diff --git "a/train_job_output.txt" "b/train_job_output.txt"
--- "a/train_job_output.txt"
+++ "b/train_job_output.txt"
@@ -473,4 +473,67 @@ command outputs:
 
                                                    

 57%|█████▋    | 6275/11074 [53:35<39:35,  2.02it/s]
 57%|█████▋    | 6276/11074 [53:35<39:43,  2.01it/s]
 57%|█████▋    | 6277/11074 [53:36<39:37,  2.02it/s]
 57%|█████▋    | 6278/11074 [53:36<39:35,  2.02it/s]
 57%|█████▋    | 6279/11074 [53:37<39:33,  2.02it/s]
 57%|█████▋    | 6280/11074 [53:37<39:33,  2.02it/s]
 57%|█████▋    | 6281/11074 [53:38<39:33,  2.02it/s]
 57%|█████▋    | 6282/11074 [53:38<39:33,  2.02it/s]
 57%|█████▋    | 6283/11074 [53:39<39:30,  2.02it/s]
 57%|█████▋    | 6284/11074 [53:39<39:30,  2.02it/s]
 57%|█████▋    | 6285/11074 [53:40<39:30,  2.02it/s]
 57%|█████▋    | 6286/11074 [53:40<39:30,  2.02it/s]
 57%|█████▋    | 6287/11074 [53:41<39:28,  2.02it/s]
 57%|█████▋    | 6288/11074 [53:41<39:26,  2.02it/s]
 57%|█████▋    | 6289/11074 [53:42<39:24,  2.02it/s]
 57%|█████▋    | 6290/11074 [53:42<39:24,  2.02it/s]
 57%|█████▋    | 6291/11074 [53:43<39:23,  2.02it/s]
 57%|█████▋    | 6292/11074 [53:43<39:22,  2.02it/s]
 57%|█████▋    | 6293/11074 [53:44<39:22,  2.02it/s]
 57%|█████▋    | 6294/11074 [53:44<39:20,  2.02it/s]
 57%|█████▋    | 6295/11074 [53:45<39:23,  2.02it/s]
 57%|█████▋    | 6296/11074 [53:45<39:22,  2.02it/s]
 57%|█████▋    | 6297/11074 [53:45<39:23,  2.02it/s]
 57%|█████▋    | 6298/11074 [53:46<39:21,  2.02it/s]
 57%|█████▋    | 6299/11074 [53:46<39:22,  2.02it/s]
 57%|█████▋    | 6300/11074 [53:47<39:18,  2.02it/s]{'loss': 3.446, 'grad_norm': 0.22482430934906006, 'learning_rate': 0.00046708218113308836, 'epoch': 7.96}
 
                                                    

 57%|█████▋    | 6300/11074 [53:47<39:18,  2.02it/s]
 57%|█████▋    | 6301/11074 [53:47<39:21,  2.02it/s]
 57%|█████▋    | 6302/11074 [53:48<39:19,  2.02it/s]
 57%|█████▋    | 6303/11074 [53:48<39:15,  2.03it/s]
 57%|█████▋    | 6304/11074 [53:49<39:18,  2.02it/s]
 57%|█████▋    | 6305/11074 [53:49<39:15,  2.02it/s]
 57%|█████▋    | 6306/11074 [53:50<39:17,  2.02it/s]
 57%|█████▋    | 6307/11074 [53:50<39:14,  2.02it/s]
 57%|█████▋    | 6308/11074 [53:51<39:16,  2.02it/s]
 57%|█████▋    | 6309/11074 [53:51<39:17,  2.02it/s]
 57%|█████▋    | 6310/11074 [53:52<39:19,  2.02it/s]
 57%|█████▋    | 6311/11074 [53:52<39:17,  2.02it/s]
 57%|█████▋    | 6312/11074 [53:53<39:16,  2.02it/s]
 57%|█████▋    | 6313/11074 [53:53<39:13,  2.02it/s]
 57%|█████▋    | 6314/11074 [53:54<39:14,  2.02it/s]
 57%|█████▋    | 6315/11074 [53:54<39:12,  2.02it/s]
 57%|█████▋    | 6316/11074 [53:55<39:13,  2.02it/s]
 57%|█████▋    | 6317/11074 [53:55<39:11,  2.02it/s]
 57%|█████▋    | 6318/11074 [53:56<39:12,  2.02it/s]
 57%|█████▋    | 6319/11074 [53:56<39:10,  2.02it/s]
 57%|█████▋    | 6320/11074 [53:57<39:07,  2.02it/s]
 57%|█████▋    | 6321/11074 [53:57<39:11,  2.02it/s]
 57%|█████▋    | 6322/11074 [53:58<39:08,  2.02it/s]
 57%|█████▋    | 6323/11074 [53:58<39:07,  2.02it/s]
 57%|█████▋    | 6324/11074 [53:59<39:05,  2.03it/s]
 57%|█████▋    | 6325/11074 [53:59<39:08,  2.02it/s]{'loss': 3.442, 'grad_norm': 0.21625582873821259, 'learning_rate': 0.0004631514046443271, 'epoch': 7.99}
 
                                                    

 57%|█████▋    | 6325/11074 [53:59<39:08,  2.02it/s]
 57%|█████▋    | 6326/11074 [54:00<39:08,  2.02it/s]
 57%|█████▋    | 6327/11074 [54:00<39:09,  2.02it/s]
 57%|█████▋    | 6328/11074 [54:01<39:07,  2.02it/s]
 57%|█████▋    | 6329/11074 [54:01<39:07,  2.02it/s]
 57%|█████▋    | 6330/11074 [54:02<39:07,  2.02it/s]
 57%|█████▋    | 6331/11074 [54:02<39:08,  2.02it/s]
 57%|█████▋    | 6332/11074 [54:03<38:43,  2.04it/s]
 57%|█████▋    | 6333/11074 [54:15<5:13:09,  3.96s/it]
 57%|█████▋    | 6334/11074 [54:15<3:50:55,  2.92s/it]
 57%|█████▋    | 6335/11074 [54:16<2:53:17,  2.19s/it]
 57%|█████▋    | 6336/11074 [54:16<2:13:04,  1.69s/it]
 57%|█████▋    | 6337/11074 [54:17<1:44:59,  1.33s/it]
 57%|█████▋    | 6338/11074 [54:17<1:28:26,  1.12s/it]
 57%|█████▋    | 6339/11074 [54:18<1:13:58,  1.07it/s]
 57%|█████▋    | 6340/11074 [54:18<1:03:35,  1.24it/s]
 57%|█████▋    | 6341/11074 [54:19<56:11,  1.40it/s]  
 57%|█████▋    | 6342/11074 [54:19<51:04,  1.54it/s]
 57%|█████▋    | 6343/11074 [54:20<47:27,  1.66it/s]
 57%|█████▋    | 6344/11074 [54:20<44:56,  1.75it/s]
 57%|��████▋    | 6345/11074 [54:21<43:08,  1.83it/s]
 57%|█████▋    | 6346/11074 [54:21<41:54,  1.88it/s]
 57%|█████▋    | 6347/11074 [54:22<40:59,  1.92it/s]
 57%|█████▋    | 6348/11074 [54:22<40:21,  1.95it/s]
 57%|█████▋    | 6349/11074 [54:23<39:55,  1.97it/s]
 57%|█████▋    | 6350/11074 [54:23<39:43,  1.98it/s]{'loss': 3.3292, 'grad_norm': 0.22343750298023224, 'learning_rate': 0.00045922291668597107, 'epoch': 8.02}
-
                                                    

 57%|█████▋    | 6350/11074 [54:23<39:43,  1.98it/s]
 57%|█████▋    | 6351/11074 [54:24<39:31,  1.99it/s]
 57%|█████▋    | 6352/11074 [54:24<39:21,  2.00it/s]
 57%|█████▋    | 6353/11074 [54:25<39:13,  2.01it/s]
 57%|█████▋    | 6354/11074 [54:25<39:08,  2.01it/s]
 57%|█████▋    | 6355/11074 [54:26<39:06,  2.01it/s]
 57%|█████▋    | 6356/11074 [54:26<39:02,  2.01it/s]
\ No newline at end of file
+
                                                    

 57%|█████▋    | 6350/11074 [54:23<39:43,  1.98it/s]
 57%|█████▋    | 6351/11074 [54:24<39:31,  1.99it/s]
 57%|█████▋    | 6352/11074 [54:24<39:21,  2.00it/s]
 57%|█████▋    | 6353/11074 [54:25<39:13,  2.01it/s]
 57%|█████▋    | 6354/11074 [54:25<39:08,  2.01it/s]
 57%|█████▋    | 6355/11074 [54:26<39:06,  2.01it/s]
 57%|█████▋    | 6356/11074 [54:26<39:02,  2.01it/s]
 57%|█████▋    | 6357/11074 [54:27<39:03,  2.01it/s]
 57%|█████▋    | 6358/11074 [54:27<38:56,  2.02it/s]
 57%|█████▋    | 6359/11074 [54:28<38:54,  2.02it/s]
 57%|█████▋    | 6360/11074 [54:28<38:50,  2.02it/s]
 57%|█████▋    | 6361/11074 [54:29<38:51,  2.02it/s]
 57%|█████▋    | 6362/11074 [54:29<38:48,  2.02it/s]
 57%|█████▋    | 6363/11074 [54:30<38:50,  2.02it/s]
 57%|█████▋    | 6364/11074 [54:30<38:48,  2.02it/s]
 57%|█████▋    | 6365/11074 [54:31<38:46,  2.02it/s]
 57%|█████▋    | 6366/11074 [54:31<38:47,  2.02it/s]
 57%|█████▋    | 6367/11074 [54:32<38:46,  2.02it/s]
 58%|█████▊    | 6368/11074 [54:32<38:46,  2.02it/s]
 58%|█████▊    | 6369/11074 [54:33<38:44,  2.02it/s]
 58%|█████▊    | 6370/11074 [54:33<38:46,  2.02it/s]
 58%|█████▊    | 6371/11074 [54:34<38:43,  2.02it/s]
 58%|█████▊    | 6372/11074 [54:34<38:45,  2.02it/s]
 58%|█████▊    | 6373/11074 [54:35<38:41,  2.03it/s]
 58%|█████▊    | 6374/11074 [54:35<38:44,  2.02it/s]
 58%|█████▊    | 6375/11074 [54:36<38:42,  2.02it/s]{'loss': 3.3099, 'grad_norm': 0.22725610435009003, 'learning_rate': 0.00045529696124192416, 'epoch': 8.05}
+
                                                    

 58%|█████▊    | 6375/11074 [54:36<38:42,  2.02it/s]
 58%|█████▊    | 6376/11074 [54:36<38:45,  2.02it/s]
 58%|█████▊    | 6377/11074 [54:37<38:42,  2.02it/s]
 58%|█████▊    | 6378/11074 [54:37<38:39,  2.02it/s]
 58%|█████▊    | 6379/11074 [54:38<38:38,  2.03it/s]
 58%|█████▊    | 6380/11074 [54:38<38:36,  2.03it/s]
 58%|█████▊    | 6381/11074 [54:39<38:38,  2.02it/s]
 58%|█████▊    | 6382/11074 [54:39<38:35,  2.03it/s]
 58%|█████▊    | 6383/11074 [54:40<38:36,  2.03it/s]
 58%|█████▊    | 6384/11074 [54:40<38:35,  2.03it/s]
 58%|█████▊    | 6385/11074 [54:41<38:35,  2.03it/s]
 58%|█████▊    | 6386/11074 [54:41<38:40,  2.02it/s]
 58%|█████▊    | 6387/11074 [54:42<38:41,  2.02it/s]
 58%|█████▊    | 6388/11074 [54:42<38:38,  2.02it/s]
 58%|█████▊    | 6389/11074 [54:43<38:37,  2.02it/s]
 58%|█████▊    | 6390/11074 [54:43<38:34,  2.02it/s]
 58%|█████▊    | 6391/11074 [54:44<38:34,  2.02it/s]
 58%|█████▊    | 6392/11074 [54:44<38:34,  2.02it/s]
 58%|█████▊    | 6393/11074 [54:45<38:32,  2.02it/s]
 58%|█████▊    | 6394/11074 [54:45<38:33,  2.02it/s]
 58%|█████▊    | 6395/11074 [54:46<38:31,  2.02it/s]
 58%|█████▊    | 6396/11074 [54:46<38:32,  2.02it/s]
 58%|█████▊    | 6397/11074 [54:47<38:30,  2.02it/s]
 58%|█████▊    | 6398/11074 [54:47<41:56,  1.86it/s]
 58%|█████▊    | 6399/11074 [54:48<40:56,  1.90it/s]
 58%|█████▊    | 6400/11074 [54:48<40:13,  1.94it/s]
                                                    
{'loss': 3.3061, 'grad_norm': 0.22684818506240845, 'learning_rate': 0.00045137378213880487, 'epoch': 8.09}
+
 58%|█████▊    | 6400/11074 [54:48<40:13,  1.94it/s]
 58%|█████▊    | 6401/11074 [54:49<39:45,  1.96it/s]
 58%|█████▊    | 6402/11074 [54:49<39:22,  1.98it/s]
 58%|█████▊    | 6403/11074 [54:50<39:06,  1.99it/s]
 58%|█████▊    | 6404/11074 [54:50<38:55,  2.00it/s]
 58%|��████▊    | 6405/11074 [54:51<38:48,  2.01it/s]
 58%|█████▊    | 6406/11074 [54:51<38:41,  2.01it/s]
 58%|█████▊    | 6407/11074 [54:52<38:36,  2.01it/s]
 58%|█████▊    | 6408/11074 [54:52<38:32,  2.02it/s]
 58%|█████▊    | 6409/11074 [54:53<38:33,  2.02it/s]
 58%|█████▊    | 6410/11074 [54:53<38:33,  2.02it/s]
 58%|█████▊    | 6411/11074 [54:54<38:28,  2.02it/s]
 58%|█████▊    | 6412/11074 [54:54<38:29,  2.02it/s]
 58%|█████▊    | 6413/11074 [54:55<38:26,  2.02it/s]
 58%|█████▊    | 6414/11074 [54:55<38:27,  2.02it/s]
 58%|█████▊    | 6415/11074 [54:56<38:25,  2.02it/s]
 58%|█████▊    | 6416/11074 [54:56<38:24,  2.02it/s]
 58%|█████▊    | 6417/11074 [54:57<38:23,  2.02it/s]
 58%|█████▊    | 6418/11074 [54:57<38:22,  2.02it/s]
 58%|█████▊    | 6419/11074 [54:58<38:22,  2.02it/s]
 58%|█████▊    | 6420/11074 [54:58<38:22,  2.02it/s]
 58%|█████▊    | 6421/11074 [54:59<38:22,  2.02it/s]
 58%|█████▊    | 6422/11074 [54:59<38:20,  2.02it/s]
 58%|█████▊    | 6423/11074 [55:00<38:18,  2.02it/s]
 58%|█████▊    | 6424/11074 [55:00<38:15,  2.03it/s]
 58%|█████▊    | 6425/11074 [55:01<38:17,  2.02it/s]{'loss': 3.3176, 'grad_norm': 0.22717179358005524, 'learning_rate': 0.00044745362303080354, 'epoch': 8.12}
+
                                                    

 58%|█████▊    | 6425/11074 [55:01<38:17,  2.02it/s]
 58%|█████▊    | 6426/11074 [55:01<38:22,  2.02it/s]
 58%|█████▊    | 6427/11074 [55:02<38:21,  2.02it/s]
 58%|█████▊    | 6428/11074 [55:02<38:19,  2.02it/s]
 58%|█████▊    | 6429/11074 [55:03<38:16,  2.02it/s]
 58%|█████▊    | 6430/11074 [55:03<38:13,  2.03it/s]
 58%|█████▊    | 6431/11074 [55:04<38:14,  2.02it/s]
 58%|█████▊    | 6432/11074 [55:04<38:12,  2.02it/s]
 58%|█████▊    | 6433/11074 [55:05<38:11,  2.03it/s]
 58%|█████▊    | 6434/11074 [55:05<38:11,  2.02it/s]
 58%|█████▊    | 6435/11074 [55:06<38:10,  2.03it/s]
 58%|█████▊    | 6436/11074 [55:06<38:11,  2.02it/s]
 58%|█████▊    | 6437/11074 [55:07<38:08,  2.03it/s]
 58%|█████▊    | 6438/11074 [55:07<38:10,  2.02it/s]
 58%|█████▊    | 6439/11074 [55:08<38:09,  2.02it/s]
 58%|█████▊    | 6440/11074 [55:08<38:07,  2.03it/s]
 58%|█████▊    | 6441/11074 [55:09<38:07,  2.03it/s]
 58%|█████▊    | 6442/11074 [55:09<38:09,  2.02it/s]
 58%|█████▊    | 6443/11074 [55:10<38:09,  2.02it/s]
 58%|█████▊    | 6444/11074 [55:10<38:10,  2.02it/s]
 58%|█████▊    | 6445/11074 [55:11<38:09,  2.02it/s]
 58%|█████▊    | 6446/11074 [55:11<38:08,  2.02it/s]
 58%|█████▊    | 6447/11074 [55:12<38:06,  2.02it/s]
 58%|█████▊    | 6448/11074 [55:12<38:07,  2.02it/s]
 58%|█████▊    | 6449/11074 [55:13<38:06,  2.02it/s]
 58%|█████▊    | 6450/11074 [55:13<38:06,  2.02it/s]{'loss': 3.3259, 'grad_norm': 0.22799476981163025, 'learning_rate': 0.00044353672738454953, 'epoch': 8.15}
+
                                                    

 58%|█████▊    | 6450/11074 [55:13<38:06,  2.02it/s]
 58%|█████▊    | 6451/11074 [55:14<38:06,  2.02it/s]
 58%|█████▊    | 6452/11074 [55:14<38:07,  2.02it/s]
 58%|█████▊    | 6453/11074 [55:15<38:06,  2.02it/s]
 58%|█████▊    | 6454/11074 [55:15<38:03,  2.02it/s]
 58%|█████▊    | 6455/11074 [55:15<38:04,  2.02it/s]
 58%|█████▊    | 6456/11074 [55:16<38:03,  2.02it/s]
 58%|█████▊    | 6457/11074 [55:16<38:03,  2.02it/s]
 58%|█████▊    | 6458/11074 [55:17<38:02,  2.02it/s]
 58%|█████▊    | 6459/11074 [55:17<37:59,  2.02it/s]
 58%|█████▊    | 6460/11074 [55:18<38:00,  2.02it/s]
 58%|█████▊    | 6461/11074 [55:18<37:58,  2.02it/s]
 58%|█████▊    | 6462/11074 [55:19<37:59,  2.02it/s]
 58%|█████▊    | 6463/11074 [55:19<37:59,  2.02it/s]
 58%|█████▊    | 6464/11074 [55:20<37:57,  2.02it/s]
 58%|█████▊    | 6465/11074 [55:20<37:57,  2.02it/s]
 58%|█████▊    | 6466/11074 [55:21<37:56,  2.02it/s]
 58%|█████▊    | 6467/11074 [55:21<37:54,  2.03it/s]
 58%|█████▊    | 6468/11074 [55:22<37:56,  2.02it/s]
 58%|█████▊    | 6469/11074 [55:22<37:53,  2.03it/s]
 58%|█████▊    | 6470/11074 [55:23<37:56,  2.02it/s]
 58%|█████▊    | 6471/11074 [55:23<37:53,  2.02it/s]
 58%|█████▊    | 6472/11074 [55:24<37:50,  2.03it/s]
 58%|█████▊    | 6473/11074 [55:24<37:52,  2.02it/s]
 58%|█████▊    | 6474/11074 [55:25<37:52,  2.02it/s]
 58%|█████▊    | 6475/11074 [55:25<37:53,  2.02it/s]{'loss': 3.3267, 'grad_norm': 0.22307640314102173, 'learning_rate': 0.0004396233384639907, 'epoch': 8.18}
+
                                                    

