diff --git "a/run-2024-07-04T00:18:50+00:00.log" "b/run-2024-07-04T00:18:50+00:00.log"
--- "a/run-2024-07-04T00:18:50+00:00.log"
+++ "b/run-2024-07-04T00:18:50+00:00.log"
@@ -1049,4 +1049,508 @@ Non-default generation parameters: {'max_length': 200, 'early_stopping': True, '
 
 31%|███       | 20008/65536 [3:26:04<13:34:02,  1.07s/it]
 31%|███       | 20009/65536 [3:26:04<11:52:08,  1.07it/s]
 31%|███       | 20010/65536 [3:26:05<10:39:27,  1.19it/s]
 31%|███       | 20011/65536 [3:26:06<9:58:44,  1.27it/s] 
 31%|███       | 20012/65536 [3:26:06<9:28:21,  1.33it/s]
 31%|███       | 20013/65536 [3:26:07<9:09:16,  1.38it/s]
 31%|███       | 20014/65536 [3:26:08<8:48:20,  1.44it/s]
 31%|███       | 20015/65536 [3:26:08<8:41:42,  1.45it/s]
 31%|███       | 20016/65536 [3:26:09<8:37:20,  1.47it/s]
 31%|███       | 20017/65536 [3:26:10<8:32:14,  1.48it/s]
 31%|███       | 20018/65536 [3:26:10<8:21:03,  1.51it/s]
 31%|███       | 20019/65536 [3:26:11<8:05:57,  1.56it/s]
 31%|███       | 20020/65536 [3:26:11<8:01:26,  1.58it/s]
                                                         
{'loss': 3.1123, 'learning_rate': 7.269805314068814e-07, 'epoch': 1235.8}
 
 31%|███       | 20020/65536 [3:26:11<8:01:26,  1.58it/s]
 31%|███       | 20021/65536 [3:26:12<7:58:15,  1.59it/s]
 31%|███       | 20022/65536 [3:26:13<7:57:30,  1.59it/s]
 31%|███       | 20023/65536 [3:26:13<7:49:58,  1.61it/s]
 31%|███       | 20024/65536 [3:26:14<8:10:13,  1.55it/s]
 31%|███       | 20025/65536 [3:26:15<8:06:17,  1.56it/s]
 31%|███       | 20026/65536 [3:26:15<8:04:43,  1.56it/s]
 31%|███       | 20027/65536 [3:26:16<8:09:04,  1.55it/s]
 31%|███       | 20028/65536 [3:26:17<8:05:34,  1.56it/s]
 31%|███       | 20029/65536 [3:26:17<8:08:14,  1.55it/s]
 31%|███       | 20030/65536 [3:26:18<8:16:05,  1.53it/s]
 31%|███       | 20031/65536 [3:26:18<8:14:32,  1.53it/s]
 31%|███       | 20032/65536 [3:26:19<8:17:26,  1.52it/s]
 31%|███       | 20033/65536 [3:26:20<8:03:42,  1.57it/s]
 31%|███       | 20034/65536 [3:26:20<8:09:50,  1.55it/s]
 31%|███       | 20035/65536 [3:26:21<8:22:35,  1.51it/s]
 31%|███       | 20036/65536 [3:26:22<8:18:50,  1.52it/s]
 31%|███       | 20037/65536 [3:26:22<8:14:18,  1.53it/s]
 31%|███       | 20038/65536 [3:26:23<8:09:25,  1.55it/s]
 31%|███       | 20039/65536 [3:26:24<8:13:28,  1.54it/s]
 31%|███       | 20040/65536 [3:26:24<8:45:33,  1.44it/s]
                                                         
{'loss': 3.113, 'learning_rate': 7.267050324476552e-07, 'epoch': 1237.04}
 
 31%|███       | 20040/65536 [3:26:24<8:45:33,  1.44it/s]
 31%|███       | 20041/65536 [3:26:25<8:38:12,  1.46it/s]
 31%|███       | 20042/65536 [3:26:26<8:41:41,  1.45it/s]
 31%|███       | 20043/65536 [3:26:27<8:35:21,  1.47it/s]
 31%|███       | 20044/65536 [3:26:27<8:37:01,  1.47it/s]
 31%|███       | 20045/65536 [3:26:28<8:24:02,  1.50it/s]
 31%|███       | 20046/65536 [3:26:28<8:12:01,  1.54it/s]
 31%|███       | 20047/65536 [3:26:29<7:55:00,  1.60it/s]
 31%|███       | 20048/65536 [3:26:30<8:11:12,  1.54it/s]
 31%|███       | 20049/65536 [3:26:30<8:08:27,  1.55it/s]
 31%|███       | 20050/65536 [3:26:31<8:12:15,  1.54it/s]
 31%|███       | 20051/65536 [3:26:32<8:13:13,  1.54it/s]
 31%|███       | 20052/65536 [3:26:32<8:12:05,  1.54it/s]
 31%|███       | 20053/65536 [3:26:33<8:22:16,  1.51it/s]
 31%|███       | 20054/65536 [3:26:34<8:17:22,  1.52it/s]
 31%|███       | 20055/65536 [3:26:34<8:18:32,  1.52it/s]
 31%|███       | 20056/65536 [3:26:35<8:28:00,  1.49it/s]
 31%|███       | 20057/65536 [3:26:36<8:21:35,  1.51it/s]
 31%|███       | 20058/65536 [3:26:36<8:06:48,  1.56it/s]
 31%|███       | 20059/65536 [3:26:37<7:58:16,  1.58it/s]
 31%|███       | 20060/65536 [3:26:37<7:56:50,  1.59it/s]
                                                         
{'loss': 3.0655, 'learning_rate': 7.26429533488429e-07, 'epoch': 1238.27}
-
 31%|███       | 20060/65536 [3:26:37<7:56:50,  1.59it/s]
 31%|███       | 20061/65536 [3:26:38<7:59:27,  1.58it/s]
 31%|███       | 20062/65536 [3:26:39<8:04:22,  1.56it/s]
 31%|███       | 20063/65536 [3:26:39<8:06:37,  1.56it/s]
 31%|███       | 20064/65536 [3:26:40<8:00:34,  1.58it/s]
\ No newline at end of file
+
 31%|███       | 20060/65536 [3:26:37<7:56:50,  1.59it/s]
 31%|███       | 20061/65536 [3:26:38<7:59:27,  1.58it/s]
 31%|███       | 20062/65536 [3:26:39<8:04:22,  1.56it/s]
 31%|███       | 20063/65536 [3:26:39<8:06:37,  1.56it/s]
 31%|███       | 20064/65536 [3:26:40<8:00:34,  1.58it/s]
 31%|███       | 20065/65536 [3:26:41<7:57:42,  1.59it/s]
 31%|███       | 20066/65536 [3:26:41<8:11:10,  1.54it/s]
 31%|███       | 20067/65536 [3:26:42<7:58:04,  1.59it/s]
 31%|███       | 20068/65536 [3:26:43<7:48:08,  1.62it/s]
 31%|███       | 20069/65536 [3:26:43<7:55:21,  1.59it/s]
 31%|███       | 20070/65536 [3:26:44<7:51:52,  1.61it/s]
 31%|███       | 20071/65536 [3:26:44<8:02:18,  1.57it/s]
 31%|███       | 20072/65536 [3:26:45<8:18:14,  1.52it/s]
 31%|███       | 20073/65536 [3:26:46<8:21:50,  1.51it/s]
 31%|███       | 20074/65536 [3:26:46<8:05:06,  1.56it/s]
 31%|███       | 20075/65536 [3:26:47<7:56:16,  1.59it/s]
 31%|███       | 20076/65536 [3:26:48<7:48:46,  1.62it/s]
 31%|███       | 20077/65536 [3:26:48<7:45:45,  1.63it/s]
 31%|███       | 20078/65536 [3:26:49<7:45:57,  1.63it/s]
 31%|███       | 20079/65536 [3:26:49<7:49:00,  1.62it/s]
 31%|███       | 20080/65536 [3:26:50<7:58:26,  1.58it/s]
                                                         
{'loss': 3.1083, 'learning_rate': 7.261540345292028e-07, 'epoch': 1239.51}
+
 31%|███       | 20080/65536 [3:26:50<7:58:26,  1.58it/s]
 31%|███       | 20081/65536 [3:26:51<8:01:33,  1.57it/s]
 31%|███       | 20082/65536 [3:26:51<7:56:08,  1.59it/s]
 31%|███       | 20083/65536 [3:26:52<7:54:55,  1.60it/s]
 31%|███       | 20084/65536 [3:26:53<7:50:45,  1.61it/s]
 31%|███       | 20085/65536 [3:26:53<7:48:15,  1.62it/s]
 31%|███       | 20086/65536 [3:26:54<8:08:15,  1.55it/s]
 31%|███       | 20087/65536 [3:26:55<8:03:23,  1.57it/s]
 31%|███       | 20088/65536 [3:26:55<8:03:59,  1.57it/s]
 31%|███       | 20089/65536 [3:26:56<8:29:21,  1.49it/s]
 31%|███       | 20090/65536 [3:26:57<8:31:47,  1.48it/s]
 31%|███       | 20091/65536 [3:26:57<8:25:04,  1.50it/s]
 31%|███       | 20092/65536 [3:26:58<8:28:40,  1.49it/s]
 31%|███       | 20093/65536 [3:26:59<8:16:58,  1.52it/s]
 31%|███       | 20094/65536 [3:26:59<8:16:59,  1.52it/s]
 31%|███       | 20095/65536 [3:27:00<8:11:21,  1.54it/s]
 31%|███       | 20096/65536 [3:27:00<8:02:26,  1.57it/s]
 31%|███       | 20097/65536 [3:27:01<7:58:56,  1.58it/s]
 31%|███       | 20098/65536 [3:27:02<7:54:23,  1.60it/s]
 31%|███       | 20099/65536 [3:27:02<7:48:57,  1.61it/s]
 31%|███       | 20100/65536 [3:27:03<8:01:26,  1.57it/s]
                                                         
{'loss': 3.0929, 'learning_rate': 7.258785355699767e-07, 'epoch': 1240.74}
+
 31%|███       | 20100/65536 [3:27:03<8:01:26,  1.57it/s]
 31%|███       | 20101/65536 [3:27:04<8:03:00,  1.57it/s]
 31%|███       | 20102/65536 [3:27:04<8:19:40,  1.52it/s]
 31%|███       | 20103/65536 [3:27:05<8:17:16,  1.52it/s]
 31%|███       | 20104/65536 [3:27:06<8:17:55,  1.52it/s]
 31%|███       | 20105/65536 [3:27:06<8:34:42,  1.47it/s]
 31%|███       | 20106/65536 [3:27:07<8:38:36,  1.46it/s]
 31%|███       | 20107/65536 [3:27:08<8:35:20,  1.47it/s]
 31%|███       | 20108/65536 [3:27:08<8:27:59,  1.49it/s]
 31%|███       | 20109/65536 [3:27:09<8:09:56,  1.55it/s]
 31%|███       | 20110/65536 [3:27:10<8:13:08,  1.54it/s]
 31%|███       | 20111/65536 [3:27:10<8:10:42,  1.54it/s]
 31%|███       | 20112/65536 [3:27:11<8:13:24,  1.53it/s]
 31%|███       | 20113/65536 [3:27:12<7:59:54,  1.58it/s]
 31%|███       | 20114/65536 [3:27:12<7:51:27,  1.61it/s]
 31%|███       | 20115/65536 [3:27:13<7:47:43,  1.62it/s]
 31%|███       | 20116/65536 [3:27:13<7:45:44,  1.63it/s]
 31%|███       | 20117/65536 [3:27:14<7:42:45,  1.64it/s]
 31%|███       | 20118/65536 [3:27:15<8:03:36,  1.57it/s]
 31%|███       | 20119/65536 [3:27:15<8:26:31,  1.49it/s]
 31%|███       | 20120/65536 [3:27:16<8:35:58,  1.47it/s]
                                                         
{'loss': 3.0803, 'learning_rate': 7.256030366107506e-07, 'epoch': 1241.98}
+
 31%|███       | 20120/65536 [3:27:16<8:35:58,  1.47it/s]
 31%|███       | 20121/65536 [3:27:17<8:52:55,  1.42it/s]
 31%|███       | 20122/65536 [3:27:18<8:42:23,  1.45it/s]
 31%|███       | 20123/65536 [3:27:18<8:37:35,  1.46it/s]
 31%|███       | 20124/65536 [3:27:19<8:32:45,  1.48it/s]
 31%|███       | 20125/65536 [3:27:19<8:21:10,  1.51it/s]
 31%|███       | 20126/65536 [3:27:20<8:03:28,  1.57it/s]
 31%|███       | 20127/65536 [3:27:21<7:59:14,  1.58it/s]
 31%|███       | 20128/65536 [3:27:21<7:50:45,  1.61it/s]
 31%|███       | 20129/65536 [3:27:22<7:51:00,  1.61it/s]
 31%|███       | 20130/65536 [3:27:23<7:46:14,  1.62it/s]
 31%|███       | 20131/65536 [3:27:23<7:56:35,  1.59it/s]
 31%|███       | 20132/65536 [3:27:24<8:05:22,  1.56it/s]
 31%|███       | 20133/65536 [3:27:24<8:02:06,  1.57it/s]
 31%|███       | 20134/65536 [3:27:25<8:04:46,  1.56it/s]
 31%|███       | 20135/65536 [3:27:26<8:09:07,  1.55it/s]
 31%|███       | 20136/65536 [3:27:26<8:06:50,  1.55it/s]
 31%|███       | 20137/65536 [3:27:27<8:19:09,  1.52it/s]
 31%|███       | 20138/65536 [3:27:28<8:11:05,  1.54it/s]
 31%|███       | 20139/65536 [3:27:28<8:18:56,  1.52it/s]
 31%|███       | 20140/65536 [3:27:29<8:13:55,  1.53it/s]
                                                         
{'loss': 3.1016, 'learning_rate': 7.253275376515244e-07, 'epoch': 1243.21}
+
 31%|███       | 20140/65536 [3:27:29<8:13:55,  1.53it/s]
 31%|███       | 20141/65536 [3:27:30<8:05:09,  1.56it/s]
 31%|███       | 20142/65536 [3:27:30<8:17:17,  1.52it/s]
 31%|███       | 20143/65536 [3:27:31<8:18:38,  1.52it/s]
 31%|███       | 20144/65536 [3:27:32<8:22:24,  1.51it/s]
 31%|███       | 20145/65536 [3:27:32<8:09:04,  1.55it/s]
 31%|███       | 20146/65536 [3:27:33<8:14:57,  1.53it/s]
 31%|███       | 20147/65536 [3:27:34<8:10:43,  1.54it/s]
 31%|███       | 20148/65536 [3:27:34<7:59:07,  1.58it/s]
 31%|███       | 20149/65536 [3:27:35<7:57:43,  1.58it/s]
 31%|███       | 20150/65536 [3:27:35<7:51:54,  1.60it/s]
 31%|███       | 20151/65536 [3:27:36<7:55:27,  1.59it/s]
 31%|███       | 20152/65536 [3:27:37<8:07:29,  1.55it/s]
 31%|███       | 20153/65536 [3:27:38<8:38:49,  1.46it/s]
 31%|███       | 20154/65536 [3:27:38<8:35:58,  1.47it/s]
 31%|███       | 20155/65536 [3:27:39<8:28:59,  1.49it/s]
 31%|███       | 20156/65536 [3:27:40<8:23:03,  1.50it/s]
 31%|███       | 20157/65536 [3:27:40<8:16:55,  1.52it/s]
 31%|███       | 20158/65536 [3:27:41<8:08:21,  1.55it/s]
 31%|███       | 20159/65536 [3:27:41<8:05:31,  1.56it/s]
 31%|███       | 20160/65536 [3:27:42<8:04:16,  1.56it/s]
                                                         
{'loss': 3.0868, 'learning_rate': 7.250520386922983e-07, 'epoch': 1244.44}
+
 31%|███       | 20160/65536 [3:27:42<8:04:16,  1.56it/s]
 31%|███       | 20161/65536 [3:27:43<8:17:47,  1.52it/s]
 31%|███       | 20162/65536 [3:27:43<8:14:14,  1.53it/s]
 31%|███       | 20163/65536 [3:27:44<8:16:25,  1.52it/s]
 31%|███       | 20164/65536 [3:27:45<8:07:25,  1.55it/s]
 31%|███       | 20165/65536 [3:27:45<8:03:02,  1.57it/s]
 31%|███       | 20166/65536 [3:27:46<8:14:43,  1.53it/s]
 31%|███       | 20167/65536 [3:27:47<8:11:56,  1.54it/s]
 31%|███       | 20168/65536 [3:27:47<8:27:39,  1.49it/s]
 31%|███       | 20169/65536 [3:27:48<8:19:07,  1.51it/s]
 31%|███       | 20170/65536 [3:27:49<8:34:24,  1.47it/s]
 31%|███       | 20171/65536 [3:27:49<8:28:50,  1.49it/s]
 31%|███       | 20172/65536 [3:27:50<8:38:25,  1.46it/s]
 31%|███       | 20173/65536 [3:27:51<8:19:14,  1.51it/s]
 31%|███       | 20174/65536 [3:27:51<8:24:44,  1.50it/s]
 31%|███       | 20175/65536 [3:27:52<8:10:31,  1.54it/s]
 31%|███       | 20176/65536 [3:27:53<8:16:12,  1.52it/s]
 31%|███       | 20177/65536 [3:27:53<7:59:55,  1.58it/s]
 31%|███       | 20178/65536 [3:27:54<8:04:41,  1.56it/s]
 31%|███       | 20179/65536 [3:27:55<8:01:47,  1.57it/s]
 31%|███       | 20180/65536 [3:27:55<8:04:04,  1.56it/s]
                                                         
{'loss': 3.053, 'learning_rate': 7.247765397330721e-07, 'epoch': 1245.68}
+
 31%|███       | 20180/65536 [3:27:55<8:04:04,  1.56it/s]
 31%|███       | 20181/65536 [3:27:56<8:07:04,  1.55it/s]
 31%|███       | 20182/65536 [3:27:56<8:12:12,  1.54it/s]
 31%|███       | 20183/65536 [3:27:57<8:05:44,  1.56it/s]
 31%|███       | 20184/65536 [3:27:58<8:08:46,  1.55it/s]
 31%|███       | 20185/65536 [3:27:58<8:12:00,  1.54it/s]
 31%|███       | 20186/65536 [3:27:59<8:37:13,  1.46it/s]
 31%|███       | 20187/65536 [3:28:00<8:52:22,  1.42it/s]
 31%|███       | 20188/65536 [3:28:01<8:39:36,  1.45it/s]
 31%|███       | 20189/65536 [3:28:01<8:22:10,  1.51it/s]
 31%|███       | 20190/65536 [3:28:02<8:07:49,  1.55it/s]
 31%|███       | 20191/65536 [3:28:02<8:12:33,  1.53it/s]
 31%|███       | 20192/65536 [3:28:03<8:15:54,  1.52it/s]
 31%|███       | 20193/65536 [3:28:04<8:05:33,  1.56it/s]
 31%|███       | 20194/65536 [3:28:04<8:01:13,  1.57it/s]
 31%|███       | 20195/65536 [3:28:05<8:02:43,  1.57it/s]
 31%|███       | 20196/65536 [3:28:06<8:18:46,  1.52it/s]
 31%|███       | 20197/65536 [3:28:06<8:04:37,  1.56it/s]
 31%|███       | 20198/65536 [3:28:07<8:00:21,  1.57it/s]
 31%|███       | 20199/65536 [3:28:08<8:01:52,  1.57it/s]
 31%|███       | 20200/65536 [3:28:08<8:01:38,  1.57it/s]
                                                         
{'loss': 3.0749, 'learning_rate': 7.24501040773846e-07, 'epoch': 1246.91}
+
 31%|███       | 20200/65536 [3:28:08<8:01:38,  1.57it/s]
 31%|███       | 20201/65536 [3:28:09<8:02:57,  1.56it/s]
 31%|███       | 20202/65536 [3:28:10<8:26:48,  1.49it/s]
 31%|███       | 20203/65536 [3:28:10<8:26:24,  1.49it/s]
 31%|███       | 20204/65536 [3:28:11<8:10:32,  1.54it/s]
 31%|███       | 20205/65536 [3:28:11<7:56:49,  1.58it/s]
 31%|███       | 20206/65536 [3:28:12<7:59:43,  1.57it/s]
 31%|███       | 20207/65536 [3:28:13<7:46:57,  1.62it/s]
 31%|███       | 20208/65536 [3:28:13<7:45:46,  1.62it/s]
 31%|███       | 20209/65536 [3:28:14<7:44:50,  1.63it/s]
 31%|███       | 20210/65536 [3:28:15<7:47:35,  1.62it/s]
 31%|███       | 20211/65536 [3:28:15<7:42:54,  1.63it/s]
 31%|███       | 20212/65536 [3:28:16<7:43:05,  1.63it/s]
 31%|███       | 20213/65536 [3:28:16<7:42:27,  1.63it/s]
 31%|███       | 20214/65536 [3:28:17<7:38:06,  1.65it/s]
 31%|███       | 20215/65536 [3:28:18<7:55:30,  1.59it/s]
 31%|███       | 20216/65536 [3:28:18<7:56:49,  1.58it/s]
 31%|███       | 20217/65536 [3:28:19<7:58:46,  1.58it/s]
 31%|███       | 20218/65536 [3:28:20<8:12:42,  1.53it/s]
 31%|███       | 20219/65536 [3:28:20<8:00:27,  1.57it/s]
 31%|███       | 20220/65536 [3:28:21<7:59:10,  1.58it/s]
                                                         
{'loss': 3.125, 'learning_rate': 7.242255418146198e-07, 'epoch': 1248.15}
+
 31%|███       | 20220/65536 [3:28:21<7:59:10,  1.58it/s]
 31%|███       | 20221/65536 [3:28:21<7:53:38,  1.59it/s]
 31%|███       | 20222/65536 [3:28:22<7:53:07,  1.60it/s]
 31%|███       | 20223/65536 [3:28:23<7:50:56,  1.60it/s]
 31%|███       | 20224/65536 [3:28:23<7:48:09,  1.61it/s]
 31%|███       | 20225/65536 [3:28:24<7:42:42,  1.63it/s]
 31%|███       | 20226/65536 [3:28:25<7:43:45,  1.63it/s]
 31%|███       | 20227/65536 [3:28:25<7:40:26,  1.64it/s]
 31%|███       | 20228/65536 [3:28:26<7:47:59,  1.61it/s]
 31%|███       | 20229/65536 [3:28:26<7:46:05,  1.62it/s]
 31%|███       | 20230/65536 [3:28:27<7:52:58,  1.60it/s]
 31%|███       | 20231/65536 [3:28:28<7:48:38,  1.61it/s]
 31%|███       | 20232/65536 [3:28:28<7:41:19,  1.64it/s]
 31%|███       | 20233/65536 [3:28:29<7:36:52,  1.65it/s]
 31%|███       | 20234/65536 [3:28:30<8:01:48,  1.57it/s]
 31%|███       | 20235/65536 [3:28:30<7:59:04,  1.58it/s]
 31%|███       | 20236/65536 [3:28:31<8:14:54,  1.53it/s]
 31%|███       | 20237/65536 [3:28:31<8:05:04,  1.56it/s]
 31%|███       | 20238/65536 [3:28:32<7:49:42,  1.61it/s]
 31%|███       | 20239/65536 [3:28:33<7:47:39,  1.61it/s]
 31%|███       | 20240/65536 [3:28:33<7:42:18,  1.63it/s]
                                                         
{'loss': 3.1073, 'learning_rate': 7.239500428553937e-07, 'epoch': 1249.38}
+
 31%|███       | 20240/65536 [3:28:33<7:42:18,  1.63it/s]
 31%|███       | 20241/65536 [3:28:34<7:50:38,  1.60it/s]
 31%|███       | 20242/65536 [3:28:35<7:50:02,  1.61it/s]
 31%|███       | 20243/65536 [3:28:35<8:03:50,  1.56it/s]
 31%|███       | 20244/65536 [3:28:36<7:51:35,  1.60it/s]
 31%|███       | 20245/65536 [3:28:36<7:54:15,  1.59it/s]
 31%|███       | 20246/65536 [3:28:37<7:49:27,  1.61it/s]
 31%|███       | 20247/65536 [3:28:38<7:45:00,  1.62it/s]
 31%|███       | 20248/65536 [3:28:38<7:47:43,  1.61it/s]
 31%|███       | 20249/65536 [3:28:39<7:40:37,  1.64it/s]
 31%|███       | 20250/65536 [3:28:39<7:30:53,  1.67it/s]
 31%|███       | 20251/65536 [3:28:40<8:16:39,  1.52it/s]
 31%|███       | 20252/65536 [3:28:41<8:12:31,  1.53it/s]
 31%|███       | 20253/65536 [3:28:41<8:03:49,  1.56it/s]
 31%|███       | 20254/65536 [3:28:42<8:09:40,  1.54it/s]
 31%|███       | 20255/65536 [3:28:43<7:59:38,  1.57it/s]
 31%|███       | 20256/65536 [3:28:43<7:50:11,  1.61it/s]
 31%|███       | 20257/65536 [3:28:44<7:48:42,  1.61it/s]
 31%|███       | 20258/65536 [3:28:45<7:48:26,  1.61it/s]
 31%|███       | 20259/65536 [3:28:45<7:47:51,  1.61it/s]
 31%|███       | 20260/65536 [3:28:46<7:47:56,  1.61it/s]
                                                         
{'loss': 3.0613, 'learning_rate': 7.236745438961675e-07, 'epoch': 1250.62}
+
 31%|███       | 20260/65536 [3:28:46<7:47:56,  1.61it/s]
 31%|███       | 20261/65536 [3:28:46<8:03:44,  1.56it/s]
 31%|███       | 20262/65536 [3:28:47<8:02:57,  1.56it/s]
 31%|███       | 20263/65536 [3:28:48<7:57:20,  1.58it/s]
 31%|███       | 20264/65536 [3:28:48<7:53:22,  1.59it/s]
 31%|███       | 20265/65536 [3:28:49<7:45:24,  1.62it/s]
 31%|███       | 20266/65536 [3:28:50<7:33:10,  1.66it/s]
 31%|███       | 20267/65536 [3:28:50<7:53:10,  1.59it/s]
 31%|███       | 20268/65536 [3:28:51<7:48:58,  1.61it/s]
 31%|███       | 20269/65536 [3:28:51<7:33:45,  1.66it/s]
 31%|███       | 20270/65536 [3:28:52<7:40:51,  1.64it/s]
 31%|███       | 20271/65536 [3:28:53<7:45:27,  1.62it/s]
 31%|███       | 20272/65536 [3:28:53<7:47:08,  1.61it/s]
 31%|███       | 20273/65536 [3:28:54<7:49:00,  1.61it/s]
 31%|███       | 20274/65536 [3:28:54<7:36:59,  1.65it/s]
 31%|███       | 20275/65536 [3:28:55<7:45:15,  1.62it/s]
 31%|███       | 20276/65536 [3:28:56<7:44:42,  1.62it/s]
 31%|███       | 20277/65536 [3:28:56<7:46:51,  1.62it/s]
 31%|███       | 20278/65536 [3:28:57<7:55:46,  1.59it/s]
 31%|███       | 20279/65536 [3:28:58<8:17:52,  1.51it/s]
 31%|███       | 20280/65536 [3:28:58<8:02:11,  1.56it/s]
                                                         
{'loss': 3.0474, 'learning_rate': 7.233990449369414e-07, 'epoch': 1251.85}
+
 31%|███       | 20280/65536 [3:28:58<8:02:11,  1.56it/s]
 31%|███       | 20281/65536 [3:28:59<7:59:28,  1.57it/s]
 31%|███       | 20282/65536 [3:29:00<7:48:27,  1.61it/s]
 31%|███       | 20283/65536 [3:29:00<8:05:45,  1.55it/s]
 31%|███       | 20284/65536 [3:29:01<8:15:00,  1.52it/s]
 31%|███       | 20285/65536 [3:29:02<8:06:29,  1.55it/s]
 31%|███       | 20286/65536 [3:29:02<8:02:35,  1.56it/s]
 31%|███       | 20287/65536 [3:29:03<8:09:20,  1.54it/s]
 31%|███       | 20288/65536 [3:29:03<7:57:27,  1.58it/s]
 31%|███       | 20289/65536 [3:29:04<7:52:44,  1.60it/s]
 31%|███       | 20290/65536 [3:29:05<7:54:32,  1.59it/s]
 31%|███       | 20291/65536 [3:29:05<7:52:25,  1.60it/s]
 31%|███       | 20292/65536 [3:29:06<7:56:04,  1.58it/s]
 31%|███       | 20293/65536 [3:29:07<7:50:00,  1.60it/s]
 31%|███       | 20294/65536 [3:29:07<7:39:30,  1.64it/s]
 31%|███       | 20295/65536 [3:29:08<7:38:14,  1.65it/s]
 31%|███       | 20296/65536 [3:29:08<7:38:32,  1.64it/s]
 31%|███       | 20297/65536 [3:29:09<7:30:15,  1.67it/s]
 31%|███       | 20298/65536 [3:29:10<7:35:47,  1.65it/s]
 31%|███       | 20299/65536 [3:29:10<7:54:42,  1.59it/s]
 31%|███       | 20300/65536 [3:29:11<7:49:08,  1.61it/s]
                                                         
{'loss': 3.0647, 'learning_rate': 7.231235459777152e-07, 'epoch': 1253.09}
+
 31%|███       | 20300/65536 [3:29:11<7:49:08,  1.61it/s]
 31%|███       | 20301/65536 [3:29:11<7:42:40,  1.63it/s]
 31%|███       | 20302/65536 [3:29:12<7:43:15,  1.63it/s]
 31%|███       | 20303/65536 [3:29:13<7:53:24,  1.59it/s]
 31%|███       | 20304/65536 [3:29:13<7:54:32,  1.59it/s]
 31%|███       | 20305/65536 [3:29:14<8:11:23,  1.53it/s]
 31%|███       | 20306/65536 [3:29:15<8:05:40,  1.55it/s]
 31%|███       | 20307/65536 [3:29:15<8:00:23,  1.57it/s]
 31%|███       | 20308/65536 [3:29:16<7:53:55,  1.59it/s]
 31%|███       | 20309/65536 [3:29:16<7:49:31,  1.61it/s]
 31%|███       | 20310/65536 [3:29:17<7:51:17,  1.60it/s]
 31%|███       | 20311/65536 [3:29:18<7:49:29,  1.61it/s]
 31%|███       | 20312/65536 [3:29:18<7:40:16,  1.64it/s]
 31%|███       | 20313/65536 [3:29:19<7:43:32,  1.63it/s]
 31%|███       | 20314/65536 [3:29:20<7:42:19,  1.63it/s]
 31%|███       | 20315/65536 [3:29:20<7:58:49,  1.57it/s]
 31%|███       | 20316/65536 [3:29:21<7:54:48,  1.59it/s]
 31%|███       | 20317/65536 [3:29:21<7:47:01,  1.61it/s]
 31%|███       | 20318/65536 [3:29:22<7:40:32,  1.64it/s]
 31%|███       | 20319/65536 [3:29:23<7:57:35,  1.58it/s]
 31%|███       | 20320/65536 [3:29:23<7:46:35,  1.62it/s]
                                                         
{'loss': 3.0641, 'learning_rate': 7.228480470184889e-07, 'epoch': 1254.32}
+
 31%|███       | 20320/65536 [3:29:23<7:46:35,  1.62it/s]
 31%|███       | 20321/65536 [3:29:24<7:39:58,  1.64it/s]
 31%|███       | 20322/65536 [3:29:24<7:32:16,  1.67it/s]
 31%|███       | 20323/65536 [3:29:25<7:39:17,  1.64it/s]
 31%|███       | 20324/65536 [3:29:26<7:50:34,  1.60it/s]
 31%|███       | 20325/65536 [3:29:26<8:00:45,  1.57it/s]
 31%|███       | 20326/65536 [3:29:27<7:57:07,  1.58it/s]
 31%|███       | 20327/65536 [3:29:28<7:48:27,  1.61it/s]
 31%|███       | 20328/65536 [3:29:28<7:44:56,  1.62it/s]
 31%|███       | 20329/65536 [3:29:29<7:50:37,  1.60it/s]
 31%|███       | 20330/65536 [3:29:30<7:50:53,  1.60it/s]
 31%|███       | 20331/65536 [3:29:30<7:47:17,  1.61it/s]
 31%|███       | 20332/65536 [3:29:31<7:55:35,  1.58it/s]
 31%|███       | 20333/65536 [3:29:31<7:56:29,  1.58it/s]
 31%|███       | 20334/65536 [3:29:32<7:37:09,  1.65it/s]
 31%|███       | 20335/65536 [3:29:33<7:44:06,  1.62it/s]
 31%|███       | 20336/65536 [3:29:33<7:49:05,  1.61it/s]
 31%|███       | 20337/65536 [3:29:34<7:46:43,  1.61it/s]
 31%|███       | 20338/65536 [3:29:34<7:44:47,  1.62it/s]
 31%|███       | 20339/65536 [3:29:35<7:58:41,  1.57it/s]
 31%|███       | 20340/65536 [3:29:36<7:49:49,  1.60it/s]
                                                         
{'loss': 3.0684, 'learning_rate': 7.225725480592628e-07, 'epoch': 1255.56}
+
 31%|███       | 20340/65536 [3:29:36<7:49:49,  1.60it/s]
 31%|███       | 20341/65536 [3:29:36<7:46:01,  1.62it/s]
 31%|███       | 20342/65536 [3:29:37<7:56:29,  1.58it/s]
 31%|███       | 20343/65536 [3:29:38<7:50:41,  1.60it/s]
 31%|███       | 20344/65536 [3:29:38<8:02:53,  1.56it/s]
 31%|███       | 20345/65536 [3:29:39<7:46:25,  1.61it/s]
 31%|███       | 20346/65536 [3:29:39<7:45:31,  1.62it/s]
 31%|███       | 20347/65536 [3:29:40<7:38:29,  1.64it/s]
 31%|███       | 20348/65536 [3:29:41<7:56:51,  1.58it/s]
 31%|███       | 20349/65536 [3:29:41<7:43:44,  1.62it/s]
 31%|███       | 20350/65536 [3:29:42<7:39:02,  1.64it/s]
 31%|███       | 20351/65536 [3:29:43<7:44:14,  1.62it/s]
 31%|███       | 20352/65536 [3:29:43<7:53:31,  1.59it/s]
 31%|███       | 20353/65536 [3:29:44<7:56:56,  1.58it/s]
 31%|███       | 20354/65536 [3:29:44<7:49:16,  1.60it/s]
 31%|███       | 20355/65536 [3:29:45<7:49:29,  1.60it/s]
 31%|███       | 20356/65536 [3:29:46<7:55:30,  1.58it/s]
 31%|███       | 20357/65536 [3:29:46<7:47:05,  1.61it/s]
 31%|███       | 20358/65536 [3:29:47<7:54:47,  1.59it/s]
 31%|███       | 20359/65536 [3:29:48<8:11:21,  1.53it/s]
 31%|███       | 20360/65536 [3:29:48<8:00:06,  1.57it/s]
                                                         
{'loss': 3.0527, 'learning_rate': 7.222970491000366e-07, 'epoch': 1256.79}
+
 31%|███       | 20360/65536 [3:29:48<8:00:06,  1.57it/s]
 31%|███       | 20361/65536 [3:29:49<7:49:27,  1.60it/s]
 31%|███       | 20362/65536 [3:29:50<7:47:55,  1.61it/s]
 31%|███       | 20363/65536 [3:29:50<7:44:14,  1.62it/s]
 31%|███       | 20364/65536 [3:29:51<7:56:54,  1.58it/s]
 31%|███       | 20365/65536 [3:29:51<7:49:22,  1.60it/s]
 31%|███       | 20366/65536 [3:29:52<7:54:25,  1.59it/s]
 31%|███       | 20367/65536 [3:29:53<8:13:31,  1.53it/s]
 31%|███       | 20368/65536 [3:29:53<8:07:47,  1.54it/s]
 31%|███       | 20369/65536 [3:29:54<8:05:07,  1.55it/s]
 31%|███       | 20370/65536 [3:29:55<7:53:32,  1.59it/s]
 31%|███       | 20371/65536 [3:29:55<7:42:39,  1.63it/s]
 31%|███       | 20372/65536 [3:29:56<7:33:16,  1.66it/s]
 31%|███       | 20373/65536 [3:29:56<7:22:19,  1.70it/s]
 31%|███       | 20374/65536 [3:29:57<7:27:31,  1.68it/s]
 31%|███       | 20375/65536 [3:29:58<7:36:55,  1.65it/s]
 31%|███       | 20376/65536 [3:29:58<7:40:02,  1.64it/s]
 31%|███       | 20377/65536 [3:29:59<7:39:06,  1.64it/s]
 31%|███       | 20378/65536 [3:29:59<7:31:18,  1.67it/s]
 31%|███       | 20379/65536 [3:30:00<7:30:22,  1.67it/s]
 31%|███       | 20380/65536 [3:30:01<7:47:24,  1.61it/s]
                                                         
{'loss': 3.1258, 'learning_rate': 7.220215501408105e-07, 'epoch': 1258.02}
+
 31%|███       | 20380/65536 [3:30:01<7:47:24,  1.61it/s]
 31%|███       | 20381/65536 [3:30:01<7:49:39,  1.60it/s]
 31%|███       | 20382/65536 [3:30:02<7:44:40,  1.62it/s]
 31%|███       | 20383/65536 [3:30:02<7:46:34,  1.61it/s]
 31%|███       | 20384/65536 [3:30:03<7:43:34,  1.62it/s]
 31%|███       | 20385/65536 [3:30:04<8:09:26,  1.54it/s]
 31%|███       | 20386/65536 [3:30:04<7:57:03,  1.58it/s]
 31%|███       | 20387/65536 [3:30:05<7:46:00,  1.61it/s]
 31%|███       | 20388/65536 [3:30:06<7:43:04,  1.62it/s]
 31%|███       | 20389/65536 [3:30:06<7:35:01,  1.65it/s]
 31%|███       | 20390/65536 [3:30:07<7:33:14,  1.66it/s]
 31%|███       | 20391/65536 [3:30:07<7:40:10,  1.64it/s]
 31%|███       | 20392/65536 [3:30:08<7:42:19,  1.63it/s]
 31%|███       | 20393/65536 [3:30:09<7:41:48,  1.63it/s]
 31%|███       | 20394/65536 [3:30:09<7:34:35,  1.66it/s]
 31%|███       | 20395/65536 [3:30:10<7:32:50,  1.66it/s]
 31%|███       | 20396/65536 [3:30:11<7:49:43,  1.60it/s]
 31%|███       | 20397/65536 [3:30:11<7:31:11,  1.67it/s]
 31%|███       | 20398/65536 [3:30:12<7:34:16,  1.66it/s]
 31%|███       | 20399/65536 [3:30:12<7:23:42,  1.70it/s]
 31%|███       | 20400/65536 [3:30:13<7:38:53,  1.64it/s]
                                                         
{'loss': 3.0887, 'learning_rate': 7.217460511815844e-07, 'epoch': 1259.26}
+
 31%|███       | 20400/65536 [3:30:13<7:38:53,  1.64it/s]
 31%|███       | 20401/65536 [3:30:13<7:37:13,  1.65it/s]
 31%|███       | 20402/65536 [3:30:14<7:27:53,  1.68it/s]
 31%|███       | 20403/65536 [3:30:15<7:30:48,  1.67it/s]
 31%|███       | 20404/65536 [3:30:15<7:30:39,  1.67it/s]
 31%|███       | 20405/65536 [3:30:16<7:42:31,  1.63it/s]
 31%|███       | 20406/65536 [3:30:17<7:39:16,  1.64it/s]
 31%|███       | 20407/65536 [3:30:17<7:34:29,  1.65it/s]
 31%|███       | 20408/65536 [3:30:18<7:43:19,  1.62it/s]
 31%|███       | 20409/65536 [3:30:18<7:43:31,  1.62it/s]
 31%|███       | 20410/65536 [3:30:19<7:45:16,  1.62it/s]
 31%|███       | 20411/65536 [3:30:20<7:50:27,  1.60it/s]
 31%|███       | 20412/65536 [3:30:20<7:40:43,  1.63it/s]
 31%|███       | 20413/65536 [3:30:21<7:59:16,  1.57it/s]
 31%|███       | 20414/65536 [3:30:22<7:51:44,  1.59it/s]
 31%|███       | 20415/65536 [3:30:22<7:39:18,  1.64it/s]
 31%|███       | 20416/65536 [3:30:23<7:31:12,  1.67it/s]
 31%|███       | 20417/65536 [3:30:23<7:25:03,  1.69it/s]
 31%|███       | 20418/65536 [3:30:24<7:36:19,  1.65it/s]
 31%|███       | 20419/65536 [3:30:25<7:46:20,  1.61it/s]
 31%|███       | 20420/65536 [3:30:25<7:39:09,  1.64it/s]
                                                         
{'loss': 3.1021, 'learning_rate': 7.214705522223582e-07, 'epoch': 1260.49}
+
 31%|███       | 20420/65536 [3:30:25<7:39:09,  1.64it/s]
 31%|███       | 20421/65536 [3:30:26<7:34:43,  1.65it/s]
 31%|███       | 20422/65536 [3:30:26<7:29:02,  1.67it/s]
 31%|███       | 20423/65536 [3:30:27<7:40:59,  1.63it/s]
 31%|███       | 20424/65536 [3:30:28<7:31:53,  1.66it/s]
 31%|███       | 20425/65536 [3:30:28<7:40:11,  1.63it/s]
 31%|███       | 20426/65536 [3:30:29<7:54:58,  1.58it/s]
 31%|███       | 20427/65536 [3:30:29<7:46:57,  1.61it/s]
 31%|███       | 20428/65536 [3:30:30<7:47:18,  1.61it/s]
 31%|███       | 20429/65536 [3:30:31<7:52:14,  1.59it/s]
 31%|███       | 20430/65536 [3:30:31<7:57:21,  1.57it/s]
 31%|███       | 20431/65536 [3:30:32<7:49:53,  1.60it/s]
 31%|███       | 20432/65536 [3:30:33<7:52:47,  1.59it/s]
 31%|███       | 20433/65536 [3:30:33<7:54:07,  1.59it/s]
 31%|███       | 20434/65536 [3:30:34<7:44:54,  1.62it/s]
 31%|███       | 20435/65536 [3:30:34<7:39:33,  1.64it/s]
 31%|███       | 20436/65536 [3:30:35<7:35:34,  1.65it/s]
 31%|███       | 20437/65536 [3:30:36<7:28:34,  1.68it/s]
 31%|███       | 20438/65536 [3:30:36<7:34:17,  1.65it/s]
 31%|███       | 20439/65536 [3:30:37<7:34:18,  1.65it/s]
 31%|███       | 20440/65536 [3:30:37<7:36:04,  1.65it/s]
                                                         
{'loss': 3.034, 'learning_rate': 7.211950532631321e-07, 'epoch': 1261.73}
+
 31%|███       | 20440/65536 [3:30:37<7:36:04,  1.65it/s]
 31%|███       | 20441/65536 [3:30:38<7:39:29,  1.64it/s]
 31%|███       | 20442/65536 [3:30:39<7:38:45,  1.64it/s]
 31%|███       | 20443/65536 [3:30:39<7:31:24,  1.66it/s]
 31%|███       | 20444/65536 [3:30:40<7:31:55,  1.66it/s]
 31%|███       | 20445/65536 [3:30:40<7:47:13,  1.61it/s]
 31%|███       | 20446/65536 [3:30:41<7:43:56,  1.62it/s]
 31%|███       | 20447/65536 [3:30:42<7:38:25,  1.64it/s]
 31%|███       | 20448/65536 [3:30:42<7:41:57,  1.63it/s]
 31%|███       | 20449/65536 [3:30:43<7:36:59,  1.64it/s]
 31%|███       | 20450/65536 [3:30:43<7:33:28,  1.66it/s]
 31%|███       | 20451/65536 [3:30:44<7:47:03,  1.61it/s]
 31%|███       | 20452/65536 [3:30:45<7:56:53,  1.58it/s]
 31%|███       | 20453/65536 [3:30:45<7:54:15,  1.58it/s]
 31%|███       | 20454/65536 [3:30:46<7:46:47,  1.61it/s]
 31%|███       | 20455/65536 [3:30:47<7:35:53,  1.65it/s]
 31%|███       | 20456/65536 [3:30:47<7:35:37,  1.65it/s]
 31%|███       | 20457/65536 [3:30:48<7:43:13,  1.62it/s]
 31%|███       | 20458/65536 [3:30:48<7:38:17,  1.64it/s]
 31%|███       | 20459/65536 [3:30:49<7:26:58,  1.68it/s]
 31%|███       | 20460/65536 [3:30:50<7:32:22,  1.66it/s]
                                                         
{'loss': 3.0488, 'learning_rate': 7.209195543039059e-07, 'epoch': 1262.96}
+
 31%|███       | 20460/65536 [3:30:50<7:32:22,  1.66it/s]
 31%|███       | 20461/65536 [3:30:50<7:46:05,  1.61it/s]
 31%|███       | 20462/65536 [3:30:51<7:50:06,  1.60it/s]
 31%|███       | 20463/65536 [3:30:52<7:52:37,  1.59it/s]
 31%|███       | 20464/65536 [3:30:52<7:39:55,  1.63it/s]
 31%|███       | 20465/65536 [3:30:53<7:40:24,  1.63it/s]
 31%|███       | 20466/65536 [3:30:53<7:37:43,  1.64it/s]
 31%|███       | 20467/65536 [3:30:54<7:56:05,  1.58it/s]
 31%|███       | 20468/65536 [3:30:55<7:56:16,  1.58it/s]
 31%|███       | 20469/65536 [3:30:55<7:48:04,  1.60it/s]
 31%|███       | 20470/65536 [3:30:56<7:42:21,  1.62it/s]
 31%|███       | 20471/65536 [3:30:56<7:39:24,  1.63it/s]
 31%|███       | 20472/65536 [3:30:57<7:36:08,  1.65it/s]
 31%|███       | 20473/65536 [3:30:58<7:39:09,  1.64it/s]
 31%|███       | 20474/65536 [3:30:58<7:27:33,  1.68it/s]
 31%|███       | 20475/65536 [3:30:59<7:25:26,  1.69it/s]
 31%|███       | 20476/65536 [3:30:59<7:29:21,  1.67it/s]
 31%|███       | 20477/65536 [3:31:00<7:36:40,  1.64it/s]
 31%|███       | 20478/65536 [3:31:01<7:39:58,  1.63it/s]
 31%|███       | 20479/65536 [3:31:01<7:36:14,  1.65it/s]
 31%|███▏      | 20480/65536 [3:31:02<7:36:34,  1.64it/s]
                                                         
{'loss': 3.1048, 'learning_rate': 7.206440553446798e-07, 'epoch': 1264.2}
+
 31%|███▏      | 20480/65536 [3:31:02<7:36:34,  1.64it/s]
 31%|███▏      | 20481/65536 [3:31:03<7:42:50,  1.62it/s]
 31%|███▏      | 20482/65536 [3:31:03<7:46:41,  1.61it/s]
 31%|███▏      | 20483/65536 [3:31:04<7:33:21,  1.66it/s]
 31%|███▏      | 20484/65536 [3:31:04<7:36:54,  1.64it/s]
 31%|███▏      | 20485/65536 [3:31:05<7:40:35,  1.63it/s]
 31%|███▏      | 20486/65536 [3:31:06<7:28:51,  1.67it/s]
 31%|███▏      | 20487/65536 [3:31:06<7:30:28,  1.67it/s]
 31%|███▏      | 20488/65536 [3:31:07<7:52:16,  1.59it/s]
 31%|███▏      | 20489/65536 [3:31:07<7:40:25,  1.63it/s]
 31%|███▏      | 20490/65536 [3:31:08<7:43:50,  1.62it/s]
 31%|███▏      | 20491/65536 [3:31:09<7:39:54,  1.63it/s]
 31%|███▏      | 20492/65536 [3:31:09<7:37:52,  1.64it/s]
 31%|███▏      | 20493/65536 [3:31:10<7:29:36,  1.67it/s]
 31%|███▏      | 20494/65536 [3:31:11<7:57:30,  1.57it/s]
 31%|███▏      | 20495/65536 [3:31:11<7:44:04,  1.62it/s]
 31%|███▏      | 20496/65536 [3:31:12<7:38:57,  1.64it/s]
 31%|███▏      | 20497/65536 [3:31:12<7:36:57,  1.64it/s]
 31%|███▏      | 20498/65536 [3:31:13<7:33:47,  1.65it/s]
 31%|███▏      | 20499/65536 [3:31:14<7:33:05,  1.66it/s]
 31%|███▏      | 20500/65536 [3:31:14<7:40:50,  1.63it/s]
                                                         
{'loss': 3.0026, 'learning_rate': 7.203685563854537e-07, 'epoch': 1265.43}
+
 31%|███▏      | 20500/65536 [3:31:14<7:40:50,  1.63it/s]
 31%|███▏      | 20501/65536 [3:31:15<7:51:18,  1.59it/s]
 31%|███▏      | 20502/65536 [3:31:15<7:45:34,  1.61it/s]
 31%|███▏      | 20503/65536 [3:31:16<7:35:44,  1.65it/s]
 31%|███▏      | 20504/65536 [3:31:17<7:24:47,  1.69it/s]
 31%|███▏      | 20505/65536 [3:31:17<7:24:48,  1.69it/s]
 31%|███▏      | 20506/65536 [3:31:18<7:29:14,  1.67it/s]
 31%|███▏      | 20507/65536 [3:31:18<7:42:58,  1.62it/s]
 31%|███▏      | 20508/65536 [3:31:19<7:38:26,  1.64it/s]
 31%|███▏      | 20509/65536 [3:31:20<7:43:19,  1.62it/s]
 31%|███▏      | 20510/65536 [3:31:20<7:49:17,  1.60it/s]
 31%|███▏      | 20511/65536 [3:31:21<7:52:00,  1.59it/s]
 31%|███▏      | 20512/65536 [3:31:22<7:50:21,  1.60it/s]
 31%|███▏      | 20513/65536 [3:31:22<7:40:52,  1.63it/s]
 31%|███▏      | 20514/65536 [3:31:23<7:35:45,  1.65it/s]
 31%|███▏      | 20515/65536 [3:31:23<7:40:45,  1.63it/s]
 31%|███▏      | 20516/65536 [3:31:24<7:45:08,  1.61it/s]
 31%|███▏      | 20517/65536 [3:31:25<7:51:14,  1.59it/s]
 31%|███▏      | 20518/65536 [3:31:25<7:45:17,  1.61it/s]
 31%|███▏      | 20519/65536 [3:31:26<7:42:53,  1.62it/s]
 31%|███▏      | 20520/65536 [3:31:26<7:41:52,  1.62it/s]
                                                         
{'loss': 2.9731, 'learning_rate': 7.200930574262274e-07, 'epoch': 1266.67}
+
 31%|███▏      | 20520/65536 [3:31:26<7:41:52,  1.62it/s]
 31%|███▏      | 20521/65536 [3:31:27<7:38:15,  1.64it/s]
 31%|███▏      | 20522/65536 [3:31:28<7:43:14,  1.62it/s]
 31%|███▏      | 20523/65536 [3:31:28<7:42:06,  1.62it/s]
 31%|███▏      | 20524/65536 [3:31:29<7:43:43,  1.62it/s]
 31%|███▏      | 20525/65536 [3:31:30<7:34:04,  1.65it/s]
 31%|███▏      | 20526/65536 [3:31:30<7:50:18,  1.60it/s]
 31%|███▏      | 20527/65536 [3:31:31<7:43:31,  1.62it/s]
 31%|███▏      | 20528/65536 [3:31:31<7:41:40,  1.62it/s]
 31%|███▏      | 20529/65536 [3:31:32<7:32:42,  1.66it/s]
 31%|███▏      | 20530/65536 [3:31:33<7:40:29,  1.63it/s]
 31%|███▏      | 20531/65536 [3:31:33<7:41:34,  1.63it/s]
 31%|███▏      | 20532/65536 [3:31:34<7:29:26,  1.67it/s]
 31%|███▏      | 20533/65536 [3:31:34<7:29:22,  1.67it/s]
 31%|███▏      | 20534/65536 [3:31:35<7:32:03,  1.66it/s]
 31%|███▏      | 20535/65536 [3:31:36<7:38:32,  1.64it/s]
 31%|███▏      | 20536/65536 [3:31:36<7:32:01,  1.66it/s]
 31%|███▏      | 20537/65536 [3:31:37<7:34:49,  1.65it/s]
 31%|███▏      | 20538/65536 [3:31:38<7:47:34,  1.60it/s]
 31%|███▏      | 20539/65536 [3:31:38<7:47:54,  1.60it/s]
 31%|███▏      | 20540/65536 [3:31:39<7:38:46,  1.63it/s]
                                                         
{'loss': 3.0784, 'learning_rate': 7.198175584670013e-07, 'epoch': 1267.9}
+
 31%|███▏      | 20540/65536 [3:31:39<7:38:46,  1.63it/s]
 31%|███▏      | 20541/65536 [3:31:39<7:41:39,  1.62it/s]
 31%|███▏      | 20542/65536 [3:31:40<7:52:59,  1.59it/s]
 31%|███▏      | 20543/65536 [3:31:41<7:54:12,  1.58it/s]
 31%|███▏      | 20544/65536 [3:31:41<7:42:05,  1.62it/s]
 31%|███▏      | 20545/65536 [3:31:42<7:38:02,  1.64it/s]
 31%|███▏      | 20546/65536 [3:31:42<7:28:20,  1.67it/s]
 31%|███▏      | 20547/65536 [3:31:43<7:26:36,  1.68it/s]
 31%|███▏      | 20548/65536 [3:31:44<7:21:19,  1.70it/s]
 31%|███▏      | 20549/65536 [3:31:44<7:22:07,  1.70it/s]
 31%|███▏      | 20550/65536 [3:31:45<7:25:13,  1.68it/s]
 31%|███▏      | 20551/65536 [3:31:45<7:27:56,  1.67it/s]
 31%|███▏      | 20552/65536 [3:31:46<7:39:01,  1.63it/s]
 31%|███▏      | 20553/65536 [3:31:47<7:46:01,  1.61it/s]
 31%|███▏      | 20554/65536 [3:31:47<7:36:46,  1.64it/s]
 31%|███▏      | 20555/65536 [3:31:48<7:38:36,  1.63it/s]
 31%|███▏      | 20556/65536 [3:31:48<7:29:11,  1.67it/s]
 31%|███▏      | 20557/65536 [3:31:49<7:55:19,  1.58it/s]
 31%|███▏      | 20558/65536 [3:31:50<7:58:17,  1.57it/s]
 31%|███▏      | 20559/65536 [3:31:50<7:45:46,  1.61it/s]
 31%|███▏      | 20560/65536 [3:31:51<7:39:11,  1.63it/s]
                                                         
{'loss': 3.0579, 'learning_rate': 7.195420595077751e-07, 'epoch': 1269.14}
+
 31%|███▏      | 20560/65536 [3:31:51<7:39:11,  1.63it/s]
 31%|███▏      | 20561/65536 [3:31:52<7:45:26,  1.61it/s]
 31%|███▏      | 20562/65536 [3:31:52<7:48:09,  1.60it/s]
 31%|███▏      | 20563/65536 [3:31:53<7:36:31,  1.64it/s]
 31%|███▏      | 20564/65536 [3:31:53<7:36:49,  1.64it/s]
 31%|███▏      | 20565/65536 [3:31:54<7:33:05,  1.65it/s]
 31%|███▏      | 20566/65536 [3:31:55<7:31:52,  1.66it/s]
 31%|███▏      | 20567/65536 [3:31:55<7:31:05,  1.66it/s]
 31%|███▏      | 20568/65536 [3:31:56<7:46:23,  1.61it/s]
 31%|███▏      | 20569/65536 [3:31:56<7:38:39,  1.63it/s]
 31%|███▏      | 20570/65536 [3:31:57<7:28:58,  1.67it/s]
 31%|███▏      | 20571/65536 [3:31:58<7:42:24,  1.62it/s]
 31%|███▏      | 20572/65536 [3:31:58<7:39:45,  1.63it/s]
 31%|███▏      | 20573/65536 [3:31:59<7:32:03,  1.66it/s]
 31%|███▏      | 20574/65536 [3:31:59<7:28:22,  1.67it/s]
 31%|███▏      | 20575/65536 [3:32:00<7:48:53,  1.60it/s]
 31%|███▏      | 20576/65536 [3:32:01<7:47:02,  1.60it/s]
 31%|███▏      | 20577/65536 [3:32:01<7:48:50,  1.60it/s]
 31%|███▏      | 20578/65536 [3:32:02<7:41:58,  1.62it/s]
 31%|███▏      | 20579/65536 [3:32:03<7:24:56,  1.68it/s]
 31%|███▏      | 20580/65536 [3:32:03<7:27:10,  1.68it/s]
                                                         
{'loss': 3.1232, 'learning_rate': 7.19266560548549e-07, 'epoch': 1270.37}
+
 31%|███▏      | 20580/65536 [3:32:03<7:27:10,  1.68it/s]
 31%|███▏      | 20581/65536 [3:32:04<7:35:31,  1.64it/s]
 31%|███▏      | 20582/65536 [3:32:04<7:37:50,  1.64it/s]
 31%|███▏      | 20583/65536 [3:32:05<7:28:36,  1.67it/s]
 31%|███▏      | 20584/65536 [3:32:06<7:26:36,  1.68it/s]
 31%|███▏      | 20585/65536 [3:32:06<7:27:31,  1.67it/s]
 31%|███▏      | 20586/65536 [3:32:07<7:29:39,  1.67it/s]
 31%|███▏      | 20587/65536 [3:32:07<7:24:57,  1.68it/s]
 31%|███▏      | 20588/65536 [3:32:08<7:24:31,  1.69it/s]
 31%|███▏      | 20589/65536 [3:32:09<7:39:38,  1.63it/s]
 31%|███▏      | 20590/65536 [3:32:09<7:43:54,  1.61it/s]
 31%|███▏      | 20591/65536 [3:32:10<8:09:24,  1.53it/s]
 31%|███▏      | 20592/65536 [3:32:11<7:57:57,  1.57it/s]
 31%|███▏      | 20593/65536 [3:32:11<8:09:15,  1.53it/s]
 31%|███▏      | 20594/65536 [3:32:12<7:56:14,  1.57it/s]
 31%|███▏      | 20595/65536 [3:32:12<7:52:31,  1.59it/s]
 31%|███▏      | 20596/65536 [3:32:13<7:39:53,  1.63it/s]
 31%|███▏      | 20597/65536 [3:32:14<7:37:36,  1.64it/s]
 31%|███▏      | 20598/65536 [3:32:14<7:31:34,  1.66it/s]
 31%|███▏      | 20599/65536 [3:32:15<7:24:18,  1.69it/s]
 31%|███▏      | 20600/65536 [3:32:15<7:23:27,  1.69it/s]
                                                         
{'loss': 3.0752, 'learning_rate': 7.189910615893227e-07, 'epoch': 1271.6}
+
 31%|███▏      | 20600/65536 [3:32:15<7:23:27,  1.69it/s]
 31%|███▏      | 20601/65536 [3:32:16<7:36:52,  1.64it/s]
 31%|███▏      | 20602/65536 [3:32:17<7:32:00,  1.66it/s]
 31%|███▏      | 20603/65536 [3:32:17<7:26:47,  1.68it/s]
 31%|███▏      | 20604/65536 [3:32:18<7:27:25,  1.67it/s]
 31%|███▏      | 20605/65536 [3:32:18<7:28:19,  1.67it/s]
 31%|███▏      | 20606/65536 [3:32:19<7:35:32,  1.64it/s]
 31%|███▏      | 20607/65536 [3:32:20<7:58:22,  1.57it/s]
 31%|███▏      | 20608/65536 [3:32:20<7:47:34,  1.60it/s]
 31%|███▏      | 20609/65536 [3:32:21<7:47:22,  1.60it/s]
 31%|███▏      | 20610/65536 [3:32:22<7:40:17,  1.63it/s]
 31%|███▏      | 20611/65536 [3:32:22<7:57:26,  1.57it/s]
 31%|███▏      | 20612/65536 [3:32:23<7:39:55,  1.63it/s]
 31%|███▏      | 20613/65536 [3:32:23<7:36:19,  1.64it/s]
 31%|███▏      | 20614/65536 [3:32:24<7:31:09,  1.66it/s]
 31%|███▏      | 20615/65536 [3:32:25<7:43:32,  1.62it/s]
 31%|███▏      | 20616/65536 [3:32:25<7:47:10,  1.60it/s]
 31%|███▏      | 20617/65536 [3:32:26<7:42:25,  1.62it/s]
 31%|███▏      | 20618/65536 [3:32:26<7:37:07,  1.64it/s]
 31%|███▏      | 20619/65536 [3:32:27<7:41:29,  1.62it/s]
 31%|███▏      | 20620/65536 [3:32:28<7:35:42,  1.64it/s]
                                                         
{'loss': 2.9859, 'learning_rate': 7.187155626300966e-07, 'epoch': 1272.84}
+
 31%|███▏      | 20620/65536 [3:32:28<7:35:42,  1.64it/s]
 31%|███▏      | 20621/65536 [3:32:28<7:37:38,  1.64it/s]
 31%|███▏      | 20622/65536 [3:32:29<7:37:24,  1.64it/s]
 31%|███▏      | 20623/65536 [3:32:30<7:48:46,  1.60it/s]
 31%|███▏      | 20624/65536 [3:32:30<7:43:46,  1.61it/s]
 31%|███▏      | 20625/65536 [3:32:31<7:43:16,  1.62it/s]
 31%|███▏      | 20626/65536 [3:32:31<7:45:42,  1.61it/s]
 31%|███▏      | 20627/65536 [3:32:32<7:39:31,  1.63it/s]
 31%|███▏      | 20628/65536 [3:32:33<7:44:17,  1.61it/s]
 31%|███▏      | 20629/65536 [3:32:33<7:41:01,  1.62it/s]
 31%|███▏      | 20630/65536 [3:32:34<7:37:26,  1.64it/s]
 31%|███▏      | 20631/65536 [3:32:35<7:42:12,  1.62it/s]
 31%|███▏      | 20632/65536 [3:32:35<7:37:12,  1.64it/s]
 31%|███▏      | 20633/65536 [3:32:36<7:38:38,  1.63it/s]
 31%|███▏      | 20634/65536 [3:32:36<7:36:12,  1.64it/s]
 31%|███▏      | 20635/65536 [3:32:37<7:30:34,  1.66it/s]
 31%|███▏      | 20636/65536 [3:32:37<7:22:26,  1.69it/s]
 31%|███▏      | 20637/65536 [3:32:38<7:20:44,  1.70it/s]
 31%|███▏      | 20638/65536 [3:32:39<7:31:03,  1.66it/s]
 31%|███▏      | 20639/65536 [3:32:39<7:47:10,  1.60it/s]
 31%|███▏      | 20640/65536 [3:32:40<7:42:57,  1.62it/s]
                                                         
{'loss': 3.0488, 'learning_rate': 7.184400636708705e-07, 'epoch': 1274.07}
+
 31%|███▏      | 20640/65536 [3:32:40<7:42:57,  1.62it/s]
 31%|███▏      | 20641/65536 [3:32:41<7:47:48,  1.60it/s]
 31%|███▏      | 20642/65536 [3:32:41<7:39:31,  1.63it/s]
 31%|███▏      | 20643/65536 [3:32:42<7:32:13,  1.65it/s]
 32%|███▏      | 20644/65536 [3:32:42<7:39:04,  1.63it/s]
 32%|███▏      | 20645/65536 [3:32:43<7:47:21,  1.60it/s]
 32%|███▏      | 20646/65536 [3:32:44<7:44:34,  1.61it/s]
 32%|███▏      | 20647/65536 [3:32:44<7:34:45,  1.65it/s]
 32%|███▏      | 20648/65536 [3:32:45<7:46:00,  1.61it/s]
 32%|███▏      | 20649/65536 [3:32:46<7:41:17,  1.62it/s]
 32%|███▏      | 20650/65536 [3:32:46<7:50:55,  1.59it/s]
 32%|███▏      | 20651/65536 [3:32:47<7:36:10,  1.64it/s]
 32%|███▏      | 20652/65536 [3:32:47<7:27:54,  1.67it/s]
 32%|███▏      | 20653/65536 [3:32:48<7:32:27,  1.65it/s]
 32%|███▏      | 20654/65536 [3:32:49<7:30:27,  1.66it/s]
 32%|███▏      | 20655/65536 [3:32:49<7:27:27,  1.67it/s]
 32%|███▏      | 20656/65536 [3:32:50<7:50:07,  1.59it/s]
 32%|███▏      | 20657/65536 [3:32:50<7:52:43,  1.58it/s]
 32%|███▏      | 20658/65536 [3:32:51<7:53:30,  1.58it/s]
 32%|███▏      | 20659/65536 [3:32:52<7:52:12,  1.58it/s]
 32%|███▏      | 20660/65536 [3:32:52<7:40:02,  1.63it/s]
                                                         
{'loss': 3.1021, 'learning_rate': 7.181645647116444e-07, 'epoch': 1275.31}
+
 32%|███▏      | 20660/65536 [3:32:52<7:40:02,  1.63it/s]
 32%|███▏      | 20661/65536 [3:32:53<7:41:14,  1.62it/s]
 32%|███▏      | 20662/65536 [3:32:53<7:34:50,  1.64it/s]
 32%|███▏      | 20663/65536 [3:32:54<7:37:59,  1.63it/s]
 32%|███▏      | 20664/65536 [3:32:55<7:39:09,  1.63it/s]
 32%|███▏      | 20665/65536 [3:32:55<7:29:44,  1.66it/s]
 32%|███▏      | 20666/65536 [3:32:56<7:32:51,  1.65it/s]
 32%|███▏      | 20667/65536 [3:32:56<7:22:17,  1.69it/s]
 32%|███▏      | 20668/65536 [3:32:57<7:24:05,  1.68it/s]
 32%|███▏      | 20669/65536 [3:32:58<7:23:47,  1.69it/s]
 32%|███▏      | 20670/65536 [3:32:58<7:21:20,  1.69it/s]
 32%|███▏      | 20671/65536 [3:32:59<7:23:21,  1.69it/s]
 32%|███▏      | 20672/65536 [3:33:00<7:43:22,  1.61it/s]
 32%|███▏      | 20673/65536 [3:33:00<7:41:30,  1.62it/s]
 32%|███▏      | 20674/65536 [3:33:01<7:39:03,  1.63it/s]
 32%|███▏      | 20675/65536 [3:33:01<7:32:47,  1.65it/s]
 32%|███▏      | 20676/65536 [3:33:02<7:34:41,  1.64it/s]
 32%|███▏      | 20677/65536 [3:33:03<7:25:14,  1.68it/s]
 32%|███▏      | 20678/65536 [3:33:03<7:32:10,  1.65it/s]
 32%|███▏      | 20679/65536 [3:33:04<7:36:29,  1.64it/s]
 32%|███▏      | 20680/65536 [3:33:04<7:34:21,  1.65it/s]
                                                         
{'loss': 3.0604, 'learning_rate': 7.178890657524183e-07, 'epoch': 1276.54}
+
 32%|███▏      | 20680/65536 [3:33:04<7:34:21,  1.65it/s]
 32%|███▏      | 20681/65536 [3:33:05<7:23:07,  1.69it/s]
 32%|███▏      | 20682/65536 [3:33:06<7:40:31,  1.62it/s]
 32%|███▏      | 20683/65536 [3:33:06<7:41:55,  1.62it/s]
 32%|███▏      | 20684/65536 [3:33:07<7:34:29,  1.64it/s]
 32%|███▏      | 20685/65536 [3:33:07<7:35:07,  1.64it/s]
 32%|███▏      | 20686/65536 [3:33:08<7:46:58,  1.60it/s]
 32%|███▏      | 20687/65536 [3:33:09<7:38:24,  1.63it/s]
 32%|███▏      | 20688/65536 [3:33:09<7:54:54,  1.57it/s]
 32%|███▏      | 20689/65536 [3:33:10<7:42:00,  1.62it/s]
 32%|███▏      | 20690/65536 [3:33:11<7:42:56,  1.61it/s]
 32%|███▏      | 20691/65536 [3:33:11<7:44:48,  1.61it/s]
 32%|███▏      | 20692/65536 [3:33:12<7:50:22,  1.59it/s]
 32%|███▏      | 20693/65536 [3:33:12<7:37:32,  1.63it/s]
 32%|███▏      | 20694/65536 [3:33:13<7:23:07,  1.69it/s]
 32%|███▏      | 20695/65536 [3:33:14<7:24:35,  1.68it/s]
 32%|███▏      | 20696/65536 [3:33:14<7:36:50,  1.64it/s]
 32%|███▏      | 20697/65536 [3:33:15<7:23:58,  1.68it/s]
 32%|███▏      | 20698/65536 [3:33:15<7:19:09,  1.70it/s]
 32%|███▏      | 20699/65536 [3:33:16<7:22:11,  1.69it/s]
 32%|███▏      | 20700/65536 [3:33:17<7:28:03,  1.67it/s]
                                                         
{'loss': 3.0458, 'learning_rate': 7.176135667931921e-07, 'epoch': 1277.78}
+
 32%|███▏      | 20700/65536 [3:33:17<7:28:03,  1.67it/s]
 32%|███▏      | 20701/65536 [3:33:17<7:38:03,  1.63it/s]
 32%|███▏      | 20702/65536 [3:33:18<7:49:59,  1.59it/s]
 32%|███▏      | 20703/65536 [3:33:18<7:44:38,  1.61it/s]
 32%|███▏      | 20704/65536 [3:33:19<7:51:54,  1.58it/s]
 32%|███▏      | 20705/65536 [3:33:20<7:43:05,  1.61it/s]
 32%|███▏      | 20706/65536 [3:33:20<7:37:33,  1.63it/s]
 32%|███▏      | 20707/65536 [3:33:21<7:37:40,  1.63it/s]
 32%|███▏      | 20708/65536 [3:33:22<7:38:39,  1.63it/s]
 32%|███▏      | 20709/65536 [3:33:22<7:37:08,  1.63it/s]
 32%|███▏      | 20710/65536 [3:33:23<7:36:16,  1.64it/s]
 32%|███▏      | 20711/65536 [3:33:23<7:29:23,  1.66it/s]
 32%|███▏      | 20712/65536 [3:33:24<7:41:32,  1.62it/s]
 32%|███▏      | 20713/65536 [3:33:25<7:33:23,  1.65it/s]
 32%|███▏      | 20714/65536 [3:33:25<7:39:42,  1.63it/s]
 32%|███▏      | 20715/65536 [3:33:26<7:34:13,  1.64it/s]
 32%|███▏      | 20716/65536 [3:33:26<7:38:30,  1.63it/s]
 32%|███▏      | 20717/65536 [3:33:27<7:39:06,  1.63it/s]
 32%|███▏      | 20718/65536 [3:33:28<7:35:20,  1.64it/s]
 32%|███▏      | 20719/65536 [3:33:28<7:30:45,  1.66it/s]
 32%|███▏      | 20720/65536 [3:33:29<7:45:05,  1.61it/s]
                                                         
{'loss': 3.0895, 'learning_rate': 7.17338067833966e-07, 'epoch': 1279.01}
+
 32%|███▏      | 20720/65536 [3:33:29<7:45:05,  1.61it/s]
 32%|███▏      | 20721/65536 [3:33:29<7:37:33,  1.63it/s]
 32%|███▏      | 20722/65536 [3:33:30<7:39:39,  1.62it/s]
 32%|███▏      | 20723/65536 [3:33:31<8:04:17,  1.54it/s]
 32%|███▏      | 20724/65536 [3:33:31<7:51:35,  1.58it/s]
 32%|███▏      | 20725/65536 [3:33:32<7:43:29,  1.61it/s]
 32%|███▏      | 20726/65536 [3:33:33<7:35:59,  1.64it/s]
 32%|███▏      | 20727/65536 [3:33:33<7:29:00,  1.66it/s]
 32%|███▏      | 20728/65536 [3:33:34<7:35:04,  1.64it/s]
 32%|███▏      | 20729/65536 [3:33:34<7:38:04,  1.63it/s]
 32%|███▏      | 20730/65536 [3:33:35<7:30:29,  1.66it/s]
 32%|███▏      | 20731/65536 [3:33:36<7:39:58,  1.62it/s]
 32%|███▏      | 20732/65536 [3:33:36<7:31:47,  1.65it/s]
 32%|███▏      | 20733/65536 [3:33:37<7:35:22,  1.64it/s]
 32%|███▏      | 20734/65536 [3:33:37<7:42:07,  1.62it/s]
 32%|███▏      | 20735/65536 [3:33:38<7:36:31,  1.64it/s]
 32%|███▏      | 20736/65536 [3:33:39<7:35:11,  1.64it/s]
 32%|███▏      | 20737/65536 [3:33:39<7:49:31,  1.59it/s]
 32%|███▏      | 20738/65536 [3:33:40<7:53:28,  1.58it/s]
 32%|███▏      | 20739/65536 [3:33:41<7:45:53,  1.60it/s]
 32%|███▏      | 20740/65536 [3:33:41<7:31:56,  1.65it/s]
                                                         
{'loss': 2.9886, 'learning_rate': 7.170625688747399e-07, 'epoch': 1280.25}
+
 32%|███▏      | 20740/65536 [3:33:41<7:31:56,  1.65it/s]
 32%|███▏      | 20741/65536 [3:33:42<7:35:14,  1.64it/s]
 32%|███▏      | 20742/65536 [3:33:42<7:48:46,  1.59it/s]
 32%|███▏      | 20743/65536 [3:33:43<7:46:47,  1.60it/s]
 32%|███▏      | 20744/65536 [3:33:44<7:40:24,  1.62it/s]
 32%|███▏      | 20745/65536 [3:33:44<7:32:19,  1.65it/s]
 32%|███▏      | 20746/65536 [3:33:45<7:32:11,  1.65it/s]
 32%|███▏      | 20747/65536 [3:33:45<7:35:08,  1.64it/s]
 32%|███▏      | 20748/65536 [3:33:46<7:36:27,  1.64it/s]
 32%|███▏      | 20749/65536 [3:33:47<7:33:18,  1.65it/s]
 32%|███▏      | 20750/65536 [3:33:47<7:22:52,  1.69it/s]
 32%|███▏      | 20751/65536 [3:33:48<7:32:03,  1.65it/s]
 32%|███▏      | 20752/65536 [3:33:49<7:33:04,  1.65it/s]
 32%|███▏      | 20753/65536 [3:33:49<7:44:57,  1.61it/s]
 32%|███▏      | 20754/65536 [3:33:50<7:47:34,  1.60it/s]
 32%|███▏      | 20755/65536 [3:33:50<7:37:59,  1.63it/s]
 32%|███▏      | 20756/65536 [3:33:51<7:38:54,  1.63it/s]
 32%|███▏      | 20757/65536 [3:33:52<7:37:35,  1.63it/s]
 32%|███▏      | 20758/65536 [3:33:52<7:47:54,  1.59it/s]
 32%|███▏      | 20759/65536 [3:33:53<7:46:18,  1.60it/s]
 32%|███▏      | 20760/65536 [3:33:54<7:53:06,  1.58it/s]
                                                         
{'loss': 3.086, 'learning_rate': 7.167870699155137e-07, 'epoch': 1281.48}
+
 32%|███▏      | 20760/65536 [3:33:54<7:53:06,  1.58it/s]
 32%|███▏      | 20761/65536 [3:33:54<7:43:20,  1.61it/s]
 32%|███▏      | 20762/65536 [3:33:55<7:29:51,  1.66it/s]
 32%|███▏      | 20763/65536 [3:33:55<7:25:04,  1.68it/s]
 32%|███▏      | 20764/65536 [3:33:56<7:31:02,  1.65it/s]
 32%|███▏      | 20765/65536 [3:33:57<7:30:06,  1.66it/s]
 32%|███▏      | 20766/65536 [3:33:57<7:37:06,  1.63it/s]
 32%|███▏      | 20767/65536 [3:33:58<7:34:18,  1.64it/s]
 32%|███▏      | 20768/65536 [3:33:58<7:31:15,  1.65it/s]
 32%|███▏      | 20769/65536 [3:33:59<7:42:48,  1.61it/s]
 32%|███▏      | 20770/65536 [3:34:00<7:57:18,  1.56it/s]
 32%|███▏      | 20771/65536 [3:34:00<7:46:57,  1.60it/s]
 32%|███▏      | 20772/65536 [3:34:01<7:40:46,  1.62it/s]
 32%|███▏      | 20773/65536 [3:34:01<7:35:18,  1.64it/s]
 32%|███▏      | 20774/65536 [3:34:02<7:33:42,  1.64it/s]
 32%|███▏      | 20775/65536 [3:34:03<7:35:24,  1.64it/s]
 32%|███▏      | 20776/65536 [3:34:03<7:38:39,  1.63it/s]
 32%|███▏      | 20777/65536 [3:34:04<7:29:39,  1.66it/s]
 32%|███▏      | 20778/65536 [3:34:04<7:19:09,  1.70it/s]
 32%|███▏      | 20779/65536 [3:34:05<7:22:46,  1.68it/s]
 32%|███▏      | 20780/65536 [3:34:06<7:27:32,  1.67it/s]
                                                         
{'loss': 3.006, 'learning_rate': 7.165115709562876e-07, 'epoch': 1282.72}
+
 32%|███▏      | 20780/65536 [3:34:06<7:27:32,  1.67it/s]
 32%|███▏      | 20781/65536 [3:34:06<7:24:46,  1.68it/s]
 32%|███▏      | 20782/65536 [3:34:07<7:30:07,  1.66it/s]
 32%|███▏      | 20783/65536 [3:34:07<7:24:55,  1.68it/s]
 32%|███▏      | 20784/65536 [3:34:08<7:40:41,  1.62it/s]
 32%|███▏      | 20785/65536 [3:34:09<7:48:23,  1.59it/s]
 32%|███▏      | 20786/65536 [3:34:09<7:37:25,  1.63it/s]
 32%|███▏      | 20787/65536 [3:34:10<7:39:15,  1.62it/s]
 32%|███▏      | 20788/65536 [3:34:11<7:33:06,  1.65it/s]
 32%|███▏      | 20789/65536 [3:34:11<7:28:03,  1.66it/s]
 32%|███▏      | 20790/65536 [3:34:12<7:46:02,  1.60it/s]
 32%|███▏      | 20791/65536 [3:34:12<7:40:56,  1.62it/s]
 32%|███▏      | 20792/65536 [3:34:13<7:42:24,  1.61it/s]
 32%|███▏      | 20793/65536 [3:34:14<7:35:50,  1.64it/s]
 32%|███▏      | 20794/65536 [3:34:14<7:34:40,  1.64it/s]
 32%|███▏      | 20795/65536 [3:34:15<7:34:19,  1.64it/s]
 32%|███▏      | 20796/65536 [3:34:15<7:33:42,  1.64it/s]
 32%|███▏      | 20797/65536 [3:34:16<7:34:35,  1.64it/s]
 32%|███▏      | 20798/65536 [3:34:17<7:24:37,  1.68it/s]
 32%|███▏      | 20799/65536 [3:34:17<7:23:36,  1.68it/s]
 32%|███▏      | 20800/65536 [3:34:18<7:24:40,  1.68it/s]
                                                         
{'loss': 3.0638, 'learning_rate': 7.162360719970613e-07, 'epoch': 1283.95}
+
 32%|███▏      | 20800/65536 [3:34:18<7:24:40,  1.68it/s]
 32%|███▏      | 20801/65536 [3:34:19<7:53:51,  1.57it/s]
 32%|███▏      | 20802/65536 [3:34:19<7:38:10,  1.63it/s]
 32%|███▏      | 20803/65536 [3:34:20<7:38:01,  1.63it/s]
 32%|███▏      | 20804/65536 [3:34:20<7:33:03,  1.65it/s]
 32%|███▏      | 20805/65536 [3:34:21<7:39:16,  1.62it/s]
 32%|███▏      | 20806/65536 [3:34:22<7:47:18,  1.60it/s]
 32%|███▏      | 20807/65536 [3:34:22<7:47:03,  1.60it/s]
 32%|███▏      | 20808/65536 [3:34:23<7:42:19,  1.61it/s]
 32%|███▏      | 20809/65536 [3:34:23<7:41:29,  1.62it/s]
 32%|███▏      | 20810/65536 [3:34:24<8:02:38,  1.54it/s]
 32%|███▏      | 20811/65536 [3:34:25<7:50:09,  1.59it/s]
 32%|███▏      | 20812/65536 [3:34:25<7:43:02,  1.61it/s]
 32%|███▏      | 20813/65536 [3:34:26<7:40:39,  1.62it/s]
 32%|███▏      | 20814/65536 [3:34:27<7:33:43,  1.64it/s]
 32%|███▏      | 20815/65536 [3:34:27<7:28:51,  1.66it/s]
 32%|███▏      | 20816/65536 [3:34:28<7:32:48,  1.65it/s]
 32%|███▏      | 20817/65536 [3:34:28<7:24:06,  1.68it/s]
 32%|███▏      | 20818/65536 [3:34:29<7:41:44,  1.61it/s]
 32%|███▏      | 20819/65536 [3:34:30<7:39:55,  1.62it/s]
 32%|███▏      | 20820/65536 [3:34:30<7:51:16,  1.58it/s]
                                                         
{'loss': 3.0063, 'learning_rate': 7.159605730378352e-07, 'epoch': 1285.19}
+
 32%|███▏      | 20820/65536 [3:34:30<7:51:16,  1.58it/s]
 32%|███▏      | 20821/65536 [3:34:31<7:37:14,  1.63it/s]
 32%|███▏      | 20822/65536 [3:34:32<7:43:30,  1.61it/s]
 32%|███▏      | 20823/65536 [3:34:32<7:39:59,  1.62it/s]
 32%|███▏      | 20824/65536 [3:34:33<7:40:02,  1.62it/s]
 32%|███▏      | 20825/65536 [3:34:33<7:28:51,  1.66it/s]
 32%|███▏      | 20826/65536 [3:34:34<7:43:01,  1.61it/s]
 32%|███▏      | 20827/65536 [3:34:35<7:45:48,  1.60it/s]
 32%|███▏      | 20828/65536 [3:34:35<7:43:06,  1.61it/s]
 32%|███▏      | 20829/65536 [3:34:36<7:38:23,  1.63it/s]
 32%|███▏      | 20830/65536 [3:34:36<7:43:43,  1.61it/s]
 32%|███▏      | 20831/65536 [3:34:37<7:43:51,  1.61it/s]
 32%|███▏      | 20832/65536 [3:34:38<7:35:38,  1.64it/s]
 32%|███▏      | 20833/65536 [3:34:38<7:24:14,  1.68it/s]
 32%|███▏      | 20834/65536 [3:34:39<7:37:37,  1.63it/s]
 32%|███▏      | 20835/65536 [3:34:40<7:46:23,  1.60it/s]
 32%|███▏      | 20836/65536 [3:34:40<7:49:02,  1.59it/s]
 32%|███▏      | 20837/65536 [3:34:41<7:40:44,  1.62it/s]
 32%|███▏      | 20838/65536 [3:34:41<7:32:49,  1.65it/s]
 32%|███▏      | 20839/65536 [3:34:42<7:41:53,  1.61it/s]
 32%|███▏      | 20840/65536 [3:34:43<7:38:37,  1.62it/s]
                                                         
{'loss': 3.0392, 'learning_rate': 7.15685074078609e-07, 'epoch': 1286.42}
+
 32%|███▏      | 20840/65536 [3:34:43<7:38:37,  1.62it/s]
 32%|███▏      | 20841/65536 [3:34:43<7:31:56,  1.65it/s]
 32%|███▏      | 20842/65536 [3:34:44<7:30:22,  1.65it/s]
 32%|███▏      | 20843/65536 [3:34:44<7:28:05,  1.66it/s]
 32%|███▏      | 20844/65536 [3:34:45<7:39:57,  1.62it/s]
 32%|███▏      | 20845/65536 [3:34:46<7:38:51,  1.62it/s]
 32%|███▏      | 20846/65536 [3:34:46<7:30:07,  1.65it/s]
 32%|███▏      | 20847/65536 [3:34:47<7:24:09,  1.68it/s]
 32%|███▏      | 20848/65536 [3:34:47<7:30:44,  1.65it/s]
 32%|███▏      | 20849/65536 [3:34:48<7:35:03,  1.64it/s]
 32%|███▏      | 20850/65536 [3:34:49<7:52:43,  1.58it/s]
 32%|███▏      | 20851/65536 [3:34:49<8:03:55,  1.54it/s]
 32%|███▏      | 20852/65536 [3:34:50<7:44:30,  1.60it/s]
 32%|███▏      | 20853/65536 [3:34:51<7:29:17,  1.66it/s]
 32%|███▏      | 20854/65536 [3:34:51<7:23:52,  1.68it/s]
 32%|███▏      | 20855/65536 [3:34:52<7:35:50,  1.63it/s]
 32%|███▏      | 20856/65536 [3:34:52<7:28:54,  1.66it/s]
 32%|███▏      | 20857/65536 [3:34:53<7:25:11,  1.67it/s]
 32%|███▏      | 20858/65536 [3:34:54<7:25:34,  1.67it/s]
 32%|███▏      | 20859/65536 [3:34:54<7:35:58,  1.63it/s]
 32%|███▏      | 20860/65536 [3:34:55<7:45:31,  1.60it/s]
                                                         
