:  67%|██████▋   | 4310/6470 [49:35<24:18,  1.48it/s]
3|4311|Loss: 0.7966083884239197:  67%|██████▋   | 4311/6470 [49:35<25:16,  1.42it/s]
3|4312|Loss: 0.8487294912338257:  67%|██████▋   | 4311/6470 [49:36<25:16,  1.42it/s]
3|4312|Loss: 0.8487294912338257:  67%|██████▋   | 4312/6470 [49:36<23:38,  1.52it/s]
3|4313|Loss: 0.69536954164505:  67%|██████▋   | 4312/6470 [49:36<23:38,  1.52it/s]  
3|4313|Loss: 0.69536954164505:  67%|██████▋   | 4313/6470 [49:37<24:25,  1.47it/s]
3|4314|Loss: 0.7250085473060608:  67%|██████▋   | 4313/6470 [49:37<24:25,  1.47it/s]
3|4314|Loss: 0.7250085473060608:  67%|██████▋   | 4314/6470 [49:37<26:00,  1.38it/s]
3|4315|Loss: 0.7312406301498413:  67%|██████▋   | 4314/6470 [49:38<26:00,  1.38it/s]
3|4315|Loss: 0.7312406301498413:  67%|██████▋   | 4315/6470 [49:38<25:00,  1.44it/s]
3|4316|Loss: 0.7610325813293457:  67%|██████▋   | 4315/6470 [49:38<25:00,  1.44it/s]
3|4316|Loss: 0.7610325813293457:  67%|██████▋   | 4316/6470 [49:39<24:47,  1.45it/s]
3|4317|Loss: 1.008573055267334:  67%|██████▋   | 4316/6470 [49:39<24:47,  1.45it/s] 
3|4317|Loss: 1.008573055267334:  67%|██████▋   | 4317/6470 [49:39<25:44,  1.39it/s]
3|4318|Loss: 0.8257875442504883:  67%|██████▋   | 4317/6470 [49:40<25:44,  1.39it/s]
3|4318|Loss: 0.8257875442504883:  67%|██████▋   | 4318/6470 [49:40<26:12,  1.37it/s]
3|4319|Loss: 0.7360857725143433:  67%|██████▋   | 4318/6470 [49:41<26:12,  1.37it/s]
3|4319|Loss: 0.7360857725143433:  67%|██████▋   | 4319/6470 [49:41<24:29,  1.46it/s]
3|4320|Loss: 0.9355405569076538:  67%|██████▋   | 4319/6470 [49:41<24:29,  1.46it/s]
3|4320|Loss: 0.9355405569076538:  67%|██████▋   | 4320/6470 [49:41<23:21,  1.53it/s]
3|4321|Loss: 0.9017580151557922:  67%|██████▋   | 4320/6470 [49:42<23:21,  1.53it/s]
3|4321|Loss: 0.9017580151557922:  67%|██████▋   | 4321/6470 [49:42<25:27,  1.41it/s]
3|4322|Loss: 0.6336872577667236:  67%|██████▋   | 4321/6470 [49:43<25:27,  1.41it/s]
3|4322|Loss: 0.6336872577667236:  67%|██████▋   | 4322/6470 [49:43<26:35,  1.35it/s]
3|4323|Loss: 0.8219755291938782:  67%|██████▋   | 4322/6470 [49:43<26:35,  1.35it/s]
3|4323|Loss: 0.8219755291938782:  67%|██████▋   | 4323/6470 [49:44<25:52,  1.38it/s]
3|4324|Loss: 1.0511832237243652:  67%|██████▋   | 4323/6470 [49:44<25:52,  1.38it/s]
3|4324|Loss: 1.0511832237243652:  67%|██████▋   | 4324/6470 [49:44<26:36,  1.34it/s]
3|4325|Loss: 0.9144800305366516:  67%|██████▋   | 4324/6470 [49:45<26:36,  1.34it/s]
3|4325|Loss: 0.9144800305366516:  67%|██████▋   | 4325/6470 [49:45<24:46,  1.44it/s]
3|4326|Loss: 0.8212078809738159:  67%|██████▋   | 4325/6470 [49:45<24:46,  1.44it/s]
3|4326|Loss: 0.8212078809738159:  67%|██████▋   | 4326/6470 [49:46<24:50,  1.44it/s]
3|4327|Loss: 0.8149762153625488:  67%|██████▋   | 4326/6470 [49:46<24:50,  1.44it/s]
3|4327|Loss: 0.8149762153625488:  67%|██████▋   | 4327/6470 [49:46<24:17,  1.47it/s]
3|4328|Loss: 1.0002309083938599:  67%|██████▋   | 4327/6470 [49:47<24:17,  1.47it/s]
3|4328|Loss: 1.0002309083938599:  67%|██████▋   | 4328/6470 [49:47<24:39,  1.45it/s]
3|4329|Loss: 0.9647908210754395:  67%|██████▋   | 4328/6470 [49:48<24:39,  1.45it/s]
3|4329|Loss: 0.9647908210754395:  67%|██████▋   | 4329/6470 [49:48<24:25,  1.46it/s]
3|4330|Loss: 0.8267464637756348:  67%|██████▋   | 4329/6470 [49:48<24:25,  1.46it/s]
3|4330|Loss: 0.8267464637756348:  67%|██████▋   | 4330/6470 [49:48<23:59,  1.49it/s]
3|4331|Loss: 0.8209522366523743:  67%|██████▋   | 4330/6470 [49:49<23:59,  1.49it/s]
3|4331|Loss: 0.8209522366523743:  67%|██████▋   | 4331/6470 [49:49<24:08,  1.48it/s]
3|4332|Loss: 0.7587580680847168:  67%|██████▋   | 4331/6470 [49:49<24:08,  1.48it/s]
3|4332|Loss: 0.7587580680847168:  67%|██████▋   | 4332/6470 [49:50<23:51,  1.49it/s]
3|4333|Loss: 0.8293451070785522:  67%|██████▋   | 4332/6470 [49:50<23:51,  1.49it/s]
3|4333|Loss: 0.8293451070785522:  67%|██████▋   | 4333/6470 [49:50<24:03,  1.48it/s]
3|4334|Loss: 0.6406849026679993:  67%|██████▋   | 4333/6470 [49:51<24:03,  1.48it/s]
3|4334|Loss: 0.6406849026679993:  67%|██████▋   | 4334/6470 [49:51<22:09,  1.61it/s]
3|4335|Loss: 0.8454365134239197:  67%|██████▋   | 4334/6470 [49:51<22:09,  1.61it/s]
3|4335|Loss: 0.8454365134239197:  67%|██████▋   | 4335/6470 [49:52<22:27,  1.58it/s]
3|4336|Loss: 0.893855094909668:  67%|██████▋   | 4335/6470 [49:52<22:27,  1.58it/s] 
3|4336|Loss: 0.893855094909668:  67%|██████▋   | 4336/6470 [49:52<23:28,  1.52it/s]
3|4337|Loss: 0.8047698736190796:  67%|██████▋   | 4336/6470 [49:53<23:28,  1.52it/s]
3|4337|Loss: 0.8047698736190796:  67%|██████▋   | 4337/6470 [49:53<25:49,  1.38it/s]
3|4338|Loss: 0.7865447998046875:  67%|██████▋   | 4337/6470 [49:54<25:49,  1.38it/s]
3|4338|Loss: 0.7865447998046875:  67%|██████▋   | 4338/6470 [49:54<26:02,  1.36it/s]
3|4339|Loss: 0.8705713152885437:  67%|██████▋   | 4338/6470 [49:54<26:02,  1.36it/s]
3|4339|Loss: 0.8705713152885437:  67%|██████▋   | 4339/6470 [49:55<25:56,  1.37it/s]
3|4340|Loss: 0.8453934788703918:  67%|██████▋   | 4339/6470 [49:55<25:56,  1.37it/s]
3|4340|Loss: 0.8453934788703918:  67%|██████▋   | 4340/6470 [49:55<24:33,  1.45it/s]
3|4341|Loss: 0.8137049078941345:  67%|██████▋   | 4340/6470 [49:56<24:33,  1.45it/s]
3|4341|Loss: 0.8137049078941345:  67%|██████▋   | 4341/6470 [49:56<23:52,  1.49it/s]
3|4342|Loss: 0.8661952018737793:  67%|██████▋   | 4341/6470 [49:56<23:52,  1.49it/s]
3|4342|Loss: 0.8661952018737793:  67%|██████▋   | 4342/6470 [49:57<24:28,  1.45it/s]
3|4343|Loss: 0.8882982730865479:  67%|██████▋   | 4342/6470 [49:57<24:28,  1.45it/s]
3|4343|Loss: 0.8882982730865479:  67%|██████▋   | 4343/6470 [49:57<24:15,  1.46it/s]
3|4344|Loss: 0.9487534761428833:  67%|██████▋   | 4343/6470 [49:58<24:15,  1.46it/s]
3|4344|Loss: 0.9487534761428833:  67%|██████▋   | 4344/6470 [49:58<24:33,  1.44it/s]
3|4345|Loss: 0.7950737476348877:  67%|██████▋   | 4344/6470 [49:58<24:33,  1.44it/s]
3|4345|Loss: 0.7950737476348877:  67%|██████▋   | 4345/6470 [49:59<24:16,  1.46it/s]
3|4346|Loss: 0.8683839440345764:  67%|██████▋   | 4345/6470 [49:59<24:16,  1.46it/s]
3|4346|Loss: 0.8683839440345764:  67%|██████▋   | 4346/6470 [49:59<22:59,  1.54it/s]
3|4347|Loss: 0.8213741779327393:  67%|██████▋   | 4346/6470 [50:00<22:59,  1.54it/s]
3|4347|Loss: 0.8213741779327393:  67%|██████▋   | 4347/6470 [50:00<23:39,  1.50it/s]
3|4348|Loss: 0.8339758515357971:  67%|██████▋   | 4347/6470 [50:00<23:39,  1.50it/s]
3|4348|Loss: 0.8339758515357971:  67%|██████▋   | 4348/6470 [50:01<23:27,  1.51it/s]
3|4349|Loss: 0.8689892888069153:  67%|██████▋   | 4348/6470 [50:01<23:27,  1.51it/s]
3|4349|Loss: 0.8689892888069153:  67%|██████▋   | 4349/6470 [50:01<25:02,  1.41it/s]
3|4350|Loss: 0.8982746005058289:  67%|██████▋   | 4349/6470 [50:02<25:02,  1.41it/s]
3|4350|Loss: 0.8982746005058289:  67%|██████▋   | 4350/6470 [50:02<26:15,  1.35it/s]
3|4351|Loss: 0.8096058368682861:  67%|██████▋   | 4350/6470 [50:03<26:15,  1.35it/s]
3|4351|Loss: 0.8096058368682861:  67%|██████▋   | 4351/6470 [50:03<26:51,  1.31it/s]
3|4352|Loss: 0.8650267124176025:  67%|██████▋   | 4351/6470 [50:04<26:51,  1.31it/s]
3|4352|Loss: 0.8650267124176025:  67%|██████▋   | 4352/6470 [50:04<27:12,  1.30it/s]
3|4353|Loss: 0.9274953603744507:  67%|██████▋   | 4352/6470 [50:04<27:12,  1.30it/s]
3|4353|Loss: 0.9274953603744507:  67%|██████▋   | 4353/6470 [50:05<26:43,  1.32it/s]
3|4354|Loss: 0.8038944005966187:  67%|██████▋   | 4353/6470 [50:05<26:43,  1.32it/s]
3|4354|Loss: 0.8038944005966187:  67%|██████▋   | 4354/6470 [50:05<25:28,  1.38it/s]
3|4355|Loss: 0.8057517409324646:  67%|██████▋   | 4354/6470 [50:06<25:28,  1.38it/s]
3|4355|Loss: 0.8057517409324646:  67%|██████▋   | 4355/6470 [50:06<24:06,  1.46it/s]
3|4356|Loss: 0.8211719393730164:  67%|██████▋   | 4355/6470 [50:06<24:06,  1.46it/s]
3|4356|Loss: 0.8211719393730164:  67%|██████▋   | 4356/6470 [50:06<23:01,  1.53it/s]
3|4357|Loss: 0.8734757304191589:  67%|██████▋   | 4356/6470 [50:07<23:01,  1.53it/s]
3|4357|Loss: 0.8734757304191589:  67%|██████▋   | 4357/6470 [50:07<23:40,  1.49it/s]
3|4358|Loss: 0.7277418971061707:  67%|██████▋   | 4357/6470 [50:08<23:40,  1.49it/s]
3|4358|Loss: 0.7277418971061707:  67%|██████▋   | 4358/6470 [50:08<23:22,  1.51it/s]
3|4359|Loss: 0.818412184715271:  67%|██████▋   | 4358/6470 [50:08<23:22,  1.51it/s] 
3|4359|Loss: 0.818412184715271:  67%|██████▋   | 4359/6470 [50:08<21:55,  1.61it/s]
3|4360|Loss: 0.8099098205566406:  67%|██████▋   | 4359/6470 [50:09<21:55,  1.61it/s]
3|4360|Loss: 0.8099098205566406:  67%|██████▋   | 4360/6470 [50:09<20:41,  1.70it/s]
3|4361|Loss: 0.8860360980033875:  67%|██████▋   | 4360/6470 [50:09<20:41,  1.70it/s]
3|4361|Loss: 0.8860360980033875:  67%|██████▋   | 4361/6470 [50:09<20:11,  1.74it/s]
3|4362|Loss: 0.8581699132919312:  67%|██████▋   | 4361/6470 [50:10<20:11,  1.74it/s]
3|4362|Loss: 0.8581699132919312:  67%|██████▋   | 4362/6470 [50:10<21:24,  1.64it/s]
3|4363|Loss: 0.9391926527023315:  67%|██████▋   | 4362/6470 [50:10<21:24,  1.64it/s]
3|4363|Loss: 0.9391926527023315:  67%|██████▋   | 4363/6470 [50:11<21:54,  1.60it/s]
3|4364|Loss: 0.7379046082496643:  67%|██████▋   | 4363/6470 [50:11<21:54,  1.60it/s]
3|4364|Loss: 0.7379046082496643:  67%|██████▋   | 4364/6470 [50:11<22:50,  1.54it/s]
3|4365|Loss: 0.7515829801559448:  67%|██████▋   | 4364/6470 [50:12<22:50,  1.54it/s]
3|4365|Loss: 0.7515829801559448:  67%|██████▋   | 4365/6470 [50:12<23:46,  1.48it/s]
3|4366|Loss: 0.8767532110214233:  67%|██████▋   | 4365/6470 [50:13<23:46,  1.48it/s]
3|4366|Loss: 0.8767532110214233:  67%|██████▋   | 4366/6470 [50:13<24:52,  1.41it/s]
3|4367|Loss: 0.9666118025779724:  67%|██████▋   | 4366/6470 [50:13<24:52,  1.41it/s]
3|4367|Loss: 0.9666118025779724:  67%|██████▋   | 4367/6470 [50:14<23:32,  1.49it/s]
3|4368|Loss: 1.0325706005096436:  67%|██████▋   | 4367/6470 [50:14<23:32,  1.49it/s]
3|4368|Loss: 1.0325706005096436:  68%|██████▊   | 4368/6470 [50:14<24:27,  1.43it/s]
3|4369|Loss: 0.8326202034950256:  68%|██████▊   | 4368/6470 [50:15<24:27,  1.43it/s]
3|4369|Loss: 0.8326202034950256:  68%|██████▊   | 4369/6470 [50:15<25:50,  1.36it/s]
3|4370|Loss: 0.7607036828994751:  68%|██████▊   | 4369/6470 [50:16<25:50,  1.36it/s]
3|4370|Loss: 0.7607036828994751:  68%|██████▊   | 4370/6470 [50:16<25:46,  1.36it/s]
3|4371|Loss: 0.8506871461868286:  68%|██████▊   | 4370/6470 [50:16<25:46,  1.36it/s]
3|4371|Loss: 0.8506871461868286:  68%|██████▊   | 4371/6470 [50:17<26:13,  1.33it/s]
3|4372|Loss: 0.9304288029670715:  68%|██████▊   | 4371/6470 [50:17<26:13,  1.33it/s]
3|4372|Loss: 0.9304288029670715:  68%|██████▊   | 4372/6470 [50:17<26:12,  1.33it/s]
3|4373|Loss: 0.8275145292282104:  68%|██████▊   | 4372/6470 [50:18<26:12,  1.33it/s]
3|4373|Loss: 0.8275145292282104:  68%|██████▊   | 4373/6470 [50:18<26:08,  1.34it/s]
3|4374|Loss: 0.6607998013496399:  68%|██████▊   | 4373/6470 [50:19<26:08,  1.34it/s]
3|4374|Loss: 0.6607998013496399:  68%|██████▊   | 4374/6470 [50:19<25:49,  1.35it/s]
3|4375|Loss: 0.7587937116622925:  68%|██████▊   | 4374/6470 [50:19<25:49,  1.35it/s]
3|4375|Loss: 0.7587937116622925:  68%|██████▊   | 4375/6470 [50:19<23:19,  1.50it/s]
3|4376|Loss: 0.8413324952125549:  68%|██████▊   | 4375/6470 [50:20<23:19,  1.50it/s]
3|4376|Loss: 0.8413324952125549:  68%|██████▊   | 4376/6470 [50:20<22:07,  1.58it/s]
3|4377|Loss: 0.7859525084495544:  68%|██████▊   | 4376/6470 [50:20<22:07,  1.58it/s]
3|4377|Loss: 0.7859525084495544:  68%|██████▊   | 4377/6470 [50:21<23:51,  1.46it/s]
3|4378|Loss: 0.8376875519752502:  68%|██████▊   | 4377/6470 [50:21<23:51,  1.46it/s]
3|4378|Loss: 0.8376875519752502:  68%|██████▊   | 4378/6470 [50:21<22:22,  1.56it/s]
3|4379|Loss: 0.8629238605499268:  68%|██████▊   | 4378/6470 [50:22<22:22,  1.56it/s]
3|4379|Loss: 0.8629238605499268:  68%|██████▊   | 4379/6470 [50:22<21:56,  1.59it/s]
3|4380|Loss: 0.901530385017395:  68%|██████▊   | 4379/6470 [50:22<21:56,  1.59it/s] 
3|4380|Loss: 0.901530385017395:  68%|██████▊   | 4380/6470 [50:22<22:06,  1.58it/s]
3|4381|Loss: 0.8251880407333374:  68%|██████▊   | 4380/6470 [50:23<22:06,  1.58it/s]
3|4381|Loss: 0.8251880407333374:  68%|██████▊   | 4381/6470 [50:23<20:36,  1.69it/s]
3|4382|Loss: 0.892918050289154:  68%|██████▊   | 4381/6470 [50:23<20:36,  1.69it/s] 
3|4382|Loss: 0.892918050289154:  68%|██████▊   | 4382/6470 [50:24<20:09,  1.73it/s]
3|4383|Loss: 0.8068866729736328:  68%|██████▊   | 4382/6470 [50:24<20:09,  1.73it/s]
3|4383|Loss: 0.8068866729736328:  68%|██████▊   | 4383/6470 [50:24<20:30,  1.70it/s]
3|4384|Loss: 0.8947396874427795:  68%|██████▊   | 4383/6470 [50:25<20:30,  1.70it/s]
3|4384|Loss: 0.8947396874427795:  68%|██████▊   | 4384/6470 [50:25<22:22,  1.55it/s]
3|4385|Loss: 0.7379641532897949:  68%|██████▊   | 4384/6470 [50:25<22:22,  1.55it/s]
3|4385|Loss: 0.7379641532897949:  68%|██████▊   | 4385/6470 [50:26<23:11,  1.50it/s]
3|4386|Loss: 0.8582692742347717:  68%|██████▊   | 4385/6470 [50:26<23:11,  1.50it/s]
3|4386|Loss: 0.8582692742347717:  68%|██████▊   | 4386/6470 [50:26<21:54,  1.59it/s]
3|4387|Loss: 0.8882591128349304:  68%|██████▊   | 4386/6470 [50:27<21:54,  1.59it/s]
3|4387|Loss: 0.8882591128349304:  68%|██████▊   | 4387/6470 [50:27<23:00,  1.51it/s]
3|4388|Loss: 1.0046809911727905:  68%|██████▊   | 4387/6470 [50:27<23:00,  1.51it/s]
3|4388|Loss: 1.0046809911727905:  68%|██████▊   | 4388/6470 [50:28<23:39,  1.47it/s]
3|4389|Loss: 0.8627837896347046:  68%|██████▊   | 4388/6470 [50:28<23:39,  1.47it/s]
3|4389|Loss: 0.8627837896347046:  68%|██████▊   | 4389/6470 [50:28<23:24,  1.48it/s]
3|4390|Loss: 0.8569216728210449:  68%|██████▊   | 4389/6470 [50:29<23:24,  1.48it/s]
3|4390|Loss: 0.8569216728210449:  68%|██████▊   | 4390/6470 [50:29<23:04,  1.50it/s]
3|4391|Loss: 0.9354680776596069:  68%|██████▊   | 4390/6470 [50:29<23:04,  1.50it/s]
3|4391|Loss: 0.9354680776596069:  68%|██████▊   | 4391/6470 [50:30<23:34,  1.47it/s]
3|4392|Loss: 0.8653577566146851:  68%|██████▊   | 4391/6470 [50:30<23:34,  1.47it/s]
3|4392|Loss: 0.8653577566146851:  68%|██████▊   | 4392/6470 [50:30<23:51,  1.45it/s]
3|4393|Loss: 0.833476185798645:  68%|██████▊   | 4392/6470 [50:31<23:51,  1.45it/s] 
3|4393|Loss: 0.833476185798645:  68%|██████▊   | 4393/6470 [50:31<24:51,  1.39it/s]
3|4394|Loss: 0.7596611380577087:  68%|██████▊   | 4393/6470 [50:32<24:51,  1.39it/s]
3|4394|Loss: 0.7596611380577087:  68%|██████▊   | 4394/6470 [50:32<25:15,  1.37it/s]
3|4395|Loss: 0.863877534866333:  68%|██████▊   | 4394/6470 [50:32<25:15,  1.37it/s] 
3|4395|Loss: 0.863877534866333:  68%|██████▊   | 4395/6470 [50:33<26:09,  1.32it/s]
3|4396|Loss: 0.8776075839996338:  68%|██████▊   | 4395/6470 [50:33<26:09,  1.32it/s]
3|4396|Loss: 0.8776075839996338:  68%|██████▊   | 4396/6470 [50:33<25:51,  1.34it/s]
3|4397|Loss: 0.9997445940971375:  68%|██████▊   | 4396/6470 [50:34<25:51,  1.34it/s]
3|4397|Loss: 0.9997445940971375:  68%|██████▊   | 4397/6470 [50:34<24:20,  1.42it/s]
3|4398|Loss: 0.9484513401985168:  68%|██████▊   | 4397/6470 [50:34<24:20,  1.42it/s]
3|4398|Loss: 0.9484513401985168:  68%|██████▊   | 4398/6470 [50:35<22:02,  1.57it/s]
3|4399|Loss: 0.8658663034439087:  68%|██████▊   | 4398/6470 [50:35<22:02,  1.57it/s]
3|4399|Loss: 0.8658663034439087:  68%|██████▊   | 4399/6470 [50:35<21:26,  1.61it/s]
3|4400|Loss: 0.9570028781890869:  68%|██████▊   | 4399/6470 [50:36<21:26,  1.61it/s]
3|4400|Loss: 0.9570028781890869:  68%|██████▊   | 4400/6470 [50:36<22:31,  1.53it/s]
3|4401|Loss: 0.9197152256965637:  68%|██████▊   | 4400/6470 [50:36<22:31,  1.53it/s]
3|4401|Loss: 0.9197152256965637:  68%|██████▊   | 4401/6470 [50:37<23:27,  1.47it/s]
3|4402|Loss: 0.794403612613678:  68%|██████▊   | 4401/6470 [50:37<23:27,  1.47it/s] 
3|4402|Loss: 0.794403612613678:  68%|██████▊   | 4402/6470 [50:37<22:19,  1.54it/s]
3|4403|Loss: 1.0176211595535278:  68%|██████▊   | 4402/6470 [50:38<22:19,  1.54it/s]
3|4403|Loss: 1.0176211595535278:  68%|██████▊   | 4403/6470 [50:38<22:32,  1.53it/s]
3|4404|Loss: 0.7941706776618958:  68%|██████▊   | 4403/6470 [50:38<22:32,  1.53it/s]
3|4404|Loss: 0.7941706776618958:  68%|██████▊   | 4404/6470 [50:39<23:24,  1.47it/s]
3|4405|Loss: 0.9453843235969543:  68%|██████▊   | 4404/6470 [50:39<23:24,  1.47it/s]
3|4405|Loss: 0.9453843235969543:  68%|██████▊   | 4405/6470 [50:39<23:49,  1.44it/s]
3|4406|Loss: 0.8812341690063477:  68%|██████▊   | 4405/6470 [50:40<23:49,  1.44it/s]
3|4406|Loss: 0.8812341690063477:  68%|██████▊   | 4406/6470 [50:40<23:13,  1.48it/s]
3|4407|Loss: 0.717313289642334:  68%|██████▊   | 4406/6470 [50:40<23:13,  1.48it/s] 
3|4407|Loss: 0.717313289642334:  68%|██████▊   | 4407/6470 [50:41<23:34,  1.46it/s]
3|4408|Loss: 0.9026874899864197:  68%|██████▊   | 4407/6470 [50:41<23:34,  1.46it/s]
3|4408|Loss: 0.9026874899864197:  68%|██████▊   | 4408/6470 [50:41<25:05,  1.37it/s]
3|4409|Loss: 0.8178417682647705:  68%|██████▊   | 4408/6470 [50:42<25:05,  1.37it/s]
3|4409|Loss: 0.8178417682647705:  68%|██████▊   | 4409/6470 [50:42<26:16,  1.31it/s]
3|4410|Loss: 0.7061216235160828:  68%|██████▊   | 4409/6470 [50:43<26:16,  1.31it/s]
3|4410|Loss: 0.7061216235160828:  68%|██████▊   | 4410/6470 [50:43<26:08,  1.31it/s]
3|4411|Loss: 0.753836989402771:  68%|██████▊   | 4410/6470 [50:43<26:08,  1.31it/s] 
3|4411|Loss: 0.753836989402771:  68%|██████▊   | 4411/6470 [50:44<24:55,  1.38it/s]
3|4412|Loss: 0.9462505578994751:  68%|██████▊   | 4411/6470 [50:44<24:55,  1.38it/s]
3|4412|Loss: 0.9462505578994751:  68%|██████▊   | 4412/6470 [50:44<24:27,  1.40it/s]
3|4413|Loss: 0.8243979215621948:  68%|██████▊   | 4412/6470 [50:45<24:27,  1.40it/s]
3|4413|Loss: 0.8243979215621948:  68%|██████▊   | 4413/6470 [50:45<24:36,  1.39it/s]
3|4414|Loss: 0.6875526309013367:  68%|██████▊   | 4413/6470 [50:46<24:36,  1.39it/s]
3|4414|Loss: 0.6875526309013367:  68%|██████▊   | 4414/6470 [50:46<26:36,  1.29it/s]
3|4415|Loss: 0.7887879610061646:  68%|██████▊   | 4414/6470 [50:47<26:36,  1.29it/s]
3|4415|Loss: 0.7887879610061646:  68%|██████▊   | 4415/6470 [50:47<27:52,  1.23it/s]
3|4416|Loss: 0.9171547293663025:  68%|██████▊   | 4415/6470 [50:47<27:52,  1.23it/s]
3|4416|Loss: 0.9171547293663025:  68%|██████▊   | 4416/6470 [50:48<26:36,  1.29it/s]
3|4417|Loss: 0.8496605753898621:  68%|██████▊   | 4416/6470 [50:48<26:36,  1.29it/s]
3|4417|Loss: 0.8496605753898621:  68%|██████▊   | 4417/6470 [50:48<25:08,  1.36it/s]
3|4418|Loss: 0.9028730988502502:  68%|██████▊   | 4417/6470 [50:49<25:08,  1.36it/s]
3|4418|Loss: 0.9028730988502502:  68%|██████▊   | 4418/6470 [50:49<24:01,  1.42it/s]
3|4419|Loss: 0.7841681241989136:  68%|██████▊   | 4418/6470 [50:49<24:01,  1.42it/s]
3|4419|Loss: 0.7841681241989136:  68%|██████▊   | 4419/6470 [50:50<23:21,  1.46it/s]
3|4420|Loss: 0.8915212154388428:  68%|██████▊   | 4419/6470 [50:50<23:21,  1.46it/s]
3|4420|Loss: 0.8915212154388428:  68%|██████▊   | 4420/6470 [50:50<23:49,  1.43it/s]
3|4421|Loss: 0.8571719527244568:  68%|██████▊   | 4420/6470 [50:51<23:49,  1.43it/s]
3|4421|Loss: 0.8571719527244568:  68%|██████▊   | 4421/6470 [50:51<25:14,  1.35it/s]
3|4422|Loss: 1.0045051574707031:  68%|██████▊   | 4421/6470 [50:52<25:14,  1.35it/s]
3|4422|Loss: 1.0045051574707031:  68%|██████▊   | 4422/6470 [50:52<25:14,  1.35it/s]
3|4423|Loss: 0.7266294360160828:  68%|██████▊   | 4422/6470 [50:52<25:14,  1.35it/s]
3|4423|Loss: 0.7266294360160828:  68%|██████▊   | 4423/6470 [50:53<24:24,  1.40it/s]
3|4424|Loss: 0.8551219701766968:  68%|██████▊   | 4423/6470 [50:53<24:24,  1.40it/s]
3|4424|Loss: 0.8551219701766968:  68%|██████▊   | 4424/6470 [50:53<24:30,  1.39it/s]
3|4425|Loss: 0.8589364290237427:  68%|██████▊   | 4424/6470 [50:54<24:30,  1.39it/s]
3|4425|Loss: 0.8589364290237427:  68%|██████▊   | 4425/6470 [50:54<23:46,  1.43it/s]
3|4426|Loss: 0.894025981426239:  68%|██████▊   | 4425/6470 [50:54<23:46,  1.43it/s] 
3|4426|Loss: 0.894025981426239:  68%|██████▊   | 4426/6470 [50:55<23:56,  1.42it/s]
3|4427|Loss: 0.9102879166603088:  68%|██████▊   | 4426/6470 [50:55<23:56,  1.42it/s]
3|4427|Loss: 0.9102879166603088:  68%|██████▊   | 4427/6470 [50:55<24:02,  1.42it/s]
3|4428|Loss: 0.9463037848472595:  68%|██████▊   | 4427/6470 [50:56<24:02,  1.42it/s]
3|4428|Loss: 0.9463037848472595:  68%|██████▊   | 4428/6470 [50:56<24:22,  1.40it/s]
3|4429|Loss: 0.9867527484893799:  68%|██████▊   | 4428/6470 [50:56<24:22,  1.40it/s]
3|4429|Loss: 0.9867527484893799:  68%|██████▊   | 4429/6470 [50:57<25:21,  1.34it/s]
3|4430|Loss: 0.8662075996398926:  68%|██████▊   | 4429/6470 [50:57<25:21,  1.34it/s]
3|4430|Loss: 0.8662075996398926:  68%|██████▊   | 4430/6470 [50:58<24:43,  1.38it/s]
3|4431|Loss: 0.7783564925193787:  68%|██████▊   | 4430/6470 [50:58<24:43,  1.38it/s]
3|4431|Loss: 0.7783564925193787:  68%|██████▊   | 4431/6470 [50:58<23:50,  1.42it/s]
3|4432|Loss: 0.828723669052124:  68%|██████▊   | 4431/6470 [50:58<23:50,  1.42it/s] 
3|4432|Loss: 0.828723669052124:  69%|██████▊   | 4432/6470 [50:59<21:10,  1.60it/s]
3|4433|Loss: 0.8196492195129395:  69%|██████▊   | 4432/6470 [50:59<21:10,  1.60it/s]
3|4433|Loss: 0.8196492195129395:  69%|██████▊   | 4433/6470 [50:59<21:23,  1.59it/s]
3|4434|Loss: 0.995599091053009:  69%|██████▊   | 4433/6470 [51:00<21:23,  1.59it/s] 
3|4434|Loss: 0.995599091053009:  69%|██████▊   | 4434/6470 [51:00<20:40,  1.64it/s]
3|4435|Loss: 0.7794119715690613:  69%|██████▊   | 4434/6470 [51:00<20:40,  1.64it/s]
3|4435|Loss: 0.7794119715690613:  69%|██████▊   | 4435/6470 [51:01<22:21,  1.52it/s]
3|4436|Loss: 0.8602868318557739:  69%|██████▊   | 4435/6470 [51:01<22:21,  1.52it/s]
3|4436|Loss: 0.8602868318557739:  69%|██████▊   | 4436/6470 [51:01<22:39,  1.50it/s]
3|4437|Loss: 0.7487741708755493:  69%|██████▊   | 4436/6470 [51:02<22:39,  1.50it/s]
3|4437|Loss: 0.7487741708755493:  69%|██████▊   | 4437/6470 [51:02<21:19,  1.59it/s]
3|4438|Loss: 0.9174363613128662:  69%|██████▊   | 4437/6470 [51:02<21:19,  1.59it/s]
3|4438|Loss: 0.9174363613128662:  69%|██████▊   | 4438/6470 [51:03<22:24,  1.51it/s]
3|4439|Loss: 0.8894107341766357:  69%|██████▊   | 4438/6470 [51:03<22:24,  1.51it/s]
3|4439|Loss: 0.8894107341766357:  69%|██████▊   | 4439/6470 [51:03<23:23,  1.45it/s]
3|4440|Loss: 0.9237701892852783:  69%|██████▊   | 4439/6470 [51:04<23:23,  1.45it/s]
3|4440|Loss: 0.9237701892852783:  69%|██████▊   | 4440/6470 [51:04<21:34,  1.57it/s]
3|4441|Loss: 0.8265312314033508:  69%|██████▊   | 4440/6470 [51:04<21:34,  1.57it/s]
3|4441|Loss: 0.8265312314033508:  69%|██████▊   | 4441/6470 [51:04<20:06,  1.68it/s]
3|4442|Loss: 0.9587068557739258:  69%|██████▊   | 4441/6470 [51:05<20:06,  1.68it/s]
3|4442|Loss: 0.9587068557739258:  69%|██████▊   | 4442/6470 [51:05<19:47,  1.71it/s]
3|4443|Loss: 0.8877466320991516:  69%|██████▊   | 4442/6470 [51:05<19:47,  1.71it/s]
3|4443|Loss: 0.8877466320991516:  69%|██████▊   | 4443/6470 [51:05<19:42,  1.71it/s]
3|4444|Loss: 0.9111198782920837:  69%|██████▊   | 4443/6470 [51:06<19:42,  1.71it/s]
3|4444|Loss: 0.9111198782920837:  69%|██████▊   | 4444/6470 [51:06<21:39,  1.56it/s]
3|4445|Loss: 0.8517327904701233:  69%|██████▊   | 4444/6470 [51:07<21:39,  1.56it/s]
3|4445|Loss: 0.8517327904701233:  69%|██████▊   | 4445/6470 [51:07<23:32,  1.43it/s]
3|4446|Loss: 0.7501040697097778:  69%|██████▊   | 4445/6470 [51:08<23:32,  1.43it/s]
3|4446|Loss: 0.7501040697097778:  69%|██████▊   | 4446/6470 [51:08<24:52,  1.36it/s]
3|4447|Loss: 0.7520206570625305:  69%|██████▊   | 4446/6470 [51:08<24:52,  1.36it/s]
3|4447|Loss: 0.7520206570625305:  69%|██████▊   | 4447/6470 [51:09<23:34,  1.43it/s]
3|4448|Loss: 0.8153076171875:  69%|██████▊   | 4447/6470 [51:09<23:34,  1.43it/s]   
3|4448|Loss: 0.8153076171875:  69%|██████▊   | 4448/6470 [51:09<22:48,  1.48it/s]
3|4449|Loss: 0.8113981485366821:  69%|██████▊   | 4448/6470 [51:09<22:48,  1.48it/s]
3|4449|Loss: 0.8113981485366821:  69%|██████▉   | 4449/6470 [51:10<21:24,  1.57it/s]
3|4450|Loss: 0.728515625:  69%|██████▉   | 4449/6470 [51:10<21:24,  1.57it/s]       
3|4450|Loss: 0.728515625:  69%|██████▉   | 4450/6470 [51:10<20:06,  1.67it/s]
3|4451|Loss: 0.9798104166984558:  69%|██████▉   | 4450/6470 [51:11<20:06,  1.67it/s]
3|4451|Loss: 0.9798104166984558:  69%|██████▉   | 4451/6470 [51:11<20:56,  1.61it/s]
3|4452|Loss: 0.8626866936683655:  69%|██████▉   | 4451/6470 [51:11<20:56,  1.61it/s]
3|4452|Loss: 0.8626866936683655:  69%|██████▉   | 4452/6470 [51:12<21:19,  1.58it/s]
3|4453|Loss: 0.7972787022590637:  69%|██████▉   | 4452/6470 [51:12<21:19,  1.58it/s]
3|4453|Loss: 0.7972787022590637:  69%|██████▉   | 4453/6470 [51:12<20:16,  1.66it/s]
3|4454|Loss: 0.9163343906402588:  69%|██████▉   | 4453/6470 [51:12<20:16,  1.66it/s]
3|4454|Loss: 0.9163343906402588:  69%|██████▉   | 4454/6470 [51:13<20:52,  1.61it/s]
3|4455|Loss: 0.8181030750274658:  69%|██████▉   | 4454/6470 [51:13<20:52,  1.61it/s]
3|4455|Loss: 0.8181030750274658:  69%|██████▉   | 4455/6470 [51:13<20:17,  1.65it/s]
3|4456|Loss: 0.8475328087806702:  69%|██████▉   | 4455/6470 [51:14<20:17,  1.65it/s]
3|4456|Loss: 0.8475328087806702:  69%|██████▉   | 4456/6470 [51:14<19:55,  1.68it/s]
3|4457|Loss: 0.834261417388916:  69%|██████▉   | 4456/6470 [51:14<19:55,  1.68it/s] 
3|4457|Loss: 0.834261417388916:  69%|██████▉   | 4457/6470 [51:14<18:58,  1.77it/s]
3|4458|Loss: 0.9497906565666199:  69%|██████▉   | 4457/6470 [51:15<18:58,  1.77it/s]
3|4458|Loss: 0.9497906565666199:  69%|██████▉   | 4458/6470 [51:15<18:29,  1.81it/s]
3|4459|Loss: 1.0021909475326538:  69%|██████▉   | 4458/6470 [51:15<18:29,  1.81it/s]
3|4459|Loss: 1.0021909475326538:  69%|██████▉   | 4459/6470 [51:16<19:53,  1.68it/s]
3|4460|Loss: 0.8776589632034302:  69%|██████▉   | 4459/6470 [51:16<19:53,  1.68it/s]
3|4460|Loss: 0.8776589632034302:  69%|██████▉   | 4460/6470 [51:16<20:04,  1.67it/s]
3|4461|Loss: 0.9205860495567322:  69%|██████▉   | 4460/6470 [51:17<20:04,  1.67it/s]
3|4461|Loss: 0.9205860495567322:  69%|██████▉   | 4461/6470 [51:17<18:51,  1.78it/s]
3|4462|Loss: 0.5930052399635315:  69%|██████▉   | 4461/6470 [51:17<18:51,  1.78it/s]
3|4462|Loss: 0.5930052399635315:  69%|██████▉   | 4462/6470 [51:17<19:11,  1.74it/s]
3|4463|Loss: 0.8490036725997925:  69%|██████▉   | 4462/6470 [51:18<19:11,  1.74it/s]
3|4463|Loss: 0.8490036725997925:  69%|██████▉   | 4463/6470 [51:18<19:07,  1.75it/s]
3|4464|Loss: 0.9447838068008423:  69%|██████▉   | 4463/6470 [51:18<19:07,  1.75it/s]
3|4464|Loss: 0.9447838068008423:  69%|██████▉   | 4464/6470 [51:19<20:29,  1.63it/s]
3|4465|Loss: 0.9380272030830383:  69%|██████▉   | 4464/6470 [51:19<20:29,  1.63it/s]
3|4465|Loss: 0.9380272030830383:  69%|██████▉   | 4465/6470 [51:19<22:12,  1.51it/s]
3|4466|Loss: 0.7738719582557678:  69%|██████▉   | 4465/6470 [51:20<22:12,  1.51it/s]
3|4466|Loss: 0.7738719582557678:  69%|██████▉   | 4466/6470 [51:20<23:28,  1.42it/s]
3|4467|Loss: 0.8405755162239075:  69%|██████▉   | 4466/6470 [51:20<23:28,  1.42it/s]
3|4467|Loss: 0.8405755162239075:  69%|██████▉   | 4467/6470 [51:21<22:05,  1.51it/s]
3|4468|Loss: 0.7736373543739319:  69%|██████▉   | 4467/6470 [51:21<22:05,  1.51it/s]
3|4468|Loss: 0.7736373543739319:  69%|██████▉   | 4468/6470 [51:21<21:03,  1.58it/s]
3|4469|Loss: 0.8144242763519287:  69%|██████▉   | 4468/6470 [51:22<21:03,  1.58it/s]
3|4469|Loss: 0.8144242763519287:  69%|██████▉   | 4469/6470 [51:22<21:53,  1.52it/s]
3|4470|Loss: 0.9434311985969543:  69%|██████▉   | 4469/6470 [51:22<21:53,  1.52it/s]
3|4470|Loss: 0.9434311985969543:  69%|██████▉   | 4470/6470 [51:23<21:33,  1.55it/s]
3|4471|Loss: 0.8334698677062988:  69%|██████▉   | 4470/6470 [51:23<21:33,  1.55it/s]
3|4471|Loss: 0.8334698677062988:  69%|██████▉   | 4471/6470 [51:23<22:33,  1.48it/s]
3|4472|Loss: 0.9923511147499084:  69%|██████▉   | 4471/6470 [51:24<22:33,  1.48it/s]
3|4472|Loss: 0.9923511147499084:  69%|██████▉   | 4472/6470 [51:24<24:11,  1.38it/s]
3|4473|Loss: 0.7913383841514587:  69%|██████▉   | 4472/6470 [51:25<24:11,  1.38it/s]
3|4473|Loss: 0.7913383841514587:  69%|██████▉   | 4473/6470 [51:25<23:38,  1.41it/s]
3|4474|Loss: 0.9666570425033569:  69%|██████▉   | 4473/6470 [51:25<23:38,  1.41it/s]
3|4474|Loss: 0.9666570425033569:  69%|██████▉   | 4474/6470 [51:26<23:20,  1.42it/s]
3|4475|Loss: 0.7527173757553101:  69%|██████▉   | 4474/6470 [51:26<23:20,  1.42it/s]
3|4475|Loss: 0.7527173757553101:  69%|██████▉   | 4475/6470 [51:26<22:19,  1.49it/s]
3|4476|Loss: 0.8596112728118896:  69%|██████▉   | 4475/6470 [51:27<22:19,  1.49it/s]
3|4476|Loss: 0.8596112728118896:  69%|██████▉   | 4476/6470 [51:27<22:25,  1.48it/s]
3|4477|Loss: 0.821125864982605:  69%|██████▉   | 4476/6470 [51:27<22:25,  1.48it/s] 
3|4477|Loss: 0.821125864982605:  69%|██████▉   | 4477/6470 [51:27<21:41,  1.53it/s]
3|4478|Loss: 0.783263087272644:  69%|██████▉   | 4477/6470 [51:28<21:41,  1.53it/s]
3|4478|Loss: 0.783263087272644:  69%|██████▉   | 4478/6470 [51:28<21:14,  1.56it/s]
3|4479|Loss: 0.6572731137275696:  69%|██████▉   | 4478/6470 [51:28<21:14,  1.56it/s]
3|4479|Loss: 0.6572731137275696:  69%|██████▉   | 4479/6470 [51:29<22:39,  1.46it/s]
3|4480|Loss: 0.9308943748474121:  69%|██████▉   | 4479/6470 [51:29<22:39,  1.46it/s]
3|4480|Loss: 0.9308943748474121:  69%|██████▉   | 4480/6470 [51:30<22:38,  1.46it/s]
3|4481|Loss: 0.9143169522285461:  69%|██████▉   | 4480/6470 [51:30<22:38,  1.46it/s]
3|4481|Loss: 0.9143169522285461:  69%|██████▉   | 4481/6470 [51:30<22:13,  1.49it/s]
3|4482|Loss: 0.9369168281555176:  69%|██████▉   | 4481/6470 [51:31<22:13,  1.49it/s]
3|4482|Loss: 0.9369168281555176:  69%|██████▉   | 4482/6470 [51:31<23:27,  1.41it/s]
3|4483|Loss: 0.9364734888076782:  69%|██████▉   | 4482/6470 [51:31<23:27,  1.41it/s]
3|4483|Loss: 0.9364734888076782:  69%|██████▉   | 4483/6470 [51:32<25:44,  1.29it/s]
3|4484|Loss: 0.91189044713974:  69%|██████▉   | 4483/6470 [51:32<25:44,  1.29it/s]  
3|4484|Loss: 0.91189044713974:  69%|██████▉   | 4484/6470 [51:33<26:50,  1.23it/s]
3|4485|Loss: 0.828411877155304:  69%|██████▉   | 4484/6470 [51:33<26:50,  1.23it/s]
3|4485|Loss: 0.828411877155304:  69%|██████▉   | 4485/6470 [51:34<26:19,  1.26it/s]
3|4486|Loss: 0.908627986907959:  69%|██████▉   | 4485/6470 [51:34<26:19,  1.26it/s]
3|4486|Loss: 0.908627986907959:  69%|██████▉   | 4486/6470 [51:34<25:43,  1.29it/s]
3|4487|Loss: 0.9003919363021851:  69%|██████▉   | 4486/6470 [51:35<25:43,  1.29it/s]
3|4487|Loss: 0.9003919363021851:  69%|██████▉   | 4487/6470 [51:35<25:18,  1.31it/s]
3|4488|Loss: 0.8267437219619751:  69%|██████▉   | 4487/6470 [51:35<25:18,  1.31it/s]
3|4488|Loss: 0.8267437219619751:  69%|██████▉   | 4488/6470 [51:36<26:27,  1.25it/s]
3|4489|Loss: 0.8312293291091919:  69%|██████▉   | 4488/6470 [51:36<26:27,  1.25it/s]
3|4489|Loss: 0.8312293291091919:  69%|██████▉   | 4489/6470 [51:36<23:58,  1.38it/s]
3|4490|Loss: 0.8426236510276794:  69%|██████▉   | 4489/6470 [51:37<23:58,  1.38it/s]
3|4490|Loss: 0.8426236510276794:  69%|██████▉   | 4490/6470 [51:37<24:29,  1.35it/s]
3|4491|Loss: 0.8123554587364197:  69%|██████▉   | 4490/6470 [51:38<24:29,  1.35it/s]
3|4491|Loss: 0.8123554587364197:  69%|██████▉   | 4491/6470 [51:38<23:59,  1.37it/s]
3|4492|Loss: 0.7801477313041687:  69%|██████▉   | 4491/6470 [51:38<23:59,  1.37it/s]
3|4492|Loss: 0.7801477313041687:  69%|██████▉   | 4492/6470 [51:39<23:39,  1.39it/s]
3|4493|Loss: 0.836573600769043:  69%|██████▉   | 4492/6470 [51:39<23:39,  1.39it/s] 
3|4493|Loss: 0.836573600769043:  69%|██████▉   | 4493/6470 [51:39<23:33,  1.40it/s]
3|4494|Loss: 0.8045174479484558:  69%|██████▉   | 4493/6470 [51:40<23:33,  1.40it/s]
3|4494|Loss: 0.8045174479484558:  69%|██████▉   | 4494/6470 [51:40<23:39,  1.39it/s]
3|4495|Loss: 0.8129112720489502:  69%|██████▉   | 4494/6470 [51:40<23:39,  1.39it/s]
3|4495|Loss: 0.8129112720489502:  69%|██████▉   | 4495/6470 [51:41<22:08,  1.49it/s]
3|4496|Loss: 0.804803192615509:  69%|██████▉   | 4495/6470 [51:41<22:08,  1.49it/s] 
3|4496|Loss: 0.804803192615509:  69%|██████▉   | 4496/6470 [51:41<23:17,  1.41it/s]
3|4497|Loss: 0.8944460153579712:  69%|██████▉   | 4496/6470 [51:42<23:17,  1.41it/s]
3|4497|Loss: 0.8944460153579712:  70%|██████▉   | 4497/6470 [51:42<23:47,  1.38it/s]
3|4498|Loss: 0.8773543238639832:  70%|██████▉   | 4497/6470 [51:43<23:47,  1.38it/s]
3|4498|Loss: 0.8773543238639832:  70%|██████▉   | 4498/6470 [51:43<24:45,  1.33it/s]
3|4499|Loss: 0.8253515958786011:  70%|██████▉   | 4498/6470 [51:43<24:45,  1.33it/s]
3|4499|Loss: 0.8253515958786011:  70%|██████▉   | 4499/6470 [51:44<24:38,  1.33it/s]
3|4500|Loss: 0.9502061009407043:  70%|██████▉   | 4499/6470 [51:44<24:38,  1.33it/s]
3|4500|Loss: 0.9502061009407043:  70%|██████▉   | 4500/6470 [51:44<23:29,  1.40it/s]
3|4501|Loss: 0.789306104183197:  70%|██████▉   | 4500/6470 [51:45<23:29,  1.40it/s] 
3|4501|Loss: 0.789306104183197:  70%|██████▉   | 4501/6470 [51:45<22:42,  1.45it/s]
3|4502|Loss: 0.7892413139343262:  70%|██████▉   | 4501/6470 [51:45<22:42,  1.45it/s]
3|4502|Loss: 0.7892413139343262:  70%|██████▉   | 4502/6470 [51:46<23:51,  1.37it/s]
3|4503|Loss: 0.8267009854316711:  70%|██████▉   | 4502/6470 [51:46<23:51,  1.37it/s]
3|4503|Loss: 0.8267009854316711:  70%|██████▉   | 4503/6470 [51:46<23:29,  1.40it/s]
3|4504|Loss: 0.8075945377349854:  70%|██████▉   | 4503/6470 [51:47<23:29,  1.40it/s]
3|4504|Loss: 0.8075945377349854:  70%|██████▉   | 4504/6470 [51:47<21:43,  1.51it/s]
3|4505|Loss: 0.9014251232147217:  70%|██████▉   | 4504/6470 [51:47<21:43,  1.51it/s]
3|4505|Loss: 0.9014251232147217:  70%|██████▉   | 4505/6470 [51:48<22:51,  1.43it/s]
3|4506|Loss: 0.8029349446296692:  70%|██████▉   | 4505/6470 [51:48<22:51,  1.43it/s]
3|4506|Loss: 0.8029349446296692:  70%|██████▉   | 4506/6470 [51:49<22:48,  1.44it/s]
3|4507|Loss: 0.9153288006782532:  70%|██████▉   | 4506/6470 [51:49<22:48,  1.44it/s]
3|4507|Loss: 0.9153288006782532:  70%|██████▉   | 4507/6470 [51:49<23:53,  1.37it/s]
3|4508|Loss: 0.8440418243408203:  70%|██████▉   | 4507/6470 [51:50<23:53,  1.37it/s]
3|4508|Loss: 0.8440418243408203:  70%|██████▉   | 4508/6470 [51:50<23:56,  1.37it/s]
3|4509|Loss: 0.8185060620307922:  70%|██████▉   | 4508/6470 [51:50<23:56,  1.37it/s]
3|4509|Loss: 0.8185060620307922:  70%|██████▉   | 4509/6470 [51:51<23:08,  1.41it/s]
3|4510|Loss: 0.9098572731018066:  70%|██████▉   | 4509/6470 [51:51<23:08,  1.41it/s]
3|4510|Loss: 0.9098572731018066:  70%|██████▉   | 4510/6470 [51:51<21:35,  1.51it/s]
3|4511|Loss: 0.8564989566802979:  70%|██████▉   | 4510/6470 [51:52<21:35,  1.51it/s]
3|4511|Loss: 0.8564989566802979:  70%|██████▉   | 4511/6470 [51:52<20:43,  1.58it/s]
3|4512|Loss: 0.6652271151542664:  70%|██████▉   | 4511/6470 [51:52<20:43,  1.58it/s]
3|4512|Loss: 0.6652271151542664:  70%|██████▉   | 4512/6470 [51:52<19:10,  1.70it/s]
3|4513|Loss: 0.7248936891555786:  70%|██████▉   | 4512/6470 [51:53<19:10,  1.70it/s]
3|4513|Loss: 0.7248936891555786:  70%|██████▉   | 4513/6470 [51:53<20:05,  1.62it/s]
3|4514|Loss: 0.7296864986419678:  70%|██████▉   | 4513/6470 [51:53<20:05,  1.62it/s]
3|4514|Loss: 0.7296864986419678:  70%|██████▉   | 4514/6470 [51:54<20:22,  1.60it/s]
3|4515|Loss: 0.8498228192329407:  70%|██████▉   | 4514/6470 [51:54<20:22,  1.60it/s]
3|4515|Loss: 0.8498228192329407:  70%|██████▉   | 4515/6470 [51:54<19:23,  1.68it/s]
3|4516|Loss: 0.9878129363059998:  70%|██████▉   | 4515/6470 [51:55<19:23,  1.68it/s]
3|4516|Loss: 0.9878129363059998:  70%|██████▉   | 4516/6470 [51:55<20:01,  1.63it/s]
3|4517|Loss: 0.9119592905044556:  70%|██████▉   | 4516/6470 [51:55<20:01,  1.63it/s]
3|4517|Loss: 0.9119592905044556:  70%|██████▉   | 4517/6470 [51:55<20:24,  1.60it/s]
3|4518|Loss: 0.7630557417869568:  70%|██████▉   | 4517/6470 [51:56<20:24,  1.60it/s]
3|4518|Loss: 0.7630557417869568:  70%|██████▉   | 4518/6470 [51:56<19:03,  1.71it/s]
3|4519|Loss: 0.806464672088623:  70%|██████▉   | 4518/6470 [51:56<19:03,  1.71it/s] 
3|4519|Loss: 0.806464672088623:  70%|██████▉   | 4519/6470 [51:57<18:59,  1.71it/s]
3|4520|Loss: 0.7413250803947449:  70%|██████▉   | 4519/6470 [51:57<18:59,  1.71it/s]
3|4520|Loss: 0.7413250803947449:  70%|██████▉   | 4520/6470 [51:57<20:17,  1.60it/s]
3|4521|Loss: 0.9500980973243713:  70%|██████▉   | 4520/6470 [51:58<20:17,  1.60it/s]
3|4521|Loss: 0.9500980973243713:  70%|██████▉   | 4521/6470 [51:58<21:10,  1.53it/s]
3|4522|Loss: 0.9045262932777405:  70%|██████▉   | 4521/6470 [51:58<21:10,  1.53it/s]
3|4522|Loss: 0.9045262932777405:  70%|██████▉   | 4522/6470 [51:59<21:41,  1.50it/s]
3|4523|Loss: 0.7352326512336731:  70%|██████▉   | 4522/6470 [51:59<21:41,  1.50it/s]
3|4523|Loss: 0.7352326512336731:  70%|██████▉   | 4523/6470 [51:59<21:18,  1.52it/s]
3|4524|Loss: 0.8325412273406982:  70%|██████▉   | 4523/6470 [52:00<21:18,  1.52it/s]
3|4524|Loss: 0.8325412273406982:  70%|██████▉   | 4524/6470 [52:00<22:33,  1.44it/s]
3|4525|Loss: 0.8310990333557129:  70%|██████▉   | 4524/6470 [52:00<22:33,  1.44it/s]
3|4525|Loss: 0.8310990333557129:  70%|██████▉   | 4525/6470 [52:01<21:53,  1.48it/s]
3|4526|Loss: 0.9801949858665466:  70%|██████▉   | 4525/6470 [52:01<21:53,  1.48it/s]
3|4526|Loss: 0.9801949858665466:  70%|██████▉   | 4526/6470 [52:01<22:08,  1.46it/s]
3|4527|Loss: 0.7197139859199524:  70%|██████▉   | 4526/6470 [52:02<22:08,  1.46it/s]
3|4527|Loss: 0.7197139859199524:  70%|██████▉   | 4527/6470 [52:02<20:53,  1.55it/s]
3|4528|Loss: 0.7987144589424133:  70%|██████▉   | 4527/6470 [52:02<20:53,  1.55it/s]
3|4528|Loss: 0.7987144589424133:  70%|██████▉   | 4528/6470 [52:03<21:27,  1.51it/s]
3|4529|Loss: 0.8034959435462952:  70%|██████▉   | 4528/6470 [52:03<21:27,  1.51it/s]
3|4529|Loss: 0.8034959435462952:  70%|███████   | 4529/6470 [52:03<20:37,  1.57it/s]
3|4530|Loss: 0.9675959348678589:  70%|███████   | 4529/6470 [52:04<20:37,  1.57it/s]
3|4530|Loss: 0.9675959348678589:  70%|███████   | 4530/6470 [52:04<20:23,  1.59it/s]
3|4531|Loss: 0.9183099269866943:  70%|███████   | 4530/6470 [52:04<20:23,  1.59it/s]
3|4531|Loss: 0.9183099269866943:  70%|███████   | 4531/6470 [52:05<20:20,  1.59it/s]
3|4532|Loss: 0.9716671705245972:  70%|███████   | 4531/6470 [52:05<20:20,  1.59it/s]
3|4532|Loss: 0.9716671705245972:  70%|███████   | 4532/6470 [52:05<18:57,  1.70it/s]
3|4533|Loss: 0.7701632976531982:  70%|███████   | 4532/6470 [52:05<18:57,  1.70it/s]
3|4533|Loss: 0.7701632976531982:  70%|███████   | 4533/6470 [52:06<19:05,  1.69it/s]
3|4534|Loss: 0.9784212112426758:  70%|███████   | 4533/6470 [52:06<19:05,  1.69it/s]
3|4534|Loss: 0.9784212112426758:  70%|███████   | 4534/6470 [52:06<19:36,  1.64it/s]
3|4535|Loss: 0.9011886715888977:  70%|███████   | 4534/6470 [52:07<19:36,  1.64it/s]
3|4535|Loss: 0.9011886715888977:  70%|███████   | 4535/6470 [52:07<19:56,  1.62it/s]
3|4536|Loss: 0.9634377956390381:  70%|███████   | 4535/6470 [52:07<19:56,  1.62it/s]
3|4536|Loss: 0.9634377956390381:  70%|███████   | 4536/6470 [52:07<19:37,  1.64it/s]
3|4537|Loss: 0.8643772602081299:  70%|███████   | 4536/6470 [52:08<19:37,  1.64it/s]
3|4537|Loss: 0.8643772602081299:  70%|███████   | 4537/6470 [52:08<18:23,  1.75it/s]
3|4538|Loss: 0.7507728338241577:  70%|███████   | 4537/6470 [52:08<18:23,  1.75it/s]
3|4538|Loss: 0.7507728338241577:  70%|███████   | 4538/6470 [52:09<19:06,  1.69it/s]
3|4539|Loss: 0.7541654706001282:  70%|███████   | 4538/6470 [52:09<19:06,  1.69it/s]
3|4539|Loss: 0.7541654706001282:  70%|███████   | 4539/6470 [52:09<21:11,  1.52it/s]
3|4540|Loss: 0.8822855949401855:  70%|███████   | 4539/6470 [52:10<21:11,  1.52it/s]
3|4540|Loss: 0.8822855949401855:  70%|███████   | 4540/6470 [52:10<22:12,  1.45it/s]
3|4541|Loss: 0.9520768523216248:  70%|███████   | 4540/6470 [52:11<22:12,  1.45it/s]
3|4541|Loss: 0.9520768523216248:  70%|███████   | 4541/6470 [52:11<22:27,  1.43it/s]
3|4542|Loss: 0.775511622428894:  70%|███████   | 4541/6470 [52:11<22:27,  1.43it/s] 
3|4542|Loss: 0.775511622428894:  70%|███████   | 4542/6470 [52:11<21:09,  1.52it/s]
3|4543|Loss: 0.9034442901611328:  70%|███████   | 4542/6470 [52:12<21:09,  1.52it/s]
3|4543|Loss: 0.9034442901611328:  70%|███████   | 4543/6470 [52:12<20:33,  1.56it/s]
3|4544|Loss: 0.9847091436386108:  70%|███████   | 4543/6470 [52:12<20:33,  1.56it/s]
3|4544|Loss: 0.9847091436386108:  70%|███████   | 4544/6470 [52:13<21:43,  1.48it/s]
3|4545|Loss: 0.9427686929702759:  70%|███████   | 4544/6470 [52:13<21:43,  1.48it/s]
3|4545|Loss: 0.9427686929702759:  70%|███████   | 4545/6470 [52:14<22:57,  1.40it/s]
3|4546|Loss: 0.8169428706169128:  70%|███████   | 4545/6470 [52:14<22:57,  1.40it/s]
3|4546|Loss: 0.8169428706169128:  70%|███████   | 4546/6470 [52:14<23:15,  1.38it/s]
3|4547|Loss: 0.7779592871665955:  70%|███████   | 4546/6470 [52:15<23:15,  1.38it/s]
3|4547|Loss: 0.7779592871665955:  70%|███████   | 4547/6470 [52:15<23:45,  1.35it/s]
3|4548|Loss: 0.852670431137085:  70%|███████   | 4547/6470 [52:16<23:45,  1.35it/s] 
3|4548|Loss: 0.852670431137085:  70%|███████   | 4548/6470 [52:16<22:37,  1.42it/s]
3|4549|Loss: 0.742594301700592:  70%|███████   | 4548/6470 [52:16<22:37,  1.42it/s]
3|4549|Loss: 0.742594301700592:  70%|███████   | 4549/6470 [52:16<21:43,  1.47it/s]
3|4550|Loss: 0.9035540819168091:  70%|███████   | 4549/6470 [52:17<21:43,  1.47it/s]
3|4550|Loss: 0.9035540819168091:  70%|███████   | 4550/6470 [52:17<21:18,  1.50it/s]
3|4551|Loss: 0.5627049207687378:  70%|███████   | 4550/6470 [52:18<21:18,  1.50it/s]
3|4551|Loss: 0.5627049207687378:  70%|███████   | 4551/6470 [52:18<24:16,  1.32it/s]
3|4552|Loss: 0.8016078472137451:  70%|███████   | 4551/6470 [52:19<24:16,  1.32it/s]
3|4552|Loss: 0.8016078472137451:  70%|███████   | 4552/6470 [52:19<25:49,  1.24it/s]
3|4553|Loss: 0.7915346622467041:  70%|███████   | 4552/6470 [52:19<25:49,  1.24it/s]
3|4553|Loss: 0.7915346622467041:  70%|███████   | 4553/6470 [52:20<26:33,  1.20it/s]
3|4554|Loss: 0.731932520866394:  70%|███████   | 4553/6470 [52:20<26:33,  1.20it/s] 
3|4554|Loss: 0.731932520866394:  70%|███████   | 4554/6470 [52:21<26:06,  1.22it/s]
3|4555|Loss: 0.9581323862075806:  70%|███████   | 4554/6470 [52:21<26:06,  1.22it/s]
3|4555|Loss: 0.9581323862075806:  70%|███████   | 4555/6470 [52:21<24:26,  1.31it/s]
3|4556|Loss: 0.7188338041305542:  70%|███████   | 4555/6470 [52:22<24:26,  1.31it/s]
3|4556|Loss: 0.7188338041305542:  70%|███████   | 4556/6470 [52:22<24:51,  1.28it/s]
3|4557|Loss: 0.8241256475448608:  70%|███████   | 4556/6470 [52:22<24:51,  1.28it/s]
3|4557|Loss: 0.8241256475448608:  70%|███████   | 4557/6470 [52:23<22:31,  1.42it/s]
3|4558|Loss: 0.9167270660400391:  70%|███████   | 4557/6470 [52:23<22:31,  1.42it/s]
3|4558|Loss: 0.9167270660400391:  70%|███████   | 4558/6470 [52:23<22:57,  1.39it/s]
3|4559|Loss: 0.8049624562263489:  70%|███████   | 4558/6470 [52:24<22:57,  1.39it/s]
3|4559|Loss: 0.8049624562263489:  70%|███████   | 4559/6470 [52:24<21:36,  1.47it/s]
3|4560|Loss: 0.945000171661377:  70%|███████   | 4559/6470 [52:24<21:36,  1.47it/s] 
3|4560|Loss: 0.945000171661377:  70%|███████   | 4560/6470 [52:25<20:47,  1.53it/s]
3|4561|Loss: 0.8409856557846069:  70%|███████   | 4560/6470 [52:25<20:47,  1.53it/s]
3|4561|Loss: 0.8409856557846069:  70%|███████   | 4561/6470 [52:25<21:04,  1.51it/s]
3|4562|Loss: 0.8501520156860352:  70%|███████   | 4561/6470 [52:26<21:04,  1.51it/s]
3|4562|Loss: 0.8501520156860352:  71%|███████   | 4562/6470 [52:26<19:58,  1.59it/s]
3|4563|Loss: 1.004025936126709:  71%|███████   | 4562/6470 [52:26<19:58,  1.59it/s] 
3|4563|Loss: 1.004025936126709:  71%|███████   | 4563/6470 [52:26<20:15,  1.57it/s]
3|4564|Loss: 0.7286268472671509:  71%|███████   | 4563/6470 [52:27<20:15,  1.57it/s]
3|4564|Loss: 0.7286268472671509:  71%|███████   | 4564/6470 [52:27<19:22,  1.64it/s]
3|4565|Loss: 0.8035542964935303:  71%|███████   | 4564/6470 [52:27<19:22,  1.64it/s]
3|4565|Loss: 0.8035542964935303:  71%|███████   | 4565/6470 [52:28<20:59,  1.51it/s]
3|4566|Loss: 0.9086487293243408:  71%|███████   | 4565/6470 [52:28<20:59,  1.51it/s]
3|4566|Loss: 0.9086487293243408:  71%|███████   | 4566/6470 [52:28<21:05,  1.50it/s]
3|4567|Loss: 0.7416207790374756:  71%|███████   | 4566/6470 [52:29<21:05,  1.50it/s]
3|4567|Loss: 0.7416207790374756:  71%|███████   | 4567/6470 [52:29<21:25,  1.48it/s]
3|4568|Loss: 0.9287115931510925:  71%|███████   | 4567/6470 [52:30<21:25,  1.48it/s]
3|4568|Loss: 0.9287115931510925:  71%|███████   | 4568/6470 [52:30<22:43,  1.39it/s]
3|4569|Loss: 1.0184653997421265:  71%|███████   | 4568/6470 [52:30<22:43,  1.39it/s]
3|4569|Loss: 1.0184653997421265:  71%|███████   | 4569/6470 [52:31<23:04,  1.37it/s]
3|4570|Loss: 0.7764055728912354:  71%|███████   | 4569/6470 [52:31<23:04,  1.37it/s]
3|4570|Loss: 0.7764055728912354:  71%|███████   | 4570/6470 [52:31<22:16,  1.42it/s]
3|4571|Loss: 0.6836720705032349:  71%|███████   | 4570/6470 [52:32<22:16,  1.42it/s]
3|4571|Loss: 0.6836720705032349:  71%|███████   | 4571/6470 [52:32<22:18,  1.42it/s]
3|4572|Loss: 0.8163097500801086:  71%|███████   | 4571/6470 [52:32<22:18,  1.42it/s]
3|4572|Loss: 0.8163097500801086:  71%|███████   | 4572/6470 [52:33<23:27,  1.35it/s]
3|4573|Loss: 0.732439398765564:  71%|███████   | 4572/6470 [52:33<23:27,  1.35it/s] 
3|4573|Loss: 0.732439398765564:  71%|███████   | 4573/6470 [52:34<22:54,  1.38it/s]
3|4574|Loss: 0.7940598726272583:  71%|███████   | 4573/6470 [52:34<22:54,  1.38it/s]
3|4574|Loss: 0.7940598726272583:  71%|███████   | 4574/6470 [52:34<22:57,  1.38it/s]
3|4575|Loss: 0.7711458802223206:  71%|███████   | 4574/6470 [52:35<22:57,  1.38it/s]
3|4575|Loss: 0.7711458802223206:  71%|███████   | 4575/6470 [52:35<22:52,  1.38it/s]
3|4576|Loss: 0.9224728345870972:  71%|███████   | 4575/6470 [52:35<22:52,  1.38it/s]
3|4576|Loss: 0.9224728345870972:  71%|███████   | 4576/6470 [52:36<21:11,  1.49it/s]
3|4577|Loss: 0.9030751585960388:  71%|███████   | 4576/6470 [52:36<21:11,  1.49it/s]
3|4577|Loss: 0.9030751585960388:  71%|███████   | 4577/6470 [52:36<20:39,  1.53it/s]
3|4578|Loss: 0.8517472147941589:  71%|███████   | 4577/6470 [52:37<20:39,  1.53it/s]
3|4578|Loss: 0.8517472147941589:  71%|███████   | 4578/6470 [52:37<21:22,  1.48it/s]
3|4579|Loss: 0.7368489503860474:  71%|███████   | 4578/6470 [52:37<21:22,  1.48it/s]
3|4579|Loss: 0.7368489503860474:  71%|███████   | 4579/6470 [52:38<20:52,  1.51it/s]
3|4580|Loss: 0.8170202374458313:  71%|███████   | 4579/6470 [52:38<20:52,  1.51it/s]
3|4580|Loss: 0.8170202374458313:  71%|███████   | 4580/6470 [52:38<21:03,  1.50it/s]
3|4581|Loss: 0.7653416991233826:  71%|███████   | 4580/6470 [52:39<21:03,  1.50it/s]
3|4581|Loss: 0.7653416991233826:  71%|███████   | 4581/6470 [52:39<20:46,  1.52it/s]
3|4582|Loss: 0.7742624282836914:  71%|███████   | 4581/6470 [52:39<20:46,  1.52it/s]
3|4582|Loss: 0.7742624282836914:  71%|███████   | 4582/6470 [52:40<21:01,  1.50it/s]
3|4583|Loss: 0.8100877404212952:  71%|███████   | 4582/6470 [52:40<21:01,  1.50it/s]
3|4583|Loss: 0.8100877404212952:  71%|███████   | 4583/6470 [52:40<21:22,  1.47it/s]
3|4584|Loss: 0.917753279209137:  71%|███████   | 4583/6470 [52:41<21:22,  1.47it/s] 
3|4584|Loss: 0.917753279209137:  71%|███████   | 4584/6470 [52:41<21:37,  1.45it/s]
3|4585|Loss: 0.8470329642295837:  71%|███████   | 4584/6470 [52:41<21:37,  1.45it/s]
3|4585|Loss: 0.8470329642295837:  71%|███████   | 4585/6470 [52:42<21:08,  1.49it/s]
3|4586|Loss: 0.7799369692802429:  71%|███████   | 4585/6470 [52:42<21:08,  1.49it/s]
3|4586|Loss: 0.7799369692802429:  71%|███████   | 4586/6470 [52:42<21:12,  1.48it/s]
3|4587|Loss: 0.8744880557060242:  71%|███████   | 4586/6470 [52:43<21:12,  1.48it/s]
3|4587|Loss: 0.8744880557060242:  71%|███████   | 4587/6470 [52:43<21:28,  1.46it/s]
3|4588|Loss: 0.9507052898406982:  71%|███████   | 4587/6470 [52:43<21:28,  1.46it/s]
3|4588|Loss: 0.9507052898406982:  71%|███████   | 4588/6470 [52:44<21:09,  1.48it/s]
3|4589|Loss: 0.996545135974884:  71%|███████   | 4588/6470 [52:44<21:09,  1.48it/s] 
3|4589|Loss: 0.996545135974884:  71%|███████   | 4589/6470 [52:44<21:17,  1.47it/s]
3|4590|Loss: 0.9762958884239197:  71%|███████   | 4589/6470 [52:45<21:17,  1.47it/s]
3|4590|Loss: 0.9762958884239197:  71%|███████   | 4590/6470 [52:45<21:44,  1.44it/s]
3|4591|Loss: 0.8196252584457397:  71%|███████   | 4590/6470 [52:45<21:44,  1.44it/s]
3|4591|Loss: 0.8196252584457397:  71%|███████   | 4591/6470 [52:46<22:48,  1.37it/s]
3|4592|Loss: 0.887272298336029:  71%|███████   | 4591/6470 [52:46<22:48,  1.37it/s] 
3|4592|Loss: 0.887272298336029:  71%|███████   | 4592/6470 [52:47<22:48,  1.37it/s]
3|4593|Loss: 0.8754202723503113:  71%|███████   | 4592/6470 [52:47<22:48,  1.37it/s]
3|4593|Loss: 0.8754202723503113:  71%|███████   | 4593/6470 [52:47<22:45,  1.38it/s]
3|4594|Loss: 0.7425225973129272:  71%|███████   | 4593/6470 [52:48<22:45,  1.38it/s]
3|4594|Loss: 0.7425225973129272:  71%|███████   | 4594/6470 [52:48<23:50,  1.31it/s]
3|4595|Loss: 0.9227339625358582:  71%|███████   | 4594/6470 [52:49<23:50,  1.31it/s]
3|4595|Loss: 0.9227339625358582:  71%|███████   | 4595/6470 [52:49<23:34,  1.33it/s]
3|4596|Loss: 0.6902854442596436:  71%|███████   | 4595/6470 [52:49<23:34,  1.33it/s]
3|4596|Loss: 0.6902854442596436:  71%|███████   | 4596/6470 [52:50<24:59,  1.25it/s]
3|4597|Loss: 0.7789745926856995:  71%|███████   | 4596/6470 [52:50<24:59,  1.25it/s]
3|4597|Loss: 0.7789745926856995:  71%|███████   | 4597/6470 [52:51<25:57,  1.20it/s]
3|4598|Loss: 0.7729101181030273:  71%|███████   | 4597/6470 [52:51<25:57,  1.20it/s]
3|4598|Loss: 0.7729101181030273:  71%|███████   | 4598/6470 [52:51<25:10,  1.24it/s]
3|4599|Loss: 0.7644093036651611:  71%|███████   | 4598/6470 [52:52<25:10,  1.24it/s]
3|4599|Loss: 0.7644093036651611:  71%|███████   | 4599/6470 [52:52<24:50,  1.26it/s]
3|4600|Loss: 0.8301568031311035:  71%|███████   | 4599/6470 [52:53<24:50,  1.26it/s]
3|4600|Loss: 0.8301568031311035:  71%|███████   | 4600/6470 [52:53<23:16,  1.34it/s]
3|4601|Loss: 0.8478307127952576:  71%|███████   | 4600/6470 [52:53<23:16,  1.34it/s]
3|4601|Loss: 0.8478307127952576:  71%|███████   | 4601/6470 [52:54<23:22,  1.33it/s]
3|4602|Loss: 0.8669175505638123:  71%|███████   | 4601/6470 [52:54<23:22,  1.33it/s]
3|4602|Loss: 0.8669175505638123:  71%|███████   | 4602/6470 [52:54<23:30,  1.32it/s]
3|4603|Loss: 0.7563997507095337:  71%|███████   | 4602/6470 [52:55<23:30,  1.32it/s]
3|4603|Loss: 0.7563997507095337:  71%|███████   | 4603/6470 [52:55<21:49,  1.43it/s]
3|4604|Loss: 0.9727127552032471:  71%|███████   | 4603/6470 [52:55<21:49,  1.43it/s]
3|4604|Loss: 0.9727127552032471:  71%|███████   | 4604/6470 [52:56<21:28,  1.45it/s]
3|4605|Loss: 0.8555230498313904:  71%|███████   | 4604/6470 [52:56<21:28,  1.45it/s]
3|4605|Loss: 0.8555230498313904:  71%|███████   | 4605/6470 [52:56<21:07,  1.47it/s]
3|4606|Loss: 0.829776406288147:  71%|███████   | 4605/6470 [52:57<21:07,  1.47it/s] 
3|4606|Loss: 0.829776406288147:  71%|███████   | 4606/6470 [52:57<20:31,  1.51it/s]
3|4607|Loss: 0.9600739479064941:  71%|███████   | 4606/6470 [52:57<20:31,  1.51it/s]
3|4607|Loss: 0.9600739479064941:  71%|███████   | 4607/6470 [52:58<21:04,  1.47it/s]
3|4608|Loss: 0.882084310054779:  71%|███████   | 4607/6470 [52:58<21:04,  1.47it/s] 
3|4608|Loss: 0.882084310054779:  71%|███████   | 4608/6470 [52:58<20:27,  1.52it/s]
3|4609|Loss: 0.841229259967804:  71%|███████   | 4608/6470 [52:59<20:27,  1.52it/s]
3|4609|Loss: 0.841229259967804:  71%|███████   | 4609/6470 [52:59<20:36,  1.51it/s]
3|4610|Loss: 0.9405117034912109:  71%|███████   | 4609/6470 [52:59<20:36,  1.51it/s]
3|4610|Loss: 0.9405117034912109:  71%|███████▏  | 4610/6470 [53:00<20:56,  1.48it/s]
3|4611|Loss: 0.9021497368812561:  71%|███████▏  | 4610/6470 [53:00<20:56,  1.48it/s]
3|4611|Loss: 0.9021497368812561:  71%|███████▏  | 4611/6470 [53:00<21:48,  1.42it/s]
3|4612|Loss: 0.7229466438293457:  71%|███████▏  | 4611/6470 [53:01<21:48,  1.42it/s]
3|4612|Loss: 0.7229466438293457:  71%|███████▏  | 4612/6470 [53:01<22:13,  1.39it/s]
3|4613|Loss: 0.8540297746658325:  71%|███████▏  | 4612/6470 [53:02<22:13,  1.39it/s]
3|4613|Loss: 0.8540297746658325:  71%|███████▏  | 4613/6470 [53:02<22:20,  1.39it/s]
3|4614|Loss: 0.9369034171104431:  71%|███████▏  | 4613/6470 [53:02<22:20,  1.39it/s]
3|4614|Loss: 0.9369034171104431:  71%|███████▏  | 4614/6470 [53:03<21:47,  1.42it/s]
3|4615|Loss: 0.7839824557304382:  71%|███████▏  | 4614/6470 [53:03<21:47,  1.42it/s]
3|4615|Loss: 0.7839824557304382:  71%|███████▏  | 4615/6470 [53:03<21:04,  1.47it/s]
3|4616|Loss: 0.9328532814979553:  71%|███████▏  | 4615/6470 [53:03<21:04,  1.47it/s]
3|4616|Loss: 0.9328532814979553:  71%|███████▏  | 4616/6470 [53:04<20:25,  1.51it/s]
3|4617|Loss: 0.8274064660072327:  71%|███████▏  | 4616/6470 [53:04<20:25,  1.51it/s]
3|4617|Loss: 0.8274064660072327:  71%|███████▏  | 4617/6470 [53:04<20:36,  1.50it/s]
3|4618|Loss: 0.9296560883522034:  71%|███████▏  | 4617/6470 [53:05<20:36,  1.50it/s]
3|4618|Loss: 0.9296560883522034:  71%|███████▏  | 4618/6470 [53:05<21:33,  1.43it/s]
3|4619|Loss: 0.9103058576583862:  71%|███████▏  | 4618/6470 [53:06<21:33,  1.43it/s]
3|4619|Loss: 0.9103058576583862:  71%|███████▏  | 4619/6470 [53:06<22:10,  1.39it/s]
3|4620|Loss: 0.690886914730072:  71%|███████▏  | 4619/6470 [53:06<22:10,  1.39it/s] 
3|4620|Loss: 0.690886914730072:  71%|███████▏  | 4620/6470 [53:07<21:33,  1.43it/s]
3|4621|Loss: 0.7888829708099365:  71%|███████▏  | 4620/6470 [53:07<21:33,  1.43it/s]
3|4621|Loss: 0.7888829708099365:  71%|███████▏  | 4621/6470 [53:07<20:47,  1.48it/s]
3|4622|Loss: 0.8400619626045227:  71%|███████▏  | 4621/6470 [53:08<20:47,  1.48it/s]
3|4622|Loss: 0.8400619626045227:  71%|███████▏  | 4622/6470 [53:08<20:27,  1.50it/s]
3|4623|Loss: 0.8061739206314087:  71%|███████▏  | 4622/6470 [53:08<20:27,  1.50it/s]
3|4623|Loss: 0.8061739206314087:  71%|███████▏  | 4623/6470 [53:08<19:53,  1.55it/s]
3|4624|Loss: 0.9496403336524963:  71%|███████▏  | 4623/6470 [53:09<19:53,  1.55it/s]
3|4624|Loss: 0.9496403336524963:  71%|███████▏  | 4624/6470 [53:09<20:35,  1.49it/s]
3|4625|Loss: 1.0482879877090454:  71%|███████▏  | 4624/6470 [53:10<20:35,  1.49it/s]
3|4625|Loss: 1.0482879877090454:  71%|███████▏  | 4625/6470 [53:10<21:58,  1.40it/s]
3|4626|Loss: 0.9874332547187805:  71%|███████▏  | 4625/6470 [53:11<21:58,  1.40it/s]
3|4626|Loss: 0.9874332547187805:  71%|███████▏  | 4626/6470 [53:11<23:25,  1.31it/s]
3|4627|Loss: 0.7806246280670166:  71%|███████▏  | 4626/6470 [53:11<23:25,  1.31it/s]
3|4627|Loss: 0.7806246280670166:  72%|███████▏  | 4627/6470 [53:12<23:29,  1.31it/s]
3|4628|Loss: 0.8547379374504089:  72%|███████▏  | 4627/6470 [53:12<23:29,  1.31it/s]
3|4628|Loss: 0.8547379374504089:  72%|███████▏  | 4628/6470 [53:12<23:54,  1.28it/s]
3|4629|Loss: 1.1024751663208008:  72%|███████▏  | 4628/6470 [53:13<23:54,  1.28it/s]
3|4629|Loss: 1.1024751663208008:  72%|███████▏  | 4629/6470 [53:13<24:34,  1.25it/s]
3|4630|Loss: 0.7231382727622986:  72%|███████▏  | 4629/6470 [53:14<24:34,  1.25it/s]
3|4630|Loss: 0.7231382727622986:  72%|███████▏  | 4630/6470 [53:14<24:10,  1.27it/s]
3|4631|Loss: 0.7648831009864807:  72%|███████▏  | 4630/6470 [53:15<24:10,  1.27it/s]
3|4631|Loss: 0.7648831009864807:  72%|███████▏  | 4631/6470 [53:15<25:22,  1.21it/s]
3|4632|Loss: 0.9011890888214111:  72%|███████▏  | 4631/6470 [53:15<25:22,  1.21it/s]
3|4632|Loss: 0.9011890888214111:  72%|███████▏  | 4632/6470 [53:16<23:32,  1.30it/s]
3|4633|Loss: 0.9320783615112305:  72%|███████▏  | 4632/6470 [53:16<23:32,  1.30it/s]
3|4633|Loss: 0.9320783615112305:  72%|███████▏  | 4633/6470 [53:16<23:44,  1.29it/s]
3|4634|Loss: 0.8357223868370056:  72%|███████▏  | 4633/6470 [53:17<23:44,  1.29it/s]
3|4634|Loss: 0.8357223868370056:  72%|███████▏  | 4634/6470 [53:17<21:32,  1.42it/s]
3|4635|Loss: 0.8555585145950317:  72%|███████▏  | 4634/6470 [53:17<21:32,  1.42it/s]
3|4635|Loss: 0.8555585145950317:  72%|███████▏  | 4635/6470 [53:18<20:34,  1.49it/s]
3|4636|Loss: 0.7655719518661499:  72%|███████▏  | 4635/6470 [53:18<20:34,  1.49it/s]
3|4636|Loss: 0.7655719518661499:  72%|███████▏  | 4636/6470 [53:18<20:18,  1.50it/s]
3|4637|Loss: 0.8510010242462158:  72%|███████▏  | 4636/6470 [53:19<20:18,  1.50it/s]
3|4637|Loss: 0.8510010242462158:  72%|███████▏  | 4637/6470 [53:19<20:31,  1.49it/s]
3|4638|Loss: 0.8229947686195374:  72%|███████▏  | 4637/6470 [53:19<20:31,  1.49it/s]
3|4638|Loss: 0.8229947686195374:  72%|███████▏  | 4638/6470 [53:20<22:07,  1.38it/s]
3|4639|Loss: 0.9169203639030457:  72%|███████▏  | 4638/6470 [53:20<22:07,  1.38it/s]
3|4639|Loss: 0.9169203639030457:  72%|███████▏  | 4639/6470 [53:21<22:37,  1.35it/s]
3|4640|Loss: 0.8949979543685913:  72%|███████▏  | 4639/6470 [53:21<22:37,  1.35it/s]
3|4640|Loss: 0.8949979543685913:  72%|███████▏  | 4640/6470 [53:21<21:04,  1.45it/s]
3|4641|Loss: 0.6473785638809204:  72%|███████▏  | 4640/6470 [53:21<21:04,  1.45it/s]
3|4641|Loss: 0.6473785638809204:  72%|███████▏  | 4641/6470 [53:22<19:36,  1.55it/s]
3|4642|Loss: 0.8748392462730408:  72%|███████▏  | 4641/6470 [53:22<19:36,  1.55it/s]
3|4642|Loss: 0.8748392462730408:  72%|███████▏  | 4642/6470 [53:22<20:02,  1.52it/s]
3|4643|Loss: 0.8679255843162537:  72%|███████▏  | 4642/6470 [53:23<20:02,  1.52it/s]
3|4643|Loss: 0.8679255843162537:  72%|███████▏  | 4643/6470 [53:23<20:52,  1.46it/s]
3|4644|Loss: 0.880174458026886:  72%|███████▏  | 4643/6470 [53:23<20:52,  1.46it/s] 
3|4644|Loss: 0.880174458026886:  72%|███████▏  | 4644/6470 [53:24<20:36,  1.48it/s]
3|4645|Loss: 0.8726981282234192:  72%|███████▏  | 4644/6470 [53:24<20:36,  1.48it/s]
3|4645|Loss: 0.8726981282234192:  72%|███████▏  | 4645/6470 [53:24<18:52,  1.61it/s]
3|4646|Loss: 0.9930883049964905:  72%|███████▏  | 4645/6470 [53:25<18:52,  1.61it/s]
3|4646|Loss: 0.9930883049964905:  72%|███████▏  | 4646/6470 [53:25<18:39,  1.63it/s]
3|4647|Loss: 0.9205095767974854:  72%|███████▏  | 4646/6470 [53:25<18:39,  1.63it/s]
3|4647|Loss: 0.9205095767974854:  72%|███████▏  | 4647/6470 [53:25<18:26,  1.65it/s]
3|4648|Loss: 0.8688082098960876:  72%|███████▏  | 4647/6470 [53:26<18:26,  1.65it/s]
3|4648|Loss: 0.8688082098960876:  72%|███████▏  | 4648/6470 [53:26<18:27,  1.64it/s]
3|4649|Loss: 0.9012031555175781:  72%|███████▏  | 4648/6470 [53:26<18:27,  1.64it/s]
3|4649|Loss: 0.9012031555175781:  72%|███████▏  | 4649/6470 [53:27<19:24,  1.56it/s]
3|4650|Loss: 0.7446413636207581:  72%|███████▏  | 4649/6470 [53:27<19:24,  1.56it/s]
3|4650|Loss: 0.7446413636207581:  72%|███████▏  | 4650/6470 [53:28<21:09,  1.43it/s]
3|4651|Loss: 0.8721596002578735:  72%|███████▏  | 4650/6470 [53:28<21:09,  1.43it/s]
3|4651|Loss: 0.8721596002578735:  72%|███████▏  | 4651/6470 [53:28<21:00,  1.44it/s]
3|4652|Loss: 0.7265838384628296:  72%|███████▏  | 4651/6470 [53:29<21:00,  1.44it/s]
3|4652|Loss: 0.7265838384628296:  72%|███████▏  | 4652/6470 [53:29<21:20,  1.42it/s]
3|4653|Loss: 0.7474611401557922:  72%|███████▏  | 4652/6470 [53:29<21:20,  1.42it/s]
3|4653|Loss: 0.7474611401557922:  72%|███████▏  | 4653/6470 [53:30<20:35,  1.47it/s]
3|4654|Loss: 0.7443689703941345:  72%|███████▏  | 4653/6470 [53:30<20:35,  1.47it/s]
3|4654|Loss: 0.7443689703941345:  72%|███████▏  | 4654/6470 [53:30<20:47,  1.46it/s]
3|4655|Loss: 0.8151041865348816:  72%|███████▏  | 4654/6470 [53:31<20:47,  1.46it/s]
3|4655|Loss: 0.8151041865348816:  72%|███████▏  | 4655/6470 [53:31<20:22,  1.48it/s]
3|4656|Loss: 0.9139161109924316:  72%|███████▏  | 4655/6470 [53:31<20:22,  1.48it/s]
3|4656|Loss: 0.9139161109924316:  72%|███████▏  | 4656/6470 [53:32<20:40,  1.46it/s]
3|4657|Loss: 0.8155737519264221:  72%|███████▏  | 4656/6470 [53:32<20:40,  1.46it/s]
3|4657|Loss: 0.8155737519264221:  72%|███████▏  | 4657/6470 [53:32<21:07,  1.43it/s]
3|4658|Loss: 1.0357861518859863:  72%|███████▏  | 4657/6470 [53:33<21:07,  1.43it/s]
3|4658|Loss: 1.0357861518859863:  72%|███████▏  | 4658/6470 [53:33<22:05,  1.37it/s]
3|4659|Loss: 0.9319137334823608:  72%|███████▏  | 4658/6470 [53:34<22:05,  1.37it/s]
3|4659|Loss: 0.9319137334823608:  72%|███████▏  | 4659/6470 [53:34<22:46,  1.32it/s]
3|4660|Loss: 0.9697213768959045:  72%|███████▏  | 4659/6470 [53:34<22:46,  1.32it/s]
3|4660|Loss: 0.9697213768959045:  72%|███████▏  | 4660/6470 [53:35<23:49,  1.27it/s]
3|4661|Loss: 0.7992984056472778:  72%|███████▏  | 4660/6470 [53:35<23:49,  1.27it/s]
3|4661|Loss: 0.7992984056472778:  72%|███████▏  | 4661/6470 [53:36<23:42,  1.27it/s]
3|4662|Loss: 0.7277357578277588:  72%|███████▏  | 4661/6470 [53:36<23:42,  1.27it/s]
3|4662|Loss: 0.7277357578277588:  72%|███████▏  | 4662/6470 [53:36<21:42,  1.39it/s]
3|4663|Loss: 0.775958240032196:  72%|███████▏  | 4662/6470 [53:37<21:42,  1.39it/s] 
3|4663|Loss: 0.775958240032196:  72%|███████▏  | 4663/6470 [53:37<20:49,  1.45it/s]
3|4664|Loss: 0.7435139417648315:  72%|███████▏  | 4663/6470 [53:37<20:49,  1.45it/s]
3|4664|Loss: 0.7435139417648315:  72%|███████▏  | 4664/6470 [53:38<20:49,  1.44it/s]
3|4665|Loss: 0.9430798888206482:  72%|███████▏  | 4664/6470 [53:38<20:49,  1.44it/s]
3|4665|Loss: 0.9430798888206482:  72%|███████▏  | 4665/6470 [53:38<20:28,  1.47it/s]
3|4666|Loss: 0.8617036938667297:  72%|███████▏  | 4665/6470 [53:39<20:28,  1.47it/s]
3|4666|Loss: 0.8617036938667297:  72%|███████▏  | 4666/6470 [53:39<20:35,  1.46it/s]
3|4667|Loss: 0.7355062365531921:  72%|███████▏  | 4666/6470 [53:39<20:35,  1.46it/s]
3|4667|Loss: 0.7355062365531921:  72%|███████▏  | 4667/6470 [53:40<20:58,  1.43it/s]
3|4668|Loss: 0.745246410369873:  72%|███████▏  | 4667/6470 [53:40<20:58,  1.43it/s] 
3|4668|Loss: 0.745246410369873:  72%|███████▏  | 4668/6470 [53:40<20:37,  1.46it/s]
3|4669|Loss: 0.7436581254005432:  72%|███████▏  | 4668/6470 [53:41<20:37,  1.46it/s]
3|4669|Loss: 0.7436581254005432:  72%|███████▏  | 4669/6470 [53:41<20:00,  1.50it/s]
3|4670|Loss: 0.8675541281700134:  72%|███████▏  | 4669/6470 [53:41<20:00,  1.50it/s]
3|4670|Loss: 0.8675541281700134:  72%|███████▏  | 4670/6470 [53:42<21:28,  1.40it/s]
3|4671|Loss: 0.7608606219291687:  72%|███████▏  | 4670/6470 [53:42<21:28,  1.40it/s]
3|4671|Loss: 0.7608606219291687:  72%|███████▏  | 4671/6470 [53:43<23:19,  1.29it/s]
3|4672|Loss: 0.78708815574646:  72%|███████▏  | 4671/6470 [53:43<23:19,  1.29it/s]  
3|4672|Loss: 0.78708815574646:  72%|███████▏  | 4672/6470 [53:43<23:20,  1.28it/s]
3|4673|Loss: 0.8577262163162231:  72%|███████▏  | 4672/6470 [53:44<23:20,  1.28it/s]
3|4673|Loss: 0.8577262163162231:  72%|███████▏  | 4673/6470 [53:44<23:48,  1.26it/s]
3|4674|Loss: 0.9039196372032166:  72%|███████▏  | 4673/6470 [53:45<23:48,  1.26it/s]
3|4674|Loss: 0.9039196372032166:  72%|███████▏  | 4674/6470 [53:45<23:37,  1.27it/s]
3|4675|Loss: 0.9753554463386536:  72%|███████▏  | 4674/6470 [53:45<23:37,  1.27it/s]
3|4675|Loss: 0.9753554463386536:  72%|███████▏  | 4675/6470 [53:46<23:40,  1.26it/s]
3|4676|Loss: 0.8511642217636108:  72%|███████▏  | 4675/6470 [53:46<23:40,  1.26it/s]
3|4676|Loss: 0.8511642217636108:  72%|███████▏  | 4676/6470 [53:47<23:01,  1.30it/s]
3|4677|Loss: 0.8795300722122192:  72%|███████▏  | 4676/6470 [53:47<23:01,  1.30it/s]
3|4677|Loss: 0.8795300722122192:  72%|███████▏  | 4677/6470 [53:47<20:59,  1.42it/s]
3|4678|Loss: 0.8499451875686646:  72%|███████▏  | 4677/6470 [53:47<20:59,  1.42it/s]
3|4678|Loss: 0.8499451875686646:  72%|███████▏  | 4678/6470 [53:48<19:45,  1.51it/s]
3|4679|Loss: 0.6692455410957336:  72%|███████▏  | 4678/6470 [53:48<19:45,  1.51it/s]
3|4679|Loss: 0.6692455410957336:  72%|███████▏  | 4679/6470 [53:48<20:53,  1.43it/s]
3|4680|Loss: 0.9493101835250854:  72%|███████▏  | 4679/6470 [53:49<20:53,  1.43it/s]
3|4680|Loss: 0.9493101835250854:  72%|███████▏  | 4680/6470 [53:49<22:10,  1.34it/s]
3|4681|Loss: 0.9018386602401733:  72%|███████▏  | 4680/6470 [53:50<22:10,  1.34it/s]
3|4681|Loss: 0.9018386602401733:  72%|███████▏  | 4681/6470 [53:50<20:50,  1.43it/s]
3|4682|Loss: 0.8412527441978455:  72%|███████▏  | 4681/6470 [53:50<20:50,  1.43it/s]
3|4682|Loss: 0.8412527441978455:  72%|███████▏  | 4682/6470 [53:51<20:06,  1.48it/s]
3|4683|Loss: 0.8799298405647278:  72%|███████▏  | 4682/6470 [53:51<20:06,  1.48it/s]
3|4683|Loss: 0.8799298405647278:  72%|███████▏  | 4683/6470 [53:51<20:28,  1.45it/s]
3|4684|Loss: 0.9238011240959167:  72%|███████▏  | 4683/6470 [53:52<20:28,  1.45it/s]
3|4684|Loss: 0.9238011240959167:  72%|███████▏  | 4684/6470 [53:52<20:18,  1.47it/s]
3|4685|Loss: 0.9494751691818237:  72%|███████▏  | 4684/6470 [53:52<20:18,  1.47it/s]
3|4685|Loss: 0.9494751691818237:  72%|███████▏  | 4685/6470 [53:53<19:55,  1.49it/s]
3|4686|Loss: 0.8465818762779236:  72%|███████▏  | 4685/6470 [53:53<19:55,  1.49it/s]
3|4686|Loss: 0.8465818762779236:  72%|███████▏  | 4686/6470 [53:53<20:26,  1.45it/s]
3|4687|Loss: 0.939544677734375:  72%|███████▏  | 4686/6470 [53:54<20:26,  1.45it/s] 
3|4687|Loss: 0.939544677734375:  72%|███████▏  | 4687/6470 [53:54<21:36,  1.38it/s]
3|4688|Loss: 0.8701545000076294:  72%|███████▏  | 4687/6470 [53:55<21:36,  1.38it/s]
3|4688|Loss: 0.8701545000076294:  72%|███████▏  | 4688/6470 [53:55<22:36,  1.31it/s]
3|4689|Loss: 1.0182461738586426:  72%|███████▏  | 4688/6470 [53:55<22:36,  1.31it/s]
3|4689|Loss: 1.0182461738586426:  72%|███████▏  | 4689/6470 [53:56<22:23,  1.33it/s]
3|4690|Loss: 0.8749181628227234:  72%|███████▏  | 4689/6470 [53:56<22:23,  1.33it/s]
3|4690|Loss: 0.8749181628227234:  72%|███████▏  | 4690/6470 [53:56<20:52,  1.42it/s]
3|4691|Loss: 0.9142802953720093:  72%|███████▏  | 4690/6470 [53:57<20:52,  1.42it/s]
3|4691|Loss: 0.9142802953720093:  73%|███████▎  | 4691/6470 [53:57<21:30,  1.38it/s]
3|4692|Loss: 0.7856045961380005:  73%|███████▎  | 4691/6470 [53:57<21:30,  1.38it/s]
3|4692|Loss: 0.7856045961380005:  73%|███████▎  | 4692/6470 [53:58<21:07,  1.40it/s]
3|4693|Loss: 0.7613310217857361:  73%|███████▎  | 4692/6470 [53:58<21:07,  1.40it/s]
3|4693|Loss: 0.7613310217857361:  73%|███████▎  | 4693/6470 [53:58<19:40,  1.50it/s]
3|4694|Loss: 0.9247888326644897:  73%|███████▎  | 4693/6470 [53:59<19:40,  1.50it/s]
3|4694|Loss: 0.9247888326644897:  73%|███████▎  | 4694/6470 [53:59<19:16,  1.54it/s]
3|4695|Loss: 0.796522855758667:  73%|███████▎  | 4694/6470 [53:59<19:16,  1.54it/s] 
3|4695|Loss: 0.796522855758667:  73%|███████▎  | 4695/6470 [53:59<17:52,  1.66it/s]
3|4696|Loss: 0.8301668167114258:  73%|███████▎  | 4695/6470 [54:00<17:52,  1.66it/s]
3|4696|Loss: 0.8301668167114258:  73%|███████▎  | 4696/6470 [54:00<19:11,  1.54it/s]
3|4697|Loss: 0.8574393391609192:  73%|███████▎  | 4696/6470 [54:01<19:11,  1.54it/s]
3|4697|Loss: 0.8574393391609192:  73%|███████▎  | 4697/6470 [54:01<19:55,  1.48it/s]
3|4698|Loss: 0.9032992720603943:  73%|███████▎  | 4697/6470 [54:01<19:55,  1.48it/s]
3|4698|Loss: 0.9032992720603943:  73%|███████▎  | 4698/6470 [54:02<20:08,  1.47it/s]
3|4699|Loss: 0.8043379187583923:  73%|███████▎  | 4698/6470 [54:02<20:08,  1.47it/s]
3|4699|Loss: 0.8043379187583923:  73%|███████▎  | 4699/6470 [54:02<21:19,  1.38it/s]
3|4700|Loss: 0.8160955905914307:  73%|███████▎  | 4699/6470 [54:03<21:19,  1.38it/s]
3|4700|Loss: 0.8160955905914307:  73%|███████▎  | 4700/6470 [54:03<21:42,  1.36it/s]
3|4701|Loss: 0.7740083932876587:  73%|███████▎  | 4700/6470 [54:04<21:42,  1.36it/s]
3|4701|Loss: 0.7740083932876587:  73%|███████▎  | 4701/6470 [54:04<22:21,  1.32it/s]
3|4702|Loss: 0.844363808631897:  73%|███████▎  | 4701/6470 [54:04<22:21,  1.32it/s] 
3|4702|Loss: 0.844363808631897:  73%|███████▎  | 4702/6470 [54:05<22:18,  1.32it/s]
3|4703|Loss: 0.9429554343223572:  73%|███████▎  | 4702/6470 [54:05<22:18,  1.32it/s]
3|4703|Loss: 0.9429554343223572:  73%|███████▎  | 4703/6470 [54:05<21:21,  1.38it/s]
3|4704|Loss: 0.8767891526222229:  73%|███████▎  | 4703/6470 [54:06<21:21,  1.38it/s]
3|4704|Loss: 0.8767891526222229:  73%|███████▎  | 4704/6470 [54:06<19:25,  1.51it/s]
3|4705|Loss: 0.856132447719574:  73%|███████▎  | 4704/6470 [54:06<19:25,  1.51it/s] 
3|4705|Loss: 0.856132447719574:  73%|███████▎  | 4705/6470 [54:06<18:01,  1.63it/s]
3|4706|Loss: 0.8379926681518555:  73%|███████▎  | 4705/6470 [54:07<18:01,  1.63it/s]
3|4706|Loss: 0.8379926681518555:  73%|███████▎  | 4706/6470 [54:07<18:49,  1.56it/s]
3|4707|Loss: 0.9734994769096375:  73%|███████▎  | 4706/6470 [54:08<18:49,  1.56it/s]
3|4707|Loss: 0.9734994769096375:  73%|███████▎  | 4707/6470 [54:08<19:31,  1.50it/s]
3|4708|Loss: 0.9080617427825928:  73%|███████▎  | 4707/6470 [54:08<19:31,  1.50it/s]
3|4708|Loss: 0.9080617427825928:  73%|███████▎  | 4708/6470 [54:08<19:15,  1.53it/s]
3|4709|Loss: 0.8238845467567444:  73%|███████▎  | 4708/6470 [54:09<19:15,  1.53it/s]
3|4709|Loss: 0.8238845467567444:  73%|███████▎  | 4709/6470 [54:09<19:47,  1.48it/s]
3|4710|Loss: 0.7582964301109314:  73%|███████▎  | 4709/6470 [54:10<19:47,  1.48it/s]
3|4710|Loss: 0.7582964301109314:  73%|███████▎  | 4710/6470 [54:10<20:08,  1.46it/s]
3|4711|Loss: 0.8569505214691162:  73%|███████▎  | 4710/6470 [54:10<20:08,  1.46it/s]
3|4711|Loss: 0.8569505214691162:  73%|███████▎  | 4711/6470 [54:11<19:51,  1.48it/s]
3|4712|Loss: 0.8905884623527527:  73%|███████▎  | 4711/6470 [54:11<19:51,  1.48it/s]
3|4712|Loss: 0.8905884623527527:  73%|███████▎  | 4712/6470 [54:11<19:19,  1.52it/s]
3|4713|Loss: 0.8988080620765686:  73%|███████▎  | 4712/6470 [54:11<19:19,  1.52it/s]
3|4713|Loss: 0.8988080620765686:  73%|███████▎  | 4713/6470 [54:12<17:47,  1.65it/s]
3|4714|Loss: 0.9027187824249268:  73%|███████▎  | 4713/6470 [54:12<17:47,  1.65it/s]
3|4714|Loss: 0.9027187824249268:  73%|███████▎  | 4714/6470 [54:12<17:20,  1.69it/s]
3|4715|Loss: 0.9812808632850647:  73%|███████▎  | 4714/6470 [54:13<17:20,  1.69it/s]
3|4715|Loss: 0.9812808632850647:  73%|███████▎  | 4715/6470 [54:13<18:59,  1.54it/s]
3|4716|Loss: 0.8843458890914917:  73%|███████▎  | 4715/6470 [54:13<18:59,  1.54it/s]
3|4716|Loss: 0.8843458890914917:  73%|███████▎  | 4716/6470 [54:14<20:34,  1.42it/s]
3|4717|Loss: 0.8605244755744934:  73%|███████▎  | 4716/6470 [54:14<20:34,  1.42it/s]
3|4717|Loss: 0.8605244755744934:  73%|███████▎  | 4717/6470 [54:15<21:07,  1.38it/s]
3|4718|Loss: 0.8806893825531006:  73%|███████▎  | 4717/6470 [54:15<21:07,  1.38it/s]
3|4718|Loss: 0.8806893825531006:  73%|███████▎  | 4718/6470 [54:15<20:40,  1.41it/s]
3|4719|Loss: 0.8595642447471619:  73%|███████▎  | 4718/6470 [54:16<20:40,  1.41it/s]
3|4719|Loss: 0.8595642447471619:  73%|███████▎  | 4719/6470 [54:16<20:33,  1.42it/s]
3|4720|Loss: 0.7426325082778931:  73%|███████▎  | 4719/6470 [54:16<20:33,  1.42it/s]
3|4720|Loss: 0.7426325082778931:  73%|███████▎  | 4720/6470 [54:17<20:05,  1.45it/s]
3|4721|Loss: 0.8352963924407959:  73%|███████▎  | 4720/6470 [54:17<20:05,  1.45it/s]
3|4721|Loss: 0.8352963924407959:  73%|███████▎  | 4721/6470 [54:17<19:12,  1.52it/s]
3|4722|Loss: 0.8925737738609314:  73%|███████▎  | 4721/6470 [54:18<19:12,  1.52it/s]
3|4722|Loss: 0.8925737738609314:  73%|███████▎  | 4722/6470 [54:18<18:47,  1.55it/s]
3|4723|Loss: 0.8117864727973938:  73%|███████▎  | 4722/6470 [54:18<18:47,  1.55it/s]
3|4723|Loss: 0.8117864727973938:  73%|███████▎  | 4723/6470 [54:18<18:39,  1.56it/s]
3|4724|Loss: 0.955819845199585:  73%|███████▎  | 4723/6470 [54:19<18:39,  1.56it/s] 
3|4724|Loss: 0.955819845199585:  73%|███████▎  | 4724/6470 [54:19<18:36,  1.56it/s]
3|4725|Loss: 0.8259096145629883:  73%|███████▎  | 4724/6470 [54:19<18:36,  1.56it/s]
3|4725|Loss: 0.8259096145629883:  73%|███████▎  | 4725/6470 [54:20<19:45,  1.47it/s]
3|4726|Loss: 0.77058345079422:  73%|███████▎  | 4725/6470 [54:20<19:45,  1.47it/s]  
3|4726|Loss: 0.77058345079422:  73%|███████▎  | 4726/6470 [54:21<20:08,  1.44it/s]
3|4727|Loss: 0.8536693453788757:  73%|███████▎  | 4726/6470 [54:21<20:08,  1.44it/s]
3|4727|Loss: 0.8536693453788757:  73%|███████▎  | 4727/6470 [54:21<20:23,  1.42it/s]
3|4728|Loss: 1.005279779434204:  73%|███████▎  | 4727/6470 [54:22<20:23,  1.42it/s] 
3|4728|Loss: 1.005279779434204:  73%|███████▎  | 4728/6470 [54:22<21:26,  1.35it/s]
3|4729|Loss: 0.9698975086212158:  73%|███████▎  | 4728/6470 [54:23<21:26,  1.35it/s]
3|4729|Loss: 0.9698975086212158:  73%|███████▎  | 4729/6470 [54:23<20:51,  1.39it/s]
3|4730|Loss: 0.8893476128578186:  73%|███████▎  | 4729/6470 [54:23<20:51,  1.39it/s]
3|4730|Loss: 0.8893476128578186:  73%|███████▎  | 4730/6470 [54:23<20:34,  1.41it/s]
3|4731|Loss: 0.751535177230835:  73%|███████▎  | 4730/6470 [54:24<20:34,  1.41it/s] 
3|4731|Loss: 0.751535177230835:  73%|███████▎  | 4731/6470 [54:24<20:44,  1.40it/s]
3|4732|Loss: 1.0497329235076904:  73%|███████▎  | 4731/6470 [54:25<20:44,  1.40it/s]
3|4732|Loss: 1.0497329235076904:  73%|███████▎  | 4732/6470 [54:25<20:55,  1.38it/s]
3|4733|Loss: 0.7730040550231934:  73%|███████▎  | 4732/6470 [54:25<20:55,  1.38it/s]
3|4733|Loss: 0.7730040550231934:  73%|███████▎  | 4733/6470 [54:26<20:51,  1.39it/s]
3|4734|Loss: 0.7804270386695862:  73%|███████▎  | 4733/6470 [54:26<20:51,  1.39it/s]
3|4734|Loss: 0.7804270386695862:  73%|███████▎  | 4734/6470 [54:26<20:59,  1.38it/s]
3|4735|Loss: 0.9038133025169373:  73%|███████▎  | 4734/6470 [54:27<20:59,  1.38it/s]
3|4735|Loss: 0.9038133025169373:  73%|███████▎  | 4735/6470 [54:27<20:27,  1.41it/s]
3|4736|Loss: 0.7952364683151245:  73%|███████▎  | 4735/6470 [54:27<20:27,  1.41it/s]
3|4736|Loss: 0.7952364683151245:  73%|███████▎  | 4736/6470 [54:28<19:40,  1.47it/s]
3|4737|Loss: 0.712520182132721:  73%|███████▎  | 4736/6470 [54:28<19:40,  1.47it/s] 
3|4737|Loss: 0.712520182132721:  73%|███████▎  | 4737/6470 [54:28<19:17,  1.50it/s]
3|4738|Loss: 0.7913542985916138:  73%|███████▎  | 4737/6470 [54:29<19:17,  1.50it/s]
3|4738|Loss: 0.7913542985916138:  73%|███████▎  | 4738/6470 [54:29<19:06,  1.51it/s]
3|4739|Loss: 0.9213119745254517:  73%|███████▎  | 4738/6470 [54:29<19:06,  1.51it/s]
3|4739|Loss: 0.9213119745254517:  73%|███████▎  | 4739/6470 [54:30<19:24,  1.49it/s]
3|4740|Loss: 0.6768003702163696:  73%|███████▎  | 4739/6470 [54:30<19:24,  1.49it/s]
3|4740|Loss: 0.6768003702163696:  73%|███████▎  | 4740/6470 [54:30<20:01,  1.44it/s]
3|4741|Loss: 0.8825712203979492:  73%|███████▎  | 4740/6470 [54:31<20:01,  1.44it/s]
3|4741|Loss: 0.8825712203979492:  73%|███████▎  | 4741/6470 [54:31<18:28,  1.56it/s]
3|4742|Loss: 0.8865416646003723:  73%|███████▎  | 4741/6470 [54:31<18:28,  1.56it/s]
3|4742|Loss: 0.8865416646003723:  73%|███████▎  | 4742/6470 [54:32<17:56,  1.60it/s]
3|4743|Loss: 0.682978630065918:  73%|███████▎  | 4742/6470 [54:32<17:56,  1.60it/s] 
3|4743|Loss: 0.682978630065918:  73%|███████▎  | 4743/6470 [54:32<17:25,  1.65it/s]
3|4744|Loss: 0.8071637153625488:  73%|███████▎  | 4743/6470 [54:32<17:25,  1.65it/s]
3|4744|Loss: 0.8071637153625488:  73%|███████▎  | 4744/6470 [54:33<16:29,  1.74it/s]
3|4745|Loss: 0.8488433361053467:  73%|███████▎  | 4744/6470 [54:33<16:29,  1.74it/s]
3|4745|Loss: 0.8488433361053467:  73%|███████▎  | 4745/6470 [54:33<18:16,  1.57it/s]
3|4746|Loss: 0.8100003004074097:  73%|███████▎  | 4745/6470 [54:34<18:16,  1.57it/s]
3|4746|Loss: 0.8100003004074097:  73%|███████▎  | 4746/6470 [54:34<19:22,  1.48it/s]
3|4747|Loss: 0.895362377166748:  73%|███████▎  | 4746/6470 [54:35<19:22,  1.48it/s] 
3|4747|Loss: 0.895362377166748:  73%|███████▎  | 4747/6470 [54:35<19:15,  1.49it/s]
3|4748|Loss: 0.809719443321228:  73%|███████▎  | 4747/6470 [54:35<19:15,  1.49it/s]
3|4748|Loss: 0.809719443321228:  73%|███████▎  | 4748/6470 [54:35<18:49,  1.52it/s]
3|4749|Loss: 0.8647510409355164:  73%|███████▎  | 4748/6470 [54:36<18:49,  1.52it/s]
3|4749|Loss: 0.8647510409355164:  73%|███████▎  | 4749/6470 [54:36<18:38,  1.54it/s]
3|4750|Loss: 0.8585472106933594:  73%|███████▎  | 4749/6470 [54:36<18:38,  1.54it/s]
3|4750|Loss: 0.8585472106933594:  73%|███████▎  | 4750/6470 [54:37<20:01,  1.43it/s]
3|4751|Loss: 0.9093403220176697:  73%|███████▎  | 4750/6470 [54:37<20:01,  1.43it/s]
3|4751|Loss: 0.9093403220176697:  73%|███████▎  | 4751/6470 [54:38<19:53,  1.44it/s]
3|4752|Loss: 0.8440134525299072:  73%|███████▎  | 4751/6470 [54:38<19:53,  1.44it/s]
3|4752|Loss: 0.8440134525299072:  73%|███████▎  | 4752/6470 [54:38<20:05,  1.43it/s]
3|4753|Loss: 0.9511359930038452:  73%|███████▎  | 4752/6470 [54:39<20:05,  1.43it/s]
3|4753|Loss: 0.9511359930038452:  73%|███████▎  | 4753/6470 [54:39<20:27,  1.40it/s]
3|4754|Loss: 0.8359319567680359:  73%|███████▎  | 4753/6470 [54:39<20:27,  1.40it/s]
3|4754|Loss: 0.8359319567680359:  73%|███████▎  | 4754/6470 [54:40<20:02,  1.43it/s]
3|4755|Loss: 0.7635496854782104:  73%|███████▎  | 4754/6470 [54:40<20:02,  1.43it/s]
3|4755|Loss: 0.7635496854782104:  73%|███████▎  | 4755/6470 [54:40<20:36,  1.39it/s]
3|4756|Loss: 0.9649666547775269:  73%|███████▎  | 4755/6470 [54:41<20:36,  1.39it/s]
3|4756|Loss: 0.9649666547775269:  74%|███████▎  | 4756/6470 [54:41<20:50,  1.37it/s]
3|4757|Loss: 0.8366045355796814:  74%|███████▎  | 4756/6470 [54:42<20:50,  1.37it/s]
3|4757|Loss: 0.8366045355796814:  74%|███████▎  | 4757/6470 [54:42<20:17,  1.41it/s]
3|4758|Loss: 0.9847621917724609:  74%|███████▎  | 4757/6470 [54:42<20:17,  1.41it/s]
3|4758|Loss: 0.9847621917724609:  74%|███████▎  | 4758/6470 [54:43<20:03,  1.42it/s]
3|4759|Loss: 0.9450536370277405:  74%|███████▎  | 4758/6470 [54:43<20:03,  1.42it/s]
3|4759|Loss: 0.9450536370277405:  74%|███████▎  | 4759/6470 [54:43<18:18,  1.56it/s]
3|4760|Loss: 0.7291787266731262:  74%|███████▎  | 4759/6470 [54:43<18:18,  1.56it/s]
3|4760|Loss: 0.7291787266731262:  74%|███████▎  | 4760/6470 [54:44<17:46,  1.60it/s]
3|4761|Loss: 0.7901504635810852:  74%|███████▎  | 4760/6470 [54:44<17:46,  1.60it/s]
3|4761|Loss: 0.7901504635810852:  74%|███████▎  | 4761/6470 [54:44<19:20,  1.47it/s]
3|4762|Loss: 0.7906946539878845:  74%|███████▎  | 4761/6470 [54:45<19:20,  1.47it/s]
3|4762|Loss: 0.7906946539878845:  74%|███████▎  | 4762/6470 [54:45<20:52,  1.36it/s]
3|4763|Loss: 0.9121223092079163:  74%|███████▎  | 4762/6470 [54:46<20:52,  1.36it/s]
3|4763|Loss: 0.9121223092079163:  74%|███████▎  | 4763/6470 [54:46<21:51,  1.30it/s]
3|4764|Loss: 0.7402034997940063:  74%|███████▎  | 4763/6470 [54:47<21:51,  1.30it/s]
3|4764|Loss: 0.7402034997940063:  74%|███████▎  | 4764/6470 [54:47<21:11,  1.34it/s]
3|4765|Loss: 0.972267210483551:  74%|███████▎  | 4764/6470 [54:47<21:11,  1.34it/s] 
3|4765|Loss: 0.972267210483551:  74%|███████▎  | 4765/6470 [54:48<20:44,  1.37it/s]
3|4766|Loss: 0.7535048723220825:  74%|███████▎  | 4765/6470 [54:48<20:44,  1.37it/s]
3|4766|Loss: 0.7535048723220825:  74%|███████▎  | 4766/6470 [54:48<20:05,  1.41it/s]
3|4767|Loss: 0.8867493271827698:  74%|███████▎  | 4766/6470 [54:49<20:05,  1.41it/s]
3|4767|Loss: 0.8867493271827698:  74%|███████▎  | 4767/6470 [54:49<20:09,  1.41it/s]
3|4768|Loss: 0.6709046959877014:  74%|███████▎  | 4767/6470 [54:49<20:09,  1.41it/s]
3|4768|Loss: 0.6709046959877014:  74%|███████▎  | 4768/6470 [54:50<19:37,  1.45it/s]
3|4769|Loss: 0.855500340461731:  74%|███████▎  | 4768/6470 [54:50<19:37,  1.45it/s] 
3|4769|Loss: 0.855500340461731:  74%|███████▎  | 4769/6470 [54:50<18:44,  1.51it/s]
3|4770|Loss: 0.8438999652862549:  74%|███████▎  | 4769/6470 [54:51<18:44,  1.51it/s]
3|4770|Loss: 0.8438999652862549:  74%|███████▎  | 4770/6470 [54:51<19:47,  1.43it/s]
3|4771|Loss: 0.8776663541793823:  74%|███████▎  | 4770/6470 [54:51<19:47,  1.43it/s]
3|4771|Loss: 0.8776663541793823:  74%|███████▎  | 4771/6470 [54:52<19:26,  1.46it/s]
3|4772|Loss: 1.0407158136367798:  74%|███████▎  | 4771/6470 [54:52<19:26,  1.46it/s]
3|4772|Loss: 1.0407158136367798:  74%|███████▍  | 4772/6470 [54:52<18:22,  1.54it/s]
3|4773|Loss: 0.7394986748695374:  74%|███████▍  | 4772/6470 [54:53<18:22,  1.54it/s]
3|4773|Loss: 0.7394986748695374:  74%|███████▍  | 4773/6470 [54:53<18:44,  1.51it/s]
3|4774|Loss: 0.9203397035598755:  74%|███████▍  | 4773/6470 [54:53<18:44,  1.51it/s]
3|4774|Loss: 0.9203397035598755:  74%|███████▍  | 4774/6470 [54:53<18:18,  1.54it/s]
3|4775|Loss: 1.038243293762207:  74%|███████▍  | 4774/6470 [54:54<18:18,  1.54it/s] 
3|4775|Loss: 1.038243293762207:  74%|███████▍  | 4775/6470 [54:54<18:05,  1.56it/s]
3|4776|Loss: 0.9193218946456909:  74%|███████▍  | 4775/6470 [54:54<18:05,  1.56it/s]
3|4776|Loss: 0.9193218946456909:  74%|███████▍  | 4776/6470 [54:55<18:00,  1.57it/s]
3|4777|Loss: 0.832348644733429:  74%|███████▍  | 4776/6470 [54:55<18:00,  1.57it/s] 
3|4777|Loss: 0.832348644733429:  74%|███████▍  | 4777/6470 [54:55<17:51,  1.58it/s]
3|4778|Loss: 0.7987624406814575:  74%|███████▍  | 4777/6470 [54:56<17:51,  1.58it/s]
3|4778|Loss: 0.7987624406814575:  74%|███████▍  | 4778/6470 [54:56<17:43,  1.59it/s]
3|4779|Loss: 0.7346562147140503:  74%|███████▍  | 4778/6470 [54:56<17:43,  1.59it/s]
3|4779|Loss: 0.7346562147140503:  74%|███████▍  | 4779/6470 [54:57<19:58,  1.41it/s]
3|4780|Loss: 0.9129914045333862:  74%|███████▍  | 4779/6470 [54:57<19:58,  1.41it/s]
3|4780|Loss: 0.9129914045333862:  74%|███████▍  | 4780/6470 [54:58<20:24,  1.38it/s]
3|4781|Loss: 0.7868322134017944:  74%|███████▍  | 4780/6470 [54:58<20:24,  1.38it/s]
3|4781|Loss: 0.7868322134017944:  74%|███████▍  | 4781/6470 [54:59<21:59,  1.28it/s]
3|4782|Loss: 0.9019266963005066:  74%|███████▍  | 4781/6470 [54:59<21:59,  1.28it/s]
3|4782|Loss: 0.9019266963005066:  74%|███████▍  | 4782/6470 [54:59<22:00,  1.28it/s]
3|4783|Loss: 0.7476363182067871:  74%|███████▍  | 4782/6470 [55:00<22:00,  1.28it/s]
3|4783|Loss: 0.7476363182067871:  74%|███████▍  | 4783/6470 [55:00<22:36,  1.24it/s]
3|4784|Loss: 0.7851139307022095:  74%|███████▍  | 4783/6470 [55:01<22:36,  1.24it/s]
3|4784|Loss: 0.7851139307022095:  74%|███████▍  | 4784/6470 [55:01<22:17,  1.26it/s]
3|4785|Loss: 0.8722739815711975:  74%|███████▍  | 4784/6470 [55:01<22:17,  1.26it/s]
3|4785|Loss: 0.8722739815711975:  74%|███████▍  | 4785/6470 [55:02<21:02,  1.33it/s]
3|4786|Loss: 1.015275239944458:  74%|███████▍  | 4785/6470 [55:02<21:02,  1.33it/s] 
3|4786|Loss: 1.015275239944458:  74%|███████▍  | 4786/6470 [55:02<19:12,  1.46it/s]
3|4787|Loss: 0.8427304029464722:  74%|███████▍  | 4786/6470 [55:02<19:12,  1.46it/s]
3|4787|Loss: 0.8427304029464722:  74%|███████▍  | 4787/6470 [55:03<19:14,  1.46it/s]
3|4788|Loss: 0.8018295764923096:  74%|███████▍  | 4787/6470 [55:03<19:14,  1.46it/s]
3|4788|Loss: 0.8018295764923096:  74%|███████▍  | 4788/6470 [55:03<19:27,  1.44it/s]
3|4789|Loss: 0.8246556520462036:  74%|███████▍  | 4788/6470 [55:04<19:27,  1.44it/s]
3|4789|Loss: 0.8246556520462036:  74%|███████▍  | 4789/6470 [55:04<19:33,  1.43it/s]
3|4790|Loss: 0.9689425826072693:  74%|███████▍  | 4789/6470 [55:05<19:33,  1.43it/s]
3|4790|Loss: 0.9689425826072693:  74%|███████▍  | 4790/6470 [55:05<19:53,  1.41it/s]
3|4791|Loss: 0.8266675472259521:  74%|███████▍  | 4790/6470 [55:05<19:53,  1.41it/s]
3|4791|Loss: 0.8266675472259521:  74%|███████▍  | 4791/6470 [55:06<18:54,  1.48it/s]
3|4792|Loss: 1.1566590070724487:  74%|███████▍  | 4791/6470 [55:06<18:54,  1.48it/s]
3|4792|Loss: 1.1566590070724487:  74%|███████▍  | 4792/6470 [55:06<19:02,  1.47it/s]
3|4793|Loss: 0.8842761516571045:  74%|███████▍  | 4792/6470 [55:07<19:02,  1.47it/s]
3|4793|Loss: 0.8842761516571045:  74%|███████▍  | 4793/6470 [55:07<18:25,  1.52it/s]
3|4794|Loss: 0.9218875169754028:  74%|███████▍  | 4793/6470 [55:07<18:25,  1.52it/s]
3|4794|Loss: 0.9218875169754028:  74%|███████▍  | 4794/6470 [55:08<18:53,  1.48it/s]
3|4795|Loss: 0.8529819250106812:  74%|███████▍  | 4794/6470 [55:08<18:53,  1.48it/s]
3|4795|Loss: 0.8529819250106812:  74%|███████▍  | 4795/6470 [55:08<19:26,  1.44it/s]
3|4796|Loss: 0.8770071268081665:  74%|███████▍  | 4795/6470 [55:09<19:26,  1.44it/s]
3|4796|Loss: 0.8770071268081665:  74%|███████▍  | 4796/6470 [55:09<20:12,  1.38it/s]
3|4797|Loss: 0.8864762783050537:  74%|███████▍  | 4796/6470 [55:10<20:12,  1.38it/s]
3|4797|Loss: 0.8864762783050537:  74%|███████▍  | 4797/6470 [55:10<21:02,  1.32it/s]
3|4798|Loss: 0.8887279033660889:  74%|███████▍  | 4797/6470 [55:10<21:02,  1.32it/s]
3|4798|Loss: 0.8887279033660889:  74%|███████▍  | 4798/6470 [55:10<19:33,  1.42it/s]
3|4799|Loss: 1.0302772521972656:  74%|███████▍  | 4798/6470 [55:11<19:33,  1.42it/s]
3|4799|Loss: 1.0302772521972656:  74%|███████▍  | 4799/6470 [55:11<19:23,  1.44it/s]
3|4800|Loss: 0.8628571629524231:  74%|███████▍  | 4799/6470 [55:12<19:23,  1.44it/s]
3|4800|Loss: 0.8628571629524231:  74%|███████▍  | 4800/6470 [55:12<18:44,  1.49it/s]
3|4801|Loss: 0.7537282705307007:  74%|███████▍  | 4800/6470 [55:12<18:44,  1.49it/s]
3|4801|Loss: 0.7537282705307007:  74%|███████▍  | 4801/6470 [55:13<19:20,  1.44it/s]
3|4802|Loss: 0.8163157105445862:  74%|███████▍  | 4801/6470 [55:13<19:20,  1.44it/s]
3|4802|Loss: 0.8163157105445862:  74%|███████▍  | 4802/6470 [55:13<17:28,  1.59it/s]
3|4803|Loss: 0.8269454836845398:  74%|███████▍  | 4802/6470 [55:13<17:28,  1.59it/s]
3|4803|Loss: 0.8269454836845398:  74%|███████▍  | 4803/6470 [55:14<18:58,  1.46it/s]
3|4804|Loss: 0.9291341304779053:  74%|███████▍  | 4803/6470 [55:14<18:58,  1.46it/s]
3|4804|Loss: 0.9291341304779053:  74%|███████▍  | 4804/6470 [55:15<19:03,  1.46it/s]
3|4805|Loss: 0.8898918628692627:  74%|███████▍  | 4804/6470 [55:15<19:03,  1.46it/s]
3|4805|Loss: 0.8898918628692627:  74%|███████▍  | 4805/6470 [55:15<18:41,  1.48it/s]
3|4806|Loss: 0.9314770102500916:  74%|███████▍  | 4805/6470 [55:16<18:41,  1.48it/s]
3|4806|Loss: 0.9314770102500916:  74%|███████▍  | 4806/6470 [55:16<18:54,  1.47it/s]
3|4807|Loss: 0.8937637805938721:  74%|███████▍  | 4806/6470 [55:16<18:54,  1.47it/s]
3|4807|Loss: 0.8937637805938721:  74%|███████▍  | 4807/6470 [55:17<18:34,  1.49it/s]
3|4808|Loss: 0.8644855618476868:  74%|███████▍  | 4807/6470 [55:17<18:34,  1.49it/s]
3|4808|Loss: 0.8644855618476868:  74%|███████▍  | 4808/6470 [55:17<17:24,  1.59it/s]
3|4809|Loss: 0.7741149663925171:  74%|███████▍  | 4808/6470 [55:17<17:24,  1.59it/s]
3|4809|Loss: 0.7741149663925171:  74%|███████▍  | 4809/6470 [55:18<17:39,  1.57it/s]
3|4810|Loss: 0.699845552444458:  74%|███████▍  | 4809/6470 [55:18<17:39,  1.57it/s] 
3|4810|Loss: 0.699845552444458:  74%|███████▍  | 4810/6470 [55:18<17:01,  1.62it/s]
3|4811|Loss: 0.9242115616798401:  74%|███████▍  | 4810/6470 [55:19<17:01,  1.62it/s]
3|4811|Loss: 0.9242115616798401:  74%|███████▍  | 4811/6470 [55:19<16:28,  1.68it/s]
3|4812|Loss: 0.7541226148605347:  74%|███████▍  | 4811/6470 [55:19<16:28,  1.68it/s]
3|4812|Loss: 0.7541226148605347:  74%|███████▍  | 4812/6470 [55:19<16:45,  1.65it/s]
3|4813|Loss: 0.8198995590209961:  74%|███████▍  | 4812/6470 [55:20<16:45,  1.65it/s]
3|4813|Loss: 0.8198995590209961:  74%|███████▍  | 4813/6470 [55:20<18:21,  1.50it/s]
3|4814|Loss: 0.7634518146514893:  74%|███████▍  | 4813/6470 [55:21<18:21,  1.50it/s]
3|4814|Loss: 0.7634518146514893:  74%|███████▍  | 4814/6470 [55:21<18:29,  1.49it/s]
3|4815|Loss: 0.7414312958717346:  74%|███████▍  | 4814/6470 [55:21<18:29,  1.49it/s]
3|4815|Loss: 0.7414312958717346:  74%|███████▍  | 4815/6470 [55:22<18:15,  1.51it/s]
3|4816|Loss: 0.7252300977706909:  74%|███████▍  | 4815/6470 [55:22<18:15,  1.51it/s]
3|4816|Loss: 0.7252300977706909:  74%|███████▍  | 4816/6470 [55:22<18:51,  1.46it/s]
3|4817|Loss: 0.6626336574554443:  74%|███████▍  | 4816/6470 [55:23<18:51,  1.46it/s]
3|4817|Loss: 0.6626336574554443:  74%|███████▍  | 4817/6470 [55:23<17:56,  1.54it/s]
3|4818|Loss: 0.8264035582542419:  74%|███████▍  | 4817/6470 [55:23<17:56,  1.54it/s]
3|4818|Loss: 0.8264035582542419:  74%|███████▍  | 4818/6470 [55:24<18:04,  1.52it/s]
3|4819|Loss: 0.9277705550193787:  74%|███████▍  | 4818/6470 [55:24<18:04,  1.52it/s]
3|4819|Loss: 0.9277705550193787:  74%|███████▍  | 4819/6470 [55:24<17:54,  1.54it/s]
3|4820|Loss: 0.9036674499511719:  74%|███████▍  | 4819/6470 [55:25<17:54,  1.54it/s]
3|4820|Loss: 0.9036674499511719:  74%|███████▍  | 4820/6470 [55:25<18:22,  1.50it/s]
3|4821|Loss: 0.8586832284927368:  74%|███████▍  | 4820/6470 [55:25<18:22,  1.50it/s]
3|4821|Loss: 0.8586832284927368:  75%|███████▍  | 4821/6470 [55:26<19:22,  1.42it/s]
3|4822|Loss: 0.9181504845619202:  75%|███████▍  | 4821/6470 [55:26<19:22,  1.42it/s]
3|4822|Loss: 0.9181504845619202:  75%|███████▍  | 4822/6470 [55:26<19:10,  1.43it/s]
3|4823|Loss: 0.8075820207595825:  75%|███████▍  | 4822/6470 [55:27<19:10,  1.43it/s]
3|4823|Loss: 0.8075820207595825:  75%|███████▍  | 4823/6470 [55:27<19:47,  1.39it/s]
3|4824|Loss: 0.8336752653121948:  75%|███████▍  | 4823/6470 [55:28<19:47,  1.39it/s]
3|4824|Loss: 0.8336752653121948:  75%|███████▍  | 4824/6470 [55:28<18:47,  1.46it/s]
3|4825|Loss: 0.7761508822441101:  75%|███████▍  | 4824/6470 [55:28<18:47,  1.46it/s]
3|4825|Loss: 0.7761508822441101:  75%|███████▍  | 4825/6470 [55:28<18:41,  1.47it/s]
3|4826|Loss: 0.9606322646141052:  75%|███████▍  | 4825/6470 [55:29<18:41,  1.47it/s]
3|4826|Loss: 0.9606322646141052:  75%|███████▍  | 4826/6470 [55:29<17:37,  1.55it/s]
3|4827|Loss: 0.9140071272850037:  75%|███████▍  | 4826/6470 [55:29<17:37,  1.55it/s]
3|4827|Loss: 0.9140071272850037:  75%|███████▍  | 4827/6470 [55:30<17:21,  1.58it/s]
3|4828|Loss: 0.7884095907211304:  75%|███████▍  | 4827/6470 [55:30<17:21,  1.58it/s]
3|4828|Loss: 0.7884095907211304:  75%|███████▍  | 4828/6470 [55:30<18:02,  1.52it/s]
3|4829|Loss: 0.7512489557266235:  75%|███████▍  | 4828/6470 [55:31<18:02,  1.52it/s]
3|4829|Loss: 0.7512489557266235:  75%|███████▍  | 4829/6470 [55:31<17:18,  1.58it/s]
3|4830|Loss: 0.9031994938850403:  75%|███████▍  | 4829/6470 [55:31<17:18,  1.58it/s]
3|4830|Loss: 0.9031994938850403:  75%|███████▍  | 4830/6470 [55:31<16:43,  1.63it/s]
3|4831|Loss: 0.8851579427719116:  75%|███████▍  | 4830/6470 [55:32<16:43,  1.63it/s]
3|4831|Loss: 0.8851579427719116:  75%|███████▍  | 4831/6470 [55:32<17:28,  1.56it/s]
3|4832|Loss: 0.8790149092674255:  75%|███████▍  | 4831/6470 [55:33<17:28,  1.56it/s]
3|4832|Loss: 0.8790149092674255:  75%|███████▍  | 4832/6470 [55:33<17:36,  1.55it/s]
3|4833|Loss: 0.8802184462547302:  75%|███████▍  | 4832/6470 [55:33<17:36,  1.55it/s]
3|4833|Loss: 0.8802184462547302:  75%|███████▍  | 4833/6470 [55:33<17:55,  1.52it/s]
3|4834|Loss: 0.8796426057815552:  75%|███████▍  | 4833/6470 [55:34<17:55,  1.52it/s]
3|4834|Loss: 0.8796426057815552:  75%|███████▍  | 4834/6470 [55:34<17:25,  1.56it/s]
3|4835|Loss: 1.015432596206665:  75%|███████▍  | 4834/6470 [55:34<17:25,  1.56it/s] 
3|4835|Loss: 1.015432596206665:  75%|███████▍  | 4835/6470 [55:35<17:10,  1.59it/s]
3|4836|Loss: 0.6504260301589966:  75%|███████▍  | 4835/6470 [55:35<17:10,  1.59it/s]
3|4836|Loss: 0.6504260301589966:  75%|███████▍  | 4836/6470 [55:35<18:14,  1.49it/s]
3|4837|Loss: 0.8426614999771118:  75%|███████▍  | 4836/6470 [55:36<18:14,  1.49it/s]
3|4837|Loss: 0.8426614999771118:  75%|███████▍  | 4837/6470 [55:36<18:52,  1.44it/s]
3|4838|Loss: 0.8597903847694397:  75%|███████▍  | 4837/6470 [55:37<18:52,  1.44it/s]
3|4838|Loss: 0.8597903847694397:  75%|███████▍  | 4838/6470 [55:37<19:11,  1.42it/s]
3|4839|Loss: 0.7826804518699646:  75%|███████▍  | 4838/6470 [55:37<19:11,  1.42it/s]
3|4839|Loss: 0.7826804518699646:  75%|███████▍  | 4839/6470 [55:38<19:23,  1.40it/s]
3|4840|Loss: 0.9280864000320435:  75%|███████▍  | 4839/6470 [55:38<19:23,  1.40it/s]
3|4840|Loss: 0.9280864000320435:  75%|███████▍  | 4840/6470 [55:38<18:57,  1.43it/s]
3|4841|Loss: 0.7830296158790588:  75%|███████▍  | 4840/6470 [55:39<18:57,  1.43it/s]
3|4841|Loss: 0.7830296158790588:  75%|███████▍  | 4841/6470 [55:39<18:08,  1.50it/s]
3|4842|Loss: 0.7121387124061584:  75%|███████▍  | 4841/6470 [55:39<18:08,  1.50it/s]
3|4842|Loss: 0.7121387124061584:  75%|███████▍  | 4842/6470 [55:40<19:00,  1.43it/s]
3|4843|Loss: 0.8714977502822876:  75%|███████▍  | 4842/6470 [55:40<19:00,  1.43it/s]
3|4843|Loss: 0.8714977502822876:  75%|███████▍  | 4843/6470 [55:40<19:16,  1.41it/s]
3|4844|Loss: 0.922014057636261:  75%|███████▍  | 4843/6470 [55:41<19:16,  1.41it/s] 
3|4844|Loss: 0.922014057636261:  75%|███████▍  | 4844/6470 [55:41<18:40,  1.45it/s]
3|4845|Loss: 0.9083646535873413:  75%|███████▍  | 4844/6470 [55:41<18:40,  1.45it/s]
3|4845|Loss: 0.9083646535873413:  75%|███████▍  | 4845/6470 [55:42<18:12,  1.49it/s]
3|4846|Loss: 0.8632359504699707:  75%|███████▍  | 4845/6470 [55:42<18:12,  1.49it/s]
3|4846|Loss: 0.8632359504699707:  75%|███████▍  | 4846/6470 [55:42<17:31,  1.54it/s]
3|4847|Loss: 0.8191646933555603:  75%|███████▍  | 4846/6470 [55:43<17:31,  1.54it/s]
3|4847|Loss: 0.8191646933555603:  75%|███████▍  | 4847/6470 [55:43<17:59,  1.50it/s]
3|4848|Loss: 0.9814507961273193:  75%|███████▍  | 4847/6470 [55:43<17:59,  1.50it/s]
3|4848|Loss: 0.9814507961273193:  75%|███████▍  | 4848/6470 [55:44<18:32,  1.46it/s]
3|4849|Loss: 0.9741269946098328:  75%|███████▍  | 4848/6470 [55:44<18:32,  1.46it/s]
3|4849|Loss: 0.9741269946098328:  75%|███████▍  | 4849/6470 [55:45<19:28,  1.39it/s]
3|4850|Loss: 0.8814737200737:  75%|███████▍  | 4849/6470 [55:45<19:28,  1.39it/s]   
3|4850|Loss: 0.8814737200737:  75%|███████▍  | 4850/6470 [55:45<20:05,  1.34it/s]
3|4851|Loss: 0.9598393440246582:  75%|███████▍  | 4850/6470 [55:46<20:05,  1.34it/s]
3|4851|Loss: 0.9598393440246582:  75%|███████▍  | 4851/6470 [55:46<19:33,  1.38it/s]
3|4852|Loss: 0.8577184677124023:  75%|███████▍  | 4851/6470 [55:46<19:33,  1.38it/s]
3|4852|Loss: 0.8577184677124023:  75%|███████▍  | 4852/6470 [55:47<18:00,  1.50it/s]
3|4853|Loss: 0.7513085603713989:  75%|███████▍  | 4852/6470 [55:47<18:00,  1.50it/s]
3|4853|Loss: 0.7513085603713989:  75%|███████▌  | 4853/6470 [55:47<16:42,  1.61it/s]
3|4854|Loss: 0.8185508847236633:  75%|███████▌  | 4853/6470 [55:47<16:42,  1.61it/s]
3|4854|Loss: 0.8185508847236633:  75%|███████▌  | 4854/6470 [55:48<16:38,  1.62it/s]
3|4855|Loss: 0.7861766219139099:  75%|███████▌  | 4854/6470 [55:48<16:38,  1.62it/s]
3|4855|Loss: 0.7861766219139099:  75%|███████▌  | 4855/6470 [55:48<18:00,  1.49it/s]
3|4856|Loss: 0.7834137082099915:  75%|███████▌  | 4855/6470 [55:49<18:00,  1.49it/s]
3|4856|Loss: 0.7834137082099915:  75%|███████▌  | 4856/6470 [55:49<16:17,  1.65it/s]
3|4857|Loss: 0.7616866827011108:  75%|███████▌  | 4856/6470 [55:49<16:17,  1.65it/s]
3|4857|Loss: 0.7616866827011108:  75%|███████▌  | 4857/6470 [55:49<16:01,  1.68it/s]
3|4858|Loss: 0.9100281000137329:  75%|███████▌  | 4857/6470 [55:50<16:01,  1.68it/s]
3|4858|Loss: 0.9100281000137329:  75%|███████▌  | 4858/6470 [55:50<16:55,  1.59it/s]
3|4859|Loss: 0.8391183018684387:  75%|███████▌  | 4858/6470 [55:51<16:55,  1.59it/s]
3|4859|Loss: 0.8391183018684387:  75%|███████▌  | 4859/6470 [55:51<18:37,  1.44it/s]
3|4860|Loss: 0.9888972043991089:  75%|███████▌  | 4859/6470 [55:51<18:37,  1.44it/s]
3|4860|Loss: 0.9888972043991089:  75%|███████▌  | 4860/6470 [55:52<18:25,  1.46it/s]
3|4861|Loss: 0.9015138149261475:  75%|███████▌  | 4860/6470 [55:52<18:25,  1.46it/s]
3|4861|Loss: 0.9015138149261475:  75%|███████▌  | 4861/6470 [55:52<18:22,  1.46it/s]
3|4862|Loss: 0.8372375965118408:  75%|███████▌  | 4861/6470 [55:53<18:22,  1.46it/s]
3|4862|Loss: 0.8372375965118408:  75%|███████▌  | 4862/6470 [55:53<19:41,  1.36it/s]
3|4863|Loss: 0.9923489689826965:  75%|███████▌  | 4862/6470 [55:54<19:41,  1.36it/s]
3|4863|Loss: 0.9923489689826965:  75%|███████▌  | 4863/6470 [55:54<20:07,  1.33it/s]
3|4864|Loss: 0.8754547834396362:  75%|███████▌  | 4863/6470 [55:54<20:07,  1.33it/s]
3|4864|Loss: 0.8754547834396362:  75%|███████▌  | 4864/6470 [55:55<18:34,  1.44it/s]
3|4865|Loss: 0.9351609349250793:  75%|███████▌  | 4864/6470 [55:55<18:34,  1.44it/s]
3|4865|Loss: 0.9351609349250793:  75%|███████▌  | 4865/6470 [55:55<17:33,  1.52it/s]
3|4866|Loss: 0.8740114569664001:  75%|███████▌  | 4865/6470 [55:56<17:33,  1.52it/s]
3|4866|Loss: 0.8740114569664001:  75%|███████▌  | 4866/6470 [55:56<17:43,  1.51it/s]
3|4867|Loss: 0.8770907521247864:  75%|███████▌  | 4866/6470 [55:56<17:43,  1.51it/s]
3|4867|Loss: 0.8770907521247864:  75%|███████▌  | 4867/6470 [55:57<18:00,  1.48it/s]
3|4868|Loss: 0.730866551399231:  75%|███████▌  | 4867/6470 [55:57<18:00,  1.48it/s] 
3|4868|Loss: 0.730866551399231:  75%|███████▌  | 4868/6470 [55:57<19:00,  1.40it/s]
3|4869|Loss: 0.7856500744819641:  75%|███████▌  | 4868/6470 [55:58<19:00,  1.40it/s]
3|4869|Loss: 0.7856500744819641:  75%|███████▌  | 4869/6470 [55:58<19:09,  1.39it/s]
3|4870|Loss: 0.8327004313468933:  75%|███████▌  | 4869/6470 [55:58<19:09,  1.39it/s]
3|4870|Loss: 0.8327004313468933:  75%|███████▌  | 4870/6470 [55:59<18:16,  1.46it/s]
3|4871|Loss: 0.9055837392807007:  75%|███████▌  | 4870/6470 [55:59<18:16,  1.46it/s]
3|4871|Loss: 0.9055837392807007:  75%|███████▌  | 4871/6470 [55:59<16:57,  1.57it/s]
3|4872|Loss: 0.7901224493980408:  75%|███████▌  | 4871/6470 [56:00<16:57,  1.57it/s]
3|4872|Loss: 0.7901224493980408:  75%|███████▌  | 4872/6470 [56:00<17:07,  1.56it/s]
3|4873|Loss: 1.0111050605773926:  75%|███████▌  | 4872/6470 [56:00<17:07,  1.56it/s]
3|4873|Loss: 1.0111050605773926:  75%|███████▌  | 4873/6470 [56:01<18:14,  1.46it/s]
3|4874|Loss: 0.7339804172515869:  75%|███████▌  | 4873/6470 [56:01<18:14,  1.46it/s]
3|4874|Loss: 0.7339804172515869:  75%|███████▌  | 4874/6470 [56:01<18:00,  1.48it/s]
3|4875|Loss: 0.8597643971443176:  75%|███████▌  | 4874/6470 [56:02<18:00,  1.48it/s]
3|4875|Loss: 0.8597643971443176:  75%|███████▌  | 4875/6470 [56:02<17:40,  1.50it/s]
3|4876|Loss: 0.8145354986190796:  75%|███████▌  | 4875/6470 [56:02<17:40,  1.50it/s]
3|4876|Loss: 0.8145354986190796:  75%|███████▌  | 4876/6470 [56:03<18:01,  1.47it/s]
3|4877|Loss: 0.8217713832855225:  75%|███████▌  | 4876/6470 [56:03<18:01,  1.47it/s]
3|4877|Loss: 0.8217713832855225:  75%|███████▌  | 4877/6470 [56:03<18:47,  1.41it/s]
3|4878|Loss: 0.8851301074028015:  75%|███████▌  | 4877/6470 [56:04<18:47,  1.41it/s]
3|4878|Loss: 0.8851301074028015:  75%|███████▌  | 4878/6470 [56:04<19:10,  1.38it/s]
3|4879|Loss: 0.8016879558563232:  75%|███████▌  | 4878/6470 [56:05<19:10,  1.38it/s]
3|4879|Loss: 0.8016879558563232:  75%|███████▌  | 4879/6470 [56:05<17:27,  1.52it/s]
3|4880|Loss: 0.8804425597190857:  75%|███████▌  | 4879/6470 [56:05<17:27,  1.52it/s]
3|4880|Loss: 0.8804425597190857:  75%|███████▌  | 4880/6470 [56:05<16:13,  1.63it/s]
3|4881|Loss: 0.8561388850212097:  75%|███████▌  | 4880/6470 [56:06<16:13,  1.63it/s]
3|4881|Loss: 0.8561388850212097:  75%|███████▌  | 4881/6470 [56:06<16:38,  1.59it/s]
3|4882|Loss: 0.7960371375083923:  75%|███████▌  | 4881/6470 [56:06<16:38,  1.59it/s]
3|4882|Loss: 0.7960371375083923:  75%|███████▌  | 4882/6470 [56:07<16:41,  1.59it/s]
3|4883|Loss: 0.7209928631782532:  75%|███████▌  | 4882/6470 [56:07<16:41,  1.59it/s]
3|4883|Loss: 0.7209928631782532:  75%|███████▌  | 4883/6470 [56:07<15:54,  1.66it/s]
3|4884|Loss: 0.816010594367981:  75%|███████▌  | 4883/6470 [56:07<15:54,  1.66it/s] 
3|4884|Loss: 0.816010594367981:  75%|███████▌  | 4884/6470 [56:08<17:12,  1.54it/s]
3|4885|Loss: 0.9710327386856079:  75%|███████▌  | 4884/6470 [56:08<17:12,  1.54it/s]
3|4885|Loss: 0.9710327386856079:  76%|███████▌  | 4885/6470 [56:09<17:50,  1.48it/s]
3|4886|Loss: 0.8822246789932251:  76%|███████▌  | 4885/6470 [56:09<17:50,  1.48it/s]
3|4886|Loss: 0.8822246789932251:  76%|███████▌  | 4886/6470 [56:09<17:39,  1.49it/s]
3|4887|Loss: 1.005514144897461:  76%|███████▌  | 4886/6470 [56:10<17:39,  1.49it/s] 
3|4887|Loss: 1.005514144897461:  76%|███████▌  | 4887/6470 [56:10<17:16,  1.53it/s]
3|4888|Loss: 0.7467966675758362:  76%|███████▌  | 4887/6470 [56:10<17:16,  1.53it/s]
3|4888|Loss: 0.7467966675758362:  76%|███████▌  | 4888/6470 [56:10<17:08,  1.54it/s]
3|4889|Loss: 0.6871287822723389:  76%|███████▌  | 4888/6470 [56:11<17:08,  1.54it/s]
3|4889|Loss: 0.6871287822723389:  76%|███████▌  | 4889/6470 [56:11<18:02,  1.46it/s]
3|4890|Loss: 0.8550317883491516:  76%|███████▌  | 4889/6470 [56:12<18:02,  1.46it/s]
3|4890|Loss: 0.8550317883491516:  76%|███████▌  | 4890/6470 [56:12<17:14,  1.53it/s]
3|4891|Loss: 0.9171727299690247:  76%|███████▌  | 4890/6470 [56:12<17:14,  1.53it/s]
3|4891|Loss: 0.9171727299690247:  76%|███████▌  | 4891/6470 [56:12<16:57,  1.55it/s]
3|4892|Loss: 0.8291922807693481:  76%|███████▌  | 4891/6470 [56:13<16:57,  1.55it/s]
3|4892|Loss: 0.8291922807693481:  76%|███████▌  | 4892/6470 [56:13<16:56,  1.55it/s]
3|4893|Loss: 0.8770602941513062:  76%|███████▌  | 4892/6470 [56:13<16:56,  1.55it/s]
3|4893|Loss: 0.8770602941513062:  76%|███████▌  | 4893/6470 [56:14<16:45,  1.57it/s]
3|4894|Loss: 0.9052172899246216:  76%|███████▌  | 4893/6470 [56:14<16:45,  1.57it/s]
3|4894|Loss: 0.9052172899246216:  76%|███████▌  | 4894/6470 [56:14<16:42,  1.57it/s]
3|4895|Loss: 0.9131078720092773:  76%|███████▌  | 4894/6470 [56:15<16:42,  1.57it/s]
3|4895|Loss: 0.9131078720092773:  76%|███████▌  | 4895/6470 [56:15<18:44,  1.40it/s]
3|4896|Loss: 0.8720307946205139:  76%|███████▌  | 4895/6470 [56:16<18:44,  1.40it/s]
3|4896|Loss: 0.8720307946205139:  76%|███████▌  | 4896/6470 [56:16<18:43,  1.40it/s]
3|4897|Loss: 0.9802442789077759:  76%|███████▌  | 4896/6470 [56:16<18:43,  1.40it/s]
3|4897|Loss: 0.9802442789077759:  76%|███████▌  | 4897/6470 [56:17<19:05,  1.37it/s]
3|4898|Loss: 0.8573774099349976:  76%|███████▌  | 4897/6470 [56:17<19:05,  1.37it/s]
3|4898|Loss: 0.8573774099349976:  76%|███████▌  | 4898/6470 [56:17<16:53,  1.55it/s]
3|4899|Loss: 0.8223207592964172:  76%|███████▌  | 4898/6470 [56:17<16:53,  1.55it/s]
3|4899|Loss: 0.8223207592964172:  76%|███████▌  | 4899/6470 [56:18<16:28,  1.59it/s]
3|4900|Loss: 0.9052829146385193:  76%|███████▌  | 4899/6470 [56:18<16:28,  1.59it/s]
3|4900|Loss: 0.9052829146385193:  76%|███████▌  | 4900/6470 [56:19<18:02,  1.45it/s]
3|4901|Loss: 0.9327067732810974:  76%|███████▌  | 4900/6470 [56:19<18:02,  1.45it/s]
3|4901|Loss: 0.9327067732810974:  76%|███████▌  | 4901/6470 [56:19<19:12,  1.36it/s]
3|4902|Loss: 1.0744212865829468:  76%|███████▌  | 4901/6470 [56:20<19:12,  1.36it/s]
3|4902|Loss: 1.0744212865829468:  76%|███████▌  | 4902/6470 [56:20<17:35,  1.49it/s]
3|4903|Loss: 0.9054214954376221:  76%|███████▌  | 4902/6470 [56:20<17:35,  1.49it/s]
3|4903|Loss: 0.9054214954376221:  76%|███████▌  | 4903/6470 [56:20<16:33,  1.58it/s]
3|4904|Loss: 0.8318618535995483:  76%|███████▌  | 4903/6470 [56:21<16:33,  1.58it/s]
3|4904|Loss: 0.8318618535995483:  76%|███████▌  | 4904/6470 [56:21<17:11,  1.52it/s]
3|4905|Loss: 0.9078357815742493:  76%|███████▌  | 4904/6470 [56:22<17:11,  1.52it/s]
3|4905|Loss: 0.9078357815742493:  76%|███████▌  | 4905/6470 [56:22<17:03,  1.53it/s]
3|4906|Loss: 0.81317138671875:  76%|███████▌  | 4905/6470 [56:22<17:03,  1.53it/s]  
3|4906|Loss: 0.81317138671875:  76%|███████▌  | 4906/6470 [56:22<16:43,  1.56it/s]
3|4907|Loss: 0.7563077807426453:  76%|███████▌  | 4906/6470 [56:23<16:43,  1.56it/s]
3|4907|Loss: 0.7563077807426453:  76%|███████▌  | 4907/6470 [56:23<16:16,  1.60it/s]
3|4908|Loss: 0.8378695249557495:  76%|███████▌  | 4907/6470 [56:23<16:16,  1.60it/s]
3|4908|Loss: 0.8378695249557495:  76%|███████▌  | 4908/6470 [56:24<16:06,  1.62it/s]
3|4909|Loss: 0.8009276986122131:  76%|███████▌  | 4908/6470 [56:24<16:06,  1.62it/s]
3|4909|Loss: 0.8009276986122131:  76%|███████▌  | 4909/6470 [56:24<16:50,  1.54it/s]
3|4910|Loss: 0.9789798259735107:  76%|███████▌  | 4909/6470 [56:25<16:50,  1.54it/s]
3|4910|Loss: 0.9789798259735107:  76%|███████▌  | 4910/6470 [56:25<16:13,  1.60it/s]
3|4911|Loss: 0.8159584403038025:  76%|███████▌  | 4910/6470 [56:25<16:13,  1.60it/s]
3|4911|Loss: 0.8159584403038025:  76%|███████▌  | 4911/6470 [56:26<16:52,  1.54it/s]
3|4912|Loss: 0.7185935974121094:  76%|███████▌  | 4911/6470 [56:26<16:52,  1.54it/s]
3|4912|Loss: 0.7185935974121094:  76%|███████▌  | 4912/6470 [56:26<16:42,  1.55it/s]
3|4913|Loss: 0.6993145942687988:  76%|███████▌  | 4912/6470 [56:27<16:42,  1.55it/s]
3|4913|Loss: 0.6993145942687988:  76%|███████▌  | 4913/6470 [56:27<17:19,  1.50it/s]
3|4914|Loss: 0.8242020010948181:  76%|███████▌  | 4913/6470 [56:27<17:19,  1.50it/s]
3|4914|Loss: 0.8242020010948181:  76%|███████▌  | 4914/6470 [56:28<18:37,  1.39it/s]
3|4915|Loss: 0.6924572587013245:  76%|███████▌  | 4914/6470 [56:28<18:37,  1.39it/s]
3|4915|Loss: 0.6924572587013245:  76%|███████▌  | 4915/6470 [56:28<18:14,  1.42it/s]
3|4916|Loss: 0.814389705657959:  76%|███████▌  | 4915/6470 [56:29<18:14,  1.42it/s] 
3|4916|Loss: 0.814389705657959:  76%|███████▌  | 4916/6470 [56:29<17:07,  1.51it/s]
3|4917|Loss: 0.929236888885498:  76%|███████▌  | 4916/6470 [56:29<17:07,  1.51it/s]
3|4917|Loss: 0.929236888885498:  76%|███████▌  | 4917/6470 [56:30<16:00,  1.62it/s]
3|4918|Loss: 0.8715023994445801:  76%|███████▌  | 4917/6470 [56:30<16:00,  1.62it/s]
3|4918|Loss: 0.8715023994445801:  76%|███████▌  | 4918/6470 [56:30<16:54,  1.53it/s]
3|4919|Loss: 0.841236412525177:  76%|███████▌  | 4918/6470 [56:31<16:54,  1.53it/s] 
3|4919|Loss: 0.841236412525177:  76%|███████▌  | 4919/6470 [56:31<16:39,  1.55it/s]
3|4920|Loss: 0.8595755696296692:  76%|███████▌  | 4919/6470 [56:31<16:39,  1.55it/s]
3|4920|Loss: 0.8595755696296692:  76%|███████▌  | 4920/6470 [56:32<16:54,  1.53it/s]
3|4921|Loss: 0.9560331702232361:  76%|███████▌  | 4920/6470 [56:32<16:54,  1.53it/s]
3|4921|Loss: 0.9560331702232361:  76%|███████▌  | 4921/6470 [56:32<16:02,  1.61it/s]
3|4922|Loss: 0.7394253015518188:  76%|███████▌  | 4921/6470 [56:33<16:02,  1.61it/s]
3|4922|Loss: 0.7394253015518188:  76%|███████▌  | 4922/6470 [56:33<16:38,  1.55it/s]
3|4923|Loss: 1.0065151453018188:  76%|███████▌  | 4922/6470 [56:33<16:38,  1.55it/s]
3|4923|Loss: 1.0065151453018188:  76%|███████▌  | 4923/6470 [56:34<16:50,  1.53it/s]
3|4924|Loss: 0.8613175749778748:  76%|███████▌  | 4923/6470 [56:34<16:50,  1.53it/s]
3|4924|Loss: 0.8613175749778748:  76%|███████▌  | 4924/6470 [56:34<17:22,  1.48it/s]
3|4925|Loss: 0.8733498454093933:  76%|███████▌  | 4924/6470 [56:35<17:22,  1.48it/s]
3|4925|Loss: 0.8733498454093933:  76%|███████▌  | 4925/6470 [56:35<17:11,  1.50it/s]
3|4926|Loss: 0.8964183330535889:  76%|███████▌  | 4925/6470 [56:35<17:11,  1.50it/s]
3|4926|Loss: 0.8964183330535889:  76%|███████▌  | 4926/6470 [56:36<17:33,  1.47it/s]
3|4927|Loss: 0.793850302696228:  76%|███████▌  | 4926/6470 [56:36<17:33,  1.47it/s] 
3|4927|Loss: 0.793850302696228:  76%|███████▌  | 4927/6470 [56:36<18:03,  1.42it/s]
3|4928|Loss: 0.9966962933540344:  76%|███████▌  | 4927/6470 [56:37<18:03,  1.42it/s]
3|4928|Loss: 0.9966962933540344:  76%|███████▌  | 4928/6470 [56:37<19:10,  1.34it/s]
3|4929|Loss: 0.8506554365158081:  76%|███████▌  | 4928/6470 [56:38<19:10,  1.34it/s]
3|4929|Loss: 0.8506554365158081:  76%|███████▌  | 4929/6470 [56:38<19:11,  1.34it/s]
3|4930|Loss: 0.773744523525238:  76%|███████▌  | 4929/6470 [56:38<19:11,  1.34it/s] 
3|4930|Loss: 0.773744523525238:  76%|███████▌  | 4930/6470 [56:39<18:17,  1.40it/s]
3|4931|Loss: 0.6979719400405884:  76%|███████▌  | 4930/6470 [56:39<18:17,  1.40it/s]
3|4931|Loss: 0.6979719400405884:  76%|███████▌  | 4931/6470 [56:39<18:36,  1.38it/s]
3|4932|Loss: 0.8844816088676453:  76%|███████▌  | 4931/6470 [56:40<18:36,  1.38it/s]
3|4932|Loss: 0.8844816088676453:  76%|███████▌  | 4932/6470 [56:40<19:30,  1.31it/s]
3|4933|Loss: 0.8448125123977661:  76%|███████▌  | 4932/6470 [56:41<19:30,  1.31it/s]
3|4933|Loss: 0.8448125123977661:  76%|███████▌  | 4933/6470 [56:41<19:21,  1.32it/s]
3|4934|Loss: 0.7563061714172363:  76%|███████▌  | 4933/6470 [56:41<19:21,  1.32it/s]
3|4934|Loss: 0.7563061714172363:  76%|███████▋  | 4934/6470 [56:42<19:05,  1.34it/s]
3|4935|Loss: 0.9469572901725769:  76%|███████▋  | 4934/6470 [56:42<19:05,  1.34it/s]
3|4935|Loss: 0.9469572901725769:  76%|███████▋  | 4935/6470 [56:42<17:26,  1.47it/s]
3|4936|Loss: 0.840653657913208:  76%|███████▋  | 4935/6470 [56:43<17:26,  1.47it/s] 
3|4936|Loss: 0.840653657913208:  76%|███████▋  | 4936/6470 [56:43<17:14,  1.48it/s]
3|4937|Loss: 0.7824903726577759:  76%|███████▋  | 4936/6470 [56:43<17:14,  1.48it/s]
3|4937|Loss: 0.7824903726577759:  76%|███████▋  | 4937/6470 [56:43<16:59,  1.50it/s]
3|4938|Loss: 0.7384597063064575:  76%|███████▋  | 4937/6470 [56:44<16:59,  1.50it/s]
3|4938|Loss: 0.7384597063064575:  76%|███████▋  | 4938/6470 [56:44<15:35,  1.64it/s]
3|4939|Loss: 0.8271847367286682:  76%|███████▋  | 4938/6470 [56:44<15:35,  1.64it/s]
3|4939|Loss: 0.8271847367286682:  76%|███████▋  | 4939/6470 [56:45<16:05,  1.59it/s]
3|4940|Loss: 0.9262198805809021:  76%|███████▋  | 4939/6470 [56:45<16:05,  1.59it/s]
3|4940|Loss: 0.9262198805809021:  76%|███████▋  | 4940/6470 [56:45<17:29,  1.46it/s]
3|4941|Loss: 0.9753795266151428:  76%|███████▋  | 4940/6470 [56:46<17:29,  1.46it/s]
3|4941|Loss: 0.9753795266151428:  76%|███████▋  | 4941/6470 [56:46<18:07,  1.41it/s]
3|4942|Loss: 0.8044973611831665:  76%|███████▋  | 4941/6470 [56:47<18:07,  1.41it/s]
3|4942|Loss: 0.8044973611831665:  76%|███████▋  | 4942/6470 [56:47<19:03,  1.34it/s]
3|4943|Loss: 0.9295850992202759:  76%|███████▋  | 4942/6470 [56:48<19:03,  1.34it/s]
3|4943|Loss: 0.9295850992202759:  76%|███████▋  | 4943/6470 [56:48<19:36,  1.30it/s]
3|4944|Loss: 0.975884735584259:  76%|███████▋  | 4943/6470 [56:48<19:36,  1.30it/s] 
3|4944|Loss: 0.975884735584259:  76%|███████▋  | 4944/6470 [56:48<18:11,  1.40it/s]
3|4945|Loss: 0.8573454022407532:  76%|███████▋  | 4944/6470 [56:49<18:11,  1.40it/s]
3|4945|Loss: 0.8573454022407532:  76%|███████▋  | 4945/6470 [56:49<17:01,  1.49it/s]
3|4946|Loss: 0.947524905204773:  76%|███████▋  | 4945/6470 [56:49<17:01,  1.49it/s] 
3|4946|Loss: 0.947524905204773:  76%|███████▋  | 4946/6470 [56:50<16:33,  1.53it/s]
3|4947|Loss: 0.79087233543396:  76%|███████▋  | 4946/6470 [56:50<16:33,  1.53it/s] 
3|4947|Loss: 0.79087233543396:  76%|███████▋  | 4947/6470 [56:50<15:37,  1.62it/s]
3|4948|Loss: 0.7272226810455322:  76%|███████▋  | 4947/6470 [56:51<15:37,  1.62it/s]
3|4948|Loss: 0.7272226810455322:  76%|███████▋  | 4948/6470 [56:51<15:28,  1.64it/s]
3|4949|Loss: 0.7165098786354065:  76%|███████▋  | 4948/6470 [56:51<15:28,  1.64it/s]
3|4949|Loss: 0.7165098786354065:  76%|███████▋  | 4949/6470 [56:51<15:14,  1.66it/s]
3|4950|Loss: 0.8125157356262207:  76%|███████▋  | 4949/6470 [56:52<15:14,  1.66it/s]
3|4950|Loss: 0.8125157356262207:  77%|███████▋  | 4950/6470 [56:52<16:35,  1.53it/s]
3|4951|Loss: 0.945091187953949:  77%|███████▋  | 4950/6470 [56:53<16:35,  1.53it/s] 
3|4951|Loss: 0.945091187953949:  77%|███████▋  | 4951/6470 [56:53<17:18,  1.46it/s]
3|4952|Loss: 0.8566177487373352:  77%|███████▋  | 4951/6470 [56:53<17:18,  1.46it/s]
3|4952|Loss: 0.8566177487373352:  77%|███████▋  | 4952/6470 [56:54<17:01,  1.49it/s]
3|4953|Loss: 0.7639652490615845:  77%|███████▋  | 4952/6470 [56:54<17:01,  1.49it/s]
3|4953|Loss: 0.7639652490615845:  77%|███████▋  | 4953/6470 [56:54<15:58,  1.58it/s]
3|4954|Loss: 0.9021787047386169:  77%|███████▋  | 4953/6470 [56:54<15:58,  1.58it/s]
3|4954|Loss: 0.9021787047386169:  77%|███████▋  | 4954/6470 [56:55<16:42,  1.51it/s]
3|4955|Loss: 1.027840495109558:  77%|███████▋  | 4954/6470 [56:55<16:42,  1.51it/s] 
3|4955|Loss: 1.027840495109558:  77%|███████▋  | 4955/6470 [56:56<17:21,  1.45it/s]
3|4956|Loss: 0.8669448494911194:  77%|███████▋  | 4955/6470 [56:56<17:21,  1.45it/s]
3|4956|Loss: 0.8669448494911194:  77%|███████▋  | 4956/6470 [56:56<17:31,  1.44it/s]
3|4957|Loss: 0.9622713923454285:  77%|███████▋  | 4956/6470 [56:57<17:31,  1.44it/s]
3|4957|Loss: 0.9622713923454285:  77%|███████▋  | 4957/6470 [56:57<18:18,  1.38it/s]
3|4958|Loss: 0.9899817705154419:  77%|███████▋  | 4957/6470 [56:58<18:18,  1.38it/s]
3|4958|Loss: 0.9899817705154419:  77%|███████▋  | 4958/6470 [56:58<19:19,  1.30it/s]
3|4959|Loss: 0.9253913760185242:  77%|███████▋  | 4958/6470 [56:58<19:19,  1.30it/s]
3|4959|Loss: 0.9253913760185242:  77%|███████▋  | 4959/6470 [56:59<18:14,  1.38it/s]
3|4960|Loss: 0.9267743825912476:  77%|███████▋  | 4959/6470 [56:59<18:14,  1.38it/s]
3|4960|Loss: 0.9267743825912476:  77%|███████▋  | 4960/6470 [56:59<18:00,  1.40it/s]
3|4961|Loss: 0.8005473613739014:  77%|███████▋  | 4960/6470 [57:00<18:00,  1.40it/s]
3|4961|Loss: 0.8005473613739014:  77%|███████▋  | 4961/6470 [57:00<18:07,  1.39it/s]
3|4962|Loss: 0.9084973335266113:  77%|███████▋  | 4961/6470 [57:00<18:07,  1.39it/s]
3|4962|Loss: 0.9084973335266113:  77%|███████▋  | 4962/6470 [57:00<16:31,  1.52it/s]
3|4963|Loss: 0.8405182361602783:  77%|███████▋  | 4962/6470 [57:01<16:31,  1.52it/s]
3|4963|Loss: 0.8405182361602783:  77%|███████▋  | 4963/6470 [57:01<17:18,  1.45it/s]
3|4964|Loss: 0.7767884731292725:  77%|███████▋  | 4963/6470 [57:02<17:18,  1.45it/s]
3|4964|Loss: 0.7767884731292725:  77%|███████▋  | 4964/6470 [57:02<18:09,  1.38it/s]
3|4965|Loss: 0.898481547832489:  77%|███████▋  | 4964/6470 [57:02<18:09,  1.38it/s] 
3|4965|Loss: 0.898481547832489:  77%|███████▋  | 4965/6470 [57:03<17:14,  1.46it/s]
3|4966|Loss: 0.788764238357544:  77%|███████▋  | 4965/6470 [57:03<17:14,  1.46it/s]
3|4966|Loss: 0.788764238357544:  77%|███████▋  | 4966/6470 [57:03<18:09,  1.38it/s]
3|4967|Loss: 0.8777102828025818:  77%|███████▋  | 4966/6470 [57:04<18:09,  1.38it/s]
3|4967|Loss: 0.8777102828025818:  77%|███████▋  | 4967/6470 [57:04<18:33,  1.35it/s]
3|4968|Loss: 0.8169289231300354:  77%|███████▋  | 4967/6470 [57:05<18:33,  1.35it/s]
3|4968|Loss: 0.8169289231300354:  77%|███████▋  | 4968/6470 [57:05<18:00,  1.39it/s]
3|4969|Loss: 0.8402268886566162:  77%|███████▋  | 4968/6470 [57:05<18:00,  1.39it/s]
3|4969|Loss: 0.8402268886566162:  77%|███████▋  | 4969/6470 [57:05<16:28,  1.52it/s]
3|4970|Loss: 0.7935939431190491:  77%|███████▋  | 4969/6470 [57:06<16:28,  1.52it/s]
3|4970|Loss: 0.7935939431190491:  77%|███████▋  | 4970/6470 [57:06<14:57,  1.67it/s]
3|4971|Loss: 0.9653741717338562:  77%|███████▋  | 4970/6470 [57:06<14:57,  1.67it/s]
3|4971|Loss: 0.9653741717338562:  77%|███████▋  | 4971/6470 [57:07<15:18,  1.63it/s]
3|4972|Loss: 0.7968331575393677:  77%|███████▋  | 4971/6470 [57:07<15:18,  1.63it/s]
3|4972|Loss: 0.7968331575393677:  77%|███████▋  | 4972/6470 [57:07<16:14,  1.54it/s]
3|4973|Loss: 0.6447461247444153:  77%|███████▋  | 4972/6470 [57:08<16:14,  1.54it/s]
3|4973|Loss: 0.6447461247444153:  77%|███████▋  | 4973/6470 [57:08<17:12,  1.45it/s]
3|4974|Loss: 0.8452959060668945:  77%|███████▋  | 4973/6470 [57:08<17:12,  1.45it/s]
3|4974|Loss: 0.8452959060668945:  77%|███████▋  | 4974/6470 [57:09<17:39,  1.41it/s]
3|4975|Loss: 0.8168702721595764:  77%|███████▋  | 4974/6470 [57:09<17:39,  1.41it/s]
3|4975|Loss: 0.8168702721595764:  77%|███████▋  | 4975/6470 [57:10<18:52,  1.32it/s]
3|4976|Loss: 0.7983850836753845:  77%|███████▋  | 4975/6470 [57:10<18:52,  1.32it/s]
3|4976|Loss: 0.7983850836753845:  77%|███████▋  | 4976/6470 [57:11<19:24,  1.28it/s]
3|4977|Loss: 0.8891298770904541:  77%|███████▋  | 4976/6470 [57:11<19:24,  1.28it/s]
3|4977|Loss: 0.8891298770904541:  77%|███████▋  | 4977/6470 [57:12<22:08,  1.12it/s]
3|4978|Loss: 0.7912653088569641:  77%|███████▋  | 4977/6470 [57:12<22:08,  1.12it/s]
3|4978|Loss: 0.7912653088569641:  77%|███████▋  | 4978/6470 [57:12<21:11,  1.17it/s]
3|4979|Loss: 0.9185793995857239:  77%|███████▋  | 4978/6470 [57:13<21:11,  1.17it/s]
3|4979|Loss: 0.9185793995857239:  77%|███████▋  | 4979/6470 [57:13<19:58,  1.24it/s]
3|4980|Loss: 0.7482042908668518:  77%|███████▋  | 4979/6470 [57:14<19:58,  1.24it/s]
3|4980|Loss: 0.7482042908668518:  77%|███████▋  | 4980/6470 [57:14<19:58,  1.24it/s]
3|4981|Loss: 0.9014149904251099:  77%|███████▋  | 4980/6470 [57:14<19:58,  1.24it/s]
3|4981|Loss: 0.9014149904251099:  77%|███████▋  | 4981/6470 [57:15<19:35,  1.27it/s]
3|4982|Loss: 0.7917768955230713:  77%|███████▋  | 4981/6470 [57:15<19:35,  1.27it/s]
3|4982|Loss: 0.7917768955230713:  77%|███████▋  | 4982/6470 [57:15<19:08,  1.30it/s]
3|4983|Loss: 0.9444038271903992:  77%|███████▋  | 4982/6470 [57:16<19:08,  1.30it/s]
3|4983|Loss: 0.9444038271903992:  77%|███████▋  | 4983/6470 [57:16<17:50,  1.39it/s]
3|4984|Loss: 0.8141318559646606:  77%|███████▋  | 4983/6470 [57:16<17:50,  1.39it/s]
3|4984|Loss: 0.8141318559646606:  77%|███████▋  | 4984/6470 [57:16<15:28,  1.60it/s]
3|4985|Loss: 0.9010860323905945:  77%|███████▋  | 4984/6470 [57:17<15:28,  1.60it/s]
3|4985|Loss: 0.9010860323905945:  77%|███████▋  | 4985/6470 [57:17<15:41,  1.58it/s]
3|4986|Loss: 0.9545506238937378:  77%|███████▋  | 4985/6470 [57:17<15:41,  1.58it/s]
3|4986|Loss: 0.9545506238937378:  77%|███████▋  | 4986/6470 [57:18<15:33,  1.59it/s]
3|4987|Loss: 0.8204387426376343:  77%|███████▋  | 4986/6470 [57:18<15:33,  1.59it/s]
3|4987|Loss: 0.8204387426376343:  77%|███████▋  | 4987/6470 [57:19<17:01,  1.45it/s]
3|4988|Loss: 0.8917048573493958:  77%|███████▋  | 4987/6470 [57:19<17:01,  1.45it/s]
3|4988|Loss: 0.8917048573493958:  77%|███████▋  | 4988/6470 [57:19<18:09,  1.36it/s]
3|4989|Loss: 0.8520647883415222:  77%|███████▋  | 4988/6470 [57:20<18:09,  1.36it/s]
3|4989|Loss: 0.8520647883415222:  77%|███████▋  | 4989/6470 [57:20<18:17,  1.35it/s]
3|4990|Loss: 0.8132786154747009:  77%|███████▋  | 4989/6470 [57:21<18:17,  1.35it/s]
3|4990|Loss: 0.8132786154747009:  77%|███████▋  | 4990/6470 [57:21<18:56,  1.30it/s]
3|4991|Loss: 0.8431674242019653:  77%|███████▋  | 4990/6470 [57:21<18:56,  1.30it/s]
3|4991|Loss: 0.8431674242019653:  77%|███████▋  | 4991/6470 [57:22<18:23,  1.34it/s]
3|4992|Loss: 0.7819938659667969:  77%|███████▋  | 4991/6470 [57:22<18:23,  1.34it/s]
3|4992|Loss: 0.7819938659667969:  77%|███████▋  | 4992/6470 [57:22<18:08,  1.36it/s]
3|4993|Loss: 0.8653611540794373:  77%|███████▋  | 4992/6470 [57:23<18:08,  1.36it/s]
3|4993|Loss: 0.8653611540794373:  77%|███████▋  | 4993/6470 [57:23<17:56,  1.37it/s]
3|4994|Loss: 0.9608369469642639:  77%|███████▋  | 4993/6470 [57:23<17:56,  1.37it/s]
3|4994|Loss: 0.9608369469642639:  77%|███████▋  | 4994/6470 [57:24<16:56,  1.45it/s]
3|4995|Loss: 0.7537895441055298:  77%|███████▋  | 4994/6470 [57:24<16:56,  1.45it/s]
3|4995|Loss: 0.7537895441055298:  77%|███████▋  | 4995/6470 [57:24<17:03,  1.44it/s]
3|4996|Loss: 0.7871744632720947:  77%|███████▋  | 4995/6470 [57:25<17:03,  1.44it/s]
3|4996|Loss: 0.7871744632720947:  77%|███████▋  | 4996/6470 [57:25<16:25,  1.50it/s]
3|4997|Loss: 1.0703462362289429:  77%|███████▋  | 4996/6470 [57:25<16:25,  1.50it/s]
3|4997|Loss: 1.0703462362289429:  77%|███████▋  | 4997/6470 [57:26<15:21,  1.60it/s]
3|4998|Loss: 0.823107898235321:  77%|███████▋  | 4997/6470 [57:26<15:21,  1.60it/s] 
3|4998|Loss: 0.823107898235321:  77%|███████▋  | 4998/6470 [57:26<15:05,  1.63it/s]
3|4999|Loss: 0.7176998257637024:  77%|███████▋  | 4998/6470 [57:26<15:05,  1.63it/s]
3|4999|Loss: 0.7176998257637024:  77%|███████▋  | 4999/6470 [57:27<14:49,  1.65it/s]
3|5000|Loss: 0.8269664645195007:  77%|███████▋  | 4999/6470 [57:27<14:49,  1.65it/s]
3|5000|Loss: 0.8269664645195007:  77%|███████▋  | 5000/6470 [57:27<15:23,  1.59it/s]
3|5001|Loss: 0.7062429189682007:  77%|███████▋  | 5000/6470 [57:28<15:23,  1.59it/s]
3|5001|Loss: 0.7062429189682007:  77%|███████▋  | 5001/6470 [57:28<16:27,  1.49it/s]
3|5002|Loss: 0.7071964740753174:  77%|███████▋  | 5001/6470 [57:29<16:27,  1.49it/s]
3|5002|Loss: 0.7071964740753174:  77%|███████▋  | 5002/6470 [57:29<17:38,  1.39it/s]
3|5003|Loss: 0.8639450669288635:  77%|███████▋  | 5002/6470 [57:29<17:38,  1.39it/s]
3|5003|Loss: 0.8639450669288635:  77%|███████▋  | 5003/6470 [57:30<17:59,  1.36it/s]
3|5004|Loss: 1.0114322900772095:  77%|███████▋  | 5003/6470 [57:30<17:59,  1.36it/s]
3|5004|Loss: 1.0114322900772095:  77%|███████▋  | 5004/6470 [57:30<18:03,  1.35it/s]
3|5005|Loss: 1.0532108545303345:  77%|███████▋  | 5004/6470 [57:31<18:03,  1.35it/s]
3|5005|Loss: 1.0532108545303345:  77%|███████▋  | 5005/6470 [57:31<17:23,  1.40it/s]
3|5006|Loss: 0.7004442811012268:  77%|███████▋  | 5005/6470 [57:31<17:23,  1.40it/s]
3|5006|Loss: 0.7004442811012268:  77%|███████▋  | 5006/6470 [57:32<16:39,  1.46it/s]
3|5007|Loss: 1.0411620140075684:  77%|███████▋  | 5006/6470 [57:32<16:39,  1.46it/s]
3|5007|Loss: 1.0411620140075684:  77%|███████▋  | 5007/6470 [57:32<16:12,  1.50it/s]
3|5008|Loss: 0.7580615282058716:  77%|███████▋  | 5007/6470 [57:33<16:12,  1.50it/s]
3|5008|Loss: 0.7580615282058716:  77%|███████▋  | 5008/6470 [57:33<16:34,  1.47it/s]
3|5009|Loss: 0.725467324256897:  77%|███████▋  | 5008/6470 [57:33<16:34,  1.47it/s] 
3|5009|Loss: 0.725467324256897:  77%|███████▋  | 5009/6470 [57:34<16:20,  1.49it/s]
3|5010|Loss: 0.7268600463867188:  77%|███████▋  | 5009/6470 [57:34<16:20,  1.49it/s]
3|5010|Loss: 0.7268600463867188:  77%|███████▋  | 5010/6470 [57:34<16:26,  1.48it/s]
3|5011|Loss: 0.8691284656524658:  77%|███████▋  | 5010/6470 [57:35<16:26,  1.48it/s]
3|5011|Loss: 0.8691284656524658:  77%|███████▋  | 5011/6470 [57:35<15:51,  1.53it/s]
3|5012|Loss: 0.7546992897987366:  77%|███████▋  | 5011/6470 [57:35<15:51,  1.53it/s]
3|5012|Loss: 0.7546992897987366:  77%|███████▋  | 5012/6470 [57:36<17:20,  1.40it/s]
3|5013|Loss: 0.848959743976593:  77%|███████▋  | 5012/6470 [57:36<17:20,  1.40it/s] 
3|5013|Loss: 0.848959743976593:  77%|███████▋  | 5013/6470 [57:37<17:43,  1.37it/s]
3|5014|Loss: 0.8538938760757446:  77%|███████▋  | 5013/6470 [57:37<17:43,  1.37it/s]
3|5014|Loss: 0.8538938760757446:  77%|███████▋  | 5014/6470 [57:37<18:07,  1.34it/s]
3|5015|Loss: 0.7892211079597473:  77%|███████▋  | 5014/6470 [57:38<18:07,  1.34it/s]
3|5015|Loss: 0.7892211079597473:  78%|███████▊  | 5015/6470 [57:38<17:03,  1.42it/s]
3|5016|Loss: 0.8484906554222107:  78%|███████▊  | 5015/6470 [57:38<17:03,  1.42it/s]
3|5016|Loss: 0.8484906554222107:  78%|███████▊  | 5016/6470 [57:39<17:25,  1.39it/s]
3|5017|Loss: 1.0008771419525146:  78%|███████▊  | 5016/6470 [57:39<17:25,  1.39it/s]
3|5017|Loss: 1.0008771419525146:  78%|███████▊  | 5017/6470 [57:40<18:09,  1.33it/s]
3|5018|Loss: 0.8441413640975952:  78%|███████▊  | 5017/6470 [57:40<18:09,  1.33it/s]
3|5018|Loss: 0.8441413640975952:  78%|███████▊  | 5018/6470 [57:40<18:17,  1.32it/s]
3|5019|Loss: 0.9726971387863159:  78%|███████▊  | 5018/6470 [57:41<18:17,  1.32it/s]
3|5019|Loss: 0.9726971387863159:  78%|███████▊  | 5019/6470 [57:41<18:15,  1.32it/s]
3|5020|Loss: 0.7771577835083008:  78%|███████▊  | 5019/6470 [57:42<18:15,  1.32it/s]
3|5020|Loss: 0.7771577835083008:  78%|███████▊  | 5020/6470 [57:42<18:12,  1.33it/s]
3|5021|Loss: 0.8051596283912659:  78%|███████▊  | 5020/6470 [57:42<18:12,  1.33it/s]
3|5021|Loss: 0.8051596283912659:  78%|███████▊  | 5021/6470 [57:43<18:07,  1.33it/s]
3|5022|Loss: 0.9415526390075684:  78%|███████▊  | 5021/6470 [57:43<18:07,  1.33it/s]
3|5022|Loss: 0.9415526390075684:  78%|███████▊  | 5022/6470 [57:43<17:20,  1.39it/s]
3|5023|Loss: 0.8596345782279968:  78%|███████▊  | 5022/6470 [57:44<17:20,  1.39it/s]
3|5023|Loss: 0.8596345782279968:  78%|███████▊  | 5023/6470 [57:44<15:38,  1.54it/s]
3|5024|Loss: 0.8495959043502808:  78%|███████▊  | 5023/6470 [57:44<15:38,  1.54it/s]
3|5024|Loss: 0.8495959043502808:  78%|███████▊  | 5024/6470 [57:44<16:05,  1.50it/s]
3|5025|Loss: 0.8053849339485168:  78%|███████▊  | 5024/6470 [57:45<16:05,  1.50it/s]
3|5025|Loss: 0.8053849339485168:  78%|███████▊  | 5025/6470 [57:45<16:02,  1.50it/s]
3|5026|Loss: 0.7798163294792175:  78%|███████▊  | 5025/6470 [57:45<16:02,  1.50it/s]
3|5026|Loss: 0.7798163294792175:  78%|███████▊  | 5026/6470 [57:46<15:18,  1.57it/s]
3|5027|Loss: 0.7292501926422119:  78%|███████▊  | 5026/6470 [57:46<15:18,  1.57it/s]
3|5027|Loss: 0.7292501926422119:  78%|███████▊  | 5027/6470 [57:46<15:04,  1.60it/s]
3|5028|Loss: 0.8621633648872375:  78%|███████▊  | 5027/6470 [57:47<15:04,  1.60it/s]
3|5028|Loss: 0.8621633648872375:  78%|███████▊  | 5028/6470 [57:47<16:00,  1.50it/s]
3|5029|Loss: 0.9478658437728882:  78%|███████▊  | 5028/6470 [57:47<16:00,  1.50it/s]
3|5029|Loss: 0.9478658437728882:  78%|███████▊  | 5029/6470 [57:48<16:04,  1.49it/s]
3|5030|Loss: 0.925214409828186:  78%|███████▊  | 5029/6470 [57:48<16:04,  1.49it/s] 
3|5030|Loss: 0.925214409828186:  78%|███████▊  | 5030/6470 [57:48<15:45,  1.52it/s]
3|5031|Loss: 0.6874963641166687:  78%|███████▊  | 5030/6470 [57:49<15:45,  1.52it/s]
3|5031|Loss: 0.6874963641166687:  78%|███████▊  | 5031/6470 [57:49<17:01,  1.41it/s]
3|5032|Loss: 0.7140448093414307:  78%|███████▊  | 5031/6470 [57:50<17:01,  1.41it/s]
3|5032|Loss: 0.7140448093414307:  78%|███████▊  | 5032/6470 [57:50<18:15,  1.31it/s]
3|5033|Loss: 1.039577841758728:  78%|███████▊  | 5032/6470 [57:50<18:15,  1.31it/s] 
3|5033|Loss: 1.039577841758728:  78%|███████▊  | 5033/6470 [57:51<17:43,  1.35it/s]
3|5034|Loss: 0.7915429472923279:  78%|███████▊  | 5033/6470 [57:51<17:43,  1.35it/s]
3|5034|Loss: 0.7915429472923279:  78%|███████▊  | 5034/6470 [57:51<17:32,  1.36it/s]
3|5035|Loss: 0.9710434675216675:  78%|███████▊  | 5034/6470 [57:52<17:32,  1.36it/s]
3|5035|Loss: 0.9710434675216675:  78%|███████▊  | 5035/6470 [57:52<17:02,  1.40it/s]
3|5036|Loss: 0.7764344215393066:  78%|███████▊  | 5035/6470 [57:53<17:02,  1.40it/s]
3|5036|Loss: 0.7764344215393066:  78%|███████▊  | 5036/6470 [57:53<16:53,  1.42it/s]
3|5037|Loss: 0.8379654884338379:  78%|███████▊  | 5036/6470 [57:53<16:53,  1.42it/s]
3|5037|Loss: 0.8379654884338379:  78%|███████▊  | 5037/6470 [57:54<17:35,  1.36it/s]
3|5038|Loss: 0.8576743602752686:  78%|███████▊  | 5037/6470 [57:54<17:35,  1.36it/s]
3|5038|Loss: 0.8576743602752686:  78%|███████▊  | 5038/6470 [57:54<18:19,  1.30it/s]
3|5039|Loss: 0.9401963949203491:  78%|███████▊  | 5038/6470 [57:55<18:19,  1.30it/s]
3|5039|Loss: 0.9401963949203491:  78%|███████▊  | 5039/6470 [57:55<18:19,  1.30it/s]
3|5040|Loss: 0.9137752056121826:  78%|███████▊  | 5039/6470 [57:56<18:19,  1.30it/s]
3|5040|Loss: 0.9137752056121826:  78%|███████▊  | 5040/6470 [57:56<17:35,  1.35it/s]
3|5041|Loss: 1.0141870975494385:  78%|███████▊  | 5040/6470 [57:56<17:35,  1.35it/s]
3|5041|Loss: 1.0141870975494385:  78%|███████▊  | 5041/6470 [57:57<16:45,  1.42it/s]
3|5042|Loss: 0.7769086360931396:  78%|███████▊  | 5041/6470 [57:57<16:45,  1.42it/s]
3|5042|Loss: 0.7769086360931396:  78%|███████▊  | 5042/6470 [57:57<17:22,  1.37it/s]
3|5043|Loss: 0.9437909126281738:  78%|███████▊  | 5042/6470 [57:58<17:22,  1.37it/s]
3|5043|Loss: 0.9437909126281738:  78%|███████▊  | 5043/6470 [57:58<17:37,  1.35it/s]
3|5044|Loss: 0.8360955119132996:  78%|███████▊  | 5043/6470 [57:59<17:37,  1.35it/s]
3|5044|Loss: 0.8360955119132996:  78%|███████▊  | 5044/6470 [57:59<17:58,  1.32it/s]
3|5045|Loss: 0.9553149938583374:  78%|███████▊  | 5044/6470 [57:59<17:58,  1.32it/s]
3|5045|Loss: 0.9553149938583374:  78%|███████▊  | 5045/6470 [58:00<18:34,  1.28it/s]
3|5046|Loss: 0.8808250427246094:  78%|███████▊  | 5045/6470 [58:00<18:34,  1.28it/s]
3|5046|Loss: 0.8808250427246094:  78%|███████▊  | 5046/6470 [58:00<17:48,  1.33it/s]
3|5047|Loss: 0.804375946521759:  78%|███████▊  | 5046/6470 [58:01<17:48,  1.33it/s] 
3|5047|Loss: 0.804375946521759:  78%|███████▊  | 5047/6470 [58:01<17:54,  1.32it/s]
3|5048|Loss: 0.804699718952179:  78%|███████▊  | 5047/6470 [58:02<17:54,  1.32it/s]
3|5048|Loss: 0.804699718952179:  78%|███████▊  | 5048/6470 [58:02<18:23,  1.29it/s]
3|5049|Loss: 0.9006465673446655:  78%|███████▊  | 5048/6470 [58:02<18:23,  1.29it/s]
3|5049|Loss: 0.9006465673446655:  78%|███████▊  | 5049/6470 [58:03<17:37,  1.34it/s]
3|5050|Loss: 0.7588563561439514:  78%|███████▊  | 5049/6470 [58:03<17:37,  1.34it/s]
3|5050|Loss: 0.7588563561439514:  78%|███████▊  | 5050/6470 [58:03<17:11,  1.38it/s]
3|5051|Loss: 0.9181487560272217:  78%|███████▊  | 5050/6470 [58:04<17:11,  1.38it/s]
3|5051|Loss: 0.9181487560272217:  78%|███████▊  | 5051/6470 [58:04<17:44,  1.33it/s]
3|5052|Loss: 0.8992993831634521:  78%|███████▊  | 5051/6470 [58:05<17:44,  1.33it/s]
3|5052|Loss: 0.8992993831634521:  78%|███████▊  | 5052/6470 [58:05<18:06,  1.31it/s]
3|5053|Loss: 0.7273048758506775:  78%|███████▊  | 5052/6470 [58:05<18:06,  1.31it/s]
3|5053|Loss: 0.7273048758506775:  78%|███████▊  | 5053/6470 [58:06<17:04,  1.38it/s]
3|5054|Loss: 0.8437669277191162:  78%|███████▊  | 5053/6470 [58:06<17:04,  1.38it/s]
3|5054|Loss: 0.8437669277191162:  78%|███████▊  | 5054/6470 [58:06<16:04,  1.47it/s]
3|5055|Loss: 0.8219665884971619:  78%|███████▊  | 5054/6470 [58:07<16:04,  1.47it/s]
3|5055|Loss: 0.8219665884971619:  78%|███████▊  | 5055/6470 [58:07<15:20,  1.54it/s]
3|5056|Loss: 0.847169041633606:  78%|███████▊  | 5055/6470 [58:07<15:20,  1.54it/s] 
3|5056|Loss: 0.847169041633606:  78%|███████▊  | 5056/6470 [58:07<14:53,  1.58it/s]
3|5057|Loss: 0.8048124313354492:  78%|███████▊  | 5056/6470 [58:08<14:53,  1.58it/s]
3|5057|Loss: 0.8048124313354492:  78%|███████▊  | 5057/6470 [58:08<15:28,  1.52it/s]
3|5058|Loss: 0.9529951810836792:  78%|███████▊  | 5057/6470 [58:09<15:28,  1.52it/s]
3|5058|Loss: 0.9529951810836792:  78%|███████▊  | 5058/6470 [58:09<16:43,  1.41it/s]
3|5059|Loss: 0.9182744026184082:  78%|███████▊  | 5058/6470 [58:09<16:43,  1.41it/s]
3|5059|Loss: 0.9182744026184082:  78%|███████▊  | 5059/6470 [58:10<16:34,  1.42it/s]
3|5060|Loss: 0.818681538105011:  78%|███████▊  | 5059/6470 [58:10<16:34,  1.42it/s] 
3|5060|Loss: 0.818681538105011:  78%|███████▊  | 5060/6470 [58:10<15:01,  1.56it/s]
3|5061|Loss: 0.9301971197128296:  78%|███████▊  | 5060/6470 [58:10<15:01,  1.56it/s]
3|5061|Loss: 0.9301971197128296:  78%|███████▊  | 5061/6470 [58:11<15:55,  1.47it/s]
3|5062|Loss: 0.7926526069641113:  78%|███████▊  | 5061/6470 [58:11<15:55,  1.47it/s]
3|5062|Loss: 0.7926526069641113:  78%|███████▊  | 5062/6470 [58:12<16:31,  1.42it/s]
3|5063|Loss: 0.8439569473266602:  78%|███████▊  | 5062/6470 [58:12<16:31,  1.42it/s]
3|5063|Loss: 0.8439569473266602:  78%|███████▊  | 5063/6470 [58:12<15:09,  1.55it/s]
3|5064|Loss: 0.9133191108703613:  78%|███████▊  | 5063/6470 [58:13<15:09,  1.55it/s]
3|5064|Loss: 0.9133191108703613:  78%|███████▊  | 5064/6470 [58:13<16:08,  1.45it/s]
3|5065|Loss: 0.8137344717979431:  78%|███████▊  | 5064/6470 [58:13<16:08,  1.45it/s]
3|5065|Loss: 0.8137344717979431:  78%|███████▊  | 5065/6470 [58:13<15:25,  1.52it/s]
3|5066|Loss: 0.7147778272628784:  78%|███████▊  | 5065/6470 [58:14<15:25,  1.52it/s]
3|5066|Loss: 0.7147778272628784:  78%|███████▊  | 5066/6470 [58:14<16:07,  1.45it/s]
3|5067|Loss: 0.8350033760070801:  78%|███████▊  | 5066/6470 [58:15<16:07,  1.45it/s]
3|5067|Loss: 0.8350033760070801:  78%|███████▊  | 5067/6470 [58:15<15:31,  1.51it/s]
3|5068|Loss: 0.9602726101875305:  78%|███████▊  | 5067/6470 [58:15<15:31,  1.51it/s]
3|5068|Loss: 0.9602726101875305:  78%|███████▊  | 5068/6470 [58:16<15:44,  1.48it/s]
3|5069|Loss: 0.7530121803283691:  78%|███████▊  | 5068/6470 [58:16<15:44,  1.48it/s]
3|5069|Loss: 0.7530121803283691:  78%|███████▊  | 5069/6470 [58:16<15:28,  1.51it/s]
3|5070|Loss: 0.9898911714553833:  78%|███████▊  | 5069/6470 [58:17<15:28,  1.51it/s]
3|5070|Loss: 0.9898911714553833:  78%|███████▊  | 5070/6470 [58:17<15:46,  1.48it/s]
3|5071|Loss: 0.8600685596466064:  78%|███████▊  | 5070/6470 [58:17<15:46,  1.48it/s]
3|5071|Loss: 0.8600685596466064:  78%|███████▊  | 5071/6470 [58:18<16:01,  1.46it/s]
3|5072|Loss: 0.8016389608383179:  78%|███████▊  | 5071/6470 [58:18<16:01,  1.46it/s]
3|5072|Loss: 0.8016389608383179:  78%|███████▊  | 5072/6470 [58:18<15:46,  1.48it/s]
3|5073|Loss: 0.7683337926864624:  78%|███████▊  | 5072/6470 [58:19<15:46,  1.48it/s]
3|5073|Loss: 0.7683337926864624:  78%|███████▊  | 5073/6470 [58:19<14:27,  1.61it/s]
3|5074|Loss: 0.8226929306983948:  78%|███████▊  | 5073/6470 [58:19<14:27,  1.61it/s]
3|5074|Loss: 0.8226929306983948:  78%|███████▊  | 5074/6470 [58:19<14:34,  1.60it/s]
3|5075|Loss: 0.9755027890205383:  78%|███████▊  | 5074/6470 [58:20<14:34,  1.60it/s]
3|5075|Loss: 0.9755027890205383:  78%|███████▊  | 5075/6470 [58:20<15:17,  1.52it/s]
3|5076|Loss: 0.778931200504303:  78%|███████▊  | 5075/6470 [58:21<15:17,  1.52it/s] 
3|5076|Loss: 0.778931200504303:  78%|███████▊  | 5076/6470 [58:21<15:49,  1.47it/s]
3|5077|Loss: 0.8545024991035461:  78%|███████▊  | 5076/6470 [58:21<15:49,  1.47it/s]
3|5077|Loss: 0.8545024991035461:  78%|███████▊  | 5077/6470 [58:22<16:11,  1.43it/s]
3|5078|Loss: 0.9172505140304565:  78%|███████▊  | 5077/6470 [58:22<16:11,  1.43it/s]
3|5078|Loss: 0.9172505140304565:  78%|███████▊  | 5078/6470 [58:22<16:31,  1.40it/s]
3|5079|Loss: 0.9396121501922607:  78%|███████▊  | 5078/6470 [58:23<16:31,  1.40it/s]
3|5079|Loss: 0.9396121501922607:  79%|███████▊  | 5079/6470 [58:23<17:12,  1.35it/s]
3|5080|Loss: 0.767128050327301:  79%|███████▊  | 5079/6470 [58:24<17:12,  1.35it/s] 
3|5080|Loss: 0.767128050327301:  79%|███████▊  | 5080/6470 [58:24<16:25,  1.41it/s]
3|5081|Loss: 0.7627641558647156:  79%|███████▊  | 5080/6470 [58:24<16:25,  1.41it/s]
3|5081|Loss: 0.7627641558647156:  79%|███████▊  | 5081/6470 [58:24<15:08,  1.53it/s]
3|5082|Loss: 0.9113729596138:  79%|███████▊  | 5081/6470 [58:25<15:08,  1.53it/s]   
3|5082|Loss: 0.9113729596138:  79%|███████▊  | 5082/6470 [58:25<15:10,  1.53it/s]
3|5083|Loss: 0.8001921772956848:  79%|███████▊  | 5082/6470 [58:25<15:10,  1.53it/s]
3|5083|Loss: 0.8001921772956848:  79%|███████▊  | 5083/6470 [58:26<15:08,  1.53it/s]
3|5084|Loss: 0.7547192573547363:  79%|███████▊  | 5083/6470 [58:26<15:08,  1.53it/s]
3|5084|Loss: 0.7547192573547363:  79%|███████▊  | 5084/6470 [58:26<16:03,  1.44it/s]
3|5085|Loss: 0.9592137932777405:  79%|███████▊  | 5084/6470 [58:27<16:03,  1.44it/s]
3|5085|Loss: 0.9592137932777405:  79%|███████▊  | 5085/6470 [58:27<15:49,  1.46it/s]
3|5086|Loss: 0.8334314227104187:  79%|███████▊  | 5085/6470 [58:27<15:49,  1.46it/s]
3|5086|Loss: 0.8334314227104187:  79%|███████▊  | 5086/6470 [58:28<16:25,  1.40it/s]
3|5087|Loss: 0.9606897234916687:  79%|███████▊  | 5086/6470 [58:28<16:25,  1.40it/s]
3|5087|Loss: 0.9606897234916687:  79%|███████▊  | 5087/6470 [58:29<16:31,  1.40it/s]
3|5088|Loss: 0.7933781743049622:  79%|███████▊  | 5087/6470 [58:29<16:31,  1.40it/s]
3|5088|Loss: 0.7933781743049622:  79%|███████▊  | 5088/6470 [58:29<17:04,  1.35it/s]
3|5089|Loss: 1.0920780897140503:  79%|███████▊  | 5088/6470 [58:30<17:04,  1.35it/s]
3|5089|Loss: 1.0920780897140503:  79%|███████▊  | 5089/6470 [58:30<17:40,  1.30it/s]
3|5090|Loss: 0.7711858749389648:  79%|███████▊  | 5089/6470 [58:31<17:40,  1.30it/s]
3|5090|Loss: 0.7711858749389648:  79%|███████▊  | 5090/6470 [58:31<17:02,  1.35it/s]
3|5091|Loss: 0.9133847951889038:  79%|███████▊  | 5090/6470 [58:31<17:02,  1.35it/s]
3|5091|Loss: 0.9133847951889038:  79%|███████▊  | 5091/6470 [58:32<17:29,  1.31it/s]
3|5092|Loss: 0.962725818157196:  79%|███████▊  | 5091/6470 [58:32<17:29,  1.31it/s] 
3|5092|Loss: 0.962725818157196:  79%|███████▊  | 5092/6470 [58:32<17:00,  1.35it/s]
3|5093|Loss: 0.8924787044525146:  79%|███████▊  | 5092/6470 [58:33<17:00,  1.35it/s]
3|5093|Loss: 0.8924787044525146:  79%|███████▊  | 5093/6470 [58:33<16:19,  1.41it/s]
3|5094|Loss: 0.8480042219161987:  79%|███████▊  | 5093/6470 [58:33<16:19,  1.41it/s]
3|5094|Loss: 0.8480042219161987:  79%|███████▊  | 5094/6470 [58:33<14:26,  1.59it/s]
3|5095|Loss: 0.9201985001564026:  79%|███████▊  | 5094/6470 [58:34<14:26,  1.59it/s]
3|5095|Loss: 0.9201985001564026:  79%|███████▊  | 5095/6470 [58:34<14:14,  1.61it/s]
3|5096|Loss: 1.0555402040481567:  79%|███████▊  | 5095/6470 [58:35<14:14,  1.61it/s]
3|5096|Loss: 1.0555402040481567:  79%|███████▉  | 5096/6470 [58:35<15:33,  1.47it/s]
3|5097|Loss: 0.7313493490219116:  79%|███████▉  | 5096/6470 [58:35<15:33,  1.47it/s]
3|5097|Loss: 0.7313493490219116:  79%|███████▉  | 5097/6470 [58:36<16:56,  1.35it/s]
3|5098|Loss: 0.7344476580619812:  79%|███████▉  | 5097/6470 [58:36<16:56,  1.35it/s]
3|5098|Loss: 0.7344476580619812:  79%|███████▉  | 5098/6470 [58:37<17:48,  1.28it/s]
3|5099|Loss: 0.9743039608001709:  79%|███████▉  | 5098/6470 [58:37<17:48,  1.28it/s]
3|5099|Loss: 0.9743039608001709:  79%|███████▉  | 5099/6470 [58:37<18:15,  1.25it/s]
3|5100|Loss: 0.831866979598999:  79%|███████▉  | 5099/6470 [58:38<18:15,  1.25it/s] 
3|5100|Loss: 0.831866979598999:  79%|███████▉  | 5100/6470 [58:38<17:20,  1.32it/s]
3|5101|Loss: 1.034611463546753:  79%|███████▉  | 5100/6470 [58:39<17:20,  1.32it/s]
3|5101|Loss: 1.034611463546753:  79%|███████▉  | 5101/6470 [58:39<17:43,  1.29it/s]
3|5102|Loss: 0.816249430179596:  79%|███████▉  | 5101/6470 [58:39<17:43,  1.29it/s]
3|5102|Loss: 0.816249430179596:  79%|███████▉  | 5102/6470 [58:40<18:41,  1.22it/s]
3|5103|Loss: 0.7965468764305115:  79%|███████▉  | 5102/6470 [58:40<18:41,  1.22it/s]
3|5103|Loss: 0.7965468764305115:  79%|███████▉  | 5103/6470 [58:40<16:29,  1.38it/s]
3|5104|Loss: 0.9301865100860596:  79%|███████▉  | 5103/6470 [58:41<16:29,  1.38it/s]
3|5104|Loss: 0.9301865100860596:  79%|███████▉  | 5104/6470 [58:41<16:10,  1.41it/s]
3|5105|Loss: 0.9448021650314331:  79%|███████▉  | 5104/6470 [58:41<16:10,  1.41it/s]
3|5105|Loss: 0.9448021650314331:  79%|███████▉  | 5105/6470 [58:42<15:15,  1.49it/s]
3|5106|Loss: 0.8921261429786682:  79%|███████▉  | 5105/6470 [58:42<15:15,  1.49it/s]
3|5106|Loss: 0.8921261429786682:  79%|███████▉  | 5106/6470 [58:42<15:14,  1.49it/s]
3|5107|Loss: 0.8970445394515991:  79%|███████▉  | 5106/6470 [58:43<15:14,  1.49it/s]
3|5107|Loss: 0.8970445394515991:  79%|███████▉  | 5107/6470 [58:43<14:51,  1.53it/s]
3|5108|Loss: 0.8512646555900574:  79%|███████▉  | 5107/6470 [58:43<14:51,  1.53it/s]
3|5108|Loss: 0.8512646555900574:  79%|███████▉  | 5108/6470 [58:44<14:32,  1.56it/s]
3|5109|Loss: 0.7866681814193726:  79%|███████▉  | 5108/6470 [58:44<14:32,  1.56it/s]
3|5109|Loss: 0.7866681814193726:  79%|███████▉  | 5109/6470 [58:44<14:59,  1.51it/s]
3|5110|Loss: 0.8812263607978821:  79%|███████▉  | 5109/6470 [58:45<14:59,  1.51it/s]
3|5110|Loss: 0.8812263607978821:  79%|███████▉  | 5110/6470 [58:45<16:08,  1.40it/s]
3|5111|Loss: 0.7745165228843689:  79%|███████▉  | 5110/6470 [58:46<16:08,  1.40it/s]
3|5111|Loss: 0.7745165228843689:  79%|███████▉  | 5111/6470 [58:46<16:19,  1.39it/s]
3|5112|Loss: 0.8309382200241089:  79%|███████▉  | 5111/6470 [58:46<16:19,  1.39it/s]
3|5112|Loss: 0.8309382200241089:  79%|███████▉  | 5112/6470 [58:46<15:43,  1.44it/s]
3|5113|Loss: 0.7757906317710876:  79%|███████▉  | 5112/6470 [58:47<15:43,  1.44it/s]
3|5113|Loss: 0.7757906317710876:  79%|███████▉  | 5113/6470 [58:47<15:13,  1.49it/s]
3|5114|Loss: 0.7410340905189514:  79%|███████▉  | 5113/6470 [58:47<15:13,  1.49it/s]
3|5114|Loss: 0.7410340905189514:  79%|███████▉  | 5114/6470 [58:48<15:28,  1.46it/s]
3|5115|Loss: 0.8245592713356018:  79%|███████▉  | 5114/6470 [58:48<15:28,  1.46it/s]
3|5115|Loss: 0.8245592713356018:  79%|███████▉  | 5115/6470 [58:48<14:16,  1.58it/s]
3|5116|Loss: 0.9382721781730652:  79%|███████▉  | 5115/6470 [58:49<14:16,  1.58it/s]
3|5116|Loss: 0.9382721781730652:  79%|███████▉  | 5116/6470 [58:49<14:03,  1.60it/s]
3|5117|Loss: 0.8832601308822632:  79%|███████▉  | 5116/6470 [58:49<14:03,  1.60it/s]
3|5117|Loss: 0.8832601308822632:  79%|███████▉  | 5117/6470 [58:50<15:12,  1.48it/s]
3|5118|Loss: 0.785717785358429:  79%|███████▉  | 5117/6470 [58:50<15:12,  1.48it/s] 
3|5118|Loss: 0.785717785358429:  79%|███████▉  | 5118/6470 [58:50<15:17,  1.47it/s]
3|5119|Loss: 0.9242696762084961:  79%|███████▉  | 5118/6470 [58:51<15:17,  1.47it/s]
3|5119|Loss: 0.9242696762084961:  79%|███████▉  | 5119/6470 [58:51<14:19,  1.57it/s]
3|5120|Loss: 0.9330076575279236:  79%|███████▉  | 5119/6470 [58:51<14:19,  1.57it/s]
3|5120|Loss: 0.9330076575279236:  79%|███████▉  | 5120/6470 [58:52<15:19,  1.47it/s]
3|5121|Loss: 0.9183092713356018:  79%|███████▉  | 5120/6470 [58:52<15:19,  1.47it/s]
3|5121|Loss: 0.9183092713356018:  79%|███████▉  | 5121/6470 [58:52<15:57,  1.41it/s]
3|5122|Loss: 0.7799316048622131:  79%|███████▉  | 5121/6470 [58:53<15:57,  1.41it/s]
3|5122|Loss: 0.7799316048622131:  79%|███████▉  | 5122/6470 [58:53<16:02,  1.40it/s]
3|5123|Loss: 0.8675235509872437:  79%|███████▉  | 5122/6470 [58:54<16:02,  1.40it/s]
3|5123|Loss: 0.8675235509872437:  79%|███████▉  | 5123/6470 [58:54<15:17,  1.47it/s]
3|5124|Loss: 0.9805240035057068:  79%|███████▉  | 5123/6470 [58:54<15:17,  1.47it/s]
3|5124|Loss: 0.9805240035057068:  79%|███████▉  | 5124/6470 [58:55<16:06,  1.39it/s]
3|5125|Loss: 0.8377623558044434:  79%|███████▉  | 5124/6470 [58:55<16:06,  1.39it/s]
3|5125|Loss: 0.8377623558044434:  79%|███████▉  | 5125/6470 [58:55<16:55,  1.32it/s]
3|5126|Loss: 1.0163670778274536:  79%|███████▉  | 5125/6470 [58:56<16:55,  1.32it/s]
3|5126|Loss: 1.0163670778274536:  79%|███████▉  | 5126/6470 [58:56<16:16,  1.38it/s]
3|5127|Loss: 0.8333978652954102:  79%|███████▉  | 5126/6470 [58:57<16:16,  1.38it/s]
3|5127|Loss: 0.8333978652954102:  79%|███████▉  | 5127/6470 [58:57<16:57,  1.32it/s]
3|5128|Loss: 0.8947325348854065:  79%|███████▉  | 5127/6470 [58:57<16:57,  1.32it/s]
3|5128|Loss: 0.8947325348854065:  79%|███████▉  | 5128/6470 [58:58<18:04,  1.24it/s]
3|5129|Loss: 0.9668785929679871:  79%|███████▉  | 5128/6470 [58:58<18:04,  1.24it/s]
3|5129|Loss: 0.9668785929679871:  79%|███████▉  | 5129/6470 [58:59<18:38,  1.20it/s]
3|5130|Loss: 0.8627701997756958:  79%|███████▉  | 5129/6470 [58:59<18:38,  1.20it/s]
3|5130|Loss: 0.8627701997756958:  79%|███████▉  | 5130/6470 [59:00<18:34,  1.20it/s]
3|5131|Loss: 0.7901954650878906:  79%|███████▉  | 5130/6470 [59:00<18:34,  1.20it/s]
3|5131|Loss: 0.7901954650878906:  79%|███████▉  | 5131/6470 [59:00<17:01,  1.31it/s]
3|5132|Loss: 0.7122366428375244:  79%|███████▉  | 5131/6470 [59:01<17:01,  1.31it/s]
3|5132|Loss: 0.7122366428375244:  79%|███████▉  | 5132/6470 [59:01<15:26,  1.44it/s]
3|5133|Loss: 0.7030075192451477:  79%|███████▉  | 5132/6470 [59:01<15:26,  1.44it/s]
3|5133|Loss: 0.7030075192451477:  79%|███████▉  | 5133/6470 [59:01<13:23,  1.66it/s]
3|5134|Loss: 0.949454128742218:  79%|███████▉  | 5133/6470 [59:01<13:23,  1.66it/s] 
3|5134|Loss: 0.949454128742218:  79%|███████▉  | 5134/6470 [59:02<14:26,  1.54it/s]
3|5135|Loss: 0.7319225668907166:  79%|███████▉  | 5134/6470 [59:02<14:26,  1.54it/s]
3|5135|Loss: 0.7319225668907166:  79%|███████▉  | 5135/6470 [59:03<14:22,  1.55it/s]
3|5136|Loss: 0.6675893068313599:  79%|███████▉  | 5135/6470 [59:03<14:22,  1.55it/s]
3|5136|Loss: 0.6675893068313599:  79%|███████▉  | 5136/6470 [59:03<15:57,  1.39it/s]
3|5137|Loss: 0.8199326395988464:  79%|███████▉  | 5136/6470 [59:04<15:57,  1.39it/s]
3|5137|Loss: 0.8199326395988464:  79%|███████▉  | 5137/6470 [59:04<15:32,  1.43it/s]
3|5138|Loss: 0.900450587272644:  79%|███████▉  | 5137/6470 [59:04<15:32,  1.43it/s] 
3|5138|Loss: 0.900450587272644:  79%|███████▉  | 5138/6470 [59:05<16:32,  1.34it/s]
3|5139|Loss: 0.8112098574638367:  79%|███████▉  | 5138/6470 [59:05<16:32,  1.34it/s]
3|5139|Loss: 0.8112098574638367:  79%|███████▉  | 5139/6470 [59:05<14:46,  1.50it/s]
3|5140|Loss: 0.7788965702056885:  79%|███████▉  | 5139/6470 [59:06<14:46,  1.50it/s]
3|5140|Loss: 0.7788965702056885:  79%|███████▉  | 5140/6470 [59:06<14:44,  1.50it/s]
3|5141|Loss: 0.8595977425575256:  79%|███████▉  | 5140/6470 [59:06<14:44,  1.50it/s]
3|5141|Loss: 0.8595977425575256:  79%|███████▉  | 5141/6470 [59:07<14:37,  1.51it/s]
3|5142|Loss: 0.7817490100860596:  79%|███████▉  | 5141/6470 [59:07<14:37,  1.51it/s]
3|5142|Loss: 0.7817490100860596:  79%|███████▉  | 5142/6470 [59:07<14:06,  1.57it/s]
3|5143|Loss: 0.8598971962928772:  79%|███████▉  | 5142/6470 [59:08<14:06,  1.57it/s]
3|5143|Loss: 0.8598971962928772:  79%|███████▉  | 5143/6470 [59:08<14:31,  1.52it/s]
3|5144|Loss: 0.8685224056243896:  79%|███████▉  | 5143/6470 [59:08<14:31,  1.52it/s]
3|5144|Loss: 0.8685224056243896:  80%|███████▉  | 5144/6470 [59:09<15:19,  1.44it/s]
3|5145|Loss: 0.6439386606216431:  80%|███████▉  | 5144/6470 [59:09<15:19,  1.44it/s]
3|5145|Loss: 0.6439386606216431:  80%|███████▉  | 5145/6470 [59:09<14:52,  1.48it/s]
3|5146|Loss: 0.6633408069610596:  80%|███████▉  | 5145/6470 [59:10<14:52,  1.48it/s]
3|5146|Loss: 0.6633408069610596:  80%|███████▉  | 5146/6470 [59:10<14:15,  1.55it/s]
3|5147|Loss: 0.8427490592002869:  80%|███████▉  | 5146/6470 [59:10<14:15,  1.55it/s]
3|5147|Loss: 0.8427490592002869:  80%|███████▉  | 5147/6470 [59:11<13:46,  1.60it/s]
3|5148|Loss: 0.7731977105140686:  80%|███████▉  | 5147/6470 [59:11<13:46,  1.60it/s]
3|5148|Loss: 0.7731977105140686:  80%|███████▉  | 5148/6470 [59:11<14:37,  1.51it/s]
3|5149|Loss: 0.8511338233947754:  80%|███████▉  | 5148/6470 [59:12<14:37,  1.51it/s]
3|5149|Loss: 0.8511338233947754:  80%|███████▉  | 5149/6470 [59:12<15:00,  1.47it/s]
3|5150|Loss: 0.8485994338989258:  80%|███████▉  | 5149/6470 [59:12<15:00,  1.47it/s]
3|5150|Loss: 0.8485994338989258:  80%|███████▉  | 5150/6470 [59:13<15:25,  1.43it/s]
3|5151|Loss: 0.8475973606109619:  80%|███████▉  | 5150/6470 [59:13<15:25,  1.43it/s]
3|5151|Loss: 0.8475973606109619:  80%|███████▉  | 5151/6470 [59:13<15:05,  1.46it/s]
3|5152|Loss: 0.7987346649169922:  80%|███████▉  | 5151/6470 [59:14<15:05,  1.46it/s]
3|5152|Loss: 0.7987346649169922:  80%|███████▉  | 5152/6470 [59:14<15:56,  1.38it/s]
3|5153|Loss: 0.8997074365615845:  80%|███████▉  | 5152/6470 [59:15<15:56,  1.38it/s]
3|5153|Loss: 0.8997074365615845:  80%|███████▉  | 5153/6470 [59:15<15:43,  1.40it/s]
3|5154|Loss: 0.7427179217338562:  80%|███████▉  | 5153/6470 [59:15<15:43,  1.40it/s]
3|5154|Loss: 0.7427179217338562:  80%|███████▉  | 5154/6470 [59:16<15:38,  1.40it/s]
3|5155|Loss: 0.8906654715538025:  80%|███████▉  | 5154/6470 [59:16<15:38,  1.40it/s]
3|5155|Loss: 0.8906654715538025:  80%|███████▉  | 5155/6470 [59:16<15:48,  1.39it/s]
3|5156|Loss: 0.8758811950683594:  80%|███████▉  | 5155/6470 [59:17<15:48,  1.39it/s]
3|5156|Loss: 0.8758811950683594:  80%|███████▉  | 5156/6470 [59:17<15:55,  1.38it/s]
3|5157|Loss: 0.7780231833457947:  80%|███████▉  | 5156/6470 [59:17<15:55,  1.38it/s]
3|5157|Loss: 0.7780231833457947:  80%|███████▉  | 5157/6470 [59:18<15:07,  1.45it/s]
3|5158|Loss: 0.8326432704925537:  80%|███████▉  | 5157/6470 [59:18<15:07,  1.45it/s]
3|5158|Loss: 0.8326432704925537:  80%|███████▉  | 5158/6470 [59:19<15:51,  1.38it/s]
3|5159|Loss: 0.9559893012046814:  80%|███████▉  | 5158/6470 [59:19<15:51,  1.38it/s]
3|5159|Loss: 0.9559893012046814:  80%|███████▉  | 5159/6470 [59:19<15:31,  1.41it/s]
3|5160|Loss: 0.9518818259239197:  80%|███████▉  | 5159/6470 [59:20<15:31,  1.41it/s]
3|5160|Loss: 0.9518818259239197:  80%|███████▉  | 5160/6470 [59:20<14:31,  1.50it/s]
3|5161|Loss: 0.9348182678222656:  80%|███████▉  | 5160/6470 [59:20<14:31,  1.50it/s]
3|5161|Loss: 0.9348182678222656:  80%|███████▉  | 5161/6470 [59:20<14:35,  1.50it/s]
3|5162|Loss: 1.0175732374191284:  80%|███████▉  | 5161/6470 [59:21<14:35,  1.50it/s]
3|5162|Loss: 1.0175732374191284:  80%|███████▉  | 5162/6470 [59:21<14:25,  1.51it/s]
3|5163|Loss: 0.8232854008674622:  80%|███████▉  | 5162/6470 [59:21<14:25,  1.51it/s]
3|5163|Loss: 0.8232854008674622:  80%|███████▉  | 5163/6470 [59:22<13:33,  1.61it/s]
3|5164|Loss: 0.7898930907249451:  80%|███████▉  | 5163/6470 [59:22<13:33,  1.61it/s]
3|5164|Loss: 0.7898930907249451:  80%|███████▉  | 5164/6470 [59:22<13:01,  1.67it/s]
3|5165|Loss: 0.7313377261161804:  80%|███████▉  | 5164/6470 [59:23<13:01,  1.67it/s]
3|5165|Loss: 0.7313377261161804:  80%|███████▉  | 5165/6470 [59:23<14:36,  1.49it/s]
3|5166|Loss: 1.0545895099639893:  80%|███████▉  | 5165/6470 [59:23<14:36,  1.49it/s]
3|5166|Loss: 1.0545895099639893:  80%|███████▉  | 5166/6470 [59:24<15:20,  1.42it/s]
3|5167|Loss: 0.8345232009887695:  80%|███████▉  | 5166/6470 [59:24<15:20,  1.42it/s]
3|5167|Loss: 0.8345232009887695:  80%|███████▉  | 5167/6470 [59:24<15:01,  1.45it/s]
3|5168|Loss: 0.7656272053718567:  80%|███████▉  | 5167/6470 [59:25<15:01,  1.45it/s]
3|5168|Loss: 0.7656272053718567:  80%|███████▉  | 5168/6470 [59:25<15:38,  1.39it/s]
3|5169|Loss: 0.9080793857574463:  80%|███████▉  | 5168/6470 [59:26<15:38,  1.39it/s]
3|5169|Loss: 0.9080793857574463:  80%|███████▉  | 5169/6470 [59:26<15:24,  1.41it/s]
3|5170|Loss: 0.8338572382926941:  80%|███████▉  | 5169/6470 [59:26<15:24,  1.41it/s]
3|5170|Loss: 0.8338572382926941:  80%|███████▉  | 5170/6470 [59:27<15:16,  1.42it/s]
3|5171|Loss: 0.6979702711105347:  80%|███████▉  | 5170/6470 [59:27<15:16,  1.42it/s]
3|5171|Loss: 0.6979702711105347:  80%|███████▉  | 5171/6470 [59:27<15:15,  1.42it/s]
3|5172|Loss: 0.9706342816352844:  80%|███████▉  | 5171/6470 [59:28<15:15,  1.42it/s]
3|5172|Loss: 0.9706342816352844:  80%|███████▉  | 5172/6470 [59:28<14:58,  1.44it/s]
3|5173|Loss: 0.882703423500061:  80%|███████▉  | 5172/6470 [59:28<14:58,  1.44it/s] 
3|5173|Loss: 0.882703423500061:  80%|███████▉  | 5173/6470 [59:29<14:39,  1.47it/s]
3|5174|Loss: 0.9528486132621765:  80%|███████▉  | 5173/6470 [59:29<14:39,  1.47it/s]
3|5174|Loss: 0.9528486132621765:  80%|███████▉  | 5174/6470 [59:29<14:05,  1.53it/s]
3|5175|Loss: 0.8808563947677612:  80%|███████▉  | 5174/6470 [59:30<14:05,  1.53it/s]
3|5175|Loss: 0.8808563947677612:  80%|███████▉  | 5175/6470 [59:30<14:44,  1.46it/s]
3|5176|Loss: 0.9130982756614685:  80%|███████▉  | 5175/6470 [59:30<14:44,  1.46it/s]
3|5176|Loss: 0.9130982756614685:  80%|████████  | 5176/6470 [59:31<13:51,  1.56it/s]
3|5177|Loss: 0.9366267919540405:  80%|████████  | 5176/6470 [59:31<13:51,  1.56it/s]
3|5177|Loss: 0.9366267919540405:  80%|████████  | 5177/6470 [59:31<13:36,  1.58it/s]
3|5178|Loss: 0.8493982553482056:  80%|████████  | 5177/6470 [59:32<13:36,  1.58it/s]
3|5178|Loss: 0.8493982553482056:  80%|████████  | 5178/6470 [59:32<13:31,  1.59it/s]
3|5179|Loss: 0.8161665201187134:  80%|████████  | 5178/6470 [59:32<13:31,  1.59it/s]
3|5179|Loss: 0.8161665201187134:  80%|████████  | 5179/6470 [59:32<12:57,  1.66it/s]
3|5180|Loss: 0.8416404724121094:  80%|████████  | 5179/6470 [59:33<12:57,  1.66it/s]
3|5180|Loss: 0.8416404724121094:  80%|████████  | 5180/6470 [59:33<14:08,  1.52it/s]
3|5181|Loss: 0.8717251420021057:  80%|████████  | 5180/6470 [59:34<14:08,  1.52it/s]
3|5181|Loss: 0.8717251420021057:  80%|████████  | 5181/6470 [59:34<14:21,  1.50it/s]
3|5182|Loss: 0.8281913995742798:  80%|████████  | 5181/6470 [59:34<14:21,  1.50it/s]
3|5182|Loss: 0.8281913995742798:  80%|████████  | 5182/6470 [59:35<14:56,  1.44it/s]
3|5183|Loss: 0.800682783126831:  80%|████████  | 5182/6470 [59:35<14:56,  1.44it/s] 
3|5183|Loss: 0.800682783126831:  80%|████████  | 5183/6470 [59:35<14:28,  1.48it/s]
3|5184|Loss: 0.900907039642334:  80%|████████  | 5183/6470 [59:36<14:28,  1.48it/s]
3|5184|Loss: 0.900907039642334:  80%|████████  | 5184/6470 [59:36<13:53,  1.54it/s]
3|5185|Loss: 0.7316337823867798:  80%|████████  | 5184/6470 [59:36<13:53,  1.54it/s]
3|5185|Loss: 0.7316337823867798:  80%|████████  | 5185/6470 [59:36<13:37,  1.57it/s]
3|5186|Loss: 0.8167920112609863:  80%|████████  | 5185/6470 [59:37<13:37,  1.57it/s]
3|5186|Loss: 0.8167920112609863:  80%|████████  | 5186/6470 [59:37<14:33,  1.47it/s]
3|5187|Loss: 0.8692594170570374:  80%|████████  | 5186/6470 [59:38<14:33,  1.47it/s]
3|5187|Loss: 0.8692594170570374:  80%|████████  | 5187/6470 [59:38<15:03,  1.42it/s]
3|5188|Loss: 0.8434693813323975:  80%|████████  | 5187/6470 [59:38<15:03,  1.42it/s]
3|5188|Loss: 0.8434693813323975:  80%|████████  | 5188/6470 [59:39<15:43,  1.36it/s]
3|5189|Loss: 0.8666332364082336:  80%|████████  | 5188/6470 [59:39<15:43,  1.36it/s]
3|5189|Loss: 0.8666332364082336:  80%|████████  | 5189/6470 [59:39<15:53,  1.34it/s]
3|5190|Loss: 0.7732710242271423:  80%|████████  | 5189/6470 [59:40<15:53,  1.34it/s]
3|5190|Loss: 0.7732710242271423:  80%|████████  | 5190/6470 [59:40<14:43,  1.45it/s]
3|5191|Loss: 0.9685274958610535:  80%|████████  | 5190/6470 [59:40<14:43,  1.45it/s]
3|5191|Loss: 0.9685274958610535:  80%|████████  | 5191/6470 [59:41<14:13,  1.50it/s]
3|5192|Loss: 0.9042936563491821:  80%|████████  | 5191/6470 [59:41<14:13,  1.50it/s]
3|5192|Loss: 0.9042936563491821:  80%|████████  | 5192/6470 [59:41<13:44,  1.55it/s]
3|5193|Loss: 0.9319804310798645:  80%|████████  | 5192/6470 [59:42<13:44,  1.55it/s]
3|5193|Loss: 0.9319804310798645:  80%|████████  | 5193/6470 [59:42<13:31,  1.57it/s]
3|5194|Loss: 0.9554021954536438:  80%|████████  | 5193/6470 [59:42<13:31,  1.57it/s]
3|5194|Loss: 0.9554021954536438:  80%|████████  | 5194/6470 [59:43<13:30,  1.57it/s]
3|5195|Loss: 0.9233604669570923:  80%|████████  | 5194/6470 [59:43<13:30,  1.57it/s]
3|5195|Loss: 0.9233604669570923:  80%|████████  | 5195/6470 [59:43<13:32,  1.57it/s]
3|5196|Loss: 0.6304003596305847:  80%|████████  | 5195/6470 [59:44<13:32,  1.57it/s]
3|5196|Loss: 0.6304003596305847:  80%|████████  | 5196/6470 [59:44<14:08,  1.50it/s]
3|5197|Loss: 0.8724638819694519:  80%|████████  | 5196/6470 [59:44<14:08,  1.50it/s]
3|5197|Loss: 0.8724638819694519:  80%|████████  | 5197/6470 [59:45<14:10,  1.50it/s]
3|5198|Loss: 1.0112485885620117:  80%|████████  | 5197/6470 [59:45<14:10,  1.50it/s]
3|5198|Loss: 1.0112485885620117:  80%|████████  | 5198/6470 [59:45<15:01,  1.41it/s]
3|5199|Loss: 0.9972830414772034:  80%|████████  | 5198/6470 [59:46<15:01,  1.41it/s]
3|5199|Loss: 0.9972830414772034:  80%|████████  | 5199/6470 [59:46<15:54,  1.33it/s]
3|5200|Loss: 0.8683314919471741:  80%|████████  | 5199/6470 [59:47<15:54,  1.33it/s]
3|5200|Loss: 0.8683314919471741:  80%|████████  | 5200/6470 [59:47<15:03,  1.41it/s]
3|5201|Loss: 0.7998859882354736:  80%|████████  | 5200/6470 [59:47<15:03,  1.41it/s]
3|5201|Loss: 0.7998859882354736:  80%|████████  | 5201/6470 [59:47<14:08,  1.49it/s]
3|5202|Loss: 0.9067902565002441:  80%|████████  | 5201/6470 [59:48<14:08,  1.49it/s]
3|5202|Loss: 0.9067902565002441:  80%|████████  | 5202/6470 [59:48<13:19,  1.59it/s]
3|5203|Loss: 0.7220520377159119:  80%|████████  | 5202/6470 [59:48<13:19,  1.59it/s]
3|5203|Loss: 0.7220520377159119:  80%|████████  | 5203/6470 [59:49<13:43,  1.54it/s]
3|5204|Loss: 0.7876049876213074:  80%|████████  | 5203/6470 [59:49<13:43,  1.54it/s]
3|5204|Loss: 0.7876049876213074:  80%|████████  | 5204/6470 [59:49<13:09,  1.60it/s]
3|5205|Loss: 0.9723494648933411:  80%|████████  | 5204/6470 [59:50<13:09,  1.60it/s]
3|5205|Loss: 0.9723494648933411:  80%|████████  | 5205/6470 [59:50<13:01,  1.62it/s]
3|5206|Loss: 0.7954037189483643:  80%|████████  | 5205/6470 [59:50<13:01,  1.62it/s]
3|5206|Loss: 0.7954037189483643:  80%|████████  | 5206/6470 [59:51<13:52,  1.52it/s]
3|5207|Loss: 0.8579304218292236:  80%|████████  | 5206/6470 [59:51<13:52,  1.52it/s]
3|5207|Loss: 0.8579304218292236:  80%|████████  | 5207/6470 [59:51<14:44,  1.43it/s]
3|5208|Loss: 0.8726305961608887:  80%|████████  | 5207/6470 [59:52<14:44,  1.43it/s]
3|5208|Loss: 0.8726305961608887:  80%|████████  | 5208/6470 [59:52<15:01,  1.40it/s]
3|5209|Loss: 0.8315904140472412:  80%|████████  | 5208/6470 [59:52<15:01,  1.40it/s]
3|5209|Loss: 0.8315904140472412:  81%|████████  | 5209/6470 [59:53<13:57,  1.51it/s]
3|5210|Loss: 0.6896117925643921:  81%|████████  | 5209/6470 [59:53<13:57,  1.51it/s]
3|5210|Loss: 0.6896117925643921:  81%|████████  | 5210/6470 [59:53<14:06,  1.49it/s]
3|5211|Loss: 0.874699056148529:  81%|████████  | 5210/6470 [59:54<14:06,  1.49it/s] 
3|5211|Loss: 0.874699056148529:  81%|████████  | 5211/6470 [59:54<14:20,  1.46it/s]
3|5212|Loss: 0.7197679281234741:  81%|████████  | 5211/6470 [59:54<14:20,  1.46it/s]
3|5212|Loss: 0.7197679281234741:  81%|████████  | 5212/6470 [59:55<14:37,  1.43it/s]
3|5213|Loss: 0.913531482219696:  81%|████████  | 5212/6470 [59:55<14:37,  1.43it/s] 
3|5213|Loss: 0.913531482219696:  81%|████████  | 5213/6470 [59:55<13:43,  1.53it/s]
3|5214|Loss: 0.7409867644309998:  81%|████████  | 5213/6470 [59:56<13:43,  1.53it/s]
3|5214|Loss: 0.7409867644309998:  81%|████████  | 5214/6470 [59:56<14:30,  1.44it/s]
3|5215|Loss: 0.8821088075637817:  81%|████████  | 5214/6470 [59:56<14:30,  1.44it/s]
3|5215|Loss: 0.8821088075637817:  81%|████████  | 5215/6470 [59:57<13:50,  1.51it/s]
3|5216|Loss: 0.9242691397666931:  81%|████████  | 5215/6470 [59:57<13:50,  1.51it/s]
3|5216|Loss: 0.9242691397666931:  81%|████████  | 5216/6470 [59:57<13:49,  1.51it/s]
3|5217|Loss: 0.8697108030319214:  81%|████████  | 5216/6470 [59:58<13:49,  1.51it/s]
3|5217|Loss: 0.8697108030319214:  81%|████████  | 5217/6470 [59:58<14:34,  1.43it/s]
3|5218|Loss: 0.8934791088104248:  81%|████████  | 5217/6470 [59:59<14:34,  1.43it/s]
3|5218|Loss: 0.8934791088104248:  81%|████████  | 5218/6470 [59:59<15:35,  1.34it/s]
3|5219|Loss: 0.8723310232162476:  81%|████████  | 5218/6470 [59:59<15:35,  1.34it/s]
3|5219|Loss: 0.8723310232162476:  81%|████████  | 5219/6470 [1:00:00<15:16,  1.37it/s]
3|5220|Loss: 0.7387566566467285:  81%|████████  | 5219/6470 [1:00:00<15:16,  1.37it/s]
3|5220|Loss: 0.7387566566467285:  81%|████████  | 5220/6470 [1:00:00<15:13,  1.37it/s]
3|5221|Loss: 0.9692279100418091:  81%|████████  | 5220/6470 [1:00:01<15:13,  1.37it/s]
3|5221|Loss: 0.9692279100418091:  81%|████████  | 5221/6470 [1:00:01<15:37,  1.33it/s]
3|5222|Loss: 0.8184153437614441:  81%|████████  | 5221/6470 [1:00:02<15:37,  1.33it/s]
3|5222|Loss: 0.8184153437614441:  81%|████████  | 5222/6470 [1:00:02<15:03,  1.38it/s]
3|5223|Loss: 0.9961475729942322:  81%|████████  | 5222/6470 [1:00:02<15:03,  1.38it/s]
3|5223|Loss: 0.9961475729942322:  81%|████████  | 5223/6470 [1:00:02<13:55,  1.49it/s]
3|5224|Loss: 0.966874897480011:  81%|████████  | 5223/6470 [1:00:03<13:55,  1.49it/s] 
3|5224|Loss: 0.966874897480011:  81%|████████  | 5224/6470 [1:00:03<14:33,  1.43it/s]
3|5225|Loss: 0.7991691827774048:  81%|████████  | 5224/6470 [1:00:04<14:33,  1.43it/s]
3|5225|Loss: 0.7991691827774048:  81%|████████  | 5225/6470 [1:00:04<13:39,  1.52it/s]
3|5226|Loss: 0.804564893245697:  81%|████████  | 5225/6470 [1:00:04<13:39,  1.52it/s] 
3|5226|Loss: 0.804564893245697:  81%|████████  | 5226/6470 [1:00:04<14:07,  1.47it/s]
3|5227|Loss: 0.8254472017288208:  81%|████████  | 5226/6470 [1:00:05<14:07,  1.47it/s]
3|5227|Loss: 0.8254472017288208:  81%|████████  | 5227/6470 [1:00:05<14:22,  1.44it/s]
3|5228|Loss: 0.8931745886802673:  81%|████████  | 5227/6470 [1:00:06<14:22,  1.44it/s]
3|5228|Loss: 0.8931745886802673:  81%|████████  | 5228/6470 [1:00:06<14:01,  1.48it/s]
3|5229|Loss: 0.931144654750824:  81%|████████  | 5228/6470 [1:00:06<14:01,  1.48it/s] 
3|5229|Loss: 0.931144654750824:  81%|████████  | 5229/6470 [1:00:07<14:32,  1.42it/s]
3|5230|Loss: 0.9112350940704346:  81%|████████  | 5229/6470 [1:00:07<14:32,  1.42it/s]
3|5230|Loss: 0.9112350940704346:  81%|████████  | 5230/6470 [1:00:07<14:39,  1.41it/s]
3|5231|Loss: 0.7461947798728943:  81%|████████  | 5230/6470 [1:00:08<14:39,  1.41it/s]
3|5231|Loss: 0.7461947798728943:  81%|████████  | 5231/6470 [1:00:08<14:46,  1.40it/s]
3|5232|Loss: 0.6874567866325378:  81%|████████  | 5231/6470 [1:00:08<14:46,  1.40it/s]
3|5232|Loss: 0.6874567866325378:  81%|████████  | 5232/6470 [1:00:09<14:06,  1.46it/s]
3|5233|Loss: 0.707848310470581:  81%|████████  | 5232/6470 [1:00:09<14:06,  1.46it/s] 
3|5233|Loss: 0.707848310470581:  81%|████████  | 5233/6470 [1:00:09<13:24,  1.54it/s]
3|5234|Loss: 0.8215243816375732:  81%|████████  | 5233/6470 [1:00:10<13:24,  1.54it/s]
3|5234|Loss: 0.8215243816375732:  81%|████████  | 5234/6470 [1:00:10<14:15,  1.45it/s]
3|5235|Loss: 0.6690436601638794:  81%|████████  | 5234/6470 [1:00:10<14:15,  1.45it/s]
3|5235|Loss: 0.6690436601638794:  81%|████████  | 5235/6470 [1:00:11<13:50,  1.49it/s]
3|5236|Loss: 0.8647758364677429:  81%|████████  | 5235/6470 [1:00:11<13:50,  1.49it/s]
3|5236|Loss: 0.8647758364677429:  81%|████████  | 5236/6470 [1:00:11<13:53,  1.48it/s]
3|5237|Loss: 0.9623829126358032:  81%|████████  | 5236/6470 [1:00:12<13:53,  1.48it/s]
3|5237|Loss: 0.9623829126358032:  81%|████████  | 5237/6470 [1:00:12<14:33,  1.41it/s]
3|5238|Loss: 0.6769880652427673:  81%|████████  | 5237/6470 [1:00:13<14:33,  1.41it/s]
3|5238|Loss: 0.6769880652427673:  81%|████████  | 5238/6470 [1:00:13<15:41,  1.31it/s]
3|5239|Loss: 0.8209384679794312:  81%|████████  | 5238/6470 [1:00:14<15:41,  1.31it/s]
3|5239|Loss: 0.8209384679794312:  81%|████████  | 5239/6470 [1:00:14<16:20,  1.26it/s]
3|5240|Loss: 0.8841723203659058:  81%|████████  | 5239/6470 [1:00:14<16:20,  1.26it/s]
3|5240|Loss: 0.8841723203659058:  81%|████████  | 5240/6470 [1:00:14<15:02,  1.36it/s]
3|5241|Loss: 0.8718283772468567:  81%|████████  | 5240/6470 [1:00:15<15:02,  1.36it/s]
3|5241|Loss: 0.8718283772468567:  81%|████████  | 5241/6470 [1:00:15<13:57,  1.47it/s]
3|5242|Loss: 0.8307850360870361:  81%|████████  | 5241/6470 [1:00:15<13:57,  1.47it/s]
3|5242|Loss: 0.8307850360870361:  81%|████████  | 5242/6470 [1:00:16<13:57,  1.47it/s]
3|5243|Loss: 0.8316788077354431:  81%|████████  | 5242/6470 [1:00:16<13:57,  1.47it/s]
3|5243|Loss: 0.8316788077354431:  81%|████████  | 5243/6470 [1:00:16<14:12,  1.44it/s]
3|5244|Loss: 0.9862226843833923:  81%|████████  | 5243/6470 [1:00:17<14:12,  1.44it/s]
3|5244|Loss: 0.9862226843833923:  81%|████████  | 5244/6470 [1:00:17<15:01,  1.36it/s]
3|5245|Loss: 0.8130837082862854:  81%|████████  | 5244/6470 [1:00:18<15:01,  1.36it/s]
3|5245|Loss: 0.8130837082862854:  81%|████████  | 5245/6470 [1:00:18<14:24,  1.42it/s]
3|5246|Loss: 0.9199342727661133:  81%|████████  | 5245/6470 [1:00:18<14:24,  1.42it/s]
3|5246|Loss: 0.9199342727661133:  81%|████████  | 5246/6470 [1:00:19<14:13,  1.43it/s]
3|5247|Loss: 0.8942586779594421:  81%|████████  | 5246/6470 [1:00:19<14:13,  1.43it/s]
3|5247|Loss: 0.8942586779594421:  81%|████████  | 5247/6470 [1:00:19<13:18,  1.53it/s]
3|5248|Loss: 0.9541565179824829:  81%|████████  | 5247/6470 [1:00:19<13:18,  1.53it/s]
3|5248|Loss: 0.9541565179824829:  81%|████████  | 5248/6470 [1:00:20<13:03,  1.56it/s]
3|5249|Loss: 0.7877764701843262:  81%|████████  | 5248/6470 [1:00:20<13:03,  1.56it/s]
3|5249|Loss: 0.7877764701843262:  81%|████████  | 5249/6470 [1:00:21<14:08,  1.44it/s]
3|5250|Loss: 0.9780922532081604:  81%|████████  | 5249/6470 [1:00:21<14:08,  1.44it/s]
3|5250|Loss: 0.9780922532081604:  81%|████████  | 5250/6470 [1:00:21<13:12,  1.54it/s]
3|5251|Loss: 0.8249624967575073:  81%|████████  | 5250/6470 [1:00:22<13:12,  1.54it/s]
3|5251|Loss: 0.8249624967575073:  81%|████████  | 5251/6470 [1:00:22<14:27,  1.41it/s]
3|5252|Loss: 0.8000364303588867:  81%|████████  | 5251/6470 [1:00:22<14:27,  1.41it/s]
3|5252|Loss: 0.8000364303588867:  81%|████████  | 5252/6470 [1:00:23<14:19,  1.42it/s]
3|5253|Loss: 0.7373054027557373:  81%|████████  | 5252/6470 [1:00:23<14:19,  1.42it/s]
3|5253|Loss: 0.7373054027557373:  81%|████████  | 5253/6470 [1:00:23<14:11,  1.43it/s]
3|5254|Loss: 0.7803927063941956:  81%|████████  | 5253/6470 [1:00:24<14:11,  1.43it/s]
3|5254|Loss: 0.7803927063941956:  81%|████████  | 5254/6470 [1:00:24<13:44,  1.48it/s]
3|5255|Loss: 0.7263193726539612:  81%|████████  | 5254/6470 [1:00:24<13:44,  1.48it/s]
3|5255|Loss: 0.7263193726539612:  81%|████████  | 5255/6470 [1:00:24<12:12,  1.66it/s]
3|5256|Loss: 1.0001548528671265:  81%|████████  | 5255/6470 [1:00:25<12:12,  1.66it/s]
3|5256|Loss: 1.0001548528671265:  81%|████████  | 5256/6470 [1:00:25<12:29,  1.62it/s]
3|5257|Loss: 0.7338492274284363:  81%|████████  | 5256/6470 [1:00:25<12:29,  1.62it/s]
3|5257|Loss: 0.7338492274284363:  81%|████████▏ | 5257/6470 [1:00:26<13:14,  1.53it/s]
3|5258|Loss: 0.7426202893257141:  81%|████████▏ | 5257/6470 [1:00:26<13:14,  1.53it/s]
3|5258|Loss: 0.7426202893257141:  81%|████████▏ | 5258/6470 [1:00:26<12:49,  1.57it/s]
3|5259|Loss: 0.9486883878707886:  81%|████████▏ | 5258/6470 [1:00:27<12:49,  1.57it/s]
3|5259|Loss: 0.9486883878707886:  81%|████████▏ | 5259/6470 [1:00:27<12:28,  1.62it/s]
3|5260|Loss: 0.9086523652076721:  81%|████████▏ | 5259/6470 [1:00:27<12:28,  1.62it/s]
3|5260|Loss: 0.9086523652076721:  81%|████████▏ | 5260/6470 [1:00:28<12:57,  1.56it/s]
3|5261|Loss: 0.9544748067855835:  81%|████████▏ | 5260/6470 [1:00:28<12:57,  1.56it/s]
3|5261|Loss: 0.9544748067855835:  81%|████████▏ | 5261/6470 [1:00:28<12:43,  1.58it/s]
3|5262|Loss: 0.8165000677108765:  81%|████████▏ | 5261/6470 [1:00:29<12:43,  1.58it/s]
3|5262|Loss: 0.8165000677108765:  81%|████████▏ | 5262/6470 [1:00:29<12:58,  1.55it/s]
3|5263|Loss: 0.7570552825927734:  81%|████████▏ | 5262/6470 [1:00:29<12:58,  1.55it/s]
3|5263|Loss: 0.7570552825927734:  81%|████████▏ | 5263/6470 [1:00:30<13:27,  1.50it/s]
3|5264|Loss: 0.7634015083312988:  81%|████████▏ | 5263/6470 [1:00:30<13:27,  1.50it/s]
3|5264|Loss: 0.7634015083312988:  81%|████████▏ | 5264/6470 [1:00:30<13:15,  1.52it/s]
3|5265|Loss: 0.8911482691764832:  81%|████████▏ | 5264/6470 [1:00:31<13:15,  1.52it/s]
3|5265|Loss: 0.8911482691764832:  81%|████████▏ | 5265/6470 [1:00:31<12:11,  1.65it/s]
3|5266|Loss: 0.845363438129425:  81%|████████▏ | 5265/6470 [1:00:31<12:11,  1.65it/s] 
3|5266|Loss: 0.845363438129425:  81%|████████▏ | 5266/6470 [1:00:31<12:39,  1.59it/s]
3|5267|Loss: 0.9848330616950989:  81%|████████▏ | 5266/6470 [1:00:32<12:39,  1.59it/s]
3|5267|Loss: 0.9848330616950989:  81%|████████▏ | 5267/6470 [1:00:32<12:19,  1.63it/s]
3|5268|Loss: 0.9930077791213989:  81%|████████▏ | 5267/6470 [1:00:32<12:19,  1.63it/s]
3|5268|Loss: 0.9930077791213989:  81%|████████▏ | 5268/6470 [1:00:33<11:59,  1.67it/s]
3|5269|Loss: 0.9012842178344727:  81%|████████▏ | 5268/6470 [1:00:33<11:59,  1.67it/s]
3|5269|Loss: 0.9012842178344727:  81%|████████▏ | 5269/6470 [1:00:33<13:05,  1.53it/s]
3|5270|Loss: 0.9769768118858337:  81%|████████▏ | 5269/6470 [1:00:34<13:05,  1.53it/s]
3|5270|Loss: 0.9769768118858337:  81%|████████▏ | 5270/6470 [1:00:34<13:33,  1.48it/s]
3|5271|Loss: 0.90716153383255:  81%|████████▏ | 5270/6470 [1:00:35<13:33,  1.48it/s]  
3|5271|Loss: 0.90716153383255:  81%|████████▏ | 5271/6470 [1:00:35<13:21,  1.50it/s]
3|5272|Loss: 0.808465838432312:  81%|████████▏ | 5271/6470 [1:00:35<13:21,  1.50it/s]
3|5272|Loss: 0.808465838432312:  81%|████████▏ | 5272/6470 [1:00:35<13:03,  1.53it/s]
3|5273|Loss: 0.9573215842247009:  81%|████████▏ | 5272/6470 [1:00:36<13:03,  1.53it/s]
3|5273|Loss: 0.9573215842247009:  81%|████████▏ | 5273/6470 [1:00:36<13:50,  1.44it/s]
3|5274|Loss: 0.8753626942634583:  81%|████████▏ | 5273/6470 [1:00:37<13:50,  1.44it/s]
3|5274|Loss: 0.8753626942634583:  82%|████████▏ | 5274/6470 [1:00:37<14:36,  1.36it/s]
3|5275|Loss: 0.92835533618927:  82%|████████▏ | 5274/6470 [1:00:37<14:36,  1.36it/s]  
3|5275|Loss: 0.92835533618927:  82%|████████▏ | 5275/6470 [1:00:38<14:49,  1.34it/s]
3|5276|Loss: 0.7974816560745239:  82%|████████▏ | 5275/6470 [1:00:38<14:49,  1.34it/s]
3|5276|Loss: 0.7974816560745239:  82%|████████▏ | 5276/6470 [1:00:39<15:23,  1.29it/s]
3|5277|Loss: 0.8270758986473083:  82%|████████▏ | 5276/6470 [1:00:39<15:23,  1.29it/s]
3|5277|Loss: 0.8270758986473083:  82%|████████▏ | 5277/6470 [1:00:39<15:16,  1.30it/s]
3|5278|Loss: 0.9638221263885498:  82%|████████▏ | 5277/6470 [1:00:40<15:16,  1.30it/s]
3|5278|Loss: 0.9638221263885498:  82%|████████▏ | 5278/6470 [1:00:40<14:09,  1.40it/s]
3|5279|Loss: 0.8059387803077698:  82%|████████▏ | 5278/6470 [1:00:40<14:09,  1.40it/s]
3|5279|Loss: 0.8059387803077698:  82%|████████▏ | 5279/6470 [1:00:40<12:20,  1.61it/s]
3|5280|Loss: 0.8301241993904114:  82%|████████▏ | 5279/6470 [1:00:41<12:20,  1.61it/s]
3|5280|Loss: 0.8301241993904114:  82%|████████▏ | 5280/6470 [1:00:41<12:06,  1.64it/s]
3|5281|Loss: 0.8975169062614441:  82%|████████▏ | 5280/6470 [1:00:41<12:06,  1.64it/s]
3|5281|Loss: 0.8975169062614441:  82%|████████▏ | 5281/6470 [1:00:42<12:49,  1.54it/s]
3|5282|Loss: 0.7864025831222534:  82%|████████▏ | 5281/6470 [1:00:42<12:49,  1.54it/s]
3|5282|Loss: 0.7864025831222534:  82%|████████▏ | 5282/6470 [1:00:42<13:15,  1.49it/s]
3|5283|Loss: 0.8904117941856384:  82%|████████▏ | 5282/6470 [1:00:43<13:15,  1.49it/s]
3|5283|Loss: 0.8904117941856384:  82%|████████▏ | 5283/6470 [1:00:43<13:40,  1.45it/s]
3|5284|Loss: 0.7147952318191528:  82%|████████▏ | 5283/6470 [1:00:44<13:40,  1.45it/s]
3|5284|Loss: 0.7147952318191528:  82%|████████▏ | 5284/6470 [1:00:44<13:58,  1.41it/s]
3|5285|Loss: 0.7284932136535645:  82%|████████▏ | 5284/6470 [1:00:44<13:58,  1.41it/s]
3|5285|Loss: 0.7284932136535645:  82%|████████▏ | 5285/6470 [1:00:45<13:34,  1.45it/s]
3|5286|Loss: 0.9374997019767761:  82%|████████▏ | 5285/6470 [1:00:45<13:34,  1.45it/s]
3|5286|Loss: 0.9374997019767761:  82%|████████▏ | 5286/6470 [1:00:45<13:55,  1.42it/s]
3|5287|Loss: 0.834403932094574:  82%|████████▏ | 5286/6470 [1:00:46<13:55,  1.42it/s] 
3|5287|Loss: 0.834403932094574:  82%|████████▏ | 5287/6470 [1:00:46<14:47,  1.33it/s]
3|5288|Loss: 0.9277181625366211:  82%|████████▏ | 5287/6470 [1:00:47<14:47,  1.33it/s]
3|5288|Loss: 0.9277181625366211:  82%|████████▏ | 5288/6470 [1:00:47<14:53,  1.32it/s]
3|5289|Loss: 0.733593225479126:  82%|████████▏ | 5288/6470 [1:00:47<14:53,  1.32it/s] 
3|5289|Loss: 0.733593225479126:  82%|████████▏ | 5289/6470 [1:00:48<14:44,  1.34it/s]
3|5290|Loss: 0.9872775673866272:  82%|████████▏ | 5289/6470 [1:00:48<14:44,  1.34it/s]
3|5290|Loss: 0.9872775673866272:  82%|████████▏ | 5290/6470 [1:00:48<13:51,  1.42it/s]
3|5291|Loss: 0.8628817796707153:  82%|████████▏ | 5290/6470 [1:00:49<13:51,  1.42it/s]
3|5291|Loss: 0.8628817796707153:  82%|████████▏ | 5291/6470 [1:00:49<12:51,  1.53it/s]
3|5292|Loss: 0.7767531871795654:  82%|████████▏ | 5291/6470 [1:00:49<12:51,  1.53it/s]
3|5292|Loss: 0.7767531871795654:  82%|████████▏ | 5292/6470 [1:00:49<12:37,  1.55it/s]
3|5293|Loss: 0.8649815917015076:  82%|████████▏ | 5292/6470 [1:00:50<12:37,  1.55it/s]
3|5293|Loss: 0.8649815917015076:  82%|████████▏ | 5293/6470 [1:00:50<12:29,  1.57it/s]
3|5294|Loss: 0.7879622578620911:  82%|████████▏ | 5293/6470 [1:00:50<12:29,  1.57it/s]
3|5294|Loss: 0.7879622578620911:  82%|████████▏ | 5294/6470 [1:00:51<12:27,  1.57it/s]
3|5295|Loss: 0.8775844573974609:  82%|████████▏ | 5294/6470 [1:00:51<12:27,  1.57it/s]
3|5295|Loss: 0.8775844573974609:  82%|████████▏ | 5295/6470 [1:00:51<12:22,  1.58it/s]
3|5296|Loss: 0.6916917562484741:  82%|████████▏ | 5295/6470 [1:00:52<12:22,  1.58it/s]
3|5296|Loss: 0.6916917562484741:  82%|████████▏ | 5296/6470 [1:00:52<13:21,  1.46it/s]
3|5297|Loss: 1.0523428916931152:  82%|████████▏ | 5296/6470 [1:00:52<13:21,  1.46it/s]
3|5297|Loss: 1.0523428916931152:  82%|████████▏ | 5297/6470 [1:00:53<12:46,  1.53it/s]
3|5298|Loss: 0.7772840857505798:  82%|████████▏ | 5297/6470 [1:00:53<12:46,  1.53it/s]
3|5298|Loss: 0.7772840857505798:  82%|████████▏ | 5298/6470 [1:00:53<12:47,  1.53it/s]
3|5299|Loss: 0.8372122049331665:  82%|████████▏ | 5298/6470 [1:00:54<12:47,  1.53it/s]
3|5299|Loss: 0.8372122049331665:  82%|████████▏ | 5299/6470 [1:00:54<12:46,  1.53it/s]
3|5300|Loss: 0.8127564191818237:  82%|████████▏ | 5299/6470 [1:00:54<12:46,  1.53it/s]
3|5300|Loss: 0.8127564191818237:  82%|████████▏ | 5300/6470 [1:00:55<13:12,  1.48it/s]
3|5301|Loss: 0.8772095441818237:  82%|████████▏ | 5300/6470 [1:00:55<13:12,  1.48it/s]
3|5301|Loss: 0.8772095441818237:  82%|████████▏ | 5301/6470 [1:00:55<13:02,  1.49it/s]
3|5302|Loss: 0.9470744729042053:  82%|████████▏ | 5301/6470 [1:00:56<13:02,  1.49it/s]
3|5302|Loss: 0.9470744729042053:  82%|████████▏ | 5302/6470 [1:00:56<13:21,  1.46it/s]
3|5303|Loss: 0.8345796465873718:  82%|████████▏ | 5302/6470 [1:00:56<13:21,  1.46it/s]
3|5303|Loss: 0.8345796465873718:  82%|████████▏ | 5303/6470 [1:00:57<13:14,  1.47it/s]
3|5304|Loss: 0.956853985786438:  82%|████████▏ | 5303/6470 [1:00:57<13:14,  1.47it/s] 
3|5304|Loss: 0.956853985786438:  82%|████████▏ | 5304/6470 [1:00:57<12:41,  1.53it/s]
3|5305|Loss: 0.9181157350540161:  82%|████████▏ | 5304/6470 [1:00:58<12:41,  1.53it/s]
3|5305|Loss: 0.9181157350540161:  82%|████████▏ | 5305/6470 [1:00:58<12:33,  1.55it/s]
3|5306|Loss: 0.7571412324905396:  82%|████████▏ | 5305/6470 [1:00:58<12:33,  1.55it/s]
3|5306|Loss: 0.7571412324905396:  82%|████████▏ | 5306/6470 [1:00:59<13:16,  1.46it/s]
3|5307|Loss: 0.881405234336853:  82%|████████▏ | 5306/6470 [1:00:59<13:16,  1.46it/s] 
3|5307|Loss: 0.881405234336853:  82%|████████▏ | 5307/6470 [1:00:59<12:55,  1.50it/s]
3|5308|Loss: 0.8769871592521667:  82%|████████▏ | 5307/6470 [1:01:00<12:55,  1.50it/s]
3|5308|Loss: 0.8769871592521667:  82%|████████▏ | 5308/6470 [1:01:00<13:34,  1.43it/s]
3|5309|Loss: 0.7988560199737549:  82%|████████▏ | 5308/6470 [1:01:01<13:34,  1.43it/s]
3|5309|Loss: 0.7988560199737549:  82%|████████▏ | 5309/6470 [1:01:01<13:19,  1.45it/s]
3|5310|Loss: 0.9278180599212646:  82%|████████▏ | 5309/6470 [1:01:01<13:19,  1.45it/s]
3|5310|Loss: 0.9278180599212646:  82%|████████▏ | 5310/6470 [1:01:02<13:15,  1.46it/s]
3|5311|Loss: 0.9846217036247253:  82%|████████▏ | 5310/6470 [1:01:02<13:15,  1.46it/s]
3|5311|Loss: 0.9846217036247253:  82%|████████▏ | 5311/6470 [1:01:02<14:02,  1.38it/s]
3|5312|Loss: 0.8295760750770569:  82%|████████▏ | 5311/6470 [1:01:03<14:02,  1.38it/s]
3|5312|Loss: 0.8295760750770569:  82%|████████▏ | 5312/6470 [1:01:03<14:46,  1.31it/s]
3|5313|Loss: 0.7362993955612183:  82%|████████▏ | 5312/6470 [1:01:04<14:46,  1.31it/s]
3|5313|Loss: 0.7362993955612183:  82%|████████▏ | 5313/6470 [1:01:04<14:04,  1.37it/s]
3|5314|Loss: 0.8911553025245667:  82%|████████▏ | 5313/6470 [1:01:04<14:04,  1.37it/s]
3|5314|Loss: 0.8911553025245667:  82%|████████▏ | 5314/6470 [1:01:04<13:23,  1.44it/s]
3|5315|Loss: 0.7384524345397949:  82%|████████▏ | 5314/6470 [1:01:05<13:23,  1.44it/s]
3|5315|Loss: 0.7384524345397949:  82%|████████▏ | 5315/6470 [1:01:05<14:08,  1.36it/s]
3|5316|Loss: 0.8457660675048828:  82%|████████▏ | 5315/6470 [1:01:06<14:08,  1.36it/s]
3|5316|Loss: 0.8457660675048828:  82%|████████▏ | 5316/6470 [1:01:06<14:48,  1.30it/s]
3|5317|Loss: 0.9540625214576721:  82%|████████▏ | 5316/6470 [1:01:07<14:48,  1.30it/s]
3|5317|Loss: 0.9540625214576721:  82%|████████▏ | 5317/6470 [1:01:07<14:00,  1.37it/s]
3|5318|Loss: 0.796349823474884:  82%|████████▏ | 5317/6470 [1:01:07<14:00,  1.37it/s] 
3|5318|Loss: 0.796349823474884:  82%|████████▏ | 5318/6470 [1:01:07<13:09,  1.46it/s]
3|5319|Loss: 0.8677312135696411:  82%|████████▏ | 5318/6470 [1:01:08<13:09,  1.46it/s]
3|5319|Loss: 0.8677312135696411:  82%|████████▏ | 5319/6470 [1:01:08<13:08,  1.46it/s]
3|5320|Loss: 1.0211067199707031:  82%|████████▏ | 5319/6470 [1:01:08<13:08,  1.46it/s]
3|5320|Loss: 1.0211067199707031:  82%|████████▏ | 5320/6470 [1:01:09<12:20,  1.55it/s]
3|5321|Loss: 0.7390457391738892:  82%|████████▏ | 5320/6470 [1:01:09<12:20,  1.55it/s]
3|5321|Loss: 0.7390457391738892:  82%|████████▏ | 5321/6470 [1:01:09<13:12,  1.45it/s]
3|5322|Loss: 0.8670291304588318:  82%|████████▏ | 5321/6470 [1:01:10<13:12,  1.45it/s]
3|5322|Loss: 0.8670291304588318:  82%|████████▏ | 5322/6470 [1:01:10<12:57,  1.48it/s]
3|5323|Loss: 0.8358902335166931:  82%|████████▏ | 5322/6470 [1:01:10<12:57,  1.48it/s]
3|5323|Loss: 0.8358902335166931:  82%|████████▏ | 5323/6470 [1:01:11<13:08,  1.45it/s]
3|5324|Loss: 0.8886798620223999:  82%|████████▏ | 5323/6470 [1:01:11<13:08,  1.45it/s]
3|5324|Loss: 0.8886798620223999:  82%|████████▏ | 5324/6470 [1:01:11<13:27,  1.42it/s]
3|5325|Loss: 0.9530531764030457:  82%|████████▏ | 5324/6470 [1:01:12<13:27,  1.42it/s]
3|5325|Loss: 0.9530531764030457:  82%|████████▏ | 5325/6470 [1:01:12<13:28,  1.42it/s]
3|5326|Loss: 1.0566034317016602:  82%|████████▏ | 5325/6470 [1:01:13<13:28,  1.42it/s]
3|5326|Loss: 1.0566034317016602:  82%|████████▏ | 5326/6470 [1:01:13<13:40,  1.39it/s]
3|5327|Loss: 0.9283453226089478:  82%|████████▏ | 5326/6470 [1:01:13<13:40,  1.39it/s]
3|5327|Loss: 0.9283453226089478:  82%|████████▏ | 5327/6470 [1:01:14<13:40,  1.39it/s]
3|5328|Loss: 0.7679874300956726:  82%|████████▏ | 5327/6470 [1:01:14<13:40,  1.39it/s]
3|5328|Loss: 0.7679874300956726:  82%|████████▏ | 5328/6470 [1:01:14<13:03,  1.46it/s]
3|5329|Loss: 0.7514890432357788:  82%|████████▏ | 5328/6470 [1:01:15<13:03,  1.46it/s]
3|5329|Loss: 0.7514890432357788:  82%|████████▏ | 5329/6470 [1:01:15<12:44,  1.49it/s]
3|5330|Loss: 0.8243802189826965:  82%|████████▏ | 5329/6470 [1:01:15<12:44,  1.49it/s]
3|5330|Loss: 0.8243802189826965:  82%|████████▏ | 5330/6470 [1:01:16<12:32,  1.52it/s]
3|5331|Loss: 0.8322830200195312:  82%|████████▏ | 5330/6470 [1:01:16<12:32,  1.52it/s]
3|5331|Loss: 0.8322830200195312:  82%|████████▏ | 5331/6470 [1:01:16<12:48,  1.48it/s]
3|5332|Loss: 0.8652581572532654:  82%|████████▏ | 5331/6470 [1:01:17<12:48,  1.48it/s]
3|5332|Loss: 0.8652581572532654:  82%|████████▏ | 5332/6470 [1:01:17<12:16,  1.54it/s]
3|5333|Loss: 0.7348904013633728:  82%|████████▏ | 5332/6470 [1:01:17<12:16,  1.54it/s]
3|5333|Loss: 0.7348904013633728:  82%|████████▏ | 5333/6470 [1:01:18<12:59,  1.46it/s]
3|5334|Loss: 0.7745769023895264:  82%|████████▏ | 5333/6470 [1:01:18<12:59,  1.46it/s]
3|5334|Loss: 0.7745769023895264:  82%|████████▏ | 5334/6470 [1:01:18<12:56,  1.46it/s]
3|5335|Loss: 0.8286004066467285:  82%|████████▏ | 5334/6470 [1:01:19<12:56,  1.46it/s]
3|5335|Loss: 0.8286004066467285:  82%|████████▏ | 5335/6470 [1:01:19<12:39,  1.49it/s]
3|5336|Loss: 0.8647153973579407:  82%|████████▏ | 5335/6470 [1:01:19<12:39,  1.49it/s]
3|5336|Loss: 0.8647153973579407:  82%|████████▏ | 5336/6470 [1:01:20<12:54,  1.46it/s]
3|5337|Loss: 0.8514962792396545:  82%|████████▏ | 5336/6470 [1:01:20<12:54,  1.46it/s]
3|5337|Loss: 0.8514962792396545:  82%|████████▏ | 5337/6470 [1:01:20<12:47,  1.48it/s]
3|5338|Loss: 0.794510006904602:  82%|████████▏ | 5337/6470 [1:01:21<12:47,  1.48it/s] 
3|5338|Loss: 0.794510006904602:  83%|████████▎ | 5338/6470 [1:01:21<12:08,  1.55it/s]
3|5339|Loss: 0.6171743869781494:  83%|████████▎ | 5338/6470 [1:01:21<12:08,  1.55it/s]
3|5339|Loss: 0.6171743869781494:  83%|████████▎ | 5339/6470 [1:01:22<12:42,  1.48it/s]
3|5340|Loss: 0.8422624468803406:  83%|████████▎ | 5339/6470 [1:01:22<12:42,  1.48it/s]
3|5340|Loss: 0.8422624468803406:  83%|████████▎ | 5340/6470 [1:01:22<13:26,  1.40it/s]
3|5341|Loss: 0.880702793598175:  83%|████████▎ | 5340/6470 [1:01:23<13:26,  1.40it/s] 
3|5341|Loss: 0.880702793598175:  83%|████████▎ | 5341/6470 [1:01:23<13:41,  1.37it/s]
3|5342|Loss: 0.9062315821647644:  83%|████████▎ | 5341/6470 [1:01:24<13:41,  1.37it/s]
3|5342|Loss: 0.9062315821647644:  83%|████████▎ | 5342/6470 [1:01:24<12:29,  1.51it/s]
3|5343|Loss: 0.8520539402961731:  83%|████████▎ | 5342/6470 [1:01:24<12:29,  1.51it/s]
3|5343|Loss: 0.8520539402961731:  83%|████████▎ | 5343/6470 [1:01:24<11:59,  1.57it/s]
3|5344|Loss: 0.7787477970123291:  83%|████████▎ | 5343/6470 [1:01:25<11:59,  1.57it/s]
3|5344|Loss: 0.7787477970123291:  83%|████████▎ | 5344/6470 [1:01:25<11:27,  1.64it/s]
3|5345|Loss: 0.8573577404022217:  83%|████████▎ | 5344/6470 [1:01:25<11:27,  1.64it/s]
3|5345|Loss: 0.8573577404022217:  83%|████████▎ | 5345/6470 [1:01:25<10:58,  1.71it/s]
3|5346|Loss: 0.9116352796554565:  83%|████████▎ | 5345/6470 [1:01:26<10:58,  1.71it/s]
3|5346|Loss: 0.9116352796554565:  83%|████████▎ | 5346/6470 [1:01:26<10:55,  1.71it/s]
3|5347|Loss: 0.8411358594894409:  83%|████████▎ | 5346/6470 [1:01:26<10:55,  1.71it/s]
3|5347|Loss: 0.8411358594894409:  83%|████████▎ | 5347/6470 [1:01:26<10:56,  1.71it/s]
3|5348|Loss: 0.9632672071456909:  83%|████████▎ | 5347/6470 [1:01:27<10:56,  1.71it/s]
3|5348|Loss: 0.9632672071456909:  83%|████████▎ | 5348/6470 [1:01:27<11:36,  1.61it/s]
3|5349|Loss: 0.8767907023429871:  83%|████████▎ | 5348/6470 [1:01:28<11:36,  1.61it/s]
3|5349|Loss: 0.8767907023429871:  83%|████████▎ | 5349/6470 [1:01:28<12:48,  1.46it/s]
3|5350|Loss: 0.9939408898353577:  83%|████████▎ | 5349/6470 [1:01:28<12:48,  1.46it/s]
3|5350|Loss: 0.9939408898353577:  83%|████████▎ | 5350/6470 [1:01:29<12:50,  1.45it/s]
3|5351|Loss: 0.8596609234809875:  83%|████████▎ | 5350/6470 [1:01:29<12:50,  1.45it/s]
3|5351|Loss: 0.8596609234809875:  83%|████████▎ | 5351/6470 [1:01:30<13:32,  1.38it/s]
3|5352|Loss: 1.0039738416671753:  83%|████████▎ | 5351/6470 [1:01:30<13:32,  1.38it/s]
3|5352|Loss: 1.0039738416671753:  83%|████████▎ | 5352/6470 [1:01:30<13:46,  1.35it/s]
3|5353|Loss: 0.8311542272567749:  83%|████████▎ | 5352/6470 [1:01:31<13:46,  1.35it/s]
3|5353|Loss: 0.8311542272567749:  83%|████████▎ | 5353/6470 [1:01:31<13:46,  1.35it/s]
3|5354|Loss: 0.8904820084571838:  83%|████████▎ | 5353/6470 [1:01:31<13:46,  1.35it/s]
3|5354|Loss: 0.8904820084571838:  83%|████████▎ | 5354/6470 [1:01:32<13:47,  1.35it/s]
3|5355|Loss: 0.772650957107544:  83%|████████▎ | 5354/6470 [1:01:32<13:47,  1.35it/s] 
3|5355|Loss: 0.772650957107544:  83%|████████▎ | 5355/6470 [1:01:33<13:43,  1.35it/s]
3|5356|Loss: 0.8431814312934875:  83%|████████▎ | 5355/6470 [1:01:33<13:43,  1.35it/s]
3|5356|Loss: 0.8431814312934875:  83%|████████▎ | 5356/6470 [1:01:33<12:51,  1.44it/s]
3|5357|Loss: 0.9387518167495728:  83%|████████▎ | 5356/6470 [1:01:33<12:51,  1.44it/s]
3|5357|Loss: 0.9387518167495728:  83%|████████▎ | 5357/6470 [1:01:34<11:59,  1.55it/s]
3|5358|Loss: 0.8631036877632141:  83%|████████▎ | 5357/6470 [1:01:34<11:59,  1.55it/s]
3|5358|Loss: 0.8631036877632141:  83%|████████▎ | 5358/6470 [1:01:34<12:27,  1.49it/s]
3|5359|Loss: 0.9295178651809692:  83%|████████▎ | 5358/6470 [1:01:35<12:27,  1.49it/s]
3|5359|Loss: 0.9295178651809692:  83%|████████▎ | 5359/6470 [1:01:35<13:06,  1.41it/s]
3|5360|Loss: 0.942631721496582:  83%|████████▎ | 5359/6470 [1:01:36<13:06,  1.41it/s] 
3|5360|Loss: 0.942631721496582:  83%|████████▎ | 5360/6470 [1:01:36<12:38,  1.46it/s]
3|5361|Loss: 0.856752336025238:  83%|████████▎ | 5360/6470 [1:01:36<12:38,  1.46it/s]
3|5361|Loss: 0.856752336025238:  83%|████████▎ | 5361/6470 [1:01:36<12:13,  1.51it/s]
3|5362|Loss: 0.7198178172111511:  83%|████████▎ | 5361/6470 [1:01:37<12:13,  1.51it/s]
3|5362|Loss: 0.7198178172111511:  83%|████████▎ | 5362/6470 [1:01:37<13:06,  1.41it/s]
3|5363|Loss: 0.8188171982765198:  83%|████████▎ | 5362/6470 [1:01:38<13:06,  1.41it/s]
3|5363|Loss: 0.8188171982765198:  83%|████████▎ | 5363/6470 [1:01:38<12:18,  1.50it/s]
3|5364|Loss: 0.8465781807899475:  83%|████████▎ | 5363/6470 [1:01:38<12:18,  1.50it/s]
3|5364|Loss: 0.8465781807899475:  83%|████████▎ | 5364/6470 [1:01:38<11:34,  1.59it/s]
3|5365|Loss: 0.9404364824295044:  83%|████████▎ | 5364/6470 [1:01:39<11:34,  1.59it/s]
3|5365|Loss: 0.9404364824295044:  83%|████████▎ | 5365/6470 [1:01:39<11:55,  1.54it/s]
3|5366|Loss: 0.8206390738487244:  83%|████████▎ | 5365/6470 [1:01:39<11:55,  1.54it/s]
3|5366|Loss: 0.8206390738487244:  83%|████████▎ | 5366/6470 [1:01:40<11:54,  1.55it/s]
3|5367|Loss: 0.9425199627876282:  83%|████████▎ | 5366/6470 [1:01:40<11:54,  1.55it/s]
3|5367|Loss: 0.9425199627876282:  83%|████████▎ | 5367/6470 [1:01:40<12:42,  1.45it/s]
3|5368|Loss: 0.7968217134475708:  83%|████████▎ | 5367/6470 [1:01:41<12:42,  1.45it/s]
3|5368|Loss: 0.7968217134475708:  83%|████████▎ | 5368/6470 [1:01:41<13:18,  1.38it/s]
3|5369|Loss: 0.8657153844833374:  83%|████████▎ | 5368/6470 [1:01:42<13:18,  1.38it/s]
3|5369|Loss: 0.8657153844833374:  83%|████████▎ | 5369/6470 [1:01:42<13:17,  1.38it/s]
3|5370|Loss: 0.965958833694458:  83%|████████▎ | 5369/6470 [1:01:42<13:17,  1.38it/s] 
3|5370|Loss: 0.965958833694458:  83%|████████▎ | 5370/6470 [1:01:43<12:53,  1.42it/s]
3|5371|Loss: 0.8193053007125854:  83%|████████▎ | 5370/6470 [1:01:43<12:53,  1.42it/s]
3|5371|Loss: 0.8193053007125854:  83%|████████▎ | 5371/6470 [1:01:43<13:28,  1.36it/s]
3|5372|Loss: 0.8649839162826538:  83%|████████▎ | 5371/6470 [1:01:44<13:28,  1.36it/s]
3|5372|Loss: 0.8649839162826538:  83%|████████▎ | 5372/6470 [1:01:44<14:15,  1.28it/s]
3|5373|Loss: 0.7739031314849854:  83%|████████▎ | 5372/6470 [1:01:45<14:15,  1.28it/s]
3|5373|Loss: 0.7739031314849854:  83%|████████▎ | 5373/6470 [1:01:45<14:35,  1.25it/s]
3|5374|Loss: 0.8281238675117493:  83%|████████▎ | 5373/6470 [1:01:46<14:35,  1.25it/s]
3|5374|Loss: 0.8281238675117493:  83%|████████▎ | 5374/6470 [1:01:46<14:38,  1.25it/s]
3|5375|Loss: 0.7107088565826416:  83%|████████▎ | 5374/6470 [1:01:46<14:38,  1.25it/s]
3|5375|Loss: 0.7107088565826416:  83%|████████▎ | 5375/6470 [1:01:46<12:40,  1.44it/s]
3|5376|Loss: 0.881642758846283:  83%|████████▎ | 5375/6470 [1:01:47<12:40,  1.44it/s] 
3|5376|Loss: 0.881642758846283:  83%|████████▎ | 5376/6470 [1:01:47<11:34,  1.57it/s]
3|5377|Loss: 0.935299813747406:  83%|████████▎ | 5376/6470 [1:01:47<11:34,  1.57it/s]
3|5377|Loss: 0.935299813747406:  83%|████████▎ | 5377/6470 [1:01:48<11:22,  1.60it/s]
3|5378|Loss: 0.9042563438415527:  83%|████████▎ | 5377/6470 [1:01:48<11:22,  1.60it/s]
3|5378|Loss: 0.9042563438415527:  83%|████████▎ | 5378/6470 [1:01:48<11:26,  1.59it/s]
3|5379|Loss: 0.9896853566169739:  83%|████████▎ | 5378/6470 [1:01:49<11:26,  1.59it/s]
3|5379|Loss: 0.9896853566169739:  83%|████████▎ | 5379/6470 [1:01:49<11:44,  1.55it/s]
3|5380|Loss: 0.9043592214584351:  83%|████████▎ | 5379/6470 [1:01:49<11:44,  1.55it/s]
3|5380|Loss: 0.9043592214584351:  83%|████████▎ | 5380/6470 [1:01:50<11:38,  1.56it/s]
3|5381|Loss: 0.7087799310684204:  83%|████████▎ | 5380/6470 [1:01:50<11:38,  1.56it/s]
3|5381|Loss: 0.7087799310684204:  83%|████████▎ | 5381/6470 [1:01:50<11:29,  1.58it/s]
3|5382|Loss: 0.6890997886657715:  83%|████████▎ | 5381/6470 [1:01:50<11:29,  1.58it/s]
3|5382|Loss: 0.6890997886657715:  83%|████████▎ | 5382/6470 [1:01:51<11:00,  1.65it/s]
3|5383|Loss: 0.8978835344314575:  83%|████████▎ | 5382/6470 [1:01:51<11:00,  1.65it/s]
3|5383|Loss: 0.8978835344314575:  83%|████████▎ | 5383/6470 [1:01:51<11:25,  1.59it/s]
3|5384|Loss: 0.9141181707382202:  83%|████████▎ | 5383/6470 [1:01:52<11:25,  1.59it/s]
3|5384|Loss: 0.9141181707382202:  83%|████████▎ | 5384/6470 [1:01:52<11:50,  1.53it/s]
3|5385|Loss: 0.8748254776000977:  83%|████████▎ | 5384/6470 [1:01:52<11:50,  1.53it/s]
3|5385|Loss: 0.8748254776000977:  83%|████████▎ | 5385/6470 [1:01:53<12:37,  1.43it/s]
3|5386|Loss: 0.8952093124389648:  83%|████████▎ | 5385/6470 [1:01:53<12:37,  1.43it/s]
3|5386|Loss: 0.8952093124389648:  83%|████████▎ | 5386/6470 [1:01:54<13:23,  1.35it/s]
3|5387|Loss: 0.7943999767303467:  83%|████████▎ | 5386/6470 [1:01:54<13:23,  1.35it/s]
3|5387|Loss: 0.7943999767303467:  83%|████████▎ | 5387/6470 [1:01:54<12:53,  1.40it/s]
3|5388|Loss: 0.8198121786117554:  83%|████████▎ | 5387/6470 [1:01:55<12:53,  1.40it/s]
3|5388|Loss: 0.8198121786117554:  83%|████████▎ | 5388/6470 [1:01:55<13:20,  1.35it/s]
3|5389|Loss: 0.9088184237480164:  83%|████████▎ | 5388/6470 [1:01:56<13:20,  1.35it/s]
3|5389|Loss: 0.9088184237480164:  83%|████████▎ | 5389/6470 [1:01:56<12:46,  1.41it/s]
3|5390|Loss: 0.7903522253036499:  83%|████████▎ | 5389/6470 [1:01:56<12:46,  1.41it/s]
3|5390|Loss: 0.7903522253036499:  83%|████████▎ | 5390/6470 [1:01:56<12:35,  1.43it/s]
3|5391|Loss: 0.9745789766311646:  83%|████████▎ | 5390/6470 [1:01:57<12:35,  1.43it/s]
3|5391|Loss: 0.9745789766311646:  83%|████████▎ | 5391/6470 [1:01:57<12:44,  1.41it/s]
3|5392|Loss: 0.8962991237640381:  83%|████████▎ | 5391/6470 [1:01:58<12:44,  1.41it/s]
3|5392|Loss: 0.8962991237640381:  83%|████████▎ | 5392/6470 [1:01:58<13:11,  1.36it/s]
3|5393|Loss: 0.8534601330757141:  83%|████████▎ | 5392/6470 [1:01:58<13:11,  1.36it/s]
3|5393|Loss: 0.8534601330757141:  83%|████████▎ | 5393/6470 [1:01:59<13:49,  1.30it/s]
3|5394|Loss: 0.9262369871139526:  83%|████████▎ | 5393/6470 [1:01:59<13:49,  1.30it/s]
3|5394|Loss: 0.9262369871139526:  83%|████████▎ | 5394/6470 [1:02:00<13:17,  1.35it/s]