 58%|█████▊    | 6475/11074 [55:25<37:53,  2.02it/s]
 58%|█████▊    | 6476/11074 [55:26<37:52,  2.02it/s]
 58%|█████▊    | 6477/11074 [55:26<37:53,  2.02it/s]
 58%|█████▊    | 6478/11074 [55:27<37:51,  2.02it/s]
 59%|█████▊    | 6479/11074 [55:27<37:52,  2.02it/s]
 59%|█████▊    | 6480/11074 [55:28<37:53,  2.02it/s]
 59%|█████▊    | 6481/11074 [55:28<37:51,  2.02it/s]
 59%|█████▊    | 6482/11074 [55:29<37:49,  2.02it/s]
 59%|█████▊    | 6483/11074 [55:29<37:48,  2.02it/s]
 59%|█████▊    | 6484/11074 [55:30<37:47,  2.02it/s]
 59%|█████▊    | 6485/11074 [55:30<37:48,  2.02it/s]
 59%|█████▊    | 6486/11074 [55:31<37:47,  2.02it/s]
 59%|█████▊    | 6487/11074 [55:31<37:45,  2.02it/s]
 59%|█████▊    | 6488/11074 [55:32<37:46,  2.02it/s]
 59%|█████▊    | 6489/11074 [55:32<37:47,  2.02it/s]
 59%|█████▊    | 6490/11074 [55:33<37:44,  2.02it/s]
 59%|█████▊    | 6491/11074 [55:33<37:44,  2.02it/s]
 59%|█████▊    | 6492/11074 [55:34<37:40,  2.03it/s]
 59%|█████▊    | 6493/11074 [55:34<37:42,  2.03it/s]
 59%|█████▊    | 6494/11074 [55:35<37:40,  2.03it/s]
 59%|█████▊    | 6495/11074 [55:35<37:42,  2.02it/s]
 59%|█████▊    | 6496/11074 [55:36<37:41,  2.02it/s]
 59%|█████▊    | 6497/11074 [55:36<37:40,  2.03it/s]
 59%|█████▊    | 6498/11074 [55:37<37:38,  2.03it/s]
 59%|█████▊    | 6499/11074 [55:37<37:36,  2.03it/s]
 59%|█████▊    | 6500/11074 [55:38<37:38,  2.03it/s]{'loss': 3.3379, 'grad_norm': 0.2257000356912613, 'learning_rate': 0.0004357136993152854, 'epoch': 8.21}
                                                    
+
 59%|█████▊    | 6500/11074 [55:38<37:38,  2.03it/s]
 59%|█████▊    | 6501/11074 [55:38<37:39,  2.02it/s]
 59%|█████▊    | 6502/11074 [55:39<37:39,  2.02it/s]
 59%|█████▊    | 6503/11074 [55:39<37:40,  2.02it/s]
 59%|█████▊    | 6504/11074 [55:40<37:40,  2.02it/s]
 59%|█████▊    | 6505/11074 [55:40<37:38,  2.02it/s]
 59%|█████▉    | 6506/11074 [55:41<37:38,  2.02it/s]
 59%|█████▉    | 6507/11074 [55:41<37:34,  2.03it/s]
 59%|█████▉    | 6508/11074 [55:42<37:36,  2.02it/s]
 59%|█████▉    | 6509/11074 [55:42<37:33,  2.03it/s]
 59%|█████▉    | 6510/11074 [55:43<37:36,  2.02it/s]
 59%|█████▉    | 6511/11074 [55:43<37:34,  2.02it/s]
 59%|█████▉    | 6512/11074 [55:44<37:33,  2.02it/s]
 59%|█████▉    | 6513/11074 [55:44<37:32,  2.02it/s]
 59%|█████▉    | 6514/11074 [55:45<37:32,  2.02it/s]
 59%|█████▉    | 6515/11074 [55:45<37:31,  2.02it/s]
 59%|█████▉    | 6516/11074 [55:46<37:30,  2.03it/s]
 59%|█████▉    | 6517/11074 [55:46<37:33,  2.02it/s]
 59%|█████▉    | 6518/11074 [55:47<37:30,  2.02it/s]
 59%|█████▉    | 6519/11074 [55:47<37:30,  2.02it/s]
 59%|█████▉    | 6520/11074 [55:48<37:28,  2.03it/s]
 59%|█████▉    | 6521/11074 [55:48<37:30,  2.02it/s]
 59%|█████▉    | 6522/11074 [55:49<37:27,  2.03it/s]
 59%|█████▉    | 6523/11074 [55:49<37:26,  2.03it/s]
 59%|█████▉    | 6524/11074 [55:50<37:25,  2.03it/s]
 59%|█████▉    | 6525/11074 [55:50<37:24,  2.03it/s]{'loss': 3.3345, 'grad_norm': 0.22320950031280518, 'learning_rate': 0.0004318080527517071, 'epoch': 8.24}
+
                                                    

 59%|█████▉    | 6525/11074 [55:50<37:24,  2.03it/s]
 59%|█████▉    | 6526/11074 [55:51<37:29,  2.02it/s]
 59%|█████▉    | 6527/11074 [55:51<37:25,  2.02it/s]
 59%|█████▉    | 6528/11074 [55:52<37:27,  2.02it/s]
 59%|█████▉    | 6529/11074 [55:52<37:24,  2.03it/s]
 59%|█████▉    | 6530/11074 [55:53<37:25,  2.02it/s]
 59%|█████▉    | 6531/11074 [55:53<37:25,  2.02it/s]
 59%|█████▉    | 6532/11074 [55:54<37:25,  2.02it/s]
 59%|█████▉    | 6533/11074 [55:54<37:26,  2.02it/s]
 59%|█████▉    | 6534/11074 [55:55<37:26,  2.02it/s]
 59%|█████▉    | 6535/11074 [55:55<37:26,  2.02it/s]
 59%|█████▉    | 6536/11074 [55:56<37:23,  2.02it/s]
 59%|█████▉    | 6537/11074 [55:56<37:22,  2.02it/s]
 59%|█████▉    | 6538/11074 [55:57<37:25,  2.02it/s]
 59%|█████▉    | 6539/11074 [55:57<37:24,  2.02it/s]
 59%|█████▉    | 6540/11074 [55:57<37:22,  2.02it/s]
 59%|█████▉    | 6541/11074 [55:58<37:23,  2.02it/s]
 59%|█████▉    | 6542/11074 [55:58<37:23,  2.02it/s]
 59%|█████▉    | 6543/11074 [55:59<37:21,  2.02it/s]
 59%|█████▉    | 6544/11074 [55:59<37:20,  2.02it/s]
 59%|███��█▉    | 6545/11074 [56:00<37:16,  2.02it/s]
 59%|█████▉    | 6546/11074 [56:00<37:18,  2.02it/s]
 59%|█████▉    | 6547/11074 [56:01<37:18,  2.02it/s]
 59%|█████▉    | 6548/11074 [56:01<37:19,  2.02it/s]
 59%|█████▉    | 6549/11074 [56:02<37:16,  2.02it/s]
 59%|█████▉    | 6550/11074 [56:02<37:20,  2.02it/s]{'loss': 3.3404, 'grad_norm': 0.22427597641944885, 'learning_rate': 0.0004279066413385646, 'epoch': 8.28}
                                                    
+
 59%|█████▉    | 6550/11074 [56:02<37:20,  2.02it/s]
 59%|█████▉    | 6551/11074 [56:03<37:19,  2.02it/s]
 59%|█████▉    | 6552/11074 [56:03<37:19,  2.02it/s]
 59%|█████▉    | 6553/11074 [56:04<37:15,  2.02it/s]
 59%|█████▉    | 6554/11074 [56:04<37:16,  2.02it/s]
 59%|█████▉    | 6555/11074 [56:05<37:12,  2.02it/s]
 59%|█████▉    | 6556/11074 [56:05<37:13,  2.02it/s]
 59%|█████▉    | 6557/11074 [56:06<37:11,  2.02it/s]
 59%|█████▉    | 6558/11074 [56:06<37:12,  2.02it/s]
 59%|█████▉    | 6559/11074 [56:07<37:12,  2.02it/s]
 59%|█████▉    | 6560/11074 [56:07<37:11,  2.02it/s]
 59%|█████▉    | 6561/11074 [56:08<37:11,  2.02it/s]
 59%|█████▉    | 6562/11074 [56:08<37:13,  2.02it/s]
 59%|█████▉    | 6563/11074 [56:09<37:11,  2.02it/s]
 59%|█████▉    | 6564/11074 [56:09<37:11,  2.02it/s]
 59%|█████▉    | 6565/11074 [56:10<37:10,  2.02it/s]
 59%|█████▉    | 6566/11074 [56:10<37:12,  2.02it/s]
 59%|█████▉    | 6567/11074 [56:11<37:08,  2.02it/s]
 59%|█████▉    | 6568/11074 [56:11<37:05,  2.02it/s]
 59%|█████▉    | 6569/11074 [56:12<37:06,  2.02it/s]
 59%|█████▉    | 6570/11074 [56:12<37:04,  2.02it/s]
 59%|█████▉    | 6571/11074 [56:13<37:06,  2.02it/s]
 59%|█████▉    | 6572/11074 [56:13<37:03,  2.03it/s]
 59%|█████▉    | 6573/11074 [56:14<37:05,  2.02it/s]
 59%|█████▉    | 6574/11074 [56:14<37:05,  2.02it/s]
 59%|█████▉    | 6575/11074 [56:15<37:05,  2.02it/s]
                                                    {'loss': 3.3407, 'grad_norm': 0.2344975769519806, 'learning_rate': 0.0004240097073781374, 'epoch': 8.31}
+
 59%|█████▉    | 6575/11074 [56:15<37:05,  2.02it/s]
 59%|█████▉    | 6576/11074 [56:15<37:05,  2.02it/s]
 59%|█████▉    | 6577/11074 [56:16<37:05,  2.02it/s]
 59%|█████▉    | 6578/11074 [56:16<37:04,  2.02it/s]
 59%|█████▉    | 6579/11074 [56:17<37:04,  2.02it/s]
 59%|█████▉    | 6580/11074 [56:17<37:02,  2.02it/s]
 59%|█████▉    | 6581/11074 [56:18<37:04,  2.02it/s]
 59%|█████▉    | 6582/11074 [56:18<37:02,  2.02it/s]
 59%|█████▉    | 6583/11074 [56:19<37:03,  2.02it/s]
 59%|█████▉    | 6584/11074 [56:19<37:01,  2.02it/s]
 59%|█████▉    | 6585/11074 [56:20<37:02,  2.02it/s]
 59%|█████▉    | 6586/11074 [56:20<37:00,  2.02it/s]
 59%|█████▉    | 6587/11074 [56:21<36:59,  2.02it/s]
 59%|█████▉    | 6588/11074 [56:21<36:58,  2.02it/s]
 59%|█████▉    | 6589/11074 [56:22<36:55,  2.02it/s]
 60%|█████▉    | 6590/11074 [56:22<36:58,  2.02it/s]
 60%|█████▉    | 6591/11074 [56:23<36:57,  2.02it/s]
 60%|█████▉    | 6592/11074 [56:23<36:59,  2.02it/s]
 60%|█████▉    | 6593/11074 [56:24<36:55,  2.02it/s]
 60%|█████▉    | 6594/11074 [56:24<36:56,  2.02it/s]
 60%|█████▉    | 6595/11074 [56:25<36:54,  2.02it/s]
 60%|█████▉    | 6596/11074 [56:25<36:55,  2.02it/s]
 60%|█████▉    | 6597/11074 [56:26<36:56,  2.02it/s]
 60%|█████▉    | 6598/11074 [56:26<36:54,  2.02it/s]
 60%|█████▉    | 6599/11074 [56:27<36:55,  2.02it/s]
 60%|█████▉    | 6600/11074 [56:27<36:52,  2.02it/s]{'loss': 3.3498, 'grad_norm': 0.22194364666938782, 'learning_rate': 0.00042011749289462644, 'epoch': 8.34}
+
                                                    

 60%|█████▉    | 6600/11074 [56:27<36:52,  2.02it/s]
 60%|█████▉    | 6601/11074 [56:28<36:53,  2.02it/s]
 60%|█████▉    | 6602/11074 [56:28<36:53,  2.02it/s]
 60%|█████▉    | 6603/11074 [56:29<36:51,  2.02it/s]
 60%|█████▉    | 6604/11074 [56:29<36:51,  2.02it/s]
 60%|█████▉    | 6605/11074 [56:30<36:47,  2.02it/s]
 60%|█████▉    | 6606/11074 [56:30<36:47,  2.02it/s]
 60%|█████▉    | 6607/11074 [56:31<36:45,  2.03it/s]
 60%|█████▉    | 6608/11074 [56:31<36:46,  2.02it/s]
 60%|█████▉    | 6609/11074 [56:32<36:48,  2.02it/s]
 60%|█████▉    | 6610/11074 [56:32<36:46,  2.02it/s]
 60%|█████▉    | 6611/11074 [56:33<36:46,  2.02it/s]
 60%|█████▉    | 6612/11074 [56:33<36:46,  2.02it/s]
 60%|█████▉    | 6613/11074 [56:34<36:46,  2.02it/s]
 60%|█████▉    | 6614/11074 [56:34<36:45,  2.02it/s]
 60%|█████▉    | 6615/11074 [56:35<36:47,  2.02it/s]
 60%|█████▉    | 6616/11074 [56:35<36:44,  2.02it/s]
 60%|█████▉    | 6617/11074 [56:36<36:44,  2.02it/s]
 60%|█████▉    | 6618/11074 [56:36<36:44,  2.02it/s]
 60%|█████▉    | 6619/11074 [56:37<36:43,  2.02it/s]
 60%|█████▉    | 6620/11074 [56:37<36:43,  2.02it/s]
 60%|█████▉    | 6621/11074 [56:38<36:43,  2.02it/s]
 60%|█████▉    | 6622/11074 [56:38<36:41,  2.02it/s]
 60%|█████▉    | 6623/11074 [56:39<36:42,  2.02it/s]
 60%|█████▉    | 6624/11074 [56:39<36:41,  2.02it/s]
 60%|█████▉    | 6625/11074 [56:40<36:41,  2.02it/s]{'loss': 3.342, 'grad_norm': 0.23584336042404175, 'learning_rate': 0.00041623023961912366, 'epoch': 8.37}
+
                                                    

 60%|█████▉    | 6625/11074 [56:40<36:41,  2.02it/s]
 60%|█████▉    | 6626/11074 [56:40<36:42,  2.02it/s]
 60%|█████▉    | 6627/11074 [56:41<36:40,  2.02it/s]
 60%|█████▉    | 6628/11074 [56:41<36:36,  2.02it/s]
 60%|█████▉    | 6629/11074 [56:42<36:37,  2.02it/s]
 60%|█████▉    | 6630/11074 [56:42<36:35,  2.02it/s]
 60%|█████▉    | 6631/11074 [56:43<36:36,  2.02it/s]
 60%|█████▉    | 6632/11074 [56:43<36:35,  2.02it/s]
 60%|█████▉    | 6633/11074 [56:43<36:32,  2.03it/s]
 60%|█████▉    | 6634/11074 [56:44<36:33,  2.02it/s]
 60%|█████▉    | 6635/11074 [56:44<36:32,  2.02it/s]
 60%|█████▉    | 6636/11074 [56:45<36:33,  2.02it/s]
 60%|█████▉    | 6637/11074 [56:45<36:30,  2.03it/s]
 60%|█████▉    | 6638/11074 [56:46<36:32,  2.02it/s]
 60%|█████▉    | 6639/11074 [56:46<36:30,  2.02it/s]
 60%|█████▉    | 6640/11074 [56:47<36:31,  2.02it/s]
 60%|█████▉    | 6641/11074 [56:47<36:28,  2.03it/s]
 60%|█████▉    | 6642/11074 [56:48<36:29,  2.02it/s]
 60%|█████▉    | 6643/11074 [56:48<36:28,  2.02it/s]
 60%|█████▉    | 6644/11074 [56:49<36:27,  2.03it/s]
 60%|██████    | 6645/11074 [56:49<36:29,  2.02it/s]
 60%|██████    | 6646/11074 [56:50<36:27,  2.02it/s]
 60%|██████    | 6647/11074 [56:50<36:27,  2.02it/s]
 60%|██████    | 6648/11074 [56:51<36:25,  2.03it/s]
 60%|██████    | 6649/11074 [56:51<36:27,  2.02it/s]
 60%|██████    | 6650/11074 [56:52<36:23,  2.03it/s]{'loss': 3.3565, 'grad_norm': 0.23263058066368103, 'learning_rate': 0.0004123481889745987, 'epoch': 8.4}
+
                                                    

 60%|██████    | 6650/11074 [56:52<36:23,  2.03it/s]
 60%|██████    | 6651/11074 [56:52<36:26,  2.02it/s]
 60%|██████    | 6652/11074 [56:53<36:25,  2.02it/s]
 60%|██████    | 6653/11074 [56:53<36:26,  2.02it/s]
 60%|██████    | 6654/11074 [56:54<36:25,  2.02it/s]
 60%|██████    | 6655/11074 [56:54<36:28,  2.02it/s]
 60%|██████    | 6656/11074 [56:55<36:25,  2.02it/s]
 60%|██████    | 6657/11074 [56:55<36:25,  2.02it/s]
 60%|██████    | 6658/11074 [56:56<36:23,  2.02it/s]
 60%|██████    | 6659/11074 [56:56<36:23,  2.02it/s]
 60%|██████    | 6660/11074 [56:57<36:22,  2.02it/s]
 60%|██████    | 6661/11074 [56:57<36:23,  2.02it/s]
 60%|██████    | 6662/11074 [56:58<36:20,  2.02it/s]
 60%|██████    | 6663/11074 [56:58<36:19,  2.02it/s]
 60%|██████    | 6664/11074 [56:59<36:20,  2.02it/s]
 60%|██████    | 6665/11074 [56:59<36:19,  2.02it/s]
 60%|██████    | 6666/11074 [57:00<36:19,  2.02it/s]
 60%|██████    | 6667/11074 [57:00<36:18,  2.02it/s]
 60%|██████    | 6668/11074 [57:01<36:19,  2.02it/s]
 60%|██████    | 6669/11074 [57:01<36:18,  2.02it/s]
 60%|██████    | 6670/11074 [57:02<36:19,  2.02it/s]
 60%|██████    | 6671/11074 [57:02<36:16,  2.02it/s]
 60%|██████    | 6672/11074 [57:03<36:17,  2.02it/s]
 60%|██████    | 6673/11074 [57:03<36:15,  2.02it/s]
 60%|██████    | 6674/11074 [57:04<36:14,  2.02it/s]
 60%|██████    | 6675/11074 [57:04<36:14,  2.02it/s]
                                                    