{'loss': 3.0259, 'learning_rate': 7.154095751193828e-07, 'epoch': 1287.65}
+
 32%|███▏      | 20860/65536 [3:34:55<7:45:31,  1.60it/s]
 32%|███▏      | 20861/65536 [3:34:55<7:35:41,  1.63it/s]
 32%|███▏      | 20862/65536 [3:34:56<7:37:02,  1.63it/s]
 32%|███▏      | 20863/65536 [3:34:57<7:37:39,  1.63it/s]
 32%|███▏      | 20864/65536 [3:34:57<7:37:48,  1.63it/s]
 32%|███▏      | 20865/65536 [3:34:58<7:37:17,  1.63it/s]
 32%|███▏      | 20866/65536 [3:34:59<7:51:27,  1.58it/s]
 32%|███▏      | 20867/65536 [3:34:59<7:39:40,  1.62it/s]
 32%|███▏      | 20868/65536 [3:35:00<7:41:54,  1.61it/s]
 32%|███▏      | 20869/65536 [3:35:00<7:33:42,  1.64it/s]
 32%|███▏      | 20870/65536 [3:35:01<7:28:51,  1.66it/s]
 32%|███▏      | 20871/65536 [3:35:02<7:48:09,  1.59it/s]
 32%|███▏      | 20872/65536 [3:35:02<7:38:31,  1.62it/s]
 32%|███▏      | 20873/65536 [3:35:03<7:38:53,  1.62it/s]
 32%|███▏      | 20874/65536 [3:35:03<7:29:18,  1.66it/s]
 32%|███▏      | 20875/65536 [3:35:04<7:19:40,  1.69it/s]
 32%|███▏      | 20876/65536 [3:35:05<7:16:26,  1.71it/s]
 32%|███▏      | 20877/65536 [3:35:05<7:12:27,  1.72it/s]
 32%|███▏      | 20878/65536 [3:35:06<7:16:26,  1.71it/s]
 32%|███▏      | 20879/65536 [3:35:06<7:48:27,  1.59it/s]
 32%|███▏      | 20880/65536 [3:35:07<7:47:52,  1.59it/s]
                                                         
{'loss': 3.0477, 'learning_rate': 7.151340761601567e-07, 'epoch': 1288.89}
+
 32%|███▏      | 20880/65536 [3:35:07<7:47:52,  1.59it/s]
 32%|███▏      | 20881/65536 [3:35:08<7:47:44,  1.59it/s]
 32%|███▏      | 20882/65536 [3:35:08<8:01:15,  1.55it/s]
 32%|███▏      | 20883/65536 [3:35:09<7:56:35,  1.56it/s]
 32%|███▏      | 20884/65536 [3:35:10<7:49:47,  1.58it/s]
 32%|███▏      | 20885/65536 [3:35:10<7:37:43,  1.63it/s]
 32%|███▏      | 20886/65536 [3:35:11<7:25:08,  1.67it/s]
 32%|███▏      | 20887/65536 [3:35:11<7:20:25,  1.69it/s]
 32%|███▏      | 20888/65536 [3:35:12<7:33:03,  1.64it/s]
 32%|███▏      | 20889/65536 [3:35:13<7:32:07,  1.65it/s]
 32%|███▏      | 20890/65536 [3:35:13<7:35:43,  1.63it/s]
 32%|███▏      | 20891/65536 [3:35:14<7:32:53,  1.64it/s]
 32%|███▏      | 20892/65536 [3:35:14<7:27:04,  1.66it/s]
 32%|███▏      | 20893/65536 [3:35:15<7:55:50,  1.56it/s]
 32%|███▏      | 20894/65536 [3:35:16<8:02:35,  1.54it/s]
 32%|███▏      | 20895/65536 [3:35:16<7:57:58,  1.56it/s]
 32%|███▏      | 20896/65536 [3:35:17<7:52:00,  1.58it/s]
 32%|███▏      | 20897/65536 [3:35:18<7:38:36,  1.62it/s]
 32%|███▏      | 20898/65536 [3:35:18<7:38:26,  1.62it/s]
 32%|███▏      | 20899/65536 [3:35:19<7:56:23,  1.56it/s]
 32%|███▏      | 20900/65536 [3:35:19<7:40:20,  1.62it/s]
                                                         
{'loss': 3.0454, 'learning_rate': 7.148585772009305e-07, 'epoch': 1290.12}
+
 32%|███▏      | 20900/65536 [3:35:19<7:40:20,  1.62it/s]
 32%|███▏      | 20901/65536 [3:35:20<7:43:27,  1.61it/s]
 32%|███▏      | 20902/65536 [3:35:21<7:35:28,  1.63it/s]
 32%|███▏      | 20903/65536 [3:35:21<7:33:43,  1.64it/s]
 32%|███▏      | 20904/65536 [3:35:22<7:29:35,  1.65it/s]
 32%|███▏      | 20905/65536 [3:35:23<7:39:12,  1.62it/s]
 32%|███▏      | 20906/65536 [3:35:23<7:40:03,  1.62it/s]
 32%|███▏      | 20907/65536 [3:35:24<7:30:53,  1.65it/s]
 32%|███▏      | 20908/65536 [3:35:24<7:49:54,  1.58it/s]
 32%|███▏      | 20909/65536 [3:35:25<7:34:15,  1.64it/s]
 32%|███▏      | 20910/65536 [3:35:26<7:22:30,  1.68it/s]
 32%|███▏      | 20911/65536 [3:35:26<7:35:49,  1.63it/s]
 32%|███▏      | 20912/65536 [3:35:27<7:26:59,  1.66it/s]
 32%|███▏      | 20913/65536 [3:35:27<7:31:42,  1.65it/s]
 32%|███▏      | 20914/65536 [3:35:28<7:24:12,  1.67it/s]
 32%|███▏      | 20915/65536 [3:35:29<7:52:30,  1.57it/s]
 32%|███▏      | 20916/65536 [3:35:29<7:43:31,  1.60it/s]
 32%|███▏      | 20917/65536 [3:35:30<7:45:46,  1.60it/s]
 32%|███▏      | 20918/65536 [3:35:31<7:37:47,  1.62it/s]
 32%|███▏      | 20919/65536 [3:35:31<7:29:51,  1.65it/s]
 32%|███▏      | 20920/65536 [3:35:32<7:34:58,  1.63it/s]
                                                         
{'loss': 3.009, 'learning_rate': 7.145830782417044e-07, 'epoch': 1291.36}
+
 32%|███▏      | 20920/65536 [3:35:32<7:34:58,  1.63it/s]
 32%|███▏      | 20921/65536 [3:35:32<7:42:30,  1.61it/s]
 32%|███▏      | 20922/65536 [3:35:33<7:36:48,  1.63it/s]
 32%|███▏      | 20923/65536 [3:35:34<7:35:24,  1.63it/s]
 32%|███▏      | 20924/65536 [3:35:34<7:43:49,  1.60it/s]
 32%|███▏      | 20925/65536 [3:35:35<7:31:28,  1.65it/s]
 32%|███▏      | 20926/65536 [3:35:35<7:34:43,  1.64it/s]
 32%|███▏      | 20927/65536 [3:35:36<7:42:45,  1.61it/s]
 32%|███▏      | 20928/65536 [3:35:37<7:35:46,  1.63it/s]
 32%|███▏      | 20929/65536 [3:35:37<7:33:21,  1.64it/s]
 32%|███▏      | 20930/65536 [3:35:38<7:21:06,  1.69it/s]
 32%|███▏      | 20931/65536 [3:35:39<7:48:50,  1.59it/s]
 32%|███▏      | 20932/65536 [3:35:39<7:48:05,  1.59it/s]
 32%|███▏      | 20933/65536 [3:35:40<7:37:25,  1.63it/s]
 32%|███▏      | 20934/65536 [3:35:40<7:32:57,  1.64it/s]
 32%|███▏      | 20935/65536 [3:35:41<7:29:10,  1.65it/s]
 32%|███▏      | 20936/65536 [3:35:42<7:29:28,  1.65it/s]
 32%|███▏      | 20937/65536 [3:35:42<7:25:12,  1.67it/s]
 32%|███▏      | 20938/65536 [3:35:43<7:23:49,  1.67it/s]
 32%|███▏      | 20939/65536 [3:35:43<7:20:21,  1.69it/s]
 32%|███▏      | 20940/65536 [3:35:44<7:17:23,  1.70it/s]
                                                         
{'loss': 3.0454, 'learning_rate': 7.143075792824782e-07, 'epoch': 1292.59}
+
 32%|███▏      | 20940/65536 [3:35:44<7:17:23,  1.70it/s]
 32%|███▏      | 20941/65536 [3:35:45<7:30:05,  1.65it/s]
 32%|███▏      | 20942/65536 [3:35:45<7:35:19,  1.63it/s]
 32%|███▏      | 20943/65536 [3:35:46<7:37:59,  1.62it/s]
 32%|███▏      | 20944/65536 [3:35:46<7:31:50,  1.64it/s]
 32%|███▏      | 20945/65536 [3:35:47<7:29:22,  1.65it/s]
 32%|███▏      | 20946/65536 [3:35:48<7:35:45,  1.63it/s]
 32%|███▏      | 20947/65536 [3:35:48<7:57:56,  1.55it/s]
 32%|███▏      | 20948/65536 [3:35:49<7:52:35,  1.57it/s]
 32%|███▏      | 20949/65536 [3:35:50<7:50:51,  1.58it/s]
 32%|███▏      | 20950/65536 [3:35:50<7:38:10,  1.62it/s]
 32%|███▏      | 20951/65536 [3:35:51<7:33:45,  1.64it/s]
 32%|███▏      | 20952/65536 [3:35:51<7:32:36,  1.64it/s]
 32%|███▏      | 20953/65536 [3:35:52<7:26:48,  1.66it/s]
 32%|███▏      | 20954/65536 [3:35:52<7:19:08,  1.69it/s]
 32%|███▏      | 20955/65536 [3:35:53<7:28:56,  1.66it/s]
 32%|███▏      | 20956/65536 [3:35:54<7:43:00,  1.60it/s]
 32%|███▏      | 20957/65536 [3:35:54<7:34:09,  1.64it/s]
 32%|███▏      | 20958/65536 [3:35:55<7:31:05,  1.65it/s]
 32%|██���▏      | 20959/65536 [3:35:56<7:34:24,  1.63it/s]
 32%|███▏      | 20960/65536 [3:35:56<7:30:07,  1.65it/s]
                                                         
{'loss': 3.0392, 'learning_rate': 7.140320803232521e-07, 'epoch': 1293.83}
+
 32%|███▏      | 20960/65536 [3:35:56<7:30:07,  1.65it/s]
 32%|███▏      | 20961/65536 [3:35:57<7:23:23,  1.68it/s]
 32%|███▏      | 20962/65536 [3:35:57<7:41:51,  1.61it/s]
 32%|███▏      | 20963/65536 [3:35:58<7:54:26,  1.57it/s]
 32%|███▏      | 20964/65536 [3:35:59<7:45:28,  1.60it/s]
 32%|███▏      | 20965/65536 [3:35:59<7:52:30,  1.57it/s]
 32%|███▏      | 20966/65536 [3:36:00<7:42:57,  1.60it/s]
 32%|███▏      | 20967/65536 [3:36:01<7:33:20,  1.64it/s]
 32%|███▏      | 20968/65536 [3:36:01<7:28:06,  1.66it/s]
 32%|███▏      | 20969/65536 [3:36:02<7:31:12,  1.65it/s]
 32%|███▏      | 20970/65536 [3:36:02<7:27:06,  1.66it/s]
 32%|███▏      | 20971/65536 [3:36:03<7:23:33,  1.67it/s]
 32%|███▏      | 20972/65536 [3:36:04<7:40:07,  1.61it/s]
 32%|███▏      | 20973/65536 [3:36:04<7:27:54,  1.66it/s]
 32%|███▏      | 20974/65536 [3:36:05<7:29:39,  1.65it/s]
 32%|███▏      | 20975/65536 [3:36:05<7:26:30,  1.66it/s]
 32%|███▏      | 20976/65536 [3:36:06<7:27:05,  1.66it/s]
 32%|███▏      | 20977/65536 [3:36:07<7:25:55,  1.67it/s]
 32%|███▏      | 20978/65536 [3:36:07<7:23:28,  1.67it/s]
 32%|███▏      | 20979/65536 [3:36:08<7:36:19,  1.63it/s]
 32%|███▏      | 20980/65536 [3:36:09<7:57:19,  1.56it/s]
                                                         
{'loss': 3.046, 'learning_rate': 7.13756581364026e-07, 'epoch': 1295.06}
+
 32%|███▏      | 20980/65536 [3:36:09<7:57:19,  1.56it/s]
 32%|███▏      | 20981/65536 [3:36:09<7:55:18,  1.56it/s]
 32%|███▏      | 20982/65536 [3:36:10<8:01:16,  1.54it/s]
 32%|███▏      | 20983/65536 [3:36:10<8:03:50,  1.53it/s]
 32%|███▏      | 20984/65536 [3:36:11<7:58:42,  1.55it/s]
 32%|███▏      | 20985/65536 [3:36:12<7:50:38,  1.58it/s]
 32%|███▏      | 20986/65536 [3:36:12<7:45:33,  1.59it/s]
 32%|███▏      | 20987/65536 [3:36:13<7:44:22,  1.60it/s]
 32%|███▏      | 20988/65536 [3:36:14<7:39:27,  1.62it/s]
 32%|███▏      | 20989/65536 [3:36:14<7:29:27,  1.65it/s]
 32%|███▏      | 20990/65536 [3:36:15<7:20:59,  1.68it/s]
 32%|███▏      | 20991/65536 [3:36:15<7:23:58,  1.67it/s]
 32%|███▏      | 20992/65536 [3:36:16<7:21:47,  1.68it/s]
 32%|███▏      | 20993/65536 [3:36:16<7:18:23,  1.69it/s]
 32%|███▏      | 20994/65536 [3:36:17<7:17:25,  1.70it/s]
 32%|███▏      | 20995/65536 [3:36:18<7:16:37,  1.70it/s]
 32%|███▏      | 20996/65536 [3:36:18<7:36:31,  1.63it/s]
 32%|███▏      | 20997/65536 [3:36:19<7:30:41,  1.65it/s]
 32%|███▏      | 20998/65536 [3:36:20<7:28:04,  1.66it/s]
 32%|███▏      | 20999/65536 [3:36:20<7:29:02,  1.65it/s]
 32%|███▏      | 21000/65536 [3:36:21<7:30:52,  1.65it/s]
                                                         
{'loss': 3.0165, 'learning_rate': 7.134810824047998e-07, 'epoch': 1296.3}
+
 32%|███▏      | 21000/65536 [3:36:21<7:30:52,  1.65it/s]
 32%|███▏      | 21001/65536 [3:36:21<7:19:05,  1.69it/s]
 32%|███▏      | 21002/65536 [3:36:22<7:16:21,  1.70it/s]
 32%|███▏      | 21003/65536 [3:36:22<7:19:22,  1.69it/s]
 32%|███▏      | 21004/65536 [3:36:23<7:37:12,  1.62it/s]
 32%|███▏      | 21005/65536 [3:36:24<7:39:43,  1.61it/s]
 32%|███▏      | 21006/65536 [3:36:24<7:37:45,  1.62it/s]
 32%|███▏      | 21007/65536 [3:36:25<7:30:06,  1.65it/s]
 32%|███▏      | 21008/65536 [3:36:26<7:40:28,  1.61it/s]
 32%|███▏      | 21009/65536 [3:36:26<7:46:15,  1.59it/s]
 32%|███▏      | 21010/65536 [3:36:27<7:41:33,  1.61it/s]
 32%|███▏      | 21011/65536 [3:36:27<7:38:44,  1.62it/s]
 32%|███▏      | 21012/65536 [3:36:28<7:42:44,  1.60it/s]
 32%|███▏      | 21013/65536 [3:36:29<7:46:13,  1.59it/s]
 32%|███▏      | 21014/65536 [3:36:29<7:32:41,  1.64it/s]
 32%|███▏      | 21015/65536 [3:36:30<7:31:14,  1.64it/s]
 32%|███▏      | 21016/65536 [3:36:31<7:47:26,  1.59it/s]
 32%|███▏      | 21017/65536 [3:36:31<7:39:24,  1.62it/s]
 32%|███▏      | 21018/65536 [3:36:32<7:32:08,  1.64it/s]
 32%|███▏      | 21019/65536 [3:36:32<7:29:28,  1.65it/s]
 32%|███▏      | 21020/65536 [3:36:33<7:33:32,  1.64it/s]
                                                         
{'loss': 3.031, 'learning_rate': 7.132055834455737e-07, 'epoch': 1297.53}
+
 32%|███▏      | 21020/65536 [3:36:33<7:33:32,  1.64it/s]
 32%|███▏      | 21021/65536 [3:36:34<7:27:41,  1.66it/s]
 32%|███▏      | 21022/65536 [3:36:34<7:25:26,  1.67it/s]
 32%|███▏      | 21023/65536 [3:36:35<7:18:37,  1.69it/s]
 32%|███▏      | 21024/65536 [3:36:35<7:35:26,  1.63it/s]
 32%|███▏      | 21025/65536 [3:36:36<7:37:24,  1.62it/s]
 32%|███▏      | 21026/65536 [3:36:37<7:24:09,  1.67it/s]
 32%|███▏      | 21027/65536 [3:36:37<7:20:46,  1.68it/s]
 32%|███▏      | 21028/65536 [3:36:38<7:37:50,  1.62it/s]
 32%|███▏      | 21029/65536 [3:36:38<7:37:30,  1.62it/s]
 32%|███▏      | 21030/65536 [3:36:39<7:22:35,  1.68it/s]
 32%|███▏      | 21031/65536 [3:36:40<7:27:16,  1.66it/s]
 32%|███▏      | 21032/65536 [3:36:40<7:30:30,  1.65it/s]
 32%|███▏      | 21033/65536 [3:36:41<7:30:52,  1.65it/s]
 32%|███▏      | 21034/65536 [3:36:42<7:39:17,  1.61it/s]
 32%|███▏      | 21035/65536 [3:36:42<7:29:21,  1.65it/s]
 32%|███▏      | 21036/65536 [3:36:43<7:43:11,  1.60it/s]
 32%|███▏      | 21037/65536 [3:36:43<7:34:42,  1.63it/s]
 32%|███▏      | 21038/65536 [3:36:44<7:35:59,  1.63it/s]
 32%|███▏      | 21039/65536 [3:36:45<7:28:39,  1.65it/s]
 32%|███▏      | 21040/65536 [3:36:45<7:23:17,  1.67it/s]
                                                         
{'loss': 3.0662, 'learning_rate': 7.129300844863475e-07, 'epoch': 1298.77}
+
 32%|███▏      | 21040/65536 [3:36:45<7:23:17,  1.67it/s]
 32%|███▏      | 21041/65536 [3:36:46<7:24:55,  1.67it/s]
 32%|███▏      | 21042/65536 [3:36:46<7:31:53,  1.64it/s]
 32%|███▏      | 21043/65536 [3:36:47<7:30:49,  1.64it/s]
 32%|███▏      | 21044/65536 [3:36:48<7:51:36,  1.57it/s]
 32%|███▏      | 21045/65536 [3:36:48<7:45:51,  1.59it/s]
 32%|███▏      | 21046/65536 [3:36:49<7:37:48,  1.62it/s]
 32%|███▏      | 21047/65536 [3:36:49<7:38:03,  1.62it/s]
 32%|███▏      | 21048/65536 [3:36:50<7:34:55,  1.63it/s]
 32%|███▏      | 21049/65536 [3:36:51<7:30:04,  1.65it/s]
 32%|███▏      | 21050/65536 [3:36:51<7:27:00,  1.66it/s]
 32%|███▏      | 21051/65536 [3:36:52<7:27:38,  1.66it/s]
 32%|███▏      | 21052/65536 [3:36:53<7:33:57,  1.63it/s]
 32%|███▏      | 21053/65536 [3:36:53<7:41:19,  1.61it/s]
 32%|███▏      | 21054/65536 [3:36:54<7:49:59,  1.58it/s]
 32%|███▏      | 21055/65536 [3:36:54<7:45:44,  1.59it/s]
 32%|███▏      | 21056/65536 [3:36:55<7:47:43,  1.58it/s]
 32%|███▏      | 21057/65536 [3:36:56<7:34:50,  1.63it/s]
 32%|███▏      | 21058/65536 [3:36:56<7:33:01,  1.64it/s]
 32%|███▏      | 21059/65536 [3:36:57<7:27:24,  1.66it/s]
 32%|███▏      | 21060/65536 [3:36:57<7:19:20,  1.69it/s]
                                                         
{'loss': 3.0765, 'learning_rate': 7.126545855271213e-07, 'epoch': 1300.0}
+
 32%|███▏      | 21060/65536 [3:36:57<7:19:20,  1.69it/s]
 32%|███▏      | 21061/65536 [3:36:58<7:39:34,  1.61it/s]
 32%|███▏      | 21062/65536 [3:36:59<7:26:30,  1.66it/s]
 32%|███▏      | 21063/65536 [3:36:59<7:33:40,  1.63it/s]
 32%|███▏      | 21064/65536 [3:37:00<7:35:27,  1.63it/s]
 32%|███▏      | 21065/65536 [3:37:00<7:26:18,  1.66it/s]
 32%|███▏      | 21066/65536 [3:37:01<7:16:19,  1.70it/s]
 32%|███▏      | 21067/65536 [3:37:02<7:14:50,  1.70it/s]
 32%|███▏      | 21068/65536 [3:37:02<7:17:13,  1.70it/s]
 32%|███▏      | 21069/65536 [3:37:03<7:37:43,  1.62it/s]
 32%|███▏      | 21070/65536 [3:37:03<7:33:27,  1.63it/s]
 32%|███▏      | 21071/65536 [3:37:04<7:33:17,  1.63it/s]
 32%|███▏      | 21072/65536 [3:37:05<7:37:20,  1.62it/s]
 32%|███▏      | 21073/65536 [3:37:05<7:33:27,  1.63it/s]
 32%|███▏      | 21074/65536 [3:37:06<7:30:32,  1.64it/s]
 32%|███▏      | 21075/65536 [3:37:07<7:40:10,  1.61it/s]
 32%|███▏      | 21076/65536 [3:37:07<7:36:06,  1.62it/s]
 32%|███▏      | 21077/65536 [3:37:08<7:46:01,  1.59it/s]
 32%|███▏      | 21078/65536 [3:37:08<7:35:12,  1.63it/s]
 32%|███▏      | 21079/65536 [3:37:09<7:25:25,  1.66it/s]
 32%|███▏      | 21080/65536 [3:37:10<7:22:49,  1.67it/s]
                                                         
{'loss': 3.014, 'learning_rate': 7.123790865678951e-07, 'epoch': 1301.23}
+
 32%|███▏      | 21080/65536 [3:37:10<7:22:49,  1.67it/s]
 32%|███▏      | 21081/65536 [3:37:10<7:22:26,  1.67it/s]
 32%|███▏      | 21082/65536 [3:37:11<7:15:01,  1.70it/s]
 32%|███▏      | 21083/65536 [3:37:11<7:19:40,  1.69it/s]
 32%|███▏      | 21084/65536 [3:37:12<7:30:22,  1.65it/s]
 32%|███▏      | 21085/65536 [3:37:13<7:35:34,  1.63it/s]
 32%|███▏      | 21086/65536 [3:37:13<7:35:50,  1.63it/s]
 32%|███▏      | 21087/65536 [3:37:14<7:31:21,  1.64it/s]
 32%|███▏      | 21088/65536 [3:37:14<7:28:23,  1.65it/s]
 32%|███▏      | 21089/65536 [3:37:15<7:34:48,  1.63it/s]
 32%|███▏      | 21090/65536 [3:37:16<7:43:00,  1.60it/s]
 32%|███▏      | 21091/65536 [3:37:16<7:33:54,  1.63it/s]
 32%|███▏      | 21092/65536 [3:37:17<7:39:38,  1.61it/s]
 32%|��██▏      | 21093/65536 [3:37:18<8:07:08,  1.52it/s]
 32%|███▏      | 21094/65536 [3:37:18<7:47:24,  1.58it/s]
 32%|███▏      | 21095/65536 [3:37:19<8:00:12,  1.54it/s]
 32%|███▏      | 21096/65536 [3:37:20<7:43:08,  1.60it/s]
 32%|███▏      | 21097/65536 [3:37:20<7:42:49,  1.60it/s]
 32%|███▏      | 21098/65536 [3:37:21<7:35:03,  1.63it/s]
 32%|███▏      | 21099/65536 [3:37:21<7:46:52,  1.59it/s]
 32%|███▏      | 21100/65536 [3:37:22<7:35:09,  1.63it/s]
                                                         
{'loss': 2.9775, 'learning_rate': 7.12103587608669e-07, 'epoch': 1302.47}
+
 32%|███▏      | 21100/65536 [3:37:22<7:35:09,  1.63it/s]
 32%|███▏      | 21101/65536 [3:37:23<7:39:11,  1.61it/s]
 32%|███▏      | 21102/65536 [3:37:23<7:33:46,  1.63it/s]
 32%|███▏      | 21103/65536 [3:37:24<7:25:46,  1.66it/s]
 32%|███▏      | 21104/65536 [3:37:24<7:21:17,  1.68it/s]
 32%|███▏      | 21105/65536 [3:37:25<7:19:48,  1.68it/s]
 32%|███▏      | 21106/65536 [3:37:26<7:33:34,  1.63it/s]
 32%|███▏      | 21107/65536 [3:37:26<7:42:10,  1.60it/s]
 32%|███▏      | 21108/65536 [3:37:27<7:33:54,  1.63it/s]
 32%|███▏      | 21109/65536 [3:37:27<7:42:14,  1.60it/s]
 32%|███▏      | 21110/65536 [3:37:28<7:28:33,  1.65it/s]
 32%|███▏      | 21111/65536 [3:37:29<7:36:42,  1.62it/s]
 32%|███▏      | 21112/65536 [3:37:29<7:32:21,  1.64it/s]
 32%|███▏      | 21113/65536 [3:37:30<7:23:24,  1.67it/s]
 32%|███▏      | 21114/65536 [3:37:30<7:23:27,  1.67it/s]
 32%|███▏      | 21115/65536 [3:37:31<7:20:49,  1.68it/s]
 32%|███▏      | 21116/65536 [3:37:32<7:19:28,  1.68it/s]
 32%|███▏      | 21117/65536 [3:37:32<7:20:18,  1.68it/s]
 32%|███▏      | 21118/65536 [3:37:33<7:26:55,  1.66it/s]
 32%|███▏      | 21119/65536 [3:37:33<7:25:03,  1.66it/s]
 32%|███▏      | 21120/65536 [3:37:34<7:37:06,  1.62it/s]
                                                         
{'loss': 3.0938, 'learning_rate': 7.118280886494429e-07, 'epoch': 1303.7}
+
 32%|███▏      | 21120/65536 [3:37:34<7:37:06,  1.62it/s]
 32%|███▏      | 21121/65536 [3:37:35<7:36:03,  1.62it/s]
 32%|███▏      | 21122/65536 [3:37:35<7:32:11,  1.64it/s]
 32%|███▏      | 21123/65536 [3:37:36<7:26:03,  1.66it/s]
 32%|███▏      | 21124/65536 [3:37:37<7:40:06,  1.61it/s]
 32%|███▏      | 21125/65536 [3:37:37<7:47:22,  1.58it/s]
 32%|███▏      | 21126/65536 [3:37:38<7:39:08,  1.61it/s]
 32%|███▏      | 21127/65536 [3:37:38<7:48:04,  1.58it/s]
 32%|███▏      | 21128/65536 [3:37:39<7:32:48,  1.63it/s]
 32%|███▏      | 21129/65536 [3:37:40<7:26:18,  1.66it/s]
 32%|███▏      | 21130/65536 [3:37:40<7:30:41,  1.64it/s]
 32%|███▏      | 21131/65536 [3:37:41<7:29:58,  1.64it/s]
 32%|███▏      | 21132/65536 [3:37:41<7:29:02,  1.65it/s]
 32%|███▏      | 21133/65536 [3:37:42<7:25:44,  1.66it/s]
 32%|███▏      | 21134/65536 [3:37:43<7:26:42,  1.66it/s]
 32%|███▏      | 21135/65536 [3:37:43<7:16:17,  1.70it/s]
 32%|███▏      | 21136/65536 [3:37:44<7:14:15,  1.70it/s]
 32%|███▏      | 21137/65536 [3:37:44<7:22:11,  1.67it/s]
 32%|███▏      | 21138/65536 [3:37:45<7:19:23,  1.68it/s]
 32%|███▏      | 21139/65536 [3:37:46<7:39:22,  1.61it/s]
 32%|███▏      | 21140/65536 [3:37:46<7:32:48,  1.63it/s]
                                                         
{'loss': 3.0424, 'learning_rate': 7.115525896902166e-07, 'epoch': 1304.94}
+
 32%|███▏      | 21140/65536 [3:37:46<7:32:48,  1.63it/s]
 32%|███▏      | 21141/65536 [3:37:47<7:35:01,  1.63it/s]
 32%|███▏      | 21142/65536 [3:37:48<7:49:42,  1.58it/s]
 32%|███▏      | 21143/65536 [3:37:48<7:41:46,  1.60it/s]
 32%|███▏      | 21144/65536 [3:37:49<7:46:08,  1.59it/s]
 32%|███▏      | 21145/65536 [3:37:49<7:40:27,  1.61it/s]
 32%|███▏      | 21146/65536 [3:37:50<7:37:53,  1.62it/s]
 32%|███▏      | 21147/65536 [3:37:51<7:34:38,  1.63it/s]
 32%|███▏      | 21148/65536 [3:37:51<7:41:09,  1.60it/s]
 32%|███▏      | 21149/65536 [3:37:52<7:43:18,  1.60it/s]
 32%|███▏      | 21150/65536 [3:37:53<7:40:58,  1.60it/s]
 32%|███▏      | 21151/65536 [3:37:53<7:45:14,  1.59it/s]
 32%|███▏      | 21152/65536 [3:37:54<7:33:43,  1.63it/s]
 32%|███▏      | 21153/65536 [3:37:54<7:41:42,  1.60it/s]
 32%|███▏      | 21154/65536 [3:37:55<7:31:12,  1.64it/s]
 32%|███▏      | 21155/65536 [3:37:56<7:31:31,  1.64it/s]
 32%|███▏      | 21156/65536 [3:37:56<7:29:16,  1.65it/s]
 32%|███▏      | 21157/65536 [3:37:57<7:26:08,  1.66it/s]
 32%|███▏      | 21158/65536 [3:37:57<7:42:38,  1.60it/s]
 32%|███▏      | 21159/65536 [3:37:58<7:39:49,  1.61it/s]
 32%|███▏      | 21160/65536 [3:37:59<7:30:41,  1.64it/s]
                                                         
{'loss': 3.053, 'learning_rate': 7.112770907309905e-07, 'epoch': 1306.17}
+
 32%|███▏      | 21160/65536 [3:37:59<7:30:41,  1.64it/s]
 32%|███▏      | 21161/65536 [3:37:59<7:38:04,  1.61it/s]
 32%|███▏      | 21162/65536 [3:38:00<7:33:41,  1.63it/s]
 32%|███▏      | 21163/65536 [3:38:01<7:29:41,  1.64it/s]
 32%|███▏      | 21164/65536 [3:38:01<7:39:14,  1.61it/s]
 32%|███▏      | 21165/65536 [3:38:02<7:35:31,  1.62it/s]
 32%|███▏      | 21166/65536 [3:38:02<7:39:13,  1.61it/s]
 32%|███▏      | 21167/65536 [3:38:03<7:36:29,  1.62it/s]
 32%|███▏      | 21168/65536 [3:38:04<7:24:55,  1.66it/s]
 32%|███▏      | 21169/65536 [3:38:04<7:25:08,  1.66it/s]
 32%|███▏      | 21170/65536 [3:38:05<7:21:22,  1.68it/s]
 32%|███▏      | 21171/65536 [3:38:05<7:31:31,  1.64it/s]
 32%|███▏      | 21172/65536 [3:38:06<7:33:27,  1.63it/s]
 32%|███▏      | 21173/65536 [3:38:07<7:32:33,  1.63it/s]
 32%|███▏      | 21174/65536 [3:38:07<7:42:30,  1.60it/s]
 32%|███▏      | 21175/65536 [3:38:08<7:45:37,  1.59it/s]
 32%|███▏      | 21176/65536 [3:38:09<7:34:47,  1.63it/s]
 32%|███▏      | 21177/65536 [3:38:09<7:33:19,  1.63it/s]
 32%|███▏      | 21178/65536 [3:38:10<7:35:29,  1.62it/s]
 32%|███▏      | 21179/65536 [3:38:10<7:26:47,  1.65it/s]
 32%|███▏      | 21180/65536 [3:38:11<7:31:45,  1.64it/s]
                                                         
{'loss': 3.0423, 'learning_rate': 7.110015917717643e-07, 'epoch': 1307.41}
+
 32%|███▏      | 21180/65536 [3:38:11<7:31:45,  1.64it/s]
 32%|███▏      | 21181/65536 [3:38:12<7:47:46,  1.58it/s]
 32%|███▏      | 21182/65536 [3:38:12<7:35:19,  1.62it/s]
 32%|███▏      | 21183/65536 [3:38:13<7:24:21,  1.66it/s]
 32%|███▏      | 21184/65536 [3:38:13<7:23:14,  1.67it/s]
 32%|███▏      | 21185/65536 [3:38:14<7:28:03,  1.65it/s]
 32%|███▏      | 21186/65536 [3:38:15<7:39:11,  1.61it/s]
 32%|███▏      | 21187/65536 [3:38:15<7:24:47,  1.66it/s]
 32%|███▏      | 21188/65536 [3:38:16<7:23:10,  1.67it/s]
 32%|███▏      | 21189/65536 [3:38:16<7:25:54,  1.66it/s]
 32%|███▏      | 21190/65536 [3:38:17<7:33:41,  1.63it/s]
 32%|███▏      | 21191/65536 [3:38:18<7:28:37,  1.65it/s]
 32%|███▏      | 21192/65536 [3:38:18<7:37:46,  1.61it/s]
 32%|███▏      | 21193/65536 [3:38:19<7:26:05,  1.66it/s]
 32%|███▏      | 21194/65536 [3:38:19<7:24:57,  1.66it/s]
 32%|███▏      | 21195/65536 [3:38:20<7:23:57,  1.66it/s]
 32%|███▏      | 21196/65536 [3:38:21<7:20:02,  1.68it/s]
 32%|███▏      | 21197/65536 [3:38:21<7:29:28,  1.64it/s]
 32%|███▏      | 21198/65536 [3:38:22<7:36:05,  1.62it/s]
 32%|███▏      | 21199/65536 [3:38:23<7:40:40,  1.60it/s]
 32%|███▏      | 21200/65536 [3:38:23<7:34:58,  1.62it/s]
                                                         
{'loss': 3.003, 'learning_rate': 7.107260928125382e-07, 'epoch': 1308.64}
+
 32%|███▏      | 21200/65536 [3:38:23<7:34:58,  1.62it/s]
 32%|███▏      | 21201/65536 [3:38:24<7:27:48,  1.65it/s]
 32%|███▏      | 21202/65536 [3:38:24<7:29:32,  1.64it/s]
 32%|███▏      | 21203/65536 [3:38:25<7:38:20,  1.61it/s]
 32%|███▏      | 21204/65536 [3:38:26<7:41:19,  1.60it/s]
 32%|███▏      | 21205/65536 [3:38:26<7:36:41,  1.62it/s]
 32%|███▏      | 21206/65536 [3:38:27<7:45:28,  1.59it/s]
 32%|███▏      | 21207/65536 [3:38:27<7:43:09,  1.60it/s]
 32%|███▏      | 21208/65536 [3:38:28<7:47:40,  1.58it/s]
 32%|███▏      | 21209/65536 [3:38:29<7:45:54,  1.59it/s]
 32%|███▏      | 21210/65536 [3:38:29<7:38:50,  1.61it/s]
 32%|███▏      | 21211/65536 [3:38:30<7:37:08,  1.62it/s]
 32%|███▏      | 21212/65536 [3:38:31<7:30:05,  1.64it/s]
 32%|███▏      | 21213/65536 [3:38:31<7:23:50,  1.66it/s]
 32%|███▏      | 21214/65536 [3:38:32<7:27:57,  1.65it/s]
 32%|███▏      | 21215/65536 [3:38:32<7:33:00,  1.63it/s]
 32%|███▏      | 21216/65536 [3:38:33<7:35:14,  1.62it/s]
 32%|███▏      | 21217/65536 [3:38:34<7:23:05,  1.67it/s]
 32%|███▏      | 21218/65536 [3:38:34<7:25:49,  1.66it/s]
 32%|███▏      | 21219/65536 [3:38:35<7:36:24,  1.62it/s]
 32%|███▏      | 21220/65536 [3:38:35<7:35:10,  1.62it/s]
                                                         
{'loss': 3.0905, 'learning_rate': 7.10450593853312e-07, 'epoch': 1309.88}
+
 32%|███▏      | 21220/65536 [3:38:35<7:35:10,  1.62it/s]
 32%|███▏      | 21221/65536 [3:38:36<7:27:18,  1.65it/s]
 32%|███▏      | 21222/65536 [3:38:37<7:19:51,  1.68it/s]
 32%|███▏      | 21223/65536 [3:38:37<7:38:48,  1.61it/s]
 32%|███▏      | 21224/65536 [3:38:38<7:26:03,  1.66it/s]
 32%|███▏      | 21225/65536 [3:38:38<7:22:50,  1.67it/s]
 32%|███▏      | 21226/65536 [3:38:39<7:24:47,  1.66it/s]
 32%|███▏      | 21227/65536 [3:38:40<7:18:57,  1.68it/s]
 32%|███▏      | 21228/65536 [3:38:40<7:24:35,  1.66it/s]
 32%|███▏      | 21229/65536 [3:38:41<7:34:55,  1.62it/s]
 32%|███▏      | 21230/65536 [3:38:41<7:29:12,  1.64it/s]
 32%|███▏      | 21231/65536 [3:38:42<7:35:58,  1.62it/s]
 32%|███▏      | 21232/65536 [3:38:43<7:38:07,  1.61it/s]
 32%|███▏      | 21233/65536 [3:38:43<7:33:39,  1.63it/s]
 32%|███▏      | 21234/65536 [3:38:44<7:27:24,  1.65it/s]
 32%|███▏      | 21235/65536 [3:38:45<7:22:31,  1.67it/s]
 32%|███▏      | 21236/65536 [3:38:45<7:45:17,  1.59it/s]
 32%|███▏      | 21237/65536 [3:38:46<7:38:45,  1.61it/s]
 32%|███▏      | 21238/65536 [3:38:46<7:29:29,  1.64it/s]
 32%|███▏      | 21239/65536 [3:38:47<7:43:07,  1.59it/s]
 32%|███▏      | 21240/65536 [3:38:48<7:36:08,  1.62it/s]
                                                         
{'loss': 3.0676, 'learning_rate': 7.101750948940859e-07, 'epoch': 1311.11}
+
 32%|███▏      | 21240/65536 [3:38:48<7:36:08,  1.62it/s]
 32%|███▏      | 21241/65536 [3:38:48<7:42:14,  1.60it/s]
 32%|███▏      | 21242/65536 [3:38:49<7:39:57,  1.61it/s]
 32%|███▏      | 21243/65536 [3:38:50<7:29:02,  1.64it/s]
 32%|███▏      | 21244/65536 [3:38:50<7:43:06,  1.59it/s]
 32%|███▏      | 21245/65536 [3:38:51<7:54:30,  1.56it/s]
 32%|███▏      | 21246/65536 [3:38:51<7:43:28,  1.59it/s]
 32%|███▏      | 21247/65536 [3:38:52<7:37:51,  1.61it/s]
 32%|███▏      | 21248/65536 [3:38:53<7:26:57,  1.65it/s]
 32%|███▏      | 21249/65536 [3:38:53<7:24:37,  1.66it/s]
 32%|███▏      | 21250/65536 [3:38:54<7:41:23,  1.60it/s]
 32%|███▏      | 21251/65536 [3:38:55<7:41:05,  1.60it/s]
 32%|███▏      | 21252/65536 [3:38:55<7:37:08,  1.61it/s]
 32%|███▏      | 21253/65536 [3:38:56<7:31:14,  1.64it/s]
 32%|███▏      | 21254/65536 [3:38:56<7:30:03,  1.64it/s]
 32%|███▏      | 21255/65536 [3:38:57<7:41:28,  1.60it/s]
 32%|███▏      | 21256/65536 [3:38:58<7:44:31,  1.59it/s]
 32%|███▏      | 21257/65536 [3:38:58<7:38:09,  1.61it/s]
 32%|███▏      | 21258/65536 [3:38:59<7:33:12,  1.63it/s]
 32%|███▏      | 21259/65536 [3:38:59<7:30:13,  1.64it/s]
 32%|███▏      | 21260/65536 [3:39:00<7:35:36,  1.62it/s]
                                                         
{'loss': 2.9869, 'learning_rate': 7.098995959348598e-07, 'epoch': 1312.35}
+
 32%|███▏      | 21260/65536 [3:39:00<7:35:36,  1.62it/s]
 32%|███▏      | 21261/65536 [3:39:01<7:33:04,  1.63it/s]
 32%|███▏      | 21262/65536 [3:39:01<7:24:38,  1.66it/s]
 32%|███▏      | 21263/65536 [3:39:02<7:32:07,  1.63it/s]
 32%|███▏      | 21264/65536 [3:39:02<7:27:23,  1.65it/s]
 32%|███▏      | 21265/65536 [3:39:03<7:28:01,  1.65it/s]
 32%|███▏      | 21266/65536 [3:39:04<7:40:04,  1.60it/s]
 32%|███▏      | 21267/65536 [3:39:04<7:36:35,  1.62it/s]
 32%|███▏      | 21268/65536 [3:39:05<7:23:09,  1.66it/s]
 32%|███▏      | 21269/65536 [3:39:06<7:21:41,  1.67it/s]
 32%|███▏      | 21270/65536 [3:39:06<7:19:09,  1.68it/s]
 32%|███▏      | 21271/65536 [3:39:07<7:33:10,  1.63it/s]
 32%|███▏      | 21272/65536 [3:39:07<7:27:30,  1.65it/s]
 32%|███▏      | 21273/65536 [3:39:08<7:26:59,  1.65it/s]
 32%|███▏      | 21274/65536 [3:39:09<7:29:51,  1.64it/s]
 32%|███▏      | 21275/65536 [3:39:09<7:23:18,  1.66it/s]
 32%|███▏      | 21276/65536 [3:39:10<7:28:01,  1.65it/s]
 32%|███▏      | 21277/65536 [3:39:10<7:21:35,  1.67it/s]
 32%|███▏      | 21278/65536 [3:39:11<7:23:34,  1.66it/s]
 32%|███▏      | 21279/65536 [3:39:12<7:23:31,  1.66it/s]
 32%|███▏      | 21280/65536 [3:39:12<7:26:03,  1.65it/s]
                                                         
{'loss': 3.082, 'learning_rate': 7.096240969756336e-07, 'epoch': 1313.58}
+
 32%|███▏      | 21280/65536 [3:39:12<7:26:03,  1.65it/s]
 32%|███▏      | 21281/65536 [3:39:13<7:28:57,  1.64it/s]
 32%|███▏      | 21282/65536 [3:39:13<7:44:27,  1.59it/s]
 32%|███▏      | 21283/65536 [3:39:14<7:57:24,  1.54it/s]
 32%|███▏      | 21284/65536 [3:39:15<7:48:50,  1.57it/s]
 32%|███▏      | 21285/65536 [3:39:15<7:25:17,  1.66it/s]
 32%|███▏      | 21286/65536 [3:39:16<7:16:34,  1.69it/s]
 32%|███▏      | 21287/65536 [3:39:17<7:39:36,  1.60it/s]
 32%|███▏      | 21288/65536 [3:39:17<7:47:53,  1.58it/s]
 32%|███▏      | 21289/65536 [3:39:18<7:46:04,  1.58it/s]
 32%|███▏      | 21290/65536 [3:39:19<7:53:27,  1.56it/s]
 32%|███▏      | 21291/65536 [3:39:19<7:48:55,  1.57it/s]
 32%|███▏      | 21292/65536 [3:39:20<7:36:43,  1.61it/s]
 32%|███▏      | 21293/65536 [3:39:20<7:35:18,  1.62it/s]
 32%|███▏      | 21294/65536 [3:39:21<7:35:47,  1.62it/s]
 32%|███▏      | 21295/65536 [3:39:22<7:32:09,  1.63it/s]
 32%|███▏      | 21296/65536 [3:39:22<7:34:16,  1.62it/s]
 32%|███▏      | 21297/65536 [3:39:23<7:25:22,  1.66it/s]
 32%|███▏      | 21298/65536 [3:39:23<7:25:13,  1.66it/s]
 32%|███▏      | 21299/65536 [3:39:24<7:18:02,  1.68it/s]
 33%|███▎      | 21300/65536 [3:39:24<7:13:31,  1.70it/s]
                                                         
{'loss': 2.9242, 'learning_rate': 7.093485980164075e-07, 'epoch': 1314.81}
+
 33%|███▎      | 21300/65536 [3:39:24<7:13:31,  1.70it/s]
 33%|███▎      | 21301/65536 [3:39:25<7:31:57,  1.63it/s]
 33%|███▎      | 21302/65536 [3:39:26<7:38:30,  1.61it/s]
 33%|███▎      | 21303/65536 [3:39:26<7:33:07,  1.63it/s]
 33%|███▎      | 21304/65536 [3:39:27<7:59:44,  1.54it/s]
 33%|███▎      | 21305/65536 [3:39:28<7:52:47,  1.56it/s]
 33%|███▎      | 21306/65536 [3:39:28<7:35:57,  1.62it/s]
 33%|███▎      | 21307/65536 [3:39:29<7:24:44,  1.66it/s]
 33%|███▎      | 21308/65536 [3:39:29<7:16:44,  1.69it/s]
 33%|███▎      | 21309/65536 [3:39:30<7:21:14,  1.67it/s]
 33%|███▎      | 21310/65536 [3:39:31<7:31:05,  1.63it/s]
 33%|███▎      | 21311/65536 [3:39:31<7:28:32,  1.64it/s]
 33%|███▎      | 21312/65536 [3:39:32<7:21:26,  1.67it/s]
 33%|███▎      | 21313/65536 [3:39:32<7:23:22,  1.66it/s]
 33%|███▎      | 21314/65536 [3:39:33<7:16:03,  1.69it/s]
 33%|███▎      | 21315/65536 [3:39:34<7:13:55,  1.70it/s]
 33%|███▎      | 21316/65536 [3:39:34<7:33:33,  1.62it/s]
 33%|███▎      | 21317/65536 [3:39:35<7:27:39,  1.65it/s]
 33%|███▎      | 21318/65536 [3:39:36<7:30:23,  1.64it/s]
 33%|███▎      | 21319/65536 [3:39:36<7:36:17,  1.62it/s]
 33%|███▎      | 21320/65536 [3:39:37<7:42:46,  1.59it/s]
                                                         
{'loss': 3.0285, 'learning_rate': 7.090730990571813e-07, 'epoch': 1316.05}
+
 33%|███▎      | 21320/65536 [3:39:37<7:42:46,  1.59it/s]
 33%|███▎      | 21321/65536 [3:39:37<7:29:44,  1.64it/s]
 33%|███▎      | 21322/65536 [3:39:38<7:17:35,  1.68it/s]
 33%|███▎      | 21323/65536 [3:39:39<7:29:45,  1.64it/s]
 33%|███▎      | 21324/65536 [3:39:39<7:32:32,  1.63it/s]
 33%|███▎      | 21325/65536 [3:39:40<7:27:12,  1.65it/s]
 33%|███▎      | 21326/65536 [3:39:40<7:16:22,  1.69it/s]
 33%|███▎      | 21327/65536 [3:39:41<7:14:18,  1.70it/s]
 33%|███▎      | 21328/65536 [3:39:42<7:20:14,  1.67it/s]
 33%|███▎      | 21329/65536 [3:39:42<7:25:53,  1.65it/s]
 33%|███▎      | 21330/65536 [3:39:43<7:32:27,  1.63it/s]
 33%|███▎      | 21331/65536 [3:39:43<7:27:13,  1.65it/s]
 33%|███▎      | 21332/65536 [3:39:44<7:35:05,  1.62it/s]
 33%|███▎      | 21333/65536 [3:39:45<7:47:32,  1.58it/s]
 33%|███▎      | 21334/65536 [3:39:45<7:38:11,  1.61it/s]
 33%|███▎      | 21335/65536 [3:39:46<7:42:00,  1.59it/s]
 33%|███▎      | 21336/65536 [3:39:47<7:52:27,  1.56it/s]
 33%|███▎      | 21337/65536 [3:39:47<7:41:10,  1.60it/s]
 33%|███▎      | 21338/65536 [3:39:48<7:44:23,  1.59it/s]
 33%|███▎      | 21339/65536 [3:39:48<7:32:44,  1.63it/s]
 33%|███▎      | 21340/65536 [3:39:49<7:26:14,  1.65it/s]
                                                         
{'loss': 3.0237, 'learning_rate': 7.087976000979551e-07, 'epoch': 1317.28}
+
 33%|███▎      | 21340/65536 [3:39:49<7:26:14,  1.65it/s]
 33%|███▎      | 21341/65536 [3:39:50<7:29:01,  1.64it/s]
 33%|███▎      | 21342/65536 [3:39:50<7:38:15,  1.61it/s]
 33%|███▎      | 21343/65536 [3:39:51<7:26:23,  1.65it/s]
 33%|███▎      | 21344/65536 [3:39:51<7:26:13,  1.65it/s]
 33%|███▎      | 21345/65536 [3:39:52<7:33:45,  1.62it/s]
 33%|███▎      | 21346/65536 [3:39:53<7:34:56,  1.62it/s]
 33%|███▎      | 21347/65536 [3:39:53<7:38:21,  1.61it/s]
 33%|███▎      | 21348/65536 [3:39:54<7:29:53,  1.64it/s]
 33%|███▎      | 21349/65536 [3:39:55<7:23:31,  1.66it/s]
 33%|███▎      | 21350/65536 [3:39:55<7:24:18,  1.66it/s]
 33%|███▎      | 21351/65536 [3:39:56<7:37:46,  1.61it/s]
 33%|███▎      | 21352/65536 [3:39:56<7:51:43,  1.56it/s]
 33%|███▎      | 21353/65536 [3:39:57<7:57:10,  1.54it/s]
 33%|███▎      | 21354/65536 [3:39:58<7:59:27,  1.54it/s]
 33%|███▎      | 21355/65536 [3:39:58<7:51:48,  1.56it/s]
 33%|███▎      | 21356/65536 [3:39:59<7:49:27,  1.57it/s]
 33%|███▎      | 21357/65536 [3:40:00<7:40:29,  1.60it/s]
 33%|███▎      | 21358/65536 [3:40:00<7:40:26,  1.60it/s]
 33%|███▎      | 21359/65536 [3:40:01<7:54:25,  1.55it/s]
 33%|███▎      | 21360/65536 [3:40:02<7:52:03,  1.56it/s]
                                                         
{'loss': 2.9958, 'learning_rate': 7.085221011387288e-07, 'epoch': 1318.52}
+
 33%|███▎      | 21360/65536 [3:40:02<7:52:03,  1.56it/s]
 33%|███▎      | 21361/65536 [3:40:02<7:34:30,  1.62it/s]
 33%|███▎      | 21362/65536 [3:40:03<7:32:03,  1.63it/s]
 33%|███▎      | 21363/65536 [3:40:03<7:37:44,  1.61it/s]
 33%|███▎      | 21364/65536 [3:40:04<7:31:09,  1.63it/s]
 33%|███▎      | 21365/65536 [3:40:05<7:26:56,  1.65it/s]
 33%|███▎      | 21366/65536 [3:40:05<7:33:32,  1.62it/s]
 33%|███▎      | 21367/65536 [3:40:06<7:28:21,  1.64it/s]
 33%|███▎      | 21368/65536 [3:40:06<7:37:07,  1.61it/s]
 33%|███▎      | 21369/65536 [3:40:07<7:37:01,  1.61it/s]
 33%|███▎      | 21370/65536 [3:40:08<7:26:20,  1.65it/s]
 33%|███▎      | 21371/65536 [3:40:08<7:27:28,  1.64it/s]
 33%|███▎      | 21372/65536 [3:40:09<7:23:10,  1.66it/s]
 33%|███▎      | 21373/65536 [3:40:09<7:24:18,  1.66it/s]
 33%|███▎      | 21374/65536 [3:40:10<7:17:59,  1.68it/s]
 33%|███▎      | 21375/65536 [3:40:11<7:17:38,  1.68it/s]
 33%|███▎      | 21376/65536 [3:40:11<7:23:39,  1.66it/s]
 33%|███▎      | 21377/65536 [3:40:12<7:23:20,  1.66it/s]
 33%|███▎      | 21378/65536 [3:40:12<7:23:30,  1.66it/s]
 33%|███▎      | 21379/65536 [3:40:13<7:24:43,  1.65it/s]
 33%|███▎      | 21380/65536 [3:40:14<7:41:28,  1.59it/s]
                                                         
{'loss': 2.9874, 'learning_rate': 7.082466021795028e-07, 'epoch': 1319.75}
+
 33%|███▎      | 21380/65536 [3:40:14<7:41:28,  1.59it/s]
 33%|███▎      | 21381/65536 [3:40:14<7:36:45,  1.61it/s]
 33%|███▎      | 21382/65536 [3:40:15<7:37:04,  1.61it/s]
 33%|███▎      | 21383/65536 [3:40:16<7:38:08,  1.61it/s]
 33%|███▎      | 21384/65536 [3:40:16<7:42:17,  1.59it/s]
 33%|███▎      | 21385/65536 [3:40:17<7:45:02,  1.58it/s]
 33%|███▎      | 21386/65536 [3:40:18<7:45:15,  1.58it/s]
 33%|███▎      | 21387/65536 [3:40:18<7:32:36,  1.63it/s]
 33%|███▎      | 21388/65536 [3:40:19<7:26:05,  1.65it/s]
 33%|███▎      | 21389/65536 [3:40:19<7:24:32,  1.66it/s]
 33%|███▎      | 21390/65536 [3:40:20<7:24:40,  1.65it/s]
 33%|███▎      | 21391/65536 [3:40:20<7:18:06,  1.68it/s]
 33%|███▎      | 21392/65536 [3:40:21<7:11:03,  1.71it/s]
 33%|███▎      | 21393/65536 [3:40:22<7:22:14,  1.66it/s]
 33%|███▎      | 21394/65536 [3:40:22<7:25:55,  1.65it/s]
 33%|███▎      | 21395/65536 [3:40:23<7:22:32,  1.66it/s]
 33%|███▎      | 21396/65536 [3:40:24<7:31:04,  1.63it/s]
 33%|███▎      | 21397/65536 [3:40:24<7:34:19,  1.62it/s]
 33%|███▎      | 21398/65536 [3:40:25<7:30:31,  1.63it/s]
 33%|███▎      | 21399/65536 [3:40:25<7:39:30,  1.60it/s]
 33%|███▎      | 21400/65536 [3:40:26<7:36:11,  1.61it/s]
                                                         
{'loss': 3.0377, 'learning_rate': 7.079711032202767e-07, 'epoch': 1320.99}
+
 33%|███▎      | 21400/65536 [3:40:26<7:36:11,  1.61it/s]
 33%|███▎      | 21401/65536 [3:40:27<7:50:03,  1.56it/s]
 33%|███▎      | 21402/65536 [3:40:27<7:50:15,  1.56it/s]
 33%|███▎      | 21403/65536 [3:40:28<7:42:42,  1.59it/s]
 33%|███▎      | 21404/65536 [3:40:29<7:39:51,  1.60it/s]
 33%|███▎      | 21405/65536 [3:40:29<7:38:51,  1.60it/s]
 33%|███▎      | 21406/65536 [3:40:30<7:31:55,  1.63it/s]
 33%|███▎      | 21407/65536 [3:40:30<7:45:03,  1.58it/s]
 33%|███▎      | 21408/65536 [3:40:31<7:49:35,  1.57it/s]
 33%|███▎      | 21409/65536 [3:40:32<7:42:24,  1.59it/s]
 33%|███▎      | 21410/65536 [3:40:32<7:34:52,  1.62it/s]
 33%|███▎      | 21411/65536 [3:40:33<7:26:36,  1.65it/s]
 33%|███▎      | 21412/65536 [3:40:33<7:26:43,  1.65it/s]
 33%|███▎      | 21413/65536 [3:40:34<7:27:35,  1.64it/s]
 33%|███▎      | 21414/65536 [3:40:35<7:42:02,  1.59it/s]
 33%|███▎      | 21415/65536 [3:40:35<7:30:07,  1.63it/s]
 33%|███▎      | 21416/65536 [3:40:36<7:27:33,  1.64it/s]
 33%|███▎      | 21417/65536 [3:40:37<7:40:17,  1.60it/s]
 33%|███▎      | 21418/65536 [3:40:37<7:21:04,  1.67it/s]
 33%|███▎      | 21419/65536 [3:40:38<7:25:21,  1.65it/s]
 33%|███▎      | 21420/65536 [3:40:38<7:25:05,  1.65it/s]
                                                         
{'loss': 3.0517, 'learning_rate': 7.076956042610505e-07, 'epoch': 1322.22}
+
 33%|███▎      | 21420/65536 [3:40:38<7:25:05,  1.65it/s]
 33%|███▎      | 21421/65536 [3:40:39<7:17:58,  1.68it/s]
 33%|███▎      | 21422/65536 [3:40:40<7:25:52,  1.65it/s]
 33%|███▎      | 21423/65536 [3:40:40<7:32:27,  1.62it/s]
 33%|███▎      | 21424/65536 [3:40:41<7:27:53,  1.64it/s]
 33%|███▎      | 21425/65536 [3:40:41<7:22:15,  1.66it/s]
 33%|███▎      | 21426/65536 [3:40:42<7:25:25,  1.65it/s]
 33%|███▎      | 21427/65536 [3:40:43<7:28:58,  1.64it/s]
 33%|███▎      | 21428/65536 [3:40:43<7:22:35,  1.66it/s]
 33%|███▎      | 21429/65536 [3:40:44<7:10:14,  1.71it/s]
 33%|███▎      | 21430/65536 [3:40:44<7:06:46,  1.72it/s]
 33%|███▎      | 21431/65536 [3:40:45<7:34:24,  1.62it/s]
 33%|███▎      | 21432/65536 [3:40:46<7:37:56,  1.61it/s]
 33%|███▎      | 21433/65536 [3:40:46<7:47:02,  1.57it/s]
 33%|███▎      | 21434/65536 [3:40:47<7:37:08,  1.61it/s]
 33%|███▎      | 21435/65536 [3:40:48<7:51:31,  1.56it/s]
 33%|███▎      | 21436/65536 [3:40:48<7:41:16,  1.59it/s]
 33%|███▎      | 21437/65536 [3:40:49<7:24:14,  1.65it/s]
 33%|███▎      | 21438/65536 [3:40:49<7:19:39,  1.67it/s]
 33%|███▎      | 21439/65536 [3:40:50<7:16:23,  1.68it/s]
 33%|███▎      | 21440/65536 [3:40:51<7:27:11,  1.64it/s]
                                                         
{'loss': 2.9989, 'learning_rate': 7.074201053018244e-07, 'epoch': 1323.46}
+
 33%|███▎      | 21440/65536 [3:40:51<7:27:11,  1.64it/s]
 33%|███▎      | 21441/65536 [3:40:51<7:32:38,  1.62it/s]
 33%|███▎      | 21442/65536 [3:40:52<7:35:41,  1.61it/s]
 33%|███▎      | 21443/65536 [3:40:52<7:33:58,  1.62it/s]
 33%|███▎      | 21444/65536 [3:40:53<7:32:16,  1.62it/s]
 33%|███▎      | 21445/65536 [3:40:54<7:20:14,  1.67it/s]
 33%|███▎      | 21446/65536 [3:40:54<7:17:36,  1.68it/s]
 33%|███▎      | 21447/65536 [3:40:55<7:17:14,  1.68it/s]
 33%|███▎      | 21448/65536 [3:40:55<7:15:01,  1.69it/s]
 33%|███▎      | 21449/65536 [3:40:56<7:32:59,  1.62it/s]
 33%|███▎      | 21450/65536 [3:40:57<7:26:14,  1.65it/s]
 33%|███▎      | 21451/65536 [3:40:57<7:25:02,  1.65it/s]
 33%|███▎      | 21452/65536 [3:40:58<7:22:48,  1.66it/s]
 33%|███▎      | 21453/65536 [3:40:59<7:41:32,  1.59it/s]
 33%|███▎      | 21454/65536 [3:40:59<7:29:41,  1.63it/s]
 33%|███▎      | 21455/65536 [3:41:00<7:32:19,  1.62it/s]
 33%|███▎      | 21456/65536 [3:41:00<7:42:55,  1.59it/s]
 33%|███▎      | 21457/65536 [3:41:01<7:38:30,  1.60it/s]
 33%|███▎      | 21458/65536 [3:41:02<7:37:28,  1.61it/s]
 33%|███▎      | 21459/65536 [3:41:02<7:30:50,  1.63it/s]
 33%|███▎      | 21460/65536 [3:41:03<7:24:19,  1.65it/s]
                                                         
{'loss': 3.0555, 'learning_rate': 7.071446063425982e-07, 'epoch': 1324.69}
+
 33%|███▎      | 21460/65536 [3:41:03<7:24:19,  1.65it/s]
 33%|███▎      | 21461/65536 [3:41:03<7:25:28,  1.65it/s]
 33%|███▎      | 21462/65536 [3:41:04<7:22:42,  1.66it/s]
 33%|███▎      | 21463/65536 [3:41:05<7:23:29,  1.66it/s]
 33%|███▎      | 21464/65536 [3:41:05<7:20:15,  1.67it/s]
 33%|███▎      | 21465/65536 [3:41:06<7:32:29,  1.62it/s]
 33%|███▎      | 21466/65536 [3:41:06<7:40:10,  1.60it/s]
 33%|███▎      | 21467/65536 [3:41:07<7:27:05,  1.64it/s]
 33%|███▎      | 21468/65536 [3:41:08<7:30:01,  1.63it/s]
 33%|███▎      | 21469/65536 [3:41:08<7:27:21,  1.64it/s]
 33%|███▎      | 21470/65536 [3:41:09<7:31:26,  1.63it/s]
 33%|███▎      | 21471/65536 [3:41:09<7:20:55,  1.67it/s]
 33%|███▎      | 21472/65536 [3:41:10<7:19:00,  1.67it/s]
 33%|███▎      | 21473/65536 [3:41:11<7:12:51,  1.70it/s]
 33%|███▎      | 21474/65536 [3:41:11<7:12:28,  1.70it/s]
 33%|███▎      | 21475/65536 [3:41:12<7:20:26,  1.67it/s]
 33%|███▎      | 21476/65536 [3:41:12<7:18:08,  1.68it/s]
 33%|███▎      | 21477/65536 [3:41:13<7:24:03,  1.65it/s]
 33%|███▎      | 21478/65536 [3:41:14<7:27:20,  1.64it/s]
 33%|███▎      | 21479/65536 [3:41:14<7:34:19,  1.62it/s]
 33%|███▎      | 21480/65536 [3:41:15<7:26:24,  1.64it/s]
                                                         
{'loss': 3.0958, 'learning_rate': 7.068691073833721e-07, 'epoch': 1325.93}
+
 33%|███▎      | 21480/65536 [3:41:15<7:26:24,  1.64it/s]
 33%|███▎      | 21481/65536 [3:41:16<7:32:58,  1.62it/s]
 33%|███▎      | 21482/65536 [3:41:16<7:45:55,  1.58it/s]
 33%|███▎      | 21483/65536 [3:41:17<7:40:36,  1.59it/s]
 33%|███▎      | 21484/65536 [3:41:17<7:36:47,  1.61it/s]
 33%|███▎      | 21485/65536 [3:41:18<7:22:27,  1.66it/s]
 33%|███▎      | 21486/65536 [3:41:19<7:27:14,  1.64it/s]
 33%|███▎      | 21487/65536 [3:41:19<7:26:36,  1.64it/s]
 33%|███▎      | 21488/65536 [3:41:20<7:25:08,  1.65it/s]
 33%|███▎      | 21489/65536 [3:41:20<7:22:50,  1.66it/s]
 33%|███▎      | 21490/65536 [3:41:21<7:13:17,  1.69it/s]
 33%|███▎      | 21491/65536 [3:41:22<7:14:03,  1.69it/s]
 33%|███▎      | 21492/65536 [3:41:22<7:08:21,  1.71it/s]
 33%|███▎      | 21493/65536 [3:41:23<7:16:57,  1.68it/s]
 33%|███▎      | 21494/65536 [3:41:23<7:33:19,  1.62it/s]
 33%|███▎      | 21495/65536 [3:41:24<7:30:50,  1.63it/s]
 33%|███▎      | 21496/65536 [3:41:25<7:26:54,  1.64it/s]
 33%|███▎      | 21497/65536 [3:41:25<7:31:38,  1.63it/s]
 33%|███▎      | 21498/65536 [3:41:26<7:46:57,  1.57it/s]
 33%|███▎      | 21499/65536 [3:41:27<7:43:01,  1.59it/s]
 33%|███▎      | 21500/65536 [3:41:27<7:40:47,  1.59it/s]
                                                         
{'loss': 3.0202, 'learning_rate': 7.06593608424146e-07, 'epoch': 1327.16}
+
 33%|███▎      | 21500/65536 [3:41:27<7:40:47,  1.59it/s]
 33%|███▎      | 21501/65536 [3:41:28<7:34:56,  1.61it/s]
 33%|███▎      | 21502/65536 [3:41:28<7:27:24,  1.64it/s]
 33%|███▎      | 21503/65536 [3:41:29<7:21:22,  1.66it/s]
 33%|███▎      | 21504/65536 [3:41:30<7:23:03,  1.66it/s]
 33%|███▎      | 21505/65536 [3:41:30<7:33:32,  1.62it/s]
 33%|███▎      | 21506/65536 [3:41:31<7:33:18,  1.62it/s]
 33%|███▎      | 21507/65536 [3:41:31<7:34:15,  1.62it/s]
 33%|███▎      | 21508/65536 [3:41:32<7:24:34,  1.65it/s]
 33%|███▎      | 21509/65536 [3:41:33<7:22:53,  1.66it/s]
 33%|███▎      | 21510/65536 [3:41:33<7:28:46,  1.64it/s]
 33%|███▎      | 21511/65536 [3:41:34<7:24:39,  1.65it/s]
 33%|███▎      | 21512/65536 [3:41:35<7:36:25,  1.61it/s]
 33%|███▎      | 21513/65536 [3:41:35<7:30:42,  1.63it/s]
 33%|███▎      | 21514/65536 [3:41:36<7:48:30,  1.57it/s]
 33%|███▎      | 21515/65536 [3:41:36<7:47:27,  1.57it/s]
 33%|███▎      | 21516/65536 [3:41:37<7:43:29,  1.58it/s]
 33%|███▎      | 21517/65536 [3:41:38<7:32:31,  1.62it/s]
 33%|███▎      | 21518/65536 [3:41:38<7:28:29,  1.64it/s]
 33%|███▎      | 21519/65536 [3:41:39<7:25:09,  1.65it/s]
 33%|███▎      | 21520/65536 [3:41:39<7:27:52,  1.64it/s]
                                                         
{'loss': 3.0313, 'learning_rate': 7.063181094649198e-07, 'epoch': 1328.4}
+
 33%|███▎      | 21520/65536 [3:41:39<7:27:52,  1.64it/s]
 33%|███▎      | 21521/65536 [3:41:40<7:20:30,  1.67it/s]
 33%|███▎      | 21522/65536 [3:41:41<7:16:42,  1.68it/s]
 33%|███▎      | 21523/65536 [3:41:41<7:34:52,  1.61it/s]
 33%|███▎      | 21524/65536 [3:41:42<7:32:39,  1.62it/s]
 33%|███▎      | 21525/65536 [3:41:43<7:32:28,  1.62it/s]
 33%|███▎      | 21526/65536 [3:41:43<7:23:31,  1.65it/s]
 33%|███▎      | 21527/65536 [3:41:44<7:19:13,  1.67it/s]
 33%|███▎      | 21528/65536 [3:41:44<7:29:14,  1.63it/s]
 33%|███▎      | 21529/65536 [3:41:45<7:23:55,  1.65it/s]
 33%|███▎      | 21530/65536 [3:41:46<7:59:28,  1.53it/s]
 33%|███▎      | 21531/65536 [3:41:46<7:49:39,  1.56it/s]
 33%|███▎      | 21532/65536 [3:41:47<7:35:07,  1.61it/s]
 33%|███▎      | 21533/65536 [3:41:47<7:33:03,  1.62it/s]
 33%|███▎      | 21534/65536 [3:41:48<7:33:28,  1.62it/s]
 33%|███▎      | 21535/65536 [3:41:49<7:28:37,  1.63it/s]
 33%|███▎      | 21536/65536 [3:41:49<7:28:57,  1.63it/s]
 33%|███▎      | 21537/65536 [3:41:50<7:30:33,  1.63it/s]
 33%|███▎      | 21538/65536 [3:41:50<7:18:03,  1.67it/s]
 33%|███▎      | 21539/65536 [3:41:51<7:23:30,  1.65it/s]
 33%|███▎      | 21540/65536 [3:41:52<7:24:34,  1.65it/s]
                                                         
{'loss': 3.0543, 'learning_rate': 7.060426105056937e-07, 'epoch': 1329.63}
+
 33%|███▎      | 21540/65536 [3:41:52<7:24:34,  1.65it/s]
 33%|███▎      | 21541/65536 [3:41:52<7:23:08,  1.65it/s]
 33%|███▎      | 21542/65536 [3:41:53<7:18:13,  1.67it/s]
 33%|███▎      | 21543/65536 [3:41:53<7:08:17,  1.71it/s]
 33%|███▎      | 21544/65536 [3:41:54<7:30:52,  1.63it/s]
 33%|███▎      | 21545/65536 [3:41:55<7:33:49,  1.62it/s]
 33%|███▎      | 21546/65536 [3:41:55<7:36:08,  1.61it/s]
 33%|███▎      | 21547/65536 [3:41:56<7:51:09,  1.56it/s]
 33%|███▎      | 21548/65536 [3:41:57<7:43:53,  1.58it/s]
 33%|███▎      | 21549/65536 [3:41:57<7:35:54,  1.61it/s]
 33%|███▎      | 21550/65536 [3:41:58<7:40:13,  1.59it/s]
 33%|███▎      | 21551/65536 [3:41:59<7:30:12,  1.63it/s]
 33%|███▎      | 21552/65536 [3:41:59<7:28:10,  1.64it/s]
 33%|███▎      | 21553/65536 [3:42:00<7:22:06,  1.66it/s]
 33%|███▎      | 21554/65536 [3:42:00<7:17:30,  1.68it/s]
 33%|███▎      | 21555/65536 [3:42:01<7:22:19,  1.66it/s]
 33%|███▎      | 21556/65536 [3:42:02<7:22:55,  1.65it/s]
 33%|███▎      | 21557/65536 [3:42:02<7:39:24,  1.60it/s]
 33%|███▎      | 21558/65536 [3:42:03<7:33:03,  1.62it/s]
 33%|███▎      | 21559/65536 [3:42:03<7:47:30,  1.57it/s]
 33%|███▎      | 21560/65536 [3:42:04<7:39:47,  1.59it/s]
                                                         
{'loss': 3.0223, 'learning_rate': 7.057671115464675e-07, 'epoch': 1330.86}
+
 33%|███▎      | 21560/65536 [3:42:04<7:39:47,  1.59it/s]
 33%|███▎      | 21561/65536 [3:42:05<7:40:31,  1.59it/s]
 33%|███▎      | 21562/65536 [3:42:05<7:31:25,  1.62it/s]
 33%|███▎      | 21563/65536 [3:42:06<7:57:48,  1.53it/s]
 33%|███▎      | 21564/65536 [3:42:07<7:55:28,  1.54it/s]
 33%|███▎      | 21565/65536 [3:42:07<7:42:27,  1.58it/s]
 33%|███▎      | 21566/65536 [3:42:08<7:38:00,  1.60it/s]
 33%|███▎      | 21567/65536 [3:42:09<7:37:36,  1.60it/s]
 33%|███▎      | 21568/65536 [3:42:09<7:28:16,  1.63it/s]
 33%|███▎      | 21569/65536 [3:42:10<7:27:27,  1.64it/s]
 33%|███▎      | 21570/65536 [3:42:10<7:23:36,  1.65it/s]
 33%|███▎      | 21571/65536 [3:42:11<7:19:48,  1.67it/s]
 33%|███▎      | 21572/65536 [3:42:11<7:22:10,  1.66it/s]
 33%|███▎      | 21573/65536 [3:42:12<7:23:21,  1.65it/s]
 33%|███▎      | 21574/65536 [3:42:13<7:27:46,  1.64it/s]
 33%|███▎      | 21575/65536 [3:42:13<7:32:44,  1.62it/s]
 33%|███▎      | 21576/65536 [3:42:14<7:30:00,  1.63it/s]
 33%|███▎      | 21577/65536 [3:42:15<7:26:13,  1.64it/s]
 33%|███▎      | 21578/65536 [3:42:15<7:34:06,  1.61it/s]
 33%|███▎      | 21579/65536 [3:42:16<7:55:19,  1.54it/s]
 33%|███▎      | 21580/65536 [3:42:17<7:43:43,  1.58it/s]
                                                         
{'loss': 2.9771, 'learning_rate': 7.054916125872414e-07, 'epoch': 1332.1}
+
 33%|███▎      | 21580/65536 [3:42:17<7:43:43,  1.58it/s]
 33%|███▎      | 21581/65536 [3:42:17<7:43:03,  1.58it/s]
 33%|███▎      | 21582/65536 [3:42:18<7:32:49,  1.62it/s]
 33%|███▎      | 21583/65536 [3:42:18<7:25:41,  1.64it/s]
 33%|███▎      | 21584/65536 [3:42:19<7:19:41,  1.67it/s]
 33%|███▎      | 21585/65536 [3:42:20<7:21:12,  1.66it/s]
 33%|███▎      | 21586/65536 [3:42:20<7:24:38,  1.65it/s]
 33%|███▎      | 21587/65536 [3:42:21<7:41:00,  1.59it/s]
 33%|███▎      | 21588/65536 [3:42:21<7:26:33,  1.64it/s]
 33%|███▎      | 21589/65536 [3:42:22<7:36:40,  1.60it/s]
 33%|███▎      | 21590/65536 [3:42:23<7:32:21,  1.62it/s]
 33%|███▎      | 21591/65536 [3:42:23<7:21:30,  1.66it/s]
 33%|███▎      | 21592/65536 [3:42:24<7:30:19,  1.63it/s]
 33%|███▎      | 21593/65536 [3:42:24<7:21:01,  1.66it/s]
 33%|███▎      | 21594/65536 [3:42:25<7:27:04,  1.64it/s]
 33%|███▎      | 21595/65536 [3:42:26<7:41:00,  1.59it/s]
 33%|███▎      | 21596/65536 [3:42:26<7:39:46,  1.59it/s]
 33%|███▎      | 21597/65536 [3:42:27<7:43:16,  1.58it/s]
 33%|███▎      | 21598/65536 [3:42:28<7:34:42,  1.61it/s]
 33%|███▎      | 21599/65536 [3:42:28<7:30:06,  1.63it/s]
 33%|███▎      | 21600/65536 [3:42:29<7:27:16,  1.64it/s]
                                                         
{'loss': 3.0458, 'learning_rate': 7.052161136280152e-07, 'epoch': 1333.33}
+
 33%|███▎      | 21600/65536 [3:42:29<7:27:16,  1.64it/s]
 33%|███▎      | 21601/65536 [3:42:29<7:19:24,  1.67it/s]
 33%|███▎      | 21602/65536 [3:42:30<7:16:59,  1.68it/s]
 33%|███▎      | 21603/65536 [3:42:31<7:11:29,  1.70it/s]
 33%|███▎      | 21604/65536 [3:42:31<7:09:27,  1.70it/s]
 33%|███▎      | 21605/65536 [3:42:32<7:21:05,  1.66it/s]
 33%|███▎      | 21606/65536 [3:42:32<7:24:49,  1.65it/s]
 33%|███▎      | 21607/65536 [3:42:33<7:34:14,  1.61it/s]
 33%|███▎      | 21608/65536 [3:42:34<7:33:28,  1.61it/s]
 33%|███▎      | 21609/65536 [3:42:34<7:30:50,  1.62it/s]
 33%|███▎      | 21610/65536 [3:42:35<7:30:09,  1.63it/s]
 33%|███▎      | 21611/65536 [3:42:36<7:45:45,  1.57it/s]
 33%|███▎      | 21612/65536 [3:42:36<7:47:36,  1.57it/s]
 33%|███▎      | 21613/65536 [3:42:37<7:45:02,  1.57it/s]
 33%|███▎      | 21614/65536 [3:42:37<7:39:44,  1.59it/s]
 33%|███▎      | 21615/65536 [3:42:38<7:38:06,  1.60it/s]
 33%|███▎      | 21616/65536 [3:42:39<7:30:23,  1.63it/s]
 33%|███▎      | 21617/65536 [3:42:39<7:38:05,  1.60it/s]
 33%|███▎      | 21618/65536 [3:42:40<7:31:43,  1.62it/s]
 33%|███▎      | 21619/65536 [3:42:40<7:29:15,  1.63it/s]
 33%|███▎      | 21620/65536 [3:42:41<7:23:02,  1.65it/s]
                                                         
{'loss': 3.0683, 'learning_rate': 7.049406146687891e-07, 'epoch': 1334.57}
+
 33%|███▎      | 21620/65536 [3:42:41<7:23:02,  1.65it/s]
 33%|███▎      | 21621/65536 [3:42:42<7:34:34,  1.61it/s]
 33%|███▎      | 21622/65536 [3:42:42<7:21:12,  1.66it/s]
 33%|███▎      | 21623/65536 [3:42:43<7:20:01,  1.66it/s]
 33%|███▎      | 21624/65536 [3:42:43<7:25:15,  1.64it/s]
 33%|███▎      | 21625/65536 [3:42:44<7:22:22,  1.65it/s]
 33%|███▎      | 21626/65536 [3:42:45<7:31:34,  1.62it/s]
 33%|███▎      | 21627/65536 [3:42:45<7:11:58,  1.69it/s]
 33%|███▎      | 21628/65536 [3:42:46<7:31:39,  1.62it/s]
 33%|███▎      | 21629/65536 [3:42:47<7:24:21,  1.65it/s]
 33%|███▎      | 21630/65536 [3:42:47<7:34:29,  1.61it/s]
 33%|███▎      | 21631/65536 [3:42:48<7:24:30,  1.65it/s]
 33%|███▎      | 21632/65536 [3:42:48<7:35:17,  1.61it/s]
 33%|███▎      | 21633/65536 [3:42:49<7:36:38,  1.60it/s]
 33%|███▎      | 21634/65536 [3:42:50<7:20:28,  1.66it/s]
 33%|███▎      | 21635/65536 [3:42:50<7:14:25,  1.68it/s]
 33%|███▎      | 21636/65536 [3:42:51<7:17:08,  1.67it/s]
 33%|███▎      | 21637/65536 [3:42:51<7:26:37,  1.64it/s]
 33%|███▎      | 21638/65536 [3:42:52<7:16:10,  1.68it/s]
 33%|███▎      | 21639/65536 [3:42:53<7:28:17,  1.63it/s]
 33%|███▎      | 21640/65536 [3:42:53<7:31:10,  1.62it/s]
                                                         
{'loss': 3.0293, 'learning_rate': 7.046651157095629e-07, 'epoch': 1335.8}
+
 33%|███▎      | 21640/65536 [3:42:53<7:31:10,  1.62it/s]
 33%|███▎      | 21641/65536 [3:42:54<7:30:50,  1.62it/s]
 33%|███▎      | 21642/65536 [3:42:54<7:24:27,  1.65it/s]
 33%|███▎      | 21643/65536 [3:42:55<7:16:24,  1.68it/s]
 33%|███▎      | 21644/65536 [3:42:56<7:45:57,  1.57it/s]
 33%|███▎      | 21645/65536 [3:42:56<7:37:52,  1.60it/s]
 33%|███▎      | 21646/65536 [3:42:57<7:27:54,  1.63it/s]
 33%|███▎      | 21647/65536 [3:42:58<7:26:15,  1.64it/s]
 33%|███▎      | 21648/65536 [3:42:58<7:26:51,  1.64it/s]
 33%|███▎      | 21649/65536 [3:42:59<7:20:58,  1.66it/s]
 33%|███▎      | 21650/65536 [3:42:59<7:17:23,  1.67it/s]
 33%|███▎      | 21651/65536 [3:43:00<7:21:49,  1.66it/s]
 33%|███▎      | 21652/65536 [3:43:01<7:25:09,  1.64it/s]
 33%|███▎      | 21653/65536 [3:43:01<7:18:36,  1.67it/s]
 33%|███▎      | 21654/65536 [3:43:02<7:26:47,  1.64it/s]
 33%|███▎      | 21655/65536 [3:43:02<7:19:32,  1.66it/s]
 33%|███▎      | 21656/65536 [3:43:03<7:35:48,  1.60it/s]
 33%|███▎      | 21657/65536 [3:43:04<7:23:59,  1.65it/s]
 33%|███▎      | 21658/65536 [3:43:04<7:17:08,  1.67it/s]
 33%|███▎      | 21659/65536 [3:43:05<7:34:27,  1.61it/s]
 33%|███▎      | 21660/65536 [3:43:06<7:47:35,  1.56it/s]
                                                         
{'loss': 3.0793, 'learning_rate': 7.043896167503367e-07, 'epoch': 1337.04}
+
 33%|███▎      | 21660/65536 [3:43:06<7:47:35,  1.56it/s]
 33%|███▎      | 21661/65536 [3:43:06<7:39:40,  1.59it/s]
 33%|███▎      | 21662/65536 [3:43:07<7:38:06,  1.60it/s]
 33%|███▎      | 21663/65536 [3:43:07<7:17:26,  1.67it/s]
 33%|███▎      | 21664/65536 [3:43:08<7:31:32,  1.62it/s]
 33%|███▎      | 21665/65536 [3:43:09<7:23:44,  1.65it/s]
 33%|███▎      | 21666/65536 [3:43:09<7:35:25,  1.61it/s]
 33%|███▎      | 21667/65536 [3:43:10<7:35:41,  1.60it/s]
 33%|███▎      | 21668/65536 [3:43:10<7:26:12,  1.64it/s]
 33%|███▎      | 21669/65536 [3:43:11<7:35:39,  1.60it/s]
 33%|███▎      | 21670/65536 [3:43:12<7:28:03,  1.63it/s]
 33%|███▎      | 21671/65536 [3:43:12<7:19:31,  1.66it/s]
 33%|███▎      | 21672/65536 [3:43:13<7:17:42,  1.67it/s]
 33%|███▎      | 21673/65536 [3:43:13<7:13:34,  1.69it/s]
 33%|███▎      | 21674/65536 [3:43:14<7:16:11,  1.68it/s]
 33%|███▎      | 21675/65536 [3:43:15<7:11:50,  1.69it/s]
 33%|███▎      | 21676/65536 [3:43:15<7:37:31,  1.60it/s]
 33%|███▎      | 21677/65536 [3:43:16<7:31:31,  1.62it/s]
 33%|███▎      | 21678/65536 [3:43:16<7:29:58,  1.62it/s]
 33%|███▎      | 21679/65536 [3:43:17<7:30:33,  1.62it/s]
 33%|███▎      | 21680/65536 [3:43:18<7:27:39,  1.63it/s]
                                                         
