Training in progress, epoch 10

logs/events.out.tfevents.1716583437.sphinx2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f57812a2fea334160330de019b81b57062751812cc44eaab2def3dddcb2e7ac
-size 71524
+oid sha256:1cd93c92e2c16a7746479214ec76207f9f52b5086780b4d21bfcd1b4f81d0399
+size 78276

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f1eb6ab312368e14bef28215379c3d8f3eecab97f677ee2d7d8c6317cb005470
+oid sha256:b9cd819eece9b23094227d17f28bc3f6ef50891a1b71638a9875dc4cf000bd69
 size 281715176

train_job_output.txt

@@ -536,4 +536,36 @@ command outputs:
 
 71%|███████   | 7850/11074 [1:06:57<26:33,  2.02it/s]
 71%|███████   | 7851/11074 [1:06:58<26:34,  2.02it/s]
 71%|███████   | 7852/11074 [1:06:58<26:34,  2.02it/s]
 71%|███████   | 7853/11074 [1:06:59<26:32,  2.02it/s]
 71%|███████   | 7854/11074 [1:06:59<26:32,  2.02it/s]
 71%|███████   | 7855/11074 [1:07:00<26:30,  2.02it/s]
 71%|███████   | 7856/11074 [1:07:00<26:30,  2.02it/s]
 71%|███████   | 7857/11074 [1:07:01<26:28,  2.02it/s]
 71%|███████   | 7858/11074 [1:07:01<26:29,  2.02it/s]
 71%|███████   | 7859/11074 [1:07:02<26:27,  2.02it/s]
 71%|███████   | 7860/11074 [1:07:02<26:27,  2.02it/s]
 71%|███████   | 7861/11074 [1:07:03<26:27,  2.02it/s]
 71%|███████   | 7862/11074 [1:07:03<26:26,  2.02it/s]
 71%|███████   | 7863/11074 [1:07:04<26:26,  2.02it/s]
 71%|███████   | 7864/11074 [1:07:04<26:24,  2.03it/s]
 71%|███████   | 7865/11074 [1:07:05<26:23,  2.03it/s]
 71%|███████   | 7866/11074 [1:07:05<26:23,  2.03it/s]
 71%|███████   | 7867/11074 [1:07:06<26:23,  2.02it/s]
 71%|███████   | 7868/11074 [1:07:06<26:22,  2.03it/s]
 71%|███████   | 7869/11074 [1:07:07<26:21,  2.03it/s]
 71%|███████   | 7870/11074 [1:07:07<26:21,  2.03it/s]
 71%|███████   | 7871/11074 [1:07:07<26:20,  2.03it/s]
 71%|███████   | 7872/11074 [1:07:08<26:20,  2.03it/s]
 71%|███████   | 7873/11074 [1:07:08<26:20,  2.03it/s]
 71%|███████   | 7874/11074 [1:07:09<26:19,  2.03it/s]
 71%|███████   | 7875/11074 [1:07:09<26:19,  2.03it/s]
                                                      {'loss': 3.3077, 'grad_norm': 0.23198536038398743, 'learning_rate': 0.0002334027221578824, 'epoch': 9.95}
 
 71%|███████   | 7875/11074 [1:07:09<26:19,  2.03it/s]
 71%|███████   | 7876/11074 [1:07:10<26:22,  2.02it/s]
 71%|███████   | 7877/11074 [1:07:10<26:21,  2.02it/s]
 71%|███████   | 7878/11074 [1:07:11<26:21,  2.02it/s]
 71%|███████   | 7879/11074 [1:07:11<26:20,  2.02it/s]
 71%|███████   | 7880/11074 [1:07:12<26:18,  2.02it/s]
 71%|███████   | 7881/11074 [1:07:12<26:18,  2.02it/s]
 71%|███████   | 7882/11074 [1:07:13<26:18,  2.02it/s]
 71%|███████   | 7883/11074 [1:07:13<26:15,  2.02it/s]
 71%|███████   | 7884/11074 [1:07:14<26:15,  2.02it/s]
 71%|███████   | 7885/11074 [1:07:14<26:13,  2.03it/s]
 71%|███████   | 7886/11074 [1:07:15<26:13,  2.03it/s]
 71%|███████   | 7887/11074 [1:07:15<26:13,  2.03it/s]
 71%|███████   | 7888/11074 [1:07:16<26:11,  2.03it/s]
 71%|███████   | 7889/11074 [1:07:16<26:12,  2.02it/s]
 71%|███████   | 7890/11074 [1:07:17<26:10,  2.03it/s]
 71%|███████▏  | 7891/11074 [1:07:17<26:11,  2.02it/s]
 71%|███████▏  | 7892/11074 [1:07:18<26:08,  2.03it/s]
 71%|███████▏  | 7893/11074 [1:07:18<26:10,  2.03it/s]
 71%|███████▏  | 7894/11074 [1:07:19<26:10,  2.03it/s]
 71%|███████▏  | 7895/11074 [1:07:19<26:09,  2.03it/s]
 71%|███████▏  | 7896/11074 [1:07:20<26:09,  2.03it/s]
 71%|███████▏  | 7897/11074 [1:07:20<26:09,  2.02it/s]
 71%|███████▏  | 7898/11074 [1:07:21<26:08,  2.03it/s]
 71%|███████▏  | 7899/11074 [1:07:21<26:06,  2.03it/s]
 71%|███████▏  | 7900/11074 [1:07:22<26:06,  2.03it/s]{'loss': 3.3093, 'grad_norm': 0.23056922852993011, 'learning_rate': 0.00023007749594435663, 'epoch': 9.98}
 
                                                      
 71%|███████▏  | 7900/11074 [1:07:22<26:06,  2.03it/s]
 71%|███████▏  | 7901/11074 [1:07:22<26:09,  2.02it/s]
 71%|███████▏  | 7902/11074 [1:07:23<26:09,  2.02it/s]
 71%|███████▏  | 7903/11074 [1:07:23<26:07,  2.02it/s]
 71%|███████▏  | 7904/11074 [1:07:24<26:06,  2.02it/s]
 71%|███████▏  | 7905/11074 [1:07:24<26:05,  2.02it/s]
 71%|███████▏  | 7906/11074 [1:07:25<26:03,  2.03it/s]
 71%|███████▏  | 7907/11074 [1:07:25<26:03,  2.03it/s]
 71%|███████▏  | 7908/11074 [1:07:26<26:01,  2.03it/s]
 71%|███████▏  | 7909/11074 [1:07:26<26:01,  2.03it/s]
 71%|███████▏  | 7910/11074 [1:07:27<26:01,  2.03it/s]
 71%|███████▏  | 7911/11074 [1:07:27<25:59,  2.03it/s]
 71%|███████▏  | 7912/11074 [1:07:28<25:59,  2.03it/s]
 71%|███████▏  | 7913/11074 [1:07:28<25:58,  2.03it/s]
 71%|███████▏  | 7914/11074 [1:07:29<26:00,  2.03it/s]
 71%|███████▏  | 7915/11074 [1:07:29<25:45,  2.04it/s]
 71%|█���█████▏  | 7916/11074 [1:07:41<3:26:15,  3.92s/it]
 71%|███████▏  | 7917/11074 [1:07:42<2:32:17,  2.89s/it]
 72%|███████▏  | 7918/11074 [1:07:42<1:54:19,  2.17s/it]
 72%|███████▏  | 7919/11074 [1:07:43<1:27:47,  1.67s/it]
 72%|███████▏  | 7920/11074 [1:07:43<1:09:13,  1.32s/it]
 72%|███████▏  | 7921/11074 [1:07:44<56:14,  1.07s/it]  
 72%|███████▏  | 7922/11074 [1:07:44<47:07,  1.11it/s]
 72%|███████▏  | 7923/11074 [1:07:45<40:57,  1.28it/s]
 72%|███████▏  | 7924/11074 [1:07:45<36:26,  1.44it/s]
 72%|███████▏  | 7925/11074 [1:07:46<33:16,  1.58it/s]
                                                      
-
 72%|███████▏  | 7925/11074 [1:07:46<33:16,  1.58it/s]
 72%|███████▏  | 7926/11074 [1:07:46<31:04,  1.69it/s]
 72%|███████▏  | 7927/11074 [1:07:47<29:32,  1.78it/s]
 72%|███████▏  | 7928/11074 [1:07:47<28:29,  1.84it/s]
 72%|███████▏  | 7929/11074 [1:07:48<27:42,  1.89it/s]
 72%|███████▏  | 7930/11074 [1:07:48<27:10,  1.93it/s]
 72%|███████▏  | 7931/11074 [1:07:49<26:47,  1.96it/s]
 72%|███████▏  | 7932/11074 [1:07:49<26:30,  1.98it/s]
 72%|███████▏  | 7933/11074 [1:07:50<26:19,  1.99it/s]
 72%|███████▏  | 7934/11074 [1:07:50<26:10,  2.00it/s]
 72%|███████▏  | 7935/11074 [1:07:51<26:03,  2.01it/s]
 72%|███████▏  | 7936/11074 [1:07:51<25:59,  2.01it/s]
 72%|███████▏  | 7937/11074 [1:07:52<25:55,  2.02it/s]
 72%|███████▏  | 7938/11074 [1:07:52<25:52,  2.02it/s]
 72%|███████▏  | 7939/11074 [1:07:53<25:50,  2.02it/s]
 72%|███████▏  | 7940/11074 [1:07:53<25:54,  2.02it/s]
+
 72%|███████▏  | 7925/11074 [1:07:46<33:16,  1.58it/s]
 72%|███████▏  | 7926/11074 [1:07:46<31:04,  1.69it/s]
 72%|███████▏  | 7927/11074 [1:07:47<29:32,  1.78it/s]
 72%|███████▏  | 7928/11074 [1:07:47<28:29,  1.84it/s]
 72%|███████▏  | 7929/11074 [1:07:48<27:42,  1.89it/s]
 72%|███████▏  | 7930/11074 [1:07:48<27:10,  1.93it/s]
 72%|███████▏  | 7931/11074 [1:07:49<26:47,  1.96it/s]
 72%|███████▏  | 7932/11074 [1:07:49<26:30,  1.98it/s]
 72%|███████▏  | 7933/11074 [1:07:50<26:19,  1.99it/s]
 72%|███████▏  | 7934/11074 [1:07:50<26:10,  2.00it/s]
 72%|███████▏  | 7935/11074 [1:07:51<26:03,  2.01it/s]
 72%|███████▏  | 7936/11074 [1:07:51<25:59,  2.01it/s]
 72%|███████▏  | 7937/11074 [1:07:52<25:55,  2.02it/s]
 72%|███████▏  | 7938/11074 [1:07:52<25:52,  2.02it/s]
 72%|███████▏  | 7939/11074 [1:07:53<25:50,  2.02it/s]
 72%|███████▏  | 7940/11074 [1:07:53<25:54,  2.02it/s]
 72%|███████▏  | 7941/11074 [1:07:54<25:57,  2.01it/s]
 72%|███████▏  | 7942/11074 [1:07:54<25:51,  2.02it/s]
 72%|███████▏  | 7943/11074 [1:07:55<25:51,  2.02it/s]
 72%|███████▏  | 7944/11074 [1:07:55<25:47,  2.02it/s]
 72%|███████▏  | 7945/11074 [1:07:55<25:47,  2.02it/s]
 72%|███████▏  | 7946/11074 [1:07:56<25:44,  2.03it/s]
 72%|███████▏  | 7947/11074 [1:07:56<25:43,  2.03it/s]
 72%|███████▏  | 7948/11074 [1:07:57<25:43,  2.02it/s]
 72%|███████▏  | 7949/11074 [1:07:57<25:41,  2.03it/s]
 72%|███████▏  | 7950/11074 [1:07:58<25:42,  2.03it/s]{'loss': 3.2027, 'grad_norm': 0.23849323391914368, 'learning_rate': 0.00022347754066845987, 'epoch': 10.04}
                                                      