{'loss': 3.3523, 'grad_norm': 0.24159778654575348, 'learning_rate': 0.00040847158206090494, 'epoch': 8.43}
+
 60%|██████    | 6675/11074 [57:04<36:14,  2.02it/s]
 60%|██████    | 6676/11074 [57:05<36:18,  2.02it/s]
 60%|██████    | 6677/11074 [57:05<36:17,  2.02it/s]
 60%|██████    | 6678/11074 [57:06<36:17,  2.02it/s]
 60%|██████    | 6679/11074 [57:06<36:17,  2.02it/s]
 60%|██████    | 6680/11074 [57:07<36:15,  2.02it/s]
 60%|██████    | 6681/11074 [57:07<36:13,  2.02it/s]
 60%|██████    | 6682/11074 [57:08<36:13,  2.02it/s]
 60%|██████    | 6683/11074 [57:08<36:13,  2.02it/s]
 60%|██████    | 6684/11074 [57:09<36:11,  2.02it/s]
 60%|██████    | 6685/11074 [57:09<36:11,  2.02it/s]
 60%|██████    | 6686/11074 [57:10<36:10,  2.02it/s]
 60%|██████    | 6687/11074 [57:10<36:13,  2.02it/s]
 60%|██████    | 6688/11074 [57:11<36:09,  2.02it/s]
 60%|██████    | 6689/11074 [57:11<36:12,  2.02it/s]
 60%|██████    | 6690/11074 [57:12<36:08,  2.02it/s]
 60%|██████    | 6691/11074 [57:12<36:08,  2.02it/s]
 60%|██████    | 6692/11074 [57:13<36:04,  2.02it/s]
 60%|██████    | 6693/11074 [57:13<36:06,  2.02it/s]
 60%|██████    | 6694/11074 [57:14<36:02,  2.03it/s]
 60%|██████    | 6695/11074 [57:14<36:03,  2.02it/s]
 60%|██████    | 6696/11074 [57:15<36:02,  2.02it/s]
 60%|██████    | 6697/11074 [57:15<36:03,  2.02it/s]
 60%|██████    | 6698/11074 [57:16<36:03,  2.02it/s]
 60%|██████    | 6699/11074 [57:16<36:04,  2.02it/s]
 61%|██████    | 6700/11074 [57:17<36:03,  2.02it/s]{'loss': 3.3539, 'grad_norm': 0.22557418048381805, 'learning_rate': 0.00040460065963980553, 'epoch': 8.46}
+
                                                    

 61%|██████    | 6700/11074 [57:17<36:03,  2.02it/s]
 61%|██████    | 6701/11074 [57:17<36:06,  2.02it/s]
 61%|██████    | 6702/11074 [57:18<36:04,  2.02it/s]
 61%|██████    | 6703/11074 [57:18<36:05,  2.02it/s]
 61%|██████    | 6704/11074 [57:19<36:03,  2.02it/s]
 61%|██████    | 6705/11074 [57:19<36:02,  2.02it/s]
 61%|██████    | 6706/11074 [57:20<36:01,  2.02it/s]
 61%|██████    | 6707/11074 [57:20<36:02,  2.02it/s]
 61%|██████    | 6708/11074 [57:21<36:00,  2.02it/s]
 61%|██████    | 6709/11074 [57:21<35:59,  2.02it/s]
 61%|██████    | 6710/11074 [57:22<35:59,  2.02it/s]
 61%|██████    | 6711/11074 [57:22<35:59,  2.02it/s]
 61%|██████    | 6712/11074 [57:23<35:56,  2.02it/s]
 61%|██████    | 6713/11074 [57:23<35:57,  2.02it/s]
 61%|██████    | 6714/11074 [57:24<35:57,  2.02it/s]
 61%|██████    | 6715/11074 [57:24<35:58,  2.02it/s]
 61%|██████    | 6716/11074 [57:25<35:55,  2.02it/s]
 61%|██████    | 6717/11074 [57:25<35:54,  2.02it/s]
 61%|██████    | 6718/11074 [57:26<35:53,  2.02it/s]
 61%|██████    | 6719/11074 [57:26<35:54,  2.02it/s]
 61%|██████    | 6720/11074 [57:27<35:51,  2.02it/s]
 61%|██████    | 6721/11074 [57:27<35:53,  2.02it/s]
 61%|██████    | 6722/11074 [57:28<35:50,  2.02it/s]
 61%|██████    | 6723/11074 [57:28<35:49,  2.02it/s]
 61%|██████    | 6724/11074 [57:28<35:49,  2.02it/s]
 61%|██████    | 6725/11074 [57:29<35:50,  2.02it/s]{'loss': 3.3676, 'grad_norm': 0.22192180156707764, 'learning_rate': 0.00040073566212002075, 'epoch': 8.5}
+
                                                    

 61%|██████    | 6725/11074 [57:29<35:50,  2.02it/s]
 61%|██████    | 6726/11074 [57:29<35:51,  2.02it/s]
 61%|██████    | 6727/11074 [57:30<35:49,  2.02it/s]
 61%|██████    | 6728/11074 [57:30<35:50,  2.02it/s]
 61%|██████    | 6729/11074 [57:31<35:49,  2.02it/s]
 61%|██████    | 6730/11074 [57:31<35:46,  2.02it/s]
 61%|██████    | 6731/11074 [57:32<35:47,  2.02it/s]
 61%|██████    | 6732/11074 [57:32<35:45,  2.02it/s]
 61%|██████    | 6733/11074 [57:33<35:46,  2.02it/s]
 61%|██████    | 6734/11074 [57:33<35:44,  2.02it/s]
 61%|██████    | 6735/11074 [57:34<35:47,  2.02it/s]
 61%|██████    | 6736/11074 [57:34<35:45,  2.02it/s]
 61%|██████    | 6737/11074 [57:35<35:46,  2.02it/s]
 61%|██████    | 6738/11074 [57:35<35:46,  2.02it/s]
 61%|██████    | 6739/11074 [57:36<35:44,  2.02it/s]
 61%|██████    | 6740/11074 [57:36<35:45,  2.02it/s]
 61%|██████    | 6741/11074 [57:37<35:44,  2.02it/s]
 61%|██████    | 6742/11074 [57:37<35:43,  2.02it/s]
 61%|██████    | 6743/11074 [57:38<35:41,  2.02it/s]
 61%|██████    | 6744/11074 [57:38<35:42,  2.02it/s]
 61%|██████    | 6745/11074 [57:39<35:41,  2.02it/s]
 61%|██████    | 6746/11074 [57:39<35:39,  2.02it/s]
 61%|██████    | 6747/11074 [57:40<35:39,  2.02it/s]
 61%|██████    | 6748/11074 [57:40<35:36,  2.02it/s]
 61%|██████    | 6749/11074 [57:41<35:37,  2.02it/s]
 61%|██████    | 6750/11074 [57:41<35:35,  2.02it/s]{'loss': 3.3597, 'grad_norm': 0.2214174121618271, 'learning_rate': 0.00039687682954229743, 'epoch': 8.53}
+
                                                    

 61%|███��██    | 6750/11074 [57:41<35:35,  2.02it/s]
 61%|██████    | 6751/11074 [57:42<35:38,  2.02it/s]
 61%|██████    | 6752/11074 [57:42<35:36,  2.02it/s]
 61%|██████    | 6753/11074 [57:43<35:38,  2.02it/s]
 61%|██████    | 6754/11074 [57:43<35:36,  2.02it/s]
 61%|██████    | 6755/11074 [57:44<35:37,  2.02it/s]
 61%|██████    | 6756/11074 [57:44<35:34,  2.02it/s]
 61%|██████    | 6757/11074 [57:45<35:36,  2.02it/s]
 61%|██████    | 6758/11074 [57:45<35:32,  2.02it/s]
 61%|██████    | 6759/11074 [57:46<35:35,  2.02it/s]
 61%|██████    | 6760/11074 [57:46<35:35,  2.02it/s]
 61%|██████    | 6761/11074 [57:47<35:36,  2.02it/s]
 61%|██████    | 6762/11074 [57:47<35:33,  2.02it/s]
 61%|██████    | 6763/11074 [57:48<35:34,  2.02it/s]
 61%|██████    | 6764/11074 [57:48<35:32,  2.02it/s]
 61%|██████    | 6765/11074 [57:49<35:33,  2.02it/s]
 61%|██████    | 6766/11074 [57:49<35:32,  2.02it/s]
 61%|██████    | 6767/11074 [57:50<35:31,  2.02it/s]
 61%|██████    | 6768/11074 [57:50<35:32,  2.02it/s]
 61%|██████    | 6769/11074 [57:51<35:32,  2.02it/s]
 61%|██████    | 6770/11074 [57:51<35:30,  2.02it/s]
 61%|██████    | 6771/11074 [57:52<35:29,  2.02it/s]
 61%|██████    | 6772/11074 [57:52<35:29,  2.02it/s]
 61%|██████    | 6773/11074 [57:53<35:27,  2.02it/s]
 61%|██████    | 6774/11074 [57:53<35:27,  2.02it/s]
 61%|██████    | 6775/11074 [57:54<35:27,  2.02it/s]
                                                    
{'loss': 3.3656, 'grad_norm': 0.22992229461669922, 'learning_rate': 0.00039302440156450036, 'epoch': 8.56}
+
 61%|██████    | 6775/11074 [57:54<35:27,  2.02it/s]
 61%|██████    | 6776/11074 [57:54<35:28,  2.02it/s]
 61%|██████    | 6777/11074 [57:55<35:28,  2.02it/s]
 61%|██████    | 6778/11074 [57:55<35:25,  2.02it/s]
 61%|██████    | 6779/11074 [57:56<35:25,  2.02it/s]
 61%|██████    | 6780/11074 [57:56<35:24,  2.02it/s]
 61%|██████    | 6781/11074 [57:57<35:23,  2.02it/s]
 61%|██████    | 6782/11074 [57:57<35:20,  2.02it/s]
 61%|██████▏   | 6783/11074 [57:58<35:21,  2.02it/s]
 61%|██████▏   | 6784/11074 [57:58<35:22,  2.02it/s]
 61%|██████▏   | 6785/11074 [57:59<35:22,  2.02it/s]
 61%|██████▏   | 6786/11074 [57:59<35:20,  2.02it/s]
 61%|██████▏   | 6787/11074 [58:00<35:20,  2.02it/s]
 61%|██████▏   | 6788/11074 [58:00<35:18,  2.02it/s]
 61%|██████▏   | 6789/11074 [58:01<35:20,  2.02it/s]
 61%|██████▏   | 6790/11074 [58:01<35:17,  2.02it/s]
 61%|██████▏   | 6791/11074 [58:02<35:17,  2.02it/s]
 61%|██████▏   | 6792/11074 [58:02<35:16,  2.02it/s]
 61%|██████▏   | 6793/11074 [58:03<35:17,  2.02it/s]
 61%|██████▏   | 6794/11074 [58:03<35:16,  2.02it/s]
 61%|██████▏   | 6795/11074 [58:04<35:17,  2.02it/s]
 61%|██████▏   | 6796/11074 [58:04<35:15,  2.02it/s]
 61%|██████▏   | 6797/11074 [58:05<35:14,  2.02it/s]
 61%|██████▏   | 6798/11074 [58:05<35:14,  2.02it/s]
 61%|██████▏   | 6799/11074 [58:06<35:12,  2.02it/s]
 61%|██████▏   | 6800/11074 [58:06<35:12,  2.02it/s]
                                                    
{'loss': 3.3653, 'grad_norm': 0.22503212094306946, 'learning_rate': 0.00038917861744672803, 'epoch': 8.59}
+
 61%|██████▏   | 6800/11074 [58:06<35:12,  2.02it/s]
 61%|██████▏   | 6801/11074 [58:07<35:15,  2.02it/s]
 61%|██████▏   | 6802/11074 [58:07<35:15,  2.02it/s]
 61%|██████▏   | 6803/11074 [58:08<35:13,  2.02it/s]
 61%|██████▏   | 6804/11074 [58:08<35:12,  2.02it/s]
 61%|██████▏   | 6805/11074 [58:09<35:09,  2.02it/s]
 61%|██████▏   | 6806/11074 [58:09<35:11,  2.02it/s]
 61%|██████▏   | 6807/11074 [58:10<35:08,  2.02it/s]
 61%|██████▏   | 6808/11074 [58:10<35:10,  2.02it/s]
 61%|██████▏   | 6809/11074 [58:11<35:07,  2.02it/s]
 61%|██████▏   | 6810/11074 [58:11<35:09,  2.02it/s]
 62%|██████▏   | 6811/11074 [58:12<35:06,  2.02it/s]
 62%|██████▏   | 6812/11074 [58:12<35:05,  2.02it/s]
 62%|██████▏   | 6813/11074 [58:13<35:05,  2.02it/s]
 62%|██████▏   | 6814/11074 [58:13<35:05,  2.02it/s]
 62%|██████▏   | 6815/11074 [58:14<35:05,  2.02it/s]
 62%|██████▏   | 6816/11074 [58:14<35:05,  2.02it/s]
 62%|██████▏   | 6817/11074 [58:15<35:06,  2.02it/s]
 62%|██████▏   | 6818/11074 [58:15<35:05,  2.02it/s]
 62%|██████▏   | 6819/11074 [58:15<35:07,  2.02it/s]
 62%|██████▏   | 6820/11074 [58:16<35:04,  2.02it/s]
 62%|██████▏   | 6821/11074 [58:16<35:04,  2.02it/s]
 62%|██████▏   | 6822/11074 [58:17<35:03,  2.02it/s]
 62%|██████▏   | 6823/11074 [58:17<35:03,  2.02it/s]
 62%|██████▏   | 6824/11074 [58:18<35:02,  2.02it/s]
 62%|██████▏   | 6825/11074 [58:18<35:02,  2.02it/s]{'loss': 3.368, 'grad_norm': 0.22349414229393005, 'learning_rate': 0.00038533971603645366, 'epoch': 8.62}
+
                                                    

 62%|██████▏   | 6825/11074 [58:18<35:02,  2.02it/s]
 62%|██████▏   | 6826/11074 [58:19<35:03,  2.02it/s]
 62%|██████▏   | 6827/11074 [58:19<35:04,  2.02it/s]
 62%|██████▏   | 6828/11074 [58:20<35:01,  2.02it/s]
 62%|██████▏   | 6829/11074 [58:20<35:01,  2.02it/s]
 62%|██████▏   | 6830/11074 [58:21<34:58,  2.02it/s]
 62%|██████▏   | 6831/11074 [58:21<34:59,  2.02it/s]
 62%|██████▏   | 6832/11074 [58:22<34:58,  2.02it/s]
 62%|██████▏   | 6833/11074 [58:22<34:56,  2.02it/s]
 62%|██████▏   | 6834/11074 [58:23<34:56,  2.02it/s]
 62%|██████▏   | 6835/11074 [58:23<34:56,  2.02it/s]
 62%|██████▏   | 6836/11074 [58:24<34:57,  2.02it/s]
 62%|██████▏   | 6837/11074 [58:24<34:55,  2.02it/s]
 62%|██████▏   | 6838/11074 [58:25<34:56,  2.02it/s]
 62%|██████▏   | 6839/11074 [58:25<34:55,  2.02it/s]
 62%|██████▏   | 6840/11074 [58:26<34:55,  2.02it/s]
 62%|██████▏   | 6841/11074 [58:26<34:54,  2.02it/s]
 62%|██████▏   | 6842/11074 [58:27<34:53,  2.02it/s]
 62%|██████▏   | 6843/11074 [58:27<34:55,  2.02it/s]
 62%|██████▏   | 6844/11074 [58:28<34:53,  2.02it/s]
 62%|██████▏   | 6845/11074 [58:28<34:52,  2.02it/s]
 62%|██████▏   | 6846/11074 [58:29<34:50,  2.02it/s]
 62%|██████▏   | 6847/11074 [58:29<34:49,  2.02it/s]
 62%|██████▏   | 6848/11074 [58:30<34:47,  2.02it/s]
 62%|██████▏   | 6849/11074 [58:30<34:46,  2.02it/s]
 62%|██████▏   | 6850/11074 [58:31<34:46,  2.02it/s]
                                                    
{'loss': 3.3679, 'grad_norm': 0.22241181135177612, 'learning_rate': 0.00038150793575369063, 'epoch': 8.65}
+
 62%|██████▏   | 6850/11074 [58:31<34:46,  2.02it/s]
 62%|██████▏   | 6851/11074 [58:31<34:49,  2.02it/s]
 62%|██████▏   | 6852/11074 [58:32<34:48,  2.02it/s]
 62%|██████▏   | 6853/11074 [58:32<34:45,  2.02it/s]
 62%|██████▏   | 6854/11074 [58:33<34:45,  2.02it/s]
 62%|██████▏   | 6855/11074 [58:33<34:44,  2.02it/s]
 62%|██████▏   | 6856/11074 [58:34<34:46,  2.02it/s]
 62%|██████▏   | 6857/11074 [58:34<34:43,  2.02it/s]
 62%|██████▏   | 6858/11074 [58:35<34:44,  2.02it/s]
 62%|██████▏   | 6859/11074 [58:35<34:41,  2.03it/s]
 62%|██████▏   | 6860/11074 [58:36<34:41,  2.02it/s]
 62%|██████▏   | 6861/11074 [58:36<34:40,  2.02it/s]
 62%|██████▏   | 6862/11074 [58:37<34:39,  2.03it/s]
 62%|██████▏   | 6863/11074 [58:37<34:40,  2.02it/s]
 62%|██████▏   | 6864/11074 [58:38<34:41,  2.02it/s]
 62%|██████▏   | 6865/11074 [58:38<34:41,  2.02it/s]
 62%|██████▏   | 6866/11074 [58:39<34:44,  2.02it/s]
 62%|██████▏   | 6867/11074 [58:39<34:44,  2.02it/s]
 62%|██████▏   | 6868/11074 [58:40<34:43,  2.02it/s]
 62%|██████▏   | 6869/11074 [58:40<34:42,  2.02it/s]
 62%|██████▏   | 6870/11074 [58:41<34:41,  2.02it/s]
 62%|██████▏   | 6871/11074 [58:41<34:37,  2.02it/s]
 62%|██████▏   | 6872/11074 [58:42<34:38,  2.02it/s]
 62%|██████▏   | 6873/11074 [58:42<34:35,  2.02it/s]
 62%|██████▏   | 6874/11074 [58:43<34:36,  2.02it/s]
 62%|██████▏   | 6875/11074 [58:43<34:34,  2.02it/s]
                                                    {'loss': 3.3735, 'grad_norm': 0.23530688881874084, 'learning_rate': 0.0003776835145761854, 'epoch': 8.69}
+
 62%|██████▏   | 6875/11074 [58:43<34:34,  2.02it/s]
 62%|██████▏   | 6876/11074 [58:44<34:36,  2.02it/s]
 62%|██████▏   | 6877/11074 [58:44<34:35,  2.02it/s]
 62%|██████▏   | 6878/11074 [58:45<34:35,  2.02it/s]
 62%|██████▏   | 6879/11074 [58:45<34:35,  2.02it/s]
 62%|██████▏   | 6880/11074 [58:46<34:34,  2.02it/s]
 62%|██████▏   | 6881/11074 [58:46<34:35,  2.02it/s]
 62%|██████▏   | 6882/11074 [58:47<34:34,  2.02it/s]
 62%|██████▏   | 6883/11074 [58:47<34:33,  2.02it/s]
 62%|██████▏   | 6884/11074 [58:48<34:31,  2.02it/s]
 62%|██████▏   | 6885/11074 [58:48<34:32,  2.02it/s]
 62%|██████▏   | 6886/11074 [58:49<34:29,  2.02it/s]
 62%|██████▏   | 6887/11074 [58:49<34:30,  2.02it/s]
 62%|██████▏   | 6888/11074 [58:50<34:28,  2.02it/s]
 62%|██████▏   | 6889/11074 [58:50<34:28,  2.02it/s]
 62%|██████▏   | 6890/11074 [58:51<34:28,  2.02it/s]
 62%|██████▏   | 6891/11074 [58:51<34:27,  2.02it/s]
 62%|██████▏   | 6892/11074 [58:52<34:26,  2.02it/s]
 62%|██████▏   | 6893/11074 [58:52<34:25,  2.02it/s]
 62%|██████▏   | 6894/11074 [58:53<34:25,  2.02it/s]
 62%|██████▏   | 6895/11074 [58:53<34:25,  2.02it/s]
 62%|██████▏   | 6896/11074 [58:54<34:24,  2.02it/s]
 62%|██████▏   | 6897/11074 [58:54<34:24,  2.02it/s]
 62%|██████▏   | 6898/11074 [58:55<34:24,  2.02it/s]
 62%|██████▏   | 6899/11074 [58:55<34:20,  2.03it/s]
 62%|██████▏   | 6900/11074 [58:56<34:21,  2.02it/s]{'loss': 3.3663, 'grad_norm': 0.23676970601081848, 'learning_rate': 0.0003738666900246377, 'epoch': 8.72}
+
                                                    