3|5395|Loss: 0.8037117123603821:  83%|████████▎ | 5394/6470 [1:02:00<13:17,  1.35it/s]
3|5395|Loss: 0.8037117123603821:  83%|████████▎ | 5395/6470 [1:02:00<12:44,  1.41it/s]
3|5396|Loss: 0.8578927516937256:  83%|████████▎ | 5395/6470 [1:02:01<12:44,  1.41it/s]
3|5396|Loss: 0.8578927516937256:  83%|████████▎ | 5396/6470 [1:02:01<12:22,  1.45it/s]
3|5397|Loss: 0.8284043669700623:  83%|████████▎ | 5396/6470 [1:02:01<12:22,  1.45it/s]
3|5397|Loss: 0.8284043669700623:  83%|████████▎ | 5397/6470 [1:02:01<11:48,  1.52it/s]
3|5398|Loss: 0.838797390460968:  83%|████████▎ | 5397/6470 [1:02:02<11:48,  1.52it/s] 
3|5398|Loss: 0.838797390460968:  83%|████████▎ | 5398/6470 [1:02:02<12:19,  1.45it/s]
3|5399|Loss: 0.9050849676132202:  83%|████████▎ | 5398/6470 [1:02:03<12:19,  1.45it/s]
3|5399|Loss: 0.9050849676132202:  83%|████████▎ | 5399/6470 [1:02:03<13:10,  1.35it/s]
3|5400|Loss: 0.7972150444984436:  83%|████████▎ | 5399/6470 [1:02:03<13:10,  1.35it/s]
3|5400|Loss: 0.7972150444984436:  83%|████████▎ | 5400/6470 [1:02:04<13:55,  1.28it/s]
3|5401|Loss: 0.9631785750389099:  83%|████████▎ | 5400/6470 [1:02:04<13:55,  1.28it/s]
3|5401|Loss: 0.9631785750389099:  83%|████████▎ | 5401/6470 [1:02:05<14:08,  1.26it/s]
3|5402|Loss: 0.885664701461792:  83%|████████▎ | 5401/6470 [1:02:05<14:08,  1.26it/s] 
3|5402|Loss: 0.885664701461792:  83%|████████▎ | 5402/6470 [1:02:05<13:44,  1.30it/s]
3|5403|Loss: 0.8466658592224121:  83%|████████▎ | 5402/6470 [1:02:06<13:44,  1.30it/s]
3|5403|Loss: 0.8466658592224121:  84%|████████▎ | 5403/6470 [1:02:06<13:33,  1.31it/s]
3|5404|Loss: 0.9033102989196777:  84%|████████▎ | 5403/6470 [1:02:07<13:33,  1.31it/s]
3|5404|Loss: 0.9033102989196777:  84%|████████▎ | 5404/6470 [1:02:07<13:24,  1.32it/s]
3|5405|Loss: 0.8455915451049805:  84%|████████▎ | 5404/6470 [1:02:07<13:24,  1.32it/s]
3|5405|Loss: 0.8455915451049805:  84%|████████▎ | 5405/6470 [1:02:08<13:10,  1.35it/s]
3|5406|Loss: 0.9249724745750427:  84%|████████▎ | 5405/6470 [1:02:08<13:10,  1.35it/s]
3|5406|Loss: 0.9249724745750427:  84%|████████▎ | 5406/6470 [1:02:08<11:59,  1.48it/s]
3|5407|Loss: 0.806555449962616:  84%|████████▎ | 5406/6470 [1:02:08<11:59,  1.48it/s] 
3|5407|Loss: 0.806555449962616:  84%|████████▎ | 5407/6470 [1:02:09<11:36,  1.53it/s]
3|5408|Loss: 0.7597967386245728:  84%|████████▎ | 5407/6470 [1:02:09<11:36,  1.53it/s]
3|5408|Loss: 0.7597967386245728:  84%|████████▎ | 5408/6470 [1:02:09<10:43,  1.65it/s]
3|5409|Loss: 0.8592969179153442:  84%|████████▎ | 5408/6470 [1:02:10<10:43,  1.65it/s]
3|5409|Loss: 0.8592969179153442:  84%|████████▎ | 5409/6470 [1:02:10<10:38,  1.66it/s]
3|5410|Loss: 0.7797710299491882:  84%|████████▎ | 5409/6470 [1:02:10<10:38,  1.66it/s]
3|5410|Loss: 0.7797710299491882:  84%|████████▎ | 5410/6470 [1:02:10<10:49,  1.63it/s]
3|5411|Loss: 0.8236188888549805:  84%|████████▎ | 5410/6470 [1:02:11<10:49,  1.63it/s]
3|5411|Loss: 0.8236188888549805:  84%|████████▎ | 5411/6470 [1:02:11<11:22,  1.55it/s]
3|5412|Loss: 0.7959810495376587:  84%|████████▎ | 5411/6470 [1:02:12<11:22,  1.55it/s]
3|5412|Loss: 0.7959810495376587:  84%|████████▎ | 5412/6470 [1:02:12<11:20,  1.55it/s]
3|5413|Loss: 0.8921020030975342:  84%|████████▎ | 5412/6470 [1:02:12<11:20,  1.55it/s]
3|5413|Loss: 0.8921020030975342:  84%|████████▎ | 5413/6470 [1:02:12<11:12,  1.57it/s]
3|5414|Loss: 0.8801165819168091:  84%|████████▎ | 5413/6470 [1:02:13<11:12,  1.57it/s]
3|5414|Loss: 0.8801165819168091:  84%|████████▎ | 5414/6470 [1:02:13<11:28,  1.53it/s]
3|5415|Loss: 0.9405547380447388:  84%|████████▎ | 5414/6470 [1:02:14<11:28,  1.53it/s]
3|5415|Loss: 0.9405547380447388:  84%|████████▎ | 5415/6470 [1:02:14<11:29,  1.53it/s]
3|5416|Loss: 0.8372762203216553:  84%|████████▎ | 5415/6470 [1:02:14<11:29,  1.53it/s]
3|5416|Loss: 0.8372762203216553:  84%|████████▎ | 5416/6470 [1:02:14<11:09,  1.57it/s]
3|5417|Loss: 0.8954459428787231:  84%|████████▎ | 5416/6470 [1:02:15<11:09,  1.57it/s]
3|5417|Loss: 0.8954459428787231:  84%|████████▎ | 5417/6470 [1:02:15<11:08,  1.58it/s]
3|5418|Loss: 0.9023708701133728:  84%|████████▎ | 5417/6470 [1:02:15<11:08,  1.58it/s]
3|5418|Loss: 0.9023708701133728:  84%|████████▎ | 5418/6470 [1:02:16<11:08,  1.57it/s]
3|5419|Loss: 0.9300809502601624:  84%|████████▎ | 5418/6470 [1:02:16<11:08,  1.57it/s]
3|5419|Loss: 0.9300809502601624:  84%|████████▍ | 5419/6470 [1:02:16<11:09,  1.57it/s]
3|5420|Loss: 0.9280336499214172:  84%|████████▍ | 5419/6470 [1:02:17<11:09,  1.57it/s]
3|5420|Loss: 0.9280336499214172:  84%|████████▍ | 5420/6470 [1:02:17<10:23,  1.68it/s]
3|5421|Loss: 0.8352074027061462:  84%|████████▍ | 5420/6470 [1:02:17<10:23,  1.68it/s]
3|5421|Loss: 0.8352074027061462:  84%|████████▍ | 5421/6470 [1:02:17<09:55,  1.76it/s]
3|5422|Loss: 0.9825011491775513:  84%|████████▍ | 5421/6470 [1:02:18<09:55,  1.76it/s]
3|5422|Loss: 0.9825011491775513:  84%|████████▍ | 5422/6470 [1:02:18<09:57,  1.75it/s]
3|5423|Loss: 0.8675035238265991:  84%|████████▍ | 5422/6470 [1:02:18<09:57,  1.75it/s]
3|5423|Loss: 0.8675035238265991:  84%|████████▍ | 5423/6470 [1:02:19<10:30,  1.66it/s]
3|5424|Loss: 0.8475914001464844:  84%|████████▍ | 5423/6470 [1:02:19<10:30,  1.66it/s]
3|5424|Loss: 0.8475914001464844:  84%|████████▍ | 5424/6470 [1:02:19<10:29,  1.66it/s]
3|5425|Loss: 0.7197932600975037:  84%|████████▍ | 5424/6470 [1:02:20<10:29,  1.66it/s]
3|5425|Loss: 0.7197932600975037:  84%|████████▍ | 5425/6470 [1:02:20<10:53,  1.60it/s]
3|5426|Loss: 0.724484920501709:  84%|████████▍ | 5425/6470 [1:02:20<10:53,  1.60it/s] 
3|5426|Loss: 0.724484920501709:  84%|████████▍ | 5426/6470 [1:02:20<10:49,  1.61it/s]
3|5427|Loss: 0.8089097738265991:  84%|████████▍ | 5426/6470 [1:02:21<10:49,  1.61it/s]
3|5427|Loss: 0.8089097738265991:  84%|████████▍ | 5427/6470 [1:02:21<11:16,  1.54it/s]
3|5428|Loss: 0.8258658647537231:  84%|████████▍ | 5427/6470 [1:02:22<11:16,  1.54it/s]
3|5428|Loss: 0.8258658647537231:  84%|████████▍ | 5428/6470 [1:02:22<11:49,  1.47it/s]
3|5429|Loss: 0.8532381653785706:  84%|████████▍ | 5428/6470 [1:02:22<11:49,  1.47it/s]
3|5429|Loss: 0.8532381653785706:  84%|████████▍ | 5429/6470 [1:02:23<12:37,  1.37it/s]
3|5430|Loss: 0.8814200162887573:  84%|████████▍ | 5429/6470 [1:02:23<12:37,  1.37it/s]
3|5430|Loss: 0.8814200162887573:  84%|████████▍ | 5430/6470 [1:02:23<12:20,  1.40it/s]
3|5431|Loss: 0.7598580718040466:  84%|████████▍ | 5430/6470 [1:02:24<12:20,  1.40it/s]
3|5431|Loss: 0.7598580718040466:  84%|████████▍ | 5431/6470 [1:02:24<11:39,  1.49it/s]
3|5432|Loss: 0.9655254483222961:  84%|████████▍ | 5431/6470 [1:02:24<11:39,  1.49it/s]
3|5432|Loss: 0.9655254483222961:  84%|████████▍ | 5432/6470 [1:02:25<11:50,  1.46it/s]
3|5433|Loss: 0.8441361784934998:  84%|████████▍ | 5432/6470 [1:02:25<11:50,  1.46it/s]
3|5433|Loss: 0.8441361784934998:  84%|████████▍ | 5433/6470 [1:02:25<11:42,  1.48it/s]
3|5434|Loss: 0.8981074690818787:  84%|████████▍ | 5433/6470 [1:02:26<11:42,  1.48it/s]
3|5434|Loss: 0.8981074690818787:  84%|████████▍ | 5434/6470 [1:02:26<11:46,  1.47it/s]
3|5435|Loss: 0.86042320728302:  84%|████████▍ | 5434/6470 [1:02:26<11:46,  1.47it/s]  
3|5435|Loss: 0.86042320728302:  84%|████████▍ | 5435/6470 [1:02:27<11:31,  1.50it/s]
3|5436|Loss: 0.8200253844261169:  84%|████████▍ | 5435/6470 [1:02:27<11:31,  1.50it/s]
3|5436|Loss: 0.8200253844261169:  84%|████████▍ | 5436/6470 [1:02:27<11:36,  1.48it/s]
3|5437|Loss: 0.8594800233840942:  84%|████████▍ | 5436/6470 [1:02:28<11:36,  1.48it/s]
3|5437|Loss: 0.8594800233840942:  84%|████████▍ | 5437/6470 [1:02:28<11:47,  1.46it/s]
3|5438|Loss: 0.8714187145233154:  84%|████████▍ | 5437/6470 [1:02:28<11:47,  1.46it/s]
3|5438|Loss: 0.8714187145233154:  84%|████████▍ | 5438/6470 [1:02:29<11:30,  1.50it/s]
3|5439|Loss: 0.7188365459442139:  84%|████████▍ | 5438/6470 [1:02:29<11:30,  1.50it/s]
3|5439|Loss: 0.7188365459442139:  84%|████████▍ | 5439/6470 [1:02:29<11:11,  1.53it/s]
3|5440|Loss: 0.8437120318412781:  84%|████████▍ | 5439/6470 [1:02:30<11:11,  1.53it/s]
3|5440|Loss: 0.8437120318412781:  84%|████████▍ | 5440/6470 [1:02:30<10:51,  1.58it/s]
3|5441|Loss: 0.7320792078971863:  84%|████████▍ | 5440/6470 [1:02:30<10:51,  1.58it/s]
3|5441|Loss: 0.7320792078971863:  84%|████████▍ | 5441/6470 [1:02:31<10:50,  1.58it/s]
3|5442|Loss: 0.9537249207496643:  84%|████████▍ | 5441/6470 [1:02:31<10:50,  1.58it/s]
3|5442|Loss: 0.9537249207496643:  84%|████████▍ | 5442/6470 [1:02:31<11:16,  1.52it/s]
3|5443|Loss: 0.8926948308944702:  84%|████████▍ | 5442/6470 [1:02:32<11:16,  1.52it/s]
3|5443|Loss: 0.8926948308944702:  84%|████████▍ | 5443/6470 [1:02:32<11:16,  1.52it/s]
3|5444|Loss: 1.1158634424209595:  84%|████████▍ | 5443/6470 [1:02:32<11:16,  1.52it/s]
3|5444|Loss: 1.1158634424209595:  84%|████████▍ | 5444/6470 [1:02:33<11:32,  1.48it/s]
3|5445|Loss: 0.8461865782737732:  84%|████████▍ | 5444/6470 [1:02:33<11:32,  1.48it/s]
3|5445|Loss: 0.8461865782737732:  84%|████████▍ | 5445/6470 [1:02:34<12:45,  1.34it/s]
3|5446|Loss: 0.7796223759651184:  84%|████████▍ | 5445/6470 [1:02:34<12:45,  1.34it/s]
3|5446|Loss: 0.7796223759651184:  84%|████████▍ | 5446/6470 [1:02:34<12:29,  1.37it/s]
3|5447|Loss: 0.7235599160194397:  84%|████████▍ | 5446/6470 [1:02:35<12:29,  1.37it/s]
3|5447|Loss: 0.7235599160194397:  84%|████████▍ | 5447/6470 [1:02:35<13:04,  1.30it/s]
3|5448|Loss: 0.8491759896278381:  84%|████████▍ | 5447/6470 [1:02:36<13:04,  1.30it/s]
3|5448|Loss: 0.8491759896278381:  84%|████████▍ | 5448/6470 [1:02:36<13:35,  1.25it/s]
3|5449|Loss: 0.8467144966125488:  84%|████████▍ | 5448/6470 [1:02:36<13:35,  1.25it/s]
3|5449|Loss: 0.8467144966125488:  84%|████████▍ | 5449/6470 [1:02:37<14:01,  1.21it/s]
3|5450|Loss: 0.7379104495048523:  84%|████████▍ | 5449/6470 [1:02:37<14:01,  1.21it/s]
3|5450|Loss: 0.7379104495048523:  84%|████████▍ | 5450/6470 [1:02:38<14:13,  1.19it/s]
3|5451|Loss: 0.772460401058197:  84%|████████▍ | 5450/6470 [1:02:38<14:13,  1.19it/s] 
3|5451|Loss: 0.772460401058197:  84%|████████▍ | 5451/6470 [1:02:38<12:55,  1.31it/s]
3|5452|Loss: 0.8376902937889099:  84%|████████▍ | 5451/6470 [1:02:39<12:55,  1.31it/s]
3|5452|Loss: 0.8376902937889099:  84%|████████▍ | 5452/6470 [1:02:39<12:33,  1.35it/s]
3|5453|Loss: 0.8202764391899109:  84%|████████▍ | 5452/6470 [1:02:39<12:33,  1.35it/s]
3|5453|Loss: 0.8202764391899109:  84%|████████▍ | 5453/6470 [1:02:40<12:24,  1.37it/s]
3|5454|Loss: 0.6918092966079712:  84%|████████▍ | 5453/6470 [1:02:40<12:24,  1.37it/s]
3|5454|Loss: 0.6918092966079712:  84%|████████▍ | 5454/6470 [1:02:40<11:43,  1.45it/s]
3|5455|Loss: 0.8016286492347717:  84%|████████▍ | 5454/6470 [1:02:41<11:43,  1.45it/s]
3|5455|Loss: 0.8016286492347717:  84%|████████▍ | 5455/6470 [1:02:41<11:37,  1.46it/s]
3|5456|Loss: 0.807198166847229:  84%|████████▍ | 5455/6470 [1:02:41<11:37,  1.46it/s] 
3|5456|Loss: 0.807198166847229:  84%|████████▍ | 5456/6470 [1:02:42<11:10,  1.51it/s]
3|5457|Loss: 0.7934078574180603:  84%|████████▍ | 5456/6470 [1:02:42<11:10,  1.51it/s]
3|5457|Loss: 0.7934078574180603:  84%|████████▍ | 5457/6470 [1:02:42<11:21,  1.49it/s]
3|5458|Loss: 0.8616894483566284:  84%|████████▍ | 5457/6470 [1:02:43<11:21,  1.49it/s]
3|5458|Loss: 0.8616894483566284:  84%|████████▍ | 5458/6470 [1:02:43<11:40,  1.45it/s]
3|5459|Loss: 0.97380131483078:  84%|████████▍ | 5458/6470 [1:02:43<11:40,  1.45it/s]  
3|5459|Loss: 0.97380131483078:  84%|████████▍ | 5459/6470 [1:02:44<11:53,  1.42it/s]
3|5460|Loss: 1.0511996746063232:  84%|████████▍ | 5459/6470 [1:02:44<11:53,  1.42it/s]
3|5460|Loss: 1.0511996746063232:  84%|████████▍ | 5460/6470 [1:02:45<12:24,  1.36it/s]
3|5461|Loss: 0.7308531403541565:  84%|████████▍ | 5460/6470 [1:02:45<12:24,  1.36it/s]
3|5461|Loss: 0.7308531403541565:  84%|████████▍ | 5461/6470 [1:02:45<12:51,  1.31it/s]
3|5462|Loss: 0.8680371642112732:  84%|████████▍ | 5461/6470 [1:02:46<12:51,  1.31it/s]
3|5462|Loss: 0.8680371642112732:  84%|████████▍ | 5462/6470 [1:02:46<12:52,  1.31it/s]
3|5463|Loss: 0.8454008102416992:  84%|████████▍ | 5462/6470 [1:02:47<12:52,  1.31it/s]
3|5463|Loss: 0.8454008102416992:  84%|████████▍ | 5463/6470 [1:02:47<12:21,  1.36it/s]
3|5464|Loss: 0.7458509802818298:  84%|████████▍ | 5463/6470 [1:02:47<12:21,  1.36it/s]
3|5464|Loss: 0.7458509802818298:  84%|████████▍ | 5464/6470 [1:02:48<12:43,  1.32it/s]
3|5465|Loss: 0.7955935001373291:  84%|████████▍ | 5464/6470 [1:02:48<12:43,  1.32it/s]
3|5465|Loss: 0.7955935001373291:  84%|████████▍ | 5465/6470 [1:02:48<12:37,  1.33it/s]
3|5466|Loss: 0.8047258853912354:  84%|████████▍ | 5465/6470 [1:02:49<12:37,  1.33it/s]
3|5466|Loss: 0.8047258853912354:  84%|████████▍ | 5466/6470 [1:02:49<12:06,  1.38it/s]
3|5467|Loss: 0.9994115829467773:  84%|████████▍ | 5466/6470 [1:02:49<12:06,  1.38it/s]
3|5467|Loss: 0.9994115829467773:  84%|████████▍ | 5467/6470 [1:02:50<12:03,  1.39it/s]
3|5468|Loss: 0.9226043820381165:  84%|████████▍ | 5467/6470 [1:02:50<12:03,  1.39it/s]
3|5468|Loss: 0.9226043820381165:  85%|████████▍ | 5468/6470 [1:02:51<12:08,  1.37it/s]
3|5469|Loss: 0.8139572143554688:  85%|████████▍ | 5468/6470 [1:02:51<12:08,  1.37it/s]
3|5469|Loss: 0.8139572143554688:  85%|████████▍ | 5469/6470 [1:02:51<12:47,  1.30it/s]
3|5470|Loss: 0.8308010697364807:  85%|████████▍ | 5469/6470 [1:02:52<12:47,  1.30it/s]
3|5470|Loss: 0.8308010697364807:  85%|████████▍ | 5470/6470 [1:02:52<12:28,  1.34it/s]
3|5471|Loss: 0.9074382185935974:  85%|████████▍ | 5470/6470 [1:02:52<12:28,  1.34it/s]
3|5471|Loss: 0.9074382185935974:  85%|████████▍ | 5471/6470 [1:02:53<12:11,  1.36it/s]
3|5472|Loss: 0.8624388575553894:  85%|████████▍ | 5471/6470 [1:02:53<12:11,  1.36it/s]
3|5472|Loss: 0.8624388575553894:  85%|████████▍ | 5472/6470 [1:02:53<12:02,  1.38it/s]
3|5473|Loss: 0.705251157283783:  85%|████████▍ | 5472/6470 [1:02:54<12:02,  1.38it/s] 
3|5473|Loss: 0.705251157283783:  85%|████████▍ | 5473/6470 [1:02:54<12:06,  1.37it/s]
3|5474|Loss: 1.0002977848052979:  85%|████████▍ | 5473/6470 [1:02:55<12:06,  1.37it/s]
3|5474|Loss: 1.0002977848052979:  85%|████████▍ | 5474/6470 [1:02:55<12:07,  1.37it/s]
3|5475|Loss: 1.0212198495864868:  85%|████████▍ | 5474/6470 [1:02:55<12:07,  1.37it/s]
3|5475|Loss: 1.0212198495864868:  85%|████████▍ | 5475/6470 [1:02:56<12:09,  1.36it/s]
3|5476|Loss: 0.9036619067192078:  85%|████████▍ | 5475/6470 [1:02:56<12:09,  1.36it/s]
3|5476|Loss: 0.9036619067192078:  85%|████████▍ | 5476/6470 [1:02:56<11:21,  1.46it/s]
3|5477|Loss: 0.904484748840332:  85%|████████▍ | 5476/6470 [1:02:57<11:21,  1.46it/s] 
3|5477|Loss: 0.904484748840332:  85%|████████▍ | 5477/6470 [1:02:57<10:46,  1.54it/s]
3|5478|Loss: 0.7546706795692444:  85%|████████▍ | 5477/6470 [1:02:57<10:46,  1.54it/s]
3|5478|Loss: 0.7546706795692444:  85%|████████▍ | 5478/6470 [1:02:57<10:38,  1.55it/s]
3|5479|Loss: 0.8722569942474365:  85%|████████▍ | 5478/6470 [1:02:58<10:38,  1.55it/s]
3|5479|Loss: 0.8722569942474365:  85%|████████▍ | 5479/6470 [1:02:58<10:05,  1.64it/s]
3|5480|Loss: 0.8517212867736816:  85%|████████▍ | 5479/6470 [1:02:58<10:05,  1.64it/s]
3|5480|Loss: 0.8517212867736816:  85%|████████▍ | 5480/6470 [1:02:59<10:05,  1.64it/s]
3|5481|Loss: 0.9102761745452881:  85%|████████▍ | 5480/6470 [1:02:59<10:05,  1.64it/s]
3|5481|Loss: 0.9102761745452881:  85%|████████▍ | 5481/6470 [1:02:59<11:05,  1.49it/s]
3|5482|Loss: 1.0217061042785645:  85%|████████▍ | 5481/6470 [1:03:00<11:05,  1.49it/s]
3|5482|Loss: 1.0217061042785645:  85%|████████▍ | 5482/6470 [1:03:00<11:06,  1.48it/s]
3|5483|Loss: 0.9119757413864136:  85%|████████▍ | 5482/6470 [1:03:00<11:06,  1.48it/s]
3|5483|Loss: 0.9119757413864136:  85%|████████▍ | 5483/6470 [1:03:01<10:43,  1.53it/s]
3|5484|Loss: 0.8420168161392212:  85%|████████▍ | 5483/6470 [1:03:01<10:43,  1.53it/s]
3|5484|Loss: 0.8420168161392212:  85%|████████▍ | 5484/6470 [1:03:01<10:31,  1.56it/s]
3|5485|Loss: 0.7895998954772949:  85%|████████▍ | 5484/6470 [1:03:02<10:31,  1.56it/s]
3|5485|Loss: 0.7895998954772949:  85%|████████▍ | 5485/6470 [1:03:02<10:50,  1.51it/s]
3|5486|Loss: 0.9391005635261536:  85%|████████▍ | 5485/6470 [1:03:02<10:50,  1.51it/s]
3|5486|Loss: 0.9391005635261536:  85%|████████▍ | 5486/6470 [1:03:03<11:03,  1.48it/s]
3|5487|Loss: 0.8700685501098633:  85%|████████▍ | 5486/6470 [1:03:03<11:03,  1.48it/s]
3|5487|Loss: 0.8700685501098633:  85%|████████▍ | 5487/6470 [1:03:03<11:19,  1.45it/s]
3|5488|Loss: 0.7648978233337402:  85%|████████▍ | 5487/6470 [1:03:04<11:19,  1.45it/s]
3|5488|Loss: 0.7648978233337402:  85%|████████▍ | 5488/6470 [1:03:04<10:42,  1.53it/s]
3|5489|Loss: 0.9236180186271667:  85%|████████▍ | 5488/6470 [1:03:04<10:42,  1.53it/s]
3|5489|Loss: 0.9236180186271667:  85%|████████▍ | 5489/6470 [1:03:05<10:42,  1.53it/s]
3|5490|Loss: 0.653940737247467:  85%|████████▍ | 5489/6470 [1:03:05<10:42,  1.53it/s] 
3|5490|Loss: 0.653940737247467:  85%|████████▍ | 5490/6470 [1:03:05<10:56,  1.49it/s]
3|5491|Loss: 0.7049970030784607:  85%|████████▍ | 5490/6470 [1:03:06<10:56,  1.49it/s]
3|5491|Loss: 0.7049970030784607:  85%|████████▍ | 5491/6470 [1:03:06<09:33,  1.71it/s]
3|5492|Loss: 0.8500523567199707:  85%|████████▍ | 5491/6470 [1:03:06<09:33,  1.71it/s]
3|5492|Loss: 0.8500523567199707:  85%|████████▍ | 5492/6470 [1:03:06<09:28,  1.72it/s]
3|5493|Loss: 0.9155876636505127:  85%|████████▍ | 5492/6470 [1:03:07<09:28,  1.72it/s]
3|5493|Loss: 0.9155876636505127:  85%|████████▍ | 5493/6470 [1:03:07<09:40,  1.68it/s]
3|5494|Loss: 0.711345911026001:  85%|████████▍ | 5493/6470 [1:03:07<09:40,  1.68it/s] 
3|5494|Loss: 0.711345911026001:  85%|████████▍ | 5494/6470 [1:03:08<10:13,  1.59it/s]
3|5495|Loss: 0.881670355796814:  85%|████████▍ | 5494/6470 [1:03:08<10:13,  1.59it/s]
3|5495|Loss: 0.881670355796814:  85%|████████▍ | 5495/6470 [1:03:08<10:11,  1.60it/s]
3|5496|Loss: 0.7779015898704529:  85%|████████▍ | 5495/6470 [1:03:09<10:11,  1.60it/s]
3|5496|Loss: 0.7779015898704529:  85%|████████▍ | 5496/6470 [1:03:09<10:05,  1.61it/s]
3|5497|Loss: 0.9111294746398926:  85%|████████▍ | 5496/6470 [1:03:09<10:05,  1.61it/s]
3|5497|Loss: 0.9111294746398926:  85%|████████▍ | 5497/6470 [1:03:10<10:34,  1.53it/s]
3|5498|Loss: 0.9072776436805725:  85%|████████▍ | 5497/6470 [1:03:10<10:34,  1.53it/s]
3|5498|Loss: 0.9072776436805725:  85%|████████▍ | 5498/6470 [1:03:10<10:58,  1.48it/s]
3|5499|Loss: 0.7955352067947388:  85%|████████▍ | 5498/6470 [1:03:11<10:58,  1.48it/s]
3|5499|Loss: 0.7955352067947388:  85%|████████▍ | 5499/6470 [1:03:11<10:16,  1.58it/s]
3|5500|Loss: 0.9614484310150146:  85%|████████▍ | 5499/6470 [1:03:11<10:16,  1.58it/s]
3|5500|Loss: 0.9614484310150146:  85%|████████▌ | 5500/6470 [1:03:12<10:52,  1.49it/s]
3|5501|Loss: 0.9012725353240967:  85%|████████▌ | 5500/6470 [1:03:12<10:52,  1.49it/s]
3|5501|Loss: 0.9012725353240967:  85%|████████▌ | 5501/6470 [1:03:12<11:21,  1.42it/s]
3|5502|Loss: 0.8643653392791748:  85%|████████▌ | 5501/6470 [1:03:13<11:21,  1.42it/s]
3|5502|Loss: 0.8643653392791748:  85%|████████▌ | 5502/6470 [1:03:13<10:57,  1.47it/s]
3|5503|Loss: 0.9061080813407898:  85%|████████▌ | 5502/6470 [1:03:13<10:57,  1.47it/s]
3|5503|Loss: 0.9061080813407898:  85%|████████▌ | 5503/6470 [1:03:14<10:54,  1.48it/s]
3|5504|Loss: 0.9066556096076965:  85%|████████▌ | 5503/6470 [1:03:14<10:54,  1.48it/s]
3|5504|Loss: 0.9066556096076965:  85%|████████▌ | 5504/6470 [1:03:14<10:42,  1.50it/s]
3|5505|Loss: 1.0495604276657104:  85%|████████▌ | 5504/6470 [1:03:15<10:42,  1.50it/s]
3|5505|Loss: 1.0495604276657104:  85%|████████▌ | 5505/6470 [1:03:15<10:04,  1.60it/s]
3|5506|Loss: 0.7560784816741943:  85%|████████▌ | 5505/6470 [1:03:15<10:04,  1.60it/s]
3|5506|Loss: 0.7560784816741943:  85%|████████▌ | 5506/6470 [1:03:16<10:21,  1.55it/s]
3|5507|Loss: 0.9718837738037109:  85%|████████▌ | 5506/6470 [1:03:16<10:21,  1.55it/s]
3|5507|Loss: 0.9718837738037109:  85%|████████▌ | 5507/6470 [1:03:16<11:15,  1.42it/s]
3|5508|Loss: 0.9309824705123901:  85%|████████▌ | 5507/6470 [1:03:17<11:15,  1.42it/s]
3|5508|Loss: 0.9309824705123901:  85%|████████▌ | 5508/6470 [1:03:17<11:49,  1.36it/s]
3|5509|Loss: 0.8554575443267822:  85%|████████▌ | 5508/6470 [1:03:18<11:49,  1.36it/s]
3|5509|Loss: 0.8554575443267822:  85%|████████▌ | 5509/6470 [1:03:18<11:30,  1.39it/s]
3|5510|Loss: 0.7099961042404175:  85%|████████▌ | 5509/6470 [1:03:18<11:30,  1.39it/s]
3|5510|Loss: 0.7099961042404175:  85%|████████▌ | 5510/6470 [1:03:19<11:26,  1.40it/s]
3|5511|Loss: 0.8467288613319397:  85%|████████▌ | 5510/6470 [1:03:19<11:26,  1.40it/s]
3|5511|Loss: 0.8467288613319397:  85%|████████▌ | 5511/6470 [1:03:19<11:47,  1.36it/s]
3|5512|Loss: 0.9642776846885681:  85%|████████▌ | 5511/6470 [1:03:20<11:47,  1.36it/s]
3|5512|Loss: 0.9642776846885681:  85%|████████▌ | 5512/6470 [1:03:20<11:30,  1.39it/s]
3|5513|Loss: 0.749759316444397:  85%|████████▌ | 5512/6470 [1:03:21<11:30,  1.39it/s] 
3|5513|Loss: 0.749759316444397:  85%|████████▌ | 5513/6470 [1:03:21<11:45,  1.36it/s]
3|5514|Loss: 1.0281496047973633:  85%|████████▌ | 5513/6470 [1:03:21<11:45,  1.36it/s]
3|5514|Loss: 1.0281496047973633:  85%|████████▌ | 5514/6470 [1:03:22<11:52,  1.34it/s]
3|5515|Loss: 0.8026512861251831:  85%|████████▌ | 5514/6470 [1:03:22<11:52,  1.34it/s]
3|5515|Loss: 0.8026512861251831:  85%|████████▌ | 5515/6470 [1:03:22<11:23,  1.40it/s]
3|5516|Loss: 0.941861093044281:  85%|████████▌ | 5515/6470 [1:03:23<11:23,  1.40it/s] 
3|5516|Loss: 0.941861093044281:  85%|████████▌ | 5516/6470 [1:03:23<10:31,  1.51it/s]
3|5517|Loss: 0.8203740119934082:  85%|████████▌ | 5516/6470 [1:03:23<10:31,  1.51it/s]
3|5517|Loss: 0.8203740119934082:  85%|████████▌ | 5517/6470 [1:03:23<10:12,  1.56it/s]
3|5518|Loss: 0.8184123635292053:  85%|████████▌ | 5517/6470 [1:03:24<10:12,  1.56it/s]
3|5518|Loss: 0.8184123635292053:  85%|████████▌ | 5518/6470 [1:03:24<10:24,  1.53it/s]
3|5519|Loss: 0.7896606922149658:  85%|████████▌ | 5518/6470 [1:03:24<10:24,  1.53it/s]
3|5519|Loss: 0.7896606922149658:  85%|████████▌ | 5519/6470 [1:03:25<09:55,  1.60it/s]
3|5520|Loss: 1.0012171268463135:  85%|████████▌ | 5519/6470 [1:03:25<09:55,  1.60it/s]
3|5520|Loss: 1.0012171268463135:  85%|████████▌ | 5520/6470 [1:03:25<09:53,  1.60it/s]
3|5521|Loss: 0.784772515296936:  85%|████████▌ | 5520/6470 [1:03:26<09:53,  1.60it/s] 
3|5521|Loss: 0.784772515296936:  85%|████████▌ | 5521/6470 [1:03:26<10:49,  1.46it/s]
3|5522|Loss: 1.021338939666748:  85%|████████▌ | 5521/6470 [1:03:27<10:49,  1.46it/s]
3|5522|Loss: 1.021338939666748:  85%|████████▌ | 5522/6470 [1:03:27<10:35,  1.49it/s]
3|5523|Loss: 0.8289474844932556:  85%|████████▌ | 5522/6470 [1:03:27<10:35,  1.49it/s]
3|5523|Loss: 0.8289474844932556:  85%|████████▌ | 5523/6470 [1:03:27<10:38,  1.48it/s]
3|5524|Loss: 1.0283254384994507:  85%|████████▌ | 5523/6470 [1:03:28<10:38,  1.48it/s]
3|5524|Loss: 1.0283254384994507:  85%|████████▌ | 5524/6470 [1:03:28<10:28,  1.51it/s]
3|5525|Loss: 0.8030485510826111:  85%|████████▌ | 5524/6470 [1:03:28<10:28,  1.51it/s]
3|5525|Loss: 0.8030485510826111:  85%|████████▌ | 5525/6470 [1:03:29<09:54,  1.59it/s]
3|5526|Loss: 1.0530894994735718:  85%|████████▌ | 5525/6470 [1:03:29<09:54,  1.59it/s]
3|5526|Loss: 1.0530894994735718:  85%|████████▌ | 5526/6470 [1:03:29<09:15,  1.70it/s]
3|5527|Loss: 0.9131104350090027:  85%|████████▌ | 5526/6470 [1:03:29<09:15,  1.70it/s]
3|5527|Loss: 0.9131104350090027:  85%|████████▌ | 5527/6470 [1:03:30<09:19,  1.68it/s]
3|5528|Loss: 0.9564473032951355:  85%|████████▌ | 5527/6470 [1:03:30<09:19,  1.68it/s]
3|5528|Loss: 0.9564473032951355:  85%|████████▌ | 5528/6470 [1:03:30<09:33,  1.64it/s]
3|5529|Loss: 0.923740804195404:  85%|████████▌ | 5528/6470 [1:03:31<09:33,  1.64it/s] 
3|5529|Loss: 0.923740804195404:  85%|████████▌ | 5529/6470 [1:03:31<10:24,  1.51it/s]
3|5530|Loss: 0.8557699918746948:  85%|████████▌ | 5529/6470 [1:03:32<10:24,  1.51it/s]
3|5530|Loss: 0.8557699918746948:  85%|████████▌ | 5530/6470 [1:03:32<10:08,  1.55it/s]
3|5531|Loss: 0.7810716032981873:  85%|████████▌ | 5530/6470 [1:03:32<10:08,  1.55it/s]
3|5531|Loss: 0.7810716032981873:  85%|████████▌ | 5531/6470 [1:03:32<09:50,  1.59it/s]
3|5532|Loss: 0.8079701066017151:  85%|████████▌ | 5531/6470 [1:03:33<09:50,  1.59it/s]
3|5532|Loss: 0.8079701066017151:  86%|████████▌ | 5532/6470 [1:03:33<10:11,  1.53it/s]
3|5533|Loss: 0.7915214896202087:  86%|████████▌ | 5532/6470 [1:03:33<10:11,  1.53it/s]
3|5533|Loss: 0.7915214896202087:  86%|████████▌ | 5533/6470 [1:03:34<10:38,  1.47it/s]
3|5534|Loss: 0.818157434463501:  86%|████████▌ | 5533/6470 [1:03:34<10:38,  1.47it/s] 
3|5534|Loss: 0.818157434463501:  86%|████████▌ | 5534/6470 [1:03:35<10:54,  1.43it/s]
3|5535|Loss: 0.7803443074226379:  86%|████████▌ | 5534/6470 [1:03:35<10:54,  1.43it/s]
3|5535|Loss: 0.7803443074226379:  86%|████████▌ | 5535/6470 [1:03:35<11:06,  1.40it/s]
3|5536|Loss: 0.9629547595977783:  86%|████████▌ | 5535/6470 [1:03:36<11:06,  1.40it/s]
3|5536|Loss: 0.9629547595977783:  86%|████████▌ | 5536/6470 [1:03:36<11:07,  1.40it/s]
3|5537|Loss: 1.0024405717849731:  86%|████████▌ | 5536/6470 [1:03:36<11:07,  1.40it/s]
3|5537|Loss: 1.0024405717849731:  86%|████████▌ | 5537/6470 [1:03:37<11:13,  1.39it/s]
3|5538|Loss: 0.7014933824539185:  86%|████████▌ | 5537/6470 [1:03:37<11:13,  1.39it/s]
3|5538|Loss: 0.7014933824539185:  86%|████████▌ | 5538/6470 [1:03:38<11:43,  1.33it/s]
3|5539|Loss: 0.8198633790016174:  86%|████████▌ | 5538/6470 [1:03:38<11:43,  1.33it/s]
3|5539|Loss: 0.8198633790016174:  86%|████████▌ | 5539/6470 [1:03:38<10:54,  1.42it/s]
3|5540|Loss: 0.8345357179641724:  86%|████████▌ | 5539/6470 [1:03:39<10:54,  1.42it/s]
3|5540|Loss: 0.8345357179641724:  86%|████████▌ | 5540/6470 [1:03:39<10:42,  1.45it/s]
3|5541|Loss: 0.7189547419548035:  86%|████████▌ | 5540/6470 [1:03:39<10:42,  1.45it/s]
3|5541|Loss: 0.7189547419548035:  86%|████████▌ | 5541/6470 [1:03:40<11:18,  1.37it/s]
3|5542|Loss: 0.848334014415741:  86%|████████▌ | 5541/6470 [1:03:40<11:18,  1.37it/s] 
3|5542|Loss: 0.848334014415741:  86%|████████▌ | 5542/6470 [1:03:41<11:59,  1.29it/s]
3|5543|Loss: 0.8790061473846436:  86%|████████▌ | 5542/6470 [1:03:41<11:59,  1.29it/s]
3|5543|Loss: 0.8790061473846436:  86%|████████▌ | 5543/6470 [1:03:41<11:32,  1.34it/s]
3|5544|Loss: 0.9157029390335083:  86%|████████▌ | 5543/6470 [1:03:42<11:32,  1.34it/s]
3|5544|Loss: 0.9157029390335083:  86%|████████▌ | 5544/6470 [1:03:42<11:01,  1.40it/s]
3|5545|Loss: 0.7589778900146484:  86%|████████▌ | 5544/6470 [1:03:42<11:01,  1.40it/s]
3|5545|Loss: 0.7589778900146484:  86%|████████▌ | 5545/6470 [1:03:43<10:54,  1.41it/s]
3|5546|Loss: 0.7478795647621155:  86%|████████▌ | 5545/6470 [1:03:43<10:54,  1.41it/s]
3|5546|Loss: 0.7478795647621155:  86%|████████▌ | 5546/6470 [1:03:43<10:34,  1.46it/s]
3|5547|Loss: 0.86927330493927:  86%|████████▌ | 5546/6470 [1:03:44<10:34,  1.46it/s]  
3|5547|Loss: 0.86927330493927:  86%|████████▌ | 5547/6470 [1:03:44<10:39,  1.44it/s]
3|5548|Loss: 0.8541109561920166:  86%|████████▌ | 5547/6470 [1:03:44<10:39,  1.44it/s]
3|5548|Loss: 0.8541109561920166:  86%|████████▌ | 5548/6470 [1:03:45<11:24,  1.35it/s]
3|5549|Loss: 0.825356662273407:  86%|████████▌ | 5548/6470 [1:03:45<11:24,  1.35it/s] 
3|5549|Loss: 0.825356662273407:  86%|████████▌ | 5549/6470 [1:03:45<10:37,  1.45it/s]
3|5550|Loss: 0.7975701689720154:  86%|████████▌ | 5549/6470 [1:03:46<10:37,  1.45it/s]
3|5550|Loss: 0.7975701689720154:  86%|████████▌ | 5550/6470 [1:03:46<11:11,  1.37it/s]
3|5551|Loss: 0.652718722820282:  86%|████████▌ | 5550/6470 [1:03:47<11:11,  1.37it/s] 
3|5551|Loss: 0.652718722820282:  86%|████████▌ | 5551/6470 [1:03:47<10:57,  1.40it/s]
3|5552|Loss: 1.0317869186401367:  86%|████████▌ | 5551/6470 [1:03:47<10:57,  1.40it/s]
3|5552|Loss: 1.0317869186401367:  86%|████████▌ | 5552/6470 [1:03:48<10:59,  1.39it/s]
3|5553|Loss: 0.8706185817718506:  86%|████████▌ | 5552/6470 [1:03:48<10:59,  1.39it/s]
3|5553|Loss: 0.8706185817718506:  86%|████████▌ | 5553/6470 [1:03:48<11:30,  1.33it/s]
3|5554|Loss: 0.8469536900520325:  86%|████████▌ | 5553/6470 [1:03:49<11:30,  1.33it/s]
3|5554|Loss: 0.8469536900520325:  86%|████████▌ | 5554/6470 [1:03:49<11:25,  1.34it/s]
3|5555|Loss: 0.886311411857605:  86%|████████▌ | 5554/6470 [1:03:50<11:25,  1.34it/s] 
3|5555|Loss: 0.886311411857605:  86%|████████▌ | 5555/6470 [1:03:50<11:13,  1.36it/s]
3|5556|Loss: 0.9801842570304871:  86%|████████▌ | 5555/6470 [1:03:50<11:13,  1.36it/s]
3|5556|Loss: 0.9801842570304871:  86%|████████▌ | 5556/6470 [1:03:50<10:09,  1.50it/s]
3|5557|Loss: 0.9337924122810364:  86%|████████▌ | 5556/6470 [1:03:51<10:09,  1.50it/s]
3|5557|Loss: 0.9337924122810364:  86%|████████▌ | 5557/6470 [1:03:51<09:35,  1.59it/s]
3|5558|Loss: 0.9652916193008423:  86%|████████▌ | 5557/6470 [1:03:51<09:35,  1.59it/s]
3|5558|Loss: 0.9652916193008423:  86%|████████▌ | 5558/6470 [1:03:51<09:15,  1.64it/s]
3|5559|Loss: 0.8583033680915833:  86%|████████▌ | 5558/6470 [1:03:52<09:15,  1.64it/s]
3|5559|Loss: 0.8583033680915833:  86%|████████▌ | 5559/6470 [1:03:52<09:14,  1.64it/s]
3|5560|Loss: 0.8392654657363892:  86%|████████▌ | 5559/6470 [1:03:52<09:14,  1.64it/s]
3|5560|Loss: 0.8392654657363892:  86%|████████▌ | 5560/6470 [1:03:53<09:02,  1.68it/s]
3|5561|Loss: 0.9451612830162048:  86%|████████▌ | 5560/6470 [1:03:53<09:02,  1.68it/s]
3|5561|Loss: 0.9451612830162048:  86%|████████▌ | 5561/6470 [1:03:53<09:30,  1.59it/s]
3|5562|Loss: 0.7267077565193176:  86%|████████▌ | 5561/6470 [1:03:54<09:30,  1.59it/s]
3|5562|Loss: 0.7267077565193176:  86%|████████▌ | 5562/6470 [1:03:54<10:29,  1.44it/s]
3|5563|Loss: 0.8057140707969666:  86%|████████▌ | 5562/6470 [1:03:55<10:29,  1.44it/s]
3|5563|Loss: 0.8057140707969666:  86%|████████▌ | 5563/6470 [1:03:55<10:28,  1.44it/s]
3|5564|Loss: 0.8227859139442444:  86%|████████▌ | 5563/6470 [1:03:55<10:28,  1.44it/s]
3|5564|Loss: 0.8227859139442444:  86%|████████▌ | 5564/6470 [1:03:56<10:34,  1.43it/s]
3|5565|Loss: 0.790158748626709:  86%|████████▌ | 5564/6470 [1:03:56<10:34,  1.43it/s] 
3|5565|Loss: 0.790158748626709:  86%|████████▌ | 5565/6470 [1:03:56<10:36,  1.42it/s]
3|5566|Loss: 0.9816008806228638:  86%|████████▌ | 5565/6470 [1:03:57<10:36,  1.42it/s]
3|5566|Loss: 0.9816008806228638:  86%|████████▌ | 5566/6470 [1:03:57<10:43,  1.41it/s]
3|5567|Loss: 0.8219060301780701:  86%|████████▌ | 5566/6470 [1:03:57<10:43,  1.41it/s]
3|5567|Loss: 0.8219060301780701:  86%|████████▌ | 5567/6470 [1:03:58<11:27,  1.31it/s]
3|5568|Loss: 0.78844153881073:  86%|████████▌ | 5567/6470 [1:03:58<11:27,  1.31it/s]  
3|5568|Loss: 0.78844153881073:  86%|████████▌ | 5568/6470 [1:03:59<11:05,  1.35it/s]
3|5569|Loss: 0.7018455266952515:  86%|████████▌ | 5568/6470 [1:03:59<11:05,  1.35it/s]
3|5569|Loss: 0.7018455266952515:  86%|████████▌ | 5569/6470 [1:03:59<10:57,  1.37it/s]
3|5570|Loss: 0.6709198355674744:  86%|████████▌ | 5569/6470 [1:04:00<10:57,  1.37it/s]
3|5570|Loss: 0.6709198355674744:  86%|████████▌ | 5570/6470 [1:04:00<10:52,  1.38it/s]
3|5571|Loss: 0.9153985381126404:  86%|████████▌ | 5570/6470 [1:04:00<10:52,  1.38it/s]
3|5571|Loss: 0.9153985381126404:  86%|████████▌ | 5571/6470 [1:04:01<10:54,  1.37it/s]
3|5572|Loss: 0.8840759992599487:  86%|████████▌ | 5571/6470 [1:04:01<10:54,  1.37it/s]
3|5572|Loss: 0.8840759992599487:  86%|████████▌ | 5572/6470 [1:04:01<10:37,  1.41it/s]
3|5573|Loss: 0.8310820460319519:  86%|████████▌ | 5572/6470 [1:04:02<10:37,  1.41it/s]
3|5573|Loss: 0.8310820460319519:  86%|████████▌ | 5573/6470 [1:04:02<10:33,  1.42it/s]
3|5574|Loss: 0.861582338809967:  86%|████████▌ | 5573/6470 [1:04:02<10:33,  1.42it/s] 
3|5574|Loss: 0.861582338809967:  86%|████████▌ | 5574/6470 [1:04:03<10:39,  1.40it/s]
3|5575|Loss: 0.9868171215057373:  86%|████████▌ | 5574/6470 [1:04:03<10:39,  1.40it/s]
3|5575|Loss: 0.9868171215057373:  86%|████████▌ | 5575/6470 [1:04:04<11:03,  1.35it/s]
3|5576|Loss: 0.857292115688324:  86%|████████▌ | 5575/6470 [1:04:04<11:03,  1.35it/s] 
3|5576|Loss: 0.857292115688324:  86%|████████▌ | 5576/6470 [1:04:04<11:05,  1.34it/s]
3|5577|Loss: 0.9784823060035706:  86%|████████▌ | 5576/6470 [1:04:05<11:05,  1.34it/s]
3|5577|Loss: 0.9784823060035706:  86%|████████▌ | 5577/6470 [1:04:05<11:14,  1.32it/s]
3|5578|Loss: 0.9130024313926697:  86%|████████▌ | 5577/6470 [1:04:06<11:14,  1.32it/s]
3|5578|Loss: 0.9130024313926697:  86%|████████▌ | 5578/6470 [1:04:06<11:12,  1.33it/s]
3|5579|Loss: 0.7978404760360718:  86%|████████▌ | 5578/6470 [1:04:06<11:12,  1.33it/s]
3|5579|Loss: 0.7978404760360718:  86%|████████▌ | 5579/6470 [1:04:07<10:33,  1.41it/s]
3|5580|Loss: 0.7749318480491638:  86%|████████▌ | 5579/6470 [1:04:07<10:33,  1.41it/s]
3|5580|Loss: 0.7749318480491638:  86%|████████▌ | 5580/6470 [1:04:07<10:42,  1.38it/s]
3|5581|Loss: 0.9386854767799377:  86%|████████▌ | 5580/6470 [1:04:08<10:42,  1.38it/s]
3|5581|Loss: 0.9386854767799377:  86%|████████▋ | 5581/6470 [1:04:08<10:16,  1.44it/s]
3|5582|Loss: 0.7585675120353699:  86%|████████▋ | 5581/6470 [1:04:08<10:16,  1.44it/s]
3|5582|Loss: 0.7585675120353699:  86%|████████▋ | 5582/6470 [1:04:09<10:19,  1.43it/s]
3|5583|Loss: 0.8654954433441162:  86%|████████▋ | 5582/6470 [1:04:09<10:19,  1.43it/s]
3|5583|Loss: 0.8654954433441162:  86%|████████▋ | 5583/6470 [1:04:09<10:24,  1.42it/s]
3|5584|Loss: 0.7764395475387573:  86%|████████▋ | 5583/6470 [1:04:10<10:24,  1.42it/s]
3|5584|Loss: 0.7764395475387573:  86%|████████▋ | 5584/6470 [1:04:10<10:29,  1.41it/s]
3|5585|Loss: 0.8532970547676086:  86%|████████▋ | 5584/6470 [1:04:10<10:29,  1.41it/s]
3|5585|Loss: 0.8532970547676086:  86%|████████▋ | 5585/6470 [1:04:11<10:34,  1.40it/s]
3|5586|Loss: 0.8762422204017639:  86%|████████▋ | 5585/6470 [1:04:11<10:34,  1.40it/s]
3|5586|Loss: 0.8762422204017639:  86%|████████▋ | 5586/6470 [1:04:12<10:39,  1.38it/s]
3|5587|Loss: 0.9117912650108337:  86%|████████▋ | 5586/6470 [1:04:12<10:39,  1.38it/s]
3|5587|Loss: 0.9117912650108337:  86%|████████▋ | 5587/6470 [1:04:12<10:19,  1.43it/s]
3|5588|Loss: 0.727790117263794:  86%|████████▋ | 5587/6470 [1:04:13<10:19,  1.43it/s] 
3|5588|Loss: 0.727790117263794:  86%|████████▋ | 5588/6470 [1:04:13<10:14,  1.44it/s]
3|5589|Loss: 0.7411515712738037:  86%|████████▋ | 5588/6470 [1:04:13<10:14,  1.44it/s]
3|5589|Loss: 0.7411515712738037:  86%|████████▋ | 5589/6470 [1:04:14<10:03,  1.46it/s]
3|5590|Loss: 0.8134216070175171:  86%|████████▋ | 5589/6470 [1:04:14<10:03,  1.46it/s]
3|5590|Loss: 0.8134216070175171:  86%|████████▋ | 5590/6470 [1:04:14<10:14,  1.43it/s]
3|5591|Loss: 0.7865297198295593:  86%|████████▋ | 5590/6470 [1:04:15<10:14,  1.43it/s]
3|5591|Loss: 0.7865297198295593:  86%|████████▋ | 5591/6470 [1:04:15<10:46,  1.36it/s]
3|5592|Loss: 0.9595744609832764:  86%|████████▋ | 5591/6470 [1:04:16<10:46,  1.36it/s]
3|5592|Loss: 0.9595744609832764:  86%|████████▋ | 5592/6470 [1:04:16<10:56,  1.34it/s]
3|5593|Loss: 0.8201380372047424:  86%|████████▋ | 5592/6470 [1:04:16<10:56,  1.34it/s]
3|5593|Loss: 0.8201380372047424:  86%|████████▋ | 5593/6470 [1:04:17<10:31,  1.39it/s]
3|5594|Loss: 0.9437952041625977:  86%|████████▋ | 5593/6470 [1:04:17<10:31,  1.39it/s]
3|5594|Loss: 0.9437952041625977:  86%|████████▋ | 5594/6470 [1:04:17<10:29,  1.39it/s]
3|5595|Loss: 0.9407263994216919:  86%|████████▋ | 5594/6470 [1:04:18<10:29,  1.39it/s]
3|5595|Loss: 0.9407263994216919:  86%|████████▋ | 5595/6470 [1:04:18<10:45,  1.35it/s]
3|5596|Loss: 0.8776676058769226:  86%|████████▋ | 5595/6470 [1:04:18<10:45,  1.35it/s]
3|5596|Loss: 0.8776676058769226:  86%|████████▋ | 5596/6470 [1:04:19<10:29,  1.39it/s]
3|5597|Loss: 0.8480654358863831:  86%|████████▋ | 5596/6470 [1:04:19<10:29,  1.39it/s]
3|5597|Loss: 0.8480654358863831:  87%|████████▋ | 5597/6470 [1:04:19<09:55,  1.47it/s]
3|5598|Loss: 0.5901679396629333:  87%|████████▋ | 5597/6470 [1:04:20<09:55,  1.47it/s]
3|5598|Loss: 0.5901679396629333:  87%|████████▋ | 5598/6470 [1:04:20<10:30,  1.38it/s]
3|5599|Loss: 0.9162421822547913:  87%|████████▋ | 5598/6470 [1:04:21<10:30,  1.38it/s]
3|5599|Loss: 0.9162421822547913:  87%|████████▋ | 5599/6470 [1:04:21<10:21,  1.40it/s]
3|5600|Loss: 0.9790744185447693:  87%|████████▋ | 5599/6470 [1:04:21<10:21,  1.40it/s]
3|5600|Loss: 0.9790744185447693:  87%|████████▋ | 5600/6470 [1:04:22<10:51,  1.34it/s]
3|5601|Loss: 0.9312111139297485:  87%|████████▋ | 5600/6470 [1:04:22<10:51,  1.34it/s]
3|5601|Loss: 0.9312111139297485:  87%|████████▋ | 5601/6470 [1:04:22<10:56,  1.32it/s]
3|5602|Loss: 0.7786903977394104:  87%|████████▋ | 5601/6470 [1:04:23<10:56,  1.32it/s]
3|5602|Loss: 0.7786903977394104:  87%|████████▋ | 5602/6470 [1:04:23<10:44,  1.35it/s]
3|5603|Loss: 0.8205520510673523:  87%|████████▋ | 5602/6470 [1:04:23<10:44,  1.35it/s]
3|5603|Loss: 0.8205520510673523:  87%|████████▋ | 5603/6470 [1:04:24<10:00,  1.44it/s]
3|5604|Loss: 0.8394652605056763:  87%|████████▋ | 5603/6470 [1:04:24<10:00,  1.44it/s]
3|5604|Loss: 0.8394652605056763:  87%|████████▋ | 5604/6470 [1:04:24<10:00,  1.44it/s]
3|5605|Loss: 0.9315105080604553:  87%|████████▋ | 5604/6470 [1:04:25<10:00,  1.44it/s]
3|5605|Loss: 0.9315105080604553:  87%|████████▋ | 5605/6470 [1:04:25<09:52,  1.46it/s]
3|5606|Loss: 0.9742752313613892:  87%|████████▋ | 5605/6470 [1:04:25<09:52,  1.46it/s]
3|5606|Loss: 0.9742752313613892:  87%|████████▋ | 5606/6470 [1:04:26<10:01,  1.44it/s]
3|5607|Loss: 0.8421295285224915:  87%|████████▋ | 5606/6470 [1:04:26<10:01,  1.44it/s]
3|5607|Loss: 0.8421295285224915:  87%|████████▋ | 5607/6470 [1:04:26<09:47,  1.47it/s]
3|5608|Loss: 0.8534747958183289:  87%|████████▋ | 5607/6470 [1:04:27<09:47,  1.47it/s]
3|5608|Loss: 0.8534747958183289:  87%|████████▋ | 5608/6470 [1:04:27<10:13,  1.40it/s]
3|5609|Loss: 0.9161173105239868:  87%|████████▋ | 5608/6470 [1:04:28<10:13,  1.40it/s]
3|5609|Loss: 0.9161173105239868:  87%|████████▋ | 5609/6470 [1:04:28<10:24,  1.38it/s]
3|5610|Loss: 0.8076371550559998:  87%|████████▋ | 5609/6470 [1:04:28<10:24,  1.38it/s]
3|5610|Loss: 0.8076371550559998:  87%|████████▋ | 5610/6470 [1:04:29<10:29,  1.37it/s]
3|5611|Loss: 0.7931740283966064:  87%|████████▋ | 5610/6470 [1:04:29<10:29,  1.37it/s]
3|5611|Loss: 0.7931740283966064:  87%|████████▋ | 5611/6470 [1:04:29<10:34,  1.35it/s]
3|5612|Loss: 0.9061524868011475:  87%|████████▋ | 5611/6470 [1:04:30<10:34,  1.35it/s]
3|5612|Loss: 0.9061524868011475:  87%|████████▋ | 5612/6470 [1:04:30<10:46,  1.33it/s]
3|5613|Loss: 0.8306801915168762:  87%|████████▋ | 5612/6470 [1:04:31<10:46,  1.33it/s]
3|5613|Loss: 0.8306801915168762:  87%|████████▋ | 5613/6470 [1:04:31<11:05,  1.29it/s]
3|5614|Loss: 0.8186039924621582:  87%|████████▋ | 5613/6470 [1:04:31<11:05,  1.29it/s]
3|5614|Loss: 0.8186039924621582:  87%|████████▋ | 5614/6470 [1:04:32<10:25,  1.37it/s]
3|5615|Loss: 0.8279553651809692:  87%|████████▋ | 5614/6470 [1:04:32<10:25,  1.37it/s]
3|5615|Loss: 0.8279553651809692:  87%|████████▋ | 5615/6470 [1:04:32<10:16,  1.39it/s]
3|5616|Loss: 0.7943663597106934:  87%|████████▋ | 5615/6470 [1:04:33<10:16,  1.39it/s]
3|5616|Loss: 0.7943663597106934:  87%|████████▋ | 5616/6470 [1:04:33<09:45,  1.46it/s]
3|5617|Loss: 0.8274350166320801:  87%|████████▋ | 5616/6470 [1:04:33<09:45,  1.46it/s]
3|5617|Loss: 0.8274350166320801:  87%|████████▋ | 5617/6470 [1:04:33<08:57,  1.59it/s]
3|5618|Loss: 0.9705597162246704:  87%|████████▋ | 5617/6470 [1:04:34<08:57,  1.59it/s]
3|5618|Loss: 0.9705597162246704:  87%|████████▋ | 5618/6470 [1:04:34<09:34,  1.48it/s]
3|5619|Loss: 0.8321879506111145:  87%|████████▋ | 5618/6470 [1:04:35<09:34,  1.48it/s]
3|5619|Loss: 0.8321879506111145:  87%|████████▋ | 5619/6470 [1:04:35<09:37,  1.47it/s]
3|5620|Loss: 0.8152641654014587:  87%|████████▋ | 5619/6470 [1:04:35<09:37,  1.47it/s]
3|5620|Loss: 0.8152641654014587:  87%|████████▋ | 5620/6470 [1:04:36<10:04,  1.41it/s]
3|5621|Loss: 0.9124243259429932:  87%|████████▋ | 5620/6470 [1:04:36<10:04,  1.41it/s]
3|5621|Loss: 0.9124243259429932:  87%|████████▋ | 5621/6470 [1:04:36<10:07,  1.40it/s]
3|5622|Loss: 0.9339120388031006:  87%|████████▋ | 5621/6470 [1:04:37<10:07,  1.40it/s]
3|5622|Loss: 0.9339120388031006:  87%|████████▋ | 5622/6470 [1:04:37<09:52,  1.43it/s]
3|5623|Loss: 0.9674290418624878:  87%|████████▋ | 5622/6470 [1:04:37<09:52,  1.43it/s]
3|5623|Loss: 0.9674290418624878:  87%|████████▋ | 5623/6470 [1:04:38<09:38,  1.47it/s]
3|5624|Loss: 0.8576710820198059:  87%|████████▋ | 5623/6470 [1:04:38<09:38,  1.47it/s]
3|5624|Loss: 0.8576710820198059:  87%|████████▋ | 5624/6470 [1:04:38<09:27,  1.49it/s]
3|5625|Loss: 0.9111303091049194:  87%|████████▋ | 5624/6470 [1:04:39<09:27,  1.49it/s]
3|5625|Loss: 0.9111303091049194:  87%|████████▋ | 5625/6470 [1:04:39<09:19,  1.51it/s]
3|5626|Loss: 0.7839757204055786:  87%|████████▋ | 5625/6470 [1:04:39<09:19,  1.51it/s]
3|5626|Loss: 0.7839757204055786:  87%|████████▋ | 5626/6470 [1:04:40<09:34,  1.47it/s]
3|5627|Loss: 0.9670740365982056:  87%|████████▋ | 5626/6470 [1:04:40<09:34,  1.47it/s]
3|5627|Loss: 0.9670740365982056:  87%|████████▋ | 5627/6470 [1:04:41<10:01,  1.40it/s]
3|5628|Loss: 0.6272996068000793:  87%|████████▋ | 5627/6470 [1:04:41<10:01,  1.40it/s]
3|5628|Loss: 0.6272996068000793:  87%|████████▋ | 5628/6470 [1:04:42<11:03,  1.27it/s]
3|5629|Loss: 0.907076358795166:  87%|████████▋ | 5628/6470 [1:04:42<11:03,  1.27it/s] 
3|5629|Loss: 0.907076358795166:  87%|████████▋ | 5629/6470 [1:04:42<10:42,  1.31it/s]
3|5630|Loss: 0.9416090846061707:  87%|████████▋ | 5629/6470 [1:04:43<10:42,  1.31it/s]
3|5630|Loss: 0.9416090846061707:  87%|████████▋ | 5630/6470 [1:04:43<10:26,  1.34it/s]
3|5631|Loss: 0.768322229385376:  87%|████████▋ | 5630/6470 [1:04:43<10:26,  1.34it/s] 
3|5631|Loss: 0.768322229385376:  87%|████████▋ | 5631/6470 [1:04:44<10:33,  1.32it/s]
3|5632|Loss: 0.8178395628929138:  87%|████████▋ | 5631/6470 [1:04:44<10:33,  1.32it/s]
3|5632|Loss: 0.8178395628929138:  87%|████████▋ | 5632/6470 [1:04:44<10:35,  1.32it/s]
3|5633|Loss: 0.838957667350769:  87%|████████▋ | 5632/6470 [1:04:45<10:35,  1.32it/s] 
3|5633|Loss: 0.838957667350769:  87%|████████▋ | 5633/6470 [1:04:45<10:24,  1.34it/s]
3|5634|Loss: 0.7910035252571106:  87%|████████▋ | 5633/6470 [1:04:46<10:24,  1.34it/s]
3|5634|Loss: 0.7910035252571106:  87%|████████▋ | 5634/6470 [1:04:46<09:48,  1.42it/s]
3|5635|Loss: 0.9267060160636902:  87%|████████▋ | 5634/6470 [1:04:46<09:48,  1.42it/s]
3|5635|Loss: 0.9267060160636902:  87%|████████▋ | 5635/6470 [1:04:46<09:41,  1.44it/s]
3|5636|Loss: 0.9778719544410706:  87%|████████▋ | 5635/6470 [1:04:47<09:41,  1.44it/s]
3|5636|Loss: 0.9778719544410706:  87%|████████▋ | 5636/6470 [1:04:47<08:53,  1.56it/s]
3|5637|Loss: 0.9272047281265259:  87%|████████▋ | 5636/6470 [1:04:47<08:53,  1.56it/s]
3|5637|Loss: 0.9272047281265259:  87%|████████▋ | 5637/6470 [1:04:48<08:37,  1.61it/s]
3|5638|Loss: 0.8014348745346069:  87%|████████▋ | 5637/6470 [1:04:48<08:37,  1.61it/s]
3|5638|Loss: 0.8014348745346069:  87%|████████▋ | 5638/6470 [1:04:48<08:27,  1.64it/s]
3|5639|Loss: 0.7540357112884521:  87%|████████▋ | 5638/6470 [1:04:49<08:27,  1.64it/s]
3|5639|Loss: 0.7540357112884521:  87%|████████▋ | 5639/6470 [1:04:49<08:53,  1.56it/s]
3|5640|Loss: 0.8838728666305542:  87%|████████▋ | 5639/6470 [1:04:49<08:53,  1.56it/s]
3|5640|Loss: 0.8838728666305542:  87%|████████▋ | 5640/6470 [1:04:50<09:28,  1.46it/s]
3|5641|Loss: 0.752101719379425:  87%|████████▋ | 5640/6470 [1:04:50<09:28,  1.46it/s] 
3|5641|Loss: 0.752101719379425:  87%|████████▋ | 5641/6470 [1:04:50<09:26,  1.46it/s]
3|5642|Loss: 1.0021954774856567:  87%|████████▋ | 5641/6470 [1:04:51<09:26,  1.46it/s]
3|5642|Loss: 1.0021954774856567:  87%|████████▋ | 5642/6470 [1:04:51<09:52,  1.40it/s]
3|5643|Loss: 0.6811604499816895:  87%|████████▋ | 5642/6470 [1:04:52<09:52,  1.40it/s]
3|5643|Loss: 0.6811604499816895:  87%|████████▋ | 5643/6470 [1:04:52<10:25,  1.32it/s]
3|5644|Loss: 0.8135943412780762:  87%|████████▋ | 5643/6470 [1:04:52<10:25,  1.32it/s]
3|5644|Loss: 0.8135943412780762:  87%|████████▋ | 5644/6470 [1:04:53<10:25,  1.32it/s]
3|5645|Loss: 0.825027346611023:  87%|████████▋ | 5644/6470 [1:04:53<10:25,  1.32it/s] 
3|5645|Loss: 0.825027346611023:  87%|████████▋ | 5645/6470 [1:04:53<10:12,  1.35it/s]
3|5646|Loss: 0.8829501271247864:  87%|████████▋ | 5645/6470 [1:04:54<10:12,  1.35it/s]
3|5646|Loss: 0.8829501271247864:  87%|████████▋ | 5646/6470 [1:04:54<09:30,  1.44it/s]
3|5647|Loss: 0.7700066566467285:  87%|████████▋ | 5646/6470 [1:04:54<09:30,  1.44it/s]
3|5647|Loss: 0.7700066566467285:  87%|████████▋ | 5647/6470 [1:04:55<09:13,  1.49it/s]
3|5648|Loss: 0.893775224685669:  87%|████████▋ | 5647/6470 [1:04:55<09:13,  1.49it/s] 
3|5648|Loss: 0.893775224685669:  87%|████████▋ | 5648/6470 [1:04:55<09:26,  1.45it/s]
3|5649|Loss: 0.8702005743980408:  87%|████████▋ | 5648/6470 [1:04:56<09:26,  1.45it/s]
3|5649|Loss: 0.8702005743980408:  87%|████████▋ | 5649/6470 [1:04:56<08:56,  1.53it/s]
3|5650|Loss: 0.8754255175590515:  87%|████████▋ | 5649/6470 [1:04:56<08:56,  1.53it/s]
3|5650|Loss: 0.8754255175590515:  87%|████████▋ | 5650/6470 [1:04:57<09:03,  1.51it/s]
3|5651|Loss: 0.84921795129776:  87%|████████▋ | 5650/6470 [1:04:57<09:03,  1.51it/s]  
3|5651|Loss: 0.84921795129776:  87%|████████▋ | 5651/6470 [1:04:57<09:21,  1.46it/s]
3|5652|Loss: 1.0510642528533936:  87%|████████▋ | 5651/6470 [1:04:58<09:21,  1.46it/s]
3|5652|Loss: 1.0510642528533936:  87%|████████▋ | 5652/6470 [1:04:58<09:33,  1.43it/s]
3|5653|Loss: 0.8292467594146729:  87%|████████▋ | 5652/6470 [1:04:59<09:33,  1.43it/s]
3|5653|Loss: 0.8292467594146729:  87%|████████▋ | 5653/6470 [1:04:59<09:44,  1.40it/s]
3|5654|Loss: 0.9348796010017395:  87%|████████▋ | 5653/6470 [1:04:59<09:44,  1.40it/s]
3|5654|Loss: 0.9348796010017395:  87%|████████▋ | 5654/6470 [1:05:00<10:00,  1.36it/s]
3|5655|Loss: 0.8858959674835205:  87%|████████▋ | 5654/6470 [1:05:00<10:00,  1.36it/s]
3|5655|Loss: 0.8858959674835205:  87%|████████▋ | 5655/6470 [1:05:00<09:56,  1.37it/s]
3|5656|Loss: 0.8798936009407043:  87%|████████▋ | 5655/6470 [1:05:01<09:56,  1.37it/s]
3|5656|Loss: 0.8798936009407043:  87%|████████▋ | 5656/6470 [1:05:01<09:47,  1.38it/s]
3|5657|Loss: 0.6955675482749939:  87%|████████▋ | 5656/6470 [1:05:01<09:47,  1.38it/s]
3|5657|Loss: 0.6955675482749939:  87%|████████▋ | 5657/6470 [1:05:02<09:47,  1.38it/s]
3|5658|Loss: 0.9075685143470764:  87%|████████▋ | 5657/6470 [1:05:02<09:47,  1.38it/s]
3|5658|Loss: 0.9075685143470764:  87%|████████▋ | 5658/6470 [1:05:03<10:12,  1.33it/s]
3|5659|Loss: 0.8973309993743896:  87%|████████▋ | 5658/6470 [1:05:03<10:12,  1.33it/s]
3|5659|Loss: 0.8973309993743896:  87%|████████▋ | 5659/6470 [1:05:03<10:32,  1.28it/s]
3|5660|Loss: 1.0148944854736328:  87%|████████▋ | 5659/6470 [1:05:04<10:32,  1.28it/s]
3|5660|Loss: 1.0148944854736328:  87%|████████▋ | 5660/6470 [1:05:04<10:24,  1.30it/s]
3|5661|Loss: 0.7789803743362427:  87%|████████▋ | 5660/6470 [1:05:05<10:24,  1.30it/s]
3|5661|Loss: 0.7789803743362427:  87%|████████▋ | 5661/6470 [1:05:05<09:51,  1.37it/s]
3|5662|Loss: 0.9693374633789062:  87%|████████▋ | 5661/6470 [1:05:05<09:51,  1.37it/s]
3|5662|Loss: 0.9693374633789062:  88%|████████▊ | 5662/6470 [1:05:06<09:39,  1.39it/s]
3|5663|Loss: 0.8513599634170532:  88%|████████▊ | 5662/6470 [1:05:06<09:39,  1.39it/s]
3|5663|Loss: 0.8513599634170532:  88%|████████▊ | 5663/6470 [1:05:06<09:23,  1.43it/s]
3|5664|Loss: 0.9658142328262329:  88%|████████▊ | 5663/6470 [1:05:07<09:23,  1.43it/s]
3|5664|Loss: 0.9658142328262329:  88%|████████▊ | 5664/6470 [1:05:07<09:19,  1.44it/s]
3|5665|Loss: 0.8444979786872864:  88%|████████▊ | 5664/6470 [1:05:07<09:19,  1.44it/s]
3|5665|Loss: 0.8444979786872864:  88%|████████▊ | 5665/6470 [1:05:08<09:29,  1.41it/s]
3|5666|Loss: 0.9209326505661011:  88%|████████▊ | 5665/6470 [1:05:08<09:29,  1.41it/s]
3|5666|Loss: 0.9209326505661011:  88%|████████▊ | 5666/6470 [1:05:08<08:49,  1.52it/s]
3|5667|Loss: 0.8342275619506836:  88%|████████▊ | 5666/6470 [1:05:08<08:49,  1.52it/s]
3|5667|Loss: 0.8342275619506836:  88%|████████▊ | 5667/6470 [1:05:09<08:34,  1.56it/s]
3|5668|Loss: 0.7935677766799927:  88%|████████▊ | 5667/6470 [1:05:09<08:34,  1.56it/s]
3|5668|Loss: 0.7935677766799927:  88%|████████▊ | 5668/6470 [1:05:09<08:32,  1.56it/s]
3|5669|Loss: 0.9330223202705383:  88%|████████▊ | 5668/6470 [1:05:10<08:32,  1.56it/s]
3|5669|Loss: 0.9330223202705383:  88%|████████▊ | 5669/6470 [1:05:10<08:51,  1.51it/s]
3|5670|Loss: 1.0409817695617676:  88%|████████▊ | 5669/6470 [1:05:10<08:51,  1.51it/s]
3|5670|Loss: 1.0409817695617676:  88%|████████▊ | 5670/6470 [1:05:11<08:45,  1.52it/s]
3|5671|Loss: 0.7692884802818298:  88%|████████▊ | 5670/6470 [1:05:11<08:45,  1.52it/s]
3|5671|Loss: 0.7692884802818298:  88%|████████▊ | 5671/6470 [1:05:11<08:33,  1.56it/s]
3|5672|Loss: 0.8344159722328186:  88%|████████▊ | 5671/6470 [1:05:12<08:33,  1.56it/s]
3|5672|Loss: 0.8344159722328186:  88%|████████▊ | 5672/6470 [1:05:12<08:27,  1.57it/s]
3|5673|Loss: 0.8857424259185791:  88%|████████▊ | 5672/6470 [1:05:12<08:27,  1.57it/s]
3|5673|Loss: 0.8857424259185791:  88%|████████▊ | 5673/6470 [1:05:13<08:44,  1.52it/s]
3|5674|Loss: 0.7890225052833557:  88%|████████▊ | 5673/6470 [1:05:13<08:44,  1.52it/s]
3|5674|Loss: 0.7890225052833557:  88%|████████▊ | 5674/6470 [1:05:13<09:12,  1.44it/s]
3|5675|Loss: 0.7756751775741577:  88%|████████▊ | 5674/6470 [1:05:14<09:12,  1.44it/s]
3|5675|Loss: 0.7756751775741577:  88%|████████▊ | 5675/6470 [1:05:14<09:20,  1.42it/s]
3|5676|Loss: 0.8176103830337524:  88%|████████▊ | 5675/6470 [1:05:15<09:20,  1.42it/s]
3|5676|Loss: 0.8176103830337524:  88%|████████▊ | 5676/6470 [1:05:15<09:19,  1.42it/s]
3|5677|Loss: 0.8488776087760925:  88%|████████▊ | 5676/6470 [1:05:15<09:19,  1.42it/s]
3|5677|Loss: 0.8488776087760925:  88%|████████▊ | 5677/6470 [1:05:16<09:34,  1.38it/s]
3|5678|Loss: 0.9332351684570312:  88%|████████▊ | 5677/6470 [1:05:16<09:34,  1.38it/s]
3|5678|Loss: 0.9332351684570312:  88%|████████▊ | 5678/6470 [1:05:16<09:40,  1.36it/s]
3|5679|Loss: 0.8523879051208496:  88%|████████▊ | 5678/6470 [1:05:17<09:40,  1.36it/s]
3|5679|Loss: 0.8523879051208496:  88%|████████▊ | 5679/6470 [1:05:17<09:57,  1.32it/s]
3|5680|Loss: 0.7662051916122437:  88%|████████▊ | 5679/6470 [1:05:18<09:57,  1.32it/s]
3|5680|Loss: 0.7662051916122437:  88%|████████▊ | 5680/6470 [1:05:18<09:34,  1.38it/s]
3|5681|Loss: 0.7525584101676941:  88%|████████▊ | 5680/6470 [1:05:18<09:34,  1.38it/s]
3|5681|Loss: 0.7525584101676941:  88%|████████▊ | 5681/6470 [1:05:19<09:30,  1.38it/s]
3|5682|Loss: 0.7313011884689331:  88%|████████▊ | 5681/6470 [1:05:19<09:30,  1.38it/s]
3|5682|Loss: 0.7313011884689331:  88%|████████▊ | 5682/6470 [1:05:19<09:58,  1.32it/s]
3|5683|Loss: 0.9558274149894714:  88%|████████▊ | 5682/6470 [1:05:20<09:58,  1.32it/s]
3|5683|Loss: 0.9558274149894714:  88%|████████▊ | 5683/6470 [1:05:20<09:39,  1.36it/s]
3|5684|Loss: 0.8701806664466858:  88%|████████▊ | 5683/6470 [1:05:20<09:39,  1.36it/s]
3|5684|Loss: 0.8701806664466858:  88%|████████▊ | 5684/6470 [1:05:21<09:11,  1.43it/s]
3|5685|Loss: 0.9139629006385803:  88%|████████▊ | 5684/6470 [1:05:21<09:11,  1.43it/s]
3|5685|Loss: 0.9139629006385803:  88%|████████▊ | 5685/6470 [1:05:21<09:13,  1.42it/s]
3|5686|Loss: 0.7584638595581055:  88%|████████▊ | 5685/6470 [1:05:22<09:13,  1.42it/s]
3|5686|Loss: 0.7584638595581055:  88%|████████▊ | 5686/6470 [1:05:22<09:46,  1.34it/s]
3|5687|Loss: 0.7655630707740784:  88%|████████▊ | 5686/6470 [1:05:23<09:46,  1.34it/s]
3|5687|Loss: 0.7655630707740784:  88%|████████▊ | 5687/6470 [1:05:23<10:03,  1.30it/s]
3|5688|Loss: 1.0165729522705078:  88%|████████▊ | 5687/6470 [1:05:23<10:03,  1.30it/s]
3|5688|Loss: 1.0165729522705078:  88%|████████▊ | 5688/6470 [1:05:24<08:51,  1.47it/s]
3|5689|Loss: 0.8043211102485657:  88%|████████▊ | 5688/6470 [1:05:24<08:51,  1.47it/s]
3|5689|Loss: 0.8043211102485657:  88%|████████▊ | 5689/6470 [1:05:24<08:48,  1.48it/s]
3|5690|Loss: 0.9933410882949829:  88%|████████▊ | 5689/6470 [1:05:25<08:48,  1.48it/s]
3|5690|Loss: 0.9933410882949829:  88%|████████▊ | 5690/6470 [1:05:25<09:01,  1.44it/s]
3|5691|Loss: 0.8220138549804688:  88%|████████▊ | 5690/6470 [1:05:25<09:01,  1.44it/s]
3|5691|Loss: 0.8220138549804688:  88%|████████▊ | 5691/6470 [1:05:26<09:04,  1.43it/s]
3|5692|Loss: 0.632286548614502:  88%|████████▊ | 5691/6470 [1:05:26<09:04,  1.43it/s] 
3|5692|Loss: 0.632286548614502:  88%|████████▊ | 5692/6470 [1:05:26<09:03,  1.43it/s]
3|5693|Loss: 0.8606443405151367:  88%|████████▊ | 5692/6470 [1:05:27<09:03,  1.43it/s]
3|5693|Loss: 0.8606443405151367:  88%|████████▊ | 5693/6470 [1:05:27<09:26,  1.37it/s]
3|5694|Loss: 0.8722657561302185:  88%|████████▊ | 5693/6470 [1:05:28<09:26,  1.37it/s]
3|5694|Loss: 0.8722657561302185:  88%|████████▊ | 5694/6470 [1:05:28<08:40,  1.49it/s]
3|5695|Loss: 0.8808448314666748:  88%|████████▊ | 5694/6470 [1:05:28<08:40,  1.49it/s]
3|5695|Loss: 0.8808448314666748:  88%|████████▊ | 5695/6470 [1:05:28<08:32,  1.51it/s]
3|5696|Loss: 0.8080055117607117:  88%|████████▊ | 5695/6470 [1:05:29<08:32,  1.51it/s]
3|5696|Loss: 0.8080055117607117:  88%|████████▊ | 5696/6470 [1:05:29<08:17,  1.56it/s]
3|5697|Loss: 0.8590239882469177:  88%|████████▊ | 5696/6470 [1:05:29<08:17,  1.56it/s]
3|5697|Loss: 0.8590239882469177:  88%|████████▊ | 5697/6470 [1:05:30<08:35,  1.50it/s]
3|5698|Loss: 0.886671245098114:  88%|████████▊ | 5697/6470 [1:05:30<08:35,  1.50it/s] 
3|5698|Loss: 0.886671245098114:  88%|████████▊ | 5698/6470 [1:05:30<08:31,  1.51it/s]
3|5699|Loss: 0.9802863001823425:  88%|████████▊ | 5698/6470 [1:05:31<08:31,  1.51it/s]
3|5699|Loss: 0.9802863001823425:  88%|████████▊ | 5699/6470 [1:05:31<08:36,  1.49it/s]
3|5700|Loss: 0.9187332987785339:  88%|████████▊ | 5699/6470 [1:05:31<08:36,  1.49it/s]
3|5700|Loss: 0.9187332987785339:  88%|████████▊ | 5700/6470 [1:05:32<08:49,  1.45it/s]
3|5701|Loss: 0.8872143030166626:  88%|████████▊ | 5700/6470 [1:05:32<08:49,  1.45it/s]
3|5701|Loss: 0.8872143030166626:  88%|████████▊ | 5701/6470 [1:05:32<08:37,  1.49it/s]
3|5702|Loss: 0.7789482474327087:  88%|████████▊ | 5701/6470 [1:05:33<08:37,  1.49it/s]
3|5702|Loss: 0.7789482474327087:  88%|████████▊ | 5702/6470 [1:05:33<09:09,  1.40it/s]
3|5703|Loss: 0.8856898546218872:  88%|████████▊ | 5702/6470 [1:05:34<09:09,  1.40it/s]
3|5703|Loss: 0.8856898546218872:  88%|████████▊ | 5703/6470 [1:05:34<09:23,  1.36it/s]
3|5704|Loss: 0.7757365703582764:  88%|████████▊ | 5703/6470 [1:05:34<09:23,  1.36it/s]
3|5704|Loss: 0.7757365703582764:  88%|████████▊ | 5704/6470 [1:05:35<09:52,  1.29it/s]
3|5705|Loss: 0.763741672039032:  88%|████████▊ | 5704/6470 [1:05:35<09:52,  1.29it/s] 
3|5705|Loss: 0.763741672039032:  88%|████████▊ | 5705/6470 [1:05:36<10:00,  1.27it/s]
3|5706|Loss: 0.8292145133018494:  88%|████████▊ | 5705/6470 [1:05:36<10:00,  1.27it/s]
3|5706|Loss: 0.8292145133018494:  88%|████████▊ | 5706/6470 [1:05:36<09:32,  1.33it/s]
3|5707|Loss: 1.0399377346038818:  88%|████████▊ | 5706/6470 [1:05:37<09:32,  1.33it/s]
3|5707|Loss: 1.0399377346038818:  88%|████████▊ | 5707/6470 [1:05:37<09:27,  1.35it/s]
3|5708|Loss: 0.9080102443695068:  88%|████████▊ | 5707/6470 [1:05:37<09:27,  1.35it/s]
3|5708|Loss: 0.9080102443695068:  88%|████████▊ | 5708/6470 [1:05:38<09:09,  1.39it/s]
3|5709|Loss: 0.9169119596481323:  88%|████████▊ | 5708/6470 [1:05:38<09:09,  1.39it/s]
3|5709|Loss: 0.9169119596481323:  88%|████████▊ | 5709/6470 [1:05:38<08:26,  1.50it/s]
3|5710|Loss: 0.7738818526268005:  88%|████████▊ | 5709/6470 [1:05:39<08:26,  1.50it/s]
3|5710|Loss: 0.7738818526268005:  88%|████████▊ | 5710/6470 [1:05:39<09:05,  1.39it/s]
3|5711|Loss: 0.8118919134140015:  88%|████████▊ | 5710/6470 [1:05:40<09:05,  1.39it/s]
3|5711|Loss: 0.8118919134140015:  88%|████████▊ | 5711/6470 [1:05:40<09:15,  1.37it/s]
3|5712|Loss: 0.8463085293769836:  88%|████████▊ | 5711/6470 [1:05:40<09:15,  1.37it/s]
3|5712|Loss: 0.8463085293769836:  88%|████████▊ | 5712/6470 [1:05:41<08:59,  1.40it/s]
3|5713|Loss: 0.871385931968689:  88%|████████▊ | 5712/6470 [1:05:41<08:59,  1.40it/s] 
3|5713|Loss: 0.871385931968689:  88%|████████▊ | 5713/6470 [1:05:41<08:42,  1.45it/s]
3|5714|Loss: 0.8231527805328369:  88%|████████▊ | 5713/6470 [1:05:42<08:42,  1.45it/s]
3|5714|Loss: 0.8231527805328369:  88%|████████▊ | 5714/6470 [1:05:42<08:30,  1.48it/s]
3|5715|Loss: 0.8150554299354553:  88%|████████▊ | 5714/6470 [1:05:42<08:30,  1.48it/s]
3|5715|Loss: 0.8150554299354553:  88%|████████▊ | 5715/6470 [1:05:43<08:32,  1.47it/s]
3|5716|Loss: 0.9684033393859863:  88%|████████▊ | 5715/6470 [1:05:43<08:32,  1.47it/s]
3|5716|Loss: 0.9684033393859863:  88%|████████▊ | 5716/6470 [1:05:43<08:45,  1.44it/s]
3|5717|Loss: 0.9417576789855957:  88%|████████▊ | 5716/6470 [1:05:44<08:45,  1.44it/s]
3|5717|Loss: 0.9417576789855957:  88%|████████▊ | 5717/6470 [1:05:44<08:54,  1.41it/s]
3|5718|Loss: 0.9433310627937317:  88%|████████▊ | 5717/6470 [1:05:44<08:54,  1.41it/s]
3|5718|Loss: 0.9433310627937317:  88%|████████▊ | 5718/6470 [1:05:45<08:52,  1.41it/s]
3|5719|Loss: 0.7994592189788818:  88%|████████▊ | 5718/6470 [1:05:45<08:52,  1.41it/s]
3|5719|Loss: 0.7994592189788818:  88%|████████▊ | 5719/6470 [1:05:46<09:16,  1.35it/s]
3|5720|Loss: 0.7277010679244995:  88%|████████▊ | 5719/6470 [1:05:46<09:16,  1.35it/s]
3|5720|Loss: 0.7277010679244995:  88%|████████▊ | 5720/6470 [1:05:46<09:41,  1.29it/s]
3|5721|Loss: 0.7692305445671082:  88%|████████▊ | 5720/6470 [1:05:47<09:41,  1.29it/s]
3|5721|Loss: 0.7692305445671082:  88%|████████▊ | 5721/6470 [1:05:47<10:00,  1.25it/s]
3|5722|Loss: 0.8730611205101013:  88%|████████▊ | 5721/6470 [1:05:48<10:00,  1.25it/s]
3|5722|Loss: 0.8730611205101013:  88%|████████▊ | 5722/6470 [1:05:48<09:34,  1.30it/s]
3|5723|Loss: 0.985154926776886:  88%|████████▊ | 5722/6470 [1:05:48<09:34,  1.30it/s] 
3|5723|Loss: 0.985154926776886:  88%|████████▊ | 5723/6470 [1:05:49<09:04,  1.37it/s]
3|5724|Loss: 0.8286880850791931:  88%|████████▊ | 5723/6470 [1:05:49<09:04,  1.37it/s]
3|5724|Loss: 0.8286880850791931:  88%|████████▊ | 5724/6470 [1:05:49<09:00,  1.38it/s]
3|5725|Loss: 0.8632512092590332:  88%|████████▊ | 5724/6470 [1:05:50<09:00,  1.38it/s]
3|5725|Loss: 0.8632512092590332:  88%|████████▊ | 5725/6470 [1:05:50<08:28,  1.46it/s]
3|5726|Loss: 0.8381445407867432:  88%|████████▊ | 5725/6470 [1:05:50<08:28,  1.46it/s]
3|5726|Loss: 0.8381445407867432:  89%|████████▊ | 5726/6470 [1:05:50<08:14,  1.50it/s]
3|5727|Loss: 1.087922215461731:  89%|████████▊ | 5726/6470 [1:05:51<08:14,  1.50it/s] 
3|5727|Loss: 1.087922215461731:  89%|████████▊ | 5727/6470 [1:05:51<07:53,  1.57it/s]
3|5728|Loss: 0.8986204266548157:  89%|████████▊ | 5727/6470 [1:05:51<07:53,  1.57it/s]
3|5728|Loss: 0.8986204266548157:  89%|████████▊ | 5728/6470 [1:05:52<07:39,  1.62it/s]
3|5729|Loss: 0.89720219373703:  89%|████████▊ | 5728/6470 [1:05:52<07:39,  1.62it/s]  
3|5729|Loss: 0.89720219373703:  89%|████████▊ | 5729/6470 [1:05:52<07:08,  1.73it/s]
3|5730|Loss: 0.8004823327064514:  89%|████████▊ | 5729/6470 [1:05:52<07:08,  1.73it/s]
3|5730|Loss: 0.8004823327064514:  89%|████████▊ | 5730/6470 [1:05:53<07:39,  1.61it/s]
3|5731|Loss: 0.8313466310501099:  89%|████████▊ | 5730/6470 [1:05:53<07:39,  1.61it/s]
3|5731|Loss: 0.8313466310501099:  89%|████████▊ | 5731/6470 [1:05:54<08:10,  1.51it/s]
3|5732|Loss: 0.9545971155166626:  89%|████████▊ | 5731/6470 [1:05:54<08:10,  1.51it/s]
3|5732|Loss: 0.9545971155166626:  89%|████████▊ | 5732/6470 [1:05:54<08:21,  1.47it/s]
3|5733|Loss: 0.7179926633834839:  89%|████████▊ | 5732/6470 [1:05:55<08:21,  1.47it/s]
3|5733|Loss: 0.7179926633834839:  89%|████████▊ | 5733/6470 [1:05:55<08:25,  1.46it/s]
3|5734|Loss: 0.951789140701294:  89%|████████▊ | 5733/6470 [1:05:55<08:25,  1.46it/s] 
3|5734|Loss: 0.951789140701294:  89%|████████▊ | 5734/6470 [1:05:56<07:43,  1.59it/s]
3|5735|Loss: 0.8311971426010132:  89%|████████▊ | 5734/6470 [1:05:56<07:43,  1.59it/s]
3|5735|Loss: 0.8311971426010132:  89%|████████▊ | 5735/6470 [1:05:56<07:36,  1.61it/s]
3|5736|Loss: 0.7887886762619019:  89%|████████▊ | 5735/6470 [1:05:56<07:36,  1.61it/s]
3|5736|Loss: 0.7887886762619019:  89%|████████▊ | 5736/6470 [1:05:57<07:40,  1.59it/s]
3|5737|Loss: 0.8748914003372192:  89%|████████▊ | 5736/6470 [1:05:57<07:40,  1.59it/s]
3|5737|Loss: 0.8748914003372192:  89%|████████▊ | 5737/6470 [1:05:57<08:00,  1.53it/s]
3|5738|Loss: 0.842217743396759:  89%|████████▊ | 5737/6470 [1:05:58<08:00,  1.53it/s] 
3|5738|Loss: 0.842217743396759:  89%|████████▊ | 5738/6470 [1:05:58<08:12,  1.49it/s]
3|5739|Loss: 0.8430097699165344:  89%|████████▊ | 5738/6470 [1:05:59<08:12,  1.49it/s]
3|5739|Loss: 0.8430097699165344:  89%|████████▊ | 5739/6470 [1:05:59<08:07,  1.50it/s]
3|5740|Loss: 1.0031867027282715:  89%|████████▊ | 5739/6470 [1:05:59<08:07,  1.50it/s]
3|5740|Loss: 1.0031867027282715:  89%|████████▊ | 5740/6470 [1:06:00<08:29,  1.43it/s]
3|5741|Loss: 0.7603650093078613:  89%|████████▊ | 5740/6470 [1:06:00<08:29,  1.43it/s]
3|5741|Loss: 0.7603650093078613:  89%|████████▊ | 5741/6470 [1:06:00<08:22,  1.45it/s]
3|5742|Loss: 0.8679574728012085:  89%|████████▊ | 5741/6470 [1:06:01<08:22,  1.45it/s]
3|5742|Loss: 0.8679574728012085:  89%|████████▊ | 5742/6470 [1:06:01<08:20,  1.45it/s]
3|5743|Loss: 0.9762956500053406:  89%|████████▊ | 5742/6470 [1:06:01<08:20,  1.45it/s]
3|5743|Loss: 0.9762956500053406:  89%|████████▉ | 5743/6470 [1:06:02<08:28,  1.43it/s]
3|5744|Loss: 0.8202367424964905:  89%|████████▉ | 5743/6470 [1:06:02<08:28,  1.43it/s]
3|5744|Loss: 0.8202367424964905:  89%|████████▉ | 5744/6470 [1:06:03<09:00,  1.34it/s]
3|5745|Loss: 0.7244725823402405:  89%|████████▉ | 5744/6470 [1:06:03<09:00,  1.34it/s]
3|5745|Loss: 0.7244725823402405:  89%|████████▉ | 5745/6470 [1:06:03<09:08,  1.32it/s]
3|5746|Loss: 1.0045768022537231:  89%|████████▉ | 5745/6470 [1:06:04<09:08,  1.32it/s]
3|5746|Loss: 1.0045768022537231:  89%|████████▉ | 5746/6470 [1:06:04<08:46,  1.38it/s]
3|5747|Loss: 0.904479444026947:  89%|████████▉ | 5746/6470 [1:06:04<08:46,  1.38it/s] 
3|5747|Loss: 0.904479444026947:  89%|████████▉ | 5747/6470 [1:06:05<08:44,  1.38it/s]
3|5748|Loss: 0.799973726272583:  89%|████████▉ | 5747/6470 [1:06:05<08:44,  1.38it/s]
3|5748|Loss: 0.799973726272583:  89%|████████▉ | 5748/6470 [1:06:05<08:15,  1.46it/s]
3|5749|Loss: 0.8105513453483582:  89%|████████▉ | 5748/6470 [1:06:06<08:15,  1.46it/s]
3|5749|Loss: 0.8105513453483582:  89%|████████▉ | 5749/6470 [1:06:06<08:10,  1.47it/s]
3|5750|Loss: 1.0554804801940918:  89%|████████▉ | 5749/6470 [1:06:06<08:10,  1.47it/s]
3|5750|Loss: 1.0554804801940918:  89%|████████▉ | 5750/6470 [1:06:07<08:20,  1.44it/s]
3|5751|Loss: 0.8291664123535156:  89%|████████▉ | 5750/6470 [1:06:07<08:20,  1.44it/s]
3|5751|Loss: 0.8291664123535156:  89%|████████▉ | 5751/6470 [1:06:07<08:07,  1.47it/s]
3|5752|Loss: 0.8566311001777649:  89%|████████▉ | 5751/6470 [1:06:08<08:07,  1.47it/s]
3|5752|Loss: 0.8566311001777649:  89%|████████▉ | 5752/6470 [1:06:08<08:07,  1.47it/s]
3|5753|Loss: 0.8365595936775208:  89%|████████▉ | 5752/6470 [1:06:08<08:07,  1.47it/s]
3|5753|Loss: 0.8365595936775208:  89%|████████▉ | 5753/6470 [1:06:09<08:01,  1.49it/s]
3|5754|Loss: 0.7721613049507141:  89%|████████▉ | 5753/6470 [1:06:09<08:01,  1.49it/s]
3|5754|Loss: 0.7721613049507141:  89%|████████▉ | 5754/6470 [1:06:09<08:09,  1.46it/s]
3|5755|Loss: 0.7650297284126282:  89%|████████▉ | 5754/6470 [1:06:10<08:09,  1.46it/s]
3|5755|Loss: 0.7650297284126282:  89%|████████▉ | 5755/6470 [1:06:10<08:03,  1.48it/s]
3|5756|Loss: 0.8926546573638916:  89%|████████▉ | 5755/6470 [1:06:10<08:03,  1.48it/s]
3|5756|Loss: 0.8926546573638916:  89%|████████▉ | 5756/6470 [1:06:11<07:50,  1.52it/s]
3|5757|Loss: 0.8421998023986816:  89%|████████▉ | 5756/6470 [1:06:11<07:50,  1.52it/s]
3|5757|Loss: 0.8421998023986816:  89%|████████▉ | 5757/6470 [1:06:11<08:20,  1.42it/s]
3|5758|Loss: 0.8677613735198975:  89%|████████▉ | 5757/6470 [1:06:12<08:20,  1.42it/s]
3|5758|Loss: 0.8677613735198975:  89%|████████▉ | 5758/6470 [1:06:12<08:27,  1.40it/s]
3|5759|Loss: 0.8311710953712463:  89%|████████▉ | 5758/6470 [1:06:13<08:27,  1.40it/s]
3|5759|Loss: 0.8311710953712463:  89%|████████▉ | 5759/6470 [1:06:13<08:26,  1.40it/s]
3|5760|Loss: 0.7418177723884583:  89%|████████▉ | 5759/6470 [1:06:13<08:26,  1.40it/s]
3|5760|Loss: 0.7418177723884583:  89%|████████▉ | 5760/6470 [1:06:14<08:36,  1.37it/s]
3|5761|Loss: 0.868628203868866:  89%|████████▉ | 5760/6470 [1:06:14<08:36,  1.37it/s] 
3|5761|Loss: 0.868628203868866:  89%|████████▉ | 5761/6470 [1:06:14<08:23,  1.41it/s]
3|5762|Loss: 0.8167557120323181:  89%|████████▉ | 5761/6470 [1:06:15<08:23,  1.41it/s]
3|5762|Loss: 0.8167557120323181:  89%|████████▉ | 5762/6470 [1:06:15<08:23,  1.41it/s]
3|5763|Loss: 0.9407743215560913:  89%|████████▉ | 5762/6470 [1:06:15<08:23,  1.41it/s]
3|5763|Loss: 0.9407743215560913:  89%|████████▉ | 5763/6470 [1:06:16<07:48,  1.51it/s]
3|5764|Loss: 0.7455393075942993:  89%|████████▉ | 5763/6470 [1:06:16<07:48,  1.51it/s]
3|5764|Loss: 0.7455393075942993:  89%|████████▉ | 5764/6470 [1:06:16<07:05,  1.66it/s]
3|5765|Loss: 0.7645673155784607:  89%|████████▉ | 5764/6470 [1:06:16<07:05,  1.66it/s]
3|5765|Loss: 0.7645673155784607:  89%|████████▉ | 5765/6470 [1:06:17<07:18,  1.61it/s]
3|5766|Loss: 0.6879115104675293:  89%|████████▉ | 5765/6470 [1:06:17<07:18,  1.61it/s]
3|5766|Loss: 0.6879115104675293:  89%|████████▉ | 5766/6470 [1:06:17<07:41,  1.52it/s]
3|5767|Loss: 0.7870230674743652:  89%|████████▉ | 5766/6470 [1:06:18<07:41,  1.52it/s]
3|5767|Loss: 0.7870230674743652:  89%|████████▉ | 5767/6470 [1:06:18<07:29,  1.56it/s]
3|5768|Loss: 0.7242147326469421:  89%|████████▉ | 5767/6470 [1:06:18<07:29,  1.56it/s]
3|5768|Loss: 0.7242147326469421:  89%|████████▉ | 5768/6470 [1:06:19<07:41,  1.52it/s]
3|5769|Loss: 0.8985348343849182:  89%|████████▉ | 5768/6470 [1:06:19<07:41,  1.52it/s]
3|5769|Loss: 0.8985348343849182:  89%|████████▉ | 5769/6470 [1:06:19<07:13,  1.62it/s]
3|5770|Loss: 0.7929829955101013:  89%|████████▉ | 5769/6470 [1:06:20<07:13,  1.62it/s]
3|5770|Loss: 0.7929829955101013:  89%|████████▉ | 5770/6470 [1:06:20<07:35,  1.54it/s]
3|5771|Loss: 0.9238324165344238:  89%|████████▉ | 5770/6470 [1:06:20<07:35,  1.54it/s]
3|5771|Loss: 0.9238324165344238:  89%|████████▉ | 5771/6470 [1:06:21<07:21,  1.58it/s]
3|5772|Loss: 0.7740545868873596:  89%|████████▉ | 5771/6470 [1:06:21<07:21,  1.58it/s]
3|5772|Loss: 0.7740545868873596:  89%|████████▉ | 5772/6470 [1:06:21<07:14,  1.61it/s]
3|5773|Loss: 0.9259027242660522:  89%|████████▉ | 5772/6470 [1:06:22<07:14,  1.61it/s]
3|5773|Loss: 0.9259027242660522:  89%|████████▉ | 5773/6470 [1:06:22<07:15,  1.60it/s]
3|5774|Loss: 0.933041512966156:  89%|████████▉ | 5773/6470 [1:06:22<07:15,  1.60it/s] 
3|5774|Loss: 0.933041512966156:  89%|████████▉ | 5774/6470 [1:06:22<07:17,  1.59it/s]
3|5775|Loss: 0.8745988011360168:  89%|████████▉ | 5774/6470 [1:06:23<07:17,  1.59it/s]
3|5775|Loss: 0.8745988011360168:  89%|████████▉ | 5775/6470 [1:06:23<07:29,  1.55it/s]
3|5776|Loss: 0.9352465271949768:  89%|████████▉ | 5775/6470 [1:06:24<07:29,  1.55it/s]
3|5776|Loss: 0.9352465271949768:  89%|████████▉ | 5776/6470 [1:06:24<07:26,  1.55it/s]
3|5777|Loss: 0.7863363027572632:  89%|████████▉ | 5776/6470 [1:06:24<07:26,  1.55it/s]
3|5777|Loss: 0.7863363027572632:  89%|████████▉ | 5777/6470 [1:06:24<06:52,  1.68it/s]
3|5778|Loss: 0.877431333065033:  89%|████████▉ | 5777/6470 [1:06:25<06:52,  1.68it/s] 
3|5778|Loss: 0.877431333065033:  89%|████████▉ | 5778/6470 [1:06:25<06:36,  1.75it/s]
3|5779|Loss: 0.8017593622207642:  89%|████████▉ | 5778/6470 [1:06:25<06:36,  1.75it/s]
3|5779|Loss: 0.8017593622207642:  89%|████████▉ | 5779/6470 [1:06:25<06:35,  1.75it/s]
3|5780|Loss: 0.8184137940406799:  89%|████████▉ | 5779/6470 [1:06:26<06:35,  1.75it/s]
3|5780|Loss: 0.8184137940406799:  89%|████████▉ | 5780/6470 [1:06:26<07:21,  1.56it/s]
3|5781|Loss: 0.9225388169288635:  89%|████████▉ | 5780/6470 [1:06:27<07:21,  1.56it/s]
3|5781|Loss: 0.9225388169288635:  89%|████████▉ | 5781/6470 [1:06:27<07:19,  1.57it/s]
3|5782|Loss: 1.0091259479522705:  89%|████████▉ | 5781/6470 [1:06:27<07:19,  1.57it/s]
3|5782|Loss: 1.0091259479522705:  89%|████████▉ | 5782/6470 [1:06:28<07:27,  1.54it/s]
3|5783|Loss: 0.9885384440422058:  89%|████████▉ | 5782/6470 [1:06:28<07:27,  1.54it/s]
3|5783|Loss: 0.9885384440422058:  89%|████████▉ | 5783/6470 [1:06:28<07:24,  1.55it/s]
3|5784|Loss: 0.9962561726570129:  89%|████████▉ | 5783/6470 [1:06:29<07:24,  1.55it/s]
3|5784|Loss: 0.9962561726570129:  89%|████████▉ | 5784/6470 [1:06:29<07:21,  1.55it/s]
3|5785|Loss: 0.8216110467910767:  89%|████████▉ | 5784/6470 [1:06:29<07:21,  1.55it/s]
3|5785|Loss: 0.8216110467910767:  89%|████████▉ | 5785/6470 [1:06:30<07:36,  1.50it/s]
3|5786|Loss: 0.8212102055549622:  89%|████████▉ | 5785/6470 [1:06:30<07:36,  1.50it/s]
3|5786|Loss: 0.8212102055549622:  89%|████████▉ | 5786/6470 [1:06:30<07:45,  1.47it/s]
3|5787|Loss: 0.9264561533927917:  89%|████████▉ | 5786/6470 [1:06:31<07:45,  1.47it/s]
3|5787|Loss: 0.9264561533927917:  89%|████████▉ | 5787/6470 [1:06:31<08:08,  1.40it/s]
3|5788|Loss: 0.816717803478241:  89%|████████▉ | 5787/6470 [1:06:31<08:08,  1.40it/s] 
3|5788|Loss: 0.816717803478241:  89%|████████▉ | 5788/6470 [1:06:32<08:18,  1.37it/s]
3|5789|Loss: 0.8385297060012817:  89%|████████▉ | 5788/6470 [1:06:32<08:18,  1.37it/s]
3|5789|Loss: 0.8385297060012817:  89%|████████▉ | 5789/6470 [1:06:33<08:18,  1.37it/s]
3|5790|Loss: 0.9629756212234497:  89%|████████▉ | 5789/6470 [1:06:33<08:18,  1.37it/s]
3|5790|Loss: 0.9629756212234497:  89%|████████▉ | 5790/6470 [1:06:33<08:16,  1.37it/s]
3|5791|Loss: 1.0487065315246582:  89%|████████▉ | 5790/6470 [1:06:34<08:16,  1.37it/s]
3|5791|Loss: 1.0487065315246582:  90%|████████▉ | 5791/6470 [1:06:34<08:31,  1.33it/s]
3|5792|Loss: 0.9099057912826538:  90%|████████▉ | 5791/6470 [1:06:35<08:31,  1.33it/s]
3|5792|Loss: 0.9099057912826538:  90%|████████▉ | 5792/6470 [1:06:35<08:45,  1.29it/s]
3|5793|Loss: 0.8922640085220337:  90%|████████▉ | 5792/6470 [1:06:35<08:45,  1.29it/s]
3|5793|Loss: 0.8922640085220337:  90%|████████▉ | 5793/6470 [1:06:36<08:52,  1.27it/s]
3|5794|Loss: 0.9409686923027039:  90%|████████▉ | 5793/6470 [1:06:36<08:52,  1.27it/s]
3|5794|Loss: 0.9409686923027039:  90%|████████▉ | 5794/6470 [1:06:36<08:39,  1.30it/s]
3|5795|Loss: 0.7883763313293457:  90%|████████▉ | 5794/6470 [1:06:37<08:39,  1.30it/s]
3|5795|Loss: 0.7883763313293457:  90%|████████▉ | 5795/6470 [1:06:37<07:53,  1.42it/s]
3|5796|Loss: 0.8769245147705078:  90%|████████▉ | 5795/6470 [1:06:37<07:53,  1.42it/s]
3|5796|Loss: 0.8769245147705078:  90%|████████▉ | 5796/6470 [1:06:38<07:35,  1.48it/s]
3|5797|Loss: 0.8210793137550354:  90%|████████▉ | 5796/6470 [1:06:38<07:35,  1.48it/s]
3|5797|Loss: 0.8210793137550354:  90%|████████▉ | 5797/6470 [1:06:38<07:07,  1.58it/s]
3|5798|Loss: 0.7289028167724609:  90%|████████▉ | 5797/6470 [1:06:38<07:07,  1.58it/s]
3|5798|Loss: 0.7289028167724609:  90%|████████▉ | 5798/6470 [1:06:39<07:25,  1.51it/s]
3|5799|Loss: 1.0194098949432373:  90%|████████▉ | 5798/6470 [1:06:39<07:25,  1.51it/s]
3|5799|Loss: 1.0194098949432373:  90%|████████▉ | 5799/6470 [1:06:40<07:57,  1.40it/s]
3|5800|Loss: 0.7720335125923157:  90%|████████▉ | 5799/6470 [1:06:40<07:57,  1.40it/s]
3|5800|Loss: 0.7720335125923157:  90%|████████▉ | 5800/6470 [1:06:41<08:31,  1.31it/s]
3|5801|Loss: 0.8547780513763428:  90%|████████▉ | 5800/6470 [1:06:41<08:31,  1.31it/s]
3|5801|Loss: 0.8547780513763428:  90%|████████▉ | 5801/6470 [1:06:41<08:47,  1.27it/s]
3|5802|Loss: 0.8803226351737976:  90%|████████▉ | 5801/6470 [1:06:42<08:47,  1.27it/s]
3|5802|Loss: 0.8803226351737976:  90%|████████▉ | 5802/6470 [1:06:42<08:21,  1.33it/s]
3|5803|Loss: 0.8427878022193909:  90%|████████▉ | 5802/6470 [1:06:42<08:21,  1.33it/s]
3|5803|Loss: 0.8427878022193909:  90%|████████▉ | 5803/6470 [1:06:43<07:54,  1.41it/s]
3|5804|Loss: 0.7414056062698364:  90%|████████▉ | 5803/6470 [1:06:43<07:54,  1.41it/s]
3|5804|Loss: 0.7414056062698364:  90%|████████▉ | 5804/6470 [1:06:43<07:39,  1.45it/s]
3|5805|Loss: 0.7133945226669312:  90%|████████▉ | 5804/6470 [1:06:44<07:39,  1.45it/s]
3|5805|Loss: 0.7133945226669312:  90%|████████▉ | 5805/6470 [1:06:44<07:44,  1.43it/s]
3|5806|Loss: 0.8155903220176697:  90%|████████▉ | 5805/6470 [1:06:44<07:44,  1.43it/s]
3|5806|Loss: 0.8155903220176697:  90%|████████▉ | 5806/6470 [1:06:45<07:37,  1.45it/s]
3|5807|Loss: 0.6233566999435425:  90%|████████▉ | 5806/6470 [1:06:45<07:37,  1.45it/s]
3|5807|Loss: 0.6233566999435425:  90%|████████▉ | 5807/6470 [1:06:46<08:11,  1.35it/s]
3|5808|Loss: 0.9383853673934937:  90%|████████▉ | 5807/6470 [1:06:46<08:11,  1.35it/s]
3|5808|Loss: 0.9383853673934937:  90%|████████▉ | 5808/6470 [1:06:46<08:36,  1.28it/s]
3|5809|Loss: 0.8188213109970093:  90%|████████▉ | 5808/6470 [1:06:47<08:36,  1.28it/s]
3|5809|Loss: 0.8188213109970093:  90%|████████▉ | 5809/6470 [1:06:47<08:40,  1.27it/s]
3|5810|Loss: 0.9241958260536194:  90%|████████▉ | 5809/6470 [1:06:48<08:40,  1.27it/s]
3|5810|Loss: 0.9241958260536194:  90%|████████▉ | 5810/6470 [1:06:48<08:34,  1.28it/s]
3|5811|Loss: 1.0188872814178467:  90%|████████▉ | 5810/6470 [1:06:48<08:34,  1.28it/s]
3|5811|Loss: 1.0188872814178467:  90%|████████▉ | 5811/6470 [1:06:49<08:12,  1.34it/s]
3|5812|Loss: 0.9732802510261536:  90%|████████▉ | 5811/6470 [1:06:49<08:12,  1.34it/s]
3|5812|Loss: 0.9732802510261536:  90%|████████▉ | 5812/6470 [1:06:49<08:19,  1.32it/s]
3|5813|Loss: 0.7947712540626526:  90%|████████▉ | 5812/6470 [1:06:50<08:19,  1.32it/s]
3|5813|Loss: 0.7947712540626526:  90%|████████▉ | 5813/6470 [1:06:50<08:17,  1.32it/s]
3|5814|Loss: 0.8302071690559387:  90%|████████▉ | 5813/6470 [1:06:51<08:17,  1.32it/s]
3|5814|Loss: 0.8302071690559387:  90%|████████▉ | 5814/6470 [1:06:51<08:07,  1.35it/s]
3|5815|Loss: 0.9774665236473083:  90%|████████▉ | 5814/6470 [1:06:51<08:07,  1.35it/s]
3|5815|Loss: 0.9774665236473083:  90%|████████▉ | 5815/6470 [1:06:52<08:06,  1.35it/s]
3|5816|Loss: 0.9677903652191162:  90%|████████▉ | 5815/6470 [1:06:52<08:06,  1.35it/s]
3|5816|Loss: 0.9677903652191162:  90%|████████▉ | 5816/6470 [1:06:52<07:29,  1.45it/s]
3|5817|Loss: 0.8257268071174622:  90%|████████▉ | 5816/6470 [1:06:53<07:29,  1.45it/s]
3|5817|Loss: 0.8257268071174622:  90%|████████▉ | 5817/6470 [1:06:53<07:00,  1.55it/s]
3|5818|Loss: 0.9340975880622864:  90%|████████▉ | 5817/6470 [1:06:53<07:00,  1.55it/s]
3|5818|Loss: 0.9340975880622864:  90%|████████▉ | 5818/6470 [1:06:54<07:19,  1.48it/s]
3|5819|Loss: 0.8753781914710999:  90%|████████▉ | 5818/6470 [1:06:54<07:19,  1.48it/s]
3|5819|Loss: 0.8753781914710999:  90%|████████▉ | 5819/6470 [1:06:54<06:58,  1.56it/s]
3|5820|Loss: 1.0430465936660767:  90%|████████▉ | 5819/6470 [1:06:54<06:58,  1.56it/s]
3|5820|Loss: 1.0430465936660767:  90%|████████▉ | 5820/6470 [1:06:55<06:53,  1.57it/s]
3|5821|Loss: 1.0462746620178223:  90%|████████▉ | 5820/6470 [1:06:55<06:53,  1.57it/s]
3|5821|Loss: 1.0462746620178223:  90%|████████▉ | 5821/6470 [1:06:55<06:51,  1.58it/s]
3|5822|Loss: 0.8647286295890808:  90%|████████▉ | 5821/6470 [1:06:56<06:51,  1.58it/s]
3|5822|Loss: 0.8647286295890808:  90%|████████▉ | 5822/6470 [1:06:56<07:29,  1.44it/s]
3|5823|Loss: 0.8442091941833496:  90%|████████▉ | 5822/6470 [1:06:57<07:29,  1.44it/s]
3|5823|Loss: 0.8442091941833496:  90%|█████████ | 5823/6470 [1:06:57<07:30,  1.44it/s]
3|5824|Loss: 0.7832273244857788:  90%|█████████ | 5823/6470 [1:06:57<07:30,  1.44it/s]
3|5824|Loss: 0.7832273244857788:  90%|█████████ | 5824/6470 [1:06:58<07:44,  1.39it/s]
3|5825|Loss: 0.8871633410453796:  90%|█████████ | 5824/6470 [1:06:58<07:44,  1.39it/s]
3|5825|Loss: 0.8871633410453796:  90%|█████████ | 5825/6470 [1:06:58<07:18,  1.47it/s]
3|5826|Loss: 0.8254615068435669:  90%|█████████ | 5825/6470 [1:06:59<07:18,  1.47it/s]
3|5826|Loss: 0.8254615068435669:  90%|█████████ | 5826/6470 [1:06:59<06:55,  1.55it/s]
3|5827|Loss: 1.0994282960891724:  90%|█████████ | 5826/6470 [1:06:59<06:55,  1.55it/s]
3|5827|Loss: 1.0994282960891724:  90%|█████████ | 5827/6470 [1:06:59<07:01,  1.53it/s]
3|5828|Loss: 0.8745267987251282:  90%|█████████ | 5827/6470 [1:07:00<07:01,  1.53it/s]
3|5828|Loss: 0.8745267987251282:  90%|█████████ | 5828/6470 [1:07:00<07:23,  1.45it/s]
3|5829|Loss: 0.7225711345672607:  90%|█████████ | 5828/6470 [1:07:01<07:23,  1.45it/s]
3|5829|Loss: 0.7225711345672607:  90%|█████████ | 5829/6470 [1:07:01<07:32,  1.42it/s]
3|5830|Loss: 0.7911397218704224:  90%|█████████ | 5829/6470 [1:07:01<07:32,  1.42it/s]
3|5830|Loss: 0.7911397218704224:  90%|█████████ | 5830/6470 [1:07:02<07:58,  1.34it/s]
3|5831|Loss: 0.8243029713630676:  90%|█████████ | 5830/6470 [1:07:02<07:58,  1.34it/s]
3|5831|Loss: 0.8243029713630676:  90%|█████████ | 5831/6470 [1:07:03<08:11,  1.30it/s]
3|5832|Loss: 0.7782472968101501:  90%|█████████ | 5831/6470 [1:07:03<08:11,  1.30it/s]
3|5832|Loss: 0.7782472968101501:  90%|█████████ | 5832/6470 [1:07:03<07:50,  1.36it/s]
3|5833|Loss: 0.9546118974685669:  90%|█████████ | 5832/6470 [1:07:04<07:50,  1.36it/s]
3|5833|Loss: 0.9546118974685669:  90%|█████████ | 5833/6470 [1:07:04<07:45,  1.37it/s]
3|5834|Loss: 0.9069724678993225:  90%|█████████ | 5833/6470 [1:07:04<07:45,  1.37it/s]
3|5834|Loss: 0.9069724678993225:  90%|█████████ | 5834/6470 [1:07:05<08:04,  1.31it/s]
3|5835|Loss: 0.8078620433807373:  90%|█████████ | 5834/6470 [1:07:05<08:04,  1.31it/s]
3|5835|Loss: 0.8078620433807373:  90%|█████████ | 5835/6470 [1:07:06<08:18,  1.27it/s]
3|5836|Loss: 0.6528212428092957:  90%|█████████ | 5835/6470 [1:07:06<08:18,  1.27it/s]
3|5836|Loss: 0.6528212428092957:  90%|█████████ | 5836/6470 [1:07:06<07:54,  1.34it/s]
3|5837|Loss: 0.9319900274276733:  90%|█████████ | 5836/6470 [1:07:07<07:54,  1.34it/s]
3|5837|Loss: 0.9319900274276733:  90%|█████████ | 5837/6470 [1:07:07<08:07,  1.30it/s]
3|5838|Loss: 0.8701339960098267:  90%|█████████ | 5837/6470 [1:07:08<08:07,  1.30it/s]
3|5838|Loss: 0.8701339960098267:  90%|█████████ | 5838/6470 [1:07:08<07:50,  1.34it/s]
3|5839|Loss: 0.8269938230514526:  90%|█████████ | 5838/6470 [1:07:08<07:50,  1.34it/s]
3|5839|Loss: 0.8269938230514526:  90%|█████████ | 5839/6470 [1:07:08<07:06,  1.48it/s]
3|5840|Loss: 0.9190990924835205:  90%|█████████ | 5839/6470 [1:07:09<07:06,  1.48it/s]
3|5840|Loss: 0.9190990924835205:  90%|█████████ | 5840/6470 [1:07:09<06:44,  1.56it/s]
3|5841|Loss: 0.8324703574180603:  90%|█████████ | 5840/6470 [1:07:09<06:44,  1.56it/s]
3|5841|Loss: 0.8324703574180603:  90%|█████████ | 5841/6470 [1:07:10<07:31,  1.39it/s]
3|5842|Loss: 0.8940468430519104:  90%|█████████ | 5841/6470 [1:07:10<07:31,  1.39it/s]
3|5842|Loss: 0.8940468430519104:  90%|█████████ | 5842/6470 [1:07:11<08:04,  1.30it/s]
3|5843|Loss: 0.9083903431892395:  90%|█████████ | 5842/6470 [1:07:11<08:04,  1.30it/s]
3|5843|Loss: 0.9083903431892395:  90%|█████████ | 5843/6470 [1:07:11<07:35,  1.38it/s]
3|5844|Loss: 0.9696868062019348:  90%|█████████ | 5843/6470 [1:07:12<07:35,  1.38it/s]
3|5844|Loss: 0.9696868062019348:  90%|█████████ | 5844/6470 [1:07:12<07:38,  1.37it/s]
3|5845|Loss: 0.790095865726471:  90%|█████████ | 5844/6470 [1:07:13<07:38,  1.37it/s] 
3|5845|Loss: 0.790095865726471:  90%|█████████ | 5845/6470 [1:07:13<07:41,  1.36it/s]
3|5846|Loss: 0.8234097361564636:  90%|█████████ | 5845/6470 [1:07:13<07:41,  1.36it/s]
3|5846|Loss: 0.8234097361564636:  90%|█████████ | 5846/6470 [1:07:14<07:22,  1.41it/s]
3|5847|Loss: 0.8203932642936707:  90%|█████████ | 5846/6470 [1:07:14<07:22,  1.41it/s]
3|5847|Loss: 0.8203932642936707:  90%|█████████ | 5847/6470 [1:07:14<07:21,  1.41it/s]
3|5848|Loss: 0.8165324926376343:  90%|█████████ | 5847/6470 [1:07:15<07:21,  1.41it/s]
3|5848|Loss: 0.8165324926376343:  90%|█████████ | 5848/6470 [1:07:15<07:27,  1.39it/s]
3|5849|Loss: 0.9126963019371033:  90%|█████████ | 5848/6470 [1:07:15<07:27,  1.39it/s]
3|5849|Loss: 0.9126963019371033:  90%|█████████ | 5849/6470 [1:07:16<07:48,  1.33it/s]
3|5850|Loss: 0.8658584952354431:  90%|█████████ | 5849/6470 [1:07:16<07:48,  1.33it/s]
3|5850|Loss: 0.8658584952354431:  90%|█████████ | 5850/6470 [1:07:17<07:50,  1.32it/s]
3|5851|Loss: 0.9358043670654297:  90%|█████████ | 5850/6470 [1:07:17<07:50,  1.32it/s]
3|5851|Loss: 0.9358043670654297:  90%|█████████ | 5851/6470 [1:07:17<07:46,  1.33it/s]
3|5852|Loss: 1.0439658164978027:  90%|█████████ | 5851/6470 [1:07:18<07:46,  1.33it/s]
3|5852|Loss: 1.0439658164978027:  90%|█████████ | 5852/6470 [1:07:18<07:01,  1.47it/s]
3|5853|Loss: 0.7581279277801514:  90%|█████████ | 5852/6470 [1:07:18<07:01,  1.47it/s]
3|5853|Loss: 0.7581279277801514:  90%|█████████ | 5853/6470 [1:07:18<06:25,  1.60it/s]
3|5854|Loss: 0.8280941247940063:  90%|█████████ | 5853/6470 [1:07:19<06:25,  1.60it/s]
3|5854|Loss: 0.8280941247940063:  90%|█████████ | 5854/6470 [1:07:19<06:55,  1.48it/s]
3|5855|Loss: 0.9439652562141418:  90%|█████████ | 5854/6470 [1:07:20<06:55,  1.48it/s]
3|5855|Loss: 0.9439652562141418:  90%|█████████ | 5855/6470 [1:07:20<07:28,  1.37it/s]
3|5856|Loss: 0.8103659749031067:  90%|█████████ | 5855/6470 [1:07:20<07:28,  1.37it/s]
3|5856|Loss: 0.8103659749031067:  91%|█████████ | 5856/6470 [1:07:21<07:33,  1.35it/s]
3|5857|Loss: 0.7044920921325684:  91%|█████████ | 5856/6470 [1:07:21<07:33,  1.35it/s]
3|5857|Loss: 0.7044920921325684:  91%|█████████ | 5857/6470 [1:07:22<07:43,  1.32it/s]
3|5858|Loss: 0.8073659539222717:  91%|█████████ | 5857/6470 [1:07:22<07:43,  1.32it/s]
3|5858|Loss: 0.8073659539222717:  91%|█████████ | 5858/6470 [1:07:22<07:36,  1.34it/s]
3|5859|Loss: 0.6451271176338196:  91%|█████████ | 5858/6470 [1:07:23<07:36,  1.34it/s]
3|5859|Loss: 0.6451271176338196:  91%|█████████ | 5859/6470 [1:07:23<07:18,  1.39it/s]
3|5860|Loss: 0.9034105539321899:  91%|█████████ | 5859/6470 [1:07:23<07:18,  1.39it/s]
3|5860|Loss: 0.9034105539321899:  91%|█████████ | 5860/6470 [1:07:24<07:16,  1.40it/s]
3|5861|Loss: 0.8678297996520996:  91%|█████████ | 5860/6470 [1:07:24<07:16,  1.40it/s]
3|5861|Loss: 0.8678297996520996:  91%|█████████ | 5861/6470 [1:07:24<07:05,  1.43it/s]
3|5862|Loss: 0.7159582376480103:  91%|█████████ | 5861/6470 [1:07:25<07:05,  1.43it/s]
3|5862|Loss: 0.7159582376480103:  91%|█████████ | 5862/6470 [1:07:25<07:23,  1.37it/s]
3|5863|Loss: 0.7612062692642212:  91%|█████████ | 5862/6470 [1:07:26<07:23,  1.37it/s]
3|5863|Loss: 0.7612062692642212:  91%|█████████ | 5863/6470 [1:07:26<07:53,  1.28it/s]
3|5864|Loss: 0.8218204379081726:  91%|█████████ | 5863/6470 [1:07:26<07:53,  1.28it/s]
3|5864|Loss: 0.8218204379081726:  91%|█████████ | 5864/6470 [1:07:27<08:00,  1.26it/s]
3|5865|Loss: 1.077357530593872:  91%|█████████ | 5864/6470 [1:07:27<08:00,  1.26it/s] 
3|5865|Loss: 1.077357530593872:  91%|█████████ | 5865/6470 [1:07:27<07:20,  1.37it/s]
3|5866|Loss: 0.9325808882713318:  91%|█████████ | 5865/6470 [1:07:28<07:20,  1.37it/s]
3|5866|Loss: 0.9325808882713318:  91%|█████████ | 5866/6470 [1:07:28<06:55,  1.45it/s]
3|5867|Loss: 0.6799401640892029:  91%|█████████ | 5866/6470 [1:07:28<06:55,  1.45it/s]
3|5867|Loss: 0.6799401640892029:  91%|█████████ | 5867/6470 [1:07:29<07:11,  1.40it/s]
3|5868|Loss: 0.7645701766014099:  91%|█████████ | 5867/6470 [1:07:29<07:11,  1.40it/s]
3|5868|Loss: 0.7645701766014099:  91%|█████████ | 5868/6470 [1:07:29<06:46,  1.48it/s]
3|5869|Loss: 0.9221740365028381:  91%|█████████ | 5868/6470 [1:07:30<06:46,  1.48it/s]
3|5869|Loss: 0.9221740365028381:  91%|█████████ | 5869/6470 [1:07:30<07:06,  1.41it/s]
3|5870|Loss: 0.9936054348945618:  91%|█████████ | 5869/6470 [1:07:31<07:06,  1.41it/s]
3|5870|Loss: 0.9936054348945618:  91%|█████████ | 5870/6470 [1:07:31<07:41,  1.30it/s]
3|5871|Loss: 0.8444037437438965:  91%|█████████ | 5870/6470 [1:07:31<07:41,  1.30it/s]
3|5871|Loss: 0.8444037437438965:  91%|█████████ | 5871/6470 [1:07:32<07:50,  1.27it/s]
3|5872|Loss: 0.997829258441925:  91%|█████████ | 5871/6470 [1:07:32<07:50,  1.27it/s] 
3|5872|Loss: 0.997829258441925:  91%|█████████ | 5872/6470 [1:07:32<07:26,  1.34it/s]
3|5873|Loss: 0.8366720080375671:  91%|█████████ | 5872/6470 [1:07:33<07:26,  1.34it/s]
3|5873|Loss: 0.8366720080375671:  91%|█████████ | 5873/6470 [1:07:33<06:42,  1.48it/s]
3|5874|Loss: 0.8854559659957886:  91%|█████████ | 5873/6470 [1:07:33<06:42,  1.48it/s]
3|5874|Loss: 0.8854559659957886:  91%|█████████ | 5874/6470 [1:07:34<07:03,  1.41it/s]
3|5875|Loss: 0.7868596315383911:  91%|█████████ | 5874/6470 [1:07:34<07:03,  1.41it/s]
3|5875|Loss: 0.7868596315383911:  91%|█████████ | 5875/6470 [1:07:35<07:29,  1.32it/s]
3|5876|Loss: 0.7577414512634277:  91%|█████████ | 5875/6470 [1:07:35<07:29,  1.32it/s]
3|5876|Loss: 0.7577414512634277:  91%|█████████ | 5876/6470 [1:07:35<07:31,  1.32it/s]
3|5877|Loss: 0.8964932560920715:  91%|█████████ | 5876/6470 [1:07:36<07:31,  1.32it/s]
3|5877|Loss: 0.8964932560920715:  91%|█████████ | 5877/6470 [1:07:36<07:47,  1.27it/s]
3|5878|Loss: 0.8295469880104065:  91%|█████████ | 5877/6470 [1:07:37<07:47,  1.27it/s]
3|5878|Loss: 0.8295469880104065:  91%|█████████ | 5878/6470 [1:07:37<07:30,  1.31it/s]
3|5879|Loss: 1.0196430683135986:  91%|█████████ | 5878/6470 [1:07:37<07:30,  1.31it/s]
3|5879|Loss: 1.0196430683135986:  91%|█████████ | 5879/6470 [1:07:38<07:09,  1.38it/s]
3|5880|Loss: 0.8516834378242493:  91%|█████████ | 5879/6470 [1:07:38<07:09,  1.38it/s]
3|5880|Loss: 0.8516834378242493:  91%|█████████ | 5880/6470 [1:07:38<07:02,  1.40it/s]
3|5881|Loss: 1.021511197090149:  91%|█████████ | 5880/6470 [1:07:39<07:02,  1.40it/s] 
3|5881|Loss: 1.021511197090149:  91%|█████████ | 5881/6470 [1:07:39<07:24,  1.33it/s]
3|5882|Loss: 0.9441297650337219:  91%|█████████ | 5881/6470 [1:07:40<07:24,  1.33it/s]
3|5882|Loss: 0.9441297650337219:  91%|█████████ | 5882/6470 [1:07:40<07:22,  1.33it/s]
3|5883|Loss: 0.7805519700050354:  91%|█████████ | 5882/6470 [1:07:40<07:22,  1.33it/s]
3|5883|Loss: 0.7805519700050354:  91%|█████████ | 5883/6470 [1:07:41<06:57,  1.41it/s]
3|5884|Loss: 0.7737455368041992:  91%|█████████ | 5883/6470 [1:07:41<06:57,  1.41it/s]
3|5884|Loss: 0.7737455368041992:  91%|█████████ | 5884/6470 [1:07:41<06:26,  1.52it/s]
3|5885|Loss: 0.9154568910598755:  91%|█████████ | 5884/6470 [1:07:41<06:26,  1.52it/s]
3|5885|Loss: 0.9154568910598755:  91%|█████████ | 5885/6470 [1:07:42<06:50,  1.42it/s]
3|5886|Loss: 0.9365459680557251:  91%|█████████ | 5885/6470 [1:07:42<06:50,  1.42it/s]
3|5886|Loss: 0.9365459680557251:  91%|█████████ | 5886/6470 [1:07:42<06:30,  1.50it/s]
3|5887|Loss: 0.7511492967605591:  91%|█████████ | 5886/6470 [1:07:43<06:30,  1.50it/s]
3|5887|Loss: 0.7511492967605591:  91%|█████████ | 5887/6470 [1:07:43<06:32,  1.48it/s]
3|5888|Loss: 0.9435613751411438:  91%|█████████ | 5887/6470 [1:07:44<06:32,  1.48it/s]
3|5888|Loss: 0.9435613751411438:  91%|█████████ | 5888/6470 [1:07:44<07:06,  1.36it/s]
3|5889|Loss: 0.727105438709259:  91%|█████████ | 5888/6470 [1:07:44<07:06,  1.36it/s] 
3|5889|Loss: 0.727105438709259:  91%|█████████ | 5889/6470 [1:07:45<07:30,  1.29it/s]
3|5890|Loss: 0.8690961599349976:  91%|█████████ | 5889/6470 [1:07:45<07:30,  1.29it/s]
3|5890|Loss: 0.8690961599349976:  91%|█████████ | 5890/6470 [1:07:46<07:12,  1.34it/s]
3|5891|Loss: 0.8507837057113647:  91%|█████████ | 5890/6470 [1:07:46<07:12,  1.34it/s]
3|5891|Loss: 0.8507837057113647:  91%|█████████ | 5891/6470 [1:07:46<06:49,  1.41it/s]
3|5892|Loss: 0.8807404041290283:  91%|█████████ | 5891/6470 [1:07:47<06:49,  1.41it/s]
3|5892|Loss: 0.8807404041290283:  91%|█████████ | 5892/6470 [1:07:47<06:29,  1.48it/s]
3|5893|Loss: 0.7849454879760742:  91%|█████████ | 5892/6470 [1:07:47<06:29,  1.48it/s]
3|5893|Loss: 0.7849454879760742:  91%|█████████ | 5893/6470 [1:07:47<06:14,  1.54it/s]
3|5894|Loss: 0.8300200700759888:  91%|█████████ | 5893/6470 [1:07:48<06:14,  1.54it/s]
3|5894|Loss: 0.8300200700759888:  91%|█████████ | 5894/6470 [1:07:48<06:19,  1.52it/s]
3|5895|Loss: 0.6846566200256348:  91%|█████████ | 5894/6470 [1:07:48<06:19,  1.52it/s]
3|5895|Loss: 0.6846566200256348:  91%|█████████ | 5895/6470 [1:07:49<06:18,  1.52it/s]
3|5896|Loss: 0.8164128065109253:  91%|█████████ | 5895/6470 [1:07:49<06:18,  1.52it/s]
3|5896|Loss: 0.8164128065109253:  91%|█████████ | 5896/6470 [1:07:49<06:29,  1.47it/s]
3|5897|Loss: 0.8008764386177063:  91%|█████████ | 5896/6470 [1:07:50<06:29,  1.47it/s]
3|5897|Loss: 0.8008764386177063:  91%|█████████ | 5897/6470 [1:07:50<06:57,  1.37it/s]
3|5898|Loss: 0.7788066864013672:  91%|█████████ | 5897/6470 [1:07:51<06:57,  1.37it/s]
3|5898|Loss: 0.7788066864013672:  91%|█████████ | 5898/6470 [1:07:51<06:47,  1.40it/s]
3|5899|Loss: 0.8118283152580261:  91%|█████████ | 5898/6470 [1:07:51<06:47,  1.40it/s]
3|5899|Loss: 0.8118283152580261:  91%|█████████ | 5899/6470 [1:07:52<06:26,  1.48it/s]
3|5900|Loss: 0.9566590785980225:  91%|█████████ | 5899/6470 [1:07:52<06:26,  1.48it/s]
3|5900|Loss: 0.9566590785980225:  91%|█████████ | 5900/6470 [1:07:52<06:18,  1.51it/s]
3|5901|Loss: 0.7427078485488892:  91%|█████████ | 5900/6470 [1:07:53<06:18,  1.51it/s]
3|5901|Loss: 0.7427078485488892:  91%|█████████ | 5901/6470 [1:07:53<06:45,  1.40it/s]
3|5902|Loss: 0.9285697340965271:  91%|█████████ | 5901/6470 [1:07:53<06:45,  1.40it/s]
3|5902|Loss: 0.9285697340965271:  91%|█████████ | 5902/6470 [1:07:54<06:30,  1.45it/s]
3|5903|Loss: 0.8034064173698425:  91%|█████████ | 5902/6470 [1:07:54<06:30,  1.45it/s]
3|5903|Loss: 0.8034064173698425:  91%|█████████ | 5903/6470 [1:07:54<06:16,  1.51it/s]
3|5904|Loss: 0.778915286064148:  91%|█████████ | 5903/6470 [1:07:55<06:16,  1.51it/s] 
3|5904|Loss: 0.778915286064148:  91%|█████████▏| 5904/6470 [1:07:55<06:11,  1.53it/s]
3|5905|Loss: 0.8696911334991455:  91%|█████████▏| 5904/6470 [1:07:55<06:11,  1.53it/s]
3|5905|Loss: 0.8696911334991455:  91%|█████████▏| 5905/6470 [1:07:56<06:46,  1.39it/s]
3|5906|Loss: 0.8949053883552551:  91%|█████████▏| 5905/6470 [1:07:56<06:46,  1.39it/s]
3|5906|Loss: 0.8949053883552551:  91%|█████████▏| 5906/6470 [1:07:56<06:50,  1.37it/s]
3|5907|Loss: 0.8592891097068787:  91%|█████████▏| 5906/6470 [1:07:57<06:50,  1.37it/s]
3|5907|Loss: 0.8592891097068787:  91%|█████████▏| 5907/6470 [1:07:57<06:56,  1.35it/s]
3|5908|Loss: 0.7751680612564087:  91%|█████████▏| 5907/6470 [1:07:58<06:56,  1.35it/s]
3|5908|Loss: 0.7751680612564087:  91%|█████████▏| 5908/6470 [1:07:58<06:44,  1.39it/s]
3|5909|Loss: 0.8771256804466248:  91%|█████████▏| 5908/6470 [1:07:58<06:44,  1.39it/s]
3|5909|Loss: 0.8771256804466248:  91%|█████████▏| 5909/6470 [1:07:59<06:40,  1.40it/s]
3|5910|Loss: 0.7928691506385803:  91%|█████████▏| 5909/6470 [1:07:59<06:40,  1.40it/s]
3|5910|Loss: 0.7928691506385803:  91%|█████████▏| 5910/6470 [1:07:59<06:29,  1.44it/s]
3|5911|Loss: 0.9640502333641052:  91%|█████████▏| 5910/6470 [1:08:00<06:29,  1.44it/s]
3|5911|Loss: 0.9640502333641052:  91%|█████████▏| 5911/6470 [1:08:00<06:32,  1.42it/s]
3|5912|Loss: 0.9681625366210938:  91%|█████████▏| 5911/6470 [1:08:00<06:32,  1.42it/s]
3|5912|Loss: 0.9681625366210938:  91%|█████████▏| 5912/6470 [1:08:01<06:50,  1.36it/s]
3|5913|Loss: 0.7676442861557007:  91%|█████████▏| 5912/6470 [1:08:01<06:50,  1.36it/s]
3|5913|Loss: 0.7676442861557007:  91%|█████████▏| 5913/6470 [1:08:02<06:53,  1.35it/s]
3|5914|Loss: 0.5493940711021423:  91%|█████████▏| 5913/6470 [1:08:02<06:53,  1.35it/s]
3|5914|Loss: 0.5493940711021423:  91%|█████████▏| 5914/6470 [1:08:02<06:58,  1.33it/s]
3|5915|Loss: 0.7530494928359985:  91%|█████████▏| 5914/6470 [1:08:03<06:58,  1.33it/s]
3|5915|Loss: 0.7530494928359985:  91%|█████████▏| 5915/6470 [1:08:03<06:41,  1.38it/s]
3|5916|Loss: 0.7770147919654846:  91%|█████████▏| 5915/6470 [1:08:03<06:41,  1.38it/s]
3|5916|Loss: 0.7770147919654846:  91%|█████████▏| 5916/6470 [1:08:04<06:34,  1.41it/s]
3|5917|Loss: 0.8237338662147522:  91%|█████████▏| 5916/6470 [1:08:04<06:34,  1.41it/s]
3|5917|Loss: 0.8237338662147522:  91%|█████████▏| 5917/6470 [1:08:04<06:35,  1.40it/s]
3|5918|Loss: 0.7716479897499084:  91%|█████████▏| 5917/6470 [1:08:05<06:35,  1.40it/s]
3|5918|Loss: 0.7716479897499084:  91%|█████████▏| 5918/6470 [1:08:05<06:48,  1.35it/s]
3|5919|Loss: 0.9051000475883484:  91%|█████████▏| 5918/6470 [1:08:06<06:48,  1.35it/s]
3|5919|Loss: 0.9051000475883484:  91%|█████████▏| 5919/6470 [1:08:06<06:37,  1.39it/s]
3|5920|Loss: 1.0459164381027222:  91%|█████████▏| 5919/6470 [1:08:06<06:37,  1.39it/s]
3|5920|Loss: 1.0459164381027222:  91%|█████████▏| 5920/6470 [1:08:07<06:48,  1.35it/s]
3|5921|Loss: 0.7679823637008667:  91%|█████████▏| 5920/6470 [1:08:07<06:48,  1.35it/s]
3|5921|Loss: 0.7679823637008667:  92%|█████████▏| 5921/6470 [1:08:07<06:38,  1.38it/s]
3|5922|Loss: 0.8498455286026001:  92%|█████████▏| 5921/6470 [1:08:08<06:38,  1.38it/s]
3|5922|Loss: 0.8498455286026001:  92%|█████████▏| 5922/6470 [1:08:08<06:22,  1.43it/s]
3|5923|Loss: 0.99330735206604:  92%|█████████▏| 5922/6470 [1:08:08<06:22,  1.43it/s]  
3|5923|Loss: 0.99330735206604:  92%|█████████▏| 5923/6470 [1:08:09<06:39,  1.37it/s]
3|5924|Loss: 0.8277057409286499:  92%|█████████▏| 5923/6470 [1:08:09<06:39,  1.37it/s]
3|5924|Loss: 0.8277057409286499:  92%|█████████▏| 5924/6470 [1:08:10<06:46,  1.34it/s]
3|5925|Loss: 0.7259768843650818:  92%|█████████▏| 5924/6470 [1:08:10<06:46,  1.34it/s]
3|5925|Loss: 0.7259768843650818:  92%|█████████▏| 5925/6470 [1:08:10<06:40,  1.36it/s]
3|5926|Loss: 0.8766594529151917:  92%|█████████▏| 5925/6470 [1:08:11<06:40,  1.36it/s]
3|5926|Loss: 0.8766594529151917:  92%|█████████▏| 5926/6470 [1:08:11<06:27,  1.40it/s]
3|5927|Loss: 0.8878114819526672:  92%|█████████▏| 5926/6470 [1:08:11<06:27,  1.40it/s]
3|5927|Loss: 0.8878114819526672:  92%|█████████▏| 5927/6470 [1:08:11<05:58,  1.52it/s]
3|5928|Loss: 0.856820821762085:  92%|█████████▏| 5927/6470 [1:08:12<05:58,  1.52it/s] 
3|5928|Loss: 0.856820821762085:  92%|█████████▏| 5928/6470 [1:08:12<05:16,  1.71it/s]
3|5929|Loss: 0.8627354502677917:  92%|█████████▏| 5928/6470 [1:08:12<05:16,  1.71it/s]
3|5929|Loss: 0.8627354502677917:  92%|█████████▏| 5929/6470 [1:08:12<05:07,  1.76it/s]
3|5930|Loss: 0.6673929691314697:  92%|█████████▏| 5929/6470 [1:08:13<05:07,  1.76it/s]
3|5930|Loss: 0.6673929691314697:  92%|█████████▏| 5930/6470 [1:08:13<04:52,  1.85it/s]
3|5931|Loss: 0.6375641226768494:  92%|█████████▏| 5930/6470 [1:08:13<04:52,  1.85it/s]
3|5931|Loss: 0.6375641226768494:  92%|█████████▏| 5931/6470 [1:08:13<04:52,  1.84it/s]
3|5932|Loss: 0.8930801153182983:  92%|█████████▏| 5931/6470 [1:08:14<04:52,  1.84it/s]
3|5932|Loss: 0.8930801153182983:  92%|█████████▏| 5932/6470 [1:08:14<05:27,  1.64it/s]
3|5933|Loss: 0.9719418287277222:  92%|█████████▏| 5932/6470 [1:08:15<05:27,  1.64it/s]
3|5933|Loss: 0.9719418287277222:  92%|█████████▏| 5933/6470 [1:08:15<05:34,  1.61it/s]
3|5934|Loss: 0.891721248626709:  92%|█████████▏| 5933/6470 [1:08:15<05:34,  1.61it/s] 
3|5934|Loss: 0.891721248626709:  92%|█████████▏| 5934/6470 [1:08:15<05:28,  1.63it/s]
3|5935|Loss: 0.9485961198806763:  92%|█████████▏| 5934/6470 [1:08:16<05:28,  1.63it/s]
3|5935|Loss: 0.9485961198806763:  92%|█████████▏| 5935/6470 [1:08:16<05:14,  1.70it/s]
3|5936|Loss: 0.9542067646980286:  92%|█████████▏| 5935/6470 [1:08:16<05:14,  1.70it/s]
3|5936|Loss: 0.9542067646980286:  92%|█████████▏| 5936/6470 [1:08:17<05:29,  1.62it/s]
3|5937|Loss: 0.9973532557487488:  92%|█████████▏| 5936/6470 [1:08:17<05:29,  1.62it/s]
3|5937|Loss: 0.9973532557487488:  92%|█████████▏| 5937/6470 [1:08:17<05:43,  1.55it/s]
3|5938|Loss: 0.8668643236160278:  92%|█████████▏| 5937/6470 [1:08:18<05:43,  1.55it/s]
3|5938|Loss: 0.8668643236160278:  92%|█████████▏| 5938/6470 [1:08:18<06:04,  1.46it/s]
3|5939|Loss: 0.8837194442749023:  92%|█████████▏| 5938/6470 [1:08:19<06:04,  1.46it/s]
3|5939|Loss: 0.8837194442749023:  92%|█████████▏| 5939/6470 [1:08:19<05:54,  1.50it/s]
3|5940|Loss: 0.9718309640884399:  92%|█████████▏| 5939/6470 [1:08:19<05:54,  1.50it/s]
3|5940|Loss: 0.9718309640884399:  92%|█████████▏| 5940/6470 [1:08:20<06:00,  1.47it/s]
3|5941|Loss: 0.9499717354774475:  92%|█████████▏| 5940/6470 [1:08:20<06:00,  1.47it/s]
3|5941|Loss: 0.9499717354774475:  92%|█████████▏| 5941/6470 [1:08:20<06:21,  1.38it/s]
3|5942|Loss: 0.7922982573509216:  92%|█████████▏| 5941/6470 [1:08:21<06:21,  1.38it/s]
3|5942|Loss: 0.7922982573509216:  92%|█████████▏| 5942/6470 [1:08:21<05:50,  1.50it/s]
3|5943|Loss: 0.7401347756385803:  92%|█████████▏| 5942/6470 [1:08:21<05:50,  1.50it/s]
3|5943|Loss: 0.7401347756385803:  92%|█████████▏| 5943/6470 [1:08:22<06:03,  1.45it/s]
3|5944|Loss: 1.022596836090088:  92%|█████████▏| 5943/6470 [1:08:22<06:03,  1.45it/s] 
3|5944|Loss: 1.022596836090088:  92%|█████████▏| 5944/6470 [1:08:22<06:25,  1.37it/s]
3|5945|Loss: 0.9639644622802734:  92%|█████████▏| 5944/6470 [1:08:23<06:25,  1.37it/s]
3|5945|Loss: 0.9639644622802734:  92%|█████████▏| 5945/6470 [1:08:23<06:27,  1.35it/s]
3|5946|Loss: 1.16544771194458:  92%|█████████▏| 5945/6470 [1:08:24<06:27,  1.35it/s]  
3|5946|Loss: 1.16544771194458:  92%|█████████▏| 5946/6470 [1:08:24<06:42,  1.30it/s]
3|5947|Loss: 0.9867485761642456:  92%|█████████▏| 5946/6470 [1:08:24<06:42,  1.30it/s]
3|5947|Loss: 0.9867485761642456:  92%|█████████▏| 5947/6470 [1:08:25<06:29,  1.34it/s]
3|5948|Loss: 0.9521284699440002:  92%|█████████▏| 5947/6470 [1:08:25<06:29,  1.34it/s]
3|5948|Loss: 0.9521284699440002:  92%|█████████▏| 5948/6470 [1:08:25<05:49,  1.49it/s]
3|5949|Loss: 0.8053604364395142:  92%|█████████▏| 5948/6470 [1:08:26<05:49,  1.49it/s]
3|5949|Loss: 0.8053604364395142:  92%|█████████▏| 5949/6470 [1:08:26<06:04,  1.43it/s]
3|5950|Loss: 0.8869266510009766:  92%|█████████▏| 5949/6470 [1:08:26<06:04,  1.43it/s]
3|5950|Loss: 0.8869266510009766:  92%|█████████▏| 5950/6470 [1:08:27<06:29,  1.34it/s]
3|5951|Loss: 0.8498296141624451:  92%|█████████▏| 5950/6470 [1:08:27<06:29,  1.34it/s]
3|5951|Loss: 0.8498296141624451:  92%|█████████▏| 5951/6470 [1:08:28<06:30,  1.33it/s]
3|5952|Loss: 0.9019834399223328:  92%|█████████▏| 5951/6470 [1:08:28<06:30,  1.33it/s]
3|5952|Loss: 0.9019834399223328:  92%|█████████▏| 5952/6470 [1:08:28<06:41,  1.29it/s]
3|5953|Loss: 0.9081060886383057:  92%|█████████▏| 5952/6470 [1:08:29<06:41,  1.29it/s]
3|5953|Loss: 0.9081060886383057:  92%|█████████▏| 5953/6470 [1:08:29<06:25,  1.34it/s]
3|5954|Loss: 0.866718053817749:  92%|█████████▏| 5953/6470 [1:08:30<06:25,  1.34it/s] 
3|5954|Loss: 0.866718053817749:  92%|█████████▏| 5954/6470 [1:08:30<06:30,  1.32it/s]
3|5955|Loss: 0.8417500853538513:  92%|█████████▏| 5954/6470 [1:08:30<06:30,  1.32it/s]
3|5955|Loss: 0.8417500853538513:  92%|█████████▏| 5955/6470 [1:08:31<06:18,  1.36it/s]
3|5956|Loss: 0.9677504897117615:  92%|█████████▏| 5955/6470 [1:08:31<06:18,  1.36it/s]
3|5956|Loss: 0.9677504897117615:  92%|█████████▏| 5956/6470 [1:08:31<06:26,  1.33it/s]
3|5957|Loss: 0.8600696921348572:  92%|█████████▏| 5956/6470 [1:08:32<06:26,  1.33it/s]
3|5957|Loss: 0.8600696921348572:  92%|█████████▏| 5957/6470 [1:08:32<06:15,  1.37it/s]
3|5958|Loss: 0.8818885087966919:  92%|█████████▏| 5957/6470 [1:08:32<06:15,  1.37it/s]
3|5958|Loss: 0.8818885087966919:  92%|█████████▏| 5958/6470 [1:08:33<06:13,  1.37it/s]
3|5959|Loss: 0.7504556775093079:  92%|█████████▏| 5958/6470 [1:08:33<06:13,  1.37it/s]
3|5959|Loss: 0.7504556775093079:  92%|█████████▏| 5959/6470 [1:08:33<05:29,  1.55it/s]
3|5960|Loss: 0.7456209659576416:  92%|█████████▏| 5959/6470 [1:08:34<05:29,  1.55it/s]
3|5960|Loss: 0.7456209659576416:  92%|█████████▏| 5960/6470 [1:08:34<05:45,  1.48it/s]
3|5961|Loss: 0.831558346748352:  92%|█████████▏| 5960/6470 [1:08:34<05:45,  1.48it/s] 
3|5961|Loss: 0.831558346748352:  92%|█████████▏| 5961/6470 [1:08:35<05:44,  1.48it/s]
3|5962|Loss: 0.8343189358711243:  92%|█████████▏| 5961/6470 [1:08:35<05:44,  1.48it/s]
3|5962|Loss: 0.8343189358711243:  92%|█████████▏| 5962/6470 [1:08:35<05:50,  1.45it/s]
3|5963|Loss: 0.7453413009643555:  92%|█████████▏| 5962/6470 [1:08:36<05:50,  1.45it/s]
3|5963|Loss: 0.7453413009643555:  92%|█████████▏| 5963/6470 [1:08:36<05:46,  1.47it/s]
3|5964|Loss: 0.926540732383728:  92%|█████████▏| 5963/6470 [1:08:36<05:46,  1.47it/s] 
3|5964|Loss: 0.926540732383728:  92%|█████████▏| 5964/6470 [1:08:37<05:52,  1.44it/s]
3|5965|Loss: 0.8527539372444153:  92%|█████████▏| 5964/6470 [1:08:37<05:52,  1.44it/s]
3|5965|Loss: 0.8527539372444153:  92%|█████████▏| 5965/6470 [1:08:38<06:04,  1.39it/s]
3|5966|Loss: 0.7653855681419373:  92%|█████████▏| 5965/6470 [1:08:38<06:04,  1.39it/s]
3|5966|Loss: 0.7653855681419373:  92%|█████████▏| 5966/6470 [1:08:38<05:54,  1.42it/s]
3|5967|Loss: 0.8327524662017822:  92%|█████████▏| 5966/6470 [1:08:39<05:54,  1.42it/s]
3|5967|Loss: 0.8327524662017822:  92%|█████████▏| 5967/6470 [1:08:39<05:55,  1.41it/s]
3|5968|Loss: 0.8690114617347717:  92%|█████████▏| 5967/6470 [1:08:39<05:55,  1.41it/s]
3|5968|Loss: 0.8690114617347717:  92%|█████████▏| 5968/6470 [1:08:40<05:50,  1.43it/s]
3|5969|Loss: 0.9570653438568115:  92%|█████████▏| 5968/6470 [1:08:40<05:50,  1.43it/s]
3|5969|Loss: 0.9570653438568115:  92%|█████████▏| 5969/6470 [1:08:40<05:54,  1.41it/s]
3|5970|Loss: 0.8292772173881531:  92%|█████████▏| 5969/6470 [1:08:41<05:54,  1.41it/s]
3|5970|Loss: 0.8292772173881531:  92%|█████████▏| 5970/6470 [1:08:41<05:47,  1.44it/s]
3|5971|Loss: 0.8911021947860718:  92%|█████████▏| 5970/6470 [1:08:41<05:47,  1.44it/s]
3|5971|Loss: 0.8911021947860718:  92%|█████████▏| 5971/6470 [1:08:42<06:02,  1.38it/s]
3|5972|Loss: 0.8745000958442688:  92%|█████████▏| 5971/6470 [1:08:42<06:02,  1.38it/s]
3|5972|Loss: 0.8745000958442688:  92%|█████████▏| 5972/6470 [1:08:43<06:17,  1.32it/s]
3|5973|Loss: 0.9194454550743103:  92%|█████████▏| 5972/6470 [1:08:43<06:17,  1.32it/s]
3|5973|Loss: 0.9194454550743103:  92%|█████████▏| 5973/6470 [1:08:43<06:16,  1.32it/s]
3|5974|Loss: 0.73177570104599:  92%|█████████▏| 5973/6470 [1:08:44<06:16,  1.32it/s]  
3|5974|Loss: 0.73177570104599:  92%|█████████▏| 5974/6470 [1:08:44<05:58,  1.38it/s]
3|5975|Loss: 0.7781137824058533:  92%|█████████▏| 5974/6470 [1:08:44<05:58,  1.38it/s]
3|5975|Loss: 0.7781137824058533:  92%|█████████▏| 5975/6470 [1:08:45<05:43,  1.44it/s]
3|5976|Loss: 0.9795777201652527:  92%|█████████▏| 5975/6470 [1:08:45<05:43,  1.44it/s]
3|5976|Loss: 0.9795777201652527:  92%|█████████▏| 5976/6470 [1:08:45<05:53,  1.40it/s]
3|5977|Loss: 0.935567319393158:  92%|█████████▏| 5976/6470 [1:08:46<05:53,  1.40it/s] 
3|5977|Loss: 0.935567319393158:  92%|█████████▏| 5977/6470 [1:08:46<05:57,  1.38it/s]
3|5978|Loss: 0.8980954885482788:  92%|█████████▏| 5977/6470 [1:08:47<05:57,  1.38it/s]
3|5978|Loss: 0.8980954885482788:  92%|█████████▏| 5978/6470 [1:08:47<05:59,  1.37it/s]
3|5979|Loss: 0.8067038655281067:  92%|█████████▏| 5978/6470 [1:08:47<05:59,  1.37it/s]
3|5979|Loss: 0.8067038655281067:  92%|█████████▏| 5979/6470 [1:08:48<06:14,  1.31it/s]
3|5980|Loss: 0.7730090618133545:  92%|█████████▏| 5979/6470 [1:08:48<06:14,  1.31it/s]
3|5980|Loss: 0.7730090618133545:  92%|█████████▏| 5980/6470 [1:08:49<06:21,  1.28it/s]
3|5981|Loss: 0.9292148947715759:  92%|█████████▏| 5980/6470 [1:08:49<06:21,  1.28it/s]
3|5981|Loss: 0.9292148947715759:  92%|█████████▏| 5981/6470 [1:08:49<06:16,  1.30it/s]
3|5982|Loss: 0.7418437600135803:  92%|█████████▏| 5981/6470 [1:08:50<06:16,  1.30it/s]
3|5982|Loss: 0.7418437600135803:  92%|█████████▏| 5982/6470 [1:08:50<06:18,  1.29it/s]
3|5983|Loss: 1.0089856386184692:  92%|█████████▏| 5982/6470 [1:08:51<06:18,  1.29it/s]
3|5983|Loss: 1.0089856386184692:  92%|█████████▏| 5983/6470 [1:08:51<06:14,  1.30it/s]
3|5984|Loss: 0.7839692234992981:  92%|█████████▏| 5983/6470 [1:08:51<06:14,  1.30it/s]
3|5984|Loss: 0.7839692234992981:  92%|█████████▏| 5984/6470 [1:08:52<06:10,  1.31it/s]
3|5985|Loss: 0.8879550099372864:  92%|█████████▏| 5984/6470 [1:08:52<06:10,  1.31it/s]
3|5985|Loss: 0.8879550099372864:  93%|█████████▎| 5985/6470 [1:08:52<05:48,  1.39it/s]
3|5986|Loss: 0.8589963316917419:  93%|█████████▎| 5985/6470 [1:08:53<05:48,  1.39it/s]
3|5986|Loss: 0.8589963316917419:  93%|█████████▎| 5986/6470 [1:08:53<05:43,  1.41it/s]
3|5987|Loss: 0.8032837510108948:  93%|█████████▎| 5986/6470 [1:08:53<05:43,  1.41it/s]
3|5987|Loss: 0.8032837510108948:  93%|█████████▎| 5987/6470 [1:08:54<05:54,  1.36it/s]
3|5988|Loss: 0.9101648926734924:  93%|█████████▎| 5987/6470 [1:08:54<05:54,  1.36it/s]
3|5988|Loss: 0.9101648926734924:  93%|█████████▎| 5988/6470 [1:08:54<05:43,  1.40it/s]
3|5989|Loss: 0.8516407608985901:  93%|█████████▎| 5988/6470 [1:08:55<05:43,  1.40it/s]
3|5989|Loss: 0.8516407608985901:  93%|█████████▎| 5989/6470 [1:08:55<05:18,  1.51it/s]
3|5990|Loss: 0.8773340582847595:  93%|█████████▎| 5989/6470 [1:08:55<05:18,  1.51it/s]
3|5990|Loss: 0.8773340582847595:  93%|█████████▎| 5990/6470 [1:08:56<05:08,  1.55it/s]
3|5991|Loss: 0.7578743100166321:  93%|█████████▎| 5990/6470 [1:08:56<05:08,  1.55it/s]
3|5991|Loss: 0.7578743100166321:  93%|█████████▎| 5991/6470 [1:08:56<04:56,  1.62it/s]
3|5992|Loss: 0.8263710141181946:  93%|█████████▎| 5991/6470 [1:08:56<04:56,  1.62it/s]
3|5992|Loss: 0.8263710141181946:  93%|█████████▎| 5992/6470 [1:08:57<05:06,  1.56it/s]
3|5993|Loss: 0.9751511216163635:  93%|█████████▎| 5992/6470 [1:08:57<05:06,  1.56it/s]
3|5993|Loss: 0.9751511216163635:  93%|█████████▎| 5993/6470 [1:08:58<05:21,  1.48it/s]
3|5994|Loss: 0.7881944179534912:  93%|█████████▎| 5993/6470 [1:08:58<05:21,  1.48it/s]
3|5994|Loss: 0.7881944179534912:  93%|█████████▎| 5994/6470 [1:08:58<05:17,  1.50it/s]
3|5995|Loss: 0.7098104953765869:  93%|█████████▎| 5994/6470 [1:08:59<05:17,  1.50it/s]
3|5995|Loss: 0.7098104953765869:  93%|█████████▎| 5995/6470 [1:08:59<05:25,  1.46it/s]
3|5996|Loss: 0.9853085875511169:  93%|█████████▎| 5995/6470 [1:08:59<05:25,  1.46it/s]
3|5996|Loss: 0.9853085875511169:  93%|█████████▎| 5996/6470 [1:09:00<05:21,  1.48it/s]
3|5997|Loss: 1.0007919073104858:  93%|█████████▎| 5996/6470 [1:09:00<05:21,  1.48it/s]
3|5997|Loss: 1.0007919073104858:  93%|█████████▎| 5997/6470 [1:09:00<05:31,  1.43it/s]
3|5998|Loss: 0.8347660303115845:  93%|█████████▎| 5997/6470 [1:09:01<05:31,  1.43it/s]
3|5998|Loss: 0.8347660303115845:  93%|█████████▎| 5998/6470 [1:09:01<05:27,  1.44it/s]
3|5999|Loss: 0.7806957364082336:  93%|█████████▎| 5998/6470 [1:09:01<05:27,  1.44it/s]
3|5999|Loss: 0.7806957364082336:  93%|█████████▎| 5999/6470 [1:09:02<05:27,  1.44it/s]
3|6000|Loss: 0.9508340358734131:  93%|█████████▎| 5999/6470 [1:09:02<05:27,  1.44it/s]
3|6000|Loss: 0.9508340358734131:  93%|█████████▎| 6000/6470 [1:09:02<05:22,  1.46it/s]
3|6001|Loss: 0.7694827914237976:  93%|█████████▎| 6000/6470 [1:09:03<05:22,  1.46it/s]
3|6001|Loss: 0.7694827914237976:  93%|█████████▎| 6001/6470 [1:09:03<05:08,  1.52it/s]
3|6002|Loss: 0.8373071551322937:  93%|█████████▎| 6001/6470 [1:09:03<05:08,  1.52it/s]
3|6002|Loss: 0.8373071551322937:  93%|█████████▎| 6002/6470 [1:09:04<05:04,  1.54it/s]
3|6003|Loss: 0.9608430862426758:  93%|█████████▎| 6002/6470 [1:09:04<05:04,  1.54it/s]
3|6003|Loss: 0.9608430862426758:  93%|█████████▎| 6003/6470 [1:09:04<05:13,  1.49it/s]
3|6004|Loss: 0.9008631110191345:  93%|█████████▎| 6003/6470 [1:09:05<05:13,  1.49it/s]
3|6004|Loss: 0.9008631110191345:  93%|█████████▎| 6004/6470 [1:09:05<04:49,  1.61it/s]
3|6005|Loss: 0.8595545291900635:  93%|█████████▎| 6004/6470 [1:09:05<04:49,  1.61it/s]
3|6005|Loss: 0.8595545291900635:  93%|█████████▎| 6005/6470 [1:09:05<04:52,  1.59it/s]
3|6006|Loss: 0.9461256861686707:  93%|█████████▎| 6005/6470 [1:09:06<04:52,  1.59it/s]
3|6006|Loss: 0.9461256861686707:  93%|█████████▎| 6006/6470 [1:09:06<04:53,  1.58it/s]
3|6007|Loss: 0.9300095438957214:  93%|█████████▎| 6006/6470 [1:09:06<04:53,  1.58it/s]
3|6007|Loss: 0.9300095438957214:  93%|█████████▎| 6007/6470 [1:09:07<05:00,  1.54it/s]
3|6008|Loss: 0.7690653800964355:  93%|█████████▎| 6007/6470 [1:09:07<05:00,  1.54it/s]
3|6008|Loss: 0.7690653800964355:  93%|█████████▎| 6008/6470 [1:09:07<05:06,  1.51it/s]
3|6009|Loss: 1.144705057144165:  93%|█████████▎| 6008/6470 [1:09:08<05:06,  1.51it/s] 
3|6009|Loss: 1.144705057144165:  93%|█████████▎| 6009/6470 [1:09:08<04:53,  1.57it/s]
3|6010|Loss: 0.9013338685035706:  93%|█████████▎| 6009/6470 [1:09:08<04:53,  1.57it/s]
3|6010|Loss: 0.9013338685035706:  93%|█████████▎| 6010/6470 [1:09:09<05:11,  1.47it/s]
3|6011|Loss: 0.8340089917182922:  93%|█████████▎| 6010/6470 [1:09:09<05:11,  1.47it/s]
3|6011|Loss: 0.8340089917182922:  93%|█████████▎| 6011/6470 [1:09:09<05:09,  1.48it/s]
3|6012|Loss: 0.9533286690711975:  93%|█████████▎| 6011/6470 [1:09:10<05:09,  1.48it/s]
3|6012|Loss: 0.9533286690711975:  93%|█████████▎| 6012/6470 [1:09:10<05:15,  1.45it/s]
3|6013|Loss: 0.9084540009498596:  93%|█████████▎| 6012/6470 [1:09:11<05:15,  1.45it/s]
3|6013|Loss: 0.9084540009498596:  93%|█████████▎| 6013/6470 [1:09:11<05:01,  1.51it/s]
3|6014|Loss: 0.8950439691543579:  93%|█████████▎| 6013/6470 [1:09:11<05:01,  1.51it/s]
3|6014|Loss: 0.8950439691543579:  93%|█████████▎| 6014/6470 [1:09:12<05:16,  1.44it/s]
3|6015|Loss: 0.8990035057067871:  93%|█████████▎| 6014/6470 [1:09:12<05:16,  1.44it/s]
3|6015|Loss: 0.8990035057067871:  93%|█████████▎| 6015/6470 [1:09:12<04:56,  1.53it/s]
3|6016|Loss: 0.8533163070678711:  93%|█████████▎| 6015/6470 [1:09:12<04:56,  1.53it/s]
3|6016|Loss: 0.8533163070678711:  93%|█████████▎| 6016/6470 [1:09:13<04:47,  1.58it/s]
3|6017|Loss: 0.7306575179100037:  93%|█████████▎| 6016/6470 [1:09:13<04:47,  1.58it/s]
3|6017|Loss: 0.7306575179100037:  93%|█████████▎| 6017/6470 [1:09:13<04:33,  1.66it/s]
3|6018|Loss: 0.8376859426498413:  93%|█████████▎| 6017/6470 [1:09:14<04:33,  1.66it/s]
3|6018|Loss: 0.8376859426498413:  93%|█████████▎| 6018/6470 [1:09:14<04:31,  1.66it/s]
3|6019|Loss: 0.8655471801757812:  93%|█████████▎| 6018/6470 [1:09:14<04:31,  1.66it/s]
3|6019|Loss: 0.8655471801757812:  93%|█████████▎| 6019/6470 [1:09:14<04:32,  1.65it/s]
3|6020|Loss: 0.8585963845252991:  93%|█████████▎| 6019/6470 [1:09:15<04:32,  1.65it/s]
3|6020|Loss: 0.8585963845252991:  93%|█████████▎| 6020/6470 [1:09:15<04:16,  1.76it/s]
3|6021|Loss: 0.9122217297554016:  93%|█████████▎| 6020/6470 [1:09:15<04:16,  1.76it/s]
3|6021|Loss: 0.9122217297554016:  93%|█████████▎| 6021/6470 [1:09:15<04:11,  1.78it/s]
3|6022|Loss: 0.9652259349822998:  93%|█████████▎| 6021/6470 [1:09:16<04:11,  1.78it/s]
3|6022|Loss: 0.9652259349822998:  93%|█████████▎| 6022/6470 [1:09:16<04:08,  1.80it/s]
3|6023|Loss: 0.8016158938407898:  93%|█████████▎| 6022/6470 [1:09:16<04:08,  1.80it/s]
3|6023|Loss: 0.8016158938407898:  93%|█████████▎| 6023/6470 [1:09:17<04:16,  1.74it/s]
3|6024|Loss: 0.8578031063079834:  93%|█████████▎| 6023/6470 [1:09:17<04:16,  1.74it/s]
3|6024|Loss: 0.8578031063079834:  93%|█████████▎| 6024/6470 [1:09:17<04:11,  1.78it/s]
3|6025|Loss: 0.841879665851593:  93%|█████████▎| 6024/6470 [1:09:18<04:11,  1.78it/s] 
3|6025|Loss: 0.841879665851593:  93%|█████████▎| 6025/6470 [1:09:18<04:23,  1.69it/s]
3|6026|Loss: 1.0308103561401367:  93%|█████████▎| 6025/6470 [1:09:18<04:23,  1.69it/s]
3|6026|Loss: 1.0308103561401367:  93%|█████████▎| 6026/6470 [1:09:18<04:29,  1.65it/s]
3|6027|Loss: 0.7418867349624634:  93%|█████████▎| 6026/6470 [1:09:19<04:29,  1.65it/s]
3|6027|Loss: 0.7418867349624634:  93%|█████████▎| 6027/6470 [1:09:19<04:25,  1.67it/s]
3|6028|Loss: 0.7092912197113037:  93%|█████████▎| 6027/6470 [1:09:19<04:25,  1.67it/s]
3|6028|Loss: 0.7092912197113037:  93%|█████████▎| 6028/6470 [1:09:20<04:28,  1.64it/s]
3|6029|Loss: 0.8782806992530823:  93%|█████████▎| 6028/6470 [1:09:20<04:28,  1.64it/s]
3|6029|Loss: 0.8782806992530823:  93%|█████████▎| 6029/6470 [1:09:20<04:26,  1.65it/s]
3|6030|Loss: 0.75539231300354:  93%|█████████▎| 6029/6470 [1:09:21<04:26,  1.65it/s]  
3|6030|Loss: 0.75539231300354:  93%|█████████▎| 6030/6470 [1:09:21<04:12,  1.74it/s]
3|6031|Loss: 0.7252823114395142:  93%|█████████▎| 6030/6470 [1:09:21<04:12,  1.74it/s]
3|6031|Loss: 0.7252823114395142:  93%|█████████▎| 6031/6470 [1:09:21<04:27,  1.64it/s]
3|6032|Loss: 0.9516870379447937:  93%|█████████▎| 6031/6470 [1:09:22<04:27,  1.64it/s]
3|6032|Loss: 0.9516870379447937:  93%|█████████▎| 6032/6470 [1:09:22<04:45,  1.54it/s]
3|6033|Loss: 0.8646754622459412:  93%|█████████▎| 6032/6470 [1:09:23<04:45,  1.54it/s]
3|6033|Loss: 0.8646754622459412:  93%|█████████▎| 6033/6470 [1:09:23<04:52,  1.49it/s]
3|6034|Loss: 0.821382462978363:  93%|█████████▎| 6033/6470 [1:09:23<04:52,  1.49it/s] 
3|6034|Loss: 0.821382462978363:  93%|█████████▎| 6034/6470 [1:09:24<04:56,  1.47it/s]
3|6035|Loss: 0.6983457207679749:  93%|█████████▎| 6034/6470 [1:09:24<04:56,  1.47it/s]
3|6035|Loss: 0.6983457207679749:  93%|█████████▎| 6035/6470 [1:09:24<04:45,  1.53it/s]
3|6036|Loss: 0.7853434681892395:  93%|█████████▎| 6035/6470 [1:09:25<04:45,  1.53it/s]
3|6036|Loss: 0.7853434681892395:  93%|█████████▎| 6036/6470 [1:09:25<04:51,  1.49it/s]
3|6037|Loss: 0.8654291033744812:  93%|█████████▎| 6036/6470 [1:09:25<04:51,  1.49it/s]
3|6037|Loss: 0.8654291033744812:  93%|█████████▎| 6037/6470 [1:09:26<05:02,  1.43it/s]
3|6038|Loss: 0.9631869792938232:  93%|█████████▎| 6037/6470 [1:09:26<05:02,  1.43it/s]
3|6038|Loss: 0.9631869792938232:  93%|█████████▎| 6038/6470 [1:09:27<05:17,  1.36it/s]
3|6039|Loss: 0.81467604637146:  93%|█████████▎| 6038/6470 [1:09:27<05:17,  1.36it/s]  
3|6039|Loss: 0.81467604637146:  93%|█████████▎| 6039/6470 [1:09:27<05:06,  1.40it/s]
3|6040|Loss: 0.8091148734092712:  93%|█████████▎| 6039/6470 [1:09:28<05:06,  1.40it/s]
3|6040|Loss: 0.8091148734092712:  93%|█████████▎| 6040/6470 [1:09:28<04:55,  1.45it/s]
3|6041|Loss: 0.8787339925765991:  93%|█████████▎| 6040/6470 [1:09:28<04:55,  1.45it/s]
3|6041|Loss: 0.8787339925765991:  93%|█████████▎| 6041/6470 [1:09:28<04:36,  1.55it/s]
3|6042|Loss: 0.7261168360710144:  93%|█████████▎| 6041/6470 [1:09:29<04:36,  1.55it/s]
3|6042|Loss: 0.7261168360710144:  93%|█████████▎| 6042/6470 [1:09:29<04:15,  1.67it/s]
3|6043|Loss: 0.9351856708526611:  93%|█████████▎| 6042/6470 [1:09:29<04:15,  1.67it/s]
3|6043|Loss: 0.9351856708526611:  93%|█████████▎| 6043/6470 [1:09:30<04:25,  1.61it/s]
3|6044|Loss: 0.8374584913253784:  93%|█████████▎| 6043/6470 [1:09:30<04:25,  1.61it/s]
3|6044|Loss: 0.8374584913253784:  93%|█████████▎| 6044/6470 [1:09:30<04:23,  1.62it/s]
3|6045|Loss: 0.7146967053413391:  93%|█████████▎| 6044/6470 [1:09:31<04:23,  1.62it/s]
3|6045|Loss: 0.7146967053413391:  93%|█████████▎| 6045/6470 [1:09:31<04:32,  1.56it/s]
3|6046|Loss: 0.9328970313072205:  93%|█████████▎| 6045/6470 [1:09:31<04:32,  1.56it/s]
3|6046|Loss: 0.9328970313072205:  93%|█████████▎| 6046/6470 [1:09:32<04:43,  1.50it/s]
3|6047|Loss: 0.7368560433387756:  93%|█████████▎| 6046/6470 [1:09:32<04:43,  1.50it/s]
3|6047|Loss: 0.7368560433387756:  93%|█████████▎| 6047/6470 [1:09:32<04:48,  1.47it/s]
3|6048|Loss: 0.9176723957061768:  93%|█████████▎| 6047/6470 [1:09:33<04:48,  1.47it/s]
3|6048|Loss: 0.9176723957061768:  93%|█████████▎| 6048/6470 [1:09:33<04:58,  1.41it/s]
3|6049|Loss: 0.7357802987098694:  93%|█████████▎| 6048/6470 [1:09:34<04:58,  1.41it/s]
3|6049|Loss: 0.7357802987098694:  93%|█████████▎| 6049/6470 [1:09:34<05:12,  1.35it/s]
3|6050|Loss: 0.9839383363723755:  93%|█████████▎| 6049/6470 [1:09:34<05:12,  1.35it/s]
3|6050|Loss: 0.9839383363723755:  94%|█████████▎| 6050/6470 [1:09:35<05:04,  1.38it/s]
3|6051|Loss: 0.8884075284004211:  94%|█████████▎| 6050/6470 [1:09:35<05:04,  1.38it/s]
3|6051|Loss: 0.8884075284004211:  94%|█████████▎| 6051/6470 [1:09:35<04:41,  1.49it/s]
3|6052|Loss: 0.8715643882751465:  94%|█████████▎| 6051/6470 [1:09:36<04:41,  1.49it/s]
3|6052|Loss: 0.8715643882751465:  94%|█████████▎| 6052/6470 [1:09:36<04:52,  1.43it/s]
3|6053|Loss: 0.9807333946228027:  94%|█████████▎| 6052/6470 [1:09:36<04:52,  1.43it/s]
3|6053|Loss: 0.9807333946228027:  94%|█████████▎| 6053/6470 [1:09:37<04:43,  1.47it/s]
3|6054|Loss: 0.7445261478424072:  94%|█████████▎| 6053/6470 [1:09:37<04:43,  1.47it/s]
3|6054|Loss: 0.7445261478424072:  94%|█████████▎| 6054/6470 [1:09:37<04:42,  1.47it/s]
3|6055|Loss: 0.8059493899345398:  94%|█████████▎| 6054/6470 [1:09:38<04:42,  1.47it/s]
3|6055|Loss: 0.8059493899345398:  94%|█████████▎| 6055/6470 [1:09:38<04:49,  1.43it/s]
3|6056|Loss: 0.7951308488845825:  94%|█████████▎| 6055/6470 [1:09:38<04:49,  1.43it/s]
3|6056|Loss: 0.7951308488845825:  94%|█████████▎| 6056/6470 [1:09:39<04:51,  1.42it/s]
3|6057|Loss: 1.0351734161376953:  94%|█████████▎| 6056/6470 [1:09:39<04:51,  1.42it/s]
3|6057|Loss: 1.0351734161376953:  94%|█████████▎| 6057/6470 [1:09:39<04:53,  1.41it/s]
3|6058|Loss: 0.7953425645828247:  94%|█████████▎| 6057/6470 [1:09:40<04:53,  1.41it/s]
3|6058|Loss: 0.7953425645828247:  94%|█████████▎| 6058/6470 [1:09:40<04:44,  1.45it/s]
3|6059|Loss: 0.8277808427810669:  94%|█████████▎| 6058/6470 [1:09:40<04:44,  1.45it/s]
3|6059|Loss: 0.8277808427810669:  94%|█████████▎| 6059/6470 [1:09:41<04:46,  1.43it/s]
3|6060|Loss: 1.0014145374298096:  94%|█████████▎| 6059/6470 [1:09:41<04:46,  1.43it/s]
3|6060|Loss: 1.0014145374298096:  94%|█████████▎| 6060/6470 [1:09:41<04:39,  1.47it/s]
3|6061|Loss: 0.9019140005111694:  94%|█████████▎| 6060/6470 [1:09:42<04:39,  1.47it/s]
3|6061|Loss: 0.9019140005111694:  94%|█████████▎| 6061/6470 [1:09:42<04:20,  1.57it/s]
3|6062|Loss: 0.7068661451339722:  94%|█████████▎| 6061/6470 [1:09:42<04:20,  1.57it/s]
3|6062|Loss: 0.7068661451339722:  94%|█████████▎| 6062/6470 [1:09:43<04:48,  1.41it/s]
3|6063|Loss: 0.8610641956329346:  94%|█████████▎| 6062/6470 [1:09:43<04:48,  1.41it/s]
3|6063|Loss: 0.8610641956329346:  94%|█████████▎| 6063/6470 [1:09:44<05:05,  1.33it/s]
3|6064|Loss: 0.7634763121604919:  94%|█████████▎| 6063/6470 [1:09:44<05:05,  1.33it/s]
3|6064|Loss: 0.7634763121604919:  94%|█████████▎| 6064/6470 [1:09:44<04:37,  1.46it/s]
3|6065|Loss: 0.7256467938423157:  94%|█████████▎| 6064/6470 [1:09:45<04:37,  1.46it/s]
3|6065|Loss: 0.7256467938423157:  94%|█████████▎| 6065/6470 [1:09:45<04:44,  1.42it/s]
3|6066|Loss: 0.9084215760231018:  94%|█████████▎| 6065/6470 [1:09:45<04:44,  1.42it/s]
3|6066|Loss: 0.9084215760231018:  94%|█████████▍| 6066/6470 [1:09:46<04:50,  1.39it/s]
3|6067|Loss: 0.8609638214111328:  94%|█████████▍| 6066/6470 [1:09:46<04:50,  1.39it/s]
3|6067|Loss: 0.8609638214111328:  94%|█████████▍| 6067/6470 [1:09:47<05:05,  1.32it/s]
3|6068|Loss: 0.8047366142272949:  94%|█████████▍| 6067/6470 [1:09:47<05:05,  1.32it/s]
3|6068|Loss: 0.8047366142272949:  94%|█████████▍| 6068/6470 [1:09:47<05:12,  1.29it/s]
3|6069|Loss: 0.9032185077667236:  94%|█████████▍| 6068/6470 [1:09:48<05:12,  1.29it/s]
3|6069|Loss: 0.9032185077667236:  94%|█████████▍| 6069/6470 [1:09:48<04:59,  1.34it/s]
3|6070|Loss: 0.8443402051925659:  94%|█████████▍| 6069/6470 [1:09:48<04:59,  1.34it/s]
3|6070|Loss: 0.8443402051925659:  94%|█████████▍| 6070/6470 [1:09:49<04:52,  1.37it/s]
3|6071|Loss: 0.7825683951377869:  94%|█████████▍| 6070/6470 [1:09:49<04:52,  1.37it/s]
3|6071|Loss: 0.7825683951377869:  94%|█████████▍| 6071/6470 [1:09:49<04:42,  1.41it/s]
3|6072|Loss: 0.8457868099212646:  94%|█████████▍| 6071/6470 [1:09:50<04:42,  1.41it/s]
3|6072|Loss: 0.8457868099212646:  94%|█████████▍| 6072/6470 [1:09:50<04:27,  1.49it/s]
3|6073|Loss: 0.9125932455062866:  94%|█████████▍| 6072/6470 [1:09:50<04:27,  1.49it/s]
3|6073|Loss: 0.9125932455062866:  94%|█████████▍| 6073/6470 [1:09:51<04:31,  1.46it/s]
3|6074|Loss: 0.9699991941452026:  94%|█████████▍| 6073/6470 [1:09:51<04:31,  1.46it/s]
3|6074|Loss: 0.9699991941452026:  94%|█████████▍| 6074/6470 [1:09:51<04:21,  1.51it/s]
3|6075|Loss: 0.8262261152267456:  94%|█████████▍| 6074/6470 [1:09:52<04:21,  1.51it/s]
3|6075|Loss: 0.8262261152267456:  94%|█████████▍| 6075/6470 [1:09:52<04:23,  1.50it/s]
3|6076|Loss: 0.9624342322349548:  94%|█████████▍| 6075/6470 [1:09:52<04:23,  1.50it/s]
3|6076|Loss: 0.9624342322349548:  94%|█████████▍| 6076/6470 [1:09:53<04:30,  1.46it/s]
3|6077|Loss: 0.9410070180892944:  94%|█████████▍| 6076/6470 [1:09:53<04:30,  1.46it/s]
3|6077|Loss: 0.9410070180892944:  94%|█████████▍| 6077/6470 [1:09:53<04:35,  1.43it/s]
3|6078|Loss: 0.8772875666618347:  94%|█████████▍| 6077/6470 [1:09:54<04:35,  1.43it/s]
3|6078|Loss: 0.8772875666618347:  94%|█████████▍| 6078/6470 [1:09:54<04:39,  1.40it/s]
3|6079|Loss: 0.9179128408432007:  94%|█████████▍| 6078/6470 [1:09:55<04:39,  1.40it/s]
3|6079|Loss: 0.9179128408432007:  94%|█████████▍| 6079/6470 [1:09:55<04:39,  1.40it/s]
3|6080|Loss: 0.9114714860916138:  94%|█████████▍| 6079/6470 [1:09:55<04:39,  1.40it/s]
3|6080|Loss: 0.9114714860916138:  94%|█████████▍| 6080/6470 [1:09:56<04:36,  1.41it/s]
3|6081|Loss: 0.7550126314163208:  94%|█████████▍| 6080/6470 [1:09:56<04:36,  1.41it/s]
3|6081|Loss: 0.7550126314163208:  94%|█████████▍| 6081/6470 [1:09:56<04:49,  1.35it/s]
3|6082|Loss: 0.854232668876648:  94%|█████████▍| 6081/6470 [1:09:57<04:49,  1.35it/s] 
3|6082|Loss: 0.854232668876648:  94%|█████████▍| 6082/6470 [1:09:57<04:42,  1.37it/s]
3|6083|Loss: 0.829127311706543:  94%|█████████▍| 6082/6470 [1:09:57<04:42,  1.37it/s]
3|6083|Loss: 0.829127311706543:  94%|█████████▍| 6083/6470 [1:09:58<04:14,  1.52it/s]
3|6084|Loss: 0.8319195508956909:  94%|█████████▍| 6083/6470 [1:09:58<04:14,  1.52it/s]
3|6084|Loss: 0.8319195508956909:  94%|█████████▍| 6084/6470 [1:09:58<04:16,  1.51it/s]
3|6085|Loss: 0.7484601736068726:  94%|█████████▍| 6084/6470 [1:09:59<04:16,  1.51it/s]
3|6085|Loss: 0.7484601736068726:  94%|█████████▍| 6085/6470 [1:09:59<04:33,  1.41it/s]
3|6086|Loss: 0.7174080014228821:  94%|█████████▍| 6085/6470 [1:10:00<04:33,  1.41it/s]
3|6086|Loss: 0.7174080014228821:  94%|█████████▍| 6086/6470 [1:10:00<04:51,  1.32it/s]
3|6087|Loss: 0.792981743812561:  94%|█████████▍| 6086/6470 [1:10:00<04:51,  1.32it/s] 
3|6087|Loss: 0.792981743812561:  94%|█████████▍| 6087/6470 [1:10:01<04:38,  1.37it/s]
3|6088|Loss: 0.8091962933540344:  94%|█████████▍| 6087/6470 [1:10:01<04:38,  1.37it/s]
3|6088|Loss: 0.8091962933540344:  94%|█████████▍| 6088/6470 [1:10:01<04:26,  1.44it/s]
3|6089|Loss: 0.8492865562438965:  94%|█████████▍| 6088/6470 [1:10:02<04:26,  1.44it/s]
3|6089|Loss: 0.8492865562438965:  94%|█████████▍| 6089/6470 [1:10:02<04:40,  1.36it/s]
3|6090|Loss: 0.8616625666618347:  94%|█████████▍| 6089/6470 [1:10:02<04:40,  1.36it/s]
3|6090|Loss: 0.8616625666618347:  94%|█████████▍| 6090/6470 [1:10:03<04:43,  1.34it/s]
3|6091|Loss: 0.9770119190216064:  94%|█████████▍| 6090/6470 [1:10:03<04:43,  1.34it/s]
3|6091|Loss: 0.9770119190216064:  94%|█████████▍| 6091/6470 [1:10:03<04:20,  1.46it/s]
3|6092|Loss: 0.8715260028839111:  94%|█████████▍| 6091/6470 [1:10:04<04:20,  1.46it/s]
3|6092|Loss: 0.8715260028839111:  94%|█████████▍| 6092/6470 [1:10:04<04:19,  1.45it/s]
3|6093|Loss: 0.8607649803161621:  94%|█████████▍| 6092/6470 [1:10:05<04:19,  1.45it/s]
3|6093|Loss: 0.8607649803161621:  94%|█████████▍| 6093/6470 [1:10:05<04:36,  1.36it/s]
3|6094|Loss: 0.7957244515419006:  94%|█████████▍| 6093/6470 [1:10:05<04:36,  1.36it/s]
3|6094|Loss: 0.7957244515419006:  94%|█████████▍| 6094/6470 [1:10:05<04:19,  1.45it/s]
3|6095|Loss: 0.8558082580566406:  94%|█████████▍| 6094/6470 [1:10:06<04:19,  1.45it/s]
3|6095|Loss: 0.8558082580566406:  94%|█████████▍| 6095/6470 [1:10:06<04:15,  1.47it/s]
3|6096|Loss: 0.7441391348838806:  94%|█████████▍| 6095/6470 [1:10:07<04:15,  1.47it/s]
3|6096|Loss: 0.7441391348838806:  94%|█████████▍| 6096/6470 [1:10:07<04:09,  1.50it/s]
3|6097|Loss: 0.8087542653083801:  94%|█████████▍| 6096/6470 [1:10:07<04:09,  1.50it/s]
3|6097|Loss: 0.8087542653083801:  94%|█████████▍| 6097/6470 [1:10:07<04:05,  1.52it/s]
3|6098|Loss: 0.9833462834358215:  94%|█████████▍| 6097/6470 [1:10:08<04:05,  1.52it/s]
3|6098|Loss: 0.9833462834358215:  94%|█████████▍| 6098/6470 [1:10:08<04:08,  1.50it/s]
3|6099|Loss: 1.0410432815551758:  94%|█████████▍| 6098/6470 [1:10:09<04:08,  1.50it/s]
3|6099|Loss: 1.0410432815551758:  94%|█████████▍| 6099/6470 [1:10:09<04:23,  1.41it/s]
3|6100|Loss: 0.7619274854660034:  94%|█████████▍| 6099/6470 [1:10:09<04:23,  1.41it/s]
3|6100|Loss: 0.7619274854660034:  94%|█████████▍| 6100/6470 [1:10:10<04:13,  1.46it/s]
3|6101|Loss: 0.9302805066108704:  94%|█████████▍| 6100/6470 [1:10:10<04:13,  1.46it/s]
3|6101|Loss: 0.9302805066108704:  94%|█████████▍| 6101/6470 [1:10:10<04:01,  1.53it/s]
3|6102|Loss: 0.7587083578109741:  94%|█████████▍| 6101/6470 [1:10:10<04:01,  1.53it/s]
3|6102|Loss: 0.7587083578109741:  94%|█████████▍| 6102/6470 [1:10:11<03:52,  1.58it/s]
3|6103|Loss: 0.9846706986427307:  94%|█████████▍| 6102/6470 [1:10:11<03:52,  1.58it/s]
3|6103|Loss: 0.9846706986427307:  94%|█████████▍| 6103/6470 [1:10:11<03:57,  1.55it/s]
3|6104|Loss: 0.8483598232269287:  94%|█████████▍| 6103/6470 [1:10:12<03:57,  1.55it/s]
3|6104|Loss: 0.8483598232269287:  94%|█████████▍| 6104/6470 [1:10:12<04:05,  1.49it/s]
3|6105|Loss: 0.8481702208518982:  94%|█████████▍| 6104/6470 [1:10:13<04:05,  1.49it/s]
3|6105|Loss: 0.8481702208518982:  94%|█████████▍| 6105/6470 [1:10:13<04:04,  1.49it/s]
3|6106|Loss: 0.7907987833023071:  94%|█████████▍| 6105/6470 [1:10:13<04:04,  1.49it/s]
3|6106|Loss: 0.7907987833023071:  94%|█████████▍| 6106/6470 [1:10:14<04:19,  1.40it/s]
3|6107|Loss: 0.8235592246055603:  94%|█████████▍| 6106/6470 [1:10:14<04:19,  1.40it/s]
3|6107|Loss: 0.8235592246055603:  94%|█████████▍| 6107/6470 [1:10:14<04:17,  1.41it/s]
3|6108|Loss: 0.8378170132637024:  94%|█████████▍| 6107/6470 [1:10:15<04:17,  1.41it/s]
3|6108|Loss: 0.8378170132637024:  94%|█████████▍| 6108/6470 [1:10:15<04:14,  1.42it/s]
3|6109|Loss: 0.7894564867019653:  94%|█████████▍| 6108/6470 [1:10:15<04:14,  1.42it/s]
3|6109|Loss: 0.7894564867019653:  94%|█████████▍| 6109/6470 [1:10:16<04:05,  1.47it/s]
3|6110|Loss: 0.9060454964637756:  94%|█████████▍| 6109/6470 [1:10:16<04:05,  1.47it/s]
3|6110|Loss: 0.9060454964637756:  94%|█████████▍| 6110/6470 [1:10:16<03:58,  1.51it/s]
3|6111|Loss: 0.7879083752632141:  94%|█████████▍| 6110/6470 [1:10:17<03:58,  1.51it/s]
3|6111|Loss: 0.7879083752632141:  94%|█████████▍| 6111/6470 [1:10:17<03:49,  1.57it/s]
3|6112|Loss: 0.9019736647605896:  94%|█████████▍| 6111/6470 [1:10:17<03:49,  1.57it/s]
3|6112|Loss: 0.9019736647605896:  94%|█████████▍| 6112/6470 [1:10:17<03:48,  1.57it/s]
3|6113|Loss: 0.960815966129303:  94%|█████████▍| 6112/6470 [1:10:18<03:48,  1.57it/s] 
3|6113|Loss: 0.960815966129303:  94%|█████████▍| 6113/6470 [1:10:18<03:54,  1.52it/s]
3|6114|Loss: 0.8105939626693726:  94%|█████████▍| 6113/6470 [1:10:19<03:54,  1.52it/s]
3|6114|Loss: 0.8105939626693726:  94%|█████████▍| 6114/6470 [1:10:19<04:01,  1.47it/s]
3|6115|Loss: 0.8377861976623535:  94%|█████████▍| 6114/6470 [1:10:19<04:01,  1.47it/s]
3|6115|Loss: 0.8377861976623535:  95%|█████████▍| 6115/6470 [1:10:20<03:59,  1.49it/s]
3|6116|Loss: 0.8331359624862671:  95%|█████████▍| 6115/6470 [1:10:20<03:59,  1.49it/s]
3|6116|Loss: 0.8331359624862671:  95%|█████████▍| 6116/6470 [1:10:20<04:15,  1.38it/s]
3|6117|Loss: 0.7645776867866516:  95%|█████████▍| 6116/6470 [1:10:21<04:15,  1.38it/s]
3|6117|Loss: 0.7645776867866516:  95%|█████████▍| 6117/6470 [1:10:21<04:02,  1.45it/s]
3|6118|Loss: 0.9576562643051147:  95%|█████████▍| 6117/6470 [1:10:21<04:02,  1.45it/s]
3|6118|Loss: 0.9576562643051147:  95%|█████████▍| 6118/6470 [1:10:21<03:39,  1.61it/s]
3|6119|Loss: 0.8117202520370483:  95%|█████████▍| 6118/6470 [1:10:22<03:39,  1.61it/s]
3|6119|Loss: 0.8117202520370483:  95%|█████████▍| 6119/6470 [1:10:22<03:33,  1.64it/s]
3|6120|Loss: 0.8538826107978821:  95%|█████████▍| 6119/6470 [1:10:22<03:33,  1.64it/s]
3|6120|Loss: 0.8538826107978821:  95%|█████████▍| 6120/6470 [1:10:23<03:43,  1.57it/s]
3|6121|Loss: 0.7594163417816162:  95%|█████████▍| 6120/6470 [1:10:23<03:43,  1.57it/s]
3|6121|Loss: 0.7594163417816162:  95%|█████████▍| 6121/6470 [1:10:23<03:54,  1.49it/s]
3|6122|Loss: 0.7687694430351257:  95%|█████████▍| 6121/6470 [1:10:24<03:54,  1.49it/s]
3|6122|Loss: 0.7687694430351257:  95%|█████████▍| 6122/6470 [1:10:24<03:42,  1.57it/s]
3|6123|Loss: 0.9323487877845764:  95%|█████████▍| 6122/6470 [1:10:24<03:42,  1.57it/s]
3|6123|Loss: 0.9323487877845764:  95%|█████████▍| 6123/6470 [1:10:25<03:33,  1.63it/s]
3|6124|Loss: 0.8487752079963684:  95%|█████████▍| 6123/6470 [1:10:25<03:33,  1.63it/s]
3|6124|Loss: 0.8487752079963684:  95%|█████████▍| 6124/6470 [1:10:25<03:33,  1.62it/s]
3|6125|Loss: 0.8567269444465637:  95%|█████████▍| 6124/6470 [1:10:26<03:33,  1.62it/s]
3|6125|Loss: 0.8567269444465637:  95%|█████████▍| 6125/6470 [1:10:26<03:51,  1.49it/s]
3|6126|Loss: 0.8053194880485535:  95%|█████████▍| 6125/6470 [1:10:26<03:51,  1.49it/s]
3|6126|Loss: 0.8053194880485535:  95%|█████████▍| 6126/6470 [1:10:27<03:58,  1.44it/s]
3|6127|Loss: 0.8275634050369263:  95%|█████████▍| 6126/6470 [1:10:27<03:58,  1.44it/s]
3|6127|Loss: 0.8275634050369263:  95%|█████████▍| 6127/6470 [1:10:28<04:03,  1.41it/s]
3|6128|Loss: 0.7272701263427734:  95%|█████████▍| 6127/6470 [1:10:28<04:03,  1.41it/s]
3|6128|Loss: 0.7272701263427734:  95%|█████████▍| 6128/6470 [1:10:28<04:03,  1.40it/s]
3|6129|Loss: 0.9419434666633606:  95%|█████████▍| 6128/6470 [1:10:29<04:03,  1.40it/s]
3|6129|Loss: 0.9419434666633606:  95%|█████████▍| 6129/6470 [1:10:29<04:05,  1.39it/s]
3|6130|Loss: 0.9535602331161499:  95%|█████████▍| 6129/6470 [1:10:29<04:05,  1.39it/s]
3|6130|Loss: 0.9535602331161499:  95%|█████████▍| 6130/6470 [1:10:30<03:56,  1.44it/s]
3|6131|Loss: 0.7337692975997925:  95%|█████████▍| 6130/6470 [1:10:30<03:56,  1.44it/s]
3|6131|Loss: 0.7337692975997925:  95%|█████████▍| 6131/6470 [1:10:30<03:49,  1.47it/s]
3|6132|Loss: 0.8825507164001465:  95%|█████████▍| 6131/6470 [1:10:31<03:49,  1.47it/s]
3|6132|Loss: 0.8825507164001465:  95%|█████████▍| 6132/6470 [1:10:31<03:40,  1.53it/s]
3|6133|Loss: 1.004589557647705:  95%|█████████▍| 6132/6470 [1:10:31<03:40,  1.53it/s] 
3|6133|Loss: 1.004589557647705:  95%|█████████▍| 6133/6470 [1:10:32<03:51,  1.46it/s]
3|6134|Loss: 0.7409904599189758:  95%|█████████▍| 6133/6470 [1:10:32<03:51,  1.46it/s]
3|6134|Loss: 0.7409904599189758:  95%|█████████▍| 6134/6470 [1:10:32<03:58,  1.41it/s]
3|6135|Loss: 0.9184789061546326:  95%|█████████▍| 6134/6470 [1:10:33<03:58,  1.41it/s]
3|6135|Loss: 0.9184789061546326:  95%|█████████▍| 6135/6470 [1:10:33<04:00,  1.39it/s]
3|6136|Loss: 0.9225282669067383:  95%|█████████▍| 6135/6470 [1:10:34<04:00,  1.39it/s]
3|6136|Loss: 0.9225282669067383:  95%|█████████▍| 6136/6470 [1:10:34<04:01,  1.38it/s]
3|6137|Loss: 1.048817753791809:  95%|█████████▍| 6136/6470 [1:10:34<04:01,  1.38it/s] 
3|6137|Loss: 1.048817753791809:  95%|█████████▍| 6137/6470 [1:10:35<04:00,  1.39it/s]
3|6138|Loss: 0.7817816734313965:  95%|█████████▍| 6137/6470 [1:10:35<04:00,  1.39it/s]
3|6138|Loss: 0.7817816734313965:  95%|█████████▍| 6138/6470 [1:10:35<04:12,  1.32it/s]
3|6139|Loss: 0.834687352180481:  95%|█████████▍| 6138/6470 [1:10:36<04:12,  1.32it/s] 
3|6139|Loss: 0.834687352180481:  95%|█████████▍| 6139/6470 [1:10:36<03:58,  1.39it/s]
3|6140|Loss: 0.8426908850669861:  95%|█████████▍| 6139/6470 [1:10:36<03:58,  1.39it/s]
3|6140|Loss: 0.8426908850669861:  95%|█████████▍| 6140/6470 [1:10:37<03:43,  1.48it/s]
3|6141|Loss: 0.8789052963256836:  95%|█████████▍| 6140/6470 [1:10:37<03:43,  1.48it/s]
3|6141|Loss: 0.8789052963256836:  95%|█████████▍| 6141/6470 [1:10:37<03:45,  1.46it/s]
3|6142|Loss: 1.022890567779541:  95%|█████████▍| 6141/6470 [1:10:38<03:45,  1.46it/s] 
3|6142|Loss: 1.022890567779541:  95%|█████████▍| 6142/6470 [1:10:38<03:42,  1.47it/s]
3|6143|Loss: 0.9235671162605286:  95%|█████████▍| 6142/6470 [1:10:38<03:42,  1.47it/s]
3|6143|Loss: 0.9235671162605286:  95%|█████████▍| 6143/6470 [1:10:39<03:43,  1.46it/s]
3|6144|Loss: 0.8183974623680115:  95%|█████████▍| 6143/6470 [1:10:39<03:43,  1.46it/s]
3|6144|Loss: 0.8183974623680115:  95%|█████████▍| 6144/6470 [1:10:39<03:45,  1.45it/s]
3|6145|Loss: 1.0324132442474365:  95%|█████████▍| 6144/6470 [1:10:40<03:45,  1.45it/s]
3|6145|Loss: 1.0324132442474365:  95%|█████████▍| 6145/6470 [1:10:40<03:39,  1.48it/s]
3|6146|Loss: 0.8492832779884338:  95%|█████████▍| 6145/6470 [1:10:40<03:39,  1.48it/s]
3|6146|Loss: 0.8492832779884338:  95%|█████████▍| 6146/6470 [1:10:41<03:34,  1.51it/s]
3|6147|Loss: 0.8959187269210815:  95%|█████████▍| 6146/6470 [1:10:41<03:34,  1.51it/s]
3|6147|Loss: 0.8959187269210815:  95%|█████████▌| 6147/6470 [1:10:41<03:25,  1.57it/s]
3|6148|Loss: 0.770493745803833:  95%|█████████▌| 6147/6470 [1:10:42<03:25,  1.57it/s] 
3|6148|Loss: 0.770493745803833:  95%|█████████▌| 6148/6470 [1:10:42<03:31,  1.52it/s]
3|6149|Loss: 0.9176044464111328:  95%|█████████▌| 6148/6470 [1:10:42<03:31,  1.52it/s]
3|6149|Loss: 0.9176044464111328:  95%|█████████▌| 6149/6470 [1:10:43<03:39,  1.46it/s]
3|6150|Loss: 0.8213489651679993:  95%|█████████▌| 6149/6470 [1:10:43<03:39,  1.46it/s]
3|6150|Loss: 0.8213489651679993:  95%|█████████▌| 6150/6470 [1:10:43<03:29,  1.53it/s]
3|6151|Loss: 0.905975341796875:  95%|█████████▌| 6150/6470 [1:10:44<03:29,  1.53it/s] 
3|6151|Loss: 0.905975341796875:  95%|█████████▌| 6151/6470 [1:10:44<03:30,  1.51it/s]
3|6152|Loss: 0.8064911961555481:  95%|█████████▌| 6151/6470 [1:10:44<03:30,  1.51it/s]
3|6152|Loss: 0.8064911961555481:  95%|█████████▌| 6152/6470 [1:10:45<03:36,  1.47it/s]
3|6153|Loss: 0.9245295524597168:  95%|█████████▌| 6152/6470 [1:10:45<03:36,  1.47it/s]
3|6153|Loss: 0.9245295524597168:  95%|█████████▌| 6153/6470 [1:10:45<03:12,  1.64it/s]
3|6154|Loss: 0.8847900032997131:  95%|█████████▌| 6153/6470 [1:10:45<03:12,  1.64it/s]
3|6154|Loss: 0.8847900032997131:  95%|█████████▌| 6154/6470 [1:10:46<03:10,  1.66it/s]
3|6155|Loss: 0.7334561347961426:  95%|█████████▌| 6154/6470 [1:10:46<03:10,  1.66it/s]
3|6155|Loss: 0.7334561347961426:  95%|█████████▌| 6155/6470 [1:10:46<03:18,  1.59it/s]
3|6156|Loss: 0.7856133580207825:  95%|█████████▌| 6155/6470 [1:10:47<03:18,  1.59it/s]
3|6156|Loss: 0.7856133580207825:  95%|█████████▌| 6156/6470 [1:10:47<03:18,  1.58it/s]
3|6157|Loss: 0.8042406439781189:  95%|█████████▌| 6156/6470 [1:10:47<03:18,  1.58it/s]
3|6157|Loss: 0.8042406439781189:  95%|█████████▌| 6157/6470 [1:10:48<03:29,  1.49it/s]
3|6158|Loss: 0.9151042103767395:  95%|█████████▌| 6157/6470 [1:10:48<03:29,  1.49it/s]
3|6158|Loss: 0.9151042103767395:  95%|█████████▌| 6158/6470 [1:10:48<03:28,  1.50it/s]
3|6159|Loss: 0.966727614402771:  95%|█████████▌| 6158/6470 [1:10:49<03:28,  1.50it/s] 
3|6159|Loss: 0.966727614402771:  95%|█████████▌| 6159/6470 [1:10:49<03:29,  1.49it/s]
3|6160|Loss: 0.8591150641441345:  95%|█████████▌| 6159/6470 [1:10:50<03:29,  1.49it/s]
3|6160|Loss: 0.8591150641441345:  95%|█████████▌| 6160/6470 [1:10:50<03:22,  1.53it/s]
3|6161|Loss: 0.838552713394165:  95%|█████████▌| 6160/6470 [1:10:50<03:22,  1.53it/s] 
3|6161|Loss: 0.838552713394165:  95%|█████████▌| 6161/6470 [1:10:50<03:05,  1.67it/s]
3|6162|Loss: 0.805021345615387:  95%|█████████▌| 6161/6470 [1:10:51<03:05,  1.67it/s]
3|6162|Loss: 0.805021345615387:  95%|█████████▌| 6162/6470 [1:10:51<03:04,  1.67it/s]
3|6163|Loss: 1.0405422449111938:  95%|█████████▌| 6162/6470 [1:10:51<03:04,  1.67it/s]
3|6163|Loss: 1.0405422449111938:  95%|█████████▌| 6163/6470 [1:10:52<03:22,  1.52it/s]
3|6164|Loss: 0.8933396339416504:  95%|█████████▌| 6163/6470 [1:10:52<03:22,  1.52it/s]
3|6164|Loss: 0.8933396339416504:  95%|█████████▌| 6164/6470 [1:10:52<03:32,  1.44it/s]
3|6165|Loss: 1.0184147357940674:  95%|█████████▌| 6164/6470 [1:10:53<03:32,  1.44it/s]
3|6165|Loss: 1.0184147357940674:  95%|█████████▌| 6165/6470 [1:10:53<03:42,  1.37it/s]
3|6166|Loss: 0.7598266005516052:  95%|█████████▌| 6165/6470 [1:10:54<03:42,  1.37it/s]
3|6166|Loss: 0.7598266005516052:  95%|█████████▌| 6166/6470 [1:10:54<03:32,  1.43it/s]
3|6167|Loss: 0.9863261580467224:  95%|█████████▌| 6166/6470 [1:10:54<03:32,  1.43it/s]
3|6167|Loss: 0.9863261580467224:  95%|█████████▌| 6167/6470 [1:10:54<03:26,  1.47it/s]
3|6168|Loss: 0.9097002744674683:  95%|█████████▌| 6167/6470 [1:10:55<03:26,  1.47it/s]
3|6168|Loss: 0.9097002744674683:  95%|█████████▌| 6168/6470 [1:10:55<03:13,  1.56it/s]
3|6169|Loss: 0.94097900390625:  95%|█████████▌| 6168/6470 [1:10:55<03:13,  1.56it/s]  
3|6169|Loss: 0.94097900390625:  95%|█████████▌| 6169/6470 [1:10:56<03:10,  1.58it/s]
3|6170|Loss: 0.892051100730896:  95%|█████████▌| 6169/6470 [1:10:56<03:10,  1.58it/s]
3|6170|Loss: 0.892051100730896:  95%|█████████▌| 6170/6470 [1:10:56<03:05,  1.61it/s]
3|6171|Loss: 0.8243042230606079:  95%|█████████▌| 6170/6470 [1:10:57<03:05,  1.61it/s]
3|6171|Loss: 0.8243042230606079:  95%|█████████▌| 6171/6470 [1:10:57<03:01,  1.64it/s]
3|6172|Loss: 0.8776293396949768:  95%|█████████▌| 6171/6470 [1:10:57<03:01,  1.64it/s]
3|6172|Loss: 0.8776293396949768:  95%|█████████▌| 6172/6470 [1:10:57<02:56,  1.69it/s]
3|6173|Loss: 1.0499461889266968:  95%|█████████▌| 6172/6470 [1:10:58<02:56,  1.69it/s]
3|6173|Loss: 1.0499461889266968:  95%|█████████▌| 6173/6470 [1:10:58<03:11,  1.55it/s]
3|6174|Loss: 0.8455565571784973:  95%|█████████▌| 6173/6470 [1:10:59<03:11,  1.55it/s]
3|6174|Loss: 0.8455565571784973:  95%|█████████▌| 6174/6470 [1:10:59<03:28,  1.42it/s]
3|6175|Loss: 0.8523632884025574:  95%|█████████▌| 6174/6470 [1:10:59<03:28,  1.42it/s]
3|6175|Loss: 0.8523632884025574:  95%|█████████▌| 6175/6470 [1:11:00<03:13,  1.53it/s]
3|6176|Loss: 0.8129269480705261:  95%|█████████▌| 6175/6470 [1:11:00<03:13,  1.53it/s]
3|6176|Loss: 0.8129269480705261:  95%|█████████▌| 6176/6470 [1:11:00<03:27,  1.41it/s]
3|6177|Loss: 0.8566468358039856:  95%|█████████▌| 6176/6470 [1:11:01<03:27,  1.41it/s]
3|6177|Loss: 0.8566468358039856:  95%|█████████▌| 6177/6470 [1:11:01<03:17,  1.48it/s]
3|6178|Loss: 0.833337664604187:  95%|█████████▌| 6177/6470 [1:11:01<03:17,  1.48it/s] 
3|6178|Loss: 0.833337664604187:  95%|█████████▌| 6178/6470 [1:11:02<03:27,  1.41it/s]
3|6179|Loss: 0.7827216386795044:  95%|█████████▌| 6178/6470 [1:11:02<03:27,  1.41it/s]
3|6179|Loss: 0.7827216386795044:  96%|█████████▌| 6179/6470 [1:11:02<03:28,  1.40it/s]
3|6180|Loss: 0.8205589652061462:  96%|█████████▌| 6179/6470 [1:11:03<03:28,  1.40it/s]
3|6180|Loss: 0.8205589652061462:  96%|█████████▌| 6180/6470 [1:11:03<03:20,  1.44it/s]
3|6181|Loss: 0.8794825673103333:  96%|█████████▌| 6180/6470 [1:11:03<03:20,  1.44it/s]
3|6181|Loss: 0.8794825673103333:  96%|█████████▌| 6181/6470 [1:11:04<03:13,  1.49it/s]
3|6182|Loss: 0.8667663931846619:  96%|█████████▌| 6181/6470 [1:11:04<03:13,  1.49it/s]
3|6182|Loss: 0.8667663931846619:  96%|█████████▌| 6182/6470 [1:11:05<03:21,  1.43it/s]
3|6183|Loss: 0.8582410216331482:  96%|█████████▌| 6182/6470 [1:11:05<03:21,  1.43it/s]
3|6183|Loss: 0.8582410216331482:  96%|█████████▌| 6183/6470 [1:11:05<03:28,  1.37it/s]
3|6184|Loss: 0.9071438908576965:  96%|█████████▌| 6183/6470 [1:11:06<03:28,  1.37it/s]
3|6184|Loss: 0.9071438908576965:  96%|█████████▌| 6184/6470 [1:11:06<03:22,  1.41it/s]
3|6185|Loss: 0.8522818088531494:  96%|█████████▌| 6184/6470 [1:11:06<03:22,  1.41it/s]
3|6185|Loss: 0.8522818088531494:  96%|█████████▌| 6185/6470 [1:11:07<03:22,  1.40it/s]
3|6186|Loss: 0.7826969623565674:  96%|█████████▌| 6185/6470 [1:11:07<03:22,  1.40it/s]
3|6186|Loss: 0.7826969623565674:  96%|█████████▌| 6186/6470 [1:11:07<03:10,  1.49it/s]
3|6187|Loss: 0.8839788436889648:  96%|█████████▌| 6186/6470 [1:11:08<03:10,  1.49it/s]
3|6187|Loss: 0.8839788436889648:  96%|█████████▌| 6187/6470 [1:11:08<03:05,  1.53it/s]
3|6188|Loss: 0.8738747239112854:  96%|█████████▌| 6187/6470 [1:11:08<03:05,  1.53it/s]
3|6188|Loss: 0.8738747239112854:  96%|█████████▌| 6188/6470 [1:11:09<03:03,  1.53it/s]
3|6189|Loss: 0.8674225211143494:  96%|█████████▌| 6188/6470 [1:11:09<03:03,  1.53it/s]
3|6189|Loss: 0.8674225211143494:  96%|█████████▌| 6189/6470 [1:11:09<03:02,  1.54it/s]
3|6190|Loss: 0.7252476215362549:  96%|█████████▌| 6189/6470 [1:11:10<03:02,  1.54it/s]
3|6190|Loss: 0.7252476215362549:  96%|█████████▌| 6190/6470 [1:11:10<03:55,  1.19it/s]
3|6191|Loss: 0.922370195388794:  96%|█████████▌| 6190/6470 [1:11:11<03:55,  1.19it/s] 
3|6191|Loss: 0.922370195388794:  96%|█████████▌| 6191/6470 [1:11:11<03:57,  1.17it/s]
3|6192|Loss: 0.8132686018943787:  96%|█████████▌| 6191/6470 [1:11:12<03:57,  1.17it/s]
3|6192|Loss: 0.8132686018943787:  96%|█████████▌| 6192/6470 [1:11:12<03:46,  1.23it/s]
3|6193|Loss: 0.7370272874832153:  96%|█████████▌| 6192/6470 [1:11:12<03:46,  1.23it/s]
3|6193|Loss: 0.7370272874832153:  96%|█████████▌| 6193/6470 [1:11:13<03:38,  1.27it/s]
3|6194|Loss: 0.9721719622612:  96%|█████████▌| 6193/6470 [1:11:13<03:38,  1.27it/s]   
3|6194|Loss: 0.9721719622612:  96%|█████████▌| 6194/6470 [1:11:13<03:19,  1.38it/s]
3|6195|Loss: 0.9666278958320618:  96%|█████████▌| 6194/6470 [1:11:14<03:19,  1.38it/s]
3|6195|Loss: 0.9666278958320618:  96%|█████████▌| 6195/6470 [1:11:14<03:17,  1.39it/s]
3|6196|Loss: 0.9647336006164551:  96%|█████████▌| 6195/6470 [1:11:14<03:17,  1.39it/s]
3|6196|Loss: 0.9647336006164551:  96%|█████████▌| 6196/6470 [1:11:15<03:11,  1.43it/s]
3|6197|Loss: 0.9787964820861816:  96%|█████████▌| 6196/6470 [1:11:15<03:11,  1.43it/s]
3|6197|Loss: 0.9787964820861816:  96%|█████████▌| 6197/6470 [1:11:15<03:13,  1.41it/s]
3|6198|Loss: 0.83018559217453:  96%|█████████▌| 6197/6470 [1:11:16<03:13,  1.41it/s]  
3|6198|Loss: 0.83018559217453:  96%|█████████▌| 6198/6470 [1:11:16<02:52,  1.57it/s]
3|6199|Loss: 0.8200609683990479:  96%|█████████▌| 6198/6470 [1:11:16<02:52,  1.57it/s]
3|6199|Loss: 0.8200609683990479:  96%|█████████▌| 6199/6470 [1:11:16<02:40,  1.69it/s]
3|6200|Loss: 0.7292908430099487:  96%|█████████▌| 6199/6470 [1:11:17<02:40,  1.69it/s]
3|6200|Loss: 0.7292908430099487:  96%|█████████▌| 6200/6470 [1:11:17<02:45,  1.63it/s]
3|6201|Loss: 0.7684630155563354:  96%|█████████▌| 6200/6470 [1:11:17<02:45,  1.63it/s]
3|6201|Loss: 0.7684630155563354:  96%|█████████▌| 6201/6470 [1:11:18<02:58,  1.51it/s]
3|6202|Loss: 0.8301876187324524:  96%|█████████▌| 6201/6470 [1:11:18<02:58,  1.51it/s]
3|6202|Loss: 0.8301876187324524:  96%|█████████▌| 6202/6470 [1:11:19<03:02,  1.47it/s]
3|6203|Loss: 1.0644978284835815:  96%|█████████▌| 6202/6470 [1:11:19<03:02,  1.47it/s]
3|6203|Loss: 1.0644978284835815:  96%|█████████▌| 6203/6470 [1:11:19<02:52,  1.55it/s]
3|6204|Loss: 0.7794234156608582:  96%|█████████▌| 6203/6470 [1:11:20<02:52,  1.55it/s]
3|6204|Loss: 0.7794234156608582:  96%|█████████▌| 6204/6470 [1:11:20<03:01,  1.46it/s]
3|6205|Loss: 0.8859331607818604:  96%|█████████▌| 6204/6470 [1:11:20<03:01,  1.46it/s]
3|6205|Loss: 0.8859331607818604:  96%|█████████▌| 6205/6470 [1:11:21<03:13,  1.37it/s]
3|6206|Loss: 0.9777465462684631:  96%|█████████▌| 6205/6470 [1:11:21<03:13,  1.37it/s]
3|6206|Loss: 0.9777465462684631:  96%|█████████▌| 6206/6470 [1:11:21<03:13,  1.37it/s]
3|6207|Loss: 0.8846268653869629:  96%|█████████▌| 6206/6470 [1:11:22<03:13,  1.37it/s]
3|6207|Loss: 0.8846268653869629:  96%|█████████▌| 6207/6470 [1:11:22<03:17,  1.33it/s]
3|6208|Loss: 0.7786688804626465:  96%|█████████▌| 6207/6470 [1:11:23<03:17,  1.33it/s]
3|6208|Loss: 0.7786688804626465:  96%|█████████▌| 6208/6470 [1:11:23<03:09,  1.38it/s]
3|6209|Loss: 0.8619087934494019:  96%|█████████▌| 6208/6470 [1:11:23<03:09,  1.38it/s]
3|6209|Loss: 0.8619087934494019:  96%|█████████▌| 6209/6470 [1:11:24<03:01,  1.44it/s]
3|6210|Loss: 0.7453673481941223:  96%|█████████▌| 6209/6470 [1:11:24<03:01,  1.44it/s]
3|6210|Loss: 0.7453673481941223:  96%|█████████▌| 6210/6470 [1:11:24<03:12,  1.35it/s]
3|6211|Loss: 0.8299837112426758:  96%|█████████▌| 6210/6470 [1:11:25<03:12,  1.35it/s]
3|6211|Loss: 0.8299837112426758:  96%|█████████▌| 6211/6470 [1:11:25<03:21,  1.28it/s]
3|6212|Loss: 0.9601916074752808:  96%|█████████▌| 6211/6470 [1:11:26<03:21,  1.28it/s]
3|6212|Loss: 0.9601916074752808:  96%|█████████▌| 6212/6470 [1:11:26<03:21,  1.28it/s]
3|6213|Loss: 0.9159295558929443:  96%|█████████▌| 6212/6470 [1:11:26<03:21,  1.28it/s]
3|6213|Loss: 0.9159295558929443:  96%|█████████▌| 6213/6470 [1:11:27<03:07,  1.37it/s]
3|6214|Loss: 0.8016574382781982:  96%|█████████▌| 6213/6470 [1:11:27<03:07,  1.37it/s]
3|6214|Loss: 0.8016574382781982:  96%|█████████▌| 6214/6470 [1:11:27<03:05,  1.38it/s]
3|6215|Loss: 0.8396309614181519:  96%|█████████▌| 6214/6470 [1:11:28<03:05,  1.38it/s]
3|6215|Loss: 0.8396309614181519:  96%|█████████▌| 6215/6470 [1:11:28<02:59,  1.42it/s]
3|6216|Loss: 0.9598156809806824:  96%|█████████▌| 6215/6470 [1:11:28<02:59,  1.42it/s]
3|6216|Loss: 0.9598156809806824:  96%|█████████▌| 6216/6470 [1:11:29<02:51,  1.48it/s]
3|6217|Loss: 0.850117027759552:  96%|█████████▌| 6216/6470 [1:11:29<02:51,  1.48it/s] 
3|6217|Loss: 0.850117027759552:  96%|█████████▌| 6217/6470 [1:11:29<02:42,  1.56it/s]
3|6218|Loss: 0.951012372970581:  96%|█████████▌| 6217/6470 [1:11:30<02:42,  1.56it/s]
3|6218|Loss: 0.951012372970581:  96%|█████████▌| 6218/6470 [1:11:30<02:38,  1.59it/s]
3|6219|Loss: 0.8846831917762756:  96%|█████████▌| 6218/6470 [1:11:30<02:38,  1.59it/s]
3|6219|Loss: 0.8846831917762756:  96%|█████████▌| 6219/6470 [1:11:30<02:38,  1.58it/s]
3|6220|Loss: 0.828415036201477:  96%|█████████▌| 6219/6470 [1:11:31<02:38,  1.58it/s] 
3|6220|Loss: 0.828415036201477:  96%|█████████▌| 6220/6470 [1:11:31<02:32,  1.63it/s]
3|6221|Loss: 0.7612879872322083:  96%|█████████▌| 6220/6470 [1:11:31<02:32,  1.63it/s]
3|6221|Loss: 0.7612879872322083:  96%|█████████▌| 6221/6470 [1:11:32<02:31,  1.64it/s]
3|6222|Loss: 1.0002275705337524:  96%|█████████▌| 6221/6470 [1:11:32<02:31,  1.64it/s]
3|6222|Loss: 1.0002275705337524:  96%|█████████▌| 6222/6470 [1:11:32<02:45,  1.50it/s]
3|6223|Loss: 0.8351907730102539:  96%|█████████▌| 6222/6470 [1:11:33<02:45,  1.50it/s]
3|6223|Loss: 0.8351907730102539:  96%|█████████▌| 6223/6470 [1:11:33<02:53,  1.43it/s]
3|6224|Loss: 0.8565739393234253:  96%|█████████▌| 6223/6470 [1:11:34<02:53,  1.43it/s]
3|6224|Loss: 0.8565739393234253:  96%|█████████▌| 6224/6470 [1:11:34<02:56,  1.39it/s]
3|6225|Loss: 0.7924126982688904:  96%|█████████▌| 6224/6470 [1:11:34<02:56,  1.39it/s]
3|6225|Loss: 0.7924126982688904:  96%|█████████▌| 6225/6470 [1:11:35<03:01,  1.35it/s]
3|6226|Loss: 0.900085985660553:  96%|█████████▌| 6225/6470 [1:11:35<03:01,  1.35it/s] 
3|6226|Loss: 0.900085985660553:  96%|█████████▌| 6226/6470 [1:11:35<02:59,  1.36it/s]
3|6227|Loss: 0.8652482032775879:  96%|█████████▌| 6226/6470 [1:11:36<02:59,  1.36it/s]
3|6227|Loss: 0.8652482032775879:  96%|█████████▌| 6227/6470 [1:11:36<02:52,  1.41it/s]
3|6228|Loss: 0.9083920121192932:  96%|█████████▌| 6227/6470 [1:11:37<02:52,  1.41it/s]
3|6228|Loss: 0.9083920121192932:  96%|█████████▋| 6228/6470 [1:11:37<02:51,  1.41it/s]
3|6229|Loss: 0.8500980734825134:  96%|█████████▋| 6228/6470 [1:11:37<02:51,  1.41it/s]
3|6229|Loss: 0.8500980734825134:  96%|█████████▋| 6229/6470 [1:11:37<02:43,  1.47it/s]
3|6230|Loss: 0.7894778847694397:  96%|█████████▋| 6229/6470 [1:11:38<02:43,  1.47it/s]
3|6230|Loss: 0.7894778847694397:  96%|█████████▋| 6230/6470 [1:11:38<02:34,  1.55it/s]
3|6231|Loss: 0.8798431158065796:  96%|█████████▋| 6230/6470 [1:11:38<02:34,  1.55it/s]
3|6231|Loss: 0.8798431158065796:  96%|█████████▋| 6231/6470 [1:11:39<02:25,  1.64it/s]
3|6232|Loss: 0.8342164158821106:  96%|█████████▋| 6231/6470 [1:11:39<02:25,  1.64it/s]
3|6232|Loss: 0.8342164158821106:  96%|█████████▋| 6232/6470 [1:11:39<02:30,  1.58it/s]
3|6233|Loss: 0.9239040613174438:  96%|█████████▋| 6232/6470 [1:11:40<02:30,  1.58it/s]
3|6233|Loss: 0.9239040613174438:  96%|█████████▋| 6233/6470 [1:11:40<02:42,  1.46it/s]
3|6234|Loss: 0.9491816759109497:  96%|█████████▋| 6233/6470 [1:11:40<02:42,  1.46it/s]
3|6234|Loss: 0.9491816759109497:  96%|█████████▋| 6234/6470 [1:11:41<02:40,  1.47it/s]
3|6235|Loss: 0.8575860857963562:  96%|█████████▋| 6234/6470 [1:11:41<02:40,  1.47it/s]
3|6235|Loss: 0.8575860857963562:  96%|█████████▋| 6235/6470 [1:11:41<02:45,  1.42it/s]
3|6236|Loss: 0.7034108638763428:  96%|█████████▋| 6235/6470 [1:11:42<02:45,  1.42it/s]
3|6236|Loss: 0.7034108638763428:  96%|█████████▋| 6236/6470 [1:11:42<02:42,  1.44it/s]
3|6237|Loss: 0.7893040180206299:  96%|█████████▋| 6236/6470 [1:11:43<02:42,  1.44it/s]
3|6237|Loss: 0.7893040180206299:  96%|█████████▋| 6237/6470 [1:11:43<02:43,  1.43it/s]
3|6238|Loss: 0.9994187951087952:  96%|█████████▋| 6237/6470 [1:11:43<02:43,  1.43it/s]
3|6238|Loss: 0.9994187951087952:  96%|█████████▋| 6238/6470 [1:11:43<02:38,  1.47it/s]
3|6239|Loss: 0.8630518317222595:  96%|█████████▋| 6238/6470 [1:11:44<02:38,  1.47it/s]
3|6239|Loss: 0.8630518317222595:  96%|█████████▋| 6239/6470 [1:11:44<02:32,  1.51it/s]
3|6240|Loss: 0.7398631572723389:  96%|█████████▋| 6239/6470 [1:11:44<02:32,  1.51it/s]
3|6240|Loss: 0.7398631572723389:  96%|█████████▋| 6240/6470 [1:11:45<02:30,  1.53it/s]
3|6241|Loss: 0.9908460378646851:  96%|█████████▋| 6240/6470 [1:11:45<02:30,  1.53it/s]
3|6241|Loss: 0.9908460378646851:  96%|█████████▋| 6241/6470 [1:11:45<02:32,  1.50it/s]
3|6242|Loss: 0.8988679051399231:  96%|█████████▋| 6241/6470 [1:11:46<02:32,  1.50it/s]
3|6242|Loss: 0.8988679051399231:  96%|█████████▋| 6242/6470 [1:11:46<02:37,  1.45it/s]
3|6243|Loss: 0.7073968052864075:  96%|█████████▋| 6242/6470 [1:11:47<02:37,  1.45it/s]
3|6243|Loss: 0.7073968052864075:  96%|█████████▋| 6243/6470 [1:11:47<02:48,  1.35it/s]
3|6244|Loss: 0.6865565180778503:  96%|█████████▋| 6243/6470 [1:11:47<02:48,  1.35it/s]
3|6244|Loss: 0.6865565180778503:  97%|█████████▋| 6244/6470 [1:11:48<02:52,  1.31it/s]
3|6245|Loss: 0.8242183327674866:  97%|█████████▋| 6244/6470 [1:11:48<02:52,  1.31it/s]
3|6245|Loss: 0.8242183327674866:  97%|█████████▋| 6245/6470 [1:11:49<02:51,  1.31it/s]
3|6246|Loss: 0.7927605509757996:  97%|█████████▋| 6245/6470 [1:11:49<02:51,  1.31it/s]
3|6246|Loss: 0.7927605509757996:  97%|█████████▋| 6246/6470 [1:11:49<02:57,  1.26it/s]
3|6247|Loss: 0.9986271858215332:  97%|█████████▋| 6246/6470 [1:11:50<02:57,  1.26it/s]
3|6247|Loss: 0.9986271858215332:  97%|█████████▋| 6247/6470 [1:11:50<02:50,  1.31it/s]
3|6248|Loss: 0.8698844313621521:  97%|█████████▋| 6247/6470 [1:11:51<02:50,  1.31it/s]
3|6248|Loss: 0.8698844313621521:  97%|█████████▋| 6248/6470 [1:11:51<02:45,  1.34it/s]
3|6249|Loss: 0.7299770712852478:  97%|█████████▋| 6248/6470 [1:11:51<02:45,  1.34it/s]
3|6249|Loss: 0.7299770712852478:  97%|█████████▋| 6249/6470 [1:11:52<02:39,  1.39it/s]
3|6250|Loss: 0.8469730019569397:  97%|█████████▋| 6249/6470 [1:11:52<02:39,  1.39it/s]
3|6250|Loss: 0.8469730019569397:  97%|█████████▋| 6250/6470 [1:11:52<02:41,  1.36it/s]
3|6251|Loss: 0.8192031979560852:  97%|█████████▋| 6250/6470 [1:11:53<02:41,  1.36it/s]
3|6251|Loss: 0.8192031979560852:  97%|█████████▋| 6251/6470 [1:11:53<02:37,  1.39it/s]
3|6252|Loss: 0.9182260036468506:  97%|█████████▋| 6251/6470 [1:11:53<02:37,  1.39it/s]
3|6252|Loss: 0.9182260036468506:  97%|█████████▋| 6252/6470 [1:11:54<02:37,  1.38it/s]
3|6253|Loss: 0.8679613471031189:  97%|█████████▋| 6252/6470 [1:11:54<02:37,  1.38it/s]
3|6253|Loss: 0.8679613471031189:  97%|█████████▋| 6253/6470 [1:11:55<02:45,  1.31it/s]
3|6254|Loss: 0.7859141826629639:  97%|█████████▋| 6253/6470 [1:11:55<02:45,  1.31it/s]
3|6254|Loss: 0.7859141826629639:  97%|█████████▋| 6254/6470 [1:11:55<02:26,  1.48it/s]
3|6255|Loss: 0.8367078900337219:  97%|█████████▋| 6254/6470 [1:11:55<02:26,  1.48it/s]
3|6255|Loss: 0.8367078900337219:  97%|█████████▋| 6255/6470 [1:11:56<02:25,  1.48it/s]
3|6256|Loss: 0.9090463519096375:  97%|█████████▋| 6255/6470 [1:11:56<02:25,  1.48it/s]
3|6256|Loss: 0.9090463519096375:  97%|█████████▋| 6256/6470 [1:11:56<02:28,  1.44it/s]
3|6257|Loss: 0.6241551637649536:  97%|█████████▋| 6256/6470 [1:11:57<02:28,  1.44it/s]
3|6257|Loss: 0.6241551637649536:  97%|█████████▋| 6257/6470 [1:11:57<02:30,  1.42it/s]
3|6258|Loss: 0.794071614742279:  97%|█████████▋| 6257/6470 [1:11:58<02:30,  1.42it/s] 
3|6258|Loss: 0.794071614742279:  97%|█████████▋| 6258/6470 [1:11:58<02:26,  1.45it/s]
3|6259|Loss: 0.9019012451171875:  97%|█████████▋| 6258/6470 [1:11:58<02:26,  1.45it/s]
3|6259|Loss: 0.9019012451171875:  97%|█████████▋| 6259/6470 [1:11:58<02:21,  1.49it/s]
3|6260|Loss: 0.7589794397354126:  97%|█████████▋| 6259/6470 [1:11:59<02:21,  1.49it/s]
3|6260|Loss: 0.7589794397354126:  97%|█████████▋| 6260/6470 [1:11:59<02:29,  1.41it/s]
3|6261|Loss: 0.7598078846931458:  97%|█████████▋| 6260/6470 [1:12:00<02:29,  1.41it/s]
3|6261|Loss: 0.7598078846931458:  97%|█████████▋| 6261/6470 [1:12:00<02:31,  1.38it/s]
3|6262|Loss: 0.730647087097168:  97%|█████████▋| 6261/6470 [1:12:00<02:31,  1.38it/s] 
3|6262|Loss: 0.730647087097168:  97%|█████████▋| 6262/6470 [1:12:01<02:25,  1.43it/s]
3|6263|Loss: 0.7178843021392822:  97%|█████████▋| 6262/6470 [1:12:01<02:25,  1.43it/s]
3|6263|Loss: 0.7178843021392822:  97%|█████████▋| 6263/6470 [1:12:02<02:38,  1.30it/s]
3|6264|Loss: 0.8360769748687744:  97%|█████████▋| 6263/6470 [1:12:02<02:38,  1.30it/s]
3|6264|Loss: 0.8360769748687744:  97%|█████████▋| 6264/6470 [1:12:02<02:39,  1.29it/s]
3|6265|Loss: 0.8165904879570007:  97%|█████████▋| 6264/6470 [1:12:03<02:39,  1.29it/s]
3|6265|Loss: 0.8165904879570007:  97%|█████████▋| 6265/6470 [1:12:03<02:30,  1.36it/s]
3|6266|Loss: 0.7475425601005554:  97%|█████████▋| 6265/6470 [1:12:03<02:30,  1.36it/s]
3|6266|Loss: 0.7475425601005554:  97%|█████████▋| 6266/6470 [1:12:04<02:27,  1.38it/s]
3|6267|Loss: 0.7569001913070679:  97%|█████████▋| 6266/6470 [1:12:04<02:27,  1.38it/s]
3|6267|Loss: 0.7569001913070679:  97%|█████████▋| 6267/6470 [1:12:04<02:18,  1.47it/s]
3|6268|Loss: 0.8720232248306274:  97%|█████████▋| 6267/6470 [1:12:05<02:18,  1.47it/s]
3|6268|Loss: 0.8720232248306274:  97%|█████████▋| 6268/6470 [1:12:05<02:16,  1.48it/s]
3|6269|Loss: 0.851111888885498:  97%|█████████▋| 6268/6470 [1:12:05<02:16,  1.48it/s] 
3|6269|Loss: 0.851111888885498:  97%|█████████▋| 6269/6470 [1:12:06<02:22,  1.41it/s]
3|6270|Loss: 0.8153732419013977:  97%|█████████▋| 6269/6470 [1:12:06<02:22,  1.41it/s]
3|6270|Loss: 0.8153732419013977:  97%|█████████▋| 6270/6470 [1:12:06<02:16,  1.46it/s]
3|6271|Loss: 0.7709442377090454:  97%|█████████▋| 6270/6470 [1:12:07<02:16,  1.46it/s]
3|6271|Loss: 0.7709442377090454:  97%|█████████▋| 6271/6470 [1:12:07<02:15,  1.47it/s]
3|6272|Loss: 0.779483437538147:  97%|█████████▋| 6271/6470 [1:12:07<02:15,  1.47it/s] 
3|6272|Loss: 0.779483437538147:  97%|█████████▋| 6272/6470 [1:12:08<02:17,  1.44it/s]
3|6273|Loss: 0.8886952996253967:  97%|█████████▋| 6272/6470 [1:12:08<02:17,  1.44it/s]
3|6273|Loss: 0.8886952996253967:  97%|█████████▋| 6273/6470 [1:12:08<02:08,  1.53it/s]
3|6274|Loss: 0.7979695796966553:  97%|█████████▋| 6273/6470 [1:12:09<02:08,  1.53it/s]
3|6274|Loss: 0.7979695796966553:  97%|█████████▋| 6274/6470 [1:12:09<02:00,  1.63it/s]
3|6275|Loss: 1.034651517868042:  97%|█████████▋| 6274/6470 [1:12:09<02:00,  1.63it/s] 
3|6275|Loss: 1.034651517868042:  97%|█████████▋| 6275/6470 [1:12:09<01:56,  1.67it/s]
3|6276|Loss: 0.8801526427268982:  97%|█████████▋| 6275/6470 [1:12:10<01:56,  1.67it/s]
3|6276|Loss: 0.8801526427268982:  97%|█████████▋| 6276/6470 [1:12:10<02:00,  1.60it/s]
3|6277|Loss: 0.8170509338378906:  97%|█████████▋| 6276/6470 [1:12:11<02:00,  1.60it/s]
3|6277|Loss: 0.8170509338378906:  97%|█████████▋| 6277/6470 [1:12:11<02:10,  1.47it/s]
3|6278|Loss: 0.8838199973106384:  97%|█████████▋| 6277/6470 [1:12:11<02:10,  1.47it/s]
3|6278|Loss: 0.8838199973106384:  97%|█████████▋| 6278/6470 [1:12:12<02:13,  1.44it/s]
3|6279|Loss: 0.8630114793777466:  97%|█████████▋| 6278/6470 [1:12:12<02:13,  1.44it/s]
3|6279|Loss: 0.8630114793777466:  97%|█████████▋| 6279/6470 [1:12:12<02:06,  1.51it/s]
3|6280|Loss: 0.8602347373962402:  97%|█████████▋| 6279/6470 [1:12:13<02:06,  1.51it/s]
3|6280|Loss: 0.8602347373962402:  97%|█████████▋| 6280/6470 [1:12:13<02:07,  1.49it/s]
3|6281|Loss: 0.9113223552703857:  97%|█████████▋| 6280/6470 [1:12:13<02:07,  1.49it/s]
3|6281|Loss: 0.9113223552703857:  97%|█████████▋| 6281/6470 [1:12:14<02:09,  1.46it/s]
3|6282|Loss: 0.910517156124115:  97%|█████████▋| 6281/6470 [1:12:14<02:09,  1.46it/s] 
3|6282|Loss: 0.910517156124115:  97%|█████████▋| 6282/6470 [1:12:14<02:16,  1.38it/s]
3|6283|Loss: 0.8950615525245667:  97%|█████████▋| 6282/6470 [1:12:15<02:16,  1.38it/s]
3|6283|Loss: 0.8950615525245667:  97%|█████████▋| 6283/6470 [1:12:15<02:13,  1.40it/s]
3|6284|Loss: 0.7100040912628174:  97%|█████████▋| 6283/6470 [1:12:16<02:13,  1.40it/s]
3|6284|Loss: 0.7100040912628174:  97%|█████████▋| 6284/6470 [1:12:16<02:12,  1.40it/s]
3|6285|Loss: 0.927492082118988:  97%|█████████▋| 6284/6470 [1:12:16<02:12,  1.40it/s] 
3|6285|Loss: 0.927492082118988:  97%|█████████▋| 6285/6470 [1:12:17<02:09,  1.43it/s]
3|6286|Loss: 0.7382283210754395:  97%|█████████▋| 6285/6470 [1:12:17<02:09,  1.43it/s]
3|6286|Loss: 0.7382283210754395:  97%|█████████▋| 6286/6470 [1:12:17<02:07,  1.44it/s]
3|6287|Loss: 0.8695564866065979:  97%|█████████▋| 6286/6470 [1:12:18<02:07,  1.44it/s]
3|6287|Loss: 0.8695564866065979:  97%|█████████▋| 6287/6470 [1:12:18<02:03,  1.48it/s]
3|6288|Loss: 0.9315126538276672:  97%|█████████▋| 6287/6470 [1:12:18<02:03,  1.48it/s]
3|6288|Loss: 0.9315126538276672:  97%|█████████▋| 6288/6470 [1:12:18<01:55,  1.57it/s]
3|6289|Loss: 0.9425129294395447:  97%|█████████▋| 6288/6470 [1:12:19<01:55,  1.57it/s]
3|6289|Loss: 0.9425129294395447:  97%|█████████▋| 6289/6470 [1:12:19<02:02,  1.48it/s]
3|6290|Loss: 0.9778281450271606:  97%|█████████▋| 6289/6470 [1:12:20<02:02,  1.48it/s]
3|6290|Loss: 0.9778281450271606:  97%|█████████▋| 6290/6470 [1:12:20<02:05,  1.43it/s]
3|6291|Loss: 0.9149274826049805:  97%|█████████▋| 6290/6470 [1:12:20<02:05,  1.43it/s]
3|6291|Loss: 0.9149274826049805:  97%|█████████▋| 6291/6470 [1:12:21<02:09,  1.38it/s]
3|6292|Loss: 0.8231788277626038:  97%|█████████▋| 6291/6470 [1:12:21<02:09,  1.38it/s]
3|6292|Loss: 0.8231788277626038:  97%|█████████▋| 6292/6470 [1:12:21<02:06,  1.40it/s]
3|6293|Loss: 0.7799203991889954:  97%|█████████▋| 6292/6470 [1:12:22<02:06,  1.40it/s]
3|6293|Loss: 0.7799203991889954:  97%|█████████▋| 6293/6470 [1:12:22<02:05,  1.42it/s]
3|6294|Loss: 0.852632462978363:  97%|█████████▋| 6293/6470 [1:12:22<02:05,  1.42it/s] 
3|6294|Loss: 0.852632462978363:  97%|█████████▋| 6294/6470 [1:12:23<01:54,  1.53it/s]
3|6295|Loss: 0.7762930393218994:  97%|█████████▋| 6294/6470 [1:12:23<01:54,  1.53it/s]
3|6295|Loss: 0.7762930393218994:  97%|█████████▋| 6295/6470 [1:12:23<01:48,  1.61it/s]
3|6296|Loss: 0.9275848865509033:  97%|█████████▋| 6295/6470 [1:12:23<01:48,  1.61it/s]
3|6296|Loss: 0.9275848865509033:  97%|█████████▋| 6296/6470 [1:12:24<01:48,  1.60it/s]
3|6297|Loss: 0.8750298023223877:  97%|█████████▋| 6296/6470 [1:12:24<01:48,  1.60it/s]
3|6297|Loss: 0.8750298023223877:  97%|█████████▋| 6297/6470 [1:12:24<01:53,  1.53it/s]
3|6298|Loss: 0.8028365969657898:  97%|█████████▋| 6297/6470 [1:12:25<01:53,  1.53it/s]
3|6298|Loss: 0.8028365969657898:  97%|█████████▋| 6298/6470 [1:12:25<01:52,  1.53it/s]
3|6299|Loss: 0.8704743385314941:  97%|█████████▋| 6298/6470 [1:12:26<01:52,  1.53it/s]
3|6299|Loss: 0.8704743385314941:  97%|█████████▋| 6299/6470 [1:12:26<01:53,  1.51it/s]
3|6300|Loss: 0.9224533438682556:  97%|█████████▋| 6299/6470 [1:12:26<01:53,  1.51it/s]
3|6300|Loss: 0.9224533438682556:  97%|█████████▋| 6300/6470 [1:12:27<02:00,  1.41it/s]
3|6301|Loss: 0.8228195309638977:  97%|█████████▋| 6300/6470 [1:12:27<02:00,  1.41it/s]
3|6301|Loss: 0.8228195309638977:  97%|█████████▋| 6301/6470 [1:12:27<02:05,  1.35it/s]
3|6302|Loss: 0.8482627868652344:  97%|█████████▋| 6301/6470 [1:12:28<02:05,  1.35it/s]
3|6302|Loss: 0.8482627868652344:  97%|█████████▋| 6302/6470 [1:12:28<02:08,  1.31it/s]
3|6303|Loss: 1.2454252243041992:  97%|█████████▋| 6302/6470 [1:12:29<02:08,  1.31it/s]
3|6303|Loss: 1.2454252243041992:  97%|█████████▋| 6303/6470 [1:12:29<02:09,  1.29it/s]
3|6304|Loss: 0.8470509052276611:  97%|█████████▋| 6303/6470 [1:12:30<02:09,  1.29it/s]
3|6304|Loss: 0.8470509052276611:  97%|█████████▋| 6304/6470 [1:12:30<02:08,  1.29it/s]
3|6305|Loss: 0.8767181634902954:  97%|█████████▋| 6304/6470 [1:12:30<02:08,  1.29it/s]
3|6305|Loss: 0.8767181634902954:  97%|█████████▋| 6305/6470 [1:12:31<02:06,  1.30it/s]
3|6306|Loss: 0.8623389601707458:  97%|█████████▋| 6305/6470 [1:12:31<02:06,  1.30it/s]
3|6306|Loss: 0.8623389601707458:  97%|█████████▋| 6306/6470 [1:12:31<01:55,  1.42it/s]
3|6307|Loss: 0.8124240040779114:  97%|█████████▋| 6306/6470 [1:12:32<01:55,  1.42it/s]
3|6307|Loss: 0.8124240040779114:  97%|█████████▋| 6307/6470 [1:12:32<01:58,  1.38it/s]
3|6308|Loss: 0.7801116108894348:  97%|█████████▋| 6307/6470 [1:12:32<01:58,  1.38it/s]
3|6308|Loss: 0.7801116108894348:  97%|█████████▋| 6308/6470 [1:12:33<01:54,  1.41it/s]
3|6309|Loss: 0.9208554625511169:  97%|█████████▋| 6308/6470 [1:12:33<01:54,  1.41it/s]
3|6309|Loss: 0.9208554625511169:  98%|█████████▊| 6309/6470 [1:12:33<01:53,  1.41it/s]
3|6310|Loss: 0.768200695514679:  98%|█████████▊| 6309/6470 [1:12:34<01:53,  1.41it/s] 
3|6310|Loss: 0.768200695514679:  98%|█████████▊| 6310/6470 [1:12:34<01:40,  1.59it/s]
3|6311|Loss: 0.9245865345001221:  98%|█████████▊| 6310/6470 [1:12:34<01:40,  1.59it/s]
3|6311|Loss: 0.9245865345001221:  98%|█████████▊| 6311/6470 [1:12:34<01:38,  1.62it/s]
3|6312|Loss: 0.7585414052009583:  98%|█████████▊| 6311/6470 [1:12:35<01:38,  1.62it/s]
3|6312|Loss: 0.7585414052009583:  98%|█████████▊| 6312/6470 [1:12:35<01:37,  1.62it/s]
3|6313|Loss: 0.8709887862205505:  98%|█████████▊| 6312/6470 [1:12:35<01:37,  1.62it/s]
3|6313|Loss: 0.8709887862205505:  98%|█████████▊| 6313/6470 [1:12:36<01:35,  1.65it/s]
3|6314|Loss: 0.7955721616744995:  98%|█████████▊| 6313/6470 [1:12:36<01:35,  1.65it/s]
3|6314|Loss: 0.7955721616744995:  98%|█████████▊| 6314/6470 [1:12:36<01:43,  1.51it/s]
3|6315|Loss: 0.917369544506073:  98%|█████████▊| 6314/6470 [1:12:37<01:43,  1.51it/s] 
3|6315|Loss: 0.917369544506073:  98%|█████████▊| 6315/6470 [1:12:37<01:51,  1.39it/s]
3|6316|Loss: 0.6660138368606567:  98%|█████████▊| 6315/6470 [1:12:38<01:51,  1.39it/s]
3|6316|Loss: 0.6660138368606567:  98%|█████████▊| 6316/6470 [1:12:38<01:48,  1.42it/s]
3|6317|Loss: 0.8093597888946533:  98%|█████████▊| 6316/6470 [1:12:38<01:48,  1.42it/s]
3|6317|Loss: 0.8093597888946533:  98%|█████████▊| 6317/6470 [1:12:38<01:43,  1.48it/s]
3|6318|Loss: 0.7844997644424438:  98%|█████████▊| 6317/6470 [1:12:39<01:43,  1.48it/s]
3|6318|Loss: 0.7844997644424438:  98%|█████████▊| 6318/6470 [1:12:39<01:44,  1.46it/s]
3|6319|Loss: 0.909643292427063:  98%|█████████▊| 6318/6470 [1:12:40<01:44,  1.46it/s] 
3|6319|Loss: 0.909643292427063:  98%|█████████▊| 6319/6470 [1:12:40<01:46,  1.42it/s]
3|6320|Loss: 0.9629778265953064:  98%|█████████▊| 6319/6470 [1:12:40<01:46,  1.42it/s]
3|6320|Loss: 0.9629778265953064:  98%|█████████▊| 6320/6470 [1:12:41<01:43,  1.44it/s]
3|6321|Loss: 0.8550435900688171:  98%|█████████▊| 6320/6470 [1:12:41<01:43,  1.44it/s]
3|6321|Loss: 0.8550435900688171:  98%|█████████▊| 6321/6470 [1:12:41<01:40,  1.48it/s]
3|6322|Loss: 0.9495555758476257:  98%|█████████▊| 6321/6470 [1:12:42<01:40,  1.48it/s]
3|6322|Loss: 0.9495555758476257:  98%|█████████▊| 6322/6470 [1:12:42<01:35,  1.55it/s]
3|6323|Loss: 1.081089973449707:  98%|█████████▊| 6322/6470 [1:12:42<01:35,  1.55it/s] 
3|6323|Loss: 1.081089973449707:  98%|█████████▊| 6323/6470 [1:12:42<01:33,  1.57it/s]
3|6324|Loss: 1.0006791353225708:  98%|█████████▊| 6323/6470 [1:12:43<01:33,  1.57it/s]
3|6324|Loss: 1.0006791353225708:  98%|█████████▊| 6324/6470 [1:12:43<01:32,  1.58it/s]
3|6325|Loss: 0.740607738494873:  98%|█████████▊| 6324/6470 [1:12:43<01:32,  1.58it/s] 
3|6325|Loss: 0.740607738494873:  98%|█████████▊| 6325/6470 [1:12:44<01:39,  1.46it/s]
3|6326|Loss: 0.8490931391716003:  98%|█████████▊| 6325/6470 [1:12:44<01:39,  1.46it/s]
3|6326|Loss: 0.8490931391716003:  98%|█████████▊| 6326/6470 [1:12:44<01:34,  1.52it/s]
3|6327|Loss: 0.8421981930732727:  98%|█████████▊| 6326/6470 [1:12:45<01:34,  1.52it/s]
3|6327|Loss: 0.8421981930732727:  98%|█████████▊| 6327/6470 [1:12:45<01:34,  1.51it/s]
3|6328|Loss: 0.8020400404930115:  98%|█████████▊| 6327/6470 [1:12:46<01:34,  1.51it/s]
3|6328|Loss: 0.8020400404930115:  98%|█████████▊| 6328/6470 [1:12:46<01:41,  1.41it/s]
3|6329|Loss: 0.8924670219421387:  98%|█████████▊| 6328/6470 [1:12:46<01:41,  1.41it/s]
3|6329|Loss: 0.8924670219421387:  98%|█████████▊| 6329/6470 [1:12:47<01:46,  1.32it/s]
3|6330|Loss: 0.8315277695655823:  98%|█████████▊| 6329/6470 [1:12:47<01:46,  1.32it/s]
3|6330|Loss: 0.8315277695655823:  98%|█████████▊| 6330/6470 [1:12:48<01:46,  1.31it/s]
3|6331|Loss: 0.8369850516319275:  98%|█████████▊| 6330/6470 [1:12:48<01:46,  1.31it/s]
3|6331|Loss: 0.8369850516319275:  98%|█████████▊| 6331/6470 [1:12:48<01:38,  1.42it/s]
3|6332|Loss: 0.7238962054252625:  98%|█████████▊| 6331/6470 [1:12:48<01:38,  1.42it/s]
3|6332|Loss: 0.7238962054252625:  98%|█████████▊| 6332/6470 [1:12:49<01:33,  1.47it/s]
3|6333|Loss: 0.8313508033752441:  98%|█████████▊| 6332/6470 [1:12:49<01:33,  1.47it/s]
3|6333|Loss: 0.8313508033752441:  98%|█████████▊| 6333/6470 [1:12:49<01:33,  1.46it/s]
3|6334|Loss: 0.7308210730552673:  98%|█████████▊| 6333/6470 [1:12:50<01:33,  1.46it/s]
3|6334|Loss: 0.7308210730552673:  98%|█████████▊| 6334/6470 [1:12:50<01:31,  1.48it/s]
3|6335|Loss: 0.8970984220504761:  98%|█████████▊| 6334/6470 [1:12:50<01:31,  1.48it/s]
3|6335|Loss: 0.8970984220504761:  98%|█████████▊| 6335/6470 [1:12:51<01:27,  1.55it/s]
3|6336|Loss: 0.6812278032302856:  98%|█████████▊| 6335/6470 [1:12:51<01:27,  1.55it/s]
3|6336|Loss: 0.6812278032302856:  98%|█████████▊| 6336/6470 [1:12:51<01:30,  1.48it/s]
3|6337|Loss: 1.014286756515503:  98%|█████████▊| 6336/6470 [1:12:52<01:30,  1.48it/s] 
3|6337|Loss: 1.014286756515503:  98%|█████████▊| 6337/6470 [1:12:52<01:34,  1.41it/s]
3|6338|Loss: 0.7038764953613281:  98%|█████████▊| 6337/6470 [1:12:53<01:34,  1.41it/s]
3|6338|Loss: 0.7038764953613281:  98%|█████████▊| 6338/6470 [1:12:53<01:40,  1.32it/s]
3|6339|Loss: 0.9018408060073853:  98%|█████████▊| 6338/6470 [1:12:54<01:40,  1.32it/s]
3|6339|Loss: 0.9018408060073853:  98%|█████████▊| 6339/6470 [1:12:54<01:37,  1.35it/s]
3|6340|Loss: 0.9554662108421326:  98%|█████████▊| 6339/6470 [1:12:54<01:37,  1.35it/s]
3|6340|Loss: 0.9554662108421326:  98%|█████████▊| 6340/6470 [1:12:54<01:28,  1.47it/s]
3|6341|Loss: 0.901656448841095:  98%|█████████▊| 6340/6470 [1:12:55<01:28,  1.47it/s] 
3|6341|Loss: 0.901656448841095:  98%|█████████▊| 6341/6470 [1:12:55<01:28,  1.46it/s]
3|6342|Loss: 0.9474043846130371:  98%|█████████▊| 6341/6470 [1:12:55<01:28,  1.46it/s]
3|6342|Loss: 0.9474043846130371:  98%|█████████▊| 6342/6470 [1:12:56<01:27,  1.47it/s]
3|6343|Loss: 0.9725271463394165:  98%|█████████▊| 6342/6470 [1:12:56<01:27,  1.47it/s]
3|6343|Loss: 0.9725271463394165:  98%|█████████▊| 6343/6470 [1:12:56<01:22,  1.53it/s]
3|6344|Loss: 0.8695995807647705:  98%|█████████▊| 6343/6470 [1:12:57<01:22,  1.53it/s]
3|6344|Loss: 0.8695995807647705:  98%|█████████▊| 6344/6470 [1:12:57<01:21,  1.54it/s]
3|6345|Loss: 0.8087141513824463:  98%|█████████▊| 6344/6470 [1:12:57<01:21,  1.54it/s]
3|6345|Loss: 0.8087141513824463:  98%|█████████▊| 6345/6470 [1:12:58<01:22,  1.51it/s]
3|6346|Loss: 0.727506160736084:  98%|█████████▊| 6345/6470 [1:12:58<01:22,  1.51it/s] 
3|6346|Loss: 0.727506160736084:  98%|█████████▊| 6346/6470 [1:12:58<01:27,  1.42it/s]
3|6347|Loss: 0.8406661748886108:  98%|█████████▊| 6346/6470 [1:12:59<01:27,  1.42it/s]
3|6347|Loss: 0.8406661748886108:  98%|█████████▊| 6347/6470 [1:12:59<01:28,  1.39it/s]
3|6348|Loss: 0.9486878514289856:  98%|█████████▊| 6347/6470 [1:13:00<01:28,  1.39it/s]
3|6348|Loss: 0.9486878514289856:  98%|█████████▊| 6348/6470 [1:13:00<01:25,  1.43it/s]
3|6349|Loss: 0.7869730591773987:  98%|█████████▊| 6348/6470 [1:13:00<01:25,  1.43it/s]
3|6349|Loss: 0.7869730591773987:  98%|█████████▊| 6349/6470 [1:13:01<01:23,  1.45it/s]
3|6350|Loss: 0.9163219332695007:  98%|█████████▊| 6349/6470 [1:13:01<01:23,  1.45it/s]
3|6350|Loss: 0.9163219332695007:  98%|█████████▊| 6350/6470 [1:13:01<01:19,  1.50it/s]
3|6351|Loss: 0.9403943419456482:  98%|█████████▊| 6350/6470 [1:13:01<01:19,  1.50it/s]
3|6351|Loss: 0.9403943419456482:  98%|█████████▊| 6351/6470 [1:13:02<01:17,  1.53it/s]
3|6352|Loss: 0.8403720855712891:  98%|█████████▊| 6351/6470 [1:13:02<01:17,  1.53it/s]
3|6352|Loss: 0.8403720855712891:  98%|█████████▊| 6352/6470 [1:13:02<01:16,  1.55it/s]
3|6353|Loss: 0.9228445291519165:  98%|█████████▊| 6352/6470 [1:13:03<01:16,  1.55it/s]
3|6353|Loss: 0.9228445291519165:  98%|█████████▊| 6353/6470 [1:13:03<01:18,  1.49it/s]
3|6354|Loss: 0.8721632957458496:  98%|█████████▊| 6353/6470 [1:13:04<01:18,  1.49it/s]
3|6354|Loss: 0.8721632957458496:  98%|█████████▊| 6354/6470 [1:13:04<01:19,  1.46it/s]
3|6355|Loss: 0.9726002216339111:  98%|█████████▊| 6354/6470 [1:13:04<01:19,  1.46it/s]
3|6355|Loss: 0.9726002216339111:  98%|█████████▊| 6355/6470 [1:13:04<01:17,  1.48it/s]
3|6356|Loss: 0.673896849155426:  98%|█████████▊| 6355/6470 [1:13:05<01:17,  1.48it/s] 
3|6356|Loss: 0.673896849155426:  98%|█████████▊| 6356/6470 [1:13:05<01:20,  1.41it/s]
3|6357|Loss: 0.7581193447113037:  98%|█████████▊| 6356/6470 [1:13:06<01:20,  1.41it/s]
3|6357|Loss: 0.7581193447113037:  98%|█████████▊| 6357/6470 [1:13:06<01:16,  1.48it/s]
3|6358|Loss: 0.7881819009780884:  98%|█████████▊| 6357/6470 [1:13:06<01:16,  1.48it/s]
3|6358|Loss: 0.7881819009780884:  98%|█████████▊| 6358/6470 [1:13:06<01:12,  1.55it/s]
3|6359|Loss: 1.0373822450637817:  98%|█████████▊| 6358/6470 [1:13:07<01:12,  1.55it/s]
3|6359|Loss: 1.0373822450637817:  98%|█████████▊| 6359/6470 [1:13:07<01:13,  1.51it/s]
3|6360|Loss: 0.9463558793067932:  98%|█████████▊| 6359/6470 [1:13:07<01:13,  1.51it/s]
3|6360|Loss: 0.9463558793067932:  98%|█████████▊| 6360/6470 [1:13:08<01:10,  1.56it/s]
3|6361|Loss: 0.7666991949081421:  98%|█████████▊| 6360/6470 [1:13:08<01:10,  1.56it/s]
3|6361|Loss: 0.7666991949081421:  98%|█████████▊| 6361/6470 [1:13:08<01:05,  1.65it/s]
3|6362|Loss: 0.9184131622314453:  98%|█████████▊| 6361/6470 [1:13:09<01:05,  1.65it/s]
3|6362|Loss: 0.9184131622314453:  98%|█████████▊| 6362/6470 [1:13:09<01:02,  1.73it/s]
3|6363|Loss: 0.9920880794525146:  98%|█████████▊| 6362/6470 [1:13:09<01:02,  1.73it/s]
3|6363|Loss: 0.9920880794525146:  98%|█████████▊| 6363/6470 [1:13:09<01:01,  1.74it/s]
3|6364|Loss: 0.8491705656051636:  98%|█████████▊| 6363/6470 [1:13:10<01:01,  1.74it/s]
3|6364|Loss: 0.8491705656051636:  98%|█████████▊| 6364/6470 [1:13:10<01:07,  1.57it/s]
3|6365|Loss: 0.8649688363075256:  98%|█████████▊| 6364/6470 [1:13:10<01:07,  1.57it/s]
3|6365|Loss: 0.8649688363075256:  98%|█████████▊| 6365/6470 [1:13:11<01:05,  1.60it/s]
3|6366|Loss: 0.8046611547470093:  98%|█████████▊| 6365/6470 [1:13:11<01:05,  1.60it/s]
3|6366|Loss: 0.8046611547470093:  98%|█████████▊| 6366/6470 [1:13:12<01:10,  1.47it/s]
3|6367|Loss: 0.8550969362258911:  98%|█████████▊| 6366/6470 [1:13:12<01:10,  1.47it/s]
3|6367|Loss: 0.8550969362258911:  98%|█████████▊| 6367/6470 [1:13:12<01:10,  1.45it/s]
3|6368|Loss: 0.9395997524261475:  98%|█████████▊| 6367/6470 [1:13:13<01:10,  1.45it/s]
3|6368|Loss: 0.9395997524261475:  98%|█████████▊| 6368/6470 [1:13:13<01:11,  1.43it/s]
3|6369|Loss: 1.05039381980896:  98%|█████████▊| 6368/6470 [1:13:13<01:11,  1.43it/s]  
3|6369|Loss: 1.05039381980896:  98%|█████████▊| 6369/6470 [1:13:14<01:11,  1.41it/s]
3|6370|Loss: 0.9044346213340759:  98%|█████████▊| 6369/6470 [1:13:14<01:11,  1.41it/s]
3|6370|Loss: 0.9044346213340759:  98%|█████████▊| 6370/6470 [1:13:14<01:11,  1.39it/s]
3|6371|Loss: 0.9431406259536743:  98%|█████████▊| 6370/6470 [1:13:15<01:11,  1.39it/s]
3|6371|Loss: 0.9431406259536743:  98%|█████████▊| 6371/6470 [1:13:15<01:05,  1.52it/s]
3|6372|Loss: 0.8515027761459351:  98%|█████████▊| 6371/6470 [1:13:15<01:05,  1.52it/s]
3|6372|Loss: 0.8515027761459351:  98%|█████████▊| 6372/6470 [1:13:16<01:04,  1.51it/s]
3|6373|Loss: 0.7719211578369141:  98%|█████████▊| 6372/6470 [1:13:16<01:04,  1.51it/s]
3|6373|Loss: 0.7719211578369141:  99%|█████████▊| 6373/6470 [1:13:16<01:08,  1.41it/s]
3|6374|Loss: 0.8915390372276306:  99%|█████████▊| 6373/6470 [1:13:17<01:08,  1.41it/s]
3|6374|Loss: 0.8915390372276306:  99%|█████████▊| 6374/6470 [1:13:17<01:09,  1.37it/s]
3|6375|Loss: 0.8734936714172363:  99%|█████████▊| 6374/6470 [1:13:18<01:09,  1.37it/s]
3|6375|Loss: 0.8734936714172363:  99%|█████████▊| 6375/6470 [1:13:18<01:11,  1.32it/s]
3|6376|Loss: 0.8198602795600891:  99%|█████████▊| 6375/6470 [1:13:18<01:11,  1.32it/s]
3|6376|Loss: 0.8198602795600891:  99%|█████████▊| 6376/6470 [1:13:19<01:08,  1.37it/s]
3|6377|Loss: 0.8516820073127747:  99%|█████████▊| 6376/6470 [1:13:19<01:08,  1.37it/s]
3|6377|Loss: 0.8516820073127747:  99%|█████████▊| 6377/6470 [1:13:19<01:05,  1.42it/s]
3|6378|Loss: 0.8458243608474731:  99%|█████████▊| 6377/6470 [1:13:20<01:05,  1.42it/s]
3|6378|Loss: 0.8458243608474731:  99%|█████████▊| 6378/6470 [1:13:20<01:04,  1.43it/s]
3|6379|Loss: 0.7805864810943604:  99%|█████████▊| 6378/6470 [1:13:20<01:04,  1.43it/s]
3|6379|Loss: 0.7805864810943604:  99%|█████████▊| 6379/6470 [1:13:21<01:04,  1.40it/s]
3|6380|Loss: 1.0524015426635742:  99%|█████████▊| 6379/6470 [1:13:21<01:04,  1.40it/s]
3|6380|Loss: 1.0524015426635742:  99%|█████████▊| 6380/6470 [1:13:21<01:02,  1.44it/s]
3|6381|Loss: 0.934540331363678:  99%|█████████▊| 6380/6470 [1:13:22<01:02,  1.44it/s] 
3|6381|Loss: 0.934540331363678:  99%|█████████▊| 6381/6470 [1:13:22<01:02,  1.43it/s]
3|6382|Loss: 0.74470055103302:  99%|█████████▊| 6381/6470 [1:13:23<01:02,  1.43it/s] 
3|6382|Loss: 0.74470055103302:  99%|█████████▊| 6382/6470 [1:13:23<01:02,  1.41it/s]
3|6383|Loss: 0.8826168179512024:  99%|█████████▊| 6382/6470 [1:13:23<01:02,  1.41it/s]
3|6383|Loss: 0.8826168179512024:  99%|█████████▊| 6383/6470 [1:13:23<00:56,  1.54it/s]
3|6384|Loss: 0.8442435264587402:  99%|█████████▊| 6383/6470 [1:13:24<00:56,  1.54it/s]
3|6384|Loss: 0.8442435264587402:  99%|█████████▊| 6384/6470 [1:13:24<00:58,  1.48it/s]
3|6385|Loss: 0.7958747148513794:  99%|█████████▊| 6384/6470 [1:13:25<00:58,  1.48it/s]
3|6385|Loss: 0.7958747148513794:  99%|█████████▊| 6385/6470 [1:13:25<01:00,  1.39it/s]
3|6386|Loss: 1.0162416696548462:  99%|█████████▊| 6385/6470 [1:13:25<01:00,  1.39it/s]
3|6386|Loss: 1.0162416696548462:  99%|█████████▊| 6386/6470 [1:13:26<00:58,  1.43it/s]
3|6387|Loss: 0.924170732498169:  99%|█████████▊| 6386/6470 [1:13:26<00:58,  1.43it/s] 
3|6387|Loss: 0.924170732498169:  99%|█████████▊| 6387/6470 [1:13:26<00:55,  1.48it/s]
3|6388|Loss: 0.8893398642539978:  99%|█████████▊| 6387/6470 [1:13:27<00:55,  1.48it/s]
3|6388|Loss: 0.8893398642539978:  99%|█████████▊| 6388/6470 [1:13:27<00:58,  1.40it/s]
3|6389|Loss: 0.8217676877975464:  99%|█████████▊| 6388/6470 [1:13:27<00:58,  1.40it/s]
3|6389|Loss: 0.8217676877975464:  99%|█████████▊| 6389/6470 [1:13:28<01:00,  1.35it/s]
3|6390|Loss: 0.8232317566871643:  99%|█████████▊| 6389/6470 [1:13:28<01:00,  1.35it/s]
3|6390|Loss: 0.8232317566871643:  99%|█████████▉| 6390/6470 [1:13:28<00:56,  1.41it/s]
3|6391|Loss: 0.8202244639396667:  99%|█████████▉| 6390/6470 [1:13:29<00:56,  1.41it/s]
3|6391|Loss: 0.8202244639396667:  99%|█████████▉| 6391/6470 [1:13:29<00:57,  1.38it/s]
3|6392|Loss: 0.9252762794494629:  99%|█████████▉| 6391/6470 [1:13:30<00:57,  1.38it/s]
3|6392|Loss: 0.9252762794494629:  99%|█████████▉| 6392/6470 [1:13:30<00:59,  1.31it/s]
3|6393|Loss: 1.0282511711120605:  99%|█████████▉| 6392/6470 [1:13:31<00:59,  1.31it/s]
3|6393|Loss: 1.0282511711120605:  99%|█████████▉| 6393/6470 [1:13:31<01:00,  1.27it/s]
3|6394|Loss: 0.8449267148971558:  99%|█████████▉| 6393/6470 [1:13:31<01:00,  1.27it/s]
3|6394|Loss: 0.8449267148971558:  99%|█████████▉| 6394/6470 [1:13:32<00:58,  1.30it/s]
3|6395|Loss: 0.7629107236862183:  99%|█████████▉| 6394/6470 [1:13:32<00:58,  1.30it/s]
3|6395|Loss: 0.7629107236862183:  99%|█████████▉| 6395/6470 [1:13:32<00:53,  1.39it/s]
3|6396|Loss: 0.8018407821655273:  99%|█████████▉| 6395/6470 [1:13:33<00:53,  1.39it/s]
3|6396|Loss: 0.8018407821655273:  99%|█████████▉| 6396/6470 [1:13:33<00:52,  1.41it/s]
3|6397|Loss: 0.9361807107925415:  99%|█████████▉| 6396/6470 [1:13:33<00:52,  1.41it/s]
3|6397|Loss: 0.9361807107925415:  99%|█████████▉| 6397/6470 [1:13:33<00:48,  1.50it/s]
3|6398|Loss: 0.8204450011253357:  99%|█████████▉| 6397/6470 [1:13:34<00:48,  1.50it/s]
3|6398|Loss: 0.8204450011253357:  99%|█████████▉| 6398/6470 [1:13:34<00:48,  1.48it/s]
3|6399|Loss: 0.8587158918380737:  99%|█████████▉| 6398/6470 [1:13:35<00:48,  1.48it/s]
3|6399|Loss: 0.8587158918380737:  99%|█████████▉| 6399/6470 [1:13:35<00:46,  1.53it/s]
3|6400|Loss: 0.8287570476531982:  99%|█████████▉| 6399/6470 [1:13:35<00:46,  1.53it/s]
3|6400|Loss: 0.8287570476531982:  99%|█████████▉| 6400/6470 [1:13:35<00:42,  1.64it/s]
3|6401|Loss: 0.8037804365158081:  99%|█████████▉| 6400/6470 [1:13:36<00:42,  1.64it/s]
3|6401|Loss: 0.8037804365158081:  99%|█████████▉| 6401/6470 [1:13:36<00:37,  1.82it/s]
3|6402|Loss: 0.8782227039337158:  99%|█████████▉| 6401/6470 [1:13:36<00:37,  1.82it/s]
3|6402|Loss: 0.8782227039337158:  99%|█████████▉| 6402/6470 [1:13:36<00:42,  1.61it/s]
3|6403|Loss: 0.811238169670105:  99%|█████████▉| 6402/6470 [1:13:37<00:42,  1.61it/s] 
3|6403|Loss: 0.811238169670105:  99%|█████████▉| 6403/6470 [1:13:37<00:45,  1.47it/s]
3|6404|Loss: 0.9227721095085144:  99%|█████████▉| 6403/6470 [1:13:38<00:45,  1.47it/s]
3|6404|Loss: 0.9227721095085144:  99%|█████████▉| 6404/6470 [1:13:38<00:46,  1.43it/s]
3|6405|Loss: 0.723484992980957:  99%|█████████▉| 6404/6470 [1:13:38<00:46,  1.43it/s] 
3|6405|Loss: 0.723484992980957:  99%|█████████▉| 6405/6470 [1:13:39<00:46,  1.41it/s]
3|6406|Loss: 0.7956424951553345:  99%|█████████▉| 6405/6470 [1:13:39<00:46,  1.41it/s]
3|6406|Loss: 0.7956424951553345:  99%|█████████▉| 6406/6470 [1:13:40<00:45,  1.41it/s]
3|6407|Loss: 0.9620336890220642:  99%|█████████▉| 6406/6470 [1:13:40<00:45,  1.41it/s]
3|6407|Loss: 0.9620336890220642:  99%|█████████▉| 6407/6470 [1:13:40<00:46,  1.34it/s]
3|6408|Loss: 0.864822506904602:  99%|█████████▉| 6407/6470 [1:13:41<00:46,  1.34it/s] 
3|6408|Loss: 0.864822506904602:  99%|█████████▉| 6408/6470 [1:13:41<00:43,  1.41it/s]
3|6409|Loss: 1.007672667503357:  99%|█████████▉| 6408/6470 [1:13:41<00:43,  1.41it/s]
3|6409|Loss: 1.007672667503357:  99%|█████████▉| 6409/6470 [1:13:42<00:44,  1.36it/s]
3|6410|Loss: 0.9425745010375977:  99%|█████████▉| 6409/6470 [1:13:42<00:44,  1.36it/s]
3|6410|Loss: 0.9425745010375977:  99%|█████████▉| 6410/6470 [1:13:43<00:44,  1.35it/s]
3|6411|Loss: 1.1615056991577148:  99%|█████████▉| 6410/6470 [1:13:43<00:44,  1.35it/s]
3|6411|Loss: 1.1615056991577148:  99%|█████████▉| 6411/6470 [1:13:43<00:40,  1.45it/s]
3|6412|Loss: 0.9082534313201904:  99%|█████████▉| 6411/6470 [1:13:43<00:40,  1.45it/s]
3|6412|Loss: 0.9082534313201904:  99%|█████████▉| 6412/6470 [1:13:44<00:39,  1.45it/s]
3|6413|Loss: 0.7457572817802429:  99%|█████████▉| 6412/6470 [1:13:44<00:39,  1.45it/s]
3|6413|Loss: 0.7457572817802429:  99%|█████████▉| 6413/6470 [1:13:44<00:40,  1.42it/s]
3|6414|Loss: 0.9648880362510681:  99%|█████████▉| 6413/6470 [1:13:45<00:40,  1.42it/s]
3|6414|Loss: 0.9648880362510681:  99%|█████████▉| 6414/6470 [1:13:45<00:40,  1.37it/s]
3|6415|Loss: 0.9560490846633911:  99%|█████████▉| 6414/6470 [1:13:46<00:40,  1.37it/s]
3|6415|Loss: 0.9560490846633911:  99%|█████████▉| 6415/6470 [1:13:46<00:41,  1.33it/s]
3|6416|Loss: 0.8651326298713684:  99%|█████████▉| 6415/6470 [1:13:47<00:41,  1.33it/s]
3|6416|Loss: 0.8651326298713684:  99%|█████████▉| 6416/6470 [1:13:47<00:40,  1.34it/s]
3|6417|Loss: 0.9123916625976562:  99%|█████████▉| 6416/6470 [1:13:47<00:40,  1.34it/s]
3|6417|Loss: 0.9123916625976562:  99%|█████████▉| 6417/6470 [1:13:48<00:39,  1.35it/s]
3|6418|Loss: 0.7668991088867188:  99%|█████████▉| 6417/6470 [1:13:48<00:39,  1.35it/s]
3|6418|Loss: 0.7668991088867188:  99%|█████████▉| 6418/6470 [1:13:48<00:36,  1.44it/s]
3|6419|Loss: 0.8626790642738342:  99%|█████████▉| 6418/6470 [1:13:48<00:36,  1.44it/s]
3|6419|Loss: 0.8626790642738342:  99%|█████████▉| 6419/6470 [1:13:49<00:34,  1.49it/s]
3|6420|Loss: 0.8721343278884888:  99%|█████████▉| 6419/6470 [1:13:49<00:34,  1.49it/s]
3|6420|Loss: 0.8721343278884888:  99%|█████████▉| 6420/6470 [1:13:49<00:33,  1.51it/s]
3|6421|Loss: 0.9381030797958374:  99%|█████████▉| 6420/6470 [1:13:50<00:33,  1.51it/s]
3|6421|Loss: 0.9381030797958374:  99%|█████████▉| 6421/6470 [1:13:50<00:32,  1.52it/s]
3|6422|Loss: 0.8419409394264221:  99%|█████████▉| 6421/6470 [1:13:50<00:32,  1.52it/s]
3|6422|Loss: 0.8419409394264221:  99%|█████████▉| 6422/6470 [1:13:51<00:32,  1.47it/s]
3|6423|Loss: 0.7920335531234741:  99%|█████████▉| 6422/6470 [1:13:51<00:32,  1.47it/s]
3|6423|Loss: 0.7920335531234741:  99%|█████████▉| 6423/6470 [1:13:51<00:31,  1.48it/s]
3|6424|Loss: 0.9859920740127563:  99%|█████████▉| 6423/6470 [1:13:52<00:31,  1.48it/s]
3|6424|Loss: 0.9859920740127563:  99%|█████████▉| 6424/6470 [1:13:52<00:30,  1.52it/s]
3|6425|Loss: 0.6978926062583923:  99%|█████████▉| 6424/6470 [1:13:52<00:30,  1.52it/s]
3|6425|Loss: 0.6978926062583923:  99%|█████████▉| 6425/6470 [1:13:53<00:29,  1.55it/s]
3|6426|Loss: 0.9147658944129944:  99%|█████████▉| 6425/6470 [1:13:53<00:29,  1.55it/s]
3|6426|Loss: 0.9147658944129944:  99%|█████████▉| 6426/6470 [1:13:53<00:28,  1.55it/s]
3|6427|Loss: 0.7304492592811584:  99%|█████████▉| 6426/6470 [1:13:54<00:28,  1.55it/s]
3|6427|Loss: 0.7304492592811584:  99%|█████████▉| 6427/6470 [1:13:54<00:29,  1.43it/s]
3|6428|Loss: 0.9244739413261414:  99%|█████████▉| 6427/6470 [1:13:55<00:29,  1.43it/s]
3|6428|Loss: 0.9244739413261414:  99%|█████████▉| 6428/6470 [1:13:55<00:28,  1.47it/s]
3|6429|Loss: 0.8753088116645813:  99%|█████████▉| 6428/6470 [1:13:55<00:28,  1.47it/s]
3|6429|Loss: 0.8753088116645813:  99%|█████████▉| 6429/6470 [1:13:55<00:26,  1.54it/s]
3|6430|Loss: 0.9291716814041138:  99%|█████████▉| 6429/6470 [1:13:56<00:26,  1.54it/s]
3|6430|Loss: 0.9291716814041138:  99%|█████████▉| 6430/6470 [1:13:56<00:27,  1.47it/s]
3|6431|Loss: 0.8375728130340576:  99%|█████████▉| 6430/6470 [1:13:56<00:27,  1.47it/s]
3|6431|Loss: 0.8375728130340576:  99%|█████████▉| 6431/6470 [1:13:57<00:24,  1.58it/s]
3|6432|Loss: 0.7072283029556274:  99%|█████████▉| 6431/6470 [1:13:57<00:24,  1.58it/s]
3|6432|Loss: 0.7072283029556274:  99%|█████████▉| 6432/6470 [1:13:57<00:21,  1.73it/s]
3|6433|Loss: 0.706291139125824:  99%|█████████▉| 6432/6470 [1:13:57<00:21,  1.73it/s] 
3|6433|Loss: 0.706291139125824:  99%|█████████▉| 6433/6470 [1:13:58<00:21,  1.71it/s]
3|6434|Loss: 0.897024929523468:  99%|█████████▉| 6433/6470 [1:13:58<00:21,  1.71it/s]
3|6434|Loss: 0.897024929523468:  99%|█████████▉| 6434/6470 [1:13:58<00:21,  1.67it/s]
3|6435|Loss: 0.8397205471992493:  99%|█████████▉| 6434/6470 [1:13:59<00:21,  1.67it/s]
3|6435|Loss: 0.8397205471992493:  99%|█████████▉| 6435/6470 [1:13:59<00:22,  1.53it/s]
3|6436|Loss: 0.8172968029975891:  99%|█████████▉| 6435/6470 [1:13:59<00:22,  1.53it/s]
3|6436|Loss: 0.8172968029975891:  99%|█████████▉| 6436/6470 [1:14:00<00:20,  1.69it/s]
3|6437|Loss: 0.8266426920890808:  99%|█████████▉| 6436/6470 [1:14:00<00:20,  1.69it/s]
3|6437|Loss: 0.8266426920890808:  99%|█████████▉| 6437/6470 [1:14:00<00:19,  1.69it/s]
3|6438|Loss: 0.7993341088294983:  99%|█████████▉| 6437/6470 [1:14:00<00:19,  1.69it/s]
3|6438|Loss: 0.7993341088294983: 100%|█████████▉| 6438/6470 [1:14:01<00:19,  1.65it/s]
3|6439|Loss: 0.919750988483429: 100%|█████████▉| 6438/6470 [1:14:01<00:19,  1.65it/s] 
3|6439|Loss: 0.919750988483429: 100%|█████████▉| 6439/6470 [1:14:01<00:19,  1.56it/s]
3|6440|Loss: 0.8705984950065613: 100%|█████████▉| 6439/6470 [1:14:02<00:19,  1.56it/s]
3|6440|Loss: 0.8705984950065613: 100%|█████████▉| 6440/6470 [1:14:02<00:19,  1.56it/s]
3|6441|Loss: 0.871526300907135: 100%|█████████▉| 6440/6470 [1:14:03<00:19,  1.56it/s] 
3|6441|Loss: 0.871526300907135: 100%|█████████▉| 6441/6470 [1:14:03<00:19,  1.48it/s]
3|6442|Loss: 0.8317717909812927: 100%|█████████▉| 6441/6470 [1:14:03<00:19,  1.48it/s]
3|6442|Loss: 0.8317717909812927: 100%|█████████▉| 6442/6470 [1:14:04<00:19,  1.43it/s]
3|6443|Loss: 0.9536888003349304: 100%|█████████▉| 6442/6470 [1:14:04<00:19,  1.43it/s]
3|6443|Loss: 0.9536888003349304: 100%|█████████▉| 6443/6470 [1:14:04<00:18,  1.46it/s]
3|6444|Loss: 0.8690505623817444: 100%|█████████▉| 6443/6470 [1:14:05<00:18,  1.46it/s]
3|6444|Loss: 0.8690505623817444: 100%|█████████▉| 6444/6470 [1:14:05<00:18,  1.38it/s]
3|6445|Loss: 0.8957414627075195: 100%|█████████▉| 6444/6470 [1:14:06<00:18,  1.38it/s]
3|6445|Loss: 0.8957414627075195: 100%|█████████▉| 6445/6470 [1:14:06<00:18,  1.32it/s]
3|6446|Loss: 0.8788154125213623: 100%|█████████▉| 6445/6470 [1:14:06<00:18,  1.32it/s]
3|6446|Loss: 0.8788154125213623: 100%|█████████▉| 6446/6470 [1:14:07<00:18,  1.32it/s]
3|6447|Loss: 0.9003119468688965: 100%|█████████▉| 6446/6470 [1:14:07<00:18,  1.32it/s]
3|6447|Loss: 0.9003119468688965: 100%|█████████▉| 6447/6470 [1:14:08<00:18,  1.27it/s]
3|6448|Loss: 0.8558903932571411: 100%|█████████▉| 6447/6470 [1:14:08<00:18,  1.27it/s]
3|6448|Loss: 0.8558903932571411: 100%|█████████▉| 6448/6470 [1:14:08<00:15,  1.38it/s]
3|6449|Loss: 0.8946914672851562: 100%|█████████▉| 6448/6470 [1:14:09<00:15,  1.38it/s]
3|6449|Loss: 0.8946914672851562: 100%|█████████▉| 6449/6470 [1:14:09<00:15,  1.35it/s]
3|6450|Loss: 0.9458813071250916: 100%|█████████▉| 6449/6470 [1:14:09<00:15,  1.35it/s]
3|6450|Loss: 0.9458813071250916: 100%|█████████▉| 6450/6470 [1:14:10<00:14,  1.39it/s]
3|6451|Loss: 0.8173273205757141: 100%|█████████▉| 6450/6470 [1:14:10<00:14,  1.39it/s]
3|6451|Loss: 0.8173273205757141: 100%|█████████▉| 6451/6470 [1:14:10<00:13,  1.44it/s]
3|6452|Loss: 0.8695840239524841: 100%|█████████▉| 6451/6470 [1:14:11<00:13,  1.44it/s]
3|6452|Loss: 0.8695840239524841: 100%|█████████▉| 6452/6470 [1:14:11<00:13,  1.37it/s]
3|6453|Loss: 0.8814790844917297: 100%|█████████▉| 6452/6470 [1:14:12<00:13,  1.37it/s]
3|6453|Loss: 0.8814790844917297: 100%|█████████▉| 6453/6470 [1:14:12<00:13,  1.29it/s]
3|6454|Loss: 0.9022554755210876: 100%|█████████▉| 6453/6470 [1:14:12<00:13,  1.29it/s]
3|6454|Loss: 0.9022554755210876: 100%|█████████▉| 6454/6470 [1:14:13<00:12,  1.33it/s]
3|6455|Loss: 0.8630134463310242: 100%|█████████▉| 6454/6470 [1:14:13<00:12,  1.33it/s]
3|6455|Loss: 0.8630134463310242: 100%|█████████▉| 6455/6470 [1:14:13<00:11,  1.35it/s]
3|6456|Loss: 0.8398890495300293: 100%|█████████▉| 6455/6470 [1:14:14<00:11,  1.35it/s]
3|6456|Loss: 0.8398890495300293: 100%|█████████▉| 6456/6470 [1:14:14<00:10,  1.35it/s]
3|6457|Loss: 1.0580905675888062: 100%|█████████▉| 6456/6470 [1:14:14<00:10,  1.35it/s]