+
 72%|███████▏  | 7950/11074 [1:07:58<25:42,  2.03it/s]
 72%|███████▏  | 7951/11074 [1:07:58<25:44,  2.02it/s]
 72%|███████▏  | 7952/11074 [1:07:59<25:42,  2.02it/s]
 72%|███████▏  | 7953/11074 [1:07:59<25:40,  2.03it/s]
 72%|███████▏  | 7954/11074 [1:08:00<25:40,  2.03it/s]
 72%|███████▏  | 7955/11074 [1:08:00<25:38,  2.03it/s]
 72%|███████▏  | 7956/11074 [1:08:01<25:38,  2.03it/s]
 72%|███████▏  | 7957/11074 [1:08:01<25:38,  2.03it/s]
 72%|███████▏  | 7958/11074 [1:08:02<25:38,  2.02it/s]
 72%|███████▏  | 7959/11074 [1:08:02<25:39,  2.02it/s]
 72%|███████▏  | 7960/11074 [1:08:03<25:36,  2.03it/s]
 72%|███████▏  | 7961/11074 [1:08:03<25:36,  2.03it/s]
 72%|███████▏  | 7962/11074 [1:08:04<25:35,  2.03it/s]
 72%|███████▏  | 7963/11074 [1:08:04<25:35,  2.03it/s]
 72%|███████▏  | 7964/11074 [1:08:05<25:34,  2.03it/s]
 72%|███████▏  | 7965/11074 [1:08:05<25:34,  2.03it/s]
 72%|███████▏  | 7966/11074 [1:08:06<25:35,  2.02it/s]
 72%|███████▏  | 7967/11074 [1:08:06<25:35,  2.02it/s]
 72%|███████▏  | 7968/11074 [1:08:07<25:35,  2.02it/s]
 72%|███████▏  | 7969/11074 [1:08:07<25:34,  2.02it/s]
 72%|███████▏  | 7970/11074 [1:08:08<25:33,  2.02it/s]
 72%|███████▏  | 7971/11074 [1:08:08<25:33,  2.02it/s]
 72%|███████▏  | 7972/11074 [1:08:09<25:32,  2.02it/s]
 72%|███████▏  | 7973/11074 [1:08:09<25:30,  2.03it/s]
 72%|███████▏  | 7974/11074 [1:08:10<25:30,  2.03it/s]
 72%|███████▏  | 7975/11074 [1:08:10<25:29,  2.03it/s]{'loss': 3.1925, 'grad_norm': 0.2428920567035675, 'learning_rate': 0.00022020322150497878, 'epoch': 10.08}
                                                      
+
 72%|███████▏  | 7975/11074 [1:08:10<25:29,  2.03it/s]
 72%|███████▏  | 7976/11074 [1:08:11<25:30,  2.02it/s]
 72%|███████▏  | 7977/11074 [1:08:11<25:29,  2.02it/s]
 72%|███████▏  | 7978/11074 [1:08:12<25:30,  2.02it/s]
 72%|███████▏  | 7979/11074 [1:08:12<25:29,  2.02it/s]
 72%|███████▏  | 7980/11074 [1:08:13<25:29,  2.02it/s]
 72%|███████▏  | 7981/11074 [1:08:13<25:28,  2.02it/s]
 72%|███████▏  | 7982/11074 [1:08:14<25:26,  2.03it/s]
 72%|███████▏  | 7983/11074 [1:08:14<25:27,  2.02it/s]
 72%|███████▏  | 7984/11074 [1:08:15<25:24,  2.03it/s]
 72%|███████▏  | 7985/11074 [1:08:15<25:25,  2.03it/s]
 72%|███████▏  | 7986/11074 [1:08:16<25:23,  2.03it/s]
 72%|███████▏  | 7987/11074 [1:08:16<25:24,  2.03it/s]
 72%|███████▏  | 7988/11074 [1:08:17<25:23,  2.03it/s]
 72%|███████▏  | 7989/11074 [1:08:17<25:20,  2.03it/s]
 72%|███████▏  | 7990/11074 [1:08:18<25:22,  2.03it/s]
 72%|███████▏  | 7991/11074 [1:08:18<25:21,  2.03it/s]
 72%|███████▏  | 7992/11074 [1:08:19<25:22,  2.02it/s]
 72%|███████▏  | 7993/11074 [1:08:19<25:20,  2.03it/s]
 72%|███████▏  | 7994/11074 [1:08:20<25:21,  2.02it/s]
 72%|███████▏  | 7995/11074 [1:08:20<25:20,  2.03it/s]
 72%|███████▏  | 7996/11074 [1:08:21<25:19,  2.03it/s]
 72%|███████▏  | 7997/11074 [1:08:21<25:19,  2.02it/s]
 72%|███████▏  | 7998/11074 [1:08:22<25:17,  2.03it/s]
 72%|███████▏  | 7999/11074 [1:08:22<25:18,  2.03it/s]
 72%|███████▏  | 8000/11074 [1:08:23<25:15,  2.03it/s]
                                                      {'loss': 3.2116, 'grad_norm': 0.23470519483089447, 'learning_rate': 0.00021694627948786466, 'epoch': 10.11}
+
 72%|███████▏  | 8000/11074 [1:08:23<25:15,  2.03it/s]
 72%|███████▏  | 8001/11074 [1:08:23<25:18,  2.02it/s]
 72%|███████▏  | 8002/11074 [1:08:24<25:17,  2.02it/s]
 72%|███████▏  | 8003/11074 [1:08:24<25:17,  2.02it/s]
 72%|███████▏  | 8004/11074 [1:08:25<25:16,  2.02it/s]
 72%|███████▏  | 8005/11074 [1:08:25<25:16,  2.02it/s]
 72%|███████▏  | 8006/11074 [1:08:26<25:16,  2.02it/s]
 72%|███████▏  | 8007/11074 [1:08:26<25:15,  2.02it/s]
 72%|███████▏  | 8008/11074 [1:08:27<25:14,  2.03it/s]
 72%|███████▏  | 8009/11074 [1:08:27<25:12,  2.03it/s]
 72%|███████▏  | 8010/11074 [1:08:28<25:12,  2.03it/s]
 72%|███████▏  | 8011/11074 [1:08:28<25:11,  2.03it/s]
 72%|███████▏  | 8012/11074 [1:08:29<25:11,  2.03it/s]
 72%|███████▏  | 8013/11074 [1:08:29<25:11,  2.02it/s]
 72%|███████▏  | 8014/11074 [1:08:30<25:11,  2.02it/s]
 72%|███████▏  | 8015/11074 [1:08:30<25:10,  2.02it/s]
 72%|███████▏  | 8016/11074 [1:08:31<25:10,  2.02it/s]
 72%|███████▏  | 8017/11074 [1:08:31<25:10,  2.02it/s]
 72%|███████▏  | 8018/11074 [1:08:32<25:08,  2.03it/s]
 72%|███████▏  | 8019/11074 [1:08:32<25:10,  2.02it/s]
 72%|███████▏  | 8020/11074 [1:08:33<25:07,  2.03it/s]
 72%|███████▏  | 8021/11074 [1:08:33<25:08,  2.02it/s]
 72%|███████▏  | 8022/11074 [1:08:34<25:06,  2.03it/s]
 72%|███████▏  | 8023/11074 [1:08:34<25:07,  2.02it/s]
 72%|███████▏  | 8024/11074 [1:08:34<25:07,  2.02it/s]
 72%|███████▏  | 8025/11074 [1:08:35<25:07,  2.02it/s]
                                                      {'loss': 3.2031, 'grad_norm': 0.24160942435264587, 'learning_rate': 0.00021370691689377887, 'epoch': 10.14}
+
 72%|███████▏  | 8025/11074 [1:08:35<25:07,  2.02it/s]
 72%|███████▏  | 8026/11074 [1:08:35<25:08,  2.02it/s]
 72%|██���████▏  | 8027/11074 [1:08:36<25:06,  2.02it/s]
 72%|███████▏  | 8028/11074 [1:08:36<25:05,  2.02it/s]
 73%|███████▎  | 8029/11074 [1:08:37<25:04,  2.02it/s]
 73%|███████▎  | 8030/11074 [1:08:37<25:03,  2.02it/s]
 73%|███████▎  | 8031/11074 [1:08:38<25:02,  2.03it/s]
 73%|███████▎  | 8032/11074 [1:08:38<25:02,  2.02it/s]
 73%|███████▎  | 8033/11074 [1:08:39<25:03,  2.02it/s]
 73%|███████▎  | 8034/11074 [1:08:39<25:02,  2.02it/s]
 73%|███████▎  | 8035/11074 [1:08:40<25:01,  2.02it/s]
 73%|███████▎  | 8036/11074 [1:08:40<25:01,  2.02it/s]
 73%|███████▎  | 8037/11074 [1:08:41<24:59,  2.03it/s]
 73%|███████▎  | 8038/11074 [1:08:41<24:59,  2.02it/s]
 73%|███████▎  | 8039/11074 [1:08:42<24:58,  2.03it/s]
 73%|███████▎  | 8040/11074 [1:08:42<24:58,  2.02it/s]
 73%|███████▎  | 8041/11074 [1:08:43<24:58,  2.02it/s]
 73%|███████▎  | 8042/11074 [1:08:43<24:57,  2.03it/s]
 73%|███████▎  | 8043/11074 [1:08:44<25:14,  2.00it/s]
 73%|███████▎  | 8044/11074 [1:08:44<25:08,  2.01it/s]
 73%|███████▎  | 8045/11074 [1:08:45<25:03,  2.01it/s]
 73%|███████▎  | 8046/11074 [1:08:45<25:00,  2.02it/s]
 73%|███████▎  | 8047/11074 [1:08:46<24:58,  2.02it/s]
 73%|███████▎  | 8048/11074 [1:08:46<24:57,  2.02it/s]
 73%|███████▎  | 8049/11074 [1:08:47<24:57,  2.02it/s]
 73%|███████▎  | 8050/11074 [1:08:47<24:55,  2.02it/s]
                                                      