 62%|██████▏   | 6900/11074 [58:56<34:21,  2.02it/s]
 62%|██████▏   | 6901/11074 [58:56<34:21,  2.02it/s]
 62%|██████▏   | 6902/11074 [58:57<34:23,  2.02it/s]
 62%|██████▏   | 6903/11074 [58:57<34:21,  2.02it/s]
 62%|██████▏   | 6904/11074 [58:58<34:21,  2.02it/s]
 62%|██████▏   | 6905/11074 [58:58<34:18,  2.03it/s]
 62%|██████▏   | 6906/11074 [58:59<34:20,  2.02it/s]
 62%|██████▏   | 6907/11074 [58:59<34:16,  2.03it/s]
 62%|██████▏   | 6908/11074 [58:59<34:17,  2.03it/s]
 62%|██████▏   | 6909/11074 [59:00<34:16,  2.03it/s]
 62%|██████▏   | 6910/11074 [59:00<34:14,  2.03it/s]
 62%|██████▏   | 6911/11074 [59:01<34:15,  2.03it/s]
 62%|██████▏   | 6912/11074 [59:01<34:15,  2.02it/s]
 62%|██████▏   | 6913/11074 [59:02<34:16,  2.02it/s]
 62%|██████▏   | 6914/11074 [59:02<34:14,  2.02it/s]
 62%|██████▏   | 6915/11074 [59:03<34:16,  2.02it/s]
 62%|██████▏   | 6916/11074 [59:03<34:13,  2.02it/s]
 62%|██████▏   | 6917/11074 [59:04<34:16,  2.02it/s]
 62%|██████▏   | 6918/11074 [59:04<34:12,  2.02it/s]
 62%|██████▏   | 6919/11074 [59:05<34:14,  2.02it/s]
 62%|██████▏   | 6920/11074 [59:05<34:11,  2.02it/s]
 62%|██████▏   | 6921/11074 [59:06<34:11,  2.02it/s]
 63%|██████▎   | 6922/11074 [59:06<34:10,  2.03it/s]
 63%|██████▎   | 6923/11074 [59:07<34:10,  2.02it/s]
 63%|██████▎   | 6924/11074 [59:07<34:09,  2.02it/s]
 63%|██████▎   | 6925/11074 [59:08<34:08,  2.03it/s]{'loss': 3.3664, 'grad_norm': 0.22845487296581268, 'learning_rate': 0.0003700576991479486, 'epoch': 8.75}
+
                                                    

 63%|██████▎   | 6925/11074 [59:08<34:08,  2.03it/s]
 63%|██████▎   | 6926/11074 [59:08<34:12,  2.02it/s]
 63%|██████▎   | 6927/11074 [59:09<34:08,  2.02it/s]
 63%|██████▎   | 6928/11074 [59:09<34:09,  2.02it/s]
 63%|██████▎   | 6929/11074 [59:10<34:05,  2.03it/s]
 63%|██████▎   | 6930/11074 [59:10<34:05,  2.03it/s]
 63%|██████▎   | 6931/11074 [59:11<34:06,  2.02it/s]
 63%|██████▎   | 6932/11074 [59:11<34:05,  2.03it/s]
 63%|██████▎   | 6933/11074 [59:12<34:06,  2.02it/s]
 63%|██████▎   | 6934/11074 [59:12<34:08,  2.02it/s]
 63%|██████▎   | 6935/11074 [59:13<34:07,  2.02it/s]
 63%|██████▎   | 6936/11074 [59:13<34:07,  2.02it/s]
 63%|██████▎   | 6937/11074 [59:14<34:06,  2.02it/s]
 63%|██████▎   | 6938/11074 [59:14<34:04,  2.02it/s]
 63%|██████▎   | 6939/11074 [59:15<34:03,  2.02it/s]
 63%|██████▎   | 6940/11074 [59:15<34:01,  2.02it/s]
 63%|██████▎   | 6941/11074 [59:16<34:02,  2.02it/s]
 63%|██████▎   | 6942/11074 [59:16<33:59,  2.03it/s]
 63%|██████▎   | 6943/11074 [59:17<34:00,  2.02it/s]
 63%|██████▎   | 6944/11074 [59:17<33:59,  2.03it/s]
 63%|██████▎   | 6945/11074 [59:18<33:59,  2.02it/s]
 63%|██████▎   | 6946/11074 [59:18<34:00,  2.02it/s]
 63%|██████▎   | 6947/11074 [59:19<33:58,  2.02it/s]
 63%|██████▎   | 6948/11074 [59:19<33:58,  2.02it/s]
 63%|██████▎   | 6949/11074 [59:20<33:55,  2.03it/s]
 63%|██████▎   | 6950/11074 [59:20<33:57,  2.02it/s]{'loss': 3.3747, 'grad_norm': 0.227634996175766, 'learning_rate': 0.00036625677850849874, 'epoch': 8.78}
+
                                                    

 63%|██████▎   | 6950/11074 [59:20<33:57,  2.02it/s]
 63%|██████▎   | 6951/11074 [59:21<34:05,  2.02it/s]
 63%|██████▎   | 6952/11074 [59:21<34:02,  2.02it/s]
 63%|██████▎   | 6953/11074 [59:22<33:59,  2.02it/s]
 63%|██████▎   | 6954/11074 [59:22<33:58,  2.02it/s]
 63%|██████▎   | 6955/11074 [59:23<33:57,  2.02it/s]
 63%|██████▎   | 6956/11074 [59:23<33:57,  2.02it/s]
 63%|██████▎   | 6957/11074 [59:24<33:54,  2.02it/s]
 63%|██████▎   | 6958/11074 [59:24<33:53,  2.02it/s]
 63%|██████▎   | 6959/11074 [59:25<33:52,  2.02it/s]
 63%|██████▎   | 6960/11074 [59:25<33:55,  2.02it/s]
 63%|██████▎   | 6961/11074 [59:26<33:52,  2.02it/s]
 63%|██████▎   | 6962/11074 [59:26<33:52,  2.02it/s]
 63%|██████▎   | 6963/11074 [59:27<33:51,  2.02it/s]
 63%|██████▎   | 6964/11074 [59:27<33:51,  2.02it/s]
 63%|██████▎   | 6965/11074 [59:28<33:50,  2.02it/s]
 63%|██████▎   | 6966/11074 [59:28<33:50,  2.02it/s]
 63%|██████▎   | 6967/11074 [59:29<33:50,  2.02it/s]
 63%|██████▎   | 6968/11074 [59:29<33:50,  2.02it/s]
 63%|██████▎   | 6969/11074 [59:30<33:49,  2.02it/s]
 63%|██████▎   | 6970/11074 [59:30<33:49,  2.02it/s]
 63%|██████▎   | 6971/11074 [59:31<33:49,  2.02it/s]
 63%|██████▎   | 6972/11074 [59:31<33:49,  2.02it/s]
 63%|██████▎   | 6973/11074 [59:32<33:49,  2.02it/s]
 63%|██████▎   | 6974/11074 [59:32<36:43,  1.86it/s]
 63%|██████▎   | 6975/11074 [59:33<35:51,  1.91it/s]
                                                    
{'loss': 3.3679, 'grad_norm': 0.22947275638580322, 'learning_rate': 0.00036246416416745573, 'epoch': 8.81}
+
 63%|██████▎   | 6975/11074 [59:33<35:51,  1.91it/s]
 63%|██████▎   | 6976/11074 [59:33<35:15,  1.94it/s]
 63%|██████▎   | 6977/11074 [59:34<34:48,  1.96it/s]
 63%|██████▎   | 6978/11074 [59:34<34:30,  1.98it/s]
 63%|██████▎   | 6979/11074 [59:35<34:15,  1.99it/s]
 63%|██████▎   | 6980/11074 [59:35<34:03,  2.00it/s]
 63%|██████▎   | 6981/11074 [59:36<33:59,  2.01it/s]
 63%|██████▎   | 6982/11074 [59:36<33:52,  2.01it/s]
 63%|██████▎   | 6983/11074 [59:37<33:50,  2.01it/s]
 63%|██████▎   | 6984/11074 [59:37<33:45,  2.02it/s]
 63%|██████▎   | 6985/11074 [59:38<33:43,  2.02it/s]
 63%|██████▎   | 6986/11074 [59:38<33:41,  2.02it/s]
 63%|██████▎   | 6987/11074 [59:39<33:39,  2.02it/s]
 63%|██████▎   | 6988/11074 [59:39<33:39,  2.02it/s]
 63%|██████▎   | 6989/11074 [59:40<33:37,  2.02it/s]
 63%|██████▎   | 6990/11074 [59:40<33:38,  2.02it/s]
 63%|██████▎   | 6991/11074 [59:41<33:35,  2.03it/s]
 63%|██████▎   | 6992/11074 [59:41<33:36,  2.02it/s]
 63%|██████▎   | 6993/11074 [59:42<33:35,  2.02it/s]
 63%|██████▎   | 6994/11074 [59:42<33:34,  2.02it/s]
 63%|██████▎   | 6995/11074 [59:43<33:34,  2.02it/s]
 63%|██████▎   | 6996/11074 [59:43<33:33,  2.03it/s]
 63%|██████▎   | 6997/11074 [59:44<33:34,  2.02it/s]
 63%|██████▎   | 6998/11074 [59:44<33:32,  2.02it/s]
 63%|██████▎   | 6999/11074 [59:45<33:33,  2.02it/s]
 63%|██████▎   | 7000/11074 [59:45<33:31,  2.03it/s]
                                                    
{'loss': 3.3742, 'grad_norm': 0.22470520436763763, 'learning_rate': 0.00035868009167011383, 'epoch': 8.84}
+
 63%|██████▎   | 7000/11074 [59:45<33:31,  2.03it/s]
 63%|██████▎   | 7001/11074 [59:46<33:35,  2.02it/s]
 63%|██████▎   | 7002/11074 [59:46<33:33,  2.02it/s]
 63%|██████▎   | 7003/11074 [59:47<33:34,  2.02it/s]
 63%|██████▎   | 7004/11074 [59:47<33:33,  2.02it/s]
 63%|██████▎   | 7005/11074 [59:48<33:32,  2.02it/s]
 63%|██████▎   | 7006/11074 [59:48<33:31,  2.02it/s]
 63%|██████▎   | 7007/11074 [59:49<33:32,  2.02it/s]
 63%|██████▎   | 7008/11074 [59:49<33:30,  2.02it/s]
 63%|██████▎   | 7009/11074 [59:50<33:30,  2.02it/s]
 63%|██████▎   | 7010/11074 [59:50<33:27,  2.02it/s]
 63%|██████▎   | 7011/11074 [59:51<33:28,  2.02it/s]
 63%|██████▎   | 7012/11074 [59:51<33:25,  2.03it/s]
 63%|██████▎   | 7013/11074 [59:52<33:25,  2.02it/s]
 63%|██████▎   | 7014/11074 [59:52<33:24,  2.02it/s]
 63%|██████▎   | 7015/11074 [59:53<33:22,  2.03it/s]
 63%|██████▎   | 7016/11074 [59:53<33:24,  2.02it/s]
 63%|██████▎   | 7017/11074 [59:54<33:22,  2.03it/s]
 63%|██████▎   | 7018/11074 [59:54<33:22,  2.03it/s]
 63%|██████▎   | 7019/11074 [59:55<33:22,  2.02it/s]
 63%|██████▎   | 7020/11074 [59:55<33:24,  2.02it/s]
 63%|██████▎   | 7021/11074 [59:55<33:21,  2.03it/s]
 63%|██████▎   | 7022/11074 [59:56<33:22,  2.02it/s]
 63%|██████▎   | 7023/11074 [59:56<33:21,  2.02it/s]
 63%|██████▎   | 7024/11074 [59:57<33:21,  2.02it/s]
 63%|██████▎   | 7025/11074 [59:57<33:20,  2.02it/s]
                                                    
{'loss': 3.3731, 'grad_norm': 0.22383926808834076, 'learning_rate': 0.00035490479603126495, 'epoch': 8.88}
+
 63%|██████▎   | 7025/11074 [59:57<33:20,  2.02it/s]
 63%|██████▎   | 7026/11074 [59:58<33:22,  2.02it/s]
 63%|██████▎   | 7027/11074 [59:58<33:20,  2.02it/s]
 63%|██████▎   | 7028/11074 [59:59<33:20,  2.02it/s]
 63%|██████▎   | 7029/11074 [59:59<33:20,  2.02it/s]
 63%|██████▎   | 7030/11074 [1:00:00<33:21,  2.02it/s]
 63%|██████▎   | 7031/11074 [1:00:00<33:21,  2.02it/s]
 64%|██████▎   | 7032/11074 [1:00:01<33:20,  2.02it/s]
 64%|██████▎   | 7033/11074 [1:00:01<33:19,  2.02it/s]
 64%|██████▎   | 7034/11074 [1:00:02<33:16,  2.02it/s]
 64%|██████▎   | 7035/11074 [1:00:02<33:17,  2.02it/s]
 64%|██████▎   | 7036/11074 [1:00:03<33:16,  2.02it/s]
 64%|██████▎   | 7037/11074 [1:00:03<33:15,  2.02it/s]
 64%|██████▎   | 7038/11074 [1:00:04<33:15,  2.02it/s]
 64%|██████▎   | 7039/11074 [1:00:04<33:16,  2.02it/s]
 64%|██████▎   | 7040/11074 [1:00:05<33:13,  2.02it/s]
 64%|██████▎   | 7041/11074 [1:00:06<36:08,  1.86it/s]
 64%|██████▎   | 7042/11074 [1:00:06<35:14,  1.91it/s]
 64%|██████▎   | 7043/11074 [1:00:07<34:35,  1.94it/s]
 64%|██████▎   | 7044/11074 [1:00:07<34:10,  1.97it/s]
 64%|██████▎   | 7045/11074 [1:00:07<33:53,  1.98it/s]
 64%|██████▎   | 7046/11074 [1:00:08<33:39,  1.99it/s]
 64%|██████▎   | 7047/11074 [1:00:08<33:30,  2.00it/s]
 64%|██████▎   | 7048/11074 [1:00:09<33:23,  2.01it/s]
 64%|██████▎   | 7049/11074 [1:00:09<33:17,  2.01it/s]
 64%|██████▎   | 7050/11074 [1:00:10<33:14,  2.02it/s]{'loss': 3.3742, 'grad_norm': 0.22696553170681, 'learning_rate': 0.00035113851172060266, 'epoch': 8.91}
+
                                                      

 64%|██████▎   | 7050/11074 [1:00:10<33:14,  2.02it/s]
 64%|██████▎   | 7051/11074 [1:00:10<33:15,  2.02it/s]
 64%|██████▎   | 7052/11074 [1:00:11<33:12,  2.02it/s]
 64%|██████▎   | 7053/11074 [1:00:11<33:09,  2.02it/s]
 64%|██████▎   | 7054/11074 [1:00:12<33:09,  2.02it/s]
 64%|██████▎   | 7055/11074 [1:00:12<33:08,  2.02it/s]
 64%|██████▎   | 7056/11074 [1:00:13<33:07,  2.02it/s]
 64%|██████▎   | 7057/11074 [1:00:13<33:06,  2.02it/s]
 64%|██████▎   | 7058/11074 [1:00:14<33:05,  2.02it/s]
 64%|██████▎   | 7059/11074 [1:00:14<33:02,  2.02it/s]
 64%|██████▍   | 7060/11074 [1:00:15<33:03,  2.02it/s]
 64%|██████▍   | 7061/11074 [1:00:15<33:01,  2.02it/s]
 64%|██████▍   | 7062/11074 [1:00:16<33:01,  2.02it/s]
 64%|██████▍   | 7063/11074 [1:00:16<33:01,  2.02it/s]
 64%|██████▍   | 7064/11074 [1:00:17<33:01,  2.02it/s]
 64%|██████▍   | 7065/11074 [1:00:17<33:01,  2.02it/s]
 64%|██████▍   | 7066/11074 [1:00:18<33:01,  2.02it/s]
 64%|██████▍   | 7067/11074 [1:00:18<33:01,  2.02it/s]
 64%|██████▍   | 7068/11074 [1:00:19<33:01,  2.02it/s]
 64%|██████▍   | 7069/11074 [1:00:19<32:59,  2.02it/s]
 64%|██████▍   | 7070/11074 [1:00:20<32:59,  2.02it/s]
 64%|██████▍   | 7071/11074 [1:00:20<32:59,  2.02it/s]
 64%|██████▍   | 7072/11074 [1:00:21<33:00,  2.02it/s]
 64%|██████▍   | 7073/11074 [1:00:21<32:59,  2.02it/s]
 64%|██████▍   | 7074/11074 [1:00:22<32:58,  2.02it/s]
 64%|██████▍   | 7075/11074 [1:00:22<32:57,  2.02it/s]
                                                      {'loss': 3.3709, 'grad_norm': 0.22354719042778015, 'learning_rate': 0.0003473814726481599, 'epoch': 8.94}
+