3|6457|Loss: 1.0580905675888062: 100%|█████████▉| 6457/6470 [1:14:15<00:09,  1.44it/s]
3|6458|Loss: 0.7692419290542603: 100%|█████████▉| 6457/6470 [1:14:15<00:09,  1.44it/s]
3|6458|Loss: 0.7692419290542603: 100%|█████████▉| 6458/6470 [1:14:15<00:08,  1.37it/s]
3|6459|Loss: 0.8917858600616455: 100%|█████████▉| 6458/6470 [1:14:16<00:08,  1.37it/s]
3|6459|Loss: 0.8917858600616455: 100%|█████████▉| 6459/6470 [1:14:16<00:07,  1.40it/s]
3|6460|Loss: 0.7720736265182495: 100%|█████████▉| 6459/6470 [1:14:17<00:07,  1.40it/s]
3|6460|Loss: 0.7720736265182495: 100%|█████████▉| 6460/6470 [1:14:17<00:06,  1.46it/s]
3|6461|Loss: 0.882988452911377: 100%|█████████▉| 6460/6470 [1:14:17<00:06,  1.46it/s] 
3|6461|Loss: 0.882988452911377: 100%|█████████▉| 6461/6470 [1:14:17<00:05,  1.50it/s]
3|6462|Loss: 0.8759465217590332: 100%|█████████▉| 6461/6470 [1:14:18<00:05,  1.50it/s]
3|6462|Loss: 0.8759465217590332: 100%|█████████▉| 6462/6470 [1:14:18<00:05,  1.58it/s]
3|6463|Loss: 0.8101745843887329: 100%|█████████▉| 6462/6470 [1:14:18<00:05,  1.58it/s]
3|6463|Loss: 0.8101745843887329: 100%|█████████▉| 6463/6470 [1:14:19<00:04,  1.60it/s]
3|6464|Loss: 0.6568958759307861: 100%|█████████▉| 6463/6470 [1:14:19<00:04,  1.60it/s]
3|6464|Loss: 0.6568958759307861: 100%|█████████▉| 6464/6470 [1:14:19<00:03,  1.55it/s]
3|6465|Loss: 0.7875169515609741: 100%|█████████▉| 6464/6470 [1:14:20<00:03,  1.55it/s]
3|6465|Loss: 0.7875169515609741: 100%|█████████▉| 6465/6470 [1:14:20<00:03,  1.49it/s]
3|6466|Loss: 0.9595587849617004: 100%|█████████▉| 6465/6470 [1:14:20<00:03,  1.49it/s]
3|6466|Loss: 0.9595587849617004: 100%|█████████▉| 6466/6470 [1:14:21<00:02,  1.56it/s]
3|6467|Loss: 0.8380613923072815: 100%|█████████▉| 6466/6470 [1:14:21<00:02,  1.56it/s]
3|6467|Loss: 0.8380613923072815: 100%|█████████▉| 6467/6470 [1:14:21<00:02,  1.46it/s]
3|6468|Loss: 1.0228890180587769: 100%|█████████▉| 6467/6470 [1:14:22<00:02,  1.46it/s]
3|6468|Loss: 1.0228890180587769: 100%|█████████▉| 6468/6470 [1:14:22<00:01,  1.46it/s]
3|6469|Loss: 0.8665008544921875: 100%|█████████▉| 6468/6470 [1:14:22<00:01,  1.46it/s]
3|6469|Loss: 0.8665008544921875: 100%|█████████▉| 6469/6470 [1:14:23<00:00,  1.41it/s]
3|6470|Loss: 0.9865991473197937: 100%|█████████▉| 6469/6470 [1:14:23<00:00,  1.41it/s]
3|6470|Loss: 0.9865991473197937: 100%|██████████| 6470/6470 [1:14:24<00:00,  1.34it/s]
3|6470|Loss: 0.9865991473197937: 100%|██████████| 6470/6470 [1:14:24<00:00,  1.45it/s]
Model checkpoint of size 25705 MB saved to output/alpaca-llama2-baseline/model_2_19410.ckpt

  0%|          | 0/6470 [00:00<?, ?it/s]
4|1|Loss: 0.842976450920105:   0%|          | 0/6470 [00:00<?, ?it/s]
4|1|Loss: 0.842976450920105:   0%|          | 1/6470 [00:00<1:01:23,  1.76it/s]
4|2|Loss: 0.8796519041061401:   0%|          | 1/6470 [00:00<1:01:23,  1.76it/s]
4|2|Loss: 0.8796519041061401:   0%|          | 2/6470 [00:01<1:05:18,  1.65it/s]
4|3|Loss: 0.8639238476753235:   0%|          | 2/6470 [00:01<1:05:18,  1.65it/s]
4|3|Loss: 0.8639238476753235:   0%|          | 3/6470 [00:01<1:13:45,  1.46it/s]
4|4|Loss: 0.8969411849975586:   0%|          | 3/6470 [00:02<1:13:45,  1.46it/s]
4|4|Loss: 0.8969411849975586:   0%|          | 4/6470 [00:02<1:13:12,  1.47it/s]
4|5|Loss: 0.9880434274673462:   0%|          | 4/6470 [00:03<1:13:12,  1.47it/s]
4|5|Loss: 0.9880434274673462:   0%|          | 5/6470 [00:03<1:14:28,  1.45it/s]
4|6|Loss: 0.932623028755188:   0%|          | 5/6470 [00:03<1:14:28,  1.45it/s] 
4|6|Loss: 0.932623028755188:   0%|          | 6/6470 [00:04<1:18:30,  1.37it/s]
4|7|Loss: 0.8634361028671265:   0%|          | 6/6470 [00:04<1:18:30,  1.37it/s]
4|7|Loss: 0.8634361028671265:   0%|          | 7/6470 [00:04<1:20:57,  1.33it/s]
4|8|Loss: 0.8513786792755127:   0%|          | 7/6470 [00:05<1:20:57,  1.33it/s]
4|8|Loss: 0.8513786792755127:   0%|          | 8/6470 [00:05<1:22:59,  1.30it/s]
4|9|Loss: 0.944691002368927:   0%|          | 8/6470 [00:06<1:22:59,  1.30it/s] 
4|9|Loss: 0.944691002368927:   0%|          | 9/6470 [00:06<1:24:02,  1.28it/s]
4|10|Loss: 0.7833819389343262:   0%|          | 9/6470 [00:07<1:24:02,  1.28it/s]
4|10|Loss: 0.7833819389343262:   0%|          | 10/6470 [00:07<1:24:18,  1.28it/s]
4|11|Loss: 0.8767737746238708:   0%|          | 10/6470 [00:07<1:24:18,  1.28it/s]
4|11|Loss: 0.8767737746238708:   0%|          | 11/6470 [00:08<1:22:00,  1.31it/s]
4|12|Loss: 0.9978099465370178:   0%|          | 11/6470 [00:08<1:22:00,  1.31it/s]
4|12|Loss: 0.9978099465370178:   0%|          | 12/6470 [00:08<1:17:28,  1.39it/s]
4|13|Loss: 0.8597204685211182:   0%|          | 12/6470 [00:09<1:17:28,  1.39it/s]
4|13|Loss: 0.8597204685211182:   0%|          | 13/6470 [00:09<1:18:14,  1.38it/s]
4|14|Loss: 0.8446393013000488:   0%|          | 13/6470 [00:09<1:18:14,  1.38it/s]
4|14|Loss: 0.8446393013000488:   0%|          | 14/6470 [00:10<1:16:12,  1.41it/s]
4|15|Loss: 0.8640486598014832:   0%|          | 14/6470 [00:10<1:16:12,  1.41it/s]
4|15|Loss: 0.8640486598014832:   0%|          | 15/6470 [00:10<1:13:54,  1.46it/s]
4|16|Loss: 0.9057064652442932:   0%|          | 15/6470 [00:11<1:13:54,  1.46it/s]
4|16|Loss: 0.9057064652442932:   0%|          | 16/6470 [00:11<1:10:05,  1.53it/s]
4|17|Loss: 0.7939582467079163:   0%|          | 16/6470 [00:11<1:10:05,  1.53it/s]
4|17|Loss: 0.7939582467079163:   0%|          | 17/6470 [00:12<1:14:30,  1.44it/s]
4|18|Loss: 0.8182328939437866:   0%|          | 17/6470 [00:12<1:14:30,  1.44it/s]
4|18|Loss: 0.8182328939437866:   0%|          | 18/6470 [00:12<1:16:35,  1.40it/s]
4|19|Loss: 0.9237533211708069:   0%|          | 18/6470 [00:13<1:16:35,  1.40it/s]
4|19|Loss: 0.9237533211708069:   0%|          | 19/6470 [00:13<1:16:13,  1.41it/s]
4|20|Loss: 0.8190556764602661:   0%|          | 19/6470 [00:13<1:16:13,  1.41it/s]
4|20|Loss: 0.8190556764602661:   0%|          | 20/6470 [00:14<1:10:06,  1.53it/s]
4|21|Loss: 0.8126195669174194:   0%|          | 20/6470 [00:14<1:10:06,  1.53it/s]
4|21|Loss: 0.8126195669174194:   0%|          | 21/6470 [00:14<1:06:07,  1.63it/s]
4|22|Loss: 0.736885666847229:   0%|          | 21/6470 [00:14<1:06:07,  1.63it/s] 
4|22|Loss: 0.736885666847229:   0%|          | 22/6470 [00:15<1:03:18,  1.70it/s]
4|23|Loss: 0.9403893351554871:   0%|          | 22/6470 [00:15<1:03:18,  1.70it/s]
4|23|Loss: 0.9403893351554871:   0%|          | 23/6470 [00:15<1:05:47,  1.63it/s]
4|24|Loss: 0.9261572957038879:   0%|          | 23/6470 [00:16<1:05:47,  1.63it/s]
4|24|Loss: 0.9261572957038879:   0%|          | 24/6470 [00:16<1:04:38,  1.66it/s]
4|25|Loss: 0.8010377883911133:   0%|          | 24/6470 [00:16<1:04:38,  1.66it/s]
4|25|Loss: 0.8010377883911133:   0%|          | 25/6470 [00:16<1:04:46,  1.66it/s]
4|26|Loss: 0.9176467061042786:   0%|          | 25/6470 [00:17<1:04:46,  1.66it/s]
4|26|Loss: 0.9176467061042786:   0%|          | 26/6470 [00:17<1:09:30,  1.55it/s]
4|27|Loss: 0.7905810475349426:   0%|          | 26/6470 [00:18<1:09:30,  1.55it/s]
4|27|Loss: 0.7905810475349426:   0%|          | 27/6470 [00:18<1:08:37,  1.56it/s]
4|28|Loss: 0.9848410487174988:   0%|          | 27/6470 [00:18<1:08:37,  1.56it/s]
4|28|Loss: 0.9848410487174988:   0%|          | 28/6470 [00:19<1:13:32,  1.46it/s]
4|29|Loss: 0.8144559264183044:   0%|          | 28/6470 [00:19<1:13:32,  1.46it/s]
4|29|Loss: 0.8144559264183044:   0%|          | 29/6470 [00:19<1:10:07,  1.53it/s]
4|30|Loss: 0.8775222897529602:   0%|          | 29/6470 [00:20<1:10:07,  1.53it/s]
4|30|Loss: 0.8775222897529602:   0%|          | 30/6470 [00:20<1:12:30,  1.48it/s]
4|31|Loss: 0.840972900390625:   0%|          | 30/6470 [00:20<1:12:30,  1.48it/s] 
4|31|Loss: 0.840972900390625:   0%|          | 31/6470 [00:21<1:14:02,  1.45it/s]
4|32|Loss: 0.9317318797111511:   0%|          | 31/6470 [00:21<1:14:02,  1.45it/s]
4|32|Loss: 0.9317318797111511:   0%|          | 32/6470 [00:21<1:09:11,  1.55it/s]
4|33|Loss: 0.9787842035293579:   0%|          | 32/6470 [00:22<1:09:11,  1.55it/s]
4|33|Loss: 0.9787842035293579:   1%|          | 33/6470 [00:22<1:10:21,  1.52it/s]
4|34|Loss: 0.9480190873146057:   1%|          | 33/6470 [00:22<1:10:21,  1.52it/s]
4|34|Loss: 0.9480190873146057:   1%|          | 34/6470 [00:22<1:05:34,  1.64it/s]
4|35|Loss: 0.7725724577903748:   1%|          | 34/6470 [00:23<1:05:34,  1.64it/s]
4|35|Loss: 0.7725724577903748:   1%|          | 35/6470 [00:23<1:02:09,  1.73it/s]
4|36|Loss: 0.7785009145736694:   1%|          | 35/6470 [00:23<1:02:09,  1.73it/s]
4|36|Loss: 0.7785009145736694:   1%|          | 36/6470 [00:23<1:00:13,  1.78it/s]
4|37|Loss: 0.8464540839195251:   1%|          | 36/6470 [00:24<1:00:13,  1.78it/s]
4|37|Loss: 0.8464540839195251:   1%|          | 37/6470 [00:24<1:04:10,  1.67it/s]
4|38|Loss: 0.7230798006057739:   1%|          | 37/6470 [00:24<1:04:10,  1.67it/s]
4|38|Loss: 0.7230798006057739:   1%|          | 38/6470 [00:25<1:03:42,  1.68it/s]
4|39|Loss: 0.8639962077140808:   1%|          | 38/6470 [00:25<1:03:42,  1.68it/s]
4|39|Loss: 0.8639962077140808:   1%|          | 39/6470 [00:25<1:02:51,  1.71it/s]
4|40|Loss: 0.860787034034729:   1%|          | 39/6470 [00:26<1:02:51,  1.71it/s] 
4|40|Loss: 0.860787034034729:   1%|          | 40/6470 [00:26<1:08:51,  1.56it/s]
4|41|Loss: 0.9738591909408569:   1%|          | 40/6470 [00:26<1:08:51,  1.56it/s]
4|41|Loss: 0.9738591909408569:   1%|          | 41/6470 [00:27<1:12:35,  1.48it/s]
4|42|Loss: 0.9271886348724365:   1%|          | 41/6470 [00:27<1:12:35,  1.48it/s]
4|42|Loss: 0.9271886348724365:   1%|          | 42/6470 [00:28<1:13:33,  1.46it/s]
4|43|Loss: 0.8400776982307434:   1%|          | 42/6470 [00:28<1:13:33,  1.46it/s]
4|43|Loss: 0.8400776982307434:   1%|          | 43/6470 [00:28<1:08:23,  1.57it/s]
4|44|Loss: 0.8351703882217407:   1%|          | 43/6470 [00:28<1:08:23,  1.57it/s]
4|44|Loss: 0.8351703882217407:   1%|          | 44/6470 [00:29<1:05:36,  1.63it/s]
4|45|Loss: 0.8831191062927246:   1%|          | 44/6470 [00:29<1:05:36,  1.63it/s]
4|45|Loss: 0.8831191062927246:   1%|          | 45/6470 [00:29<1:05:29,  1.63it/s]
4|46|Loss: 0.9162454605102539:   1%|          | 45/6470 [00:30<1:05:29,  1.63it/s]
4|46|Loss: 0.9162454605102539:   1%|          | 46/6470 [00:30<1:07:43,  1.58it/s]
4|47|Loss: 0.8059762120246887:   1%|          | 46/6470 [00:30<1:07:43,  1.58it/s]
4|47|Loss: 0.8059762120246887:   1%|          | 47/6470 [00:31<1:09:47,  1.53it/s]
4|48|Loss: 0.8208907246589661:   1%|          | 47/6470 [00:31<1:09:47,  1.53it/s]
4|48|Loss: 0.8208907246589661:   1%|          | 48/6470 [00:31<1:13:52,  1.45it/s]
4|49|Loss: 0.8478868007659912:   1%|          | 48/6470 [00:32<1:13:52,  1.45it/s]
4|49|Loss: 0.8478868007659912:   1%|          | 49/6470 [00:32<1:17:21,  1.38it/s]
4|50|Loss: 0.8815130591392517:   1%|          | 49/6470 [00:33<1:17:21,  1.38it/s]
4|50|Loss: 0.8815130591392517:   1%|          | 50/6470 [00:33<1:15:42,  1.41it/s]
4|51|Loss: 0.9947001934051514:   1%|          | 50/6470 [00:33<1:15:42,  1.41it/s]
4|51|Loss: 0.9947001934051514:   1%|          | 51/6470 [00:34<1:16:20,  1.40it/s]
4|52|Loss: 0.7557477951049805:   1%|          | 51/6470 [00:34<1:16:20,  1.40it/s]
4|52|Loss: 0.7557477951049805:   1%|          | 52/6470 [00:34<1:20:47,  1.32it/s]
4|53|Loss: 0.9021939039230347:   1%|          | 52/6470 [00:35<1:20:47,  1.32it/s]
4|53|Loss: 0.9021939039230347:   1%|          | 53/6470 [00:35<1:23:44,  1.28it/s]
4|54|Loss: 0.8868181109428406:   1%|          | 53/6470 [00:36<1:23:44,  1.28it/s]
4|54|Loss: 0.8868181109428406:   1%|          | 54/6470 [00:36<1:25:55,  1.24it/s]
4|55|Loss: 0.8820644617080688:   1%|          | 54/6470 [00:37<1:25:55,  1.24it/s]
4|55|Loss: 0.8820644617080688:   1%|          | 55/6470 [00:37<1:25:56,  1.24it/s]
4|56|Loss: 0.7993134260177612:   1%|          | 55/6470 [00:37<1:25:56,  1.24it/s]
4|56|Loss: 0.7993134260177612:   1%|          | 56/6470 [00:38<1:21:44,  1.31it/s]
4|57|Loss: 0.9133981466293335:   1%|          | 56/6470 [00:38<1:21:44,  1.31it/s]
4|57|Loss: 0.9133981466293335:   1%|          | 57/6470 [00:38<1:19:16,  1.35it/s]
4|58|Loss: 0.8640066981315613:   1%|          | 57/6470 [00:39<1:19:16,  1.35it/s]
4|58|Loss: 0.8640066981315613:   1%|          | 58/6470 [00:39<1:21:50,  1.31it/s]
4|59|Loss: 0.8900414109230042:   1%|          | 58/6470 [00:40<1:21:50,  1.31it/s]
4|59|Loss: 0.8900414109230042:   1%|          | 59/6470 [00:40<1:22:22,  1.30it/s]
4|60|Loss: 0.984295129776001:   1%|          | 59/6470 [00:40<1:22:22,  1.30it/s] 
4|60|Loss: 0.984295129776001:   1%|          | 60/6470 [00:41<1:23:54,  1.27it/s]
4|61|Loss: 0.9196212291717529:   1%|          | 60/6470 [00:41<1:23:54,  1.27it/s]
4|61|Loss: 0.9196212291717529:   1%|          | 61/6470 [00:41<1:20:05,  1.33it/s]
4|62|Loss: 0.8907889723777771:   1%|          | 61/6470 [00:42<1:20:05,  1.33it/s]
4|62|Loss: 0.8907889723777771:   1%|          | 62/6470 [00:42<1:18:45,  1.36it/s]
4|63|Loss: 0.8255078792572021:   1%|          | 62/6470 [00:42<1:18:45,  1.36it/s]
4|63|Loss: 0.8255078792572021:   1%|          | 63/6470 [00:43<1:18:27,  1.36it/s]
4|64|Loss: 0.8513321876525879:   1%|          | 63/6470 [00:43<1:18:27,  1.36it/s]
4|64|Loss: 0.8513321876525879:   1%|          | 64/6470 [00:44<1:22:06,  1.30it/s]
4|65|Loss: 1.1642303466796875:   1%|          | 64/6470 [00:44<1:22:06,  1.30it/s]
4|65|Loss: 1.1642303466796875:   1%|          | 65/6470 [00:44<1:14:57,  1.42it/s]
4|66|Loss: 0.9421647787094116:   1%|          | 65/6470 [00:45<1:14:57,  1.42it/s]
4|66|Loss: 0.9421647787094116:   1%|          | 66/6470 [00:45<1:13:52,  1.44it/s]
4|67|Loss: 0.8381659984588623:   1%|          | 66/6470 [00:45<1:13:52,  1.44it/s]
4|67|Loss: 0.8381659984588623:   1%|          | 67/6470 [00:46<1:14:31,  1.43it/s]
4|68|Loss: 0.6542391180992126:   1%|          | 67/6470 [00:46<1:14:31,  1.43it/s]
4|68|Loss: 0.6542391180992126:   1%|          | 68/6470 [00:46<1:12:15,  1.48it/s]
4|69|Loss: 0.8852368593215942:   1%|          | 68/6470 [00:47<1:12:15,  1.48it/s]
4|69|Loss: 0.8852368593215942:   1%|          | 69/6470 [00:47<1:13:38,  1.45it/s]
4|70|Loss: 0.7682881355285645:   1%|          | 69/6470 [00:47<1:13:38,  1.45it/s]
4|70|Loss: 0.7682881355285645:   1%|          | 70/6470 [00:48<1:09:57,  1.52it/s]
4|71|Loss: 0.8220092058181763:   1%|          | 70/6470 [00:48<1:09:57,  1.52it/s]
4|71|Loss: 0.8220092058181763:   1%|          | 71/6470 [00:48<1:14:17,  1.44it/s]
4|72|Loss: 0.8359040021896362:   1%|          | 71/6470 [00:49<1:14:17,  1.44it/s]
4|72|Loss: 0.8359040021896362:   1%|          | 72/6470 [00:49<1:14:02,  1.44it/s]
4|73|Loss: 0.7692123651504517:   1%|          | 72/6470 [00:49<1:14:02,  1.44it/s]
4|73|Loss: 0.7692123651504517:   1%|          | 73/6470 [00:50<1:18:25,  1.36it/s]
4|74|Loss: 0.9805923700332642:   1%|          | 73/6470 [00:50<1:18:25,  1.36it/s]
4|74|Loss: 0.9805923700332642:   1%|          | 74/6470 [00:51<1:18:48,  1.35it/s]
4|75|Loss: 0.8324101567268372:   1%|          | 74/6470 [00:51<1:18:48,  1.35it/s]
4|75|Loss: 0.8324101567268372:   1%|          | 75/6470 [00:51<1:14:37,  1.43it/s]
4|76|Loss: 0.8341798782348633:   1%|          | 75/6470 [00:52<1:14:37,  1.43it/s]
4|76|Loss: 0.8341798782348633:   1%|          | 76/6470 [00:52<1:13:50,  1.44it/s]
4|77|Loss: 0.8015097975730896:   1%|          | 76/6470 [00:52<1:13:50,  1.44it/s]
4|77|Loss: 0.8015097975730896:   1%|          | 77/6470 [00:53<1:21:21,  1.31it/s]
4|78|Loss: 0.9733512997627258:   1%|          | 77/6470 [00:53<1:21:21,  1.31it/s]
4|78|Loss: 0.9733512997627258:   1%|          | 78/6470 [00:54<1:20:03,  1.33it/s]
4|79|Loss: 0.7341492176055908:   1%|          | 78/6470 [00:54<1:20:03,  1.33it/s]
4|79|Loss: 0.7341492176055908:   1%|          | 79/6470 [00:54<1:16:31,  1.39it/s]
4|80|Loss: 0.804531455039978:   1%|          | 79/6470 [00:54<1:16:31,  1.39it/s] 
4|80|Loss: 0.804531455039978:   1%|          | 80/6470 [00:55<1:11:41,  1.49it/s]
4|81|Loss: 0.8222463130950928:   1%|          | 80/6470 [00:55<1:11:41,  1.49it/s]
4|81|Loss: 0.8222463130950928:   1%|▏         | 81/6470 [00:55<1:05:25,  1.63it/s]
4|82|Loss: 0.9602496027946472:   1%|▏         | 81/6470 [00:56<1:05:25,  1.63it/s]
4|82|Loss: 0.9602496027946472:   1%|▏         | 82/6470 [00:56<1:07:31,  1.58it/s]
4|83|Loss: 0.7460122108459473:   1%|▏         | 82/6470 [00:56<1:07:31,  1.58it/s]
4|83|Loss: 0.7460122108459473:   1%|▏         | 83/6470 [00:57<1:10:10,  1.52it/s]
4|84|Loss: 1.0244274139404297:   1%|▏         | 83/6470 [00:57<1:10:10,  1.52it/s]
4|84|Loss: 1.0244274139404297:   1%|▏         | 84/6470 [00:57<1:10:02,  1.52it/s]
4|85|Loss: 0.7721815705299377:   1%|▏         | 84/6470 [00:58<1:10:02,  1.52it/s]
4|85|Loss: 0.7721815705299377:   1%|▏         | 85/6470 [00:58<1:08:59,  1.54it/s]
4|86|Loss: 0.9657255411148071:   1%|▏         | 85/6470 [00:58<1:08:59,  1.54it/s]
4|86|Loss: 0.9657255411148071:   1%|▏         | 86/6470 [00:58<1:07:54,  1.57it/s]
4|87|Loss: 0.8018760085105896:   1%|▏         | 86/6470 [00:59<1:07:54,  1.57it/s]
4|87|Loss: 0.8018760085105896:   1%|▏         | 87/6470 [00:59<1:09:31,  1.53it/s]
4|88|Loss: 0.8635978698730469:   1%|▏         | 87/6470 [01:00<1:09:31,  1.53it/s]
4|88|Loss: 0.8635978698730469:   1%|▏         | 88/6470 [01:00<1:11:14,  1.49it/s]
4|89|Loss: 0.7198193669319153:   1%|▏         | 88/6470 [01:00<1:11:14,  1.49it/s]
4|89|Loss: 0.7198193669319153:   1%|▏         | 89/6470 [01:00<1:06:38,  1.60it/s]
4|90|Loss: 0.866510272026062:   1%|▏         | 89/6470 [01:01<1:06:38,  1.60it/s] 
4|90|Loss: 0.866510272026062:   1%|▏         | 90/6470 [01:01<1:08:44,  1.55it/s]
4|91|Loss: 0.8477224707603455:   1%|▏         | 90/6470 [01:01<1:08:44,  1.55it/s]
4|91|Loss: 0.8477224707603455:   1%|▏         | 91/6470 [01:02<1:02:09,  1.71it/s]
4|92|Loss: 0.8908907175064087:   1%|▏         | 91/6470 [01:02<1:02:09,  1.71it/s]
4|92|Loss: 0.8908907175064087:   1%|▏         | 92/6470 [01:02<1:03:49,  1.67it/s]
4|93|Loss: 0.6356934905052185:   1%|▏         | 92/6470 [01:03<1:03:49,  1.67it/s]
4|93|Loss: 0.6356934905052185:   1%|▏         | 93/6470 [01:03<1:07:01,  1.59it/s]
4|94|Loss: 0.8985481858253479:   1%|▏         | 93/6470 [01:03<1:07:01,  1.59it/s]
4|94|Loss: 0.8985481858253479:   1%|▏         | 94/6470 [01:04<1:09:54,  1.52it/s]
4|95|Loss: 0.9491608738899231:   1%|▏         | 94/6470 [01:04<1:09:54,  1.52it/s]
4|95|Loss: 0.9491608738899231:   1%|▏         | 95/6470 [01:04<1:13:36,  1.44it/s]
4|96|Loss: 0.8753862977027893:   1%|▏         | 95/6470 [01:05<1:13:36,  1.44it/s]
4|96|Loss: 0.8753862977027893:   1%|▏         | 96/6470 [01:05<1:14:34,  1.42it/s]
4|97|Loss: 0.9226110577583313:   1%|▏         | 96/6470 [01:05<1:14:34,  1.42it/s]
4|97|Loss: 0.9226110577583313:   1%|▏         | 97/6470 [01:06<1:14:45,  1.42it/s]
4|98|Loss: 0.8131093978881836:   1%|▏         | 97/6470 [01:06<1:14:45,  1.42it/s]
4|98|Loss: 0.8131093978881836:   2%|▏         | 98/6470 [01:06<1:13:04,  1.45it/s]
4|99|Loss: 0.9437870979309082:   2%|▏         | 98/6470 [01:07<1:13:04,  1.45it/s]
4|99|Loss: 0.9437870979309082:   2%|▏         | 99/6470 [01:07<1:04:30,  1.65it/s]
4|100|Loss: 0.7440471649169922:   2%|▏         | 99/6470 [01:07<1:04:30,  1.65it/s]
4|100|Loss: 0.7440471649169922:   2%|▏         | 100/6470 [01:07<1:01:42,  1.72it/s]
4|101|Loss: 0.8073016405105591:   2%|▏         | 100/6470 [01:08<1:01:42,  1.72it/s]
4|101|Loss: 0.8073016405105591:   2%|▏         | 101/6470 [01:08<58:21,  1.82it/s]  
4|102|Loss: 0.8929783701896667:   2%|▏         | 101/6470 [01:08<58:21,  1.82it/s]
4|102|Loss: 0.8929783701896667:   2%|▏         | 102/6470 [01:09<1:02:26,  1.70it/s]
4|103|Loss: 0.7658019661903381:   2%|▏         | 102/6470 [01:09<1:02:26,  1.70it/s]
4|103|Loss: 0.7658019661903381:   2%|▏         | 103/6470 [01:09<1:04:05,  1.66it/s]
4|104|Loss: 0.873196542263031:   2%|▏         | 103/6470 [01:10<1:04:05,  1.66it/s] 
4|104|Loss: 0.873196542263031:   2%|▏         | 104/6470 [01:10<1:10:47,  1.50it/s]
4|105|Loss: 0.8357477188110352:   2%|▏         | 104/6470 [01:10<1:10:47,  1.50it/s]
4|105|Loss: 0.8357477188110352:   2%|▏         | 105/6470 [01:11<1:08:45,  1.54it/s]
4|106|Loss: 0.721905529499054:   2%|▏         | 105/6470 [01:11<1:08:45,  1.54it/s] 
4|106|Loss: 0.721905529499054:   2%|▏         | 106/6470 [01:11<1:07:43,  1.57it/s]
4|107|Loss: 0.7793887853622437:   2%|▏         | 106/6470 [01:12<1:07:43,  1.57it/s]
4|107|Loss: 0.7793887853622437:   2%|▏         | 107/6470 [01:12<1:04:59,  1.63it/s]
4|108|Loss: 0.7575123906135559:   2%|▏         | 107/6470 [01:12<1:04:59,  1.63it/s]
4|108|Loss: 0.7575123906135559:   2%|▏         | 108/6470 [01:13<1:09:45,  1.52it/s]
4|109|Loss: 0.735136866569519:   2%|▏         | 108/6470 [01:13<1:09:45,  1.52it/s] 
4|109|Loss: 0.735136866569519:   2%|▏         | 109/6470 [01:13<1:05:30,  1.62it/s]
4|110|Loss: 0.8879120349884033:   2%|▏         | 109/6470 [01:13<1:05:30,  1.62it/s]
4|110|Loss: 0.8879120349884033:   2%|▏         | 110/6470 [01:14<1:00:04,  1.76it/s]
4|111|Loss: 0.7433105111122131:   2%|▏         | 110/6470 [01:14<1:00:04,  1.76it/s]
4|111|Loss: 0.7433105111122131:   2%|▏         | 111/6470 [01:14<1:03:57,  1.66it/s]
4|112|Loss: 0.9777402877807617:   2%|▏         | 111/6470 [01:15<1:03:57,  1.66it/s]
4|112|Loss: 0.9777402877807617:   2%|▏         | 112/6470 [01:15<1:07:46,  1.56it/s]
4|113|Loss: 0.8710318207740784:   2%|▏         | 112/6470 [01:15<1:07:46,  1.56it/s]
4|113|Loss: 0.8710318207740784:   2%|▏         | 113/6470 [01:16<1:08:10,  1.55it/s]
4|114|Loss: 0.9015688300132751:   2%|▏         | 113/6470 [01:16<1:08:10,  1.55it/s]
4|114|Loss: 0.9015688300132751:   2%|▏         | 114/6470 [01:16<1:10:59,  1.49it/s]
4|115|Loss: 0.7720767855644226:   2%|▏         | 114/6470 [01:17<1:10:59,  1.49it/s]
4|115|Loss: 0.7720767855644226:   2%|▏         | 115/6470 [01:17<1:06:10,  1.60it/s]
4|116|Loss: 0.9097927808761597:   2%|▏         | 115/6470 [01:17<1:06:10,  1.60it/s]
4|116|Loss: 0.9097927808761597:   2%|▏         | 116/6470 [01:17<1:03:59,  1.65it/s]
4|117|Loss: 0.7902886867523193:   2%|▏         | 116/6470 [01:18<1:03:59,  1.65it/s]
4|117|Loss: 0.7902886867523193:   2%|▏         | 117/6470 [01:18<1:03:15,  1.67it/s]
4|118|Loss: 0.957690417766571:   2%|▏         | 117/6470 [01:18<1:03:15,  1.67it/s] 
4|118|Loss: 0.957690417766571:   2%|▏         | 118/6470 [01:19<1:04:12,  1.65it/s]
4|119|Loss: 0.918847382068634:   2%|▏         | 118/6470 [01:19<1:04:12,  1.65it/s]
4|119|Loss: 0.918847382068634:   2%|▏         | 119/6470 [01:19<1:05:04,  1.63it/s]
4|120|Loss: 0.9374220371246338:   2%|▏         | 119/6470 [01:20<1:05:04,  1.63it/s]
4|120|Loss: 0.9374220371246338:   2%|▏         | 120/6470 [01:20<1:11:21,  1.48it/s]
4|121|Loss: 0.9351565837860107:   2%|▏         | 120/6470 [01:21<1:11:21,  1.48it/s]
4|121|Loss: 0.9351565837860107:   2%|▏         | 121/6470 [01:21<1:17:11,  1.37it/s]
4|122|Loss: 0.8723272085189819:   2%|▏         | 121/6470 [01:21<1:17:11,  1.37it/s]
4|122|Loss: 0.8723272085189819:   2%|▏         | 122/6470 [01:22<1:18:29,  1.35it/s]
4|123|Loss: 0.6725482940673828:   2%|▏         | 122/6470 [01:22<1:18:29,  1.35it/s]
4|123|Loss: 0.6725482940673828:   2%|▏         | 123/6470 [01:23<1:24:02,  1.26it/s]
4|124|Loss: 0.9284910559654236:   2%|▏         | 123/6470 [01:23<1:24:02,  1.26it/s]
4|124|Loss: 0.9284910559654236:   2%|▏         | 124/6470 [01:23<1:25:25,  1.24it/s]
4|125|Loss: 0.8379891514778137:   2%|▏         | 124/6470 [01:24<1:25:25,  1.24it/s]
4|125|Loss: 0.8379891514778137:   2%|▏         | 125/6470 [01:24<1:23:35,  1.27it/s]
4|126|Loss: 0.7448510527610779:   2%|▏         | 125/6470 [01:25<1:23:35,  1.27it/s]
4|126|Loss: 0.7448510527610779:   2%|▏         | 126/6470 [01:25<1:20:54,  1.31it/s]
4|127|Loss: 0.9075839519500732:   2%|▏         | 126/6470 [01:25<1:20:54,  1.31it/s]
4|127|Loss: 0.9075839519500732:   2%|▏         | 127/6470 [01:25<1:12:31,  1.46it/s]
4|128|Loss: 0.8799372911453247:   2%|▏         | 127/6470 [01:26<1:12:31,  1.46it/s]
4|128|Loss: 0.8799372911453247:   2%|▏         | 128/6470 [01:26<1:11:30,  1.48it/s]
4|129|Loss: 0.8836232423782349:   2%|▏         | 128/6470 [01:26<1:11:30,  1.48it/s]
4|129|Loss: 0.8836232423782349:   2%|▏         | 129/6470 [01:27<1:13:44,  1.43it/s]
4|130|Loss: 0.7636163234710693:   2%|▏         | 129/6470 [01:27<1:13:44,  1.43it/s]
4|130|Loss: 0.7636163234710693:   2%|▏         | 130/6470 [01:27<1:12:55,  1.45it/s]
4|131|Loss: 0.7738972902297974:   2%|▏         | 130/6470 [01:28<1:12:55,  1.45it/s]
4|131|Loss: 0.7738972902297974:   2%|▏         | 131/6470 [01:28<1:09:41,  1.52it/s]
4|132|Loss: 0.6901195645332336:   2%|▏         | 131/6470 [01:28<1:09:41,  1.52it/s]
4|132|Loss: 0.6901195645332336:   2%|▏         | 132/6470 [01:29<1:11:34,  1.48it/s]
4|133|Loss: 0.7700890302658081:   2%|▏         | 132/6470 [01:29<1:11:34,  1.48it/s]
4|133|Loss: 0.7700890302658081:   2%|▏         | 133/6470 [01:29<1:10:57,  1.49it/s]
4|134|Loss: 0.9209126234054565:   2%|▏         | 133/6470 [01:30<1:10:57,  1.49it/s]
4|134|Loss: 0.9209126234054565:   2%|▏         | 134/6470 [01:30<1:14:14,  1.42it/s]
4|135|Loss: 0.6941614747047424:   2%|▏         | 134/6470 [01:31<1:14:14,  1.42it/s]
4|135|Loss: 0.6941614747047424:   2%|▏         | 135/6470 [01:31<1:20:21,  1.31it/s]
4|136|Loss: 0.9305478930473328:   2%|▏         | 135/6470 [01:32<1:20:21,  1.31it/s]
4|136|Loss: 0.9305478930473328:   2%|▏         | 136/6470 [01:32<1:21:07,  1.30it/s]
4|137|Loss: 0.869378924369812:   2%|▏         | 136/6470 [01:32<1:21:07,  1.30it/s] 
4|137|Loss: 0.869378924369812:   2%|▏         | 137/6470 [01:33<1:19:20,  1.33it/s]
4|138|Loss: 0.8558931946754456:   2%|▏         | 137/6470 [01:33<1:19:20,  1.33it/s]
4|138|Loss: 0.8558931946754456:   2%|▏         | 138/6470 [01:33<1:17:46,  1.36it/s]
4|139|Loss: 0.8210052251815796:   2%|▏         | 138/6470 [01:34<1:17:46,  1.36it/s]
4|139|Loss: 0.8210052251815796:   2%|▏         | 139/6470 [01:34<1:13:33,  1.43it/s]
4|140|Loss: 0.7307015657424927:   2%|▏         | 139/6470 [01:34<1:13:33,  1.43it/s]
4|140|Loss: 0.7307015657424927:   2%|▏         | 140/6470 [01:35<1:14:00,  1.43it/s]
4|141|Loss: 0.8408999443054199:   2%|▏         | 140/6470 [01:35<1:14:00,  1.43it/s]
4|141|Loss: 0.8408999443054199:   2%|▏         | 141/6470 [01:35<1:12:30,  1.45it/s]
4|142|Loss: 0.7044350504875183:   2%|▏         | 141/6470 [01:36<1:12:30,  1.45it/s]
4|142|Loss: 0.7044350504875183:   2%|▏         | 142/6470 [01:36<1:13:21,  1.44it/s]
4|143|Loss: 0.854456901550293:   2%|▏         | 142/6470 [01:36<1:13:21,  1.44it/s] 
4|143|Loss: 0.854456901550293:   2%|▏         | 143/6470 [01:37<1:09:54,  1.51it/s]
4|144|Loss: 0.8875671029090881:   2%|▏         | 143/6470 [01:37<1:09:54,  1.51it/s]
4|144|Loss: 0.8875671029090881:   2%|▏         | 144/6470 [01:37<1:06:55,  1.58it/s]
4|145|Loss: 0.9885174632072449:   2%|▏         | 144/6470 [01:37<1:06:55,  1.58it/s]
4|145|Loss: 0.9885174632072449:   2%|▏         | 145/6470 [01:38<1:05:07,  1.62it/s]
4|146|Loss: 0.9837185740470886:   2%|▏         | 145/6470 [01:38<1:05:07,  1.62it/s]
4|146|Loss: 0.9837185740470886:   2%|▏         | 146/6470 [01:39<1:11:01,  1.48it/s]
4|147|Loss: 0.935299277305603:   2%|▏         | 146/6470 [01:39<1:11:01,  1.48it/s] 
4|147|Loss: 0.935299277305603:   2%|▏         | 147/6470 [01:39<1:09:51,  1.51it/s]
4|148|Loss: 0.953173816204071:   2%|▏         | 147/6470 [01:40<1:09:51,  1.51it/s]
4|148|Loss: 0.953173816204071:   2%|▏         | 148/6470 [01:40<1:11:19,  1.48it/s]
4|149|Loss: 0.9285251498222351:   2%|▏         | 148/6470 [01:40<1:11:19,  1.48it/s]
4|149|Loss: 0.9285251498222351:   2%|▏         | 149/6470 [01:40<1:09:22,  1.52it/s]
4|150|Loss: 0.8897383213043213:   2%|▏         | 149/6470 [01:41<1:09:22,  1.52it/s]
4|150|Loss: 0.8897383213043213:   2%|▏         | 150/6470 [01:41<1:13:56,  1.42it/s]
4|151|Loss: 0.8720823526382446:   2%|▏         | 150/6470 [01:42<1:13:56,  1.42it/s]
4|151|Loss: 0.8720823526382446:   2%|▏         | 151/6470 [01:42<1:18:18,  1.34it/s]
4|152|Loss: 0.9432239532470703:   2%|▏         | 151/6470 [01:43<1:18:18,  1.34it/s]
4|152|Loss: 0.9432239532470703:   2%|▏         | 152/6470 [01:43<1:20:14,  1.31it/s]
4|153|Loss: 0.7979288101196289:   2%|▏         | 152/6470 [01:43<1:20:14,  1.31it/s]
4|153|Loss: 0.7979288101196289:   2%|▏         | 153/6470 [01:44<1:17:38,  1.36it/s]
4|154|Loss: 0.8299213647842407:   2%|▏         | 153/6470 [01:44<1:17:38,  1.36it/s]
4|154|Loss: 0.8299213647842407:   2%|▏         | 154/6470 [01:44<1:20:23,  1.31it/s]
4|155|Loss: 0.6231734752655029:   2%|▏         | 154/6470 [01:45<1:20:23,  1.31it/s]
4|155|Loss: 0.6231734752655029:   2%|▏         | 155/6470 [01:45<1:17:37,  1.36it/s]
4|156|Loss: 0.7143435478210449:   2%|▏         | 155/6470 [01:45<1:17:37,  1.36it/s]
4|156|Loss: 0.7143435478210449:   2%|▏         | 156/6470 [01:46<1:09:37,  1.51it/s]
4|157|Loss: 0.839207649230957:   2%|▏         | 156/6470 [01:46<1:09:37,  1.51it/s] 
4|157|Loss: 0.839207649230957:   2%|▏         | 157/6470 [01:46<1:10:21,  1.50it/s]
4|158|Loss: 0.8956536650657654:   2%|▏         | 157/6470 [01:47<1:10:21,  1.50it/s]
4|158|Loss: 0.8956536650657654:   2%|▏         | 158/6470 [01:47<1:10:45,  1.49it/s]
4|159|Loss: 0.8254982829093933:   2%|▏         | 158/6470 [01:47<1:10:45,  1.49it/s]
4|159|Loss: 0.8254982829093933:   2%|▏         | 159/6470 [01:47<1:05:51,  1.60it/s]
4|160|Loss: 0.8361284136772156:   2%|▏         | 159/6470 [01:48<1:05:51,  1.60it/s]
4|160|Loss: 0.8361284136772156:   2%|▏         | 160/6470 [01:48<1:06:34,  1.58it/s]
4|161|Loss: 0.7519571185112:   2%|▏         | 160/6470 [01:49<1:06:34,  1.58it/s]   
4|161|Loss: 0.7519571185112:   2%|▏         | 161/6470 [01:49<1:07:07,  1.57it/s]
4|162|Loss: 0.7936009168624878:   2%|▏         | 161/6470 [01:49<1:07:07,  1.57it/s]
4|162|Loss: 0.7936009168624878:   3%|▎         | 162/6470 [01:49<1:07:16,  1.56it/s]
4|163|Loss: 1.019658088684082:   3%|▎         | 162/6470 [01:50<1:07:16,  1.56it/s] 
4|163|Loss: 1.019658088684082:   3%|▎         | 163/6470 [01:50<1:12:38,  1.45it/s]
4|164|Loss: 1.1122654676437378:   3%|▎         | 163/6470 [01:51<1:12:38,  1.45it/s]
4|164|Loss: 1.1122654676437378:   3%|▎         | 164/6470 [01:51<1:14:10,  1.42it/s]
4|165|Loss: 0.9091439843177795:   3%|▎         | 164/6470 [01:51<1:14:10,  1.42it/s]
4|165|Loss: 0.9091439843177795:   3%|▎         | 165/6470 [01:52<1:10:07,  1.50it/s]
4|166|Loss: 0.877219557762146:   3%|▎         | 165/6470 [01:52<1:10:07,  1.50it/s] 
4|166|Loss: 0.877219557762146:   3%|▎         | 166/6470 [01:52<1:07:06,  1.57it/s]
4|167|Loss: 0.8205627799034119:   3%|▎         | 166/6470 [01:52<1:07:06,  1.57it/s]
4|167|Loss: 0.8205627799034119:   3%|▎         | 167/6470 [01:53<1:05:29,  1.60it/s]
4|168|Loss: 0.6787866353988647:   3%|▎         | 167/6470 [01:53<1:05:29,  1.60it/s]
4|168|Loss: 0.6787866353988647:   3%|▎         | 168/6470 [01:53<1:03:18,  1.66it/s]
4|169|Loss: 0.8053088784217834:   3%|▎         | 168/6470 [01:54<1:03:18,  1.66it/s]
4|169|Loss: 0.8053088784217834:   3%|▎         | 169/6470 [01:54<1:00:04,  1.75it/s]
4|170|Loss: 0.8458805084228516:   3%|▎         | 169/6470 [01:54<1:00:04,  1.75it/s]
4|170|Loss: 0.8458805084228516:   3%|▎         | 170/6470 [01:54<1:04:56,  1.62it/s]
4|171|Loss: 0.7284415364265442:   3%|▎         | 170/6470 [01:55<1:04:56,  1.62it/s]
4|171|Loss: 0.7284415364265442:   3%|▎         | 171/6470 [01:55<1:08:08,  1.54it/s]
4|172|Loss: 0.952519416809082:   3%|▎         | 171/6470 [01:56<1:08:08,  1.54it/s] 
4|172|Loss: 0.952519416809082:   3%|▎         | 172/6470 [01:56<1:10:31,  1.49it/s]
4|173|Loss: 0.7979698777198792:   3%|▎         | 172/6470 [01:56<1:10:31,  1.49it/s]
4|173|Loss: 0.7979698777198792:   3%|▎         | 173/6470 [01:57<1:12:47,  1.44it/s]
4|174|Loss: 1.000237226486206:   3%|▎         | 173/6470 [01:57<1:12:47,  1.44it/s] 
4|174|Loss: 1.000237226486206:   3%|▎         | 174/6470 [01:57<1:11:26,  1.47it/s]
4|175|Loss: 0.8048010468482971:   3%|▎         | 174/6470 [01:58<1:11:26,  1.47it/s]
4|175|Loss: 0.8048010468482971:   3%|▎         | 175/6470 [01:58<1:09:20,  1.51it/s]
4|176|Loss: 0.9260432720184326:   3%|▎         | 175/6470 [01:58<1:09:20,  1.51it/s]
4|176|Loss: 0.9260432720184326:   3%|▎         | 176/6470 [01:59<1:10:51,  1.48it/s]
4|177|Loss: 0.832891047000885:   3%|▎         | 176/6470 [01:59<1:10:51,  1.48it/s] 
4|177|Loss: 0.832891047000885:   3%|▎         | 177/6470 [01:59<1:09:47,  1.50it/s]
4|178|Loss: 0.7619110941886902:   3%|▎         | 177/6470 [02:00<1:09:47,  1.50it/s]
4|178|Loss: 0.7619110941886902:   3%|▎         | 178/6470 [02:00<1:14:32,  1.41it/s]
4|179|Loss: 0.768792986869812:   3%|▎         | 178/6470 [02:01<1:14:32,  1.41it/s] 
4|179|Loss: 0.768792986869812:   3%|▎         | 179/6470 [02:01<1:13:08,  1.43it/s]
4|180|Loss: 0.9422692656517029:   3%|▎         | 179/6470 [02:01<1:13:08,  1.43it/s]
4|180|Loss: 0.9422692656517029:   3%|▎         | 180/6470 [02:01<1:10:24,  1.49it/s]
4|181|Loss: 1.071350336074829:   3%|▎         | 180/6470 [02:02<1:10:24,  1.49it/s] 
4|181|Loss: 1.071350336074829:   3%|▎         | 181/6470 [02:02<1:06:29,  1.58it/s]
4|182|Loss: 0.7999327778816223:   3%|▎         | 181/6470 [02:02<1:06:29,  1.58it/s]
4|182|Loss: 0.7999327778816223:   3%|▎         | 182/6470 [02:03<1:05:28,  1.60it/s]
4|183|Loss: 0.8578002452850342:   3%|▎         | 182/6470 [02:03<1:05:28,  1.60it/s]
4|183|Loss: 0.8578002452850342:   3%|▎         | 183/6470 [02:03<1:02:01,  1.69it/s]
4|184|Loss: 0.9212861061096191:   3%|▎         | 183/6470 [02:03<1:02:01,  1.69it/s]
4|184|Loss: 0.9212861061096191:   3%|▎         | 184/6470 [02:04<1:02:03,  1.69it/s]
4|185|Loss: 0.8769069314002991:   3%|▎         | 184/6470 [02:04<1:02:03,  1.69it/s]
4|185|Loss: 0.8769069314002991:   3%|▎         | 185/6470 [02:04<1:05:23,  1.60it/s]
4|186|Loss: 0.6880494952201843:   3%|▎         | 185/6470 [02:05<1:05:23,  1.60it/s]
4|186|Loss: 0.6880494952201843:   3%|▎         | 186/6470 [02:05<1:08:40,  1.53it/s]
4|187|Loss: 0.7682406902313232:   3%|▎         | 186/6470 [02:05<1:08:40,  1.53it/s]
4|187|Loss: 0.7682406902313232:   3%|▎         | 187/6470 [02:06<1:08:09,  1.54it/s]
4|188|Loss: 0.8176014423370361:   3%|▎         | 187/6470 [02:06<1:08:09,  1.54it/s]
4|188|Loss: 0.8176014423370361:   3%|▎         | 188/6470 [02:06<1:07:37,  1.55it/s]
4|189|Loss: 0.8676770329475403:   3%|▎         | 188/6470 [02:07<1:07:37,  1.55it/s]
4|189|Loss: 0.8676770329475403:   3%|▎         | 189/6470 [02:07<1:02:50,  1.67it/s]
4|190|Loss: 0.8770904541015625:   3%|▎         | 189/6470 [02:07<1:02:50,  1.67it/s]
4|190|Loss: 0.8770904541015625:   3%|▎         | 190/6470 [02:07<1:00:00,  1.74it/s]
4|191|Loss: 0.8869099617004395:   3%|▎         | 190/6470 [02:08<1:00:00,  1.74it/s]
4|191|Loss: 0.8869099617004395:   3%|▎         | 191/6470 [02:08<1:02:56,  1.66it/s]
4|192|Loss: 0.7541584968566895:   3%|▎         | 191/6470 [02:08<1:02:56,  1.66it/s]
4|192|Loss: 0.7541584968566895:   3%|▎         | 192/6470 [02:09<1:00:48,  1.72it/s]
4|193|Loss: 0.6982062458992004:   3%|▎         | 192/6470 [02:09<1:00:48,  1.72it/s]
4|193|Loss: 0.6982062458992004:   3%|▎         | 193/6470 [02:09<1:01:48,  1.69it/s]
4|194|Loss: 0.7874768376350403:   3%|▎         | 193/6470 [02:10<1:01:48,  1.69it/s]
4|194|Loss: 0.7874768376350403:   3%|▎         | 194/6470 [02:10<1:07:08,  1.56it/s]
4|195|Loss: 0.9185791015625:   3%|▎         | 194/6470 [02:10<1:07:08,  1.56it/s]   
4|195|Loss: 0.9185791015625:   3%|▎         | 195/6470 [02:10<1:01:42,  1.69it/s]
4|196|Loss: 1.0169709920883179:   3%|▎         | 195/6470 [02:11<1:01:42,  1.69it/s]
4|196|Loss: 1.0169709920883179:   3%|▎         | 196/6470 [02:11<57:12,  1.83it/s]  
4|197|Loss: 0.8912311792373657:   3%|▎         | 196/6470 [02:11<57:12,  1.83it/s]
4|197|Loss: 0.8912311792373657:   3%|▎         | 197/6470 [02:12<1:00:27,  1.73it/s]
4|198|Loss: 0.7753319144248962:   3%|▎         | 197/6470 [02:12<1:00:27,  1.73it/s]
4|198|Loss: 0.7753319144248962:   3%|▎         | 198/6470 [02:12<1:06:56,  1.56it/s]
4|199|Loss: 0.9386486411094666:   3%|▎         | 198/6470 [02:13<1:06:56,  1.56it/s]
4|199|Loss: 0.9386486411094666:   3%|▎         | 199/6470 [02:13<1:13:29,  1.42it/s]
4|200|Loss: 0.7842043042182922:   3%|▎         | 199/6470 [02:14<1:13:29,  1.42it/s]
4|200|Loss: 0.7842043042182922:   3%|▎         | 200/6470 [02:14<1:12:33,  1.44it/s]
4|201|Loss: 0.7943671345710754:   3%|▎         | 200/6470 [02:14<1:12:33,  1.44it/s]
4|201|Loss: 0.7943671345710754:   3%|▎         | 201/6470 [02:15<1:17:39,  1.35it/s]
4|202|Loss: 0.778384804725647:   3%|▎         | 201/6470 [02:15<1:17:39,  1.35it/s] 
4|202|Loss: 0.778384804725647:   3%|▎         | 202/6470 [02:15<1:14:40,  1.40it/s]
4|203|Loss: 0.7002699971199036:   3%|▎         | 202/6470 [02:16<1:14:40,  1.40it/s]
4|203|Loss: 0.7002699971199036:   3%|▎         | 203/6470 [02:16<1:14:50,  1.40it/s]
4|204|Loss: 0.736176073551178:   3%|▎         | 203/6470 [02:16<1:14:50,  1.40it/s] 
4|204|Loss: 0.736176073551178:   3%|▎         | 204/6470 [02:17<1:14:47,  1.40it/s]
4|205|Loss: 0.9585609436035156:   3%|▎         | 204/6470 [02:17<1:14:47,  1.40it/s]
4|205|Loss: 0.9585609436035156:   3%|▎         | 205/6470 [02:18<1:15:48,  1.38it/s]
4|206|Loss: 0.7915048003196716:   3%|▎         | 205/6470 [02:18<1:15:48,  1.38it/s]
4|206|Loss: 0.7915048003196716:   3%|▎         | 206/6470 [02:18<1:15:24,  1.38it/s]
4|207|Loss: 0.8646544814109802:   3%|▎         | 206/6470 [02:19<1:15:24,  1.38it/s]
4|207|Loss: 0.8646544814109802:   3%|▎         | 207/6470 [02:19<1:11:50,  1.45it/s]
4|208|Loss: 0.8604469299316406:   3%|▎         | 207/6470 [02:19<1:11:50,  1.45it/s]
4|208|Loss: 0.8604469299316406:   3%|▎         | 208/6470 [02:20<1:11:25,  1.46it/s]
4|209|Loss: 0.8722066283226013:   3%|▎         | 208/6470 [02:20<1:11:25,  1.46it/s]
4|209|Loss: 0.8722066283226013:   3%|▎         | 209/6470 [02:20<1:09:46,  1.50it/s]
4|210|Loss: 0.9972177147865295:   3%|▎         | 209/6470 [02:20<1:09:46,  1.50it/s]
4|210|Loss: 0.9972177147865295:   3%|▎         | 210/6470 [02:21<1:05:59,  1.58it/s]
4|211|Loss: 0.7965084314346313:   3%|▎         | 210/6470 [02:21<1:05:59,  1.58it/s]
4|211|Loss: 0.7965084314346313:   3%|▎         | 211/6470 [02:21<1:05:44,  1.59it/s]
4|212|Loss: 0.8501994609832764:   3%|▎         | 211/6470 [02:22<1:05:44,  1.59it/s]
4|212|Loss: 0.8501994609832764:   3%|▎         | 212/6470 [02:22<1:05:37,  1.59it/s]
4|213|Loss: 0.8365415930747986:   3%|▎         | 212/6470 [02:22<1:05:37,  1.59it/s]
4|213|Loss: 0.8365415930747986:   3%|▎         | 213/6470 [02:23<1:04:10,  1.63it/s]
4|214|Loss: 0.9633393287658691:   3%|▎         | 213/6470 [02:23<1:04:10,  1.63it/s]
4|214|Loss: 0.9633393287658691:   3%|▎         | 214/6470 [02:23<1:06:14,  1.57it/s]
4|215|Loss: 0.7407233119010925:   3%|▎         | 214/6470 [02:24<1:06:14,  1.57it/s]
4|215|Loss: 0.7407233119010925:   3%|▎         | 215/6470 [02:24<1:09:42,  1.50it/s]
4|216|Loss: 0.9659907221794128:   3%|▎         | 215/6470 [02:24<1:09:42,  1.50it/s]
4|216|Loss: 0.9659907221794128:   3%|▎         | 216/6470 [02:25<1:11:05,  1.47it/s]
4|217|Loss: 0.8087554574012756:   3%|▎         | 216/6470 [02:25<1:11:05,  1.47it/s]
4|217|Loss: 0.8087554574012756:   3%|▎         | 217/6470 [02:25<1:13:02,  1.43it/s]
4|218|Loss: 0.8451250791549683:   3%|▎         | 217/6470 [02:26<1:13:02,  1.43it/s]
4|218|Loss: 0.8451250791549683:   3%|▎         | 218/6470 [02:26<1:13:41,  1.41it/s]
4|219|Loss: 0.634231686592102:   3%|▎         | 218/6470 [02:27<1:13:41,  1.41it/s] 
4|219|Loss: 0.634231686592102:   3%|▎         | 219/6470 [02:27<1:14:55,  1.39it/s]
4|220|Loss: 0.9791411757469177:   3%|▎         | 219/6470 [02:27<1:14:55,  1.39it/s]
4|220|Loss: 0.9791411757469177:   3%|▎         | 220/6470 [02:28<1:12:55,  1.43it/s]
4|221|Loss: 0.7912452816963196:   3%|▎         | 220/6470 [02:28<1:12:55,  1.43it/s]
4|221|Loss: 0.7912452816963196:   3%|▎         | 221/6470 [02:28<1:07:59,  1.53it/s]
4|222|Loss: 0.8704697489738464:   3%|▎         | 221/6470 [02:28<1:07:59,  1.53it/s]
4|222|Loss: 0.8704697489738464:   3%|▎         | 222/6470 [02:29<1:08:19,  1.52it/s]
4|223|Loss: 0.7538946866989136:   3%|▎         | 222/6470 [02:29<1:08:19,  1.52it/s]
4|223|Loss: 0.7538946866989136:   3%|▎         | 223/6470 [02:30<1:12:07,  1.44it/s]
4|224|Loss: 0.8163914084434509:   3%|▎         | 223/6470 [02:30<1:12:07,  1.44it/s]
4|224|Loss: 0.8163914084434509:   3%|▎         | 224/6470 [02:30<1:16:38,  1.36it/s]
4|225|Loss: 0.850550651550293:   3%|▎         | 224/6470 [02:31<1:16:38,  1.36it/s] 
4|225|Loss: 0.850550651550293:   3%|▎         | 225/6470 [02:31<1:16:26,  1.36it/s]
4|226|Loss: 0.7272345423698425:   3%|▎         | 225/6470 [02:31<1:16:26,  1.36it/s]
4|226|Loss: 0.7272345423698425:   3%|▎         | 226/6470 [02:32<1:15:58,  1.37it/s]
4|227|Loss: 0.8046213388442993:   3%|▎         | 226/6470 [02:32<1:15:58,  1.37it/s]
4|227|Loss: 0.8046213388442993:   4%|▎         | 227/6470 [02:32<1:13:50,  1.41it/s]
4|228|Loss: 0.8628395795822144:   4%|▎         | 227/6470 [02:33<1:13:50,  1.41it/s]
4|228|Loss: 0.8628395795822144:   4%|▎         | 228/6470 [02:33<1:07:51,  1.53it/s]
4|229|Loss: 0.81227046251297:   4%|▎         | 228/6470 [02:33<1:07:51,  1.53it/s]  
4|229|Loss: 0.81227046251297:   4%|▎         | 229/6470 [02:34<1:07:55,  1.53it/s]
4|230|Loss: 0.897490918636322:   4%|▎         | 229/6470 [02:34<1:07:55,  1.53it/s]
4|230|Loss: 0.897490918636322:   4%|▎         | 230/6470 [02:34<1:10:19,  1.48it/s]
4|231|Loss: 0.8221742510795593:   4%|▎         | 230/6470 [02:35<1:10:19,  1.48it/s]
4|231|Loss: 0.8221742510795593:   4%|▎         | 231/6470 [02:35<1:08:41,  1.51it/s]
4|232|Loss: 0.8471366167068481:   4%|▎         | 231/6470 [02:35<1:08:41,  1.51it/s]
4|232|Loss: 0.8471366167068481:   4%|▎         | 232/6470 [02:36<1:07:09,  1.55it/s]
4|233|Loss: 0.9283358454704285:   4%|▎         | 232/6470 [02:36<1:07:09,  1.55it/s]
4|233|Loss: 0.9283358454704285:   4%|▎         | 233/6470 [02:36<1:02:10,  1.67it/s]
4|234|Loss: 0.8982976078987122:   4%|▎         | 233/6470 [02:36<1:02:10,  1.67it/s]
4|234|Loss: 0.8982976078987122:   4%|▎         | 234/6470 [02:37<1:02:09,  1.67it/s]
4|235|Loss: 0.7865694761276245:   4%|▎         | 234/6470 [02:37<1:02:09,  1.67it/s]
4|235|Loss: 0.7865694761276245:   4%|▎         | 235/6470 [02:37<59:41,  1.74it/s]  
4|236|Loss: 0.9340478777885437:   4%|▎         | 235/6470 [02:38<59:41,  1.74it/s]
4|236|Loss: 0.9340478777885437:   4%|▎         | 236/6470 [02:38<1:02:24,  1.66it/s]
4|237|Loss: 0.7612951993942261:   4%|▎         | 236/6470 [02:38<1:02:24,  1.66it/s]
4|237|Loss: 0.7612951993942261:   4%|▎         | 237/6470 [02:39<1:03:46,  1.63it/s]
4|238|Loss: 0.9428735971450806:   4%|▎         | 237/6470 [02:39<1:03:46,  1.63it/s]
4|238|Loss: 0.9428735971450806:   4%|▎         | 238/6470 [02:39<1:01:38,  1.69it/s]
4|239|Loss: 0.7053114771842957:   4%|▎         | 238/6470 [02:39<1:01:38,  1.69it/s]
4|239|Loss: 0.7053114771842957:   4%|▎         | 239/6470 [02:40<1:07:36,  1.54it/s]
4|240|Loss: 0.6947530508041382:   4%|▎         | 239/6470 [02:40<1:07:36,  1.54it/s]
4|240|Loss: 0.6947530508041382:   4%|▎         | 240/6470 [02:41<1:12:14,  1.44it/s]
4|241|Loss: 0.9965132474899292:   4%|▎         | 240/6470 [02:41<1:12:14,  1.44it/s]
4|241|Loss: 0.9965132474899292:   4%|▎         | 241/6470 [02:41<1:14:18,  1.40it/s]
4|242|Loss: 1.0337623357772827:   4%|▎         | 241/6470 [02:42<1:14:18,  1.40it/s]
4|242|Loss: 1.0337623357772827:   4%|▎         | 242/6470 [02:42<1:12:57,  1.42it/s]
4|243|Loss: 0.8556263446807861:   4%|▎         | 242/6470 [02:42<1:12:57,  1.42it/s]
4|243|Loss: 0.8556263446807861:   4%|▍         | 243/6470 [02:43<1:13:18,  1.42it/s]
4|244|Loss: 1.0740097761154175:   4%|▍         | 243/6470 [02:43<1:13:18,  1.42it/s]
4|244|Loss: 1.0740097761154175:   4%|▍         | 244/6470 [02:44<1:14:03,  1.40it/s]
4|245|Loss: 0.8959453701972961:   4%|▍         | 244/6470 [02:44<1:14:03,  1.40it/s]
4|245|Loss: 0.8959453701972961:   4%|▍         | 245/6470 [02:44<1:15:15,  1.38it/s]
4|246|Loss: 0.679437518119812:   4%|▍         | 245/6470 [02:45<1:15:15,  1.38it/s] 
4|246|Loss: 0.679437518119812:   4%|▍         | 246/6470 [02:45<1:18:51,  1.32it/s]
4|247|Loss: 0.8851613998413086:   4%|▍         | 246/6470 [02:46<1:18:51,  1.32it/s]
4|247|Loss: 0.8851613998413086:   4%|▍         | 247/6470 [02:46<1:16:06,  1.36it/s]
4|248|Loss: 0.7496599555015564:   4%|▍         | 247/6470 [02:46<1:16:06,  1.36it/s]
4|248|Loss: 0.7496599555015564:   4%|▍         | 248/6470 [02:47<1:17:00,  1.35it/s]
4|249|Loss: 0.814005434513092:   4%|▍         | 248/6470 [02:47<1:17:00,  1.35it/s] 
4|249|Loss: 0.814005434513092:   4%|▍         | 249/6470 [02:47<1:17:12,  1.34it/s]
4|250|Loss: 0.8654466271400452:   4%|▍         | 249/6470 [02:48<1:17:12,  1.34it/s]
4|250|Loss: 0.8654466271400452:   4%|▍         | 250/6470 [02:48<1:11:24,  1.45it/s]
4|251|Loss: 0.8620339035987854:   4%|▍         | 250/6470 [02:48<1:11:24,  1.45it/s]
4|251|Loss: 0.8620339035987854:   4%|▍         | 251/6470 [02:48<1:08:45,  1.51it/s]
4|252|Loss: 0.7331733107566833:   4%|▍         | 251/6470 [02:49<1:08:45,  1.51it/s]
4|252|Loss: 0.7331733107566833:   4%|▍         | 252/6470 [02:49<1:05:51,  1.57it/s]
4|253|Loss: 0.7826414704322815:   4%|▍         | 252/6470 [02:49<1:05:51,  1.57it/s]
4|253|Loss: 0.7826414704322815:   4%|▍         | 253/6470 [02:50<1:09:19,  1.49it/s]
4|254|Loss: 0.9534955024719238:   4%|▍         | 253/6470 [02:50<1:09:19,  1.49it/s]
4|254|Loss: 0.9534955024719238:   4%|▍         | 254/6470 [02:51<1:13:58,  1.40it/s]
4|255|Loss: 0.802097737789154:   4%|▍         | 254/6470 [02:51<1:13:58,  1.40it/s] 
4|255|Loss: 0.802097737789154:   4%|▍         | 255/6470 [02:51<1:18:17,  1.32it/s]
4|256|Loss: 0.9850189089775085:   4%|▍         | 255/6470 [02:52<1:18:17,  1.32it/s]
4|256|Loss: 0.9850189089775085:   4%|▍         | 256/6470 [02:52<1:12:55,  1.42it/s]
4|257|Loss: 0.8174751996994019:   4%|▍         | 256/6470 [02:52<1:12:55,  1.42it/s]
4|257|Loss: 0.8174751996994019:   4%|▍         | 257/6470 [02:53<1:11:27,  1.45it/s]
4|258|Loss: 1.056493878364563:   4%|▍         | 257/6470 [02:53<1:11:27,  1.45it/s] 
4|258|Loss: 1.056493878364563:   4%|▍         | 258/6470 [02:53<1:12:29,  1.43it/s]
4|259|Loss: 0.9386934041976929:   4%|▍         | 258/6470 [02:54<1:12:29,  1.43it/s]
4|259|Loss: 0.9386934041976929:   4%|▍         | 259/6470 [02:54<1:11:08,  1.46it/s]
4|260|Loss: 0.9335175156593323:   4%|▍         | 259/6470 [02:54<1:11:08,  1.46it/s]
4|260|Loss: 0.9335175156593323:   4%|▍         | 260/6470 [02:55<1:12:30,  1.43it/s]
4|261|Loss: 0.8992388248443604:   4%|▍         | 260/6470 [02:55<1:12:30,  1.43it/s]
4|261|Loss: 0.8992388248443604:   4%|▍         | 261/6470 [02:55<1:08:52,  1.50it/s]
4|262|Loss: 0.885325014591217:   4%|▍         | 261/6470 [02:56<1:08:52,  1.50it/s] 
4|262|Loss: 0.885325014591217:   4%|▍         | 262/6470 [02:56<1:08:57,  1.50it/s]
4|263|Loss: 0.7827742695808411:   4%|▍         | 262/6470 [02:56<1:08:57,  1.50it/s]
4|263|Loss: 0.7827742695808411:   4%|▍         | 263/6470 [02:57<1:13:11,  1.41it/s]
4|264|Loss: 0.795526921749115:   4%|▍         | 263/6470 [02:57<1:13:11,  1.41it/s] 
4|264|Loss: 0.795526921749115:   4%|▍         | 264/6470 [02:58<1:11:43,  1.44it/s]
4|265|Loss: 0.7967053651809692:   4%|▍         | 264/6470 [02:58<1:11:43,  1.44it/s]
4|265|Loss: 0.7967053651809692:   4%|▍         | 265/6470 [02:58<1:09:11,  1.49it/s]
4|266|Loss: 0.9095574021339417:   4%|▍         | 265/6470 [02:59<1:09:11,  1.49it/s]
4|266|Loss: 0.9095574021339417:   4%|▍         | 266/6470 [02:59<1:10:24,  1.47it/s]
4|267|Loss: 0.8519188165664673:   4%|▍         | 266/6470 [02:59<1:10:24,  1.47it/s]
4|267|Loss: 0.8519188165664673:   4%|▍         | 267/6470 [03:00<1:14:03,  1.40it/s]
4|268|Loss: 0.5747879147529602:   4%|▍         | 267/6470 [03:00<1:14:03,  1.40it/s]
4|268|Loss: 0.5747879147529602:   4%|▍         | 268/6470 [03:01<1:19:37,  1.30it/s]
4|269|Loss: 1.0632296800613403:   4%|▍         | 268/6470 [03:01<1:19:37,  1.30it/s]
4|269|Loss: 1.0632296800613403:   4%|▍         | 269/6470 [03:01<1:17:35,  1.33it/s]
4|270|Loss: 0.6869205832481384:   4%|▍         | 269/6470 [03:02<1:17:35,  1.33it/s]
4|270|Loss: 0.6869205832481384:   4%|▍         | 270/6470 [03:02<1:15:47,  1.36it/s]
4|271|Loss: 0.8632187247276306:   4%|▍         | 270/6470 [03:02<1:15:47,  1.36it/s]
4|271|Loss: 0.8632187247276306:   4%|▍         | 271/6470 [03:03<1:14:47,  1.38it/s]
4|272|Loss: 0.7892105579376221:   4%|▍         | 271/6470 [03:03<1:14:47,  1.38it/s]
4|272|Loss: 0.7892105579376221:   4%|▍         | 272/6470 [03:03<1:17:53,  1.33it/s]
4|273|Loss: 0.8930026292800903:   4%|▍         | 272/6470 [03:04<1:17:53,  1.33it/s]
4|273|Loss: 0.8930026292800903:   4%|▍         | 273/6470 [03:04<1:18:32,  1.31it/s]
4|274|Loss: 0.7720378637313843:   4%|▍         | 273/6470 [03:05<1:18:32,  1.31it/s]
4|274|Loss: 0.7720378637313843:   4%|▍         | 274/6470 [03:05<1:14:17,  1.39it/s]
4|275|Loss: 0.8865452408790588:   4%|▍         | 274/6470 [03:05<1:14:17,  1.39it/s]
4|275|Loss: 0.8865452408790588:   4%|▍         | 275/6470 [03:06<1:16:37,  1.35it/s]
4|276|Loss: 0.8546810746192932:   4%|▍         | 275/6470 [03:06<1:16:37,  1.35it/s]
4|276|Loss: 0.8546810746192932:   4%|▍         | 276/6470 [03:06<1:10:18,  1.47it/s]
4|277|Loss: 0.7886399626731873:   4%|▍         | 276/6470 [03:07<1:10:18,  1.47it/s]
4|277|Loss: 0.7886399626731873:   4%|▍         | 277/6470 [03:07<1:09:20,  1.49it/s]
4|278|Loss: 0.8689574599266052:   4%|▍         | 277/6470 [03:07<1:09:20,  1.49it/s]
4|278|Loss: 0.8689574599266052:   4%|▍         | 278/6470 [03:07<1:08:23,  1.51it/s]
4|279|Loss: 0.9158604741096497:   4%|▍         | 278/6470 [03:08<1:08:23,  1.51it/s]
4|279|Loss: 0.9158604741096497:   4%|▍         | 279/6470 [03:08<1:07:01,  1.54it/s]
4|280|Loss: 0.8517120480537415:   4%|▍         | 279/6470 [03:08<1:07:01,  1.54it/s]
4|280|Loss: 0.8517120480537415:   4%|▍         | 280/6470 [03:09<1:04:28,  1.60it/s]
4|281|Loss: 0.7891515493392944:   4%|▍         | 280/6470 [03:09<1:04:28,  1.60it/s]
4|281|Loss: 0.7891515493392944:   4%|▍         | 281/6470 [03:09<1:02:54,  1.64it/s]
4|282|Loss: 0.9141330122947693:   4%|▍         | 281/6470 [03:10<1:02:54,  1.64it/s]
4|282|Loss: 0.9141330122947693:   4%|▍         | 282/6470 [03:10<1:03:35,  1.62it/s]
4|283|Loss: 0.9042336940765381:   4%|▍         | 282/6470 [03:10<1:03:35,  1.62it/s]
4|283|Loss: 0.9042336940765381:   4%|▍         | 283/6470 [03:11<1:04:35,  1.60it/s]
4|284|Loss: 0.7731878161430359:   4%|▍         | 283/6470 [03:11<1:04:35,  1.60it/s]
4|284|Loss: 0.7731878161430359:   4%|▍         | 284/6470 [03:11<1:05:00,  1.59it/s]
4|285|Loss: 1.005062222480774:   4%|▍         | 284/6470 [03:12<1:05:00,  1.59it/s] 
4|285|Loss: 1.005062222480774:   4%|▍         | 285/6470 [03:12<1:03:40,  1.62it/s]
4|286|Loss: 0.8780251145362854:   4%|▍         | 285/6470 [03:12<1:03:40,  1.62it/s]
4|286|Loss: 0.8780251145362854:   4%|▍         | 286/6470 [03:12<1:05:47,  1.57it/s]
4|287|Loss: 0.650295078754425:   4%|▍         | 286/6470 [03:13<1:05:47,  1.57it/s] 
4|287|Loss: 0.650295078754425:   4%|▍         | 287/6470 [03:13<1:16:28,  1.35it/s]
4|288|Loss: 0.9395031929016113:   4%|▍         | 287/6470 [03:14<1:16:28,  1.35it/s]
4|288|Loss: 0.9395031929016113:   4%|▍         | 288/6470 [03:14<1:15:48,  1.36it/s]
4|289|Loss: 0.9363237023353577:   4%|▍         | 288/6470 [03:15<1:15:48,  1.36it/s]
4|289|Loss: 0.9363237023353577:   4%|▍         | 289/6470 [03:15<1:12:07,  1.43it/s]
4|290|Loss: 1.0070732831954956:   4%|▍         | 289/6470 [03:15<1:12:07,  1.43it/s]
4|290|Loss: 1.0070732831954956:   4%|▍         | 290/6470 [03:15<1:10:12,  1.47it/s]
4|291|Loss: 0.8446189761161804:   4%|▍         | 290/6470 [03:16<1:10:12,  1.47it/s]
4|291|Loss: 0.8446189761161804:   4%|▍         | 291/6470 [03:16<1:07:23,  1.53it/s]
4|292|Loss: 1.022555947303772:   4%|▍         | 291/6470 [03:16<1:07:23,  1.53it/s] 
4|292|Loss: 1.022555947303772:   5%|▍         | 292/6470 [03:17<1:12:20,  1.42it/s]
4|293|Loss: 0.8769460916519165:   5%|▍         | 292/6470 [03:17<1:12:20,  1.42it/s]
4|293|Loss: 0.8769460916519165:   5%|▍         | 293/6470 [03:18<1:14:55,  1.37it/s]
4|294|Loss: 0.9452952146530151:   5%|▍         | 293/6470 [03:18<1:14:55,  1.37it/s]
4|294|Loss: 0.9452952146530151:   5%|▍         | 294/6470 [03:18<1:15:34,  1.36it/s]
4|295|Loss: 0.8135729432106018:   5%|▍         | 294/6470 [03:19<1:15:34,  1.36it/s]
4|295|Loss: 0.8135729432106018:   5%|▍         | 295/6470 [03:19<1:13:00,  1.41it/s]
4|296|Loss: 0.7764501571655273:   5%|▍         | 295/6470 [03:19<1:13:00,  1.41it/s]
4|296|Loss: 0.7764501571655273:   5%|▍         | 296/6470 [03:19<1:06:08,  1.56it/s]
4|297|Loss: 0.8537107110023499:   5%|▍         | 296/6470 [03:20<1:06:08,  1.56it/s]
4|297|Loss: 0.8537107110023499:   5%|▍         | 297/6470 [03:20<1:10:29,  1.46it/s]
4|298|Loss: 0.9248733520507812:   5%|▍         | 297/6470 [03:21<1:10:29,  1.46it/s]
4|298|Loss: 0.9248733520507812:   5%|▍         | 298/6470 [03:21<1:09:02,  1.49it/s]
4|299|Loss: 0.8931965827941895:   5%|▍         | 298/6470 [03:21<1:09:02,  1.49it/s]
4|299|Loss: 0.8931965827941895:   5%|▍         | 299/6470 [03:22<1:11:50,  1.43it/s]
4|300|Loss: 0.8888973593711853:   5%|▍         | 299/6470 [03:22<1:11:50,  1.43it/s]
4|300|Loss: 0.8888973593711853:   5%|▍         | 300/6470 [03:22<1:07:56,  1.51it/s]
4|301|Loss: 0.9966812133789062:   5%|▍         | 300/6470 [03:23<1:07:56,  1.51it/s]
4|301|Loss: 0.9966812133789062:   5%|▍         | 301/6470 [03:23<1:04:59,  1.58it/s]
4|302|Loss: 0.781220018863678:   5%|▍         | 301/6470 [03:23<1:04:59,  1.58it/s] 
4|302|Loss: 0.781220018863678:   5%|▍         | 302/6470 [03:23<1:01:41,  1.67it/s]
4|303|Loss: 0.8725008368492126:   5%|▍         | 302/6470 [03:24<1:01:41,  1.67it/s]
4|303|Loss: 0.8725008368492126:   5%|▍         | 303/6470 [03:24<1:00:33,  1.70it/s]
4|304|Loss: 0.9035179018974304:   5%|▍         | 303/6470 [03:24<1:00:33,  1.70it/s]
4|304|Loss: 0.9035179018974304:   5%|▍         | 304/6470 [03:25<1:04:13,  1.60it/s]
4|305|Loss: 0.8132511377334595:   5%|▍         | 304/6470 [03:25<1:04:13,  1.60it/s]
4|305|Loss: 0.8132511377334595:   5%|▍         | 305/6470 [03:25<1:06:55,  1.54it/s]
4|306|Loss: 0.897075891494751:   5%|▍         | 305/6470 [03:26<1:06:55,  1.54it/s] 
4|306|Loss: 0.897075891494751:   5%|▍         | 306/6470 [03:26<1:06:25,  1.55it/s]
4|307|Loss: 0.7254000902175903:   5%|▍         | 306/6470 [03:26<1:06:25,  1.55it/s]
4|307|Loss: 0.7254000902175903:   5%|▍         | 307/6470 [03:27<1:05:26,  1.57it/s]
4|308|Loss: 0.7672451138496399:   5%|▍         | 307/6470 [03:27<1:05:26,  1.57it/s]
4|308|Loss: 0.7672451138496399:   5%|▍         | 308/6470 [03:27<1:01:30,  1.67it/s]
4|309|Loss: 0.9754061102867126:   5%|▍         | 308/6470 [03:27<1:01:30,  1.67it/s]
4|309|Loss: 0.9754061102867126:   5%|▍         | 309/6470 [03:28<1:00:12,  1.71it/s]
4|310|Loss: 0.9117968082427979:   5%|▍         | 309/6470 [03:28<1:00:12,  1.71it/s]
4|310|Loss: 0.9117968082427979:   5%|▍         | 310/6470 [03:28<1:00:56,  1.68it/s]
4|311|Loss: 0.6857099533081055:   5%|▍         | 310/6470 [03:29<1:00:56,  1.68it/s]
4|311|Loss: 0.6857099533081055:   5%|▍         | 311/6470 [03:29<1:03:54,  1.61it/s]
4|312|Loss: 0.887910783290863:   5%|▍         | 311/6470 [03:29<1:03:54,  1.61it/s] 
4|312|Loss: 0.887910783290863:   5%|▍         | 312/6470 [03:30<1:06:19,  1.55it/s]
4|313|Loss: 0.9055298566818237:   5%|▍         | 312/6470 [03:30<1:06:19,  1.55it/s]
4|313|Loss: 0.9055298566818237:   5%|▍         | 313/6470 [03:30<1:06:36,  1.54it/s]
4|314|Loss: 0.9002060294151306:   5%|▍         | 313/6470 [03:31<1:06:36,  1.54it/s]
4|314|Loss: 0.9002060294151306:   5%|▍         | 314/6470 [03:31<1:08:49,  1.49it/s]
4|315|Loss: 0.8139662742614746:   5%|▍         | 314/6470 [03:31<1:08:49,  1.49it/s]
4|315|Loss: 0.8139662742614746:   5%|▍         | 315/6470 [03:32<1:07:59,  1.51it/s]
4|316|Loss: 1.0240052938461304:   5%|▍         | 315/6470 [03:32<1:07:59,  1.51it/s]
4|316|Loss: 1.0240052938461304:   5%|▍         | 316/6470 [03:32<1:08:47,  1.49it/s]
4|317|Loss: 0.8922907114028931:   5%|▍         | 316/6470 [03:33<1:08:47,  1.49it/s]
4|317|Loss: 0.8922907114028931:   5%|▍         | 317/6470 [03:33<1:13:00,  1.40it/s]
4|318|Loss: 0.9053603410720825:   5%|▍         | 317/6470 [03:34<1:13:00,  1.40it/s]
4|318|Loss: 0.9053603410720825:   5%|▍         | 318/6470 [03:34<1:12:14,  1.42it/s]
4|319|Loss: 0.9067437052726746:   5%|▍         | 318/6470 [03:34<1:12:14,  1.42it/s]
4|319|Loss: 0.9067437052726746:   5%|▍         | 319/6470 [03:35<1:11:47,  1.43it/s]
4|320|Loss: 0.9185081124305725:   5%|▍         | 319/6470 [03:35<1:11:47,  1.43it/s]
4|320|Loss: 0.9185081124305725:   5%|▍         | 320/6470 [03:35<1:09:52,  1.47it/s]
4|321|Loss: 0.8914803266525269:   5%|▍         | 320/6470 [03:36<1:09:52,  1.47it/s]
4|321|Loss: 0.8914803266525269:   5%|▍         | 321/6470 [03:36<1:08:33,  1.49it/s]
4|322|Loss: 0.8776376843452454:   5%|▍         | 321/6470 [03:36<1:08:33,  1.49it/s]
4|322|Loss: 0.8776376843452454:   5%|▍         | 322/6470 [03:36<1:07:04,  1.53it/s]
4|323|Loss: 0.8681143522262573:   5%|▍         | 322/6470 [03:37<1:07:04,  1.53it/s]
4|323|Loss: 0.8681143522262573:   5%|▍         | 323/6470 [03:37<1:06:36,  1.54it/s]
4|324|Loss: 0.8987706899642944:   5%|▍         | 323/6470 [03:37<1:06:36,  1.54it/s]
4|324|Loss: 0.8987706899642944:   5%|▌         | 324/6470 [03:38<1:03:13,  1.62it/s]
4|325|Loss: 0.7604486346244812:   5%|▌         | 324/6470 [03:38<1:03:13,  1.62it/s]
4|325|Loss: 0.7604486346244812:   5%|▌         | 325/6470 [03:38<1:05:06,  1.57it/s]
4|326|Loss: 0.8849904537200928:   5%|▌         | 325/6470 [03:39<1:05:06,  1.57it/s]
4|326|Loss: 0.8849904537200928:   5%|▌         | 326/6470 [03:39<1:08:11,  1.50it/s]
4|327|Loss: 0.7565373778343201:   5%|▌         | 326/6470 [03:39<1:08:11,  1.50it/s]
4|327|Loss: 0.7565373778343201:   5%|▌         | 327/6470 [03:40<1:08:05,  1.50it/s]
4|328|Loss: 0.7812575697898865:   5%|▌         | 327/6470 [03:40<1:08:05,  1.50it/s]
4|328|Loss: 0.7812575697898865:   5%|▌         | 328/6470 [03:40<1:06:04,  1.55it/s]
4|329|Loss: 0.9450837969779968:   5%|▌         | 328/6470 [03:41<1:06:04,  1.55it/s]
4|329|Loss: 0.9450837969779968:   5%|▌         | 329/6470 [03:41<1:05:37,  1.56it/s]
4|330|Loss: 0.8506705164909363:   5%|▌         | 329/6470 [03:41<1:05:37,  1.56it/s]
4|330|Loss: 0.8506705164909363:   5%|▌         | 330/6470 [03:42<1:07:24,  1.52it/s]
4|331|Loss: 0.8347153663635254:   5%|▌         | 330/6470 [03:42<1:07:24,  1.52it/s]
4|331|Loss: 0.8347153663635254:   5%|▌         | 331/6470 [03:42<1:04:59,  1.57it/s]
4|332|Loss: 0.7657395601272583:   5%|▌         | 331/6470 [03:43<1:04:59,  1.57it/s]
4|332|Loss: 0.7657395601272583:   5%|▌         | 332/6470 [03:43<1:08:19,  1.50it/s]
4|333|Loss: 0.7553629875183105:   5%|▌         | 332/6470 [03:43<1:08:19,  1.50it/s]
4|333|Loss: 0.7553629875183105:   5%|▌         | 333/6470 [03:44<1:10:47,  1.45it/s]
4|334|Loss: 0.7993601560592651:   5%|▌         | 333/6470 [03:44<1:10:47,  1.45it/s]
4|334|Loss: 0.7993601560592651:   5%|▌         | 334/6470 [03:44<1:07:27,  1.52it/s]
4|335|Loss: 0.798594057559967:   5%|▌         | 334/6470 [03:45<1:07:27,  1.52it/s] 
4|335|Loss: 0.798594057559967:   5%|▌         | 335/6470 [03:45<1:04:51,  1.58it/s]
4|336|Loss: 1.0473212003707886:   5%|▌         | 335/6470 [03:45<1:04:51,  1.58it/s]
4|336|Loss: 1.0473212003707886:   5%|▌         | 336/6470 [03:45<1:04:44,  1.58it/s]
4|337|Loss: 0.7763532996177673:   5%|▌         | 336/6470 [03:46<1:04:44,  1.58it/s]
4|337|Loss: 0.7763532996177673:   5%|▌         | 337/6470 [03:46<1:10:23,  1.45it/s]
4|338|Loss: 0.7424699664115906:   5%|▌         | 337/6470 [03:47<1:10:23,  1.45it/s]
4|338|Loss: 0.7424699664115906:   5%|▌         | 338/6470 [03:47<1:15:51,  1.35it/s]
4|339|Loss: 1.2917243242263794:   5%|▌         | 338/6470 [03:48<1:15:51,  1.35it/s]
4|339|Loss: 1.2917243242263794:   5%|▌         | 339/6470 [03:48<1:23:39,  1.22it/s]
4|340|Loss: 0.9317504167556763:   5%|▌         | 339/6470 [03:49<1:23:39,  1.22it/s]
4|340|Loss: 0.9317504167556763:   5%|▌         | 340/6470 [03:49<1:22:22,  1.24it/s]
4|341|Loss: 0.9495648741722107:   5%|▌         | 340/6470 [03:49<1:22:22,  1.24it/s]
4|341|Loss: 0.9495648741722107:   5%|▌         | 341/6470 [03:50<1:23:19,  1.23it/s]
4|342|Loss: 0.8831507563591003:   5%|▌         | 341/6470 [03:50<1:23:19,  1.23it/s]
4|342|Loss: 0.8831507563591003:   5%|▌         | 342/6470 [03:51<1:21:04,  1.26it/s]
4|343|Loss: 0.927944004535675:   5%|▌         | 342/6470 [03:51<1:21:04,  1.26it/s] 
4|343|Loss: 0.927944004535675:   5%|▌         | 343/6470 [03:51<1:19:46,  1.28it/s]
4|344|Loss: 0.938123881816864:   5%|▌         | 343/6470 [03:52<1:19:46,  1.28it/s]
4|344|Loss: 0.938123881816864:   5%|▌         | 344/6470 [03:52<1:16:10,  1.34it/s]
4|345|Loss: 0.8432558178901672:   5%|▌         | 344/6470 [03:52<1:16:10,  1.34it/s]
4|345|Loss: 0.8432558178901672:   5%|▌         | 345/6470 [03:53<1:11:17,  1.43it/s]
4|346|Loss: 0.8685911893844604:   5%|▌         | 345/6470 [03:53<1:11:17,  1.43it/s]
4|346|Loss: 0.8685911893844604:   5%|▌         | 346/6470 [03:53<1:08:46,  1.48it/s]
4|347|Loss: 0.8480214476585388:   5%|▌         | 346/6470 [03:54<1:08:46,  1.48it/s]
4|347|Loss: 0.8480214476585388:   5%|▌         | 347/6470 [03:54<1:09:59,  1.46it/s]
4|348|Loss: 0.9047216176986694:   5%|▌         | 347/6470 [03:54<1:09:59,  1.46it/s]
4|348|Loss: 0.9047216176986694:   5%|▌         | 348/6470 [03:55<1:09:22,  1.47it/s]
4|349|Loss: 0.8983693718910217:   5%|▌         | 348/6470 [03:55<1:09:22,  1.47it/s]
4|349|Loss: 0.8983693718910217:   5%|▌         | 349/6470 [03:55<1:13:12,  1.39it/s]
4|350|Loss: 0.8752785921096802:   5%|▌         | 349/6470 [03:56<1:13:12,  1.39it/s]
4|350|Loss: 0.8752785921096802:   5%|▌         | 350/6470 [03:56<1:14:37,  1.37it/s]
4|351|Loss: 0.7556217908859253:   5%|▌         | 350/6470 [03:57<1:14:37,  1.37it/s]
4|351|Loss: 0.7556217908859253:   5%|▌         | 351/6470 [03:57<1:16:04,  1.34it/s]
4|352|Loss: 0.8860787153244019:   5%|▌         | 351/6470 [03:57<1:16:04,  1.34it/s]
4|352|Loss: 0.8860787153244019:   5%|▌         | 352/6470 [03:58<1:15:23,  1.35it/s]
4|353|Loss: 0.8641173243522644:   5%|▌         | 352/6470 [03:58<1:15:23,  1.35it/s]
4|353|Loss: 0.8641173243522644:   5%|▌         | 353/6470 [03:58<1:15:28,  1.35it/s]
4|354|Loss: 0.8129788637161255:   5%|▌         | 353/6470 [03:59<1:15:28,  1.35it/s]
4|354|Loss: 0.8129788637161255:   5%|▌         | 354/6470 [03:59<1:14:53,  1.36it/s]
4|355|Loss: 0.8613389134407043:   5%|▌         | 354/6470 [04:00<1:14:53,  1.36it/s]
4|355|Loss: 0.8613389134407043:   5%|▌         | 355/6470 [04:00<1:17:05,  1.32it/s]
4|356|Loss: 0.7255241274833679:   5%|▌         | 355/6470 [04:00<1:17:05,  1.32it/s]
4|356|Loss: 0.7255241274833679:   6%|▌         | 356/6470 [04:00<1:11:29,  1.43it/s]
4|357|Loss: 0.8037804961204529:   6%|▌         | 356/6470 [04:01<1:11:29,  1.43it/s]
4|357|Loss: 0.8037804961204529:   6%|▌         | 357/6470 [04:01<1:08:10,  1.49it/s]
4|358|Loss: 0.9044307470321655:   6%|▌         | 357/6470 [04:01<1:08:10,  1.49it/s]
4|358|Loss: 0.9044307470321655:   6%|▌         | 358/6470 [04:02<1:08:36,  1.48it/s]
4|359|Loss: 0.8060927987098694:   6%|▌         | 358/6470 [04:02<1:08:36,  1.48it/s]
4|359|Loss: 0.8060927987098694:   6%|▌         | 359/6470 [04:03<1:14:30,  1.37it/s]
4|360|Loss: 0.8183608651161194:   6%|▌         | 359/6470 [04:03<1:14:30,  1.37it/s]
4|360|Loss: 0.8183608651161194:   6%|▌         | 360/6470 [04:03<1:13:26,  1.39it/s]
4|361|Loss: 0.9096218347549438:   6%|▌         | 360/6470 [04:04<1:13:26,  1.39it/s]
4|361|Loss: 0.9096218347549438:   6%|▌         | 361/6470 [04:04<1:13:05,  1.39it/s]
4|362|Loss: 0.7696358561515808:   6%|▌         | 361/6470 [04:04<1:13:05,  1.39it/s]
4|362|Loss: 0.7696358561515808:   6%|▌         | 362/6470 [04:05<1:12:36,  1.40it/s]
4|363|Loss: 0.9202736020088196:   6%|▌         | 362/6470 [04:05<1:12:36,  1.40it/s]
4|363|Loss: 0.9202736020088196:   6%|▌         | 363/6470 [04:05<1:08:58,  1.48it/s]
4|364|Loss: 0.7980878353118896:   6%|▌         | 363/6470 [04:06<1:08:58,  1.48it/s]
4|364|Loss: 0.7980878353118896:   6%|▌         | 364/6470 [04:06<1:06:18,  1.53it/s]
4|365|Loss: 0.9175423979759216:   6%|▌         | 364/6470 [04:06<1:06:18,  1.53it/s]
4|365|Loss: 0.9175423979759216:   6%|▌         | 365/6470 [04:07<1:08:21,  1.49it/s]
4|366|Loss: 0.9693196415901184:   6%|▌         | 365/6470 [04:07<1:08:21,  1.49it/s]
4|366|Loss: 0.9693196415901184:   6%|▌         | 366/6470 [04:07<1:06:49,  1.52it/s]
4|367|Loss: 0.9453479647636414:   6%|▌         | 366/6470 [04:08<1:06:49,  1.52it/s]
4|367|Loss: 0.9453479647636414:   6%|▌         | 367/6470 [04:08<1:07:29,  1.51it/s]
4|368|Loss: 0.8617566823959351:   6%|▌         | 367/6470 [04:08<1:07:29,  1.51it/s]
4|368|Loss: 0.8617566823959351:   6%|▌         | 368/6470 [04:09<1:09:40,  1.46it/s]
4|369|Loss: 0.8721685409545898:   6%|▌         | 368/6470 [04:09<1:09:40,  1.46it/s]
4|369|Loss: 0.8721685409545898:   6%|▌         | 369/6470 [04:09<1:09:04,  1.47it/s]
4|370|Loss: 0.9386441707611084:   6%|▌         | 369/6470 [04:10<1:09:04,  1.47it/s]
4|370|Loss: 0.9386441707611084:   6%|▌         | 370/6470 [04:10<1:10:52,  1.43it/s]
4|371|Loss: 0.5942056775093079:   6%|▌         | 370/6470 [04:11<1:10:52,  1.43it/s]
4|371|Loss: 0.5942056775093079:   6%|▌         | 371/6470 [04:11<1:15:33,  1.35it/s]
4|372|Loss: 0.8546057343482971:   6%|▌         | 371/6470 [04:11<1:15:33,  1.35it/s]
4|372|Loss: 0.8546057343482971:   6%|▌         | 372/6470 [04:12<1:12:55,  1.39it/s]
4|373|Loss: 0.6858841180801392:   6%|▌         | 372/6470 [04:12<1:12:55,  1.39it/s]
4|373|Loss: 0.6858841180801392:   6%|▌         | 373/6470 [04:12<1:11:52,  1.41it/s]
4|374|Loss: 0.9124189019203186:   6%|▌         | 373/6470 [04:13<1:11:52,  1.41it/s]
4|374|Loss: 0.9124189019203186:   6%|▌         | 374/6470 [04:13<1:12:24,  1.40it/s]
4|375|Loss: 0.8015941977500916:   6%|▌         | 374/6470 [04:13<1:12:24,  1.40it/s]
4|375|Loss: 0.8015941977500916:   6%|▌         | 375/6470 [04:14<1:15:38,  1.34it/s]
4|376|Loss: 0.8779891729354858:   6%|▌         | 375/6470 [04:14<1:15:38,  1.34it/s]
4|376|Loss: 0.8779891729354858:   6%|▌         | 376/6470 [04:14<1:11:29,  1.42it/s]
4|377|Loss: 0.829638659954071:   6%|▌         | 376/6470 [04:15<1:11:29,  1.42it/s] 
4|377|Loss: 0.829638659954071:   6%|▌         | 377/6470 [04:15<1:14:54,  1.36it/s]
4|378|Loss: 0.7979871034622192:   6%|▌         | 377/6470 [04:16<1:14:54,  1.36it/s]
4|378|Loss: 0.7979871034622192:   6%|▌         | 378/6470 [04:16<1:13:08,  1.39it/s]
4|379|Loss: 0.8287539482116699:   6%|▌         | 378/6470 [04:16<1:13:08,  1.39it/s]
4|379|Loss: 0.8287539482116699:   6%|▌         | 379/6470 [04:17<1:15:13,  1.35it/s]
4|380|Loss: 0.8147699236869812:   6%|▌         | 379/6470 [04:17<1:15:13,  1.35it/s]
4|380|Loss: 0.8147699236869812:   6%|▌         | 380/6470 [04:17<1:15:33,  1.34it/s]
4|381|Loss: 0.7621407508850098:   6%|▌         | 380/6470 [04:18<1:15:33,  1.34it/s]
4|381|Loss: 0.7621407508850098:   6%|▌         | 381/6470 [04:18<1:13:04,  1.39it/s]
4|382|Loss: 0.8386391401290894:   6%|▌         | 381/6470 [04:18<1:13:04,  1.39it/s]
4|382|Loss: 0.8386391401290894:   6%|▌         | 382/6470 [04:19<1:09:03,  1.47it/s]
4|383|Loss: 0.6057456731796265:   6%|▌         | 382/6470 [04:19<1:09:03,  1.47it/s]
4|383|Loss: 0.6057456731796265:   6%|▌         | 383/6470 [04:19<1:06:55,  1.52it/s]
4|384|Loss: 0.9769071936607361:   6%|▌         | 383/6470 [04:20<1:06:55,  1.52it/s]
4|384|Loss: 0.9769071936607361:   6%|▌         | 384/6470 [04:20<1:09:59,  1.45it/s]
4|385|Loss: 0.8457320928573608:   6%|▌         | 384/6470 [04:20<1:09:59,  1.45it/s]
4|385|Loss: 0.8457320928573608:   6%|▌         | 385/6470 [04:21<1:11:54,  1.41it/s]
4|386|Loss: 0.9777165055274963:   6%|▌         | 385/6470 [04:21<1:11:54,  1.41it/s]
4|386|Loss: 0.9777165055274963:   6%|▌         | 386/6470 [04:22<1:14:30,  1.36it/s]
4|387|Loss: 0.8624398112297058:   6%|▌         | 386/6470 [04:22<1:14:30,  1.36it/s]
4|387|Loss: 0.8624398112297058:   6%|▌         | 387/6470 [04:22<1:10:03,  1.45it/s]
4|388|Loss: 0.7432591915130615:   6%|▌         | 387/6470 [04:23<1:10:03,  1.45it/s]
4|388|Loss: 0.7432591915130615:   6%|▌         | 388/6470 [04:23<1:09:16,  1.46it/s]
4|389|Loss: 0.7438311576843262:   6%|▌         | 388/6470 [04:23<1:09:16,  1.46it/s]
4|389|Loss: 0.7438311576843262:   6%|▌         | 389/6470 [04:23<1:07:50,  1.49it/s]
4|390|Loss: 0.8736722469329834:   6%|▌         | 389/6470 [04:24<1:07:50,  1.49it/s]
4|390|Loss: 0.8736722469329834:   6%|▌         | 390/6470 [04:24<1:08:27,  1.48it/s]
4|391|Loss: 0.948802649974823:   6%|▌         | 390/6470 [04:25<1:08:27,  1.48it/s] 
4|391|Loss: 0.948802649974823:   6%|▌         | 391/6470 [04:25<1:09:18,  1.46it/s]
4|392|Loss: 0.7824133634567261:   6%|▌         | 391/6470 [04:25<1:09:18,  1.46it/s]
4|392|Loss: 0.7824133634567261:   6%|▌         | 392/6470 [04:25<1:05:28,  1.55it/s]
4|393|Loss: 0.9020522236824036:   6%|▌         | 392/6470 [04:26<1:05:28,  1.55it/s]
4|393|Loss: 0.9020522236824036:   6%|▌         | 393/6470 [04:26<1:01:23,  1.65it/s]
4|394|Loss: 0.7753596305847168:   6%|▌         | 393/6470 [04:26<1:01:23,  1.65it/s]
4|394|Loss: 0.7753596305847168:   6%|▌         | 394/6470 [04:27<1:06:11,  1.53it/s]
4|395|Loss: 0.8031766414642334:   6%|▌         | 394/6470 [04:27<1:06:11,  1.53it/s]
4|395|Loss: 0.8031766414642334:   6%|▌         | 395/6470 [04:28<1:11:30,  1.42it/s]
4|396|Loss: 0.8695865273475647:   6%|▌         | 395/6470 [04:28<1:11:30,  1.42it/s]
4|396|Loss: 0.8695865273475647:   6%|▌         | 396/6470 [04:28<1:13:03,  1.39it/s]
4|397|Loss: 0.9033439755439758:   6%|▌         | 396/6470 [04:29<1:13:03,  1.39it/s]
4|397|Loss: 0.9033439755439758:   6%|▌         | 397/6470 [04:29<1:09:44,  1.45it/s]
4|398|Loss: 0.7607093453407288:   6%|▌         | 397/6470 [04:29<1:09:44,  1.45it/s]
4|398|Loss: 0.7607093453407288:   6%|▌         | 398/6470 [04:30<1:11:39,  1.41it/s]
4|399|Loss: 0.8288683295249939:   6%|▌         | 398/6470 [04:30<1:11:39,  1.41it/s]
4|399|Loss: 0.8288683295249939:   6%|▌         | 399/6470 [04:30<1:12:04,  1.40it/s]
4|400|Loss: 0.8121162056922913:   6%|▌         | 399/6470 [04:31<1:12:04,  1.40it/s]
4|400|Loss: 0.8121162056922913:   6%|▌         | 400/6470 [04:31<1:14:48,  1.35it/s]
4|401|Loss: 0.8721898198127747:   6%|▌         | 400/6470 [04:32<1:14:48,  1.35it/s]
4|401|Loss: 0.8721898198127747:   6%|▌         | 401/6470 [04:32<1:17:38,  1.30it/s]
4|402|Loss: 1.0538134574890137:   6%|▌         | 401/6470 [04:32<1:17:38,  1.30it/s]
4|402|Loss: 1.0538134574890137:   6%|▌         | 402/6470 [04:33<1:13:53,  1.37it/s]
4|403|Loss: 0.819261372089386:   6%|▌         | 402/6470 [04:33<1:13:53,  1.37it/s] 
4|403|Loss: 0.819261372089386:   6%|▌         | 403/6470 [04:33<1:10:56,  1.43it/s]
4|404|Loss: 0.7405571341514587:   6%|▌         | 403/6470 [04:34<1:10:56,  1.43it/s]
4|404|Loss: 0.7405571341514587:   6%|▌         | 404/6470 [04:34<1:09:06,  1.46it/s]
4|405|Loss: 0.9233825206756592:   6%|▌         | 404/6470 [04:34<1:09:06,  1.46it/s]
4|405|Loss: 0.9233825206756592:   6%|▋         | 405/6470 [04:35<1:07:44,  1.49it/s]
4|406|Loss: 0.9167363047599792:   6%|▋         | 405/6470 [04:35<1:07:44,  1.49it/s]
4|406|Loss: 0.9167363047599792:   6%|▋         | 406/6470 [04:35<1:04:45,  1.56it/s]
4|407|Loss: 0.8309144377708435:   6%|▋         | 406/6470 [04:36<1:04:45,  1.56it/s]
4|407|Loss: 0.8309144377708435:   6%|▋         | 407/6470 [04:36<1:09:37,  1.45it/s]
4|408|Loss: 0.7953073382377625:   6%|▋         | 407/6470 [04:36<1:09:37,  1.45it/s]
4|408|Loss: 0.7953073382377625:   6%|▋         | 408/6470 [04:37<1:08:12,  1.48it/s]
4|409|Loss: 0.7888311147689819:   6%|▋         | 408/6470 [04:37<1:08:12,  1.48it/s]
4|409|Loss: 0.7888311147689819:   6%|▋         | 409/6470 [04:37<1:11:34,  1.41it/s]
4|410|Loss: 0.9130426645278931:   6%|▋         | 409/6470 [04:38<1:11:34,  1.41it/s]
4|410|Loss: 0.9130426645278931:   6%|▋         | 410/6470 [04:38<1:09:07,  1.46it/s]
4|411|Loss: 0.808208703994751:   6%|▋         | 410/6470 [04:38<1:09:07,  1.46it/s] 
4|411|Loss: 0.808208703994751:   6%|▋         | 411/6470 [04:39<1:07:04,  1.51it/s]
4|412|Loss: 0.8765904903411865:   6%|▋         | 411/6470 [04:39<1:07:04,  1.51it/s]
4|412|Loss: 0.8765904903411865:   6%|▋         | 412/6470 [04:39<1:11:24,  1.41it/s]
4|413|Loss: 0.8541499972343445:   6%|▋         | 412/6470 [04:40<1:11:24,  1.41it/s]
4|413|Loss: 0.8541499972343445:   6%|▋         | 413/6470 [04:40<1:07:46,  1.49it/s]
4|414|Loss: 0.8912510871887207:   6%|▋         | 413/6470 [04:40<1:07:46,  1.49it/s]
4|414|Loss: 0.8912510871887207:   6%|▋         | 414/6470 [04:41<1:10:36,  1.43it/s]
4|415|Loss: 0.9100499749183655:   6%|▋         | 414/6470 [04:41<1:10:36,  1.43it/s]
4|415|Loss: 0.9100499749183655:   6%|▋         | 415/6470 [04:41<1:09:33,  1.45it/s]
4|416|Loss: 0.586733341217041:   6%|▋         | 415/6470 [04:42<1:09:33,  1.45it/s] 
4|416|Loss: 0.586733341217041:   6%|▋         | 416/6470 [04:42<1:12:31,  1.39it/s]
4|417|Loss: 0.875532865524292:   6%|▋         | 416/6470 [04:43<1:12:31,  1.39it/s]
4|417|Loss: 0.875532865524292:   6%|▋         | 417/6470 [04:43<1:12:45,  1.39it/s]
4|418|Loss: 0.8819096684455872:   6%|▋         | 417/6470 [04:43<1:12:45,  1.39it/s]
4|418|Loss: 0.8819096684455872:   6%|▋         | 418/6470 [04:44<1:10:50,  1.42it/s]
4|419|Loss: 0.8639142513275146:   6%|▋         | 418/6470 [04:44<1:10:50,  1.42it/s]
4|419|Loss: 0.8639142513275146:   6%|▋         | 419/6470 [04:44<1:11:10,  1.42it/s]
4|420|Loss: 0.7573629021644592:   6%|▋         | 419/6470 [04:45<1:11:10,  1.42it/s]
4|420|Loss: 0.7573629021644592:   6%|▋         | 420/6470 [04:45<1:13:42,  1.37it/s]
4|421|Loss: 1.1659494638442993:   6%|▋         | 420/6470 [04:46<1:13:42,  1.37it/s]
4|421|Loss: 1.1659494638442993:   7%|▋         | 421/6470 [04:46<1:10:41,  1.43it/s]
4|422|Loss: 0.7722458243370056:   7%|▋         | 421/6470 [04:46<1:10:41,  1.43it/s]
4|422|Loss: 0.7722458243370056:   7%|▋         | 422/6470 [04:46<1:10:13,  1.44it/s]
4|423|Loss: 0.7980624437332153:   7%|▋         | 422/6470 [04:47<1:10:13,  1.44it/s]
4|423|Loss: 0.7980624437332153:   7%|▋         | 423/6470 [04:47<1:10:29,  1.43it/s]
4|424|Loss: 0.8667947053909302:   7%|▋         | 423/6470 [04:48<1:10:29,  1.43it/s]
4|424|Loss: 0.8667947053909302:   7%|▋         | 424/6470 [04:48<1:06:24,  1.52it/s]
4|425|Loss: 0.7821043729782104:   7%|▋         | 424/6470 [04:48<1:06:24,  1.52it/s]
4|425|Loss: 0.7821043729782104:   7%|▋         | 425/6470 [04:48<1:03:50,  1.58it/s]
4|426|Loss: 0.8150492906570435:   7%|▋         | 425/6470 [04:49<1:03:50,  1.58it/s]
4|426|Loss: 0.8150492906570435:   7%|▋         | 426/6470 [04:49<1:08:22,  1.47it/s]
4|427|Loss: 0.8808761835098267:   7%|▋         | 426/6470 [04:49<1:08:22,  1.47it/s]
4|427|Loss: 0.8808761835098267:   7%|▋         | 427/6470 [04:50<1:01:45,  1.63it/s]
4|428|Loss: 0.6963031888008118:   7%|▋         | 427/6470 [04:50<1:01:45,  1.63it/s]
4|428|Loss: 0.6963031888008118:   7%|▋         | 428/6470 [04:50<1:00:01,  1.68it/s]
4|429|Loss: 0.8012087941169739:   7%|▋         | 428/6470 [04:51<1:00:01,  1.68it/s]
4|429|Loss: 0.8012087941169739:   7%|▋         | 429/6470 [04:51<1:04:39,  1.56it/s]
4|430|Loss: 0.9454526305198669:   7%|▋         | 429/6470 [04:51<1:04:39,  1.56it/s]
4|430|Loss: 0.9454526305198669:   7%|▋         | 430/6470 [04:52<1:07:46,  1.49it/s]
4|431|Loss: 0.8243968486785889:   7%|▋         | 430/6470 [04:52<1:07:46,  1.49it/s]
4|431|Loss: 0.8243968486785889:   7%|▋         | 431/6470 [04:52<1:07:23,  1.49it/s]
4|432|Loss: 0.7671219110488892:   7%|▋         | 431/6470 [04:53<1:07:23,  1.49it/s]
4|432|Loss: 0.7671219110488892:   7%|▋         | 432/6470 [04:53<1:04:55,  1.55it/s]
4|433|Loss: 0.8419528007507324:   7%|▋         | 432/6470 [04:53<1:04:55,  1.55it/s]
4|433|Loss: 0.8419528007507324:   7%|▋         | 433/6470 [04:54<1:06:47,  1.51it/s]
4|434|Loss: 0.9541530609130859:   7%|▋         | 433/6470 [04:54<1:06:47,  1.51it/s]
4|434|Loss: 0.9541530609130859:   7%|▋         | 434/6470 [04:54<1:08:58,  1.46it/s]
4|435|Loss: 0.8200101852416992:   7%|▋         | 434/6470 [04:55<1:08:58,  1.46it/s]
4|435|Loss: 0.8200101852416992:   7%|▋         | 435/6470 [04:55<1:08:13,  1.47it/s]
4|436|Loss: 0.8740925192832947:   7%|▋         | 435/6470 [04:55<1:08:13,  1.47it/s]
4|436|Loss: 0.8740925192832947:   7%|▋         | 436/6470 [04:56<1:11:06,  1.41it/s]
4|437|Loss: 0.866047203540802:   7%|▋         | 436/6470 [04:56<1:11:06,  1.41it/s] 
4|437|Loss: 0.866047203540802:   7%|▋         | 437/6470 [04:56<1:11:47,  1.40it/s]
4|438|Loss: 1.0819069147109985:   7%|▋         | 437/6470 [04:57<1:11:47,  1.40it/s]
4|438|Loss: 1.0819069147109985:   7%|▋         | 438/6470 [04:57<1:15:30,  1.33it/s]
4|439|Loss: 0.8094387650489807:   7%|▋         | 438/6470 [04:58<1:15:30,  1.33it/s]
4|439|Loss: 0.8094387650489807:   7%|▋         | 439/6470 [04:58<1:16:04,  1.32it/s]
4|440|Loss: 0.7468510270118713:   7%|▋         | 439/6470 [04:58<1:16:04,  1.32it/s]
4|440|Loss: 0.7468510270118713:   7%|▋         | 440/6470 [04:59<1:13:17,  1.37it/s]
4|441|Loss: 0.9710734486579895:   7%|▋         | 440/6470 [04:59<1:13:17,  1.37it/s]
4|441|Loss: 0.9710734486579895:   7%|▋         | 441/6470 [04:59<1:12:11,  1.39it/s]
4|442|Loss: 0.7889138460159302:   7%|▋         | 441/6470 [05:00<1:12:11,  1.39it/s]
4|442|Loss: 0.7889138460159302:   7%|▋         | 442/6470 [05:00<1:12:44,  1.38it/s]
4|443|Loss: 0.7918215990066528:   7%|▋         | 442/6470 [05:01<1:12:44,  1.38it/s]
4|443|Loss: 0.7918215990066528:   7%|▋         | 443/6470 [05:01<1:13:39,  1.36it/s]
4|444|Loss: 0.8509422540664673:   7%|▋         | 443/6470 [05:01<1:13:39,  1.36it/s]
4|444|Loss: 0.8509422540664673:   7%|▋         | 444/6470 [05:02<1:11:45,  1.40it/s]
4|445|Loss: 0.9281165599822998:   7%|▋         | 444/6470 [05:02<1:11:45,  1.40it/s]
4|445|Loss: 0.9281165599822998:   7%|▋         | 445/6470 [05:02<1:11:43,  1.40it/s]
4|446|Loss: 0.8668643236160278:   7%|▋         | 445/6470 [05:03<1:11:43,  1.40it/s]
4|446|Loss: 0.8668643236160278:   7%|▋         | 446/6470 [05:03<1:09:37,  1.44it/s]
4|447|Loss: 0.8769717812538147:   7%|▋         | 446/6470 [05:03<1:09:37,  1.44it/s]
4|447|Loss: 0.8769717812538147:   7%|▋         | 447/6470 [05:04<1:07:28,  1.49it/s]
4|448|Loss: 0.8588736057281494:   7%|▋         | 447/6470 [05:04<1:07:28,  1.49it/s]
4|448|Loss: 0.8588736057281494:   7%|▋         | 448/6470 [05:04<1:10:27,  1.42it/s]
4|449|Loss: 0.8263460397720337:   7%|▋         | 448/6470 [05:05<1:10:27,  1.42it/s]
4|449|Loss: 0.8263460397720337:   7%|▋         | 449/6470 [05:05<1:08:04,  1.47it/s]
4|450|Loss: 0.9367730617523193:   7%|▋         | 449/6470 [05:05<1:08:04,  1.47it/s]
4|450|Loss: 0.9367730617523193:   7%|▋         | 450/6470 [05:06<1:08:09,  1.47it/s]
4|451|Loss: 0.951444685459137:   7%|▋         | 450/6470 [05:06<1:08:09,  1.47it/s] 
4|451|Loss: 0.951444685459137:   7%|▋         | 451/6470 [05:06<1:09:42,  1.44it/s]
4|452|Loss: 0.7977384924888611:   7%|▋         | 451/6470 [05:07<1:09:42,  1.44it/s]
4|452|Loss: 0.7977384924888611:   7%|▋         | 452/6470 [05:07<1:10:58,  1.41it/s]
4|453|Loss: 0.8795256614685059:   7%|▋         | 452/6470 [05:08<1:10:58,  1.41it/s]
4|453|Loss: 0.8795256614685059:   7%|▋         | 453/6470 [05:08<1:13:05,  1.37it/s]
4|454|Loss: 0.6890608072280884:   7%|▋         | 453/6470 [05:08<1:13:05,  1.37it/s]
4|454|Loss: 0.6890608072280884:   7%|▋         | 454/6470 [05:09<1:11:42,  1.40it/s]
4|455|Loss: 0.8192330598831177:   7%|▋         | 454/6470 [05:09<1:11:42,  1.40it/s]
4|455|Loss: 0.8192330598831177:   7%|▋         | 455/6470 [05:09<1:08:55,  1.45it/s]
4|456|Loss: 1.1444549560546875:   7%|▋         | 455/6470 [05:10<1:08:55,  1.45it/s]
4|456|Loss: 1.1444549560546875:   7%|▋         | 456/6470 [05:10<1:02:49,  1.60it/s]
4|457|Loss: 1.0279462337493896:   7%|▋         | 456/6470 [05:10<1:02:49,  1.60it/s]
4|457|Loss: 1.0279462337493896:   7%|▋         | 457/6470 [05:10<59:48,  1.68it/s]  
4|458|Loss: 0.8864352107048035:   7%|▋         | 457/6470 [05:11<59:48,  1.68it/s]
4|458|Loss: 0.8864352107048035:   7%|▋         | 458/6470 [05:11<1:02:10,  1.61it/s]
4|459|Loss: 0.8996264338493347:   7%|▋         | 458/6470 [05:11<1:02:10,  1.61it/s]
4|459|Loss: 0.8996264338493347:   7%|▋         | 459/6470 [05:11<1:01:48,  1.62it/s]
4|460|Loss: 0.8371911644935608:   7%|▋         | 459/6470 [05:12<1:01:48,  1.62it/s]
4|460|Loss: 0.8371911644935608:   7%|▋         | 460/6470 [05:12<1:01:42,  1.62it/s]
4|461|Loss: 0.9864640235900879:   7%|▋         | 460/6470 [05:13<1:01:42,  1.62it/s]
4|461|Loss: 0.9864640235900879:   7%|▋         | 461/6470 [05:13<1:06:13,  1.51it/s]
4|462|Loss: 0.8904075026512146:   7%|▋         | 461/6470 [05:13<1:06:13,  1.51it/s]
4|462|Loss: 0.8904075026512146:   7%|▋         | 462/6470 [05:14<1:05:11,  1.54it/s]
4|463|Loss: 1.0396363735198975:   7%|▋         | 462/6470 [05:14<1:05:11,  1.54it/s]
4|463|Loss: 1.0396363735198975:   7%|▋         | 463/6470 [05:14<1:04:21,  1.56it/s]
4|464|Loss: 0.8065412044525146:   7%|▋         | 463/6470 [05:15<1:04:21,  1.56it/s]
4|464|Loss: 0.8065412044525146:   7%|▋         | 464/6470 [05:15<1:09:21,  1.44it/s]
4|465|Loss: 0.9667661786079407:   7%|▋         | 464/6470 [05:15<1:09:21,  1.44it/s]
4|465|Loss: 0.9667661786079407:   7%|▋         | 465/6470 [05:16<1:08:42,  1.46it/s]
4|466|Loss: 0.8162146210670471:   7%|▋         | 465/6470 [05:16<1:08:42,  1.46it/s]
4|466|Loss: 0.8162146210670471:   7%|▋         | 466/6470 [05:16<1:13:59,  1.35it/s]
4|467|Loss: 0.8313375115394592:   7%|▋         | 466/6470 [05:17<1:13:59,  1.35it/s]
4|467|Loss: 0.8313375115394592:   7%|▋         | 467/6470 [05:17<1:15:01,  1.33it/s]
4|468|Loss: 0.7558957934379578:   7%|▋         | 467/6470 [05:18<1:15:01,  1.33it/s]
4|468|Loss: 0.7558957934379578:   7%|▋         | 468/6470 [05:18<1:16:29,  1.31it/s]
4|469|Loss: 0.7227720022201538:   7%|▋         | 468/6470 [05:18<1:16:29,  1.31it/s]
4|469|Loss: 0.7227720022201538:   7%|▋         | 469/6470 [05:19<1:10:41,  1.41it/s]
4|470|Loss: 0.8468207120895386:   7%|▋         | 469/6470 [05:19<1:10:41,  1.41it/s]
4|470|Loss: 0.8468207120895386:   7%|▋         | 470/6470 [05:19<1:10:03,  1.43it/s]
4|471|Loss: 0.8366920948028564:   7%|▋         | 470/6470 [05:20<1:10:03,  1.43it/s]
4|471|Loss: 0.8366920948028564:   7%|▋         | 471/6470 [05:20<1:07:15,  1.49it/s]
4|472|Loss: 0.9996499419212341:   7%|▋         | 471/6470 [05:20<1:07:15,  1.49it/s]
4|472|Loss: 0.9996499419212341:   7%|▋         | 472/6470 [05:20<1:03:48,  1.57it/s]
4|473|Loss: 0.8617734313011169:   7%|▋         | 472/6470 [05:21<1:03:48,  1.57it/s]
4|473|Loss: 0.8617734313011169:   7%|▋         | 473/6470 [05:21<1:01:53,  1.62it/s]
4|474|Loss: 0.8449796438217163:   7%|▋         | 473/6470 [05:21<1:01:53,  1.62it/s]
4|474|Loss: 0.8449796438217163:   7%|▋         | 474/6470 [05:22<1:01:44,  1.62it/s]
4|475|Loss: 0.7462202906608582:   7%|▋         | 474/6470 [05:22<1:01:44,  1.62it/s]
4|475|Loss: 0.7462202906608582:   7%|▋         | 475/6470 [05:22<1:02:14,  1.61it/s]
4|476|Loss: 0.8017908334732056:   7%|▋         | 475/6470 [05:23<1:02:14,  1.61it/s]
4|476|Loss: 0.8017908334732056:   7%|▋         | 476/6470 [05:23<1:02:19,  1.60it/s]
4|477|Loss: 0.7632579207420349:   7%|▋         | 476/6470 [05:23<1:02:19,  1.60it/s]
4|477|Loss: 0.7632579207420349:   7%|▋         | 477/6470 [05:24<1:01:08,  1.63it/s]
4|478|Loss: 0.8369094133377075:   7%|▋         | 477/6470 [05:24<1:01:08,  1.63it/s]
4|478|Loss: 0.8369094133377075:   7%|▋         | 478/6470 [05:24<1:07:33,  1.48it/s]
4|479|Loss: 0.7972222566604614:   7%|▋         | 478/6470 [05:25<1:07:33,  1.48it/s]
4|479|Loss: 0.7972222566604614:   7%|▋         | 479/6470 [05:25<1:09:38,  1.43it/s]
4|480|Loss: 0.8994036316871643:   7%|▋         | 479/6470 [05:25<1:09:38,  1.43it/s]
4|480|Loss: 0.8994036316871643:   7%|▋         | 480/6470 [05:26<1:10:43,  1.41it/s]
4|481|Loss: 0.8126974105834961:   7%|▋         | 480/6470 [05:26<1:10:43,  1.41it/s]
4|481|Loss: 0.8126974105834961:   7%|▋         | 481/6470 [05:27<1:10:49,  1.41it/s]
4|482|Loss: 0.8722034692764282:   7%|▋         | 481/6470 [05:27<1:10:49,  1.41it/s]
4|482|Loss: 0.8722034692764282:   7%|▋         | 482/6470 [05:27<1:11:51,  1.39it/s]
4|483|Loss: 0.7824392318725586:   7%|▋         | 482/6470 [05:28<1:11:51,  1.39it/s]
4|483|Loss: 0.7824392318725586:   7%|▋         | 483/6470 [05:28<1:12:29,  1.38it/s]
4|484|Loss: 0.9134522676467896:   7%|▋         | 483/6470 [05:28<1:12:29,  1.38it/s]
4|484|Loss: 0.9134522676467896:   7%|▋         | 484/6470 [05:29<1:12:41,  1.37it/s]
4|485|Loss: 0.8047006726264954:   7%|▋         | 484/6470 [05:29<1:12:41,  1.37it/s]
4|485|Loss: 0.8047006726264954:   7%|▋         | 485/6470 [05:30<1:14:39,  1.34it/s]
4|486|Loss: 0.8576589226722717:   7%|▋         | 485/6470 [05:30<1:14:39,  1.34it/s]
4|486|Loss: 0.8576589226722717:   8%|▊         | 486/6470 [05:30<1:20:15,  1.24it/s]
4|487|Loss: 0.7512041926383972:   8%|▊         | 486/6470 [05:31<1:20:15,  1.24it/s]
4|487|Loss: 0.7512041926383972:   8%|▊         | 487/6470 [05:31<1:16:06,  1.31it/s]
4|488|Loss: 0.8144888877868652:   8%|▊         | 487/6470 [05:32<1:16:06,  1.31it/s]
4|488|Loss: 0.8144888877868652:   8%|▊         | 488/6470 [05:32<1:14:09,  1.34it/s]
4|489|Loss: 0.8691936731338501:   8%|▊         | 488/6470 [05:32<1:14:09,  1.34it/s]
4|489|Loss: 0.8691936731338501:   8%|▊         | 489/6470 [05:33<1:13:10,  1.36it/s]
4|490|Loss: 0.8130617737770081:   8%|▊         | 489/6470 [05:33<1:13:10,  1.36it/s]
4|490|Loss: 0.8130617737770081:   8%|▊         | 490/6470 [05:33<1:15:30,  1.32it/s]
4|491|Loss: 0.9610170125961304:   8%|▊         | 490/6470 [05:34<1:15:30,  1.32it/s]
4|491|Loss: 0.9610170125961304:   8%|▊         | 491/6470 [05:34<1:11:51,  1.39it/s]
4|492|Loss: 0.770138680934906:   8%|▊         | 491/6470 [05:34<1:11:51,  1.39it/s] 
4|492|Loss: 0.770138680934906:   8%|▊         | 492/6470 [05:35<1:14:55,  1.33it/s]
4|493|Loss: 0.9233449101448059:   8%|▊         | 492/6470 [05:35<1:14:55,  1.33it/s]
4|493|Loss: 0.9233449101448059:   8%|▊         | 493/6470 [05:35<1:11:57,  1.38it/s]
4|494|Loss: 0.869408905506134:   8%|▊         | 493/6470 [05:36<1:11:57,  1.38it/s] 
4|494|Loss: 0.869408905506134:   8%|▊         | 494/6470 [05:36<1:11:33,  1.39it/s]
4|495|Loss: 0.7434275150299072:   8%|▊         | 494/6470 [05:37<1:11:33,  1.39it/s]
4|495|Loss: 0.7434275150299072:   8%|▊         | 495/6470 [05:37<1:12:10,  1.38it/s]
4|496|Loss: 0.8481338620185852:   8%|▊         | 495/6470 [05:37<1:12:10,  1.38it/s]
4|496|Loss: 0.8481338620185852:   8%|▊         | 496/6470 [05:38<1:14:52,  1.33it/s]
4|497|Loss: 0.7594103813171387:   8%|▊         | 496/6470 [05:38<1:14:52,  1.33it/s]
4|497|Loss: 0.7594103813171387:   8%|▊         | 497/6470 [05:38<1:12:44,  1.37it/s]
4|498|Loss: 0.838447093963623:   8%|▊         | 497/6470 [05:39<1:12:44,  1.37it/s] 
4|498|Loss: 0.838447093963623:   8%|▊         | 498/6470 [05:39<1:07:49,  1.47it/s]
4|499|Loss: 0.7984289526939392:   8%|▊         | 498/6470 [05:39<1:07:49,  1.47it/s]
4|499|Loss: 0.7984289526939392:   8%|▊         | 499/6470 [05:40<1:11:41,  1.39it/s]
4|500|Loss: 0.8178704977035522:   8%|▊         | 499/6470 [05:40<1:11:41,  1.39it/s]
4|500|Loss: 0.8178704977035522:   8%|▊         | 500/6470 [05:40<1:10:24,  1.41it/s]
4|501|Loss: 0.8694911003112793:   8%|▊         | 500/6470 [05:41<1:10:24,  1.41it/s]
4|501|Loss: 0.8694911003112793:   8%|▊         | 501/6470 [05:41<1:07:47,  1.47it/s]
4|502|Loss: 0.8899133801460266:   8%|▊         | 501/6470 [05:41<1:07:47,  1.47it/s]
4|502|Loss: 0.8899133801460266:   8%|▊         | 502/6470 [05:42<1:02:01,  1.60it/s]
4|503|Loss: 0.7353939414024353:   8%|▊         | 502/6470 [05:42<1:02:01,  1.60it/s]
4|503|Loss: 0.7353939414024353:   8%|▊         | 503/6470 [05:42<1:01:23,  1.62it/s]
4|504|Loss: 0.7652784585952759:   8%|▊         | 503/6470 [05:43<1:01:23,  1.62it/s]
4|504|Loss: 0.7652784585952759:   8%|▊         | 504/6470 [05:43<1:03:59,  1.55it/s]
4|505|Loss: 0.7338755130767822:   8%|▊         | 504/6470 [05:43<1:03:59,  1.55it/s]
4|505|Loss: 0.7338755130767822:   8%|▊         | 505/6470 [05:44<1:04:27,  1.54it/s]
4|506|Loss: 0.8771697878837585:   8%|▊         | 505/6470 [05:44<1:04:27,  1.54it/s]
4|506|Loss: 0.8771697878837585:   8%|▊         | 506/6470 [05:44<1:09:23,  1.43it/s]
4|507|Loss: 0.7962554693222046:   8%|▊         | 506/6470 [05:45<1:09:23,  1.43it/s]
4|507|Loss: 0.7962554693222046:   8%|▊         | 507/6470 [05:45<1:06:03,  1.50it/s]
4|508|Loss: 0.8556451797485352:   8%|▊         | 507/6470 [05:45<1:06:03,  1.50it/s]
4|508|Loss: 0.8556451797485352:   8%|▊         | 508/6470 [05:46<1:02:55,  1.58it/s]
4|509|Loss: 0.8600943088531494:   8%|▊         | 508/6470 [05:46<1:02:55,  1.58it/s]
4|509|Loss: 0.8600943088531494:   8%|▊         | 509/6470 [05:46<1:02:51,  1.58it/s]
4|510|Loss: 0.8440943360328674:   8%|▊         | 509/6470 [05:47<1:02:51,  1.58it/s]
4|510|Loss: 0.8440943360328674:   8%|▊         | 510/6470 [05:47<1:03:08,  1.57it/s]
4|511|Loss: 0.8324994444847107:   8%|▊         | 510/6470 [05:47<1:03:08,  1.57it/s]
4|511|Loss: 0.8324994444847107:   8%|▊         | 511/6470 [05:48<1:08:29,  1.45it/s]
4|512|Loss: 0.9997717142105103:   8%|▊         | 511/6470 [05:48<1:08:29,  1.45it/s]
4|512|Loss: 0.9997717142105103:   8%|▊         | 512/6470 [05:48<1:13:35,  1.35it/s]
4|513|Loss: 0.8272351622581482:   8%|▊         | 512/6470 [05:49<1:13:35,  1.35it/s]
4|513|Loss: 0.8272351622581482:   8%|▊         | 513/6470 [05:49<1:14:09,  1.34it/s]
4|514|Loss: 0.9191269874572754:   8%|▊         | 513/6470 [05:50<1:14:09,  1.34it/s]
4|514|Loss: 0.9191269874572754:   8%|▊         | 514/6470 [05:50<1:13:10,  1.36it/s]
4|515|Loss: 0.9331917762756348:   8%|▊         | 514/6470 [05:50<1:13:10,  1.36it/s]
4|515|Loss: 0.9331917762756348:   8%|▊         | 515/6470 [05:51<1:10:55,  1.40it/s]
4|516|Loss: 0.8062582015991211:   8%|▊         | 515/6470 [05:51<1:10:55,  1.40it/s]
4|516|Loss: 0.8062582015991211:   8%|▊         | 516/6470 [05:51<1:14:30,  1.33it/s]
4|517|Loss: 0.9624560475349426:   8%|▊         | 516/6470 [05:52<1:14:30,  1.33it/s]
4|517|Loss: 0.9624560475349426:   8%|▊         | 517/6470 [05:52<1:18:21,  1.27it/s]
4|518|Loss: 0.7446183562278748:   8%|▊         | 517/6470 [05:53<1:18:21,  1.27it/s]
4|518|Loss: 0.7446183562278748:   8%|▊         | 518/6470 [05:53<1:15:41,  1.31it/s]
4|519|Loss: 0.8363354802131653:   8%|▊         | 518/6470 [05:53<1:15:41,  1.31it/s]
4|519|Loss: 0.8363354802131653:   8%|▊         | 519/6470 [05:54<1:17:53,  1.27it/s]
4|520|Loss: 0.8803063035011292:   8%|▊         | 519/6470 [05:54<1:17:53,  1.27it/s]
4|520|Loss: 0.8803063035011292:   8%|▊         | 520/6470 [05:55<1:17:19,  1.28it/s]
4|521|Loss: 0.9797031283378601:   8%|▊         | 520/6470 [05:55<1:17:19,  1.28it/s]
4|521|Loss: 0.9797031283378601:   8%|▊         | 521/6470 [05:55<1:13:39,  1.35it/s]
4|522|Loss: 0.8287940621376038:   8%|▊         | 521/6470 [05:56<1:13:39,  1.35it/s]
4|522|Loss: 0.8287940621376038:   8%|▊         | 522/6470 [05:56<1:09:01,  1.44it/s]
4|523|Loss: 1.0632729530334473:   8%|▊         | 522/6470 [05:56<1:09:01,  1.44it/s]
4|523|Loss: 1.0632729530334473:   8%|▊         | 523/6470 [05:57<1:09:47,  1.42it/s]
4|524|Loss: 0.9470463991165161:   8%|▊         | 523/6470 [05:57<1:09:47,  1.42it/s]
4|524|Loss: 0.9470463991165161:   8%|▊         | 524/6470 [05:57<1:08:24,  1.45it/s]
4|525|Loss: 0.9607980847358704:   8%|▊         | 524/6470 [05:58<1:08:24,  1.45it/s]
4|525|Loss: 0.9607980847358704:   8%|▊         | 525/6470 [05:58<1:08:21,  1.45it/s]
4|526|Loss: 0.8160211443901062:   8%|▊         | 525/6470 [05:58<1:08:21,  1.45it/s]
4|526|Loss: 0.8160211443901062:   8%|▊         | 526/6470 [05:59<1:09:30,  1.43it/s]
4|527|Loss: 0.8015625476837158:   8%|▊         | 526/6470 [05:59<1:09:30,  1.43it/s]
4|527|Loss: 0.8015625476837158:   8%|▊         | 527/6470 [05:59<1:05:05,  1.52it/s]
4|528|Loss: 0.8335279226303101:   8%|▊         | 527/6470 [06:00<1:05:05,  1.52it/s]
4|528|Loss: 0.8335279226303101:   8%|▊         | 528/6470 [06:00<1:02:26,  1.59it/s]
4|529|Loss: 0.8166312575340271:   8%|▊         | 528/6470 [06:00<1:02:26,  1.59it/s]
4|529|Loss: 0.8166312575340271:   8%|▊         | 529/6470 [06:00<1:02:19,  1.59it/s]
4|530|Loss: 0.8956173658370972:   8%|▊         | 529/6470 [06:01<1:02:19,  1.59it/s]
4|530|Loss: 0.8956173658370972:   8%|▊         | 530/6470 [06:01<1:04:54,  1.53it/s]
4|531|Loss: 0.7639479637145996:   8%|▊         | 530/6470 [06:01<1:04:54,  1.53it/s]
4|531|Loss: 0.7639479637145996:   8%|▊         | 531/6470 [06:02<1:01:55,  1.60it/s]
4|532|Loss: 0.7945771217346191:   8%|▊         | 531/6470 [06:02<1:01:55,  1.60it/s]
4|532|Loss: 0.7945771217346191:   8%|▊         | 532/6470 [06:02<1:03:04,  1.57it/s]
4|533|Loss: 0.8468698859214783:   8%|▊         | 532/6470 [06:03<1:03:04,  1.57it/s]
4|533|Loss: 0.8468698859214783:   8%|▊         | 533/6470 [06:03<1:03:42,  1.55it/s]
4|534|Loss: 0.7822661995887756:   8%|▊         | 533/6470 [06:03<1:03:42,  1.55it/s]
4|534|Loss: 0.7822661995887756:   8%|▊         | 534/6470 [06:04<1:09:53,  1.42it/s]
4|535|Loss: 0.8089991807937622:   8%|▊         | 534/6470 [06:04<1:09:53,  1.42it/s]
4|535|Loss: 0.8089991807937622:   8%|▊         | 535/6470 [06:05<1:09:39,  1.42it/s]
4|536|Loss: 0.9042100310325623:   8%|▊         | 535/6470 [06:05<1:09:39,  1.42it/s]
4|536|Loss: 0.9042100310325623:   8%|▊         | 536/6470 [06:05<1:07:11,  1.47it/s]
4|537|Loss: 0.9148304462432861:   8%|▊         | 536/6470 [06:06<1:07:11,  1.47it/s]
4|537|Loss: 0.9148304462432861:   8%|▊         | 537/6470 [06:06<1:11:39,  1.38it/s]
4|538|Loss: 0.9701744914054871:   8%|▊         | 537/6470 [06:06<1:11:39,  1.38it/s]
4|538|Loss: 0.9701744914054871:   8%|▊         | 538/6470 [06:07<1:06:34,  1.49it/s]
4|539|Loss: 0.8525527715682983:   8%|▊         | 538/6470 [06:07<1:06:34,  1.49it/s]
4|539|Loss: 0.8525527715682983:   8%|▊         | 539/6470 [06:07<1:08:19,  1.45it/s]
4|540|Loss: 0.8543971180915833:   8%|▊         | 539/6470 [06:08<1:08:19,  1.45it/s]
4|540|Loss: 0.8543971180915833:   8%|▊         | 540/6470 [06:08<1:08:09,  1.45it/s]
4|541|Loss: 0.7054691314697266:   8%|▊         | 540/6470 [06:08<1:08:09,  1.45it/s]
4|541|Loss: 0.7054691314697266:   8%|▊         | 541/6470 [06:09<1:03:58,  1.54it/s]
4|542|Loss: 0.9665873646736145:   8%|▊         | 541/6470 [06:09<1:03:58,  1.54it/s]
4|542|Loss: 0.9665873646736145:   8%|▊         | 542/6470 [06:09<1:03:09,  1.56it/s]
4|543|Loss: 0.8402475118637085:   8%|▊         | 542/6470 [06:10<1:03:09,  1.56it/s]
4|543|Loss: 0.8402475118637085:   8%|▊         | 543/6470 [06:10<1:04:50,  1.52it/s]
4|544|Loss: 0.837721049785614:   8%|▊         | 543/6470 [06:10<1:04:50,  1.52it/s] 
4|544|Loss: 0.837721049785614:   8%|▊         | 544/6470 [06:11<1:06:54,  1.48it/s]
4|545|Loss: 0.8495702147483826:   8%|▊         | 544/6470 [06:11<1:06:54,  1.48it/s]
4|545|Loss: 0.8495702147483826:   8%|▊         | 545/6470 [06:11<1:07:38,  1.46it/s]
4|546|Loss: 0.969497561454773:   8%|▊         | 545/6470 [06:12<1:07:38,  1.46it/s] 
4|546|Loss: 0.969497561454773:   8%|▊         | 546/6470 [06:12<1:11:36,  1.38it/s]
4|547|Loss: 0.6911611557006836:   8%|▊         | 546/6470 [06:12<1:11:36,  1.38it/s]
4|547|Loss: 0.6911611557006836:   8%|▊         | 547/6470 [06:13<1:06:05,  1.49it/s]
4|548|Loss: 0.9053394794464111:   8%|▊         | 547/6470 [06:13<1:06:05,  1.49it/s]
4|548|Loss: 0.9053394794464111:   8%|▊         | 548/6470 [06:13<1:00:58,  1.62it/s]
4|549|Loss: 0.7060228586196899:   8%|▊         | 548/6470 [06:13<1:00:58,  1.62it/s]
4|549|Loss: 0.7060228586196899:   8%|▊         | 549/6470 [06:14<1:00:54,  1.62it/s]
4|550|Loss: 0.8536136150360107:   8%|▊         | 549/6470 [06:14<1:00:54,  1.62it/s]
4|550|Loss: 0.8536136150360107:   9%|▊         | 550/6470 [06:14<1:01:41,  1.60it/s]
4|551|Loss: 0.8736287355422974:   9%|▊         | 550/6470 [06:15<1:01:41,  1.60it/s]
4|551|Loss: 0.8736287355422974:   9%|▊         | 551/6470 [06:15<1:06:43,  1.48it/s]
4|552|Loss: 0.7739514708518982:   9%|▊         | 551/6470 [06:16<1:06:43,  1.48it/s]
4|552|Loss: 0.7739514708518982:   9%|▊         | 552/6470 [06:16<1:08:56,  1.43it/s]
4|553|Loss: 0.872195303440094:   9%|▊         | 552/6470 [06:16<1:08:56,  1.43it/s] 
4|553|Loss: 0.872195303440094:   9%|▊         | 553/6470 [06:16<1:03:18,  1.56it/s]
4|554|Loss: 1.0492990016937256:   9%|▊         | 553/6470 [06:17<1:03:18,  1.56it/s]
4|554|Loss: 1.0492990016937256:   9%|▊         | 554/6470 [06:17<1:01:41,  1.60it/s]
4|555|Loss: 0.8118475079536438:   9%|▊         | 554/6470 [06:17<1:01:41,  1.60it/s]
4|555|Loss: 0.8118475079536438:   9%|▊         | 555/6470 [06:18<1:08:48,  1.43it/s]
4|556|Loss: 0.8114059567451477:   9%|▊         | 555/6470 [06:18<1:08:48,  1.43it/s]
4|556|Loss: 0.8114059567451477:   9%|▊         | 556/6470 [06:19<1:10:29,  1.40it/s]
4|557|Loss: 0.7770113348960876:   9%|▊         | 556/6470 [06:19<1:10:29,  1.40it/s]
4|557|Loss: 0.7770113348960876:   9%|▊         | 557/6470 [06:19<1:12:14,  1.36it/s]
4|558|Loss: 0.89928138256073:   9%|▊         | 557/6470 [06:20<1:12:14,  1.36it/s]  
4|558|Loss: 0.89928138256073:   9%|▊         | 558/6470 [06:20<1:12:38,  1.36it/s]
4|559|Loss: 0.7670905590057373:   9%|▊         | 558/6470 [06:21<1:12:38,  1.36it/s]
4|559|Loss: 0.7670905590057373:   9%|▊         | 559/6470 [06:21<1:09:06,  1.43it/s]
4|560|Loss: 0.9466400146484375:   9%|▊         | 559/6470 [06:21<1:09:06,  1.43it/s]
4|560|Loss: 0.9466400146484375:   9%|▊         | 560/6470 [06:21<1:06:32,  1.48it/s]
4|561|Loss: 0.6905640959739685:   9%|▊         | 560/6470 [06:22<1:06:32,  1.48it/s]
4|561|Loss: 0.6905640959739685:   9%|▊         | 561/6470 [06:22<1:01:02,  1.61it/s]
4|562|Loss: 0.6385645866394043:   9%|▊         | 561/6470 [06:22<1:01:02,  1.61it/s]
4|562|Loss: 0.6385645866394043:   9%|▊         | 562/6470 [06:23<1:05:18,  1.51it/s]
4|563|Loss: 0.8524729013442993:   9%|▊         | 562/6470 [06:23<1:05:18,  1.51it/s]
4|563|Loss: 0.8524729013442993:   9%|▊         | 563/6470 [06:23<1:09:59,  1.41it/s]
4|564|Loss: 0.8532735705375671:   9%|▊         | 563/6470 [06:24<1:09:59,  1.41it/s]
4|564|Loss: 0.8532735705375671:   9%|▊         | 564/6470 [06:24<1:14:14,  1.33it/s]
4|565|Loss: 0.7656765580177307:   9%|▊         | 564/6470 [06:25<1:14:14,  1.33it/s]
4|565|Loss: 0.7656765580177307:   9%|▊         | 565/6470 [06:25<1:10:43,  1.39it/s]
4|566|Loss: 0.728011965751648:   9%|▊         | 565/6470 [06:25<1:10:43,  1.39it/s] 
4|566|Loss: 0.728011965751648:   9%|▊         | 566/6470 [06:26<1:06:43,  1.47it/s]
4|567|Loss: 0.8514506816864014:   9%|▊         | 566/6470 [06:26<1:06:43,  1.47it/s]
4|567|Loss: 0.8514506816864014:   9%|▉         | 567/6470 [06:26<1:03:50,  1.54it/s]
4|568|Loss: 0.7366008162498474:   9%|▉         | 567/6470 [06:27<1:03:50,  1.54it/s]
4|568|Loss: 0.7366008162498474:   9%|▉         | 568/6470 [06:27<1:10:56,  1.39it/s]
4|569|Loss: 0.7903854846954346:   9%|▉         | 568/6470 [06:28<1:10:56,  1.39it/s]
4|569|Loss: 0.7903854846954346:   9%|▉         | 569/6470 [06:28<1:13:34,  1.34it/s]
4|570|Loss: 0.6880658268928528:   9%|▉         | 569/6470 [06:28<1:13:34,  1.34it/s]
4|570|Loss: 0.6880658268928528:   9%|▉         | 570/6470 [06:28<1:09:42,  1.41it/s]
4|571|Loss: 0.9570927023887634:   9%|▉         | 570/6470 [06:29<1:09:42,  1.41it/s]
4|571|Loss: 0.9570927023887634:   9%|▉         | 571/6470 [06:29<1:02:56,  1.56it/s]
4|572|Loss: 0.9422966837882996:   9%|▉         | 571/6470 [06:29<1:02:56,  1.56it/s]
4|572|Loss: 0.9422966837882996:   9%|▉         | 572/6470 [06:30<1:03:14,  1.55it/s]
4|573|Loss: 0.7751229405403137:   9%|▉         | 572/6470 [06:30<1:03:14,  1.55it/s]
4|573|Loss: 0.7751229405403137:   9%|▉         | 573/6470 [06:30<1:08:38,  1.43it/s]
4|574|Loss: 0.8066540956497192:   9%|▉         | 573/6470 [06:31<1:08:38,  1.43it/s]
4|574|Loss: 0.8066540956497192:   9%|▉         | 574/6470 [06:31<1:09:42,  1.41it/s]
4|575|Loss: 0.7936137318611145:   9%|▉         | 574/6470 [06:32<1:09:42,  1.41it/s]
4|575|Loss: 0.7936137318611145:   9%|▉         | 575/6470 [06:32<1:06:29,  1.48it/s]
4|576|Loss: 0.7828314304351807:   9%|▉         | 575/6470 [06:32<1:06:29,  1.48it/s]
4|576|Loss: 0.7828314304351807:   9%|▉         | 576/6470 [06:32<1:06:39,  1.47it/s]
4|577|Loss: 0.7804200649261475:   9%|▉         | 576/6470 [06:33<1:06:39,  1.47it/s]
4|577|Loss: 0.7804200649261475:   9%|▉         | 577/6470 [06:33<1:11:00,  1.38it/s]
4|578|Loss: 0.8930811882019043:   9%|▉         | 577/6470 [06:34<1:11:00,  1.38it/s]
4|578|Loss: 0.8930811882019043:   9%|▉         | 578/6470 [06:34<1:10:11,  1.40it/s]
4|579|Loss: 0.802952229976654:   9%|▉         | 578/6470 [06:34<1:10:11,  1.40it/s] 
4|579|Loss: 0.802952229976654:   9%|▉         | 579/6470 [06:35<1:12:56,  1.35it/s]
4|580|Loss: 0.7898054718971252:   9%|▉         | 579/6470 [06:35<1:12:56,  1.35it/s]
4|580|Loss: 0.7898054718971252:   9%|▉         | 580/6470 [06:36<1:15:49,  1.29it/s]
4|581|Loss: 0.66684490442276:   9%|▉         | 580/6470 [06:36<1:15:49,  1.29it/s]  
4|581|Loss: 0.66684490442276:   9%|▉         | 581/6470 [06:37<1:21:21,  1.21it/s]
4|582|Loss: 0.7517269849777222:   9%|▉         | 581/6470 [06:37<1:21:21,  1.21it/s]
4|582|Loss: 0.7517269849777222:   9%|▉         | 582/6470 [06:37<1:16:59,  1.27it/s]
4|583|Loss: 0.8275296092033386:   9%|▉         | 582/6470 [06:38<1:16:59,  1.27it/s]
4|583|Loss: 0.8275296092033386:   9%|▉         | 583/6470 [06:38<1:12:23,  1.36it/s]
4|584|Loss: 0.8616757988929749:   9%|▉         | 583/6470 [06:38<1:12:23,  1.36it/s]
4|584|Loss: 0.8616757988929749:   9%|▉         | 584/6470 [06:38<1:08:14,  1.44it/s]
4|585|Loss: 0.8785579800605774:   9%|▉         | 584/6470 [06:39<1:08:14,  1.44it/s]
4|585|Loss: 0.8785579800605774:   9%|▉         | 585/6470 [06:39<1:07:48,  1.45it/s]
4|586|Loss: 1.0012898445129395:   9%|▉         | 585/6470 [06:40<1:07:48,  1.45it/s]
4|586|Loss: 1.0012898445129395:   9%|▉         | 586/6470 [06:40<1:08:56,  1.42it/s]
4|587|Loss: 0.7612372040748596:   9%|▉         | 586/6470 [06:40<1:08:56,  1.42it/s]
4|587|Loss: 0.7612372040748596:   9%|▉         | 587/6470 [06:41<1:09:51,  1.40it/s]
4|588|Loss: 0.7652074694633484:   9%|▉         | 587/6470 [06:41<1:09:51,  1.40it/s]
4|588|Loss: 0.7652074694633484:   9%|▉         | 588/6470 [06:41<1:05:04,  1.51it/s]
4|589|Loss: 0.8142117261886597:   9%|▉         | 588/6470 [06:42<1:05:04,  1.51it/s]
4|589|Loss: 0.8142117261886597:   9%|▉         | 589/6470 [06:42<1:03:42,  1.54it/s]
4|590|Loss: 0.7417024374008179:   9%|▉         | 589/6470 [06:42<1:03:42,  1.54it/s]
4|590|Loss: 0.7417024374008179:   9%|▉         | 590/6470 [06:42<1:03:25,  1.55it/s]
4|591|Loss: 0.7267164587974548:   9%|▉         | 590/6470 [06:43<1:03:25,  1.55it/s]
4|591|Loss: 0.7267164587974548:   9%|▉         | 591/6470 [06:43<1:08:41,  1.43it/s]
4|592|Loss: 0.845228910446167:   9%|▉         | 591/6470 [06:44<1:08:41,  1.43it/s] 
4|592|Loss: 0.845228910446167:   9%|▉         | 592/6470 [06:44<1:14:10,  1.32it/s]
4|593|Loss: 0.9111377000808716:   9%|▉         | 592/6470 [06:45<1:14:10,  1.32it/s]
4|593|Loss: 0.9111377000808716:   9%|▉         | 593/6470 [06:45<1:14:54,  1.31it/s]
4|594|Loss: 0.7443769574165344:   9%|▉         | 593/6470 [06:45<1:14:54,  1.31it/s]
4|594|Loss: 0.7443769574165344:   9%|▉         | 594/6470 [06:46<1:20:27,  1.22it/s]
4|595|Loss: 0.8219528794288635:   9%|▉         | 594/6470 [06:46<1:20:27,  1.22it/s]
4|595|Loss: 0.8219528794288635:   9%|▉         | 595/6470 [06:47<1:19:07,  1.24it/s]
4|596|Loss: 0.9309501051902771:   9%|▉         | 595/6470 [06:47<1:19:07,  1.24it/s]
4|596|Loss: 0.9309501051902771:   9%|▉         | 596/6470 [06:47<1:11:56,  1.36it/s]
4|597|Loss: 0.6358407735824585:   9%|▉         | 596/6470 [06:48<1:11:56,  1.36it/s]
4|597|Loss: 0.6358407735824585:   9%|▉         | 597/6470 [06:48<1:13:03,  1.34it/s]
4|598|Loss: 0.8603106141090393:   9%|▉         | 597/6470 [06:48<1:13:03,  1.34it/s]
4|598|Loss: 0.8603106141090393:   9%|▉         | 598/6470 [06:49<1:13:38,  1.33it/s]
4|599|Loss: 0.9102588891983032:   9%|▉         | 598/6470 [06:49<1:13:38,  1.33it/s]
4|599|Loss: 0.9102588891983032:   9%|▉         | 599/6470 [06:49<1:11:17,  1.37it/s]
4|600|Loss: 0.7703083753585815:   9%|▉         | 599/6470 [06:50<1:11:17,  1.37it/s]
4|600|Loss: 0.7703083753585815:   9%|▉         | 600/6470 [06:50<1:15:05,  1.30it/s]
4|601|Loss: 0.8536580204963684:   9%|▉         | 600/6470 [06:51<1:15:05,  1.30it/s]
4|601|Loss: 0.8536580204963684:   9%|▉         | 601/6470 [06:51<1:12:39,  1.35it/s]
4|602|Loss: 0.7892427444458008:   9%|▉         | 601/6470 [06:51<1:12:39,  1.35it/s]
4|602|Loss: 0.7892427444458008:   9%|▉         | 602/6470 [06:52<1:11:05,  1.38it/s]
4|603|Loss: 0.8766369223594666:   9%|▉         | 602/6470 [06:52<1:11:05,  1.38it/s]
4|603|Loss: 0.8766369223594666:   9%|▉         | 603/6470 [06:52<1:11:43,  1.36it/s]
4|604|Loss: 0.8089293837547302:   9%|▉         | 603/6470 [06:53<1:11:43,  1.36it/s]
4|604|Loss: 0.8089293837547302:   9%|▉         | 604/6470 [06:53<1:15:20,  1.30it/s]
4|605|Loss: 0.863814115524292:   9%|▉         | 604/6470 [06:54<1:15:20,  1.30it/s] 
4|605|Loss: 0.863814115524292:   9%|▉         | 605/6470 [06:54<1:15:46,  1.29it/s]
4|606|Loss: 0.8670828938484192:   9%|▉         | 605/6470 [06:54<1:15:46,  1.29it/s]
4|606|Loss: 0.8670828938484192:   9%|▉         | 606/6470 [06:55<1:14:47,  1.31it/s]
4|607|Loss: 0.8630663156509399:   9%|▉         | 606/6470 [06:55<1:14:47,  1.31it/s]
4|607|Loss: 0.8630663156509399:   9%|▉         | 607/6470 [06:56<1:14:00,  1.32it/s]
4|608|Loss: 0.8196888566017151:   9%|▉         | 607/6470 [06:56<1:14:00,  1.32it/s]
4|608|Loss: 0.8196888566017151:   9%|▉         | 608/6470 [06:56<1:10:36,  1.38it/s]
4|609|Loss: 0.7113568186759949:   9%|▉         | 608/6470 [06:57<1:10:36,  1.38it/s]
4|609|Loss: 0.7113568186759949:   9%|▉         | 609/6470 [06:57<1:16:47,  1.27it/s]
4|610|Loss: 0.8095282316207886:   9%|▉         | 609/6470 [06:58<1:16:47,  1.27it/s]
4|610|Loss: 0.8095282316207886:   9%|▉         | 610/6470 [06:58<1:13:32,  1.33it/s]
4|611|Loss: 0.9115529656410217:   9%|▉         | 610/6470 [06:58<1:13:32,  1.33it/s]
4|611|Loss: 0.9115529656410217:   9%|▉         | 611/6470 [06:58<1:09:25,  1.41it/s]
4|612|Loss: 0.9126222133636475:   9%|▉         | 611/6470 [06:59<1:09:25,  1.41it/s]
4|612|Loss: 0.9126222133636475:   9%|▉         | 612/6470 [06:59<1:08:41,  1.42it/s]
4|613|Loss: 0.8105072975158691:   9%|▉         | 612/6470 [06:59<1:08:41,  1.42it/s]
4|613|Loss: 0.8105072975158691:   9%|▉         | 613/6470 [07:00<1:07:17,  1.45it/s]
4|614|Loss: 0.8219324946403503:   9%|▉         | 613/6470 [07:00<1:07:17,  1.45it/s]
4|614|Loss: 0.8219324946403503:   9%|▉         | 614/6470 [07:00<1:07:12,  1.45it/s]
4|615|Loss: 0.958234965801239:   9%|▉         | 614/6470 [07:01<1:07:12,  1.45it/s] 
4|615|Loss: 0.958234965801239:  10%|▉         | 615/6470 [07:01<1:06:10,  1.47it/s]
4|616|Loss: 0.8630852103233337:  10%|▉         | 615/6470 [07:01<1:06:10,  1.47it/s]
4|616|Loss: 0.8630852103233337:  10%|▉         | 616/6470 [07:02<1:07:08,  1.45it/s]
4|617|Loss: 0.9636411666870117:  10%|▉         | 616/6470 [07:02<1:07:08,  1.45it/s]
4|617|Loss: 0.9636411666870117:  10%|▉         | 617/6470 [07:03<1:08:35,  1.42it/s]
4|618|Loss: 0.6547530293464661:  10%|▉         | 617/6470 [07:03<1:08:35,  1.42it/s]
4|618|Loss: 0.6547530293464661:  10%|▉         | 618/6470 [07:03<1:13:02,  1.34it/s]
4|619|Loss: 0.7617889642715454:  10%|▉         | 618/6470 [07:04<1:13:02,  1.34it/s]
4|619|Loss: 0.7617889642715454:  10%|▉         | 619/6470 [07:04<1:13:35,  1.33it/s]
4|620|Loss: 0.873986005783081:  10%|▉         | 619/6470 [07:05<1:13:35,  1.33it/s] 
4|620|Loss: 0.873986005783081:  10%|▉         | 620/6470 [07:05<1:10:20,  1.39it/s]
4|621|Loss: 0.9029675722122192:  10%|▉         | 620/6470 [07:05<1:10:20,  1.39it/s]
4|621|Loss: 0.9029675722122192:  10%|▉         | 621/6470 [07:05<1:07:50,  1.44it/s]
4|622|Loss: 0.7982426881790161:  10%|▉         | 621/6470 [07:06<1:07:50,  1.44it/s]
4|622|Loss: 0.7982426881790161:  10%|▉         | 622/6470 [07:06<1:04:16,  1.52it/s]
4|623|Loss: 0.7578356862068176:  10%|▉         | 622/6470 [07:06<1:04:16,  1.52it/s]
4|623|Loss: 0.7578356862068176:  10%|▉         | 623/6470 [07:07<1:06:54,  1.46it/s]
4|624|Loss: 0.7699041962623596:  10%|▉         | 623/6470 [07:07<1:06:54,  1.46it/s]
4|624|Loss: 0.7699041962623596:  10%|▉         | 624/6470 [07:07<1:06:45,  1.46it/s]
4|625|Loss: 0.8300021290779114:  10%|▉         | 624/6470 [07:08<1:06:45,  1.46it/s]
4|625|Loss: 0.8300021290779114:  10%|▉         | 625/6470 [07:08<1:11:08,  1.37it/s]
4|626|Loss: 0.915056049823761:  10%|▉         | 625/6470 [07:09<1:11:08,  1.37it/s] 
4|626|Loss: 0.915056049823761:  10%|▉         | 626/6470 [07:09<1:11:24,  1.36it/s]
4|627|Loss: 0.8699127435684204:  10%|▉         | 626/6470 [07:09<1:11:24,  1.36it/s]
4|627|Loss: 0.8699127435684204:  10%|▉         | 627/6470 [07:10<1:11:10,  1.37it/s]
4|628|Loss: 1.1378717422485352:  10%|▉         | 627/6470 [07:10<1:11:10,  1.37it/s]
4|628|Loss: 1.1378717422485352:  10%|▉         | 628/6470 [07:10<1:08:37,  1.42it/s]
4|629|Loss: 0.8057413697242737:  10%|▉         | 628/6470 [07:11<1:08:37,  1.42it/s]
4|629|Loss: 0.8057413697242737:  10%|▉         | 629/6470 [07:11<1:08:06,  1.43it/s]
4|630|Loss: 0.9431830644607544:  10%|▉         | 629/6470 [07:11<1:08:06,  1.43it/s]
4|630|Loss: 0.9431830644607544:  10%|▉         | 630/6470 [07:12<1:06:40,  1.46it/s]
4|631|Loss: 0.8064647912979126:  10%|▉         | 630/6470 [07:12<1:06:40,  1.46it/s]
4|631|Loss: 0.8064647912979126:  10%|▉         | 631/6470 [07:12<1:03:08,  1.54it/s]
4|632|Loss: 0.9229406714439392:  10%|▉         | 631/6470 [07:13<1:03:08,  1.54it/s]
4|632|Loss: 0.9229406714439392:  10%|▉         | 632/6470 [07:13<1:05:50,  1.48it/s]
4|633|Loss: 0.946967363357544:  10%|▉         | 632/6470 [07:13<1:05:50,  1.48it/s] 
4|633|Loss: 0.946967363357544:  10%|▉         | 633/6470 [07:14<1:07:11,  1.45it/s]
4|634|Loss: 1.0416147708892822:  10%|▉         | 633/6470 [07:14<1:07:11,  1.45it/s]
4|634|Loss: 1.0416147708892822:  10%|▉         | 634/6470 [07:14<1:03:18,  1.54it/s]
4|635|Loss: 0.7097258567810059:  10%|▉         | 634/6470 [07:15<1:03:18,  1.54it/s]
4|635|Loss: 0.7097258567810059:  10%|▉         | 635/6470 [07:15<1:01:48,  1.57it/s]
4|636|Loss: 0.8732407093048096:  10%|▉         | 635/6470 [07:15<1:01:48,  1.57it/s]
4|636|Loss: 0.8732407093048096:  10%|▉         | 636/6470 [07:16<1:04:14,  1.51it/s]
4|637|Loss: 0.9596625566482544:  10%|▉         | 636/6470 [07:16<1:04:14,  1.51it/s]
4|637|Loss: 0.9596625566482544:  10%|▉         | 637/6470 [07:16<1:06:47,  1.46it/s]
4|638|Loss: 0.9304170608520508:  10%|▉         | 637/6470 [07:17<1:06:47,  1.46it/s]
4|638|Loss: 0.9304170608520508:  10%|▉         | 638/6470 [07:17<1:09:46,  1.39it/s]
4|639|Loss: 0.9621264934539795:  10%|▉         | 638/6470 [07:18<1:09:46,  1.39it/s]
4|639|Loss: 0.9621264934539795:  10%|▉         | 639/6470 [07:18<1:10:53,  1.37it/s]
4|640|Loss: 0.8835097551345825:  10%|▉         | 639/6470 [07:18<1:10:53,  1.37it/s]
4|640|Loss: 0.8835097551345825:  10%|▉         | 640/6470 [07:19<1:11:39,  1.36it/s]
4|641|Loss: 0.8808602690696716:  10%|▉         | 640/6470 [07:19<1:11:39,  1.36it/s]
4|641|Loss: 0.8808602690696716:  10%|▉         | 641/6470 [07:19<1:10:58,  1.37it/s]
4|642|Loss: 0.657883882522583:  10%|▉         | 641/6470 [07:20<1:10:58,  1.37it/s] 
4|642|Loss: 0.657883882522583:  10%|▉         | 642/6470 [07:20<1:09:59,  1.39it/s]
4|643|Loss: 0.8326571583747864:  10%|▉         | 642/6470 [07:21<1:09:59,  1.39it/s]
4|643|Loss: 0.8326571583747864:  10%|▉         | 643/6470 [07:21<1:12:54,  1.33it/s]
4|644|Loss: 0.8035922646522522:  10%|▉         | 643/6470 [07:21<1:12:54,  1.33it/s]
4|644|Loss: 0.8035922646522522:  10%|▉         | 644/6470 [07:22<1:18:34,  1.24it/s]
4|645|Loss: 0.8075597286224365:  10%|▉         | 644/6470 [07:22<1:18:34,  1.24it/s]
4|645|Loss: 0.8075597286224365:  10%|▉         | 645/6470 [07:22<1:13:02,  1.33it/s]
4|646|Loss: 0.8152990937232971:  10%|▉         | 645/6470 [07:23<1:13:02,  1.33it/s]
4|646|Loss: 0.8152990937232971:  10%|▉         | 646/6470 [07:23<1:07:42,  1.43it/s]
4|647|Loss: 0.9108411073684692:  10%|▉         | 646/6470 [07:23<1:07:42,  1.43it/s]
4|647|Loss: 0.9108411073684692:  10%|█         | 647/6470 [07:24<1:05:33,  1.48it/s]
4|648|Loss: 0.7918006777763367:  10%|█         | 647/6470 [07:24<1:05:33,  1.48it/s]
4|648|Loss: 0.7918006777763367:  10%|█         | 648/6470 [07:24<1:06:41,  1.46it/s]
4|649|Loss: 0.8349222540855408:  10%|█         | 648/6470 [07:25<1:06:41,  1.46it/s]
4|649|Loss: 0.8349222540855408:  10%|█         | 649/6470 [07:25<1:07:59,  1.43it/s]
4|650|Loss: 0.7909739017486572:  10%|█         | 649/6470 [07:26<1:07:59,  1.43it/s]
4|650|Loss: 0.7909739017486572:  10%|█         | 650/6470 [07:26<1:10:37,  1.37it/s]
4|651|Loss: 0.8896297812461853:  10%|█         | 650/6470 [07:26<1:10:37,  1.37it/s]
4|651|Loss: 0.8896297812461853:  10%|█         | 651/6470 [07:27<1:08:59,  1.41it/s]
4|652|Loss: 0.8200868964195251:  10%|█         | 651/6470 [07:27<1:08:59,  1.41it/s]
4|652|Loss: 0.8200868964195251:  10%|█         | 652/6470 [07:27<1:08:16,  1.42it/s]
4|653|Loss: 0.8959755897521973:  10%|█         | 652/6470 [07:28<1:08:16,  1.42it/s]
4|653|Loss: 0.8959755897521973:  10%|█         | 653/6470 [07:28<1:12:26,  1.34it/s]
4|654|Loss: 0.9065054655075073:  10%|█         | 653/6470 [07:29<1:12:26,  1.34it/s]
4|654|Loss: 0.9065054655075073:  10%|█         | 654/6470 [07:29<1:13:23,  1.32it/s]
4|655|Loss: 0.6913820505142212:  10%|█         | 654/6470 [07:29<1:13:23,  1.32it/s]
4|655|Loss: 0.6913820505142212:  10%|█         | 655/6470 [07:29<1:08:00,  1.43it/s]
4|656|Loss: 1.0312751531600952:  10%|█         | 655/6470 [07:30<1:08:00,  1.43it/s]
4|656|Loss: 1.0312751531600952:  10%|█         | 656/6470 [07:30<1:05:02,  1.49it/s]
4|657|Loss: 0.8354698419570923:  10%|█         | 656/6470 [07:31<1:05:02,  1.49it/s]
4|657|Loss: 0.8354698419570923:  10%|█         | 657/6470 [07:31<1:09:44,  1.39it/s]
4|658|Loss: 0.8844390511512756:  10%|█         | 657/6470 [07:31<1:09:44,  1.39it/s]
4|658|Loss: 0.8844390511512756:  10%|█         | 658/6470 [07:32<1:06:24,  1.46it/s]
4|659|Loss: 0.7315028309822083:  10%|█         | 658/6470 [07:32<1:06:24,  1.46it/s]
4|659|Loss: 0.7315028309822083:  10%|█         | 659/6470 [07:32<1:05:41,  1.47it/s]
4|660|Loss: 0.9870471358299255:  10%|█         | 659/6470 [07:33<1:05:41,  1.47it/s]
4|660|Loss: 0.9870471358299255:  10%|█         | 660/6470 [07:33<1:10:03,  1.38it/s]
4|661|Loss: 0.7347702980041504:  10%|█         | 660/6470 [07:33<1:10:03,  1.38it/s]
4|661|Loss: 0.7347702980041504:  10%|█         | 661/6470 [07:34<1:11:18,  1.36it/s]
4|662|Loss: 0.7127193212509155:  10%|█         | 661/6470 [07:34<1:11:18,  1.36it/s]
4|662|Loss: 0.7127193212509155:  10%|█         | 662/6470 [07:34<1:10:20,  1.38it/s]
4|663|Loss: 0.8484251499176025:  10%|█         | 662/6470 [07:35<1:10:20,  1.38it/s]
4|663|Loss: 0.8484251499176025:  10%|█         | 663/6470 [07:35<1:09:28,  1.39it/s]
4|664|Loss: 0.8661319017410278:  10%|█         | 663/6470 [07:36<1:09:28,  1.39it/s]
4|664|Loss: 0.8661319017410278:  10%|█         | 664/6470 [07:36<1:09:03,  1.40it/s]
4|665|Loss: 1.0155340433120728:  10%|█         | 664/6470 [07:36<1:09:03,  1.40it/s]
4|665|Loss: 1.0155340433120728:  10%|█         | 665/6470 [07:36<1:03:37,  1.52it/s]
4|666|Loss: 0.8564288020133972:  10%|█         | 665/6470 [07:37<1:03:37,  1.52it/s]
4|666|Loss: 0.8564288020133972:  10%|█         | 666/6470 [07:37<1:00:33,  1.60it/s]
4|667|Loss: 1.0609711408615112:  10%|█         | 666/6470 [07:37<1:00:33,  1.60it/s]
4|667|Loss: 1.0609711408615112:  10%|█         | 667/6470 [07:38<1:02:04,  1.56it/s]
4|668|Loss: 0.8460641503334045:  10%|█         | 667/6470 [07:38<1:02:04,  1.56it/s]
4|668|Loss: 0.8460641503334045:  10%|█         | 668/6470 [07:38<58:03,  1.67it/s]  
4|669|Loss: 0.7796790599822998:  10%|█         | 668/6470 [07:38<58:03,  1.67it/s]
4|669|Loss: 0.7796790599822998:  10%|█         | 669/6470 [07:39<59:28,  1.63it/s]
4|670|Loss: 0.7948797345161438:  10%|█         | 669/6470 [07:39<59:28,  1.63it/s]
4|670|Loss: 0.7948797345161438:  10%|█         | 670/6470 [07:39<1:00:36,  1.60it/s]
4|671|Loss: 0.8851611614227295:  10%|█         | 670/6470 [07:40<1:00:36,  1.60it/s]
4|671|Loss: 0.8851611614227295:  10%|█         | 671/6470 [07:40<1:03:06,  1.53it/s]
4|672|Loss: 0.716272234916687:  10%|█         | 671/6470 [07:41<1:03:06,  1.53it/s] 
4|672|Loss: 0.716272234916687:  10%|█         | 672/6470 [07:41<1:00:16,  1.60it/s]
4|673|Loss: 0.8161384463310242:  10%|█         | 672/6470 [07:41<1:00:16,  1.60it/s]
4|673|Loss: 0.8161384463310242:  10%|█         | 673/6470 [07:41<56:59,  1.70it/s]  
4|674|Loss: 0.8834587931632996:  10%|█         | 673/6470 [07:42<56:59,  1.70it/s]
4|674|Loss: 0.8834587931632996:  10%|█         | 674/6470 [07:42<57:15,  1.69it/s]
4|675|Loss: 0.9336728453636169:  10%|█         | 674/6470 [07:42<57:15,  1.69it/s]
4|675|Loss: 0.9336728453636169:  10%|█         | 675/6470 [07:42<58:39,  1.65it/s]
4|676|Loss: 0.8723056316375732:  10%|█         | 675/6470 [07:43<58:39,  1.65it/s]
4|676|Loss: 0.8723056316375732:  10%|█         | 676/6470 [07:43<58:08,  1.66it/s]
4|677|Loss: 0.9132482409477234:  10%|█         | 676/6470 [07:43<58:08,  1.66it/s]
4|677|Loss: 0.9132482409477234:  10%|█         | 677/6470 [07:44<1:01:06,  1.58it/s]
4|678|Loss: 0.8392953276634216:  10%|█         | 677/6470 [07:44<1:01:06,  1.58it/s]
4|678|Loss: 0.8392953276634216:  10%|█         | 678/6470 [07:45<1:04:19,  1.50it/s]
4|679|Loss: 0.8733817934989929:  10%|█         | 678/6470 [07:45<1:04:19,  1.50it/s]
4|679|Loss: 0.8733817934989929:  10%|█         | 679/6470 [07:45<1:05:42,  1.47it/s]
4|680|Loss: 0.8893956542015076:  10%|█         | 679/6470 [07:46<1:05:42,  1.47it/s]
4|680|Loss: 0.8893956542015076:  11%|█         | 680/6470 [07:46<1:07:28,  1.43it/s]
4|681|Loss: 0.8383374214172363:  11%|█         | 680/6470 [07:46<1:07:28,  1.43it/s]
4|681|Loss: 0.8383374214172363:  11%|█         | 681/6470 [07:47<1:10:01,  1.38it/s]
4|682|Loss: 0.946418821811676:  11%|█         | 681/6470 [07:47<1:10:01,  1.38it/s] 
4|682|Loss: 0.946418821811676:  11%|█         | 682/6470 [07:47<1:04:07,  1.50it/s]
4|683|Loss: 0.8507554531097412:  11%|█         | 682/6470 [07:48<1:04:07,  1.50it/s]
4|683|Loss: 0.8507554531097412:  11%|█         | 683/6470 [07:48<1:07:04,  1.44it/s]
4|684|Loss: 1.0339716672897339:  11%|█         | 683/6470 [07:48<1:07:04,  1.44it/s]
4|684|Loss: 1.0339716672897339:  11%|█         | 684/6470 [07:49<1:02:40,  1.54it/s]
4|685|Loss: 0.8010479807853699:  11%|█         | 684/6470 [07:49<1:02:40,  1.54it/s]
4|685|Loss: 0.8010479807853699:  11%|█         | 685/6470 [07:49<1:03:19,  1.52it/s]
4|686|Loss: 0.7566506862640381:  11%|█         | 685/6470 [07:50<1:03:19,  1.52it/s]
4|686|Loss: 0.7566506862640381:  11%|█         | 686/6470 [07:50<1:00:20,  1.60it/s]
4|687|Loss: 1.0070658922195435:  11%|█         | 686/6470 [07:50<1:00:20,  1.60it/s]
4|687|Loss: 1.0070658922195435:  11%|█         | 687/6470 [07:50<57:27,  1.68it/s]  
4|688|Loss: 0.8856123685836792:  11%|█         | 687/6470 [07:51<57:27,  1.68it/s]
4|688|Loss: 0.8856123685836792:  11%|█         | 688/6470 [07:51<59:23,  1.62it/s]
4|689|Loss: 0.8293054103851318:  11%|█         | 688/6470 [07:51<59:23,  1.62it/s]
4|689|Loss: 0.8293054103851318:  11%|█         | 689/6470 [07:52<1:03:07,  1.53it/s]
4|690|Loss: 1.030730128288269:  11%|█         | 689/6470 [07:52<1:03:07,  1.53it/s] 
4|690|Loss: 1.030730128288269:  11%|█         | 690/6470 [07:52<1:01:54,  1.56it/s]
4|691|Loss: 0.7245435118675232:  11%|█         | 690/6470 [07:53<1:01:54,  1.56it/s]
4|691|Loss: 0.7245435118675232:  11%|█         | 691/6470 [07:53<1:03:44,  1.51it/s]
4|692|Loss: 0.8280851244926453:  11%|█         | 691/6470 [07:53<1:03:44,  1.51it/s]
4|692|Loss: 0.8280851244926453:  11%|█         | 692/6470 [07:54<1:01:12,  1.57it/s]
4|693|Loss: 0.9207046627998352:  11%|█         | 692/6470 [07:54<1:01:12,  1.57it/s]
4|693|Loss: 0.9207046627998352:  11%|█         | 693/6470 [07:54<1:02:13,  1.55it/s]
4|694|Loss: 1.054381012916565:  11%|█         | 693/6470 [07:55<1:02:13,  1.55it/s] 
4|694|Loss: 1.054381012916565:  11%|█         | 694/6470 [07:55<1:04:32,  1.49it/s]
4|695|Loss: 0.8664572834968567:  11%|█         | 694/6470 [07:55<1:04:32,  1.49it/s]
4|695|Loss: 0.8664572834968567:  11%|█         | 695/6470 [07:56<1:06:18,  1.45it/s]
4|696|Loss: 0.7885241508483887:  11%|█         | 695/6470 [07:56<1:06:18,  1.45it/s]
4|696|Loss: 0.7885241508483887:  11%|█         | 696/6470 [07:57<1:07:25,  1.43it/s]
4|697|Loss: 0.8571537137031555:  11%|█         | 696/6470 [07:57<1:07:25,  1.43it/s]
4|697|Loss: 0.8571537137031555:  11%|█         | 697/6470 [07:57<1:10:24,  1.37it/s]
4|698|Loss: 0.8468345403671265:  11%|█         | 697/6470 [07:58<1:10:24,  1.37it/s]
4|698|Loss: 0.8468345403671265:  11%|█         | 698/6470 [07:58<1:13:56,  1.30it/s]
4|699|Loss: 0.8626419901847839:  11%|█         | 698/6470 [07:59<1:13:56,  1.30it/s]
4|699|Loss: 0.8626419901847839:  11%|█         | 699/6470 [07:59<1:13:00,  1.32it/s]
4|700|Loss: 0.9144642353057861:  11%|█         | 699/6470 [07:59<1:13:00,  1.32it/s]
4|700|Loss: 0.9144642353057861:  11%|█         | 700/6470 [07:59<1:08:17,  1.41it/s]
4|701|Loss: 0.5836280584335327:  11%|█         | 700/6470 [08:00<1:08:17,  1.41it/s]
4|701|Loss: 0.5836280584335327:  11%|█         | 701/6470 [08:00<1:07:53,  1.42it/s]
4|702|Loss: 0.8990941643714905:  11%|█         | 701/6470 [08:01<1:07:53,  1.42it/s]
4|702|Loss: 0.8990941643714905:  11%|█         | 702/6470 [08:01<1:06:02,  1.46it/s]
4|703|Loss: 0.9304386377334595:  11%|█         | 702/6470 [08:01<1:06:02,  1.46it/s]
4|703|Loss: 0.9304386377334595:  11%|█         | 703/6470 [08:02<1:08:59,  1.39it/s]
4|704|Loss: 0.7156645655632019:  11%|█         | 703/6470 [08:02<1:08:59,  1.39it/s]
4|704|Loss: 0.7156645655632019:  11%|█         | 704/6470 [08:02<1:11:23,  1.35it/s]
4|705|Loss: 0.7411702275276184:  11%|█         | 704/6470 [08:03<1:11:23,  1.35it/s]
4|705|Loss: 0.7411702275276184:  11%|█         | 705/6470 [08:03<1:11:33,  1.34it/s]
4|706|Loss: 0.8649008870124817:  11%|█         | 705/6470 [08:04<1:11:33,  1.34it/s]
4|706|Loss: 0.8649008870124817:  11%|█         | 706/6470 [08:04<1:10:39,  1.36it/s]
4|707|Loss: 0.9338008165359497:  11%|█         | 706/6470 [08:04<1:10:39,  1.36it/s]
4|707|Loss: 0.9338008165359497:  11%|█         | 707/6470 [08:05<1:09:35,  1.38it/s]
4|708|Loss: 0.8548884391784668:  11%|█         | 707/6470 [08:05<1:09:35,  1.38it/s]
4|708|Loss: 0.8548884391784668:  11%|█         | 708/6470 [08:05<1:09:34,  1.38it/s]
4|709|Loss: 0.8344454765319824:  11%|█         | 708/6470 [08:06<1:09:34,  1.38it/s]
4|709|Loss: 0.8344454765319824:  11%|█         | 709/6470 [08:06<1:07:13,  1.43it/s]
4|710|Loss: 0.8105350732803345:  11%|█         | 709/6470 [08:06<1:07:13,  1.43it/s]
4|710|Loss: 0.8105350732803345:  11%|█         | 710/6470 [08:07<1:10:20,  1.36it/s]
4|711|Loss: 0.9066046476364136:  11%|█         | 710/6470 [08:07<1:10:20,  1.36it/s]
4|711|Loss: 0.9066046476364136:  11%|█         | 711/6470 [08:07<1:08:43,  1.40it/s]
4|712|Loss: 0.8190558552742004:  11%|█         | 711/6470 [08:08<1:08:43,  1.40it/s]
4|712|Loss: 0.8190558552742004:  11%|█         | 712/6470 [08:08<1:08:34,  1.40it/s]
4|713|Loss: 0.8399686813354492:  11%|█         | 712/6470 [08:09<1:08:34,  1.40it/s]
4|713|Loss: 0.8399686813354492:  11%|█         | 713/6470 [08:09<1:04:04,  1.50it/s]
4|714|Loss: 0.9632976055145264:  11%|█         | 713/6470 [08:09<1:04:04,  1.50it/s]
4|714|Loss: 0.9632976055145264:  11%|█         | 714/6470 [08:10<1:07:37,  1.42it/s]
4|715|Loss: 0.8042215704917908:  11%|█         | 714/6470 [08:10<1:07:37,  1.42it/s]
4|715|Loss: 0.8042215704917908:  11%|█         | 715/6470 [08:10<1:06:39,  1.44it/s]
4|716|Loss: 0.8195074796676636:  11%|█         | 715/6470 [08:11<1:06:39,  1.44it/s]
4|716|Loss: 0.8195074796676636:  11%|█         | 716/6470 [08:11<1:06:24,  1.44it/s]
4|717|Loss: 0.8401109576225281:  11%|█         | 716/6470 [08:11<1:06:24,  1.44it/s]
4|717|Loss: 0.8401109576225281:  11%|█         | 717/6470 [08:12<1:05:18,  1.47it/s]
4|718|Loss: 0.904136598110199:  11%|█         | 717/6470 [08:12<1:05:18,  1.47it/s] 
4|718|Loss: 0.904136598110199:  11%|█         | 718/6470 [08:12<1:06:36,  1.44it/s]
4|719|Loss: 0.812764048576355:  11%|█         | 718/6470 [08:13<1:06:36,  1.44it/s]
4|719|Loss: 0.812764048576355:  11%|█         | 719/6470 [08:13<1:03:06,  1.52it/s]
4|720|Loss: 0.8296414017677307:  11%|█         | 719/6470 [08:13<1:03:06,  1.52it/s]
4|720|Loss: 0.8296414017677307:  11%|█         | 720/6470 [08:13<1:01:50,  1.55it/s]
4|721|Loss: 0.8984166979789734:  11%|█         | 720/6470 [08:14<1:01:50,  1.55it/s]
4|721|Loss: 0.8984166979789734:  11%|█         | 721/6470 [08:14<1:01:10,  1.57it/s]
4|722|Loss: 0.7366175055503845:  11%|█         | 721/6470 [08:14<1:01:10,  1.57it/s]
4|722|Loss: 0.7366175055503845:  11%|█         | 722/6470 [08:15<1:01:06,  1.57it/s]
4|723|Loss: 0.785010039806366:  11%|█         | 722/6470 [08:15<1:01:06,  1.57it/s] 
4|723|Loss: 0.785010039806366:  11%|█         | 723/6470 [08:15<1:00:50,  1.57it/s]
4|724|Loss: 0.567616879940033:  11%|█         | 723/6470 [08:16<1:00:50,  1.57it/s]
4|724|Loss: 0.567616879940033:  11%|█         | 724/6470 [08:16<1:02:51,  1.52it/s]
4|725|Loss: 0.8696458339691162:  11%|█         | 724/6470 [08:16<1:02:51,  1.52it/s]
4|725|Loss: 0.8696458339691162:  11%|█         | 725/6470 [08:17<1:02:54,  1.52it/s]
4|726|Loss: 0.9095626473426819:  11%|█         | 725/6470 [08:17<1:02:54,  1.52it/s]
4|726|Loss: 0.9095626473426819:  11%|█         | 726/6470 [08:17<1:05:52,  1.45it/s]
4|727|Loss: 0.7837868928909302:  11%|█         | 726/6470 [08:18<1:05:52,  1.45it/s]
4|727|Loss: 0.7837868928909302:  11%|█         | 727/6470 [08:18<1:03:52,  1.50it/s]
4|728|Loss: 0.8395044207572937:  11%|█         | 727/6470 [08:18<1:03:52,  1.50it/s]
4|728|Loss: 0.8395044207572937:  11%|█▏        | 728/6470 [08:19<1:02:17,  1.54it/s]
4|729|Loss: 0.8370721340179443:  11%|█▏        | 728/6470 [08:19<1:02:17,  1.54it/s]
4|729|Loss: 0.8370721340179443:  11%|█▏        | 729/6470 [08:19<1:01:23,  1.56it/s]
4|730|Loss: 0.8230490684509277:  11%|█▏        | 729/6470 [08:20<1:01:23,  1.56it/s]
4|730|Loss: 0.8230490684509277:  11%|█▏        | 730/6470 [08:20<58:12,  1.64it/s]  
4|731|Loss: 0.9389750361442566:  11%|█▏        | 730/6470 [08:20<58:12,  1.64it/s]
4|731|Loss: 0.9389750361442566:  11%|█▏        | 731/6470 [08:20<57:10,  1.67it/s]
4|732|Loss: 0.9013743996620178:  11%|█▏        | 731/6470 [08:21<57:10,  1.67it/s]
4|732|Loss: 0.9013743996620178:  11%|█▏        | 732/6470 [08:21<1:02:22,  1.53it/s]
4|733|Loss: 0.8996532559394836:  11%|█▏        | 732/6470 [08:22<1:02:22,  1.53it/s]
4|733|Loss: 0.8996532559394836:  11%|█▏        | 733/6470 [08:22<1:00:10,  1.59it/s]
4|734|Loss: 0.9112755060195923:  11%|█▏        | 733/6470 [08:22<1:00:10,  1.59it/s]
4|734|Loss: 0.9112755060195923:  11%|█▏        | 734/6470 [08:22<59:21,  1.61it/s]  
4|735|Loss: 0.927734375:  11%|█▏        | 734/6470 [08:23<59:21,  1.61it/s]       
4|735|Loss: 0.927734375:  11%|█▏        | 735/6470 [08:23<59:19,  1.61it/s]
4|736|Loss: 0.7718807458877563:  11%|█▏        | 735/6470 [08:23<59:19,  1.61it/s]
4|736|Loss: 0.7718807458877563:  11%|█▏        | 736/6470 [08:24<58:12,  1.64it/s]
4|737|Loss: 0.7887811660766602:  11%|█▏        | 736/6470 [08:24<58:12,  1.64it/s]
4|737|Loss: 0.7887811660766602:  11%|█▏        | 737/6470 [08:24<1:01:10,  1.56it/s]
4|738|Loss: 0.6851943731307983:  11%|█▏        | 737/6470 [08:25<1:01:10,  1.56it/s]
4|738|Loss: 0.6851943731307983:  11%|█▏        | 738/6470 [08:25<1:00:24,  1.58it/s]
4|739|Loss: 0.934954047203064:  11%|█▏        | 738/6470 [08:25<1:00:24,  1.58it/s] 
4|739|Loss: 0.934954047203064:  11%|█▏        | 739/6470 [08:26<1:00:21,  1.58it/s]
4|740|Loss: 0.7741696834564209:  11%|█▏        | 739/6470 [08:26<1:00:21,  1.58it/s]
4|740|Loss: 0.7741696834564209:  11%|█▏        | 740/6470 [08:26<1:00:40,  1.57it/s]
4|741|Loss: 0.8366391658782959:  11%|█▏        | 740/6470 [08:27<1:00:40,  1.57it/s]
4|741|Loss: 0.8366391658782959:  11%|█▏        | 741/6470 [08:27<1:03:20,  1.51it/s]
4|742|Loss: 0.9918961524963379:  11%|█▏        | 741/6470 [08:27<1:03:20,  1.51it/s]
4|742|Loss: 0.9918961524963379:  11%|█▏        | 742/6470 [08:28<1:07:37,  1.41it/s]
4|743|Loss: 1.0059069395065308:  11%|█▏        | 742/6470 [08:28<1:07:37,  1.41it/s]
4|743|Loss: 1.0059069395065308:  11%|█▏        | 743/6470 [08:28<1:02:24,  1.53it/s]
4|744|Loss: 0.9012850522994995:  11%|█▏        | 743/6470 [08:29<1:02:24,  1.53it/s]
4|744|Loss: 0.9012850522994995:  11%|█▏        | 744/6470 [08:29<1:02:21,  1.53it/s]
4|745|Loss: 0.7589672803878784:  11%|█▏        | 744/6470 [08:29<1:02:21,  1.53it/s]
4|745|Loss: 0.7589672803878784:  12%|█▏        | 745/6470 [08:29<59:50,  1.59it/s]  
4|746|Loss: 0.8690328598022461:  12%|█▏        | 745/6470 [08:30<59:50,  1.59it/s]
4|746|Loss: 0.8690328598022461:  12%|█▏        | 746/6470 [08:30<1:03:51,  1.49it/s]
4|747|Loss: 1.056978464126587:  12%|█▏        | 746/6470 [08:31<1:03:51,  1.49it/s] 
4|747|Loss: 1.056978464126587:  12%|█▏        | 747/6470 [08:31<1:08:58,  1.38it/s]
4|748|Loss: 0.8341920375823975:  12%|█▏        | 747/6470 [08:31<1:08:58,  1.38it/s]
4|748|Loss: 0.8341920375823975:  12%|█▏        | 748/6470 [08:32<1:05:01,  1.47it/s]
4|749|Loss: 1.0095751285552979:  12%|█▏        | 748/6470 [08:32<1:05:01,  1.47it/s]
4|749|Loss: 1.0095751285552979:  12%|█▏        | 749/6470 [08:32<1:00:02,  1.59it/s]
4|750|Loss: 0.777266263961792:  12%|█▏        | 749/6470 [08:32<1:00:02,  1.59it/s] 
4|750|Loss: 0.777266263961792:  12%|█▏        | 750/6470 [08:33<59:34,  1.60it/s]  
4|751|Loss: 0.8704924583435059:  12%|█▏        | 750/6470 [08:33<59:34,  1.60it/s]
4|751|Loss: 0.8704924583435059:  12%|█▏        | 751/6470 [08:34<1:02:47,  1.52it/s]
4|752|Loss: 0.9889553785324097:  12%|█▏        | 751/6470 [08:34<1:02:47,  1.52it/s]
4|752|Loss: 0.9889553785324097:  12%|█▏        | 752/6470 [08:34<59:31,  1.60it/s]  
4|753|Loss: 0.7768442034721375:  12%|█▏        | 752/6470 [08:34<59:31,  1.60it/s]
4|753|Loss: 0.7768442034721375:  12%|█▏        | 753/6470 [08:35<1:00:30,  1.57it/s]
4|754|Loss: 0.9181543588638306:  12%|█▏        | 753/6470 [08:35<1:00:30,  1.57it/s]
4|754|Loss: 0.9181543588638306:  12%|█▏        | 754/6470 [08:36<1:06:31,  1.43it/s]
4|755|Loss: 0.8724762201309204:  12%|█▏        | 754/6470 [08:36<1:06:31,  1.43it/s]
4|755|Loss: 0.8724762201309204:  12%|█▏        | 755/6470 [08:36<1:06:35,  1.43it/s]
4|756|Loss: 0.897565484046936:  12%|█▏        | 755/6470 [08:37<1:06:35,  1.43it/s] 
4|756|Loss: 0.897565484046936:  12%|█▏        | 756/6470 [08:37<1:09:19,  1.37it/s]
4|757|Loss: 0.8292321562767029:  12%|█▏        | 756/6470 [08:37<1:09:19,  1.37it/s]
4|757|Loss: 0.8292321562767029:  12%|█▏        | 757/6470 [08:38<1:09:25,  1.37it/s]
4|758|Loss: 0.9212679862976074:  12%|█▏        | 757/6470 [08:38<1:09:25,  1.37it/s]
4|758|Loss: 0.9212679862976074:  12%|█▏        | 758/6470 [08:39<1:09:53,  1.36it/s]
4|759|Loss: 0.7422360777854919:  12%|█▏        | 758/6470 [08:39<1:09:53,  1.36it/s]
4|759|Loss: 0.7422360777854919:  12%|█▏        | 759/6470 [08:39<1:09:27,  1.37it/s]
4|760|Loss: 0.8727343678474426:  12%|█▏        | 759/6470 [08:40<1:09:27,  1.37it/s]
4|760|Loss: 0.8727343678474426:  12%|█▏        | 760/6470 [08:40<1:11:36,  1.33it/s]
4|761|Loss: 0.7663893699645996:  12%|█▏        | 760/6470 [08:41<1:11:36,  1.33it/s]
4|761|Loss: 0.7663893699645996:  12%|█▏        | 761/6470 [08:41<1:14:07,  1.28it/s]
4|762|Loss: 0.9755934476852417:  12%|█▏        | 761/6470 [08:41<1:14:07,  1.28it/s]
4|762|Loss: 0.9755934476852417:  12%|█▏        | 762/6470 [08:42<1:12:49,  1.31it/s]
4|763|Loss: 0.9055172204971313:  12%|█▏        | 762/6470 [08:42<1:12:49,  1.31it/s]
4|763|Loss: 0.9055172204971313:  12%|█▏        | 763/6470 [08:42<1:09:30,  1.37it/s]
4|764|Loss: 0.9122471213340759:  12%|█▏        | 763/6470 [08:43<1:09:30,  1.37it/s]
4|764|Loss: 0.9122471213340759:  12%|█▏        | 764/6470 [08:43<1:09:05,  1.38it/s]
4|765|Loss: 0.8867809772491455:  12%|█▏        | 764/6470 [08:43<1:09:05,  1.38it/s]
4|765|Loss: 0.8867809772491455:  12%|█▏        | 765/6470 [08:44<1:10:35,  1.35it/s]
4|766|Loss: 0.8909686803817749:  12%|█▏        | 765/6470 [08:44<1:10:35,  1.35it/s]
4|766|Loss: 0.8909686803817749:  12%|█▏        | 766/6470 [08:44<1:08:46,  1.38it/s]
4|767|Loss: 0.8558785319328308:  12%|█▏        | 766/6470 [08:45<1:08:46,  1.38it/s]
4|767|Loss: 0.8558785319328308:  12%|█▏        | 767/6470 [08:45<1:02:05,  1.53it/s]
4|768|Loss: 0.7754786610603333:  12%|█▏        | 767/6470 [08:45<1:02:05,  1.53it/s]
4|768|Loss: 0.7754786610603333:  12%|█▏        | 768/6470 [08:46<1:04:46,  1.47it/s]
4|769|Loss: 0.8090490102767944:  12%|█▏        | 768/6470 [08:46<1:04:46,  1.47it/s]
4|769|Loss: 0.8090490102767944:  12%|█▏        | 769/6470 [08:46<1:06:49,  1.42it/s]
4|770|Loss: 0.960839033126831:  12%|█▏        | 769/6470 [08:47<1:06:49,  1.42it/s] 
4|770|Loss: 0.960839033126831:  12%|█▏        | 770/6470 [08:47<1:10:32,  1.35it/s]
4|771|Loss: 0.9215344786643982:  12%|█▏        | 770/6470 [08:48<1:10:32,  1.35it/s]
4|771|Loss: 0.9215344786643982:  12%|█▏        | 771/6470 [08:48<1:10:14,  1.35it/s]
4|772|Loss: 0.8767651319503784:  12%|█▏        | 771/6470 [08:48<1:10:14,  1.35it/s]
4|772|Loss: 0.8767651319503784:  12%|█▏        | 772/6470 [08:49<1:11:31,  1.33it/s]
4|773|Loss: 1.0227510929107666:  12%|█▏        | 772/6470 [08:49<1:11:31,  1.33it/s]
4|773|Loss: 1.0227510929107666:  12%|█▏        | 773/6470 [08:50<1:14:01,  1.28it/s]
4|774|Loss: 0.7040738463401794:  12%|█▏        | 773/6470 [08:50<1:14:01,  1.28it/s]
4|774|Loss: 0.7040738463401794:  12%|█▏        | 774/6470 [08:50<1:10:47,  1.34it/s]
4|775|Loss: 0.7237310409545898:  12%|█▏        | 774/6470 [08:51<1:10:47,  1.34it/s]
4|775|Loss: 0.7237310409545898:  12%|█▏        | 775/6470 [08:51<1:12:23,  1.31it/s]
4|776|Loss: 0.8127559423446655:  12%|█▏        | 775/6470 [08:52<1:12:23,  1.31it/s]
4|776|Loss: 0.8127559423446655:  12%|█▏        | 776/6470 [08:52<1:09:41,  1.36it/s]
4|777|Loss: 0.9070510864257812:  12%|█▏        | 776/6470 [08:52<1:09:41,  1.36it/s]
4|777|Loss: 0.9070510864257812:  12%|█▏        | 777/6470 [08:53<1:09:28,  1.37it/s]
4|778|Loss: 0.980015218257904:  12%|█▏        | 777/6470 [08:53<1:09:28,  1.37it/s] 
4|778|Loss: 0.980015218257904:  12%|█▏        | 778/6470 [08:53<1:11:02,  1.34it/s]
4|779|Loss: 1.0239126682281494:  12%|█▏        | 778/6470 [08:54<1:11:02,  1.34it/s]
4|779|Loss: 1.0239126682281494:  12%|█▏        | 779/6470 [08:54<1:08:22,  1.39it/s]
4|780|Loss: 0.8319789171218872:  12%|█▏        | 779/6470 [08:54<1:08:22,  1.39it/s]
4|780|Loss: 0.8319789171218872:  12%|█▏        | 780/6470 [08:55<1:07:12,  1.41it/s]
4|781|Loss: 0.7807947993278503:  12%|█▏        | 780/6470 [08:55<1:07:12,  1.41it/s]
4|781|Loss: 0.7807947993278503:  12%|█▏        | 781/6470 [08:55<1:07:58,  1.39it/s]
4|782|Loss: 0.9210984706878662:  12%|█▏        | 781/6470 [08:56<1:07:58,  1.39it/s]
4|782|Loss: 0.9210984706878662:  12%|█▏        | 782/6470 [08:56<1:02:50,  1.51it/s]
4|783|Loss: 1.0119216442108154:  12%|█▏        | 782/6470 [08:56<1:02:50,  1.51it/s]
4|783|Loss: 1.0119216442108154:  12%|█▏        | 783/6470 [08:57<1:07:29,  1.40it/s]
4|784|Loss: 0.8018428087234497:  12%|█▏        | 783/6470 [08:57<1:07:29,  1.40it/s]
4|784|Loss: 0.8018428087234497:  12%|█▏        | 784/6470 [08:58<1:10:23,  1.35it/s]
4|785|Loss: 0.7671635746955872:  12%|█▏        | 784/6470 [08:58<1:10:23,  1.35it/s]
4|785|Loss: 0.7671635746955872:  12%|█▏        | 785/6470 [08:58<1:08:27,  1.38it/s]
4|786|Loss: 0.8368623852729797:  12%|█▏        | 785/6470 [08:59<1:08:27,  1.38it/s]
4|786|Loss: 0.8368623852729797:  12%|█▏        | 786/6470 [08:59<1:04:01,  1.48it/s]
4|787|Loss: 0.9156110286712646:  12%|█▏        | 786/6470 [08:59<1:04:01,  1.48it/s]
4|787|Loss: 0.9156110286712646:  12%|█▏        | 787/6470 [09:00<1:06:16,  1.43it/s]
4|788|Loss: 0.950316309928894:  12%|█▏        | 787/6470 [09:00<1:06:16,  1.43it/s] 
4|788|Loss: 0.950316309928894:  12%|█▏        | 788/6470 [09:00<1:05:07,  1.45it/s]
4|789|Loss: 0.8366470336914062:  12%|█▏        | 788/6470 [09:01<1:05:07,  1.45it/s]
4|789|Loss: 0.8366470336914062:  12%|█▏        | 789/6470 [09:01<1:06:03,  1.43it/s]
4|790|Loss: 0.8814763426780701:  12%|█▏        | 789/6470 [09:01<1:06:03,  1.43it/s]
4|790|Loss: 0.8814763426780701:  12%|█▏        | 790/6470 [09:02<1:05:12,  1.45it/s]
4|791|Loss: 0.8505458235740662:  12%|█▏        | 790/6470 [09:02<1:05:12,  1.45it/s]
4|791|Loss: 0.8505458235740662:  12%|█▏        | 791/6470 [09:02<1:05:27,  1.45it/s]
4|792|Loss: 0.7637280821800232:  12%|█▏        | 791/6470 [09:03<1:05:27,  1.45it/s]
4|792|Loss: 0.7637280821800232:  12%|█▏        | 792/6470 [09:03<1:08:29,  1.38it/s]
4|793|Loss: 0.9660606384277344:  12%|█▏        | 792/6470 [09:03<1:08:29,  1.38it/s]
4|793|Loss: 0.9660606384277344:  12%|█▏        | 793/6470 [09:04<1:07:18,  1.41it/s]
4|794|Loss: 0.8791938424110413:  12%|█▏        | 793/6470 [09:04<1:07:18,  1.41it/s]
4|794|Loss: 0.8791938424110413:  12%|█▏        | 794/6470 [09:04<1:06:45,  1.42it/s]
4|795|Loss: 0.8489652276039124:  12%|█▏        | 794/6470 [09:05<1:06:45,  1.42it/s]
4|795|Loss: 0.8489652276039124:  12%|█▏        | 795/6470 [09:05<1:07:29,  1.40it/s]
4|796|Loss: 0.899830162525177:  12%|█▏        | 795/6470 [09:06<1:07:29,  1.40it/s] 
4|796|Loss: 0.899830162525177:  12%|█▏        | 796/6470 [09:06<1:09:22,  1.36it/s]
4|797|Loss: 0.8320515155792236:  12%|█▏        | 796/6470 [09:06<1:09:22,  1.36it/s]
4|797|Loss: 0.8320515155792236:  12%|█▏        | 797/6470 [09:07<1:07:10,  1.41it/s]
4|798|Loss: 0.9642683863639832:  12%|█▏        | 797/6470 [09:07<1:07:10,  1.41it/s]
4|798|Loss: 0.9642683863639832:  12%|█▏        | 798/6470 [09:07<1:07:25,  1.40it/s]
4|799|Loss: 0.806451678276062:  12%|█▏        | 798/6470 [09:08<1:07:25,  1.40it/s] 
4|799|Loss: 0.806451678276062:  12%|█▏        | 799/6470 [09:08<1:08:31,  1.38it/s]
4|800|Loss: 0.7719393372535706:  12%|█▏        | 799/6470 [09:08<1:08:31,  1.38it/s]
4|800|Loss: 0.7719393372535706:  12%|█▏        | 800/6470 [09:09<1:06:22,  1.42it/s]
4|801|Loss: 0.9010899662971497:  12%|█▏        | 800/6470 [09:09<1:06:22,  1.42it/s]
4|801|Loss: 0.9010899662971497:  12%|█▏        | 801/6470 [09:09<1:04:25,  1.47it/s]
4|802|Loss: 0.870127260684967:  12%|█▏        | 801/6470 [09:10<1:04:25,  1.47it/s] 
4|802|Loss: 0.870127260684967:  12%|█▏        | 802/6470 [09:10<1:00:17,  1.57it/s]
4|803|Loss: 0.8341349959373474:  12%|█▏        | 802/6470 [09:10<1:00:17,  1.57it/s]
4|803|Loss: 0.8341349959373474:  12%|█▏        | 803/6470 [09:10<58:04,  1.63it/s]  
4|804|Loss: 0.8626793622970581:  12%|█▏        | 803/6470 [09:11<58:04,  1.63it/s]
4|804|Loss: 0.8626793622970581:  12%|█▏        | 804/6470 [09:11<59:54,  1.58it/s]
4|805|Loss: 0.9473139047622681:  12%|█▏        | 804/6470 [09:12<59:54,  1.58it/s]
4|805|Loss: 0.9473139047622681:  12%|█▏        | 805/6470 [09:12<1:03:44,  1.48it/s]
4|806|Loss: 0.8368760347366333:  12%|█▏        | 805/6470 [09:12<1:03:44,  1.48it/s]
4|806|Loss: 0.8368760347366333:  12%|█▏        | 806/6470 [09:13<1:06:11,  1.43it/s]
4|807|Loss: 0.8253923058509827:  12%|█▏        | 806/6470 [09:13<1:06:11,  1.43it/s]
4|807|Loss: 0.8253923058509827:  12%|█▏        | 807/6470 [09:13<1:04:03,  1.47it/s]
4|808|Loss: 0.8796742558479309:  12%|█▏        | 807/6470 [09:14<1:04:03,  1.47it/s]
4|808|Loss: 0.8796742558479309:  12%|█▏        | 808/6470 [09:14<1:02:48,  1.50it/s]
4|809|Loss: 0.7712900638580322:  12%|█▏        | 808/6470 [09:14<1:02:48,  1.50it/s]
4|809|Loss: 0.7712900638580322:  13%|█▎        | 809/6470 [09:15<59:43,  1.58it/s]  
4|810|Loss: 0.8283520936965942:  13%|█▎        | 809/6470 [09:15<59:43,  1.58it/s]
4|810|Loss: 0.8283520936965942:  13%|█▎        | 810/6470 [09:15<1:05:35,  1.44it/s]
4|811|Loss: 0.7228625416755676:  13%|█▎        | 810/6470 [09:16<1:05:35,  1.44it/s]
4|811|Loss: 0.7228625416755676:  13%|█▎        | 811/6470 [09:16<1:09:15,  1.36it/s]
4|812|Loss: 0.8861839771270752:  13%|█▎        | 811/6470 [09:17<1:09:15,  1.36it/s]
4|812|Loss: 0.8861839771270752:  13%|█▎        | 812/6470 [09:17<1:10:12,  1.34it/s]
4|813|Loss: 0.7963733077049255:  13%|█▎        | 812/6470 [09:17<1:10:12,  1.34it/s]
4|813|Loss: 0.7963733077049255:  13%|█▎        | 813/6470 [09:18<1:08:07,  1.38it/s]
4|814|Loss: 0.8187485337257385:  13%|█▎        | 813/6470 [09:18<1:08:07,  1.38it/s]
4|814|Loss: 0.8187485337257385:  13%|█▎        | 814/6470 [09:18<1:07:11,  1.40it/s]
4|815|Loss: 0.99185711145401:  13%|█▎        | 814/6470 [09:19<1:07:11,  1.40it/s]  
4|815|Loss: 0.99185711145401:  13%|█▎        | 815/6470 [09:19<1:04:19,  1.47it/s]
4|816|Loss: 0.6783308982849121:  13%|█▎        | 815/6470 [09:19<1:04:19,  1.47it/s]
4|816|Loss: 0.6783308982849121:  13%|█▎        | 816/6470 [09:20<1:10:36,  1.33it/s]
4|817|Loss: 0.7960963845252991:  13%|█▎        | 816/6470 [09:20<1:10:36,  1.33it/s]
4|817|Loss: 0.7960963845252991:  13%|█▎        | 817/6470 [09:21<1:09:15,  1.36it/s]
4|818|Loss: 0.8205969929695129:  13%|█▎        | 817/6470 [09:21<1:09:15,  1.36it/s]
4|818|Loss: 0.8205969929695129:  13%|█▎        | 818/6470 [09:21<1:03:34,  1.48it/s]
4|819|Loss: 0.8193539381027222:  13%|█▎        | 818/6470 [09:21<1:03:34,  1.48it/s]
4|819|Loss: 0.8193539381027222:  13%|█▎        | 819/6470 [09:22<59:22,  1.59it/s]  
4|820|Loss: 0.8976587057113647:  13%|█▎        | 819/6470 [09:22<59:22,  1.59it/s]
4|820|Loss: 0.8976587057113647:  13%|█▎        | 820/6470 [09:22<1:03:24,  1.49it/s]
4|821|Loss: 0.7217183709144592:  13%|█▎        | 820/6470 [09:23<1:03:24,  1.49it/s]
4|821|Loss: 0.7217183709144592:  13%|█▎        | 821/6470 [09:23<1:07:57,  1.39it/s]
4|822|Loss: 0.9725372195243835:  13%|█▎        | 821/6470 [09:24<1:07:57,  1.39it/s]
4|822|Loss: 0.9725372195243835:  13%|█▎        | 822/6470 [09:24<1:04:53,  1.45it/s]
4|823|Loss: 0.8570530414581299:  13%|█▎        | 822/6470 [09:24<1:04:53,  1.45it/s]
4|823|Loss: 0.8570530414581299:  13%|█▎        | 823/6470 [09:24<1:01:33,  1.53it/s]
4|824|Loss: 0.781684160232544:  13%|█▎        | 823/6470 [09:25<1:01:33,  1.53it/s] 
4|824|Loss: 0.781684160232544:  13%|█▎        | 824/6470 [09:25<59:21,  1.59it/s]  
4|825|Loss: 0.626707911491394:  13%|█▎        | 824/6470 [09:25<59:21,  1.59it/s]
4|825|Loss: 0.626707911491394:  13%|█▎        | 825/6470 [09:26<58:43,  1.60it/s]
4|826|Loss: 0.7349064946174622:  13%|█▎        | 825/6470 [09:26<58:43,  1.60it/s]
4|826|Loss: 0.7349064946174622:  13%|█▎        | 826/6470 [09:26<1:03:27,  1.48it/s]
4|827|Loss: 0.808635413646698:  13%|█▎        | 826/6470 [09:27<1:03:27,  1.48it/s] 
4|827|Loss: 0.808635413646698:  13%|█▎        | 827/6470 [09:27<1:03:10,  1.49it/s]
4|828|Loss: 0.8626516461372375:  13%|█▎        | 827/6470 [09:27<1:03:10,  1.49it/s]
4|828|Loss: 0.8626516461372375:  13%|█▎        | 828/6470 [09:28<1:03:38,  1.48it/s]
4|829|Loss: 0.8223057389259338:  13%|█▎        | 828/6470 [09:28<1:03:38,  1.48it/s]
4|829|Loss: 0.8223057389259338:  13%|█▎        | 829/6470 [09:28<1:02:21,  1.51it/s]
4|830|Loss: 0.9254129528999329:  13%|█▎        | 829/6470 [09:29<1:02:21,  1.51it/s]
4|830|Loss: 0.9254129528999329:  13%|█▎        | 830/6470 [09:29<1:04:04,  1.47it/s]
4|831|Loss: 0.9097999930381775:  13%|█▎        | 830/6470 [09:30<1:04:04,  1.47it/s]
4|831|Loss: 0.9097999930381775:  13%|█▎        | 831/6470 [09:30<1:07:17,  1.40it/s]
4|832|Loss: 0.7282499074935913:  13%|█▎        | 831/6470 [09:30<1:07:17,  1.40it/s]
4|832|Loss: 0.7282499074935913:  13%|█▎        | 832/6470 [09:31<1:13:12,  1.28it/s]
4|833|Loss: 0.8729293942451477:  13%|█▎        | 832/6470 [09:31<1:13:12,  1.28it/s]
4|833|Loss: 0.8729293942451477:  13%|█▎        | 833/6470 [09:32<1:13:12,  1.28it/s]
4|834|Loss: 0.9795957207679749:  13%|█▎        | 833/6470 [09:32<1:13:12,  1.28it/s]
4|834|Loss: 0.9795957207679749:  13%|█▎        | 834/6470 [09:32<1:12:16,  1.30it/s]
4|835|Loss: 0.6907895803451538:  13%|█▎        | 834/6470 [09:33<1:12:16,  1.30it/s]
4|835|Loss: 0.6907895803451538:  13%|█▎        | 835/6470 [09:33<1:14:03,  1.27it/s]
4|836|Loss: 0.7433293461799622:  13%|█▎        | 835/6470 [09:34<1:14:03,  1.27it/s]
4|836|Loss: 0.7433293461799622:  13%|█▎        | 836/6470 [09:34<1:16:38,  1.23it/s]
4|837|Loss: 0.9204885959625244:  13%|█▎        | 836/6470 [09:34<1:16:38,  1.23it/s]
4|837|Loss: 0.9204885959625244:  13%|█▎        | 837/6470 [09:35<1:15:17,  1.25it/s]
4|838|Loss: 0.9420616030693054:  13%|█▎        | 837/6470 [09:35<1:15:17,  1.25it/s]
4|838|Loss: 0.9420616030693054:  13%|█▎        | 838/6470 [09:36<1:14:49,  1.25it/s]
4|839|Loss: 0.7428956627845764:  13%|█▎        | 838/6470 [09:36<1:14:49,  1.25it/s]
4|839|Loss: 0.7428956627845764:  13%|█▎        | 839/6470 [09:36<1:13:30,  1.28it/s]
4|840|Loss: 0.9988808035850525:  13%|█▎        | 839/6470 [09:37<1:13:30,  1.28it/s]
4|840|Loss: 0.9988808035850525:  13%|█▎        | 840/6470 [09:37<1:14:18,  1.26it/s]
4|841|Loss: 0.8447731137275696:  13%|█▎        | 840/6470 [09:38<1:14:18,  1.26it/s]
4|841|Loss: 0.8447731137275696:  13%|█▎        | 841/6470 [09:38<1:13:34,  1.28it/s]
4|842|Loss: 0.680664598941803:  13%|█▎        | 841/6470 [09:38<1:13:34,  1.28it/s] 
4|842|Loss: 0.680664598941803:  13%|█▎        | 842/6470 [09:39<1:11:41,  1.31it/s]
4|843|Loss: 0.6619172692298889:  13%|█▎        | 842/6470 [09:39<1:11:41,  1.31it/s]
4|843|Loss: 0.6619172692298889:  13%|█▎        | 843/6470 [09:39<1:04:15,  1.46it/s]
4|844|Loss: 0.6900275945663452:  13%|█▎        | 843/6470 [09:40<1:04:15,  1.46it/s]
4|844|Loss: 0.6900275945663452:  13%|█▎        | 844/6470 [09:40<1:05:55,  1.42it/s]
4|845|Loss: 1.0004215240478516:  13%|█▎        | 844/6470 [09:40<1:05:55,  1.42it/s]
4|845|Loss: 1.0004215240478516:  13%|█▎        | 845/6470 [09:40<1:03:14,  1.48it/s]
4|846|Loss: 0.8297514319419861:  13%|█▎        | 845/6470 [09:41<1:03:14,  1.48it/s]
4|846|Loss: 0.8297514319419861:  13%|█▎        | 846/6470 [09:41<1:01:46,  1.52it/s]
4|847|Loss: 0.9394015669822693:  13%|█▎        | 846/6470 [09:41<1:01:46,  1.52it/s]
4|847|Loss: 0.9394015669822693:  13%|█▎        | 847/6470 [09:42<1:00:59,  1.54it/s]
4|848|Loss: 0.8144643306732178:  13%|█▎        | 847/6470 [09:42<1:00:59,  1.54it/s]
4|848|Loss: 0.8144643306732178:  13%|█▎        | 848/6470 [09:42<1:02:05,  1.51it/s]
4|849|Loss: 0.806350827217102:  13%|█▎        | 848/6470 [09:43<1:02:05,  1.51it/s] 
4|849|Loss: 0.806350827217102:  13%|█▎        | 849/6470 [09:43<1:06:11,  1.42it/s]
4|850|Loss: 0.8897442817687988:  13%|█▎        | 849/6470 [09:44<1:06:11,  1.42it/s]
4|850|Loss: 0.8897442817687988:  13%|█▎        | 850/6470 [09:44<1:05:49,  1.42it/s]
4|851|Loss: 0.9276465773582458:  13%|█▎        | 850/6470 [09:44<1:05:49,  1.42it/s]
4|851|Loss: 0.9276465773582458:  13%|█▎        | 851/6470 [09:44<59:54,  1.56it/s]  
4|852|Loss: 1.0663713216781616:  13%|█▎        | 851/6470 [09:45<59:54,  1.56it/s]
4|852|Loss: 1.0663713216781616:  13%|█▎        | 852/6470 [09:45<52:58,  1.77it/s]
4|853|Loss: 0.7650870680809021:  13%|█▎        | 852/6470 [09:45<52:58,  1.77it/s]
4|853|Loss: 0.7650870680809021:  13%|█▎        | 853/6470 [09:45<55:57,  1.67it/s]
4|854|Loss: 0.8481807112693787:  13%|█▎        | 853/6470 [09:46<55:57,  1.67it/s]
4|854|Loss: 0.8481807112693787:  13%|█▎        | 854/6470 [09:46<57:51,  1.62it/s]
4|855|Loss: 0.880273163318634:  13%|█▎        | 854/6470 [09:47<57:51,  1.62it/s] 
4|855|Loss: 0.880273163318634:  13%|█▎        | 855/6470 [09:47<1:02:54,  1.49it/s]
4|856|Loss: 0.9735884666442871:  13%|█▎        | 855/6470 [09:47<1:02:54,  1.49it/s]
4|856|Loss: 0.9735884666442871:  13%|█▎        | 856/6470 [09:48<1:06:20,  1.41it/s]
4|857|Loss: 0.9965649247169495:  13%|█▎        | 856/6470 [09:48<1:06:20,  1.41it/s]
4|857|Loss: 0.9965649247169495:  13%|█▎        | 857/6470 [09:48<1:04:01,  1.46it/s]
4|858|Loss: 0.8863618969917297:  13%|█▎        | 857/6470 [09:49<1:04:01,  1.46it/s]
4|858|Loss: 0.8863618969917297:  13%|█▎        | 858/6470 [09:49<1:08:01,  1.38it/s]
4|859|Loss: 0.8858360052108765:  13%|█▎        | 858/6470 [09:50<1:08:01,  1.38it/s]
4|859|Loss: 0.8858360052108765:  13%|█▎        | 859/6470 [09:50<1:13:22,  1.27it/s]
4|860|Loss: 0.938647449016571:  13%|█▎        | 859/6470 [09:51<1:13:22,  1.27it/s] 
4|860|Loss: 0.938647449016571:  13%|█▎        | 860/6470 [09:51<1:10:23,  1.33it/s]
4|861|Loss: 0.8147906064987183:  13%|█▎        | 860/6470 [09:51<1:10:23,  1.33it/s]
4|861|Loss: 0.8147906064987183:  13%|█▎        | 861/6470 [09:52<1:11:27,  1.31it/s]
4|862|Loss: 0.7949696183204651:  13%|█▎        | 861/6470 [09:52<1:11:27,  1.31it/s]
4|862|Loss: 0.7949696183204651:  13%|█▎        | 862/6470 [09:52<1:10:20,  1.33it/s]
4|863|Loss: 0.9124081134796143:  13%|█▎        | 862/6470 [09:53<1:10:20,  1.33it/s]
4|863|Loss: 0.9124081134796143:  13%|█▎        | 863/6470 [09:53<1:03:54,  1.46it/s]
4|864|Loss: 0.9471604228019714:  13%|█▎        | 863/6470 [09:53<1:03:54,  1.46it/s]
4|864|Loss: 0.9471604228019714:  13%|█▎        | 864/6470 [09:54<1:03:45,  1.47it/s]
4|865|Loss: 0.8333532810211182:  13%|█▎        | 864/6470 [09:54<1:03:45,  1.47it/s]
4|865|Loss: 0.8333532810211182:  13%|█▎        | 865/6470 [09:54<58:53,  1.59it/s]  
4|866|Loss: 0.8807799220085144:  13%|█▎        | 865/6470 [09:54<58:53,  1.59it/s]
4|866|Loss: 0.8807799220085144:  13%|█▎        | 866/6470 [09:55<59:17,  1.58it/s]
4|867|Loss: 0.7697609663009644:  13%|█▎        | 866/6470 [09:55<59:17,  1.58it/s]
4|867|Loss: 0.7697609663009644:  13%|█▎        | 867/6470 [09:55<1:02:29,  1.49it/s]
4|868|Loss: 0.8906927704811096:  13%|█▎        | 867/6470 [09:56<1:02:29,  1.49it/s]
4|868|Loss: 0.8906927704811096:  13%|█▎        | 868/6470 [09:56<1:03:52,  1.46it/s]
4|869|Loss: 0.8443086743354797:  13%|█▎        | 868/6470 [09:57<1:03:52,  1.46it/s]
4|869|Loss: 0.8443086743354797:  13%|█▎        | 869/6470 [09:57<1:02:52,  1.48it/s]
4|870|Loss: 0.9410082697868347:  13%|█▎        | 869/6470 [09:57<1:02:52,  1.48it/s]
4|870|Loss: 0.9410082697868347:  13%|█▎        | 870/6470 [09:57<1:02:02,  1.50it/s]
4|871|Loss: 0.8213318586349487:  13%|█▎        | 870/6470 [09:58<1:02:02,  1.50it/s]
4|871|Loss: 0.8213318586349487:  13%|█▎        | 871/6470 [09:58<1:05:34,  1.42it/s]
4|872|Loss: 0.9063781499862671:  13%|█▎        | 871/6470 [09:59<1:05:34,  1.42it/s]
4|872|Loss: 0.9063781499862671:  13%|█▎        | 872/6470 [09:59<1:02:58,  1.48it/s]
4|873|Loss: 0.930366039276123:  13%|█▎        | 872/6470 [09:59<1:02:58,  1.48it/s] 
4|873|Loss: 0.930366039276123:  13%|█▎        | 873/6470 [09:59<58:00,  1.61it/s]  
4|874|Loss: 0.8911449313163757:  13%|█▎        | 873/6470 [10:00<58:00,  1.61it/s]
4|874|Loss: 0.8911449313163757:  14%|█▎        | 874/6470 [10:00<59:01,  1.58it/s]
4|875|Loss: 0.7813485264778137:  14%|█▎        | 874/6470 [10:00<59:01,  1.58it/s]
4|875|Loss: 0.7813485264778137:  14%|█▎        | 875/6470 [10:01<1:00:55,  1.53it/s]
4|876|Loss: 0.8644701838493347:  14%|█▎        | 875/6470 [10:01<1:00:55,  1.53it/s]
4|876|Loss: 0.8644701838493347:  14%|█▎        | 876/6470 [10:01<1:00:12,  1.55it/s]
4|877|Loss: 1.0239944458007812:  14%|█▎        | 876/6470 [10:02<1:00:12,  1.55it/s]
4|877|Loss: 1.0239944458007812:  14%|█▎        | 877/6470 [10:02<59:48,  1.56it/s]  
4|878|Loss: 0.975278913974762:  14%|█▎        | 877/6470 [10:02<59:48,  1.56it/s] 
4|878|Loss: 0.975278913974762:  14%|█▎        | 878/6470 [10:03<1:02:20,  1.49it/s]
4|879|Loss: 0.9459845423698425:  14%|█▎        | 878/6470 [10:03<1:02:20,  1.49it/s]
4|879|Loss: 0.9459845423698425:  14%|█▎        | 879/6470 [10:03<1:03:37,  1.46it/s]
4|880|Loss: 0.7225390672683716:  14%|█▎        | 879/6470 [10:04<1:03:37,  1.46it/s]
4|880|Loss: 0.7225390672683716:  14%|█▎        | 880/6470 [10:04<1:00:39,  1.54it/s]
4|881|Loss: 1.0610744953155518:  14%|█▎        | 880/6470 [10:04<1:00:39,  1.54it/s]
4|881|Loss: 1.0610744953155518:  14%|█▎        | 881/6470 [10:05<1:04:13,  1.45it/s]
4|882|Loss: 0.8141238689422607:  14%|█▎        | 881/6470 [10:05<1:04:13,  1.45it/s]
4|882|Loss: 0.8141238689422607:  14%|█▎        | 882/6470 [10:05<1:05:25,  1.42it/s]
4|883|Loss: 0.6978872418403625:  14%|█▎        | 882/6470 [10:06<1:05:25,  1.42it/s]
4|883|Loss: 0.6978872418403625:  14%|█▎        | 883/6470 [10:06<1:12:03,  1.29it/s]
4|884|Loss: 0.7826932668685913:  14%|█▎        | 883/6470 [10:07<1:12:03,  1.29it/s]
4|884|Loss: 0.7826932668685913:  14%|█▎        | 884/6470 [10:07<1:06:30,  1.40it/s]
4|885|Loss: 0.8665060997009277:  14%|█▎        | 884/6470 [10:07<1:06:30,  1.40it/s]
4|885|Loss: 0.8665060997009277:  14%|█▎        | 885/6470 [10:08<1:08:21,  1.36it/s]
4|886|Loss: 0.7608529329299927:  14%|█▎        | 885/6470 [10:08<1:08:21,  1.36it/s]
4|886|Loss: 0.7608529329299927:  14%|█▎        | 886/6470 [10:09<1:09:12,  1.34it/s]
4|887|Loss: 0.8085174560546875:  14%|█▎        | 886/6470 [10:09<1:09:12,  1.34it/s]
4|887|Loss: 0.8085174560546875:  14%|█▎        | 887/6470 [10:09<1:11:11,  1.31it/s]
4|888|Loss: 0.7030293941497803:  14%|█▎        | 887/6470 [10:10<1:11:11,  1.31it/s]
4|888|Loss: 0.7030293941497803:  14%|█▎        | 888/6470 [10:10<1:11:17,  1.31it/s]
4|889|Loss: 0.7638757228851318:  14%|█▎        | 888/6470 [10:11<1:11:17,  1.31it/s]
4|889|Loss: 0.7638757228851318:  14%|█▎        | 889/6470 [10:11<1:08:35,  1.36it/s]
4|890|Loss: 0.799130380153656:  14%|█▎        | 889/6470 [10:11<1:08:35,  1.36it/s] 
4|890|Loss: 0.799130380153656:  14%|█▍        | 890/6470 [10:11<1:04:35,  1.44it/s]
4|891|Loss: 0.8060853481292725:  14%|█▍        | 890/6470 [10:12<1:04:35,  1.44it/s]
4|891|Loss: 0.8060853481292725:  14%|█▍        | 891/6470 [10:12<1:05:58,  1.41it/s]
4|892|Loss: 0.8092361688613892:  14%|█▍        | 891/6470 [10:13<1:05:58,  1.41it/s]
4|892|Loss: 0.8092361688613892:  14%|█▍        | 892/6470 [10:13<1:02:03,  1.50it/s]
4|893|Loss: 0.8123543858528137:  14%|█▍        | 892/6470 [10:13<1:02:03,  1.50it/s]
4|893|Loss: 0.8123543858528137:  14%|█▍        | 893/6470 [10:14<1:06:43,  1.39it/s]
4|894|Loss: 0.7748059630393982:  14%|█▍        | 893/6470 [10:14<1:06:43,  1.39it/s]
4|894|Loss: 0.7748059630393982:  14%|█▍        | 894/6470 [10:14<1:04:42,  1.44it/s]
4|895|Loss: 0.7631065845489502:  14%|█▍        | 894/6470 [10:15<1:04:42,  1.44it/s]
4|895|Loss: 0.7631065845489502:  14%|█▍        | 895/6470 [10:15<1:04:56,  1.43it/s]
4|896|Loss: 0.8156224489212036:  14%|█▍        | 895/6470 [10:15<1:04:56,  1.43it/s]
4|896|Loss: 0.8156224489212036:  14%|█▍        | 896/6470 [10:15<1:01:50,  1.50it/s]
4|897|Loss: 1.036381721496582:  14%|█▍        | 896/6470 [10:16<1:01:50,  1.50it/s] 
4|897|Loss: 1.036381721496582:  14%|█▍        | 897/6470 [10:16<1:00:35,  1.53it/s]
4|898|Loss: 0.9200835824012756:  14%|█▍        | 897/6470 [10:17<1:00:35,  1.53it/s]
4|898|Loss: 0.9200835824012756:  14%|█▍        | 898/6470 [10:17<1:05:13,  1.42it/s]
4|899|Loss: 0.9210972189903259:  14%|█▍        | 898/6470 [10:17<1:05:13,  1.42it/s]
4|899|Loss: 0.9210972189903259:  14%|█▍        | 899/6470 [10:18<1:08:39,  1.35it/s]
4|900|Loss: 0.977364182472229:  14%|█▍        | 899/6470 [10:18<1:08:39,  1.35it/s] 
4|900|Loss: 0.977364182472229:  14%|█▍        | 900/6470 [10:18<1:02:36,  1.48it/s]
4|901|Loss: 0.7524566650390625:  14%|█▍        | 900/6470 [10:19<1:02:36,  1.48it/s]
4|901|Loss: 0.7524566650390625:  14%|█▍        | 901/6470 [10:19<1:02:30,  1.48it/s]
4|902|Loss: 0.9164242148399353:  14%|█▍        | 901/6470 [10:19<1:02:30,  1.48it/s]
4|902|Loss: 0.9164242148399353:  14%|█▍        | 902/6470 [10:20<1:02:07,  1.49it/s]
4|903|Loss: 0.914955735206604:  14%|█▍        | 902/6470 [10:20<1:02:07,  1.49it/s] 
4|903|Loss: 0.914955735206604:  14%|█▍        | 903/6470 [10:20<1:03:38,  1.46it/s]
4|904|Loss: 0.9822510480880737:  14%|█▍        | 903/6470 [10:21<1:03:38,  1.46it/s]
4|904|Loss: 0.9822510480880737:  14%|█▍        | 904/6470 [10:21<1:03:07,  1.47it/s]
4|905|Loss: 0.8885790109634399:  14%|█▍        | 904/6470 [10:21<1:03:07,  1.47it/s]
4|905|Loss: 0.8885790109634399:  14%|█▍        | 905/6470 [10:22<1:06:07,  1.40it/s]
4|906|Loss: 0.533463716506958:  14%|█▍        | 905/6470 [10:22<1:06:07,  1.40it/s] 
4|906|Loss: 0.533463716506958:  14%|█▍        | 906/6470 [10:23<1:23:24,  1.11it/s]
4|907|Loss: 0.8314794898033142:  14%|█▍        | 906/6470 [10:24<1:23:24,  1.11it/s]
4|907|Loss: 0.8314794898033142:  14%|█▍        | 907/6470 [10:24<1:23:03,  1.12it/s]
4|908|Loss: 0.7634106874465942:  14%|█▍        | 907/6470 [10:24<1:23:03,  1.12it/s]
4|908|Loss: 0.7634106874465942:  14%|█▍        | 908/6470 [10:25<1:19:07,  1.17it/s]
4|909|Loss: 0.8841617703437805:  14%|█▍        | 908/6470 [10:25<1:19:07,  1.17it/s]
4|909|Loss: 0.8841617703437805:  14%|█▍        | 909/6470 [10:25<1:13:57,  1.25it/s]
4|910|Loss: 0.7693203687667847:  14%|█▍        | 909/6470 [10:26<1:13:57,  1.25it/s]
4|910|Loss: 0.7693203687667847:  14%|█▍        | 910/6470 [10:26<1:13:44,  1.26it/s]
4|911|Loss: 0.8867447972297668:  14%|█▍        | 910/6470 [10:27<1:13:44,  1.26it/s]
4|911|Loss: 0.8867447972297668:  14%|█▍        | 911/6470 [10:27<1:10:04,  1.32it/s]
4|912|Loss: 0.7773526310920715:  14%|█▍        | 911/6470 [10:27<1:10:04,  1.32it/s]
4|912|Loss: 0.7773526310920715:  14%|█▍        | 912/6470 [10:28<1:07:56,  1.36it/s]
4|913|Loss: 0.8950821161270142:  14%|█▍        | 912/6470 [10:28<1:07:56,  1.36it/s]
4|913|Loss: 0.8950821161270142:  14%|█▍        | 913/6470 [10:28<1:07:55,  1.36it/s]
4|914|Loss: 0.7473062872886658:  14%|█▍        | 913/6470 [10:29<1:07:55,  1.36it/s]
4|914|Loss: 0.7473062872886658:  14%|█▍        | 914/6470 [10:29<1:06:11,  1.40it/s]
4|915|Loss: 0.8653797507286072:  14%|█▍        | 914/6470 [10:29<1:06:11,  1.40it/s]
4|915|Loss: 0.8653797507286072:  14%|█▍        | 915/6470 [10:30<1:02:48,  1.47it/s]
4|916|Loss: 0.7665649056434631:  14%|█▍        | 915/6470 [10:30<1:02:48,  1.47it/s]
4|916|Loss: 0.7665649056434631:  14%|█▍        | 916/6470 [10:30<1:03:38,  1.45it/s]
4|917|Loss: 0.7673107981681824:  14%|█▍        | 916/6470 [10:31<1:03:38,  1.45it/s]
4|917|Loss: 0.7673107981681824:  14%|█▍        | 917/6470 [10:31<1:00:53,  1.52it/s]
4|918|Loss: 0.8912990093231201:  14%|█▍        | 917/6470 [10:31<1:00:53,  1.52it/s]
4|918|Loss: 0.8912990093231201:  14%|█▍        | 918/6470 [10:32<1:01:15,  1.51it/s]
4|919|Loss: 0.9910474419593811:  14%|█▍        | 918/6470 [10:32<1:01:15,  1.51it/s]
4|919|Loss: 0.9910474419593811:  14%|█▍        | 919/6470 [10:32<1:05:44,  1.41it/s]
4|920|Loss: 0.7751802802085876:  14%|█▍        | 919/6470 [10:33<1:05:44,  1.41it/s]
4|920|Loss: 0.7751802802085876:  14%|█▍        | 920/6470 [10:33<1:02:12,  1.49it/s]
4|921|Loss: 0.8471085429191589:  14%|█▍        | 920/6470 [10:33<1:02:12,  1.49it/s]
4|921|Loss: 0.8471085429191589:  14%|█▍        | 921/6470 [10:34<1:04:32,  1.43it/s]
4|922|Loss: 0.8177189230918884:  14%|█▍        | 921/6470 [10:34<1:04:32,  1.43it/s]
4|922|Loss: 0.8177189230918884:  14%|█▍        | 922/6470 [10:35<1:07:18,  1.37it/s]
4|923|Loss: 0.8601343035697937:  14%|█▍        | 922/6470 [10:35<1:07:18,  1.37it/s]
4|923|Loss: 0.8601343035697937:  14%|█▍        | 923/6470 [10:35<1:07:13,  1.38it/s]
4|924|Loss: 0.7144466042518616:  14%|█▍        | 923/6470 [10:36<1:07:13,  1.38it/s]
4|924|Loss: 0.7144466042518616:  14%|█▍        | 924/6470 [10:36<1:09:14,  1.33it/s]
4|925|Loss: 0.8232856392860413:  14%|█▍        | 924/6470 [10:36<1:09:14,  1.33it/s]
4|925|Loss: 0.8232856392860413:  14%|█▍        | 925/6470 [10:37<1:06:57,  1.38it/s]
4|926|Loss: 0.8525924682617188:  14%|█▍        | 925/6470 [10:37<1:06:57,  1.38it/s]
4|926|Loss: 0.8525924682617188:  14%|█▍        | 926/6470 [10:37<1:06:30,  1.39it/s]
4|927|Loss: 0.695129930973053:  14%|█▍        | 926/6470 [10:38<1:06:30,  1.39it/s] 
4|927|Loss: 0.695129930973053:  14%|█▍        | 927/6470 [10:38<1:06:50,  1.38it/s]
4|928|Loss: 0.6032745242118835:  14%|█▍        | 927/6470 [10:39<1:06:50,  1.38it/s]
4|928|Loss: 0.6032745242118835:  14%|█▍        | 928/6470 [10:39<1:10:10,  1.32it/s]
4|929|Loss: 0.724494218826294:  14%|█▍        | 928/6470 [10:39<1:10:10,  1.32it/s] 
4|929|Loss: 0.724494218826294:  14%|█▍        | 929/6470 [10:40<1:07:49,  1.36it/s]
4|930|Loss: 0.9174321293830872:  14%|█▍        | 929/6470 [10:40<1:07:49,  1.36it/s]
4|930|Loss: 0.9174321293830872:  14%|█▍        | 930/6470 [10:40<1:04:30,  1.43it/s]
4|931|Loss: 0.9894833564758301:  14%|█▍        | 930/6470 [10:41<1:04:30,  1.43it/s]
4|931|Loss: 0.9894833564758301:  14%|█▍        | 931/6470 [10:41<1:04:12,  1.44it/s]
4|932|Loss: 0.9673815965652466:  14%|█▍        | 931/6470 [10:41<1:04:12,  1.44it/s]
4|932|Loss: 0.9673815965652466:  14%|█▍        | 932/6470 [10:42<1:05:25,  1.41it/s]
4|933|Loss: 0.8317245841026306:  14%|█▍        | 932/6470 [10:42<1:05:25,  1.41it/s]
4|933|Loss: 0.8317245841026306:  14%|█▍        | 933/6470 [10:42<1:06:18,  1.39it/s]
4|934|Loss: 1.0438050031661987:  14%|█▍        | 933/6470 [10:43<1:06:18,  1.39it/s]
4|934|Loss: 1.0438050031661987:  14%|█▍        | 934/6470 [10:43<1:06:04,  1.40it/s]
4|935|Loss: 0.9000564217567444:  14%|█▍        | 934/6470 [10:44<1:06:04,  1.40it/s]
4|935|Loss: 0.9000564217567444:  14%|█▍        | 935/6470 [10:44<1:05:37,  1.41it/s]
4|936|Loss: 0.8532413244247437:  14%|█▍        | 935/6470 [10:44<1:05:37,  1.41it/s]
4|936|Loss: 0.8532413244247437:  14%|█▍        | 936/6470 [10:44<1:03:34,  1.45it/s]
4|937|Loss: 0.8771119117736816:  14%|█▍        | 936/6470 [10:45<1:03:34,  1.45it/s]
4|937|Loss: 0.8771119117736816:  14%|█▍        | 937/6470 [10:45<1:02:07,  1.48it/s]
4|938|Loss: 0.8565288782119751:  14%|█▍        | 937/6470 [10:46<1:02:07,  1.48it/s]
4|938|Loss: 0.8565288782119751:  14%|█▍        | 938/6470 [10:46<1:03:19,  1.46it/s]
4|939|Loss: 0.8542532920837402:  14%|█▍        | 938/6470 [10:46<1:03:19,  1.46it/s]
4|939|Loss: 0.8542532920837402:  15%|█▍        | 939/6470 [10:47<1:06:47,  1.38it/s]
4|940|Loss: 0.9333556294441223:  15%|█▍        | 939/6470 [10:47<1:06:47,  1.38it/s]
4|940|Loss: 0.9333556294441223:  15%|█▍        | 940/6470 [10:48<1:10:13,  1.31it/s]
4|941|Loss: 0.9523352980613708:  15%|█▍        | 940/6470 [10:48<1:10:13,  1.31it/s]
4|941|Loss: 0.9523352980613708:  15%|█▍        | 941/6470 [10:48<1:06:56,  1.38it/s]
4|942|Loss: 0.7832011580467224:  15%|█▍        | 941/6470 [10:49<1:06:56,  1.38it/s]
4|942|Loss: 0.7832011580467224:  15%|█▍        | 942/6470 [10:49<1:05:25,  1.41it/s]
4|943|Loss: 0.8750805258750916:  15%|█▍        | 942/6470 [10:49<1:05:25,  1.41it/s]
4|943|Loss: 0.8750805258750916:  15%|█▍        | 943/6470 [10:49<1:03:54,  1.44it/s]
4|944|Loss: 0.7571044564247131:  15%|█▍        | 943/6470 [10:50<1:03:54,  1.44it/s]
4|944|Loss: 0.7571044564247131:  15%|█▍        | 944/6470 [10:50<1:05:55,  1.40it/s]
4|945|Loss: 0.9778153896331787:  15%|█▍        | 944/6470 [10:51<1:05:55,  1.40it/s]
4|945|Loss: 0.9778153896331787:  15%|█▍        | 945/6470 [10:51<1:02:17,  1.48it/s]
4|946|Loss: 0.7724426984786987:  15%|█▍        | 945/6470 [10:51<1:02:17,  1.48it/s]
4|946|Loss: 0.7724426984786987:  15%|█▍        | 946/6470 [10:51<56:27,  1.63it/s]  
4|947|Loss: 0.7915774583816528:  15%|█▍        | 946/6470 [10:52<56:27,  1.63it/s]
4|947|Loss: 0.7915774583816528:  15%|█▍        | 947/6470 [10:52<51:49,  1.78it/s]
4|948|Loss: 0.9316443800926208:  15%|█▍        | 947/6470 [10:52<51:49,  1.78it/s]
4|948|Loss: 0.9316443800926208:  15%|█▍        | 948/6470 [10:52<47:07,  1.95it/s]
4|949|Loss: 0.8017778396606445:  15%|█▍        | 948/6470 [10:52<47:07,  1.95it/s]
4|949|Loss: 0.8017778396606445:  15%|█▍        | 949/6470 [10:53<46:58,  1.96it/s]
4|950|Loss: 0.8293717503547668:  15%|█▍        | 949/6470 [10:53<46:58,  1.96it/s]
4|950|Loss: 0.8293717503547668:  15%|█▍        | 950/6470 [10:53<47:48,  1.92it/s]
4|951|Loss: 0.929050087928772:  15%|█▍        | 950/6470 [10:54<47:48,  1.92it/s] 
4|951|Loss: 0.929050087928772:  15%|█▍        | 951/6470 [10:54<52:01,  1.77it/s]
4|952|Loss: 0.8095255494117737:  15%|█▍        | 951/6470 [10:54<52:01,  1.77it/s]
4|952|Loss: 0.8095255494117737:  15%|█▍        | 952/6470 [10:55<59:26,  1.55it/s]
4|953|Loss: 0.8950626254081726:  15%|█▍        | 952/6470 [10:55<59:26,  1.55it/s]
4|953|Loss: 0.8950626254081726:  15%|█▍        | 953/6470 [10:55<1:00:43,  1.51it/s]
4|954|Loss: 0.8188770413398743:  15%|█▍        | 953/6470 [10:56<1:00:43,  1.51it/s]
4|954|Loss: 0.8188770413398743:  15%|█▍        | 954/6470 [10:56<1:01:47,  1.49it/s]
4|955|Loss: 0.893410861492157:  15%|█▍        | 954/6470 [10:56<1:01:47,  1.49it/s] 
4|955|Loss: 0.893410861492157:  15%|█▍        | 955/6470 [10:57<58:20,  1.58it/s]  
4|956|Loss: 0.8598806262016296:  15%|█▍        | 955/6470 [10:57<58:20,  1.58it/s]
4|956|Loss: 0.8598806262016296:  15%|█▍        | 956/6470 [10:57<57:14,  1.61it/s]
4|957|Loss: 0.8526962995529175:  15%|█▍        | 956/6470 [10:58<57:14,  1.61it/s]
4|957|Loss: 0.8526962995529175:  15%|█▍        | 957/6470 [10:58<56:29,  1.63it/s]
4|958|Loss: 0.9732218384742737:  15%|█▍        | 957/6470 [10:58<56:29,  1.63it/s]
4|958|Loss: 0.9732218384742737:  15%|█▍        | 958/6470 [10:59<58:57,  1.56it/s]
4|959|Loss: 0.6993734836578369:  15%|█▍        | 958/6470 [10:59<58:57,  1.56it/s]
4|959|Loss: 0.6993734836578369:  15%|█▍        | 959/6470 [10:59<58:51,  1.56it/s]
4|960|Loss: 0.7528086304664612:  15%|█▍        | 959/6470 [11:00<58:51,  1.56it/s]
4|960|Loss: 0.7528086304664612:  15%|█▍        | 960/6470 [11:00<1:00:29,  1.52it/s]
4|961|Loss: 1.0905901193618774:  15%|█▍        | 960/6470 [11:00<1:00:29,  1.52it/s]
4|961|Loss: 1.0905901193618774:  15%|█▍        | 961/6470 [11:01<1:01:45,  1.49it/s]
4|962|Loss: 0.8022099733352661:  15%|█▍        | 961/6470 [11:01<1:01:45,  1.49it/s]
4|962|Loss: 0.8022099733352661:  15%|█▍        | 962/6470 [11:01<1:03:15,  1.45it/s]
4|963|Loss: 0.7439893484115601:  15%|█▍        | 962/6470 [11:02<1:03:15,  1.45it/s]
4|963|Loss: 0.7439893484115601:  15%|█▍        | 963/6470 [11:02<1:01:02,  1.50it/s]
4|964|Loss: 0.7250324487686157:  15%|█▍        | 963/6470 [11:02<1:01:02,  1.50it/s]
4|964|Loss: 0.7250324487686157:  15%|█▍        | 964/6470 [11:03<1:01:28,  1.49it/s]
4|965|Loss: 0.8363497853279114:  15%|█▍        | 964/6470 [11:03<1:01:28,  1.49it/s]
4|965|Loss: 0.8363497853279114:  15%|█▍        | 965/6470 [11:03<56:56,  1.61it/s]  
4|966|Loss: 0.8605722784996033:  15%|█▍        | 965/6470 [11:03<56:56,  1.61it/s]
4|966|Loss: 0.8605722784996033:  15%|█▍        | 966/6470 [11:04<58:15,  1.57it/s]
4|967|Loss: 0.8741376996040344:  15%|█▍        | 966/6470 [11:04<58:15,  1.57it/s]
4|967|Loss: 0.8741376996040344:  15%|█▍        | 967/6470 [11:05<1:05:04,  1.41it/s]
4|968|Loss: 0.7924889326095581:  15%|█▍        | 967/6470 [11:05<1:05:04,  1.41it/s]
4|968|Loss: 0.7924889326095581:  15%|█▍        | 968/6470 [11:05<1:06:10,  1.39it/s]
4|969|Loss: 0.9367766380310059:  15%|█▍        | 968/6470 [11:06<1:06:10,  1.39it/s]
4|969|Loss: 0.9367766380310059:  15%|█▍        | 969/6470 [11:06<1:05:48,  1.39it/s]
4|970|Loss: 0.9062616229057312:  15%|█▍        | 969/6470 [11:06<1:05:48,  1.39it/s]
4|970|Loss: 0.9062616229057312:  15%|█▍        | 970/6470 [11:07<1:01:05,  1.50it/s]
4|971|Loss: 0.9260297417640686:  15%|█▍        | 970/6470 [11:07<1:01:05,  1.50it/s]
4|971|Loss: 0.9260297417640686:  15%|█▌        | 971/6470 [11:07<1:01:28,  1.49it/s]
4|972|Loss: 0.7238841652870178:  15%|█▌        | 971/6470 [11:08<1:01:28,  1.49it/s]
4|972|Loss: 0.7238841652870178:  15%|█▌        | 972/6470 [11:08<1:06:14,  1.38it/s]
4|973|Loss: 0.8853570818901062:  15%|█▌        | 972/6470 [11:09<1:06:14,  1.38it/s]
4|973|Loss: 0.8853570818901062:  15%|█▌        | 973/6470 [11:09<1:04:53,  1.41it/s]
4|974|Loss: 0.8749572038650513:  15%|█▌        | 973/6470 [11:09<1:04:53,  1.41it/s]
4|974|Loss: 0.8749572038650513:  15%|█▌        | 974/6470 [11:09<1:00:01,  1.53it/s]
4|975|Loss: 0.7627046704292297:  15%|█▌        | 974/6470 [11:10<1:00:01,  1.53it/s]
4|975|Loss: 0.7627046704292297:  15%|█▌        | 975/6470 [11:10<1:00:53,  1.50it/s]
4|976|Loss: 0.8579162955284119:  15%|█▌        | 975/6470 [11:10<1:00:53,  1.50it/s]
4|976|Loss: 0.8579162955284119:  15%|█▌        | 976/6470 [11:11<1:02:46,  1.46it/s]
4|977|Loss: 0.9143304824829102:  15%|█▌        | 976/6470 [11:11<1:02:46,  1.46it/s]
4|977|Loss: 0.9143304824829102:  15%|█▌        | 977/6470 [11:11<1:00:56,  1.50it/s]
4|978|Loss: 1.0316706895828247:  15%|█▌        | 977/6470 [11:12<1:00:56,  1.50it/s]
4|978|Loss: 1.0316706895828247:  15%|█▌        | 978/6470 [11:12<1:01:54,  1.48it/s]
4|979|Loss: 0.7873315215110779:  15%|█▌        | 978/6470 [11:13<1:01:54,  1.48it/s]
4|979|Loss: 0.7873315215110779:  15%|█▌        | 979/6470 [11:13<1:02:39,  1.46it/s]
4|980|Loss: 0.9151755571365356:  15%|█▌        | 979/6470 [11:13<1:02:39,  1.46it/s]
4|980|Loss: 0.9151755571365356:  15%|█▌        | 980/6470 [11:14<1:03:24,  1.44it/s]
4|981|Loss: 0.8900563716888428:  15%|█▌        | 980/6470 [11:14<1:03:24,  1.44it/s]
4|981|Loss: 0.8900563716888428:  15%|█▌        | 981/6470 [11:14<1:02:16,  1.47it/s]
4|982|Loss: 0.8442134857177734:  15%|█▌        | 981/6470 [11:15<1:02:16,  1.47it/s]
4|982|Loss: 0.8442134857177734:  15%|█▌        | 982/6470 [11:15<1:02:59,  1.45it/s]
4|983|Loss: 0.8546673655509949:  15%|█▌        | 982/6470 [11:15<1:02:59,  1.45it/s]
4|983|Loss: 0.8546673655509949:  15%|█▌        | 983/6470 [11:16<1:04:15,  1.42it/s]
4|984|Loss: 0.8062826991081238:  15%|█▌        | 983/6470 [11:16<1:04:15,  1.42it/s]
4|984|Loss: 0.8062826991081238:  15%|█▌        | 984/6470 [11:16<1:05:14,  1.40it/s]
4|985|Loss: 0.8191549777984619:  15%|█▌        | 984/6470 [11:17<1:05:14,  1.40it/s]
4|985|Loss: 0.8191549777984619:  15%|█▌        | 985/6470 [11:17<1:02:28,  1.46it/s]
4|986|Loss: 0.860664963722229:  15%|█▌        | 985/6470 [11:17<1:02:28,  1.46it/s] 
4|986|Loss: 0.860664963722229:  15%|█▌        | 986/6470 [11:18<1:00:47,  1.50it/s]
4|987|Loss: 0.7559508085250854:  15%|█▌        | 986/6470 [11:18<1:00:47,  1.50it/s]
4|987|Loss: 0.7559508085250854:  15%|█▌        | 987/6470 [11:18<1:01:24,  1.49it/s]
4|988|Loss: 0.7063226103782654:  15%|█▌        | 987/6470 [11:19<1:01:24,  1.49it/s]
4|988|Loss: 0.7063226103782654:  15%|█▌        | 988/6470 [11:19<1:00:57,  1.50it/s]
4|989|Loss: 0.7051473259925842:  15%|█▌        | 988/6470 [11:19<1:00:57,  1.50it/s]
4|989|Loss: 0.7051473259925842:  15%|█▌        | 989/6470 [11:20<1:04:28,  1.42it/s]
4|990|Loss: 0.710129976272583:  15%|█▌        | 989/6470 [11:20<1:04:28,  1.42it/s] 
4|990|Loss: 0.710129976272583:  15%|█▌        | 990/6470 [11:21<1:08:48,  1.33it/s]
4|991|Loss: 0.9522443413734436:  15%|█▌        | 990/6470 [11:21<1:08:48,  1.33it/s]
4|991|Loss: 0.9522443413734436:  15%|█▌        | 991/6470 [11:21<1:11:20,  1.28it/s]
4|992|Loss: 0.8729733824729919:  15%|█▌        | 991/6470 [11:22<1:11:20,  1.28it/s]
4|992|Loss: 0.8729733824729919:  15%|█▌        | 992/6470 [11:22<1:08:35,  1.33it/s]
4|993|Loss: 0.8163725137710571:  15%|█▌        | 992/6470 [11:23<1:08:35,  1.33it/s]
4|993|Loss: 0.8163725137710571:  15%|█▌        | 993/6470 [11:23<1:10:38,  1.29it/s]
4|994|Loss: 0.7660446763038635:  15%|█▌        | 993/6470 [11:23<1:10:38,  1.29it/s]
4|994|Loss: 0.7660446763038635:  15%|█▌        | 994/6470 [11:24<1:10:44,  1.29it/s]
4|995|Loss: 0.8593114018440247:  15%|█▌        | 994/6470 [11:24<1:10:44,  1.29it/s]
4|995|Loss: 0.8593114018440247:  15%|█▌        | 995/6470 [11:24<1:07:36,  1.35it/s]
4|996|Loss: 0.8526545166969299:  15%|█▌        | 995/6470 [11:25<1:07:36,  1.35it/s]
4|996|Loss: 0.8526545166969299:  15%|█▌        | 996/6470 [11:25<1:02:42,  1.45it/s]
4|997|Loss: 0.9579797983169556:  15%|█▌        | 996/6470 [11:25<1:02:42,  1.45it/s]
4|997|Loss: 0.9579797983169556:  15%|█▌        | 997/6470 [11:26<1:03:15,  1.44it/s]
4|998|Loss: 0.8550007939338684:  15%|█▌        | 997/6470 [11:26<1:03:15,  1.44it/s]
4|998|Loss: 0.8550007939338684:  15%|█▌        | 998/6470 [11:26<1:03:51,  1.43it/s]
4|999|Loss: 0.7915089726448059:  15%|█▌        | 998/6470 [11:27<1:03:51,  1.43it/s]
4|999|Loss: 0.7915089726448059:  15%|█▌        | 999/6470 [11:27<1:02:36,  1.46it/s]
4|1000|Loss: 0.9073920249938965:  15%|█▌        | 999/6470 [11:27<1:02:36,  1.46it/s]
4|1000|Loss: 0.9073920249938965:  15%|█▌        | 1000/6470 [11:28<58:53,  1.55it/s] 
4|1001|Loss: 0.8644552826881409:  15%|█▌        | 1000/6470 [11:28<58:53,  1.55it/s]
4|1001|Loss: 0.8644552826881409:  15%|█▌        | 1001/6470 [11:28<1:00:31,  1.51it/s]
4|1002|Loss: 0.8694905638694763:  15%|█▌        | 1001/6470 [11:29<1:00:31,  1.51it/s]
4|1002|Loss: 0.8694905638694763:  15%|█▌        | 1002/6470 [11:29<1:02:11,  1.47it/s]
4|1003|Loss: 0.8223936557769775:  15%|█▌        | 1002/6470 [11:29<1:02:11,  1.47it/s]
4|1003|Loss: 0.8223936557769775:  16%|█▌        | 1003/6470 [11:30<57:18,  1.59it/s]  
4|1004|Loss: 0.8544475436210632:  16%|█▌        | 1003/6470 [11:30<57:18,  1.59it/s]
4|1004|Loss: 0.8544475436210632:  16%|█▌        | 1004/6470 [11:30<58:31,  1.56it/s]
4|1005|Loss: 0.8284269571304321:  16%|█▌        | 1004/6470 [11:31<58:31,  1.56it/s]
4|1005|Loss: 0.8284269571304321:  16%|█▌        | 1005/6470 [11:31<1:02:34,  1.46it/s]
4|1006|Loss: 0.9000557661056519:  16%|█▌        | 1005/6470 [11:31<1:02:34,  1.46it/s]
4|1006|Loss: 0.9000557661056519:  16%|█▌        | 1006/6470 [11:32<1:02:18,  1.46it/s]
4|1007|Loss: 0.8060659766197205:  16%|█▌        | 1006/6470 [11:32<1:02:18,  1.46it/s]
4|1007|Loss: 0.8060659766197205:  16%|█▌        | 1007/6470 [11:32<1:01:06,  1.49it/s]
4|1008|Loss: 0.9749665260314941:  16%|█▌        | 1007/6470 [11:33<1:01:06,  1.49it/s]
4|1008|Loss: 0.9749665260314941:  16%|█▌        | 1008/6470 [11:33<1:03:35,  1.43it/s]
4|1009|Loss: 0.856551468372345:  16%|█▌        | 1008/6470 [11:33<1:03:35,  1.43it/s] 
4|1009|Loss: 0.856551468372345:  16%|█▌        | 1009/6470 [11:34<1:04:10,  1.42it/s]
4|1010|Loss: 0.9732398390769958:  16%|█▌        | 1009/6470 [11:34<1:04:10,  1.42it/s]
4|1010|Loss: 0.9732398390769958:  16%|█▌        | 1010/6470 [11:34<1:02:23,  1.46it/s]
4|1011|Loss: 0.7589432597160339:  16%|█▌        | 1010/6470 [11:35<1:02:23,  1.46it/s]
4|1011|Loss: 0.7589432597160339:  16%|█▌        | 1011/6470 [11:35<1:05:09,  1.40it/s]
4|1012|Loss: 0.8830009698867798:  16%|█▌        | 1011/6470 [11:36<1:05:09,  1.40it/s]
4|1012|Loss: 0.8830009698867798:  16%|█▌        | 1012/6470 [11:36<1:05:41,  1.38it/s]
4|1013|Loss: 0.8843675851821899:  16%|█▌        | 1012/6470 [11:36<1:05:41,  1.38it/s]
4|1013|Loss: 0.8843675851821899:  16%|█▌        | 1013/6470 [11:37<1:02:21,  1.46it/s]
4|1014|Loss: 0.7557592988014221:  16%|█▌        | 1013/6470 [11:37<1:02:21,  1.46it/s]
4|1014|Loss: 0.7557592988014221:  16%|█▌        | 1014/6470 [11:37<1:02:08,  1.46it/s]
4|1015|Loss: 0.8061057329177856:  16%|█▌        | 1014/6470 [11:38<1:02:08,  1.46it/s]
4|1015|Loss: 0.8061057329177856:  16%|█▌        | 1015/6470 [11:38<1:03:23,  1.43it/s]
4|1016|Loss: 0.815472424030304:  16%|█▌        | 1015/6470 [11:38<1:03:23,  1.43it/s] 
4|1016|Loss: 0.815472424030304:  16%|█▌        | 1016/6470 [11:39<59:49,  1.52it/s]  
4|1017|Loss: 0.8966578245162964:  16%|█▌        | 1016/6470 [11:39<59:49,  1.52it/s]
4|1017|Loss: 0.8966578245162964:  16%|█▌        | 1017/6470 [11:39<1:03:17,  1.44it/s]
4|1018|Loss: 0.8807060718536377:  16%|█▌        | 1017/6470 [11:40<1:03:17,  1.44it/s]
4|1018|Loss: 0.8807060718536377:  16%|█▌        | 1018/6470 [11:40<1:05:28,  1.39it/s]
4|1019|Loss: 0.7812473177909851:  16%|█▌        | 1018/6470 [11:40<1:05:28,  1.39it/s]
4|1019|Loss: 0.7812473177909851:  16%|█▌        | 1019/6470 [11:41<1:01:25,  1.48it/s]
4|1020|Loss: 0.8121165633201599:  16%|█▌        | 1019/6470 [11:41<1:01:25,  1.48it/s]
4|1020|Loss: 0.8121165633201599:  16%|█▌        | 1020/6470 [11:41<59:43,  1.52it/s]  
4|1021|Loss: 0.8792045712471008:  16%|█▌        | 1020/6470 [11:42<59:43,  1.52it/s]
4|1021|Loss: 0.8792045712471008:  16%|█▌        | 1021/6470 [11:42<1:02:43,  1.45it/s]
4|1022|Loss: 0.8416616916656494:  16%|█▌        | 1021/6470 [11:43<1:02:43,  1.45it/s]
4|1022|Loss: 0.8416616916656494:  16%|█▌        | 1022/6470 [11:43<1:06:25,  1.37it/s]
4|1023|Loss: 1.0071396827697754:  16%|█▌        | 1022/6470 [11:43<1:06:25,  1.37it/s]
4|1023|Loss: 1.0071396827697754:  16%|█▌        | 1023/6470 [11:44<1:03:44,  1.42it/s]
4|1024|Loss: 0.889798641204834:  16%|█▌        | 1023/6470 [11:44<1:03:44,  1.42it/s] 
4|1024|Loss: 0.889798641204834:  16%|█▌        | 1024/6470 [11:44<1:05:04,  1.39it/s]
4|1025|Loss: 0.6730508804321289:  16%|█▌        | 1024/6470 [11:45<1:05:04,  1.39it/s]
4|1025|Loss: 0.6730508804321289:  16%|█▌        | 1025/6470 [11:45<1:06:18,  1.37it/s]
4|1026|Loss: 0.7086597681045532:  16%|█▌        | 1025/6470 [11:45<1:06:18,  1.37it/s]
4|1026|Loss: 0.7086597681045532:  16%|█▌        | 1026/6470 [11:46<1:01:48,  1.47it/s]
4|1027|Loss: 0.9481022953987122:  16%|█▌        | 1026/6470 [11:46<1:01:48,  1.47it/s]
4|1027|Loss: 0.9481022953987122:  16%|█▌        | 1027/6470 [11:46<1:03:20,  1.43it/s]
4|1028|Loss: 0.8139882683753967:  16%|█▌        | 1027/6470 [11:47<1:03:20,  1.43it/s]
4|1028|Loss: 0.8139882683753967:  16%|█▌        | 1028/6470 [11:47<1:04:56,  1.40it/s]
4|1029|Loss: 0.8501840233802795:  16%|█▌        | 1028/6470 [11:48<1:04:56,  1.40it/s]
4|1029|Loss: 0.8501840233802795:  16%|█▌        | 1029/6470 [11:48<1:05:35,  1.38it/s]
4|1030|Loss: 0.742058277130127:  16%|█▌        | 1029/6470 [11:48<1:05:35,  1.38it/s] 
4|1030|Loss: 0.742058277130127:  16%|█▌        | 1030/6470 [11:49<1:05:45,  1.38it/s]
4|1031|Loss: 0.8874790668487549:  16%|█▌        | 1030/6470 [11:49<1:05:45,  1.38it/s]
4|1031|Loss: 0.8874790668487549:  16%|█▌        | 1031/6470 [11:49<1:06:03,  1.37it/s]
4|1032|Loss: 0.8896465301513672:  16%|█▌        | 1031/6470 [11:50<1:06:03,  1.37it/s]
4|1032|Loss: 0.8896465301513672:  16%|█▌        | 1032/6470 [11:50<1:06:15,  1.37it/s]
4|1033|Loss: 0.8674302697181702:  16%|█▌        | 1032/6470 [11:50<1:06:15,  1.37it/s]
4|1033|Loss: 0.8674302697181702:  16%|█▌        | 1033/6470 [11:51<1:06:14,  1.37it/s]
4|1034|Loss: 0.8745684623718262:  16%|█▌        | 1033/6470 [11:51<1:06:14,  1.37it/s]
4|1034|Loss: 0.8745684623718262:  16%|█▌        | 1034/6470 [11:51<1:05:35,  1.38it/s]
4|1035|Loss: 0.7436184287071228:  16%|█▌        | 1034/6470 [11:52<1:05:35,  1.38it/s]
4|1035|Loss: 0.7436184287071228:  16%|█▌        | 1035/6470 [11:52<1:04:54,  1.40it/s]
4|1036|Loss: 0.8595554232597351:  16%|█▌        | 1035/6470 [11:53<1:04:54,  1.40it/s]
4|1036|Loss: 0.8595554232597351:  16%|█▌        | 1036/6470 [11:53<1:05:11,  1.39it/s]
4|1037|Loss: 0.8879755139350891:  16%|█▌        | 1036/6470 [11:53<1:05:11,  1.39it/s]
4|1037|Loss: 0.8879755139350891:  16%|█▌        | 1037/6470 [11:54<1:05:49,  1.38it/s]
4|1038|Loss: 0.7538600564002991:  16%|█▌        | 1037/6470 [11:54<1:05:49,  1.38it/s]
4|1038|Loss: 0.7538600564002991:  16%|█▌        | 1038/6470 [11:55<1:10:06,  1.29it/s]
4|1039|Loss: 0.6613647937774658:  16%|█▌        | 1038/6470 [11:55<1:10:06,  1.29it/s]
4|1039|Loss: 0.6613647937774658:  16%|█▌        | 1039/6470 [11:55<1:07:37,  1.34it/s]
4|1040|Loss: 1.4183605909347534:  16%|█▌        | 1039/6470 [11:56<1:07:37,  1.34it/s]
4|1040|Loss: 1.4183605909347534:  16%|█▌        | 1040/6470 [11:56<1:08:06,  1.33it/s]
4|1041|Loss: 0.8437736630439758:  16%|█▌        | 1040/6470 [11:56<1:08:06,  1.33it/s]
4|1041|Loss: 0.8437736630439758:  16%|█▌        | 1041/6470 [11:57<1:07:22,  1.34it/s]
4|1042|Loss: 0.6906378269195557:  16%|█▌        | 1041/6470 [11:57<1:07:22,  1.34it/s]
4|1042|Loss: 0.6906378269195557:  16%|█▌        | 1042/6470 [11:57<1:00:46,  1.49it/s]
4|1043|Loss: 0.9026693105697632:  16%|█▌        | 1042/6470 [11:58<1:00:46,  1.49it/s]
4|1043|Loss: 0.9026693105697632:  16%|█▌        | 1043/6470 [11:58<1:02:01,  1.46it/s]
4|1044|Loss: 0.5871402621269226:  16%|█▌        | 1043/6470 [11:59<1:02:01,  1.46it/s]
4|1044|Loss: 0.5871402621269226:  16%|█▌        | 1044/6470 [11:59<1:22:38,  1.09it/s]
4|1045|Loss: 0.787755012512207:  16%|█▌        | 1044/6470 [12:00<1:22:38,  1.09it/s] 
4|1045|Loss: 0.787755012512207:  16%|█▌        | 1045/6470 [12:00<1:20:43,  1.12it/s]
4|1046|Loss: 0.757678747177124:  16%|█▌        | 1045/6470 [12:01<1:20:43,  1.12it/s]
4|1046|Loss: 0.757678747177124:  16%|█▌        | 1046/6470 [12:01<1:17:03,  1.17it/s]
4|1047|Loss: 0.7970135807991028:  16%|█▌        | 1046/6470 [12:01<1:17:03,  1.17it/s]
4|1047|Loss: 0.7970135807991028:  16%|█▌        | 1047/6470 [12:02<1:11:53,  1.26it/s]
4|1048|Loss: 0.8154210448265076:  16%|█▌        | 1047/6470 [12:02<1:11:53,  1.26it/s]
4|1048|Loss: 0.8154210448265076:  16%|█▌        | 1048/6470 [12:02<1:08:59,  1.31it/s]
4|1049|Loss: 0.8011289834976196:  16%|█▌        | 1048/6470 [12:03<1:08:59,  1.31it/s]
4|1049|Loss: 0.8011289834976196:  16%|█▌        | 1049/6470 [12:03<1:07:24,  1.34it/s]
4|1050|Loss: 0.7860892415046692:  16%|█▌        | 1049/6470 [12:03<1:07:24,  1.34it/s]
4|1050|Loss: 0.7860892415046692:  16%|█▌        | 1050/6470 [12:04<1:09:40,  1.30it/s]
4|1051|Loss: 0.7397510409355164:  16%|█▌        | 1050/6470 [12:04<1:09:40,  1.30it/s]
4|1051|Loss: 0.7397510409355164:  16%|█▌        | 1051/6470 [12:04<1:01:22,  1.47it/s]
4|1052|Loss: 0.8264791369438171:  16%|█▌        | 1051/6470 [12:05<1:01:22,  1.47it/s]
4|1052|Loss: 0.8264791369438171:  16%|█▋        | 1052/6470 [12:05<54:43,  1.65it/s]  
4|1053|Loss: 0.9141926765441895:  16%|█▋        | 1052/6470 [12:05<54:43,  1.65it/s]
4|1053|Loss: 0.9141926765441895:  16%|█▋        | 1053/6470 [12:05<51:45,  1.74it/s]
4|1054|Loss: 0.8738675713539124:  16%|█▋        | 1053/6470 [12:06<51:45,  1.74it/s]
4|1054|Loss: 0.8738675713539124:  16%|█▋        | 1054/6470 [12:06<58:07,  1.55it/s]
4|1055|Loss: 0.8015424609184265:  16%|█▋        | 1054/6470 [12:06<58:07,  1.55it/s]
4|1055|Loss: 0.8015424609184265:  16%|█▋        | 1055/6470 [12:07<56:43,  1.59it/s]
4|1056|Loss: 0.9238875508308411:  16%|█▋        | 1055/6470 [12:07<56:43,  1.59it/s]
4|1056|Loss: 0.9238875508308411:  16%|█▋        | 1056/6470 [12:07<58:08,  1.55it/s]
4|1057|Loss: 0.8496376872062683:  16%|█▋        | 1056/6470 [12:08<58:08,  1.55it/s]
4|1057|Loss: 0.8496376872062683:  16%|█▋        | 1057/6470 [12:08<58:42,  1.54it/s]
4|1058|Loss: 0.8285358548164368:  16%|█▋        | 1057/6470 [12:08<58:42,  1.54it/s]
4|1058|Loss: 0.8285358548164368:  16%|█▋        | 1058/6470 [12:09<58:00,  1.55it/s]
4|1059|Loss: 0.8042712211608887:  16%|█▋        | 1058/6470 [12:09<58:00,  1.55it/s]
4|1059|Loss: 0.8042712211608887:  16%|█▋        | 1059/6470 [12:09<56:28,  1.60it/s]
4|1060|Loss: 0.8650579452514648:  16%|█▋        | 1059/6470 [12:10<56:28,  1.60it/s]
4|1060|Loss: 0.8650579452514648:  16%|█▋        | 1060/6470 [12:10<1:01:58,  1.45it/s]
4|1061|Loss: 0.9422592520713806:  16%|█▋        | 1060/6470 [12:11<1:01:58,  1.45it/s]
4|1061|Loss: 0.9422592520713806:  16%|█▋        | 1061/6470 [12:11<1:06:17,  1.36it/s]
4|1062|Loss: 0.7497854828834534:  16%|█▋        | 1061/6470 [12:11<1:06:17,  1.36it/s]
4|1062|Loss: 0.7497854828834534:  16%|█▋        | 1062/6470 [12:12<1:04:58,  1.39it/s]
4|1063|Loss: 0.8133577108383179:  16%|█▋        | 1062/6470 [12:12<1:04:58,  1.39it/s]
4|1063|Loss: 0.8133577108383179:  16%|█▋        | 1063/6470 [12:12<1:06:58,  1.35it/s]
4|1064|Loss: 0.8083115220069885:  16%|█▋        | 1063/6470 [12:13<1:06:58,  1.35it/s]
4|1064|Loss: 0.8083115220069885:  16%|█▋        | 1064/6470 [12:13<1:01:52,  1.46it/s]
4|1065|Loss: 0.7845370173454285:  16%|█▋        | 1064/6470 [12:13<1:01:52,  1.46it/s]
4|1065|Loss: 0.7845370173454285:  16%|█▋        | 1065/6470 [12:14<59:30,  1.51it/s]  
4|1066|Loss: 0.8898282051086426:  16%|█▋        | 1065/6470 [12:14<59:30,  1.51it/s]
4|1066|Loss: 0.8898282051086426:  16%|█▋        | 1066/6470 [12:14<1:01:15,  1.47it/s]
4|1067|Loss: 0.8817991018295288:  16%|█▋        | 1066/6470 [12:15<1:01:15,  1.47it/s]
4|1067|Loss: 0.8817991018295288:  16%|█▋        | 1067/6470 [12:15<56:51,  1.58it/s]  
4|1068|Loss: 0.7938597798347473:  16%|█▋        | 1067/6470 [12:15<56:51,  1.58it/s]
4|1068|Loss: 0.7938597798347473:  17%|█▋        | 1068/6470 [12:15<54:06,  1.66it/s]
4|1069|Loss: 0.9238497018814087:  17%|█▋        | 1068/6470 [12:16<54:06,  1.66it/s]
4|1069|Loss: 0.9238497018814087:  17%|█▋        | 1069/6470 [12:16<56:06,  1.60it/s]
4|1070|Loss: 0.9691464900970459:  17%|█▋        | 1069/6470 [12:16<56:06,  1.60it/s]
4|1070|Loss: 0.9691464900970459:  17%|█▋        | 1070/6470 [12:17<54:58,  1.64it/s]
4|1071|Loss: 0.7230635285377502:  17%|█▋        | 1070/6470 [12:17<54:58,  1.64it/s]
4|1071|Loss: 0.7230635285377502:  17%|█▋        | 1071/6470 [12:17<55:12,  1.63it/s]
4|1072|Loss: 0.7176463603973389:  17%|█▋        | 1071/6470 [12:18<55:12,  1.63it/s]
4|1072|Loss: 0.7176463603973389:  17%|█▋        | 1072/6470 [12:18<57:05,  1.58it/s]
4|1073|Loss: 0.8159472942352295:  17%|█▋        | 1072/6470 [12:18<57:05,  1.58it/s]
4|1073|Loss: 0.8159472942352295:  17%|█▋        | 1073/6470 [12:19<1:01:45,  1.46it/s]
4|1074|Loss: 0.7316073775291443:  17%|█▋        | 1073/6470 [12:19<1:01:45,  1.46it/s]
4|1074|Loss: 0.7316073775291443:  17%|█▋        | 1074/6470 [12:20<1:06:08,  1.36it/s]
4|1075|Loss: 0.7668071985244751:  17%|█▋        | 1074/6470 [12:20<1:06:08,  1.36it/s]
4|1075|Loss: 0.7668071985244751:  17%|█▋        | 1075/6470 [12:20<1:04:56,  1.38it/s]