{'loss': 3.0518, 'learning_rate': 7.041141177911105e-07, 'epoch': 1338.27}
+
 33%|███▎      | 21680/65536 [3:43:18<7:27:39,  1.63it/s]
 33%|███▎      | 21681/65536 [3:43:18<7:25:54,  1.64it/s]
 33%|███▎      | 21682/65536 [3:43:19<7:26:34,  1.64it/s]
 33%|███▎      | 21683/65536 [3:43:20<7:19:53,  1.66it/s]
 33%|███▎      | 21684/65536 [3:43:20<7:19:34,  1.66it/s]
 33%|███▎      | 21685/65536 [3:43:21<7:23:45,  1.65it/s]
 33%|███▎      | 21686/65536 [3:43:21<7:28:38,  1.63it/s]
 33%|███▎      | 21687/65536 [3:43:22<7:35:32,  1.60it/s]
 33%|███▎      | 21688/65536 [3:43:23<7:35:58,  1.60it/s]
 33%|███▎      | 21689/65536 [3:43:23<7:40:31,  1.59it/s]
 33%|███▎      | 21690/65536 [3:43:24<7:24:23,  1.64it/s]
 33%|███▎      | 21691/65536 [3:43:24<7:24:06,  1.65it/s]
 33%|███▎      | 21692/65536 [3:43:25<7:49:25,  1.56it/s]
 33%|███▎      | 21693/65536 [3:43:26<7:44:08,  1.57it/s]
 33%|███▎      | 21694/65536 [3:43:26<7:28:56,  1.63it/s]
 33%|███▎      | 21695/65536 [3:43:27<7:32:00,  1.62it/s]
 33%|███▎      | 21696/65536 [3:43:28<7:23:01,  1.65it/s]
 33%|███▎      | 21697/65536 [3:43:28<7:24:29,  1.64it/s]
 33%|███▎      | 21698/65536 [3:43:29<7:22:48,  1.65it/s]
 33%|███▎      | 21699/65536 [3:43:29<7:13:04,  1.69it/s]
 33%|███▎      | 21700/65536 [3:43:30<7:26:05,  1.64it/s]
                                                         
{'loss': 2.9899, 'learning_rate': 7.038386188318843e-07, 'epoch': 1339.51}
+
 33%|███▎      | 21700/65536 [3:43:30<7:26:05,  1.64it/s]
 33%|███▎      | 21701/65536 [3:43:31<7:17:07,  1.67it/s]
 33%|███▎      | 21702/65536 [3:43:31<7:13:59,  1.68it/s]
 33%|███▎      | 21703/65536 [3:43:32<7:16:20,  1.67it/s]
 33%|███▎      | 21704/65536 [3:43:32<7:11:50,  1.69it/s]
 33%|███▎      | 21705/65536 [3:43:33<7:23:26,  1.65it/s]
 33%|███▎      | 21706/65536 [3:43:34<7:29:39,  1.62it/s]
 33%|███▎      | 21707/65536 [3:43:34<7:25:05,  1.64it/s]
 33%|███▎      | 21708/65536 [3:43:35<7:34:32,  1.61it/s]
 33%|███▎      | 21709/65536 [3:43:36<7:57:55,  1.53it/s]
 33%|███▎      | 21710/65536 [3:43:36<7:48:40,  1.56it/s]
 33%|███▎      | 21711/65536 [3:43:37<7:44:27,  1.57it/s]
 33%|███▎      | 21712/65536 [3:43:37<7:38:51,  1.59it/s]
 33%|███▎      | 21713/65536 [3:43:38<7:28:34,  1.63it/s]
 33%|███▎      | 21714/65536 [3:43:39<7:21:25,  1.65it/s]
 33%|███▎      | 21715/65536 [3:43:39<7:14:57,  1.68it/s]
 33%|███▎      | 21716/65536 [3:43:40<7:24:10,  1.64it/s]
 33%|███▎      | 21717/65536 [3:43:40<7:18:35,  1.67it/s]
 33%|███▎      | 21718/65536 [3:43:41<7:24:18,  1.64it/s]
 33%|███▎      | 21719/65536 [3:43:42<7:25:39,  1.64it/s]
 33%|███▎      | 21720/65536 [3:43:42<7:27:07,  1.63it/s]
                                                         
{'loss': 3.0019, 'learning_rate': 7.035631198726582e-07, 'epoch': 1340.74}
+
 33%|███▎      | 21720/65536 [3:43:42<7:27:07,  1.63it/s]
 33%|███▎      | 21721/65536 [3:43:43<7:16:58,  1.67it/s]
 33%|███▎      | 21722/65536 [3:43:43<7:27:19,  1.63it/s]
 33%|███▎      | 21723/65536 [3:43:44<7:36:29,  1.60it/s]
 33%|███▎      | 21724/65536 [3:43:45<7:24:32,  1.64it/s]
 33%|███▎      | 21725/65536 [3:43:45<7:45:47,  1.57it/s]
 33%|███▎      | 21726/65536 [3:43:46<7:37:32,  1.60it/s]
 33%|███▎      | 21727/65536 [3:43:47<7:33:12,  1.61it/s]
 33%|███▎      | 21728/65536 [3:43:47<7:40:17,  1.59it/s]
 33%|███▎      | 21729/65536 [3:43:48<7:37:59,  1.59it/s]
 33%|███▎      | 21730/65536 [3:43:48<7:33:32,  1.61it/s]
 33%|███▎      | 21731/65536 [3:43:49<7:30:51,  1.62it/s]
 33%|███▎      | 21732/65536 [3:43:50<7:24:28,  1.64it/s]
 33%|███▎      | 21733/65536 [3:43:50<7:22:41,  1.65it/s]
 33%|███▎      | 21734/65536 [3:43:51<7:21:41,  1.65it/s]
 33%|███▎      | 21735/65536 [3:43:51<7:14:15,  1.68it/s]
 33%|███▎      | 21736/65536 [3:43:52<7:25:09,  1.64it/s]
 33%|███▎      | 21737/65536 [3:43:53<7:21:57,  1.65it/s]
 33%|███▎      | 21738/65536 [3:43:53<7:18:53,  1.66it/s]
 33%|███▎      | 21739/65536 [3:43:54<7:24:06,  1.64it/s]
 33%|███▎      | 21740/65536 [3:43:54<7:18:40,  1.66it/s]
                                                         
{'loss': 2.99, 'learning_rate': 7.032876209134321e-07, 'epoch': 1341.98}
+
 33%|███▎      | 21740/65536 [3:43:54<7:18:40,  1.66it/s]
 33%|███▎      | 21741/65536 [3:43:55<7:37:18,  1.60it/s]
 33%|███▎      | 21742/65536 [3:43:56<7:41:11,  1.58it/s]
 33%|███▎      | 21743/65536 [3:43:56<7:36:13,  1.60it/s]
 33%|███▎      | 21744/65536 [3:43:57<7:21:45,  1.65it/s]
 33%|███▎      | 21745/65536 [3:43:58<7:20:18,  1.66it/s]
 33%|███▎      | 21746/65536 [3:43:58<7:32:25,  1.61it/s]
 33%|███▎      | 21747/65536 [3:43:59<7:27:16,  1.63it/s]
 33%|███▎      | 21748/65536 [3:43:59<7:28:06,  1.63it/s]
 33%|███▎      | 21749/65536 [3:44:00<7:16:42,  1.67it/s]
 33%|███▎      | 21750/65536 [3:44:01<7:22:38,  1.65it/s]
 33%|███▎      | 21751/65536 [3:44:01<7:19:41,  1.66it/s]
 33%|███▎      | 21752/65536 [3:44:02<7:18:56,  1.66it/s]
 33%|███▎      | 21753/65536 [3:44:02<7:18:51,  1.66it/s]
 33%|███▎      | 21754/65536 [3:44:03<7:21:33,  1.65it/s]
 33%|███▎      | 21755/65536 [3:44:04<7:18:22,  1.66it/s]
 33%|███▎      | 21756/65536 [3:44:04<7:13:40,  1.68it/s]
 33%|███▎      | 21757/65536 [3:44:05<7:37:02,  1.60it/s]
 33%|███▎      | 21758/65536 [3:44:06<7:31:19,  1.62it/s]
 33%|███▎      | 21759/65536 [3:44:06<7:28:15,  1.63it/s]
 33%|███▎      | 21760/65536 [3:44:07<7:24:49,  1.64it/s]
                                                         
{'loss': 3.0385, 'learning_rate': 7.030121219542059e-07, 'epoch': 1343.21}
+
 33%|███▎      | 21760/65536 [3:44:07<7:24:49,  1.64it/s]
 33%|███▎      | 21761/65536 [3:44:07<7:26:50,  1.63it/s]
 33%|███▎      | 21762/65536 [3:44:08<7:23:51,  1.64it/s]
 33%|███▎      | 21763/65536 [3:44:09<7:25:11,  1.64it/s]
 33%|███▎      | 21764/65536 [3:44:09<7:33:40,  1.61it/s]
 33%|███▎      | 21765/65536 [3:44:10<7:22:06,  1.65it/s]
 33%|███▎      | 21766/65536 [3:44:10<7:19:54,  1.66it/s]
 33%|███▎      | 21767/65536 [3:44:11<7:22:00,  1.65it/s]
 33%|███▎      | 21768/65536 [3:44:12<7:14:43,  1.68it/s]
 33%|███▎      | 21769/65536 [3:44:12<7:08:00,  1.70it/s]
 33%|███▎      | 21770/65536 [3:44:13<7:13:22,  1.68it/s]
 33%|███▎      | 21771/65536 [3:44:13<7:10:47,  1.69it/s]
 33%|███▎      | 21772/65536 [3:44:14<7:36:41,  1.60it/s]
 33%|███▎      | 21773/65536 [3:44:15<7:43:08,  1.57it/s]
 33%|███▎      | 21774/65536 [3:44:15<7:34:34,  1.60it/s]
 33%|███▎      | 21775/65536 [3:44:16<7:36:59,  1.60it/s]
 33%|███▎      | 21776/65536 [3:44:17<7:36:12,  1.60it/s]
 33%|███▎      | 21777/65536 [3:44:17<7:32:55,  1.61it/s]
 33%|███▎      | 21778/65536 [3:44:18<7:32:02,  1.61it/s]
 33%|███▎      | 21779/65536 [3:44:18<7:33:20,  1.61it/s]
 33%|███▎      | 21780/65536 [3:44:19<7:38:32,  1.59it/s]
                                                         
{'loss': 3.0483, 'learning_rate': 7.027366229949798e-07, 'epoch': 1344.44}
+
 33%|███▎      | 21780/65536 [3:44:19<7:38:32,  1.59it/s]
 33%|███▎      | 21781/65536 [3:44:20<7:28:48,  1.62it/s]
 33%|███▎      | 21782/65536 [3:44:20<7:36:48,  1.60it/s]
 33%|███▎      | 21783/65536 [3:44:21<7:37:41,  1.59it/s]
 33%|███▎      | 21784/65536 [3:44:21<7:29:23,  1.62it/s]
 33%|███▎      | 21785/65536 [3:44:22<7:25:58,  1.64it/s]
 33%|███▎      | 21786/65536 [3:44:23<7:27:50,  1.63it/s]
 33%|███▎      | 21787/65536 [3:44:23<7:13:34,  1.68it/s]
 33%|███▎      | 21788/65536 [3:44:24<7:13:46,  1.68it/s]
 33%|███▎      | 21789/65536 [3:44:24<7:06:29,  1.71it/s]
 33%|███▎      | 21790/65536 [3:44:25<7:33:08,  1.61it/s]
 33%|███▎      | 21791/65536 [3:44:26<7:33:41,  1.61it/s]
 33%|███▎      | 21792/65536 [3:44:26<7:28:41,  1.62it/s]
 33%|███▎      | 21793/65536 [3:44:27<7:23:31,  1.64it/s]
 33%|███▎      | 21794/65536 [3:44:28<7:30:54,  1.62it/s]
 33%|███▎      | 21795/65536 [3:44:28<7:25:13,  1.64it/s]
 33%|███▎      | 21796/65536 [3:44:29<7:27:45,  1.63it/s]
 33%|███▎      | 21797/65536 [3:44:29<7:36:31,  1.60it/s]
 33%|███▎      | 21798/65536 [3:44:30<7:42:33,  1.58it/s]
 33%|███▎      | 21799/65536 [3:44:31<7:28:35,  1.62it/s]
 33%|███▎      | 21800/65536 [3:44:31<7:19:27,  1.66it/s]
                                                         
{'loss': 3.0463, 'learning_rate': 7.024611240357536e-07, 'epoch': 1345.68}
+
 33%|███▎      | 21800/65536 [3:44:31<7:19:27,  1.66it/s]
 33%|███▎      | 21801/65536 [3:44:32<7:09:44,  1.70it/s]
 33%|███▎      | 21802/65536 [3:44:32<7:19:29,  1.66it/s]
 33%|███▎      | 21803/65536 [3:44:33<7:15:34,  1.67it/s]
 33%|███▎      | 21804/65536 [3:44:34<7:15:40,  1.67it/s]
 33%|███▎      | 21805/65536 [3:44:34<7:16:45,  1.67it/s]
 33%|███▎      | 21806/65536 [3:44:35<7:35:54,  1.60it/s]
 33%|███▎      | 21807/65536 [3:44:35<7:27:06,  1.63it/s]
 33%|███▎      | 21808/65536 [3:44:36<7:17:02,  1.67it/s]
 33%|███▎      | 21809/65536 [3:44:37<7:10:34,  1.69it/s]
 33%|███▎      | 21810/65536 [3:44:37<7:08:21,  1.70it/s]
 33%|███▎      | 21811/65536 [3:44:38<7:18:07,  1.66it/s]
 33%|███▎      | 21812/65536 [3:44:38<7:25:32,  1.64it/s]
 33%|███▎      | 21813/65536 [3:44:39<7:17:14,  1.67it/s]
 33%|███▎      | 21814/65536 [3:44:40<7:19:07,  1.66it/s]
 33%|███▎      | 21815/65536 [3:44:40<7:25:20,  1.64it/s]
 33%|███▎      | 21816/65536 [3:44:41<7:26:19,  1.63it/s]
 33%|███▎      | 21817/65536 [3:44:42<7:35:10,  1.60it/s]
 33%|███▎      | 21818/65536 [3:44:42<7:34:39,  1.60it/s]
 33%|███▎      | 21819/65536 [3:44:43<7:38:28,  1.59it/s]
 33%|███▎      | 21820/65536 [3:44:43<7:32:34,  1.61it/s]
                                                         
{'loss': 3.0503, 'learning_rate': 7.021856250765275e-07, 'epoch': 1346.91}
+
 33%|███▎      | 21820/65536 [3:44:43<7:32:34,  1.61it/s]
 33%|███▎      | 21821/65536 [3:44:44<7:35:35,  1.60it/s]
 33%|███▎      | 21822/65536 [3:44:45<7:47:35,  1.56it/s]
 33%|███▎      | 21823/65536 [3:44:45<7:52:46,  1.54it/s]
 33%|███▎      | 21824/65536 [3:44:46<7:45:45,  1.56it/s]
 33%|███▎      | 21825/65536 [3:44:47<7:36:22,  1.60it/s]
 33%|███▎      | 21826/65536 [3:44:47<7:42:43,  1.57it/s]
 33%|███▎      | 21827/65536 [3:44:48<7:37:30,  1.59it/s]
 33%|███▎      | 21828/65536 [3:44:49<7:34:20,  1.60it/s]
 33%|███▎      | 21829/65536 [3:44:49<7:28:31,  1.62it/s]
 33%|███▎      | 21830/65536 [3:44:50<7:32:07,  1.61it/s]
 33%|███▎      | 21831/65536 [3:44:50<7:25:35,  1.63it/s]
 33%|███▎      | 21832/65536 [3:44:51<7:15:55,  1.67it/s]
 33%|███▎      | 21833/65536 [3:44:52<7:19:54,  1.66it/s]
 33%|███▎      | 21834/65536 [3:44:52<7:10:21,  1.69it/s]
 33%|███▎      | 21835/65536 [3:44:53<7:22:35,  1.65it/s]
 33%|███▎      | 21836/65536 [3:44:53<7:17:29,  1.66it/s]
 33%|███▎      | 21837/65536 [3:44:54<7:25:23,  1.64it/s]
 33%|███▎      | 21838/65536 [3:44:55<7:39:26,  1.59it/s]
 33%|███▎      | 21839/65536 [3:44:55<7:33:49,  1.60it/s]
 33%|███▎      | 21840/65536 [3:44:56<7:43:07,  1.57it/s]
                                                         
{'loss': 2.9832, 'learning_rate': 7.019101261173014e-07, 'epoch': 1348.15}
+
 33%|███▎      | 21840/65536 [3:44:56<7:43:07,  1.57it/s]
 33%|███▎      | 21841/65536 [3:44:56<7:37:56,  1.59it/s]
 33%|███▎      | 21842/65536 [3:44:57<7:34:35,  1.60it/s]
 33%|███▎      | 21843/65536 [3:44:58<7:17:47,  1.66it/s]
 33%|███▎      | 21844/65536 [3:44:58<7:26:56,  1.63it/s]
 33%|███▎      | 21845/65536 [3:44:59<7:21:17,  1.65it/s]
 33%|███▎      | 21846/65536 [3:44:59<7:16:56,  1.67it/s]
 33%|███▎      | 21847/65536 [3:45:00<7:27:12,  1.63it/s]
 33%|███▎      | 21848/65536 [3:45:01<7:10:45,  1.69it/s]
 33%|███▎      | 21849/65536 [3:45:01<7:10:54,  1.69it/s]
 33%|███▎      | 21850/65536 [3:45:02<7:26:10,  1.63it/s]
 33%|███▎      | 21851/65536 [3:45:03<7:28:30,  1.62it/s]
 33%|███▎      | 21852/65536 [3:45:03<7:28:47,  1.62it/s]
 33%|███▎      | 21853/65536 [3:45:04<7:18:06,  1.66it/s]
 33%|███▎      | 21854/65536 [3:45:04<7:36:09,  1.60it/s]
 33%|███▎      | 21855/65536 [3:45:05<7:32:20,  1.61it/s]
 33%|███▎      | 21856/65536 [3:45:06<7:24:17,  1.64it/s]
 33%|███▎      | 21857/65536 [3:45:06<7:17:00,  1.67it/s]
 33%|███▎      | 21858/65536 [3:45:07<7:17:29,  1.66it/s]
 33%|███▎      | 21859/65536 [3:45:07<7:19:11,  1.66it/s]
 33%|███▎      | 21860/65536 [3:45:08<7:12:31,  1.68it/s]
                                                         
{'loss': 2.9928, 'learning_rate': 7.016346271580752e-07, 'epoch': 1349.38}
+
 33%|███▎      | 21860/65536 [3:45:08<7:12:31,  1.68it/s]
 33%|███▎      | 21861/65536 [3:45:09<7:22:28,  1.65it/s]
 33%|███▎      | 21862/65536 [3:45:09<7:37:17,  1.59it/s]
 33%|███▎      | 21863/65536 [3:45:10<7:33:44,  1.60it/s]
 33%|███▎      | 21864/65536 [3:45:10<7:23:37,  1.64it/s]
 33%|███▎      | 21865/65536 [3:45:11<7:21:45,  1.65it/s]
 33%|███▎      | 21866/65536 [3:45:12<7:20:37,  1.65it/s]
 33%|███▎      | 21867/65536 [3:45:12<7:14:43,  1.67it/s]
 33%|███▎      | 21868/65536 [3:45:13<7:12:19,  1.68it/s]
 33%|███▎      | 21869/65536 [3:45:13<7:19:25,  1.66it/s]
 33%|███▎      | 21870/65536 [3:45:14<7:33:35,  1.60it/s]
 33%|███▎      | 21871/65536 [3:45:15<7:57:25,  1.52it/s]
 33%|███▎      | 21872/65536 [3:45:15<7:38:38,  1.59it/s]
 33%|███▎      | 21873/65536 [3:45:16<7:37:52,  1.59it/s]
 33%|███▎      | 21874/65536 [3:45:17<7:34:53,  1.60it/s]
 33%|███▎      | 21875/65536 [3:45:17<7:43:01,  1.57it/s]
 33%|███▎      | 21876/65536 [3:45:18<7:33:46,  1.60it/s]
 33%|███▎      | 21877/65536 [3:45:19<7:34:32,  1.60it/s]
 33%|███▎      | 21878/65536 [3:45:19<7:23:43,  1.64it/s]
 33%|███▎      | 21879/65536 [3:45:20<7:21:28,  1.65it/s]
 33%|███▎      | 21880/65536 [3:45:20<7:23:44,  1.64it/s]
                                                         
{'loss': 3.0193, 'learning_rate': 7.01359128198849e-07, 'epoch': 1350.62}
+
 33%|███▎      | 21880/65536 [3:45:20<7:23:44,  1.64it/s]
 33%|███▎      | 21881/65536 [3:45:21<7:13:37,  1.68it/s]
 33%|███▎      | 21882/65536 [3:45:22<7:24:03,  1.64it/s]
 33%|███▎      | 21883/65536 [3:45:22<7:31:39,  1.61it/s]
 33%|███▎      | 21884/65536 [3:45:23<7:25:03,  1.63it/s]
 33%|███▎      | 21885/65536 [3:45:23<7:21:05,  1.65it/s]
 33%|███▎      | 21886/65536 [3:45:24<7:17:13,  1.66it/s]
 33%|███▎      | 21887/65536 [3:45:25<7:24:25,  1.64it/s]
 33%|███▎      | 21888/65536 [3:45:25<7:20:56,  1.65it/s]
 33%|███▎      | 21889/65536 [3:45:26<7:23:52,  1.64it/s]
 33%|███▎      | 21890/65536 [3:45:26<7:30:44,  1.61it/s]
 33%|███▎      | 21891/65536 [3:45:27<7:15:18,  1.67it/s]
 33%|███▎      | 21892/65536 [3:45:28<7:08:54,  1.70it/s]
 33%|███▎      | 21893/65536 [3:45:28<7:19:53,  1.65it/s]
 33%|███▎      | 21894/65536 [3:45:29<7:26:22,  1.63it/s]
 33%|███▎      | 21895/65536 [3:45:30<7:43:12,  1.57it/s]
 33%|███▎      | 21896/65536 [3:45:30<7:44:40,  1.57it/s]
 33%|███▎      | 21897/65536 [3:45:31<7:35:31,  1.60it/s]
 33%|███▎      | 21898/65536 [3:45:31<7:32:22,  1.61it/s]
 33%|███▎      | 21899/65536 [3:45:32<7:29:16,  1.62it/s]
 33%|███▎      | 21900/65536 [3:45:33<7:33:59,  1.60it/s]
                                                         
{'loss': 3.0242, 'learning_rate': 7.010836292396228e-07, 'epoch': 1351.85}
+
 33%|███▎      | 21900/65536 [3:45:33<7:33:59,  1.60it/s]
 33%|███▎      | 21901/65536 [3:45:33<7:22:41,  1.64it/s]
 33%|███▎      | 21902/65536 [3:45:34<7:06:48,  1.70it/s]
 33%|███▎      | 21903/65536 [3:45:34<7:18:18,  1.66it/s]
 33%|███▎      | 21904/65536 [3:45:35<7:20:49,  1.65it/s]
 33%|███▎      | 21905/65536 [3:45:36<7:21:39,  1.65it/s]
 33%|███▎      | 21906/65536 [3:45:36<7:34:09,  1.60it/s]
 33%|███▎      | 21907/65536 [3:45:37<7:20:37,  1.65it/s]
 33%|███▎      | 21908/65536 [3:45:37<7:23:38,  1.64it/s]
 33%|███▎      | 21909/65536 [3:45:38<7:24:08,  1.64it/s]
 33%|███▎      | 21910/65536 [3:45:39<7:23:44,  1.64it/s]
 33%|███▎      | 21911/65536 [3:45:39<7:34:34,  1.60it/s]
 33%|███▎      | 21912/65536 [3:45:40<7:31:06,  1.61it/s]
 33%|███▎      | 21913/65536 [3:45:41<7:24:40,  1.64it/s]
 33%|███▎      | 21914/65536 [3:45:41<7:15:29,  1.67it/s]
 33%|███▎      | 21915/65536 [3:45:42<7:11:16,  1.69it/s]
 33%|███▎      | 21916/65536 [3:45:42<7:15:30,  1.67it/s]
 33%|███▎      | 21917/65536 [3:45:43<7:26:17,  1.63it/s]
 33%|███▎      | 21918/65536 [3:45:44<7:21:51,  1.65it/s]
 33%|███▎      | 21919/65536 [3:45:44<7:29:02,  1.62it/s]
 33%|███▎      | 21920/65536 [3:45:45<7:23:58,  1.64it/s]
                                                         
{'loss': 3.0383, 'learning_rate': 7.008081302803967e-07, 'epoch': 1353.09}
+
 33%|███▎      | 21920/65536 [3:45:45<7:23:58,  1.64it/s]
 33%|███▎      | 21921/65536 [3:45:45<7:39:53,  1.58it/s]
 33%|███▎      | 21922/65536 [3:45:46<7:24:36,  1.63it/s]
 33%|███▎      | 21923/65536 [3:45:47<7:19:09,  1.66it/s]
 33%|███▎      | 21924/65536 [3:45:47<7:25:45,  1.63it/s]
 33%|███▎      | 21925/65536 [3:45:48<7:27:07,  1.63it/s]
 33%|███▎      | 21926/65536 [3:45:48<7:23:07,  1.64it/s]
 33%|███▎      | 21927/65536 [3:45:49<7:18:31,  1.66it/s]
 33%|███▎      | 21928/65536 [3:45:50<7:23:46,  1.64it/s]
 33%|███▎      | 21929/65536 [3:45:50<7:29:47,  1.62it/s]
 33%|███▎      | 21930/65536 [3:45:51<7:32:25,  1.61it/s]
 33%|███▎      | 21931/65536 [3:45:52<7:35:17,  1.60it/s]
 33%|███▎      | 21932/65536 [3:45:52<7:28:04,  1.62it/s]
 33%|███▎      | 21933/65536 [3:45:53<7:26:29,  1.63it/s]
 33%|███▎      | 21934/65536 [3:45:53<7:21:07,  1.65it/s]
 33%|███▎      | 21935/65536 [3:45:54<7:30:19,  1.61it/s]
 33%|███▎      | 21936/65536 [3:45:55<7:23:50,  1.64it/s]
 33%|███▎      | 21937/65536 [3:45:55<7:19:59,  1.65it/s]
 33%|███▎      | 21938/65536 [3:45:56<7:26:38,  1.63it/s]
 33%|███▎      | 21939/65536 [3:45:56<7:32:43,  1.60it/s]
 33%|███▎      | 21940/65536 [3:45:57<7:25:13,  1.63it/s]
                                                         
{'loss': 3.0641, 'learning_rate': 7.005326313211704e-07, 'epoch': 1354.32}
+
 33%|███▎      | 21940/65536 [3:45:57<7:25:13,  1.63it/s]
 33%|███▎      | 21941/65536 [3:45:58<7:31:13,  1.61it/s]
 33%|███▎      | 21942/65536 [3:45:58<7:23:37,  1.64it/s]
 33%|███▎      | 21943/65536 [3:45:59<7:16:28,  1.66it/s]
 33%|███▎      | 21944/65536 [3:45:59<7:15:14,  1.67it/s]
 33%|███▎      | 21945/65536 [3:46:00<7:20:08,  1.65it/s]
 33%|███▎      | 21946/65536 [3:46:01<7:23:56,  1.64it/s]
 33%|███▎      | 21947/65536 [3:46:01<7:24:18,  1.64it/s]
 33%|███▎      | 21948/65536 [3:46:02<7:30:38,  1.61it/s]
 33%|███▎      | 21949/65536 [3:46:03<7:25:28,  1.63it/s]
 33%|███▎      | 21950/65536 [3:46:03<7:24:56,  1.63it/s]
 33%|███▎      | 21951/65536 [3:46:04<7:17:51,  1.66it/s]
 33%|███▎      | 21952/65536 [3:46:04<7:29:31,  1.62it/s]
 33%|███▎      | 21953/65536 [3:46:05<7:25:19,  1.63it/s]
 33%|███▎      | 21954/65536 [3:46:06<7:26:26,  1.63it/s]
 34%|███▎      | 21955/65536 [3:46:06<7:11:07,  1.68it/s]
 34%|███▎      | 21956/65536 [3:46:07<7:18:33,  1.66it/s]
 34%|███▎      | 21957/65536 [3:46:07<7:23:47,  1.64it/s]
 34%|███▎      | 21958/65536 [3:46:08<7:34:54,  1.60it/s]
 34%|███▎      | 21959/65536 [3:46:09<7:35:46,  1.59it/s]
 34%|███▎      | 21960/65536 [3:46:09<7:16:23,  1.66it/s]
                                                         
{'loss': 2.9852, 'learning_rate': 7.002571323619443e-07, 'epoch': 1355.56}
+
 34%|███▎      | 21960/65536 [3:46:09<7:16:23,  1.66it/s]
 34%|███▎      | 21961/65536 [3:46:10<7:10:49,  1.69it/s]
 34%|███▎      | 21962/65536 [3:46:11<7:28:33,  1.62it/s]
 34%|███▎      | 21963/65536 [3:46:11<7:39:42,  1.58it/s]
 34%|███▎      | 21964/65536 [3:46:12<7:34:27,  1.60it/s]
 34%|███▎      | 21965/65536 [3:46:12<7:26:36,  1.63it/s]
 34%|███▎      | 21966/65536 [3:46:13<7:17:43,  1.66it/s]
 34%|███▎      | 21967/65536 [3:46:14<7:10:22,  1.69it/s]
 34%|███▎      | 21968/65536 [3:46:14<7:25:29,  1.63it/s]
 34%|███▎      | 21969/65536 [3:46:15<7:17:12,  1.66it/s]
 34%|███▎      | 21970/65536 [3:46:15<7:17:26,  1.66it/s]
 34%|███▎      | 21971/65536 [3:46:16<7:25:22,  1.63it/s]
 34%|███▎      | 21972/65536 [3:46:17<7:23:15,  1.64it/s]
 34%|███▎      | 21973/65536 [3:46:17<7:24:44,  1.63it/s]
 34%|███▎      | 21974/65536 [3:46:18<7:22:54,  1.64it/s]
 34%|███▎      | 21975/65536 [3:46:18<7:15:34,  1.67it/s]
 34%|███▎      | 21976/65536 [3:46:19<7:07:30,  1.70it/s]
 34%|███▎      | 21977/65536 [3:46:20<7:12:42,  1.68it/s]
 34%|███▎      | 21978/65536 [3:46:20<7:12:57,  1.68it/s]
 34%|███▎      | 21979/65536 [3:46:21<7:23:19,  1.64it/s]
 34%|███▎      | 21980/65536 [3:46:21<7:16:21,  1.66it/s]
                                                         
{'loss': 3.0579, 'learning_rate': 6.999816334027183e-07, 'epoch': 1356.79}
+
 34%|███▎      | 21980/65536 [3:46:21<7:16:21,  1.66it/s]
 34%|███▎      | 21981/65536 [3:46:22<7:37:01,  1.59it/s]
 34%|███▎      | 21982/65536 [3:46:23<7:30:08,  1.61it/s]
 34%|███▎      | 21983/65536 [3:46:23<7:21:06,  1.65it/s]
 34%|███▎      | 21984/65536 [3:46:24<7:31:34,  1.61it/s]
 34%|███▎      | 21985/65536 [3:46:25<7:26:39,  1.63it/s]
 34%|███▎      | 21986/65536 [3:46:25<7:26:53,  1.62it/s]
 34%|███▎      | 21987/65536 [3:46:26<7:30:43,  1.61it/s]
 34%|███▎      | 21988/65536 [3:46:26<7:26:32,  1.63it/s]
 34%|███▎      | 21989/65536 [3:46:27<7:27:20,  1.62it/s]
 34%|███▎      | 21990/65536 [3:46:28<7:31:50,  1.61it/s]
 34%|███▎      | 21991/65536 [3:46:28<7:23:50,  1.64it/s]
 34%|███▎      | 21992/65536 [3:46:29<7:23:53,  1.63it/s]
 34%|███▎      | 21993/65536 [3:46:30<7:34:47,  1.60it/s]
 34%|███▎      | 21994/65536 [3:46:30<7:33:47,  1.60it/s]
 34%|███▎      | 21995/65536 [3:46:31<7:28:15,  1.62it/s]
 34%|███▎      | 21996/65536 [3:46:31<7:22:02,  1.64it/s]
 34%|███▎      | 21997/65536 [3:46:32<7:12:02,  1.68it/s]
 34%|███▎      | 21998/65536 [3:46:32<7:06:12,  1.70it/s]
 34%|███▎      | 21999/65536 [3:46:33<7:07:03,  1.70it/s]
 34%|███▎      | 22000/65536 [3:46:34<7:25:30,  1.63it/s]
                                                         
{'loss': 3.1143, 'learning_rate': 6.99706134443492e-07, 'epoch': 1358.02}
+
 34%|███▎      | 22000/65536 [3:46:34<7:25:30,  1.63it/s]
 34%|███▎      | 22001/65536 [3:46:34<7:30:49,  1.61it/s]
 34%|███▎      | 22002/65536 [3:46:35<7:26:17,  1.63it/s]
 34%|███▎      | 22003/65536 [3:46:36<7:20:24,  1.65it/s]
 34%|███▎      | 22004/65536 [3:46:36<7:17:20,  1.66it/s]
 34%|███▎      | 22005/65536 [3:46:37<7:18:25,  1.65it/s]
 34%|███▎      | 22006/65536 [3:46:37<7:11:54,  1.68it/s]
 34%|███▎      | 22007/65536 [3:46:38<7:09:18,  1.69it/s]
 34%|███▎      | 22008/65536 [3:46:38<7:10:34,  1.68it/s]
 34%|███▎      | 22009/65536 [3:46:39<7:04:46,  1.71it/s]
 34%|███▎      | 22010/65536 [3:46:40<7:01:14,  1.72it/s]
 34%|███▎      | 22011/65536 [3:46:40<7:19:10,  1.65it/s]
 34%|███▎      | 22012/65536 [3:46:41<7:25:49,  1.63it/s]
 34%|███▎      | 22013/65536 [3:46:41<7:12:17,  1.68it/s]
 34%|███▎      | 22014/65536 [3:46:42<7:19:19,  1.65it/s]
 34%|███▎      | 22015/65536 [3:46:43<7:20:59,  1.64it/s]
 34%|███▎      | 22016/65536 [3:46:43<7:40:52,  1.57it/s]
 34%|███▎      | 22017/65536 [3:46:44<7:38:18,  1.58it/s]
 34%|███▎      | 22018/65536 [3:46:45<7:38:03,  1.58it/s]
 34%|███▎      | 22019/65536 [3:46:45<7:36:08,  1.59it/s]
 34%|███▎      | 22020/65536 [3:46:46<7:30:29,  1.61it/s]
                                                         
{'loss': 2.9811, 'learning_rate': 6.994306354842659e-07, 'epoch': 1359.26}
+
 34%|███▎      | 22020/65536 [3:46:46<7:30:29,  1.61it/s]
 34%|███▎      | 22021/65536 [3:46:46<7:24:04,  1.63it/s]
 34%|███▎      | 22022/65536 [3:46:47<7:31:10,  1.61it/s]
 34%|███▎      | 22023/65536 [3:46:48<7:27:07,  1.62it/s]
 34%|███▎      | 22024/65536 [3:46:48<7:23:59,  1.63it/s]
 34%|███▎      | 22025/65536 [3:46:49<7:23:00,  1.64it/s]
 34%|███▎      | 22026/65536 [3:46:50<7:20:00,  1.65it/s]
 34%|███▎      | 22027/65536 [3:46:50<7:20:09,  1.65it/s]
 34%|███▎      | 22028/65536 [3:46:51<7:22:21,  1.64it/s]
 34%|███▎      | 22029/65536 [3:46:51<7:22:10,  1.64it/s]
 34%|███▎      | 22030/65536 [3:46:52<7:35:53,  1.59it/s]
 34%|███▎      | 22031/65536 [3:46:53<7:25:16,  1.63it/s]
 34%|███▎      | 22032/65536 [3:46:53<7:16:34,  1.66it/s]
 34%|███▎      | 22033/65536 [3:46:54<7:37:00,  1.59it/s]
 34%|███▎      | 22034/65536 [3:46:55<7:39:14,  1.58it/s]
 34%|███▎      | 22035/65536 [3:46:55<7:32:38,  1.60it/s]
 34%|███��      | 22036/65536 [3:46:56<7:27:33,  1.62it/s]
 34%|███▎      | 22037/65536 [3:46:56<7:26:28,  1.62it/s]
 34%|███▎      | 22038/65536 [3:46:57<7:25:32,  1.63it/s]
 34%|███▎      | 22039/65536 [3:46:58<7:29:38,  1.61it/s]
 34%|███▎      | 22040/65536 [3:46:58<7:24:56,  1.63it/s]
                                                         
{'loss': 2.9968, 'learning_rate': 6.991551365250397e-07, 'epoch': 1360.49}
+
 34%|███▎      | 22040/65536 [3:46:58<7:24:56,  1.63it/s]
 34%|███▎      | 22041/65536 [3:46:59<7:35:49,  1.59it/s]
 34%|███▎      | 22042/65536 [3:46:59<7:29:32,  1.61it/s]
 34%|███▎      | 22043/65536 [3:47:00<7:29:00,  1.61it/s]
 34%|███▎      | 22044/65536 [3:47:01<7:24:46,  1.63it/s]
 34%|███▎      | 22045/65536 [3:47:01<7:25:41,  1.63it/s]
 34%|███▎      | 22046/65536 [3:47:02<7:34:25,  1.60it/s]
 34%|███▎      | 22047/65536 [3:47:03<7:31:55,  1.60it/s]
 34%|███▎      | 22048/65536 [3:47:03<7:25:11,  1.63it/s]
 34%|███▎      | 22049/65536 [3:47:04<7:46:54,  1.55it/s]
 34%|███▎      | 22050/65536 [3:47:04<7:28:05,  1.62it/s]
 34%|███▎      | 22051/65536 [3:47:05<7:40:38,  1.57it/s]
 34%|███▎      | 22052/65536 [3:47:06<7:40:46,  1.57it/s]
 34%|███▎      | 22053/65536 [3:47:06<7:32:31,  1.60it/s]
 34%|███▎      | 22054/65536 [3:47:07<7:21:17,  1.64it/s]
 34%|███▎      | 22055/65536 [3:47:08<7:23:57,  1.63it/s]
 34%|███▎      | 22056/65536 [3:47:08<7:26:15,  1.62it/s]
 34%|███▎      | 22057/65536 [3:47:09<7:18:04,  1.65it/s]
 34%|███▎      | 22058/65536 [3:47:09<7:19:00,  1.65it/s]
 34%|███▎      | 22059/65536 [3:47:10<7:27:11,  1.62it/s]
 34%|███▎      | 22060/65536 [3:47:11<7:25:20,  1.63it/s]
                                                         
{'loss': 2.9376, 'learning_rate': 6.988796375658136e-07, 'epoch': 1361.73}
+
 34%|███▎      | 22060/65536 [3:47:11<7:25:20,  1.63it/s]
 34%|███▎      | 22061/65536 [3:47:11<7:21:13,  1.64it/s]
 34%|███▎      | 22062/65536 [3:47:12<7:25:31,  1.63it/s]
 34%|███▎      | 22063/65536 [3:47:12<7:21:52,  1.64it/s]
 34%|███▎      | 22064/65536 [3:47:13<7:22:45,  1.64it/s]
 34%|███▎      | 22065/65536 [3:47:14<7:40:39,  1.57it/s]
 34%|███▎      | 22066/65536 [3:47:14<7:30:46,  1.61it/s]
 34%|███▎      | 22067/65536 [3:47:15<7:31:15,  1.61it/s]
 34%|███▎      | 22068/65536 [3:47:16<7:26:14,  1.62it/s]
 34%|███▎      | 22069/65536 [3:47:16<7:26:11,  1.62it/s]
 34%|███▎      | 22070/65536 [3:47:17<7:35:37,  1.59it/s]
 34%|███▎      | 22071/65536 [3:47:17<7:36:50,  1.59it/s]
 34%|███▎      | 22072/65536 [3:47:18<7:23:11,  1.63it/s]
 34%|███▎      | 22073/65536 [3:47:19<7:19:17,  1.65it/s]
 34%|███▎      | 22074/65536 [3:47:19<7:23:38,  1.63it/s]
 34%|███▎      | 22075/65536 [3:47:20<7:30:19,  1.61it/s]
 34%|███▎      | 22076/65536 [3:47:20<7:19:17,  1.65it/s]
 34%|███▎      | 22077/65536 [3:47:21<7:13:01,  1.67it/s]
 34%|███▎      | 22078/65536 [3:47:22<7:20:39,  1.64it/s]
 34%|███▎      | 22079/65536 [3:47:22<7:18:12,  1.65it/s]
 34%|███▎      | 22080/65536 [3:47:23<7:18:38,  1.65it/s]
                                                         
{'loss': 2.7708, 'learning_rate': 6.986041386065874e-07, 'epoch': 1362.96}
+
 34%|███▎      | 22080/65536 [3:47:23<7:18:38,  1.65it/s]
 34%|███▎      | 22081/65536 [3:47:23<7:18:04,  1.65it/s]
 34%|███▎      | 22082/65536 [3:47:24<7:14:39,  1.67it/s]
 34%|███▎      | 22083/65536 [3:47:25<7:12:45,  1.67it/s]
 34%|███▎      | 22084/65536 [3:47:25<7:13:18,  1.67it/s]
 34%|███▎      | 22085/65536 [3:47:26<7:18:46,  1.65it/s]
 34%|███▎      | 22086/65536 [3:47:27<7:35:41,  1.59it/s]
 34%|███▎      | 22087/65536 [3:47:27<7:26:28,  1.62it/s]
 34%|███▎      | 22088/65536 [3:47:28<7:15:19,  1.66it/s]
 34%|███▎      | 22089/65536 [3:47:28<7:31:45,  1.60it/s]
 34%|███▎      | 22090/65536 [3:47:29<7:38:31,  1.58it/s]
 34%|███▎      | 22091/65536 [3:47:30<7:25:47,  1.62it/s]
 34%|███▎      | 22092/65536 [3:47:30<7:20:16,  1.64it/s]
 34%|███▎      | 22093/65536 [3:47:31<7:28:24,  1.61it/s]
 34%|███▎      | 22094/65536 [3:47:31<7:28:38,  1.61it/s]
 34%|███▎      | 22095/65536 [3:47:32<7:25:01,  1.63it/s]
 34%|███▎      | 22096/65536 [3:47:33<7:21:22,  1.64it/s]
 34%|███▎      | 22097/65536 [3:47:33<7:39:21,  1.58it/s]
 34%|███▎      | 22098/65536 [3:47:34<7:31:01,  1.61it/s]
 34%|███▎      | 22099/65536 [3:47:35<7:31:44,  1.60it/s]
 34%|███▎      | 22100/65536 [3:47:35<7:31:35,  1.60it/s]
                                                         
{'loss': 2.7872, 'learning_rate': 6.983286396473613e-07, 'epoch': 1364.2}
+
 34%|███▎      | 22100/65536 [3:47:35<7:31:35,  1.60it/s]
 34%|███▎      | 22101/65536 [3:47:36<7:24:10,  1.63it/s]
 34%|███▎      | 22102/65536 [3:47:36<7:33:02,  1.60it/s]
 34%|███▎      | 22103/65536 [3:47:37<7:23:17,  1.63it/s]
 34%|███▎      | 22104/65536 [3:47:38<7:31:09,  1.60it/s]
 34%|███▎      | 22105/65536 [3:47:38<7:21:31,  1.64it/s]
 34%|███▎      | 22106/65536 [3:47:39<7:23:44,  1.63it/s]
 34%|███▎      | 22107/65536 [3:47:40<7:28:33,  1.61it/s]
 34%|███▎      | 22108/65536 [3:47:40<7:23:19,  1.63it/s]
 34%|███▎      | 22109/65536 [3:47:41<7:23:17,  1.63it/s]
 34%|███▎      | 22110/65536 [3:47:41<7:23:16,  1.63it/s]
 34%|███▎      | 22111/65536 [3:47:42<7:15:50,  1.66it/s]
 34%|███▎      | 22112/65536 [3:47:42<7:06:33,  1.70it/s]
 34%|███▎      | 22113/65536 [3:47:43<7:15:01,  1.66it/s]
 34%|███▎      | 22114/65536 [3:47:44<7:44:19,  1.56it/s]
 34%|███▎      | 22115/65536 [3:47:45<7:51:28,  1.53it/s]
 34%|███▎      | 22116/65536 [3:47:45<7:48:06,  1.55it/s]
 34%|███▎      | 22117/65536 [3:47:46<7:40:46,  1.57it/s]
 34%|███▎      | 22118/65536 [3:47:46<7:23:01,  1.63it/s]
 34%|███▍      | 22119/65536 [3:47:47<7:16:47,  1.66it/s]
 34%|███▍      | 22120/65536 [3:47:48<7:11:19,  1.68it/s]
                                                         
{'loss': 2.9491, 'learning_rate': 6.980531406881352e-07, 'epoch': 1365.43}
+
 34%|███▍      | 22120/65536 [3:47:48<7:11:19,  1.68it/s]
 34%|███▍      | 22121/65536 [3:47:48<7:17:03,  1.66it/s]
 34%|███▍      | 22122/65536 [3:47:49<7:16:22,  1.66it/s]
 34%|███▍      | 22123/65536 [3:47:49<7:17:04,  1.66it/s]
 34%|███▍      | 22124/65536 [3:47:50<7:12:13,  1.67it/s]
 34%|███▍      | 22125/65536 [3:47:51<7:11:18,  1.68it/s]
 34%|███▍      | 22126/65536 [3:47:51<7:12:33,  1.67it/s]
 34%|███▍      | 22127/65536 [3:47:52<7:15:03,  1.66it/s]
 34%|███▍      | 22128/65536 [3:47:52<7:06:52,  1.69it/s]
 34%|███▍      | 22129/65536 [3:47:53<7:10:46,  1.68it/s]
 34%|███▍      | 22130/65536 [3:47:54<7:21:28,  1.64it/s]
 34%|███▍      | 22131/65536 [3:47:54<7:19:07,  1.65it/s]
 34%|███▍      | 22132/65536 [3:47:55<7:31:05,  1.60it/s]
 34%|███▍      | 22133/65536 [3:47:55<7:23:37,  1.63it/s]
 34%|███▍      | 22134/65536 [3:47:56<7:21:12,  1.64it/s]
 34%|███▍      | 22135/65536 [3:47:57<7:15:43,  1.66it/s]
 34%|███▍      | 22136/65536 [3:47:57<7:16:22,  1.66it/s]
 34%|███▍      | 22137/65536 [3:47:58<7:26:14,  1.62it/s]
 34%|███▍      | 22138/65536 [3:47:58<7:23:02,  1.63it/s]
 34%|███▍      | 22139/65536 [3:47:59<7:24:29,  1.63it/s]
 34%|███▍      | 22140/65536 [3:48:00<7:19:39,  1.65it/s]
                                                         
{'loss': 2.9494, 'learning_rate': 6.977776417289089e-07, 'epoch': 1366.67}
+
 34%|███▍      | 22140/65536 [3:48:00<7:19:39,  1.65it/s]
 34%|███▍      | 22141/65536 [3:48:00<7:20:12,  1.64it/s]
 34%|███▍      | 22142/65536 [3:48:01<7:18:40,  1.65it/s]
 34%|███▍      | 22143/65536 [3:48:01<7:27:05,  1.62it/s]
 34%|███▍      | 22144/65536 [3:48:02<7:20:21,  1.64it/s]
 34%|███▍      | 22145/65536 [3:48:03<7:23:14,  1.63it/s]
 34%|███▍      | 22146/65536 [3:48:03<7:34:42,  1.59it/s]
 34%|███▍      | 22147/65536 [3:48:04<7:27:20,  1.62it/s]
 34%|███▍      | 22148/65536 [3:48:05<7:20:41,  1.64it/s]
 34%|███▍      | 22149/65536 [3:48:05<7:27:43,  1.62it/s]
 34%|███▍      | 22150/65536 [3:48:06<7:21:52,  1.64it/s]
 34%|███▍      | 22151/65536 [3:48:06<7:11:25,  1.68it/s]
 34%|███▍      | 22152/65536 [3:48:07<7:17:41,  1.65it/s]
 34%|███▍      | 22153/65536 [3:48:08<7:07:40,  1.69it/s]
 34%|███▍      | 22154/65536 [3:48:08<7:11:15,  1.68it/s]
 34%|███▍      | 22155/65536 [3:48:09<7:07:05,  1.69it/s]
 34%|███▍      | 22156/65536 [3:48:09<7:11:03,  1.68it/s]
 34%|███▍      | 22157/65536 [3:48:10<7:06:31,  1.70it/s]
 34%|███▍      | 22158/65536 [3:48:11<7:11:56,  1.67it/s]
 34%|███▍      | 22159/65536 [3:48:11<7:34:28,  1.59it/s]
 34%|███▍      | 22160/65536 [3:48:12<7:45:10,  1.55it/s]
                                                         
{'loss': 2.952, 'learning_rate': 6.975021427696829e-07, 'epoch': 1367.9}
+
 34%|███▍      | 22160/65536 [3:48:12<7:45:10,  1.55it/s]
 34%|███▍      | 22161/65536 [3:48:12<7:33:09,  1.60it/s]
 34%|███▍      | 22162/65536 [3:48:13<7:43:28,  1.56it/s]
 34%|███▍      | 22163/65536 [3:48:14<7:34:51,  1.59it/s]
 34%|███▍      | 22164/65536 [3:48:14<7:24:55,  1.62it/s]
 34%|███▍      | 22165/65536 [3:48:15<7:18:19,  1.65it/s]
 34%|███▍      | 22166/65536 [3:48:16<7:19:14,  1.65it/s]
 34%|███▍      | 22167/65536 [3:48:16<7:18:31,  1.65it/s]
 34%|███▍      | 22168/65536 [3:48:17<7:32:53,  1.60it/s]
 34%|███▍      | 22169/65536 [3:48:17<7:26:58,  1.62it/s]
 34%|███▍      | 22170/65536 [3:48:18<7:26:15,  1.62it/s]
 34%|███▍      | 22171/65536 [3:48:19<7:20:36,  1.64it/s]
 34%|███▍      | 22172/65536 [3:48:19<7:27:56,  1.61it/s]
 34%|███▍      | 22173/65536 [3:48:20<7:17:47,  1.65it/s]
 34%|███▍      | 22174/65536 [3:48:20<7:17:48,  1.65it/s]
 34%|███▍      | 22175/65536 [3:48:21<7:27:41,  1.61it/s]
 34%|███▍      | 22176/65536 [3:48:22<7:26:46,  1.62it/s]
 34%|███▍      | 22177/65536 [3:48:22<7:24:07,  1.63it/s]
 34%|███▍      | 22178/65536 [3:48:23<7:35:52,  1.59it/s]
 34%|███▍      | 22179/65536 [3:48:24<7:24:05,  1.63it/s]
 34%|███▍      | 22180/65536 [3:48:24<7:21:31,  1.64it/s]
                                                         
{'loss': 2.8546, 'learning_rate': 6.972266438104567e-07, 'epoch': 1369.14}
+
 34%|███▍      | 22180/65536 [3:48:24<7:21:31,  1.64it/s]
 34%|███▍      | 22181/65536 [3:48:25<7:27:42,  1.61it/s]
 34%|███▍      | 22182/65536 [3:48:25<7:27:45,  1.61it/s]
 34%|███▍      | 22183/65536 [3:48:26<7:22:08,  1.63it/s]
 34%|███▍      | 22184/65536 [3:48:27<7:28:55,  1.61it/s]
 34%|███▍      | 22185/65536 [3:48:27<7:29:05,  1.61it/s]
 34%|███▍      | 22186/65536 [3:48:28<7:21:45,  1.64it/s]
 34%|███▍      | 22187/65536 [3:48:28<7:22:42,  1.63it/s]
 34%|███▍      | 22188/65536 [3:48:29<7:16:28,  1.66it/s]
 34%|███▍      | 22189/65536 [3:48:30<7:18:22,  1.65it/s]
 34%|███▍      | 22190/65536 [3:48:30<7:21:23,  1.64it/s]
 34%|███▍      | 22191/65536 [3:48:31<7:12:16,  1.67it/s]
 34%|███▍      | 22192/65536 [3:48:32<7:27:29,  1.61it/s]
 34%|███▍      | 22193/65536 [3:48:32<7:25:51,  1.62it/s]
 34%|███▍      | 22194/65536 [3:48:33<7:16:40,  1.65it/s]
 34%|███▍      | 22195/65536 [3:48:33<7:30:04,  1.60it/s]
 34%|███▍      | 22196/65536 [3:48:34<7:32:02,  1.60it/s]
 34%|███▍      | 22197/65536 [3:48:35<7:23:16,  1.63it/s]
 34%|███▍      | 22198/65536 [3:48:35<7:24:19,  1.63it/s]
 34%|███▍      | 22199/65536 [3:48:36<7:27:39,  1.61it/s]
 34%|███▍      | 22200/65536 [3:48:37<7:29:38,  1.61it/s]
                                                         
{'loss': 2.9859, 'learning_rate': 6.969511448512306e-07, 'epoch': 1370.37}
+
 34%|███▍      | 22200/65536 [3:48:37<7:29:38,  1.61it/s]
 34%|███▍      | 22201/65536 [3:48:37<7:21:21,  1.64it/s]
 34%|███▍      | 22202/65536 [3:48:38<7:29:23,  1.61it/s]
 34%|███▍      | 22203/65536 [3:48:38<7:32:11,  1.60it/s]
 34%|███▍      | 22204/65536 [3:48:39<7:22:32,  1.63it/s]
 34%|███▍      | 22205/65536 [3:48:40<7:29:06,  1.61it/s]
 34%|███▍      | 22206/65536 [3:48:40<7:27:21,  1.61it/s]
 34%|███▍      | 22207/65536 [3:48:41<7:21:50,  1.63it/s]
 34%|███▍      | 22208/65536 [3:48:41<7:16:12,  1.66it/s]
 34%|███▍      | 22209/65536 [3:48:42<7:22:21,  1.63it/s]
 34%|███▍      | 22210/65536 [3:48:43<7:20:40,  1.64it/s]
 34%|███▍      | 22211/65536 [3:48:43<7:33:29,  1.59it/s]
 34%|███▍      | 22212/65536 [3:48:44<7:41:45,  1.56it/s]
 34%|███▍      | 22213/65536 [3:48:45<7:33:40,  1.59it/s]
 34%|███▍      | 22214/65536 [3:48:45<7:34:05,  1.59it/s]
 34%|███▍      | 22215/65536 [3:48:46<7:30:48,  1.60it/s]
 34%|███▍      | 22216/65536 [3:48:46<7:16:59,  1.65it/s]
 34%|███▍      | 22217/65536 [3:48:47<7:24:54,  1.62it/s]
 34%|███▍      | 22218/65536 [3:48:48<7:15:43,  1.66it/s]
 34%|███▍      | 22219/65536 [3:48:48<7:15:48,  1.66it/s]
 34%|███▍      | 22220/65536 [3:48:49<7:13:02,  1.67it/s]
                                                         
{'loss': 3.0157, 'learning_rate': 6.966756458920044e-07, 'epoch': 1371.6}
+
 34%|███▍      | 22220/65536 [3:48:49<7:13:02,  1.67it/s]
 34%|███▍      | 22221/65536 [3:48:49<7:15:24,  1.66it/s]
 34%|███▍      | 22222/65536 [3:48:50<7:13:27,  1.67it/s]
 34%|███▍      | 22223/65536 [3:48:51<7:09:31,  1.68it/s]
 34%|███▍      | 22224/65536 [3:48:51<7:20:14,  1.64it/s]
 34%|███▍      | 22225/65536 [3:48:52<7:17:38,  1.65it/s]
 34%|███▍      | 22226/65536 [3:48:52<7:24:01,  1.63it/s]
 34%|███▍      | 22227/65536 [3:48:53<7:32:00,  1.60it/s]
 34%|███▍      | 22228/65536 [3:48:54<7:23:27,  1.63it/s]
 34%|███▍      | 22229/65536 [3:48:54<7:18:45,  1.65it/s]
 34%|███▍      | 22230/65536 [3:48:55<7:12:12,  1.67it/s]
 34%|███▍      | 22231/65536 [3:48:55<7:11:46,  1.67it/s]
 34%|███▍      | 22232/65536 [3:48:56<7:20:16,  1.64it/s]
 34%|███▍      | 22233/65536 [3:48:57<7:20:28,  1.64it/s]
 34%|███▍      | 22234/65536 [3:48:57<7:22:31,  1.63it/s]
 34%|███▍      | 22235/65536 [3:48:58<7:32:01,  1.60it/s]
 34%|███▍      | 22236/65536 [3:48:59<7:27:36,  1.61it/s]
 34%|███▍      | 22237/65536 [3:48:59<7:21:25,  1.63it/s]
 34%|███▍      | 22238/65536 [3:49:00<7:15:27,  1.66it/s]
 34%|███▍      | 22239/65536 [3:49:00<7:29:37,  1.60it/s]
 34%|███▍      | 22240/65536 [3:49:01<7:17:18,  1.65it/s]
                                                         
{'loss': 3.0706, 'learning_rate': 6.964001469327782e-07, 'epoch': 1372.84}
+
 34%|███▍      | 22240/65536 [3:49:01<7:17:18,  1.65it/s]
 34%|███▍      | 22241/65536 [3:49:02<7:30:39,  1.60it/s]
 34%|███▍      | 22242/65536 [3:49:02<7:23:03,  1.63it/s]
 34%|███▍      | 22243/65536 [3:49:03<7:31:23,  1.60it/s]
 34%|███▍      | 22244/65536 [3:49:04<7:45:16,  1.55it/s]
 34%|███▍      | 22245/65536 [3:49:04<7:37:54,  1.58it/s]
 34%|███▍      | 22246/65536 [3:49:05<7:31:37,  1.60it/s]
 34%|███▍      | 22247/65536 [3:49:05<7:20:48,  1.64it/s]
 34%|███▍      | 22248/65536 [3:49:06<7:19:52,  1.64it/s]
 34%|███▍      | 22249/65536 [3:49:07<7:16:18,  1.65it/s]
 34%|███▍      | 22250/65536 [3:49:07<7:21:32,  1.63it/s]
 34%|███▍      | 22251/65536 [3:49:08<7:20:55,  1.64it/s]
 34%|███▍      | 22252/65536 [3:49:08<7:17:20,  1.65it/s]
 34%|███▍      | 22253/65536 [3:49:09<7:09:03,  1.68it/s]
 34%|███▍      | 22254/65536 [3:49:10<7:23:54,  1.63it/s]
 34%|███▍      | 22255/65536 [3:49:10<7:19:23,  1.64it/s]
 34%|███▍      | 22256/65536 [3:49:11<7:14:43,  1.66it/s]
 34%|███▍      | 22257/65536 [3:49:11<7:28:51,  1.61it/s]
 34%|███▍      | 22258/65536 [3:49:12<7:12:54,  1.67it/s]
 34%|███▍      | 22259/65536 [3:49:13<7:22:33,  1.63it/s]
 34%|███▍      | 22260/65536 [3:49:13<7:12:35,  1.67it/s]
                                                         
{'loss': 3.0858, 'learning_rate': 6.961246479735521e-07, 'epoch': 1374.07}
+
 34%|███▍      | 22260/65536 [3:49:13<7:12:35,  1.67it/s]
 34%|███▍      | 22261/65536 [3:49:14<7:09:48,  1.68it/s]
 34%|███▍      | 22262/65536 [3:49:14<7:06:50,  1.69it/s]
 34%|███▍      | 22263/65536 [3:49:15<7:07:20,  1.69it/s]
 34%|███▍      | 22264/65536 [3:49:16<7:17:49,  1.65it/s]
 34%|███▍      | 22265/65536 [3:49:16<7:23:32,  1.63it/s]
 34%|███▍      | 22266/65536 [3:49:17<7:20:21,  1.64it/s]
 34%|███▍      | 22267/65536 [3:49:18<7:21:46,  1.63it/s]
 34%|███▍      | 22268/65536 [3:49:18<7:27:25,  1.61it/s]
 34%|███▍      | 22269/65536 [3:49:19<7:24:53,  1.62it/s]
 34%|███▍      | 22270/65536 [3:49:19<7:23:52,  1.62it/s]
 34%|███▍      | 22271/65536 [3:49:20<7:22:56,  1.63it/s]
 34%|███▍      | 22272/65536 [3:49:21<7:29:25,  1.60it/s]
 34%|███▍      | 22273/65536 [3:49:21<7:38:11,  1.57it/s]
 34%|███▍      | 22274/65536 [3:49:22<7:27:43,  1.61it/s]
 34%|███▍      | 22275/65536 [3:49:22<7:22:51,  1.63it/s]
 34%|███▍      | 22276/65536 [3:49:23<7:31:04,  1.60it/s]
 34%|███▍      | 22277/65536 [3:49:24<7:36:36,  1.58it/s]
 34%|███▍      | 22278/65536 [3:49:24<7:24:05,  1.62it/s]
 34%|███▍      | 22279/65536 [3:49:25<7:28:47,  1.61it/s]
 34%|███▍      | 22280/65536 [3:49:26<7:19:12,  1.64it/s]
                                                         
{'loss': 2.9814, 'learning_rate': 6.958491490143259e-07, 'epoch': 1375.31}
+
 34%|███▍      | 22280/65536 [3:49:26<7:19:12,  1.64it/s]
 34%|███▍      | 22281/65536 [3:49:26<7:08:58,  1.68it/s]
 34%|███▍      | 22282/65536 [3:49:27<7:09:55,  1.68it/s]
 34%|███▍      | 22283/65536 [3:49:27<7:15:12,  1.66it/s]
 34%|███▍      | 22284/65536 [3:49:28<7:24:37,  1.62it/s]
 34%|███▍      | 22285/65536 [3:49:29<7:25:07,  1.62it/s]
 34%|███▍      | 22286/65536 [3:49:29<7:19:06,  1.64it/s]
 34%|███▍      | 22287/65536 [3:49:30<7:32:11,  1.59it/s]
 34%|███▍      | 22288/65536 [3:49:30<7:18:22,  1.64it/s]
 34%|███▍      | 22289/65536 [3:49:31<7:11:24,  1.67it/s]
 34%|███▍      | 22290/65536 [3:49:32<7:21:39,  1.63it/s]
 34%|███▍      | 22291/65536 [3:49:32<7:09:13,  1.68it/s]
 34%|███▍      | 22292/65536 [3:49:33<7:42:02,  1.56it/s]
 34%|███▍      | 22293/65536 [3:49:34<7:41:47,  1.56it/s]
 34%|███▍      | 22294/65536 [3:49:34<7:42:10,  1.56it/s]
 34%|███▍      | 22295/65536 [3:49:35<7:33:11,  1.59it/s]
 34%|███▍      | 22296/65536 [3:49:36<7:42:38,  1.56it/s]
 34%|███▍      | 22297/65536 [3:49:36<7:33:36,  1.59it/s]
 34%|███▍      | 22298/65536 [3:49:37<7:19:35,  1.64it/s]
 34%|███▍      | 22299/65536 [3:49:37<7:17:08,  1.65it/s]
 34%|███▍      | 22300/65536 [3:49:38<7:08:12,  1.68it/s]
                                                         
{'loss': 3.0067, 'learning_rate': 6.955736500550998e-07, 'epoch': 1376.54}
+
 34%|███▍      | 22300/65536 [3:49:38<7:08:12,  1.68it/s]
 34%|███▍      | 22301/65536 [3:49:38<7:11:34,  1.67it/s]
 34%|███▍      | 22302/65536 [3:49:39<7:21:46,  1.63it/s]
 34%|███▍      | 22303/65536 [3:49:40<7:14:12,  1.66it/s]
 34%|███▍      | 22304/65536 [3:49:40<7:26:24,  1.61it/s]
 34%|███▍      | 22305/65536 [3:49:41<7:21:18,  1.63it/s]
 34%|███▍      | 22306/65536 [3:49:42<7:16:42,  1.65it/s]
 34%|███▍      | 22307/65536 [3:49:42<7:10:52,  1.67it/s]
 34%|███▍      | 22308/65536 [3:49:43<7:28:44,  1.61it/s]
 34%|███▍      | 22309/65536 [3:49:43<7:25:07,  1.62it/s]
 34%|███▍      | 22310/65536 [3:49:44<7:32:17,  1.59it/s]
 34%|███▍      | 22311/65536 [3:49:45<7:22:31,  1.63it/s]
 34%|███▍      | 22312/65536 [3:49:45<7:17:35,  1.65it/s]
 34%|███▍      | 22313/65536 [3:49:46<7:20:27,  1.64it/s]
 34%|███▍      | 22314/65536 [3:49:46<7:21:44,  1.63it/s]
 34%|███▍      | 22315/65536 [3:49:47<7:27:09,  1.61it/s]
 34%|███▍      | 22316/65536 [3:49:48<7:22:45,  1.63it/s]
 34%|███▍      | 22317/65536 [3:49:48<7:16:39,  1.65it/s]
 34%|███▍      | 22318/65536 [3:49:49<7:27:41,  1.61it/s]
 34%|███▍      | 22319/65536 [3:49:50<7:33:53,  1.59it/s]
 34%|███▍      | 22320/65536 [3:49:50<7:24:09,  1.62it/s]
                                                         
{'loss': 2.993, 'learning_rate': 6.952981510958736e-07, 'epoch': 1377.78}
+
 34%|███▍      | 22320/65536 [3:49:50<7:24:09,  1.62it/s]
 34%|███▍      | 22321/65536 [3:49:51<7:32:47,  1.59it/s]
 34%|███▍      | 22322/65536 [3:49:51<7:22:46,  1.63it/s]
 34%|███▍      | 22323/65536 [3:49:52<7:21:20,  1.63it/s]
 34%|███▍      | 22324/65536 [3:49:53<7:37:45,  1.57it/s]
 34%|███▍      | 22325/65536 [3:49:53<7:29:54,  1.60it/s]
 34%|███▍      | 22326/65536 [3:49:54<7:35:02,  1.58it/s]
 34%|███▍      | 22327/65536 [3:49:55<7:16:38,  1.65it/s]
 34%|███▍      | 22328/65536 [3:49:55<7:14:30,  1.66it/s]
 34%|███▍      | 22329/65536 [3:49:56<7:14:45,  1.66it/s]
 34%|███▍      | 22330/65536 [3:49:56<7:09:54,  1.68it/s]
 34%|███▍      | 22331/65536 [3:49:57<7:11:38,  1.67it/s]
 34%|███▍      | 22332/65536 [3:49:58<7:12:31,  1.66it/s]
 34%|███▍      | 22333/65536 [3:49:58<7:06:11,  1.69it/s]
 34%|███▍      | 22334/65536 [3:49:59<7:08:11,  1.68it/s]
 34%|███▍      | 22335/65536 [3:49:59<7:09:44,  1.68it/s]
 34%|███▍      | 22336/65536 [3:50:00<7:15:16,  1.65it/s]
 34%|███▍      | 22337/65536 [3:50:01<7:18:29,  1.64it/s]
 34%|███▍      | 22338/65536 [3:50:01<7:22:24,  1.63it/s]
 34%|███▍      | 22339/65536 [3:50:02<7:15:53,  1.65it/s]
 34%|███▍      | 22340/65536 [3:50:02<7:35:01,  1.58it/s]
                                                         
{'loss': 3.0182, 'learning_rate': 6.950226521366475e-07, 'epoch': 1379.01}
+
 34%|███▍      | 22340/65536 [3:50:02<7:35:01,  1.58it/s]
 34%|███▍      | 22341/65536 [3:50:03<7:34:52,  1.58it/s]
 34%|███▍      | 22342/65536 [3:50:04<7:28:50,  1.60it/s]
 34%|███▍      | 22343/65536 [3:50:04<7:26:13,  1.61it/s]
 34%|███▍      | 22344/65536 [3:50:05<7:28:05,  1.61it/s]
 34%|███▍      | 22345/65536 [3:50:05<7:18:42,  1.64it/s]
 34%|███▍      | 22346/65536 [3:50:06<7:21:02,  1.63it/s]
 34%|███▍      | 22347/65536 [3:50:07<7:20:36,  1.63it/s]
 34%|███▍      | 22348/65536 [3:50:07<7:09:53,  1.67it/s]
 34%|███▍      | 22349/65536 [3:50:08<7:12:23,  1.66it/s]
 34%|███▍      | 22350/65536 [3:50:08<7:15:36,  1.65it/s]
 34%|███▍      | 22351/65536 [3:50:09<7:35:11,  1.58it/s]
 34%|███▍      | 22352/65536 [3:50:10<7:26:55,  1.61it/s]
 34%|███▍      | 22353/65536 [3:50:10<7:17:25,  1.65it/s]
 34%|███▍      | 22354/65536 [3:50:11<7:17:38,  1.64it/s]
 34%|███▍      | 22355/65536 [3:50:12<7:14:32,  1.66it/s]
 34%|███▍      | 22356/65536 [3:50:12<7:24:02,  1.62it/s]
 34%|███▍      | 22357/65536 [3:50:13<7:41:08,  1.56it/s]
 34%|███▍      | 22358/65536 [3:50:14<7:31:51,  1.59it/s]
 34%|███▍      | 22359/65536 [3:50:14<7:33:34,  1.59it/s]
 34%|███▍      | 22360/65536 [3:50:15<7:31:55,  1.59it/s]
                                                         
{'loss': 2.9773, 'learning_rate': 6.947471531774214e-07, 'epoch': 1380.25}
+
 34%|███▍      | 22360/65536 [3:50:15<7:31:55,  1.59it/s]
 34%|███▍      | 22361/65536 [3:50:15<7:34:47,  1.58it/s]
 34%|███▍      | 22362/65536 [3:50:16<7:27:05,  1.61it/s]
 34%|███▍      | 22363/65536 [3:50:17<7:23:46,  1.62it/s]
 34%|███▍      | 22364/65536 [3:50:17<7:19:57,  1.64it/s]
 34%|███▍      | 22365/65536 [3:50:18<7:21:20,  1.63it/s]
 34%|███▍      | 22366/65536 [3:50:18<7:13:04,  1.66it/s]
 34%|███▍      | 22367/65536 [3:50:19<7:13:33,  1.66it/s]
 34%|███▍      | 22368/65536 [3:50:20<7:06:53,  1.69it/s]
 34%|███▍      | 22369/65536 [3:50:20<7:23:45,  1.62it/s]
 34%|███▍      | 22370/65536 [3:50:21<7:25:34,  1.61it/s]
 34%|███▍      | 22371/65536 [3:50:22<7:27:01,  1.61it/s]
 34%|███▍      | 22372/65536 [3:50:22<7:18:36,  1.64it/s]
 34%|███▍      | 22373/65536 [3:50:23<7:38:14,  1.57it/s]
 34%|███▍      | 22374/65536 [3:50:23<7:48:06,  1.54it/s]
 34%|███▍      | 22375/65536 [3:50:24<7:36:59,  1.57it/s]
 34%|███▍      | 22376/65536 [3:50:25<7:28:05,  1.61it/s]
 34%|███▍      | 22377/65536 [3:50:25<7:17:29,  1.64it/s]
 34%|███▍      | 22378/65536 [3:50:26<7:12:51,  1.66it/s]
 34%|███▍      | 22379/65536 [3:50:26<7:09:44,  1.67it/s]
 34%|███▍      | 22380/65536 [3:50:27<7:09:29,  1.67it/s]
                                                         
{'loss': 2.9697, 'learning_rate': 6.944716542181952e-07, 'epoch': 1381.48}
+
 34%|███▍      | 22380/65536 [3:50:27<7:09:29,  1.67it/s]
 34%|███▍      | 22381/65536 [3:50:28<7:20:42,  1.63it/s]
 34%|███▍      | 22382/65536 [3:50:28<7:09:13,  1.68it/s]
 34%|███▍      | 22383/65536 [3:50:29<7:08:13,  1.68it/s]
 34%|███▍      | 22384/65536 [3:50:29<7:12:19,  1.66it/s]
 34%|███▍      | 22385/65536 [3:50:30<7:11:19,  1.67it/s]
 34%|███▍      | 22386/65536 [3:50:31<7:22:16,  1.63it/s]
 34%|███▍      | 22387/65536 [3:50:31<7:27:00,  1.61it/s]
 34%|███▍      | 22388/65536 [3:50:32<7:20:36,  1.63it/s]
 34%|███▍      | 22389/65536 [3:50:33<7:25:05,  1.62it/s]
 34%|███▍      | 22390/65536 [3:50:33<7:15:18,  1.65it/s]
 34%|███▍      | 22391/65536 [3:50:34<7:12:06,  1.66it/s]
 34%|███▍      | 22392/65536 [3:50:34<7:07:53,  1.68it/s]
 34%|███▍      | 22393/65536 [3:50:35<7:09:40,  1.67it/s]
 34%|███▍      | 22394/65536 [3:50:35<7:12:12,  1.66it/s]
 34%|███▍      | 22395/65536 [3:50:36<7:15:58,  1.65it/s]
 34%|███▍      | 22396/65536 [3:50:37<7:10:14,  1.67it/s]
 34%|███▍      | 22397/65536 [3:50:37<7:32:50,  1.59it/s]
 34%|███▍      | 22398/65536 [3:50:38<7:22:59,  1.62it/s]
 34%|███▍      | 22399/65536 [3:50:39<7:18:15,  1.64it/s]
 34%|███▍      | 22400/65536 [3:50:39<7:19:14,  1.64it/s]
                                                         
{'loss': 3.0496, 'learning_rate': 6.941961552589691e-07, 'epoch': 1382.72}
+
 34%|███▍      | 22400/65536 [3:50:39<7:19:14,  1.64it/s]
 34%|███▍      | 22401/65536 [3:50:40<7:24:22,  1.62it/s]
 34%|███▍      | 22402/65536 [3:50:40<7:19:07,  1.64it/s]
 34%|███▍      | 22403/65536 [3:50:41<7:19:02,  1.64it/s]
 34%|███▍      | 22404/65536 [3:50:42<7:29:57,  1.60it/s]
 34%|███▍      | 22405/65536 [3:50:42<7:39:02,  1.57it/s]
 34%|███▍      | 22406/65536 [3:50:43<7:43:11,  1.55it/s]
 34%|███▍      | 22407/65536 [3:50:44<7:44:42,  1.55it/s]
 34%|███▍      | 22408/65536 [3:50:44<7:42:26,  1.55it/s]
 34%|███▍      | 22409/65536 [3:50:45<7:36:40,  1.57it/s]
 34%|███▍      | 22410/65536 [3:50:46<7:30:52,  1.59it/s]
 34%|███▍      | 22411/65536 [3:50:46<7:24:08,  1.62it/s]
 34%|███▍      | 22412/65536 [3:50:47<7:28:04,  1.60it/s]
 34%|███▍      | 22413/65536 [3:50:47<7:24:21,  1.62it/s]
 34%|███▍      | 22414/65536 [3:50:48<7:10:37,  1.67it/s]
 34%|███▍      | 22415/65536 [3:50:49<7:08:44,  1.68it/s]
 34%|███▍      | 22416/65536 [3:50:49<7:02:41,  1.70it/s]
 34%|███▍      | 22417/65536 [3:50:50<7:05:24,  1.69it/s]
 34%|███▍      | 22418/65536 [3:50:50<7:20:03,  1.63it/s]
 34%|███▍      | 22419/65536 [3:50:51<7:23:57,  1.62it/s]
 34%|███▍      | 22420/65536 [3:50:52<7:23:59,  1.62it/s]
                                                         
{'loss': 3.0031, 'learning_rate': 6.939206562997428e-07, 'epoch': 1383.95}
+
 34%|███▍      | 22420/65536 [3:50:52<7:23:59,  1.62it/s]
 34%|███▍      | 22421/65536 [3:50:52<7:38:02,  1.57it/s]
 34%|███▍      | 22422/65536 [3:50:53<7:26:16,  1.61it/s]
 34%|███▍      | 22423/65536 [3:50:53<7:10:50,  1.67it/s]
 34%|███▍      | 22424/65536 [3:50:54<7:09:51,  1.67it/s]
 34%|███▍      | 22425/65536 [3:50:55<7:11:01,  1.67it/s]
 34%|███▍      | 22426/65536 [3:50:55<7:21:15,  1.63it/s]
 34%|███▍      | 22427/65536 [3:50:56<7:38:18,  1.57it/s]
 34%|███▍      | 22428/65536 [3:50:57<7:41:08,  1.56it/s]
 34%|███▍      | 22429/65536 [3:50:57<7:34:01,  1.58it/s]
 34%|███▍      | 22430/65536 [3:50:58<7:26:46,  1.61it/s]
 34%|███▍      | 22431/65536 [3:50:58<7:21:26,  1.63it/s]
 34%|███▍      | 22432/65536 [3:50:59<7:10:10,  1.67it/s]
 34%|███▍      | 22433/65536 [3:51:00<7:32:46,  1.59it/s]
 34%|███▍      | 22434/65536 [3:51:00<7:28:34,  1.60it/s]
 34%|███▍      | 22435/65536 [3:51:01<7:14:39,  1.65it/s]
 34%|███▍      | 22436/65536 [3:51:01<7:11:46,  1.66it/s]
 34%|███▍      | 22437/65536 [3:51:02<7:12:56,  1.66it/s]
 34%|███▍      | 22438/65536 [3:51:03<7:27:09,  1.61it/s]
 34%|███▍      | 22439/65536 [3:51:03<7:49:31,  1.53it/s]
 34%|███▍      | 22440/65536 [3:51:04<7:37:00,  1.57it/s]
                                                         
{'loss': 3.0418, 'learning_rate': 6.936451573405167e-07, 'epoch': 1385.19}
+
 34%|███▍      | 22440/65536 [3:51:04<7:37:00,  1.57it/s]
 34%|███▍      | 22441/65536 [3:51:05<7:28:24,  1.60it/s]
 34%|███▍      | 22442/65536 [3:51:05<7:16:57,  1.64it/s]
 34%|███▍      | 22443/65536 [3:51:06<7:15:00,  1.65it/s]
 34%|███▍      | 22444/65536 [3:51:06<7:19:18,  1.63it/s]
 34%|███▍      | 22445/65536 [3:51:07<7:15:15,  1.65it/s]
 34%|███▍      | 22446/65536 [3:51:08<7:19:50,  1.63it/s]
 34%|███▍      | 22447/65536 [3:51:08<7:14:23,  1.65it/s]
 34%|███▍      | 22448/65536 [3:51:09<7:14:54,  1.65it/s]
 34%|███▍      | 22449/65536 [3:51:09<7:19:59,  1.63it/s]
 34%|███▍      | 22450/65536 [3:51:10<7:19:48,  1.63it/s]
 34%|███▍      | 22451/65536 [3:51:11<7:11:14,  1.67it/s]
 34%|███▍      | 22452/65536 [3:51:11<7:17:22,  1.64it/s]
 34%|███▍      | 22453/65536 [3:51:12<7:19:09,  1.64it/s]
 34%|███▍      | 22454/65536 [3:51:13<7:35:44,  1.58it/s]
 34%|███▍      | 22455/65536 [3:51:13<7:31:05,  1.59it/s]
 34%|███▍      | 22456/65536 [3:51:14<7:20:09,  1.63it/s]
 34%|███▍      | 22457/65536 [3:51:14<7:21:40,  1.63it/s]
 34%|███▍      | 22458/65536 [3:51:15<7:20:03,  1.63it/s]
 34%|███▍      | 22459/65536 [3:51:16<7:23:39,  1.62it/s]
 34%|███▍      | 22460/65536 [3:51:16<7:39:47,  1.56it/s]
                                                         
{'loss': 3.0117, 'learning_rate': 6.933696583812906e-07, 'epoch': 1386.42}
+
 34%|███▍      | 22460/65536 [3:51:16<7:39:47,  1.56it/s]
 34%|███▍      | 22461/65536 [3:51:17<7:27:09,  1.61it/s]
 34%|███▍      | 22462/65536 [3:51:17<7:19:14,  1.63it/s]
 34%|███▍      | 22463/65536 [3:51:18<7:12:00,  1.66it/s]
 34%|███▍      | 22464/65536 [3:51:19<7:37:47,  1.57it/s]
 34%|███▍      | 22465/65536 [3:51:19<7:32:42,  1.59it/s]
 34%|███▍      | 22466/65536 [3:51:20<7:24:35,  1.61it/s]
 34%|███▍      | 22467/65536 [3:51:21<7:16:09,  1.65it/s]
 34%|███▍      | 22468/65536 [3:51:21<7:12:08,  1.66it/s]
 34%|███▍      | 22469/65536 [3:51:22<7:09:39,  1.67it/s]
 34%|███▍      | 22470/65536 [3:51:22<7:28:27,  1.60it/s]
 34%|███▍      | 22471/65536 [3:51:23<7:33:27,  1.58it/s]
 34%|███▍      | 22472/65536 [3:51:24<7:28:46,  1.60it/s]
 34%|███▍      | 22473/65536 [3:51:24<7:25:36,  1.61it/s]
 34%|███▍      | 22474/65536 [3:51:25<7:18:29,  1.64it/s]
 34%|███▍      | 22475/65536 [3:51:26<7:17:33,  1.64it/s]
 34%|███▍      | 22476/65536 [3:51:26<7:08:41,  1.67it/s]
 34%|███▍      | 22477/65536 [3:51:27<7:11:28,  1.66it/s]
 34%|███▍      | 22478/65536 [3:51:27<7:12:22,  1.66it/s]
 34%|███▍      | 22479/65536 [3:51:28<7:14:07,  1.65it/s]
 34%|███▍      | 22480/65536 [3:51:29<7:16:47,  1.64it/s]
                                                         
{'loss': 2.9879, 'learning_rate': 6.930941594220643e-07, 'epoch': 1387.65}
+
 34%|███▍      | 22480/65536 [3:51:29<7:16:47,  1.64it/s]
 34%|███▍      | 22481/65536 [3:51:29<7:13:48,  1.65it/s]
 34%|███▍      | 22482/65536 [3:51:30<7:10:09,  1.67it/s]
 34%|███▍      | 22483/65536 [3:51:30<7:16:41,  1.64it/s]
 34%|███▍      | 22484/65536 [3:51:31<7:30:43,  1.59it/s]
 34%|███▍      | 22485/65536 [3:51:32<7:22:28,  1.62it/s]
 34%|███▍      | 22486/65536 [3:51:32<7:35:13,  1.58it/s]
 34%|███▍      | 22487/65536 [3:51:33<7:24:33,  1.61it/s]
 34%|███▍      | 22488/65536 [3:51:33<7:22:12,  1.62it/s]
 34%|███▍      | 22489/65536 [3:51:34<7:25:03,  1.61it/s]
 34%|███▍      | 22490/65536 [3:51:35<7:16:32,  1.64it/s]
 34%|███▍      | 22491/65536 [3:51:35<7:16:48,  1.64it/s]
 34%|███▍      | 22492/65536 [3:51:36<7:16:50,  1.64it/s]
 34%|███▍      | 22493/65536 [3:51:37<7:20:19,  1.63it/s]
 34%|███▍      | 22494/65536 [3:51:37<7:32:26,  1.59it/s]
 34%|███▍      | 22495/65536 [3:51:38<7:19:29,  1.63it/s]
 34%|███▍      | 22496/65536 [3:51:38<7:19:01,  1.63it/s]
 34%|███▍      | 22497/65536 [3:51:39<7:10:50,  1.66it/s]
 34%|███▍      | 22498/65536 [3:51:40<7:10:32,  1.67it/s]
 34%|███▍      | 22499/65536 [3:51:40<7:04:08,  1.69it/s]
 34%|███▍      | 22500/65536 [3:51:41<7:15:29,  1.65it/s]
                                                         
{'loss': 3.0156, 'learning_rate': 6.928186604628382e-07, 'epoch': 1388.89}
+
 34%|███▍      | 22500/65536 [3:51:41<7:15:29,  1.65it/s]
 34%|███▍      | 22501/65536 [3:51:41<7:23:06,  1.62it/s]
 34%|███▍      | 22502/65536 [3:51:42<7:36:53,  1.57it/s]
 34%|███▍      | 22503/65536 [3:51:43<7:32:46,  1.58it/s]
 34%|███▍      | 22504/65536 [3:51:43<7:27:38,  1.60it/s]
 34%|███▍      | 22505/65536 [3:51:44<7:17:13,  1.64it/s]
 34%|███▍      | 22506/65536 [3:51:44<7:13:56,  1.65it/s]
 34%|███▍      | 22507/65536 [3:51:45<7:25:44,  1.61it/s]
 34%|███▍      | 22508/65536 [3:51:46<7:21:10,  1.63it/s]
 34%|███▍      | 22509/65536 [3:51:46<7:28:54,  1.60it/s]
 34%|███▍      | 22510/65536 [3:51:47<7:21:56,  1.62it/s]
 34%|███▍      | 22511/65536 [3:51:48<7:40:00,  1.56it/s]
 34%|███▍      | 22512/65536 [3:51:48<7:39:02,  1.56it/s]
 34%|███▍      | 22513/65536 [3:51:49<7:32:03,  1.59it/s]
 34%|███▍      | 22514/65536 [3:51:50<7:19:52,  1.63it/s]
 34%|███▍      | 22515/65536 [3:51:50<7:20:13,  1.63it/s]
 34%|███▍      | 22516/65536 [3:51:51<7:17:55,  1.64it/s]
 34%|███▍      | 22517/65536 [3:51:51<7:18:24,  1.64it/s]
 34%|███▍      | 22518/65536 [3:51:52<7:07:25,  1.68it/s]
 34%|███▍      | 22519/65536 [3:51:53<7:19:08,  1.63it/s]
 34%|███▍      | 22520/65536 [3:51:53<7:23:42,  1.62it/s]
                                                         