 64%|██████▍   | 7075/11074 [1:00:22<32:57,  2.02it/s]
 64%|██████▍   | 7076/11074 [1:00:23<33:00,  2.02it/s]
 64%|██████▍   | 7077/11074 [1:00:23<32:57,  2.02it/s]
 64%|██████▍   | 7078/11074 [1:00:24<32:57,  2.02it/s]
 64%|██████▍   | 7079/11074 [1:00:24<32:56,  2.02it/s]
 64%|██████▍   | 7080/11074 [1:00:25<32:55,  2.02it/s]
 64%|██████▍   | 7081/11074 [1:00:25<32:55,  2.02it/s]
 64%|██████▍   | 7082/11074 [1:00:26<32:54,  2.02it/s]
 64%|██████▍   | 7083/11074 [1:00:26<32:52,  2.02it/s]
 64%|██████▍   | 7084/11074 [1:00:27<32:51,  2.02it/s]
 64%|██████▍   | 7085/11074 [1:00:27<32:51,  2.02it/s]
 64%|██████▍   | 7086/11074 [1:00:28<32:49,  2.03it/s]
 64%|███��██▍   | 7087/11074 [1:00:28<32:50,  2.02it/s]
 64%|██████▍   | 7088/11074 [1:00:29<32:46,  2.03it/s]
 64%|██████▍   | 7089/11074 [1:00:29<32:48,  2.02it/s]
 64%|██████▍   | 7090/11074 [1:00:30<32:49,  2.02it/s]
 64%|██████▍   | 7091/11074 [1:00:30<32:49,  2.02it/s]
 64%|██████▍   | 7092/11074 [1:00:31<32:48,  2.02it/s]
 64%|██████▍   | 7093/11074 [1:00:31<32:49,  2.02it/s]
 64%|██████▍   | 7094/11074 [1:00:32<32:46,  2.02it/s]
 64%|██████▍   | 7095/11074 [1:00:32<32:47,  2.02it/s]
 64%|██████▍   | 7096/11074 [1:00:33<32:46,  2.02it/s]
 64%|██████▍   | 7097/11074 [1:00:33<32:46,  2.02it/s]
 64%|██████▍   | 7098/11074 [1:00:34<32:46,  2.02it/s]
 64%|██████▍   | 7099/11074 [1:00:34<32:46,  2.02it/s]
 64%|██████▍   | 7100/11074 [1:00:35<32:46,  2.02it/s]{'loss': 3.3739, 'grad_norm': 0.22911973297595978, 'learning_rate': 0.0003436339121497822, 'epoch': 8.97}
+
                                                      

 64%|██████▍   | 7100/11074 [1:00:35<32:46,  2.02it/s]
 64%|██████▍   | 7101/11074 [1:00:35<32:48,  2.02it/s]
 64%|██████▍   | 7102/11074 [1:00:36<32:46,  2.02it/s]
 64%|██████▍   | 7103/11074 [1:00:36<32:44,  2.02it/s]
 64%|██████▍   | 7104/11074 [1:00:37<32:43,  2.02it/s]
 64%|██████▍   | 7105/11074 [1:00:37<32:45,  2.02it/s]
 64%|██████▍   | 7106/11074 [1:00:38<32:41,  2.02it/s]
 64%|██████▍   | 7107/11074 [1:00:38<32:42,  2.02it/s]
 64%|██████▍   | 7108/11074 [1:00:39<32:41,  2.02it/s]
 64%|██████▍   | 7109/11074 [1:00:39<32:40,  2.02it/s]
 64%|██████▍   | 7110/11074 [1:00:40<32:38,  2.02it/s]
 64%|██████▍   | 7111/11074 [1:00:40<32:35,  2.03it/s]
 64%|██████▍   | 7112/11074 [1:00:41<32:37,  2.02it/s]
 64%|██████▍   | 7113/11074 [1:00:41<32:33,  2.03it/s]
 64%|██████▍   | 7114/11074 [1:00:42<32:34,  2.03it/s]
 64%|██████▍   | 7115/11074 [1:00:42<32:34,  2.03it/s]
 64%|██████▍   | 7116/11074 [1:00:43<32:34,  2.02it/s]
 64%|██████▍   | 7117/11074 [1:00:43<32:33,  2.03it/s]
 64%|██████▍   | 7118/11074 [1:00:44<32:34,  2.02it/s]
 64%|██████▍   | 7119/11074 [1:00:44<32:33,  2.02it/s]
 64%|██████▍   | 7120/11074 [1:00:45<32:31,  2.03it/s]
 64%|██████▍   | 7121/11074 [1:00:45<32:33,  2.02it/s]
 64%|██████▍   | 7122/11074 [1:00:46<32:32,  2.02it/s]
 64%|██████▍   | 7123/11074 [1:00:46<33:21,  1.97it/s]
 64%|██████▍   | 7124/11074 [1:00:58<4:20:29,  3.96s/it]
 64%|██████▍   | 7125/11074 [1:00:59<3:12:13,  2.92s/it]{'loss': 3.371, 'grad_norm': 0.2374652773141861, 'learning_rate': 0.00033989606297263575, 'epoch': 9.0}
                                                        
+
 64%|██████▍   | 7125/11074 [1:00:59<3:12:13,  2.92s/it]
 64%|██████▍   | 7126/11074 [1:00:59<2:24:22,  2.19s/it]
 64%|██████▍   | 7127/11074 [1:01:00<1:50:48,  1.68s/it]
 64%|██████▍   | 7128/11074 [1:01:00<1:27:20,  1.33s/it]
 64%|██████▍   | 7129/11074 [1:01:01<1:10:52,  1.08s/it]
 64%|██████▍   | 7130/11074 [1:01:01<59:22,  1.11it/s]  
 64%|██████▍   | 7131/11074 [1:01:02<51:19,  1.28it/s]
 64%|██████▍   | 7132/11074 [1:01:02<45:40,  1.44it/s]
 64%|██████▍   | 7133/11074 [1:01:03<41:40,  1.58it/s]
 64%|██████▍   | 7134/11074 [1:01:03<38:58,  1.68it/s]
 64%|██████▍   | 7135/11074 [1:01:04<37:05,  1.77it/s]
 64%|██████▍   | 7136/11074 [1:01:04<35:40,  1.84it/s]
 64%|██████▍   | 7137/11074 [1:01:05<34:39,  1.89it/s]
 64%|██████▍   | 7138/11074 [1:01:05<34:00,  1.93it/s]
 64%|██████▍   | 7139/11074 [1:01:06<33:29,  1.96it/s]
 64%|██████▍   | 7140/11074 [1:01:06<33:11,  1.98it/s]
 64%|██████▍   | 7141/11074 [1:01:07<32:59,  1.99it/s]
 64%|██████▍   | 7142/11074 [1:01:07<32:48,  2.00it/s]
 65%|██████▍   | 7143/11074 [1:01:08<32:48,  2.00it/s]
 65%|██████▍   | 7144/11074 [1:01:08<32:42,  2.00it/s]
 65%|██████▍   | 7145/11074 [1:01:09<32:39,  2.01it/s]
 65%|██████▍   | 7146/11074 [1:01:09<32:32,  2.01it/s]
 65%|██████▍   | 7147/11074 [1:01:10<32:30,  2.01it/s]
 65%|██████▍   | 7148/11074 [1:01:10<32:27,  2.02it/s]
 65%|██████▍   | 7149/11074 [1:01:11<32:27,  2.02it/s]
 65%|██████▍   | 7150/11074 [1:01:11<32:24,  2.02it/s]{'loss': 3.238, 'grad_norm': 0.23618361353874207, 'learning_rate': 0.00033616815726075246, 'epoch': 9.03}
+
                                                      

 65%|██████▍   | 7150/11074 [1:01:11<32:24,  2.02it/s]
 65%|██████▍   | 7151/11074 [1:01:11<32:32,  2.01it/s]
 65%|██████▍   | 7152/11074 [1:01:12<32:28,  2.01it/s]
 65%|██████▍   | 7153/11074 [1:01:12<32:23,  2.02it/s]
 65%|██████▍   | 7154/11074 [1:01:13<32:21,  2.02it/s]
 65%|██████▍   | 7155/11074 [1:01:13<32:18,  2.02it/s]
 65%|██████▍   | 7156/11074 [1:01:14<32:18,  2.02it/s]
 65%|██████▍   | 7157/11074 [1:01:14<32:15,  2.02it/s]
 65%|██████▍   | 7158/11074 [1:01:15<32:15,  2.02it/s]
 65%|██████▍   | 7159/11074 [1:01:15<32:13,  2.03it/s]
 65%|██████▍   | 7160/11074 [1:01:16<32:12,  2.03it/s]
 65%|██████▍   | 7161/11074 [1:01:16<32:11,  2.03it/s]
 65%|██████▍   | 7162/11074 [1:01:17<32:08,  2.03it/s]
 65%|██████▍   | 7163/11074 [1:01:17<32:09,  2.03it/s]
 65%|██████▍   | 7164/11074 [1:01:18<32:08,  2.03it/s]
 65%|██████▍   | 7165/11074 [1:01:18<32:09,  2.03it/s]
 65%|██████▍   | 7166/11074 [1:01:19<32:07,  2.03it/s]
 65%|██████▍   | 7167/11074 [1:01:19<32:08,  2.03it/s]
 65%|██████▍   | 7168/11074 [1:01:20<32:08,  2.03it/s]
 65%|██████▍   | 7169/11074 [1:01:20<32:06,  2.03it/s]
 65%|██████▍   | 7170/11074 [1:01:21<32:06,  2.03it/s]
 65%|██████▍   | 7171/11074 [1:01:21<32:05,  2.03it/s]
 65%|██████▍   | 7172/11074 [1:01:22<32:08,  2.02it/s]
 65%|██████▍   | 7173/11074 [1:01:22<32:06,  2.02it/s]
 65%|██████▍   | 7174/11074 [1:01:23<32:06,  2.02it/s]
 65%|██████▍   | 7175/11074 [1:01:23<32:05,  2.02it/s]
                                                      {'loss': 3.2443, 'grad_norm': 0.22908008098602295, 'learning_rate': 0.000332450426540612, 'epoch': 9.07}
+
 65%|██████▍   | 7175/11074 [1:01:23<32:05,  2.02it/s]
 65%|██████▍   | 7176/11074 [1:01:24<32:07,  2.02it/s]
 65%|██████▍   | 7177/11074 [1:01:24<32:06,  2.02it/s]
 65%|██████▍   | 7178/11074 [1:01:25<32:04,  2.02it/s]
 65%|██████▍   | 7179/11074 [1:01:25<32:04,  2.02it/s]
 65%|██████▍   | 7180/11074 [1:01:26<32:01,  2.03it/s]
 65%|██████▍   | 7181/11074 [1:01:26<32:03,  2.02it/s]
 65%|██████▍   | 7182/11074 [1:01:27<32:00,  2.03it/s]
 65%|██████▍   | 7183/11074 [1:01:27<32:01,  2.02it/s]
 65%|██████▍   | 7184/11074 [1:01:28<31:59,  2.03it/s]
 65%|██████▍   | 7185/11074 [1:01:28<32:00,  2.03it/s]
 65%|██████▍   | 7186/11074 [1:01:29<31:59,  2.03it/s]
 65%|██████▍   | 7187/11074 [1:01:29<31:57,  2.03it/s]
 65%|██████▍   | 7188/11074 [1:01:30<31:59,  2.02it/s]
 65%|██████▍   | 7189/11074 [1:01:30<32:00,  2.02it/s]
 65%|██████▍   | 7190/11074 [1:01:31<32:00,  2.02it/s]
 65%|██████▍   | 7191/11074 [1:01:31<31:57,  2.03it/s]
 65%|██████▍   | 7192/11074 [1:01:32<31:57,  2.02it/s]
 65%|██████▍   | 7193/11074 [1:01:32<31:55,  2.03it/s]
 65%|██████▍   | 7194/11074 [1:01:33<31:55,  2.03it/s]
 65%|██████▍   | 7195/11074 [1:01:33<31:53,  2.03it/s]
 65%|██████▍   | 7196/11074 [1:01:34<31:52,  2.03it/s]
 65%|██████▍   | 7197/11074 [1:01:34<31:53,  2.03it/s]
 65%|██████▍   | 7198/11074 [1:01:35<31:53,  2.03it/s]
 65%|██████▌   | 7199/11074 [1:01:35<31:53,  2.03it/s]
 65%|██████▌   | 7200/11074 [1:01:36<31:52,  2.03it/s]{'loss': 3.2534, 'grad_norm': 0.2305447906255722, 'learning_rate': 0.0003287431017067631, 'epoch': 9.1}
+
                                                      

 65%|██████▌   | 7200/11074 [1:01:36<31:52,  2.03it/s]
 65%|██████▌   | 7201/11074 [1:01:36<31:57,  2.02it/s]
 65%|██████▌   | 7202/11074 [1:01:37<31:54,  2.02it/s]
 65%|██████▌   | 7203/11074 [1:01:37<31:54,  2.02it/s]
 65%|██████▌   | 7204/11074 [1:01:38<31:53,  2.02it/s]
 65%|██████▌   | 7205/11074 [1:01:38<31:54,  2.02it/s]
 65%|██████▌   | 7206/11074 [1:01:39<31:54,  2.02it/s]
 65%|██████▌   | 7207/11074 [1:01:39<31:52,  2.02it/s]
 65%|██████▌   | 7208/11074 [1:01:40<31:50,  2.02it/s]
 65%|██████▌   | 7209/11074 [1:01:40<31:50,  2.02it/s]
 65%|██████▌   | 7210/11074 [1:01:41<31:50,  2.02it/s]
 65%|██████▌   | 7211/11074 [1:01:41<31:50,  2.02it/s]
 65%|██████▌   | 7212/11074 [1:01:42<31:49,  2.02it/s]
 65%|██████▌   | 7213/11074 [1:01:42<31:49,  2.02it/s]
 65%|██████▌   | 7214/11074 [1:01:43<31:49,  2.02it/s]
 65%|██████▌   | 7215/11074 [1:01:43<31:48,  2.02it/s]
 65%|██████▌   | 7216/11074 [1:01:44<31:47,  2.02it/s]
 65%|██████▌   | 7217/11074 [1:01:44<31:47,  2.02it/s]
 65%|██████▌   | 7218/11074 [1:01:45<31:47,  2.02it/s]
 65%|██████▌   | 7219/11074 [1:01:45<31:46,  2.02it/s]
 65%|██████▌   | 7220/11074 [1:01:46<31:45,  2.02it/s]
 65%|██████▌   | 7221/11074 [1:01:46<31:44,  2.02it/s]
 65%|██████▌   | 7222/11074 [1:01:47<31:44,  2.02it/s]
 65%|██████▌   | 7223/11074 [1:01:47<31:44,  2.02it/s]
 65%|██████▌   | 7224/11074 [1:01:48<31:44,  2.02it/s]
 65%|██████▌   | 7225/11074 [1:01:48<31:43,  2.02it/s]{'loss': 3.2564, 'grad_norm': 0.23095867037773132, 'learning_rate': 0.00032504641300748314, 'epoch': 9.13}
+
                                                      

 65%|██████▌   | 7225/11074 [1:01:48<31:43,  2.02it/s]
 65%|██████▌   | 7226/11074 [1:01:49<31:44,  2.02it/s]
 65%|██████▌   | 7227/11074 [1:01:49<31:41,  2.02it/s]
 65%|██████▌   | 7228/11074 [1:01:50<31:41,  2.02it/s]
 65%|██████▌   | 7229/11074 [1:01:50<31:38,  2.03it/s]
 65%|██████▌   | 7230/11074 [1:01:51<31:39,  2.02it/s]
 65%|██████▌   | 7231/11074 [1:01:51<31:36,  2.03it/s]
 65%|██████▌   | 7232/11074 [1:01:52<31:37,  2.02it/s]
 65%|██████▌   | 7233/11074 [1:01:52<31:37,  2.02it/s]
 65%|██████▌   | 7234/11074 [1:01:53<31:36,  2.02it/s]
 65%|██████▌   | 7235/11074 [1:01:53<31:36,  2.02it/s]
 65%|██████▌   | 7236/11074 [1:01:53<31:34,  2.03it/s]
 65%|██████▌   | 7237/11074 [1:01:54<31:34,  2.03it/s]
 65%|██████▌   | 7238/11074 [1:01:54<31:34,  2.02it/s]
 65%|██████▌   | 7239/11074 [1:01:55<31:35,  2.02it/s]
 65%|██████▌   | 7240/11074 [1:01:55<31:33,  2.02it/s]
 65%|██████▌   | 7241/11074 [1:01:56<31:33,  2.02it/s]
 65%|██████▌   | 7242/11074 [1:01:56<31:33,  2.02it/s]
 65%|██████▌   | 7243/11074 [1:01:57<31:33,  2.02it/s]
 65%|██████▌   | 7244/11074 [1:01:57<31:32,  2.02it/s]
 65%|██████▌   | 7245/11074 [1:01:58<31:33,  2.02it/s]
 65%|██████▌   | 7246/11074 [1:01:58<31:32,  2.02it/s]
 65%|██████▌   | 7247/11074 [1:01:59<31:30,  2.02it/s]
 65%|██████▌   | 7248/11074 [1:01:59<31:30,  2.02it/s]
 65%|██████▌   | 7249/11074 [1:02:00<31:27,  2.03it/s]
 65%|██████▌   | 7250/11074 [1:02:00<31:28,  2.02it/s]{'loss': 3.2607, 'grad_norm': 0.23604585230350494, 'learning_rate': 0.00032136059003047833, 'epoch': 9.16}
+
                                                      

 65%|██████▌   | 7250/11074 [1:02:00<31:28,  2.02it/s]
 65%|██████▌   | 7251/11074 [1:02:01<31:30,  2.02it/s]
 65%|██████▌   | 7252/11074 [1:02:01<31:29,  2.02it/s]
 65%|██████▌   | 7253/11074 [1:02:02<31:27,  2.02it/s]
 66%|██████▌   | 7254/11074 [1:02:02<31:24,  2.03it/s]
 66%|██████▌   | 7255/11074 [1:02:03<31:25,  2.03it/s]
 66%|██████▌   | 7256/11074 [1:02:03<31:23,  2.03it/s]
 66%|██████▌   | 7257/11074 [1:02:04<31:23,  2.03it/s]
 66%|██████▌   | 7258/11074 [1:02:04<31:22,  2.03it/s]
 66%|██████▌   | 7259/11074 [1:02:05<31:23,  2.03it/s]
 66%|██████▌   | 7260/11074 [1:02:05<31:28,  2.02it/s]
 66%|██████▌   | 7261/11074 [1:02:06<31:26,  2.02it/s]
 66%|██████▌   | 7262/11074 [1:02:06<31:23,  2.02it/s]
 66%|██████▌   | 7263/11074 [1:02:07<31:22,  2.02it/s]
 66%|██████▌   | 7264/11074 [1:02:07<31:20,  2.03it/s]
 66%|██████▌   | 7265/11074 [1:02:08<31:21,  2.02it/s]
 66%|██████▌   | 7266/11074 [1:02:08<31:20,  2.03it/s]
 66%|██████▌   | 7267/11074 [1:02:09<31:20,  2.02it/s]
 66%|██████▌   | 7268/11074 [1:02:09<31:18,  2.03it/s]
 66%|██████▌   | 7269/11074 [1:02:10<31:18,  2.03it/s]
 66%|██████▌   | 7270/11074 [1:02:10<31:18,  2.03it/s]
 66%|██████▌   | 7271/11074 [1:02:11<31:16,  2.03it/s]
 66%|██████▌   | 7272/11074 [1:02:11<31:17,  2.03it/s]
 66%|██████▌   | 7273/11074 [1:02:12<31:16,  2.03it/s]
 66%|██████▌   | 7274/11074 [1:02:12<31:17,  2.02it/s]
 66%|██████▌   | 7275/11074 [1:02:13<31:15,  2.03it/s]
                                                      