4|1076|Loss: 0.8717873096466064:  17%|█▋        | 1075/6470 [12:21<1:04:56,  1.38it/s]
4|1076|Loss: 0.8717873096466064:  17%|█▋        | 1076/6470 [12:21<1:02:45,  1.43it/s]
4|1077|Loss: 0.8923238515853882:  17%|█▋        | 1076/6470 [12:21<1:02:45,  1.43it/s]
4|1077|Loss: 0.8923238515853882:  17%|█▋        | 1077/6470 [12:21<59:53,  1.50it/s]  
4|1078|Loss: 0.8956469297409058:  17%|█▋        | 1077/6470 [12:22<59:53,  1.50it/s]
4|1078|Loss: 0.8956469297409058:  17%|█▋        | 1078/6470 [12:22<1:00:47,  1.48it/s]
4|1079|Loss: 0.7675153613090515:  17%|█▋        | 1078/6470 [12:23<1:00:47,  1.48it/s]
4|1079|Loss: 0.7675153613090515:  17%|█▋        | 1079/6470 [12:23<59:06,  1.52it/s]  
4|1080|Loss: 0.804187536239624:  17%|█▋        | 1079/6470 [12:23<59:06,  1.52it/s] 
4|1080|Loss: 0.804187536239624:  17%|█▋        | 1080/6470 [12:23<58:07,  1.55it/s]
4|1081|Loss: 0.7050427198410034:  17%|█▋        | 1080/6470 [12:24<58:07,  1.55it/s]
4|1081|Loss: 0.7050427198410034:  17%|█▋        | 1081/6470 [12:24<59:58,  1.50it/s]
4|1082|Loss: 0.8238878846168518:  17%|█▋        | 1081/6470 [12:25<59:58,  1.50it/s]
4|1082|Loss: 0.8238878846168518:  17%|█▋        | 1082/6470 [12:25<1:01:45,  1.45it/s]
4|1083|Loss: 0.9786762595176697:  17%|█▋        | 1082/6470 [12:25<1:01:45,  1.45it/s]
4|1083|Loss: 0.9786762595176697:  17%|█▋        | 1083/6470 [12:25<1:01:02,  1.47it/s]
4|1084|Loss: 0.9393355846405029:  17%|█▋        | 1083/6470 [12:26<1:01:02,  1.47it/s]
4|1084|Loss: 0.9393355846405029:  17%|█▋        | 1084/6470 [12:26<1:05:15,  1.38it/s]
4|1085|Loss: 0.8766855597496033:  17%|█▋        | 1084/6470 [12:27<1:05:15,  1.38it/s]
4|1085|Loss: 0.8766855597496033:  17%|█▋        | 1085/6470 [12:27<1:03:22,  1.42it/s]
4|1086|Loss: 0.8884727954864502:  17%|█▋        | 1085/6470 [12:27<1:03:22,  1.42it/s]
4|1086|Loss: 0.8884727954864502:  17%|█▋        | 1086/6470 [12:28<1:02:40,  1.43it/s]
4|1087|Loss: 0.8285242915153503:  17%|█▋        | 1086/6470 [12:28<1:02:40,  1.43it/s]
4|1087|Loss: 0.8285242915153503:  17%|█▋        | 1087/6470 [12:28<1:06:02,  1.36it/s]
4|1088|Loss: 0.8804519176483154:  17%|█▋        | 1087/6470 [12:29<1:06:02,  1.36it/s]
4|1088|Loss: 0.8804519176483154:  17%|█▋        | 1088/6470 [12:29<1:07:00,  1.34it/s]
4|1089|Loss: 0.8296354413032532:  17%|█▋        | 1088/6470 [12:30<1:07:00,  1.34it/s]
4|1089|Loss: 0.8296354413032532:  17%|█▋        | 1089/6470 [12:30<1:07:59,  1.32it/s]
4|1090|Loss: 0.8598198294639587:  17%|█▋        | 1089/6470 [12:30<1:07:59,  1.32it/s]
4|1090|Loss: 0.8598198294639587:  17%|█▋        | 1090/6470 [12:31<1:02:19,  1.44it/s]
4|1091|Loss: 0.8972225189208984:  17%|█▋        | 1090/6470 [12:31<1:02:19,  1.44it/s]
4|1091|Loss: 0.8972225189208984:  17%|█▋        | 1091/6470 [12:31<1:04:39,  1.39it/s]
4|1092|Loss: 0.7049787640571594:  17%|█▋        | 1091/6470 [12:32<1:04:39,  1.39it/s]
4|1092|Loss: 0.7049787640571594:  17%|█▋        | 1092/6470 [12:32<1:05:03,  1.38it/s]
4|1093|Loss: 0.8491768836975098:  17%|█▋        | 1092/6470 [12:32<1:05:03,  1.38it/s]
4|1093|Loss: 0.8491768836975098:  17%|█▋        | 1093/6470 [12:33<1:03:11,  1.42it/s]
4|1094|Loss: 0.8838499188423157:  17%|█▋        | 1093/6470 [12:33<1:03:11,  1.42it/s]
4|1094|Loss: 0.8838499188423157:  17%|█▋        | 1094/6470 [12:34<1:06:08,  1.35it/s]
4|1095|Loss: 0.6575332880020142:  17%|█▋        | 1094/6470 [12:34<1:06:08,  1.35it/s]
4|1095|Loss: 0.6575332880020142:  17%|█▋        | 1095/6470 [12:34<1:09:23,  1.29it/s]
4|1096|Loss: 0.790121853351593:  17%|█▋        | 1095/6470 [12:35<1:09:23,  1.29it/s] 
4|1096|Loss: 0.790121853351593:  17%|█▋        | 1096/6470 [12:35<1:09:32,  1.29it/s]
4|1097|Loss: 0.7943540215492249:  17%|█▋        | 1096/6470 [12:36<1:09:32,  1.29it/s]
4|1097|Loss: 0.7943540215492249:  17%|█▋        | 1097/6470 [12:36<1:08:33,  1.31it/s]
4|1098|Loss: 0.8272974491119385:  17%|█▋        | 1097/6470 [12:36<1:08:33,  1.31it/s]
4|1098|Loss: 0.8272974491119385:  17%|█▋        | 1098/6470 [12:37<1:09:44,  1.28it/s]
4|1099|Loss: 1.119917392730713:  17%|█▋        | 1098/6470 [12:37<1:09:44,  1.28it/s] 
4|1099|Loss: 1.119917392730713:  17%|█▋        | 1099/6470 [12:38<1:10:29,  1.27it/s]
4|1100|Loss: 0.8233281373977661:  17%|█▋        | 1099/6470 [12:38<1:10:29,  1.27it/s]
4|1100|Loss: 0.8233281373977661:  17%|█▋        | 1100/6470 [12:38<1:04:08,  1.40it/s]
4|1101|Loss: 0.7062161564826965:  17%|█▋        | 1100/6470 [12:38<1:04:08,  1.40it/s]
4|1101|Loss: 0.7062161564826965:  17%|█▋        | 1101/6470 [12:39<1:00:01,  1.49it/s]
4|1102|Loss: 0.9539668560028076:  17%|█▋        | 1101/6470 [12:39<1:00:01,  1.49it/s]
4|1102|Loss: 0.9539668560028076:  17%|█▋        | 1102/6470 [12:39<54:49,  1.63it/s]  
4|1103|Loss: 1.0186635255813599:  17%|█▋        | 1102/6470 [12:40<54:49,  1.63it/s]
4|1103|Loss: 1.0186635255813599:  17%|█▋        | 1103/6470 [12:40<56:16,  1.59it/s]
4|1104|Loss: 0.8610950112342834:  17%|█▋        | 1103/6470 [12:40<56:16,  1.59it/s]
4|1104|Loss: 0.8610950112342834:  17%|█▋        | 1104/6470 [12:41<59:08,  1.51it/s]
4|1105|Loss: 1.009630560874939:  17%|█▋        | 1104/6470 [12:41<59:08,  1.51it/s] 
4|1105|Loss: 1.009630560874939:  17%|█▋        | 1105/6470 [12:41<1:00:42,  1.47it/s]
4|1106|Loss: 0.9097191691398621:  17%|█▋        | 1105/6470 [12:42<1:00:42,  1.47it/s]
4|1106|Loss: 0.9097191691398621:  17%|█▋        | 1106/6470 [12:42<1:01:18,  1.46it/s]
4|1107|Loss: 0.8251364827156067:  17%|█▋        | 1106/6470 [12:42<1:01:18,  1.46it/s]
4|1107|Loss: 0.8251364827156067:  17%|█▋        | 1107/6470 [12:43<1:02:21,  1.43it/s]
4|1108|Loss: 0.8348315358161926:  17%|█▋        | 1107/6470 [12:43<1:02:21,  1.43it/s]
4|1108|Loss: 0.8348315358161926:  17%|█▋        | 1108/6470 [12:43<54:57,  1.63it/s]  
4|1109|Loss: 0.8552108407020569:  17%|█▋        | 1108/6470 [12:43<54:57,  1.63it/s]
4|1109|Loss: 0.8552108407020569:  17%|█▋        | 1109/6470 [12:44<56:02,  1.59it/s]
4|1110|Loss: 1.003089427947998:  17%|█▋        | 1109/6470 [12:44<56:02,  1.59it/s] 
4|1110|Loss: 1.003089427947998:  17%|█▋        | 1110/6470 [12:45<1:01:41,  1.45it/s]
4|1111|Loss: 0.8357264995574951:  17%|█▋        | 1110/6470 [12:45<1:01:41,  1.45it/s]
4|1111|Loss: 0.8357264995574951:  17%|█▋        | 1111/6470 [12:45<1:01:23,  1.46it/s]
4|1112|Loss: 0.9836392402648926:  17%|█▋        | 1111/6470 [12:46<1:01:23,  1.46it/s]
4|1112|Loss: 0.9836392402648926:  17%|█▋        | 1112/6470 [12:46<1:02:10,  1.44it/s]
4|1113|Loss: 0.8291463255882263:  17%|█▋        | 1112/6470 [12:46<1:02:10,  1.44it/s]
4|1113|Loss: 0.8291463255882263:  17%|█▋        | 1113/6470 [12:47<1:02:49,  1.42it/s]
4|1114|Loss: 0.8289681077003479:  17%|█▋        | 1113/6470 [12:47<1:02:49,  1.42it/s]
4|1114|Loss: 0.8289681077003479:  17%|█▋        | 1114/6470 [12:48<1:03:38,  1.40it/s]
4|1115|Loss: 0.8648988604545593:  17%|█▋        | 1114/6470 [12:48<1:03:38,  1.40it/s]
4|1115|Loss: 0.8648988604545593:  17%|█▋        | 1115/6470 [12:48<1:07:19,  1.33it/s]
4|1116|Loss: 0.7272676229476929:  17%|█▋        | 1115/6470 [12:49<1:07:19,  1.33it/s]
4|1116|Loss: 0.7272676229476929:  17%|█▋        | 1116/6470 [12:49<1:05:50,  1.36it/s]
4|1117|Loss: 0.7011650204658508:  17%|█▋        | 1116/6470 [12:49<1:05:50,  1.36it/s]
4|1117|Loss: 0.7011650204658508:  17%|█▋        | 1117/6470 [12:50<1:06:37,  1.34it/s]
4|1118|Loss: 0.9045040607452393:  17%|█▋        | 1117/6470 [12:50<1:06:37,  1.34it/s]
4|1118|Loss: 0.9045040607452393:  17%|█▋        | 1118/6470 [12:50<1:03:25,  1.41it/s]
4|1119|Loss: 0.8953994512557983:  17%|█▋        | 1118/6470 [12:51<1:03:25,  1.41it/s]
4|1119|Loss: 0.8953994512557983:  17%|█▋        | 1119/6470 [12:51<1:02:21,  1.43it/s]
4|1120|Loss: 0.8986858129501343:  17%|█▋        | 1119/6470 [12:51<1:02:21,  1.43it/s]
4|1120|Loss: 0.8986858129501343:  17%|█▋        | 1120/6470 [12:52<1:01:17,  1.45it/s]
4|1121|Loss: 0.7771047949790955:  17%|█▋        | 1120/6470 [12:52<1:01:17,  1.45it/s]
4|1121|Loss: 0.7771047949790955:  17%|█▋        | 1121/6470 [12:52<59:38,  1.49it/s]  
4|1122|Loss: 0.837285578250885:  17%|█▋        | 1121/6470 [12:53<59:38,  1.49it/s] 
4|1122|Loss: 0.837285578250885:  17%|█▋        | 1122/6470 [12:53<1:00:08,  1.48it/s]
4|1123|Loss: 0.9321138858795166:  17%|█▋        | 1122/6470 [12:53<1:00:08,  1.48it/s]
4|1123|Loss: 0.9321138858795166:  17%|█▋        | 1123/6470 [12:54<1:01:03,  1.46it/s]
4|1124|Loss: 0.9149914383888245:  17%|█▋        | 1123/6470 [12:54<1:01:03,  1.46it/s]
4|1124|Loss: 0.9149914383888245:  17%|█▋        | 1124/6470 [12:55<1:02:41,  1.42it/s]
4|1125|Loss: 0.9647625088691711:  17%|█▋        | 1124/6470 [12:55<1:02:41,  1.42it/s]
4|1125|Loss: 0.9647625088691711:  17%|█▋        | 1125/6470 [12:55<1:06:58,  1.33it/s]
4|1126|Loss: 0.962514340877533:  17%|█▋        | 1125/6470 [12:56<1:06:58,  1.33it/s] 
4|1126|Loss: 0.962514340877533:  17%|█▋        | 1126/6470 [12:56<1:09:35,  1.28it/s]
4|1127|Loss: 0.7988858819007874:  17%|█▋        | 1126/6470 [12:57<1:09:35,  1.28it/s]
4|1127|Loss: 0.7988858819007874:  17%|█▋        | 1127/6470 [12:57<1:04:08,  1.39it/s]
4|1128|Loss: 0.9397700428962708:  17%|█▋        | 1127/6470 [12:57<1:04:08,  1.39it/s]
4|1128|Loss: 0.9397700428962708:  17%|█▋        | 1128/6470 [12:57<1:01:11,  1.45it/s]
4|1129|Loss: 0.8813607096672058:  17%|█▋        | 1128/6470 [12:58<1:01:11,  1.45it/s]
4|1129|Loss: 0.8813607096672058:  17%|█▋        | 1129/6470 [12:58<1:01:13,  1.45it/s]
4|1130|Loss: 0.8182563185691833:  17%|█▋        | 1129/6470 [12:59<1:01:13,  1.45it/s]
4|1130|Loss: 0.8182563185691833:  17%|█▋        | 1130/6470 [12:59<1:01:40,  1.44it/s]
4|1131|Loss: 0.8418611288070679:  17%|█▋        | 1130/6470 [12:59<1:01:40,  1.44it/s]
4|1131|Loss: 0.8418611288070679:  17%|█▋        | 1131/6470 [13:00<1:04:41,  1.38it/s]
4|1132|Loss: 1.0103569030761719:  17%|█▋        | 1131/6470 [13:00<1:04:41,  1.38it/s]
4|1132|Loss: 1.0103569030761719:  17%|█▋        | 1132/6470 [13:00<1:05:41,  1.35it/s]
4|1133|Loss: 0.9420429468154907:  17%|█▋        | 1132/6470 [13:01<1:05:41,  1.35it/s]
4|1133|Loss: 0.9420429468154907:  18%|█▊        | 1133/6470 [13:01<1:03:12,  1.41it/s]
4|1134|Loss: 0.7389736771583557:  18%|█▊        | 1133/6470 [13:01<1:03:12,  1.41it/s]
4|1134|Loss: 0.7389736771583557:  18%|█▊        | 1134/6470 [13:02<1:05:07,  1.37it/s]
4|1135|Loss: 0.7355836033821106:  18%|█▊        | 1134/6470 [13:02<1:05:07,  1.37it/s]
4|1135|Loss: 0.7355836033821106:  18%|█▊        | 1135/6470 [13:03<1:03:19,  1.40it/s]
4|1136|Loss: 0.8007780909538269:  18%|█▊        | 1135/6470 [13:03<1:03:19,  1.40it/s]
4|1136|Loss: 0.8007780909538269:  18%|█▊        | 1136/6470 [13:03<1:04:29,  1.38it/s]
4|1137|Loss: 0.6964758634567261:  18%|█▊        | 1136/6470 [13:04<1:04:29,  1.38it/s]
4|1137|Loss: 0.6964758634567261:  18%|█▊        | 1137/6470 [13:04<1:06:21,  1.34it/s]
4|1138|Loss: 0.6445840001106262:  18%|█▊        | 1137/6470 [13:04<1:06:21,  1.34it/s]
4|1138|Loss: 0.6445840001106262:  18%|█▊        | 1138/6470 [13:05<1:06:44,  1.33it/s]
4|1139|Loss: 0.7160335779190063:  18%|█▊        | 1138/6470 [13:05<1:06:44,  1.33it/s]
4|1139|Loss: 0.7160335779190063:  18%|█▊        | 1139/6470 [13:06<1:05:52,  1.35it/s]
4|1140|Loss: 0.8373574614524841:  18%|█▊        | 1139/6470 [13:06<1:05:52,  1.35it/s]
4|1140|Loss: 0.8373574614524841:  18%|█▊        | 1140/6470 [13:06<1:08:22,  1.30it/s]
4|1141|Loss: 0.8431854248046875:  18%|█▊        | 1140/6470 [13:07<1:08:22,  1.30it/s]
4|1141|Loss: 0.8431854248046875:  18%|█▊        | 1141/6470 [13:07<1:03:38,  1.40it/s]
4|1142|Loss: 0.8427936434745789:  18%|█▊        | 1141/6470 [13:07<1:03:38,  1.40it/s]
4|1142|Loss: 0.8427936434745789:  18%|█▊        | 1142/6470 [13:08<1:02:15,  1.43it/s]
4|1143|Loss: 0.8503150939941406:  18%|█▊        | 1142/6470 [13:08<1:02:15,  1.43it/s]
4|1143|Loss: 0.8503150939941406:  18%|█▊        | 1143/6470 [13:08<1:04:43,  1.37it/s]
4|1144|Loss: 0.9358304142951965:  18%|█▊        | 1143/6470 [13:09<1:04:43,  1.37it/s]
4|1144|Loss: 0.9358304142951965:  18%|█▊        | 1144/6470 [13:09<1:07:25,  1.32it/s]
4|1145|Loss: 0.8872235417366028:  18%|█▊        | 1144/6470 [13:10<1:07:25,  1.32it/s]
4|1145|Loss: 0.8872235417366028:  18%|█▊        | 1145/6470 [13:10<1:06:43,  1.33it/s]
4|1146|Loss: 0.7179492712020874:  18%|█▊        | 1145/6470 [13:10<1:06:43,  1.33it/s]
4|1146|Loss: 0.7179492712020874:  18%|█▊        | 1146/6470 [13:11<1:03:31,  1.40it/s]
4|1147|Loss: 0.816292405128479:  18%|█▊        | 1146/6470 [13:11<1:03:31,  1.40it/s] 
4|1147|Loss: 0.816292405128479:  18%|█▊        | 1147/6470 [13:11<1:00:13,  1.47it/s]
4|1148|Loss: 0.8620112538337708:  18%|█▊        | 1147/6470 [13:12<1:00:13,  1.47it/s]
4|1148|Loss: 0.8620112538337708:  18%|█▊        | 1148/6470 [13:12<1:00:51,  1.46it/s]
4|1149|Loss: 0.79555344581604:  18%|█▊        | 1148/6470 [13:12<1:00:51,  1.46it/s]  
4|1149|Loss: 0.79555344581604:  18%|█▊        | 1149/6470 [13:13<59:35,  1.49it/s]  
4|1150|Loss: 0.8160027861595154:  18%|█▊        | 1149/6470 [13:13<59:35,  1.49it/s]
4|1150|Loss: 0.8160027861595154:  18%|█▊        | 1150/6470 [13:13<56:13,  1.58it/s]
4|1151|Loss: 1.04756760597229:  18%|█▊        | 1150/6470 [13:13<56:13,  1.58it/s]  
4|1151|Loss: 1.04756760597229:  18%|█▊        | 1151/6470 [13:14<55:16,  1.60it/s]
4|1152|Loss: 0.7874574661254883:  18%|█▊        | 1151/6470 [13:14<55:16,  1.60it/s]
4|1152|Loss: 0.7874574661254883:  18%|█▊        | 1152/6470 [13:14<55:30,  1.60it/s]
4|1153|Loss: 0.8902987241744995:  18%|█▊        | 1152/6470 [13:15<55:30,  1.60it/s]
4|1153|Loss: 0.8902987241744995:  18%|█▊        | 1153/6470 [13:15<53:27,  1.66it/s]
4|1154|Loss: 0.9705976247787476:  18%|█▊        | 1153/6470 [13:15<53:27,  1.66it/s]
4|1154|Loss: 0.9705976247787476:  18%|█▊        | 1154/6470 [13:16<55:48,  1.59it/s]
4|1155|Loss: 0.8868423104286194:  18%|█▊        | 1154/6470 [13:16<55:48,  1.59it/s]
4|1155|Loss: 0.8868423104286194:  18%|█▊        | 1155/6470 [13:16<1:00:32,  1.46it/s]
4|1156|Loss: 0.876124382019043:  18%|█▊        | 1155/6470 [13:17<1:00:32,  1.46it/s] 
4|1156|Loss: 0.876124382019043:  18%|█▊        | 1156/6470 [13:17<1:01:42,  1.44it/s]
4|1157|Loss: 0.7946659326553345:  18%|█▊        | 1156/6470 [13:17<1:01:42,  1.44it/s]
4|1157|Loss: 0.7946659326553345:  18%|█▊        | 1157/6470 [13:18<1:00:38,  1.46it/s]
4|1158|Loss: 0.958156406879425:  18%|█▊        | 1157/6470 [13:18<1:00:38,  1.46it/s] 
4|1158|Loss: 0.958156406879425:  18%|█▊        | 1158/6470 [13:18<59:03,  1.50it/s]  
4|1159|Loss: 0.7747871279716492:  18%|█▊        | 1158/6470 [13:19<59:03,  1.50it/s]
4|1159|Loss: 0.7747871279716492:  18%|█▊        | 1159/6470 [13:19<1:05:08,  1.36it/s]
4|1160|Loss: 0.7178664803504944:  18%|█▊        | 1159/6470 [13:20<1:05:08,  1.36it/s]
4|1160|Loss: 0.7178664803504944:  18%|█▊        | 1160/6470 [13:20<1:00:41,  1.46it/s]
4|1161|Loss: 0.7209776043891907:  18%|█▊        | 1160/6470 [13:20<1:00:41,  1.46it/s]
4|1161|Loss: 0.7209776043891907:  18%|█▊        | 1161/6470 [13:20<56:56,  1.55it/s]  
4|1162|Loss: 1.0040161609649658:  18%|█▊        | 1161/6470 [13:21<56:56,  1.55it/s]
4|1162|Loss: 1.0040161609649658:  18%|█▊        | 1162/6470 [13:21<56:28,  1.57it/s]
4|1163|Loss: 0.8755653500556946:  18%|█▊        | 1162/6470 [13:21<56:28,  1.57it/s]
4|1163|Loss: 0.8755653500556946:  18%|█▊        | 1163/6470 [13:22<58:23,  1.51it/s]
4|1164|Loss: 0.8158448338508606:  18%|█▊        | 1163/6470 [13:22<58:23,  1.51it/s]
4|1164|Loss: 0.8158448338508606:  18%|█▊        | 1164/6470 [13:22<1:00:33,  1.46it/s]
4|1165|Loss: 0.9013898372650146:  18%|█▊        | 1164/6470 [13:23<1:00:33,  1.46it/s]
4|1165|Loss: 0.9013898372650146:  18%|█▊        | 1165/6470 [13:23<1:01:25,  1.44it/s]
4|1166|Loss: 0.7900277972221375:  18%|█▊        | 1165/6470 [13:24<1:01:25,  1.44it/s]
4|1166|Loss: 0.7900277972221375:  18%|█▊        | 1166/6470 [13:24<1:05:18,  1.35it/s]
4|1167|Loss: 0.9059910774230957:  18%|█▊        | 1166/6470 [13:24<1:05:18,  1.35it/s]
4|1167|Loss: 0.9059910774230957:  18%|█▊        | 1167/6470 [13:25<1:03:28,  1.39it/s]
4|1168|Loss: 0.8099995255470276:  18%|█▊        | 1167/6470 [13:25<1:03:28,  1.39it/s]
4|1168|Loss: 0.8099995255470276:  18%|█▊        | 1168/6470 [13:26<1:05:14,  1.35it/s]
4|1169|Loss: 0.8415782451629639:  18%|█▊        | 1168/6470 [13:26<1:05:14,  1.35it/s]
4|1169|Loss: 0.8415782451629639:  18%|█▊        | 1169/6470 [13:26<1:04:56,  1.36it/s]
4|1170|Loss: 0.8219819068908691:  18%|█▊        | 1169/6470 [13:27<1:04:56,  1.36it/s]
4|1170|Loss: 0.8219819068908691:  18%|█▊        | 1170/6470 [13:27<1:05:12,  1.35it/s]
4|1171|Loss: 0.8748964667320251:  18%|█▊        | 1170/6470 [13:27<1:05:12,  1.35it/s]
4|1171|Loss: 0.8748964667320251:  18%|█▊        | 1171/6470 [13:28<1:00:57,  1.45it/s]
4|1172|Loss: 0.6781657934188843:  18%|█▊        | 1171/6470 [13:28<1:00:57,  1.45it/s]
4|1172|Loss: 0.6781657934188843:  18%|█▊        | 1172/6470 [13:28<58:43,  1.50it/s]  
4|1173|Loss: 0.840421199798584:  18%|█▊        | 1172/6470 [13:29<58:43,  1.50it/s] 
4|1173|Loss: 0.840421199798584:  18%|█▊        | 1173/6470 [13:29<59:16,  1.49it/s]
4|1174|Loss: 0.8485116958618164:  18%|█▊        | 1173/6470 [13:29<59:16,  1.49it/s]
4|1174|Loss: 0.8485116958618164:  18%|█▊        | 1174/6470 [13:30<58:53,  1.50it/s]
4|1175|Loss: 0.9261621832847595:  18%|█▊        | 1174/6470 [13:30<58:53,  1.50it/s]
4|1175|Loss: 0.9261621832847595:  18%|█▊        | 1175/6470 [13:30<1:00:25,  1.46it/s]
4|1176|Loss: 0.8478693962097168:  18%|█▊        | 1175/6470 [13:31<1:00:25,  1.46it/s]
4|1176|Loss: 0.8478693962097168:  18%|█▊        | 1176/6470 [13:31<58:20,  1.51it/s]  
4|1177|Loss: 0.949170708656311:  18%|█▊        | 1176/6470 [13:31<58:20,  1.51it/s] 
4|1177|Loss: 0.949170708656311:  18%|█▊        | 1177/6470 [13:31<57:23,  1.54it/s]
4|1178|Loss: 0.8507958650588989:  18%|█▊        | 1177/6470 [13:32<57:23,  1.54it/s]
4|1178|Loss: 0.8507958650588989:  18%|█▊        | 1178/6470 [13:32<55:48,  1.58it/s]
4|1179|Loss: 1.0030568838119507:  18%|█▊        | 1178/6470 [13:32<55:48,  1.58it/s]
4|1179|Loss: 1.0030568838119507:  18%|█▊        | 1179/6470 [13:33<55:05,  1.60it/s]
4|1180|Loss: 0.8569468855857849:  18%|█▊        | 1179/6470 [13:33<55:05,  1.60it/s]
4|1180|Loss: 0.8569468855857849:  18%|█▊        | 1180/6470 [13:33<1:00:11,  1.46it/s]
4|1181|Loss: 0.8960720300674438:  18%|█▊        | 1180/6470 [13:34<1:00:11,  1.46it/s]
4|1181|Loss: 0.8960720300674438:  18%|█▊        | 1181/6470 [13:34<1:03:35,  1.39it/s]
4|1182|Loss: 0.7845765352249146:  18%|█▊        | 1181/6470 [13:35<1:03:35,  1.39it/s]
4|1182|Loss: 0.7845765352249146:  18%|█▊        | 1182/6470 [13:35<1:02:16,  1.42it/s]
4|1183|Loss: 0.8495768308639526:  18%|█▊        | 1182/6470 [13:35<1:02:16,  1.42it/s]
4|1183|Loss: 0.8495768308639526:  18%|█▊        | 1183/6470 [13:36<1:02:42,  1.41it/s]
4|1184|Loss: 0.8822211623191833:  18%|█▊        | 1183/6470 [13:36<1:02:42,  1.41it/s]
4|1184|Loss: 0.8822211623191833:  18%|█▊        | 1184/6470 [13:36<1:01:28,  1.43it/s]
4|1185|Loss: 0.7892702221870422:  18%|█▊        | 1184/6470 [13:37<1:01:28,  1.43it/s]
4|1185|Loss: 0.7892702221870422:  18%|█▊        | 1185/6470 [13:37<1:01:56,  1.42it/s]
4|1186|Loss: 0.8724396824836731:  18%|█▊        | 1185/6470 [13:38<1:01:56,  1.42it/s]
4|1186|Loss: 0.8724396824836731:  18%|█▊        | 1186/6470 [13:38<1:05:43,  1.34it/s]
4|1187|Loss: 0.8245931267738342:  18%|█▊        | 1186/6470 [13:38<1:05:43,  1.34it/s]
4|1187|Loss: 0.8245931267738342:  18%|█▊        | 1187/6470 [13:39<1:06:19,  1.33it/s]
4|1188|Loss: 0.8670570850372314:  18%|█▊        | 1187/6470 [13:39<1:06:19,  1.33it/s]
4|1188|Loss: 0.8670570850372314:  18%|█▊        | 1188/6470 [13:39<1:05:12,  1.35it/s]
4|1189|Loss: 0.8106701374053955:  18%|█▊        | 1188/6470 [13:40<1:05:12,  1.35it/s]
4|1189|Loss: 0.8106701374053955:  18%|█▊        | 1189/6470 [13:40<1:02:19,  1.41it/s]
4|1190|Loss: 0.8957489728927612:  18%|█▊        | 1189/6470 [13:40<1:02:19,  1.41it/s]
4|1190|Loss: 0.8957489728927612:  18%|█▊        | 1190/6470 [13:41<1:01:36,  1.43it/s]
4|1191|Loss: 0.7996885776519775:  18%|█▊        | 1190/6470 [13:41<1:01:36,  1.43it/s]
4|1191|Loss: 0.7996885776519775:  18%|█▊        | 1191/6470 [13:41<1:00:21,  1.46it/s]
4|1192|Loss: 0.8977517485618591:  18%|█▊        | 1191/6470 [13:42<1:00:21,  1.46it/s]
4|1192|Loss: 0.8977517485618591:  18%|█▊        | 1192/6470 [13:42<55:15,  1.59it/s]  
4|1193|Loss: 0.8070699572563171:  18%|█▊        | 1192/6470 [13:42<55:15,  1.59it/s]
4|1193|Loss: 0.8070699572563171:  18%|█▊        | 1193/6470 [13:42<51:41,  1.70it/s]
4|1194|Loss: 0.8088200688362122:  18%|█▊        | 1193/6470 [13:43<51:41,  1.70it/s]
4|1194|Loss: 0.8088200688362122:  18%|█▊        | 1194/6470 [13:43<47:02,  1.87it/s]
4|1195|Loss: 0.8691139817237854:  18%|█▊        | 1194/6470 [13:43<47:02,  1.87it/s]
4|1195|Loss: 0.8691139817237854:  18%|█▊        | 1195/6470 [13:43<48:17,  1.82it/s]
4|1196|Loss: 0.7849957942962646:  18%|█▊        | 1195/6470 [13:44<48:17,  1.82it/s]
4|1196|Loss: 0.7849957942962646:  18%|█▊        | 1196/6470 [13:44<49:32,  1.77it/s]
4|1197|Loss: 0.8991182446479797:  18%|█▊        | 1196/6470 [13:44<49:32,  1.77it/s]
4|1197|Loss: 0.8991182446479797:  19%|█▊        | 1197/6470 [13:45<51:21,  1.71it/s]
4|1198|Loss: 0.8128604292869568:  19%|█▊        | 1197/6470 [13:45<51:21,  1.71it/s]
4|1198|Loss: 0.8128604292869568:  19%|█▊        | 1198/6470 [13:45<55:15,  1.59it/s]
4|1199|Loss: 0.8286313414573669:  19%|█▊        | 1198/6470 [13:46<55:15,  1.59it/s]
4|1199|Loss: 0.8286313414573669:  19%|█▊        | 1199/6470 [13:46<58:31,  1.50it/s]
4|1200|Loss: 0.9218854904174805:  19%|█▊        | 1199/6470 [13:46<58:31,  1.50it/s]
4|1200|Loss: 0.9218854904174805:  19%|█▊        | 1200/6470 [13:47<1:01:49,  1.42it/s]
4|1201|Loss: 0.9072476029396057:  19%|█▊        | 1200/6470 [13:47<1:01:49,  1.42it/s]
4|1201|Loss: 0.9072476029396057:  19%|█▊        | 1201/6470 [13:48<1:03:28,  1.38it/s]
4|1202|Loss: 0.6602113246917725:  19%|█▊        | 1201/6470 [13:48<1:03:28,  1.38it/s]
4|1202|Loss: 0.6602113246917725:  19%|█▊        | 1202/6470 [13:48<1:00:46,  1.44it/s]
4|1203|Loss: 0.7779841423034668:  19%|█▊        | 1202/6470 [13:49<1:00:46,  1.44it/s]
4|1203|Loss: 0.7779841423034668:  19%|█▊        | 1203/6470 [13:49<59:11,  1.48it/s]  
4|1204|Loss: 0.7353869080543518:  19%|█▊        | 1203/6470 [13:49<59:11,  1.48it/s]
4|1204|Loss: 0.7353869080543518:  19%|█▊        | 1204/6470 [13:49<56:23,  1.56it/s]
4|1205|Loss: 0.9186777472496033:  19%|█▊        | 1204/6470 [13:50<56:23,  1.56it/s]
4|1205|Loss: 0.9186777472496033:  19%|█▊        | 1205/6470 [13:50<53:10,  1.65it/s]
4|1206|Loss: 0.855867862701416:  19%|█▊        | 1205/6470 [13:50<53:10,  1.65it/s] 
4|1206|Loss: 0.855867862701416:  19%|█▊        | 1206/6470 [13:51<54:34,  1.61it/s]
4|1207|Loss: 0.9147845506668091:  19%|█▊        | 1206/6470 [13:51<54:34,  1.61it/s]
4|1207|Loss: 0.9147845506668091:  19%|█▊        | 1207/6470 [13:51<57:23,  1.53it/s]
4|1208|Loss: 0.8118721842765808:  19%|█▊        | 1207/6470 [13:52<57:23,  1.53it/s]
4|1208|Loss: 0.8118721842765808:  19%|█▊        | 1208/6470 [13:52<55:31,  1.58it/s]
4|1209|Loss: 0.8989771604537964:  19%|█▊        | 1208/6470 [13:52<55:31,  1.58it/s]
4|1209|Loss: 0.8989771604537964:  19%|█▊        | 1209/6470 [13:53<55:12,  1.59it/s]
4|1210|Loss: 0.848281741142273:  19%|█▊        | 1209/6470 [13:53<55:12,  1.59it/s] 
4|1210|Loss: 0.848281741142273:  19%|█▊        | 1210/6470 [13:53<54:21,  1.61it/s]
4|1211|Loss: 0.8743929862976074:  19%|█▊        | 1210/6470 [13:54<54:21,  1.61it/s]
4|1211|Loss: 0.8743929862976074:  19%|█▊        | 1211/6470 [13:54<56:47,  1.54it/s]
4|1212|Loss: 0.7257616519927979:  19%|█▊        | 1211/6470 [13:54<56:47,  1.54it/s]
4|1212|Loss: 0.7257616519927979:  19%|█▊        | 1212/6470 [13:55<1:00:15,  1.45it/s]
4|1213|Loss: 0.89715176820755:  19%|█▊        | 1212/6470 [13:55<1:00:15,  1.45it/s]  
4|1213|Loss: 0.89715176820755:  19%|█▊        | 1213/6470 [13:55<1:01:06,  1.43it/s]
4|1214|Loss: 0.8716588616371155:  19%|█▊        | 1213/6470 [13:56<1:01:06,  1.43it/s]
4|1214|Loss: 0.8716588616371155:  19%|█▉        | 1214/6470 [13:56<1:02:14,  1.41it/s]
4|1215|Loss: 0.8648504614830017:  19%|█▉        | 1214/6470 [13:57<1:02:14,  1.41it/s]
4|1215|Loss: 0.8648504614830017:  19%|█▉        | 1215/6470 [13:57<1:05:07,  1.34it/s]
4|1216|Loss: 0.7879828810691833:  19%|█▉        | 1215/6470 [13:57<1:05:07,  1.34it/s]
4|1216|Loss: 0.7879828810691833:  19%|█▉        | 1216/6470 [13:58<1:09:28,  1.26it/s]
4|1217|Loss: 0.7553598284721375:  19%|█▉        | 1216/6470 [13:58<1:09:28,  1.26it/s]
4|1217|Loss: 0.7553598284721375:  19%|█▉        | 1217/6470 [13:59<1:09:31,  1.26it/s]
4|1218|Loss: 0.8335325121879578:  19%|█▉        | 1217/6470 [13:59<1:09:31,  1.26it/s]
4|1218|Loss: 0.8335325121879578:  19%|█▉        | 1218/6470 [13:59<1:09:23,  1.26it/s]
4|1219|Loss: 0.9225555062294006:  19%|█▉        | 1218/6470 [14:00<1:09:23,  1.26it/s]
4|1219|Loss: 0.9225555062294006:  19%|█▉        | 1219/6470 [14:00<1:08:12,  1.28it/s]
4|1220|Loss: 0.792395830154419:  19%|█▉        | 1219/6470 [14:01<1:08:12,  1.28it/s] 
4|1220|Loss: 0.792395830154419:  19%|█▉        | 1220/6470 [14:01<1:05:01,  1.35it/s]
4|1221|Loss: 0.8654518127441406:  19%|█▉        | 1220/6470 [14:01<1:05:01,  1.35it/s]
4|1221|Loss: 0.8654518127441406:  19%|█▉        | 1221/6470 [14:02<1:04:30,  1.36it/s]
4|1222|Loss: 0.9376047253608704:  19%|█▉        | 1221/6470 [14:02<1:04:30,  1.36it/s]
4|1222|Loss: 0.9376047253608704:  19%|█▉        | 1222/6470 [14:02<1:04:33,  1.35it/s]
4|1223|Loss: 0.8500437140464783:  19%|█▉        | 1222/6470 [14:03<1:04:33,  1.35it/s]
4|1223|Loss: 0.8500437140464783:  19%|█▉        | 1223/6470 [14:03<59:41,  1.47it/s]  
4|1224|Loss: 0.8633531332015991:  19%|█▉        | 1223/6470 [14:03<59:41,  1.47it/s]
4|1224|Loss: 0.8633531332015991:  19%|█▉        | 1224/6470 [14:04<1:00:10,  1.45it/s]
4|1225|Loss: 0.7618931531906128:  19%|█▉        | 1224/6470 [14:04<1:00:10,  1.45it/s]
4|1225|Loss: 0.7618931531906128:  19%|█▉        | 1225/6470 [14:04<59:14,  1.48it/s]  
4|1226|Loss: 0.9779971837997437:  19%|█▉        | 1225/6470 [14:05<59:14,  1.48it/s]
4|1226|Loss: 0.9779971837997437:  19%|█▉        | 1226/6470 [14:05<57:33,  1.52it/s]
4|1227|Loss: 1.0293607711791992:  19%|█▉        | 1226/6470 [14:05<57:33,  1.52it/s]
4|1227|Loss: 1.0293607711791992:  19%|█▉        | 1227/6470 [14:05<55:12,  1.58it/s]
4|1228|Loss: 0.8393576741218567:  19%|█▉        | 1227/6470 [14:06<55:12,  1.58it/s]
4|1228|Loss: 0.8393576741218567:  19%|█▉        | 1228/6470 [14:06<54:38,  1.60it/s]
4|1229|Loss: 0.8539953827857971:  19%|█▉        | 1228/6470 [14:06<54:38,  1.60it/s]
4|1229|Loss: 0.8539953827857971:  19%|█▉        | 1229/6470 [14:07<56:27,  1.55it/s]
4|1230|Loss: 0.8169878721237183:  19%|█▉        | 1229/6470 [14:07<56:27,  1.55it/s]
4|1230|Loss: 0.8169878721237183:  19%|█▉        | 1230/6470 [14:07<57:51,  1.51it/s]
4|1231|Loss: 0.93282550573349:  19%|█▉        | 1230/6470 [14:08<57:51,  1.51it/s]  
4|1231|Loss: 0.93282550573349:  19%|█▉        | 1231/6470 [14:08<55:14,  1.58it/s]
4|1232|Loss: 0.8756558895111084:  19%|█▉        | 1231/6470 [14:08<55:14,  1.58it/s]
4|1232|Loss: 0.8756558895111084:  19%|█▉        | 1232/6470 [14:09<56:10,  1.55it/s]
4|1233|Loss: 0.9439890384674072:  19%|█▉        | 1232/6470 [14:09<56:10,  1.55it/s]
4|1233|Loss: 0.9439890384674072:  19%|█▉        | 1233/6470 [14:09<55:13,  1.58it/s]
4|1234|Loss: 0.8904650807380676:  19%|█▉        | 1233/6470 [14:10<55:13,  1.58it/s]
4|1234|Loss: 0.8904650807380676:  19%|█▉        | 1234/6470 [14:10<58:44,  1.49it/s]
4|1235|Loss: 0.736190140247345:  19%|█▉        | 1234/6470 [14:10<58:44,  1.49it/s] 
4|1235|Loss: 0.736190140247345:  19%|█▉        | 1235/6470 [14:11<1:00:15,  1.45it/s]
4|1236|Loss: 0.8296991586685181:  19%|█▉        | 1235/6470 [14:11<1:00:15,  1.45it/s]
4|1236|Loss: 0.8296991586685181:  19%|█▉        | 1236/6470 [14:11<1:00:33,  1.44it/s]
4|1237|Loss: 0.8439557552337646:  19%|█▉        | 1236/6470 [14:12<1:00:33,  1.44it/s]
4|1237|Loss: 0.8439557552337646:  19%|█▉        | 1237/6470 [14:12<59:03,  1.48it/s]  
4|1238|Loss: 0.8674412369728088:  19%|█▉        | 1237/6470 [14:12<59:03,  1.48it/s]
4|1238|Loss: 0.8674412369728088:  19%|█▉        | 1238/6470 [14:13<1:00:16,  1.45it/s]
4|1239|Loss: 0.8967831134796143:  19%|█▉        | 1238/6470 [14:13<1:00:16,  1.45it/s]
4|1239|Loss: 0.8967831134796143:  19%|█▉        | 1239/6470 [14:14<1:01:40,  1.41it/s]
4|1240|Loss: 0.8590019941329956:  19%|█▉        | 1239/6470 [14:14<1:01:40,  1.41it/s]
4|1240|Loss: 0.8590019941329956:  19%|█▉        | 1240/6470 [14:14<1:04:39,  1.35it/s]
4|1241|Loss: 0.7296431064605713:  19%|█▉        | 1240/6470 [14:15<1:04:39,  1.35it/s]
4|1241|Loss: 0.7296431064605713:  19%|█▉        | 1241/6470 [14:15<1:08:06,  1.28it/s]
4|1242|Loss: 0.9746744632720947:  19%|█▉        | 1241/6470 [14:16<1:08:06,  1.28it/s]
4|1242|Loss: 0.9746744632720947:  19%|█▉        | 1242/6470 [14:16<1:05:25,  1.33it/s]
4|1243|Loss: 0.8517704606056213:  19%|█▉        | 1242/6470 [14:16<1:05:25,  1.33it/s]
4|1243|Loss: 0.8517704606056213:  19%|█▉        | 1243/6470 [14:17<1:02:26,  1.40it/s]
4|1244|Loss: 0.790075421333313:  19%|█▉        | 1243/6470 [14:17<1:02:26,  1.40it/s] 
4|1244|Loss: 0.790075421333313:  19%|█▉        | 1244/6470 [14:17<1:02:45,  1.39it/s]
4|1245|Loss: 1.0240106582641602:  19%|█▉        | 1244/6470 [14:18<1:02:45,  1.39it/s]
4|1245|Loss: 1.0240106582641602:  19%|█▉        | 1245/6470 [14:18<1:01:19,  1.42it/s]
4|1246|Loss: 1.053914189338684:  19%|█▉        | 1245/6470 [14:18<1:01:19,  1.42it/s] 
4|1246|Loss: 1.053914189338684:  19%|█▉        | 1246/6470 [14:19<1:01:43,  1.41it/s]
4|1247|Loss: 1.0004475116729736:  19%|█▉        | 1246/6470 [14:19<1:01:43,  1.41it/s]
4|1247|Loss: 1.0004475116729736:  19%|█▉        | 1247/6470 [14:19<1:00:06,  1.45it/s]
4|1248|Loss: 0.8138680458068848:  19%|█▉        | 1247/6470 [14:20<1:00:06,  1.45it/s]
4|1248|Loss: 0.8138680458068848:  19%|█▉        | 1248/6470 [14:20<1:00:00,  1.45it/s]
4|1249|Loss: 0.8439292907714844:  19%|█▉        | 1248/6470 [14:20<1:00:00,  1.45it/s]
4|1249|Loss: 0.8439292907714844:  19%|█▉        | 1249/6470 [14:21<1:02:05,  1.40it/s]
4|1250|Loss: 1.0012097358703613:  19%|█▉        | 1249/6470 [14:21<1:02:05,  1.40it/s]
4|1250|Loss: 1.0012097358703613:  19%|█▉        | 1250/6470 [14:21<1:01:15,  1.42it/s]
4|1251|Loss: 0.8487805128097534:  19%|█▉        | 1250/6470 [14:22<1:01:15,  1.42it/s]
4|1251|Loss: 0.8487805128097534:  19%|█▉        | 1251/6470 [14:22<56:48,  1.53it/s]  
4|1252|Loss: 0.8568201661109924:  19%|█▉        | 1251/6470 [14:22<56:48,  1.53it/s]
4|1252|Loss: 0.8568201661109924:  19%|█▉        | 1252/6470 [14:23<57:03,  1.52it/s]
4|1253|Loss: 0.7989258766174316:  19%|█▉        | 1252/6470 [14:23<57:03,  1.52it/s]
4|1253|Loss: 0.7989258766174316:  19%|█▉        | 1253/6470 [14:23<1:00:54,  1.43it/s]
4|1254|Loss: 0.9389048218727112:  19%|█▉        | 1253/6470 [14:24<1:00:54,  1.43it/s]
4|1254|Loss: 0.9389048218727112:  19%|█▉        | 1254/6470 [14:24<59:53,  1.45it/s]  
4|1255|Loss: 0.8206451535224915:  19%|█▉        | 1254/6470 [14:25<59:53,  1.45it/s]
4|1255|Loss: 0.8206451535224915:  19%|█▉        | 1255/6470 [14:25<1:00:22,  1.44it/s]
4|1256|Loss: 0.8497723937034607:  19%|█▉        | 1255/6470 [14:25<1:00:22,  1.44it/s]
4|1256|Loss: 0.8497723937034607:  19%|█▉        | 1256/6470 [14:25<57:29,  1.51it/s]  
4|1257|Loss: 0.862179696559906:  19%|█▉        | 1256/6470 [14:26<57:29,  1.51it/s] 
4|1257|Loss: 0.862179696559906:  19%|█▉        | 1257/6470 [14:26<58:46,  1.48it/s]
4|1258|Loss: 0.835406482219696:  19%|█▉        | 1257/6470 [14:27<58:46,  1.48it/s]
4|1258|Loss: 0.835406482219696:  19%|█▉        | 1258/6470 [14:27<1:00:27,  1.44it/s]
4|1259|Loss: 0.8818531632423401:  19%|█▉        | 1258/6470 [14:27<1:00:27,  1.44it/s]
4|1259|Loss: 0.8818531632423401:  19%|█▉        | 1259/6470 [14:27<57:03,  1.52it/s]  
4|1260|Loss: 0.8983325958251953:  19%|█▉        | 1259/6470 [14:28<57:03,  1.52it/s]
4|1260|Loss: 0.8983325958251953:  19%|█▉        | 1260/6470 [14:28<57:19,  1.51it/s]
4|1261|Loss: 0.8198553919792175:  19%|█▉        | 1260/6470 [14:28<57:19,  1.51it/s]
4|1261|Loss: 0.8198553919792175:  19%|█▉        | 1261/6470 [14:29<56:46,  1.53it/s]
4|1262|Loss: 0.8791651129722595:  19%|█▉        | 1261/6470 [14:29<56:46,  1.53it/s]
4|1262|Loss: 0.8791651129722595:  20%|█▉        | 1262/6470 [14:29<58:16,  1.49it/s]
4|1263|Loss: 0.7984826564788818:  20%|█▉        | 1262/6470 [14:30<58:16,  1.49it/s]
4|1263|Loss: 0.7984826564788818:  20%|█▉        | 1263/6470 [14:30<59:24,  1.46it/s]
4|1264|Loss: 0.7363306879997253:  20%|█▉        | 1263/6470 [14:31<59:24,  1.46it/s]
4|1264|Loss: 0.7363306879997253:  20%|█▉        | 1264/6470 [14:31<58:38,  1.48it/s]
4|1265|Loss: 0.8277715444564819:  20%|█▉        | 1264/6470 [14:31<58:38,  1.48it/s]
4|1265|Loss: 0.8277715444564819:  20%|█▉        | 1265/6470 [14:32<1:00:56,  1.42it/s]
4|1266|Loss: 0.82776939868927:  20%|█▉        | 1265/6470 [14:32<1:00:56,  1.42it/s]  
4|1266|Loss: 0.82776939868927:  20%|█▉        | 1266/6470 [14:32<56:19,  1.54it/s]  
4|1267|Loss: 0.9727630615234375:  20%|█▉        | 1266/6470 [14:32<56:19,  1.54it/s]
4|1267|Loss: 0.9727630615234375:  20%|█▉        | 1267/6470 [14:33<56:09,  1.54it/s]
4|1268|Loss: 0.8936065435409546:  20%|█▉        | 1267/6470 [14:33<56:09,  1.54it/s]
4|1268|Loss: 0.8936065435409546:  20%|█▉        | 1268/6470 [14:34<1:00:25,  1.43it/s]
4|1269|Loss: 0.8921341300010681:  20%|█▉        | 1268/6470 [14:34<1:00:25,  1.43it/s]
4|1269|Loss: 0.8921341300010681:  20%|█▉        | 1269/6470 [14:34<54:22,  1.59it/s]  
4|1270|Loss: 0.6613940000534058:  20%|█▉        | 1269/6470 [14:34<54:22,  1.59it/s]
4|1270|Loss: 0.6613940000534058:  20%|█▉        | 1270/6470 [14:35<54:38,  1.59it/s]
4|1271|Loss: 0.7891893982887268:  20%|█▉        | 1270/6470 [14:35<54:38,  1.59it/s]
4|1271|Loss: 0.7891893982887268:  20%|█▉        | 1271/6470 [14:36<1:00:11,  1.44it/s]
4|1272|Loss: 0.8300808072090149:  20%|█▉        | 1271/6470 [14:36<1:00:11,  1.44it/s]
4|1272|Loss: 0.8300808072090149:  20%|█▉        | 1272/6470 [14:36<59:47,  1.45it/s]  
4|1273|Loss: 0.8114977478981018:  20%|█▉        | 1272/6470 [14:37<59:47,  1.45it/s]
4|1273|Loss: 0.8114977478981018:  20%|█▉        | 1273/6470 [14:37<59:51,  1.45it/s]
4|1274|Loss: 0.810142457485199:  20%|█▉        | 1273/6470 [14:37<59:51,  1.45it/s] 
4|1274|Loss: 0.810142457485199:  20%|█▉        | 1274/6470 [14:38<1:03:49,  1.36it/s]
4|1275|Loss: 0.8699889779090881:  20%|█▉        | 1274/6470 [14:38<1:03:49,  1.36it/s]
4|1275|Loss: 0.8699889779090881:  20%|█▉        | 1275/6470 [14:38<1:00:22,  1.43it/s]
4|1276|Loss: 0.9465272426605225:  20%|█▉        | 1275/6470 [14:39<1:00:22,  1.43it/s]
4|1276|Loss: 0.9465272426605225:  20%|█▉        | 1276/6470 [14:39<58:05,  1.49it/s]  
4|1277|Loss: 0.8383333086967468:  20%|█▉        | 1276/6470 [14:39<58:05,  1.49it/s]
4|1277|Loss: 0.8383333086967468:  20%|█▉        | 1277/6470 [14:40<58:29,  1.48it/s]
4|1278|Loss: 0.7640020251274109:  20%|█▉        | 1277/6470 [14:40<58:29,  1.48it/s]
4|1278|Loss: 0.7640020251274109:  20%|█▉        | 1278/6470 [14:40<52:07,  1.66it/s]
4|1279|Loss: 0.8897969722747803:  20%|█▉        | 1278/6470 [14:40<52:07,  1.66it/s]
4|1279|Loss: 0.8897969722747803:  20%|█▉        | 1279/6470 [14:41<51:29,  1.68it/s]
4|1280|Loss: 0.9850212335586548:  20%|█▉        | 1279/6470 [14:41<51:29,  1.68it/s]
4|1280|Loss: 0.9850212335586548:  20%|█▉        | 1280/6470 [14:41<52:27,  1.65it/s]
4|1281|Loss: 0.9681529402732849:  20%|█▉        | 1280/6470 [14:42<52:27,  1.65it/s]
4|1281|Loss: 0.9681529402732849:  20%|█▉        | 1281/6470 [14:42<50:31,  1.71it/s]
4|1282|Loss: 0.921444833278656:  20%|█▉        | 1281/6470 [14:42<50:31,  1.71it/s] 
4|1282|Loss: 0.921444833278656:  20%|█▉        | 1282/6470 [14:42<52:31,  1.65it/s]
4|1283|Loss: 0.7702853083610535:  20%|█▉        | 1282/6470 [14:43<52:31,  1.65it/s]
4|1283|Loss: 0.7702853083610535:  20%|█▉        | 1283/6470 [14:43<53:13,  1.62it/s]
4|1284|Loss: 0.8147178292274475:  20%|█▉        | 1283/6470 [14:43<53:13,  1.62it/s]
4|1284|Loss: 0.8147178292274475:  20%|█▉        | 1284/6470 [14:44<50:56,  1.70it/s]
4|1285|Loss: 0.7710078358650208:  20%|█▉        | 1284/6470 [14:44<50:56,  1.70it/s]
4|1285|Loss: 0.7710078358650208:  20%|█▉        | 1285/6470 [14:44<47:45,  1.81it/s]
4|1286|Loss: 1.0023071765899658:  20%|█▉        | 1285/6470 [14:44<47:45,  1.81it/s]
4|1286|Loss: 1.0023071765899658:  20%|█▉        | 1286/6470 [14:45<51:19,  1.68it/s]
4|1287|Loss: 0.9132363200187683:  20%|█▉        | 1286/6470 [14:45<51:19,  1.68it/s]
4|1287|Loss: 0.9132363200187683:  20%|█▉        | 1287/6470 [14:46<56:32,  1.53it/s]
4|1288|Loss: 0.7886067032814026:  20%|█▉        | 1287/6470 [14:46<56:32,  1.53it/s]
4|1288|Loss: 0.7886067032814026:  20%|█▉        | 1288/6470 [14:46<57:06,  1.51it/s]
4|1289|Loss: 0.7977677583694458:  20%|█▉        | 1288/6470 [14:47<57:06,  1.51it/s]
4|1289|Loss: 0.7977677583694458:  20%|█▉        | 1289/6470 [14:47<59:42,  1.45it/s]
4|1290|Loss: 0.685207724571228:  20%|█▉        | 1289/6470 [14:47<59:42,  1.45it/s] 
4|1290|Loss: 0.685207724571228:  20%|█▉        | 1290/6470 [14:48<57:58,  1.49it/s]
4|1291|Loss: 1.0076208114624023:  20%|█▉        | 1290/6470 [14:48<57:58,  1.49it/s]
4|1291|Loss: 1.0076208114624023:  20%|█▉        | 1291/6470 [14:48<53:08,  1.62it/s]
4|1292|Loss: 1.0266507863998413:  20%|█▉        | 1291/6470 [14:48<53:08,  1.62it/s]
4|1292|Loss: 1.0266507863998413:  20%|█▉        | 1292/6470 [14:49<52:50,  1.63it/s]
4|1293|Loss: 0.7723492383956909:  20%|█▉        | 1292/6470 [14:49<52:50,  1.63it/s]
4|1293|Loss: 0.7723492383956909:  20%|█▉        | 1293/6470 [14:50<57:00,  1.51it/s]
4|1294|Loss: 0.8876163959503174:  20%|█▉        | 1293/6470 [14:50<57:00,  1.51it/s]
4|1294|Loss: 0.8876163959503174:  20%|██        | 1294/6470 [14:50<59:36,  1.45it/s]
4|1295|Loss: 0.9133251905441284:  20%|██        | 1294/6470 [14:51<59:36,  1.45it/s]
4|1295|Loss: 0.9133251905441284:  20%|██        | 1295/6470 [14:51<58:33,  1.47it/s]
4|1296|Loss: 0.8916351199150085:  20%|██        | 1295/6470 [14:51<58:33,  1.47it/s]
4|1296|Loss: 0.8916351199150085:  20%|██        | 1296/6470 [14:51<54:16,  1.59it/s]
4|1297|Loss: 0.8658648133277893:  20%|██        | 1296/6470 [14:52<54:16,  1.59it/s]
4|1297|Loss: 0.8658648133277893:  20%|██        | 1297/6470 [14:52<55:28,  1.55it/s]
4|1298|Loss: 0.6915831565856934:  20%|██        | 1297/6470 [14:53<55:28,  1.55it/s]
4|1298|Loss: 0.6915831565856934:  20%|██        | 1298/6470 [14:53<58:02,  1.49it/s]
4|1299|Loss: 0.9313825964927673:  20%|██        | 1298/6470 [14:53<58:02,  1.49it/s]
4|1299|Loss: 0.9313825964927673:  20%|██        | 1299/6470 [14:54<59:54,  1.44it/s]
4|1300|Loss: 0.8767231702804565:  20%|██        | 1299/6470 [14:54<59:54,  1.44it/s]
4|1300|Loss: 0.8767231702804565:  20%|██        | 1300/6470 [14:54<58:31,  1.47it/s]
4|1301|Loss: 0.6941562294960022:  20%|██        | 1300/6470 [14:55<58:31,  1.47it/s]
4|1301|Loss: 0.6941562294960022:  20%|██        | 1301/6470 [14:55<51:37,  1.67it/s]
4|1302|Loss: 0.7104596495628357:  20%|██        | 1301/6470 [14:55<51:37,  1.67it/s]
4|1302|Loss: 0.7104596495628357:  20%|██        | 1302/6470 [14:55<47:35,  1.81it/s]
4|1303|Loss: 0.6067735552787781:  20%|██        | 1302/6470 [14:55<47:35,  1.81it/s]
4|1303|Loss: 0.6067735552787781:  20%|██        | 1303/6470 [14:56<50:21,  1.71it/s]
4|1304|Loss: 0.8693833947181702:  20%|██        | 1303/6470 [14:56<50:21,  1.71it/s]
4|1304|Loss: 0.8693833947181702:  20%|██        | 1304/6470 [14:56<53:42,  1.60it/s]
4|1305|Loss: 0.9941726326942444:  20%|██        | 1304/6470 [14:57<53:42,  1.60it/s]
4|1305|Loss: 0.9941726326942444:  20%|██        | 1305/6470 [14:57<57:37,  1.49it/s]
4|1306|Loss: 0.8295412063598633:  20%|██        | 1305/6470 [14:58<57:37,  1.49it/s]
4|1306|Loss: 0.8295412063598633:  20%|██        | 1306/6470 [14:58<1:02:35,  1.37it/s]
4|1307|Loss: 0.9328165054321289:  20%|██        | 1306/6470 [14:59<1:02:35,  1.37it/s]
4|1307|Loss: 0.9328165054321289:  20%|██        | 1307/6470 [14:59<1:03:26,  1.36it/s]
4|1308|Loss: 0.9264994859695435:  20%|██        | 1307/6470 [14:59<1:03:26,  1.36it/s]
4|1308|Loss: 0.9264994859695435:  20%|██        | 1308/6470 [15:00<1:06:04,  1.30it/s]
4|1309|Loss: 0.7674645185470581:  20%|██        | 1308/6470 [15:00<1:06:04,  1.30it/s]
4|1309|Loss: 0.7674645185470581:  20%|██        | 1309/6470 [15:00<1:06:23,  1.30it/s]
4|1310|Loss: 0.8440990447998047:  20%|██        | 1309/6470 [15:01<1:06:23,  1.30it/s]
4|1310|Loss: 0.8440990447998047:  20%|██        | 1310/6470 [15:01<1:07:38,  1.27it/s]
4|1311|Loss: 0.8753658533096313:  20%|██        | 1310/6470 [15:02<1:07:38,  1.27it/s]
4|1311|Loss: 0.8753658533096313:  20%|██        | 1311/6470 [15:02<1:09:19,  1.24it/s]
4|1312|Loss: 0.8569031953811646:  20%|██        | 1311/6470 [15:03<1:09:19,  1.24it/s]
4|1312|Loss: 0.8569031953811646:  20%|██        | 1312/6470 [15:03<1:08:08,  1.26it/s]
4|1313|Loss: 0.7597556710243225:  20%|██        | 1312/6470 [15:03<1:08:08,  1.26it/s]
4|1313|Loss: 0.7597556710243225:  20%|██        | 1313/6470 [15:04<1:04:43,  1.33it/s]
4|1314|Loss: 0.9863846898078918:  20%|██        | 1313/6470 [15:04<1:04:43,  1.33it/s]
4|1314|Loss: 0.9863846898078918:  20%|██        | 1314/6470 [15:04<1:01:50,  1.39it/s]
4|1315|Loss: 0.8532379865646362:  20%|██        | 1314/6470 [15:05<1:01:50,  1.39it/s]
4|1315|Loss: 0.8532379865646362:  20%|██        | 1315/6470 [15:05<1:01:40,  1.39it/s]
4|1316|Loss: 0.8452471494674683:  20%|██        | 1315/6470 [15:05<1:01:40,  1.39it/s]
4|1316|Loss: 0.8452471494674683:  20%|██        | 1316/6470 [15:06<1:01:59,  1.39it/s]
4|1317|Loss: 0.7334734201431274:  20%|██        | 1316/6470 [15:06<1:01:59,  1.39it/s]
4|1317|Loss: 0.7334734201431274:  20%|██        | 1317/6470 [15:06<1:04:20,  1.33it/s]
4|1318|Loss: 0.9537350535392761:  20%|██        | 1317/6470 [15:07<1:04:20,  1.33it/s]
4|1318|Loss: 0.9537350535392761:  20%|██        | 1318/6470 [15:07<1:02:28,  1.37it/s]
4|1319|Loss: 0.8525237441062927:  20%|██        | 1318/6470 [15:08<1:02:28,  1.37it/s]
4|1319|Loss: 0.8525237441062927:  20%|██        | 1319/6470 [15:08<1:02:11,  1.38it/s]
4|1320|Loss: 0.8127214312553406:  20%|██        | 1319/6470 [15:08<1:02:11,  1.38it/s]
4|1320|Loss: 0.8127214312553406:  20%|██        | 1320/6470 [15:09<1:05:08,  1.32it/s]
4|1321|Loss: 0.7753505110740662:  20%|██        | 1320/6470 [15:09<1:05:08,  1.32it/s]
4|1321|Loss: 0.7753505110740662:  20%|██        | 1321/6470 [15:09<1:00:44,  1.41it/s]
4|1322|Loss: 0.9012823104858398:  20%|██        | 1321/6470 [15:10<1:00:44,  1.41it/s]
4|1322|Loss: 0.9012823104858398:  20%|██        | 1322/6470 [15:10<57:40,  1.49it/s]  
4|1323|Loss: 0.7957640886306763:  20%|██        | 1322/6470 [15:10<57:40,  1.49it/s]
4|1323|Loss: 0.7957640886306763:  20%|██        | 1323/6470 [15:11<58:00,  1.48it/s]
4|1324|Loss: 0.9919148683547974:  20%|██        | 1323/6470 [15:11<58:00,  1.48it/s]
4|1324|Loss: 0.9919148683547974:  20%|██        | 1324/6470 [15:11<1:00:35,  1.42it/s]
4|1325|Loss: 0.9161640405654907:  20%|██        | 1324/6470 [15:12<1:00:35,  1.42it/s]
4|1325|Loss: 0.9161640405654907:  20%|██        | 1325/6470 [15:12<58:39,  1.46it/s]  
4|1326|Loss: 0.7420185804367065:  20%|██        | 1325/6470 [15:12<58:39,  1.46it/s]
4|1326|Loss: 0.7420185804367065:  20%|██        | 1326/6470 [15:13<59:08,  1.45it/s]
4|1327|Loss: 0.7468443512916565:  20%|██        | 1326/6470 [15:13<59:08,  1.45it/s]
4|1327|Loss: 0.7468443512916565:  21%|██        | 1327/6470 [15:13<58:03,  1.48it/s]
4|1328|Loss: 0.7357584238052368:  21%|██        | 1327/6470 [15:14<58:03,  1.48it/s]
4|1328|Loss: 0.7357584238052368:  21%|██        | 1328/6470 [15:14<1:02:33,  1.37it/s]
4|1329|Loss: 0.993873119354248:  21%|██        | 1328/6470 [15:15<1:02:33,  1.37it/s] 
4|1329|Loss: 0.993873119354248:  21%|██        | 1329/6470 [15:15<1:04:46,  1.32it/s]
4|1330|Loss: 0.8763928413391113:  21%|██        | 1329/6470 [15:15<1:04:46,  1.32it/s]
4|1330|Loss: 0.8763928413391113:  21%|██        | 1330/6470 [15:16<1:02:46,  1.36it/s]
4|1331|Loss: 0.6707334518432617:  21%|██        | 1330/6470 [15:16<1:02:46,  1.36it/s]
4|1331|Loss: 0.6707334518432617:  21%|██        | 1331/6470 [15:17<1:04:57,  1.32it/s]
4|1332|Loss: 0.8793916702270508:  21%|██        | 1331/6470 [15:17<1:04:57,  1.32it/s]
4|1332|Loss: 0.8793916702270508:  21%|██        | 1332/6470 [15:17<1:03:12,  1.35it/s]
4|1333|Loss: 0.9345943927764893:  21%|██        | 1332/6470 [15:18<1:03:12,  1.35it/s]
4|1333|Loss: 0.9345943927764893:  21%|██        | 1333/6470 [15:18<1:05:06,  1.31it/s]
4|1334|Loss: 0.7536391019821167:  21%|██        | 1333/6470 [15:18<1:05:06,  1.31it/s]
4|1334|Loss: 0.7536391019821167:  21%|██        | 1334/6470 [15:19<1:03:18,  1.35it/s]
4|1335|Loss: 0.8498930931091309:  21%|██        | 1334/6470 [15:19<1:03:18,  1.35it/s]
4|1335|Loss: 0.8498930931091309:  21%|██        | 1335/6470 [15:19<1:00:11,  1.42it/s]
4|1336|Loss: 0.8709877133369446:  21%|██        | 1335/6470 [15:20<1:00:11,  1.42it/s]
4|1336|Loss: 0.8709877133369446:  21%|██        | 1336/6470 [15:20<1:00:32,  1.41it/s]
4|1337|Loss: 0.7858380079269409:  21%|██        | 1336/6470 [15:20<1:00:32,  1.41it/s]
4|1337|Loss: 0.7858380079269409:  21%|██        | 1337/6470 [15:21<1:02:34,  1.37it/s]
4|1338|Loss: 0.8399467468261719:  21%|██        | 1337/6470 [15:21<1:02:34,  1.37it/s]
4|1338|Loss: 0.8399467468261719:  21%|██        | 1338/6470 [15:21<57:16,  1.49it/s]  
4|1339|Loss: 0.8615923523902893:  21%|██        | 1338/6470 [15:22<57:16,  1.49it/s]
4|1339|Loss: 0.8615923523902893:  21%|██        | 1339/6470 [15:22<58:29,  1.46it/s]
4|1340|Loss: 1.059794306755066:  21%|██        | 1339/6470 [15:23<58:29,  1.46it/s] 
4|1340|Loss: 1.059794306755066:  21%|██        | 1340/6470 [15:23<1:00:37,  1.41it/s]
4|1341|Loss: 0.8584391474723816:  21%|██        | 1340/6470 [15:23<1:00:37,  1.41it/s]
4|1341|Loss: 0.8584391474723816:  21%|██        | 1341/6470 [15:24<1:02:47,  1.36it/s]
4|1342|Loss: 0.9132997393608093:  21%|██        | 1341/6470 [15:24<1:02:47,  1.36it/s]
4|1342|Loss: 0.9132997393608093:  21%|██        | 1342/6470 [15:24<1:05:02,  1.31it/s]
4|1343|Loss: 0.757929801940918:  21%|██        | 1342/6470 [15:25<1:05:02,  1.31it/s] 
4|1343|Loss: 0.757929801940918:  21%|██        | 1343/6470 [15:25<1:04:44,  1.32it/s]
4|1344|Loss: 0.8425294160842896:  21%|██        | 1343/6470 [15:26<1:04:44,  1.32it/s]
4|1344|Loss: 0.8425294160842896:  21%|██        | 1344/6470 [15:26<1:04:08,  1.33it/s]
4|1345|Loss: 0.9153439998626709:  21%|██        | 1344/6470 [15:26<1:04:08,  1.33it/s]
4|1345|Loss: 0.9153439998626709:  21%|██        | 1345/6470 [15:27<1:01:30,  1.39it/s]
4|1346|Loss: 0.8425769209861755:  21%|██        | 1345/6470 [15:27<1:01:30,  1.39it/s]
4|1346|Loss: 0.8425769209861755:  21%|██        | 1346/6470 [15:27<1:01:14,  1.39it/s]
4|1347|Loss: 0.9814355373382568:  21%|██        | 1346/6470 [15:28<1:01:14,  1.39it/s]
4|1347|Loss: 0.9814355373382568:  21%|██        | 1347/6470 [15:28<59:21,  1.44it/s]  
4|1348|Loss: 0.8072153329849243:  21%|██        | 1347/6470 [15:28<59:21,  1.44it/s]
4|1348|Loss: 0.8072153329849243:  21%|██        | 1348/6470 [15:29<56:28,  1.51it/s]
4|1349|Loss: 1.0309666395187378:  21%|██        | 1348/6470 [15:29<56:28,  1.51it/s]
4|1349|Loss: 1.0309666395187378:  21%|██        | 1349/6470 [15:29<55:38,  1.53it/s]
4|1350|Loss: 0.9430229663848877:  21%|██        | 1349/6470 [15:30<55:38,  1.53it/s]
4|1350|Loss: 0.9430229663848877:  21%|██        | 1350/6470 [15:30<56:27,  1.51it/s]
4|1351|Loss: 0.8627822995185852:  21%|██        | 1350/6470 [15:30<56:27,  1.51it/s]
4|1351|Loss: 0.8627822995185852:  21%|██        | 1351/6470 [15:30<54:17,  1.57it/s]
4|1352|Loss: 0.7337519526481628:  21%|██        | 1351/6470 [15:31<54:17,  1.57it/s]
4|1352|Loss: 0.7337519526481628:  21%|██        | 1352/6470 [15:31<52:44,  1.62it/s]
4|1353|Loss: 0.8824042677879333:  21%|██        | 1352/6470 [15:31<52:44,  1.62it/s]
4|1353|Loss: 0.8824042677879333:  21%|██        | 1353/6470 [15:32<54:18,  1.57it/s]
4|1354|Loss: 0.8028934001922607:  21%|██        | 1353/6470 [15:32<54:18,  1.57it/s]
4|1354|Loss: 0.8028934001922607:  21%|██        | 1354/6470 [15:32<52:15,  1.63it/s]
4|1355|Loss: 1.048599362373352:  21%|██        | 1354/6470 [15:33<52:15,  1.63it/s] 
4|1355|Loss: 1.048599362373352:  21%|██        | 1355/6470 [15:33<55:28,  1.54it/s]
4|1356|Loss: 0.823309063911438:  21%|██        | 1355/6470 [15:33<55:28,  1.54it/s]
4|1356|Loss: 0.823309063911438:  21%|██        | 1356/6470 [15:34<58:02,  1.47it/s]
4|1357|Loss: 0.9528776407241821:  21%|██        | 1356/6470 [15:34<58:02,  1.47it/s]
4|1357|Loss: 0.9528776407241821:  21%|██        | 1357/6470 [15:35<1:00:52,  1.40it/s]
4|1358|Loss: 0.8697288632392883:  21%|██        | 1357/6470 [15:35<1:00:52,  1.40it/s]
4|1358|Loss: 0.8697288632392883:  21%|██        | 1358/6470 [15:35<57:15,  1.49it/s]  
4|1359|Loss: 0.8499498963356018:  21%|██        | 1358/6470 [15:35<57:15,  1.49it/s]
4|1359|Loss: 0.8499498963356018:  21%|██        | 1359/6470 [15:36<58:00,  1.47it/s]
4|1360|Loss: 0.9024938344955444:  21%|██        | 1359/6470 [15:36<58:00,  1.47it/s]
4|1360|Loss: 0.9024938344955444:  21%|██        | 1360/6470 [15:37<59:40,  1.43it/s]
4|1361|Loss: 0.8343844413757324:  21%|██        | 1360/6470 [15:37<59:40,  1.43it/s]
4|1361|Loss: 0.8343844413757324:  21%|██        | 1361/6470 [15:37<1:00:55,  1.40it/s]
4|1362|Loss: 0.8839397430419922:  21%|██        | 1361/6470 [15:38<1:00:55,  1.40it/s]
4|1362|Loss: 0.8839397430419922:  21%|██        | 1362/6470 [15:38<1:01:38,  1.38it/s]
4|1363|Loss: 0.6914199590682983:  21%|██        | 1362/6470 [15:38<1:01:38,  1.38it/s]
4|1363|Loss: 0.6914199590682983:  21%|██        | 1363/6470 [15:39<1:03:51,  1.33it/s]
4|1364|Loss: 0.757544994354248:  21%|██        | 1363/6470 [15:39<1:03:51,  1.33it/s] 
4|1364|Loss: 0.757544994354248:  21%|██        | 1364/6470 [15:40<1:03:15,  1.35it/s]
4|1365|Loss: 0.8384606242179871:  21%|██        | 1364/6470 [15:40<1:03:15,  1.35it/s]
4|1365|Loss: 0.8384606242179871:  21%|██        | 1365/6470 [15:40<1:04:19,  1.32it/s]
4|1366|Loss: 0.8703166246414185:  21%|██        | 1365/6470 [15:41<1:04:19,  1.32it/s]
4|1366|Loss: 0.8703166246414185:  21%|██        | 1366/6470 [15:41<1:02:18,  1.37it/s]
4|1367|Loss: 0.8357501029968262:  21%|██        | 1366/6470 [15:41<1:02:18,  1.37it/s]
4|1367|Loss: 0.8357501029968262:  21%|██        | 1367/6470 [15:42<1:01:50,  1.38it/s]
4|1368|Loss: 0.8006846904754639:  21%|██        | 1367/6470 [15:42<1:01:50,  1.38it/s]
4|1368|Loss: 0.8006846904754639:  21%|██        | 1368/6470 [15:43<1:05:35,  1.30it/s]
4|1369|Loss: 0.8585836887359619:  21%|██        | 1368/6470 [15:43<1:05:35,  1.30it/s]
4|1369|Loss: 0.8585836887359619:  21%|██        | 1369/6470 [15:43<1:01:07,  1.39it/s]
4|1370|Loss: 0.6781396865844727:  21%|██        | 1369/6470 [15:44<1:01:07,  1.39it/s]
4|1370|Loss: 0.6781396865844727:  21%|██        | 1370/6470 [15:44<1:01:44,  1.38it/s]
4|1371|Loss: 0.7019432187080383:  21%|██        | 1370/6470 [15:44<1:01:44,  1.38it/s]
4|1371|Loss: 0.7019432187080383:  21%|██        | 1371/6470 [15:45<1:02:45,  1.35it/s]
4|1372|Loss: 0.8164404630661011:  21%|██        | 1371/6470 [15:45<1:02:45,  1.35it/s]
4|1372|Loss: 0.8164404630661011:  21%|██        | 1372/6470 [15:45<58:15,  1.46it/s]  
4|1373|Loss: 0.8048011660575867:  21%|██        | 1372/6470 [15:46<58:15,  1.46it/s]
4|1373|Loss: 0.8048011660575867:  21%|██        | 1373/6470 [15:46<50:44,  1.67it/s]
4|1374|Loss: 0.894252598285675:  21%|██        | 1373/6470 [15:46<50:44,  1.67it/s] 
4|1374|Loss: 0.894252598285675:  21%|██        | 1374/6470 [15:46<50:12,  1.69it/s]
4|1375|Loss: 0.8372253179550171:  21%|██        | 1374/6470 [15:47<50:12,  1.69it/s]
4|1375|Loss: 0.8372253179550171:  21%|██▏       | 1375/6470 [15:47<53:37,  1.58it/s]
4|1376|Loss: 0.7544767260551453:  21%|██▏       | 1375/6470 [15:47<53:37,  1.58it/s]
4|1376|Loss: 0.7544767260551453:  21%|██▏       | 1376/6470 [15:48<54:01,  1.57it/s]
4|1377|Loss: 0.7764586806297302:  21%|██▏       | 1376/6470 [15:48<54:01,  1.57it/s]
4|1377|Loss: 0.7764586806297302:  21%|██▏       | 1377/6470 [15:48<57:44,  1.47it/s]
4|1378|Loss: 0.9545477032661438:  21%|██▏       | 1377/6470 [15:49<57:44,  1.47it/s]
4|1378|Loss: 0.9545477032661438:  21%|██▏       | 1378/6470 [15:49<59:01,  1.44it/s]
4|1379|Loss: 0.9263178706169128:  21%|██▏       | 1378/6470 [15:50<59:01,  1.44it/s]
4|1379|Loss: 0.9263178706169128:  21%|██▏       | 1379/6470 [15:50<1:01:04,  1.39it/s]
4|1380|Loss: 0.9105200171470642:  21%|██▏       | 1379/6470 [15:50<1:01:04,  1.39it/s]
4|1380|Loss: 0.9105200171470642:  21%|██▏       | 1380/6470 [15:51<58:13,  1.46it/s]  
4|1381|Loss: 0.7340583205223083:  21%|██▏       | 1380/6470 [15:51<58:13,  1.46it/s]
4|1381|Loss: 0.7340583205223083:  21%|██▏       | 1381/6470 [15:51<53:52,  1.57it/s]
4|1382|Loss: 0.8382861614227295:  21%|██▏       | 1381/6470 [15:51<53:52,  1.57it/s]
4|1382|Loss: 0.8382861614227295:  21%|██▏       | 1382/6470 [15:52<55:32,  1.53it/s]
4|1383|Loss: 0.858846127986908:  21%|██▏       | 1382/6470 [15:52<55:32,  1.53it/s] 
4|1383|Loss: 0.858846127986908:  21%|██▏       | 1383/6470 [15:52<54:07,  1.57it/s]
4|1384|Loss: 0.8031982779502869:  21%|██▏       | 1383/6470 [15:53<54:07,  1.57it/s]
4|1384|Loss: 0.8031982779502869:  21%|██▏       | 1384/6470 [15:53<1:01:10,  1.39it/s]
4|1385|Loss: 0.948434591293335:  21%|██▏       | 1384/6470 [15:54<1:01:10,  1.39it/s] 
4|1385|Loss: 0.948434591293335:  21%|██▏       | 1385/6470 [15:54<1:11:17,  1.19it/s]
4|1386|Loss: 0.8931353092193604:  21%|██▏       | 1385/6470 [15:55<1:11:17,  1.19it/s]
4|1386|Loss: 0.8931353092193604:  21%|██▏       | 1386/6470 [15:55<1:10:31,  1.20it/s]
4|1387|Loss: 0.6880521774291992:  21%|██▏       | 1386/6470 [15:56<1:10:31,  1.20it/s]
4|1387|Loss: 0.6880521774291992:  21%|██▏       | 1387/6470 [15:56<1:11:06,  1.19it/s]
4|1388|Loss: 0.8842582702636719:  21%|██▏       | 1387/6470 [15:56<1:11:06,  1.19it/s]
4|1388|Loss: 0.8842582702636719:  21%|██▏       | 1388/6470 [15:57<1:08:27,  1.24it/s]
4|1389|Loss: 0.7526479959487915:  21%|██▏       | 1388/6470 [15:57<1:08:27,  1.24it/s]
4|1389|Loss: 0.7526479959487915:  21%|██▏       | 1389/6470 [15:58<1:07:38,  1.25it/s]
4|1390|Loss: 0.8735277652740479:  21%|██▏       | 1389/6470 [15:58<1:07:38,  1.25it/s]
4|1390|Loss: 0.8735277652740479:  21%|██▏       | 1390/6470 [15:58<1:06:17,  1.28it/s]
4|1391|Loss: 0.6800696849822998:  21%|██▏       | 1390/6470 [15:59<1:06:17,  1.28it/s]
4|1391|Loss: 0.6800696849822998:  21%|██▏       | 1391/6470 [15:59<1:01:07,  1.38it/s]
4|1392|Loss: 0.8658488988876343:  21%|██▏       | 1391/6470 [15:59<1:01:07,  1.38it/s]
4|1392|Loss: 0.8658488988876343:  22%|██▏       | 1392/6470 [16:00<59:56,  1.41it/s]  
4|1393|Loss: 0.8334054350852966:  22%|██▏       | 1392/6470 [16:00<59:56,  1.41it/s]
4|1393|Loss: 0.8334054350852966:  22%|██▏       | 1393/6470 [16:00<1:00:29,  1.40it/s]
4|1394|Loss: 0.8276174664497375:  22%|██▏       | 1393/6470 [16:01<1:00:29,  1.40it/s]
4|1394|Loss: 0.8276174664497375:  22%|██▏       | 1394/6470 [16:01<57:50,  1.46it/s]  
4|1395|Loss: 0.9892457127571106:  22%|██▏       | 1394/6470 [16:01<57:50,  1.46it/s]
4|1395|Loss: 0.9892457127571106:  22%|██▏       | 1395/6470 [16:02<58:20,  1.45it/s]
4|1396|Loss: 1.0782792568206787:  22%|██▏       | 1395/6470 [16:02<58:20,  1.45it/s]
4|1396|Loss: 1.0782792568206787:  22%|██▏       | 1396/6470 [16:02<57:44,  1.46it/s]
4|1397|Loss: 0.8829829096794128:  22%|██▏       | 1396/6470 [16:03<57:44,  1.46it/s]
4|1397|Loss: 0.8829829096794128:  22%|██▏       | 1397/6470 [16:03<58:15,  1.45it/s]
4|1398|Loss: 0.849245548248291:  22%|██▏       | 1397/6470 [16:03<58:15,  1.45it/s] 
4|1398|Loss: 0.849245548248291:  22%|██▏       | 1398/6470 [16:04<59:42,  1.42it/s]
4|1399|Loss: 0.8711948990821838:  22%|██▏       | 1398/6470 [16:04<59:42,  1.42it/s]
4|1399|Loss: 0.8711948990821838:  22%|██▏       | 1399/6470 [16:04<58:52,  1.44it/s]
4|1400|Loss: 0.9212440252304077:  22%|██▏       | 1399/6470 [16:05<58:52,  1.44it/s]
4|1400|Loss: 0.9212440252304077:  22%|██▏       | 1400/6470 [16:05<59:27,  1.42it/s]
4|1401|Loss: 0.8924782276153564:  22%|██▏       | 1400/6470 [16:06<59:27,  1.42it/s]
4|1401|Loss: 0.8924782276153564:  22%|██▏       | 1401/6470 [16:06<1:03:33,  1.33it/s]
4|1402|Loss: 0.8462803959846497:  22%|██▏       | 1401/6470 [16:06<1:03:33,  1.33it/s]
4|1402|Loss: 0.8462803959846497:  22%|██▏       | 1402/6470 [16:07<58:19,  1.45it/s]  
4|1403|Loss: 1.0867407321929932:  22%|██▏       | 1402/6470 [16:07<58:19,  1.45it/s]
4|1403|Loss: 1.0867407321929932:  22%|██▏       | 1403/6470 [16:07<53:43,  1.57it/s]
4|1404|Loss: 0.7773236036300659:  22%|██▏       | 1403/6470 [16:07<53:43,  1.57it/s]
4|1404|Loss: 0.7773236036300659:  22%|██▏       | 1404/6470 [16:08<54:30,  1.55it/s]
4|1405|Loss: 0.8603386282920837:  22%|██▏       | 1404/6470 [16:08<54:30,  1.55it/s]
4|1405|Loss: 0.8603386282920837:  22%|██▏       | 1405/6470 [16:09<58:58,  1.43it/s]
4|1406|Loss: 0.8370085954666138:  22%|██▏       | 1405/6470 [16:09<58:58,  1.43it/s]
4|1406|Loss: 0.8370085954666138:  22%|██▏       | 1406/6470 [16:09<58:46,  1.44it/s]
4|1407|Loss: 0.8365761637687683:  22%|██▏       | 1406/6470 [16:10<58:46,  1.44it/s]
4|1407|Loss: 0.8365761637687683:  22%|██▏       | 1407/6470 [16:10<59:26,  1.42it/s]
4|1408|Loss: 0.8379773497581482:  22%|██▏       | 1407/6470 [16:10<59:26,  1.42it/s]
4|1408|Loss: 0.8379773497581482:  22%|██▏       | 1408/6470 [16:11<59:47,  1.41it/s]
4|1409|Loss: 0.8883919715881348:  22%|██▏       | 1408/6470 [16:11<59:47,  1.41it/s]
4|1409|Loss: 0.8883919715881348:  22%|██▏       | 1409/6470 [16:11<58:33,  1.44it/s]
4|1410|Loss: 0.903110921382904:  22%|██▏       | 1409/6470 [16:12<58:33,  1.44it/s] 
4|1410|Loss: 0.903110921382904:  22%|██▏       | 1410/6470 [16:12<56:51,  1.48it/s]
4|1411|Loss: 1.0381413698196411:  22%|██▏       | 1410/6470 [16:12<56:51,  1.48it/s]
4|1411|Loss: 1.0381413698196411:  22%|██▏       | 1411/6470 [16:13<58:05,  1.45it/s]
4|1412|Loss: 0.8362187743186951:  22%|██▏       | 1411/6470 [16:13<58:05,  1.45it/s]
4|1412|Loss: 0.8362187743186951:  22%|██▏       | 1412/6470 [16:13<57:06,  1.48it/s]
4|1413|Loss: 0.7754561305046082:  22%|██▏       | 1412/6470 [16:14<57:06,  1.48it/s]
4|1413|Loss: 0.7754561305046082:  22%|██▏       | 1413/6470 [16:14<58:16,  1.45it/s]
4|1414|Loss: 0.7927761673927307:  22%|██▏       | 1413/6470 [16:14<58:16,  1.45it/s]
4|1414|Loss: 0.7927761673927307:  22%|██▏       | 1414/6470 [16:15<56:17,  1.50it/s]
4|1415|Loss: 0.7594006061553955:  22%|██▏       | 1414/6470 [16:15<56:17,  1.50it/s]
4|1415|Loss: 0.7594006061553955:  22%|██▏       | 1415/6470 [16:15<55:12,  1.53it/s]
4|1416|Loss: 0.8627522587776184:  22%|██▏       | 1415/6470 [16:16<55:12,  1.53it/s]
4|1416|Loss: 0.8627522587776184:  22%|██▏       | 1416/6470 [16:16<54:25,  1.55it/s]
4|1417|Loss: 0.8608571887016296:  22%|██▏       | 1416/6470 [16:16<54:25,  1.55it/s]
4|1417|Loss: 0.8608571887016296:  22%|██▏       | 1417/6470 [16:17<53:41,  1.57it/s]
4|1418|Loss: 0.9619858860969543:  22%|██▏       | 1417/6470 [16:17<53:41,  1.57it/s]
4|1418|Loss: 0.9619858860969543:  22%|██▏       | 1418/6470 [16:17<55:48,  1.51it/s]
4|1419|Loss: 0.8234186172485352:  22%|██▏       | 1418/6470 [16:18<55:48,  1.51it/s]
4|1419|Loss: 0.8234186172485352:  22%|██▏       | 1419/6470 [16:18<58:01,  1.45it/s]
4|1420|Loss: 0.6555097699165344:  22%|██▏       | 1419/6470 [16:18<58:01,  1.45it/s]
4|1420|Loss: 0.6555097699165344:  22%|██▏       | 1420/6470 [16:19<59:17,  1.42it/s]
4|1421|Loss: 0.8371988534927368:  22%|██▏       | 1420/6470 [16:19<59:17,  1.42it/s]
4|1421|Loss: 0.8371988534927368:  22%|██▏       | 1421/6470 [16:19<1:00:33,  1.39it/s]
4|1422|Loss: 0.8507896065711975:  22%|██▏       | 1421/6470 [16:20<1:00:33,  1.39it/s]
4|1422|Loss: 0.8507896065711975:  22%|██▏       | 1422/6470 [16:20<57:28,  1.46it/s]  
4|1423|Loss: 0.921496570110321:  22%|██▏       | 1422/6470 [16:20<57:28,  1.46it/s] 
4|1423|Loss: 0.921496570110321:  22%|██▏       | 1423/6470 [16:21<54:47,  1.54it/s]
4|1424|Loss: 0.8450042605400085:  22%|██▏       | 1423/6470 [16:21<54:47,  1.54it/s]
4|1424|Loss: 0.8450042605400085:  22%|██▏       | 1424/6470 [16:21<53:47,  1.56it/s]
4|1425|Loss: 0.7273864150047302:  22%|██▏       | 1424/6470 [16:22<53:47,  1.56it/s]
4|1425|Loss: 0.7273864150047302:  22%|██▏       | 1425/6470 [16:22<50:36,  1.66it/s]
4|1426|Loss: 0.8491367697715759:  22%|██▏       | 1425/6470 [16:22<50:36,  1.66it/s]
4|1426|Loss: 0.8491367697715759:  22%|██▏       | 1426/6470 [16:22<48:25,  1.74it/s]
4|1427|Loss: 0.8891156911849976:  22%|██▏       | 1426/6470 [16:23<48:25,  1.74it/s]
4|1427|Loss: 0.8891156911849976:  22%|██▏       | 1427/6470 [16:23<53:27,  1.57it/s]
4|1428|Loss: 0.9045450687408447:  22%|██▏       | 1427/6470 [16:24<53:27,  1.57it/s]
4|1428|Loss: 0.9045450687408447:  22%|██▏       | 1428/6470 [16:24<58:23,  1.44it/s]
4|1429|Loss: 0.8493912816047668:  22%|██▏       | 1428/6470 [16:24<58:23,  1.44it/s]
4|1429|Loss: 0.8493912816047668:  22%|██▏       | 1429/6470 [16:25<57:32,  1.46it/s]
4|1430|Loss: 0.9538874626159668:  22%|██▏       | 1429/6470 [16:25<57:32,  1.46it/s]
4|1430|Loss: 0.9538874626159668:  22%|██▏       | 1430/6470 [16:25<53:00,  1.58it/s]
4|1431|Loss: 0.847013533115387:  22%|██▏       | 1430/6470 [16:25<53:00,  1.58it/s] 
4|1431|Loss: 0.847013533115387:  22%|██▏       | 1431/6470 [16:26<51:21,  1.63it/s]
4|1432|Loss: 0.5866710543632507:  22%|██▏       | 1431/6470 [16:26<51:21,  1.63it/s]
4|1432|Loss: 0.5866710543632507:  22%|██▏       | 1432/6470 [16:26<49:26,  1.70it/s]
4|1433|Loss: 0.9606860280036926:  22%|██▏       | 1432/6470 [16:27<49:26,  1.70it/s]
4|1433|Loss: 0.9606860280036926:  22%|██▏       | 1433/6470 [16:27<53:14,  1.58it/s]
4|1434|Loss: 0.8474152684211731:  22%|██▏       | 1433/6470 [16:27<53:14,  1.58it/s]
4|1434|Loss: 0.8474152684211731:  22%|██▏       | 1434/6470 [16:28<58:01,  1.45it/s]
4|1435|Loss: 0.9047629237174988:  22%|██▏       | 1434/6470 [16:28<58:01,  1.45it/s]
4|1435|Loss: 0.9047629237174988:  22%|██▏       | 1435/6470 [16:28<57:31,  1.46it/s]
4|1436|Loss: 0.7222684621810913:  22%|██▏       | 1435/6470 [16:29<57:31,  1.46it/s]
4|1436|Loss: 0.7222684621810913:  22%|██▏       | 1436/6470 [16:29<54:57,  1.53it/s]
4|1437|Loss: 0.9272922873497009:  22%|██▏       | 1436/6470 [16:29<54:57,  1.53it/s]
4|1437|Loss: 0.9272922873497009:  22%|██▏       | 1437/6470 [16:30<55:52,  1.50it/s]
4|1438|Loss: 0.6871901154518127:  22%|██▏       | 1437/6470 [16:30<55:52,  1.50it/s]
4|1438|Loss: 0.6871901154518127:  22%|██▏       | 1438/6470 [16:31<1:00:22,  1.39it/s]
4|1439|Loss: 0.8801203370094299:  22%|██▏       | 1438/6470 [16:31<1:00:22,  1.39it/s]
4|1439|Loss: 0.8801203370094299:  22%|██▏       | 1439/6470 [16:31<58:46,  1.43it/s]  
4|1440|Loss: 0.820647120475769:  22%|██▏       | 1439/6470 [16:31<58:46,  1.43it/s] 
4|1440|Loss: 0.820647120475769:  22%|██▏       | 1440/6470 [16:32<51:38,  1.62it/s]
4|1441|Loss: 0.8444275259971619:  22%|██▏       | 1440/6470 [16:32<51:38,  1.62it/s]
4|1441|Loss: 0.8444275259971619:  22%|██▏       | 1441/6470 [16:32<54:01,  1.55it/s]
4|1442|Loss: 1.0290619134902954:  22%|██▏       | 1441/6470 [16:33<54:01,  1.55it/s]
4|1442|Loss: 1.0290619134902954:  22%|██▏       | 1442/6470 [16:33<54:26,  1.54it/s]
4|1443|Loss: 0.9717246890068054:  22%|██▏       | 1442/6470 [16:33<54:26,  1.54it/s]
4|1443|Loss: 0.9717246890068054:  22%|██▏       | 1443/6470 [16:34<56:15,  1.49it/s]
4|1444|Loss: 0.8278617262840271:  22%|██▏       | 1443/6470 [16:34<56:15,  1.49it/s]
4|1444|Loss: 0.8278617262840271:  22%|██▏       | 1444/6470 [16:34<58:05,  1.44it/s]
4|1445|Loss: 0.8970044851303101:  22%|██▏       | 1444/6470 [16:35<58:05,  1.44it/s]
4|1445|Loss: 0.8970044851303101:  22%|██▏       | 1445/6470 [16:35<1:02:57,  1.33it/s]
4|1446|Loss: 0.908950924873352:  22%|██▏       | 1445/6470 [16:36<1:02:57,  1.33it/s] 
4|1446|Loss: 0.908950924873352:  22%|██▏       | 1446/6470 [16:36<1:01:53,  1.35it/s]
4|1447|Loss: 0.7806700468063354:  22%|██▏       | 1446/6470 [16:36<1:01:53,  1.35it/s]
4|1447|Loss: 0.7806700468063354:  22%|██▏       | 1447/6470 [16:37<58:53,  1.42it/s]  
4|1448|Loss: 0.7620108127593994:  22%|██▏       | 1447/6470 [16:37<58:53,  1.42it/s]
4|1448|Loss: 0.7620108127593994:  22%|██▏       | 1448/6470 [16:37<56:47,  1.47it/s]
4|1449|Loss: 0.8781470060348511:  22%|██▏       | 1448/6470 [16:38<56:47,  1.47it/s]
4|1449|Loss: 0.8781470060348511:  22%|██▏       | 1449/6470 [16:38<57:57,  1.44it/s]
4|1450|Loss: 0.6656816005706787:  22%|██▏       | 1449/6470 [16:38<57:57,  1.44it/s]
4|1450|Loss: 0.6656816005706787:  22%|██▏       | 1450/6470 [16:39<56:59,  1.47it/s]
4|1451|Loss: 0.8320015668869019:  22%|██▏       | 1450/6470 [16:39<56:59,  1.47it/s]
4|1451|Loss: 0.8320015668869019:  22%|██▏       | 1451/6470 [16:39<57:40,  1.45it/s]
4|1452|Loss: 0.8023990988731384:  22%|██▏       | 1451/6470 [16:40<57:40,  1.45it/s]
4|1452|Loss: 0.8023990988731384:  22%|██▏       | 1452/6470 [16:40<55:35,  1.50it/s]
4|1453|Loss: 0.7936318516731262:  22%|██▏       | 1452/6470 [16:40<55:35,  1.50it/s]
4|1453|Loss: 0.7936318516731262:  22%|██▏       | 1453/6470 [16:41<54:40,  1.53it/s]
4|1454|Loss: 0.909133791923523:  22%|██▏       | 1453/6470 [16:41<54:40,  1.53it/s] 
4|1454|Loss: 0.909133791923523:  22%|██▏       | 1454/6470 [16:41<57:22,  1.46it/s]
4|1455|Loss: 0.8168936371803284:  22%|██▏       | 1454/6470 [16:42<57:22,  1.46it/s]
4|1455|Loss: 0.8168936371803284:  22%|██▏       | 1455/6470 [16:42<56:42,  1.47it/s]
4|1456|Loss: 1.0179507732391357:  22%|██▏       | 1455/6470 [16:42<56:42,  1.47it/s]
4|1456|Loss: 1.0179507732391357:  23%|██▎       | 1456/6470 [16:43<57:48,  1.45it/s]
4|1457|Loss: 0.729522705078125:  23%|██▎       | 1456/6470 [16:43<57:48,  1.45it/s] 
4|1457|Loss: 0.729522705078125:  23%|██▎       | 1457/6470 [16:43<55:57,  1.49it/s]
4|1458|Loss: 0.8791593313217163:  23%|██▎       | 1457/6470 [16:44<55:57,  1.49it/s]
4|1458|Loss: 0.8791593313217163:  23%|██▎       | 1458/6470 [16:44<52:36,  1.59it/s]
4|1459|Loss: 0.9265181422233582:  23%|██▎       | 1458/6470 [16:44<52:36,  1.59it/s]
4|1459|Loss: 0.9265181422233582:  23%|██▎       | 1459/6470 [16:45<53:37,  1.56it/s]
4|1460|Loss: 0.8917483687400818:  23%|██▎       | 1459/6470 [16:45<53:37,  1.56it/s]
4|1460|Loss: 0.8917483687400818:  23%|██▎       | 1460/6470 [16:45<54:01,  1.55it/s]
4|1461|Loss: 0.9010875821113586:  23%|██▎       | 1460/6470 [16:46<54:01,  1.55it/s]
4|1461|Loss: 0.9010875821113586:  23%|██▎       | 1461/6470 [16:46<54:04,  1.54it/s]
4|1462|Loss: 0.8266081213951111:  23%|██▎       | 1461/6470 [16:46<54:04,  1.54it/s]
4|1462|Loss: 0.8266081213951111:  23%|██▎       | 1462/6470 [16:47<55:15,  1.51it/s]
4|1463|Loss: 0.8363049030303955:  23%|██▎       | 1462/6470 [16:47<55:15,  1.51it/s]
4|1463|Loss: 0.8363049030303955:  23%|██▎       | 1463/6470 [16:47<52:20,  1.59it/s]
4|1464|Loss: 0.7888957858085632:  23%|██▎       | 1463/6470 [16:47<52:20,  1.59it/s]
4|1464|Loss: 0.7888957858085632:  23%|██▎       | 1464/6470 [16:48<48:53,  1.71it/s]
4|1465|Loss: 0.7833036780357361:  23%|██▎       | 1464/6470 [16:48<48:53,  1.71it/s]
4|1465|Loss: 0.7833036780357361:  23%|██▎       | 1465/6470 [16:48<49:23,  1.69it/s]
4|1466|Loss: 0.8847360610961914:  23%|██▎       | 1465/6470 [16:49<49:23,  1.69it/s]
4|1466|Loss: 0.8847360610961914:  23%|██▎       | 1466/6470 [16:49<51:46,  1.61it/s]
4|1467|Loss: 0.8930149078369141:  23%|██▎       | 1466/6470 [16:49<51:46,  1.61it/s]
4|1467|Loss: 0.8930149078369141:  23%|██▎       | 1467/6470 [16:50<57:22,  1.45it/s]
4|1468|Loss: 0.7354456186294556:  23%|██▎       | 1467/6470 [16:50<57:22,  1.45it/s]
4|1468|Loss: 0.7354456186294556:  23%|██▎       | 1468/6470 [16:50<57:18,  1.45it/s]
4|1469|Loss: 0.8198925852775574:  23%|██▎       | 1468/6470 [16:51<57:18,  1.45it/s]
4|1469|Loss: 0.8198925852775574:  23%|██▎       | 1469/6470 [16:51<58:12,  1.43it/s]
4|1470|Loss: 0.9012765288352966:  23%|██▎       | 1469/6470 [16:52<58:12,  1.43it/s]
4|1470|Loss: 0.9012765288352966:  23%|██▎       | 1470/6470 [16:52<57:30,  1.45it/s]
4|1471|Loss: 0.916797399520874:  23%|██▎       | 1470/6470 [16:52<57:30,  1.45it/s] 
4|1471|Loss: 0.916797399520874:  23%|██▎       | 1471/6470 [16:53<1:00:30,  1.38it/s]
4|1472|Loss: 0.8931427001953125:  23%|██▎       | 1471/6470 [16:53<1:00:30,  1.38it/s]
4|1472|Loss: 0.8931427001953125:  23%|██▎       | 1472/6470 [16:54<1:03:54,  1.30it/s]
4|1473|Loss: 0.9017096161842346:  23%|██▎       | 1472/6470 [16:54<1:03:54,  1.30it/s]
4|1473|Loss: 0.9017096161842346:  23%|██▎       | 1473/6470 [16:54<1:05:02,  1.28it/s]
4|1474|Loss: 0.8983568549156189:  23%|██▎       | 1473/6470 [16:55<1:05:02,  1.28it/s]
4|1474|Loss: 0.8983568549156189:  23%|██▎       | 1474/6470 [16:55<1:02:30,  1.33it/s]
4|1475|Loss: 1.0788767337799072:  23%|██▎       | 1474/6470 [16:55<1:02:30,  1.33it/s]
4|1475|Loss: 1.0788767337799072:  23%|██▎       | 1475/6470 [16:56<1:01:55,  1.34it/s]
4|1476|Loss: 0.9974474310874939:  23%|██▎       | 1475/6470 [16:56<1:01:55,  1.34it/s]
4|1476|Loss: 0.9974474310874939:  23%|██▎       | 1476/6470 [16:56<59:58,  1.39it/s]  
4|1477|Loss: 0.8535791039466858:  23%|██▎       | 1476/6470 [16:57<59:58,  1.39it/s]
4|1477|Loss: 0.8535791039466858:  23%|██▎       | 1477/6470 [16:57<54:12,  1.54it/s]
4|1478|Loss: 0.9237060546875:  23%|██▎       | 1477/6470 [16:57<54:12,  1.54it/s]   
4|1478|Loss: 0.9237060546875:  23%|██▎       | 1478/6470 [16:58<54:08,  1.54it/s]
4|1479|Loss: 0.9096367955207825:  23%|██▎       | 1478/6470 [16:58<54:08,  1.54it/s]
4|1479|Loss: 0.9096367955207825:  23%|██▎       | 1479/6470 [16:58<55:33,  1.50it/s]
4|1480|Loss: 0.8465705513954163:  23%|██▎       | 1479/6470 [16:59<55:33,  1.50it/s]
4|1480|Loss: 0.8465705513954163:  23%|██▎       | 1480/6470 [16:59<52:32,  1.58it/s]
4|1481|Loss: 0.9579105973243713:  23%|██▎       | 1480/6470 [16:59<52:32,  1.58it/s]
4|1481|Loss: 0.9579105973243713:  23%|██▎       | 1481/6470 [16:59<53:14,  1.56it/s]
4|1482|Loss: 0.8255606293678284:  23%|██▎       | 1481/6470 [17:00<53:14,  1.56it/s]
4|1482|Loss: 0.8255606293678284:  23%|██▎       | 1482/6470 [17:00<49:50,  1.67it/s]
4|1483|Loss: 0.8452683091163635:  23%|██▎       | 1482/6470 [17:00<49:50,  1.67it/s]
4|1483|Loss: 0.8452683091163635:  23%|██▎       | 1483/6470 [17:01<51:02,  1.63it/s]
4|1484|Loss: 0.7470980286598206:  23%|██▎       | 1483/6470 [17:01<51:02,  1.63it/s]
4|1484|Loss: 0.7470980286598206:  23%|██▎       | 1484/6470 [17:01<53:55,  1.54it/s]
4|1485|Loss: 0.7783463597297668:  23%|██▎       | 1484/6470 [17:02<53:55,  1.54it/s]
4|1485|Loss: 0.7783463597297668:  23%|██▎       | 1485/6470 [17:02<53:43,  1.55it/s]
4|1486|Loss: 0.8031324744224548:  23%|██▎       | 1485/6470 [17:02<53:43,  1.55it/s]
4|1486|Loss: 0.8031324744224548:  23%|██▎       | 1486/6470 [17:03<52:54,  1.57it/s]
4|1487|Loss: 0.9800593852996826:  23%|██▎       | 1486/6470 [17:03<52:54,  1.57it/s]
4|1487|Loss: 0.9800593852996826:  23%|██▎       | 1487/6470 [17:03<56:31,  1.47it/s]
4|1488|Loss: 0.7968854904174805:  23%|██▎       | 1487/6470 [17:04<56:31,  1.47it/s]
4|1488|Loss: 0.7968854904174805:  23%|██▎       | 1488/6470 [17:04<53:49,  1.54it/s]
4|1489|Loss: 0.8688319325447083:  23%|██▎       | 1488/6470 [17:04<53:49,  1.54it/s]
4|1489|Loss: 0.8688319325447083:  23%|██▎       | 1489/6470 [17:05<56:11,  1.48it/s]
4|1490|Loss: 0.9098078608512878:  23%|██▎       | 1489/6470 [17:05<56:11,  1.48it/s]
4|1490|Loss: 0.9098078608512878:  23%|██▎       | 1490/6470 [17:05<55:51,  1.49it/s]
4|1491|Loss: 0.8275672197341919:  23%|██▎       | 1490/6470 [17:06<55:51,  1.49it/s]
4|1491|Loss: 0.8275672197341919:  23%|██▎       | 1491/6470 [17:06<56:12,  1.48it/s]
4|1492|Loss: 0.7462578415870667:  23%|██▎       | 1491/6470 [17:07<56:12,  1.48it/s]
4|1492|Loss: 0.7462578415870667:  23%|██▎       | 1492/6470 [17:07<1:00:14,  1.38it/s]
4|1493|Loss: 0.768170952796936:  23%|██▎       | 1492/6470 [17:07<1:00:14,  1.38it/s] 
4|1493|Loss: 0.768170952796936:  23%|██▎       | 1493/6470 [17:08<1:00:41,  1.37it/s]
4|1494|Loss: 0.9396277070045471:  23%|██▎       | 1493/6470 [17:08<1:00:41,  1.37it/s]
4|1494|Loss: 0.9396277070045471:  23%|██▎       | 1494/6470 [17:08<1:01:38,  1.35it/s]
4|1495|Loss: 0.9507834911346436:  23%|██▎       | 1494/6470 [17:09<1:01:38,  1.35it/s]
4|1495|Loss: 0.9507834911346436:  23%|██▎       | 1495/6470 [17:09<1:01:44,  1.34it/s]
4|1496|Loss: 0.6539605259895325:  23%|██▎       | 1495/6470 [17:10<1:01:44,  1.34it/s]
4|1496|Loss: 0.6539605259895325:  23%|██▎       | 1496/6470 [17:11<1:25:31,  1.03s/it]
4|1497|Loss: 0.8999956846237183:  23%|██▎       | 1496/6470 [17:11<1:25:31,  1.03s/it]
4|1497|Loss: 0.8999956846237183:  23%|██▎       | 1497/6470 [17:12<1:19:49,  1.04it/s]
4|1498|Loss: 0.8956628441810608:  23%|██▎       | 1497/6470 [17:12<1:19:49,  1.04it/s]
4|1498|Loss: 0.8956628441810608:  23%|██▎       | 1498/6470 [17:12<1:12:42,  1.14it/s]
4|1499|Loss: 0.9310555458068848:  23%|██▎       | 1498/6470 [17:13<1:12:42,  1.14it/s]
4|1499|Loss: 0.9310555458068848:  23%|██▎       | 1499/6470 [17:13<1:03:24,  1.31it/s]
4|1500|Loss: 0.8902484774589539:  23%|██▎       | 1499/6470 [17:13<1:03:24,  1.31it/s]
4|1500|Loss: 0.8902484774589539:  23%|██▎       | 1500/6470 [17:14<1:02:26,  1.33it/s]
4|1501|Loss: 1.0152137279510498:  23%|██▎       | 1500/6470 [17:14<1:02:26,  1.33it/s]
4|1501|Loss: 1.0152137279510498:  23%|██▎       | 1501/6470 [17:14<59:20,  1.40it/s]  
4|1502|Loss: 0.8423166275024414:  23%|██▎       | 1501/6470 [17:15<59:20,  1.40it/s]
4|1502|Loss: 0.8423166275024414:  23%|██▎       | 1502/6470 [17:15<57:13,  1.45it/s]
4|1503|Loss: 0.8451981544494629:  23%|██▎       | 1502/6470 [17:15<57:13,  1.45it/s]
4|1503|Loss: 0.8451981544494629:  23%|██▎       | 1503/6470 [17:16<1:00:33,  1.37it/s]
4|1504|Loss: 0.9787277579307556:  23%|██▎       | 1503/6470 [17:16<1:00:33,  1.37it/s]
4|1504|Loss: 0.9787277579307556:  23%|██▎       | 1504/6470 [17:16<57:54,  1.43it/s]  
4|1505|Loss: 0.6229701042175293:  23%|██▎       | 1504/6470 [17:17<57:54,  1.43it/s]
4|1505|Loss: 0.6229701042175293:  23%|██▎       | 1505/6470 [17:17<54:44,  1.51it/s]
4|1506|Loss: 0.8634077310562134:  23%|██▎       | 1505/6470 [17:17<54:44,  1.51it/s]
4|1506|Loss: 0.8634077310562134:  23%|██▎       | 1506/6470 [17:18<55:16,  1.50it/s]
4|1507|Loss: 0.881844162940979:  23%|██▎       | 1506/6470 [17:18<55:16,  1.50it/s] 
4|1507|Loss: 0.881844162940979:  23%|██▎       | 1507/6470 [17:18<54:32,  1.52it/s]
4|1508|Loss: 0.8968343138694763:  23%|██▎       | 1507/6470 [17:19<54:32,  1.52it/s]
4|1508|Loss: 0.8968343138694763:  23%|██▎       | 1508/6470 [17:19<53:23,  1.55it/s]
4|1509|Loss: 0.961733877658844:  23%|██▎       | 1508/6470 [17:19<53:23,  1.55it/s] 
4|1509|Loss: 0.961733877658844:  23%|██▎       | 1509/6470 [17:20<56:08,  1.47it/s]
4|1510|Loss: 0.9747005701065063:  23%|██▎       | 1509/6470 [17:20<56:08,  1.47it/s]
4|1510|Loss: 0.9747005701065063:  23%|██▎       | 1510/6470 [17:20<57:52,  1.43it/s]
4|1511|Loss: 0.7866418361663818:  23%|██▎       | 1510/6470 [17:21<57:52,  1.43it/s]
4|1511|Loss: 0.7866418361663818:  23%|██▎       | 1511/6470 [17:21<56:49,  1.45it/s]
4|1512|Loss: 0.9250443577766418:  23%|██▎       | 1511/6470 [17:21<56:49,  1.45it/s]
4|1512|Loss: 0.9250443577766418:  23%|██▎       | 1512/6470 [17:22<53:54,  1.53it/s]
4|1513|Loss: 0.9358748197555542:  23%|██▎       | 1512/6470 [17:22<53:54,  1.53it/s]
4|1513|Loss: 0.9358748197555542:  23%|██▎       | 1513/6470 [17:22<57:32,  1.44it/s]
4|1514|Loss: 0.7873204946517944:  23%|██▎       | 1513/6470 [17:23<57:32,  1.44it/s]
4|1514|Loss: 0.7873204946517944:  23%|██▎       | 1514/6470 [17:23<58:31,  1.41it/s]
4|1515|Loss: 0.861022412776947:  23%|██▎       | 1514/6470 [17:24<58:31,  1.41it/s] 
4|1515|Loss: 0.861022412776947:  23%|██▎       | 1515/6470 [17:24<1:01:49,  1.34it/s]
4|1516|Loss: 0.9631212949752808:  23%|██▎       | 1515/6470 [17:24<1:01:49,  1.34it/s]
4|1516|Loss: 0.9631212949752808:  23%|██▎       | 1516/6470 [17:25<1:00:01,  1.38it/s]
4|1517|Loss: 0.9022251963615417:  23%|██▎       | 1516/6470 [17:25<1:00:01,  1.38it/s]
4|1517|Loss: 0.9022251963615417:  23%|██▎       | 1517/6470 [17:25<57:44,  1.43it/s]  
4|1518|Loss: 0.731726884841919:  23%|██▎       | 1517/6470 [17:26<57:44,  1.43it/s] 
4|1518|Loss: 0.731726884841919:  23%|██▎       | 1518/6470 [17:26<1:00:33,  1.36it/s]
4|1519|Loss: 0.8500784635543823:  23%|██▎       | 1518/6470 [17:26<1:00:33,  1.36it/s]
4|1519|Loss: 0.8500784635543823:  23%|██▎       | 1519/6470 [17:27<59:00,  1.40it/s]  
4|1520|Loss: 0.8926512002944946:  23%|██▎       | 1519/6470 [17:27<59:00,  1.40it/s]
4|1520|Loss: 0.8926512002944946:  23%|██▎       | 1520/6470 [17:28<1:01:14,  1.35it/s]
4|1521|Loss: 0.8953981399536133:  23%|██▎       | 1520/6470 [17:28<1:01:14,  1.35it/s]
4|1521|Loss: 0.8953981399536133:  24%|██▎       | 1521/6470 [17:28<1:01:46,  1.34it/s]
4|1522|Loss: 0.798130989074707:  24%|██▎       | 1521/6470 [17:29<1:01:46,  1.34it/s] 
4|1522|Loss: 0.798130989074707:  24%|██▎       | 1522/6470 [17:29<58:32,  1.41it/s]  
4|1523|Loss: 0.8852675557136536:  24%|██▎       | 1522/6470 [17:29<58:32,  1.41it/s]
4|1523|Loss: 0.8852675557136536:  24%|██▎       | 1523/6470 [17:30<57:42,  1.43it/s]
4|1524|Loss: 0.8747386336326599:  24%|██▎       | 1523/6470 [17:30<57:42,  1.43it/s]
4|1524|Loss: 0.8747386336326599:  24%|██▎       | 1524/6470 [17:30<58:46,  1.40it/s]
4|1525|Loss: 0.7657264471054077:  24%|██▎       | 1524/6470 [17:31<58:46,  1.40it/s]
4|1525|Loss: 0.7657264471054077:  24%|██▎       | 1525/6470 [17:31<1:00:34,  1.36it/s]
4|1526|Loss: 0.9476300477981567:  24%|██▎       | 1525/6470 [17:32<1:00:34,  1.36it/s]
4|1526|Loss: 0.9476300477981567:  24%|██▎       | 1526/6470 [17:32<59:03,  1.40it/s]  
4|1527|Loss: 0.9002445340156555:  24%|██▎       | 1526/6470 [17:32<59:03,  1.40it/s]
4|1527|Loss: 0.9002445340156555:  24%|██▎       | 1527/6470 [17:32<56:34,  1.46it/s]
4|1528|Loss: 0.8446332216262817:  24%|██▎       | 1527/6470 [17:33<56:34,  1.46it/s]
4|1528|Loss: 0.8446332216262817:  24%|██▎       | 1528/6470 [17:33<53:57,  1.53it/s]
4|1529|Loss: 0.8977029919624329:  24%|██▎       | 1528/6470 [17:33<53:57,  1.53it/s]
4|1529|Loss: 0.8977029919624329:  24%|██▎       | 1529/6470 [17:34<55:23,  1.49it/s]
4|1530|Loss: 0.5832058191299438:  24%|██▎       | 1529/6470 [17:34<55:23,  1.49it/s]
4|1530|Loss: 0.5832058191299438:  24%|██▎       | 1530/6470 [17:35<59:53,  1.37it/s]
4|1531|Loss: 0.6275193095207214:  24%|██▎       | 1530/6470 [17:35<59:53,  1.37it/s]
4|1531|Loss: 0.6275193095207214:  24%|██▎       | 1531/6470 [17:35<58:44,  1.40it/s]
4|1532|Loss: 0.8776074051856995:  24%|██▎       | 1531/6470 [17:36<58:44,  1.40it/s]
4|1532|Loss: 0.8776074051856995:  24%|██▎       | 1532/6470 [17:36<1:00:48,  1.35it/s]
4|1533|Loss: 0.9645852446556091:  24%|██▎       | 1532/6470 [17:36<1:00:48,  1.35it/s]
4|1533|Loss: 0.9645852446556091:  24%|██▎       | 1533/6470 [17:37<1:03:55,  1.29it/s]
4|1534|Loss: 0.9517320394515991:  24%|██▎       | 1533/6470 [17:37<1:03:55,  1.29it/s]
4|1534|Loss: 0.9517320394515991:  24%|██▎       | 1534/6470 [17:38<1:04:40,  1.27it/s]
4|1535|Loss: 0.8200682401657104:  24%|██▎       | 1534/6470 [17:38<1:04:40,  1.27it/s]
4|1535|Loss: 0.8200682401657104:  24%|██▎       | 1535/6470 [17:38<56:30,  1.46it/s]  
4|1536|Loss: 0.8756021857261658:  24%|██▎       | 1535/6470 [17:38<56:30,  1.46it/s]
4|1536|Loss: 0.8756021857261658:  24%|██▎       | 1536/6470 [17:39<50:14,  1.64it/s]
4|1537|Loss: 0.8637983202934265:  24%|██▎       | 1536/6470 [17:39<50:14,  1.64it/s]
4|1537|Loss: 0.8637983202934265:  24%|██▍       | 1537/6470 [17:39<49:47,  1.65it/s]
4|1538|Loss: 0.7628921270370483:  24%|██▍       | 1537/6470 [17:40<49:47,  1.65it/s]
4|1538|Loss: 0.7628921270370483:  24%|██▍       | 1538/6470 [17:40<50:38,  1.62it/s]
4|1539|Loss: 0.6706499457359314:  24%|██▍       | 1538/6470 [17:40<50:38,  1.62it/s]
4|1539|Loss: 0.6706499457359314:  24%|██▍       | 1539/6470 [17:41<57:27,  1.43it/s]
4|1540|Loss: 0.9136953353881836:  24%|██▍       | 1539/6470 [17:41<57:27,  1.43it/s]
4|1540|Loss: 0.9136953353881836:  24%|██▍       | 1540/6470 [17:41<57:18,  1.43it/s]
4|1541|Loss: 0.7160319685935974:  24%|██▍       | 1540/6470 [17:42<57:18,  1.43it/s]
4|1541|Loss: 0.7160319685935974:  24%|██▍       | 1541/6470 [17:42<57:47,  1.42it/s]
4|1542|Loss: 0.7850564122200012:  24%|██▍       | 1541/6470 [17:42<57:47,  1.42it/s]
4|1542|Loss: 0.7850564122200012:  24%|██▍       | 1542/6470 [17:43<54:14,  1.51it/s]
4|1543|Loss: 0.7495784163475037:  24%|██▍       | 1542/6470 [17:43<54:14,  1.51it/s]
4|1543|Loss: 0.7495784163475037:  24%|██▍       | 1543/6470 [17:43<57:45,  1.42it/s]
4|1544|Loss: 0.7618375420570374:  24%|██▍       | 1543/6470 [17:44<57:45,  1.42it/s]
4|1544|Loss: 0.7618375420570374:  24%|██▍       | 1544/6470 [17:44<57:04,  1.44it/s]
4|1545|Loss: 0.9593650102615356:  24%|██▍       | 1544/6470 [17:45<57:04,  1.44it/s]
4|1545|Loss: 0.9593650102615356:  24%|██▍       | 1545/6470 [17:45<56:46,  1.45it/s]
4|1546|Loss: 0.7390050888061523:  24%|██▍       | 1545/6470 [17:45<56:46,  1.45it/s]
4|1546|Loss: 0.7390050888061523:  24%|██▍       | 1546/6470 [17:46<55:52,  1.47it/s]
4|1547|Loss: 0.7524166703224182:  24%|██▍       | 1546/6470 [17:46<55:52,  1.47it/s]
4|1547|Loss: 0.7524166703224182:  24%|██▍       | 1547/6470 [17:46<56:59,  1.44it/s]
4|1548|Loss: 0.8656867742538452:  24%|██▍       | 1547/6470 [17:47<56:59,  1.44it/s]
4|1548|Loss: 0.8656867742538452:  24%|██▍       | 1548/6470 [17:47<56:00,  1.46it/s]
4|1549|Loss: 0.8218249678611755:  24%|██▍       | 1548/6470 [17:47<56:00,  1.46it/s]
4|1549|Loss: 0.8218249678611755:  24%|██▍       | 1549/6470 [17:47<51:44,  1.59it/s]
4|1550|Loss: 0.8769387602806091:  24%|██▍       | 1549/6470 [17:48<51:44,  1.59it/s]
4|1550|Loss: 0.8769387602806091:  24%|██▍       | 1550/6470 [17:48<52:52,  1.55it/s]
4|1551|Loss: 0.7571201324462891:  24%|██▍       | 1550/6470 [17:49<52:52,  1.55it/s]
4|1551|Loss: 0.7571201324462891:  24%|██▍       | 1551/6470 [17:49<58:50,  1.39it/s]
4|1552|Loss: 0.9789953827857971:  24%|██▍       | 1551/6470 [17:49<58:50,  1.39it/s]
4|1552|Loss: 0.9789953827857971:  24%|██▍       | 1552/6470 [17:50<55:31,  1.48it/s]
4|1553|Loss: 0.8531922698020935:  24%|██▍       | 1552/6470 [17:50<55:31,  1.48it/s]
4|1553|Loss: 0.8531922698020935:  24%|██▍       | 1553/6470 [17:50<53:47,  1.52it/s]
4|1554|Loss: 0.8501192927360535:  24%|██▍       | 1553/6470 [17:50<53:47,  1.52it/s]
4|1554|Loss: 0.8501192927360535:  24%|██▍       | 1554/6470 [17:51<50:21,  1.63it/s]
4|1555|Loss: 0.859642744064331:  24%|██▍       | 1554/6470 [17:51<50:21,  1.63it/s] 
4|1555|Loss: 0.859642744064331:  24%|██▍       | 1555/6470 [17:51<48:52,  1.68it/s]
4|1556|Loss: 0.850127100944519:  24%|██▍       | 1555/6470 [17:52<48:52,  1.68it/s]
4|1556|Loss: 0.850127100944519:  24%|██▍       | 1556/6470 [17:52<47:50,  1.71it/s]
4|1557|Loss: 0.8376286625862122:  24%|██▍       | 1556/6470 [17:52<47:50,  1.71it/s]
4|1557|Loss: 0.8376286625862122:  24%|██▍       | 1557/6470 [17:52<50:51,  1.61it/s]
4|1558|Loss: 0.8487767577171326:  24%|██▍       | 1557/6470 [17:53<50:51,  1.61it/s]
4|1558|Loss: 0.8487767577171326:  24%|██▍       | 1558/6470 [17:53<54:50,  1.49it/s]
4|1559|Loss: 0.840708315372467:  24%|██▍       | 1558/6470 [17:54<54:50,  1.49it/s] 
4|1559|Loss: 0.840708315372467:  24%|██▍       | 1559/6470 [17:54<57:04,  1.43it/s]
4|1560|Loss: 0.7872896790504456:  24%|██▍       | 1559/6470 [17:54<57:04,  1.43it/s]
4|1560|Loss: 0.7872896790504456:  24%|██▍       | 1560/6470 [17:55<1:01:34,  1.33it/s]
4|1561|Loss: 0.8757839798927307:  24%|██▍       | 1560/6470 [17:55<1:01:34,  1.33it/s]
4|1561|Loss: 0.8757839798927307:  24%|██▍       | 1561/6470 [17:56<1:03:14,  1.29it/s]
4|1562|Loss: 0.82609623670578:  24%|██▍       | 1561/6470 [17:56<1:03:14,  1.29it/s]  
4|1562|Loss: 0.82609623670578:  24%|██▍       | 1562/6470 [17:56<1:02:43,  1.30it/s]
4|1563|Loss: 0.8380099534988403:  24%|██▍       | 1562/6470 [17:57<1:02:43,  1.30it/s]
4|1563|Loss: 0.8380099534988403:  24%|██▍       | 1563/6470 [17:57<1:05:27,  1.25it/s]
4|1564|Loss: 0.7837841510772705:  24%|██▍       | 1563/6470 [17:58<1:05:27,  1.25it/s]
4|1564|Loss: 0.7837841510772705:  24%|██▍       | 1564/6470 [17:58<1:05:57,  1.24it/s]
4|1565|Loss: 0.8457039594650269:  24%|██▍       | 1564/6470 [17:59<1:05:57,  1.24it/s]
4|1565|Loss: 0.8457039594650269:  24%|██▍       | 1565/6470 [17:59<1:02:51,  1.30it/s]
4|1566|Loss: 0.7337409257888794:  24%|██▍       | 1565/6470 [17:59<1:02:51,  1.30it/s]
4|1566|Loss: 0.7337409257888794:  24%|██▍       | 1566/6470 [18:00<1:03:52,  1.28it/s]
4|1567|Loss: 0.8420631885528564:  24%|██▍       | 1566/6470 [18:00<1:03:52,  1.28it/s]
4|1567|Loss: 0.8420631885528564:  24%|██▍       | 1567/6470 [18:00<1:03:36,  1.28it/s]
4|1568|Loss: 0.7732219696044922:  24%|██▍       | 1567/6470 [18:01<1:03:36,  1.28it/s]
4|1568|Loss: 0.7732219696044922:  24%|██▍       | 1568/6470 [18:01<1:02:00,  1.32it/s]
4|1569|Loss: 0.7048576474189758:  24%|██▍       | 1568/6470 [18:02<1:02:00,  1.32it/s]
4|1569|Loss: 0.7048576474189758:  24%|██▍       | 1569/6470 [18:02<1:00:40,  1.35it/s]
4|1570|Loss: 0.858283519744873:  24%|██▍       | 1569/6470 [18:02<1:00:40,  1.35it/s] 
4|1570|Loss: 0.858283519744873:  24%|██▍       | 1570/6470 [18:02<57:26,  1.42it/s]  
4|1571|Loss: 0.9693609476089478:  24%|██▍       | 1570/6470 [18:03<57:26,  1.42it/s]
4|1571|Loss: 0.9693609476089478:  24%|██▍       | 1571/6470 [18:03<53:53,  1.51it/s]
4|1572|Loss: 0.7311405539512634:  24%|██▍       | 1571/6470 [18:03<53:53,  1.51it/s]
4|1572|Loss: 0.7311405539512634:  24%|██▍       | 1572/6470 [18:04<53:01,  1.54it/s]
4|1573|Loss: 1.0087565183639526:  24%|██▍       | 1572/6470 [18:04<53:01,  1.54it/s]
4|1573|Loss: 1.0087565183639526:  24%|██▍       | 1573/6470 [18:04<50:43,  1.61it/s]
4|1574|Loss: 0.8457419872283936:  24%|██▍       | 1573/6470 [18:05<50:43,  1.61it/s]
4|1574|Loss: 0.8457419872283936:  24%|██▍       | 1574/6470 [18:05<50:41,  1.61it/s]
4|1575|Loss: 0.8235962390899658:  24%|██▍       | 1574/6470 [18:05<50:41,  1.61it/s]
4|1575|Loss: 0.8235962390899658:  24%|██▍       | 1575/6470 [18:05<51:14,  1.59it/s]
4|1576|Loss: 0.8924225568771362:  24%|██▍       | 1575/6470 [18:06<51:14,  1.59it/s]
4|1576|Loss: 0.8924225568771362:  24%|██▍       | 1576/6470 [18:06<52:49,  1.54it/s]
4|1577|Loss: 0.8342528343200684:  24%|██▍       | 1576/6470 [18:07<52:49,  1.54it/s]
4|1577|Loss: 0.8342528343200684:  24%|██▍       | 1577/6470 [18:07<54:03,  1.51it/s]
4|1578|Loss: 0.9081286787986755:  24%|██▍       | 1577/6470 [18:07<54:03,  1.51it/s]
4|1578|Loss: 0.9081286787986755:  24%|██▍       | 1578/6470 [18:07<52:52,  1.54it/s]
4|1579|Loss: 0.7932062745094299:  24%|██▍       | 1578/6470 [18:08<52:52,  1.54it/s]
4|1579|Loss: 0.7932062745094299:  24%|██▍       | 1579/6470 [18:08<47:03,  1.73it/s]
4|1580|Loss: 1.016637921333313:  24%|██▍       | 1579/6470 [18:08<47:03,  1.73it/s] 
4|1580|Loss: 1.016637921333313:  24%|██▍       | 1580/6470 [18:09<50:52,  1.60it/s]
4|1581|Loss: 0.8952864408493042:  24%|██▍       | 1580/6470 [18:09<50:52,  1.60it/s]
4|1581|Loss: 0.8952864408493042:  24%|██▍       | 1581/6470 [18:09<50:04,  1.63it/s]
4|1582|Loss: 0.824140191078186:  24%|██▍       | 1581/6470 [18:10<50:04,  1.63it/s] 
4|1582|Loss: 0.824140191078186:  24%|██▍       | 1582/6470 [18:10<52:13,  1.56it/s]
4|1583|Loss: 0.8408617973327637:  24%|██▍       | 1582/6470 [18:10<52:13,  1.56it/s]
4|1583|Loss: 0.8408617973327637:  24%|██▍       | 1583/6470 [18:11<51:42,  1.58it/s]
4|1584|Loss: 0.8079413175582886:  24%|██▍       | 1583/6470 [18:11<51:42,  1.58it/s]
4|1584|Loss: 0.8079413175582886:  24%|██▍       | 1584/6470 [18:11<49:50,  1.63it/s]
4|1585|Loss: 0.8902248740196228:  24%|██▍       | 1584/6470 [18:12<49:50,  1.63it/s]
4|1585|Loss: 0.8902248740196228:  24%|██▍       | 1585/6470 [18:12<53:09,  1.53it/s]
4|1586|Loss: 0.9549282193183899:  24%|██▍       | 1585/6470 [18:12<53:09,  1.53it/s]
4|1586|Loss: 0.9549282193183899:  25%|██▍       | 1586/6470 [18:13<55:50,  1.46it/s]
4|1587|Loss: 0.9089211225509644:  25%|██▍       | 1586/6470 [18:13<55:50,  1.46it/s]
4|1587|Loss: 0.9089211225509644:  25%|██▍       | 1587/6470 [18:13<56:33,  1.44it/s]
4|1588|Loss: 0.9950436353683472:  25%|██▍       | 1587/6470 [18:14<56:33,  1.44it/s]
4|1588|Loss: 0.9950436353683472:  25%|██▍       | 1588/6470 [18:14<56:33,  1.44it/s]
4|1589|Loss: 0.6523668766021729:  25%|██▍       | 1588/6470 [18:14<56:33,  1.44it/s]
4|1589|Loss: 0.6523668766021729:  25%|██▍       | 1589/6470 [18:15<55:33,  1.46it/s]
4|1590|Loss: 0.6201691031455994:  25%|██▍       | 1589/6470 [18:15<55:33,  1.46it/s]
4|1590|Loss: 0.6201691031455994:  25%|██▍       | 1590/6470 [18:15<55:45,  1.46it/s]
4|1591|Loss: 0.8339006900787354:  25%|██▍       | 1590/6470 [18:16<55:45,  1.46it/s]
4|1591|Loss: 0.8339006900787354:  25%|██▍       | 1591/6470 [18:16<56:11,  1.45it/s]
4|1592|Loss: 0.734774649143219:  25%|██▍       | 1591/6470 [18:16<56:11,  1.45it/s] 
4|1592|Loss: 0.734774649143219:  25%|██▍       | 1592/6470 [18:17<56:30,  1.44it/s]
4|1593|Loss: 0.9687778949737549:  25%|██▍       | 1592/6470 [18:17<56:30,  1.44it/s]
4|1593|Loss: 0.9687778949737549:  25%|██▍       | 1593/6470 [18:17<53:39,  1.52it/s]
4|1594|Loss: 1.027458667755127:  25%|██▍       | 1593/6470 [18:18<53:39,  1.52it/s] 
4|1594|Loss: 1.027458667755127:  25%|██▍       | 1594/6470 [18:18<54:49,  1.48it/s]
4|1595|Loss: 0.7316398620605469:  25%|██▍       | 1594/6470 [18:19<54:49,  1.48it/s]
4|1595|Loss: 0.7316398620605469:  25%|██▍       | 1595/6470 [18:19<1:01:20,  1.32it/s]
4|1596|Loss: 0.9501203298568726:  25%|██▍       | 1595/6470 [18:19<1:01:20,  1.32it/s]
4|1596|Loss: 0.9501203298568726:  25%|██▍       | 1596/6470 [18:20<1:02:11,  1.31it/s]
4|1597|Loss: 0.8099412322044373:  25%|██▍       | 1596/6470 [18:20<1:02:11,  1.31it/s]
4|1597|Loss: 0.8099412322044373:  25%|██▍       | 1597/6470 [18:21<1:01:23,  1.32it/s]
4|1598|Loss: 0.8008682131767273:  25%|██▍       | 1597/6470 [18:21<1:01:23,  1.32it/s]
4|1598|Loss: 0.8008682131767273:  25%|██▍       | 1598/6470 [18:21<58:53,  1.38it/s]  
4|1599|Loss: 0.7835513949394226:  25%|██▍       | 1598/6470 [18:22<58:53,  1.38it/s]
4|1599|Loss: 0.7835513949394226:  25%|██▍       | 1599/6470 [18:22<56:16,  1.44it/s]
4|1600|Loss: 0.8463979363441467:  25%|██▍       | 1599/6470 [18:22<56:16,  1.44it/s]
4|1600|Loss: 0.8463979363441467:  25%|██▍       | 1600/6470 [18:23<56:29,  1.44it/s]
4|1601|Loss: 0.9443172216415405:  25%|██▍       | 1600/6470 [18:23<56:29,  1.44it/s]
4|1601|Loss: 0.9443172216415405:  25%|██▍       | 1601/6470 [18:23<56:50,  1.43it/s]
4|1602|Loss: 0.867770791053772:  25%|██▍       | 1601/6470 [18:24<56:50,  1.43it/s] 
4|1602|Loss: 0.867770791053772:  25%|██▍       | 1602/6470 [18:24<54:20,  1.49it/s]
4|1603|Loss: 0.8314900994300842:  25%|██▍       | 1602/6470 [18:24<54:20,  1.49it/s]
4|1603|Loss: 0.8314900994300842:  25%|██▍       | 1603/6470 [18:25<56:51,  1.43it/s]
4|1604|Loss: 0.9046644568443298:  25%|██▍       | 1603/6470 [18:25<56:51,  1.43it/s]
4|1604|Loss: 0.9046644568443298:  25%|██▍       | 1604/6470 [18:25<57:33,  1.41it/s]
4|1605|Loss: 1.018994927406311:  25%|██▍       | 1604/6470 [18:26<57:33,  1.41it/s] 
4|1605|Loss: 1.018994927406311:  25%|██▍       | 1605/6470 [18:26<55:48,  1.45it/s]
4|1606|Loss: 0.8560733795166016:  25%|██▍       | 1605/6470 [18:26<55:48,  1.45it/s]
4|1606|Loss: 0.8560733795166016:  25%|██▍       | 1606/6470 [18:27<58:10,  1.39it/s]
4|1607|Loss: 0.8550980091094971:  25%|██▍       | 1606/6470 [18:27<58:10,  1.39it/s]
4|1607|Loss: 0.8550980091094971:  25%|██▍       | 1607/6470 [18:27<58:20,  1.39it/s]
4|1608|Loss: 0.9318583607673645:  25%|██▍       | 1607/6470 [18:28<58:20,  1.39it/s]
4|1608|Loss: 0.9318583607673645:  25%|██▍       | 1608/6470 [18:28<59:35,  1.36it/s]
4|1609|Loss: 0.9306344389915466:  25%|██▍       | 1608/6470 [18:29<59:35,  1.36it/s]
4|1609|Loss: 0.9306344389915466:  25%|██▍       | 1609/6470 [18:29<1:02:49,  1.29it/s]
4|1610|Loss: 0.9344881176948547:  25%|██▍       | 1609/6470 [18:30<1:02:49,  1.29it/s]
4|1610|Loss: 0.9344881176948547:  25%|██▍       | 1610/6470 [18:30<1:02:11,  1.30it/s]
4|1611|Loss: 0.8209435939788818:  25%|██▍       | 1610/6470 [18:30<1:02:11,  1.30it/s]
4|1611|Loss: 0.8209435939788818:  25%|██▍       | 1611/6470 [18:31<1:02:32,  1.29it/s]
4|1612|Loss: 0.7352089881896973:  25%|██▍       | 1611/6470 [18:31<1:02:32,  1.29it/s]
4|1612|Loss: 0.7352089881896973:  25%|██▍       | 1612/6470 [18:31<58:24,  1.39it/s]  
4|1613|Loss: 0.9154921174049377:  25%|██▍       | 1612/6470 [18:32<58:24,  1.39it/s]
4|1613|Loss: 0.9154921174049377:  25%|██▍       | 1613/6470 [18:32<56:09,  1.44it/s]
4|1614|Loss: 0.8524718880653381:  25%|██▍       | 1613/6470 [18:32<56:09,  1.44it/s]
4|1614|Loss: 0.8524718880653381:  25%|██▍       | 1614/6470 [18:33<59:32,  1.36it/s]
4|1615|Loss: 0.8220890760421753:  25%|██▍       | 1614/6470 [18:33<59:32,  1.36it/s]
4|1615|Loss: 0.8220890760421753:  25%|██▍       | 1615/6470 [18:33<56:05,  1.44it/s]
4|1616|Loss: 0.7415697574615479:  25%|██▍       | 1615/6470 [18:34<56:05,  1.44it/s]
4|1616|Loss: 0.7415697574615479:  25%|██▍       | 1616/6470 [18:34<55:51,  1.45it/s]
4|1617|Loss: 0.9379788041114807:  25%|██▍       | 1616/6470 [18:34<55:51,  1.45it/s]
4|1617|Loss: 0.9379788041114807:  25%|██▍       | 1617/6470 [18:35<56:12,  1.44it/s]
4|1618|Loss: 0.8498147130012512:  25%|██▍       | 1617/6470 [18:35<56:12,  1.44it/s]
4|1618|Loss: 0.8498147130012512:  25%|██▌       | 1618/6470 [18:35<57:07,  1.42it/s]
4|1619|Loss: 0.7742123603820801:  25%|██▌       | 1618/6470 [18:36<57:07,  1.42it/s]
4|1619|Loss: 0.7742123603820801:  25%|██▌       | 1619/6470 [18:36<54:55,  1.47it/s]
4|1620|Loss: 0.7650577425956726:  25%|██▌       | 1619/6470 [18:36<54:55,  1.47it/s]
4|1620|Loss: 0.7650577425956726:  25%|██▌       | 1620/6470 [18:37<55:36,  1.45it/s]
4|1621|Loss: 0.6908910274505615:  25%|██▌       | 1620/6470 [18:37<55:36,  1.45it/s]
4|1621|Loss: 0.6908910274505615:  25%|██▌       | 1621/6470 [18:37<56:22,  1.43it/s]
4|1622|Loss: 0.8687531352043152:  25%|██▌       | 1621/6470 [18:38<56:22,  1.43it/s]
4|1622|Loss: 0.8687531352043152:  25%|██▌       | 1622/6470 [18:38<55:21,  1.46it/s]
4|1623|Loss: 0.7861837148666382:  25%|██▌       | 1622/6470 [18:39<55:21,  1.46it/s]
4|1623|Loss: 0.7861837148666382:  25%|██▌       | 1623/6470 [18:39<58:32,  1.38it/s]
4|1624|Loss: 0.8619651198387146:  25%|██▌       | 1623/6470 [18:39<58:32,  1.38it/s]
4|1624|Loss: 0.8619651198387146:  25%|██▌       | 1624/6470 [18:40<1:01:21,  1.32it/s]
4|1625|Loss: 0.8940495848655701:  25%|██▌       | 1624/6470 [18:40<1:01:21,  1.32it/s]
4|1625|Loss: 0.8940495848655701:  25%|██▌       | 1625/6470 [18:40<58:57,  1.37it/s]  
4|1626|Loss: 0.9394274950027466:  25%|██▌       | 1625/6470 [18:41<58:57,  1.37it/s]
4|1626|Loss: 0.9394274950027466:  25%|██▌       | 1626/6470 [18:41<1:00:26,  1.34it/s]
4|1627|Loss: 0.8556287884712219:  25%|██▌       | 1626/6470 [18:42<1:00:26,  1.34it/s]
4|1627|Loss: 0.8556287884712219:  25%|██▌       | 1627/6470 [18:42<57:56,  1.39it/s]  
4|1628|Loss: 0.8655862212181091:  25%|██▌       | 1627/6470 [18:42<57:56,  1.39it/s]
4|1628|Loss: 0.8655862212181091:  25%|██▌       | 1628/6470 [18:42<55:17,  1.46it/s]
4|1629|Loss: 0.8154762983322144:  25%|██▌       | 1628/6470 [18:43<55:17,  1.46it/s]
4|1629|Loss: 0.8154762983322144:  25%|██▌       | 1629/6470 [18:43<55:54,  1.44it/s]
4|1630|Loss: 0.7867793440818787:  25%|██▌       | 1629/6470 [18:44<55:54,  1.44it/s]
4|1630|Loss: 0.7867793440818787:  25%|██▌       | 1630/6470 [18:44<56:55,  1.42it/s]
4|1631|Loss: 0.6473994851112366:  25%|██▌       | 1630/6470 [18:44<56:55,  1.42it/s]
4|1631|Loss: 0.6473994851112366:  25%|██▌       | 1631/6470 [18:45<55:40,  1.45it/s]
4|1632|Loss: 0.912738561630249:  25%|██▌       | 1631/6470 [18:45<55:40,  1.45it/s] 
4|1632|Loss: 0.912738561630249:  25%|██▌       | 1632/6470 [18:45<55:34,  1.45it/s]
4|1633|Loss: 1.0745974779129028:  25%|██▌       | 1632/6470 [18:46<55:34,  1.45it/s]
4|1633|Loss: 1.0745974779129028:  25%|██▌       | 1633/6470 [18:46<52:30,  1.54it/s]
4|1634|Loss: 0.9602437615394592:  25%|██▌       | 1633/6470 [18:46<52:30,  1.54it/s]
4|1634|Loss: 0.9602437615394592:  25%|██▌       | 1634/6470 [18:46<50:45,  1.59it/s]
4|1635|Loss: 0.9546098113059998:  25%|██▌       | 1634/6470 [18:47<50:45,  1.59it/s]
4|1635|Loss: 0.9546098113059998:  25%|██▌       | 1635/6470 [18:47<54:20,  1.48it/s]
4|1636|Loss: 0.8883384466171265:  25%|██▌       | 1635/6470 [18:48<54:20,  1.48it/s]
4|1636|Loss: 0.8883384466171265:  25%|██▌       | 1636/6470 [18:48<55:37,  1.45it/s]
4|1637|Loss: 0.8068671822547913:  25%|██▌       | 1636/6470 [18:48<55:37,  1.45it/s]
4|1637|Loss: 0.8068671822547913:  25%|██▌       | 1637/6470 [18:49<57:01,  1.41it/s]
4|1638|Loss: 0.8688090443611145:  25%|██▌       | 1637/6470 [18:49<57:01,  1.41it/s]
4|1638|Loss: 0.8688090443611145:  25%|██▌       | 1638/6470 [18:49<59:30,  1.35it/s]
4|1639|Loss: 0.8881090879440308:  25%|██▌       | 1638/6470 [18:50<59:30,  1.35it/s]
4|1639|Loss: 0.8881090879440308:  25%|██▌       | 1639/6470 [18:50<1:04:15,  1.25it/s]
4|1640|Loss: 0.7180601358413696:  25%|██▌       | 1639/6470 [18:51<1:04:15,  1.25it/s]
4|1640|Loss: 0.7180601358413696:  25%|██▌       | 1640/6470 [18:51<59:58,  1.34it/s]  
4|1641|Loss: 0.8436205387115479:  25%|██▌       | 1640/6470 [18:51<59:58,  1.34it/s]
4|1641|Loss: 0.8436205387115479:  25%|██▌       | 1641/6470 [18:52<58:58,  1.36it/s]
4|1642|Loss: 1.0918992757797241:  25%|██▌       | 1641/6470 [18:52<58:58,  1.36it/s]
4|1642|Loss: 1.0918992757797241:  25%|██▌       | 1642/6470 [18:53<1:00:56,  1.32it/s]
4|1643|Loss: 0.8009170889854431:  25%|██▌       | 1642/6470 [18:53<1:00:56,  1.32it/s]
4|1643|Loss: 0.8009170889854431:  25%|██▌       | 1643/6470 [18:53<55:20,  1.45it/s]  
4|1644|Loss: 0.8671074509620667:  25%|██▌       | 1643/6470 [18:53<55:20,  1.45it/s]
4|1644|Loss: 0.8671074509620667:  25%|██▌       | 1644/6470 [18:54<55:02,  1.46it/s]
4|1645|Loss: 0.9014434814453125:  25%|██▌       | 1644/6470 [18:54<55:02,  1.46it/s]
4|1645|Loss: 0.9014434814453125:  25%|██▌       | 1645/6470 [18:55<56:40,  1.42it/s]
4|1646|Loss: 0.7395609021186829:  25%|██▌       | 1645/6470 [18:55<56:40,  1.42it/s]
4|1646|Loss: 0.7395609021186829:  25%|██▌       | 1646/6470 [18:55<53:11,  1.51it/s]
4|1647|Loss: 0.8284623622894287:  25%|██▌       | 1646/6470 [18:55<53:11,  1.51it/s]
4|1647|Loss: 0.8284623622894287:  25%|██▌       | 1647/6470 [18:56<49:55,  1.61it/s]
4|1648|Loss: 0.897815465927124:  25%|██▌       | 1647/6470 [18:56<49:55,  1.61it/s] 
4|1648|Loss: 0.897815465927124:  25%|██▌       | 1648/6470 [18:56<49:21,  1.63it/s]
4|1649|Loss: 0.8242469429969788:  25%|██▌       | 1648/6470 [18:57<49:21,  1.63it/s]
4|1649|Loss: 0.8242469429969788:  25%|██▌       | 1649/6470 [18:57<49:49,  1.61it/s]
4|1650|Loss: 0.7036905884742737:  25%|██▌       | 1649/6470 [18:57<49:49,  1.61it/s]
4|1650|Loss: 0.7036905884742737:  26%|██▌       | 1650/6470 [18:57<49:16,  1.63it/s]
4|1651|Loss: 0.923531711101532:  26%|██▌       | 1650/6470 [18:58<49:16,  1.63it/s] 
4|1651|Loss: 0.923531711101532:  26%|██▌       | 1651/6470 [18:58<48:27,  1.66it/s]
4|1652|Loss: 0.7503345608711243:  26%|██▌       | 1651/6470 [18:58<48:27,  1.66it/s]
4|1652|Loss: 0.7503345608711243:  26%|██▌       | 1652/6470 [18:59<49:07,  1.63it/s]
4|1653|Loss: 0.8646734356880188:  26%|██▌       | 1652/6470 [18:59<49:07,  1.63it/s]
4|1653|Loss: 0.8646734356880188:  26%|██▌       | 1653/6470 [18:59<53:41,  1.50it/s]
4|1654|Loss: 0.8260945081710815:  26%|██▌       | 1653/6470 [19:00<53:41,  1.50it/s]
4|1654|Loss: 0.8260945081710815:  26%|██▌       | 1654/6470 [19:00<56:55,  1.41it/s]
4|1655|Loss: 0.8570563793182373:  26%|██▌       | 1654/6470 [19:01<56:55,  1.41it/s]
4|1655|Loss: 0.8570563793182373:  26%|██▌       | 1655/6470 [19:01<55:38,  1.44it/s]
4|1656|Loss: 0.6905332803726196:  26%|██▌       | 1655/6470 [19:01<55:38,  1.44it/s]
4|1656|Loss: 0.6905332803726196:  26%|██▌       | 1656/6470 [19:02<58:34,  1.37it/s]
4|1657|Loss: 0.9228308200836182:  26%|██▌       | 1656/6470 [19:02<58:34,  1.37it/s]
4|1657|Loss: 0.9228308200836182:  26%|██▌       | 1657/6470 [19:02<53:58,  1.49it/s]
4|1658|Loss: 0.8304902911186218:  26%|██▌       | 1657/6470 [19:03<53:58,  1.49it/s]
4|1658|Loss: 0.8304902911186218:  26%|██▌       | 1658/6470 [19:03<51:55,  1.54it/s]
4|1659|Loss: 0.894251823425293:  26%|██▌       | 1658/6470 [19:03<51:55,  1.54it/s] 
4|1659|Loss: 0.894251823425293:  26%|██▌       | 1659/6470 [19:03<49:31,  1.62it/s]
4|1660|Loss: 0.817552387714386:  26%|██▌       | 1659/6470 [19:04<49:31,  1.62it/s]
4|1660|Loss: 0.817552387714386:  26%|██▌       | 1660/6470 [19:04<49:07,  1.63it/s]
4|1661|Loss: 0.9877128601074219:  26%|██▌       | 1660/6470 [19:04<49:07,  1.63it/s]
4|1661|Loss: 0.9877128601074219:  26%|██▌       | 1661/6470 [19:04<46:10,  1.74it/s]
4|1662|Loss: 0.868121325969696:  26%|██▌       | 1661/6470 [19:05<46:10,  1.74it/s] 
4|1662|Loss: 0.868121325969696:  26%|██▌       | 1662/6470 [19:05<48:44,  1.64it/s]
4|1663|Loss: 0.9540746212005615:  26%|██▌       | 1662/6470 [19:06<48:44,  1.64it/s]
4|1663|Loss: 0.9540746212005615:  26%|██▌       | 1663/6470 [19:06<47:43,  1.68it/s]
4|1664|Loss: 0.7014524340629578:  26%|██▌       | 1663/6470 [19:06<47:43,  1.68it/s]
4|1664|Loss: 0.7014524340629578:  26%|██▌       | 1664/6470 [19:06<51:08,  1.57it/s]
4|1665|Loss: 0.6669350862503052:  26%|██▌       | 1664/6470 [19:07<51:08,  1.57it/s]
4|1665|Loss: 0.6669350862503052:  26%|██▌       | 1665/6470 [19:07<55:57,  1.43it/s]
4|1666|Loss: 0.8367882966995239:  26%|██▌       | 1665/6470 [19:08<55:57,  1.43it/s]
4|1666|Loss: 0.8367882966995239:  26%|██▌       | 1666/6470 [19:08<57:08,  1.40it/s]
4|1667|Loss: 0.8472088575363159:  26%|██▌       | 1666/6470 [19:09<57:08,  1.40it/s]
4|1667|Loss: 0.8472088575363159:  26%|██▌       | 1667/6470 [19:09<1:02:46,  1.28it/s]
4|1668|Loss: 0.9505273699760437:  26%|██▌       | 1667/6470 [19:10<1:02:46,  1.28it/s]
4|1668|Loss: 0.9505273699760437:  26%|██▌       | 1668/6470 [19:10<1:05:04,  1.23it/s]
4|1669|Loss: 0.8232879042625427:  26%|██▌       | 1668/6470 [19:10<1:05:04,  1.23it/s]
4|1669|Loss: 0.8232879042625427:  26%|██▌       | 1669/6470 [19:11<1:08:29,  1.17it/s]
4|1670|Loss: 0.9955267310142517:  26%|██▌       | 1669/6470 [19:11<1:08:29,  1.17it/s]
4|1670|Loss: 0.9955267310142517:  26%|██▌       | 1670/6470 [19:11<1:02:15,  1.28it/s]
4|1671|Loss: 0.9047880172729492:  26%|██▌       | 1670/6470 [19:12<1:02:15,  1.28it/s]
4|1671|Loss: 0.9047880172729492:  26%|██▌       | 1671/6470 [19:12<1:02:13,  1.29it/s]
4|1672|Loss: 0.805004358291626:  26%|██▌       | 1671/6470 [19:13<1:02:13,  1.29it/s] 
4|1672|Loss: 0.805004358291626:  26%|██▌       | 1672/6470 [19:13<59:47,  1.34it/s]  
4|1673|Loss: 0.840079665184021:  26%|██▌       | 1672/6470 [19:13<59:47,  1.34it/s]
4|1673|Loss: 0.840079665184021:  26%|██▌       | 1673/6470 [19:14<58:55,  1.36it/s]
4|1674|Loss: 0.7618969678878784:  26%|██▌       | 1673/6470 [19:14<58:55,  1.36it/s]
4|1674|Loss: 0.7618969678878784:  26%|██▌       | 1674/6470 [19:14<56:42,  1.41it/s]
4|1675|Loss: 0.9193207025527954:  26%|██▌       | 1674/6470 [19:15<56:42,  1.41it/s]
4|1675|Loss: 0.9193207025527954:  26%|██▌       | 1675/6470 [19:15<54:57,  1.45it/s]
4|1676|Loss: 0.9388230443000793:  26%|██▌       | 1675/6470 [19:15<54:57,  1.45it/s]
4|1676|Loss: 0.9388230443000793:  26%|██▌       | 1676/6470 [19:16<57:14,  1.40it/s]
4|1677|Loss: 0.7637675404548645:  26%|██▌       | 1676/6470 [19:16<57:14,  1.40it/s]
4|1677|Loss: 0.7637675404548645:  26%|██▌       | 1677/6470 [19:16<57:27,  1.39it/s]
4|1678|Loss: 0.8975827693939209:  26%|██▌       | 1677/6470 [19:17<57:27,  1.39it/s]
4|1678|Loss: 0.8975827693939209:  26%|██▌       | 1678/6470 [19:17<55:46,  1.43it/s]
4|1679|Loss: 0.7741879820823669:  26%|██▌       | 1678/6470 [19:17<55:46,  1.43it/s]
4|1679|Loss: 0.7741879820823669:  26%|██▌       | 1679/6470 [19:18<52:58,  1.51it/s]
4|1680|Loss: 0.8658508062362671:  26%|██▌       | 1679/6470 [19:18<52:58,  1.51it/s]
4|1680|Loss: 0.8658508062362671:  26%|██▌       | 1680/6470 [19:18<56:30,  1.41it/s]
4|1681|Loss: 0.7761675715446472:  26%|██▌       | 1680/6470 [19:19<56:30,  1.41it/s]
4|1681|Loss: 0.7761675715446472:  26%|██▌       | 1681/6470 [19:19<56:11,  1.42it/s]
4|1682|Loss: 0.8127060532569885:  26%|██▌       | 1681/6470 [19:20<56:11,  1.42it/s]
4|1682|Loss: 0.8127060532569885:  26%|██▌       | 1682/6470 [19:20<56:24,  1.41it/s]
4|1683|Loss: 0.8384395837783813:  26%|██▌       | 1682/6470 [19:20<56:24,  1.41it/s]
4|1683|Loss: 0.8384395837783813:  26%|██▌       | 1683/6470 [19:21<56:55,  1.40it/s]
4|1684|Loss: 0.7851855158805847:  26%|██▌       | 1683/6470 [19:21<56:55,  1.40it/s]
4|1684|Loss: 0.7851855158805847:  26%|██▌       | 1684/6470 [19:21<56:48,  1.40it/s]
4|1685|Loss: 0.7803656458854675:  26%|██▌       | 1684/6470 [19:22<56:48,  1.40it/s]
4|1685|Loss: 0.7803656458854675:  26%|██▌       | 1685/6470 [19:22<55:25,  1.44it/s]
4|1686|Loss: 0.8141944408416748:  26%|██▌       | 1685/6470 [19:22<55:25,  1.44it/s]
4|1686|Loss: 0.8141944408416748:  26%|██▌       | 1686/6470 [19:23<56:02,  1.42it/s]
4|1687|Loss: 0.7426332235336304:  26%|██▌       | 1686/6470 [19:23<56:02,  1.42it/s]
4|1687|Loss: 0.7426332235336304:  26%|██▌       | 1687/6470 [19:23<58:19,  1.37it/s]
4|1688|Loss: 0.7985743284225464:  26%|██▌       | 1687/6470 [19:24<58:19,  1.37it/s]
4|1688|Loss: 0.7985743284225464:  26%|██▌       | 1688/6470 [19:24<56:00,  1.42it/s]
4|1689|Loss: 0.9038804173469543:  26%|██▌       | 1688/6470 [19:25<56:00,  1.42it/s]
4|1689|Loss: 0.9038804173469543:  26%|██▌       | 1689/6470 [19:25<57:39,  1.38it/s]
4|1690|Loss: 0.857738733291626:  26%|██▌       | 1689/6470 [19:25<57:39,  1.38it/s] 
4|1690|Loss: 0.857738733291626:  26%|██▌       | 1690/6470 [19:26<57:41,  1.38it/s]
4|1691|Loss: 0.8287769556045532:  26%|██▌       | 1690/6470 [19:26<57:41,  1.38it/s]
4|1691|Loss: 0.8287769556045532:  26%|██▌       | 1691/6470 [19:26<55:59,  1.42it/s]
4|1692|Loss: 0.6548422574996948:  26%|██▌       | 1691/6470 [19:27<55:59,  1.42it/s]
4|1692|Loss: 0.6548422574996948:  26%|██▌       | 1692/6470 [19:27<55:40,  1.43it/s]
4|1693|Loss: 0.8475661873817444:  26%|██▌       | 1692/6470 [19:27<55:40,  1.43it/s]
4|1693|Loss: 0.8475661873817444:  26%|██▌       | 1693/6470 [19:28<55:34,  1.43it/s]
4|1694|Loss: 0.841600775718689:  26%|██▌       | 1693/6470 [19:28<55:34,  1.43it/s] 
4|1694|Loss: 0.841600775718689:  26%|██▌       | 1694/6470 [19:28<53:14,  1.50it/s]
4|1695|Loss: 0.8288318514823914:  26%|██▌       | 1694/6470 [19:29<53:14,  1.50it/s]
4|1695|Loss: 0.8288318514823914:  26%|██▌       | 1695/6470 [19:29<52:21,  1.52it/s]
4|1696|Loss: 0.7905077338218689:  26%|██▌       | 1695/6470 [19:29<52:21,  1.52it/s]
4|1696|Loss: 0.7905077338218689:  26%|██▌       | 1696/6470 [19:30<53:38,  1.48it/s]
4|1697|Loss: 1.044878363609314:  26%|██▌       | 1696/6470 [19:30<53:38,  1.48it/s] 
4|1697|Loss: 1.044878363609314:  26%|██▌       | 1697/6470 [19:30<50:42,  1.57it/s]
4|1698|Loss: 0.928451657295227:  26%|██▌       | 1697/6470 [19:31<50:42,  1.57it/s]
4|1698|Loss: 0.928451657295227:  26%|██▌       | 1698/6470 [19:31<51:17,  1.55it/s]
4|1699|Loss: 0.7297936677932739:  26%|██▌       | 1698/6470 [19:31<51:17,  1.55it/s]
4|1699|Loss: 0.7297936677932739:  26%|██▋       | 1699/6470 [19:31<51:31,  1.54it/s]
4|1700|Loss: 0.9791350960731506:  26%|██▋       | 1699/6470 [19:32<51:31,  1.54it/s]
4|1700|Loss: 0.9791350960731506:  26%|██▋       | 1700/6470 [19:32<55:40,  1.43it/s]
4|1701|Loss: 0.8384065628051758:  26%|██▋       | 1700/6470 [19:33<55:40,  1.43it/s]
4|1701|Loss: 0.8384065628051758:  26%|██▋       | 1701/6470 [19:33<59:52,  1.33it/s]
4|1702|Loss: 0.836021900177002:  26%|██▋       | 1701/6470 [19:34<59:52,  1.33it/s] 
4|1702|Loss: 0.836021900177002:  26%|██▋       | 1702/6470 [19:34<1:00:44,  1.31it/s]
4|1703|Loss: 0.7965652346611023:  26%|██▋       | 1702/6470 [19:34<1:00:44,  1.31it/s]
4|1703|Loss: 0.7965652346611023:  26%|██▋       | 1703/6470 [19:35<59:30,  1.33it/s]  
4|1704|Loss: 1.0170835256576538:  26%|██▋       | 1703/6470 [19:35<59:30,  1.33it/s]
4|1704|Loss: 1.0170835256576538:  26%|██▋       | 1704/6470 [19:35<58:58,  1.35it/s]
4|1705|Loss: 0.8406426906585693:  26%|██▋       | 1704/6470 [19:36<58:58,  1.35it/s]
4|1705|Loss: 0.8406426906585693:  26%|██▋       | 1705/6470 [19:36<1:00:02,  1.32it/s]
4|1706|Loss: 0.9412937760353088:  26%|██▋       | 1705/6470 [19:37<1:00:02,  1.32it/s]
4|1706|Loss: 0.9412937760353088:  26%|██▋       | 1706/6470 [19:37<1:00:52,  1.30it/s]
4|1707|Loss: 0.9397771954536438:  26%|██▋       | 1706/6470 [19:37<1:00:52,  1.30it/s]
4|1707|Loss: 0.9397771954536438:  26%|██▋       | 1707/6470 [19:38<1:02:12,  1.28it/s]
4|1708|Loss: 0.7805087566375732:  26%|██▋       | 1707/6470 [19:38<1:02:12,  1.28it/s]
4|1708|Loss: 0.7805087566375732:  26%|██▋       | 1708/6470 [19:39<1:01:56,  1.28it/s]
4|1709|Loss: 0.8825286626815796:  26%|██▋       | 1708/6470 [19:39<1:01:56,  1.28it/s]
4|1709|Loss: 0.8825286626815796:  26%|██▋       | 1709/6470 [19:39<1:00:55,  1.30it/s]
4|1710|Loss: 0.810508668422699:  26%|██▋       | 1709/6470 [19:40<1:00:55,  1.30it/s] 
4|1710|Loss: 0.810508668422699:  26%|██▋       | 1710/6470 [19:40<55:46,  1.42it/s]  
4|1711|Loss: 0.8784802556037903:  26%|██▋       | 1710/6470 [19:40<55:46,  1.42it/s]
4|1711|Loss: 0.8784802556037903:  26%|██▋       | 1711/6470 [19:40<53:46,  1.47it/s]
4|1712|Loss: 0.7512598037719727:  26%|██▋       | 1711/6470 [19:41<53:46,  1.47it/s]
4|1712|Loss: 0.7512598037719727:  26%|██▋       | 1712/6470 [19:41<54:04,  1.47it/s]
4|1713|Loss: 0.9210033416748047:  26%|██▋       | 1712/6470 [19:42<54:04,  1.47it/s]
4|1713|Loss: 0.9210033416748047:  26%|██▋       | 1713/6470 [19:42<52:52,  1.50it/s]
4|1714|Loss: 0.7742351293563843:  26%|██▋       | 1713/6470 [19:42<52:52,  1.50it/s]
4|1714|Loss: 0.7742351293563843:  26%|██▋       | 1714/6470 [19:42<50:55,  1.56it/s]
4|1715|Loss: 0.6838952302932739:  26%|██▋       | 1714/6470 [19:43<50:55,  1.56it/s]
4|1715|Loss: 0.6838952302932739:  27%|██▋       | 1715/6470 [19:43<48:49,  1.62it/s]
4|1716|Loss: 0.8510857224464417:  27%|██▋       | 1715/6470 [19:43<48:49,  1.62it/s]
4|1716|Loss: 0.8510857224464417:  27%|██▋       | 1716/6470 [19:44<48:40,  1.63it/s]
4|1717|Loss: 0.8957886099815369:  27%|██▋       | 1716/6470 [19:44<48:40,  1.63it/s]
4|1717|Loss: 0.8957886099815369:  27%|██▋       | 1717/6470 [19:44<49:12,  1.61it/s]
4|1718|Loss: 0.8651508688926697:  27%|██▋       | 1717/6470 [19:45<49:12,  1.61it/s]
4|1718|Loss: 0.8651508688926697:  27%|██▋       | 1718/6470 [19:45<54:09,  1.46it/s]
4|1719|Loss: 0.8189400434494019:  27%|██▋       | 1718/6470 [19:45<54:09,  1.46it/s]
4|1719|Loss: 0.8189400434494019:  27%|██▋       | 1719/6470 [19:46<53:17,  1.49it/s]
4|1720|Loss: 0.9690211415290833:  27%|██▋       | 1719/6470 [19:46<53:17,  1.49it/s]
4|1720|Loss: 0.9690211415290833:  27%|██▋       | 1720/6470 [19:46<54:20,  1.46it/s]
4|1721|Loss: 0.8281456828117371:  27%|██▋       | 1720/6470 [19:47<54:20,  1.46it/s]
4|1721|Loss: 0.8281456828117371:  27%|██▋       | 1721/6470 [19:47<56:02,  1.41it/s]
4|1722|Loss: 0.762343168258667:  27%|██▋       | 1721/6470 [19:48<56:02,  1.41it/s] 
4|1722|Loss: 0.762343168258667:  27%|██▋       | 1722/6470 [19:48<55:10,  1.43it/s]
4|1723|Loss: 0.9182395339012146:  27%|██▋       | 1722/6470 [19:48<55:10,  1.43it/s]
4|1723|Loss: 0.9182395339012146:  27%|██▋       | 1723/6470 [19:48<53:20,  1.48it/s]
4|1724|Loss: 0.8719038963317871:  27%|██▋       | 1723/6470 [19:49<53:20,  1.48it/s]
4|1724|Loss: 0.8719038963317871:  27%|██▋       | 1724/6470 [19:49<57:08,  1.38it/s]
4|1725|Loss: 0.8520486354827881:  27%|██▋       | 1724/6470 [19:50<57:08,  1.38it/s]
4|1725|Loss: 0.8520486354827881:  27%|██▋       | 1725/6470 [19:50<56:52,  1.39it/s]
4|1726|Loss: 0.776641845703125:  27%|██▋       | 1725/6470 [19:50<56:52,  1.39it/s] 
4|1726|Loss: 0.776641845703125:  27%|██▋       | 1726/6470 [19:51<54:25,  1.45it/s]
4|1727|Loss: 1.0266121625900269:  27%|██▋       | 1726/6470 [19:51<54:25,  1.45it/s]
4|1727|Loss: 1.0266121625900269:  27%|██▋       | 1727/6470 [19:51<52:04,  1.52it/s]
4|1728|Loss: 0.8446393013000488:  27%|██▋       | 1727/6470 [19:52<52:04,  1.52it/s]
4|1728|Loss: 0.8446393013000488:  27%|██▋       | 1728/6470 [19:52<55:04,  1.44it/s]
4|1729|Loss: 0.9571623206138611:  27%|██▋       | 1728/6470 [19:52<55:04,  1.44it/s]
4|1729|Loss: 0.9571623206138611:  27%|██▋       | 1729/6470 [19:53<56:28,  1.40it/s]
4|1730|Loss: 0.8472070693969727:  27%|██▋       | 1729/6470 [19:53<56:28,  1.40it/s]
4|1730|Loss: 0.8472070693969727:  27%|██▋       | 1730/6470 [19:53<57:25,  1.38it/s]
4|1731|Loss: 0.8336933851242065:  27%|██▋       | 1730/6470 [19:54<57:25,  1.38it/s]
4|1731|Loss: 0.8336933851242065:  27%|██▋       | 1731/6470 [19:54<59:01,  1.34it/s]
4|1732|Loss: 0.7552826404571533:  27%|██▋       | 1731/6470 [19:55<59:01,  1.34it/s]
4|1732|Loss: 0.7552826404571533:  27%|██▋       | 1732/6470 [19:55<59:21,  1.33it/s]
4|1733|Loss: 0.7421073913574219:  27%|██▋       | 1732/6470 [19:56<59:21,  1.33it/s]
4|1733|Loss: 0.7421073913574219:  27%|██▋       | 1733/6470 [19:56<1:01:56,  1.27it/s]
4|1734|Loss: 0.9081241488456726:  27%|██▋       | 1733/6470 [19:56<1:01:56,  1.27it/s]
4|1734|Loss: 0.9081241488456726:  27%|██▋       | 1734/6470 [19:57<59:55,  1.32it/s]  
4|1735|Loss: 0.8740701079368591:  27%|██▋       | 1734/6470 [19:57<59:55,  1.32it/s]
4|1735|Loss: 0.8740701079368591:  27%|██▋       | 1735/6470 [19:57<1:01:05,  1.29it/s]
4|1736|Loss: 0.8181964755058289:  27%|██▋       | 1735/6470 [19:58<1:01:05,  1.29it/s]
4|1736|Loss: 0.8181964755058289:  27%|██▋       | 1736/6470 [19:58<58:48,  1.34it/s]  
4|1737|Loss: 0.949666440486908:  27%|██▋       | 1736/6470 [19:58<58:48,  1.34it/s] 
4|1737|Loss: 0.949666440486908:  27%|██▋       | 1737/6470 [19:59<58:06,  1.36it/s]
4|1738|Loss: 0.8842636942863464:  27%|██▋       | 1737/6470 [19:59<58:06,  1.36it/s]
4|1738|Loss: 0.8842636942863464:  27%|██▋       | 1738/6470 [19:59<55:54,  1.41it/s]
4|1739|Loss: 0.8340721130371094:  27%|██▋       | 1738/6470 [20:00<55:54,  1.41it/s]
4|1739|Loss: 0.8340721130371094:  27%|██▋       | 1739/6470 [20:00<53:38,  1.47it/s]
4|1740|Loss: 0.896604597568512:  27%|██▋       | 1739/6470 [20:00<53:38,  1.47it/s] 
4|1740|Loss: 0.896604597568512:  27%|██▋       | 1740/6470 [20:01<54:53,  1.44it/s]
4|1741|Loss: 0.8411521911621094:  27%|██▋       | 1740/6470 [20:01<54:53,  1.44it/s]
4|1741|Loss: 0.8411521911621094:  27%|██▋       | 1741/6470 [20:02<58:15,  1.35it/s]
4|1742|Loss: 0.8669642210006714:  27%|██▋       | 1741/6470 [20:02<58:15,  1.35it/s]
4|1742|Loss: 0.8669642210006714:  27%|██▋       | 1742/6470 [20:02<56:41,  1.39it/s]
4|1743|Loss: 0.8882712125778198:  27%|██▋       | 1742/6470 [20:03<56:41,  1.39it/s]
4|1743|Loss: 0.8882712125778198:  27%|██▋       | 1743/6470 [20:03<56:39,  1.39it/s]
4|1744|Loss: 0.8475100994110107:  27%|██▋       | 1743/6470 [20:03<56:39,  1.39it/s]
4|1744|Loss: 0.8475100994110107:  27%|██▋       | 1744/6470 [20:04<55:19,  1.42it/s]
4|1745|Loss: 0.9222412705421448:  27%|██▋       | 1744/6470 [20:04<55:19,  1.42it/s]
4|1745|Loss: 0.9222412705421448:  27%|██▋       | 1745/6470 [20:04<53:47,  1.46it/s]
4|1746|Loss: 0.9232587218284607:  27%|██▋       | 1745/6470 [20:05<53:47,  1.46it/s]
4|1746|Loss: 0.9232587218284607:  27%|██▋       | 1746/6470 [20:05<54:35,  1.44it/s]
4|1747|Loss: 0.7983933091163635:  27%|██▋       | 1746/6470 [20:05<54:35,  1.44it/s]
4|1747|Loss: 0.7983933091163635:  27%|██▋       | 1747/6470 [20:06<55:30,  1.42it/s]
4|1748|Loss: 1.0069301128387451:  27%|██▋       | 1747/6470 [20:06<55:30,  1.42it/s]
4|1748|Loss: 1.0069301128387451:  27%|██▋       | 1748/6470 [20:07<56:24,  1.40it/s]
4|1749|Loss: 0.7573292851448059:  27%|██▋       | 1748/6470 [20:07<56:24,  1.40it/s]
4|1749|Loss: 0.7573292851448059:  27%|██▋       | 1749/6470 [20:07<56:12,  1.40it/s]
4|1750|Loss: 0.9169867038726807:  27%|██▋       | 1749/6470 [20:08<56:12,  1.40it/s]
4|1750|Loss: 0.9169867038726807:  27%|██▋       | 1750/6470 [20:08<55:46,  1.41it/s]
4|1751|Loss: 1.0275800228118896:  27%|██▋       | 1750/6470 [20:08<55:46,  1.41it/s]
4|1751|Loss: 1.0275800228118896:  27%|██▋       | 1751/6470 [20:09<53:02,  1.48it/s]
4|1752|Loss: 0.8741987943649292:  27%|██▋       | 1751/6470 [20:09<53:02,  1.48it/s]
4|1752|Loss: 0.8741987943649292:  27%|██▋       | 1752/6470 [20:09<53:01,  1.48it/s]
4|1753|Loss: 0.8764017820358276:  27%|██▋       | 1752/6470 [20:10<53:01,  1.48it/s]
4|1753|Loss: 0.8764017820358276:  27%|██▋       | 1753/6470 [20:10<52:01,  1.51it/s]
4|1754|Loss: 0.7868450880050659:  27%|██▋       | 1753/6470 [20:10<52:01,  1.51it/s]
4|1754|Loss: 0.7868450880050659:  27%|██▋       | 1754/6470 [20:10<50:59,  1.54it/s]
4|1755|Loss: 0.9273702502250671:  27%|██▋       | 1754/6470 [20:11<50:59,  1.54it/s]
4|1755|Loss: 0.9273702502250671:  27%|██▋       | 1755/6470 [20:11<50:16,  1.56it/s]
4|1756|Loss: 1.0677757263183594:  27%|██▋       | 1755/6470 [20:11<50:16,  1.56it/s]
4|1756|Loss: 1.0677757263183594:  27%|██▋       | 1756/6470 [20:12<50:07,  1.57it/s]
4|1757|Loss: 0.8319781422615051:  27%|██▋       | 1756/6470 [20:12<50:07,  1.57it/s]
4|1757|Loss: 0.8319781422615051:  27%|██▋       | 1757/6470 [20:12<50:13,  1.56it/s]
4|1758|Loss: 0.7888391017913818:  27%|██▋       | 1757/6470 [20:13<50:13,  1.56it/s]
4|1758|Loss: 0.7888391017913818:  27%|██▋       | 1758/6470 [20:13<49:19,  1.59it/s]
4|1759|Loss: 0.8427560329437256:  27%|██▋       | 1758/6470 [20:13<49:19,  1.59it/s]
4|1759|Loss: 0.8427560329437256:  27%|██▋       | 1759/6470 [20:14<51:13,  1.53it/s]
4|1760|Loss: 0.8437458276748657:  27%|██▋       | 1759/6470 [20:14<51:13,  1.53it/s]
4|1760|Loss: 0.8437458276748657:  27%|██▋       | 1760/6470 [20:14<53:45,  1.46it/s]
4|1761|Loss: 0.8660284876823425:  27%|██▋       | 1760/6470 [20:15<53:45,  1.46it/s]
4|1761|Loss: 0.8660284876823425:  27%|██▋       | 1761/6470 [20:15<50:31,  1.55it/s]
4|1762|Loss: 0.7630033493041992:  27%|██▋       | 1761/6470 [20:15<50:31,  1.55it/s]
4|1762|Loss: 0.7630033493041992:  27%|██▋       | 1762/6470 [20:16<54:03,  1.45it/s]
4|1763|Loss: 0.9783360362052917:  27%|██▋       | 1762/6470 [20:16<54:03,  1.45it/s]
4|1763|Loss: 0.9783360362052917:  27%|██▋       | 1763/6470 [20:17<55:25,  1.42it/s]
4|1764|Loss: 0.8797367215156555:  27%|██▋       | 1763/6470 [20:17<55:25,  1.42it/s]
4|1764|Loss: 0.8797367215156555:  27%|██▋       | 1764/6470 [20:17<51:48,  1.51it/s]
4|1765|Loss: 0.8435550332069397:  27%|██▋       | 1764/6470 [20:17<51:48,  1.51it/s]
4|1765|Loss: 0.8435550332069397:  27%|██▋       | 1765/6470 [20:18<53:26,  1.47it/s]
4|1766|Loss: 0.8544901013374329:  27%|██▋       | 1765/6470 [20:18<53:26,  1.47it/s]
4|1766|Loss: 0.8544901013374329:  27%|██▋       | 1766/6470 [20:19<54:48,  1.43it/s]
4|1767|Loss: 0.863072395324707:  27%|██▋       | 1766/6470 [20:19<54:48,  1.43it/s] 
4|1767|Loss: 0.863072395324707:  27%|██▋       | 1767/6470 [20:19<54:49,  1.43it/s]
4|1768|Loss: 0.895713210105896:  27%|██▋       | 1767/6470 [20:20<54:49,  1.43it/s]
4|1768|Loss: 0.895713210105896:  27%|██▋       | 1768/6470 [20:20<55:41,  1.41it/s]
4|1769|Loss: 0.8324809670448303:  27%|██▋       | 1768/6470 [20:20<55:41,  1.41it/s]
4|1769|Loss: 0.8324809670448303:  27%|██▋       | 1769/6470 [20:21<56:21,  1.39it/s]
4|1770|Loss: 0.8321967124938965:  27%|██▋       | 1769/6470 [20:21<56:21,  1.39it/s]
4|1770|Loss: 0.8321967124938965:  27%|██▋       | 1770/6470 [20:21<56:02,  1.40it/s]
4|1771|Loss: 0.8974344730377197:  27%|██▋       | 1770/6470 [20:22<56:02,  1.40it/s]
4|1771|Loss: 0.8974344730377197:  27%|██▋       | 1771/6470 [20:22<57:57,  1.35it/s]
4|1772|Loss: 0.931296169757843:  27%|██▋       | 1771/6470 [20:23<57:57,  1.35it/s] 
4|1772|Loss: 0.931296169757843:  27%|██▋       | 1772/6470 [20:23<56:00,  1.40it/s]
4|1773|Loss: 0.9341274499893188:  27%|██▋       | 1772/6470 [20:23<56:00,  1.40it/s]
4|1773|Loss: 0.9341274499893188:  27%|██▋       | 1773/6470 [20:24<56:08,  1.39it/s]
4|1774|Loss: 0.9168385863304138:  27%|██▋       | 1773/6470 [20:24<56:08,  1.39it/s]
4|1774|Loss: 0.9168385863304138:  27%|██▋       | 1774/6470 [20:24<57:00,  1.37it/s]
4|1775|Loss: 0.8855481743812561:  27%|██▋       | 1774/6470 [20:25<57:00,  1.37it/s]
4|1775|Loss: 0.8855481743812561:  27%|██▋       | 1775/6470 [20:25<57:25,  1.36it/s]
4|1776|Loss: 0.7970375418663025:  27%|██▋       | 1775/6470 [20:25<57:25,  1.36it/s]
4|1776|Loss: 0.7970375418663025:  27%|██▋       | 1776/6470 [20:26<55:17,  1.42it/s]
4|1777|Loss: 0.8171983361244202:  27%|██▋       | 1776/6470 [20:26<55:17,  1.42it/s]
4|1777|Loss: 0.8171983361244202:  27%|██▋       | 1777/6470 [20:27<57:36,  1.36it/s]
4|1778|Loss: 0.8354752659797668:  27%|██▋       | 1777/6470 [20:27<57:36,  1.36it/s]
4|1778|Loss: 0.8354752659797668:  27%|██▋       | 1778/6470 [20:27<58:16,  1.34it/s]
4|1779|Loss: 0.7833172082901001:  27%|██▋       | 1778/6470 [20:28<58:16,  1.34it/s]
4|1779|Loss: 0.7833172082901001:  27%|██▋       | 1779/6470 [20:28<52:47,  1.48it/s]
4|1780|Loss: 1.0354259014129639:  27%|██▋       | 1779/6470 [20:28<52:47,  1.48it/s]
4|1780|Loss: 1.0354259014129639:  28%|██▊       | 1780/6470 [20:28<50:35,  1.55it/s]
4|1781|Loss: 1.0100493431091309:  28%|██▊       | 1780/6470 [20:29<50:35,  1.55it/s]
4|1781|Loss: 1.0100493431091309:  28%|██▊       | 1781/6470 [20:29<52:08,  1.50it/s]
4|1782|Loss: 0.9254074692726135:  28%|██▊       | 1781/6470 [20:29<52:08,  1.50it/s]
4|1782|Loss: 0.9254074692726135:  28%|██▊       | 1782/6470 [20:30<51:29,  1.52it/s]
4|1783|Loss: 0.7807343006134033:  28%|██▊       | 1782/6470 [20:30<51:29,  1.52it/s]
4|1783|Loss: 0.7807343006134033:  28%|██▊       | 1783/6470 [20:30<50:51,  1.54it/s]
4|1784|Loss: 0.9060741662979126:  28%|██▊       | 1783/6470 [20:31<50:51,  1.54it/s]
4|1784|Loss: 0.9060741662979126:  28%|██▊       | 1784/6470 [20:31<53:32,  1.46it/s]
4|1785|Loss: 0.8615307211875916:  28%|██▊       | 1784/6470 [20:32<53:32,  1.46it/s]
4|1785|Loss: 0.8615307211875916:  28%|██▊       | 1785/6470 [20:32<54:20,  1.44it/s]
4|1786|Loss: 0.6911086440086365:  28%|██▊       | 1785/6470 [20:32<54:20,  1.44it/s]
4|1786|Loss: 0.6911086440086365:  28%|██▊       | 1786/6470 [20:32<52:00,  1.50it/s]
4|1787|Loss: 0.9494352340698242:  28%|██▊       | 1786/6470 [20:33<52:00,  1.50it/s]
4|1787|Loss: 0.9494352340698242:  28%|██▊       | 1787/6470 [20:33<54:43,  1.43it/s]
4|1788|Loss: 0.8100360035896301:  28%|██▊       | 1787/6470 [20:34<54:43,  1.43it/s]
4|1788|Loss: 0.8100360035896301:  28%|██▊       | 1788/6470 [20:34<58:07,  1.34it/s]
4|1789|Loss: 0.8993328809738159:  28%|██▊       | 1788/6470 [20:35<58:07,  1.34it/s]
4|1789|Loss: 0.8993328809738159:  28%|██▊       | 1789/6470 [20:35<58:37,  1.33it/s]
4|1790|Loss: 0.8159666657447815:  28%|██▊       | 1789/6470 [20:35<58:37,  1.33it/s]
4|1790|Loss: 0.8159666657447815:  28%|██▊       | 1790/6470 [20:36<59:27,  1.31it/s]
4|1791|Loss: 0.8223394751548767:  28%|██▊       | 1790/6470 [20:36<59:27,  1.31it/s]
4|1791|Loss: 0.8223394751548767:  28%|██▊       | 1791/6470 [20:36<55:13,  1.41it/s]
4|1792|Loss: 0.9051958322525024:  28%|██▊       | 1791/6470 [20:37<55:13,  1.41it/s]
4|1792|Loss: 0.9051958322525024:  28%|██▊       | 1792/6470 [20:37<52:02,  1.50it/s]
4|1793|Loss: 0.9064533710479736:  28%|██▊       | 1792/6470 [20:37<52:02,  1.50it/s]
4|1793|Loss: 0.9064533710479736:  28%|██▊       | 1793/6470 [20:38<52:47,  1.48it/s]
4|1794|Loss: 0.8269526362419128:  28%|██▊       | 1793/6470 [20:38<52:47,  1.48it/s]
4|1794|Loss: 0.8269526362419128:  28%|██▊       | 1794/6470 [20:38<49:50,  1.56it/s]
4|1795|Loss: 0.8822922706604004:  28%|██▊       | 1794/6470 [20:38<49:50,  1.56it/s]
4|1795|Loss: 0.8822922706604004:  28%|██▊       | 1795/6470 [20:39<47:03,  1.66it/s]
4|1796|Loss: 0.9441354274749756:  28%|██▊       | 1795/6470 [20:39<47:03,  1.66it/s]
4|1796|Loss: 0.9441354274749756:  28%|██▊       | 1796/6470 [20:39<48:28,  1.61it/s]
4|1797|Loss: 0.9332219362258911:  28%|██▊       | 1796/6470 [20:40<48:28,  1.61it/s]
4|1797|Loss: 0.9332219362258911:  28%|██▊       | 1797/6470 [20:40<50:28,  1.54it/s]
4|1798|Loss: 0.7928685545921326:  28%|██▊       | 1797/6470 [20:40<50:28,  1.54it/s]
4|1798|Loss: 0.7928685545921326:  28%|██▊       | 1798/6470 [20:40<47:08,  1.65it/s]
4|1799|Loss: 0.9254889488220215:  28%|██▊       | 1798/6470 [20:41<47:08,  1.65it/s]
4|1799|Loss: 0.9254889488220215:  28%|██▊       | 1799/6470 [20:41<48:44,  1.60it/s]
4|1800|Loss: 0.7521435618400574:  28%|██▊       | 1799/6470 [20:42<48:44,  1.60it/s]
4|1800|Loss: 0.7521435618400574:  28%|██▊       | 1800/6470 [20:42<49:42,  1.57it/s]
4|1801|Loss: 0.9388090372085571:  28%|██▊       | 1800/6470 [20:42<49:42,  1.57it/s]
4|1801|Loss: 0.9388090372085571:  28%|██▊       | 1801/6470 [20:43<51:55,  1.50it/s]
4|1802|Loss: 1.0436993837356567:  28%|██▊       | 1801/6470 [20:43<51:55,  1.50it/s]
4|1802|Loss: 1.0436993837356567:  28%|██▊       | 1802/6470 [20:43<51:57,  1.50it/s]
4|1803|Loss: 0.8157460689544678:  28%|██▊       | 1802/6470 [20:44<51:57,  1.50it/s]
4|1803|Loss: 0.8157460689544678:  28%|██▊       | 1803/6470 [20:44<54:56,  1.42it/s]
4|1804|Loss: 0.8731012940406799:  28%|██▊       | 1803/6470 [20:44<54:56,  1.42it/s]
4|1804|Loss: 0.8731012940406799:  28%|██▊       | 1804/6470 [20:45<50:58,  1.53it/s]
4|1805|Loss: 0.7706075310707092:  28%|██▊       | 1804/6470 [20:45<50:58,  1.53it/s]
4|1805|Loss: 0.7706075310707092:  28%|██▊       | 1805/6470 [20:45<46:14,  1.68it/s]
4|1806|Loss: 0.972233235836029:  28%|██▊       | 1805/6470 [20:45<46:14,  1.68it/s] 
4|1806|Loss: 0.972233235836029:  28%|██▊       | 1806/6470 [20:46<48:22,  1.61it/s]
4|1807|Loss: 0.7453673481941223:  28%|██▊       | 1806/6470 [20:46<48:22,  1.61it/s]
4|1807|Loss: 0.7453673481941223:  28%|██▊       | 1807/6470 [20:46<51:05,  1.52it/s]
4|1808|Loss: 0.8234537243843079:  28%|██▊       | 1807/6470 [20:47<51:05,  1.52it/s]
4|1808|Loss: 0.8234537243843079:  28%|██▊       | 1808/6470 [20:47<52:13,  1.49it/s]
4|1809|Loss: 0.8795611262321472:  28%|██▊       | 1808/6470 [20:48<52:13,  1.49it/s]
4|1809|Loss: 0.8795611262321472:  28%|██▊       | 1809/6470 [20:48<51:49,  1.50it/s]
4|1810|Loss: 0.8594475388526917:  28%|██▊       | 1809/6470 [20:48<51:49,  1.50it/s]
4|1810|Loss: 0.8594475388526917:  28%|██▊       | 1810/6470 [20:48<50:44,  1.53it/s]
4|1811|Loss: 0.9680242538452148:  28%|██▊       | 1810/6470 [20:49<50:44,  1.53it/s]
4|1811|Loss: 0.9680242538452148:  28%|██▊       | 1811/6470 [20:49<51:30,  1.51it/s]
4|1812|Loss: 0.7776432037353516:  28%|██▊       | 1811/6470 [20:50<51:30,  1.51it/s]
4|1812|Loss: 0.7776432037353516:  28%|██▊       | 1812/6470 [20:50<54:07,  1.43it/s]
4|1813|Loss: 0.9771491289138794:  28%|██▊       | 1812/6470 [20:50<54:07,  1.43it/s]
4|1813|Loss: 0.9771491289138794:  28%|██▊       | 1813/6470 [20:51<53:14,  1.46it/s]
4|1814|Loss: 0.7014493346214294:  28%|██▊       | 1813/6470 [20:51<53:14,  1.46it/s]
4|1814|Loss: 0.7014493346214294:  28%|██▊       | 1814/6470 [20:51<57:51,  1.34it/s]
4|1815|Loss: 0.821960985660553:  28%|██▊       | 1814/6470 [20:52<57:51,  1.34it/s] 
4|1815|Loss: 0.821960985660553:  28%|██▊       | 1815/6470 [20:52<56:32,  1.37it/s]
4|1816|Loss: 0.937110960483551:  28%|██▊       | 1815/6470 [20:52<56:32,  1.37it/s]
4|1816|Loss: 0.937110960483551:  28%|██▊       | 1816/6470 [20:53<53:55,  1.44it/s]
4|1817|Loss: 0.7593527436256409:  28%|██▊       | 1816/6470 [20:53<53:55,  1.44it/s]
4|1817|Loss: 0.7593527436256409:  28%|██▊       | 1817/6470 [20:54<56:07,  1.38it/s]
4|1818|Loss: 1.0639146566390991:  28%|██▊       | 1817/6470 [20:54<56:07,  1.38it/s]
4|1818|Loss: 1.0639146566390991:  28%|██▊       | 1818/6470 [20:54<58:39,  1.32it/s]
4|1819|Loss: 0.7422980666160583:  28%|██▊       | 1818/6470 [20:55<58:39,  1.32it/s]
4|1819|Loss: 0.7422980666160583:  28%|██▊       | 1819/6470 [20:55<58:36,  1.32it/s]
4|1820|Loss: 0.7893144488334656:  28%|██▊       | 1819/6470 [20:55<58:36,  1.32it/s]
4|1820|Loss: 0.7893144488334656:  28%|██▊       | 1820/6470 [20:56<55:12,  1.40it/s]
4|1821|Loss: 0.9274815320968628:  28%|██▊       | 1820/6470 [20:56<55:12,  1.40it/s]
4|1821|Loss: 0.9274815320968628:  28%|██▊       | 1821/6470 [20:56<52:51,  1.47it/s]
4|1822|Loss: 0.8158808350563049:  28%|██▊       | 1821/6470 [20:57<52:51,  1.47it/s]
4|1822|Loss: 0.8158808350563049:  28%|██▊       | 1822/6470 [20:57<52:52,  1.47it/s]
4|1823|Loss: 0.8193114399909973:  28%|██▊       | 1822/6470 [20:57<52:52,  1.47it/s]
4|1823|Loss: 0.8193114399909973:  28%|██▊       | 1823/6470 [20:58<53:23,  1.45it/s]
4|1824|Loss: 0.8162183165550232:  28%|██▊       | 1823/6470 [20:58<53:23,  1.45it/s]
4|1824|Loss: 0.8162183165550232:  28%|██▊       | 1824/6470 [20:58<54:16,  1.43it/s]
4|1825|Loss: 1.0389395952224731:  28%|██▊       | 1824/6470 [20:59<54:16,  1.43it/s]
4|1825|Loss: 1.0389395952224731:  28%|██▊       | 1825/6470 [20:59<54:22,  1.42it/s]
4|1826|Loss: 0.8066931366920471:  28%|██▊       | 1825/6470 [21:00<54:22,  1.42it/s]
4|1826|Loss: 0.8066931366920471:  28%|██▊       | 1826/6470 [21:00<54:27,  1.42it/s]
4|1827|Loss: 0.8802249431610107:  28%|██▊       | 1826/6470 [21:00<54:27,  1.42it/s]
4|1827|Loss: 0.8802249431610107:  28%|██▊       | 1827/6470 [21:00<53:11,  1.45it/s]
4|1828|Loss: 0.8627556562423706:  28%|██▊       | 1827/6470 [21:01<53:11,  1.45it/s]
4|1828|Loss: 0.8627556562423706:  28%|██▊       | 1828/6470 [21:01<50:22,  1.54it/s]
4|1829|Loss: 0.9031362533569336:  28%|██▊       | 1828/6470 [21:01<50:22,  1.54it/s]
4|1829|Loss: 0.9031362533569336:  28%|██▊       | 1829/6470 [21:02<50:58,  1.52it/s]
4|1830|Loss: 0.9964420199394226:  28%|██▊       | 1829/6470 [21:02<50:58,  1.52it/s]
4|1830|Loss: 0.9964420199394226:  28%|██▊       | 1830/6470 [21:02<52:13,  1.48it/s]
4|1831|Loss: 0.9974526166915894:  28%|██▊       | 1830/6470 [21:03<52:13,  1.48it/s]
4|1831|Loss: 0.9974526166915894:  28%|██▊       | 1831/6470 [21:03<52:56,  1.46it/s]
4|1832|Loss: 0.9568227529525757:  28%|██▊       | 1831/6470 [21:04<52:56,  1.46it/s]
4|1832|Loss: 0.9568227529525757:  28%|██▊       | 1832/6470 [21:04<53:20,  1.45it/s]
4|1833|Loss: 0.6811861991882324:  28%|██▊       | 1832/6470 [21:04<53:20,  1.45it/s]
4|1833|Loss: 0.6811861991882324:  28%|██▊       | 1833/6470 [21:05<55:30,  1.39it/s]
4|1834|Loss: 0.8280749917030334:  28%|██▊       | 1833/6470 [21:05<55:30,  1.39it/s]
4|1834|Loss: 0.8280749917030334:  28%|██▊       | 1834/6470 [21:05<53:17,  1.45it/s]
4|1835|Loss: 0.9298710823059082:  28%|██▊       | 1834/6470 [21:06<53:17,  1.45it/s]
4|1835|Loss: 0.9298710823059082:  28%|██▊       | 1835/6470 [21:06<53:45,  1.44it/s]
4|1836|Loss: 0.7435839772224426:  28%|██▊       | 1835/6470 [21:06<53:45,  1.44it/s]
4|1836|Loss: 0.7435839772224426:  28%|██▊       | 1836/6470 [21:07<55:43,  1.39it/s]
4|1837|Loss: 0.984586238861084:  28%|██▊       | 1836/6470 [21:07<55:43,  1.39it/s] 
4|1837|Loss: 0.984586238861084:  28%|██▊       | 1837/6470 [21:07<53:33,  1.44it/s]
4|1838|Loss: 0.7894555330276489:  28%|██▊       | 1837/6470 [21:08<53:33,  1.44it/s]
4|1838|Loss: 0.7894555330276489:  28%|██▊       | 1838/6470 [21:08<53:16,  1.45it/s]
4|1839|Loss: 0.6948344707489014:  28%|██▊       | 1838/6470 [21:08<53:16,  1.45it/s]
4|1839|Loss: 0.6948344707489014:  28%|██▊       | 1839/6470 [21:09<52:27,  1.47it/s]
4|1840|Loss: 0.8183412551879883:  28%|██▊       | 1839/6470 [21:09<52:27,  1.47it/s]
4|1840|Loss: 0.8183412551879883:  28%|██▊       | 1840/6470 [21:09<51:08,  1.51it/s]
4|1841|Loss: 0.7725339531898499:  28%|██▊       | 1840/6470 [21:10<51:08,  1.51it/s]
4|1841|Loss: 0.7725339531898499:  28%|██▊       | 1841/6470 [21:10<56:29,  1.37it/s]
4|1842|Loss: 0.9009012579917908:  28%|██▊       | 1841/6470 [21:11<56:29,  1.37it/s]
4|1842|Loss: 0.9009012579917908:  28%|██▊       | 1842/6470 [21:11<57:58,  1.33it/s]
4|1843|Loss: 0.9887906908988953:  28%|██▊       | 1842/6470 [21:11<57:58,  1.33it/s]
4|1843|Loss: 0.9887906908988953:  28%|██▊       | 1843/6470 [21:12<55:34,  1.39it/s]
4|1844|Loss: 0.8819817900657654:  28%|██▊       | 1843/6470 [21:12<55:34,  1.39it/s]
4|1844|Loss: 0.8819817900657654:  29%|██▊       | 1844/6470 [21:12<54:40,  1.41it/s]
4|1845|Loss: 0.7997370958328247:  29%|██▊       | 1844/6470 [21:13<54:40,  1.41it/s]
4|1845|Loss: 0.7997370958328247:  29%|██▊       | 1845/6470 [21:13<51:18,  1.50it/s]
4|1846|Loss: 0.9510247111320496:  29%|██▊       | 1845/6470 [21:13<51:18,  1.50it/s]
4|1846|Loss: 0.9510247111320496:  29%|██▊       | 1846/6470 [21:14<53:47,  1.43it/s]
4|1847|Loss: 0.9130887985229492:  29%|██▊       | 1846/6470 [21:14<53:47,  1.43it/s]
4|1847|Loss: 0.9130887985229492:  29%|██▊       | 1847/6470 [21:14<55:27,  1.39it/s]
4|1848|Loss: 0.8203567266464233:  29%|██▊       | 1847/6470 [21:15<55:27,  1.39it/s]
4|1848|Loss: 0.8203567266464233:  29%|██▊       | 1848/6470 [21:15<53:08,  1.45it/s]
4|1849|Loss: 0.8265296816825867:  29%|██▊       | 1848/6470 [21:15<53:08,  1.45it/s]
4|1849|Loss: 0.8265296816825867:  29%|██▊       | 1849/6470 [21:16<53:01,  1.45it/s]
4|1850|Loss: 0.8199575543403625:  29%|██▊       | 1849/6470 [21:16<53:01,  1.45it/s]
4|1850|Loss: 0.8199575543403625:  29%|██▊       | 1850/6470 [21:16<51:01,  1.51it/s]
4|1851|Loss: 1.019454002380371:  29%|██▊       | 1850/6470 [21:17<51:01,  1.51it/s] 
4|1851|Loss: 1.019454002380371:  29%|██▊       | 1851/6470 [21:17<52:07,  1.48it/s]
4|1852|Loss: 0.6706997156143188:  29%|██▊       | 1851/6470 [21:18<52:07,  1.48it/s]
4|1852|Loss: 0.6706997156143188:  29%|██▊       | 1852/6470 [21:18<53:31,  1.44it/s]
4|1853|Loss: 0.7167715430259705:  29%|██▊       | 1852/6470 [21:18<53:31,  1.44it/s]
4|1853|Loss: 0.7167715430259705:  29%|██▊       | 1853/6470 [21:18<52:44,  1.46it/s]
4|1854|Loss: 0.8275408744812012:  29%|██▊       | 1853/6470 [21:19<52:44,  1.46it/s]
4|1854|Loss: 0.8275408744812012:  29%|██▊       | 1854/6470 [21:19<49:45,  1.55it/s]
4|1855|Loss: 0.8142374753952026:  29%|██▊       | 1854/6470 [21:19<49:45,  1.55it/s]
4|1855|Loss: 0.8142374753952026:  29%|██▊       | 1855/6470 [21:20<46:16,  1.66it/s]
4|1856|Loss: 0.8699679970741272:  29%|██▊       | 1855/6470 [21:20<46:16,  1.66it/s]
4|1856|Loss: 0.8699679970741272:  29%|██▊       | 1856/6470 [21:20<45:53,  1.68it/s]
4|1857|Loss: 0.8812215328216553:  29%|██▊       | 1856/6470 [21:21<45:53,  1.68it/s]
4|1857|Loss: 0.8812215328216553:  29%|██▊       | 1857/6470 [21:21<47:52,  1.61it/s]
4|1858|Loss: 0.8314998745918274:  29%|██▊       | 1857/6470 [21:21<47:52,  1.61it/s]
4|1858|Loss: 0.8314998745918274:  29%|██▊       | 1858/6470 [21:22<49:36,  1.55it/s]
4|1859|Loss: 0.7345759272575378:  29%|██▊       | 1858/6470 [21:22<49:36,  1.55it/s]
4|1859|Loss: 0.7345759272575378:  29%|██▊       | 1859/6470 [21:22<51:01,  1.51it/s]
4|1860|Loss: 0.9159004092216492:  29%|██▊       | 1859/6470 [21:23<51:01,  1.51it/s]
4|1860|Loss: 0.9159004092216492:  29%|██▊       | 1860/6470 [21:23<51:57,  1.48it/s]
4|1861|Loss: 0.7110506892204285:  29%|██▊       | 1860/6470 [21:23<51:57,  1.48it/s]
4|1861|Loss: 0.7110506892204285:  29%|██▉       | 1861/6470 [21:24<53:07,  1.45it/s]
4|1862|Loss: 0.7877581715583801:  29%|██▉       | 1861/6470 [21:24<53:07,  1.45it/s]
4|1862|Loss: 0.7877581715583801:  29%|██▉       | 1862/6470 [21:24<52:23,  1.47it/s]
4|1863|Loss: 0.811417818069458:  29%|██▉       | 1862/6470 [21:25<52:23,  1.47it/s] 
4|1863|Loss: 0.811417818069458:  29%|██▉       | 1863/6470 [21:25<54:54,  1.40it/s]
4|1864|Loss: 0.7623588442802429:  29%|██▉       | 1863/6470 [21:26<54:54,  1.40it/s]
4|1864|Loss: 0.7623588442802429:  29%|██▉       | 1864/6470 [21:26<54:13,  1.42it/s]
4|1865|Loss: 0.8667361736297607:  29%|██▉       | 1864/6470 [21:26<54:13,  1.42it/s]
4|1865|Loss: 0.8667361736297607:  29%|██▉       | 1865/6470 [21:26<53:57,  1.42it/s]
4|1866|Loss: 0.7580469846725464:  29%|██▉       | 1865/6470 [21:27<53:57,  1.42it/s]
4|1866|Loss: 0.7580469846725464:  29%|██▉       | 1866/6470 [21:27<59:45,  1.28it/s]
4|1867|Loss: 0.777840256690979:  29%|██▉       | 1866/6470 [21:28<59:45,  1.28it/s] 
4|1867|Loss: 0.777840256690979:  29%|██▉       | 1867/6470 [21:28<58:20,  1.31it/s]
4|1868|Loss: 0.8095723390579224:  29%|██▉       | 1867/6470 [21:29<58:20,  1.31it/s]
4|1868|Loss: 0.8095723390579224:  29%|██▉       | 1868/6470 [21:29<57:21,  1.34it/s]
4|1869|Loss: 0.7219981551170349:  29%|██▉       | 1868/6470 [21:29<57:21,  1.34it/s]
4|1869|Loss: 0.7219981551170349:  29%|██▉       | 1869/6470 [21:29<52:57,  1.45it/s]
4|1870|Loss: 0.8782597780227661:  29%|██▉       | 1869/6470 [21:30<52:57,  1.45it/s]
4|1870|Loss: 0.8782597780227661:  29%|██▉       | 1870/6470 [21:30<53:09,  1.44it/s]
4|1871|Loss: 0.7882781028747559:  29%|██▉       | 1870/6470 [21:31<53:09,  1.44it/s]
4|1871|Loss: 0.7882781028747559:  29%|██▉       | 1871/6470 [21:31<50:55,  1.51it/s]
4|1872|Loss: 0.8866963982582092:  29%|██▉       | 1871/6470 [21:31<50:55,  1.51it/s]
4|1872|Loss: 0.8866963982582092:  29%|██▉       | 1872/6470 [21:31<51:54,  1.48it/s]
4|1873|Loss: 0.820008397102356:  29%|██▉       | 1872/6470 [21:32<51:54,  1.48it/s] 
4|1873|Loss: 0.820008397102356:  29%|██▉       | 1873/6470 [21:32<53:10,  1.44it/s]
4|1874|Loss: 0.766694188117981:  29%|██▉       | 1873/6470 [21:33<53:10,  1.44it/s]
4|1874|Loss: 0.766694188117981:  29%|██▉       | 1874/6470 [21:33<56:00,  1.37it/s]
4|1875|Loss: 0.7419295907020569:  29%|██▉       | 1874/6470 [21:33<56:00,  1.37it/s]
4|1875|Loss: 0.7419295907020569:  29%|██▉       | 1875/6470 [21:34<53:41,  1.43it/s]
4|1876|Loss: 0.7569512128829956:  29%|██▉       | 1875/6470 [21:34<53:41,  1.43it/s]
4|1876|Loss: 0.7569512128829956:  29%|██▉       | 1876/6470 [21:34<53:35,  1.43it/s]
4|1877|Loss: 0.9090360999107361:  29%|██▉       | 1876/6470 [21:35<53:35,  1.43it/s]
4|1877|Loss: 0.9090360999107361:  29%|██▉       | 1877/6470 [21:35<54:11,  1.41it/s]
4|1878|Loss: 0.8627462387084961:  29%|██▉       | 1877/6470 [21:35<54:11,  1.41it/s]
4|1878|Loss: 0.8627462387084961:  29%|██▉       | 1878/6470 [21:36<54:35,  1.40it/s]
4|1879|Loss: 0.8084076046943665:  29%|██▉       | 1878/6470 [21:36<54:35,  1.40it/s]
4|1879|Loss: 0.8084076046943665:  29%|██▉       | 1879/6470 [21:37<55:00,  1.39it/s]
4|1880|Loss: 0.8375855088233948:  29%|██▉       | 1879/6470 [21:37<55:00,  1.39it/s]
4|1880|Loss: 0.8375855088233948:  29%|██▉       | 1880/6470 [21:37<57:00,  1.34it/s]
4|1881|Loss: 0.834040105342865:  29%|██▉       | 1880/6470 [21:38<57:00,  1.34it/s] 
4|1881|Loss: 0.834040105342865:  29%|██▉       | 1881/6470 [21:38<53:52,  1.42it/s]
4|1882|Loss: 0.8370835781097412:  29%|██▉       | 1881/6470 [21:38<53:52,  1.42it/s]
4|1882|Loss: 0.8370835781097412:  29%|██▉       | 1882/6470 [21:38<50:01,  1.53it/s]
4|1883|Loss: 0.9775484800338745:  29%|██▉       | 1882/6470 [21:39<50:01,  1.53it/s]
4|1883|Loss: 0.9775484800338745:  29%|██▉       | 1883/6470 [21:39<50:20,  1.52it/s]
4|1884|Loss: 0.8292372822761536:  29%|██▉       | 1883/6470 [21:40<50:20,  1.52it/s]
4|1884|Loss: 0.8292372822761536:  29%|██▉       | 1884/6470 [21:40<51:53,  1.47it/s]
4|1885|Loss: 0.7131568789482117:  29%|██▉       | 1884/6470 [21:40<51:53,  1.47it/s]
4|1885|Loss: 0.7131568789482117:  29%|██▉       | 1885/6470 [21:41<53:06,  1.44it/s]
4|1886|Loss: 0.6607435941696167:  29%|██▉       | 1885/6470 [21:41<53:06,  1.44it/s]
4|1886|Loss: 0.6607435941696167:  29%|██▉       | 1886/6470 [21:41<50:57,  1.50it/s]
4|1887|Loss: 0.9124108552932739:  29%|██▉       | 1886/6470 [21:42<50:57,  1.50it/s]
4|1887|Loss: 0.9124108552932739:  29%|██▉       | 1887/6470 [21:42<53:00,  1.44it/s]
4|1888|Loss: 0.914461612701416:  29%|██▉       | 1887/6470 [21:42<53:00,  1.44it/s] 
4|1888|Loss: 0.914461612701416:  29%|██▉       | 1888/6470 [21:43<52:36,  1.45it/s]
4|1889|Loss: 0.9768285155296326:  29%|██▉       | 1888/6470 [21:43<52:36,  1.45it/s]
4|1889|Loss: 0.9768285155296326:  29%|██▉       | 1889/6470 [21:43<53:16,  1.43it/s]
4|1890|Loss: 0.7896490693092346:  29%|██▉       | 1889/6470 [21:44<53:16,  1.43it/s]
4|1890|Loss: 0.7896490693092346:  29%|██▉       | 1890/6470 [21:44<54:02,  1.41it/s]
4|1891|Loss: 0.7540168166160583:  29%|██▉       | 1890/6470 [21:44<54:02,  1.41it/s]
4|1891|Loss: 0.7540168166160583:  29%|██▉       | 1891/6470 [21:45<52:36,  1.45it/s]
4|1892|Loss: 0.9173733592033386:  29%|██▉       | 1891/6470 [21:45<52:36,  1.45it/s]
4|1892|Loss: 0.9173733592033386:  29%|██▉       | 1892/6470 [21:45<52:59,  1.44it/s]
4|1893|Loss: 0.9308333396911621:  29%|██▉       | 1892/6470 [21:46<52:59,  1.44it/s]
4|1893|Loss: 0.9308333396911621:  29%|██▉       | 1893/6470 [21:46<47:25,  1.61it/s]
4|1894|Loss: 0.7881566882133484:  29%|██▉       | 1893/6470 [21:46<47:25,  1.61it/s]
4|1894|Loss: 0.7881566882133484:  29%|██▉       | 1894/6470 [21:47<48:49,  1.56it/s]
4|1895|Loss: 0.8418033123016357:  29%|██▉       | 1894/6470 [21:47<48:49,  1.56it/s]
4|1895|Loss: 0.8418033123016357:  29%|██▉       | 1895/6470 [21:47<52:56,  1.44it/s]
4|1896|Loss: 0.8992854356765747:  29%|██▉       | 1895/6470 [21:48<52:56,  1.44it/s]
4|1896|Loss: 0.8992854356765747:  29%|██▉       | 1896/6470 [21:48<54:12,  1.41it/s]
4|1897|Loss: 0.8756310343742371:  29%|██▉       | 1896/6470 [21:49<54:12,  1.41it/s]
4|1897|Loss: 0.8756310343742371:  29%|██▉       | 1897/6470 [21:49<54:38,  1.39it/s]
4|1898|Loss: 1.1784862279891968:  29%|██▉       | 1897/6470 [21:49<54:38,  1.39it/s]
4|1898|Loss: 1.1784862279891968:  29%|██▉       | 1898/6470 [21:49<52:08,  1.46it/s]
4|1899|Loss: 0.8797754645347595:  29%|██▉       | 1898/6470 [21:50<52:08,  1.46it/s]
4|1899|Loss: 0.8797754645347595:  29%|██▉       | 1899/6470 [21:50<48:53,  1.56it/s]
4|1900|Loss: 0.8116114735603333:  29%|██▉       | 1899/6470 [21:50<48:53,  1.56it/s]
4|1900|Loss: 0.8116114735603333:  29%|██▉       | 1900/6470 [21:51<47:01,  1.62it/s]
4|1901|Loss: 0.7366209626197815:  29%|██▉       | 1900/6470 [21:51<47:01,  1.62it/s]
4|1901|Loss: 0.7366209626197815:  29%|██▉       | 1901/6470 [21:51<50:46,  1.50it/s]
4|1902|Loss: 0.8956875801086426:  29%|██▉       | 1901/6470 [21:52<50:46,  1.50it/s]
4|1902|Loss: 0.8956875801086426:  29%|██▉       | 1902/6470 [21:52<51:05,  1.49it/s]
4|1903|Loss: 0.8036653995513916:  29%|██▉       | 1902/6470 [21:52<51:05,  1.49it/s]
4|1903|Loss: 0.8036653995513916:  29%|██▉       | 1903/6470 [21:53<52:31,  1.45it/s]
4|1904|Loss: 1.0050934553146362:  29%|██▉       | 1903/6470 [21:53<52:31,  1.45it/s]
4|1904|Loss: 1.0050934553146362:  29%|██▉       | 1904/6470 [21:53<51:45,  1.47it/s]
4|1905|Loss: 0.8637972474098206:  29%|██▉       | 1904/6470 [21:54<51:45,  1.47it/s]
4|1905|Loss: 0.8637972474098206:  29%|██▉       | 1905/6470 [21:54<54:37,  1.39it/s]
4|1906|Loss: 0.7654964327812195:  29%|██▉       | 1905/6470 [21:55<54:37,  1.39it/s]
4|1906|Loss: 0.7654964327812195:  29%|██▉       | 1906/6470 [21:55<54:01,  1.41it/s]
4|1907|Loss: 0.7738914489746094:  29%|██▉       | 1906/6470 [21:55<54:01,  1.41it/s]
4|1907|Loss: 0.7738914489746094:  29%|██▉       | 1907/6470 [21:56<55:08,  1.38it/s]
4|1908|Loss: 0.7473039031028748:  29%|██▉       | 1907/6470 [21:56<55:08,  1.38it/s]
4|1908|Loss: 0.7473039031028748:  29%|██▉       | 1908/6470 [21:56<55:05,  1.38it/s]
4|1909|Loss: 0.9051706790924072:  29%|██▉       | 1908/6470 [21:57<55:05,  1.38it/s]
4|1909|Loss: 0.9051706790924072:  30%|██▉       | 1909/6470 [21:57<57:39,  1.32it/s]
4|1910|Loss: 1.048052191734314:  30%|██▉       | 1909/6470 [21:58<57:39,  1.32it/s] 
4|1910|Loss: 1.048052191734314:  30%|██▉       | 1910/6470 [21:58<57:53,  1.31it/s]
4|1911|Loss: 0.8471843004226685:  30%|██▉       | 1910/6470 [21:58<57:53,  1.31it/s]
4|1911|Loss: 0.8471843004226685:  30%|██▉       | 1911/6470 [21:59<58:16,  1.30it/s]
4|1912|Loss: 0.9827444553375244:  30%|██▉       | 1911/6470 [21:59<58:16,  1.30it/s]
4|1912|Loss: 0.9827444553375244:  30%|██▉       | 1912/6470 [22:00<59:32,  1.28it/s]
4|1913|Loss: 0.8032179474830627:  30%|██▉       | 1912/6470 [22:00<59:32,  1.28it/s]
4|1913|Loss: 0.8032179474830627:  30%|██▉       | 1913/6470 [22:00<55:05,  1.38it/s]
4|1914|Loss: 0.8483278155326843:  30%|██▉       | 1913/6470 [22:01<55:05,  1.38it/s]
4|1914|Loss: 0.8483278155326843:  30%|██▉       | 1914/6470 [22:01<51:17,  1.48it/s]
4|1915|Loss: 0.9054534435272217:  30%|██▉       | 1914/6470 [22:01<51:17,  1.48it/s]
4|1915|Loss: 0.9054534435272217:  30%|██▉       | 1915/6470 [22:01<47:19,  1.60it/s]
4|1916|Loss: 0.6242806911468506:  30%|██▉       | 1915/6470 [22:02<47:19,  1.60it/s]
4|1916|Loss: 0.6242806911468506:  30%|██▉       | 1916/6470 [22:02<45:42,  1.66it/s]
4|1917|Loss: 0.7935655117034912:  30%|██▉       | 1916/6470 [22:02<45:42,  1.66it/s]
4|1917|Loss: 0.7935655117034912:  30%|██▉       | 1917/6470 [22:02<45:03,  1.68it/s]
4|1918|Loss: 0.7651440501213074:  30%|██▉       | 1917/6470 [22:03<45:03,  1.68it/s]
4|1918|Loss: 0.7651440501213074:  30%|██▉       | 1918/6470 [22:03<49:48,  1.52it/s]
4|1919|Loss: 0.9772026538848877:  30%|██▉       | 1918/6470 [22:04<49:48,  1.52it/s]
4|1919|Loss: 0.9772026538848877:  30%|██▉       | 1919/6470 [22:04<48:18,  1.57it/s]
4|1920|Loss: 0.9262956976890564:  30%|██▉       | 1919/6470 [22:04<48:18,  1.57it/s]
4|1920|Loss: 0.9262956976890564:  30%|██▉       | 1920/6470 [22:04<47:19,  1.60it/s]
4|1921|Loss: 0.8662319779396057:  30%|██▉       | 1920/6470 [22:05<47:19,  1.60it/s]
4|1921|Loss: 0.8662319779396057:  30%|██▉       | 1921/6470 [22:05<49:40,  1.53it/s]
4|1922|Loss: 0.9016924500465393:  30%|██▉       | 1921/6470 [22:05<49:40,  1.53it/s]
4|1922|Loss: 0.9016924500465393:  30%|██▉       | 1922/6470 [22:06<49:30,  1.53it/s]
4|1923|Loss: 0.9555888772010803:  30%|██▉       | 1922/6470 [22:06<49:30,  1.53it/s]
4|1923|Loss: 0.9555888772010803:  30%|██▉       | 1923/6470 [22:06<49:14,  1.54it/s]
4|1924|Loss: 0.9524062275886536:  30%|██▉       | 1923/6470 [22:07<49:14,  1.54it/s]
4|1924|Loss: 0.9524062275886536:  30%|██▉       | 1924/6470 [22:07<49:08,  1.54it/s]
4|1925|Loss: 0.9121432900428772:  30%|██▉       | 1924/6470 [22:07<49:08,  1.54it/s]
4|1925|Loss: 0.9121432900428772:  30%|██▉       | 1925/6470 [22:08<45:32,  1.66it/s]
4|1926|Loss: 0.9593412280082703:  30%|██▉       | 1925/6470 [22:08<45:32,  1.66it/s]
4|1926|Loss: 0.9593412280082703:  30%|██▉       | 1926/6470 [22:08<45:04,  1.68it/s]
4|1927|Loss: 0.8577075004577637:  30%|██▉       | 1926/6470 [22:08<45:04,  1.68it/s]
4|1927|Loss: 0.8577075004577637:  30%|██▉       | 1927/6470 [22:09<45:57,  1.65it/s]
4|1928|Loss: 1.0454907417297363:  30%|██▉       | 1927/6470 [22:09<45:57,  1.65it/s]
4|1928|Loss: 1.0454907417297363:  30%|██▉       | 1928/6470 [22:10<50:13,  1.51it/s]
4|1929|Loss: 0.7646608352661133:  30%|██▉       | 1928/6470 [22:10<50:13,  1.51it/s]
4|1929|Loss: 0.7646608352661133:  30%|██▉       | 1929/6470 [22:10<50:14,  1.51it/s]
4|1930|Loss: 0.9637550711631775:  30%|██▉       | 1929/6470 [22:11<50:14,  1.51it/s]
4|1930|Loss: 0.9637550711631775:  30%|██▉       | 1930/6470 [22:11<54:25,  1.39it/s]
4|1931|Loss: 0.9223082065582275:  30%|██▉       | 1930/6470 [22:11<54:25,  1.39it/s]
4|1931|Loss: 0.9223082065582275:  30%|██▉       | 1931/6470 [22:12<51:38,  1.46it/s]
4|1932|Loss: 0.9436281323432922:  30%|██▉       | 1931/6470 [22:12<51:38,  1.46it/s]
4|1932|Loss: 0.9436281323432922:  30%|██▉       | 1932/6470 [22:12<51:55,  1.46it/s]
4|1933|Loss: 0.9066883325576782:  30%|██▉       | 1932/6470 [22:13<51:55,  1.46it/s]
4|1933|Loss: 0.9066883325576782:  30%|██▉       | 1933/6470 [22:13<53:04,  1.42it/s]
4|1934|Loss: 0.9476138949394226:  30%|██▉       | 1933/6470 [22:13<53:04,  1.42it/s]
4|1934|Loss: 0.9476138949394226:  30%|██▉       | 1934/6470 [22:14<51:12,  1.48it/s]
4|1935|Loss: 0.8479983806610107:  30%|██▉       | 1934/6470 [22:14<51:12,  1.48it/s]
4|1935|Loss: 0.8479983806610107:  30%|██▉       | 1935/6470 [22:14<51:43,  1.46it/s]
4|1936|Loss: 0.7955566048622131:  30%|██▉       | 1935/6470 [22:15<51:43,  1.46it/s]
4|1936|Loss: 0.7955566048622131:  30%|██▉       | 1936/6470 [22:15<49:24,  1.53it/s]
4|1937|Loss: 0.8195719122886658:  30%|██▉       | 1936/6470 [22:15<49:24,  1.53it/s]
4|1937|Loss: 0.8195719122886658:  30%|██▉       | 1937/6470 [22:16<49:49,  1.52it/s]
4|1938|Loss: 0.872355043888092:  30%|██▉       | 1937/6470 [22:16<49:49,  1.52it/s] 
4|1938|Loss: 0.872355043888092:  30%|██▉       | 1938/6470 [22:16<48:44,  1.55it/s]
4|1939|Loss: 1.0018088817596436:  30%|██▉       | 1938/6470 [22:17<48:44,  1.55it/s]
4|1939|Loss: 1.0018088817596436:  30%|██▉       | 1939/6470 [22:17<51:15,  1.47it/s]
4|1940|Loss: 0.6975380182266235:  30%|██▉       | 1939/6470 [22:18<51:15,  1.47it/s]
4|1940|Loss: 0.6975380182266235:  30%|██▉       | 1940/6470 [22:18<56:00,  1.35it/s]
4|1941|Loss: 0.7845549583435059:  30%|██▉       | 1940/6470 [22:18<56:00,  1.35it/s]
4|1941|Loss: 0.7845549583435059:  30%|███       | 1941/6470 [22:19<55:07,  1.37it/s]
4|1942|Loss: 0.980762779712677:  30%|███       | 1941/6470 [22:19<55:07,  1.37it/s] 
4|1942|Loss: 0.980762779712677:  30%|███       | 1942/6470 [22:19<52:01,  1.45it/s]
4|1943|Loss: 0.8910800218582153:  30%|███       | 1942/6470 [22:20<52:01,  1.45it/s]
4|1943|Loss: 0.8910800218582153:  30%|███       | 1943/6470 [22:20<54:46,  1.38it/s]
4|1944|Loss: 0.6055861115455627:  30%|███       | 1943/6470 [22:21<54:46,  1.38it/s]
4|1944|Loss: 0.6055861115455627:  30%|███       | 1944/6470 [22:21<58:25,  1.29it/s]
4|1945|Loss: 0.9477340579032898:  30%|███       | 1944/6470 [22:21<58:25,  1.29it/s]
4|1945|Loss: 0.9477340579032898:  30%|███       | 1945/6470 [22:22<56:48,  1.33it/s]
4|1946|Loss: 1.036691665649414:  30%|███       | 1945/6470 [22:22<56:48,  1.33it/s] 
4|1946|Loss: 1.036691665649414:  30%|███       | 1946/6470 [22:22<54:15,  1.39it/s]
4|1947|Loss: 0.7465227246284485:  30%|███       | 1946/6470 [22:23<54:15,  1.39it/s]
4|1947|Loss: 0.7465227246284485:  30%|███       | 1947/6470 [22:23<54:34,  1.38it/s]
4|1948|Loss: 0.7987275123596191:  30%|███       | 1947/6470 [22:23<54:34,  1.38it/s]
4|1948|Loss: 0.7987275123596191:  30%|███       | 1948/6470 [22:24<54:22,  1.39it/s]
4|1949|Loss: 0.8648666143417358:  30%|███       | 1948/6470 [22:24<54:22,  1.39it/s]
4|1949|Loss: 0.8648666143417358:  30%|███       | 1949/6470 [22:24<54:30,  1.38it/s]
4|1950|Loss: 0.8976735472679138:  30%|███       | 1949/6470 [22:25<54:30,  1.38it/s]
4|1950|Loss: 0.8976735472679138:  30%|███       | 1950/6470 [22:25<53:03,  1.42it/s]
4|1951|Loss: 0.8912503719329834:  30%|███       | 1950/6470 [22:25<53:03,  1.42it/s]
4|1951|Loss: 0.8912503719329834:  30%|███       | 1951/6470 [22:26<51:38,  1.46it/s]
4|1952|Loss: 0.8361139893531799:  30%|███       | 1951/6470 [22:26<51:38,  1.46it/s]
4|1952|Loss: 0.8361139893531799:  30%|███       | 1952/6470 [22:26<52:42,  1.43it/s]
4|1953|Loss: 0.947472870349884:  30%|███       | 1952/6470 [22:27<52:42,  1.43it/s] 
4|1953|Loss: 0.947472870349884:  30%|███       | 1953/6470 [22:27<55:41,  1.35it/s]
4|1954|Loss: 0.7973987460136414:  30%|███       | 1953/6470 [22:28<55:41,  1.35it/s]
4|1954|Loss: 0.7973987460136414:  30%|███       | 1954/6470 [22:28<56:28,  1.33it/s]
4|1955|Loss: 0.9203108549118042:  30%|███       | 1954/6470 [22:29<56:28,  1.33it/s]
4|1955|Loss: 0.9203108549118042:  30%|███       | 1955/6470 [22:29<57:07,  1.32it/s]
4|1956|Loss: 0.8142745494842529:  30%|███       | 1955/6470 [22:29<57:07,  1.32it/s]
4|1956|Loss: 0.8142745494842529:  30%|███       | 1956/6470 [22:30<55:11,  1.36it/s]
4|1957|Loss: 0.8253104090690613:  30%|███       | 1956/6470 [22:30<55:11,  1.36it/s]
4|1957|Loss: 0.8253104090690613:  30%|███       | 1957/6470 [22:30<54:20,  1.38it/s]
4|1958|Loss: 0.7822139263153076:  30%|███       | 1957/6470 [22:31<54:20,  1.38it/s]
4|1958|Loss: 0.7822139263153076:  30%|███       | 1958/6470 [22:31<55:47,  1.35it/s]
4|1959|Loss: 0.9790982604026794:  30%|███       | 1958/6470 [22:31<55:47,  1.35it/s]
4|1959|Loss: 0.9790982604026794:  30%|███       | 1959/6470 [22:32<55:25,  1.36it/s]
4|1960|Loss: 1.0388462543487549:  30%|███       | 1959/6470 [22:32<55:25,  1.36it/s]
4|1960|Loss: 1.0388462543487549:  30%|███       | 1960/6470 [22:32<55:17,  1.36it/s]
4|1961|Loss: 0.8388144969940186:  30%|███       | 1960/6470 [22:33<55:17,  1.36it/s]
4|1961|Loss: 0.8388144969940186:  30%|███       | 1961/6470 [22:33<53:20,  1.41it/s]
4|1962|Loss: 0.9970776438713074:  30%|███       | 1961/6470 [22:33<53:20,  1.41it/s]
4|1962|Loss: 0.9970776438713074:  30%|███       | 1962/6470 [22:34<51:42,  1.45it/s]
4|1963|Loss: 0.933558464050293:  30%|███       | 1962/6470 [22:34<51:42,  1.45it/s] 
4|1963|Loss: 0.933558464050293:  30%|███       | 1963/6470 [22:34<51:49,  1.45it/s]
4|1964|Loss: 0.8455994725227356:  30%|███       | 1963/6470 [22:35<51:49,  1.45it/s]
4|1964|Loss: 0.8455994725227356:  30%|███       | 1964/6470 [22:35<54:52,  1.37it/s]
4|1965|Loss: 0.8662053942680359:  30%|███       | 1964/6470 [22:36<54:52,  1.37it/s]
4|1965|Loss: 0.8662053942680359:  30%|███       | 1965/6470 [22:36<58:18,  1.29it/s]
4|1966|Loss: 0.8824430108070374:  30%|███       | 1965/6470 [22:37<58:18,  1.29it/s]
4|1966|Loss: 0.8824430108070374:  30%|███       | 1966/6470 [22:37<54:48,  1.37it/s]
4|1967|Loss: 0.8113633394241333:  30%|███       | 1966/6470 [22:37<54:48,  1.37it/s]
4|1967|Loss: 0.8113633394241333:  30%|███       | 1967/6470 [22:37<51:51,  1.45it/s]
4|1968|Loss: 0.8523104786872864:  30%|███       | 1967/6470 [22:38<51:51,  1.45it/s]
4|1968|Loss: 0.8523104786872864:  30%|███       | 1968/6470 [22:38<50:11,  1.50it/s]
4|1969|Loss: 0.7579963207244873:  30%|███       | 1968/6470 [22:38<50:11,  1.50it/s]
4|1969|Loss: 0.7579963207244873:  30%|███       | 1969/6470 [22:38<45:58,  1.63it/s]
4|1970|Loss: 0.8231564164161682:  30%|███       | 1969/6470 [22:39<45:58,  1.63it/s]
4|1970|Loss: 0.8231564164161682:  30%|███       | 1970/6470 [22:39<46:43,  1.61it/s]
4|1971|Loss: 0.8210597038269043:  30%|███       | 1970/6470 [22:40<46:43,  1.61it/s]
4|1971|Loss: 0.8210597038269043:  30%|███       | 1971/6470 [22:40<47:37,  1.57it/s]
4|1972|Loss: 0.9494770169258118:  30%|███       | 1971/6470 [22:40<47:37,  1.57it/s]
4|1972|Loss: 0.9494770169258118:  30%|███       | 1972/6470 [22:40<47:26,  1.58it/s]
4|1973|Loss: 0.8442137837409973:  30%|███       | 1972/6470 [22:41<47:26,  1.58it/s]
4|1973|Loss: 0.8442137837409973:  30%|███       | 1973/6470 [22:41<49:27,  1.52it/s]
4|1974|Loss: 1.0106127262115479:  30%|███       | 1973/6470 [22:42<49:27,  1.52it/s]
4|1974|Loss: 1.0106127262115479:  31%|███       | 1974/6470 [22:42<52:14,  1.43it/s]
4|1975|Loss: 0.955231249332428:  31%|███       | 1974/6470 [22:42<52:14,  1.43it/s] 
4|1975|Loss: 0.955231249332428:  31%|███       | 1975/6470 [22:43<51:47,  1.45it/s]
4|1976|Loss: 0.8480939865112305:  31%|███       | 1975/6470 [22:43<51:47,  1.45it/s]
4|1976|Loss: 0.8480939865112305:  31%|███       | 1976/6470 [22:43<49:31,  1.51it/s]
4|1977|Loss: 0.8905743956565857:  31%|███       | 1976/6470 [22:44<49:31,  1.51it/s]
4|1977|Loss: 0.8905743956565857:  31%|███       | 1977/6470 [22:44<48:23,  1.55it/s]
4|1978|Loss: 0.8079555630683899:  31%|███       | 1977/6470 [22:44<48:23,  1.55it/s]
4|1978|Loss: 0.8079555630683899:  31%|███       | 1978/6470 [22:45<51:58,  1.44it/s]
4|1979|Loss: 0.9089406728744507:  31%|███       | 1978/6470 [22:45<51:58,  1.44it/s]
4|1979|Loss: 0.9089406728744507:  31%|███       | 1979/6470 [22:45<51:26,  1.45it/s]
4|1980|Loss: 0.8989969491958618:  31%|███       | 1979/6470 [22:46<51:26,  1.45it/s]
4|1980|Loss: 0.8989969491958618:  31%|███       | 1980/6470 [22:46<52:13,  1.43it/s]
4|1981|Loss: 0.8475199937820435:  31%|███       | 1980/6470 [22:46<52:13,  1.43it/s]
4|1981|Loss: 0.8475199937820435:  31%|███       | 1981/6470 [22:47<48:41,  1.54it/s]
4|1982|Loss: 0.8559815883636475:  31%|███       | 1981/6470 [22:47<48:41,  1.54it/s]
4|1982|Loss: 0.8559815883636475:  31%|███       | 1982/6470 [22:47<49:15,  1.52it/s]
4|1983|Loss: 0.6726281642913818:  31%|███       | 1982/6470 [22:48<49:15,  1.52it/s]
4|1983|Loss: 0.6726281642913818:  31%|███       | 1983/6470 [22:48<52:50,  1.42it/s]
4|1984|Loss: 0.7570635080337524:  31%|███       | 1983/6470 [22:48<52:50,  1.42it/s]
4|1984|Loss: 0.7570635080337524:  31%|███       | 1984/6470 [22:49<51:51,  1.44it/s]
4|1985|Loss: 0.782625138759613:  31%|███       | 1984/6470 [22:49<51:51,  1.44it/s] 
4|1985|Loss: 0.782625138759613:  31%|███       | 1985/6470 [22:49<49:27,  1.51it/s]
4|1986|Loss: 0.8113782405853271:  31%|███       | 1985/6470 [22:50<49:27,  1.51it/s]
4|1986|Loss: 0.8113782405853271:  31%|███       | 1986/6470 [22:50<48:54,  1.53it/s]
4|1987|Loss: 0.9069193601608276:  31%|███       | 1986/6470 [22:50<48:54,  1.53it/s]
4|1987|Loss: 0.9069193601608276:  31%|███       | 1987/6470 [22:51<52:29,  1.42it/s]
4|1988|Loss: 0.88331538438797:  31%|███       | 1987/6470 [22:51<52:29,  1.42it/s]  
4|1988|Loss: 0.88331538438797:  31%|███       | 1988/6470 [22:51<50:09,  1.49it/s]
4|1989|Loss: 0.9058666825294495:  31%|███       | 1988/6470 [22:52<50:09,  1.49it/s]
4|1989|Loss: 0.9058666825294495:  31%|███       | 1989/6470 [22:52<53:14,  1.40it/s]
4|1990|Loss: 0.8554758429527283:  31%|███       | 1989/6470 [22:53<53:14,  1.40it/s]
4|1990|Loss: 0.8554758429527283:  31%|███       | 1990/6470 [22:53<52:50,  1.41it/s]
4|1991|Loss: 0.809515118598938:  31%|███       | 1990/6470 [22:53<52:50,  1.41it/s] 
4|1991|Loss: 0.809515118598938:  31%|███       | 1991/6470 [22:54<51:26,  1.45it/s]
4|1992|Loss: 0.8997756838798523:  31%|███       | 1991/6470 [22:54<51:26,  1.45it/s]
4|1992|Loss: 0.8997756838798523:  31%|███       | 1992/6470 [22:54<51:28,  1.45it/s]
4|1993|Loss: 0.9497854709625244:  31%|███       | 1992/6470 [22:55<51:28,  1.45it/s]
4|1993|Loss: 0.9497854709625244:  31%|███       | 1993/6470 [22:55<48:14,  1.55it/s]
4|1994|Loss: 0.7086987495422363:  31%|███       | 1993/6470 [22:55<48:14,  1.55it/s]
4|1994|Loss: 0.7086987495422363:  31%|███       | 1994/6470 [22:56<51:26,  1.45it/s]
4|1995|Loss: 0.6559023857116699:  31%|███       | 1994/6470 [22:56<51:26,  1.45it/s]
4|1995|Loss: 0.6559023857116699:  31%|███       | 1995/6470 [22:56<52:33,  1.42it/s]
4|1996|Loss: 0.9239739775657654:  31%|███       | 1995/6470 [22:57<52:33,  1.42it/s]
4|1996|Loss: 0.9239739775657654:  31%|███       | 1996/6470 [22:57<51:04,  1.46it/s]
4|1997|Loss: 0.886748194694519:  31%|███       | 1996/6470 [22:57<51:04,  1.46it/s] 
4|1997|Loss: 0.886748194694519:  31%|███       | 1997/6470 [22:58<50:00,  1.49it/s]
4|1998|Loss: 0.7400410175323486:  31%|███       | 1997/6470 [22:58<50:00,  1.49it/s]
4|1998|Loss: 0.7400410175323486:  31%|███       | 1998/6470 [22:58<50:25,  1.48it/s]
4|1999|Loss: 0.8282197713851929:  31%|███       | 1998/6470 [22:59<50:25,  1.48it/s]
4|1999|Loss: 0.8282197713851929:  31%|███       | 1999/6470 [22:59<49:30,  1.51it/s]
4|2000|Loss: 0.8645465970039368:  31%|███       | 1999/6470 [22:59<49:30,  1.51it/s]
4|2000|Loss: 0.8645465970039368:  31%|███       | 2000/6470 [23:00<53:43,  1.39it/s]
4|2001|Loss: 0.897335946559906:  31%|███       | 2000/6470 [23:00<53:43,  1.39it/s] 
4|2001|Loss: 0.897335946559906:  31%|███       | 2001/6470 [23:01<55:06,  1.35it/s]
4|2002|Loss: 0.8176151514053345:  31%|███       | 2001/6470 [23:01<55:06,  1.35it/s]
4|2002|Loss: 0.8176151514053345:  31%|███       | 2002/6470 [23:01<55:16,  1.35it/s]
4|2003|Loss: 0.9806758165359497:  31%|███       | 2002/6470 [23:02<55:16,  1.35it/s]
4|2003|Loss: 0.9806758165359497:  31%|███       | 2003/6470 [23:02<51:25,  1.45it/s]
4|2004|Loss: 0.7365200519561768:  31%|███       | 2003/6470 [23:02<51:25,  1.45it/s]
4|2004|Loss: 0.7365200519561768:  31%|███       | 2004/6470 [23:03<55:29,  1.34it/s]
4|2005|Loss: 0.8978888392448425:  31%|███       | 2004/6470 [23:03<55:29,  1.34it/s]
4|2005|Loss: 0.8978888392448425:  31%|███       | 2005/6470 [23:03<54:24,  1.37it/s]
4|2006|Loss: 0.7623540759086609:  31%|███       | 2005/6470 [23:04<54:24,  1.37it/s]
4|2006|Loss: 0.7623540759086609:  31%|███       | 2006/6470 [23:04<55:37,  1.34it/s]
4|2007|Loss: 0.7707806825637817:  31%|███       | 2006/6470 [23:05<55:37,  1.34it/s]
4|2007|Loss: 0.7707806825637817:  31%|███       | 2007/6470 [23:05<53:40,  1.39it/s]
4|2008|Loss: 0.5411999225616455:  31%|███       | 2007/6470 [23:05<53:40,  1.39it/s]
4|2008|Loss: 0.5411999225616455:  31%|███       | 2008/6470 [23:06<59:10,  1.26it/s]
4|2009|Loss: 0.7459651231765747:  31%|███       | 2008/6470 [23:06<59:10,  1.26it/s]
4|2009|Loss: 0.7459651231765747:  31%|███       | 2009/6470 [23:07<1:00:09,  1.24it/s]
4|2010|Loss: 0.949044942855835:  31%|███       | 2009/6470 [23:07<1:00:09,  1.24it/s] 
4|2010|Loss: 0.949044942855835:  31%|███       | 2010/6470 [23:07<54:24,  1.37it/s]  
4|2011|Loss: 1.0001308917999268:  31%|███       | 2010/6470 [23:08<54:24,  1.37it/s]
4|2011|Loss: 1.0001308917999268:  31%|███       | 2011/6470 [23:08<54:37,  1.36it/s]
4|2012|Loss: 0.8546924591064453:  31%|███       | 2011/6470 [23:08<54:37,  1.36it/s]
4|2012|Loss: 0.8546924591064453:  31%|███       | 2012/6470 [23:09<51:36,  1.44it/s]
4|2013|Loss: 1.0718356370925903:  31%|███       | 2012/6470 [23:09<51:36,  1.44it/s]
4|2013|Loss: 1.0718356370925903:  31%|███       | 2013/6470 [23:09<49:34,  1.50it/s]
4|2014|Loss: 0.8240240812301636:  31%|███       | 2013/6470 [23:10<49:34,  1.50it/s]
4|2014|Loss: 0.8240240812301636:  31%|███       | 2014/6470 [23:10<47:45,  1.55it/s]
4|2015|Loss: 0.9779603481292725:  31%|███       | 2014/6470 [23:10<47:45,  1.55it/s]
4|2015|Loss: 0.9779603481292725:  31%|███       | 2015/6470 [23:10<45:07,  1.65it/s]
4|2016|Loss: 0.9448079466819763:  31%|███       | 2015/6470 [23:11<45:07,  1.65it/s]
4|2016|Loss: 0.9448079466819763:  31%|███       | 2016/6470 [23:11<49:03,  1.51it/s]
4|2017|Loss: 1.013729214668274:  31%|███       | 2016/6470 [23:12<49:03,  1.51it/s] 
4|2017|Loss: 1.013729214668274:  31%|███       | 2017/6470 [23:12<52:57,  1.40it/s]
4|2018|Loss: 0.8588390946388245:  31%|███       | 2017/6470 [23:12<52:57,  1.40it/s]
4|2018|Loss: 0.8588390946388245:  31%|███       | 2018/6470 [23:13<57:42,  1.29it/s]
4|2019|Loss: 0.9609619975090027:  31%|███       | 2018/6470 [23:13<57:42,  1.29it/s]
4|2019|Loss: 0.9609619975090027:  31%|███       | 2019/6470 [23:13<54:19,  1.37it/s]
4|2020|Loss: 0.823514997959137:  31%|███       | 2019/6470 [23:14<54:19,  1.37it/s] 
4|2020|Loss: 0.823514997959137:  31%|███       | 2020/6470 [23:14<54:28,  1.36it/s]
4|2021|Loss: 0.723356306552887:  31%|███       | 2020/6470 [23:15<54:28,  1.36it/s]
4|2021|Loss: 0.723356306552887:  31%|███       | 2021/6470 [23:15<54:23,  1.36it/s]
4|2022|Loss: 0.7529743909835815:  31%|███       | 2021/6470 [23:15<54:23,  1.36it/s]
4|2022|Loss: 0.7529743909835815:  31%|███▏      | 2022/6470 [23:16<54:34,  1.36it/s]
4|2023|Loss: 0.9077969789505005:  31%|███▏      | 2022/6470 [23:16<54:34,  1.36it/s]
4|2023|Loss: 0.9077969789505005:  31%|███▏      | 2023/6470 [23:16<50:16,  1.47it/s]
4|2024|Loss: 0.9128643274307251:  31%|███▏      | 2023/6470 [23:17<50:16,  1.47it/s]
4|2024|Loss: 0.9128643274307251:  31%|███▏      | 2024/6470 [23:17<49:48,  1.49it/s]
4|2025|Loss: 1.048644781112671:  31%|███▏      | 2024/6470 [23:17<49:48,  1.49it/s] 
4|2025|Loss: 1.048644781112671:  31%|███▏      | 2025/6470 [23:17<48:14,  1.54it/s]
4|2026|Loss: 0.8734451532363892:  31%|███▏      | 2025/6470 [23:18<48:14,  1.54it/s]
4|2026|Loss: 0.8734451532363892:  31%|███▏      | 2026/6470 [23:18<49:25,  1.50it/s]
4|2027|Loss: 0.784744143486023:  31%|███▏      | 2026/6470 [23:19<49:25,  1.50it/s] 
4|2027|Loss: 0.784744143486023:  31%|███▏      | 2027/6470 [23:19<50:21,  1.47it/s]
4|2028|Loss: 0.8553978800773621:  31%|███▏      | 2027/6470 [23:19<50:21,  1.47it/s]
4|2028|Loss: 0.8553978800773621:  31%|███▏      | 2028/6470 [23:20<49:24,  1.50it/s]
4|2029|Loss: 0.8896812796592712:  31%|███▏      | 2028/6470 [23:20<49:24,  1.50it/s]
4|2029|Loss: 0.8896812796592712:  31%|███▏      | 2029/6470 [23:20<52:29,  1.41it/s]
4|2030|Loss: 0.9026502966880798:  31%|███▏      | 2029/6470 [23:21<52:29,  1.41it/s]
4|2030|Loss: 0.9026502966880798:  31%|███▏      | 2030/6470 [23:21<53:13,  1.39it/s]
4|2031|Loss: 0.766434907913208:  31%|███▏      | 2030/6470 [23:21<53:13,  1.39it/s] 
4|2031|Loss: 0.766434907913208:  31%|███▏      | 2031/6470 [23:22<53:01,  1.40it/s]
4|2032|Loss: 0.8284757733345032:  31%|███▏      | 2031/6470 [23:22<53:01,  1.40it/s]
4|2032|Loss: 0.8284757733345032:  31%|███▏      | 2032/6470 [23:22<50:25,  1.47it/s]
4|2033|Loss: 0.8018821477890015:  31%|███▏      | 2032/6470 [23:23<50:25,  1.47it/s]
4|2033|Loss: 0.8018821477890015:  31%|███▏      | 2033/6470 [23:23<47:12,  1.57it/s]
4|2034|Loss: 0.922044038772583:  31%|███▏      | 2033/6470 [23:23<47:12,  1.57it/s] 
4|2034|Loss: 0.922044038772583:  31%|███▏      | 2034/6470 [23:24<48:38,  1.52it/s]
4|2035|Loss: 0.8760151267051697:  31%|███▏      | 2034/6470 [23:24<48:38,  1.52it/s]
4|2035|Loss: 0.8760151267051697:  31%|███▏      | 2035/6470 [23:24<47:53,  1.54it/s]
4|2036|Loss: 0.805840015411377:  31%|███▏      | 2035/6470 [23:25<47:53,  1.54it/s] 
4|2036|Loss: 0.805840015411377:  31%|███▏      | 2036/6470 [23:25<50:22,  1.47it/s]
4|2037|Loss: 1.069947361946106:  31%|███▏      | 2036/6470 [23:25<50:22,  1.47it/s]
4|2037|Loss: 1.069947361946106:  31%|███▏      | 2037/6470 [23:26<52:19,  1.41it/s]
4|2038|Loss: 0.9310866594314575:  31%|███▏      | 2037/6470 [23:26<52:19,  1.41it/s]
4|2038|Loss: 0.9310866594314575:  31%|███▏      | 2038/6470 [23:26<46:55,  1.57it/s]
4|2039|Loss: 0.9766828417778015:  31%|███▏      | 2038/6470 [23:27<46:55,  1.57it/s]
4|2039|Loss: 0.9766828417778015:  32%|███▏      | 2039/6470 [23:27<44:57,  1.64it/s]
4|2040|Loss: 0.8162636756896973:  32%|███▏      | 2039/6470 [23:27<44:57,  1.64it/s]
4|2040|Loss: 0.8162636756896973:  32%|███▏      | 2040/6470 [23:28<49:45,  1.48it/s]
4|2041|Loss: 0.9151429533958435:  32%|███▏      | 2040/6470 [23:28<49:45,  1.48it/s]
4|2041|Loss: 0.9151429533958435:  32%|███▏      | 2041/6470 [23:28<52:05,  1.42it/s]
4|2042|Loss: 0.8476812243461609:  32%|███▏      | 2041/6470 [23:29<52:05,  1.42it/s]
4|2042|Loss: 0.8476812243461609:  32%|███▏      | 2042/6470 [23:29<56:11,  1.31it/s]
4|2043|Loss: 0.8127094507217407:  32%|███▏      | 2042/6470 [23:30<56:11,  1.31it/s]
4|2043|Loss: 0.8127094507217407:  32%|███▏      | 2043/6470 [23:30<54:59,  1.34it/s]
4|2044|Loss: 0.8913736939430237:  32%|███▏      | 2043/6470 [23:30<54:59,  1.34it/s]
4|2044|Loss: 0.8913736939430237:  32%|███▏      | 2044/6470 [23:31<53:49,  1.37it/s]
4|2045|Loss: 0.8050386309623718:  32%|███▏      | 2044/6470 [23:31<53:49,  1.37it/s]
4|2045|Loss: 0.8050386309623718:  32%|███▏      | 2045/6470 [23:32<56:29,  1.31it/s]
4|2046|Loss: 0.8578764200210571:  32%|███▏      | 2045/6470 [23:32<56:29,  1.31it/s]
4|2046|Loss: 0.8578764200210571:  32%|███▏      | 2046/6470 [23:32<51:42,  1.43it/s]
4|2047|Loss: 0.859957754611969:  32%|███▏      | 2046/6470 [23:32<51:42,  1.43it/s] 
4|2047|Loss: 0.859957754611969:  32%|███▏      | 2047/6470 [23:33<49:24,  1.49it/s]
4|2048|Loss: 0.9065712094306946:  32%|███▏      | 2047/6470 [23:33<49:24,  1.49it/s]
4|2048|Loss: 0.9065712094306946:  32%|███▏      | 2048/6470 [23:33<48:24,  1.52it/s]
4|2049|Loss: 0.8318325877189636:  32%|███▏      | 2048/6470 [23:34<48:24,  1.52it/s]
4|2049|Loss: 0.8318325877189636:  32%|███▏      | 2049/6470 [23:34<48:56,  1.51it/s]
4|2050|Loss: 0.9357560276985168:  32%|███▏      | 2049/6470 [23:34<48:56,  1.51it/s]
4|2050|Loss: 0.9357560276985168:  32%|███▏      | 2050/6470 [23:35<50:18,  1.46it/s]
4|2051|Loss: 0.8178078532218933:  32%|███▏      | 2050/6470 [23:35<50:18,  1.46it/s]
4|2051|Loss: 0.8178078532218933:  32%|███▏      | 2051/6470 [23:35<48:41,  1.51it/s]
4|2052|Loss: 1.0298975706100464:  32%|███▏      | 2051/6470 [23:36<48:41,  1.51it/s]
4|2052|Loss: 1.0298975706100464:  32%|███▏      | 2052/6470 [23:36<50:49,  1.45it/s]
4|2053|Loss: 0.7620338201522827:  32%|███▏      | 2052/6470 [23:36<50:49,  1.45it/s]
4|2053|Loss: 0.7620338201522827:  32%|███▏      | 2053/6470 [23:37<48:10,  1.53it/s]
4|2054|Loss: 0.7737091183662415:  32%|███▏      | 2053/6470 [23:37<48:10,  1.53it/s]
4|2054|Loss: 0.7737091183662415:  32%|███▏      | 2054/6470 [23:37<49:08,  1.50it/s]
4|2055|Loss: 0.6980210542678833:  32%|███▏      | 2054/6470 [23:38<49:08,  1.50it/s]
4|2055|Loss: 0.6980210542678833:  32%|███▏      | 2055/6470 [23:38<44:20,  1.66it/s]
4|2056|Loss: 0.8349354267120361:  32%|███▏      | 2055/6470 [23:38<44:20,  1.66it/s]
4|2056|Loss: 0.8349354267120361:  32%|███▏      | 2056/6470 [23:38<46:02,  1.60it/s]
4|2057|Loss: 0.785027265548706:  32%|███▏      | 2056/6470 [23:39<46:02,  1.60it/s] 
4|2057|Loss: 0.785027265548706:  32%|███▏      | 2057/6470 [23:39<48:05,  1.53it/s]
4|2058|Loss: 0.8964502811431885:  32%|███▏      | 2057/6470 [23:40<48:05,  1.53it/s]
4|2058|Loss: 0.8964502811431885:  32%|███▏      | 2058/6470 [23:40<48:05,  1.53it/s]
4|2059|Loss: 0.7888472080230713:  32%|███▏      | 2058/6470 [23:40<48:05,  1.53it/s]
4|2059|Loss: 0.7888472080230713:  32%|███▏      | 2059/6470 [23:41<49:43,  1.48it/s]
4|2060|Loss: 0.9092928171157837:  32%|███▏      | 2059/6470 [23:41<49:43,  1.48it/s]
4|2060|Loss: 0.9092928171157837:  32%|███▏      | 2060/6470 [23:41<50:37,  1.45it/s]
4|2061|Loss: 0.7335973978042603:  32%|███▏      | 2060/6470 [23:42<50:37,  1.45it/s]
4|2061|Loss: 0.7335973978042603:  32%|███▏      | 2061/6470 [23:42<49:46,  1.48it/s]
4|2062|Loss: 0.8163229823112488:  32%|███▏      | 2061/6470 [23:42<49:46,  1.48it/s]
4|2062|Loss: 0.8163229823112488:  32%|███▏      | 2062/6470 [23:43<48:26,  1.52it/s]
4|2063|Loss: 0.739257276058197:  32%|███▏      | 2062/6470 [23:43<48:26,  1.52it/s] 
4|2063|Loss: 0.739257276058197:  32%|███▏      | 2063/6470 [23:43<47:48,  1.54it/s]
4|2064|Loss: 0.7758857011795044:  32%|███▏      | 2063/6470 [23:44<47:48,  1.54it/s]
4|2064|Loss: 0.7758857011795044:  32%|███▏      | 2064/6470 [23:44<51:15,  1.43it/s]
4|2065|Loss: 0.9121431708335876:  32%|███▏      | 2064/6470 [23:44<51:15,  1.43it/s]
4|2065|Loss: 0.9121431708335876:  32%|███▏      | 2065/6470 [23:45<50:46,  1.45it/s]
4|2066|Loss: 0.8131252527236938:  32%|███▏      | 2065/6470 [23:45<50:46,  1.45it/s]
4|2066|Loss: 0.8131252527236938:  32%|███▏      | 2066/6470 [23:45<51:25,  1.43it/s]
4|2067|Loss: 0.7080682516098022:  32%|███▏      | 2066/6470 [23:46<51:25,  1.43it/s]
4|2067|Loss: 0.7080682516098022:  32%|███▏      | 2067/6470 [23:46<52:16,  1.40it/s]
4|2068|Loss: 0.8959097862243652:  32%|███▏      | 2067/6470 [23:47<52:16,  1.40it/s]
4|2068|Loss: 0.8959097862243652:  32%|███▏      | 2068/6470 [23:47<52:07,  1.41it/s]
4|2069|Loss: 0.888610303401947:  32%|███▏      | 2068/6470 [23:47<52:07,  1.41it/s] 
4|2069|Loss: 0.888610303401947:  32%|███▏      | 2069/6470 [23:47<49:45,  1.47it/s]
4|2070|Loss: 0.9461647272109985:  32%|███▏      | 2069/6470 [23:48<49:45,  1.47it/s]
4|2070|Loss: 0.9461647272109985:  32%|███▏      | 2070/6470 [23:48<49:47,  1.47it/s]
4|2071|Loss: 0.7779116630554199:  32%|███▏      | 2070/6470 [23:49<49:47,  1.47it/s]
4|2071|Loss: 0.7779116630554199:  32%|███▏      | 2071/6470 [23:49<53:17,  1.38it/s]
4|2072|Loss: 0.9316633343696594:  32%|███▏      | 2071/6470 [23:49<53:17,  1.38it/s]
4|2072|Loss: 0.9316633343696594:  32%|███▏      | 2072/6470 [23:50<53:35,  1.37it/s]
4|2073|Loss: 0.9135439395904541:  32%|███▏      | 2072/6470 [23:50<53:35,  1.37it/s]
4|2073|Loss: 0.9135439395904541:  32%|███▏      | 2073/6470 [23:50<54:33,  1.34it/s]
4|2074|Loss: 0.9617398977279663:  32%|███▏      | 2073/6470 [23:51<54:33,  1.34it/s]
4|2074|Loss: 0.9617398977279663:  32%|███▏      | 2074/6470 [23:51<54:01,  1.36it/s]
4|2075|Loss: 0.7911979556083679:  32%|███▏      | 2074/6470 [23:52<54:01,  1.36it/s]
4|2075|Loss: 0.7911979556083679:  32%|███▏      | 2075/6470 [23:52<54:16,  1.35it/s]
4|2076|Loss: 0.8520219326019287:  32%|███▏      | 2075/6470 [23:52<54:16,  1.35it/s]
4|2076|Loss: 0.8520219326019287:  32%|███▏      | 2076/6470 [23:53<52:21,  1.40it/s]
4|2077|Loss: 0.9747695326805115:  32%|███▏      | 2076/6470 [23:53<52:21,  1.40it/s]
4|2077|Loss: 0.9747695326805115:  32%|███▏      | 2077/6470 [23:53<52:34,  1.39it/s]
4|2078|Loss: 0.7678924202919006:  32%|███▏      | 2077/6470 [23:54<52:34,  1.39it/s]
4|2078|Loss: 0.7678924202919006:  32%|███▏      | 2078/6470 [23:54<50:58,  1.44it/s]
4|2079|Loss: 0.9125065207481384:  32%|███▏      | 2078/6470 [23:54<50:58,  1.44it/s]
4|2079|Loss: 0.9125065207481384:  32%|███▏      | 2079/6470 [23:55<49:12,  1.49it/s]
4|2080|Loss: 0.8005847930908203:  32%|███▏      | 2079/6470 [23:55<49:12,  1.49it/s]
4|2080|Loss: 0.8005847930908203:  32%|███▏      | 2080/6470 [23:55<46:04,  1.59it/s]
4|2081|Loss: 0.9181869626045227:  32%|███▏      | 2080/6470 [23:55<46:04,  1.59it/s]
4|2081|Loss: 0.9181869626045227:  32%|███▏      | 2081/6470 [23:56<44:39,  1.64it/s]
4|2082|Loss: 0.8284009099006653:  32%|███▏      | 2081/6470 [23:56<44:39,  1.64it/s]
4|2082|Loss: 0.8284009099006653:  32%|███▏      | 2082/6470 [23:56<44:46,  1.63it/s]
4|2083|Loss: 0.9138684272766113:  32%|███▏      | 2082/6470 [23:57<44:46,  1.63it/s]
4|2083|Loss: 0.9138684272766113:  32%|███▏      | 2083/6470 [23:57<45:26,  1.61it/s]
4|2084|Loss: 0.905182421207428:  32%|███▏      | 2083/6470 [23:57<45:26,  1.61it/s] 
4|2084|Loss: 0.905182421207428:  32%|███▏      | 2084/6470 [23:58<45:56,  1.59it/s]
4|2085|Loss: 0.8473004698753357:  32%|███▏      | 2084/6470 [23:58<45:56,  1.59it/s]
4|2085|Loss: 0.8473004698753357:  32%|███▏      | 2085/6470 [23:58<47:18,  1.54it/s]
4|2086|Loss: 0.7981886863708496:  32%|███▏      | 2085/6470 [23:59<47:18,  1.54it/s]
4|2086|Loss: 0.7981886863708496:  32%|███▏      | 2086/6470 [23:59<50:02,  1.46it/s]
4|2087|Loss: 0.8163976073265076:  32%|███▏      | 2086/6470 [23:59<50:02,  1.46it/s]
4|2087|Loss: 0.8163976073265076:  32%|███▏      | 2087/6470 [24:00<47:46,  1.53it/s]
4|2088|Loss: 0.720483660697937:  32%|███▏      | 2087/6470 [24:00<47:46,  1.53it/s] 
4|2088|Loss: 0.720483660697937:  32%|███▏      | 2088/6470 [24:00<46:22,  1.57it/s]
4|2089|Loss: 0.831855058670044:  32%|███▏      | 2088/6470 [24:01<46:22,  1.57it/s]
4|2089|Loss: 0.831855058670044:  32%|███▏      | 2089/6470 [24:01<48:18,  1.51it/s]
4|2090|Loss: 0.7940918207168579:  32%|███▏      | 2089/6470 [24:01<48:18,  1.51it/s]
4|2090|Loss: 0.7940918207168579:  32%|███▏      | 2090/6470 [24:02<47:58,  1.52it/s]
4|2091|Loss: 0.8204848170280457:  32%|███▏      | 2090/6470 [24:02<47:58,  1.52it/s]
4|2091|Loss: 0.8204848170280457:  32%|███▏      | 2091/6470 [24:02<50:04,  1.46it/s]
4|2092|Loss: 0.9260705709457397:  32%|███▏      | 2091/6470 [24:03<50:04,  1.46it/s]
4|2092|Loss: 0.9260705709457397:  32%|███▏      | 2092/6470 [24:03<49:59,  1.46it/s]
4|2093|Loss: 0.7710797786712646:  32%|███▏      | 2092/6470 [24:03<49:59,  1.46it/s]
4|2093|Loss: 0.7710797786712646:  32%|███▏      | 2093/6470 [24:04<48:48,  1.49it/s]
4|2094|Loss: 0.7990472316741943:  32%|███▏      | 2093/6470 [24:04<48:48,  1.49it/s]
4|2094|Loss: 0.7990472316741943:  32%|███▏      | 2094/6470 [24:04<49:38,  1.47it/s]
4|2095|Loss: 0.8364152312278748:  32%|███▏      | 2094/6470 [24:05<49:38,  1.47it/s]
4|2095|Loss: 0.8364152312278748:  32%|███▏      | 2095/6470 [24:05<55:14,  1.32it/s]
4|2096|Loss: 0.6908319592475891:  32%|███▏      | 2095/6470 [24:06<55:14,  1.32it/s]
4|2096|Loss: 0.6908319592475891:  32%|███▏      | 2096/6470 [24:06<55:29,  1.31it/s]
4|2097|Loss: 0.8542303442955017:  32%|███▏      | 2096/6470 [24:06<55:29,  1.31it/s]
4|2097|Loss: 0.8542303442955017:  32%|███▏      | 2097/6470 [24:07<54:14,  1.34it/s]
4|2098|Loss: 0.8412453532218933:  32%|███▏      | 2097/6470 [24:07<54:14,  1.34it/s]
4|2098|Loss: 0.8412453532218933:  32%|███▏      | 2098/6470 [24:07<51:44,  1.41it/s]
4|2099|Loss: 0.8293986320495605:  32%|███▏      | 2098/6470 [24:08<51:44,  1.41it/s]
4|2099|Loss: 0.8293986320495605:  32%|███▏      | 2099/6470 [24:08<53:47,  1.35it/s]
4|2100|Loss: 0.730252742767334:  32%|███▏      | 2099/6470 [24:09<53:47,  1.35it/s] 
4|2100|Loss: 0.730252742767334:  32%|███▏      | 2100/6470 [24:09<56:23,  1.29it/s]
4|2101|Loss: 0.7494897842407227:  32%|███▏      | 2100/6470 [24:10<56:23,  1.29it/s]
4|2101|Loss: 0.7494897842407227:  32%|███▏      | 2101/6470 [24:10<56:04,  1.30it/s]
4|2102|Loss: 0.8208202123641968:  32%|███▏      | 2101/6470 [24:10<56:04,  1.30it/s]
4|2102|Loss: 0.8208202123641968:  32%|███▏      | 2102/6470 [24:11<54:35,  1.33it/s]
4|2103|Loss: 0.7988668084144592:  32%|███▏      | 2102/6470 [24:11<54:35,  1.33it/s]
4|2103|Loss: 0.7988668084144592:  33%|███▎      | 2103/6470 [24:11<52:00,  1.40it/s]
4|2104|Loss: 0.795892059803009:  33%|███▎      | 2103/6470 [24:12<52:00,  1.40it/s] 
4|2104|Loss: 0.795892059803009:  33%|███▎      | 2104/6470 [24:12<48:16,  1.51it/s]
4|2105|Loss: 0.8885003924369812:  33%|███▎      | 2104/6470 [24:12<48:16,  1.51it/s]
4|2105|Loss: 0.8885003924369812:  33%|███▎      | 2105/6470 [24:13<50:36,  1.44it/s]
4|2106|Loss: 0.8447149395942688:  33%|███▎      | 2105/6470 [24:13<50:36,  1.44it/s]
4|2106|Loss: 0.8447149395942688:  33%|███▎      | 2106/6470 [24:13<50:05,  1.45it/s]
4|2107|Loss: 0.9243671894073486:  33%|███▎      | 2106/6470 [24:13<50:05,  1.45it/s]
4|2107|Loss: 0.9243671894073486:  33%|███▎      | 2107/6470 [24:14<45:42,  1.59it/s]
4|2108|Loss: 0.9142055511474609:  33%|███▎      | 2107/6470 [24:14<45:42,  1.59it/s]
4|2108|Loss: 0.9142055511474609:  33%|███▎      | 2108/6470 [24:14<47:38,  1.53it/s]
4|2109|Loss: 0.8410170674324036:  33%|███▎      | 2108/6470 [24:15<47:38,  1.53it/s]
4|2109|Loss: 0.8410170674324036:  33%|███▎      | 2109/6470 [24:15<47:43,  1.52it/s]
4|2110|Loss: 0.8793728947639465:  33%|███▎      | 2109/6470 [24:15<47:43,  1.52it/s]
4|2110|Loss: 0.8793728947639465:  33%|███▎      | 2110/6470 [24:16<43:56,  1.65it/s]
4|2111|Loss: 0.8269516229629517:  33%|███▎      | 2110/6470 [24:16<43:56,  1.65it/s]
4|2111|Loss: 0.8269516229629517:  33%|███▎      | 2111/6470 [24:16<45:23,  1.60it/s]
4|2112|Loss: 0.7871347069740295:  33%|███▎      | 2111/6470 [24:17<45:23,  1.60it/s]
4|2112|Loss: 0.7871347069740295:  33%|███▎      | 2112/6470 [24:17<43:30,  1.67it/s]
4|2113|Loss: 0.7054605484008789:  33%|███▎      | 2112/6470 [24:17<43:30,  1.67it/s]
4|2113|Loss: 0.7054605484008789:  33%|███▎      | 2113/6470 [24:17<44:47,  1.62it/s]
4|2114|Loss: 0.9861898422241211:  33%|███▎      | 2113/6470 [24:18<44:47,  1.62it/s]
4|2114|Loss: 0.9861898422241211:  33%|███▎      | 2114/6470 [24:18<44:37,  1.63it/s]
4|2115|Loss: 0.6723126769065857:  33%|███▎      | 2114/6470 [24:18<44:37,  1.63it/s]
4|2115|Loss: 0.6723126769065857:  33%|███▎      | 2115/6470 [24:18<39:15,  1.85it/s]
4|2116|Loss: 0.8406572341918945:  33%|███▎      | 2115/6470 [24:19<39:15,  1.85it/s]
4|2116|Loss: 0.8406572341918945:  33%|███▎      | 2116/6470 [24:19<39:48,  1.82it/s]
4|2117|Loss: 0.7939596772193909:  33%|███▎      | 2116/6470 [24:19<39:48,  1.82it/s]
4|2117|Loss: 0.7939596772193909:  33%|███▎      | 2117/6470 [24:20<43:04,  1.68it/s]
4|2118|Loss: 0.9648858904838562:  33%|███▎      | 2117/6470 [24:20<43:04,  1.68it/s]
4|2118|Loss: 0.9648858904838562:  33%|███▎      | 2118/6470 [24:20<42:07,  1.72it/s]
4|2119|Loss: 0.7303317189216614:  33%|███▎      | 2118/6470 [24:21<42:07,  1.72it/s]
4|2119|Loss: 0.7303317189216614:  33%|███▎      | 2119/6470 [24:21<44:42,  1.62it/s]
4|2120|Loss: 0.9529659748077393:  33%|███▎      | 2119/6470 [24:21<44:42,  1.62it/s]
4|2120|Loss: 0.9529659748077393:  33%|███▎      | 2120/6470 [24:22<49:26,  1.47it/s]
4|2121|Loss: 1.0182160139083862:  33%|███▎      | 2120/6470 [24:22<49:26,  1.47it/s]
4|2121|Loss: 1.0182160139083862:  33%|███▎      | 2121/6470 [24:23<52:45,  1.37it/s]
4|2122|Loss: 0.8301339149475098:  33%|███▎      | 2121/6470 [24:23<52:45,  1.37it/s]
4|2122|Loss: 0.8301339149475098:  33%|███▎      | 2122/6470 [24:23<48:26,  1.50it/s]
4|2123|Loss: 0.977480947971344:  33%|███▎      | 2122/6470 [24:23<48:26,  1.50it/s] 
4|2123|Loss: 0.977480947971344:  33%|███▎      | 2123/6470 [24:24<45:17,  1.60it/s]
4|2124|Loss: 0.7250961661338806:  33%|███▎      | 2123/6470 [24:24<45:17,  1.60it/s]
4|2124|Loss: 0.7250961661338806:  33%|███▎      | 2124/6470 [24:24<40:07,  1.81it/s]
4|2125|Loss: 0.8467872738838196:  33%|███▎      | 2124/6470 [24:24<40:07,  1.81it/s]
4|2125|Loss: 0.8467872738838196:  33%|███▎      | 2125/6470 [24:25<40:35,  1.78it/s]
4|2126|Loss: 0.8356219530105591:  33%|███▎      | 2125/6470 [24:25<40:35,  1.78it/s]
4|2126|Loss: 0.8356219530105591:  33%|███▎      | 2126/6470 [24:25<40:49,  1.77it/s]
4|2127|Loss: 0.9286181926727295:  33%|███▎      | 2126/6470 [24:26<40:49,  1.77it/s]
4|2127|Loss: 0.9286181926727295:  33%|███▎      | 2127/6470 [24:26<43:12,  1.68it/s]
4|2128|Loss: 0.9004514217376709:  33%|███▎      | 2127/6470 [24:26<43:12,  1.68it/s]
4|2128|Loss: 0.9004514217376709:  33%|███▎      | 2128/6470 [24:26<41:14,  1.75it/s]
4|2129|Loss: 0.963580846786499:  33%|███▎      | 2128/6470 [24:27<41:14,  1.75it/s] 
4|2129|Loss: 0.963580846786499:  33%|███▎      | 2129/6470 [24:27<42:56,  1.68it/s]
4|2130|Loss: 0.6851654052734375:  33%|███▎      | 2129/6470 [24:27<42:56,  1.68it/s]
4|2130|Loss: 0.6851654052734375:  33%|███▎      | 2130/6470 [24:28<45:45,  1.58it/s]
4|2131|Loss: 0.9080808758735657:  33%|███▎      | 2130/6470 [24:28<45:45,  1.58it/s]
4|2131|Loss: 0.9080808758735657:  33%|███▎      | 2131/6470 [24:28<45:25,  1.59it/s]
4|2132|Loss: 0.78355872631073:  33%|███▎      | 2131/6470 [24:29<45:25,  1.59it/s]  
4|2132|Loss: 0.78355872631073:  33%|███▎      | 2132/6470 [24:29<42:16,  1.71it/s]
4|2133|Loss: 1.0581597089767456:  33%|███▎      | 2132/6470 [24:29<42:16,  1.71it/s]
4|2133|Loss: 1.0581597089767456:  33%|███▎      | 2133/6470 [24:29<44:28,  1.63it/s]
4|2134|Loss: 0.7858485579490662:  33%|███▎      | 2133/6470 [24:30<44:28,  1.63it/s]
4|2134|Loss: 0.7858485579490662:  33%|███▎      | 2134/6470 [24:30<44:33,  1.62it/s]
4|2135|Loss: 0.8969031572341919:  33%|███▎      | 2134/6470 [24:30<44:33,  1.62it/s]
4|2135|Loss: 0.8969031572341919:  33%|███▎      | 2135/6470 [24:31<42:29,  1.70it/s]
4|2136|Loss: 0.6929829120635986:  33%|███▎      | 2135/6470 [24:31<42:29,  1.70it/s]
4|2136|Loss: 0.6929829120635986:  33%|███▎      | 2136/6470 [24:31<47:51,  1.51it/s]
4|2137|Loss: 0.93425053358078:  33%|███▎      | 2136/6470 [24:32<47:51,  1.51it/s]  
4|2137|Loss: 0.93425053358078:  33%|███▎      | 2137/6470 [24:32<44:20,  1.63it/s]
4|2138|Loss: 0.9072687029838562:  33%|███▎      | 2137/6470 [24:32<44:20,  1.63it/s]
4|2138|Loss: 0.9072687029838562:  33%|███▎      | 2138/6470 [24:32<39:15,  1.84it/s]
4|2139|Loss: 0.8473454117774963:  33%|███▎      | 2138/6470 [24:33<39:15,  1.84it/s]
4|2139|Loss: 0.8473454117774963:  33%|███▎      | 2139/6470 [24:33<42:54,  1.68it/s]
4|2140|Loss: 0.9170658588409424:  33%|███▎      | 2139/6470 [24:34<42:54,  1.68it/s]
4|2140|Loss: 0.9170658588409424:  33%|███▎      | 2140/6470 [24:34<48:12,  1.50it/s]
4|2141|Loss: 0.9047236442565918:  33%|███▎      | 2140/6470 [24:34<48:12,  1.50it/s]
4|2141|Loss: 0.9047236442565918:  33%|███▎      | 2141/6470 [24:35<50:04,  1.44it/s]
4|2142|Loss: 0.9159268140792847:  33%|███▎      | 2141/6470 [24:35<50:04,  1.44it/s]
4|2142|Loss: 0.9159268140792847:  33%|███▎      | 2142/6470 [24:35<50:17,  1.43it/s]
4|2143|Loss: 0.8508005738258362:  33%|███▎      | 2142/6470 [24:36<50:17,  1.43it/s]
4|2143|Loss: 0.8508005738258362:  33%|███▎      | 2143/6470 [24:36<50:25,  1.43it/s]
4|2144|Loss: 0.7917227745056152:  33%|███▎      | 2143/6470 [24:36<50:25,  1.43it/s]
4|2144|Loss: 0.7917227745056152:  33%|███▎      | 2144/6470 [24:37<49:00,  1.47it/s]
4|2145|Loss: 0.858158528804779:  33%|███▎      | 2144/6470 [24:37<49:00,  1.47it/s] 
4|2145|Loss: 0.858158528804779:  33%|███▎      | 2145/6470 [24:38<52:42,  1.37it/s]
4|2146|Loss: 0.9261927604675293:  33%|███▎      | 2145/6470 [24:38<52:42,  1.37it/s]
4|2146|Loss: 0.9261927604675293:  33%|███▎      | 2146/6470 [24:38<54:05,  1.33it/s]
4|2147|Loss: 0.7581233978271484:  33%|███▎      | 2146/6470 [24:39<54:05,  1.33it/s]
4|2147|Loss: 0.7581233978271484:  33%|███▎      | 2147/6470 [24:39<52:18,  1.38it/s]
4|2148|Loss: 0.9490328431129456:  33%|███▎      | 2147/6470 [24:39<52:18,  1.38it/s]
4|2148|Loss: 0.9490328431129456:  33%|███▎      | 2148/6470 [24:40<50:09,  1.44it/s]
4|2149|Loss: 0.7669259905815125:  33%|███▎      | 2148/6470 [24:40<50:09,  1.44it/s]
4|2149|Loss: 0.7669259905815125:  33%|███▎      | 2149/6470 [24:40<48:58,  1.47it/s]
4|2150|Loss: 0.795310378074646:  33%|███▎      | 2149/6470 [24:41<48:58,  1.47it/s] 
4|2150|Loss: 0.795310378074646:  33%|███▎      | 2150/6470 [24:41<48:06,  1.50it/s]
4|2151|Loss: 0.7800361514091492:  33%|███▎      | 2150/6470 [24:41<48:06,  1.50it/s]
4|2151|Loss: 0.7800361514091492:  33%|███▎      | 2151/6470 [24:41<45:13,  1.59it/s]
4|2152|Loss: 0.672710657119751:  33%|███▎      | 2151/6470 [24:42<45:13,  1.59it/s] 
4|2152|Loss: 0.672710657119751:  33%|███▎      | 2152/6470 [24:43<59:30,  1.21it/s]
4|2153|Loss: 0.8209889531135559:  33%|███▎      | 2152/6470 [24:43<59:30,  1.21it/s]
4|2153|Loss: 0.8209889531135559:  33%|███▎      | 2153/6470 [24:44<1:01:17,  1.17it/s]
4|2154|Loss: 0.9596039056777954:  33%|███▎      | 2153/6470 [24:44<1:01:17,  1.17it/s]
4|2154|Loss: 0.9596039056777954:  33%|███▎      | 2154/6470 [24:44<1:00:27,  1.19it/s]
4|2155|Loss: 0.813860297203064:  33%|███▎      | 2154/6470 [24:45<1:00:27,  1.19it/s] 
4|2155|Loss: 0.813860297203064:  33%|███▎      | 2155/6470 [24:45<57:00,  1.26it/s]  
4|2156|Loss: 0.9234336018562317:  33%|███▎      | 2155/6470 [24:45<57:00,  1.26it/s]
4|2156|Loss: 0.9234336018562317:  33%|███▎      | 2156/6470 [24:46<51:31,  1.40it/s]