+
 73%|███████▎  | 8050/11074 [1:08:47<24:55,  2.02it/s]
 73%|███████▎  | 8051/11074 [1:08:48<24:56,  2.02it/s]
 73%|███████▎  | 8052/11074 [1:08:48<24:55,  2.02it/s]
 73%|███████▎  | 8053/11074 [1:08:49<24:55,  2.02it/s]
 73%|███████▎  | 8054/11074 [1:08:49<24:53,  2.02it/s]
 73%|███████▎  | 8055/11074 [1:08:50<24:52,  2.02it/s]
 73%|███████▎  | 8056/11074 [1:08:50<24:52,  2.02it/s]
 73%|███████▎  | 8057/11074 [1:08:51<24:49,  2.03it/s]
 73%|███████▎  | 8058/11074 [1:08:51<24:49,  2.02it/s]
 73%|███████▎  | 8059/11074 [1:08:52<24:48,  2.02it/s]
 73%|███████▎  | 8060/11074 [1:08:52<24:48,  2.02it/s]
 73%|███████▎  | 8061/11074 [1:08:53<24:47,  2.03it/s]
 73%|███████▎  | 8062/11074 [1:08:53<24:47,  2.02it/s]
 73%|███████▎  | 8063/11074 [1:08:54<24:47,  2.02it/s]
 73%|███████▎  | 8064/11074 [1:08:54<24:44,  2.03it/s]
 73%|███████▎  | 8065/11074 [1:08:55<24:45,  2.03it/s]
 73%|███████▎  | 8066/11074 [1:08:55<24:43,  2.03it/s]
 73%|███████▎  | 8067/11074 [1:08:56<24:44,  2.03it/s]
 73%|███████▎  | 8068/11074 [1:08:56<24:43,  2.03it/s]
 73%|███████▎  | 8069/11074 [1:08:57<24:43,  2.03it/s]
 73%|███████▎  | 8070/11074 [1:08:57<24:42,  2.03it/s]
 73%|███████▎  | 8071/11074 [1:08:58<24:42,  2.03it/s]
 73%|███████▎  | 8072/11074 [1:08:58<24:44,  2.02it/s]
 73%|███████▎  | 8073/11074 [1:08:59<24:44,  2.02it/s]
 73%|███████▎  | 8074/11074 [1:08:59<24:43,  2.02it/s]
 73%|███████▎  | 8075/11074 [1:09:00<24:43,  2.02it/s]
                                                      
+
 73%|███████▎  | 8075/11074 [1:09:00<24:43,  2.02it/s]
 73%|███████▎  | 8076/11074 [1:09:00<24:43,  2.02it/s]
 73%|███████▎  | 8077/11074 [1:09:01<24:42,  2.02it/s]
 73%|███████▎  | 8078/11074 [1:09:01<24:41,  2.02it/s]
 73%|███████▎  | 8079/11074 [1:09:02<24:38,  2.03it/s]
 73%|███████▎  | 8080/11074 [1:09:02<24:38,  2.02it/s]
 73%|███████▎  | 8081/11074 [1:09:03<24:35,  2.03it/s]
 73%|███████▎  | 8082/11074 [1:09:03<24:37,  2.03it/s]
 73%|███████▎  | 8083/11074 [1:09:04<24:36,  2.03it/s]
 73%|███████▎  | 8084/11074 [1:09:04<24:35,  2.03it/s]
 73%|███████▎  | 8085/11074 [1:09:05<24:34,  2.03it/s]
 73%|███████▎  | 8086/11074 [1:09:05<24:33,  2.03it/s]
 73%|███████▎  | 8087/11074 [1:09:06<24:34,  2.03it/s]
 73%|███████▎  | 8088/11074 [1:09:06<24:31,  2.03it/s]
 73%|███████▎  | 8089/11074 [1:09:07<24:32,  2.03it/s]
 73%|███████▎  | 8090/11074 [1:09:07<24:31,  2.03it/s]
 73%|███████▎  | 8091/11074 [1:09:08<24:32,  2.03it/s]
 73%|███████▎  | 8092/11074 [1:09:08<24:31,  2.03it/s]
 73%|███████▎  | 8093/11074 [1:09:09<24:31,  2.03it/s]
 73%|███████▎  | 8094/11074 [1:09:09<24:31,  2.03it/s]
 73%|███████▎  | 8095/11074 [1:09:10<24:31,  2.02it/s]
 73%|███████▎  | 8096/11074 [1:09:10<24:30,  2.02it/s]
 73%|███████▎  | 8097/11074 [1:09:11<24:31,  2.02it/s]
 73%|███████▎  | 8098/11074 [1:09:11<24:30,  2.02it/s]
 73%|███████▎  | 8099/11074 [1:09:12<24:28,  2.03it/s]
 73%|███████▎  | 8100/11074 [1:09:12<24:29,  2.02it/s]{'loss': 3.2194, 'grad_norm': 0.23723545670509338, 'learning_rate': 0.00020409631196450457, 'epoch': 10.23}
+
                                                      
 73%|███████▎  | 8100/11074 [1:09:12<24:29,  2.02it/s]
 73%|███████▎  | 8101/11074 [1:09:13<24:28,  2.02it/s]
 73%|███████▎  | 8102/11074 [1:09:13<24:28,  2.02it/s]
 73%|███████▎  | 8103/11074 [1:09:14<24:26,  2.03it/s]
 73%|███████▎  | 8104/11074 [1:09:14<24:26,  2.03it/s]
 73%|███████▎  | 8105/11074 [1:09:15<24:24,  2.03it/s]
 73%|███████▎  | 8106/11074 [1:09:15<24:25,  2.03it/s]
 73%|███████▎  | 8107/11074 [1:09:16<24:25,  2.02it/s]
 73%|███████▎  | 8108/11074 [1:09:16<24:24,  2.03it/s]
 73%|███████▎  | 8109/11074 [1:09:16<24:24,  2.02it/s]
 73%|███████▎  | 8110/11074 [1:09:17<24:22,  2.03it/s]
 73%|███████▎  | 8111/11074 [1:09:17<24:23,  2.03it/s]
 73%|███████▎  | 8112/11074 [1:09:18<24:21,  2.03it/s]
 73%|███████▎  | 8113/11074 [1:09:18<24:22,  2.03it/s]
 73%|███████▎  | 8114/11074 [1:09:19<24:20,  2.03it/s]
 73%|███████▎  | 8115/11074 [1:09:19<24:19,  2.03it/s]
 73%|███████▎  | 8116/11074 [1:09:20<24:20,  2.03it/s]
 73%|███████▎  | 8117/11074 [1:09:20<24:17,  2.03it/s]
 73%|███████▎  | 8118/11074 [1:09:21<24:19,  2.03it/s]
 73%|███████▎  | 8119/11074 [1:09:21<24:17,  2.03it/s]
 73%|███████▎  | 8120/11074 [1:09:22<24:17,  2.03it/s]
 73%|███████▎  | 8121/11074 [1:09:22<24:17,  2.03it/s]
 73%|███████▎  | 8122/11074 [1:09:23<24:17,  2.03it/s]
 73%|███████▎  | 8123/11074 [1:09:23<24:17,  2.02it/s]
 73%|███████▎  | 8124/11074 [1:09:24<24:15,  2.03it/s]
 73%|███████▎  | 8125/11074 [1:09:24<24:16,  2.02it/s]
                                                      
+
 73%|███████▎  | 8125/11074 [1:09:24<24:16,  2.02it/s]
 73%|███████▎  | 8126/11074 [1:09:25<24:16,  2.02it/s]
 73%|███████▎  | 8127/11074 [1:09:25<24:18,  2.02it/s]
 73%|███████▎  | 8128/11074 [1:09:26<24:15,  2.02it/s]
 73%|███████▎  | 8129/11074 [1:09:26<24:14,  2.02it/s]
 73%|███████▎  | 8130/11074 [1:09:27<24:13,  2.03it/s]
 73%|███████▎  | 8131/11074 [1:09:27<24:14,  2.02it/s]
 73%|███████▎  | 8132/11074 [1:09:28<24:13,  2.02it/s]
 73%|███████▎  | 8133/11074 [1:09:28<24:14,  2.02it/s]
 73%|███████▎  | 8134/11074 [1:09:29<24:13,  2.02it/s]
 73%|███████▎  | 8135/11074 [1:09:29<24:12,  2.02it/s]
 73%|███████▎  | 8136/11074 [1:09:30<24:11,  2.02it/s]
 73%|███████▎  | 8137/11074 [1:09:30<24:11,  2.02it/s]
 73%|███████▎  | 8138/11074 [1:09:31<24:11,  2.02it/s]
 73%|███████▎  | 8139/11074 [1:09:31<24:10,  2.02it/s]
 74%|███████▎  | 8140/11074 [1:09:32<24:10,  2.02it/s]
 74%|███████▎  | 8141/11074 [1:09:32<24:10,  2.02it/s]
 74%|███████▎  | 8142/11074 [1:09:33<24:08,  2.02it/s]
 74%|███████▎  | 8143/11074 [1:09:33<24:07,  2.02it/s]
 74%|███████▎  | 8144/11074 [1:09:34<24:06,  2.03it/s]
 74%|███████▎  | 8145/11074 [1:09:34<24:06,  2.03it/s]
 74%|███████▎  | 8146/11074 [1:09:35<24:05,  2.03it/s]
 74%|███████▎  | 8147/11074 [1:09:35<24:05,  2.03it/s]
 74%|███████▎  | 8148/11074 [1:09:36<24:05,  2.02it/s]
 74%|███████▎  | 8149/11074 [1:09:36<24:04,  2.02it/s]
 74%|███████▎  | 8150/11074 [1:09:37<24:04,  2.02it/s]
                                                      {'loss': 3.2225, 'grad_norm': 0.23790352046489716, 'learning_rate': 0.00019778079782860702, 'epoch': 10.3}
+
 74%|███████▎  | 8150/11074 [1:09:37<24:04,  2.02it/s]
 74%|███████▎  | 8151/11074 [1:09:37<24:05,  2.02it/s]
 74%|███████▎  | 8152/11074 [1:09:38<24:05,  2.02it/s]
 74%|███████▎  | 8153/11074 [1:09:38<24:04,  2.02it/s]
 74%|███████▎  | 8154/11074 [1:09:39<24:04,  2.02it/s]
 74%|███████▎  | 8155/11074 [1:09:39<24:03,  2.02it/s]
 74%|███████▎  | 8156/11074 [1:09:40<24:01,  2.02it/s]
 74%|███████▎  | 8157/11074 [1:09:40<24:00,  2.03it/s]
 74%|███████▎  | 8158/11074 [1:09:41<23:59,  2.03it/s]
 74%|███████▎  | 8159/11074 [1:09:41<23:58,  2.03it/s]
 74%|███████▎  | 8160/11074 [1:09:42<23:58,  2.03it/s]
 74%|███████▎  | 8161/11074 [1:09:42<23:56,  2.03it/s]
 74%|███████▎  | 8162/11074 [1:09:43<23:56,  2.03it/s]
 74%|███████▎  | 8163/11074 [1:09:43<23:55,  2.03it/s]
 74%|███████▎  | 8164/11074 [1:09:44<23:55,  2.03it/s]
 74%|███████▎  | 8165/11074 [1:09:44<23:55,  2.03it/s]
 74%|███████▎  | 8166/11074 [1:09:45<23:54,  2.03it/s]
 74%|███████▎  | 8167/11074 [1:09:45<23:55,  2.03it/s]
 74%|███████▍  | 8168/11074 [1:09:46<23:53,  2.03it/s]
 74%|███████▍  | 8169/11074 [1:09:46<23:53,  2.03it/s]
 74%|███████▍  | 8170/11074 [1:09:47<23:51,  2.03it/s]
 74%|███████▍  | 8171/11074 [1:09:47<23:49,  2.03it/s]
 74%|███████▍  | 8172/11074 [1:09:48<23:51,  2.03it/s]
 74%|███████▍  | 8173/11074 [1:09:48<23:49,  2.03it/s]
 74%|███████▍  | 8174/11074 [1:09:49<23:49,  2.03it/s]
 74%|███████▍  | 8175/11074 [1:09:49<23:49,  2.03it/s]
                                                      {'loss': 3.2276, 'grad_norm': 0.2349643111228943, 'learning_rate': 0.00019465109757138034, 'epoch': 10.33}
+
 74%|███████▍  | 8175/11074 [1:09:49<23:49,  2.03it/s]
 74%|███████▍  | 8176/11074 [1:09:50<23:50,  2.03it/s]
 74%|███████▍  | 8177/11074 [1:09:50<23:51,  2.02it/s]
 74%|███████▍  | 8178/11074 [1:09:51<23:51,  2.02it/s]
 74%|███████▍  | 8179/11074 [1:09:51<23:50,  2.02it/s]
 74%|███████▍  | 8180/11074 [1:09:52<23:48,  2.03it/s]
 74%|███████▍  | 8181/11074 [1:09:52<23:48,  2.02it/s]
 74%|███████▍  | 8182/11074 [1:09:53<23:47,  2.03it/s]
 74%|███████▍  | 8183/11074 [1:09:53<23:47,  2.03it/s]
 74%|███████▍  | 8184/11074 [1:09:54<23:47,  2.02it/s]
 74%|███████▍  | 8185/11074 [1:09:54<23:47,  2.02it/s]
 74%|███████▍  | 8186/11074 [1:09:55<23:46,  2.02it/s]
 74%|███████▍  | 8187/11074 [1:09:55<23:45,  2.03it/s]
 74%|███████▍  | 8188/11074 [1:09:56<23:43,  2.03it/s]
 74%|███████▍  | 8189/11074 [1:09:56<23:45,  2.02it/s]
 74%|███████▍  | 8190/11074 [1:09:56<23:44,  2.03it/s]
 74%|███████▍  | 8191/11074 [1:09:57<23:43,  2.03it/s]
 74%|███████▍  | 8192/11074 [1:09:57<23:43,  2.02it/s]
 74%|███████▍  | 8193/11074 [1:09:58<23:44,  2.02it/s]
 74%|███████▍  | 8194/11074 [1:09:58<23:43,  2.02it/s]
 74%|███████▍  | 8195/11074 [1:09:59<23:41,  2.03it/s]
 74%|███████▍  | 8196/11074 [1:09:59<23:41,  2.02it/s]
 74%|███████▍  | 8197/11074 [1:10:00<23:39,  2.03it/s]
 74%|███████▍  | 8198/11074 [1:10:00<23:39,  2.03it/s]
 74%|███████▍  | 8199/11074 [1:10:01<23:39,  2.03it/s]
 74%|███████▍  | 8200/11074 [1:10:01<23:39,  2.02it/s]
                                                      