{'loss': 3.2684, 'grad_norm': 0.23639388382434845, 'learning_rate': 0.00031768586168862524, 'epoch': 9.19}
+
 66%|██████▌   | 7275/11074 [1:02:13<31:15,  2.03it/s]
 66%|██████▌   | 7276/11074 [1:02:13<31:18,  2.02it/s]
 66%|██████▌   | 7277/11074 [1:02:14<31:15,  2.02it/s]
 66%|██████▌   | 7278/11074 [1:02:14<31:16,  2.02it/s]
 66%|██████▌   | 7279/11074 [1:02:15<31:13,  2.03it/s]
 66%|██████▌   | 7280/11074 [1:02:15<31:14,  2.02it/s]
 66%|██████▌   | 7281/11074 [1:02:16<31:15,  2.02it/s]
 66%|██████▌   | 7282/11074 [1:02:16<31:16,  2.02it/s]
 66%|██████▌   | 7283/11074 [1:02:17<31:12,  2.02it/s]
 66%|██████▌   | 7284/11074 [1:02:17<31:12,  2.02it/s]
 66%|██████▌   | 7285/11074 [1:02:18<31:12,  2.02it/s]
 66%|██████▌   | 7286/11074 [1:02:18<31:12,  2.02it/s]
 66%|██████▌   | 7287/11074 [1:02:19<31:12,  2.02it/s]
 66%|██████▌   | 7288/11074 [1:02:19<31:11,  2.02it/s]
 66%|██████▌   | 7289/11074 [1:02:20<31:10,  2.02it/s]
 66%|██████▌   | 7290/11074 [1:02:20<31:09,  2.02it/s]
 66%|██████▌   | 7291/11074 [1:02:21<31:08,  2.02it/s]
 66%|██████▌   | 7292/11074 [1:02:21<31:09,  2.02it/s]
 66%|██████▌   | 7293/11074 [1:02:22<31:08,  2.02it/s]
 66%|██████▌   | 7294/11074 [1:02:22<31:09,  2.02it/s]
 66%|██████▌   | 7295/11074 [1:02:23<31:08,  2.02it/s]
 66%|██████▌   | 7296/11074 [1:02:23<31:08,  2.02it/s]
 66%|██████▌   | 7297/11074 [1:02:24<31:08,  2.02it/s]
 66%|██████▌   | 7298/11074 [1:02:24<31:06,  2.02it/s]
 66%|██████▌   | 7299/11074 [1:02:25<31:06,  2.02it/s]
 66%|██████▌   | 7300/11074 [1:02:25<31:05,  2.02it/s]{'loss': 3.2719, 'grad_norm': 0.22805973887443542, 'learning_rate': 0.0003140224562057532, 'epoch': 9.22}
+
                                                      

 66%|██████▌   | 7300/11074 [1:02:25<31:05,  2.02it/s]
 66%|██████▌   | 7301/11074 [1:02:26<31:07,  2.02it/s]
 66%|██████▌   | 7302/11074 [1:02:26<31:06,  2.02it/s]
 66%|██████▌   | 7303/11074 [1:02:27<31:06,  2.02it/s]
 66%|██████▌   | 7304/11074 [1:02:27<31:04,  2.02it/s]
 66%|██████▌   | 7305/11074 [1:02:28<31:03,  2.02it/s]
 66%|██████▌   | 7306/11074 [1:02:28<31:02,  2.02it/s]
 66%|██████▌   | 7307/11074 [1:02:29<31:03,  2.02it/s]
 66%|██████▌   | 7308/11074 [1:02:29<31:03,  2.02it/s]
 66%|██████▌   | 7309/11074 [1:02:30<31:02,  2.02it/s]
 66%|██████▌   | 7310/11074 [1:02:30<31:01,  2.02it/s]
 66%|██████▌   | 7311/11074 [1:02:31<31:02,  2.02it/s]
 66%|██████▌   | 7312/11074 [1:02:31<31:00,  2.02it/s]
 66%|██████▌   | 7313/11074 [1:02:32<30:59,  2.02it/s]
 66%|██████▌   | 7314/11074 [1:02:32<30:58,  2.02it/s]
 66%|██████▌   | 7315/11074 [1:02:33<30:57,  2.02it/s]
 66%|██████▌   | 7316/11074 [1:02:33<30:58,  2.02it/s]
 66%|██████▌   | 7317/11074 [1:02:34<30:57,  2.02it/s]
 66%|██████▌   | 7318/11074 [1:02:34<31:01,  2.02it/s]
 66%|██████▌   | 7319/11074 [1:02:35<30:59,  2.02it/s]
 66%|██████▌   | 7320/11074 [1:02:35<30:58,  2.02it/s]
 66%|██████▌   | 7321/11074 [1:02:36<30:56,  2.02it/s]
 66%|██████▌   | 7322/11074 [1:02:36<30:55,  2.02it/s]
 66%|██████▌   | 7323/11074 [1:02:36<30:54,  2.02it/s]
 66%|██████▌   | 7324/11074 [1:02:37<30:52,  2.02it/s]
 66%|██████▌   | 7325/11074 [1:02:37<30:53,  2.02it/s]{'loss': 3.2819, 'grad_norm': 0.23239585757255554, 'learning_rate': 0.0003103706011024705, 'epoch': 9.25}
+
                                                      

 66%|██████▌   | 7325/11074 [1:02:37<30:53,  2.02it/s]
 66%|██████▌   | 7326/11074 [1:02:38<30:53,  2.02it/s]
 66%|██████▌   | 7327/11074 [1:02:38<30:54,  2.02it/s]
 66%|██████▌   | 7328/11074 [1:02:39<30:50,  2.02it/s]
 66%|██████▌   | 7329/11074 [1:02:39<30:49,  2.02it/s]
 66%|██████▌   | 7330/11074 [1:02:40<30:49,  2.02it/s]
 66%|██████▌   | 7331/11074 [1:02:40<30:49,  2.02it/s]
 66%|██████▌   | 7332/11074 [1:02:41<30:48,  2.02it/s]
 66%|██████▌   | 7333/11074 [1:02:41<30:46,  2.03it/s]
 66%|██████▌   | 7334/11074 [1:02:42<30:48,  2.02it/s]
 66%|██████▌   | 7335/11074 [1:02:42<30:48,  2.02it/s]
 66%|██████▌   | 7336/11074 [1:02:43<30:48,  2.02it/s]
 66%|██████▋   | 7337/11074 [1:02:43<30:47,  2.02it/s]
 66%|██████▋   | 7338/11074 [1:02:44<30:47,  2.02it/s]
 66%|██████▋   | 7339/11074 [1:02:44<30:46,  2.02it/s]
 66%|██████▋   | 7340/11074 [1:02:45<30:47,  2.02it/s]
 66%|██████▋   | 7341/11074 [1:02:45<30:45,  2.02it/s]
 66%|██████▋   | 7342/11074 [1:02:46<30:46,  2.02it/s]
 66%|██████▋   | 7343/11074 [1:02:46<30:46,  2.02it/s]
 66%|██████▋   | 7344/11074 [1:02:47<30:45,  2.02it/s]
 66%|██████▋   | 7345/11074 [1:02:47<30:44,  2.02it/s]
 66%|██████▋   | 7346/11074 [1:02:48<30:42,  2.02it/s]
 66%|██████▋   | 7347/11074 [1:02:48<30:41,  2.02it/s]
 66%|██████▋   | 7348/11074 [1:02:49<30:40,  2.02it/s]
 66%|██████▋   | 7349/11074 [1:02:49<30:41,  2.02it/s]
 66%|██��███▋   | 7350/11074 [1:02:50<30:38,  2.03it/s]{'loss': 3.2731, 'grad_norm': 0.23187832534313202, 'learning_rate': 0.00030673052318203417, 'epoch': 9.29}
+
                                                      

 66%|██████▋   | 7350/11074 [1:02:50<30:38,  2.03it/s]
 66%|██████▋   | 7351/11074 [1:02:50<30:40,  2.02it/s]
 66%|██████▋   | 7352/11074 [1:02:51<30:39,  2.02it/s]
 66%|██████▋   | 7353/11074 [1:02:51<30:41,  2.02it/s]
 66%|██████▋   | 7354/11074 [1:02:52<30:38,  2.02it/s]
 66%|██████▋   | 7355/11074 [1:02:52<30:37,  2.02it/s]
 66%|██████▋   | 7356/11074 [1:02:53<30:36,  2.03it/s]
 66%|██████▋   | 7357/11074 [1:02:53<30:37,  2.02it/s]
 66%|██████▋   | 7358/11074 [1:02:54<30:36,  2.02it/s]
 66%|██████▋   | 7359/11074 [1:02:54<30:36,  2.02it/s]
 66%|██████▋   | 7360/11074 [1:02:55<30:35,  2.02it/s]
 66%|██████▋   | 7361/11074 [1:02:55<30:34,  2.02it/s]
 66%|██████▋   | 7362/11074 [1:02:56<30:32,  2.03it/s]
 66%|██████▋   | 7363/11074 [1:02:56<30:32,  2.03it/s]
 66%|██████▋   | 7364/11074 [1:02:57<30:33,  2.02it/s]
 67%|██████▋   | 7365/11074 [1:02:57<30:34,  2.02it/s]
 67%|██████▋   | 7366/11074 [1:02:58<30:32,  2.02it/s]
 67%|██████▋   | 7367/11074 [1:02:58<30:32,  2.02it/s]
 67%|██████▋   | 7368/11074 [1:02:59<30:32,  2.02it/s]
 67%|██████▋   | 7369/11074 [1:02:59<30:31,  2.02it/s]
 67%|██████▋   | 7370/11074 [1:03:00<30:30,  2.02it/s]
 67%|██████▋   | 7371/11074 [1:03:00<30:28,  2.03it/s]
 67%|██████▋   | 7372/11074 [1:03:01<30:28,  2.02it/s]
 67%|██████▋   | 7373/11074 [1:03:01<30:27,  2.02it/s]
 67%|██████▋   | 7374/11074 [1:03:02<30:26,  2.03it/s]
 67%|██████▋   | 7375/11074 [1:03:02<30:27,  2.02it/s]
                                                      
{'loss': 3.2916, 'grad_norm': 0.23099683225154877, 'learning_rate': 0.0003031024485162637, 'epoch': 9.32}
+
 67%|██████▋   | 7375/11074 [1:03:02<30:27,  2.02it/s]
 67%|██████▋   | 7376/11074 [1:03:03<30:31,  2.02it/s]
 67%|██████▋   | 7377/11074 [1:03:03<30:29,  2.02it/s]
 67%|██████▋   | 7378/11074 [1:03:04<30:29,  2.02it/s]
 67%|██████▋   | 7379/11074 [1:03:04<30:28,  2.02it/s]
 67%|██████▋   | 7380/11074 [1:03:05<30:29,  2.02it/s]
 67%|██████▋   | 7381/11074 [1:03:05<30:26,  2.02it/s]
 67%|██████▋   | 7382/11074 [1:03:06<30:25,  2.02it/s]
 67%|██████▋   | 7383/11074 [1:03:06<30:25,  2.02it/s]
 67%|██████▋   | 7384/11074 [1:03:07<30:24,  2.02it/s]
 67%|██████▋   | 7385/11074 [1:03:07<30:23,  2.02it/s]
 67%|██████▋   | 7386/11074 [1:03:08<30:22,  2.02it/s]
 67%|██████▋   | 7387/11074 [1:03:08<30:21,  2.02it/s]
 67%|██████▋   | 7388/11074 [1:03:09<30:21,  2.02it/s]
 67%|██████▋   | 7389/11074 [1:03:09<30:23,  2.02it/s]
 67%|██████▋   | 7390/11074 [1:03:10<30:22,  2.02it/s]
 67%|██████▋   | 7391/11074 [1:03:10<30:19,  2.02it/s]
 67%|██████▋   | 7392/11074 [1:03:11<30:20,  2.02it/s]
 67%|██████▋   | 7393/11074 [1:03:11<30:16,  2.03it/s]
 67%|██████▋   | 7394/11074 [1:03:12<30:17,  2.02it/s]
 67%|██████▋   | 7395/11074 [1:03:12<30:17,  2.02it/s]
 67%|██████▋   | 7396/11074 [1:03:13<30:16,  2.02it/s]
 67%|██████▋   | 7397/11074 [1:03:13<30:15,  2.03it/s]
 67%|██████▋   | 7398/11074 [1:03:14<30:15,  2.03it/s]
 67%|██████▋   | 7399/11074 [1:03:14<30:15,  2.02it/s]
 67%|██████▋   | 7400/11074 [1:03:15<30:16,  2.02it/s]{'loss': 3.2809, 'grad_norm': 0.23284220695495605, 'learning_rate': 0.00029948660243150096, 'epoch': 9.35}
+
                                                      

 67%|██████▋   | 7400/11074 [1:03:15<30:16,  2.02it/s]
 67%|██████▋   | 7401/11074 [1:03:15<30:16,  2.02it/s]
 67%|██████▋   | 7402/11074 [1:03:16<30:16,  2.02it/s]
 67%|██████▋   | 7403/11074 [1:03:16<30:13,  2.02it/s]
 67%|██████▋   | 7404/11074 [1:03:17<30:14,  2.02it/s]
 67%|██████▋   | 7405/11074 [1:03:17<30:13,  2.02it/s]
 67%|██████▋   | 7406/11074 [1:03:18<30:13,  2.02it/s]
 67%|██████▋   | 7407/11074 [1:03:18<30:11,  2.02it/s]
 67%|██████▋   | 7408/11074 [1:03:19<30:11,  2.02it/s]
 67%|██████▋   | 7409/11074 [1:03:19<30:09,  2.03it/s]
 67%|██████▋   | 7410/11074 [1:03:19<30:06,  2.03it/s]
 67%|██████▋   | 7411/11074 [1:03:20<30:09,  2.02it/s]
 67%|██████▋   | 7412/11074 [1:03:20<30:07,  2.03it/s]
 67%|██████▋   | 7413/11074 [1:03:21<30:08,  2.02it/s]
 67%|████��█▋   | 7414/11074 [1:03:21<30:05,  2.03it/s]
 67%|██████▋   | 7415/11074 [1:03:22<30:06,  2.03it/s]
 67%|██████▋   | 7416/11074 [1:03:22<30:05,  2.03it/s]
 67%|██████▋   | 7417/11074 [1:03:23<30:05,  2.03it/s]
 67%|██████▋   | 7418/11074 [1:03:23<30:04,  2.03it/s]
 67%|██████▋   | 7419/11074 [1:03:24<30:05,  2.02it/s]
 67%|██████▋   | 7420/11074 [1:03:24<30:03,  2.03it/s]
 67%|██████▋   | 7421/11074 [1:03:25<30:01,  2.03it/s]
 67%|██████▋   | 7422/11074 [1:03:25<30:03,  2.02it/s]
 67%|██████▋   | 7423/11074 [1:03:26<30:06,  2.02it/s]
 67%|██████▋   | 7424/11074 [1:03:26<30:05,  2.02it/s]
 67%|██████▋   | 7425/11074 [1:03:27<30:01,  2.03it/s]{'loss': 3.2872, 'grad_norm': 0.24848097562789917, 'learning_rate': 0.0002958832094946151, 'epoch': 9.38}
+
                                                      

 67%|██████▋   | 7425/11074 [1:03:27<30:01,  2.03it/s]
 67%|██████▋   | 7426/11074 [1:03:27<30:03,  2.02it/s]
 67%|██████▋   | 7427/11074 [1:03:28<30:01,  2.02it/s]
 67%|██████▋   | 7428/11074 [1:03:28<30:02,  2.02it/s]
 67%|██████▋   | 7429/11074 [1:03:29<30:00,  2.02it/s]
 67%|██████▋   | 7430/11074 [1:03:29<30:01,  2.02it/s]
 67%|██████▋   | 7431/11074 [1:03:30<29:58,  2.03it/s]
 67%|██████▋   | 7432/11074 [1:03:30<30:01,  2.02it/s]
 67%|██████▋   | 7433/11074 [1:03:31<29:58,  2.02it/s]
 67%|██████▋   | 7434/11074 [1:03:31<29:59,  2.02it/s]
 67%|██████▋   | 7435/11074 [1:03:32<29:57,  2.02it/s]
 67%|██████▋   | 7436/11074 [1:03:32<29:57,  2.02it/s]
 67%|██████▋   | 7437/11074 [1:03:33<29:56,  2.02it/s]
 67%|██████▋   | 7438/11074 [1:03:33<29:58,  2.02it/s]
 67%|██████▋   | 7439/11074 [1:03:34<29:56,  2.02it/s]
 67%|██████▋   | 7440/11074 [1:03:34<29:57,  2.02it/s]
 67%|██████▋   | 7441/11074 [1:03:35<29:55,  2.02it/s]
 67%|██████▋   | 7442/11074 [1:03:35<29:54,  2.02it/s]
 67%|██████▋   | 7443/11074 [1:03:36<29:54,  2.02it/s]
 67%|██████▋   | 7444/11074 [1:03:36<29:50,  2.03it/s]
 67%|██████▋   | 7445/11074 [1:03:37<29:51,  2.03it/s]
 67%|██████▋   | 7446/11074 [1:03:37<29:50,  2.03it/s]
 67%|██████▋   | 7447/11074 [1:03:38<29:50,  2.03it/s]
 67%|██████▋   | 7448/11074 [1:03:38<29:49,  2.03it/s]
 67%|██████▋   | 7449/11074 [1:03:39<29:46,  2.03it/s]
 67%|██████▋   | 7450/11074 [1:03:39<29:48,  2.03it/s]{'loss': 3.296, 'grad_norm': 0.22972343862056732, 'learning_rate': 0.0002922924934990568, 'epoch': 9.41}
+
                                                      

 67%|██████▋   | 7450/11074 [1:03:39<29:48,  2.03it/s]
 67%|██████▋   | 7451/11074 [1:03:40<29:47,  2.03it/s]
 67%|██████▋   | 7452/11074 [1:03:40<29:48,  2.03it/s]
 67%|██████▋   | 7453/11074 [1:03:41<29:47,  2.03it/s]
 67%|██████▋   | 7454/11074 [1:03:41<29:47,  2.03it/s]
 67%|██████▋   | 7455/11074 [1:03:42<29:46,  2.03it/s]
 67%|██████▋   | 7456/11074 [1:03:42<29:47,  2.02it/s]
 67%|██████▋   | 7457/11074 [1:03:43<29:47,  2.02it/s]
 67%|██████▋   | 7458/11074 [1:03:43<29:45,  2.03it/s]
 67%|██████▋   | 7459/11074 [1:03:44<29:44,  2.03it/s]
 67%|██████▋   | 7460/11074 [1:03:44<29:43,  2.03it/s]
 67%|██████▋   | 7461/11074 [1:03:45<29:43,  2.03it/s]
 67%|██████▋   | 7462/11074 [1:03:45<29:42,  2.03it/s]
 67%|██████▋   | 7463/11074 [1:03:46<29:41,  2.03it/s]
 67%|██████▋   | 7464/11074 [1:03:46<29:42,  2.03it/s]
 67%|██████▋   | 7465/11074 [1:03:47<29:41,  2.03it/s]
 67%|██████▋   | 7466/11074 [1:03:47<29:42,  2.02it/s]
 67%|██████▋   | 7467/11074 [1:03:48<29:41,  2.03it/s]
 67%|██████▋   | 7468/11074 [1:03:48<29:40,  2.02it/s]
 67%|██████▋   | 7469/11074 [1:03:49<29:40,  2.03it/s]
 67%|██████▋   | 7470/11074 [1:03:49<29:40,  2.02it/s]
 67%|██████▋   | 7471/11074 [1:03:50<29:38,  2.03it/s]
 67%|██████▋   | 7472/11074 [1:03:50<29:39,  2.02it/s]
 67%|██████▋   | 7473/11074 [1:03:51<29:39,  2.02it/s]
 67%|██████▋   | 7474/11074 [1:03:51<29:39,  2.02it/s]
 68%|██████▊   | 7475/11074 [1:03:52<29:39,  2.02it/s]{'loss': 3.2915, 'grad_norm': 0.23844318091869354, 'learning_rate': 0.0002887146774509584, 'epoch': 9.44}
+
                                                      