4|2157|Loss: 0.8612055778503418:  33%|███▎      | 2156/6470 [24:46<51:31,  1.40it/s]
4|2157|Loss: 0.8612055778503418:  33%|███▎      | 2157/6470 [24:46<50:23,  1.43it/s]
4|2158|Loss: 0.8211762309074402:  33%|███▎      | 2157/6470 [24:47<50:23,  1.43it/s]
4|2158|Loss: 0.8211762309074402:  33%|███▎      | 2158/6470 [24:47<47:50,  1.50it/s]
4|2159|Loss: 0.9693946242332458:  33%|███▎      | 2158/6470 [24:47<47:50,  1.50it/s]
4|2159|Loss: 0.9693946242332458:  33%|███▎      | 2159/6470 [24:48<48:31,  1.48it/s]
4|2160|Loss: 0.8397229909896851:  33%|███▎      | 2159/6470 [24:48<48:31,  1.48it/s]
4|2160|Loss: 0.8397229909896851:  33%|███▎      | 2160/6470 [24:48<51:01,  1.41it/s]
4|2161|Loss: 0.8920655250549316:  33%|███▎      | 2160/6470 [24:49<51:01,  1.41it/s]
4|2161|Loss: 0.8920655250549316:  33%|███▎      | 2161/6470 [24:49<51:27,  1.40it/s]
4|2162|Loss: 0.8703429102897644:  33%|███▎      | 2161/6470 [24:50<51:27,  1.40it/s]
4|2162|Loss: 0.8703429102897644:  33%|███▎      | 2162/6470 [24:50<51:59,  1.38it/s]
4|2163|Loss: 0.7976162433624268:  33%|███▎      | 2162/6470 [24:50<51:59,  1.38it/s]
4|2163|Loss: 0.7976162433624268:  33%|███▎      | 2163/6470 [24:51<55:26,  1.29it/s]
4|2164|Loss: 0.8324254155158997:  33%|███▎      | 2163/6470 [24:51<55:26,  1.29it/s]
4|2164|Loss: 0.8324254155158997:  33%|███▎      | 2164/6470 [24:52<57:43,  1.24it/s]
4|2165|Loss: 0.8396873474121094:  33%|███▎      | 2164/6470 [24:52<57:43,  1.24it/s]
4|2165|Loss: 0.8396873474121094:  33%|███▎      | 2165/6470 [24:52<53:25,  1.34it/s]
4|2166|Loss: 0.8124959468841553:  33%|███▎      | 2165/6470 [24:53<53:25,  1.34it/s]
4|2166|Loss: 0.8124959468841553:  33%|███▎      | 2166/6470 [24:53<50:15,  1.43it/s]
4|2167|Loss: 0.7576066851615906:  33%|███▎      | 2166/6470 [24:53<50:15,  1.43it/s]
4|2167|Loss: 0.7576066851615906:  33%|███▎      | 2167/6470 [24:53<48:25,  1.48it/s]
4|2168|Loss: 0.9259908199310303:  33%|███▎      | 2167/6470 [24:54<48:25,  1.48it/s]
4|2168|Loss: 0.9259908199310303:  34%|███▎      | 2168/6470 [24:54<49:30,  1.45it/s]
4|2169|Loss: 0.8446519374847412:  34%|███▎      | 2168/6470 [24:55<49:30,  1.45it/s]
4|2169|Loss: 0.8446519374847412:  34%|███▎      | 2169/6470 [24:55<46:55,  1.53it/s]
4|2170|Loss: 0.7217705249786377:  34%|███▎      | 2169/6470 [24:55<46:55,  1.53it/s]
4|2170|Loss: 0.7217705249786377:  34%|███▎      | 2170/6470 [24:55<46:07,  1.55it/s]
4|2171|Loss: 1.0772428512573242:  34%|███▎      | 2170/6470 [24:56<46:07,  1.55it/s]
4|2171|Loss: 1.0772428512573242:  34%|███▎      | 2171/6470 [24:56<47:59,  1.49it/s]
4|2172|Loss: 0.786206841468811:  34%|███▎      | 2171/6470 [24:56<47:59,  1.49it/s] 
4|2172|Loss: 0.786206841468811:  34%|███▎      | 2172/6470 [24:57<45:07,  1.59it/s]
4|2173|Loss: 1.0449758768081665:  34%|███▎      | 2172/6470 [24:57<45:07,  1.59it/s]
4|2173|Loss: 1.0449758768081665:  34%|███▎      | 2173/6470 [24:57<44:37,  1.60it/s]
4|2174|Loss: 0.871609628200531:  34%|███▎      | 2173/6470 [24:58<44:37,  1.60it/s] 
4|2174|Loss: 0.871609628200531:  34%|███▎      | 2174/6470 [24:58<46:34,  1.54it/s]
4|2175|Loss: 0.8313965201377869:  34%|███▎      | 2174/6470 [24:58<46:34,  1.54it/s]
4|2175|Loss: 0.8313965201377869:  34%|███▎      | 2175/6470 [24:59<50:51,  1.41it/s]
4|2176|Loss: 0.9214211106300354:  34%|███▎      | 2175/6470 [24:59<50:51,  1.41it/s]
4|2176|Loss: 0.9214211106300354:  34%|███▎      | 2176/6470 [25:00<51:33,  1.39it/s]
4|2177|Loss: 1.0265825986862183:  34%|███▎      | 2176/6470 [25:00<51:33,  1.39it/s]
4|2177|Loss: 1.0265825986862183:  34%|███▎      | 2177/6470 [25:00<51:07,  1.40it/s]
4|2178|Loss: 0.837944746017456:  34%|███▎      | 2177/6470 [25:01<51:07,  1.40it/s] 
4|2178|Loss: 0.837944746017456:  34%|███▎      | 2178/6470 [25:01<51:19,  1.39it/s]
4|2179|Loss: 0.9257259368896484:  34%|███▎      | 2178/6470 [25:01<51:19,  1.39it/s]
4|2179|Loss: 0.9257259368896484:  34%|███▎      | 2179/6470 [25:02<55:30,  1.29it/s]
4|2180|Loss: 0.9517266154289246:  34%|███▎      | 2179/6470 [25:02<55:30,  1.29it/s]
4|2180|Loss: 0.9517266154289246:  34%|███▎      | 2180/6470 [25:03<55:40,  1.28it/s]
4|2181|Loss: 0.8229021430015564:  34%|███▎      | 2180/6470 [25:03<55:40,  1.28it/s]
4|2181|Loss: 0.8229021430015564:  34%|███▎      | 2181/6470 [25:03<54:23,  1.31it/s]
4|2182|Loss: 0.850830078125:  34%|███▎      | 2181/6470 [25:04<54:23,  1.31it/s]    
4|2182|Loss: 0.850830078125:  34%|███▎      | 2182/6470 [25:04<53:39,  1.33it/s]
4|2183|Loss: 0.8555358052253723:  34%|███▎      | 2182/6470 [25:05<53:39,  1.33it/s]
4|2183|Loss: 0.8555358052253723:  34%|███▎      | 2183/6470 [25:05<50:11,  1.42it/s]
4|2184|Loss: 0.847645103931427:  34%|███▎      | 2183/6470 [25:05<50:11,  1.42it/s] 
4|2184|Loss: 0.847645103931427:  34%|███▍      | 2184/6470 [25:05<50:19,  1.42it/s]
4|2185|Loss: 0.7663849592208862:  34%|███▍      | 2184/6470 [25:06<50:19,  1.42it/s]
4|2185|Loss: 0.7663849592208862:  34%|███▍      | 2185/6470 [25:06<51:11,  1.39it/s]
4|2186|Loss: 0.7325804233551025:  34%|███▍      | 2185/6470 [25:07<51:11,  1.39it/s]
4|2186|Loss: 0.7325804233551025:  34%|███▍      | 2186/6470 [25:07<52:31,  1.36it/s]
4|2187|Loss: 0.832801878452301:  34%|███▍      | 2186/6470 [25:07<52:31,  1.36it/s] 
4|2187|Loss: 0.832801878452301:  34%|███▍      | 2187/6470 [25:08<52:55,  1.35it/s]
4|2188|Loss: 0.8129233121871948:  34%|███▍      | 2187/6470 [25:08<52:55,  1.35it/s]
4|2188|Loss: 0.8129233121871948:  34%|███▍      | 2188/6470 [25:08<52:48,  1.35it/s]
4|2189|Loss: 0.8816561102867126:  34%|███▍      | 2188/6470 [25:09<52:48,  1.35it/s]
4|2189|Loss: 0.8816561102867126:  34%|███▍      | 2189/6470 [25:09<48:48,  1.46it/s]
4|2190|Loss: 0.9041827917098999:  34%|███▍      | 2189/6470 [25:09<48:48,  1.46it/s]
4|2190|Loss: 0.9041827917098999:  34%|███▍      | 2190/6470 [25:10<46:08,  1.55it/s]
4|2191|Loss: 0.6920529007911682:  34%|███▍      | 2190/6470 [25:10<46:08,  1.55it/s]
4|2191|Loss: 0.6920529007911682:  34%|███▍      | 2191/6470 [25:10<45:14,  1.58it/s]
4|2192|Loss: 0.4869540333747864:  34%|███▍      | 2191/6470 [25:11<45:14,  1.58it/s]
4|2192|Loss: 0.4869540333747864:  34%|███▍      | 2192/6470 [25:11<45:07,  1.58it/s]
4|2193|Loss: 0.7579411268234253:  34%|███▍      | 2192/6470 [25:11<45:07,  1.58it/s]
4|2193|Loss: 0.7579411268234253:  34%|███▍      | 2193/6470 [25:12<47:01,  1.52it/s]
4|2194|Loss: 0.8556092977523804:  34%|███▍      | 2193/6470 [25:12<47:01,  1.52it/s]
4|2194|Loss: 0.8556092977523804:  34%|███▍      | 2194/6470 [25:12<46:46,  1.52it/s]
4|2195|Loss: 0.8002336025238037:  34%|███▍      | 2194/6470 [25:13<46:46,  1.52it/s]
4|2195|Loss: 0.8002336025238037:  34%|███▍      | 2195/6470 [25:13<46:27,  1.53it/s]
4|2196|Loss: 0.8554802536964417:  34%|███▍      | 2195/6470 [25:13<46:27,  1.53it/s]
4|2196|Loss: 0.8554802536964417:  34%|███▍      | 2196/6470 [25:14<49:30,  1.44it/s]
4|2197|Loss: 0.9313228130340576:  34%|███▍      | 2196/6470 [25:14<49:30,  1.44it/s]
4|2197|Loss: 0.9313228130340576:  34%|███▍      | 2197/6470 [25:14<48:24,  1.47it/s]
4|2198|Loss: 0.9146140217781067:  34%|███▍      | 2197/6470 [25:15<48:24,  1.47it/s]
4|2198|Loss: 0.9146140217781067:  34%|███▍      | 2198/6470 [25:15<48:19,  1.47it/s]
4|2199|Loss: 0.7717278003692627:  34%|███▍      | 2198/6470 [25:15<48:19,  1.47it/s]
4|2199|Loss: 0.7717278003692627:  34%|███▍      | 2199/6470 [25:16<47:45,  1.49it/s]
4|2200|Loss: 0.9857050776481628:  34%|███▍      | 2199/6470 [25:16<47:45,  1.49it/s]
4|2200|Loss: 0.9857050776481628:  34%|███▍      | 2200/6470 [25:16<48:10,  1.48it/s]
4|2201|Loss: 0.9082742929458618:  34%|███▍      | 2200/6470 [25:17<48:10,  1.48it/s]
4|2201|Loss: 0.9082742929458618:  34%|███▍      | 2201/6470 [25:17<46:41,  1.52it/s]
4|2202|Loss: 0.827246367931366:  34%|███▍      | 2201/6470 [25:17<46:41,  1.52it/s] 
4|2202|Loss: 0.827246367931366:  34%|███▍      | 2202/6470 [25:18<48:07,  1.48it/s]
4|2203|Loss: 0.6524040102958679:  34%|███▍      | 2202/6470 [25:18<48:07,  1.48it/s]
4|2203|Loss: 0.6524040102958679:  34%|███▍      | 2203/6470 [25:18<47:41,  1.49it/s]
4|2204|Loss: 0.8235466480255127:  34%|███▍      | 2203/6470 [25:19<47:41,  1.49it/s]
4|2204|Loss: 0.8235466480255127:  34%|███▍      | 2204/6470 [25:19<48:47,  1.46it/s]
4|2205|Loss: 0.8518890142440796:  34%|███▍      | 2204/6470 [25:19<48:47,  1.46it/s]
4|2205|Loss: 0.8518890142440796:  34%|███▍      | 2205/6470 [25:20<47:57,  1.48it/s]
4|2206|Loss: 0.8889066576957703:  34%|███▍      | 2205/6470 [25:20<47:57,  1.48it/s]
4|2206|Loss: 0.8889066576957703:  34%|███▍      | 2206/6470 [25:20<47:11,  1.51it/s]
4|2207|Loss: 0.9072248935699463:  34%|███▍      | 2206/6470 [25:21<47:11,  1.51it/s]
4|2207|Loss: 0.9072248935699463:  34%|███▍      | 2207/6470 [25:21<49:17,  1.44it/s]
4|2208|Loss: 0.6829559803009033:  34%|███▍      | 2207/6470 [25:21<49:17,  1.44it/s]
4|2208|Loss: 0.6829559803009033:  34%|███▍      | 2208/6470 [25:22<49:59,  1.42it/s]
4|2209|Loss: 0.9011268019676208:  34%|███▍      | 2208/6470 [25:22<49:59,  1.42it/s]
4|2209|Loss: 0.9011268019676208:  34%|███▍      | 2209/6470 [25:22<47:21,  1.50it/s]
4|2210|Loss: 0.8447605967521667:  34%|███▍      | 2209/6470 [25:23<47:21,  1.50it/s]
4|2210|Loss: 0.8447605967521667:  34%|███▍      | 2210/6470 [25:23<50:07,  1.42it/s]
4|2211|Loss: 1.0707523822784424:  34%|███▍      | 2210/6470 [25:24<50:07,  1.42it/s]
4|2211|Loss: 1.0707523822784424:  34%|███▍      | 2211/6470 [25:24<51:18,  1.38it/s]
4|2212|Loss: 0.7809857726097107:  34%|███▍      | 2211/6470 [25:24<51:18,  1.38it/s]
4|2212|Loss: 0.7809857726097107:  34%|███▍      | 2212/6470 [25:24<46:47,  1.52it/s]
4|2213|Loss: 0.7884233593940735:  34%|███▍      | 2212/6470 [25:25<46:47,  1.52it/s]
4|2213|Loss: 0.7884233593940735:  34%|███▍      | 2213/6470 [25:25<44:31,  1.59it/s]
4|2214|Loss: 0.840106725692749:  34%|███▍      | 2213/6470 [25:25<44:31,  1.59it/s] 
4|2214|Loss: 0.840106725692749:  34%|███▍      | 2214/6470 [25:26<44:07,  1.61it/s]
4|2215|Loss: 0.854461669921875:  34%|███▍      | 2214/6470 [25:26<44:07,  1.61it/s]
4|2215|Loss: 0.854461669921875:  34%|███▍      | 2215/6470 [25:26<41:45,  1.70it/s]
4|2216|Loss: 0.8133284449577332:  34%|███▍      | 2215/6470 [25:26<41:45,  1.70it/s]
4|2216|Loss: 0.8133284449577332:  34%|███▍      | 2216/6470 [25:27<40:57,  1.73it/s]
4|2217|Loss: 0.9716269969940186:  34%|███▍      | 2216/6470 [25:27<40:57,  1.73it/s]
4|2217|Loss: 0.9716269969940186:  34%|███▍      | 2217/6470 [25:27<44:55,  1.58it/s]
4|2218|Loss: 0.8058330416679382:  34%|███▍      | 2217/6470 [25:28<44:55,  1.58it/s]
4|2218|Loss: 0.8058330416679382:  34%|███▍      | 2218/6470 [25:28<46:57,  1.51it/s]
4|2219|Loss: 0.8404097557067871:  34%|███▍      | 2218/6470 [25:29<46:57,  1.51it/s]
4|2219|Loss: 0.8404097557067871:  34%|███▍      | 2219/6470 [25:29<50:24,  1.41it/s]
4|2220|Loss: 0.8261759281158447:  34%|███▍      | 2219/6470 [25:29<50:24,  1.41it/s]
4|2220|Loss: 0.8261759281158447:  34%|███▍      | 2220/6470 [25:30<53:35,  1.32it/s]
4|2221|Loss: 0.8712875247001648:  34%|███▍      | 2220/6470 [25:30<53:35,  1.32it/s]
4|2221|Loss: 0.8712875247001648:  34%|███▍      | 2221/6470 [25:31<53:40,  1.32it/s]
4|2222|Loss: 0.8612055778503418:  34%|███▍      | 2221/6470 [25:31<53:40,  1.32it/s]
4|2222|Loss: 0.8612055778503418:  34%|███▍      | 2222/6470 [25:31<51:41,  1.37it/s]
4|2223|Loss: 0.9806853532791138:  34%|███▍      | 2222/6470 [25:32<51:41,  1.37it/s]
4|2223|Loss: 0.9806853532791138:  34%|███▍      | 2223/6470 [25:32<53:23,  1.33it/s]
4|2224|Loss: 0.9452604651451111:  34%|███▍      | 2223/6470 [25:33<53:23,  1.33it/s]
4|2224|Loss: 0.9452604651451111:  34%|███▍      | 2224/6470 [25:33<54:40,  1.29it/s]
4|2225|Loss: 0.7445110082626343:  34%|███▍      | 2224/6470 [25:33<54:40,  1.29it/s]
4|2225|Loss: 0.7445110082626343:  34%|███▍      | 2225/6470 [25:34<54:32,  1.30it/s]
4|2226|Loss: 0.8281140327453613:  34%|███▍      | 2225/6470 [25:34<54:32,  1.30it/s]
4|2226|Loss: 0.8281140327453613:  34%|███▍      | 2226/6470 [25:34<54:54,  1.29it/s]
4|2227|Loss: 0.825366199016571:  34%|███▍      | 2226/6470 [25:35<54:54,  1.29it/s] 
4|2227|Loss: 0.825366199016571:  34%|███▍      | 2227/6470 [25:35<56:45,  1.25it/s]
4|2228|Loss: 0.6622498631477356:  34%|███▍      | 2227/6470 [25:36<56:45,  1.25it/s]
4|2228|Loss: 0.6622498631477356:  34%|███▍      | 2228/6470 [25:36<55:38,  1.27it/s]
4|2229|Loss: 0.8989021182060242:  34%|███▍      | 2228/6470 [25:36<55:38,  1.27it/s]
4|2229|Loss: 0.8989021182060242:  34%|███▍      | 2229/6470 [25:37<54:34,  1.30it/s]
4|2230|Loss: 0.8664821982383728:  34%|███▍      | 2229/6470 [25:37<54:34,  1.30it/s]
4|2230|Loss: 0.8664821982383728:  34%|███▍      | 2230/6470 [25:38<54:10,  1.30it/s]
4|2231|Loss: 0.7440911531448364:  34%|███▍      | 2230/6470 [25:38<54:10,  1.30it/s]
4|2231|Loss: 0.7440911531448364:  34%|███▍      | 2231/6470 [25:38<55:28,  1.27it/s]
4|2232|Loss: 0.8427796959877014:  34%|███▍      | 2231/6470 [25:39<55:28,  1.27it/s]
4|2232|Loss: 0.8427796959877014:  34%|███▍      | 2232/6470 [25:39<55:55,  1.26it/s]
4|2233|Loss: 0.8691596388816833:  34%|███▍      | 2232/6470 [25:40<55:55,  1.26it/s]
4|2233|Loss: 0.8691596388816833:  35%|███▍      | 2233/6470 [25:40<57:14,  1.23it/s]
4|2234|Loss: 0.8276323080062866:  35%|███▍      | 2233/6470 [25:41<57:14,  1.23it/s]
4|2234|Loss: 0.8276323080062866:  35%|███▍      | 2234/6470 [25:41<58:37,  1.20it/s]
4|2235|Loss: 0.871457040309906:  35%|███▍      | 2234/6470 [25:41<58:37,  1.20it/s] 
4|2235|Loss: 0.871457040309906:  35%|███▍      | 2235/6470 [25:42<55:42,  1.27it/s]
4|2236|Loss: 0.7905287742614746:  35%|███▍      | 2235/6470 [25:42<55:42,  1.27it/s]
4|2236|Loss: 0.7905287742614746:  35%|███▍      | 2236/6470 [25:42<52:32,  1.34it/s]
4|2237|Loss: 0.7648356556892395:  35%|███▍      | 2236/6470 [25:43<52:32,  1.34it/s]
4|2237|Loss: 0.7648356556892395:  35%|███▍      | 2237/6470 [25:43<51:48,  1.36it/s]
4|2238|Loss: 0.7437036633491516:  35%|███▍      | 2237/6470 [25:43<51:48,  1.36it/s]
4|2238|Loss: 0.7437036633491516:  35%|███▍      | 2238/6470 [25:44<52:06,  1.35it/s]
4|2239|Loss: 0.9164398908615112:  35%|███▍      | 2238/6470 [25:44<52:06,  1.35it/s]
4|2239|Loss: 0.9164398908615112:  35%|███▍      | 2239/6470 [25:44<50:42,  1.39it/s]
4|2240|Loss: 0.7640029191970825:  35%|███▍      | 2239/6470 [25:45<50:42,  1.39it/s]
4|2240|Loss: 0.7640029191970825:  35%|███▍      | 2240/6470 [25:45<52:23,  1.35it/s]
4|2241|Loss: 0.7573419213294983:  35%|███▍      | 2240/6470 [25:46<52:23,  1.35it/s]
4|2241|Loss: 0.7573419213294983:  35%|███▍      | 2241/6470 [25:46<48:58,  1.44it/s]
4|2242|Loss: 1.0127568244934082:  35%|███▍      | 2241/6470 [25:46<48:58,  1.44it/s]
4|2242|Loss: 1.0127568244934082:  35%|███▍      | 2242/6470 [25:46<47:16,  1.49it/s]
4|2243|Loss: 0.751832127571106:  35%|███▍      | 2242/6470 [25:47<47:16,  1.49it/s] 
4|2243|Loss: 0.751832127571106:  35%|███▍      | 2243/6470 [25:47<45:33,  1.55it/s]
4|2244|Loss: 0.7831716537475586:  35%|███▍      | 2243/6470 [25:47<45:33,  1.55it/s]
4|2244|Loss: 0.7831716537475586:  35%|███▍      | 2244/6470 [25:48<46:12,  1.52it/s]
4|2245|Loss: 1.0897530317306519:  35%|███▍      | 2244/6470 [25:48<46:12,  1.52it/s]
4|2245|Loss: 1.0897530317306519:  35%|███▍      | 2245/6470 [25:48<47:55,  1.47it/s]
4|2246|Loss: 0.7707498669624329:  35%|███▍      | 2245/6470 [25:49<47:55,  1.47it/s]
4|2246|Loss: 0.7707498669624329:  35%|███▍      | 2246/6470 [25:49<47:36,  1.48it/s]
4|2247|Loss: 0.8448951244354248:  35%|███▍      | 2246/6470 [25:49<47:36,  1.48it/s]
4|2247|Loss: 0.8448951244354248:  35%|███▍      | 2247/6470 [25:50<46:49,  1.50it/s]
4|2248|Loss: 0.8503365516662598:  35%|███▍      | 2247/6470 [25:50<46:49,  1.50it/s]
4|2248|Loss: 0.8503365516662598:  35%|███▍      | 2248/6470 [25:50<45:56,  1.53it/s]
4|2249|Loss: 0.8166772723197937:  35%|███▍      | 2248/6470 [25:51<45:56,  1.53it/s]
4|2249|Loss: 0.8166772723197937:  35%|███▍      | 2249/6470 [25:51<45:33,  1.54it/s]
4|2250|Loss: 0.5808057188987732:  35%|███▍      | 2249/6470 [25:51<45:33,  1.54it/s]
4|2250|Loss: 0.5808057188987732:  35%|███▍      | 2250/6470 [25:52<47:14,  1.49it/s]
4|2251|Loss: 0.8467090129852295:  35%|███▍      | 2250/6470 [25:52<47:14,  1.49it/s]
4|2251|Loss: 0.8467090129852295:  35%|███▍      | 2251/6470 [25:53<50:46,  1.39it/s]
4|2252|Loss: 0.7813425660133362:  35%|███▍      | 2251/6470 [25:53<50:46,  1.39it/s]
4|2252|Loss: 0.7813425660133362:  35%|███▍      | 2252/6470 [25:53<51:12,  1.37it/s]
4|2253|Loss: 0.9817906022071838:  35%|███▍      | 2252/6470 [25:54<51:12,  1.37it/s]
4|2253|Loss: 0.9817906022071838:  35%|███▍      | 2253/6470 [25:54<51:20,  1.37it/s]
4|2254|Loss: 1.0364537239074707:  35%|███▍      | 2253/6470 [25:54<51:20,  1.37it/s]
4|2254|Loss: 1.0364537239074707:  35%|███▍      | 2254/6470 [25:55<54:37,  1.29it/s]
4|2255|Loss: 0.6874307990074158:  35%|███▍      | 2254/6470 [25:55<54:37,  1.29it/s]
4|2255|Loss: 0.6874307990074158:  35%|███▍      | 2255/6470 [25:56<54:31,  1.29it/s]
4|2256|Loss: 0.9277668595314026:  35%|███▍      | 2255/6470 [25:56<54:31,  1.29it/s]
4|2256|Loss: 0.9277668595314026:  35%|███▍      | 2256/6470 [25:56<52:05,  1.35it/s]
4|2257|Loss: 0.7902965545654297:  35%|███▍      | 2256/6470 [25:57<52:05,  1.35it/s]
4|2257|Loss: 0.7902965545654297:  35%|███▍      | 2257/6470 [25:57<49:25,  1.42it/s]
4|2258|Loss: 0.7946399450302124:  35%|███▍      | 2257/6470 [25:57<49:25,  1.42it/s]
4|2258|Loss: 0.7946399450302124:  35%|███▍      | 2258/6470 [25:58<49:03,  1.43it/s]
4|2259|Loss: 0.8584116697311401:  35%|███▍      | 2258/6470 [25:58<49:03,  1.43it/s]
4|2259|Loss: 0.8584116697311401:  35%|███▍      | 2259/6470 [25:58<51:44,  1.36it/s]
4|2260|Loss: 0.8640910387039185:  35%|███▍      | 2259/6470 [25:59<51:44,  1.36it/s]
4|2260|Loss: 0.8640910387039185:  35%|███▍      | 2260/6470 [25:59<53:29,  1.31it/s]
4|2261|Loss: 0.7899385094642639:  35%|███▍      | 2260/6470 [26:00<53:29,  1.31it/s]
4|2261|Loss: 0.7899385094642639:  35%|███▍      | 2261/6470 [26:00<50:03,  1.40it/s]
4|2262|Loss: 0.8089903593063354:  35%|███▍      | 2261/6470 [26:00<50:03,  1.40it/s]
4|2262|Loss: 0.8089903593063354:  35%|███▍      | 2262/6470 [26:00<47:06,  1.49it/s]
4|2263|Loss: 1.08916437625885:  35%|███▍      | 2262/6470 [26:01<47:06,  1.49it/s]  
4|2263|Loss: 1.08916437625885:  35%|███▍      | 2263/6470 [26:01<44:03,  1.59it/s]
4|2264|Loss: 0.9297960996627808:  35%|███▍      | 2263/6470 [26:01<44:03,  1.59it/s]
4|2264|Loss: 0.9297960996627808:  35%|███▍      | 2264/6470 [26:01<41:43,  1.68it/s]
4|2265|Loss: 0.903470516204834:  35%|███▍      | 2264/6470 [26:02<41:43,  1.68it/s] 
4|2265|Loss: 0.903470516204834:  35%|███▌      | 2265/6470 [26:02<45:31,  1.54it/s]
4|2266|Loss: 0.9239636659622192:  35%|███▌      | 2265/6470 [26:03<45:31,  1.54it/s]
4|2266|Loss: 0.9239636659622192:  35%|███▌      | 2266/6470 [26:03<44:43,  1.57it/s]
4|2267|Loss: 0.9224288463592529:  35%|███▌      | 2266/6470 [26:03<44:43,  1.57it/s]
4|2267|Loss: 0.9224288463592529:  35%|███▌      | 2267/6470 [26:04<48:00,  1.46it/s]
4|2268|Loss: 0.7953397631645203:  35%|███▌      | 2267/6470 [26:04<48:00,  1.46it/s]
4|2268|Loss: 0.7953397631645203:  35%|███▌      | 2268/6470 [26:04<47:37,  1.47it/s]
4|2269|Loss: 0.8349854350090027:  35%|███▌      | 2268/6470 [26:05<47:37,  1.47it/s]
4|2269|Loss: 0.8349854350090027:  35%|███▌      | 2269/6470 [26:05<46:36,  1.50it/s]
4|2270|Loss: 0.822634756565094:  35%|███▌      | 2269/6470 [26:05<46:36,  1.50it/s] 
4|2270|Loss: 0.822634756565094:  35%|███▌      | 2270/6470 [26:06<45:01,  1.55it/s]
4|2271|Loss: 0.8336824774742126:  35%|███▌      | 2270/6470 [26:06<45:01,  1.55it/s]
4|2271|Loss: 0.8336824774742126:  35%|███▌      | 2271/6470 [26:06<47:55,  1.46it/s]
4|2272|Loss: 0.906268298625946:  35%|███▌      | 2271/6470 [26:07<47:55,  1.46it/s] 
4|2272|Loss: 0.906268298625946:  35%|███▌      | 2272/6470 [26:07<46:15,  1.51it/s]
4|2273|Loss: 0.8581667542457581:  35%|███▌      | 2272/6470 [26:07<46:15,  1.51it/s]
4|2273|Loss: 0.8581667542457581:  35%|███▌      | 2273/6470 [26:08<48:10,  1.45it/s]
4|2274|Loss: 0.7376345992088318:  35%|███▌      | 2273/6470 [26:08<48:10,  1.45it/s]
4|2274|Loss: 0.7376345992088318:  35%|███▌      | 2274/6470 [26:08<49:19,  1.42it/s]
4|2275|Loss: 0.9015800952911377:  35%|███▌      | 2274/6470 [26:09<49:19,  1.42it/s]
4|2275|Loss: 0.9015800952911377:  35%|███▌      | 2275/6470 [26:09<50:08,  1.39it/s]
4|2276|Loss: 0.6662036776542664:  35%|███▌      | 2275/6470 [26:09<50:08,  1.39it/s]
4|2276|Loss: 0.6662036776542664:  35%|███▌      | 2276/6470 [26:10<43:29,  1.61it/s]
4|2277|Loss: 0.9091639518737793:  35%|███▌      | 2276/6470 [26:10<43:29,  1.61it/s]
4|2277|Loss: 0.9091639518737793:  35%|███▌      | 2277/6470 [26:10<42:18,  1.65it/s]
4|2278|Loss: 0.855488657951355:  35%|███▌      | 2277/6470 [26:11<42:18,  1.65it/s] 
4|2278|Loss: 0.855488657951355:  35%|███▌      | 2278/6470 [26:11<47:51,  1.46it/s]
4|2279|Loss: 0.7737156748771667:  35%|███▌      | 2278/6470 [26:11<47:51,  1.46it/s]
4|2279|Loss: 0.7737156748771667:  35%|███▌      | 2279/6470 [26:12<48:17,  1.45it/s]
4|2280|Loss: 0.9022200703620911:  35%|███▌      | 2279/6470 [26:12<48:17,  1.45it/s]
4|2280|Loss: 0.9022200703620911:  35%|███▌      | 2280/6470 [26:12<48:46,  1.43it/s]
4|2281|Loss: 0.9171152710914612:  35%|███▌      | 2280/6470 [26:13<48:46,  1.43it/s]
4|2281|Loss: 0.9171152710914612:  35%|███▌      | 2281/6470 [26:13<47:57,  1.46it/s]
4|2282|Loss: 0.8268243074417114:  35%|███▌      | 2281/6470 [26:13<47:57,  1.46it/s]
4|2282|Loss: 0.8268243074417114:  35%|███▌      | 2282/6470 [26:14<48:06,  1.45it/s]
4|2283|Loss: 0.8118753433227539:  35%|███▌      | 2282/6470 [26:14<48:06,  1.45it/s]
4|2283|Loss: 0.8118753433227539:  35%|███▌      | 2283/6470 [26:15<50:11,  1.39it/s]
4|2284|Loss: 0.8531127572059631:  35%|███▌      | 2283/6470 [26:15<50:11,  1.39it/s]
4|2284|Loss: 0.8531127572059631:  35%|███▌      | 2284/6470 [26:15<48:59,  1.42it/s]
4|2285|Loss: 0.9920998811721802:  35%|███▌      | 2284/6470 [26:16<48:59,  1.42it/s]
4|2285|Loss: 0.9920998811721802:  35%|███▌      | 2285/6470 [26:16<51:33,  1.35it/s]
4|2286|Loss: 0.9506955742835999:  35%|███▌      | 2285/6470 [26:16<51:33,  1.35it/s]
4|2286|Loss: 0.9506955742835999:  35%|███▌      | 2286/6470 [26:17<50:23,  1.38it/s]
4|2287|Loss: 0.7002646327018738:  35%|███▌      | 2286/6470 [26:17<50:23,  1.38it/s]
4|2287|Loss: 0.7002646327018738:  35%|███▌      | 2287/6470 [26:17<49:47,  1.40it/s]
4|2288|Loss: 0.7217997312545776:  35%|███▌      | 2287/6470 [26:18<49:47,  1.40it/s]
4|2288|Loss: 0.7217997312545776:  35%|███▌      | 2288/6470 [26:18<48:27,  1.44it/s]
4|2289|Loss: 0.7325387597084045:  35%|███▌      | 2288/6470 [26:19<48:27,  1.44it/s]
4|2289|Loss: 0.7325387597084045:  35%|███▌      | 2289/6470 [26:19<49:55,  1.40it/s]
4|2290|Loss: 0.8526397943496704:  35%|███▌      | 2289/6470 [26:19<49:55,  1.40it/s]
4|2290|Loss: 0.8526397943496704:  35%|███▌      | 2290/6470 [26:20<50:36,  1.38it/s]
4|2291|Loss: 0.8383030891418457:  35%|███▌      | 2290/6470 [26:20<50:36,  1.38it/s]
4|2291|Loss: 0.8383030891418457:  35%|███▌      | 2291/6470 [26:20<52:45,  1.32it/s]
4|2292|Loss: 0.6942654252052307:  35%|███▌      | 2291/6470 [26:21<52:45,  1.32it/s]
4|2292|Loss: 0.6942654252052307:  35%|███▌      | 2292/6470 [26:21<53:47,  1.29it/s]
4|2293|Loss: 0.9251349568367004:  35%|███▌      | 2292/6470 [26:22<53:47,  1.29it/s]
4|2293|Loss: 0.9251349568367004:  35%|███▌      | 2293/6470 [26:22<52:37,  1.32it/s]
4|2294|Loss: 0.9140558242797852:  35%|███▌      | 2293/6470 [26:22<52:37,  1.32it/s]
4|2294|Loss: 0.9140558242797852:  35%|███▌      | 2294/6470 [26:23<49:20,  1.41it/s]
4|2295|Loss: 0.9978214502334595:  35%|███▌      | 2294/6470 [26:23<49:20,  1.41it/s]
4|2295|Loss: 0.9978214502334595:  35%|███▌      | 2295/6470 [26:23<47:12,  1.47it/s]
4|2296|Loss: 0.9178362488746643:  35%|███▌      | 2295/6470 [26:24<47:12,  1.47it/s]
4|2296|Loss: 0.9178362488746643:  35%|███▌      | 2296/6470 [26:24<46:28,  1.50it/s]
4|2297|Loss: 0.7553878426551819:  35%|███▌      | 2296/6470 [26:24<46:28,  1.50it/s]
4|2297|Loss: 0.7553878426551819:  36%|███▌      | 2297/6470 [26:24<44:23,  1.57it/s]
4|2298|Loss: 0.7822608947753906:  36%|███▌      | 2297/6470 [26:25<44:23,  1.57it/s]
4|2298|Loss: 0.7822608947753906:  36%|███▌      | 2298/6470 [26:25<47:16,  1.47it/s]
4|2299|Loss: 0.9886520504951477:  36%|███▌      | 2298/6470 [26:26<47:16,  1.47it/s]
4|2299|Loss: 0.9886520504951477:  36%|███▌      | 2299/6470 [26:26<46:25,  1.50it/s]
4|2300|Loss: 0.860092282295227:  36%|███▌      | 2299/6470 [26:26<46:25,  1.50it/s] 
4|2300|Loss: 0.860092282295227:  36%|███▌      | 2300/6470 [26:26<44:11,  1.57it/s]
4|2301|Loss: 0.7819538116455078:  36%|███▌      | 2300/6470 [26:27<44:11,  1.57it/s]
4|2301|Loss: 0.7819538116455078:  36%|███▌      | 2301/6470 [26:27<43:09,  1.61it/s]
4|2302|Loss: 0.7251710295677185:  36%|███▌      | 2301/6470 [26:27<43:09,  1.61it/s]
4|2302|Loss: 0.7251710295677185:  36%|███▌      | 2302/6470 [26:28<46:30,  1.49it/s]
4|2303|Loss: 0.740121066570282:  36%|███▌      | 2302/6470 [26:28<46:30,  1.49it/s] 
4|2303|Loss: 0.740121066570282:  36%|███▌      | 2303/6470 [26:28<44:50,  1.55it/s]
4|2304|Loss: 1.1120884418487549:  36%|███▌      | 2303/6470 [26:29<44:50,  1.55it/s]
4|2304|Loss: 1.1120884418487549:  36%|███▌      | 2304/6470 [26:29<45:43,  1.52it/s]
4|2305|Loss: 0.8297561407089233:  36%|███▌      | 2304/6470 [26:29<45:43,  1.52it/s]
4|2305|Loss: 0.8297561407089233:  36%|███▌      | 2305/6470 [26:30<45:26,  1.53it/s]
4|2306|Loss: 0.9640076756477356:  36%|███▌      | 2305/6470 [26:30<45:26,  1.53it/s]
4|2306|Loss: 0.9640076756477356:  36%|███▌      | 2306/6470 [26:30<43:27,  1.60it/s]
4|2307|Loss: 0.8097771406173706:  36%|███▌      | 2306/6470 [26:31<43:27,  1.60it/s]
4|2307|Loss: 0.8097771406173706:  36%|███▌      | 2307/6470 [26:31<45:00,  1.54it/s]
4|2308|Loss: 0.8063058853149414:  36%|███▌      | 2307/6470 [26:31<45:00,  1.54it/s]
4|2308|Loss: 0.8063058853149414:  36%|███▌      | 2308/6470 [26:32<45:21,  1.53it/s]
4|2309|Loss: 0.906632125377655:  36%|███▌      | 2308/6470 [26:32<45:21,  1.53it/s] 
4|2309|Loss: 0.906632125377655:  36%|███▌      | 2309/6470 [26:32<41:59,  1.65it/s]
4|2310|Loss: 0.7502802014350891:  36%|███▌      | 2309/6470 [26:32<41:59,  1.65it/s]
4|2310|Loss: 0.7502802014350891:  36%|███▌      | 2310/6470 [26:33<41:50,  1.66it/s]
4|2311|Loss: 0.8276744484901428:  36%|███▌      | 2310/6470 [26:33<41:50,  1.66it/s]
4|2311|Loss: 0.8276744484901428:  36%|███▌      | 2311/6470 [26:33<44:17,  1.56it/s]
4|2312|Loss: 0.9247234463691711:  36%|███▌      | 2311/6470 [26:34<44:17,  1.56it/s]
4|2312|Loss: 0.9247234463691711:  36%|███▌      | 2312/6470 [26:34<48:36,  1.43it/s]
4|2313|Loss: 0.83284592628479:  36%|███▌      | 2312/6470 [26:35<48:36,  1.43it/s]  
4|2313|Loss: 0.83284592628479:  36%|███▌      | 2313/6470 [26:35<47:43,  1.45it/s]
4|2314|Loss: 1.147347331047058:  36%|███▌      | 2313/6470 [26:35<47:43,  1.45it/s]
4|2314|Loss: 1.147347331047058:  36%|███▌      | 2314/6470 [26:36<51:48,  1.34it/s]
4|2315|Loss: 0.9151080846786499:  36%|███▌      | 2314/6470 [26:36<51:48,  1.34it/s]
4|2315|Loss: 0.9151080846786499:  36%|███▌      | 2315/6470 [26:36<50:36,  1.37it/s]
4|2316|Loss: 0.7259162664413452:  36%|███▌      | 2315/6470 [26:37<50:36,  1.37it/s]
4|2316|Loss: 0.7259162664413452:  36%|███▌      | 2316/6470 [26:37<50:02,  1.38it/s]
4|2317|Loss: 0.7900801301002502:  36%|███▌      | 2316/6470 [26:38<50:02,  1.38it/s]
4|2317|Loss: 0.7900801301002502:  36%|███▌      | 2317/6470 [26:38<48:40,  1.42it/s]
4|2318|Loss: 0.9213837385177612:  36%|███▌      | 2317/6470 [26:38<48:40,  1.42it/s]
4|2318|Loss: 0.9213837385177612:  36%|███▌      | 2318/6470 [26:39<50:31,  1.37it/s]
4|2319|Loss: 0.8131837248802185:  36%|███▌      | 2318/6470 [26:39<50:31,  1.37it/s]
4|2319|Loss: 0.8131837248802185:  36%|███▌      | 2319/6470 [26:39<49:34,  1.40it/s]
4|2320|Loss: 0.7328616976737976:  36%|███▌      | 2319/6470 [26:40<49:34,  1.40it/s]
4|2320|Loss: 0.7328616976737976:  36%|███▌      | 2320/6470 [26:40<52:36,  1.31it/s]
4|2321|Loss: 0.8421913981437683:  36%|███▌      | 2320/6470 [26:41<52:36,  1.31it/s]
4|2321|Loss: 0.8421913981437683:  36%|███▌      | 2321/6470 [26:41<52:50,  1.31it/s]
4|2322|Loss: 0.8626614809036255:  36%|███▌      | 2321/6470 [26:41<52:50,  1.31it/s]
4|2322|Loss: 0.8626614809036255:  36%|███▌      | 2322/6470 [26:42<53:27,  1.29it/s]
4|2323|Loss: 0.9139224886894226:  36%|███▌      | 2322/6470 [26:42<53:27,  1.29it/s]
4|2323|Loss: 0.9139224886894226:  36%|███▌      | 2323/6470 [26:42<52:25,  1.32it/s]
4|2324|Loss: 0.8695687055587769:  36%|███▌      | 2323/6470 [26:43<52:25,  1.32it/s]
4|2324|Loss: 0.8695687055587769:  36%|███▌      | 2324/6470 [26:43<49:50,  1.39it/s]
4|2325|Loss: 0.8630300760269165:  36%|███▌      | 2324/6470 [26:43<49:50,  1.39it/s]
4|2325|Loss: 0.8630300760269165:  36%|███▌      | 2325/6470 [26:44<49:12,  1.40it/s]
4|2326|Loss: 0.879313588142395:  36%|███▌      | 2325/6470 [26:44<49:12,  1.40it/s] 
4|2326|Loss: 0.879313588142395:  36%|███▌      | 2326/6470 [26:45<49:34,  1.39it/s]
4|2327|Loss: 0.958696722984314:  36%|███▌      | 2326/6470 [26:45<49:34,  1.39it/s]
4|2327|Loss: 0.958696722984314:  36%|███▌      | 2327/6470 [26:45<48:08,  1.43it/s]
4|2328|Loss: 0.8307132720947266:  36%|███▌      | 2327/6470 [26:46<48:08,  1.43it/s]
4|2328|Loss: 0.8307132720947266:  36%|███▌      | 2328/6470 [26:46<49:28,  1.40it/s]
4|2329|Loss: 0.785518229007721:  36%|███▌      | 2328/6470 [26:46<49:28,  1.40it/s] 
4|2329|Loss: 0.785518229007721:  36%|███▌      | 2329/6470 [26:47<46:03,  1.50it/s]
4|2330|Loss: 0.9953432679176331:  36%|███▌      | 2329/6470 [26:47<46:03,  1.50it/s]
4|2330|Loss: 0.9953432679176331:  36%|███▌      | 2330/6470 [26:47<46:16,  1.49it/s]
4|2331|Loss: 0.9330703020095825:  36%|███▌      | 2330/6470 [26:48<46:16,  1.49it/s]
4|2331|Loss: 0.9330703020095825:  36%|███▌      | 2331/6470 [26:48<47:57,  1.44it/s]
4|2332|Loss: 0.8462033867835999:  36%|███▌      | 2331/6470 [26:48<47:57,  1.44it/s]
4|2332|Loss: 0.8462033867835999:  36%|███▌      | 2332/6470 [26:48<44:10,  1.56it/s]
4|2333|Loss: 0.8419988751411438:  36%|███▌      | 2332/6470 [26:49<44:10,  1.56it/s]
4|2333|Loss: 0.8419988751411438:  36%|███▌      | 2333/6470 [26:49<42:57,  1.61it/s]
4|2334|Loss: 0.8523630499839783:  36%|███▌      | 2333/6470 [26:49<42:57,  1.61it/s]
4|2334|Loss: 0.8523630499839783:  36%|███▌      | 2334/6470 [26:50<44:36,  1.55it/s]
4|2335|Loss: 0.890676736831665:  36%|███▌      | 2334/6470 [26:50<44:36,  1.55it/s] 
4|2335|Loss: 0.890676736831665:  36%|███▌      | 2335/6470 [26:51<48:44,  1.41it/s]
4|2336|Loss: 0.9667097926139832:  36%|███▌      | 2335/6470 [26:51<48:44,  1.41it/s]
4|2336|Loss: 0.9667097926139832:  36%|███▌      | 2336/6470 [26:51<47:32,  1.45it/s]
4|2337|Loss: 0.8645651340484619:  36%|███▌      | 2336/6470 [26:52<47:32,  1.45it/s]
4|2337|Loss: 0.8645651340484619:  36%|███▌      | 2337/6470 [26:52<44:46,  1.54it/s]
4|2338|Loss: 0.7298940420150757:  36%|███▌      | 2337/6470 [26:52<44:46,  1.54it/s]
4|2338|Loss: 0.7298940420150757:  36%|███▌      | 2338/6470 [26:52<45:06,  1.53it/s]
4|2339|Loss: 1.0246927738189697:  36%|███▌      | 2338/6470 [26:53<45:06,  1.53it/s]
4|2339|Loss: 1.0246927738189697:  36%|███▌      | 2339/6470 [26:53<42:53,  1.61it/s]
4|2340|Loss: 0.9145281910896301:  36%|███▌      | 2339/6470 [26:53<42:53,  1.61it/s]
4|2340|Loss: 0.9145281910896301:  36%|███▌      | 2340/6470 [26:54<43:47,  1.57it/s]
4|2341|Loss: 0.9735265970230103:  36%|███▌      | 2340/6470 [26:54<43:47,  1.57it/s]
4|2341|Loss: 0.9735265970230103:  36%|███▌      | 2341/6470 [26:55<48:06,  1.43it/s]
4|2342|Loss: 0.9068770408630371:  36%|███▌      | 2341/6470 [26:55<48:06,  1.43it/s]
4|2342|Loss: 0.9068770408630371:  36%|███▌      | 2342/6470 [26:55<50:32,  1.36it/s]
4|2343|Loss: 0.9344233274459839:  36%|███▌      | 2342/6470 [26:56<50:32,  1.36it/s]
4|2343|Loss: 0.9344233274459839:  36%|███▌      | 2343/6470 [26:56<47:29,  1.45it/s]
4|2344|Loss: 0.7163518071174622:  36%|███▌      | 2343/6470 [26:56<47:29,  1.45it/s]
4|2344|Loss: 0.7163518071174622:  36%|███▌      | 2344/6470 [26:56<43:45,  1.57it/s]
4|2345|Loss: 1.0300170183181763:  36%|███▌      | 2344/6470 [26:57<43:45,  1.57it/s]
4|2345|Loss: 1.0300170183181763:  36%|███▌      | 2345/6470 [26:57<45:02,  1.53it/s]
4|2346|Loss: 0.8090922832489014:  36%|███▌      | 2345/6470 [26:58<45:02,  1.53it/s]
4|2346|Loss: 0.8090922832489014:  36%|███▋      | 2346/6470 [26:58<46:26,  1.48it/s]
4|2347|Loss: 0.8356795310974121:  36%|███▋      | 2346/6470 [26:58<46:26,  1.48it/s]
4|2347|Loss: 0.8356795310974121:  36%|███▋      | 2347/6470 [26:59<47:33,  1.44it/s]
4|2348|Loss: 0.8083667755126953:  36%|███▋      | 2347/6470 [26:59<47:33,  1.44it/s]
4|2348|Loss: 0.8083667755126953:  36%|███▋      | 2348/6470 [26:59<48:02,  1.43it/s]
4|2349|Loss: 0.8323621153831482:  36%|███▋      | 2348/6470 [27:00<48:02,  1.43it/s]
4|2349|Loss: 0.8323621153831482:  36%|███▋      | 2349/6470 [27:00<48:08,  1.43it/s]
4|2350|Loss: 0.8520404696464539:  36%|███▋      | 2349/6470 [27:00<48:08,  1.43it/s]
4|2350|Loss: 0.8520404696464539:  36%|███▋      | 2350/6470 [27:01<51:12,  1.34it/s]
4|2351|Loss: 0.9059495329856873:  36%|███▋      | 2350/6470 [27:01<51:12,  1.34it/s]
4|2351|Loss: 0.9059495329856873:  36%|███▋      | 2351/6470 [27:02<50:16,  1.37it/s]
4|2352|Loss: 0.8276561498641968:  36%|███▋      | 2351/6470 [27:02<50:16,  1.37it/s]
4|2352|Loss: 0.8276561498641968:  36%|███▋      | 2352/6470 [27:02<51:57,  1.32it/s]
4|2353|Loss: 0.8449144959449768:  36%|███▋      | 2352/6470 [27:03<51:57,  1.32it/s]
4|2353|Loss: 0.8449144959449768:  36%|███▋      | 2353/6470 [27:03<50:36,  1.36it/s]
4|2354|Loss: 0.9793384075164795:  36%|███▋      | 2353/6470 [27:03<50:36,  1.36it/s]
4|2354|Loss: 0.9793384075164795:  36%|███▋      | 2354/6470 [27:04<50:19,  1.36it/s]
4|2355|Loss: 0.911169707775116:  36%|███▋      | 2354/6470 [27:04<50:19,  1.36it/s] 
4|2355|Loss: 0.911169707775116:  36%|███▋      | 2355/6470 [27:04<46:48,  1.47it/s]
4|2356|Loss: 0.7813147306442261:  36%|███▋      | 2355/6470 [27:05<46:48,  1.47it/s]
4|2356|Loss: 0.7813147306442261:  36%|███▋      | 2356/6470 [27:05<44:59,  1.52it/s]
4|2357|Loss: 0.8265165090560913:  36%|███▋      | 2356/6470 [27:05<44:59,  1.52it/s]
4|2357|Loss: 0.8265165090560913:  36%|███▋      | 2357/6470 [27:06<46:06,  1.49it/s]
4|2358|Loss: 0.8138137459754944:  36%|███▋      | 2357/6470 [27:06<46:06,  1.49it/s]
4|2358|Loss: 0.8138137459754944:  36%|███▋      | 2358/6470 [27:06<44:41,  1.53it/s]
4|2359|Loss: 0.739495575428009:  36%|███▋      | 2358/6470 [27:07<44:41,  1.53it/s] 
4|2359|Loss: 0.739495575428009:  36%|███▋      | 2359/6470 [27:07<41:07,  1.67it/s]
4|2360|Loss: 0.9194206595420837:  36%|███▋      | 2359/6470 [27:07<41:07,  1.67it/s]
4|2360|Loss: 0.9194206595420837:  36%|███▋      | 2360/6470 [27:07<42:13,  1.62it/s]
4|2361|Loss: 0.8563571572303772:  36%|███▋      | 2360/6470 [27:08<42:13,  1.62it/s]
4|2361|Loss: 0.8563571572303772:  36%|███▋      | 2361/6470 [27:08<45:48,  1.49it/s]
4|2362|Loss: 0.7958081960678101:  36%|███▋      | 2361/6470 [27:09<45:48,  1.49it/s]
4|2362|Loss: 0.7958081960678101:  37%|███▋      | 2362/6470 [27:09<46:03,  1.49it/s]
4|2363|Loss: 0.8875599503517151:  37%|███▋      | 2362/6470 [27:09<46:03,  1.49it/s]
4|2363|Loss: 0.8875599503517151:  37%|███▋      | 2363/6470 [27:10<49:51,  1.37it/s]
4|2364|Loss: 0.9013055562973022:  37%|███▋      | 2363/6470 [27:10<49:51,  1.37it/s]
4|2364|Loss: 0.9013055562973022:  37%|███▋      | 2364/6470 [27:11<51:08,  1.34it/s]
4|2365|Loss: 0.7308958172798157:  37%|███▋      | 2364/6470 [27:11<51:08,  1.34it/s]
4|2365|Loss: 0.7308958172798157:  37%|███▋      | 2365/6470 [27:11<48:04,  1.42it/s]
4|2366|Loss: 0.7934659123420715:  37%|███▋      | 2365/6470 [27:11<48:04,  1.42it/s]
4|2366|Loss: 0.7934659123420715:  37%|███▋      | 2366/6470 [27:12<46:03,  1.49it/s]
4|2367|Loss: 0.8665151596069336:  37%|███▋      | 2366/6470 [27:12<46:03,  1.49it/s]
4|2367|Loss: 0.8665151596069336:  37%|███▋      | 2367/6470 [27:12<42:16,  1.62it/s]
4|2368|Loss: 1.0191106796264648:  37%|███▋      | 2367/6470 [27:13<42:16,  1.62it/s]
4|2368|Loss: 1.0191106796264648:  37%|███▋      | 2368/6470 [27:13<45:26,  1.50it/s]
4|2369|Loss: 0.8188210725784302:  37%|███▋      | 2368/6470 [27:13<45:26,  1.50it/s]
4|2369|Loss: 0.8188210725784302:  37%|███▋      | 2369/6470 [27:14<46:04,  1.48it/s]
4|2370|Loss: 0.8933667540550232:  37%|███▋      | 2369/6470 [27:14<46:04,  1.48it/s]
4|2370|Loss: 0.8933667540550232:  37%|███▋      | 2370/6470 [27:14<44:15,  1.54it/s]
4|2371|Loss: 0.8523345589637756:  37%|███▋      | 2370/6470 [27:15<44:15,  1.54it/s]
4|2371|Loss: 0.8523345589637756:  37%|███▋      | 2371/6470 [27:15<45:19,  1.51it/s]
4|2372|Loss: 0.8636322617530823:  37%|███▋      | 2371/6470 [27:15<45:19,  1.51it/s]
4|2372|Loss: 0.8636322617530823:  37%|███▋      | 2372/6470 [27:16<46:21,  1.47it/s]
4|2373|Loss: 0.6814942955970764:  37%|███▋      | 2372/6470 [27:16<46:21,  1.47it/s]
4|2373|Loss: 0.6814942955970764:  37%|███▋      | 2373/6470 [27:16<45:55,  1.49it/s]
4|2374|Loss: 0.8249413371086121:  37%|███▋      | 2373/6470 [27:17<45:55,  1.49it/s]
4|2374|Loss: 0.8249413371086121:  37%|███▋      | 2374/6470 [27:17<46:50,  1.46it/s]
4|2375|Loss: 0.6549088954925537:  37%|███▋      | 2374/6470 [27:17<46:50,  1.46it/s]
4|2375|Loss: 0.6549088954925537:  37%|███▋      | 2375/6470 [27:18<47:49,  1.43it/s]
4|2376|Loss: 0.8875394463539124:  37%|███▋      | 2375/6470 [27:18<47:49,  1.43it/s]
4|2376|Loss: 0.8875394463539124:  37%|███▋      | 2376/6470 [27:19<48:27,  1.41it/s]
4|2377|Loss: 0.8613852262496948:  37%|███▋      | 2376/6470 [27:19<48:27,  1.41it/s]
4|2377|Loss: 0.8613852262496948:  37%|███▋      | 2377/6470 [27:19<45:55,  1.49it/s]
4|2378|Loss: 0.8273707032203674:  37%|███▋      | 2377/6470 [27:19<45:55,  1.49it/s]
4|2378|Loss: 0.8273707032203674:  37%|███▋      | 2378/6470 [27:20<44:55,  1.52it/s]
4|2379|Loss: 0.7294257879257202:  37%|███▋      | 2378/6470 [27:20<44:55,  1.52it/s]
4|2379|Loss: 0.7294257879257202:  37%|███▋      | 2379/6470 [27:21<49:47,  1.37it/s]
4|2380|Loss: 0.7695900201797485:  37%|███▋      | 2379/6470 [27:21<49:47,  1.37it/s]
4|2380|Loss: 0.7695900201797485:  37%|███▋      | 2380/6470 [27:21<48:13,  1.41it/s]
4|2381|Loss: 0.9606547355651855:  37%|███▋      | 2380/6470 [27:22<48:13,  1.41it/s]
4|2381|Loss: 0.9606547355651855:  37%|███▋      | 2381/6470 [27:22<44:26,  1.53it/s]
4|2382|Loss: 0.7747709155082703:  37%|███▋      | 2381/6470 [27:22<44:26,  1.53it/s]
4|2382|Loss: 0.7747709155082703:  37%|███▋      | 2382/6470 [27:23<46:22,  1.47it/s]
4|2383|Loss: 0.9067155122756958:  37%|███▋      | 2382/6470 [27:23<46:22,  1.47it/s]
4|2383|Loss: 0.9067155122756958:  37%|███▋      | 2383/6470 [27:23<47:53,  1.42it/s]
4|2384|Loss: 0.865463137626648:  37%|███▋      | 2383/6470 [27:24<47:53,  1.42it/s] 
4|2384|Loss: 0.865463137626648:  37%|███▋      | 2384/6470 [27:24<45:50,  1.49it/s]
4|2385|Loss: 0.7764427065849304:  37%|███▋      | 2384/6470 [27:24<45:50,  1.49it/s]
4|2385|Loss: 0.7764427065849304:  37%|███▋      | 2385/6470 [27:25<44:28,  1.53it/s]
4|2386|Loss: 0.8540049195289612:  37%|███▋      | 2385/6470 [27:25<44:28,  1.53it/s]
4|2386|Loss: 0.8540049195289612:  37%|███▋      | 2386/6470 [27:25<41:58,  1.62it/s]
4|2387|Loss: 0.904803991317749:  37%|███▋      | 2386/6470 [27:25<41:58,  1.62it/s] 
4|2387|Loss: 0.904803991317749:  37%|███▋      | 2387/6470 [27:26<38:58,  1.75it/s]
4|2388|Loss: 0.8791874051094055:  37%|███▋      | 2387/6470 [27:26<38:58,  1.75it/s]
4|2388|Loss: 0.8791874051094055:  37%|███▋      | 2388/6470 [27:26<40:58,  1.66it/s]
4|2389|Loss: 0.7717767357826233:  37%|███▋      | 2388/6470 [27:27<40:58,  1.66it/s]
4|2389|Loss: 0.7717767357826233:  37%|███▋      | 2389/6470 [27:27<42:20,  1.61it/s]
4|2390|Loss: 0.8786667585372925:  37%|███▋      | 2389/6470 [27:27<42:20,  1.61it/s]
4|2390|Loss: 0.8786667585372925:  37%|███▋      | 2390/6470 [27:28<45:49,  1.48it/s]
4|2391|Loss: 0.709302544593811:  37%|███▋      | 2390/6470 [27:28<45:49,  1.48it/s] 
4|2391|Loss: 0.709302544593811:  37%|███▋      | 2391/6470 [27:28<46:58,  1.45it/s]
4|2392|Loss: 0.8592853546142578:  37%|███▋      | 2391/6470 [27:29<46:58,  1.45it/s]
4|2392|Loss: 0.8592853546142578:  37%|███▋      | 2392/6470 [27:29<47:44,  1.42it/s]
4|2393|Loss: 0.8094492554664612:  37%|███▋      | 2392/6470 [27:30<47:44,  1.42it/s]
4|2393|Loss: 0.8094492554664612:  37%|███▋      | 2393/6470 [27:30<46:57,  1.45it/s]
4|2394|Loss: 0.7087453603744507:  37%|███▋      | 2393/6470 [27:30<46:57,  1.45it/s]
4|2394|Loss: 0.7087453603744507:  37%|███▋      | 2394/6470 [27:30<45:32,  1.49it/s]
4|2395|Loss: 0.797099769115448:  37%|███▋      | 2394/6470 [27:31<45:32,  1.49it/s] 
4|2395|Loss: 0.797099769115448:  37%|███▋      | 2395/6470 [27:31<46:13,  1.47it/s]
4|2396|Loss: 0.923176646232605:  37%|███▋      | 2395/6470 [27:31<46:13,  1.47it/s]
4|2396|Loss: 0.923176646232605:  37%|███▋      | 2396/6470 [27:32<44:43,  1.52it/s]
4|2397|Loss: 0.9863001108169556:  37%|███▋      | 2396/6470 [27:32<44:43,  1.52it/s]
4|2397|Loss: 0.9863001108169556:  37%|███▋      | 2397/6470 [27:32<43:05,  1.58it/s]
4|2398|Loss: 0.7144279479980469:  37%|███▋      | 2397/6470 [27:33<43:05,  1.58it/s]
4|2398|Loss: 0.7144279479980469:  37%|███▋      | 2398/6470 [27:33<42:38,  1.59it/s]
4|2399|Loss: 0.7409393787384033:  37%|███▋      | 2398/6470 [27:33<42:38,  1.59it/s]
4|2399|Loss: 0.7409393787384033:  37%|███▋      | 2399/6470 [27:34<45:23,  1.49it/s]
4|2400|Loss: 0.7964674830436707:  37%|███▋      | 2399/6470 [27:34<45:23,  1.49it/s]
4|2400|Loss: 0.7964674830436707:  37%|███▋      | 2400/6470 [27:34<48:14,  1.41it/s]
4|2401|Loss: 0.973440945148468:  37%|███▋      | 2400/6470 [27:35<48:14,  1.41it/s] 
4|2401|Loss: 0.973440945148468:  37%|███▋      | 2401/6470 [27:35<50:45,  1.34it/s]
4|2402|Loss: 0.852741539478302:  37%|███▋      | 2401/6470 [27:36<50:45,  1.34it/s]
4|2402|Loss: 0.852741539478302:  37%|███▋      | 2402/6470 [27:36<50:58,  1.33it/s]
4|2403|Loss: 0.8546597361564636:  37%|███▋      | 2402/6470 [27:36<50:58,  1.33it/s]
4|2403|Loss: 0.8546597361564636:  37%|███▋      | 2403/6470 [27:37<48:51,  1.39it/s]
4|2404|Loss: 0.8904701471328735:  37%|███▋      | 2403/6470 [27:37<48:51,  1.39it/s]
4|2404|Loss: 0.8904701471328735:  37%|███▋      | 2404/6470 [27:38<50:55,  1.33it/s]
4|2405|Loss: 0.9662697315216064:  37%|███▋      | 2404/6470 [27:38<50:55,  1.33it/s]
4|2405|Loss: 0.9662697315216064:  37%|███▋      | 2405/6470 [27:38<51:07,  1.33it/s]
4|2406|Loss: 0.852314293384552:  37%|███▋      | 2405/6470 [27:39<51:07,  1.33it/s] 
4|2406|Loss: 0.852314293384552:  37%|███▋      | 2406/6470 [27:39<49:10,  1.38it/s]
4|2407|Loss: 1.0221850872039795:  37%|███▋      | 2406/6470 [27:39<49:10,  1.38it/s]
4|2407|Loss: 1.0221850872039795:  37%|███▋      | 2407/6470 [27:40<48:19,  1.40it/s]
4|2408|Loss: 0.9332008957862854:  37%|███▋      | 2407/6470 [27:40<48:19,  1.40it/s]
4|2408|Loss: 0.9332008957862854:  37%|███▋      | 2408/6470 [27:40<48:32,  1.39it/s]
4|2409|Loss: 0.858849287033081:  37%|███▋      | 2408/6470 [27:41<48:32,  1.39it/s] 
4|2409|Loss: 0.858849287033081:  37%|███▋      | 2409/6470 [27:41<47:01,  1.44it/s]
4|2410|Loss: 0.8033087253570557:  37%|███▋      | 2409/6470 [27:41<47:01,  1.44it/s]
4|2410|Loss: 0.8033087253570557:  37%|███▋      | 2410/6470 [27:42<48:58,  1.38it/s]
4|2411|Loss: 0.7845147848129272:  37%|███▋      | 2410/6470 [27:42<48:58,  1.38it/s]
4|2411|Loss: 0.7845147848129272:  37%|███▋      | 2411/6470 [27:43<49:53,  1.36it/s]
4|2412|Loss: 0.8311080932617188:  37%|███▋      | 2411/6470 [27:43<49:53,  1.36it/s]
4|2412|Loss: 0.8311080932617188:  37%|███▋      | 2412/6470 [27:43<51:08,  1.32it/s]
4|2413|Loss: 0.9705636501312256:  37%|███▋      | 2412/6470 [27:44<51:08,  1.32it/s]
4|2413|Loss: 0.9705636501312256:  37%|███▋      | 2413/6470 [27:44<52:55,  1.28it/s]
4|2414|Loss: 0.823206901550293:  37%|███▋      | 2413/6470 [27:45<52:55,  1.28it/s] 
4|2414|Loss: 0.823206901550293:  37%|███▋      | 2414/6470 [27:45<48:53,  1.38it/s]
4|2415|Loss: 0.7132694125175476:  37%|███▋      | 2414/6470 [27:45<48:53,  1.38it/s]
4|2415|Loss: 0.7132694125175476:  37%|███▋      | 2415/6470 [27:45<47:57,  1.41it/s]
4|2416|Loss: 0.8406519293785095:  37%|███▋      | 2415/6470 [27:46<47:57,  1.41it/s]
4|2416|Loss: 0.8406519293785095:  37%|███▋      | 2416/6470 [27:46<50:17,  1.34it/s]
4|2417|Loss: 0.8390610218048096:  37%|███▋      | 2416/6470 [27:47<50:17,  1.34it/s]
4|2417|Loss: 0.8390610218048096:  37%|███▋      | 2417/6470 [27:47<50:34,  1.34it/s]
4|2418|Loss: 0.7893118858337402:  37%|███▋      | 2417/6470 [27:47<50:34,  1.34it/s]
4|2418|Loss: 0.7893118858337402:  37%|███▋      | 2418/6470 [27:48<50:16,  1.34it/s]
4|2419|Loss: 0.9267103672027588:  37%|███▋      | 2418/6470 [27:48<50:16,  1.34it/s]
4|2419|Loss: 0.9267103672027588:  37%|███▋      | 2419/6470 [27:49<52:18,  1.29it/s]
4|2420|Loss: 0.7945143580436707:  37%|███▋      | 2419/6470 [27:49<52:18,  1.29it/s]
4|2420|Loss: 0.7945143580436707:  37%|███▋      | 2420/6470 [27:50<54:05,  1.25it/s]
4|2421|Loss: 0.7809000611305237:  37%|███▋      | 2420/6470 [27:50<54:05,  1.25it/s]
4|2421|Loss: 0.7809000611305237:  37%|███▋      | 2421/6470 [27:50<50:06,  1.35it/s]
4|2422|Loss: 0.7094658017158508:  37%|███▋      | 2421/6470 [27:51<50:06,  1.35it/s]
4|2422|Loss: 0.7094658017158508:  37%|███▋      | 2422/6470 [27:51<51:50,  1.30it/s]
4|2423|Loss: 0.8133271932601929:  37%|███▋      | 2422/6470 [27:51<51:50,  1.30it/s]
4|2423|Loss: 0.8133271932601929:  37%|███▋      | 2423/6470 [27:52<48:57,  1.38it/s]
4|2424|Loss: 0.8555712103843689:  37%|███▋      | 2423/6470 [27:52<48:57,  1.38it/s]
4|2424|Loss: 0.8555712103843689:  37%|███▋      | 2424/6470 [27:52<48:20,  1.39it/s]
4|2425|Loss: 0.8998197913169861:  37%|███▋      | 2424/6470 [27:53<48:20,  1.39it/s]
4|2425|Loss: 0.8998197913169861:  37%|███▋      | 2425/6470 [27:53<49:41,  1.36it/s]
4|2426|Loss: 0.9841943979263306:  37%|███▋      | 2425/6470 [27:54<49:41,  1.36it/s]
4|2426|Loss: 0.9841943979263306:  37%|███▋      | 2426/6470 [27:54<51:21,  1.31it/s]
4|2427|Loss: 0.8410402536392212:  37%|███▋      | 2426/6470 [27:54<51:21,  1.31it/s]
4|2427|Loss: 0.8410402536392212:  38%|███▊      | 2427/6470 [27:55<49:18,  1.37it/s]
4|2428|Loss: 0.860463559627533:  38%|███▊      | 2427/6470 [27:55<49:18,  1.37it/s] 
4|2428|Loss: 0.860463559627533:  38%|███▊      | 2428/6470 [27:55<49:56,  1.35it/s]
4|2429|Loss: 0.6829807758331299:  38%|███▊      | 2428/6470 [27:56<49:56,  1.35it/s]
4|2429|Loss: 0.6829807758331299:  38%|███▊      | 2429/6470 [27:56<48:38,  1.38it/s]
4|2430|Loss: 0.908462405204773:  38%|███▊      | 2429/6470 [27:56<48:38,  1.38it/s] 
4|2430|Loss: 0.908462405204773:  38%|███▊      | 2430/6470 [27:57<48:07,  1.40it/s]
4|2431|Loss: 0.7760615944862366:  38%|███▊      | 2430/6470 [27:57<48:07,  1.40it/s]
4|2431|Loss: 0.7760615944862366:  38%|███▊      | 2431/6470 [27:57<45:47,  1.47it/s]
4|2432|Loss: 0.8801712989807129:  38%|███▊      | 2431/6470 [27:58<45:47,  1.47it/s]
4|2432|Loss: 0.8801712989807129:  38%|███▊      | 2432/6470 [27:58<47:46,  1.41it/s]
4|2433|Loss: 0.9720852375030518:  38%|███▊      | 2432/6470 [27:59<47:46,  1.41it/s]
4|2433|Loss: 0.9720852375030518:  38%|███▊      | 2433/6470 [27:59<49:43,  1.35it/s]
4|2434|Loss: 0.8452842831611633:  38%|███▊      | 2433/6470 [27:59<49:43,  1.35it/s]
4|2434|Loss: 0.8452842831611633:  38%|███▊      | 2434/6470 [28:00<50:13,  1.34it/s]
4|2435|Loss: 0.9136643409729004:  38%|███▊      | 2434/6470 [28:00<50:13,  1.34it/s]
4|2435|Loss: 0.9136643409729004:  38%|███▊      | 2435/6470 [28:00<49:40,  1.35it/s]
4|2436|Loss: 0.8469223380088806:  38%|███▊      | 2435/6470 [28:01<49:40,  1.35it/s]
4|2436|Loss: 0.8469223380088806:  38%|███▊      | 2436/6470 [28:01<50:29,  1.33it/s]
4|2437|Loss: 0.8695041537284851:  38%|███▊      | 2436/6470 [28:02<50:29,  1.33it/s]
4|2437|Loss: 0.8695041537284851:  38%|███▊      | 2437/6470 [28:02<47:34,  1.41it/s]
4|2438|Loss: 1.0611813068389893:  38%|███▊      | 2437/6470 [28:02<47:34,  1.41it/s]
4|2438|Loss: 1.0611813068389893:  38%|███▊      | 2438/6470 [28:02<47:31,  1.41it/s]
4|2439|Loss: 0.7781732082366943:  38%|███▊      | 2438/6470 [28:03<47:31,  1.41it/s]
4|2439|Loss: 0.7781732082366943:  38%|███▊      | 2439/6470 [28:03<48:20,  1.39it/s]
4|2440|Loss: 0.956251859664917:  38%|███▊      | 2439/6470 [28:04<48:20,  1.39it/s] 
4|2440|Loss: 0.956251859664917:  38%|███▊      | 2440/6470 [28:04<49:08,  1.37it/s]
4|2441|Loss: 0.9195763468742371:  38%|███▊      | 2440/6470 [28:04<49:08,  1.37it/s]
4|2441|Loss: 0.9195763468742371:  38%|███▊      | 2441/6470 [28:05<46:33,  1.44it/s]
4|2442|Loss: 0.7926046848297119:  38%|███▊      | 2441/6470 [28:05<46:33,  1.44it/s]
4|2442|Loss: 0.7926046848297119:  38%|███▊      | 2442/6470 [28:05<46:17,  1.45it/s]
4|2443|Loss: 0.9520668983459473:  38%|███▊      | 2442/6470 [28:06<46:17,  1.45it/s]
4|2443|Loss: 0.9520668983459473:  38%|███▊      | 2443/6470 [28:06<45:17,  1.48it/s]
4|2444|Loss: 0.7423722147941589:  38%|███▊      | 2443/6470 [28:06<45:17,  1.48it/s]
4|2444|Loss: 0.7423722147941589:  38%|███▊      | 2444/6470 [28:06<42:40,  1.57it/s]
4|2445|Loss: 0.7093287706375122:  38%|███▊      | 2444/6470 [28:07<42:40,  1.57it/s]
4|2445|Loss: 0.7093287706375122:  38%|███▊      | 2445/6470 [28:07<42:21,  1.58it/s]
4|2446|Loss: 0.9626492261886597:  38%|███▊      | 2445/6470 [28:07<42:21,  1.58it/s]
4|2446|Loss: 0.9626492261886597:  38%|███▊      | 2446/6470 [28:08<44:09,  1.52it/s]
4|2447|Loss: 1.0009764432907104:  38%|███▊      | 2446/6470 [28:08<44:09,  1.52it/s]
4|2447|Loss: 1.0009764432907104:  38%|███▊      | 2447/6470 [28:08<45:19,  1.48it/s]
4|2448|Loss: 0.9036517143249512:  38%|███▊      | 2447/6470 [28:09<45:19,  1.48it/s]
4|2448|Loss: 0.9036517143249512:  38%|███▊      | 2448/6470 [28:09<44:54,  1.49it/s]
4|2449|Loss: 0.8267871737480164:  38%|███▊      | 2448/6470 [28:10<44:54,  1.49it/s]
4|2449|Loss: 0.8267871737480164:  38%|███▊      | 2449/6470 [28:10<44:26,  1.51it/s]
4|2450|Loss: 0.8790780305862427:  38%|███▊      | 2449/6470 [28:10<44:26,  1.51it/s]
4|2450|Loss: 0.8790780305862427:  38%|███▊      | 2450/6470 [28:10<44:04,  1.52it/s]
4|2451|Loss: 0.8337484002113342:  38%|███▊      | 2450/6470 [28:11<44:04,  1.52it/s]
4|2451|Loss: 0.8337484002113342:  38%|███▊      | 2451/6470 [28:11<47:12,  1.42it/s]
4|2452|Loss: 0.8688197731971741:  38%|███▊      | 2451/6470 [28:12<47:12,  1.42it/s]
4|2452|Loss: 0.8688197731971741:  38%|███▊      | 2452/6470 [28:12<46:56,  1.43it/s]
4|2453|Loss: 0.9472456574440002:  38%|███▊      | 2452/6470 [28:12<46:56,  1.43it/s]
4|2453|Loss: 0.9472456574440002:  38%|███▊      | 2453/6470 [28:13<44:40,  1.50it/s]
4|2454|Loss: 0.9431171417236328:  38%|███▊      | 2453/6470 [28:13<44:40,  1.50it/s]
4|2454|Loss: 0.9431171417236328:  38%|███▊      | 2454/6470 [28:13<42:33,  1.57it/s]
4|2455|Loss: 0.8961688280105591:  38%|███▊      | 2454/6470 [28:13<42:33,  1.57it/s]
4|2455|Loss: 0.8961688280105591:  38%|███▊      | 2455/6470 [28:14<41:33,  1.61it/s]
4|2456|Loss: 0.8481785655021667:  38%|███▊      | 2455/6470 [28:14<41:33,  1.61it/s]
4|2456|Loss: 0.8481785655021667:  38%|███▊      | 2456/6470 [28:14<41:25,  1.61it/s]
4|2457|Loss: 0.8254988193511963:  38%|███▊      | 2456/6470 [28:15<41:25,  1.61it/s]
4|2457|Loss: 0.8254988193511963:  38%|███▊      | 2457/6470 [28:15<45:08,  1.48it/s]
4|2458|Loss: 0.9711526036262512:  38%|███▊      | 2457/6470 [28:16<45:08,  1.48it/s]
4|2458|Loss: 0.9711526036262512:  38%|███▊      | 2458/6470 [28:16<47:50,  1.40it/s]
4|2459|Loss: 0.7845105528831482:  38%|███▊      | 2458/6470 [28:16<47:50,  1.40it/s]
4|2459|Loss: 0.7845105528831482:  38%|███▊      | 2459/6470 [28:17<47:58,  1.39it/s]
4|2460|Loss: 0.8403721451759338:  38%|███▊      | 2459/6470 [28:17<47:58,  1.39it/s]
4|2460|Loss: 0.8403721451759338:  38%|███▊      | 2460/6470 [28:17<50:11,  1.33it/s]
4|2461|Loss: 0.9198833107948303:  38%|███▊      | 2460/6470 [28:18<50:11,  1.33it/s]
4|2461|Loss: 0.9198833107948303:  38%|███▊      | 2461/6470 [28:18<51:54,  1.29it/s]
4|2462|Loss: 0.8341482281684875:  38%|███▊      | 2461/6470 [28:19<51:54,  1.29it/s]
4|2462|Loss: 0.8341482281684875:  38%|███▊      | 2462/6470 [28:19<47:49,  1.40it/s]
4|2463|Loss: 0.8790964484214783:  38%|███▊      | 2462/6470 [28:19<47:49,  1.40it/s]
4|2463|Loss: 0.8790964484214783:  38%|███▊      | 2463/6470 [28:19<44:22,  1.50it/s]
4|2464|Loss: 0.6400729417800903:  38%|███▊      | 2463/6470 [28:20<44:22,  1.50it/s]
4|2464|Loss: 0.6400729417800903:  38%|███▊      | 2464/6470 [28:20<51:03,  1.31it/s]
4|2465|Loss: 0.8049164414405823:  38%|███▊      | 2464/6470 [28:21<51:03,  1.31it/s]
4|2465|Loss: 0.8049164414405823:  38%|███▊      | 2465/6470 [28:21<52:12,  1.28it/s]
4|2466|Loss: 0.8862101435661316:  38%|███▊      | 2465/6470 [28:22<52:12,  1.28it/s]
4|2466|Loss: 0.8862101435661316:  38%|███▊      | 2466/6470 [28:22<51:35,  1.29it/s]
4|2467|Loss: 0.9324992895126343:  38%|███▊      | 2466/6470 [28:22<51:35,  1.29it/s]
4|2467|Loss: 0.9324992895126343:  38%|███▊      | 2467/6470 [28:23<46:48,  1.43it/s]
4|2468|Loss: 0.8683539628982544:  38%|███▊      | 2467/6470 [28:23<46:48,  1.43it/s]
4|2468|Loss: 0.8683539628982544:  38%|███▊      | 2468/6470 [28:23<44:33,  1.50it/s]
4|2469|Loss: 0.9979678392410278:  38%|███▊      | 2468/6470 [28:24<44:33,  1.50it/s]
4|2469|Loss: 0.9979678392410278:  38%|███▊      | 2469/6470 [28:24<46:22,  1.44it/s]
4|2470|Loss: 0.9764471650123596:  38%|███▊      | 2469/6470 [28:24<46:22,  1.44it/s]
4|2470|Loss: 0.9764471650123596:  38%|███▊      | 2470/6470 [28:25<45:05,  1.48it/s]
4|2471|Loss: 0.9028224349021912:  38%|███▊      | 2470/6470 [28:25<45:05,  1.48it/s]
4|2471|Loss: 0.9028224349021912:  38%|███▊      | 2471/6470 [28:25<45:49,  1.45it/s]
4|2472|Loss: 0.854907751083374:  38%|███▊      | 2471/6470 [28:26<45:49,  1.45it/s] 
4|2472|Loss: 0.854907751083374:  38%|███▊      | 2472/6470 [28:26<47:00,  1.42it/s]
4|2473|Loss: 0.8584671020507812:  38%|███▊      | 2472/6470 [28:26<47:00,  1.42it/s]
4|2473|Loss: 0.8584671020507812:  38%|███▊      | 2473/6470 [28:27<48:48,  1.36it/s]
4|2474|Loss: 0.8972154855728149:  38%|███▊      | 2473/6470 [28:27<48:48,  1.36it/s]
4|2474|Loss: 0.8972154855728149:  38%|███▊      | 2474/6470 [28:28<50:36,  1.32it/s]
4|2475|Loss: 0.8282039761543274:  38%|███▊      | 2474/6470 [28:28<50:36,  1.32it/s]
4|2475|Loss: 0.8282039761543274:  38%|███▊      | 2475/6470 [28:28<51:22,  1.30it/s]
4|2476|Loss: 0.9898305535316467:  38%|███▊      | 2475/6470 [28:29<51:22,  1.30it/s]
4|2476|Loss: 0.9898305535316467:  38%|███▊      | 2476/6470 [28:29<49:30,  1.34it/s]
4|2477|Loss: 0.9542176723480225:  38%|███▊      | 2476/6470 [28:29<49:30,  1.34it/s]
4|2477|Loss: 0.9542176723480225:  38%|███▊      | 2477/6470 [28:30<44:24,  1.50it/s]
4|2478|Loss: 0.9400883913040161:  38%|███▊      | 2477/6470 [28:30<44:24,  1.50it/s]
4|2478|Loss: 0.9400883913040161:  38%|███▊      | 2478/6470 [28:30<46:10,  1.44it/s]
4|2479|Loss: 0.8849931955337524:  38%|███▊      | 2478/6470 [28:31<46:10,  1.44it/s]
4|2479|Loss: 0.8849931955337524:  38%|███▊      | 2479/6470 [28:31<49:52,  1.33it/s]
4|2480|Loss: 0.8617416620254517:  38%|███▊      | 2479/6470 [28:32<49:52,  1.33it/s]
4|2480|Loss: 0.8617416620254517:  38%|███▊      | 2480/6470 [28:32<45:53,  1.45it/s]
4|2481|Loss: 0.7779640555381775:  38%|███▊      | 2480/6470 [28:32<45:53,  1.45it/s]
4|2481|Loss: 0.7779640555381775:  38%|███▊      | 2481/6470 [28:32<47:00,  1.41it/s]
4|2482|Loss: 0.7686675190925598:  38%|███▊      | 2481/6470 [28:33<47:00,  1.41it/s]
4|2482|Loss: 0.7686675190925598:  38%|███▊      | 2482/6470 [28:33<48:02,  1.38it/s]
4|2483|Loss: 0.8632173538208008:  38%|███▊      | 2482/6470 [28:34<48:02,  1.38it/s]
4|2483|Loss: 0.8632173538208008:  38%|███▊      | 2483/6470 [28:34<46:30,  1.43it/s]
4|2484|Loss: 0.7782403230667114:  38%|███▊      | 2483/6470 [28:34<46:30,  1.43it/s]
4|2484|Loss: 0.7782403230667114:  38%|███▊      | 2484/6470 [28:35<46:44,  1.42it/s]
4|2485|Loss: 1.0764485597610474:  38%|███▊      | 2484/6470 [28:35<46:44,  1.42it/s]
4|2485|Loss: 1.0764485597610474:  38%|███▊      | 2485/6470 [28:35<47:00,  1.41it/s]
4|2486|Loss: 0.7172025442123413:  38%|███▊      | 2485/6470 [28:36<47:00,  1.41it/s]
4|2486|Loss: 0.7172025442123413:  38%|███▊      | 2486/6470 [28:36<46:47,  1.42it/s]
4|2487|Loss: 0.7594199776649475:  38%|███▊      | 2486/6470 [28:36<46:47,  1.42it/s]
4|2487|Loss: 0.7594199776649475:  38%|███▊      | 2487/6470 [28:37<44:38,  1.49it/s]
4|2488|Loss: 0.7832355499267578:  38%|███▊      | 2487/6470 [28:37<44:38,  1.49it/s]
4|2488|Loss: 0.7832355499267578:  38%|███▊      | 2488/6470 [28:37<41:36,  1.59it/s]
4|2489|Loss: 0.8573847413063049:  38%|███▊      | 2488/6470 [28:37<41:36,  1.59it/s]
4|2489|Loss: 0.8573847413063049:  38%|███▊      | 2489/6470 [28:38<40:59,  1.62it/s]
4|2490|Loss: 0.8437413573265076:  38%|███▊      | 2489/6470 [28:38<40:59,  1.62it/s]
4|2490|Loss: 0.8437413573265076:  38%|███▊      | 2490/6470 [28:38<39:33,  1.68it/s]
4|2491|Loss: 0.9485379457473755:  38%|███▊      | 2490/6470 [28:39<39:33,  1.68it/s]
4|2491|Loss: 0.9485379457473755:  39%|███▊      | 2491/6470 [28:39<35:57,  1.84it/s]
4|2492|Loss: 0.7834175229072571:  39%|███▊      | 2491/6470 [28:39<35:57,  1.84it/s]
4|2492|Loss: 0.7834175229072571:  39%|███▊      | 2492/6470 [28:39<34:59,  1.89it/s]
4|2493|Loss: 0.6859128475189209:  39%|███▊      | 2492/6470 [28:40<34:59,  1.89it/s]
4|2493|Loss: 0.6859128475189209:  39%|███▊      | 2493/6470 [28:40<37:49,  1.75it/s]
4|2494|Loss: 0.8453224301338196:  39%|███▊      | 2493/6470 [28:40<37:49,  1.75it/s]
4|2494|Loss: 0.8453224301338196:  39%|███▊      | 2494/6470 [28:40<39:13,  1.69it/s]
4|2495|Loss: 0.8478284478187561:  39%|███▊      | 2494/6470 [28:41<39:13,  1.69it/s]
4|2495|Loss: 0.8478284478187561:  39%|███▊      | 2495/6470 [28:41<41:35,  1.59it/s]
4|2496|Loss: 0.9665764570236206:  39%|███▊      | 2495/6470 [28:42<41:35,  1.59it/s]
4|2496|Loss: 0.9665764570236206:  39%|███▊      | 2496/6470 [28:42<40:47,  1.62it/s]
4|2497|Loss: 0.7832886576652527:  39%|███▊      | 2496/6470 [28:42<40:47,  1.62it/s]
4|2497|Loss: 0.7832886576652527:  39%|███▊      | 2497/6470 [28:43<44:26,  1.49it/s]
4|2498|Loss: 0.8188220262527466:  39%|███▊      | 2497/6470 [28:43<44:26,  1.49it/s]
4|2498|Loss: 0.8188220262527466:  39%|███▊      | 2498/6470 [28:43<43:46,  1.51it/s]
4|2499|Loss: 0.9289789199829102:  39%|███▊      | 2498/6470 [28:44<43:46,  1.51it/s]
4|2499|Loss: 0.9289789199829102:  39%|███▊      | 2499/6470 [28:44<45:27,  1.46it/s]
4|2500|Loss: 0.7837309241294861:  39%|███▊      | 2499/6470 [28:44<45:27,  1.46it/s]
4|2500|Loss: 0.7837309241294861:  39%|███▊      | 2500/6470 [28:45<46:41,  1.42it/s]
4|2501|Loss: 0.8054723143577576:  39%|███▊      | 2500/6470 [28:45<46:41,  1.42it/s]
4|2501|Loss: 0.8054723143577576:  39%|███▊      | 2501/6470 [28:46<48:07,  1.37it/s]
4|2502|Loss: 0.8508807420730591:  39%|███▊      | 2501/6470 [28:46<48:07,  1.37it/s]
4|2502|Loss: 0.8508807420730591:  39%|███▊      | 2502/6470 [28:46<46:39,  1.42it/s]
4|2503|Loss: 0.8394007086753845:  39%|███▊      | 2502/6470 [28:47<46:39,  1.42it/s]
4|2503|Loss: 0.8394007086753845:  39%|███▊      | 2503/6470 [28:47<48:40,  1.36it/s]
4|2504|Loss: 0.7121551632881165:  39%|███▊      | 2503/6470 [28:47<48:40,  1.36it/s]
4|2504|Loss: 0.7121551632881165:  39%|███▊      | 2504/6470 [28:47<43:44,  1.51it/s]
4|2505|Loss: 0.7866058349609375:  39%|███▊      | 2504/6470 [28:48<43:44,  1.51it/s]
4|2505|Loss: 0.7866058349609375:  39%|███▊      | 2505/6470 [28:48<41:25,  1.60it/s]
4|2506|Loss: 0.7835664749145508:  39%|███▊      | 2505/6470 [28:48<41:25,  1.60it/s]
4|2506|Loss: 0.7835664749145508:  39%|███▊      | 2506/6470 [28:49<41:26,  1.59it/s]
4|2507|Loss: 0.8260937929153442:  39%|███▊      | 2506/6470 [28:49<41:26,  1.59it/s]
4|2507|Loss: 0.8260937929153442:  39%|███▊      | 2507/6470 [28:49<39:40,  1.66it/s]
4|2508|Loss: 0.7008786201477051:  39%|███▊      | 2507/6470 [28:50<39:40,  1.66it/s]
4|2508|Loss: 0.7008786201477051:  39%|███▉      | 2508/6470 [28:50<39:32,  1.67it/s]
4|2509|Loss: 0.7451592683792114:  39%|███▉      | 2508/6470 [28:50<39:32,  1.67it/s]
4|2509|Loss: 0.7451592683792114:  39%|███▉      | 2509/6470 [28:50<41:38,  1.59it/s]
4|2510|Loss: 0.8440259695053101:  39%|███▉      | 2509/6470 [28:51<41:38,  1.59it/s]
4|2510|Loss: 0.8440259695053101:  39%|███▉      | 2510/6470 [28:51<45:48,  1.44it/s]
4|2511|Loss: 0.9155480265617371:  39%|███▉      | 2510/6470 [28:52<45:48,  1.44it/s]
4|2511|Loss: 0.9155480265617371:  39%|███▉      | 2511/6470 [28:52<43:46,  1.51it/s]
4|2512|Loss: 0.7245793342590332:  39%|███▉      | 2511/6470 [28:52<43:46,  1.51it/s]
4|2512|Loss: 0.7245793342590332:  39%|███▉      | 2512/6470 [28:53<44:29,  1.48it/s]
4|2513|Loss: 0.8676828742027283:  39%|███▉      | 2512/6470 [28:53<44:29,  1.48it/s]
4|2513|Loss: 0.8676828742027283:  39%|███▉      | 2513/6470 [28:53<45:42,  1.44it/s]
4|2514|Loss: 0.783491849899292:  39%|███▉      | 2513/6470 [28:54<45:42,  1.44it/s] 
4|2514|Loss: 0.783491849899292:  39%|███▉      | 2514/6470 [28:54<44:46,  1.47it/s]
4|2515|Loss: 0.9794952273368835:  39%|███▉      | 2514/6470 [28:54<44:46,  1.47it/s]
4|2515|Loss: 0.9794952273368835:  39%|███▉      | 2515/6470 [28:55<43:05,  1.53it/s]
4|2516|Loss: 0.7726014852523804:  39%|███▉      | 2515/6470 [28:55<43:05,  1.53it/s]
4|2516|Loss: 0.7726014852523804:  39%|███▉      | 2516/6470 [28:55<46:35,  1.41it/s]
4|2517|Loss: 0.7575877904891968:  39%|███▉      | 2516/6470 [28:56<46:35,  1.41it/s]
4|2517|Loss: 0.7575877904891968:  39%|███▉      | 2517/6470 [28:56<48:05,  1.37it/s]
4|2518|Loss: 0.9068241715431213:  39%|███▉      | 2517/6470 [28:57<48:05,  1.37it/s]
4|2518|Loss: 0.9068241715431213:  39%|███▉      | 2518/6470 [28:57<44:45,  1.47it/s]
4|2519|Loss: 0.6170851588249207:  39%|███▉      | 2518/6470 [28:57<44:45,  1.47it/s]
4|2519|Loss: 0.6170851588249207:  39%|███▉      | 2519/6470 [28:57<44:57,  1.46it/s]
4|2520|Loss: 0.8691152930259705:  39%|███▉      | 2519/6470 [28:58<44:57,  1.46it/s]
4|2520|Loss: 0.8691152930259705:  39%|███▉      | 2520/6470 [28:58<47:02,  1.40it/s]
4|2521|Loss: 0.790969729423523:  39%|███▉      | 2520/6470 [28:59<47:02,  1.40it/s] 
4|2521|Loss: 0.790969729423523:  39%|███▉      | 2521/6470 [28:59<45:55,  1.43it/s]
4|2522|Loss: 0.768771231174469:  39%|███▉      | 2521/6470 [28:59<45:55,  1.43it/s]
4|2522|Loss: 0.768771231174469:  39%|███▉      | 2522/6470 [29:00<48:15,  1.36it/s]
4|2523|Loss: 0.9084380865097046:  39%|███▉      | 2522/6470 [29:00<48:15,  1.36it/s]
4|2523|Loss: 0.9084380865097046:  39%|███▉      | 2523/6470 [29:01<50:47,  1.29it/s]
4|2524|Loss: 0.7615324258804321:  39%|███▉      | 2523/6470 [29:01<50:47,  1.29it/s]
4|2524|Loss: 0.7615324258804321:  39%|███▉      | 2524/6470 [29:01<44:30,  1.48it/s]
4|2525|Loss: 0.7408257126808167:  39%|███▉      | 2524/6470 [29:01<44:30,  1.48it/s]
4|2525|Loss: 0.7408257126808167:  39%|███▉      | 2525/6470 [29:02<47:23,  1.39it/s]
4|2526|Loss: 0.8590965270996094:  39%|███▉      | 2525/6470 [29:02<47:23,  1.39it/s]
4|2526|Loss: 0.8590965270996094:  39%|███▉      | 2526/6470 [29:03<46:14,  1.42it/s]
4|2527|Loss: 0.7587172985076904:  39%|███▉      | 2526/6470 [29:03<46:14,  1.42it/s]
4|2527|Loss: 0.7587172985076904:  39%|███▉      | 2527/6470 [29:03<44:26,  1.48it/s]
4|2528|Loss: 0.8696693778038025:  39%|███▉      | 2527/6470 [29:03<44:26,  1.48it/s]
4|2528|Loss: 0.8696693778038025:  39%|███▉      | 2528/6470 [29:04<43:20,  1.52it/s]
4|2529|Loss: 0.8668779134750366:  39%|███▉      | 2528/6470 [29:04<43:20,  1.52it/s]
4|2529|Loss: 0.8668779134750366:  39%|███▉      | 2529/6470 [29:05<45:13,  1.45it/s]
4|2530|Loss: 1.0192993879318237:  39%|███▉      | 2529/6470 [29:05<45:13,  1.45it/s]
4|2530|Loss: 1.0192993879318237:  39%|███▉      | 2530/6470 [29:05<44:51,  1.46it/s]
4|2531|Loss: 0.8647346496582031:  39%|███▉      | 2530/6470 [29:06<44:51,  1.46it/s]
4|2531|Loss: 0.8647346496582031:  39%|███▉      | 2531/6470 [29:06<45:23,  1.45it/s]
4|2532|Loss: 0.748397946357727:  39%|███▉      | 2531/6470 [29:06<45:23,  1.45it/s] 
4|2532|Loss: 0.748397946357727:  39%|███▉      | 2532/6470 [29:07<44:30,  1.47it/s]
4|2533|Loss: 0.8844664692878723:  39%|███▉      | 2532/6470 [29:07<44:30,  1.47it/s]
4|2533|Loss: 0.8844664692878723:  39%|███▉      | 2533/6470 [29:07<47:19,  1.39it/s]
4|2534|Loss: 0.6959921717643738:  39%|███▉      | 2533/6470 [29:08<47:19,  1.39it/s]
4|2534|Loss: 0.6959921717643738:  39%|███▉      | 2534/6470 [29:08<46:22,  1.41it/s]
4|2535|Loss: 0.8944178819656372:  39%|███▉      | 2534/6470 [29:08<46:22,  1.41it/s]
4|2535|Loss: 0.8944178819656372:  39%|███▉      | 2535/6470 [29:09<47:28,  1.38it/s]
4|2536|Loss: 0.7657452821731567:  39%|███▉      | 2535/6470 [29:09<47:28,  1.38it/s]
4|2536|Loss: 0.7657452821731567:  39%|███▉      | 2536/6470 [29:10<47:31,  1.38it/s]
4|2537|Loss: 0.8771572709083557:  39%|███▉      | 2536/6470 [29:10<47:31,  1.38it/s]
4|2537|Loss: 0.8771572709083557:  39%|███▉      | 2537/6470 [29:10<45:08,  1.45it/s]
4|2538|Loss: 0.9178365468978882:  39%|███▉      | 2537/6470 [29:10<45:08,  1.45it/s]
4|2538|Loss: 0.9178365468978882:  39%|███▉      | 2538/6470 [29:11<44:04,  1.49it/s]
4|2539|Loss: 0.8802880644798279:  39%|███▉      | 2538/6470 [29:11<44:04,  1.49it/s]
4|2539|Loss: 0.8802880644798279:  39%|███▉      | 2539/6470 [29:11<44:30,  1.47it/s]
4|2540|Loss: 0.6533037424087524:  39%|███▉      | 2539/6470 [29:12<44:30,  1.47it/s]
4|2540|Loss: 0.6533037424087524:  39%|███▉      | 2540/6470 [29:12<50:32,  1.30it/s]
4|2541|Loss: 0.910732090473175:  39%|███▉      | 2540/6470 [29:13<50:32,  1.30it/s] 
4|2541|Loss: 0.910732090473175:  39%|███▉      | 2541/6470 [29:13<49:34,  1.32it/s]
4|2542|Loss: 0.7457060217857361:  39%|███▉      | 2541/6470 [29:14<49:34,  1.32it/s]
4|2542|Loss: 0.7457060217857361:  39%|███▉      | 2542/6470 [29:14<49:54,  1.31it/s]
4|2543|Loss: 0.8135789036750793:  39%|███▉      | 2542/6470 [29:14<49:54,  1.31it/s]
4|2543|Loss: 0.8135789036750793:  39%|███▉      | 2543/6470 [29:15<49:04,  1.33it/s]
4|2544|Loss: 0.8911083936691284:  39%|███▉      | 2543/6470 [29:15<49:04,  1.33it/s]
4|2544|Loss: 0.8911083936691284:  39%|███▉      | 2544/6470 [29:15<46:07,  1.42it/s]
4|2545|Loss: 0.8332126140594482:  39%|███▉      | 2544/6470 [29:16<46:07,  1.42it/s]
4|2545|Loss: 0.8332126140594482:  39%|███▉      | 2545/6470 [29:16<48:24,  1.35it/s]
4|2546|Loss: 0.9342643022537231:  39%|███▉      | 2545/6470 [29:17<48:24,  1.35it/s]
4|2546|Loss: 0.9342643022537231:  39%|███▉      | 2546/6470 [29:17<50:31,  1.29it/s]
4|2547|Loss: 0.8048157691955566:  39%|███▉      | 2546/6470 [29:17<50:31,  1.29it/s]
4|2547|Loss: 0.8048157691955566:  39%|███▉      | 2547/6470 [29:18<48:28,  1.35it/s]
4|2548|Loss: 0.8542177081108093:  39%|███▉      | 2547/6470 [29:18<48:28,  1.35it/s]
4|2548|Loss: 0.8542177081108093:  39%|███▉      | 2548/6470 [29:18<47:21,  1.38it/s]
4|2549|Loss: 1.0095770359039307:  39%|███▉      | 2548/6470 [29:19<47:21,  1.38it/s]
4|2549|Loss: 1.0095770359039307:  39%|███▉      | 2549/6470 [29:19<47:21,  1.38it/s]
4|2550|Loss: 0.718665599822998:  39%|███▉      | 2549/6470 [29:19<47:21,  1.38it/s] 
4|2550|Loss: 0.718665599822998:  39%|███▉      | 2550/6470 [29:20<47:42,  1.37it/s]
4|2551|Loss: 0.9171953797340393:  39%|███▉      | 2550/6470 [29:20<47:42,  1.37it/s]
4|2551|Loss: 0.9171953797340393:  39%|███▉      | 2551/6470 [29:20<47:47,  1.37it/s]
4|2552|Loss: 0.8566244840621948:  39%|███▉      | 2551/6470 [29:21<47:47,  1.37it/s]
4|2552|Loss: 0.8566244840621948:  39%|███▉      | 2552/6470 [29:21<45:52,  1.42it/s]
4|2553|Loss: 0.7630581259727478:  39%|███▉      | 2552/6470 [29:21<45:52,  1.42it/s]
4|2553|Loss: 0.7630581259727478:  39%|███▉      | 2553/6470 [29:22<45:32,  1.43it/s]
4|2554|Loss: 0.8534225225448608:  39%|███▉      | 2553/6470 [29:22<45:32,  1.43it/s]
4|2554|Loss: 0.8534225225448608:  39%|███▉      | 2554/6470 [29:22<42:53,  1.52it/s]
4|2555|Loss: 0.9121439456939697:  39%|███▉      | 2554/6470 [29:23<42:53,  1.52it/s]
4|2555|Loss: 0.9121439456939697:  39%|███▉      | 2555/6470 [29:23<44:39,  1.46it/s]
4|2556|Loss: 0.8005577325820923:  39%|███▉      | 2555/6470 [29:24<44:39,  1.46it/s]
4|2556|Loss: 0.8005577325820923:  40%|███▉      | 2556/6470 [29:24<48:01,  1.36it/s]
4|2557|Loss: 0.7610353231430054:  40%|███▉      | 2556/6470 [29:24<48:01,  1.36it/s]
4|2557|Loss: 0.7610353231430054:  40%|███▉      | 2557/6470 [29:25<49:44,  1.31it/s]
4|2558|Loss: 0.8119009733200073:  40%|███▉      | 2557/6470 [29:25<49:44,  1.31it/s]
4|2558|Loss: 0.8119009733200073:  40%|███▉      | 2558/6470 [29:25<46:36,  1.40it/s]
4|2559|Loss: 0.9094318151473999:  40%|███▉      | 2558/6470 [29:26<46:36,  1.40it/s]
4|2559|Loss: 0.9094318151473999:  40%|███▉      | 2559/6470 [29:26<44:33,  1.46it/s]
4|2560|Loss: 0.8678093552589417:  40%|███▉      | 2559/6470 [29:26<44:33,  1.46it/s]
4|2560|Loss: 0.8678093552589417:  40%|███▉      | 2560/6470 [29:27<43:24,  1.50it/s]
4|2561|Loss: 0.8309741616249084:  40%|███▉      | 2560/6470 [29:27<43:24,  1.50it/s]
4|2561|Loss: 0.8309741616249084:  40%|███▉      | 2561/6470 [29:27<45:11,  1.44it/s]
4|2562|Loss: 0.8951329588890076:  40%|███▉      | 2561/6470 [29:28<45:11,  1.44it/s]
4|2562|Loss: 0.8951329588890076:  40%|███▉      | 2562/6470 [29:28<47:38,  1.37it/s]
4|2563|Loss: 0.9363940954208374:  40%|███▉      | 2562/6470 [29:29<47:38,  1.37it/s]
4|2563|Loss: 0.9363940954208374:  40%|███▉      | 2563/6470 [29:29<45:12,  1.44it/s]
4|2564|Loss: 0.8290565609931946:  40%|███▉      | 2563/6470 [29:29<45:12,  1.44it/s]
4|2564|Loss: 0.8290565609931946:  40%|███▉      | 2564/6470 [29:30<45:15,  1.44it/s]
4|2565|Loss: 0.8325002193450928:  40%|███▉      | 2564/6470 [29:30<45:15,  1.44it/s]
4|2565|Loss: 0.8325002193450928:  40%|███▉      | 2565/6470 [29:30<47:01,  1.38it/s]
4|2566|Loss: 0.7314720153808594:  40%|███▉      | 2565/6470 [29:31<47:01,  1.38it/s]
4|2566|Loss: 0.7314720153808594:  40%|███▉      | 2566/6470 [29:31<41:54,  1.55it/s]
4|2567|Loss: 0.9202091097831726:  40%|███▉      | 2566/6470 [29:31<41:54,  1.55it/s]
4|2567|Loss: 0.9202091097831726:  40%|███▉      | 2567/6470 [29:31<42:13,  1.54it/s]
4|2568|Loss: 0.8963514566421509:  40%|███▉      | 2567/6470 [29:32<42:13,  1.54it/s]
4|2568|Loss: 0.8963514566421509:  40%|███▉      | 2568/6470 [29:32<45:20,  1.43it/s]
4|2569|Loss: 0.7302442789077759:  40%|███▉      | 2568/6470 [29:33<45:20,  1.43it/s]
4|2569|Loss: 0.7302442789077759:  40%|███▉      | 2569/6470 [29:33<49:14,  1.32it/s]
4|2570|Loss: 0.9823480248451233:  40%|███▉      | 2569/6470 [29:34<49:14,  1.32it/s]
4|2570|Loss: 0.9823480248451233:  40%|███▉      | 2570/6470 [29:34<51:06,  1.27it/s]
4|2571|Loss: 0.919849157333374:  40%|███▉      | 2570/6470 [29:34<51:06,  1.27it/s] 
4|2571|Loss: 0.919849157333374:  40%|███▉      | 2571/6470 [29:35<49:03,  1.32it/s]
4|2572|Loss: 0.7564213275909424:  40%|███▉      | 2571/6470 [29:35<49:03,  1.32it/s]
4|2572|Loss: 0.7564213275909424:  40%|███▉      | 2572/6470 [29:35<44:43,  1.45it/s]
4|2573|Loss: 0.8070312738418579:  40%|███▉      | 2572/6470 [29:36<44:43,  1.45it/s]
4|2573|Loss: 0.8070312738418579:  40%|███▉      | 2573/6470 [29:36<42:13,  1.54it/s]
4|2574|Loss: 0.7237153649330139:  40%|███▉      | 2573/6470 [29:36<42:13,  1.54it/s]
4|2574|Loss: 0.7237153649330139:  40%|███▉      | 2574/6470 [29:36<38:52,  1.67it/s]
4|2575|Loss: 0.8633424043655396:  40%|███▉      | 2574/6470 [29:37<38:52,  1.67it/s]
4|2575|Loss: 0.8633424043655396:  40%|███▉      | 2575/6470 [29:37<40:18,  1.61it/s]
4|2576|Loss: 0.9611497521400452:  40%|███▉      | 2575/6470 [29:37<40:18,  1.61it/s]
4|2576|Loss: 0.9611497521400452:  40%|███▉      | 2576/6470 [29:38<42:32,  1.53it/s]
4|2577|Loss: 0.9816674590110779:  40%|███▉      | 2576/6470 [29:38<42:32,  1.53it/s]
4|2577|Loss: 0.9816674590110779:  40%|███▉      | 2577/6470 [29:38<41:13,  1.57it/s]
4|2578|Loss: 0.8922643661499023:  40%|███▉      | 2577/6470 [29:39<41:13,  1.57it/s]
4|2578|Loss: 0.8922643661499023:  40%|███▉      | 2578/6470 [29:39<41:00,  1.58it/s]
4|2579|Loss: 0.8355798125267029:  40%|███▉      | 2578/6470 [29:39<41:00,  1.58it/s]
4|2579|Loss: 0.8355798125267029:  40%|███▉      | 2579/6470 [29:40<44:28,  1.46it/s]
4|2580|Loss: 0.9199223518371582:  40%|███▉      | 2579/6470 [29:40<44:28,  1.46it/s]
4|2580|Loss: 0.9199223518371582:  40%|███▉      | 2580/6470 [29:40<41:36,  1.56it/s]
4|2581|Loss: 0.814507246017456:  40%|███▉      | 2580/6470 [29:41<41:36,  1.56it/s] 
4|2581|Loss: 0.814507246017456:  40%|███▉      | 2581/6470 [29:41<40:45,  1.59it/s]
4|2582|Loss: 0.9184525609016418:  40%|███▉      | 2581/6470 [29:41<40:45,  1.59it/s]
4|2582|Loss: 0.9184525609016418:  40%|███▉      | 2582/6470 [29:42<44:25,  1.46it/s]
4|2583|Loss: 0.8860101699829102:  40%|███▉      | 2582/6470 [29:42<44:25,  1.46it/s]
4|2583|Loss: 0.8860101699829102:  40%|███▉      | 2583/6470 [29:43<47:51,  1.35it/s]
4|2584|Loss: 0.91457200050354:  40%|███▉      | 2583/6470 [29:43<47:51,  1.35it/s]  
4|2584|Loss: 0.91457200050354:  40%|███▉      | 2584/6470 [29:43<48:30,  1.34it/s]
4|2585|Loss: 0.9731251001358032:  40%|███▉      | 2584/6470 [29:44<48:30,  1.34it/s]
4|2585|Loss: 0.9731251001358032:  40%|███▉      | 2585/6470 [29:44<48:31,  1.33it/s]
4|2586|Loss: 0.8706725835800171:  40%|███▉      | 2585/6470 [29:44<48:31,  1.33it/s]
4|2586|Loss: 0.8706725835800171:  40%|███▉      | 2586/6470 [29:45<46:41,  1.39it/s]
4|2587|Loss: 0.8401134610176086:  40%|███▉      | 2586/6470 [29:45<46:41,  1.39it/s]
4|2587|Loss: 0.8401134610176086:  40%|███▉      | 2587/6470 [29:45<45:00,  1.44it/s]
4|2588|Loss: 0.8341503143310547:  40%|███▉      | 2587/6470 [29:46<45:00,  1.44it/s]
4|2588|Loss: 0.8341503143310547:  40%|████      | 2588/6470 [29:46<42:52,  1.51it/s]
4|2589|Loss: 0.9445655941963196:  40%|████      | 2588/6470 [29:46<42:52,  1.51it/s]
4|2589|Loss: 0.9445655941963196:  40%|████      | 2589/6470 [29:46<37:58,  1.70it/s]
4|2590|Loss: 0.8841106295585632:  40%|████      | 2589/6470 [29:47<37:58,  1.70it/s]
4|2590|Loss: 0.8841106295585632:  40%|████      | 2590/6470 [29:47<39:30,  1.64it/s]
4|2591|Loss: 1.0034290552139282:  40%|████      | 2590/6470 [29:47<39:30,  1.64it/s]
4|2591|Loss: 1.0034290552139282:  40%|████      | 2591/6470 [29:48<41:54,  1.54it/s]
4|2592|Loss: 0.906288743019104:  40%|████      | 2591/6470 [29:48<41:54,  1.54it/s] 
4|2592|Loss: 0.906288743019104:  40%|████      | 2592/6470 [29:49<45:15,  1.43it/s]
4|2593|Loss: 0.7286803722381592:  40%|████      | 2592/6470 [29:49<45:15,  1.43it/s]
4|2593|Loss: 0.7286803722381592:  40%|████      | 2593/6470 [29:49<46:30,  1.39it/s]
4|2594|Loss: 0.8572188019752502:  40%|████      | 2593/6470 [29:50<46:30,  1.39it/s]
4|2594|Loss: 0.8572188019752502:  40%|████      | 2594/6470 [29:50<44:19,  1.46it/s]
4|2595|Loss: 0.7897096872329712:  40%|████      | 2594/6470 [29:50<44:19,  1.46it/s]
4|2595|Loss: 0.7897096872329712:  40%|████      | 2595/6470 [29:50<42:11,  1.53it/s]
4|2596|Loss: 0.8452712893486023:  40%|████      | 2595/6470 [29:51<42:11,  1.53it/s]
4|2596|Loss: 0.8452712893486023:  40%|████      | 2596/6470 [29:51<40:41,  1.59it/s]
4|2597|Loss: 0.833747148513794:  40%|████      | 2596/6470 [29:51<40:41,  1.59it/s] 
4|2597|Loss: 0.833747148513794:  40%|████      | 2597/6470 [29:52<42:03,  1.53it/s]
4|2598|Loss: 0.9071580767631531:  40%|████      | 2597/6470 [29:52<42:03,  1.53it/s]
4|2598|Loss: 0.9071580767631531:  40%|████      | 2598/6470 [29:53<44:41,  1.44it/s]
4|2599|Loss: 0.877157986164093:  40%|████      | 2598/6470 [29:53<44:41,  1.44it/s] 
4|2599|Loss: 0.877157986164093:  40%|████      | 2599/6470 [29:53<48:11,  1.34it/s]
4|2600|Loss: 0.7262440323829651:  40%|████      | 2599/6470 [29:54<48:11,  1.34it/s]
4|2600|Loss: 0.7262440323829651:  40%|████      | 2600/6470 [29:54<49:34,  1.30it/s]
4|2601|Loss: 0.8201925158500671:  40%|████      | 2600/6470 [29:55<49:34,  1.30it/s]
4|2601|Loss: 0.8201925158500671:  40%|████      | 2601/6470 [29:55<49:13,  1.31it/s]
4|2602|Loss: 0.8900036811828613:  40%|████      | 2601/6470 [29:55<49:13,  1.31it/s]
4|2602|Loss: 0.8900036811828613:  40%|████      | 2602/6470 [29:56<47:11,  1.37it/s]
4|2603|Loss: 0.9756745100021362:  40%|████      | 2602/6470 [29:56<47:11,  1.37it/s]
4|2603|Loss: 0.9756745100021362:  40%|████      | 2603/6470 [29:56<45:07,  1.43it/s]
4|2604|Loss: 0.8971748948097229:  40%|████      | 2603/6470 [29:57<45:07,  1.43it/s]
4|2604|Loss: 0.8971748948097229:  40%|████      | 2604/6470 [29:57<46:08,  1.40it/s]
4|2605|Loss: 0.9753994941711426:  40%|████      | 2604/6470 [29:57<46:08,  1.40it/s]
4|2605|Loss: 0.9753994941711426:  40%|████      | 2605/6470 [29:58<44:58,  1.43it/s]
4|2606|Loss: 0.7593100666999817:  40%|████      | 2605/6470 [29:58<44:58,  1.43it/s]
4|2606|Loss: 0.7593100666999817:  40%|████      | 2606/6470 [29:58<42:42,  1.51it/s]
4|2607|Loss: 0.8010984659194946:  40%|████      | 2606/6470 [29:59<42:42,  1.51it/s]
4|2607|Loss: 0.8010984659194946:  40%|████      | 2607/6470 [29:59<43:40,  1.47it/s]
4|2608|Loss: 1.0894856452941895:  40%|████      | 2607/6470 [29:59<43:40,  1.47it/s]
4|2608|Loss: 1.0894856452941895:  40%|████      | 2608/6470 [30:00<41:41,  1.54it/s]
4|2609|Loss: 0.7733575701713562:  40%|████      | 2608/6470 [30:00<41:41,  1.54it/s]
4|2609|Loss: 0.7733575701713562:  40%|████      | 2609/6470 [30:00<40:45,  1.58it/s]
4|2610|Loss: 0.7882952690124512:  40%|████      | 2609/6470 [30:01<40:45,  1.58it/s]
4|2610|Loss: 0.7882952690124512:  40%|████      | 2610/6470 [30:01<41:46,  1.54it/s]
4|2611|Loss: 0.844866156578064:  40%|████      | 2610/6470 [30:01<41:46,  1.54it/s] 
4|2611|Loss: 0.844866156578064:  40%|████      | 2611/6470 [30:01<40:06,  1.60it/s]
4|2612|Loss: 0.8931939005851746:  40%|████      | 2611/6470 [30:02<40:06,  1.60it/s]
4|2612|Loss: 0.8931939005851746:  40%|████      | 2612/6470 [30:02<41:22,  1.55it/s]
4|2613|Loss: 0.9063134789466858:  40%|████      | 2612/6470 [30:02<41:22,  1.55it/s]
4|2613|Loss: 0.9063134789466858:  40%|████      | 2613/6470 [30:03<41:39,  1.54it/s]
4|2614|Loss: 0.7232816219329834:  40%|████      | 2613/6470 [30:03<41:39,  1.54it/s]
4|2614|Loss: 0.7232816219329834:  40%|████      | 2614/6470 [30:04<44:19,  1.45it/s]
4|2615|Loss: 0.8929024934768677:  40%|████      | 2614/6470 [30:04<44:19,  1.45it/s]
4|2615|Loss: 0.8929024934768677:  40%|████      | 2615/6470 [30:04<43:47,  1.47it/s]
4|2616|Loss: 0.9042818546295166:  40%|████      | 2615/6470 [30:05<43:47,  1.47it/s]
4|2616|Loss: 0.9042818546295166:  40%|████      | 2616/6470 [30:05<47:16,  1.36it/s]
4|2617|Loss: 0.9050862193107605:  40%|████      | 2616/6470 [30:06<47:16,  1.36it/s]
4|2617|Loss: 0.9050862193107605:  40%|████      | 2617/6470 [30:06<50:11,  1.28it/s]
4|2618|Loss: 0.8035609126091003:  40%|████      | 2617/6470 [30:06<50:11,  1.28it/s]
4|2618|Loss: 0.8035609126091003:  40%|████      | 2618/6470 [30:07<48:09,  1.33it/s]
4|2619|Loss: 0.9913867115974426:  40%|████      | 2618/6470 [30:07<48:09,  1.33it/s]
4|2619|Loss: 0.9913867115974426:  40%|████      | 2619/6470 [30:07<45:54,  1.40it/s]
4|2620|Loss: 0.962490975856781:  40%|████      | 2619/6470 [30:08<45:54,  1.40it/s] 
4|2620|Loss: 0.962490975856781:  40%|████      | 2620/6470 [30:08<47:17,  1.36it/s]
4|2621|Loss: 0.8102064728736877:  40%|████      | 2620/6470 [30:08<47:17,  1.36it/s]
4|2621|Loss: 0.8102064728736877:  41%|████      | 2621/6470 [30:09<44:46,  1.43it/s]
4|2622|Loss: 0.7966203093528748:  41%|████      | 2621/6470 [30:09<44:46,  1.43it/s]
4|2622|Loss: 0.7966203093528748:  41%|████      | 2622/6470 [30:09<44:16,  1.45it/s]
4|2623|Loss: 0.843611478805542:  41%|████      | 2622/6470 [30:10<44:16,  1.45it/s] 
4|2623|Loss: 0.843611478805542:  41%|████      | 2623/6470 [30:10<45:04,  1.42it/s]
4|2624|Loss: 0.8220425248146057:  41%|████      | 2623/6470 [30:10<45:04,  1.42it/s]
4|2624|Loss: 0.8220425248146057:  41%|████      | 2624/6470 [30:11<41:21,  1.55it/s]
4|2625|Loss: 0.6617922186851501:  41%|████      | 2624/6470 [30:11<41:21,  1.55it/s]
4|2625|Loss: 0.6617922186851501:  41%|████      | 2625/6470 [30:11<39:26,  1.62it/s]
4|2626|Loss: 0.9541898965835571:  41%|████      | 2625/6470 [30:11<39:26,  1.62it/s]
4|2626|Loss: 0.9541898965835571:  41%|████      | 2626/6470 [30:12<37:41,  1.70it/s]
4|2627|Loss: 0.8697957396507263:  41%|████      | 2626/6470 [30:12<37:41,  1.70it/s]
4|2627|Loss: 0.8697957396507263:  41%|████      | 2627/6470 [30:12<39:27,  1.62it/s]
4|2628|Loss: 0.9055120348930359:  41%|████      | 2627/6470 [30:13<39:27,  1.62it/s]
4|2628|Loss: 0.9055120348930359:  41%|████      | 2628/6470 [30:13<42:01,  1.52it/s]
4|2629|Loss: 0.9208864569664001:  41%|████      | 2628/6470 [30:13<42:01,  1.52it/s]
4|2629|Loss: 0.9208864569664001:  41%|████      | 2629/6470 [30:14<41:27,  1.54it/s]
4|2630|Loss: 1.0648603439331055:  41%|████      | 2629/6470 [30:14<41:27,  1.54it/s]
4|2630|Loss: 1.0648603439331055:  41%|████      | 2630/6470 [30:14<43:25,  1.47it/s]
4|2631|Loss: 0.8991758823394775:  41%|████      | 2630/6470 [30:15<43:25,  1.47it/s]
4|2631|Loss: 0.8991758823394775:  41%|████      | 2631/6470 [30:15<41:52,  1.53it/s]
4|2632|Loss: 0.8168326616287231:  41%|████      | 2631/6470 [30:15<41:52,  1.53it/s]
4|2632|Loss: 0.8168326616287231:  41%|████      | 2632/6470 [30:16<42:36,  1.50it/s]
4|2633|Loss: 0.90645432472229:  41%|████      | 2632/6470 [30:16<42:36,  1.50it/s]  
4|2633|Loss: 0.90645432472229:  41%|████      | 2633/6470 [30:16<42:23,  1.51it/s]
4|2634|Loss: 0.8170623183250427:  41%|████      | 2633/6470 [30:17<42:23,  1.51it/s]
4|2634|Loss: 0.8170623183250427:  41%|████      | 2634/6470 [30:17<41:05,  1.56it/s]
4|2635|Loss: 0.8685286045074463:  41%|████      | 2634/6470 [30:17<41:05,  1.56it/s]
4|2635|Loss: 0.8685286045074463:  41%|████      | 2635/6470 [30:18<43:11,  1.48it/s]
4|2636|Loss: 0.948515772819519:  41%|████      | 2635/6470 [30:18<43:11,  1.48it/s] 
4|2636|Loss: 0.948515772819519:  41%|████      | 2636/6470 [30:18<42:52,  1.49it/s]
4|2637|Loss: 0.7449028491973877:  41%|████      | 2636/6470 [30:19<42:52,  1.49it/s]
4|2637|Loss: 0.7449028491973877:  41%|████      | 2637/6470 [30:19<43:43,  1.46it/s]
4|2638|Loss: 0.7670207619667053:  41%|████      | 2637/6470 [30:19<43:43,  1.46it/s]
4|2638|Loss: 0.7670207619667053:  41%|████      | 2638/6470 [30:20<40:51,  1.56it/s]
4|2639|Loss: 0.8382909893989563:  41%|████      | 2638/6470 [30:20<40:51,  1.56it/s]
4|2639|Loss: 0.8382909893989563:  41%|████      | 2639/6470 [30:20<40:03,  1.59it/s]
4|2640|Loss: 0.7450718283653259:  41%|████      | 2639/6470 [30:21<40:03,  1.59it/s]
4|2640|Loss: 0.7450718283653259:  41%|████      | 2640/6470 [30:21<44:04,  1.45it/s]
4|2641|Loss: 0.8147445917129517:  41%|████      | 2640/6470 [30:22<44:04,  1.45it/s]
4|2641|Loss: 0.8147445917129517:  41%|████      | 2641/6470 [30:22<45:04,  1.42it/s]
4|2642|Loss: 0.8798008561134338:  41%|████      | 2641/6470 [30:22<45:04,  1.42it/s]
4|2642|Loss: 0.8798008561134338:  41%|████      | 2642/6470 [30:23<46:26,  1.37it/s]
4|2643|Loss: 0.8956874012947083:  41%|████      | 2642/6470 [30:23<46:26,  1.37it/s]
4|2643|Loss: 0.8956874012947083:  41%|████      | 2643/6470 [30:23<48:12,  1.32it/s]
4|2644|Loss: 0.7773423194885254:  41%|████      | 2643/6470 [30:24<48:12,  1.32it/s]
4|2644|Loss: 0.7773423194885254:  41%|████      | 2644/6470 [30:24<46:25,  1.37it/s]
4|2645|Loss: 0.8602707982063293:  41%|████      | 2644/6470 [30:24<46:25,  1.37it/s]
4|2645|Loss: 0.8602707982063293:  41%|████      | 2645/6470 [30:25<41:49,  1.52it/s]
4|2646|Loss: 0.6972025036811829:  41%|████      | 2645/6470 [30:25<41:49,  1.52it/s]
4|2646|Loss: 0.6972025036811829:  41%|████      | 2646/6470 [30:25<44:19,  1.44it/s]
4|2647|Loss: 0.9061998724937439:  41%|████      | 2646/6470 [30:26<44:19,  1.44it/s]
4|2647|Loss: 0.9061998724937439:  41%|████      | 2647/6470 [30:26<44:00,  1.45it/s]
4|2648|Loss: 0.8877052068710327:  41%|████      | 2647/6470 [30:26<44:00,  1.45it/s]
4|2648|Loss: 0.8877052068710327:  41%|████      | 2648/6470 [30:27<40:55,  1.56it/s]
4|2649|Loss: 0.8150610327720642:  41%|████      | 2648/6470 [30:27<40:55,  1.56it/s]
4|2649|Loss: 0.8150610327720642:  41%|████      | 2649/6470 [30:27<41:12,  1.55it/s]
4|2650|Loss: 0.8376047611236572:  41%|████      | 2649/6470 [30:28<41:12,  1.55it/s]
4|2650|Loss: 0.8376047611236572:  41%|████      | 2650/6470 [30:28<42:48,  1.49it/s]
4|2651|Loss: 0.8090202808380127:  41%|████      | 2650/6470 [30:28<42:48,  1.49it/s]
4|2651|Loss: 0.8090202808380127:  41%|████      | 2651/6470 [30:29<40:52,  1.56it/s]
4|2652|Loss: 0.9441580772399902:  41%|████      | 2651/6470 [30:29<40:52,  1.56it/s]
4|2652|Loss: 0.9441580772399902:  41%|████      | 2652/6470 [30:29<36:14,  1.76it/s]
4|2653|Loss: 0.8442374467849731:  41%|████      | 2652/6470 [30:29<36:14,  1.76it/s]
4|2653|Loss: 0.8442374467849731:  41%|████      | 2653/6470 [30:30<36:20,  1.75it/s]
4|2654|Loss: 0.7534660696983337:  41%|████      | 2653/6470 [30:30<36:20,  1.75it/s]
4|2654|Loss: 0.7534660696983337:  41%|████      | 2654/6470 [30:30<37:13,  1.71it/s]
4|2655|Loss: 0.591178834438324:  41%|████      | 2654/6470 [30:31<37:13,  1.71it/s] 
4|2655|Loss: 0.591178834438324:  41%|████      | 2655/6470 [30:31<42:27,  1.50it/s]
4|2656|Loss: 0.8694398999214172:  41%|████      | 2655/6470 [30:31<42:27,  1.50it/s]
4|2656|Loss: 0.8694398999214172:  41%|████      | 2656/6470 [30:32<44:42,  1.42it/s]
4|2657|Loss: 0.9430749416351318:  41%|████      | 2656/6470 [30:32<44:42,  1.42it/s]
4|2657|Loss: 0.9430749416351318:  41%|████      | 2657/6470 [30:32<43:01,  1.48it/s]
4|2658|Loss: 0.9079700112342834:  41%|████      | 2657/6470 [30:33<43:01,  1.48it/s]
4|2658|Loss: 0.9079700112342834:  41%|████      | 2658/6470 [30:33<40:40,  1.56it/s]
4|2659|Loss: 1.129369854927063:  41%|████      | 2658/6470 [30:33<40:40,  1.56it/s] 
4|2659|Loss: 1.129369854927063:  41%|████      | 2659/6470 [30:34<40:20,  1.57it/s]
4|2660|Loss: 0.7276546955108643:  41%|████      | 2659/6470 [30:34<40:20,  1.57it/s]
4|2660|Loss: 0.7276546955108643:  41%|████      | 2660/6470 [30:34<41:29,  1.53it/s]
4|2661|Loss: 0.6849134564399719:  41%|████      | 2660/6470 [30:35<41:29,  1.53it/s]
4|2661|Loss: 0.6849134564399719:  41%|████      | 2661/6470 [30:35<41:08,  1.54it/s]
4|2662|Loss: 0.9138608574867249:  41%|████      | 2661/6470 [30:35<41:08,  1.54it/s]
4|2662|Loss: 0.9138608574867249:  41%|████      | 2662/6470 [30:36<42:34,  1.49it/s]
4|2663|Loss: 0.9035637378692627:  41%|████      | 2662/6470 [30:36<42:34,  1.49it/s]
4|2663|Loss: 0.9035637378692627:  41%|████      | 2663/6470 [30:36<44:45,  1.42it/s]
4|2664|Loss: 0.6999139785766602:  41%|████      | 2663/6470 [30:37<44:45,  1.42it/s]
4|2664|Loss: 0.6999139785766602:  41%|████      | 2664/6470 [30:37<47:24,  1.34it/s]
4|2665|Loss: 0.8256599307060242:  41%|████      | 2664/6470 [30:38<47:24,  1.34it/s]
4|2665|Loss: 0.8256599307060242:  41%|████      | 2665/6470 [30:38<47:51,  1.33it/s]
4|2666|Loss: 0.8738723993301392:  41%|████      | 2665/6470 [30:38<47:51,  1.33it/s]
4|2666|Loss: 0.8738723993301392:  41%|████      | 2666/6470 [30:39<47:32,  1.33it/s]
4|2667|Loss: 0.7946416735649109:  41%|████      | 2666/6470 [30:39<47:32,  1.33it/s]
4|2667|Loss: 0.7946416735649109:  41%|████      | 2667/6470 [30:40<47:31,  1.33it/s]
4|2668|Loss: 0.8098956346511841:  41%|████      | 2667/6470 [30:40<47:31,  1.33it/s]
4|2668|Loss: 0.8098956346511841:  41%|████      | 2668/6470 [30:40<48:41,  1.30it/s]
4|2669|Loss: 0.8407041430473328:  41%|████      | 2668/6470 [30:41<48:41,  1.30it/s]
4|2669|Loss: 0.8407041430473328:  41%|████▏     | 2669/6470 [30:41<47:05,  1.35it/s]
4|2670|Loss: 0.9057063460350037:  41%|████▏     | 2669/6470 [30:41<47:05,  1.35it/s]
4|2670|Loss: 0.9057063460350037:  41%|████▏     | 2670/6470 [30:42<45:08,  1.40it/s]
4|2671|Loss: 0.8928134441375732:  41%|████▏     | 2670/6470 [30:42<45:08,  1.40it/s]
4|2671|Loss: 0.8928134441375732:  41%|████▏     | 2671/6470 [30:42<41:46,  1.52it/s]
4|2672|Loss: 0.8961728811264038:  41%|████▏     | 2671/6470 [30:43<41:46,  1.52it/s]
4|2672|Loss: 0.8961728811264038:  41%|████▏     | 2672/6470 [30:43<43:52,  1.44it/s]
4|2673|Loss: 1.0125008821487427:  41%|████▏     | 2672/6470 [30:43<43:52,  1.44it/s]
4|2673|Loss: 1.0125008821487427:  41%|████▏     | 2673/6470 [30:44<43:40,  1.45it/s]
4|2674|Loss: 0.8295149207115173:  41%|████▏     | 2673/6470 [30:44<43:40,  1.45it/s]
4|2674|Loss: 0.8295149207115173:  41%|████▏     | 2674/6470 [30:44<44:01,  1.44it/s]
4|2675|Loss: 0.8558054566383362:  41%|████▏     | 2674/6470 [30:45<44:01,  1.44it/s]
4|2675|Loss: 0.8558054566383362:  41%|████▏     | 2675/6470 [30:45<43:12,  1.46it/s]
4|2676|Loss: 0.7776371240615845:  41%|████▏     | 2675/6470 [30:45<43:12,  1.46it/s]
4|2676|Loss: 0.7776371240615845:  41%|████▏     | 2676/6470 [30:46<40:55,  1.54it/s]
4|2677|Loss: 0.8438200354576111:  41%|████▏     | 2676/6470 [30:46<40:55,  1.54it/s]
4|2677|Loss: 0.8438200354576111:  41%|████▏     | 2677/6470 [30:46<42:45,  1.48it/s]
4|2678|Loss: 0.7904869914054871:  41%|████▏     | 2677/6470 [30:47<42:45,  1.48it/s]
4|2678|Loss: 0.7904869914054871:  41%|████▏     | 2678/6470 [30:47<44:20,  1.43it/s]
4|2679|Loss: 0.8628963828086853:  41%|████▏     | 2678/6470 [30:48<44:20,  1.43it/s]
4|2679|Loss: 0.8628963828086853:  41%|████▏     | 2679/6470 [30:48<45:16,  1.40it/s]
4|2680|Loss: 1.002294898033142:  41%|████▏     | 2679/6470 [30:48<45:16,  1.40it/s] 
4|2680|Loss: 1.002294898033142:  41%|████▏     | 2680/6470 [30:49<45:15,  1.40it/s]
4|2681|Loss: 0.7932887077331543:  41%|████▏     | 2680/6470 [30:49<45:15,  1.40it/s]
4|2681|Loss: 0.7932887077331543:  41%|████▏     | 2681/6470 [30:49<45:40,  1.38it/s]
4|2682|Loss: 0.8763391375541687:  41%|████▏     | 2681/6470 [30:50<45:40,  1.38it/s]
4|2682|Loss: 0.8763391375541687:  41%|████▏     | 2682/6470 [30:50<43:39,  1.45it/s]
4|2683|Loss: 0.7308701276779175:  41%|████▏     | 2682/6470 [30:50<43:39,  1.45it/s]
4|2683|Loss: 0.7308701276779175:  41%|████▏     | 2683/6470 [30:51<43:50,  1.44it/s]
4|2684|Loss: 0.9635121822357178:  41%|████▏     | 2683/6470 [30:51<43:50,  1.44it/s]
4|2684|Loss: 0.9635121822357178:  41%|████▏     | 2684/6470 [30:51<41:28,  1.52it/s]
4|2685|Loss: 0.7590333819389343:  41%|████▏     | 2684/6470 [30:52<41:28,  1.52it/s]
4|2685|Loss: 0.7590333819389343:  41%|████▏     | 2685/6470 [30:52<42:25,  1.49it/s]
4|2686|Loss: 0.918323814868927:  41%|████▏     | 2685/6470 [30:52<42:25,  1.49it/s] 
4|2686|Loss: 0.918323814868927:  42%|████▏     | 2686/6470 [30:53<45:29,  1.39it/s]
4|2687|Loss: 0.8221240043640137:  42%|████▏     | 2686/6470 [30:53<45:29,  1.39it/s]
4|2687|Loss: 0.8221240043640137:  42%|████▏     | 2687/6470 [30:54<47:46,  1.32it/s]
4|2688|Loss: 0.8061175346374512:  42%|████▏     | 2687/6470 [30:54<47:46,  1.32it/s]
4|2688|Loss: 0.8061175346374512:  42%|████▏     | 2688/6470 [30:54<48:29,  1.30it/s]
4|2689|Loss: 0.7813634276390076:  42%|████▏     | 2688/6470 [30:55<48:29,  1.30it/s]
4|2689|Loss: 0.7813634276390076:  42%|████▏     | 2689/6470 [30:55<45:38,  1.38it/s]
4|2690|Loss: 0.8141845464706421:  42%|████▏     | 2689/6470 [30:55<45:38,  1.38it/s]
4|2690|Loss: 0.8141845464706421:  42%|████▏     | 2690/6470 [30:55<39:58,  1.58it/s]
4|2691|Loss: 1.0038896799087524:  42%|████▏     | 2690/6470 [30:56<39:58,  1.58it/s]
4|2691|Loss: 1.0038896799087524:  42%|████▏     | 2691/6470 [30:56<38:52,  1.62it/s]
4|2692|Loss: 0.874116837978363:  42%|████▏     | 2691/6470 [30:56<38:52,  1.62it/s] 
4|2692|Loss: 0.874116837978363:  42%|████▏     | 2692/6470 [30:57<40:32,  1.55it/s]
4|2693|Loss: 0.7299767732620239:  42%|████▏     | 2692/6470 [30:57<40:32,  1.55it/s]
4|2693|Loss: 0.7299767732620239:  42%|████▏     | 2693/6470 [30:57<39:01,  1.61it/s]
4|2694|Loss: 0.8373119235038757:  42%|████▏     | 2693/6470 [30:58<39:01,  1.61it/s]
4|2694|Loss: 0.8373119235038757:  42%|████▏     | 2694/6470 [30:58<38:32,  1.63it/s]
4|2695|Loss: 0.9804526567459106:  42%|████▏     | 2694/6470 [30:58<38:32,  1.63it/s]
4|2695|Loss: 0.9804526567459106:  42%|████▏     | 2695/6470 [30:59<39:55,  1.58it/s]
4|2696|Loss: 0.9107450246810913:  42%|████▏     | 2695/6470 [30:59<39:55,  1.58it/s]
4|2696|Loss: 0.9107450246810913:  42%|████▏     | 2696/6470 [30:59<42:35,  1.48it/s]
4|2697|Loss: 0.9222400784492493:  42%|████▏     | 2696/6470 [31:00<42:35,  1.48it/s]
4|2697|Loss: 0.9222400784492493:  42%|████▏     | 2697/6470 [31:00<45:03,  1.40it/s]
4|2698|Loss: 0.874335527420044:  42%|████▏     | 2697/6470 [31:01<45:03,  1.40it/s] 
4|2698|Loss: 0.874335527420044:  42%|████▏     | 2698/6470 [31:01<46:45,  1.34it/s]
4|2699|Loss: 0.7021035552024841:  42%|████▏     | 2698/6470 [31:01<46:45,  1.34it/s]
4|2699|Loss: 0.7021035552024841:  42%|████▏     | 2699/6470 [31:02<47:00,  1.34it/s]
4|2700|Loss: 0.7090786099433899:  42%|████▏     | 2699/6470 [31:02<47:00,  1.34it/s]
4|2700|Loss: 0.7090786099433899:  42%|████▏     | 2700/6470 [31:02<46:47,  1.34it/s]
4|2701|Loss: 0.8394060134887695:  42%|████▏     | 2700/6470 [31:03<46:47,  1.34it/s]
4|2701|Loss: 0.8394060134887695:  42%|████▏     | 2701/6470 [31:03<43:08,  1.46it/s]
4|2702|Loss: 0.7481182217597961:  42%|████▏     | 2701/6470 [31:03<43:08,  1.46it/s]
4|2702|Loss: 0.7481182217597961:  42%|████▏     | 2702/6470 [31:04<43:13,  1.45it/s]
4|2703|Loss: 0.7976328134536743:  42%|████▏     | 2702/6470 [31:04<43:13,  1.45it/s]
4|2703|Loss: 0.7976328134536743:  42%|████▏     | 2703/6470 [31:04<44:27,  1.41it/s]
4|2704|Loss: 0.9416763186454773:  42%|████▏     | 2703/6470 [31:05<44:27,  1.41it/s]
4|2704|Loss: 0.9416763186454773:  42%|████▏     | 2704/6470 [31:05<42:59,  1.46it/s]
4|2705|Loss: 0.6994132995605469:  42%|████▏     | 2704/6470 [31:05<42:59,  1.46it/s]
4|2705|Loss: 0.6994132995605469:  42%|████▏     | 2705/6470 [31:06<45:20,  1.38it/s]
4|2706|Loss: 0.858385443687439:  42%|████▏     | 2705/6470 [31:06<45:20,  1.38it/s] 
4|2706|Loss: 0.858385443687439:  42%|████▏     | 2706/6470 [31:06<42:47,  1.47it/s]
4|2707|Loss: 0.8509538173675537:  42%|████▏     | 2706/6470 [31:07<42:47,  1.47it/s]
4|2707|Loss: 0.8509538173675537:  42%|████▏     | 2707/6470 [31:07<39:49,  1.57it/s]
4|2708|Loss: 0.8593268990516663:  42%|████▏     | 2707/6470 [31:07<39:49,  1.57it/s]
4|2708|Loss: 0.8593268990516663:  42%|████▏     | 2708/6470 [31:08<41:48,  1.50it/s]
4|2709|Loss: 0.8567767143249512:  42%|████▏     | 2708/6470 [31:08<41:48,  1.50it/s]
4|2709|Loss: 0.8567767143249512:  42%|████▏     | 2709/6470 [31:09<44:49,  1.40it/s]
4|2710|Loss: 0.760430097579956:  42%|████▏     | 2709/6470 [31:09<44:49,  1.40it/s] 
4|2710|Loss: 0.760430097579956:  42%|████▏     | 2710/6470 [31:09<46:31,  1.35it/s]
4|2711|Loss: 0.822170615196228:  42%|████▏     | 2710/6470 [31:10<46:31,  1.35it/s]
4|2711|Loss: 0.822170615196228:  42%|████▏     | 2711/6470 [31:10<48:10,  1.30it/s]
4|2712|Loss: 0.8835933804512024:  42%|████▏     | 2711/6470 [31:11<48:10,  1.30it/s]
4|2712|Loss: 0.8835933804512024:  42%|████▏     | 2712/6470 [31:11<48:54,  1.28it/s]
4|2713|Loss: 0.826603353023529:  42%|████▏     | 2712/6470 [31:11<48:54,  1.28it/s] 
4|2713|Loss: 0.826603353023529:  42%|████▏     | 2713/6470 [31:12<45:57,  1.36it/s]
4|2714|Loss: 0.8933793306350708:  42%|████▏     | 2713/6470 [31:12<45:57,  1.36it/s]
4|2714|Loss: 0.8933793306350708:  42%|████▏     | 2714/6470 [31:12<44:04,  1.42it/s]
4|2715|Loss: 0.8270345330238342:  42%|████▏     | 2714/6470 [31:13<44:04,  1.42it/s]
4|2715|Loss: 0.8270345330238342:  42%|████▏     | 2715/6470 [31:13<42:52,  1.46it/s]
4|2716|Loss: 0.8563097715377808:  42%|████▏     | 2715/6470 [31:13<42:52,  1.46it/s]
4|2716|Loss: 0.8563097715377808:  42%|████▏     | 2716/6470 [31:14<43:48,  1.43it/s]
4|2717|Loss: 0.7716497778892517:  42%|████▏     | 2716/6470 [31:14<43:48,  1.43it/s]
4|2717|Loss: 0.7716497778892517:  42%|████▏     | 2717/6470 [31:15<49:26,  1.27it/s]
4|2718|Loss: 0.9903417825698853:  42%|████▏     | 2717/6470 [31:15<49:26,  1.27it/s]
4|2718|Loss: 0.9903417825698853:  42%|████▏     | 2718/6470 [31:15<47:27,  1.32it/s]
4|2719|Loss: 0.8571117520332336:  42%|████▏     | 2718/6470 [31:16<47:27,  1.32it/s]
4|2719|Loss: 0.8571117520332336:  42%|████▏     | 2719/6470 [31:16<44:53,  1.39it/s]
4|2720|Loss: 0.7069836854934692:  42%|████▏     | 2719/6470 [31:16<44:53,  1.39it/s]
4|2720|Loss: 0.7069836854934692:  42%|████▏     | 2720/6470 [31:17<43:25,  1.44it/s]
4|2721|Loss: 0.8668768405914307:  42%|████▏     | 2720/6470 [31:17<43:25,  1.44it/s]
4|2721|Loss: 0.8668768405914307:  42%|████▏     | 2721/6470 [31:17<43:22,  1.44it/s]
4|2722|Loss: 0.8593670129776001:  42%|████▏     | 2721/6470 [31:18<43:22,  1.44it/s]
4|2722|Loss: 0.8593670129776001:  42%|████▏     | 2722/6470 [31:18<42:41,  1.46it/s]
4|2723|Loss: 0.782269299030304:  42%|████▏     | 2722/6470 [31:18<42:41,  1.46it/s] 
4|2723|Loss: 0.782269299030304:  42%|████▏     | 2723/6470 [31:19<41:51,  1.49it/s]
4|2724|Loss: 0.8009978532791138:  42%|████▏     | 2723/6470 [31:19<41:51,  1.49it/s]
4|2724|Loss: 0.8009978532791138:  42%|████▏     | 2724/6470 [31:19<42:22,  1.47it/s]
4|2725|Loss: 0.8256916403770447:  42%|████▏     | 2724/6470 [31:20<42:22,  1.47it/s]
4|2725|Loss: 0.8256916403770447:  42%|████▏     | 2725/6470 [31:20<41:53,  1.49it/s]
4|2726|Loss: 1.0837085247039795:  42%|████▏     | 2725/6470 [31:20<41:53,  1.49it/s]
4|2726|Loss: 1.0837085247039795:  42%|████▏     | 2726/6470 [31:21<44:26,  1.40it/s]
4|2727|Loss: 0.7291275858879089:  42%|████▏     | 2726/6470 [31:21<44:26,  1.40it/s]
4|2727|Loss: 0.7291275858879089:  42%|████▏     | 2727/6470 [31:21<43:44,  1.43it/s]
4|2728|Loss: 1.088271975517273:  42%|████▏     | 2727/6470 [31:22<43:44,  1.43it/s] 
4|2728|Loss: 1.088271975517273:  42%|████▏     | 2728/6470 [31:22<43:28,  1.43it/s]
4|2729|Loss: 0.7197456359863281:  42%|████▏     | 2728/6470 [31:22<43:28,  1.43it/s]
4|2729|Loss: 0.7197456359863281:  42%|████▏     | 2729/6470 [31:23<39:49,  1.57it/s]
4|2730|Loss: 0.8700398206710815:  42%|████▏     | 2729/6470 [31:23<39:49,  1.57it/s]
4|2730|Loss: 0.8700398206710815:  42%|████▏     | 2730/6470 [31:23<39:53,  1.56it/s]
4|2731|Loss: 0.6844701766967773:  42%|████▏     | 2730/6470 [31:24<39:53,  1.56it/s]
4|2731|Loss: 0.6844701766967773:  42%|████▏     | 2731/6470 [31:24<38:45,  1.61it/s]
4|2732|Loss: 0.8592666983604431:  42%|████▏     | 2731/6470 [31:24<38:45,  1.61it/s]
4|2732|Loss: 0.8592666983604431:  42%|████▏     | 2732/6470 [31:24<35:56,  1.73it/s]
4|2733|Loss: 0.933268666267395:  42%|████▏     | 2732/6470 [31:25<35:56,  1.73it/s] 
4|2733|Loss: 0.933268666267395:  42%|████▏     | 2733/6470 [31:25<37:51,  1.65it/s]
4|2734|Loss: 0.7387155890464783:  42%|████▏     | 2733/6470 [31:25<37:51,  1.65it/s]
4|2734|Loss: 0.7387155890464783:  42%|████▏     | 2734/6470 [31:26<41:06,  1.51it/s]
4|2735|Loss: 0.9893833994865417:  42%|████▏     | 2734/6470 [31:26<41:06,  1.51it/s]
4|2735|Loss: 0.9893833994865417:  42%|████▏     | 2735/6470 [31:27<42:53,  1.45it/s]
4|2736|Loss: 1.0927094221115112:  42%|████▏     | 2735/6470 [31:27<42:53,  1.45it/s]
4|2736|Loss: 1.0927094221115112:  42%|████▏     | 2736/6470 [31:27<40:27,  1.54it/s]
4|2737|Loss: 0.8216795325279236:  42%|████▏     | 2736/6470 [31:27<40:27,  1.54it/s]
4|2737|Loss: 0.8216795325279236:  42%|████▏     | 2737/6470 [31:28<39:51,  1.56it/s]
4|2738|Loss: 0.7656626105308533:  42%|████▏     | 2737/6470 [31:28<39:51,  1.56it/s]
4|2738|Loss: 0.7656626105308533:  42%|████▏     | 2738/6470 [31:28<42:46,  1.45it/s]
4|2739|Loss: 0.956218957901001:  42%|████▏     | 2738/6470 [31:29<42:46,  1.45it/s] 
4|2739|Loss: 0.956218957901001:  42%|████▏     | 2739/6470 [31:29<42:26,  1.47it/s]
4|2740|Loss: 0.8775528073310852:  42%|████▏     | 2739/6470 [31:30<42:26,  1.47it/s]
4|2740|Loss: 0.8775528073310852:  42%|████▏     | 2740/6470 [31:30<43:00,  1.45it/s]
4|2741|Loss: 0.8949333429336548:  42%|████▏     | 2740/6470 [31:30<43:00,  1.45it/s]
4|2741|Loss: 0.8949333429336548:  42%|████▏     | 2741/6470 [31:31<44:02,  1.41it/s]
4|2742|Loss: 0.9449975490570068:  42%|████▏     | 2741/6470 [31:31<44:02,  1.41it/s]
4|2742|Loss: 0.9449975490570068:  42%|████▏     | 2742/6470 [31:31<43:19,  1.43it/s]
4|2743|Loss: 0.6897639036178589:  42%|████▏     | 2742/6470 [31:32<43:19,  1.43it/s]
4|2743|Loss: 0.6897639036178589:  42%|████▏     | 2743/6470 [31:32<46:20,  1.34it/s]
4|2744|Loss: 0.8393939137458801:  42%|████▏     | 2743/6470 [31:33<46:20,  1.34it/s]
4|2744|Loss: 0.8393939137458801:  42%|████▏     | 2744/6470 [31:33<48:48,  1.27it/s]
4|2745|Loss: 0.8314783573150635:  42%|████▏     | 2744/6470 [31:33<48:48,  1.27it/s]
4|2745|Loss: 0.8314783573150635:  42%|████▏     | 2745/6470 [31:34<48:25,  1.28it/s]
4|2746|Loss: 1.0106101036071777:  42%|████▏     | 2745/6470 [31:34<48:25,  1.28it/s]
4|2746|Loss: 1.0106101036071777:  42%|████▏     | 2746/6470 [31:35<47:36,  1.30it/s]
4|2747|Loss: 0.8685994148254395:  42%|████▏     | 2746/6470 [31:35<47:36,  1.30it/s]
4|2747|Loss: 0.8685994148254395:  42%|████▏     | 2747/6470 [31:35<43:37,  1.42it/s]
4|2748|Loss: 0.8793854117393494:  42%|████▏     | 2747/6470 [31:35<43:37,  1.42it/s]
4|2748|Loss: 0.8793854117393494:  42%|████▏     | 2748/6470 [31:36<44:08,  1.41it/s]
4|2749|Loss: 0.878246545791626:  42%|████▏     | 2748/6470 [31:36<44:08,  1.41it/s] 
4|2749|Loss: 0.878246545791626:  42%|████▏     | 2749/6470 [31:37<45:01,  1.38it/s]
4|2750|Loss: 0.8630394339561462:  42%|████▏     | 2749/6470 [31:37<45:01,  1.38it/s]
4|2750|Loss: 0.8630394339561462:  43%|████▎     | 2750/6470 [31:37<43:32,  1.42it/s]
4|2751|Loss: 0.8134967088699341:  43%|████▎     | 2750/6470 [31:38<43:32,  1.42it/s]
4|2751|Loss: 0.8134967088699341:  43%|████▎     | 2751/6470 [31:38<46:29,  1.33it/s]
4|2752|Loss: 0.9665976762771606:  43%|████▎     | 2751/6470 [31:39<46:29,  1.33it/s]
4|2752|Loss: 0.9665976762771606:  43%|████▎     | 2752/6470 [31:39<46:59,  1.32it/s]
4|2753|Loss: 0.9405809044837952:  43%|████▎     | 2752/6470 [31:39<46:59,  1.32it/s]
4|2753|Loss: 0.9405809044837952:  43%|████▎     | 2753/6470 [31:40<48:21,  1.28it/s]
4|2754|Loss: 0.775165319442749:  43%|████▎     | 2753/6470 [31:40<48:21,  1.28it/s] 
4|2754|Loss: 0.775165319442749:  43%|████▎     | 2754/6470 [31:40<46:25,  1.33it/s]
4|2755|Loss: 0.9700626730918884:  43%|████▎     | 2754/6470 [31:41<46:25,  1.33it/s]
4|2755|Loss: 0.9700626730918884:  43%|████▎     | 2755/6470 [31:41<44:18,  1.40it/s]
4|2756|Loss: 1.0026410818099976:  43%|████▎     | 2755/6470 [31:41<44:18,  1.40it/s]
4|2756|Loss: 1.0026410818099976:  43%|████▎     | 2756/6470 [31:42<41:57,  1.48it/s]
4|2757|Loss: 0.9629730582237244:  43%|████▎     | 2756/6470 [31:42<41:57,  1.48it/s]
4|2757|Loss: 0.9629730582237244:  43%|████▎     | 2757/6470 [31:42<42:21,  1.46it/s]
4|2758|Loss: 0.9660306572914124:  43%|████▎     | 2757/6470 [31:43<42:21,  1.46it/s]
4|2758|Loss: 0.9660306572914124:  43%|████▎     | 2758/6470 [31:43<45:02,  1.37it/s]
4|2759|Loss: 0.948849618434906:  43%|████▎     | 2758/6470 [31:44<45:02,  1.37it/s] 
4|2759|Loss: 0.948849618434906:  43%|████▎     | 2759/6470 [31:44<43:58,  1.41it/s]
4|2760|Loss: 0.9364660382270813:  43%|████▎     | 2759/6470 [31:44<43:58,  1.41it/s]
4|2760|Loss: 0.9364660382270813:  43%|████▎     | 2760/6470 [31:45<46:18,  1.34it/s]
4|2761|Loss: 0.904754638671875:  43%|████▎     | 2760/6470 [31:45<46:18,  1.34it/s] 
4|2761|Loss: 0.904754638671875:  43%|████▎     | 2761/6470 [31:45<43:39,  1.42it/s]
4|2762|Loss: 0.8037460446357727:  43%|████▎     | 2761/6470 [31:46<43:39,  1.42it/s]
4|2762|Loss: 0.8037460446357727:  43%|████▎     | 2762/6470 [31:46<43:38,  1.42it/s]
4|2763|Loss: 0.9041699767112732:  43%|████▎     | 2762/6470 [31:46<43:38,  1.42it/s]
4|2763|Loss: 0.9041699767112732:  43%|████▎     | 2763/6470 [31:47<42:39,  1.45it/s]
4|2764|Loss: 0.8712949156761169:  43%|████▎     | 2763/6470 [31:47<42:39,  1.45it/s]
4|2764|Loss: 0.8712949156761169:  43%|████▎     | 2764/6470 [31:47<43:05,  1.43it/s]
4|2765|Loss: 0.9424376487731934:  43%|████▎     | 2764/6470 [31:48<43:05,  1.43it/s]
4|2765|Loss: 0.9424376487731934:  43%|████▎     | 2765/6470 [31:48<45:35,  1.35it/s]
4|2766|Loss: 0.870049238204956:  43%|████▎     | 2765/6470 [31:49<45:35,  1.35it/s] 
4|2766|Loss: 0.870049238204956:  43%|████▎     | 2766/6470 [31:49<44:26,  1.39it/s]
4|2767|Loss: 0.8412116765975952:  43%|████▎     | 2766/6470 [31:49<44:26,  1.39it/s]
4|2767|Loss: 0.8412116765975952:  43%|████▎     | 2767/6470 [31:49<41:13,  1.50it/s]
4|2768|Loss: 0.8622733354568481:  43%|████▎     | 2767/6470 [31:50<41:13,  1.50it/s]
4|2768|Loss: 0.8622733354568481:  43%|████▎     | 2768/6470 [31:50<42:31,  1.45it/s]
4|2769|Loss: 0.8578030467033386:  43%|████▎     | 2768/6470 [31:51<42:31,  1.45it/s]
4|2769|Loss: 0.8578030467033386:  43%|████▎     | 2769/6470 [31:51<43:52,  1.41it/s]
4|2770|Loss: 0.941494882106781:  43%|████▎     | 2769/6470 [31:51<43:52,  1.41it/s] 
4|2770|Loss: 0.941494882106781:  43%|████▎     | 2770/6470 [31:51<41:04,  1.50it/s]
4|2771|Loss: 0.7981681823730469:  43%|████▎     | 2770/6470 [31:52<41:04,  1.50it/s]
4|2771|Loss: 0.7981681823730469:  43%|████▎     | 2771/6470 [31:52<39:41,  1.55it/s]
4|2772|Loss: 0.9019415974617004:  43%|████▎     | 2771/6470 [31:52<39:41,  1.55it/s]
4|2772|Loss: 0.9019415974617004:  43%|████▎     | 2772/6470 [31:53<36:49,  1.67it/s]
4|2773|Loss: 0.8105897903442383:  43%|████▎     | 2772/6470 [31:53<36:49,  1.67it/s]
4|2773|Loss: 0.8105897903442383:  43%|████▎     | 2773/6470 [31:53<40:23,  1.53it/s]
4|2774|Loss: 1.1228657960891724:  43%|████▎     | 2773/6470 [31:54<40:23,  1.53it/s]
4|2774|Loss: 1.1228657960891724:  43%|████▎     | 2774/6470 [31:54<40:57,  1.50it/s]
4|2775|Loss: 0.7106021046638489:  43%|████▎     | 2774/6470 [31:54<40:57,  1.50it/s]
4|2775|Loss: 0.7106021046638489:  43%|████▎     | 2775/6470 [31:55<40:10,  1.53it/s]
4|2776|Loss: 0.741682231426239:  43%|████▎     | 2775/6470 [31:55<40:10,  1.53it/s] 
4|2776|Loss: 0.741682231426239:  43%|████▎     | 2776/6470 [31:55<39:08,  1.57it/s]
4|2777|Loss: 0.9889843463897705:  43%|████▎     | 2776/6470 [31:56<39:08,  1.57it/s]
4|2777|Loss: 0.9889843463897705:  43%|████▎     | 2777/6470 [31:56<38:56,  1.58it/s]
4|2778|Loss: 0.8073187470436096:  43%|████▎     | 2777/6470 [31:56<38:56,  1.58it/s]
4|2778|Loss: 0.8073187470436096:  43%|████▎     | 2778/6470 [31:56<39:07,  1.57it/s]
4|2779|Loss: 0.8466177582740784:  43%|████▎     | 2778/6470 [31:57<39:07,  1.57it/s]
4|2779|Loss: 0.8466177582740784:  43%|████▎     | 2779/6470 [31:57<40:13,  1.53it/s]
4|2780|Loss: 0.761501133441925:  43%|████▎     | 2779/6470 [31:58<40:13,  1.53it/s] 
4|2780|Loss: 0.761501133441925:  43%|████▎     | 2780/6470 [31:58<38:29,  1.60it/s]
4|2781|Loss: 1.0142861604690552:  43%|████▎     | 2780/6470 [31:58<38:29,  1.60it/s]
4|2781|Loss: 1.0142861604690552:  43%|████▎     | 2781/6470 [31:58<37:02,  1.66it/s]
4|2782|Loss: 0.8983966708183289:  43%|████▎     | 2781/6470 [31:59<37:02,  1.66it/s]
4|2782|Loss: 0.8983966708183289:  43%|████▎     | 2782/6470 [31:59<38:16,  1.61it/s]
4|2783|Loss: 0.8836619853973389:  43%|████▎     | 2782/6470 [31:59<38:16,  1.61it/s]
4|2783|Loss: 0.8836619853973389:  43%|████▎     | 2783/6470 [32:00<41:33,  1.48it/s]
4|2784|Loss: 0.8790408372879028:  43%|████▎     | 2783/6470 [32:00<41:33,  1.48it/s]
4|2784|Loss: 0.8790408372879028:  43%|████▎     | 2784/6470 [32:01<43:44,  1.40it/s]
4|2785|Loss: 0.9768219590187073:  43%|████▎     | 2784/6470 [32:01<43:44,  1.40it/s]
4|2785|Loss: 0.9768219590187073:  43%|████▎     | 2785/6470 [32:01<43:56,  1.40it/s]
4|2786|Loss: 0.7413330674171448:  43%|████▎     | 2785/6470 [32:02<43:56,  1.40it/s]
4|2786|Loss: 0.7413330674171448:  43%|████▎     | 2786/6470 [32:02<45:08,  1.36it/s]
4|2787|Loss: 0.7135530710220337:  43%|████▎     | 2786/6470 [32:02<45:08,  1.36it/s]
4|2787|Loss: 0.7135530710220337:  43%|████▎     | 2787/6470 [32:03<44:07,  1.39it/s]
4|2788|Loss: 0.8374594449996948:  43%|████▎     | 2787/6470 [32:03<44:07,  1.39it/s]
4|2788|Loss: 0.8374594449996948:  43%|████▎     | 2788/6470 [32:03<43:31,  1.41it/s]
4|2789|Loss: 0.8225371837615967:  43%|████▎     | 2788/6470 [32:04<43:31,  1.41it/s]
4|2789|Loss: 0.8225371837615967:  43%|████▎     | 2789/6470 [32:04<42:03,  1.46it/s]
4|2790|Loss: 0.8960080146789551:  43%|████▎     | 2789/6470 [32:04<42:03,  1.46it/s]
4|2790|Loss: 0.8960080146789551:  43%|████▎     | 2790/6470 [32:05<40:12,  1.53it/s]
4|2791|Loss: 0.7370190620422363:  43%|████▎     | 2790/6470 [32:05<40:12,  1.53it/s]
4|2791|Loss: 0.7370190620422363:  43%|████▎     | 2791/6470 [32:05<39:15,  1.56it/s]
4|2792|Loss: 0.8232542872428894:  43%|████▎     | 2791/6470 [32:06<39:15,  1.56it/s]
4|2792|Loss: 0.8232542872428894:  43%|████▎     | 2792/6470 [32:06<40:24,  1.52it/s]
4|2793|Loss: 0.9820930361747742:  43%|████▎     | 2792/6470 [32:06<40:24,  1.52it/s]
4|2793|Loss: 0.9820930361747742:  43%|████▎     | 2793/6470 [32:07<43:12,  1.42it/s]
4|2794|Loss: 0.8808406591415405:  43%|████▎     | 2793/6470 [32:07<43:12,  1.42it/s]
4|2794|Loss: 0.8808406591415405:  43%|████▎     | 2794/6470 [32:08<43:46,  1.40it/s]
4|2795|Loss: 0.8928840160369873:  43%|████▎     | 2794/6470 [32:08<43:46,  1.40it/s]
4|2795|Loss: 0.8928840160369873:  43%|████▎     | 2795/6470 [32:08<43:32,  1.41it/s]
4|2796|Loss: 0.7475858330726624:  43%|████▎     | 2795/6470 [32:09<43:32,  1.41it/s]
4|2796|Loss: 0.7475858330726624:  43%|████▎     | 2796/6470 [32:09<40:30,  1.51it/s]
4|2797|Loss: 0.866552472114563:  43%|████▎     | 2796/6470 [32:09<40:30,  1.51it/s] 
4|2797|Loss: 0.866552472114563:  43%|████▎     | 2797/6470 [32:09<41:09,  1.49it/s]
4|2798|Loss: 0.7902401089668274:  43%|████▎     | 2797/6470 [32:10<41:09,  1.49it/s]
4|2798|Loss: 0.7902401089668274:  43%|████▎     | 2798/6470 [32:10<42:37,  1.44it/s]
4|2799|Loss: 0.8688929080963135:  43%|████▎     | 2798/6470 [32:11<42:37,  1.44it/s]
4|2799|Loss: 0.8688929080963135:  43%|████▎     | 2799/6470 [32:11<44:05,  1.39it/s]
4|2800|Loss: 1.0152300596237183:  43%|████▎     | 2799/6470 [32:11<44:05,  1.39it/s]
4|2800|Loss: 1.0152300596237183:  43%|████▎     | 2800/6470 [32:12<45:30,  1.34it/s]
4|2801|Loss: 0.8807397484779358:  43%|████▎     | 2800/6470 [32:12<45:30,  1.34it/s]
4|2801|Loss: 0.8807397484779358:  43%|████▎     | 2801/6470 [32:12<42:38,  1.43it/s]
4|2802|Loss: 0.8747648596763611:  43%|████▎     | 2801/6470 [32:13<42:38,  1.43it/s]
4|2802|Loss: 0.8747648596763611:  43%|████▎     | 2802/6470 [32:13<40:53,  1.50it/s]
4|2803|Loss: 0.8848884701728821:  43%|████▎     | 2802/6470 [32:13<40:53,  1.50it/s]
4|2803|Loss: 0.8848884701728821:  43%|████▎     | 2803/6470 [32:13<37:30,  1.63it/s]
4|2804|Loss: 0.8658369183540344:  43%|████▎     | 2803/6470 [32:14<37:30,  1.63it/s]
4|2804|Loss: 0.8658369183540344:  43%|████▎     | 2804/6470 [32:14<35:56,  1.70it/s]
4|2805|Loss: 0.9048359990119934:  43%|████▎     | 2804/6470 [32:14<35:56,  1.70it/s]
4|2805|Loss: 0.9048359990119934:  43%|████▎     | 2805/6470 [32:15<36:13,  1.69it/s]
4|2806|Loss: 0.9044755101203918:  43%|████▎     | 2805/6470 [32:15<36:13,  1.69it/s]
4|2806|Loss: 0.9044755101203918:  43%|████▎     | 2806/6470 [32:15<38:21,  1.59it/s]
4|2807|Loss: 0.5790828466415405:  43%|████▎     | 2806/6470 [32:16<38:21,  1.59it/s]
4|2807|Loss: 0.5790828466415405:  43%|████▎     | 2807/6470 [32:16<43:26,  1.41it/s]
4|2808|Loss: 0.8672373294830322:  43%|████▎     | 2807/6470 [32:17<43:26,  1.41it/s]
4|2808|Loss: 0.8672373294830322:  43%|████▎     | 2808/6470 [32:17<43:03,  1.42it/s]
4|2809|Loss: 0.803973913192749:  43%|████▎     | 2808/6470 [32:17<43:03,  1.42it/s] 
4|2809|Loss: 0.803973913192749:  43%|████▎     | 2809/6470 [32:17<39:51,  1.53it/s]
4|2810|Loss: 0.7699677348136902:  43%|████▎     | 2809/6470 [32:18<39:51,  1.53it/s]
4|2810|Loss: 0.7699677348136902:  43%|████▎     | 2810/6470 [32:18<40:43,  1.50it/s]
4|2811|Loss: 0.7264534831047058:  43%|████▎     | 2810/6470 [32:19<40:43,  1.50it/s]
4|2811|Loss: 0.7264534831047058:  43%|████▎     | 2811/6470 [32:19<42:07,  1.45it/s]
4|2812|Loss: 0.9617061614990234:  43%|████▎     | 2811/6470 [32:19<42:07,  1.45it/s]
4|2812|Loss: 0.9617061614990234:  43%|████▎     | 2812/6470 [32:20<43:14,  1.41it/s]
4|2813|Loss: 0.7274693846702576:  43%|████▎     | 2812/6470 [32:20<43:14,  1.41it/s]
4|2813|Loss: 0.7274693846702576:  43%|████▎     | 2813/6470 [32:20<45:28,  1.34it/s]
4|2814|Loss: 0.8074889779090881:  43%|████▎     | 2813/6470 [32:21<45:28,  1.34it/s]
4|2814|Loss: 0.8074889779090881:  43%|████▎     | 2814/6470 [32:21<40:39,  1.50it/s]
4|2815|Loss: 0.8858130574226379:  43%|████▎     | 2814/6470 [32:21<40:39,  1.50it/s]
4|2815|Loss: 0.8858130574226379:  44%|████▎     | 2815/6470 [32:22<40:26,  1.51it/s]
4|2816|Loss: 0.9399641752243042:  44%|████▎     | 2815/6470 [32:22<40:26,  1.51it/s]
4|2816|Loss: 0.9399641752243042:  44%|████▎     | 2816/6470 [32:22<39:37,  1.54it/s]
4|2817|Loss: 0.8048489093780518:  44%|████▎     | 2816/6470 [32:23<39:37,  1.54it/s]
4|2817|Loss: 0.8048489093780518:  44%|████▎     | 2817/6470 [32:23<37:20,  1.63it/s]
4|2818|Loss: 0.8220731019973755:  44%|████▎     | 2817/6470 [32:23<37:20,  1.63it/s]
4|2818|Loss: 0.8220731019973755:  44%|████▎     | 2818/6470 [32:23<38:16,  1.59it/s]
4|2819|Loss: 0.7225515246391296:  44%|████▎     | 2818/6470 [32:24<38:16,  1.59it/s]
4|2819|Loss: 0.7225515246391296:  44%|████▎     | 2819/6470 [32:24<40:05,  1.52it/s]
4|2820|Loss: 0.8362818956375122:  44%|████▎     | 2819/6470 [32:24<40:05,  1.52it/s]
4|2820|Loss: 0.8362818956375122:  44%|████▎     | 2820/6470 [32:25<38:42,  1.57it/s]
4|2821|Loss: 0.7939505577087402:  44%|████▎     | 2820/6470 [32:25<38:42,  1.57it/s]
4|2821|Loss: 0.7939505577087402:  44%|████▎     | 2821/6470 [32:25<39:21,  1.55it/s]
4|2822|Loss: 0.8255807757377625:  44%|████▎     | 2821/6470 [32:26<39:21,  1.55it/s]
4|2822|Loss: 0.8255807757377625:  44%|████▎     | 2822/6470 [32:26<41:03,  1.48it/s]
4|2823|Loss: 0.8213534355163574:  44%|████▎     | 2822/6470 [32:26<41:03,  1.48it/s]
4|2823|Loss: 0.8213534355163574:  44%|████▎     | 2823/6470 [32:27<38:06,  1.59it/s]
4|2824|Loss: 0.7756478786468506:  44%|████▎     | 2823/6470 [32:27<38:06,  1.59it/s]
4|2824|Loss: 0.7756478786468506:  44%|████▎     | 2824/6470 [32:27<39:01,  1.56it/s]
4|2825|Loss: 0.8825778365135193:  44%|████▎     | 2824/6470 [32:28<39:01,  1.56it/s]
4|2825|Loss: 0.8825778365135193:  44%|████▎     | 2825/6470 [32:28<39:05,  1.55it/s]
4|2826|Loss: 0.7572380900382996:  44%|████▎     | 2825/6470 [32:28<39:05,  1.55it/s]
4|2826|Loss: 0.7572380900382996:  44%|████▎     | 2826/6470 [32:29<37:35,  1.62it/s]
4|2827|Loss: 0.8604668378829956:  44%|████▎     | 2826/6470 [32:29<37:35,  1.62it/s]
4|2827|Loss: 0.8604668378829956:  44%|████▎     | 2827/6470 [32:29<39:14,  1.55it/s]
4|2828|Loss: 0.8745421767234802:  44%|████▎     | 2827/6470 [32:30<39:14,  1.55it/s]
4|2828|Loss: 0.8745421767234802:  44%|████▎     | 2828/6470 [32:30<41:09,  1.47it/s]
4|2829|Loss: 0.8924166560173035:  44%|████▎     | 2828/6470 [32:30<41:09,  1.47it/s]
4|2829|Loss: 0.8924166560173035:  44%|████▎     | 2829/6470 [32:31<40:10,  1.51it/s]
4|2830|Loss: 0.8169374465942383:  44%|████▎     | 2829/6470 [32:31<40:10,  1.51it/s]
4|2830|Loss: 0.8169374465942383:  44%|████▎     | 2830/6470 [32:31<42:31,  1.43it/s]
4|2831|Loss: 0.8974819779396057:  44%|████▎     | 2830/6470 [32:32<42:31,  1.43it/s]
4|2831|Loss: 0.8974819779396057:  44%|████▍     | 2831/6470 [32:32<41:50,  1.45it/s]
4|2832|Loss: 0.7434715628623962:  44%|████▍     | 2831/6470 [32:32<41:50,  1.45it/s]
4|2832|Loss: 0.7434715628623962:  44%|████▍     | 2832/6470 [32:33<44:17,  1.37it/s]
4|2833|Loss: 0.8021933436393738:  44%|████▍     | 2832/6470 [32:33<44:17,  1.37it/s]
4|2833|Loss: 0.8021933436393738:  44%|████▍     | 2833/6470 [32:34<42:01,  1.44it/s]
4|2834|Loss: 0.9347271919250488:  44%|████▍     | 2833/6470 [32:34<42:01,  1.44it/s]
4|2834|Loss: 0.9347271919250488:  44%|████▍     | 2834/6470 [32:34<42:17,  1.43it/s]
4|2835|Loss: 0.8293018937110901:  44%|████▍     | 2834/6470 [32:35<42:17,  1.43it/s]
4|2835|Loss: 0.8293018937110901:  44%|████▍     | 2835/6470 [32:35<44:20,  1.37it/s]
4|2836|Loss: 0.8117507696151733:  44%|████▍     | 2835/6470 [32:35<44:20,  1.37it/s]
4|2836|Loss: 0.8117507696151733:  44%|████▍     | 2836/6470 [32:36<44:52,  1.35it/s]
4|2837|Loss: 0.8084666132926941:  44%|████▍     | 2836/6470 [32:36<44:52,  1.35it/s]
4|2837|Loss: 0.8084666132926941:  44%|████▍     | 2837/6470 [32:37<47:26,  1.28it/s]
4|2838|Loss: 0.8667699098587036:  44%|████▍     | 2837/6470 [32:37<47:26,  1.28it/s]
4|2838|Loss: 0.8667699098587036:  44%|████▍     | 2838/6470 [32:37<45:53,  1.32it/s]
4|2839|Loss: 0.8295738697052002:  44%|████▍     | 2838/6470 [32:38<45:53,  1.32it/s]
4|2839|Loss: 0.8295738697052002:  44%|████▍     | 2839/6470 [32:38<42:56,  1.41it/s]
4|2840|Loss: 0.8277149200439453:  44%|████▍     | 2839/6470 [32:38<42:56,  1.41it/s]
4|2840|Loss: 0.8277149200439453:  44%|████▍     | 2840/6470 [32:39<41:23,  1.46it/s]
4|2841|Loss: 0.9013161063194275:  44%|████▍     | 2840/6470 [32:39<41:23,  1.46it/s]
4|2841|Loss: 0.9013161063194275:  44%|████▍     | 2841/6470 [32:39<39:43,  1.52it/s]
4|2842|Loss: 1.0005360841751099:  44%|████▍     | 2841/6470 [32:40<39:43,  1.52it/s]
4|2842|Loss: 1.0005360841751099:  44%|████▍     | 2842/6470 [32:40<42:19,  1.43it/s]
4|2843|Loss: 0.8591291308403015:  44%|████▍     | 2842/6470 [32:40<42:19,  1.43it/s]
4|2843|Loss: 0.8591291308403015:  44%|████▍     | 2843/6470 [32:41<43:20,  1.39it/s]
4|2844|Loss: 0.7378475666046143:  44%|████▍     | 2843/6470 [32:41<43:20,  1.39it/s]
4|2844|Loss: 0.7378475666046143:  44%|████▍     | 2844/6470 [32:42<44:27,  1.36it/s]
4|2845|Loss: 0.7855281829833984:  44%|████▍     | 2844/6470 [32:42<44:27,  1.36it/s]
4|2845|Loss: 0.7855281829833984:  44%|████▍     | 2845/6470 [32:42<43:19,  1.39it/s]
4|2846|Loss: 0.8179274797439575:  44%|████▍     | 2845/6470 [32:43<43:19,  1.39it/s]
4|2846|Loss: 0.8179274797439575:  44%|████▍     | 2846/6470 [32:43<41:51,  1.44it/s]
4|2847|Loss: 0.8342918157577515:  44%|████▍     | 2846/6470 [32:43<41:51,  1.44it/s]
4|2847|Loss: 0.8342918157577515:  44%|████▍     | 2847/6470 [32:44<42:37,  1.42it/s]
4|2848|Loss: 0.9320430755615234:  44%|████▍     | 2847/6470 [32:44<42:37,  1.42it/s]
4|2848|Loss: 0.9320430755615234:  44%|████▍     | 2848/6470 [32:44<44:08,  1.37it/s]
4|2849|Loss: 0.8735846877098083:  44%|████▍     | 2848/6470 [32:45<44:08,  1.37it/s]
4|2849|Loss: 0.8735846877098083:  44%|████▍     | 2849/6470 [32:45<46:55,  1.29it/s]
4|2850|Loss: 0.6638769507408142:  44%|████▍     | 2849/6470 [32:46<46:55,  1.29it/s]
4|2850|Loss: 0.6638769507408142:  44%|████▍     | 2850/6470 [32:46<44:29,  1.36it/s]
4|2851|Loss: 0.7427553534507751:  44%|████▍     | 2850/6470 [32:46<44:29,  1.36it/s]
4|2851|Loss: 0.7427553534507751:  44%|████▍     | 2851/6470 [32:47<43:27,  1.39it/s]
4|2852|Loss: 0.9477441310882568:  44%|████▍     | 2851/6470 [32:47<43:27,  1.39it/s]
4|2852|Loss: 0.9477441310882568:  44%|████▍     | 2852/6470 [32:47<43:53,  1.37it/s]
4|2853|Loss: 0.9231675863265991:  44%|████▍     | 2852/6470 [32:48<43:53,  1.37it/s]
4|2853|Loss: 0.9231675863265991:  44%|████▍     | 2853/6470 [32:48<43:48,  1.38it/s]
4|2854|Loss: 0.9845844507217407:  44%|████▍     | 2853/6470 [32:48<43:48,  1.38it/s]
4|2854|Loss: 0.9845844507217407:  44%|████▍     | 2854/6470 [32:49<43:48,  1.38it/s]
4|2855|Loss: 0.7785212993621826:  44%|████▍     | 2854/6470 [32:49<43:48,  1.38it/s]
4|2855|Loss: 0.7785212993621826:  44%|████▍     | 2855/6470 [32:49<42:34,  1.42it/s]
4|2856|Loss: 0.8016602396965027:  44%|████▍     | 2855/6470 [32:50<42:34,  1.42it/s]
4|2856|Loss: 0.8016602396965027:  44%|████▍     | 2856/6470 [32:50<42:35,  1.41it/s]
4|2857|Loss: 0.9303609728813171:  44%|████▍     | 2856/6470 [32:51<42:35,  1.41it/s]
4|2857|Loss: 0.9303609728813171:  44%|████▍     | 2857/6470 [32:51<42:38,  1.41it/s]
4|2858|Loss: 0.8026066422462463:  44%|████▍     | 2857/6470 [32:51<42:38,  1.41it/s]
4|2858|Loss: 0.8026066422462463:  44%|████▍     | 2858/6470 [32:52<42:54,  1.40it/s]
4|2859|Loss: 0.9078967571258545:  44%|████▍     | 2858/6470 [32:52<42:54,  1.40it/s]
4|2859|Loss: 0.9078967571258545:  44%|████▍     | 2859/6470 [32:52<44:59,  1.34it/s]
4|2860|Loss: 0.7496269941329956:  44%|████▍     | 2859/6470 [32:53<44:59,  1.34it/s]
4|2860|Loss: 0.7496269941329956:  44%|████▍     | 2860/6470 [32:53<43:34,  1.38it/s]
4|2861|Loss: 0.8275596499443054:  44%|████▍     | 2860/6470 [32:53<43:34,  1.38it/s]
4|2861|Loss: 0.8275596499443054:  44%|████▍     | 2861/6470 [32:54<44:54,  1.34it/s]
4|2862|Loss: 0.8771052956581116:  44%|████▍     | 2861/6470 [32:54<44:54,  1.34it/s]
4|2862|Loss: 0.8771052956581116:  44%|████▍     | 2862/6470 [32:55<44:43,  1.34it/s]
4|2863|Loss: 0.8776903748512268:  44%|████▍     | 2862/6470 [32:55<44:43,  1.34it/s]
4|2863|Loss: 0.8776903748512268:  44%|████▍     | 2863/6470 [32:55<42:07,  1.43it/s]
4|2864|Loss: 0.8785709738731384:  44%|████▍     | 2863/6470 [32:56<42:07,  1.43it/s]
4|2864|Loss: 0.8785709738731384:  44%|████▍     | 2864/6470 [32:56<40:04,  1.50it/s]
4|2865|Loss: 0.8712491989135742:  44%|████▍     | 2864/6470 [32:56<40:04,  1.50it/s]
4|2865|Loss: 0.8712491989135742:  44%|████▍     | 2865/6470 [32:56<39:25,  1.52it/s]
4|2866|Loss: 0.7052245736122131:  44%|████▍     | 2865/6470 [32:57<39:25,  1.52it/s]
4|2866|Loss: 0.7052245736122131:  44%|████▍     | 2866/6470 [32:57<42:10,  1.42it/s]
4|2867|Loss: 0.7832331657409668:  44%|████▍     | 2866/6470 [32:58<42:10,  1.42it/s]
4|2867|Loss: 0.7832331657409668:  44%|████▍     | 2867/6470 [32:58<41:35,  1.44it/s]
4|2868|Loss: 0.9771039485931396:  44%|████▍     | 2867/6470 [32:58<41:35,  1.44it/s]
4|2868|Loss: 0.9771039485931396:  44%|████▍     | 2868/6470 [32:59<43:10,  1.39it/s]
4|2869|Loss: 0.8058513402938843:  44%|████▍     | 2868/6470 [32:59<43:10,  1.39it/s]
4|2869|Loss: 0.8058513402938843:  44%|████▍     | 2869/6470 [32:59<43:20,  1.38it/s]
4|2870|Loss: 0.924401581287384:  44%|████▍     | 2869/6470 [33:00<43:20,  1.38it/s] 
4|2870|Loss: 0.924401581287384:  44%|████▍     | 2870/6470 [33:00<43:22,  1.38it/s]
4|2871|Loss: 0.7381061315536499:  44%|████▍     | 2870/6470 [33:01<43:22,  1.38it/s]
4|2871|Loss: 0.7381061315536499:  44%|████▍     | 2871/6470 [33:01<44:51,  1.34it/s]
4|2872|Loss: 0.8995988368988037:  44%|████▍     | 2871/6470 [33:01<44:51,  1.34it/s]
4|2872|Loss: 0.8995988368988037:  44%|████▍     | 2872/6470 [33:02<45:48,  1.31it/s]
4|2873|Loss: 0.9608500003814697:  44%|████▍     | 2872/6470 [33:02<45:48,  1.31it/s]
4|2873|Loss: 0.9608500003814697:  44%|████▍     | 2873/6470 [33:02<44:14,  1.35it/s]
4|2874|Loss: 0.907085120677948:  44%|████▍     | 2873/6470 [33:03<44:14,  1.35it/s] 
4|2874|Loss: 0.907085120677948:  44%|████▍     | 2874/6470 [33:03<42:24,  1.41it/s]
4|2875|Loss: 0.7882010340690613:  44%|████▍     | 2874/6470 [33:03<42:24,  1.41it/s]
4|2875|Loss: 0.7882010340690613:  44%|████▍     | 2875/6470 [33:04<41:06,  1.46it/s]
4|2876|Loss: 0.629085123538971:  44%|████▍     | 2875/6470 [33:04<41:06,  1.46it/s] 
4|2876|Loss: 0.629085123538971:  44%|████▍     | 2876/6470 [33:04<41:38,  1.44it/s]
4|2877|Loss: 0.8766711950302124:  44%|████▍     | 2876/6470 [33:05<41:38,  1.44it/s]
4|2877|Loss: 0.8766711950302124:  44%|████▍     | 2877/6470 [33:05<40:43,  1.47it/s]
4|2878|Loss: 0.9245201945304871:  44%|████▍     | 2877/6470 [33:05<40:43,  1.47it/s]
4|2878|Loss: 0.9245201945304871:  44%|████▍     | 2878/6470 [33:06<39:09,  1.53it/s]
4|2879|Loss: 0.7677004933357239:  44%|████▍     | 2878/6470 [33:06<39:09,  1.53it/s]
4|2879|Loss: 0.7677004933357239:  44%|████▍     | 2879/6470 [33:06<34:41,  1.72it/s]
4|2880|Loss: 0.8625683784484863:  44%|████▍     | 2879/6470 [33:06<34:41,  1.72it/s]
4|2880|Loss: 0.8625683784484863:  45%|████▍     | 2880/6470 [33:07<34:44,  1.72it/s]
4|2881|Loss: 0.8295807838439941:  45%|████▍     | 2880/6470 [33:07<34:44,  1.72it/s]
4|2881|Loss: 0.8295807838439941:  45%|████▍     | 2881/6470 [33:07<34:59,  1.71it/s]
4|2882|Loss: 0.8167261481285095:  45%|████▍     | 2881/6470 [33:08<34:59,  1.71it/s]
4|2882|Loss: 0.8167261481285095:  45%|████▍     | 2882/6470 [33:08<34:49,  1.72it/s]
4|2883|Loss: 0.7380146384239197:  45%|████▍     | 2882/6470 [33:08<34:49,  1.72it/s]
4|2883|Loss: 0.7380146384239197:  45%|████▍     | 2883/6470 [33:08<35:14,  1.70it/s]
4|2884|Loss: 0.7551263570785522:  45%|████▍     | 2883/6470 [33:09<35:14,  1.70it/s]
4|2884|Loss: 0.7551263570785522:  45%|████▍     | 2884/6470 [33:09<35:38,  1.68it/s]
4|2885|Loss: 0.7390180230140686:  45%|████▍     | 2884/6470 [33:09<35:38,  1.68it/s]
4|2885|Loss: 0.7390180230140686:  45%|████▍     | 2885/6470 [33:10<38:27,  1.55it/s]
4|2886|Loss: 0.9057059288024902:  45%|████▍     | 2885/6470 [33:10<38:27,  1.55it/s]
4|2886|Loss: 0.9057059288024902:  45%|████▍     | 2886/6470 [33:10<38:38,  1.55it/s]
4|2887|Loss: 1.005518913269043:  45%|████▍     | 2886/6470 [33:11<38:38,  1.55it/s] 
4|2887|Loss: 1.005518913269043:  45%|████▍     | 2887/6470 [33:11<37:27,  1.59it/s]
4|2888|Loss: 0.8595943450927734:  45%|████▍     | 2887/6470 [33:11<37:27,  1.59it/s]
4|2888|Loss: 0.8595943450927734:  45%|████▍     | 2888/6470 [33:12<40:14,  1.48it/s]
4|2889|Loss: 0.9389667510986328:  45%|████▍     | 2888/6470 [33:12<40:14,  1.48it/s]
4|2889|Loss: 0.9389667510986328:  45%|████▍     | 2889/6470 [33:12<40:21,  1.48it/s]
4|2890|Loss: 0.8676097393035889:  45%|████▍     | 2889/6470 [33:13<40:21,  1.48it/s]
4|2890|Loss: 0.8676097393035889:  45%|████▍     | 2890/6470 [33:13<39:26,  1.51it/s]
4|2891|Loss: 0.7866263389587402:  45%|████▍     | 2890/6470 [33:13<39:26,  1.51it/s]
4|2891|Loss: 0.7866263389587402:  45%|████▍     | 2891/6470 [33:14<38:39,  1.54it/s]
4|2892|Loss: 0.8766245245933533:  45%|████▍     | 2891/6470 [33:14<38:39,  1.54it/s]
4|2892|Loss: 0.8766245245933533:  45%|████▍     | 2892/6470 [33:14<37:07,  1.61it/s]
4|2893|Loss: 0.8220292329788208:  45%|████▍     | 2892/6470 [33:15<37:07,  1.61it/s]
4|2893|Loss: 0.8220292329788208:  45%|████▍     | 2893/6470 [33:15<39:16,  1.52it/s]
4|2894|Loss: 0.7959883213043213:  45%|████▍     | 2893/6470 [33:15<39:16,  1.52it/s]
4|2894|Loss: 0.7959883213043213:  45%|████▍     | 2894/6470 [33:16<39:36,  1.50it/s]
4|2895|Loss: 0.7505074143409729:  45%|████▍     | 2894/6470 [33:16<39:36,  1.50it/s]
4|2895|Loss: 0.7505074143409729:  45%|████▍     | 2895/6470 [33:16<41:29,  1.44it/s]
4|2896|Loss: 0.7741387486457825:  45%|████▍     | 2895/6470 [33:17<41:29,  1.44it/s]
4|2896|Loss: 0.7741387486457825:  45%|████▍     | 2896/6470 [33:17<43:30,  1.37it/s]
4|2897|Loss: 0.8925822377204895:  45%|████▍     | 2896/6470 [33:18<43:30,  1.37it/s]
4|2897|Loss: 0.8925822377204895:  45%|████▍     | 2897/6470 [33:18<41:51,  1.42it/s]
4|2898|Loss: 0.7967286109924316:  45%|████▍     | 2897/6470 [33:18<41:51,  1.42it/s]
4|2898|Loss: 0.7967286109924316:  45%|████▍     | 2898/6470 [33:18<38:55,  1.53it/s]
4|2899|Loss: 0.9489246606826782:  45%|████▍     | 2898/6470 [33:19<38:55,  1.53it/s]
4|2899|Loss: 0.9489246606826782:  45%|████▍     | 2899/6470 [33:19<39:51,  1.49it/s]
4|2900|Loss: 0.8739636540412903:  45%|████▍     | 2899/6470 [33:19<39:51,  1.49it/s]
4|2900|Loss: 0.8739636540412903:  45%|████▍     | 2900/6470 [33:20<35:54,  1.66it/s]
4|2901|Loss: 0.7860097885131836:  45%|████▍     | 2900/6470 [33:20<35:54,  1.66it/s]
4|2901|Loss: 0.7860097885131836:  45%|████▍     | 2901/6470 [33:20<38:41,  1.54it/s]
4|2902|Loss: 0.8542734384536743:  45%|████▍     | 2901/6470 [33:21<38:41,  1.54it/s]
4|2902|Loss: 0.8542734384536743:  45%|████▍     | 2902/6470 [33:21<38:10,  1.56it/s]
4|2903|Loss: 0.9512685537338257:  45%|████▍     | 2902/6470 [33:21<38:10,  1.56it/s]
4|2903|Loss: 0.9512685537338257:  45%|████▍     | 2903/6470 [33:22<39:06,  1.52it/s]
4|2904|Loss: 0.7599295973777771:  45%|████▍     | 2903/6470 [33:22<39:06,  1.52it/s]
4|2904|Loss: 0.7599295973777771:  45%|████▍     | 2904/6470 [33:23<42:36,  1.40it/s]
4|2905|Loss: 0.7858834266662598:  45%|████▍     | 2904/6470 [33:23<42:36,  1.40it/s]
4|2905|Loss: 0.7858834266662598:  45%|████▍     | 2905/6470 [33:23<44:37,  1.33it/s]
4|2906|Loss: 0.7982285022735596:  45%|████▍     | 2905/6470 [33:24<44:37,  1.33it/s]
4|2906|Loss: 0.7982285022735596:  45%|████▍     | 2906/6470 [33:24<43:01,  1.38it/s]
4|2907|Loss: 0.9500080943107605:  45%|████▍     | 2906/6470 [33:24<43:01,  1.38it/s]
4|2907|Loss: 0.9500080943107605:  45%|████▍     | 2907/6470 [33:25<42:46,  1.39it/s]
4|2908|Loss: 0.8339924812316895:  45%|████▍     | 2907/6470 [33:25<42:46,  1.39it/s]
4|2908|Loss: 0.8339924812316895:  45%|████▍     | 2908/6470 [33:26<43:58,  1.35it/s]
4|2909|Loss: 0.9610311985015869:  45%|████▍     | 2908/6470 [33:26<43:58,  1.35it/s]
4|2909|Loss: 0.9610311985015869:  45%|████▍     | 2909/6470 [33:26<44:18,  1.34it/s]
4|2910|Loss: 0.831539511680603:  45%|████▍     | 2909/6470 [33:27<44:18,  1.34it/s] 
4|2910|Loss: 0.831539511680603:  45%|████▍     | 2910/6470 [33:27<41:33,  1.43it/s]
4|2911|Loss: 0.8909163475036621:  45%|████▍     | 2910/6470 [33:27<41:33,  1.43it/s]
4|2911|Loss: 0.8909163475036621:  45%|████▍     | 2911/6470 [33:28<41:08,  1.44it/s]
4|2912|Loss: 0.845899760723114:  45%|████▍     | 2911/6470 [33:28<41:08,  1.44it/s] 
4|2912|Loss: 0.845899760723114:  45%|████▌     | 2912/6470 [33:28<43:11,  1.37it/s]
4|2913|Loss: 0.8927456736564636:  45%|████▌     | 2912/6470 [33:29<43:11,  1.37it/s]
4|2913|Loss: 0.8927456736564636:  45%|████▌     | 2913/6470 [33:29<42:27,  1.40it/s]
4|2914|Loss: 0.690100371837616:  45%|████▌     | 2913/6470 [33:29<42:27,  1.40it/s] 
4|2914|Loss: 0.690100371837616:  45%|████▌     | 2914/6470 [33:30<40:49,  1.45it/s]
4|2915|Loss: 0.9067685604095459:  45%|████▌     | 2914/6470 [33:30<40:49,  1.45it/s]
4|2915|Loss: 0.9067685604095459:  45%|████▌     | 2915/6470 [33:30<41:09,  1.44it/s]
4|2916|Loss: 0.820165753364563:  45%|████▌     | 2915/6470 [33:31<41:09,  1.44it/s] 
4|2916|Loss: 0.820165753364563:  45%|████▌     | 2916/6470 [33:31<42:43,  1.39it/s]
4|2917|Loss: 0.7702983021736145:  45%|████▌     | 2916/6470 [33:32<42:43,  1.39it/s]
4|2917|Loss: 0.7702983021736145:  45%|████▌     | 2917/6470 [33:32<44:32,  1.33it/s]
4|2918|Loss: 0.945527195930481:  45%|████▌     | 2917/6470 [33:32<44:32,  1.33it/s] 
4|2918|Loss: 0.945527195930481:  45%|████▌     | 2918/6470 [33:33<45:34,  1.30it/s]
4|2919|Loss: 0.8827894926071167:  45%|████▌     | 2918/6470 [33:33<45:34,  1.30it/s]
4|2919|Loss: 0.8827894926071167:  45%|████▌     | 2919/6470 [33:34<45:19,  1.31it/s]
4|2920|Loss: 0.8169025778770447:  45%|████▌     | 2919/6470 [33:34<45:19,  1.31it/s]
4|2920|Loss: 0.8169025778770447:  45%|████▌     | 2920/6470 [33:34<46:22,  1.28it/s]
4|2921|Loss: 0.8336226940155029:  45%|████▌     | 2920/6470 [33:35<46:22,  1.28it/s]
4|2921|Loss: 0.8336226940155029:  45%|████▌     | 2921/6470 [33:35<46:37,  1.27it/s]
4|2922|Loss: 0.8608562350273132:  45%|████▌     | 2921/6470 [33:36<46:37,  1.27it/s]
4|2922|Loss: 0.8608562350273132:  45%|████▌     | 2922/6470 [33:36<45:58,  1.29it/s]
4|2923|Loss: 0.7455742359161377:  45%|████▌     | 2922/6470 [33:36<45:58,  1.29it/s]
4|2923|Loss: 0.7455742359161377:  45%|████▌     | 2923/6470 [33:37<42:36,  1.39it/s]
4|2924|Loss: 0.8298683762550354:  45%|████▌     | 2923/6470 [33:37<42:36,  1.39it/s]
4|2924|Loss: 0.8298683762550354:  45%|████▌     | 2924/6470 [33:37<40:32,  1.46it/s]
4|2925|Loss: 0.8623422980308533:  45%|████▌     | 2924/6470 [33:38<40:32,  1.46it/s]
4|2925|Loss: 0.8623422980308533:  45%|████▌     | 2925/6470 [33:38<42:40,  1.38it/s]
4|2926|Loss: 0.8513919115066528:  45%|████▌     | 2925/6470 [33:38<42:40,  1.38it/s]
4|2926|Loss: 0.8513919115066528:  45%|████▌     | 2926/6470 [33:39<43:18,  1.36it/s]
4|2927|Loss: 0.7741838097572327:  45%|████▌     | 2926/6470 [33:39<43:18,  1.36it/s]
4|2927|Loss: 0.7741838097572327:  45%|████▌     | 2927/6470 [33:39<44:03,  1.34it/s]
4|2928|Loss: 0.8955402374267578:  45%|████▌     | 2927/6470 [33:40<44:03,  1.34it/s]
4|2928|Loss: 0.8955402374267578:  45%|████▌     | 2928/6470 [33:40<44:21,  1.33it/s]
4|2929|Loss: 0.8685778975486755:  45%|████▌     | 2928/6470 [33:41<44:21,  1.33it/s]
4|2929|Loss: 0.8685778975486755:  45%|████▌     | 2929/6470 [33:41<42:53,  1.38it/s]
4|2930|Loss: 0.6799058318138123:  45%|████▌     | 2929/6470 [33:41<42:53,  1.38it/s]
4|2930|Loss: 0.6799058318138123:  45%|████▌     | 2930/6470 [33:42<45:18,  1.30it/s]
4|2931|Loss: 0.8415924310684204:  45%|████▌     | 2930/6470 [33:42<45:18,  1.30it/s]
4|2931|Loss: 0.8415924310684204:  45%|████▌     | 2931/6470 [33:43<47:29,  1.24it/s]
4|2932|Loss: 0.9047327637672424:  45%|████▌     | 2931/6470 [33:43<47:29,  1.24it/s]
4|2932|Loss: 0.9047327637672424:  45%|████▌     | 2932/6470 [33:43<46:22,  1.27it/s]
4|2933|Loss: 0.853512704372406:  45%|████▌     | 2932/6470 [33:44<46:22,  1.27it/s] 
4|2933|Loss: 0.853512704372406:  45%|████▌     | 2933/6470 [33:44<44:49,  1.32it/s]
4|2934|Loss: 0.8333901166915894:  45%|████▌     | 2933/6470 [33:45<44:49,  1.32it/s]
4|2934|Loss: 0.8333901166915894:  45%|████▌     | 2934/6470 [33:45<42:58,  1.37it/s]
4|2935|Loss: 0.8542785048484802:  45%|████▌     | 2934/6470 [33:45<42:58,  1.37it/s]
4|2935|Loss: 0.8542785048484802:  45%|████▌     | 2935/6470 [33:45<38:45,  1.52it/s]
4|2936|Loss: 0.8841435313224792:  45%|████▌     | 2935/6470 [33:46<38:45,  1.52it/s]
4|2936|Loss: 0.8841435313224792:  45%|████▌     | 2936/6470 [33:46<39:02,  1.51it/s]
4|2937|Loss: 0.8688011169433594:  45%|████▌     | 2936/6470 [33:46<39:02,  1.51it/s]
4|2937|Loss: 0.8688011169433594:  45%|████▌     | 2937/6470 [33:47<40:25,  1.46it/s]
4|2938|Loss: 0.7282875180244446:  45%|████▌     | 2937/6470 [33:47<40:25,  1.46it/s]
4|2938|Loss: 0.7282875180244446:  45%|████▌     | 2938/6470 [33:47<39:36,  1.49it/s]
4|2939|Loss: 0.7722945213317871:  45%|████▌     | 2938/6470 [33:48<39:36,  1.49it/s]
4|2939|Loss: 0.7722945213317871:  45%|████▌     | 2939/6470 [33:48<39:49,  1.48it/s]
4|2940|Loss: 0.9588299989700317:  45%|████▌     | 2939/6470 [33:48<39:49,  1.48it/s]
4|2940|Loss: 0.9588299989700317:  45%|████▌     | 2940/6470 [33:49<41:38,  1.41it/s]
4|2941|Loss: 0.8384700417518616:  45%|████▌     | 2940/6470 [33:49<41:38,  1.41it/s]
4|2941|Loss: 0.8384700417518616:  45%|████▌     | 2941/6470 [33:49<40:51,  1.44it/s]
4|2942|Loss: 0.6517471075057983:  45%|████▌     | 2941/6470 [33:50<40:51,  1.44it/s]
4|2942|Loss: 0.6517471075057983:  45%|████▌     | 2942/6470 [33:50<38:50,  1.51it/s]
4|2943|Loss: 0.88958740234375:  45%|████▌     | 2942/6470 [33:50<38:50,  1.51it/s]  
4|2943|Loss: 0.88958740234375:  45%|████▌     | 2943/6470 [33:51<39:04,  1.50it/s]
4|2944|Loss: 0.920791506767273:  45%|████▌     | 2943/6470 [33:51<39:04,  1.50it/s]
4|2944|Loss: 0.920791506767273:  46%|████▌     | 2944/6470 [33:51<39:48,  1.48it/s]
4|2945|Loss: 0.8278760313987732:  46%|████▌     | 2944/6470 [33:52<39:48,  1.48it/s]
4|2945|Loss: 0.8278760313987732:  46%|████▌     | 2945/6470 [33:52<39:24,  1.49it/s]
4|2946|Loss: 0.888286292552948:  46%|████▌     | 2945/6470 [33:52<39:24,  1.49it/s] 
4|2946|Loss: 0.888286292552948:  46%|████▌     | 2946/6470 [33:53<40:27,  1.45it/s]
4|2947|Loss: 1.0668859481811523:  46%|████▌     | 2946/6470 [33:53<40:27,  1.45it/s]
4|2947|Loss: 1.0668859481811523:  46%|████▌     | 2947/6470 [33:54<42:14,  1.39it/s]
4|2948|Loss: 1.037858247756958:  46%|████▌     | 2947/6470 [33:54<42:14,  1.39it/s] 
4|2948|Loss: 1.037858247756958:  46%|████▌     | 2948/6470 [33:54<42:46,  1.37it/s]
4|2949|Loss: 0.9920596480369568:  46%|████▌     | 2948/6470 [33:55<42:46,  1.37it/s]
4|2949|Loss: 0.9920596480369568:  46%|████▌     | 2949/6470 [33:55<41:19,  1.42it/s]
4|2950|Loss: 0.7898473143577576:  46%|████▌     | 2949/6470 [33:55<41:19,  1.42it/s]
4|2950|Loss: 0.7898473143577576:  46%|████▌     | 2950/6470 [33:56<38:17,  1.53it/s]
4|2951|Loss: 0.8489124178886414:  46%|████▌     | 2950/6470 [33:56<38:17,  1.53it/s]
4|2951|Loss: 0.8489124178886414:  46%|████▌     | 2951/6470 [33:56<37:21,  1.57it/s]
4|2952|Loss: 0.8177356719970703:  46%|████▌     | 2951/6470 [33:56<37:21,  1.57it/s]
4|2952|Loss: 0.8177356719970703:  46%|████▌     | 2952/6470 [33:57<36:59,  1.58it/s]
4|2953|Loss: 0.8342192769050598:  46%|████▌     | 2952/6470 [33:57<36:59,  1.58it/s]
4|2953|Loss: 0.8342192769050598:  46%|████▌     | 2953/6470 [33:57<33:28,  1.75it/s]
4|2954|Loss: 0.9518558382987976:  46%|████▌     | 2953/6470 [33:57<33:28,  1.75it/s]
4|2954|Loss: 0.9518558382987976:  46%|████▌     | 2954/6470 [33:58<33:03,  1.77it/s]
4|2955|Loss: 0.8359385132789612:  46%|████▌     | 2954/6470 [33:58<33:03,  1.77it/s]
4|2955|Loss: 0.8359385132789612:  46%|████▌     | 2955/6470 [33:58<33:14,  1.76it/s]
4|2956|Loss: 0.7734492421150208:  46%|████▌     | 2955/6470 [33:59<33:14,  1.76it/s]
4|2956|Loss: 0.7734492421150208:  46%|████▌     | 2956/6470 [33:59<35:56,  1.63it/s]
4|2957|Loss: 0.8302319645881653:  46%|████▌     | 2956/6470 [33:59<35:56,  1.63it/s]
4|2957|Loss: 0.8302319645881653:  46%|████▌     | 2957/6470 [34:00<39:29,  1.48it/s]
4|2958|Loss: 0.8697474598884583:  46%|████▌     | 2957/6470 [34:00<39:29,  1.48it/s]
4|2958|Loss: 0.8697474598884583:  46%|████▌     | 2958/6470 [34:01<42:49,  1.37it/s]
4|2959|Loss: 0.8416253924369812:  46%|████▌     | 2958/6470 [34:01<42:49,  1.37it/s]
4|2959|Loss: 0.8416253924369812:  46%|████▌     | 2959/6470 [34:01<43:38,  1.34it/s]
4|2960|Loss: 0.955554723739624:  46%|████▌     | 2959/6470 [34:02<43:38,  1.34it/s] 
4|2960|Loss: 0.955554723739624:  46%|████▌     | 2960/6470 [34:02<42:18,  1.38it/s]
4|2961|Loss: 0.9817626476287842:  46%|████▌     | 2960/6470 [34:02<42:18,  1.38it/s]
4|2961|Loss: 0.9817626476287842:  46%|████▌     | 2961/6470 [34:03<39:00,  1.50it/s]
4|2962|Loss: 0.9843916893005371:  46%|████▌     | 2961/6470 [34:03<39:00,  1.50it/s]
4|2962|Loss: 0.9843916893005371:  46%|████▌     | 2962/6470 [34:03<38:57,  1.50it/s]
4|2963|Loss: 0.8501640558242798:  46%|████▌     | 2962/6470 [34:04<38:57,  1.50it/s]
4|2963|Loss: 0.8501640558242798:  46%|████▌     | 2963/6470 [34:04<41:10,  1.42it/s]
4|2964|Loss: 0.8937742114067078:  46%|████▌     | 2963/6470 [34:05<41:10,  1.42it/s]
4|2964|Loss: 0.8937742114067078:  46%|████▌     | 2964/6470 [34:05<43:04,  1.36it/s]
4|2965|Loss: 0.9233523011207581:  46%|████▌     | 2964/6470 [34:05<43:04,  1.36it/s]
4|2965|Loss: 0.9233523011207581:  46%|████▌     | 2965/6470 [34:06<43:19,  1.35it/s]
4|2966|Loss: 0.8457316756248474:  46%|████▌     | 2965/6470 [34:06<43:19,  1.35it/s]
4|2966|Loss: 0.8457316756248474:  46%|████▌     | 2966/6470 [34:06<43:18,  1.35it/s]
4|2967|Loss: 0.7448788285255432:  46%|████▌     | 2966/6470 [34:07<43:18,  1.35it/s]
4|2967|Loss: 0.7448788285255432:  46%|████▌     | 2967/6470 [34:07<44:17,  1.32it/s]
4|2968|Loss: 0.7632741332054138:  46%|████▌     | 2967/6470 [34:08<44:17,  1.32it/s]
4|2968|Loss: 0.7632741332054138:  46%|████▌     | 2968/6470 [34:08<46:10,  1.26it/s]
4|2969|Loss: 0.8986844420433044:  46%|████▌     | 2968/6470 [34:09<46:10,  1.26it/s]
4|2969|Loss: 0.8986844420433044:  46%|████▌     | 2969/6470 [34:09<46:02,  1.27it/s]
4|2970|Loss: 0.7779124975204468:  46%|████▌     | 2969/6470 [34:09<46:02,  1.27it/s]
4|2970|Loss: 0.7779124975204468:  46%|████▌     | 2970/6470 [34:10<44:42,  1.30it/s]
4|2971|Loss: 0.798572838306427:  46%|████▌     | 2970/6470 [34:10<44:42,  1.30it/s] 
4|2971|Loss: 0.798572838306427:  46%|████▌     | 2971/6470 [34:10<41:57,  1.39it/s]
4|2972|Loss: 0.8516960144042969:  46%|████▌     | 2971/6470 [34:11<41:57,  1.39it/s]
4|2972|Loss: 0.8516960144042969:  46%|████▌     | 2972/6470 [34:11<42:40,  1.37it/s]
4|2973|Loss: 0.8089843392372131:  46%|████▌     | 2972/6470 [34:11<42:40,  1.37it/s]
4|2973|Loss: 0.8089843392372131:  46%|████▌     | 2973/6470 [34:12<43:00,  1.36it/s]
4|2974|Loss: 0.6898893117904663:  46%|████▌     | 2973/6470 [34:12<43:00,  1.36it/s]
4|2974|Loss: 0.6898893117904663:  46%|████▌     | 2974/6470 [34:13<46:01,  1.27it/s]
4|2975|Loss: 0.900773823261261:  46%|████▌     | 2974/6470 [34:13<46:01,  1.27it/s] 
4|2975|Loss: 0.900773823261261:  46%|████▌     | 2975/6470 [34:13<46:06,  1.26it/s]
4|2976|Loss: 0.8939606547355652:  46%|████▌     | 2975/6470 [34:14<46:06,  1.26it/s]
4|2976|Loss: 0.8939606547355652:  46%|████▌     | 2976/6470 [34:14<43:39,  1.33it/s]
4|2977|Loss: 0.8574098348617554:  46%|████▌     | 2976/6470 [34:14<43:39,  1.33it/s]
4|2977|Loss: 0.8574098348617554:  46%|████▌     | 2977/6470 [34:15<42:34,  1.37it/s]
4|2978|Loss: 0.7966160774230957:  46%|████▌     | 2977/6470 [34:15<42:34,  1.37it/s]
4|2978|Loss: 0.7966160774230957:  46%|████▌     | 2978/6470 [34:15<41:02,  1.42it/s]
4|2979|Loss: 0.8689497709274292:  46%|████▌     | 2978/6470 [34:16<41:02,  1.42it/s]
4|2979|Loss: 0.8689497709274292:  46%|████▌     | 2979/6470 [34:16<41:16,  1.41it/s]
4|2980|Loss: 0.8217418193817139:  46%|████▌     | 2979/6470 [34:17<41:16,  1.41it/s]
4|2980|Loss: 0.8217418193817139:  46%|████▌     | 2980/6470 [34:17<40:24,  1.44it/s]
4|2981|Loss: 0.83758944272995:  46%|████▌     | 2980/6470 [34:17<40:24,  1.44it/s]  
4|2981|Loss: 0.83758944272995:  46%|████▌     | 2981/6470 [34:18<40:21,  1.44it/s]
4|2982|Loss: 0.835986316204071:  46%|████▌     | 2981/6470 [34:18<40:21,  1.44it/s]
4|2982|Loss: 0.835986316204071:  46%|████▌     | 2982/6470 [34:18<40:25,  1.44it/s]
4|2983|Loss: 0.8600596785545349:  46%|████▌     | 2982/6470 [34:19<40:25,  1.44it/s]
4|2983|Loss: 0.8600596785545349:  46%|████▌     | 2983/6470 [34:19<38:45,  1.50it/s]
4|2984|Loss: 0.9624643325805664:  46%|████▌     | 2983/6470 [34:19<38:45,  1.50it/s]
4|2984|Loss: 0.9624643325805664:  46%|████▌     | 2984/6470 [34:20<39:46,  1.46it/s]
4|2985|Loss: 0.7579944133758545:  46%|████▌     | 2984/6470 [34:20<39:46,  1.46it/s]
4|2985|Loss: 0.7579944133758545:  46%|████▌     | 2985/6470 [34:20<41:04,  1.41it/s]
4|2986|Loss: 0.9469441771507263:  46%|████▌     | 2985/6470 [34:21<41:04,  1.41it/s]
4|2986|Loss: 0.9469441771507263:  46%|████▌     | 2986/6470 [34:21<43:04,  1.35it/s]
4|2987|Loss: 0.7838510274887085:  46%|████▌     | 2986/6470 [34:22<43:04,  1.35it/s]
4|2987|Loss: 0.7838510274887085:  46%|████▌     | 2987/6470 [34:22<45:16,  1.28it/s]
4|2988|Loss: 0.6676851511001587:  46%|████▌     | 2987/6470 [34:23<45:16,  1.28it/s]
4|2988|Loss: 0.6676851511001587:  46%|████▌     | 2988/6470 [34:23<48:31,  1.20it/s]
4|2989|Loss: 0.8263030648231506:  46%|████▌     | 2988/6470 [34:23<48:31,  1.20it/s]
4|2989|Loss: 0.8263030648231506:  46%|████▌     | 2989/6470 [34:24<46:26,  1.25it/s]
4|2990|Loss: 0.8113517165184021:  46%|████▌     | 2989/6470 [34:24<46:26,  1.25it/s]
4|2990|Loss: 0.8113517165184021:  46%|████▌     | 2990/6470 [34:24<44:32,  1.30it/s]
4|2991|Loss: 0.9376996159553528:  46%|████▌     | 2990/6470 [34:25<44:32,  1.30it/s]
4|2991|Loss: 0.9376996159553528:  46%|████▌     | 2991/6470 [34:25<42:33,  1.36it/s]
4|2992|Loss: 0.7247190475463867:  46%|████▌     | 2991/6470 [34:25<42:33,  1.36it/s]
4|2992|Loss: 0.7247190475463867:  46%|████▌     | 2992/6470 [34:26<40:35,  1.43it/s]
4|2993|Loss: 0.8255023956298828:  46%|████▌     | 2992/6470 [34:26<40:35,  1.43it/s]
4|2993|Loss: 0.8255023956298828:  46%|████▋     | 2993/6470 [34:26<41:41,  1.39it/s]
4|2994|Loss: 0.7269138097763062:  46%|████▋     | 2993/6470 [34:27<41:41,  1.39it/s]
4|2994|Loss: 0.7269138097763062:  46%|████▋     | 2994/6470 [34:27<42:12,  1.37it/s]
4|2995|Loss: 0.9968114495277405:  46%|████▋     | 2994/6470 [34:28<42:12,  1.37it/s]
4|2995|Loss: 0.9968114495277405:  46%|████▋     | 2995/6470 [34:28<41:07,  1.41it/s]
4|2996|Loss: 0.7783024311065674:  46%|████▋     | 2995/6470 [34:28<41:07,  1.41it/s]
4|2996|Loss: 0.7783024311065674:  46%|████▋     | 2996/6470 [34:29<41:11,  1.41it/s]
4|2997|Loss: 0.81413334608078:  46%|████▋     | 2996/6470 [34:29<41:11,  1.41it/s]  
4|2997|Loss: 0.81413334608078:  46%|████▋     | 2997/6470 [34:29<39:23,  1.47it/s]
4|2998|Loss: 0.7751531600952148:  46%|████▋     | 2997/6470 [34:29<39:23,  1.47it/s]
4|2998|Loss: 0.7751531600952148:  46%|████▋     | 2998/6470 [34:30<38:09,  1.52it/s]
4|2999|Loss: 0.7174471020698547:  46%|████▋     | 2998/6470 [34:30<38:09,  1.52it/s]
4|2999|Loss: 0.7174471020698547:  46%|████▋     | 2999/6470 [34:30<36:00,  1.61it/s]
4|3000|Loss: 0.8639363646507263:  46%|████▋     | 2999/6470 [34:31<36:00,  1.61it/s]
4|3000|Loss: 0.8639363646507263:  46%|████▋     | 3000/6470 [34:31<34:57,  1.65it/s]
4|3001|Loss: 0.9099380373954773:  46%|████▋     | 3000/6470 [34:31<34:57,  1.65it/s]
4|3001|Loss: 0.9099380373954773:  46%|████▋     | 3001/6470 [34:31<33:46,  1.71it/s]
4|3002|Loss: 0.8681091666221619:  46%|████▋     | 3001/6470 [34:32<33:46,  1.71it/s]
4|3002|Loss: 0.8681091666221619:  46%|████▋     | 3002/6470 [34:32<35:05,  1.65it/s]
4|3003|Loss: 0.8739357590675354:  46%|████▋     | 3002/6470 [34:32<35:05,  1.65it/s]
4|3003|Loss: 0.8739357590675354:  46%|████▋     | 3003/6470 [34:33<39:10,  1.47it/s]
4|3004|Loss: 0.9613308906555176:  46%|████▋     | 3003/6470 [34:33<39:10,  1.47it/s]
4|3004|Loss: 0.9613308906555176:  46%|████▋     | 3004/6470 [34:34<39:08,  1.48it/s]
4|3005|Loss: 0.7711144089698792:  46%|████▋     | 3004/6470 [34:34<39:08,  1.48it/s]
4|3005|Loss: 0.7711144089698792:  46%|████▋     | 3005/6470 [34:34<39:17,  1.47it/s]
4|3006|Loss: 0.8013899922370911:  46%|████▋     | 3005/6470 [34:35<39:17,  1.47it/s]
4|3006|Loss: 0.8013899922370911:  46%|████▋     | 3006/6470 [34:35<40:14,  1.43it/s]
4|3007|Loss: 0.839569091796875:  46%|████▋     | 3006/6470 [34:35<40:14,  1.43it/s] 
4|3007|Loss: 0.839569091796875:  46%|████▋     | 3007/6470 [34:36<37:01,  1.56it/s]
4|3008|Loss: 0.9553045630455017:  46%|████▋     | 3007/6470 [34:36<37:01,  1.56it/s]
4|3008|Loss: 0.9553045630455017:  46%|████▋     | 3008/6470 [34:36<37:32,  1.54it/s]
4|3009|Loss: 0.7509669661521912:  46%|████▋     | 3008/6470 [34:37<37:32,  1.54it/s]
4|3009|Loss: 0.7509669661521912:  47%|████▋     | 3009/6470 [34:37<36:54,  1.56it/s]
4|3010|Loss: 0.7564026713371277:  47%|████▋     | 3009/6470 [34:37<36:54,  1.56it/s]
4|3010|Loss: 0.7564026713371277:  47%|████▋     | 3010/6470 [34:37<34:10,  1.69it/s]
4|3011|Loss: 0.7121766805648804:  47%|████▋     | 3010/6470 [34:38<34:10,  1.69it/s]
4|3011|Loss: 0.7121766805648804:  47%|████▋     | 3011/6470 [34:38<35:07,  1.64it/s]
4|3012|Loss: 0.7469057440757751:  47%|████▋     | 3011/6470 [34:38<35:07,  1.64it/s]
4|3012|Loss: 0.7469057440757751:  47%|████▋     | 3012/6470 [34:39<35:48,  1.61it/s]
4|3013|Loss: 0.8329811692237854:  47%|████▋     | 3012/6470 [34:39<35:48,  1.61it/s]
4|3013|Loss: 0.8329811692237854:  47%|████▋     | 3013/6470 [34:39<37:24,  1.54it/s]
4|3014|Loss: 0.8527758121490479:  47%|████▋     | 3013/6470 [34:40<37:24,  1.54it/s]
4|3014|Loss: 0.8527758121490479:  47%|████▋     | 3014/6470 [34:40<38:28,  1.50it/s]
4|3015|Loss: 0.9336625337600708:  47%|████▋     | 3014/6470 [34:40<38:28,  1.50it/s]
4|3015|Loss: 0.9336625337600708:  47%|████▋     | 3015/6470 [34:40<34:19,  1.68it/s]
4|3016|Loss: 0.9756202697753906:  47%|████▋     | 3015/6470 [34:41<34:19,  1.68it/s]
4|3016|Loss: 0.9756202697753906:  47%|████▋     | 3016/6470 [34:41<32:42,  1.76it/s]
4|3017|Loss: 0.942620038986206:  47%|████▋     | 3016/6470 [34:41<32:42,  1.76it/s] 
4|3017|Loss: 0.942620038986206:  47%|████▋     | 3017/6470 [34:41<31:43,  1.81it/s]
4|3018|Loss: 0.8213345408439636:  47%|████▋     | 3017/6470 [34:42<31:43,  1.81it/s]
4|3018|Loss: 0.8213345408439636:  47%|████▋     | 3018/6470 [34:42<31:50,  1.81it/s]
4|3019|Loss: 0.9051763415336609:  47%|████▋     | 3018/6470 [34:42<31:50,  1.81it/s]
4|3019|Loss: 0.9051763415336609:  47%|████▋     | 3019/6470 [34:43<34:30,  1.67it/s]
4|3020|Loss: 0.9027541875839233:  47%|████▋     | 3019/6470 [34:43<34:30,  1.67it/s]
4|3020|Loss: 0.9027541875839233:  47%|████▋     | 3020/6470 [34:43<34:22,  1.67it/s]
4|3021|Loss: 0.7387325167655945:  47%|████▋     | 3020/6470 [34:44<34:22,  1.67it/s]
4|3021|Loss: 0.7387325167655945:  47%|████▋     | 3021/6470 [34:44<36:07,  1.59it/s]
4|3022|Loss: 0.6752788424491882:  47%|████▋     | 3021/6470 [34:44<36:07,  1.59it/s]
4|3022|Loss: 0.6752788424491882:  47%|████▋     | 3022/6470 [34:45<38:02,  1.51it/s]
4|3023|Loss: 1.1178981065750122:  47%|████▋     | 3022/6470 [34:45<38:02,  1.51it/s]
4|3023|Loss: 1.1178981065750122:  47%|████▋     | 3023/6470 [34:45<38:03,  1.51it/s]
4|3024|Loss: 0.7196041941642761:  47%|████▋     | 3023/6470 [34:46<38:03,  1.51it/s]
4|3024|Loss: 0.7196041941642761:  47%|████▋     | 3024/6470 [34:46<37:23,  1.54it/s]
4|3025|Loss: 0.7676811218261719:  47%|████▋     | 3024/6470 [34:46<37:23,  1.54it/s]
4|3025|Loss: 0.7676811218261719:  47%|████▋     | 3025/6470 [34:47<34:36,  1.66it/s]
4|3026|Loss: 1.0694056749343872:  47%|████▋     | 3025/6470 [34:47<34:36,  1.66it/s]
4|3026|Loss: 1.0694056749343872:  47%|████▋     | 3026/6470 [34:47<38:13,  1.50it/s]
4|3027|Loss: 0.6488687992095947:  47%|████▋     | 3026/6470 [34:48<38:13,  1.50it/s]
4|3027|Loss: 0.6488687992095947:  47%|████▋     | 3027/6470 [34:48<37:30,  1.53it/s]
4|3028|Loss: 0.7573962211608887:  47%|████▋     | 3027/6470 [34:48<37:30,  1.53it/s]
4|3028|Loss: 0.7573962211608887:  47%|████▋     | 3028/6470 [34:49<39:08,  1.47it/s]
4|3029|Loss: 0.8278483748435974:  47%|████▋     | 3028/6470 [34:49<39:08,  1.47it/s]
4|3029|Loss: 0.8278483748435974:  47%|████▋     | 3029/6470 [34:50<41:36,  1.38it/s]
4|3030|Loss: 0.782589316368103:  47%|████▋     | 3029/6470 [34:50<41:36,  1.38it/s] 
4|3030|Loss: 0.782589316368103:  47%|████▋     | 3030/6470 [34:50<40:00,  1.43it/s]
4|3031|Loss: 0.872583270072937:  47%|████▋     | 3030/6470 [34:51<40:00,  1.43it/s]
4|3031|Loss: 0.872583270072937:  47%|████▋     | 3031/6470 [34:51<41:48,  1.37it/s]
4|3032|Loss: 0.919030487537384:  47%|████▋     | 3031/6470 [34:51<41:48,  1.37it/s]
4|3032|Loss: 0.919030487537384:  47%|████▋     | 3032/6470 [34:52<42:18,  1.35it/s]
4|3033|Loss: 0.8969086408615112:  47%|████▋     | 3032/6470 [34:52<42:18,  1.35it/s]
4|3033|Loss: 0.8969086408615112:  47%|████▋     | 3033/6470 [34:53<43:39,  1.31it/s]
4|3034|Loss: 0.8247503638267517:  47%|████▋     | 3033/6470 [34:53<43:39,  1.31it/s]
4|3034|Loss: 0.8247503638267517:  47%|████▋     | 3034/6470 [34:53<41:04,  1.39it/s]
4|3035|Loss: 0.7496980428695679:  47%|████▋     | 3034/6470 [34:54<41:04,  1.39it/s]
4|3035|Loss: 0.7496980428695679:  47%|████▋     | 3035/6470 [34:54<40:17,  1.42it/s]
4|3036|Loss: 0.9158408641815186:  47%|████▋     | 3035/6470 [34:54<40:17,  1.42it/s]
4|3036|Loss: 0.9158408641815186:  47%|████▋     | 3036/6470 [34:54<37:48,  1.51it/s]
4|3037|Loss: 0.8585296273231506:  47%|████▋     | 3036/6470 [34:55<37:48,  1.51it/s]
4|3037|Loss: 0.8585296273231506:  47%|████▋     | 3037/6470 [34:55<38:21,  1.49it/s]
4|3038|Loss: 0.8736211061477661:  47%|████▋     | 3037/6470 [34:55<38:21,  1.49it/s]
4|3038|Loss: 0.8736211061477661:  47%|████▋     | 3038/6470 [34:56<37:55,  1.51it/s]
4|3039|Loss: 0.7999679446220398:  47%|████▋     | 3038/6470 [34:56<37:55,  1.51it/s]
4|3039|Loss: 0.7999679446220398:  47%|████▋     | 3039/6470 [34:56<38:54,  1.47it/s]
4|3040|Loss: 0.7848322987556458:  47%|████▋     | 3039/6470 [34:57<38:54,  1.47it/s]
4|3040|Loss: 0.7848322987556458:  47%|████▋     | 3040/6470 [34:57<39:24,  1.45it/s]
4|3041|Loss: 0.932232677936554:  47%|████▋     | 3040/6470 [34:58<39:24,  1.45it/s] 
4|3041|Loss: 0.932232677936554:  47%|████▋     | 3041/6470 [34:58<40:03,  1.43it/s]
4|3042|Loss: 0.7400449514389038:  47%|████▋     | 3041/6470 [34:58<40:03,  1.43it/s]
4|3042|Loss: 0.7400449514389038:  47%|████▋     | 3042/6470 [34:58<35:34,  1.61it/s]
4|3043|Loss: 0.9469819664955139:  47%|████▋     | 3042/6470 [34:59<35:34,  1.61it/s]
4|3043|Loss: 0.9469819664955139:  47%|████▋     | 3043/6470 [34:59<34:47,  1.64it/s]
4|3044|Loss: 0.8281616568565369:  47%|████▋     | 3043/6470 [34:59<34:47,  1.64it/s]
4|3044|Loss: 0.8281616568565369:  47%|████▋     | 3044/6470 [34:59<34:20,  1.66it/s]
4|3045|Loss: 0.8260532021522522:  47%|████▋     | 3044/6470 [35:00<34:20,  1.66it/s]
4|3045|Loss: 0.8260532021522522:  47%|████▋     | 3045/6470 [35:00<36:01,  1.58it/s]
4|3046|Loss: 0.7641817927360535:  47%|████▋     | 3045/6470 [35:01<36:01,  1.58it/s]
4|3046|Loss: 0.7641817927360535:  47%|████▋     | 3046/6470 [35:01<37:46,  1.51it/s]
4|3047|Loss: 0.9353156685829163:  47%|████▋     | 3046/6470 [35:01<37:46,  1.51it/s]
4|3047|Loss: 0.9353156685829163:  47%|████▋     | 3047/6470 [35:02<37:19,  1.53it/s]
4|3048|Loss: 0.8167466521263123:  47%|████▋     | 3047/6470 [35:02<37:19,  1.53it/s]
4|3048|Loss: 0.8167466521263123:  47%|████▋     | 3048/6470 [35:02<36:02,  1.58it/s]
4|3049|Loss: 0.7628387808799744:  47%|████▋     | 3048/6470 [35:03<36:02,  1.58it/s]
4|3049|Loss: 0.7628387808799744:  47%|████▋     | 3049/6470 [35:03<38:00,  1.50it/s]
4|3050|Loss: 0.8594337701797485:  47%|████▋     | 3049/6470 [35:03<38:00,  1.50it/s]
4|3050|Loss: 0.8594337701797485:  47%|████▋     | 3050/6470 [35:04<39:36,  1.44it/s]
4|3051|Loss: 0.7378522753715515:  47%|████▋     | 3050/6470 [35:04<39:36,  1.44it/s]
4|3051|Loss: 0.7378522753715515:  47%|████▋     | 3051/6470 [35:04<39:11,  1.45it/s]
4|3052|Loss: 0.7361066341400146:  47%|████▋     | 3051/6470 [35:05<39:11,  1.45it/s]
4|3052|Loss: 0.7361066341400146:  47%|████▋     | 3052/6470 [35:05<34:16,  1.66it/s]
4|3053|Loss: 0.8526771068572998:  47%|████▋     | 3052/6470 [35:05<34:16,  1.66it/s]
4|3053|Loss: 0.8526771068572998:  47%|████▋     | 3053/6470 [35:05<33:24,  1.70it/s]
4|3054|Loss: 0.8658121228218079:  47%|████▋     | 3053/6470 [35:06<33:24,  1.70it/s]
4|3054|Loss: 0.8658121228218079:  47%|████▋     | 3054/6470 [35:06<36:50,  1.55it/s]
4|3055|Loss: 0.8597270846366882:  47%|████▋     | 3054/6470 [35:06<36:50,  1.55it/s]
4|3055|Loss: 0.8597270846366882:  47%|████▋     | 3055/6470 [35:07<35:53,  1.59it/s]
4|3056|Loss: 0.8804486393928528:  47%|████▋     | 3055/6470 [35:07<35:53,  1.59it/s]
4|3056|Loss: 0.8804486393928528:  47%|████▋     | 3056/6470 [35:07<33:32,  1.70it/s]
4|3057|Loss: 1.0227305889129639:  47%|████▋     | 3056/6470 [35:07<33:32,  1.70it/s]
4|3057|Loss: 1.0227305889129639:  47%|████▋     | 3057/6470 [35:08<33:53,  1.68it/s]
4|3058|Loss: 0.8106628656387329:  47%|████▋     | 3057/6470 [35:08<33:53,  1.68it/s]
4|3058|Loss: 0.8106628656387329:  47%|████▋     | 3058/6470 [35:09<36:40,  1.55it/s]
4|3059|Loss: 1.0216573476791382:  47%|████▋     | 3058/6470 [35:09<36:40,  1.55it/s]
4|3059|Loss: 1.0216573476791382:  47%|████▋     | 3059/6470 [35:09<37:10,  1.53it/s]
4|3060|Loss: 0.746041476726532:  47%|████▋     | 3059/6470 [35:10<37:10,  1.53it/s] 
4|3060|Loss: 0.746041476726532:  47%|████▋     | 3060/6470 [35:10<36:11,  1.57it/s]
4|3061|Loss: 0.8502535223960876:  47%|████▋     | 3060/6470 [35:10<36:11,  1.57it/s]
4|3061|Loss: 0.8502535223960876:  47%|████▋     | 3061/6470 [35:10<35:41,  1.59it/s]
4|3062|Loss: 0.8610579371452332:  47%|████▋     | 3061/6470 [35:11<35:41,  1.59it/s]
4|3062|Loss: 0.8610579371452332:  47%|████▋     | 3062/6470 [35:11<34:54,  1.63it/s]
4|3063|Loss: 0.865786075592041:  47%|████▋     | 3062/6470 [35:11<34:54,  1.63it/s] 
4|3063|Loss: 0.865786075592041:  47%|████▋     | 3063/6470 [35:12<38:10,  1.49it/s]
4|3064|Loss: 0.880277693271637:  47%|████▋     | 3063/6470 [35:12<38:10,  1.49it/s]
4|3064|Loss: 0.880277693271637:  47%|████▋     | 3064/6470 [35:13<40:57,  1.39it/s]
4|3065|Loss: 0.9046739935874939:  47%|████▋     | 3064/6470 [35:13<40:57,  1.39it/s]
4|3065|Loss: 0.9046739935874939:  47%|████▋     | 3065/6470 [35:13<38:35,  1.47it/s]
4|3066|Loss: 0.8593568205833435:  47%|████▋     | 3065/6470 [35:14<38:35,  1.47it/s]
4|3066|Loss: 0.8593568205833435:  47%|████▋     | 3066/6470 [35:14<38:41,  1.47it/s]
4|3067|Loss: 0.8268012404441833:  47%|████▋     | 3066/6470 [35:14<38:41,  1.47it/s]
4|3067|Loss: 0.8268012404441833:  47%|████▋     | 3067/6470 [35:14<34:52,  1.63it/s]
4|3068|Loss: 0.7912957668304443:  47%|████▋     | 3067/6470 [35:15<34:52,  1.63it/s]
4|3068|Loss: 0.7912957668304443:  47%|████▋     | 3068/6470 [35:15<34:18,  1.65it/s]
4|3069|Loss: 0.8587507605552673:  47%|████▋     | 3068/6470 [35:15<34:18,  1.65it/s]
4|3069|Loss: 0.8587507605552673:  47%|████▋     | 3069/6470 [35:16<35:35,  1.59it/s]
4|3070|Loss: 0.9203960299491882:  47%|████▋     | 3069/6470 [35:16<35:35,  1.59it/s]
4|3070|Loss: 0.9203960299491882:  47%|████▋     | 3070/6470 [35:16<33:48,  1.68it/s]
4|3071|Loss: 0.9450226426124573:  47%|████▋     | 3070/6470 [35:16<33:48,  1.68it/s]
4|3071|Loss: 0.9450226426124573:  47%|████▋     | 3071/6470 [35:17<33:59,  1.67it/s]
4|3072|Loss: 0.8527266383171082:  47%|████▋     | 3071/6470 [35:17<33:59,  1.67it/s]
4|3072|Loss: 0.8527266383171082:  47%|████▋     | 3072/6470 [35:17<33:59,  1.67it/s]
4|3073|Loss: 0.7706876993179321:  47%|████▋     | 3072/6470 [35:18<33:59,  1.67it/s]
4|3073|Loss: 0.7706876993179321:  47%|████▋     | 3073/6470 [35:18<36:26,  1.55it/s]
4|3074|Loss: 0.8154168128967285:  47%|████▋     | 3073/6470 [35:18<36:26,  1.55it/s]
4|3074|Loss: 0.8154168128967285:  48%|████▊     | 3074/6470 [35:19<36:42,  1.54it/s]
4|3075|Loss: 0.8928614854812622:  48%|████▊     | 3074/6470 [35:19<36:42,  1.54it/s]
4|3075|Loss: 0.8928614854812622:  48%|████▊     | 3075/6470 [35:19<36:30,  1.55it/s]
4|3076|Loss: 0.873189389705658:  48%|████▊     | 3075/6470 [35:20<36:30,  1.55it/s] 
4|3076|Loss: 0.873189389705658:  48%|████▊     | 3076/6470 [35:20<37:22,  1.51it/s]
4|3077|Loss: 0.8480650782585144:  48%|████▊     | 3076/6470 [35:20<37:22,  1.51it/s]
4|3077|Loss: 0.8480650782585144:  48%|████▊     | 3077/6470 [35:20<32:47,  1.72it/s]
4|3078|Loss: 0.9540652632713318:  48%|████▊     | 3077/6470 [35:21<32:47,  1.72it/s]
4|3078|Loss: 0.9540652632713318:  48%|████▊     | 3078/6470 [35:21<34:28,  1.64it/s]
4|3079|Loss: 0.8401398062705994:  48%|████▊     | 3078/6470 [35:22<34:28,  1.64it/s]
4|3079|Loss: 0.8401398062705994:  48%|████▊     | 3079/6470 [35:22<37:07,  1.52it/s]
4|3080|Loss: 0.9508559107780457:  48%|████▊     | 3079/6470 [35:22<37:07,  1.52it/s]
4|3080|Loss: 0.9508559107780457:  48%|████▊     | 3080/6470 [35:23<40:13,  1.40it/s]
4|3081|Loss: 0.8610957860946655:  48%|████▊     | 3080/6470 [35:23<40:13,  1.40it/s]
4|3081|Loss: 0.8610957860946655:  48%|████▊     | 3081/6470 [35:23<37:17,  1.51it/s]
4|3082|Loss: 0.744477391242981:  48%|████▊     | 3081/6470 [35:24<37:17,  1.51it/s] 
4|3082|Loss: 0.744477391242981:  48%|████▊     | 3082/6470 [35:24<37:43,  1.50it/s]
4|3083|Loss: 0.8978681564331055:  48%|████▊     | 3082/6470 [35:24<37:43,  1.50it/s]
4|3083|Loss: 0.8978681564331055:  48%|████▊     | 3083/6470 [35:25<37:45,  1.50it/s]
4|3084|Loss: 0.9101147055625916:  48%|████▊     | 3083/6470 [35:25<37:45,  1.50it/s]
4|3084|Loss: 0.9101147055625916:  48%|████▊     | 3084/6470 [35:25<36:49,  1.53it/s]
4|3085|Loss: 0.8843303918838501:  48%|████▊     | 3084/6470 [35:26<36:49,  1.53it/s]
4|3085|Loss: 0.8843303918838501:  48%|████▊     | 3085/6470 [35:26<39:25,  1.43it/s]
4|3086|Loss: 0.9796037673950195:  48%|████▊     | 3085/6470 [35:26<39:25,  1.43it/s]
4|3086|Loss: 0.9796037673950195:  48%|████▊     | 3086/6470 [35:27<39:20,  1.43it/s]
4|3087|Loss: 0.8082077503204346:  48%|████▊     | 3086/6470 [35:27<39:20,  1.43it/s]
4|3087|Loss: 0.8082077503204346:  48%|████▊     | 3087/6470 [35:27<38:07,  1.48it/s]
4|3088|Loss: 0.8845893740653992:  48%|████▊     | 3087/6470 [35:28<38:07,  1.48it/s]
4|3088|Loss: 0.8845893740653992:  48%|████▊     | 3088/6470 [35:28<38:56,  1.45it/s]
4|3089|Loss: 0.9171678423881531:  48%|████▊     | 3088/6470 [35:29<38:56,  1.45it/s]
4|3089|Loss: 0.9171678423881531:  48%|████▊     | 3089/6470 [35:29<40:06,  1.41it/s]
4|3090|Loss: 0.9291893839836121:  48%|████▊     | 3089/6470 [35:29<40:06,  1.41it/s]
4|3090|Loss: 0.9291893839836121:  48%|████▊     | 3090/6470 [35:30<40:44,  1.38it/s]
4|3091|Loss: 0.7845757007598877:  48%|████▊     | 3090/6470 [35:30<40:44,  1.38it/s]
4|3091|Loss: 0.7845757007598877:  48%|████▊     | 3091/6470 [35:30<41:10,  1.37it/s]
4|3092|Loss: 0.757977306842804:  48%|████▊     | 3091/6470 [35:31<41:10,  1.37it/s] 
4|3092|Loss: 0.757977306842804:  48%|████▊     | 3092/6470 [35:31<40:53,  1.38it/s]
4|3093|Loss: 0.9128460884094238:  48%|████▊     | 3092/6470 [35:31<40:53,  1.38it/s]
4|3093|Loss: 0.9128460884094238:  48%|████▊     | 3093/6470 [35:32<40:32,  1.39it/s]
4|3094|Loss: 0.9440573453903198:  48%|████▊     | 3093/6470 [35:32<40:32,  1.39it/s]
4|3094|Loss: 0.9440573453903198:  48%|████▊     | 3094/6470 [35:32<38:36,  1.46it/s]
4|3095|Loss: 0.9215968251228333:  48%|████▊     | 3094/6470 [35:33<38:36,  1.46it/s]
4|3095|Loss: 0.9215968251228333:  48%|████▊     | 3095/6470 [35:33<40:34,  1.39it/s]
4|3096|Loss: 0.665402889251709:  48%|████▊     | 3095/6470 [35:34<40:34,  1.39it/s] 
4|3096|Loss: 0.665402889251709:  48%|████▊     | 3096/6470 [35:34<38:31,  1.46it/s]
4|3097|Loss: 0.9087477326393127:  48%|████▊     | 3096/6470 [35:34<38:31,  1.46it/s]
4|3097|Loss: 0.9087477326393127:  48%|████▊     | 3097/6470 [35:34<35:50,  1.57it/s]
4|3098|Loss: 0.8295897245407104:  48%|████▊     | 3097/6470 [35:35<35:50,  1.57it/s]
4|3098|Loss: 0.8295897245407104:  48%|████▊     | 3098/6470 [35:35<35:13,  1.60it/s]
4|3099|Loss: 0.8481648564338684:  48%|████▊     | 3098/6470 [35:35<35:13,  1.60it/s]
4|3099|Loss: 0.8481648564338684:  48%|████▊     | 3099/6470 [35:36<36:36,  1.53it/s]
4|3100|Loss: 0.8541324734687805:  48%|████▊     | 3099/6470 [35:36<36:36,  1.53it/s]
4|3100|Loss: 0.8541324734687805:  48%|████▊     | 3100/6470 [35:36<36:50,  1.52it/s]
4|3101|Loss: 0.788773775100708:  48%|████▊     | 3100/6470 [35:37<36:50,  1.52it/s] 
4|3101|Loss: 0.788773775100708:  48%|████▊     | 3101/6470 [35:37<38:04,  1.47it/s]
4|3102|Loss: 0.8328900933265686:  48%|████▊     | 3101/6470 [35:37<38:04,  1.47it/s]
4|3102|Loss: 0.8328900933265686:  48%|████▊     | 3102/6470 [35:38<40:27,  1.39it/s]
4|3103|Loss: 0.7381132245063782:  48%|████▊     | 3102/6470 [35:38<40:27,  1.39it/s]
4|3103|Loss: 0.7381132245063782:  48%|████▊     | 3103/6470 [35:38<37:17,  1.50it/s]
4|3104|Loss: 0.9898390173912048:  48%|████▊     | 3103/6470 [35:39<37:17,  1.50it/s]
4|3104|Loss: 0.9898390173912048:  48%|████▊     | 3104/6470 [35:39<38:39,  1.45it/s]
4|3105|Loss: 0.9140828847885132:  48%|████▊     | 3104/6470 [35:40<38:39,  1.45it/s]
4|3105|Loss: 0.9140828847885132:  48%|████▊     | 3105/6470 [35:40<39:25,  1.42it/s]
4|3106|Loss: 0.8211818933486938:  48%|████▊     | 3105/6470 [35:40<39:25,  1.42it/s]
4|3106|Loss: 0.8211818933486938:  48%|████▊     | 3106/6470 [35:41<38:11,  1.47it/s]
4|3107|Loss: 0.8804680109024048:  48%|████▊     | 3106/6470 [35:41<38:11,  1.47it/s]
4|3107|Loss: 0.8804680109024048:  48%|████▊     | 3107/6470 [35:41<36:42,  1.53it/s]
4|3108|Loss: 0.8366358876228333:  48%|████▊     | 3107/6470 [35:41<36:42,  1.53it/s]
4|3108|Loss: 0.8366358876228333:  48%|████▊     | 3108/6470 [35:42<37:33,  1.49it/s]
4|3109|Loss: 0.817358672618866:  48%|████▊     | 3108/6470 [35:42<37:33,  1.49it/s] 
4|3109|Loss: 0.817358672618866:  48%|████▊     | 3109/6470 [35:42<37:03,  1.51it/s]
4|3110|Loss: 0.964709460735321:  48%|████▊     | 3109/6470 [35:43<37:03,  1.51it/s]
4|3110|Loss: 0.964709460735321:  48%|████▊     | 3110/6470 [35:43<37:22,  1.50it/s]
4|3111|Loss: 0.9059783220291138:  48%|████▊     | 3110/6470 [35:43<37:22,  1.50it/s]
4|3111|Loss: 0.9059783220291138:  48%|████▊     | 3111/6470 [35:44<34:36,  1.62it/s]
4|3112|Loss: 0.906390368938446:  48%|████▊     | 3111/6470 [35:44<34:36,  1.62it/s] 
4|3112|Loss: 0.906390368938446:  48%|████▊     | 3112/6470 [35:44<33:38,  1.66it/s]
4|3113|Loss: 0.7755308747291565:  48%|████▊     | 3112/6470 [35:45<33:38,  1.66it/s]
4|3113|Loss: 0.7755308747291565:  48%|████▊     | 3113/6470 [35:45<35:04,  1.60it/s]
4|3114|Loss: 0.7769559621810913:  48%|████▊     | 3113/6470 [35:45<35:04,  1.60it/s]
4|3114|Loss: 0.7769559621810913:  48%|████▊     | 3114/6470 [35:45<33:59,  1.65it/s]
4|3115|Loss: 0.8028562664985657:  48%|████▊     | 3114/6470 [35:46<33:59,  1.65it/s]
4|3115|Loss: 0.8028562664985657:  48%|████▊     | 3115/6470 [35:46<32:25,  1.72it/s]
4|3116|Loss: 0.8449306488037109:  48%|████▊     | 3115/6470 [35:46<32:25,  1.72it/s]
4|3116|Loss: 0.8449306488037109:  48%|████▊     | 3116/6470 [35:47<34:19,  1.63it/s]
4|3117|Loss: 0.9198228716850281:  48%|████▊     | 3116/6470 [35:47<34:19,  1.63it/s]
4|3117|Loss: 0.9198228716850281:  48%|████▊     | 3117/6470 [35:48<38:07,  1.47it/s]
4|3118|Loss: 0.9054644703865051:  48%|████▊     | 3117/6470 [35:48<38:07,  1.47it/s]
4|3118|Loss: 0.9054644703865051:  48%|████▊     | 3118/6470 [35:48<39:46,  1.40it/s]
4|3119|Loss: 0.7094411849975586:  48%|████▊     | 3118/6470 [35:49<39:46,  1.40it/s]
4|3119|Loss: 0.7094411849975586:  48%|████▊     | 3119/6470 [35:49<40:25,  1.38it/s]
4|3120|Loss: 0.9956381916999817:  48%|████▊     | 3119/6470 [35:49<40:25,  1.38it/s]
4|3120|Loss: 0.9956381916999817:  48%|████▊     | 3120/6470 [35:50<39:23,  1.42it/s]
4|3121|Loss: 0.754501223564148:  48%|████▊     | 3120/6470 [35:50<39:23,  1.42it/s] 
4|3121|Loss: 0.754501223564148:  48%|████▊     | 3121/6470 [35:51<41:23,  1.35it/s]
4|3122|Loss: 0.9496797323226929:  48%|████▊     | 3121/6470 [35:51<41:23,  1.35it/s]
4|3122|Loss: 0.9496797323226929:  48%|████▊     | 3122/6470 [35:51<39:23,  1.42it/s]
4|3123|Loss: 0.7147274613380432:  48%|████▊     | 3122/6470 [35:51<39:23,  1.42it/s]
4|3123|Loss: 0.7147274613380432:  48%|████▊     | 3123/6470 [35:52<37:07,  1.50it/s]
4|3124|Loss: 0.9284067749977112:  48%|████▊     | 3123/6470 [35:52<37:07,  1.50it/s]
4|3124|Loss: 0.9284067749977112:  48%|████▊     | 3124/6470 [35:53<39:21,  1.42it/s]
4|3125|Loss: 0.980836808681488:  48%|████▊     | 3124/6470 [35:53<39:21,  1.42it/s] 
4|3125|Loss: 0.980836808681488:  48%|████▊     | 3125/6470 [35:53<35:24,  1.57it/s]
4|3126|Loss: 0.956321656703949:  48%|████▊     | 3125/6470 [35:53<35:24,  1.57it/s]
4|3126|Loss: 0.956321656703949:  48%|████▊     | 3126/6470 [35:54<38:20,  1.45it/s]
4|3127|Loss: 0.747153103351593:  48%|████▊     | 3126/6470 [35:54<38:20,  1.45it/s]
4|3127|Loss: 0.747153103351593:  48%|████▊     | 3127/6470 [35:54<37:47,  1.47it/s]
4|3128|Loss: 1.0027289390563965:  48%|████▊     | 3127/6470 [35:55<37:47,  1.47it/s]
4|3128|Loss: 1.0027289390563965:  48%|████▊     | 3128/6470 [35:55<38:22,  1.45it/s]
4|3129|Loss: 0.8573580384254456:  48%|████▊     | 3128/6470 [35:56<38:22,  1.45it/s]
4|3129|Loss: 0.8573580384254456:  48%|████▊     | 3129/6470 [35:56<39:30,  1.41it/s]
4|3130|Loss: 0.9258915185928345:  48%|████▊     | 3129/6470 [35:56<39:30,  1.41it/s]
4|3130|Loss: 0.9258915185928345:  48%|████▊     | 3130/6470 [35:57<40:15,  1.38it/s]
4|3131|Loss: 0.7682569622993469:  48%|████▊     | 3130/6470 [35:57<40:15,  1.38it/s]
4|3131|Loss: 0.7682569622993469:  48%|████▊     | 3131/6470 [35:57<40:03,  1.39it/s]
4|3132|Loss: 0.9265590310096741:  48%|████▊     | 3131/6470 [35:58<40:03,  1.39it/s]
4|3132|Loss: 0.9265590310096741:  48%|████▊     | 3132/6470 [35:58<36:23,  1.53it/s]
4|3133|Loss: 0.7332105040550232:  48%|████▊     | 3132/6470 [35:58<36:23,  1.53it/s]
4|3133|Loss: 0.7332105040550232:  48%|████▊     | 3133/6470 [35:59<38:31,  1.44it/s]
4|3134|Loss: 0.950172483921051:  48%|████▊     | 3133/6470 [35:59<38:31,  1.44it/s] 
4|3134|Loss: 0.950172483921051:  48%|████▊     | 3134/6470 [36:00<41:01,  1.36it/s]
4|3135|Loss: 0.8180995583534241:  48%|████▊     | 3134/6470 [36:00<41:01,  1.36it/s]
4|3135|Loss: 0.8180995583534241:  48%|████▊     | 3135/6470 [36:00<39:09,  1.42it/s]
4|3136|Loss: 0.9076516628265381:  48%|████▊     | 3135/6470 [36:01<39:09,  1.42it/s]
4|3136|Loss: 0.9076516628265381:  48%|████▊     | 3136/6470 [36:01<39:23,  1.41it/s]
4|3137|Loss: 0.7511683106422424:  48%|████▊     | 3136/6470 [36:01<39:23,  1.41it/s]
4|3137|Loss: 0.7511683106422424:  48%|████▊     | 3137/6470 [36:01<36:59,  1.50it/s]
4|3138|Loss: 0.7648765444755554:  48%|████▊     | 3137/6470 [36:02<36:59,  1.50it/s]
4|3138|Loss: 0.7648765444755554:  49%|████▊     | 3138/6470 [36:02<35:47,  1.55it/s]
4|3139|Loss: 0.9871973991394043:  49%|████▊     | 3138/6470 [36:02<35:47,  1.55it/s]
4|3139|Loss: 0.9871973991394043:  49%|████▊     | 3139/6470 [36:03<38:32,  1.44it/s]
4|3140|Loss: 0.8796419501304626:  49%|████▊     | 3139/6470 [36:03<38:32,  1.44it/s]
4|3140|Loss: 0.8796419501304626:  49%|████▊     | 3140/6470 [36:04<39:53,  1.39it/s]
4|3141|Loss: 0.7894520163536072:  49%|████▊     | 3140/6470 [36:04<39:53,  1.39it/s]
4|3141|Loss: 0.7894520163536072:  49%|████▊     | 3141/6470 [36:04<42:06,  1.32it/s]
4|3142|Loss: 0.847733199596405:  49%|████▊     | 3141/6470 [36:05<42:06,  1.32it/s] 
4|3142|Loss: 0.847733199596405:  49%|████▊     | 3142/6470 [36:05<45:16,  1.23it/s]
4|3143|Loss: 0.6996661424636841:  49%|████▊     | 3142/6470 [36:06<45:16,  1.23it/s]
4|3143|Loss: 0.6996661424636841:  49%|████▊     | 3143/6470 [36:06<43:09,  1.28it/s]
4|3144|Loss: 0.7518172264099121:  49%|████▊     | 3143/6470 [36:06<43:09,  1.28it/s]
4|3144|Loss: 0.7518172264099121:  49%|████▊     | 3144/6470 [36:07<40:42,  1.36it/s]
4|3145|Loss: 0.7519165873527527:  49%|████▊     | 3144/6470 [36:07<40:42,  1.36it/s]
4|3145|Loss: 0.7519165873527527:  49%|████▊     | 3145/6470 [36:07<38:54,  1.42it/s]
4|3146|Loss: 0.8448566198348999:  49%|████▊     | 3145/6470 [36:08<38:54,  1.42it/s]
4|3146|Loss: 0.8448566198348999:  49%|████▊     | 3146/6470 [36:08<37:49,  1.46it/s]
4|3147|Loss: 0.8435980081558228:  49%|████▊     | 3146/6470 [36:08<37:49,  1.46it/s]
4|3147|Loss: 0.8435980081558228:  49%|████▊     | 3147/6470 [36:09<40:01,  1.38it/s]
4|3148|Loss: 0.7704911828041077:  49%|████▊     | 3147/6470 [36:09<40:01,  1.38it/s]
4|3148|Loss: 0.7704911828041077:  49%|████▊     | 3148/6470 [36:09<38:00,  1.46it/s]
4|3149|Loss: 0.873062789440155:  49%|████▊     | 3148/6470 [36:10<38:00,  1.46it/s] 
4|3149|Loss: 0.873062789440155:  49%|████▊     | 3149/6470 [36:10<37:40,  1.47it/s]
4|3150|Loss: 0.834511399269104:  49%|████▊     | 3149/6470 [36:10<37:40,  1.47it/s]
4|3150|Loss: 0.834511399269104:  49%|████▊     | 3150/6470 [36:11<36:27,  1.52it/s]
4|3151|Loss: 0.8675932884216309:  49%|████▊     | 3150/6470 [36:11<36:27,  1.52it/s]
4|3151|Loss: 0.8675932884216309:  49%|████▊     | 3151/6470 [36:11<38:06,  1.45it/s]
4|3152|Loss: 0.8844642639160156:  49%|████▊     | 3151/6470 [36:12<38:06,  1.45it/s]
4|3152|Loss: 0.8844642639160156:  49%|████▊     | 3152/6470 [36:12<36:08,  1.53it/s]
4|3153|Loss: 0.9045023918151855:  49%|████▊     | 3152/6470 [36:12<36:08,  1.53it/s]
4|3153|Loss: 0.9045023918151855:  49%|████▊     | 3153/6470 [36:13<36:56,  1.50it/s]
4|3154|Loss: 0.7857155799865723:  49%|████▊     | 3153/6470 [36:13<36:56,  1.50it/s]
4|3154|Loss: 0.7857155799865723:  49%|████▊     | 3154/6470 [36:13<36:49,  1.50it/s]
4|3155|Loss: 0.8214718103408813:  49%|████▊     | 3154/6470 [36:14<36:49,  1.50it/s]
4|3155|Loss: 0.8214718103408813:  49%|████▉     | 3155/6470 [36:14<38:27,  1.44it/s]
4|3156|Loss: 0.9617576599121094:  49%|████▉     | 3155/6470 [36:15<38:27,  1.44it/s]
4|3156|Loss: 0.9617576599121094:  49%|████▉     | 3156/6470 [36:15<40:05,  1.38it/s]
4|3157|Loss: 0.8289988040924072:  49%|████▉     | 3156/6470 [36:15<40:05,  1.38it/s]
4|3157|Loss: 0.8289988040924072:  49%|████▉     | 3157/6470 [36:16<38:57,  1.42it/s]
4|3158|Loss: 0.855518639087677:  49%|████▉     | 3157/6470 [36:16<38:57,  1.42it/s] 
4|3158|Loss: 0.855518639087677:  49%|████▉     | 3158/6470 [36:16<37:48,  1.46it/s]
4|3159|Loss: 0.7424394488334656:  49%|████▉     | 3158/6470 [36:17<37:48,  1.46it/s]
4|3159|Loss: 0.7424394488334656:  49%|████▉     | 3159/6470 [36:17<32:58,  1.67it/s]
4|3160|Loss: 0.8054604530334473:  49%|████▉     | 3159/6470 [36:17<32:58,  1.67it/s]
4|3160|Loss: 0.8054604530334473:  49%|████▉     | 3160/6470 [36:17<33:15,  1.66it/s]
4|3161|Loss: 0.774505078792572:  49%|████▉     | 3160/6470 [36:18<33:15,  1.66it/s] 
4|3161|Loss: 0.774505078792572:  49%|████▉     | 3161/6470 [36:18<35:21,  1.56it/s]
4|3162|Loss: 0.6267873048782349:  49%|████▉     | 3161/6470 [36:18<35:21,  1.56it/s]
4|3162|Loss: 0.6267873048782349:  49%|████▉     | 3162/6470 [36:19<39:53,  1.38it/s]
4|3163|Loss: 0.7541956901550293:  49%|████▉     | 3162/6470 [36:19<39:53,  1.38it/s]
4|3163|Loss: 0.7541956901550293:  49%|████▉     | 3163/6470 [36:20<41:10,  1.34it/s]
4|3164|Loss: 0.844079315662384:  49%|████▉     | 3163/6470 [36:20<41:10,  1.34it/s] 
4|3164|Loss: 0.844079315662384:  49%|████▉     | 3164/6470 [36:20<39:34,  1.39it/s]
4|3165|Loss: 0.8125748038291931:  49%|████▉     | 3164/6470 [36:21<39:34,  1.39it/s]
4|3165|Loss: 0.8125748038291931:  49%|████▉     | 3165/6470 [36:21<38:04,  1.45it/s]
4|3166|Loss: 0.897236704826355:  49%|████▉     | 3165/6470 [36:21<38:04,  1.45it/s] 
4|3166|Loss: 0.897236704826355:  49%|████▉     | 3166/6470 [36:22<38:37,  1.43it/s]
4|3167|Loss: 0.8661870956420898:  49%|████▉     | 3166/6470 [36:22<38:37,  1.43it/s]
4|3167|Loss: 0.8661870956420898:  49%|████▉     | 3167/6470 [36:22<37:59,  1.45it/s]
4|3168|Loss: 0.6824731826782227:  49%|████▉     | 3167/6470 [36:23<37:59,  1.45it/s]
4|3168|Loss: 0.6824731826782227:  49%|████▉     | 3168/6470 [36:23<35:23,  1.56it/s]
4|3169|Loss: 0.8109896779060364:  49%|████▉     | 3168/6470 [36:23<35:23,  1.56it/s]
4|3169|Loss: 0.8109896779060364:  49%|████▉     | 3169/6470 [36:23<34:04,  1.61it/s]
4|3170|Loss: 0.9112845063209534:  49%|████▉     | 3169/6470 [36:24<34:04,  1.61it/s]
4|3170|Loss: 0.9112845063209534:  49%|████▉     | 3170/6470 [36:24<33:59,  1.62it/s]
4|3171|Loss: 0.8766263723373413:  49%|████▉     | 3170/6470 [36:24<33:59,  1.62it/s]
4|3171|Loss: 0.8766263723373413:  49%|████▉     | 3171/6470 [36:25<35:47,  1.54it/s]
4|3172|Loss: 0.9441684484481812:  49%|████▉     | 3171/6470 [36:25<35:47,  1.54it/s]
4|3172|Loss: 0.9441684484481812:  49%|████▉     | 3172/6470 [36:26<38:44,  1.42it/s]
4|3173|Loss: 0.9032938480377197:  49%|████▉     | 3172/6470 [36:26<38:44,  1.42it/s]
4|3173|Loss: 0.9032938480377197:  49%|████▉     | 3173/6470 [36:26<39:39,  1.39it/s]
4|3174|Loss: 0.8687101602554321:  49%|████▉     | 3173/6470 [36:27<39:39,  1.39it/s]
4|3174|Loss: 0.8687101602554321:  49%|████▉     | 3174/6470 [36:27<40:43,  1.35it/s]
4|3175|Loss: 0.8794140815734863:  49%|████▉     | 3174/6470 [36:28<40:43,  1.35it/s]
4|3175|Loss: 0.8794140815734863:  49%|████▉     | 3175/6470 [36:28<42:21,  1.30it/s]
4|3176|Loss: 0.9070129990577698:  49%|████▉     | 3175/6470 [36:28<42:21,  1.30it/s]
4|3176|Loss: 0.9070129990577698:  49%|████▉     | 3176/6470 [36:29<41:56,  1.31it/s]
4|3177|Loss: 0.7887113094329834:  49%|████▉     | 3176/6470 [36:29<41:56,  1.31it/s]
4|3177|Loss: 0.7887113094329834:  49%|████▉     | 3177/6470 [36:29<39:52,  1.38it/s]
4|3178|Loss: 0.8122051954269409:  49%|████▉     | 3177/6470 [36:30<39:52,  1.38it/s]
4|3178|Loss: 0.8122051954269409:  49%|████▉     | 3178/6470 [36:30<40:21,  1.36it/s]
4|3179|Loss: 0.8785909414291382:  49%|████▉     | 3178/6470 [36:31<40:21,  1.36it/s]
4|3179|Loss: 0.8785909414291382:  49%|████▉     | 3179/6470 [36:31<41:52,  1.31it/s]
4|3180|Loss: 0.8879855275154114:  49%|████▉     | 3179/6470 [36:31<41:52,  1.31it/s]
4|3180|Loss: 0.8879855275154114:  49%|████▉     | 3180/6470 [36:32<40:32,  1.35it/s]
4|3181|Loss: 0.6701139807701111:  49%|████▉     | 3180/6470 [36:32<40:32,  1.35it/s]
4|3181|Loss: 0.6701139807701111:  49%|████▉     | 3181/6470 [36:32<41:27,  1.32it/s]
4|3182|Loss: 0.8718101382255554:  49%|████▉     | 3181/6470 [36:33<41:27,  1.32it/s]
4|3182|Loss: 0.8718101382255554:  49%|████▉     | 3182/6470 [36:33<42:18,  1.30it/s]
4|3183|Loss: 0.8981324434280396:  49%|████▉     | 3182/6470 [36:34<42:18,  1.30it/s]
4|3183|Loss: 0.8981324434280396:  49%|████▉     | 3183/6470 [36:34<40:24,  1.36it/s]
4|3184|Loss: 0.7959244251251221:  49%|████▉     | 3183/6470 [36:34<40:24,  1.36it/s]
4|3184|Loss: 0.7959244251251221:  49%|████▉     | 3184/6470 [36:35<42:58,  1.27it/s]
4|3185|Loss: 0.9428572654724121:  49%|████▉     | 3184/6470 [36:35<42:58,  1.27it/s]
4|3185|Loss: 0.9428572654724121:  49%|████▉     | 3185/6470 [36:36<41:07,  1.33it/s]
4|3186|Loss: 0.9022926092147827:  49%|████▉     | 3185/6470 [36:36<41:07,  1.33it/s]
4|3186|Loss: 0.9022926092147827:  49%|████▉     | 3186/6470 [36:36<41:37,  1.31it/s]
4|3187|Loss: 0.7841314077377319:  49%|████▉     | 3186/6470 [36:37<41:37,  1.31it/s]
4|3187|Loss: 0.7841314077377319:  49%|████▉     | 3187/6470 [36:37<38:50,  1.41it/s]
4|3188|Loss: 0.7166464328765869:  49%|████▉     | 3187/6470 [36:37<38:50,  1.41it/s]
4|3188|Loss: 0.7166464328765869:  49%|████▉     | 3188/6470 [36:38<38:27,  1.42it/s]
4|3189|Loss: 0.9181209206581116:  49%|████▉     | 3188/6470 [36:38<38:27,  1.42it/s]
4|3189|Loss: 0.9181209206581116:  49%|████▉     | 3189/6470 [36:38<40:15,  1.36it/s]
4|3190|Loss: 0.8206124305725098:  49%|████▉     | 3189/6470 [36:39<40:15,  1.36it/s]
4|3190|Loss: 0.8206124305725098:  49%|████▉     | 3190/6470 [36:39<40:04,  1.36it/s]
4|3191|Loss: 0.7044476866722107:  49%|████▉     | 3190/6470 [36:40<40:04,  1.36it/s]
4|3191|Loss: 0.7044476866722107:  49%|████▉     | 3191/6470 [36:40<39:50,  1.37it/s]
4|3192|Loss: 0.8631013631820679:  49%|████▉     | 3191/6470 [36:40<39:50,  1.37it/s]
4|3192|Loss: 0.8631013631820679:  49%|████▉     | 3192/6470 [36:41<39:34,  1.38it/s]
4|3193|Loss: 0.915241003036499:  49%|████▉     | 3192/6470 [36:41<39:34,  1.38it/s] 
4|3193|Loss: 0.915241003036499:  49%|████▉     | 3193/6470 [36:41<40:53,  1.34it/s]
4|3194|Loss: 0.8646999597549438:  49%|████▉     | 3193/6470 [36:42<40:53,  1.34it/s]
4|3194|Loss: 0.8646999597549438:  49%|████▉     | 3194/6470 [36:42<42:15,  1.29it/s]
4|3195|Loss: 0.785424530506134:  49%|████▉     | 3194/6470 [36:43<42:15,  1.29it/s] 
4|3195|Loss: 0.785424530506134:  49%|████▉     | 3195/6470 [36:43<40:50,  1.34it/s]
4|3196|Loss: 0.8837847709655762:  49%|████▉     | 3195/6470 [36:43<40:50,  1.34it/s]
4|3196|Loss: 0.8837847709655762:  49%|████▉     | 3196/6470 [36:43<37:02,  1.47it/s]
4|3197|Loss: 0.8961407542228699:  49%|████▉     | 3196/6470 [36:44<37:02,  1.47it/s]
4|3197|Loss: 0.8961407542228699:  49%|████▉     | 3197/6470 [36:44<38:56,  1.40it/s]
4|3198|Loss: 0.9133493900299072:  49%|████▉     | 3197/6470 [36:45<38:56,  1.40it/s]
4|3198|Loss: 0.9133493900299072:  49%|████▉     | 3198/6470 [36:45<41:03,  1.33it/s]
4|3199|Loss: 0.6823874115943909:  49%|████▉     | 3198/6470 [36:46<41:03,  1.33it/s]
4|3199|Loss: 0.6823874115943909:  49%|████▉     | 3199/6470 [36:46<43:30,  1.25it/s]
4|3200|Loss: 0.89045250415802:  49%|████▉     | 3199/6470 [36:46<43:30,  1.25it/s]  
4|3200|Loss: 0.89045250415802:  49%|████▉     | 3200/6470 [36:47<41:43,  1.31it/s]
4|3201|Loss: 0.8923505544662476:  49%|████▉     | 3200/6470 [36:47<41:43,  1.31it/s]
4|3201|Loss: 0.8923505544662476:  49%|████▉     | 3201/6470 [36:47<41:32,  1.31it/s]
4|3202|Loss: 0.7900094389915466:  49%|████▉     | 3201/6470 [36:48<41:32,  1.31it/s]
4|3202|Loss: 0.7900094389915466:  49%|████▉     | 3202/6470 [36:48<39:51,  1.37it/s]
4|3203|Loss: 0.8133283257484436:  49%|████▉     | 3202/6470 [36:48<39:51,  1.37it/s]
4|3203|Loss: 0.8133283257484436:  50%|████▉     | 3203/6470 [36:49<37:55,  1.44it/s]
4|3204|Loss: 0.8759604692459106:  50%|████▉     | 3203/6470 [36:49<37:55,  1.44it/s]
4|3204|Loss: 0.8759604692459106:  50%|████▉     | 3204/6470 [36:49<40:03,  1.36it/s]
4|3205|Loss: 0.9267756938934326:  50%|████▉     | 3204/6470 [36:50<40:03,  1.36it/s]
4|3205|Loss: 0.9267756938934326:  50%|████▉     | 3205/6470 [36:50<40:32,  1.34it/s]
4|3206|Loss: 0.9060119390487671:  50%|████▉     | 3205/6470 [36:51<40:32,  1.34it/s]
4|3206|Loss: 0.9060119390487671:  50%|████▉     | 3206/6470 [36:51<40:28,  1.34it/s]
4|3207|Loss: 1.0785654783248901:  50%|████▉     | 3206/6470 [36:51<40:28,  1.34it/s]
4|3207|Loss: 1.0785654783248901:  50%|████▉     | 3207/6470 [36:52<41:54,  1.30it/s]
4|3208|Loss: 0.8063724040985107:  50%|████▉     | 3207/6470 [36:52<41:54,  1.30it/s]
4|3208|Loss: 0.8063724040985107:  50%|████▉     | 3208/6470 [36:52<39:43,  1.37it/s]
4|3209|Loss: 0.863993227481842:  50%|████▉     | 3208/6470 [36:53<39:43,  1.37it/s] 
4|3209|Loss: 0.863993227481842:  50%|████▉     | 3209/6470 [36:53<39:16,  1.38it/s]
4|3210|Loss: 0.7944101095199585:  50%|████▉     | 3209/6470 [36:54<39:16,  1.38it/s]
4|3210|Loss: 0.7944101095199585:  50%|████▉     | 3210/6470 [36:54<41:51,  1.30it/s]
4|3211|Loss: 0.86615389585495:  50%|████▉     | 3210/6470 [36:54<41:51,  1.30it/s]  
4|3211|Loss: 0.86615389585495:  50%|████▉     | 3211/6470 [36:55<41:57,  1.29it/s]
4|3212|Loss: 0.875656008720398:  50%|████▉     | 3211/6470 [36:55<41:57,  1.29it/s]
4|3212|Loss: 0.875656008720398:  50%|████▉     | 3212/6470 [36:55<40:06,  1.35it/s]
4|3213|Loss: 0.799767792224884:  50%|████▉     | 3212/6470 [36:56<40:06,  1.35it/s]
4|3213|Loss: 0.799767792224884:  50%|████▉     | 3213/6470 [36:56<38:28,  1.41it/s]
4|3214|Loss: 0.7741274237632751:  50%|████▉     | 3213/6470 [36:56<38:28,  1.41it/s]
4|3214|Loss: 0.7741274237632751:  50%|████▉     | 3214/6470 [36:57<35:40,  1.52it/s]
4|3215|Loss: 0.7233193516731262:  50%|████▉     | 3214/6470 [36:57<35:40,  1.52it/s]
4|3215|Loss: 0.7233193516731262:  50%|████▉     | 3215/6470 [36:57<35:02,  1.55it/s]
4|3216|Loss: 0.9231025576591492:  50%|████▉     | 3215/6470 [36:58<35:02,  1.55it/s]
4|3216|Loss: 0.9231025576591492:  50%|████▉     | 3216/6470 [36:58<38:27,  1.41it/s]
4|3217|Loss: 0.6706188917160034:  50%|████▉     | 3216/6470 [36:59<38:27,  1.41it/s]
4|3217|Loss: 0.6706188917160034:  50%|████▉     | 3217/6470 [36:59<36:36,  1.48it/s]
4|3218|Loss: 0.8487903475761414:  50%|████▉     | 3217/6470 [36:59<36:36,  1.48it/s]
4|3218|Loss: 0.8487903475761414:  50%|████▉     | 3218/6470 [36:59<38:02,  1.42it/s]
4|3219|Loss: 0.8477334380149841:  50%|████▉     | 3218/6470 [37:00<38:02,  1.42it/s]
4|3219|Loss: 0.8477334380149841:  50%|████▉     | 3219/6470 [37:00<38:30,  1.41it/s]
4|3220|Loss: 0.9334090352058411:  50%|████▉     | 3219/6470 [37:01<38:30,  1.41it/s]
4|3220|Loss: 0.9334090352058411:  50%|████▉     | 3220/6470 [37:01<38:53,  1.39it/s]
4|3221|Loss: 0.899800181388855:  50%|████▉     | 3220/6470 [37:01<38:53,  1.39it/s] 
4|3221|Loss: 0.899800181388855:  50%|████▉     | 3221/6470 [37:02<37:57,  1.43it/s]
4|3222|Loss: 0.8179644346237183:  50%|████▉     | 3221/6470 [37:02<37:57,  1.43it/s]
4|3222|Loss: 0.8179644346237183:  50%|████▉     | 3222/6470 [37:02<38:08,  1.42it/s]
4|3223|Loss: 0.7731238007545471:  50%|████▉     | 3222/6470 [37:03<38:08,  1.42it/s]
4|3223|Loss: 0.7731238007545471:  50%|████▉     | 3223/6470 [37:03<33:42,  1.61it/s]
4|3224|Loss: 0.9756788015365601:  50%|████▉     | 3223/6470 [37:03<33:42,  1.61it/s]
4|3224|Loss: 0.9756788015365601:  50%|████▉     | 3224/6470 [37:03<34:16,  1.58it/s]
4|3225|Loss: 1.1025052070617676:  50%|████▉     | 3224/6470 [37:04<34:16,  1.58it/s]
4|3225|Loss: 1.1025052070617676:  50%|████▉     | 3225/6470 [37:04<37:28,  1.44it/s]
4|3226|Loss: 0.770388662815094:  50%|████▉     | 3225/6470 [37:05<37:28,  1.44it/s] 
4|3226|Loss: 0.770388662815094:  50%|████▉     | 3226/6470 [37:05<38:08,  1.42it/s]
4|3227|Loss: 0.9175771474838257:  50%|████▉     | 3226/6470 [37:05<38:08,  1.42it/s]
4|3227|Loss: 0.9175771474838257:  50%|████▉     | 3227/6470 [37:06<38:35,  1.40it/s]
4|3228|Loss: 0.7518549561500549:  50%|████▉     | 3227/6470 [37:06<38:35,  1.40it/s]
4|3228|Loss: 0.7518549561500549:  50%|████▉     | 3228/6470 [37:06<38:32,  1.40it/s]
4|3229|Loss: 1.0176178216934204:  50%|████▉     | 3228/6470 [37:07<38:32,  1.40it/s]
4|3229|Loss: 1.0176178216934204:  50%|████▉     | 3229/6470 [37:07<39:00,  1.38it/s]
4|3230|Loss: 0.829028844833374:  50%|████▉     | 3229/6470 [37:08<39:00,  1.38it/s] 
4|3230|Loss: 0.829028844833374:  50%|████▉     | 3230/6470 [37:08<37:16,  1.45it/s]
4|3231|Loss: 0.999894917011261:  50%|████▉     | 3230/6470 [37:08<37:16,  1.45it/s]
4|3231|Loss: 0.999894917011261:  50%|████▉     | 3231/6470 [37:08<34:38,  1.56it/s]
4|3232|Loss: 0.8748205900192261:  50%|████▉     | 3231/6470 [37:09<34:38,  1.56it/s]
4|3232|Loss: 0.8748205900192261:  50%|████▉     | 3232/6470 [37:09<36:43,  1.47it/s]
4|3233|Loss: 0.8845731616020203:  50%|████▉     | 3232/6470 [37:10<36:43,  1.47it/s]
4|3233|Loss: 0.8845731616020203:  50%|████▉     | 3233/6470 [37:10<37:32,  1.44it/s]
4|3234|Loss: 0.8930560946464539:  50%|████▉     | 3233/6470 [37:10<37:32,  1.44it/s]
4|3234|Loss: 0.8930560946464539:  50%|████▉     | 3234/6470 [37:10<36:48,  1.47it/s]
4|3235|Loss: 0.9503575563430786:  50%|████▉     | 3234/6470 [37:11<36:48,  1.47it/s]
4|3235|Loss: 0.9503575563430786:  50%|█████     | 3235/6470 [37:11<37:18,  1.45it/s]
4|3236|Loss: 0.8349155187606812:  50%|█████     | 3235/6470 [37:12<37:18,  1.45it/s]
4|3236|Loss: 0.8349155187606812:  50%|█████     | 3236/6470 [37:12<37:59,  1.42it/s]
4|3237|Loss: 0.7918470501899719:  50%|█████     | 3236/6470 [37:12<37:59,  1.42it/s]
4|3237|Loss: 0.7918470501899719:  50%|█████     | 3237/6470 [37:13<38:04,  1.42it/s]
4|3238|Loss: 0.8570900559425354:  50%|█████     | 3237/6470 [37:13<38:04,  1.42it/s]
4|3238|Loss: 0.8570900559425354:  50%|█████     | 3238/6470 [37:13<39:07,  1.38it/s]
4|3239|Loss: 0.6456478834152222:  50%|█████     | 3238/6470 [37:14<39:07,  1.38it/s]
4|3239|Loss: 0.6456478834152222:  50%|█████     | 3239/6470 [37:14<42:42,  1.26it/s]
4|3240|Loss: 0.5684023499488831:  50%|█████     | 3239/6470 [37:15<42:42,  1.26it/s]
4|3240|Loss: 0.5684023499488831:  50%|█████     | 3240/6470 [37:15<43:04,  1.25it/s]
4|3241|Loss: 0.8712909817695618:  50%|█████     | 3240/6470 [37:16<43:04,  1.25it/s]
4|3241|Loss: 0.8712909817695618:  50%|█████     | 3241/6470 [37:16<42:17,  1.27it/s]
4|3242|Loss: 0.7237332463264465:  50%|█████     | 3241/6470 [37:16<42:17,  1.27it/s]
4|3242|Loss: 0.7237332463264465:  50%|█████     | 3242/6470 [37:17<41:28,  1.30it/s]
4|3243|Loss: 0.8856683373451233:  50%|█████     | 3242/6470 [37:17<41:28,  1.30it/s]
4|3243|Loss: 0.8856683373451233:  50%|█████     | 3243/6470 [37:17<39:43,  1.35it/s]
4|3244|Loss: 0.8758058547973633:  50%|█████     | 3243/6470 [37:18<39:43,  1.35it/s]
4|3244|Loss: 0.8758058547973633:  50%|█████     | 3244/6470 [37:18<39:02,  1.38it/s]