+
 74%|███████▍  | 8200/11074 [1:10:01<23:39,  2.02it/s]
 74%|███████▍  | 8201/11074 [1:10:02<23:39,  2.02it/s]
 74%|███████▍  | 8202/11074 [1:10:02<23:39,  2.02it/s]
 74%|███████▍  | 8203/11074 [1:10:03<23:38,  2.02it/s]
 74%|███████▍  | 8204/11074 [1:10:03<23:38,  2.02it/s]
 74%|███████▍  | 8205/11074 [1:10:04<23:37,  2.02it/s]
 74%|███████▍  | 8206/11074 [1:10:04<23:34,  2.03it/s]
 74%|███████▍  | 8207/11074 [1:10:05<23:35,  2.03it/s]
 74%|███████▍  | 8208/11074 [1:10:05<23:32,  2.03it/s]
 74%|███████▍  | 8209/11074 [1:10:06<23:33,  2.03it/s]
 74%|███████▍  | 8210/11074 [1:10:06<23:32,  2.03it/s]
 74%|███████▍  | 8211/11074 [1:10:07<23:32,  2.03it/s]
 74%|███████▍  | 8212/11074 [1:10:07<23:33,  2.03it/s]
 74%|███████▍  | 8213/11074 [1:10:08<23:32,  2.03it/s]
 74%|███████▍  | 8214/11074 [1:10:08<23:32,  2.02it/s]
 74%|███████▍  | 8215/11074 [1:10:09<23:33,  2.02it/s]
 74%|███████▍  | 8216/11074 [1:10:09<23:32,  2.02it/s]
 74%|███████▍  | 8217/11074 [1:10:10<23:31,  2.02it/s]
 74%|███████▍  | 8218/11074 [1:10:10<23:31,  2.02it/s]
 74%|███████▍  | 8219/11074 [1:10:11<23:29,  2.03it/s]
 74%|█████��█▍  | 8220/11074 [1:10:11<23:30,  2.02it/s]
 74%|███████▍  | 8221/11074 [1:10:12<23:29,  2.02it/s]
 74%|███████▍  | 8222/11074 [1:10:12<23:28,  2.02it/s]
 74%|███████▍  | 8223/11074 [1:10:13<23:27,  2.03it/s]
 74%|███████▍  | 8224/11074 [1:10:13<23:27,  2.02it/s]
 74%|███████▍  | 8225/11074 [1:10:14<23:27,  2.02it/s]{'loss': 3.2261, 'grad_norm': 0.24224504828453064, 'learning_rate': 0.00018844878253702113, 'epoch': 10.39}
                                                      
+
 74%|███████▍  | 8225/11074 [1:10:14<23:27,  2.02it/s]
 74%|███████▍  | 8226/11074 [1:10:14<23:29,  2.02it/s]
 74%|███████▍  | 8227/11074 [1:10:15<23:26,  2.02it/s]
 74%|███████▍  | 8228/11074 [1:10:15<23:27,  2.02it/s]
 74%|███████▍  | 8229/11074 [1:10:16<23:24,  2.02it/s]
 74%|███████▍  | 8230/11074 [1:10:16<23:25,  2.02it/s]
 74%|███████▍  | 8231/11074 [1:10:17<23:25,  2.02it/s]
 74%|███████▍  | 8232/11074 [1:10:17<23:24,  2.02it/s]
 74%|███████▍  | 8233/11074 [1:10:18<23:23,  2.02it/s]
 74%|███████▍  | 8234/11074 [1:10:18<23:23,  2.02it/s]
 74%|███████▍  | 8235/11074 [1:10:19<23:22,  2.02it/s]
 74%|███████▍  | 8236/11074 [1:10:19<23:21,  2.03it/s]
 74%|███████▍  | 8237/11074 [1:10:20<23:21,  2.02it/s]
 74%|███████▍  | 8238/11074 [1:10:20<23:19,  2.03it/s]
 74%|███████▍  | 8239/11074 [1:10:21<23:20,  2.02it/s]
 74%|███████▍  | 8240/11074 [1:10:21<23:18,  2.03it/s]
 74%|███████▍  | 8241/11074 [1:10:22<23:20,  2.02it/s]
 74%|███████▍  | 8242/11074 [1:10:22<23:18,  2.03it/s]
 74%|███████▍  | 8243/11074 [1:10:23<23:18,  2.02it/s]
 74%|███████▍  | 8244/11074 [1:10:23<23:16,  2.03it/s]
 74%|███████▍  | 8245/11074 [1:10:24<23:15,  2.03it/s]
 74%|███████▍  | 8246/11074 [1:10:24<25:18,  1.86it/s]
 74%|███████▍  | 8247/11074 [1:10:25<24:41,  1.91it/s]
 74%|███████▍  | 8248/11074 [1:10:25<24:15,  1.94it/s]
 74%|███████▍  | 8249/11074 [1:10:26<23:56,  1.97it/s]
 74%|███████▍  | 8250/11074 [1:10:26<23:44,  1.98it/s]{'loss': 3.2294, 'grad_norm': 0.24246717989444733, 'learning_rate': 0.00018537655296280926, 'epoch': 10.42}
+
                                                      
 74%|███████▍  | 8250/11074 [1:10:26<23:44,  1.98it/s]
 75%|███████▍  | 8251/11074 [1:10:27<23:37,  1.99it/s]
 75%|███████▍  | 8252/11074 [1:10:27<23:31,  2.00it/s]
 75%|███████▍  | 8253/11074 [1:10:28<23:23,  2.01it/s]
 75%|███████▍  | 8254/11074 [1:10:28<23:22,  2.01it/s]
 75%|███████▍  | 8255/11074 [1:10:29<23:16,  2.02it/s]
 75%|███████▍  | 8256/11074 [1:10:29<23:14,  2.02it/s]
 75%|███████▍  | 8257/11074 [1:10:30<23:12,  2.02it/s]
 75%|███████▍  | 8258/11074 [1:10:30<23:12,  2.02it/s]
 75%|███████▍  | 8259/11074 [1:10:31<23:12,  2.02it/s]
 75%|███████▍  | 8260/11074 [1:10:31<23:09,  2.02it/s]
 75%|███████▍  | 8261/11074 [1:10:32<23:09,  2.02it/s]
 75%|███████▍  | 8262/11074 [1:10:32<23:07,  2.03it/s]
 75%|███████▍  | 8263/11074 [1:10:33<23:08,  2.02it/s]
 75%|███████▍  | 8264/11074 [1:10:33<23:06,  2.03it/s]
 75%|███████▍  | 8265/11074 [1:10:34<23:06,  2.03it/s]
 75%|███████▍  | 8266/11074 [1:10:34<23:07,  2.02it/s]
 75%|███████▍  | 8267/11074 [1:10:35<23:06,  2.03it/s]
 75%|███████▍  | 8268/11074 [1:10:35<23:05,  2.03it/s]
 75%|███████▍  | 8269/11074 [1:10:36<23:04,  2.03it/s]
 75%|███████▍  | 8270/11074 [1:10:36<23:04,  2.03it/s]
 75%|███████▍  | 8271/11074 [1:10:37<23:04,  2.03it/s]
 75%|███████▍  | 8272/11074 [1:10:37<23:03,  2.02it/s]
 75%|███████▍  | 8273/11074 [1:10:38<23:02,  2.03it/s]
 75%|███████▍  | 8274/11074 [1:10:38<23:01,  2.03it/s]
 75%|███████▍  | 8275/11074 [1:10:39<23:00,  2.03it/s]{'loss': 3.2352, 'grad_norm': 0.23462679982185364, 'learning_rate': 0.00018232386349094988, 'epoch': 10.45}
+
                                                      