 68%|██████▊   | 7475/11074 [1:03:52<29:39,  2.02it/s]
 68%|██████▊   | 7476/11074 [1:03:52<29:40,  2.02it/s]
 68%|██████▊   | 7477/11074 [1:03:53<29:41,  2.02it/s]
 68%|██████▊   | 7478/11074 [1:03:53<29:39,  2.02it/s]
 68%|██████▊   | 7479/11074 [1:03:54<29:38,  2.02it/s]
 68%|██████▊   | 7480/11074 [1:03:54<29:36,  2.02it/s]
 68%|██████▊   | 7481/11074 [1:03:55<29:38,  2.02it/s]
 68%|██████▊   | 7482/11074 [1:03:55<29:38,  2.02it/s]
 68%|██████▊   | 7483/11074 [1:03:56<29:37,  2.02it/s]
 68%|██████▊   | 7484/11074 [1:03:56<29:35,  2.02it/s]
 68%|██████▊   | 7485/11074 [1:03:57<29:35,  2.02it/s]
 68%|██████▊   | 7486/11074 [1:03:57<29:34,  2.02it/s]
 68%|██████▊   | 7487/11074 [1:03:58<29:33,  2.02it/s]
 68%|██████▊   | 7488/11074 [1:03:58<29:33,  2.02it/s]
 68%|██████▊   | 7489/11074 [1:03:59<29:30,  2.02it/s]
 68%|██████▊   | 7490/11074 [1:03:59<29:30,  2.02it/s]
 68%|██████▊   | 7491/11074 [1:04:00<29:27,  2.03it/s]
 68%|██████▊   | 7492/11074 [1:04:00<29:27,  2.03it/s]
 68%|██████▊   | 7493/11074 [1:04:00<29:27,  2.03it/s]
 68%|██████▊   | 7494/11074 [1:04:01<29:27,  2.03it/s]
 68%|██████▊   | 7495/11074 [1:04:01<29:25,  2.03it/s]
 68%|██████▊   | 7496/11074 [1:04:02<29:28,  2.02it/s]
 68%|██████▊   | 7497/11074 [1:04:02<29:27,  2.02it/s]
 68%|██████▊   | 7498/11074 [1:04:03<29:27,  2.02it/s]
 68%|██████▊   | 7499/11074 [1:04:03<29:26,  2.02it/s]
 68%|██████▊   | 7500/11074 [1:04:04<29:27,  2.02it/s]{'loss': 3.2901, 'grad_norm': 0.23053638637065887, 'learning_rate': 0.00028514998355528415, 'epoch': 9.48}
+
                                                      

 68%|██████▊   | 7500/11074 [1:04:04<29:27,  2.02it/s]
 68%|██████▊   | 7501/11074 [1:04:04<29:29,  2.02it/s]
 68%|██████▊   | 7502/11074 [1:04:05<29:27,  2.02it/s]
 68%|██████▊   | 7503/11074 [1:04:05<29:25,  2.02it/s]
 68%|██████▊   | 7504/11074 [1:04:06<29:25,  2.02it/s]
 68%|██████▊   | 7505/11074 [1:04:06<29:21,  2.03it/s]
 68%|██████▊   | 7506/11074 [1:04:07<29:21,  2.02it/s]
 68%|██████▊   | 7507/11074 [1:04:07<29:21,  2.03it/s]
 68%|██████▊   | 7508/11074 [1:04:08<29:22,  2.02it/s]
 68%|██████▊   | 7509/11074 [1:04:08<29:20,  2.03it/s]
 68%|██████▊   | 7510/11074 [1:04:09<29:21,  2.02it/s]
 68%|██████▊   | 7511/11074 [1:04:09<29:21,  2.02it/s]
 68%|██████▊   | 7512/11074 [1:04:10<29:20,  2.02it/s]
 68%|██████▊   | 7513/11074 [1:04:10<29:18,  2.02it/s]
 68%|██████▊   | 7514/11074 [1:04:11<29:16,  2.03it/s]
 68%|██████▊   | 7515/11074 [1:04:11<29:16,  2.03it/s]
 68%|██████▊   | 7516/11074 [1:04:12<29:16,  2.03it/s]
 68%|██████▊   | 7517/11074 [1:04:12<29:17,  2.02it/s]
 68%|██████▊   | 7518/11074 [1:04:13<29:16,  2.02it/s]
 68%|██████▊   | 7519/11074 [1:04:13<29:14,  2.03it/s]
 68%|██████▊   | 7520/11074 [1:04:14<29:15,  2.02it/s]
 68%|██████▊   | 7521/11074 [1:04:14<29:13,  2.03it/s]
 68%|██████▊   | 7522/11074 [1:04:15<29:14,  2.02it/s]
 68%|██████▊   | 7523/11074 [1:04:15<29:12,  2.03it/s]
 68%|██████▊   | 7524/11074 [1:04:16<29:13,  2.02it/s]
 68%|██████▊   | 7525/11074 [1:04:16<29:14,  2.02it/s]{'loss': 3.2977, 'grad_norm': 0.23594748973846436, 'learning_rate': 0.00028159863320202937, 'epoch': 9.51}
+
                                                      

 68%|██████▊   | 7525/11074 [1:04:16<29:14,  2.02it/s]
 68%|██████▊   | 7526/11074 [1:04:17<29:15,  2.02it/s]
 68%|██████▊   | 7527/11074 [1:04:17<29:13,  2.02it/s]
 68%|██████▊   | 7528/11074 [1:04:18<29:13,  2.02it/s]
 68%|██████▊   | 7529/11074 [1:04:18<29:11,  2.02it/s]
 68%|██████▊   | 7530/11074 [1:04:19<29:11,  2.02it/s]
 68%|██████▊   | 7531/11074 [1:04:19<29:08,  2.03it/s]
 68%|██████▊   | 7532/11074 [1:04:20<29:07,  2.03it/s]
 68%|██████▊   | 7533/11074 [1:04:20<29:07,  2.03it/s]
 68%|██████▊   | 7534/11074 [1:04:21<29:05,  2.03it/s]
 68%|██████▊   | 7535/11074 [1:04:21<29:07,  2.03it/s]
 68%|██████▊   | 7536/11074 [1:04:22<29:05,  2.03it/s]
 68%|██████▊   | 7537/11074 [1:04:22<29:05,  2.03it/s]
 68%|██████▊   | 7538/11074 [1:04:23<29:05,  2.03it/s]
 68%|██████▊   | 7539/11074 [1:04:23<29:04,  2.03it/s]
 68%|██████▊   | 7540/11074 [1:04:24<29:05,  2.02it/s]
 68%|██████▊   | 7541/11074 [1:04:24<29:04,  2.03it/s]
 68%|██████▊   | 7542/11074 [1:04:25<29:05,  2.02it/s]
 68%|██████▊   | 7543/11074 [1:04:25<29:04,  2.02it/s]
 68%|██████▊   | 7544/11074 [1:04:26<29:04,  2.02it/s]
 68%|██████▊   | 7545/11074 [1:04:26<29:04,  2.02it/s]
 68%|██████▊   | 7546/11074 [1:04:27<29:04,  2.02it/s]
 68%|██████▊   | 7547/11074 [1:04:27<29:03,  2.02it/s]
 68%|██████▊   | 7548/11074 [1:04:28<29:02,  2.02it/s]
 68%|██████▊   | 7549/11074 [1:04:28<29:00,  2.03it/s]
 68%|██████▊   | 7550/11074 [1:04:29<29:00,  2.02it/s]{'loss': 3.302, 'grad_norm': 0.23429641127586365, 'learning_rate': 0.000278060846952472, 'epoch': 9.54}
+
                                                      

 68%|██████▊   | 7550/11074 [1:04:29<29:00,  2.02it/s]
 68%|██████▊   | 7551/11074 [1:04:29<29:00,  2.02it/s]
 68%|██████▊   | 7552/11074 [1:04:30<29:01,  2.02it/s]
 68%|██████▊   | 7553/11074 [1:04:30<28:59,  2.02it/s]
 68%|██████▊   | 7554/11074 [1:04:31<28:58,  2.02it/s]
 68%|██████▊   | 7555/11074 [1:04:31<28:58,  2.02it/s]
 68%|██████▊   | 7556/11074 [1:04:32<28:58,  2.02it/s]
 68%|██████▊   | 7557/11074 [1:04:32<28:57,  2.02it/s]
 68%|██████▊   | 7558/11074 [1:04:33<28:56,  2.02it/s]
 68%|██████▊   | 7559/11074 [1:04:33<28:58,  2.02it/s]
 68%|██████▊   | 7560/11074 [1:04:34<28:57,  2.02it/s]
 68%|██████▊   | 7561/11074 [1:04:34<28:56,  2.02it/s]
 68%|██████▊   | 7562/11074 [1:04:35<28:55,  2.02it/s]
 68%|██████▊   | 7563/11074 [1:04:35<28:55,  2.02it/s]
 68%|██████▊   | 7564/11074 [1:04:36<28:55,  2.02it/s]
 68%|██████▊   | 7565/11074 [1:04:36<28:54,  2.02it/s]
 68%|██████▊   | 7566/11074 [1:04:37<28:53,  2.02it/s]
 68%|██████▊   | 7567/11074 [1:04:37<28:53,  2.02it/s]
 68%|██████▊   | 7568/11074 [1:04:38<28:52,  2.02it/s]
 68%|██████▊   | 7569/11074 [1:04:38<28:53,  2.02it/s]
 68%|██████▊   | 7570/11074 [1:04:39<28:52,  2.02it/s]
 68%|██████▊   | 7571/11074 [1:04:39<28:51,  2.02it/s]
 68%|██████▊   | 7572/11074 [1:04:40<28:49,  2.02it/s]
 68%|██████▊   | 7573/11074 [1:04:40<28:50,  2.02it/s]
 68%|██████▊   | 7574/11074 [1:04:41<28:47,  2.03it/s]
 68%|██████▊   | 7575/11074 [1:04:41<28:46,  2.03it/s]{'loss': 3.2947, 'grad_norm': 0.2263978272676468, 'learning_rate': 0.0002745368445254728, 'epoch': 9.57}
                                                      
+
 68%|██████▊   | 7575/11074 [1:04:41<28:46,  2.03it/s]
 68%|██████▊   | 7576/11074 [1:04:41<28:48,  2.02it/s]
 68%|██████▊   | 7577/11074 [1:04:42<28:49,  2.02it/s]
 68%|██████▊   | 7578/11074 [1:04:42<28:47,  2.02it/s]
 68%|██████▊   | 7579/11074 [1:04:43<28:46,  2.02it/s]
 68%|██████▊   | 7580/11074 [1:04:43<28:46,  2.02it/s]
 68%|██████▊   | 7581/11074 [1:04:44<28:45,  2.02it/s]
 68%|██████▊   | 7582/11074 [1:04:44<28:45,  2.02it/s]
 68%|██████▊   | 7583/11074 [1:04:45<28:45,  2.02it/s]
 68%|██████▊   | 7584/11074 [1:04:45<28:44,  2.02it/s]
 68%|██████▊   | 7585/11074 [1:04:46<28:41,  2.03it/s]
 69%|██████▊   | 7586/11074 [1:04:46<28:42,  2.02it/s]
 69%|██████▊   | 7587/11074 [1:04:47<28:40,  2.03it/s]
 69%|██████▊   | 7588/11074 [1:04:47<28:41,  2.02it/s]
 69%|██████▊   | 7589/11074 [1:04:48<28:41,  2.02it/s]
 69%|██████▊   | 7590/11074 [1:04:48<28:41,  2.02it/s]
 69%|██████▊   | 7591/11074 [1:04:49<28:40,  2.02it/s]
 69%|██████▊   | 7592/11074 [1:04:49<28:40,  2.02it/s]
 69%|██████▊   | 7593/11074 [1:04:50<28:40,  2.02it/s]
 69%|██████▊   | 7594/11074 [1:04:50<28:39,  2.02it/s]
 69%|██████▊   | 7595/11074 [1:04:51<28:40,  2.02it/s]
 69%|██████▊   | 7596/11074 [1:04:51<28:38,  2.02it/s]
 69%|██████▊   | 7597/11074 [1:04:52<28:39,  2.02it/s]
 69%|██████▊   | 7598/11074 [1:04:52<28:38,  2.02it/s]
 69%|██████▊   | 7599/11074 [1:04:53<28:38,  2.02it/s]
 69%|██████▊   | 7600/11074 [1:04:53<28:37,  2.02it/s]{'loss': 3.2955, 'grad_norm': 0.24187149107456207, 'learning_rate': 0.00027102684478383, 'epoch': 9.6}
+
                                                      

 69%|██████▊   | 7600/11074 [1:04:53<28:37,  2.02it/s]
 69%|██████▊   | 7601/11074 [1:04:54<28:41,  2.02it/s]
 69%|██████▊   | 7602/11074 [1:04:54<28:36,  2.02it/s]
 69%|██████▊   | 7603/11074 [1:04:55<28:36,  2.02it/s]
 69%|██████▊   | 7604/11074 [1:04:55<28:35,  2.02it/s]
 69%|██████▊   | 7605/11074 [1:04:56<28:35,  2.02it/s]
 69%|██████▊   | 7606/11074 [1:04:56<28:34,  2.02it/s]
 69%|██████▊   | 7607/11074 [1:04:57<28:33,  2.02it/s]
 69%|██████▊   | 7608/11074 [1:04:57<28:34,  2.02it/s]
 69%|██████▊   | 7609/11074 [1:04:58<28:32,  2.02it/s]
 69%|██████▊   | 7610/11074 [1:04:58<31:06,  1.86it/s]
 69%|██████▊   | 7611/11074 [1:04:59<30:18,  1.90it/s]
 69%|██████▊   | 7612/11074 [1:04:59<29:45,  1.94it/s]
 69%|██████▊   | 7613/11074 [1:05:00<29:22,  1.96it/s]
 69%|██████▉   | 7614/11074 [1:05:00<29:05,  1.98it/s]
 69%|██████▉   | 7615/11074 [1:05:01<28:54,  1.99it/s]
 69%|██████▉   | 7616/11074 [1:05:01<28:46,  2.00it/s]
 69%|██████▉   | 7617/11074 [1:05:02<28:39,  2.01it/s]
 69%|██████▉   | 7618/11074 [1:05:02<28:33,  2.02it/s]
 69%|██████▉   | 7619/11074 [1:05:03<28:30,  2.02it/s]
 69%|██████▉   | 7620/11074 [1:05:03<28:27,  2.02it/s]
 69%|██████▉   | 7621/11074 [1:05:04<28:28,  2.02it/s]
 69%|██████▉   | 7622/11074 [1:05:04<28:25,  2.02it/s]
 69%|██████▉   | 7623/11074 [1:05:05<28:24,  2.02it/s]
 69%|██████▉   | 7624/11074 [1:05:05<28:24,  2.02it/s]
 69%|██████▉   | 7625/11074 [1:05:06<28:21,  2.03it/s]{'loss': 3.2966, 'grad_norm': 0.235133096575737, 'learning_rate': 0.0002675310657206874, 'epoch': 9.63}
                                                      
+
 69%|██████▉   | 7625/11074 [1:05:06<28:21,  2.03it/s]
 69%|██████▉   | 7626/11074 [1:05:06<28:24,  2.02it/s]
 69%|██████▉   | 7627/11074 [1:05:07<28:21,  2.03it/s]
 69%|██████▉   | 7628/11074 [1:05:07<28:21,  2.02it/s]
 69%|██████▉   | 7629/11074 [1:05:08<28:20,  2.03it/s]
 69%|██████▉   | 7630/11074 [1:05:08<28:20,  2.03it/s]
 69%|██████▉   | 7631/11074 [1:05:09<28:20,  2.03it/s]
 69%|██████▉   | 7632/11074 [1:05:09<28:18,  2.03it/s]
 69%|██████▉   | 7633/11074 [1:05:10<28:19,  2.02it/s]
 69%|██████▉   | 7634/11074 [1:05:10<28:17,  2.03it/s]
 69%|██████▉   | 7635/11074 [1:05:11<28:18,  2.02it/s]
 69%|██████▉   | 7636/11074 [1:05:11<28:17,  2.03it/s]
 69%|██████▉   | 7637/11074 [1:05:12<28:17,  2.02it/s]
 69%|██████▉   | 7638/11074 [1:05:12<28:17,  2.02it/s]
 69%|██████▉   | 7639/11074 [1:05:13<28:16,  2.03it/s]
 69%|██████▉   | 7640/11074 [1:05:13<28:13,  2.03it/s]
 69%|██████▉   | 7641/11074 [1:05:14<28:14,  2.03it/s]
 69%|██████▉   | 7642/11074 [1:05:14<28:14,  2.03it/s]
 69%|██████▉   | 7643/11074 [1:05:15<28:13,  2.03it/s]
 69%|██████▉   | 7644/11074 [1:05:15<28:13,  2.02it/s]
 69%|██████▉   | 7645/11074 [1:05:16<28:12,  2.03it/s]
 69%|██████▉   | 7646/11074 [1:05:16<28:12,  2.02it/s]
 69%|██████▉   | 7647/11074 [1:05:17<28:11,  2.03it/s]
 69%|██████▉   | 7648/11074 [1:05:17<28:10,  2.03it/s]
 69%|██████▉   | 7649/11074 [1:05:18<28:09,  2.03it/s]
 69%|██████▉   | 7650/11074 [1:05:18<28:09,  2.03it/s]{'loss': 3.3054, 'grad_norm': 0.23101946711540222, 'learning_rate': 0.0002640497244459946, 'epoch': 9.67}
                                                      