{'loss': 2.9637, 'learning_rate': 6.92543161503612e-07, 'epoch': 1390.12}
+
 34%|███▍      | 22520/65536 [3:51:53<7:23:42,  1.62it/s]
 34%|███▍      | 22521/65536 [3:51:54<7:29:58,  1.59it/s]
 34%|███▍      | 22522/65536 [3:51:54<7:32:19,  1.58it/s]
 34%|███▍      | 22523/65536 [3:51:55<7:25:58,  1.61it/s]
 34%|███▍      | 22524/65536 [3:51:56<7:22:44,  1.62it/s]
 34%|███▍      | 22525/65536 [3:51:56<7:23:27,  1.62it/s]
 34%|███▍      | 22526/65536 [3:51:57<7:13:36,  1.65it/s]
 34%|███▍      | 22527/65536 [3:51:57<7:08:31,  1.67it/s]
 34%|███▍      | 22528/65536 [3:51:58<6:59:57,  1.71it/s]
 34%|███▍      | 22529/65536 [3:51:59<7:19:23,  1.63it/s]
 34%|███▍      | 22530/65536 [3:51:59<7:32:59,  1.58it/s]
 34%|███▍      | 22531/65536 [3:52:00<7:29:55,  1.59it/s]
 34%|███▍      | 22532/65536 [3:52:01<7:19:44,  1.63it/s]
 34%|███▍      | 22533/65536 [3:52:01<7:12:38,  1.66it/s]
 34%|███▍      | 22534/65536 [3:52:02<7:16:44,  1.64it/s]
 34%|███▍      | 22535/65536 [3:52:02<7:23:58,  1.61it/s]
 34%|███▍      | 22536/65536 [3:52:03<7:25:14,  1.61it/s]
 34%|███▍      | 22537/65536 [3:52:04<7:31:24,  1.59it/s]
 34%|███▍      | 22538/65536 [3:52:04<7:22:47,  1.62it/s]
 34%|███▍      | 22539/65536 [3:52:05<7:17:15,  1.64it/s]
 34%|███▍      | 22540/65536 [3:52:05<7:06:42,  1.68it/s]
                                                         
{'loss': 3.0093, 'learning_rate': 6.922676625443859e-07, 'epoch': 1391.36}
+
 34%|███▍      | 22540/65536 [3:52:05<7:06:42,  1.68it/s]
 34%|███▍      | 22541/65536 [3:52:06<7:20:27,  1.63it/s]
 34%|███▍      | 22542/65536 [3:52:07<7:09:11,  1.67it/s]
 34%|███▍      | 22543/65536 [3:52:07<7:08:20,  1.67it/s]
 34%|███▍      | 22544/65536 [3:52:08<7:07:04,  1.68it/s]
 34%|███▍      | 22545/65536 [3:52:08<7:14:57,  1.65it/s]
 34%|███▍      | 22546/65536 [3:52:09<7:15:43,  1.64it/s]
 34%|███▍      | 22547/65536 [3:52:10<7:06:22,  1.68it/s]
 34%|███▍      | 22548/65536 [3:52:10<7:19:43,  1.63it/s]
 34%|███▍      | 22549/65536 [3:52:11<7:27:25,  1.60it/s]
 34%|███▍      | 22550/65536 [3:52:12<7:20:17,  1.63it/s]
 34%|███▍      | 22551/65536 [3:52:12<7:33:20,  1.58it/s]
 34%|███▍      | 22552/65536 [3:52:13<7:40:27,  1.56it/s]
 34%|███▍      | 22553/65536 [3:52:13<7:30:47,  1.59it/s]
 34%|███▍      | 22554/65536 [3:52:14<7:18:46,  1.63it/s]
 34%|███▍      | 22555/65536 [3:52:15<7:01:20,  1.70it/s]
 34%|███▍      | 22556/65536 [3:52:15<7:14:36,  1.65it/s]
 34%|███▍      | 22557/65536 [3:52:16<7:18:34,  1.63it/s]
 34%|███▍      | 22558/65536 [3:52:16<7:22:00,  1.62it/s]
 34%|███▍      | 22559/65536 [3:52:17<7:22:11,  1.62it/s]
 34%|███▍      | 22560/65536 [3:52:18<7:13:07,  1.65it/s]
                                                         
{'loss': 3.0227, 'learning_rate': 6.919921635851597e-07, 'epoch': 1392.59}
+
 34%|███▍      | 22560/65536 [3:52:18<7:13:07,  1.65it/s]
 34%|███▍      | 22561/65536 [3:52:18<7:15:06,  1.65it/s]
 34%|███▍      | 22562/65536 [3:52:19<7:23:51,  1.61it/s]
 34%|███▍      | 22563/65536 [3:52:20<7:16:16,  1.64it/s]
 34%|███▍      | 22564/65536 [3:52:20<7:07:29,  1.68it/s]
 34%|███▍      | 22565/65536 [3:52:21<7:10:47,  1.66it/s]
 34%|███▍      | 22566/65536 [3:52:21<7:07:55,  1.67it/s]
 34%|███▍      | 22567/65536 [3:52:22<7:34:40,  1.58it/s]
 34%|███▍      | 22568/65536 [3:52:23<7:26:19,  1.60it/s]
 34%|███▍      | 22569/65536 [3:52:23<7:24:21,  1.61it/s]
 34%|███▍      | 22570/65536 [3:52:24<7:18:28,  1.63it/s]
 34%|███▍      | 22571/65536 [3:52:24<7:17:28,  1.64it/s]
 34%|███▍      | 22572/65536 [3:52:25<7:16:05,  1.64it/s]
 34%|███▍      | 22573/65536 [3:52:26<7:25:35,  1.61it/s]
 34%|███▍      | 22574/65536 [3:52:26<7:16:45,  1.64it/s]
 34%|███▍      | 22575/65536 [3:52:27<7:17:30,  1.64it/s]
 34%|███▍      | 22576/65536 [3:52:27<7:17:12,  1.64it/s]
 34%|███▍      | 22577/65536 [3:52:28<7:12:17,  1.66it/s]
 34%|███▍      | 22578/65536 [3:52:29<7:21:14,  1.62it/s]
 34%|███▍      | 22579/65536 [3:52:29<7:13:51,  1.65it/s]
 34%|███▍      | 22580/65536 [3:52:30<7:12:14,  1.66it/s]
                                                         
{'loss': 3.0063, 'learning_rate': 6.917166646259336e-07, 'epoch': 1393.83}
+
 34%|███▍      | 22580/65536 [3:52:30<7:12:14,  1.66it/s]
 34%|███▍      | 22581/65536 [3:52:31<7:12:27,  1.66it/s]
 34%|███▍      | 22582/65536 [3:52:31<7:14:36,  1.65it/s]
 34%|███▍      | 22583/65536 [3:52:32<7:24:34,  1.61it/s]
 34%|███▍      | 22584/65536 [3:52:32<7:23:39,  1.61it/s]
 34%|███▍      | 22585/65536 [3:52:33<7:18:11,  1.63it/s]
 34%|███▍      | 22586/65536 [3:52:34<7:16:16,  1.64it/s]
 34%|███▍      | 22587/65536 [3:52:34<7:05:30,  1.68it/s]
 34%|███▍      | 22588/65536 [3:52:35<7:03:04,  1.69it/s]
 34%|███▍      | 22589/65536 [3:52:35<7:08:16,  1.67it/s]
 34%|███▍      | 22590/65536 [3:52:36<7:07:53,  1.67it/s]
 34%|███▍      | 22591/65536 [3:52:37<7:21:16,  1.62it/s]
 34%|███▍      | 22592/65536 [3:52:37<7:22:06,  1.62it/s]
 34%|███▍      | 22593/65536 [3:52:38<7:14:34,  1.65it/s]
 34%|███▍      | 22594/65536 [3:52:38<7:12:17,  1.66it/s]
 34%|███▍      | 22595/65536 [3:52:39<7:25:07,  1.61it/s]
 34%|███▍      | 22596/65536 [3:52:40<7:22:05,  1.62it/s]
 34%|███▍      | 22597/65536 [3:52:40<7:21:43,  1.62it/s]
 34%|███▍      | 22598/65536 [3:52:41<7:27:25,  1.60it/s]
 34%|███▍      | 22599/65536 [3:52:41<7:12:20,  1.66it/s]
 34%|███▍      | 22600/65536 [3:52:42<7:25:14,  1.61it/s]
                                                         
{'loss': 3.0236, 'learning_rate': 6.914411656667076e-07, 'epoch': 1395.06}
+
 34%|███▍      | 22600/65536 [3:52:42<7:25:14,  1.61it/s]
 34%|███▍      | 22601/65536 [3:52:43<7:20:28,  1.62it/s]
 34%|███▍      | 22602/65536 [3:52:43<7:10:53,  1.66it/s]
 34%|███▍      | 22603/65536 [3:52:44<7:20:52,  1.62it/s]
 34%|███▍      | 22604/65536 [3:52:45<7:19:11,  1.63it/s]
 34%|███▍      | 22605/65536 [3:52:45<7:18:19,  1.63it/s]
 34%|███▍      | 22606/65536 [3:52:46<7:20:31,  1.62it/s]
 34%|███▍      | 22607/65536 [3:52:46<7:18:43,  1.63it/s]
 34%|███▍      | 22608/65536 [3:52:47<7:12:10,  1.66it/s]
 34%|███▍      | 22609/65536 [3:52:48<7:15:02,  1.64it/s]
 35%|███▍      | 22610/65536 [3:52:48<7:19:55,  1.63it/s]
 35%|███▍      | 22611/65536 [3:52:49<7:35:09,  1.57it/s]
 35%|███▍      | 22612/65536 [3:52:50<7:29:47,  1.59it/s]
 35%|███▍      | 22613/65536 [3:52:50<7:19:05,  1.63it/s]
 35%|███▍      | 22614/65536 [3:52:51<7:20:23,  1.62it/s]
 35%|███▍      | 22615/65536 [3:52:51<7:14:35,  1.65it/s]
 35%|███▍      | 22616/65536 [3:52:52<7:33:24,  1.58it/s]
 35%|███▍      | 22617/65536 [3:52:53<7:38:08,  1.56it/s]
 35%|███▍      | 22618/65536 [3:52:53<7:27:12,  1.60it/s]
 35%|███▍      | 22619/65536 [3:52:54<7:18:34,  1.63it/s]
 35%|███▍      | 22620/65536 [3:52:54<7:21:23,  1.62it/s]
                                                         
{'loss': 2.9828, 'learning_rate': 6.911656667074813e-07, 'epoch': 1396.3}
+
 35%|███▍      | 22620/65536 [3:52:54<7:21:23,  1.62it/s]
 35%|███▍      | 22621/65536 [3:52:55<7:29:33,  1.59it/s]
 35%|███▍      | 22622/65536 [3:52:56<7:24:32,  1.61it/s]
 35%|███▍      | 22623/65536 [3:52:56<7:36:40,  1.57it/s]
 35%|███▍      | 22624/65536 [3:52:57<7:37:26,  1.56it/s]
 35%|███▍      | 22625/65536 [3:52:58<7:21:45,  1.62it/s]
 35%|███▍      | 22626/65536 [3:52:58<7:13:45,  1.65it/s]
 35%|███▍      | 22627/65536 [3:52:59<7:15:11,  1.64it/s]
 35%|███▍      | 22628/65536 [3:52:59<7:19:43,  1.63it/s]
 35%|███▍      | 22629/65536 [3:53:00<7:21:07,  1.62it/s]
 35%|███▍      | 22630/65536 [3:53:01<7:17:23,  1.63it/s]
 35%|███▍      | 22631/65536 [3:53:01<7:15:53,  1.64it/s]
 35%|███▍      | 22632/65536 [3:53:02<7:28:54,  1.59it/s]
 35%|███▍      | 22633/65536 [3:53:03<7:28:53,  1.59it/s]
 35%|███▍      | 22634/65536 [3:53:03<7:19:00,  1.63it/s]
 35%|███▍      | 22635/65536 [3:53:04<7:18:41,  1.63it/s]
 35%|███▍      | 22636/65536 [3:53:04<7:24:56,  1.61it/s]
 35%|███▍      | 22637/65536 [3:53:05<7:20:25,  1.62it/s]
 35%|███▍      | 22638/65536 [3:53:06<7:14:55,  1.64it/s]
 35%|███▍      | 22639/65536 [3:53:06<7:24:34,  1.61it/s]
 35%|███▍      | 22640/65536 [3:53:07<7:17:50,  1.63it/s]
                                                         
{'loss': 3.0113, 'learning_rate': 6.908901677482552e-07, 'epoch': 1397.53}
+
 35%|███▍      | 22640/65536 [3:53:07<7:17:50,  1.63it/s]
 35%|███▍      | 22641/65536 [3:53:07<7:10:31,  1.66it/s]
 35%|███▍      | 22642/65536 [3:53:08<7:08:19,  1.67it/s]
 35%|███▍      | 22643/65536 [3:53:09<7:06:09,  1.68it/s]
 35%|███▍      | 22644/65536 [3:53:09<7:06:12,  1.68it/s]
 35%|███▍      | 22645/65536 [3:53:10<7:17:57,  1.63it/s]
 35%|███▍      | 22646/65536 [3:53:10<7:18:55,  1.63it/s]
 35%|███▍      | 22647/65536 [3:53:11<7:26:45,  1.60it/s]
 35%|███▍      | 22648/65536 [3:53:12<7:31:10,  1.58it/s]
 35%|███▍      | 22649/65536 [3:53:12<7:34:28,  1.57it/s]
 35%|███▍      | 22650/65536 [3:53:13<7:25:17,  1.61it/s]
 35%|███▍      | 22651/65536 [3:53:14<7:24:45,  1.61it/s]
 35%|███▍      | 22652/65536 [3:53:14<7:18:33,  1.63it/s]
 35%|███▍      | 22653/65536 [3:53:15<7:28:04,  1.60it/s]
 35%|███▍      | 22654/65536 [3:53:15<7:14:18,  1.65it/s]
 35%|███▍      | 22655/65536 [3:53:16<7:15:15,  1.64it/s]
 35%|███▍      | 22656/65536 [3:53:17<7:18:34,  1.63it/s]
 35%|███▍      | 22657/65536 [3:53:17<7:12:00,  1.65it/s]
 35%|███▍      | 22658/65536 [3:53:18<7:29:30,  1.59it/s]
 35%|███▍      | 22659/65536 [3:53:19<7:28:43,  1.59it/s]
 35%|███▍      | 22660/65536 [3:53:19<7:25:13,  1.61it/s]
                                                         
{'loss': 3.0021, 'learning_rate': 6.90614668789029e-07, 'epoch': 1398.77}
+
 35%|███▍      | 22660/65536 [3:53:19<7:25:13,  1.61it/s]
 35%|███▍      | 22661/65536 [3:53:20<7:18:18,  1.63it/s]
 35%|███▍      | 22662/65536 [3:53:20<7:05:08,  1.68it/s]
 35%|███▍      | 22663/65536 [3:53:21<7:11:50,  1.65it/s]
 35%|███▍      | 22664/65536 [3:53:22<7:22:37,  1.61it/s]
 35%|███▍      | 22665/65536 [3:53:22<7:16:41,  1.64it/s]
 35%|███▍      | 22666/65536 [3:53:23<7:10:07,  1.66it/s]
 35%|███▍      | 22667/65536 [3:53:23<7:27:14,  1.60it/s]
 35%|███▍      | 22668/65536 [3:53:24<7:20:20,  1.62it/s]
 35%|███▍      | 22669/65536 [3:53:25<7:25:19,  1.60it/s]
 35%|███▍      | 22670/65536 [3:53:25<7:23:39,  1.61it/s]
 35%|███▍      | 22671/65536 [3:53:26<7:21:17,  1.62it/s]
 35%|███▍      | 22672/65536 [3:53:27<7:20:01,  1.62it/s]
 35%|███▍      | 22673/65536 [3:53:27<7:23:31,  1.61it/s]
 35%|███▍      | 22674/65536 [3:53:28<7:47:44,  1.53it/s]
 35%|███▍      | 22675/65536 [3:53:29<7:55:13,  1.50it/s]
 35%|███▍      | 22676/65536 [3:53:29<7:42:49,  1.54it/s]
 35%|███▍      | 22677/65536 [3:53:30<7:34:03,  1.57it/s]
 35%|███▍      | 22678/65536 [3:53:30<7:23:23,  1.61it/s]
 35%|███▍      | 22679/65536 [3:53:31<7:17:01,  1.63it/s]
 35%|███▍      | 22680/65536 [3:53:32<7:22:17,  1.61it/s]
                                                         
{'loss': 2.9998, 'learning_rate': 6.903391698298028e-07, 'epoch': 1400.0}
+
 35%|███▍      | 22680/65536 [3:53:32<7:22:17,  1.61it/s]
 35%|███▍      | 22681/65536 [3:53:32<7:29:14,  1.59it/s]
 35%|███▍      | 22682/65536 [3:53:33<7:24:53,  1.61it/s]
 35%|███▍      | 22683/65536 [3:53:34<7:22:59,  1.61it/s]
 35%|███▍      | 22684/65536 [3:53:34<7:26:20,  1.60it/s]
 35%|███▍      | 22685/65536 [3:53:35<7:21:06,  1.62it/s]
 35%|███▍      | 22686/65536 [3:53:35<7:28:17,  1.59it/s]
 35%|███▍      | 22687/65536 [3:53:36<7:21:01,  1.62it/s]
 35%|███▍      | 22688/65536 [3:53:37<7:13:38,  1.65it/s]
 35%|███▍      | 22689/65536 [3:53:37<7:12:30,  1.65it/s]
 35%|███▍      | 22690/65536 [3:53:38<7:16:33,  1.64it/s]
 35%|███▍      | 22691/65536 [3:53:38<7:16:32,  1.64it/s]
 35%|███▍      | 22692/65536 [3:53:39<7:02:42,  1.69it/s]
 35%|███▍      | 22693/65536 [3:53:40<6:59:35,  1.70it/s]
 35%|███▍      | 22694/65536 [3:53:40<7:13:45,  1.65it/s]
 35%|███▍      | 22695/65536 [3:53:41<7:14:05,  1.64it/s]
 35%|███▍      | 22696/65536 [3:53:41<7:15:46,  1.64it/s]
 35%|███▍      | 22697/65536 [3:53:42<7:38:39,  1.56it/s]
 35%|███▍      | 22698/65536 [3:53:43<7:23:53,  1.61it/s]
 35%|███▍      | 22699/65536 [3:53:43<7:21:37,  1.62it/s]
 35%|███▍      | 22700/65536 [3:53:44<7:24:55,  1.60it/s]
                                                         
{'loss': 3.0051, 'learning_rate': 6.900636708705766e-07, 'epoch': 1401.23}
+
 35%|███▍      | 22700/65536 [3:53:44<7:24:55,  1.60it/s]
 35%|███▍      | 22701/65536 [3:53:45<7:21:58,  1.62it/s]
 35%|███▍      | 22702/65536 [3:53:45<7:13:57,  1.65it/s]
 35%|███▍      | 22703/65536 [3:53:46<7:09:51,  1.66it/s]
 35%|███▍      | 22704/65536 [3:53:46<7:07:55,  1.67it/s]
 35%|███▍      | 22705/65536 [3:53:47<7:07:34,  1.67it/s]
 35%|███▍      | 22706/65536 [3:53:48<7:11:31,  1.65it/s]
 35%|███▍      | 22707/65536 [3:53:48<7:15:36,  1.64it/s]
 35%|███▍      | 22708/65536 [3:53:49<7:11:28,  1.65it/s]
 35%|███▍      | 22709/65536 [3:53:49<7:08:16,  1.67it/s]
 35%|███▍      | 22710/65536 [3:53:50<6:59:41,  1.70it/s]
 35%|███▍      | 22711/65536 [3:53:51<7:10:29,  1.66it/s]
 35%|███▍      | 22712/65536 [3:53:51<7:27:12,  1.60it/s]
 35%|███▍      | 22713/65536 [3:53:52<7:35:02,  1.57it/s]
 35%|███▍      | 22714/65536 [3:53:52<7:22:19,  1.61it/s]
 35%|███▍      | 22715/65536 [3:53:53<7:14:51,  1.64it/s]
 35%|███▍      | 22716/65536 [3:53:54<7:34:43,  1.57it/s]
 35%|███▍      | 22717/65536 [3:53:54<7:17:30,  1.63it/s]
 35%|███▍      | 22718/65536 [3:53:55<7:07:11,  1.67it/s]
 35%|███▍      | 22719/65536 [3:53:55<7:14:06,  1.64it/s]
 35%|███▍      | 22720/65536 [3:53:56<7:12:25,  1.65it/s]
                                                         
{'loss': 3.0361, 'learning_rate': 6.897881719113505e-07, 'epoch': 1402.47}
+
 35%|███▍      | 22720/65536 [3:53:56<7:12:25,  1.65it/s]
 35%|███▍      | 22721/65536 [3:53:57<7:12:48,  1.65it/s]
 35%|███▍      | 22722/65536 [3:53:57<7:11:47,  1.65it/s]
 35%|███▍      | 22723/65536 [3:53:58<7:28:15,  1.59it/s]
 35%|███▍      | 22724/65536 [3:53:59<7:26:50,  1.60it/s]
 35%|███▍      | 22725/65536 [3:53:59<7:30:18,  1.58it/s]
 35%|███▍      | 22726/65536 [3:54:00<7:29:56,  1.59it/s]
 35%|███▍      | 22727/65536 [3:54:01<7:32:51,  1.58it/s]
 35%|███▍      | 22728/65536 [3:54:01<7:24:53,  1.60it/s]
 35%|███▍      | 22729/65536 [3:54:02<7:30:34,  1.58it/s]
 35%|███▍      | 22730/65536 [3:54:02<7:31:40,  1.58it/s]
 35%|███▍      | 22731/65536 [3:54:03<7:18:28,  1.63it/s]
 35%|███▍      | 22732/65536 [3:54:04<7:19:11,  1.62it/s]
 35%|███▍      | 22733/65536 [3:54:04<7:23:29,  1.61it/s]
 35%|███▍      | 22734/65536 [3:54:05<7:09:15,  1.66it/s]
 35%|███▍      | 22735/65536 [3:54:05<7:14:08,  1.64it/s]
 35%|███▍      | 22736/65536 [3:54:06<7:17:30,  1.63it/s]
 35%|███▍      | 22737/65536 [3:54:07<7:17:03,  1.63it/s]
 35%|███▍      | 22738/65536 [3:54:07<7:13:13,  1.65it/s]
 35%|███▍      | 22739/65536 [3:54:08<7:04:14,  1.68it/s]
 35%|███▍      | 22740/65536 [3:54:08<7:11:39,  1.65it/s]
                                                         
{'loss': 3.0125, 'learning_rate': 6.895126729521244e-07, 'epoch': 1403.7}
+
 35%|███▍      | 22740/65536 [3:54:08<7:11:39,  1.65it/s]
 35%|███▍      | 22741/65536 [3:54:09<7:18:52,  1.63it/s]
 35%|███▍      | 22742/65536 [3:54:10<7:17:52,  1.63it/s]
 35%|███▍      | 22743/65536 [3:54:10<7:29:24,  1.59it/s]
 35%|███▍      | 22744/65536 [3:54:11<7:25:41,  1.60it/s]
 35%|███▍      | 22745/65536 [3:54:12<7:35:56,  1.56it/s]
 35%|███▍      | 22746/65536 [3:54:12<7:31:48,  1.58it/s]
 35%|███▍      | 22747/65536 [3:54:13<7:17:10,  1.63it/s]
 35%|███▍      | 22748/65536 [3:54:13<7:21:01,  1.62it/s]
 35%|███▍      | 22749/65536 [3:54:14<7:21:48,  1.61it/s]
 35%|███▍      | 22750/65536 [3:54:15<7:13:48,  1.64it/s]
 35%|███▍      | 22751/65536 [3:54:15<7:27:45,  1.59it/s]
 35%|███▍      | 22752/65536 [3:54:16<7:24:03,  1.61it/s]
 35%|███▍      | 22753/65536 [3:54:17<7:20:03,  1.62it/s]
 35%|███▍      | 22754/65536 [3:54:17<7:11:51,  1.65it/s]
 35%|███▍      | 22755/65536 [3:54:18<7:11:20,  1.65it/s]
 35%|███▍      | 22756/65536 [3:54:18<7:14:24,  1.64it/s]
 35%|███▍      | 22757/65536 [3:54:19<7:12:44,  1.65it/s]
 35%|███▍      | 22758/65536 [3:54:20<7:17:55,  1.63it/s]
 35%|███▍      | 22759/65536 [3:54:20<7:08:30,  1.66it/s]
 35%|███▍      | 22760/65536 [3:54:21<7:16:56,  1.63it/s]
                                                         
{'loss': 2.9742, 'learning_rate': 6.892371739928981e-07, 'epoch': 1404.94}
+
 35%|███▍      | 22760/65536 [3:54:21<7:16:56,  1.63it/s]
 35%|███▍      | 22761/65536 [3:54:21<7:12:00,  1.65it/s]
 35%|███▍      | 22762/65536 [3:54:22<7:24:08,  1.61it/s]
 35%|███▍      | 22763/65536 [3:54:23<7:18:30,  1.63it/s]
 35%|███▍      | 22764/65536 [3:54:23<7:22:51,  1.61it/s]
 35%|███▍      | 22765/65536 [3:54:24<7:15:07,  1.64it/s]
 35%|███▍      | 22766/65536 [3:54:24<7:13:39,  1.64it/s]
 35%|███▍      | 22767/65536 [3:54:25<7:22:32,  1.61it/s]
 35%|███▍      | 22768/65536 [3:54:26<7:26:40,  1.60it/s]
 35%|███▍      | 22769/65536 [3:54:26<7:18:29,  1.63it/s]
 35%|███▍      | 22770/65536 [3:54:27<7:18:06,  1.63it/s]
 35%|███▍      | 22771/65536 [3:54:28<7:12:47,  1.65it/s]
 35%|███▍      | 22772/65536 [3:54:28<7:25:40,  1.60it/s]
 35%|███▍      | 22773/65536 [3:54:29<7:14:17,  1.64it/s]
 35%|███▍      | 22774/65536 [3:54:29<7:21:44,  1.61it/s]
 35%|███▍      | 22775/65536 [3:54:30<7:23:30,  1.61it/s]
 35%|███▍      | 22776/65536 [3:54:31<7:32:03,  1.58it/s]
 35%|███▍      | 22777/65536 [3:54:31<7:14:10,  1.64it/s]
 35%|███▍      | 22778/65536 [3:54:32<7:29:20,  1.59it/s]
 35%|███▍      | 22779/65536 [3:54:33<7:35:05,  1.57it/s]
 35%|███▍      | 22780/65536 [3:54:33<7:26:36,  1.60it/s]
                                                         
{'loss': 2.9415, 'learning_rate': 6.88961675033672e-07, 'epoch': 1406.17}
+
 35%|███▍      | 22780/65536 [3:54:33<7:26:36,  1.60it/s]
 35%|███▍      | 22781/65536 [3:54:34<7:23:41,  1.61it/s]
 35%|███▍      | 22782/65536 [3:54:34<7:31:07,  1.58it/s]
 35%|███▍      | 22783/65536 [3:54:35<7:30:31,  1.58it/s]
 35%|███▍      | 22784/65536 [3:54:36<7:20:57,  1.62it/s]
 35%|███▍      | 22785/65536 [3:54:36<7:30:03,  1.58it/s]
 35%|███▍      | 22786/65536 [3:54:37<7:29:17,  1.59it/s]
 35%|███▍      | 22787/65536 [3:54:38<7:16:48,  1.63it/s]
 35%|███▍      | 22788/65536 [3:54:38<7:06:17,  1.67it/s]
 35%|███▍      | 22789/65536 [3:54:39<7:02:01,  1.69it/s]
 35%|███▍      | 22790/65536 [3:54:39<7:02:47,  1.69it/s]
 35%|███▍      | 22791/65536 [3:54:40<7:05:03,  1.68it/s]
 35%|███▍      | 22792/65536 [3:54:41<7:11:49,  1.65it/s]
 35%|███▍      | 22793/65536 [3:54:41<7:13:11,  1.64it/s]
 35%|███▍      | 22794/65536 [3:54:42<7:25:50,  1.60it/s]
 35%|███▍      | 22795/65536 [3:54:42<7:16:24,  1.63it/s]
 35%|███▍      | 22796/65536 [3:54:43<7:09:24,  1.66it/s]
 35%|███▍      | 22797/65536 [3:54:44<7:10:43,  1.65it/s]
 35%|███▍      | 22798/65536 [3:54:44<7:07:14,  1.67it/s]
 35%|███▍      | 22799/65536 [3:54:45<7:15:39,  1.63it/s]
 35%|███▍      | 22800/65536 [3:54:45<7:27:49,  1.59it/s]
                                                         
{'loss': 2.9931, 'learning_rate': 6.886861760744458e-07, 'epoch': 1407.41}
+
 35%|███▍      | 22800/65536 [3:54:45<7:27:49,  1.59it/s]
 35%|███▍      | 22801/65536 [3:54:46<7:23:46,  1.60it/s]
 35%|███▍      | 22802/65536 [3:54:47<7:18:30,  1.62it/s]
 35%|███▍      | 22803/65536 [3:54:47<7:23:37,  1.61it/s]
 35%|███▍      | 22804/65536 [3:54:48<7:26:53,  1.59it/s]
 35%|███▍      | 22805/65536 [3:54:49<7:22:58,  1.61it/s]
 35%|███▍      | 22806/65536 [3:54:49<7:24:52,  1.60it/s]
 35%|███▍      | 22807/65536 [3:54:50<7:28:42,  1.59it/s]
 35%|███▍      | 22808/65536 [3:54:50<7:23:09,  1.61it/s]
 35%|███▍      | 22809/65536 [3:54:51<7:20:35,  1.62it/s]
 35%|███▍      | 22810/65536 [3:54:52<7:35:18,  1.56it/s]
 35%|███▍      | 22811/65536 [3:54:52<7:27:20,  1.59it/s]
 35%|███▍      | 22812/65536 [3:54:53<7:24:23,  1.60it/s]
 35%|███▍      | 22813/65536 [3:54:54<7:17:57,  1.63it/s]
 35%|███▍      | 22814/65536 [3:54:54<7:27:44,  1.59it/s]
 35%|███▍      | 22815/65536 [3:54:55<7:19:38,  1.62it/s]
 35%|███▍      | 22816/65536 [3:54:55<7:12:39,  1.65it/s]
 35%|███▍      | 22817/65536 [3:54:56<7:13:40,  1.64it/s]
 35%|███▍      | 22818/65536 [3:54:57<7:17:50,  1.63it/s]
 35%|███▍      | 22819/65536 [3:54:57<7:05:05,  1.67it/s]
 35%|███▍      | 22820/65536 [3:54:58<7:06:54,  1.67it/s]
                                                         
{'loss': 2.9967, 'learning_rate': 6.884106771152197e-07, 'epoch': 1408.64}
+
 35%|███▍      | 22820/65536 [3:54:58<7:06:54,  1.67it/s]
 35%|███▍      | 22821/65536 [3:54:58<7:09:15,  1.66it/s]
 35%|███▍      | 22822/65536 [3:54:59<7:03:54,  1.68it/s]
 35%|███▍      | 22823/65536 [3:55:00<7:03:33,  1.68it/s]
 35%|███▍      | 22824/65536 [3:55:00<7:15:03,  1.64it/s]
 35%|███▍      | 22825/65536 [3:55:01<7:24:45,  1.60it/s]
 35%|███▍      | 22826/65536 [3:55:02<7:49:12,  1.52it/s]
 35%|███▍      | 22827/65536 [3:55:02<7:36:20,  1.56it/s]
 35%|███▍      | 22828/65536 [3:55:03<7:27:20,  1.59it/s]
 35%|███▍      | 22829/65536 [3:55:03<7:16:09,  1.63it/s]
 35%|███▍      | 22830/65536 [3:55:04<7:14:53,  1.64it/s]
 35%|███▍      | 22831/65536 [3:55:05<7:19:40,  1.62it/s]
 35%|███▍      | 22832/65536 [3:55:05<7:29:15,  1.58it/s]
 35%|███▍      | 22833/65536 [3:55:06<7:27:52,  1.59it/s]
 35%|███▍      | 22834/65536 [3:55:07<7:34:02,  1.57it/s]
 35%|███▍      | 22835/65536 [3:55:07<7:27:53,  1.59it/s]
 35%|███▍      | 22836/65536 [3:55:08<7:19:34,  1.62it/s]
 35%|███▍      | 22837/65536 [3:55:08<7:18:11,  1.62it/s]
 35%|███▍      | 22838/65536 [3:55:09<7:17:54,  1.63it/s]
 35%|███▍      | 22839/65536 [3:55:10<7:14:43,  1.64it/s]
 35%|███▍      | 22840/65536 [3:55:10<7:22:49,  1.61it/s]
                                                         
{'loss': 2.971, 'learning_rate': 6.881351781559936e-07, 'epoch': 1409.88}
+
 35%|███▍      | 22840/65536 [3:55:10<7:22:49,  1.61it/s]
 35%|███▍      | 22841/65536 [3:55:11<7:08:48,  1.66it/s]
 35%|███▍      | 22842/65536 [3:55:11<7:05:26,  1.67it/s]
 35%|███▍      | 22843/65536 [3:55:12<7:16:46,  1.63it/s]
 35%|███▍      | 22844/65536 [3:55:13<7:18:24,  1.62it/s]
 35%|███▍      | 22845/65536 [3:55:13<7:12:10,  1.65it/s]
 35%|███▍      | 22846/65536 [3:55:14<7:02:56,  1.68it/s]
 35%|███▍      | 22847/65536 [3:55:14<7:15:20,  1.63it/s]
 35%|███▍      | 22848/65536 [3:55:15<7:12:20,  1.65it/s]
 35%|███▍      | 22849/65536 [3:55:16<7:14:26,  1.64it/s]
 35%|███▍      | 22850/65536 [3:55:16<7:09:48,  1.66it/s]
 35%|███▍      | 22851/65536 [3:55:17<7:13:28,  1.64it/s]
 35%|███▍      | 22852/65536 [3:55:17<7:06:51,  1.67it/s]
 35%|███▍      | 22853/65536 [3:55:18<7:20:07,  1.62it/s]
 35%|███▍      | 22854/65536 [3:55:19<7:16:34,  1.63it/s]
 35%|███▍      | 22855/65536 [3:55:19<7:12:55,  1.64it/s]
 35%|███▍      | 22856/65536 [3:55:20<7:19:38,  1.62it/s]
 35%|███▍      | 22857/65536 [3:55:21<7:12:02,  1.65it/s]
 35%|███▍      | 22858/65536 [3:55:21<7:21:21,  1.61it/s]
 35%|███▍      | 22859/65536 [3:55:22<7:44:17,  1.53it/s]
 35%|███▍      | 22860/65536 [3:55:23<7:31:08,  1.58it/s]
                                                         
{'loss': 2.9922, 'learning_rate': 6.878596791967674e-07, 'epoch': 1411.11}
+
 35%|███▍      | 22860/65536 [3:55:23<7:31:08,  1.58it/s]
 35%|███▍      | 22861/65536 [3:55:23<7:28:11,  1.59it/s]
 35%|███▍      | 22862/65536 [3:55:24<7:28:34,  1.59it/s]
 35%|███▍      | 22863/65536 [3:55:24<7:20:40,  1.61it/s]
 35%|███▍      | 22864/65536 [3:55:25<7:14:31,  1.64it/s]
 35%|███▍      | 22865/65536 [3:55:26<7:12:12,  1.65it/s]
 35%|███▍      | 22866/65536 [3:55:26<7:13:05,  1.64it/s]
 35%|███▍      | 22867/65536 [3:55:27<7:08:12,  1.66it/s]
 35%|███▍      | 22868/65536 [3:55:27<7:10:57,  1.65it/s]
 35%|███▍      | 22869/65536 [3:55:28<7:04:48,  1.67it/s]
 35%|███▍      | 22870/65536 [3:55:29<7:09:17,  1.66it/s]
 35%|███▍      | 22871/65536 [3:55:29<6:59:26,  1.70it/s]
 35%|███▍      | 22872/65536 [3:55:30<7:02:31,  1.68it/s]
 35%|███▍      | 22873/65536 [3:55:30<7:02:38,  1.68it/s]
 35%|███▍      | 22874/65536 [3:55:31<7:02:02,  1.68it/s]
 35%|███▍      | 22875/65536 [3:55:32<7:24:21,  1.60it/s]
 35%|███▍      | 22876/65536 [3:55:32<7:26:18,  1.59it/s]
 35%|███▍      | 22877/65536 [3:55:33<7:21:32,  1.61it/s]
 35%|███▍      | 22878/65536 [3:55:33<7:18:34,  1.62it/s]
 35%|███▍      | 22879/65536 [3:55:34<7:05:01,  1.67it/s]
 35%|███▍      | 22880/65536 [3:55:35<7:18:37,  1.62it/s]
                                                         
{'loss': 3.0975, 'learning_rate': 6.875841802375413e-07, 'epoch': 1412.35}
+
 35%|███▍      | 22880/65536 [3:55:35<7:18:37,  1.62it/s]
 35%|███▍      | 22881/65536 [3:55:35<7:05:29,  1.67it/s]
 35%|███▍      | 22882/65536 [3:55:36<7:17:27,  1.63it/s]
 35%|███▍      | 22883/65536 [3:55:37<7:12:55,  1.64it/s]
 35%|███▍      | 22884/65536 [3:55:37<7:06:09,  1.67it/s]
 35%|███▍      | 22885/65536 [3:55:38<7:01:16,  1.69it/s]
 35%|███▍      | 22886/65536 [3:55:38<7:06:41,  1.67it/s]
 35%|███▍      | 22887/65536 [3:55:39<7:13:23,  1.64it/s]
 35%|███▍      | 22888/65536 [3:55:40<7:15:06,  1.63it/s]
 35%|███▍      | 22889/65536 [3:55:40<7:22:13,  1.61it/s]
 35%|███▍      | 22890/65536 [3:55:41<7:18:48,  1.62it/s]
 35%|███▍      | 22891/65536 [3:55:41<7:39:18,  1.55it/s]
 35%|███▍      | 22892/65536 [3:55:42<7:38:15,  1.55it/s]
 35%|███▍      | 22893/65536 [3:55:43<7:44:53,  1.53it/s]
 35%|███▍      | 22894/65536 [3:55:43<7:33:41,  1.57it/s]
 35%|███▍      | 22895/65536 [3:55:44<7:20:49,  1.61it/s]
 35%|███▍      | 22896/65536 [3:55:45<7:10:13,  1.65it/s]
 35%|███▍      | 22897/65536 [3:55:45<7:16:10,  1.63it/s]
 35%|███▍      | 22898/65536 [3:55:46<7:23:51,  1.60it/s]
 35%|███▍      | 22899/65536 [3:55:46<7:14:04,  1.64it/s]
 35%|███▍      | 22900/65536 [3:55:47<7:09:50,  1.65it/s]
                                                         
{'loss': 2.9825, 'learning_rate': 6.873086812783151e-07, 'epoch': 1413.58}
+
 35%|███▍      | 22900/65536 [3:55:47<7:09:50,  1.65it/s]
 35%|███▍      | 22901/65536 [3:55:48<7:01:58,  1.68it/s]
 35%|███▍      | 22902/65536 [3:55:48<6:59:38,  1.69it/s]
 35%|███▍      | 22903/65536 [3:55:49<7:16:42,  1.63it/s]
 35%|███▍      | 22904/65536 [3:55:49<7:16:39,  1.63it/s]
 35%|███▍      | 22905/65536 [3:55:50<7:16:22,  1.63it/s]
 35%|███▍      | 22906/65536 [3:55:51<7:07:38,  1.66it/s]
 35%|███▍      | 22907/65536 [3:55:51<7:29:20,  1.58it/s]
 35%|███▍      | 22908/65536 [3:55:52<7:26:55,  1.59it/s]
 35%|███▍      | 22909/65536 [3:55:53<7:24:27,  1.60it/s]
 35%|███▍      | 22910/65536 [3:55:53<7:13:58,  1.64it/s]
 35%|███▍      | 22911/65536 [3:55:54<7:10:17,  1.65it/s]
 35%|███▍      | 22912/65536 [3:55:54<7:08:46,  1.66it/s]
 35%|███▍      | 22913/65536 [3:55:55<7:05:12,  1.67it/s]
 35%|███▍      | 22914/65536 [3:55:56<7:12:41,  1.64it/s]
 35%|███▍      | 22915/65536 [3:55:56<7:06:44,  1.66it/s]
 35%|███▍      | 22916/65536 [3:55:57<7:12:50,  1.64it/s]
 35%|███▍      | 22917/65536 [3:55:57<7:23:03,  1.60it/s]
 35%|███▍      | 22918/65536 [3:55:58<7:13:35,  1.64it/s]
 35%|███▍      | 22919/65536 [3:55:59<7:21:44,  1.61it/s]
 35%|███▍      | 22920/65536 [3:55:59<7:09:36,  1.65it/s]
                                                         
{'loss': 3.023, 'learning_rate': 6.87033182319089e-07, 'epoch': 1414.81}
+
 35%|███▍      | 22920/65536 [3:55:59<7:09:36,  1.65it/s]
 35%|███▍      | 22921/65536 [3:56:00<7:06:41,  1.66it/s]
 35%|███▍      | 22922/65536 [3:56:00<7:00:00,  1.69it/s]
 35%|███▍      | 22923/65536 [3:56:01<7:00:12,  1.69it/s]
 35%|███▍      | 22924/65536 [3:56:02<7:14:33,  1.63it/s]
 35%|███▍      | 22925/65536 [3:56:02<7:12:39,  1.64it/s]
 35%|███▍      | 22926/65536 [3:56:03<7:10:48,  1.65it/s]
 35%|███▍      | 22927/65536 [3:56:03<7:11:01,  1.65it/s]
 35%|███▍      | 22928/65536 [3:56:04<7:23:18,  1.60it/s]
 35%|███▍      | 22929/65536 [3:56:05<7:33:18,  1.57it/s]
 35%|███▍      | 22930/65536 [3:56:05<7:14:37,  1.63it/s]
 35%|███▍      | 22931/65536 [3:56:06<7:16:45,  1.63it/s]
 35%|███▍      | 22932/65536 [3:56:07<7:16:09,  1.63it/s]
 35%|███▍      | 22933/65536 [3:56:07<7:15:37,  1.63it/s]
 35%|███▍      | 22934/65536 [3:56:08<7:05:38,  1.67it/s]
 35%|███▍      | 22935/65536 [3:56:08<7:08:46,  1.66it/s]
 35%|███▍      | 22936/65536 [3:56:09<7:14:27,  1.63it/s]
 35%|███▍      | 22937/65536 [3:56:10<7:16:29,  1.63it/s]
 35%|███▌      | 22938/65536 [3:56:10<7:22:07,  1.61it/s]
 35%|███▌      | 22939/65536 [3:56:11<7:24:23,  1.60it/s]
 35%|███▌      | 22940/65536 [3:56:12<7:39:12,  1.55it/s]
                                                         
{'loss': 3.0236, 'learning_rate': 6.86757683359863e-07, 'epoch': 1416.05}
+
 35%|███▌      | 22940/65536 [3:56:12<7:39:12,  1.55it/s]
 35%|███▌      | 22941/65536 [3:56:12<7:27:56,  1.58it/s]
 35%|███▌      | 22942/65536 [3:56:13<7:16:04,  1.63it/s]
 35%|███▌      | 22943/65536 [3:56:13<7:06:21,  1.66it/s]
 35%|███▌      | 22944/65536 [3:56:14<7:01:26,  1.68it/s]
 35%|███▌      | 22945/65536 [3:56:14<7:01:39,  1.68it/s]
 35%|███▌      | 22946/65536 [3:56:15<7:02:21,  1.68it/s]
 35%|███▌      | 22947/65536 [3:56:16<7:10:23,  1.65it/s]
 35%|███▌      | 22948/65536 [3:56:16<7:15:59,  1.63it/s]
 35%|███▌      | 22949/65536 [3:56:17<7:21:31,  1.61it/s]
 35%|███▌      | 22950/65536 [3:56:18<7:15:21,  1.63it/s]
 35%|███▌      | 22951/65536 [3:56:18<7:22:49,  1.60it/s]
 35%|███▌      | 22952/65536 [3:56:19<7:29:06,  1.58it/s]
 35%|███▌      | 22953/65536 [3:56:19<7:20:15,  1.61it/s]
 35%|███▌      | 22954/65536 [3:56:20<7:21:50,  1.61it/s]
 35%|███▌      | 22955/65536 [3:56:21<7:24:51,  1.60it/s]
 35%|███▌      | 22956/65536 [3:56:21<7:38:30,  1.55it/s]
 35%|███▌      | 22957/65536 [3:56:22<7:33:43,  1.56it/s]
 35%|███▌      | 22958/65536 [3:56:23<7:25:03,  1.59it/s]
 35%|███▌      | 22959/65536 [3:56:23<7:12:08,  1.64it/s]
 35%|███▌      | 22960/65536 [3:56:24<7:16:01,  1.63it/s]
                                                         
{'loss': 3.0156, 'learning_rate': 6.864821844006366e-07, 'epoch': 1417.28}
+
 35%|███▌      | 22960/65536 [3:56:24<7:16:01,  1.63it/s]
 35%|███▌      | 22961/65536 [3:56:24<7:08:02,  1.66it/s]
 35%|███▌      | 22962/65536 [3:56:25<7:12:31,  1.64it/s]
 35%|███▌      | 22963/65536 [3:56:26<7:06:22,  1.66it/s]
 35%|███▌      | 22964/65536 [3:56:26<7:03:11,  1.68it/s]
 35%|███▌      | 22965/65536 [3:56:27<7:06:04,  1.67it/s]
 35%|███▌      | 22966/65536 [3:56:27<7:02:11,  1.68it/s]
 35%|███▌      | 22967/65536 [3:56:28<7:14:06,  1.63it/s]
 35%|███▌      | 22968/65536 [3:56:29<7:05:18,  1.67it/s]
 35%|███▌      | 22969/65536 [3:56:29<7:11:29,  1.64it/s]
 35%|███▌      | 22970/65536 [3:56:30<7:05:12,  1.67it/s]
 35%|███▌      | 22971/65536 [3:56:30<7:13:56,  1.63it/s]
 35%|███▌      | 22972/65536 [3:56:31<7:26:55,  1.59it/s]
 35%|███▌      | 22973/65536 [3:56:32<7:30:25,  1.57it/s]
 35%|███▌      | 22974/65536 [3:56:32<7:23:51,  1.60it/s]
 35%|███▌      | 22975/65536 [3:56:33<7:15:56,  1.63it/s]
 35%|███▌      | 22976/65536 [3:56:34<7:16:12,  1.63it/s]
 35%|███▌      | 22977/65536 [3:56:34<7:15:49,  1.63it/s]
 35%|███▌      | 22978/65536 [3:56:35<7:16:50,  1.62it/s]
 35%|███▌      | 22979/65536 [3:56:35<7:20:20,  1.61it/s]
 35%|███▌      | 22980/65536 [3:56:36<7:17:39,  1.62it/s]
                                                         
{'loss': 2.9708, 'learning_rate': 6.862066854414106e-07, 'epoch': 1418.52}
+
 35%|███▌      | 22980/65536 [3:56:36<7:17:39,  1.62it/s]
 35%|███▌      | 22981/65536 [3:56:37<7:12:26,  1.64it/s]
 35%|███▌      | 22982/65536 [3:56:37<7:14:57,  1.63it/s]
 35%|███▌      | 22983/65536 [3:56:38<7:14:02,  1.63it/s]
 35%|███▌      | 22984/65536 [3:56:38<7:07:34,  1.66it/s]
 35%|███▌      | 22985/65536 [3:56:39<7:07:41,  1.66it/s]
 35%|███▌      | 22986/65536 [3:56:40<7:13:59,  1.63it/s]
 35%|███▌      | 22987/65536 [3:56:40<7:11:27,  1.64it/s]
 35%|███▌      | 22988/65536 [3:56:41<7:17:21,  1.62it/s]
 35%|███▌      | 22989/65536 [3:56:42<7:16:22,  1.63it/s]
 35%|███▌      | 22990/65536 [3:56:42<7:15:22,  1.63it/s]
 35%|███▌      | 22991/65536 [3:56:43<7:17:13,  1.62it/s]
 35%|███▌      | 22992/65536 [3:56:43<7:13:57,  1.63it/s]
 35%|███▌      | 22993/65536 [3:56:44<7:11:23,  1.64it/s]
 35%|███▌      | 22994/65536 [3:56:45<7:02:46,  1.68it/s]
 35%|███▌      | 22995/65536 [3:56:45<7:09:10,  1.65it/s]
 35%|███▌      | 22996/65536 [3:56:46<7:24:41,  1.59it/s]
 35%|███▌      | 22997/65536 [3:56:46<7:13:53,  1.63it/s]
 35%|███▌      | 22998/65536 [3:56:47<7:11:02,  1.64it/s]
 35%|███▌      | 22999/65536 [3:56:48<7:08:21,  1.66it/s]
 35%|███▌      | 23000/65536 [3:56:48<7:00:50,  1.68it/s]
                                                         
{'loss': 3.0535, 'learning_rate': 6.859311864821844e-07, 'epoch': 1419.75}
+
 35%|███▌      | 23000/65536 [3:56:48<7:00:50,  1.68it/s]
 35%|███▌      | 23001/65536 [3:56:49<7:18:53,  1.62it/s]
 35%|███▌      | 23002/65536 [3:56:50<7:17:34,  1.62it/s]
 35%|███▌      | 23003/65536 [3:56:50<7:26:24,  1.59it/s]
 35%|███▌      | 23004/65536 [3:56:51<7:16:22,  1.62it/s]
 35%|███▌      | 23005/65536 [3:56:51<7:18:39,  1.62it/s]
 35%|███▌      | 23006/65536 [3:56:52<7:15:20,  1.63it/s]
 35%|███▌      | 23007/65536 [3:56:53<7:07:36,  1.66it/s]
 35%|███▌      | 23008/65536 [3:56:53<7:05:36,  1.67it/s]
 35%|███▌      | 23009/65536 [3:56:54<7:06:37,  1.66it/s]
 35%|███▌      | 23010/65536 [3:56:54<7:20:52,  1.61it/s]
 35%|███▌      | 23011/65536 [3:56:55<7:26:39,  1.59it/s]
 35%|███▌      | 23012/65536 [3:56:56<7:26:48,  1.59it/s]
 35%|███▌      | 23013/65536 [3:56:56<7:19:15,  1.61it/s]
 35%|███▌      | 23014/65536 [3:56:57<7:24:28,  1.59it/s]
 35%|███▌      | 23015/65536 [3:56:58<7:20:30,  1.61it/s]
 35%|███▌      | 23016/65536 [3:56:58<7:18:49,  1.61it/s]
 35%|███▌      | 23017/65536 [3:56:59<7:09:32,  1.65it/s]
 35%|███▌      | 23018/65536 [3:56:59<7:10:32,  1.65it/s]
 35%|███▌      | 23019/65536 [3:57:00<7:05:45,  1.66it/s]
 35%|███▌      | 23020/65536 [3:57:01<7:00:54,  1.68it/s]
                                                         
{'loss': 2.9992, 'learning_rate': 6.856556875229582e-07, 'epoch': 1420.99}
+
 35%|███▌      | 23020/65536 [3:57:01<7:00:54,  1.68it/s]
 35%|███▌      | 23021/65536 [3:57:01<7:28:26,  1.58it/s]
 35%|███▌      | 23022/65536 [3:57:02<7:27:54,  1.58it/s]
 35%|███▌      | 23023/65536 [3:57:02<7:13:11,  1.64it/s]
 35%|███▌      | 23024/65536 [3:57:03<7:18:41,  1.62it/s]
 35%|███▌      | 23025/65536 [3:57:04<7:18:04,  1.62it/s]
 35%|███▌      | 23026/65536 [3:57:04<7:17:50,  1.62it/s]
 35%|███▌      | 23027/65536 [3:57:05<7:06:06,  1.66it/s]
 35%|███▌      | 23028/65536 [3:57:06<7:27:37,  1.58it/s]
 35%|███▌      | 23029/65536 [3:57:06<7:31:23,  1.57it/s]
 35%|███▌      | 23030/65536 [3:57:07<7:25:11,  1.59it/s]
 35%|███▌      | 23031/65536 [3:57:07<7:17:03,  1.62it/s]
 35%|███▌      | 23032/65536 [3:57:08<7:05:48,  1.66it/s]
 35%|███▌      | 23033/65536 [3:57:09<7:06:37,  1.66it/s]
 35%|███▌      | 23034/65536 [3:57:09<7:17:30,  1.62it/s]
 35%|███▌      | 23035/65536 [3:57:10<7:20:30,  1.61it/s]
 35%|███▌      | 23036/65536 [3:57:10<7:09:01,  1.65it/s]
 35%|███▌      | 23037/65536 [3:57:11<7:16:21,  1.62it/s]
 35%|███▌      | 23038/65536 [3:57:12<7:21:39,  1.60it/s]
 35%|███▌      | 23039/65536 [3:57:12<7:30:25,  1.57it/s]
 35%|███▌      | 23040/65536 [3:57:13<7:25:43,  1.59it/s]
                                                         
{'loss': 2.9647, 'learning_rate': 6.85380188563732e-07, 'epoch': 1422.22}
+
 35%|███▌      | 23040/65536 [3:57:13<7:25:43,  1.59it/s]
 35%|███▌      | 23041/65536 [3:57:14<7:28:35,  1.58it/s]
 35%|███▌      | 23042/65536 [3:57:14<7:17:09,  1.62it/s]
 35%|███▌      | 23043/65536 [3:57:15<7:19:00,  1.61it/s]
 35%|███▌      | 23044/65536 [3:57:15<7:19:19,  1.61it/s]
 35%|███▌      | 23045/65536 [3:57:16<7:18:38,  1.61it/s]
 35%|███▌      | 23046/65536 [3:57:17<7:14:08,  1.63it/s]
 35%|███▌      | 23047/65536 [3:57:17<7:17:24,  1.62it/s]
 35%|███▌      | 23048/65536 [3:57:18<7:09:22,  1.65it/s]
 35%|███▌      | 23049/65536 [3:57:18<7:04:44,  1.67it/s]
 35%|███▌      | 23050/65536 [3:57:19<7:15:42,  1.63it/s]
 35%|███▌      | 23051/65536 [3:57:20<7:06:56,  1.66it/s]
 35%|███▌      | 23052/65536 [3:57:20<7:11:18,  1.64it/s]
 35%|███▌      | 23053/65536 [3:57:21<7:22:06,  1.60it/s]
 35%|███▌      | 23054/65536 [3:57:22<7:18:47,  1.61it/s]
 35%|███▌      | 23055/65536 [3:57:22<7:10:25,  1.64it/s]
 35%|███▌      | 23056/65536 [3:57:23<7:19:55,  1.61it/s]
 35%|███▌      | 23057/65536 [3:57:23<7:21:37,  1.60it/s]
 35%|███▌      | 23058/65536 [3:57:24<7:14:45,  1.63it/s]
 35%|███▌      | 23059/65536 [3:57:25<7:07:30,  1.66it/s]
 35%|███▌      | 23060/65536 [3:57:25<7:26:18,  1.59it/s]
                                                         
{'loss': 2.9628, 'learning_rate': 6.851046896045059e-07, 'epoch': 1423.46}
+
 35%|███▌      | 23060/65536 [3:57:25<7:26:18,  1.59it/s]
 35%|███▌      | 23061/65536 [3:57:26<7:30:29,  1.57it/s]
 35%|███▌      | 23062/65536 [3:57:27<7:24:27,  1.59it/s]
 35%|███▌      | 23063/65536 [3:57:27<7:17:17,  1.62it/s]
 35%|███▌      | 23064/65536 [3:57:28<7:11:14,  1.64it/s]
 35%|███▌      | 23065/65536 [3:57:28<7:01:51,  1.68it/s]
 35%|███▌      | 23066/65536 [3:57:29<7:14:15,  1.63it/s]
 35%|███▌      | 23067/65536 [3:57:30<7:12:12,  1.64it/s]
 35%|███▌      | 23068/65536 [3:57:30<6:59:20,  1.69it/s]
 35%|███▌      | 23069/65536 [3:57:31<7:21:32,  1.60it/s]
 35%|███▌      | 23070/65536 [3:57:31<7:12:47,  1.64it/s]
 35%|███▌      | 23071/65536 [3:57:32<7:08:51,  1.65it/s]
 35%|███▌      | 23072/65536 [3:57:33<7:16:17,  1.62it/s]
 35%|███▌      | 23073/65536 [3:57:33<7:18:41,  1.61it/s]
 35%|███▌      | 23074/65536 [3:57:34<7:09:07,  1.65it/s]
 35%|███▌      | 23075/65536 [3:57:34<7:11:35,  1.64it/s]
 35%|███▌      | 23076/65536 [3:57:35<7:16:31,  1.62it/s]
 35%|███▌      | 23077/65536 [3:57:36<7:10:47,  1.64it/s]
 35%|███▌      | 23078/65536 [3:57:36<7:10:49,  1.64it/s]
 35%|███▌      | 23079/65536 [3:57:37<7:11:17,  1.64it/s]
 35%|███▌      | 23080/65536 [3:57:38<7:10:17,  1.64it/s]
                                                         
{'loss': 3.0062, 'learning_rate': 6.848291906452798e-07, 'epoch': 1424.69}
+
 35%|███▌      | 23080/65536 [3:57:38<7:10:17,  1.64it/s]
 35%|███▌      | 23081/65536 [3:57:38<7:16:30,  1.62it/s]
 35%|███▌      | 23082/65536 [3:57:39<7:13:07,  1.63it/s]
 35%|███▌      | 23083/65536 [3:57:39<7:09:06,  1.65it/s]
 35%|███▌      | 23084/65536 [3:57:40<7:03:14,  1.67it/s]
 35%|███▌      | 23085/65536 [3:57:41<7:06:44,  1.66it/s]
 35%|███▌      | 23086/65536 [3:57:41<7:12:08,  1.64it/s]
 35%|███▌      | 23087/65536 [3:57:42<7:17:47,  1.62it/s]
 35%|███▌      | 23088/65536 [3:57:42<7:15:12,  1.63it/s]
 35%|███▌      | 23089/65536 [3:57:43<7:09:45,  1.65it/s]
 35%|███▌      | 23090/65536 [3:57:44<7:03:01,  1.67it/s]
 35%|███▌      | 23091/65536 [3:57:44<7:21:23,  1.60it/s]
 35%|███▌      | 23092/65536 [3:57:45<7:30:26,  1.57it/s]
 35%|███▌      | 23093/65536 [3:57:46<7:21:47,  1.60it/s]
 35%|███▌      | 23094/65536 [3:57:46<7:21:53,  1.60it/s]
 35%|███▌      | 23095/65536 [3:57:47<7:16:34,  1.62it/s]
 35%|███▌      | 23096/65536 [3:57:47<7:11:53,  1.64it/s]
 35%|███▌      | 23097/65536 [3:57:48<7:12:31,  1.64it/s]
 35%|███▌      | 23098/65536 [3:57:49<7:14:11,  1.63it/s]
 35%|███▌      | 23099/65536 [3:57:49<7:12:05,  1.64it/s]
 35%|███▌      | 23100/65536 [3:57:50<7:10:28,  1.64it/s]
                                                         
{'loss': 2.943, 'learning_rate': 6.845536916860536e-07, 'epoch': 1425.93}
+
 35%|███▌      | 23100/65536 [3:57:50<7:10:28,  1.64it/s]
 35%|███▌      | 23101/65536 [3:57:50<7:13:37,  1.63it/s]
 35%|███▌      | 23102/65536 [3:57:51<7:33:03,  1.56it/s]
 35%|███▌      | 23103/65536 [3:57:52<7:26:59,  1.58it/s]
 35%|███▌      | 23104/65536 [3:57:52<7:18:29,  1.61it/s]
 35%|███▌      | 23105/65536 [3:57:53<7:16:38,  1.62it/s]
 35%|███▌      | 23106/65536 [3:57:54<7:08:01,  1.65it/s]
 35%|███▌      | 23107/65536 [3:57:54<7:16:24,  1.62it/s]
 35%|███▌      | 23108/65536 [3:57:55<7:10:14,  1.64it/s]
 35%|███▌      | 23109/65536 [3:57:55<7:13:14,  1.63it/s]
 35%|███▌      | 23110/65536 [3:57:56<7:04:34,  1.67it/s]
 35%|███▌      | 23111/65536 [3:57:57<7:14:39,  1.63it/s]
 35%|███▌      | 23112/65536 [3:57:57<7:27:49,  1.58it/s]
 35%|███▌      | 23113/65536 [3:57:58<7:13:02,  1.63it/s]
 35%|███▌      | 23114/65536 [3:57:59<7:26:40,  1.58it/s]
 35%|███▌      | 23115/65536 [3:57:59<7:14:51,  1.63it/s]
 35%|███▌      | 23116/65536 [3:58:00<7:05:59,  1.66it/s]
 35%|███▌      | 23117/65536 [3:58:00<7:00:53,  1.68it/s]
 35%|███▌      | 23118/65536 [3:58:01<7:13:04,  1.63it/s]
 35%|███▌      | 23119/65536 [3:58:01<7:09:44,  1.65it/s]
 35%|███▌      | 23120/65536 [3:58:02<7:05:22,  1.66it/s]
                                                         
{'loss': 3.0331, 'learning_rate': 6.842781927268275e-07, 'epoch': 1427.16}
+
 35%|███▌      | 23120/65536 [3:58:02<7:05:22,  1.66it/s]
 35%|███▌      | 23121/65536 [3:58:03<7:05:29,  1.66it/s]
 35%|███▌      | 23122/65536 [3:58:03<7:14:32,  1.63it/s]
 35%|███▌      | 23123/65536 [3:58:04<7:18:57,  1.61it/s]
 35%|███▌      | 23124/65536 [3:58:05<7:17:13,  1.62it/s]
 35%|███▌      | 23125/65536 [3:58:05<7:04:47,  1.66it/s]
 35%|███▌      | 23126/65536 [3:58:06<7:09:06,  1.65it/s]
 35%|███▌      | 23127/65536 [3:58:06<7:08:45,  1.65it/s]
 35%|███▌      | 23128/65536 [3:58:07<7:06:11,  1.66it/s]
 35%|███▌      | 23129/65536 [3:58:08<7:01:53,  1.68it/s]
 35%|███▌      | 23130/65536 [3:58:08<7:15:54,  1.62it/s]
 35%|███▌      | 23131/65536 [3:58:09<7:15:46,  1.62it/s]
 35%|███▌      | 23132/65536 [3:58:09<7:21:51,  1.60it/s]
 35%|███▌      | 23133/65536 [3:58:10<7:12:32,  1.63it/s]
 35%|███▌      | 23134/65536 [3:58:11<7:23:13,  1.59it/s]
 35%|███▌      | 23135/65536 [3:58:11<7:24:34,  1.59it/s]
 35%|███▌      | 23136/65536 [3:58:12<7:13:24,  1.63it/s]
 35%|███▌      | 23137/65536 [3:58:12<7:06:26,  1.66it/s]
 35%|███▌      | 23138/65536 [3:58:13<7:24:25,  1.59it/s]
 35%|███▌      | 23139/65536 [3:58:14<7:33:15,  1.56it/s]
 35%|███▌      | 23140/65536 [3:58:14<7:23:26,  1.59it/s]
                                                         
{'loss': 3.0069, 'learning_rate': 6.840026937676013e-07, 'epoch': 1428.4}
+
 35%|███▌      | 23140/65536 [3:58:14<7:23:26,  1.59it/s]
 35%|███▌      | 23141/65536 [3:58:15<7:20:21,  1.60it/s]
 35%|███▌      | 23142/65536 [3:58:16<7:09:28,  1.65it/s]
 35%|███▌      | 23143/65536 [3:58:16<7:00:09,  1.68it/s]
 35%|███▌      | 23144/65536 [3:58:17<6:53:01,  1.71it/s]
 35%|███▌      | 23145/65536 [3:58:17<6:58:38,  1.69it/s]
 35%|███▌      | 23146/65536 [3:58:18<7:16:11,  1.62it/s]
 35%|███▌      | 23147/65536 [3:58:19<7:11:51,  1.64it/s]
 35%|███▌      | 23148/65536 [3:58:19<7:00:10,  1.68it/s]
 35%|███▌      | 23149/65536 [3:58:20<7:06:54,  1.65it/s]
 35%|███▌      | 23150/65536 [3:58:20<7:19:04,  1.61it/s]
 35%|███▌      | 23151/65536 [3:58:21<7:13:25,  1.63it/s]
 35%|███▌      | 23152/65536 [3:58:22<7:10:37,  1.64it/s]
 35%|███▌      | 23153/65536 [3:58:22<7:02:24,  1.67it/s]
 35%|███▌      | 23154/65536 [3:58:23<6:54:46,  1.70it/s]
 35%|███▌      | 23155/65536 [3:58:24<7:17:57,  1.61it/s]
 35%|███▌      | 23156/65536 [3:58:24<7:29:00,  1.57it/s]
 35%|███▌      | 23157/65536 [3:58:25<7:19:23,  1.61it/s]
 35%|███▌      | 23158/65536 [3:58:25<7:13:17,  1.63it/s]
 35%|███▌      | 23159/65536 [3:58:26<7:11:17,  1.64it/s]
 35%|███▌      | 23160/65536 [3:58:27<7:15:04,  1.62it/s]
                                                         
{'loss': 3.1004, 'learning_rate': 6.837271948083752e-07, 'epoch': 1429.63}
+
 35%|███▌      | 23160/65536 [3:58:27<7:15:04,  1.62it/s]
 35%|███▌      | 23161/65536 [3:58:27<7:18:57,  1.61it/s]
 35%|███▌      | 23162/65536 [3:58:28<7:12:15,  1.63it/s]
 35%|███▌      | 23163/65536 [3:58:28<7:13:40,  1.63it/s]
 35%|███▌      | 23164/65536 [3:58:29<7:13:04,  1.63it/s]
 35%|███▌      | 23165/65536 [3:58:30<7:19:45,  1.61it/s]
 35%|███▌      | 23166/65536 [3:58:30<7:09:20,  1.64it/s]
 35%|███▌      | 23167/65536 [3:58:31<7:23:00,  1.59it/s]
 35%|███▌      | 23168/65536 [3:58:32<7:12:30,  1.63it/s]
 35%|███▌      | 23169/65536 [3:58:32<7:12:56,  1.63it/s]
 35%|███▌      | 23170/65536 [3:58:33<7:19:21,  1.61it/s]
 35%|███▌      | 23171/65536 [3:58:33<7:17:59,  1.61it/s]
 35%|███▌      | 23172/65536 [3:58:34<7:14:15,  1.63it/s]
 35%|███▌      | 23173/65536 [3:58:35<7:11:08,  1.64it/s]
 35%|███▌      | 23174/65536 [3:58:35<7:18:43,  1.61it/s]
 35%|███▌      | 23175/65536 [3:58:36<7:12:12,  1.63it/s]
 35%|███▌      | 23176/65536 [3:58:36<7:08:32,  1.65it/s]
 35%|███▌      | 23177/65536 [3:58:37<6:58:47,  1.69it/s]
 35%|███▌      | 23178/65536 [3:58:38<7:08:47,  1.65it/s]
 35%|███▌      | 23179/65536 [3:58:38<7:15:17,  1.62it/s]
 35%|███▌      | 23180/65536 [3:58:39<7:27:45,  1.58it/s]
                                                         
{'loss': 2.9466, 'learning_rate': 6.83451695849149e-07, 'epoch': 1430.86}
+
 35%|███▌      | 23180/65536 [3:58:39<7:27:45,  1.58it/s]
 35%|███▌      | 23181/65536 [3:58:40<7:23:12,  1.59it/s]
 35%|███▌      | 23182/65536 [3:58:40<7:14:23,  1.63it/s]
 35%|███▌      | 23183/65536 [3:58:41<7:28:25,  1.57it/s]
 35%|███▌      | 23184/65536 [3:58:41<7:24:48,  1.59it/s]
 35%|███▌      | 23185/65536 [3:58:42<7:25:32,  1.58it/s]
 35%|███▌      | 23186/65536 [3:58:43<7:18:39,  1.61it/s]
 35%|███▌      | 23187/65536 [3:58:43<7:07:50,  1.65it/s]
 35%|███▌      | 23188/65536 [3:58:44<7:14:06,  1.63it/s]
 35%|███▌      | 23189/65536 [3:58:45<7:15:37,  1.62it/s]
 35%|███▌      | 23190/65536 [3:58:45<7:15:29,  1.62it/s]
 35%|███▌      | 23191/65536 [3:58:46<7:16:51,  1.62it/s]
 35%|███▌      | 23192/65536 [3:58:46<7:19:17,  1.61it/s]
 35%|███▌      | 23193/65536 [3:58:47<7:10:18,  1.64it/s]
 35%|███▌      | 23194/65536 [3:58:48<7:03:56,  1.66it/s]
 35%|███▌      | 23195/65536 [3:58:48<7:02:38,  1.67it/s]
 35%|███▌      | 23196/65536 [3:58:49<7:16:31,  1.62it/s]
 35%|███▌      | 23197/65536 [3:58:49<7:05:28,  1.66it/s]
 35%|███▌      | 23198/65536 [3:58:50<7:12:40,  1.63it/s]
 35%|███▌      | 23199/65536 [3:58:51<7:19:38,  1.60it/s]
 35%|███▌      | 23200/65536 [3:58:51<7:20:57,  1.60it/s]
                                                         
{'loss': 3.0308, 'learning_rate': 6.831761968899229e-07, 'epoch': 1432.1}
+
 35%|███▌      | 23200/65536 [3:58:51<7:20:57,  1.60it/s]
 35%|███▌      | 23201/65536 [3:58:52<7:14:48,  1.62it/s]
 35%|███▌      | 23202/65536 [3:58:52<7:15:51,  1.62it/s]
 35%|███▌      | 23203/65536 [3:58:53<7:29:49,  1.57it/s]
 35%|███▌      | 23204/65536 [3:58:54<7:28:44,  1.57it/s]
 35%|███▌      | 23205/65536 [3:58:54<7:25:07,  1.58it/s]
 35%|███▌      | 23206/65536 [3:58:55<7:09:09,  1.64it/s]
 35%|███▌      | 23207/65536 [3:58:56<7:13:57,  1.63it/s]
 35%|███▌      | 23208/65536 [3:58:56<7:09:59,  1.64it/s]
 35%|███▌      | 23209/65536 [3:58:57<7:12:39,  1.63it/s]
 35%|███▌      | 23210/65536 [3:58:57<7:09:26,  1.64it/s]
 35%|███▌      | 23211/65536 [3:58:58<7:14:20,  1.62it/s]
 35%|███▌      | 23212/65536 [3:58:59<7:05:31,  1.66it/s]
 35%|███▌      | 23213/65536 [3:58:59<6:55:47,  1.70it/s]
 35%|███▌      | 23214/65536 [3:59:00<6:55:56,  1.70it/s]
 35%|███▌      | 23215/65536 [3:59:00<7:13:34,  1.63it/s]
 35%|███▌      | 23216/65536 [3:59:01<7:09:19,  1.64it/s]
 35%|███▌      | 23217/65536 [3:59:02<7:12:39,  1.63it/s]
 35%|███▌      | 23218/65536 [3:59:02<7:06:01,  1.66it/s]
 35%|███▌      | 23219/65536 [3:59:03<7:01:47,  1.67it/s]
 35%|███▌      | 23220/65536 [3:59:03<7:03:38,  1.66it/s]
                                                         
{'loss': 3.0082, 'learning_rate': 6.829006979306967e-07, 'epoch': 1433.33}
+
 35%|███▌      | 23220/65536 [3:59:03<7:03:38,  1.66it/s]
 35%|███▌      | 23221/65536 [3:59:04<6:58:37,  1.68it/s]
 35%|███▌      | 23222/65536 [3:59:05<6:58:52,  1.68it/s]
 35%|███▌      | 23223/65536 [3:59:05<7:09:54,  1.64it/s]
 35%|███▌      | 23224/65536 [3:59:06<7:13:53,  1.63it/s]
 35%|███▌      | 23225/65536 [3:59:06<7:09:06,  1.64it/s]
 35%|███▌      | 23226/65536 [3:59:07<7:14:21,  1.62it/s]
 35%|███▌      | 23227/65536 [3:59:08<7:12:59,  1.63it/s]
 35%|███▌      | 23228/65536 [3:59:08<7:15:24,  1.62it/s]
 35%|███▌      | 23229/65536 [3:59:09<7:07:48,  1.65it/s]
 35%|███▌      | 23230/65536 [3:59:10<7:11:43,  1.63it/s]
 35%|███▌      | 23231/65536 [3:59:10<7:34:00,  1.55it/s]
 35%|███▌      | 23232/65536 [3:59:11<7:23:14,  1.59it/s]
 35%|███▌      | 23233/65536 [3:59:11<7:12:09,  1.63it/s]
 35%|███▌      | 23234/65536 [3:59:12<6:59:16,  1.68it/s]
 35%|███▌      | 23235/65536 [3:59:13<7:05:51,  1.66it/s]
 35%|███▌      | 23236/65536 [3:59:13<7:10:03,  1.64it/s]
 35%|███▌      | 23237/65536 [3:59:14<7:07:57,  1.65it/s]
 35%|███▌      | 23238/65536 [3:59:14<7:15:26,  1.62it/s]
 35%|███▌      | 23239/65536 [3:59:15<7:11:53,  1.63it/s]
 35%|███▌      | 23240/65536 [3:59:16<7:02:24,  1.67it/s]
                                                         
{'loss': 2.9983, 'learning_rate': 6.826251989714705e-07, 'epoch': 1434.57}
+
 35%|███▌      | 23240/65536 [3:59:16<7:02:24,  1.67it/s]
 35%|███▌      | 23241/65536 [3:59:16<7:03:54,  1.66it/s]
 35%|███▌      | 23242/65536 [3:59:17<7:01:22,  1.67it/s]
 35%|███▌      | 23243/65536 [3:59:17<7:03:55,  1.66it/s]
 35%|███▌      | 23244/65536 [3:59:18<7:16:47,  1.61it/s]
 35%|███▌      | 23245/65536 [3:59:19<7:34:10,  1.55it/s]
 35%|███▌      | 23246/65536 [3:59:19<7:30:46,  1.56it/s]
 35%|█���█▌      | 23247/65536 [3:59:20<7:15:27,  1.62it/s]
 35%|███▌      | 23248/65536 [3:59:21<7:31:20,  1.56it/s]
 35%|███▌      | 23249/65536 [3:59:21<7:19:20,  1.60it/s]
 35%|███▌      | 23250/65536 [3:59:22<7:15:05,  1.62it/s]
 35%|███▌      | 23251/65536 [3:59:22<7:07:00,  1.65it/s]
 35%|███▌      | 23252/65536 [3:59:23<7:13:12,  1.63it/s]
 35%|███▌      | 23253/65536 [3:59:24<7:14:30,  1.62it/s]
 35%|███▌      | 23254/65536 [3:59:24<7:16:49,  1.61it/s]
 35%|███▌      | 23255/65536 [3:59:25<7:30:29,  1.56it/s]
 35%|███▌      | 23256/65536 [3:59:26<7:34:15,  1.55it/s]
 35%|███▌      | 23257/65536 [3:59:26<7:30:01,  1.57it/s]
 35%|███▌      | 23258/65536 [3:59:27<7:20:10,  1.60it/s]
 35%|███▌      | 23259/65536 [3:59:28<7:16:42,  1.61it/s]
 35%|███▌      | 23260/65536 [3:59:28<7:01:25,  1.67it/s]
                                                         
{'loss': 2.9632, 'learning_rate': 6.823497000122444e-07, 'epoch': 1435.8}
+
 35%|███▌      | 23260/65536 [3:59:28<7:01:25,  1.67it/s]
 35%|███▌      | 23261/65536 [3:59:29<7:06:49,  1.65it/s]
 35%|███▌      | 23262/65536 [3:59:29<7:03:54,  1.66it/s]
 35%|███▌      | 23263/65536 [3:59:30<6:58:10,  1.68it/s]
 35%|███▌      | 23264/65536 [3:59:31<7:16:44,  1.61it/s]
 35%|███▌      | 23265/65536 [3:59:31<7:02:31,  1.67it/s]
 36%|███▌      | 23266/65536 [3:59:32<6:56:04,  1.69it/s]
 36%|███▌      | 23267/65536 [3:59:32<7:05:07,  1.66it/s]
 36%|███▌      | 23268/65536 [3:59:33<7:02:45,  1.67it/s]
 36%|███▌      | 23269/65536 [3:59:34<7:08:03,  1.65it/s]
 36%|███▌      | 23270/65536 [3:59:34<7:18:12,  1.61it/s]
 36%|███▌      | 23271/65536 [3:59:35<7:15:29,  1.62it/s]
 36%|███▌      | 23272/65536 [3:59:35<7:23:13,  1.59it/s]
 36%|███▌      | 23273/65536 [3:59:36<7:14:17,  1.62it/s]
 36%|███▌      | 23274/65536 [3:59:37<7:08:22,  1.64it/s]
 36%|███▌      | 23275/65536 [3:59:37<7:11:37,  1.63it/s]
 36%|███▌      | 23276/65536 [3:59:38<7:04:19,  1.66it/s]
 36%|███▌      | 23277/65536 [3:59:39<7:25:32,  1.58it/s]
 36%|███▌      | 23278/65536 [3:59:39<7:18:35,  1.61it/s]
 36%|███▌      | 23279/65536 [3:59:40<7:11:01,  1.63it/s]
 36%|███▌      | 23280/65536 [3:59:40<7:25:49,  1.58it/s]
                                                         
{'loss': 2.986, 'learning_rate': 6.820742010530181e-07, 'epoch': 1437.04}
+
 36%|███▌      | 23280/65536 [3:59:40<7:25:49,  1.58it/s]
 36%|███▌      | 23281/65536 [3:59:41<7:17:31,  1.61it/s]
 36%|███▌      | 23282/65536 [3:59:42<7:19:41,  1.60it/s]
 36%|███▌      | 23283/65536 [3:59:42<7:14:53,  1.62it/s]
 36%|███▌      | 23284/65536 [3:59:43<7:08:26,  1.64it/s]
 36%|███▌      | 23285/65536 [3:59:43<7:06:51,  1.65it/s]
 36%|███▌      | 23286/65536 [3:59:44<7:11:46,  1.63it/s]
 36%|███▌      | 23287/65536 [3:59:45<7:12:59,  1.63it/s]
 36%|███▌      | 23288/65536 [3:59:45<7:19:43,  1.60it/s]
 36%|███▌      | 23289/65536 [3:59:46<7:25:03,  1.58it/s]
 36%|███▌      | 23290/65536 [3:59:47<7:25:13,  1.58it/s]
 36%|███▌      | 23291/65536 [3:59:47<7:10:21,  1.64it/s]
 36%|███▌      | 23292/65536 [3:59:48<7:19:46,  1.60it/s]
 36%|███▌      | 23293/65536 [3:59:48<7:20:10,  1.60it/s]
 36%|███▌      | 23294/65536 [3:59:49<7:06:15,  1.65it/s]
 36%|███▌      | 23295/65536 [3:59:50<7:00:16,  1.68it/s]
 36%|███▌      | 23296/65536 [3:59:50<7:08:13,  1.64it/s]
 36%|███▌      | 23297/65536 [3:59:51<7:06:04,  1.65it/s]
 36%|███▌      | 23298/65536 [3:59:51<7:12:28,  1.63it/s]
 36%|███▌      | 23299/65536 [3:59:52<7:10:41,  1.63it/s]
 36%|███▌      | 23300/65536 [3:59:53<7:01:10,  1.67it/s]
                                                         
{'loss': 3.0113, 'learning_rate': 6.81798702093792e-07, 'epoch': 1438.27}
+
 36%|███▌      | 23300/65536 [3:59:53<7:01:10,  1.67it/s]
 36%|███▌      | 23301/65536 [3:59:53<7:00:07,  1.68it/s]
 36%|███▌      | 23302/65536 [3:59:54<7:00:02,  1.68it/s]
 36%|███▌      | 23303/65536 [3:59:54<7:00:05,  1.68it/s]
 36%|███▌      | 23304/65536 [3:59:55<6:53:24,  1.70it/s]
 36%|███▌      | 23305/65536 [3:59:56<7:06:49,  1.65it/s]
 36%|███▌      | 23306/65536 [3:59:56<7:08:54,  1.64it/s]
 36%|███▌      | 23307/65536 [3:59:57<7:06:03,  1.65it/s]
 36%|███▌      | 23308/65536 [3:59:57<7:17:13,  1.61it/s]
 36%|███▌      | 23309/65536 [3:59:58<7:19:27,  1.60it/s]
 36%|███▌      | 23310/65536 [3:59:59<7:32:51,  1.55it/s]
 36%|███▌      | 23311/65536 [3:59:59<7:29:43,  1.56it/s]
 36%|███▌      | 23312/65536 [4:00:00<7:31:13,  1.56it/s]
 36%|███▌      | 23313/65536 [4:00:01<7:18:43,  1.60it/s]
 36%|███▌      | 23314/65536 [4:00:01<7:10:36,  1.63it/s]
 36%|███▌      | 23315/65536 [4:00:02<7:18:31,  1.60it/s]
 36%|███▌      | 23316/65536 [4:00:02<7:10:24,  1.63it/s]
 36%|███▌      | 23317/65536 [4:00:03<7:10:37,  1.63it/s]
 36%|███▌      | 23318/65536 [4:00:04<7:02:23,  1.67it/s]
 36%|███▌      | 23319/65536 [4:00:04<7:01:06,  1.67it/s]
 36%|███▌      | 23320/65536 [4:00:05<7:03:18,  1.66it/s]
                                                         
{'loss': 2.9418, 'learning_rate': 6.81523203134566e-07, 'epoch': 1439.51}
+
 36%|███▌      | 23320/65536 [4:00:05<7:03:18,  1.66it/s]
 36%|███▌      | 23321/65536 [4:00:05<7:07:04,  1.65it/s]
 36%|███▌      | 23322/65536 [4:00:06<7:29:02,  1.57it/s]
 36%|███▌      | 23323/65536 [4:00:07<7:28:43,  1.57it/s]
 36%|███▌      | 23324/65536 [4:00:07<7:08:38,  1.64it/s]
 36%|███▌      | 23325/65536 [4:00:08<7:05:44,  1.65it/s]
 36%|███▌      | 23326/65536 [4:00:09<7:15:42,  1.61it/s]
 36%|███▌      | 23327/65536 [4:00:09<7:17:30,  1.61it/s]
 36%|███▌      | 23328/65536 [4:00:10<7:17:46,  1.61it/s]
 36%|███▌      | 23329/65536 [4:00:11<7:39:59,  1.53it/s]
 36%|███▌      | 23330/65536 [4:00:11<7:33:46,  1.55it/s]
 36%|███▌      | 23331/65536 [4:00:12<7:24:59,  1.58it/s]
 36%|███▌      | 23332/65536 [4:00:12<7:14:39,  1.62it/s]
 36%|███▌      | 23333/65536 [4:00:13<7:25:55,  1.58it/s]
 36%|███▌      | 23334/65536 [4:00:14<7:18:10,  1.61it/s]
 36%|███▌      | 23335/65536 [4:00:14<7:09:55,  1.64it/s]
 36%|███▌      | 23336/65536 [4:00:15<7:13:10,  1.62it/s]
 36%|███▌      | 23337/65536 [4:00:15<7:07:26,  1.65it/s]
 36%|███▌      | 23338/65536 [4:00:16<7:08:45,  1.64it/s]
 36%|███▌      | 23339/65536 [4:00:17<7:19:47,  1.60it/s]
 36%|███▌      | 23340/65536 [4:00:17<7:11:40,  1.63it/s]
                                                         