 75%|███████▍  | 8275/11074 [1:10:39<23:00,  2.03it/s]
 75%|███████▍  | 8276/11074 [1:10:39<23:02,  2.02it/s]
 75%|███████▍  | 8277/11074 [1:10:40<23:02,  2.02it/s]
 75%|███████▍  | 8278/11074 [1:10:40<23:01,  2.02it/s]
 75%|███████▍  | 8279/11074 [1:10:41<23:00,  2.02it/s]
 75%|███████▍  | 8280/11074 [1:10:41<22:59,  2.02it/s]
 75%|███████▍  | 8281/11074 [1:10:42<22:58,  2.03it/s]
 75%|███████▍  | 8282/11074 [1:10:42<22:58,  2.02it/s]
 75%|███████▍  | 8283/11074 [1:10:43<22:57,  2.03it/s]
 75%|███████▍  | 8284/11074 [1:10:43<22:56,  2.03it/s]
 75%|███████▍  | 8285/11074 [1:10:44<22:56,  2.03it/s]
 75%|███████▍  | 8286/11074 [1:10:44<22:56,  2.02it/s]
 75%|███████▍  | 8287/11074 [1:10:45<22:56,  2.02it/s]
 75%|███████▍  | 8288/11074 [1:10:45<22:56,  2.02it/s]
 75%|███████▍  | 8289/11074 [1:10:46<22:54,  2.03it/s]
 75%|███████▍  | 8290/11074 [1:10:46<22:54,  2.03it/s]
 75%|███████▍  | 8291/11074 [1:10:47<22:52,  2.03it/s]
 75%|███████▍  | 8292/11074 [1:10:47<22:53,  2.03it/s]
 75%|███████▍  | 8293/11074 [1:10:48<22:51,  2.03it/s]
 75%|███████▍  | 8294/11074 [1:10:48<22:54,  2.02it/s]
 75%|███████▍  | 8295/11074 [1:10:48<22:52,  2.02it/s]
 75%|███████▍  | 8296/11074 [1:10:49<22:51,  2.03it/s]
 75%|███████▍  | 8297/11074 [1:10:49<22:49,  2.03it/s]
 75%|███████▍  | 8298/11074 [1:10:50<22:47,  2.03it/s]
 75%|███████▍  | 8299/11074 [1:10:50<22:49,  2.03it/s]
 75%|███████▍  | 8300/11074 [1:10:51<22:48,  2.03it/s]
                                                      
+
 75%|███████▍  | 8300/11074 [1:10:51<22:48,  2.03it/s]
 75%|███████▍  | 8301/11074 [1:10:51<22:50,  2.02it/s]
 75%|███████▍  | 8302/11074 [1:10:52<22:49,  2.02it/s]
 75%|███████▍  | 8303/11074 [1:10:52<22:50,  2.02it/s]
 75%|███████▍  | 8304/11074 [1:10:53<22:49,  2.02it/s]
 75%|███████▍  | 8305/11074 [1:10:53<22:48,  2.02it/s]
 75%|███████▌  | 8306/11074 [1:10:54<22:47,  2.02it/s]
 75%|███████▌  | 8307/11074 [1:10:54<22:48,  2.02it/s]
 75%|███████▌  | 8308/11074 [1:10:55<22:46,  2.02it/s]
 75%|███████▌  | 8309/11074 [1:10:55<22:46,  2.02it/s]
 75%|███████▌  | 8310/11074 [1:10:56<22:44,  2.03it/s]
 75%|███████▌  | 8311/11074 [1:10:56<22:44,  2.03it/s]
 75%|███████▌  | 8312/11074 [1:10:57<22:43,  2.03it/s]
 75%|███████▌  | 8313/11074 [1:10:57<22:43,  2.02it/s]
 75%|███████▌  | 8314/11074 [1:10:58<22:43,  2.02it/s]
 75%|███████▌  | 8315/11074 [1:10:58<22:42,  2.02it/s]
 75%|███████▌  | 8316/11074 [1:10:59<22:41,  2.03it/s]
 75%|███████▌  | 8317/11074 [1:10:59<22:41,  2.03it/s]
 75%|███████▌  | 8318/11074 [1:11:00<22:39,  2.03it/s]
 75%|███████▌  | 8319/11074 [1:11:00<22:39,  2.03it/s]
 75%|███████▌  | 8320/11074 [1:11:01<22:39,  2.03it/s]
 75%|███████▌  | 8321/11074 [1:11:01<22:37,  2.03it/s]
 75%|███████▌  | 8322/11074 [1:11:02<22:38,  2.03it/s]
 75%|███████▌  | 8323/11074 [1:11:02<22:36,  2.03it/s]
 75%|███████▌  | 8324/11074 [1:11:03<22:36,  2.03it/s]
 75%|███████▌  | 8325/11074 [1:11:03<22:35,  2.03it/s]
                                                      {'loss': 3.244, 'grad_norm': 0.24260075390338898, 'learning_rate': 0.0001762778619941043, 'epoch': 10.52}
+
 75%|███████▌  | 8325/11074 [1:11:03<22:35,  2.03it/s]
 75%|███████▌  | 8326/11074 [1:11:04<24:35,  1.86it/s]
 75%|███████▌  | 8327/11074 [1:11:04<23:58,  1.91it/s]
 75%|███████▌  | 8328/11074 [1:11:05<23:32,  1.94it/s]
 75%|███████▌  | 8329/11074 [1:11:05<23:14,  1.97it/s]
 75%|███████▌  | 8330/11074 [1:11:06<23:02,  1.99it/s]
 75%|███████▌  | 8331/11074 [1:11:06<22:53,  2.00it/s]
 75%|███████▌  | 8332/11074 [1:11:07<22:47,  2.00it/s]
 75%|███████▌  | 8333/11074 [1:11:07<22:42,  2.01it/s]
 75%|███████▌  | 8334/11074 [1:11:08<22:37,  2.02it/s]
 75%|███████▌  | 8335/11074 [1:11:08<22:36,  2.02it/s]
 75%|███████▌  | 8336/11074 [1:11:09<22:35,  2.02it/s]
 75%|███████▌  | 8337/11074 [1:11:09<22:34,  2.02it/s]
 75%|███████▌  | 8338/11074 [1:11:10<22:32,  2.02it/s]
 75%|███████▌  | 8339/11074 [1:11:10<22:32,  2.02it/s]
 75%|███████▌  | 8340/11074 [1:11:11<22:31,  2.02it/s]
 75%|███████▌  | 8341/11074 [1:11:11<22:31,  2.02it/s]
 75%|███████▌  | 8342/11074 [1:11:12<22:29,  2.02it/s]
 75%|███████▌  | 8343/11074 [1:11:12<22:29,  2.02it/s]
 75%|███████▌  | 8344/11074 [1:11:13<22:27,  2.03it/s]
 75%|███████▌  | 8345/11074 [1:11:13<22:26,  2.03it/s]
 75%|███████▌  | 8346/11074 [1:11:14<22:27,  2.03it/s]
 75%|███████▌  | 8347/11074 [1:11:14<22:24,  2.03it/s]
 75%|███████▌  | 8348/11074 [1:11:15<22:25,  2.03it/s]
 75%|███████▌  | 8349/11074 [1:11:15<22:24,  2.03it/s]
 75%|███████▌  | 8350/11074 [1:11:16<22:24,  2.03it/s]{'loss': 3.2382, 'grad_norm': 0.23930463194847107, 'learning_rate': 0.0001732849254639809, 'epoch': 10.55}
                                                      
+
 75%|███████▌  | 8350/11074 [1:11:16<22:24,  2.03it/s]
 75%|███████▌  | 8351/11074 [1:11:16<22:25,  2.02it/s]
 75%|███████▌  | 8352/11074 [1:11:17<22:26,  2.02it/s]
 75%|███████▌  | 8353/11074 [1:11:17<22:25,  2.02it/s]
 75%|███████▌  | 8354/11074 [1:11:18<22:24,  2.02it/s]
 75%|███████▌  | 8355/11074 [1:11:18<22:22,  2.03it/s]
 75%|███████▌  | 8356/11074 [1:11:19<22:22,  2.02it/s]
 75%|███████▌  | 8357/11074 [1:11:19<22:21,  2.03it/s]
 75%|███████▌  | 8358/11074 [1:11:20<22:19,  2.03it/s]
 75%|███████▌  | 8359/11074 [1:11:20<22:19,  2.03it/s]
 75%|███████▌  | 8360/11074 [1:11:21<22:17,  2.03it/s]
 76%|███████▌  | 8361/11074 [1:11:21<22:17,  2.03it/s]
 76%|███████▌  | 8362/11074 [1:11:22<22:17,  2.03it/s]
 76%|███████▌  | 8363/11074 [1:11:22<22:18,  2.03it/s]
 76%|███████▌  | 8364/11074 [1:11:23<22:17,  2.03it/s]
 76%|███████▌  | 8365/11074 [1:11:23<22:18,  2.02it/s]
 76%|███████▌  | 8366/11074 [1:11:24<22:15,  2.03it/s]
 76%|███████▌  | 8367/11074 [1:11:24<22:16,  2.03it/s]
 76%|███████▌  | 8368/11074 [1:11:25<22:16,  2.02it/s]
 76%|███████▌  | 8369/11074 [1:11:25<22:16,  2.02it/s]
 76%|███████▌  | 8370/11074 [1:11:26<22:16,  2.02it/s]
 76%|███████▌  | 8371/11074 [1:11:26<22:15,  2.02it/s]
 76%|███████▌  | 8372/11074 [1:11:27<22:15,  2.02it/s]
 76%|███████▌  | 8373/11074 [1:11:27<22:14,  2.02it/s]
 76%|███████▌  | 8374/11074 [1:11:28<22:15,  2.02it/s]
 76%|███████▌  | 8375/11074 [1:11:28<22:13,  2.02it/s]{'loss': 3.2304, 'grad_norm': 0.23772986233234406, 'learning_rate': 0.000170312280002615, 'epoch': 10.58}
                                                      
+
 76%|███████▌  | 8375/11074 [1:11:28<22:13,  2.02it/s]
 76%|███████▌  | 8376/11074 [1:11:29<22:15,  2.02it/s]
 76%|███████▌  | 8377/11074 [1:11:29<22:13,  2.02it/s]
 76%|███████▌  | 8378/11074 [1:11:30<22:13,  2.02it/s]
 76%|███████▌  | 8379/11074 [1:11:30<22:11,  2.02it/s]
 76%|███████▌  | 8380/11074 [1:11:31<22:12,  2.02it/s]
 76%|███████▌  | 8381/11074 [1:11:31<22:11,  2.02it/s]
 76%|███████▌  | 8382/11074 [1:11:32<22:11,  2.02it/s]
 76%|███████▌  | 8383/11074 [1:11:32<22:09,  2.02it/s]
 76%|███████▌  | 8384/11074 [1:11:33<22:09,  2.02it/s]
 76%|███████▌  | 8385/11074 [1:11:33<22:07,  2.03it/s]
 76%|███████▌  | 8386/11074 [1:11:34<22:07,  2.02it/s]
 76%|███████▌  | 8387/11074 [1:11:34<22:07,  2.02it/s]
 76%|███████▌  | 8388/11074 [1:11:35<22:06,  2.02it/s]
 76%|███████▌  | 8389/11074 [1:11:35<22:06,  2.02it/s]
 76%|███████▌  | 8390/11074 [1:11:36<22:04,  2.03it/s]
 76%|███████▌  | 8391/11074 [1:11:36<22:05,  2.02it/s]
 76%|███████▌  | 8392/11074 [1:11:37<22:03,  2.03it/s]
 76%|███████▌  | 8393/11074 [1:11:37<22:04,  2.02it/s]
 76%|███████▌  | 8394/11074 [1:11:38<22:02,  2.03it/s]
 76%|███████▌  | 8395/11074 [1:11:38<22:02,  2.03it/s]
 76%|███████▌  | 8396/11074 [1:11:39<22:01,  2.03it/s]
 76%|███████▌  | 8397/11074 [1:11:39<22:00,  2.03it/s]
 76%|███████▌  | 8398/11074 [1:11:39<22:01,  2.03it/s]
 76%|███████▌  | 8399/11074 [1:11:40<22:00,  2.03it/s]
 76%|███████▌  | 8400/11074 [1:11:40<22:00,  2.02it/s]{'loss': 3.2414, 'grad_norm': 0.23746074736118317, 'learning_rate': 0.00016736011023005699, 'epoch': 10.61}
+
                                                      