4|3245|Loss: 0.8526461124420166:  50%|█████     | 3244/6470 [37:18<39:02,  1.38it/s]
4|3245|Loss: 0.8526461124420166:  50%|█████     | 3245/6470 [37:18<34:19,  1.57it/s]
4|3246|Loss: 0.8445988893508911:  50%|█████     | 3245/6470 [37:19<34:19,  1.57it/s]
4|3246|Loss: 0.8445988893508911:  50%|█████     | 3246/6470 [37:19<33:36,  1.60it/s]
4|3247|Loss: 0.8437147736549377:  50%|█████     | 3246/6470 [37:19<33:36,  1.60it/s]
4|3247|Loss: 0.8437147736549377:  50%|█████     | 3247/6470 [37:20<33:53,  1.58it/s]
4|3248|Loss: 0.8588768243789673:  50%|█████     | 3247/6470 [37:20<33:53,  1.58it/s]
4|3248|Loss: 0.8588768243789673:  50%|█████     | 3248/6470 [37:20<34:03,  1.58it/s]
4|3249|Loss: 1.0622135400772095:  50%|█████     | 3248/6470 [37:21<34:03,  1.58it/s]
4|3249|Loss: 1.0622135400772095:  50%|█████     | 3249/6470 [37:21<36:04,  1.49it/s]
4|3250|Loss: 0.8026235103607178:  50%|█████     | 3249/6470 [37:22<36:04,  1.49it/s]
4|3250|Loss: 0.8026235103607178:  50%|█████     | 3250/6470 [37:22<37:31,  1.43it/s]
4|3251|Loss: 0.8059868216514587:  50%|█████     | 3250/6470 [37:22<37:31,  1.43it/s]
4|3251|Loss: 0.8059868216514587:  50%|█████     | 3251/6470 [37:23<39:01,  1.37it/s]
4|3252|Loss: 0.9167526364326477:  50%|█████     | 3251/6470 [37:23<39:01,  1.37it/s]
4|3252|Loss: 0.9167526364326477:  50%|█████     | 3252/6470 [37:23<39:24,  1.36it/s]
4|3253|Loss: 0.8372097015380859:  50%|█████     | 3252/6470 [37:24<39:24,  1.36it/s]
4|3253|Loss: 0.8372097015380859:  50%|█████     | 3253/6470 [37:24<39:05,  1.37it/s]
4|3254|Loss: 0.8741200566291809:  50%|█████     | 3253/6470 [37:25<39:05,  1.37it/s]
4|3254|Loss: 0.8741200566291809:  50%|█████     | 3254/6470 [37:25<40:39,  1.32it/s]
4|3255|Loss: 0.9371716380119324:  50%|█████     | 3254/6470 [37:25<40:39,  1.32it/s]
4|3255|Loss: 0.9371716380119324:  50%|█████     | 3255/6470 [37:26<40:21,  1.33it/s]
4|3256|Loss: 0.8686883449554443:  50%|█████     | 3255/6470 [37:26<40:21,  1.33it/s]
4|3256|Loss: 0.8686883449554443:  50%|█████     | 3256/6470 [37:26<39:36,  1.35it/s]
4|3257|Loss: 0.6207017302513123:  50%|█████     | 3256/6470 [37:27<39:36,  1.35it/s]
4|3257|Loss: 0.6207017302513123:  50%|█████     | 3257/6470 [37:27<41:47,  1.28it/s]
4|3258|Loss: 0.8173765540122986:  50%|█████     | 3257/6470 [37:28<41:47,  1.28it/s]
4|3258|Loss: 0.8173765540122986:  50%|█████     | 3258/6470 [37:28<40:30,  1.32it/s]
4|3259|Loss: 0.84710294008255:  50%|█████     | 3258/6470 [37:28<40:30,  1.32it/s]  
4|3259|Loss: 0.84710294008255:  50%|█████     | 3259/6470 [37:29<39:26,  1.36it/s]
4|3260|Loss: 0.8680245280265808:  50%|█████     | 3259/6470 [37:29<39:26,  1.36it/s]
4|3260|Loss: 0.8680245280265808:  50%|█████     | 3260/6470 [37:29<36:33,  1.46it/s]
4|3261|Loss: 0.81072598695755:  50%|█████     | 3260/6470 [37:30<36:33,  1.46it/s]  
4|3261|Loss: 0.81072598695755:  50%|█████     | 3261/6470 [37:30<34:20,  1.56it/s]
4|3262|Loss: 0.9032940864562988:  50%|█████     | 3261/6470 [37:30<34:20,  1.56it/s]
4|3262|Loss: 0.9032940864562988:  50%|█████     | 3262/6470 [37:30<34:02,  1.57it/s]
4|3263|Loss: 0.8414977192878723:  50%|█████     | 3262/6470 [37:31<34:02,  1.57it/s]
4|3263|Loss: 0.8414977192878723:  50%|█████     | 3263/6470 [37:31<34:59,  1.53it/s]
4|3264|Loss: 0.8264813423156738:  50%|█████     | 3263/6470 [37:31<34:59,  1.53it/s]
4|3264|Loss: 0.8264813423156738:  50%|█████     | 3264/6470 [37:32<36:12,  1.48it/s]
4|3265|Loss: 0.8434021472930908:  50%|█████     | 3264/6470 [37:32<36:12,  1.48it/s]
4|3265|Loss: 0.8434021472930908:  50%|█████     | 3265/6470 [37:33<37:48,  1.41it/s]
4|3266|Loss: 0.8867198824882507:  50%|█████     | 3265/6470 [37:33<37:48,  1.41it/s]
4|3266|Loss: 0.8867198824882507:  50%|█████     | 3266/6470 [37:33<38:07,  1.40it/s]
4|3267|Loss: 0.9129109382629395:  50%|█████     | 3266/6470 [37:34<38:07,  1.40it/s]
4|3267|Loss: 0.9129109382629395:  50%|█████     | 3267/6470 [37:34<39:00,  1.37it/s]
4|3268|Loss: 0.8172990679740906:  50%|█████     | 3267/6470 [37:34<39:00,  1.37it/s]
4|3268|Loss: 0.8172990679740906:  51%|█████     | 3268/6470 [37:35<36:38,  1.46it/s]
4|3269|Loss: 0.8480780720710754:  51%|█████     | 3268/6470 [37:35<36:38,  1.46it/s]
4|3269|Loss: 0.8480780720710754:  51%|█████     | 3269/6470 [37:35<36:14,  1.47it/s]
4|3270|Loss: 0.8397312164306641:  51%|█████     | 3269/6470 [37:36<36:14,  1.47it/s]
4|3270|Loss: 0.8397312164306641:  51%|█████     | 3270/6470 [37:36<35:06,  1.52it/s]
4|3271|Loss: 0.8212153315544128:  51%|█████     | 3270/6470 [37:36<35:06,  1.52it/s]
4|3271|Loss: 0.8212153315544128:  51%|█████     | 3271/6470 [37:37<35:36,  1.50it/s]
4|3272|Loss: 0.821696937084198:  51%|█████     | 3271/6470 [37:37<35:36,  1.50it/s] 
4|3272|Loss: 0.821696937084198:  51%|█████     | 3272/6470 [37:37<36:34,  1.46it/s]
4|3273|Loss: 0.8676657676696777:  51%|█████     | 3272/6470 [37:38<36:34,  1.46it/s]
4|3273|Loss: 0.8676657676696777:  51%|█████     | 3273/6470 [37:38<36:14,  1.47it/s]
4|3274|Loss: 1.0003618001937866:  51%|█████     | 3273/6470 [37:38<36:14,  1.47it/s]
4|3274|Loss: 1.0003618001937866:  51%|█████     | 3274/6470 [37:39<36:20,  1.47it/s]
4|3275|Loss: 0.8659982085227966:  51%|█████     | 3274/6470 [37:39<36:20,  1.47it/s]
4|3275|Loss: 0.8659982085227966:  51%|█████     | 3275/6470 [37:40<38:16,  1.39it/s]
4|3276|Loss: 0.9016239047050476:  51%|█████     | 3275/6470 [37:40<38:16,  1.39it/s]
4|3276|Loss: 0.9016239047050476:  51%|█████     | 3276/6470 [37:40<38:50,  1.37it/s]
4|3277|Loss: 0.8164737224578857:  51%|█████     | 3276/6470 [37:41<38:50,  1.37it/s]
4|3277|Loss: 0.8164737224578857:  51%|█████     | 3277/6470 [37:41<40:28,  1.31it/s]
4|3278|Loss: 0.7775152921676636:  51%|█████     | 3277/6470 [37:42<40:28,  1.31it/s]
4|3278|Loss: 0.7775152921676636:  51%|█████     | 3278/6470 [37:42<39:22,  1.35it/s]
4|3279|Loss: 0.8513988256454468:  51%|█████     | 3278/6470 [37:42<39:22,  1.35it/s]
4|3279|Loss: 0.8513988256454468:  51%|█████     | 3279/6470 [37:42<37:31,  1.42it/s]
4|3280|Loss: 0.8588368892669678:  51%|█████     | 3279/6470 [37:43<37:31,  1.42it/s]
4|3280|Loss: 0.8588368892669678:  51%|█████     | 3280/6470 [37:43<36:01,  1.48it/s]
4|3281|Loss: 0.7783251404762268:  51%|█████     | 3280/6470 [37:43<36:01,  1.48it/s]
4|3281|Loss: 0.7783251404762268:  51%|█████     | 3281/6470 [37:44<35:22,  1.50it/s]
4|3282|Loss: 0.8546714782714844:  51%|█████     | 3281/6470 [37:44<35:22,  1.50it/s]
4|3282|Loss: 0.8546714782714844:  51%|█████     | 3282/6470 [37:44<36:57,  1.44it/s]
4|3283|Loss: 0.927109956741333:  51%|█████     | 3282/6470 [37:45<36:57,  1.44it/s] 
4|3283|Loss: 0.927109956741333:  51%|█████     | 3283/6470 [37:45<35:56,  1.48it/s]
4|3284|Loss: 0.8516689538955688:  51%|█████     | 3283/6470 [37:45<35:56,  1.48it/s]
4|3284|Loss: 0.8516689538955688:  51%|█████     | 3284/6470 [37:46<34:54,  1.52it/s]
4|3285|Loss: 0.701988160610199:  51%|█████     | 3284/6470 [37:46<34:54,  1.52it/s] 
4|3285|Loss: 0.701988160610199:  51%|█████     | 3285/6470 [37:46<37:07,  1.43it/s]
4|3286|Loss: 0.7891569137573242:  51%|█████     | 3285/6470 [37:47<37:07,  1.43it/s]
4|3286|Loss: 0.7891569137573242:  51%|█████     | 3286/6470 [37:47<35:09,  1.51it/s]
4|3287|Loss: 0.6579357981681824:  51%|█████     | 3286/6470 [37:47<35:09,  1.51it/s]
4|3287|Loss: 0.6579357981681824:  51%|█████     | 3287/6470 [37:48<35:09,  1.51it/s]
4|3288|Loss: 0.71636563539505:  51%|█████     | 3287/6470 [37:48<35:09,  1.51it/s]  
4|3288|Loss: 0.71636563539505:  51%|█████     | 3288/6470 [37:48<34:40,  1.53it/s]
4|3289|Loss: 0.7151591181755066:  51%|█████     | 3288/6470 [37:49<34:40,  1.53it/s]
4|3289|Loss: 0.7151591181755066:  51%|█████     | 3289/6470 [37:49<34:01,  1.56it/s]
4|3290|Loss: 0.9851264953613281:  51%|█████     | 3289/6470 [37:49<34:01,  1.56it/s]
4|3290|Loss: 0.9851264953613281:  51%|█████     | 3290/6470 [37:50<35:01,  1.51it/s]
4|3291|Loss: 0.9181452989578247:  51%|█████     | 3290/6470 [37:50<35:01,  1.51it/s]
4|3291|Loss: 0.9181452989578247:  51%|█████     | 3291/6470 [37:50<33:18,  1.59it/s]
4|3292|Loss: 0.9403197169303894:  51%|█████     | 3291/6470 [37:51<33:18,  1.59it/s]
4|3292|Loss: 0.9403197169303894:  51%|█████     | 3292/6470 [37:51<36:29,  1.45it/s]
4|3293|Loss: 0.8333126902580261:  51%|█████     | 3292/6470 [37:52<36:29,  1.45it/s]
4|3293|Loss: 0.8333126902580261:  51%|█████     | 3293/6470 [37:52<39:21,  1.35it/s]
4|3294|Loss: 0.7804670929908752:  51%|█████     | 3293/6470 [37:52<39:21,  1.35it/s]
4|3294|Loss: 0.7804670929908752:  51%|█████     | 3294/6470 [37:53<38:17,  1.38it/s]
4|3295|Loss: 0.7602798342704773:  51%|█████     | 3294/6470 [37:53<38:17,  1.38it/s]
4|3295|Loss: 0.7602798342704773:  51%|█████     | 3295/6470 [37:53<33:39,  1.57it/s]
4|3296|Loss: 0.8526623845100403:  51%|█████     | 3295/6470 [37:53<33:39,  1.57it/s]
4|3296|Loss: 0.8526623845100403:  51%|█████     | 3296/6470 [37:54<36:39,  1.44it/s]
4|3297|Loss: 0.8122076392173767:  51%|█████     | 3296/6470 [37:54<36:39,  1.44it/s]
4|3297|Loss: 0.8122076392173767:  51%|█████     | 3297/6470 [37:55<38:02,  1.39it/s]
4|3298|Loss: 0.6717491149902344:  51%|█████     | 3297/6470 [37:55<38:02,  1.39it/s]
4|3298|Loss: 0.6717491149902344:  51%|█████     | 3298/6470 [37:55<39:54,  1.32it/s]
4|3299|Loss: 0.9157122373580933:  51%|█████     | 3298/6470 [37:56<39:54,  1.32it/s]
4|3299|Loss: 0.9157122373580933:  51%|█████     | 3299/6470 [37:56<41:14,  1.28it/s]
4|3300|Loss: 0.8575469255447388:  51%|█████     | 3299/6470 [37:57<41:14,  1.28it/s]
4|3300|Loss: 0.8575469255447388:  51%|█████     | 3300/6470 [37:57<38:56,  1.36it/s]
4|3301|Loss: 0.8338969349861145:  51%|█████     | 3300/6470 [37:57<38:56,  1.36it/s]
4|3301|Loss: 0.8338969349861145:  51%|█████     | 3301/6470 [37:58<38:28,  1.37it/s]
4|3302|Loss: 0.9520491361618042:  51%|█████     | 3301/6470 [37:58<38:28,  1.37it/s]
4|3302|Loss: 0.9520491361618042:  51%|█████     | 3302/6470 [37:58<37:24,  1.41it/s]
4|3303|Loss: 0.7794733643531799:  51%|█████     | 3302/6470 [37:59<37:24,  1.41it/s]
4|3303|Loss: 0.7794733643531799:  51%|█████     | 3303/6470 [37:59<37:34,  1.40it/s]
4|3304|Loss: 0.7470465302467346:  51%|█████     | 3303/6470 [37:59<37:34,  1.40it/s]
4|3304|Loss: 0.7470465302467346:  51%|█████     | 3304/6470 [38:00<37:35,  1.40it/s]
4|3305|Loss: 0.9955165386199951:  51%|█████     | 3304/6470 [38:00<37:35,  1.40it/s]
4|3305|Loss: 0.9955165386199951:  51%|█████     | 3305/6470 [38:00<34:16,  1.54it/s]
4|3306|Loss: 0.796608030796051:  51%|█████     | 3305/6470 [38:01<34:16,  1.54it/s] 
4|3306|Loss: 0.796608030796051:  51%|█████     | 3306/6470 [38:01<34:46,  1.52it/s]
4|3307|Loss: 0.797109842300415:  51%|█████     | 3306/6470 [38:01<34:46,  1.52it/s]
4|3307|Loss: 0.797109842300415:  51%|█████     | 3307/6470 [38:02<36:05,  1.46it/s]
4|3308|Loss: 0.9212254881858826:  51%|█████     | 3307/6470 [38:02<36:05,  1.46it/s]
4|3308|Loss: 0.9212254881858826:  51%|█████     | 3308/6470 [38:02<36:27,  1.45it/s]
4|3309|Loss: 0.9210308194160461:  51%|█████     | 3308/6470 [38:03<36:27,  1.45it/s]
4|3309|Loss: 0.9210308194160461:  51%|█████     | 3309/6470 [38:03<37:46,  1.39it/s]
4|3310|Loss: 0.8845818638801575:  51%|█████     | 3309/6470 [38:04<37:46,  1.39it/s]
4|3310|Loss: 0.8845818638801575:  51%|█████     | 3310/6470 [38:04<38:28,  1.37it/s]
4|3311|Loss: 0.9810335040092468:  51%|█████     | 3310/6470 [38:04<38:28,  1.37it/s]
4|3311|Loss: 0.9810335040092468:  51%|█████     | 3311/6470 [38:05<37:26,  1.41it/s]
4|3312|Loss: 0.6587655544281006:  51%|█████     | 3311/6470 [38:05<37:26,  1.41it/s]
4|3312|Loss: 0.6587655544281006:  51%|█████     | 3312/6470 [38:05<34:39,  1.52it/s]
4|3313|Loss: 0.7805339097976685:  51%|█████     | 3312/6470 [38:05<34:39,  1.52it/s]
4|3313|Loss: 0.7805339097976685:  51%|█████     | 3313/6470 [38:06<33:52,  1.55it/s]
4|3314|Loss: 0.9908300638198853:  51%|█████     | 3313/6470 [38:06<33:52,  1.55it/s]
4|3314|Loss: 0.9908300638198853:  51%|█████     | 3314/6470 [38:06<34:29,  1.53it/s]
4|3315|Loss: 0.8660349249839783:  51%|█████     | 3314/6470 [38:07<34:29,  1.53it/s]
4|3315|Loss: 0.8660349249839783:  51%|█████     | 3315/6470 [38:07<37:10,  1.41it/s]
4|3316|Loss: 0.8714761137962341:  51%|█████     | 3315/6470 [38:08<37:10,  1.41it/s]
4|3316|Loss: 0.8714761137962341:  51%|█████▏    | 3316/6470 [38:08<36:07,  1.45it/s]
4|3317|Loss: 0.8902134299278259:  51%|█████▏    | 3316/6470 [38:08<36:07,  1.45it/s]
4|3317|Loss: 0.8902134299278259:  51%|█████▏    | 3317/6470 [38:09<36:24,  1.44it/s]
4|3318|Loss: 0.8860241770744324:  51%|█████▏    | 3317/6470 [38:09<36:24,  1.44it/s]
4|3318|Loss: 0.8860241770744324:  51%|█████▏    | 3318/6470 [38:09<37:21,  1.41it/s]
4|3319|Loss: 0.809786319732666:  51%|█████▏    | 3318/6470 [38:10<37:21,  1.41it/s] 
4|3319|Loss: 0.809786319732666:  51%|█████▏    | 3319/6470 [38:10<38:41,  1.36it/s]
4|3320|Loss: 0.8141931295394897:  51%|█████▏    | 3319/6470 [38:11<38:41,  1.36it/s]
4|3320|Loss: 0.8141931295394897:  51%|█████▏    | 3320/6470 [38:11<35:59,  1.46it/s]
4|3321|Loss: 0.8481045961380005:  51%|█████▏    | 3320/6470 [38:11<35:59,  1.46it/s]
4|3321|Loss: 0.8481045961380005:  51%|█████▏    | 3321/6470 [38:11<36:05,  1.45it/s]
4|3322|Loss: 0.8086684346199036:  51%|█████▏    | 3321/6470 [38:12<36:05,  1.45it/s]
4|3322|Loss: 0.8086684346199036:  51%|█████▏    | 3322/6470 [38:12<36:33,  1.44it/s]
4|3323|Loss: 0.7782829999923706:  51%|█████▏    | 3322/6470 [38:13<36:33,  1.44it/s]
4|3323|Loss: 0.7782829999923706:  51%|█████▏    | 3323/6470 [38:13<35:38,  1.47it/s]
4|3324|Loss: 0.6853114366531372:  51%|█████▏    | 3323/6470 [38:13<35:38,  1.47it/s]
4|3324|Loss: 0.6853114366531372:  51%|█████▏    | 3324/6470 [38:13<33:18,  1.57it/s]
4|3325|Loss: 0.8549388647079468:  51%|█████▏    | 3324/6470 [38:14<33:18,  1.57it/s]
4|3325|Loss: 0.8549388647079468:  51%|█████▏    | 3325/6470 [38:14<32:38,  1.61it/s]
4|3326|Loss: 0.779552161693573:  51%|█████▏    | 3325/6470 [38:14<32:38,  1.61it/s] 
4|3326|Loss: 0.779552161693573:  51%|█████▏    | 3326/6470 [38:15<34:46,  1.51it/s]
4|3327|Loss: 0.7900068759918213:  51%|█████▏    | 3326/6470 [38:15<34:46,  1.51it/s]
4|3327|Loss: 0.7900068759918213:  51%|█████▏    | 3327/6470 [38:15<36:20,  1.44it/s]
4|3328|Loss: 0.7772973775863647:  51%|█████▏    | 3327/6470 [38:16<36:20,  1.44it/s]
4|3328|Loss: 0.7772973775863647:  51%|█████▏    | 3328/6470 [38:16<38:50,  1.35it/s]
4|3329|Loss: 0.9060758352279663:  51%|█████▏    | 3328/6470 [38:17<38:50,  1.35it/s]
4|3329|Loss: 0.9060758352279663:  51%|█████▏    | 3329/6470 [38:17<36:05,  1.45it/s]
4|3330|Loss: 0.7492671012878418:  51%|█████▏    | 3329/6470 [38:17<36:05,  1.45it/s]
4|3330|Loss: 0.7492671012878418:  51%|█████▏    | 3330/6470 [38:18<37:09,  1.41it/s]
4|3331|Loss: 0.9365234375:  51%|█████▏    | 3330/6470 [38:18<37:09,  1.41it/s]      
4|3331|Loss: 0.9365234375:  51%|█████▏    | 3331/6470 [38:18<37:34,  1.39it/s]
4|3332|Loss: 0.868656575679779:  51%|█████▏    | 3331/6470 [38:19<37:34,  1.39it/s]
4|3332|Loss: 0.868656575679779:  51%|█████▏    | 3332/6470 [38:19<37:56,  1.38it/s]
4|3333|Loss: 0.9070738554000854:  51%|█████▏    | 3332/6470 [38:20<37:56,  1.38it/s]
4|3333|Loss: 0.9070738554000854:  52%|█████▏    | 3333/6470 [38:20<38:06,  1.37it/s]
4|3334|Loss: 0.774604856967926:  52%|█████▏    | 3333/6470 [38:20<38:06,  1.37it/s] 
4|3334|Loss: 0.774604856967926:  52%|█████▏    | 3334/6470 [38:21<37:46,  1.38it/s]
4|3335|Loss: 0.7508108615875244:  52%|█████▏    | 3334/6470 [38:21<37:46,  1.38it/s]
4|3335|Loss: 0.7508108615875244:  52%|█████▏    | 3335/6470 [38:21<39:16,  1.33it/s]
4|3336|Loss: 0.855521023273468:  52%|█████▏    | 3335/6470 [38:22<39:16,  1.33it/s] 
4|3336|Loss: 0.855521023273468:  52%|█████▏    | 3336/6470 [38:22<39:28,  1.32it/s]
4|3337|Loss: 0.9068973064422607:  52%|█████▏    | 3336/6470 [38:23<39:28,  1.32it/s]
4|3337|Loss: 0.9068973064422607:  52%|█████▏    | 3337/6470 [38:23<38:45,  1.35it/s]
4|3338|Loss: 0.814018189907074:  52%|█████▏    | 3337/6470 [38:23<38:45,  1.35it/s] 
4|3338|Loss: 0.814018189907074:  52%|█████▏    | 3338/6470 [38:24<39:36,  1.32it/s]
4|3339|Loss: 0.8426417112350464:  52%|█████▏    | 3338/6470 [38:24<39:36,  1.32it/s]
4|3339|Loss: 0.8426417112350464:  52%|█████▏    | 3339/6470 [38:24<39:00,  1.34it/s]
4|3340|Loss: 0.8184524774551392:  52%|█████▏    | 3339/6470 [38:25<39:00,  1.34it/s]
4|3340|Loss: 0.8184524774551392:  52%|█████▏    | 3340/6470 [38:25<38:14,  1.36it/s]
4|3341|Loss: 0.8307162523269653:  52%|█████▏    | 3340/6470 [38:25<38:14,  1.36it/s]
4|3341|Loss: 0.8307162523269653:  52%|█████▏    | 3341/6470 [38:26<38:06,  1.37it/s]
4|3342|Loss: 0.7354995012283325:  52%|█████▏    | 3341/6470 [38:26<38:06,  1.37it/s]
4|3342|Loss: 0.7354995012283325:  52%|█████▏    | 3342/6470 [38:27<40:28,  1.29it/s]
4|3343|Loss: 0.9739823341369629:  52%|█████▏    | 3342/6470 [38:27<40:28,  1.29it/s]
4|3343|Loss: 0.9739823341369629:  52%|█████▏    | 3343/6470 [38:27<37:37,  1.38it/s]
4|3344|Loss: 0.9468988180160522:  52%|█████▏    | 3343/6470 [38:28<37:37,  1.38it/s]
4|3344|Loss: 0.9468988180160522:  52%|█████▏    | 3344/6470 [38:28<38:17,  1.36it/s]
4|3345|Loss: 0.8613275289535522:  52%|█████▏    | 3344/6470 [38:28<38:17,  1.36it/s]
4|3345|Loss: 0.8613275289535522:  52%|█████▏    | 3345/6470 [38:29<38:50,  1.34it/s]
4|3346|Loss: 0.8738090991973877:  52%|█████▏    | 3345/6470 [38:29<38:50,  1.34it/s]
4|3346|Loss: 0.8738090991973877:  52%|█████▏    | 3346/6470 [38:29<37:17,  1.40it/s]
4|3347|Loss: 0.904463529586792:  52%|█████▏    | 3346/6470 [38:30<37:17,  1.40it/s] 
4|3347|Loss: 0.904463529586792:  52%|█████▏    | 3347/6470 [38:30<36:37,  1.42it/s]
4|3348|Loss: 0.77499920129776:  52%|█████▏    | 3347/6470 [38:30<36:37,  1.42it/s] 
4|3348|Loss: 0.77499920129776:  52%|█████▏    | 3348/6470 [38:31<35:08,  1.48it/s]
4|3349|Loss: 0.7851600050926208:  52%|█████▏    | 3348/6470 [38:31<35:08,  1.48it/s]
4|3349|Loss: 0.7851600050926208:  52%|█████▏    | 3349/6470 [38:31<35:13,  1.48it/s]
4|3350|Loss: 0.7967857718467712:  52%|█████▏    | 3349/6470 [38:32<35:13,  1.48it/s]
4|3350|Loss: 0.7967857718467712:  52%|█████▏    | 3350/6470 [38:32<35:43,  1.46it/s]
4|3351|Loss: 0.7213861346244812:  52%|█████▏    | 3350/6470 [38:32<35:43,  1.46it/s]
4|3351|Loss: 0.7213861346244812:  52%|█████▏    | 3351/6470 [38:33<33:11,  1.57it/s]
4|3352|Loss: 0.8272823095321655:  52%|█████▏    | 3351/6470 [38:33<33:11,  1.57it/s]
4|3352|Loss: 0.8272823095321655:  52%|█████▏    | 3352/6470 [38:33<32:38,  1.59it/s]
4|3353|Loss: 0.7970638871192932:  52%|█████▏    | 3352/6470 [38:34<32:38,  1.59it/s]
4|3353|Loss: 0.7970638871192932:  52%|█████▏    | 3353/6470 [38:34<29:36,  1.75it/s]
4|3354|Loss: 0.9192444682121277:  52%|█████▏    | 3353/6470 [38:34<29:36,  1.75it/s]
4|3354|Loss: 0.9192444682121277:  52%|█████▏    | 3354/6470 [38:34<31:22,  1.66it/s]
4|3355|Loss: 0.8697241544723511:  52%|█████▏    | 3354/6470 [38:35<31:22,  1.66it/s]
4|3355|Loss: 0.8697241544723511:  52%|█████▏    | 3355/6470 [38:35<35:22,  1.47it/s]
4|3356|Loss: 0.8712183833122253:  52%|█████▏    | 3355/6470 [38:36<35:22,  1.47it/s]
4|3356|Loss: 0.8712183833122253:  52%|█████▏    | 3356/6470 [38:36<35:39,  1.46it/s]
4|3357|Loss: 0.9011980891227722:  52%|█████▏    | 3356/6470 [38:36<35:39,  1.46it/s]
4|3357|Loss: 0.9011980891227722:  52%|█████▏    | 3357/6470 [38:37<35:37,  1.46it/s]
4|3358|Loss: 0.8425817489624023:  52%|█████▏    | 3357/6470 [38:37<35:37,  1.46it/s]
4|3358|Loss: 0.8425817489624023:  52%|█████▏    | 3358/6470 [38:37<35:09,  1.48it/s]
4|3359|Loss: 0.8844918012619019:  52%|█████▏    | 3358/6470 [38:38<35:09,  1.48it/s]
4|3359|Loss: 0.8844918012619019:  52%|█████▏    | 3359/6470 [38:38<33:00,  1.57it/s]
4|3360|Loss: 0.7377932667732239:  52%|█████▏    | 3359/6470 [38:38<33:00,  1.57it/s]
4|3360|Loss: 0.7377932667732239:  52%|█████▏    | 3360/6470 [38:39<35:23,  1.46it/s]
4|3361|Loss: 1.0266560316085815:  52%|█████▏    | 3360/6470 [38:39<35:23,  1.46it/s]
4|3361|Loss: 1.0266560316085815:  52%|█████▏    | 3361/6470 [38:39<38:53,  1.33it/s]
4|3362|Loss: 0.7725381255149841:  52%|█████▏    | 3361/6470 [38:40<38:53,  1.33it/s]
4|3362|Loss: 0.7725381255149841:  52%|█████▏    | 3362/6470 [38:40<37:52,  1.37it/s]
4|3363|Loss: 0.8079379200935364:  52%|█████▏    | 3362/6470 [38:41<37:52,  1.37it/s]
4|3363|Loss: 0.8079379200935364:  52%|█████▏    | 3363/6470 [38:41<37:40,  1.37it/s]
4|3364|Loss: 0.9288814663887024:  52%|█████▏    | 3363/6470 [38:41<37:40,  1.37it/s]
4|3364|Loss: 0.9288814663887024:  52%|█████▏    | 3364/6470 [38:42<37:27,  1.38it/s]
4|3365|Loss: 0.8823853135108948:  52%|█████▏    | 3364/6470 [38:42<37:27,  1.38it/s]
4|3365|Loss: 0.8823853135108948:  52%|█████▏    | 3365/6470 [38:42<39:01,  1.33it/s]
4|3366|Loss: 1.001084804534912:  52%|█████▏    | 3365/6470 [38:43<39:01,  1.33it/s] 
4|3366|Loss: 1.001084804534912:  52%|█████▏    | 3366/6470 [38:43<38:00,  1.36it/s]
4|3367|Loss: 0.7974871397018433:  52%|█████▏    | 3366/6470 [38:44<38:00,  1.36it/s]
4|3367|Loss: 0.7974871397018433:  52%|█████▏    | 3367/6470 [38:44<37:19,  1.39it/s]
4|3368|Loss: 0.7882670164108276:  52%|█████▏    | 3367/6470 [38:44<37:19,  1.39it/s]
4|3368|Loss: 0.7882670164108276:  52%|█████▏    | 3368/6470 [38:44<35:38,  1.45it/s]
4|3369|Loss: 0.8600456714630127:  52%|█████▏    | 3368/6470 [38:45<35:38,  1.45it/s]
4|3369|Loss: 0.8600456714630127:  52%|█████▏    | 3369/6470 [38:45<34:03,  1.52it/s]
4|3370|Loss: 0.8814557194709778:  52%|█████▏    | 3369/6470 [38:45<34:03,  1.52it/s]
4|3370|Loss: 0.8814557194709778:  52%|█████▏    | 3370/6470 [38:46<31:38,  1.63it/s]
4|3371|Loss: 0.9331870675086975:  52%|█████▏    | 3370/6470 [38:46<31:38,  1.63it/s]
4|3371|Loss: 0.9331870675086975:  52%|█████▏    | 3371/6470 [38:46<32:37,  1.58it/s]
4|3372|Loss: 0.8863943219184875:  52%|█████▏    | 3371/6470 [38:47<32:37,  1.58it/s]
4|3372|Loss: 0.8863943219184875:  52%|█████▏    | 3372/6470 [38:47<35:48,  1.44it/s]
4|3373|Loss: 0.8824209570884705:  52%|█████▏    | 3372/6470 [38:47<35:48,  1.44it/s]
4|3373|Loss: 0.8824209570884705:  52%|█████▏    | 3373/6470 [38:48<35:31,  1.45it/s]
4|3374|Loss: 0.7548010349273682:  52%|█████▏    | 3373/6470 [38:48<35:31,  1.45it/s]
4|3374|Loss: 0.7548010349273682:  52%|█████▏    | 3374/6470 [38:48<33:48,  1.53it/s]
4|3375|Loss: 0.7971217632293701:  52%|█████▏    | 3374/6470 [38:49<33:48,  1.53it/s]
4|3375|Loss: 0.7971217632293701:  52%|█████▏    | 3375/6470 [38:49<30:00,  1.72it/s]
4|3376|Loss: 0.8199325203895569:  52%|█████▏    | 3375/6470 [38:49<30:00,  1.72it/s]
4|3376|Loss: 0.8199325203895569:  52%|█████▏    | 3376/6470 [38:49<26:32,  1.94it/s]
4|3377|Loss: 0.875916600227356:  52%|█████▏    | 3376/6470 [38:49<26:32,  1.94it/s] 
4|3377|Loss: 0.875916600227356:  52%|█████▏    | 3377/6470 [38:50<30:14,  1.70it/s]
4|3378|Loss: 0.7605977058410645:  52%|█████▏    | 3377/6470 [38:50<30:14,  1.70it/s]
4|3378|Loss: 0.7605977058410645:  52%|█████▏    | 3378/6470 [38:51<33:00,  1.56it/s]
4|3379|Loss: 0.8416406512260437:  52%|█████▏    | 3378/6470 [38:51<33:00,  1.56it/s]
4|3379|Loss: 0.8416406512260437:  52%|█████▏    | 3379/6470 [38:51<34:37,  1.49it/s]
4|3380|Loss: 0.7944134473800659:  52%|█████▏    | 3379/6470 [38:52<34:37,  1.49it/s]
4|3380|Loss: 0.7944134473800659:  52%|█████▏    | 3380/6470 [38:52<37:48,  1.36it/s]
4|3381|Loss: 0.8798865675926208:  52%|█████▏    | 3380/6470 [38:53<37:48,  1.36it/s]
4|3381|Loss: 0.8798865675926208:  52%|█████▏    | 3381/6470 [38:53<39:08,  1.32it/s]
4|3382|Loss: 0.8569936752319336:  52%|█████▏    | 3381/6470 [38:53<39:08,  1.32it/s]
4|3382|Loss: 0.8569936752319336:  52%|█████▏    | 3382/6470 [38:54<36:47,  1.40it/s]
4|3383|Loss: 0.7390755414962769:  52%|█████▏    | 3382/6470 [38:54<36:47,  1.40it/s]
4|3383|Loss: 0.7390755414962769:  52%|█████▏    | 3383/6470 [38:54<34:02,  1.51it/s]
4|3384|Loss: 0.7936553359031677:  52%|█████▏    | 3383/6470 [38:54<34:02,  1.51it/s]
4|3384|Loss: 0.7936553359031677:  52%|█████▏    | 3384/6470 [38:55<31:01,  1.66it/s]
4|3385|Loss: 0.9439486265182495:  52%|█████▏    | 3384/6470 [38:55<31:01,  1.66it/s]
4|3385|Loss: 0.9439486265182495:  52%|█████▏    | 3385/6470 [38:55<30:55,  1.66it/s]
4|3386|Loss: 0.7912605404853821:  52%|█████▏    | 3385/6470 [38:56<30:55,  1.66it/s]
4|3386|Loss: 0.7912605404853821:  52%|█████▏    | 3386/6470 [38:56<30:44,  1.67it/s]
4|3387|Loss: 0.8851237893104553:  52%|█████▏    | 3386/6470 [38:56<30:44,  1.67it/s]
4|3387|Loss: 0.8851237893104553:  52%|█████▏    | 3387/6470 [38:56<31:04,  1.65it/s]
4|3388|Loss: 1.0535260438919067:  52%|█████▏    | 3387/6470 [38:57<31:04,  1.65it/s]
4|3388|Loss: 1.0535260438919067:  52%|█████▏    | 3388/6470 [38:57<32:56,  1.56it/s]
4|3389|Loss: 0.8502979278564453:  52%|█████▏    | 3388/6470 [38:58<32:56,  1.56it/s]
4|3389|Loss: 0.8502979278564453:  52%|█████▏    | 3389/6470 [38:58<33:18,  1.54it/s]
4|3390|Loss: 0.8483249545097351:  52%|█████▏    | 3389/6470 [38:58<33:18,  1.54it/s]
4|3390|Loss: 0.8483249545097351:  52%|█████▏    | 3390/6470 [38:58<32:53,  1.56it/s]
4|3391|Loss: 0.8735290765762329:  52%|█████▏    | 3390/6470 [38:59<32:53,  1.56it/s]
4|3391|Loss: 0.8735290765762329:  52%|█████▏    | 3391/6470 [38:59<31:29,  1.63it/s]
4|3392|Loss: 0.9029989838600159:  52%|█████▏    | 3391/6470 [38:59<31:29,  1.63it/s]
4|3392|Loss: 0.9029989838600159:  52%|█████▏    | 3392/6470 [39:00<31:34,  1.62it/s]
4|3393|Loss: 0.790218710899353:  52%|█████▏    | 3392/6470 [39:00<31:34,  1.62it/s] 
4|3393|Loss: 0.790218710899353:  52%|█████▏    | 3393/6470 [39:00<33:15,  1.54it/s]
4|3394|Loss: 0.7998635172843933:  52%|█████▏    | 3393/6470 [39:01<33:15,  1.54it/s]
4|3394|Loss: 0.7998635172843933:  52%|█████▏    | 3394/6470 [39:01<34:51,  1.47it/s]
4|3395|Loss: 0.786223292350769:  52%|█████▏    | 3394/6470 [39:02<34:51,  1.47it/s] 
4|3395|Loss: 0.786223292350769:  52%|█████▏    | 3395/6470 [39:02<35:58,  1.42it/s]
4|3396|Loss: 0.8337278962135315:  52%|█████▏    | 3395/6470 [39:02<35:58,  1.42it/s]
4|3396|Loss: 0.8337278962135315:  52%|█████▏    | 3396/6470 [39:03<38:45,  1.32it/s]
4|3397|Loss: 0.7068788409233093:  52%|█████▏    | 3396/6470 [39:03<38:45,  1.32it/s]
4|3397|Loss: 0.7068788409233093:  53%|█████▎    | 3397/6470 [39:03<35:35,  1.44it/s]
4|3398|Loss: 0.796896755695343:  53%|█████▎    | 3397/6470 [39:04<35:35,  1.44it/s] 
4|3398|Loss: 0.796896755695343:  53%|█████▎    | 3398/6470 [39:04<34:45,  1.47it/s]
4|3399|Loss: 0.7687798738479614:  53%|█████▎    | 3398/6470 [39:04<34:45,  1.47it/s]
4|3399|Loss: 0.7687798738479614:  53%|█████▎    | 3399/6470 [39:05<36:11,  1.41it/s]
4|3400|Loss: 0.9365480542182922:  53%|█████▎    | 3399/6470 [39:05<36:11,  1.41it/s]
4|3400|Loss: 0.9365480542182922:  53%|█████▎    | 3400/6470 [39:05<35:28,  1.44it/s]
4|3401|Loss: 0.8294758200645447:  53%|█████▎    | 3400/6470 [39:06<35:28,  1.44it/s]
4|3401|Loss: 0.8294758200645447:  53%|█████▎    | 3401/6470 [39:06<35:20,  1.45it/s]
4|3402|Loss: 0.6980554461479187:  53%|█████▎    | 3401/6470 [39:07<35:20,  1.45it/s]
4|3402|Loss: 0.6980554461479187:  53%|█████▎    | 3402/6470 [39:07<37:45,  1.35it/s]
4|3403|Loss: 0.9104499220848083:  53%|█████▎    | 3402/6470 [39:07<37:45,  1.35it/s]
4|3403|Loss: 0.9104499220848083:  53%|█████▎    | 3403/6470 [39:08<36:51,  1.39it/s]
4|3404|Loss: 0.8307786583900452:  53%|█████▎    | 3403/6470 [39:08<36:51,  1.39it/s]
4|3404|Loss: 0.8307786583900452:  53%|█████▎    | 3404/6470 [39:08<38:25,  1.33it/s]
4|3405|Loss: 0.8861106038093567:  53%|█████▎    | 3404/6470 [39:09<38:25,  1.33it/s]
4|3405|Loss: 0.8861106038093567:  53%|█████▎    | 3405/6470 [39:09<36:26,  1.40it/s]
4|3406|Loss: 0.9924896359443665:  53%|█████▎    | 3405/6470 [39:09<36:26,  1.40it/s]
4|3406|Loss: 0.9924896359443665:  53%|█████▎    | 3406/6470 [39:10<35:00,  1.46it/s]
4|3407|Loss: 0.7857413291931152:  53%|█████▎    | 3406/6470 [39:10<35:00,  1.46it/s]
4|3407|Loss: 0.7857413291931152:  53%|█████▎    | 3407/6470 [39:10<35:09,  1.45it/s]
4|3408|Loss: 0.7619854807853699:  53%|█████▎    | 3407/6470 [39:11<35:09,  1.45it/s]
4|3408|Loss: 0.7619854807853699:  53%|█████▎    | 3408/6470 [39:11<33:56,  1.50it/s]
4|3409|Loss: 0.8779125213623047:  53%|█████▎    | 3408/6470 [39:11<33:56,  1.50it/s]
4|3409|Loss: 0.8779125213623047:  53%|█████▎    | 3409/6470 [39:11<31:47,  1.60it/s]
4|3410|Loss: 0.9614130854606628:  53%|█████▎    | 3409/6470 [39:12<31:47,  1.60it/s]
4|3410|Loss: 0.9614130854606628:  53%|█████▎    | 3410/6470 [39:12<31:04,  1.64it/s]
4|3411|Loss: 0.7690126895904541:  53%|█████▎    | 3410/6470 [39:12<31:04,  1.64it/s]
4|3411|Loss: 0.7690126895904541:  53%|█████▎    | 3411/6470 [39:13<34:04,  1.50it/s]
4|3412|Loss: 0.9651131629943848:  53%|█████▎    | 3411/6470 [39:13<34:04,  1.50it/s]
4|3412|Loss: 0.9651131629943848:  53%|█████▎    | 3412/6470 [39:13<30:57,  1.65it/s]
4|3413|Loss: 0.8720667958259583:  53%|█████▎    | 3412/6470 [39:14<30:57,  1.65it/s]
4|3413|Loss: 0.8720667958259583:  53%|█████▎    | 3413/6470 [39:14<28:28,  1.79it/s]
4|3414|Loss: 0.7928345203399658:  53%|█████▎    | 3413/6470 [39:14<28:28,  1.79it/s]
4|3414|Loss: 0.7928345203399658:  53%|█████▎    | 3414/6470 [39:15<31:21,  1.62it/s]
4|3415|Loss: 0.7622019648551941:  53%|█████▎    | 3414/6470 [39:15<31:21,  1.62it/s]
4|3415|Loss: 0.7622019648551941:  53%|█████▎    | 3415/6470 [39:15<32:23,  1.57it/s]
4|3416|Loss: 0.919336199760437:  53%|█████▎    | 3415/6470 [39:16<32:23,  1.57it/s] 
4|3416|Loss: 0.919336199760437:  53%|█████▎    | 3416/6470 [39:16<33:32,  1.52it/s]
4|3417|Loss: 0.9185112714767456:  53%|█████▎    | 3416/6470 [39:16<33:32,  1.52it/s]
4|3417|Loss: 0.9185112714767456:  53%|█████▎    | 3417/6470 [39:17<34:53,  1.46it/s]
4|3418|Loss: 0.7794866561889648:  53%|█████▎    | 3417/6470 [39:17<34:53,  1.46it/s]
4|3418|Loss: 0.7794866561889648:  53%|█████▎    | 3418/6470 [39:18<37:14,  1.37it/s]
4|3419|Loss: 0.8968887329101562:  53%|█████▎    | 3418/6470 [39:18<37:14,  1.37it/s]
4|3419|Loss: 0.8968887329101562:  53%|█████▎    | 3419/6470 [39:18<37:56,  1.34it/s]
4|3420|Loss: 0.9308320879936218:  53%|█████▎    | 3419/6470 [39:19<37:56,  1.34it/s]
4|3420|Loss: 0.9308320879936218:  53%|█████▎    | 3420/6470 [39:19<34:20,  1.48it/s]
4|3421|Loss: 0.78780597448349:  53%|█████▎    | 3420/6470 [39:19<34:20,  1.48it/s]  
4|3421|Loss: 0.78780597448349:  53%|█████▎    | 3421/6470 [39:19<32:26,  1.57it/s]
4|3422|Loss: 0.784142792224884:  53%|█████▎    | 3421/6470 [39:20<32:26,  1.57it/s]
4|3422|Loss: 0.784142792224884:  53%|█████▎    | 3422/6470 [39:20<31:37,  1.61it/s]
4|3423|Loss: 0.7368594408035278:  53%|█████▎    | 3422/6470 [39:20<31:37,  1.61it/s]
4|3423|Loss: 0.7368594408035278:  53%|█████▎    | 3423/6470 [39:21<31:45,  1.60it/s]
4|3424|Loss: 0.7712526917457581:  53%|█████▎    | 3423/6470 [39:21<31:45,  1.60it/s]
4|3424|Loss: 0.7712526917457581:  53%|█████▎    | 3424/6470 [39:21<32:05,  1.58it/s]
4|3425|Loss: 0.6142553091049194:  53%|█████▎    | 3424/6470 [39:22<32:05,  1.58it/s]
4|3425|Loss: 0.6142553091049194:  53%|█████▎    | 3425/6470 [39:22<34:47,  1.46it/s]
4|3426|Loss: 0.6824474334716797:  53%|█████▎    | 3425/6470 [39:23<34:47,  1.46it/s]
4|3426|Loss: 0.6824474334716797:  53%|█████▎    | 3426/6470 [39:23<37:12,  1.36it/s]
4|3427|Loss: 0.9292914867401123:  53%|█████▎    | 3426/6470 [39:23<37:12,  1.36it/s]
4|3427|Loss: 0.9292914867401123:  53%|█████▎    | 3427/6470 [39:23<35:15,  1.44it/s]
4|3428|Loss: 0.9125828742980957:  53%|█████▎    | 3427/6470 [39:24<35:15,  1.44it/s]
4|3428|Loss: 0.9125828742980957:  53%|█████▎    | 3428/6470 [39:24<34:53,  1.45it/s]
4|3429|Loss: 0.7209791541099548:  53%|█████▎    | 3428/6470 [39:25<34:53,  1.45it/s]
4|3429|Loss: 0.7209791541099548:  53%|█████▎    | 3429/6470 [39:25<35:24,  1.43it/s]
4|3430|Loss: 0.8576123714447021:  53%|█████▎    | 3429/6470 [39:25<35:24,  1.43it/s]
4|3430|Loss: 0.8576123714447021:  53%|█████▎    | 3430/6470 [39:26<35:58,  1.41it/s]
4|3431|Loss: 0.8088034987449646:  53%|█████▎    | 3430/6470 [39:26<35:58,  1.41it/s]
4|3431|Loss: 0.8088034987449646:  53%|█████▎    | 3431/6470 [39:26<37:09,  1.36it/s]
4|3432|Loss: 0.8207801580429077:  53%|█████▎    | 3431/6470 [39:27<37:09,  1.36it/s]
4|3432|Loss: 0.8207801580429077:  53%|█████▎    | 3432/6470 [39:27<35:30,  1.43it/s]
4|3433|Loss: 0.8326327204704285:  53%|█████▎    | 3432/6470 [39:27<35:30,  1.43it/s]
4|3433|Loss: 0.8326327204704285:  53%|█████▎    | 3433/6470 [39:28<36:13,  1.40it/s]
4|3434|Loss: 0.7048571109771729:  53%|█████▎    | 3433/6470 [39:28<36:13,  1.40it/s]
4|3434|Loss: 0.7048571109771729:  53%|█████▎    | 3434/6470 [39:29<37:23,  1.35it/s]
4|3435|Loss: 0.825647234916687:  53%|█████▎    | 3434/6470 [39:29<37:23,  1.35it/s] 
4|3435|Loss: 0.825647234916687:  53%|█████▎    | 3435/6470 [39:29<37:33,  1.35it/s]
4|3436|Loss: 0.7215661406517029:  53%|█████▎    | 3435/6470 [39:30<37:33,  1.35it/s]
4|3436|Loss: 0.7215661406517029:  53%|█████▎    | 3436/6470 [39:30<37:24,  1.35it/s]
4|3437|Loss: 0.7292193174362183:  53%|█████▎    | 3436/6470 [39:30<37:24,  1.35it/s]
4|3437|Loss: 0.7292193174362183:  53%|█████▎    | 3437/6470 [39:31<37:15,  1.36it/s]
4|3438|Loss: 0.8891244530677795:  53%|█████▎    | 3437/6470 [39:31<37:15,  1.36it/s]
4|3438|Loss: 0.8891244530677795:  53%|█████▎    | 3438/6470 [39:31<34:29,  1.47it/s]
4|3439|Loss: 0.7801520824432373:  53%|█████▎    | 3438/6470 [39:32<34:29,  1.47it/s]
4|3439|Loss: 0.7801520824432373:  53%|█████▎    | 3439/6470 [39:32<30:25,  1.66it/s]
4|3440|Loss: 0.826469898223877:  53%|█████▎    | 3439/6470 [39:32<30:25,  1.66it/s] 
4|3440|Loss: 0.826469898223877:  53%|█████▎    | 3440/6470 [39:32<31:32,  1.60it/s]
4|3441|Loss: 0.8452416062355042:  53%|█████▎    | 3440/6470 [39:33<31:32,  1.60it/s]
4|3441|Loss: 0.8452416062355042:  53%|█████▎    | 3441/6470 [39:33<33:27,  1.51it/s]
4|3442|Loss: 0.9870466589927673:  53%|█████▎    | 3441/6470 [39:34<33:27,  1.51it/s]
4|3442|Loss: 0.9870466589927673:  53%|█████▎    | 3442/6470 [39:34<34:12,  1.48it/s]
4|3443|Loss: 0.8677384853363037:  53%|█████▎    | 3442/6470 [39:34<34:12,  1.48it/s]
4|3443|Loss: 0.8677384853363037:  53%|█████▎    | 3443/6470 [39:35<34:29,  1.46it/s]
4|3444|Loss: 0.823283851146698:  53%|█████▎    | 3443/6470 [39:35<34:29,  1.46it/s] 
4|3444|Loss: 0.823283851146698:  53%|█████▎    | 3444/6470 [39:35<35:26,  1.42it/s]
4|3445|Loss: 1.0291706323623657:  53%|█████▎    | 3444/6470 [39:36<35:26,  1.42it/s]
4|3445|Loss: 1.0291706323623657:  53%|█████▎    | 3445/6470 [39:36<35:00,  1.44it/s]
4|3446|Loss: 0.923632800579071:  53%|█████▎    | 3445/6470 [39:36<35:00,  1.44it/s] 
4|3446|Loss: 0.923632800579071:  53%|█████▎    | 3446/6470 [39:37<33:54,  1.49it/s]
4|3447|Loss: 0.986533522605896:  53%|█████▎    | 3446/6470 [39:37<33:54,  1.49it/s]
4|3447|Loss: 0.986533522605896:  53%|█████▎    | 3447/6470 [39:37<33:00,  1.53it/s]
4|3448|Loss: 0.8139650821685791:  53%|█████▎    | 3447/6470 [39:38<33:00,  1.53it/s]
4|3448|Loss: 0.8139650821685791:  53%|█████▎    | 3448/6470 [39:38<35:33,  1.42it/s]
4|3449|Loss: 0.8621708750724792:  53%|█████▎    | 3448/6470 [39:38<35:33,  1.42it/s]
4|3449|Loss: 0.8621708750724792:  53%|█████▎    | 3449/6470 [39:39<34:15,  1.47it/s]
4|3450|Loss: 0.9096428155899048:  53%|█████▎    | 3449/6470 [39:39<34:15,  1.47it/s]
4|3450|Loss: 0.9096428155899048:  53%|█████▎    | 3450/6470 [39:39<34:26,  1.46it/s]
4|3451|Loss: 0.76573646068573:  53%|█████▎    | 3450/6470 [39:40<34:26,  1.46it/s]  
4|3451|Loss: 0.76573646068573:  53%|█████▎    | 3451/6470 [39:40<33:17,  1.51it/s]
4|3452|Loss: 0.9510759115219116:  53%|█████▎    | 3451/6470 [39:40<33:17,  1.51it/s]
4|3452|Loss: 0.9510759115219116:  53%|█████▎    | 3452/6470 [39:40<30:36,  1.64it/s]
4|3453|Loss: 0.8456254601478577:  53%|█████▎    | 3452/6470 [39:41<30:36,  1.64it/s]
4|3453|Loss: 0.8456254601478577:  53%|█████▎    | 3453/6470 [39:41<31:10,  1.61it/s]
4|3454|Loss: 0.9288134574890137:  53%|█████▎    | 3453/6470 [39:42<31:10,  1.61it/s]
4|3454|Loss: 0.9288134574890137:  53%|█████▎    | 3454/6470 [39:42<31:02,  1.62it/s]
4|3455|Loss: 0.8636517524719238:  53%|█████▎    | 3454/6470 [39:42<31:02,  1.62it/s]
4|3455|Loss: 0.8636517524719238:  53%|█████▎    | 3455/6470 [39:42<30:21,  1.66it/s]
4|3456|Loss: 0.8889269828796387:  53%|█████▎    | 3455/6470 [39:43<30:21,  1.66it/s]
4|3456|Loss: 0.8889269828796387:  53%|█████▎    | 3456/6470 [39:43<30:41,  1.64it/s]
4|3457|Loss: 0.8540742993354797:  53%|█████▎    | 3456/6470 [39:43<30:41,  1.64it/s]
4|3457|Loss: 0.8540742993354797:  53%|█████▎    | 3457/6470 [39:44<30:46,  1.63it/s]
4|3458|Loss: 1.08565092086792:  53%|█████▎    | 3457/6470 [39:44<30:46,  1.63it/s]  
4|3458|Loss: 1.08565092086792:  53%|█████▎    | 3458/6470 [39:44<32:08,  1.56it/s]
4|3459|Loss: 0.9982101321220398:  53%|█████▎    | 3458/6470 [39:45<32:08,  1.56it/s]
4|3459|Loss: 0.9982101321220398:  53%|█████▎    | 3459/6470 [39:45<31:31,  1.59it/s]
4|3460|Loss: 0.745162308216095:  53%|█████▎    | 3459/6470 [39:45<31:31,  1.59it/s] 
4|3460|Loss: 0.745162308216095:  53%|█████▎    | 3460/6470 [39:45<30:44,  1.63it/s]
4|3461|Loss: 0.8637896776199341:  53%|█████▎    | 3460/6470 [39:46<30:44,  1.63it/s]
4|3461|Loss: 0.8637896776199341:  53%|█████▎    | 3461/6470 [39:46<31:49,  1.58it/s]
4|3462|Loss: 0.8762800097465515:  53%|█████▎    | 3461/6470 [39:47<31:49,  1.58it/s]
4|3462|Loss: 0.8762800097465515:  54%|█████▎    | 3462/6470 [39:47<32:07,  1.56it/s]
4|3463|Loss: 0.8381730318069458:  54%|█████▎    | 3462/6470 [39:47<32:07,  1.56it/s]
4|3463|Loss: 0.8381730318069458:  54%|█████▎    | 3463/6470 [39:48<33:12,  1.51it/s]
4|3464|Loss: 0.6963462233543396:  54%|█████▎    | 3463/6470 [39:48<33:12,  1.51it/s]
4|3464|Loss: 0.6963462233543396:  54%|█████▎    | 3464/6470 [39:48<32:52,  1.52it/s]
4|3465|Loss: 0.8698346018791199:  54%|█████▎    | 3464/6470 [39:48<32:52,  1.52it/s]
4|3465|Loss: 0.8698346018791199:  54%|█████▎    | 3465/6470 [39:49<30:57,  1.62it/s]
4|3466|Loss: 0.7858514189720154:  54%|█████▎    | 3465/6470 [39:49<30:57,  1.62it/s]
4|3466|Loss: 0.7858514189720154:  54%|█████▎    | 3466/6470 [39:49<30:49,  1.62it/s]
4|3467|Loss: 0.8880854249000549:  54%|█████▎    | 3466/6470 [39:50<30:49,  1.62it/s]
4|3467|Loss: 0.8880854249000549:  54%|█████▎    | 3467/6470 [39:50<30:34,  1.64it/s]
4|3468|Loss: 0.7074086666107178:  54%|█████▎    | 3467/6470 [39:50<30:34,  1.64it/s]
4|3468|Loss: 0.7074086666107178:  54%|█████▎    | 3468/6470 [39:51<32:03,  1.56it/s]
4|3469|Loss: 0.8029505610466003:  54%|█████▎    | 3468/6470 [39:51<32:03,  1.56it/s]
4|3469|Loss: 0.8029505610466003:  54%|█████▎    | 3469/6470 [39:51<33:31,  1.49it/s]
4|3470|Loss: 0.9300367832183838:  54%|█████▎    | 3469/6470 [39:52<33:31,  1.49it/s]
4|3470|Loss: 0.9300367832183838:  54%|█████▎    | 3470/6470 [39:52<34:38,  1.44it/s]
4|3471|Loss: 0.7275543212890625:  54%|█████▎    | 3470/6470 [39:53<34:38,  1.44it/s]
4|3471|Loss: 0.7275543212890625:  54%|█████▎    | 3471/6470 [39:53<35:25,  1.41it/s]
4|3472|Loss: 0.7456332445144653:  54%|█████▎    | 3471/6470 [39:53<35:25,  1.41it/s]
4|3472|Loss: 0.7456332445144653:  54%|█████▎    | 3472/6470 [39:54<36:06,  1.38it/s]
4|3473|Loss: 1.0281696319580078:  54%|█████▎    | 3472/6470 [39:54<36:06,  1.38it/s]
4|3473|Loss: 1.0281696319580078:  54%|█████▎    | 3473/6470 [39:54<36:20,  1.37it/s]
4|3474|Loss: 0.7349193096160889:  54%|█████▎    | 3473/6470 [39:55<36:20,  1.37it/s]
4|3474|Loss: 0.7349193096160889:  54%|█████▎    | 3474/6470 [39:55<35:03,  1.42it/s]
4|3475|Loss: 0.8188333511352539:  54%|█████▎    | 3474/6470 [39:55<35:03,  1.42it/s]
4|3475|Loss: 0.8188333511352539:  54%|█████▎    | 3475/6470 [39:56<35:09,  1.42it/s]
4|3476|Loss: 1.0132181644439697:  54%|█████▎    | 3475/6470 [39:56<35:09,  1.42it/s]
4|3476|Loss: 1.0132181644439697:  54%|█████▎    | 3476/6470 [39:56<34:12,  1.46it/s]
4|3477|Loss: 0.8540611267089844:  54%|█████▎    | 3476/6470 [39:57<34:12,  1.46it/s]
4|3477|Loss: 0.8540611267089844:  54%|█████▎    | 3477/6470 [39:57<33:22,  1.49it/s]
4|3478|Loss: 0.7688742876052856:  54%|█████▎    | 3477/6470 [39:57<33:22,  1.49it/s]
4|3478|Loss: 0.7688742876052856:  54%|█████▍    | 3478/6470 [39:58<32:41,  1.53it/s]
4|3479|Loss: 0.8180579543113708:  54%|█████▍    | 3478/6470 [39:58<32:41,  1.53it/s]
4|3479|Loss: 0.8180579543113708:  54%|█████▍    | 3479/6470 [39:58<33:06,  1.51it/s]
4|3480|Loss: 0.8804883360862732:  54%|█████▍    | 3479/6470 [39:59<33:06,  1.51it/s]
4|3480|Loss: 0.8804883360862732:  54%|█████▍    | 3480/6470 [39:59<33:43,  1.48it/s]
4|3481|Loss: 0.8601986169815063:  54%|█████▍    | 3480/6470 [39:59<33:43,  1.48it/s]
4|3481|Loss: 0.8601986169815063:  54%|█████▍    | 3481/6470 [40:00<35:41,  1.40it/s]
4|3482|Loss: 0.8724331259727478:  54%|█████▍    | 3481/6470 [40:00<35:41,  1.40it/s]
4|3482|Loss: 0.8724331259727478:  54%|█████▍    | 3482/6470 [40:01<35:59,  1.38it/s]
4|3483|Loss: 0.7306387424468994:  54%|█████▍    | 3482/6470 [40:01<35:59,  1.38it/s]
4|3483|Loss: 0.7306387424468994:  54%|█████▍    | 3483/6470 [40:01<35:43,  1.39it/s]
4|3484|Loss: 0.9779432415962219:  54%|█████▍    | 3483/6470 [40:02<35:43,  1.39it/s]
4|3484|Loss: 0.9779432415962219:  54%|█████▍    | 3484/6470 [40:02<34:30,  1.44it/s]
4|3485|Loss: 0.8489745855331421:  54%|█████▍    | 3484/6470 [40:02<34:30,  1.44it/s]
4|3485|Loss: 0.8489745855331421:  54%|█████▍    | 3485/6470 [40:03<35:32,  1.40it/s]
4|3486|Loss: 0.8681430220603943:  54%|█████▍    | 3485/6470 [40:03<35:32,  1.40it/s]
4|3486|Loss: 0.8681430220603943:  54%|█████▍    | 3486/6470 [40:03<34:44,  1.43it/s]
4|3487|Loss: 0.6194289326667786:  54%|█████▍    | 3486/6470 [40:04<34:44,  1.43it/s]
4|3487|Loss: 0.6194289326667786:  54%|█████▍    | 3487/6470 [40:04<32:26,  1.53it/s]
4|3488|Loss: 0.8059717416763306:  54%|█████▍    | 3487/6470 [40:04<32:26,  1.53it/s]
4|3488|Loss: 0.8059717416763306:  54%|█████▍    | 3488/6470 [40:04<29:59,  1.66it/s]
4|3489|Loss: 0.7993921041488647:  54%|█████▍    | 3488/6470 [40:05<29:59,  1.66it/s]
4|3489|Loss: 0.7993921041488647:  54%|█████▍    | 3489/6470 [40:05<29:59,  1.66it/s]
4|3490|Loss: 0.8792860507965088:  54%|█████▍    | 3489/6470 [40:05<29:59,  1.66it/s]
4|3490|Loss: 0.8792860507965088:  54%|█████▍    | 3490/6470 [40:06<30:31,  1.63it/s]
4|3491|Loss: 0.7048196792602539:  54%|█████▍    | 3490/6470 [40:06<30:31,  1.63it/s]
4|3491|Loss: 0.7048196792602539:  54%|█████▍    | 3491/6470 [40:06<28:58,  1.71it/s]
4|3492|Loss: 0.8850967884063721:  54%|█████▍    | 3491/6470 [40:06<28:58,  1.71it/s]
4|3492|Loss: 0.8850967884063721:  54%|█████▍    | 3492/6470 [40:07<29:55,  1.66it/s]
4|3493|Loss: 0.7747045755386353:  54%|█████▍    | 3492/6470 [40:07<29:55,  1.66it/s]
4|3493|Loss: 0.7747045755386353:  54%|█████▍    | 3493/6470 [40:07<30:25,  1.63it/s]
4|3494|Loss: 0.8013189435005188:  54%|█████▍    | 3493/6470 [40:08<30:25,  1.63it/s]
4|3494|Loss: 0.8013189435005188:  54%|█████▍    | 3494/6470 [40:08<29:36,  1.68it/s]
4|3495|Loss: 0.7831031680107117:  54%|█████▍    | 3494/6470 [40:08<29:36,  1.68it/s]
4|3495|Loss: 0.7831031680107117:  54%|█████▍    | 3495/6470 [40:09<29:54,  1.66it/s]
4|3496|Loss: 0.8903800249099731:  54%|█████▍    | 3495/6470 [40:09<29:54,  1.66it/s]
4|3496|Loss: 0.8903800249099731:  54%|█████▍    | 3496/6470 [40:09<29:51,  1.66it/s]
4|3497|Loss: 0.6763122081756592:  54%|█████▍    | 3496/6470 [40:10<29:51,  1.66it/s]
4|3497|Loss: 0.6763122081756592:  54%|█████▍    | 3497/6470 [40:10<31:15,  1.59it/s]
4|3498|Loss: 0.754479706287384:  54%|█████▍    | 3497/6470 [40:10<31:15,  1.59it/s] 
4|3498|Loss: 0.754479706287384:  54%|█████▍    | 3498/6470 [40:11<34:06,  1.45it/s]
4|3499|Loss: 1.0270106792449951:  54%|█████▍    | 3498/6470 [40:11<34:06,  1.45it/s]
4|3499|Loss: 1.0270106792449951:  54%|█████▍    | 3499/6470 [40:11<33:50,  1.46it/s]
4|3500|Loss: 0.9133346676826477:  54%|█████▍    | 3499/6470 [40:12<33:50,  1.46it/s]
4|3500|Loss: 0.9133346676826477:  54%|█████▍    | 3500/6470 [40:12<34:10,  1.45it/s]
4|3501|Loss: 0.7818475961685181:  54%|█████▍    | 3500/6470 [40:12<34:10,  1.45it/s]
4|3501|Loss: 0.7818475961685181:  54%|█████▍    | 3501/6470 [40:13<32:37,  1.52it/s]
4|3502|Loss: 0.815127432346344:  54%|█████▍    | 3501/6470 [40:13<32:37,  1.52it/s] 
4|3502|Loss: 0.815127432346344:  54%|█████▍    | 3502/6470 [40:13<31:47,  1.56it/s]
4|3503|Loss: 0.741568386554718:  54%|█████▍    | 3502/6470 [40:14<31:47,  1.56it/s]
4|3503|Loss: 0.741568386554718:  54%|█████▍    | 3503/6470 [40:14<30:56,  1.60it/s]
4|3504|Loss: 0.8026458024978638:  54%|█████▍    | 3503/6470 [40:14<30:56,  1.60it/s]
4|3504|Loss: 0.8026458024978638:  54%|█████▍    | 3504/6470 [40:15<33:29,  1.48it/s]
4|3505|Loss: 0.7459940314292908:  54%|█████▍    | 3504/6470 [40:15<33:29,  1.48it/s]
4|3505|Loss: 0.7459940314292908:  54%|█████▍    | 3505/6470 [40:15<35:52,  1.38it/s]
4|3506|Loss: 0.9455702304840088:  54%|█████▍    | 3505/6470 [40:16<35:52,  1.38it/s]
4|3506|Loss: 0.9455702304840088:  54%|█████▍    | 3506/6470 [40:16<37:46,  1.31it/s]
4|3507|Loss: 0.7615021467208862:  54%|█████▍    | 3506/6470 [40:17<37:46,  1.31it/s]
4|3507|Loss: 0.7615021467208862:  54%|█████▍    | 3507/6470 [40:17<37:21,  1.32it/s]
4|3508|Loss: 0.8488526344299316:  54%|█████▍    | 3507/6470 [40:17<37:21,  1.32it/s]
4|3508|Loss: 0.8488526344299316:  54%|█████▍    | 3508/6470 [40:18<36:34,  1.35it/s]
4|3509|Loss: 0.8467634320259094:  54%|█████▍    | 3508/6470 [40:18<36:34,  1.35it/s]
4|3509|Loss: 0.8467634320259094:  54%|█████▍    | 3509/6470 [40:18<35:17,  1.40it/s]
4|3510|Loss: 0.7714034914970398:  54%|█████▍    | 3509/6470 [40:19<35:17,  1.40it/s]
4|3510|Loss: 0.7714034914970398:  54%|█████▍    | 3510/6470 [40:19<33:19,  1.48it/s]
4|3511|Loss: 0.8726098537445068:  54%|█████▍    | 3510/6470 [40:19<33:19,  1.48it/s]
4|3511|Loss: 0.8726098537445068:  54%|█████▍    | 3511/6470 [40:20<32:19,  1.53it/s]
4|3512|Loss: 0.987365186214447:  54%|█████▍    | 3511/6470 [40:20<32:19,  1.53it/s] 
4|3512|Loss: 0.987365186214447:  54%|█████▍    | 3512/6470 [40:20<32:05,  1.54it/s]
4|3513|Loss: 0.8019505143165588:  54%|█████▍    | 3512/6470 [40:21<32:05,  1.54it/s]
4|3513|Loss: 0.8019505143165588:  54%|█████▍    | 3513/6470 [40:21<32:41,  1.51it/s]
4|3514|Loss: 0.8993614315986633:  54%|█████▍    | 3513/6470 [40:21<32:41,  1.51it/s]
4|3514|Loss: 0.8993614315986633:  54%|█████▍    | 3514/6470 [40:22<32:29,  1.52it/s]
4|3515|Loss: 0.9542896747589111:  54%|█████▍    | 3514/6470 [40:22<32:29,  1.52it/s]
4|3515|Loss: 0.9542896747589111:  54%|█████▍    | 3515/6470 [40:22<30:34,  1.61it/s]
4|3516|Loss: 0.9116461277008057:  54%|█████▍    | 3515/6470 [40:22<30:34,  1.61it/s]
4|3516|Loss: 0.9116461277008057:  54%|█████▍    | 3516/6470 [40:23<29:08,  1.69it/s]
4|3517|Loss: 0.7216329574584961:  54%|█████▍    | 3516/6470 [40:23<29:08,  1.69it/s]
4|3517|Loss: 0.7216329574584961:  54%|█████▍    | 3517/6470 [40:23<32:02,  1.54it/s]
4|3518|Loss: 0.9240599870681763:  54%|█████▍    | 3517/6470 [40:24<32:02,  1.54it/s]
4|3518|Loss: 0.9240599870681763:  54%|█████▍    | 3518/6470 [40:24<33:40,  1.46it/s]
4|3519|Loss: 0.7717686295509338:  54%|█████▍    | 3518/6470 [40:25<33:40,  1.46it/s]
4|3519|Loss: 0.7717686295509338:  54%|█████▍    | 3519/6470 [40:25<35:32,  1.38it/s]
4|3520|Loss: 0.931678295135498:  54%|█████▍    | 3519/6470 [40:25<35:32,  1.38it/s] 
4|3520|Loss: 0.931678295135498:  54%|█████▍    | 3520/6470 [40:26<33:20,  1.47it/s]
4|3521|Loss: 0.7438163757324219:  54%|█████▍    | 3520/6470 [40:26<33:20,  1.47it/s]
4|3521|Loss: 0.7438163757324219:  54%|█████▍    | 3521/6470 [40:26<31:43,  1.55it/s]
4|3522|Loss: 0.7197880148887634:  54%|█████▍    | 3521/6470 [40:26<31:43,  1.55it/s]
4|3522|Loss: 0.7197880148887634:  54%|█████▍    | 3522/6470 [40:27<30:01,  1.64it/s]
4|3523|Loss: 0.7977215647697449:  54%|█████▍    | 3522/6470 [40:27<30:01,  1.64it/s]
4|3523|Loss: 0.7977215647697449:  54%|█████▍    | 3523/6470 [40:27<30:03,  1.63it/s]
4|3524|Loss: 0.8427924513816833:  54%|█████▍    | 3523/6470 [40:28<30:03,  1.63it/s]
4|3524|Loss: 0.8427924513816833:  54%|█████▍    | 3524/6470 [40:28<31:56,  1.54it/s]
4|3525|Loss: 0.9168586134910583:  54%|█████▍    | 3524/6470 [40:28<31:56,  1.54it/s]
4|3525|Loss: 0.9168586134910583:  54%|█████▍    | 3525/6470 [40:29<31:54,  1.54it/s]
4|3526|Loss: 0.8255256414413452:  54%|█████▍    | 3525/6470 [40:29<31:54,  1.54it/s]
4|3526|Loss: 0.8255256414413452:  54%|█████▍    | 3526/6470 [40:29<30:52,  1.59it/s]
4|3527|Loss: 0.7900121808052063:  54%|█████▍    | 3526/6470 [40:30<30:52,  1.59it/s]
4|3527|Loss: 0.7900121808052063:  55%|█████▍    | 3527/6470 [40:30<31:35,  1.55it/s]
4|3528|Loss: 0.9336801171302795:  55%|█████▍    | 3527/6470 [40:30<31:35,  1.55it/s]
4|3528|Loss: 0.9336801171302795:  55%|█████▍    | 3528/6470 [40:31<33:59,  1.44it/s]
4|3529|Loss: 0.7250402569770813:  55%|█████▍    | 3528/6470 [40:31<33:59,  1.44it/s]
4|3529|Loss: 0.7250402569770813:  55%|█████▍    | 3529/6470 [40:31<33:54,  1.45it/s]
4|3530|Loss: 0.9673498272895813:  55%|█████▍    | 3529/6470 [40:32<33:54,  1.45it/s]
4|3530|Loss: 0.9673498272895813:  55%|█████▍    | 3530/6470 [40:32<32:22,  1.51it/s]
4|3531|Loss: 0.7262666821479797:  55%|█████▍    | 3530/6470 [40:32<32:22,  1.51it/s]
4|3531|Loss: 0.7262666821479797:  55%|█████▍    | 3531/6470 [40:33<33:09,  1.48it/s]
4|3532|Loss: 0.7466476559638977:  55%|█████▍    | 3531/6470 [40:33<33:09,  1.48it/s]
4|3532|Loss: 0.7466476559638977:  55%|█████▍    | 3532/6470 [40:33<34:11,  1.43it/s]
4|3533|Loss: 0.9263187646865845:  55%|█████▍    | 3532/6470 [40:34<34:11,  1.43it/s]
4|3533|Loss: 0.9263187646865845:  55%|█████▍    | 3533/6470 [40:34<34:43,  1.41it/s]
4|3534|Loss: 0.8966261744499207:  55%|█████▍    | 3533/6470 [40:35<34:43,  1.41it/s]
4|3534|Loss: 0.8966261744499207:  55%|█████▍    | 3534/6470 [40:35<33:41,  1.45it/s]
4|3535|Loss: 0.8344441056251526:  55%|█████▍    | 3534/6470 [40:35<33:41,  1.45it/s]
4|3535|Loss: 0.8344441056251526:  55%|█████▍    | 3535/6470 [40:36<34:12,  1.43it/s]
4|3536|Loss: 0.8020026087760925:  55%|█████▍    | 3535/6470 [40:36<34:12,  1.43it/s]
4|3536|Loss: 0.8020026087760925:  55%|█████▍    | 3536/6470 [40:36<34:44,  1.41it/s]
4|3537|Loss: 0.7573655247688293:  55%|█████▍    | 3536/6470 [40:37<34:44,  1.41it/s]
4|3537|Loss: 0.7573655247688293:  55%|█████▍    | 3537/6470 [40:37<33:14,  1.47it/s]
4|3538|Loss: 0.8192301392555237:  55%|█████▍    | 3537/6470 [40:37<33:14,  1.47it/s]
4|3538|Loss: 0.8192301392555237:  55%|█████▍    | 3538/6470 [40:38<34:37,  1.41it/s]
4|3539|Loss: 0.7726025581359863:  55%|█████▍    | 3538/6470 [40:38<34:37,  1.41it/s]
4|3539|Loss: 0.7726025581359863:  55%|█████▍    | 3539/6470 [40:38<30:53,  1.58it/s]
4|3540|Loss: 0.6308738589286804:  55%|█████▍    | 3539/6470 [40:39<30:53,  1.58it/s]
4|3540|Loss: 0.6308738589286804:  55%|█████▍    | 3540/6470 [40:39<32:26,  1.51it/s]
4|3541|Loss: 0.7349188327789307:  55%|█████▍    | 3540/6470 [40:39<32:26,  1.51it/s]
4|3541|Loss: 0.7349188327789307:  55%|█████▍    | 3541/6470 [40:40<33:46,  1.45it/s]
4|3542|Loss: 1.0594862699508667:  55%|█████▍    | 3541/6470 [40:40<33:46,  1.45it/s]
4|3542|Loss: 1.0594862699508667:  55%|█████▍    | 3542/6470 [40:40<33:13,  1.47it/s]
4|3543|Loss: 0.8207024335861206:  55%|█████▍    | 3542/6470 [40:41<33:13,  1.47it/s]
4|3543|Loss: 0.8207024335861206:  55%|█████▍    | 3543/6470 [40:41<34:31,  1.41it/s]
4|3544|Loss: 0.9236169457435608:  55%|█████▍    | 3543/6470 [40:41<34:31,  1.41it/s]
4|3544|Loss: 0.9236169457435608:  55%|█████▍    | 3544/6470 [40:42<33:44,  1.45it/s]
4|3545|Loss: 0.6828188300132751:  55%|█████▍    | 3544/6470 [40:42<33:44,  1.45it/s]
4|3545|Loss: 0.6828188300132751:  55%|█████▍    | 3545/6470 [40:42<32:50,  1.48it/s]
4|3546|Loss: 0.7517382502555847:  55%|█████▍    | 3545/6470 [40:43<32:50,  1.48it/s]
4|3546|Loss: 0.7517382502555847:  55%|█████▍    | 3546/6470 [40:43<33:01,  1.48it/s]
4|3547|Loss: 0.971153974533081:  55%|█████▍    | 3546/6470 [40:43<33:01,  1.48it/s] 
4|3547|Loss: 0.971153974533081:  55%|█████▍    | 3547/6470 [40:44<33:30,  1.45it/s]
4|3548|Loss: 0.8812020421028137:  55%|█████▍    | 3547/6470 [40:44<33:30,  1.45it/s]
4|3548|Loss: 0.8812020421028137:  55%|█████▍    | 3548/6470 [40:45<34:48,  1.40it/s]
4|3549|Loss: 0.8484141826629639:  55%|█████▍    | 3548/6470 [40:45<34:48,  1.40it/s]
4|3549|Loss: 0.8484141826629639:  55%|█████▍    | 3549/6470 [40:45<31:56,  1.52it/s]
4|3550|Loss: 0.9226987957954407:  55%|█████▍    | 3549/6470 [40:45<31:56,  1.52it/s]
4|3550|Loss: 0.9226987957954407:  55%|█████▍    | 3550/6470 [40:46<30:49,  1.58it/s]
4|3551|Loss: 0.9966177344322205:  55%|█████▍    | 3550/6470 [40:46<30:49,  1.58it/s]
4|3551|Loss: 0.9966177344322205:  55%|█████▍    | 3551/6470 [40:46<32:09,  1.51it/s]
4|3552|Loss: 0.8413431644439697:  55%|█████▍    | 3551/6470 [40:47<32:09,  1.51it/s]
4|3552|Loss: 0.8413431644439697:  55%|█████▍    | 3552/6470 [40:47<33:24,  1.46it/s]
4|3553|Loss: 0.8330753445625305:  55%|█████▍    | 3552/6470 [40:48<33:24,  1.46it/s]
4|3553|Loss: 0.8330753445625305:  55%|█████▍    | 3553/6470 [40:48<34:48,  1.40it/s]
4|3554|Loss: 0.8479963541030884:  55%|█████▍    | 3553/6470 [40:48<34:48,  1.40it/s]
4|3554|Loss: 0.8479963541030884:  55%|█████▍    | 3554/6470 [40:49<34:52,  1.39it/s]
4|3555|Loss: 0.9194208383560181:  55%|█████▍    | 3554/6470 [40:49<34:52,  1.39it/s]
4|3555|Loss: 0.9194208383560181:  55%|█████▍    | 3555/6470 [40:49<34:57,  1.39it/s]
4|3556|Loss: 0.8084192276000977:  55%|█████▍    | 3555/6470 [40:50<34:57,  1.39it/s]
4|3556|Loss: 0.8084192276000977:  55%|█████▍    | 3556/6470 [40:50<36:17,  1.34it/s]
4|3557|Loss: 0.9528322815895081:  55%|█████▍    | 3556/6470 [40:51<36:17,  1.34it/s]
4|3557|Loss: 0.9528322815895081:  55%|█████▍    | 3557/6470 [40:51<35:17,  1.38it/s]
4|3558|Loss: 0.8166235089302063:  55%|█████▍    | 3557/6470 [40:51<35:17,  1.38it/s]
4|3558|Loss: 0.8166235089302063:  55%|█████▍    | 3558/6470 [40:51<33:13,  1.46it/s]
4|3559|Loss: 0.9390171766281128:  55%|█████▍    | 3558/6470 [40:52<33:13,  1.46it/s]
4|3559|Loss: 0.9390171766281128:  55%|█████▌    | 3559/6470 [40:52<31:26,  1.54it/s]
4|3560|Loss: 0.8318540453910828:  55%|█████▌    | 3559/6470 [40:52<31:26,  1.54it/s]
4|3560|Loss: 0.8318540453910828:  55%|█████▌    | 3560/6470 [40:53<32:20,  1.50it/s]
4|3561|Loss: 0.8189377784729004:  55%|█████▌    | 3560/6470 [40:53<32:20,  1.50it/s]
4|3561|Loss: 0.8189377784729004:  55%|█████▌    | 3561/6470 [40:53<31:16,  1.55it/s]
4|3562|Loss: 0.9088383913040161:  55%|█████▌    | 3561/6470 [40:54<31:16,  1.55it/s]
4|3562|Loss: 0.9088383913040161:  55%|█████▌    | 3562/6470 [40:54<30:40,  1.58it/s]
4|3563|Loss: 0.889196515083313:  55%|█████▌    | 3562/6470 [40:54<30:40,  1.58it/s] 
4|3563|Loss: 0.889196515083313:  55%|█████▌    | 3563/6470 [40:55<32:33,  1.49it/s]
4|3564|Loss: 0.8711661100387573:  55%|█████▌    | 3563/6470 [40:55<32:33,  1.49it/s]
4|3564|Loss: 0.8711661100387573:  55%|█████▌    | 3564/6470 [40:55<33:19,  1.45it/s]
4|3565|Loss: 0.9198086261749268:  55%|█████▌    | 3564/6470 [40:56<33:19,  1.45it/s]
4|3565|Loss: 0.9198086261749268:  55%|█████▌    | 3565/6470 [40:56<32:48,  1.48it/s]
4|3566|Loss: 0.836773693561554:  55%|█████▌    | 3565/6470 [40:56<32:48,  1.48it/s] 
4|3566|Loss: 0.836773693561554:  55%|█████▌    | 3566/6470 [40:56<29:01,  1.67it/s]
4|3567|Loss: 0.8153600692749023:  55%|█████▌    | 3566/6470 [40:57<29:01,  1.67it/s]
4|3567|Loss: 0.8153600692749023:  55%|█████▌    | 3567/6470 [40:57<28:37,  1.69it/s]
4|3568|Loss: 0.8718982934951782:  55%|█████▌    | 3567/6470 [40:57<28:37,  1.69it/s]
4|3568|Loss: 0.8718982934951782:  55%|█████▌    | 3568/6470 [40:57<27:04,  1.79it/s]
4|3569|Loss: 0.8756692409515381:  55%|█████▌    | 3568/6470 [40:58<27:04,  1.79it/s]
4|3569|Loss: 0.8756692409515381:  55%|█████▌    | 3569/6470 [40:58<29:47,  1.62it/s]
4|3570|Loss: 1.0701797008514404:  55%|█████▌    | 3569/6470 [40:59<29:47,  1.62it/s]
4|3570|Loss: 1.0701797008514404:  55%|█████▌    | 3570/6470 [40:59<30:23,  1.59it/s]
4|3571|Loss: 0.7697128057479858:  55%|█████▌    | 3570/6470 [40:59<30:23,  1.59it/s]
4|3571|Loss: 0.7697128057479858:  55%|█████▌    | 3571/6470 [41:00<33:35,  1.44it/s]
4|3572|Loss: 0.8755505084991455:  55%|█████▌    | 3571/6470 [41:00<33:35,  1.44it/s]
4|3572|Loss: 0.8755505084991455:  55%|█████▌    | 3572/6470 [41:00<33:37,  1.44it/s]
4|3573|Loss: 0.9686654210090637:  55%|█████▌    | 3572/6470 [41:01<33:37,  1.44it/s]
4|3573|Loss: 0.9686654210090637:  55%|█████▌    | 3573/6470 [41:01<32:49,  1.47it/s]
4|3574|Loss: 0.8714965581893921:  55%|█████▌    | 3573/6470 [41:01<32:49,  1.47it/s]
4|3574|Loss: 0.8714965581893921:  55%|█████▌    | 3574/6470 [41:02<32:15,  1.50it/s]
4|3575|Loss: 0.595045268535614:  55%|█████▌    | 3574/6470 [41:02<32:15,  1.50it/s] 
4|3575|Loss: 0.595045268535614:  55%|█████▌    | 3575/6470 [41:02<32:30,  1.48it/s]
4|3576|Loss: 0.9096358418464661:  55%|█████▌    | 3575/6470 [41:03<32:30,  1.48it/s]
4|3576|Loss: 0.9096358418464661:  55%|█████▌    | 3576/6470 [41:03<31:53,  1.51it/s]
4|3577|Loss: 0.9061631560325623:  55%|█████▌    | 3576/6470 [41:03<31:53,  1.51it/s]
4|3577|Loss: 0.9061631560325623:  55%|█████▌    | 3577/6470 [41:04<30:11,  1.60it/s]
4|3578|Loss: 0.8514615297317505:  55%|█████▌    | 3577/6470 [41:04<30:11,  1.60it/s]
4|3578|Loss: 0.8514615297317505:  55%|█████▌    | 3578/6470 [41:04<31:05,  1.55it/s]
4|3579|Loss: 0.8520486950874329:  55%|█████▌    | 3578/6470 [41:05<31:05,  1.55it/s]
4|3579|Loss: 0.8520486950874329:  55%|█████▌    | 3579/6470 [41:05<33:45,  1.43it/s]
4|3580|Loss: 0.9029906392097473:  55%|█████▌    | 3579/6470 [41:06<33:45,  1.43it/s]
4|3580|Loss: 0.9029906392097473:  55%|█████▌    | 3580/6470 [41:06<34:40,  1.39it/s]
4|3581|Loss: 0.772458553314209:  55%|█████▌    | 3580/6470 [41:06<34:40,  1.39it/s] 
4|3581|Loss: 0.772458553314209:  55%|█████▌    | 3581/6470 [41:07<35:43,  1.35it/s]
4|3582|Loss: 0.7825564742088318:  55%|█████▌    | 3581/6470 [41:07<35:43,  1.35it/s]
4|3582|Loss: 0.7825564742088318:  55%|█████▌    | 3582/6470 [41:07<35:46,  1.35it/s]
4|3583|Loss: 0.8417935371398926:  55%|█████▌    | 3582/6470 [41:08<35:46,  1.35it/s]
4|3583|Loss: 0.8417935371398926:  55%|█████▌    | 3583/6470 [41:08<35:33,  1.35it/s]
4|3584|Loss: 0.8512863516807556:  55%|█████▌    | 3583/6470 [41:09<35:33,  1.35it/s]
4|3584|Loss: 0.8512863516807556:  55%|█████▌    | 3584/6470 [41:09<34:18,  1.40it/s]
4|3585|Loss: 1.0077650547027588:  55%|█████▌    | 3584/6470 [41:09<34:18,  1.40it/s]
4|3585|Loss: 1.0077650547027588:  55%|█████▌    | 3585/6470 [41:09<33:11,  1.45it/s]
4|3586|Loss: 0.7731304168701172:  55%|█████▌    | 3585/6470 [41:10<33:11,  1.45it/s]
4|3586|Loss: 0.7731304168701172:  55%|█████▌    | 3586/6470 [41:10<31:49,  1.51it/s]
4|3587|Loss: 0.8296316266059875:  55%|█████▌    | 3586/6470 [41:10<31:49,  1.51it/s]
4|3587|Loss: 0.8296316266059875:  55%|█████▌    | 3587/6470 [41:11<33:11,  1.45it/s]
4|3588|Loss: 0.7780246734619141:  55%|█████▌    | 3587/6470 [41:11<33:11,  1.45it/s]
4|3588|Loss: 0.7780246734619141:  55%|█████▌    | 3588/6470 [41:11<32:10,  1.49it/s]
4|3589|Loss: 0.7904701232910156:  55%|█████▌    | 3588/6470 [41:12<32:10,  1.49it/s]
4|3589|Loss: 0.7904701232910156:  55%|█████▌    | 3589/6470 [41:12<32:34,  1.47it/s]
4|3590|Loss: 1.0235519409179688:  55%|█████▌    | 3589/6470 [41:13<32:34,  1.47it/s]
4|3590|Loss: 1.0235519409179688:  55%|█████▌    | 3590/6470 [41:13<34:20,  1.40it/s]
4|3591|Loss: 0.9138000011444092:  55%|█████▌    | 3590/6470 [41:13<34:20,  1.40it/s]
4|3591|Loss: 0.9138000011444092:  56%|█████▌    | 3591/6470 [41:14<33:49,  1.42it/s]
4|3592|Loss: 0.7848579287528992:  56%|█████▌    | 3591/6470 [41:14<33:49,  1.42it/s]
4|3592|Loss: 0.7848579287528992:  56%|█████▌    | 3592/6470 [41:14<34:47,  1.38it/s]
4|3593|Loss: 0.8237571716308594:  56%|█████▌    | 3592/6470 [41:15<34:47,  1.38it/s]
4|3593|Loss: 0.8237571716308594:  56%|█████▌    | 3593/6470 [41:15<34:04,  1.41it/s]
4|3594|Loss: 0.8642352819442749:  56%|█████▌    | 3593/6470 [41:15<34:04,  1.41it/s]
4|3594|Loss: 0.8642352819442749:  56%|█████▌    | 3594/6470 [41:16<33:45,  1.42it/s]
4|3595|Loss: 0.7600821256637573:  56%|█████▌    | 3594/6470 [41:16<33:45,  1.42it/s]
4|3595|Loss: 0.7600821256637573:  56%|█████▌    | 3595/6470 [41:16<33:57,  1.41it/s]
4|3596|Loss: 0.8481242656707764:  56%|█████▌    | 3595/6470 [41:17<33:57,  1.41it/s]
4|3596|Loss: 0.8481242656707764:  56%|█████▌    | 3596/6470 [41:17<35:43,  1.34it/s]
4|3597|Loss: 0.8825022578239441:  56%|█████▌    | 3596/6470 [41:18<35:43,  1.34it/s]
4|3597|Loss: 0.8825022578239441:  56%|█████▌    | 3597/6470 [41:18<33:28,  1.43it/s]
4|3598|Loss: 0.8706780076026917:  56%|█████▌    | 3597/6470 [41:18<33:28,  1.43it/s]
4|3598|Loss: 0.8706780076026917:  56%|█████▌    | 3598/6470 [41:18<31:31,  1.52it/s]
4|3599|Loss: 0.8769126534461975:  56%|█████▌    | 3598/6470 [41:19<31:31,  1.52it/s]
4|3599|Loss: 0.8769126534461975:  56%|█████▌    | 3599/6470 [41:19<28:56,  1.65it/s]
4|3600|Loss: 0.9213070869445801:  56%|█████▌    | 3599/6470 [41:19<28:56,  1.65it/s]
4|3600|Loss: 0.9213070869445801:  56%|█████▌    | 3600/6470 [41:20<30:27,  1.57it/s]
4|3601|Loss: 0.7437219619750977:  56%|█████▌    | 3600/6470 [41:20<30:27,  1.57it/s]
4|3601|Loss: 0.7437219619750977:  56%|█████▌    | 3601/6470 [41:20<32:02,  1.49it/s]
4|3602|Loss: 0.886860728263855:  56%|█████▌    | 3601/6470 [41:21<32:02,  1.49it/s] 
4|3602|Loss: 0.886860728263855:  56%|█████▌    | 3602/6470 [41:21<32:51,  1.45it/s]
4|3603|Loss: 0.8253402709960938:  56%|█████▌    | 3602/6470 [41:22<32:51,  1.45it/s]
4|3603|Loss: 0.8253402709960938:  56%|█████▌    | 3603/6470 [41:22<33:24,  1.43it/s]
4|3604|Loss: 0.7825141549110413:  56%|█████▌    | 3603/6470 [41:22<33:24,  1.43it/s]
4|3604|Loss: 0.7825141549110413:  56%|█████▌    | 3604/6470 [41:23<35:58,  1.33it/s]
4|3605|Loss: 0.773938775062561:  56%|█████▌    | 3604/6470 [41:23<35:58,  1.33it/s] 
4|3605|Loss: 0.773938775062561:  56%|█████▌    | 3605/6470 [41:24<37:58,  1.26it/s]
4|3606|Loss: 0.8583020567893982:  56%|█████▌    | 3605/6470 [41:24<37:58,  1.26it/s]
4|3606|Loss: 0.8583020567893982:  56%|█████▌    | 3606/6470 [41:24<37:42,  1.27it/s]
4|3607|Loss: 0.9242621064186096:  56%|█████▌    | 3606/6470 [41:25<37:42,  1.27it/s]
4|3607|Loss: 0.9242621064186096:  56%|█████▌    | 3607/6470 [41:25<34:24,  1.39it/s]
4|3608|Loss: 0.9018869996070862:  56%|█████▌    | 3607/6470 [41:25<34:24,  1.39it/s]
4|3608|Loss: 0.9018869996070862:  56%|█████▌    | 3608/6470 [41:26<34:06,  1.40it/s]
4|3609|Loss: 0.9134706854820251:  56%|█████▌    | 3608/6470 [41:26<34:06,  1.40it/s]
4|3609|Loss: 0.9134706854820251:  56%|█████▌    | 3609/6470 [41:26<31:37,  1.51it/s]
4|3610|Loss: 0.8997606039047241:  56%|█████▌    | 3609/6470 [41:27<31:37,  1.51it/s]
4|3610|Loss: 0.8997606039047241:  56%|█████▌    | 3610/6470 [41:27<32:29,  1.47it/s]
4|3611|Loss: 0.8495148420333862:  56%|█████▌    | 3610/6470 [41:27<32:29,  1.47it/s]
4|3611|Loss: 0.8495148420333862:  56%|█████▌    | 3611/6470 [41:28<34:58,  1.36it/s]
4|3612|Loss: 0.8478195071220398:  56%|█████▌    | 3611/6470 [41:28<34:58,  1.36it/s]
4|3612|Loss: 0.8478195071220398:  56%|█████▌    | 3612/6470 [41:28<34:06,  1.40it/s]
4|3613|Loss: 0.7625243663787842:  56%|█████▌    | 3612/6470 [41:29<34:06,  1.40it/s]
4|3613|Loss: 0.7625243663787842:  56%|█████▌    | 3613/6470 [41:29<35:20,  1.35it/s]
4|3614|Loss: 0.8510451316833496:  56%|█████▌    | 3613/6470 [41:30<35:20,  1.35it/s]
4|3614|Loss: 0.8510451316833496:  56%|█████▌    | 3614/6470 [41:30<34:37,  1.37it/s]
4|3615|Loss: 0.9042329788208008:  56%|█████▌    | 3614/6470 [41:30<34:37,  1.37it/s]
4|3615|Loss: 0.9042329788208008:  56%|█████▌    | 3615/6470 [41:31<33:24,  1.42it/s]
4|3616|Loss: 0.8548603057861328:  56%|█████▌    | 3615/6470 [41:31<33:24,  1.42it/s]
4|3616|Loss: 0.8548603057861328:  56%|█████▌    | 3616/6470 [41:31<33:16,  1.43it/s]
4|3617|Loss: 0.8961483240127563:  56%|█████▌    | 3616/6470 [41:32<33:16,  1.43it/s]
4|3617|Loss: 0.8961483240127563:  56%|█████▌    | 3617/6470 [41:32<32:34,  1.46it/s]
4|3618|Loss: 0.7922411561012268:  56%|█████▌    | 3617/6470 [41:32<32:34,  1.46it/s]
4|3618|Loss: 0.7922411561012268:  56%|█████▌    | 3618/6470 [41:33<34:24,  1.38it/s]
4|3619|Loss: 0.8143090605735779:  56%|█████▌    | 3618/6470 [41:33<34:24,  1.38it/s]
4|3619|Loss: 0.8143090605735779:  56%|█████▌    | 3619/6470 [41:33<33:38,  1.41it/s]
4|3620|Loss: 0.7728475332260132:  56%|█████▌    | 3619/6470 [41:34<33:38,  1.41it/s]
4|3620|Loss: 0.7728475332260132:  56%|█████▌    | 3620/6470 [41:34<35:22,  1.34it/s]
4|3621|Loss: 0.7019244432449341:  56%|█████▌    | 3620/6470 [41:35<35:22,  1.34it/s]
4|3621|Loss: 0.7019244432449341:  56%|█████▌    | 3621/6470 [41:35<37:47,  1.26it/s]
4|3622|Loss: 0.7861546277999878:  56%|█████▌    | 3621/6470 [41:36<37:47,  1.26it/s]
4|3622|Loss: 0.7861546277999878:  56%|█████▌    | 3622/6470 [41:36<38:27,  1.23it/s]
4|3623|Loss: 0.7053247094154358:  56%|█████▌    | 3622/6470 [41:36<38:27,  1.23it/s]
4|3623|Loss: 0.7053247094154358:  56%|█████▌    | 3623/6470 [41:37<39:01,  1.22it/s]
4|3624|Loss: 0.8498436808586121:  56%|█████▌    | 3623/6470 [41:37<39:01,  1.22it/s]
4|3624|Loss: 0.8498436808586121:  56%|█████▌    | 3624/6470 [41:38<38:18,  1.24it/s]
4|3625|Loss: 0.7376804947853088:  56%|█████▌    | 3624/6470 [41:38<38:18,  1.24it/s]
4|3625|Loss: 0.7376804947853088:  56%|█████▌    | 3625/6470 [41:39<39:22,  1.20it/s]
4|3626|Loss: 1.005092740058899:  56%|█████▌    | 3625/6470 [41:39<39:22,  1.20it/s] 
4|3626|Loss: 1.005092740058899:  56%|█████▌    | 3626/6470 [41:39<38:35,  1.23it/s]
4|3627|Loss: 0.8188207745552063:  56%|█████▌    | 3626/6470 [41:40<38:35,  1.23it/s]
4|3627|Loss: 0.8188207745552063:  56%|█████▌    | 3627/6470 [41:40<36:13,  1.31it/s]
4|3628|Loss: 0.7946205139160156:  56%|█████▌    | 3627/6470 [41:40<36:13,  1.31it/s]
4|3628|Loss: 0.7946205139160156:  56%|█████▌    | 3628/6470 [41:41<35:29,  1.33it/s]
4|3629|Loss: 0.9269483685493469:  56%|█████▌    | 3628/6470 [41:41<35:29,  1.33it/s]
4|3629|Loss: 0.9269483685493469:  56%|█████▌    | 3629/6470 [41:41<34:14,  1.38it/s]
4|3630|Loss: 0.7262157201766968:  56%|█████▌    | 3629/6470 [41:42<34:14,  1.38it/s]
4|3630|Loss: 0.7262157201766968:  56%|█████▌    | 3630/6470 [41:42<33:06,  1.43it/s]
4|3631|Loss: 0.8492720127105713:  56%|█████▌    | 3630/6470 [41:42<33:06,  1.43it/s]
4|3631|Loss: 0.8492720127105713:  56%|█████▌    | 3631/6470 [41:43<31:59,  1.48it/s]
4|3632|Loss: 0.8465943932533264:  56%|█████▌    | 3631/6470 [41:43<31:59,  1.48it/s]
4|3632|Loss: 0.8465943932533264:  56%|█████▌    | 3632/6470 [41:43<33:47,  1.40it/s]
4|3633|Loss: 0.7685363292694092:  56%|█████▌    | 3632/6470 [41:44<33:47,  1.40it/s]
4|3633|Loss: 0.7685363292694092:  56%|█████▌    | 3633/6470 [41:44<33:25,  1.41it/s]
4|3634|Loss: 0.7798237800598145:  56%|█████▌    | 3633/6470 [41:44<33:25,  1.41it/s]
4|3634|Loss: 0.7798237800598145:  56%|█████▌    | 3634/6470 [41:45<32:29,  1.45it/s]
4|3635|Loss: 1.016860008239746:  56%|█████▌    | 3634/6470 [41:45<32:29,  1.45it/s] 
4|3635|Loss: 1.016860008239746:  56%|█████▌    | 3635/6470 [41:46<34:03,  1.39it/s]
4|3636|Loss: 0.9432926774024963:  56%|█████▌    | 3635/6470 [41:46<34:03,  1.39it/s]
4|3636|Loss: 0.9432926774024963:  56%|█████▌    | 3636/6470 [41:46<32:14,  1.46it/s]
4|3637|Loss: 0.8746628165245056:  56%|█████▌    | 3636/6470 [41:46<32:14,  1.46it/s]
4|3637|Loss: 0.8746628165245056:  56%|█████▌    | 3637/6470 [41:47<32:30,  1.45it/s]
4|3638|Loss: 0.8433881998062134:  56%|█████▌    | 3637/6470 [41:47<32:30,  1.45it/s]
4|3638|Loss: 0.8433881998062134:  56%|█████▌    | 3638/6470 [41:48<33:27,  1.41it/s]
4|3639|Loss: 0.7416402697563171:  56%|█████▌    | 3638/6470 [41:48<33:27,  1.41it/s]
4|3639|Loss: 0.7416402697563171:  56%|█████▌    | 3639/6470 [41:48<32:14,  1.46it/s]
4|3640|Loss: 0.7121410965919495:  56%|█████▌    | 3639/6470 [41:48<32:14,  1.46it/s]
4|3640|Loss: 0.7121410965919495:  56%|█████▋    | 3640/6470 [41:49<28:32,  1.65it/s]
4|3641|Loss: 0.7799116373062134:  56%|█████▋    | 3640/6470 [41:49<28:32,  1.65it/s]
4|3641|Loss: 0.7799116373062134:  56%|█████▋    | 3641/6470 [41:49<29:09,  1.62it/s]
4|3642|Loss: 0.8376131653785706:  56%|█████▋    | 3641/6470 [41:50<29:09,  1.62it/s]
4|3642|Loss: 0.8376131653785706:  56%|█████▋    | 3642/6470 [41:50<32:22,  1.46it/s]
4|3643|Loss: 0.7028353810310364:  56%|█████▋    | 3642/6470 [41:50<32:22,  1.46it/s]
4|3643|Loss: 0.7028353810310364:  56%|█████▋    | 3643/6470 [41:51<31:26,  1.50it/s]
4|3644|Loss: 0.8704059720039368:  56%|█████▋    | 3643/6470 [41:51<31:26,  1.50it/s]
4|3644|Loss: 0.8704059720039368:  56%|█████▋    | 3644/6470 [41:51<29:20,  1.61it/s]
4|3645|Loss: 0.8671706914901733:  56%|█████▋    | 3644/6470 [41:52<29:20,  1.61it/s]
4|3645|Loss: 0.8671706914901733:  56%|█████▋    | 3645/6470 [41:52<29:06,  1.62it/s]
4|3646|Loss: 0.799089252948761:  56%|█████▋    | 3645/6470 [41:52<29:06,  1.62it/s] 
4|3646|Loss: 0.799089252948761:  56%|█████▋    | 3646/6470 [41:53<32:01,  1.47it/s]
4|3647|Loss: 0.6906313300132751:  56%|█████▋    | 3646/6470 [41:53<32:01,  1.47it/s]
4|3647|Loss: 0.6906313300132751:  56%|█████▋    | 3647/6470 [41:53<33:55,  1.39it/s]
4|3648|Loss: 0.8468391299247742:  56%|█████▋    | 3647/6470 [41:54<33:55,  1.39it/s]
4|3648|Loss: 0.8468391299247742:  56%|█████▋    | 3648/6470 [41:54<34:00,  1.38it/s]
4|3649|Loss: 0.910459578037262:  56%|█████▋    | 3648/6470 [41:55<34:00,  1.38it/s] 
4|3649|Loss: 0.910459578037262:  56%|█████▋    | 3649/6470 [41:55<33:45,  1.39it/s]
4|3650|Loss: 0.7144112586975098:  56%|█████▋    | 3649/6470 [41:55<33:45,  1.39it/s]
4|3650|Loss: 0.7144112586975098:  56%|█████▋    | 3650/6470 [41:55<30:41,  1.53it/s]
4|3651|Loss: 0.9083315134048462:  56%|█████▋    | 3650/6470 [41:56<30:41,  1.53it/s]
4|3651|Loss: 0.9083315134048462:  56%|█████▋    | 3651/6470 [41:56<32:03,  1.47it/s]
4|3652|Loss: 0.9250433444976807:  56%|█████▋    | 3651/6470 [41:57<32:03,  1.47it/s]
4|3652|Loss: 0.9250433444976807:  56%|█████▋    | 3652/6470 [41:57<31:45,  1.48it/s]
4|3653|Loss: 1.1042338609695435:  56%|█████▋    | 3652/6470 [41:57<31:45,  1.48it/s]
4|3653|Loss: 1.1042338609695435:  56%|█████▋    | 3653/6470 [41:57<30:52,  1.52it/s]
4|3654|Loss: 0.9125043749809265:  56%|█████▋    | 3653/6470 [41:58<30:52,  1.52it/s]
4|3654|Loss: 0.9125043749809265:  56%|█████▋    | 3654/6470 [41:58<29:32,  1.59it/s]
4|3655|Loss: 0.8669745922088623:  56%|█████▋    | 3654/6470 [41:58<29:32,  1.59it/s]
4|3655|Loss: 0.8669745922088623:  56%|█████▋    | 3655/6470 [41:59<30:07,  1.56it/s]
4|3656|Loss: 0.7407629489898682:  56%|█████▋    | 3655/6470 [41:59<30:07,  1.56it/s]
4|3656|Loss: 0.7407629489898682:  57%|█████▋    | 3656/6470 [41:59<30:26,  1.54it/s]
4|3657|Loss: 0.8862390518188477:  57%|█████▋    | 3656/6470 [42:00<30:26,  1.54it/s]
4|3657|Loss: 0.8862390518188477:  57%|█████▋    | 3657/6470 [42:00<30:04,  1.56it/s]
4|3658|Loss: 0.7455252408981323:  57%|█████▋    | 3657/6470 [42:00<30:04,  1.56it/s]
4|3658|Loss: 0.7455252408981323:  57%|█████▋    | 3658/6470 [42:01<31:11,  1.50it/s]
4|3659|Loss: 0.8849384784698486:  57%|█████▋    | 3658/6470 [42:01<31:11,  1.50it/s]
4|3659|Loss: 0.8849384784698486:  57%|█████▋    | 3659/6470 [42:01<29:40,  1.58it/s]
4|3660|Loss: 0.6731756329536438:  57%|█████▋    | 3659/6470 [42:02<29:40,  1.58it/s]
4|3660|Loss: 0.6731756329536438:  57%|█████▋    | 3660/6470 [42:02<33:32,  1.40it/s]
4|3661|Loss: 0.8717361092567444:  57%|█████▋    | 3660/6470 [42:03<33:32,  1.40it/s]
4|3661|Loss: 0.8717361092567444:  57%|█████▋    | 3661/6470 [42:03<33:25,  1.40it/s]
4|3662|Loss: 0.738893985748291:  57%|█████▋    | 3661/6470 [42:03<33:25,  1.40it/s] 
4|3662|Loss: 0.738893985748291:  57%|█████▋    | 3662/6470 [42:03<30:08,  1.55it/s]
4|3663|Loss: 0.7839496731758118:  57%|█████▋    | 3662/6470 [42:04<30:08,  1.55it/s]
4|3663|Loss: 0.7839496731758118:  57%|█████▋    | 3663/6470 [42:04<30:42,  1.52it/s]
4|3664|Loss: 0.8812026977539062:  57%|█████▋    | 3663/6470 [42:04<30:42,  1.52it/s]
4|3664|Loss: 0.8812026977539062:  57%|█████▋    | 3664/6470 [42:05<31:35,  1.48it/s]
4|3665|Loss: 0.8910933136940002:  57%|█████▋    | 3664/6470 [42:05<31:35,  1.48it/s]
4|3665|Loss: 0.8910933136940002:  57%|█████▋    | 3665/6470 [42:05<30:57,  1.51it/s]
4|3666|Loss: 0.8787503242492676:  57%|█████▋    | 3665/6470 [42:06<30:57,  1.51it/s]
4|3666|Loss: 0.8787503242492676:  57%|█████▋    | 3666/6470 [42:06<32:13,  1.45it/s]
4|3667|Loss: 0.8522810339927673:  57%|█████▋    | 3666/6470 [42:07<32:13,  1.45it/s]
4|3667|Loss: 0.8522810339927673:  57%|█████▋    | 3667/6470 [42:07<31:47,  1.47it/s]
4|3668|Loss: 0.8405801653862:  57%|█████▋    | 3667/6470 [42:07<31:47,  1.47it/s]   
4|3668|Loss: 0.8405801653862:  57%|█████▋    | 3668/6470 [42:07<31:03,  1.50it/s]
4|3669|Loss: 0.9210814833641052:  57%|█████▋    | 3668/6470 [42:08<31:03,  1.50it/s]
4|3669|Loss: 0.9210814833641052:  57%|█████▋    | 3669/6470 [42:08<31:41,  1.47it/s]
4|3670|Loss: 0.8623163104057312:  57%|█████▋    | 3669/6470 [42:09<31:41,  1.47it/s]
4|3670|Loss: 0.8623163104057312:  57%|█████▋    | 3670/6470 [42:09<32:25,  1.44it/s]
4|3671|Loss: 0.8692792057991028:  57%|█████▋    | 3670/6470 [42:09<32:25,  1.44it/s]
4|3671|Loss: 0.8692792057991028:  57%|█████▋    | 3671/6470 [42:10<32:56,  1.42it/s]
4|3672|Loss: 1.0058083534240723:  57%|█████▋    | 3671/6470 [42:10<32:56,  1.42it/s]
4|3672|Loss: 1.0058083534240723:  57%|█████▋    | 3672/6470 [42:10<33:28,  1.39it/s]
4|3673|Loss: 0.8303094506263733:  57%|█████▋    | 3672/6470 [42:11<33:28,  1.39it/s]
4|3673|Loss: 0.8303094506263733:  57%|█████▋    | 3673/6470 [42:11<35:10,  1.33it/s]
4|3674|Loss: 0.7981029152870178:  57%|█████▋    | 3673/6470 [42:12<35:10,  1.33it/s]
4|3674|Loss: 0.7981029152870178:  57%|█████▋    | 3674/6470 [42:12<36:33,  1.27it/s]
4|3675|Loss: 0.841729462146759:  57%|█████▋    | 3674/6470 [42:12<36:33,  1.27it/s] 
4|3675|Loss: 0.841729462146759:  57%|█████▋    | 3675/6470 [42:13<35:49,  1.30it/s]
4|3676|Loss: 0.7965283989906311:  57%|█████▋    | 3675/6470 [42:13<35:49,  1.30it/s]
4|3676|Loss: 0.7965283989906311:  57%|█████▋    | 3676/6470 [42:13<33:21,  1.40it/s]
4|3677|Loss: 0.9561305046081543:  57%|█████▋    | 3676/6470 [42:14<33:21,  1.40it/s]
4|3677|Loss: 0.9561305046081543:  57%|█████▋    | 3677/6470 [42:14<30:52,  1.51it/s]
4|3678|Loss: 0.9946858882904053:  57%|█████▋    | 3677/6470 [42:14<30:52,  1.51it/s]
4|3678|Loss: 0.9946858882904053:  57%|█████▋    | 3678/6470 [42:15<29:58,  1.55it/s]
4|3679|Loss: 0.7371154427528381:  57%|█████▋    | 3678/6470 [42:15<29:58,  1.55it/s]
4|3679|Loss: 0.7371154427528381:  57%|█████▋    | 3679/6470 [42:15<29:49,  1.56it/s]
4|3680|Loss: 1.0643795728683472:  57%|█████▋    | 3679/6470 [42:16<29:49,  1.56it/s]
4|3680|Loss: 1.0643795728683472:  57%|█████▋    | 3680/6470 [42:16<32:33,  1.43it/s]
4|3681|Loss: 0.8313150405883789:  57%|█████▋    | 3680/6470 [42:16<32:33,  1.43it/s]
4|3681|Loss: 0.8313150405883789:  57%|█████▋    | 3681/6470 [42:17<31:50,  1.46it/s]
4|3682|Loss: 0.7682956457138062:  57%|█████▋    | 3681/6470 [42:17<31:50,  1.46it/s]
4|3682|Loss: 0.7682956457138062:  57%|█████▋    | 3682/6470 [42:17<32:03,  1.45it/s]
4|3683|Loss: 1.000616431236267:  57%|█████▋    | 3682/6470 [42:18<32:03,  1.45it/s] 
4|3683|Loss: 1.000616431236267:  57%|█████▋    | 3683/6470 [42:18<34:17,  1.35it/s]
4|3684|Loss: 0.8524717688560486:  57%|█████▋    | 3683/6470 [42:19<34:17,  1.35it/s]
4|3684|Loss: 0.8524717688560486:  57%|█████▋    | 3684/6470 [42:19<34:29,  1.35it/s]
4|3685|Loss: 0.8700491189956665:  57%|█████▋    | 3684/6470 [42:19<34:29,  1.35it/s]
4|3685|Loss: 0.8700491189956665:  57%|█████▋    | 3685/6470 [42:20<36:45,  1.26it/s]
4|3686|Loss: 0.9425395131111145:  57%|█████▋    | 3685/6470 [42:20<36:45,  1.26it/s]
4|3686|Loss: 0.9425395131111145:  57%|█████▋    | 3686/6470 [42:20<34:18,  1.35it/s]
4|3687|Loss: 1.0483620166778564:  57%|█████▋    | 3686/6470 [42:21<34:18,  1.35it/s]
4|3687|Loss: 1.0483620166778564:  57%|█████▋    | 3687/6470 [42:21<33:52,  1.37it/s]
4|3688|Loss: 0.7214339375495911:  57%|█████▋    | 3687/6470 [42:22<33:52,  1.37it/s]
4|3688|Loss: 0.7214339375495911:  57%|█████▋    | 3688/6470 [42:22<35:33,  1.30it/s]
4|3689|Loss: 0.8621068596839905:  57%|█████▋    | 3688/6470 [42:22<35:33,  1.30it/s]
4|3689|Loss: 0.8621068596839905:  57%|█████▋    | 3689/6470 [42:23<35:12,  1.32it/s]
4|3690|Loss: 0.974840521812439:  57%|█████▋    | 3689/6470 [42:23<35:12,  1.32it/s] 
4|3690|Loss: 0.974840521812439:  57%|█████▋    | 3690/6470 [42:24<36:10,  1.28it/s]
4|3691|Loss: 0.7777655720710754:  57%|█████▋    | 3690/6470 [42:24<36:10,  1.28it/s]
4|3691|Loss: 0.7777655720710754:  57%|█████▋    | 3691/6470 [42:24<34:55,  1.33it/s]
4|3692|Loss: 0.9675838351249695:  57%|█████▋    | 3691/6470 [42:25<34:55,  1.33it/s]
4|3692|Loss: 0.9675838351249695:  57%|█████▋    | 3692/6470 [42:25<35:28,  1.31it/s]
4|3693|Loss: 1.006062388420105:  57%|█████▋    | 3692/6470 [42:25<35:28,  1.31it/s] 
4|3693|Loss: 1.006062388420105:  57%|█████▋    | 3693/6470 [42:26<34:18,  1.35it/s]
4|3694|Loss: 0.9932117462158203:  57%|█████▋    | 3693/6470 [42:26<34:18,  1.35it/s]
4|3694|Loss: 0.9932117462158203:  57%|█████▋    | 3694/6470 [42:27<35:18,  1.31it/s]
4|3695|Loss: 0.796998918056488:  57%|█████▋    | 3694/6470 [42:27<35:18,  1.31it/s] 
4|3695|Loss: 0.796998918056488:  57%|█████▋    | 3695/6470 [42:27<33:13,  1.39it/s]
4|3696|Loss: 0.972528874874115:  57%|█████▋    | 3695/6470 [42:28<33:13,  1.39it/s]
4|3696|Loss: 0.972528874874115:  57%|█████▋    | 3696/6470 [42:28<34:17,  1.35it/s]
4|3697|Loss: 0.9066139459609985:  57%|█████▋    | 3696/6470 [42:28<34:17,  1.35it/s]
4|3697|Loss: 0.9066139459609985:  57%|█████▋    | 3697/6470 [42:29<32:37,  1.42it/s]
4|3698|Loss: 0.9103774428367615:  57%|█████▋    | 3697/6470 [42:29<32:37,  1.42it/s]
4|3698|Loss: 0.9103774428367615:  57%|█████▋    | 3698/6470 [42:29<33:13,  1.39it/s]
4|3699|Loss: 1.1510635614395142:  57%|█████▋    | 3698/6470 [42:30<33:13,  1.39it/s]
4|3699|Loss: 1.1510635614395142:  57%|█████▋    | 3699/6470 [42:30<34:40,  1.33it/s]
4|3700|Loss: 0.8092687129974365:  57%|█████▋    | 3699/6470 [42:31<34:40,  1.33it/s]
4|3700|Loss: 0.8092687129974365:  57%|█████▋    | 3700/6470 [42:31<33:00,  1.40it/s]
4|3701|Loss: 0.9318899512290955:  57%|█████▋    | 3700/6470 [42:31<33:00,  1.40it/s]
4|3701|Loss: 0.9318899512290955:  57%|█████▋    | 3701/6470 [42:32<34:44,  1.33it/s]
4|3702|Loss: 0.8145060539245605:  57%|█████▋    | 3701/6470 [42:32<34:44,  1.33it/s]
4|3702|Loss: 0.8145060539245605:  57%|█████▋    | 3702/6470 [42:32<33:59,  1.36it/s]
4|3703|Loss: 0.8407431244850159:  57%|█████▋    | 3702/6470 [42:33<33:59,  1.36it/s]
4|3703|Loss: 0.8407431244850159:  57%|█████▋    | 3703/6470 [42:33<31:54,  1.45it/s]
4|3704|Loss: 0.9094580411911011:  57%|█████▋    | 3703/6470 [42:33<31:54,  1.45it/s]
4|3704|Loss: 0.9094580411911011:  57%|█████▋    | 3704/6470 [42:34<30:42,  1.50it/s]
4|3705|Loss: 0.8866117596626282:  57%|█████▋    | 3704/6470 [42:34<30:42,  1.50it/s]
4|3705|Loss: 0.8866117596626282:  57%|█████▋    | 3705/6470 [42:34<30:08,  1.53it/s]
4|3706|Loss: 0.7901387214660645:  57%|█████▋    | 3705/6470 [42:35<30:08,  1.53it/s]
4|3706|Loss: 0.7901387214660645:  57%|█████▋    | 3706/6470 [42:35<29:44,  1.55it/s]
4|3707|Loss: 0.7498211860656738:  57%|█████▋    | 3706/6470 [42:35<29:44,  1.55it/s]
4|3707|Loss: 0.7498211860656738:  57%|█████▋    | 3707/6470 [42:36<30:53,  1.49it/s]
4|3708|Loss: 0.827441394329071:  57%|█████▋    | 3707/6470 [42:36<30:53,  1.49it/s] 
4|3708|Loss: 0.827441394329071:  57%|█████▋    | 3708/6470 [42:36<29:49,  1.54it/s]
4|3709|Loss: 0.6916144490242004:  57%|█████▋    | 3708/6470 [42:36<29:49,  1.54it/s]
4|3709|Loss: 0.6916144490242004:  57%|█████▋    | 3709/6470 [42:37<29:26,  1.56it/s]
4|3710|Loss: 0.7880531549453735:  57%|█████▋    | 3709/6470 [42:37<29:26,  1.56it/s]
4|3710|Loss: 0.7880531549453735:  57%|█████▋    | 3710/6470 [42:37<30:31,  1.51it/s]
4|3711|Loss: 0.9215866327285767:  57%|█████▋    | 3710/6470 [42:38<30:31,  1.51it/s]
4|3711|Loss: 0.9215866327285767:  57%|█████▋    | 3711/6470 [42:38<31:31,  1.46it/s]
4|3712|Loss: 0.8840569853782654:  57%|█████▋    | 3711/6470 [42:39<31:31,  1.46it/s]
4|3712|Loss: 0.8840569853782654:  57%|█████▋    | 3712/6470 [42:39<32:11,  1.43it/s]
4|3713|Loss: 0.7916556000709534:  57%|█████▋    | 3712/6470 [42:39<32:11,  1.43it/s]
4|3713|Loss: 0.7916556000709534:  57%|█████▋    | 3713/6470 [42:40<32:38,  1.41it/s]
4|3714|Loss: 0.9715606570243835:  57%|█████▋    | 3713/6470 [42:40<32:38,  1.41it/s]
4|3714|Loss: 0.9715606570243835:  57%|█████▋    | 3714/6470 [42:40<31:39,  1.45it/s]
4|3715|Loss: 0.806067705154419:  57%|█████▋    | 3714/6470 [42:41<31:39,  1.45it/s] 
4|3715|Loss: 0.806067705154419:  57%|█████▋    | 3715/6470 [42:41<30:34,  1.50it/s]
4|3716|Loss: 1.0627821683883667:  57%|█████▋    | 3715/6470 [42:41<30:34,  1.50it/s]
4|3716|Loss: 1.0627821683883667:  57%|█████▋    | 3716/6470 [42:42<32:41,  1.40it/s]
4|3717|Loss: 0.8335879445075989:  57%|█████▋    | 3716/6470 [42:42<32:41,  1.40it/s]
4|3717|Loss: 0.8335879445075989:  57%|█████▋    | 3717/6470 [42:42<32:07,  1.43it/s]
4|3718|Loss: 0.6219912171363831:  57%|█████▋    | 3717/6470 [42:43<32:07,  1.43it/s]
4|3718|Loss: 0.6219912171363831:  57%|█████▋    | 3718/6470 [42:43<32:18,  1.42it/s]
4|3719|Loss: 0.7920423150062561:  57%|█████▋    | 3718/6470 [42:44<32:18,  1.42it/s]
4|3719|Loss: 0.7920423150062561:  57%|█████▋    | 3719/6470 [42:44<31:29,  1.46it/s]
4|3720|Loss: 0.9236470460891724:  57%|█████▋    | 3719/6470 [42:44<31:29,  1.46it/s]
4|3720|Loss: 0.9236470460891724:  57%|█████▋    | 3720/6470 [42:45<32:23,  1.41it/s]
4|3721|Loss: 0.8885452747344971:  57%|█████▋    | 3720/6470 [42:45<32:23,  1.41it/s]
4|3721|Loss: 0.8885452747344971:  58%|█████▊    | 3721/6470 [42:45<34:17,  1.34it/s]
4|3722|Loss: 0.8824482560157776:  58%|█████▊    | 3721/6470 [42:46<34:17,  1.34it/s]
4|3722|Loss: 0.8824482560157776:  58%|█████▊    | 3722/6470 [42:46<32:29,  1.41it/s]
4|3723|Loss: 0.9896669387817383:  58%|█████▊    | 3722/6470 [42:46<32:29,  1.41it/s]
4|3723|Loss: 0.9896669387817383:  58%|█████▊    | 3723/6470 [42:47<31:19,  1.46it/s]
4|3724|Loss: 0.8421213626861572:  58%|█████▊    | 3723/6470 [42:47<31:19,  1.46it/s]
4|3724|Loss: 0.8421213626861572:  58%|█████▊    | 3724/6470 [42:47<32:03,  1.43it/s]
4|3725|Loss: 0.7236597537994385:  58%|█████▊    | 3724/6470 [42:48<32:03,  1.43it/s]
4|3725|Loss: 0.7236597537994385:  58%|█████▊    | 3725/6470 [42:48<30:36,  1.49it/s]
4|3726|Loss: 0.9723801612854004:  58%|█████▊    | 3725/6470 [42:48<30:36,  1.49it/s]
4|3726|Loss: 0.9723801612854004:  58%|█████▊    | 3726/6470 [42:49<29:49,  1.53it/s]
4|3727|Loss: 0.7713074684143066:  58%|█████▊    | 3726/6470 [42:49<29:49,  1.53it/s]
4|3727|Loss: 0.7713074684143066:  58%|█████▊    | 3727/6470 [42:49<28:24,  1.61it/s]
4|3728|Loss: 0.7835971117019653:  58%|█████▊    | 3727/6470 [42:49<28:24,  1.61it/s]
4|3728|Loss: 0.7835971117019653:  58%|█████▊    | 3728/6470 [42:50<27:05,  1.69it/s]
4|3729|Loss: 0.7376130223274231:  58%|█████▊    | 3728/6470 [42:50<27:05,  1.69it/s]
4|3729|Loss: 0.7376130223274231:  58%|█████▊    | 3729/6470 [42:50<29:51,  1.53it/s]
4|3730|Loss: 0.7681059837341309:  58%|█████▊    | 3729/6470 [42:51<29:51,  1.53it/s]
4|3730|Loss: 0.7681059837341309:  58%|█████▊    | 3730/6470 [42:51<30:57,  1.47it/s]
4|3731|Loss: 0.8362907767295837:  58%|█████▊    | 3730/6470 [42:52<30:57,  1.47it/s]
4|3731|Loss: 0.8362907767295837:  58%|█████▊    | 3731/6470 [42:52<32:37,  1.40it/s]
4|3732|Loss: 0.8419870138168335:  58%|█████▊    | 3731/6470 [42:52<32:37,  1.40it/s]
4|3732|Loss: 0.8419870138168335:  58%|█████▊    | 3732/6470 [42:53<31:49,  1.43it/s]
4|3733|Loss: 0.8514582514762878:  58%|█████▊    | 3732/6470 [42:53<31:49,  1.43it/s]
4|3733|Loss: 0.8514582514762878:  58%|█████▊    | 3733/6470 [42:53<29:44,  1.53it/s]
4|3734|Loss: 0.7919553518295288:  58%|█████▊    | 3733/6470 [42:54<29:44,  1.53it/s]
4|3734|Loss: 0.7919553518295288:  58%|█████▊    | 3734/6470 [42:54<29:15,  1.56it/s]
4|3735|Loss: 0.9128385186195374:  58%|█████▊    | 3734/6470 [42:54<29:15,  1.56it/s]
4|3735|Loss: 0.9128385186195374:  58%|█████▊    | 3735/6470 [42:55<30:51,  1.48it/s]
4|3736|Loss: 0.8509026169776917:  58%|█████▊    | 3735/6470 [42:55<30:51,  1.48it/s]
4|3736|Loss: 0.8509026169776917:  58%|█████▊    | 3736/6470 [42:55<32:42,  1.39it/s]
4|3737|Loss: 0.8335858583450317:  58%|█████▊    | 3736/6470 [42:56<32:42,  1.39it/s]
4|3737|Loss: 0.8335858583450317:  58%|█████▊    | 3737/6470 [42:56<32:47,  1.39it/s]
4|3738|Loss: 0.8274828195571899:  58%|█████▊    | 3737/6470 [42:56<32:47,  1.39it/s]
4|3738|Loss: 0.8274828195571899:  58%|█████▊    | 3738/6470 [42:57<31:33,  1.44it/s]
4|3739|Loss: 0.848664402961731:  58%|█████▊    | 3738/6470 [42:57<31:33,  1.44it/s] 
4|3739|Loss: 0.848664402961731:  58%|█████▊    | 3739/6470 [42:57<30:47,  1.48it/s]
4|3740|Loss: 0.7502426505088806:  58%|█████▊    | 3739/6470 [42:58<30:47,  1.48it/s]
4|3740|Loss: 0.7502426505088806:  58%|█████▊    | 3740/6470 [42:58<31:00,  1.47it/s]
4|3741|Loss: 0.9804295897483826:  58%|█████▊    | 3740/6470 [42:58<31:00,  1.47it/s]
4|3741|Loss: 0.9804295897483826:  58%|█████▊    | 3741/6470 [42:59<32:22,  1.40it/s]
4|3742|Loss: 0.9454293847084045:  58%|█████▊    | 3741/6470 [42:59<32:22,  1.40it/s]
4|3742|Loss: 0.9454293847084045:  58%|█████▊    | 3742/6470 [43:00<31:51,  1.43it/s]
4|3743|Loss: 0.9352672696113586:  58%|█████▊    | 3742/6470 [43:00<31:51,  1.43it/s]
4|3743|Loss: 0.9352672696113586:  58%|█████▊    | 3743/6470 [43:00<30:46,  1.48it/s]
4|3744|Loss: 0.9397262930870056:  58%|█████▊    | 3743/6470 [43:01<30:46,  1.48it/s]
4|3744|Loss: 0.9397262930870056:  58%|█████▊    | 3744/6470 [43:01<32:53,  1.38it/s]
4|3745|Loss: 0.8650669455528259:  58%|█████▊    | 3744/6470 [43:01<32:53,  1.38it/s]
4|3745|Loss: 0.8650669455528259:  58%|█████▊    | 3745/6470 [43:02<31:36,  1.44it/s]
4|3746|Loss: 0.7785030603408813:  58%|█████▊    | 3745/6470 [43:02<31:36,  1.44it/s]
4|3746|Loss: 0.7785030603408813:  58%|█████▊    | 3746/6470 [43:02<32:53,  1.38it/s]
4|3747|Loss: 0.8532139658927917:  58%|█████▊    | 3746/6470 [43:03<32:53,  1.38it/s]
4|3747|Loss: 0.8532139658927917:  58%|█████▊    | 3747/6470 [43:03<32:57,  1.38it/s]
4|3748|Loss: 0.8479702472686768:  58%|█████▊    | 3747/6470 [43:03<32:57,  1.38it/s]
4|3748|Loss: 0.8479702472686768:  58%|█████▊    | 3748/6470 [43:04<31:40,  1.43it/s]
4|3749|Loss: 0.8757022619247437:  58%|█████▊    | 3748/6470 [43:04<31:40,  1.43it/s]
4|3749|Loss: 0.8757022619247437:  58%|█████▊    | 3749/6470 [43:04<30:07,  1.51it/s]
4|3750|Loss: 0.8710229396820068:  58%|█████▊    | 3749/6470 [43:05<30:07,  1.51it/s]
4|3750|Loss: 0.8710229396820068:  58%|█████▊    | 3750/6470 [43:05<30:56,  1.47it/s]
4|3751|Loss: 0.8081379532814026:  58%|█████▊    | 3750/6470 [43:06<30:56,  1.47it/s]
4|3751|Loss: 0.8081379532814026:  58%|█████▊    | 3751/6470 [43:06<32:48,  1.38it/s]
4|3752|Loss: 0.9072872996330261:  58%|█████▊    | 3751/6470 [43:06<32:48,  1.38it/s]
4|3752|Loss: 0.9072872996330261:  58%|█████▊    | 3752/6470 [43:07<33:24,  1.36it/s]
4|3753|Loss: 0.7579441070556641:  58%|█████▊    | 3752/6470 [43:07<33:24,  1.36it/s]
4|3753|Loss: 0.7579441070556641:  58%|█████▊    | 3753/6470 [43:07<31:22,  1.44it/s]
4|3754|Loss: 0.8532399535179138:  58%|█████▊    | 3753/6470 [43:08<31:22,  1.44it/s]
4|3754|Loss: 0.8532399535179138:  58%|█████▊    | 3754/6470 [43:08<29:26,  1.54it/s]
4|3755|Loss: 0.8989830017089844:  58%|█████▊    | 3754/6470 [43:08<29:26,  1.54it/s]
4|3755|Loss: 0.8989830017089844:  58%|█████▊    | 3755/6470 [43:08<29:41,  1.52it/s]
4|3756|Loss: 0.7457661032676697:  58%|█████▊    | 3755/6470 [43:09<29:41,  1.52it/s]
4|3756|Loss: 0.7457661032676697:  58%|█████▊    | 3756/6470 [43:09<27:35,  1.64it/s]
4|3757|Loss: 0.8441165089607239:  58%|█████▊    | 3756/6470 [43:09<27:35,  1.64it/s]
4|3757|Loss: 0.8441165089607239:  58%|█████▊    | 3757/6470 [43:10<29:04,  1.56it/s]
4|3758|Loss: 0.6977617144584656:  58%|█████▊    | 3757/6470 [43:10<29:04,  1.56it/s]
4|3758|Loss: 0.6977617144584656:  58%|█████▊    | 3758/6470 [43:10<27:27,  1.65it/s]
4|3759|Loss: 0.9730989933013916:  58%|█████▊    | 3758/6470 [43:11<27:27,  1.65it/s]
4|3759|Loss: 0.9730989933013916:  58%|█████▊    | 3759/6470 [43:11<27:02,  1.67it/s]
4|3760|Loss: 0.8591583371162415:  58%|█████▊    | 3759/6470 [43:11<27:02,  1.67it/s]
4|3760|Loss: 0.8591583371162415:  58%|█████▊    | 3760/6470 [43:11<27:28,  1.64it/s]
4|3761|Loss: 0.540724515914917:  58%|█████▊    | 3760/6470 [43:12<27:28,  1.64it/s] 
4|3761|Loss: 0.540724515914917:  58%|█████▊    | 3761/6470 [43:12<27:51,  1.62it/s]
4|3762|Loss: 0.9101772904396057:  58%|█████▊    | 3761/6470 [43:12<27:51,  1.62it/s]
4|3762|Loss: 0.9101772904396057:  58%|█████▊    | 3762/6470 [43:13<29:17,  1.54it/s]
4|3763|Loss: 0.8215609788894653:  58%|█████▊    | 3762/6470 [43:13<29:17,  1.54it/s]
4|3763|Loss: 0.8215609788894653:  58%|█████▊    | 3763/6470 [43:14<31:37,  1.43it/s]
4|3764|Loss: 1.0204787254333496:  58%|█████▊    | 3763/6470 [43:14<31:37,  1.43it/s]
4|3764|Loss: 1.0204787254333496:  58%|█████▊    | 3764/6470 [43:14<32:31,  1.39it/s]
4|3765|Loss: 0.8268813490867615:  58%|█████▊    | 3764/6470 [43:15<32:31,  1.39it/s]
4|3765|Loss: 0.8268813490867615:  58%|█████▊    | 3765/6470 [43:15<32:56,  1.37it/s]
4|3766|Loss: 0.8734686970710754:  58%|█████▊    | 3765/6470 [43:15<32:56,  1.37it/s]
4|3766|Loss: 0.8734686970710754:  58%|█████▊    | 3766/6470 [43:16<29:16,  1.54it/s]
4|3767|Loss: 0.7125259041786194:  58%|█████▊    | 3766/6470 [43:16<29:16,  1.54it/s]
4|3767|Loss: 0.7125259041786194:  58%|█████▊    | 3767/6470 [43:16<30:14,  1.49it/s]
4|3768|Loss: 0.8533012270927429:  58%|█████▊    | 3767/6470 [43:17<30:14,  1.49it/s]
4|3768|Loss: 0.8533012270927429:  58%|█████▊    | 3768/6470 [43:17<30:08,  1.49it/s]
4|3769|Loss: 0.8807781338691711:  58%|█████▊    | 3768/6470 [43:17<30:08,  1.49it/s]
4|3769|Loss: 0.8807781338691711:  58%|█████▊    | 3769/6470 [43:18<28:56,  1.56it/s]
4|3770|Loss: 0.8401188850402832:  58%|█████▊    | 3769/6470 [43:18<28:56,  1.56it/s]
4|3770|Loss: 0.8401188850402832:  58%|█████▊    | 3770/6470 [43:18<28:45,  1.56it/s]
4|3771|Loss: 0.8007761240005493:  58%|█████▊    | 3770/6470 [43:19<28:45,  1.56it/s]
4|3771|Loss: 0.8007761240005493:  58%|█████▊    | 3771/6470 [43:19<28:37,  1.57it/s]
4|3772|Loss: 0.7535245418548584:  58%|█████▊    | 3771/6470 [43:19<28:37,  1.57it/s]
4|3772|Loss: 0.7535245418548584:  58%|█████▊    | 3772/6470 [43:20<29:50,  1.51it/s]
4|3773|Loss: 0.9496066570281982:  58%|█████▊    | 3772/6470 [43:20<29:50,  1.51it/s]
4|3773|Loss: 0.9496066570281982:  58%|█████▊    | 3773/6470 [43:20<32:49,  1.37it/s]
4|3774|Loss: 0.8585931658744812:  58%|█████▊    | 3773/6470 [43:21<32:49,  1.37it/s]
4|3774|Loss: 0.8585931658744812:  58%|█████▊    | 3774/6470 [43:21<34:10,  1.32it/s]
4|3775|Loss: 0.9625368118286133:  58%|█████▊    | 3774/6470 [43:22<34:10,  1.32it/s]
4|3775|Loss: 0.9625368118286133:  58%|█████▊    | 3775/6470 [43:22<34:40,  1.30it/s]
4|3776|Loss: 0.7749220728874207:  58%|█████▊    | 3775/6470 [43:23<34:40,  1.30it/s]
4|3776|Loss: 0.7749220728874207:  58%|█████▊    | 3776/6470 [43:23<35:41,  1.26it/s]
4|3777|Loss: 0.8145262598991394:  58%|█████▊    | 3776/6470 [43:23<35:41,  1.26it/s]
4|3777|Loss: 0.8145262598991394:  58%|█████▊    | 3777/6470 [43:24<35:55,  1.25it/s]
4|3778|Loss: 0.9213932752609253:  58%|█████▊    | 3777/6470 [43:24<35:55,  1.25it/s]
4|3778|Loss: 0.9213932752609253:  58%|█████▊    | 3778/6470 [43:24<33:31,  1.34it/s]
4|3779|Loss: 0.9438085556030273:  58%|█████▊    | 3778/6470 [43:25<33:31,  1.34it/s]
4|3779|Loss: 0.9438085556030273:  58%|█████▊    | 3779/6470 [43:25<33:04,  1.36it/s]
4|3780|Loss: 1.1228832006454468:  58%|█████▊    | 3779/6470 [43:26<33:04,  1.36it/s]
4|3780|Loss: 1.1228832006454468:  58%|█████▊    | 3780/6470 [43:26<39:23,  1.14it/s]
4|3781|Loss: 0.8791785836219788:  58%|█████▊    | 3780/6470 [43:27<39:23,  1.14it/s]
4|3781|Loss: 0.8791785836219788:  58%|█████▊    | 3781/6470 [43:27<38:03,  1.18it/s]
4|3782|Loss: 0.627271294593811:  58%|█████▊    | 3781/6470 [43:27<38:03,  1.18it/s] 
4|3782|Loss: 0.627271294593811:  58%|█████▊    | 3782/6470 [43:28<35:11,  1.27it/s]
4|3783|Loss: 0.8564165830612183:  58%|█████▊    | 3782/6470 [43:28<35:11,  1.27it/s]
4|3783|Loss: 0.8564165830612183:  58%|█████▊    | 3783/6470 [43:28<33:55,  1.32it/s]
4|3784|Loss: 0.9343886971473694:  58%|█████▊    | 3783/6470 [43:29<33:55,  1.32it/s]
4|3784|Loss: 0.9343886971473694:  58%|█████▊    | 3784/6470 [43:29<34:39,  1.29it/s]
4|3785|Loss: 0.7586124539375305:  58%|█████▊    | 3784/6470 [43:30<34:39,  1.29it/s]
4|3785|Loss: 0.7586124539375305:  59%|█████▊    | 3785/6470 [43:30<33:25,  1.34it/s]
4|3786|Loss: 0.7907019853591919:  59%|█████▊    | 3785/6470 [43:30<33:25,  1.34it/s]
4|3786|Loss: 0.7907019853591919:  59%|█████▊    | 3786/6470 [43:31<32:57,  1.36it/s]
4|3787|Loss: 0.8650499582290649:  59%|█████▊    | 3786/6470 [43:31<32:57,  1.36it/s]
4|3787|Loss: 0.8650499582290649:  59%|█████▊    | 3787/6470 [43:31<29:11,  1.53it/s]
4|3788|Loss: 0.8786200284957886:  59%|█████▊    | 3787/6470 [43:31<29:11,  1.53it/s]
4|3788|Loss: 0.8786200284957886:  59%|█████▊    | 3788/6470 [43:32<29:08,  1.53it/s]
4|3789|Loss: 0.9635533690452576:  59%|█████▊    | 3788/6470 [43:32<29:08,  1.53it/s]
4|3789|Loss: 0.9635533690452576:  59%|█████▊    | 3789/6470 [43:33<32:14,  1.39it/s]
4|3790|Loss: 0.8899326920509338:  59%|█████▊    | 3789/6470 [43:33<32:14,  1.39it/s]
4|3790|Loss: 0.8899326920509338:  59%|█████▊    | 3790/6470 [43:33<32:37,  1.37it/s]
4|3791|Loss: 1.1005321741104126:  59%|█████▊    | 3790/6470 [43:34<32:37,  1.37it/s]
4|3791|Loss: 1.1005321741104126:  59%|█████▊    | 3791/6470 [43:34<31:30,  1.42it/s]
4|3792|Loss: 0.7506980895996094:  59%|█████▊    | 3791/6470 [43:34<31:30,  1.42it/s]
4|3792|Loss: 0.7506980895996094:  59%|█████▊    | 3792/6470 [43:35<29:09,  1.53it/s]
4|3793|Loss: 0.9254788160324097:  59%|█████▊    | 3792/6470 [43:35<29:09,  1.53it/s]
4|3793|Loss: 0.9254788160324097:  59%|█████▊    | 3793/6470 [43:35<28:31,  1.56it/s]
4|3794|Loss: 0.9164032340049744:  59%|█████▊    | 3793/6470 [43:35<28:31,  1.56it/s]
4|3794|Loss: 0.9164032340049744:  59%|█████▊    | 3794/6470 [43:36<28:15,  1.58it/s]
4|3795|Loss: 0.8645742535591125:  59%|█████▊    | 3794/6470 [43:36<28:15,  1.58it/s]
4|3795|Loss: 0.8645742535591125:  59%|█████▊    | 3795/6470 [43:36<29:35,  1.51it/s]
4|3796|Loss: 0.8778492212295532:  59%|█████▊    | 3795/6470 [43:37<29:35,  1.51it/s]
4|3796|Loss: 0.8778492212295532:  59%|█████▊    | 3796/6470 [43:37<30:41,  1.45it/s]
4|3797|Loss: 0.704780101776123:  59%|█████▊    | 3796/6470 [43:38<30:41,  1.45it/s] 
4|3797|Loss: 0.704780101776123:  59%|█████▊    | 3797/6470 [43:38<32:25,  1.37it/s]
4|3798|Loss: 0.9827437996864319:  59%|█████▊    | 3797/6470 [43:38<32:25,  1.37it/s]
4|3798|Loss: 0.9827437996864319:  59%|█████▊    | 3798/6470 [43:39<34:06,  1.31it/s]
4|3799|Loss: 0.8015987873077393:  59%|█████▊    | 3798/6470 [43:39<34:06,  1.31it/s]
4|3799|Loss: 0.8015987873077393:  59%|█████▊    | 3799/6470 [43:40<33:40,  1.32it/s]
4|3800|Loss: 0.8253445029258728:  59%|█████▊    | 3799/6470 [43:40<33:40,  1.32it/s]
4|3800|Loss: 0.8253445029258728:  59%|█████▊    | 3800/6470 [43:40<33:21,  1.33it/s]
4|3801|Loss: 0.9236355423927307:  59%|█████▊    | 3800/6470 [43:41<33:21,  1.33it/s]
4|3801|Loss: 0.9236355423927307:  59%|█████▊    | 3801/6470 [43:41<31:28,  1.41it/s]
4|3802|Loss: 0.9641608595848083:  59%|█████▊    | 3801/6470 [43:41<31:28,  1.41it/s]
4|3802|Loss: 0.9641608595848083:  59%|█████▉    | 3802/6470 [43:42<31:09,  1.43it/s]
4|3803|Loss: 0.8093549013137817:  59%|█████▉    | 3802/6470 [43:42<31:09,  1.43it/s]
4|3803|Loss: 0.8093549013137817:  59%|█████▉    | 3803/6470 [43:42<31:42,  1.40it/s]
4|3804|Loss: 0.9075517654418945:  59%|█████▉    | 3803/6470 [43:43<31:42,  1.40it/s]
4|3804|Loss: 0.9075517654418945:  59%|█████▉    | 3804/6470 [43:43<31:05,  1.43it/s]
4|3805|Loss: 0.8245128393173218:  59%|█████▉    | 3804/6470 [43:43<31:05,  1.43it/s]
4|3805|Loss: 0.8245128393173218:  59%|█████▉    | 3805/6470 [43:44<32:01,  1.39it/s]
4|3806|Loss: 0.8588785529136658:  59%|█████▉    | 3805/6470 [43:44<32:01,  1.39it/s]
4|3806|Loss: 0.8588785529136658:  59%|█████▉    | 3806/6470 [43:45<33:38,  1.32it/s]
4|3807|Loss: 0.8959906101226807:  59%|█████▉    | 3806/6470 [43:45<33:38,  1.32it/s]
4|3807|Loss: 0.8959906101226807:  59%|█████▉    | 3807/6470 [43:45<32:47,  1.35it/s]
4|3808|Loss: 0.8852109909057617:  59%|█████▉    | 3807/6470 [43:46<32:47,  1.35it/s]
4|3808|Loss: 0.8852109909057617:  59%|█████▉    | 3808/6470 [43:46<29:26,  1.51it/s]
4|3809|Loss: 0.758042573928833:  59%|█████▉    | 3808/6470 [43:46<29:26,  1.51it/s] 
4|3809|Loss: 0.758042573928833:  59%|█████▉    | 3809/6470 [43:46<25:40,  1.73it/s]
4|3810|Loss: 0.8471161127090454:  59%|█████▉    | 3809/6470 [43:47<25:40,  1.73it/s]
4|3810|Loss: 0.8471161127090454:  59%|█████▉    | 3810/6470 [43:47<25:52,  1.71it/s]
4|3811|Loss: 0.8141685724258423:  59%|█████▉    | 3810/6470 [43:47<25:52,  1.71it/s]
4|3811|Loss: 0.8141685724258423:  59%|█████▉    | 3811/6470 [43:47<26:40,  1.66it/s]
4|3812|Loss: 0.9292060136795044:  59%|█████▉    | 3811/6470 [43:48<26:40,  1.66it/s]
4|3812|Loss: 0.9292060136795044:  59%|█████▉    | 3812/6470 [43:48<25:59,  1.70it/s]
4|3813|Loss: 0.7177771329879761:  59%|█████▉    | 3812/6470 [43:48<25:59,  1.70it/s]
4|3813|Loss: 0.7177771329879761:  59%|█████▉    | 3813/6470 [43:49<25:42,  1.72it/s]
4|3814|Loss: 0.9820955991744995:  59%|█████▉    | 3813/6470 [43:49<25:42,  1.72it/s]
4|3814|Loss: 0.9820955991744995:  59%|█████▉    | 3814/6470 [43:49<28:20,  1.56it/s]
4|3815|Loss: 0.8078807592391968:  59%|█████▉    | 3814/6470 [43:50<28:20,  1.56it/s]
4|3815|Loss: 0.8078807592391968:  59%|█████▉    | 3815/6470 [43:50<28:39,  1.54it/s]
4|3816|Loss: 0.8991219997406006:  59%|█████▉    | 3815/6470 [43:50<28:39,  1.54it/s]
4|3816|Loss: 0.8991219997406006:  59%|█████▉    | 3816/6470 [43:51<28:16,  1.56it/s]
4|3817|Loss: 0.8515557646751404:  59%|█████▉    | 3816/6470 [43:51<28:16,  1.56it/s]
4|3817|Loss: 0.8515557646751404:  59%|█████▉    | 3817/6470 [43:51<29:11,  1.51it/s]
4|3818|Loss: 0.8482585549354553:  59%|█████▉    | 3817/6470 [43:52<29:11,  1.51it/s]
4|3818|Loss: 0.8482585549354553:  59%|█████▉    | 3818/6470 [43:52<31:41,  1.39it/s]
4|3819|Loss: 0.7957654595375061:  59%|█████▉    | 3818/6470 [43:53<31:41,  1.39it/s]
4|3819|Loss: 0.7957654595375061:  59%|█████▉    | 3819/6470 [43:53<32:01,  1.38it/s]
4|3820|Loss: 0.9430489540100098:  59%|█████▉    | 3819/6470 [43:53<32:01,  1.38it/s]
4|3820|Loss: 0.9430489540100098:  59%|█████▉    | 3820/6470 [43:54<30:17,  1.46it/s]
4|3821|Loss: 0.8740174770355225:  59%|█████▉    | 3820/6470 [43:54<30:17,  1.46it/s]
4|3821|Loss: 0.8740174770355225:  59%|█████▉    | 3821/6470 [43:54<29:18,  1.51it/s]
4|3822|Loss: 0.8256378769874573:  59%|█████▉    | 3821/6470 [43:55<29:18,  1.51it/s]
4|3822|Loss: 0.8256378769874573:  59%|█████▉    | 3822/6470 [43:55<31:18,  1.41it/s]
4|3823|Loss: 0.8089891672134399:  59%|█████▉    | 3822/6470 [43:55<31:18,  1.41it/s]
4|3823|Loss: 0.8089891672134399:  59%|█████▉    | 3823/6470 [43:56<31:05,  1.42it/s]
4|3824|Loss: 0.8607043623924255:  59%|█████▉    | 3823/6470 [43:56<31:05,  1.42it/s]
4|3824|Loss: 0.8607043623924255:  59%|█████▉    | 3824/6470 [43:56<28:34,  1.54it/s]
4|3825|Loss: 0.8354561924934387:  59%|█████▉    | 3824/6470 [43:57<28:34,  1.54it/s]
4|3825|Loss: 0.8354561924934387:  59%|█████▉    | 3825/6470 [43:57<30:31,  1.44it/s]
4|3826|Loss: 0.8712939620018005:  59%|█████▉    | 3825/6470 [43:57<30:31,  1.44it/s]
4|3826|Loss: 0.8712939620018005:  59%|█████▉    | 3826/6470 [43:58<31:26,  1.40it/s]
4|3827|Loss: 0.9907554388046265:  59%|█████▉    | 3826/6470 [43:58<31:26,  1.40it/s]
4|3827|Loss: 0.9907554388046265:  59%|█████▉    | 3827/6470 [43:58<31:48,  1.38it/s]
4|3828|Loss: 0.7859669923782349:  59%|█████▉    | 3827/6470 [43:59<31:48,  1.38it/s]
4|3828|Loss: 0.7859669923782349:  59%|█████▉    | 3828/6470 [43:59<31:58,  1.38it/s]
4|3829|Loss: 0.8133395314216614:  59%|█████▉    | 3828/6470 [44:00<31:58,  1.38it/s]
4|3829|Loss: 0.8133395314216614:  59%|█████▉    | 3829/6470 [44:00<33:07,  1.33it/s]
4|3830|Loss: 0.7421013712882996:  59%|█████▉    | 3829/6470 [44:00<33:07,  1.33it/s]
4|3830|Loss: 0.7421013712882996:  59%|█████▉    | 3830/6470 [44:01<33:11,  1.33it/s]
4|3831|Loss: 0.7622151374816895:  59%|█████▉    | 3830/6470 [44:01<33:11,  1.33it/s]
4|3831|Loss: 0.7622151374816895:  59%|█████▉    | 3831/6470 [44:02<33:00,  1.33it/s]
4|3832|Loss: 0.8454450368881226:  59%|█████▉    | 3831/6470 [44:02<33:00,  1.33it/s]
4|3832|Loss: 0.8454450368881226:  59%|█████▉    | 3832/6470 [44:02<33:22,  1.32it/s]
4|3833|Loss: 0.8913394212722778:  59%|█████▉    | 3832/6470 [44:03<33:22,  1.32it/s]
4|3833|Loss: 0.8913394212722778:  59%|█████▉    | 3833/6470 [44:03<30:14,  1.45it/s]
4|3834|Loss: 1.0143065452575684:  59%|█████▉    | 3833/6470 [44:03<30:14,  1.45it/s]
4|3834|Loss: 1.0143065452575684:  59%|█████▉    | 3834/6470 [44:03<29:09,  1.51it/s]
4|3835|Loss: 0.7632611393928528:  59%|█████▉    | 3834/6470 [44:04<29:09,  1.51it/s]
4|3835|Loss: 0.7632611393928528:  59%|█████▉    | 3835/6470 [44:04<31:06,  1.41it/s]
4|3836|Loss: 0.8407850861549377:  59%|█████▉    | 3835/6470 [44:05<31:06,  1.41it/s]
4|3836|Loss: 0.8407850861549377:  59%|█████▉    | 3836/6470 [44:05<30:40,  1.43it/s]
4|3837|Loss: 0.7700560092926025:  59%|█████▉    | 3836/6470 [44:05<30:40,  1.43it/s]
4|3837|Loss: 0.7700560092926025:  59%|█████▉    | 3837/6470 [44:06<28:54,  1.52it/s]
4|3838|Loss: 0.8723681569099426:  59%|█████▉    | 3837/6470 [44:06<28:54,  1.52it/s]
4|3838|Loss: 0.8723681569099426:  59%|█████▉    | 3838/6470 [44:06<29:29,  1.49it/s]
4|3839|Loss: 0.7612298727035522:  59%|█████▉    | 3838/6470 [44:07<29:29,  1.49it/s]
4|3839|Loss: 0.7612298727035522:  59%|█████▉    | 3839/6470 [44:07<30:18,  1.45it/s]
4|3840|Loss: 0.9057568907737732:  59%|█████▉    | 3839/6470 [44:07<30:18,  1.45it/s]
4|3840|Loss: 0.9057568907737732:  59%|█████▉    | 3840/6470 [44:08<31:05,  1.41it/s]
4|3841|Loss: 0.9494091272354126:  59%|█████▉    | 3840/6470 [44:08<31:05,  1.41it/s]
4|3841|Loss: 0.9494091272354126:  59%|█████▉    | 3841/6470 [44:09<32:56,  1.33it/s]
4|3842|Loss: 0.8749588131904602:  59%|█████▉    | 3841/6470 [44:09<32:56,  1.33it/s]
4|3842|Loss: 0.8749588131904602:  59%|█████▉    | 3842/6470 [44:09<30:12,  1.45it/s]
4|3843|Loss: 0.8291823863983154:  59%|█████▉    | 3842/6470 [44:09<30:12,  1.45it/s]
4|3843|Loss: 0.8291823863983154:  59%|█████▉    | 3843/6470 [44:10<30:34,  1.43it/s]
4|3844|Loss: 0.927445113658905:  59%|█████▉    | 3843/6470 [44:10<30:34,  1.43it/s] 
4|3844|Loss: 0.927445113658905:  59%|█████▉    | 3844/6470 [44:11<32:36,  1.34it/s]
4|3845|Loss: 0.8775197267532349:  59%|█████▉    | 3844/6470 [44:11<32:36,  1.34it/s]
4|3845|Loss: 0.8775197267532349:  59%|█████▉    | 3845/6470 [44:11<33:05,  1.32it/s]
4|3846|Loss: 0.8182960748672485:  59%|█████▉    | 3845/6470 [44:12<33:05,  1.32it/s]
4|3846|Loss: 0.8182960748672485:  59%|█████▉    | 3846/6470 [44:12<32:50,  1.33it/s]
4|3847|Loss: 0.8727887272834778:  59%|█████▉    | 3846/6470 [44:13<32:50,  1.33it/s]
4|3847|Loss: 0.8727887272834778:  59%|█████▉    | 3847/6470 [44:13<33:52,  1.29it/s]
4|3848|Loss: 0.7823371887207031:  59%|█████▉    | 3847/6470 [44:13<33:52,  1.29it/s]
4|3848|Loss: 0.7823371887207031:  59%|█████▉    | 3848/6470 [44:14<33:20,  1.31it/s]
4|3849|Loss: 0.747201144695282:  59%|█████▉    | 3848/6470 [44:14<33:20,  1.31it/s] 
4|3849|Loss: 0.747201144695282:  59%|█████▉    | 3849/6470 [44:14<31:13,  1.40it/s]
4|3850|Loss: 0.8061021566390991:  59%|█████▉    | 3849/6470 [44:15<31:13,  1.40it/s]
4|3850|Loss: 0.8061021566390991:  60%|█████▉    | 3850/6470 [44:15<31:40,  1.38it/s]
4|3851|Loss: 0.8788695335388184:  60%|█████▉    | 3850/6470 [44:16<31:40,  1.38it/s]
4|3851|Loss: 0.8788695335388184:  60%|█████▉    | 3851/6470 [44:16<31:39,  1.38it/s]
4|3852|Loss: 0.9842081665992737:  60%|█████▉    | 3851/6470 [44:16<31:39,  1.38it/s]
4|3852|Loss: 0.9842081665992737:  60%|█████▉    | 3852/6470 [44:17<31:51,  1.37it/s]
4|3853|Loss: 0.7883073687553406:  60%|█████▉    | 3852/6470 [44:17<31:51,  1.37it/s]
4|3853|Loss: 0.7883073687553406:  60%|█████▉    | 3853/6470 [44:17<30:46,  1.42it/s]
4|3854|Loss: 0.6905721426010132:  60%|█████▉    | 3853/6470 [44:18<30:46,  1.42it/s]
4|3854|Loss: 0.6905721426010132:  60%|█████▉    | 3854/6470 [44:18<28:26,  1.53it/s]
4|3855|Loss: 0.873854398727417:  60%|█████▉    | 3854/6470 [44:18<28:26,  1.53it/s] 
4|3855|Loss: 0.873854398727417:  60%|█████▉    | 3855/6470 [44:18<27:56,  1.56it/s]
4|3856|Loss: 1.093563199043274:  60%|█████▉    | 3855/6470 [44:19<27:56,  1.56it/s]
4|3856|Loss: 1.093563199043274:  60%|█████▉    | 3856/6470 [44:19<27:02,  1.61it/s]
4|3857|Loss: 0.7661356329917908:  60%|█████▉    | 3856/6470 [44:19<27:02,  1.61it/s]
4|3857|Loss: 0.7661356329917908:  60%|█████▉    | 3857/6470 [44:20<28:07,  1.55it/s]
4|3858|Loss: 0.9210701584815979:  60%|█████▉    | 3857/6470 [44:20<28:07,  1.55it/s]
4|3858|Loss: 0.9210701584815979:  60%|█████▉    | 3858/6470 [44:20<29:22,  1.48it/s]
4|3859|Loss: 0.8137157559394836:  60%|█████▉    | 3858/6470 [44:21<29:22,  1.48it/s]
4|3859|Loss: 0.8137157559394836:  60%|█████▉    | 3859/6470 [44:21<28:33,  1.52it/s]
4|3860|Loss: 1.2003332376480103:  60%|█████▉    | 3859/6470 [44:21<28:33,  1.52it/s]
4|3860|Loss: 1.2003332376480103:  60%|█████▉    | 3860/6470 [44:22<28:56,  1.50it/s]
4|3861|Loss: 0.8797494173049927:  60%|█████▉    | 3860/6470 [44:22<28:56,  1.50it/s]
4|3861|Loss: 0.8797494173049927:  60%|█████▉    | 3861/6470 [44:22<30:25,  1.43it/s]
4|3862|Loss: 0.9429518580436707:  60%|█████▉    | 3861/6470 [44:23<30:25,  1.43it/s]
4|3862|Loss: 0.9429518580436707:  60%|█████▉    | 3862/6470 [44:23<31:10,  1.39it/s]
4|3863|Loss: 0.9339184761047363:  60%|█████▉    | 3862/6470 [44:24<31:10,  1.39it/s]
4|3863|Loss: 0.9339184761047363:  60%|█████▉    | 3863/6470 [44:24<32:42,  1.33it/s]
4|3864|Loss: 0.8760456442832947:  60%|█████▉    | 3863/6470 [44:24<32:42,  1.33it/s]
4|3864|Loss: 0.8760456442832947:  60%|█████▉    | 3864/6470 [44:25<32:52,  1.32it/s]
4|3865|Loss: 1.0457829236984253:  60%|█████▉    | 3864/6470 [44:25<32:52,  1.32it/s]
4|3865|Loss: 1.0457829236984253:  60%|█████▉    | 3865/6470 [44:26<32:36,  1.33it/s]
4|3866|Loss: 0.7172491550445557:  60%|█████▉    | 3865/6470 [44:26<32:36,  1.33it/s]
4|3866|Loss: 0.7172491550445557:  60%|█████▉    | 3866/6470 [44:26<30:30,  1.42it/s]
4|3867|Loss: 0.9702215790748596:  60%|█████▉    | 3866/6470 [44:26<30:30,  1.42it/s]
4|3867|Loss: 0.9702215790748596:  60%|█████▉    | 3867/6470 [44:27<29:12,  1.49it/s]
4|3868|Loss: 0.8941277265548706:  60%|█████▉    | 3867/6470 [44:27<29:12,  1.49it/s]
4|3868|Loss: 0.8941277265548706:  60%|█████▉    | 3868/6470 [44:27<28:29,  1.52it/s]
4|3869|Loss: 0.9085036516189575:  60%|█████▉    | 3868/6470 [44:28<28:29,  1.52it/s]
4|3869|Loss: 0.9085036516189575:  60%|█████▉    | 3869/6470 [44:28<29:08,  1.49it/s]
4|3870|Loss: 0.9563606381416321:  60%|█████▉    | 3869/6470 [44:29<29:08,  1.49it/s]
4|3870|Loss: 0.9563606381416321:  60%|█████▉    | 3870/6470 [44:29<30:40,  1.41it/s]
4|3871|Loss: 0.9056933522224426:  60%|█████▉    | 3870/6470 [44:29<30:40,  1.41it/s]
4|3871|Loss: 0.9056933522224426:  60%|█████▉    | 3871/6470 [44:29<29:18,  1.48it/s]
4|3872|Loss: 0.9446385502815247:  60%|█████▉    | 3871/6470 [44:30<29:18,  1.48it/s]
4|3872|Loss: 0.9446385502815247:  60%|█████▉    | 3872/6470 [44:30<27:59,  1.55it/s]
4|3873|Loss: 0.8563364744186401:  60%|█████▉    | 3872/6470 [44:30<27:59,  1.55it/s]
4|3873|Loss: 0.8563364744186401:  60%|█████▉    | 3873/6470 [44:31<29:44,  1.46it/s]
4|3874|Loss: 0.7924991846084595:  60%|█████▉    | 3873/6470 [44:31<29:44,  1.46it/s]
4|3874|Loss: 0.7924991846084595:  60%|█████▉    | 3874/6470 [44:32<31:18,  1.38it/s]
4|3875|Loss: 0.6763499975204468:  60%|█████▉    | 3874/6470 [44:32<31:18,  1.38it/s]
4|3875|Loss: 0.6763499975204468:  60%|█████▉    | 3875/6470 [44:32<31:17,  1.38it/s]
4|3876|Loss: 0.8520745038986206:  60%|█████▉    | 3875/6470 [44:33<31:17,  1.38it/s]
4|3876|Loss: 0.8520745038986206:  60%|█████▉    | 3876/6470 [44:33<31:01,  1.39it/s]
4|3877|Loss: 0.8668984770774841:  60%|█████▉    | 3876/6470 [44:33<31:01,  1.39it/s]
4|3877|Loss: 0.8668984770774841:  60%|█████▉    | 3877/6470 [44:34<29:12,  1.48it/s]
4|3878|Loss: 0.9162919521331787:  60%|█████▉    | 3877/6470 [44:34<29:12,  1.48it/s]
4|3878|Loss: 0.9162919521331787:  60%|█████▉    | 3878/6470 [44:34<30:56,  1.40it/s]
4|3879|Loss: 0.7314371466636658:  60%|█████▉    | 3878/6470 [44:35<30:56,  1.40it/s]
4|3879|Loss: 0.7314371466636658:  60%|█████▉    | 3879/6470 [44:35<30:25,  1.42it/s]
4|3880|Loss: 0.7718188762664795:  60%|█████▉    | 3879/6470 [44:36<30:25,  1.42it/s]
4|3880|Loss: 0.7718188762664795:  60%|█████▉    | 3880/6470 [44:36<31:27,  1.37it/s]
4|3881|Loss: 0.8231309056282043:  60%|█████▉    | 3880/6470 [44:36<31:27,  1.37it/s]
4|3881|Loss: 0.8231309056282043:  60%|█████▉    | 3881/6470 [44:36<29:27,  1.47it/s]
4|3882|Loss: 0.8392137885093689:  60%|█████▉    | 3881/6470 [44:37<29:27,  1.47it/s]
4|3882|Loss: 0.8392137885093689:  60%|██████    | 3882/6470 [44:37<27:38,  1.56it/s]
4|3883|Loss: 0.9199573397636414:  60%|██████    | 3882/6470 [44:37<27:38,  1.56it/s]
4|3883|Loss: 0.9199573397636414:  60%|██████    | 3883/6470 [44:38<28:07,  1.53it/s]
4|3884|Loss: 0.7825987935066223:  60%|██████    | 3883/6470 [44:38<28:07,  1.53it/s]
4|3884|Loss: 0.7825987935066223:  60%|██████    | 3884/6470 [44:38<29:12,  1.48it/s]
4|3885|Loss: 0.8812535405158997:  60%|██████    | 3884/6470 [44:39<29:12,  1.48it/s]
4|3885|Loss: 0.8812535405158997:  60%|██████    | 3885/6470 [44:39<27:51,  1.55it/s]
4|3886|Loss: 0.8817286491394043:  60%|██████    | 3885/6470 [44:39<27:51,  1.55it/s]
4|3886|Loss: 0.8817286491394043:  60%|██████    | 3886/6470 [44:40<28:05,  1.53it/s]
4|3887|Loss: 0.6911445260047913:  60%|██████    | 3886/6470 [44:40<28:05,  1.53it/s]
4|3887|Loss: 0.6911445260047913:  60%|██████    | 3887/6470 [44:40<28:03,  1.53it/s]
4|3888|Loss: 0.9064791202545166:  60%|██████    | 3887/6470 [44:41<28:03,  1.53it/s]
4|3888|Loss: 0.9064791202545166:  60%|██████    | 3888/6470 [44:41<28:32,  1.51it/s]
4|3889|Loss: 0.7469190359115601:  60%|██████    | 3888/6470 [44:41<28:32,  1.51it/s]
4|3889|Loss: 0.7469190359115601:  60%|██████    | 3889/6470 [44:42<30:00,  1.43it/s]
4|3890|Loss: 0.7403603792190552:  60%|██████    | 3889/6470 [44:42<30:00,  1.43it/s]
4|3890|Loss: 0.7403603792190552:  60%|██████    | 3890/6470 [44:43<30:42,  1.40it/s]
4|3891|Loss: 0.847289502620697:  60%|██████    | 3890/6470 [44:43<30:42,  1.40it/s] 
4|3891|Loss: 0.847289502620697:  60%|██████    | 3891/6470 [44:43<30:41,  1.40it/s]
4|3892|Loss: 0.8679195046424866:  60%|██████    | 3891/6470 [44:44<30:41,  1.40it/s]
4|3892|Loss: 0.8679195046424866:  60%|██████    | 3892/6470 [44:44<32:01,  1.34it/s]
4|3893|Loss: 0.7858823537826538:  60%|██████    | 3892/6470 [44:44<32:01,  1.34it/s]
4|3893|Loss: 0.7858823537826538:  60%|██████    | 3893/6470 [44:45<30:37,  1.40it/s]
4|3894|Loss: 0.7911801934242249:  60%|██████    | 3893/6470 [44:45<30:37,  1.40it/s]
4|3894|Loss: 0.7911801934242249:  60%|██████    | 3894/6470 [44:46<31:31,  1.36it/s]
4|3895|Loss: 0.7824901342391968:  60%|██████    | 3894/6470 [44:46<31:31,  1.36it/s]
4|3895|Loss: 0.7824901342391968:  60%|██████    | 3895/6470 [44:46<32:18,  1.33it/s]
4|3896|Loss: 0.7185816168785095:  60%|██████    | 3895/6470 [44:47<32:18,  1.33it/s]
4|3896|Loss: 0.7185816168785095:  60%|██████    | 3896/6470 [44:47<29:39,  1.45it/s]
4|3897|Loss: 0.6976187229156494:  60%|██████    | 3896/6470 [44:47<29:39,  1.45it/s]
4|3897|Loss: 0.6976187229156494:  60%|██████    | 3897/6470 [44:48<30:58,  1.38it/s]
4|3898|Loss: 0.8260564208030701:  60%|██████    | 3897/6470 [44:48<30:58,  1.38it/s]
4|3898|Loss: 0.8260564208030701:  60%|██████    | 3898/6470 [44:48<31:42,  1.35it/s]
4|3899|Loss: 0.9277220964431763:  60%|██████    | 3898/6470 [44:49<31:42,  1.35it/s]
4|3899|Loss: 0.9277220964431763:  60%|██████    | 3899/6470 [44:49<31:42,  1.35it/s]
4|3900|Loss: 0.8320173025131226:  60%|██████    | 3899/6470 [44:50<31:42,  1.35it/s]
4|3900|Loss: 0.8320173025131226:  60%|██████    | 3900/6470 [44:50<32:49,  1.30it/s]
4|3901|Loss: 0.6453326940536499:  60%|██████    | 3900/6470 [44:50<32:49,  1.30it/s]
4|3901|Loss: 0.6453326940536499:  60%|██████    | 3901/6470 [44:50<28:39,  1.49it/s]
4|3902|Loss: 0.7861287593841553:  60%|██████    | 3901/6470 [44:51<28:39,  1.49it/s]
4|3902|Loss: 0.7861287593841553:  60%|██████    | 3902/6470 [44:51<28:57,  1.48it/s]
4|3903|Loss: 1.0057569742202759:  60%|██████    | 3902/6470 [44:52<28:57,  1.48it/s]
4|3903|Loss: 1.0057569742202759:  60%|██████    | 3903/6470 [44:52<30:06,  1.42it/s]
4|3904|Loss: 0.8526224493980408:  60%|██████    | 3903/6470 [44:52<30:06,  1.42it/s]
4|3904|Loss: 0.8526224493980408:  60%|██████    | 3904/6470 [44:53<32:06,  1.33it/s]
4|3905|Loss: 0.8855797052383423:  60%|██████    | 3904/6470 [44:53<32:06,  1.33it/s]
4|3905|Loss: 0.8855797052383423:  60%|██████    | 3905/6470 [44:54<32:22,  1.32it/s]
4|3906|Loss: 0.8016452193260193:  60%|██████    | 3905/6470 [44:54<32:22,  1.32it/s]
4|3906|Loss: 0.8016452193260193:  60%|██████    | 3906/6470 [44:54<30:36,  1.40it/s]
4|3907|Loss: 0.8607970476150513:  60%|██████    | 3906/6470 [44:55<30:36,  1.40it/s]
4|3907|Loss: 0.8607970476150513:  60%|██████    | 3907/6470 [44:55<30:03,  1.42it/s]
4|3908|Loss: 0.7687088251113892:  60%|██████    | 3907/6470 [44:55<30:03,  1.42it/s]
4|3908|Loss: 0.7687088251113892:  60%|██████    | 3908/6470 [44:56<31:34,  1.35it/s]
4|3909|Loss: 0.6639057993888855:  60%|██████    | 3908/6470 [44:56<31:34,  1.35it/s]
4|3909|Loss: 0.6639057993888855:  60%|██████    | 3909/6470 [44:56<30:54,  1.38it/s]
4|3910|Loss: 0.8263819813728333:  60%|██████    | 3909/6470 [44:57<30:54,  1.38it/s]
4|3910|Loss: 0.8263819813728333:  60%|██████    | 3910/6470 [44:57<30:27,  1.40it/s]
4|3911|Loss: 0.9269905090332031:  60%|██████    | 3910/6470 [44:57<30:27,  1.40it/s]
4|3911|Loss: 0.9269905090332031:  60%|██████    | 3911/6470 [44:58<31:15,  1.36it/s]
4|3912|Loss: 0.9145866632461548:  60%|██████    | 3911/6470 [44:58<31:15,  1.36it/s]
4|3912|Loss: 0.9145866632461548:  60%|██████    | 3912/6470 [44:59<32:02,  1.33it/s]
4|3913|Loss: 0.820926308631897:  60%|██████    | 3912/6470 [44:59<32:02,  1.33it/s] 
4|3913|Loss: 0.820926308631897:  60%|██████    | 3913/6470 [44:59<32:01,  1.33it/s]
4|3914|Loss: 0.830262303352356:  60%|██████    | 3913/6470 [45:00<32:01,  1.33it/s]
4|3914|Loss: 0.830262303352356:  60%|██████    | 3914/6470 [45:00<33:04,  1.29it/s]
4|3915|Loss: 0.9381231069564819:  60%|██████    | 3914/6470 [45:01<33:04,  1.29it/s]
4|3915|Loss: 0.9381231069564819:  61%|██████    | 3915/6470 [45:01<31:18,  1.36it/s]
4|3916|Loss: 0.961631178855896:  61%|██████    | 3915/6470 [45:01<31:18,  1.36it/s] 
4|3916|Loss: 0.961631178855896:  61%|██████    | 3916/6470 [45:01<28:46,  1.48it/s]
4|3917|Loss: 0.8102588653564453:  61%|██████    | 3916/6470 [45:02<28:46,  1.48it/s]
4|3917|Loss: 0.8102588653564453:  61%|██████    | 3917/6470 [45:02<29:53,  1.42it/s]
4|3918|Loss: 0.8407054543495178:  61%|██████    | 3917/6470 [45:03<29:53,  1.42it/s]
4|3918|Loss: 0.8407054543495178:  61%|██████    | 3918/6470 [45:03<31:04,  1.37it/s]
4|3919|Loss: 0.862087607383728:  61%|██████    | 3918/6470 [45:03<31:04,  1.37it/s] 
4|3919|Loss: 0.862087607383728:  61%|██████    | 3919/6470 [45:04<29:43,  1.43it/s]
4|3920|Loss: 0.9430422186851501:  61%|██████    | 3919/6470 [45:04<29:43,  1.43it/s]
4|3920|Loss: 0.9430422186851501:  61%|██████    | 3920/6470 [45:04<31:15,  1.36it/s]
4|3921|Loss: 0.8968868255615234:  61%|██████    | 3920/6470 [45:05<31:15,  1.36it/s]
4|3921|Loss: 0.8968868255615234:  61%|██████    | 3921/6470 [45:05<29:39,  1.43it/s]
4|3922|Loss: 0.8515955209732056:  61%|██████    | 3921/6470 [45:05<29:39,  1.43it/s]
4|3922|Loss: 0.8515955209732056:  61%|██████    | 3922/6470 [45:06<28:34,  1.49it/s]
4|3923|Loss: 0.8847337365150452:  61%|██████    | 3922/6470 [45:06<28:34,  1.49it/s]
4|3923|Loss: 0.8847337365150452:  61%|██████    | 3923/6470 [45:06<29:17,  1.45it/s]
4|3924|Loss: 0.8981786370277405:  61%|██████    | 3923/6470 [45:07<29:17,  1.45it/s]
4|3924|Loss: 0.8981786370277405:  61%|██████    | 3924/6470 [45:07<30:56,  1.37it/s]
4|3925|Loss: 0.8767763376235962:  61%|██████    | 3924/6470 [45:08<30:56,  1.37it/s]
4|3925|Loss: 0.8767763376235962:  61%|██████    | 3925/6470 [45:08<30:05,  1.41it/s]
4|3926|Loss: 0.845580518245697:  61%|██████    | 3925/6470 [45:08<30:05,  1.41it/s] 
4|3926|Loss: 0.845580518245697:  61%|██████    | 3926/6470 [45:08<27:58,  1.52it/s]
4|3927|Loss: 0.888813316822052:  61%|██████    | 3926/6470 [45:09<27:58,  1.52it/s]
4|3927|Loss: 0.888813316822052:  61%|██████    | 3927/6470 [45:09<28:05,  1.51it/s]
4|3928|Loss: 0.8663212060928345:  61%|██████    | 3927/6470 [45:09<28:05,  1.51it/s]
4|3928|Loss: 0.8663212060928345:  61%|██████    | 3928/6470 [45:10<27:52,  1.52it/s]
4|3929|Loss: 0.6221962571144104:  61%|██████    | 3928/6470 [45:10<27:52,  1.52it/s]
4|3929|Loss: 0.6221962571144104:  61%|██████    | 3929/6470 [45:11<31:14,  1.36it/s]
4|3930|Loss: 1.0397334098815918:  61%|██████    | 3929/6470 [45:11<31:14,  1.36it/s]
4|3930|Loss: 1.0397334098815918:  61%|██████    | 3930/6470 [45:11<32:06,  1.32it/s]
4|3931|Loss: 0.9274260997772217:  61%|██████    | 3930/6470 [45:12<32:06,  1.32it/s]
4|3931|Loss: 0.9274260997772217:  61%|██████    | 3931/6470 [45:12<30:16,  1.40it/s]
4|3932|Loss: 0.9933461546897888:  61%|██████    | 3931/6470 [45:12<30:16,  1.40it/s]
4|3932|Loss: 0.9933461546897888:  61%|██████    | 3932/6470 [45:13<29:12,  1.45it/s]
4|3933|Loss: 0.8965539336204529:  61%|██████    | 3932/6470 [45:13<29:12,  1.45it/s]
4|3933|Loss: 0.8965539336204529:  61%|██████    | 3933/6470 [45:13<28:35,  1.48it/s]
4|3934|Loss: 0.868899405002594:  61%|██████    | 3933/6470 [45:14<28:35,  1.48it/s] 
4|3934|Loss: 0.868899405002594:  61%|██████    | 3934/6470 [45:14<28:07,  1.50it/s]
4|3935|Loss: 1.022264003753662:  61%|██████    | 3934/6470 [45:14<28:07,  1.50it/s]
4|3935|Loss: 1.022264003753662:  61%|██████    | 3935/6470 [45:15<28:55,  1.46it/s]
4|3936|Loss: 0.9166377782821655:  61%|██████    | 3935/6470 [45:15<28:55,  1.46it/s]
4|3936|Loss: 0.9166377782821655:  61%|██████    | 3936/6470 [45:15<30:11,  1.40it/s]
4|3937|Loss: 0.8623753190040588:  61%|██████    | 3936/6470 [45:16<30:11,  1.40it/s]
4|3937|Loss: 0.8623753190040588:  61%|██████    | 3937/6470 [45:16<31:15,  1.35it/s]
4|3938|Loss: 0.7047150731086731:  61%|██████    | 3937/6470 [45:17<31:15,  1.35it/s]
4|3938|Loss: 0.7047150731086731:  61%|██████    | 3938/6470 [45:17<31:15,  1.35it/s]
4|3939|Loss: 0.907598078250885:  61%|██████    | 3938/6470 [45:17<31:15,  1.35it/s] 
4|3939|Loss: 0.907598078250885:  61%|██████    | 3939/6470 [45:18<31:05,  1.36it/s]
4|3940|Loss: 0.7979221940040588:  61%|██████    | 3939/6470 [45:18<31:05,  1.36it/s]
4|3940|Loss: 0.7979221940040588:  61%|██████    | 3940/6470 [45:18<30:41,  1.37it/s]
4|3941|Loss: 0.8277887105941772:  61%|██████    | 3940/6470 [45:19<30:41,  1.37it/s]
4|3941|Loss: 0.8277887105941772:  61%|██████    | 3941/6470 [45:19<29:45,  1.42it/s]
4|3942|Loss: 0.7781174778938293:  61%|██████    | 3941/6470 [45:19<29:45,  1.42it/s]
4|3942|Loss: 0.7781174778938293:  61%|██████    | 3942/6470 [45:20<29:35,  1.42it/s]
4|3943|Loss: 0.8395277261734009:  61%|██████    | 3942/6470 [45:20<29:35,  1.42it/s]
4|3943|Loss: 0.8395277261734009:  61%|██████    | 3943/6470 [45:21<30:06,  1.40it/s]
4|3944|Loss: 1.0459563732147217:  61%|██████    | 3943/6470 [45:21<30:06,  1.40it/s]
4|3944|Loss: 1.0459563732147217:  61%|██████    | 3944/6470 [45:21<30:30,  1.38it/s]
4|3945|Loss: 0.9750518798828125:  61%|██████    | 3944/6470 [45:22<30:30,  1.38it/s]
4|3945|Loss: 0.9750518798828125:  61%|██████    | 3945/6470 [45:22<30:50,  1.36it/s]
4|3946|Loss: 0.7210524678230286:  61%|██████    | 3945/6470 [45:22<30:50,  1.36it/s]
4|3946|Loss: 0.7210524678230286:  61%|██████    | 3946/6470 [45:23<31:08,  1.35it/s]
4|3947|Loss: 0.7652103304862976:  61%|██████    | 3946/6470 [45:23<31:08,  1.35it/s]
4|3947|Loss: 0.7652103304862976:  61%|██████    | 3947/6470 [45:24<32:45,  1.28it/s]
4|3948|Loss: 0.9233611226081848:  61%|██████    | 3947/6470 [45:24<32:45,  1.28it/s]
4|3948|Loss: 0.9233611226081848:  61%|██████    | 3948/6470 [45:25<33:40,  1.25it/s]
4|3949|Loss: 0.7792701125144958:  61%|██████    | 3948/6470 [45:25<33:40,  1.25it/s]
4|3949|Loss: 0.7792701125144958:  61%|██████    | 3949/6470 [45:25<32:11,  1.31it/s]
4|3950|Loss: 0.6709499359130859:  61%|██████    | 3949/6470 [45:26<32:11,  1.31it/s]
4|3950|Loss: 0.6709499359130859:  61%|██████    | 3950/6470 [45:26<31:12,  1.35it/s]
4|3951|Loss: 0.7671945095062256:  61%|██████    | 3950/6470 [45:26<31:12,  1.35it/s]
4|3951|Loss: 0.7671945095062256:  61%|██████    | 3951/6470 [45:26<29:11,  1.44it/s]
4|3952|Loss: 0.8384528756141663:  61%|██████    | 3951/6470 [45:27<29:11,  1.44it/s]
4|3952|Loss: 0.8384528756141663:  61%|██████    | 3952/6470 [45:27<27:47,  1.51it/s]
4|3953|Loss: 0.9373489618301392:  61%|██████    | 3952/6470 [45:27<27:47,  1.51it/s]
4|3953|Loss: 0.9373489618301392:  61%|██████    | 3953/6470 [45:28<26:46,  1.57it/s]
4|3954|Loss: 0.717210054397583:  61%|██████    | 3953/6470 [45:28<26:46,  1.57it/s] 
4|3954|Loss: 0.717210054397583:  61%|██████    | 3954/6470 [45:28<29:00,  1.45it/s]
4|3955|Loss: 0.9207310676574707:  61%|██████    | 3954/6470 [45:29<29:00,  1.45it/s]
4|3955|Loss: 0.9207310676574707:  61%|██████    | 3955/6470 [45:29<28:21,  1.48it/s]
4|3956|Loss: 0.784231424331665:  61%|██████    | 3955/6470 [45:29<28:21,  1.48it/s] 
4|3956|Loss: 0.784231424331665:  61%|██████    | 3956/6470 [45:30<29:16,  1.43it/s]
4|3957|Loss: 0.8263221979141235:  61%|██████    | 3956/6470 [45:30<29:16,  1.43it/s]
4|3957|Loss: 0.8263221979141235:  61%|██████    | 3957/6470 [45:30<27:38,  1.52it/s]
4|3958|Loss: 0.7787681818008423:  61%|██████    | 3957/6470 [45:31<27:38,  1.52it/s]
4|3958|Loss: 0.7787681818008423:  61%|██████    | 3958/6470 [45:31<27:33,  1.52it/s]
4|3959|Loss: 1.015121340751648:  61%|██████    | 3958/6470 [45:31<27:33,  1.52it/s] 
4|3959|Loss: 1.015121340751648:  61%|██████    | 3959/6470 [45:32<28:58,  1.44it/s]
4|3960|Loss: 0.8045918345451355:  61%|██████    | 3959/6470 [45:32<28:58,  1.44it/s]
4|3960|Loss: 0.8045918345451355:  61%|██████    | 3960/6470 [45:33<31:14,  1.34it/s]
4|3961|Loss: 0.6982255578041077:  61%|██████    | 3960/6470 [45:33<31:14,  1.34it/s]
4|3961|Loss: 0.6982255578041077:  61%|██████    | 3961/6470 [45:33<31:16,  1.34it/s]
4|3962|Loss: 0.8020386695861816:  61%|██████    | 3961/6470 [45:34<31:16,  1.34it/s]
4|3962|Loss: 0.8020386695861816:  61%|██████    | 3962/6470 [45:34<30:43,  1.36it/s]
4|3963|Loss: 0.9105920195579529:  61%|██████    | 3962/6470 [45:34<30:43,  1.36it/s]
4|3963|Loss: 0.9105920195579529:  61%|██████▏   | 3963/6470 [45:35<28:07,  1.49it/s]
4|3964|Loss: 0.8839805722236633:  61%|██████▏   | 3963/6470 [45:35<28:07,  1.49it/s]
4|3964|Loss: 0.8839805722236633:  61%|██████▏   | 3964/6470 [45:35<24:32,  1.70it/s]
4|3965|Loss: 0.8408362865447998:  61%|██████▏   | 3964/6470 [45:35<24:32,  1.70it/s]
4|3965|Loss: 0.8408362865447998:  61%|██████▏   | 3965/6470 [45:36<24:06,  1.73it/s]
4|3966|Loss: 0.8021918535232544:  61%|██████▏   | 3965/6470 [45:36<24:06,  1.73it/s]
4|3966|Loss: 0.8021918535232544:  61%|██████▏   | 3966/6470 [45:36<26:41,  1.56it/s]
4|3967|Loss: 0.9399999976158142:  61%|██████▏   | 3966/6470 [45:37<26:41,  1.56it/s]
4|3967|Loss: 0.9399999976158142:  61%|██████▏   | 3967/6470 [45:37<27:54,  1.50it/s]
4|3968|Loss: 0.8992472290992737:  61%|██████▏   | 3967/6470 [45:38<27:54,  1.50it/s]
4|3968|Loss: 0.8992472290992737:  61%|██████▏   | 3968/6470 [45:38<27:31,  1.52it/s]
4|3969|Loss: 0.7108038663864136:  61%|██████▏   | 3968/6470 [45:38<27:31,  1.52it/s]
4|3969|Loss: 0.7108038663864136:  61%|██████▏   | 3969/6470 [45:38<24:24,  1.71it/s]
4|3970|Loss: 0.8403446674346924:  61%|██████▏   | 3969/6470 [45:39<24:24,  1.71it/s]
4|3970|Loss: 0.8403446674346924:  61%|██████▏   | 3970/6470 [45:39<29:48,  1.40it/s]
4|3971|Loss: 0.838613748550415:  61%|██████▏   | 3970/6470 [45:40<29:48,  1.40it/s] 
4|3971|Loss: 0.838613748550415:  61%|██████▏   | 3971/6470 [45:40<29:40,  1.40it/s]
4|3972|Loss: 0.8690195083618164:  61%|██████▏   | 3971/6470 [45:40<29:40,  1.40it/s]
4|3972|Loss: 0.8690195083618164:  61%|██████▏   | 3972/6470 [45:40<27:46,  1.50it/s]
4|3973|Loss: 0.7407141327857971:  61%|██████▏   | 3972/6470 [45:41<27:46,  1.50it/s]
4|3973|Loss: 0.7407141327857971:  61%|██████▏   | 3973/6470 [45:41<28:12,  1.48it/s]
4|3974|Loss: 0.6896393895149231:  61%|██████▏   | 3973/6470 [45:42<28:12,  1.48it/s]
4|3974|Loss: 0.6896393895149231:  61%|██████▏   | 3974/6470 [45:42<27:21,  1.52it/s]
4|3975|Loss: 0.8674529790878296:  61%|██████▏   | 3974/6470 [45:42<27:21,  1.52it/s]
4|3975|Loss: 0.8674529790878296:  61%|██████▏   | 3975/6470 [45:42<26:43,  1.56it/s]
4|3976|Loss: 0.8231500387191772:  61%|██████▏   | 3975/6470 [45:43<26:43,  1.56it/s]
4|3976|Loss: 0.8231500387191772:  61%|██████▏   | 3976/6470 [45:43<27:40,  1.50it/s]
4|3977|Loss: 0.9171624183654785:  61%|██████▏   | 3976/6470 [45:44<27:40,  1.50it/s]
4|3977|Loss: 0.9171624183654785:  61%|██████▏   | 3977/6470 [45:44<28:18,  1.47it/s]
4|3978|Loss: 0.759689211845398:  61%|██████▏   | 3977/6470 [45:44<28:18,  1.47it/s] 
4|3978|Loss: 0.759689211845398:  61%|██████▏   | 3978/6470 [45:45<28:31,  1.46it/s]
4|3979|Loss: 0.8194977641105652:  61%|██████▏   | 3978/6470 [45:45<28:31,  1.46it/s]
4|3979|Loss: 0.8194977641105652:  61%|██████▏   | 3979/6470 [45:45<28:38,  1.45it/s]
4|3980|Loss: 0.7720625400543213:  61%|██████▏   | 3979/6470 [45:46<28:38,  1.45it/s]
4|3980|Loss: 0.7720625400543213:  62%|██████▏   | 3980/6470 [45:46<26:14,  1.58it/s]
4|3981|Loss: 0.7311335802078247:  62%|██████▏   | 3980/6470 [45:46<26:14,  1.58it/s]
4|3981|Loss: 0.7311335802078247:  62%|██████▏   | 3981/6470 [45:46<26:55,  1.54it/s]
4|3982|Loss: 0.8059977293014526:  62%|██████▏   | 3981/6470 [45:47<26:55,  1.54it/s]
4|3982|Loss: 0.8059977293014526:  62%|██████▏   | 3982/6470 [45:47<26:17,  1.58it/s]
4|3983|Loss: 0.9800171852111816:  62%|██████▏   | 3982/6470 [45:47<26:17,  1.58it/s]
4|3983|Loss: 0.9800171852111816:  62%|██████▏   | 3983/6470 [45:48<28:01,  1.48it/s]
4|3984|Loss: 0.9264430999755859:  62%|██████▏   | 3983/6470 [45:48<28:01,  1.48it/s]
4|3984|Loss: 0.9264430999755859:  62%|██████▏   | 3984/6470 [45:48<27:29,  1.51it/s]
4|3985|Loss: 0.6865536570549011:  62%|██████▏   | 3984/6470 [45:49<27:29,  1.51it/s]
4|3985|Loss: 0.6865536570549011:  62%|██████▏   | 3985/6470 [45:49<25:30,  1.62it/s]
4|3986|Loss: 0.7310142517089844:  62%|██████▏   | 3985/6470 [45:49<25:30,  1.62it/s]
4|3986|Loss: 0.7310142517089844:  62%|██████▏   | 3986/6470 [45:50<25:07,  1.65it/s]
4|3987|Loss: 0.950942873954773:  62%|██████▏   | 3986/6470 [45:50<25:07,  1.65it/s] 
4|3987|Loss: 0.950942873954773:  62%|██████▏   | 3987/6470 [45:50<26:33,  1.56it/s]
4|3988|Loss: 0.9994721412658691:  62%|██████▏   | 3987/6470 [45:51<26:33,  1.56it/s]
4|3988|Loss: 0.9994721412658691:  62%|██████▏   | 3988/6470 [45:51<28:49,  1.44it/s]
4|3989|Loss: 0.9909621477127075:  62%|██████▏   | 3988/6470 [45:51<28:49,  1.44it/s]
4|3989|Loss: 0.9909621477127075:  62%|██████▏   | 3989/6470 [45:52<28:41,  1.44it/s]
4|3990|Loss: 0.9223608374595642:  62%|██████▏   | 3989/6470 [45:52<28:41,  1.44it/s]
4|3990|Loss: 0.9223608374595642:  62%|██████▏   | 3990/6470 [45:52<28:00,  1.48it/s]
4|3991|Loss: 0.9197975993156433:  62%|██████▏   | 3990/6470 [45:53<28:00,  1.48it/s]
4|3991|Loss: 0.9197975993156433:  62%|██████▏   | 3991/6470 [45:53<27:22,  1.51it/s]
4|3992|Loss: 0.8309251070022583:  62%|██████▏   | 3991/6470 [45:53<27:22,  1.51it/s]
4|3992|Loss: 0.8309251070022583:  62%|██████▏   | 3992/6470 [45:54<27:56,  1.48it/s]
4|3993|Loss: 1.1209460496902466:  62%|██████▏   | 3992/6470 [45:54<27:56,  1.48it/s]
4|3993|Loss: 1.1209460496902466:  62%|██████▏   | 3993/6470 [45:54<27:48,  1.48it/s]
4|3994|Loss: 0.7286131381988525:  62%|██████▏   | 3993/6470 [45:55<27:48,  1.48it/s]
4|3994|Loss: 0.7286131381988525:  62%|██████▏   | 3994/6470 [45:55<28:19,  1.46it/s]
4|3995|Loss: 0.8749399185180664:  62%|██████▏   | 3994/6470 [45:56<28:19,  1.46it/s]
4|3995|Loss: 0.8749399185180664:  62%|██████▏   | 3995/6470 [45:56<29:36,  1.39it/s]
4|3996|Loss: 0.8949646949768066:  62%|██████▏   | 3995/6470 [45:56<29:36,  1.39it/s]
4|3996|Loss: 0.8949646949768066:  62%|██████▏   | 3996/6470 [45:57<28:50,  1.43it/s]
4|3997|Loss: 0.7909115552902222:  62%|██████▏   | 3996/6470 [45:57<28:50,  1.43it/s]
4|3997|Loss: 0.7909115552902222:  62%|██████▏   | 3997/6470 [45:57<26:08,  1.58it/s]
4|3998|Loss: 0.9297264814376831:  62%|██████▏   | 3997/6470 [45:57<26:08,  1.58it/s]
4|3998|Loss: 0.9297264814376831:  62%|██████▏   | 3998/6470 [45:58<27:45,  1.48it/s]
4|3999|Loss: 0.8610674738883972:  62%|██████▏   | 3998/6470 [45:58<27:45,  1.48it/s]
4|3999|Loss: 0.8610674738883972:  62%|██████▏   | 3999/6470 [45:59<28:56,  1.42it/s]
4|4000|Loss: 0.892440676689148:  62%|██████▏   | 3999/6470 [45:59<28:56,  1.42it/s] 
4|4000|Loss: 0.892440676689148:  62%|██████▏   | 4000/6470 [45:59<28:16,  1.46it/s]
4|4001|Loss: 0.8495218753814697:  62%|██████▏   | 4000/6470 [46:00<28:16,  1.46it/s]
4|4001|Loss: 0.8495218753814697:  62%|██████▏   | 4001/6470 [46:00<27:26,  1.50it/s]
4|4002|Loss: 0.77555912733078:  62%|██████▏   | 4001/6470 [46:00<27:26,  1.50it/s]  
4|4002|Loss: 0.77555912733078:  62%|██████▏   | 4002/6470 [46:00<27:03,  1.52it/s]
4|4003|Loss: 0.7103727459907532:  62%|██████▏   | 4002/6470 [46:01<27:03,  1.52it/s]
4|4003|Loss: 0.7103727459907532:  62%|██████▏   | 4003/6470 [46:01<26:39,  1.54it/s]
4|4004|Loss: 0.7813448905944824:  62%|██████▏   | 4003/6470 [46:01<26:39,  1.54it/s]
4|4004|Loss: 0.7813448905944824:  62%|██████▏   | 4004/6470 [46:02<26:31,  1.55it/s]
4|4005|Loss: 0.8269938230514526:  62%|██████▏   | 4004/6470 [46:02<26:31,  1.55it/s]
4|4005|Loss: 0.8269938230514526:  62%|██████▏   | 4005/6470 [46:02<26:15,  1.56it/s]
4|4006|Loss: 0.872194766998291:  62%|██████▏   | 4005/6470 [46:03<26:15,  1.56it/s] 
4|4006|Loss: 0.872194766998291:  62%|██████▏   | 4006/6470 [46:03<28:05,  1.46it/s]
4|4007|Loss: 0.8466739058494568:  62%|██████▏   | 4006/6470 [46:04<28:05,  1.46it/s]
4|4007|Loss: 0.8466739058494568:  62%|██████▏   | 4007/6470 [46:04<27:49,  1.48it/s]
4|4008|Loss: 0.791348397731781:  62%|██████▏   | 4007/6470 [46:04<27:49,  1.48it/s] 
4|4008|Loss: 0.791348397731781:  62%|██████▏   | 4008/6470 [46:04<26:11,  1.57it/s]
4|4009|Loss: 0.9551416039466858:  62%|██████▏   | 4008/6470 [46:05<26:11,  1.57it/s]
4|4009|Loss: 0.9551416039466858:  62%|██████▏   | 4009/6470 [46:05<26:47,  1.53it/s]
4|4010|Loss: 0.8071557879447937:  62%|██████▏   | 4009/6470 [46:05<26:47,  1.53it/s]
4|4010|Loss: 0.8071557879447937:  62%|██████▏   | 4010/6470 [46:06<24:17,  1.69it/s]
4|4011|Loss: 0.7323261499404907:  62%|██████▏   | 4010/6470 [46:06<24:17,  1.69it/s]
4|4011|Loss: 0.7323261499404907:  62%|██████▏   | 4011/6470 [46:06<23:32,  1.74it/s]
4|4012|Loss: 0.8467714190483093:  62%|██████▏   | 4011/6470 [46:06<23:32,  1.74it/s]
4|4012|Loss: 0.8467714190483093:  62%|██████▏   | 4012/6470 [46:07<24:00,  1.71it/s]
4|4013|Loss: 0.7217314839363098:  62%|██████▏   | 4012/6470 [46:07<24:00,  1.71it/s]
4|4013|Loss: 0.7217314839363098:  62%|██████▏   | 4013/6470 [46:07<24:22,  1.68it/s]
4|4014|Loss: 0.7354925274848938:  62%|██████▏   | 4013/6470 [46:08<24:22,  1.68it/s]
4|4014|Loss: 0.7354925274848938:  62%|██████▏   | 4014/6470 [46:08<24:15,  1.69it/s]
4|4015|Loss: 0.8161743879318237:  62%|██████▏   | 4014/6470 [46:08<24:15,  1.69it/s]
4|4015|Loss: 0.8161743879318237:  62%|██████▏   | 4015/6470 [46:09<26:57,  1.52it/s]
4|4016|Loss: 0.8919298648834229:  62%|██████▏   | 4015/6470 [46:09<26:57,  1.52it/s]
4|4016|Loss: 0.8919298648834229:  62%|██████▏   | 4016/6470 [46:09<26:06,  1.57it/s]
4|4017|Loss: 0.9804605841636658:  62%|██████▏   | 4016/6470 [46:10<26:06,  1.57it/s]
4|4017|Loss: 0.9804605841636658:  62%|██████▏   | 4017/6470 [46:10<25:34,  1.60it/s]
4|4018|Loss: 0.9322944283485413:  62%|██████▏   | 4017/6470 [46:10<25:34,  1.60it/s]
4|4018|Loss: 0.9322944283485413:  62%|██████▏   | 4018/6470 [46:11<25:43,  1.59it/s]
4|4019|Loss: 0.9849165081977844:  62%|██████▏   | 4018/6470 [46:11<25:43,  1.59it/s]
4|4019|Loss: 0.9849165081977844:  62%|██████▏   | 4019/6470 [46:11<25:40,  1.59it/s]
4|4020|Loss: 0.9099736213684082:  62%|██████▏   | 4019/6470 [46:12<25:40,  1.59it/s]
4|4020|Loss: 0.9099736213684082:  62%|██████▏   | 4020/6470 [46:12<26:22,  1.55it/s]
4|4021|Loss: 0.8786588907241821:  62%|██████▏   | 4020/6470 [46:12<26:22,  1.55it/s]
4|4021|Loss: 0.8786588907241821:  62%|██████▏   | 4021/6470 [46:13<28:33,  1.43it/s]
4|4022|Loss: 0.9092994332313538:  62%|██████▏   | 4021/6470 [46:13<28:33,  1.43it/s]
4|4022|Loss: 0.9092994332313538:  62%|██████▏   | 4022/6470 [46:14<30:32,  1.34it/s]
4|4023|Loss: 0.9541762471199036:  62%|██████▏   | 4022/6470 [46:14<30:32,  1.34it/s]
4|4023|Loss: 0.9541762471199036:  62%|██████▏   | 4023/6470 [46:14<30:44,  1.33it/s]
4|4024|Loss: 0.7354934811592102:  62%|██████▏   | 4023/6470 [46:15<30:44,  1.33it/s]
4|4024|Loss: 0.7354934811592102:  62%|██████▏   | 4024/6470 [46:15<29:06,  1.40it/s]
4|4025|Loss: 0.7428290843963623:  62%|██████▏   | 4024/6470 [46:15<29:06,  1.40it/s]
4|4025|Loss: 0.7428290843963623:  62%|██████▏   | 4025/6470 [46:16<28:40,  1.42it/s]
4|4026|Loss: 0.8258658647537231:  62%|██████▏   | 4025/6470 [46:16<28:40,  1.42it/s]
4|4026|Loss: 0.8258658647537231:  62%|██████▏   | 4026/6470 [46:16<29:57,  1.36it/s]
4|4027|Loss: 0.7712565064430237:  62%|██████▏   | 4026/6470 [46:17<29:57,  1.36it/s]
4|4027|Loss: 0.7712565064430237:  62%|██████▏   | 4027/6470 [46:17<29:21,  1.39it/s]
4|4028|Loss: 0.8274399638175964:  62%|██████▏   | 4027/6470 [46:17<29:21,  1.39it/s]
4|4028|Loss: 0.8274399638175964:  62%|██████▏   | 4028/6470 [46:18<28:20,  1.44it/s]
4|4029|Loss: 0.7570286989212036:  62%|██████▏   | 4028/6470 [46:18<28:20,  1.44it/s]
4|4029|Loss: 0.7570286989212036:  62%|██████▏   | 4029/6470 [46:19<29:26,  1.38it/s]
4|4030|Loss: 0.8775110244750977:  62%|██████▏   | 4029/6470 [46:19<29:26,  1.38it/s]
4|4030|Loss: 0.8775110244750977:  62%|██████▏   | 4030/6470 [46:19<29:47,  1.36it/s]
4|4031|Loss: 0.9592087268829346:  62%|██████▏   | 4030/6470 [46:20<29:47,  1.36it/s]
4|4031|Loss: 0.9592087268829346:  62%|██████▏   | 4031/6470 [46:20<27:02,  1.50it/s]
4|4032|Loss: 0.8518093824386597:  62%|██████▏   | 4031/6470 [46:20<27:02,  1.50it/s]
4|4032|Loss: 0.8518093824386597:  62%|██████▏   | 4032/6470 [46:20<27:07,  1.50it/s]
4|4033|Loss: 0.870681643486023:  62%|██████▏   | 4032/6470 [46:21<27:07,  1.50it/s] 
4|4033|Loss: 0.870681643486023:  62%|██████▏   | 4033/6470 [46:21<26:29,  1.53it/s]
4|4034|Loss: 0.8096957206726074:  62%|██████▏   | 4033/6470 [46:21<26:29,  1.53it/s]
4|4034|Loss: 0.8096957206726074:  62%|██████▏   | 4034/6470 [46:22<26:49,  1.51it/s]
4|4035|Loss: 0.8245786428451538:  62%|██████▏   | 4034/6470 [46:22<26:49,  1.51it/s]
4|4035|Loss: 0.8245786428451538:  62%|██████▏   | 4035/6470 [46:22<26:45,  1.52it/s]
4|4036|Loss: 0.8685106635093689:  62%|██████▏   | 4035/6470 [46:23<26:45,  1.52it/s]
4|4036|Loss: 0.8685106635093689:  62%|██████▏   | 4036/6470 [46:23<25:23,  1.60it/s]
4|4037|Loss: 0.7501534223556519:  62%|██████▏   | 4036/6470 [46:23<25:23,  1.60it/s]
4|4037|Loss: 0.7501534223556519:  62%|██████▏   | 4037/6470 [46:23<24:11,  1.68it/s]
4|4038|Loss: 0.9099655151367188:  62%|██████▏   | 4037/6470 [46:24<24:11,  1.68it/s]
4|4038|Loss: 0.9099655151367188:  62%|██████▏   | 4038/6470 [46:24<25:03,  1.62it/s]
4|4039|Loss: 0.9949376583099365:  62%|██████▏   | 4038/6470 [46:25<25:03,  1.62it/s]
4|4039|Loss: 0.9949376583099365:  62%|██████▏   | 4039/6470 [46:25<24:53,  1.63it/s]
4|4040|Loss: 0.9369961619377136:  62%|██████▏   | 4039/6470 [46:25<24:53,  1.63it/s]
4|4040|Loss: 0.9369961619377136:  62%|██████▏   | 4040/6470 [46:25<26:05,  1.55it/s]
4|4041|Loss: 0.8808861970901489:  62%|██████▏   | 4040/6470 [46:26<26:05,  1.55it/s]
4|4041|Loss: 0.8808861970901489:  62%|██████▏   | 4041/6470 [46:26<27:00,  1.50it/s]
4|4042|Loss: 0.8770684003829956:  62%|██████▏   | 4041/6470 [46:27<27:00,  1.50it/s]
4|4042|Loss: 0.8770684003829956:  62%|██████▏   | 4042/6470 [46:27<27:45,  1.46it/s]
4|4043|Loss: 0.8786738514900208:  62%|██████▏   | 4042/6470 [46:27<27:45,  1.46it/s]
4|4043|Loss: 0.8786738514900208:  62%|██████▏   | 4043/6470 [46:28<28:27,  1.42it/s]
4|4044|Loss: 0.824293851852417:  62%|██████▏   | 4043/6470 [46:28<28:27,  1.42it/s] 
4|4044|Loss: 0.824293851852417:  63%|██████▎   | 4044/6470 [46:28<27:31,  1.47it/s]
4|4045|Loss: 0.9349974989891052:  63%|██████▎   | 4044/6470 [46:29<27:31,  1.47it/s]
4|4045|Loss: 0.9349974989891052:  63%|██████▎   | 4045/6470 [46:29<27:23,  1.48it/s]
4|4046|Loss: 0.8082659244537354:  63%|██████▎   | 4045/6470 [46:29<27:23,  1.48it/s]
4|4046|Loss: 0.8082659244537354:  63%|██████▎   | 4046/6470 [46:30<30:09,  1.34it/s]
4|4047|Loss: 0.9736841320991516:  63%|██████▎   | 4046/6470 [46:30<30:09,  1.34it/s]
4|4047|Loss: 0.9736841320991516:  63%|██████▎   | 4047/6470 [46:31<30:48,  1.31it/s]
4|4048|Loss: 0.7807070016860962:  63%|██████▎   | 4047/6470 [46:31<30:48,  1.31it/s]
4|4048|Loss: 0.7807070016860962:  63%|██████▎   | 4048/6470 [46:31<28:42,  1.41it/s]
4|4049|Loss: 0.9201956987380981:  63%|██████▎   | 4048/6470 [46:32<28:42,  1.41it/s]
4|4049|Loss: 0.9201956987380981:  63%|██████▎   | 4049/6470 [46:32<27:27,  1.47it/s]
4|4050|Loss: 0.7701704502105713:  63%|██████▎   | 4049/6470 [46:32<27:27,  1.47it/s]
4|4050|Loss: 0.7701704502105713:  63%|██████▎   | 4050/6470 [46:33<27:49,  1.45it/s]
4|4051|Loss: 0.8985232710838318:  63%|██████▎   | 4050/6470 [46:33<27:49,  1.45it/s]
4|4051|Loss: 0.8985232710838318:  63%|██████▎   | 4051/6470 [46:33<26:52,  1.50it/s]
4|4052|Loss: 0.7050785422325134:  63%|██████▎   | 4051/6470 [46:34<26:52,  1.50it/s]
4|4052|Loss: 0.7050785422325134:  63%|██████▎   | 4052/6470 [46:34<26:13,  1.54it/s]
4|4053|Loss: 0.9360411167144775:  63%|██████▎   | 4052/6470 [46:34<26:13,  1.54it/s]
4|4053|Loss: 0.9360411167144775:  63%|██████▎   | 4053/6470 [46:35<28:09,  1.43it/s]
4|4054|Loss: 0.6843005418777466:  63%|██████▎   | 4053/6470 [46:35<28:09,  1.43it/s]
4|4054|Loss: 0.6843005418777466:  63%|██████▎   | 4054/6470 [46:35<29:06,  1.38it/s]
4|4055|Loss: 0.8337646722793579:  63%|██████▎   | 4054/6470 [46:36<29:06,  1.38it/s]
4|4055|Loss: 0.8337646722793579:  63%|██████▎   | 4055/6470 [46:36<27:18,  1.47it/s]
4|4056|Loss: 0.850727379322052:  63%|██████▎   | 4055/6470 [46:36<27:18,  1.47it/s] 
4|4056|Loss: 0.850727379322052:  63%|██████▎   | 4056/6470 [46:37<26:20,  1.53it/s]
4|4057|Loss: 0.8007107377052307:  63%|██████▎   | 4056/6470 [46:37<26:20,  1.53it/s]
4|4057|Loss: 0.8007107377052307:  63%|██████▎   | 4057/6470 [46:37<26:38,  1.51it/s]
4|4058|Loss: 0.8453443050384521:  63%|██████▎   | 4057/6470 [46:38<26:38,  1.51it/s]
4|4058|Loss: 0.8453443050384521:  63%|██████▎   | 4058/6470 [46:38<27:34,  1.46it/s]
4|4059|Loss: 0.8198257088661194:  63%|██████▎   | 4058/6470 [46:38<27:34,  1.46it/s]
4|4059|Loss: 0.8198257088661194:  63%|██████▎   | 4059/6470 [46:39<28:46,  1.40it/s]
4|4060|Loss: 1.0467448234558105:  63%|██████▎   | 4059/6470 [46:39<28:46,  1.40it/s]
4|4060|Loss: 1.0467448234558105:  63%|██████▎   | 4060/6470 [46:39<27:10,  1.48it/s]
4|4061|Loss: 0.773165762424469:  63%|██████▎   | 4060/6470 [46:40<27:10,  1.48it/s] 
4|4061|Loss: 0.773165762424469:  63%|██████▎   | 4061/6470 [46:40<27:29,  1.46it/s]
4|4062|Loss: 0.9161434173583984:  63%|██████▎   | 4061/6470 [46:40<27:29,  1.46it/s]
4|4062|Loss: 0.9161434173583984:  63%|██████▎   | 4062/6470 [46:41<27:51,  1.44it/s]
4|4063|Loss: 0.8467137217521667:  63%|██████▎   | 4062/6470 [46:41<27:51,  1.44it/s]
4|4063|Loss: 0.8467137217521667:  63%|██████▎   | 4063/6470 [46:41<26:42,  1.50it/s]
4|4064|Loss: 0.8655803203582764:  63%|██████▎   | 4063/6470 [46:42<26:42,  1.50it/s]
4|4064|Loss: 0.8655803203582764:  63%|██████▎   | 4064/6470 [46:42<28:20,  1.41it/s]
4|4065|Loss: 0.8797495365142822:  63%|██████▎   | 4064/6470 [46:43<28:20,  1.41it/s]
4|4065|Loss: 0.8797495365142822:  63%|██████▎   | 4065/6470 [46:43<25:38,  1.56it/s]
4|4066|Loss: 0.8174363374710083:  63%|██████▎   | 4065/6470 [46:43<25:38,  1.56it/s]
4|4066|Loss: 0.8174363374710083:  63%|██████▎   | 4066/6470 [46:43<24:30,  1.63it/s]
4|4067|Loss: 0.861368715763092:  63%|██████▎   | 4066/6470 [46:44<24:30,  1.63it/s] 
4|4067|Loss: 0.861368715763092:  63%|██████▎   | 4067/6470 [46:44<25:42,  1.56it/s]
4|4068|Loss: 0.852109432220459:  63%|██████▎   | 4067/6470 [46:44<25:42,  1.56it/s]
4|4068|Loss: 0.852109432220459:  63%|██████▎   | 4068/6470 [46:44<24:52,  1.61it/s]
4|4069|Loss: 0.7719897031784058:  63%|██████▎   | 4068/6470 [46:45<24:52,  1.61it/s]
4|4069|Loss: 0.7719897031784058:  63%|██████▎   | 4069/6470 [46:45<24:48,  1.61it/s]
4|4070|Loss: 0.8624722361564636:  63%|██████▎   | 4069/6470 [46:45<24:48,  1.61it/s]
4|4070|Loss: 0.8624722361564636:  63%|██████▎   | 4070/6470 [46:46<24:13,  1.65it/s]
4|4071|Loss: 0.8301749229431152:  63%|██████▎   | 4070/6470 [46:46<24:13,  1.65it/s]
4|4071|Loss: 0.8301749229431152:  63%|██████▎   | 4071/6470 [46:46<23:38,  1.69it/s]
4|4072|Loss: 0.6483832001686096:  63%|██████▎   | 4071/6470 [46:47<23:38,  1.69it/s]
4|4072|Loss: 0.6483832001686096:  63%|██████▎   | 4072/6470 [46:47<23:30,  1.70it/s]
4|4073|Loss: 0.9974369406700134:  63%|██████▎   | 4072/6470 [46:47<23:30,  1.70it/s]
4|4073|Loss: 0.9974369406700134:  63%|██████▎   | 4073/6470 [46:48<24:51,  1.61it/s]
4|4074|Loss: 0.8480671048164368:  63%|██████▎   | 4073/6470 [46:48<24:51,  1.61it/s]
4|4074|Loss: 0.8480671048164368:  63%|██████▎   | 4074/6470 [46:48<26:08,  1.53it/s]
4|4075|Loss: 0.7670466899871826:  63%|██████▎   | 4074/6470 [46:49<26:08,  1.53it/s]
4|4075|Loss: 0.7670466899871826:  63%|██████▎   | 4075/6470 [46:49<27:07,  1.47it/s]
4|4076|Loss: 0.7281518578529358:  63%|██████▎   | 4075/6470 [46:49<27:07,  1.47it/s]
4|4076|Loss: 0.7281518578529358:  63%|██████▎   | 4076/6470 [46:50<26:17,  1.52it/s]
4|4077|Loss: 0.9139636754989624:  63%|██████▎   | 4076/6470 [46:50<26:17,  1.52it/s]
4|4077|Loss: 0.9139636754989624:  63%|██████▎   | 4077/6470 [46:50<26:36,  1.50it/s]
4|4078|Loss: 0.8793008923530579:  63%|██████▎   | 4077/6470 [46:51<26:36,  1.50it/s]
4|4078|Loss: 0.8793008923530579:  63%|██████▎   | 4078/6470 [46:51<27:55,  1.43it/s]
4|4079|Loss: 1.0354211330413818:  63%|██████▎   | 4078/6470 [46:51<27:55,  1.43it/s]
4|4079|Loss: 1.0354211330413818:  63%|██████▎   | 4079/6470 [46:52<26:33,  1.50it/s]
4|4080|Loss: 0.9184946417808533:  63%|██████▎   | 4079/6470 [46:52<26:33,  1.50it/s]
4|4080|Loss: 0.9184946417808533:  63%|██████▎   | 4080/6470 [46:52<26:55,  1.48it/s]
4|4081|Loss: 0.9459701180458069:  63%|██████▎   | 4080/6470 [46:53<26:55,  1.48it/s]
4|4081|Loss: 0.9459701180458069:  63%|██████▎   | 4081/6470 [46:53<28:39,  1.39it/s]
4|4082|Loss: 0.8930283784866333:  63%|██████▎   | 4081/6470 [46:54<28:39,  1.39it/s]
4|4082|Loss: 0.8930283784866333:  63%|██████▎   | 4082/6470 [46:54<27:22,  1.45it/s]
4|4083|Loss: 0.9170499444007874:  63%|██████▎   | 4082/6470 [46:54<27:22,  1.45it/s]
4|4083|Loss: 0.9170499444007874:  63%|██████▎   | 4083/6470 [46:54<27:32,  1.44it/s]
4|4084|Loss: 0.8033315539360046:  63%|██████▎   | 4083/6470 [46:55<27:32,  1.44it/s]
4|4084|Loss: 0.8033315539360046:  63%|██████▎   | 4084/6470 [46:55<26:58,  1.47it/s]
4|4085|Loss: 0.8961849212646484:  63%|██████▎   | 4084/6470 [46:55<26:58,  1.47it/s]
4|4085|Loss: 0.8961849212646484:  63%|██████▎   | 4085/6470 [46:56<26:18,  1.51it/s]
4|4086|Loss: 0.6845283508300781:  63%|██████▎   | 4085/6470 [46:56<26:18,  1.51it/s]
4|4086|Loss: 0.6845283508300781:  63%|██████▎   | 4086/6470 [46:56<25:51,  1.54it/s]
4|4087|Loss: 0.8522915840148926:  63%|██████▎   | 4086/6470 [46:57<25:51,  1.54it/s]
4|4087|Loss: 0.8522915840148926:  63%|██████▎   | 4087/6470 [46:57<25:39,  1.55it/s]
4|4088|Loss: 0.7620285749435425:  63%|██████▎   | 4087/6470 [46:57<25:39,  1.55it/s]
4|4088|Loss: 0.7620285749435425:  63%|██████▎   | 4088/6470 [46:58<26:37,  1.49it/s]
4|4089|Loss: 0.8488078713417053:  63%|██████▎   | 4088/6470 [46:58<26:37,  1.49it/s]
4|4089|Loss: 0.8488078713417053:  63%|██████▎   | 4089/6470 [46:58<26:40,  1.49it/s]
4|4090|Loss: 0.8118802905082703:  63%|██████▎   | 4089/6470 [46:59<26:40,  1.49it/s]
4|4090|Loss: 0.8118802905082703:  63%|██████▎   | 4090/6470 [46:59<28:10,  1.41it/s]
4|4091|Loss: 0.7640630006790161:  63%|██████▎   | 4090/6470 [47:00<28:10,  1.41it/s]
4|4091|Loss: 0.7640630006790161:  63%|██████▎   | 4091/6470 [47:00<27:38,  1.43it/s]
4|4092|Loss: 0.7834492921829224:  63%|██████▎   | 4091/6470 [47:00<27:38,  1.43it/s]
4|4092|Loss: 0.7834492921829224:  63%|██████▎   | 4092/6470 [47:00<26:02,  1.52it/s]
4|4093|Loss: 0.7950746417045593:  63%|██████▎   | 4092/6470 [47:01<26:02,  1.52it/s]
4|4093|Loss: 0.7950746417045593:  63%|██████▎   | 4093/6470 [47:01<26:27,  1.50it/s]
4|4094|Loss: 0.8659148812294006:  63%|██████▎   | 4093/6470 [47:02<26:27,  1.50it/s]
4|4094|Loss: 0.8659148812294006:  63%|██████▎   | 4094/6470 [47:02<27:12,  1.46it/s]
4|4095|Loss: 0.8428889513015747:  63%|██████▎   | 4094/6470 [47:02<27:12,  1.46it/s]
4|4095|Loss: 0.8428889513015747:  63%|██████▎   | 4095/6470 [47:03<28:54,  1.37it/s]
4|4096|Loss: 0.9040808081626892:  63%|██████▎   | 4095/6470 [47:03<28:54,  1.37it/s]
4|4096|Loss: 0.9040808081626892:  63%|██████▎   | 4096/6470 [47:03<28:24,  1.39it/s]
4|4097|Loss: 0.9014389514923096:  63%|██████▎   | 4096/6470 [47:04<28:24,  1.39it/s]
4|4097|Loss: 0.9014389514923096:  63%|██████▎   | 4097/6470 [47:04<28:18,  1.40it/s]
4|4098|Loss: 0.6913611888885498:  63%|██████▎   | 4097/6470 [47:05<28:18,  1.40it/s]
4|4098|Loss: 0.6913611888885498:  63%|██████▎   | 4098/6470 [47:05<29:03,  1.36it/s]
4|4099|Loss: 0.8537148833274841:  63%|██████▎   | 4098/6470 [47:05<29:03,  1.36it/s]
4|4099|Loss: 0.8537148833274841:  63%|██████▎   | 4099/6470 [47:06<28:53,  1.37it/s]
4|4100|Loss: 0.7901749014854431:  63%|██████▎   | 4099/6470 [47:06<28:53,  1.37it/s]
4|4100|Loss: 0.7901749014854431:  63%|██████▎   | 4100/6470 [47:06<28:51,  1.37it/s]
4|4101|Loss: 0.8875311613082886:  63%|██████▎   | 4100/6470 [47:07<28:51,  1.37it/s]
4|4101|Loss: 0.8875311613082886:  63%|██████▎   | 4101/6470 [47:07<28:52,  1.37it/s]
4|4102|Loss: 0.9012254476547241:  63%|██████▎   | 4101/6470 [47:07<28:52,  1.37it/s]
4|4102|Loss: 0.9012254476547241:  63%|██████▎   | 4102/6470 [47:08<27:33,  1.43it/s]
4|4103|Loss: 0.6711946129798889:  63%|██████▎   | 4102/6470 [47:08<27:33,  1.43it/s]
4|4103|Loss: 0.6711946129798889:  63%|██████▎   | 4103/6470 [47:08<27:14,  1.45it/s]
4|4104|Loss: 0.7788019180297852:  63%|██████▎   | 4103/6470 [47:09<27:14,  1.45it/s]
4|4104|Loss: 0.7788019180297852:  63%|██████▎   | 4104/6470 [47:09<27:48,  1.42it/s]
4|4105|Loss: 0.9442777633666992:  63%|██████▎   | 4104/6470 [47:10<27:48,  1.42it/s]
4|4105|Loss: 0.9442777633666992:  63%|██████▎   | 4105/6470 [47:10<29:09,  1.35it/s]
4|4106|Loss: 0.9544830918312073:  63%|██████▎   | 4105/6470 [47:10<29:09,  1.35it/s]
4|4106|Loss: 0.9544830918312073:  63%|██████▎   | 4106/6470 [47:11<29:22,  1.34it/s]
4|4107|Loss: 0.8545206189155579:  63%|██████▎   | 4106/6470 [47:11<29:22,  1.34it/s]
4|4107|Loss: 0.8545206189155579:  63%|██████▎   | 4107/6470 [47:11<28:08,  1.40it/s]
4|4108|Loss: 0.7763469815254211:  63%|██████▎   | 4107/6470 [47:12<28:08,  1.40it/s]
4|4108|Loss: 0.7763469815254211:  63%|██████▎   | 4108/6470 [47:12<29:00,  1.36it/s]
4|4109|Loss: 0.6877281665802002:  63%|██████▎   | 4108/6470 [47:12<29:00,  1.36it/s]
4|4109|Loss: 0.6877281665802002:  64%|██████▎   | 4109/6470 [47:13<27:07,  1.45it/s]
4|4110|Loss: 0.684569776058197:  64%|██████▎   | 4109/6470 [47:13<27:07,  1.45it/s] 
4|4110|Loss: 0.684569776058197:  64%|██████▎   | 4110/6470 [47:13<28:13,  1.39it/s]
4|4111|Loss: 0.8475383520126343:  64%|██████▎   | 4110/6470 [47:14<28:13,  1.39it/s]
4|4111|Loss: 0.8475383520126343:  64%|██████▎   | 4111/6470 [47:14<30:24,  1.29it/s]
4|4112|Loss: 0.8076359629631042:  64%|██████▎   | 4111/6470 [47:15<30:24,  1.29it/s]
4|4112|Loss: 0.8076359629631042:  64%|██████▎   | 4112/6470 [47:15<28:56,  1.36it/s]
4|4113|Loss: 0.8833273649215698:  64%|██████▎   | 4112/6470 [47:15<28:56,  1.36it/s]
4|4113|Loss: 0.8833273649215698:  64%|██████▎   | 4113/6470 [47:16<27:07,  1.45it/s]
4|4114|Loss: 0.8725364208221436:  64%|██████▎   | 4113/6470 [47:16<27:07,  1.45it/s]
4|4114|Loss: 0.8725364208221436:  64%|██████▎   | 4114/6470 [47:16<26:14,  1.50it/s]
4|4115|Loss: 0.9460287094116211:  64%|██████▎   | 4114/6470 [47:17<26:14,  1.50it/s]
4|4115|Loss: 0.9460287094116211:  64%|██████▎   | 4115/6470 [47:17<24:23,  1.61it/s]
4|4116|Loss: 0.9444074630737305:  64%|██████▎   | 4115/6470 [47:17<24:23,  1.61it/s]
4|4116|Loss: 0.9444074630737305:  64%|██████▎   | 4116/6470 [47:17<23:57,  1.64it/s]
4|4117|Loss: 0.8718785047531128:  64%|██████▎   | 4116/6470 [47:18<23:57,  1.64it/s]
4|4117|Loss: 0.8718785047531128:  64%|██████▎   | 4117/6470 [47:18<25:58,  1.51it/s]
4|4118|Loss: 0.8476491570472717:  64%|██████▎   | 4117/6470 [47:19<25:58,  1.51it/s]
4|4118|Loss: 0.8476491570472717:  64%|██████▎   | 4118/6470 [47:19<26:55,  1.46it/s]
4|4119|Loss: 0.7818748950958252:  64%|██████▎   | 4118/6470 [47:19<26:55,  1.46it/s]
4|4119|Loss: 0.7818748950958252:  64%|██████▎   | 4119/6470 [47:20<27:11,  1.44it/s]
4|4120|Loss: 0.8433142304420471:  64%|██████▎   | 4119/6470 [47:20<27:11,  1.44it/s]
4|4120|Loss: 0.8433142304420471:  64%|██████▎   | 4120/6470 [47:20<26:04,  1.50it/s]
4|4121|Loss: 0.9252025485038757:  64%|██████▎   | 4120/6470 [47:20<26:04,  1.50it/s]
4|4121|Loss: 0.9252025485038757:  64%|██████▎   | 4121/6470 [47:21<24:36,  1.59it/s]
4|4122|Loss: 0.8844904899597168:  64%|██████▎   | 4121/6470 [47:21<24:36,  1.59it/s]
4|4122|Loss: 0.8844904899597168:  64%|██████▎   | 4122/6470 [47:21<25:26,  1.54it/s]
4|4123|Loss: 0.8003552556037903:  64%|██████▎   | 4122/6470 [47:22<25:26,  1.54it/s]
4|4123|Loss: 0.8003552556037903:  64%|██████▎   | 4123/6470 [47:22<26:14,  1.49it/s]
4|4124|Loss: 0.8580290675163269:  64%|██████▎   | 4123/6470 [47:23<26:14,  1.49it/s]
4|4124|Loss: 0.8580290675163269:  64%|██████▎   | 4124/6470 [47:23<28:34,  1.37it/s]
4|4125|Loss: 0.8406279683113098:  64%|██████▎   | 4124/6470 [47:23<28:34,  1.37it/s]
4|4125|Loss: 0.8406279683113098:  64%|██████▍   | 4125/6470 [47:24<28:11,  1.39it/s]
4|4126|Loss: 0.895037829875946:  64%|██████▍   | 4125/6470 [47:24<28:11,  1.39it/s] 
4|4126|Loss: 0.895037829875946:  64%|██████▍   | 4126/6470 [47:24<28:36,  1.37it/s]
4|4127|Loss: 0.8822884559631348:  64%|██████▍   | 4126/6470 [47:25<28:36,  1.37it/s]
4|4127|Loss: 0.8822884559631348:  64%|██████▍   | 4127/6470 [47:25<29:28,  1.32it/s]
4|4128|Loss: 0.8124468326568604:  64%|██████▍   | 4127/6470 [47:26<29:28,  1.32it/s]
4|4128|Loss: 0.8124468326568604:  64%|██████▍   | 4128/6470 [47:26<28:33,  1.37it/s]
4|4129|Loss: 0.8720030784606934:  64%|██████▍   | 4128/6470 [47:26<28:33,  1.37it/s]
4|4129|Loss: 0.8720030784606934:  64%|██████▍   | 4129/6470 [47:27<27:30,  1.42it/s]
4|4130|Loss: 0.7682129144668579:  64%|██████▍   | 4129/6470 [47:27<27:30,  1.42it/s]
4|4130|Loss: 0.7682129144668579:  64%|██████▍   | 4130/6470 [47:27<25:00,  1.56it/s]
4|4131|Loss: 0.8060957193374634:  64%|██████▍   | 4130/6470 [47:27<25:00,  1.56it/s]
4|4131|Loss: 0.8060957193374634:  64%|██████▍   | 4131/6470 [47:28<25:28,  1.53it/s]
4|4132|Loss: 0.7615203261375427:  64%|██████▍   | 4131/6470 [47:28<25:28,  1.53it/s]
4|4132|Loss: 0.7615203261375427:  64%|██████▍   | 4132/6470 [47:28<25:41,  1.52it/s]
4|4133|Loss: 0.8738293051719666:  64%|██████▍   | 4132/6470 [47:29<25:41,  1.52it/s]
4|4133|Loss: 0.8738293051719666:  64%|██████▍   | 4133/6470 [47:29<25:30,  1.53it/s]
4|4134|Loss: 0.7399559020996094:  64%|██████▍   | 4133/6470 [47:30<25:30,  1.53it/s]
4|4134|Loss: 0.7399559020996094:  64%|██████▍   | 4134/6470 [47:30<28:24,  1.37it/s]
4|4135|Loss: 0.9022970795631409:  64%|██████▍   | 4134/6470 [47:30<28:24,  1.37it/s]
4|4135|Loss: 0.9022970795631409:  64%|██████▍   | 4135/6470 [47:31<26:57,  1.44it/s]
4|4136|Loss: 0.9694058895111084:  64%|██████▍   | 4135/6470 [47:31<26:57,  1.44it/s]
4|4136|Loss: 0.9694058895111084:  64%|██████▍   | 4136/6470 [47:31<24:58,  1.56it/s]
4|4137|Loss: 0.7110732793807983:  64%|██████▍   | 4136/6470 [47:31<24:58,  1.56it/s]
4|4137|Loss: 0.7110732793807983:  64%|██████▍   | 4137/6470 [47:32<25:30,  1.52it/s]
4|4138|Loss: 0.9132853150367737:  64%|██████▍   | 4137/6470 [47:32<25:30,  1.52it/s]
4|4138|Loss: 0.9132853150367737:  64%|██████▍   | 4138/6470 [47:32<24:19,  1.60it/s]
4|4139|Loss: 0.8825333118438721:  64%|██████▍   | 4138/6470 [47:33<24:19,  1.60it/s]
4|4139|Loss: 0.8825333118438721:  64%|██████▍   | 4139/6470 [47:33<24:09,  1.61it/s]
4|4140|Loss: 0.8805046677589417:  64%|██████▍   | 4139/6470 [47:33<24:09,  1.61it/s]
4|4140|Loss: 0.8805046677589417:  64%|██████▍   | 4140/6470 [47:34<24:06,  1.61it/s]
4|4141|Loss: 1.1056509017944336:  64%|██████▍   | 4140/6470 [47:34<24:06,  1.61it/s]
4|4141|Loss: 1.1056509017944336:  64%|██████▍   | 4141/6470 [47:34<23:00,  1.69it/s]
4|4142|Loss: 0.9251053929328918:  64%|██████▍   | 4141/6470 [47:34<23:00,  1.69it/s]
4|4142|Loss: 0.9251053929328918:  64%|██████▍   | 4142/6470 [47:35<23:45,  1.63it/s]
4|4143|Loss: 0.9754139184951782:  64%|██████▍   | 4142/6470 [47:35<23:45,  1.63it/s]
4|4143|Loss: 0.9754139184951782:  64%|██████▍   | 4143/6470 [47:36<25:55,  1.50it/s]
4|4144|Loss: 0.7645040154457092:  64%|██████▍   | 4143/6470 [47:36<25:55,  1.50it/s]
4|4144|Loss: 0.7645040154457092:  64%|██████▍   | 4144/6470 [47:36<25:10,  1.54it/s]
4|4145|Loss: 1.0133554935455322:  64%|██████▍   | 4144/6470 [47:37<25:10,  1.54it/s]
4|4145|Loss: 1.0133554935455322:  64%|██████▍   | 4145/6470 [47:37<27:06,  1.43it/s]
4|4146|Loss: 0.777685821056366:  64%|██████▍   | 4145/6470 [47:37<27:06,  1.43it/s] 
4|4146|Loss: 0.777685821056366:  64%|██████▍   | 4146/6470 [47:37<24:28,  1.58it/s]
4|4147|Loss: 0.7521678805351257:  64%|██████▍   | 4146/6470 [47:38<24:28,  1.58it/s]
4|4147|Loss: 0.7521678805351257:  64%|██████▍   | 4147/6470 [47:38<23:31,  1.65it/s]
4|4148|Loss: 0.8383780717849731:  64%|██████▍   | 4147/6470 [47:38<23:31,  1.65it/s]
4|4148|Loss: 0.8383780717849731:  64%|██████▍   | 4148/6470 [47:39<25:33,  1.51it/s]
4|4149|Loss: 0.788178563117981:  64%|██████▍   | 4148/6470 [47:39<25:33,  1.51it/s] 
4|4149|Loss: 0.788178563117981:  64%|██████▍   | 4149/6470 [47:39<25:43,  1.50it/s]
4|4150|Loss: 0.7462183833122253:  64%|██████▍   | 4149/6470 [47:40<25:43,  1.50it/s]
4|4150|Loss: 0.7462183833122253:  64%|██████▍   | 4150/6470 [47:40<26:14,  1.47it/s]
4|4151|Loss: 0.890908420085907:  64%|██████▍   | 4150/6470 [47:41<26:14,  1.47it/s] 
4|4151|Loss: 0.890908420085907:  64%|██████▍   | 4151/6470 [47:41<25:08,  1.54it/s]
4|4152|Loss: 0.870109498500824:  64%|██████▍   | 4151/6470 [47:41<25:08,  1.54it/s]
4|4152|Loss: 0.870109498500824:  64%|██████▍   | 4152/6470 [47:41<25:25,  1.52it/s]
4|4153|Loss: 0.7456689476966858:  64%|██████▍   | 4152/6470 [47:42<25:25,  1.52it/s]
4|4153|Loss: 0.7456689476966858:  64%|██████▍   | 4153/6470 [47:42<24:51,  1.55it/s]
4|4154|Loss: 0.976244330406189:  64%|██████▍   | 4153/6470 [47:42<24:51,  1.55it/s] 
4|4154|Loss: 0.976244330406189:  64%|██████▍   | 4154/6470 [47:43<23:01,  1.68it/s]
4|4155|Loss: 0.7771211862564087:  64%|██████▍   | 4154/6470 [47:43<23:01,  1.68it/s]
4|4155|Loss: 0.7771211862564087:  64%|██████▍   | 4155/6470 [47:43<22:11,  1.74it/s]
4|4156|Loss: 0.7809815406799316:  64%|██████▍   | 4155/6470 [47:43<22:11,  1.74it/s]
4|4156|Loss: 0.7809815406799316:  64%|██████▍   | 4156/6470 [47:44<23:19,  1.65it/s]
4|4157|Loss: 0.8077517151832581:  64%|██████▍   | 4156/6470 [47:44<23:19,  1.65it/s]
4|4157|Loss: 0.8077517151832581:  64%|██████▍   | 4157/6470 [47:44<24:29,  1.57it/s]
4|4158|Loss: 0.9171396493911743:  64%|██████▍   | 4157/6470 [47:45<24:29,  1.57it/s]
4|4158|Loss: 0.9171396493911743:  64%|██████▍   | 4158/6470 [47:45<23:49,  1.62it/s]
4|4159|Loss: 0.9575285911560059:  64%|██████▍   | 4158/6470 [47:45<23:49,  1.62it/s]
4|4159|Loss: 0.9575285911560059:  64%|██████▍   | 4159/6470 [47:46<25:16,  1.52it/s]
4|4160|Loss: 0.9884058237075806:  64%|██████▍   | 4159/6470 [47:46<25:16,  1.52it/s]
4|4160|Loss: 0.9884058237075806:  64%|██████▍   | 4160/6470 [47:47<27:29,  1.40it/s]
4|4161|Loss: 0.905518114566803:  64%|██████▍   | 4160/6470 [47:47<27:29,  1.40it/s] 
4|4161|Loss: 0.905518114566803:  64%|██████▍   | 4161/6470 [47:47<27:43,  1.39it/s]
4|4162|Loss: 0.8333752751350403:  64%|██████▍   | 4161/6470 [47:48<27:43,  1.39it/s]
4|4162|Loss: 0.8333752751350403:  64%|██████▍   | 4162/6470 [47:48<28:44,  1.34it/s]
4|4163|Loss: 0.7966734766960144:  64%|██████▍   | 4162/6470 [47:49<28:44,  1.34it/s]
4|4163|Loss: 0.7966734766960144:  64%|██████▍   | 4163/6470 [47:49<30:05,  1.28it/s]
4|4164|Loss: 1.041796326637268:  64%|██████▍   | 4163/6470 [47:49<30:05,  1.28it/s] 
4|4164|Loss: 1.041796326637268:  64%|██████▍   | 4164/6470 [47:50<29:53,  1.29it/s]
4|4165|Loss: 0.9255674481391907:  64%|██████▍   | 4164/6470 [47:50<29:53,  1.29it/s]
4|4165|Loss: 0.9255674481391907:  64%|██████▍   | 4165/6470 [47:51<29:23,  1.31it/s]
4|4166|Loss: 0.8669823408126831:  64%|██████▍   | 4165/6470 [47:51<29:23,  1.31it/s]
4|4166|Loss: 0.8669823408126831:  64%|██████▍   | 4166/6470 [47:51<28:49,  1.33it/s]
4|4167|Loss: 0.7774103283882141:  64%|██████▍   | 4166/6470 [47:52<28:49,  1.33it/s]
4|4167|Loss: 0.7774103283882141:  64%|██████▍   | 4167/6470 [47:52<28:42,  1.34it/s]
4|4168|Loss: 0.8652842044830322:  64%|██████▍   | 4167/6470 [47:52<28:42,  1.34it/s]
4|4168|Loss: 0.8652842044830322:  64%|██████▍   | 4168/6470 [47:53<30:02,  1.28it/s]
4|4169|Loss: 0.840764045715332:  64%|██████▍   | 4168/6470 [47:53<30:02,  1.28it/s] 
4|4169|Loss: 0.840764045715332:  64%|██████▍   | 4169/6470 [47:54<30:29,  1.26it/s]
4|4170|Loss: 0.699165940284729:  64%|██████▍   | 4169/6470 [47:54<30:29,  1.26it/s]
4|4170|Loss: 0.699165940284729:  64%|██████▍   | 4170/6470 [47:54<29:55,  1.28it/s]
4|4171|Loss: 0.7722670435905457:  64%|██████▍   | 4170/6470 [47:55<29:55,  1.28it/s]
4|4171|Loss: 0.7722670435905457:  64%|██████▍   | 4171/6470 [47:55<29:21,  1.31it/s]
4|4172|Loss: 0.8196941018104553:  64%|██████▍   | 4171/6470 [47:56<29:21,  1.31it/s]
4|4172|Loss: 0.8196941018104553:  64%|██████▍   | 4172/6470 [47:56<27:32,  1.39it/s]
4|4173|Loss: 0.9420164823532104:  64%|██████▍   | 4172/6470 [47:56<27:32,  1.39it/s]
4|4173|Loss: 0.9420164823532104:  64%|██████▍   | 4173/6470 [47:56<27:33,  1.39it/s]
4|4174|Loss: 0.8269534111022949:  64%|██████▍   | 4173/6470 [47:57<27:33,  1.39it/s]
4|4174|Loss: 0.8269534111022949:  65%|██████▍   | 4174/6470 [47:57<28:35,  1.34it/s]
4|4175|Loss: 0.860180675983429:  65%|██████▍   | 4174/6470 [47:58<28:35,  1.34it/s] 
4|4175|Loss: 0.860180675983429:  65%|██████▍   | 4175/6470 [47:58<28:41,  1.33it/s]
4|4176|Loss: 0.9083618521690369:  65%|██████▍   | 4175/6470 [47:58<28:41,  1.33it/s]
4|4176|Loss: 0.9083618521690369:  65%|██████▍   | 4176/6470 [47:59<28:15,  1.35it/s]
4|4177|Loss: 0.7679341435432434:  65%|██████▍   | 4176/6470 [47:59<28:15,  1.35it/s]
4|4177|Loss: 0.7679341435432434:  65%|██████▍   | 4177/6470 [47:59<28:04,  1.36it/s]
4|4178|Loss: 0.7916658520698547:  65%|██████▍   | 4177/6470 [48:00<28:04,  1.36it/s]
4|4178|Loss: 0.7916658520698547:  65%|██████▍   | 4178/6470 [48:00<27:01,  1.41it/s]
4|4179|Loss: 0.69209885597229:  65%|██████▍   | 4178/6470 [48:01<27:01,  1.41it/s]  
4|4179|Loss: 0.69209885597229:  65%|██████▍   | 4179/6470 [48:01<27:49,  1.37it/s]
4|4180|Loss: 1.0120495557785034:  65%|██████▍   | 4179/6470 [48:01<27:49,  1.37it/s]
4|4180|Loss: 1.0120495557785034:  65%|██████▍   | 4180/6470 [48:02<28:16,  1.35it/s]
4|4181|Loss: 0.7695060968399048:  65%|██████▍   | 4180/6470 [48:02<28:16,  1.35it/s]
4|4181|Loss: 0.7695060968399048:  65%|██████▍   | 4181/6470 [48:02<28:28,  1.34it/s]
4|4182|Loss: 0.8712665438652039:  65%|██████▍   | 4181/6470 [48:03<28:28,  1.34it/s]
4|4182|Loss: 0.8712665438652039:  65%|██████▍   | 4182/6470 [48:03<28:08,  1.36it/s]
4|4183|Loss: 0.9227702617645264:  65%|██████▍   | 4182/6470 [48:04<28:08,  1.36it/s]
4|4183|Loss: 0.9227702617645264:  65%|██████▍   | 4183/6470 [48:04<26:18,  1.45it/s]
4|4184|Loss: 0.9108471274375916:  65%|██████▍   | 4183/6470 [48:04<26:18,  1.45it/s]
4|4184|Loss: 0.9108471274375916:  65%|██████▍   | 4184/6470 [48:04<25:06,  1.52it/s]
4|4185|Loss: 0.8016926646232605:  65%|██████▍   | 4184/6470 [48:05<25:06,  1.52it/s]
4|4185|Loss: 0.8016926646232605:  65%|██████▍   | 4185/6470 [48:05<26:40,  1.43it/s]
4|4186|Loss: 0.8937231302261353:  65%|██████▍   | 4185/6470 [48:06<26:40,  1.43it/s]
4|4186|Loss: 0.8937231302261353:  65%|██████▍   | 4186/6470 [48:06<28:11,  1.35it/s]
4|4187|Loss: 0.9261650443077087:  65%|██████▍   | 4186/6470 [48:06<28:11,  1.35it/s]
4|4187|Loss: 0.9261650443077087:  65%|██████▍   | 4187/6470 [48:07<27:22,  1.39it/s]
4|4188|Loss: 0.8974308967590332:  65%|██████▍   | 4187/6470 [48:07<27:22,  1.39it/s]
4|4188|Loss: 0.8974308967590332:  65%|██████▍   | 4188/6470 [48:07<26:27,  1.44it/s]
4|4189|Loss: 0.7524132132530212:  65%|██████▍   | 4188/6470 [48:08<26:27,  1.44it/s]
4|4189|Loss: 0.7524132132530212:  65%|██████▍   | 4189/6470 [48:08<26:40,  1.43it/s]
4|4190|Loss: 0.9182500839233398:  65%|██████▍   | 4189/6470 [48:08<26:40,  1.43it/s]
4|4190|Loss: 0.9182500839233398:  65%|██████▍   | 4190/6470 [48:09<25:35,  1.48it/s]
4|4191|Loss: 0.7512722611427307:  65%|██████▍   | 4190/6470 [48:09<25:35,  1.48it/s]
4|4191|Loss: 0.7512722611427307:  65%|██████▍   | 4191/6470 [48:09<23:54,  1.59it/s]
4|4192|Loss: 0.8947802782058716:  65%|██████▍   | 4191/6470 [48:09<23:54,  1.59it/s]
4|4192|Loss: 0.8947802782058716:  65%|██████▍   | 4192/6470 [48:10<21:22,  1.78it/s]
4|4193|Loss: 0.8765146732330322:  65%|██████▍   | 4192/6470 [48:10<21:22,  1.78it/s]
4|4193|Loss: 0.8765146732330322:  65%|██████▍   | 4193/6470 [48:10<21:35,  1.76it/s]
4|4194|Loss: 0.83631831407547:  65%|██████▍   | 4193/6470 [48:10<21:35,  1.76it/s]  
4|4194|Loss: 0.83631831407547:  65%|██████▍   | 4194/6470 [48:11<22:23,  1.69it/s]
4|4195|Loss: 0.8702410459518433:  65%|██████▍   | 4194/6470 [48:11<22:23,  1.69it/s]
4|4195|Loss: 0.8702410459518433:  65%|██████▍   | 4195/6470 [48:11<22:44,  1.67it/s]
4|4196|Loss: 0.8955897092819214:  65%|██████▍   | 4195/6470 [48:12<22:44,  1.67it/s]
4|4196|Loss: 0.8955897092819214:  65%|██████▍   | 4196/6470 [48:12<24:49,  1.53it/s]
4|4197|Loss: 0.90317702293396:  65%|██████▍   | 4196/6470 [48:13<24:49,  1.53it/s]  
4|4197|Loss: 0.90317702293396:  65%|██████▍   | 4197/6470 [48:13<26:46,  1.41it/s]
4|4198|Loss: 0.6572849154472351:  65%|██████▍   | 4197/6470 [48:13<26:46,  1.41it/s]
4|4198|Loss: 0.6572849154472351:  65%|██████▍   | 4198/6470 [48:14<28:24,  1.33it/s]
4|4199|Loss: 1.0612629652023315:  65%|██████▍   | 4198/6470 [48:14<28:24,  1.33it/s]
4|4199|Loss: 1.0612629652023315:  65%|██████▍   | 4199/6470 [48:15<27:29,  1.38it/s]
4|4200|Loss: 0.8324121236801147:  65%|██████▍   | 4199/6470 [48:15<27:29,  1.38it/s]
4|4200|Loss: 0.8324121236801147:  65%|██████▍   | 4200/6470 [48:15<25:57,  1.46it/s]
4|4201|Loss: 1.0415103435516357:  65%|██████▍   | 4200/6470 [48:15<25:57,  1.46it/s]
4|4201|Loss: 1.0415103435516357:  65%|██████▍   | 4201/6470 [48:16<26:08,  1.45it/s]
4|4202|Loss: 0.8561010360717773:  65%|██████▍   | 4201/6470 [48:16<26:08,  1.45it/s]
4|4202|Loss: 0.8561010360717773:  65%|██████▍   | 4202/6470 [48:17<27:25,  1.38it/s]
4|4203|Loss: 0.8465114831924438:  65%|██████▍   | 4202/6470 [48:17<27:25,  1.38it/s]
4|4203|Loss: 0.8465114831924438:  65%|██████▍   | 4203/6470 [48:17<27:59,  1.35it/s]
4|4204|Loss: 0.9818064570426941:  65%|██████▍   | 4203/6470 [48:18<27:59,  1.35it/s]
4|4204|Loss: 0.9818064570426941:  65%|██████▍   | 4204/6470 [48:18<27:02,  1.40it/s]
4|4205|Loss: 0.8062164187431335:  65%|██████▍   | 4204/6470 [48:18<27:02,  1.40it/s]
4|4205|Loss: 0.8062164187431335:  65%|██████▍   | 4205/6470 [48:19<27:28,  1.37it/s]
4|4206|Loss: 0.8672266602516174:  65%|██████▍   | 4205/6470 [48:19<27:28,  1.37it/s]
4|4206|Loss: 0.8672266602516174:  65%|██████▌   | 4206/6470 [48:20<28:56,  1.30it/s]
4|4207|Loss: 0.8586353659629822:  65%|██████▌   | 4206/6470 [48:20<28:56,  1.30it/s]
4|4207|Loss: 0.8586353659629822:  65%|██████▌   | 4207/6470 [48:20<29:02,  1.30it/s]
4|4208|Loss: 0.8477082252502441:  65%|██████▌   | 4207/6470 [48:21<29:02,  1.30it/s]
4|4208|Loss: 0.8477082252502441:  65%|██████▌   | 4208/6470 [48:21<28:50,  1.31it/s]
4|4209|Loss: 0.7287297248840332:  65%|██████▌   | 4208/6470 [48:22<28:50,  1.31it/s]
4|4209|Loss: 0.7287297248840332:  65%|██████▌   | 4209/6470 [48:22<28:32,  1.32it/s]
4|4210|Loss: 0.9407314658164978:  65%|██████▌   | 4209/6470 [48:22<28:32,  1.32it/s]
4|4210|Loss: 0.9407314658164978:  65%|██████▌   | 4210/6470 [48:23<28:23,  1.33it/s]
4|4211|Loss: 0.724667489528656:  65%|██████▌   | 4210/6470 [48:23<28:23,  1.33it/s] 
4|4211|Loss: 0.724667489528656:  65%|██████▌   | 4211/6470 [48:23<28:10,  1.34it/s]
4|4212|Loss: 0.8600268363952637:  65%|██████▌   | 4211/6470 [48:24<28:10,  1.34it/s]
4|4212|Loss: 0.8600268363952637:  65%|██████▌   | 4212/6470 [48:24<29:19,  1.28it/s]
4|4213|Loss: 0.8172827363014221:  65%|██████▌   | 4212/6470 [48:25<29:19,  1.28it/s]
4|4213|Loss: 0.8172827363014221:  65%|██████▌   | 4213/6470 [48:25<30:37,  1.23it/s]
4|4214|Loss: 0.7879059314727783:  65%|██████▌   | 4213/6470 [48:26<30:37,  1.23it/s]
4|4214|Loss: 0.7879059314727783:  65%|██████▌   | 4214/6470 [48:26<31:03,  1.21it/s]
4|4215|Loss: 0.7622803449630737:  65%|██████▌   | 4214/6470 [48:26<31:03,  1.21it/s]
4|4215|Loss: 0.7622803449630737:  65%|██████▌   | 4215/6470 [48:27<30:48,  1.22it/s]
4|4216|Loss: 0.7993326187133789:  65%|██████▌   | 4215/6470 [48:27<30:48,  1.22it/s]
4|4216|Loss: 0.7993326187133789:  65%|██████▌   | 4216/6470 [48:28<31:02,  1.21it/s]
4|4217|Loss: 0.7419492602348328:  65%|██████▌   | 4216/6470 [48:28<31:02,  1.21it/s]
4|4217|Loss: 0.7419492602348328:  65%|██████▌   | 4217/6470 [48:28<30:01,  1.25it/s]
4|4218|Loss: 0.7953969240188599:  65%|██████▌   | 4217/6470 [48:29<30:01,  1.25it/s]
4|4218|Loss: 0.7953969240188599:  65%|██████▌   | 4218/6470 [48:29<28:08,  1.33it/s]
4|4219|Loss: 0.8301716446876526:  65%|██████▌   | 4218/6470 [48:29<28:08,  1.33it/s]
4|4219|Loss: 0.8301716446876526:  65%|██████▌   | 4219/6470 [48:30<26:14,  1.43it/s]
4|4220|Loss: 0.890287458896637:  65%|██████▌   | 4219/6470 [48:30<26:14,  1.43it/s] 
4|4220|Loss: 0.890287458896637:  65%|██████▌   | 4220/6470 [48:30<24:17,  1.54it/s]
4|4221|Loss: 0.8167757987976074:  65%|██████▌   | 4220/6470 [48:31<24:17,  1.54it/s]
4|4221|Loss: 0.8167757987976074:  65%|██████▌   | 4221/6470 [48:31<28:10,  1.33it/s]
4|4222|Loss: 0.8994839191436768:  65%|██████▌   | 4221/6470 [48:32<28:10,  1.33it/s]
4|4222|Loss: 0.8994839191436768:  65%|██████▌   | 4222/6470 [48:32<28:35,  1.31it/s]
4|4223|Loss: 0.8411340713500977:  65%|██████▌   | 4222/6470 [48:32<28:35,  1.31it/s]
4|4223|Loss: 0.8411340713500977:  65%|██████▌   | 4223/6470 [48:33<27:08,  1.38it/s]
4|4224|Loss: 0.6953790783882141:  65%|██████▌   | 4223/6470 [48:33<27:08,  1.38it/s]
4|4224|Loss: 0.6953790783882141:  65%|██████▌   | 4224/6470 [48:33<25:05,  1.49it/s]
4|4225|Loss: 0.722107470035553:  65%|██████▌   | 4224/6470 [48:33<25:05,  1.49it/s] 
4|4225|Loss: 0.722107470035553:  65%|██████▌   | 4225/6470 [48:34<23:59,  1.56it/s]
4|4226|Loss: 0.9107485413551331:  65%|██████▌   | 4225/6470 [48:34<23:59,  1.56it/s]
4|4226|Loss: 0.9107485413551331:  65%|██████▌   | 4226/6470 [48:34<23:46,  1.57it/s]
4|4227|Loss: 0.8585646748542786:  65%|██████▌   | 4226/6470 [48:35<23:46,  1.57it/s]
4|4227|Loss: 0.8585646748542786:  65%|██████▌   | 4227/6470 [48:35<23:15,  1.61it/s]
4|4228|Loss: 0.8582096099853516:  65%|██████▌   | 4227/6470 [48:35<23:15,  1.61it/s]
4|4228|Loss: 0.8582096099853516:  65%|██████▌   | 4228/6470 [48:35<22:20,  1.67it/s]
4|4229|Loss: 0.9115251302719116:  65%|██████▌   | 4228/6470 [48:36<22:20,  1.67it/s]
4|4229|Loss: 0.9115251302719116:  65%|██████▌   | 4229/6470 [48:36<22:20,  1.67it/s]
4|4230|Loss: 0.8637511730194092:  65%|██████▌   | 4229/6470 [48:36<22:20,  1.67it/s]
4|4230|Loss: 0.8637511730194092:  65%|██████▌   | 4230/6470 [48:37<23:22,  1.60it/s]
4|4231|Loss: 0.7660258412361145:  65%|██████▌   | 4230/6470 [48:37<23:22,  1.60it/s]
4|4231|Loss: 0.7660258412361145:  65%|██████▌   | 4231/6470 [48:37<24:40,  1.51it/s]
4|4232|Loss: 0.8018331527709961:  65%|██████▌   | 4231/6470 [48:38<24:40,  1.51it/s]
4|4232|Loss: 0.8018331527709961:  65%|██████▌   | 4232/6470 [48:38<26:40,  1.40it/s]
4|4233|Loss: 0.8453015089035034:  65%|██████▌   | 4232/6470 [48:39<26:40,  1.40it/s]
4|4233|Loss: 0.8453015089035034:  65%|██████▌   | 4233/6470 [48:39<24:42,  1.51it/s]
4|4234|Loss: 1.105349063873291:  65%|██████▌   | 4233/6470 [48:39<24:42,  1.51it/s] 
4|4234|Loss: 1.105349063873291:  65%|██████▌   | 4234/6470 [48:40<25:51,  1.44it/s]
4|4235|Loss: 0.8006485104560852:  65%|██████▌   | 4234/6470 [48:40<25:51,  1.44it/s]
4|4235|Loss: 0.8006485104560852:  65%|██████▌   | 4235/6470 [48:40<26:22,  1.41it/s]
4|4236|Loss: 0.8740247488021851:  65%|██████▌   | 4235/6470 [48:41<26:22,  1.41it/s]
4|4236|Loss: 0.8740247488021851:  65%|██████▌   | 4236/6470 [48:41<27:04,  1.38it/s]
4|4237|Loss: 0.835543692111969:  65%|██████▌   | 4236/6470 [48:42<27:04,  1.38it/s] 
4|4237|Loss: 0.835543692111969:  65%|██████▌   | 4237/6470 [48:42<28:09,  1.32it/s]
4|4238|Loss: 0.7672966122627258:  65%|██████▌   | 4237/6470 [48:42<28:09,  1.32it/s]
4|4238|Loss: 0.7672966122627258:  66%|██████▌   | 4238/6470 [48:43<26:16,  1.42it/s]
4|4239|Loss: 0.8142989873886108:  66%|██████▌   | 4238/6470 [48:43<26:16,  1.42it/s]
4|4239|Loss: 0.8142989873886108:  66%|██████▌   | 4239/6470 [48:43<25:11,  1.48it/s]
4|4240|Loss: 0.7741950750350952:  66%|██████▌   | 4239/6470 [48:43<25:11,  1.48it/s]
4|4240|Loss: 0.7741950750350952:  66%|██████▌   | 4240/6470 [48:44<23:34,  1.58it/s]
4|4241|Loss: 0.8437491059303284:  66%|██████▌   | 4240/6470 [48:44<23:34,  1.58it/s]
4|4241|Loss: 0.8437491059303284:  66%|██████▌   | 4241/6470 [48:44<23:51,  1.56it/s]
4|4242|Loss: 0.9368898272514343:  66%|██████▌   | 4241/6470 [48:45<23:51,  1.56it/s]
4|4242|Loss: 0.9368898272514343:  66%|██████▌   | 4242/6470 [48:45<24:03,  1.54it/s]
4|4243|Loss: 0.9859724044799805:  66%|██████▌   | 4242/6470 [48:45<24:03,  1.54it/s]
4|4243|Loss: 0.9859724044799805:  66%|██████▌   | 4243/6470 [48:46<26:01,  1.43it/s]
4|4244|Loss: 0.8974977135658264:  66%|██████▌   | 4243/6470 [48:46<26:01,  1.43it/s]
4|4244|Loss: 0.8974977135658264:  66%|██████▌   | 4244/6470 [48:47<26:53,  1.38it/s]
4|4245|Loss: 0.9439343214035034:  66%|██████▌   | 4244/6470 [48:47<26:53,  1.38it/s]
4|4245|Loss: 0.9439343214035034:  66%|██████▌   | 4245/6470 [48:47<27:42,  1.34it/s]
4|4246|Loss: 0.7471295595169067:  66%|██████▌   | 4245/6470 [48:48<27:42,  1.34it/s]
4|4246|Loss: 0.7471295595169067:  66%|██████▌   | 4246/6470 [48:48<28:46,  1.29it/s]
4|4247|Loss: 0.7786429524421692:  66%|██████▌   | 4246/6470 [48:49<28:46,  1.29it/s]
4|4247|Loss: 0.7786429524421692:  66%|██████▌   | 4247/6470 [48:49<28:17,  1.31it/s]
4|4248|Loss: 0.9422416687011719:  66%|██████▌   | 4247/6470 [48:49<28:17,  1.31it/s]
4|4248|Loss: 0.9422416687011719:  66%|██████▌   | 4248/6470 [48:50<26:02,  1.42it/s]
4|4249|Loss: 0.790943443775177:  66%|██████▌   | 4248/6470 [48:50<26:02,  1.42it/s] 
4|4249|Loss: 0.790943443775177:  66%|██████▌   | 4249/6470 [48:50<24:49,  1.49it/s]
4|4250|Loss: 0.8389037251472473:  66%|██████▌   | 4249/6470 [48:51<24:49,  1.49it/s]
4|4250|Loss: 0.8389037251472473:  66%|██████▌   | 4250/6470 [48:51<25:23,  1.46it/s]
4|4251|Loss: 0.9037699103355408:  66%|██████▌   | 4250/6470 [48:51<25:23,  1.46it/s]
4|4251|Loss: 0.9037699103355408:  66%|██████▌   | 4251/6470 [48:51<24:34,  1.50it/s]
4|4252|Loss: 0.7471085786819458:  66%|██████▌   | 4251/6470 [48:52<24:34,  1.50it/s]
4|4252|Loss: 0.7471085786819458:  66%|██████▌   | 4252/6470 [48:52<24:10,  1.53it/s]
4|4253|Loss: 0.7533858418464661:  66%|██████▌   | 4252/6470 [48:52<24:10,  1.53it/s]
4|4253|Loss: 0.7533858418464661:  66%|██████▌   | 4253/6470 [48:53<24:05,  1.53it/s]
4|4254|Loss: 0.7221762537956238:  66%|██████▌   | 4253/6470 [48:53<24:05,  1.53it/s]
4|4254|Loss: 0.7221762537956238:  66%|██████▌   | 4254/6470 [48:53<23:21,  1.58it/s]
4|4255|Loss: 0.9084705114364624:  66%|██████▌   | 4254/6470 [48:54<23:21,  1.58it/s]
4|4255|Loss: 0.9084705114364624:  66%|██████▌   | 4255/6470 [48:54<23:19,  1.58it/s]
4|4256|Loss: 0.8528757095336914:  66%|██████▌   | 4255/6470 [48:54<23:19,  1.58it/s]
4|4256|Loss: 0.8528757095336914:  66%|██████▌   | 4256/6470 [48:55<22:24,  1.65it/s]
4|4257|Loss: 0.8368319272994995:  66%|██████▌   | 4256/6470 [48:55<22:24,  1.65it/s]
4|4257|Loss: 0.8368319272994995:  66%|██████▌   | 4257/6470 [48:55<20:01,  1.84it/s]
4|4258|Loss: 0.8393028974533081:  66%|██████▌   | 4257/6470 [48:55<20:01,  1.84it/s]
4|4258|Loss: 0.8393028974533081:  66%|██████▌   | 4258/6470 [48:55<20:23,  1.81it/s]
4|4259|Loss: 0.7853648066520691:  66%|██████▌   | 4258/6470 [48:56<20:23,  1.81it/s]
4|4259|Loss: 0.7853648066520691:  66%|██████▌   | 4259/6470 [48:56<21:19,  1.73it/s]
4|4260|Loss: 0.742037296295166:  66%|██████▌   | 4259/6470 [48:56<21:19,  1.73it/s] 
4|4260|Loss: 0.742037296295166:  66%|██████▌   | 4260/6470 [48:57<21:49,  1.69it/s]
4|4261|Loss: 0.7849909067153931:  66%|██████▌   | 4260/6470 [48:57<21:49,  1.69it/s]
4|4261|Loss: 0.7849909067153931:  66%|██████▌   | 4261/6470 [48:57<23:05,  1.59it/s]
4|4262|Loss: 0.8441873788833618:  66%|██████▌   | 4261/6470 [48:58<23:05,  1.59it/s]
4|4262|Loss: 0.8441873788833618:  66%|██████▌   | 4262/6470 [48:58<23:27,  1.57it/s]
4|4263|Loss: 0.9216715097427368:  66%|██████▌   | 4262/6470 [48:58<23:27,  1.57it/s]
4|4263|Loss: 0.9216715097427368:  66%|██████▌   | 4263/6470 [48:59<23:59,  1.53it/s]
4|4264|Loss: 0.9187239408493042:  66%|██████▌   | 4263/6470 [48:59<23:59,  1.53it/s]
4|4264|Loss: 0.9187239408493042:  66%|██████▌   | 4264/6470 [49:00<24:28,  1.50it/s]
4|4265|Loss: 0.7224924564361572:  66%|██████▌   | 4264/6470 [49:00<24:28,  1.50it/s]
4|4265|Loss: 0.7224924564361572:  66%|██████▌   | 4265/6470 [49:00<25:46,  1.43it/s]
4|4266|Loss: 0.9717899560928345:  66%|██████▌   | 4265/6470 [49:01<25:46,  1.43it/s]
4|4266|Loss: 0.9717899560928345:  66%|██████▌   | 4266/6470 [49:01<24:22,  1.51it/s]
4|4267|Loss: 0.8364339470863342:  66%|██████▌   | 4266/6470 [49:01<24:22,  1.51it/s]
4|4267|Loss: 0.8364339470863342:  66%|██████▌   | 4267/6470 [49:01<23:48,  1.54it/s]
4|4268|Loss: 0.7202034592628479:  66%|██████▌   | 4267/6470 [49:02<23:48,  1.54it/s]
4|4268|Loss: 0.7202034592628479:  66%|██████▌   | 4268/6470 [49:02<24:38,  1.49it/s]
4|4269|Loss: 0.8248695731163025:  66%|██████▌   | 4268/6470 [49:03<24:38,  1.49it/s]
4|4269|Loss: 0.8248695731163025:  66%|██████▌   | 4269/6470 [49:03<25:10,  1.46it/s]
4|4270|Loss: 0.9316927790641785:  66%|██████▌   | 4269/6470 [49:03<25:10,  1.46it/s]
4|4270|Loss: 0.9316927790641785:  66%|██████▌   | 4270/6470 [49:04<25:40,  1.43it/s]
4|4271|Loss: 1.0453184843063354:  66%|██████▌   | 4270/6470 [49:04<25:40,  1.43it/s]
4|4271|Loss: 1.0453184843063354:  66%|██████▌   | 4271/6470 [49:04<26:41,  1.37it/s]
4|4272|Loss: 0.9842862486839294:  66%|██████▌   | 4271/6470 [49:05<26:41,  1.37it/s]
4|4272|Loss: 0.9842862486839294:  66%|██████▌   | 4272/6470 [49:05<27:03,  1.35it/s]
4|4273|Loss: 0.8435925245285034:  66%|██████▌   | 4272/6470 [49:06<27:03,  1.35it/s]
4|4273|Loss: 0.8435925245285034:  66%|██████▌   | 4273/6470 [49:06<27:28,  1.33it/s]
4|4274|Loss: 0.8781992793083191:  66%|██████▌   | 4273/6470 [49:06<27:28,  1.33it/s]
4|4274|Loss: 0.8781992793083191:  66%|██████▌   | 4274/6470 [49:07<24:56,  1.47it/s]
4|4275|Loss: 0.8622012138366699:  66%|██████▌   | 4274/6470 [49:07<24:56,  1.47it/s]
4|4275|Loss: 0.8622012138366699:  66%|██████▌   | 4275/6470 [49:07<23:49,  1.53it/s]
4|4276|Loss: 0.7477272748947144:  66%|██████▌   | 4275/6470 [49:07<23:49,  1.53it/s]
4|4276|Loss: 0.7477272748947144:  66%|██████▌   | 4276/6470 [49:08<23:04,  1.58it/s]
4|4277|Loss: 0.813969075679779:  66%|██████▌   | 4276/6470 [49:08<23:04,  1.58it/s] 
4|4277|Loss: 0.813969075679779:  66%|██████▌   | 4277/6470 [49:08<23:50,  1.53it/s]
4|4278|Loss: 0.7720096707344055:  66%|██████▌   | 4277/6470 [49:09<23:50,  1.53it/s]
4|4278|Loss: 0.7720096707344055:  66%|██████▌   | 4278/6470 [49:09<24:49,  1.47it/s]
4|4279|Loss: 0.8971874117851257:  66%|██████▌   | 4278/6470 [49:09<24:49,  1.47it/s]
4|4279|Loss: 0.8971874117851257:  66%|██████▌   | 4279/6470 [49:10<23:37,  1.55it/s]
4|4280|Loss: 0.820777177810669:  66%|██████▌   | 4279/6470 [49:10<23:37,  1.55it/s] 
4|4280|Loss: 0.820777177810669:  66%|██████▌   | 4280/6470 [49:10<23:17,  1.57it/s]
4|4281|Loss: 0.8832514882087708:  66%|██████▌   | 4280/6470 [49:11<23:17,  1.57it/s]
4|4281|Loss: 0.8832514882087708:  66%|██████▌   | 4281/6470 [49:11<22:33,  1.62it/s]
4|4282|Loss: 0.6803121566772461:  66%|██████▌   | 4281/6470 [49:11<22:33,  1.62it/s]
4|4282|Loss: 0.6803121566772461:  66%|██████▌   | 4282/6470 [49:12<23:11,  1.57it/s]
4|4283|Loss: 0.9847087264060974:  66%|██████▌   | 4282/6470 [49:12<23:11,  1.57it/s]
4|4283|Loss: 0.9847087264060974:  66%|██████▌   | 4283/6470 [49:12<22:56,  1.59it/s]
4|4284|Loss: 0.9072124361991882:  66%|██████▌   | 4283/6470 [49:13<22:56,  1.59it/s]
4|4284|Loss: 0.9072124361991882:  66%|██████▌   | 4284/6470 [49:13<23:54,  1.52it/s]
4|4285|Loss: 0.8716176748275757:  66%|██████▌   | 4284/6470 [49:13<23:54,  1.52it/s]
4|4285|Loss: 0.8716176748275757:  66%|██████▌   | 4285/6470 [49:14<24:50,  1.47it/s]
4|4286|Loss: 0.810494065284729:  66%|██████▌   | 4285/6470 [49:14<24:50,  1.47it/s] 
4|4286|Loss: 0.810494065284729:  66%|██████▌   | 4286/6470 [49:14<25:08,  1.45it/s]
4|4287|Loss: 0.8936150670051575:  66%|██████▌   | 4286/6470 [49:15<25:08,  1.45it/s]
4|4287|Loss: 0.8936150670051575:  66%|██████▋   | 4287/6470 [49:15<24:47,  1.47it/s]
4|4288|Loss: 0.7923367619514465:  66%|██████▋   | 4287/6470 [49:15<24:47,  1.47it/s]
4|4288|Loss: 0.7923367619514465:  66%|██████▋   | 4288/6470 [49:16<25:47,  1.41it/s]
4|4289|Loss: 0.9320260286331177:  66%|██████▋   | 4288/6470 [49:16<25:47,  1.41it/s]
4|4289|Loss: 0.9320260286331177:  66%|██████▋   | 4289/6470 [49:17<27:46,  1.31it/s]
4|4290|Loss: 0.7448864579200745:  66%|██████▋   | 4289/6470 [49:17<27:46,  1.31it/s]
4|4290|Loss: 0.7448864579200745:  66%|██████▋   | 4290/6470 [49:17<27:02,  1.34it/s]
4|4291|Loss: 0.885708212852478:  66%|██████▋   | 4290/6470 [49:18<27:02,  1.34it/s] 
4|4291|Loss: 0.885708212852478:  66%|██████▋   | 4291/6470 [49:18<28:15,  1.29it/s]
4|4292|Loss: 0.8584770560264587:  66%|██████▋   | 4291/6470 [49:19<28:15,  1.29it/s]
4|4292|Loss: 0.8584770560264587:  66%|██████▋   | 4292/6470 [49:19<28:42,  1.26it/s]
4|4293|Loss: 1.015482783317566:  66%|██████▋   | 4292/6470 [49:19<28:42,  1.26it/s] 
4|4293|Loss: 1.015482783317566:  66%|██████▋   | 4293/6470 [49:20<25:49,  1.40it/s]
4|4294|Loss: 0.7221761345863342:  66%|██████▋   | 4293/6470 [49:20<25:49,  1.40it/s]
4|4294|Loss: 0.7221761345863342:  66%|██████▋   | 4294/6470 [49:20<27:25,  1.32it/s]
4|4295|Loss: 0.736690104007721:  66%|██████▋   | 4294/6470 [49:21<27:25,  1.32it/s] 
4|4295|Loss: 0.736690104007721:  66%|██████▋   | 4295/6470 [49:21<26:22,  1.37it/s]
4|4296|Loss: 0.8722991943359375:  66%|██████▋   | 4295/6470 [49:21<26:22,  1.37it/s]
4|4296|Loss: 0.8722991943359375:  66%|██████▋   | 4296/6470 [49:22<26:56,  1.35it/s]
4|4297|Loss: 0.8198211789131165:  66%|██████▋   | 4296/6470 [49:22<26:56,  1.35it/s]
4|4297|Loss: 0.8198211789131165:  66%|██████▋   | 4297/6470 [49:23<27:09,  1.33it/s]
4|4298|Loss: 0.826953649520874:  66%|██████▋   | 4297/6470 [49:23<27:09,  1.33it/s] 
4|4298|Loss: 0.826953649520874:  66%|██████▋   | 4298/6470 [49:23<27:11,  1.33it/s]
4|4299|Loss: 0.9886803030967712:  66%|██████▋   | 4298/6470 [49:24<27:11,  1.33it/s]
4|4299|Loss: 0.9886803030967712:  66%|██████▋   | 4299/6470 [49:24<25:15,  1.43it/s]
4|4300|Loss: 0.9670624732971191:  66%|██████▋   | 4299/6470 [49:24<25:15,  1.43it/s]
4|4300|Loss: 0.9670624732971191:  66%|██████▋   | 4300/6470 [49:25<24:25,  1.48it/s]
4|4301|Loss: 0.9937139749526978:  66%|██████▋   | 4300/6470 [49:25<24:25,  1.48it/s]
4|4301|Loss: 0.9937139749526978:  66%|██████▋   | 4301/6470 [49:25<24:37,  1.47it/s]
4|4302|Loss: 0.7490519285202026:  66%|██████▋   | 4301/6470 [49:26<24:37,  1.47it/s]
4|4302|Loss: 0.7490519285202026:  66%|██████▋   | 4302/6470 [49:26<25:14,  1.43it/s]
4|4303|Loss: 0.7259457111358643:  66%|██████▋   | 4302/6470 [49:27<25:14,  1.43it/s]
4|4303|Loss: 0.7259457111358643:  67%|██████▋   | 4303/6470 [49:27<27:40,  1.30it/s]
4|4304|Loss: 1.0088176727294922:  67%|██████▋   | 4303/6470 [49:27<27:40,  1.30it/s]
4|4304|Loss: 1.0088176727294922:  67%|██████▋   | 4304/6470 [49:28<28:53,  1.25it/s]
4|4305|Loss: 0.8930096626281738:  67%|██████▋   | 4304/6470 [49:28<28:53,  1.25it/s]
4|4305|Loss: 0.8930096626281738:  67%|██████▋   | 4305/6470 [49:28<26:52,  1.34it/s]
4|4306|Loss: 0.9353178143501282:  67%|██████▋   | 4305/6470 [49:29<26:52,  1.34it/s]
4|4306|Loss: 0.9353178143501282:  67%|██████▋   | 4306/6470 [49:29<23:54,  1.51it/s]
4|4307|Loss: 0.7726438045501709:  67%|██████▋   | 4306/6470 [49:29<23:54,  1.51it/s]
4|4307|Loss: 0.7726438045501709:  67%|██████▋   | 4307/6470 [49:29<22:45,  1.58it/s]
4|4308|Loss: 0.9326627254486084:  67%|██████▋   | 4307/6470 [49:30<22:45,  1.58it/s]
4|4308|Loss: 0.9326627254486084:  67%|██████▋   | 4308/6470 [49:30<24:39,  1.46it/s]
4|4309|Loss: 0.6745235919952393:  67%|██████▋   | 4308/6470 [49:31<24:39,  1.46it/s]
4|4309|Loss: 0.6745235919952393:  67%|██████▋   | 4309/6470 [49:31<26:22,  1.37it/s]
4|4310|Loss: 0.7785921096801758:  67%|██████▋   | 4309/6470 [49:32<26:22,  1.37it/s]
4|4310|Loss: 0.7785921096801758:  67%|██████▋   | 4310/6470 [49:32<25:18,  1.42it/s]
4|4311|Loss: 0.8050792217254639:  67%|██████▋   | 4310/6470 [49:32<25:18,  1.42it/s]
4|4311|Loss: 0.8050792217254639:  67%|██████▋   | 4311/6470 [49:32<24:19,  1.48it/s]
4|4312|Loss: 0.6502352952957153:  67%|██████▋   | 4311/6470 [49:33<24:19,  1.48it/s]
4|4312|Loss: 0.6502352952957153:  67%|██████▋   | 4312/6470 [49:33<23:19,  1.54it/s]
4|4313|Loss: 0.8267408013343811:  67%|██████▋   | 4312/6470 [49:33<23:19,  1.54it/s]
4|4313|Loss: 0.8267408013343811:  67%|██████▋   | 4313/6470 [49:34<23:04,  1.56it/s]
4|4314|Loss: 0.8956865072250366:  67%|██████▋   | 4313/6470 [49:34<23:04,  1.56it/s]
4|4314|Loss: 0.8956865072250366:  67%|██████▋   | 4314/6470 [49:34<23:51,  1.51it/s]
4|4315|Loss: 0.8637946844100952:  67%|██████▋   | 4314/6470 [49:35<23:51,  1.51it/s]
4|4315|Loss: 0.8637946844100952:  67%|██████▋   | 4315/6470 [49:35<24:36,  1.46it/s]
4|4316|Loss: 0.8673280477523804:  67%|██████▋   | 4315/6470 [49:35<24:36,  1.46it/s]
4|4316|Loss: 0.8673280477523804:  67%|██████▋   | 4316/6470 [49:36<25:36,  1.40it/s]
4|4317|Loss: 0.870734691619873:  67%|██████▋   | 4316/6470 [49:36<25:36,  1.40it/s] 
4|4317|Loss: 0.870734691619873:  67%|██████▋   | 4317/6470 [49:37<26:46,  1.34it/s]
4|4318|Loss: 0.8537553548812866:  67%|██████▋   | 4317/6470 [49:37<26:46,  1.34it/s]
4|4318|Loss: 0.8537553548812866:  67%|██████▋   | 4318/6470 [49:37<26:49,  1.34it/s]
4|4319|Loss: 0.797544538974762:  67%|██████▋   | 4318/6470 [49:38<26:49,  1.34it/s] 
4|4319|Loss: 0.797544538974762:  67%|██████▋   | 4319/6470 [49:38<26:45,  1.34it/s]
4|4320|Loss: 0.7903146743774414:  67%|██████▋   | 4319/6470 [49:39<26:45,  1.34it/s]
4|4320|Loss: 0.7903146743774414:  67%|██████▋   | 4320/6470 [49:39<25:55,  1.38it/s]
4|4321|Loss: 0.7406868934631348:  67%|██████▋   | 4320/6470 [49:39<25:55,  1.38it/s]
4|4321|Loss: 0.7406868934631348:  67%|██████▋   | 4321/6470 [49:39<24:28,  1.46it/s]
4|4322|Loss: 0.8495491147041321:  67%|██████▋   | 4321/6470 [49:40<24:28,  1.46it/s]
4|4322|Loss: 0.8495491147041321:  67%|██████▋   | 4322/6470 [49:40<25:15,  1.42it/s]
4|4323|Loss: 0.8071660399436951:  67%|██████▋   | 4322/6470 [49:41<25:15,  1.42it/s]
4|4323|Loss: 0.8071660399436951:  67%|██████▋   | 4323/6470 [49:41<24:57,  1.43it/s]
4|4324|Loss: 0.8452357649803162:  67%|██████▋   | 4323/6470 [49:41<24:57,  1.43it/s]
4|4324|Loss: 0.8452357649803162:  67%|██████▋   | 4324/6470 [49:41<24:20,  1.47it/s]
4|4325|Loss: 0.7103251814842224:  67%|██████▋   | 4324/6470 [49:42<24:20,  1.47it/s]
4|4325|Loss: 0.7103251814842224:  67%|██████▋   | 4325/6470 [49:42<25:32,  1.40it/s]
4|4326|Loss: 0.8939455151557922:  67%|██████▋   | 4325/6470 [49:43<25:32,  1.40it/s]
4|4326|Loss: 0.8939455151557922:  67%|██████▋   | 4326/6470 [49:43<24:38,  1.45it/s]
4|4327|Loss: 0.8242651224136353:  67%|██████▋   | 4326/6470 [49:43<24:38,  1.45it/s]
4|4327|Loss: 0.8242651224136353:  67%|██████▋   | 4327/6470 [49:44<25:34,  1.40it/s]
4|4328|Loss: 0.8175494074821472:  67%|██████▋   | 4327/6470 [49:44<25:34,  1.40it/s]
4|4328|Loss: 0.8175494074821472:  67%|██████▋   | 4328/6470 [49:44<25:11,  1.42it/s]
4|4329|Loss: 0.8323487043380737:  67%|██████▋   | 4328/6470 [49:45<25:11,  1.42it/s]
4|4329|Loss: 0.8323487043380737:  67%|██████▋   | 4329/6470 [49:45<23:29,  1.52it/s]
4|4330|Loss: 0.8415228128433228:  67%|██████▋   | 4329/6470 [49:45<23:29,  1.52it/s]
4|4330|Loss: 0.8415228128433228:  67%|██████▋   | 4330/6470 [49:46<23:54,  1.49it/s]
4|4331|Loss: 0.738459587097168:  67%|██████▋   | 4330/6470 [49:46<23:54,  1.49it/s] 
4|4331|Loss: 0.738459587097168:  67%|██████▋   | 4331/6470 [49:46<23:51,  1.49it/s]
4|4332|Loss: 0.857046365737915:  67%|██████▋   | 4331/6470 [49:47<23:51,  1.49it/s]
4|4332|Loss: 0.857046365737915:  67%|██████▋   | 4332/6470 [49:47<25:12,  1.41it/s]