{'loss': 2.9326, 'learning_rate': 6.812477041753397e-07, 'epoch': 1440.74}
+
 36%|███▌      | 23340/65536 [4:00:17<7:11:40,  1.63it/s]
 36%|███▌      | 23341/65536 [4:00:18<7:03:35,  1.66it/s]
 36%|███▌      | 23342/65536 [4:00:19<7:14:43,  1.62it/s]
 36%|███▌      | 23343/65536 [4:00:19<7:05:32,  1.65it/s]
 36%|███▌      | 23344/65536 [4:00:20<7:07:01,  1.65it/s]
 36%|███▌      | 23345/65536 [4:00:20<7:18:28,  1.60it/s]
 36%|███▌      | 23346/65536 [4:00:21<7:06:03,  1.65it/s]
 36%|███▌      | 23347/65536 [4:00:22<7:11:08,  1.63it/s]
 36%|███▌      | 23348/65536 [4:00:22<7:02:01,  1.67it/s]
 36%|███▌      | 23349/65536 [4:00:23<7:07:14,  1.65it/s]
 36%|███▌      | 23350/65536 [4:00:23<7:09:23,  1.64it/s]
 36%|███▌      | 23351/65536 [4:00:24<7:10:28,  1.63it/s]
 36%|███▌      | 23352/65536 [4:00:25<7:11:49,  1.63it/s]
 36%|███▌      | 23353/65536 [4:00:25<7:08:34,  1.64it/s]
 36%|███▌      | 23354/65536 [4:00:26<7:21:45,  1.59it/s]
 36%|███▌      | 23355/65536 [4:00:27<7:10:44,  1.63it/s]
 36%|███▌      | 23356/65536 [4:00:27<7:27:02,  1.57it/s]
 36%|███▌      | 23357/65536 [4:00:28<7:24:51,  1.58it/s]
 36%|███▌      | 23358/65536 [4:00:28<7:21:26,  1.59it/s]
 36%|███▌      | 23359/65536 [4:00:29<7:06:00,  1.65it/s]
 36%|███▌      | 23360/65536 [4:00:30<7:03:37,  1.66it/s]
                                                         
{'loss': 2.9788, 'learning_rate': 6.809722052161136e-07, 'epoch': 1441.98}
+
 36%|███▌      | 23360/65536 [4:00:30<7:03:37,  1.66it/s]
 36%|███▌      | 23361/65536 [4:00:30<7:19:43,  1.60it/s]
 36%|███▌      | 23362/65536 [4:00:31<7:05:18,  1.65it/s]
 36%|███▌      | 23363/65536 [4:00:31<7:08:59,  1.64it/s]
 36%|███▌      | 23364/65536 [4:00:32<7:03:53,  1.66it/s]
 36%|███▌      | 23365/65536 [4:00:33<7:07:15,  1.64it/s]
 36%|███▌      | 23366/65536 [4:00:33<7:24:29,  1.58it/s]
 36%|███▌      | 23367/65536 [4:00:34<7:22:55,  1.59it/s]
 36%|███▌      | 23368/65536 [4:00:35<7:15:34,  1.61it/s]
 36%|███▌      | 23369/65536 [4:00:35<7:17:58,  1.60it/s]
 36%|███▌      | 23370/65536 [4:00:36<7:23:36,  1.58it/s]
 36%|███▌      | 23371/65536 [4:00:37<7:32:21,  1.55it/s]
 36%|███▌      | 23372/65536 [4:00:37<7:13:16,  1.62it/s]
 36%|███▌      | 23373/65536 [4:00:38<7:05:55,  1.65it/s]
 36%|███▌      | 23374/65536 [4:00:38<6:57:02,  1.68it/s]
 36%|███▌      | 23375/65536 [4:00:39<6:58:47,  1.68it/s]
 36%|███▌      | 23376/65536 [4:00:39<6:56:28,  1.69it/s]
 36%|███▌      | 23377/65536 [4:00:40<7:09:26,  1.64it/s]
 36%|███▌      | 23378/65536 [4:00:41<7:11:48,  1.63it/s]
 36%|███▌      | 23379/65536 [4:00:41<7:14:27,  1.62it/s]
 36%|███▌      | 23380/65536 [4:00:42<7:23:55,  1.58it/s]
                                                         
{'loss': 3.0468, 'learning_rate': 6.806967062568874e-07, 'epoch': 1443.21}
+
 36%|███▌      | 23380/65536 [4:00:42<7:23:55,  1.58it/s]
 36%|███▌      | 23381/65536 [4:00:43<7:13:37,  1.62it/s]
 36%|███▌      | 23382/65536 [4:00:43<7:07:19,  1.64it/s]
 36%|███▌      | 23383/65536 [4:00:44<7:05:53,  1.65it/s]
 36%|███▌      | 23384/65536 [4:00:44<7:05:39,  1.65it/s]
 36%|███▌      | 23385/65536 [4:00:45<7:08:55,  1.64it/s]
 36%|███▌      | 23386/65536 [4:00:46<7:21:21,  1.59it/s]
 36%|███▌      | 23387/65536 [4:00:46<7:13:36,  1.62it/s]
 36%|███▌      | 23388/65536 [4:00:47<7:12:59,  1.62it/s]
 36%|███▌      | 23389/65536 [4:00:47<7:06:57,  1.65it/s]
 36%|███▌      | 23390/65536 [4:00:48<7:13:02,  1.62it/s]
 36%|███▌      | 23391/65536 [4:00:49<7:05:48,  1.65it/s]
 36%|███▌      | 23392/65536 [4:00:49<7:12:45,  1.62it/s]
 36%|███▌      | 23393/65536 [4:00:50<7:21:13,  1.59it/s]
 36%|███▌      | 23394/65536 [4:00:51<7:11:45,  1.63it/s]
 36%|███▌      | 23395/65536 [4:00:51<7:13:52,  1.62it/s]
 36%|███▌      | 23396/65536 [4:00:52<7:11:24,  1.63it/s]
 36%|███▌      | 23397/65536 [4:00:52<7:05:27,  1.65it/s]
 36%|███▌      | 23398/65536 [4:00:53<7:10:38,  1.63it/s]
 36%|███▌      | 23399/65536 [4:00:54<6:59:46,  1.67it/s]
 36%|███▌      | 23400/65536 [4:00:54<6:59:37,  1.67it/s]
                                                         
{'loss': 3.0129, 'learning_rate': 6.804212072976613e-07, 'epoch': 1444.44}
+
 36%|███▌      | 23400/65536 [4:00:54<6:59:37,  1.67it/s]
 36%|███▌      | 23401/65536 [4:00:55<7:02:20,  1.66it/s]
 36%|███▌      | 23402/65536 [4:00:55<7:06:27,  1.65it/s]
 36%|███▌      | 23403/65536 [4:00:56<7:07:43,  1.64it/s]
 36%|███▌      | 23404/65536 [4:00:57<7:14:17,  1.62it/s]
 36%|███▌      | 23405/65536 [4:00:57<7:10:59,  1.63it/s]
 36%|███▌      | 23406/65536 [4:00:58<7:19:36,  1.60it/s]
 36%|███▌      | 23407/65536 [4:00:58<7:05:48,  1.65it/s]
 36%|███▌      | 23408/65536 [4:00:59<7:17:24,  1.61it/s]
 36%|███▌      | 23409/65536 [4:01:00<7:19:56,  1.60it/s]
 36%|███▌      | 23410/65536 [4:01:00<7:20:48,  1.59it/s]
 36%|███▌      | 23411/65536 [4:01:01<7:10:57,  1.63it/s]
 36%|███▌      | 23412/65536 [4:01:02<7:10:19,  1.63it/s]
 36%|███▌      | 23413/65536 [4:01:02<7:03:00,  1.66it/s]
 36%|███▌      | 23414/65536 [4:01:03<6:58:17,  1.68it/s]
 36%|███▌      | 23415/65536 [4:01:03<6:55:19,  1.69it/s]
 36%|███▌      | 23416/65536 [4:01:04<6:59:28,  1.67it/s]
 36%|███▌      | 23417/65536 [4:01:05<7:01:27,  1.67it/s]
 36%|███▌      | 23418/65536 [4:01:05<7:14:01,  1.62it/s]
 36%|███▌      | 23419/65536 [4:01:06<7:15:20,  1.61it/s]
 36%|███▌      | 23420/65536 [4:01:06<7:11:36,  1.63it/s]
                                                         
{'loss': 2.9964, 'learning_rate': 6.801457083384351e-07, 'epoch': 1445.68}
+
 36%|███▌      | 23420/65536 [4:01:06<7:11:36,  1.63it/s]
 36%|███▌      | 23421/65536 [4:01:07<7:20:22,  1.59it/s]
 36%|███▌      | 23422/65536 [4:01:08<7:22:22,  1.59it/s]
 36%|███▌      | 23423/65536 [4:01:08<7:20:00,  1.60it/s]
 36%|███▌      | 23424/65536 [4:01:09<7:21:28,  1.59it/s]
 36%|███▌      | 23425/65536 [4:01:10<7:14:07,  1.62it/s]
 36%|███▌      | 23426/65536 [4:01:10<7:23:21,  1.58it/s]
 36%|███▌      | 23427/65536 [4:01:11<7:15:56,  1.61it/s]
 36%|███▌      | 23428/65536 [4:01:11<7:12:33,  1.62it/s]
 36%|███▌      | 23429/65536 [4:01:12<7:15:11,  1.61it/s]
 36%|███▌      | 23430/65536 [4:01:13<7:15:14,  1.61it/s]
 36%|███▌      | 23431/65536 [4:01:13<7:18:34,  1.60it/s]
 36%|███▌      | 23432/65536 [4:01:14<7:24:16,  1.58it/s]
 36%|███▌      | 23433/65536 [4:01:15<7:23:23,  1.58it/s]
 36%|███▌      | 23434/65536 [4:01:15<7:21:33,  1.59it/s]
 36%|███▌      | 23435/65536 [4:01:16<7:19:14,  1.60it/s]
 36%|███▌      | 23436/65536 [4:01:16<7:16:41,  1.61it/s]
 36%|███▌      | 23437/65536 [4:01:17<7:13:54,  1.62it/s]
 36%|███▌      | 23438/65536 [4:01:18<7:04:46,  1.65it/s]
 36%|███▌      | 23439/65536 [4:01:18<7:09:36,  1.63it/s]
 36%|███▌      | 23440/65536 [4:01:19<7:01:04,  1.67it/s]
                                                         
{'loss': 3.0094, 'learning_rate': 6.79870209379209e-07, 'epoch': 1446.91}
+
 36%|███▌      | 23440/65536 [4:01:19<7:01:04,  1.67it/s]
 36%|███▌      | 23441/65536 [4:01:19<6:54:25,  1.69it/s]
 36%|███▌      | 23442/65536 [4:01:20<7:18:50,  1.60it/s]
 36%|███▌      | 23443/65536 [4:01:21<7:09:59,  1.63it/s]
 36%|███▌      | 23444/65536 [4:01:21<7:06:36,  1.64it/s]
 36%|███▌      | 23445/65536 [4:01:22<7:01:08,  1.67it/s]
 36%|███▌      | 23446/65536 [4:01:22<7:03:30,  1.66it/s]
 36%|███▌      | 23447/65536 [4:01:23<7:04:31,  1.65it/s]
 36%|███▌      | 23448/65536 [4:01:24<6:57:55,  1.68it/s]
 36%|███▌      | 23449/65536 [4:01:24<7:01:14,  1.67it/s]
 36%|███▌      | 23450/65536 [4:01:25<6:56:56,  1.68it/s]
 36%|███▌      | 23451/65536 [4:01:25<6:58:39,  1.68it/s]
 36%|███▌      | 23452/65536 [4:01:26<7:13:14,  1.62it/s]
 36%|███▌      | 23453/65536 [4:01:27<7:16:43,  1.61it/s]
 36%|███▌      | 23454/65536 [4:01:27<7:10:57,  1.63it/s]
 36%|███▌      | 23455/65536 [4:01:28<7:13:29,  1.62it/s]
 36%|███▌      | 23456/65536 [4:01:29<7:20:37,  1.59it/s]
 36%|███▌      | 23457/65536 [4:01:29<7:09:53,  1.63it/s]
 36%|███▌      | 23458/65536 [4:01:30<7:41:40,  1.52it/s]
 36%|███▌      | 23459/65536 [4:01:31<7:28:54,  1.56it/s]
 36%|███▌      | 23460/65536 [4:01:31<7:19:45,  1.59it/s]
                                                         
{'loss': 2.9579, 'learning_rate': 6.795947104199829e-07, 'epoch': 1448.15}
+
 36%|███▌      | 23460/65536 [4:01:31<7:19:45,  1.59it/s]
 36%|███▌      | 23461/65536 [4:01:32<7:18:32,  1.60it/s]
 36%|███▌      | 23462/65536 [4:01:32<7:23:03,  1.58it/s]
 36%|███▌      | 23463/65536 [4:01:33<7:13:11,  1.62it/s]
 36%|███▌      | 23464/65536 [4:01:34<7:13:56,  1.62it/s]
 36%|███▌      | 23465/65536 [4:01:34<7:17:47,  1.60it/s]
 36%|███▌      | 23466/65536 [4:01:35<7:09:19,  1.63it/s]
 36%|███▌      | 23467/65536 [4:01:36<7:21:00,  1.59it/s]
 36%|███▌      | 23468/65536 [4:01:36<7:11:02,  1.63it/s]
 36%|███▌      | 23469/65536 [4:01:37<7:06:52,  1.64it/s]
 36%|███▌      | 23470/65536 [4:01:37<6:56:51,  1.68it/s]
 36%|███▌      | 23471/65536 [4:01:38<6:48:16,  1.72it/s]
 36%|███▌      | 23472/65536 [4:01:38<7:03:28,  1.66it/s]
 36%|███▌      | 23473/65536 [4:01:39<6:55:33,  1.69it/s]
 36%|███▌      | 23474/65536 [4:01:40<7:16:46,  1.61it/s]
 36%|███▌      | 23475/65536 [4:01:40<7:14:13,  1.61it/s]
 36%|███▌      | 23476/65536 [4:01:41<7:11:55,  1.62it/s]
 36%|███▌      | 23477/65536 [4:01:42<7:19:20,  1.60it/s]
 36%|███▌      | 23478/65536 [4:01:42<7:14:19,  1.61it/s]
 36%|███▌      | 23479/65536 [4:01:43<7:18:41,  1.60it/s]
 36%|███▌      | 23480/65536 [4:01:43<7:12:23,  1.62it/s]
                                                         
{'loss': 3.0166, 'learning_rate': 6.793192114607567e-07, 'epoch': 1449.38}
+
 36%|███▌      | 23480/65536 [4:01:43<7:12:23,  1.62it/s]
 36%|███▌      | 23481/65536 [4:01:44<6:59:35,  1.67it/s]
 36%|███▌      | 23482/65536 [4:01:45<7:11:05,  1.63it/s]
 36%|███▌      | 23483/65536 [4:01:45<7:09:35,  1.63it/s]
 36%|███▌      | 23484/65536 [4:01:46<7:12:00,  1.62it/s]
 36%|███▌      | 23485/65536 [4:01:47<7:12:39,  1.62it/s]
 36%|███▌      | 23486/65536 [4:01:47<7:04:05,  1.65it/s]
 36%|███▌      | 23487/65536 [4:01:48<6:49:46,  1.71it/s]
 36%|███▌      | 23488/65536 [4:01:48<6:51:48,  1.70it/s]
 36%|███▌      | 23489/65536 [4:01:49<6:57:07,  1.68it/s]
 36%|███▌      | 23490/65536 [4:01:49<7:09:03,  1.63it/s]
 36%|███▌      | 23491/65536 [4:01:50<7:19:55,  1.59it/s]
 36%|███▌      | 23492/65536 [4:01:51<7:26:28,  1.57it/s]
 36%|███▌      | 23493/65536 [4:01:51<7:17:08,  1.60it/s]
 36%|███▌      | 23494/65536 [4:01:52<7:07:39,  1.64it/s]
 36%|███▌      | 23495/65536 [4:01:53<7:12:00,  1.62it/s]
 36%|███▌      | 23496/65536 [4:01:53<7:10:38,  1.63it/s]
 36%|███▌      | 23497/65536 [4:01:54<7:05:00,  1.65it/s]
 36%|███▌      | 23498/65536 [4:01:54<7:04:23,  1.65it/s]
 36%|███▌      | 23499/65536 [4:01:55<7:05:24,  1.65it/s]
 36%|███▌      | 23500/65536 [4:01:56<7:01:57,  1.66it/s]
                                                         
{'loss': 3.0441, 'learning_rate': 6.790437125015305e-07, 'epoch': 1450.62}
+
 36%|███▌      | 23500/65536 [4:01:56<7:01:57,  1.66it/s]
 36%|███▌      | 23501/65536 [4:01:56<6:55:04,  1.69it/s]
 36%|███▌      | 23502/65536 [4:01:57<7:06:14,  1.64it/s]
 36%|███▌      | 23503/65536 [4:01:58<7:23:37,  1.58it/s]
 36%|███▌      | 23504/65536 [4:01:58<7:20:26,  1.59it/s]
 36%|███▌      | 23505/65536 [4:01:59<7:15:57,  1.61it/s]
 36%|███▌      | 23506/65536 [4:01:59<7:11:56,  1.62it/s]
 36%|███▌      | 23507/65536 [4:02:00<7:19:18,  1.59it/s]
 36%|███▌      | 23508/65536 [4:02:01<7:27:58,  1.56it/s]
 36%|███▌      | 23509/65536 [4:02:01<7:18:22,  1.60it/s]
 36%|███▌      | 23510/65536 [4:02:02<7:11:41,  1.62it/s]
 36%|███▌      | 23511/65536 [4:02:02<7:13:20,  1.62it/s]
 36%|███▌      | 23512/65536 [4:02:03<7:09:00,  1.63it/s]
 36%|███▌      | 23513/65536 [4:02:04<7:02:58,  1.66it/s]
 36%|███▌      | 23514/65536 [4:02:04<7:01:05,  1.66it/s]
 36%|███▌      | 23515/65536 [4:02:05<7:12:57,  1.62it/s]
 36%|███▌      | 23516/65536 [4:02:05<6:56:28,  1.68it/s]
 36%|███▌      | 23517/65536 [4:02:06<7:05:28,  1.65it/s]
 36%|███▌      | 23518/65536 [4:02:07<7:10:21,  1.63it/s]
 36%|███▌      | 23519/65536 [4:02:07<7:19:49,  1.59it/s]
 36%|███▌      | 23520/65536 [4:02:08<7:09:55,  1.63it/s]
                                                         
{'loss': 2.9766, 'learning_rate': 6.787682135423043e-07, 'epoch': 1451.85}
+
 36%|███▌      | 23520/65536 [4:02:08<7:09:55,  1.63it/s]
 36%|███▌      | 23521/65536 [4:02:09<7:13:15,  1.62it/s]
 36%|███▌      | 23522/65536 [4:02:09<7:13:41,  1.61it/s]
 36%|███▌      | 23523/65536 [4:02:10<7:24:40,  1.57it/s]
 36%|███▌      | 23524/65536 [4:02:11<7:23:08,  1.58it/s]
 36%|███▌      | 23525/65536 [4:02:11<7:16:26,  1.60it/s]
 36%|███▌      | 23526/65536 [4:02:12<7:20:12,  1.59it/s]
 36%|███▌      | 23527/65536 [4:02:12<7:10:19,  1.63it/s]
 36%|███▌      | 23528/65536 [4:02:13<6:54:27,  1.69it/s]
 36%|███▌      | 23529/65536 [4:02:14<7:00:12,  1.67it/s]
 36%|███▌      | 23530/65536 [4:02:14<6:58:59,  1.67it/s]
 36%|███▌      | 23531/65536 [4:02:15<6:55:56,  1.68it/s]
 36%|███▌      | 23532/65536 [4:02:15<6:59:06,  1.67it/s]
 36%|███▌      | 23533/65536 [4:02:16<7:06:56,  1.64it/s]
 36%|███▌      | 23534/65536 [4:02:17<7:04:26,  1.65it/s]
 36%|███▌      | 23535/65536 [4:02:17<7:04:12,  1.65it/s]
 36%|███▌      | 23536/65536 [4:02:18<7:08:30,  1.63it/s]
 36%|███▌      | 23537/65536 [4:02:18<7:10:13,  1.63it/s]
 36%|███▌      | 23538/65536 [4:02:19<7:04:40,  1.65it/s]
 36%|███▌      | 23539/65536 [4:02:20<7:16:34,  1.60it/s]
 36%|███▌      | 23540/65536 [4:02:20<7:11:24,  1.62it/s]
                                                         
{'loss': 3.0122, 'learning_rate': 6.784927145830782e-07, 'epoch': 1453.09}
+
 36%|███▌      | 23540/65536 [4:02:20<7:11:24,  1.62it/s]
 36%|███▌      | 23541/65536 [4:02:21<7:12:05,  1.62it/s]
 36%|███▌      | 23542/65536 [4:02:21<7:07:25,  1.64it/s]
 36%|███▌      | 23543/65536 [4:02:22<7:03:58,  1.65it/s]
 36%|███▌      | 23544/65536 [4:02:23<7:05:44,  1.64it/s]
 36%|███▌      | 23545/65536 [4:02:23<6:58:57,  1.67it/s]
 36%|███▌      | 23546/65536 [4:02:24<6:57:27,  1.68it/s]
 36%|███▌      | 23547/65536 [4:02:24<6:48:12,  1.71it/s]
 36%|███▌      | 23548/65536 [4:02:25<6:51:00,  1.70it/s]
 36%|███▌      | 23549/65536 [4:02:26<7:04:14,  1.65it/s]
 36%|███▌      | 23550/65536 [4:02:26<7:16:54,  1.60it/s]
 36%|███▌      | 23551/65536 [4:02:27<7:08:08,  1.63it/s]
 36%|███▌      | 23552/65536 [4:02:27<6:54:02,  1.69it/s]
 36%|███▌      | 23553/65536 [4:02:28<7:05:24,  1.64it/s]
 36%|███▌      | 23554/65536 [4:02:29<7:09:39,  1.63it/s]
 36%|███▌      | 23555/65536 [4:02:29<7:18:42,  1.59it/s]
 36%|███▌      | 23556/65536 [4:02:30<7:11:06,  1.62it/s]
 36%|███▌      | 23557/65536 [4:02:31<7:07:03,  1.64it/s]
 36%|███▌      | 23558/65536 [4:02:31<7:04:20,  1.65it/s]
 36%|███▌      | 23559/65536 [4:02:32<7:04:39,  1.65it/s]
 36%|███▌      | 23560/65536 [4:02:32<6:55:24,  1.68it/s]
                                                         
{'loss': 3.0115, 'learning_rate': 6.782172156238519e-07, 'epoch': 1454.32}
+
 36%|███▌      | 23560/65536 [4:02:32<6:55:24,  1.68it/s]
 36%|███▌      | 23561/65536 [4:02:33<7:03:38,  1.65it/s]
 36%|███▌      | 23562/65536 [4:02:34<6:57:49,  1.67it/s]
 36%|███▌      | 23563/65536 [4:02:34<6:54:43,  1.69it/s]
 36%|███▌      | 23564/65536 [4:02:35<6:47:50,  1.72it/s]
 36%|███▌      | 23565/65536 [4:02:35<6:49:44,  1.71it/s]
 36%|███▌      | 23566/65536 [4:02:36<7:20:23,  1.59it/s]
 36%|███▌      | 23567/65536 [4:02:37<7:14:46,  1.61it/s]
 36%|███▌      | 23568/65536 [4:02:37<7:11:30,  1.62it/s]
 36%|███▌      | 23569/65536 [4:02:38<7:06:46,  1.64it/s]
 36%|███▌      | 23570/65536 [4:02:38<7:21:07,  1.59it/s]
 36%|███▌      | 23571/65536 [4:02:39<7:18:25,  1.60it/s]
 36%|███▌      | 23572/65536 [4:02:40<7:37:49,  1.53it/s]
 36%|███▌      | 23573/65536 [4:02:40<7:25:37,  1.57it/s]
 36%|███▌      | 23574/65536 [4:02:41<7:18:02,  1.60it/s]
 36%|███▌      | 23575/65536 [4:02:42<7:14:26,  1.61it/s]
 36%|███▌      | 23576/65536 [4:02:42<7:05:37,  1.64it/s]
 36%|███▌      | 23577/65536 [4:02:43<7:00:37,  1.66it/s]
 36%|███▌      | 23578/65536 [4:02:43<7:10:30,  1.62it/s]
 36%|███▌      | 23579/65536 [4:02:44<7:08:07,  1.63it/s]
 36%|███▌      | 23580/65536 [4:02:45<7:17:33,  1.60it/s]
                                                         
{'loss': 3.0546, 'learning_rate': 6.779417166646258e-07, 'epoch': 1455.56}
+
 36%|███▌      | 23580/65536 [4:02:45<7:17:33,  1.60it/s]
 36%|███▌      | 23581/65536 [4:02:45<7:06:53,  1.64it/s]
 36%|███▌      | 23582/65536 [4:02:46<7:08:10,  1.63it/s]
 36%|███▌      | 23583/65536 [4:02:46<7:02:02,  1.66it/s]
 36%|███▌      | 23584/65536 [4:02:47<7:00:34,  1.66it/s]
 36%|███▌      | 23585/65536 [4:02:48<7:09:59,  1.63it/s]
 36%|███▌      | 23586/65536 [4:02:48<7:21:55,  1.58it/s]
 36%|███▌      | 23587/65536 [4:02:49<7:15:39,  1.60it/s]
 36%|███▌      | 23588/65536 [4:02:50<7:28:37,  1.56it/s]
 36%|███▌      | 23589/65536 [4:02:50<7:24:01,  1.57it/s]
 36%|███▌      | 23590/65536 [4:02:51<7:11:24,  1.62it/s]
 36%|███▌      | 23591/65536 [4:02:51<7:04:59,  1.64it/s]
 36%|███▌      | 23592/65536 [4:02:52<7:06:20,  1.64it/s]
 36%|███▌      | 23593/65536 [4:02:53<6:56:04,  1.68it/s]
 36%|███▌      | 23594/65536 [4:02:53<6:55:15,  1.68it/s]
 36%|███▌      | 23595/65536 [4:02:54<7:04:20,  1.65it/s]
 36%|███▌      | 23596/65536 [4:02:54<6:55:50,  1.68it/s]
 36%|███▌      | 23597/65536 [4:02:55<6:53:13,  1.69it/s]
 36%|███▌      | 23598/65536 [4:02:56<7:15:51,  1.60it/s]
 36%|███▌      | 23599/65536 [4:02:56<7:13:29,  1.61it/s]
 36%|███▌      | 23600/65536 [4:02:57<7:04:06,  1.65it/s]
                                                         
{'loss': 2.9454, 'learning_rate': 6.776662177053997e-07, 'epoch': 1456.79}
+
 36%|███▌      | 23600/65536 [4:02:57<7:04:06,  1.65it/s]
 36%|███▌      | 23601/65536 [4:02:57<7:04:32,  1.65it/s]
 36%|███▌      | 23602/65536 [4:02:58<6:57:36,  1.67it/s]
 36%|███▌      | 23603/65536 [4:02:59<7:11:32,  1.62it/s]
 36%|███▌      | 23604/65536 [4:02:59<7:28:16,  1.56it/s]
 36%|███▌      | 23605/65536 [4:03:00<7:21:18,  1.58it/s]
 36%|███▌      | 23606/65536 [4:03:01<7:16:12,  1.60it/s]
 36%|███▌      | 23607/65536 [4:03:01<7:11:14,  1.62it/s]
 36%|███▌      | 23608/65536 [4:03:02<6:58:34,  1.67it/s]
 36%|███▌      | 23609/65536 [4:03:02<6:59:37,  1.67it/s]
 36%|███▌      | 23610/65536 [4:03:03<7:01:48,  1.66it/s]
 36%|███▌      | 23611/65536 [4:03:04<7:01:33,  1.66it/s]
 36%|███▌      | 23612/65536 [4:03:04<6:55:34,  1.68it/s]
 36%|███▌      | 23613/65536 [4:03:05<7:03:19,  1.65it/s]
 36%|███▌      | 23614/65536 [4:03:05<6:58:55,  1.67it/s]
 36%|███▌      | 23615/65536 [4:03:06<7:14:05,  1.61it/s]
 36%|███▌      | 23616/65536 [4:03:07<7:20:29,  1.59it/s]
 36%|███▌      | 23617/65536 [4:03:07<7:14:46,  1.61it/s]
 36%|███▌      | 23618/65536 [4:03:08<7:22:33,  1.58it/s]
 36%|███▌      | 23619/65536 [4:03:09<7:16:52,  1.60it/s]
 36%|███▌      | 23620/65536 [4:03:09<7:24:11,  1.57it/s]
                                                         
{'loss': 3.0118, 'learning_rate': 6.773907187461735e-07, 'epoch': 1458.02}
+
 36%|███▌      | 23620/65536 [4:03:09<7:24:11,  1.57it/s]
 36%|███▌      | 23621/65536 [4:03:10<7:12:29,  1.62it/s]
 36%|███▌      | 23622/65536 [4:03:10<7:04:04,  1.65it/s]
 36%|███▌      | 23623/65536 [4:03:11<6:56:57,  1.68it/s]
 36%|███▌      | 23624/65536 [4:03:12<6:57:08,  1.67it/s]
 36%|███▌      | 23625/65536 [4:03:12<7:03:26,  1.65it/s]
 36%|███▌      | 23626/65536 [4:03:13<7:03:23,  1.65it/s]
 36%|███▌      | 23627/65536 [4:03:13<7:09:24,  1.63it/s]
 36%|███▌      | 23628/65536 [4:03:14<7:02:25,  1.65it/s]
 36%|███▌      | 23629/65536 [4:03:15<7:03:57,  1.65it/s]
 36%|███▌      | 23630/65536 [4:03:15<7:04:51,  1.64it/s]
 36%|███▌      | 23631/65536 [4:03:16<7:14:23,  1.61it/s]
 36%|███▌      | 23632/65536 [4:03:17<7:10:56,  1.62it/s]
 36%|███▌      | 23633/65536 [4:03:17<7:07:26,  1.63it/s]
 36%|███▌      | 23634/65536 [4:03:18<7:07:12,  1.63it/s]
 36%|███▌      | 23635/65536 [4:03:18<7:13:55,  1.61it/s]
 36%|███▌      | 23636/65536 [4:03:19<7:24:55,  1.57it/s]
 36%|███▌      | 23637/65536 [4:03:20<7:29:35,  1.55it/s]
 36%|███▌      | 23638/65536 [4:03:20<7:27:40,  1.56it/s]
 36%|███▌      | 23639/65536 [4:03:21<7:18:38,  1.59it/s]
 36%|███▌      | 23640/65536 [4:03:22<7:16:34,  1.60it/s]
                                                         
{'loss': 3.0519, 'learning_rate': 6.771152197869475e-07, 'epoch': 1459.26}
+
 36%|███▌      | 23640/65536 [4:03:22<7:16:34,  1.60it/s]
 36%|███▌      | 23641/65536 [4:03:22<7:14:49,  1.61it/s]
 36%|███▌      | 23642/65536 [4:03:23<7:13:02,  1.61it/s]
 36%|███▌      | 23643/65536 [4:03:23<7:04:46,  1.64it/s]
 36%|███▌      | 23644/65536 [4:03:24<7:11:00,  1.62it/s]
 36%|███▌      | 23645/65536 [4:03:25<7:16:11,  1.60it/s]
 36%|███▌      | 23646/65536 [4:03:25<7:11:55,  1.62it/s]
 36%|███▌      | 23647/65536 [4:03:26<7:14:54,  1.61it/s]
 36%|███▌      | 23648/65536 [4:03:27<7:13:18,  1.61it/s]
 36%|███▌      | 23649/65536 [4:03:27<7:03:09,  1.65it/s]
 36%|███▌      | 23650/65536 [4:03:28<7:06:07,  1.64it/s]
 36%|███▌      | 23651/65536 [4:03:28<7:09:49,  1.62it/s]
 36%|███▌      | 23652/65536 [4:03:29<7:05:03,  1.64it/s]
 36%|███▌      | 23653/65536 [4:03:30<7:33:01,  1.54it/s]
 36%|███▌      | 23654/65536 [4:03:30<7:14:12,  1.61it/s]
 36%|███▌      | 23655/65536 [4:03:31<7:10:47,  1.62it/s]
 36%|███▌      | 23656/65536 [4:03:31<7:10:29,  1.62it/s]
 36%|███▌      | 23657/65536 [4:03:32<7:12:53,  1.61it/s]
 36%|███▌      | 23658/65536 [4:03:33<7:10:50,  1.62it/s]
 36%|███▌      | 23659/65536 [4:03:33<7:04:39,  1.64it/s]
 36%|███▌      | 23660/65536 [4:03:34<7:06:41,  1.64it/s]
                                                         
{'loss': 2.9304, 'learning_rate': 6.768397208277213e-07, 'epoch': 1460.49}
+
 36%|███▌      | 23660/65536 [4:03:34<7:06:41,  1.64it/s]
 36%|███▌      | 23661/65536 [4:03:34<7:04:10,  1.65it/s]
 36%|███▌      | 23662/65536 [4:03:35<7:04:33,  1.64it/s]
 36%|███▌      | 23663/65536 [4:03:36<7:10:56,  1.62it/s]
 36%|███▌      | 23664/65536 [4:03:36<7:15:30,  1.60it/s]
 36%|███▌      | 23665/65536 [4:03:37<7:04:59,  1.64it/s]
 36%|███▌      | 23666/65536 [4:03:38<7:13:35,  1.61it/s]
 36%|███▌      | 23667/65536 [4:03:38<7:03:51,  1.65it/s]
 36%|███▌      | 23668/65536 [4:03:39<6:52:44,  1.69it/s]
 36%|███▌      | 23669/65536 [4:03:39<7:01:02,  1.66it/s]
 36%|███▌      | 23670/65536 [4:03:40<7:06:56,  1.63it/s]
 36%|███▌      | 23671/65536 [4:03:41<7:04:03,  1.65it/s]
 36%|███▌      | 23672/65536 [4:03:41<7:00:22,  1.66it/s]
 36%|███▌      | 23673/65536 [4:03:42<7:01:46,  1.65it/s]
 36%|███▌      | 23674/65536 [4:03:42<7:02:16,  1.65it/s]
 36%|███▌      | 23675/65536 [4:03:43<6:59:43,  1.66it/s]
 36%|███▌      | 23676/65536 [4:03:44<7:00:00,  1.66it/s]
 36%|███▌      | 23677/65536 [4:03:44<6:59:19,  1.66it/s]
 36%|███▌      | 23678/65536 [4:03:45<6:58:13,  1.67it/s]
 36%|███▌      | 23679/65536 [4:03:45<7:02:52,  1.65it/s]
 36%|███▌      | 23680/65536 [4:03:46<7:24:59,  1.57it/s]
                                                         
{'loss': 2.9727, 'learning_rate': 6.765642218684952e-07, 'epoch': 1461.73}
+
 36%|███▌      | 23680/65536 [4:03:46<7:24:59,  1.57it/s]
 36%|███▌      | 23681/65536 [4:03:47<7:19:16,  1.59it/s]
 36%|███▌      | 23682/65536 [4:03:47<7:20:44,  1.58it/s]
 36%|███▌      | 23683/65536 [4:03:48<7:10:54,  1.62it/s]
 36%|███▌      | 23684/65536 [4:03:49<7:08:54,  1.63it/s]
 36%|███▌      | 23685/65536 [4:03:49<7:29:50,  1.55it/s]
 36%|███▌      | 23686/65536 [4:03:50<7:20:18,  1.58it/s]
 36%|███▌      | 23687/65536 [4:03:50<7:15:41,  1.60it/s]
 36%|███▌      | 23688/65536 [4:03:51<7:09:44,  1.62it/s]
 36%|███▌      | 23689/65536 [4:03:52<7:01:44,  1.65it/s]
 36%|███▌      | 23690/65536 [4:03:52<7:07:50,  1.63it/s]
 36%|███▌      | 23691/65536 [4:03:53<7:05:24,  1.64it/s]
 36%|███▌      | 23692/65536 [4:03:53<6:59:32,  1.66it/s]
 36%|███▌      | 23693/65536 [4:03:54<7:01:52,  1.65it/s]
 36%|███▌      | 23694/65536 [4:03:55<6:58:06,  1.67it/s]
 36%|███▌      | 23695/65536 [4:03:55<6:52:02,  1.69it/s]
 36%|███▌      | 23696/65536 [4:03:56<6:49:57,  1.70it/s]
 36%|███▌      | 23697/65536 [4:03:56<6:52:58,  1.69it/s]
 36%|███▌      | 23698/65536 [4:03:57<6:58:02,  1.67it/s]
 36%|███▌      | 23699/65536 [4:03:58<7:17:55,  1.59it/s]
 36%|███▌      | 23700/65536 [4:03:58<7:28:57,  1.55it/s]
                                                         
{'loss': 2.9836, 'learning_rate': 6.762887229092691e-07, 'epoch': 1462.96}
+
 36%|███▌      | 23700/65536 [4:03:58<7:28:57,  1.55it/s]
 36%|███▌      | 23701/65536 [4:03:59<7:36:36,  1.53it/s]
 36%|███▌      | 23702/65536 [4:04:00<7:25:15,  1.57it/s]
 36%|███▌      | 23703/65536 [4:04:00<7:20:59,  1.58it/s]
 36%|███▌      | 23704/65536 [4:04:01<7:18:10,  1.59it/s]
 36%|███▌      | 23705/65536 [4:04:02<7:17:43,  1.59it/s]
 36%|███▌      | 23706/65536 [4:04:02<7:12:43,  1.61it/s]
 36%|███▌      | 23707/65536 [4:04:03<7:06:46,  1.63it/s]
 36%|███▌      | 23708/65536 [4:04:03<7:14:48,  1.60it/s]
 36%|███▌      | 23709/65536 [4:04:04<7:15:17,  1.60it/s]
 36%|███▌      | 23710/65536 [4:04:05<7:16:43,  1.60it/s]
 36%|███▌      | 23711/65536 [4:04:05<7:08:53,  1.63it/s]
 36%|███▌      | 23712/65536 [4:04:06<6:59:09,  1.66it/s]
 36%|███▌      | 23713/65536 [4:04:06<6:54:46,  1.68it/s]
 36%|███▌      | 23714/65536 [4:04:07<6:59:21,  1.66it/s]
 36%|███▌      | 23715/65536 [4:04:08<7:00:32,  1.66it/s]
 36%|███▌      | 23716/65536 [4:04:08<7:03:01,  1.65it/s]
 36%|███▌      | 23717/65536 [4:04:09<7:18:11,  1.59it/s]
 36%|███▌      | 23718/65536 [4:04:10<7:17:54,  1.59it/s]
 36%|███▌      | 23719/65536 [4:04:10<7:12:08,  1.61it/s]
 36%|██��▌      | 23720/65536 [4:04:11<7:09:52,  1.62it/s]
                                                         
{'loss': 3.0426, 'learning_rate': 6.760132239500429e-07, 'epoch': 1464.2}
+
 36%|███▌      | 23720/65536 [4:04:11<7:09:52,  1.62it/s]
 36%|███▌      | 23721/65536 [4:04:11<7:06:34,  1.63it/s]
 36%|███▌      | 23722/65536 [4:04:12<7:08:44,  1.63it/s]
 36%|███▌      | 23723/65536 [4:04:13<7:16:18,  1.60it/s]
 36%|███▌      | 23724/65536 [4:04:13<7:24:59,  1.57it/s]
 36%|███▌      | 23725/65536 [4:04:14<7:16:29,  1.60it/s]
 36%|███▌      | 23726/65536 [4:04:15<7:17:31,  1.59it/s]
 36%|███▌      | 23727/65536 [4:04:15<7:11:16,  1.62it/s]
 36%|███▌      | 23728/65536 [4:04:16<7:02:01,  1.65it/s]
 36%|███▌      | 23729/65536 [4:04:16<6:57:55,  1.67it/s]
 36%|███▌      | 23730/65536 [4:04:17<6:56:17,  1.67it/s]
 36%|███▌      | 23731/65536 [4:04:18<7:00:38,  1.66it/s]
 36%|███▌      | 23732/65536 [4:04:18<6:57:13,  1.67it/s]
 36%|███▌      | 23733/65536 [4:04:19<7:00:55,  1.66it/s]
 36%|███▌      | 23734/65536 [4:04:19<7:16:34,  1.60it/s]
 36%|███▌      | 23735/65536 [4:04:20<7:07:44,  1.63it/s]
 36%|███▌      | 23736/65536 [4:04:21<6:57:56,  1.67it/s]
 36%|███▌      | 23737/65536 [4:04:21<6:51:24,  1.69it/s]
 36%|███▌      | 23738/65536 [4:04:22<6:57:38,  1.67it/s]
 36%|███▌      | 23739/65536 [4:04:22<7:02:34,  1.65it/s]
 36%|███▌      | 23740/65536 [4:04:23<6:59:00,  1.66it/s]
                                                         
{'loss': 3.0034, 'learning_rate': 6.757377249908168e-07, 'epoch': 1465.43}
+
 36%|███▌      | 23740/65536 [4:04:23<6:59:00,  1.66it/s]
 36%|███▌      | 23741/65536 [4:04:24<7:05:25,  1.64it/s]
 36%|███▌      | 23742/65536 [4:04:24<7:06:04,  1.63it/s]
 36%|███▌      | 23743/65536 [4:04:25<6:59:45,  1.66it/s]
 36%|███▌      | 23744/65536 [4:04:25<7:06:54,  1.63it/s]
 36%|███▌      | 23745/65536 [4:04:26<7:15:29,  1.60it/s]
 36%|███▌      | 23746/65536 [4:04:27<7:09:42,  1.62it/s]
 36%|███▌      | 23747/65536 [4:04:27<7:09:51,  1.62it/s]
 36%|███▌      | 23748/65536 [4:04:28<7:21:27,  1.58it/s]
 36%|███▌      | 23749/65536 [4:04:29<7:08:39,  1.62it/s]
 36%|███▌      | 23750/65536 [4:04:29<7:19:14,  1.59it/s]
 36%|███▌      | 23751/65536 [4:04:30<7:13:31,  1.61it/s]
 36%|███▌      | 23752/65536 [4:04:30<7:17:38,  1.59it/s]
 36%|███▌      | 23753/65536 [4:04:31<7:11:41,  1.61it/s]
 36%|███▌      | 23754/65536 [4:04:32<7:11:18,  1.61it/s]
 36%|███▌      | 23755/65536 [4:04:32<7:07:21,  1.63it/s]
 36%|███▌      | 23756/65536 [4:04:33<7:15:37,  1.60it/s]
 36%|███▋      | 23757/65536 [4:04:34<7:24:49,  1.57it/s]
 36%|███▋      | 23758/65536 [4:04:34<7:13:03,  1.61it/s]
 36%|███▋      | 23759/65536 [4:04:35<7:01:08,  1.65it/s]
 36%|███▋      | 23760/65536 [4:04:35<7:06:19,  1.63it/s]
                                                         
{'loss': 2.9911, 'learning_rate': 6.754622260315905e-07, 'epoch': 1466.67}
+
 36%|███▋      | 23760/65536 [4:04:35<7:06:19,  1.63it/s]
 36%|███▋      | 23761/65536 [4:04:36<7:17:25,  1.59it/s]
 36%|███▋      | 23762/65536 [4:04:37<7:05:49,  1.63it/s]
 36%|███▋      | 23763/65536 [4:04:37<6:58:08,  1.67it/s]
 36%|███▋      | 23764/65536 [4:04:38<7:01:46,  1.65it/s]
 36%|███▋      | 23765/65536 [4:04:38<6:55:24,  1.68it/s]
 36%|███▋      | 23766/65536 [4:04:39<7:08:50,  1.62it/s]
 36%|███▋      | 23767/65536 [4:04:40<7:12:20,  1.61it/s]
 36%|███▋      | 23768/65536 [4:04:40<6:58:11,  1.66it/s]
 36%|███▋      | 23769/65536 [4:04:41<6:57:40,  1.67it/s]
 36%|███▋      | 23770/65536 [4:04:41<7:06:37,  1.63it/s]
 36%|███▋      | 23771/65536 [4:04:42<7:04:10,  1.64it/s]
 36%|███▋      | 23772/65536 [4:04:43<7:09:00,  1.62it/s]
 36%|███▋      | 23773/65536 [4:04:43<7:04:00,  1.64it/s]
 36%|███▋      | 23774/65536 [4:04:44<7:05:29,  1.64it/s]
 36%|███▋      | 23775/65536 [4:04:45<7:19:20,  1.58it/s]
 36%|███▋      | 23776/65536 [4:04:45<7:02:14,  1.65it/s]
 36%|███▋      | 23777/65536 [4:04:46<7:04:43,  1.64it/s]
 36%|███▋      | 23778/65536 [4:04:46<7:00:32,  1.65it/s]
 36%|███▋      | 23779/65536 [4:04:47<6:58:09,  1.66it/s]
 36%|███▋      | 23780/65536 [4:04:48<6:58:57,  1.66it/s]
                                                         
{'loss': 2.9945, 'learning_rate': 6.751867270723644e-07, 'epoch': 1467.9}
+
 36%|███▋      | 23780/65536 [4:04:48<6:58:57,  1.66it/s]
 36%|███▋      | 23781/65536 [4:04:48<7:12:12,  1.61it/s]
 36%|███▋      | 23782/65536 [4:04:49<7:23:17,  1.57it/s]
 36%|███▋      | 23783/65536 [4:04:50<7:28:16,  1.55it/s]
 36%|███▋      | 23784/65536 [4:04:50<7:12:52,  1.61it/s]
 36%|███▋      | 23785/65536 [4:04:51<7:08:25,  1.62it/s]
 36%|███▋      | 23786/65536 [4:04:51<7:08:45,  1.62it/s]
 36%|███▋      | 23787/65536 [4:04:52<7:03:51,  1.64it/s]
 36%|███▋      | 23788/65536 [4:04:53<7:02:54,  1.65it/s]
 36%|███▋      | 23789/65536 [4:04:53<7:12:31,  1.61it/s]
 36%|███▋      | 23790/65536 [4:04:54<7:06:52,  1.63it/s]
 36%|███▋      | 23791/65536 [4:04:54<6:58:54,  1.66it/s]
 36%|███▋      | 23792/65536 [4:04:55<7:00:04,  1.66it/s]
 36%|███▋      | 23793/65536 [4:04:56<7:12:52,  1.61it/s]
 36%|███▋      | 23794/65536 [4:04:56<7:05:21,  1.64it/s]
 36%|███▋      | 23795/65536 [4:04:57<6:58:49,  1.66it/s]
 36%|███▋      | 23796/65536 [4:04:57<6:49:57,  1.70it/s]
 36%|███▋      | 23797/65536 [4:04:58<7:04:00,  1.64it/s]
 36%|███▋      | 23798/65536 [4:04:59<7:27:46,  1.55it/s]
 36%|███▋      | 23799/65536 [4:04:59<7:25:49,  1.56it/s]
 36%|███▋      | 23800/65536 [4:05:00<7:20:46,  1.58it/s]
                                                         
{'loss': 2.9749, 'learning_rate': 6.749112281131383e-07, 'epoch': 1469.14}
+
 36%|███▋      | 23800/65536 [4:05:00<7:20:46,  1.58it/s]
 36%|███▋      | 23801/65536 [4:05:01<7:22:32,  1.57it/s]
 36%|███▋      | 23802/65536 [4:05:01<7:13:40,  1.60it/s]
 36%|███▋      | 23803/65536 [4:05:02<7:00:43,  1.65it/s]
 36%|███▋      | 23804/65536 [4:05:02<7:19:14,  1.58it/s]
 36%|███▋      | 23805/65536 [4:05:03<7:18:39,  1.59it/s]
 36%|███▋      | 23806/65536 [4:05:04<7:12:04,  1.61it/s]
 36%|███▋      | 23807/65536 [4:05:04<7:07:47,  1.63it/s]
 36%|███▋      | 23808/65536 [4:05:05<7:04:54,  1.64it/s]
 36%|███▋      | 23809/65536 [4:05:06<7:06:49,  1.63it/s]
 36%|███▋      | 23810/65536 [4:05:06<7:02:29,  1.65it/s]
 36%|███▋      | 23811/65536 [4:05:07<6:57:31,  1.67it/s]
 36%|███▋      | 23812/65536 [4:05:07<6:58:19,  1.66it/s]
 36%|███▋      | 23813/65536 [4:05:08<6:55:03,  1.68it/s]
 36%|███▋      | 23814/65536 [4:05:08<6:49:26,  1.70it/s]
 36%|███▋      | 23815/65536 [4:05:09<7:08:03,  1.62it/s]
 36%|███▋      | 23816/65536 [4:05:10<7:01:32,  1.65it/s]
 36%|███▋      | 23817/65536 [4:05:10<6:58:32,  1.66it/s]
 36%|███▋      | 23818/65536 [4:05:11<7:14:45,  1.60it/s]
 36%|███▋      | 23819/65536 [4:05:12<7:13:40,  1.60it/s]
 36%|███▋      | 23820/65536 [4:05:12<7:06:19,  1.63it/s]
                                                         
{'loss': 2.9992, 'learning_rate': 6.74635729153912e-07, 'epoch': 1470.37}
+
 36%|███▋      | 23820/65536 [4:05:12<7:06:19,  1.63it/s]
 36%|███▋      | 23821/65536 [4:05:13<7:11:05,  1.61it/s]
 36%|███▋      | 23822/65536 [4:05:13<7:02:07,  1.65it/s]
 36%|███▋      | 23823/65536 [4:05:14<6:59:12,  1.66it/s]
 36%|███▋      | 23824/65536 [4:05:15<6:53:32,  1.68it/s]
 36%|███▋      | 23825/65536 [4:05:15<6:44:30,  1.72it/s]
 36%|███▋      | 23826/65536 [4:05:16<6:58:43,  1.66it/s]
 36%|███▋      | 23827/65536 [4:05:16<6:53:24,  1.68it/s]
 36%|███▋      | 23828/65536 [4:05:17<7:09:55,  1.62it/s]
 36%|███▋      | 23829/65536 [4:05:18<6:59:39,  1.66it/s]
 36%|███▋      | 23830/65536 [4:05:18<7:03:01,  1.64it/s]
 36%|███▋      | 23831/65536 [4:05:19<7:16:47,  1.59it/s]
 36%|███▋      | 23832/65536 [4:05:19<7:04:43,  1.64it/s]
 36%|███▋      | 23833/65536 [4:05:20<7:01:06,  1.65it/s]
 36%|███▋      | 23834/65536 [4:05:21<7:01:52,  1.65it/s]
 36%|███▋      | 23835/65536 [4:05:21<6:55:37,  1.67it/s]
 36%|███▋      | 23836/65536 [4:05:22<6:53:47,  1.68it/s]
 36%|███▋      | 23837/65536 [4:05:23<7:18:47,  1.58it/s]
 36%|███▋      | 23838/65536 [4:05:23<7:10:00,  1.62it/s]
 36%|███▋      | 23839/65536 [4:05:24<7:01:41,  1.65it/s]
 36%|███▋      | 23840/65536 [4:05:24<7:05:16,  1.63it/s]
                                                         
{'loss': 2.9742, 'learning_rate': 6.743602301946859e-07, 'epoch': 1471.6}
+
 36%|███▋      | 23840/65536 [4:05:24<7:05:16,  1.63it/s]
 36%|███▋      | 23841/65536 [4:05:25<6:56:27,  1.67it/s]
 36%|███▋      | 23842/65536 [4:05:26<7:11:06,  1.61it/s]
 36%|███▋      | 23843/65536 [4:05:26<7:10:22,  1.61it/s]
 36%|███▋      | 23844/65536 [4:05:27<7:04:42,  1.64it/s]
 36%|███▋      | 23845/65536 [4:05:27<7:04:41,  1.64it/s]
 36%|███▋      | 23846/65536 [4:05:28<7:07:54,  1.62it/s]
 36%|███▋      | 23847/65536 [4:05:29<7:43:21,  1.50it/s]
 36%|███▋      | 23848/65536 [4:05:29<7:27:06,  1.55it/s]
 36%|███▋      | 23849/65536 [4:05:30<7:10:35,  1.61it/s]
 36%|███▋      | 23850/65536 [4:05:31<7:01:49,  1.65it/s]
 36%|███▋      | 23851/65536 [4:05:31<7:02:58,  1.64it/s]
 36%|███▋      | 23852/65536 [4:05:32<7:07:47,  1.62it/s]
 36%|███▋      | 23853/65536 [4:05:32<6:56:14,  1.67it/s]
 36%|���██▋      | 23854/65536 [4:05:33<7:01:17,  1.65it/s]
 36%|███▋      | 23855/65536 [4:05:34<6:57:51,  1.66it/s]
 36%|███▋      | 23856/65536 [4:05:34<6:47:36,  1.70it/s]
 36%|███▋      | 23857/65536 [4:05:35<6:57:35,  1.66it/s]
 36%|███▋      | 23858/65536 [4:05:35<7:03:59,  1.64it/s]
 36%|███▋      | 23859/65536 [4:05:36<7:13:47,  1.60it/s]
 36%|███▋      | 23860/65536 [4:05:37<7:06:49,  1.63it/s]
                                                         
{'loss': 3.0011, 'learning_rate': 6.740847312354597e-07, 'epoch': 1472.84}
+
 36%|███▋      | 23860/65536 [4:05:37<7:06:49,  1.63it/s]
 36%|███▋      | 23861/65536 [4:05:37<7:08:14,  1.62it/s]
 36%|███▋      | 23862/65536 [4:05:38<7:03:22,  1.64it/s]
 36%|███▋      | 23863/65536 [4:05:39<7:17:03,  1.59it/s]
 36%|███▋      | 23864/65536 [4:05:39<7:22:09,  1.57it/s]
 36%|███▋      | 23865/65536 [4:05:40<7:27:10,  1.55it/s]
 36%|███▋      | 23866/65536 [4:05:40<7:14:16,  1.60it/s]
 36%|███▋      | 23867/65536 [4:05:41<7:08:11,  1.62it/s]
 36%|███▋      | 23868/65536 [4:05:42<6:59:37,  1.65it/s]
 36%|███▋      | 23869/65536 [4:05:42<6:50:58,  1.69it/s]
 36%|███▋      | 23870/65536 [4:05:43<7:04:08,  1.64it/s]
 36%|███▋      | 23871/65536 [4:05:43<7:01:36,  1.65it/s]
 36%|███▋      | 23872/65536 [4:05:44<6:53:19,  1.68it/s]
 36%|███▋      | 23873/65536 [4:05:45<6:50:15,  1.69it/s]
 36%|███▋      | 23874/65536 [4:05:45<6:58:42,  1.66it/s]
 36%|███▋      | 23875/65536 [4:05:46<7:03:44,  1.64it/s]
 36%|███▋      | 23876/65536 [4:05:46<6:59:45,  1.65it/s]
 36%|███▋      | 23877/65536 [4:05:47<7:04:21,  1.64it/s]
 36%|███▋      | 23878/65536 [4:05:48<7:03:10,  1.64it/s]
 36%|███▋      | 23879/65536 [4:05:48<7:28:25,  1.55it/s]
 36%|███▋      | 23880/65536 [4:05:49<7:23:54,  1.56it/s]
                                                         
{'loss': 2.9287, 'learning_rate': 6.738092322762336e-07, 'epoch': 1474.07}
+
 36%|███▋      | 23880/65536 [4:05:49<7:23:54,  1.56it/s]
 36%|███▋      | 23881/65536 [4:05:50<7:11:34,  1.61it/s]
 36%|███▋      | 23882/65536 [4:05:50<7:06:07,  1.63it/s]
 36%|███▋      | 23883/65536 [4:05:51<7:10:06,  1.61it/s]
 36%|███▋      | 23884/65536 [4:05:51<7:05:27,  1.63it/s]
 36%|███▋      | 23885/65536 [4:05:52<6:56:58,  1.66it/s]
 36%|███▋      | 23886/65536 [4:05:53<7:08:09,  1.62it/s]
 36%|███▋      | 23887/65536 [4:05:53<7:03:27,  1.64it/s]
 36%|███▋      | 23888/65536 [4:05:54<7:04:10,  1.64it/s]
 36%|███▋      | 23889/65536 [4:05:54<7:02:14,  1.64it/s]
 36%|███▋      | 23890/65536 [4:05:55<7:04:06,  1.64it/s]
 36%|███▋      | 23891/65536 [4:05:56<7:25:05,  1.56it/s]
 36%|███▋      | 23892/65536 [4:05:56<7:14:13,  1.60it/s]
 36%|███▋      | 23893/65536 [4:05:57<7:14:52,  1.60it/s]
 36%|███▋      | 23894/65536 [4:05:58<7:07:07,  1.62it/s]
 36%|███▋      | 23895/65536 [4:05:58<7:16:27,  1.59it/s]
 36%|███▋      | 23896/65536 [4:05:59<7:37:00,  1.52it/s]
 36%|███▋      | 23897/65536 [4:06:00<7:26:23,  1.55it/s]
 36%|███▋      | 23898/65536 [4:06:00<7:22:06,  1.57it/s]
 36%|███▋      | 23899/65536 [4:06:01<7:10:57,  1.61it/s]
 36%|███▋      | 23900/65536 [4:06:01<6:59:42,  1.65it/s]
                                                         
{'loss': 2.9665, 'learning_rate': 6.735337333170074e-07, 'epoch': 1475.31}
+
 36%|███▋      | 23900/65536 [4:06:01<6:59:42,  1.65it/s]
 36%|███▋      | 23901/65536 [4:06:02<7:03:40,  1.64it/s]
 36%|███▋      | 23902/65536 [4:06:03<7:00:23,  1.65it/s]
 36%|███▋      | 23903/65536 [4:06:03<6:53:26,  1.68it/s]
 36%|███▋      | 23904/65536 [4:06:04<6:54:12,  1.68it/s]
 36%|███▋      | 23905/65536 [4:06:04<7:00:06,  1.65it/s]
 36%|███▋      | 23906/65536 [4:06:05<7:02:00,  1.64it/s]
 36%|███▋      | 23907/65536 [4:06:06<7:03:08,  1.64it/s]
 36%|███▋      | 23908/65536 [4:06:06<6:56:25,  1.67it/s]
 36%|███▋      | 23909/65536 [4:06:07<6:59:55,  1.65it/s]
 36%|███▋      | 23910/65536 [4:06:07<7:05:45,  1.63it/s]
 36%|███▋      | 23911/65536 [4:06:08<6:55:12,  1.67it/s]
 36%|███▋      | 23912/65536 [4:06:09<7:11:24,  1.61it/s]
 36%|███▋      | 23913/65536 [4:06:09<7:05:57,  1.63it/s]
 36%|███▋      | 23914/65536 [4:06:10<7:01:16,  1.65it/s]
 36%|███▋      | 23915/65536 [4:06:10<7:00:13,  1.65it/s]
 36%|███▋      | 23916/65536 [4:06:11<7:15:25,  1.59it/s]
 36%|███▋      | 23917/65536 [4:06:12<7:03:29,  1.64it/s]
 36%|███▋      | 23918/65536 [4:06:12<6:56:07,  1.67it/s]
 36%|███▋      | 23919/65536 [4:06:13<6:54:31,  1.67it/s]
 36%|███▋      | 23920/65536 [4:06:13<6:56:15,  1.67it/s]
                                                         
{'loss': 2.973, 'learning_rate': 6.732582343577813e-07, 'epoch': 1476.54}
+
 36%|███▋      | 23920/65536 [4:06:13<6:56:15,  1.67it/s]
 37%|███▋      | 23921/65536 [4:06:14<7:00:35,  1.65it/s]
 37%|███▋      | 23922/65536 [4:06:15<7:03:29,  1.64it/s]
 37%|███▋      | 23923/65536 [4:06:15<7:13:32,  1.60it/s]
 37%|███▋      | 23924/65536 [4:06:16<7:05:56,  1.63it/s]
 37%|███▋      | 23925/65536 [4:06:17<6:57:53,  1.66it/s]
 37%|███▋      | 23926/65536 [4:06:17<7:10:45,  1.61it/s]
 37%|███▋      | 23927/65536 [4:06:18<7:01:10,  1.65it/s]
 37%|███▋      | 23928/65536 [4:06:18<7:22:08,  1.57it/s]
 37%|███▋      | 23929/65536 [4:06:19<7:14:32,  1.60it/s]
 37%|███▋      | 23930/65536 [4:06:20<7:12:09,  1.60it/s]
 37%|███▋      | 23931/65536 [4:06:20<7:08:57,  1.62it/s]
 37%|███▋      | 23932/65536 [4:06:21<7:03:04,  1.64it/s]
 37%|███▋      | 23933/65536 [4:06:22<7:05:56,  1.63it/s]
 37%|███▋      | 23934/65536 [4:06:22<6:57:09,  1.66it/s]
 37%|███▋      | 23935/65536 [4:06:23<6:54:50,  1.67it/s]
 37%|███▋      | 23936/65536 [4:06:23<7:00:10,  1.65it/s]
 37%|███▋      | 23937/65536 [4:06:24<7:10:10,  1.61it/s]
 37%|███▋      | 23938/65536 [4:06:25<7:11:06,  1.61it/s]
 37%|███▋      | 23939/65536 [4:06:25<6:59:10,  1.65it/s]
 37%|███▋      | 23940/65536 [4:06:26<6:58:22,  1.66it/s]
                                                         
{'loss': 2.9416, 'learning_rate': 6.729827353985553e-07, 'epoch': 1477.78}
+
 37%|███▋      | 23940/65536 [4:06:26<6:58:22,  1.66it/s]
 37%|███▋      | 23941/65536 [4:06:26<7:16:20,  1.59it/s]
 37%|███▋      | 23942/65536 [4:06:27<7:13:13,  1.60it/s]
 37%|███▋      | 23943/65536 [4:06:28<7:30:57,  1.54it/s]
 37%|███▋      | 23944/65536 [4:06:28<7:50:34,  1.47it/s]
 37%|███▋      | 23945/65536 [4:06:29<7:44:12,  1.49it/s]
 37%|███▋      | 23946/65536 [4:06:30<7:27:25,  1.55it/s]
 37%|███▋      | 23947/65536 [4:06:30<7:23:23,  1.56it/s]
 37%|███▋      | 23948/65536 [4:06:31<7:23:06,  1.56it/s]
 37%|███▋      | 23949/65536 [4:06:32<7:10:25,  1.61it/s]
 37%|███▋      | 23950/65536 [4:06:32<7:09:11,  1.61it/s]
 37%|███▋      | 23951/65536 [4:06:33<7:15:30,  1.59it/s]
 37%|███▋      | 23952/65536 [4:06:33<6:54:28,  1.67it/s]
 37%|███▋      | 23953/65536 [4:06:34<6:45:56,  1.71it/s]
 37%|███▋      | 23954/65536 [4:06:34<6:42:18,  1.72it/s]
 37%|███▋      | 23955/65536 [4:06:35<6:48:16,  1.70it/s]
 37%|███▋      | 23956/65536 [4:06:36<6:48:14,  1.70it/s]
 37%|███▋      | 23957/65536 [4:06:36<6:50:54,  1.69it/s]
 37%|███▋      | 23958/65536 [4:06:37<6:46:05,  1.71it/s]
 37%|███▋      | 23959/65536 [4:06:38<7:01:25,  1.64it/s]
 37%|███▋      | 23960/65536 [4:06:38<7:15:30,  1.59it/s]
                                                         
{'loss': 3.0906, 'learning_rate': 6.72707236439329e-07, 'epoch': 1479.01}
+
 37%|███▋      | 23960/65536 [4:06:38<7:15:30,  1.59it/s]
 37%|███▋      | 23961/65536 [4:06:39<7:11:58,  1.60it/s]
 37%|███▋      | 23962/65536 [4:06:39<6:59:26,  1.65it/s]
 37%|███▋      | 23963/65536 [4:06:40<7:03:55,  1.63it/s]
 37%|███▋      | 23964/65536 [4:06:41<7:01:49,  1.64it/s]
 37%|███▋      | 23965/65536 [4:06:41<7:08:29,  1.62it/s]
 37%|███▋      | 23966/65536 [4:06:42<7:15:20,  1.59it/s]
 37%|███▋      | 23967/65536 [4:06:42<7:07:25,  1.62it/s]
 37%|███▋      | 23968/65536 [4:06:43<6:59:10,  1.65it/s]
 37%|███▋      | 23969/65536 [4:06:44<7:00:43,  1.65it/s]
 37%|███▋      | 23970/65536 [4:06:44<7:02:14,  1.64it/s]
 37%|███▋      | 23971/65536 [4:06:45<6:59:00,  1.65it/s]
 37%|███▋      | 23972/65536 [4:06:46<7:04:09,  1.63it/s]
 37%|███▋      | 23973/65536 [4:06:46<6:58:04,  1.66it/s]
 37%|███▋      | 23974/65536 [4:06:47<7:06:45,  1.62it/s]
 37%|███▋      | 23975/65536 [4:06:47<7:07:56,  1.62it/s]
 37%|███▋      | 23976/65536 [4:06:48<7:07:14,  1.62it/s]
 37%|███▋      | 23977/65536 [4:06:49<7:17:34,  1.58it/s]
 37%|███▋      | 23978/65536 [4:06:49<7:10:26,  1.61it/s]
 37%|███▋      | 23979/65536 [4:06:50<7:16:53,  1.59it/s]
 37%|███▋      | 23980/65536 [4:06:50<7:08:09,  1.62it/s]
                                                         
{'loss': 2.9303, 'learning_rate': 6.724317374801029e-07, 'epoch': 1480.25}
+
 37%|███▋      | 23980/65536 [4:06:50<7:08:09,  1.62it/s]
 37%|███▋      | 23981/65536 [4:06:51<7:00:47,  1.65it/s]
 37%|███▋      | 23982/65536 [4:06:52<7:00:23,  1.65it/s]
 37%|███▋      | 23983/65536 [4:06:52<7:02:20,  1.64it/s]
 37%|███▋      | 23984/65536 [4:06:53<7:02:11,  1.64it/s]
 37%|███▋      | 23985/65536 [4:06:54<7:18:29,  1.58it/s]
 37%|███▋      | 23986/65536 [4:06:54<7:04:08,  1.63it/s]
 37%|███▋      | 23987/65536 [4:06:55<7:10:29,  1.61it/s]
 37%|███▋      | 23988/65536 [4:06:55<7:09:28,  1.61it/s]
 37%|███▋      | 23989/65536 [4:06:56<7:08:03,  1.62it/s]
 37%|███▋      | 23990/65536 [4:06:57<7:01:58,  1.64it/s]
 37%|███▋      | 23991/65536 [4:06:57<7:00:01,  1.65it/s]
 37%|███▋      | 23992/65536 [4:06:58<6:51:47,  1.68it/s]
 37%|███▋      | 23993/65536 [4:06:58<7:12:20,  1.60it/s]
 37%|███▋      | 23994/65536 [4:06:59<7:31:00,  1.54it/s]
 37%|███▋      | 23995/65536 [4:07:00<7:36:42,  1.52it/s]
 37%|███▋      | 23996/65536 [4:07:00<7:23:30,  1.56it/s]
 37%|███▋      | 23997/65536 [4:07:01<7:19:01,  1.58it/s]
 37%|███▋      | 23998/65536 [4:07:02<7:17:28,  1.58it/s]
 37%|███▋      | 23999/65536 [4:07:02<7:07:56,  1.62it/s]
 37%|███▋      | 24000/65536 [4:07:03<6:57:09,  1.66it/s]
                                                         
{'loss': 2.9875, 'learning_rate': 6.721562385208767e-07, 'epoch': 1481.48}
+
 37%|███▋      | 24000/65536 [4:07:03<6:57:09,  1.66it/s]
 37%|███▋      | 24001/65536 [4:07:03<6:51:15,  1.68it/s]
 37%|███▋      | 24002/65536 [4:07:04<6:50:17,  1.69it/s]
 37%|███▋      | 24003/65536 [4:07:05<6:48:53,  1.69it/s]
 37%|███▋      | 24004/65536 [4:07:05<6:49:57,  1.69it/s]
 37%|███▋      | 24005/65536 [4:07:06<6:55:32,  1.67it/s]
 37%|███▋      | 24006/65536 [4:07:06<6:59:23,  1.65it/s]
 37%|███▋      | 24007/65536 [4:07:07<6:55:30,  1.67it/s]
 37%|███▋      | 24008/65536 [4:07:08<7:10:05,  1.61it/s]
 37%|███▋      | 24009/65536 [4:07:08<7:15:04,  1.59it/s]
 37%|███▋      | 24010/65536 [4:07:09<7:13:10,  1.60it/s]
 37%|███▋      | 24011/65536 [4:07:10<7:09:35,  1.61it/s]
 37%|███▋      | 24012/65536 [4:07:10<7:02:41,  1.64it/s]
 37%|███▋      | 24013/65536 [4:07:11<7:03:25,  1.63it/s]
 37%|███▋      | 24014/65536 [4:07:11<7:07:41,  1.62it/s]
 37%|███▋      | 24015/65536 [4:07:12<7:07:46,  1.62it/s]
 37%|███▋      | 24016/65536 [4:07:13<7:12:38,  1.60it/s]
 37%|███▋      | 24017/65536 [4:07:13<7:05:38,  1.63it/s]
 37%|███▋      | 24018/65536 [4:07:14<7:03:20,  1.63it/s]
 37%|███▋      | 24019/65536 [4:07:14<7:03:13,  1.63it/s]
 37%|███▋      | 24020/65536 [4:07:15<6:53:36,  1.67it/s]
                                                         
{'loss': 2.9577, 'learning_rate': 6.718807395616506e-07, 'epoch': 1482.72}
+
 37%|███▋      | 24020/65536 [4:07:15<6:53:36,  1.67it/s]
 37%|███▋      | 24021/65536 [4:07:16<6:56:09,  1.66it/s]
 37%|███▋      | 24022/65536 [4:07:16<6:50:24,  1.69it/s]
 37%|███▋      | 24023/65536 [4:07:17<6:47:47,  1.70it/s]
 37%|███▋      | 24024/65536 [4:07:17<7:01:17,  1.64it/s]
 37%|███▋      | 24025/65536 [4:07:18<7:19:14,  1.58it/s]
 37%|███▋      | 24026/65536 [4:07:19<7:02:04,  1.64it/s]
 37%|███▋      | 24027/65536 [4:07:19<6:55:50,  1.66it/s]
 37%|███▋      | 24028/65536 [4:07:20<7:00:57,  1.64it/s]
 37%|███▋      | 24029/65536 [4:07:21<7:11:51,  1.60it/s]
 37%|███▋      | 24030/65536 [4:07:21<7:12:03,  1.60it/s]
 37%|███▋      | 24031/65536 [4:07:22<7:01:31,  1.64it/s]
 37%|███▋      | 24032/65536 [4:07:22<7:04:41,  1.63it/s]
 37%|███▋      | 24033/65536 [4:07:23<6:57:17,  1.66it/s]
 37%|███▋      | 24034/65536 [4:07:24<6:54:38,  1.67it/s]
 37%|███▋      | 24035/65536 [4:07:24<7:10:58,  1.60it/s]
 37%|███▋      | 24036/65536 [4:07:25<7:00:51,  1.64it/s]
 37%|███▋      | 24037/65536 [4:07:25<6:54:37,  1.67it/s]
 37%|███▋      | 24038/65536 [4:07:26<6:50:55,  1.68it/s]
 37%|███▋      | 24039/65536 [4:07:27<7:08:14,  1.61it/s]
 37%|███▋      | 24040/65536 [4:07:27<6:57:03,  1.66it/s]
                                                         
{'loss': 3.0099, 'learning_rate': 6.716052406024243e-07, 'epoch': 1483.95}
+
 37%|███▋      | 24040/65536 [4:07:27<6:57:03,  1.66it/s]
 37%|███▋      | 24041/65536 [4:07:28<7:12:39,  1.60it/s]
 37%|███▋      | 24042/65536 [4:07:28<7:03:45,  1.63it/s]
 37%|███▋      | 24043/65536 [4:07:29<6:56:20,  1.66it/s]
 37%|███▋      | 24044/65536 [4:07:30<7:01:52,  1.64it/s]
 37%|███▋      | 24045/65536 [4:07:30<7:06:25,  1.62it/s]
 37%|███▋      | 24046/65536 [4:07:31<7:07:28,  1.62it/s]
 37%|███▋      | 24047/65536 [4:07:32<7:10:17,  1.61it/s]
 37%|███▋      | 24048/65536 [4:07:32<7:09:42,  1.61it/s]
 37%|███▋      | 24049/65536 [4:07:33<7:07:47,  1.62it/s]
 37%|███▋      | 24050/65536 [4:07:33<6:58:50,  1.65it/s]
 37%|███▋      | 24051/65536 [4:07:34<7:16:28,  1.58it/s]
 37%|███▋      | 24052/65536 [4:07:35<7:10:47,  1.60it/s]
 37%|███▋      | 24053/65536 [4:07:35<7:01:23,  1.64it/s]
 37%|███▋      | 24054/65536 [4:07:36<7:03:34,  1.63it/s]
 37%|███▋      | 24055/65536 [4:07:36<6:53:34,  1.67it/s]
 37%|███▋      | 24056/65536 [4:07:37<7:07:59,  1.62it/s]
 37%|███▋      | 24057/65536 [4:07:38<7:16:07,  1.59it/s]
 37%|███▋      | 24058/65536 [4:07:38<7:35:11,  1.52it/s]
 37%|███▋      | 24059/65536 [4:07:39<7:27:00,  1.55it/s]
 37%|███▋      | 24060/65536 [4:07:40<7:10:19,  1.61it/s]
                                                         
{'loss': 2.9635, 'learning_rate': 6.713297416431982e-07, 'epoch': 1485.19}
+
 37%|███▋      | 24060/65536 [4:07:40<7:10:19,  1.61it/s]
 37%|███▋      | 24061/65536 [4:07:40<7:10:17,  1.61it/s]
 37%|███▋      | 24062/65536 [4:07:41<7:17:34,  1.58it/s]
 37%|███▋      | 24063/65536 [4:07:42<7:10:24,  1.61it/s]
 37%|███▋      | 24064/65536 [4:07:42<7:04:11,  1.63it/s]
 37%|███▋      | 24065/65536 [4:07:43<6:56:15,  1.66it/s]
 37%|███▋      | 24066/65536 [4:07:43<6:57:44,  1.65it/s]
 37%|███▋      | 24067/65536 [4:07:44<7:05:02,  1.63it/s]
 37%|███▋      | 24068/65536 [4:07:45<6:55:58,  1.66it/s]
 37%|███▋      | 24069/65536 [4:07:45<6:54:56,  1.67it/s]
 37%|███▋      | 24070/65536 [4:07:46<6:53:10,  1.67it/s]
 37%|███▋      | 24071/65536 [4:07:46<6:47:01,  1.70it/s]
 37%|███▋      | 24072/65536 [4:07:47<6:49:27,  1.69it/s]
 37%|███▋      | 24073/65536 [4:07:48<6:57:51,  1.65it/s]
 37%|███▋      | 24074/65536 [4:07:48<7:14:30,  1.59it/s]
 37%|███▋      | 24075/65536 [4:07:49<7:20:30,  1.57it/s]
 37%|███▋      | 24076/65536 [4:07:49<7:16:53,  1.58it/s]
 37%|███▋      | 24077/65536 [4:07:50<7:22:25,  1.56it/s]
 37%|███▋      | 24078/65536 [4:07:51<7:09:01,  1.61it/s]
 37%|███▋      | 24079/65536 [4:07:51<7:03:28,  1.63it/s]
 37%|███▋      | 24080/65536 [4:07:52<7:05:33,  1.62it/s]
                                                         
{'loss': 3.0368, 'learning_rate': 6.710542426839721e-07, 'epoch': 1486.42}
+
 37%|███▋      | 24080/65536 [4:07:52<7:05:33,  1.62it/s]
 37%|███▋      | 24081/65536 [4:07:53<7:00:01,  1.64it/s]
 37%|███▋      | 24082/65536 [4:07:53<6:55:34,  1.66it/s]
 37%|███▋      | 24083/65536 [4:07:54<6:51:27,  1.68it/s]
 37%|███▋      | 24084/65536 [4:07:54<6:51:37,  1.68it/s]
 37%|███▋      | 24085/65536 [4:07:55<6:51:17,  1.68it/s]
 37%|███▋      | 24086/65536 [4:07:56<7:14:33,  1.59it/s]
 37%|███▋      | 24087/65536 [4:07:56<7:14:23,  1.59it/s]
 37%|███▋      | 24088/65536 [4:07:57<7:04:57,  1.63it/s]
 37%|███▋      | 24089/65536 [4:07:57<7:07:28,  1.62it/s]
 37%|███▋      | 24090/65536 [4:07:58<7:16:20,  1.58it/s]
 37%|███▋      | 24091/65536 [4:07:59<7:02:33,  1.63it/s]
 37%|███▋      | 24092/65536 [4:07:59<7:03:14,  1.63it/s]
 37%|███▋      | 24093/65536 [4:08:00<7:15:24,  1.59it/s]
 37%|███▋      | 24094/65536 [4:08:01<7:11:37,  1.60it/s]
 37%|███▋      | 24095/65536 [4:08:01<7:16:11,  1.58it/s]
 37%|███▋      | 24096/65536 [4:08:02<7:14:54,  1.59it/s]
 37%|███▋      | 24097/65536 [4:08:02<7:11:39,  1.60it/s]
 37%|███▋      | 24098/65536 [4:08:03<7:06:07,  1.62it/s]
 37%|███▋      | 24099/65536 [4:08:04<7:02:12,  1.64it/s]
 37%|███▋      | 24100/65536 [4:08:04<6:53:36,  1.67it/s]
                                                         
{'loss': 2.8998, 'learning_rate': 6.707787437247458e-07, 'epoch': 1487.65}
+
 37%|███▋      | 24100/65536 [4:08:04<6:53:36,  1.67it/s]
 37%|███▋      | 24101/65536 [4:08:05<6:56:23,  1.66it/s]
 37%|███▋      | 24102/65536 [4:08:05<6:59:55,  1.64it/s]
 37%|███▋      | 24103/65536 [4:08:06<7:08:30,  1.61it/s]
 37%|███▋      | 24104/65536 [4:08:07<7:16:51,  1.58it/s]
 37%|███▋      | 24105/65536 [4:08:07<7:06:21,  1.62it/s]
 37%|███▋      | 24106/65536 [4:08:08<7:25:50,  1.55it/s]
 37%|███▋      | 24107/65536 [4:08:09<7:31:09,  1.53it/s]
 37%|███▋      | 24108/65536 [4:08:09<7:22:59,  1.56it/s]
 37%|███▋      | 24109/65536 [4:08:10<7:13:30,  1.59it/s]
 37%|███▋      | 24110/65536 [4:08:11<7:10:02,  1.61it/s]
 37%|███▋      | 24111/65536 [4:08:11<7:13:11,  1.59it/s]
 37%|███▋      | 24112/65536 [4:08:12<7:06:55,  1.62it/s]
 37%|███▋      | 24113/65536 [4:08:12<6:57:10,  1.65it/s]
 37%|███▋      | 24114/65536 [4:08:13<6:57:30,  1.65it/s]
 37%|███▋      | 24115/65536 [4:08:14<6:52:06,  1.68it/s]
 37%|███▋      | 24116/65536 [4:08:14<7:07:29,  1.61it/s]
 37%|███▋      | 24117/65536 [4:08:15<7:07:23,  1.62it/s]
 37%|███▋      | 24118/65536 [4:08:15<6:55:31,  1.66it/s]
 37%|███▋      | 24119/65536 [4:08:16<6:50:50,  1.68it/s]
 37%|███▋      | 24120/65536 [4:08:17<6:55:46,  1.66it/s]
                                                         
{'loss': 2.9467, 'learning_rate': 6.705032447655197e-07, 'epoch': 1488.89}
+
 37%|███▋      | 24120/65536 [4:08:17<6:55:46,  1.66it/s]
 37%|███▋      | 24121/65536 [4:08:17<7:00:41,  1.64it/s]
 37%|███▋      | 24122/65536 [4:08:18<7:09:22,  1.61it/s]
 37%|███▋      | 24123/65536 [4:08:18<7:04:00,  1.63it/s]
 37%|███▋      | 24124/65536 [4:08:19<6:54:36,  1.66it/s]
 37%|███▋      | 24125/65536 [4:08:20<6:53:28,  1.67it/s]
 37%|███▋      | 24126/65536 [4:08:20<6:46:50,  1.70it/s]
 37%|███▋      | 24127/65536 [4:08:21<7:01:56,  1.64it/s]
 37%|███▋      | 24128/65536 [4:08:22<7:09:32,  1.61it/s]
 37%|███▋      | 24129/65536 [4:08:22<7:01:27,  1.64it/s]
 37%|███▋      | 24130/65536 [4:08:23<6:57:54,  1.65it/s]
 37%|███▋      | 24131/65536 [4:08:23<6:52:25,  1.67it/s]
 37%|███▋      | 24132/65536 [4:08:24<6:52:51,  1.67it/s]
 37%|███▋      | 24133/65536 [4:08:24<6:52:55,  1.67it/s]
 37%|███▋      | 24134/65536 [4:08:25<6:48:28,  1.69it/s]
 37%|███▋      | 24135/65536 [4:08:26<6:56:34,  1.66it/s]
 37%|███▋      | 24136/65536 [4:08:26<6:55:06,  1.66it/s]
 37%|███▋      | 24137/65536 [4:08:27<6:48:32,  1.69it/s]
 37%|███▋      | 24138/65536 [4:08:28<7:12:34,  1.60it/s]
 37%|███▋      | 24139/65536 [4:08:28<7:36:02,  1.51it/s]
 37%|███▋      | 24140/65536 [4:08:29<7:22:35,  1.56it/s]
                                                         
{'loss': 3.0443, 'learning_rate': 6.702277458062935e-07, 'epoch': 1490.12}
+
 37%|███▋      | 24140/65536 [4:08:29<7:22:35,  1.56it/s]
 37%|███▋      | 24141/65536 [4:08:30<7:21:32,  1.56it/s]
 37%|███▋      | 24142/65536 [4:08:30<7:11:16,  1.60it/s]
 37%|███▋      | 24143/65536 [4:08:31<7:04:19,  1.63it/s]
 37%|███▋      | 24144/65536 [4:08:31<7:03:30,  1.63it/s]
 37%|███▋      | 24145/65536 [4:08:32<7:12:43,  1.59it/s]
 37%|███▋      | 24146/65536 [4:08:33<7:10:43,  1.60it/s]
 37%|███▋      | 24147/65536 [4:08:33<7:07:22,  1.61it/s]
 37%|███▋      | 24148/65536 [4:08:34<6:53:23,  1.67it/s]
 37%|███▋      | 24149/65536 [4:08:34<7:06:30,  1.62it/s]
 37%|███▋      | 24150/65536 [4:08:35<6:58:54,  1.65it/s]
 37%|███▋      | 24151/65536 [4:08:36<7:09:06,  1.61it/s]
 37%|███▋      | 24152/65536 [4:08:36<7:00:27,  1.64it/s]
 37%|███▋      | 24153/65536 [4:08:37<7:03:21,  1.63it/s]
 37%|███▋      | 24154/65536 [4:08:37<7:05:38,  1.62it/s]
 37%|███▋      | 24155/65536 [4:08:38<7:29:46,  1.53it/s]
 37%|███▋      | 24156/65536 [4:08:39<7:19:42,  1.57it/s]
 37%|███▋      | 24157/65536 [4:08:39<7:27:37,  1.54it/s]
 37%|███▋      | 24158/65536 [4:08:40<7:28:03,  1.54it/s]
 37%|███▋      | 24159/65536 [4:08:41<7:16:29,  1.58it/s]
 37%|███▋      | 24160/65536 [4:08:41<7:04:08,  1.63it/s]
                                                         
{'loss': 2.9395, 'learning_rate': 6.699522468470674e-07, 'epoch': 1491.36}
+
 37%|███▋      | 24160/65536 [4:08:41<7:04:08,  1.63it/s]
 37%|███▋      | 24161/65536 [4:08:42<6:54:35,  1.66it/s]
 37%|███▋      | 24162/65536 [4:08:42<6:45:35,  1.70it/s]
 37%|███▋      | 24163/65536 [4:08:43<6:44:52,  1.70it/s]
 37%|███▋      | 24164/65536 [4:08:44<6:42:56,  1.71it/s]
 37%|███▋      | 24165/65536 [4:08:44<6:54:52,  1.66it/s]
 37%|███▋      | 24166/65536 [4:08:45<7:04:09,  1.63it/s]
 37%|███▋      | 24167/65536 [4:08:45<7:01:38,  1.64it/s]
 37%|███▋      | 24168/65536 [4:08:46<6:58:37,  1.65it/s]
 37%|███▋      | 24169/65536 [4:08:47<7:04:43,  1.62it/s]
 37%|███▋      | 24170/65536 [4:08:47<6:51:47,  1.67it/s]
 37%|███▋      | 24171/65536 [4:08:48<7:08:47,  1.61it/s]
 37%|███▋      | 24172/65536 [4:08:49<7:13:47,  1.59it/s]
 37%|███▋      | 24173/65536 [4:08:49<7:07:41,  1.61it/s]
 37%|███▋      | 24174/65536 [4:08:50<6:53:52,  1.67it/s]
 37%|███▋      | 24175/65536 [4:08:50<6:57:19,  1.65it/s]
 37%|███▋      | 24176/65536 [4:08:51<6:46:35,  1.70it/s]
 37%|███▋      | 24177/65536 [4:08:52<6:43:55,  1.71it/s]
 37%|███▋      | 24178/65536 [4:08:52<6:40:19,  1.72it/s]
 37%|███▋      | 24179/65536 [4:08:53<6:53:37,  1.67it/s]
 37%|███▋      | 24180/65536 [4:08:53<6:59:36,  1.64it/s]
                                                         