 76%|███████▌  | 8400/11074 [1:11:40<22:00,  2.02it/s]
 76%|███████▌  | 8401/11074 [1:11:41<22:01,  2.02it/s]
 76%|███████▌  | 8402/11074 [1:11:41<22:02,  2.02it/s]
 76%|███████▌  | 8403/11074 [1:11:42<22:00,  2.02it/s]
 76%|███████▌  | 8404/11074 [1:11:42<22:01,  2.02it/s]
 76%|███████▌  | 8405/11074 [1:11:43<21:58,  2.02it/s]
 76%|███████▌  | 8406/11074 [1:11:43<21:59,  2.02it/s]
 76%|███████▌  | 8407/11074 [1:11:44<21:57,  2.02it/s]
 76%|███████▌  | 8408/11074 [1:11:44<21:57,  2.02it/s]
 76%|███████▌  | 8409/11074 [1:11:45<21:57,  2.02it/s]
 76%|█���█████▌  | 8410/11074 [1:11:45<21:57,  2.02it/s]
 76%|███████▌  | 8411/11074 [1:11:46<21:57,  2.02it/s]
 76%|███████▌  | 8412/11074 [1:11:46<21:56,  2.02it/s]
 76%|███████▌  | 8413/11074 [1:11:47<21:56,  2.02it/s]
 76%|███████▌  | 8414/11074 [1:11:47<21:55,  2.02it/s]
 76%|███████▌  | 8415/11074 [1:11:48<21:55,  2.02it/s]
 76%|███████▌  | 8416/11074 [1:11:48<21:54,  2.02it/s]
 76%|███████▌  | 8417/11074 [1:11:49<21:54,  2.02it/s]
 76%|███████▌  | 8418/11074 [1:11:49<21:53,  2.02it/s]
 76%|███████▌  | 8419/11074 [1:11:50<21:54,  2.02it/s]
 76%|███████▌  | 8420/11074 [1:11:50<21:53,  2.02it/s]
 76%|███████▌  | 8421/11074 [1:11:51<21:51,  2.02it/s]
 76%|███████▌  | 8422/11074 [1:11:51<21:51,  2.02it/s]
 76%|███████▌  | 8423/11074 [1:11:52<21:51,  2.02it/s]
 76%|███████▌  | 8424/11074 [1:11:52<21:52,  2.02it/s]
 76%|███████▌  | 8425/11074 [1:11:53<21:50,  2.02it/s]{'loss': 3.2383, 'grad_norm': 0.23675471544265747, 'learning_rate': 0.0001644285994946871, 'epoch': 10.64}
                                                      
+
 76%|███████▌  | 8425/11074 [1:11:53<21:50,  2.02it/s]
 76%|███████▌  | 8426/11074 [1:11:53<21:56,  2.01it/s]
 76%|███████▌  | 8427/11074 [1:11:54<21:52,  2.02it/s]
 76%|███████▌  | 8428/11074 [1:11:54<21:50,  2.02it/s]
 76%|███████▌  | 8429/11074 [1:11:55<21:47,  2.02it/s]
 76%|███████▌  | 8430/11074 [1:11:55<21:48,  2.02it/s]
 76%|███████▌  | 8431/11074 [1:11:56<21:46,  2.02it/s]
 76%|███████▌  | 8432/11074 [1:11:56<21:46,  2.02it/s]
 76%|███████▌  | 8433/11074 [1:11:57<21:45,  2.02it/s]
 76%|███████▌  | 8434/11074 [1:11:57<21:43,  2.03it/s]
 76%|███████▌  | 8435/11074 [1:11:58<21:43,  2.02it/s]
 76%|███████▌  | 8436/11074 [1:11:58<21:41,  2.03it/s]
 76%|███████▌  | 8437/11074 [1:11:59<21:41,  2.03it/s]
 76%|███████▌  | 8438/11074 [1:11:59<21:41,  2.03it/s]
 76%|███████▌  | 8439/11074 [1:12:00<21:40,  2.03it/s]
 76%|███████▌  | 8440/11074 [1:12:00<21:40,  2.02it/s]
 76%|███████▌  | 8441/11074 [1:12:01<21:38,  2.03it/s]
 76%|███████▌  | 8442/11074 [1:12:01<21:39,  2.03it/s]
 76%|███████▌  | 8443/11074 [1:12:02<21:37,  2.03it/s]
 76%|███████▋  | 8444/11074 [1:12:02<21:37,  2.03it/s]
 76%|███████▋  | 8445/11074 [1:12:03<21:37,  2.03it/s]
 76%|███████▋  | 8446/11074 [1:12:03<21:38,  2.02it/s]
 76%|███████▋  | 8447/11074 [1:12:04<21:38,  2.02it/s]
 76%|███████▋  | 8448/11074 [1:12:04<21:38,  2.02it/s]
 76%|███████▋  | 8449/11074 [1:12:05<21:36,  2.02it/s]
 76%|███████▋  | 8450/11074 [1:12:05<21:34,  2.03it/s]
                                                      
+
 76%|███████▋  | 8450/11074 [1:12:05<21:34,  2.03it/s]
 76%|███████▋  | 8451/11074 [1:12:06<21:37,  2.02it/s]
 76%|███████▋  | 8452/11074 [1:12:06<21:34,  2.03it/s]
 76%|███████▋  | 8453/11074 [1:12:07<21:34,  2.02it/s]
 76%|███████▋  | 8454/11074 [1:12:07<21:33,  2.03it/s]
 76%|███████▋  | 8455/11074 [1:12:08<21:33,  2.02it/s]
 76%|███████▋  | 8456/11074 [1:12:08<21:33,  2.02it/s]
 76%|███████▋  | 8457/11074 [1:12:09<21:34,  2.02it/s]
 76%|███████▋  | 8458/11074 [1:12:09<21:33,  2.02it/s]
 76%|███████▋  | 8459/11074 [1:12:10<21:32,  2.02it/s]
 76%|███████▋  | 8460/11074 [1:12:10<21:30,  2.03it/s]
 76%|███████▋  | 8461/11074 [1:12:11<21:31,  2.02it/s]
 76%|███████▋  | 8462/11074 [1:12:11<21:29,  2.02it/s]
 76%|███████▋  | 8463/11074 [1:12:12<21:30,  2.02it/s]
 76%|███████▋  | 8464/11074 [1:12:12<21:29,  2.02it/s]
 76%|███████▋  | 8465/11074 [1:12:13<21:28,  2.02it/s]
 76%|███████▋  | 8466/11074 [1:12:13<21:28,  2.02it/s]
 76%|███████▋  | 8467/11074 [1:12:14<21:27,  2.03it/s]
 76%|███████▋  | 8468/11074 [1:12:14<21:27,  2.02it/s]
 76%|███████▋  | 8469/11074 [1:12:15<21:26,  2.03it/s]
 76%|███████▋  | 8470/11074 [1:12:15<21:26,  2.02it/s]
 76%|███████▋  | 8471/11074 [1:12:16<21:24,  2.03it/s]
 77%|███████▋  | 8472/11074 [1:12:16<21:24,  2.03it/s]
 77%|███████▋  | 8473/11074 [1:12:17<21:24,  2.03it/s]
 77%|███████▋  | 8474/11074 [1:12:17<21:23,  2.02it/s]
 77%|███████▋  | 8475/11074 [1:12:18<21:23,  2.03it/s]{'loss': 3.2461, 'grad_norm': 0.23413671553134918, 'learning_rate': 0.00015862828210244434, 'epoch': 10.71}
                                                      
+
 77%|███████▋  | 8475/11074 [1:12:18<21:23,  2.03it/s]
 77%|███████▋  | 8476/11074 [1:12:18<21:25,  2.02it/s]
 77%|███████▋  | 8477/11074 [1:12:19<21:23,  2.02it/s]
 77%|███████▋  | 8478/11074 [1:12:19<21:23,  2.02it/s]
 77%|███████▋  | 8479/11074 [1:12:20<21:22,  2.02it/s]
 77%|███████▋  | 8480/11074 [1:12:20<21:21,  2.02it/s]
 77%|███████▋  | 8481/11074 [1:12:21<21:21,  2.02it/s]
 77%|███████▋  | 8482/11074 [1:12:21<21:20,  2.02it/s]
 77%|███████▋  | 8483/11074 [1:12:22<21:20,  2.02it/s]
 77%|███████▋  | 8484/11074 [1:12:22<21:19,  2.02it/s]
 77%|███████▋  | 8485/11074 [1:12:22<21:20,  2.02it/s]
 77%|███████▋  | 8486/11074 [1:12:23<21:19,  2.02it/s]
 77%|███████▋  | 8487/11074 [1:12:23<21:18,  2.02it/s]
 77%|███████▋  | 8488/11074 [1:12:24<21:17,  2.03it/s]
 77%|███████▋  | 8489/11074 [1:12:24<21:16,  2.02it/s]
 77%|███████▋  | 8490/11074 [1:12:25<21:15,  2.03it/s]
 77%|███████▋  | 8491/11074 [1:12:25<21:16,  2.02it/s]
 77%|███████▋  | 8492/11074 [1:12:26<21:14,  2.03it/s]
 77%|███████▋  | 8493/11074 [1:12:26<21:14,  2.02it/s]
 77%|███████▋  | 8494/11074 [1:12:27<21:13,  2.03it/s]
 77%|███████▋  | 8495/11074 [1:12:27<21:14,  2.02it/s]
 77%|███████▋  | 8496/11074 [1:12:28<21:13,  2.02it/s]
 77%|███████▋  | 8497/11074 [1:12:28<21:12,  2.03it/s]
 77%|███████▋  | 8498/11074 [1:12:29<21:12,  2.02it/s]
 77%|███████▋  | 8499/11074 [1:12:29<21:12,  2.02it/s]
 77%|███████▋  | 8500/11074 [1:12:30<21:12,  2.02it/s]
                                                      {'loss': 3.2367, 'grad_norm': 0.23599669337272644, 'learning_rate': 0.0001557598356819, 'epoch': 10.74}
+
 77%|███████▋  | 8500/11074 [1:12:30<21:12,  2.02it/s]
 77%|███████▋  | 8501/11074 [1:12:30<21:14,  2.02it/s]
 77%|███████▋  | 8502/11074 [1:12:31<21:12,  2.02it/s]
 77%|███████▋  | 8503/11074 [1:12:31<21:12,  2.02it/s]
 77%|███████▋  | 8504/11074 [1:12:32<21:10,  2.02it/s]
 77%|███████▋  | 8505/11074 [1:12:32<21:09,  2.02it/s]
 77%|███████▋  | 8506/11074 [1:12:33<21:09,  2.02it/s]
 77%|███████▋  | 8507/11074 [1:12:33<21:07,  2.02it/s]
 77%|███████▋  | 8508/11074 [1:12:34<21:08,  2.02it/s]
 77%|███████▋  | 8509/11074 [1:12:34<21:08,  2.02it/s]
 77%|███████▋  | 8510/11074 [1:12:35<21:07,  2.02it/s]
 77%|███████▋  | 8511/11074 [1:12:35<21:06,  2.02it/s]
 77%|███████▋  | 8512/11074 [1:12:36<21:06,  2.02it/s]
 77%|███████▋  | 8513/11074 [1:12:36<21:03,  2.03it/s]
 77%|███████▋  | 8514/11074 [1:12:37<21:03,  2.03it/s]
 77%|███████▋  | 8515/11074 [1:12:37<21:03,  2.03it/s]
 77%|███████▋  | 8516/11074 [1:12:38<21:03,  2.02it/s]
 77%|███████▋  | 8517/11074 [1:12:38<21:02,  2.03it/s]
 77%|███████▋  | 8518/11074 [1:12:39<21:01,  2.03it/s]
 77%|███████▋  | 8519/11074 [1:12:39<21:01,  2.03it/s]
 77%|███████▋  | 8520/11074 [1:12:40<20:59,  2.03it/s]
 77%|███████▋  | 8521/11074 [1:12:40<21:00,  2.03it/s]
 77%|███████▋  | 8522/11074 [1:12:41<20:58,  2.03it/s]
 77%|███████▋  | 8523/11074 [1:12:41<20:59,  2.03it/s]
 77%|███████▋  | 8524/11074 [1:12:42<20:58,  2.03it/s]
 77%|███████▋  | 8525/11074 [1:12:42<20:58,  2.03it/s]{'loss': 3.2446, 'grad_norm': 0.2339746206998825, 'learning_rate': 0.00015291276874882887, 'epoch': 10.77}
+
                                                      