4|4333|Loss: 0.8845815658569336:  67%|██████▋   | 4332/6470 [49:47<25:12,  1.41it/s]
4|4333|Loss: 0.8845815658569336:  67%|██████▋   | 4333/6470 [49:48<22:42,  1.57it/s]
4|4334|Loss: 0.7959943413734436:  67%|██████▋   | 4333/6470 [49:48<22:42,  1.57it/s]
4|4334|Loss: 0.7959943413734436:  67%|██████▋   | 4334/6470 [49:48<22:49,  1.56it/s]
4|4335|Loss: 0.909938633441925:  67%|██████▋   | 4334/6470 [49:49<22:49,  1.56it/s] 
4|4335|Loss: 0.909938633441925:  67%|██████▋   | 4335/6470 [49:49<22:57,  1.55it/s]
4|4336|Loss: 0.7517824769020081:  67%|██████▋   | 4335/6470 [49:49<22:57,  1.55it/s]
4|4336|Loss: 0.7517824769020081:  67%|██████▋   | 4336/6470 [49:50<23:40,  1.50it/s]
4|4337|Loss: 0.9622775912284851:  67%|██████▋   | 4336/6470 [49:50<23:40,  1.50it/s]
4|4337|Loss: 0.9622775912284851:  67%|██████▋   | 4337/6470 [49:50<23:24,  1.52it/s]
4|4338|Loss: 0.9047793745994568:  67%|██████▋   | 4337/6470 [49:51<23:24,  1.52it/s]
4|4338|Loss: 0.9047793745994568:  67%|██████▋   | 4338/6470 [49:51<22:57,  1.55it/s]
4|4339|Loss: 0.8071101307868958:  67%|██████▋   | 4338/6470 [49:51<22:57,  1.55it/s]
4|4339|Loss: 0.8071101307868958:  67%|██████▋   | 4339/6470 [49:52<23:40,  1.50it/s]
4|4340|Loss: 0.8629710078239441:  67%|██████▋   | 4339/6470 [49:52<23:40,  1.50it/s]
4|4340|Loss: 0.8629710078239441:  67%|██████▋   | 4340/6470 [49:52<24:52,  1.43it/s]
4|4341|Loss: 0.7907595038414001:  67%|██████▋   | 4340/6470 [49:53<24:52,  1.43it/s]
4|4341|Loss: 0.7907595038414001:  67%|██████▋   | 4341/6470 [49:53<25:32,  1.39it/s]
4|4342|Loss: 0.9720207452774048:  67%|██████▋   | 4341/6470 [49:54<25:32,  1.39it/s]
4|4342|Loss: 0.9720207452774048:  67%|██████▋   | 4342/6470 [49:54<26:32,  1.34it/s]
4|4343|Loss: 1.0065053701400757:  67%|██████▋   | 4342/6470 [49:54<26:32,  1.34it/s]
4|4343|Loss: 1.0065053701400757:  67%|██████▋   | 4343/6470 [49:55<27:05,  1.31it/s]
4|4344|Loss: 0.7880330085754395:  67%|██████▋   | 4343/6470 [49:55<27:05,  1.31it/s]
4|4344|Loss: 0.7880330085754395:  67%|██████▋   | 4344/6470 [49:55<26:11,  1.35it/s]
4|4345|Loss: 0.7937482595443726:  67%|██████▋   | 4344/6470 [49:56<26:11,  1.35it/s]
4|4345|Loss: 0.7937482595443726:  67%|██████▋   | 4345/6470 [49:56<24:57,  1.42it/s]
4|4346|Loss: 0.7265950441360474:  67%|██████▋   | 4345/6470 [49:56<24:57,  1.42it/s]
4|4346|Loss: 0.7265950441360474:  67%|██████▋   | 4346/6470 [49:57<26:12,  1.35it/s]
4|4347|Loss: 0.8351992964744568:  67%|██████▋   | 4346/6470 [49:57<26:12,  1.35it/s]
4|4347|Loss: 0.8351992964744568:  67%|██████▋   | 4347/6470 [49:58<25:47,  1.37it/s]
4|4348|Loss: 0.8778743743896484:  67%|██████▋   | 4347/6470 [49:58<25:47,  1.37it/s]
4|4348|Loss: 0.8778743743896484:  67%|██████▋   | 4348/6470 [49:58<23:18,  1.52it/s]
4|4349|Loss: 0.789212167263031:  67%|██████▋   | 4348/6470 [49:58<23:18,  1.52it/s] 
4|4349|Loss: 0.789212167263031:  67%|██████▋   | 4349/6470 [49:59<22:28,  1.57it/s]
4|4350|Loss: 0.707612931728363:  67%|██████▋   | 4349/6470 [49:59<22:28,  1.57it/s]
4|4350|Loss: 0.707612931728363:  67%|██████▋   | 4350/6470 [49:59<23:03,  1.53it/s]
4|4351|Loss: 0.8883987069129944:  67%|██████▋   | 4350/6470 [50:00<23:03,  1.53it/s]
4|4351|Loss: 0.8883987069129944:  67%|██████▋   | 4351/6470 [50:00<24:27,  1.44it/s]
4|4352|Loss: 0.841454029083252:  67%|██████▋   | 4351/6470 [50:00<24:27,  1.44it/s] 
4|4352|Loss: 0.841454029083252:  67%|██████▋   | 4352/6470 [50:01<24:03,  1.47it/s]
4|4353|Loss: 0.9657116532325745:  67%|██████▋   | 4352/6470 [50:01<24:03,  1.47it/s]
4|4353|Loss: 0.9657116532325745:  67%|██████▋   | 4353/6470 [50:02<25:06,  1.41it/s]
4|4354|Loss: 0.9082397222518921:  67%|██████▋   | 4353/6470 [50:02<25:06,  1.41it/s]
4|4354|Loss: 0.9082397222518921:  67%|██████▋   | 4354/6470 [50:02<25:28,  1.38it/s]
4|4355|Loss: 0.9877038598060608:  67%|██████▋   | 4354/6470 [50:03<25:28,  1.38it/s]
4|4355|Loss: 0.9877038598060608:  67%|██████▋   | 4355/6470 [50:03<25:36,  1.38it/s]
4|4356|Loss: 0.8030431270599365:  67%|██████▋   | 4355/6470 [50:03<25:36,  1.38it/s]
4|4356|Loss: 0.8030431270599365:  67%|██████▋   | 4356/6470 [50:04<24:52,  1.42it/s]
4|4357|Loss: 0.8989540934562683:  67%|██████▋   | 4356/6470 [50:04<24:52,  1.42it/s]
4|4357|Loss: 0.8989540934562683:  67%|██████▋   | 4357/6470 [50:04<23:24,  1.50it/s]
4|4358|Loss: 0.8956286907196045:  67%|██████▋   | 4357/6470 [50:05<23:24,  1.50it/s]
4|4358|Loss: 0.8956286907196045:  67%|██████▋   | 4358/6470 [50:05<21:25,  1.64it/s]
4|4359|Loss: 0.9199204444885254:  67%|██████▋   | 4358/6470 [50:05<21:25,  1.64it/s]
4|4359|Loss: 0.9199204444885254:  67%|██████▋   | 4359/6470 [50:05<21:21,  1.65it/s]
4|4360|Loss: 0.8223109841346741:  67%|██████▋   | 4359/6470 [50:06<21:21,  1.65it/s]
4|4360|Loss: 0.8223109841346741:  67%|██████▋   | 4360/6470 [50:06<22:19,  1.58it/s]
4|4361|Loss: 0.7993110418319702:  67%|██████▋   | 4360/6470 [50:06<22:19,  1.58it/s]
4|4361|Loss: 0.7993110418319702:  67%|██████▋   | 4361/6470 [50:07<25:14,  1.39it/s]
4|4362|Loss: 1.0477849245071411:  67%|██████▋   | 4361/6470 [50:07<25:14,  1.39it/s]
4|4362|Loss: 1.0477849245071411:  67%|██████▋   | 4362/6470 [50:08<26:58,  1.30it/s]
4|4363|Loss: 0.8765361309051514:  67%|██████▋   | 4362/6470 [50:08<26:58,  1.30it/s]
4|4363|Loss: 0.8765361309051514:  67%|██████▋   | 4363/6470 [50:09<26:55,  1.30it/s]
4|4364|Loss: 0.764451801776886:  67%|██████▋   | 4363/6470 [50:09<26:55,  1.30it/s] 
4|4364|Loss: 0.764451801776886:  67%|██████▋   | 4364/6470 [50:09<25:35,  1.37it/s]
4|4365|Loss: 0.8776881694793701:  67%|██████▋   | 4364/6470 [50:10<25:35,  1.37it/s]
4|4365|Loss: 0.8776881694793701:  67%|██████▋   | 4365/6470 [50:10<25:58,  1.35it/s]
4|4366|Loss: 0.8722429871559143:  67%|██████▋   | 4365/6470 [50:10<25:58,  1.35it/s]
4|4366|Loss: 0.8722429871559143:  67%|██████▋   | 4366/6470 [50:11<26:10,  1.34it/s]
4|4367|Loss: 0.7369515895843506:  67%|██████▋   | 4366/6470 [50:11<26:10,  1.34it/s]
4|4367|Loss: 0.7369515895843506:  67%|██████▋   | 4367/6470 [50:11<25:20,  1.38it/s]
4|4368|Loss: 0.9081917405128479:  67%|██████▋   | 4367/6470 [50:12<25:20,  1.38it/s]
4|4368|Loss: 0.9081917405128479:  68%|██████▊   | 4368/6470 [50:12<24:16,  1.44it/s]
4|4369|Loss: 0.7265067100524902:  68%|██████▊   | 4368/6470 [50:12<24:16,  1.44it/s]
4|4369|Loss: 0.7265067100524902:  68%|██████▊   | 4369/6470 [50:13<24:38,  1.42it/s]
4|4370|Loss: 0.8263580799102783:  68%|██████▊   | 4369/6470 [50:13<24:38,  1.42it/s]
4|4370|Loss: 0.8263580799102783:  68%|██████▊   | 4370/6470 [50:13<23:40,  1.48it/s]
4|4371|Loss: 0.7298837304115295:  68%|██████▊   | 4370/6470 [50:14<23:40,  1.48it/s]
4|4371|Loss: 0.7298837304115295:  68%|██████▊   | 4371/6470 [50:14<25:11,  1.39it/s]
4|4372|Loss: 0.8778977394104004:  68%|██████▊   | 4371/6470 [50:15<25:11,  1.39it/s]
4|4372|Loss: 0.8778977394104004:  68%|██████▊   | 4372/6470 [50:15<24:56,  1.40it/s]
4|4373|Loss: 0.9455558061599731:  68%|██████▊   | 4372/6470 [50:15<24:56,  1.40it/s]
4|4373|Loss: 0.9455558061599731:  68%|██████▊   | 4373/6470 [50:15<23:56,  1.46it/s]
4|4374|Loss: 0.8888671398162842:  68%|██████▊   | 4373/6470 [50:16<23:56,  1.46it/s]
4|4374|Loss: 0.8888671398162842:  68%|██████▊   | 4374/6470 [50:16<24:19,  1.44it/s]
4|4375|Loss: 0.8806811571121216:  68%|██████▊   | 4374/6470 [50:17<24:19,  1.44it/s]
4|4375|Loss: 0.8806811571121216:  68%|██████▊   | 4375/6470 [50:17<23:28,  1.49it/s]
4|4376|Loss: 1.0745540857315063:  68%|██████▊   | 4375/6470 [50:17<23:28,  1.49it/s]
4|4376|Loss: 1.0745540857315063:  68%|██████▊   | 4376/6470 [50:17<21:27,  1.63it/s]
4|4377|Loss: 0.7435641288757324:  68%|██████▊   | 4376/6470 [50:18<21:27,  1.63it/s]
4|4377|Loss: 0.7435641288757324:  68%|██████▊   | 4377/6470 [50:18<21:50,  1.60it/s]
4|4378|Loss: 0.9361343383789062:  68%|██████▊   | 4377/6470 [50:18<21:50,  1.60it/s]
4|4378|Loss: 0.9361343383789062:  68%|██████▊   | 4378/6470 [50:19<22:11,  1.57it/s]
4|4379|Loss: 0.8747238516807556:  68%|██████▊   | 4378/6470 [50:19<22:11,  1.57it/s]
4|4379|Loss: 0.8747238516807556:  68%|██████▊   | 4379/6470 [50:19<23:48,  1.46it/s]
4|4380|Loss: 0.8870347142219543:  68%|██████▊   | 4379/6470 [50:20<23:48,  1.46it/s]
4|4380|Loss: 0.8870347142219543:  68%|██████▊   | 4380/6470 [50:20<23:13,  1.50it/s]
4|4381|Loss: 0.8211275935173035:  68%|██████▊   | 4380/6470 [50:20<23:13,  1.50it/s]
4|4381|Loss: 0.8211275935173035:  68%|██████▊   | 4381/6470 [50:21<24:06,  1.44it/s]
4|4382|Loss: 0.8201189637184143:  68%|██████▊   | 4381/6470 [50:21<24:06,  1.44it/s]
4|4382|Loss: 0.8201189637184143:  68%|██████▊   | 4382/6470 [50:21<23:26,  1.48it/s]
4|4383|Loss: 0.7419016361236572:  68%|██████▊   | 4382/6470 [50:22<23:26,  1.48it/s]
4|4383|Loss: 0.7419016361236572:  68%|██████▊   | 4383/6470 [50:22<25:00,  1.39it/s]
4|4384|Loss: 0.7087686657905579:  68%|██████▊   | 4383/6470 [50:23<25:00,  1.39it/s]
4|4384|Loss: 0.7087686657905579:  68%|██████▊   | 4384/6470 [50:23<23:13,  1.50it/s]
4|4385|Loss: 0.8919483423233032:  68%|██████▊   | 4384/6470 [50:23<23:13,  1.50it/s]
4|4385|Loss: 0.8919483423233032:  68%|██████▊   | 4385/6470 [50:23<22:14,  1.56it/s]
4|4386|Loss: 0.7659514546394348:  68%|██████▊   | 4385/6470 [50:24<22:14,  1.56it/s]
4|4386|Loss: 0.7659514546394348:  68%|██████▊   | 4386/6470 [50:24<23:07,  1.50it/s]
4|4387|Loss: 0.6561182737350464:  68%|██████▊   | 4386/6470 [50:24<23:07,  1.50it/s]
4|4387|Loss: 0.6561182737350464:  68%|██████▊   | 4387/6470 [50:25<22:52,  1.52it/s]
4|4388|Loss: 0.9087164402008057:  68%|██████▊   | 4387/6470 [50:25<22:52,  1.52it/s]
4|4388|Loss: 0.9087164402008057:  68%|██████▊   | 4388/6470 [50:26<24:25,  1.42it/s]
4|4389|Loss: 0.9097620844841003:  68%|██████▊   | 4388/6470 [50:26<24:25,  1.42it/s]
4|4389|Loss: 0.9097620844841003:  68%|██████▊   | 4389/6470 [50:26<25:02,  1.38it/s]
4|4390|Loss: 0.9175925850868225:  68%|██████▊   | 4389/6470 [50:27<25:02,  1.38it/s]
4|4390|Loss: 0.9175925850868225:  68%|██████▊   | 4390/6470 [50:27<23:53,  1.45it/s]
4|4391|Loss: 0.8427369594573975:  68%|██████▊   | 4390/6470 [50:27<23:53,  1.45it/s]
4|4391|Loss: 0.8427369594573975:  68%|██████▊   | 4391/6470 [50:28<24:15,  1.43it/s]
4|4392|Loss: 0.7907987833023071:  68%|██████▊   | 4391/6470 [50:28<24:15,  1.43it/s]
4|4392|Loss: 0.7907987833023071:  68%|██████▊   | 4392/6470 [50:28<22:58,  1.51it/s]
4|4393|Loss: 0.8216391801834106:  68%|██████▊   | 4392/6470 [50:29<22:58,  1.51it/s]
4|4393|Loss: 0.8216391801834106:  68%|██████▊   | 4393/6470 [50:29<23:27,  1.48it/s]
4|4394|Loss: 0.7408694624900818:  68%|██████▊   | 4393/6470 [50:29<23:27,  1.48it/s]
4|4394|Loss: 0.7408694624900818:  68%|██████▊   | 4394/6470 [50:30<24:05,  1.44it/s]
4|4395|Loss: 0.7916402816772461:  68%|██████▊   | 4394/6470 [50:30<24:05,  1.44it/s]
4|4395|Loss: 0.7916402816772461:  68%|██████▊   | 4395/6470 [50:31<25:30,  1.36it/s]
4|4396|Loss: 0.776521623134613:  68%|██████▊   | 4395/6470 [50:31<25:30,  1.36it/s] 
4|4396|Loss: 0.776521623134613:  68%|██████▊   | 4396/6470 [50:31<25:01,  1.38it/s]
4|4397|Loss: 0.9560453295707703:  68%|██████▊   | 4396/6470 [50:32<25:01,  1.38it/s]
4|4397|Loss: 0.9560453295707703:  68%|██████▊   | 4397/6470 [50:32<23:58,  1.44it/s]
4|4398|Loss: 0.8567553758621216:  68%|██████▊   | 4397/6470 [50:32<23:58,  1.44it/s]
4|4398|Loss: 0.8567553758621216:  68%|██████▊   | 4398/6470 [50:33<24:07,  1.43it/s]
4|4399|Loss: 0.7354570627212524:  68%|██████▊   | 4398/6470 [50:33<24:07,  1.43it/s]
4|4399|Loss: 0.7354570627212524:  68%|██████▊   | 4399/6470 [50:33<25:00,  1.38it/s]
4|4400|Loss: 0.9769583940505981:  68%|██████▊   | 4399/6470 [50:34<25:00,  1.38it/s]
4|4400|Loss: 0.9769583940505981:  68%|██████▊   | 4400/6470 [50:34<24:16,  1.42it/s]
4|4401|Loss: 0.775583028793335:  68%|██████▊   | 4400/6470 [50:34<24:16,  1.42it/s] 
4|4401|Loss: 0.775583028793335:  68%|██████▊   | 4401/6470 [50:35<24:44,  1.39it/s]
4|4402|Loss: 0.824725329875946:  68%|██████▊   | 4401/6470 [50:35<24:44,  1.39it/s]
4|4402|Loss: 0.824725329875946:  68%|██████▊   | 4402/6470 [50:36<26:12,  1.32it/s]
4|4403|Loss: 0.8931930065155029:  68%|██████▊   | 4402/6470 [50:36<26:12,  1.32it/s]
4|4403|Loss: 0.8931930065155029:  68%|██████▊   | 4403/6470 [50:36<25:07,  1.37it/s]
4|4404|Loss: 0.8493110537528992:  68%|██████▊   | 4403/6470 [50:37<25:07,  1.37it/s]
4|4404|Loss: 0.8493110537528992:  68%|██████▊   | 4404/6470 [50:37<22:45,  1.51it/s]
4|4405|Loss: 0.8535102605819702:  68%|██████▊   | 4404/6470 [50:37<22:45,  1.51it/s]
4|4405|Loss: 0.8535102605819702:  68%|██████▊   | 4405/6470 [50:37<23:00,  1.50it/s]
4|4406|Loss: 0.8101459741592407:  68%|██████▊   | 4405/6470 [50:38<23:00,  1.50it/s]
4|4406|Loss: 0.8101459741592407:  68%|██████▊   | 4406/6470 [50:38<22:14,  1.55it/s]
4|4407|Loss: 0.962783932685852:  68%|██████▊   | 4406/6470 [50:38<22:14,  1.55it/s] 
4|4407|Loss: 0.962783932685852:  68%|██████▊   | 4407/6470 [50:39<21:58,  1.56it/s]
4|4408|Loss: 0.8066208958625793:  68%|██████▊   | 4407/6470 [50:39<21:58,  1.56it/s]
4|4408|Loss: 0.8066208958625793:  68%|██████▊   | 4408/6470 [50:39<22:32,  1.52it/s]
4|4409|Loss: 0.9316547513008118:  68%|██████▊   | 4408/6470 [50:40<22:32,  1.52it/s]
4|4409|Loss: 0.9316547513008118:  68%|██████▊   | 4409/6470 [50:40<24:06,  1.42it/s]
4|4410|Loss: 0.9143887758255005:  68%|██████▊   | 4409/6470 [50:41<24:06,  1.42it/s]
4|4410|Loss: 0.9143887758255005:  68%|██████▊   | 4410/6470 [50:41<23:45,  1.45it/s]
4|4411|Loss: 0.7759113311767578:  68%|██████▊   | 4410/6470 [50:41<23:45,  1.45it/s]
4|4411|Loss: 0.7759113311767578:  68%|██████▊   | 4411/6470 [50:41<22:37,  1.52it/s]
4|4412|Loss: 0.757459282875061:  68%|██████▊   | 4411/6470 [50:42<22:37,  1.52it/s] 
4|4412|Loss: 0.757459282875061:  68%|██████▊   | 4412/6470 [50:42<24:05,  1.42it/s]
4|4413|Loss: 0.7845170497894287:  68%|██████▊   | 4412/6470 [50:43<24:05,  1.42it/s]
4|4413|Loss: 0.7845170497894287:  68%|██████▊   | 4413/6470 [50:43<24:49,  1.38it/s]
4|4414|Loss: 0.9800362586975098:  68%|██████▊   | 4413/6470 [50:43<24:49,  1.38it/s]
4|4414|Loss: 0.9800362586975098:  68%|██████▊   | 4414/6470 [50:44<24:58,  1.37it/s]
4|4415|Loss: 0.9460758566856384:  68%|██████▊   | 4414/6470 [50:44<24:58,  1.37it/s]
4|4415|Loss: 0.9460758566856384:  68%|██████▊   | 4415/6470 [50:44<23:08,  1.48it/s]
4|4416|Loss: 0.9471723437309265:  68%|██████▊   | 4415/6470 [50:45<23:08,  1.48it/s]
4|4416|Loss: 0.9471723437309265:  68%|██████▊   | 4416/6470 [50:45<23:22,  1.46it/s]
4|4417|Loss: 0.6868957877159119:  68%|██████▊   | 4416/6470 [50:45<23:22,  1.46it/s]
4|4417|Loss: 0.6868957877159119:  68%|██████▊   | 4417/6470 [50:46<26:16,  1.30it/s]
4|4418|Loss: 0.7665352821350098:  68%|██████▊   | 4417/6470 [50:46<26:16,  1.30it/s]
4|4418|Loss: 0.7665352821350098:  68%|██████▊   | 4418/6470 [50:47<25:05,  1.36it/s]
4|4419|Loss: 0.9930862188339233:  68%|██████▊   | 4418/6470 [50:47<25:05,  1.36it/s]
4|4419|Loss: 0.9930862188339233:  68%|██████▊   | 4419/6470 [50:47<23:55,  1.43it/s]
4|4420|Loss: 0.8048067092895508:  68%|██████▊   | 4419/6470 [50:48<23:55,  1.43it/s]
4|4420|Loss: 0.8048067092895508:  68%|██████▊   | 4420/6470 [50:48<24:35,  1.39it/s]
4|4421|Loss: 0.8852115273475647:  68%|██████▊   | 4420/6470 [50:48<24:35,  1.39it/s]
4|4421|Loss: 0.8852115273475647:  68%|██████▊   | 4421/6470 [50:49<24:56,  1.37it/s]
4|4422|Loss: 0.7964687347412109:  68%|██████▊   | 4421/6470 [50:49<24:56,  1.37it/s]
4|4422|Loss: 0.7964687347412109:  68%|██████▊   | 4422/6470 [50:50<25:08,  1.36it/s]
4|4423|Loss: 0.8004652261734009:  68%|██████▊   | 4422/6470 [50:50<25:08,  1.36it/s]
4|4423|Loss: 0.8004652261734009:  68%|██████▊   | 4423/6470 [50:50<25:10,  1.35it/s]
4|4424|Loss: 0.7611228823661804:  68%|██████▊   | 4423/6470 [50:51<25:10,  1.35it/s]
4|4424|Loss: 0.7611228823661804:  68%|██████▊   | 4424/6470 [50:51<25:41,  1.33it/s]
4|4425|Loss: 0.8701388835906982:  68%|██████▊   | 4424/6470 [50:51<25:41,  1.33it/s]
4|4425|Loss: 0.8701388835906982:  68%|██████▊   | 4425/6470 [50:52<24:45,  1.38it/s]
4|4426|Loss: 1.0890405178070068:  68%|██████▊   | 4425/6470 [50:52<24:45,  1.38it/s]
4|4426|Loss: 1.0890405178070068:  68%|██████▊   | 4426/6470 [50:52<23:24,  1.45it/s]
4|4427|Loss: 0.955902099609375:  68%|██████▊   | 4426/6470 [50:53<23:24,  1.45it/s] 
4|4427|Loss: 0.955902099609375:  68%|██████▊   | 4427/6470 [50:53<23:43,  1.44it/s]
4|4428|Loss: 0.9535766839981079:  68%|██████▊   | 4427/6470 [50:53<23:43,  1.44it/s]
4|4428|Loss: 0.9535766839981079:  68%|██████▊   | 4428/6470 [50:54<22:26,  1.52it/s]
4|4429|Loss: 0.8772927522659302:  68%|██████▊   | 4428/6470 [50:54<22:26,  1.52it/s]
4|4429|Loss: 0.8772927522659302:  68%|██████▊   | 4429/6470 [50:54<22:33,  1.51it/s]
4|4430|Loss: 0.7300875186920166:  68%|██████▊   | 4429/6470 [50:55<22:33,  1.51it/s]
4|4430|Loss: 0.7300875186920166:  68%|██████▊   | 4430/6470 [50:55<21:32,  1.58it/s]
4|4431|Loss: 0.898024320602417:  68%|██████▊   | 4430/6470 [50:55<21:32,  1.58it/s] 
4|4431|Loss: 0.898024320602417:  68%|██████▊   | 4431/6470 [50:56<22:06,  1.54it/s]
4|4432|Loss: 0.7923678755760193:  68%|██████▊   | 4431/6470 [50:56<22:06,  1.54it/s]
4|4432|Loss: 0.7923678755760193:  69%|██████▊   | 4432/6470 [50:56<23:52,  1.42it/s]
4|4433|Loss: 0.8499745726585388:  69%|██████▊   | 4432/6470 [50:57<23:52,  1.42it/s]
4|4433|Loss: 0.8499745726585388:  69%|██████▊   | 4433/6470 [50:57<24:17,  1.40it/s]
4|4434|Loss: 0.8976961374282837:  69%|██████▊   | 4433/6470 [50:57<24:17,  1.40it/s]
4|4434|Loss: 0.8976961374282837:  69%|██████▊   | 4434/6470 [50:58<23:08,  1.47it/s]
4|4435|Loss: 0.8065029978752136:  69%|██████▊   | 4434/6470 [50:58<23:08,  1.47it/s]
4|4435|Loss: 0.8065029978752136:  69%|██████▊   | 4435/6470 [50:58<22:24,  1.51it/s]
4|4436|Loss: 0.884679913520813:  69%|██████▊   | 4435/6470 [50:59<22:24,  1.51it/s] 
4|4436|Loss: 0.884679913520813:  69%|██████▊   | 4436/6470 [50:59<20:54,  1.62it/s]
4|4437|Loss: 0.9641660451889038:  69%|██████▊   | 4436/6470 [50:59<20:54,  1.62it/s]
4|4437|Loss: 0.9641660451889038:  69%|██████▊   | 4437/6470 [50:59<21:15,  1.59it/s]
4|4438|Loss: 0.8159005641937256:  69%|██████▊   | 4437/6470 [51:00<21:15,  1.59it/s]
4|4438|Loss: 0.8159005641937256:  69%|██████▊   | 4438/6470 [51:00<21:07,  1.60it/s]
4|4439|Loss: 0.91697096824646:  69%|██████▊   | 4438/6470 [51:00<21:07,  1.60it/s]  
4|4439|Loss: 0.91697096824646:  69%|██████▊   | 4439/6470 [51:01<21:09,  1.60it/s]
4|4440|Loss: 0.9235385656356812:  69%|██████▊   | 4439/6470 [51:01<21:09,  1.60it/s]
4|4440|Loss: 0.9235385656356812:  69%|██████▊   | 4440/6470 [51:01<22:49,  1.48it/s]
4|4441|Loss: 0.7812525629997253:  69%|██████▊   | 4440/6470 [51:02<22:49,  1.48it/s]
4|4441|Loss: 0.7812525629997253:  69%|██████▊   | 4441/6470 [51:02<24:23,  1.39it/s]
4|4442|Loss: 0.954553484916687:  69%|██████▊   | 4441/6470 [51:03<24:23,  1.39it/s] 
4|4442|Loss: 0.954553484916687:  69%|██████▊   | 4442/6470 [51:03<24:46,  1.36it/s]
4|4443|Loss: 1.0891207456588745:  69%|██████▊   | 4442/6470 [51:03<24:46,  1.36it/s]
4|4443|Loss: 1.0891207456588745:  69%|██████▊   | 4443/6470 [51:04<24:07,  1.40it/s]
4|4444|Loss: 0.7976869940757751:  69%|██████▊   | 4443/6470 [51:04<24:07,  1.40it/s]
4|4444|Loss: 0.7976869940757751:  69%|██████▊   | 4444/6470 [51:04<23:20,  1.45it/s]
4|4445|Loss: 0.7770623564720154:  69%|██████▊   | 4444/6470 [51:05<23:20,  1.45it/s]
4|4445|Loss: 0.7770623564720154:  69%|██████▊   | 4445/6470 [51:05<23:36,  1.43it/s]
4|4446|Loss: 0.8962391018867493:  69%|██████▊   | 4445/6470 [51:06<23:36,  1.43it/s]
4|4446|Loss: 0.8962391018867493:  69%|██████▊   | 4446/6470 [51:06<23:47,  1.42it/s]
4|4447|Loss: 0.8194432854652405:  69%|██████▊   | 4446/6470 [51:06<23:47,  1.42it/s]
4|4447|Loss: 0.8194432854652405:  69%|██████▊   | 4447/6470 [51:06<22:13,  1.52it/s]
4|4448|Loss: 0.7680251002311707:  69%|██████▊   | 4447/6470 [51:07<22:13,  1.52it/s]
4|4448|Loss: 0.7680251002311707:  69%|██████▊   | 4448/6470 [51:07<22:18,  1.51it/s]
4|4449|Loss: 0.9230664968490601:  69%|██████▊   | 4448/6470 [51:07<22:18,  1.51it/s]
4|4449|Loss: 0.9230664968490601:  69%|██████▉   | 4449/6470 [51:08<22:06,  1.52it/s]
4|4450|Loss: 0.9191620349884033:  69%|██████▉   | 4449/6470 [51:08<22:06,  1.52it/s]
4|4450|Loss: 0.9191620349884033:  69%|██████▉   | 4450/6470 [51:08<23:04,  1.46it/s]
4|4451|Loss: 0.9195088744163513:  69%|██████▉   | 4450/6470 [51:09<23:04,  1.46it/s]
4|4451|Loss: 0.9195088744163513:  69%|██████▉   | 4451/6470 [51:09<22:43,  1.48it/s]
4|4452|Loss: 0.9875311255455017:  69%|██████▉   | 4451/6470 [51:09<22:43,  1.48it/s]
4|4452|Loss: 0.9875311255455017:  69%|██████▉   | 4452/6470 [51:10<22:14,  1.51it/s]
4|4453|Loss: 0.9388822317123413:  69%|██████▉   | 4452/6470 [51:10<22:14,  1.51it/s]
4|4453|Loss: 0.9388822317123413:  69%|██████▉   | 4453/6470 [51:10<21:59,  1.53it/s]
4|4454|Loss: 0.7735918164253235:  69%|██████▉   | 4453/6470 [51:11<21:59,  1.53it/s]
4|4454|Loss: 0.7735918164253235:  69%|██████▉   | 4454/6470 [51:11<21:25,  1.57it/s]
4|4455|Loss: 0.8256568312644958:  69%|██████▉   | 4454/6470 [51:11<21:25,  1.57it/s]
4|4455|Loss: 0.8256568312644958:  69%|██████▉   | 4455/6470 [51:12<23:00,  1.46it/s]
4|4456|Loss: 0.8053279519081116:  69%|██████▉   | 4455/6470 [51:12<23:00,  1.46it/s]
4|4456|Loss: 0.8053279519081116:  69%|██████▉   | 4456/6470 [51:13<23:33,  1.42it/s]
4|4457|Loss: 0.7730520963668823:  69%|██████▉   | 4456/6470 [51:13<23:33,  1.42it/s]
4|4457|Loss: 0.7730520963668823:  69%|██████▉   | 4457/6470 [51:13<24:31,  1.37it/s]
4|4458|Loss: 0.8512037992477417:  69%|██████▉   | 4457/6470 [51:14<24:31,  1.37it/s]
4|4458|Loss: 0.8512037992477417:  69%|██████▉   | 4458/6470 [51:14<24:27,  1.37it/s]
4|4459|Loss: 0.8045632243156433:  69%|██████▉   | 4458/6470 [51:14<24:27,  1.37it/s]
4|4459|Loss: 0.8045632243156433:  69%|██████▉   | 4459/6470 [51:15<23:38,  1.42it/s]
4|4460|Loss: 0.8390247225761414:  69%|██████▉   | 4459/6470 [51:15<23:38,  1.42it/s]
4|4460|Loss: 0.8390247225761414:  69%|██████▉   | 4460/6470 [51:15<22:47,  1.47it/s]
4|4461|Loss: 0.8825727701187134:  69%|██████▉   | 4460/6470 [51:16<22:47,  1.47it/s]
4|4461|Loss: 0.8825727701187134:  69%|██████▉   | 4461/6470 [51:16<23:39,  1.41it/s]
4|4462|Loss: 0.8950175642967224:  69%|██████▉   | 4461/6470 [51:16<23:39,  1.41it/s]
4|4462|Loss: 0.8950175642967224:  69%|██████▉   | 4462/6470 [51:17<23:10,  1.44it/s]
4|4463|Loss: 0.819124698638916:  69%|██████▉   | 4462/6470 [51:17<23:10,  1.44it/s] 
4|4463|Loss: 0.819124698638916:  69%|██████▉   | 4463/6470 [51:17<21:21,  1.57it/s]
4|4464|Loss: 0.8937012553215027:  69%|██████▉   | 4463/6470 [51:18<21:21,  1.57it/s]
4|4464|Loss: 0.8937012553215027:  69%|██████▉   | 4464/6470 [51:18<22:58,  1.46it/s]
4|4465|Loss: 0.7805197834968567:  69%|██████▉   | 4464/6470 [51:19<22:58,  1.46it/s]
4|4465|Loss: 0.7805197834968567:  69%|██████▉   | 4465/6470 [51:19<25:15,  1.32it/s]
4|4466|Loss: 1.034263014793396:  69%|██████▉   | 4465/6470 [51:19<25:15,  1.32it/s] 
4|4466|Loss: 1.034263014793396:  69%|██████▉   | 4466/6470 [51:20<23:12,  1.44it/s]
4|4467|Loss: 0.7031592726707458:  69%|██████▉   | 4466/6470 [51:20<23:12,  1.44it/s]
4|4467|Loss: 0.7031592726707458:  69%|██████▉   | 4467/6470 [51:20<22:54,  1.46it/s]
4|4468|Loss: 0.6968472599983215:  69%|██████▉   | 4467/6470 [51:21<22:54,  1.46it/s]
4|4468|Loss: 0.6968472599983215:  69%|██████▉   | 4468/6470 [51:21<24:07,  1.38it/s]
4|4469|Loss: 0.8275812864303589:  69%|██████▉   | 4468/6470 [51:21<24:07,  1.38it/s]
4|4469|Loss: 0.8275812864303589:  69%|██████▉   | 4469/6470 [51:22<25:34,  1.30it/s]
4|4470|Loss: 0.7988432049751282:  69%|██████▉   | 4469/6470 [51:22<25:34,  1.30it/s]
4|4470|Loss: 0.7988432049751282:  69%|██████▉   | 4470/6470 [51:23<26:21,  1.26it/s]
4|4471|Loss: 0.7752504944801331:  69%|██████▉   | 4470/6470 [51:23<26:21,  1.26it/s]
4|4471|Loss: 0.7752504944801331:  69%|██████▉   | 4471/6470 [51:23<25:56,  1.28it/s]
4|4472|Loss: 0.9194456338882446:  69%|██████▉   | 4471/6470 [51:24<25:56,  1.28it/s]
4|4472|Loss: 0.9194456338882446:  69%|██████▉   | 4472/6470 [51:24<23:41,  1.41it/s]
4|4473|Loss: 0.9704551100730896:  69%|██████▉   | 4472/6470 [51:24<23:41,  1.41it/s]
4|4473|Loss: 0.9704551100730896:  69%|██████▉   | 4473/6470 [51:25<24:13,  1.37it/s]
4|4474|Loss: 0.8511502742767334:  69%|██████▉   | 4473/6470 [51:25<24:13,  1.37it/s]
4|4474|Loss: 0.8511502742767334:  69%|██████▉   | 4474/6470 [51:25<23:32,  1.41it/s]
4|4475|Loss: 0.9104996919631958:  69%|██████▉   | 4474/6470 [51:26<23:32,  1.41it/s]
4|4475|Loss: 0.9104996919631958:  69%|██████▉   | 4475/6470 [51:26<23:23,  1.42it/s]
4|4476|Loss: 0.7929596900939941:  69%|██████▉   | 4475/6470 [51:26<23:23,  1.42it/s]
4|4476|Loss: 0.7929596900939941:  69%|██████▉   | 4476/6470 [51:27<22:23,  1.48it/s]
4|4477|Loss: 0.9504302144050598:  69%|██████▉   | 4476/6470 [51:27<22:23,  1.48it/s]
4|4477|Loss: 0.9504302144050598:  69%|██████▉   | 4477/6470 [51:27<23:11,  1.43it/s]
4|4478|Loss: 0.7879992127418518:  69%|██████▉   | 4477/6470 [51:28<23:11,  1.43it/s]
4|4478|Loss: 0.7879992127418518:  69%|██████▉   | 4478/6470 [51:28<22:56,  1.45it/s]
4|4479|Loss: 0.8388940691947937:  69%|██████▉   | 4478/6470 [51:29<22:56,  1.45it/s]
4|4479|Loss: 0.8388940691947937:  69%|██████▉   | 4479/6470 [51:29<22:13,  1.49it/s]
4|4480|Loss: 0.9644992351531982:  69%|██████▉   | 4479/6470 [51:29<22:13,  1.49it/s]
4|4480|Loss: 0.9644992351531982:  69%|██████▉   | 4480/6470 [51:29<21:40,  1.53it/s]
4|4481|Loss: 0.8839465379714966:  69%|██████▉   | 4480/6470 [51:30<21:40,  1.53it/s]
4|4481|Loss: 0.8839465379714966:  69%|██████▉   | 4481/6470 [51:30<21:31,  1.54it/s]
4|4482|Loss: 0.8298225402832031:  69%|██████▉   | 4481/6470 [51:30<21:31,  1.54it/s]
4|4482|Loss: 0.8298225402832031:  69%|██████▉   | 4482/6470 [51:31<21:30,  1.54it/s]
4|4483|Loss: 0.9475154280662537:  69%|██████▉   | 4482/6470 [51:31<21:30,  1.54it/s]
4|4483|Loss: 0.9475154280662537:  69%|██████▉   | 4483/6470 [51:31<22:20,  1.48it/s]
4|4484|Loss: 1.0212836265563965:  69%|██████▉   | 4483/6470 [51:32<22:20,  1.48it/s]
4|4484|Loss: 1.0212836265563965:  69%|██████▉   | 4484/6470 [51:32<22:16,  1.49it/s]
4|4485|Loss: 0.8719453811645508:  69%|██████▉   | 4484/6470 [51:32<22:16,  1.49it/s]
4|4485|Loss: 0.8719453811645508:  69%|██████▉   | 4485/6470 [51:33<21:56,  1.51it/s]
4|4486|Loss: 1.0022644996643066:  69%|██████▉   | 4485/6470 [51:33<21:56,  1.51it/s]
4|4486|Loss: 1.0022644996643066:  69%|██████▉   | 4486/6470 [51:33<21:43,  1.52it/s]
4|4487|Loss: 0.865595817565918:  69%|██████▉   | 4486/6470 [51:34<21:43,  1.52it/s] 
4|4487|Loss: 0.865595817565918:  69%|██████▉   | 4487/6470 [51:34<23:04,  1.43it/s]
4|4488|Loss: 0.8435850739479065:  69%|██████▉   | 4487/6470 [51:35<23:04,  1.43it/s]
4|4488|Loss: 0.8435850739479065:  69%|██████▉   | 4488/6470 [51:35<25:00,  1.32it/s]
4|4489|Loss: 0.8931689858436584:  69%|██████▉   | 4488/6470 [51:35<25:00,  1.32it/s]
4|4489|Loss: 0.8931689858436584:  69%|██████▉   | 4489/6470 [51:36<23:33,  1.40it/s]
4|4490|Loss: 0.752346932888031:  69%|██████▉   | 4489/6470 [51:36<23:33,  1.40it/s] 
4|4490|Loss: 0.752346932888031:  69%|██████▉   | 4490/6470 [51:36<22:35,  1.46it/s]
4|4491|Loss: 0.8794933557510376:  69%|██████▉   | 4490/6470 [51:37<22:35,  1.46it/s]
4|4491|Loss: 0.8794933557510376:  69%|██████▉   | 4491/6470 [51:37<22:55,  1.44it/s]
4|4492|Loss: 0.7906728386878967:  69%|██████▉   | 4491/6470 [51:37<22:55,  1.44it/s]
4|4492|Loss: 0.7906728386878967:  69%|██████▉   | 4492/6470 [51:38<23:16,  1.42it/s]
4|4493|Loss: 0.7166542410850525:  69%|██████▉   | 4492/6470 [51:38<23:16,  1.42it/s]
4|4493|Loss: 0.7166542410850525:  69%|██████▉   | 4493/6470 [51:39<24:01,  1.37it/s]
4|4494|Loss: 0.9074644446372986:  69%|██████▉   | 4493/6470 [51:39<24:01,  1.37it/s]
4|4494|Loss: 0.9074644446372986:  69%|██████▉   | 4494/6470 [51:39<23:55,  1.38it/s]
4|4495|Loss: 0.7960349917411804:  69%|██████▉   | 4494/6470 [51:40<23:55,  1.38it/s]
4|4495|Loss: 0.7960349917411804:  69%|██████▉   | 4495/6470 [51:40<23:58,  1.37it/s]
4|4496|Loss: 0.8359219431877136:  69%|██████▉   | 4495/6470 [51:40<23:58,  1.37it/s]
4|4496|Loss: 0.8359219431877136:  69%|██████▉   | 4496/6470 [51:41<24:12,  1.36it/s]
4|4497|Loss: 0.9534873962402344:  69%|██████▉   | 4496/6470 [51:41<24:12,  1.36it/s]
4|4497|Loss: 0.9534873962402344:  70%|██████▉   | 4497/6470 [51:41<24:22,  1.35it/s]
4|4498|Loss: 0.8265590071678162:  70%|██████▉   | 4497/6470 [51:42<24:22,  1.35it/s]
4|4498|Loss: 0.8265590071678162:  70%|██████▉   | 4498/6470 [51:42<23:25,  1.40it/s]
4|4499|Loss: 0.7675435543060303:  70%|██████▉   | 4498/6470 [51:42<23:25,  1.40it/s]
4|4499|Loss: 0.7675435543060303:  70%|██████▉   | 4499/6470 [51:43<22:06,  1.49it/s]
4|4500|Loss: 0.6855849623680115:  70%|██████▉   | 4499/6470 [51:43<22:06,  1.49it/s]
4|4500|Loss: 0.6855849623680115:  70%|██████▉   | 4500/6470 [51:44<23:46,  1.38it/s]
4|4501|Loss: 0.7691370844841003:  70%|██████▉   | 4500/6470 [51:44<23:46,  1.38it/s]
4|4501|Loss: 0.7691370844841003:  70%|██████▉   | 4501/6470 [51:44<24:55,  1.32it/s]
4|4502|Loss: 0.889499843120575:  70%|██████▉   | 4501/6470 [51:45<24:55,  1.32it/s] 
4|4502|Loss: 0.889499843120575:  70%|██████▉   | 4502/6470 [51:45<24:48,  1.32it/s]
4|4503|Loss: 0.841326117515564:  70%|██████▉   | 4502/6470 [51:46<24:48,  1.32it/s]
4|4503|Loss: 0.841326117515564:  70%|██████▉   | 4503/6470 [51:46<23:24,  1.40it/s]
4|4504|Loss: 0.7574508190155029:  70%|██████▉   | 4503/6470 [51:46<23:24,  1.40it/s]
4|4504|Loss: 0.7574508190155029:  70%|██████▉   | 4504/6470 [51:46<22:19,  1.47it/s]
4|4505|Loss: 0.8301917910575867:  70%|██████▉   | 4504/6470 [51:47<22:19,  1.47it/s]
4|4505|Loss: 0.8301917910575867:  70%|██████▉   | 4505/6470 [51:47<20:38,  1.59it/s]
4|4506|Loss: 0.8095895051956177:  70%|██████▉   | 4505/6470 [51:47<20:38,  1.59it/s]
4|4506|Loss: 0.8095895051956177:  70%|██████▉   | 4506/6470 [51:47<19:54,  1.64it/s]
4|4507|Loss: 0.9584812521934509:  70%|██████▉   | 4506/6470 [51:48<19:54,  1.64it/s]
4|4507|Loss: 0.9584812521934509:  70%|██████▉   | 4507/6470 [51:48<19:11,  1.70it/s]
4|4508|Loss: 0.8503178954124451:  70%|██████▉   | 4507/6470 [51:48<19:11,  1.70it/s]
4|4508|Loss: 0.8503178954124451:  70%|██████▉   | 4508/6470 [51:49<20:07,  1.63it/s]
4|4509|Loss: 0.8433417677879333:  70%|██████▉   | 4508/6470 [51:49<20:07,  1.63it/s]
4|4509|Loss: 0.8433417677879333:  70%|██████▉   | 4509/6470 [51:49<20:33,  1.59it/s]
4|4510|Loss: 0.7378162145614624:  70%|██████▉   | 4509/6470 [51:50<20:33,  1.59it/s]
4|4510|Loss: 0.7378162145614624:  70%|██████▉   | 4510/6470 [51:50<21:09,  1.54it/s]
4|4511|Loss: 0.844809353351593:  70%|██████▉   | 4510/6470 [51:50<21:09,  1.54it/s] 
4|4511|Loss: 0.844809353351593:  70%|██████▉   | 4511/6470 [51:51<21:54,  1.49it/s]
4|4512|Loss: 0.7538424134254456:  70%|██████▉   | 4511/6470 [51:51<21:54,  1.49it/s]
4|4512|Loss: 0.7538424134254456:  70%|██████▉   | 4512/6470 [51:51<22:13,  1.47it/s]
4|4513|Loss: 0.7710492014884949:  70%|██████▉   | 4512/6470 [51:52<22:13,  1.47it/s]
4|4513|Loss: 0.7710492014884949:  70%|██████▉   | 4513/6470 [51:52<21:55,  1.49it/s]
4|4514|Loss: 0.8452832698822021:  70%|██████▉   | 4513/6470 [51:52<21:55,  1.49it/s]
4|4514|Loss: 0.8452832698822021:  70%|██████▉   | 4514/6470 [51:53<21:21,  1.53it/s]
4|4515|Loss: 0.7803671956062317:  70%|██████▉   | 4514/6470 [51:53<21:21,  1.53it/s]
4|4515|Loss: 0.7803671956062317:  70%|██████▉   | 4515/6470 [51:53<20:26,  1.59it/s]
4|4516|Loss: 0.8109829425811768:  70%|██████▉   | 4515/6470 [51:54<20:26,  1.59it/s]
4|4516|Loss: 0.8109829425811768:  70%|██████▉   | 4516/6470 [51:54<21:15,  1.53it/s]
4|4517|Loss: 0.9160956144332886:  70%|██████▉   | 4516/6470 [51:54<21:15,  1.53it/s]
4|4517|Loss: 0.9160956144332886:  70%|██████▉   | 4517/6470 [51:55<21:14,  1.53it/s]
4|4518|Loss: 0.8304955959320068:  70%|██████▉   | 4517/6470 [51:55<21:14,  1.53it/s]
4|4518|Loss: 0.8304955959320068:  70%|██████▉   | 4518/6470 [51:55<20:02,  1.62it/s]
4|4519|Loss: 1.0693081617355347:  70%|██████▉   | 4518/6470 [51:56<20:02,  1.62it/s]
4|4519|Loss: 1.0693081617355347:  70%|██████▉   | 4519/6470 [51:56<20:28,  1.59it/s]
4|4520|Loss: 0.9445161819458008:  70%|██████▉   | 4519/6470 [51:56<20:28,  1.59it/s]
4|4520|Loss: 0.9445161819458008:  70%|██████▉   | 4520/6470 [51:57<21:12,  1.53it/s]
4|4521|Loss: 0.9597280025482178:  70%|██████▉   | 4520/6470 [51:57<21:12,  1.53it/s]
4|4521|Loss: 0.9597280025482178:  70%|██████▉   | 4521/6470 [51:57<21:44,  1.49it/s]
4|4522|Loss: 0.8308113813400269:  70%|██████▉   | 4521/6470 [51:58<21:44,  1.49it/s]
4|4522|Loss: 0.8308113813400269:  70%|██████▉   | 4522/6470 [51:58<21:21,  1.52it/s]
4|4523|Loss: 0.8504016399383545:  70%|██████▉   | 4522/6470 [51:58<21:21,  1.52it/s]
4|4523|Loss: 0.8504016399383545:  70%|██████▉   | 4523/6470 [51:58<20:43,  1.57it/s]
4|4524|Loss: 0.9052096605300903:  70%|██████▉   | 4523/6470 [51:59<20:43,  1.57it/s]
4|4524|Loss: 0.9052096605300903:  70%|██████▉   | 4524/6470 [51:59<22:18,  1.45it/s]
4|4525|Loss: 0.944460928440094:  70%|██████▉   | 4524/6470 [52:00<22:18,  1.45it/s] 
4|4525|Loss: 0.944460928440094:  70%|██████▉   | 4525/6470 [52:00<23:02,  1.41it/s]
4|4526|Loss: 0.8962773680686951:  70%|██████▉   | 4525/6470 [52:00<23:02,  1.41it/s]
4|4526|Loss: 0.8962773680686951:  70%|██████▉   | 4526/6470 [52:01<21:38,  1.50it/s]
4|4527|Loss: 0.7341315150260925:  70%|██████▉   | 4526/6470 [52:01<21:38,  1.50it/s]
4|4527|Loss: 0.7341315150260925:  70%|██████▉   | 4527/6470 [52:01<20:58,  1.54it/s]
4|4528|Loss: 0.8180252313613892:  70%|██████▉   | 4527/6470 [52:02<20:58,  1.54it/s]
4|4528|Loss: 0.8180252313613892:  70%|██████▉   | 4528/6470 [52:02<20:52,  1.55it/s]
4|4529|Loss: 0.845702588558197:  70%|██████▉   | 4528/6470 [52:02<20:52,  1.55it/s] 
4|4529|Loss: 0.845702588558197:  70%|███████   | 4529/6470 [52:03<22:28,  1.44it/s]
4|4530|Loss: 0.8159728050231934:  70%|███████   | 4529/6470 [52:03<22:28,  1.44it/s]
4|4530|Loss: 0.8159728050231934:  70%|███████   | 4530/6470 [52:03<22:56,  1.41it/s]
4|4531|Loss: 0.826083242893219:  70%|███████   | 4530/6470 [52:04<22:56,  1.41it/s] 
4|4531|Loss: 0.826083242893219:  70%|███████   | 4531/6470 [52:04<23:07,  1.40it/s]
4|4532|Loss: 0.9018639922142029:  70%|███████   | 4531/6470 [52:04<23:07,  1.40it/s]
4|4532|Loss: 0.9018639922142029:  70%|███████   | 4532/6470 [52:05<22:34,  1.43it/s]
4|4533|Loss: 0.8366577625274658:  70%|███████   | 4532/6470 [52:05<22:34,  1.43it/s]
4|4533|Loss: 0.8366577625274658:  70%|███████   | 4533/6470 [52:05<22:40,  1.42it/s]
4|4534|Loss: 0.792203962802887:  70%|███████   | 4533/6470 [52:06<22:40,  1.42it/s] 
4|4534|Loss: 0.792203962802887:  70%|███████   | 4534/6470 [52:06<22:19,  1.45it/s]
4|4535|Loss: 0.9205098152160645:  70%|███████   | 4534/6470 [52:07<22:19,  1.45it/s]
4|4535|Loss: 0.9205098152160645:  70%|███████   | 4535/6470 [52:07<21:49,  1.48it/s]
4|4536|Loss: 0.7875399589538574:  70%|███████   | 4535/6470 [52:07<21:49,  1.48it/s]
4|4536|Loss: 0.7875399589538574:  70%|███████   | 4536/6470 [52:07<21:27,  1.50it/s]
4|4537|Loss: 0.7802667617797852:  70%|███████   | 4536/6470 [52:08<21:27,  1.50it/s]
4|4537|Loss: 0.7802667617797852:  70%|███████   | 4537/6470 [52:08<20:41,  1.56it/s]
4|4538|Loss: 0.7875348329544067:  70%|███████   | 4537/6470 [52:08<20:41,  1.56it/s]
4|4538|Loss: 0.7875348329544067:  70%|███████   | 4538/6470 [52:09<20:01,  1.61it/s]
4|4539|Loss: 0.8778296709060669:  70%|███████   | 4538/6470 [52:09<20:01,  1.61it/s]
4|4539|Loss: 0.8778296709060669:  70%|███████   | 4539/6470 [52:09<20:48,  1.55it/s]
4|4540|Loss: 0.786550760269165:  70%|███████   | 4539/6470 [52:10<20:48,  1.55it/s] 
4|4540|Loss: 0.786550760269165:  70%|███████   | 4540/6470 [52:10<21:48,  1.48it/s]
4|4541|Loss: 0.7672348618507385:  70%|███████   | 4540/6470 [52:10<21:48,  1.48it/s]
4|4541|Loss: 0.7672348618507385:  70%|███████   | 4541/6470 [52:11<22:55,  1.40it/s]
4|4542|Loss: 0.7430941462516785:  70%|███████   | 4541/6470 [52:11<22:55,  1.40it/s]
4|4542|Loss: 0.7430941462516785:  70%|███████   | 4542/6470 [52:12<22:58,  1.40it/s]
4|4543|Loss: 0.9596502780914307:  70%|███████   | 4542/6470 [52:12<22:58,  1.40it/s]
4|4543|Loss: 0.9596502780914307:  70%|███████   | 4543/6470 [52:12<23:01,  1.39it/s]
4|4544|Loss: 0.8120697140693665:  70%|███████   | 4543/6470 [52:13<23:01,  1.39it/s]
4|4544|Loss: 0.8120697140693665:  70%|███████   | 4544/6470 [52:13<24:16,  1.32it/s]
4|4545|Loss: 0.8908904194831848:  70%|███████   | 4544/6470 [52:14<24:16,  1.32it/s]
4|4545|Loss: 0.8908904194831848:  70%|███████   | 4545/6470 [52:14<22:49,  1.41it/s]
4|4546|Loss: 0.8880435824394226:  70%|███████   | 4545/6470 [52:14<22:49,  1.41it/s]
4|4546|Loss: 0.8880435824394226:  70%|███████   | 4546/6470 [52:14<23:03,  1.39it/s]
4|4547|Loss: 0.936894953250885:  70%|███████   | 4546/6470 [52:15<23:03,  1.39it/s] 
4|4547|Loss: 0.936894953250885:  70%|███████   | 4547/6470 [52:15<22:08,  1.45it/s]
4|4548|Loss: 0.9203047156333923:  70%|███████   | 4547/6470 [52:15<22:08,  1.45it/s]
4|4548|Loss: 0.9203047156333923:  70%|███████   | 4548/6470 [52:16<21:59,  1.46it/s]
4|4549|Loss: 0.7545390725135803:  70%|███████   | 4548/6470 [52:16<21:59,  1.46it/s]
4|4549|Loss: 0.7545390725135803:  70%|███████   | 4549/6470 [52:17<23:06,  1.39it/s]
4|4550|Loss: 0.9527040123939514:  70%|███████   | 4549/6470 [52:17<23:06,  1.39it/s]
4|4550|Loss: 0.9527040123939514:  70%|███████   | 4550/6470 [52:17<22:45,  1.41it/s]
4|4551|Loss: 0.945491373538971:  70%|███████   | 4550/6470 [52:18<22:45,  1.41it/s] 
4|4551|Loss: 0.945491373538971:  70%|███████   | 4551/6470 [52:18<23:19,  1.37it/s]
4|4552|Loss: 1.0023839473724365:  70%|███████   | 4551/6470 [52:18<23:19,  1.37it/s]
4|4552|Loss: 1.0023839473724365:  70%|███████   | 4552/6470 [52:19<22:48,  1.40it/s]
4|4553|Loss: 0.9545847177505493:  70%|███████   | 4552/6470 [52:19<22:48,  1.40it/s]
4|4553|Loss: 0.9545847177505493:  70%|███████   | 4553/6470 [52:19<20:39,  1.55it/s]
4|4554|Loss: 0.7469464540481567:  70%|███████   | 4553/6470 [52:20<20:39,  1.55it/s]
4|4554|Loss: 0.7469464540481567:  70%|███████   | 4554/6470 [52:20<20:57,  1.52it/s]
4|4555|Loss: 0.8783155083656311:  70%|███████   | 4554/6470 [52:20<20:57,  1.52it/s]
4|4555|Loss: 0.8783155083656311:  70%|███████   | 4555/6470 [52:21<20:41,  1.54it/s]
4|4556|Loss: 0.8909435868263245:  70%|███████   | 4555/6470 [52:21<20:41,  1.54it/s]
4|4556|Loss: 0.8909435868263245:  70%|███████   | 4556/6470 [52:21<21:03,  1.51it/s]
4|4557|Loss: 0.8069470524787903:  70%|███████   | 4556/6470 [52:22<21:03,  1.51it/s]
4|4557|Loss: 0.8069470524787903:  70%|███████   | 4557/6470 [52:22<22:48,  1.40it/s]
4|4558|Loss: 0.830435574054718:  70%|███████   | 4557/6470 [52:22<22:48,  1.40it/s] 
4|4558|Loss: 0.830435574054718:  70%|███████   | 4558/6470 [52:23<23:08,  1.38it/s]
4|4559|Loss: 0.7536154985427856:  70%|███████   | 4558/6470 [52:23<23:08,  1.38it/s]
4|4559|Loss: 0.7536154985427856:  70%|███████   | 4559/6470 [52:23<21:54,  1.45it/s]
4|4560|Loss: 0.5401438474655151:  70%|███████   | 4559/6470 [52:24<21:54,  1.45it/s]
4|4560|Loss: 0.5401438474655151:  70%|███████   | 4560/6470 [52:24<21:19,  1.49it/s]
4|4561|Loss: 0.712579071521759:  70%|███████   | 4560/6470 [52:24<21:19,  1.49it/s] 
4|4561|Loss: 0.712579071521759:  70%|███████   | 4561/6470 [52:25<19:40,  1.62it/s]
4|4562|Loss: 0.8641008734703064:  70%|███████   | 4561/6470 [52:25<19:40,  1.62it/s]
4|4562|Loss: 0.8641008734703064:  71%|███████   | 4562/6470 [52:25<19:53,  1.60it/s]
4|4563|Loss: 1.0160726308822632:  71%|███████   | 4562/6470 [52:26<19:53,  1.60it/s]
4|4563|Loss: 1.0160726308822632:  71%|███████   | 4563/6470 [52:26<21:34,  1.47it/s]
4|4564|Loss: 0.8386381268501282:  71%|███████   | 4563/6470 [52:26<21:34,  1.47it/s]
4|4564|Loss: 0.8386381268501282:  71%|███████   | 4564/6470 [52:27<22:27,  1.41it/s]
4|4565|Loss: 0.8595520257949829:  71%|███████   | 4564/6470 [52:27<22:27,  1.41it/s]
4|4565|Loss: 0.8595520257949829:  71%|███████   | 4565/6470 [52:27<21:22,  1.49it/s]
4|4566|Loss: 0.7938348054885864:  71%|███████   | 4565/6470 [52:28<21:22,  1.49it/s]
4|4566|Loss: 0.7938348054885864:  71%|███████   | 4566/6470 [52:28<22:05,  1.44it/s]
4|4567|Loss: 0.8216462135314941:  71%|███████   | 4566/6470 [52:29<22:05,  1.44it/s]
4|4567|Loss: 0.8216462135314941:  71%|███████   | 4567/6470 [52:29<22:45,  1.39it/s]
4|4568|Loss: 0.8650895953178406:  71%|███████   | 4567/6470 [52:29<22:45,  1.39it/s]
4|4568|Loss: 0.8650895953178406:  71%|███████   | 4568/6470 [52:30<22:50,  1.39it/s]
4|4569|Loss: 0.7596763372421265:  71%|███████   | 4568/6470 [52:30<22:50,  1.39it/s]
4|4569|Loss: 0.7596763372421265:  71%|███████   | 4569/6470 [52:30<21:19,  1.49it/s]
4|4570|Loss: 0.8398606777191162:  71%|███████   | 4569/6470 [52:30<21:19,  1.49it/s]
4|4570|Loss: 0.8398606777191162:  71%|███████   | 4570/6470 [52:31<20:39,  1.53it/s]
4|4571|Loss: 0.9028359055519104:  71%|███████   | 4570/6470 [52:31<20:39,  1.53it/s]
4|4571|Loss: 0.9028359055519104:  71%|███████   | 4571/6470 [52:31<19:50,  1.59it/s]
4|4572|Loss: 0.7414831519126892:  71%|███████   | 4571/6470 [52:32<19:50,  1.59it/s]
4|4572|Loss: 0.7414831519126892:  71%|███████   | 4572/6470 [52:32<21:27,  1.47it/s]
4|4573|Loss: 0.8574419021606445:  71%|███████   | 4572/6470 [52:33<21:27,  1.47it/s]
4|4573|Loss: 0.8574419021606445:  71%|███████   | 4573/6470 [52:33<21:23,  1.48it/s]
4|4574|Loss: 0.862375020980835:  71%|███████   | 4573/6470 [52:33<21:23,  1.48it/s] 
4|4574|Loss: 0.862375020980835:  71%|███████   | 4574/6470 [52:34<22:28,  1.41it/s]
4|4575|Loss: 0.8406622409820557:  71%|███████   | 4574/6470 [52:34<22:28,  1.41it/s]
4|4575|Loss: 0.8406622409820557:  71%|███████   | 4575/6470 [52:34<22:52,  1.38it/s]
4|4576|Loss: 0.7229712605476379:  71%|███████   | 4575/6470 [52:35<22:52,  1.38it/s]
4|4576|Loss: 0.7229712605476379:  71%|███████   | 4576/6470 [52:35<22:15,  1.42it/s]
4|4577|Loss: 0.8348909020423889:  71%|███████   | 4576/6470 [52:35<22:15,  1.42it/s]
4|4577|Loss: 0.8348909020423889:  71%|███████   | 4577/6470 [52:36<22:20,  1.41it/s]
4|4578|Loss: 0.8099386096000671:  71%|███████   | 4577/6470 [52:36<22:20,  1.41it/s]
4|4578|Loss: 0.8099386096000671:  71%|███████   | 4578/6470 [52:37<23:30,  1.34it/s]
4|4579|Loss: 0.979045569896698:  71%|███████   | 4578/6470 [52:37<23:30,  1.34it/s] 
4|4579|Loss: 0.979045569896698:  71%|███████   | 4579/6470 [52:37<23:24,  1.35it/s]
4|4580|Loss: 0.7917277812957764:  71%|███████   | 4579/6470 [52:38<23:24,  1.35it/s]
4|4580|Loss: 0.7917277812957764:  71%|███████   | 4580/6470 [52:38<23:23,  1.35it/s]
4|4581|Loss: 0.8153460025787354:  71%|███████   | 4580/6470 [52:38<23:23,  1.35it/s]
4|4581|Loss: 0.8153460025787354:  71%|███████   | 4581/6470 [52:39<23:07,  1.36it/s]
4|4582|Loss: 0.7712732553482056:  71%|███████   | 4581/6470 [52:39<23:07,  1.36it/s]
4|4582|Loss: 0.7712732553482056:  71%|███████   | 4582/6470 [52:39<20:56,  1.50it/s]
4|4583|Loss: 0.9286259412765503:  71%|███████   | 4582/6470 [52:40<20:56,  1.50it/s]
4|4583|Loss: 0.9286259412765503:  71%|███████   | 4583/6470 [52:40<19:48,  1.59it/s]
4|4584|Loss: 0.8226892948150635:  71%|███████   | 4583/6470 [52:40<19:48,  1.59it/s]
4|4584|Loss: 0.8226892948150635:  71%|███████   | 4584/6470 [52:40<18:52,  1.67it/s]
4|4585|Loss: 0.8659882545471191:  71%|███████   | 4584/6470 [52:41<18:52,  1.67it/s]
4|4585|Loss: 0.8659882545471191:  71%|███████   | 4585/6470 [52:41<20:37,  1.52it/s]
4|4586|Loss: 0.7302765250205994:  71%|███████   | 4585/6470 [52:42<20:37,  1.52it/s]
4|4586|Loss: 0.7302765250205994:  71%|███████   | 4586/6470 [52:42<22:47,  1.38it/s]
4|4587|Loss: 0.9067957997322083:  71%|███████   | 4586/6470 [52:42<22:47,  1.38it/s]
4|4587|Loss: 0.9067957997322083:  71%|███████   | 4587/6470 [52:43<23:17,  1.35it/s]
4|4588|Loss: 0.9149060249328613:  71%|███████   | 4587/6470 [52:43<23:17,  1.35it/s]
4|4588|Loss: 0.9149060249328613:  71%|███████   | 4588/6470 [52:44<23:15,  1.35it/s]
4|4589|Loss: 0.8810340762138367:  71%|███████   | 4588/6470 [52:44<23:15,  1.35it/s]
4|4589|Loss: 0.8810340762138367:  71%|███████   | 4589/6470 [52:44<22:55,  1.37it/s]
4|4590|Loss: 0.8468950390815735:  71%|███████   | 4589/6470 [52:45<22:55,  1.37it/s]
4|4590|Loss: 0.8468950390815735:  71%|███████   | 4590/6470 [52:45<23:22,  1.34it/s]
4|4591|Loss: 0.7716556787490845:  71%|███████   | 4590/6470 [52:45<23:22,  1.34it/s]
4|4591|Loss: 0.7716556787490845:  71%|███████   | 4591/6470 [52:46<22:41,  1.38it/s]
4|4592|Loss: 0.801480233669281:  71%|███████   | 4591/6470 [52:46<22:41,  1.38it/s] 
4|4592|Loss: 0.801480233669281:  71%|███████   | 4592/6470 [52:46<21:45,  1.44it/s]
4|4593|Loss: 1.02222740650177:  71%|███████   | 4592/6470 [52:47<21:45,  1.44it/s] 
4|4593|Loss: 1.02222740650177:  71%|███████   | 4593/6470 [52:47<22:21,  1.40it/s]
4|4594|Loss: 0.9724076986312866:  71%|███████   | 4593/6470 [52:47<22:21,  1.40it/s]
4|4594|Loss: 0.9724076986312866:  71%|███████   | 4594/6470 [52:48<20:35,  1.52it/s]
4|4595|Loss: 0.8157760500907898:  71%|███████   | 4594/6470 [52:48<20:35,  1.52it/s]
4|4595|Loss: 0.8157760500907898:  71%|███████   | 4595/6470 [52:48<20:48,  1.50it/s]
4|4596|Loss: 0.9546658992767334:  71%|███████   | 4595/6470 [52:49<20:48,  1.50it/s]
4|4596|Loss: 0.9546658992767334:  71%|███████   | 4596/6470 [52:49<20:51,  1.50it/s]
4|4597|Loss: 0.76051926612854:  71%|███████   | 4596/6470 [52:49<20:51,  1.50it/s]  
4|4597|Loss: 0.76051926612854:  71%|███████   | 4597/6470 [52:50<20:22,  1.53it/s]
4|4598|Loss: 0.8321889042854309:  71%|███████   | 4597/6470 [52:50<20:22,  1.53it/s]
4|4598|Loss: 0.8321889042854309:  71%|███████   | 4598/6470 [52:50<20:36,  1.51it/s]
4|4599|Loss: 0.8174641132354736:  71%|███████   | 4598/6470 [52:51<20:36,  1.51it/s]
4|4599|Loss: 0.8174641132354736:  71%|███████   | 4599/6470 [52:51<20:21,  1.53it/s]
4|4600|Loss: 0.8538057804107666:  71%|███████   | 4599/6470 [52:51<20:21,  1.53it/s]
4|4600|Loss: 0.8538057804107666:  71%|███████   | 4600/6470 [52:52<20:34,  1.51it/s]
4|4601|Loss: 0.9250651597976685:  71%|███████   | 4600/6470 [52:52<20:34,  1.51it/s]
4|4601|Loss: 0.9250651597976685:  71%|███████   | 4601/6470 [52:52<21:18,  1.46it/s]
4|4602|Loss: 0.9682584404945374:  71%|███████   | 4601/6470 [52:53<21:18,  1.46it/s]
4|4602|Loss: 0.9682584404945374:  71%|███████   | 4602/6470 [52:53<22:33,  1.38it/s]
4|4603|Loss: 0.8450188636779785:  71%|███████   | 4602/6470 [52:54<22:33,  1.38it/s]
4|4603|Loss: 0.8450188636779785:  71%|███████   | 4603/6470 [52:54<23:40,  1.31it/s]
4|4604|Loss: 0.9235641956329346:  71%|███████   | 4603/6470 [52:54<23:40,  1.31it/s]
4|4604|Loss: 0.9235641956329346:  71%|███████   | 4604/6470 [52:55<24:24,  1.27it/s]
4|4605|Loss: 0.7962605357170105:  71%|███████   | 4604/6470 [52:55<24:24,  1.27it/s]
4|4605|Loss: 0.7962605357170105:  71%|███████   | 4605/6470 [52:56<24:04,  1.29it/s]
4|4606|Loss: 0.7272421717643738:  71%|███████   | 4605/6470 [52:56<24:04,  1.29it/s]
4|4606|Loss: 0.7272421717643738:  71%|███████   | 4606/6470 [52:56<23:25,  1.33it/s]
4|4607|Loss: 0.9328186511993408:  71%|███████   | 4606/6470 [52:57<23:25,  1.33it/s]
4|4607|Loss: 0.9328186511993408:  71%|███████   | 4607/6470 [52:57<21:56,  1.42it/s]
4|4608|Loss: 0.9339596629142761:  71%|███████   | 4607/6470 [52:57<21:56,  1.42it/s]
4|4608|Loss: 0.9339596629142761:  71%|███████   | 4608/6470 [52:58<22:01,  1.41it/s]
4|4609|Loss: 0.7482330799102783:  71%|███████   | 4608/6470 [52:58<22:01,  1.41it/s]
4|4609|Loss: 0.7482330799102783:  71%|███████   | 4609/6470 [52:58<21:39,  1.43it/s]
4|4610|Loss: 0.8749150037765503:  71%|███████   | 4609/6470 [52:59<21:39,  1.43it/s]
4|4610|Loss: 0.8749150037765503:  71%|███████▏  | 4610/6470 [52:59<20:26,  1.52it/s]
4|4611|Loss: 0.7459282875061035:  71%|███████▏  | 4610/6470 [52:59<20:26,  1.52it/s]
4|4611|Loss: 0.7459282875061035:  71%|███████▏  | 4611/6470 [52:59<19:45,  1.57it/s]
4|4612|Loss: 0.819897472858429:  71%|███████▏  | 4611/6470 [53:00<19:45,  1.57it/s] 
4|4612|Loss: 0.819897472858429:  71%|███████▏  | 4612/6470 [53:00<19:37,  1.58it/s]
4|4613|Loss: 0.8411470651626587:  71%|███████▏  | 4612/6470 [53:00<19:37,  1.58it/s]
4|4613|Loss: 0.8411470651626587:  71%|███████▏  | 4613/6470 [53:01<20:21,  1.52it/s]
4|4614|Loss: 0.7925348281860352:  71%|███████▏  | 4613/6470 [53:01<20:21,  1.52it/s]
4|4614|Loss: 0.7925348281860352:  71%|███████▏  | 4614/6470 [53:01<21:10,  1.46it/s]
4|4615|Loss: 0.8780031204223633:  71%|███████▏  | 4614/6470 [53:02<21:10,  1.46it/s]
4|4615|Loss: 0.8780031204223633:  71%|███████▏  | 4615/6470 [53:02<22:50,  1.35it/s]
4|4616|Loss: 0.814267635345459:  71%|███████▏  | 4615/6470 [53:03<22:50,  1.35it/s] 
4|4616|Loss: 0.814267635345459:  71%|███████▏  | 4616/6470 [53:03<22:58,  1.35it/s]
4|4617|Loss: 0.9229968786239624:  71%|███████▏  | 4616/6470 [53:04<22:58,  1.35it/s]
4|4617|Loss: 0.9229968786239624:  71%|███████▏  | 4617/6470 [53:04<23:40,  1.30it/s]
4|4618|Loss: 0.7902088761329651:  71%|███████▏  | 4617/6470 [53:04<23:40,  1.30it/s]
4|4618|Loss: 0.7902088761329651:  71%|███████▏  | 4618/6470 [53:05<23:23,  1.32it/s]
4|4619|Loss: 0.8756420612335205:  71%|███████▏  | 4618/6470 [53:05<23:23,  1.32it/s]
4|4619|Loss: 0.8756420612335205:  71%|███████▏  | 4619/6470 [53:05<23:28,  1.31it/s]
4|4620|Loss: 0.7110111117362976:  71%|███████▏  | 4619/6470 [53:06<23:28,  1.31it/s]
4|4620|Loss: 0.7110111117362976:  71%|███████▏  | 4620/6470 [53:06<22:37,  1.36it/s]
4|4621|Loss: 0.8255094289779663:  71%|███████▏  | 4620/6470 [53:07<22:37,  1.36it/s]
4|4621|Loss: 0.8255094289779663:  71%|███████▏  | 4621/6470 [53:07<23:17,  1.32it/s]
4|4622|Loss: 0.834750771522522:  71%|███████▏  | 4621/6470 [53:07<23:17,  1.32it/s] 
4|4622|Loss: 0.834750771522522:  71%|███████▏  | 4622/6470 [53:08<23:19,  1.32it/s]
4|4623|Loss: 0.8769238591194153:  71%|███████▏  | 4622/6470 [53:08<23:19,  1.32it/s]
4|4623|Loss: 0.8769238591194153:  71%|███████▏  | 4623/6470 [53:09<24:02,  1.28it/s]
4|4624|Loss: 0.7209511399269104:  71%|███████▏  | 4623/6470 [53:09<24:02,  1.28it/s]
4|4624|Loss: 0.7209511399269104:  71%|███████▏  | 4624/6470 [53:09<23:03,  1.33it/s]
4|4625|Loss: 1.0131257772445679:  71%|███████▏  | 4624/6470 [53:10<23:03,  1.33it/s]
4|4625|Loss: 1.0131257772445679:  71%|███████▏  | 4625/6470 [53:10<22:35,  1.36it/s]
4|4626|Loss: 0.7515577673912048:  71%|███████▏  | 4625/6470 [53:10<22:35,  1.36it/s]
4|4626|Loss: 0.7515577673912048:  71%|███████▏  | 4626/6470 [53:11<22:34,  1.36it/s]
4|4627|Loss: 0.8089302182197571:  71%|███████▏  | 4626/6470 [53:11<22:34,  1.36it/s]
4|4627|Loss: 0.8089302182197571:  72%|███████▏  | 4627/6470 [53:11<21:43,  1.41it/s]
4|4628|Loss: 0.9683706164360046:  72%|███████▏  | 4627/6470 [53:12<21:43,  1.41it/s]
4|4628|Loss: 0.9683706164360046:  72%|███████▏  | 4628/6470 [53:12<21:01,  1.46it/s]
4|4629|Loss: 0.7728967666625977:  72%|███████▏  | 4628/6470 [53:12<21:01,  1.46it/s]
4|4629|Loss: 0.7728967666625977:  72%|███████▏  | 4629/6470 [53:13<20:36,  1.49it/s]
4|4630|Loss: 0.9143815040588379:  72%|███████▏  | 4629/6470 [53:13<20:36,  1.49it/s]
4|4630|Loss: 0.9143815040588379:  72%|███████▏  | 4630/6470 [53:13<20:07,  1.52it/s]
4|4631|Loss: 0.8399773240089417:  72%|███████▏  | 4630/6470 [53:14<20:07,  1.52it/s]
4|4631|Loss: 0.8399773240089417:  72%|███████▏  | 4631/6470 [53:14<19:56,  1.54it/s]
4|4632|Loss: 0.7334964275360107:  72%|███████▏  | 4631/6470 [53:14<19:56,  1.54it/s]
4|4632|Loss: 0.7334964275360107:  72%|███████▏  | 4632/6470 [53:14<20:19,  1.51it/s]
4|4633|Loss: 0.7675691843032837:  72%|███████▏  | 4632/6470 [53:15<20:19,  1.51it/s]
4|4633|Loss: 0.7675691843032837:  72%|███████▏  | 4633/6470 [53:15<21:47,  1.40it/s]
4|4634|Loss: 0.8746357560157776:  72%|███████▏  | 4633/6470 [53:16<21:47,  1.40it/s]
4|4634|Loss: 0.8746357560157776:  72%|███████▏  | 4634/6470 [53:16<21:26,  1.43it/s]
4|4635|Loss: 0.7260056138038635:  72%|███████▏  | 4634/6470 [53:16<21:26,  1.43it/s]
4|4635|Loss: 0.7260056138038635:  72%|███████▏  | 4635/6470 [53:17<20:08,  1.52it/s]
4|4636|Loss: 0.9854909777641296:  72%|███████▏  | 4635/6470 [53:17<20:08,  1.52it/s]
4|4636|Loss: 0.9854909777641296:  72%|███████▏  | 4636/6470 [53:17<21:15,  1.44it/s]
4|4637|Loss: 0.7566123008728027:  72%|███████▏  | 4636/6470 [53:18<21:15,  1.44it/s]
4|4637|Loss: 0.7566123008728027:  72%|███████▏  | 4637/6470 [53:18<21:49,  1.40it/s]
4|4638|Loss: 0.7096319794654846:  72%|███████▏  | 4637/6470 [53:19<21:49,  1.40it/s]
4|4638|Loss: 0.7096319794654846:  72%|███████▏  | 4638/6470 [53:19<23:19,  1.31it/s]
4|4639|Loss: 0.7896082997322083:  72%|███████▏  | 4638/6470 [53:19<23:19,  1.31it/s]
4|4639|Loss: 0.7896082997322083:  72%|███████▏  | 4639/6470 [53:20<22:30,  1.36it/s]
4|4640|Loss: 0.9367972016334534:  72%|███████▏  | 4639/6470 [53:20<22:30,  1.36it/s]
4|4640|Loss: 0.9367972016334534:  72%|███████▏  | 4640/6470 [53:20<20:23,  1.50it/s]
4|4641|Loss: 0.8235946297645569:  72%|███████▏  | 4640/6470 [53:21<20:23,  1.50it/s]
4|4641|Loss: 0.8235946297645569:  72%|███████▏  | 4641/6470 [53:21<21:22,  1.43it/s]
4|4642|Loss: 0.9099363088607788:  72%|███████▏  | 4641/6470 [53:21<21:22,  1.43it/s]
4|4642|Loss: 0.9099363088607788:  72%|███████▏  | 4642/6470 [53:22<22:50,  1.33it/s]
4|4643|Loss: 0.825912594795227:  72%|███████▏  | 4642/6470 [53:22<22:50,  1.33it/s] 
4|4643|Loss: 0.825912594795227:  72%|███████▏  | 4643/6470 [53:22<21:59,  1.38it/s]
4|4644|Loss: 0.9210795164108276:  72%|███████▏  | 4643/6470 [53:23<21:59,  1.38it/s]
4|4644|Loss: 0.9210795164108276:  72%|███████▏  | 4644/6470 [53:23<22:18,  1.36it/s]
4|4645|Loss: 0.8424873352050781:  72%|███████▏  | 4644/6470 [53:24<22:18,  1.36it/s]
4|4645|Loss: 0.8424873352050781:  72%|███████▏  | 4645/6470 [53:24<22:14,  1.37it/s]
4|4646|Loss: 0.8337149024009705:  72%|███████▏  | 4645/6470 [53:24<22:14,  1.37it/s]
4|4646|Loss: 0.8337149024009705:  72%|███████▏  | 4646/6470 [53:25<21:21,  1.42it/s]
4|4647|Loss: 0.7336611747741699:  72%|███████▏  | 4646/6470 [53:25<21:21,  1.42it/s]
4|4647|Loss: 0.7336611747741699:  72%|███████▏  | 4647/6470 [53:25<22:06,  1.37it/s]
4|4648|Loss: 0.9006807804107666:  72%|███████▏  | 4647/6470 [53:26<22:06,  1.37it/s]
4|4648|Loss: 0.9006807804107666:  72%|███████▏  | 4648/6470 [53:26<20:45,  1.46it/s]
4|4649|Loss: 0.8171284198760986:  72%|███████▏  | 4648/6470 [53:26<20:45,  1.46it/s]
4|4649|Loss: 0.8171284198760986:  72%|███████▏  | 4649/6470 [53:27<20:05,  1.51it/s]
4|4650|Loss: 0.9310250282287598:  72%|███████▏  | 4649/6470 [53:27<20:05,  1.51it/s]
4|4650|Loss: 0.9310250282287598:  72%|███████▏  | 4650/6470 [53:27<21:37,  1.40it/s]
4|4651|Loss: 0.8638067841529846:  72%|███████▏  | 4650/6470 [53:28<21:37,  1.40it/s]
4|4651|Loss: 0.8638067841529846:  72%|███████▏  | 4651/6470 [53:28<21:05,  1.44it/s]
4|4652|Loss: 0.9357994198799133:  72%|███████▏  | 4651/6470 [53:28<21:05,  1.44it/s]
4|4652|Loss: 0.9357994198799133:  72%|███████▏  | 4652/6470 [53:29<20:19,  1.49it/s]
4|4653|Loss: 1.0642467737197876:  72%|███████▏  | 4652/6470 [53:29<20:19,  1.49it/s]
4|4653|Loss: 1.0642467737197876:  72%|███████▏  | 4653/6470 [53:29<20:48,  1.46it/s]
4|4654|Loss: 0.8927034139633179:  72%|███████▏  | 4653/6470 [53:30<20:48,  1.46it/s]
4|4654|Loss: 0.8927034139633179:  72%|███████▏  | 4654/6470 [53:30<19:57,  1.52it/s]
4|4655|Loss: 0.8706539273262024:  72%|███████▏  | 4654/6470 [53:30<19:57,  1.52it/s]
4|4655|Loss: 0.8706539273262024:  72%|███████▏  | 4655/6470 [53:31<20:04,  1.51it/s]
4|4656|Loss: 0.8643465638160706:  72%|███████▏  | 4655/6470 [53:31<20:04,  1.51it/s]
4|4656|Loss: 0.8643465638160706:  72%|███████▏  | 4656/6470 [53:31<21:09,  1.43it/s]
4|4657|Loss: 1.032421588897705:  72%|███████▏  | 4656/6470 [53:32<21:09,  1.43it/s] 
4|4657|Loss: 1.032421588897705:  72%|███████▏  | 4657/6470 [53:32<19:58,  1.51it/s]
4|4658|Loss: 0.8093680739402771:  72%|███████▏  | 4657/6470 [53:32<19:58,  1.51it/s]
4|4658|Loss: 0.8093680739402771:  72%|███████▏  | 4658/6470 [53:33<20:09,  1.50it/s]
4|4659|Loss: 0.8220897316932678:  72%|███████▏  | 4658/6470 [53:33<20:09,  1.50it/s]
4|4659|Loss: 0.8220897316932678:  72%|███████▏  | 4659/6470 [53:33<20:47,  1.45it/s]
4|4660|Loss: 0.6920766234397888:  72%|███████▏  | 4659/6470 [53:34<20:47,  1.45it/s]
4|4660|Loss: 0.6920766234397888:  72%|███████▏  | 4660/6470 [53:34<21:08,  1.43it/s]
4|4661|Loss: 0.5988373160362244:  72%|███████▏  | 4660/6470 [53:34<21:08,  1.43it/s]
4|4661|Loss: 0.5988373160362244:  72%|███████▏  | 4661/6470 [53:35<19:21,  1.56it/s]
4|4662|Loss: 0.7148672342300415:  72%|███████▏  | 4661/6470 [53:35<19:21,  1.56it/s]
4|4662|Loss: 0.7148672342300415:  72%|███████▏  | 4662/6470 [53:35<21:12,  1.42it/s]
4|4663|Loss: 0.7173004746437073:  72%|███████▏  | 4662/6470 [53:36<21:12,  1.42it/s]
4|4663|Loss: 0.7173004746437073:  72%|███████▏  | 4663/6470 [53:36<22:53,  1.32it/s]
4|4664|Loss: 0.8125957250595093:  72%|███████▏  | 4663/6470 [53:37<22:53,  1.32it/s]
4|4664|Loss: 0.8125957250595093:  72%|███████▏  | 4664/6470 [53:37<22:04,  1.36it/s]
4|4665|Loss: 0.7907884120941162:  72%|███████▏  | 4664/6470 [53:37<22:04,  1.36it/s]
4|4665|Loss: 0.7907884120941162:  72%|███████▏  | 4665/6470 [53:38<21:08,  1.42it/s]
4|4666|Loss: 0.8900992274284363:  72%|███████▏  | 4665/6470 [53:38<21:08,  1.42it/s]
4|4666|Loss: 0.8900992274284363:  72%|███████▏  | 4666/6470 [53:38<21:19,  1.41it/s]
4|4667|Loss: 0.831620454788208:  72%|███████▏  | 4666/6470 [53:39<21:19,  1.41it/s] 
4|4667|Loss: 0.831620454788208:  72%|███████▏  | 4667/6470 [53:39<21:30,  1.40it/s]
4|4668|Loss: 0.8250548839569092:  72%|███████▏  | 4667/6470 [53:39<21:30,  1.40it/s]
4|4668|Loss: 0.8250548839569092:  72%|███████▏  | 4668/6470 [53:40<20:02,  1.50it/s]
4|4669|Loss: 0.8416902422904968:  72%|███████▏  | 4668/6470 [53:40<20:02,  1.50it/s]
4|4669|Loss: 0.8416902422904968:  72%|███████▏  | 4669/6470 [53:40<21:05,  1.42it/s]
4|4670|Loss: 0.9528346061706543:  72%|███████▏  | 4669/6470 [53:41<21:05,  1.42it/s]
4|4670|Loss: 0.9528346061706543:  72%|███████▏  | 4670/6470 [53:41<19:00,  1.58it/s]
4|4671|Loss: 0.7999779582023621:  72%|███████▏  | 4670/6470 [53:41<19:00,  1.58it/s]
4|4671|Loss: 0.7999779582023621:  72%|███████▏  | 4671/6470 [53:42<19:02,  1.57it/s]
4|4672|Loss: 0.8085564970970154:  72%|███████▏  | 4671/6470 [53:42<19:02,  1.57it/s]
4|4672|Loss: 0.8085564970970154:  72%|███████▏  | 4672/6470 [53:42<19:51,  1.51it/s]
4|4673|Loss: 0.8771370053291321:  72%|███████▏  | 4672/6470 [53:43<19:51,  1.51it/s]
4|4673|Loss: 0.8771370053291321:  72%|███████▏  | 4673/6470 [53:43<19:19,  1.55it/s]
4|4674|Loss: 0.8404961824417114:  72%|███████▏  | 4673/6470 [53:43<19:19,  1.55it/s]
4|4674|Loss: 0.8404961824417114:  72%|███████▏  | 4674/6470 [53:44<18:59,  1.58it/s]
4|4675|Loss: 0.8957096338272095:  72%|███████▏  | 4674/6470 [53:44<18:59,  1.58it/s]
4|4675|Loss: 0.8957096338272095:  72%|███████▏  | 4675/6470 [53:44<17:51,  1.68it/s]
4|4676|Loss: 0.8665108680725098:  72%|███████▏  | 4675/6470 [53:44<17:51,  1.68it/s]
4|4676|Loss: 0.8665108680725098:  72%|███████▏  | 4676/6470 [53:45<18:58,  1.58it/s]
4|4677|Loss: 0.9393686652183533:  72%|███████▏  | 4676/6470 [53:45<18:58,  1.58it/s]
4|4677|Loss: 0.9393686652183533:  72%|███████▏  | 4677/6470 [53:46<20:04,  1.49it/s]
4|4678|Loss: 0.7613070011138916:  72%|███████▏  | 4677/6470 [53:46<20:04,  1.49it/s]
4|4678|Loss: 0.7613070011138916:  72%|███████▏  | 4678/6470 [53:46<20:27,  1.46it/s]
4|4679|Loss: 0.8912798166275024:  72%|███████▏  | 4678/6470 [53:47<20:27,  1.46it/s]
4|4679|Loss: 0.8912798166275024:  72%|███████▏  | 4679/6470 [53:47<20:35,  1.45it/s]
4|4680|Loss: 0.9450133442878723:  72%|███████▏  | 4679/6470 [53:47<20:35,  1.45it/s]
4|4680|Loss: 0.9450133442878723:  72%|███████▏  | 4680/6470 [53:48<21:34,  1.38it/s]
4|4681|Loss: 0.9974814653396606:  72%|███████▏  | 4680/6470 [53:48<21:34,  1.38it/s]
4|4681|Loss: 0.9974814653396606:  72%|███████▏  | 4681/6470 [53:48<20:40,  1.44it/s]
4|4682|Loss: 1.0353561639785767:  72%|███████▏  | 4681/6470 [53:49<20:40,  1.44it/s]
4|4682|Loss: 1.0353561639785767:  72%|███████▏  | 4682/6470 [53:49<21:17,  1.40it/s]
4|4683|Loss: 0.8998412489891052:  72%|███████▏  | 4682/6470 [53:50<21:17,  1.40it/s]
4|4683|Loss: 0.8998412489891052:  72%|███████▏  | 4683/6470 [53:50<21:45,  1.37it/s]
4|4684|Loss: 0.6813721656799316:  72%|███████▏  | 4683/6470 [53:50<21:45,  1.37it/s]
4|4684|Loss: 0.6813721656799316:  72%|███████▏  | 4684/6470 [53:51<22:38,  1.31it/s]
4|4685|Loss: 0.7851876020431519:  72%|███████▏  | 4684/6470 [53:51<22:38,  1.31it/s]
4|4685|Loss: 0.7851876020431519:  72%|███████▏  | 4685/6470 [53:51<22:33,  1.32it/s]
4|4686|Loss: 0.6977708339691162:  72%|███████▏  | 4685/6470 [53:52<22:33,  1.32it/s]
4|4686|Loss: 0.6977708339691162:  72%|███████▏  | 4686/6470 [53:52<22:14,  1.34it/s]
4|4687|Loss: 0.8826645016670227:  72%|███████▏  | 4686/6470 [53:53<22:14,  1.34it/s]
4|4687|Loss: 0.8826645016670227:  72%|███████▏  | 4687/6470 [53:53<20:03,  1.48it/s]
4|4688|Loss: 0.8576434254646301:  72%|███████▏  | 4687/6470 [53:53<20:03,  1.48it/s]
4|4688|Loss: 0.8576434254646301:  72%|███████▏  | 4688/6470 [53:53<19:11,  1.55it/s]
4|4689|Loss: 0.7509968280792236:  72%|███████▏  | 4688/6470 [53:54<19:11,  1.55it/s]
4|4689|Loss: 0.7509968280792236:  72%|███████▏  | 4689/6470 [53:54<18:08,  1.64it/s]
4|4690|Loss: 0.8969170451164246:  72%|███████▏  | 4689/6470 [53:54<18:08,  1.64it/s]
4|4690|Loss: 0.8969170451164246:  72%|███████▏  | 4690/6470 [53:55<19:49,  1.50it/s]
4|4691|Loss: 1.031553864479065:  72%|███████▏  | 4690/6470 [53:55<19:49,  1.50it/s] 
4|4691|Loss: 1.031553864479065:  73%|███████▎  | 4691/6470 [53:55<20:02,  1.48it/s]
4|4692|Loss: 0.8562113642692566:  73%|███████▎  | 4691/6470 [53:56<20:02,  1.48it/s]
4|4692|Loss: 0.8562113642692566:  73%|███████▎  | 4692/6470 [53:56<21:12,  1.40it/s]
4|4693|Loss: 0.921814501285553:  73%|███████▎  | 4692/6470 [53:57<21:12,  1.40it/s] 
4|4693|Loss: 0.921814501285553:  73%|███████▎  | 4693/6470 [53:57<20:52,  1.42it/s]
4|4694|Loss: 0.7399651408195496:  73%|███████▎  | 4693/6470 [53:57<20:52,  1.42it/s]
4|4694|Loss: 0.7399651408195496:  73%|███████▎  | 4694/6470 [53:58<22:34,  1.31it/s]
4|4695|Loss: 0.8995039463043213:  73%|███████▎  | 4694/6470 [53:58<22:34,  1.31it/s]
4|4695|Loss: 0.8995039463043213:  73%|███████▎  | 4695/6470 [53:59<23:16,  1.27it/s]
4|4696|Loss: 0.8993062376976013:  73%|███████▎  | 4695/6470 [53:59<23:16,  1.27it/s]
4|4696|Loss: 0.8993062376976013:  73%|███████▎  | 4696/6470 [53:59<23:16,  1.27it/s]
4|4697|Loss: 0.8795874118804932:  73%|███████▎  | 4696/6470 [54:00<23:16,  1.27it/s]
4|4697|Loss: 0.8795874118804932:  73%|███████▎  | 4697/6470 [54:00<22:08,  1.33it/s]
4|4698|Loss: 0.8446608185768127:  73%|███████▎  | 4697/6470 [54:00<22:08,  1.33it/s]
4|4698|Loss: 0.8446608185768127:  73%|███████▎  | 4698/6470 [54:01<21:54,  1.35it/s]
4|4699|Loss: 0.8198106288909912:  73%|███████▎  | 4698/6470 [54:01<21:54,  1.35it/s]
4|4699|Loss: 0.8198106288909912:  73%|███████▎  | 4699/6470 [54:01<21:36,  1.37it/s]
4|4700|Loss: 0.7784202694892883:  73%|███████▎  | 4699/6470 [54:02<21:36,  1.37it/s]
4|4700|Loss: 0.7784202694892883:  73%|███████▎  | 4700/6470 [54:02<21:30,  1.37it/s]
4|4701|Loss: 0.8689161539077759:  73%|███████▎  | 4700/6470 [54:03<21:30,  1.37it/s]
4|4701|Loss: 0.8689161539077759:  73%|███████▎  | 4701/6470 [54:03<21:41,  1.36it/s]
4|4702|Loss: 0.9417010545730591:  73%|███████▎  | 4701/6470 [54:03<21:41,  1.36it/s]
4|4702|Loss: 0.9417010545730591:  73%|███████▎  | 4702/6470 [54:03<20:09,  1.46it/s]
4|4703|Loss: 0.8611212968826294:  73%|███████▎  | 4702/6470 [54:04<20:09,  1.46it/s]
4|4703|Loss: 0.8611212968826294:  73%|███████▎  | 4703/6470 [54:04<19:56,  1.48it/s]
4|4704|Loss: 0.9021600484848022:  73%|███████▎  | 4703/6470 [54:05<19:56,  1.48it/s]
4|4704|Loss: 0.9021600484848022:  73%|███████▎  | 4704/6470 [54:05<20:21,  1.45it/s]
4|4705|Loss: 0.9293960928916931:  73%|███████▎  | 4704/6470 [54:05<20:21,  1.45it/s]
4|4705|Loss: 0.9293960928916931:  73%|███████▎  | 4705/6470 [54:05<19:45,  1.49it/s]
4|4706|Loss: 0.9182998538017273:  73%|███████▎  | 4705/6470 [54:06<19:45,  1.49it/s]
4|4706|Loss: 0.9182998538017273:  73%|███████▎  | 4706/6470 [54:06<18:28,  1.59it/s]
4|4707|Loss: 0.7893214821815491:  73%|███████▎  | 4706/6470 [54:06<18:28,  1.59it/s]
4|4707|Loss: 0.7893214821815491:  73%|███████▎  | 4707/6470 [54:07<18:57,  1.55it/s]
4|4708|Loss: 0.7849457859992981:  73%|███████▎  | 4707/6470 [54:07<18:57,  1.55it/s]
4|4708|Loss: 0.7849457859992981:  73%|███████▎  | 4708/6470 [54:07<20:15,  1.45it/s]
4|4709|Loss: 0.7581861615180969:  73%|███████▎  | 4708/6470 [54:08<20:15,  1.45it/s]
4|4709|Loss: 0.7581861615180969:  73%|███████▎  | 4709/6470 [54:08<21:08,  1.39it/s]
4|4710|Loss: 0.727106511592865:  73%|███████▎  | 4709/6470 [54:09<21:08,  1.39it/s] 
4|4710|Loss: 0.727106511592865:  73%|███████▎  | 4710/6470 [54:09<22:25,  1.31it/s]
4|4711|Loss: 0.8999203443527222:  73%|███████▎  | 4710/6470 [54:10<22:25,  1.31it/s]
4|4711|Loss: 0.8999203443527222:  73%|███████▎  | 4711/6470 [54:10<21:21,  1.37it/s]
4|4712|Loss: 0.8020443916320801:  73%|███████▎  | 4711/6470 [54:10<21:21,  1.37it/s]
4|4712|Loss: 0.8020443916320801:  73%|███████▎  | 4712/6470 [54:10<20:05,  1.46it/s]
4|4713|Loss: 0.7137840390205383:  73%|███████▎  | 4712/6470 [54:11<20:05,  1.46it/s]
4|4713|Loss: 0.7137840390205383:  73%|███████▎  | 4713/6470 [54:11<19:33,  1.50it/s]
4|4714|Loss: 0.8243182301521301:  73%|███████▎  | 4713/6470 [54:11<19:33,  1.50it/s]
4|4714|Loss: 0.8243182301521301:  73%|███████▎  | 4714/6470 [54:12<19:46,  1.48it/s]
4|4715|Loss: 0.8380000591278076:  73%|███████▎  | 4714/6470 [54:12<19:46,  1.48it/s]
4|4715|Loss: 0.8380000591278076:  73%|███████▎  | 4715/6470 [54:12<19:25,  1.51it/s]
4|4716|Loss: 0.9369767904281616:  73%|███████▎  | 4715/6470 [54:13<19:25,  1.51it/s]
4|4716|Loss: 0.9369767904281616:  73%|███████▎  | 4716/6470 [54:13<19:10,  1.52it/s]
4|4717|Loss: 0.7952037453651428:  73%|███████▎  | 4716/6470 [54:13<19:10,  1.52it/s]
4|4717|Loss: 0.7952037453651428:  73%|███████▎  | 4717/6470 [54:14<18:51,  1.55it/s]
4|4718|Loss: 0.8729479312896729:  73%|███████▎  | 4717/6470 [54:14<18:51,  1.55it/s]
4|4718|Loss: 0.8729479312896729:  73%|███████▎  | 4718/6470 [54:14<19:49,  1.47it/s]
4|4719|Loss: 0.692220151424408:  73%|███████▎  | 4718/6470 [54:15<19:49,  1.47it/s] 
4|4719|Loss: 0.692220151424408:  73%|███████▎  | 4719/6470 [54:15<19:37,  1.49it/s]
4|4720|Loss: 0.9707943797111511:  73%|███████▎  | 4719/6470 [54:15<19:37,  1.49it/s]
4|4720|Loss: 0.9707943797111511:  73%|███████▎  | 4720/6470 [54:16<18:53,  1.54it/s]
4|4721|Loss: 0.817155659198761:  73%|███████▎  | 4720/6470 [54:16<18:53,  1.54it/s] 
4|4721|Loss: 0.817155659198761:  73%|███████▎  | 4721/6470 [54:16<19:23,  1.50it/s]
4|4722|Loss: 0.7711057066917419:  73%|███████▎  | 4721/6470 [54:17<19:23,  1.50it/s]
4|4722|Loss: 0.7711057066917419:  73%|███████▎  | 4722/6470 [54:17<19:43,  1.48it/s]
4|4723|Loss: 0.9685916304588318:  73%|███████▎  | 4722/6470 [54:17<19:43,  1.48it/s]
4|4723|Loss: 0.9685916304588318:  73%|███████▎  | 4723/6470 [54:17<17:32,  1.66it/s]
4|4724|Loss: 0.9739160537719727:  73%|███████▎  | 4723/6470 [54:18<17:32,  1.66it/s]
4|4724|Loss: 0.9739160537719727:  73%|███████▎  | 4724/6470 [54:18<18:43,  1.55it/s]
4|4725|Loss: 1.007994294166565:  73%|███████▎  | 4724/6470 [54:19<18:43,  1.55it/s] 
4|4725|Loss: 1.007994294166565:  73%|███████▎  | 4725/6470 [54:19<19:38,  1.48it/s]
4|4726|Loss: 1.0681393146514893:  73%|███████▎  | 4725/6470 [54:19<19:38,  1.48it/s]
4|4726|Loss: 1.0681393146514893:  73%|███████▎  | 4726/6470 [54:20<19:53,  1.46it/s]
4|4727|Loss: 0.7123746275901794:  73%|███████▎  | 4726/6470 [54:20<19:53,  1.46it/s]
4|4727|Loss: 0.7123746275901794:  73%|███████▎  | 4727/6470 [54:20<19:37,  1.48it/s]
4|4728|Loss: 0.8251991868019104:  73%|███████▎  | 4727/6470 [54:21<19:37,  1.48it/s]
4|4728|Loss: 0.8251991868019104:  73%|███████▎  | 4728/6470 [54:21<18:33,  1.56it/s]
4|4729|Loss: 0.9546728730201721:  73%|███████▎  | 4728/6470 [54:21<18:33,  1.56it/s]
4|4729|Loss: 0.9546728730201721:  73%|███████▎  | 4729/6470 [54:21<18:12,  1.59it/s]
4|4730|Loss: 0.6912336349487305:  73%|███████▎  | 4729/6470 [54:22<18:12,  1.59it/s]
4|4730|Loss: 0.6912336349487305:  73%|███████▎  | 4730/6470 [54:22<18:43,  1.55it/s]
4|4731|Loss: 0.8392252922058105:  73%|███████▎  | 4730/6470 [54:23<18:43,  1.55it/s]
4|4731|Loss: 0.8392252922058105:  73%|███████▎  | 4731/6470 [54:23<19:50,  1.46it/s]
4|4732|Loss: 0.7064083814620972:  73%|███████▎  | 4731/6470 [54:23<19:50,  1.46it/s]
4|4732|Loss: 0.7064083814620972:  73%|███████▎  | 4732/6470 [54:24<22:10,  1.31it/s]
4|4733|Loss: 0.84307861328125:  73%|███████▎  | 4732/6470 [54:24<22:10,  1.31it/s]  
4|4733|Loss: 0.84307861328125:  73%|███████▎  | 4733/6470 [54:25<22:34,  1.28it/s]
4|4734|Loss: 0.804624617099762:  73%|███████▎  | 4733/6470 [54:25<22:34,  1.28it/s]
4|4734|Loss: 0.804624617099762:  73%|███████▎  | 4734/6470 [54:25<21:15,  1.36it/s]
4|4735|Loss: 0.9141554236412048:  73%|███████▎  | 4734/6470 [54:26<21:15,  1.36it/s]
4|4735|Loss: 0.9141554236412048:  73%|███████▎  | 4735/6470 [54:26<21:00,  1.38it/s]
4|4736|Loss: 0.8159888386726379:  73%|███████▎  | 4735/6470 [54:26<21:00,  1.38it/s]
4|4736|Loss: 0.8159888386726379:  73%|███████▎  | 4736/6470 [54:27<19:37,  1.47it/s]
4|4737|Loss: 0.8135400414466858:  73%|███████▎  | 4736/6470 [54:27<19:37,  1.47it/s]
4|4737|Loss: 0.8135400414466858:  73%|███████▎  | 4737/6470 [54:27<19:41,  1.47it/s]
4|4738|Loss: 0.7498804926872253:  73%|███████▎  | 4737/6470 [54:28<19:41,  1.47it/s]
4|4738|Loss: 0.7498804926872253:  73%|███████▎  | 4738/6470 [54:28<19:21,  1.49it/s]
4|4739|Loss: 0.6807885766029358:  73%|███████▎  | 4738/6470 [54:28<19:21,  1.49it/s]
4|4739|Loss: 0.6807885766029358:  73%|███████▎  | 4739/6470 [54:29<21:25,  1.35it/s]
4|4740|Loss: 0.8738738894462585:  73%|███████▎  | 4739/6470 [54:29<21:25,  1.35it/s]
4|4740|Loss: 0.8738738894462585:  73%|███████▎  | 4740/6470 [54:30<21:04,  1.37it/s]
4|4741|Loss: 0.8712378144264221:  73%|███████▎  | 4740/6470 [54:30<21:04,  1.37it/s]
4|4741|Loss: 0.8712378144264221:  73%|███████▎  | 4741/6470 [54:30<20:05,  1.43it/s]
4|4742|Loss: 0.8914344906806946:  73%|███████▎  | 4741/6470 [54:30<20:05,  1.43it/s]
4|4742|Loss: 0.8914344906806946:  73%|███████▎  | 4742/6470 [54:31<18:59,  1.52it/s]
4|4743|Loss: 0.8181264996528625:  73%|███████▎  | 4742/6470 [54:31<18:59,  1.52it/s]
4|4743|Loss: 0.8181264996528625:  73%|███████▎  | 4743/6470 [54:31<19:20,  1.49it/s]
4|4744|Loss: 0.8627751469612122:  73%|███████▎  | 4743/6470 [54:32<19:20,  1.49it/s]
4|4744|Loss: 0.8627751469612122:  73%|███████▎  | 4744/6470 [54:32<19:17,  1.49it/s]
4|4745|Loss: 0.7917808294296265:  73%|███████▎  | 4744/6470 [54:32<19:17,  1.49it/s]
4|4745|Loss: 0.7917808294296265:  73%|███████▎  | 4745/6470 [54:33<18:09,  1.58it/s]
4|4746|Loss: 0.9244984984397888:  73%|███████▎  | 4745/6470 [54:33<18:09,  1.58it/s]
4|4746|Loss: 0.9244984984397888:  73%|███████▎  | 4746/6470 [54:33<18:38,  1.54it/s]
4|4747|Loss: 0.8195461630821228:  73%|███████▎  | 4746/6470 [54:34<18:38,  1.54it/s]
4|4747|Loss: 0.8195461630821228:  73%|███████▎  | 4747/6470 [54:34<19:54,  1.44it/s]
4|4748|Loss: 0.9916669130325317:  73%|███████▎  | 4747/6470 [54:34<19:54,  1.44it/s]
4|4748|Loss: 0.9916669130325317:  73%|███████▎  | 4748/6470 [54:35<19:45,  1.45it/s]
4|4749|Loss: 0.8579373955726624:  73%|███████▎  | 4748/6470 [54:35<19:45,  1.45it/s]
4|4749|Loss: 0.8579373955726624:  73%|███████▎  | 4749/6470 [54:35<18:15,  1.57it/s]
4|4750|Loss: 0.7613548040390015:  73%|███████▎  | 4749/6470 [54:36<18:15,  1.57it/s]
4|4750|Loss: 0.7613548040390015:  73%|███████▎  | 4750/6470 [54:36<19:31,  1.47it/s]
4|4751|Loss: 0.8400887250900269:  73%|███████▎  | 4750/6470 [54:37<19:31,  1.47it/s]
4|4751|Loss: 0.8400887250900269:  73%|███████▎  | 4751/6470 [54:37<20:04,  1.43it/s]
4|4752|Loss: 0.7670267224311829:  73%|███████▎  | 4751/6470 [54:37<20:04,  1.43it/s]
4|4752|Loss: 0.7670267224311829:  73%|███████▎  | 4752/6470 [54:38<20:06,  1.42it/s]
4|4753|Loss: 0.8389711976051331:  73%|███████▎  | 4752/6470 [54:38<20:06,  1.42it/s]
4|4753|Loss: 0.8389711976051331:  73%|███████▎  | 4753/6470 [54:38<19:42,  1.45it/s]
4|4754|Loss: 0.8301419615745544:  73%|███████▎  | 4753/6470 [54:39<19:42,  1.45it/s]
4|4754|Loss: 0.8301419615745544:  73%|███████▎  | 4754/6470 [54:39<19:40,  1.45it/s]
4|4755|Loss: 0.9652034044265747:  73%|███████▎  | 4754/6470 [54:39<19:40,  1.45it/s]
4|4755|Loss: 0.9652034044265747:  73%|███████▎  | 4755/6470 [54:40<19:09,  1.49it/s]
4|4756|Loss: 0.8821964263916016:  73%|███████▎  | 4755/6470 [54:40<19:09,  1.49it/s]
4|4756|Loss: 0.8821964263916016:  74%|███████▎  | 4756/6470 [54:40<20:33,  1.39it/s]
4|4757|Loss: 0.9320924282073975:  74%|███████▎  | 4756/6470 [54:41<20:33,  1.39it/s]
4|4757|Loss: 0.9320924282073975:  74%|███████▎  | 4757/6470 [54:41<20:23,  1.40it/s]
4|4758|Loss: 0.7402974963188171:  74%|███████▎  | 4757/6470 [54:41<20:23,  1.40it/s]
4|4758|Loss: 0.7402974963188171:  74%|███████▎  | 4758/6470 [54:42<18:30,  1.54it/s]
4|4759|Loss: 0.9555983543395996:  74%|███████▎  | 4758/6470 [54:42<18:30,  1.54it/s]
4|4759|Loss: 0.9555983543395996:  74%|███████▎  | 4759/6470 [54:42<17:25,  1.64it/s]
4|4760|Loss: 0.7521060705184937:  74%|███████▎  | 4759/6470 [54:42<17:25,  1.64it/s]
4|4760|Loss: 0.7521060705184937:  74%|███████▎  | 4760/6470 [54:43<16:53,  1.69it/s]
4|4761|Loss: 0.7630214691162109:  74%|███████▎  | 4760/6470 [54:43<16:53,  1.69it/s]
4|4761|Loss: 0.7630214691162109:  74%|███████▎  | 4761/6470 [54:43<17:56,  1.59it/s]
4|4762|Loss: 0.7161372900009155:  74%|███████▎  | 4761/6470 [54:44<17:56,  1.59it/s]
4|4762|Loss: 0.7161372900009155:  74%|███████▎  | 4762/6470 [54:44<16:20,  1.74it/s]
4|4763|Loss: 0.8464148640632629:  74%|███████▎  | 4762/6470 [54:44<16:20,  1.74it/s]
4|4763|Loss: 0.8464148640632629:  74%|███████▎  | 4763/6470 [54:44<17:05,  1.66it/s]
4|4764|Loss: 0.8789920210838318:  74%|███████▎  | 4763/6470 [54:45<17:05,  1.66it/s]
4|4764|Loss: 0.8789920210838318:  74%|███████▎  | 4764/6470 [54:45<18:15,  1.56it/s]
4|4765|Loss: 0.9561019539833069:  74%|███████▎  | 4764/6470 [54:46<18:15,  1.56it/s]
4|4765|Loss: 0.9561019539833069:  74%|███████▎  | 4765/6470 [54:46<18:59,  1.50it/s]
4|4766|Loss: 0.9899804592132568:  74%|███████▎  | 4765/6470 [54:46<18:59,  1.50it/s]
4|4766|Loss: 0.9899804592132568:  74%|███████▎  | 4766/6470 [54:47<18:45,  1.51it/s]
4|4767|Loss: 0.774136483669281:  74%|███████▎  | 4766/6470 [54:47<18:45,  1.51it/s] 
4|4767|Loss: 0.774136483669281:  74%|███████▎  | 4767/6470 [54:47<17:30,  1.62it/s]
4|4768|Loss: 0.8632081151008606:  74%|███████▎  | 4767/6470 [54:47<17:30,  1.62it/s]
4|4768|Loss: 0.8632081151008606:  74%|███████▎  | 4768/6470 [54:48<18:10,  1.56it/s]
4|4769|Loss: 0.8336371183395386:  74%|███████▎  | 4768/6470 [54:48<18:10,  1.56it/s]
4|4769|Loss: 0.8336371183395386:  74%|███████▎  | 4769/6470 [54:48<16:32,  1.71it/s]
4|4770|Loss: 1.0080337524414062:  74%|███████▎  | 4769/6470 [54:49<16:32,  1.71it/s]
4|4770|Loss: 1.0080337524414062:  74%|███████▎  | 4770/6470 [54:49<16:30,  1.72it/s]
4|4771|Loss: 0.834230363368988:  74%|███████▎  | 4770/6470 [54:49<16:30,  1.72it/s] 
4|4771|Loss: 0.834230363368988:  74%|███████▎  | 4771/6470 [54:50<18:06,  1.56it/s]
4|4772|Loss: 0.8216177225112915:  74%|███████▎  | 4771/6470 [54:50<18:06,  1.56it/s]
4|4772|Loss: 0.8216177225112915:  74%|███████▍  | 4772/6470 [54:50<18:17,  1.55it/s]
4|4773|Loss: 0.8276100158691406:  74%|███████▍  | 4772/6470 [54:51<18:17,  1.55it/s]
4|4773|Loss: 0.8276100158691406:  74%|███████▍  | 4773/6470 [54:51<18:44,  1.51it/s]
4|4774|Loss: 0.8694531917572021:  74%|███████▍  | 4773/6470 [54:51<18:44,  1.51it/s]
4|4774|Loss: 0.8694531917572021:  74%|███████▍  | 4774/6470 [54:52<19:06,  1.48it/s]
4|4775|Loss: 0.8214740753173828:  74%|███████▍  | 4774/6470 [54:52<19:06,  1.48it/s]
4|4775|Loss: 0.8214740753173828:  74%|███████▍  | 4775/6470 [54:52<18:46,  1.50it/s]
4|4776|Loss: 0.7584953904151917:  74%|███████▍  | 4775/6470 [54:53<18:46,  1.50it/s]
4|4776|Loss: 0.7584953904151917:  74%|███████▍  | 4776/6470 [54:53<19:46,  1.43it/s]
4|4777|Loss: 0.8636472225189209:  74%|███████▍  | 4776/6470 [54:53<19:46,  1.43it/s]
4|4777|Loss: 0.8636472225189209:  74%|███████▍  | 4777/6470 [54:54<19:36,  1.44it/s]
4|4778|Loss: 0.8273850679397583:  74%|███████▍  | 4777/6470 [54:54<19:36,  1.44it/s]
4|4778|Loss: 0.8273850679397583:  74%|███████▍  | 4778/6470 [54:54<19:09,  1.47it/s]
4|4779|Loss: 0.6826574206352234:  74%|███████▍  | 4778/6470 [54:55<19:09,  1.47it/s]
4|4779|Loss: 0.6826574206352234:  74%|███████▍  | 4779/6470 [54:55<19:14,  1.47it/s]
4|4780|Loss: 0.7067158222198486:  74%|███████▍  | 4779/6470 [54:55<19:14,  1.47it/s]
4|4780|Loss: 0.7067158222198486:  74%|███████▍  | 4780/6470 [54:56<18:13,  1.55it/s]
4|4781|Loss: 0.8847092390060425:  74%|███████▍  | 4780/6470 [54:56<18:13,  1.55it/s]
4|4781|Loss: 0.8847092390060425:  74%|███████▍  | 4781/6470 [54:56<17:31,  1.61it/s]
4|4782|Loss: 0.9496911764144897:  74%|███████▍  | 4781/6470 [54:57<17:31,  1.61it/s]
4|4782|Loss: 0.9496911764144897:  74%|███████▍  | 4782/6470 [54:57<17:23,  1.62it/s]
4|4783|Loss: 0.7364794015884399:  74%|███████▍  | 4782/6470 [54:57<17:23,  1.62it/s]
4|4783|Loss: 0.7364794015884399:  74%|███████▍  | 4783/6470 [54:57<16:53,  1.66it/s]
4|4784|Loss: 0.7516095042228699:  74%|███████▍  | 4783/6470 [54:58<16:53,  1.66it/s]
4|4784|Loss: 0.7516095042228699:  74%|███████▍  | 4784/6470 [54:58<17:31,  1.60it/s]
4|4785|Loss: 0.8044872283935547:  74%|███████▍  | 4784/6470 [54:58<17:31,  1.60it/s]
4|4785|Loss: 0.8044872283935547:  74%|███████▍  | 4785/6470 [54:59<18:08,  1.55it/s]
4|4786|Loss: 0.9035224914550781:  74%|███████▍  | 4785/6470 [54:59<18:08,  1.55it/s]
4|4786|Loss: 0.9035224914550781:  74%|███████▍  | 4786/6470 [54:59<18:12,  1.54it/s]
4|4787|Loss: 0.8692567348480225:  74%|███████▍  | 4786/6470 [55:00<18:12,  1.54it/s]
4|4787|Loss: 0.8692567348480225:  74%|███████▍  | 4787/6470 [55:00<19:22,  1.45it/s]
4|4788|Loss: 0.7472654581069946:  74%|███████▍  | 4787/6470 [55:01<19:22,  1.45it/s]
4|4788|Loss: 0.7472654581069946:  74%|███████▍  | 4788/6470 [55:01<20:28,  1.37it/s]
4|4789|Loss: 0.7946784496307373:  74%|███████▍  | 4788/6470 [55:01<20:28,  1.37it/s]
4|4789|Loss: 0.7946784496307373:  74%|███████▍  | 4789/6470 [55:02<20:30,  1.37it/s]
4|4790|Loss: 0.8506479263305664:  74%|███████▍  | 4789/6470 [55:02<20:30,  1.37it/s]
4|4790|Loss: 0.8506479263305664:  74%|███████▍  | 4790/6470 [55:03<21:07,  1.33it/s]
4|4791|Loss: 0.7147849202156067:  74%|███████▍  | 4790/6470 [55:03<21:07,  1.33it/s]
4|4791|Loss: 0.7147849202156067:  74%|███████▍  | 4791/6470 [55:03<19:27,  1.44it/s]
4|4792|Loss: 0.697162926197052:  74%|███████▍  | 4791/6470 [55:03<19:27,  1.44it/s] 
4|4792|Loss: 0.697162926197052:  74%|███████▍  | 4792/6470 [55:04<18:14,  1.53it/s]
4|4793|Loss: 0.8252164721488953:  74%|███████▍  | 4792/6470 [55:04<18:14,  1.53it/s]
4|4793|Loss: 0.8252164721488953:  74%|███████▍  | 4793/6470 [55:04<17:59,  1.55it/s]
4|4794|Loss: 0.9499621391296387:  74%|███████▍  | 4793/6470 [55:05<17:59,  1.55it/s]
4|4794|Loss: 0.9499621391296387:  74%|███████▍  | 4794/6470 [55:05<17:45,  1.57it/s]
4|4795|Loss: 0.8539236187934875:  74%|███████▍  | 4794/6470 [55:05<17:45,  1.57it/s]
4|4795|Loss: 0.8539236187934875:  74%|███████▍  | 4795/6470 [55:06<17:43,  1.57it/s]
4|4796|Loss: 0.8916439414024353:  74%|███████▍  | 4795/6470 [55:06<17:43,  1.57it/s]
4|4796|Loss: 0.8916439414024353:  74%|███████▍  | 4796/6470 [55:06<17:35,  1.59it/s]
4|4797|Loss: 0.9382479786872864:  74%|███████▍  | 4796/6470 [55:07<17:35,  1.59it/s]
4|4797|Loss: 0.9382479786872864:  74%|███████▍  | 4797/6470 [55:07<17:59,  1.55it/s]
4|4798|Loss: 1.1233619451522827:  74%|███████▍  | 4797/6470 [55:07<17:59,  1.55it/s]
4|4798|Loss: 1.1233619451522827:  74%|███████▍  | 4798/6470 [55:07<18:07,  1.54it/s]
4|4799|Loss: 0.8770172595977783:  74%|███████▍  | 4798/6470 [55:08<18:07,  1.54it/s]
4|4799|Loss: 0.8770172595977783:  74%|███████▍  | 4799/6470 [55:08<17:36,  1.58it/s]
4|4800|Loss: 0.819541871547699:  74%|███████▍  | 4799/6470 [55:08<17:36,  1.58it/s] 
4|4800|Loss: 0.819541871547699:  74%|███████▍  | 4800/6470 [55:09<17:10,  1.62it/s]
4|4801|Loss: 0.820248544216156:  74%|███████▍  | 4800/6470 [55:09<17:10,  1.62it/s]
4|4801|Loss: 0.820248544216156:  74%|███████▍  | 4801/6470 [55:09<17:39,  1.57it/s]
4|4802|Loss: 0.7604376673698425:  74%|███████▍  | 4801/6470 [55:10<17:39,  1.57it/s]
4|4802|Loss: 0.7604376673698425:  74%|███████▍  | 4802/6470 [55:10<17:47,  1.56it/s]
4|4803|Loss: 0.8617876768112183:  74%|███████▍  | 4802/6470 [55:10<17:47,  1.56it/s]
4|4803|Loss: 0.8617876768112183:  74%|███████▍  | 4803/6470 [55:11<17:37,  1.58it/s]
4|4804|Loss: 0.9272846579551697:  74%|███████▍  | 4803/6470 [55:11<17:37,  1.58it/s]
4|4804|Loss: 0.9272846579551697:  74%|███████▍  | 4804/6470 [55:11<17:40,  1.57it/s]
4|4805|Loss: 1.064475178718567:  74%|███████▍  | 4804/6470 [55:12<17:40,  1.57it/s] 
4|4805|Loss: 1.064475178718567:  74%|███████▍  | 4805/6470 [55:12<17:05,  1.62it/s]
4|4806|Loss: 0.7890700101852417:  74%|███████▍  | 4805/6470 [55:12<17:05,  1.62it/s]
4|4806|Loss: 0.7890700101852417:  74%|███████▍  | 4806/6470 [55:12<16:46,  1.65it/s]
4|4807|Loss: 0.7671107649803162:  74%|███████▍  | 4806/6470 [55:13<16:46,  1.65it/s]
4|4807|Loss: 0.7671107649803162:  74%|███████▍  | 4807/6470 [55:13<17:42,  1.57it/s]
4|4808|Loss: 0.8017193675041199:  74%|███████▍  | 4807/6470 [55:14<17:42,  1.57it/s]
4|4808|Loss: 0.8017193675041199:  74%|███████▍  | 4808/6470 [55:14<18:37,  1.49it/s]
4|4809|Loss: 0.8354602456092834:  74%|███████▍  | 4808/6470 [55:14<18:37,  1.49it/s]
4|4809|Loss: 0.8354602456092834:  74%|███████▍  | 4809/6470 [55:14<17:40,  1.57it/s]
4|4810|Loss: 1.0002808570861816:  74%|███████▍  | 4809/6470 [55:15<17:40,  1.57it/s]
4|4810|Loss: 1.0002808570861816:  74%|███████▍  | 4810/6470 [55:15<16:25,  1.68it/s]
4|4811|Loss: 0.8090311288833618:  74%|███████▍  | 4810/6470 [55:15<16:25,  1.68it/s]
4|4811|Loss: 0.8090311288833618:  74%|███████▍  | 4811/6470 [55:15<15:39,  1.77it/s]
4|4812|Loss: 0.9130647778511047:  74%|███████▍  | 4811/6470 [55:16<15:39,  1.77it/s]
4|4812|Loss: 0.9130647778511047:  74%|███████▍  | 4812/6470 [55:16<14:51,  1.86it/s]
4|4813|Loss: 1.1262873411178589:  74%|███████▍  | 4812/6470 [55:16<14:51,  1.86it/s]
4|4813|Loss: 1.1262873411178589:  74%|███████▍  | 4813/6470 [55:17<15:56,  1.73it/s]
4|4814|Loss: 1.061269760131836:  74%|███████▍  | 4813/6470 [55:17<15:56,  1.73it/s] 
4|4814|Loss: 1.061269760131836:  74%|███████▍  | 4814/6470 [55:17<16:37,  1.66it/s]
4|4815|Loss: 0.8320038914680481:  74%|███████▍  | 4814/6470 [55:18<16:37,  1.66it/s]
4|4815|Loss: 0.8320038914680481:  74%|███████▍  | 4815/6470 [55:18<17:33,  1.57it/s]
4|4816|Loss: 0.9445046186447144:  74%|███████▍  | 4815/6470 [55:18<17:33,  1.57it/s]
4|4816|Loss: 0.9445046186447144:  74%|███████▍  | 4816/6470 [55:19<17:36,  1.57it/s]
4|4817|Loss: 0.7946761250495911:  74%|███████▍  | 4816/6470 [55:19<17:36,  1.57it/s]
4|4817|Loss: 0.7946761250495911:  74%|███████▍  | 4817/6470 [55:19<16:59,  1.62it/s]
4|4818|Loss: 0.9094535708427429:  74%|███████▍  | 4817/6470 [55:20<16:59,  1.62it/s]
4|4818|Loss: 0.9094535708427429:  74%|███████▍  | 4818/6470 [55:20<17:42,  1.55it/s]
4|4819|Loss: 0.7123270034790039:  74%|███████▍  | 4818/6470 [55:20<17:42,  1.55it/s]
4|4819|Loss: 0.7123270034790039:  74%|███████▍  | 4819/6470 [55:21<18:33,  1.48it/s]
4|4820|Loss: 0.7459617853164673:  74%|███████▍  | 4819/6470 [55:21<18:33,  1.48it/s]
4|4820|Loss: 0.7459617853164673:  74%|███████▍  | 4820/6470 [55:21<19:28,  1.41it/s]
4|4821|Loss: 0.8855133652687073:  74%|███████▍  | 4820/6470 [55:22<19:28,  1.41it/s]
4|4821|Loss: 0.8855133652687073:  75%|███████▍  | 4821/6470 [55:22<20:14,  1.36it/s]
4|4822|Loss: 0.9054085612297058:  75%|███████▍  | 4821/6470 [55:23<20:14,  1.36it/s]
4|4822|Loss: 0.9054085612297058:  75%|███████▍  | 4822/6470 [55:23<19:34,  1.40it/s]
4|4823|Loss: 0.7452511191368103:  75%|███████▍  | 4822/6470 [55:23<19:34,  1.40it/s]
4|4823|Loss: 0.7452511191368103:  75%|███████▍  | 4823/6470 [55:24<21:16,  1.29it/s]
4|4824|Loss: 0.8937634825706482:  75%|███████▍  | 4823/6470 [55:24<21:16,  1.29it/s]
4|4824|Loss: 0.8937634825706482:  75%|███████▍  | 4824/6470 [55:25<21:35,  1.27it/s]
4|4825|Loss: 0.7282530069351196:  75%|███████▍  | 4824/6470 [55:25<21:35,  1.27it/s]
4|4825|Loss: 0.7282530069351196:  75%|███████▍  | 4825/6470 [55:25<22:02,  1.24it/s]
4|4826|Loss: 0.5779432654380798:  75%|███████▍  | 4825/6470 [55:26<22:02,  1.24it/s]
4|4826|Loss: 0.5779432654380798:  75%|███████▍  | 4826/6470 [55:26<20:57,  1.31it/s]
4|4827|Loss: 0.9049009680747986:  75%|███████▍  | 4826/6470 [55:27<20:57,  1.31it/s]
4|4827|Loss: 0.9049009680747986:  75%|███████▍  | 4827/6470 [55:27<21:29,  1.27it/s]
4|4828|Loss: 0.6981201171875:  75%|███████▍  | 4827/6470 [55:27<21:29,  1.27it/s]   
4|4828|Loss: 0.6981201171875:  75%|███████▍  | 4828/6470 [55:28<20:36,  1.33it/s]
4|4829|Loss: 0.8439971804618835:  75%|███████▍  | 4828/6470 [55:28<20:36,  1.33it/s]
4|4829|Loss: 0.8439971804618835:  75%|███████▍  | 4829/6470 [55:28<19:11,  1.43it/s]
4|4830|Loss: 0.8196513652801514:  75%|███████▍  | 4829/6470 [55:29<19:11,  1.43it/s]
4|4830|Loss: 0.8196513652801514:  75%|███████▍  | 4830/6470 [55:29<18:59,  1.44it/s]
4|4831|Loss: 0.755342960357666:  75%|███████▍  | 4830/6470 [55:29<18:59,  1.44it/s] 
4|4831|Loss: 0.755342960357666:  75%|███████▍  | 4831/6470 [55:30<19:44,  1.38it/s]
4|4832|Loss: 0.827242910861969:  75%|███████▍  | 4831/6470 [55:30<19:44,  1.38it/s]
4|4832|Loss: 0.827242910861969:  75%|███████▍  | 4832/6470 [55:30<18:08,  1.51it/s]
4|4833|Loss: 0.8314845561981201:  75%|███████▍  | 4832/6470 [55:31<18:08,  1.51it/s]
4|4833|Loss: 0.8314845561981201:  75%|███████▍  | 4833/6470 [55:31<18:11,  1.50it/s]
4|4834|Loss: 0.7390056848526001:  75%|███████▍  | 4833/6470 [55:31<18:11,  1.50it/s]
4|4834|Loss: 0.7390056848526001:  75%|███████▍  | 4834/6470 [55:32<18:51,  1.45it/s]
4|4835|Loss: 0.7616946697235107:  75%|███████▍  | 4834/6470 [55:32<18:51,  1.45it/s]
4|4835|Loss: 0.7616946697235107:  75%|███████▍  | 4835/6470 [55:32<17:46,  1.53it/s]
4|4836|Loss: 0.7507926225662231:  75%|███████▍  | 4835/6470 [55:33<17:46,  1.53it/s]
4|4836|Loss: 0.7507926225662231:  75%|███████▍  | 4836/6470 [55:33<18:53,  1.44it/s]
4|4837|Loss: 0.9056960940361023:  75%|███████▍  | 4836/6470 [55:33<18:53,  1.44it/s]
4|4837|Loss: 0.9056960940361023:  75%|███████▍  | 4837/6470 [55:34<20:13,  1.35it/s]
4|4838|Loss: 0.847160279750824:  75%|███████▍  | 4837/6470 [55:34<20:13,  1.35it/s] 
4|4838|Loss: 0.847160279750824:  75%|███████▍  | 4838/6470 [55:35<20:29,  1.33it/s]
4|4839|Loss: 0.7730237245559692:  75%|███████▍  | 4838/6470 [55:35<20:29,  1.33it/s]
4|4839|Loss: 0.7730237245559692:  75%|███████▍  | 4839/6470 [55:35<19:11,  1.42it/s]
4|4840|Loss: 0.8309831619262695:  75%|███████▍  | 4839/6470 [55:36<19:11,  1.42it/s]
4|4840|Loss: 0.8309831619262695:  75%|███████▍  | 4840/6470 [55:36<19:07,  1.42it/s]
4|4841|Loss: 0.8705542683601379:  75%|███████▍  | 4840/6470 [55:36<19:07,  1.42it/s]
4|4841|Loss: 0.8705542683601379:  75%|███████▍  | 4841/6470 [55:37<19:29,  1.39it/s]
4|4842|Loss: 0.8372061848640442:  75%|███████▍  | 4841/6470 [55:37<19:29,  1.39it/s]
4|4842|Loss: 0.8372061848640442:  75%|███████▍  | 4842/6470 [55:37<18:27,  1.47it/s]
4|4843|Loss: 0.8525254130363464:  75%|███████▍  | 4842/6470 [55:38<18:27,  1.47it/s]
4|4843|Loss: 0.8525254130363464:  75%|███████▍  | 4843/6470 [55:38<17:50,  1.52it/s]
4|4844|Loss: 0.7067893743515015:  75%|███████▍  | 4843/6470 [55:38<17:50,  1.52it/s]
4|4844|Loss: 0.7067893743515015:  75%|███████▍  | 4844/6470 [55:39<18:54,  1.43it/s]
4|4845|Loss: 0.9098767638206482:  75%|███████▍  | 4844/6470 [55:39<18:54,  1.43it/s]
4|4845|Loss: 0.9098767638206482:  75%|███████▍  | 4845/6470 [55:39<18:49,  1.44it/s]
4|4846|Loss: 0.806847333908081:  75%|███████▍  | 4845/6470 [55:40<18:49,  1.44it/s] 
4|4846|Loss: 0.806847333908081:  75%|███████▍  | 4846/6470 [55:40<19:04,  1.42it/s]
4|4847|Loss: 0.754947304725647:  75%|███████▍  | 4846/6470 [55:40<19:04,  1.42it/s]
4|4847|Loss: 0.754947304725647:  75%|███████▍  | 4847/6470 [55:41<19:28,  1.39it/s]
4|4848|Loss: 0.8457390666007996:  75%|███████▍  | 4847/6470 [55:41<19:28,  1.39it/s]
4|4848|Loss: 0.8457390666007996:  75%|███████▍  | 4848/6470 [55:42<20:23,  1.33it/s]
4|4849|Loss: 0.9344731569290161:  75%|███████▍  | 4848/6470 [55:42<20:23,  1.33it/s]
4|4849|Loss: 0.9344731569290161:  75%|███████▍  | 4849/6470 [55:42<19:52,  1.36it/s]
4|4850|Loss: 0.9936631917953491:  75%|███████▍  | 4849/6470 [55:43<19:52,  1.36it/s]
4|4850|Loss: 0.9936631917953491:  75%|███████▍  | 4850/6470 [55:43<19:05,  1.41it/s]
4|4851|Loss: 0.8948091864585876:  75%|███████▍  | 4850/6470 [55:43<19:05,  1.41it/s]
4|4851|Loss: 0.8948091864585876:  75%|███████▍  | 4851/6470 [55:44<20:21,  1.33it/s]
4|4852|Loss: 0.8742392659187317:  75%|███████▍  | 4851/6470 [55:44<20:21,  1.33it/s]
4|4852|Loss: 0.8742392659187317:  75%|███████▍  | 4852/6470 [55:44<19:31,  1.38it/s]
4|4853|Loss: 0.725887656211853:  75%|███████▍  | 4852/6470 [55:45<19:31,  1.38it/s] 
4|4853|Loss: 0.725887656211853:  75%|███████▌  | 4853/6470 [55:45<19:57,  1.35it/s]
4|4854|Loss: 0.7746119499206543:  75%|███████▌  | 4853/6470 [55:46<19:57,  1.35it/s]
4|4854|Loss: 0.7746119499206543:  75%|███████▌  | 4854/6470 [55:46<19:26,  1.38it/s]
4|4855|Loss: 0.922356128692627:  75%|███████▌  | 4854/6470 [55:46<19:26,  1.38it/s] 
4|4855|Loss: 0.922356128692627:  75%|███████▌  | 4855/6470 [55:47<18:37,  1.44it/s]
4|4856|Loss: 0.8156102895736694:  75%|███████▌  | 4855/6470 [55:47<18:37,  1.44it/s]
4|4856|Loss: 0.8156102895736694:  75%|███████▌  | 4856/6470 [55:47<16:57,  1.59it/s]
4|4857|Loss: 0.8579955101013184:  75%|███████▌  | 4856/6470 [55:47<16:57,  1.59it/s]
4|4857|Loss: 0.8579955101013184:  75%|███████▌  | 4857/6470 [55:48<17:35,  1.53it/s]
4|4858|Loss: 0.6803469657897949:  75%|███████▌  | 4857/6470 [55:48<17:35,  1.53it/s]
4|4858|Loss: 0.6803469657897949:  75%|███████▌  | 4858/6470 [55:49<18:20,  1.46it/s]
4|4859|Loss: 0.8389129042625427:  75%|███████▌  | 4858/6470 [55:49<18:20,  1.46it/s]
4|4859|Loss: 0.8389129042625427:  75%|███████▌  | 4859/6470 [55:49<18:42,  1.43it/s]
4|4860|Loss: 0.7529618144035339:  75%|███████▌  | 4859/6470 [55:50<18:42,  1.43it/s]
4|4860|Loss: 0.7529618144035339:  75%|███████▌  | 4860/6470 [55:50<18:00,  1.49it/s]
4|4861|Loss: 0.7855164408683777:  75%|███████▌  | 4860/6470 [55:50<18:00,  1.49it/s]
4|4861|Loss: 0.7855164408683777:  75%|███████▌  | 4861/6470 [55:50<17:16,  1.55it/s]
4|4862|Loss: 0.770932137966156:  75%|███████▌  | 4861/6470 [55:51<17:16,  1.55it/s] 
4|4862|Loss: 0.770932137966156:  75%|███████▌  | 4862/6470 [55:51<16:47,  1.60it/s]
4|4863|Loss: 0.895272970199585:  75%|███████▌  | 4862/6470 [55:51<16:47,  1.60it/s]
4|4863|Loss: 0.895272970199585:  75%|███████▌  | 4863/6470 [55:52<16:50,  1.59it/s]
4|4864|Loss: 0.8723468780517578:  75%|███████▌  | 4863/6470 [55:52<16:50,  1.59it/s]
4|4864|Loss: 0.8723468780517578:  75%|███████▌  | 4864/6470 [55:52<17:22,  1.54it/s]
4|4865|Loss: 0.9566315412521362:  75%|███████▌  | 4864/6470 [55:53<17:22,  1.54it/s]
4|4865|Loss: 0.9566315412521362:  75%|███████▌  | 4865/6470 [55:53<18:37,  1.44it/s]
4|4866|Loss: 0.7915664911270142:  75%|███████▌  | 4865/6470 [55:54<18:37,  1.44it/s]
4|4866|Loss: 0.7915664911270142:  75%|███████▌  | 4866/6470 [55:54<17:45,  1.50it/s]
4|4867|Loss: 0.9685488939285278:  75%|███████▌  | 4866/6470 [55:54<17:45,  1.50it/s]
4|4867|Loss: 0.9685488939285278:  75%|███████▌  | 4867/6470 [55:55<18:55,  1.41it/s]
4|4868|Loss: 0.9231764674186707:  75%|███████▌  | 4867/6470 [55:55<18:55,  1.41it/s]
4|4868|Loss: 0.9231764674186707:  75%|███████▌  | 4868/6470 [55:55<18:48,  1.42it/s]
4|4869|Loss: 0.8177639842033386:  75%|███████▌  | 4868/6470 [55:56<18:48,  1.42it/s]
4|4869|Loss: 0.8177639842033386:  75%|███████▌  | 4869/6470 [55:56<18:58,  1.41it/s]
4|4870|Loss: 0.7990959286689758:  75%|███████▌  | 4869/6470 [55:56<18:58,  1.41it/s]
4|4870|Loss: 0.7990959286689758:  75%|███████▌  | 4870/6470 [55:57<18:00,  1.48it/s]
4|4871|Loss: 0.7387761473655701:  75%|███████▌  | 4870/6470 [55:57<18:00,  1.48it/s]
4|4871|Loss: 0.7387761473655701:  75%|███████▌  | 4871/6470 [55:57<16:24,  1.62it/s]
4|4872|Loss: 0.7405458688735962:  75%|███████▌  | 4871/6470 [55:57<16:24,  1.62it/s]
4|4872|Loss: 0.7405458688735962:  75%|███████▌  | 4872/6470 [55:58<15:49,  1.68it/s]
4|4873|Loss: 0.8670928478240967:  75%|███████▌  | 4872/6470 [55:58<15:49,  1.68it/s]
4|4873|Loss: 0.8670928478240967:  75%|███████▌  | 4873/6470 [55:58<16:27,  1.62it/s]
4|4874|Loss: 1.0677529573440552:  75%|███████▌  | 4873/6470 [55:59<16:27,  1.62it/s]
4|4874|Loss: 1.0677529573440552:  75%|███████▌  | 4874/6470 [55:59<15:54,  1.67it/s]
4|4875|Loss: 0.9193353056907654:  75%|███████▌  | 4874/6470 [55:59<15:54,  1.67it/s]
4|4875|Loss: 0.9193353056907654:  75%|███████▌  | 4875/6470 [55:59<16:00,  1.66it/s]
4|4876|Loss: 0.639075517654419:  75%|███████▌  | 4875/6470 [56:00<16:00,  1.66it/s] 
4|4876|Loss: 0.639075517654419:  75%|███████▌  | 4876/6470 [56:00<15:54,  1.67it/s]
4|4877|Loss: 0.8449187874794006:  75%|███████▌  | 4876/6470 [56:00<15:54,  1.67it/s]
4|4877|Loss: 0.8449187874794006:  75%|███████▌  | 4877/6470 [56:01<15:44,  1.69it/s]
4|4878|Loss: 0.8775473237037659:  75%|███████▌  | 4877/6470 [56:01<15:44,  1.69it/s]
4|4878|Loss: 0.8775473237037659:  75%|███████▌  | 4878/6470 [56:01<17:13,  1.54it/s]
4|4879|Loss: 0.7277828454971313:  75%|███████▌  | 4878/6470 [56:02<17:13,  1.54it/s]
4|4879|Loss: 0.7277828454971313:  75%|███████▌  | 4879/6470 [56:02<16:42,  1.59it/s]
4|4880|Loss: 0.919498085975647:  75%|███████▌  | 4879/6470 [56:02<16:42,  1.59it/s] 
4|4880|Loss: 0.919498085975647:  75%|███████▌  | 4880/6470 [56:03<16:59,  1.56it/s]
4|4881|Loss: 0.8370033502578735:  75%|███████▌  | 4880/6470 [56:03<16:59,  1.56it/s]
4|4881|Loss: 0.8370033502578735:  75%|███████▌  | 4881/6470 [56:03<18:02,  1.47it/s]
4|4882|Loss: 0.8960676789283752:  75%|███████▌  | 4881/6470 [56:04<18:02,  1.47it/s]
4|4882|Loss: 0.8960676789283752:  75%|███████▌  | 4882/6470 [56:04<18:35,  1.42it/s]
4|4883|Loss: 0.9736860394477844:  75%|███████▌  | 4882/6470 [56:05<18:35,  1.42it/s]
4|4883|Loss: 0.9736860394477844:  75%|███████▌  | 4883/6470 [56:05<18:55,  1.40it/s]
4|4884|Loss: 0.7238934636116028:  75%|███████▌  | 4883/6470 [56:05<18:55,  1.40it/s]
4|4884|Loss: 0.7238934636116028:  75%|███████▌  | 4884/6470 [56:06<19:35,  1.35it/s]
4|4885|Loss: 0.8349787592887878:  75%|███████▌  | 4884/6470 [56:06<19:35,  1.35it/s]
4|4885|Loss: 0.8349787592887878:  76%|███████▌  | 4885/6470 [56:07<20:15,  1.30it/s]
4|4886|Loss: 0.8926721215248108:  76%|███████▌  | 4885/6470 [56:07<20:15,  1.30it/s]
4|4886|Loss: 0.8926721215248108:  76%|███████▌  | 4886/6470 [56:07<19:33,  1.35it/s]
4|4887|Loss: 0.8641876578330994:  76%|███████▌  | 4886/6470 [56:08<19:33,  1.35it/s]
4|4887|Loss: 0.8641876578330994:  76%|███████▌  | 4887/6470 [56:08<19:20,  1.36it/s]
4|4888|Loss: 0.7995777130126953:  76%|███████▌  | 4887/6470 [56:08<19:20,  1.36it/s]
4|4888|Loss: 0.7995777130126953:  76%|███████▌  | 4888/6470 [56:09<18:11,  1.45it/s]
4|4889|Loss: 0.8340409398078918:  76%|███████▌  | 4888/6470 [56:09<18:11,  1.45it/s]
4|4889|Loss: 0.8340409398078918:  76%|███████▌  | 4889/6470 [56:09<18:19,  1.44it/s]
4|4890|Loss: 0.8604069352149963:  76%|███████▌  | 4889/6470 [56:10<18:19,  1.44it/s]
4|4890|Loss: 0.8604069352149963:  76%|███████▌  | 4890/6470 [56:10<18:42,  1.41it/s]
4|4891|Loss: 0.7309494018554688:  76%|███████▌  | 4890/6470 [56:10<18:42,  1.41it/s]
4|4891|Loss: 0.7309494018554688:  76%|███████▌  | 4891/6470 [56:11<18:41,  1.41it/s]
4|4892|Loss: 0.8966953754425049:  76%|███████▌  | 4891/6470 [56:11<18:41,  1.41it/s]
4|4892|Loss: 0.8966953754425049:  76%|███████▌  | 4892/6470 [56:12<19:34,  1.34it/s]
4|4893|Loss: 0.8390319347381592:  76%|███████▌  | 4892/6470 [56:12<19:34,  1.34it/s]
4|4893|Loss: 0.8390319347381592:  76%|███████▌  | 4893/6470 [56:12<19:09,  1.37it/s]
4|4894|Loss: 0.7911163568496704:  76%|███████▌  | 4893/6470 [56:13<19:09,  1.37it/s]
4|4894|Loss: 0.7911163568496704:  76%|███████▌  | 4894/6470 [56:13<19:02,  1.38it/s]
4|4895|Loss: 0.8860380053520203:  76%|███████▌  | 4894/6470 [56:13<19:02,  1.38it/s]
4|4895|Loss: 0.8860380053520203:  76%|███████▌  | 4895/6470 [56:14<19:09,  1.37it/s]
4|4896|Loss: 1.0233619213104248:  76%|███████▌  | 4895/6470 [56:14<19:09,  1.37it/s]
4|4896|Loss: 1.0233619213104248:  76%|███████▌  | 4896/6470 [56:14<18:43,  1.40it/s]
4|4897|Loss: 0.8934272527694702:  76%|███████▌  | 4896/6470 [56:15<18:43,  1.40it/s]
4|4897|Loss: 0.8934272527694702:  76%|███████▌  | 4897/6470 [56:15<18:51,  1.39it/s]
4|4898|Loss: 0.7802104949951172:  76%|███████▌  | 4897/6470 [56:15<18:51,  1.39it/s]
4|4898|Loss: 0.7802104949951172:  76%|███████▌  | 4898/6470 [56:16<18:26,  1.42it/s]
4|4899|Loss: 0.8656177520751953:  76%|███████▌  | 4898/6470 [56:16<18:26,  1.42it/s]
4|4899|Loss: 0.8656177520751953:  76%|███████▌  | 4899/6470 [56:17<19:07,  1.37it/s]
4|4900|Loss: 0.8902344107627869:  76%|███████▌  | 4899/6470 [56:17<19:07,  1.37it/s]
4|4900|Loss: 0.8902344107627869:  76%|███████▌  | 4900/6470 [56:17<18:36,  1.41it/s]
4|4901|Loss: 0.8688321113586426:  76%|███████▌  | 4900/6470 [56:18<18:36,  1.41it/s]
4|4901|Loss: 0.8688321113586426:  76%|███████▌  | 4901/6470 [56:18<17:25,  1.50it/s]
4|4902|Loss: 0.8573628664016724:  76%|███████▌  | 4901/6470 [56:18<17:25,  1.50it/s]
4|4902|Loss: 0.8573628664016724:  76%|███████▌  | 4902/6470 [56:18<17:39,  1.48it/s]
4|4903|Loss: 0.7859606742858887:  76%|███████▌  | 4902/6470 [56:19<17:39,  1.48it/s]
4|4903|Loss: 0.7859606742858887:  76%|███████▌  | 4903/6470 [56:19<19:33,  1.34it/s]
4|4904|Loss: 0.8628138303756714:  76%|███████▌  | 4903/6470 [56:20<19:33,  1.34it/s]
4|4904|Loss: 0.8628138303756714:  76%|███████▌  | 4904/6470 [56:20<19:55,  1.31it/s]
4|4905|Loss: 0.8847334384918213:  76%|███████▌  | 4904/6470 [56:21<19:55,  1.31it/s]
4|4905|Loss: 0.8847334384918213:  76%|███████▌  | 4905/6470 [56:21<19:07,  1.36it/s]
4|4906|Loss: 1.005781650543213:  76%|███████▌  | 4905/6470 [56:21<19:07,  1.36it/s] 
4|4906|Loss: 1.005781650543213:  76%|███████▌  | 4906/6470 [56:21<18:22,  1.42it/s]
4|4907|Loss: 0.8845903873443604:  76%|███████▌  | 4906/6470 [56:22<18:22,  1.42it/s]
4|4907|Loss: 0.8845903873443604:  76%|███████▌  | 4907/6470 [56:22<18:15,  1.43it/s]
4|4908|Loss: 0.8173673748970032:  76%|███████▌  | 4907/6470 [56:23<18:15,  1.43it/s]
4|4908|Loss: 0.8173673748970032:  76%|███████▌  | 4908/6470 [56:23<18:26,  1.41it/s]
4|4909|Loss: 0.7567065954208374:  76%|███████▌  | 4908/6470 [56:23<18:26,  1.41it/s]
4|4909|Loss: 0.7567065954208374:  76%|███████▌  | 4909/6470 [56:24<19:22,  1.34it/s]
4|4910|Loss: 0.7051061391830444:  76%|███████▌  | 4909/6470 [56:24<19:22,  1.34it/s]
4|4910|Loss: 0.7051061391830444:  76%|███████▌  | 4910/6470 [56:24<17:44,  1.47it/s]
4|4911|Loss: 0.6979080438613892:  76%|███████▌  | 4910/6470 [56:25<17:44,  1.47it/s]
4|4911|Loss: 0.6979080438613892:  76%|███████▌  | 4911/6470 [56:25<18:21,  1.41it/s]
4|4912|Loss: 0.9249285459518433:  76%|███████▌  | 4911/6470 [56:25<18:21,  1.41it/s]
4|4912|Loss: 0.9249285459518433:  76%|███████▌  | 4912/6470 [56:26<17:55,  1.45it/s]
4|4913|Loss: 1.009455919265747:  76%|███████▌  | 4912/6470 [56:26<17:55,  1.45it/s] 
4|4913|Loss: 1.009455919265747:  76%|███████▌  | 4913/6470 [56:26<18:34,  1.40it/s]
4|4914|Loss: 0.8958720564842224:  76%|███████▌  | 4913/6470 [56:27<18:34,  1.40it/s]
4|4914|Loss: 0.8958720564842224:  76%|███████▌  | 4914/6470 [56:27<18:57,  1.37it/s]
4|4915|Loss: 0.9528336524963379:  76%|███████▌  | 4914/6470 [56:28<18:57,  1.37it/s]
4|4915|Loss: 0.9528336524963379:  76%|███████▌  | 4915/6470 [56:28<18:18,  1.42it/s]
4|4916|Loss: 0.8700599074363708:  76%|███████▌  | 4915/6470 [56:28<18:18,  1.42it/s]
4|4916|Loss: 0.8700599074363708:  76%|███████▌  | 4916/6470 [56:28<17:34,  1.47it/s]
4|4917|Loss: 0.783218264579773:  76%|███████▌  | 4916/6470 [56:29<17:34,  1.47it/s] 
4|4917|Loss: 0.783218264579773:  76%|███████▌  | 4917/6470 [56:29<16:50,  1.54it/s]
4|4918|Loss: 0.7981314063072205:  76%|███████▌  | 4917/6470 [56:29<16:50,  1.54it/s]
4|4918|Loss: 0.7981314063072205:  76%|███████▌  | 4918/6470 [56:30<17:53,  1.45it/s]
4|4919|Loss: 0.8395183086395264:  76%|███████▌  | 4918/6470 [56:30<17:53,  1.45it/s]
4|4919|Loss: 0.8395183086395264:  76%|███████▌  | 4919/6470 [56:31<17:46,  1.45it/s]
4|4920|Loss: 0.922631561756134:  76%|███████▌  | 4919/6470 [56:31<17:46,  1.45it/s] 
4|4920|Loss: 0.922631561756134:  76%|███████▌  | 4920/6470 [56:31<16:35,  1.56it/s]
4|4921|Loss: 0.8787832260131836:  76%|███████▌  | 4920/6470 [56:31<16:35,  1.56it/s]
4|4921|Loss: 0.8787832260131836:  76%|███████▌  | 4921/6470 [56:32<16:53,  1.53it/s]
4|4922|Loss: 0.8691994547843933:  76%|███████▌  | 4921/6470 [56:32<16:53,  1.53it/s]
4|4922|Loss: 0.8691994547843933:  76%|███████▌  | 4922/6470 [56:32<17:19,  1.49it/s]
4|4923|Loss: 0.8374605178833008:  76%|███████▌  | 4922/6470 [56:33<17:19,  1.49it/s]
4|4923|Loss: 0.8374605178833008:  76%|███████▌  | 4923/6470 [56:33<17:11,  1.50it/s]
4|4924|Loss: 1.025558590888977:  76%|███████▌  | 4923/6470 [56:34<17:11,  1.50it/s] 
4|4924|Loss: 1.025558590888977:  76%|███████▌  | 4924/6470 [56:34<17:53,  1.44it/s]
4|4925|Loss: 0.8198798894882202:  76%|███████▌  | 4924/6470 [56:34<17:53,  1.44it/s]
4|4925|Loss: 0.8198798894882202:  76%|███████▌  | 4925/6470 [56:35<18:51,  1.37it/s]
4|4926|Loss: 0.7929001450538635:  76%|███████▌  | 4925/6470 [56:35<18:51,  1.37it/s]
4|4926|Loss: 0.7929001450538635:  76%|███████▌  | 4926/6470 [56:35<17:14,  1.49it/s]
4|4927|Loss: 0.8043982982635498:  76%|███████▌  | 4926/6470 [56:36<17:14,  1.49it/s]
4|4927|Loss: 0.8043982982635498:  76%|███████▌  | 4927/6470 [56:36<17:00,  1.51it/s]
4|4928|Loss: 0.849673330783844:  76%|███████▌  | 4927/6470 [56:36<17:00,  1.51it/s] 
4|4928|Loss: 0.849673330783844:  76%|███████▌  | 4928/6470 [56:37<18:16,  1.41it/s]
4|4929|Loss: 0.744522750377655:  76%|███████▌  | 4928/6470 [56:37<18:16,  1.41it/s]
4|4929|Loss: 0.744522750377655:  76%|███████▌  | 4929/6470 [56:37<17:33,  1.46it/s]
4|4930|Loss: 0.832578718662262:  76%|███████▌  | 4929/6470 [56:38<17:33,  1.46it/s]
4|4930|Loss: 0.832578718662262:  76%|███████▌  | 4930/6470 [56:38<18:21,  1.40it/s]
4|4931|Loss: 0.7476569414138794:  76%|███████▌  | 4930/6470 [56:39<18:21,  1.40it/s]
4|4931|Loss: 0.7476569414138794:  76%|███████▌  | 4931/6470 [56:39<18:08,  1.41it/s]
4|4932|Loss: 0.7492802143096924:  76%|███████▌  | 4931/6470 [56:39<18:08,  1.41it/s]
4|4932|Loss: 0.7492802143096924:  76%|███████▌  | 4932/6470 [56:39<17:12,  1.49it/s]
4|4933|Loss: 0.8943567872047424:  76%|███████▌  | 4932/6470 [56:40<17:12,  1.49it/s]
4|4933|Loss: 0.8943567872047424:  76%|███████▌  | 4933/6470 [56:40<17:48,  1.44it/s]
4|4934|Loss: 0.8904349207878113:  76%|███████▌  | 4933/6470 [56:41<17:48,  1.44it/s]
4|4934|Loss: 0.8904349207878113:  76%|███████▋  | 4934/6470 [56:41<18:13,  1.40it/s]
4|4935|Loss: 0.8784456253051758:  76%|███████▋  | 4934/6470 [56:41<18:13,  1.40it/s]
4|4935|Loss: 0.8784456253051758:  76%|███████▋  | 4935/6470 [56:42<18:46,  1.36it/s]
4|4936|Loss: 0.7560994625091553:  76%|███████▋  | 4935/6470 [56:42<18:46,  1.36it/s]
4|4936|Loss: 0.7560994625091553:  76%|███████▋  | 4936/6470 [56:42<18:52,  1.35it/s]
4|4937|Loss: 0.8663104772567749:  76%|███████▋  | 4936/6470 [56:43<18:52,  1.35it/s]
4|4937|Loss: 0.8663104772567749:  76%|███████▋  | 4937/6470 [56:43<18:48,  1.36it/s]
4|4938|Loss: 0.8937925696372986:  76%|███████▋  | 4937/6470 [56:44<18:48,  1.36it/s]
4|4938|Loss: 0.8937925696372986:  76%|███████▋  | 4938/6470 [56:44<18:45,  1.36it/s]
4|4939|Loss: 0.9263280630111694:  76%|███████▋  | 4938/6470 [56:44<18:45,  1.36it/s]
4|4939|Loss: 0.9263280630111694:  76%|███████▋  | 4939/6470 [56:44<17:49,  1.43it/s]
4|4940|Loss: 0.8026006817817688:  76%|███████▋  | 4939/6470 [56:45<17:49,  1.43it/s]
4|4940|Loss: 0.8026006817817688:  76%|███████▋  | 4940/6470 [56:45<17:13,  1.48it/s]
4|4941|Loss: 0.7288078665733337:  76%|███████▋  | 4940/6470 [56:45<17:13,  1.48it/s]
4|4941|Loss: 0.7288078665733337:  76%|███████▋  | 4941/6470 [56:46<17:35,  1.45it/s]
4|4942|Loss: 0.9039919972419739:  76%|███████▋  | 4941/6470 [56:46<17:35,  1.45it/s]
4|4942|Loss: 0.9039919972419739:  76%|███████▋  | 4942/6470 [56:47<18:43,  1.36it/s]
4|4943|Loss: 0.7043366432189941:  76%|███████▋  | 4942/6470 [56:47<18:43,  1.36it/s]
4|4943|Loss: 0.7043366432189941:  76%|███████▋  | 4943/6470 [56:48<19:30,  1.30it/s]
4|4944|Loss: 0.7851274609565735:  76%|███████▋  | 4943/6470 [56:48<19:30,  1.30it/s]
4|4944|Loss: 0.7851274609565735:  76%|███████▋  | 4944/6470 [56:48<18:49,  1.35it/s]
4|4945|Loss: 0.9526014924049377:  76%|███████▋  | 4944/6470 [56:49<18:49,  1.35it/s]
4|4945|Loss: 0.9526014924049377:  76%|███████▋  | 4945/6470 [56:49<18:39,  1.36it/s]
4|4946|Loss: 0.7635068297386169:  76%|███████▋  | 4945/6470 [56:49<18:39,  1.36it/s]
4|4946|Loss: 0.7635068297386169:  76%|███████▋  | 4946/6470 [56:49<16:24,  1.55it/s]
4|4947|Loss: 0.8749199509620667:  76%|███████▋  | 4946/6470 [56:50<16:24,  1.55it/s]
4|4947|Loss: 0.8749199509620667:  76%|███████▋  | 4947/6470 [56:50<15:49,  1.60it/s]
4|4948|Loss: 0.7997658848762512:  76%|███████▋  | 4947/6470 [56:50<15:49,  1.60it/s]
4|4948|Loss: 0.7997658848762512:  76%|███████▋  | 4948/6470 [56:51<17:03,  1.49it/s]
4|4949|Loss: 0.656624436378479:  76%|███████▋  | 4948/6470 [56:51<17:03,  1.49it/s] 
4|4949|Loss: 0.656624436378479:  76%|███████▋  | 4949/6470 [56:51<16:31,  1.53it/s]
4|4950|Loss: 0.927152156829834:  76%|███████▋  | 4949/6470 [56:52<16:31,  1.53it/s]
4|4950|Loss: 0.927152156829834:  77%|███████▋  | 4950/6470 [56:52<16:41,  1.52it/s]
4|4951|Loss: 0.9219260811805725:  77%|███████▋  | 4950/6470 [56:52<16:41,  1.52it/s]
4|4951|Loss: 0.9219260811805725:  77%|███████▋  | 4951/6470 [56:53<17:04,  1.48it/s]
4|4952|Loss: 0.8683992028236389:  77%|███████▋  | 4951/6470 [56:53<17:04,  1.48it/s]
4|4952|Loss: 0.8683992028236389:  77%|███████▋  | 4952/6470 [56:53<16:28,  1.54it/s]
4|4953|Loss: 0.8947715163230896:  77%|███████▋  | 4952/6470 [56:54<16:28,  1.54it/s]
4|4953|Loss: 0.8947715163230896:  77%|███████▋  | 4953/6470 [56:54<15:33,  1.62it/s]
4|4954|Loss: 0.8539491295814514:  77%|███████▋  | 4953/6470 [56:54<15:33,  1.62it/s]
4|4954|Loss: 0.8539491295814514:  77%|███████▋  | 4954/6470 [56:55<17:06,  1.48it/s]
4|4955|Loss: 0.80016028881073:  77%|███████▋  | 4954/6470 [56:55<17:06,  1.48it/s]  
4|4955|Loss: 0.80016028881073:  77%|███████▋  | 4955/6470 [56:55<17:10,  1.47it/s]
4|4956|Loss: 0.8945826292037964:  77%|███████▋  | 4955/6470 [56:56<17:10,  1.47it/s]
4|4956|Loss: 0.8945826292037964:  77%|███████▋  | 4956/6470 [56:56<17:24,  1.45it/s]
4|4957|Loss: 0.8398132920265198:  77%|███████▋  | 4956/6470 [56:56<17:24,  1.45it/s]
4|4957|Loss: 0.8398132920265198:  77%|███████▋  | 4957/6470 [56:57<17:11,  1.47it/s]
4|4958|Loss: 0.7345154285430908:  77%|███████▋  | 4957/6470 [56:57<17:11,  1.47it/s]
4|4958|Loss: 0.7345154285430908:  77%|███████▋  | 4958/6470 [56:57<16:42,  1.51it/s]
4|4959|Loss: 0.7744976878166199:  77%|███████▋  | 4958/6470 [56:58<16:42,  1.51it/s]
4|4959|Loss: 0.7744976878166199:  77%|███████▋  | 4959/6470 [56:58<16:19,  1.54it/s]
4|4960|Loss: 0.8748197555541992:  77%|███████▋  | 4959/6470 [56:58<16:19,  1.54it/s]
4|4960|Loss: 0.8748197555541992:  77%|███████▋  | 4960/6470 [56:59<16:54,  1.49it/s]
4|4961|Loss: 0.6778329014778137:  77%|███████▋  | 4960/6470 [56:59<16:54,  1.49it/s]
4|4961|Loss: 0.6778329014778137:  77%|███████▋  | 4961/6470 [56:59<15:18,  1.64it/s]
4|4962|Loss: 1.0089282989501953:  77%|███████▋  | 4961/6470 [56:59<15:18,  1.64it/s]
4|4962|Loss: 1.0089282989501953:  77%|███████▋  | 4962/6470 [57:00<15:11,  1.65it/s]
4|4963|Loss: 1.109334945678711:  77%|███████▋  | 4962/6470 [57:00<15:11,  1.65it/s] 
4|4963|Loss: 1.109334945678711:  77%|███████▋  | 4963/6470 [57:00<16:27,  1.53it/s]
4|4964|Loss: 1.0085684061050415:  77%|███████▋  | 4963/6470 [57:01<16:27,  1.53it/s]
4|4964|Loss: 1.0085684061050415:  77%|███████▋  | 4964/6470 [57:01<16:38,  1.51it/s]
4|4965|Loss: 0.9056776762008667:  77%|███████▋  | 4964/6470 [57:02<16:38,  1.51it/s]
4|4965|Loss: 0.9056776762008667:  77%|███████▋  | 4965/6470 [57:02<16:59,  1.48it/s]
4|4966|Loss: 1.0311776399612427:  77%|███████▋  | 4965/6470 [57:02<16:59,  1.48it/s]
4|4966|Loss: 1.0311776399612427:  77%|███████▋  | 4966/6470 [57:03<17:59,  1.39it/s]
4|4967|Loss: 0.9698285460472107:  77%|███████▋  | 4966/6470 [57:03<17:59,  1.39it/s]
4|4967|Loss: 0.9698285460472107:  77%|███████▋  | 4967/6470 [57:03<17:36,  1.42it/s]
4|4968|Loss: 0.9711194038391113:  77%|███████▋  | 4967/6470 [57:04<17:36,  1.42it/s]
4|4968|Loss: 0.9711194038391113:  77%|███████▋  | 4968/6470 [57:04<17:35,  1.42it/s]
4|4969|Loss: 0.8933271169662476:  77%|███████▋  | 4968/6470 [57:04<17:35,  1.42it/s]
4|4969|Loss: 0.8933271169662476:  77%|███████▋  | 4969/6470 [57:05<16:58,  1.47it/s]
4|4970|Loss: 0.7778911590576172:  77%|███████▋  | 4969/6470 [57:05<16:58,  1.47it/s]
4|4970|Loss: 0.7778911590576172:  77%|███████▋  | 4970/6470 [57:05<16:25,  1.52it/s]
4|4971|Loss: 0.9299737811088562:  77%|███████▋  | 4970/6470 [57:06<16:25,  1.52it/s]
4|4971|Loss: 0.9299737811088562:  77%|███████▋  | 4971/6470 [57:06<15:52,  1.57it/s]
4|4972|Loss: 0.832990288734436:  77%|███████▋  | 4971/6470 [57:06<15:52,  1.57it/s] 
4|4972|Loss: 0.832990288734436:  77%|███████▋  | 4972/6470 [57:07<15:49,  1.58it/s]
4|4973|Loss: 0.9030870795249939:  77%|███████▋  | 4972/6470 [57:07<15:49,  1.58it/s]
4|4973|Loss: 0.9030870795249939:  77%|███████▋  | 4973/6470 [57:07<16:30,  1.51it/s]
4|4974|Loss: 0.7371790409088135:  77%|███████▋  | 4973/6470 [57:08<16:30,  1.51it/s]
4|4974|Loss: 0.7371790409088135:  77%|███████▋  | 4974/6470 [57:08<15:22,  1.62it/s]
4|4975|Loss: 0.9626116156578064:  77%|███████▋  | 4974/6470 [57:08<15:22,  1.62it/s]
4|4975|Loss: 0.9626116156578064:  77%|███████▋  | 4975/6470 [57:08<16:08,  1.54it/s]
4|4976|Loss: 0.9021382927894592:  77%|███████▋  | 4975/6470 [57:09<16:08,  1.54it/s]
4|4976|Loss: 0.9021382927894592:  77%|███████▋  | 4976/6470 [57:09<16:46,  1.48it/s]
4|4977|Loss: 0.7558035254478455:  77%|███████▋  | 4976/6470 [57:10<16:46,  1.48it/s]
4|4977|Loss: 0.7558035254478455:  77%|███████▋  | 4977/6470 [57:10<15:40,  1.59it/s]
4|4978|Loss: 0.7370538115501404:  77%|███████▋  | 4977/6470 [57:10<15:40,  1.59it/s]
4|4978|Loss: 0.7370538115501404:  77%|███████▋  | 4978/6470 [57:10<14:23,  1.73it/s]
4|4979|Loss: 0.8879435658454895:  77%|███████▋  | 4978/6470 [57:11<14:23,  1.73it/s]
4|4979|Loss: 0.8879435658454895:  77%|███████▋  | 4979/6470 [57:11<15:37,  1.59it/s]
4|4980|Loss: 1.0746556520462036:  77%|███████▋  | 4979/6470 [57:11<15:37,  1.59it/s]
4|4980|Loss: 1.0746556520462036:  77%|███████▋  | 4980/6470 [57:12<17:06,  1.45it/s]
4|4981|Loss: 0.9561203718185425:  77%|███████▋  | 4980/6470 [57:12<17:06,  1.45it/s]
4|4981|Loss: 0.9561203718185425:  77%|███████▋  | 4981/6470 [57:12<17:23,  1.43it/s]
4|4982|Loss: 0.9026101231575012:  77%|███████▋  | 4981/6470 [57:13<17:23,  1.43it/s]
4|4982|Loss: 0.9026101231575012:  77%|███████▋  | 4982/6470 [57:13<16:50,  1.47it/s]
4|4983|Loss: 0.7957987189292908:  77%|███████▋  | 4982/6470 [57:14<16:50,  1.47it/s]
4|4983|Loss: 0.7957987189292908:  77%|███████▋  | 4983/6470 [57:14<17:02,  1.45it/s]
4|4984|Loss: 0.8659811615943909:  77%|███████▋  | 4983/6470 [57:14<17:02,  1.45it/s]
4|4984|Loss: 0.8659811615943909:  77%|███████▋  | 4984/6470 [57:15<17:57,  1.38it/s]
4|4985|Loss: 0.7984158396720886:  77%|███████▋  | 4984/6470 [57:15<17:57,  1.38it/s]
4|4985|Loss: 0.7984158396720886:  77%|███████▋  | 4985/6470 [57:15<18:34,  1.33it/s]
4|4986|Loss: 0.7280601263046265:  77%|███████▋  | 4985/6470 [57:16<18:34,  1.33it/s]
4|4986|Loss: 0.7280601263046265:  77%|███████▋  | 4986/6470 [57:16<17:12,  1.44it/s]
4|4987|Loss: 0.9033480882644653:  77%|███████▋  | 4986/6470 [57:16<17:12,  1.44it/s]
4|4987|Loss: 0.9033480882644653:  77%|███████▋  | 4987/6470 [57:17<17:49,  1.39it/s]
4|4988|Loss: 0.9700132608413696:  77%|███████▋  | 4987/6470 [57:17<17:49,  1.39it/s]
4|4988|Loss: 0.9700132608413696:  77%|███████▋  | 4988/6470 [57:18<18:12,  1.36it/s]
4|4989|Loss: 0.8080180883407593:  77%|███████▋  | 4988/6470 [57:18<18:12,  1.36it/s]
4|4989|Loss: 0.8080180883407593:  77%|███████▋  | 4989/6470 [57:18<16:52,  1.46it/s]
4|4990|Loss: 0.8082366585731506:  77%|███████▋  | 4989/6470 [57:18<16:52,  1.46it/s]
4|4990|Loss: 0.8082366585731506:  77%|███████▋  | 4990/6470 [57:19<16:40,  1.48it/s]
4|4991|Loss: 1.0013912916183472:  77%|███████▋  | 4990/6470 [57:19<16:40,  1.48it/s]
4|4991|Loss: 1.0013912916183472:  77%|███████▋  | 4991/6470 [57:19<16:34,  1.49it/s]
4|4992|Loss: 0.8727368712425232:  77%|███████▋  | 4991/6470 [57:20<16:34,  1.49it/s]
4|4992|Loss: 0.8727368712425232:  77%|███████▋  | 4992/6470 [57:20<16:40,  1.48it/s]
4|4993|Loss: 0.9825754761695862:  77%|███████▋  | 4992/6470 [57:21<16:40,  1.48it/s]
4|4993|Loss: 0.9825754761695862:  77%|███████▋  | 4993/6470 [57:21<17:42,  1.39it/s]
4|4994|Loss: 0.9731944799423218:  77%|███████▋  | 4993/6470 [57:21<17:42,  1.39it/s]
4|4994|Loss: 0.9731944799423218:  77%|███████▋  | 4994/6470 [57:22<18:00,  1.37it/s]
4|4995|Loss: 0.8238959312438965:  77%|███████▋  | 4994/6470 [57:22<18:00,  1.37it/s]
4|4995|Loss: 0.8238959312438965:  77%|███████▋  | 4995/6470 [57:22<18:00,  1.37it/s]
4|4996|Loss: 0.9032740592956543:  77%|███████▋  | 4995/6470 [57:23<18:00,  1.37it/s]
4|4996|Loss: 0.9032740592956543:  77%|███████▋  | 4996/6470 [57:23<18:42,  1.31it/s]
4|4997|Loss: 0.8990569710731506:  77%|███████▋  | 4996/6470 [57:24<18:42,  1.31it/s]
4|4997|Loss: 0.8990569710731506:  77%|███████▋  | 4997/6470 [57:24<19:32,  1.26it/s]
4|4998|Loss: 0.9673358798027039:  77%|███████▋  | 4997/6470 [57:25<19:32,  1.26it/s]
4|4998|Loss: 0.9673358798027039:  77%|███████▋  | 4998/6470 [57:25<20:25,  1.20it/s]
4|4999|Loss: 0.8532389402389526:  77%|███████▋  | 4998/6470 [57:25<20:25,  1.20it/s]
4|4999|Loss: 0.8532389402389526:  77%|███████▋  | 4999/6470 [57:26<19:16,  1.27it/s]
4|5000|Loss: 0.8266541957855225:  77%|███████▋  | 4999/6470 [57:26<19:16,  1.27it/s]
4|5000|Loss: 0.8266541957855225:  77%|███████▋  | 5000/6470 [57:27<19:27,  1.26it/s]
4|5001|Loss: 0.9168586134910583:  77%|███████▋  | 5000/6470 [57:27<19:27,  1.26it/s]
4|5001|Loss: 0.9168586134910583:  77%|███████▋  | 5001/6470 [57:27<19:03,  1.28it/s]
4|5002|Loss: 0.6946068406105042:  77%|███████▋  | 5001/6470 [57:28<19:03,  1.28it/s]
4|5002|Loss: 0.6946068406105042:  77%|███████▋  | 5002/6470 [57:28<20:01,  1.22it/s]
4|5003|Loss: 0.8314956426620483:  77%|███████▋  | 5002/6470 [57:29<20:01,  1.22it/s]
4|5003|Loss: 0.8314956426620483:  77%|███████▋  | 5003/6470 [57:29<20:09,  1.21it/s]
4|5004|Loss: 0.731572687625885:  77%|███████▋  | 5003/6470 [57:29<20:09,  1.21it/s] 
4|5004|Loss: 0.731572687625885:  77%|███████▋  | 5004/6470 [57:30<19:24,  1.26it/s]
4|5005|Loss: 1.0302165746688843:  77%|███████▋  | 5004/6470 [57:30<19:24,  1.26it/s]
4|5005|Loss: 1.0302165746688843:  77%|███████▋  | 5005/6470 [57:30<18:01,  1.35it/s]
4|5006|Loss: 0.7859295606613159:  77%|███████▋  | 5005/6470 [57:31<18:01,  1.35it/s]
4|5006|Loss: 0.7859295606613159:  77%|███████▋  | 5006/6470 [57:31<18:20,  1.33it/s]
4|5007|Loss: 0.6426793336868286:  77%|███████▋  | 5006/6470 [57:32<18:20,  1.33it/s]
4|5007|Loss: 0.6426793336868286:  77%|███████▋  | 5007/6470 [57:32<18:09,  1.34it/s]
4|5008|Loss: 0.8901721835136414:  77%|███████▋  | 5007/6470 [57:32<18:09,  1.34it/s]
4|5008|Loss: 0.8901721835136414:  77%|███████▋  | 5008/6470 [57:32<16:43,  1.46it/s]
4|5009|Loss: 0.9679474830627441:  77%|███████▋  | 5008/6470 [57:33<16:43,  1.46it/s]
4|5009|Loss: 0.9679474830627441:  77%|███████▋  | 5009/6470 [57:33<17:11,  1.42it/s]
4|5010|Loss: 0.7888182997703552:  77%|███████▋  | 5009/6470 [57:34<17:11,  1.42it/s]
4|5010|Loss: 0.7888182997703552:  77%|███████▋  | 5010/6470 [57:34<17:26,  1.39it/s]
4|5011|Loss: 0.8808599710464478:  77%|███████▋  | 5010/6470 [57:34<17:26,  1.39it/s]
4|5011|Loss: 0.8808599710464478:  77%|███████▋  | 5011/6470 [57:35<18:15,  1.33it/s]
4|5012|Loss: 0.7247717380523682:  77%|███████▋  | 5011/6470 [57:35<18:15,  1.33it/s]
4|5012|Loss: 0.7247717380523682:  77%|███████▋  | 5012/6470 [57:35<16:10,  1.50it/s]
4|5013|Loss: 0.8200996518135071:  77%|███████▋  | 5012/6470 [57:36<16:10,  1.50it/s]
4|5013|Loss: 0.8200996518135071:  77%|███████▋  | 5013/6470 [57:36<16:01,  1.51it/s]
4|5014|Loss: 0.9666403532028198:  77%|███████▋  | 5013/6470 [57:36<16:01,  1.51it/s]
4|5014|Loss: 0.9666403532028198:  77%|███████▋  | 5014/6470 [57:36<15:35,  1.56it/s]
4|5015|Loss: 0.8680049180984497:  77%|███████▋  | 5014/6470 [57:37<15:35,  1.56it/s]
4|5015|Loss: 0.8680049180984497:  78%|███████▊  | 5015/6470 [57:37<15:21,  1.58it/s]
4|5016|Loss: 0.8262397050857544:  78%|███████▊  | 5015/6470 [57:37<15:21,  1.58it/s]
4|5016|Loss: 0.8262397050857544:  78%|███████▊  | 5016/6470 [57:38<15:15,  1.59it/s]
4|5017|Loss: 0.8681541681289673:  78%|███████▊  | 5016/6470 [57:38<15:15,  1.59it/s]
4|5017|Loss: 0.8681541681289673:  78%|███████▊  | 5017/6470 [57:39<16:23,  1.48it/s]
4|5018|Loss: 0.6842133402824402:  78%|███████▊  | 5017/6470 [57:39<16:23,  1.48it/s]
4|5018|Loss: 0.6842133402824402:  78%|███████▊  | 5018/6470 [57:39<17:00,  1.42it/s]
4|5019|Loss: 0.8295401334762573:  78%|███████▊  | 5018/6470 [57:40<17:00,  1.42it/s]
4|5019|Loss: 0.8295401334762573:  78%|███████▊  | 5019/6470 [57:40<17:57,  1.35it/s]
4|5020|Loss: 0.8628738522529602:  78%|███████▊  | 5019/6470 [57:41<17:57,  1.35it/s]
4|5020|Loss: 0.8628738522529602:  78%|███████▊  | 5020/6470 [57:41<17:25,  1.39it/s]
4|5021|Loss: 0.7835136651992798:  78%|███████▊  | 5020/6470 [57:41<17:25,  1.39it/s]
4|5021|Loss: 0.7835136651992798:  78%|███████▊  | 5021/6470 [57:42<18:23,  1.31it/s]
4|5022|Loss: 0.7696712017059326:  78%|███████▊  | 5021/6470 [57:42<18:23,  1.31it/s]
4|5022|Loss: 0.7696712017059326:  78%|███████▊  | 5022/6470 [57:43<19:31,  1.24it/s]
4|5023|Loss: 1.07378089427948:  78%|███████▊  | 5022/6470 [57:43<19:31,  1.24it/s]  
4|5023|Loss: 1.07378089427948:  78%|███████▊  | 5023/6470 [57:43<18:37,  1.30it/s]
4|5024|Loss: 0.6939328908920288:  78%|███████▊  | 5023/6470 [57:44<18:37,  1.30it/s]
4|5024|Loss: 0.6939328908920288:  78%|███████▊  | 5024/6470 [57:44<16:05,  1.50it/s]
4|5025|Loss: 0.9276309609413147:  78%|███████▊  | 5024/6470 [57:44<16:05,  1.50it/s]
4|5025|Loss: 0.9276309609413147:  78%|███████▊  | 5025/6470 [57:44<15:30,  1.55it/s]
4|5026|Loss: 0.7567861080169678:  78%|███████▊  | 5025/6470 [57:45<15:30,  1.55it/s]
4|5026|Loss: 0.7567861080169678:  78%|███████▊  | 5026/6470 [57:45<16:00,  1.50it/s]
4|5027|Loss: 1.0152828693389893:  78%|███████▊  | 5026/6470 [57:45<16:00,  1.50it/s]
4|5027|Loss: 1.0152828693389893:  78%|███████▊  | 5027/6470 [57:46<16:30,  1.46it/s]
4|5028|Loss: 0.9561901092529297:  78%|███████▊  | 5027/6470 [57:46<16:30,  1.46it/s]
4|5028|Loss: 0.9561901092529297:  78%|███████▊  | 5028/6470 [57:46<16:13,  1.48it/s]
4|5029|Loss: 0.8131071925163269:  78%|███████▊  | 5028/6470 [57:47<16:13,  1.48it/s]
4|5029|Loss: 0.8131071925163269:  78%|███████▊  | 5029/6470 [57:47<15:48,  1.52it/s]
4|5030|Loss: 0.7681052088737488:  78%|███████▊  | 5029/6470 [57:47<15:48,  1.52it/s]
4|5030|Loss: 0.7681052088737488:  78%|███████▊  | 5030/6470 [57:48<15:35,  1.54it/s]
4|5031|Loss: 0.8241778612136841:  78%|███████▊  | 5030/6470 [57:48<15:35,  1.54it/s]
4|5031|Loss: 0.8241778612136841:  78%|███████▊  | 5031/6470 [57:48<15:52,  1.51it/s]
4|5032|Loss: 0.8652206659317017:  78%|███████▊  | 5031/6470 [57:49<15:52,  1.51it/s]
4|5032|Loss: 0.8652206659317017:  78%|███████▊  | 5032/6470 [57:49<15:49,  1.51it/s]
4|5033|Loss: 0.8429542183876038:  78%|███████▊  | 5032/6470 [57:49<15:49,  1.51it/s]
4|5033|Loss: 0.8429542183876038:  78%|███████▊  | 5033/6470 [57:50<15:40,  1.53it/s]
4|5034|Loss: 0.8720356822013855:  78%|███████▊  | 5033/6470 [57:50<15:40,  1.53it/s]
4|5034|Loss: 0.8720356822013855:  78%|███████▊  | 5034/6470 [57:50<15:35,  1.53it/s]
4|5035|Loss: 0.7806670665740967:  78%|███████▊  | 5034/6470 [57:51<15:35,  1.53it/s]
4|5035|Loss: 0.7806670665740967:  78%|███████▊  | 5035/6470 [57:51<17:21,  1.38it/s]
4|5036|Loss: 0.8152865767478943:  78%|███████▊  | 5035/6470 [57:52<17:21,  1.38it/s]
4|5036|Loss: 0.8152865767478943:  78%|███████▊  | 5036/6470 [57:52<17:45,  1.35it/s]
4|5037|Loss: 0.8046586513519287:  78%|███████▊  | 5036/6470 [57:52<17:45,  1.35it/s]
4|5037|Loss: 0.8046586513519287:  78%|███████▊  | 5037/6470 [57:53<17:30,  1.36it/s]
4|5038|Loss: 0.7812503576278687:  78%|███████▊  | 5037/6470 [57:53<17:30,  1.36it/s]
4|5038|Loss: 0.7812503576278687:  78%|███████▊  | 5038/6470 [57:53<16:24,  1.45it/s]
4|5039|Loss: 0.9091424345970154:  78%|███████▊  | 5038/6470 [57:54<16:24,  1.45it/s]
4|5039|Loss: 0.9091424345970154:  78%|███████▊  | 5039/6470 [57:54<16:36,  1.44it/s]
4|5040|Loss: 0.8313340544700623:  78%|███████▊  | 5039/6470 [57:54<16:36,  1.44it/s]
4|5040|Loss: 0.8313340544700623:  78%|███████▊  | 5040/6470 [57:55<16:14,  1.47it/s]
4|5041|Loss: 0.8858184814453125:  78%|███████▊  | 5040/6470 [57:55<16:14,  1.47it/s]
4|5041|Loss: 0.8858184814453125:  78%|███████▊  | 5041/6470 [57:55<16:26,  1.45it/s]
4|5042|Loss: 0.8576423525810242:  78%|███████▊  | 5041/6470 [57:56<16:26,  1.45it/s]
4|5042|Loss: 0.8576423525810242:  78%|███████▊  | 5042/6470 [57:56<15:27,  1.54it/s]
4|5043|Loss: 0.9267550110816956:  78%|███████▊  | 5042/6470 [57:56<15:27,  1.54it/s]
4|5043|Loss: 0.9267550110816956:  78%|███████▊  | 5043/6470 [57:57<16:22,  1.45it/s]
4|5044|Loss: 0.6618790626525879:  78%|███████▊  | 5043/6470 [57:57<16:22,  1.45it/s]
4|5044|Loss: 0.6618790626525879:  78%|███████▊  | 5044/6470 [57:57<16:45,  1.42it/s]
4|5045|Loss: 0.7953003644943237:  78%|███████▊  | 5044/6470 [57:58<16:45,  1.42it/s]
4|5045|Loss: 0.7953003644943237:  78%|███████▊  | 5045/6470 [57:58<16:46,  1.42it/s]
4|5046|Loss: 0.8548218011856079:  78%|███████▊  | 5045/6470 [57:58<16:46,  1.42it/s]
4|5046|Loss: 0.8548218011856079:  78%|███████▊  | 5046/6470 [57:59<17:13,  1.38it/s]
4|5047|Loss: 0.9121347069740295:  78%|███████▊  | 5046/6470 [57:59<17:13,  1.38it/s]
4|5047|Loss: 0.9121347069740295:  78%|███████▊  | 5047/6470 [58:00<17:41,  1.34it/s]
4|5048|Loss: 0.8260786533355713:  78%|███████▊  | 5047/6470 [58:00<17:41,  1.34it/s]
4|5048|Loss: 0.8260786533355713:  78%|███████▊  | 5048/6470 [58:00<17:40,  1.34it/s]
4|5049|Loss: 0.8678615093231201:  78%|███████▊  | 5048/6470 [58:01<17:40,  1.34it/s]
4|5049|Loss: 0.8678615093231201:  78%|███████▊  | 5049/6470 [58:01<17:43,  1.34it/s]
4|5050|Loss: 0.7927007079124451:  78%|███████▊  | 5049/6470 [58:02<17:43,  1.34it/s]
4|5050|Loss: 0.7927007079124451:  78%|███████▊  | 5050/6470 [58:02<17:31,  1.35it/s]
4|5051|Loss: 0.7795061469078064:  78%|███████▊  | 5050/6470 [58:02<17:31,  1.35it/s]
4|5051|Loss: 0.7795061469078064:  78%|███████▊  | 5051/6470 [58:03<17:57,  1.32it/s]
4|5052|Loss: 0.7897335886955261:  78%|███████▊  | 5051/6470 [58:03<17:57,  1.32it/s]
4|5052|Loss: 0.7897335886955261:  78%|███████▊  | 5052/6470 [58:03<18:26,  1.28it/s]
4|5053|Loss: 0.6312689185142517:  78%|███████▊  | 5052/6470 [58:04<18:26,  1.28it/s]
4|5053|Loss: 0.6312689185142517:  78%|███████▊  | 5053/6470 [58:04<17:41,  1.33it/s]
4|5054|Loss: 0.8464726209640503:  78%|███████▊  | 5053/6470 [58:05<17:41,  1.33it/s]
4|5054|Loss: 0.8464726209640503:  78%|███████▊  | 5054/6470 [58:05<16:50,  1.40it/s]
4|5055|Loss: 0.8447027206420898:  78%|███████▊  | 5054/6470 [58:05<16:50,  1.40it/s]
4|5055|Loss: 0.8447027206420898:  78%|███████▊  | 5055/6470 [58:06<17:13,  1.37it/s]
4|5056|Loss: 0.8664023280143738:  78%|███████▊  | 5055/6470 [58:06<17:13,  1.37it/s]
4|5056|Loss: 0.8664023280143738:  78%|███████▊  | 5056/6470 [58:06<17:12,  1.37it/s]
4|5057|Loss: 0.6891796588897705:  78%|███████▊  | 5056/6470 [58:07<17:12,  1.37it/s]
4|5057|Loss: 0.6891796588897705:  78%|███████▊  | 5057/6470 [58:07<18:01,  1.31it/s]
4|5058|Loss: 0.9386840462684631:  78%|███████▊  | 5057/6470 [58:08<18:01,  1.31it/s]
4|5058|Loss: 0.9386840462684631:  78%|███████▊  | 5058/6470 [58:08<16:52,  1.40it/s]
4|5059|Loss: 0.8683137893676758:  78%|███████▊  | 5058/6470 [58:08<16:52,  1.40it/s]
4|5059|Loss: 0.8683137893676758:  78%|███████▊  | 5059/6470 [58:09<17:26,  1.35it/s]
4|5060|Loss: 0.990283191204071:  78%|███████▊  | 5059/6470 [58:09<17:26,  1.35it/s] 
4|5060|Loss: 0.990283191204071:  78%|███████▊  | 5060/6470 [58:09<17:08,  1.37it/s]
4|5061|Loss: 0.937000572681427:  78%|███████▊  | 5060/6470 [58:10<17:08,  1.37it/s]
4|5061|Loss: 0.937000572681427:  78%|███████▊  | 5061/6470 [58:10<17:01,  1.38it/s]
4|5062|Loss: 0.8119738101959229:  78%|███████▊  | 5061/6470 [58:10<17:01,  1.38it/s]
4|5062|Loss: 0.8119738101959229:  78%|███████▊  | 5062/6470 [58:11<16:02,  1.46it/s]
4|5063|Loss: 0.756872296333313:  78%|███████▊  | 5062/6470 [58:11<16:02,  1.46it/s] 
4|5063|Loss: 0.756872296333313:  78%|███████▊  | 5063/6470 [58:11<15:27,  1.52it/s]
4|5064|Loss: 0.8631477952003479:  78%|███████▊  | 5063/6470 [58:12<15:27,  1.52it/s]
4|5064|Loss: 0.8631477952003479:  78%|███████▊  | 5064/6470 [58:12<15:07,  1.55it/s]
4|5065|Loss: 0.8196681141853333:  78%|███████▊  | 5064/6470 [58:12<15:07,  1.55it/s]
4|5065|Loss: 0.8196681141853333:  78%|███████▊  | 5065/6470 [58:12<14:45,  1.59it/s]
4|5066|Loss: 0.7770520448684692:  78%|███████▊  | 5065/6470 [58:13<14:45,  1.59it/s]
4|5066|Loss: 0.7770520448684692:  78%|███████▊  | 5066/6470 [58:13<15:40,  1.49it/s]
4|5067|Loss: 0.9726824760437012:  78%|███████▊  | 5066/6470 [58:14<15:40,  1.49it/s]
4|5067|Loss: 0.9726824760437012:  78%|███████▊  | 5067/6470 [58:14<16:02,  1.46it/s]
4|5068|Loss: 1.0320152044296265:  78%|███████▊  | 5067/6470 [58:14<16:02,  1.46it/s]
4|5068|Loss: 1.0320152044296265:  78%|███████▊  | 5068/6470 [58:14<15:42,  1.49it/s]
4|5069|Loss: 0.84360271692276:  78%|███████▊  | 5068/6470 [58:15<15:42,  1.49it/s]  
4|5069|Loss: 0.84360271692276:  78%|███████▊  | 5069/6470 [58:15<15:55,  1.47it/s]
4|5070|Loss: 0.7404342889785767:  78%|███████▊  | 5069/6470 [58:16<15:55,  1.47it/s]
4|5070|Loss: 0.7404342889785767:  78%|███████▊  | 5070/6470 [58:16<16:14,  1.44it/s]
4|5071|Loss: 0.9173383116722107:  78%|███████▊  | 5070/6470 [58:16<16:14,  1.44it/s]
4|5071|Loss: 0.9173383116722107:  78%|███████▊  | 5071/6470 [58:17<15:36,  1.49it/s]
4|5072|Loss: 0.8124404549598694:  78%|███████▊  | 5071/6470 [58:17<15:36,  1.49it/s]
4|5072|Loss: 0.8124404549598694:  78%|███████▊  | 5072/6470 [58:17<16:53,  1.38it/s]
4|5073|Loss: 0.8357177972793579:  78%|███████▊  | 5072/6470 [58:18<16:53,  1.38it/s]
4|5073|Loss: 0.8357177972793579:  78%|███████▊  | 5073/6470 [58:18<18:00,  1.29it/s]
4|5074|Loss: 0.9192845821380615:  78%|███████▊  | 5073/6470 [58:19<18:00,  1.29it/s]
4|5074|Loss: 0.9192845821380615:  78%|███████▊  | 5074/6470 [58:19<18:03,  1.29it/s]
4|5075|Loss: 0.8810371160507202:  78%|███████▊  | 5074/6470 [58:19<18:03,  1.29it/s]
4|5075|Loss: 0.8810371160507202:  78%|███████▊  | 5075/6470 [58:20<17:19,  1.34it/s]
4|5076|Loss: 0.5069101452827454:  78%|███████▊  | 5075/6470 [58:20<17:19,  1.34it/s]
4|5076|Loss: 0.5069101452827454:  78%|███████▊  | 5076/6470 [58:21<18:19,  1.27it/s]
4|5077|Loss: 0.896409273147583:  78%|███████▊  | 5076/6470 [58:21<18:19,  1.27it/s] 
4|5077|Loss: 0.896409273147583:  78%|███████▊  | 5077/6470 [58:21<17:44,  1.31it/s]
4|5078|Loss: 0.8161977529525757:  78%|███████▊  | 5077/6470 [58:22<17:44,  1.31it/s]
4|5078|Loss: 0.8161977529525757:  78%|███████▊  | 5078/6470 [58:22<17:09,  1.35it/s]
4|5079|Loss: 0.9008742570877075:  78%|███████▊  | 5078/6470 [58:22<17:09,  1.35it/s]
4|5079|Loss: 0.9008742570877075:  79%|███████▊  | 5079/6470 [58:23<16:08,  1.44it/s]
4|5080|Loss: 0.84093177318573:  79%|███████▊  | 5079/6470 [58:23<16:08,  1.44it/s]  
4|5080|Loss: 0.84093177318573:  79%|███████▊  | 5080/6470 [58:23<14:37,  1.58it/s]
4|5081|Loss: 0.7264305949211121:  79%|███████▊  | 5080/6470 [58:23<14:37,  1.58it/s]
4|5081|Loss: 0.7264305949211121:  79%|███████▊  | 5081/6470 [58:24<15:16,  1.52it/s]
4|5082|Loss: 0.7825376391410828:  79%|███████▊  | 5081/6470 [58:24<15:16,  1.52it/s]
4|5082|Loss: 0.7825376391410828:  79%|███████▊  | 5082/6470 [58:25<15:45,  1.47it/s]
4|5083|Loss: 0.8417724967002869:  79%|███████▊  | 5082/6470 [58:25<15:45,  1.47it/s]
4|5083|Loss: 0.8417724967002869:  79%|███████▊  | 5083/6470 [58:25<15:15,  1.51it/s]
4|5084|Loss: 0.8349036574363708:  79%|███████▊  | 5083/6470 [58:26<15:15,  1.51it/s]
4|5084|Loss: 0.8349036574363708:  79%|███████▊  | 5084/6470 [58:26<15:34,  1.48it/s]
4|5085|Loss: 0.7948243618011475:  79%|███████▊  | 5084/6470 [58:26<15:34,  1.48it/s]
4|5085|Loss: 0.7948243618011475:  79%|███████▊  | 5085/6470 [58:26<14:46,  1.56it/s]
4|5086|Loss: 0.8838038444519043:  79%|███████▊  | 5085/6470 [58:27<14:46,  1.56it/s]
4|5086|Loss: 0.8838038444519043:  79%|███████▊  | 5086/6470 [58:27<15:39,  1.47it/s]
4|5087|Loss: 0.9357263445854187:  79%|███████▊  | 5086/6470 [58:28<15:39,  1.47it/s]
4|5087|Loss: 0.9357263445854187:  79%|███████▊  | 5087/6470 [58:28<16:40,  1.38it/s]
4|5088|Loss: 0.9355886578559875:  79%|███████▊  | 5087/6470 [58:28<16:40,  1.38it/s]
4|5088|Loss: 0.9355886578559875:  79%|███████▊  | 5088/6470 [58:29<15:40,  1.47it/s]
4|5089|Loss: 0.9278197884559631:  79%|███████▊  | 5088/6470 [58:29<15:40,  1.47it/s]
4|5089|Loss: 0.9278197884559631:  79%|███████▊  | 5089/6470 [58:29<15:11,  1.52it/s]
4|5090|Loss: 0.8570868372917175:  79%|███████▊  | 5089/6470 [58:30<15:11,  1.52it/s]
4|5090|Loss: 0.8570868372917175:  79%|███████▊  | 5090/6470 [58:30<15:23,  1.49it/s]
4|5091|Loss: 0.985718846321106:  79%|███████▊  | 5090/6470 [58:30<15:23,  1.49it/s] 
4|5091|Loss: 0.985718846321106:  79%|███████▊  | 5091/6470 [58:31<14:56,  1.54it/s]
4|5092|Loss: 0.8738917112350464:  79%|███████▊  | 5091/6470 [58:31<14:56,  1.54it/s]
4|5092|Loss: 0.8738917112350464:  79%|███████▊  | 5092/6470 [58:31<15:06,  1.52it/s]
4|5093|Loss: 1.1364301443099976:  79%|███████▊  | 5092/6470 [58:32<15:06,  1.52it/s]
4|5093|Loss: 1.1364301443099976:  79%|███████▊  | 5093/6470 [58:32<16:15,  1.41it/s]
4|5094|Loss: 0.8633962273597717:  79%|███████▊  | 5093/6470 [58:32<16:15,  1.41it/s]
4|5094|Loss: 0.8633962273597717:  79%|███████▊  | 5094/6470 [58:33<17:10,  1.34it/s]
4|5095|Loss: 0.9241289496421814:  79%|███████▊  | 5094/6470 [58:33<17:10,  1.34it/s]
4|5095|Loss: 0.9241289496421814:  79%|███████▊  | 5095/6470 [58:33<16:04,  1.43it/s]
4|5096|Loss: 0.8054289221763611:  79%|███████▊  | 5095/6470 [58:34<16:04,  1.43it/s]
4|5096|Loss: 0.8054289221763611:  79%|███████▉  | 5096/6470 [58:34<15:27,  1.48it/s]
4|5097|Loss: 0.7949948310852051:  79%|███████▉  | 5096/6470 [58:34<15:27,  1.48it/s]
4|5097|Loss: 0.7949948310852051:  79%|███████▉  | 5097/6470 [58:35<16:22,  1.40it/s]
4|5098|Loss: 0.9359644055366516:  79%|███████▉  | 5097/6470 [58:35<16:22,  1.40it/s]
4|5098|Loss: 0.9359644055366516:  79%|███████▉  | 5098/6470 [58:35<15:34,  1.47it/s]
4|5099|Loss: 0.861738383769989:  79%|███████▉  | 5098/6470 [58:36<15:34,  1.47it/s] 
4|5099|Loss: 0.861738383769989:  79%|███████▉  | 5099/6470 [58:36<14:47,  1.54it/s]
4|5100|Loss: 1.1953922510147095:  79%|███████▉  | 5099/6470 [58:36<14:47,  1.54it/s]
4|5100|Loss: 1.1953922510147095:  79%|███████▉  | 5100/6470 [58:37<13:56,  1.64it/s]
4|5101|Loss: 0.8029088973999023:  79%|███████▉  | 5100/6470 [58:37<13:56,  1.64it/s]
4|5101|Loss: 0.8029088973999023:  79%|███████▉  | 5101/6470 [58:37<13:49,  1.65it/s]
4|5102|Loss: 0.8170313239097595:  79%|███████▉  | 5101/6470 [58:38<13:49,  1.65it/s]
4|5102|Loss: 0.8170313239097595:  79%|███████▉  | 5102/6470 [58:38<14:00,  1.63it/s]
4|5103|Loss: 0.8156003355979919:  79%|███████▉  | 5102/6470 [58:38<14:00,  1.63it/s]
4|5103|Loss: 0.8156003355979919:  79%|███████▉  | 5103/6470 [58:38<14:30,  1.57it/s]
4|5104|Loss: 0.7748910188674927:  79%|███████▉  | 5103/6470 [58:39<14:30,  1.57it/s]
4|5104|Loss: 0.7748910188674927:  79%|███████▉  | 5104/6470 [58:39<14:27,  1.57it/s]
4|5105|Loss: 0.9661323428153992:  79%|███████▉  | 5104/6470 [58:39<14:27,  1.57it/s]
4|5105|Loss: 0.9661323428153992:  79%|███████▉  | 5105/6470 [58:40<15:01,  1.51it/s]
4|5106|Loss: 0.9058589935302734:  79%|███████▉  | 5105/6470 [58:40<15:01,  1.51it/s]
4|5106|Loss: 0.9058589935302734:  79%|███████▉  | 5106/6470 [58:40<14:44,  1.54it/s]
4|5107|Loss: 1.030626893043518:  79%|███████▉  | 5106/6470 [58:41<14:44,  1.54it/s] 
4|5107|Loss: 1.030626893043518:  79%|███████▉  | 5107/6470 [58:41<15:06,  1.50it/s]
4|5108|Loss: 0.8089854717254639:  79%|███████▉  | 5107/6470 [58:42<15:06,  1.50it/s]
4|5108|Loss: 0.8089854717254639:  79%|███████▉  | 5108/6470 [58:42<15:25,  1.47it/s]
4|5109|Loss: 0.9721469879150391:  79%|███████▉  | 5108/6470 [58:42<15:25,  1.47it/s]
4|5109|Loss: 0.9721469879150391:  79%|███████▉  | 5109/6470 [58:42<14:56,  1.52it/s]
4|5110|Loss: 0.9754757881164551:  79%|███████▉  | 5109/6470 [58:43<14:56,  1.52it/s]
4|5110|Loss: 0.9754757881164551:  79%|███████▉  | 5110/6470 [58:43<16:11,  1.40it/s]
4|5111|Loss: 0.8053843975067139:  79%|███████▉  | 5110/6470 [58:44<16:11,  1.40it/s]
4|5111|Loss: 0.8053843975067139:  79%|███████▉  | 5111/6470 [58:44<15:43,  1.44it/s]
4|5112|Loss: 0.8202046155929565:  79%|███████▉  | 5111/6470 [58:44<15:43,  1.44it/s]
4|5112|Loss: 0.8202046155929565:  79%|███████▉  | 5112/6470 [58:45<15:53,  1.42it/s]
4|5113|Loss: 0.9527742862701416:  79%|███████▉  | 5112/6470 [58:45<15:53,  1.42it/s]
4|5113|Loss: 0.9527742862701416:  79%|███████▉  | 5113/6470 [58:46<16:39,  1.36it/s]
4|5114|Loss: 0.9694646000862122:  79%|███████▉  | 5113/6470 [58:46<16:39,  1.36it/s]
4|5114|Loss: 0.9694646000862122:  79%|███████▉  | 5114/6470 [58:46<17:16,  1.31it/s]
4|5115|Loss: 0.8055293560028076:  79%|███████▉  | 5114/6470 [58:47<17:16,  1.31it/s]
4|5115|Loss: 0.8055293560028076:  79%|███████▉  | 5115/6470 [58:47<17:19,  1.30it/s]
4|5116|Loss: 0.9284994006156921:  79%|███████▉  | 5115/6470 [58:47<17:19,  1.30it/s]
4|5116|Loss: 0.9284994006156921:  79%|███████▉  | 5116/6470 [58:48<16:30,  1.37it/s]
4|5117|Loss: 1.0036190748214722:  79%|███████▉  | 5116/6470 [58:48<16:30,  1.37it/s]
4|5117|Loss: 1.0036190748214722:  79%|███████▉  | 5117/6470 [58:48<15:50,  1.42it/s]
4|5118|Loss: 0.8596605658531189:  79%|███████▉  | 5117/6470 [58:49<15:50,  1.42it/s]
4|5118|Loss: 0.8596605658531189:  79%|███████▉  | 5118/6470 [58:49<16:39,  1.35it/s]
4|5119|Loss: 0.7829815149307251:  79%|███████▉  | 5118/6470 [58:50<16:39,  1.35it/s]
4|5119|Loss: 0.7829815149307251:  79%|███████▉  | 5119/6470 [58:50<16:12,  1.39it/s]
4|5120|Loss: 0.8807271122932434:  79%|███████▉  | 5119/6470 [58:50<16:12,  1.39it/s]
4|5120|Loss: 0.8807271122932434:  79%|███████▉  | 5120/6470 [58:51<15:31,  1.45it/s]
4|5121|Loss: 0.7910178303718567:  79%|███████▉  | 5120/6470 [58:51<15:31,  1.45it/s]
4|5121|Loss: 0.7910178303718567:  79%|███████▉  | 5121/6470 [58:51<14:09,  1.59it/s]
4|5122|Loss: 0.7623552680015564:  79%|███████▉  | 5121/6470 [58:51<14:09,  1.59it/s]
4|5122|Loss: 0.7623552680015564:  79%|███████▉  | 5122/6470 [58:52<13:50,  1.62it/s]
4|5123|Loss: 0.7713221907615662:  79%|███████▉  | 5122/6470 [58:52<13:50,  1.62it/s]
4|5123|Loss: 0.7713221907615662:  79%|███████▉  | 5123/6470 [58:52<14:17,  1.57it/s]
4|5124|Loss: 1.0681830644607544:  79%|███████▉  | 5123/6470 [58:53<14:17,  1.57it/s]
4|5124|Loss: 1.0681830644607544:  79%|███████▉  | 5124/6470 [58:53<14:01,  1.60it/s]
4|5125|Loss: 0.8908470273017883:  79%|███████▉  | 5124/6470 [58:53<14:01,  1.60it/s]
4|5125|Loss: 0.8908470273017883:  79%|███████▉  | 5125/6470 [58:53<14:03,  1.59it/s]
4|5126|Loss: 0.6584687829017639:  79%|███████▉  | 5125/6470 [58:54<14:03,  1.59it/s]
4|5126|Loss: 0.6584687829017639:  79%|███████▉  | 5126/6470 [58:54<14:02,  1.59it/s]
4|5127|Loss: 0.9006479382514954:  79%|███████▉  | 5126/6470 [58:54<14:02,  1.59it/s]
4|5127|Loss: 0.9006479382514954:  79%|███████▉  | 5127/6470 [58:55<13:49,  1.62it/s]
4|5128|Loss: 0.8024110794067383:  79%|███████▉  | 5127/6470 [58:55<13:49,  1.62it/s]
4|5128|Loss: 0.8024110794067383:  79%|███████▉  | 5128/6470 [58:55<13:24,  1.67it/s]
4|5129|Loss: 0.803051233291626:  79%|███████▉  | 5128/6470 [58:56<13:24,  1.67it/s] 
4|5129|Loss: 0.803051233291626:  79%|███████▉  | 5129/6470 [58:56<12:57,  1.73it/s]
4|5130|Loss: 0.9594665765762329:  79%|███████▉  | 5129/6470 [58:56<12:57,  1.73it/s]
4|5130|Loss: 0.9594665765762329:  79%|███████▉  | 5130/6470 [58:56<12:50,  1.74it/s]
4|5131|Loss: 0.9605333805084229:  79%|███████▉  | 5130/6470 [58:57<12:50,  1.74it/s]
4|5131|Loss: 0.9605333805084229:  79%|███████▉  | 5131/6470 [58:57<14:05,  1.58it/s]
4|5132|Loss: 0.878328800201416:  79%|███████▉  | 5131/6470 [58:58<14:05,  1.58it/s] 
4|5132|Loss: 0.878328800201416:  79%|███████▉  | 5132/6470 [58:58<14:22,  1.55it/s]
4|5133|Loss: 0.7039601802825928:  79%|███████▉  | 5132/6470 [58:58<14:22,  1.55it/s]
4|5133|Loss: 0.7039601802825928:  79%|███████▉  | 5133/6470 [58:58<14:04,  1.58it/s]
4|5134|Loss: 0.9669439196586609:  79%|███████▉  | 5133/6470 [58:59<14:04,  1.58it/s]
4|5134|Loss: 0.9669439196586609:  79%|███████▉  | 5134/6470 [58:59<14:39,  1.52it/s]
4|5135|Loss: 0.7134276032447815:  79%|███████▉  | 5134/6470 [59:00<14:39,  1.52it/s]
4|5135|Loss: 0.7134276032447815:  79%|███████▉  | 5135/6470 [59:00<14:13,  1.56it/s]
4|5136|Loss: 0.927394688129425:  79%|███████▉  | 5135/6470 [59:00<14:13,  1.56it/s] 
4|5136|Loss: 0.927394688129425:  79%|███████▉  | 5136/6470 [59:00<14:08,  1.57it/s]
4|5137|Loss: 0.8755682706832886:  79%|███████▉  | 5136/6470 [59:01<14:08,  1.57it/s]
4|5137|Loss: 0.8755682706832886:  79%|███████▉  | 5137/6470 [59:01<14:11,  1.57it/s]
4|5138|Loss: 0.725695013999939:  79%|███████▉  | 5137/6470 [59:01<14:11,  1.57it/s] 
4|5138|Loss: 0.725695013999939:  79%|███████▉  | 5138/6470 [59:02<13:49,  1.61it/s]
4|5139|Loss: 0.903626561164856:  79%|███████▉  | 5138/6470 [59:02<13:49,  1.61it/s]
4|5139|Loss: 0.903626561164856:  79%|███████▉  | 5139/6470 [59:02<14:26,  1.54it/s]
4|5140|Loss: 0.813061535358429:  79%|███████▉  | 5139/6470 [59:03<14:26,  1.54it/s]
4|5140|Loss: 0.813061535358429:  79%|███████▉  | 5140/6470 [59:03<14:28,  1.53it/s]
4|5141|Loss: 0.8834786415100098:  79%|███████▉  | 5140/6470 [59:03<14:28,  1.53it/s]
4|5141|Loss: 0.8834786415100098:  79%|███████▉  | 5141/6470 [59:04<14:14,  1.56it/s]
4|5142|Loss: 0.9481170773506165:  79%|███████▉  | 5141/6470 [59:04<14:14,  1.56it/s]
4|5142|Loss: 0.9481170773506165:  79%|███████▉  | 5142/6470 [59:04<14:43,  1.50it/s]
4|5143|Loss: 0.8615903258323669:  79%|███████▉  | 5142/6470 [59:05<14:43,  1.50it/s]
4|5143|Loss: 0.8615903258323669:  79%|███████▉  | 5143/6470 [59:05<13:52,  1.59it/s]
4|5144|Loss: 0.8826559782028198:  79%|███████▉  | 5143/6470 [59:05<13:52,  1.59it/s]
4|5144|Loss: 0.8826559782028198:  80%|███████▉  | 5144/6470 [59:06<14:16,  1.55it/s]
4|5145|Loss: 0.855290949344635:  80%|███████▉  | 5144/6470 [59:06<14:16,  1.55it/s] 
4|5145|Loss: 0.855290949344635:  80%|███████▉  | 5145/6470 [59:06<14:43,  1.50it/s]
4|5146|Loss: 0.7588210105895996:  80%|███████▉  | 5145/6470 [59:07<14:43,  1.50it/s]
4|5146|Loss: 0.7588210105895996:  80%|███████▉  | 5146/6470 [59:07<15:52,  1.39it/s]
4|5147|Loss: 0.7807047963142395:  80%|███████▉  | 5146/6470 [59:07<15:52,  1.39it/s]
4|5147|Loss: 0.7807047963142395:  80%|███████▉  | 5147/6470 [59:08<14:59,  1.47it/s]
4|5148|Loss: 0.7191442847251892:  80%|███████▉  | 5147/6470 [59:08<14:59,  1.47it/s]
4|5148|Loss: 0.7191442847251892:  80%|███████▉  | 5148/6470 [59:08<15:00,  1.47it/s]
4|5149|Loss: 0.9101015329360962:  80%|███████▉  | 5148/6470 [59:09<15:00,  1.47it/s]
4|5149|Loss: 0.9101015329360962:  80%|███████▉  | 5149/6470 [59:09<15:20,  1.44it/s]
4|5150|Loss: 0.8207805752754211:  80%|███████▉  | 5149/6470 [59:10<15:20,  1.44it/s]
4|5150|Loss: 0.8207805752754211:  80%|███████▉  | 5150/6470 [59:10<15:37,  1.41it/s]
4|5151|Loss: 0.7844575643539429:  80%|███████▉  | 5150/6470 [59:10<15:37,  1.41it/s]
4|5151|Loss: 0.7844575643539429:  80%|███████▉  | 5151/6470 [59:11<15:48,  1.39it/s]
4|5152|Loss: 1.1543536186218262:  80%|███████▉  | 5151/6470 [59:11<15:48,  1.39it/s]
4|5152|Loss: 1.1543536186218262:  80%|███████▉  | 5152/6470 [59:11<16:20,  1.34it/s]
4|5153|Loss: 0.8565201163291931:  80%|███████▉  | 5152/6470 [59:12<16:20,  1.34it/s]
4|5153|Loss: 0.8565201163291931:  80%|███████▉  | 5153/6470 [59:12<17:01,  1.29it/s]
4|5154|Loss: 0.8596993684768677:  80%|███████▉  | 5153/6470 [59:13<17:01,  1.29it/s]
4|5154|Loss: 0.8596993684768677:  80%|███████▉  | 5154/6470 [59:13<16:56,  1.29it/s]
4|5155|Loss: 0.878129780292511:  80%|███████▉  | 5154/6470 [59:13<16:56,  1.29it/s] 
4|5155|Loss: 0.878129780292511:  80%|███████▉  | 5155/6470 [59:14<17:06,  1.28it/s]
4|5156|Loss: 0.8819767832756042:  80%|███████▉  | 5155/6470 [59:14<17:06,  1.28it/s]
4|5156|Loss: 0.8819767832756042:  80%|███████▉  | 5156/6470 [59:14<16:27,  1.33it/s]
4|5157|Loss: 0.9233368635177612:  80%|███████▉  | 5156/6470 [59:15<16:27,  1.33it/s]
4|5157|Loss: 0.9233368635177612:  80%|███████▉  | 5157/6470 [59:15<16:34,  1.32it/s]
4|5158|Loss: 0.9194642305374146:  80%|███████▉  | 5157/6470 [59:16<16:34,  1.32it/s]
4|5158|Loss: 0.9194642305374146:  80%|███████▉  | 5158/6470 [59:16<16:02,  1.36it/s]
4|5159|Loss: 0.8815572261810303:  80%|███████▉  | 5158/6470 [59:16<16:02,  1.36it/s]
4|5159|Loss: 0.8815572261810303:  80%|███████▉  | 5159/6470 [59:17<16:34,  1.32it/s]
4|5160|Loss: 0.7328341007232666:  80%|███████▉  | 5159/6470 [59:17<16:34,  1.32it/s]
4|5160|Loss: 0.7328341007232666:  80%|███████▉  | 5160/6470 [59:17<15:59,  1.37it/s]
4|5161|Loss: 0.8520920872688293:  80%|███████▉  | 5160/6470 [59:18<15:59,  1.37it/s]
4|5161|Loss: 0.8520920872688293:  80%|███████▉  | 5161/6470 [59:18<15:19,  1.42it/s]
4|5162|Loss: 0.9991848468780518:  80%|███████▉  | 5161/6470 [59:18<15:19,  1.42it/s]
4|5162|Loss: 0.9991848468780518:  80%|███████▉  | 5162/6470 [59:19<14:56,  1.46it/s]
4|5163|Loss: 0.9313516020774841:  80%|███████▉  | 5162/6470 [59:19<14:56,  1.46it/s]
4|5163|Loss: 0.9313516020774841:  80%|███████▉  | 5163/6470 [59:20<15:46,  1.38it/s]
4|5164|Loss: 0.9195056557655334:  80%|███████▉  | 5163/6470 [59:20<15:46,  1.38it/s]
4|5164|Loss: 0.9195056557655334:  80%|███████▉  | 5164/6470 [59:20<15:13,  1.43it/s]
4|5165|Loss: 0.7509218454360962:  80%|███████▉  | 5164/6470 [59:21<15:13,  1.43it/s]
4|5165|Loss: 0.7509218454360962:  80%|███████▉  | 5165/6470 [59:21<15:20,  1.42it/s]
4|5166|Loss: 1.0416250228881836:  80%|███████▉  | 5165/6470 [59:21<15:20,  1.42it/s]
4|5166|Loss: 1.0416250228881836:  80%|███████▉  | 5166/6470 [59:21<14:30,  1.50it/s]
4|5167|Loss: 0.9818312525749207:  80%|███████▉  | 5166/6470 [59:22<14:30,  1.50it/s]
4|5167|Loss: 0.9818312525749207:  80%|███████▉  | 5167/6470 [59:22<13:30,  1.61it/s]
4|5168|Loss: 0.9001089334487915:  80%|███████▉  | 5167/6470 [59:22<13:30,  1.61it/s]
4|5168|Loss: 0.9001089334487915:  80%|███████▉  | 5168/6470 [59:23<14:46,  1.47it/s]
4|5169|Loss: 0.7714403867721558:  80%|███████▉  | 5168/6470 [59:23<14:46,  1.47it/s]
4|5169|Loss: 0.7714403867721558:  80%|███████▉  | 5169/6470 [59:24<15:18,  1.42it/s]
4|5170|Loss: 0.7638231515884399:  80%|███████▉  | 5169/6470 [59:24<15:18,  1.42it/s]
4|5170|Loss: 0.7638231515884399:  80%|███████▉  | 5170/6470 [59:24<14:33,  1.49it/s]
4|5171|Loss: 0.8727486729621887:  80%|███████▉  | 5170/6470 [59:24<14:33,  1.49it/s]
4|5171|Loss: 0.8727486729621887:  80%|███████▉  | 5171/6470 [59:25<14:17,  1.52it/s]
4|5172|Loss: 0.8396283388137817:  80%|███████▉  | 5171/6470 [59:25<14:17,  1.52it/s]
4|5172|Loss: 0.8396283388137817:  80%|███████▉  | 5172/6470 [59:25<14:11,  1.52it/s]
4|5173|Loss: 0.8221694827079773:  80%|███████▉  | 5172/6470 [59:26<14:11,  1.52it/s]
4|5173|Loss: 0.8221694827079773:  80%|███████▉  | 5173/6470 [59:26<14:41,  1.47it/s]
4|5174|Loss: 0.8352049589157104:  80%|███████▉  | 5173/6470 [59:27<14:41,  1.47it/s]
4|5174|Loss: 0.8352049589157104:  80%|███████▉  | 5174/6470 [59:27<14:17,  1.51it/s]
4|5175|Loss: 0.8514794707298279:  80%|███████▉  | 5174/6470 [59:27<14:17,  1.51it/s]
4|5175|Loss: 0.8514794707298279:  80%|███████▉  | 5175/6470 [59:27<13:15,  1.63it/s]
4|5176|Loss: 1.1128109693527222:  80%|███████▉  | 5175/6470 [59:28<13:15,  1.63it/s]
4|5176|Loss: 1.1128109693527222:  80%|████████  | 5176/6470 [59:28<13:05,  1.65it/s]
4|5177|Loss: 0.7331191897392273:  80%|████████  | 5176/6470 [59:28<13:05,  1.65it/s]
4|5177|Loss: 0.7331191897392273:  80%|████████  | 5177/6470 [59:29<13:49,  1.56it/s]
4|5178|Loss: 0.8655452728271484:  80%|████████  | 5177/6470 [59:29<13:49,  1.56it/s]
4|5178|Loss: 0.8655452728271484:  80%|████████  | 5178/6470 [59:29<13:23,  1.61it/s]
4|5179|Loss: 0.7477294206619263:  80%|████████  | 5178/6470 [59:29<13:23,  1.61it/s]
4|5179|Loss: 0.7477294206619263:  80%|████████  | 5179/6470 [59:30<12:54,  1.67it/s]
4|5180|Loss: 0.9064770340919495:  80%|████████  | 5179/6470 [59:30<12:54,  1.67it/s]
4|5180|Loss: 0.9064770340919495:  80%|████████  | 5180/6470 [59:30<13:31,  1.59it/s]
4|5181|Loss: 0.8929927349090576:  80%|████████  | 5180/6470 [59:31<13:31,  1.59it/s]
4|5181|Loss: 0.8929927349090576:  80%|████████  | 5181/6470 [59:31<14:13,  1.51it/s]
4|5182|Loss: 0.8522764444351196:  80%|████████  | 5181/6470 [59:31<14:13,  1.51it/s]
4|5182|Loss: 0.8522764444351196:  80%|████████  | 5182/6470 [59:32<12:54,  1.66it/s]
4|5183|Loss: 0.9588817954063416:  80%|████████  | 5182/6470 [59:32<12:54,  1.66it/s]
4|5183|Loss: 0.9588817954063416:  80%|████████  | 5183/6470 [59:32<12:50,  1.67it/s]
4|5184|Loss: 0.919905424118042:  80%|████████  | 5183/6470 [59:33<12:50,  1.67it/s] 
4|5184|Loss: 0.919905424118042:  80%|████████  | 5184/6470 [59:33<12:45,  1.68it/s]
4|5185|Loss: 0.7880465388298035:  80%|████████  | 5184/6470 [59:33<12:45,  1.68it/s]
4|5185|Loss: 0.7880465388298035:  80%|████████  | 5185/6470 [59:34<14:12,  1.51it/s]
4|5186|Loss: 0.8084467649459839:  80%|████████  | 5185/6470 [59:34<14:12,  1.51it/s]
4|5186|Loss: 0.8084467649459839:  80%|████████  | 5186/6470 [59:34<15:12,  1.41it/s]
4|5187|Loss: 0.7518342733383179:  80%|████████  | 5186/6470 [59:35<15:12,  1.41it/s]
4|5187|Loss: 0.7518342733383179:  80%|████████  | 5187/6470 [59:35<13:40,  1.56it/s]
4|5188|Loss: 0.8610370755195618:  80%|████████  | 5187/6470 [59:35<13:40,  1.56it/s]
4|5188|Loss: 0.8610370755195618:  80%|████████  | 5188/6470 [59:36<13:51,  1.54it/s]
4|5189|Loss: 0.7758896946907043:  80%|████████  | 5188/6470 [59:36<13:51,  1.54it/s]
4|5189|Loss: 0.7758896946907043:  80%|████████  | 5189/6470 [59:36<13:50,  1.54it/s]
4|5190|Loss: 0.8975029587745667:  80%|████████  | 5189/6470 [59:37<13:50,  1.54it/s]
4|5190|Loss: 0.8975029587745667:  80%|████████  | 5190/6470 [59:37<14:04,  1.51it/s]
4|5191|Loss: 0.8659451603889465:  80%|████████  | 5190/6470 [59:37<14:04,  1.51it/s]
4|5191|Loss: 0.8659451603889465:  80%|████████  | 5191/6470 [59:38<14:01,  1.52it/s]
4|5192|Loss: 0.8618780970573425:  80%|████████  | 5191/6470 [59:38<14:01,  1.52it/s]
4|5192|Loss: 0.8618780970573425:  80%|████████  | 5192/6470 [59:38<13:27,  1.58it/s]
4|5193|Loss: 0.8904775381088257:  80%|████████  | 5192/6470 [59:38<13:27,  1.58it/s]
4|5193|Loss: 0.8904775381088257:  80%|████████  | 5193/6470 [59:39<13:23,  1.59it/s]
4|5194|Loss: 0.9402180910110474:  80%|████████  | 5193/6470 [59:39<13:23,  1.59it/s]
4|5194|Loss: 0.9402180910110474:  80%|████████  | 5194/6470 [59:39<14:02,  1.51it/s]
4|5195|Loss: 0.9596627950668335:  80%|████████  | 5194/6470 [59:40<14:02,  1.51it/s]
4|5195|Loss: 0.9596627950668335:  80%|████████  | 5195/6470 [59:40<15:12,  1.40it/s]
4|5196|Loss: 0.8144809007644653:  80%|████████  | 5195/6470 [59:41<15:12,  1.40it/s]
4|5196|Loss: 0.8144809007644653:  80%|████████  | 5196/6470 [59:41<15:48,  1.34it/s]
4|5197|Loss: 0.8774430751800537:  80%|████████  | 5196/6470 [59:42<15:48,  1.34it/s]
4|5197|Loss: 0.8774430751800537:  80%|████████  | 5197/6470 [59:42<16:28,  1.29it/s]
4|5198|Loss: 0.9161203503608704:  80%|████████  | 5197/6470 [59:42<16:28,  1.29it/s]
4|5198|Loss: 0.9161203503608704:  80%|████████  | 5198/6470 [59:43<17:20,  1.22it/s]
4|5199|Loss: 0.6958012580871582:  80%|████████  | 5198/6470 [59:43<17:20,  1.22it/s]
4|5199|Loss: 0.6958012580871582:  80%|████████  | 5199/6470 [59:44<17:31,  1.21it/s]
4|5200|Loss: 0.8317328691482544:  80%|████████  | 5199/6470 [59:44<17:31,  1.21it/s]
4|5200|Loss: 0.8317328691482544:  80%|████████  | 5200/6470 [59:44<16:52,  1.25it/s]
4|5201|Loss: 0.8161988854408264:  80%|████████  | 5200/6470 [59:45<16:52,  1.25it/s]
4|5201|Loss: 0.8161988854408264:  80%|████████  | 5201/6470 [59:45<16:14,  1.30it/s]
4|5202|Loss: 0.7722294330596924:  80%|████████  | 5201/6470 [59:46<16:14,  1.30it/s]
4|5202|Loss: 0.7722294330596924:  80%|████████  | 5202/6470 [59:46<16:28,  1.28it/s]
4|5203|Loss: 0.898504912853241:  80%|████████  | 5202/6470 [59:46<16:28,  1.28it/s] 
4|5203|Loss: 0.898504912853241:  80%|████████  | 5203/6470 [59:47<17:01,  1.24it/s]
4|5204|Loss: 0.8166742920875549:  80%|████████  | 5203/6470 [59:47<17:01,  1.24it/s]
4|5204|Loss: 0.8166742920875549:  80%|████████  | 5204/6470 [59:48<17:02,  1.24it/s]
4|5205|Loss: 0.896591305732727:  80%|████████  | 5204/6470 [59:48<17:02,  1.24it/s] 
4|5205|Loss: 0.896591305732727:  80%|████████  | 5205/6470 [59:49<17:19,  1.22it/s]
4|5206|Loss: 0.8052717447280884:  80%|████████  | 5205/6470 [59:49<17:19,  1.22it/s]
4|5206|Loss: 0.8052717447280884:  80%|████████  | 5206/6470 [59:49<17:32,  1.20it/s]
4|5207|Loss: 0.9771667718887329:  80%|████████  | 5206/6470 [59:50<17:32,  1.20it/s]
4|5207|Loss: 0.9771667718887329:  80%|████████  | 5207/6470 [59:50<15:39,  1.34it/s]
4|5208|Loss: 0.8656381964683533:  80%|████████  | 5207/6470 [59:50<15:39,  1.34it/s]
4|5208|Loss: 0.8656381964683533:  80%|████████  | 5208/6470 [59:51<15:15,  1.38it/s]
4|5209|Loss: 1.0398547649383545:  80%|████████  | 5208/6470 [59:51<15:15,  1.38it/s]
4|5209|Loss: 1.0398547649383545:  81%|████████  | 5209/6470 [59:51<15:11,  1.38it/s]
4|5210|Loss: 0.9160886406898499:  81%|████████  | 5209/6470 [59:52<15:11,  1.38it/s]
4|5210|Loss: 0.9160886406898499:  81%|████████  | 5210/6470 [59:52<13:56,  1.51it/s]
4|5211|Loss: 0.8345825672149658:  81%|████████  | 5210/6470 [59:52<13:56,  1.51it/s]
4|5211|Loss: 0.8345825672149658:  81%|████████  | 5211/6470 [59:52<12:46,  1.64it/s]
4|5212|Loss: 0.842951774597168:  81%|████████  | 5211/6470 [59:53<12:46,  1.64it/s] 
4|5212|Loss: 0.842951774597168:  81%|████████  | 5212/6470 [59:53<12:26,  1.69it/s]
4|5213|Loss: 0.7801068425178528:  81%|████████  | 5212/6470 [59:53<12:26,  1.69it/s]
4|5213|Loss: 0.7801068425178528:  81%|████████  | 5213/6470 [59:54<13:47,  1.52it/s]
4|5214|Loss: 1.0545337200164795:  81%|████████  | 5213/6470 [59:54<13:47,  1.52it/s]
4|5214|Loss: 1.0545337200164795:  81%|████████  | 5214/6470 [59:55<15:03,  1.39it/s]
4|5215|Loss: 0.959973931312561:  81%|████████  | 5214/6470 [59:55<15:03,  1.39it/s] 
4|5215|Loss: 0.959973931312561:  81%|████████  | 5215/6470 [59:55<16:00,  1.31it/s]
4|5216|Loss: 0.9144185185432434:  81%|████████  | 5215/6470 [59:56<16:00,  1.31it/s]
4|5216|Loss: 0.9144185185432434:  81%|████████  | 5216/6470 [59:56<15:25,  1.36it/s]
4|5217|Loss: 0.7982520461082458:  81%|████████  | 5216/6470 [59:56<15:25,  1.36it/s]
4|5217|Loss: 0.7982520461082458:  81%|████████  | 5217/6470 [59:57<14:27,  1.44it/s]
4|5218|Loss: 0.8727814555168152:  81%|████████  | 5217/6470 [59:57<14:27,  1.44it/s]
4|5218|Loss: 0.8727814555168152:  81%|████████  | 5218/6470 [59:57<15:06,  1.38it/s]
4|5219|Loss: 0.8850293159484863:  81%|████████  | 5218/6470 [59:58<15:06,  1.38it/s]
4|5219|Loss: 0.8850293159484863:  81%|████████  | 5219/6470 [59:58<15:21,  1.36it/s]
4|5220|Loss: 0.9184722900390625:  81%|████████  | 5219/6470 [59:59<15:21,  1.36it/s]
4|5220|Loss: 0.9184722900390625:  81%|████████  | 5220/6470 [59:59<15:21,  1.36it/s]
4|5221|Loss: 0.9179351925849915:  81%|████████  | 5220/6470 [59:59<15:21,  1.36it/s]
4|5221|Loss: 0.9179351925849915:  81%|████████  | 5221/6470 [1:00:00<15:11,  1.37it/s]
4|5222|Loss: 0.8572213649749756:  81%|████████  | 5221/6470 [1:00:00<15:11,  1.37it/s]
4|5222|Loss: 0.8572213649749756:  81%|████████  | 5222/6470 [1:00:01<15:44,  1.32it/s]
4|5223|Loss: 0.7570160031318665:  81%|████████  | 5222/6470 [1:00:01<15:44,  1.32it/s]
4|5223|Loss: 0.7570160031318665:  81%|████████  | 5223/6470 [1:00:01<14:58,  1.39it/s]
4|5224|Loss: 0.8547374606132507:  81%|████████  | 5223/6470 [1:00:02<14:58,  1.39it/s]
4|5224|Loss: 0.8547374606132507:  81%|████████  | 5224/6470 [1:00:02<15:08,  1.37it/s]
4|5225|Loss: 1.0042502880096436:  81%|████████  | 5224/6470 [1:00:02<15:08,  1.37it/s]
4|5225|Loss: 1.0042502880096436:  81%|████████  | 5225/6470 [1:00:03<16:08,  1.29it/s]
4|5226|Loss: 0.8078832030296326:  81%|████████  | 5225/6470 [1:00:03<16:08,  1.29it/s]
4|5226|Loss: 0.8078832030296326:  81%|████████  | 5226/6470 [1:00:04<16:12,  1.28it/s]
4|5227|Loss: 0.8855261206626892:  81%|████████  | 5226/6470 [1:00:04<16:12,  1.28it/s]
4|5227|Loss: 0.8855261206626892:  81%|████████  | 5227/6470 [1:00:04<16:01,  1.29it/s]
4|5228|Loss: 0.8905199766159058:  81%|████████  | 5227/6470 [1:00:05<16:01,  1.29it/s]
4|5228|Loss: 0.8905199766159058:  81%|████████  | 5228/6470 [1:00:05<16:26,  1.26it/s]
4|5229|Loss: 0.8457589149475098:  81%|████████  | 5228/6470 [1:00:06<16:26,  1.26it/s]
4|5229|Loss: 0.8457589149475098:  81%|████████  | 5229/6470 [1:00:06<15:31,  1.33it/s]
4|5230|Loss: 0.5770717859268188:  81%|████████  | 5229/6470 [1:00:06<15:31,  1.33it/s]
4|5230|Loss: 0.5770717859268188:  81%|████████  | 5230/6470 [1:00:07<16:18,  1.27it/s]
4|5231|Loss: 0.999793291091919:  81%|████████  | 5230/6470 [1:00:07<16:18,  1.27it/s] 
4|5231|Loss: 0.999793291091919:  81%|████████  | 5231/6470 [1:00:08<16:51,  1.22it/s]
4|5232|Loss: 0.9619486331939697:  81%|████████  | 5231/6470 [1:00:08<16:51,  1.22it/s]
4|5232|Loss: 0.9619486331939697:  81%|████████  | 5232/6470 [1:00:08<16:23,  1.26it/s]
4|5233|Loss: 0.9177193641662598:  81%|████████  | 5232/6470 [1:00:09<16:23,  1.26it/s]
4|5233|Loss: 0.9177193641662598:  81%|████████  | 5233/6470 [1:00:09<16:05,  1.28it/s]
4|5234|Loss: 0.9957057237625122:  81%|████████  | 5233/6470 [1:00:10<16:05,  1.28it/s]
4|5234|Loss: 0.9957057237625122:  81%|████████  | 5234/6470 [1:00:10<15:49,  1.30it/s]
4|5235|Loss: 0.7921785116195679:  81%|████████  | 5234/6470 [1:00:10<15:49,  1.30it/s]
4|5235|Loss: 0.7921785116195679:  81%|████████  | 5235/6470 [1:00:11<15:24,  1.34it/s]
4|5236|Loss: 0.7766781449317932:  81%|████████  | 5235/6470 [1:00:11<15:24,  1.34it/s]
4|5236|Loss: 0.7766781449317932:  81%|████████  | 5236/6470 [1:00:11<13:50,  1.49it/s]
4|5237|Loss: 1.0093411207199097:  81%|████████  | 5236/6470 [1:00:11<13:50,  1.49it/s]
4|5237|Loss: 1.0093411207199097:  81%|████████  | 5237/6470 [1:00:12<13:16,  1.55it/s]
4|5238|Loss: 0.9497146010398865:  81%|████████  | 5237/6470 [1:00:12<13:16,  1.55it/s]
4|5238|Loss: 0.9497146010398865:  81%|████████  | 5238/6470 [1:00:12<14:16,  1.44it/s]
4|5239|Loss: 0.8123862147331238:  81%|████████  | 5238/6470 [1:00:13<14:16,  1.44it/s]
4|5239|Loss: 0.8123862147331238:  81%|████████  | 5239/6470 [1:00:13<13:57,  1.47it/s]
4|5240|Loss: 0.8911153078079224:  81%|████████  | 5239/6470 [1:00:13<13:57,  1.47it/s]
4|5240|Loss: 0.8911153078079224:  81%|████████  | 5240/6470 [1:00:14<13:32,  1.51it/s]
4|5241|Loss: 0.9662575721740723:  81%|████████  | 5240/6470 [1:00:14<13:32,  1.51it/s]
4|5241|Loss: 0.9662575721740723:  81%|████████  | 5241/6470 [1:00:14<13:43,  1.49it/s]
4|5242|Loss: 0.7667221426963806:  81%|████████  | 5241/6470 [1:00:15<13:43,  1.49it/s]
4|5242|Loss: 0.7667221426963806:  81%|████████  | 5242/6470 [1:00:15<13:29,  1.52it/s]
4|5243|Loss: 0.7921144962310791:  81%|████████  | 5242/6470 [1:00:15<13:29,  1.52it/s]
4|5243|Loss: 0.7921144962310791:  81%|████████  | 5243/6470 [1:00:16<13:19,  1.53it/s]
4|5244|Loss: 0.8024658560752869:  81%|████████  | 5243/6470 [1:00:16<13:19,  1.53it/s]
4|5244|Loss: 0.8024658560752869:  81%|████████  | 5244/6470 [1:00:16<13:42,  1.49it/s]
4|5245|Loss: 0.8473566770553589:  81%|████████  | 5244/6470 [1:00:17<13:42,  1.49it/s]
4|5245|Loss: 0.8473566770553589:  81%|████████  | 5245/6470 [1:00:17<13:30,  1.51it/s]
4|5246|Loss: 0.8253738284111023:  81%|████████  | 5245/6470 [1:00:17<13:30,  1.51it/s]
4|5246|Loss: 0.8253738284111023:  81%|████████  | 5246/6470 [1:00:18<13:47,  1.48it/s]
4|5247|Loss: 1.0375126600265503:  81%|████████  | 5246/6470 [1:00:18<13:47,  1.48it/s]
4|5247|Loss: 1.0375126600265503:  81%|████████  | 5247/6470 [1:00:18<13:59,  1.46it/s]
4|5248|Loss: 0.8927651047706604:  81%|████████  | 5247/6470 [1:00:19<13:59,  1.46it/s]
4|5248|Loss: 0.8927651047706604:  81%|████████  | 5248/6470 [1:00:19<13:30,  1.51it/s]
4|5249|Loss: 0.8896225690841675:  81%|████████  | 5248/6470 [1:00:19<13:30,  1.51it/s]
4|5249|Loss: 0.8896225690841675:  81%|████████  | 5249/6470 [1:00:20<14:13,  1.43it/s]
4|5250|Loss: 0.9023972153663635:  81%|████████  | 5249/6470 [1:00:20<14:13,  1.43it/s]
4|5250|Loss: 0.9023972153663635:  81%|████████  | 5250/6470 [1:00:21<14:59,  1.36it/s]
4|5251|Loss: 0.8933891654014587:  81%|████████  | 5250/6470 [1:00:21<14:59,  1.36it/s]
4|5251|Loss: 0.8933891654014587:  81%|████████  | 5251/6470 [1:00:21<14:20,  1.42it/s]
4|5252|Loss: 0.8723601698875427:  81%|████████  | 5251/6470 [1:00:22<14:20,  1.42it/s]
4|5252|Loss: 0.8723601698875427:  81%|████████  | 5252/6470 [1:00:22<14:33,  1.39it/s]
4|5253|Loss: 0.8087601065635681:  81%|████████  | 5252/6470 [1:00:22<14:33,  1.39it/s]
4|5253|Loss: 0.8087601065635681:  81%|████████  | 5253/6470 [1:00:23<15:17,  1.33it/s]
4|5254|Loss: 0.8098496794700623:  81%|████████  | 5253/6470 [1:00:23<15:17,  1.33it/s]
4|5254|Loss: 0.8098496794700623:  81%|████████  | 5254/6470 [1:00:24<15:20,  1.32it/s]
4|5255|Loss: 0.7941953539848328:  81%|████████  | 5254/6470 [1:00:24<15:20,  1.32it/s]
4|5255|Loss: 0.7941953539848328:  81%|████████  | 5255/6470 [1:00:24<15:47,  1.28it/s]
4|5256|Loss: 0.9976592659950256:  81%|████████  | 5255/6470 [1:00:25<15:47,  1.28it/s]
4|5256|Loss: 0.9976592659950256:  81%|████████  | 5256/6470 [1:00:25<15:16,  1.32it/s]
4|5257|Loss: 0.7921872735023499:  81%|████████  | 5256/6470 [1:00:26<15:16,  1.32it/s]
4|5257|Loss: 0.7921872735023499:  81%|████████▏ | 5257/6470 [1:00:26<15:42,  1.29it/s]
4|5258|Loss: 0.8888357877731323:  81%|████████▏ | 5257/6470 [1:00:26<15:42,  1.29it/s]
4|5258|Loss: 0.8888357877731323:  81%|████████▏ | 5258/6470 [1:00:27<16:06,  1.25it/s]
4|5259|Loss: 0.8609402775764465:  81%|████████▏ | 5258/6470 [1:00:27<16:06,  1.25it/s]
4|5259|Loss: 0.8609402775764465:  81%|████████▏ | 5259/6470 [1:00:28<16:19,  1.24it/s]
4|5260|Loss: 0.8118771910667419:  81%|████████▏ | 5259/6470 [1:00:28<16:19,  1.24it/s]
4|5260|Loss: 0.8118771910667419:  81%|████████▏ | 5260/6470 [1:00:28<15:26,  1.31it/s]
4|5261|Loss: 0.9153761267662048:  81%|████████▏ | 5260/6470 [1:00:29<15:26,  1.31it/s]
4|5261|Loss: 0.9153761267662048:  81%|████████▏ | 5261/6470 [1:00:29<15:33,  1.30it/s]
4|5262|Loss: 0.8337977528572083:  81%|████████▏ | 5261/6470 [1:00:30<15:33,  1.30it/s]
4|5262|Loss: 0.8337977528572083:  81%|████████▏ | 5262/6470 [1:00:30<15:18,  1.32it/s]
4|5263|Loss: 0.8635382652282715:  81%|████████▏ | 5262/6470 [1:00:30<15:18,  1.32it/s]
4|5263|Loss: 0.8635382652282715:  81%|████████▏ | 5263/6470 [1:00:31<15:49,  1.27it/s]
4|5264|Loss: 0.8330244421958923:  81%|████████▏ | 5263/6470 [1:00:31<15:49,  1.27it/s]
4|5264|Loss: 0.8330244421958923:  81%|████████▏ | 5264/6470 [1:00:31<15:11,  1.32it/s]
4|5265|Loss: 0.9750083684921265:  81%|████████▏ | 5264/6470 [1:00:32<15:11,  1.32it/s]
4|5265|Loss: 0.9750083684921265:  81%|████████▏ | 5265/6470 [1:00:32<14:53,  1.35it/s]
4|5266|Loss: 0.9880324602127075:  81%|████████▏ | 5265/6470 [1:00:32<14:53,  1.35it/s]
4|5266|Loss: 0.9880324602127075:  81%|████████▏ | 5266/6470 [1:00:33<13:38,  1.47it/s]
4|5267|Loss: 0.8475685715675354:  81%|████████▏ | 5266/6470 [1:00:33<13:38,  1.47it/s]
4|5267|Loss: 0.8475685715675354:  81%|████████▏ | 5267/6470 [1:00:33<13:11,  1.52it/s]
4|5268|Loss: 0.9348300099372864:  81%|████████▏ | 5267/6470 [1:00:34<13:11,  1.52it/s]
4|5268|Loss: 0.9348300099372864:  81%|████████▏ | 5268/6470 [1:00:34<14:07,  1.42it/s]
4|5269|Loss: 0.8371416926383972:  81%|████████▏ | 5268/6470 [1:00:34<14:07,  1.42it/s]
4|5269|Loss: 0.8371416926383972:  81%|████████▏ | 5269/6470 [1:00:35<14:03,  1.42it/s]
4|5270|Loss: 1.0906283855438232:  81%|████████▏ | 5269/6470 [1:00:35<14:03,  1.42it/s]
4|5270|Loss: 1.0906283855438232:  81%|████████▏ | 5270/6470 [1:00:35<13:21,  1.50it/s]
4|5271|Loss: 0.6266773343086243:  81%|████████▏ | 5270/6470 [1:00:36<13:21,  1.50it/s]
4|5271|Loss: 0.6266773343086243:  81%|████████▏ | 5271/6470 [1:00:36<13:23,  1.49it/s]
4|5272|Loss: 0.9006472229957581:  81%|████████▏ | 5271/6470 [1:00:36<13:23,  1.49it/s]
4|5272|Loss: 0.9006472229957581:  81%|████████▏ | 5272/6470 [1:00:37<13:36,  1.47it/s]
4|5273|Loss: 0.833823561668396:  81%|████████▏ | 5272/6470 [1:00:37<13:36,  1.47it/s] 
4|5273|Loss: 0.833823561668396:  81%|████████▏ | 5273/6470 [1:00:37<13:25,  1.49it/s]
4|5274|Loss: 0.7796515822410583:  81%|████████▏ | 5273/6470 [1:00:38<13:25,  1.49it/s]
4|5274|Loss: 0.7796515822410583:  82%|████████▏ | 5274/6470 [1:00:38<12:34,  1.59it/s]
4|5275|Loss: 1.0400861501693726:  82%|████████▏ | 5274/6470 [1:00:38<12:34,  1.59it/s]
4|5275|Loss: 1.0400861501693726:  82%|████████▏ | 5275/6470 [1:00:39<12:59,  1.53it/s]
4|5276|Loss: 0.914261519908905:  82%|████████▏ | 5275/6470 [1:00:39<12:59,  1.53it/s] 
4|5276|Loss: 0.914261519908905:  82%|████████▏ | 5276/6470 [1:00:39<14:13,  1.40it/s]
4|5277|Loss: 0.8067284226417542:  82%|████████▏ | 5276/6470 [1:00:40<14:13,  1.40it/s]
4|5277|Loss: 0.8067284226417542:  82%|████████▏ | 5277/6470 [1:00:40<14:33,  1.37it/s]
4|5278|Loss: 0.9254749417304993:  82%|████████▏ | 5277/6470 [1:00:41<14:33,  1.37it/s]
4|5278|Loss: 0.9254749417304993:  82%|████████▏ | 5278/6470 [1:00:41<14:22,  1.38it/s]
4|5279|Loss: 0.7723962664604187:  82%|████████▏ | 5278/6470 [1:00:41<14:22,  1.38it/s]
4|5279|Loss: 0.7723962664604187:  82%|████████▏ | 5279/6470 [1:00:42<14:41,  1.35it/s]
4|5280|Loss: 0.7968570590019226:  82%|████████▏ | 5279/6470 [1:00:42<14:41,  1.35it/s]
4|5280|Loss: 0.7968570590019226:  82%|████████▏ | 5280/6470 [1:00:42<13:39,  1.45it/s]
4|5281|Loss: 0.8755025863647461:  82%|████████▏ | 5280/6470 [1:00:43<13:39,  1.45it/s]
4|5281|Loss: 0.8755025863647461:  82%|████████▏ | 5281/6470 [1:00:43<13:43,  1.44it/s]
4|5282|Loss: 0.8502376079559326:  82%|████████▏ | 5281/6470 [1:00:43<13:43,  1.44it/s]
4|5282|Loss: 0.8502376079559326:  82%|████████▏ | 5282/6470 [1:00:44<13:26,  1.47it/s]
4|5283|Loss: 0.8838154673576355:  82%|████████▏ | 5282/6470 [1:00:44<13:26,  1.47it/s]
4|5283|Loss: 0.8838154673576355:  82%|████████▏ | 5283/6470 [1:00:44<12:55,  1.53it/s]
4|5284|Loss: 0.8741408586502075:  82%|████████▏ | 5283/6470 [1:00:45<12:55,  1.53it/s]
4|5284|Loss: 0.8741408586502075:  82%|████████▏ | 5284/6470 [1:00:45<11:54,  1.66it/s]
4|5285|Loss: 0.9048657417297363:  82%|████████▏ | 5284/6470 [1:00:45<11:54,  1.66it/s]
4|5285|Loss: 0.9048657417297363:  82%|████████▏ | 5285/6470 [1:00:45<12:33,  1.57it/s]
4|5286|Loss: 0.980754017829895:  82%|████████▏ | 5285/6470 [1:00:46<12:33,  1.57it/s] 
4|5286|Loss: 0.980754017829895:  82%|████████▏ | 5286/6470 [1:00:46<13:17,  1.48it/s]
4|5287|Loss: 0.9218782186508179:  82%|████████▏ | 5286/6470 [1:00:47<13:17,  1.48it/s]
4|5287|Loss: 0.9218782186508179:  82%|████████▏ | 5287/6470 [1:00:47<13:33,  1.45it/s]
4|5288|Loss: 1.0667299032211304:  82%|████████▏ | 5287/6470 [1:00:47<13:33,  1.45it/s]
4|5288|Loss: 1.0667299032211304:  82%|████████▏ | 5288/6470 [1:00:47<12:07,  1.62it/s]
4|5289|Loss: 0.829990804195404:  82%|████████▏ | 5288/6470 [1:00:48<12:07,  1.62it/s] 
4|5289|Loss: 0.829990804195404:  82%|████████▏ | 5289/6470 [1:00:48<12:55,  1.52it/s]
4|5290|Loss: 0.9182056188583374:  82%|████████▏ | 5289/6470 [1:00:49<12:55,  1.52it/s]
4|5290|Loss: 0.9182056188583374:  82%|████████▏ | 5290/6470 [1:00:49<13:23,  1.47it/s]
4|5291|Loss: 0.7821910977363586:  82%|████████▏ | 5290/6470 [1:00:49<13:23,  1.47it/s]
4|5291|Loss: 0.7821910977363586:  82%|████████▏ | 5291/6470 [1:00:50<14:14,  1.38it/s]
4|5292|Loss: 0.6718327403068542:  82%|████████▏ | 5291/6470 [1:00:50<14:14,  1.38it/s]
4|5292|Loss: 0.6718327403068542:  82%|████████▏ | 5292/6470 [1:00:50<14:52,  1.32it/s]
4|5293|Loss: 0.8525445461273193:  82%|████████▏ | 5292/6470 [1:00:51<14:52,  1.32it/s]
4|5293|Loss: 0.8525445461273193:  82%|████████▏ | 5293/6470 [1:00:51<15:17,  1.28it/s]
4|5294|Loss: 0.7705305814743042:  82%|████████▏ | 5293/6470 [1:00:52<15:17,  1.28it/s]
4|5294|Loss: 0.7705305814743042:  82%|████████▏ | 5294/6470 [1:00:52<14:42,  1.33it/s]
4|5295|Loss: 0.844495952129364:  82%|████████▏ | 5294/6470 [1:00:52<14:42,  1.33it/s] 
4|5295|Loss: 0.844495952129364:  82%|████████▏ | 5295/6470 [1:00:53<14:32,  1.35it/s]
4|5296|Loss: 0.8035966157913208:  82%|████████▏ | 5295/6470 [1:00:53<14:32,  1.35it/s]
4|5296|Loss: 0.8035966157913208:  82%|████████▏ | 5296/6470 [1:00:53<14:01,  1.40it/s]
4|5297|Loss: 1.1475173234939575:  82%|████████▏ | 5296/6470 [1:00:54<14:01,  1.40it/s]
4|5297|Loss: 1.1475173234939575:  82%|████████▏ | 5297/6470 [1:00:54<14:49,  1.32it/s]
4|5298|Loss: 0.8547452688217163:  82%|████████▏ | 5297/6470 [1:00:55<14:49,  1.32it/s]
4|5298|Loss: 0.8547452688217163:  82%|████████▏ | 5298/6470 [1:00:55<14:45,  1.32it/s]
4|5299|Loss: 0.8300338983535767:  82%|████████▏ | 5298/6470 [1:00:55<14:45,  1.32it/s]
4|5299|Loss: 0.8300338983535767:  82%|████████▏ | 5299/6470 [1:00:56<13:43,  1.42it/s]
4|5300|Loss: 0.855673611164093:  82%|████████▏ | 5299/6470 [1:00:56<13:43,  1.42it/s] 
4|5300|Loss: 0.855673611164093:  82%|████████▏ | 5300/6470 [1:00:56<13:33,  1.44it/s]
4|5301|Loss: 0.9817913770675659:  82%|████████▏ | 5300/6470 [1:00:57<13:33,  1.44it/s]
4|5301|Loss: 0.9817913770675659:  82%|████████▏ | 5301/6470 [1:00:57<13:44,  1.42it/s]
4|5302|Loss: 0.8854382038116455:  82%|████████▏ | 5301/6470 [1:00:57<13:44,  1.42it/s]
4|5302|Loss: 0.8854382038116455:  82%|████████▏ | 5302/6470 [1:00:58<14:01,  1.39it/s]
4|5303|Loss: 0.7393633723258972:  82%|████████▏ | 5302/6470 [1:00:58<14:01,  1.39it/s]
4|5303|Loss: 0.7393633723258972:  82%|████████▏ | 5303/6470 [1:00:58<13:16,  1.46it/s]
4|5304|Loss: 0.8347699642181396:  82%|████████▏ | 5303/6470 [1:00:59<13:16,  1.46it/s]
4|5304|Loss: 0.8347699642181396:  82%|████████▏ | 5304/6470 [1:00:59<12:30,  1.55it/s]
4|5305|Loss: 0.9572120904922485:  82%|████████▏ | 5304/6470 [1:00:59<12:30,  1.55it/s]
4|5305|Loss: 0.9572120904922485:  82%|████████▏ | 5305/6470 [1:00:59<12:21,  1.57it/s]
4|5306|Loss: 0.8755220174789429:  82%|████████▏ | 5305/6470 [1:01:00<12:21,  1.57it/s]
4|5306|Loss: 0.8755220174789429:  82%|████████▏ | 5306/6470 [1:01:00<12:08,  1.60it/s]
4|5307|Loss: 0.8138145804405212:  82%|████████▏ | 5306/6470 [1:01:00<12:08,  1.60it/s]
4|5307|Loss: 0.8138145804405212:  82%|████████▏ | 5307/6470 [1:01:01<12:26,  1.56it/s]
4|5308|Loss: 0.9624340534210205:  82%|████████▏ | 5307/6470 [1:01:01<12:26,  1.56it/s]
4|5308|Loss: 0.9624340534210205:  82%|████████▏ | 5308/6470 [1:01:01<12:45,  1.52it/s]
4|5309|Loss: 0.8365225791931152:  82%|████████▏ | 5308/6470 [1:01:02<12:45,  1.52it/s]
4|5309|Loss: 0.8365225791931152:  82%|████████▏ | 5309/6470 [1:01:02<13:48,  1.40it/s]
4|5310|Loss: 0.8283350467681885:  82%|████████▏ | 5309/6470 [1:01:03<13:48,  1.40it/s]
4|5310|Loss: 0.8283350467681885:  82%|████████▏ | 5310/6470 [1:01:03<14:33,  1.33it/s]
4|5311|Loss: 0.8608365654945374:  82%|████████▏ | 5310/6470 [1:01:04<14:33,  1.33it/s]
4|5311|Loss: 0.8608365654945374:  82%|████████▏ | 5311/6470 [1:01:04<13:29,  1.43it/s]
4|5312|Loss: 0.7334270477294922:  82%|████████▏ | 5311/6470 [1:01:04<13:29,  1.43it/s]
4|5312|Loss: 0.7334270477294922:  82%|████████▏ | 5312/6470 [1:01:04<12:43,  1.52it/s]
4|5313|Loss: 0.8896994590759277:  82%|████████▏ | 5312/6470 [1:01:05<12:43,  1.52it/s]
4|5313|Loss: 0.8896994590759277:  82%|████████▏ | 5313/6470 [1:01:05<12:26,  1.55it/s]
4|5314|Loss: 0.8143372535705566:  82%|████████▏ | 5313/6470 [1:01:05<12:26,  1.55it/s]
4|5314|Loss: 0.8143372535705566:  82%|████████▏ | 5314/6470 [1:01:06<12:23,  1.55it/s]
4|5315|Loss: 0.7742084860801697:  82%|████████▏ | 5314/6470 [1:01:06<12:23,  1.55it/s]
4|5315|Loss: 0.7742084860801697:  82%|████████▏ | 5315/6470 [1:01:06<13:06,  1.47it/s]
4|5316|Loss: 0.9240959286689758:  82%|████████▏ | 5315/6470 [1:01:07<13:06,  1.47it/s]
4|5316|Loss: 0.9240959286689758:  82%|████████▏ | 5316/6470 [1:01:07<13:36,  1.41it/s]
4|5317|Loss: 0.9307997226715088:  82%|████████▏ | 5316/6470 [1:01:08<13:36,  1.41it/s]
4|5317|Loss: 0.9307997226715088:  82%|████████▏ | 5317/6470 [1:01:08<13:49,  1.39it/s]
4|5318|Loss: 0.7782829999923706:  82%|████████▏ | 5317/6470 [1:01:08<13:49,  1.39it/s]
4|5318|Loss: 0.7782829999923706:  82%|████████▏ | 5318/6470 [1:01:09<13:58,  1.37it/s]
4|5319|Loss: 0.7772274613380432:  82%|████████▏ | 5318/6470 [1:01:09<13:58,  1.37it/s]
4|5319|Loss: 0.7772274613380432:  82%|████████▏ | 5319/6470 [1:01:09<14:29,  1.32it/s]
4|5320|Loss: 0.7789367437362671:  82%|████████▏ | 5319/6470 [1:01:10<14:29,  1.32it/s]
4|5320|Loss: 0.7789367437362671:  82%|████████▏ | 5320/6470 [1:01:10<14:29,  1.32it/s]
4|5321|Loss: 0.86069655418396:  82%|████████▏ | 5320/6470 [1:01:11<14:29,  1.32it/s]  
4|5321|Loss: 0.86069655418396:  82%|████████▏ | 5321/6470 [1:01:11<14:22,  1.33it/s]
4|5322|Loss: 0.6756643652915955:  82%|████████▏ | 5321/6470 [1:01:11<14:22,  1.33it/s]
4|5322|Loss: 0.6756643652915955:  82%|████████▏ | 5322/6470 [1:01:11<13:14,  1.44it/s]
4|5323|Loss: 0.979109525680542:  82%|████████▏ | 5322/6470 [1:01:12<13:14,  1.44it/s] 
4|5323|Loss: 0.979109525680542:  82%|████████▏ | 5323/6470 [1:01:12<13:46,  1.39it/s]
4|5324|Loss: 0.8013931512832642:  82%|████████▏ | 5323/6470 [1:01:13<13:46,  1.39it/s]
4|5324|Loss: 0.8013931512832642:  82%|████████▏ | 5324/6470 [1:01:13<13:29,  1.41it/s]
4|5325|Loss: 0.7796511054039001:  82%|████████▏ | 5324/6470 [1:01:13<13:29,  1.41it/s]
4|5325|Loss: 0.7796511054039001:  82%|████████▏ | 5325/6470 [1:01:14<12:57,  1.47it/s]
4|5326|Loss: 0.8922280073165894:  82%|████████▏ | 5325/6470 [1:01:14<12:57,  1.47it/s]
4|5326|Loss: 0.8922280073165894:  82%|████████▏ | 5326/6470 [1:01:14<11:30,  1.66it/s]
4|5327|Loss: 0.9691025018692017:  82%|████████▏ | 5326/6470 [1:01:14<11:30,  1.66it/s]
4|5327|Loss: 0.9691025018692017:  82%|████████▏ | 5327/6470 [1:01:15<11:09,  1.71it/s]
4|5328|Loss: 0.8354679942131042:  82%|████████▏ | 5327/6470 [1:01:15<11:09,  1.71it/s]
4|5328|Loss: 0.8354679942131042:  82%|████████▏ | 5328/6470 [1:01:15<12:39,  1.50it/s]
4|5329|Loss: 0.886969804763794:  82%|████████▏ | 5328/6470 [1:01:16<12:39,  1.50it/s] 
4|5329|Loss: 0.886969804763794:  82%|████████▏ | 5329/6470 [1:01:16<12:50,  1.48it/s]
4|5330|Loss: 0.8860260248184204:  82%|████████▏ | 5329/6470 [1:01:16<12:50,  1.48it/s]
4|5330|Loss: 0.8860260248184204:  82%|████████▏ | 5330/6470 [1:01:17<13:04,  1.45it/s]
4|5331|Loss: 0.9704067707061768:  82%|████████▏ | 5330/6470 [1:01:17<13:04,  1.45it/s]
4|5331|Loss: 0.9704067707061768:  82%|████████▏ | 5331/6470 [1:01:17<13:11,  1.44it/s]
4|5332|Loss: 1.058279037475586:  82%|████████▏ | 5331/6470 [1:01:18<13:11,  1.44it/s] 
4|5332|Loss: 1.058279037475586:  82%|████████▏ | 5332/6470 [1:01:18<13:23,  1.42it/s]
4|5333|Loss: 0.8158504366874695:  82%|████████▏ | 5332/6470 [1:01:19<13:23,  1.42it/s]
4|5333|Loss: 0.8158504366874695:  82%|████████▏ | 5333/6470 [1:01:19<13:59,  1.36it/s]
4|5334|Loss: 0.9124143719673157:  82%|████████▏ | 5333/6470 [1:01:19<13:59,  1.36it/s]
4|5334|Loss: 0.9124143719673157:  82%|████████▏ | 5334/6470 [1:01:20<13:56,  1.36it/s]
4|5335|Loss: 0.8549036383628845:  82%|████████▏ | 5334/6470 [1:01:20<13:56,  1.36it/s]
4|5335|Loss: 0.8549036383628845:  82%|████████▏ | 5335/6470 [1:01:20<12:43,  1.49it/s]
4|5336|Loss: 0.9030700325965881:  82%|████████▏ | 5335/6470 [1:01:21<12:43,  1.49it/s]
4|5336|Loss: 0.9030700325965881:  82%|████████▏ | 5336/6470 [1:01:21<13:23,  1.41it/s]
4|5337|Loss: 0.8955391645431519:  82%|████████▏ | 5336/6470 [1:01:22<13:23,  1.41it/s]
4|5337|Loss: 0.8955391645431519:  82%|████████▏ | 5337/6470 [1:01:22<13:36,  1.39it/s]
4|5338|Loss: 0.9022790193557739:  82%|████████▏ | 5337/6470 [1:01:22<13:36,  1.39it/s]
4|5338|Loss: 0.9022790193557739:  83%|████████▎ | 5338/6470 [1:01:23<13:32,  1.39it/s]
4|5339|Loss: 0.9206547737121582:  83%|████████▎ | 5338/6470 [1:01:23<13:32,  1.39it/s]
4|5339|Loss: 0.9206547737121582:  83%|████████▎ | 5339/6470 [1:01:23<12:19,  1.53it/s]
4|5340|Loss: 0.7891648411750793:  83%|████████▎ | 5339/6470 [1:01:23<12:19,  1.53it/s]
4|5340|Loss: 0.7891648411750793:  83%|████████▎ | 5340/6470 [1:01:24<12:31,  1.50it/s]
4|5341|Loss: 0.8832612633705139:  83%|████████▎ | 5340/6470 [1:01:24<12:31,  1.50it/s]
4|5341|Loss: 0.8832612633705139:  83%|████████▎ | 5341/6470 [1:01:24<12:18,  1.53it/s]
4|5342|Loss: 0.8630808591842651:  83%|████████▎ | 5341/6470 [1:01:25<12:18,  1.53it/s]
4|5342|Loss: 0.8630808591842651:  83%|████████▎ | 5342/6470 [1:01:25<12:26,  1.51it/s]
4|5343|Loss: 0.6979808807373047:  83%|████████▎ | 5342/6470 [1:01:25<12:26,  1.51it/s]
4|5343|Loss: 0.6979808807373047:  83%|████████▎ | 5343/6470 [1:01:26<13:03,  1.44it/s]
4|5344|Loss: 1.0284638404846191:  83%|████████▎ | 5343/6470 [1:01:26<13:03,  1.44it/s]
4|5344|Loss: 1.0284638404846191:  83%|████████▎ | 5344/6470 [1:01:27<13:20,  1.41it/s]
4|5345|Loss: 0.8034330606460571:  83%|████████▎ | 5344/6470 [1:01:27<13:20,  1.41it/s]
4|5345|Loss: 0.8034330606460571:  83%|████████▎ | 5345/6470 [1:01:27<13:28,  1.39it/s]
4|5346|Loss: 0.9795219302177429:  83%|████████▎ | 5345/6470 [1:01:28<13:28,  1.39it/s]
4|5346|Loss: 0.9795219302177429:  83%|████████▎ | 5346/6470 [1:01:28<13:26,  1.39it/s]
4|5347|Loss: 0.8085363507270813:  83%|████████▎ | 5346/6470 [1:01:28<13:26,  1.39it/s]
4|5347|Loss: 0.8085363507270813:  83%|████████▎ | 5347/6470 [1:01:29<12:13,  1.53it/s]
4|5348|Loss: 0.8381254076957703:  83%|████████▎ | 5347/6470 [1:01:29<12:13,  1.53it/s]
4|5348|Loss: 0.8381254076957703:  83%|████████▎ | 5348/6470 [1:01:29<12:20,  1.51it/s]
4|5349|Loss: 1.0022492408752441:  83%|████████▎ | 5348/6470 [1:01:30<12:20,  1.51it/s]
4|5349|Loss: 1.0022492408752441:  83%|████████▎ | 5349/6470 [1:01:30<11:57,  1.56it/s]
4|5350|Loss: 0.8342162370681763:  83%|████████▎ | 5349/6470 [1:01:30<11:57,  1.56it/s]
4|5350|Loss: 0.8342162370681763:  83%|████████▎ | 5350/6470 [1:01:31<12:40,  1.47it/s]
4|5351|Loss: 0.9014940857887268:  83%|████████▎ | 5350/6470 [1:01:31<12:40,  1.47it/s]
4|5351|Loss: 0.9014940857887268:  83%|████████▎ | 5351/6470 [1:01:31<13:22,  1.39it/s]
4|5352|Loss: 0.883217453956604:  83%|████████▎ | 5351/6470 [1:01:32<13:22,  1.39it/s] 
4|5352|Loss: 0.883217453956604:  83%|████████▎ | 5352/6470 [1:01:32<12:34,  1.48it/s]
4|5353|Loss: 0.9227480292320251:  83%|████████▎ | 5352/6470 [1:01:32<12:34,  1.48it/s]
4|5353|Loss: 0.9227480292320251:  83%|████████▎ | 5353/6470 [1:01:33<13:16,  1.40it/s]
4|5354|Loss: 0.865113377571106:  83%|████████▎ | 5353/6470 [1:01:33<13:16,  1.40it/s] 
4|5354|Loss: 0.865113377571106:  83%|████████▎ | 5354/6470 [1:01:33<13:27,  1.38it/s]
4|5355|Loss: 0.8288686871528625:  83%|████████▎ | 5354/6470 [1:01:34<13:27,  1.38it/s]
4|5355|Loss: 0.8288686871528625:  83%|████████▎ | 5355/6470 [1:01:34<13:19,  1.39it/s]
4|5356|Loss: 0.7883799076080322:  83%|████████▎ | 5355/6470 [1:01:35<13:19,  1.39it/s]
4|5356|Loss: 0.7883799076080322:  83%|████████▎ | 5356/6470 [1:01:35<13:14,  1.40it/s]
4|5357|Loss: 0.8665569424629211:  83%|████████▎ | 5356/6470 [1:01:35<13:14,  1.40it/s]
4|5357|Loss: 0.8665569424629211:  83%|████████▎ | 5357/6470 [1:01:36<13:54,  1.33it/s]
4|5358|Loss: 0.8555825352668762:  83%|████████▎ | 5357/6470 [1:01:36<13:54,  1.33it/s]
4|5358|Loss: 0.8555825352668762:  83%|████████▎ | 5358/6470 [1:01:36<13:29,  1.37it/s]
4|5359|Loss: 1.0898802280426025:  83%|████████▎ | 5358/6470 [1:01:37<13:29,  1.37it/s]
4|5359|Loss: 1.0898802280426025:  83%|████████▎ | 5359/6470 [1:01:37<12:49,  1.44it/s]
4|5360|Loss: 0.9843154549598694:  83%|████████▎ | 5359/6470 [1:01:37<12:49,  1.44it/s]
4|5360|Loss: 0.9843154549598694:  83%|████████▎ | 5360/6470 [1:01:38<12:00,  1.54it/s]
4|5361|Loss: 0.8364855647087097:  83%|████████▎ | 5360/6470 [1:01:38<12:00,  1.54it/s]
4|5361|Loss: 0.8364855647087097:  83%|████████▎ | 5361/6470 [1:01:38<12:40,  1.46it/s]
4|5362|Loss: 0.8135276436805725:  83%|████████▎ | 5361/6470 [1:01:39<12:40,  1.46it/s]
4|5362|Loss: 0.8135276436805725:  83%|████████▎ | 5362/6470 [1:01:39<13:02,  1.42it/s]
4|5363|Loss: 0.7319884300231934:  83%|████████▎ | 5362/6470 [1:01:39<13:02,  1.42it/s]
4|5363|Loss: 0.7319884300231934:  83%|████████▎ | 5363/6470 [1:01:40<12:46,  1.44it/s]
4|5364|Loss: 0.745254635810852:  83%|████████▎ | 5363/6470 [1:01:40<12:46,  1.44it/s] 
4|5364|Loss: 0.745254635810852:  83%|████████▎ | 5364/6470 [1:01:40<12:45,  1.44it/s]
4|5365|Loss: 1.0119905471801758:  83%|████████▎ | 5364/6470 [1:01:41<12:45,  1.44it/s]
4|5365|Loss: 1.0119905471801758:  83%|████████▎ | 5365/6470 [1:01:41<12:48,  1.44it/s]
4|5366|Loss: 0.7919880747795105:  83%|████████▎ | 5365/6470 [1:01:42<12:48,  1.44it/s]
4|5366|Loss: 0.7919880747795105:  83%|████████▎ | 5366/6470 [1:01:42<12:16,  1.50it/s]
4|5367|Loss: 0.7629648447036743:  83%|████████▎ | 5366/6470 [1:01:42<12:16,  1.50it/s]
4|5367|Loss: 0.7629648447036743:  83%|████████▎ | 5367/6470 [1:01:42<11:22,  1.62it/s]
4|5368|Loss: 0.727993905544281:  83%|████████▎ | 5367/6470 [1:01:43<11:22,  1.62it/s] 
4|5368|Loss: 0.727993905544281:  83%|████████▎ | 5368/6470 [1:01:43<11:12,  1.64it/s]
4|5369|Loss: 0.9013887643814087:  83%|████████▎ | 5368/6470 [1:01:43<11:12,  1.64it/s]
4|5369|Loss: 0.9013887643814087:  83%|████████▎ | 5369/6470 [1:01:44<11:45,  1.56it/s]
4|5370|Loss: 0.9043580889701843:  83%|████████▎ | 5369/6470 [1:01:44<11:45,  1.56it/s]
4|5370|Loss: 0.9043580889701843:  83%|████████▎ | 5370/6470 [1:01:44<11:01,  1.66it/s]
4|5371|Loss: 0.6145941615104675:  83%|████████▎ | 5370/6470 [1:01:44<11:01,  1.66it/s]
4|5371|Loss: 0.6145941615104675:  83%|████████▎ | 5371/6470 [1:01:45<12:00,  1.53it/s]
4|5372|Loss: 0.8691035509109497:  83%|████████▎ | 5371/6470 [1:01:45<12:00,  1.53it/s]
4|5372|Loss: 0.8691035509109497:  83%|████████▎ | 5372/6470 [1:01:46<13:02,  1.40it/s]
4|5373|Loss: 0.8859005570411682:  83%|████████▎ | 5372/6470 [1:01:46<13:02,  1.40it/s]
4|5373|Loss: 0.8859005570411682:  83%|████████▎ | 5373/6470 [1:01:46<12:46,  1.43it/s]
4|5374|Loss: 0.7862345576286316:  83%|████████▎ | 5373/6470 [1:01:47<12:46,  1.43it/s]
4|5374|Loss: 0.7862345576286316:  83%|████████▎ | 5374/6470 [1:01:47<11:33,  1.58it/s]
4|5375|Loss: 0.8773336410522461:  83%|████████▎ | 5374/6470 [1:01:47<11:33,  1.58it/s]
4|5375|Loss: 0.8773336410522461:  83%|████████▎ | 5375/6470 [1:01:48<11:49,  1.54it/s]
4|5376|Loss: 0.8959782719612122:  83%|████████▎ | 5375/6470 [1:01:48<11:49,  1.54it/s]
4|5376|Loss: 0.8959782719612122:  83%|████████▎ | 5376/6470 [1:01:48<11:50,  1.54it/s]
4|5377|Loss: 0.7790184617042542:  83%|████████▎ | 5376/6470 [1:01:49<11:50,  1.54it/s]
4|5377|Loss: 0.7790184617042542:  83%|████████▎ | 5377/6470 [1:01:49<11:43,  1.55it/s]
4|5378|Loss: 0.9960232377052307:  83%|████████▎ | 5377/6470 [1:01:49<11:43,  1.55it/s]
4|5378|Loss: 0.9960232377052307:  83%|████████▎ | 5378/6470 [1:01:49<11:42,  1.55it/s]
4|5379|Loss: 0.8457274436950684:  83%|████████▎ | 5378/6470 [1:01:50<11:42,  1.55it/s]
4|5379|Loss: 0.8457274436950684:  83%|████████▎ | 5379/6470 [1:01:50<12:20,  1.47it/s]
4|5380|Loss: 0.9167494773864746:  83%|████████▎ | 5379/6470 [1:01:51<12:20,  1.47it/s]
4|5380|Loss: 0.9167494773864746:  83%|████████▎ | 5380/6470 [1:01:51<12:47,  1.42it/s]
4|5381|Loss: 0.8764505982398987:  83%|████████▎ | 5380/6470 [1:01:51<12:47,  1.42it/s]
4|5381|Loss: 0.8764505982398987:  83%|████████▎ | 5381/6470 [1:01:52<13:24,  1.35it/s]
4|5382|Loss: 0.9057294130325317:  83%|████████▎ | 5381/6470 [1:01:52<13:24,  1.35it/s]
4|5382|Loss: 0.9057294130325317:  83%|████████▎ | 5382/6470 [1:01:52<13:06,  1.38it/s]
4|5383|Loss: 0.842068612575531:  83%|████████▎ | 5382/6470 [1:01:53<13:06,  1.38it/s] 
4|5383|Loss: 0.842068612575531:  83%|████████▎ | 5383/6470 [1:01:53<13:02,  1.39it/s]
4|5384|Loss: 0.72841477394104:  83%|████████▎ | 5383/6470 [1:01:54<13:02,  1.39it/s] 
4|5384|Loss: 0.72841477394104:  83%|████████▎ | 5384/6470 [1:01:54<12:36,  1.44it/s]
4|5385|Loss: 0.8467239141464233:  83%|████████▎ | 5384/6470 [1:01:54<12:36,  1.44it/s]
4|5385|Loss: 0.8467239141464233:  83%|████████▎ | 5385/6470 [1:01:55<12:46,  1.42it/s]
4|5386|Loss: 0.8684961795806885:  83%|████████▎ | 5385/6470 [1:01:55<12:46,  1.42it/s]
4|5386|Loss: 0.8684961795806885:  83%|████████▎ | 5386/6470 [1:01:55<13:16,  1.36it/s]
4|5387|Loss: 0.9052664637565613:  83%|████████▎ | 5386/6470 [1:01:56<13:16,  1.36it/s]
4|5387|Loss: 0.9052664637565613:  83%|████████▎ | 5387/6470 [1:01:56<12:58,  1.39it/s]
4|5388|Loss: 0.7092805504798889:  83%|████████▎ | 5387/6470 [1:01:56<12:58,  1.39it/s]
4|5388|Loss: 0.7092805504798889:  83%|████████▎ | 5388/6470 [1:01:57<12:27,  1.45it/s]
4|5389|Loss: 0.6537929177284241:  83%|████████▎ | 5388/6470 [1:01:57<12:27,  1.45it/s]
4|5389|Loss: 0.6537929177284241:  83%|████████▎ | 5389/6470 [1:01:57<11:52,  1.52it/s]
4|5390|Loss: 0.8336069583892822:  83%|████████▎ | 5389/6470 [1:01:58<11:52,  1.52it/s]
4|5390|Loss: 0.8336069583892822:  83%|████████▎ | 5390/6470 [1:01:58<12:07,  1.49it/s]
4|5391|Loss: 0.7986586689949036:  83%|████████▎ | 5390/6470 [1:01:58<12:07,  1.49it/s]
4|5391|Loss: 0.7986586689949036:  83%|████████▎ | 5391/6470 [1:01:59<12:03,  1.49it/s]
4|5392|Loss: 0.8213573694229126:  83%|████████▎ | 5391/6470 [1:01:59<12:03,  1.49it/s]
4|5392|Loss: 0.8213573694229126:  83%|████████▎ | 5392/6470 [1:01:59<12:16,  1.46it/s]
4|5393|Loss: 0.9505894780158997:  83%|████████▎ | 5392/6470 [1:02:00<12:16,  1.46it/s]
4|5393|Loss: 0.9505894780158997:  83%|████████▎ | 5393/6470 [1:02:00<11:45,  1.53it/s]
4|5394|Loss: 0.758443295955658:  83%|████████▎ | 5393/6470 [1:02:00<11:45,  1.53it/s] 
4|5394|Loss: 0.758443295955658:  83%|████████▎ | 5394/6470 [1:02:01<11:50,  1.51it/s]
4|5395|Loss: 0.9293937683105469:  83%|████████▎ | 5394/6470 [1:02:01<11:50,  1.51it/s]
4|5395|Loss: 0.9293937683105469:  83%|████████▎ | 5395/6470 [1:02:01<12:19,  1.45it/s]
4|5396|Loss: 0.9163896441459656:  83%|████████▎ | 5395/6470 [1:02:02<12:19,  1.45it/s]
4|5396|Loss: 0.9163896441459656:  83%|████████▎ | 5396/6470 [1:02:02<11:32,  1.55it/s]
4|5397|Loss: 0.8108393549919128:  83%|████████▎ | 5396/6470 [1:02:02<11:32,  1.55it/s]
4|5397|Loss: 0.8108393549919128:  83%|████████▎ | 5397/6470 [1:02:03<11:42,  1.53it/s]
4|5398|Loss: 0.9758893251419067:  83%|████████▎ | 5397/6470 [1:02:03<11:42,  1.53it/s]
4|5398|Loss: 0.9758893251419067:  83%|████████▎ | 5398/6470 [1:02:03<12:10,  1.47it/s]
4|5399|Loss: 0.8879103660583496:  83%|████████▎ | 5398/6470 [1:02:04<12:10,  1.47it/s]
4|5399|Loss: 0.8879103660583496:  83%|████████▎ | 5399/6470 [1:02:04<12:04,  1.48it/s]
4|5400|Loss: 0.8412831425666809:  83%|████████▎ | 5399/6470 [1:02:04<12:04,  1.48it/s]
4|5400|Loss: 0.8412831425666809:  83%|████████▎ | 5400/6470 [1:02:05<11:50,  1.51it/s]
4|5401|Loss: 0.9392765164375305:  83%|████████▎ | 5400/6470 [1:02:05<11:50,  1.51it/s]
4|5401|Loss: 0.9392765164375305:  83%|████████▎ | 5401/6470 [1:02:05<11:41,  1.52it/s]
4|5402|Loss: 0.9344819784164429:  83%|████████▎ | 5401/6470 [1:02:06<11:41,  1.52it/s]
4|5402|Loss: 0.9344819784164429:  83%|████████▎ | 5402/6470 [1:02:06<11:07,  1.60it/s]
4|5403|Loss: 0.929360568523407:  83%|████████▎ | 5402/6470 [1:02:06<11:07,  1.60it/s] 
4|5403|Loss: 0.929360568523407:  84%|████████▎ | 5403/6470 [1:02:07<12:07,  1.47it/s]
4|5404|Loss: 0.8407564163208008:  84%|████████▎ | 5403/6470 [1:02:07<12:07,  1.47it/s]
4|5404|Loss: 0.8407564163208008:  84%|████████▎ | 5404/6470 [1:02:07<13:05,  1.36it/s]
4|5405|Loss: 0.80021733045578:  84%|████████▎ | 5404/6470 [1:02:08<13:05,  1.36it/s]  
4|5405|Loss: 0.80021733045578:  84%|████████▎ | 5405/6470 [1:02:08<13:17,  1.33it/s]
4|5406|Loss: 0.919126570224762:  84%|████████▎ | 5405/6470 [1:02:09<13:17,  1.33it/s]
4|5406|Loss: 0.919126570224762:  84%|████████▎ | 5406/6470 [1:02:09<12:46,  1.39it/s]
4|5407|Loss: 0.9287687540054321:  84%|████████▎ | 5406/6470 [1:02:09<12:46,  1.39it/s]
4|5407|Loss: 0.9287687540054321:  84%|████████▎ | 5407/6470 [1:02:10<13:12,  1.34it/s]
4|5408|Loss: 0.9730002880096436:  84%|████████▎ | 5407/6470 [1:02:10<13:12,  1.34it/s]
4|5408|Loss: 0.9730002880096436:  84%|████████▎ | 5408/6470 [1:02:11<13:55,  1.27it/s]
4|5409|Loss: 0.9539046883583069:  84%|████████▎ | 5408/6470 [1:02:11<13:55,  1.27it/s]
4|5409|Loss: 0.9539046883583069:  84%|████████▎ | 5409/6470 [1:02:11<14:22,  1.23it/s]
4|5410|Loss: 0.8997607827186584:  84%|████████▎ | 5409/6470 [1:02:12<14:22,  1.23it/s]
4|5410|Loss: 0.8997607827186584:  84%|████████▎ | 5410/6470 [1:02:12<14:09,  1.25it/s]
4|5411|Loss: 1.0172529220581055:  84%|████████▎ | 5410/6470 [1:02:13<14:09,  1.25it/s]
4|5411|Loss: 1.0172529220581055:  84%|████████▎ | 5411/6470 [1:02:13<14:20,  1.23it/s]
4|5412|Loss: 0.7576919794082642:  84%|████████▎ | 5411/6470 [1:02:13<14:20,  1.23it/s]
4|5412|Loss: 0.7576919794082642:  84%|████████▎ | 5412/6470 [1:02:14<13:18,  1.33it/s]
4|5413|Loss: 0.9038805365562439:  84%|████████▎ | 5412/6470 [1:02:14<13:18,  1.33it/s]
4|5413|Loss: 0.9038805365562439:  84%|████████▎ | 5413/6470 [1:02:14<12:20,  1.43it/s]
4|5414|Loss: 0.9429956078529358:  84%|████████▎ | 5413/6470 [1:02:15<12:20,  1.43it/s]
4|5414|Loss: 0.9429956078529358:  84%|████████▎ | 5414/6470 [1:02:15<12:53,  1.36it/s]
4|5415|Loss: 0.8021408915519714:  84%|████████▎ | 5414/6470 [1:02:16<12:53,  1.36it/s]
4|5415|Loss: 0.8021408915519714:  84%|████████▎ | 5415/6470 [1:02:16<13:02,  1.35it/s]