{'loss': 3.0041, 'learning_rate': 6.696767478878413e-07, 'epoch': 1492.59}
+
 37%|███▋      | 24180/65536 [4:08:53<6:59:36,  1.64it/s]
 37%|███▋      | 24181/65536 [4:08:54<7:14:32,  1.59it/s]
 37%|███▋      | 24182/65536 [4:08:55<7:09:06,  1.61it/s]
 37%|███▋      | 24183/65536 [4:08:55<7:03:30,  1.63it/s]
 37%|███▋      | 24184/65536 [4:08:56<6:54:20,  1.66it/s]
 37%|███▋      | 24185/65536 [4:08:56<6:59:55,  1.64it/s]
 37%|███▋      | 24186/65536 [4:08:57<6:59:08,  1.64it/s]
 37%|███▋      | 24187/65536 [4:08:58<7:14:36,  1.59it/s]
 37%|███▋      | 24188/65536 [4:08:58<7:10:44,  1.60it/s]
 37%|███▋      | 24189/65536 [4:08:59<7:04:01,  1.63it/s]
 37%|███���      | 24190/65536 [4:09:00<7:12:25,  1.59it/s]
 37%|███▋      | 24191/65536 [4:09:00<7:07:38,  1.61it/s]
 37%|███▋      | 24192/65536 [4:09:01<7:02:12,  1.63it/s]
 37%|███▋      | 24193/65536 [4:09:01<7:02:24,  1.63it/s]
 37%|███▋      | 24194/65536 [4:09:02<7:01:20,  1.64it/s]
 37%|███▋      | 24195/65536 [4:09:03<6:54:57,  1.66it/s]
 37%|███▋      | 24196/65536 [4:09:03<6:55:37,  1.66it/s]
 37%|███▋      | 24197/65536 [4:09:04<6:58:56,  1.64it/s]
 37%|███▋      | 24198/65536 [4:09:04<7:05:54,  1.62it/s]
 37%|███▋      | 24199/65536 [4:09:05<6:59:54,  1.64it/s]
 37%|███▋      | 24200/65536 [4:09:06<7:15:06,  1.58it/s]
                                                         
{'loss': 2.9951, 'learning_rate': 6.694012489286151e-07, 'epoch': 1493.83}
+
 37%|███▋      | 24200/65536 [4:09:06<7:15:06,  1.58it/s]
 37%|███▋      | 24201/65536 [4:09:06<7:03:57,  1.62it/s]
 37%|███▋      | 24202/65536 [4:09:07<7:01:40,  1.63it/s]
 37%|███▋      | 24203/65536 [4:09:08<7:14:25,  1.59it/s]
 37%|███▋      | 24204/65536 [4:09:08<6:58:16,  1.65it/s]
 37%|███▋      | 24205/65536 [4:09:09<6:52:35,  1.67it/s]
 37%|███▋      | 24206/65536 [4:09:09<6:52:06,  1.67it/s]
 37%|███▋      | 24207/65536 [4:09:10<6:54:18,  1.66it/s]
 37%|███▋      | 24208/65536 [4:09:10<6:52:20,  1.67it/s]
 37%|███▋      | 24209/65536 [4:09:11<7:11:04,  1.60it/s]
 37%|███▋      | 24210/65536 [4:09:12<7:18:25,  1.57it/s]
 37%|███▋      | 24211/65536 [4:09:12<7:14:14,  1.59it/s]
 37%|███▋      | 24212/65536 [4:09:13<7:02:37,  1.63it/s]
 37%|███▋      | 24213/65536 [4:09:14<6:51:08,  1.68it/s]
 37%|███▋      | 24214/65536 [4:09:14<6:57:22,  1.65it/s]
 37%|███▋      | 24215/65536 [4:09:15<6:56:26,  1.65it/s]
 37%|███▋      | 24216/65536 [4:09:15<7:07:30,  1.61it/s]
 37%|███▋      | 24217/65536 [4:09:16<7:04:49,  1.62it/s]
 37%|███▋      | 24218/65536 [4:09:17<7:00:09,  1.64it/s]
 37%|███▋      | 24219/65536 [4:09:17<7:00:26,  1.64it/s]
 37%|███▋      | 24220/65536 [4:09:18<7:08:32,  1.61it/s]
                                                         
{'loss': 3.0352, 'learning_rate': 6.69125749969389e-07, 'epoch': 1495.06}
+
 37%|███▋      | 24220/65536 [4:09:18<7:08:32,  1.61it/s]
 37%|███▋      | 24221/65536 [4:09:19<7:05:53,  1.62it/s]
 37%|███▋      | 24222/65536 [4:09:19<7:04:47,  1.62it/s]
 37%|███▋      | 24223/65536 [4:09:20<7:17:56,  1.57it/s]
 37%|███▋      | 24224/65536 [4:09:20<7:10:46,  1.60it/s]
 37%|███▋      | 24225/65536 [4:09:21<7:01:07,  1.63it/s]
 37%|███▋      | 24226/65536 [4:09:22<6:53:40,  1.66it/s]
 37%|███▋      | 24227/65536 [4:09:22<6:50:00,  1.68it/s]
 37%|███▋      | 24228/65536 [4:09:23<6:42:41,  1.71it/s]
 37%|███▋      | 24229/65536 [4:09:23<6:51:09,  1.67it/s]
 37%|███▋      | 24230/65536 [4:09:24<6:57:21,  1.65it/s]
 37%|███▋      | 24231/65536 [4:09:25<7:02:56,  1.63it/s]
 37%|███▋      | 24232/65536 [4:09:25<7:13:36,  1.59it/s]
 37%|███▋      | 24233/65536 [4:09:26<7:18:44,  1.57it/s]
 37%|███▋      | 24234/65536 [4:09:27<7:12:55,  1.59it/s]
 37%|███▋      | 24235/65536 [4:09:27<6:54:46,  1.66it/s]
 37%|███▋      | 24236/65536 [4:09:28<7:02:19,  1.63it/s]
 37%|███▋      | 24237/65536 [4:09:28<7:21:41,  1.56it/s]
 37%|███▋      | 24238/65536 [4:09:29<7:07:10,  1.61it/s]
 37%|███▋      | 24239/65536 [4:09:30<6:54:23,  1.66it/s]
 37%|███▋      | 24240/65536 [4:09:30<7:06:56,  1.61it/s]
                                                         
{'loss': 2.9081, 'learning_rate': 6.688502510101628e-07, 'epoch': 1496.3}
+
 37%|███▋      | 24240/65536 [4:09:30<7:06:56,  1.61it/s]
 37%|███▋      | 24241/65536 [4:09:31<7:05:57,  1.62it/s]
 37%|███▋      | 24242/65536 [4:09:31<6:56:22,  1.65it/s]
 37%|███▋      | 24243/65536 [4:09:32<7:03:51,  1.62it/s]
 37%|███▋      | 24244/65536 [4:09:33<7:04:26,  1.62it/s]
 37%|███▋      | 24245/65536 [4:09:33<7:00:36,  1.64it/s]
 37%|███▋      | 24246/65536 [4:09:34<6:59:06,  1.64it/s]
 37%|███▋      | 24247/65536 [4:09:35<7:02:26,  1.63it/s]
 37%|███▋      | 24248/65536 [4:09:35<7:01:39,  1.63it/s]
 37%|███▋      | 24249/65536 [4:09:36<7:03:59,  1.62it/s]
 37%|███▋      | 24250/65536 [4:09:36<6:58:02,  1.65it/s]
 37%|███▋      | 24251/65536 [4:09:37<6:57:46,  1.65it/s]
 37%|███▋      | 24252/65536 [4:09:38<7:13:06,  1.59it/s]
 37%|███▋      | 24253/65536 [4:09:38<7:06:07,  1.61it/s]
 37%|███▋      | 24254/65536 [4:09:39<6:59:50,  1.64it/s]
 37%|███▋      | 24255/65536 [4:09:39<6:58:24,  1.64it/s]
 37%|███▋      | 24256/65536 [4:09:40<6:54:42,  1.66it/s]
 37%|███▋      | 24257/65536 [4:09:41<6:51:17,  1.67it/s]
 37%|███▋      | 24258/65536 [4:09:41<6:59:36,  1.64it/s]
 37%|███▋      | 24259/65536 [4:09:42<6:56:57,  1.65it/s]
 37%|███▋      | 24260/65536 [4:09:42<7:06:08,  1.61it/s]
                                                         
{'loss': 3.0379, 'learning_rate': 6.685747520509367e-07, 'epoch': 1497.53}
+
 37%|███▋      | 24260/65536 [4:09:42<7:06:08,  1.61it/s]
 37%|███▋      | 24261/65536 [4:09:43<7:12:54,  1.59it/s]
 37%|███▋      | 24262/65536 [4:09:44<7:10:14,  1.60it/s]
 37%|███▋      | 24263/65536 [4:09:44<7:07:51,  1.61it/s]
 37%|███▋      | 24264/65536 [4:09:45<7:12:48,  1.59it/s]
 37%|███▋      | 24265/65536 [4:09:46<7:09:36,  1.60it/s]
 37%|███▋      | 24266/65536 [4:09:46<7:02:14,  1.63it/s]
 37%|███▋      | 24267/65536 [4:09:47<7:02:14,  1.63it/s]
 37%|███▋      | 24268/65536 [4:09:48<7:22:02,  1.56it/s]
 37%|███▋      | 24269/65536 [4:09:48<7:12:38,  1.59it/s]
 37%|███▋      | 24270/65536 [4:09:49<7:12:54,  1.59it/s]
 37%|███▋      | 24271/65536 [4:09:49<6:59:22,  1.64it/s]
 37%|███▋      | 24272/65536 [4:09:50<7:04:35,  1.62it/s]
 37%|███▋      | 24273/65536 [4:09:51<6:55:52,  1.65it/s]
 37%|███▋      | 24274/65536 [4:09:51<6:50:52,  1.67it/s]
 37%|███▋      | 24275/65536 [4:09:52<6:55:46,  1.65it/s]
 37%|███▋      | 24276/65536 [4:09:52<6:53:48,  1.66it/s]
 37%|███▋      | 24277/65536 [4:09:53<6:56:48,  1.65it/s]
 37%|███▋      | 24278/65536 [4:09:54<6:50:40,  1.67it/s]
 37%|███▋      | 24279/65536 [4:09:54<7:00:00,  1.64it/s]
 37%|███▋      | 24280/65536 [4:09:55<6:57:39,  1.65it/s]
                                                         
{'loss': 2.9549, 'learning_rate': 6.682992530917105e-07, 'epoch': 1498.77}
+
 37%|███▋      | 24280/65536 [4:09:55<6:57:39,  1.65it/s]
 37%|███▋      | 24281/65536 [4:09:55<7:14:28,  1.58it/s]
 37%|███▋      | 24282/65536 [4:09:56<7:02:00,  1.63it/s]
 37%|███▋      | 24283/65536 [4:09:57<6:55:40,  1.65it/s]
 37%|███▋      | 24284/65536 [4:09:57<7:02:38,  1.63it/s]
 37%|███▋      | 24285/65536 [4:09:58<7:05:00,  1.62it/s]
 37%|███▋      | 24286/65536 [4:09:58<7:04:03,  1.62it/s]
 37%|███▋      | 24287/65536 [4:09:59<6:53:46,  1.66it/s]
 37%|███▋      | 24288/65536 [4:10:00<6:45:09,  1.70it/s]
 37%|███▋      | 24289/65536 [4:10:00<6:50:48,  1.67it/s]
 37%|███▋      | 24290/65536 [4:10:01<6:46:07,  1.69it/s]
 37%|███▋      | 24291/65536 [4:10:01<6:57:16,  1.65it/s]
 37%|███▋      | 24292/65536 [4:10:02<6:48:07,  1.68it/s]
 37%|███▋      | 24293/65536 [4:10:03<7:04:07,  1.62it/s]
 37%|███▋      | 24294/65536 [4:10:03<7:04:29,  1.62it/s]
 37%|███▋      | 24295/65536 [4:10:04<6:54:57,  1.66it/s]
 37%|███▋      | 24296/65536 [4:10:04<6:52:28,  1.67it/s]
 37%|███▋      | 24297/65536 [4:10:05<6:54:42,  1.66it/s]
 37%|███▋      | 24298/65536 [4:10:06<7:10:25,  1.60it/s]
 37%|███▋      | 24299/65536 [4:10:06<7:01:07,  1.63it/s]
 37%|███▋      | 24300/65536 [4:10:07<7:00:01,  1.64it/s]
                                                         
{'loss': 3.0366, 'learning_rate': 6.680237541324844e-07, 'epoch': 1500.0}
+
 37%|███▋      | 24300/65536 [4:10:07<7:00:01,  1.64it/s]
 37%|███▋      | 24301/65536 [4:10:08<7:11:03,  1.59it/s]
 37%|███▋      | 24302/65536 [4:10:08<7:13:11,  1.59it/s]
 37%|███▋      | 24303/65536 [4:10:09<7:08:17,  1.60it/s]
 37%|███▋      | 24304/65536 [4:10:09<7:07:52,  1.61it/s]
 37%|███▋      | 24305/65536 [4:10:10<7:00:57,  1.63it/s]
 37%|███▋      | 24306/65536 [4:10:11<6:55:10,  1.66it/s]
 37%|███▋      | 24307/65536 [4:10:11<7:00:41,  1.63it/s]
 37%|███▋      | 24308/65536 [4:10:12<7:14:11,  1.58it/s]
 37%|███▋      | 24309/65536 [4:10:13<7:05:51,  1.61it/s]
 37%|███▋      | 24310/65536 [4:10:13<7:02:37,  1.63it/s]
 37%|███▋      | 24311/65536 [4:10:14<7:08:58,  1.60it/s]
 37%|███▋      | 24312/65536 [4:10:14<7:05:31,  1.61it/s]
 37%|███▋      | 24313/65536 [4:10:15<7:03:09,  1.62it/s]
 37%|███▋      | 24314/65536 [4:10:16<7:00:37,  1.63it/s]
 37%|███▋      | 24315/65536 [4:10:16<7:01:09,  1.63it/s]
 37%|███▋      | 24316/65536 [4:10:17<6:56:03,  1.65it/s]
 37%|███▋      | 24317/65536 [4:10:18<7:12:30,  1.59it/s]
 37%|███▋      | 24318/65536 [4:10:18<7:20:02,  1.56it/s]
 37%|███▋      | 24319/65536 [4:10:19<7:09:50,  1.60it/s]
 37%|███▋      | 24320/65536 [4:10:19<7:08:07,  1.60it/s]
                                                         
{'loss': 2.9446, 'learning_rate': 6.677482551732582e-07, 'epoch': 1501.23}
+
 37%|███▋      | 24320/65536 [4:10:19<7:08:07,  1.60it/s]
 37%|███▋      | 24321/65536 [4:10:20<7:02:08,  1.63it/s]
 37%|███▋      | 24322/65536 [4:10:21<7:04:09,  1.62it/s]
 37%|███▋      | 24323/65536 [4:10:21<6:54:05,  1.66it/s]
 37%|��██▋      | 24324/65536 [4:10:22<6:53:11,  1.66it/s]
 37%|███▋      | 24325/65536 [4:10:22<6:55:02,  1.65it/s]
 37%|███▋      | 24326/65536 [4:10:23<6:56:04,  1.65it/s]
 37%|███▋      | 24327/65536 [4:10:24<6:50:45,  1.67it/s]
 37%|███▋      | 24328/65536 [4:10:24<6:51:14,  1.67it/s]
 37%|███▋      | 24329/65536 [4:10:25<6:51:50,  1.67it/s]
 37%|███▋      | 24330/65536 [4:10:25<6:52:12,  1.67it/s]
 37%|███▋      | 24331/65536 [4:10:26<6:56:42,  1.65it/s]
 37%|███▋      | 24332/65536 [4:10:27<6:57:42,  1.64it/s]
 37%|███▋      | 24333/65536 [4:10:27<7:20:02,  1.56it/s]
 37%|███▋      | 24334/65536 [4:10:28<7:17:09,  1.57it/s]
 37%|███▋      | 24335/65536 [4:10:29<7:16:10,  1.57it/s]
 37%|███▋      | 24336/65536 [4:10:29<7:06:15,  1.61it/s]
 37%|███▋      | 24337/65536 [4:10:30<6:49:11,  1.68it/s]
 37%|███▋      | 24338/65536 [4:10:30<6:55:58,  1.65it/s]
 37%|███▋      | 24339/65536 [4:10:31<7:07:31,  1.61it/s]
 37%|███▋      | 24340/65536 [4:10:32<7:03:37,  1.62it/s]
                                                         
{'loss': 2.9547, 'learning_rate': 6.67472756214032e-07, 'epoch': 1502.47}
+
 37%|███▋      | 24340/65536 [4:10:32<7:03:37,  1.62it/s]
 37%|███▋      | 24341/65536 [4:10:32<7:12:52,  1.59it/s]
 37%|███▋      | 24342/65536 [4:10:33<7:07:18,  1.61it/s]
 37%|███▋      | 24343/65536 [4:10:33<6:56:46,  1.65it/s]
 37%|███▋      | 24344/65536 [4:10:34<7:03:28,  1.62it/s]
 37%|███▋      | 24345/65536 [4:10:35<7:00:07,  1.63it/s]
 37%|███▋      | 24346/65536 [4:10:35<6:58:34,  1.64it/s]
 37%|███▋      | 24347/65536 [4:10:36<6:49:18,  1.68it/s]
 37%|███▋      | 24348/65536 [4:10:36<6:52:29,  1.66it/s]
 37%|███▋      | 24349/65536 [4:10:37<7:15:18,  1.58it/s]
 37%|███▋      | 24350/65536 [4:10:38<7:15:16,  1.58it/s]
 37%|███▋      | 24351/65536 [4:10:38<7:07:12,  1.61it/s]
 37%|███▋      | 24352/65536 [4:10:39<6:50:16,  1.67it/s]
 37%|███▋      | 24353/65536 [4:10:40<6:51:38,  1.67it/s]
 37%|███▋      | 24354/65536 [4:10:40<6:51:57,  1.67it/s]
 37%|███▋      | 24355/65536 [4:10:41<6:54:26,  1.66it/s]
 37%|███▋      | 24356/65536 [4:10:41<7:00:43,  1.63it/s]
 37%|███▋      | 24357/65536 [4:10:42<7:01:11,  1.63it/s]
 37%|███▋      | 24358/65536 [4:10:43<7:01:00,  1.63it/s]
 37%|███▋      | 24359/65536 [4:10:43<7:00:57,  1.63it/s]
 37%|███▋      | 24360/65536 [4:10:44<6:54:43,  1.65it/s]
                                                         
{'loss': 2.9982, 'learning_rate': 6.671972572548059e-07, 'epoch': 1503.7}
+
 37%|███▋      | 24360/65536 [4:10:44<6:54:43,  1.65it/s]
 37%|███▋      | 24361/65536 [4:10:44<6:53:49,  1.66it/s]
 37%|███▋      | 24362/65536 [4:10:45<7:00:21,  1.63it/s]
 37%|███▋      | 24363/65536 [4:10:46<7:06:24,  1.61it/s]
 37%|███▋      | 24364/65536 [4:10:46<7:06:47,  1.61it/s]
 37%|███▋      | 24365/65536 [4:10:47<7:09:10,  1.60it/s]
 37%|███▋      | 24366/65536 [4:10:48<7:11:16,  1.59it/s]
 37%|███▋      | 24367/65536 [4:10:48<7:04:23,  1.62it/s]
 37%|███▋      | 24368/65536 [4:10:49<7:06:01,  1.61it/s]
 37%|███▋      | 24369/65536 [4:10:49<7:02:04,  1.63it/s]
 37%|███▋      | 24370/65536 [4:10:50<6:54:22,  1.66it/s]
 37%|███▋      | 24371/65536 [4:10:51<6:50:54,  1.67it/s]
 37%|███▋      | 24372/65536 [4:10:51<6:49:40,  1.67it/s]
 37%|███▋      | 24373/65536 [4:10:52<6:46:05,  1.69it/s]
 37%|███▋      | 24374/65536 [4:10:52<6:40:18,  1.71it/s]
 37%|███▋      | 24375/65536 [4:10:53<6:39:58,  1.72it/s]
 37%|███▋      | 24376/65536 [4:10:54<7:01:47,  1.63it/s]
 37%|███▋      | 24377/65536 [4:10:54<7:13:54,  1.58it/s]
 37%|███▋      | 24378/65536 [4:10:55<7:10:15,  1.59it/s]
 37%|███▋      | 24379/65536 [4:10:56<7:13:00,  1.58it/s]
 37%|███▋      | 24380/65536 [4:10:56<7:16:45,  1.57it/s]
                                                         
{'loss': 3.0066, 'learning_rate': 6.669217582955797e-07, 'epoch': 1504.94}
+
 37%|███▋      | 24380/65536 [4:10:56<7:16:45,  1.57it/s]
 37%|███▋      | 24381/65536 [4:10:57<7:03:31,  1.62it/s]
 37%|███▋      | 24382/65536 [4:10:57<7:12:15,  1.59it/s]
 37%|███▋      | 24383/65536 [4:10:58<7:03:57,  1.62it/s]
 37%|███▋      | 24384/65536 [4:10:59<7:15:32,  1.57it/s]
 37%|███▋      | 24385/65536 [4:10:59<7:14:30,  1.58it/s]
 37%|███▋      | 24386/65536 [4:11:00<7:10:27,  1.59it/s]
 37%|███▋      | 24387/65536 [4:11:01<7:08:02,  1.60it/s]
 37%|███▋      | 24388/65536 [4:11:01<6:58:31,  1.64it/s]
 37%|███▋      | 24389/65536 [4:11:02<6:54:07,  1.66it/s]
 37%|███▋      | 24390/65536 [4:11:02<6:53:37,  1.66it/s]
 37%|███▋      | 24391/65536 [4:11:03<6:49:34,  1.67it/s]
 37%|███▋      | 24392/65536 [4:11:03<6:53:38,  1.66it/s]
 37%|███▋      | 24393/65536 [4:11:04<6:54:50,  1.65it/s]
 37%|███▋      | 24394/65536 [4:11:05<6:50:42,  1.67it/s]
 37%|███▋      | 24395/65536 [4:11:05<6:52:00,  1.66it/s]
 37%|███▋      | 24396/65536 [4:11:06<7:00:59,  1.63it/s]
 37%|███▋      | 24397/65536 [4:11:07<7:05:23,  1.61it/s]
 37%|███▋      | 24398/65536 [4:11:07<7:08:59,  1.60it/s]
 37%|███▋      | 24399/65536 [4:11:08<7:15:45,  1.57it/s]
 37%|███▋      | 24400/65536 [4:11:08<7:13:33,  1.58it/s]
                                                         
{'loss': 3.0048, 'learning_rate': 6.666462593363536e-07, 'epoch': 1506.17}
+
 37%|███▋      | 24400/65536 [4:11:08<7:13:33,  1.58it/s]
 37%|███▋      | 24401/65536 [4:11:09<7:04:47,  1.61it/s]
 37%|███▋      | 24402/65536 [4:11:10<6:58:41,  1.64it/s]
 37%|███▋      | 24403/65536 [4:11:10<7:04:47,  1.61it/s]
 37%|███▋      | 24404/65536 [4:11:11<6:54:46,  1.65it/s]
 37%|███▋      | 24405/65536 [4:11:11<6:50:13,  1.67it/s]
 37%|███▋      | 24406/65536 [4:11:12<6:52:37,  1.66it/s]
 37%|███▋      | 24407/65536 [4:11:13<7:02:22,  1.62it/s]
 37%|███▋      | 24408/65536 [4:11:13<7:02:39,  1.62it/s]
 37%|███▋      | 24409/65536 [4:11:14<6:54:44,  1.65it/s]
 37%|███▋      | 24410/65536 [4:11:15<6:50:59,  1.67it/s]
 37%|███▋      | 24411/65536 [4:11:15<7:02:50,  1.62it/s]
 37%|███▋      | 24412/65536 [4:11:16<7:05:14,  1.61it/s]
 37%|███▋      | 24413/65536 [4:11:16<7:09:37,  1.60it/s]
 37%|███▋      | 24414/65536 [4:11:17<7:17:10,  1.57it/s]
 37%|███▋      | 24415/65536 [4:11:18<7:09:25,  1.60it/s]
 37%|███▋      | 24416/65536 [4:11:18<7:01:56,  1.62it/s]
 37%|███▋      | 24417/65536 [4:11:19<6:57:32,  1.64it/s]
 37%|███▋      | 24418/65536 [4:11:19<6:54:23,  1.65it/s]
 37%|███▋      | 24419/65536 [4:11:20<6:53:01,  1.66it/s]
 37%|███▋      | 24420/65536 [4:11:21<6:55:56,  1.65it/s]
                                                         
{'loss': 3.0346, 'learning_rate': 6.663707603771275e-07, 'epoch': 1507.41}
+
 37%|███▋      | 24420/65536 [4:11:21<6:55:56,  1.65it/s]
 37%|███▋      | 24421/65536 [4:11:21<6:51:34,  1.66it/s]
 37%|███▋      | 24422/65536 [4:11:22<6:59:58,  1.63it/s]
 37%|███▋      | 24423/65536 [4:11:23<6:53:45,  1.66it/s]
 37%|███▋      | 24424/65536 [4:11:23<7:00:26,  1.63it/s]
 37%|███▋      | 24425/65536 [4:11:24<6:58:50,  1.64it/s]
 37%|███▋      | 24426/65536 [4:11:24<7:13:53,  1.58it/s]
 37%|███▋      | 24427/65536 [4:11:25<7:08:51,  1.60it/s]
 37%|███▋      | 24428/65536 [4:11:26<7:05:52,  1.61it/s]
 37%|███▋      | 24429/65536 [4:11:26<7:01:41,  1.62it/s]
 37%|███▋      | 24430/65536 [4:11:27<7:11:25,  1.59it/s]
 37%|███▋      | 24431/65536 [4:11:28<7:04:37,  1.61it/s]
 37%|███▋      | 24432/65536 [4:11:28<6:55:38,  1.65it/s]
 37%|███▋      | 24433/65536 [4:11:29<6:48:37,  1.68it/s]
 37%|███▋      | 24434/65536 [4:11:29<6:57:27,  1.64it/s]
 37%|███▋      | 24435/65536 [4:11:30<6:49:34,  1.67it/s]
 37%|███▋      | 24436/65536 [4:11:30<6:45:17,  1.69it/s]
 37%|███▋      | 24437/65536 [4:11:31<6:39:42,  1.71it/s]
 37%|███▋      | 24438/65536 [4:11:32<6:47:56,  1.68it/s]
 37%|███▋      | 24439/65536 [4:11:32<6:52:50,  1.66it/s]
 37%|███▋      | 24440/65536 [4:11:33<6:59:02,  1.63it/s]
                                                         
{'loss': 2.9453, 'learning_rate': 6.660952614179013e-07, 'epoch': 1508.64}
+
 37%|███▋      | 24440/65536 [4:11:33<6:59:02,  1.63it/s]
 37%|███▋      | 24441/65536 [4:11:34<7:04:34,  1.61it/s]
 37%|███▋      | 24442/65536 [4:11:34<7:13:06,  1.58it/s]
 37%|███▋      | 24443/65536 [4:11:35<7:11:14,  1.59it/s]
 37%|███▋      | 24444/65536 [4:11:36<7:24:05,  1.54it/s]
 37%|███▋      | 24445/65536 [4:11:36<7:15:21,  1.57it/s]
 37%|███▋      | 24446/65536 [4:11:37<7:16:29,  1.57it/s]
 37%|███▋      | 24447/65536 [4:11:37<7:10:23,  1.59it/s]
 37%|███▋      | 24448/65536 [4:11:38<7:07:32,  1.60it/s]
 37%|███▋      | 24449/65536 [4:11:39<6:57:22,  1.64it/s]
 37%|███▋      | 24450/65536 [4:11:39<7:02:00,  1.62it/s]
 37%|███▋      | 24451/65536 [4:11:40<7:01:02,  1.63it/s]
 37%|███▋      | 24452/65536 [4:11:40<7:03:28,  1.62it/s]
 37%|███▋      | 24453/65536 [4:11:41<7:06:08,  1.61it/s]
 37%|███▋      | 24454/65536 [4:11:42<6:52:29,  1.66it/s]
 37%|███▋      | 24455/65536 [4:11:42<6:45:15,  1.69it/s]
 37%|███▋      | 24456/65536 [4:11:43<6:52:19,  1.66it/s]
 37%|███▋      | 24457/65536 [4:11:43<6:59:03,  1.63it/s]
 37%|███▋      | 24458/65536 [4:11:44<6:59:56,  1.63it/s]
 37%|███▋      | 24459/65536 [4:11:45<7:08:47,  1.60it/s]
 37%|███▋      | 24460/65536 [4:11:45<7:03:24,  1.62it/s]
                                                         
{'loss': 2.9708, 'learning_rate': 6.658197624586752e-07, 'epoch': 1509.88}
+
 37%|███▋      | 24460/65536 [4:11:45<7:03:24,  1.62it/s]
 37%|███▋      | 24461/65536 [4:11:46<7:05:43,  1.61it/s]
 37%|███▋      | 24462/65536 [4:11:47<6:59:23,  1.63it/s]
 37%|███▋      | 24463/65536 [4:11:47<7:11:48,  1.59it/s]
 37%|███▋      | 24464/65536 [4:11:48<7:04:44,  1.61it/s]
 37%|███▋      | 24465/65536 [4:11:48<6:54:10,  1.65it/s]
 37%|███▋      | 24466/65536 [4:11:49<6:51:27,  1.66it/s]
 37%|███▋      | 24467/65536 [4:11:50<6:56:15,  1.64it/s]
 37%|███▋      | 24468/65536 [4:11:50<7:06:14,  1.61it/s]
 37%|███▋      | 24469/65536 [4:11:51<7:01:27,  1.62it/s]
 37%|███▋      | 24470/65536 [4:11:51<7:00:04,  1.63it/s]
 37%|███▋      | 24471/65536 [4:11:52<6:58:10,  1.64it/s]
 37%|███▋      | 24472/65536 [4:11:53<7:02:35,  1.62it/s]
 37%|███▋      | 24473/65536 [4:11:53<7:04:18,  1.61it/s]
 37%|███▋      | 24474/65536 [4:11:54<7:06:50,  1.60it/s]
 37%|███▋      | 24475/65536 [4:11:55<7:05:26,  1.61it/s]
 37%|███▋      | 24476/65536 [4:11:55<7:05:16,  1.61it/s]
 37%|███▋      | 24477/65536 [4:11:56<7:10:09,  1.59it/s]
 37%|███▋      | 24478/65536 [4:11:56<6:59:45,  1.63it/s]
 37%|███▋      | 24479/65536 [4:11:57<7:20:06,  1.55it/s]
 37%|███▋      | 24480/65536 [4:11:58<7:16:46,  1.57it/s]
                                                         
{'loss': 2.9684, 'learning_rate': 6.65544263499449e-07, 'epoch': 1511.11}
+
 37%|███▋      | 24480/65536 [4:11:58<7:16:46,  1.57it/s]
 37%|███▋      | 24481/65536 [4:11:58<7:21:28,  1.55it/s]
 37%|███▋      | 24482/65536 [4:11:59<7:06:23,  1.60it/s]
 37%|███▋      | 24483/65536 [4:12:00<7:03:34,  1.62it/s]
 37%|███▋      | 24484/65536 [4:12:00<6:59:30,  1.63it/s]
 37%|███▋      | 24485/65536 [4:12:01<6:48:09,  1.68it/s]
 37%|███▋      | 24486/65536 [4:12:01<6:49:09,  1.67it/s]
 37%|███▋      | 24487/65536 [4:12:02<7:00:20,  1.63it/s]
 37%|███▋      | 24488/65536 [4:12:03<6:48:56,  1.67it/s]
 37%|███▋      | 24489/65536 [4:12:03<6:44:22,  1.69it/s]
 37%|███▋      | 24490/65536 [4:12:04<6:50:25,  1.67it/s]
 37%|███▋      | 24491/65536 [4:12:04<6:52:23,  1.66it/s]
 37%|███▋      | 24492/65536 [4:12:05<6:59:44,  1.63it/s]
 37%|███▋      | 24493/65536 [4:12:06<7:07:15,  1.60it/s]
 37%|███▋      | 24494/65536 [4:12:06<7:03:07,  1.62it/s]
 37%|███▋      | 24495/65536 [4:12:07<7:12:37,  1.58it/s]
 37%|███▋      | 24496/65536 [4:12:08<7:04:06,  1.61it/s]
 37%|███▋      | 24497/65536 [4:12:08<7:10:29,  1.59it/s]
 37%|███▋      | 24498/65536 [4:12:09<6:55:32,  1.65it/s]
 37%|███▋      | 24499/65536 [4:12:09<6:47:39,  1.68it/s]
 37%|███▋      | 24500/65536 [4:12:10<6:54:14,  1.65it/s]
                                                         
{'loss': 2.9751, 'learning_rate': 6.652687645402229e-07, 'epoch': 1512.35}
+
 37%|███▋      | 24500/65536 [4:12:10<6:54:14,  1.65it/s]
 37%|███▋      | 24501/65536 [4:12:11<6:52:43,  1.66it/s]
 37%|███▋      | 24502/65536 [4:12:11<6:58:01,  1.64it/s]
 37%|███▋      | 24503/65536 [4:12:12<7:04:45,  1.61it/s]
 37%|███▋      | 24504/65536 [4:12:12<7:06:20,  1.60it/s]
 37%|███▋      | 24505/65536 [4:12:13<7:06:45,  1.60it/s]
 37%|███▋      | 24506/65536 [4:12:14<6:58:08,  1.64it/s]
 37%|███▋      | 24507/65536 [4:12:14<6:51:18,  1.66it/s]
 37%|███▋      | 24508/65536 [4:12:15<6:53:14,  1.65it/s]
 37%|███▋      | 24509/65536 [4:12:15<6:51:04,  1.66it/s]
 37%|███▋      | 24510/65536 [4:12:16<7:19:56,  1.55it/s]
 37%|███▋      | 24511/65536 [4:12:17<7:22:06,  1.55it/s]
 37%|███▋      | 24512/65536 [4:12:17<7:17:12,  1.56it/s]
 37%|███▋      | 24513/65536 [4:12:18<7:17:13,  1.56it/s]
 37%|███▋      | 24514/65536 [4:12:19<7:09:28,  1.59it/s]
 37%|███▋      | 24515/65536 [4:12:19<7:07:17,  1.60it/s]
 37%|███▋      | 24516/65536 [4:12:20<7:05:41,  1.61it/s]
 37%|███▋      | 24517/65536 [4:12:20<6:54:28,  1.65it/s]
 37%|███▋      | 24518/65536 [4:12:21<6:44:39,  1.69it/s]
 37%|███▋      | 24519/65536 [4:12:22<6:45:58,  1.68it/s]
 37%|███▋      | 24520/65536 [4:12:22<7:02:23,  1.62it/s]
                                                         
{'loss': 2.9727, 'learning_rate': 6.649932655809967e-07, 'epoch': 1513.58}
+
 37%|███▋      | 24520/65536 [4:12:22<7:02:23,  1.62it/s]
 37%|███▋      | 24521/65536 [4:12:23<7:01:20,  1.62it/s]
 37%|███▋      | 24522/65536 [4:12:24<6:55:42,  1.64it/s]
 37%|███▋      | 24523/65536 [4:12:24<6:57:04,  1.64it/s]
 37%|███▋      | 24524/65536 [4:12:25<7:05:16,  1.61it/s]
 37%|███▋      | 24525/65536 [4:12:25<7:02:41,  1.62it/s]
 37%|███▋      | 24526/65536 [4:12:26<7:03:29,  1.61it/s]
 37%|███▋      | 24527/65536 [4:12:27<7:12:28,  1.58it/s]
 37%|███▋      | 24528/65536 [4:12:27<7:11:47,  1.58it/s]
 37%|███▋      | 24529/65536 [4:12:28<7:10:45,  1.59it/s]
 37%|███▋      | 24530/65536 [4:12:29<7:05:53,  1.60it/s]
 37%|███▋      | 24531/65536 [4:12:29<7:09:27,  1.59it/s]
 37%|███▋      | 24532/65536 [4:12:30<7:04:37,  1.61it/s]
 37%|███▋      | 24533/65536 [4:12:30<7:02:13,  1.62it/s]
 37%|███▋      | 24534/65536 [4:12:31<7:14:50,  1.57it/s]
 37%|███▋      | 24535/65536 [4:12:32<7:10:52,  1.59it/s]
 37%|███▋      | 24536/65536 [4:12:32<7:04:11,  1.61it/s]
 37%|███▋      | 24537/65536 [4:12:33<6:53:42,  1.65it/s]
 37%|███▋      | 24538/65536 [4:12:33<6:58:25,  1.63it/s]
 37%|███▋      | 24539/65536 [4:12:34<7:04:26,  1.61it/s]
 37%|███▋      | 24540/65536 [4:12:35<6:59:34,  1.63it/s]
                                                         
{'loss': 2.9699, 'learning_rate': 6.647177666217706e-07, 'epoch': 1514.81}
+
 37%|███▋      | 24540/65536 [4:12:35<6:59:34,  1.63it/s]
 37%|███▋      | 24541/65536 [4:12:35<6:56:02,  1.64it/s]
 37%|███▋      | 24542/65536 [4:12:36<6:56:16,  1.64it/s]
 37%|███▋      | 24543/65536 [4:12:37<7:01:13,  1.62it/s]
 37%|███▋      | 24544/65536 [4:12:37<7:25:59,  1.53it/s]
 37%|███▋      | 24545/65536 [4:12:38<7:06:27,  1.60it/s]
 37%|███▋      | 24546/65536 [4:12:38<7:07:10,  1.60it/s]
 37%|███▋      | 24547/65536 [4:12:39<6:59:44,  1.63it/s]
 37%|███▋      | 24548/65536 [4:12:40<6:53:31,  1.65it/s]
 37%|███▋      | 24549/65536 [4:12:40<7:01:13,  1.62it/s]
 37%|███▋      | 24550/65536 [4:12:41<6:53:32,  1.65it/s]
 37%|███▋      | 24551/65536 [4:12:41<6:49:30,  1.67it/s]
 37%|███▋      | 24552/65536 [4:12:42<7:12:16,  1.58it/s]
 37%|███▋      | 24553/65536 [4:12:43<7:26:30,  1.53it/s]
 37%|███▋      | 24554/65536 [4:12:43<7:13:07,  1.58it/s]
 37%|███▋      | 24555/65536 [4:12:44<7:08:12,  1.60it/s]
 37%|███▋      | 24556/65536 [4:12:45<7:03:10,  1.61it/s]
 37%|███▋      | 24557/65536 [4:12:45<7:12:18,  1.58it/s]
 37%|███▋      | 24558/65536 [4:12:46<7:11:09,  1.58it/s]
 37%|███▋      | 24559/65536 [4:12:47<7:08:16,  1.59it/s]
 37%|███▋      | 24560/65536 [4:12:47<7:20:21,  1.55it/s]
                                                         
{'loss': 2.9626, 'learning_rate': 6.644422676625444e-07, 'epoch': 1516.05}
+
 37%|███▋      | 24560/65536 [4:12:47<7:20:21,  1.55it/s]
 37%|███▋      | 24561/65536 [4:12:48<7:11:34,  1.58it/s]
 37%|███▋      | 24562/65536 [4:12:49<7:13:38,  1.57it/s]
 37%|███▋      | 24563/65536 [4:12:49<7:08:28,  1.59it/s]
 37%|███▋      | 24564/65536 [4:12:50<6:53:22,  1.65it/s]
 37%|███▋      | 24565/65536 [4:12:50<6:46:24,  1.68it/s]
 37%|███▋      | 24566/65536 [4:12:51<6:54:56,  1.65it/s]
 37%|███▋      | 24567/65536 [4:12:51<6:54:48,  1.65it/s]
 37%|███▋      | 24568/65536 [4:12:52<6:52:41,  1.65it/s]
 37%|███▋      | 24569/65536 [4:12:53<6:47:37,  1.68it/s]
 37%|███▋      | 24570/65536 [4:12:53<6:52:12,  1.66it/s]
 37%|███▋      | 24571/65536 [4:12:54<6:59:16,  1.63it/s]
 37%|███▋      | 24572/65536 [4:12:55<6:54:21,  1.65it/s]
 37%|███▋      | 24573/65536 [4:12:55<7:08:29,  1.59it/s]
 37%|███▋      | 24574/65536 [4:12:56<7:11:43,  1.58it/s]
 37%|███▋      | 24575/65536 [4:12:56<7:01:37,  1.62it/s]
 38%|███▊      | 24576/65536 [4:12:57<7:19:25,  1.55it/s]
 38%|███▊      | 24577/65536 [4:12:58<7:25:01,  1.53it/s]
 38%|███▊      | 24578/65536 [4:12:58<7:11:27,  1.58it/s]
 38%|███▊      | 24579/65536 [4:12:59<7:08:14,  1.59it/s]
 38%|███▊      | 24580/65536 [4:13:00<7:13:15,  1.58it/s]
                                                         
{'loss': 3.0035, 'learning_rate': 6.641667687033182e-07, 'epoch': 1517.28}
+
 38%|███▊      | 24580/65536 [4:13:00<7:13:15,  1.58it/s]
 38%|███▊      | 24581/65536 [4:13:00<7:07:14,  1.60it/s]
 38%|███▊      | 24582/65536 [4:13:01<7:00:03,  1.62it/s]
 38%|███▊      | 24583/65536 [4:13:01<6:57:48,  1.63it/s]
 38%|███▊      | 24584/65536 [4:13:02<6:58:51,  1.63it/s]
 38%|███▊      | 24585/65536 [4:13:03<6:58:36,  1.63it/s]
 38%|███▊      | 24586/65536 [4:13:03<6:50:26,  1.66it/s]
 38%|███▊      | 24587/65536 [4:13:04<7:01:12,  1.62it/s]
 38%|███▊      | 24588/65536 [4:13:05<7:05:41,  1.60it/s]
 38%|███▊      | 24589/65536 [4:13:05<6:59:46,  1.63it/s]
 38%|███▊      | 24590/65536 [4:13:06<7:01:59,  1.62it/s]
 38%|███▊      | 24591/65536 [4:13:06<6:57:53,  1.63it/s]
 38%|███▊      | 24592/65536 [4:13:07<7:06:33,  1.60it/s]
 38%|███▊      | 24593/65536 [4:13:08<7:10:51,  1.58it/s]
 38%|███▊      | 24594/65536 [4:13:08<7:14:18,  1.57it/s]
 38%|███▊      | 24595/65536 [4:13:09<7:03:04,  1.61it/s]
 38%|███▊      | 24596/65536 [4:13:09<6:57:15,  1.64it/s]
 38%|███▊      | 24597/65536 [4:13:10<7:03:58,  1.61it/s]
 38%|███▊      | 24598/65536 [4:13:11<7:05:24,  1.60it/s]
 38%|███▊      | 24599/65536 [4:13:11<6:58:51,  1.63it/s]
 38%|███▊      | 24600/65536 [4:13:12<7:08:28,  1.59it/s]
                                                         
{'loss': 2.9803, 'learning_rate': 6.638912697440921e-07, 'epoch': 1518.52}
+
 38%|███▊      | 24600/65536 [4:13:12<7:08:28,  1.59it/s]
 38%|███▊      | 24601/65536 [4:13:13<6:57:27,  1.63it/s]
 38%|███▊      | 24602/65536 [4:13:13<7:02:18,  1.62it/s]
 38%|███▊      | 24603/65536 [4:13:14<6:51:14,  1.66it/s]
 38%|███▊      | 24604/65536 [4:13:14<6:52:51,  1.65it/s]
 38%|███▊      | 24605/65536 [4:13:15<6:46:52,  1.68it/s]
 38%|███▊      | 24606/65536 [4:13:16<6:46:59,  1.68it/s]
 38%|███▊      | 24607/65536 [4:13:16<6:40:19,  1.70it/s]
 38%|███▊      | 24608/65536 [4:13:17<6:57:22,  1.63it/s]
 38%|███▊      | 24609/65536 [4:13:17<7:00:12,  1.62it/s]
 38%|███▊      | 24610/65536 [4:13:18<6:54:02,  1.65it/s]
 38%|███▊      | 24611/65536 [4:13:19<6:52:35,  1.65it/s]
 38%|███▊      | 24612/65536 [4:13:19<6:55:48,  1.64it/s]
 38%|███▊      | 24613/65536 [4:13:20<7:01:12,  1.62it/s]
 38%|███▊      | 24614/65536 [4:13:21<7:07:52,  1.59it/s]
 38%|███▊      | 24615/65536 [4:13:21<7:04:31,  1.61it/s]
 38%|███▊      | 24616/65536 [4:13:22<6:59:23,  1.63it/s]
 38%|███▊      | 24617/65536 [4:13:22<7:00:41,  1.62it/s]
 38%|███▊      | 24618/65536 [4:13:23<6:59:52,  1.62it/s]
 38%|███▊      | 24619/65536 [4:13:24<6:45:01,  1.68it/s]
 38%|███▊      | 24620/65536 [4:13:24<6:50:49,  1.66it/s]
                                                         
{'loss': 2.9758, 'learning_rate': 6.636157707848659e-07, 'epoch': 1519.75}
+
 38%|███▊      | 24620/65536 [4:13:24<6:50:49,  1.66it/s]
 38%|███▊      | 24621/65536 [4:13:25<6:53:19,  1.65it/s]
 38%|███▊      | 24622/65536 [4:13:25<7:06:14,  1.60it/s]
 38%|███▊      | 24623/65536 [4:13:26<7:13:02,  1.57it/s]
 38%|███▊      | 24624/65536 [4:13:27<7:02:55,  1.61it/s]
 38%|███▊      | 24625/65536 [4:13:27<7:16:33,  1.56it/s]
 38%|███▊      | 24626/65536 [4:13:28<7:13:45,  1.57it/s]
 38%|███▊      | 24627/65536 [4:13:29<7:04:39,  1.61it/s]
 38%|███▊      | 24628/65536 [4:13:29<6:57:15,  1.63it/s]
 38%|███▊      | 24629/65536 [4:13:30<6:52:51,  1.65it/s]
 38%|███▊      | 24630/65536 [4:13:30<7:10:14,  1.58it/s]
 38%|███▊      | 24631/65536 [4:13:31<6:56:49,  1.64it/s]
 38%|███▊      | 24632/65536 [4:13:32<6:55:49,  1.64it/s]
 38%|███▊      | 24633/65536 [4:13:32<6:55:17,  1.64it/s]
 38%|███▊      | 24634/65536 [4:13:33<6:51:11,  1.66it/s]
 38%|███▊      | 24635/65536 [4:13:33<6:55:48,  1.64it/s]
 38%|███▊      | 24636/65536 [4:13:34<7:04:09,  1.61it/s]
 38%|███▊      | 24637/65536 [4:13:35<6:58:51,  1.63it/s]
 38%|███▊      | 24638/65536 [4:13:35<7:05:00,  1.60it/s]
 38%|███▊      | 24639/65536 [4:13:36<7:00:27,  1.62it/s]
 38%|███▊      | 24640/65536 [4:13:37<6:55:12,  1.64it/s]
                                                         
{'loss': 2.9671, 'learning_rate': 6.633402718256397e-07, 'epoch': 1520.99}
+
 38%|███▊      | 24640/65536 [4:13:37<6:55:12,  1.64it/s]
 38%|███▊      | 24641/65536 [4:13:37<7:17:12,  1.56it/s]
 38%|███▊      | 24642/65536 [4:13:38<7:14:10,  1.57it/s]
 38%|███▊      | 24643/65536 [4:13:38<7:02:44,  1.61it/s]
 38%|███▊      | 24644/65536 [4:13:39<6:57:33,  1.63it/s]
 38%|███▊      | 24645/65536 [4:13:40<7:10:35,  1.58it/s]
 38%|███▊      | 24646/65536 [4:13:40<7:23:15,  1.54it/s]
 38%|███▊      | 24647/65536 [4:13:41<7:16:20,  1.56it/s]
 38%|███▊      | 24648/65536 [4:13:42<7:07:19,  1.59it/s]
 38%|███▊      | 24649/65536 [4:13:42<7:00:34,  1.62it/s]
 38%|███▊      | 24650/65536 [4:13:43<6:55:27,  1.64it/s]
 38%|███▊      | 24651/65536 [4:13:43<6:54:18,  1.64it/s]
 38%|███▊      | 24652/65536 [4:13:44<6:54:25,  1.64it/s]
 38%|███▊      | 24653/65536 [4:13:45<6:55:08,  1.64it/s]
 38%|███▊      | 24654/65536 [4:13:45<6:54:38,  1.64it/s]
 38%|███▊      | 24655/65536 [4:13:46<6:48:25,  1.67it/s]
 38%|███▊      | 24656/65536 [4:13:46<6:41:56,  1.70it/s]
 38%|███▊      | 24657/65536 [4:13:47<6:56:38,  1.64it/s]
 38%|███▊      | 24658/65536 [4:13:48<7:00:42,  1.62it/s]
 38%|███▊      | 24659/65536 [4:13:48<7:01:01,  1.62it/s]
 38%|███▊      | 24660/65536 [4:13:49<7:07:54,  1.59it/s]
                                                         
{'loss': 3.0017, 'learning_rate': 6.630647728664134e-07, 'epoch': 1522.22}
+
 38%|███▊      | 24660/65536 [4:13:49<7:07:54,  1.59it/s]
 38%|███▊      | 24661/65536 [4:13:50<7:03:45,  1.61it/s]
 38%|███▊      | 24662/65536 [4:13:50<6:56:07,  1.64it/s]
 38%|███▊      | 24663/65536 [4:13:51<6:51:15,  1.66it/s]
 38%|███▊      | 24664/65536 [4:13:51<6:49:29,  1.66it/s]
 38%|███▊      | 24665/65536 [4:13:52<6:38:43,  1.71it/s]
 38%|███▊      | 24666/65536 [4:13:53<7:01:09,  1.62it/s]
 38%|███▊      | 24667/65536 [4:13:53<6:52:07,  1.65it/s]
 38%|███▊      | 24668/65536 [4:13:54<6:54:02,  1.65it/s]
 38%|███▊      | 24669/65536 [4:13:54<6:48:29,  1.67it/s]
 38%|███▊      | 24670/65536 [4:13:55<7:02:24,  1.61it/s]
 38%|███▊      | 24671/65536 [4:13:56<7:01:45,  1.61it/s]
 38%|███▊      | 24672/65536 [4:13:56<6:57:57,  1.63it/s]
 38%|███▊      | 24673/65536 [4:13:57<7:06:43,  1.60it/s]
 38%|███▊      | 24674/65536 [4:13:57<6:59:27,  1.62it/s]
 38%|███▊      | 24675/65536 [4:13:58<7:17:02,  1.56it/s]
 38%|███▊      | 24676/65536 [4:13:59<7:15:17,  1.56it/s]
 38%|███▊      | 24677/65536 [4:13:59<7:13:23,  1.57it/s]
 38%|███▊      | 24678/65536 [4:14:00<6:59:48,  1.62it/s]
 38%|███▊      | 24679/65536 [4:14:01<6:51:57,  1.65it/s]
 38%|███▊      | 24680/65536 [4:14:01<6:50:19,  1.66it/s]
                                                         
{'loss': 2.9997, 'learning_rate': 6.627892739071874e-07, 'epoch': 1523.46}
+
 38%|███▊      | 24680/65536 [4:14:01<6:50:19,  1.66it/s]
 38%|███▊      | 24681/65536 [4:14:02<6:50:45,  1.66it/s]
 38%|███▊      | 24682/65536 [4:14:02<6:57:12,  1.63it/s]
 38%|███▊      | 24683/65536 [4:14:03<7:09:44,  1.58it/s]
 38%|███▊      | 24684/65536 [4:14:04<6:59:48,  1.62it/s]
 38%|███▊      | 24685/65536 [4:14:04<7:05:00,  1.60it/s]
 38%|███▊      | 24686/65536 [4:14:05<7:00:25,  1.62it/s]
 38%|███▊      | 24687/65536 [4:14:06<6:55:33,  1.64it/s]
 38%|███▊      | 24688/65536 [4:14:06<6:59:11,  1.62it/s]
 38%|███▊      | 24689/65536 [4:14:07<7:12:38,  1.57it/s]
 38%|███▊      | 24690/65536 [4:14:07<7:10:06,  1.58it/s]
 38%|███▊      | 24691/65536 [4:14:08<6:59:24,  1.62it/s]
 38%|███▊      | 24692/65536 [4:14:09<7:04:39,  1.60it/s]
 38%|███▊      | 24693/65536 [4:14:09<7:01:54,  1.61it/s]
 38%|███▊      | 24694/65536 [4:14:10<7:05:41,  1.60it/s]
 38%|███▊      | 24695/65536 [4:14:11<6:58:47,  1.63it/s]
 38%|███▊      | 24696/65536 [4:14:11<6:57:06,  1.63it/s]
 38%|███▊      | 24697/65536 [4:14:12<6:58:44,  1.63it/s]
 38%|███▊      | 24698/65536 [4:14:12<6:52:52,  1.65it/s]
 38%|███▊      | 24699/65536 [4:14:13<7:00:20,  1.62it/s]
 38%|███▊      | 24700/65536 [4:14:14<7:04:12,  1.60it/s]
                                                         
{'loss': 2.9376, 'learning_rate': 6.625137749479613e-07, 'epoch': 1524.69}
+
 38%|███▊      | 24700/65536 [4:14:14<7:04:12,  1.60it/s]
 38%|███▊      | 24701/65536 [4:14:14<6:57:48,  1.63it/s]
 38%|███▊      | 24702/65536 [4:14:15<6:56:05,  1.64it/s]
 38%|███▊      | 24703/65536 [4:14:15<7:01:47,  1.61it/s]
 38%|███▊      | 24704/65536 [4:14:16<7:01:36,  1.61it/s]
 38%|███▊      | 24705/65536 [4:14:17<7:03:20,  1.61it/s]
 38%|███▊      | 24706/65536 [4:14:17<7:22:12,  1.54it/s]
 38%|███▊      | 24707/65536 [4:14:18<7:11:38,  1.58it/s]
 38%|███▊      | 24708/65536 [4:14:19<7:03:18,  1.61it/s]
 38%|███▊      | 24709/65536 [4:14:19<6:55:14,  1.64it/s]
 38%|███▊      | 24710/65536 [4:14:20<6:53:57,  1.64it/s]
 38%|███▊      | 24711/65536 [4:14:20<6:53:13,  1.65it/s]
 38%|███▊      | 24712/65536 [4:14:21<7:02:10,  1.61it/s]
 38%|███▊      | 24713/65536 [4:14:22<7:03:06,  1.61it/s]
 38%|███▊      | 24714/65536 [4:14:22<7:01:54,  1.61it/s]
 38%|███▊      | 24715/65536 [4:14:23<7:05:37,  1.60it/s]
 38%|███▊      | 24716/65536 [4:14:24<7:10:45,  1.58it/s]
 38%|███▊      | 24717/65536 [4:14:24<7:05:42,  1.60it/s]
 38%|███▊      | 24718/65536 [4:14:25<7:02:23,  1.61it/s]
 38%|███▊      | 24719/65536 [4:14:25<7:01:43,  1.61it/s]
 38%|███▊      | 24720/65536 [4:14:26<7:00:41,  1.62it/s]
                                                         
{'loss': 2.9338, 'learning_rate': 6.622382759887351e-07, 'epoch': 1525.93}
+
 38%|███▊      | 24720/65536 [4:14:26<7:00:41,  1.62it/s]
 38%|███▊      | 24721/65536 [4:14:27<6:53:42,  1.64it/s]
 38%|███▊      | 24722/65536 [4:14:27<7:05:19,  1.60it/s]
 38%|███▊      | 24723/65536 [4:14:28<7:12:48,  1.57it/s]
 38%|███▊      | 24724/65536 [4:14:29<7:04:40,  1.60it/s]
 38%|███▊      | 24725/65536 [4:14:29<6:57:02,  1.63it/s]
 38%|███▊      | 24726/65536 [4:14:30<7:04:42,  1.60it/s]
 38%|███▊      | 24727/65536 [4:14:30<7:05:43,  1.60it/s]
 38%|███▊      | 24728/65536 [4:14:31<7:05:09,  1.60it/s]
 38%|███▊      | 24729/65536 [4:14:32<7:05:40,  1.60it/s]
 38%|███▊      | 24730/65536 [4:14:32<7:01:08,  1.61it/s]
 38%|███▊      | 24731/65536 [4:14:33<6:51:41,  1.65it/s]
 38%|███▊      | 24732/65536 [4:14:33<6:39:27,  1.70it/s]
 38%|███▊      | 24733/65536 [4:14:34<6:52:03,  1.65it/s]
 38%|███▊      | 24734/65536 [4:14:35<6:47:57,  1.67it/s]
 38%|███▊      | 24735/65536 [4:14:35<6:47:10,  1.67it/s]
 38%|███▊      | 24736/65536 [4:14:36<6:46:21,  1.67it/s]
 38%|███▊      | 24737/65536 [4:14:36<6:40:47,  1.70it/s]
 38%|███▊      | 24738/65536 [4:14:37<6:59:29,  1.62it/s]
 38%|███▊      | 24739/65536 [4:14:38<7:04:22,  1.60it/s]
 38%|███▊      | 24740/65536 [4:14:38<6:54:32,  1.64it/s]
                                                         
{'loss': 3.0424, 'learning_rate': 6.61962777029509e-07, 'epoch': 1527.16}
+
 38%|███▊      | 24740/65536 [4:14:38<6:54:32,  1.64it/s]
 38%|███▊      | 24741/65536 [4:14:39<6:56:49,  1.63it/s]
 38%|███▊      | 24742/65536 [4:14:40<7:03:58,  1.60it/s]
 38%|███▊      | 24743/65536 [4:14:40<7:12:20,  1.57it/s]
 38%|███▊      | 24744/65536 [4:14:41<7:16:00,  1.56it/s]
 38%|███▊      | 24745/65536 [4:14:41<7:10:52,  1.58it/s]
 38%|███▊      | 24746/65536 [4:14:42<7:04:30,  1.60it/s]
 38%|███▊      | 24747/65536 [4:14:43<6:57:38,  1.63it/s]
 38%|███▊      | 24748/65536 [4:14:43<6:54:58,  1.64it/s]
 38%|███▊      | 24749/65536 [4:14:44<6:52:58,  1.65it/s]
 38%|███▊      | 24750/65536 [4:14:44<6:44:00,  1.68it/s]
 38%|███▊      | 24751/65536 [4:14:45<6:38:05,  1.71it/s]
 38%|███▊      | 24752/65536 [4:14:46<6:47:21,  1.67it/s]
 38%|███▊      | 24753/65536 [4:14:46<6:53:18,  1.64it/s]
 38%|███▊      | 24754/65536 [4:14:47<7:18:00,  1.55it/s]
 38%|███▊      | 24755/65536 [4:14:48<7:09:09,  1.58it/s]
 38%|███▊      | 24756/65536 [4:14:48<7:04:11,  1.60it/s]
 38%|███▊      | 24757/65536 [4:14:49<6:55:53,  1.63it/s]
 38%|███▊      | 24758/65536 [4:14:49<6:45:12,  1.68it/s]
 38%|███▊      | 24759/65536 [4:14:50<6:47:25,  1.67it/s]
 38%|███▊      | 24760/65536 [4:14:51<6:56:45,  1.63it/s]
                                                         
{'loss': 3.0168, 'learning_rate': 6.616872780702828e-07, 'epoch': 1528.4}
+
 38%|███▊      | 24760/65536 [4:14:51<6:56:45,  1.63it/s]
 38%|███▊      | 24761/65536 [4:14:51<6:48:10,  1.66it/s]
 38%|███▊      | 24762/65536 [4:14:52<7:04:18,  1.60it/s]
 38%|███▊      | 24763/65536 [4:14:52<6:52:58,  1.65it/s]
 38%|███▊      | 24764/65536 [4:14:53<6:51:37,  1.65it/s]
 38%|███▊      | 24765/65536 [4:14:54<6:43:54,  1.68it/s]
 38%|███▊      | 24766/65536 [4:14:54<6:52:12,  1.65it/s]
 38%|███▊      | 24767/65536 [4:14:55<6:50:05,  1.66it/s]
 38%|███▊      | 24768/65536 [4:14:55<7:02:30,  1.61it/s]
 38%|███▊      | 24769/65536 [4:14:56<7:07:01,  1.59it/s]
 38%|███▊      | 24770/65536 [4:14:57<7:10:22,  1.58it/s]
 38%|███▊      | 24771/65536 [4:14:57<7:03:27,  1.60it/s]
 38%|███▊      | 24772/65536 [4:14:58<6:59:36,  1.62it/s]
 38%|███▊      | 24773/65536 [4:14:59<6:57:11,  1.63it/s]
 38%|███▊      | 24774/65536 [4:14:59<6:51:57,  1.65it/s]
 38%|███▊      | 24775/65536 [4:15:00<6:46:39,  1.67it/s]
 38%|███▊      | 24776/65536 [4:15:00<6:48:20,  1.66it/s]
 38%|███▊      | 24777/65536 [4:15:01<6:40:21,  1.70it/s]
 38%|███▊      | 24778/65536 [4:15:02<6:59:17,  1.62it/s]
 38%|███▊      | 24779/65536 [4:15:02<6:49:42,  1.66it/s]
 38%|███▊      | 24780/65536 [4:15:03<6:45:47,  1.67it/s]
                                                         
{'loss': 2.9788, 'learning_rate': 6.614117791110567e-07, 'epoch': 1529.63}
+
 38%|███▊      | 24780/65536 [4:15:03<6:45:47,  1.67it/s]
 38%|███▊      | 24781/65536 [4:15:03<6:54:37,  1.64it/s]
 38%|███▊      | 24782/65536 [4:15:04<7:00:36,  1.61it/s]
 38%|███▊      | 24783/65536 [4:15:05<6:59:58,  1.62it/s]
 38%|███▊      | 24784/65536 [4:15:05<6:50:34,  1.65it/s]
 38%|███▊      | 24785/65536 [4:15:06<7:03:57,  1.60it/s]
 38%|███▊      | 24786/65536 [4:15:07<7:06:52,  1.59it/s]
 38%|███▊      | 24787/65536 [4:15:07<7:12:42,  1.57it/s]
 38%|███▊      | 24788/65536 [4:15:08<7:09:38,  1.58it/s]
 38%|███▊      | 24789/65536 [4:15:08<7:07:23,  1.59it/s]
 38%|███▊      | 24790/65536 [4:15:09<6:52:38,  1.65it/s]
 38%|███▊      | 24791/65536 [4:15:10<6:53:39,  1.64it/s]
 38%|███▊      | 24792/65536 [4:15:10<6:55:16,  1.64it/s]
 38%|███▊      | 24793/65536 [4:15:11<7:07:22,  1.59it/s]
 38%|███▊      | 24794/65536 [4:15:12<7:09:07,  1.58it/s]
 38%|███▊      | 24795/65536 [4:15:12<7:08:19,  1.59it/s]
 38%|███▊      | 24796/65536 [4:15:13<7:00:26,  1.61it/s]
 38%|███▊      | 24797/65536 [4:15:13<6:57:18,  1.63it/s]
 38%|███▊      | 24798/65536 [4:15:14<7:02:44,  1.61it/s]
 38%|███▊      | 24799/65536 [4:15:15<7:03:58,  1.60it/s]
 38%|███▊      | 24800/65536 [4:15:15<6:56:35,  1.63it/s]
                                                         
{'loss': 2.968, 'learning_rate': 6.611362801518306e-07, 'epoch': 1530.86}
+
 38%|███▊      | 24800/65536 [4:15:15<6:56:35,  1.63it/s]
 38%|███▊      | 24801/65536 [4:15:16<6:50:13,  1.65it/s]
 38%|███▊      | 24802/65536 [4:15:16<6:40:26,  1.70it/s]
 38%|███▊      | 24803/65536 [4:15:17<6:53:54,  1.64it/s]
 38%|███▊      | 24804/65536 [4:15:18<7:02:06,  1.61it/s]
 38%|███▊      | 24805/65536 [4:15:18<7:02:16,  1.61it/s]
 38%|███▊      | 24806/65536 [4:15:19<6:52:11,  1.65it/s]
 38%|███▊      | 24807/65536 [4:15:19<6:50:10,  1.65it/s]
 38%|███▊      | 24808/65536 [4:15:20<6:52:59,  1.64it/s]
 38%|███▊      | 24809/65536 [4:15:21<6:54:03,  1.64it/s]
 38%|███▊      | 24810/65536 [4:15:21<7:08:17,  1.58it/s]
 38%|███▊      | 24811/65536 [4:15:22<6:58:07,  1.62it/s]
 38%|███▊      | 24812/65536 [4:15:23<6:54:06,  1.64it/s]
 38%|███▊      | 24813/65536 [4:15:23<6:59:11,  1.62it/s]
 38%|███▊      | 24814/65536 [4:15:24<7:00:26,  1.61it/s]
 38%|███▊      | 24815/65536 [4:15:24<6:51:41,  1.65it/s]
 38%|███▊      | 24816/65536 [4:15:25<7:00:57,  1.61it/s]
 38%|███▊      | 24817/65536 [4:15:26<6:59:40,  1.62it/s]
 38%|███▊      | 24818/65536 [4:15:26<6:50:47,  1.65it/s]
 38%|███▊      | 24819/65536 [4:15:27<6:59:11,  1.62it/s]
 38%|███▊      | 24820/65536 [4:15:27<6:59:17,  1.62it/s]
                                                         
{'loss': 3.0261, 'learning_rate': 6.608607811926044e-07, 'epoch': 1532.1}
+
 38%|███▊      | 24820/65536 [4:15:27<6:59:17,  1.62it/s]
 38%|███▊      | 24821/65536 [4:15:28<6:59:58,  1.62it/s]
 38%|███▊      | 24822/65536 [4:15:29<6:51:10,  1.65it/s]
 38%|███▊      | 24823/65536 [4:15:29<6:48:34,  1.66it/s]
 38%|███▊      | 24824/65536 [4:15:30<6:48:42,  1.66it/s]
 38%|███▊      | 24825/65536 [4:15:30<6:48:15,  1.66it/s]
 38%|███▊      | 24826/65536 [4:15:31<6:46:23,  1.67it/s]
 38%|███▊      | 24827/65536 [4:15:32<6:52:01,  1.65it/s]
 38%|███▊      | 24828/65536 [4:15:32<6:52:58,  1.64it/s]
 38%|███▊      | 24829/65536 [4:15:33<6:49:34,  1.66it/s]
 38%|███▊      | 24830/65536 [4:15:34<6:59:07,  1.62it/s]
 38%|███▊      | 24831/65536 [4:15:34<6:52:00,  1.65it/s]
 38%|███▊      | 24832/65536 [4:15:35<6:51:35,  1.65it/s]
 38%|███▊      | 24833/65536 [4:15:35<6:59:40,  1.62it/s]
 38%|███▊      | 24834/65536 [4:15:36<6:55:35,  1.63it/s]
 38%|███▊      | 24835/65536 [4:15:37<7:07:43,  1.59it/s]
 38%|███▊      | 24836/65536 [4:15:37<6:57:25,  1.63it/s]
 38%|███▊      | 24837/65536 [4:15:38<6:52:56,  1.64it/s]
 38%|███▊      | 24838/65536 [4:15:38<6:46:24,  1.67it/s]
 38%|███▊      | 24839/65536 [4:15:39<6:47:07,  1.67it/s]
 38%|███▊      | 24840/65536 [4:15:40<6:38:22,  1.70it/s]
                                                         
{'loss': 3.0065, 'learning_rate': 6.605852822333782e-07, 'epoch': 1533.33}
+
 38%|███▊      | 24840/65536 [4:15:40<6:38:22,  1.70it/s]
 38%|███▊      | 24841/65536 [4:15:40<6:44:36,  1.68it/s]
 38%|███▊      | 24842/65536 [4:15:41<6:53:29,  1.64it/s]
 38%|███▊      | 24843/65536 [4:15:41<6:53:30,  1.64it/s]
 38%|███▊      | 24844/65536 [4:15:42<6:48:39,  1.66it/s]
 38%|███▊      | 24845/65536 [4:15:43<6:46:58,  1.67it/s]
 38%|███▊      | 24846/65536 [4:15:43<6:51:09,  1.65it/s]
 38%|███▊      | 24847/65536 [4:15:44<6:58:50,  1.62it/s]
 38%|███▊      | 24848/65536 [4:15:45<7:05:03,  1.60it/s]
 38%|███▊      | 24849/65536 [4:15:45<6:55:34,  1.63it/s]
 38%|███▊      | 24850/65536 [4:15:46<7:08:19,  1.58it/s]
 38%|███▊      | 24851/65536 [4:15:46<7:29:19,  1.51it/s]
 38%|███▊      | 24852/65536 [4:15:47<7:22:49,  1.53it/s]
 38%|███▊      | 24853/65536 [4:15:48<7:17:21,  1.55it/s]
 38%|███▊      | 24854/65536 [4:15:48<7:06:10,  1.59it/s]
 38%|███▊      | 24855/65536 [4:15:49<7:05:05,  1.59it/s]
 38%|███▊      | 24856/65536 [4:15:50<6:59:36,  1.62it/s]
 38%|███▊      | 24857/65536 [4:15:50<7:04:01,  1.60it/s]
 38%|███▊      | 24858/65536 [4:15:51<7:03:29,  1.60it/s]
 38%|███▊      | 24859/65536 [4:15:51<6:55:55,  1.63it/s]
 38%|███▊      | 24860/65536 [4:15:52<6:54:51,  1.63it/s]
                                                         
{'loss': 2.9672, 'learning_rate': 6.60309783274152e-07, 'epoch': 1534.57}
+
 38%|███▊      | 24860/65536 [4:15:52<6:54:51,  1.63it/s]
 38%|███▊      | 24861/65536 [4:15:53<7:00:07,  1.61it/s]
 38%|███▊      | 24862/65536 [4:15:53<7:00:07,  1.61it/s]
 38%|███▊      | 24863/65536 [4:15:54<6:55:26,  1.63it/s]
 38%|███▊      | 24864/65536 [4:15:54<6:55:41,  1.63it/s]
 38%|███▊      | 24865/65536 [4:15:55<6:55:38,  1.63it/s]
 38%|███▊      | 24866/65536 [4:15:56<6:57:46,  1.62it/s]
 38%|███▊      | 24867/65536 [4:15:56<6:49:33,  1.65it/s]
 38%|███▊      | 24868/65536 [4:15:57<7:01:31,  1.61it/s]
 38%|███▊      | 24869/65536 [4:15:58<6:59:53,  1.61it/s]
 38%|███▊      | 24870/65536 [4:15:58<6:54:45,  1.63it/s]
 38%|███▊      | 24871/65536 [4:15:59<6:52:49,  1.64it/s]
 38%|███▊      | 24872/65536 [4:15:59<6:55:06,  1.63it/s]
 38%|███▊      | 24873/65536 [4:16:00<6:58:50,  1.62it/s]
 38%|███▊      | 24874/65536 [4:16:01<6:56:38,  1.63it/s]
 38%|███▊      | 24875/65536 [4:16:01<6:57:11,  1.62it/s]
 38%|███▊      | 24876/65536 [4:16:02<7:08:17,  1.58it/s]
 38%|███▊      | 24877/65536 [4:16:03<7:00:56,  1.61it/s]
 38%|███▊      | 24878/65536 [4:16:03<6:50:46,  1.65it/s]
 38%|███▊      | 24879/65536 [4:16:04<6:53:28,  1.64it/s]
 38%|███▊      | 24880/65536 [4:16:04<6:43:11,  1.68it/s]
                                                         
{'loss': 2.9691, 'learning_rate': 6.600342843149259e-07, 'epoch': 1535.8}
+
 38%|███▊      | 24880/65536 [4:16:04<6:43:11,  1.68it/s]
 38%|███▊      | 24881/65536 [4:16:05<6:42:56,  1.68it/s]
 38%|███▊      | 24882/65536 [4:16:06<6:54:10,  1.64it/s]
 38%|███▊      | 24883/65536 [4:16:06<7:00:51,  1.61it/s]
 38%|███▊      | 24884/65536 [4:16:07<7:10:48,  1.57it/s]
 38%|███▊      | 24885/65536 [4:16:07<7:00:38,  1.61it/s]
 38%|███▊      | 24886/65536 [4:16:08<6:54:17,  1.64it/s]
 38%|███▊      | 24887/65536 [4:16:09<6:49:19,  1.66it/s]
 38%|███▊      | 24888/65536 [4:16:09<6:47:49,  1.66it/s]
 38%|███▊      | 24889/65536 [4:16:10<6:50:32,  1.65it/s]
 38%|███▊      | 24890/65536 [4:16:10<6:54:07,  1.64it/s]
 38%|███▊      | 24891/65536 [4:16:11<6:50:12,  1.65it/s]
 38%|███▊      | 24892/65536 [4:16:12<6:58:04,  1.62it/s]
 38%|███▊      | 24893/65536 [4:16:12<6:44:40,  1.67it/s]
 38%|███▊      | 24894/65536 [4:16:13<6:43:57,  1.68it/s]
 38%|███▊      | 24895/65536 [4:16:13<6:41:29,  1.69it/s]
 38%|███▊      | 24896/65536 [4:16:14<6:35:39,  1.71it/s]
 38%|███▊      | 24897/65536 [4:16:15<6:52:35,  1.64it/s]
 38%|███▊      | 24898/65536 [4:16:15<7:11:54,  1.57it/s]
 38%|███▊      | 24899/65536 [4:16:16<7:10:19,  1.57it/s]
 38%|███▊      | 24900/65536 [4:16:17<7:25:29,  1.52it/s]
                                                         
{'loss': 2.9562, 'learning_rate': 6.597587853556996e-07, 'epoch': 1537.04}
+
 38%|███▊      | 24900/65536 [4:16:17<7:25:29,  1.52it/s]
 38%|███▊      | 24901/65536 [4:16:17<7:10:00,  1.57it/s]
 38%|███▊      | 24902/65536 [4:16:18<6:57:43,  1.62it/s]
 38%|███▊      | 24903/65536 [4:16:18<6:57:32,  1.62it/s]
 38%|███▊      | 24904/65536 [4:16:19<6:53:52,  1.64it/s]
 38%|███▊      | 24905/65536 [4:16:20<6:48:50,  1.66it/s]
 38%|███▊      | 24906/65536 [4:16:20<6:54:21,  1.63it/s]
 38%|███▊      | 24907/65536 [4:16:21<6:47:08,  1.66it/s]
 38%|███▊      | 24908/65536 [4:16:21<6:46:31,  1.67it/s]
 38%|███▊      | 24909/65536 [4:16:22<6:46:05,  1.67it/s]
 38%|███▊      | 24910/65536 [4:16:23<6:50:09,  1.65it/s]
 38%|███▊      | 24911/65536 [4:16:23<6:47:51,  1.66it/s]
 38%|███▊      | 24912/65536 [4:16:24<6:46:38,  1.66it/s]
 38%|███▊      | 24913/65536 [4:16:25<7:03:18,  1.60it/s]
 38%|███▊      | 24914/65536 [4:16:25<7:09:54,  1.57it/s]
 38%|███▊      | 24915/65536 [4:16:26<7:04:15,  1.60it/s]
 38%|███▊      | 24916/65536 [4:16:26<7:07:33,  1.58it/s]
 38%|███▊      | 24917/65536 [4:16:27<7:13:47,  1.56it/s]
 38%|███▊      | 24918/65536 [4:16:28<7:11:19,  1.57it/s]
 38%|███▊      | 24919/65536 [4:16:28<6:57:19,  1.62it/s]
 38%|███▊      | 24920/65536 [4:16:29<6:54:54,  1.63it/s]
                                                         
{'loss': 2.9616, 'learning_rate': 6.594832863964735e-07, 'epoch': 1538.27}
+
 38%|███▊      | 24920/65536 [4:16:29<6:54:54,  1.63it/s]
 38%|███▊      | 24921/65536 [4:16:30<6:54:22,  1.63it/s]
 38%|███▊      | 24922/65536 [4:16:30<6:53:48,  1.64it/s]
 38%|███▊      | 24923/65536 [4:16:31<6:48:36,  1.66it/s]
 38%|███▊      | 24924/65536 [4:16:31<7:01:11,  1.61it/s]
 38%|███▊      | 24925/65536 [4:16:32<6:52:41,  1.64it/s]
 38%|███▊      | 24926/65536 [4:16:33<6:56:49,  1.62it/s]
 38%|███▊      | 24927/65536 [4:16:33<6:54:48,  1.63it/s]
 38%|███▊      | 24928/65536 [4:16:34<6:52:43,  1.64it/s]
 38%|███▊      | 24929/65536 [4:16:34<6:52:46,  1.64it/s]
 38%|███▊      | 24930/65536 [4:16:35<6:46:26,  1.67it/s]
 38%|███▊      | 24931/65536 [4:16:36<6:49:07,  1.65it/s]
 38%|███▊      | 24932/65536 [4:16:36<7:03:59,  1.60it/s]
 38%|███▊      | 24933/65536 [4:16:37<7:10:27,  1.57it/s]
 38%|███▊      | 24934/65536 [4:16:38<7:17:47,  1.55it/s]
 38%|███▊      | 24935/65536 [4:16:38<7:02:26,  1.60it/s]
 38%|███▊      | 24936/65536 [4:16:39<6:48:25,  1.66it/s]
 38%|███▊      | 24937/65536 [4:16:39<6:59:35,  1.61it/s]
 38%|███▊      | 24938/65536 [4:16:40<6:48:51,  1.65it/s]
 38%|███▊      | 24939/65536 [4:16:41<6:48:33,  1.66it/s]
 38%|███▊      | 24940/65536 [4:16:41<6:42:31,  1.68it/s]
                                                         
{'loss': 3.0074, 'learning_rate': 6.592077874372474e-07, 'epoch': 1539.51}
+
 38%|███▊      | 24940/65536 [4:16:41<6:42:31,  1.68it/s]
 38%|███▊      | 24941/65536 [4:16:42<6:39:40,  1.69it/s]
 38%|███▊      | 24942/65536 [4:16:42<6:47:52,  1.66it/s]
 38%|███▊      | 24943/65536 [4:16:43<6:46:00,  1.67it/s]
 38%|███▊      | 24944/65536 [4:16:44<6:56:27,  1.62it/s]
 38%|███▊      | 24945/65536 [4:16:44<6:48:33,  1.66it/s]
 38%|███▊      | 24946/65536 [4:16:45<6:50:52,  1.65it/s]
 38%|███▊      | 24947/65536 [4:16:45<6:50:10,  1.65it/s]
 38%|███▊      | 24948/65536 [4:16:46<6:51:17,  1.64it/s]
 38%|███▊      | 24949/65536 [4:16:47<7:09:18,  1.58it/s]
 38%|███▊      | 24950/65536 [4:16:47<7:07:46,  1.58it/s]
 38%|███▊      | 24951/65536 [4:16:48<7:06:47,  1.58it/s]
 38%|███▊      | 24952/65536 [4:16:49<6:55:41,  1.63it/s]
 38%|███▊      | 24953/65536 [4:16:49<6:58:58,  1.61it/s]
 38%|███▊      | 24954/65536 [4:16:50<6:48:20,  1.66it/s]
 38%|███▊      | 24955/65536 [4:16:50<6:50:41,  1.65it/s]
 38%|███▊      | 24956/65536 [4:16:51<6:50:44,  1.65it/s]
 38%|███▊      | 24957/65536 [4:16:52<6:50:33,  1.65it/s]
 38%|███▊      | 24958/65536 [4:16:52<7:00:54,  1.61it/s]
 38%|███▊      | 24959/65536 [4:16:53<6:53:56,  1.63it/s]
 38%|███▊      | 24960/65536 [4:16:53<6:45:57,  1.67it/s]
                                                         
{'loss': 2.9765, 'learning_rate': 6.589322884780212e-07, 'epoch': 1540.74}
+
 38%|███▊      | 24960/65536 [4:16:53<6:45:57,  1.67it/s]
 38%|███▊      | 24961/65536 [4:16:54<7:00:46,  1.61it/s]
 38%|███▊      | 24962/65536 [4:16:55<6:53:51,  1.63it/s]
 38%|███▊      | 24963/65536 [4:16:55<6:47:45,  1.66it/s]
 38%|███▊      | 24964/65536 [4:16:56<6:55:55,  1.63it/s]
 38%|███▊      | 24965/65536 [4:16:57<7:03:14,  1.60it/s]
 38%|███▊      | 24966/65536 [4:16:57<6:55:45,  1.63it/s]
 38%|███▊      | 24967/65536 [4:16:58<7:03:40,  1.60it/s]
 38%|███▊      | 24968/65536 [4:16:58<6:54:51,  1.63it/s]
 38%|███▊      | 24969/65536 [4:16:59<7:05:41,  1.59it/s]
 38%|███▊      | 24970/65536 [4:17:00<7:16:11,  1.55it/s]
 38%|███▊      | 24971/65536 [4:17:00<7:09:44,  1.57it/s]
 38%|███▊      | 24972/65536 [4:17:01<6:57:33,  1.62it/s]
 38%|███▊      | 24973/65536 [4:17:01<6:49:45,  1.65it/s]
 38%|███▊      | 24974/65536 [4:17:02<6:41:02,  1.69it/s]
 38%|███▊      | 24975/65536 [4:17:03<6:54:52,  1.63it/s]
 38%|███▊      | 24976/65536 [4:17:03<6:53:36,  1.63it/s]
 38%|███▊      | 24977/65536 [4:17:04<6:54:36,  1.63it/s]
 38%|███▊      | 24978/65536 [4:17:04<6:51:02,  1.64it/s]
 38%|███▊      | 24979/65536 [4:17:05<6:53:56,  1.63it/s]
 38%|███▊      | 24980/65536 [4:17:06<6:52:33,  1.64it/s]
                                                         
{'loss': 2.9252, 'learning_rate': 6.586567895187951e-07, 'epoch': 1541.98}
+
 38%|███▊      | 24980/65536 [4:17:06<6:52:33,  1.64it/s]
 38%|███▊      | 24981/65536 [4:17:06<7:09:25,  1.57it/s]
 38%|███▊      | 24982/65536 [4:17:07<6:59:07,  1.61it/s]
 38%|███▊      | 24983/65536 [4:17:08<7:08:00,  1.58it/s]
 38%|███▊      | 24984/65536 [4:17:08<7:04:18,  1.59it/s]
 38%|███▊      | 24985/65536 [4:17:09<7:08:20,  1.58it/s]
 38%|███▊      | 24986/65536 [4:17:10<7:19:40,  1.54it/s]
 38%|███▊      | 24987/65536 [4:17:10<7:07:37,  1.58it/s]
 38%|███▊      | 24988/65536 [4:17:11<7:02:29,  1.60it/s]
 38%|███▊      | 24989/65536 [4:17:11<6:51:11,  1.64it/s]
 38%|███▊      | 24990/65536 [4:17:12<6:42:34,  1.68it/s]
 38%|███▊      | 24991/65536 [4:17:13<6:48:02,  1.66it/s]
 38%|███▊      | 24992/65536 [4:17:13<6:52:28,  1.64it/s]
 38%|███▊      | 24993/65536 [4:17:14<6:46:13,  1.66it/s]
 38%|███▊      | 24994/65536 [4:17:14<6:38:41,  1.69it/s]
 38%|███▊      | 24995/65536 [4:17:15<6:39:18,  1.69it/s]
 38%|███▊      | 24996/65536 [4:17:16<6:45:28,  1.67it/s]
 38%|███▊      | 24997/65536 [4:17:16<6:53:42,  1.63it/s]
 38%|███▊      | 24998/65536 [4:17:17<7:01:35,  1.60it/s]
 38%|███▊      | 24999/65536 [4:17:17<6:51:10,  1.64it/s]
 38%|███▊      | 25000/65536 [4:17:18<6:50:54,  1.64it/s]
                                                         