 77%|███████▋  | 8525/11074 [1:12:42<20:58,  2.03it/s]
 77%|███████▋  | 8526/11074 [1:12:43<20:59,  2.02it/s]
 77%|███████▋  | 8527/11074 [1:12:43<20:59,  2.02it/s]
 77%|███████▋  | 8528/11074 [1:12:44<20:58,  2.02it/s]
 77%|███████▋  | 8529/11074 [1:12:44<20:57,  2.02it/s]
 77%|███████▋  | 8530/11074 [1:12:45<20:57,  2.02it/s]
 77%|███████▋  | 8531/11074 [1:12:45<20:55,  2.03it/s]
 77%|███████▋  | 8532/11074 [1:12:46<20:55,  2.02it/s]
 77%|███████▋  | 8533/11074 [1:12:46<20:54,  2.03it/s]
 77%|███████▋  | 8534/11074 [1:12:47<20:54,  2.02it/s]
 77%|███████▋  | 8535/11074 [1:12:47<20:53,  2.03it/s]
 77%|███████▋  | 8536/11074 [1:12:48<20:53,  2.02it/s]
 77%|███████▋  | 8537/11074 [1:12:48<20:54,  2.02it/s]
 77%|███████▋  | 8538/11074 [1:12:49<20:53,  2.02it/s]
 77%|███████▋  | 8539/11074 [1:12:49<20:54,  2.02it/s]
 77%|███████▋  | 8540/11074 [1:12:50<20:55,  2.02it/s]
 77%|███████▋  | 8541/11074 [1:12:50<20:54,  2.02it/s]
 77%|███████▋  | 8542/11074 [1:12:51<20:52,  2.02it/s]
 77%|███████▋  | 8543/11074 [1:12:51<20:52,  2.02it/s]
 77%|███████▋  | 8544/11074 [1:12:52<20:51,  2.02it/s]
 77%|███████▋  | 8545/11074 [1:12:52<20:51,  2.02it/s]
 77%|███████▋  | 8546/11074 [1:12:53<20:50,  2.02it/s]
 77%|███████▋  | 8547/11074 [1:12:53<20:48,  2.02it/s]
 77%|███████▋  | 8548/11074 [1:12:54<20:48,  2.02it/s]
 77%|███████▋  | 8549/11074 [1:12:54<20:47,  2.02it/s]
 77%|███████▋  | 8550/11074 [1:12:55<20:47,  2.02it/s]{'loss': 3.24, 'grad_norm': 0.24058009684085846, 'learning_rate': 0.00015008725812406143, 'epoch': 10.8}
+
                                                      
 77%|███████▋  | 8550/11074 [1:12:55<20:47,  2.02it/s]
 77%|███████▋  | 8551/11074 [1:12:55<20:48,  2.02it/s]
 77%|███████▋  | 8552/11074 [1:12:56<20:48,  2.02it/s]
 77%|███████▋  | 8553/11074 [1:12:56<20:47,  2.02it/s]
 77%|███████▋  | 8554/11074 [1:12:57<20:46,  2.02it/s]
 77%|███████▋  | 8555/11074 [1:12:57<20:46,  2.02it/s]
 77%|███████▋  | 8556/11074 [1:12:58<20:44,  2.02it/s]
 77%|███████▋  | 8557/11074 [1:12:58<20:43,  2.02it/s]
 77%|███████▋  | 8558/11074 [1:12:59<20:42,  2.03it/s]
 77%|███████▋  | 8559/11074 [1:12:59<20:41,  2.03it/s]
 77%|███████▋  | 8560/11074 [1:13:00<20:41,  2.03it/s]
 77%|███████▋  | 8561/11074 [1:13:00<20:41,  2.02it/s]
 77%|███████▋  | 8562/11074 [1:13:01<20:41,  2.02it/s]
 77%|███████▋  | 8563/11074 [1:13:01<20:40,  2.02it/s]
 77%|███████▋  | 8564/11074 [1:13:02<20:40,  2.02it/s]
 77%|███████▋  | 8565/11074 [1:13:02<20:38,  2.03it/s]
 77%|███████▋  | 8566/11074 [1:13:03<20:39,  2.02it/s]
 77%|███████▋  | 8567/11074 [1:13:03<20:36,  2.03it/s]
 77%|███████▋  | 8568/11074 [1:13:04<20:37,  2.03it/s]
 77%|███████▋  | 8569/11074 [1:13:04<20:36,  2.03it/s]
 77%|███████▋  | 8570/11074 [1:13:04<20:36,  2.02it/s]
 77%|███████▋  | 8571/11074 [1:13:05<20:36,  2.02it/s]
 77%|███████▋  | 8572/11074 [1:13:05<20:34,  2.03it/s]
 77%|███████▋  | 8573/11074 [1:13:06<20:36,  2.02it/s]
 77%|███████▋  | 8574/11074 [1:13:06<20:33,  2.03it/s]
 77%|███████▋  | 8575/11074 [1:13:07<20:34,  2.02it/s]{'loss': 3.2475, 'grad_norm': 0.23284657299518585, 'learning_rate': 0.00014728347928964547, 'epoch': 10.83}
+
                                                      
 77%|███████▋  | 8575/11074 [1:13:07<20:34,  2.02it/s]
 77%|███████▋  | 8576/11074 [1:13:07<20:34,  2.02it/s]
 77%|███████▋  | 8577/11074 [1:13:08<20:33,  2.02it/s]
 77%|███████▋  | 8578/11074 [1:13:08<20:33,  2.02it/s]
 77%|███████▋  | 8579/11074 [1:13:09<20:33,  2.02it/s]
 77%|███████▋  | 8580/11074 [1:13:09<20:32,  2.02it/s]
 77%|███████▋  | 8581/11074 [1:13:10<20:31,  2.02it/s]
 77%|███████▋  | 8582/11074 [1:13:10<20:31,  2.02it/s]
 78%|███████▊  | 8583/11074 [1:13:11<20:29,  2.03it/s]
 78%|███████▊  | 8584/11074 [1:13:11<20:29,  2.02it/s]
 78%|███████▊  | 8585/11074 [1:13:12<20:29,  2.02it/s]
 78%|███████▊  | 8586/11074 [1:13:12<20:30,  2.02it/s]
 78%|███████▊  | 8587/11074 [1:13:13<20:28,  2.02it/s]
 78%|███████▊  | 8588/11074 [1:13:13<20:30,  2.02it/s]
 78%|███████▊  | 8589/11074 [1:13:14<20:37,  2.01it/s]
 78%|███████▊  | 8590/11074 [1:13:14<20:33,  2.01it/s]
 78%|███████▊  | 8591/11074 [1:13:15<20:30,  2.02it/s]
 78%|███████▊  | 8592/11074 [1:13:15<20:29,  2.02it/s]
 78%|███████▊  | 8593/11074 [1:13:16<20:28,  2.02it/s]
 78%|███████▊  | 8594/11074 [1:13:16<20:27,  2.02it/s]
 78%|███████▊  | 8595/11074 [1:13:17<20:26,  2.02it/s]
 78%|███████▊  | 8596/11074 [1:13:17<20:26,  2.02it/s]
 78%|███████▊  | 8597/11074 [1:13:18<20:25,  2.02it/s]
 78%|███████▊  | 8598/11074 [1:13:18<20:25,  2.02it/s]
 78%|███████▊  | 8599/11074 [1:13:19<20:24,  2.02it/s]
 78%|███████▊  | 8600/11074 [1:13:19<20:23,  2.02it/s]{'loss': 3.2479, 'grad_norm': 0.2400185614824295, 'learning_rate': 0.00014450160637794786, 'epoch': 10.87}
+
                                                      
 78%|███████▊  | 8600/11074 [1:13:19<20:23,  2.02it/s]
 78%|███████▊  | 8601/11074 [1:13:20<20:24,  2.02it/s]
 78%|███████▊  | 8602/11074 [1:13:20<20:22,  2.02it/s]
 78%|███████▊  | 8603/11074 [1:13:21<20:21,  2.02it/s]
 78%|███████▊  | 8604/11074 [1:13:21<20:20,  2.02it/s]
 78%|███████▊  | 8605/11074 [1:13:22<20:21,  2.02it/s]
 78%|███████▊  | 8606/11074 [1:13:22<20:20,  2.02it/s]
 78%|███████▊  | 8607/11074 [1:13:23<20:19,  2.02it/s]
 78%|███████▊  | 8608/11074 [1:13:23<20:17,  2.02it/s]
 78%|███████▊  | 8609/11074 [1:13:24<20:18,  2.02it/s]
 78%|███████▊  | 8610/11074 [1:13:24<20:17,  2.02it/s]
 78%|███████▊  | 8611/11074 [1:13:25<20:18,  2.02it/s]
 78%|███████▊  | 8612/11074 [1:13:25<20:16,  2.02it/s]
 78%|███████▊  | 8613/11074 [1:13:26<20:16,  2.02it/s]
 78%|███████▊  | 8614/11074 [1:13:26<20:14,  2.02it/s]
 78%|███████▊  | 8615/11074 [1:13:27<20:15,  2.02it/s]
 78%|███████▊  | 8616/11074 [1:13:27<20:14,  2.02it/s]
 78%|███████▊  | 8617/11074 [1:13:28<20:14,  2.02it/s]
 78%|███████▊  | 8618/11074 [1:13:28<20:13,  2.02it/s]
 78%|███████▊  | 8619/11074 [1:13:29<20:11,  2.03it/s]
 78%|███████▊  | 8620/11074 [1:13:29<20:12,  2.02it/s]
 78%|███████▊  | 8621/11074 [1:13:30<20:10,  2.03it/s]
 78%|███████▊  | 8622/11074 [1:13:30<20:11,  2.02it/s]
 78%|███████▊  | 8623/11074 [1:13:31<20:09,  2.03it/s]
 78%|███████▊  | 8624/11074 [1:13:31<20:09,  2.03it/s]
 78%|███████▊  | 8625/11074 [1:13:32<20:08,  2.03it/s]{'loss': 3.2424, 'grad_norm': 0.24160033464431763, 'learning_rate': 0.00014174181216083863, 'epoch': 10.9}
+
                                                      