4|5416|Loss: 0.8141990900039673:  84%|████████▎ | 5415/6470 [1:02:16<13:02,  1.35it/s]
4|5416|Loss: 0.8141990900039673:  84%|████████▎ | 5416/6470 [1:02:17<12:58,  1.35it/s]
4|5417|Loss: 0.7478246092796326:  84%|████████▎ | 5416/6470 [1:02:17<12:58,  1.35it/s]
4|5417|Loss: 0.7478246092796326:  84%|████████▎ | 5417/6470 [1:02:17<13:10,  1.33it/s]
4|5418|Loss: 0.7776753902435303:  84%|████████▎ | 5417/6470 [1:02:18<13:10,  1.33it/s]
4|5418|Loss: 0.7776753902435303:  84%|████████▎ | 5418/6470 [1:02:18<13:09,  1.33it/s]
4|5419|Loss: 0.8254616856575012:  84%|████████▎ | 5418/6470 [1:02:18<13:09,  1.33it/s]
4|5419|Loss: 0.8254616856575012:  84%|████████▍ | 5419/6470 [1:02:19<12:18,  1.42it/s]
4|5420|Loss: 0.8389224410057068:  84%|████████▍ | 5419/6470 [1:02:19<12:18,  1.42it/s]
4|5420|Loss: 0.8389224410057068:  84%|████████▍ | 5420/6470 [1:02:20<13:15,  1.32it/s]
4|5421|Loss: 0.8414849638938904:  84%|████████▍ | 5420/6470 [1:02:20<13:15,  1.32it/s]
4|5421|Loss: 0.8414849638938904:  84%|████████▍ | 5421/6470 [1:02:20<12:27,  1.40it/s]
4|5422|Loss: 0.7279184460639954:  84%|████████▍ | 5421/6470 [1:02:21<12:27,  1.40it/s]
4|5422|Loss: 0.7279184460639954:  84%|████████▍ | 5422/6470 [1:02:21<11:42,  1.49it/s]
4|5423|Loss: 0.8141688108444214:  84%|████████▍ | 5422/6470 [1:02:21<11:42,  1.49it/s]
4|5423|Loss: 0.8141688108444214:  84%|████████▍ | 5423/6470 [1:02:21<11:57,  1.46it/s]
4|5424|Loss: 0.8309804201126099:  84%|████████▍ | 5423/6470 [1:02:22<11:57,  1.46it/s]
4|5424|Loss: 0.8309804201126099:  84%|████████▍ | 5424/6470 [1:02:22<12:14,  1.42it/s]
4|5425|Loss: 0.8040080070495605:  84%|████████▍ | 5424/6470 [1:02:23<12:14,  1.42it/s]
4|5425|Loss: 0.8040080070495605:  84%|████████▍ | 5425/6470 [1:02:23<11:28,  1.52it/s]
4|5426|Loss: 0.8307139277458191:  84%|████████▍ | 5425/6470 [1:02:23<11:28,  1.52it/s]
4|5426|Loss: 0.8307139277458191:  84%|████████▍ | 5426/6470 [1:02:23<11:09,  1.56it/s]
4|5427|Loss: 0.7677178382873535:  84%|████████▍ | 5426/6470 [1:02:24<11:09,  1.56it/s]
4|5427|Loss: 0.7677178382873535:  84%|████████▍ | 5427/6470 [1:02:24<12:00,  1.45it/s]
4|5428|Loss: 0.7920476794242859:  84%|████████▍ | 5427/6470 [1:02:25<12:00,  1.45it/s]
4|5428|Loss: 0.7920476794242859:  84%|████████▍ | 5428/6470 [1:02:25<12:24,  1.40it/s]
4|5429|Loss: 0.9424245357513428:  84%|████████▍ | 5428/6470 [1:02:25<12:24,  1.40it/s]
4|5429|Loss: 0.9424245357513428:  84%|████████▍ | 5429/6470 [1:02:26<12:24,  1.40it/s]
4|5430|Loss: 0.8486069440841675:  84%|████████▍ | 5429/6470 [1:02:26<12:24,  1.40it/s]
4|5430|Loss: 0.8486069440841675:  84%|████████▍ | 5430/6470 [1:02:26<11:46,  1.47it/s]
4|5431|Loss: 0.7971507906913757:  84%|████████▍ | 5430/6470 [1:02:27<11:46,  1.47it/s]
4|5431|Loss: 0.7971507906913757:  84%|████████▍ | 5431/6470 [1:02:27<11:30,  1.50it/s]
4|5432|Loss: 0.8223437666893005:  84%|████████▍ | 5431/6470 [1:02:27<11:30,  1.50it/s]
4|5432|Loss: 0.8223437666893005:  84%|████████▍ | 5432/6470 [1:02:28<11:46,  1.47it/s]
4|5433|Loss: 0.7978593111038208:  84%|████████▍ | 5432/6470 [1:02:28<11:46,  1.47it/s]
4|5433|Loss: 0.7978593111038208:  84%|████████▍ | 5433/6470 [1:02:28<10:51,  1.59it/s]
4|5434|Loss: 0.9185479283332825:  84%|████████▍ | 5433/6470 [1:02:28<10:51,  1.59it/s]
4|5434|Loss: 0.9185479283332825:  84%|████████▍ | 5434/6470 [1:02:29<10:59,  1.57it/s]
4|5435|Loss: 0.7733719348907471:  84%|████████▍ | 5434/6470 [1:02:29<10:59,  1.57it/s]
4|5435|Loss: 0.7733719348907471:  84%|████████▍ | 5435/6470 [1:02:30<11:58,  1.44it/s]
4|5436|Loss: 0.917096734046936:  84%|████████▍ | 5435/6470 [1:02:30<11:58,  1.44it/s] 
4|5436|Loss: 0.917096734046936:  84%|████████▍ | 5436/6470 [1:02:30<12:49,  1.34it/s]
4|5437|Loss: 0.7299826145172119:  84%|████████▍ | 5436/6470 [1:02:31<12:49,  1.34it/s]
4|5437|Loss: 0.7299826145172119:  84%|████████▍ | 5437/6470 [1:02:31<12:18,  1.40it/s]
4|5438|Loss: 0.7016807198524475:  84%|████████▍ | 5437/6470 [1:02:31<12:18,  1.40it/s]
4|5438|Loss: 0.7016807198524475:  84%|████████▍ | 5438/6470 [1:02:32<10:42,  1.61it/s]
4|5439|Loss: 0.9624154567718506:  84%|████████▍ | 5438/6470 [1:02:32<10:42,  1.61it/s]
4|5439|Loss: 0.9624154567718506:  84%|████████▍ | 5439/6470 [1:02:32<11:19,  1.52it/s]
4|5440|Loss: 0.6546148061752319:  84%|████████▍ | 5439/6470 [1:02:33<11:19,  1.52it/s]
4|5440|Loss: 0.6546148061752319:  84%|████████▍ | 5440/6470 [1:02:33<12:20,  1.39it/s]
4|5441|Loss: 0.8355376720428467:  84%|████████▍ | 5440/6470 [1:02:34<12:20,  1.39it/s]
4|5441|Loss: 0.8355376720428467:  84%|████████▍ | 5441/6470 [1:02:34<12:05,  1.42it/s]
4|5442|Loss: 0.8538981080055237:  84%|████████▍ | 5441/6470 [1:02:34<12:05,  1.42it/s]
4|5442|Loss: 0.8538981080055237:  84%|████████▍ | 5442/6470 [1:02:35<12:06,  1.41it/s]
4|5443|Loss: 0.8787450790405273:  84%|████████▍ | 5442/6470 [1:02:35<12:06,  1.41it/s]
4|5443|Loss: 0.8787450790405273:  84%|████████▍ | 5443/6470 [1:02:35<11:47,  1.45it/s]
4|5444|Loss: 0.8218817710876465:  84%|████████▍ | 5443/6470 [1:02:36<11:47,  1.45it/s]
4|5444|Loss: 0.8218817710876465:  84%|████████▍ | 5444/6470 [1:02:36<11:26,  1.50it/s]
4|5445|Loss: 0.7873104810714722:  84%|████████▍ | 5444/6470 [1:02:36<11:26,  1.50it/s]
4|5445|Loss: 0.7873104810714722:  84%|████████▍ | 5445/6470 [1:02:37<12:23,  1.38it/s]
4|5446|Loss: 0.8412153124809265:  84%|████████▍ | 5445/6470 [1:02:37<12:23,  1.38it/s]
4|5446|Loss: 0.8412153124809265:  84%|████████▍ | 5446/6470 [1:02:37<12:54,  1.32it/s]
4|5447|Loss: 0.8810749053955078:  84%|████████▍ | 5446/6470 [1:02:38<12:54,  1.32it/s]
4|5447|Loss: 0.8810749053955078:  84%|████████▍ | 5447/6470 [1:02:38<11:21,  1.50it/s]
4|5448|Loss: 0.8721242547035217:  84%|████████▍ | 5447/6470 [1:02:38<11:21,  1.50it/s]
4|5448|Loss: 0.8721242547035217:  84%|████████▍ | 5448/6470 [1:02:39<11:57,  1.42it/s]
4|5449|Loss: 0.8148045539855957:  84%|████████▍ | 5448/6470 [1:02:39<11:57,  1.42it/s]
4|5449|Loss: 0.8148045539855957:  84%|████████▍ | 5449/6470 [1:02:39<11:56,  1.42it/s]
4|5450|Loss: 0.7985900640487671:  84%|████████▍ | 5449/6470 [1:02:40<11:56,  1.42it/s]
4|5450|Loss: 0.7985900640487671:  84%|████████▍ | 5450/6470 [1:02:40<10:35,  1.60it/s]
4|5451|Loss: 0.7681195735931396:  84%|████████▍ | 5450/6470 [1:02:40<10:35,  1.60it/s]
4|5451|Loss: 0.7681195735931396:  84%|████████▍ | 5451/6470 [1:02:40<10:23,  1.63it/s]
4|5452|Loss: 0.8409696817398071:  84%|████████▍ | 5451/6470 [1:02:41<10:23,  1.63it/s]
4|5452|Loss: 0.8409696817398071:  84%|████████▍ | 5452/6470 [1:02:42<12:46,  1.33it/s]
4|5453|Loss: 0.824584424495697:  84%|████████▍ | 5452/6470 [1:02:42<12:46,  1.33it/s] 
4|5453|Loss: 0.824584424495697:  84%|████████▍ | 5453/6470 [1:02:42<12:49,  1.32it/s]
4|5454|Loss: 0.8163696527481079:  84%|████████▍ | 5453/6470 [1:02:43<12:49,  1.32it/s]
4|5454|Loss: 0.8163696527481079:  84%|████████▍ | 5454/6470 [1:02:43<11:51,  1.43it/s]
4|5455|Loss: 0.9103495478630066:  84%|████████▍ | 5454/6470 [1:02:43<11:51,  1.43it/s]
4|5455|Loss: 0.9103495478630066:  84%|████████▍ | 5455/6470 [1:02:43<11:28,  1.47it/s]
4|5456|Loss: 0.8749591112136841:  84%|████████▍ | 5455/6470 [1:02:44<11:28,  1.47it/s]
4|5456|Loss: 0.8749591112136841:  84%|████████▍ | 5456/6470 [1:02:44<12:02,  1.40it/s]
4|5457|Loss: 0.8574510216712952:  84%|████████▍ | 5456/6470 [1:02:45<12:02,  1.40it/s]
4|5457|Loss: 0.8574510216712952:  84%|████████▍ | 5457/6470 [1:02:45<12:42,  1.33it/s]
4|5458|Loss: 0.8018299341201782:  84%|████████▍ | 5457/6470 [1:02:45<12:42,  1.33it/s]
4|5458|Loss: 0.8018299341201782:  84%|████████▍ | 5458/6470 [1:02:46<11:20,  1.49it/s]
4|5459|Loss: 0.8659040927886963:  84%|████████▍ | 5458/6470 [1:02:46<11:20,  1.49it/s]
4|5459|Loss: 0.8659040927886963:  84%|████████▍ | 5459/6470 [1:02:46<11:52,  1.42it/s]
4|5460|Loss: 0.9105167388916016:  84%|████████▍ | 5459/6470 [1:02:47<11:52,  1.42it/s]
4|5460|Loss: 0.9105167388916016:  84%|████████▍ | 5460/6470 [1:02:47<11:48,  1.43it/s]
4|5461|Loss: 0.7817488312721252:  84%|████████▍ | 5460/6470 [1:02:47<11:48,  1.43it/s]
4|5461|Loss: 0.7817488312721252:  84%|████████▍ | 5461/6470 [1:02:48<11:25,  1.47it/s]
4|5462|Loss: 0.7640607357025146:  84%|████████▍ | 5461/6470 [1:02:48<11:25,  1.47it/s]
4|5462|Loss: 0.7640607357025146:  84%|████████▍ | 5462/6470 [1:02:48<11:49,  1.42it/s]
4|5463|Loss: 0.9095670580863953:  84%|████████▍ | 5462/6470 [1:02:49<11:49,  1.42it/s]
4|5463|Loss: 0.9095670580863953:  84%|████████▍ | 5463/6470 [1:02:49<11:27,  1.46it/s]
4|5464|Loss: 0.7893204689025879:  84%|████████▍ | 5463/6470 [1:02:49<11:27,  1.46it/s]
4|5464|Loss: 0.7893204689025879:  84%|████████▍ | 5464/6470 [1:02:50<11:30,  1.46it/s]
4|5465|Loss: 0.721819281578064:  84%|████████▍ | 5464/6470 [1:02:50<11:30,  1.46it/s] 
4|5465|Loss: 0.721819281578064:  84%|████████▍ | 5465/6470 [1:02:51<12:37,  1.33it/s]
4|5466|Loss: 0.8805919885635376:  84%|████████▍ | 5465/6470 [1:02:51<12:37,  1.33it/s]
4|5466|Loss: 0.8805919885635376:  84%|████████▍ | 5466/6470 [1:02:51<12:45,  1.31it/s]
4|5467|Loss: 1.0618362426757812:  84%|████████▍ | 5466/6470 [1:02:52<12:45,  1.31it/s]
4|5467|Loss: 1.0618362426757812:  84%|████████▍ | 5467/6470 [1:02:52<12:29,  1.34it/s]
4|5468|Loss: 0.9615338444709778:  84%|████████▍ | 5467/6470 [1:02:53<12:29,  1.34it/s]
4|5468|Loss: 0.9615338444709778:  85%|████████▍ | 5468/6470 [1:02:53<12:00,  1.39it/s]
4|5469|Loss: 0.7217934727668762:  85%|████████▍ | 5468/6470 [1:02:53<12:00,  1.39it/s]
4|5469|Loss: 0.7217934727668762:  85%|████████▍ | 5469/6470 [1:02:54<11:50,  1.41it/s]
4|5470|Loss: 0.7630050778388977:  85%|████████▍ | 5469/6470 [1:02:54<11:50,  1.41it/s]
4|5470|Loss: 0.7630050778388977:  85%|████████▍ | 5470/6470 [1:02:54<11:18,  1.47it/s]
4|5471|Loss: 0.8570151329040527:  85%|████████▍ | 5470/6470 [1:02:55<11:18,  1.47it/s]
4|5471|Loss: 0.8570151329040527:  85%|████████▍ | 5471/6470 [1:02:55<11:24,  1.46it/s]
4|5472|Loss: 0.8062229156494141:  85%|████████▍ | 5471/6470 [1:02:55<11:24,  1.46it/s]
4|5472|Loss: 0.8062229156494141:  85%|████████▍ | 5472/6470 [1:02:56<11:37,  1.43it/s]
4|5473|Loss: 0.8239633440971375:  85%|████████▍ | 5472/6470 [1:02:56<11:37,  1.43it/s]
4|5473|Loss: 0.8239633440971375:  85%|████████▍ | 5473/6470 [1:02:56<11:19,  1.47it/s]
4|5474|Loss: 0.8458354473114014:  85%|████████▍ | 5473/6470 [1:02:57<11:19,  1.47it/s]
4|5474|Loss: 0.8458354473114014:  85%|████████▍ | 5474/6470 [1:02:57<11:48,  1.41it/s]
4|5475|Loss: 0.7451619505882263:  85%|████████▍ | 5474/6470 [1:02:57<11:48,  1.41it/s]
4|5475|Loss: 0.7451619505882263:  85%|████████▍ | 5475/6470 [1:02:58<11:53,  1.40it/s]
4|5476|Loss: 0.7465468049049377:  85%|████████▍ | 5475/6470 [1:02:58<11:53,  1.40it/s]
4|5476|Loss: 0.7465468049049377:  85%|████████▍ | 5476/6470 [1:02:58<11:33,  1.43it/s]
4|5477|Loss: 0.9718735218048096:  85%|████████▍ | 5476/6470 [1:02:59<11:33,  1.43it/s]
4|5477|Loss: 0.9718735218048096:  85%|████████▍ | 5477/6470 [1:02:59<11:10,  1.48it/s]
4|5478|Loss: 0.8813289403915405:  85%|████████▍ | 5477/6470 [1:02:59<11:10,  1.48it/s]
4|5478|Loss: 0.8813289403915405:  85%|████████▍ | 5478/6470 [1:03:00<11:14,  1.47it/s]
4|5479|Loss: 0.839292049407959:  85%|████████▍ | 5478/6470 [1:03:00<11:14,  1.47it/s] 
4|5479|Loss: 0.839292049407959:  85%|████████▍ | 5479/6470 [1:03:00<11:00,  1.50it/s]
4|5480|Loss: 0.8155122399330139:  85%|████████▍ | 5479/6470 [1:03:01<11:00,  1.50it/s]
4|5480|Loss: 0.8155122399330139:  85%|████████▍ | 5480/6470 [1:03:01<11:12,  1.47it/s]
4|5481|Loss: 0.8119028806686401:  85%|████████▍ | 5480/6470 [1:03:01<11:12,  1.47it/s]
4|5481|Loss: 0.8119028806686401:  85%|████████▍ | 5481/6470 [1:03:02<11:32,  1.43it/s]
4|5482|Loss: 0.8304290175437927:  85%|████████▍ | 5481/6470 [1:03:02<11:32,  1.43it/s]
4|5482|Loss: 0.8304290175437927:  85%|████████▍ | 5482/6470 [1:03:03<11:35,  1.42it/s]
4|5483|Loss: 0.8344739079475403:  85%|████████▍ | 5482/6470 [1:03:03<11:35,  1.42it/s]
4|5483|Loss: 0.8344739079475403:  85%|████████▍ | 5483/6470 [1:03:03<11:19,  1.45it/s]
4|5484|Loss: 0.8029978275299072:  85%|████████▍ | 5483/6470 [1:03:04<11:19,  1.45it/s]
4|5484|Loss: 0.8029978275299072:  85%|████████▍ | 5484/6470 [1:03:04<12:10,  1.35it/s]
4|5485|Loss: 0.7156702876091003:  85%|████████▍ | 5484/6470 [1:03:04<12:10,  1.35it/s]
4|5485|Loss: 0.7156702876091003:  85%|████████▍ | 5485/6470 [1:03:05<11:53,  1.38it/s]
4|5486|Loss: 0.9484162926673889:  85%|████████▍ | 5485/6470 [1:03:05<11:53,  1.38it/s]
4|5486|Loss: 0.9484162926673889:  85%|████████▍ | 5486/6470 [1:03:06<12:13,  1.34it/s]
4|5487|Loss: 0.7363654971122742:  85%|████████▍ | 5486/6470 [1:03:06<12:13,  1.34it/s]
4|5487|Loss: 0.7363654971122742:  85%|████████▍ | 5487/6470 [1:03:06<12:29,  1.31it/s]
4|5488|Loss: 0.794741153717041:  85%|████████▍ | 5487/6470 [1:03:07<12:29,  1.31it/s] 
4|5488|Loss: 0.794741153717041:  85%|████████▍ | 5488/6470 [1:03:07<12:17,  1.33it/s]
4|5489|Loss: 0.7792112231254578:  85%|████████▍ | 5488/6470 [1:03:07<12:17,  1.33it/s]
4|5489|Loss: 0.7792112231254578:  85%|████████▍ | 5489/6470 [1:03:08<11:41,  1.40it/s]
4|5490|Loss: 0.8357272148132324:  85%|████████▍ | 5489/6470 [1:03:08<11:41,  1.40it/s]
4|5490|Loss: 0.8357272148132324:  85%|████████▍ | 5490/6470 [1:03:08<11:15,  1.45it/s]
4|5491|Loss: 0.8437575697898865:  85%|████████▍ | 5490/6470 [1:03:09<11:15,  1.45it/s]
4|5491|Loss: 0.8437575697898865:  85%|████████▍ | 5491/6470 [1:03:09<11:23,  1.43it/s]
4|5492|Loss: 0.8451677560806274:  85%|████████▍ | 5491/6470 [1:03:09<11:23,  1.43it/s]
4|5492|Loss: 0.8451677560806274:  85%|████████▍ | 5492/6470 [1:03:10<11:06,  1.47it/s]
4|5493|Loss: 0.689678430557251:  85%|████████▍ | 5492/6470 [1:03:10<11:06,  1.47it/s] 
4|5493|Loss: 0.689678430557251:  85%|████████▍ | 5493/6470 [1:03:10<11:12,  1.45it/s]
4|5494|Loss: 0.8993496894836426:  85%|████████▍ | 5493/6470 [1:03:11<11:12,  1.45it/s]
4|5494|Loss: 0.8993496894836426:  85%|████████▍ | 5494/6470 [1:03:11<11:03,  1.47it/s]
4|5495|Loss: 0.9582251310348511:  85%|████████▍ | 5494/6470 [1:03:11<11:03,  1.47it/s]
4|5495|Loss: 0.9582251310348511:  85%|████████▍ | 5495/6470 [1:03:12<11:18,  1.44it/s]
4|5496|Loss: 0.8201049566268921:  85%|████████▍ | 5495/6470 [1:03:12<11:18,  1.44it/s]
4|5496|Loss: 0.8201049566268921:  85%|████████▍ | 5496/6470 [1:03:13<11:54,  1.36it/s]
4|5497|Loss: 0.7724302411079407:  85%|████████▍ | 5496/6470 [1:03:13<11:54,  1.36it/s]
4|5497|Loss: 0.7724302411079407:  85%|████████▍ | 5497/6470 [1:03:13<12:04,  1.34it/s]
4|5498|Loss: 0.8467398881912231:  85%|████████▍ | 5497/6470 [1:03:14<12:04,  1.34it/s]
4|5498|Loss: 0.8467398881912231:  85%|████████▍ | 5498/6470 [1:03:14<11:41,  1.39it/s]
4|5499|Loss: 1.0307682752609253:  85%|████████▍ | 5498/6470 [1:03:14<11:41,  1.39it/s]
4|5499|Loss: 1.0307682752609253:  85%|████████▍ | 5499/6470 [1:03:15<11:39,  1.39it/s]
4|5500|Loss: 0.7346673011779785:  85%|████████▍ | 5499/6470 [1:03:15<11:39,  1.39it/s]
4|5500|Loss: 0.7346673011779785:  85%|████████▌ | 5500/6470 [1:03:15<11:44,  1.38it/s]
4|5501|Loss: 0.806285560131073:  85%|████████▌ | 5500/6470 [1:03:16<11:44,  1.38it/s] 
4|5501|Loss: 0.806285560131073:  85%|████████▌ | 5501/6470 [1:03:16<11:11,  1.44it/s]
4|5502|Loss: 0.7643365263938904:  85%|████████▌ | 5501/6470 [1:03:16<11:11,  1.44it/s]
4|5502|Loss: 0.7643365263938904:  85%|████████▌ | 5502/6470 [1:03:17<10:21,  1.56it/s]
4|5503|Loss: 0.794999897480011:  85%|████████▌ | 5502/6470 [1:03:17<10:21,  1.56it/s] 
4|5503|Loss: 0.794999897480011:  85%|████████▌ | 5503/6470 [1:03:17<10:33,  1.53it/s]
4|5504|Loss: 0.98381108045578:  85%|████████▌ | 5503/6470 [1:03:18<10:33,  1.53it/s] 
4|5504|Loss: 0.98381108045578:  85%|████████▌ | 5504/6470 [1:03:18<10:19,  1.56it/s]
4|5505|Loss: 0.8533247113227844:  85%|████████▌ | 5504/6470 [1:03:18<10:19,  1.56it/s]
4|5505|Loss: 0.8533247113227844:  85%|████████▌ | 5505/6470 [1:03:19<10:14,  1.57it/s]
4|5506|Loss: 0.7271530628204346:  85%|████████▌ | 5505/6470 [1:03:19<10:14,  1.57it/s]
4|5506|Loss: 0.7271530628204346:  85%|████████▌ | 5506/6470 [1:03:19<10:30,  1.53it/s]
4|5507|Loss: 0.7886069416999817:  85%|████████▌ | 5506/6470 [1:03:20<10:30,  1.53it/s]
4|5507|Loss: 0.7886069416999817:  85%|████████▌ | 5507/6470 [1:03:20<10:45,  1.49it/s]
4|5508|Loss: 0.8534508347511292:  85%|████████▌ | 5507/6470 [1:03:20<10:45,  1.49it/s]
4|5508|Loss: 0.8534508347511292:  85%|████████▌ | 5508/6470 [1:03:20<10:09,  1.58it/s]
4|5509|Loss: 0.7820492386817932:  85%|████████▌ | 5508/6470 [1:03:21<10:09,  1.58it/s]
4|5509|Loss: 0.7820492386817932:  85%|████████▌ | 5509/6470 [1:03:21<09:43,  1.65it/s]
4|5510|Loss: 0.8875691890716553:  85%|████████▌ | 5509/6470 [1:03:21<09:43,  1.65it/s]
4|5510|Loss: 0.8875691890716553:  85%|████████▌ | 5510/6470 [1:03:22<09:47,  1.63it/s]
4|5511|Loss: 0.7962566614151001:  85%|████████▌ | 5510/6470 [1:03:22<09:47,  1.63it/s]
4|5511|Loss: 0.7962566614151001:  85%|████████▌ | 5511/6470 [1:03:22<09:41,  1.65it/s]
4|5512|Loss: 0.9458292722702026:  85%|████████▌ | 5511/6470 [1:03:23<09:41,  1.65it/s]
4|5512|Loss: 0.9458292722702026:  85%|████████▌ | 5512/6470 [1:03:23<10:38,  1.50it/s]
4|5513|Loss: 0.8834631443023682:  85%|████████▌ | 5512/6470 [1:03:23<10:38,  1.50it/s]
4|5513|Loss: 0.8834631443023682:  85%|████████▌ | 5513/6470 [1:03:24<10:01,  1.59it/s]
4|5514|Loss: 0.861652672290802:  85%|████████▌ | 5513/6470 [1:03:24<10:01,  1.59it/s] 
4|5514|Loss: 0.861652672290802:  85%|████████▌ | 5514/6470 [1:03:24<09:49,  1.62it/s]
4|5515|Loss: 0.6448217630386353:  85%|████████▌ | 5514/6470 [1:03:25<09:49,  1.62it/s]
4|5515|Loss: 0.6448217630386353:  85%|████████▌ | 5515/6470 [1:03:25<11:10,  1.42it/s]
4|5516|Loss: 0.9423741698265076:  85%|████████▌ | 5515/6470 [1:03:26<11:10,  1.42it/s]
4|5516|Loss: 0.9423741698265076:  85%|████████▌ | 5516/6470 [1:03:26<11:34,  1.37it/s]
4|5517|Loss: 0.8388961553573608:  85%|████████▌ | 5516/6470 [1:03:26<11:34,  1.37it/s]
4|5517|Loss: 0.8388961553573608:  85%|████████▌ | 5517/6470 [1:03:27<12:18,  1.29it/s]
4|5518|Loss: 0.9731640219688416:  85%|████████▌ | 5517/6470 [1:03:27<12:18,  1.29it/s]
4|5518|Loss: 0.9731640219688416:  85%|████████▌ | 5518/6470 [1:03:28<12:32,  1.27it/s]
4|5519|Loss: 0.8235923051834106:  85%|████████▌ | 5518/6470 [1:03:28<12:32,  1.27it/s]
4|5519|Loss: 0.8235923051834106:  85%|████████▌ | 5519/6470 [1:03:28<11:28,  1.38it/s]
4|5520|Loss: 0.8281850218772888:  85%|████████▌ | 5519/6470 [1:03:28<11:28,  1.38it/s]
4|5520|Loss: 0.8281850218772888:  85%|████████▌ | 5520/6470 [1:03:29<10:50,  1.46it/s]
4|5521|Loss: 0.8026223182678223:  85%|████████▌ | 5520/6470 [1:03:29<10:50,  1.46it/s]
4|5521|Loss: 0.8026223182678223:  85%|████████▌ | 5521/6470 [1:03:29<10:57,  1.44it/s]
4|5522|Loss: 0.9189721345901489:  85%|████████▌ | 5521/6470 [1:03:30<10:57,  1.44it/s]
4|5522|Loss: 0.9189721345901489:  85%|████████▌ | 5522/6470 [1:03:30<11:04,  1.43it/s]
4|5523|Loss: 0.7263886332511902:  85%|████████▌ | 5522/6470 [1:03:31<11:04,  1.43it/s]
4|5523|Loss: 0.7263886332511902:  85%|████████▌ | 5523/6470 [1:03:31<10:23,  1.52it/s]
4|5524|Loss: 0.8562096357345581:  85%|████████▌ | 5523/6470 [1:03:31<10:23,  1.52it/s]
4|5524|Loss: 0.8562096357345581:  85%|████████▌ | 5524/6470 [1:03:31<09:41,  1.63it/s]
4|5525|Loss: 0.8568517565727234:  85%|████████▌ | 5524/6470 [1:03:32<09:41,  1.63it/s]
4|5525|Loss: 0.8568517565727234:  85%|████████▌ | 5525/6470 [1:03:32<09:10,  1.72it/s]
4|5526|Loss: 0.775397539138794:  85%|████████▌ | 5525/6470 [1:03:32<09:10,  1.72it/s] 
4|5526|Loss: 0.775397539138794:  85%|████████▌ | 5526/6470 [1:03:32<09:53,  1.59it/s]
4|5527|Loss: 0.9334325194358826:  85%|████████▌ | 5526/6470 [1:03:33<09:53,  1.59it/s]
4|5527|Loss: 0.9334325194358826:  85%|████████▌ | 5527/6470 [1:03:33<10:39,  1.47it/s]
4|5528|Loss: 0.9603980183601379:  85%|████████▌ | 5527/6470 [1:03:34<10:39,  1.47it/s]
4|5528|Loss: 0.9603980183601379:  85%|████████▌ | 5528/6470 [1:03:34<10:59,  1.43it/s]
4|5529|Loss: 0.8366430401802063:  85%|████████▌ | 5528/6470 [1:03:34<10:59,  1.43it/s]
4|5529|Loss: 0.8366430401802063:  85%|████████▌ | 5529/6470 [1:03:35<11:21,  1.38it/s]
4|5530|Loss: 0.9857214689254761:  85%|████████▌ | 5529/6470 [1:03:35<11:21,  1.38it/s]
4|5530|Loss: 0.9857214689254761:  85%|████████▌ | 5530/6470 [1:03:36<12:00,  1.30it/s]
4|5531|Loss: 0.8394119739532471:  85%|████████▌ | 5530/6470 [1:03:36<12:00,  1.30it/s]
4|5531|Loss: 0.8394119739532471:  85%|████████▌ | 5531/6470 [1:03:37<12:41,  1.23it/s]
4|5532|Loss: 0.9548252820968628:  85%|████████▌ | 5531/6470 [1:03:37<12:41,  1.23it/s]
4|5532|Loss: 0.9548252820968628:  86%|████████▌ | 5532/6470 [1:03:37<12:03,  1.30it/s]
4|5533|Loss: 0.7669537663459778:  86%|████████▌ | 5532/6470 [1:03:38<12:03,  1.30it/s]
4|5533|Loss: 0.7669537663459778:  86%|████████▌ | 5533/6470 [1:03:38<11:44,  1.33it/s]
4|5534|Loss: 0.798028826713562:  86%|████████▌ | 5533/6470 [1:03:38<11:44,  1.33it/s] 
4|5534|Loss: 0.798028826713562:  86%|████████▌ | 5534/6470 [1:03:39<12:00,  1.30it/s]
4|5535|Loss: 0.8664695620536804:  86%|████████▌ | 5534/6470 [1:03:39<12:00,  1.30it/s]
4|5535|Loss: 0.8664695620536804:  86%|████████▌ | 5535/6470 [1:03:40<11:48,  1.32it/s]
4|5536|Loss: 0.7487422823905945:  86%|████████▌ | 5535/6470 [1:03:40<11:48,  1.32it/s]
4|5536|Loss: 0.7487422823905945:  86%|████████▌ | 5536/6470 [1:03:40<10:11,  1.53it/s]
4|5537|Loss: 0.8458485007286072:  86%|████████▌ | 5536/6470 [1:03:40<10:11,  1.53it/s]
4|5537|Loss: 0.8458485007286072:  86%|████████▌ | 5537/6470 [1:03:40<09:19,  1.67it/s]
4|5538|Loss: 0.8120483756065369:  86%|████████▌ | 5537/6470 [1:03:41<09:19,  1.67it/s]
4|5538|Loss: 0.8120483756065369:  86%|████████▌ | 5538/6470 [1:03:41<10:08,  1.53it/s]
4|5539|Loss: 0.7812610268592834:  86%|████████▌ | 5538/6470 [1:03:42<10:08,  1.53it/s]
4|5539|Loss: 0.7812610268592834:  86%|████████▌ | 5539/6470 [1:03:42<11:01,  1.41it/s]
4|5540|Loss: 0.7395873069763184:  86%|████████▌ | 5539/6470 [1:03:42<11:01,  1.41it/s]
4|5540|Loss: 0.7395873069763184:  86%|████████▌ | 5540/6470 [1:03:43<10:41,  1.45it/s]
4|5541|Loss: 0.8213240504264832:  86%|████████▌ | 5540/6470 [1:03:43<10:41,  1.45it/s]
4|5541|Loss: 0.8213240504264832:  86%|████████▌ | 5541/6470 [1:03:43<10:38,  1.45it/s]
4|5542|Loss: 0.7843310832977295:  86%|████████▌ | 5541/6470 [1:03:44<10:38,  1.45it/s]
4|5542|Loss: 0.7843310832977295:  86%|████████▌ | 5542/6470 [1:03:44<10:16,  1.51it/s]
4|5543|Loss: 0.8485066890716553:  86%|████████▌ | 5542/6470 [1:03:44<10:16,  1.51it/s]
4|5543|Loss: 0.8485066890716553:  86%|████████▌ | 5543/6470 [1:03:45<10:28,  1.47it/s]
4|5544|Loss: 0.9577354788780212:  86%|████████▌ | 5543/6470 [1:03:45<10:28,  1.47it/s]
4|5544|Loss: 0.9577354788780212:  86%|████████▌ | 5544/6470 [1:03:45<10:46,  1.43it/s]
4|5545|Loss: 0.675262451171875:  86%|████████▌ | 5544/6470 [1:03:46<10:46,  1.43it/s] 
4|5545|Loss: 0.675262451171875:  86%|████████▌ | 5545/6470 [1:03:46<11:08,  1.38it/s]
4|5546|Loss: 0.8148969411849976:  86%|████████▌ | 5545/6470 [1:03:47<11:08,  1.38it/s]
4|5546|Loss: 0.8148969411849976:  86%|████████▌ | 5546/6470 [1:03:47<11:15,  1.37it/s]
4|5547|Loss: 0.8579314947128296:  86%|████████▌ | 5546/6470 [1:03:47<11:15,  1.37it/s]
4|5547|Loss: 0.8579314947128296:  86%|████████▌ | 5547/6470 [1:03:47<10:13,  1.50it/s]
4|5548|Loss: 0.7862932682037354:  86%|████████▌ | 5547/6470 [1:03:48<10:13,  1.50it/s]
4|5548|Loss: 0.7862932682037354:  86%|████████▌ | 5548/6470 [1:03:48<09:12,  1.67it/s]
4|5549|Loss: 0.8624528050422668:  86%|████████▌ | 5548/6470 [1:03:48<09:12,  1.67it/s]
4|5549|Loss: 0.8624528050422668:  86%|████████▌ | 5549/6470 [1:03:48<08:14,  1.86it/s]
4|5550|Loss: 0.8264391422271729:  86%|████████▌ | 5549/6470 [1:03:49<08:14,  1.86it/s]
4|5550|Loss: 0.8264391422271729:  86%|████████▌ | 5550/6470 [1:03:49<09:11,  1.67it/s]
4|5551|Loss: 0.9491906762123108:  86%|████████▌ | 5550/6470 [1:03:49<09:11,  1.67it/s]
4|5551|Loss: 0.9491906762123108:  86%|████████▌ | 5551/6470 [1:03:50<09:35,  1.60it/s]
4|5552|Loss: 0.856585681438446:  86%|████████▌ | 5551/6470 [1:03:50<09:35,  1.60it/s] 
4|5552|Loss: 0.856585681438446:  86%|████████▌ | 5552/6470 [1:03:50<09:18,  1.64it/s]
4|5553|Loss: 0.8059036731719971:  86%|████████▌ | 5552/6470 [1:03:51<09:18,  1.64it/s]
4|5553|Loss: 0.8059036731719971:  86%|████████▌ | 5553/6470 [1:03:51<09:42,  1.58it/s]
4|5554|Loss: 0.8434630632400513:  86%|████████▌ | 5553/6470 [1:03:51<09:42,  1.58it/s]
4|5554|Loss: 0.8434630632400513:  86%|████████▌ | 5554/6470 [1:03:52<09:48,  1.56it/s]
4|5555|Loss: 1.166551947593689:  86%|████████▌ | 5554/6470 [1:03:52<09:48,  1.56it/s] 
4|5555|Loss: 1.166551947593689:  86%|████████▌ | 5555/6470 [1:03:53<11:28,  1.33it/s]
4|5556|Loss: 1.0589560270309448:  86%|████████▌ | 5555/6470 [1:03:53<11:28,  1.33it/s]
4|5556|Loss: 1.0589560270309448:  86%|████████▌ | 5556/6470 [1:03:53<11:19,  1.34it/s]
4|5557|Loss: 0.7231041789054871:  86%|████████▌ | 5556/6470 [1:03:54<11:19,  1.34it/s]
4|5557|Loss: 0.7231041789054871:  86%|████████▌ | 5557/6470 [1:03:54<10:19,  1.47it/s]
4|5558|Loss: 0.9230114817619324:  86%|████████▌ | 5557/6470 [1:03:54<10:19,  1.47it/s]
4|5558|Loss: 0.9230114817619324:  86%|████████▌ | 5558/6470 [1:03:55<10:20,  1.47it/s]
4|5559|Loss: 0.8652376532554626:  86%|████████▌ | 5558/6470 [1:03:55<10:20,  1.47it/s]
4|5559|Loss: 0.8652376532554626:  86%|████████▌ | 5559/6470 [1:03:55<10:10,  1.49it/s]
4|5560|Loss: 0.8555908203125:  86%|████████▌ | 5559/6470 [1:03:56<10:10,  1.49it/s]   
4|5560|Loss: 0.8555908203125:  86%|████████▌ | 5560/6470 [1:03:56<09:53,  1.53it/s]
4|5561|Loss: 0.8522549867630005:  86%|████████▌ | 5560/6470 [1:03:56<09:53,  1.53it/s]
4|5561|Loss: 0.8522549867630005:  86%|████████▌ | 5561/6470 [1:03:57<10:28,  1.45it/s]
4|5562|Loss: 0.7676652669906616:  86%|████████▌ | 5561/6470 [1:03:57<10:28,  1.45it/s]
4|5562|Loss: 0.7676652669906616:  86%|████████▌ | 5562/6470 [1:03:57<10:36,  1.43it/s]
4|5563|Loss: 0.8194695115089417:  86%|████████▌ | 5562/6470 [1:03:58<10:36,  1.43it/s]
4|5563|Loss: 0.8194695115089417:  86%|████████▌ | 5563/6470 [1:03:58<10:45,  1.41it/s]
4|5564|Loss: 0.8587514162063599:  86%|████████▌ | 5563/6470 [1:03:59<10:45,  1.41it/s]
4|5564|Loss: 0.8587514162063599:  86%|████████▌ | 5564/6470 [1:03:59<10:53,  1.39it/s]
4|5565|Loss: 0.8586857318878174:  86%|████████▌ | 5564/6470 [1:03:59<10:53,  1.39it/s]
4|5565|Loss: 0.8586857318878174:  86%|████████▌ | 5565/6470 [1:03:59<10:09,  1.49it/s]
4|5566|Loss: 0.9174668192863464:  86%|████████▌ | 5565/6470 [1:04:00<10:09,  1.49it/s]
4|5566|Loss: 0.9174668192863464:  86%|████████▌ | 5566/6470 [1:04:00<10:29,  1.44it/s]
4|5567|Loss: 0.966064453125:  86%|████████▌ | 5566/6470 [1:04:01<10:29,  1.44it/s]    
4|5567|Loss: 0.966064453125:  86%|████████▌ | 5567/6470 [1:04:01<10:36,  1.42it/s]
4|5568|Loss: 0.8239301443099976:  86%|████████▌ | 5567/6470 [1:04:01<10:36,  1.42it/s]
4|5568|Loss: 0.8239301443099976:  86%|████████▌ | 5568/6470 [1:04:02<10:44,  1.40it/s]
4|5569|Loss: 0.8076831102371216:  86%|████████▌ | 5568/6470 [1:04:02<10:44,  1.40it/s]
4|5569|Loss: 0.8076831102371216:  86%|████████▌ | 5569/6470 [1:04:02<10:31,  1.43it/s]
4|5570|Loss: 0.7962424755096436:  86%|████████▌ | 5569/6470 [1:04:03<10:31,  1.43it/s]
4|5570|Loss: 0.7962424755096436:  86%|████████▌ | 5570/6470 [1:04:03<10:34,  1.42it/s]
4|5571|Loss: 0.7410039305686951:  86%|████████▌ | 5570/6470 [1:04:03<10:34,  1.42it/s]
4|5571|Loss: 0.7410039305686951:  86%|████████▌ | 5571/6470 [1:04:04<10:34,  1.42it/s]
4|5572|Loss: 0.7722504138946533:  86%|████████▌ | 5571/6470 [1:04:04<10:34,  1.42it/s]
4|5572|Loss: 0.7722504138946533:  86%|████████▌ | 5572/6470 [1:04:04<10:38,  1.41it/s]
4|5573|Loss: 0.807293713092804:  86%|████████▌ | 5572/6470 [1:04:05<10:38,  1.41it/s] 
4|5573|Loss: 0.807293713092804:  86%|████████▌ | 5573/6470 [1:04:05<11:02,  1.35it/s]
4|5574|Loss: 0.8469552397727966:  86%|████████▌ | 5573/6470 [1:04:06<11:02,  1.35it/s]
4|5574|Loss: 0.8469552397727966:  86%|████████▌ | 5574/6470 [1:04:06<10:46,  1.39it/s]
4|5575|Loss: 0.8551591634750366:  86%|████████▌ | 5574/6470 [1:04:06<10:46,  1.39it/s]
4|5575|Loss: 0.8551591634750366:  86%|████████▌ | 5575/6470 [1:04:07<11:12,  1.33it/s]
4|5576|Loss: 0.7723190188407898:  86%|████████▌ | 5575/6470 [1:04:07<11:12,  1.33it/s]
4|5576|Loss: 0.7723190188407898:  86%|████████▌ | 5576/6470 [1:04:07<10:15,  1.45it/s]
4|5577|Loss: 0.7536985874176025:  86%|████████▌ | 5576/6470 [1:04:08<10:15,  1.45it/s]
4|5577|Loss: 0.7536985874176025:  86%|████████▌ | 5577/6470 [1:04:08<10:14,  1.45it/s]
4|5578|Loss: 1.0441322326660156:  86%|████████▌ | 5577/6470 [1:04:08<10:14,  1.45it/s]
4|5578|Loss: 1.0441322326660156:  86%|████████▌ | 5578/6470 [1:04:08<09:29,  1.57it/s]
4|5579|Loss: 0.8427739143371582:  86%|████████▌ | 5578/6470 [1:04:09<09:29,  1.57it/s]
4|5579|Loss: 0.8427739143371582:  86%|████████▌ | 5579/6470 [1:04:09<09:57,  1.49it/s]
4|5580|Loss: 0.992893636226654:  86%|████████▌ | 5579/6470 [1:04:10<09:57,  1.49it/s] 
4|5580|Loss: 0.992893636226654:  86%|████████▌ | 5580/6470 [1:04:10<10:18,  1.44it/s]
4|5581|Loss: 0.6761234998703003:  86%|████████▌ | 5580/6470 [1:04:10<10:18,  1.44it/s]
4|5581|Loss: 0.6761234998703003:  86%|████████▋ | 5581/6470 [1:04:11<10:27,  1.42it/s]
4|5582|Loss: 1.029318928718567:  86%|████████▋ | 5581/6470 [1:04:11<10:27,  1.42it/s] 
4|5582|Loss: 1.029318928718567:  86%|████████▋ | 5582/6470 [1:04:11<10:14,  1.44it/s]
4|5583|Loss: 0.9748085737228394:  86%|████████▋ | 5582/6470 [1:04:12<10:14,  1.44it/s]
4|5583|Loss: 0.9748085737228394:  86%|████████▋ | 5583/6470 [1:04:12<09:56,  1.49it/s]
4|5584|Loss: 0.780400276184082:  86%|████████▋ | 5583/6470 [1:04:12<09:56,  1.49it/s] 
4|5584|Loss: 0.780400276184082:  86%|████████▋ | 5584/6470 [1:04:13<10:06,  1.46it/s]
4|5585|Loss: 0.9707053899765015:  86%|████████▋ | 5584/6470 [1:04:13<10:06,  1.46it/s]
4|5585|Loss: 0.9707053899765015:  86%|████████▋ | 5585/6470 [1:04:13<09:34,  1.54it/s]
4|5586|Loss: 0.8101202845573425:  86%|████████▋ | 5585/6470 [1:04:14<09:34,  1.54it/s]
4|5586|Loss: 0.8101202845573425:  86%|████████▋ | 5586/6470 [1:04:14<09:14,  1.59it/s]
4|5587|Loss: 0.8359014987945557:  86%|████████▋ | 5586/6470 [1:04:14<09:14,  1.59it/s]
4|5587|Loss: 0.8359014987945557:  86%|████████▋ | 5587/6470 [1:04:15<09:53,  1.49it/s]
4|5588|Loss: 0.8654014468193054:  86%|████████▋ | 5587/6470 [1:04:15<09:53,  1.49it/s]
4|5588|Loss: 0.8654014468193054:  86%|████████▋ | 5588/6470 [1:04:15<09:15,  1.59it/s]
4|5589|Loss: 0.8134134411811829:  86%|████████▋ | 5588/6470 [1:04:16<09:15,  1.59it/s]
4|5589|Loss: 0.8134134411811829:  86%|████████▋ | 5589/6470 [1:04:16<09:26,  1.56it/s]
4|5590|Loss: 0.961785614490509:  86%|████████▋ | 5589/6470 [1:04:16<09:26,  1.56it/s] 
4|5590|Loss: 0.961785614490509:  86%|████████▋ | 5590/6470 [1:04:16<09:31,  1.54it/s]
4|5591|Loss: 0.6981995701789856:  86%|████████▋ | 5590/6470 [1:04:17<09:31,  1.54it/s]
4|5591|Loss: 0.6981995701789856:  86%|████████▋ | 5591/6470 [1:04:17<10:00,  1.46it/s]
4|5592|Loss: 0.9519415497779846:  86%|████████▋ | 5591/6470 [1:04:18<10:00,  1.46it/s]
4|5592|Loss: 0.9519415497779846:  86%|████████▋ | 5592/6470 [1:04:18<10:19,  1.42it/s]
4|5593|Loss: 0.825251042842865:  86%|████████▋ | 5592/6470 [1:04:18<10:19,  1.42it/s] 
4|5593|Loss: 0.825251042842865:  86%|████████▋ | 5593/6470 [1:04:19<10:09,  1.44it/s]
4|5594|Loss: 0.7803194522857666:  86%|████████▋ | 5593/6470 [1:04:19<10:09,  1.44it/s]
4|5594|Loss: 0.7803194522857666:  86%|████████▋ | 5594/6470 [1:04:19<10:06,  1.44it/s]
4|5595|Loss: 0.8210451006889343:  86%|████████▋ | 5594/6470 [1:04:20<10:06,  1.44it/s]
4|5595|Loss: 0.8210451006889343:  86%|████████▋ | 5595/6470 [1:04:20<10:18,  1.41it/s]
4|5596|Loss: 1.0474562644958496:  86%|████████▋ | 5595/6470 [1:04:21<10:18,  1.41it/s]
4|5596|Loss: 1.0474562644958496:  86%|████████▋ | 5596/6470 [1:04:21<10:57,  1.33it/s]
4|5597|Loss: 0.7568280100822449:  86%|████████▋ | 5596/6470 [1:04:21<10:57,  1.33it/s]
4|5597|Loss: 0.7568280100822449:  87%|████████▋ | 5597/6470 [1:04:22<10:36,  1.37it/s]
4|5598|Loss: 0.740389347076416:  87%|████████▋ | 5597/6470 [1:04:22<10:36,  1.37it/s] 
4|5598|Loss: 0.740389347076416:  87%|████████▋ | 5598/6470 [1:04:22<10:42,  1.36it/s]
4|5599|Loss: 0.8089853525161743:  87%|████████▋ | 5598/6470 [1:04:23<10:42,  1.36it/s]
4|5599|Loss: 0.8089853525161743:  87%|████████▋ | 5599/6470 [1:04:23<10:25,  1.39it/s]
4|5600|Loss: 0.6496333479881287:  87%|████████▋ | 5599/6470 [1:04:24<10:25,  1.39it/s]
4|5600|Loss: 0.6496333479881287:  87%|████████▋ | 5600/6470 [1:04:24<10:53,  1.33it/s]
4|5601|Loss: 0.7518072128295898:  87%|████████▋ | 5600/6470 [1:04:24<10:53,  1.33it/s]
4|5601|Loss: 0.7518072128295898:  87%|████████▋ | 5601/6470 [1:04:25<11:22,  1.27it/s]
4|5602|Loss: 0.8556758165359497:  87%|████████▋ | 5601/6470 [1:04:25<11:22,  1.27it/s]
4|5602|Loss: 0.8556758165359497:  87%|████████▋ | 5602/6470 [1:04:26<11:16,  1.28it/s]
4|5603|Loss: 0.876048743724823:  87%|████████▋ | 5602/6470 [1:04:26<11:16,  1.28it/s] 
4|5603|Loss: 0.876048743724823:  87%|████████▋ | 5603/6470 [1:04:26<10:40,  1.35it/s]
4|5604|Loss: 0.904827892780304:  87%|████████▋ | 5603/6470 [1:04:27<10:40,  1.35it/s]
4|5604|Loss: 0.904827892780304:  87%|████████▋ | 5604/6470 [1:04:27<10:07,  1.43it/s]
4|5605|Loss: 0.9643979668617249:  87%|████████▋ | 5604/6470 [1:04:27<10:07,  1.43it/s]
4|5605|Loss: 0.9643979668617249:  87%|████████▋ | 5605/6470 [1:04:28<10:29,  1.37it/s]
4|5606|Loss: 0.9988409876823425:  87%|████████▋ | 5605/6470 [1:04:28<10:29,  1.37it/s]
4|5606|Loss: 0.9988409876823425:  87%|████████▋ | 5606/6470 [1:04:28<11:06,  1.30it/s]
4|5607|Loss: 0.8652279376983643:  87%|████████▋ | 5606/6470 [1:04:29<11:06,  1.30it/s]
4|5607|Loss: 0.8652279376983643:  87%|████████▋ | 5607/6470 [1:04:29<10:42,  1.34it/s]
4|5608|Loss: 0.7536744475364685:  87%|████████▋ | 5607/6470 [1:04:29<10:42,  1.34it/s]
4|5608|Loss: 0.7536744475364685:  87%|████████▋ | 5608/6470 [1:04:30<10:00,  1.44it/s]
4|5609|Loss: 0.8725913166999817:  87%|████████▋ | 5608/6470 [1:04:30<10:00,  1.44it/s]
4|5609|Loss: 0.8725913166999817:  87%|████████▋ | 5609/6470 [1:04:30<09:37,  1.49it/s]
4|5610|Loss: 0.8423779606819153:  87%|████████▋ | 5609/6470 [1:04:31<09:37,  1.49it/s]
4|5610|Loss: 0.8423779606819153:  87%|████████▋ | 5610/6470 [1:04:31<08:31,  1.68it/s]
4|5611|Loss: 0.7733155488967896:  87%|████████▋ | 5610/6470 [1:04:31<08:31,  1.68it/s]
4|5611|Loss: 0.7733155488967896:  87%|████████▋ | 5611/6470 [1:04:31<08:51,  1.62it/s]
4|5612|Loss: 0.834929883480072:  87%|████████▋ | 5611/6470 [1:04:32<08:51,  1.62it/s] 
4|5612|Loss: 0.834929883480072:  87%|████████▋ | 5612/6470 [1:04:32<09:37,  1.49it/s]
4|5613|Loss: 0.7623133063316345:  87%|████████▋ | 5612/6470 [1:04:33<09:37,  1.49it/s]
4|5613|Loss: 0.7623133063316345:  87%|████████▋ | 5613/6470 [1:04:33<10:00,  1.43it/s]
4|5614|Loss: 0.7310618162155151:  87%|████████▋ | 5613/6470 [1:04:33<10:00,  1.43it/s]
4|5614|Loss: 0.7310618162155151:  87%|████████▋ | 5614/6470 [1:04:34<09:47,  1.46it/s]
4|5615|Loss: 0.9740559458732605:  87%|████████▋ | 5614/6470 [1:04:34<09:47,  1.46it/s]
4|5615|Loss: 0.9740559458732605:  87%|████████▋ | 5615/6470 [1:04:34<09:28,  1.50it/s]
4|5616|Loss: 0.7918403744697571:  87%|████████▋ | 5615/6470 [1:04:35<09:28,  1.50it/s]
4|5616|Loss: 0.7918403744697571:  87%|████████▋ | 5616/6470 [1:04:35<08:53,  1.60it/s]
4|5617|Loss: 0.84578537940979:  87%|████████▋ | 5616/6470 [1:04:35<08:53,  1.60it/s]  
4|5617|Loss: 0.84578537940979:  87%|████████▋ | 5617/6470 [1:04:35<08:31,  1.67it/s]
4|5618|Loss: 0.8840591311454773:  87%|████████▋ | 5617/6470 [1:04:36<08:31,  1.67it/s]
4|5618|Loss: 0.8840591311454773:  87%|████████▋ | 5618/6470 [1:04:36<09:32,  1.49it/s]
4|5619|Loss: 0.9596675634384155:  87%|████████▋ | 5618/6470 [1:04:37<09:32,  1.49it/s]
4|5619|Loss: 0.9596675634384155:  87%|████████▋ | 5619/6470 [1:04:37<09:52,  1.44it/s]
4|5620|Loss: 0.8533629775047302:  87%|████████▋ | 5619/6470 [1:04:37<09:52,  1.44it/s]
4|5620|Loss: 0.8533629775047302:  87%|████████▋ | 5620/6470 [1:04:38<09:55,  1.43it/s]
4|5621|Loss: 0.8057583570480347:  87%|████████▋ | 5620/6470 [1:04:38<09:55,  1.43it/s]
4|5621|Loss: 0.8057583570480347:  87%|████████▋ | 5621/6470 [1:04:38<10:03,  1.41it/s]
4|5622|Loss: 0.8425519466400146:  87%|████████▋ | 5621/6470 [1:04:39<10:03,  1.41it/s]
4|5622|Loss: 0.8425519466400146:  87%|████████▋ | 5622/6470 [1:04:39<09:37,  1.47it/s]
4|5623|Loss: 0.9083443880081177:  87%|████████▋ | 5622/6470 [1:04:39<09:37,  1.47it/s]
4|5623|Loss: 0.9083443880081177:  87%|████████▋ | 5623/6470 [1:04:40<10:08,  1.39it/s]
4|5624|Loss: 0.905585765838623:  87%|████████▋ | 5623/6470 [1:04:40<10:08,  1.39it/s] 
4|5624|Loss: 0.905585765838623:  87%|████████▋ | 5624/6470 [1:04:41<10:23,  1.36it/s]
4|5625|Loss: 0.7518163919448853:  87%|████████▋ | 5624/6470 [1:04:41<10:23,  1.36it/s]
4|5625|Loss: 0.7518163919448853:  87%|████████▋ | 5625/6470 [1:04:41<10:18,  1.37it/s]
4|5626|Loss: 0.863603413105011:  87%|████████▋ | 5625/6470 [1:04:42<10:18,  1.37it/s] 
4|5626|Loss: 0.863603413105011:  87%|████████▋ | 5626/6470 [1:04:42<10:54,  1.29it/s]
4|5627|Loss: 0.9804803133010864:  87%|████████▋ | 5626/6470 [1:04:43<10:54,  1.29it/s]
4|5627|Loss: 0.9804803133010864:  87%|████████▋ | 5627/6470 [1:04:43<10:33,  1.33it/s]
4|5628|Loss: 0.8956101536750793:  87%|████████▋ | 5627/6470 [1:04:43<10:33,  1.33it/s]
4|5628|Loss: 0.8956101536750793:  87%|████████▋ | 5628/6470 [1:04:44<10:05,  1.39it/s]
4|5629|Loss: 0.7888578772544861:  87%|████████▋ | 5628/6470 [1:04:44<10:05,  1.39it/s]
4|5629|Loss: 0.7888578772544861:  87%|████████▋ | 5629/6470 [1:04:44<09:26,  1.49it/s]
4|5630|Loss: 1.1264716386795044:  87%|████████▋ | 5629/6470 [1:04:44<09:26,  1.49it/s]
4|5630|Loss: 1.1264716386795044:  87%|████████▋ | 5630/6470 [1:04:45<09:56,  1.41it/s]
4|5631|Loss: 0.698549211025238:  87%|████████▋ | 5630/6470 [1:04:45<09:56,  1.41it/s] 
4|5631|Loss: 0.698549211025238:  87%|████████▋ | 5631/6470 [1:04:45<09:25,  1.48it/s]
4|5632|Loss: 0.869684100151062:  87%|████████▋ | 5631/6470 [1:04:46<09:25,  1.48it/s]
4|5632|Loss: 0.869684100151062:  87%|████████▋ | 5632/6470 [1:04:46<09:10,  1.52it/s]
4|5633|Loss: 0.7920954823493958:  87%|████████▋ | 5632/6470 [1:04:46<09:10,  1.52it/s]
4|5633|Loss: 0.7920954823493958:  87%|████████▋ | 5633/6470 [1:04:47<08:52,  1.57it/s]
4|5634|Loss: 0.785047173500061:  87%|████████▋ | 5633/6470 [1:04:47<08:52,  1.57it/s] 
4|5634|Loss: 0.785047173500061:  87%|████████▋ | 5634/6470 [1:04:47<08:36,  1.62it/s]
4|5635|Loss: 0.8446725010871887:  87%|████████▋ | 5634/6470 [1:04:48<08:36,  1.62it/s]
4|5635|Loss: 0.8446725010871887:  87%|████████▋ | 5635/6470 [1:04:48<08:13,  1.69it/s]
4|5636|Loss: 0.9039078950881958:  87%|████████▋ | 5635/6470 [1:04:48<08:13,  1.69it/s]
4|5636|Loss: 0.9039078950881958:  87%|████████▋ | 5636/6470 [1:04:48<07:40,  1.81it/s]
4|5637|Loss: 0.8562875390052795:  87%|████████▋ | 5636/6470 [1:04:49<07:40,  1.81it/s]
4|5637|Loss: 0.8562875390052795:  87%|████████▋ | 5637/6470 [1:04:49<07:51,  1.77it/s]
4|5638|Loss: 0.8824474811553955:  87%|████████▋ | 5637/6470 [1:04:49<07:51,  1.77it/s]
4|5638|Loss: 0.8824474811553955:  87%|████████▋ | 5638/6470 [1:04:49<08:09,  1.70it/s]
4|5639|Loss: 0.8212046027183533:  87%|████████▋ | 5638/6470 [1:04:50<08:09,  1.70it/s]
4|5639|Loss: 0.8212046027183533:  87%|████████▋ | 5639/6470 [1:04:50<08:34,  1.62it/s]
4|5640|Loss: 0.8052201271057129:  87%|████████▋ | 5639/6470 [1:04:51<08:34,  1.62it/s]
4|5640|Loss: 0.8052201271057129:  87%|████████▋ | 5640/6470 [1:04:51<08:53,  1.56it/s]
4|5641|Loss: 1.0581051111221313:  87%|████████▋ | 5640/6470 [1:04:51<08:53,  1.56it/s]
4|5641|Loss: 1.0581051111221313:  87%|████████▋ | 5641/6470 [1:04:52<08:55,  1.55it/s]
4|5642|Loss: 0.8083364963531494:  87%|████████▋ | 5641/6470 [1:04:52<08:55,  1.55it/s]
4|5642|Loss: 0.8083364963531494:  87%|████████▋ | 5642/6470 [1:04:52<08:23,  1.65it/s]
4|5643|Loss: 0.8397467732429504:  87%|████████▋ | 5642/6470 [1:04:52<08:23,  1.65it/s]
4|5643|Loss: 0.8397467732429504:  87%|████████▋ | 5643/6470 [1:04:53<08:40,  1.59it/s]
4|5644|Loss: 0.8194484710693359:  87%|████████▋ | 5643/6470 [1:04:53<08:40,  1.59it/s]
4|5644|Loss: 0.8194484710693359:  87%|████████▋ | 5644/6470 [1:04:53<09:10,  1.50it/s]
4|5645|Loss: 0.9282674193382263:  87%|████████▋ | 5644/6470 [1:04:54<09:10,  1.50it/s]
4|5645|Loss: 0.9282674193382263:  87%|████████▋ | 5645/6470 [1:04:54<08:58,  1.53it/s]
4|5646|Loss: 0.8977730870246887:  87%|████████▋ | 5645/6470 [1:04:54<08:58,  1.53it/s]
4|5646|Loss: 0.8977730870246887:  87%|████████▋ | 5646/6470 [1:04:55<08:39,  1.59it/s]
4|5647|Loss: 0.7637205123901367:  87%|████████▋ | 5646/6470 [1:04:55<08:39,  1.59it/s]
4|5647|Loss: 0.7637205123901367:  87%|████████▋ | 5647/6470 [1:04:55<08:01,  1.71it/s]
4|5648|Loss: 0.9971038699150085:  87%|████████▋ | 5647/6470 [1:04:55<08:01,  1.71it/s]
4|5648|Loss: 0.9971038699150085:  87%|████████▋ | 5648/6470 [1:04:56<07:58,  1.72it/s]
4|5649|Loss: 0.8710372447967529:  87%|████████▋ | 5648/6470 [1:04:56<07:58,  1.72it/s]
4|5649|Loss: 0.8710372447967529:  87%|████████▋ | 5649/6470 [1:04:57<08:53,  1.54it/s]
4|5650|Loss: 0.9227839708328247:  87%|████████▋ | 5649/6470 [1:04:57<08:53,  1.54it/s]
4|5650|Loss: 0.9227839708328247:  87%|████████▋ | 5650/6470 [1:04:57<09:40,  1.41it/s]
4|5651|Loss: 0.8419498801231384:  87%|████████▋ | 5650/6470 [1:04:58<09:40,  1.41it/s]
4|5651|Loss: 0.8419498801231384:  87%|████████▋ | 5651/6470 [1:04:58<09:50,  1.39it/s]
4|5652|Loss: 0.8822875618934631:  87%|████████▋ | 5651/6470 [1:04:59<09:50,  1.39it/s]
4|5652|Loss: 0.8822875618934631:  87%|████████▋ | 5652/6470 [1:04:59<10:10,  1.34it/s]
4|5653|Loss: 0.8986346125602722:  87%|████████▋ | 5652/6470 [1:04:59<10:10,  1.34it/s]
4|5653|Loss: 0.8986346125602722:  87%|████████▋ | 5653/6470 [1:05:00<09:53,  1.38it/s]
4|5654|Loss: 0.9264266490936279:  87%|████████▋ | 5653/6470 [1:05:00<09:53,  1.38it/s]
4|5654|Loss: 0.9264266490936279:  87%|████████▋ | 5654/6470 [1:05:00<09:26,  1.44it/s]
4|5655|Loss: 0.7808989882469177:  87%|████████▋ | 5654/6470 [1:05:01<09:26,  1.44it/s]
4|5655|Loss: 0.7808989882469177:  87%|████████▋ | 5655/6470 [1:05:01<08:57,  1.52it/s]
4|5656|Loss: 0.8704715967178345:  87%|████████▋ | 5655/6470 [1:05:01<08:57,  1.52it/s]
4|5656|Loss: 0.8704715967178345:  87%|████████▋ | 5656/6470 [1:05:02<09:08,  1.48it/s]
4|5657|Loss: 0.8649201393127441:  87%|████████▋ | 5656/6470 [1:05:02<09:08,  1.48it/s]
4|5657|Loss: 0.8649201393127441:  87%|████████▋ | 5657/6470 [1:05:02<09:22,  1.44it/s]
4|5658|Loss: 0.8135840892791748:  87%|████████▋ | 5657/6470 [1:05:03<09:22,  1.44it/s]
4|5658|Loss: 0.8135840892791748:  87%|████████▋ | 5658/6470 [1:05:03<09:31,  1.42it/s]
4|5659|Loss: 0.7642753720283508:  87%|████████▋ | 5658/6470 [1:05:03<09:31,  1.42it/s]
4|5659|Loss: 0.7642753720283508:  87%|████████▋ | 5659/6470 [1:05:04<09:51,  1.37it/s]
4|5660|Loss: 0.626018762588501:  87%|████████▋ | 5659/6470 [1:05:04<09:51,  1.37it/s] 
4|5660|Loss: 0.626018762588501:  87%|████████▋ | 5660/6470 [1:05:04<09:00,  1.50it/s]
4|5661|Loss: 0.9032893776893616:  87%|████████▋ | 5660/6470 [1:05:05<09:00,  1.50it/s]
4|5661|Loss: 0.9032893776893616:  87%|████████▋ | 5661/6470 [1:05:05<08:42,  1.55it/s]
4|5662|Loss: 0.7555872797966003:  87%|████████▋ | 5661/6470 [1:05:05<08:42,  1.55it/s]
4|5662|Loss: 0.7555872797966003:  88%|████████▊ | 5662/6470 [1:05:06<09:43,  1.39it/s]
4|5663|Loss: 0.8405686616897583:  88%|████████▊ | 5662/6470 [1:05:06<09:43,  1.39it/s]
4|5663|Loss: 0.8405686616897583:  88%|████████▊ | 5663/6470 [1:05:07<10:09,  1.32it/s]
4|5664|Loss: 0.7316079139709473:  88%|████████▊ | 5663/6470 [1:05:07<10:09,  1.32it/s]
4|5664|Loss: 0.7316079139709473:  88%|████████▊ | 5664/6470 [1:05:07<10:24,  1.29it/s]
4|5665|Loss: 0.8611360788345337:  88%|████████▊ | 5664/6470 [1:05:08<10:24,  1.29it/s]
4|5665|Loss: 0.8611360788345337:  88%|████████▊ | 5665/6470 [1:05:08<10:02,  1.34it/s]
4|5666|Loss: 0.9037789106369019:  88%|████████▊ | 5665/6470 [1:05:09<10:02,  1.34it/s]
4|5666|Loss: 0.9037789106369019:  88%|████████▊ | 5666/6470 [1:05:09<10:24,  1.29it/s]
4|5667|Loss: 1.0121797323226929:  88%|████████▊ | 5666/6470 [1:05:09<10:24,  1.29it/s]
4|5667|Loss: 1.0121797323226929:  88%|████████▊ | 5667/6470 [1:05:10<09:28,  1.41it/s]
4|5668|Loss: 0.882663369178772:  88%|████████▊ | 5667/6470 [1:05:10<09:28,  1.41it/s] 
4|5668|Loss: 0.882663369178772:  88%|████████▊ | 5668/6470 [1:05:10<09:00,  1.48it/s]
4|5669|Loss: 0.8726756572723389:  88%|████████▊ | 5668/6470 [1:05:10<09:00,  1.48it/s]
4|5669|Loss: 0.8726756572723389:  88%|████████▊ | 5669/6470 [1:05:11<08:33,  1.56it/s]
4|5670|Loss: 0.9065880179405212:  88%|████████▊ | 5669/6470 [1:05:11<08:33,  1.56it/s]
4|5670|Loss: 0.9065880179405212:  88%|████████▊ | 5670/6470 [1:05:11<08:59,  1.48it/s]
4|5671|Loss: 0.8704764246940613:  88%|████████▊ | 5670/6470 [1:05:12<08:59,  1.48it/s]
4|5671|Loss: 0.8704764246940613:  88%|████████▊ | 5671/6470 [1:05:12<09:00,  1.48it/s]
4|5672|Loss: 0.7822456359863281:  88%|████████▊ | 5671/6470 [1:05:13<09:00,  1.48it/s]
4|5672|Loss: 0.7822456359863281:  88%|████████▊ | 5672/6470 [1:05:13<09:31,  1.40it/s]
4|5673|Loss: 0.8725358843803406:  88%|████████▊ | 5672/6470 [1:05:13<09:31,  1.40it/s]
4|5673|Loss: 0.8725358843803406:  88%|████████▊ | 5673/6470 [1:05:14<09:42,  1.37it/s]
4|5674|Loss: 0.877973735332489:  88%|████████▊ | 5673/6470 [1:05:14<09:42,  1.37it/s] 
4|5674|Loss: 0.877973735332489:  88%|████████▊ | 5674/6470 [1:05:14<10:03,  1.32it/s]
4|5675|Loss: 0.813615083694458:  88%|████████▊ | 5674/6470 [1:05:15<10:03,  1.32it/s]
4|5675|Loss: 0.813615083694458:  88%|████████▊ | 5675/6470 [1:05:15<10:23,  1.27it/s]
4|5676|Loss: 0.7414073944091797:  88%|████████▊ | 5675/6470 [1:05:16<10:23,  1.27it/s]
4|5676|Loss: 0.7414073944091797:  88%|████████▊ | 5676/6470 [1:05:16<09:57,  1.33it/s]
4|5677|Loss: 0.8756024837493896:  88%|████████▊ | 5676/6470 [1:05:16<09:57,  1.33it/s]
4|5677|Loss: 0.8756024837493896:  88%|████████▊ | 5677/6470 [1:05:17<09:17,  1.42it/s]
4|5678|Loss: 0.8810397386550903:  88%|████████▊ | 5677/6470 [1:05:17<09:17,  1.42it/s]
4|5678|Loss: 0.8810397386550903:  88%|████████▊ | 5678/6470 [1:05:17<08:54,  1.48it/s]
4|5679|Loss: 0.9308157563209534:  88%|████████▊ | 5678/6470 [1:05:18<08:54,  1.48it/s]
4|5679|Loss: 0.9308157563209534:  88%|████████▊ | 5679/6470 [1:05:18<08:33,  1.54it/s]
4|5680|Loss: 0.8071205019950867:  88%|████████▊ | 5679/6470 [1:05:18<08:33,  1.54it/s]
4|5680|Loss: 0.8071205019950867:  88%|████████▊ | 5680/6470 [1:05:18<08:11,  1.61it/s]
4|5681|Loss: 0.7721097469329834:  88%|████████▊ | 5680/6470 [1:05:19<08:11,  1.61it/s]
4|5681|Loss: 0.7721097469329834:  88%|████████▊ | 5681/6470 [1:05:19<08:47,  1.50it/s]
4|5682|Loss: 0.722072184085846:  88%|████████▊ | 5681/6470 [1:05:20<08:47,  1.50it/s] 
4|5682|Loss: 0.722072184085846:  88%|████████▊ | 5682/6470 [1:05:20<09:01,  1.45it/s]
4|5683|Loss: 0.9005164504051208:  88%|████████▊ | 5682/6470 [1:05:20<09:01,  1.45it/s]
4|5683|Loss: 0.9005164504051208:  88%|████████▊ | 5683/6470 [1:05:21<09:24,  1.39it/s]
4|5684|Loss: 0.7785091400146484:  88%|████████▊ | 5683/6470 [1:05:21<09:24,  1.39it/s]
4|5684|Loss: 0.7785091400146484:  88%|████████▊ | 5684/6470 [1:05:22<09:53,  1.32it/s]
4|5685|Loss: 0.8355541825294495:  88%|████████▊ | 5684/6470 [1:05:22<09:53,  1.32it/s]
4|5685|Loss: 0.8355541825294495:  88%|████████▊ | 5685/6470 [1:05:22<09:01,  1.45it/s]
4|5686|Loss: 0.9348334074020386:  88%|████████▊ | 5685/6470 [1:05:22<09:01,  1.45it/s]
4|5686|Loss: 0.9348334074020386:  88%|████████▊ | 5686/6470 [1:05:23<08:28,  1.54it/s]
4|5687|Loss: 0.7315670251846313:  88%|████████▊ | 5686/6470 [1:05:23<08:28,  1.54it/s]
4|5687|Loss: 0.7315670251846313:  88%|████████▊ | 5687/6470 [1:05:23<09:10,  1.42it/s]
4|5688|Loss: 0.8952364921569824:  88%|████████▊ | 5687/6470 [1:05:24<09:10,  1.42it/s]
4|5688|Loss: 0.8952364921569824:  88%|████████▊ | 5688/6470 [1:05:24<09:04,  1.44it/s]
4|5689|Loss: 0.8853887319564819:  88%|████████▊ | 5688/6470 [1:05:24<09:04,  1.44it/s]
4|5689|Loss: 0.8853887319564819:  88%|████████▊ | 5689/6470 [1:05:25<09:01,  1.44it/s]
4|5690|Loss: 0.827126681804657:  88%|████████▊ | 5689/6470 [1:05:25<09:01,  1.44it/s] 
4|5690|Loss: 0.827126681804657:  88%|████████▊ | 5690/6470 [1:05:25<08:28,  1.53it/s]
4|5691|Loss: 0.9172577261924744:  88%|████████▊ | 5690/6470 [1:05:26<08:28,  1.53it/s]
4|5691|Loss: 0.9172577261924744:  88%|████████▊ | 5691/6470 [1:05:26<08:36,  1.51it/s]
4|5692|Loss: 0.8178424835205078:  88%|████████▊ | 5691/6470 [1:05:26<08:36,  1.51it/s]
4|5692|Loss: 0.8178424835205078:  88%|████████▊ | 5692/6470 [1:05:27<09:03,  1.43it/s]
4|5693|Loss: 0.7841958999633789:  88%|████████▊ | 5692/6470 [1:05:27<09:03,  1.43it/s]
4|5693|Loss: 0.7841958999633789:  88%|████████▊ | 5693/6470 [1:05:27<08:40,  1.49it/s]
4|5694|Loss: 0.8778724074363708:  88%|████████▊ | 5693/6470 [1:05:28<08:40,  1.49it/s]
4|5694|Loss: 0.8778724074363708:  88%|████████▊ | 5694/6470 [1:05:28<09:00,  1.44it/s]
4|5695|Loss: 0.8170106410980225:  88%|████████▊ | 5694/6470 [1:05:29<09:00,  1.44it/s]
4|5695|Loss: 0.8170106410980225:  88%|████████▊ | 5695/6470 [1:05:29<09:07,  1.42it/s]
4|5696|Loss: 0.8283505439758301:  88%|████████▊ | 5695/6470 [1:05:29<09:07,  1.42it/s]
4|5696|Loss: 0.8283505439758301:  88%|████████▊ | 5696/6470 [1:05:29<08:19,  1.55it/s]
4|5697|Loss: 0.8365218043327332:  88%|████████▊ | 5696/6470 [1:05:30<08:19,  1.55it/s]
4|5697|Loss: 0.8365218043327332:  88%|████████▊ | 5697/6470 [1:05:30<07:56,  1.62it/s]
4|5698|Loss: 0.8125573396682739:  88%|████████▊ | 5697/6470 [1:05:30<07:56,  1.62it/s]
4|5698|Loss: 0.8125573396682739:  88%|████████▊ | 5698/6470 [1:05:31<07:55,  1.62it/s]
4|5699|Loss: 0.747828483581543:  88%|████████▊ | 5698/6470 [1:05:31<07:55,  1.62it/s] 
4|5699|Loss: 0.747828483581543:  88%|████████▊ | 5699/6470 [1:05:31<08:29,  1.51it/s]
4|5700|Loss: 0.7764890193939209:  88%|████████▊ | 5699/6470 [1:05:32<08:29,  1.51it/s]
4|5700|Loss: 0.7764890193939209:  88%|████████▊ | 5700/6470 [1:05:32<08:27,  1.52it/s]
4|5701|Loss: 0.8455911874771118:  88%|████████▊ | 5700/6470 [1:05:32<08:27,  1.52it/s]
4|5701|Loss: 0.8455911874771118:  88%|████████▊ | 5701/6470 [1:05:33<09:07,  1.41it/s]
4|5702|Loss: 0.7670038938522339:  88%|████████▊ | 5701/6470 [1:05:33<09:07,  1.41it/s]
4|5702|Loss: 0.7670038938522339:  88%|████████▊ | 5702/6470 [1:05:34<09:03,  1.41it/s]
4|5703|Loss: 0.8742368221282959:  88%|████████▊ | 5702/6470 [1:05:34<09:03,  1.41it/s]
4|5703|Loss: 0.8742368221282959:  88%|████████▊ | 5703/6470 [1:05:34<08:48,  1.45it/s]
4|5704|Loss: 0.9636357426643372:  88%|████████▊ | 5703/6470 [1:05:35<08:48,  1.45it/s]
4|5704|Loss: 0.9636357426643372:  88%|████████▊ | 5704/6470 [1:05:35<08:49,  1.45it/s]
4|5705|Loss: 0.798542320728302:  88%|████████▊ | 5704/6470 [1:05:35<08:49,  1.45it/s] 
4|5705|Loss: 0.798542320728302:  88%|████████▊ | 5705/6470 [1:05:36<09:00,  1.42it/s]
4|5706|Loss: 0.7258426547050476:  88%|████████▊ | 5705/6470 [1:05:36<09:00,  1.42it/s]
4|5706|Loss: 0.7258426547050476:  88%|████████▊ | 5706/6470 [1:05:36<09:01,  1.41it/s]
4|5707|Loss: 0.7435672879219055:  88%|████████▊ | 5706/6470 [1:05:37<09:01,  1.41it/s]
4|5707|Loss: 0.7435672879219055:  88%|████████▊ | 5707/6470 [1:05:37<08:59,  1.41it/s]
4|5708|Loss: 0.7373263239860535:  88%|████████▊ | 5707/6470 [1:05:37<08:59,  1.41it/s]
4|5708|Loss: 0.7373263239860535:  88%|████████▊ | 5708/6470 [1:05:38<08:40,  1.46it/s]
4|5709|Loss: 0.8041394352912903:  88%|████████▊ | 5708/6470 [1:05:38<08:40,  1.46it/s]
4|5709|Loss: 0.8041394352912903:  88%|████████▊ | 5709/6470 [1:05:38<08:22,  1.51it/s]
4|5710|Loss: 0.8807421326637268:  88%|████████▊ | 5709/6470 [1:05:39<08:22,  1.51it/s]
4|5710|Loss: 0.8807421326637268:  88%|████████▊ | 5710/6470 [1:05:39<08:54,  1.42it/s]
4|5711|Loss: 0.8864864110946655:  88%|████████▊ | 5710/6470 [1:05:40<08:54,  1.42it/s]
4|5711|Loss: 0.8864864110946655:  88%|████████▊ | 5711/6470 [1:05:40<09:19,  1.36it/s]
4|5712|Loss: 0.8640985488891602:  88%|████████▊ | 5711/6470 [1:05:40<09:19,  1.36it/s]
4|5712|Loss: 0.8640985488891602:  88%|████████▊ | 5712/6470 [1:05:41<09:21,  1.35it/s]
4|5713|Loss: 0.8566646575927734:  88%|████████▊ | 5712/6470 [1:05:41<09:21,  1.35it/s]
4|5713|Loss: 0.8566646575927734:  88%|████████▊ | 5713/6470 [1:05:41<08:50,  1.43it/s]
4|5714|Loss: 0.9835306406021118:  88%|████████▊ | 5713/6470 [1:05:42<08:50,  1.43it/s]
4|5714|Loss: 0.9835306406021118:  88%|████████▊ | 5714/6470 [1:05:42<09:06,  1.38it/s]
4|5715|Loss: 0.8585401177406311:  88%|████████▊ | 5714/6470 [1:05:42<09:06,  1.38it/s]
4|5715|Loss: 0.8585401177406311:  88%|████████▊ | 5715/6470 [1:05:43<08:28,  1.48it/s]
4|5716|Loss: 0.750442385673523:  88%|████████▊ | 5715/6470 [1:05:43<08:28,  1.48it/s] 
4|5716|Loss: 0.750442385673523:  88%|████████▊ | 5716/6470 [1:05:43<09:10,  1.37it/s]
4|5717|Loss: 0.9672996997833252:  88%|████████▊ | 5716/6470 [1:05:44<09:10,  1.37it/s]
4|5717|Loss: 0.9672996997833252:  88%|████████▊ | 5717/6470 [1:05:44<09:01,  1.39it/s]
4|5718|Loss: 0.9893020391464233:  88%|████████▊ | 5717/6470 [1:05:44<09:01,  1.39it/s]
4|5718|Loss: 0.9893020391464233:  88%|████████▊ | 5718/6470 [1:05:45<08:51,  1.41it/s]
4|5719|Loss: 0.8865644335746765:  88%|████████▊ | 5718/6470 [1:05:45<08:51,  1.41it/s]
4|5719|Loss: 0.8865644335746765:  88%|████████▊ | 5719/6470 [1:05:46<08:58,  1.39it/s]
4|5720|Loss: 0.8720780611038208:  88%|████████▊ | 5719/6470 [1:05:46<08:58,  1.39it/s]
4|5720|Loss: 0.8720780611038208:  88%|████████▊ | 5720/6470 [1:05:46<08:38,  1.45it/s]
4|5721|Loss: 0.7457260489463806:  88%|████████▊ | 5720/6470 [1:05:47<08:38,  1.45it/s]
4|5721|Loss: 0.7457260489463806:  88%|████████▊ | 5721/6470 [1:05:47<08:06,  1.54it/s]
4|5722|Loss: 0.7813069820404053:  88%|████████▊ | 5721/6470 [1:05:47<08:06,  1.54it/s]
4|5722|Loss: 0.7813069820404053:  88%|████████▊ | 5722/6470 [1:05:47<08:19,  1.50it/s]
4|5723|Loss: 0.8851341009140015:  88%|████████▊ | 5722/6470 [1:05:48<08:19,  1.50it/s]
4|5723|Loss: 0.8851341009140015:  88%|████████▊ | 5723/6470 [1:05:48<08:35,  1.45it/s]
4|5724|Loss: 0.7881647944450378:  88%|████████▊ | 5723/6470 [1:05:49<08:35,  1.45it/s]
4|5724|Loss: 0.7881647944450378:  88%|████████▊ | 5724/6470 [1:05:49<09:02,  1.38it/s]
4|5725|Loss: 0.6846570372581482:  88%|████████▊ | 5724/6470 [1:05:49<09:02,  1.38it/s]
4|5725|Loss: 0.6846570372581482:  88%|████████▊ | 5725/6470 [1:05:50<09:02,  1.37it/s]
4|5726|Loss: 0.6650670170783997:  88%|████████▊ | 5725/6470 [1:05:50<09:02,  1.37it/s]
4|5726|Loss: 0.6650670170783997:  89%|████████▊ | 5726/6470 [1:05:50<08:55,  1.39it/s]
4|5727|Loss: 0.7764103412628174:  89%|████████▊ | 5726/6470 [1:05:51<08:55,  1.39it/s]
4|5727|Loss: 0.7764103412628174:  89%|████████▊ | 5727/6470 [1:05:51<08:28,  1.46it/s]
4|5728|Loss: 0.8246321082115173:  89%|████████▊ | 5727/6470 [1:05:51<08:28,  1.46it/s]
4|5728|Loss: 0.8246321082115173:  89%|████████▊ | 5728/6470 [1:05:52<08:27,  1.46it/s]
4|5729|Loss: 0.813304603099823:  89%|████████▊ | 5728/6470 [1:05:52<08:27,  1.46it/s] 
4|5729|Loss: 0.813304603099823:  89%|████████▊ | 5729/6470 [1:05:52<08:19,  1.48it/s]
4|5730|Loss: 0.9850856065750122:  89%|████████▊ | 5729/6470 [1:05:53<08:19,  1.48it/s]
4|5730|Loss: 0.9850856065750122:  89%|████████▊ | 5730/6470 [1:05:53<08:29,  1.45it/s]
4|5731|Loss: 0.8778709173202515:  89%|████████▊ | 5730/6470 [1:05:53<08:29,  1.45it/s]
4|5731|Loss: 0.8778709173202515:  89%|████████▊ | 5731/6470 [1:05:54<08:40,  1.42it/s]
4|5732|Loss: 0.8502200245857239:  89%|████████▊ | 5731/6470 [1:05:54<08:40,  1.42it/s]
4|5732|Loss: 0.8502200245857239:  89%|████████▊ | 5732/6470 [1:05:54<08:24,  1.46it/s]
4|5733|Loss: 0.7012243866920471:  89%|████████▊ | 5732/6470 [1:05:55<08:24,  1.46it/s]
4|5733|Loss: 0.7012243866920471:  89%|████████▊ | 5733/6470 [1:05:55<08:09,  1.50it/s]
4|5734|Loss: 0.6990755200386047:  89%|████████▊ | 5733/6470 [1:05:55<08:09,  1.50it/s]
4|5734|Loss: 0.6990755200386047:  89%|████████▊ | 5734/6470 [1:05:56<07:31,  1.63it/s]
4|5735|Loss: 0.8801969289779663:  89%|████████▊ | 5734/6470 [1:05:56<07:31,  1.63it/s]
4|5735|Loss: 0.8801969289779663:  89%|████████▊ | 5735/6470 [1:05:56<07:39,  1.60it/s]
4|5736|Loss: 0.6468503475189209:  89%|████████▊ | 5735/6470 [1:05:57<07:39,  1.60it/s]
4|5736|Loss: 0.6468503475189209:  89%|████████▊ | 5736/6470 [1:05:57<07:41,  1.59it/s]
4|5737|Loss: 0.8049125671386719:  89%|████████▊ | 5736/6470 [1:05:57<07:41,  1.59it/s]
4|5737|Loss: 0.8049125671386719:  89%|████████▊ | 5737/6470 [1:05:57<07:22,  1.66it/s]
4|5738|Loss: 0.7261297702789307:  89%|████████▊ | 5737/6470 [1:05:58<07:22,  1.66it/s]
4|5738|Loss: 0.7261297702789307:  89%|████████▊ | 5738/6470 [1:05:58<07:20,  1.66it/s]
4|5739|Loss: 0.7909930944442749:  89%|████████▊ | 5738/6470 [1:05:58<07:20,  1.66it/s]
4|5739|Loss: 0.7909930944442749:  89%|████████▊ | 5739/6470 [1:05:59<07:47,  1.56it/s]
4|5740|Loss: 0.8260838985443115:  89%|████████▊ | 5739/6470 [1:05:59<07:47,  1.56it/s]
4|5740|Loss: 0.8260838985443115:  89%|████████▊ | 5740/6470 [1:05:59<08:12,  1.48it/s]
4|5741|Loss: 0.8335594534873962:  89%|████████▊ | 5740/6470 [1:06:00<08:12,  1.48it/s]
4|5741|Loss: 0.8335594534873962:  89%|████████▊ | 5741/6470 [1:06:00<08:10,  1.49it/s]
4|5742|Loss: 0.8448888659477234:  89%|████████▊ | 5741/6470 [1:06:01<08:10,  1.49it/s]
4|5742|Loss: 0.8448888659477234:  89%|████████▊ | 5742/6470 [1:06:01<08:12,  1.48it/s]
4|5743|Loss: 0.8948875069618225:  89%|████████▊ | 5742/6470 [1:06:01<08:12,  1.48it/s]
4|5743|Loss: 0.8948875069618225:  89%|████████▉ | 5743/6470 [1:06:02<08:17,  1.46it/s]
4|5744|Loss: 0.901512861251831:  89%|████████▉ | 5743/6470 [1:06:02<08:17,  1.46it/s] 
4|5744|Loss: 0.901512861251831:  89%|████████▉ | 5744/6470 [1:06:02<07:53,  1.53it/s]
4|5745|Loss: 0.9409863352775574:  89%|████████▉ | 5744/6470 [1:06:03<07:53,  1.53it/s]
4|5745|Loss: 0.9409863352775574:  89%|████████▉ | 5745/6470 [1:06:03<08:35,  1.41it/s]
4|5746|Loss: 0.7473952770233154:  89%|████████▉ | 5745/6470 [1:06:03<08:35,  1.41it/s]
4|5746|Loss: 0.7473952770233154:  89%|████████▉ | 5746/6470 [1:06:04<09:09,  1.32it/s]
4|5747|Loss: 0.7546517848968506:  89%|████████▉ | 5746/6470 [1:06:04<09:09,  1.32it/s]
4|5747|Loss: 0.7546517848968506:  89%|████████▉ | 5747/6470 [1:06:05<08:54,  1.35it/s]
4|5748|Loss: 0.8918938040733337:  89%|████████▉ | 5747/6470 [1:06:05<08:54,  1.35it/s]
4|5748|Loss: 0.8918938040733337:  89%|████████▉ | 5748/6470 [1:06:05<08:43,  1.38it/s]
4|5749|Loss: 0.9802494049072266:  89%|████████▉ | 5748/6470 [1:06:06<08:43,  1.38it/s]
4|5749|Loss: 0.9802494049072266:  89%|████████▉ | 5749/6470 [1:06:06<08:08,  1.48it/s]
4|5750|Loss: 0.7795991897583008:  89%|████████▉ | 5749/6470 [1:06:06<08:08,  1.48it/s]
4|5750|Loss: 0.7795991897583008:  89%|████████▉ | 5750/6470 [1:06:07<08:13,  1.46it/s]
4|5751|Loss: 0.8066162467002869:  89%|████████▉ | 5750/6470 [1:06:07<08:13,  1.46it/s]
4|5751|Loss: 0.8066162467002869:  89%|████████▉ | 5751/6470 [1:06:07<08:45,  1.37it/s]
4|5752|Loss: 0.8085744380950928:  89%|████████▉ | 5751/6470 [1:06:08<08:45,  1.37it/s]
4|5752|Loss: 0.8085744380950928:  89%|████████▉ | 5752/6470 [1:06:08<08:35,  1.39it/s]
4|5753|Loss: 0.9286581873893738:  89%|████████▉ | 5752/6470 [1:06:08<08:35,  1.39it/s]
4|5753|Loss: 0.9286581873893738:  89%|████████▉ | 5753/6470 [1:06:09<08:05,  1.48it/s]
4|5754|Loss: 0.8592936396598816:  89%|████████▉ | 5753/6470 [1:06:09<08:05,  1.48it/s]
4|5754|Loss: 0.8592936396598816:  89%|████████▉ | 5754/6470 [1:06:09<07:44,  1.54it/s]
4|5755|Loss: 0.9338645935058594:  89%|████████▉ | 5754/6470 [1:06:10<07:44,  1.54it/s]
4|5755|Loss: 0.9338645935058594:  89%|████████▉ | 5755/6470 [1:06:10<07:24,  1.61it/s]
4|5756|Loss: 0.9899160861968994:  89%|████████▉ | 5755/6470 [1:06:10<07:24,  1.61it/s]
4|5756|Loss: 0.9899160861968994:  89%|████████▉ | 5756/6470 [1:06:10<07:24,  1.61it/s]
4|5757|Loss: 0.8072271347045898:  89%|████████▉ | 5756/6470 [1:06:11<07:24,  1.61it/s]
4|5757|Loss: 0.8072271347045898:  89%|████████▉ | 5757/6470 [1:06:11<07:24,  1.60it/s]
4|5758|Loss: 1.0394954681396484:  89%|████████▉ | 5757/6470 [1:06:11<07:24,  1.60it/s]
4|5758|Loss: 1.0394954681396484:  89%|████████▉ | 5758/6470 [1:06:12<07:42,  1.54it/s]
4|5759|Loss: 0.8476165533065796:  89%|████████▉ | 5758/6470 [1:06:12<07:42,  1.54it/s]
4|5759|Loss: 0.8476165533065796:  89%|████████▉ | 5759/6470 [1:06:12<07:11,  1.65it/s]
4|5760|Loss: 0.6771001219749451:  89%|████████▉ | 5759/6470 [1:06:13<07:11,  1.65it/s]
4|5760|Loss: 0.6771001219749451:  89%|████████▉ | 5760/6470 [1:06:13<07:43,  1.53it/s]
4|5761|Loss: 0.8470649123191833:  89%|████████▉ | 5760/6470 [1:06:13<07:43,  1.53it/s]
4|5761|Loss: 0.8470649123191833:  89%|████████▉ | 5761/6470 [1:06:14<08:24,  1.40it/s]
4|5762|Loss: 0.8083457946777344:  89%|████████▉ | 5761/6470 [1:06:14<08:24,  1.40it/s]
4|5762|Loss: 0.8083457946777344:  89%|████████▉ | 5762/6470 [1:06:15<08:20,  1.42it/s]
4|5763|Loss: 0.640852153301239:  89%|████████▉ | 5762/6470 [1:06:15<08:20,  1.42it/s] 
4|5763|Loss: 0.640852153301239:  89%|████████▉ | 5763/6470 [1:06:15<08:36,  1.37it/s]
4|5764|Loss: 0.7695097923278809:  89%|████████▉ | 5763/6470 [1:06:16<08:36,  1.37it/s]
4|5764|Loss: 0.7695097923278809:  89%|████████▉ | 5764/6470 [1:06:16<08:05,  1.45it/s]
4|5765|Loss: 0.8483077883720398:  89%|████████▉ | 5764/6470 [1:06:16<08:05,  1.45it/s]
4|5765|Loss: 0.8483077883720398:  89%|████████▉ | 5765/6470 [1:06:16<07:30,  1.57it/s]
4|5766|Loss: 0.7359940409660339:  89%|████████▉ | 5765/6470 [1:06:17<07:30,  1.57it/s]
4|5766|Loss: 0.7359940409660339:  89%|████████▉ | 5766/6470 [1:06:17<07:05,  1.65it/s]
4|5767|Loss: 0.782547116279602:  89%|████████▉ | 5766/6470 [1:06:17<07:05,  1.65it/s] 
4|5767|Loss: 0.782547116279602:  89%|████████▉ | 5767/6470 [1:06:17<06:45,  1.74it/s]
4|5768|Loss: 0.8900516033172607:  89%|████████▉ | 5767/6470 [1:06:18<06:45,  1.74it/s]
4|5768|Loss: 0.8900516033172607:  89%|████████▉ | 5768/6470 [1:06:18<07:02,  1.66it/s]
4|5769|Loss: 0.8824855089187622:  89%|████████▉ | 5768/6470 [1:06:18<07:02,  1.66it/s]
4|5769|Loss: 0.8824855089187622:  89%|████████▉ | 5769/6470 [1:06:19<07:03,  1.65it/s]
4|5770|Loss: 0.9714422821998596:  89%|████████▉ | 5769/6470 [1:06:19<07:03,  1.65it/s]
4|5770|Loss: 0.9714422821998596:  89%|████████▉ | 5770/6470 [1:06:19<07:22,  1.58it/s]
4|5771|Loss: 0.8242508172988892:  89%|████████▉ | 5770/6470 [1:06:20<07:22,  1.58it/s]
4|5771|Loss: 0.8242508172988892:  89%|████████▉ | 5771/6470 [1:06:20<07:38,  1.53it/s]
4|5772|Loss: 0.6774401068687439:  89%|████████▉ | 5771/6470 [1:06:21<07:38,  1.53it/s]
4|5772|Loss: 0.6774401068687439:  89%|████████▉ | 5772/6470 [1:06:21<07:51,  1.48it/s]
4|5773|Loss: 0.8210787773132324:  89%|████████▉ | 5772/6470 [1:06:21<07:51,  1.48it/s]
4|5773|Loss: 0.8210787773132324:  89%|████████▉ | 5773/6470 [1:06:22<08:21,  1.39it/s]
4|5774|Loss: 0.8769708871841431:  89%|████████▉ | 5773/6470 [1:06:22<08:21,  1.39it/s]
4|5774|Loss: 0.8769708871841431:  89%|████████▉ | 5774/6470 [1:06:22<08:06,  1.43it/s]
4|5775|Loss: 0.8553553223609924:  89%|████████▉ | 5774/6470 [1:06:23<08:06,  1.43it/s]
4|5775|Loss: 0.8553553223609924:  89%|████████▉ | 5775/6470 [1:06:23<07:50,  1.48it/s]
4|5776|Loss: 0.9282664060592651:  89%|████████▉ | 5775/6470 [1:06:23<07:50,  1.48it/s]
4|5776|Loss: 0.9282664060592651:  89%|████████▉ | 5776/6470 [1:06:24<07:27,  1.55it/s]
4|5777|Loss: 0.8470020294189453:  89%|████████▉ | 5776/6470 [1:06:24<07:27,  1.55it/s]
4|5777|Loss: 0.8470020294189453:  89%|████████▉ | 5777/6470 [1:06:24<07:36,  1.52it/s]
4|5778|Loss: 0.8424216508865356:  89%|████████▉ | 5777/6470 [1:06:25<07:36,  1.52it/s]
4|5778|Loss: 0.8424216508865356:  89%|████████▉ | 5778/6470 [1:06:25<07:32,  1.53it/s]
4|5779|Loss: 0.9088069796562195:  89%|████████▉ | 5778/6470 [1:06:25<07:32,  1.53it/s]
4|5779|Loss: 0.9088069796562195:  89%|████████▉ | 5779/6470 [1:06:26<07:45,  1.48it/s]
4|5780|Loss: 0.9197485446929932:  89%|████████▉ | 5779/6470 [1:06:26<07:45,  1.48it/s]
4|5780|Loss: 0.9197485446929932:  89%|████████▉ | 5780/6470 [1:06:26<07:42,  1.49it/s]
4|5781|Loss: 1.0064297914505005:  89%|████████▉ | 5780/6470 [1:06:27<07:42,  1.49it/s]
4|5781|Loss: 1.0064297914505005:  89%|████████▉ | 5781/6470 [1:06:27<07:46,  1.48it/s]
4|5782|Loss: 0.9864251017570496:  89%|████████▉ | 5781/6470 [1:06:27<07:46,  1.48it/s]
4|5782|Loss: 0.9864251017570496:  89%|████████▉ | 5782/6470 [1:06:28<08:05,  1.42it/s]
4|5783|Loss: 0.934562087059021:  89%|████████▉ | 5782/6470 [1:06:28<08:05,  1.42it/s] 
4|5783|Loss: 0.934562087059021:  89%|████████▉ | 5783/6470 [1:06:28<07:31,  1.52it/s]
4|5784|Loss: 0.7416456937789917:  89%|████████▉ | 5783/6470 [1:06:29<07:31,  1.52it/s]
4|5784|Loss: 0.7416456937789917:  89%|████████▉ | 5784/6470 [1:06:29<07:20,  1.56it/s]
4|5785|Loss: 0.7697302103042603:  89%|████████▉ | 5784/6470 [1:06:29<07:20,  1.56it/s]
4|5785|Loss: 0.7697302103042603:  89%|████████▉ | 5785/6470 [1:06:29<07:08,  1.60it/s]
4|5786|Loss: 0.9136227965354919:  89%|████████▉ | 5785/6470 [1:06:30<07:08,  1.60it/s]
4|5786|Loss: 0.9136227965354919:  89%|████████▉ | 5786/6470 [1:06:30<07:08,  1.59it/s]
4|5787|Loss: 0.9279154539108276:  89%|████████▉ | 5786/6470 [1:06:30<07:08,  1.59it/s]
4|5787|Loss: 0.9279154539108276:  89%|████████▉ | 5787/6470 [1:06:31<07:22,  1.54it/s]
4|5788|Loss: 0.9048105478286743:  89%|████████▉ | 5787/6470 [1:06:31<07:22,  1.54it/s]
4|5788|Loss: 0.9048105478286743:  89%|████████▉ | 5788/6470 [1:06:31<07:10,  1.59it/s]
4|5789|Loss: 0.8963901996612549:  89%|████████▉ | 5788/6470 [1:06:32<07:10,  1.59it/s]
4|5789|Loss: 0.8963901996612549:  89%|████████▉ | 5789/6470 [1:06:32<07:05,  1.60it/s]
4|5790|Loss: 0.7105109095573425:  89%|████████▉ | 5789/6470 [1:06:32<07:05,  1.60it/s]
4|5790|Loss: 0.7105109095573425:  89%|████████▉ | 5790/6470 [1:06:33<07:47,  1.46it/s]
4|5791|Loss: 0.759306013584137:  89%|████████▉ | 5790/6470 [1:06:33<07:47,  1.46it/s] 
4|5791|Loss: 0.759306013584137:  90%|████████▉ | 5791/6470 [1:06:34<08:25,  1.34it/s]
4|5792|Loss: 0.9126802086830139:  90%|████████▉ | 5791/6470 [1:06:34<08:25,  1.34it/s]
4|5792|Loss: 0.9126802086830139:  90%|████████▉ | 5792/6470 [1:06:34<08:12,  1.38it/s]
4|5793|Loss: 1.0218398571014404:  90%|████████▉ | 5792/6470 [1:06:35<08:12,  1.38it/s]
4|5793|Loss: 1.0218398571014404:  90%|████████▉ | 5793/6470 [1:06:35<07:32,  1.50it/s]
4|5794|Loss: 0.8968210816383362:  90%|████████▉ | 5793/6470 [1:06:35<07:32,  1.50it/s]
4|5794|Loss: 0.8968210816383362:  90%|████████▉ | 5794/6470 [1:06:36<07:29,  1.50it/s]
4|5795|Loss: 0.9787809252738953:  90%|████████▉ | 5794/6470 [1:06:36<07:29,  1.50it/s]
4|5795|Loss: 0.9787809252738953:  90%|████████▉ | 5795/6470 [1:06:36<07:51,  1.43it/s]
4|5796|Loss: 0.8444362878799438:  90%|████████▉ | 5795/6470 [1:06:37<07:51,  1.43it/s]
4|5796|Loss: 0.8444362878799438:  90%|████████▉ | 5796/6470 [1:06:37<07:56,  1.42it/s]
4|5797|Loss: 0.735904335975647:  90%|████████▉ | 5796/6470 [1:06:37<07:56,  1.42it/s] 
4|5797|Loss: 0.735904335975647:  90%|████████▉ | 5797/6470 [1:06:38<07:43,  1.45it/s]
4|5798|Loss: 0.7045434713363647:  90%|████████▉ | 5797/6470 [1:06:38<07:43,  1.45it/s]
4|5798|Loss: 0.7045434713363647:  90%|████████▉ | 5798/6470 [1:06:38<08:02,  1.39it/s]
4|5799|Loss: 0.8973232507705688:  90%|████████▉ | 5798/6470 [1:06:39<08:02,  1.39it/s]
4|5799|Loss: 0.8973232507705688:  90%|████████▉ | 5799/6470 [1:06:39<07:41,  1.45it/s]
4|5800|Loss: 0.9068982601165771:  90%|████████▉ | 5799/6470 [1:06:39<07:41,  1.45it/s]
4|5800|Loss: 0.9068982601165771:  90%|████████▉ | 5800/6470 [1:06:40<07:43,  1.45it/s]
4|5801|Loss: 0.8856887817382812:  90%|████████▉ | 5800/6470 [1:06:40<07:43,  1.45it/s]
4|5801|Loss: 0.8856887817382812:  90%|████████▉ | 5801/6470 [1:06:40<07:37,  1.46it/s]
4|5802|Loss: 0.9446250796318054:  90%|████████▉ | 5801/6470 [1:06:41<07:37,  1.46it/s]
4|5802|Loss: 0.9446250796318054:  90%|████████▉ | 5802/6470 [1:06:41<07:43,  1.44it/s]
4|5803|Loss: 0.873088538646698:  90%|████████▉ | 5802/6470 [1:06:42<07:43,  1.44it/s] 
4|5803|Loss: 0.873088538646698:  90%|████████▉ | 5803/6470 [1:06:42<08:06,  1.37it/s]
4|5804|Loss: 0.8654688596725464:  90%|████████▉ | 5803/6470 [1:06:42<08:06,  1.37it/s]
4|5804|Loss: 0.8654688596725464:  90%|████████▉ | 5804/6470 [1:06:43<07:52,  1.41it/s]
4|5805|Loss: 0.8131459951400757:  90%|████████▉ | 5804/6470 [1:06:43<07:52,  1.41it/s]
4|5805|Loss: 0.8131459951400757:  90%|████████▉ | 5805/6470 [1:06:43<07:05,  1.56it/s]
4|5806|Loss: 0.9509456157684326:  90%|████████▉ | 5805/6470 [1:06:43<07:05,  1.56it/s]
4|5806|Loss: 0.9509456157684326:  90%|████████▉ | 5806/6470 [1:06:44<06:41,  1.65it/s]
4|5807|Loss: 0.5554808974266052:  90%|████████▉ | 5806/6470 [1:06:44<06:41,  1.65it/s]
4|5807|Loss: 0.5554808974266052:  90%|████████▉ | 5807/6470 [1:06:44<06:36,  1.67it/s]
4|5808|Loss: 0.9397416114807129:  90%|████████▉ | 5807/6470 [1:06:45<06:36,  1.67it/s]
4|5808|Loss: 0.9397416114807129:  90%|████████▉ | 5808/6470 [1:06:45<06:52,  1.60it/s]
4|5809|Loss: 0.8628865480422974:  90%|████████▉ | 5808/6470 [1:06:45<06:52,  1.60it/s]
4|5809|Loss: 0.8628865480422974:  90%|████████▉ | 5809/6470 [1:06:46<07:08,  1.54it/s]
4|5810|Loss: 0.9986758828163147:  90%|████████▉ | 5809/6470 [1:06:46<07:08,  1.54it/s]
4|5810|Loss: 0.9986758828163147:  90%|████████▉ | 5810/6470 [1:06:46<07:32,  1.46it/s]
4|5811|Loss: 0.7194409370422363:  90%|████████▉ | 5810/6470 [1:06:47<07:32,  1.46it/s]
4|5811|Loss: 0.7194409370422363:  90%|████████▉ | 5811/6470 [1:06:47<07:31,  1.46it/s]
4|5812|Loss: 0.9696739315986633:  90%|████████▉ | 5811/6470 [1:06:47<07:31,  1.46it/s]
4|5812|Loss: 0.9696739315986633:  90%|████████▉ | 5812/6470 [1:06:48<07:32,  1.45it/s]
4|5813|Loss: 0.8631317019462585:  90%|████████▉ | 5812/6470 [1:06:48<07:32,  1.45it/s]
4|5813|Loss: 0.8631317019462585:  90%|████████▉ | 5813/6470 [1:06:49<07:39,  1.43it/s]
4|5814|Loss: 0.9063059091567993:  90%|████████▉ | 5813/6470 [1:06:49<07:39,  1.43it/s]
4|5814|Loss: 0.9063059091567993:  90%|████████▉ | 5814/6470 [1:06:49<07:45,  1.41it/s]
4|5815|Loss: 0.8952885866165161:  90%|████████▉ | 5814/6470 [1:06:50<07:45,  1.41it/s]
4|5815|Loss: 0.8952885866165161:  90%|████████▉ | 5815/6470 [1:06:50<06:50,  1.60it/s]
4|5816|Loss: 0.8636143207550049:  90%|████████▉ | 5815/6470 [1:06:50<06:50,  1.60it/s]
4|5816|Loss: 0.8636143207550049:  90%|████████▉ | 5816/6470 [1:06:50<06:29,  1.68it/s]
4|5817|Loss: 0.837220311164856:  90%|████████▉ | 5816/6470 [1:06:51<06:29,  1.68it/s] 
4|5817|Loss: 0.837220311164856:  90%|████████▉ | 5817/6470 [1:06:51<06:31,  1.67it/s]
4|5818|Loss: 0.8104321956634521:  90%|████████▉ | 5817/6470 [1:06:51<06:31,  1.67it/s]
4|5818|Loss: 0.8104321956634521:  90%|████████▉ | 5818/6470 [1:06:52<07:11,  1.51it/s]
4|5819|Loss: 0.8713356256484985:  90%|████████▉ | 5818/6470 [1:06:52<07:11,  1.51it/s]
4|5819|Loss: 0.8713356256484985:  90%|████████▉ | 5819/6470 [1:06:52<06:47,  1.60it/s]
4|5820|Loss: 0.8598433136940002:  90%|████████▉ | 5819/6470 [1:06:52<06:47,  1.60it/s]
4|5820|Loss: 0.8598433136940002:  90%|████████▉ | 5820/6470 [1:06:53<06:30,  1.67it/s]
4|5821|Loss: 0.7149296402931213:  90%|████████▉ | 5820/6470 [1:06:53<06:30,  1.67it/s]
4|5821|Loss: 0.7149296402931213:  90%|████████▉ | 5821/6470 [1:06:53<06:34,  1.64it/s]
4|5822|Loss: 0.7514413595199585:  90%|████████▉ | 5821/6470 [1:06:54<06:34,  1.64it/s]
4|5822|Loss: 0.7514413595199585:  90%|████████▉ | 5822/6470 [1:06:54<06:50,  1.58it/s]
4|5823|Loss: 0.790042519569397:  90%|████████▉ | 5822/6470 [1:06:54<06:50,  1.58it/s] 
4|5823|Loss: 0.790042519569397:  90%|█████████ | 5823/6470 [1:06:55<06:24,  1.68it/s]
4|5824|Loss: 0.7930729985237122:  90%|█████████ | 5823/6470 [1:06:55<06:24,  1.68it/s]
4|5824|Loss: 0.7930729985237122:  90%|█████████ | 5824/6470 [1:06:55<06:22,  1.69it/s]
4|5825|Loss: 0.9187297821044922:  90%|█████████ | 5824/6470 [1:06:56<06:22,  1.69it/s]
4|5825|Loss: 0.9187297821044922:  90%|█████████ | 5825/6470 [1:06:56<06:59,  1.54it/s]
4|5826|Loss: 0.8963586091995239:  90%|█████████ | 5825/6470 [1:06:56<06:59,  1.54it/s]
4|5826|Loss: 0.8963586091995239:  90%|█████████ | 5826/6470 [1:06:57<07:15,  1.48it/s]
4|5827|Loss: 0.7737133502960205:  90%|█████████ | 5826/6470 [1:06:57<07:15,  1.48it/s]
4|5827|Loss: 0.7737133502960205:  90%|█████████ | 5827/6470 [1:06:57<07:01,  1.53it/s]
4|5828|Loss: 0.870266854763031:  90%|█████████ | 5827/6470 [1:06:58<07:01,  1.53it/s] 
4|5828|Loss: 0.870266854763031:  90%|█████████ | 5828/6470 [1:06:58<07:11,  1.49it/s]
4|5829|Loss: 0.9428573250770569:  90%|█████████ | 5828/6470 [1:06:58<07:11,  1.49it/s]
4|5829|Loss: 0.9428573250770569:  90%|█████████ | 5829/6470 [1:06:59<07:18,  1.46it/s]
4|5830|Loss: 0.8204689621925354:  90%|█████████ | 5829/6470 [1:06:59<07:18,  1.46it/s]
4|5830|Loss: 0.8204689621925354:  90%|█████████ | 5830/6470 [1:06:59<06:26,  1.66it/s]
4|5831|Loss: 0.956851601600647:  90%|█████████ | 5830/6470 [1:06:59<06:26,  1.66it/s] 
4|5831|Loss: 0.956851601600647:  90%|█████████ | 5831/6470 [1:07:00<06:34,  1.62it/s]
4|5832|Loss: 0.6578289270401001:  90%|█████████ | 5831/6470 [1:07:00<06:34,  1.62it/s]
4|5832|Loss: 0.6578289270401001:  90%|█████████ | 5832/6470 [1:07:00<06:24,  1.66it/s]
4|5833|Loss: 0.814663827419281:  90%|█████████ | 5832/6470 [1:07:01<06:24,  1.66it/s] 
4|5833|Loss: 0.814663827419281:  90%|█████████ | 5833/6470 [1:07:01<06:40,  1.59it/s]
4|5834|Loss: 0.7781964540481567:  90%|█████████ | 5833/6470 [1:07:01<06:40,  1.59it/s]
4|5834|Loss: 0.7781964540481567:  90%|█████████ | 5834/6470 [1:07:02<07:01,  1.51it/s]
4|5835|Loss: 0.9083695411682129:  90%|█████████ | 5834/6470 [1:07:02<07:01,  1.51it/s]
4|5835|Loss: 0.9083695411682129:  90%|█████████ | 5835/6470 [1:07:02<06:57,  1.52it/s]
4|5836|Loss: 0.8449520468711853:  90%|█████████ | 5835/6470 [1:07:03<06:57,  1.52it/s]
4|5836|Loss: 0.8449520468711853:  90%|█████████ | 5836/6470 [1:07:03<07:00,  1.51it/s]
4|5837|Loss: 0.7700117230415344:  90%|█████████ | 5836/6470 [1:07:03<07:00,  1.51it/s]
4|5837|Loss: 0.7700117230415344:  90%|█████████ | 5837/6470 [1:07:04<06:29,  1.62it/s]
4|5838|Loss: 0.8416574597358704:  90%|█████████ | 5837/6470 [1:07:04<06:29,  1.62it/s]
4|5838|Loss: 0.8416574597358704:  90%|█████████ | 5838/6470 [1:07:04<06:17,  1.67it/s]
4|5839|Loss: 0.8365164399147034:  90%|█████████ | 5838/6470 [1:07:04<06:17,  1.67it/s]
4|5839|Loss: 0.8365164399147034:  90%|█████████ | 5839/6470 [1:07:05<06:37,  1.59it/s]
4|5840|Loss: 0.895771324634552:  90%|█████████ | 5839/6470 [1:07:05<06:37,  1.59it/s] 
4|5840|Loss: 0.895771324634552:  90%|█████████ | 5840/6470 [1:07:06<06:58,  1.51it/s]
4|5841|Loss: 0.9108067154884338:  90%|█████████ | 5840/6470 [1:07:06<06:58,  1.51it/s]
4|5841|Loss: 0.9108067154884338:  90%|█████████ | 5841/6470 [1:07:06<07:12,  1.45it/s]
4|5842|Loss: 0.8181130290031433:  90%|█████████ | 5841/6470 [1:07:07<07:12,  1.45it/s]
4|5842|Loss: 0.8181130290031433:  90%|█████████ | 5842/6470 [1:07:07<07:22,  1.42it/s]
4|5843|Loss: 0.969720184803009:  90%|█████████ | 5842/6470 [1:07:07<07:22,  1.42it/s] 
4|5843|Loss: 0.969720184803009:  90%|█████████ | 5843/6470 [1:07:08<06:45,  1.54it/s]
4|5844|Loss: 0.7643157839775085:  90%|█████████ | 5843/6470 [1:07:08<06:45,  1.54it/s]
4|5844|Loss: 0.7643157839775085:  90%|█████████ | 5844/6470 [1:07:08<07:11,  1.45it/s]
4|5845|Loss: 0.7471016645431519:  90%|█████████ | 5844/6470 [1:07:09<07:11,  1.45it/s]
4|5845|Loss: 0.7471016645431519:  90%|█████████ | 5845/6470 [1:07:09<07:34,  1.37it/s]
4|5846|Loss: 0.9928527474403381:  90%|█████████ | 5845/6470 [1:07:10<07:34,  1.37it/s]
4|5846|Loss: 0.9928527474403381:  90%|█████████ | 5846/6470 [1:07:10<07:48,  1.33it/s]
4|5847|Loss: 0.9155241847038269:  90%|█████████ | 5846/6470 [1:07:10<07:48,  1.33it/s]
4|5847|Loss: 0.9155241847038269:  90%|█████████ | 5847/6470 [1:07:11<07:25,  1.40it/s]
4|5848|Loss: 0.7746149897575378:  90%|█████████ | 5847/6470 [1:07:11<07:25,  1.40it/s]
4|5848|Loss: 0.7746149897575378:  90%|█████████ | 5848/6470 [1:07:11<06:58,  1.48it/s]
4|5849|Loss: 0.877376914024353:  90%|█████████ | 5848/6470 [1:07:12<06:58,  1.48it/s] 
4|5849|Loss: 0.877376914024353:  90%|█████████ | 5849/6470 [1:07:12<07:25,  1.39it/s]
4|5850|Loss: 0.8207370042800903:  90%|█████████ | 5849/6470 [1:07:12<07:25,  1.39it/s]
4|5850|Loss: 0.8207370042800903:  90%|█████████ | 5850/6470 [1:07:13<07:08,  1.45it/s]
4|5851|Loss: 0.8601794242858887:  90%|█████████ | 5850/6470 [1:07:13<07:08,  1.45it/s]
4|5851|Loss: 0.8601794242858887:  90%|█████████ | 5851/6470 [1:07:13<06:32,  1.58it/s]
4|5852|Loss: 0.7820687890052795:  90%|█████████ | 5851/6470 [1:07:13<06:32,  1.58it/s]
4|5852|Loss: 0.7820687890052795:  90%|█████████ | 5852/6470 [1:07:14<06:35,  1.56it/s]
4|5853|Loss: 0.894254744052887:  90%|█████████ | 5852/6470 [1:07:14<06:35,  1.56it/s] 
4|5853|Loss: 0.894254744052887:  90%|█████████ | 5853/6470 [1:07:14<06:34,  1.56it/s]
4|5854|Loss: 0.884554386138916:  90%|█████████ | 5853/6470 [1:07:15<06:34,  1.56it/s]
4|5854|Loss: 0.884554386138916:  90%|█████████ | 5854/6470 [1:07:15<06:48,  1.51it/s]
4|5855|Loss: 0.7240207195281982:  90%|█████████ | 5854/6470 [1:07:16<06:48,  1.51it/s]
4|5855|Loss: 0.7240207195281982:  90%|█████████ | 5855/6470 [1:07:16<07:18,  1.40it/s]
4|5856|Loss: 0.8551921844482422:  90%|█████████ | 5855/6470 [1:07:16<07:18,  1.40it/s]
4|5856|Loss: 0.8551921844482422:  91%|█████████ | 5856/6470 [1:07:17<07:34,  1.35it/s]
4|5857|Loss: 0.8735833168029785:  91%|█████████ | 5856/6470 [1:07:17<07:34,  1.35it/s]
4|5857|Loss: 0.8735833168029785:  91%|█████████ | 5857/6470 [1:07:18<07:55,  1.29it/s]
4|5858|Loss: 0.895144522190094:  91%|█████████ | 5857/6470 [1:07:18<07:55,  1.29it/s] 
4|5858|Loss: 0.895144522190094:  91%|█████████ | 5858/6470 [1:07:18<07:49,  1.30it/s]
4|5859|Loss: 0.8353539705276489:  91%|█████████ | 5858/6470 [1:07:19<07:49,  1.30it/s]
4|5859|Loss: 0.8353539705276489:  91%|█████████ | 5859/6470 [1:07:19<07:41,  1.32it/s]
4|5860|Loss: 0.8046647310256958:  91%|█████████ | 5859/6470 [1:07:19<07:41,  1.32it/s]
4|5860|Loss: 0.8046647310256958:  91%|█████████ | 5860/6470 [1:07:20<06:59,  1.45it/s]
4|5861|Loss: 0.6692799925804138:  91%|█████████ | 5860/6470 [1:07:20<06:59,  1.45it/s]
4|5861|Loss: 0.6692799925804138:  91%|█████████ | 5861/6470 [1:07:20<06:16,  1.62it/s]
4|5862|Loss: 0.798510730266571:  91%|█████████ | 5861/6470 [1:07:20<06:16,  1.62it/s] 
4|5862|Loss: 0.798510730266571:  91%|█████████ | 5862/6470 [1:07:21<06:25,  1.58it/s]
4|5863|Loss: 1.0208570957183838:  91%|█████████ | 5862/6470 [1:07:21<06:25,  1.58it/s]
4|5863|Loss: 1.0208570957183838:  91%|█████████ | 5863/6470 [1:07:21<06:38,  1.52it/s]
4|5864|Loss: 0.8852779269218445:  91%|█████████ | 5863/6470 [1:07:22<06:38,  1.52it/s]
4|5864|Loss: 0.8852779269218445:  91%|█████████ | 5864/6470 [1:07:22<06:29,  1.56it/s]
4|5865|Loss: 0.881401002407074:  91%|█████████ | 5864/6470 [1:07:22<06:29,  1.56it/s] 
4|5865|Loss: 0.881401002407074:  91%|█████████ | 5865/6470 [1:07:23<06:25,  1.57it/s]
4|5866|Loss: 0.8852307200431824:  91%|█████████ | 5865/6470 [1:07:23<06:25,  1.57it/s]
4|5866|Loss: 0.8852307200431824:  91%|█████████ | 5866/6470 [1:07:24<06:54,  1.46it/s]
4|5867|Loss: 0.7909078598022461:  91%|█████████ | 5866/6470 [1:07:24<06:54,  1.46it/s]
4|5867|Loss: 0.7909078598022461:  91%|█████████ | 5867/6470 [1:07:24<06:50,  1.47it/s]
4|5868|Loss: 0.8707373738288879:  91%|█████████ | 5867/6470 [1:07:25<06:50,  1.47it/s]
4|5868|Loss: 0.8707373738288879:  91%|█████████ | 5868/6470 [1:07:25<06:41,  1.50it/s]
4|5869|Loss: 0.7755657434463501:  91%|█████████ | 5868/6470 [1:07:25<06:41,  1.50it/s]
4|5869|Loss: 0.7755657434463501:  91%|█████████ | 5869/6470 [1:07:26<06:46,  1.48it/s]
4|5870|Loss: 0.8630641102790833:  91%|█████████ | 5869/6470 [1:07:26<06:46,  1.48it/s]
4|5870|Loss: 0.8630641102790833:  91%|█████████ | 5870/6470 [1:07:26<06:21,  1.57it/s]
4|5871|Loss: 0.8789224624633789:  91%|█████████ | 5870/6470 [1:07:26<06:21,  1.57it/s]
4|5871|Loss: 0.8789224624633789:  91%|█████████ | 5871/6470 [1:07:27<06:25,  1.55it/s]
4|5872|Loss: 0.769707441329956:  91%|█████████ | 5871/6470 [1:07:27<06:25,  1.55it/s] 
4|5872|Loss: 0.769707441329956:  91%|█████████ | 5872/6470 [1:07:27<06:36,  1.51it/s]
4|5873|Loss: 0.8225814700126648:  91%|█████████ | 5872/6470 [1:07:28<06:36,  1.51it/s]
4|5873|Loss: 0.8225814700126648:  91%|█████████ | 5873/6470 [1:07:28<06:55,  1.44it/s]
4|5874|Loss: 0.7532844543457031:  91%|█████████ | 5873/6470 [1:07:29<06:55,  1.44it/s]
4|5874|Loss: 0.7532844543457031:  91%|█████████ | 5874/6470 [1:07:29<07:07,  1.40it/s]
4|5875|Loss: 0.8114569783210754:  91%|█████████ | 5874/6470 [1:07:29<07:07,  1.40it/s]
4|5875|Loss: 0.8114569783210754:  91%|█████████ | 5875/6470 [1:07:30<06:54,  1.44it/s]
4|5876|Loss: 0.8483425378799438:  91%|█████████ | 5875/6470 [1:07:30<06:54,  1.44it/s]
4|5876|Loss: 0.8483425378799438:  91%|█████████ | 5876/6470 [1:07:30<07:03,  1.40it/s]
4|5877|Loss: 0.8844137787818909:  91%|█████████ | 5876/6470 [1:07:31<07:03,  1.40it/s]
4|5877|Loss: 0.8844137787818909:  91%|█████████ | 5877/6470 [1:07:31<07:05,  1.39it/s]
4|5878|Loss: 0.8090977668762207:  91%|█████████ | 5877/6470 [1:07:31<07:05,  1.39it/s]
4|5878|Loss: 0.8090977668762207:  91%|█████████ | 5878/6470 [1:07:32<06:34,  1.50it/s]
4|5879|Loss: 1.0056145191192627:  91%|█████████ | 5878/6470 [1:07:32<06:34,  1.50it/s]
4|5879|Loss: 1.0056145191192627:  91%|█████████ | 5879/6470 [1:07:32<06:32,  1.50it/s]
4|5880|Loss: 0.8172983527183533:  91%|█████████ | 5879/6470 [1:07:33<06:32,  1.50it/s]
4|5880|Loss: 0.8172983527183533:  91%|█████████ | 5880/6470 [1:07:33<06:51,  1.43it/s]
4|5881|Loss: 0.9104998111724854:  91%|█████████ | 5880/6470 [1:07:33<06:51,  1.43it/s]
4|5881|Loss: 0.9104998111724854:  91%|█████████ | 5881/6470 [1:07:34<06:55,  1.42it/s]
4|5882|Loss: 0.7906849384307861:  91%|█████████ | 5881/6470 [1:07:34<06:55,  1.42it/s]
4|5882|Loss: 0.7906849384307861:  91%|█████████ | 5882/6470 [1:07:35<07:16,  1.35it/s]
4|5883|Loss: 0.8360736966133118:  91%|█████████ | 5882/6470 [1:07:35<07:16,  1.35it/s]
4|5883|Loss: 0.8360736966133118:  91%|█████████ | 5883/6470 [1:07:35<07:19,  1.34it/s]
4|5884|Loss: 0.832545816898346:  91%|█████████ | 5883/6470 [1:07:36<07:19,  1.34it/s] 
4|5884|Loss: 0.832545816898346:  91%|█████████ | 5884/6470 [1:07:36<07:35,  1.29it/s]
4|5885|Loss: 0.9469177722930908:  91%|█████████ | 5884/6470 [1:07:37<07:35,  1.29it/s]
4|5885|Loss: 0.9469177722930908:  91%|█████████ | 5885/6470 [1:07:37<07:10,  1.36it/s]
4|5886|Loss: 0.8021740913391113:  91%|█████████ | 5885/6470 [1:07:37<07:10,  1.36it/s]
4|5886|Loss: 0.8021740913391113:  91%|█████████ | 5886/6470 [1:07:37<06:47,  1.43it/s]
4|5887|Loss: 0.7913456559181213:  91%|█████████ | 5886/6470 [1:07:38<06:47,  1.43it/s]
4|5887|Loss: 0.7913456559181213:  91%|█████████ | 5887/6470 [1:07:38<06:26,  1.51it/s]
4|5888|Loss: 0.9544277787208557:  91%|█████████ | 5887/6470 [1:07:38<06:26,  1.51it/s]
4|5888|Loss: 0.9544277787208557:  91%|█████████ | 5888/6470 [1:07:39<06:35,  1.47it/s]
4|5889|Loss: 0.7715787887573242:  91%|█████████ | 5888/6470 [1:07:39<06:35,  1.47it/s]
4|5889|Loss: 0.7715787887573242:  91%|█████████ | 5889/6470 [1:07:39<06:40,  1.45it/s]
4|5890|Loss: 0.8862987160682678:  91%|█████████ | 5889/6470 [1:07:40<06:40,  1.45it/s]
4|5890|Loss: 0.8862987160682678:  91%|█████████ | 5890/6470 [1:07:40<06:20,  1.52it/s]
4|5891|Loss: 0.985813558101654:  91%|█████████ | 5890/6470 [1:07:40<06:20,  1.52it/s] 
4|5891|Loss: 0.985813558101654:  91%|█████████ | 5891/6470 [1:07:41<06:11,  1.56it/s]
4|5892|Loss: 0.8415957093238831:  91%|█████████ | 5891/6470 [1:07:41<06:11,  1.56it/s]
4|5892|Loss: 0.8415957093238831:  91%|█████████ | 5892/6470 [1:07:41<06:00,  1.60it/s]
4|5893|Loss: 0.78767329454422:  91%|█████████ | 5892/6470 [1:07:42<06:00,  1.60it/s]  
4|5893|Loss: 0.78767329454422:  91%|█████████ | 5893/6470 [1:07:42<06:34,  1.46it/s]
4|5894|Loss: 0.7409297227859497:  91%|█████████ | 5893/6470 [1:07:43<06:34,  1.46it/s]
4|5894|Loss: 0.7409297227859497:  91%|█████████ | 5894/6470 [1:07:43<07:01,  1.37it/s]
4|5895|Loss: 0.8893964886665344:  91%|█████████ | 5894/6470 [1:07:43<07:01,  1.37it/s]
4|5895|Loss: 0.8893964886665344:  91%|█████████ | 5895/6470 [1:07:43<06:17,  1.53it/s]
4|5896|Loss: 1.0182722806930542:  91%|█████████ | 5895/6470 [1:07:44<06:17,  1.53it/s]
4|5896|Loss: 1.0182722806930542:  91%|█████████ | 5896/6470 [1:07:44<06:40,  1.43it/s]
4|5897|Loss: 0.8340421915054321:  91%|█████████ | 5896/6470 [1:07:45<06:40,  1.43it/s]
4|5897|Loss: 0.8340421915054321:  91%|█████████ | 5897/6470 [1:07:45<06:40,  1.43it/s]
4|5898|Loss: 0.7525167465209961:  91%|█████████ | 5897/6470 [1:07:45<06:40,  1.43it/s]
4|5898|Loss: 0.7525167465209961:  91%|█████████ | 5898/6470 [1:07:45<06:03,  1.57it/s]
4|5899|Loss: 0.8646316528320312:  91%|█████████ | 5898/6470 [1:07:46<06:03,  1.57it/s]
4|5899|Loss: 0.8646316528320312:  91%|█████████ | 5899/6470 [1:07:46<05:40,  1.67it/s]
4|5900|Loss: 0.9710851907730103:  91%|█████████ | 5899/6470 [1:07:46<05:40,  1.67it/s]
4|5900|Loss: 0.9710851907730103:  91%|█████████ | 5900/6470 [1:07:47<05:44,  1.66it/s]
4|5901|Loss: 0.8761318325996399:  91%|█████████ | 5900/6470 [1:07:47<05:44,  1.66it/s]
4|5901|Loss: 0.8761318325996399:  91%|█████████ | 5901/6470 [1:07:47<05:49,  1.63it/s]
4|5902|Loss: 0.8966166973114014:  91%|█████████ | 5901/6470 [1:07:48<05:49,  1.63it/s]
4|5902|Loss: 0.8966166973114014:  91%|█████████ | 5902/6470 [1:07:48<06:22,  1.49it/s]
4|5903|Loss: 0.8167960047721863:  91%|█████████ | 5902/6470 [1:07:48<06:22,  1.49it/s]
4|5903|Loss: 0.8167960047721863:  91%|█████████ | 5903/6470 [1:07:49<06:31,  1.45it/s]
4|5904|Loss: 0.7938125729560852:  91%|█████████ | 5903/6470 [1:07:49<06:31,  1.45it/s]
4|5904|Loss: 0.7938125729560852:  91%|█████████▏| 5904/6470 [1:07:49<06:37,  1.42it/s]
4|5905|Loss: 0.8500313758850098:  91%|█████████▏| 5904/6470 [1:07:50<06:37,  1.42it/s]
4|5905|Loss: 0.8500313758850098:  91%|█████████▏| 5905/6470 [1:07:50<06:37,  1.42it/s]
4|5906|Loss: 0.8791595697402954:  91%|█████████▏| 5905/6470 [1:07:51<06:37,  1.42it/s]
4|5906|Loss: 0.8791595697402954:  91%|█████████▏| 5906/6470 [1:07:51<06:53,  1.36it/s]
4|5907|Loss: 0.751883327960968:  91%|█████████▏| 5906/6470 [1:07:51<06:53,  1.36it/s] 
4|5907|Loss: 0.751883327960968:  91%|█████████▏| 5907/6470 [1:07:52<07:17,  1.29it/s]
4|5908|Loss: 0.9931353330612183:  91%|█████████▏| 5907/6470 [1:07:52<07:17,  1.29it/s]
4|5908|Loss: 0.9931353330612183:  91%|█████████▏| 5908/6470 [1:07:53<07:27,  1.25it/s]
4|5909|Loss: 0.8234349489212036:  91%|█████████▏| 5908/6470 [1:07:53<07:27,  1.25it/s]
4|5909|Loss: 0.8234349489212036:  91%|█████████▏| 5909/6470 [1:07:53<07:21,  1.27it/s]
4|5910|Loss: 0.7355732917785645:  91%|█████████▏| 5909/6470 [1:07:54<07:21,  1.27it/s]
4|5910|Loss: 0.7355732917785645:  91%|█████████▏| 5910/6470 [1:07:54<07:00,  1.33it/s]
4|5911|Loss: 0.8888057470321655:  91%|█████████▏| 5910/6470 [1:07:54<07:00,  1.33it/s]
4|5911|Loss: 0.8888057470321655:  91%|█████████▏| 5911/6470 [1:07:55<07:01,  1.32it/s]
4|5912|Loss: 0.7579474449157715:  91%|█████████▏| 5911/6470 [1:07:55<07:01,  1.32it/s]
4|5912|Loss: 0.7579474449157715:  91%|█████████▏| 5912/6470 [1:07:55<06:30,  1.43it/s]
4|5913|Loss: 0.7752354741096497:  91%|█████████▏| 5912/6470 [1:07:56<06:30,  1.43it/s]
4|5913|Loss: 0.7752354741096497:  91%|█████████▏| 5913/6470 [1:07:56<06:28,  1.43it/s]
4|5914|Loss: 0.7998836040496826:  91%|█████████▏| 5913/6470 [1:07:57<06:28,  1.43it/s]
4|5914|Loss: 0.7998836040496826:  91%|█████████▏| 5914/6470 [1:07:57<06:36,  1.40it/s]
4|5915|Loss: 0.8134136199951172:  91%|█████████▏| 5914/6470 [1:07:57<06:36,  1.40it/s]
4|5915|Loss: 0.8134136199951172:  91%|█████████▏| 5915/6470 [1:07:58<06:37,  1.40it/s]
4|5916|Loss: 0.8570687770843506:  91%|█████████▏| 5915/6470 [1:07:58<06:37,  1.40it/s]
4|5916|Loss: 0.8570687770843506:  91%|█████████▏| 5916/6470 [1:07:58<06:40,  1.38it/s]
4|5917|Loss: 0.8569145202636719:  91%|█████████▏| 5916/6470 [1:07:59<06:40,  1.38it/s]
4|5917|Loss: 0.8569145202636719:  91%|█████████▏| 5917/6470 [1:07:59<06:12,  1.48it/s]
4|5918|Loss: 0.7899817228317261:  91%|█████████▏| 5917/6470 [1:07:59<06:12,  1.48it/s]
4|5918|Loss: 0.7899817228317261:  91%|█████████▏| 5918/6470 [1:08:00<06:17,  1.46it/s]
4|5919|Loss: 0.8880482316017151:  91%|█████████▏| 5918/6470 [1:08:00<06:17,  1.46it/s]
4|5919|Loss: 0.8880482316017151:  91%|█████████▏| 5919/6470 [1:08:00<06:46,  1.35it/s]
4|5920|Loss: 0.9085546135902405:  91%|█████████▏| 5919/6470 [1:08:01<06:46,  1.35it/s]
4|5920|Loss: 0.9085546135902405:  91%|█████████▏| 5920/6470 [1:08:01<06:25,  1.43it/s]
4|5921|Loss: 0.8713395595550537:  91%|█████████▏| 5920/6470 [1:08:01<06:25,  1.43it/s]
4|5921|Loss: 0.8713395595550537:  92%|█████████▏| 5921/6470 [1:08:02<06:19,  1.45it/s]
4|5922|Loss: 0.6590200662612915:  92%|█████████▏| 5921/6470 [1:08:02<06:19,  1.45it/s]
4|5922|Loss: 0.6590200662612915:  92%|█████████▏| 5922/6470 [1:08:02<06:21,  1.44it/s]
4|5923|Loss: 1.068518042564392:  92%|█████████▏| 5922/6470 [1:08:03<06:21,  1.44it/s] 
4|5923|Loss: 1.068518042564392:  92%|█████████▏| 5923/6470 [1:08:03<06:58,  1.31it/s]
4|5924|Loss: 0.8046746253967285:  92%|█████████▏| 5923/6470 [1:08:04<06:58,  1.31it/s]
4|5924|Loss: 0.8046746253967285:  92%|█████████▏| 5924/6470 [1:08:04<06:51,  1.33it/s]
4|5925|Loss: 0.8217115998268127:  92%|█████████▏| 5924/6470 [1:08:04<06:51,  1.33it/s]
4|5925|Loss: 0.8217115998268127:  92%|█████████▏| 5925/6470 [1:08:05<06:29,  1.40it/s]
4|5926|Loss: 0.9392023682594299:  92%|█████████▏| 5925/6470 [1:08:05<06:29,  1.40it/s]
4|5926|Loss: 0.9392023682594299:  92%|█████████▏| 5926/6470 [1:08:05<06:07,  1.48it/s]
4|5927|Loss: 0.7827367186546326:  92%|█████████▏| 5926/6470 [1:08:06<06:07,  1.48it/s]
4|5927|Loss: 0.7827367186546326:  92%|█████████▏| 5927/6470 [1:08:06<06:12,  1.46it/s]
4|5928|Loss: 0.7958965301513672:  92%|█████████▏| 5927/6470 [1:08:06<06:12,  1.46it/s]
4|5928|Loss: 0.7958965301513672:  92%|█████████▏| 5928/6470 [1:08:07<06:07,  1.48it/s]
4|5929|Loss: 0.787585973739624:  92%|█████████▏| 5928/6470 [1:08:07<06:07,  1.48it/s] 
4|5929|Loss: 0.787585973739624:  92%|█████████▏| 5929/6470 [1:08:07<06:28,  1.39it/s]
4|5930|Loss: 0.7784920930862427:  92%|█████████▏| 5929/6470 [1:08:08<06:28,  1.39it/s]
4|5930|Loss: 0.7784920930862427:  92%|█████████▏| 5930/6470 [1:08:08<06:34,  1.37it/s]
4|5931|Loss: 0.8318185806274414:  92%|█████████▏| 5930/6470 [1:08:09<06:34,  1.37it/s]
4|5931|Loss: 0.8318185806274414:  92%|█████████▏| 5931/6470 [1:08:09<06:50,  1.31it/s]
4|5932|Loss: 0.9175713062286377:  92%|█████████▏| 5931/6470 [1:08:10<06:50,  1.31it/s]
4|5932|Loss: 0.9175713062286377:  92%|█████████▏| 5932/6470 [1:08:10<06:58,  1.28it/s]
4|5933|Loss: 0.6738778352737427:  92%|█████████▏| 5932/6470 [1:08:10<06:58,  1.28it/s]
4|5933|Loss: 0.6738778352737427:  92%|█████████▏| 5933/6470 [1:08:11<06:52,  1.30it/s]
4|5934|Loss: 0.8999945521354675:  92%|█████████▏| 5933/6470 [1:08:11<06:52,  1.30it/s]
4|5934|Loss: 0.8999945521354675:  92%|█████████▏| 5934/6470 [1:08:11<06:48,  1.31it/s]
4|5935|Loss: 0.9111395478248596:  92%|█████████▏| 5934/6470 [1:08:12<06:48,  1.31it/s]
4|5935|Loss: 0.9111395478248596:  92%|█████████▏| 5935/6470 [1:08:12<07:07,  1.25it/s]
4|5936|Loss: 0.8399041295051575:  92%|█████████▏| 5935/6470 [1:08:13<07:07,  1.25it/s]
4|5936|Loss: 0.8399041295051575:  92%|█████████▏| 5936/6470 [1:08:13<07:18,  1.22it/s]
4|5937|Loss: 0.8353222012519836:  92%|█████████▏| 5936/6470 [1:08:14<07:18,  1.22it/s]
4|5937|Loss: 0.8353222012519836:  92%|█████████▏| 5937/6470 [1:08:14<06:56,  1.28it/s]
4|5938|Loss: 0.9189784526824951:  92%|█████████▏| 5937/6470 [1:08:14<06:56,  1.28it/s]
4|5938|Loss: 0.9189784526824951:  92%|█████████▏| 5938/6470 [1:08:15<06:57,  1.27it/s]
4|5939|Loss: 0.8693698048591614:  92%|█████████▏| 5938/6470 [1:08:15<06:57,  1.27it/s]
4|5939|Loss: 0.8693698048591614:  92%|█████████▏| 5939/6470 [1:08:15<06:53,  1.28it/s]
4|5940|Loss: 0.7869265675544739:  92%|█████████▏| 5939/6470 [1:08:16<06:53,  1.28it/s]
4|5940|Loss: 0.7869265675544739:  92%|█████████▏| 5940/6470 [1:08:16<06:49,  1.30it/s]
4|5941|Loss: 0.8182691335678101:  92%|█████████▏| 5940/6470 [1:08:17<06:49,  1.30it/s]
4|5941|Loss: 0.8182691335678101:  92%|█████████▏| 5941/6470 [1:08:17<06:28,  1.36it/s]
4|5942|Loss: 0.8743159174919128:  92%|█████████▏| 5941/6470 [1:08:17<06:28,  1.36it/s]
4|5942|Loss: 0.8743159174919128:  92%|█████████▏| 5942/6470 [1:08:17<05:37,  1.57it/s]
4|5943|Loss: 0.8493452072143555:  92%|█████████▏| 5942/6470 [1:08:18<05:37,  1.57it/s]
4|5943|Loss: 0.8493452072143555:  92%|█████████▏| 5943/6470 [1:08:18<05:37,  1.56it/s]
4|5944|Loss: 0.7690730690956116:  92%|█████████▏| 5943/6470 [1:08:18<05:37,  1.56it/s]
4|5944|Loss: 0.7690730690956116:  92%|█████████▏| 5944/6470 [1:08:18<05:06,  1.72it/s]
4|5945|Loss: 0.9478254914283752:  92%|█████████▏| 5944/6470 [1:08:19<05:06,  1.72it/s]
4|5945|Loss: 0.9478254914283752:  92%|█████████▏| 5945/6470 [1:08:19<05:20,  1.64it/s]
4|5946|Loss: 0.8484398722648621:  92%|█████████▏| 5945/6470 [1:08:19<05:20,  1.64it/s]
4|5946|Loss: 0.8484398722648621:  92%|█████████▏| 5946/6470 [1:08:20<05:48,  1.50it/s]
4|5947|Loss: 0.9097622632980347:  92%|█████████▏| 5946/6470 [1:08:20<05:48,  1.50it/s]
4|5947|Loss: 0.9097622632980347:  92%|█████████▏| 5947/6470 [1:08:21<06:13,  1.40it/s]
4|5948|Loss: 0.8139669299125671:  92%|█████████▏| 5947/6470 [1:08:21<06:13,  1.40it/s]
4|5948|Loss: 0.8139669299125671:  92%|█████████▏| 5948/6470 [1:08:21<06:15,  1.39it/s]
4|5949|Loss: 0.9066190719604492:  92%|█████████▏| 5948/6470 [1:08:22<06:15,  1.39it/s]
4|5949|Loss: 0.9066190719604492:  92%|█████████▏| 5949/6470 [1:08:22<05:56,  1.46it/s]
4|5950|Loss: 0.8689008355140686:  92%|█████████▏| 5949/6470 [1:08:22<05:56,  1.46it/s]
4|5950|Loss: 0.8689008355140686:  92%|█████████▏| 5950/6470 [1:08:23<06:02,  1.44it/s]
4|5951|Loss: 0.7433667778968811:  92%|█████████▏| 5950/6470 [1:08:23<06:02,  1.44it/s]
4|5951|Loss: 0.7433667778968811:  92%|█████████▏| 5951/6470 [1:08:23<06:05,  1.42it/s]
4|5952|Loss: 0.8280028700828552:  92%|█████████▏| 5951/6470 [1:08:24<06:05,  1.42it/s]
4|5952|Loss: 0.8280028700828552:  92%|█████████▏| 5952/6470 [1:08:24<06:05,  1.42it/s]
4|5953|Loss: 0.9001571536064148:  92%|█████████▏| 5952/6470 [1:08:24<06:05,  1.42it/s]
4|5953|Loss: 0.9001571536064148:  92%|█████████▏| 5953/6470 [1:08:25<06:03,  1.42it/s]
4|5954|Loss: 0.9360631704330444:  92%|█████████▏| 5953/6470 [1:08:25<06:03,  1.42it/s]
4|5954|Loss: 0.9360631704330444:  92%|█████████▏| 5954/6470 [1:08:26<06:08,  1.40it/s]
4|5955|Loss: 0.7441890835762024:  92%|█████████▏| 5954/6470 [1:08:26<06:08,  1.40it/s]
4|5955|Loss: 0.7441890835762024:  92%|█████████▏| 5955/6470 [1:08:26<06:11,  1.38it/s]
4|5956|Loss: 0.6839708089828491:  92%|█████████▏| 5955/6470 [1:08:27<06:11,  1.38it/s]
4|5956|Loss: 0.6839708089828491:  92%|█████████▏| 5956/6470 [1:08:27<05:45,  1.49it/s]
4|5957|Loss: 0.7826859951019287:  92%|█████████▏| 5956/6470 [1:08:27<05:45,  1.49it/s]
4|5957|Loss: 0.7826859951019287:  92%|█████████▏| 5957/6470 [1:08:28<05:47,  1.47it/s]
4|5958|Loss: 0.7963674068450928:  92%|█████████▏| 5957/6470 [1:08:28<05:47,  1.47it/s]
4|5958|Loss: 0.7963674068450928:  92%|█████████▏| 5958/6470 [1:08:28<05:25,  1.57it/s]
4|5959|Loss: 0.9574162364006042:  92%|█████████▏| 5958/6470 [1:08:28<05:25,  1.57it/s]
4|5959|Loss: 0.9574162364006042:  92%|█████████▏| 5959/6470 [1:08:29<05:30,  1.54it/s]
4|5960|Loss: 0.9606701731681824:  92%|█████████▏| 5959/6470 [1:08:29<05:30,  1.54it/s]
4|5960|Loss: 0.9606701731681824:  92%|█████████▏| 5960/6470 [1:08:29<05:40,  1.50it/s]
4|5961|Loss: 0.9552485942840576:  92%|█████████▏| 5960/6470 [1:08:30<05:40,  1.50it/s]
4|5961|Loss: 0.9552485942840576:  92%|█████████▏| 5961/6470 [1:08:30<05:49,  1.46it/s]
4|5962|Loss: 0.8169940710067749:  92%|█████████▏| 5961/6470 [1:08:31<05:49,  1.46it/s]
4|5962|Loss: 0.8169940710067749:  92%|█████████▏| 5962/6470 [1:08:31<05:59,  1.41it/s]
4|5963|Loss: 0.7546512484550476:  92%|█████████▏| 5962/6470 [1:08:31<05:59,  1.41it/s]
4|5963|Loss: 0.7546512484550476:  92%|█████████▏| 5963/6470 [1:08:32<06:10,  1.37it/s]
4|5964|Loss: 0.7638089656829834:  92%|█████████▏| 5963/6470 [1:08:32<06:10,  1.37it/s]
4|5964|Loss: 0.7638089656829834:  92%|█████████▏| 5964/6470 [1:08:32<06:14,  1.35it/s]
4|5965|Loss: 0.7185320258140564:  92%|█████████▏| 5964/6470 [1:08:33<06:14,  1.35it/s]
4|5965|Loss: 0.7185320258140564:  92%|█████████▏| 5965/6470 [1:08:33<06:10,  1.36it/s]
4|5966|Loss: 0.8494575619697571:  92%|█████████▏| 5965/6470 [1:08:34<06:10,  1.36it/s]
4|5966|Loss: 0.8494575619697571:  92%|█████████▏| 5966/6470 [1:08:34<05:46,  1.45it/s]
4|5967|Loss: 0.8722947835922241:  92%|█████████▏| 5966/6470 [1:08:34<05:46,  1.45it/s]
4|5967|Loss: 0.8722947835922241:  92%|█████████▏| 5967/6470 [1:08:34<05:36,  1.49it/s]
4|5968|Loss: 0.8098838329315186:  92%|█████████▏| 5967/6470 [1:08:35<05:36,  1.49it/s]
4|5968|Loss: 0.8098838329315186:  92%|█████████▏| 5968/6470 [1:08:35<05:32,  1.51it/s]
4|5969|Loss: 0.8678873777389526:  92%|█████████▏| 5968/6470 [1:08:35<05:32,  1.51it/s]
4|5969|Loss: 0.8678873777389526:  92%|█████████▏| 5969/6470 [1:08:36<05:35,  1.50it/s]
4|5970|Loss: 0.8462567925453186:  92%|█████████▏| 5969/6470 [1:08:36<05:35,  1.50it/s]
4|5970|Loss: 0.8462567925453186:  92%|█████████▏| 5970/6470 [1:08:36<05:32,  1.51it/s]
4|5971|Loss: 0.883394718170166:  92%|█████████▏| 5970/6470 [1:08:37<05:32,  1.51it/s] 
4|5971|Loss: 0.883394718170166:  92%|█████████▏| 5971/6470 [1:08:37<05:25,  1.54it/s]
4|5972|Loss: 0.9812776446342468:  92%|█████████▏| 5971/6470 [1:08:37<05:25,  1.54it/s]
4|5972|Loss: 0.9812776446342468:  92%|█████████▏| 5972/6470 [1:08:38<05:28,  1.51it/s]
4|5973|Loss: 0.8985892534255981:  92%|█████████▏| 5972/6470 [1:08:38<05:28,  1.51it/s]
4|5973|Loss: 0.8985892534255981:  92%|█████████▏| 5973/6470 [1:08:38<05:37,  1.47it/s]
4|5974|Loss: 0.7875968217849731:  92%|█████████▏| 5973/6470 [1:08:39<05:37,  1.47it/s]
4|5974|Loss: 0.7875968217849731:  92%|█████████▏| 5974/6470 [1:08:39<05:41,  1.45it/s]
4|5975|Loss: 1.0151221752166748:  92%|█████████▏| 5974/6470 [1:08:39<05:41,  1.45it/s]
4|5975|Loss: 1.0151221752166748:  92%|█████████▏| 5975/6470 [1:08:40<05:27,  1.51it/s]
4|5976|Loss: 0.7593562602996826:  92%|█████████▏| 5975/6470 [1:08:40<05:27,  1.51it/s]
4|5976|Loss: 0.7593562602996826:  92%|█████████▏| 5976/6470 [1:08:40<05:32,  1.49it/s]
4|5977|Loss: 0.9552438855171204:  92%|█████████▏| 5976/6470 [1:08:41<05:32,  1.49it/s]
4|5977|Loss: 0.9552438855171204:  92%|█████████▏| 5977/6470 [1:08:41<05:56,  1.38it/s]
4|5978|Loss: 0.8726069927215576:  92%|█████████▏| 5977/6470 [1:08:42<05:56,  1.38it/s]
4|5978|Loss: 0.8726069927215576:  92%|█████████▏| 5978/6470 [1:08:42<05:33,  1.48it/s]
4|5979|Loss: 0.9118170738220215:  92%|█████████▏| 5978/6470 [1:08:42<05:33,  1.48it/s]
4|5979|Loss: 0.9118170738220215:  92%|█████████▏| 5979/6470 [1:08:42<05:19,  1.54it/s]
4|5980|Loss: 0.8732903003692627:  92%|█████████▏| 5979/6470 [1:08:43<05:19,  1.54it/s]
4|5980|Loss: 0.8732903003692627:  92%|█████████▏| 5980/6470 [1:08:43<05:44,  1.42it/s]
4|5981|Loss: 0.9018115401268005:  92%|█████████▏| 5980/6470 [1:08:44<05:44,  1.42it/s]
4|5981|Loss: 0.9018115401268005:  92%|█████████▏| 5981/6470 [1:08:44<05:28,  1.49it/s]
4|5982|Loss: 0.8089448809623718:  92%|█████████▏| 5981/6470 [1:08:44<05:28,  1.49it/s]
4|5982|Loss: 0.8089448809623718:  92%|█████████▏| 5982/6470 [1:08:44<05:17,  1.54it/s]
4|5983|Loss: 0.72676020860672:  92%|█████████▏| 5982/6470 [1:08:45<05:17,  1.54it/s]  
4|5983|Loss: 0.72676020860672:  92%|█████████▏| 5983/6470 [1:08:45<05:22,  1.51it/s]
4|5984|Loss: 0.7679568529129028:  92%|█████████▏| 5983/6470 [1:08:45<05:22,  1.51it/s]
4|5984|Loss: 0.7679568529129028:  92%|█████████▏| 5984/6470 [1:08:46<05:05,  1.59it/s]
4|5985|Loss: 0.7412164211273193:  92%|█████████▏| 5984/6470 [1:08:46<05:05,  1.59it/s]
4|5985|Loss: 0.7412164211273193:  93%|█████████▎| 5985/6470 [1:08:46<05:21,  1.51it/s]
4|5986|Loss: 0.9066667556762695:  93%|█████████▎| 5985/6470 [1:08:47<05:21,  1.51it/s]
4|5986|Loss: 0.9066667556762695:  93%|█████████▎| 5986/6470 [1:08:47<05:29,  1.47it/s]
4|5987|Loss: 0.7409600019454956:  93%|█████████▎| 5986/6470 [1:08:47<05:29,  1.47it/s]
4|5987|Loss: 0.7409600019454956:  93%|█████████▎| 5987/6470 [1:08:48<05:08,  1.56it/s]
4|5988|Loss: 0.8160114288330078:  93%|█████████▎| 5987/6470 [1:08:48<05:08,  1.56it/s]
4|5988|Loss: 0.8160114288330078:  93%|█████████▎| 5988/6470 [1:08:48<05:04,  1.58it/s]
4|5989|Loss: 0.9936967492103577:  93%|█████████▎| 5988/6470 [1:08:49<05:04,  1.58it/s]
4|5989|Loss: 0.9936967492103577:  93%|█████████▎| 5989/6470 [1:08:49<05:16,  1.52it/s]
4|5990|Loss: 0.8042305111885071:  93%|█████████▎| 5989/6470 [1:08:49<05:16,  1.52it/s]
4|5990|Loss: 0.8042305111885071:  93%|█████████▎| 5990/6470 [1:08:50<05:24,  1.48it/s]
4|5991|Loss: 0.7289588451385498:  93%|█████████▎| 5990/6470 [1:08:50<05:24,  1.48it/s]
4|5991|Loss: 0.7289588451385498:  93%|█████████▎| 5991/6470 [1:08:50<05:27,  1.46it/s]
4|5992|Loss: 0.6873399019241333:  93%|█████████▎| 5991/6470 [1:08:51<05:27,  1.46it/s]
4|5992|Loss: 0.6873399019241333:  93%|█████████▎| 5992/6470 [1:08:51<05:40,  1.40it/s]
4|5993|Loss: 0.9408512711524963:  93%|█████████▎| 5992/6470 [1:08:52<05:40,  1.40it/s]
4|5993|Loss: 0.9408512711524963:  93%|█████████▎| 5993/6470 [1:08:52<05:44,  1.38it/s]
4|5994|Loss: 0.9025408029556274:  93%|█████████▎| 5993/6470 [1:08:52<05:44,  1.38it/s]
4|5994|Loss: 0.9025408029556274:  93%|█████████▎| 5994/6470 [1:08:53<05:44,  1.38it/s]
4|5995|Loss: 0.8735524415969849:  93%|█████████▎| 5994/6470 [1:08:53<05:44,  1.38it/s]
4|5995|Loss: 0.8735524415969849:  93%|█████████▎| 5995/6470 [1:08:53<05:46,  1.37it/s]
4|5996|Loss: 0.8005646467208862:  93%|█████████▎| 5995/6470 [1:08:54<05:46,  1.37it/s]
4|5996|Loss: 0.8005646467208862:  93%|█████████▎| 5996/6470 [1:08:54<05:43,  1.38it/s]
4|5997|Loss: 0.9952308535575867:  93%|█████████▎| 5996/6470 [1:08:55<05:43,  1.38it/s]
4|5997|Loss: 0.9952308535575867:  93%|█████████▎| 5997/6470 [1:08:55<05:26,  1.45it/s]
4|5998|Loss: 0.987228274345398:  93%|█████████▎| 5997/6470 [1:08:55<05:26,  1.45it/s] 
4|5998|Loss: 0.987228274345398:  93%|█████████▎| 5998/6470 [1:08:55<05:17,  1.49it/s]
4|5999|Loss: 0.7688559293746948:  93%|█████████▎| 5998/6470 [1:08:56<05:17,  1.49it/s]
4|5999|Loss: 0.7688559293746948:  93%|█████████▎| 5999/6470 [1:08:56<05:13,  1.50it/s]
4|6000|Loss: 0.9036151766777039:  93%|█████████▎| 5999/6470 [1:08:56<05:13,  1.50it/s]
4|6000|Loss: 0.9036151766777039:  93%|█████████▎| 6000/6470 [1:08:57<05:27,  1.44it/s]
4|6001|Loss: 0.8017425537109375:  93%|█████████▎| 6000/6470 [1:08:57<05:27,  1.44it/s]
4|6001|Loss: 0.8017425537109375:  93%|█████████▎| 6001/6470 [1:08:57<05:08,  1.52it/s]
4|6002|Loss: 0.7250027656555176:  93%|█████████▎| 6001/6470 [1:08:58<05:08,  1.52it/s]
4|6002|Loss: 0.7250027656555176:  93%|█████████▎| 6002/6470 [1:08:58<05:22,  1.45it/s]
4|6003|Loss: 0.7084266543388367:  93%|█████████▎| 6002/6470 [1:08:59<05:22,  1.45it/s]
4|6003|Loss: 0.7084266543388367:  93%|█████████▎| 6003/6470 [1:08:59<05:31,  1.41it/s]
4|6004|Loss: 1.1429656744003296:  93%|█████████▎| 6003/6470 [1:08:59<05:31,  1.41it/s]
4|6004|Loss: 1.1429656744003296:  93%|█████████▎| 6004/6470 [1:09:00<05:17,  1.47it/s]
4|6005|Loss: 0.7248157262802124:  93%|█████████▎| 6004/6470 [1:09:00<05:17,  1.47it/s]
4|6005|Loss: 0.7248157262802124:  93%|█████████▎| 6005/6470 [1:09:00<05:21,  1.45it/s]
4|6006|Loss: 0.8426775932312012:  93%|█████████▎| 6005/6470 [1:09:01<05:21,  1.45it/s]
4|6006|Loss: 0.8426775932312012:  93%|█████████▎| 6006/6470 [1:09:01<05:10,  1.49it/s]
4|6007|Loss: 0.8348516821861267:  93%|█████████▎| 6006/6470 [1:09:01<05:10,  1.49it/s]
4|6007|Loss: 0.8348516821861267:  93%|█████████▎| 6007/6470 [1:09:02<05:15,  1.47it/s]
4|6008|Loss: 0.8025376796722412:  93%|█████████▎| 6007/6470 [1:09:02<05:15,  1.47it/s]
4|6008|Loss: 0.8025376796722412:  93%|█████████▎| 6008/6470 [1:09:02<05:06,  1.51it/s]
4|6009|Loss: 0.9006686806678772:  93%|█████████▎| 6008/6470 [1:09:03<05:06,  1.51it/s]
4|6009|Loss: 0.9006686806678772:  93%|█████████▎| 6009/6470 [1:09:03<05:12,  1.48it/s]
4|6010|Loss: 1.0462268590927124:  93%|█████████▎| 6009/6470 [1:09:03<05:12,  1.48it/s]
4|6010|Loss: 1.0462268590927124:  93%|█████████▎| 6010/6470 [1:09:04<05:09,  1.49it/s]
4|6011|Loss: 0.8171911835670471:  93%|█████████▎| 6010/6470 [1:09:04<05:09,  1.49it/s]
4|6011|Loss: 0.8171911835670471:  93%|█████████▎| 6011/6470 [1:09:04<05:14,  1.46it/s]
4|6012|Loss: 0.8695363998413086:  93%|█████████▎| 6011/6470 [1:09:05<05:14,  1.46it/s]
4|6012|Loss: 0.8695363998413086:  93%|█████████▎| 6012/6470 [1:09:05<05:34,  1.37it/s]
4|6013|Loss: 0.7692939639091492:  93%|█████████▎| 6012/6470 [1:09:06<05:34,  1.37it/s]
4|6013|Loss: 0.7692939639091492:  93%|█████████▎| 6013/6470 [1:09:06<05:35,  1.36it/s]
4|6014|Loss: 0.8802269697189331:  93%|█████████▎| 6013/6470 [1:09:06<05:35,  1.36it/s]
4|6014|Loss: 0.8802269697189331:  93%|█████████▎| 6014/6470 [1:09:06<05:11,  1.46it/s]
4|6015|Loss: 0.8728970289230347:  93%|█████████▎| 6014/6470 [1:09:07<05:11,  1.46it/s]
4|6015|Loss: 0.8728970289230347:  93%|█████████▎| 6015/6470 [1:09:07<04:58,  1.52it/s]
4|6016|Loss: 0.8435537219047546:  93%|█████████▎| 6015/6470 [1:09:07<04:58,  1.52it/s]
4|6016|Loss: 0.8435537219047546:  93%|█████████▎| 6016/6470 [1:09:08<04:53,  1.55it/s]
4|6017|Loss: 0.8492822051048279:  93%|█████████▎| 6016/6470 [1:09:08<04:53,  1.55it/s]
4|6017|Loss: 0.8492822051048279:  93%|█████████▎| 6017/6470 [1:09:08<04:30,  1.67it/s]
4|6018|Loss: 0.8465617299079895:  93%|█████████▎| 6017/6470 [1:09:08<04:30,  1.67it/s]
4|6018|Loss: 0.8465617299079895:  93%|█████████▎| 6018/6470 [1:09:09<04:16,  1.76it/s]
4|6019|Loss: 0.8544676899909973:  93%|█████████▎| 6018/6470 [1:09:09<04:16,  1.76it/s]
4|6019|Loss: 0.8544676899909973:  93%|█████████▎| 6019/6470 [1:09:09<04:47,  1.57it/s]
4|6020|Loss: 0.9207445383071899:  93%|█████████▎| 6019/6470 [1:09:10<04:47,  1.57it/s]
4|6020|Loss: 0.9207445383071899:  93%|█████████▎| 6020/6470 [1:09:10<05:01,  1.49it/s]
4|6021|Loss: 0.8035752177238464:  93%|█████████▎| 6020/6470 [1:09:11<05:01,  1.49it/s]
4|6021|Loss: 0.8035752177238464:  93%|█████████▎| 6021/6470 [1:09:11<04:55,  1.52it/s]
4|6022|Loss: 0.8560282588005066:  93%|█████████▎| 6021/6470 [1:09:11<04:55,  1.52it/s]
4|6022|Loss: 0.8560282588005066:  93%|█████████▎| 6022/6470 [1:09:12<05:02,  1.48it/s]
4|6023|Loss: 0.765717625617981:  93%|█████████▎| 6022/6470 [1:09:12<05:02,  1.48it/s] 
4|6023|Loss: 0.765717625617981:  93%|█████████▎| 6023/6470 [1:09:12<04:42,  1.58it/s]
4|6024|Loss: 0.8394030928611755:  93%|█████████▎| 6023/6470 [1:09:12<04:42,  1.58it/s]
4|6024|Loss: 0.8394030928611755:  93%|█████████▎| 6024/6470 [1:09:13<05:01,  1.48it/s]
4|6025|Loss: 0.8334452509880066:  93%|█████████▎| 6024/6470 [1:09:13<05:01,  1.48it/s]
4|6025|Loss: 0.8334452509880066:  93%|█████████▎| 6025/6470 [1:09:14<05:12,  1.42it/s]
4|6026|Loss: 0.9791674017906189:  93%|█████████▎| 6025/6470 [1:09:14<05:12,  1.42it/s]
4|6026|Loss: 0.9791674017906189:  93%|█████████▎| 6026/6470 [1:09:14<05:04,  1.46it/s]
4|6027|Loss: 0.8364368081092834:  93%|█████████▎| 6026/6470 [1:09:15<05:04,  1.46it/s]
4|6027|Loss: 0.8364368081092834:  93%|█████████▎| 6027/6470 [1:09:15<04:51,  1.52it/s]
4|6028|Loss: 0.7630462050437927:  93%|█████████▎| 6027/6470 [1:09:15<04:51,  1.52it/s]
4|6028|Loss: 0.7630462050437927:  93%|█████████▎| 6028/6470 [1:09:15<04:37,  1.59it/s]
4|6029|Loss: 0.7362497448921204:  93%|█████████▎| 6028/6470 [1:09:16<04:37,  1.59it/s]
4|6029|Loss: 0.7362497448921204:  93%|█████████▎| 6029/6470 [1:09:16<04:16,  1.72it/s]
4|6030|Loss: 0.8868088126182556:  93%|█████████▎| 6029/6470 [1:09:16<04:16,  1.72it/s]
4|6030|Loss: 0.8868088126182556:  93%|█████████▎| 6030/6470 [1:09:17<04:37,  1.58it/s]
4|6031|Loss: 0.9176155924797058:  93%|█████████▎| 6030/6470 [1:09:17<04:37,  1.58it/s]
4|6031|Loss: 0.9176155924797058:  93%|█████████▎| 6031/6470 [1:09:17<04:44,  1.54it/s]
4|6032|Loss: 1.0445939302444458:  93%|█████████▎| 6031/6470 [1:09:18<04:44,  1.54it/s]
4|6032|Loss: 1.0445939302444458:  93%|█████████▎| 6032/6470 [1:09:18<05:02,  1.45it/s]
4|6033|Loss: 0.7533928751945496:  93%|█████████▎| 6032/6470 [1:09:18<05:02,  1.45it/s]
4|6033|Loss: 0.7533928751945496:  93%|█████████▎| 6033/6470 [1:09:19<04:53,  1.49it/s]
4|6034|Loss: 0.9738783240318298:  93%|█████████▎| 6033/6470 [1:09:19<04:53,  1.49it/s]
4|6034|Loss: 0.9738783240318298:  93%|█████████▎| 6034/6470 [1:09:19<04:47,  1.52it/s]
4|6035|Loss: 0.8409231305122375:  93%|█████████▎| 6034/6470 [1:09:20<04:47,  1.52it/s]
4|6035|Loss: 0.8409231305122375:  93%|█████████▎| 6035/6470 [1:09:20<04:55,  1.47it/s]
4|6036|Loss: 0.6449398398399353:  93%|█████████▎| 6035/6470 [1:09:20<04:55,  1.47it/s]
4|6036|Loss: 0.6449398398399353:  93%|█████████▎| 6036/6470 [1:09:21<05:04,  1.43it/s]
4|6037|Loss: 0.8738656044006348:  93%|█████████▎| 6036/6470 [1:09:21<05:04,  1.43it/s]
4|6037|Loss: 0.8738656044006348:  93%|█████████▎| 6037/6470 [1:09:22<05:24,  1.33it/s]
4|6038|Loss: 0.7859193682670593:  93%|█████████▎| 6037/6470 [1:09:22<05:24,  1.33it/s]
4|6038|Loss: 0.7859193682670593:  93%|█████████▎| 6038/6470 [1:09:22<05:28,  1.32it/s]
4|6039|Loss: 0.9020253419876099:  93%|█████████▎| 6038/6470 [1:09:23<05:28,  1.32it/s]
4|6039|Loss: 0.9020253419876099:  93%|█████████▎| 6039/6470 [1:09:23<05:24,  1.33it/s]
4|6040|Loss: 0.8152529001235962:  93%|█████████▎| 6039/6470 [1:09:24<05:24,  1.33it/s]
4|6040|Loss: 0.8152529001235962:  93%|█████████▎| 6040/6470 [1:09:24<05:27,  1.31it/s]
4|6041|Loss: 0.8000958561897278:  93%|█████████▎| 6040/6470 [1:09:24<05:27,  1.31it/s]
4|6041|Loss: 0.8000958561897278:  93%|█████████▎| 6041/6470 [1:09:25<05:15,  1.36it/s]
4|6042|Loss: 0.8441416025161743:  93%|█████████▎| 6041/6470 [1:09:25<05:15,  1.36it/s]
4|6042|Loss: 0.8441416025161743:  93%|█████████▎| 6042/6470 [1:09:25<05:14,  1.36it/s]
4|6043|Loss: 0.9996549487113953:  93%|█████████▎| 6042/6470 [1:09:26<05:14,  1.36it/s]
4|6043|Loss: 0.9996549487113953:  93%|█████████▎| 6043/6470 [1:09:26<05:03,  1.41it/s]
4|6044|Loss: 0.8040387034416199:  93%|█████████▎| 6043/6470 [1:09:26<05:03,  1.41it/s]
4|6044|Loss: 0.8040387034416199:  93%|█████████▎| 6044/6470 [1:09:27<05:05,  1.39it/s]
4|6045|Loss: 0.9435188174247742:  93%|█████████▎| 6044/6470 [1:09:27<05:05,  1.39it/s]
4|6045|Loss: 0.9435188174247742:  93%|█████████▎| 6045/6470 [1:09:27<04:59,  1.42it/s]
4|6046|Loss: 0.8345128297805786:  93%|█████████▎| 6045/6470 [1:09:28<04:59,  1.42it/s]
4|6046|Loss: 0.8345128297805786:  93%|█████████▎| 6046/6470 [1:09:28<05:06,  1.38it/s]
4|6047|Loss: 0.8358153104782104:  93%|█████████▎| 6046/6470 [1:09:29<05:06,  1.38it/s]
4|6047|Loss: 0.8358153104782104:  93%|█████████▎| 6047/6470 [1:09:29<05:06,  1.38it/s]
4|6048|Loss: 0.9468458294868469:  93%|█████████▎| 6047/6470 [1:09:29<05:06,  1.38it/s]
4|6048|Loss: 0.9468458294868469:  93%|█████████▎| 6048/6470 [1:09:30<05:07,  1.37it/s]
4|6049|Loss: 0.913678765296936:  93%|█████████▎| 6048/6470 [1:09:30<05:07,  1.37it/s] 
4|6049|Loss: 0.913678765296936:  93%|█████████▎| 6049/6470 [1:09:30<04:58,  1.41it/s]
4|6050|Loss: 0.8988642692565918:  93%|█████████▎| 6049/6470 [1:09:31<04:58,  1.41it/s]
4|6050|Loss: 0.8988642692565918:  94%|█████████▎| 6050/6470 [1:09:31<05:04,  1.38it/s]
4|6051|Loss: 0.7630721926689148:  94%|█████████▎| 6050/6470 [1:09:32<05:04,  1.38it/s]
4|6051|Loss: 0.7630721926689148:  94%|█████████▎| 6051/6470 [1:09:32<05:21,  1.30it/s]
4|6052|Loss: 0.738398015499115:  94%|█████████▎| 6051/6470 [1:09:32<05:21,  1.30it/s] 
4|6052|Loss: 0.738398015499115:  94%|█████████▎| 6052/6470 [1:09:33<05:11,  1.34it/s]
4|6053|Loss: 0.8453391194343567:  94%|█████████▎| 6052/6470 [1:09:33<05:11,  1.34it/s]
4|6053|Loss: 0.8453391194343567:  94%|█████████▎| 6053/6470 [1:09:33<04:58,  1.40it/s]
4|6054|Loss: 0.757186233997345:  94%|█████████▎| 6053/6470 [1:09:34<04:58,  1.40it/s] 
4|6054|Loss: 0.757186233997345:  94%|█████████▎| 6054/6470 [1:09:34<05:24,  1.28it/s]
4|6055|Loss: 0.9834375977516174:  94%|█████████▎| 6054/6470 [1:09:35<05:24,  1.28it/s]
4|6055|Loss: 0.9834375977516174:  94%|█████████▎| 6055/6470 [1:09:35<05:15,  1.32it/s]
4|6056|Loss: 0.8022002577781677:  94%|█████████▎| 6055/6470 [1:09:35<05:15,  1.32it/s]
4|6056|Loss: 0.8022002577781677:  94%|█████████▎| 6056/6470 [1:09:36<05:09,  1.34it/s]
4|6057|Loss: 0.8311367034912109:  94%|█████████▎| 6056/6470 [1:09:36<05:09,  1.34it/s]
4|6057|Loss: 0.8311367034912109:  94%|█████████▎| 6057/6470 [1:09:36<05:08,  1.34it/s]
4|6058|Loss: 0.856209933757782:  94%|█████████▎| 6057/6470 [1:09:37<05:08,  1.34it/s] 
4|6058|Loss: 0.856209933757782:  94%|█████████▎| 6058/6470 [1:09:37<04:38,  1.48it/s]
4|6059|Loss: 0.9196791648864746:  94%|█████████▎| 6058/6470 [1:09:37<04:38,  1.48it/s]
4|6059|Loss: 0.9196791648864746:  94%|█████████▎| 6059/6470 [1:09:37<04:21,  1.57it/s]
4|6060|Loss: 0.8109744787216187:  94%|█████████▎| 6059/6470 [1:09:38<04:21,  1.57it/s]
4|6060|Loss: 0.8109744787216187:  94%|█████████▎| 6060/6470 [1:09:38<04:13,  1.62it/s]
4|6061|Loss: 0.7652552723884583:  94%|█████████▎| 6060/6470 [1:09:38<04:13,  1.62it/s]
4|6061|Loss: 0.7652552723884583:  94%|█████████▎| 6061/6470 [1:09:39<04:15,  1.60it/s]
4|6062|Loss: 0.7683399319648743:  94%|█████████▎| 6061/6470 [1:09:39<04:15,  1.60it/s]
4|6062|Loss: 0.7683399319648743:  94%|█████████▎| 6062/6470 [1:09:39<04:37,  1.47it/s]
4|6063|Loss: 0.961487889289856:  94%|█████████▎| 6062/6470 [1:09:40<04:37,  1.47it/s] 
4|6063|Loss: 0.961487889289856:  94%|█████████▎| 6063/6470 [1:09:40<04:46,  1.42it/s]
4|6064|Loss: 0.8390442132949829:  94%|█████████▎| 6063/6470 [1:09:41<04:46,  1.42it/s]
4|6064|Loss: 0.8390442132949829:  94%|█████████▎| 6064/6470 [1:09:41<04:45,  1.42it/s]
4|6065|Loss: 0.7574905157089233:  94%|█████████▎| 6064/6470 [1:09:41<04:45,  1.42it/s]
4|6065|Loss: 0.7574905157089233:  94%|█████████▎| 6065/6470 [1:09:42<04:56,  1.37it/s]
4|6066|Loss: 0.8789322376251221:  94%|█████████▎| 6065/6470 [1:09:42<04:56,  1.37it/s]
4|6066|Loss: 0.8789322376251221:  94%|█████████▍| 6066/6470 [1:09:43<04:59,  1.35it/s]
4|6067|Loss: 0.8383719325065613:  94%|█████████▍| 6066/6470 [1:09:43<04:59,  1.35it/s]
4|6067|Loss: 0.8383719325065613:  94%|█████████▍| 6067/6470 [1:09:43<04:55,  1.37it/s]
4|6068|Loss: 0.9086387753486633:  94%|█████████▍| 6067/6470 [1:09:44<04:55,  1.37it/s]
4|6068|Loss: 0.9086387753486633:  94%|█████████▍| 6068/6470 [1:09:44<04:59,  1.34it/s]
4|6069|Loss: 0.9532317519187927:  94%|█████████▍| 6068/6470 [1:09:44<04:59,  1.34it/s]
4|6069|Loss: 0.9532317519187927:  94%|█████████▍| 6069/6470 [1:09:45<04:37,  1.44it/s]
4|6070|Loss: 0.8911250233650208:  94%|█████████▍| 6069/6470 [1:09:45<04:37,  1.44it/s]
4|6070|Loss: 0.8911250233650208:  94%|█████████▍| 6070/6470 [1:09:45<04:26,  1.50it/s]
4|6071|Loss: 1.0914716720581055:  94%|█████████▍| 6070/6470 [1:09:46<04:26,  1.50it/s]
4|6071|Loss: 1.0914716720581055:  94%|█████████▍| 6071/6470 [1:09:46<04:22,  1.52it/s]
4|6072|Loss: 0.7171242237091064:  94%|█████████▍| 6071/6470 [1:09:46<04:22,  1.52it/s]
4|6072|Loss: 0.7171242237091064:  94%|█████████▍| 6072/6470 [1:09:47<04:30,  1.47it/s]
4|6073|Loss: 0.8634912967681885:  94%|█████████▍| 6072/6470 [1:09:47<04:30,  1.47it/s]
4|6073|Loss: 0.8634912967681885:  94%|█████████▍| 6073/6470 [1:09:47<04:26,  1.49it/s]
4|6074|Loss: 0.7337097525596619:  94%|█████████▍| 6073/6470 [1:09:48<04:26,  1.49it/s]
4|6074|Loss: 0.7337097525596619:  94%|█████████▍| 6074/6470 [1:09:48<04:32,  1.45it/s]
4|6075|Loss: 0.819387674331665:  94%|█████████▍| 6074/6470 [1:09:48<04:32,  1.45it/s] 
4|6075|Loss: 0.819387674331665:  94%|█████████▍| 6075/6470 [1:09:49<04:38,  1.42it/s]
4|6076|Loss: 0.8664456605911255:  94%|█████████▍| 6075/6470 [1:09:49<04:38,  1.42it/s]
4|6076|Loss: 0.8664456605911255:  94%|█████████▍| 6076/6470 [1:09:49<04:33,  1.44it/s]
4|6077|Loss: 0.7763469219207764:  94%|█████████▍| 6076/6470 [1:09:50<04:33,  1.44it/s]
4|6077|Loss: 0.7763469219207764:  94%|█████████▍| 6077/6470 [1:09:50<04:31,  1.45it/s]
4|6078|Loss: 0.7861276865005493:  94%|█████████▍| 6077/6470 [1:09:50<04:31,  1.45it/s]
4|6078|Loss: 0.7861276865005493:  94%|█████████▍| 6078/6470 [1:09:51<04:49,  1.35it/s]
4|6079|Loss: 0.689812958240509:  94%|█████████▍| 6078/6470 [1:09:51<04:49,  1.35it/s] 
4|6079|Loss: 0.689812958240509:  94%|█████████▍| 6079/6470 [1:09:52<05:06,  1.28it/s]
4|6080|Loss: 0.7665607929229736:  94%|█████████▍| 6079/6470 [1:09:52<05:06,  1.28it/s]
4|6080|Loss: 0.7665607929229736:  94%|█████████▍| 6080/6470 [1:09:52<04:45,  1.37it/s]
4|6081|Loss: 0.9398572444915771:  94%|█████████▍| 6080/6470 [1:09:53<04:45,  1.37it/s]
4|6081|Loss: 0.9398572444915771:  94%|█████████▍| 6081/6470 [1:09:53<04:28,  1.45it/s]
4|6082|Loss: 0.8038252592086792:  94%|█████████▍| 6081/6470 [1:09:53<04:28,  1.45it/s]
4|6082|Loss: 0.8038252592086792:  94%|█████████▍| 6082/6470 [1:09:54<04:44,  1.37it/s]
4|6083|Loss: 0.8628314733505249:  94%|█████████▍| 6082/6470 [1:09:54<04:44,  1.37it/s]
4|6083|Loss: 0.8628314733505249:  94%|█████████▍| 6083/6470 [1:09:55<04:48,  1.34it/s]
4|6084|Loss: 0.7888734340667725:  94%|█████████▍| 6083/6470 [1:09:55<04:48,  1.34it/s]
4|6084|Loss: 0.7888734340667725:  94%|█████████▍| 6084/6470 [1:09:55<04:58,  1.29it/s]
4|6085|Loss: 0.9264776706695557:  94%|█████████▍| 6084/6470 [1:09:56<04:58,  1.29it/s]
4|6085|Loss: 0.9264776706695557:  94%|█████████▍| 6085/6470 [1:09:56<04:35,  1.40it/s]
4|6086|Loss: 0.8906835317611694:  94%|█████████▍| 6085/6470 [1:09:56<04:35,  1.40it/s]
4|6086|Loss: 0.8906835317611694:  94%|█████████▍| 6086/6470 [1:09:57<04:57,  1.29it/s]
4|6087|Loss: 0.756490170955658:  94%|█████████▍| 6086/6470 [1:09:57<04:57,  1.29it/s] 
4|6087|Loss: 0.756490170955658:  94%|█████████▍| 6087/6470 [1:09:58<04:59,  1.28it/s]
4|6088|Loss: 0.7970340251922607:  94%|█████████▍| 6087/6470 [1:09:58<04:59,  1.28it/s]
4|6088|Loss: 0.7970340251922607:  94%|█████████▍| 6088/6470 [1:09:58<04:55,  1.29it/s]
4|6089|Loss: 0.7707123160362244:  94%|█████████▍| 6088/6470 [1:09:59<04:55,  1.29it/s]
4|6089|Loss: 0.7707123160362244:  94%|█████████▍| 6089/6470 [1:09:59<04:50,  1.31it/s]
4|6090|Loss: 0.8946478366851807:  94%|█████████▍| 6089/6470 [1:10:00<04:50,  1.31it/s]
4|6090|Loss: 0.8946478366851807:  94%|█████████▍| 6090/6470 [1:10:00<04:42,  1.34it/s]
4|6091|Loss: 0.7326043844223022:  94%|█████████▍| 6090/6470 [1:10:00<04:42,  1.34it/s]
4|6091|Loss: 0.7326043844223022:  94%|█████████▍| 6091/6470 [1:10:00<04:14,  1.49it/s]
4|6092|Loss: 0.6683090329170227:  94%|█████████▍| 6091/6470 [1:10:01<04:14,  1.49it/s]
4|6092|Loss: 0.6683090329170227:  94%|█████████▍| 6092/6470 [1:10:01<04:10,  1.51it/s]
4|6093|Loss: 0.7749210596084595:  94%|█████████▍| 6092/6470 [1:10:01<04:10,  1.51it/s]
4|6093|Loss: 0.7749210596084595:  94%|█████████▍| 6093/6470 [1:10:02<04:09,  1.51it/s]
4|6094|Loss: 0.7310872673988342:  94%|█████████▍| 6093/6470 [1:10:02<04:09,  1.51it/s]
4|6094|Loss: 0.7310872673988342:  94%|█████████▍| 6094/6470 [1:10:02<04:06,  1.52it/s]
4|6095|Loss: 0.9905081391334534:  94%|█████████▍| 6094/6470 [1:10:03<04:06,  1.52it/s]
4|6095|Loss: 0.9905081391334534:  94%|█████████▍| 6095/6470 [1:10:03<04:23,  1.42it/s]
4|6096|Loss: 0.7741861939430237:  94%|█████████▍| 6095/6470 [1:10:04<04:23,  1.42it/s]
4|6096|Loss: 0.7741861939430237:  94%|█████████▍| 6096/6470 [1:10:04<04:20,  1.44it/s]
4|6097|Loss: 0.7888373732566833:  94%|█████████▍| 6096/6470 [1:10:04<04:20,  1.44it/s]
4|6097|Loss: 0.7888373732566833:  94%|█████████▍| 6097/6470 [1:10:05<04:21,  1.43it/s]
4|6098|Loss: 0.8949198126792908:  94%|█████████▍| 6097/6470 [1:10:05<04:21,  1.43it/s]
4|6098|Loss: 0.8949198126792908:  94%|█████████▍| 6098/6470 [1:10:05<04:16,  1.45it/s]
4|6099|Loss: 0.8462947607040405:  94%|█████████▍| 6098/6470 [1:10:06<04:16,  1.45it/s]
4|6099|Loss: 0.8462947607040405:  94%|█████████▍| 6099/6470 [1:10:06<04:04,  1.52it/s]
4|6100|Loss: 0.8964821100234985:  94%|█████████▍| 6099/6470 [1:10:06<04:04,  1.52it/s]
4|6100|Loss: 0.8964821100234985:  94%|█████████▍| 6100/6470 [1:10:06<03:54,  1.57it/s]
4|6101|Loss: 0.8983182311058044:  94%|█████████▍| 6100/6470 [1:10:07<03:54,  1.57it/s]
4|6101|Loss: 0.8983182311058044:  94%|█████████▍| 6101/6470 [1:10:07<04:02,  1.52it/s]
4|6102|Loss: 0.9000130891799927:  94%|█████████▍| 6101/6470 [1:10:08<04:02,  1.52it/s]
4|6102|Loss: 0.9000130891799927:  94%|█████████▍| 6102/6470 [1:10:08<04:19,  1.42it/s]
4|6103|Loss: 0.8544273972511292:  94%|█████████▍| 6102/6470 [1:10:08<04:19,  1.42it/s]
4|6103|Loss: 0.8544273972511292:  94%|█████████▍| 6103/6470 [1:10:09<04:15,  1.44it/s]
4|6104|Loss: 0.7527489066123962:  94%|█████████▍| 6103/6470 [1:10:09<04:15,  1.44it/s]
4|6104|Loss: 0.7527489066123962:  94%|█████████▍| 6104/6470 [1:10:09<04:03,  1.50it/s]
4|6105|Loss: 0.7932942509651184:  94%|█████████▍| 6104/6470 [1:10:10<04:03,  1.50it/s]
4|6105|Loss: 0.7932942509651184:  94%|█████████▍| 6105/6470 [1:10:10<03:59,  1.52it/s]
4|6106|Loss: 1.0159356594085693:  94%|█████████▍| 6105/6470 [1:10:10<03:59,  1.52it/s]
4|6106|Loss: 1.0159356594085693:  94%|█████████▍| 6106/6470 [1:10:10<03:51,  1.57it/s]
4|6107|Loss: 0.8924363255500793:  94%|█████████▍| 6106/6470 [1:10:11<03:51,  1.57it/s]
4|6107|Loss: 0.8924363255500793:  94%|█████████▍| 6107/6470 [1:10:11<04:00,  1.51it/s]
4|6108|Loss: 0.9283016920089722:  94%|█████████▍| 6107/6470 [1:10:11<04:00,  1.51it/s]
4|6108|Loss: 0.9283016920089722:  94%|█████████▍| 6108/6470 [1:10:12<03:52,  1.56it/s]
4|6109|Loss: 0.8128795027732849:  94%|█████████▍| 6108/6470 [1:10:12<03:52,  1.56it/s]
4|6109|Loss: 0.8128795027732849:  94%|█████████▍| 6109/6470 [1:10:12<03:55,  1.53it/s]
4|6110|Loss: 0.7840362191200256:  94%|█████████▍| 6109/6470 [1:10:13<03:55,  1.53it/s]
4|6110|Loss: 0.7840362191200256:  94%|█████████▍| 6110/6470 [1:10:13<03:49,  1.57it/s]
4|6111|Loss: 0.8804923295974731:  94%|█████████▍| 6110/6470 [1:10:13<03:49,  1.57it/s]
4|6111|Loss: 0.8804923295974731:  94%|█████████▍| 6111/6470 [1:10:14<03:48,  1.57it/s]
4|6112|Loss: 1.013104796409607:  94%|█████████▍| 6111/6470 [1:10:14<03:48,  1.57it/s] 
4|6112|Loss: 1.013104796409607:  94%|█████████▍| 6112/6470 [1:10:14<03:49,  1.56it/s]
4|6113|Loss: 0.7876835465431213:  94%|█████████▍| 6112/6470 [1:10:15<03:49,  1.56it/s]
4|6113|Loss: 0.7876835465431213:  94%|█████████▍| 6113/6470 [1:10:15<03:58,  1.50it/s]
4|6114|Loss: 0.9602184891700745:  94%|█████████▍| 6113/6470 [1:10:15<03:58,  1.50it/s]
4|6114|Loss: 0.9602184891700745:  94%|█████████▍| 6114/6470 [1:10:16<03:49,  1.55it/s]
4|6115|Loss: 0.7322949767112732:  94%|█████████▍| 6114/6470 [1:10:16<03:49,  1.55it/s]
4|6115|Loss: 0.7322949767112732:  95%|█████████▍| 6115/6470 [1:10:16<04:00,  1.47it/s]
4|6116|Loss: 0.75902259349823:  95%|█████████▍| 6115/6470 [1:10:17<04:00,  1.47it/s]  
4|6116|Loss: 0.75902259349823:  95%|█████████▍| 6116/6470 [1:10:17<03:55,  1.50it/s]
4|6117|Loss: 0.7145909070968628:  95%|█████████▍| 6116/6470 [1:10:17<03:55,  1.50it/s]
4|6117|Loss: 0.7145909070968628:  95%|█████████▍| 6117/6470 [1:10:18<03:58,  1.48it/s]
4|6118|Loss: 0.8099766373634338:  95%|█████████▍| 6117/6470 [1:10:18<03:58,  1.48it/s]
4|6118|Loss: 0.8099766373634338:  95%|█████████▍| 6118/6470 [1:10:18<03:55,  1.49it/s]
4|6119|Loss: 0.9362022280693054:  95%|█████████▍| 6118/6470 [1:10:19<03:55,  1.49it/s]
4|6119|Loss: 0.9362022280693054:  95%|█████████▍| 6119/6470 [1:10:19<04:00,  1.46it/s]
4|6120|Loss: 0.8402705788612366:  95%|█████████▍| 6119/6470 [1:10:19<04:00,  1.46it/s]
4|6120|Loss: 0.8402705788612366:  95%|█████████▍| 6120/6470 [1:10:20<04:03,  1.44it/s]
4|6121|Loss: 0.8772727847099304:  95%|█████████▍| 6120/6470 [1:10:20<04:03,  1.44it/s]
4|6121|Loss: 0.8772727847099304:  95%|█████████▍| 6121/6470 [1:10:20<03:53,  1.49it/s]
4|6122|Loss: 0.7663717865943909:  95%|█████████▍| 6121/6470 [1:10:21<03:53,  1.49it/s]
4|6122|Loss: 0.7663717865943909:  95%|█████████▍| 6122/6470 [1:10:21<03:56,  1.47it/s]
4|6123|Loss: 0.8438661694526672:  95%|█████████▍| 6122/6470 [1:10:21<03:56,  1.47it/s]
4|6123|Loss: 0.8438661694526672:  95%|█████████▍| 6123/6470 [1:10:22<03:52,  1.49it/s]
4|6124|Loss: 0.8240165114402771:  95%|█████████▍| 6123/6470 [1:10:22<03:52,  1.49it/s]
4|6124|Loss: 0.8240165114402771:  95%|█████████▍| 6124/6470 [1:10:22<03:37,  1.59it/s]
4|6125|Loss: 0.8008472323417664:  95%|█████████▍| 6124/6470 [1:10:23<03:37,  1.59it/s]
4|6125|Loss: 0.8008472323417664:  95%|█████████▍| 6125/6470 [1:10:23<03:52,  1.48it/s]
4|6126|Loss: 0.8154539465904236:  95%|█████████▍| 6125/6470 [1:10:23<03:52,  1.48it/s]
4|6126|Loss: 0.8154539465904236:  95%|█████████▍| 6126/6470 [1:10:24<04:02,  1.42it/s]
4|6127|Loss: 0.841961145401001:  95%|█████████▍| 6126/6470 [1:10:24<04:02,  1.42it/s] 
4|6127|Loss: 0.841961145401001:  95%|█████████▍| 6127/6470 [1:10:25<04:05,  1.40it/s]
4|6128|Loss: 0.8376932144165039:  95%|█████████▍| 6127/6470 [1:10:25<04:05,  1.40it/s]
4|6128|Loss: 0.8376932144165039:  95%|█████████▍| 6128/6470 [1:10:25<04:21,  1.31it/s]
4|6129|Loss: 0.8084527254104614:  95%|█████████▍| 6128/6470 [1:10:26<04:21,  1.31it/s]
4|6129|Loss: 0.8084527254104614:  95%|█████████▍| 6129/6470 [1:10:26<04:09,  1.37it/s]
4|6130|Loss: 0.875861406326294:  95%|█████████▍| 6129/6470 [1:10:27<04:09,  1.37it/s] 
4|6130|Loss: 0.875861406326294:  95%|█████████▍| 6130/6470 [1:10:27<04:15,  1.33it/s]
4|6131|Loss: 0.647371232509613:  95%|█████████▍| 6130/6470 [1:10:27<04:15,  1.33it/s]
4|6131|Loss: 0.647371232509613:  95%|█████████▍| 6131/6470 [1:10:28<04:20,  1.30it/s]
4|6132|Loss: 0.6263558864593506:  95%|█████████▍| 6131/6470 [1:10:28<04:20,  1.30it/s]
4|6132|Loss: 0.6263558864593506:  95%|█████████▍| 6132/6470 [1:10:28<03:48,  1.48it/s]
4|6133|Loss: 0.7926957011222839:  95%|█████████▍| 6132/6470 [1:10:28<03:48,  1.48it/s]
4|6133|Loss: 0.7926957011222839:  95%|█████████▍| 6133/6470 [1:10:29<03:36,  1.56it/s]
4|6134|Loss: 0.7378073334693909:  95%|█████████▍| 6133/6470 [1:10:29<03:36,  1.56it/s]
4|6134|Loss: 0.7378073334693909:  95%|█████████▍| 6134/6470 [1:10:29<03:25,  1.63it/s]
4|6135|Loss: 0.7905066609382629:  95%|█████████▍| 6134/6470 [1:10:30<03:25,  1.63it/s]
4|6135|Loss: 0.7905066609382629:  95%|█████████▍| 6135/6470 [1:10:30<03:25,  1.63it/s]
4|6136|Loss: 0.7796239256858826:  95%|█████████▍| 6135/6470 [1:10:30<03:25,  1.63it/s]
4|6136|Loss: 0.7796239256858826:  95%|█████████▍| 6136/6470 [1:10:31<03:26,  1.62it/s]
4|6137|Loss: 0.8278911113739014:  95%|█████████▍| 6136/6470 [1:10:31<03:26,  1.62it/s]
4|6137|Loss: 0.8278911113739014:  95%|█████████▍| 6137/6470 [1:10:31<03:38,  1.52it/s]
4|6138|Loss: 0.802751898765564:  95%|█████████▍| 6137/6470 [1:10:32<03:38,  1.52it/s] 
4|6138|Loss: 0.802751898765564:  95%|█████████▍| 6138/6470 [1:10:32<03:40,  1.51it/s]
4|6139|Loss: 0.8638434410095215:  95%|█████████▍| 6138/6470 [1:10:32<03:40,  1.51it/s]
4|6139|Loss: 0.8638434410095215:  95%|█████████▍| 6139/6470 [1:10:33<03:35,  1.53it/s]
4|6140|Loss: 0.7366065979003906:  95%|█████████▍| 6139/6470 [1:10:33<03:35,  1.53it/s]
4|6140|Loss: 0.7366065979003906:  95%|█████████▍| 6140/6470 [1:10:33<03:33,  1.55it/s]
4|6141|Loss: 0.7961093783378601:  95%|█████████▍| 6140/6470 [1:10:34<03:33,  1.55it/s]
4|6141|Loss: 0.7961093783378601:  95%|█████████▍| 6141/6470 [1:10:34<03:37,  1.52it/s]
4|6142|Loss: 0.9076427221298218:  95%|█████████▍| 6141/6470 [1:10:34<03:37,  1.52it/s]
4|6142|Loss: 0.9076427221298218:  95%|█████████▍| 6142/6470 [1:10:35<03:30,  1.55it/s]
4|6143|Loss: 0.7873204946517944:  95%|█████████▍| 6142/6470 [1:10:35<03:30,  1.55it/s]
4|6143|Loss: 0.7873204946517944:  95%|█████████▍| 6143/6470 [1:10:35<03:45,  1.45it/s]
4|6144|Loss: 0.8387572169303894:  95%|█████████▍| 6143/6470 [1:10:36<03:45,  1.45it/s]
4|6144|Loss: 0.8387572169303894:  95%|█████████▍| 6144/6470 [1:10:36<03:56,  1.38it/s]
4|6145|Loss: 0.8940992951393127:  95%|█████████▍| 6144/6470 [1:10:37<03:56,  1.38it/s]
4|6145|Loss: 0.8940992951393127:  95%|█████████▍| 6145/6470 [1:10:37<03:58,  1.36it/s]
4|6146|Loss: 0.7512078881263733:  95%|█████████▍| 6145/6470 [1:10:37<03:58,  1.36it/s]
4|6146|Loss: 0.7512078881263733:  95%|█████████▍| 6146/6470 [1:10:38<03:57,  1.36it/s]
4|6147|Loss: 0.8535864353179932:  95%|█████████▍| 6146/6470 [1:10:38<03:57,  1.36it/s]
4|6147|Loss: 0.8535864353179932:  95%|█████████▌| 6147/6470 [1:10:38<03:58,  1.35it/s]
4|6148|Loss: 0.8848670721054077:  95%|█████████▌| 6147/6470 [1:10:39<03:58,  1.35it/s]
4|6148|Loss: 0.8848670721054077:  95%|█████████▌| 6148/6470 [1:10:39<04:11,  1.28it/s]
4|6149|Loss: 0.8322415351867676:  95%|█████████▌| 6148/6470 [1:10:40<04:11,  1.28it/s]
4|6149|Loss: 0.8322415351867676:  95%|█████████▌| 6149/6470 [1:10:40<04:01,  1.33it/s]
4|6150|Loss: 0.8234928846359253:  95%|█████████▌| 6149/6470 [1:10:40<04:01,  1.33it/s]
4|6150|Loss: 0.8234928846359253:  95%|█████████▌| 6150/6470 [1:10:41<03:47,  1.41it/s]
4|6151|Loss: 0.9698575735092163:  95%|█████████▌| 6150/6470 [1:10:41<03:47,  1.41it/s]
4|6151|Loss: 0.9698575735092163:  95%|█████████▌| 6151/6470 [1:10:41<03:46,  1.41it/s]
4|6152|Loss: 0.9286788105964661:  95%|█████████▌| 6151/6470 [1:10:42<03:46,  1.41it/s]
4|6152|Loss: 0.9286788105964661:  95%|█████████▌| 6152/6470 [1:10:42<03:50,  1.38it/s]
4|6153|Loss: 0.8651163578033447:  95%|█████████▌| 6152/6470 [1:10:42<03:50,  1.38it/s]
4|6153|Loss: 0.8651163578033447:  95%|█████████▌| 6153/6470 [1:10:43<04:05,  1.29it/s]
4|6154|Loss: 0.8559015393257141:  95%|█████████▌| 6153/6470 [1:10:43<04:05,  1.29it/s]
4|6154|Loss: 0.8559015393257141:  95%|█████████▌| 6154/6470 [1:10:44<04:04,  1.29it/s]
4|6155|Loss: 0.8539849519729614:  95%|█████████▌| 6154/6470 [1:10:44<04:04,  1.29it/s]
4|6155|Loss: 0.8539849519729614:  95%|█████████▌| 6155/6470 [1:10:44<04:04,  1.29it/s]
4|6156|Loss: 0.7690209746360779:  95%|█████████▌| 6155/6470 [1:10:45<04:04,  1.29it/s]
4|6156|Loss: 0.7690209746360779:  95%|█████████▌| 6156/6470 [1:10:45<03:59,  1.31it/s]
4|6157|Loss: 0.9351364970207214:  95%|█████████▌| 6156/6470 [1:10:46<03:59,  1.31it/s]
4|6157|Loss: 0.9351364970207214:  95%|█████████▌| 6157/6470 [1:10:46<03:39,  1.43it/s]
4|6158|Loss: 0.8383760452270508:  95%|█████████▌| 6157/6470 [1:10:46<03:39,  1.43it/s]
4|6158|Loss: 0.8383760452270508:  95%|█████████▌| 6158/6470 [1:10:46<03:37,  1.43it/s]
4|6159|Loss: 0.9749800562858582:  95%|█████████▌| 6158/6470 [1:10:47<03:37,  1.43it/s]
4|6159|Loss: 0.9749800562858582:  95%|█████████▌| 6159/6470 [1:10:47<03:27,  1.50it/s]
4|6160|Loss: 0.8297544717788696:  95%|█████████▌| 6159/6470 [1:10:47<03:27,  1.50it/s]
4|6160|Loss: 0.8297544717788696:  95%|█████████▌| 6160/6470 [1:10:48<03:29,  1.48it/s]
4|6161|Loss: 0.7600976824760437:  95%|█████████▌| 6160/6470 [1:10:48<03:29,  1.48it/s]
4|6161|Loss: 0.7600976824760437:  95%|█████████▌| 6161/6470 [1:10:48<03:34,  1.44it/s]
4|6162|Loss: 0.7810320258140564:  95%|█████████▌| 6161/6470 [1:10:49<03:34,  1.44it/s]
4|6162|Loss: 0.7810320258140564:  95%|█████████▌| 6162/6470 [1:10:49<03:43,  1.38it/s]
4|6163|Loss: 0.8650866150856018:  95%|█████████▌| 6162/6470 [1:10:50<03:43,  1.38it/s]
4|6163|Loss: 0.8650866150856018:  95%|█████████▌| 6163/6470 [1:10:50<03:53,  1.32it/s]
4|6164|Loss: 0.8231378197669983:  95%|█████████▌| 6163/6470 [1:10:50<03:53,  1.32it/s]
4|6164|Loss: 0.8231378197669983:  95%|█████████▌| 6164/6470 [1:10:51<03:43,  1.37it/s]
4|6165|Loss: 0.876702606678009:  95%|█████████▌| 6164/6470 [1:10:51<03:43,  1.37it/s] 
4|6165|Loss: 0.876702606678009:  95%|█████████▌| 6165/6470 [1:10:52<03:47,  1.34it/s]
4|6166|Loss: 0.8053126931190491:  95%|█████████▌| 6165/6470 [1:10:52<03:47,  1.34it/s]
4|6166|Loss: 0.8053126931190491:  95%|█████████▌| 6166/6470 [1:10:52<03:36,  1.41it/s]
4|6167|Loss: 0.9465623497962952:  95%|█████████▌| 6166/6470 [1:10:52<03:36,  1.41it/s]
4|6167|Loss: 0.9465623497962952:  95%|█████████▌| 6167/6470 [1:10:53<03:22,  1.50it/s]
4|6168|Loss: 0.8806082606315613:  95%|█████████▌| 6167/6470 [1:10:53<03:22,  1.50it/s]
4|6168|Loss: 0.8806082606315613:  95%|█████████▌| 6168/6470 [1:10:54<03:31,  1.43it/s]
4|6169|Loss: 0.8210669755935669:  95%|█████████▌| 6168/6470 [1:10:54<03:31,  1.43it/s]
4|6169|Loss: 0.8210669755935669:  95%|█████████▌| 6169/6470 [1:10:54<03:37,  1.39it/s]
4|6170|Loss: 0.7905459403991699:  95%|█████████▌| 6169/6470 [1:10:55<03:37,  1.39it/s]
4|6170|Loss: 0.7905459403991699:  95%|█████████▌| 6170/6470 [1:10:55<03:27,  1.44it/s]
4|6171|Loss: 0.8427238464355469:  95%|█████████▌| 6170/6470 [1:10:55<03:27,  1.44it/s]
4|6171|Loss: 0.8427238464355469:  95%|█████████▌| 6171/6470 [1:10:56<03:19,  1.50it/s]
4|6172|Loss: 0.8659218549728394:  95%|█████████▌| 6171/6470 [1:10:56<03:19,  1.50it/s]
4|6172|Loss: 0.8659218549728394:  95%|█████████▌| 6172/6470 [1:10:56<03:16,  1.52it/s]
4|6173|Loss: 0.7746943235397339:  95%|█████████▌| 6172/6470 [1:10:57<03:16,  1.52it/s]
4|6173|Loss: 0.7746943235397339:  95%|█████████▌| 6173/6470 [1:10:57<03:18,  1.50it/s]
4|6174|Loss: 0.8279815912246704:  95%|█████████▌| 6173/6470 [1:10:57<03:18,  1.50it/s]
4|6174|Loss: 0.8279815912246704:  95%|█████████▌| 6174/6470 [1:10:57<03:06,  1.59it/s]
4|6175|Loss: 0.9108752608299255:  95%|█████████▌| 6174/6470 [1:10:58<03:06,  1.59it/s]
4|6175|Loss: 0.9108752608299255:  95%|█████████▌| 6175/6470 [1:10:58<03:04,  1.60it/s]
4|6176|Loss: 0.8795948624610901:  95%|█████████▌| 6175/6470 [1:10:58<03:04,  1.60it/s]
4|6176|Loss: 0.8795948624610901:  95%|█████████▌| 6176/6470 [1:10:59<03:00,  1.63it/s]
4|6177|Loss: 0.944110095500946:  95%|█████████▌| 6176/6470 [1:10:59<03:00,  1.63it/s] 
4|6177|Loss: 0.944110095500946:  95%|█████████▌| 6177/6470 [1:10:59<03:00,  1.63it/s]
4|6178|Loss: 0.5944720506668091:  95%|█████████▌| 6177/6470 [1:11:00<03:00,  1.63it/s]
4|6178|Loss: 0.5944720506668091:  95%|█████████▌| 6178/6470 [1:11:00<03:05,  1.58it/s]
4|6179|Loss: 0.9479370713233948:  95%|█████████▌| 6178/6470 [1:11:00<03:05,  1.58it/s]
4|6179|Loss: 0.9479370713233948:  96%|█████████▌| 6179/6470 [1:11:01<03:10,  1.53it/s]
4|6180|Loss: 0.8099252581596375:  96%|█████████▌| 6179/6470 [1:11:01<03:10,  1.53it/s]
4|6180|Loss: 0.8099252581596375:  96%|█████████▌| 6180/6470 [1:11:01<03:09,  1.53it/s]
4|6181|Loss: 0.9028608202934265:  96%|█████████▌| 6180/6470 [1:11:02<03:09,  1.53it/s]
4|6181|Loss: 0.9028608202934265:  96%|█████████▌| 6181/6470 [1:11:02<03:12,  1.50it/s]
4|6182|Loss: 0.9116464853286743:  96%|█████████▌| 6181/6470 [1:11:02<03:12,  1.50it/s]
4|6182|Loss: 0.9116464853286743:  96%|█████████▌| 6182/6470 [1:11:03<03:22,  1.42it/s]
4|6183|Loss: 0.8392344117164612:  96%|█████████▌| 6182/6470 [1:11:03<03:22,  1.42it/s]
4|6183|Loss: 0.8392344117164612:  96%|█████████▌| 6183/6470 [1:11:03<03:25,  1.40it/s]
4|6184|Loss: 0.9361543655395508:  96%|█████████▌| 6183/6470 [1:11:04<03:25,  1.40it/s]
4|6184|Loss: 0.9361543655395508:  96%|█████████▌| 6184/6470 [1:11:04<03:34,  1.33it/s]
4|6185|Loss: 0.9148417115211487:  96%|█████████▌| 6184/6470 [1:11:05<03:34,  1.33it/s]
4|6185|Loss: 0.9148417115211487:  96%|█████████▌| 6185/6470 [1:11:05<03:32,  1.34it/s]
4|6186|Loss: 0.7887914776802063:  96%|█████████▌| 6185/6470 [1:11:05<03:32,  1.34it/s]
4|6186|Loss: 0.7887914776802063:  96%|█████████▌| 6186/6470 [1:11:06<03:27,  1.37it/s]
4|6187|Loss: 0.929862916469574:  96%|█████████▌| 6186/6470 [1:11:06<03:27,  1.37it/s] 
4|6187|Loss: 0.929862916469574:  96%|█████████▌| 6187/6470 [1:11:06<03:26,  1.37it/s]
4|6188|Loss: 0.7546600699424744:  96%|█████████▌| 6187/6470 [1:11:07<03:26,  1.37it/s]
4|6188|Loss: 0.7546600699424744:  96%|█████████▌| 6188/6470 [1:11:07<03:26,  1.37it/s]
4|6189|Loss: 0.7041491866111755:  96%|█████████▌| 6188/6470 [1:11:08<03:26,  1.37it/s]
4|6189|Loss: 0.7041491866111755:  96%|█████████▌| 6189/6470 [1:11:08<03:30,  1.34it/s]
4|6190|Loss: 0.8244168162345886:  96%|█████████▌| 6189/6470 [1:11:08<03:30,  1.34it/s]
4|6190|Loss: 0.8244168162345886:  96%|█████████▌| 6190/6470 [1:11:09<03:23,  1.37it/s]
4|6191|Loss: 0.747187614440918:  96%|█████████▌| 6190/6470 [1:11:09<03:23,  1.37it/s] 
4|6191|Loss: 0.747187614440918:  96%|█████████▌| 6191/6470 [1:11:09<03:19,  1.40it/s]
4|6192|Loss: 0.850456714630127:  96%|█████████▌| 6191/6470 [1:11:10<03:19,  1.40it/s]
4|6192|Loss: 0.850456714630127:  96%|█████████▌| 6192/6470 [1:11:10<03:18,  1.40it/s]
4|6193|Loss: 0.9531475305557251:  96%|█████████▌| 6192/6470 [1:11:10<03:18,  1.40it/s]
4|6193|Loss: 0.9531475305557251:  96%|█████████▌| 6193/6470 [1:11:11<03:27,  1.33it/s]
4|6194|Loss: 0.6945731043815613:  96%|█████████▌| 6193/6470 [1:11:11<03:27,  1.33it/s]
4|6194|Loss: 0.6945731043815613:  96%|█████████▌| 6194/6470 [1:11:12<03:26,  1.34it/s]
4|6195|Loss: 0.9164279699325562:  96%|█████████▌| 6194/6470 [1:11:12<03:26,  1.34it/s]
4|6195|Loss: 0.9164279699325562:  96%|█████████▌| 6195/6470 [1:11:12<03:21,  1.36it/s]
4|6196|Loss: 0.784969687461853:  96%|█████████▌| 6195/6470 [1:11:13<03:21,  1.36it/s] 
4|6196|Loss: 0.784969687461853:  96%|█████████▌| 6196/6470 [1:11:13<03:14,  1.41it/s]
4|6197|Loss: 0.8576146960258484:  96%|█████████▌| 6196/6470 [1:11:13<03:14,  1.41it/s]
4|6197|Loss: 0.8576146960258484:  96%|█████████▌| 6197/6470 [1:11:14<03:15,  1.40it/s]
4|6198|Loss: 0.827036440372467:  96%|█████████▌| 6197/6470 [1:11:14<03:15,  1.40it/s] 
4|6198|Loss: 0.827036440372467:  96%|█████████▌| 6198/6470 [1:11:14<03:09,  1.44it/s]
4|6199|Loss: 0.717014729976654:  96%|█████████▌| 6198/6470 [1:11:15<03:09,  1.44it/s]
4|6199|Loss: 0.717014729976654:  96%|█████████▌| 6199/6470 [1:11:15<03:11,  1.42it/s]
4|6200|Loss: 0.6279972791671753:  96%|█████████▌| 6199/6470 [1:11:15<03:11,  1.42it/s]
4|6200|Loss: 0.6279972791671753:  96%|█████████▌| 6200/6470 [1:11:16<02:49,  1.59it/s]
4|6201|Loss: 0.8197119235992432:  96%|█████████▌| 6200/6470 [1:11:16<02:49,  1.59it/s]
4|6201|Loss: 0.8197119235992432:  96%|█████████▌| 6201/6470 [1:11:16<02:41,  1.66it/s]
4|6202|Loss: 0.956285834312439:  96%|█████████▌| 6201/6470 [1:11:16<02:41,  1.66it/s] 
4|6202|Loss: 0.956285834312439:  96%|█████████▌| 6202/6470 [1:11:17<02:47,  1.60it/s]
4|6203|Loss: 1.0253421068191528:  96%|█████████▌| 6202/6470 [1:11:17<02:47,  1.60it/s]
4|6203|Loss: 1.0253421068191528:  96%|█████████▌| 6203/6470 [1:11:17<02:55,  1.52it/s]
4|6204|Loss: 1.005739450454712:  96%|█████████▌| 6203/6470 [1:11:18<02:55,  1.52it/s] 
4|6204|Loss: 1.005739450454712:  96%|█████████▌| 6204/6470 [1:11:18<02:59,  1.48it/s]
4|6205|Loss: 0.8381006121635437:  96%|█████████▌| 6204/6470 [1:11:19<02:59,  1.48it/s]
4|6205|Loss: 0.8381006121635437:  96%|█████████▌| 6205/6470 [1:11:19<03:04,  1.44it/s]
4|6206|Loss: 0.7416532039642334:  96%|█████████▌| 6205/6470 [1:11:19<03:04,  1.44it/s]
4|6206|Loss: 0.7416532039642334:  96%|█████████▌| 6206/6470 [1:11:20<03:11,  1.38it/s]
4|6207|Loss: 1.1253260374069214:  96%|█████████▌| 6206/6470 [1:11:20<03:11,  1.38it/s]
4|6207|Loss: 1.1253260374069214:  96%|█████████▌| 6207/6470 [1:11:21<03:13,  1.36it/s]
4|6208|Loss: 0.7802602052688599:  96%|█████████▌| 6207/6470 [1:11:21<03:13,  1.36it/s]
4|6208|Loss: 0.7802602052688599:  96%|█████████▌| 6208/6470 [1:11:21<03:07,  1.40it/s]
4|6209|Loss: 0.7798445820808411:  96%|█████████▌| 6208/6470 [1:11:22<03:07,  1.40it/s]
4|6209|Loss: 0.7798445820808411:  96%|█████████▌| 6209/6470 [1:11:22<02:59,  1.46it/s]
4|6210|Loss: 0.7097339630126953:  96%|█████████▌| 6209/6470 [1:11:22<02:59,  1.46it/s]
4|6210|Loss: 0.7097339630126953:  96%|█████████▌| 6210/6470 [1:11:22<02:58,  1.46it/s]
4|6211|Loss: 0.9575574398040771:  96%|█████████▌| 6210/6470 [1:11:23<02:58,  1.46it/s]
4|6211|Loss: 0.9575574398040771:  96%|█████████▌| 6211/6470 [1:11:23<03:01,  1.43it/s]
4|6212|Loss: 0.7669925093650818:  96%|█████████▌| 6211/6470 [1:11:24<03:01,  1.43it/s]
4|6212|Loss: 0.7669925093650818:  96%|█████████▌| 6212/6470 [1:11:24<02:58,  1.45it/s]
4|6213|Loss: 0.8800654411315918:  96%|█████████▌| 6212/6470 [1:11:24<02:58,  1.45it/s]
4|6213|Loss: 0.8800654411315918:  96%|█████████▌| 6213/6470 [1:11:24<02:37,  1.63it/s]
4|6214|Loss: 1.0297377109527588:  96%|█████████▌| 6213/6470 [1:11:25<02:37,  1.63it/s]
4|6214|Loss: 1.0297377109527588:  96%|█████████▌| 6214/6470 [1:11:25<02:23,  1.79it/s]
4|6215|Loss: 0.7738948464393616:  96%|█████████▌| 6214/6470 [1:11:25<02:23,  1.79it/s]
4|6215|Loss: 0.7738948464393616:  96%|█████████▌| 6215/6470 [1:11:25<02:29,  1.70it/s]
4|6216|Loss: 0.9008165597915649:  96%|█████████▌| 6215/6470 [1:11:26<02:29,  1.70it/s]
4|6216|Loss: 0.9008165597915649:  96%|█████████▌| 6216/6470 [1:11:26<02:46,  1.53it/s]
4|6217|Loss: 0.8721452355384827:  96%|█████████▌| 6216/6470 [1:11:27<02:46,  1.53it/s]
4|6217|Loss: 0.8721452355384827:  96%|█████████▌| 6217/6470 [1:11:27<02:56,  1.43it/s]
4|6218|Loss: 0.8507803082466125:  96%|█████████▌| 6217/6470 [1:11:27<02:56,  1.43it/s]
4|6218|Loss: 0.8507803082466125:  96%|█████████▌| 6218/6470 [1:11:28<02:51,  1.47it/s]
4|6219|Loss: 0.7428719997406006:  96%|█████████▌| 6218/6470 [1:11:28<02:51,  1.47it/s]
4|6219|Loss: 0.7428719997406006:  96%|█████████▌| 6219/6470 [1:11:28<02:38,  1.58it/s]
4|6220|Loss: 0.8495916128158569:  96%|█████████▌| 6219/6470 [1:11:29<02:38,  1.58it/s]
4|6220|Loss: 0.8495916128158569:  96%|█████████▌| 6220/6470 [1:11:29<02:48,  1.48it/s]
4|6221|Loss: 0.6500343084335327:  96%|█████████▌| 6220/6470 [1:11:29<02:48,  1.48it/s]
4|6221|Loss: 0.6500343084335327:  96%|█████████▌| 6221/6470 [1:11:30<02:54,  1.42it/s]
4|6222|Loss: 0.9407484531402588:  96%|█████████▌| 6221/6470 [1:11:30<02:54,  1.42it/s]
4|6222|Loss: 0.9407484531402588:  96%|█████████▌| 6222/6470 [1:11:30<02:57,  1.39it/s]
4|6223|Loss: 0.8585346937179565:  96%|█████████▌| 6222/6470 [1:11:31<02:57,  1.39it/s]
4|6223|Loss: 0.8585346937179565:  96%|█████████▌| 6223/6470 [1:11:31<03:03,  1.35it/s]
4|6224|Loss: 0.7750955820083618:  96%|█████████▌| 6223/6470 [1:11:32<03:03,  1.35it/s]
4|6224|Loss: 0.7750955820083618:  96%|█████████▌| 6224/6470 [1:11:32<03:04,  1.33it/s]
4|6225|Loss: 0.758459746837616:  96%|█████████▌| 6224/6470 [1:11:32<03:04,  1.33it/s] 
4|6225|Loss: 0.758459746837616:  96%|█████████▌| 6225/6470 [1:11:33<02:56,  1.39it/s]
4|6226|Loss: 0.7429743409156799:  96%|█████████▌| 6225/6470 [1:11:33<02:56,  1.39it/s]
4|6226|Loss: 0.7429743409156799:  96%|█████████▌| 6226/6470 [1:11:33<02:44,  1.48it/s]
4|6227|Loss: 0.7575544118881226:  96%|█████████▌| 6226/6470 [1:11:34<02:44,  1.48it/s]
4|6227|Loss: 0.7575544118881226:  96%|█████████▌| 6227/6470 [1:11:34<02:34,  1.57it/s]
4|6228|Loss: 0.9518830180168152:  96%|█████████▌| 6227/6470 [1:11:34<02:34,  1.57it/s]
4|6228|Loss: 0.9518830180168152:  96%|█████████▋| 6228/6470 [1:11:34<02:38,  1.53it/s]
4|6229|Loss: 0.7856777906417847:  96%|█████████▋| 6228/6470 [1:11:35<02:38,  1.53it/s]
4|6229|Loss: 0.7856777906417847:  96%|█████████▋| 6229/6470 [1:11:35<02:49,  1.42it/s]
4|6230|Loss: 0.7202689051628113:  96%|█████████▋| 6229/6470 [1:11:36<02:49,  1.42it/s]
4|6230|Loss: 0.7202689051628113:  96%|█████████▋| 6230/6470 [1:11:36<02:47,  1.43it/s]
4|6231|Loss: 0.7472038269042969:  96%|█████████▋| 6230/6470 [1:11:36<02:47,  1.43it/s]
4|6231|Loss: 0.7472038269042969:  96%|█████████▋| 6231/6470 [1:11:37<02:48,  1.41it/s]
4|6232|Loss: 0.875828742980957:  96%|█████████▋| 6231/6470 [1:11:37<02:48,  1.41it/s] 
4|6232|Loss: 0.875828742980957:  96%|█████████▋| 6232/6470 [1:11:37<02:49,  1.41it/s]
4|6233|Loss: 0.9081286191940308:  96%|█████████▋| 6232/6470 [1:11:38<02:49,  1.41it/s]
4|6233|Loss: 0.9081286191940308:  96%|█████████▋| 6233/6470 [1:11:38<02:43,  1.45it/s]
4|6234|Loss: 1.0015625953674316:  96%|█████████▋| 6233/6470 [1:11:38<02:43,  1.45it/s]
4|6234|Loss: 1.0015625953674316:  96%|█████████▋| 6234/6470 [1:11:39<02:49,  1.39it/s]
4|6235|Loss: 0.9840333461761475:  96%|█████████▋| 6234/6470 [1:11:39<02:49,  1.39it/s]
4|6235|Loss: 0.9840333461761475:  96%|█████████▋| 6235/6470 [1:11:40<02:50,  1.38it/s]
4|6236|Loss: 0.8613670468330383:  96%|█████████▋| 6235/6470 [1:11:40<02:50,  1.38it/s]
4|6236|Loss: 0.8613670468330383:  96%|█████████▋| 6236/6470 [1:11:40<02:51,  1.37it/s]
4|6237|Loss: 0.887116551399231:  96%|█████████▋| 6236/6470 [1:11:41<02:51,  1.37it/s] 
4|6237|Loss: 0.887116551399231:  96%|█████████▋| 6237/6470 [1:11:41<02:46,  1.40it/s]
4|6238|Loss: 0.6935297846794128:  96%|█████████▋| 6237/6470 [1:11:41<02:46,  1.40it/s]
4|6238|Loss: 0.6935297846794128:  96%|█████████▋| 6238/6470 [1:11:42<02:52,  1.34it/s]
4|6239|Loss: 0.7956103682518005:  96%|█████████▋| 6238/6470 [1:11:42<02:52,  1.34it/s]
4|6239|Loss: 0.7956103682518005:  96%|█████████▋| 6239/6470 [1:11:42<02:44,  1.40it/s]
4|6240|Loss: 0.7769796252250671:  96%|█████████▋| 6239/6470 [1:11:43<02:44,  1.40it/s]
4|6240|Loss: 0.7769796252250671:  96%|█████████▋| 6240/6470 [1:11:43<02:37,  1.46it/s]
4|6241|Loss: 0.7798170447349548:  96%|█████████▋| 6240/6470 [1:11:43<02:37,  1.46it/s]
4|6241|Loss: 0.7798170447349548:  96%|█████████▋| 6241/6470 [1:11:44<02:37,  1.46it/s]
4|6242|Loss: 1.0000604391098022:  96%|█████████▋| 6241/6470 [1:11:44<02:37,  1.46it/s]
4|6242|Loss: 1.0000604391098022:  96%|█████████▋| 6242/6470 [1:11:45<02:39,  1.43it/s]
4|6243|Loss: 0.8800994157791138:  96%|█████████▋| 6242/6470 [1:11:45<02:39,  1.43it/s]
4|6243|Loss: 0.8800994157791138:  96%|█████████▋| 6243/6470 [1:11:45<02:40,  1.41it/s]
4|6244|Loss: 0.8564263582229614:  96%|█████████▋| 6243/6470 [1:11:46<02:40,  1.41it/s]
4|6244|Loss: 0.8564263582229614:  97%|█████████▋| 6244/6470 [1:11:46<02:40,  1.41it/s]
4|6245|Loss: 0.9470219016075134:  97%|█████████▋| 6244/6470 [1:11:46<02:40,  1.41it/s]
4|6245|Loss: 0.9470219016075134:  97%|█████████▋| 6245/6470 [1:11:47<02:36,  1.44it/s]
4|6246|Loss: 0.9656223654747009:  97%|█████████▋| 6245/6470 [1:11:47<02:36,  1.44it/s]
4|6246|Loss: 0.9656223654747009:  97%|█████████▋| 6246/6470 [1:11:47<02:37,  1.42it/s]
4|6247|Loss: 0.9669636487960815:  97%|█████████▋| 6246/6470 [1:11:48<02:37,  1.42it/s]
4|6247|Loss: 0.9669636487960815:  97%|█████████▋| 6247/6470 [1:11:48<02:39,  1.39it/s]
4|6248|Loss: 0.869750440120697:  97%|█████████▋| 6247/6470 [1:11:48<02:39,  1.39it/s] 
4|6248|Loss: 0.869750440120697:  97%|█████████▋| 6248/6470 [1:11:49<02:35,  1.43it/s]
4|6249|Loss: 0.8539206981658936:  97%|█████████▋| 6248/6470 [1:11:49<02:35,  1.43it/s]
4|6249|Loss: 0.8539206981658936:  97%|█████████▋| 6249/6470 [1:11:49<02:30,  1.47it/s]
4|6250|Loss: 0.860297441482544:  97%|█████████▋| 6249/6470 [1:11:50<02:30,  1.47it/s] 
4|6250|Loss: 0.860297441482544:  97%|█████████▋| 6250/6470 [1:11:50<02:32,  1.44it/s]
4|6251|Loss: 0.7279242277145386:  97%|█████████▋| 6250/6470 [1:11:50<02:32,  1.44it/s]
4|6251|Loss: 0.7279242277145386:  97%|█████████▋| 6251/6470 [1:11:51<02:22,  1.54it/s]
4|6252|Loss: 0.9809336066246033:  97%|█████████▋| 6251/6470 [1:11:51<02:22,  1.54it/s]
4|6252|Loss: 0.9809336066246033:  97%|█████████▋| 6252/6470 [1:11:51<02:22,  1.53it/s]
4|6253|Loss: 0.8203427791595459:  97%|█████████▋| 6252/6470 [1:11:52<02:22,  1.53it/s]
4|6253|Loss: 0.8203427791595459:  97%|█████████▋| 6253/6470 [1:11:52<02:32,  1.43it/s]
4|6254|Loss: 0.8528268933296204:  97%|█████████▋| 6253/6470 [1:11:53<02:32,  1.43it/s]
4|6254|Loss: 0.8528268933296204:  97%|█████████▋| 6254/6470 [1:11:53<02:29,  1.45it/s]
4|6255|Loss: 0.9027780294418335:  97%|█████████▋| 6254/6470 [1:11:53<02:29,  1.45it/s]
4|6255|Loss: 0.9027780294418335:  97%|█████████▋| 6255/6470 [1:11:54<02:30,  1.43it/s]
4|6256|Loss: 0.7978670001029968:  97%|█████████▋| 6255/6470 [1:11:54<02:30,  1.43it/s]
4|6256|Loss: 0.7978670001029968:  97%|█████████▋| 6256/6470 [1:11:54<02:36,  1.37it/s]
4|6257|Loss: 0.8066977858543396:  97%|█████████▋| 6256/6470 [1:11:55<02:36,  1.37it/s]
4|6257|Loss: 0.8066977858543396:  97%|█████████▋| 6257/6470 [1:11:55<02:31,  1.41it/s]
4|6258|Loss: 0.8557270765304565:  97%|█████████▋| 6257/6470 [1:11:55<02:31,  1.41it/s]
4|6258|Loss: 0.8557270765304565:  97%|█████████▋| 6258/6470 [1:11:56<02:28,  1.43it/s]
4|6259|Loss: 0.819918692111969:  97%|█████████▋| 6258/6470 [1:11:56<02:28,  1.43it/s] 
4|6259|Loss: 0.819918692111969:  97%|█████████▋| 6259/6470 [1:11:56<02:24,  1.46it/s]
4|6260|Loss: 0.94019615650177:  97%|█████████▋| 6259/6470 [1:11:57<02:24,  1.46it/s] 
4|6260|Loss: 0.94019615650177:  97%|█████████▋| 6260/6470 [1:11:57<02:14,  1.56it/s]
4|6261|Loss: 0.8580524921417236:  97%|█████████▋| 6260/6470 [1:11:57<02:14,  1.56it/s]
4|6261|Loss: 0.8580524921417236:  97%|█████████▋| 6261/6470 [1:11:57<02:11,  1.59it/s]
4|6262|Loss: 0.859519362449646:  97%|█████████▋| 6261/6470 [1:11:58<02:11,  1.59it/s] 
4|6262|Loss: 0.859519362449646:  97%|█████████▋| 6262/6470 [1:11:58<02:16,  1.52it/s]
4|6263|Loss: 0.9907464385032654:  97%|█████████▋| 6262/6470 [1:11:59<02:16,  1.52it/s]
4|6263|Loss: 0.9907464385032654:  97%|█████████▋| 6263/6470 [1:11:59<02:24,  1.44it/s]
4|6264|Loss: 0.8637651801109314:  97%|█████████▋| 6263/6470 [1:11:59<02:24,  1.44it/s]
4|6264|Loss: 0.8637651801109314:  97%|█████████▋| 6264/6470 [1:11:59<02:12,  1.55it/s]
4|6265|Loss: 0.7972844243049622:  97%|█████████▋| 6264/6470 [1:12:00<02:12,  1.55it/s]
4|6265|Loss: 0.7972844243049622:  97%|█████████▋| 6265/6470 [1:12:00<02:06,  1.63it/s]
4|6266|Loss: 0.7892667651176453:  97%|█████████▋| 6265/6470 [1:12:00<02:06,  1.63it/s]
4|6266|Loss: 0.7892667651176453:  97%|█████████▋| 6266/6470 [1:12:01<02:03,  1.65it/s]
4|6267|Loss: 0.7372175455093384:  97%|█████████▋| 6266/6470 [1:12:01<02:03,  1.65it/s]
4|6267|Loss: 0.7372175455093384:  97%|█████████▋| 6267/6470 [1:12:01<02:09,  1.56it/s]
4|6268|Loss: 0.8009100556373596:  97%|█████████▋| 6267/6470 [1:12:02<02:09,  1.56it/s]
4|6268|Loss: 0.8009100556373596:  97%|█████████▋| 6268/6470 [1:12:02<02:13,  1.51it/s]
4|6269|Loss: 1.0473235845565796:  97%|█████████▋| 6268/6470 [1:12:02<02:13,  1.51it/s]
4|6269|Loss: 1.0473235845565796:  97%|█████████▋| 6269/6470 [1:12:03<02:05,  1.61it/s]
4|6270|Loss: 0.8408974409103394:  97%|█████████▋| 6269/6470 [1:12:03<02:05,  1.61it/s]
4|6270|Loss: 0.8408974409103394:  97%|█████████▋| 6270/6470 [1:12:03<02:12,  1.51it/s]
4|6271|Loss: 0.7910164594650269:  97%|█████████▋| 6270/6470 [1:12:04<02:12,  1.51it/s]
4|6271|Loss: 0.7910164594650269:  97%|█████████▋| 6271/6470 [1:12:04<02:16,  1.46it/s]
4|6272|Loss: 0.8070160746574402:  97%|█████████▋| 6271/6470 [1:12:04<02:16,  1.46it/s]
4|6272|Loss: 0.8070160746574402:  97%|█████████▋| 6272/6470 [1:12:05<02:16,  1.45it/s]
4|6273|Loss: 0.8704990148544312:  97%|█████████▋| 6272/6470 [1:12:05<02:16,  1.45it/s]
4|6273|Loss: 0.8704990148544312:  97%|█████████▋| 6273/6470 [1:12:05<02:16,  1.44it/s]
4|6274|Loss: 0.8074026703834534:  97%|█████████▋| 6273/6470 [1:12:06<02:16,  1.44it/s]
4|6274|Loss: 0.8074026703834534:  97%|█████████▋| 6274/6470 [1:12:06<02:18,  1.41it/s]
4|6275|Loss: 0.8237279057502747:  97%|█████████▋| 6274/6470 [1:12:07<02:18,  1.41it/s]
4|6275|Loss: 0.8237279057502747:  97%|█████████▋| 6275/6470 [1:12:07<02:14,  1.45it/s]
4|6276|Loss: 0.750552773475647:  97%|█████████▋| 6275/6470 [1:12:07<02:14,  1.45it/s] 
4|6276|Loss: 0.750552773475647:  97%|█████████▋| 6276/6470 [1:12:08<02:17,  1.41it/s]
4|6277|Loss: 0.7820428013801575:  97%|█████████▋| 6276/6470 [1:12:08<02:17,  1.41it/s]
4|6277|Loss: 0.7820428013801575:  97%|█████████▋| 6277/6470 [1:12:08<02:21,  1.36it/s]
4|6278|Loss: 0.7576196193695068:  97%|█████████▋| 6277/6470 [1:12:09<02:21,  1.36it/s]
4|6278|Loss: 0.7576196193695068:  97%|█████████▋| 6278/6470 [1:12:09<02:17,  1.40it/s]
4|6279|Loss: 0.8303259015083313:  97%|█████████▋| 6278/6470 [1:12:09<02:17,  1.40it/s]
4|6279|Loss: 0.8303259015083313:  97%|█████████▋| 6279/6470 [1:12:10<02:15,  1.41it/s]
4|6280|Loss: 0.8713784217834473:  97%|█████████▋| 6279/6470 [1:12:10<02:15,  1.41it/s]
4|6280|Loss: 0.8713784217834473:  97%|█████████▋| 6280/6470 [1:12:11<02:15,  1.40it/s]
4|6281|Loss: 0.8827458620071411:  97%|█████████▋| 6280/6470 [1:12:11<02:15,  1.40it/s]
4|6281|Loss: 0.8827458620071411:  97%|█████████▋| 6281/6470 [1:12:11<02:14,  1.40it/s]
4|6282|Loss: 0.8232699036598206:  97%|█████████▋| 6281/6470 [1:12:12<02:14,  1.40it/s]
4|6282|Loss: 0.8232699036598206:  97%|█████████▋| 6282/6470 [1:12:12<02:13,  1.41it/s]
4|6283|Loss: 0.9696846008300781:  97%|█████████▋| 6282/6470 [1:12:12<02:13,  1.41it/s]
4|6283|Loss: 0.9696846008300781:  97%|█████████▋| 6283/6470 [1:12:13<02:18,  1.35it/s]
4|6284|Loss: 0.8135743737220764:  97%|█████████▋| 6283/6470 [1:12:13<02:18,  1.35it/s]
4|6284|Loss: 0.8135743737220764:  97%|█████████▋| 6284/6470 [1:12:13<02:15,  1.37it/s]
4|6285|Loss: 0.7206255197525024:  97%|█████████▋| 6284/6470 [1:12:14<02:15,  1.37it/s]
4|6285|Loss: 0.7206255197525024:  97%|█████████▋| 6285/6470 [1:12:14<02:16,  1.36it/s]
4|6286|Loss: 0.8456516861915588:  97%|█████████▋| 6285/6470 [1:12:15<02:16,  1.36it/s]
4|6286|Loss: 0.8456516861915588:  97%|█████████▋| 6286/6470 [1:12:15<02:16,  1.35it/s]
4|6287|Loss: 0.7958701848983765:  97%|█████████▋| 6286/6470 [1:12:15<02:16,  1.35it/s]
4|6287|Loss: 0.7958701848983765:  97%|█████████▋| 6287/6470 [1:12:16<02:14,  1.36it/s]
4|6288|Loss: 0.8430447578430176:  97%|█████████▋| 6287/6470 [1:12:16<02:14,  1.36it/s]
4|6288|Loss: 0.8430447578430176:  97%|█████████▋| 6288/6470 [1:12:16<02:07,  1.43it/s]
4|6289|Loss: 1.062593936920166:  97%|█████████▋| 6288/6470 [1:12:17<02:07,  1.43it/s] 
4|6289|Loss: 1.062593936920166:  97%|█████████▋| 6289/6470 [1:12:17<01:57,  1.54it/s]
4|6290|Loss: 0.9540524482727051:  97%|█████████▋| 6289/6470 [1:12:17<01:57,  1.54it/s]
4|6290|Loss: 0.9540524482727051:  97%|█████████▋| 6290/6470 [1:12:17<01:54,  1.57it/s]
4|6291|Loss: 0.8325383067131042:  97%|█████████▋| 6290/6470 [1:12:18<01:54,  1.57it/s]
4|6291|Loss: 0.8325383067131042:  97%|█████████▋| 6291/6470 [1:12:18<01:58,  1.51it/s]
4|6292|Loss: 0.876939058303833:  97%|█████████▋| 6291/6470 [1:12:19<01:58,  1.51it/s] 
4|6292|Loss: 0.876939058303833:  97%|█████████▋| 6292/6470 [1:12:19<02:02,  1.46it/s]
4|6293|Loss: 0.7879711985588074:  97%|█████████▋| 6292/6470 [1:12:19<02:02,  1.46it/s]
4|6293|Loss: 0.7879711985588074:  97%|█████████▋| 6293/6470 [1:12:20<02:00,  1.47it/s]
4|6294|Loss: 0.8727638721466064:  97%|█████████▋| 6293/6470 [1:12:20<02:00,  1.47it/s]
4|6294|Loss: 0.8727638721466064:  97%|█████████▋| 6294/6470 [1:12:20<02:00,  1.46it/s]
4|6295|Loss: 0.7840375900268555:  97%|█████████▋| 6294/6470 [1:12:21<02:00,  1.46it/s]
4|6295|Loss: 0.7840375900268555:  97%|█████████▋| 6295/6470 [1:12:21<01:58,  1.48it/s]
4|6296|Loss: 0.8838284015655518:  97%|█████████▋| 6295/6470 [1:12:21<01:58,  1.48it/s]
4|6296|Loss: 0.8838284015655518:  97%|█████████▋| 6296/6470 [1:12:22<01:58,  1.47it/s]
4|6297|Loss: 0.7817201614379883:  97%|█████████▋| 6296/6470 [1:12:22<01:58,  1.47it/s]
4|6297|Loss: 0.7817201614379883:  97%|█████████▋| 6297/6470 [1:12:22<01:55,  1.50it/s]
4|6298|Loss: 0.9474047422409058:  97%|█████████▋| 6297/6470 [1:12:23<01:55,  1.50it/s]
4|6298|Loss: 0.9474047422409058:  97%|█████████▋| 6298/6470 [1:12:23<01:53,  1.52it/s]
4|6299|Loss: 0.9511132836341858:  97%|█████████▋| 6298/6470 [1:12:23<01:53,  1.52it/s]
4|6299|Loss: 0.9511132836341858:  97%|█████████▋| 6299/6470 [1:12:23<01:49,  1.56it/s]
4|6300|Loss: 0.8349970579147339:  97%|█████████▋| 6299/6470 [1:12:24<01:49,  1.56it/s]
4|6300|Loss: 0.8349970579147339:  97%|█████████▋| 6300/6470 [1:12:24<01:48,  1.57it/s]
4|6301|Loss: 0.8039699792861938:  97%|█████████▋| 6300/6470 [1:12:24<01:48,  1.57it/s]
4|6301|Loss: 0.8039699792861938:  97%|█████████▋| 6301/6470 [1:12:25<01:50,  1.53it/s]
4|6302|Loss: 0.9062774777412415:  97%|█████████▋| 6301/6470 [1:12:25<01:50,  1.53it/s]
4|6302|Loss: 0.9062774777412415:  97%|█████████▋| 6302/6470 [1:12:25<01:42,  1.64it/s]
4|6303|Loss: 0.8625025153160095:  97%|█████████▋| 6302/6470 [1:12:26<01:42,  1.64it/s]
4|6303|Loss: 0.8625025153160095:  97%|█████████▋| 6303/6470 [1:12:26<01:47,  1.55it/s]
4|6304|Loss: 0.9953086376190186:  97%|█████████▋| 6303/6470 [1:12:26<01:47,  1.55it/s]
4|6304|Loss: 0.9953086376190186:  97%|█████████▋| 6304/6470 [1:12:27<01:46,  1.57it/s]
4|6305|Loss: 0.9305282235145569:  97%|█████████▋| 6304/6470 [1:12:27<01:46,  1.57it/s]
4|6305|Loss: 0.9305282235145569:  97%|█████████▋| 6305/6470 [1:12:27<01:49,  1.51it/s]
4|6306|Loss: 1.0024166107177734:  97%|█████████▋| 6305/6470 [1:12:28<01:49,  1.51it/s]
4|6306|Loss: 1.0024166107177734:  97%|█████████▋| 6306/6470 [1:12:28<01:48,  1.52it/s]
4|6307|Loss: 0.7615693807601929:  97%|█████████▋| 6306/6470 [1:12:28<01:48,  1.52it/s]
4|6307|Loss: 0.7615693807601929:  97%|█████████▋| 6307/6470 [1:12:29<01:41,  1.61it/s]
4|6308|Loss: 0.8484529256820679:  97%|█████████▋| 6307/6470 [1:12:29<01:41,  1.61it/s]
4|6308|Loss: 0.8484529256820679:  97%|█████████▋| 6308/6470 [1:12:29<01:40,  1.62it/s]
4|6309|Loss: 0.8156482577323914:  97%|█████████▋| 6308/6470 [1:12:30<01:40,  1.62it/s]
4|6309|Loss: 0.8156482577323914:  98%|█████████▊| 6309/6470 [1:12:30<01:46,  1.51it/s]
4|6310|Loss: 0.8614839315414429:  98%|█████████▊| 6309/6470 [1:12:30<01:46,  1.51it/s]
4|6310|Loss: 0.8614839315414429:  98%|█████████▊| 6310/6470 [1:12:31<01:46,  1.50it/s]
4|6311|Loss: 0.9068460464477539:  98%|█████████▊| 6310/6470 [1:12:31<01:46,  1.50it/s]
4|6311|Loss: 0.9068460464477539:  98%|█████████▊| 6311/6470 [1:12:31<01:34,  1.68it/s]
4|6312|Loss: 1.0286900997161865:  98%|█████████▊| 6311/6470 [1:12:31<01:34,  1.68it/s]
4|6312|Loss: 1.0286900997161865:  98%|█████████▊| 6312/6470 [1:12:32<01:41,  1.55it/s]
4|6313|Loss: 0.8353854417800903:  98%|█████████▊| 6312/6470 [1:12:32<01:41,  1.55it/s]
4|6313|Loss: 0.8353854417800903:  98%|█████████▊| 6313/6470 [1:12:32<01:40,  1.57it/s]
4|6314|Loss: 0.8095079064369202:  98%|█████████▊| 6313/6470 [1:12:33<01:40,  1.57it/s]
4|6314|Loss: 0.8095079064369202:  98%|█████████▊| 6314/6470 [1:12:33<01:33,  1.66it/s]
4|6315|Loss: 0.753724217414856:  98%|█████████▊| 6314/6470 [1:12:33<01:33,  1.66it/s] 
4|6315|Loss: 0.753724217414856:  98%|█████████▊| 6315/6470 [1:12:34<01:33,  1.67it/s]
4|6316|Loss: 0.9391338229179382:  98%|█████████▊| 6315/6470 [1:12:34<01:33,  1.67it/s]
4|6316|Loss: 0.9391338229179382:  98%|█████████▊| 6316/6470 [1:12:34<01:36,  1.59it/s]
4|6317|Loss: 0.7989506721496582:  98%|█████████▊| 6316/6470 [1:12:35<01:36,  1.59it/s]
4|6317|Loss: 0.7989506721496582:  98%|█████████▊| 6317/6470 [1:12:35<01:36,  1.59it/s]
4|6318|Loss: 0.9896664023399353:  98%|█████████▊| 6317/6470 [1:12:35<01:36,  1.59it/s]
4|6318|Loss: 0.9896664023399353:  98%|█████████▊| 6318/6470 [1:12:36<01:41,  1.50it/s]
4|6319|Loss: 0.8470518589019775:  98%|█████████▊| 6318/6470 [1:12:36<01:41,  1.50it/s]
4|6319|Loss: 0.8470518589019775:  98%|█████████▊| 6319/6470 [1:12:36<01:43,  1.46it/s]
4|6320|Loss: 0.89473956823349:  98%|█████████▊| 6319/6470 [1:12:37<01:43,  1.46it/s]  
4|6320|Loss: 0.89473956823349:  98%|█████████▊| 6320/6470 [1:12:37<01:44,  1.44it/s]
4|6321|Loss: 0.8775365948677063:  98%|█████████▊| 6320/6470 [1:12:37<01:44,  1.44it/s]
4|6321|Loss: 0.8775365948677063:  98%|█████████▊| 6321/6470 [1:12:38<01:45,  1.41it/s]
4|6322|Loss: 0.7923882007598877:  98%|█████████▊| 6321/6470 [1:12:38<01:45,  1.41it/s]
4|6322|Loss: 0.7923882007598877:  98%|█████████▊| 6322/6470 [1:12:38<01:39,  1.49it/s]
4|6323|Loss: 0.745912492275238:  98%|█████████▊| 6322/6470 [1:12:39<01:39,  1.49it/s] 
4|6323|Loss: 0.745912492275238:  98%|█████████▊| 6323/6470 [1:12:39<01:44,  1.41it/s]
4|6324|Loss: 0.7511091232299805:  98%|█████████▊| 6323/6470 [1:12:40<01:44,  1.41it/s]
4|6324|Loss: 0.7511091232299805:  98%|█████████▊| 6324/6470 [1:12:40<01:42,  1.42it/s]
4|6325|Loss: 0.8201135993003845:  98%|█████████▊| 6324/6470 [1:12:40<01:42,  1.42it/s]
4|6325|Loss: 0.8201135993003845:  98%|█████████▊| 6325/6470 [1:12:40<01:38,  1.47it/s]
4|6326|Loss: 0.9208643436431885:  98%|█████████▊| 6325/6470 [1:12:41<01:38,  1.47it/s]
4|6326|Loss: 0.9208643436431885:  98%|█████████▊| 6326/6470 [1:12:41<01:40,  1.44it/s]
4|6327|Loss: 0.8749865889549255:  98%|█████████▊| 6326/6470 [1:12:42<01:40,  1.44it/s]
4|6327|Loss: 0.8749865889549255:  98%|█████████▊| 6327/6470 [1:12:42<01:35,  1.49it/s]
4|6328|Loss: 0.8147117495536804:  98%|█████████▊| 6327/6470 [1:12:42<01:35,  1.49it/s]
4|6328|Loss: 0.8147117495536804:  98%|█████████▊| 6328/6470 [1:12:42<01:33,  1.52it/s]
4|6329|Loss: 1.01534104347229:  98%|█████████▊| 6328/6470 [1:12:43<01:33,  1.52it/s]  
4|6329|Loss: 1.01534104347229:  98%|█████████▊| 6329/6470 [1:12:43<01:38,  1.43it/s]
4|6330|Loss: 0.5615472793579102:  98%|█████████▊| 6329/6470 [1:12:44<01:38,  1.43it/s]
4|6330|Loss: 0.5615472793579102:  98%|█████████▊| 6330/6470 [1:12:44<01:50,  1.26it/s]
4|6331|Loss: 0.8734564781188965:  98%|█████████▊| 6330/6470 [1:12:45<01:50,  1.26it/s]
4|6331|Loss: 0.8734564781188965:  98%|█████████▊| 6331/6470 [1:12:45<01:52,  1.23it/s]
4|6332|Loss: 0.7791019678115845:  98%|█████████▊| 6331/6470 [1:12:46<01:52,  1.23it/s]
4|6332|Loss: 0.7791019678115845:  98%|█████████▊| 6332/6470 [1:12:46<01:49,  1.27it/s]
4|6333|Loss: 0.8482419848442078:  98%|█████████▊| 6332/6470 [1:12:46<01:49,  1.27it/s]
4|6333|Loss: 0.8482419848442078:  98%|█████████▊| 6333/6470 [1:12:47<01:42,  1.33it/s]
4|6334|Loss: 0.8085796236991882:  98%|█████████▊| 6333/6470 [1:12:47<01:42,  1.33it/s]
4|6334|Loss: 0.8085796236991882:  98%|█████████▊| 6334/6470 [1:12:47<01:39,  1.37it/s]
4|6335|Loss: 0.8437502980232239:  98%|█████████▊| 6334/6470 [1:12:48<01:39,  1.37it/s]
4|6335|Loss: 0.8437502980232239:  98%|█████████▊| 6335/6470 [1:12:48<01:37,  1.38it/s]
4|6336|Loss: 0.9177798628807068:  98%|█████████▊| 6335/6470 [1:12:48<01:37,  1.38it/s]
4|6336|Loss: 0.9177798628807068:  98%|█████████▊| 6336/6470 [1:12:49<01:34,  1.42it/s]
4|6337|Loss: 0.960302472114563:  98%|█████████▊| 6336/6470 [1:12:49<01:34,  1.42it/s] 
4|6337|Loss: 0.960302472114563:  98%|█████████▊| 6337/6470 [1:12:49<01:33,  1.42it/s]
4|6338|Loss: 0.7984208464622498:  98%|█████████▊| 6337/6470 [1:12:50<01:33,  1.42it/s]
4|6338|Loss: 0.7984208464622498:  98%|█████████▊| 6338/6470 [1:12:50<01:37,  1.36it/s]
4|6339|Loss: 0.8730974793434143:  98%|█████████▊| 6338/6470 [1:12:50<01:37,  1.36it/s]
4|6339|Loss: 0.8730974793434143:  98%|█████████▊| 6339/6470 [1:12:51<01:33,  1.40it/s]
4|6340|Loss: 0.7701090574264526:  98%|█████████▊| 6339/6470 [1:12:51<01:33,  1.40it/s]
4|6340|Loss: 0.7701090574264526:  98%|█████████▊| 6340/6470 [1:12:51<01:32,  1.41it/s]
4|6341|Loss: 0.8387483954429626:  98%|█████████▊| 6340/6470 [1:12:52<01:32,  1.41it/s]
4|6341|Loss: 0.8387483954429626:  98%|█████████▊| 6341/6470 [1:12:52<01:26,  1.49it/s]
4|6342|Loss: 0.8354222774505615:  98%|█████████▊| 6341/6470 [1:12:52<01:26,  1.49it/s]
4|6342|Loss: 0.8354222774505615:  98%|█████████▊| 6342/6470 [1:12:53<01:25,  1.49it/s]
4|6343|Loss: 0.8035545945167542:  98%|█████████▊| 6342/6470 [1:12:53<01:25,  1.49it/s]
4|6343|Loss: 0.8035545945167542:  98%|█████████▊| 6343/6470 [1:12:53<01:21,  1.56it/s]
4|6344|Loss: 0.9461942315101624:  98%|█████████▊| 6343/6470 [1:12:54<01:21,  1.56it/s]
4|6344|Loss: 0.9461942315101624:  98%|█████████▊| 6344/6470 [1:12:54<01:21,  1.55it/s]
4|6345|Loss: 0.8973810076713562:  98%|█████████▊| 6344/6470 [1:12:54<01:21,  1.55it/s]
4|6345|Loss: 0.8973810076713562:  98%|█████████▊| 6345/6470 [1:12:54<01:15,  1.65it/s]
4|6346|Loss: 0.8165213465690613:  98%|█████████▊| 6345/6470 [1:12:55<01:15,  1.65it/s]
4|6346|Loss: 0.8165213465690613:  98%|█████████▊| 6346/6470 [1:12:55<01:14,  1.66it/s]
4|6347|Loss: 0.7671244740486145:  98%|█████████▊| 6346/6470 [1:12:55<01:14,  1.66it/s]
4|6347|Loss: 0.7671244740486145:  98%|█████████▊| 6347/6470 [1:12:56<01:13,  1.67it/s]
4|6348|Loss: 0.9361950159072876:  98%|█████████▊| 6347/6470 [1:12:56<01:13,  1.67it/s]
4|6348|Loss: 0.9361950159072876:  98%|█████████▊| 6348/6470 [1:12:56<01:16,  1.59it/s]
4|6349|Loss: 0.8976227045059204:  98%|█████████▊| 6348/6470 [1:12:57<01:16,  1.59it/s]
4|6349|Loss: 0.8976227045059204:  98%|█████████▊| 6349/6470 [1:12:57<01:19,  1.51it/s]
4|6350|Loss: 0.8204894661903381:  98%|█████████▊| 6349/6470 [1:12:58<01:19,  1.51it/s]
4|6350|Loss: 0.8204894661903381:  98%|█████████▊| 6350/6470 [1:12:58<01:25,  1.40it/s]
4|6351|Loss: 0.9542508125305176:  98%|█████████▊| 6350/6470 [1:12:58<01:25,  1.40it/s]
4|6351|Loss: 0.9542508125305176:  98%|█████████▊| 6351/6470 [1:12:59<01:30,  1.32it/s]
4|6352|Loss: 0.8774175047874451:  98%|█████████▊| 6351/6470 [1:12:59<01:30,  1.32it/s]
4|6352|Loss: 0.8774175047874451:  98%|█████████▊| 6352/6470 [1:13:00<01:29,  1.32it/s]
4|6353|Loss: 0.7074962258338928:  98%|█████████▊| 6352/6470 [1:13:00<01:29,  1.32it/s]
4|6353|Loss: 0.7074962258338928:  98%|█████████▊| 6353/6470 [1:13:00<01:20,  1.46it/s]
4|6354|Loss: 0.8547914028167725:  98%|█████████▊| 6353/6470 [1:13:00<01:20,  1.46it/s]
4|6354|Loss: 0.8547914028167725:  98%|█████████▊| 6354/6470 [1:13:01<01:22,  1.41it/s]
4|6355|Loss: 0.8479170799255371:  98%|█████████▊| 6354/6470 [1:13:01<01:22,  1.41it/s]
4|6355|Loss: 0.8479170799255371:  98%|█████████▊| 6355/6470 [1:13:01<01:18,  1.47it/s]
4|6356|Loss: 0.7659082412719727:  98%|█████████▊| 6355/6470 [1:13:02<01:18,  1.47it/s]
4|6356|Loss: 0.7659082412719727:  98%|█████████▊| 6356/6470 [1:13:02<01:11,  1.60it/s]
4|6357|Loss: 0.8109172582626343:  98%|█████████▊| 6356/6470 [1:13:02<01:11,  1.60it/s]
4|6357|Loss: 0.8109172582626343:  98%|█████████▊| 6357/6470 [1:13:03<01:09,  1.62it/s]
4|6358|Loss: 0.7174674272537231:  98%|█████████▊| 6357/6470 [1:13:03<01:09,  1.62it/s]
4|6358|Loss: 0.7174674272537231:  98%|█████████▊| 6358/6470 [1:13:03<01:16,  1.46it/s]
4|6359|Loss: 0.8381268382072449:  98%|█████████▊| 6358/6470 [1:13:04<01:16,  1.46it/s]
4|6359|Loss: 0.8381268382072449:  98%|█████████▊| 6359/6470 [1:13:04<01:20,  1.38it/s]
4|6360|Loss: 0.7992921471595764:  98%|█████████▊| 6359/6470 [1:13:05<01:20,  1.38it/s]
4|6360|Loss: 0.7992921471595764:  98%|█████████▊| 6360/6470 [1:13:05<01:18,  1.41it/s]
4|6361|Loss: 0.8668595552444458:  98%|█████████▊| 6360/6470 [1:13:05<01:18,  1.41it/s]
4|6361|Loss: 0.8668595552444458:  98%|█████████▊| 6361/6470 [1:13:06<01:16,  1.42it/s]
4|6362|Loss: 0.877947211265564:  98%|█████████▊| 6361/6470 [1:13:06<01:16,  1.42it/s] 
4|6362|Loss: 0.877947211265564:  98%|█████████▊| 6362/6470 [1:13:06<01:12,  1.50it/s]
4|6363|Loss: 0.8966139554977417:  98%|█████████▊| 6362/6470 [1:13:06<01:12,  1.50it/s]
4|6363|Loss: 0.8966139554977417:  98%|█████████▊| 6363/6470 [1:13:07<01:09,  1.54it/s]
4|6364|Loss: 0.9310742020606995:  98%|█████████▊| 6363/6470 [1:13:07<01:09,  1.54it/s]
4|6364|Loss: 0.9310742020606995:  98%|█████████▊| 6364/6470 [1:13:07<01:05,  1.62it/s]
4|6365|Loss: 0.8972743153572083:  98%|█████████▊| 6364/6470 [1:13:08<01:05,  1.62it/s]
4|6365|Loss: 0.8972743153572083:  98%|█████████▊| 6365/6470 [1:13:08<01:04,  1.62it/s]
4|6366|Loss: 0.818126380443573:  98%|█████████▊| 6365/6470 [1:13:08<01:04,  1.62it/s] 
4|6366|Loss: 0.818126380443573:  98%|█████████▊| 6366/6470 [1:13:08<01:03,  1.63it/s]
4|6367|Loss: 0.740280270576477:  98%|█████████▊| 6366/6470 [1:13:09<01:03,  1.63it/s]
4|6367|Loss: 0.740280270576477:  98%|█████████▊| 6367/6470 [1:13:09<01:10,  1.47it/s]
4|6368|Loss: 0.8870997428894043:  98%|█████████▊| 6367/6470 [1:13:10<01:10,  1.47it/s]
4|6368|Loss: 0.8870997428894043:  98%|█████████▊| 6368/6470 [1:13:10<01:16,  1.33it/s]
4|6369|Loss: 0.7272089719772339:  98%|█████████▊| 6368/6470 [1:13:11<01:16,  1.33it/s]
4|6369|Loss: 0.7272089719772339:  98%|█████████▊| 6369/6470 [1:13:11<01:28,  1.15it/s]
4|6370|Loss: 0.8062151670455933:  98%|█████████▊| 6369/6470 [1:13:12<01:28,  1.15it/s]
4|6370|Loss: 0.8062151670455933:  98%|█████████▊| 6370/6470 [1:13:12<01:19,  1.26it/s]
4|6371|Loss: 0.9471073746681213:  98%|█████████▊| 6370/6470 [1:13:12<01:19,  1.26it/s]
4|6371|Loss: 0.9471073746681213:  98%|█████████▊| 6371/6470 [1:13:13<01:14,  1.32it/s]
4|6372|Loss: 0.837145209312439:  98%|█████████▊| 6371/6470 [1:13:13<01:14,  1.32it/s] 
4|6372|Loss: 0.837145209312439:  98%|█████████▊| 6372/6470 [1:13:13<01:11,  1.37it/s]
4|6373|Loss: 0.7891498804092407:  98%|█████████▊| 6372/6470 [1:13:14<01:11,  1.37it/s]
4|6373|Loss: 0.7891498804092407:  99%|█████████▊| 6373/6470 [1:13:14<01:12,  1.34it/s]
4|6374|Loss: 0.8350854516029358:  99%|█████████▊| 6373/6470 [1:13:15<01:12,  1.34it/s]
4|6374|Loss: 0.8350854516029358:  99%|█████████▊| 6374/6470 [1:13:15<01:13,  1.31it/s]
4|6375|Loss: 0.8419808149337769:  99%|█████████▊| 6374/6470 [1:13:15<01:13,  1.31it/s]
4|6375|Loss: 0.8419808149337769:  99%|█████████▊| 6375/6470 [1:13:16<01:08,  1.38it/s]
4|6376|Loss: 0.8943275213241577:  99%|█████████▊| 6375/6470 [1:13:16<01:08,  1.38it/s]
4|6376|Loss: 0.8943275213241577:  99%|█████████▊| 6376/6470 [1:13:17<01:18,  1.19it/s]
4|6377|Loss: 0.7173782587051392:  99%|█████████▊| 6376/6470 [1:13:17<01:18,  1.19it/s]
4|6377|Loss: 0.7173782587051392:  99%|█████████▊| 6377/6470 [1:13:18<01:17,  1.19it/s]
4|6378|Loss: 0.722756028175354:  99%|█████████▊| 6377/6470 [1:13:18<01:17,  1.19it/s] 
4|6378|Loss: 0.722756028175354:  99%|█████████▊| 6378/6470 [1:13:18<01:14,  1.23it/s]
4|6379|Loss: 0.9732484221458435:  99%|█████████▊| 6378/6470 [1:13:19<01:14,  1.23it/s]
4|6379|Loss: 0.9732484221458435:  99%|█████████▊| 6379/6470 [1:13:19<01:11,  1.26it/s]
4|6380|Loss: 0.7606319785118103:  99%|█████████▊| 6379/6470 [1:13:19<01:11,  1.26it/s]
4|6380|Loss: 0.7606319785118103:  99%|█████████▊| 6380/6470 [1:13:19<01:01,  1.47it/s]
4|6381|Loss: 0.7563952207565308:  99%|█████████▊| 6380/6470 [1:13:20<01:01,  1.47it/s]
4|6381|Loss: 0.7563952207565308:  99%|█████████▊| 6381/6470 [1:13:20<01:03,  1.41it/s]
4|6382|Loss: 0.8247649669647217:  99%|█████████▊| 6381/6470 [1:13:21<01:03,  1.41it/s]
4|6382|Loss: 0.8247649669647217:  99%|█████████▊| 6382/6470 [1:13:21<01:04,  1.37it/s]
4|6383|Loss: 0.756372332572937:  99%|█████████▊| 6382/6470 [1:13:21<01:04,  1.37it/s] 
4|6383|Loss: 0.756372332572937:  99%|█████████▊| 6383/6470 [1:13:22<01:00,  1.45it/s]
4|6384|Loss: 0.9479536414146423:  99%|█████████▊| 6383/6470 [1:13:22<01:00,  1.45it/s]
4|6384|Loss: 0.9479536414146423:  99%|█████████▊| 6384/6470 [1:13:22<00:56,  1.52it/s]
4|6385|Loss: 0.855840265750885:  99%|█████████▊| 6384/6470 [1:13:23<00:56,  1.52it/s] 
4|6385|Loss: 0.855840265750885:  99%|█████████▊| 6385/6470 [1:13:23<00:54,  1.55it/s]
4|6386|Loss: 1.2726945877075195:  99%|█████████▊| 6385/6470 [1:13:23<00:54,  1.55it/s]
4|6386|Loss: 1.2726945877075195:  99%|█████████▊| 6386/6470 [1:13:23<00:55,  1.52it/s]
4|6387|Loss: 0.8701688647270203:  99%|█████████▊| 6386/6470 [1:13:24<00:55,  1.52it/s]
4|6387|Loss: 0.8701688647270203:  99%|█████████▊| 6387/6470 [1:13:24<00:53,  1.54it/s]
4|6388|Loss: 0.852098286151886:  99%|█████████▊| 6387/6470 [1:13:24<00:53,  1.54it/s] 
4|6388|Loss: 0.852098286151886:  99%|█████████▊| 6388/6470 [1:13:25<00:54,  1.51it/s]
4|6389|Loss: 0.8880857825279236:  99%|█████████▊| 6388/6470 [1:13:25<00:54,  1.51it/s]
4|6389|Loss: 0.8880857825279236:  99%|█████████▊| 6389/6470 [1:13:25<00:53,  1.53it/s]
4|6390|Loss: 0.8646162152290344:  99%|█████████▊| 6389/6470 [1:13:26<00:53,  1.53it/s]
4|6390|Loss: 0.8646162152290344:  99%|█████████▉| 6390/6470 [1:13:26<00:49,  1.62it/s]
4|6391|Loss: 0.9941978454589844:  99%|█████████▉| 6390/6470 [1:13:26<00:49,  1.62it/s]
4|6391|Loss: 0.9941978454589844:  99%|█████████▉| 6391/6470 [1:13:26<00:46,  1.68it/s]
4|6392|Loss: 1.0466808080673218:  99%|█████████▉| 6391/6470 [1:13:27<00:46,  1.68it/s]
4|6392|Loss: 1.0466808080673218:  99%|█████████▉| 6392/6470 [1:13:27<00:50,  1.54it/s]
4|6393|Loss: 0.8370721936225891:  99%|█████████▉| 6392/6470 [1:13:28<00:50,  1.54it/s]
4|6393|Loss: 0.8370721936225891:  99%|█████████▉| 6393/6470 [1:13:28<00:51,  1.48it/s]
4|6394|Loss: 0.8610981702804565:  99%|█████████▉| 6393/6470 [1:13:28<00:51,  1.48it/s]
4|6394|Loss: 0.8610981702804565:  99%|█████████▉| 6394/6470 [1:13:29<00:49,  1.52it/s]
4|6395|Loss: 0.9213030934333801:  99%|█████████▉| 6394/6470 [1:13:29<00:49,  1.52it/s]
4|6395|Loss: 0.9213030934333801:  99%|█████████▉| 6395/6470 [1:13:29<00:50,  1.49it/s]
4|6396|Loss: 0.8499007225036621:  99%|█████████▉| 6395/6470 [1:13:30<00:50,  1.49it/s]
4|6396|Loss: 0.8499007225036621:  99%|█████████▉| 6396/6470 [1:13:30<00:51,  1.45it/s]
4|6397|Loss: 0.8116262555122375:  99%|█████████▉| 6396/6470 [1:13:30<00:51,  1.45it/s]
4|6397|Loss: 0.8116262555122375:  99%|█████████▉| 6397/6470 [1:13:31<00:48,  1.51it/s]
4|6398|Loss: 0.8843669295310974:  99%|█████████▉| 6397/6470 [1:13:31<00:48,  1.51it/s]
4|6398|Loss: 0.8843669295310974:  99%|█████████▉| 6398/6470 [1:13:31<00:48,  1.47it/s]
4|6399|Loss: 0.751837968826294:  99%|█████████▉| 6398/6470 [1:13:32<00:48,  1.47it/s] 
4|6399|Loss: 0.751837968826294:  99%|█████████▉| 6399/6470 [1:13:32<00:49,  1.43it/s]
4|6400|Loss: 0.7085820436477661:  99%|█████████▉| 6399/6470 [1:13:33<00:49,  1.43it/s]
4|6400|Loss: 0.7085820436477661:  99%|█████████▉| 6400/6470 [1:13:33<00:50,  1.39it/s]
4|6401|Loss: 0.7604133486747742:  99%|█████████▉| 6400/6470 [1:13:33<00:50,  1.39it/s]
4|6401|Loss: 0.7604133486747742:  99%|█████████▉| 6401/6470 [1:13:34<00:51,  1.35it/s]
4|6402|Loss: 0.8349245190620422:  99%|█████████▉| 6401/6470 [1:13:34<00:51,  1.35it/s]
4|6402|Loss: 0.8349245190620422:  99%|█████████▉| 6402/6470 [1:13:34<00:47,  1.43it/s]
4|6403|Loss: 0.7507418394088745:  99%|█████████▉| 6402/6470 [1:13:35<00:47,  1.43it/s]
4|6403|Loss: 0.7507418394088745:  99%|█████████▉| 6403/6470 [1:13:35<00:46,  1.45it/s]
4|6404|Loss: 0.9082905650138855:  99%|█████████▉| 6403/6470 [1:13:35<00:46,  1.45it/s]
4|6404|Loss: 0.9082905650138855:  99%|█████████▉| 6404/6470 [1:13:36<00:45,  1.44it/s]
4|6405|Loss: 0.8595359921455383:  99%|█████████▉| 6404/6470 [1:13:36<00:45,  1.44it/s]
4|6405|Loss: 0.8595359921455383:  99%|█████████▉| 6405/6470 [1:13:36<00:45,  1.42it/s]
4|6406|Loss: 0.757109522819519:  99%|█████████▉| 6405/6470 [1:13:37<00:45,  1.42it/s] 
4|6406|Loss: 0.757109522819519:  99%|█████████▉| 6406/6470 [1:13:37<00:45,  1.41it/s]
4|6407|Loss: 0.788987398147583:  99%|█████████▉| 6406/6470 [1:13:38<00:45,  1.41it/s]
4|6407|Loss: 0.788987398147583:  99%|█████████▉| 6407/6470 [1:13:38<00:45,  1.40it/s]
4|6408|Loss: 0.9026356935501099:  99%|█████████▉| 6407/6470 [1:13:38<00:45,  1.40it/s]
4|6408|Loss: 0.9026356935501099:  99%|█████████▉| 6408/6470 [1:13:39<00:44,  1.39it/s]
4|6409|Loss: 0.816234827041626:  99%|█████████▉| 6408/6470 [1:13:39<00:44,  1.39it/s] 
4|6409|Loss: 0.816234827041626:  99%|█████████▉| 6409/6470 [1:13:39<00:45,  1.35it/s]
4|6410|Loss: 0.7758328318595886:  99%|█████████▉| 6409/6470 [1:13:40<00:45,  1.35it/s]
4|6410|Loss: 0.7758328318595886:  99%|█████████▉| 6410/6470 [1:13:40<00:42,  1.40it/s]
4|6411|Loss: 0.5520209670066833:  99%|█████████▉| 6410/6470 [1:13:40<00:42,  1.40it/s]
4|6411|Loss: 0.5520209670066833:  99%|█████████▉| 6411/6470 [1:13:41<00:44,  1.32it/s]
4|6412|Loss: 0.8304684162139893:  99%|█████████▉| 6411/6470 [1:13:41<00:44,  1.32it/s]
4|6412|Loss: 0.8304684162139893:  99%|█████████▉| 6412/6470 [1:13:42<00:41,  1.38it/s]
4|6413|Loss: 0.886077880859375:  99%|█████████▉| 6412/6470 [1:13:42<00:41,  1.38it/s] 
4|6413|Loss: 0.886077880859375:  99%|█████████▉| 6413/6470 [1:13:42<00:40,  1.39it/s]
4|6414|Loss: 0.9994151592254639:  99%|█████████▉| 6413/6470 [1:13:43<00:40,  1.39it/s]
4|6414|Loss: 0.9994151592254639:  99%|█████████▉| 6414/6470 [1:13:43<00:38,  1.44it/s]
4|6415|Loss: 0.8234975337982178:  99%|█████████▉| 6414/6470 [1:13:43<00:38,  1.44it/s]
4|6415|Loss: 0.8234975337982178:  99%|█████████▉| 6415/6470 [1:13:44<00:38,  1.44it/s]
4|6416|Loss: 0.80176842212677:  99%|█████████▉| 6415/6470 [1:13:44<00:38,  1.44it/s]  
4|6416|Loss: 0.80176842212677:  99%|█████████▉| 6416/6470 [1:13:44<00:37,  1.42it/s]
4|6417|Loss: 0.8671579360961914:  99%|█████████▉| 6416/6470 [1:13:45<00:37,  1.42it/s]
4|6417|Loss: 0.8671579360961914:  99%|█████████▉| 6417/6470 [1:13:45<00:36,  1.44it/s]
4|6418|Loss: 0.8335331082344055:  99%|█████████▉| 6417/6470 [1:13:45<00:36,  1.44it/s]
4|6418|Loss: 0.8335331082344055:  99%|█████████▉| 6418/6470 [1:13:46<00:34,  1.51it/s]
4|6419|Loss: 0.8490811586380005:  99%|█████████▉| 6418/6470 [1:13:46<00:34,  1.51it/s]
4|6419|Loss: 0.8490811586380005:  99%|█████████▉| 6419/6470 [1:13:46<00:35,  1.42it/s]
4|6420|Loss: 0.6868557333946228:  99%|█████████▉| 6419/6470 [1:13:47<00:35,  1.42it/s]
4|6420|Loss: 0.6868557333946228:  99%|█████████▉| 6420/6470 [1:13:47<00:34,  1.44it/s]
4|6421|Loss: 0.7373573184013367:  99%|█████████▉| 6420/6470 [1:13:47<00:34,  1.44it/s]
4|6421|Loss: 0.7373573184013367:  99%|█████████▉| 6421/6470 [1:13:48<00:33,  1.48it/s]
4|6422|Loss: 0.9699812531471252:  99%|█████████▉| 6421/6470 [1:13:48<00:33,  1.48it/s]
4|6422|Loss: 0.9699812531471252:  99%|█████████▉| 6422/6470 [1:13:48<00:31,  1.53it/s]
4|6423|Loss: 0.7578108906745911:  99%|█████████▉| 6422/6470 [1:13:49<00:31,  1.53it/s]
4|6423|Loss: 0.7578108906745911:  99%|█████████▉| 6423/6470 [1:13:49<00:30,  1.54it/s]
4|6424|Loss: 0.8526813983917236:  99%|█████████▉| 6423/6470 [1:13:49<00:30,  1.54it/s]
4|6424|Loss: 0.8526813983917236:  99%|█████████▉| 6424/6470 [1:13:50<00:30,  1.51it/s]
4|6425|Loss: 0.874487042427063:  99%|█████████▉| 6424/6470 [1:13:50<00:30,  1.51it/s] 
4|6425|Loss: 0.874487042427063:  99%|█████████▉| 6425/6470 [1:13:50<00:29,  1.52it/s]
4|6426|Loss: 0.8263739347457886:  99%|█████████▉| 6425/6470 [1:13:51<00:29,  1.52it/s]
4|6426|Loss: 0.8263739347457886:  99%|█████████▉| 6426/6470 [1:13:51<00:28,  1.54it/s]
4|6427|Loss: 0.8882048726081848:  99%|█████████▉| 6426/6470 [1:13:51<00:28,  1.54it/s]
4|6427|Loss: 0.8882048726081848:  99%|█████████▉| 6427/6470 [1:13:51<00:27,  1.55it/s]
4|6428|Loss: 0.9226167798042297:  99%|█████████▉| 6427/6470 [1:13:52<00:27,  1.55it/s]
4|6428|Loss: 0.9226167798042297:  99%|█████████▉| 6428/6470 [1:13:52<00:25,  1.66it/s]
4|6429|Loss: 0.8666799068450928:  99%|█████████▉| 6428/6470 [1:13:52<00:25,  1.66it/s]
4|6429|Loss: 0.8666799068450928:  99%|█████████▉| 6429/6470 [1:13:53<00:25,  1.62it/s]
4|6430|Loss: 0.8798783421516418:  99%|█████████▉| 6429/6470 [1:13:53<00:25,  1.62it/s]
4|6430|Loss: 0.8798783421516418:  99%|█████████▉| 6430/6470 [1:13:53<00:26,  1.53it/s]
4|6431|Loss: 0.9133914709091187:  99%|█████████▉| 6430/6470 [1:13:54<00:26,  1.53it/s]
4|6431|Loss: 0.9133914709091187:  99%|█████████▉| 6431/6470 [1:13:54<00:26,  1.46it/s]
4|6432|Loss: 0.7864576578140259:  99%|█████████▉| 6431/6470 [1:13:55<00:26,  1.46it/s]
4|6432|Loss: 0.7864576578140259:  99%|█████████▉| 6432/6470 [1:13:55<00:26,  1.43it/s]
4|6433|Loss: 0.8583894968032837:  99%|█████████▉| 6432/6470 [1:13:55<00:26,  1.43it/s]
4|6433|Loss: 0.8583894968032837:  99%|█████████▉| 6433/6470 [1:13:56<00:25,  1.43it/s]
4|6434|Loss: 0.8504859209060669:  99%|█████████▉| 6433/6470 [1:13:56<00:25,  1.43it/s]
4|6434|Loss: 0.8504859209060669:  99%|█████████▉| 6434/6470 [1:13:56<00:22,  1.60it/s]
4|6435|Loss: 0.9984176754951477:  99%|█████████▉| 6434/6470 [1:13:56<00:22,  1.60it/s]
4|6435|Loss: 0.9984176754951477:  99%|█████████▉| 6435/6470 [1:13:57<00:22,  1.58it/s]
4|6436|Loss: 0.9540390372276306:  99%|█████████▉| 6435/6470 [1:13:57<00:22,  1.58it/s]
4|6436|Loss: 0.9540390372276306:  99%|█████████▉| 6436/6470 [1:13:57<00:20,  1.68it/s]
4|6437|Loss: 0.9225359559059143:  99%|█████████▉| 6436/6470 [1:13:58<00:20,  1.68it/s]
4|6437|Loss: 0.9225359559059143:  99%|█████████▉| 6437/6470 [1:13:58<00:20,  1.58it/s]
4|6438|Loss: 0.8868812918663025:  99%|█████████▉| 6437/6470 [1:13:58<00:20,  1.58it/s]
4|6438|Loss: 0.8868812918663025: 100%|█████████▉| 6438/6470 [1:13:59<00:21,  1.47it/s]
4|6439|Loss: 0.8185425996780396: 100%|█████████▉| 6438/6470 [1:13:59<00:21,  1.47it/s]
4|6439|Loss: 0.8185425996780396: 100%|█████████▉| 6439/6470 [1:13:59<00:21,  1.42it/s]
4|6440|Loss: 0.782100260257721: 100%|█████████▉| 6439/6470 [1:14:00<00:21,  1.42it/s] 
4|6440|Loss: 0.782100260257721: 100%|█████████▉| 6440/6470 [1:14:00<00:21,  1.37it/s]
4|6441|Loss: 0.8274407982826233: 100%|█████████▉| 6440/6470 [1:14:01<00:21,  1.37it/s]
4|6441|Loss: 0.8274407982826233: 100%|█████████▉| 6441/6470 [1:14:01<00:21,  1.37it/s]
4|6442|Loss: 0.8336226344108582: 100%|█████████▉| 6441/6470 [1:14:01<00:21,  1.37it/s]
4|6442|Loss: 0.8336226344108582: 100%|█████████▉| 6442/6470 [1:14:02<00:20,  1.37it/s]
4|6443|Loss: 1.0698919296264648: 100%|█████████▉| 6442/6470 [1:14:02<00:20,  1.37it/s]
4|6443|Loss: 1.0698919296264648: 100%|█████████▉| 6443/6470 [1:14:02<00:19,  1.36it/s]
4|6444|Loss: 0.8630589246749878: 100%|█████████▉| 6443/6470 [1:14:03<00:19,  1.36it/s]
4|6444|Loss: 0.8630589246749878: 100%|█████████▉| 6444/6470 [1:14:03<00:19,  1.36it/s]
4|6445|Loss: 0.7591606974601746: 100%|█████████▉| 6444/6470 [1:14:04<00:19,  1.36it/s]
4|6445|Loss: 0.7591606974601746: 100%|█████████▉| 6445/6470 [1:14:04<00:17,  1.45it/s]
4|6446|Loss: 0.9220765829086304: 100%|█████████▉| 6445/6470 [1:14:04<00:17,  1.45it/s]
4|6446|Loss: 0.9220765829086304: 100%|█████████▉| 6446/6470 [1:14:04<00:14,  1.61it/s]
4|6447|Loss: 0.8064724802970886: 100%|█████████▉| 6446/6470 [1:14:05<00:14,  1.61it/s]
4|6447|Loss: 0.8064724802970886: 100%|█████████▉| 6447/6470 [1:14:05<00:13,  1.71it/s]
4|6448|Loss: 0.8223397731781006: 100%|█████████▉| 6447/6470 [1:14:05<00:13,  1.71it/s]
4|6448|Loss: 0.8223397731781006: 100%|█████████▉| 6448/6470 [1:14:05<00:12,  1.70it/s]
4|6449|Loss: 0.8036291003227234: 100%|█████████▉| 6448/6470 [1:14:06<00:12,  1.70it/s]
4|6449|Loss: 0.8036291003227234: 100%|█████████▉| 6449/6470 [1:14:06<00:12,  1.66it/s]
4|6450|Loss: 0.8363302946090698: 100%|█████████▉| 6449/6470 [1:14:06<00:12,  1.66it/s]
4|6450|Loss: 0.8363302946090698: 100%|█████████▉| 6450/6470 [1:14:07<00:12,  1.63it/s]
4|6451|Loss: 0.9817918539047241: 100%|█████████▉| 6450/6470 [1:14:07<00:12,  1.63it/s]
4|6451|Loss: 0.9817918539047241: 100%|█████████▉| 6451/6470 [1:14:07<00:11,  1.62it/s]
4|6452|Loss: 0.9931104779243469: 100%|█████████▉| 6451/6470 [1:14:08<00:11,  1.62it/s]
4|6452|Loss: 0.9931104779243469: 100%|█████████▉| 6452/6470 [1:14:08<00:10,  1.68it/s]
4|6453|Loss: 0.916088342666626: 100%|█████████▉| 6452/6470 [1:14:08<00:10,  1.68it/s] 
4|6453|Loss: 0.916088342666626: 100%|█████████▉| 6453/6470 [1:14:08<00:10,  1.68it/s]
4|6454|Loss: 1.0340352058410645: 100%|█████████▉| 6453/6470 [1:14:09<00:10,  1.68it/s]
4|6454|Loss: 1.0340352058410645: 100%|█████████▉| 6454/6470 [1:14:09<00:09,  1.73it/s]
4|6455|Loss: 0.8354172706604004: 100%|█████████▉| 6454/6470 [1:14:09<00:09,  1.73it/s]
4|6455|Loss: 0.8354172706604004: 100%|█████████▉| 6455/6470 [1:14:09<00:07,  1.89it/s]
4|6456|Loss: 0.9108781814575195: 100%|█████████▉| 6455/6470 [1:14:10<00:07,  1.89it/s]
4|6456|Loss: 0.9108781814575195: 100%|█████████▉| 6456/6470 [1:14:10<00:07,  1.77it/s]
4|6457|Loss: 0.8686513900756836: 100%|█████████▉| 6456/6470 [1:14:10<00:07,  1.77it/s]
4|6457|Loss: 0.8686513900756836: 100%|█████████▉| 6457/6470 [1:14:11<00:08,  1.59it/s]
4|6458|Loss: 0.767105221748352: 100%|█████████▉| 6457/6470 [1:14:11<00:08,  1.59it/s] 
4|6458|Loss: 0.767105221748352: 100%|█████████▉| 6458/6470 [1:14:11<00:07,  1.55it/s]
4|6459|Loss: 0.7560315132141113: 100%|█████████▉| 6458/6470 [1:14:12<00:07,  1.55it/s]
4|6459|Loss: 0.7560315132141113: 100%|█████████▉| 6459/6470 [1:14:12<00:07,  1.44it/s]
4|6460|Loss: 0.7381058931350708: 100%|█████████▉| 6459/6470 [1:14:13<00:07,  1.44it/s]
4|6460|Loss: 0.7381058931350708: 100%|█████████▉| 6460/6470 [1:14:13<00:06,  1.44it/s]
4|6461|Loss: 0.9017964005470276: 100%|█████████▉| 6460/6470 [1:14:13<00:06,  1.44it/s]
4|6461|Loss: 0.9017964005470276: 100%|█████████▉| 6461/6470 [1:14:14<00:06,  1.48it/s]
4|6462|Loss: 0.9245341420173645: 100%|█████████▉| 6461/6470 [1:14:14<00:06,  1.48it/s]
4|6462|Loss: 0.9245341420173645: 100%|█████████▉| 6462/6470 [1:14:14<00:05,  1.45it/s]
4|6463|Loss: 0.7983295321464539: 100%|█████████▉| 6462/6470 [1:14:15<00:05,  1.45it/s]
4|6463|Loss: 0.7983295321464539: 100%|█████████▉| 6463/6470 [1:14:15<00:04,  1.48it/s]
4|6464|Loss: 0.9489600658416748: 100%|█████████▉| 6463/6470 [1:14:15<00:04,  1.48it/s]
4|6464|Loss: 0.9489600658416748: 100%|█████████▉| 6464/6470 [1:14:16<00:03,  1.51it/s]
4|6465|Loss: 0.8887046575546265: 100%|█████████▉| 6464/6470 [1:14:16<00:03,  1.51it/s]
4|6465|Loss: 0.8887046575546265: 100%|█████████▉| 6465/6470 [1:14:16<00:03,  1.42it/s]
4|6466|Loss: 0.785558819770813: 100%|█████████▉| 6465/6470 [1:14:17<00:03,  1.42it/s] 
4|6466|Loss: 0.785558819770813: 100%|█████████▉| 6466/6470 [1:14:17<00:03,  1.33it/s]
4|6467|Loss: 0.8075568079948425: 100%|█████████▉| 6466/6470 [1:14:18<00:03,  1.33it/s]
4|6467|Loss: 0.8075568079948425: 100%|█████████▉| 6467/6470 [1:14:18<00:02,  1.39it/s]
4|6468|Loss: 0.8465657234191895: 100%|█████████▉| 6467/6470 [1:14:18<00:02,  1.39it/s]
4|6468|Loss: 0.8465657234191895: 100%|█████████▉| 6468/6470 [1:14:18<00:01,  1.48it/s]
4|6469|Loss: 0.7640355825424194: 100%|█████████▉| 6468/6470 [1:14:19<00:01,  1.48it/s]
4|6469|Loss: 0.7640355825424194: 100%|█████████▉| 6469/6470 [1:14:19<00:00,  1.46it/s]
4|6470|Loss: 1.0264137983322144: 100%|█████████▉| 6469/6470 [1:14:20<00:00,  1.46it/s]
4|6470|Loss: 1.0264137983322144: 100%|██████████| 6470/6470 [1:14:20<00:00,  1.39it/s]
4|6470|Loss: 1.0264137983322144: 100%|██████████| 6470/6470 [1:14:20<00:00,  1.45it/s]
Model checkpoint of size 25705 MB saved to output/alpaca-llama2-baseline/model_3_25880.ckpt
wandb: - 0.004 MB of 0.004 MB uploaded
wandb: \ 0.004 MB of 0.004 MB uploaded
wandb: | 0.010 MB of 0.026 MB uploaded (0.001 MB deduped)
wandb: / 0.026 MB of 0.026 MB uploaded (0.001 MB deduped)
wandb: - 0.026 MB of 0.026 MB uploaded (0.001 MB deduped)
wandb: \ 0.026 MB of 0.026 MB uploaded (0.001 MB deduped)
wandb:                                                                                
wandb: W&B sync reduced upload amount by 4.6%             
wandb: 
wandb: Run history:
wandb: gpu_resources ▄▇▅▇▄▃▃▅▃▅▆▅▅▄▅▄▆▆▄▇▆▃▆▃▂▇▅▇▄▇▂▂▄█▅▄▅▁▄▁
wandb:          loss █▄▄▃▃▂▃▃▄▃▃▂▃▂▂▃▃▃▂▂▂▄▂▂▃▂▃▁▃▂▂▄▂▃▃▁▃▁▃▃
wandb:            lr ▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
wandb: 
wandb: Run summary:
wandb: gpu_resources 44202082304
wandb:          loss 1.02641
wandb:            lr 2e-05
wandb: 
wandb: 🚀 View run auspicious-lamp-5 at: https://wandb.ai/laurence_r/colorful-llama/runs/iu28me1d
wandb: ️⚡ View job at https://wandb.ai/laurence_r/colorful-llama/jobs/QXJ0aWZhY3RDb2xsZWN0aW9uOjE0MDgxNDM2OQ==/version_details/v4
wandb: Synced 5 W&B file(s), 0 media file(s), 2 artifact file(s) and 0 other file(s)
wandb: Find logs at: ./wandb/run-20240218_170205-iu28me1d/logs