{'loss': 2.9334, 'learning_rate': 6.583812905595689e-07, 'epoch': 1543.21}
+
 38%|███▊      | 25000/65536 [4:17:18<6:50:54,  1.64it/s]
 38%|███▊      | 25001/65536 [4:17:19<6:49:15,  1.65it/s]
 38%|███▊      | 25002/65536 [4:17:19<6:46:06,  1.66it/s]
 38%|███▊      | 25003/65536 [4:17:20<6:46:36,  1.66it/s]
 38%|███▊      | 25004/65536 [4:17:20<6:50:18,  1.65it/s]
 38%|███▊      | 25005/65536 [4:17:21<6:56:35,  1.62it/s]
 38%|███▊      | 25006/65536 [4:17:22<6:45:42,  1.66it/s]
 38%|███▊      | 25007/65536 [4:17:22<6:57:27,  1.62it/s]
 38%|███▊      | 25008/65536 [4:17:23<6:55:52,  1.62it/s]
 38%|███▊      | 25009/65536 [4:17:23<6:47:33,  1.66it/s]
 38%|███▊      | 25010/65536 [4:17:24<6:52:16,  1.64it/s]
 38%|███▊      | 25011/65536 [4:17:25<6:53:10,  1.63it/s]
 38%|███▊      | 25012/65536 [4:17:25<6:55:54,  1.62it/s]
 38%|███▊      | 25013/65536 [4:17:26<7:20:38,  1.53it/s]
 38%|███▊      | 25014/65536 [4:17:27<7:06:17,  1.58it/s]
 38%|███▊      | 25015/65536 [4:17:27<6:58:09,  1.62it/s]
 38%|███▊      | 25016/65536 [4:17:28<6:52:55,  1.64it/s]
 38%|███▊      | 25017/65536 [4:17:28<6:55:12,  1.63it/s]
 38%|███▊      | 25018/65536 [4:17:29<6:48:32,  1.65it/s]
 38%|███▊      | 25019/65536 [4:17:30<6:53:56,  1.63it/s]
 38%|███▊      | 25020/65536 [4:17:30<6:57:57,  1.62it/s]
                                                         
{'loss': 2.9412, 'learning_rate': 6.581057916003428e-07, 'epoch': 1544.44}
+
 38%|███▊      | 25020/65536 [4:17:30<6:57:57,  1.62it/s]
 38%|███▊      | 25021/65536 [4:17:31<6:54:39,  1.63it/s]
 38%|███▊      | 25022/65536 [4:17:32<7:01:01,  1.60it/s]
 38%|███▊      | 25023/65536 [4:17:32<6:57:24,  1.62it/s]
 38%|███▊      | 25024/65536 [4:17:33<6:57:50,  1.62it/s]
 38%|███▊      | 25025/65536 [4:17:33<6:49:47,  1.65it/s]
 38%|███▊      | 25026/65536 [4:17:34<6:50:58,  1.64it/s]
 38%|███▊      | 25027/65536 [4:17:35<6:56:19,  1.62it/s]
 38%|███▊      | 25028/65536 [4:17:35<6:50:29,  1.64it/s]
 38%|███▊      | 25029/65536 [4:17:36<6:49:25,  1.65it/s]
 38%|███▊      | 25030/65536 [4:17:37<7:07:29,  1.58it/s]
 38%|███▊      | 25031/65536 [4:17:37<6:54:57,  1.63it/s]
 38%|███▊      | 25032/65536 [4:17:38<6:53:06,  1.63it/s]
 38%|███▊      | 25033/65536 [4:17:38<6:52:33,  1.64it/s]
 38%|███▊      | 25034/65536 [4:17:39<7:05:46,  1.59it/s]
 38%|███▊      | 25035/65536 [4:17:40<6:54:34,  1.63it/s]
 38%|███▊      | 25036/65536 [4:17:40<6:54:53,  1.63it/s]
 38%|███▊      | 25037/65536 [4:17:41<6:44:49,  1.67it/s]
 38%|███▊      | 25038/65536 [4:17:41<6:55:08,  1.63it/s]
 38%|███▊      | 25039/65536 [4:17:42<6:49:35,  1.65it/s]
 38%|███▊      | 25040/65536 [4:17:43<6:46:39,  1.66it/s]
                                                         
{'loss': 2.9712, 'learning_rate': 6.578302926411167e-07, 'epoch': 1545.68}
+
 38%|███▊      | 25040/65536 [4:17:43<6:46:39,  1.66it/s]
 38%|███▊      | 25041/65536 [4:17:43<6:51:11,  1.64it/s]
 38%|███▊      | 25042/65536 [4:17:44<6:56:22,  1.62it/s]
 38%|███▊      | 25043/65536 [4:17:44<6:53:35,  1.63it/s]
 38%|███▊      | 25044/65536 [4:17:45<6:53:35,  1.63it/s]
 38%|███▊      | 25045/65536 [4:17:46<6:46:52,  1.66it/s]
 38%|███▊      | 25046/65536 [4:17:46<6:53:55,  1.63it/s]
 38%|███▊      | 25047/65536 [4:17:47<6:45:34,  1.66it/s]
 38%|███▊      | 25048/65536 [4:17:47<6:47:10,  1.66it/s]
 38%|███▊      | 25049/65536 [4:17:48<6:48:44,  1.65it/s]
 38%|███▊      | 25050/65536 [4:17:49<6:45:49,  1.66it/s]
 38%|███▊      | 25051/65536 [4:17:49<6:46:41,  1.66it/s]
 38%|███▊      | 25052/65536 [4:17:50<7:05:50,  1.58it/s]
 38%|███▊      | 25053/65536 [4:17:51<6:55:56,  1.62it/s]
 38%|███▊      | 25054/65536 [4:17:51<6:49:01,  1.65it/s]
 38%|███▊      | 25055/65536 [4:17:52<6:51:37,  1.64it/s]
 38%|███▊      | 25056/65536 [4:17:52<7:11:53,  1.56it/s]
 38%|███▊      | 25057/65536 [4:17:53<7:13:04,  1.56it/s]
 38%|███▊      | 25058/65536 [4:17:54<7:13:57,  1.55it/s]
 38%|███▊      | 25059/65536 [4:17:54<7:06:38,  1.58it/s]
 38%|███▊      | 25060/65536 [4:17:55<7:01:07,  1.60it/s]
                                                         
{'loss': 3.0192, 'learning_rate': 6.575547936818905e-07, 'epoch': 1546.91}
+
 38%|███▊      | 25060/65536 [4:17:55<7:01:07,  1.60it/s]
 38%|███▊      | 25061/65536 [4:17:56<6:53:06,  1.63it/s]
 38%|███▊      | 25062/65536 [4:17:56<7:07:34,  1.58it/s]
 38%|███▊      | 25063/65536 [4:17:57<7:01:18,  1.60it/s]
 38%|███▊      | 25064/65536 [4:17:57<6:58:21,  1.61it/s]
 38%|███▊      | 25065/65536 [4:17:58<6:52:23,  1.64it/s]
 38%|███▊      | 25066/65536 [4:17:59<6:40:57,  1.68it/s]
 38%|███▊      | 25067/65536 [4:17:59<6:41:22,  1.68it/s]
 38%|███▊      | 25068/65536 [4:18:00<6:46:28,  1.66it/s]
 38%|███▊      | 25069/65536 [4:18:00<6:55:11,  1.62it/s]
 38%|███▊      | 25070/65536 [4:18:01<6:53:48,  1.63it/s]
 38%|███▊      | 25071/65536 [4:18:02<7:10:21,  1.57it/s]
 38%|███▊      | 25072/65536 [4:18:02<6:52:42,  1.63it/s]
 38%|███▊      | 25073/65536 [4:18:03<6:45:45,  1.66it/s]
 38%|███▊      | 25074/65536 [4:18:04<6:54:35,  1.63it/s]
 38%|███▊      | 25075/65536 [4:18:04<6:52:50,  1.63it/s]
 38%|███▊      | 25076/65536 [4:18:05<6:45:37,  1.66it/s]
 38%|███▊      | 25077/65536 [4:18:05<6:43:25,  1.67it/s]
 38%|███▊      | 25078/65536 [4:18:06<7:05:49,  1.58it/s]
 38%|███▊      | 25079/65536 [4:18:07<7:03:46,  1.59it/s]
 38%|███▊      | 25080/65536 [4:18:07<6:50:44,  1.64it/s]
                                                         
{'loss': 3.0106, 'learning_rate': 6.572792947226644e-07, 'epoch': 1548.15}
+
 38%|███▊      | 25080/65536 [4:18:07<6:50:44,  1.64it/s]
 38%|███▊      | 25081/65536 [4:18:08<6:48:48,  1.65it/s]
 38%|███▊      | 25082/65536 [4:18:08<6:42:17,  1.68it/s]
 38%|███▊      | 25083/65536 [4:18:09<6:49:37,  1.65it/s]
 38%|███▊      | 25084/65536 [4:18:10<7:05:16,  1.59it/s]
 38%|███▊      | 25085/65536 [4:18:10<6:54:31,  1.63it/s]
 38%|███▊      | 25086/65536 [4:18:11<6:48:36,  1.65it/s]
 38%|███▊      | 25087/65536 [4:18:11<6:43:20,  1.67it/s]
 38%|███▊      | 25088/65536 [4:18:12<6:37:47,  1.69it/s]
 38%|███▊      | 25089/65536 [4:18:13<6:45:12,  1.66it/s]
 38%|███▊      | 25090/65536 [4:18:13<6:42:18,  1.68it/s]
 38%|███▊      | 25091/65536 [4:18:14<6:34:43,  1.71it/s]
 38%|███▊      | 25092/65536 [4:18:14<6:40:52,  1.68it/s]
 38%|███▊      | 25093/65536 [4:18:15<6:54:11,  1.63it/s]
 38%|███▊      | 25094/65536 [4:18:16<7:21:11,  1.53it/s]
 38%|███▊      | 25095/65536 [4:18:16<7:10:27,  1.57it/s]
 38%|███▊      | 25096/65536 [4:18:17<7:09:34,  1.57it/s]
 38%|███▊      | 25097/65536 [4:18:18<7:14:30,  1.55it/s]
 38%|███▊      | 25098/65536 [4:18:18<7:02:24,  1.60it/s]
 38%|███▊      | 25099/65536 [4:18:19<6:51:30,  1.64it/s]
 38%|███▊      | 25100/65536 [4:18:19<6:50:55,  1.64it/s]
                                                         
{'loss': 2.9342, 'learning_rate': 6.570037957634381e-07, 'epoch': 1549.38}
+
 38%|███▊      | 25100/65536 [4:18:19<6:50:55,  1.64it/s]
 38%|███▊      | 25101/65536 [4:18:20<6:41:34,  1.68it/s]
 38%|███▊      | 25102/65536 [4:18:21<6:29:13,  1.73it/s]
 38%|███▊      | 25103/65536 [4:18:21<6:28:30,  1.73it/s]
 38%|███▊      | 25104/65536 [4:18:22<6:33:43,  1.71it/s]
 38%|███▊      | 25105/65536 [4:18:22<6:44:33,  1.67it/s]
 38%|███▊      | 25106/65536 [4:18:23<6:42:33,  1.67it/s]
 38%|███▊      | 25107/65536 [4:18:24<6:47:01,  1.66it/s]
 38%|███▊      | 25108/65536 [4:18:24<6:56:17,  1.62it/s]
 38%|███▊      | 25109/65536 [4:18:25<7:29:54,  1.50it/s]
 38%|███▊      | 25110/65536 [4:18:26<7:32:08,  1.49it/s]
 38%|███▊      | 25111/65536 [4:18:26<7:32:18,  1.49it/s]
 38%|███▊      | 25112/65536 [4:18:27<7:12:47,  1.56it/s]
 38%|███▊      | 25113/65536 [4:18:28<7:06:12,  1.58it/s]
 38%|███▊      | 25114/65536 [4:18:28<6:58:27,  1.61it/s]
 38%|███▊      | 25115/65536 [4:18:29<7:15:02,  1.55it/s]
 38%|███▊      | 25116/65536 [4:18:29<7:05:34,  1.58it/s]
 38%|███▊      | 25117/65536 [4:18:30<6:52:05,  1.63it/s]
 38%|███▊      | 25118/65536 [4:18:31<6:48:38,  1.65it/s]
 38%|███▊      | 25119/65536 [4:18:31<6:44:27,  1.67it/s]
 38%|███▊      | 25120/65536 [4:18:32<6:48:31,  1.65it/s]
                                                         
{'loss': 2.9943, 'learning_rate': 6.56728296804212e-07, 'epoch': 1550.62}
+
 38%|███▊      | 25120/65536 [4:18:32<6:48:31,  1.65it/s]
 38%|███▊      | 25121/65536 [4:18:32<6:47:28,  1.65it/s]
 38%|███▊      | 25122/65536 [4:18:33<6:47:33,  1.65it/s]
 38%|███▊      | 25123/65536 [4:18:34<6:49:47,  1.64it/s]
 38%|███▊      | 25124/65536 [4:18:34<6:48:44,  1.65it/s]
 38%|███▊      | 25125/65536 [4:18:35<6:58:52,  1.61it/s]
 38%|███▊      | 25126/65536 [4:18:36<6:59:02,  1.61it/s]
 38%|███▊      | 25127/65536 [4:18:36<7:09:59,  1.57it/s]
 38%|███▊      | 25128/65536 [4:18:37<7:05:39,  1.58it/s]
 38%|███▊      | 25129/65536 [4:18:38<7:20:34,  1.53it/s]
 38%|███▊      | 25130/65536 [4:18:38<7:04:00,  1.59it/s]
 38%|███▊      | 25131/65536 [4:18:39<6:51:13,  1.64it/s]
 38%|███▊      | 25132/65536 [4:18:39<6:43:50,  1.67it/s]
 38%|███▊      | 25133/65536 [4:18:40<6:40:14,  1.68it/s]
 38%|███▊      | 25134/65536 [4:18:40<6:49:59,  1.64it/s]
 38%|███▊      | 25135/65536 [4:18:41<6:44:15,  1.67it/s]
 38%|███▊      | 25136/65536 [4:18:42<6:45:55,  1.66it/s]
 38%|███▊      | 25137/65536 [4:18:42<6:57:45,  1.61it/s]
 38%|███▊      | 25138/65536 [4:18:43<6:44:01,  1.67it/s]
 38%|███▊      | 25139/65536 [4:18:43<6:47:41,  1.65it/s]
 38%|███▊      | 25140/65536 [4:18:44<6:49:47,  1.64it/s]
                                                         
{'loss': 2.8885, 'learning_rate': 6.564527978449859e-07, 'epoch': 1551.85}
+
 38%|███▊      | 25140/65536 [4:18:44<6:49:47,  1.64it/s]
 38%|███▊      | 25141/65536 [4:18:45<6:45:13,  1.66it/s]
 38%|███▊      | 25142/65536 [4:18:45<6:54:33,  1.62it/s]
 38%|███▊      | 25143/65536 [4:18:46<7:11:26,  1.56it/s]
 38%|███▊      | 25144/65536 [4:18:47<7:01:57,  1.60it/s]
 38%|███▊      | 25145/65536 [4:18:47<6:45:34,  1.66it/s]
 38%|███▊      | 25146/65536 [4:18:48<6:44:35,  1.66it/s]
 38%|███▊      | 25147/65536 [4:18:48<6:48:50,  1.65it/s]
 38%|███▊      | 25148/65536 [4:18:49<6:56:05,  1.62it/s]
 38%|███▊      | 25149/65536 [4:18:50<6:53:53,  1.63it/s]
 38%|███▊      | 25150/65536 [4:18:50<7:02:48,  1.59it/s]
 38%|███▊      | 25151/65536 [4:18:51<6:53:57,  1.63it/s]
 38%|███▊      | 25152/65536 [4:18:52<6:56:16,  1.62it/s]
 38%|███▊      | 25153/65536 [4:18:52<6:51:11,  1.64it/s]
 38%|███▊      | 25154/65536 [4:18:53<6:42:10,  1.67it/s]
 38%|███▊      | 25155/65536 [4:18:53<6:38:13,  1.69it/s]
 38%|███▊      | 25156/65536 [4:18:54<6:46:57,  1.65it/s]
 38%|███▊      | 25157/65536 [4:18:55<6:55:13,  1.62it/s]
 38%|███▊      | 25158/65536 [4:18:55<6:48:02,  1.65it/s]
 38%|███▊      | 25159/65536 [4:18:56<7:04:20,  1.59it/s]
 38%|███▊      | 25160/65536 [4:18:56<6:57:40,  1.61it/s]
                                                         
{'loss': 2.9782, 'learning_rate': 6.561772988857598e-07, 'epoch': 1553.09}
+
 38%|███▊      | 25160/65536 [4:18:56<6:57:40,  1.61it/s]
 38%|███▊      | 25161/65536 [4:18:57<7:05:50,  1.58it/s]
 38%|███▊      | 25162/65536 [4:18:58<7:12:59,  1.55it/s]
 38%|███▊      | 25163/65536 [4:18:58<7:14:33,  1.55it/s]
 38%|███▊      | 25164/65536 [4:18:59<7:05:16,  1.58it/s]
 38%|███▊      | 25165/65536 [4:19:00<6:51:40,  1.63it/s]
 38%|███▊      | 25166/65536 [4:19:00<6:55:15,  1.62it/s]
 38%|███▊      | 25167/65536 [4:19:01<7:00:08,  1.60it/s]
 38%|███▊      | 25168/65536 [4:19:01<6:51:56,  1.63it/s]
 38%|███▊      | 25169/65536 [4:19:02<6:59:17,  1.60it/s]
 38%|███▊      | 25170/65536 [4:19:03<6:58:36,  1.61it/s]
 38%|███▊      | 25171/65536 [4:19:03<6:51:14,  1.64it/s]
 38%|███▊      | 25172/65536 [4:19:04<6:44:26,  1.66it/s]
 38%|███▊      | 25173/65536 [4:19:04<6:42:30,  1.67it/s]
 38%|███▊      | 25174/65536 [4:19:05<6:53:12,  1.63it/s]
 38%|███▊      | 25175/65536 [4:19:06<6:58:23,  1.61it/s]
 38%|███▊      | 25176/65536 [4:19:06<6:50:32,  1.64it/s]
 38%|███▊      | 25177/65536 [4:19:07<6:32:19,  1.71it/s]
 38%|███▊      | 25178/65536 [4:19:07<6:34:48,  1.70it/s]
 38%|███▊      | 25179/65536 [4:19:08<6:57:57,  1.61it/s]
 38%|███▊      | 25180/65536 [4:19:09<7:03:40,  1.59it/s]
                                                         
{'loss': 2.9691, 'learning_rate': 6.559017999265336e-07, 'epoch': 1554.32}
+
 38%|███▊      | 25180/65536 [4:19:09<7:03:40,  1.59it/s]
 38%|███▊      | 25181/65536 [4:19:09<7:00:47,  1.60it/s]
 38%|███▊      | 25182/65536 [4:19:10<7:07:31,  1.57it/s]
 38%|███▊      | 25183/65536 [4:19:11<6:58:01,  1.61it/s]
 38%|███▊      | 25184/65536 [4:19:11<6:58:25,  1.61it/s]
 38%|███▊      | 25185/65536 [4:19:12<6:53:05,  1.63it/s]
 38%|███▊      | 25186/65536 [4:19:12<6:55:17,  1.62it/s]
 38%|███▊      | 25187/65536 [4:19:13<6:50:28,  1.64it/s]
 38%|███▊      | 25188/65536 [4:19:14<6:41:09,  1.68it/s]
 38%|███▊      | 25189/65536 [4:19:14<6:42:54,  1.67it/s]
 38%|███▊      | 25190/65536 [4:19:15<6:44:34,  1.66it/s]
 38%|███▊      | 25191/65536 [4:19:15<6:49:04,  1.64it/s]
 38%|███▊      | 25192/65536 [4:19:16<7:08:24,  1.57it/s]
 38%|███▊      | 25193/65536 [4:19:17<7:12:13,  1.56it/s]
 38%|███▊      | 25194/65536 [4:19:17<6:59:00,  1.60it/s]
 38%|███▊      | 25195/65536 [4:19:18<6:56:24,  1.61it/s]
 38%|███▊      | 25196/65536 [4:19:19<6:53:24,  1.63it/s]
 38%|███▊      | 25197/65536 [4:19:19<6:53:36,  1.63it/s]
 38%|███▊      | 25198/65536 [4:19:20<6:48:37,  1.65it/s]
 38%|███▊      | 25199/65536 [4:19:20<6:53:14,  1.63it/s]
 38%|███▊      | 25200/65536 [4:19:21<6:52:29,  1.63it/s]
                                                         
{'loss': 2.9399, 'learning_rate': 6.556263009673074e-07, 'epoch': 1555.56}
+
 38%|███▊      | 25200/65536 [4:19:21<6:52:29,  1.63it/s]
 38%|███▊      | 25201/65536 [4:19:22<6:38:29,  1.69it/s]
 38%|███▊      | 25202/65536 [4:19:22<6:50:12,  1.64it/s]
 38%|███▊      | 25203/65536 [4:19:23<6:45:08,  1.66it/s]
 38%|███▊      | 25204/65536 [4:19:23<6:44:54,  1.66it/s]
 38%|███▊      | 25205/65536 [4:19:24<6:52:30,  1.63it/s]
 38%|███▊      | 25206/65536 [4:19:25<6:37:30,  1.69it/s]
 38%|███▊      | 25207/65536 [4:19:25<6:39:25,  1.68it/s]
 38%|███▊      | 25208/65536 [4:19:26<7:01:32,  1.59it/s]
 38%|███▊      | 25209/65536 [4:19:27<6:56:00,  1.62it/s]
 38%|███▊      | 25210/65536 [4:19:27<6:48:05,  1.65it/s]
 38%|███▊      | 25211/65536 [4:19:28<6:50:26,  1.64it/s]
 38%|███▊      | 25212/65536 [4:19:28<6:58:53,  1.60it/s]
 38%|███▊      | 25213/65536 [4:19:29<7:02:46,  1.59it/s]
 38%|███▊      | 25214/65536 [4:19:30<7:06:43,  1.57it/s]
 38%|███▊      | 25215/65536 [4:19:30<6:56:51,  1.61it/s]
 38%|███▊      | 25216/65536 [4:19:31<7:00:22,  1.60it/s]
 38%|███▊      | 25217/65536 [4:19:31<6:55:28,  1.62it/s]
 38%|███▊      | 25218/65536 [4:19:32<6:53:16,  1.63it/s]
 38%|███▊      | 25219/65536 [4:19:33<6:54:50,  1.62it/s]
 38%|███▊      | 25220/65536 [4:19:33<6:53:47,  1.62it/s]
                                                         
{'loss': 2.9602, 'learning_rate': 6.553508020080813e-07, 'epoch': 1556.79}
+
 38%|███▊      | 25220/65536 [4:19:33<6:53:47,  1.62it/s]
 38%|███▊      | 25221/65536 [4:19:34<6:54:43,  1.62it/s]
 38%|███▊      | 25222/65536 [4:19:35<6:48:30,  1.64it/s]
 38%|███▊      | 25223/65536 [4:19:35<6:47:32,  1.65it/s]
 38%|███▊      | 25224/65536 [4:19:36<7:03:31,  1.59it/s]
 38%|███▊      | 25225/65536 [4:19:36<6:58:53,  1.60it/s]
 38%|███▊      | 25226/65536 [4:19:37<6:53:46,  1.62it/s]
 38%|███▊      | 25227/65536 [4:19:38<7:05:45,  1.58it/s]
 38%|███▊      | 25228/65536 [4:19:38<6:51:55,  1.63it/s]
 38%|███▊      | 25229/65536 [4:19:39<6:44:37,  1.66it/s]
 38%|███▊      | 25230/65536 [4:19:39<6:43:16,  1.67it/s]
 38%|███▊      | 25231/65536 [4:19:40<6:42:54,  1.67it/s]
 39%|███▊      | 25232/65536 [4:19:41<6:45:07,  1.66it/s]
 39%|███▊      | 25233/65536 [4:19:41<6:50:17,  1.64it/s]
 39%|███▊      | 25234/65536 [4:19:42<6:49:18,  1.64it/s]
 39%|███▊      | 25235/65536 [4:19:42<6:45:22,  1.66it/s]
 39%|███▊      | 25236/65536 [4:19:43<6:50:51,  1.63it/s]
 39%|███▊      | 25237/65536 [4:19:44<6:44:19,  1.66it/s]
 39%|███▊      | 25238/65536 [4:19:44<6:51:53,  1.63it/s]
 39%|███▊      | 25239/65536 [4:19:45<6:52:37,  1.63it/s]
 39%|███▊      | 25240/65536 [4:19:46<7:11:15,  1.56it/s]
                                                         
{'loss': 2.9456, 'learning_rate': 6.550753030488551e-07, 'epoch': 1558.02}
+
 39%|███▊      | 25240/65536 [4:19:46<7:11:15,  1.56it/s]
 39%|███▊      | 25241/65536 [4:19:46<7:06:07,  1.58it/s]
 39%|███▊      | 25242/65536 [4:19:47<6:55:21,  1.62it/s]
 39%|███▊      | 25243/65536 [4:19:47<6:51:32,  1.63it/s]
 39%|███▊      | 25244/65536 [4:19:48<6:50:52,  1.63it/s]
 39%|███▊      | 25245/65536 [4:19:49<6:52:37,  1.63it/s]
 39%|███▊      | 25246/65536 [4:19:49<6:56:47,  1.61it/s]
 39%|███▊      | 25247/65536 [4:19:50<6:45:04,  1.66it/s]
 39%|███▊      | 25248/65536 [4:19:50<6:40:03,  1.68it/s]
 39%|███▊      | 25249/65536 [4:19:51<6:42:37,  1.67it/s]
 39%|███▊      | 25250/65536 [4:19:52<6:42:54,  1.67it/s]
 39%|███▊      | 25251/65536 [4:19:52<6:38:59,  1.68it/s]
 39%|███▊      | 25252/65536 [4:19:53<6:50:51,  1.63it/s]
 39%|███▊      | 25253/65536 [4:19:54<6:58:15,  1.61it/s]
 39%|███▊      | 25254/65536 [4:19:54<6:46:52,  1.65it/s]
 39%|███▊      | 25255/65536 [4:19:55<6:44:45,  1.66it/s]
 39%|███▊      | 25256/65536 [4:19:55<6:59:51,  1.60it/s]
 39%|███▊      | 25257/65536 [4:19:56<6:55:14,  1.62it/s]
 39%|███▊      | 25258/65536 [4:19:57<6:47:47,  1.65it/s]
 39%|███▊      | 25259/65536 [4:19:57<6:59:41,  1.60it/s]
 39%|███▊      | 25260/65536 [4:19:58<7:02:58,  1.59it/s]
                                                         
{'loss': 3.0105, 'learning_rate': 6.54799804089629e-07, 'epoch': 1559.26}
+
 39%|███▊      | 25260/65536 [4:19:58<7:02:58,  1.59it/s]
 39%|███▊      | 25261/65536 [4:19:59<7:04:49,  1.58it/s]
 39%|███▊      | 25262/65536 [4:19:59<7:01:26,  1.59it/s]
 39%|███▊      | 25263/65536 [4:20:00<6:58:56,  1.60it/s]
 39%|███▊      | 25264/65536 [4:20:00<6:50:18,  1.64it/s]
 39%|███▊      | 25265/65536 [4:20:01<6:47:11,  1.65it/s]
 39%|███▊      | 25266/65536 [4:20:02<6:43:08,  1.66it/s]
 39%|██��▊      | 25267/65536 [4:20:02<6:40:11,  1.68it/s]
 39%|███▊      | 25268/65536 [4:20:03<6:42:24,  1.67it/s]
 39%|███▊      | 25269/65536 [4:20:03<6:44:28,  1.66it/s]
 39%|███▊      | 25270/65536 [4:20:04<7:03:19,  1.59it/s]
 39%|███▊      | 25271/65536 [4:20:05<6:53:09,  1.62it/s]
 39%|███▊      | 25272/65536 [4:20:05<6:47:26,  1.65it/s]
 39%|███▊      | 25273/65536 [4:20:06<7:17:28,  1.53it/s]
 39%|███▊      | 25274/65536 [4:20:07<7:12:32,  1.55it/s]
 39%|███▊      | 25275/65536 [4:20:07<6:57:03,  1.61it/s]
 39%|███▊      | 25276/65536 [4:20:08<7:03:04,  1.59it/s]
 39%|███▊      | 25277/65536 [4:20:08<7:02:07,  1.59it/s]
 39%|███▊      | 25278/65536 [4:20:09<7:01:02,  1.59it/s]
 39%|███▊      | 25279/65536 [4:20:10<6:55:53,  1.61it/s]
 39%|███▊      | 25280/65536 [4:20:10<6:55:26,  1.61it/s]
                                                         
{'loss': 2.9561, 'learning_rate': 6.545243051304027e-07, 'epoch': 1560.49}
+
 39%|███▊      | 25280/65536 [4:20:10<6:55:26,  1.61it/s]
 39%|███▊      | 25281/65536 [4:20:11<6:51:45,  1.63it/s]
 39%|███▊      | 25282/65536 [4:20:11<6:53:59,  1.62it/s]
 39%|███▊      | 25283/65536 [4:20:12<6:56:04,  1.61it/s]
 39%|███▊      | 25284/65536 [4:20:13<6:52:45,  1.63it/s]
 39%|███▊      | 25285/65536 [4:20:13<6:42:48,  1.67it/s]
 39%|███▊      | 25286/65536 [4:20:14<6:48:08,  1.64it/s]
 39%|███▊      | 25287/65536 [4:20:14<6:39:13,  1.68it/s]
 39%|███▊      | 25288/65536 [4:20:15<6:28:34,  1.73it/s]
 39%|███▊      | 25289/65536 [4:20:16<7:03:35,  1.58it/s]
 39%|███▊      | 25290/65536 [4:20:16<6:58:27,  1.60it/s]
 39%|███▊      | 25291/65536 [4:20:17<6:53:07,  1.62it/s]
 39%|███▊      | 25292/65536 [4:20:18<6:59:46,  1.60it/s]
 39%|███▊      | 25293/65536 [4:20:18<6:51:01,  1.63it/s]
 39%|███▊      | 25294/65536 [4:20:19<6:49:21,  1.64it/s]
 39%|███▊      | 25295/65536 [4:20:19<6:58:31,  1.60it/s]
 39%|███▊      | 25296/65536 [4:20:20<6:53:34,  1.62it/s]
 39%|███▊      | 25297/65536 [4:20:21<6:41:09,  1.67it/s]
 39%|███▊      | 25298/65536 [4:20:21<6:48:23,  1.64it/s]
 39%|███▊      | 25299/65536 [4:20:22<6:45:31,  1.65it/s]
 39%|███▊      | 25300/65536 [4:20:22<6:47:46,  1.64it/s]
                                                         
{'loss': 2.9641, 'learning_rate': 6.542488061711767e-07, 'epoch': 1561.73}
+
 39%|███▊      | 25300/65536 [4:20:22<6:47:46,  1.64it/s]
 39%|███▊      | 25301/65536 [4:20:23<6:43:40,  1.66it/s]
 39%|███▊      | 25302/65536 [4:20:24<6:38:24,  1.68it/s]
 39%|███▊      | 25303/65536 [4:20:24<6:41:50,  1.67it/s]
 39%|███▊      | 25304/65536 [4:20:25<6:42:06,  1.67it/s]
 39%|███▊      | 25305/65536 [4:20:26<7:05:41,  1.58it/s]
 39%|███▊      | 25306/65536 [4:20:26<6:52:33,  1.63it/s]
 39%|███▊      | 25307/65536 [4:20:27<6:48:15,  1.64it/s]
 39%|███▊      | 25308/65536 [4:20:27<6:42:03,  1.67it/s]
 39%|███▊      | 25309/65536 [4:20:28<6:32:46,  1.71it/s]
 39%|███▊      | 25310/65536 [4:20:28<6:42:01,  1.67it/s]
 39%|███▊      | 25311/65536 [4:20:29<6:38:11,  1.68it/s]
 39%|███▊      | 25312/65536 [4:20:30<6:40:46,  1.67it/s]
 39%|███▊      | 25313/65536 [4:20:30<7:02:34,  1.59it/s]
 39%|███▊      | 25314/65536 [4:20:31<6:53:45,  1.62it/s]
 39%|███▊      | 25315/65536 [4:20:32<6:59:49,  1.60it/s]
 39%|███▊      | 25316/65536 [4:20:32<7:02:00,  1.59it/s]
 39%|███▊      | 25317/65536 [4:20:33<6:47:35,  1.64it/s]
 39%|███▊      | 25318/65536 [4:20:33<6:46:18,  1.65it/s]
 39%|███▊      | 25319/65536 [4:20:34<6:41:45,  1.67it/s]
 39%|███▊      | 25320/65536 [4:20:35<6:43:04,  1.66it/s]
                                                         
{'loss': 3.0078, 'learning_rate': 6.539733072119506e-07, 'epoch': 1562.96}
+
 39%|███▊      | 25320/65536 [4:20:35<6:43:04,  1.66it/s]
 39%|███▊      | 25321/65536 [4:20:35<6:59:04,  1.60it/s]
 39%|███▊      | 25322/65536 [4:20:36<7:04:08,  1.58it/s]
 39%|███▊      | 25323/65536 [4:20:36<6:50:48,  1.63it/s]
 39%|███▊      | 25324/65536 [4:20:37<6:49:11,  1.64it/s]
 39%|███▊      | 25325/65536 [4:20:38<6:50:17,  1.63it/s]
 39%|███▊      | 25326/65536 [4:20:38<6:56:25,  1.61it/s]
 39%|███▊      | 25327/65536 [4:20:39<6:49:00,  1.64it/s]
 39%|███▊      | 25328/65536 [4:20:40<6:54:30,  1.62it/s]
 39%|███▊      | 25329/65536 [4:20:40<6:51:55,  1.63it/s]
 39%|███▊      | 25330/65536 [4:20:41<6:49:52,  1.63it/s]
 39%|███▊      | 25331/65536 [4:20:41<6:41:58,  1.67it/s]
 39%|███▊      | 25332/65536 [4:20:42<6:49:09,  1.64it/s]
 39%|███▊      | 25333/65536 [4:20:43<6:49:32,  1.64it/s]
 39%|███▊      | 25334/65536 [4:20:43<6:48:13,  1.64it/s]
 39%|███▊      | 25335/65536 [4:20:44<6:43:38,  1.66it/s]
 39%|███▊      | 25336/65536 [4:20:44<6:38:47,  1.68it/s]
 39%|███▊      | 25337/65536 [4:20:45<6:50:22,  1.63it/s]
 39%|███▊      | 25338/65536 [4:20:46<6:47:53,  1.64it/s]
 39%|███▊      | 25339/65536 [4:20:46<6:44:20,  1.66it/s]
 39%|███▊      | 25340/65536 [4:20:47<6:55:09,  1.61it/s]
                                                         
{'loss': 2.9818, 'learning_rate': 6.536978082527244e-07, 'epoch': 1564.2}
+
 39%|███▊      | 25340/65536 [4:20:47<6:55:09,  1.61it/s]
 39%|███▊      | 25341/65536 [4:20:48<6:59:52,  1.60it/s]
 39%|███▊      | 25342/65536 [4:20:48<6:50:25,  1.63it/s]
 39%|███▊      | 25343/65536 [4:20:49<6:46:16,  1.65it/s]
 39%|███▊      | 25344/65536 [4:20:49<6:44:34,  1.66it/s]
 39%|███▊      | 25345/65536 [4:20:50<6:39:52,  1.68it/s]
 39%|███▊      | 25346/65536 [4:20:50<6:44:29,  1.66it/s]
 39%|███▊      | 25347/65536 [4:20:51<6:43:07,  1.66it/s]
 39%|███▊      | 25348/65536 [4:20:52<6:35:00,  1.70it/s]
 39%|███▊      | 25349/65536 [4:20:52<6:41:33,  1.67it/s]
 39%|███▊      | 25350/65536 [4:20:53<6:43:13,  1.66it/s]
 39%|███▊      | 25351/65536 [4:20:53<6:45:09,  1.65it/s]
 39%|███▊      | 25352/65536 [4:20:54<6:55:27,  1.61it/s]
 39%|███▊      | 25353/65536 [4:20:55<6:58:29,  1.60it/s]
 39%|███▊      | 25354/65536 [4:20:55<7:13:50,  1.54it/s]
 39%|███▊      | 25355/65536 [4:20:56<7:06:59,  1.57it/s]
 39%|███▊      | 25356/65536 [4:20:57<7:08:28,  1.56it/s]
 39%|███▊      | 25357/65536 [4:20:57<7:00:08,  1.59it/s]
 39%|███▊      | 25358/65536 [4:20:58<7:02:28,  1.59it/s]
 39%|███▊      | 25359/65536 [4:20:59<6:47:22,  1.64it/s]
 39%|███▊      | 25360/65536 [4:20:59<6:48:06,  1.64it/s]
                                                         
{'loss': 3.0106, 'learning_rate': 6.534223092934983e-07, 'epoch': 1565.43}
+
 39%|███▊      | 25360/65536 [4:20:59<6:48:06,  1.64it/s]
 39%|███▊      | 25361/65536 [4:21:00<6:43:16,  1.66it/s]
 39%|███▊      | 25362/65536 [4:21:00<6:42:13,  1.66it/s]
 39%|███▊      | 25363/65536 [4:21:01<6:49:17,  1.64it/s]
 39%|███▊      | 25364/65536 [4:21:02<6:41:25,  1.67it/s]
 39%|███▊      | 25365/65536 [4:21:02<6:39:54,  1.67it/s]
 39%|███▊      | 25366/65536 [4:21:03<7:02:18,  1.59it/s]
 39%|███▊      | 25367/65536 [4:21:03<6:50:09,  1.63it/s]
 39%|███▊      | 25368/65536 [4:21:04<6:57:03,  1.61it/s]
 39%|███▊      | 25369/65536 [4:21:05<6:51:43,  1.63it/s]
 39%|███▊      | 25370/65536 [4:21:05<7:07:12,  1.57it/s]
 39%|███▊      | 25371/65536 [4:21:06<6:59:06,  1.60it/s]
 39%|███▊      | 25372/65536 [4:21:07<7:00:44,  1.59it/s]
 39%|███▊      | 25373/65536 [4:21:07<7:04:22,  1.58it/s]
 39%|███▊      | 25374/65536 [4:21:08<7:04:47,  1.58it/s]
 39%|███▊      | 25375/65536 [4:21:09<7:22:00,  1.51it/s]
 39%|███▊      | 25376/65536 [4:21:09<7:17:20,  1.53it/s]
 39%|███▊      | 25377/65536 [4:21:10<7:02:33,  1.58it/s]
 39%|███▊      | 25378/65536 [4:21:10<6:48:23,  1.64it/s]
 39%|███▊      | 25379/65536 [4:21:11<6:54:27,  1.61it/s]
 39%|███▊      | 25380/65536 [4:21:12<6:49:29,  1.63it/s]
                                                         
{'loss': 2.9646, 'learning_rate': 6.53146810334272e-07, 'epoch': 1566.67}
+
 39%|███▊      | 25380/65536 [4:21:12<6:49:29,  1.63it/s]
 39%|███▊      | 25381/65536 [4:21:12<6:43:29,  1.66it/s]
 39%|███▊      | 25382/65536 [4:21:13<6:43:44,  1.66it/s]
 39%|███▊      | 25383/65536 [4:21:13<7:02:06,  1.59it/s]
 39%|███▊      | 25384/65536 [4:21:14<6:51:55,  1.62it/s]
 39%|███▊      | 25385/65536 [4:21:15<6:51:45,  1.63it/s]
 39%|███▊      | 25386/65536 [4:21:15<7:03:09,  1.58it/s]
 39%|███▊      | 25387/65536 [4:21:16<7:00:28,  1.59it/s]
 39%|███▊      | 25388/65536 [4:21:17<6:50:38,  1.63it/s]
 39%|███▊      | 25389/65536 [4:21:17<6:53:23,  1.62it/s]
 39%|███▊      | 25390/65536 [4:21:18<7:01:40,  1.59it/s]
 39%|███▊      | 25391/65536 [4:21:18<6:56:34,  1.61it/s]
 39%|███▊      | 25392/65536 [4:21:19<7:01:52,  1.59it/s]
 39%|███▊      | 25393/65536 [4:21:20<7:13:36,  1.54it/s]
 39%|███▊      | 25394/65536 [4:21:20<7:04:09,  1.58it/s]
 39%|███▊      | 25395/65536 [4:21:21<6:53:48,  1.62it/s]
 39%|███▉      | 25396/65536 [4:21:22<6:50:53,  1.63it/s]
 39%|███▉      | 25397/65536 [4:21:22<6:45:11,  1.65it/s]
 39%|███▉      | 25398/65536 [4:21:23<6:42:21,  1.66it/s]
 39%|███▉      | 25399/65536 [4:21:23<6:42:33,  1.66it/s]
 39%|███▉      | 25400/65536 [4:21:24<6:46:06,  1.65it/s]
                                                         
{'loss': 2.9095, 'learning_rate': 6.528713113750459e-07, 'epoch': 1567.9}
+
 39%|███▉      | 25400/65536 [4:21:24<6:46:06,  1.65it/s]
 39%|███▉      | 25401/65536 [4:21:25<6:57:15,  1.60it/s]
 39%|███▉      | 25402/65536 [4:21:25<7:09:06,  1.56it/s]
 39%|███▉      | 25403/65536 [4:21:26<6:57:45,  1.60it/s]
 39%|███▉      | 25404/65536 [4:21:27<6:57:25,  1.60it/s]
 39%|███▉      | 25405/65536 [4:21:27<6:53:49,  1.62it/s]
 39%|███▉      | 25406/65536 [4:21:28<6:51:45,  1.62it/s]
 39%|███▉      | 25407/65536 [4:21:28<6:50:59,  1.63it/s]
 39%|███▉      | 25408/65536 [4:21:29<7:00:55,  1.59it/s]
 39%|███▉      | 25409/65536 [4:21:30<6:46:28,  1.65it/s]
 39%|███▉      | 25410/65536 [4:21:30<6:51:02,  1.63it/s]
 39%|███▉      | 25411/65536 [4:21:31<6:58:55,  1.60it/s]
 39%|███▉      | 25412/65536 [4:21:31<6:59:08,  1.60it/s]
 39%|███▉      | 25413/65536 [4:21:32<7:02:51,  1.58it/s]
 39%|███▉      | 25414/65536 [4:21:33<7:03:28,  1.58it/s]
 39%|███▉      | 25415/65536 [4:21:33<6:52:52,  1.62it/s]
 39%|███▉      | 25416/65536 [4:21:34<6:46:17,  1.65it/s]
 39%|███▉      | 25417/65536 [4:21:35<6:48:52,  1.64it/s]
 39%|███▉      | 25418/65536 [4:21:35<7:09:23,  1.56it/s]
 39%|███▉      | 25419/65536 [4:21:36<6:59:24,  1.59it/s]
 39%|███▉      | 25420/65536 [4:21:36<7:02:46,  1.58it/s]
                                                         
{'loss': 2.9554, 'learning_rate': 6.525958124158198e-07, 'epoch': 1569.14}
+
 39%|███▉      | 25420/65536 [4:21:36<7:02:46,  1.58it/s]
 39%|███▉      | 25421/65536 [4:21:37<6:56:13,  1.61it/s]
 39%|███▉      | 25422/65536 [4:21:38<6:53:28,  1.62it/s]
 39%|███▉      | 25423/65536 [4:21:38<6:59:01,  1.60it/s]
 39%|███▉      | 25424/65536 [4:21:39<6:49:39,  1.63it/s]
 39%|███▉      | 25425/65536 [4:21:40<6:50:46,  1.63it/s]
 39%|███▉      | 25426/65536 [4:21:40<6:51:51,  1.62it/s]
 39%|███▉      | 25427/65536 [4:21:41<6:55:31,  1.61it/s]
 39%|███▉      | 25428/65536 [4:21:41<6:56:54,  1.60it/s]
 39%|███▉      | 25429/65536 [4:21:42<6:52:16,  1.62it/s]
 39%|███▉      | 25430/65536 [4:21:43<6:45:34,  1.65it/s]
 39%|███▉      | 25431/65536 [4:21:43<6:48:25,  1.64it/s]
 39%|███▉      | 25432/65536 [4:21:44<6:57:10,  1.60it/s]
 39%|███▉      | 25433/65536 [4:21:44<6:52:45,  1.62it/s]
 39%|███▉      | 25434/65536 [4:21:45<6:46:12,  1.65it/s]
 39%|███▉      | 25435/65536 [4:21:46<7:05:55,  1.57it/s]
 39%|███▉      | 25436/65536 [4:21:46<6:56:52,  1.60it/s]
 39%|███▉      | 25437/65536 [4:21:47<6:54:31,  1.61it/s]
 39%|███▉      | 25438/65536 [4:21:48<6:53:22,  1.62it/s]
 39%|███▉      | 25439/65536 [4:21:48<6:57:00,  1.60it/s]
 39%|███▉      | 25440/65536 [4:21:49<7:05:10,  1.57it/s]
                                                         
{'loss': 2.8864, 'learning_rate': 6.523203134565935e-07, 'epoch': 1570.37}
+
 39%|███▉      | 25440/65536 [4:21:49<7:05:10,  1.57it/s]
 39%|███▉      | 25441/65536 [4:21:50<6:57:52,  1.60it/s]
 39%|███▉      | 25442/65536 [4:21:50<6:53:12,  1.62it/s]
 39%|███▉      | 25443/65536 [4:21:51<6:47:56,  1.64it/s]
 39%|███▉      | 25444/65536 [4:21:51<6:48:22,  1.64it/s]
 39%|███▉      | 25445/65536 [4:21:52<6:45:23,  1.65it/s]
 39%|███▉      | 25446/65536 [4:21:52<6:42:39,  1.66it/s]
 39%|███▉      | 25447/65536 [4:21:53<6:51:29,  1.62it/s]
 39%|███▉      | 25448/65536 [4:21:54<6:43:59,  1.65it/s]
 39%|███▉      | 25449/65536 [4:21:54<6:41:43,  1.66it/s]
 39%|███▉      | 25450/65536 [4:21:55<6:42:02,  1.66it/s]
 39%|███▉      | 25451/65536 [4:21:56<6:59:55,  1.59it/s]
 39%|███▉      | 25452/65536 [4:21:56<6:47:28,  1.64it/s]
 39%|███▉      | 25453/65536 [4:21:57<6:56:34,  1.60it/s]
 39%|███▉      | 25454/65536 [4:21:57<6:46:26,  1.64it/s]
 39%|███▉      | 25455/65536 [4:21:58<6:54:37,  1.61it/s]
 39%|███▉      | 25456/65536 [4:21:59<6:48:14,  1.64it/s]
 39%|███▉      | 25457/65536 [4:21:59<6:41:20,  1.66it/s]
 39%|███▉      | 25458/65536 [4:22:00<6:35:01,  1.69it/s]
 39%|███▉      | 25459/65536 [4:22:00<6:32:26,  1.70it/s]
 39%|███▉      | 25460/65536 [4:22:01<6:44:10,  1.65it/s]
                                                         
{'loss': 2.9818, 'learning_rate': 6.520448144973674e-07, 'epoch': 1571.6}
+
 39%|███▉      | 25460/65536 [4:22:01<6:44:10,  1.65it/s]
 39%|███▉      | 25461/65536 [4:22:02<6:42:22,  1.66it/s]
 39%|███▉      | 25462/65536 [4:22:02<6:39:00,  1.67it/s]
 39%|███▉      | 25463/65536 [4:22:03<6:49:42,  1.63it/s]
 39%|███▉      | 25464/65536 [4:22:04<7:07:27,  1.56it/s]
 39%|███▉      | 25465/65536 [4:22:04<7:03:54,  1.58it/s]
 39%|███▉      | 25466/65536 [4:22:05<6:57:50,  1.60it/s]
 39%|███▉      | 25467/65536 [4:22:05<7:02:17,  1.58it/s]
 39%|███▉      | 25468/65536 [4:22:06<7:01:06,  1.59it/s]
 39%|███▉      | 25469/65536 [4:22:07<6:53:36,  1.61it/s]
 39%|███▉      | 25470/65536 [4:22:07<6:41:25,  1.66it/s]
 39%|███▉      | 25471/65536 [4:22:08<6:45:14,  1.65it/s]
 39%|███▉      | 25472/65536 [4:22:08<6:53:38,  1.61it/s]
 39%|███▉      | 25473/65536 [4:22:09<6:46:54,  1.64it/s]
 39%|███▉      | 25474/65536 [4:22:10<6:40:40,  1.67it/s]
 39%|███▉      | 25475/65536 [4:22:10<6:54:49,  1.61it/s]
 39%|███▉      | 25476/65536 [4:22:11<6:48:48,  1.63it/s]
 39%|███▉      | 25477/65536 [4:22:12<6:52:14,  1.62it/s]
 39%|███▉      | 25478/65536 [4:22:12<6:59:34,  1.59it/s]
 39%|███▉      | 25479/65536 [4:22:13<7:03:02,  1.58it/s]
 39%|███▉      | 25480/65536 [4:22:13<7:03:30,  1.58it/s]
                                                         
{'loss': 2.9394, 'learning_rate': 6.517693155381412e-07, 'epoch': 1572.84}
+
 39%|███▉      | 25480/65536 [4:22:13<7:03:30,  1.58it/s]
 39%|███▉      | 25481/65536 [4:22:14<6:50:05,  1.63it/s]
 39%|███▉      | 25482/65536 [4:22:15<6:46:21,  1.64it/s]
 39%|███▉      | 25483/65536 [4:22:15<6:54:06,  1.61it/s]
 39%|███▉      | 25484/65536 [4:22:16<6:51:39,  1.62it/s]
 39%|███▉      | 25485/65536 [4:22:16<6:40:58,  1.66it/s]
 39%|███▉      | 25486/65536 [4:22:17<6:49:51,  1.63it/s]
 39%|███▉      | 25487/65536 [4:22:18<6:52:48,  1.62it/s]
 39%|███▉      | 25488/65536 [4:22:18<7:06:08,  1.57it/s]
 39%|███▉      | 25489/65536 [4:22:19<6:59:35,  1.59it/s]
 39%|███▉      | 25490/65536 [4:22:20<6:55:46,  1.61it/s]
 39%|███▉      | 25491/65536 [4:22:20<6:43:27,  1.65it/s]
 39%|███▉      | 25492/65536 [4:22:21<6:40:49,  1.67it/s]
 39%|███▉      | 25493/65536 [4:22:21<6:40:07,  1.67it/s]
 39%|███▉      | 25494/65536 [4:22:22<6:42:53,  1.66it/s]
 39%|███▉      | 25495/65536 [4:22:23<6:40:04,  1.67it/s]
 39%|███▉      | 25496/65536 [4:22:23<6:42:47,  1.66it/s]
 39%|███▉      | 25497/65536 [4:22:24<6:51:52,  1.62it/s]
 39%|███▉      | 25498/65536 [4:22:24<6:43:53,  1.65it/s]
 39%|███▉      | 25499/65536 [4:22:25<6:56:36,  1.60it/s]
 39%|███▉      | 25500/65536 [4:22:26<6:46:50,  1.64it/s]
                                                         
{'loss': 2.9811, 'learning_rate': 6.514938165789151e-07, 'epoch': 1574.07}
+
 39%|███▉      | 25500/65536 [4:22:26<6:46:50,  1.64it/s]
 39%|███▉      | 25501/65536 [4:22:26<6:43:53,  1.65it/s]
 39%|███▉      | 25502/65536 [4:22:27<6:45:55,  1.64it/s]
 39%|███▉      | 25503/65536 [4:22:27<6:44:57,  1.65it/s]
 39%|███▉      | 25504/65536 [4:22:28<6:46:14,  1.64it/s]
 39%|███▉      | 25505/65536 [4:22:29<6:43:02,  1.66it/s]
 39%|███▉      | 25506/65536 [4:22:29<6:52:47,  1.62it/s]
 39%|███▉      | 25507/65536 [4:22:30<6:52:30,  1.62it/s]
 39%|███▉      | 25508/65536 [4:22:31<6:46:28,  1.64it/s]
 39%|███▉      | 25509/65536 [4:22:31<6:38:50,  1.67it/s]
 39%|███▉      | 25510/65536 [4:22:32<6:36:01,  1.68it/s]
 39%|███▉      | 25511/65536 [4:22:32<6:47:51,  1.64it/s]
 39%|███▉      | 25512/65536 [4:22:33<6:54:58,  1.61it/s]
 39%|███▉      | 25513/65536 [4:22:34<6:57:01,  1.60it/s]
 39%|███▉      | 25514/65536 [4:22:34<6:58:35,  1.59it/s]
 39%|███▉      | 25515/65536 [4:22:35<6:59:27,  1.59it/s]
 39%|███▉      | 25516/65536 [4:22:36<7:21:44,  1.51it/s]
 39%|███▉      | 25517/65536 [4:22:36<7:11:50,  1.54it/s]
 39%|███▉      | 25518/65536 [4:22:37<7:00:23,  1.59it/s]
 39%|███▉      | 25519/65536 [4:22:37<6:52:24,  1.62it/s]
 39%|███▉      | 25520/65536 [4:22:38<6:54:22,  1.61it/s]
                                                         
{'loss': 2.963, 'learning_rate': 6.51218317619689e-07, 'epoch': 1575.31}
+
 39%|███▉      | 25520/65536 [4:22:38<6:54:22,  1.61it/s]
 39%|███▉      | 25521/65536 [4:22:39<7:08:07,  1.56it/s]
 39%|███▉      | 25522/65536 [4:22:39<7:00:10,  1.59it/s]
 39%|███▉      | 25523/65536 [4:22:40<7:00:58,  1.58it/s]
 39%|███▉      | 25524/65536 [4:22:41<6:47:50,  1.64it/s]
 39%|███▉      | 25525/65536 [4:22:41<6:49:51,  1.63it/s]
 39%|███▉      | 25526/65536 [4:22:42<6:52:02,  1.62it/s]
 39%|███▉      | 25527/65536 [4:22:42<6:42:45,  1.66it/s]
 39%|███▉      | 25528/65536 [4:22:43<6:39:56,  1.67it/s]
 39%|███▉      | 25529/65536 [4:22:44<6:36:42,  1.68it/s]
 39%|███▉      | 25530/65536 [4:22:44<6:36:54,  1.68it/s]
 39%|███▉      | 25531/65536 [4:22:45<6:52:03,  1.62it/s]
 39%|███▉      | 25532/65536 [4:22:46<7:08:57,  1.55it/s]
 39%|███▉      | 25533/65536 [4:22:46<7:06:56,  1.56it/s]
 39%|███▉      | 25534/65536 [4:22:47<6:59:37,  1.59it/s]
 39%|███▉      | 25535/65536 [4:22:47<6:51:37,  1.62it/s]
 39%|███▉      | 25536/65536 [4:22:48<6:48:01,  1.63it/s]
 39%|███▉      | 25537/65536 [4:22:49<6:46:30,  1.64it/s]
 39%|███▉      | 25538/65536 [4:22:49<6:45:29,  1.64it/s]
 39%|███▉      | 25539/65536 [4:22:50<6:37:01,  1.68it/s]
 39%|███▉      | 25540/65536 [4:22:50<6:46:56,  1.64it/s]
                                                         
{'loss': 2.9593, 'learning_rate': 6.509428186604628e-07, 'epoch': 1576.54}
+
 39%|███▉      | 25540/65536 [4:22:50<6:46:56,  1.64it/s]
 39%|███▉      | 25541/65536 [4:22:51<6:47:23,  1.64it/s]
 39%|███▉      | 25542/65536 [4:22:52<6:58:51,  1.59it/s]
 39%|███▉      | 25543/65536 [4:22:52<6:51:03,  1.62it/s]
 39%|███▉      | 25544/65536 [4:22:53<7:00:46,  1.58it/s]
 39%|███▉      | 25545/65536 [4:22:53<6:44:40,  1.65it/s]
 39%|███▉      | 25546/65536 [4:22:54<6:42:46,  1.65it/s]
 39%|███▉      | 25547/65536 [4:22:55<6:38:00,  1.67it/s]
 39%|███▉      | 25548/65536 [4:22:55<6:48:21,  1.63it/s]
 39%|███▉      | 25549/65536 [4:22:56<6:46:33,  1.64it/s]
 39%|███▉      | 25550/65536 [4:22:56<6:43:50,  1.65it/s]
 39%|███▉      | 25551/65536 [4:22:57<6:35:17,  1.69it/s]
 39%|███▉      | 25552/65536 [4:22:58<6:31:29,  1.70it/s]
 39%|███▉      | 25553/65536 [4:22:58<6:34:14,  1.69it/s]
 39%|███▉      | 25554/65536 [4:22:59<6:28:30,  1.72it/s]
 39%|███▉      | 25555/65536 [4:22:59<6:33:37,  1.69it/s]
 39%|███▉      | 25556/65536 [4:23:00<6:41:28,  1.66it/s]
 39%|███▉      | 25557/65536 [4:23:01<6:51:54,  1.62it/s]
 39%|███▉      | 25558/65536 [4:23:01<6:52:20,  1.62it/s]
 39%|███▉      | 25559/65536 [4:23:02<6:50:58,  1.62it/s]
 39%|███▉      | 25560/65536 [4:23:03<6:57:55,  1.59it/s]
                                                         
{'loss': 2.9009, 'learning_rate': 6.506673197012367e-07, 'epoch': 1577.78}
+
 39%|███▉      | 25560/65536 [4:23:03<6:57:55,  1.59it/s]
 39%|███▉      | 25561/65536 [4:23:03<6:51:14,  1.62it/s]
 39%|███▉      | 25562/65536 [4:23:04<7:05:03,  1.57it/s]
 39%|███▉      | 25563/65536 [4:23:04<6:52:05,  1.62it/s]
 39%|███▉      | 25564/65536 [4:23:05<7:01:20,  1.58it/s]
 39%|███▉      | 25565/65536 [4:23:06<7:08:17,  1.56it/s]
 39%|███▉      | 25566/65536 [4:23:06<7:04:49,  1.57it/s]
 39%|███▉      | 25567/65536 [4:23:07<6:51:34,  1.62it/s]
 39%|███▉      | 25568/65536 [4:23:07<6:41:29,  1.66it/s]
 39%|███▉      | 25569/65536 [4:23:08<6:46:32,  1.64it/s]
 39%|███▉      | 25570/65536 [4:23:09<6:52:59,  1.61it/s]
 39%|███▉      | 25571/65536 [4:23:09<6:58:19,  1.59it/s]
 39%|███▉      | 25572/65536 [4:23:10<6:52:39,  1.61it/s]
 39%|███▉      | 25573/65536 [4:23:11<6:42:29,  1.65it/s]
 39%|███▉      | 25574/65536 [4:23:11<6:42:52,  1.65it/s]
 39%|███▉      | 25575/65536 [4:23:12<6:37:36,  1.68it/s]
 39%|███▉      | 25576/65536 [4:23:12<6:40:53,  1.66it/s]
 39%|███▉      | 25577/65536 [4:23:13<6:43:15,  1.65it/s]
 39%|███▉      | 25578/65536 [4:23:14<6:49:40,  1.63it/s]
 39%|███▉      | 25579/65536 [4:23:14<6:44:29,  1.65it/s]
 39%|███▉      | 25580/65536 [4:23:15<6:52:03,  1.62it/s]
                                                         
{'loss': 2.9137, 'learning_rate': 6.503918207420105e-07, 'epoch': 1579.01}
+
 39%|███▉      | 25580/65536 [4:23:15<6:52:03,  1.62it/s]
 39%|███▉      | 25581/65536 [4:23:15<6:47:41,  1.63it/s]
 39%|███▉      | 25582/65536 [4:23:16<6:43:45,  1.65it/s]
 39%|███▉      | 25583/65536 [4:23:17<6:42:52,  1.65it/s]
 39%|███▉      | 25584/65536 [4:23:17<6:42:45,  1.65it/s]
 39%|███▉      | 25585/65536 [4:23:18<6:49:27,  1.63it/s]
 39%|███▉      | 25586/65536 [4:23:19<6:51:15,  1.62it/s]
 39%|███▉      | 25587/65536 [4:23:19<6:52:56,  1.61it/s]
 39%|███▉      | 25588/65536 [4:23:20<6:40:35,  1.66it/s]
 39%|███▉      | 25589/65536 [4:23:20<6:43:58,  1.65it/s]
 39%|███▉      | 25590/65536 [4:23:21<6:45:30,  1.64it/s]
 39%|███▉      | 25591/65536 [4:23:22<6:48:05,  1.63it/s]
 39%|███▉      | 25592/65536 [4:23:22<6:49:52,  1.62it/s]
 39%|███▉      | 25593/65536 [4:23:23<6:58:36,  1.59it/s]
 39%|███▉      | 25594/65536 [4:23:23<6:54:14,  1.61it/s]
 39%|███▉      | 25595/65536 [4:23:24<6:49:56,  1.62it/s]
 39%|███▉      | 25596/65536 [4:23:25<6:49:36,  1.63it/s]
 39%|███▉      | 25597/65536 [4:23:25<7:05:29,  1.56it/s]
 39%|███▉      | 25598/65536 [4:23:26<7:01:18,  1.58it/s]
 39%|███▉      | 25599/65536 [4:23:27<6:59:56,  1.59it/s]
 39%|███▉      | 25600/65536 [4:23:27<6:53:34,  1.61it/s]
                                                         
{'loss': 2.8327, 'learning_rate': 6.501163217827844e-07, 'epoch': 1580.25}
+
 39%|███▉      | 25600/65536 [4:23:27<6:53:34,  1.61it/s]
 39%|███▉      | 25601/65536 [4:23:28<6:49:11,  1.63it/s]
 39%|███▉      | 25602/65536 [4:23:28<6:47:00,  1.64it/s]
 39%|███▉      | 25603/65536 [4:23:29<6:47:22,  1.63it/s]
 39%|███▉      | 25604/65536 [4:23:30<6:52:45,  1.61it/s]
 39%|███▉      | 25605/65536 [4:23:30<6:57:19,  1.59it/s]
 39%|███▉      | 25606/65536 [4:23:31<6:56:25,  1.60it/s]
 39%|███▉      | 25607/65536 [4:23:32<6:53:50,  1.61it/s]
 39%|███▉      | 25608/65536 [4:23:32<6:52:37,  1.61it/s]
 39%|███▉      | 25609/65536 [4:23:33<6:45:38,  1.64it/s]
 39%|███▉      | 25610/65536 [4:23:33<6:45:17,  1.64it/s]
 39%|███▉      | 25611/65536 [4:23:34<6:36:05,  1.68it/s]
 39%|███▉      | 25612/65536 [4:23:35<6:36:28,  1.68it/s]
 39%|███▉      | 25613/65536 [4:23:35<6:47:03,  1.63it/s]
 39%|███▉      | 25614/65536 [4:23:36<6:42:32,  1.65it/s]
 39%|███▉      | 25615/65536 [4:23:36<6:54:08,  1.61it/s]
 39%|███▉      | 25616/65536 [4:23:37<6:56:09,  1.60it/s]
 39%|███▉      | 25617/65536 [4:23:38<6:53:02,  1.61it/s]
 39%|███▉      | 25618/65536 [4:23:38<6:42:58,  1.65it/s]
 39%|███▉      | 25619/65536 [4:23:39<6:55:10,  1.60it/s]
 39%|███▉      | 25620/65536 [4:23:39<6:46:29,  1.64it/s]
                                                         
{'loss': 2.9212, 'learning_rate': 6.498408228235582e-07, 'epoch': 1581.48}
+
 39%|███▉      | 25620/65536 [4:23:39<6:46:29,  1.64it/s]
 39%|███▉      | 25621/65536 [4:23:40<6:54:09,  1.61it/s]
 39%|███▉      | 25622/65536 [4:23:41<6:58:40,  1.59it/s]
 39%|███▉      | 25623/65536 [4:23:41<6:59:51,  1.58it/s]
 39%|███▉      | 25624/65536 [4:23:42<6:56:23,  1.60it/s]
 39%|███▉      | 25625/65536 [4:23:43<6:46:27,  1.64it/s]
 39%|███▉      | 25626/65536 [4:23:43<6:40:14,  1.66it/s]
 39%|███▉      | 25627/65536 [4:23:44<6:39:59,  1.66it/s]
 39%|███▉      | 25628/65536 [4:23:44<6:37:13,  1.67it/s]
 39%|███▉      | 25629/65536 [4:23:45<6:58:10,  1.59it/s]
 39%|███▉      | 25630/65536 [4:23:46<6:59:40,  1.58it/s]
 39%|███▉      | 25631/65536 [4:23:46<6:55:58,  1.60it/s]
 39%|███▉      | 25632/65536 [4:23:47<6:51:14,  1.62it/s]
 39%|███▉      | 25633/65536 [4:23:47<6:44:01,  1.65it/s]
 39%|███▉      | 25634/65536 [4:23:48<6:40:40,  1.66it/s]
 39%|███▉      | 25635/65536 [4:23:49<6:55:32,  1.60it/s]
 39%|███▉      | 25636/65536 [4:23:49<6:59:50,  1.58it/s]
 39%|███▉      | 25637/65536 [4:23:50<6:44:33,  1.64it/s]
 39%|███▉      | 25638/65536 [4:23:51<6:47:35,  1.63it/s]
 39%|███▉      | 25639/65536 [4:23:51<6:48:28,  1.63it/s]
 39%|███▉      | 25640/65536 [4:23:52<6:46:20,  1.64it/s]
                                                         
{'loss': 2.8549, 'learning_rate': 6.495653238643321e-07, 'epoch': 1582.72}
+
 39%|███▉      | 25640/65536 [4:23:52<6:46:20,  1.64it/s]
 39%|███▉      | 25641/65536 [4:23:52<6:40:45,  1.66it/s]
 39%|███▉      | 25642/65536 [4:23:53<6:38:57,  1.67it/s]
 39%|███▉      | 25643/65536 [4:23:54<6:46:19,  1.64it/s]
 39%|███▉      | 25644/65536 [4:23:54<6:42:20,  1.65it/s]
 39%|███▉      | 25645/65536 [4:23:55<6:46:56,  1.63it/s]
 39%|███▉      | 25646/65536 [4:23:55<6:47:40,  1.63it/s]
 39%|███▉      | 25647/65536 [4:23:56<6:59:14,  1.59it/s]
 39%|███▉      | 25648/65536 [4:23:57<6:45:00,  1.64it/s]
 39%|███▉      | 25649/65536 [4:23:57<6:44:00,  1.65it/s]
 39%|███▉      | 25650/65536 [4:23:58<6:41:24,  1.66it/s]
 39%|███▉      | 25651/65536 [4:23:58<6:34:34,  1.68it/s]
 39%|███▉      | 25652/65536 [4:23:59<6:46:08,  1.64it/s]
 39%|███▉      | 25653/65536 [4:24:00<6:52:10,  1.61it/s]
 39%|███▉      | 25654/65536 [4:24:00<6:51:30,  1.62it/s]
 39%|███▉      | 25655/65536 [4:24:01<6:46:53,  1.63it/s]
 39%|███▉      | 25656/65536 [4:24:02<6:37:51,  1.67it/s]
 39%|███▉      | 25657/65536 [4:24:02<6:33:59,  1.69it/s]
 39%|███▉      | 25658/65536 [4:24:03<6:41:54,  1.65it/s]
 39%|███▉      | 25659/65536 [4:24:03<6:43:20,  1.65it/s]
 39%|███▉      | 25660/65536 [4:24:04<6:42:15,  1.65it/s]
                                                         
{'loss': 2.889, 'learning_rate': 6.492898249051059e-07, 'epoch': 1583.95}
+
 39%|███▉      | 25660/65536 [4:24:04<6:42:15,  1.65it/s]
 39%|███▉      | 25661/65536 [4:24:05<6:58:42,  1.59it/s]
 39%|███▉      | 25662/65536 [4:24:05<6:51:58,  1.61it/s]
 39%|███▉      | 25663/65536 [4:24:06<7:09:33,  1.55it/s]
 39%|███▉      | 25664/65536 [4:24:07<6:57:46,  1.59it/s]
 39%|███▉      | 25665/65536 [4:24:07<6:54:25,  1.60it/s]
 39%|███▉      | 25666/65536 [4:24:08<6:45:27,  1.64it/s]
 39%|███▉      | 25667/65536 [4:24:08<6:48:55,  1.62it/s]
 39%|███▉      | 25668/65536 [4:24:09<6:47:29,  1.63it/s]
 39%|███▉      | 25669/65536 [4:24:10<6:54:11,  1.60it/s]
 39%|███▉      | 25670/65536 [4:24:10<6:54:08,  1.60it/s]
 39%|███▉      | 25671/65536 [4:24:11<6:45:12,  1.64it/s]
 39%|███▉      | 25672/65536 [4:24:11<6:38:50,  1.67it/s]
 39%|███▉      | 25673/65536 [4:24:12<6:42:15,  1.65it/s]
 39%|███▉      | 25674/65536 [4:24:13<6:40:48,  1.66it/s]
 39%|███▉      | 25675/65536 [4:24:13<6:41:33,  1.65it/s]
 39%|███▉      | 25676/65536 [4:24:14<6:43:58,  1.64it/s]
 39%|███▉      | 25677/65536 [4:24:14<6:38:19,  1.67it/s]
 39%|███▉      | 25678/65536 [4:24:15<6:49:29,  1.62it/s]
 39%|███▉      | 25679/65536 [4:24:16<6:48:29,  1.63it/s]
 39%|███▉      | 25680/65536 [4:24:16<6:59:12,  1.58it/s]
                                                         
{'loss': 2.8057, 'learning_rate': 6.490143259458797e-07, 'epoch': 1585.19}
+
 39%|███▉      | 25680/65536 [4:24:16<6:59:12,  1.58it/s]
 39%|███▉      | 25681/65536 [4:24:17<6:45:59,  1.64it/s]
 39%|███▉      | 25682/65536 [4:24:18<6:44:00,  1.64it/s]
 39%|███▉      | 25683/65536 [4:24:18<6:50:49,  1.62it/s]
 39%|███▉      | 25684/65536 [4:24:19<6:38:07,  1.67it/s]
 39%|███▉      | 25685/65536 [4:24:19<6:48:55,  1.62it/s]
 39%|███▉      | 25686/65536 [4:24:20<6:52:41,  1.61it/s]
 39%|███▉      | 25687/65536 [4:24:21<6:54:39,  1.60it/s]
 39%|███▉      | 25688/65536 [4:24:21<6:53:19,  1.61it/s]
 39%|███▉      | 25689/65536 [4:24:22<7:04:36,  1.56it/s]
 39%|███▉      | 25690/65536 [4:24:23<6:53:48,  1.60it/s]
 39%|███▉      | 25691/65536 [4:24:23<6:45:49,  1.64it/s]
 39%|███▉      | 25692/65536 [4:24:24<6:43:54,  1.64it/s]
 39%|███▉      | 25693/65536 [4:24:24<6:38:39,  1.67it/s]
 39%|███▉      | 25694/65536 [4:24:25<7:04:50,  1.56it/s]
 39%|███▉      | 25695/65536 [4:24:26<6:54:37,  1.60it/s]
 39%|███▉      | 25696/65536 [4:24:26<7:02:02,  1.57it/s]
 39%|███▉      | 25697/65536 [4:24:27<6:51:00,  1.62it/s]
 39%|███▉      | 25698/65536 [4:24:27<6:41:32,  1.65it/s]
 39%|███▉      | 25699/65536 [4:24:28<7:00:59,  1.58it/s]
 39%|███▉      | 25700/65536 [4:24:29<7:01:35,  1.57it/s]
                                                         
{'loss': 2.7912, 'learning_rate': 6.487388269866536e-07, 'epoch': 1586.42}
+
 39%|███▉      | 25700/65536 [4:24:29<7:01:35,  1.57it/s]
 39%|███▉      | 25701/65536 [4:24:29<6:44:36,  1.64it/s]
 39%|███▉      | 25702/65536 [4:24:30<6:44:52,  1.64it/s]
 39%|███▉      | 25703/65536 [4:24:31<6:44:48,  1.64it/s]
 39%|███▉      | 25704/65536 [4:24:31<6:35:10,  1.68it/s]
 39%|███▉      | 25705/65536 [4:24:32<6:30:38,  1.70it/s]
 39%|███▉      | 25706/65536 [4:24:32<6:30:15,  1.70it/s]
 39%|███▉      | 25707/65536 [4:24:33<6:37:25,  1.67it/s]
 39%|███▉      | 25708/65536 [4:24:33<6:37:34,  1.67it/s]
 39%|███▉      | 25709/65536 [4:24:34<6:31:50,  1.69it/s]
 39%|███▉      | 25710/65536 [4:24:35<6:46:05,  1.63it/s]
 39%|███▉      | 25711/65536 [4:24:35<6:45:41,  1.64it/s]
 39%|███▉      | 25712/65536 [4:24:36<6:55:26,  1.60it/s]
 39%|███▉      | 25713/65536 [4:24:37<7:02:03,  1.57it/s]
 39%|███▉      | 25714/65536 [4:24:37<6:59:24,  1.58it/s]
 39%|███▉      | 25715/65536 [4:24:38<6:47:22,  1.63it/s]
 39%|███▉      | 25716/65536 [4:24:38<6:55:09,  1.60it/s]
 39%|███▉      | 25717/65536 [4:24:39<7:06:57,  1.55it/s]
 39%|███▉      | 25718/65536 [4:24:40<6:52:16,  1.61it/s]
 39%|███▉      | 25719/65536 [4:24:40<6:48:46,  1.62it/s]
 39%|███▉      | 25720/65536 [4:24:41<6:44:39,  1.64it/s]
                                                         
{'loss': 2.8488, 'learning_rate': 6.484633280274273e-07, 'epoch': 1587.65}
+
 39%|███▉      | 25720/65536 [4:24:41<6:44:39,  1.64it/s]
 39%|███▉      | 25721/65536 [4:24:42<6:40:15,  1.66it/s]
 39%|███▉      | 25722/65536 [4:24:42<6:28:29,  1.71it/s]
 39%|███▉      | 25723/65536 [4:24:43<6:26:14,  1.72it/s]
 39%|███▉      | 25724/65536 [4:24:43<6:26:17,  1.72it/s]
 39%|███▉      | 25725/65536 [4:24:44<6:35:43,  1.68it/s]
 39%|███▉      | 25726/65536 [4:24:45<6:50:14,  1.62it/s]
 39%|███▉      | 25727/65536 [4:24:45<6:47:56,  1.63it/s]
 39%|███▉      | 25728/65536 [4:24:46<6:57:17,  1.59it/s]
 39%|███▉      | 25729/65536 [4:24:46<6:59:12,  1.58it/s]
 39%|███▉      | 25730/65536 [4:24:47<6:46:22,  1.63it/s]
 39%|███▉      | 25731/65536 [4:24:48<6:36:08,  1.67it/s]
 39%|███▉      | 25732/65536 [4:24:48<6:32:24,  1.69it/s]
 39%|███▉      | 25733/65536 [4:24:49<6:51:33,  1.61it/s]
 39%|███▉      | 25734/65536 [4:24:49<6:39:31,  1.66it/s]
 39%|███▉      | 25735/65536 [4:24:50<6:38:28,  1.66it/s]
 39%|███▉      | 25736/65536 [4:24:51<6:43:08,  1.65it/s]
 39%|███▉      | 25737/65536 [4:24:51<6:47:24,  1.63it/s]
 39%|███▉      | 25738/65536 [4:24:52<6:35:17,  1.68it/s]
 39%|███▉      | 25739/65536 [4:24:52<6:37:21,  1.67it/s]
 39%|███▉      | 25740/65536 [4:24:53<6:35:22,  1.68it/s]
                                                         
{'loss': 2.8975, 'learning_rate': 6.481878290682012e-07, 'epoch': 1588.89}
+
 39%|███▉      | 25740/65536 [4:24:53<6:35:22,  1.68it/s]
 39%|███▉      | 25741/65536 [4:24:54<6:33:09,  1.69it/s]
 39%|███▉      | 25742/65536 [4:24:54<6:52:35,  1.61it/s]
 39%|███▉      | 25743/65536 [4:24:55<6:52:11,  1.61it/s]
 39%|███▉      | 25744/65536 [4:24:55<6:50:56,  1.61it/s]
 39%|███▉      | 25745/65536 [4:24:56<7:03:47,  1.56it/s]
 39%|███▉      | 25746/65536 [4:24:57<6:58:28,  1.58it/s]
 39%|███▉      | 25747/65536 [4:24:57<6:51:21,  1.61it/s]
 39%|███▉      | 25748/65536 [4:24:58<6:55:20,  1.60it/s]
 39%|███▉      | 25749/65536 [4:24:59<6:50:46,  1.61it/s]
 39%|███▉      | 25750/65536 [4:24:59<6:47:06,  1.63it/s]
 39%|███▉      | 25751/65536 [4:25:00<6:46:44,  1.63it/s]
 39%|███▉      | 25752/65536 [4:25:00<6:42:27,  1.65it/s]
 39%|███▉      | 25753/65536 [4:25:01<6:47:10,  1.63it/s]
 39%|███▉      | 25754/65536 [4:25:02<6:33:33,  1.68it/s]
 39%|███▉      | 25755/65536 [4:25:02<6:38:53,  1.66it/s]
 39%|███▉      | 25756/65536 [4:25:03<6:32:30,  1.69it/s]
 39%|███▉      | 25757/65536 [4:25:03<6:40:42,  1.65it/s]
 39%|███▉      | 25758/65536 [4:25:04<6:41:23,  1.65it/s]
 39%|███▉      | 25759/65536 [4:25:05<6:52:28,  1.61it/s]
 39%|███▉      | 25760/65536 [4:25:05<6:51:29,  1.61it/s]
                                                         
{'loss': 2.7681, 'learning_rate': 6.47912330108975e-07, 'epoch': 1590.12}
+
 39%|███▉      | 25760/65536 [4:25:05<6:51:29,  1.61it/s]
 39%|███▉      | 25761/65536 [4:25:06<6:54:55,  1.60it/s]
 39%|███▉      | 25762/65536 [4:25:07<6:50:20,  1.62it/s]
 39%|███▉      | 25763/65536 [4:25:07<6:48:58,  1.62it/s]
 39%|███▉      | 25764/65536 [4:25:08<6:51:34,  1.61it/s]
 39%|███▉      | 25765/65536 [4:25:08<6:45:59,  1.63it/s]
 39%|███▉      | 25766/65536 [4:25:09<6:36:39,  1.67it/s]
 39%|███▉      | 25767/65536 [4:25:10<6:34:52,  1.68it/s]
 39%|███▉      | 25768/65536 [4:25:10<6:36:40,  1.67it/s]
 39%|███▉      | 25769/65536 [4:25:11<6:44:40,  1.64it/s]
 39%|███▉      | 25770/65536 [4:25:11<6:54:08,  1.60it/s]
 39%|███▉      | 25771/65536 [4:25:12<6:54:53,  1.60it/s]
 39%|███▉      | 25772/65536 [4:25:13<7:01:39,  1.57it/s]
 39%|███▉      | 25773/65536 [4:25:13<7:00:44,  1.58it/s]
 39%|███▉      | 25774/65536 [4:25:14<6:52:58,  1.60it/s]
 39%|███▉      | 25775/65536 [4:25:15<7:11:28,  1.54it/s]
 39%|███▉      | 25776/65536 [4:25:15<7:12:01,  1.53it/s]
 39%|███▉      | 25777/65536 [4:25:16<7:16:22,  1.52it/s]
 39%|███▉      | 25778/65536 [4:25:17<7:05:51,  1.56it/s]
 39%|███▉      | 25779/65536 [4:25:17<7:02:41,  1.57it/s]
 39%|███▉      | 25780/65536 [4:25:18<7:11:29,  1.54it/s]
                                                         
{'loss': 2.3341, 'learning_rate': 6.476368311497489e-07, 'epoch': 1591.36}
+
 39%|███▉      | 25780/65536 [4:25:18<7:11:29,  1.54it/s]
 39%|███▉      | 25781/65536 [4:25:18<6:56:40,  1.59it/s]
 39%|███▉      | 25782/65536 [4:25:19<6:41:07,  1.65it/s]
 39%|███▉      | 25783/65536 [4:25:20<6:38:38,  1.66it/s]
 39%|███▉      | 25784/65536 [4:25:20<6:31:47,  1.69it/s]
 39%|███▉      | 25785/65536 [4:25:21<6:37:31,  1.67it/s]
 39%|███▉      | 25786/65536 [4:25:21<6:33:09,  1.69it/s]
 39%|███▉      | 25787/65536 [4:25:22<6:33:57,  1.68it/s]
 39%|███▉      | 25788/65536 [4:25:23<6:27:08,  1.71it/s]
 39%|███▉      | 25789/65536 [4:25:23<6:31:28,  1.69it/s]
 39%|███▉      | 25790/65536 [4:25:24<6:27:24,  1.71it/s]
 39%|███▉      | 25791/65536 [4:25:24<6:59:31,  1.58it/s]
 39%|███▉      | 25792/65536 [4:25:25<6:50:08,  1.62it/s]
 39%|███▉      | 25793/65536 [4:25:26<6:44:36,  1.64it/s]
 39%|███▉      | 25794/65536 [4:25:26<6:44:20,  1.64it/s]
 39%|███▉      | 25795/65536 [4:25:27<6:41:09,  1.65it/s]
 39%|███▉      | 25796/65536 [4:25:27<6:42:02,  1.65it/s]
 39%|███▉      | 25797/65536 [4:25:28<6:54:09,  1.60it/s]
 39%|███▉      | 25798/65536 [4:25:29<6:50:08,  1.61it/s]
 39%|███▉      | 25799/65536 [4:25:29<6:41:43,  1.65it/s]
 39%|███▉      | 25800/65536 [4:25:30<6:52:40,  1.60it/s]
                                                         
{'loss': 2.1036, 'learning_rate': 6.473613321905228e-07, 'epoch': 1592.59}
+
 39%|███▉      | 25800/65536 [4:25:30<6:52:40,  1.60it/s]
 39%|███▉      | 25801/65536 [4:25:31<6:53:04,  1.60it/s]
 39%|███▉      | 25802/65536 [4:25:31<6:57:23,  1.59it/s]
 39%|███▉      | 25803/65536 [4:25:32<6:51:57,  1.61it/s]
 39%|███▉      | 25804/65536 [4:25:32<6:45:29,  1.63it/s]
 39%|███▉      | 25805/65536 [4:25:33<6:32:08,  1.69it/s]
 39%|███▉      | 25806/65536 [4:25:34<6:30:56,  1.69it/s]
 39%|███▉      | 25807/65536 [4:25:34<6:47:46,  1.62it/s]
 39%|███▉      | 25808/65536 [4:25:35<6:51:28,  1.61it/s]
 39%|███▉      | 25809/65536 [4:25:36<6:49:03,  1.62it/s]
 39%|███▉      | 25810/65536 [4:25:36<6:49:01,  1.62it/s]
 39%|███▉      | 25811/65536 [4:25:37<6:42:25,  1.65it/s]
 39%|███▉      | 25812/65536 [4:25:37<6:36:58,  1.67it/s]
 39%|███▉      | 25813/65536 [4:25:38<6:48:19,  1.62it/s]
 39%|███▉      | 25814/65536 [4:25:39<6:40:29,  1.65it/s]
 39%|███▉      | 25815/65536 [4:25:39<6:48:08,  1.62it/s]
 39%|███▉      | 25816/65536 [4:25:40<6:51:45,  1.61it/s]
 39%|███▉      | 25817/65536 [4:25:40<6:43:51,  1.64it/s]
 39%|███▉      | 25818/65536 [4:25:41<6:44:26,  1.64it/s]
 39%|███▉      | 25819/65536 [4:25:42<6:46:39,  1.63it/s]
 39%|███▉      | 25820/65536 [4:25:42<6:42:20,  1.65it/s]
                                                         