 78%|███████▊  | 8625/11074 [1:13:32<20:08,  2.03it/s]
 78%|███████▊  | 8626/11074 [1:13:32<20:14,  2.02it/s]
 78%|███████▊  | 8627/11074 [1:13:33<20:11,  2.02it/s]
 78%|███████▊  | 8628/11074 [1:13:33<20:10,  2.02it/s]
 78%|███████▊  | 8629/11074 [1:13:34<20:08,  2.02it/s]
 78%|███████▊  | 8630/11074 [1:13:34<20:09,  2.02it/s]
 78%|███████▊  | 8631/11074 [1:13:35<20:06,  2.02it/s]
 78%|███████▊  | 8632/11074 [1:13:35<20:08,  2.02it/s]
 78%|███████▊  | 8633/11074 [1:13:36<20:06,  2.02it/s]
 78%|███████▊  | 8634/11074 [1:13:36<20:05,  2.02it/s]
 78%|███████▊  | 8635/11074 [1:13:37<20:04,  2.02it/s]
 78%|███████▊  | 8636/11074 [1:13:37<20:04,  2.02it/s]
 78%|███████▊  | 8637/11074 [1:13:38<20:02,  2.03it/s]
 78%|███████▊  | 8638/11074 [1:13:38<20:02,  2.03it/s]
 78%|███████▊  | 8639/11074 [1:13:39<20:03,  2.02it/s]
 78%|███████▊  | 8640/11074 [1:13:39<20:03,  2.02it/s]
 78%|███████▊  | 8641/11074 [1:13:40<20:02,  2.02it/s]
 78%|███████▊  | 8642/11074 [1:13:40<20:03,  2.02it/s]
 78%|███████▊  | 8643/11074 [1:13:41<20:02,  2.02it/s]
 78%|███████▊  | 8644/11074 [1:13:41<20:02,  2.02it/s]
 78%|███████▊  | 8645/11074 [1:13:42<20:00,  2.02it/s]
 78%|███████▊  | 8646/11074 [1:13:42<20:01,  2.02it/s]
 78%|███████▊  | 8647/11074 [1:13:43<20:00,  2.02it/s]
 78%|███████▊  | 8648/11074 [1:13:43<20:00,  2.02it/s]
 78%|███████▊  | 8649/11074 [1:13:44<19:59,  2.02it/s]
 78%|███████▊  | 8650/11074 [1:13:44<19:59,  2.02it/s]
                                                      {'loss': 3.2466, 'grad_norm': 0.23945370316505432, 'learning_rate': 0.00013900426803896234, 'epoch': 10.93}
+
 78%|███████▊  | 8650/11074 [1:13:44<19:59,  2.02it/s]
 78%|███████▊  | 8651/11074 [1:13:45<19:59,  2.02it/s]
 78%|███████▊  | 8652/11074 [1:13:45<19:59,  2.02it/s]
 78%|███████▊  | 8653/11074 [1:13:46<19:57,  2.02it/s]
 78%|███████▊  | 8654/11074 [1:13:46<19:56,  2.02it/s]
 78%|███████▊  | 8655/11074 [1:13:47<19:55,  2.02it/s]
 78%|███████▊  | 8656/11074 [1:13:47<19:56,  2.02it/s]
 78%|███████▊  | 8657/11074 [1:13:48<19:55,  2.02it/s]
 78%|███████▊  | 8658/11074 [1:13:48<19:56,  2.02it/s]
 78%|███████▊  | 8659/11074 [1:13:49<19:54,  2.02it/s]
 78%|███████▊  | 8660/11074 [1:13:49<19:55,  2.02it/s]
 78%|███████▊  | 8661/11074 [1:13:49<19:53,  2.02it/s]
 78%|███████▊  | 8662/11074 [1:13:50<19:52,  2.02it/s]
 78%|███████▊  | 8663/11074 [1:13:50<19:52,  2.02it/s]
 78%|███████▊  | 8664/11074 [1:13:51<19:50,  2.02it/s]
 78%|███████▊  | 8665/11074 [1:13:51<19:51,  2.02it/s]
 78%|███████▊  | 8666/11074 [1:13:52<19:51,  2.02it/s]
 78%|███████▊  | 8667/11074 [1:13:52<19:50,  2.02it/s]
 78%|███████▊  | 8668/11074 [1:13:53<19:48,  2.02it/s]
 78%|███████▊  | 8669/11074 [1:13:53<19:49,  2.02it/s]
 78%|███████▊  | 8670/11074 [1:13:54<19:48,  2.02it/s]
 78%|███████▊  | 8671/11074 [1:13:54<19:48,  2.02it/s]
 78%|███████▊  | 8672/11074 [1:13:55<19:48,  2.02it/s]
 78%|███████▊  | 8673/11074 [1:13:55<19:47,  2.02it/s]
 78%|███████▊  | 8674/11074 [1:13:56<19:47,  2.02it/s]
 78%|███████▊  | 8675/11074 [1:13:56<19:46,  2.02it/s]
                                                      
+
 78%|███████▊  | 8675/11074 [1:13:56<19:46,  2.02it/s]
 78%|███████▊  | 8676/11074 [1:13:57<19:46,  2.02it/s]
 78%|███████▊  | 8677/11074 [1:13:57<19:45,  2.02it/s]
 78%|███████▊  | 8678/11074 [1:13:58<19:45,  2.02it/s]
 78%|███████▊  | 8679/11074 [1:13:58<19:43,  2.02it/s]
 78%|███████▊  | 8680/11074 [1:13:59<19:44,  2.02it/s]
 78%|███████▊  | 8681/11074 [1:13:59<19:43,  2.02it/s]
 78%|███████▊  | 8682/11074 [1:14:00<19:42,  2.02it/s]
 78%|███████▊  | 8683/11074 [1:14:00<19:42,  2.02it/s]
 78%|███████▊  | 8684/11074 [1:14:01<19:41,  2.02it/s]
 78%|███████▊  | 8685/11074 [1:14:01<19:41,  2.02it/s]
 78%|███████▊  | 8686/11074 [1:14:02<19:42,  2.02it/s]
 78%|███████▊  | 8687/11074 [1:14:02<19:41,  2.02it/s]
 78%|███████▊  | 8688/11074 [1:14:03<19:40,  2.02it/s]
 78%|███████▊  | 8689/11074 [1:14:03<19:39,  2.02it/s]
 78%|███████▊  | 8690/11074 [1:14:04<19:40,  2.02it/s]
 78%|███████▊  | 8691/11074 [1:14:04<19:40,  2.02it/s]
 78%|███████▊  | 8692/11074 [1:14:05<19:38,  2.02it/s]
 78%|███████▊  | 8693/11074 [1:14:05<19:37,  2.02it/s]
 79%|███████▊  | 8694/11074 [1:14:06<19:36,  2.02it/s]
 79%|███████▊  | 8695/11074 [1:14:06<19:36,  2.02it/s]
 79%|███████▊  | 8696/11074 [1:14:07<19:35,  2.02it/s]
 79%|███████▊  | 8697/11074 [1:14:07<19:35,  2.02it/s]
 79%|███████▊  | 8698/11074 [1:14:08<19:35,  2.02it/s]
 79%|███████▊  | 8699/11074 [1:14:08<19:34,  2.02it/s]
 79%|███████▊  | 8700/11074 [1:14:09<19:33,  2.02it/s]{'loss': 3.2503, 'grad_norm': 0.24186570942401886, 'learning_rate': 0.000133596608763568, 'epoch': 10.99}
+
                                                      
 79%|███████▊  | 8700/11074 [1:14:09<19:33,  2.02it/s]
 79%|███████▊  | 8701/11074 [1:14:09<19:34,  2.02it/s]
 79%|███████▊  | 8702/11074 [1:14:10<19:34,  2.02it/s]
 79%|███████▊  | 8703/11074 [1:14:10<19:33,  2.02it/s]
 79%|███████▊  | 8704/11074 [1:14:11<19:33,  2.02it/s]
 79%|███████▊  | 8705/11074 [1:14:11<19:31,  2.02it/s]
 79%|███████▊  | 8706/11074 [1:14:12<20:00,  1.97it/s]
 79%|███████▊  | 8707/11074 [1:14:24<2:38:17,  4.01s/it]
 79%|███████▊  | 8708/11074 [1:14:24<1:56:41,  2.96s/it]
 79%|███████▊  | 8709/11074 [1:14:25<1:27:30,  2.22s/it]
 79%|███████▊  | 8710/11074 [1:14:25<1:07:12,  1.71s/it]
 79%|███████▊  | 8711/11074 [1:14:26<52:50,  1.34s/it]  
 79%|███████▊  | 8712/11074 [1:14:26<42:50,  1.09s/it]
 79%|███████▊  | 8713/11074 [1:14:27<35:46,  1.10it/s]
 79%|███████▊  | 8714/11074 [1:14:27<30:52,  1.27it/s]
 79%|███████▊  | 8715/11074 [1:14:28<27:25,  1.43it/s]
 79%|███████▊  | 8716/11074 [1:14:28<25:00,  1.57it/s]
 79%|███████▊  | 8717/11074 [1:14:29<23:20,  1.68it/s]
 79%|███████▊  | 8718/11074 [1:14:29<22:07,  1.77it/s]
 79%|███████▊  | 8719/11074 [1:14:30<21:26,  1.83it/s]
 79%|███████▊  | 8720/11074 [1:14:30<20:52,  1.88it/s]
 79%|███████▉  | 8721/11074 [1:14:31<20:27,  1.92it/s]
 79%|███████▉  | 8722/11074 [1:14:31<20:06,  1.95it/s]
 79%|███████▉  | 8723/11074 [1:14:32<19:52,  1.97it/s]
 79%|███████▉  | 8724/11074 [1:14:32<19:42,  1.99it/s]
 79%|███████▉  | 8725/11074 [1:14:33<19:37,  2.00it/s]
                                                      {'loss': 3.1825, 'grad_norm': 0.2393781691789627, 'learning_rate': 0.0001309268294598309, 'epoch': 11.02}
+
 79%|███████▉  | 8725/11074 [1:14:33<19:37,  2.00it/s]
 79%|███████▉  | 8726/11074 [1:14:33<19:32,  2.00it/s]
 79%|███████▉  | 8727/11074 [1:14:34<19:28,  2.01it/s]
 79%|███████▉  | 8728/11074 [1:14:34<19:27,  2.01it/s]
 79%|███████▉  | 8729/11074 [1:14:35<19:24,  2.01it/s]
 79%|███████▉  | 8730/11074 [1:14:35<19:22,  2.02it/s]
 79%|███████▉  | 8731/11074 [1:14:36<19:23,  2.01it/s]
 79%|███████▉  | 8732/11074 [1:14:36<19:22,  2.01it/s]
 79%|███████▉  | 8733/11074 [1:14:37<19:21,  2.01it/s]
 79%|███████▉  | 8734/11074 [1:14:37<19:20,  2.02it/s]
 79%|███████▉  | 8735/11074 [1:14:38<19:17,  2.02it/s]
 79%|███████▉  | 8736/11074 [1:14:38<19:15,  2.02it/s]
 79%|███████▉  | 8737/11074 [1:14:39<19:15,  2.02it/s]