+
 69%|██████▉   | 7650/11074 [1:05:18<28:09,  2.03it/s]
 69%|██████▉   | 7651/11074 [1:05:19<28:13,  2.02it/s]
 69%|██████▉   | 7652/11074 [1:05:19<28:13,  2.02it/s]
 69%|██████▉   | 7653/11074 [1:05:20<28:10,  2.02it/s]
 69%|██████▉   | 7654/11074 [1:05:20<28:08,  2.02it/s]
 69%|██████▉   | 7655/11074 [1:05:21<28:08,  2.03it/s]
 69%|██████▉   | 7656/11074 [1:05:21<28:08,  2.02it/s]
 69%|██████▉   | 7657/11074 [1:05:22<28:09,  2.02it/s]
 69%|██████▉   | 7658/11074 [1:05:22<28:07,  2.02it/s]
 69%|██████▉   | 7659/11074 [1:05:23<28:07,  2.02it/s]
 69%|██████▉   | 7660/11074 [1:05:23<28:06,  2.02it/s]
 69%|██████▉   | 7661/11074 [1:05:24<28:06,  2.02it/s]
 69%|██████▉   | 7662/11074 [1:05:24<28:05,  2.02it/s]
 69%|██████▉   | 7663/11074 [1:05:25<28:04,  2.02it/s]
 69%|██████▉   | 7664/11074 [1:05:25<28:04,  2.02it/s]
 69%|██████▉   | 7665/11074 [1:05:26<28:04,  2.02it/s]
 69%|██████▉   | 7666/11074 [1:05:26<28:02,  2.03it/s]
 69%|██████▉   | 7667/11074 [1:05:27<28:02,  2.02it/s]
 69%|██████▉   | 7668/11074 [1:05:27<28:02,  2.02it/s]
 69%|██████▉   | 7669/11074 [1:05:28<28:03,  2.02it/s]
 69%|██████▉   | 7670/11074 [1:05:28<28:01,  2.02it/s]
 69%|██████▉   | 7671/11074 [1:05:29<28:01,  2.02it/s]
 69%|██████▉   | 7672/11074 [1:05:29<28:01,  2.02it/s]
 69%|██████▉   | 7673/11074 [1:05:30<27:59,  2.03it/s]
 69%|██████▉   | 7674/11074 [1:05:30<27:58,  2.03it/s]
 69%|██████▉   | 7675/11074 [1:05:31<27:57,  2.03it/s]{'loss': 3.305, 'grad_norm': 0.23394663631916046, 'learning_rate': 0.0002605830371730229, 'epoch': 9.7}
+
                                                      

 69%|██████▉   | 7675/11074 [1:05:31<27:57,  2.03it/s]
 69%|██████▉   | 7676/11074 [1:05:31<28:01,  2.02it/s]
 69%|██████▉   | 7677/11074 [1:05:32<27:58,  2.02it/s]
 69%|██████▉   | 7678/11074 [1:05:32<27:57,  2.02it/s]
 69%|██████▉   | 7679/11074 [1:05:33<27:56,  2.03it/s]
 69%|██████▉   | 7680/11074 [1:05:33<27:55,  2.03it/s]
 69%|██████▉   | 7681/11074 [1:05:34<27:54,  2.03it/s]
 69%|██████▉   | 7682/11074 [1:05:34<27:54,  2.03it/s]
 69%|██████▉   | 7683/11074 [1:05:35<30:19,  1.86it/s]
 69%|██████▉   | 7684/11074 [1:05:35<29:36,  1.91it/s]
 69%|██████▉   | 7685/11074 [1:05:36<29:05,  1.94it/s]
 69%|██████▉   | 7686/11074 [1:05:36<28:43,  1.97it/s]
 69%|██████▉   | 7687/11074 [1:05:37<28:26,  1.98it/s]
 69%|██████▉   | 7688/11074 [1:05:37<28:16,  2.00it/s]
 69%|██████▉   | 7689/11074 [1:05:38<28:09,  2.00it/s]
 69%|██████▉   | 7690/11074 [1:05:38<28:04,  2.01it/s]
 69%|██████▉   | 7691/11074 [1:05:39<27:58,  2.02it/s]
 69%|██████▉   | 7692/11074 [1:05:39<27:56,  2.02it/s]
 69%|██████▉   | 7693/11074 [1:05:40<27:51,  2.02it/s]
 69%|██████▉   | 7694/11074 [1:05:40<27:50,  2.02it/s]
 69%|██████▉   | 7695/11074 [1:05:41<27:49,  2.02it/s]
 69%|██████▉   | 7696/11074 [1:05:41<27:49,  2.02it/s]
 70%|██████▉   | 7697/11074 [1:05:42<27:48,  2.02it/s]
 70%|██████▉   | 7698/11074 [1:05:42<27:46,  2.03it/s]
 70%|██████▉   | 7699/11074 [1:05:43<27:46,  2.02it/s]
 70%|██████▉   | 7700/11074 [1:05:43<27:44,  2.03it/s]{'loss': 3.2982, 'grad_norm': 0.22738361358642578, 'learning_rate': 0.00025713121920493834, 'epoch': 9.73}
+
                                                      

 70%|██████▉   | 7700/11074 [1:05:43<27:44,  2.03it/s]
 70%|██████▉   | 7701/11074 [1:05:44<27:49,  2.02it/s]
 70%|██████▉   | 7702/11074 [1:05:44<27:46,  2.02it/s]
 70%|██████▉   | 7703/11074 [1:05:45<27:47,  2.02it/s]
 70%|██████▉   | 7704/11074 [1:05:45<27:42,  2.03it/s]
 70%|██████▉   | 7705/11074 [1:05:46<27:43,  2.03it/s]
 70%|██████▉   | 7706/11074 [1:05:46<27:41,  2.03it/s]
 70%|██████▉   | 7707/11074 [1:05:46<27:39,  2.03it/s]
 70%|██████▉   | 7708/11074 [1:05:47<27:40,  2.03it/s]
 70%|██████▉   | 7709/11074 [1:05:47<27:38,  2.03it/s]
 70%|██████▉   | 7710/11074 [1:05:48<27:39,  2.03it/s]
 70%|██████▉   | 7711/11074 [1:05:48<27:37,  2.03it/s]
 70%|██████▉   | 7712/11074 [1:05:49<27:38,  2.03it/s]
 70%|██████▉   | 7713/11074 [1:05:49<27:38,  2.03it/s]
 70%|██████▉   | 7714/11074 [1:05:50<27:40,  2.02it/s]
 70%|██████▉   | 7715/11074 [1:05:50<27:39,  2.02it/s]
 70%|██████▉   | 7716/11074 [1:05:51<27:37,  2.03it/s]
 70%|██████▉   | 7717/11074 [1:05:51<27:38,  2.02it/s]
 70%|██████▉   | 7718/11074 [1:05:52<27:35,  2.03it/s]
 70%|██████▉   | 7719/11074 [1:05:52<27:36,  2.03it/s]
 70%|██████▉   | 7720/11074 [1:05:53<27:34,  2.03it/s]
 70%|██████▉   | 7721/11074 [1:05:53<27:37,  2.02it/s]
 70%|██████▉   | 7722/11074 [1:05:54<27:36,  2.02it/s]
 70%|██████▉   | 7723/11074 [1:05:54<27:35,  2.02it/s]
 70%|██████▉   | 7724/11074 [1:05:55<27:34,  2.03it/s]
 70%|██████▉   | 7725/11074 [1:05:55<27:32,  2.03it/s]
                                                      {'loss': 3.2983, 'grad_norm': 0.23312132060527802, 'learning_rate': 0.00025369448492142864, 'epoch': 9.76}
+

 70%|██████▉   | 7725/11074 [1:05:55<27:32,  2.03it/s]
 70%|██████▉   | 7726/11074 [1:05:56<27:34,  2.02it/s]
 70%|██████▉   | 7727/11074 [1:05:56<27:35,  2.02it/s]
 70%|██████▉   | 7728/11074 [1:05:57<27:34,  2.02it/s]
 70%|██████▉   | 7729/11074 [1:05:57<27:32,  2.02it/s]
 70%|██████▉   | 7730/11074 [1:05:58<27:31,  2.02it/s]
 70%|██████▉   | 7731/11074 [1:05:58<27:29,  2.03it/s]
 70%|██████▉   | 7732/11074 [1:05:59<27:30,  2.02it/s]
 70%|██████▉   | 7733/11074 [1:05:59<27:28,  2.03it/s]
 70%|██████▉   | 7734/11074 [1:06:00<27:30,  2.02it/s]
 70%|██████▉   | 7735/11074 [1:06:00<27:28,  2.03it/s]
 70%|██████▉   | 7736/11074 [1:06:01<27:26,  2.03it/s]
 70%|██████▉   | 7737/11074 [1:06:01<27:27,  2.03it/s]
 70%|██████▉   | 7738/11074 [1:06:02<27:25,  2.03it/s]
 70%|██████▉   | 7739/11074 [1:06:02<27:24,  2.03it/s]
 70%|██████▉   | 7740/11074 [1:06:03<27:23,  2.03it/s]
 70%|██████▉   | 7741/11074 [1:06:03<27:21,  2.03it/s]
 70%|██████▉   | 7742/11074 [1:06:04<27:24,  2.03it/s]
 70%|██████▉   | 7743/11074 [1:06:04<27:23,  2.03it/s]
 70%|██████▉   | 7744/11074 [1:06:05<27:23,  2.03it/s]
 70%|██████▉   | 7745/11074 [1:06:05<27:22,  2.03it/s]
 70%|██████▉   | 7746/11074 [1:06:06<27:22,  2.03it/s]
 70%|██████▉   | 7747/11074 [1:06:06<27:21,  2.03it/s]
 70%|██████▉   | 7748/11074 [1:06:07<27:22,  2.03it/s]
 70%|██████▉   | 7749/11074 [1:06:07<27:21,  2.03it/s]
 70%|██████▉   | 7750/11074 [1:06:08<27:19,  2.03it/s]
                                                      {'loss': 3.3089, 'grad_norm': 0.2382747232913971, 'learning_rate': 0.0002502730477653896, 'epoch': 9.79}
+

 70%|██████▉   | 7750/11074 [1:06:08<27:19,  2.03it/s]
 70%|██████▉   | 7751/11074 [1:06:08<27:23,  2.02it/s]
 70%|███████   | 7752/11074 [1:06:09<27:23,  2.02it/s]
 70%|███████   | 7753/11074 [1:06:09<27:22,  2.02it/s]
 70%|███████   | 7754/11074 [1:06:10<27:19,  2.02it/s]
 70%|███████   | 7755/11074 [1:06:10<27:18,  2.03it/s]
 70%|███████   | 7756/11074 [1:06:11<27:17,  2.03it/s]
 70%|███████   | 7757/11074 [1:06:11<27:17,  2.03it/s]
 70%|███████   | 7758/11074 [1:06:12<27:17,  2.03it/s]
 70%|███████   | 7759/11074 [1:06:12<27:17,  2.02it/s]
 70%|███████   | 7760/11074 [1:06:13<27:18,  2.02it/s]
 70%|███████   | 7761/11074 [1:06:13<27:18,  2.02it/s]
 70%|███████   | 7762/11074 [1:06:14<27:17,  2.02it/s]
 70%|███████   | 7763/11074 [1:06:14<27:18,  2.02it/s]
 70%|███████   | 7764/11074 [1:06:15<27:16,  2.02it/s]
 70%|███████   | 7765/11074 [1:06:15<27:15,  2.02it/s]
 70%|███████   | 7766/11074 [1:06:16<27:15,  2.02it/s]
 70%|███████   | 7767/11074 [1:06:16<27:13,  2.02it/s]
 70%|███████   | 7768/11074 [1:06:17<27:13,  2.02it/s]
 70%|███████   | 7769/11074 [1:06:17<27:10,  2.03it/s]
 70%|███████   | 7770/11074 [1:06:18<27:10,  2.03it/s]
 70%|███████   | 7771/11074 [1:06:18<27:10,  2.03it/s]
 70%|███████   | 7772/11074 [1:06:19<27:07,  2.03it/s]
 70%|███████   | 7773/11074 [1:06:19<27:10,  2.02it/s]
 70%|███████   | 7774/11074 [1:06:20<27:09,  2.03it/s]
 70%|███████   | 7775/11074 [1:06:20<27:09,  2.02it/s]{'loss': 3.3003, 'grad_norm': 0.2325059324502945, 'learning_rate': 0.0002468671202296693, 'epoch': 9.82}
+
                                                      

 70%|███████   | 7775/11074 [1:06:20<27:09,  2.02it/s]
 70%|███████   | 7776/11074 [1:06:21<27:09,  2.02it/s]
 70%|███████   | 7777/11074 [1:06:21<27:09,  2.02it/s]
 70%|███████   | 7778/11074 [1:06:22<27:07,  2.03it/s]
 70%|███████   | 7779/11074 [1:06:22<27:08,  2.02it/s]
 70%|███████   | 7780/11074 [1:06:23<27:08,  2.02it/s]
 70%|███████   | 7781/11074 [1:06:23<27:06,  2.02it/s]
 70%|███████   | 7782/11074 [1:06:24<27:05,  2.03it/s]
 70%|███████   | 7783/11074 [1:06:24<27:04,  2.03it/s]
 70%|███████   | 7784/11074 [1:06:25<27:05,  2.02it/s]
 70%|███████   | 7785/11074 [1:06:25<27:04,  2.02it/s]
 70%|███████   | 7786/11074 [1:06:26<27:05,  2.02it/s]
 70%|███████   | 7787/11074 [1:06:26<27:03,  2.02it/s]
 70%|███████   | 7788/11074 [1:06:26<27:03,  2.02it/s]
 70%|███████   | 7789/11074 [1:06:27<27:02,  2.02it/s]
 70%|███████   | 7790/11074 [1:06:27<27:02,  2.02it/s]
 70%|███████   | 7791/11074 [1:06:28<27:01,  2.02it/s]
 70%|███████   | 7792/11074 [1:06:28<26:58,  2.03it/s]
 70%|███████   | 7793/11074 [1:06:29<26:59,  2.03it/s]
 70%|███████   | 7794/11074 [1:06:29<26:58,  2.03it/s]
 70%|███████   | 7795/11074 [1:06:30<26:58,  2.03it/s]
 70%|███████   | 7796/11074 [1:06:30<26:58,  2.03it/s]
 70%|███████   | 7797/11074 [1:06:31<26:57,  2.03it/s]
 70%|███████   | 7798/11074 [1:06:31<26:58,  2.02it/s]
 70%|███████   | 7799/11074 [1:06:32<26:56,  2.03it/s]
 70%|███████   | 7800/11074 [1:06:32<26:55,  2.03it/s]{'loss': 3.307, 'grad_norm': 0.23049074411392212, 'learning_rate': 0.00024347691384386978, 'epoch': 9.85}
+
                                                      

 70%|███████   | 7800/11074 [1:06:32<26:55,  2.03it/s]
 70%|███████   | 7801/11074 [1:06:33<26:56,  2.02it/s]
 70%|███████   | 7802/11074 [1:06:33<26:55,  2.02it/s]
 70%|███████   | 7803/11074 [1:06:34<26:57,  2.02it/s]
 70%|███████   | 7804/11074 [1:06:34<26:56,  2.02it/s]
 70%|███████   | 7805/11074 [1:06:35<26:54,  2.02it/s]
 70%|███████   | 7806/11074 [1:06:35<26:55,  2.02it/s]
 70%|███████   | 7807/11074 [1:06:36<26:53,  2.02it/s]
 71%|███████   | 7808/11074 [1:06:36<26:55,  2.02it/s]
 71%|███████   | 7809/11074 [1:06:37<26:51,  2.03it/s]
 71%|███████   | 7810/11074 [1:06:37<26:52,  2.02it/s]
 71%|███████   | 7811/11074 [1:06:38<26:50,  2.03it/s]
 71%|███████   | 7812/11074 [1:06:38<26:50,  2.03it/s]
 71%|███████   | 7813/11074 [1:06:39<26:50,  2.03it/s]
 71%|███████   | 7814/11074 [1:06:39<26:47,  2.03it/s]
 71%|███████   | 7815/11074 [1:06:40<26:47,  2.03it/s]
 71%|███████   | 7816/11074 [1:06:40<26:46,  2.03it/s]
 71%|███████   | 7817/11074 [1:06:41<26:47,  2.03it/s]
 71%|███████   | 7818/11074 [1:06:41<26:45,  2.03it/s]
 71%|███████   | 7819/11074 [1:06:42<26:47,  2.03it/s]
 71%|███████   | 7820/11074 [1:06:42<26:47,  2.02it/s]
 71%|███████   | 7821/11074 [1:06:43<26:47,  2.02it/s]
 71%|███████   | 7822/11074 [1:06:43<26:46,  2.02it/s]
 71%|███████   | 7823/11074 [1:06:44<26:45,  2.03it/s]
 71%|███████   | 7824/11074 [1:06:44<26:45,  2.02it/s]
 71%|███████   | 7825/11074 [1:06:45<26:44,  2.02it/s]{'loss': 3.3118, 'grad_norm': 0.23580317199230194, 'learning_rate': 0.0002401026391612111, 'epoch': 9.89}
                                                      
+
 71%|███████   | 7825/11074 [1:06:45<26:44,  2.02it/s]
 71%|███████   | 7826/11074 [1:06:45<26:46,  2.02it/s]
 71%|███████   | 7827/11074 [1:06:46<26:44,  2.02it/s]
 71%|███████   | 7828/11074 [1:06:46<26:45,  2.02it/s]
 71%|███████   | 7829/11074 [1:06:47<26:45,  2.02it/s]
 71%|███████   | 7830/11074 [1:06:47<26:44,  2.02it/s]
 71%|███████   | 7831/11074 [1:06:48<26:43,  2.02it/s]
 71%|███████   | 7832/11074 [1:06:48<26:43,  2.02it/s]
 71%|███████   | 7833/11074 [1:06:49<26:42,  2.02it/s]
 71%|███████   | 7834/11074 [1:06:49<26:42,  2.02it/s]
 71%|███████   | 7835/11074 [1:06:50<26:40,  2.02it/s]
 71%|███████   | 7836/11074 [1:06:50<26:39,  2.02it/s]
 71%|███████   | 7837/11074 [1:06:51<26:39,  2.02it/s]
 71%|███████   | 7838/11074 [1:06:51<26:40,  2.02it/s]
 71%|███████   | 7839/11074 [1:06:52<26:40,  2.02it/s]
 71%|███████   | 7840/11074 [1:06:52<26:39,  2.02it/s]
 71%|███████   | 7841/11074 [1:06:53<26:37,  2.02it/s]
 71%|███████   | 7842/11074 [1:06:53<26:38,  2.02it/s]
 71%|███████   | 7843/11074 [1:06:54<26:36,  2.02it/s]
 71%|███████   | 7844/11074 [1:06:54<26:36,  2.02it/s]
 71%|███████   | 7845/11074 [1:06:55<26:35,  2.02it/s]
 71%|███████   | 7846/11074 [1:06:55<26:35,  2.02it/s]
 71%|███████   | 7847/11074 [1:06:56<26:35,  2.02it/s]
 71%|███████   | 7848/11074 [1:06:56<26:33,  2.02it/s]
 71%|███████   | 7849/11074 [1:06:57<26:33,  2.02it/s]
 71%|███████   | 7850/11074 [1:06:57<26:33,  2.02it/s]
                                                      
{'loss': 3.3118, 'grad_norm': 0.23138053715229034, 'learning_rate': 0.00023674450574545343, 'epoch': 9.92}
+
 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]
                                                      
{'loss': 3.2656, 'grad_norm': 0.24008791148662567, 'learning_rate': 0.0002267690336224168, 'epoch': 10.01}
+
 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]
\ No newline at end of file