{'loss': 2.0129, 'learning_rate': 6.470858332312966e-07, 'epoch': 1593.83}
+
 39%|███▉      | 25820/65536 [4:25:42<6:42:20,  1.65it/s]
 39%|███▉      | 25821/65536 [4:25:43<6:48:41,  1.62it/s]
 39%|███▉      | 25822/65536 [4:25:43<6:38:28,  1.66it/s]
 39%|███▉      | 25823/65536 [4:25:44<6:51:22,  1.61it/s]
 39%|███▉      | 25824/65536 [4:25:45<6:40:19,  1.65it/s]
 39%|███▉      | 25825/65536 [4:25:45<6:36:07,  1.67it/s]
 39%|███▉      | 25826/65536 [4:25:46<6:41:54,  1.65it/s]
 39%|███▉      | 25827/65536 [4:25:46<6:40:44,  1.65it/s]
 39%|███▉      | 25828/65536 [4:25:47<6:58:19,  1.58it/s]
 39%|███▉      | 25829/65536 [4:25:48<6:52:52,  1.60it/s]
 39%|███▉      | 25830/65536 [4:25:48<6:50:25,  1.61it/s]
 39%|███▉      | 25831/65536 [4:25:49<6:46:14,  1.63it/s]
 39%|███▉      | 25832/65536 [4:25:50<6:52:26,  1.60it/s]
 39%|███▉      | 25833/65536 [4:25:50<6:51:38,  1.61it/s]
 39%|███▉      | 25834/65536 [4:25:51<6:47:24,  1.62it/s]
 39%|███▉      | 25835/65536 [4:25:51<6:52:32,  1.60it/s]
 39%|███▉      | 25836/65536 [4:25:52<6:47:32,  1.62it/s]
 39%|███▉      | 25837/65536 [4:25:53<6:44:04,  1.64it/s]
 39%|███▉      | 25838/65536 [4:25:53<6:49:53,  1.61it/s]
 39%|███▉      | 25839/65536 [4:25:54<6:37:29,  1.66it/s]
 39%|███▉      | 25840/65536 [4:25:55<7:08:16,  1.54it/s]
                                                         
{'loss': 2.0491, 'learning_rate': 6.468103342720705e-07, 'epoch': 1595.06}
+
 39%|███▉      | 25840/65536 [4:25:55<7:08:16,  1.54it/s]
 39%|███▉      | 25841/65536 [4:25:55<6:57:55,  1.58it/s]
 39%|███▉      | 25842/65536 [4:25:56<6:54:06,  1.60it/s]
 39%|███▉      | 25843/65536 [4:25:56<6:45:52,  1.63it/s]
 39%|███▉      | 25844/65536 [4:25:57<6:45:28,  1.63it/s]
 39%|███▉      | 25845/65536 [4:25:58<6:45:40,  1.63it/s]
 39%|███▉      | 25846/65536 [4:25:58<6:36:13,  1.67it/s]
 39%|███▉      | 25847/65536 [4:25:59<6:44:49,  1.63it/s]
 39%|███▉      | 25848/65536 [4:25:59<6:40:07,  1.65it/s]
 39%|███▉      | 25849/65536 [4:26:00<6:39:30,  1.66it/s]
 39%|███▉      | 25850/65536 [4:26:01<6:42:10,  1.64it/s]
 39%|███▉      | 25851/65536 [4:26:01<6:54:30,  1.60it/s]
 39%|███▉      | 25852/65536 [4:26:02<6:53:00,  1.60it/s]
 39%|███▉      | 25853/65536 [4:26:03<6:57:25,  1.58it/s]
 39%|███▉      | 25854/65536 [4:26:03<6:45:40,  1.63it/s]
 39%|███▉      | 25855/65536 [4:26:04<6:40:23,  1.65it/s]
 39%|███▉      | 25856/65536 [4:26:04<6:51:38,  1.61it/s]
 39%|███▉      | 25857/65536 [4:26:05<6:44:18,  1.64it/s]
 39%|███▉      | 25858/65536 [4:26:06<6:43:45,  1.64it/s]
 39%|███▉      | 25859/65536 [4:26:06<6:37:40,  1.66it/s]
 39%|███▉      | 25860/65536 [4:26:07<6:38:48,  1.66it/s]
                                                         
{'loss': 2.0864, 'learning_rate': 6.465348353128443e-07, 'epoch': 1596.3}
+
 39%|███▉      | 25860/65536 [4:26:07<6:38:48,  1.66it/s]
 39%|███▉      | 25861/65536 [4:26:07<6:41:44,  1.65it/s]
 39%|███▉      | 25862/65536 [4:26:08<6:43:32,  1.64it/s]
 39%|███▉      | 25863/65536 [4:26:09<6:40:21,  1.65it/s]
 39%|███▉      | 25864/65536 [4:26:09<6:40:47,  1.65it/s]
 39%|███▉      | 25865/65536 [4:26:10<6:37:58,  1.66it/s]
 39%|███▉      | 25866/65536 [4:26:10<6:34:59,  1.67it/s]
 39%|███▉      | 25867/65536 [4:26:11<6:42:39,  1.64it/s]
 39%|███▉      | 25868/65536 [4:26:12<6:42:22,  1.64it/s]
 39%|███▉      | 25869/65536 [4:26:12<6:44:54,  1.63it/s]
 39%|███▉      | 25870/65536 [4:26:13<6:45:51,  1.63it/s]
 39%|███▉      | 25871/65536 [4:26:13<6:41:06,  1.65it/s]
 39%|███▉      | 25872/65536 [4:26:14<7:09:28,  1.54it/s]
 39%|███▉      | 25873/65536 [4:26:15<6:57:55,  1.58it/s]
 39%|█��█▉      | 25874/65536 [4:26:15<6:43:11,  1.64it/s]
 39%|███▉      | 25875/65536 [4:26:16<6:38:27,  1.66it/s]
 39%|███▉      | 25876/65536 [4:26:17<6:47:11,  1.62it/s]
 39%|███▉      | 25877/65536 [4:26:17<6:41:49,  1.64it/s]
 39%|███▉      | 25878/65536 [4:26:18<6:48:44,  1.62it/s]
 39%|███▉      | 25879/65536 [4:26:18<6:51:19,  1.61it/s]
 39%|███▉      | 25880/65536 [4:26:19<6:43:18,  1.64it/s]
                                                         
{'loss': 2.1008, 'learning_rate': 6.462593363536182e-07, 'epoch': 1597.53}
+
 39%|███▉      | 25880/65536 [4:26:19<6:43:18,  1.64it/s]
 39%|███▉      | 25881/65536 [4:26:20<6:35:52,  1.67it/s]
 39%|███▉      | 25882/65536 [4:26:20<6:28:48,  1.70it/s]
 39%|███▉      | 25883/65536 [4:26:21<6:29:51,  1.70it/s]
 39%|███▉      | 25884/65536 [4:26:21<6:33:52,  1.68it/s]
 39%|███▉      | 25885/65536 [4:26:22<6:25:51,  1.71it/s]
 39%|███▉      | 25886/65536 [4:26:23<6:36:24,  1.67it/s]
 40%|███▉      | 25887/65536 [4:26:23<6:44:06,  1.64it/s]
 40%|███▉      | 25888/65536 [4:26:24<6:49:15,  1.61it/s]
 40%|███▉      | 25889/65536 [4:26:24<6:49:39,  1.61it/s]
 40%|███▉      | 25890/65536 [4:26:25<6:51:51,  1.60it/s]
 40%|███▉      | 25891/65536 [4:26:26<6:47:26,  1.62it/s]
 40%|███▉      | 25892/65536 [4:26:26<6:37:11,  1.66it/s]
 40%|███▉      | 25893/65536 [4:26:27<6:30:13,  1.69it/s]
 40%|███▉      | 25894/65536 [4:26:28<6:41:41,  1.64it/s]
 40%|███▉      | 25895/65536 [4:26:28<6:39:30,  1.65it/s]
 40%|███▉      | 25896/65536 [4:26:29<6:27:24,  1.71it/s]
 40%|███▉      | 25897/65536 [4:26:29<6:40:20,  1.65it/s]
 40%|███▉      | 25898/65536 [4:26:30<6:41:54,  1.64it/s]
 40%|███▉      | 25899/65536 [4:26:30<6:32:48,  1.68it/s]
 40%|███▉      | 25900/65536 [4:26:31<6:38:36,  1.66it/s]
                                                         
{'loss': 1.9265, 'learning_rate': 6.459838373943921e-07, 'epoch': 1598.77}
+
 40%|███▉      | 25900/65536 [4:26:31<6:38:36,  1.66it/s]
 40%|███▉      | 25901/65536 [4:26:32<6:42:47,  1.64it/s]
 40%|███▉      | 25902/65536 [4:26:32<6:43:32,  1.64it/s]
 40%|███▉      | 25903/65536 [4:26:33<6:50:02,  1.61it/s]
 40%|███▉      | 25904/65536 [4:26:34<7:04:02,  1.56it/s]
 40%|███▉      | 25905/65536 [4:26:34<7:02:14,  1.56it/s]
 40%|███▉      | 25906/65536 [4:26:35<6:54:11,  1.59it/s]
 40%|███▉      | 25907/65536 [4:26:36<6:52:26,  1.60it/s]
 40%|███▉      | 25908/65536 [4:26:36<6:42:44,  1.64it/s]
 40%|███▉      | 25909/65536 [4:26:37<6:33:45,  1.68it/s]
 40%|███▉      | 25910/65536 [4:26:37<6:35:07,  1.67it/s]
 40%|███▉      | 25911/65536 [4:26:38<6:50:52,  1.61it/s]
 40%|███▉      | 25912/65536 [4:26:39<7:00:43,  1.57it/s]
 40%|███▉      | 25913/65536 [4:26:39<6:52:52,  1.60it/s]
 40%|███▉      | 25914/65536 [4:26:40<6:44:30,  1.63it/s]
 40%|███▉      | 25915/65536 [4:26:40<6:51:29,  1.60it/s]
 40%|███▉      | 25916/65536 [4:26:41<6:43:23,  1.64it/s]
 40%|███▉      | 25917/65536 [4:26:42<6:30:57,  1.69it/s]
 40%|███▉      | 25918/65536 [4:26:42<6:30:44,  1.69it/s]
 40%|███▉      | 25919/65536 [4:26:43<6:27:50,  1.70it/s]
 40%|███▉      | 25920/65536 [4:26:43<6:42:55,  1.64it/s]
                                                         
{'loss': 1.8548, 'learning_rate': 6.457083384351659e-07, 'epoch': 1600.0}
+
 40%|███▉      | 25920/65536 [4:26:43<6:42:55,  1.64it/s]
 40%|███▉      | 25921/65536 [4:26:44<7:02:02,  1.56it/s]
 40%|███▉      | 25922/65536 [4:26:45<6:58:02,  1.58it/s]
 40%|███▉      | 25923/65536 [4:26:45<7:10:46,  1.53it/s]
 40%|███▉      | 25924/65536 [4:26:46<7:02:28,  1.56it/s]
 40%|███▉      | 25925/65536 [4:26:47<7:03:50,  1.56it/s]
 40%|███▉      | 25926/65536 [4:26:47<6:52:29,  1.60it/s]
 40%|███▉      | 25927/65536 [4:26:48<6:41:16,  1.65it/s]
 40%|███▉      | 25928/65536 [4:26:48<6:42:28,  1.64it/s]
 40%|███▉      | 25929/65536 [4:26:49<6:37:20,  1.66it/s]
 40%|███▉      | 25930/65536 [4:26:50<6:34:11,  1.67it/s]
 40%|███▉      | 25931/65536 [4:26:50<6:33:59,  1.68it/s]
 40%|███▉      | 25932/65536 [4:26:51<6:38:34,  1.66it/s]
 40%|███▉      | 25933/65536 [4:26:51<6:36:00,  1.67it/s]
 40%|███▉      | 25934/65536 [4:26:52<6:39:38,  1.65it/s]
 40%|███▉      | 25935/65536 [4:26:53<6:38:57,  1.65it/s]
 40%|███▉      | 25936/65536 [4:26:53<6:47:50,  1.62it/s]
 40%|███▉      | 25937/65536 [4:26:54<6:57:19,  1.58it/s]
 40%|███▉      | 25938/65536 [4:26:55<6:53:14,  1.60it/s]
 40%|███▉      | 25939/65536 [4:26:55<6:45:31,  1.63it/s]
 40%|███▉      | 25940/65536 [4:26:56<6:39:54,  1.65it/s]
                                                         
{'loss': 1.8596, 'learning_rate': 6.454328394759398e-07, 'epoch': 1601.23}
+
 40%|███▉      | 25940/65536 [4:26:56<6:39:54,  1.65it/s]
 40%|███▉      | 25941/65536 [4:26:56<6:37:38,  1.66it/s]
 40%|███▉      | 25942/65536 [4:26:57<6:44:18,  1.63it/s]
 40%|███▉      | 25943/65536 [4:26:58<6:40:24,  1.65it/s]
 40%|███▉      | 25944/65536 [4:26:58<6:34:15,  1.67it/s]
 40%|███▉      | 25945/65536 [4:26:59<6:32:29,  1.68it/s]
 40%|███▉      | 25946/65536 [4:26:59<6:50:13,  1.61it/s]
 40%|███▉      | 25947/65536 [4:27:00<6:50:40,  1.61it/s]
 40%|███▉      | 25948/65536 [4:27:01<6:50:47,  1.61it/s]
 40%|███▉      | 25949/65536 [4:27:01<6:48:25,  1.62it/s]
 40%|███▉      | 25950/65536 [4:27:02<6:39:26,  1.65it/s]
 40%|███▉      | 25951/65536 [4:27:02<6:39:50,  1.65it/s]
 40%|███▉      | 25952/65536 [4:27:03<6:45:38,  1.63it/s]
 40%|███▉      | 25953/65536 [4:27:04<6:54:43,  1.59it/s]
 40%|███▉      | 25954/65536 [4:27:04<6:43:14,  1.64it/s]
 40%|███▉      | 25955/65536 [4:27:05<6:37:13,  1.66it/s]
 40%|███▉      | 25956/65536 [4:27:06<6:38:23,  1.66it/s]
 40%|███▉      | 25957/65536 [4:27:06<6:56:26,  1.58it/s]
 40%|███▉      | 25958/65536 [4:27:07<6:53:26,  1.60it/s]
 40%|███▉      | 25959/65536 [4:27:07<6:44:41,  1.63it/s]
 40%|███▉      | 25960/65536 [4:27:08<6:44:13,  1.63it/s]
                                                         
{'loss': 1.7976, 'learning_rate': 6.451573405167136e-07, 'epoch': 1602.47}
+
 40%|███▉      | 25960/65536 [4:27:08<6:44:13,  1.63it/s]
 40%|███▉      | 25961/65536 [4:27:09<6:41:45,  1.64it/s]
 40%|███▉      | 25962/65536 [4:27:09<6:40:41,  1.65it/s]
 40%|███▉      | 25963/65536 [4:27:10<6:37:36,  1.66it/s]
 40%|███▉      | 25964/65536 [4:27:10<6:34:28,  1.67it/s]
 40%|███▉      | 25965/65536 [4:27:11<6:37:51,  1.66it/s]
 40%|███▉      | 25966/65536 [4:27:12<6:41:18,  1.64it/s]
 40%|███▉      | 25967/65536 [4:27:12<7:03:23,  1.56it/s]
 40%|███▉      | 25968/65536 [4:27:13<6:53:02,  1.60it/s]
 40%|███▉      | 25969/65536 [4:27:14<7:01:07,  1.57it/s]
 40%|███▉      | 25970/65536 [4:27:14<6:57:38,  1.58it/s]
 40%|███▉      | 25971/65536 [4:27:15<7:01:32,  1.56it/s]
 40%|███▉      | 25972/65536 [4:27:16<7:04:05,  1.55it/s]
 40%|███▉      | 25973/65536 [4:27:16<6:47:43,  1.62it/s]
 40%|███▉      | 25974/65536 [4:27:17<6:42:03,  1.64it/s]
 40%|███▉      | 25975/65536 [4:27:17<6:40:54,  1.64it/s]
 40%|███▉      | 25976/65536 [4:27:18<6:39:21,  1.65it/s]
 40%|███▉      | 25977/65536 [4:27:19<6:45:02,  1.63it/s]
 40%|███▉      | 25978/65536 [4:27:19<6:38:50,  1.65it/s]
 40%|███▉      | 25979/65536 [4:27:20<6:42:10,  1.64it/s]
 40%|███▉      | 25980/65536 [4:27:20<6:43:17,  1.63it/s]
                                                         
{'loss': 1.7973, 'learning_rate': 6.448818415574874e-07, 'epoch': 1603.7}
+
 40%|███▉      | 25980/65536 [4:27:20<6:43:17,  1.63it/s]
 40%|███▉      | 25981/65536 [4:27:21<6:37:10,  1.66it/s]
 40%|███▉      | 25982/65536 [4:27:22<6:33:11,  1.68it/s]
 40%|███▉      | 25983/65536 [4:27:22<6:45:27,  1.63it/s]
 40%|███▉      | 25984/65536 [4:27:23<6:46:59,  1.62it/s]
 40%|███▉      | 25985/65536 [4:27:23<6:55:15,  1.59it/s]
 40%|███▉      | 25986/65536 [4:27:24<6:48:58,  1.61it/s]
 40%|███▉      | 25987/65536 [4:27:25<6:45:23,  1.63it/s]
 40%|███▉      | 25988/65536 [4:27:25<6:43:29,  1.63it/s]
 40%|███▉      | 25989/65536 [4:27:26<6:41:32,  1.64it/s]
 40%|███▉      | 25990/65536 [4:27:26<6:35:02,  1.67it/s]
 40%|███▉      | 25991/65536 [4:27:27<6:37:53,  1.66it/s]
 40%|███▉      | 25992/65536 [4:27:28<6:30:43,  1.69it/s]
 40%|███▉      | 25993/65536 [4:27:28<6:37:57,  1.66it/s]
 40%|███▉      | 25994/65536 [4:27:29<6:36:10,  1.66it/s]
 40%|███▉      | 25995/65536 [4:27:29<6:44:08,  1.63it/s]
 40%|███▉      | 25996/65536 [4:27:30<6:37:47,  1.66it/s]
 40%|███▉      | 25997/65536 [4:27:31<6:35:30,  1.67it/s]
 40%|███▉      | 25998/65536 [4:27:31<6:53:21,  1.59it/s]
 40%|███▉      | 25999/65536 [4:27:32<6:51:11,  1.60it/s]
 40%|███▉      | 26000/65536 [4:27:33<6:49:19,  1.61it/s]
                                                         
{'loss': 1.8429, 'learning_rate': 6.446063425982611e-07, 'epoch': 1604.94}
+
 40%|███▉      | 26000/65536 [4:27:33<6:49:19,  1.61it/s]
 40%|███▉      | 26001/65536 [4:27:33<6:44:31,  1.63it/s]
 40%|███▉      | 26002/65536 [4:27:34<6:51:54,  1.60it/s]
 40%|███▉      | 26003/65536 [4:27:34<6:37:59,  1.66it/s]
 40%|███▉      | 26004/65536 [4:27:35<6:37:59,  1.66it/s]
 40%|███▉      | 26005/65536 [4:27:36<6:52:56,  1.60it/s]
 40%|███▉      | 26006/65536 [4:27:36<6:55:16,  1.59it/s]
 40%|███▉      | 26007/65536 [4:27:37<6:51:42,  1.60it/s]
 40%|███▉      | 26008/65536 [4:27:38<6:52:00,  1.60it/s]
 40%|███▉      | 26009/65536 [4:27:38<6:49:13,  1.61it/s]
 40%|███▉      | 26010/65536 [4:27:39<6:46:43,  1.62it/s]
 40%|███▉      | 26011/65536 [4:27:39<6:44:41,  1.63it/s]
 40%|███▉      | 26012/65536 [4:27:40<6:35:49,  1.66it/s]
 40%|███▉      | 26013/65536 [4:27:41<6:35:59,  1.66it/s]
 40%|███▉      | 26014/65536 [4:27:41<6:44:02,  1.63it/s]
 40%|███▉      | 26015/65536 [4:27:42<6:39:43,  1.65it/s]
 40%|███▉      | 26016/65536 [4:27:42<6:36:04,  1.66it/s]
 40%|███▉      | 26017/65536 [4:27:43<6:34:48,  1.67it/s]
 40%|███▉      | 26018/65536 [4:27:44<6:46:09,  1.62it/s]
 40%|███▉      | 26019/65536 [4:27:44<6:39:24,  1.65it/s]
 40%|███▉      | 26020/65536 [4:27:45<6:32:33,  1.68it/s]
                                                         
{'loss': 1.8667, 'learning_rate': 6.443308436390351e-07, 'epoch': 1606.17}
+
 40%|███▉      | 26020/65536 [4:27:45<6:32:33,  1.68it/s]
 40%|███▉      | 26021/65536 [4:27:45<6:34:38,  1.67it/s]
 40%|███▉      | 26022/65536 [4:27:46<6:45:44,  1.62it/s]
 40%|███▉      | 26023/65536 [4:27:47<6:52:06,  1.60it/s]
 40%|███▉      | 26024/65536 [4:27:47<6:42:23,  1.64it/s]
 40%|███▉      | 26025/65536 [4:27:48<6:38:08,  1.65it/s]
 40%|███▉      | 26026/65536 [4:27:48<6:34:03,  1.67it/s]
 40%|███▉      | 26027/65536 [4:27:49<6:36:54,  1.66it/s]
 40%|███▉      | 26028/65536 [4:27:50<6:48:34,  1.61it/s]
 40%|███▉      | 26029/65536 [4:27:50<6:52:03,  1.60it/s]
 40%|███▉      | 26030/65536 [4:27:51<6:46:23,  1.62it/s]
 40%|███▉      | 26031/65536 [4:27:51<6:32:33,  1.68it/s]
 40%|███▉      | 26032/65536 [4:27:52<6:39:25,  1.65it/s]
 40%|███▉      | 26033/65536 [4:27:53<6:45:19,  1.62it/s]
 40%|███▉      | 26034/65536 [4:27:53<6:57:13,  1.58it/s]
 40%|███▉      | 26035/65536 [4:27:54<7:08:00,  1.54it/s]
 40%|███▉      | 26036/65536 [4:27:55<7:00:47,  1.56it/s]
 40%|███▉      | 26037/65536 [4:27:55<6:56:24,  1.58it/s]
 40%|███▉      | 26038/65536 [4:27:56<6:47:54,  1.61it/s]
 40%|███▉      | 26039/65536 [4:27:57<6:42:42,  1.63it/s]
 40%|███▉      | 26040/65536 [4:27:57<6:43:59,  1.63it/s]
                                                         
{'loss': 1.8682, 'learning_rate': 6.44055344679809e-07, 'epoch': 1607.41}
+
 40%|███▉      | 26040/65536 [4:27:57<6:43:59,  1.63it/s]
 40%|███▉      | 26041/65536 [4:27:58<6:41:17,  1.64it/s]
 40%|███▉      | 26042/65536 [4:27:58<6:36:06,  1.66it/s]
 40%|███▉      | 26043/65536 [4:27:59<6:35:09,  1.67it/s]
 40%|███▉      | 26044/65536 [4:28:00<6:34:17,  1.67it/s]
 40%|███▉      | 26045/65536 [4:28:00<6:32:32,  1.68it/s]
 40%|███▉      | 26046/65536 [4:28:01<6:28:51,  1.69it/s]
 40%|███▉      | 26047/65536 [4:28:01<6:25:28,  1.71it/s]
 40%|███▉      | 26048/65536 [4:28:02<6:25:37,  1.71it/s]
 40%|███▉      | 26049/65536 [4:28:02<6:34:20,  1.67it/s]
 40%|███▉      | 26050/65536 [4:28:03<6:54:00,  1.59it/s]
 40%|███▉      | 26051/65536 [4:28:04<6:43:54,  1.63it/s]
 40%|███▉      | 26052/65536 [4:28:04<6:39:26,  1.65it/s]
 40%|███▉      | 26053/65536 [4:28:05<6:38:58,  1.65it/s]
 40%|███▉      | 26054/65536 [4:28:06<6:33:22,  1.67it/s]
 40%|███▉      | 26055/65536 [4:28:06<6:32:39,  1.68it/s]
 40%|███▉      | 26056/65536 [4:28:07<6:32:07,  1.68it/s]
 40%|███▉      | 26057/65536 [4:28:07<6:24:09,  1.71it/s]
 40%|███▉      | 26058/65536 [4:28:08<6:34:30,  1.67it/s]
 40%|███▉      | 26059/65536 [4:28:09<6:38:57,  1.65it/s]
 40%|███▉      | 26060/65536 [4:28:09<6:33:57,  1.67it/s]
                                                         
{'loss': 1.8548, 'learning_rate': 6.437798457205828e-07, 'epoch': 1608.64}
+
 40%|███▉      | 26060/65536 [4:28:09<6:33:57,  1.67it/s]
 40%|███▉      | 26061/65536 [4:28:10<6:39:49,  1.65it/s]
 40%|███▉      | 26062/65536 [4:28:10<6:36:58,  1.66it/s]
 40%|███▉      | 26063/65536 [4:28:11<6:37:33,  1.65it/s]
 40%|███▉      | 26064/65536 [4:28:12<6:35:33,  1.66it/s]
 40%|███▉      | 26065/65536 [4:28:12<6:44:18,  1.63it/s]
 40%|███▉      | 26066/65536 [4:28:13<7:10:33,  1.53it/s]
 40%|███▉      | 26067/65536 [4:28:14<6:57:18,  1.58it/s]
 40%|███▉      | 26068/65536 [4:28:14<6:51:24,  1.60it/s]
 40%|███▉      | 26069/65536 [4:28:15<6:56:07,  1.58it/s]
 40%|███▉      | 26070/65536 [4:28:15<6:48:47,  1.61it/s]
 40%|███▉      | 26071/65536 [4:28:16<6:40:55,  1.64it/s]
 40%|███▉      | 26072/65536 [4:28:17<6:51:50,  1.60it/s]
 40%|███▉      | 26073/65536 [4:28:17<7:01:39,  1.56it/s]
 40%|███▉      | 26074/65536 [4:28:18<6:55:22,  1.58it/s]
 40%|███▉      | 26075/65536 [4:28:19<6:53:44,  1.59it/s]
 40%|███▉      | 26076/65536 [4:28:19<6:48:01,  1.61it/s]
 40%|███▉      | 26077/65536 [4:28:20<6:47:49,  1.61it/s]
 40%|███▉      | 26078/65536 [4:28:20<6:50:41,  1.60it/s]
 40%|███▉      | 26079/65536 [4:28:21<6:40:40,  1.64it/s]
 40%|███▉      | 26080/65536 [4:28:22<6:36:28,  1.66it/s]
                                                         
{'loss': 1.787, 'learning_rate': 6.435043467613567e-07, 'epoch': 1609.88}
+
 40%|███▉      | 26080/65536 [4:28:22<6:36:28,  1.66it/s]
 40%|███▉      | 26081/65536 [4:28:22<6:38:11,  1.65it/s]
 40%|███▉      | 26082/65536 [4:28:23<6:31:49,  1.68it/s]
 40%|███▉      | 26083/65536 [4:28:23<6:47:50,  1.61it/s]
 40%|███▉      | 26084/65536 [4:28:24<6:46:35,  1.62it/s]
 40%|███▉      | 26085/65536 [4:28:25<6:39:42,  1.64it/s]
 40%|███▉      | 26086/65536 [4:28:25<6:39:24,  1.65it/s]
 40%|███▉      | 26087/65536 [4:28:26<6:28:16,  1.69it/s]
 40%|███▉      | 26088/65536 [4:28:26<6:21:39,  1.72it/s]
 40%|███▉      | 26089/65536 [4:28:27<6:21:14,  1.72it/s]
 40%|███▉      | 26090/65536 [4:28:28<6:30:57,  1.68it/s]
 40%|███▉      | 26091/65536 [4:28:28<6:43:42,  1.63it/s]
 40%|███▉      | 26092/65536 [4:28:29<6:35:52,  1.66it/s]
 40%|███▉      | 26093/65536 [4:28:29<6:45:51,  1.62it/s]
 40%|███▉      | 26094/65536 [4:28:30<6:50:30,  1.60it/s]
 40%|███▉      | 26095/65536 [4:28:31<6:53:08,  1.59it/s]
 40%|███▉      | 26096/65536 [4:28:31<6:43:22,  1.63it/s]
 40%|███▉      | 26097/65536 [4:28:32<6:44:52,  1.62it/s]
 40%|███▉      | 26098/65536 [4:28:33<6:48:23,  1.61it/s]
 40%|███▉      | 26099/65536 [4:28:33<7:07:15,  1.54it/s]
 40%|███▉      | 26100/65536 [4:28:34<7:02:06,  1.56it/s]
                                                         
{'loss': 1.8336, 'learning_rate': 6.432288478021305e-07, 'epoch': 1611.11}
+
 40%|███▉      | 26100/65536 [4:28:34<7:02:06,  1.56it/s]
 40%|███▉      | 26101/65536 [4:28:34<6:53:27,  1.59it/s]
 40%|███▉      | 26102/65536 [4:28:35<6:49:03,  1.61it/s]
 40%|███▉      | 26103/65536 [4:28:36<6:40:46,  1.64it/s]
 40%|███▉      | 26104/65536 [4:28:36<6:46:32,  1.62it/s]
 40%|███▉      | 26105/65536 [4:28:37<6:40:30,  1.64it/s]
 40%|███▉      | 26106/65536 [4:28:37<6:33:23,  1.67it/s]
 40%|███▉      | 26107/65536 [4:28:38<6:24:58,  1.71it/s]
 40%|███▉      | 26108/65536 [4:28:39<6:21:33,  1.72it/s]
 40%|███▉      | 26109/65536 [4:28:39<6:27:50,  1.69it/s]
 40%|███▉      | 26110/65536 [4:28:40<6:32:09,  1.68it/s]
 40%|███▉      | 26111/65536 [4:28:40<6:48:52,  1.61it/s]
 40%|███▉      | 26112/65536 [4:28:41<6:47:39,  1.61it/s]
 40%|███▉      | 26113/65536 [4:28:42<6:42:46,  1.63it/s]
 40%|███▉      | 26114/65536 [4:28:42<6:41:10,  1.64it/s]
 40%|███▉      | 26115/65536 [4:28:43<6:50:09,  1.60it/s]
 40%|███▉      | 26116/65536 [4:28:44<6:49:17,  1.61it/s]
 40%|███▉      | 26117/65536 [4:28:44<6:44:48,  1.62it/s]
 40%|███▉      | 26118/65536 [4:28:45<6:43:24,  1.63it/s]
 40%|███▉      | 26119/65536 [4:28:45<6:58:10,  1.57it/s]
 40%|███▉      | 26120/65536 [4:28:46<6:45:18,  1.62it/s]
                                                         
{'loss': 1.7849, 'learning_rate': 6.429533488429044e-07, 'epoch': 1612.35}
+
 40%|███▉      | 26120/65536 [4:28:46<6:45:18,  1.62it/s]
 40%|███▉      | 26121/65536 [4:28:47<6:37:26,  1.65it/s]
 40%|███▉      | 26122/65536 [4:28:47<6:43:36,  1.63it/s]
 40%|███▉      | 26123/65536 [4:28:48<6:43:43,  1.63it/s]
 40%|███▉      | 26124/65536 [4:28:49<6:47:34,  1.61it/s]
 40%|███▉      | 26125/65536 [4:28:49<6:40:13,  1.64it/s]
 40%|███▉      | 26126/65536 [4:28:50<6:40:19,  1.64it/s]
 40%|███▉      | 26127/65536 [4:28:50<6:51:53,  1.59it/s]
 40%|███▉      | 26128/65536 [4:28:51<6:34:30,  1.66it/s]
 40%|███▉      | 26129/65536 [4:28:52<6:46:00,  1.62it/s]
 40%|███▉      | 26130/65536 [4:28:52<6:33:31,  1.67it/s]
 40%|███▉      | 26131/65536 [4:28:53<6:52:36,  1.59it/s]
 40%|███▉      | 26132/65536 [4:28:53<6:49:51,  1.60it/s]
 40%|███▉      | 26133/65536 [4:28:54<6:41:37,  1.64it/s]
 40%|███▉      | 26134/65536 [4:28:55<6:36:28,  1.66it/s]
 40%|███▉      | 26135/65536 [4:28:55<6:37:19,  1.65it/s]
 40%|███▉      | 26136/65536 [4:28:56<6:34:32,  1.66it/s]
 40%|███▉      | 26137/65536 [4:28:56<6:33:00,  1.67it/s]
 40%|███▉      | 26138/65536 [4:28:57<6:47:07,  1.61it/s]
 40%|███▉      | 26139/65536 [4:28:58<6:50:16,  1.60it/s]
 40%|███▉      | 26140/65536 [4:28:58<6:55:16,  1.58it/s]
                                                         
{'loss': 1.7917, 'learning_rate': 6.426778498836783e-07, 'epoch': 1613.58}
+
 40%|███▉      | 26140/65536 [4:28:58<6:55:16,  1.58it/s]
 40%|███▉      | 26141/65536 [4:28:59<6:49:24,  1.60it/s]
 40%|███▉      | 26142/65536 [4:29:00<6:52:42,  1.59it/s]
 40%|███▉      | 26143/65536 [4:29:00<6:52:57,  1.59it/s]
 40%|███▉      | 26144/65536 [4:29:01<6:49:35,  1.60it/s]
 40%|███▉      | 26145/65536 [4:29:01<6:34:49,  1.66it/s]
 40%|███▉      | 26146/65536 [4:29:02<6:38:58,  1.65it/s]
 40%|███▉      | 26147/65536 [4:29:03<6:46:59,  1.61it/s]
 40%|███▉      | 26148/65536 [4:29:03<6:50:50,  1.60it/s]
 40%|███▉      | 26149/65536 [4:29:04<6:57:21,  1.57it/s]
 40%|███▉      | 26150/65536 [4:29:05<6:52:30,  1.59it/s]
 40%|███▉      | 26151/65536 [4:29:05<6:48:34,  1.61it/s]
 40%|███▉      | 26152/65536 [4:29:06<6:44:13,  1.62it/s]
 40%|███▉      | 26153/65536 [4:29:06<6:45:07,  1.62it/s]
 40%|███▉      | 26154/65536 [4:29:07<6:40:25,  1.64it/s]
 40%|███▉      | 26155/65536 [4:29:08<6:41:10,  1.64it/s]
 40%|███▉      | 26156/65536 [4:29:08<6:35:39,  1.66it/s]
 40%|███▉      | 26157/65536 [4:29:09<6:46:52,  1.61it/s]
 40%|███▉      | 26158/65536 [4:29:09<6:39:48,  1.64it/s]
 40%|███▉      | 26159/65536 [4:29:10<6:44:05,  1.62it/s]
 40%|███▉      | 26160/65536 [4:29:11<6:39:48,  1.64it/s]
                                                         
{'loss': 1.7625, 'learning_rate': 6.424023509244521e-07, 'epoch': 1614.81}
+
 40%|███▉      | 26160/65536 [4:29:11<6:39:48,  1.64it/s]
 40%|███▉      | 26161/65536 [4:29:11<6:38:24,  1.65it/s]
 40%|███▉      | 26162/65536 [4:29:12<6:29:23,  1.69it/s]
 40%|███▉      | 26163/65536 [4:29:12<6:41:38,  1.63it/s]
 40%|███▉      | 26164/65536 [4:29:13<6:46:17,  1.62it/s]
 40%|███▉      | 26165/65536 [4:29:14<6:38:46,  1.65it/s]
 40%|███▉      | 26166/65536 [4:29:14<6:50:57,  1.60it/s]
 40%|███▉      | 26167/65536 [4:29:15<6:58:20,  1.57it/s]
 40%|███▉      | 26168/65536 [4:29:16<6:50:53,  1.60it/s]
 40%|███▉      | 26169/65536 [4:29:16<6:45:22,  1.62it/s]
 40%|███▉      | 26170/65536 [4:29:17<6:31:36,  1.68it/s]
 40%|███▉      | 26171/65536 [4:29:17<6:29:04,  1.69it/s]
 40%|███▉      | 26172/65536 [4:29:18<6:34:31,  1.66it/s]
 40%|███▉      | 26173/65536 [4:29:19<6:34:07,  1.66it/s]
 40%|███▉      | 26174/65536 [4:29:19<6:37:20,  1.65it/s]
 40%|███▉      | 26175/65536 [4:29:20<6:48:56,  1.60it/s]
 40%|███▉      | 26176/65536 [4:29:20<6:44:17,  1.62it/s]
 40%|███▉      | 26177/65536 [4:29:21<6:52:20,  1.59it/s]
 40%|███▉      | 26178/65536 [4:29:22<6:42:03,  1.63it/s]
 40%|███▉      | 26179/65536 [4:29:22<6:41:19,  1.63it/s]
 40%|███▉      | 26180/65536 [4:29:23<6:55:57,  1.58it/s]
                                                         
{'loss': 1.8147, 'learning_rate': 6.421268519652259e-07, 'epoch': 1616.05}
+
 40%|███▉      | 26180/65536 [4:29:23<6:55:57,  1.58it/s]
 40%|███▉      | 26181/65536 [4:29:24<6:52:16,  1.59it/s]
 40%|███▉      | 26182/65536 [4:29:24<6:45:28,  1.62it/s]
 40%|███▉      | 26183/65536 [4:29:25<6:43:23,  1.63it/s]
 40%|███▉      | 26184/65536 [4:29:25<6:49:52,  1.60it/s]
 40%|███▉      | 26185/65536 [4:29:26<6:56:23,  1.58it/s]
 40%|███▉      | 26186/65536 [4:29:27<7:05:47,  1.54it/s]
 40%|███▉      | 26187/65536 [4:29:27<6:50:56,  1.60it/s]
 40%|███▉      | 26188/65536 [4:29:28<6:42:58,  1.63it/s]
 40%|███▉      | 26189/65536 [4:29:29<6:36:45,  1.65it/s]
 40%|███▉      | 26190/65536 [4:29:29<6:40:51,  1.64it/s]
 40%|███▉      | 26191/65536 [4:29:30<6:40:25,  1.64it/s]
 40%|███▉      | 26192/65536 [4:29:30<6:37:27,  1.65it/s]
 40%|███▉      | 26193/65536 [4:29:31<6:40:39,  1.64it/s]
 40%|███▉      | 26194/65536 [4:29:32<6:47:10,  1.61it/s]
 40%|███▉      | 26195/65536 [4:29:32<6:43:50,  1.62it/s]
 40%|███▉      | 26196/65536 [4:29:33<6:51:57,  1.59it/s]
 40%|███▉      | 26197/65536 [4:29:34<6:48:47,  1.60it/s]
 40%|███▉      | 26198/65536 [4:29:34<6:51:55,  1.59it/s]
 40%|███▉      | 26199/65536 [4:29:35<6:53:03,  1.59it/s]
 40%|███▉      | 26200/65536 [4:29:35<6:56:02,  1.58it/s]
                                                         
{'loss': 1.749, 'learning_rate': 6.418513530059997e-07, 'epoch': 1617.28}
+
 40%|███▉      | 26200/65536 [4:29:35<6:56:02,  1.58it/s]
 40%|███▉      | 26201/65536 [4:29:36<6:57:22,  1.57it/s]
 40%|███▉      | 26202/65536 [4:29:37<6:47:39,  1.61it/s]
 40%|███▉      | 26203/65536 [4:29:37<6:48:42,  1.60it/s]
 40%|███▉      | 26204/65536 [4:29:38<6:55:46,  1.58it/s]
 40%|███▉      | 26205/65536 [4:29:39<6:46:10,  1.61it/s]
 40%|███▉      | 26206/65536 [4:29:39<6:50:12,  1.60it/s]
 40%|███▉      | 26207/65536 [4:29:40<6:48:11,  1.61it/s]
 40%|███▉      | 26208/65536 [4:29:40<6:44:54,  1.62it/s]
 40%|███▉      | 26209/65536 [4:29:41<6:47:26,  1.61it/s]
 40%|███▉      | 26210/65536 [4:29:42<6:44:26,  1.62it/s]
 40%|███▉      | 26211/65536 [4:29:42<6:37:37,  1.65it/s]
 40%|███▉      | 26212/65536 [4:29:43<6:51:56,  1.59it/s]
 40%|███▉      | 26213/65536 [4:29:43<6:41:32,  1.63it/s]
 40%|███▉      | 26214/65536 [4:29:44<6:44:13,  1.62it/s]
 40%|████      | 26215/65536 [4:29:45<6:57:26,  1.57it/s]
 40%|████      | 26216/65536 [4:29:45<7:00:48,  1.56it/s]
 40%|████      | 26217/65536 [4:29:46<7:01:42,  1.55it/s]
 40%|████      | 26218/65536 [4:29:47<6:54:05,  1.58it/s]
 40%|████      | 26219/65536 [4:29:47<6:54:15,  1.58it/s]
 40%|████      | 26220/65536 [4:29:48<6:43:41,  1.62it/s]
                                                         
{'loss': 1.7624, 'learning_rate': 6.415758540467736e-07, 'epoch': 1618.52}
+
 40%|████      | 26220/65536 [4:29:48<6:43:41,  1.62it/s]
 40%|████      | 26221/65536 [4:29:49<6:41:41,  1.63it/s]
 40%|████      | 26222/65536 [4:29:49<6:50:14,  1.60it/s]
 40%|████      | 26223/65536 [4:29:50<6:49:19,  1.60it/s]
 40%|████      | 26224/65536 [4:29:51<7:09:33,  1.53it/s]
 40%|████      | 26225/65536 [4:29:51<6:59:36,  1.56it/s]
 40%|████      | 26226/65536 [4:29:52<6:50:41,  1.60it/s]
 40%|████      | 26227/65536 [4:29:52<6:37:54,  1.65it/s]
 40%|████      | 26228/65536 [4:29:53<6:47:49,  1.61it/s]
 40%|████      | 26229/65536 [4:29:53<6:35:28,  1.66it/s]
 40%|████      | 26230/65536 [4:29:54<6:44:48,  1.62it/s]
 40%|████      | 26231/65536 [4:29:55<6:50:09,  1.60it/s]
 40%|████      | 26232/65536 [4:29:55<6:52:54,  1.59it/s]
 40%|████      | 26233/65536 [4:29:56<6:53:56,  1.58it/s]
 40%|████      | 26234/65536 [4:29:57<6:43:24,  1.62it/s]
 40%|████      | 26235/65536 [4:29:57<6:35:54,  1.65it/s]
 40%|████      | 26236/65536 [4:29:58<6:33:27,  1.66it/s]
 40%|████      | 26237/65536 [4:29:58<6:29:40,  1.68it/s]
 40%|████      | 26238/65536 [4:29:59<6:42:27,  1.63it/s]
 40%|████      | 26239/65536 [4:30:00<6:40:39,  1.63it/s]
 40%|████      | 26240/65536 [4:30:00<6:32:58,  1.67it/s]
                                                         
{'loss': 1.7598, 'learning_rate': 6.413003550875474e-07, 'epoch': 1619.75}
+
 40%|████      | 26240/65536 [4:30:00<6:32:58,  1.67it/s]
 40%|████      | 26241/65536 [4:30:01<6:41:21,  1.63it/s]
 40%|████      | 26242/65536 [4:30:01<6:36:57,  1.65it/s]
 40%|████      | 26243/65536 [4:30:02<6:37:32,  1.65it/s]
 40%|████      | 26244/65536 [4:30:03<6:38:47,  1.64it/s]
 40%|████      | 26245/65536 [4:30:03<7:01:05,  1.56it/s]
 40%|████      | 26246/65536 [4:30:04<6:53:04,  1.59it/s]
 40%|████      | 26247/65536 [4:30:05<6:51:49,  1.59it/s]
 40%|████      | 26248/65536 [4:30:05<6:46:25,  1.61it/s]
 40%|████      | 26249/65536 [4:30:06<6:48:05,  1.60it/s]
 40%|████      | 26250/65536 [4:30:07<6:52:18,  1.59it/s]
 40%|████      | 26251/65536 [4:30:07<6:49:11,  1.60it/s]
 40%|████      | 26252/65536 [4:30:08<6:52:29,  1.59it/s]
 40%|████      | 26253/65536 [4:30:08<6:38:34,  1.64it/s]
 40%|████      | 26254/65536 [4:30:09<6:40:52,  1.63it/s]
 40%|████      | 26255/65536 [4:30:10<6:40:44,  1.63it/s]
 40%|████      | 26256/65536 [4:30:10<6:40:15,  1.64it/s]
 40%|████      | 26257/65536 [4:30:11<6:36:15,  1.65it/s]
 40%|████      | 26258/65536 [4:30:11<6:30:37,  1.68it/s]
 40%|████      | 26259/65536 [4:30:12<6:32:24,  1.67it/s]
 40%|████      | 26260/65536 [4:30:13<6:34:37,  1.66it/s]
                                                         
{'loss': 1.7928, 'learning_rate': 6.410248561283212e-07, 'epoch': 1620.99}
+
 40%|████      | 26260/65536 [4:30:13<6:34:37,  1.66it/s]
 40%|████      | 26261/65536 [4:30:13<6:49:33,  1.60it/s]
 40%|████      | 26262/65536 [4:30:14<6:42:16,  1.63it/s]
 40%|████      | 26263/65536 [4:30:14<6:46:25,  1.61it/s]
 40%|████      | 26264/65536 [4:30:15<6:36:37,  1.65it/s]
 40%|████      | 26265/65536 [4:30:16<6:43:09,  1.62it/s]
 40%|████      | 26266/65536 [4:30:16<6:34:56,  1.66it/s]
 40%|████      | 26267/65536 [4:30:17<6:24:03,  1.70it/s]
 40%|████      | 26268/65536 [4:30:18<6:50:09,  1.60it/s]
 40%|████      | 26269/65536 [4:30:18<6:41:42,  1.63it/s]
 40%|████      | 26270/65536 [4:30:19<6:38:31,  1.64it/s]
 40%|████      | 26271/65536 [4:30:19<6:42:43,  1.62it/s]
 40%|████      | 26272/65536 [4:30:20<6:36:36,  1.65it/s]
 40%|████      | 26273/65536 [4:30:21<6:38:04,  1.64it/s]
 40%|████      | 26274/65536 [4:30:21<6:43:17,  1.62it/s]
 40%|████      | 26275/65536 [4:30:22<6:58:57,  1.56it/s]
 40%|████      | 26276/65536 [4:30:23<7:10:54,  1.52it/s]
 40%|████      | 26277/65536 [4:30:23<7:18:59,  1.49it/s]
 40%|████      | 26278/65536 [4:30:24<7:05:28,  1.54it/s]
 40%|████      | 26279/65536 [4:30:24<7:00:08,  1.56it/s]
 40%|████      | 26280/65536 [4:30:25<6:54:20,  1.58it/s]
                                                         
{'loss': 1.7973, 'learning_rate': 6.407493571690951e-07, 'epoch': 1622.22}
+
 40%|████      | 26280/65536 [4:30:25<6:54:20,  1.58it/s]
 40%|████      | 26281/65536 [4:30:26<7:01:50,  1.55it/s]
 40%|████      | 26282/65536 [4:30:26<6:53:21,  1.58it/s]
 40%|████      | 26283/65536 [4:30:27<6:43:37,  1.62it/s]
 40%|████      | 26284/65536 [4:30:28<6:43:43,  1.62it/s]
 40%|████      | 26285/65536 [4:30:28<6:33:35,  1.66it/s]
 40%|████      | 26286/65536 [4:30:29<6:30:49,  1.67it/s]
 40%|████      | 26287/65536 [4:30:29<6:32:49,  1.67it/s]
 40%|████      | 26288/65536 [4:30:30<6:34:40,  1.66it/s]
 40%|████      | 26289/65536 [4:30:31<6:29:41,  1.68it/s]
 40%|████      | 26290/65536 [4:30:31<6:37:55,  1.64it/s]
 40%|████      | 26291/65536 [4:30:32<6:54:30,  1.58it/s]
 40%|████      | 26292/65536 [4:30:32<6:42:50,  1.62it/s]
 40%|████      | 26293/65536 [4:30:33<6:47:29,  1.61it/s]
 40%|████      | 26294/65536 [4:30:34<6:46:50,  1.61it/s]
 40%|████      | 26295/65536 [4:30:34<6:51:58,  1.59it/s]
 40%|████      | 26296/65536 [4:30:35<6:38:59,  1.64it/s]
 40%|████      | 26297/65536 [4:30:35<6:37:07,  1.65it/s]
 40%|████      | 26298/65536 [4:30:36<6:45:54,  1.61it/s]
 40%|████      | 26299/65536 [4:30:37<6:41:25,  1.63it/s]
 40%|████      | 26300/65536 [4:30:37<6:49:59,  1.59it/s]
                                                         
{'loss': 1.7961, 'learning_rate': 6.404738582098689e-07, 'epoch': 1623.46}
+
 40%|████      | 26300/65536 [4:30:37<6:49:59,  1.59it/s]
 40%|████      | 26301/65536 [4:30:38<6:42:39,  1.62it/s]
 40%|████      | 26302/65536 [4:30:39<6:42:01,  1.63it/s]
 40%|████      | 26303/65536 [4:30:39<6:44:08,  1.62it/s]
 40%|████      | 26304/65536 [4:30:40<6:42:56,  1.62it/s]
 40%|████      | 26305/65536 [4:30:40<6:32:40,  1.67it/s]
 40%|████      | 26306/65536 [4:30:41<6:26:09,  1.69it/s]
 40%|████      | 26307/65536 [4:30:42<6:25:18,  1.70it/s]
 40%|████      | 26308/65536 [4:30:42<6:21:53,  1.71it/s]
 40%|████      | 26309/65536 [4:30:43<6:40:21,  1.63it/s]
 40%|████      | 26310/65536 [4:30:43<6:38:47,  1.64it/s]
 40%|████      | 26311/65536 [4:30:44<6:33:06,  1.66it/s]
 40%|████      | 26312/65536 [4:30:45<6:31:46,  1.67it/s]
 40%|████      | 26313/65536 [4:30:45<6:31:12,  1.67it/s]
 40%|████      | 26314/65536 [4:30:46<6:37:08,  1.65it/s]
 40%|████      | 26315/65536 [4:30:46<6:36:51,  1.65it/s]
 40%|████      | 26316/65536 [4:30:47<6:35:31,  1.65it/s]
 40%|████      | 26317/65536 [4:30:48<6:31:11,  1.67it/s]
 40%|████      | 26318/65536 [4:30:48<6:33:16,  1.66it/s]
 40%|████      | 26319/65536 [4:30:49<6:32:30,  1.67it/s]
 40%|████      | 26320/65536 [4:30:50<7:00:59,  1.55it/s]
                                                         
{'loss': 1.811, 'learning_rate': 6.401983592506428e-07, 'epoch': 1624.69}
+
 40%|████      | 26320/65536 [4:30:50<7:00:59,  1.55it/s]
 40%|████      | 26321/65536 [4:30:50<6:58:54,  1.56it/s]
 40%|████      | 26322/65536 [4:30:51<7:07:09,  1.53it/s]
 40%|████      | 26323/65536 [4:30:51<6:47:49,  1.60it/s]
 40%|████      | 26324/65536 [4:30:52<6:45:30,  1.61it/s]
 40%|████      | 26325/65536 [4:30:53<6:41:54,  1.63it/s]
 40%|████      | 26326/65536 [4:30:53<6:57:19,  1.57it/s]
 40%|████      | 26327/65536 [4:30:54<6:56:21,  1.57it/s]
 40%|████      | 26328/65536 [4:30:55<6:47:35,  1.60it/s]
 40%|████      | 26329/65536 [4:30:55<6:44:59,  1.61it/s]
 40%|████      | 26330/65536 [4:30:56<6:49:42,  1.59it/s]
 40%|████      | 26331/65536 [4:30:56<6:50:51,  1.59it/s]
 40%|████      | 26332/65536 [4:30:57<6:45:30,  1.61it/s]
 40%|████      | 26333/65536 [4:30:58<6:46:25,  1.61it/s]
 40%|████      | 26334/65536 [4:30:58<6:42:17,  1.62it/s]
 40%|████      | 26335/65536 [4:30:59<6:44:03,  1.62it/s]
 40%|████      | 26336/65536 [4:30:59<6:36:20,  1.65it/s]
 40%|████      | 26337/65536 [4:31:00<6:34:31,  1.66it/s]
 40%|████      | 26338/65536 [4:31:01<6:38:49,  1.64it/s]
 40%|████      | 26339/65536 [4:31:01<6:37:18,  1.64it/s]
 40%|████      | 26340/65536 [4:31:02<6:39:29,  1.64it/s]
                                                         
{'loss': 1.7704, 'learning_rate': 6.399228602914166e-07, 'epoch': 1625.93}
+
 40%|████      | 26340/65536 [4:31:02<6:39:29,  1.64it/s]
 40%|████      | 26341/65536 [4:31:03<6:37:10,  1.64it/s]
 40%|████      | 26342/65536 [4:31:03<6:51:20,  1.59it/s]
 40%|████      | 26343/65536 [4:31:04<6:44:05,  1.62it/s]
 40%|██���█      | 26344/65536 [4:31:04<6:53:24,  1.58it/s]
 40%|████      | 26345/65536 [4:31:05<6:48:05,  1.60it/s]
 40%|████      | 26346/65536 [4:31:06<6:55:44,  1.57it/s]
 40%|████      | 26347/65536 [4:31:06<6:49:54,  1.59it/s]
 40%|████      | 26348/65536 [4:31:07<6:50:20,  1.59it/s]
 40%|████      | 26349/65536 [4:31:08<6:40:02,  1.63it/s]
 40%|████      | 26350/65536 [4:31:08<6:35:11,  1.65it/s]
 40%|████      | 26351/65536 [4:31:09<6:27:24,  1.69it/s]
 40%|████      | 26352/65536 [4:31:09<6:27:06,  1.69it/s]
 40%|████      | 26353/65536 [4:31:10<6:42:03,  1.62it/s]
 40%|████      | 26354/65536 [4:31:11<6:52:35,  1.58it/s]
 40%|████      | 26355/65536 [4:31:11<6:46:52,  1.60it/s]
 40%|████      | 26356/65536 [4:31:12<6:41:55,  1.62it/s]
 40%|████      | 26357/65536 [4:31:12<6:32:33,  1.66it/s]
 40%|████      | 26358/65536 [4:31:13<6:45:58,  1.61it/s]
 40%|████      | 26359/65536 [4:31:14<6:41:14,  1.63it/s]
 40%|████      | 26360/65536 [4:31:14<6:42:46,  1.62it/s]
                                                         
{'loss': 1.7649, 'learning_rate': 6.396473613321905e-07, 'epoch': 1627.16}
+
 40%|████      | 26360/65536 [4:31:14<6:42:46,  1.62it/s]
 40%|████      | 26361/65536 [4:31:15<6:42:03,  1.62it/s]
 40%|████      | 26362/65536 [4:31:16<6:41:13,  1.63it/s]
 40%|████      | 26363/65536 [4:31:16<6:41:36,  1.63it/s]
 40%|████      | 26364/65536 [4:31:17<6:37:11,  1.64it/s]
 40%|████      | 26365/65536 [4:31:17<6:38:43,  1.64it/s]
 40%|████      | 26366/65536 [4:31:18<6:44:37,  1.61it/s]
 40%|████      | 26367/65536 [4:31:19<6:47:00,  1.60it/s]
 40%|████      | 26368/65536 [4:31:19<6:46:28,  1.61it/s]
 40%|████      | 26369/65536 [4:31:20<6:42:47,  1.62it/s]
 40%|████      | 26370/65536 [4:31:20<6:42:09,  1.62it/s]
 40%|████      | 26371/65536 [4:31:21<6:37:18,  1.64it/s]
 40%|████      | 26372/65536 [4:31:22<6:38:03,  1.64it/s]
 40%|████      | 26373/65536 [4:31:22<6:33:45,  1.66it/s]
 40%|████      | 26374/65536 [4:31:23<6:48:41,  1.60it/s]
 40%|████      | 26375/65536 [4:31:24<6:47:59,  1.60it/s]
 40%|████      | 26376/65536 [4:31:24<6:46:57,  1.60it/s]
 40%|████      | 26377/65536 [4:31:25<6:41:57,  1.62it/s]
 40%|████      | 26378/65536 [4:31:25<6:30:15,  1.67it/s]
 40%|████      | 26379/65536 [4:31:26<6:37:14,  1.64it/s]
 40%|████      | 26380/65536 [4:31:27<6:43:20,  1.62it/s]
                                                         
{'loss': 1.7882, 'learning_rate': 6.393718623729643e-07, 'epoch': 1628.4}
+
 40%|████      | 26380/65536 [4:31:27<6:43:20,  1.62it/s]
 40%|████      | 26381/65536 [4:31:27<6:37:54,  1.64it/s]
 40%|████      | 26382/65536 [4:31:28<6:36:35,  1.65it/s]
 40%|████      | 26383/65536 [4:31:28<6:47:46,  1.60it/s]
 40%|████      | 26384/65536 [4:31:29<6:38:53,  1.64it/s]
 40%|████      | 26385/65536 [4:31:30<6:35:56,  1.65it/s]
 40%|████      | 26386/65536 [4:31:30<6:45:29,  1.61it/s]
 40%|████      | 26387/65536 [4:31:31<6:35:12,  1.65it/s]
 40%|████      | 26388/65536 [4:31:31<6:37:58,  1.64it/s]
 40%|████      | 26389/65536 [4:31:32<6:35:38,  1.65it/s]
 40%|████      | 26390/65536 [4:31:33<6:51:11,  1.59it/s]
 40%|████      | 26391/65536 [4:31:33<6:48:54,  1.60it/s]
 40%|████      | 26392/65536 [4:31:34<6:50:44,  1.59it/s]
 40%|████      | 26393/65536 [4:31:35<6:55:54,  1.57it/s]
 40%|████      | 26394/65536 [4:31:35<6:46:56,  1.60it/s]
 40%|████      | 26395/65536 [4:31:36<6:45:05,  1.61it/s]
 40%|████      | 26396/65536 [4:31:36<6:41:12,  1.63it/s]
 40%|████      | 26397/65536 [4:31:37<6:35:17,  1.65it/s]
 40%|████      | 26398/65536 [4:31:38<6:29:37,  1.67it/s]
 40%|████      | 26399/65536 [4:31:38<6:27:27,  1.68it/s]
 40%|████      | 26400/65536 [4:31:39<6:25:16,  1.69it/s]
                                                         
{'loss': 1.7661, 'learning_rate': 6.390963634137382e-07, 'epoch': 1629.63}
+
 40%|████      | 26400/65536 [4:31:39<6:25:16,  1.69it/s]
 40%|████      | 26401/65536 [4:31:39<6:38:05,  1.64it/s]
 40%|████      | 26402/65536 [4:31:40<6:40:09,  1.63it/s]
 40%|████      | 26403/65536 [4:31:41<6:47:11,  1.60it/s]
 40%|████      | 26404/65536 [4:31:41<6:48:55,  1.59it/s]
 40%|████      | 26405/65536 [4:31:42<6:42:13,  1.62it/s]
 40%|████      | 26406/65536 [4:31:43<6:37:26,  1.64it/s]
 40%|████      | 26407/65536 [4:31:43<6:49:53,  1.59it/s]
 40%|████      | 26408/65536 [4:31:44<6:43:14,  1.62it/s]
 40%|████      | 26409/65536 [4:31:44<6:45:03,  1.61it/s]
 40%|████      | 26410/65536 [4:31:45<6:50:09,  1.59it/s]
 40%|████      | 26411/65536 [4:31:46<6:41:55,  1.62it/s]
 40%|████      | 26412/65536 [4:31:46<6:36:02,  1.65it/s]
 40%|████      | 26413/65536 [4:31:47<6:37:32,  1.64it/s]
 40%|████      | 26414/65536 [4:31:47<6:32:27,  1.66it/s]
 40%|████      | 26415/65536 [4:31:48<6:39:01,  1.63it/s]
 40%|████      | 26416/65536 [4:31:49<6:41:05,  1.63it/s]
 40%|████      | 26417/65536 [4:31:49<6:36:28,  1.64it/s]
 40%|████      | 26418/65536 [4:31:50<6:40:25,  1.63it/s]
 40%|████      | 26419/65536 [4:31:51<6:36:13,  1.65it/s]
 40%|████      | 26420/65536 [4:31:51<6:43:46,  1.61it/s]
                                                         
{'loss': 1.7519, 'learning_rate': 6.388208644545121e-07, 'epoch': 1630.86}
+
 40%|████      | 26420/65536 [4:31:51<6:43:46,  1.61it/s]
 40%|████      | 26421/65536 [4:31:52<6:38:38,  1.64it/s]
 40%|████      | 26422/65536 [4:31:52<6:45:57,  1.61it/s]
 40%|████      | 26423/65536 [4:31:53<7:05:08,  1.53it/s]
 40%|████      | 26424/65536 [4:31:54<6:49:26,  1.59it/s]
 40%|████      | 26425/65536 [4:31:54<6:43:44,  1.61it/s]
 40%|████      | 26426/65536 [4:31:55<6:38:04,  1.64it/s]
 40%|████      | 26427/65536 [4:31:56<6:54:39,  1.57it/s]
 40%|████      | 26428/65536 [4:31:56<6:50:26,  1.59it/s]
 40%|████      | 26429/65536 [4:31:57<6:42:03,  1.62it/s]
 40%|████      | 26430/65536 [4:31:57<6:51:09,  1.59it/s]
 40%|████      | 26431/65536 [4:31:58<6:46:35,  1.60it/s]
 40%|████      | 26432/65536 [4:31:59<6:47:56,  1.60it/s]
 40%|████      | 26433/65536 [4:31:59<6:49:16,  1.59it/s]
 40%|████      | 26434/65536 [4:32:00<6:47:52,  1.60it/s]
 40%|████      | 26435/65536 [4:32:01<6:40:30,  1.63it/s]
 40%|████      | 26436/65536 [4:32:01<6:40:27,  1.63it/s]
 40%|████      | 26437/65536 [4:32:02<6:42:36,  1.62it/s]
 40%|████      | 26438/65536 [4:32:02<6:38:28,  1.64it/s]
 40%|████      | 26439/65536 [4:32:03<6:46:30,  1.60it/s]
 40%|████      | 26440/65536 [4:32:04<6:35:34,  1.65it/s]
                                                         
{'loss': 1.7378, 'learning_rate': 6.385453654952859e-07, 'epoch': 1632.1}
+
 40%|████      | 26440/65536 [4:32:04<6:35:34,  1.65it/s]
 40%|████      | 26441/65536 [4:32:04<6:49:45,  1.59it/s]
 40%|████      | 26442/65536 [4:32:05<6:45:34,  1.61it/s]
 40%|████      | 26443/65536 [4:32:05<6:42:19,  1.62it/s]
 40%|████      | 26444/65536 [4:32:06<6:42:15,  1.62it/s]
 40%|████      | 26445/65536 [4:32:07<6:33:00,  1.66it/s]
 40%|████      | 26446/65536 [4:32:07<6:31:59,  1.66it/s]
 40%|████      | 26447/65536 [4:32:08<6:34:43,  1.65it/s]
 40%|████      | 26448/65536 [4:32:09<6:35:56,  1.65it/s]
 40%|████      | 26449/65536 [4:32:09<6:39:42,  1.63it/s]
 40%|████      | 26450/65536 [4:32:10<6:58:35,  1.56it/s]
 40%|████      | 26451/65536 [4:32:10<6:43:59,  1.61it/s]
 40%|████      | 26452/65536 [4:32:11<6:32:45,  1.66it/s]
 40%|████      | 26453/65536 [4:32:12<6:36:12,  1.64it/s]
 40%|████      | 26454/65536 [4:32:12<6:40:30,  1.63it/s]
 40%|████      | 26455/65536 [4:32:13<6:48:45,  1.59it/s]
 40%|████      | 26456/65536 [4:32:13<6:45:37,  1.61it/s]
 40%|████      | 26457/65536 [4:32:14<6:37:56,  1.64it/s]
 40%|████      | 26458/65536 [4:32:15<6:40:28,  1.63it/s]
 40%|████      | 26459/65536 [4:32:15<6:31:44,  1.66it/s]
 40%|████      | 26460/65536 [4:32:16<6:33:00,  1.66it/s]
                                                         
{'loss': 1.7445, 'learning_rate': 6.382698665360597e-07, 'epoch': 1633.33}
+
 40%|████      | 26460/65536 [4:32:16<6:33:00,  1.66it/s]
 40%|████      | 26461/65536 [4:32:16<6:35:47,  1.65it/s]
 40%|████      | 26462/65536 [4:32:17<6:41:09,  1.62it/s]
 40%|████      | 26463/65536 [4:32:18<6:35:26,  1.65it/s]
 40%|████      | 26464/65536 [4:32:18<6:42:26,  1.62it/s]
 40%|████      | 26465/65536 [4:32:19<6:41:01,  1.62it/s]
 40%|████      | 26466/65536 [4:32:20<6:40:04,  1.63it/s]
 40%|████      | 26467/65536 [4:32:20<6:42:27,  1.62it/s]
 40%|████      | 26468/65536 [4:32:21<6:47:10,  1.60it/s]
 40%|████      | 26469/65536 [4:32:22<7:02:24,  1.54it/s]
 40%|████      | 26470/65536 [4:32:22<6:51:09,  1.58it/s]
 40%|████      | 26471/65536 [4:32:23<6:54:42,  1.57it/s]
 40%|████      | 26472/65536 [4:32:23<6:48:41,  1.59it/s]
 40%|████      | 26473/65536 [4:32:24<6:46:57,  1.60it/s]
 40%|████      | 26474/65536 [4:32:25<6:49:01,  1.59it/s]
 40%|████      | 26475/65536 [4:32:25<6:50:05,  1.59it/s]
 40%|████      | 26476/65536 [4:32:26<6:41:21,  1.62it/s]
 40%|████      | 26477/65536 [4:32:27<6:45:59,  1.60it/s]
 40%|████      | 26478/65536 [4:32:27<6:43:41,  1.61it/s]
 40%|████      | 26479/65536 [4:32:28<6:41:32,  1.62it/s]
 40%|████      | 26480/65536 [4:32:28<6:41:39,  1.62it/s]
                                                         
{'loss': 1.7006, 'learning_rate': 6.379943675768335e-07, 'epoch': 1634.57}
+
 40%|████      | 26480/65536 [4:32:28<6:41:39,  1.62it/s]
 40%|████      | 26481/65536 [4:32:29<6:39:23,  1.63it/s]
 40%|████      | 26482/65536 [4:32:30<6:36:07,  1.64it/s]
 40%|████      | 26483/65536 [4:32:30<6:44:17,  1.61it/s]
 40%|████      | 26484/65536 [4:32:31<6:33:48,  1.65it/s]
 40%|████      | 26485/65536 [4:32:31<6:40:24,  1.63it/s]
 40%|████      | 26486/65536 [4:32:32<6:35:35,  1.65it/s]
 40%|████      | 26487/65536 [4:32:33<6:34:59,  1.65it/s]
 40%|████      | 26488/65536 [4:32:33<6:55:12,  1.57it/s]
 40%|████      | 26489/65536 [4:32:34<6:47:11,  1.60it/s]
 40%|████      | 26490/65536 [4:32:35<6:48:52,  1.59it/s]
 40%|████      | 26491/65536 [4:32:35<6:43:31,  1.61it/s]
 40%|████      | 26492/65536 [4:32:36<6:33:47,  1.65it/s]
 40%|████      | 26493/65536 [4:32:36<6:41:36,  1.62it/s]
 40%|████      | 26494/65536 [4:32:37<6:43:48,  1.61it/s]
 40%|████      | 26495/65536 [4:32:38<6:30:39,  1.67it/s]
 40%|████      | 26496/65536 [4:32:38<6:28:01,  1.68it/s]
 40%|████      | 26497/65536 [4:32:39<6:21:49,  1.70it/s]
 40%|████      | 26498/65536 [4:32:39<6:23:10,  1.70it/s]
 40%|████      | 26499/65536 [4:32:40<6:32:53,  1.66it/s]
 40%|████      | 26500/65536 [4:32:41<6:29:15,  1.67it/s]
                                                         
{'loss': 1.8112, 'learning_rate': 6.377188686176074e-07, 'epoch': 1635.8}
+
 40%|████      | 26500/65536 [4:32:41<6:29:15,  1.67it/s]
 40%|████      | 26501/65536 [4:32:41<6:28:07,  1.68it/s]
 40%|████      | 26502/65536 [4:32:42<6:40:09,  1.63it/s]
 40%|████      | 26503/65536 [4:32:42<6:42:22,  1.62it/s]
 40%|████      | 26504/65536 [4:32:43<6:47:01,  1.60it/s]
 40%|████      | 26505/65536 [4:32:44<6:31:44,  1.66it/s]
 40%|████      | 26506/65536 [4:32:44<6:27:20,  1.68it/s]
 40%|████      | 26507/65536 [4:32:45<6:30:37,  1.67it/s]
 40%|████      | 26508/65536 [4:32:45<6:35:03,  1.65it/s]
 40%|████      | 26509/65536 [4:32:46<6:35:08,  1.65it/s]
 40%|████      | 26510/65536 [4:32:47<6:34:36,  1.65it/s]
 40%|████      | 26511/65536 [4:32:47<6:37:24,  1.64it/s]
 40%|████      | 26512/65536 [4:32:48<6:41:57,  1.62it/s]
 40%|████      | 26513/65536 [4:32:48<6:42:35,  1.62it/s]
 40%|████      | 26514/65536 [4:32:49<6:47:15,  1.60it/s]
 40%|████      | 26515/65536 [4:32:50<6:44:48,  1.61it/s]
 40%|████      | 26516/65536 [4:32:50<6:48:42,  1.59it/s]
 40%|████      | 26517/65536 [4:32:51<6:47:21,  1.60it/s]
 40%|████      | 26518/65536 [4:32:52<6:46:07,  1.60it/s]
 40%|████      | 26519/65536 [4:32:52<6:39:07,  1.63it/s]
 40%|████      | 26520/65536 [4:32:53<6:41:22,  1.62it/s]
                                                         
{'loss': 1.7643, 'learning_rate': 6.374433696583811e-07, 'epoch': 1637.04}
+
 40%|████      | 26520/65536 [4:32:53<6:41:22,  1.62it/s]
 40%|████      | 26521/65536 [4:32:53<6:34:58,  1.65it/s]
 40%|████      | 26522/65536 [4:32:54<6:40:50,  1.62it/s]
 40%|████      | 26523/65536 [4:32:55<6:41:40,  1.62it/s]
 40%|████      | 26524/65536 [4:32:55<6:51:52,  1.58it/s]
 40%|████      | 26525/65536 [4:32:56<6:54:46,  1.57it/s]
 40%|████      | 26526/65536 [4:32:57<6:45:38,  1.60it/s]
 40%|████      | 26527/65536 [4:32:57<6:46:36,  1.60it/s]
 40%|████      | 26528/65536 [4:32:58<6:45:57,  1.60it/s]
 40%|████      | 26529/65536 [4:32:58<6:36:47,  1.64it/s]
 40%|████      | 26530/65536 [4:32:59<6:26:50,  1.68it/s]
 40%|████      | 26531/65536 [4:33:00<6:27:11,  1.68it/s]
 40%|████      | 26532/65536 [4:33:00<6:24:57,  1.69it/s]
 40%|████      | 26533/65536 [4:33:01<6:32:09,  1.66it/s]
 40%|████      | 26534/65536 [4:33:01<6:31:35,  1.66it/s]
 40%|████      | 26535/65536 [4:33:02<6:34:23,  1.65it/s]
 40%|████      | 26536/65536 [4:33:03<6:51:25,  1.58it/s]
 40%|████      | 26537/65536 [4:33:03<6:40:03,  1.62it/s]
 40%|████      | 26538/65536 [4:33:04<6:36:28,  1.64it/s]
 40%|████      | 26539/65536 [4:33:05<6:38:48,  1.63it/s]
 40%|████      | 26540/65536 [4:33:05<6:46:53,  1.60it/s]
                                                         
{'loss': 1.7564, 'learning_rate': 6.37167870699155e-07, 'epoch': 1638.27}
+
 40%|████      | 26540/65536 [4:33:05<6:46:53,  1.60it/s]
 40%|████      | 26541/65536 [4:33:06<6:57:55,  1.56it/s]
 40%|████      | 26542/65536 [4:33:06<6:48:39,  1.59it/s]
 41%|████      | 26543/65536 [4:33:07<6:59:07,  1.55it/s]
 41%|████      | 26544/65536 [4:33:08<6:49:21,  1.59it/s]
 41%|████      | 26545/65536 [4:33:08<6:47:11,  1.60it/s]
 41%|████      | 26546/65536 [4:33:09<6:41:12,  1.62it/s]
 41%|████      | 26547/65536 [4:33:10<6:38:00,  1.63it/s]
 41%|████      | 26548/65536 [4:33:10<6:36:16,  1.64it/s]
 41%|████      | 26549/65536 [4:33:11<6:36:32,  1.64it/s]
 41%|████      | 26550/65536 [4:33:11<6:33:21,  1.65it/s]
 41%|████      | 26551/65536 [4:33:12<6:26:02,  1.68it/s]
 41%|████      | 26552/65536 [4:33:13<6:44:22,  1.61it/s]
 41%|████      | 26553/65536 [4:33:13<6:39:13,  1.63it/s]
 41%|████      | 26554/65536 [4:33:14<6:34:18,  1.65it/s]
 41%|████      | 26555/65536 [4:33:14<6:33:24,  1.65it/s]
 41%|████      | 26556/65536 [4:33:15<6:41:25,  1.62it/s]
 41%|████      | 26557/65536 [4:33:16<6:36:27,  1.64it/s]
 41%|████      | 26558/65536 [4:33:16<6:28:43,  1.67it/s]
 41%|████      | 26559/65536 [4:33:17<6:25:56,  1.68it/s]
 41%|████      | 26560/65536 [4:33:17<6:35:00,  1.64it/s]
                                                         
{'loss': 1.7366, 'learning_rate': 6.368923717399289e-07, 'epoch': 1639.51}
+
 41%|████      | 26560/65536 [4:33:17<6:35:00,  1.64it/s]
 41%|████      | 26561/65536 [4:33:18<6:44:43,  1.60it/s]
 41%|████      | 26562/65536 [4:33:19<6:35:07,  1.64it/s]
 41%|████      | 26563/65536 [4:33:19<6:51:24,  1.58it/s]
 41%|████      | 26564/65536 [4:33:20<6:47:44,  1.59it/s]
 41%|████      | 26565/65536 [4:33:21<6:37:51,  1.63it/s]
 41%|████      | 26566/65536 [4:33:21<6:47:41,  1.59it/s]
 41%|████      | 26567/65536 [4:33:22<6:40:03,  1.62it/s]
 41%|████      | 26568/65536 [4:33:22<6:35:09,  1.64it/s]
 41%|████      | 26569/65536 [4:33:23<6:51:24,  1.58it/s]
 41%|████      | 26570/65536 [4:33:24<6:48:29,  1.59it/s]
 41%|████      | 26571/65536 [4:33:24<6:46:54,  1.60it/s]
 41%|████      | 26572/65536 [4:33:25<6:48:23,  1.59it/s]
 41%|████      | 26573/65536 [4:33:26<6:39:18,  1.63it/s]
 41%|████      | 26574/65536 [4:33:26<6:29:13,  1.67it/s]
 41%|████      | 26575/65536 [4:33:27<6:44:20,  1.61it/s]
 41%|████      | 26576/65536 [4:33:27<6:40:45,  1.62it/s]
 41%|████      | 26577/65536 [4:33:28<6:44:17,  1.61it/s]
 41%|████      | 26578/65536 [4:33:29<6:37:39,  1.63it/s]
 41%|████      | 26579/65536 [4:33:29<6:42:42,  1.61it/s]
 41%|████      | 26580/65536 [4:33:30<6:42:18,  1.61it/s]
                                                         
{'loss': 1.7218, 'learning_rate': 6.366168727807027e-07, 'epoch': 1640.74}
+
 41%|████      | 26580/65536 [4:33:30<6:42:18,  1.61it/s]
 41%|████      | 26581/65536 [4:33:30<6:40:08,  1.62it/s]
 41%|████      | 26582/65536 [4:33:31<6:33:51,  1.65it/s]
 41%|████      | 26583/65536 [4:33:32<6:31:11,  1.66it/s]
 41%|████      | 26584/65536 [4:33:32<6:25:04,  1.69it/s]
 41%|████      | 26585/65536 [4:33:33<6:42:16,  1.61it/s]
 41%|████      | 26586/65536 [4:33:33<6:30:59,  1.66it/s]
 41%|████      | 26587/65536 [4:33:34<6:28:03,  1.67it/s]
 41%|████      | 26588/65536 [4:33:35<6:26:15,  1.68it/s]
 41%|████      | 26589/65536 [4:33:35<6:25:50,  1.68it/s]
 41%|████      | 26590/65536 [4:33:36<6:34:21,  1.65it/s]
 41%|████      | 26591/65536 [4:33:37<6:50:03,  1.58it/s]
 41%|████      | 26592/65536 [4:33:37<6:53:51,  1.57it/s]
 41%|████      | 26593/65536 [4:33:38<6:52:32,  1.57it/s]
 41%|████      | 26594/65536 [4:33:38<6:53:56,  1.57it/s]
 41%|████      | 26595/65536 [4:33:39<6:58:23,  1.55it/s]
 41%|████      | 26596/65536 [4:33:40<6:53:50,  1.57it/s]
 41%|████      | 26597/65536 [4:33:40<6:45:07,  1.60it/s]
 41%|████      | 26598/65536 [4:33:41<6:38:58,  1.63it/s]
 41%|████      | 26599/65536 [4:33:42<6:29:54,  1.66it/s]
 41%|████      | 26600/65536 [4:33:42<6:26:13,  1.68it/s]
                                                         
{'loss': 1.7528, 'learning_rate': 6.363413738214766e-07, 'epoch': 1641.98}
+
 41%|████      | 26600/65536 [4:33:42<6:26:13,  1.68it/s]
 41%|████      | 26601/65536 [4:33:43<6:36:40,  1.64it/s]
 41%|████      | 26602/65536 [4:33:43<6:30:08,  1.66it/s]
 41%|████      | 26603/65536 [4:33:44<6:30:56,  1.66it/s]
 41%|████      | 26604/65536 [4:33:45<6:36:51,  1.63it/s]
 41%|████      | 26605/65536 [4:33:45<6:28:53,  1.67it/s]
 41%|████      | 26606/65536 [4:33:46<6:45:26,  1.60it/s]
 41%|████      | 26607/65536 [4:33:46<6:36:02,  1.64it/s]
 41%|████      | 26608/65536 [4:33:47<6:30:56,  1.66it/s]
 41%|████      | 26609/65536 [4:33:48<6:32:55,  1.65it/s]
 41%|████      | 26610/65536 [4:33:48<6:38:23,  1.63it/s]
 41%|████      | 26611/65536 [4:33:49<6:23:29,  1.69it/s]
 41%|████      | 26612/65536 [4:33:49<6:20:45,  1.70it/s]
 41%|████      | 26613/65536 [4:33:50<6:33:40,  1.65it/s]
 41%|████      | 26614/65536 [4:33:51<6:32:41,  1.65it/s]
 41%|████      | 26615/65536 [4:33:51<6:33:03,  1.65it/s]
 41%|████      | 26616/65536 [4:33:52<6:39:16,  1.62it/s]
 41%|████      | 26617/65536 [4:33:53<6:55:49,  1.56it/s]
 41%|████      | 26618/65536 [4:33:53<6:49:08,  1.59it/s]
 41%|████      | 26619/65536 [4:33:54<6:38:59,  1.63it/s]
 41%|████      | 26620/65536 [4:33:54<6:43:00,  1.61it/s]
                                                         
{'loss': 1.7528, 'learning_rate': 6.360658748622505e-07, 'epoch': 1643.21}
+
 41%|████      | 26620/65536 [4:33:54<6:43:00,  1.61it/s]
 41%|████      | 26621/65536 [4:33:55<6:39:32,  1.62it/s]
 41%|████      | 26622/65536 [4:33:56<6:34:47,  1.64it/s]
 41%|████      | 26623/65536 [4:33:56<6:32:12,  1.65it/s]
 41%|████      | 26624/65536 [4:33:57<6:39:36,  1.62it/s]
 41%|████      | 26625/65536 [4:33:57<6:30:28,  1.66it/s]
 41%|████      | 26626/65536 [4:33:58<6:43:21,  1.61it/s]
 41%|████      | 26627/65536 [4:33:59<6:41:36,  1.61it/s]
 41%|████      | 26628/65536 [4:33:59<6:37:50,  1.63it/s]
 41%|████      | 26629/65536 [4:34:00<6:41:46,  1.61it/s]
 41%|████      | 26630/65536 [4:34:01<6:43:58,  1.61it/s]
 41%|████      | 26631/65536 [4:34:01<6:42:47,  1.61it/s]
 41%|████      | 26632/65536 [4:34:02<6:31:05,  1.66it/s]
 41%|████      | 26633/65536 [4:34:02<6:41:46,  1.61it/s]
 41%|████      | 26634/65536 [4:34:03<6:57:45,  1.55it/s]
 41%|████      | 26635/65536 [4:34:04<6:48:34,  1.59it/s]
 41%|████      | 26636/65536 [4:34:04<6:47:58,  1.59it/s]
 41%|████      | 26637/65536 [4:34:05<6:36:35,  1.63it/s]
 41%|████      | 26638/65536 [4:34:05<6:27:18,  1.67it/s]
 41%|████      | 26639/65536 [4:34:06<6:27:30,  1.67it/s]
 41%|████      | 26640/65536 [4:34:07<6:28:14,  1.67it/s]
                                                         
{'loss': 1.7733, 'learning_rate': 6.357903759030244e-07, 'epoch': 1644.44}
+
 41%|████      | 26640/65536 [4:34:07<6:28:14,  1.67it/s]
 41%|████      | 26641/65536 [4:34:07<6:42:42,  1.61it/s]
 41%|████      | 26642/65536 [4:34:08<6:41:19,  1.62it/s]
 41%|████      | 26643/65536 [4:34:08<6:38:40,  1.63it/s]
 41%|████      | 26644/65536 [4:34:09<6:35:19,  1.64it/s]
 41%|████      | 26645/65536 [4:34:10<6:30:36,  1.66it/s]
 41%|████      | 26646/65536 [4:34:10<6:37:37,  1.63it/s]
 41%|████      | 26647/65536 [4:34:11<6:28:23,  1.67it/s]
 41%|████      | 26648/65536 [4:34:11<6:28:20,  1.67it/s]
 41%|████      | 26649/65536 [4:34:12<6:28:49,  1.67it/s]
 41%|████      | 26650/65536 [4:34:13<6:32:47,  1.65it/s]
 41%|████      | 26651/65536 [4:34:13<6:21:06,  1.70it/s]
 41%|████      | 26652/65536 [4:34:14<6:34:41,  1.64it/s]
 41%|████      | 26653/65536 [4:34:15<6:36:28,  1.63it/s]
 41%|████      | 26654/65536 [4:34:15<6:32:59,  1.65it/s]
 41%|████      | 26655/65536 [4:34:16<6:38:56,  1.62it/s]
 41%|████      | 26656/65536 [4:34:16<6:32:55,  1.65it/s]
 41%|████      | 26657/65536 [4:34:17<6:40:30,  1.62it/s]
 41%|████      | 26658/65536 [4:34:18<6:31:42,  1.65it/s]
 41%|████      | 26659/65536 [4:34:18<6:35:55,  1.64it/s]
 41%|████      | 26660/65536 [4:34:19<6:30:15,  1.66it/s]
                                                         
{'loss': 1.7352, 'learning_rate': 6.355148769437983e-07, 'epoch': 1645.68}
+
 41%|████      | 26660/65536 [4:34:19<6:30:15,  1.66it/s]
 41%|████      | 26661/65536 [4:34:19<6:30:10,  1.66it/s]
 41%|████      | 26662/65536 [4:34:20<6:28:03,  1.67it/s]
 41%|████      | 26663/65536 [4:34:21<6:35:12,  1.64it/s]
 41%|████      | 26664/65536 [4:34:21<6:48:48,  1.58it/s]
 41%|████      | 26665/65536 [4:34:22<6:41:52,  1.61it/s]
 41%|████      | 26666/65536 [4:34:23<6:55:09,  1.56it/s]
 41%|████      | 26667/65536 [4:34:23<6:56:19,  1.56it/s]
 41%|████      | 26668/65536 [4:34:24<6:42:15,  1.61it/s]
 41%|████      | 26669/65536 [4:34:24<6:48:44,  1.58it/s]
 41%|████      | 26670/65536 [4:34:25<6:42:20,  1.61it/s]
 41%|████      | 26671/65536 [4:34:26<6:41:12,  1.61it/s]
 41%|████      | 26672/65536 [4:34:26<6:30:49,  1.66it/s]
 41%|████      | 26673/65536 [4:34:27<6:37:53,  1.63it/s]
 41%|████      | 26674/65536 [4:34:27<6:37:13,  1.63it/s]
 41%|████      | 26675/65536 [4:34:28<6:28:49,  1.67it/s]
 41%|████      | 26676/65536 [4:34:29<6:32:47,  1.65it/s]
 41%|████      | 26677/65536 [4:34:29<6:38:02,  1.63it/s]
 41%|████      | 26678/65536 [4:34:30<6:42:15,  1.61it/s]
 41%|████      | 26679/65536 [4:34:31<6:36:17,  1.63it/s]
 41%|████      | 26680/65536 [4:34:31<6:38:34,  1.62it/s]
                                                         
{'loss': 1.7101, 'learning_rate': 6.352393779845721e-07, 'epoch': 1646.91}
+
 41%|████      | 26680/65536 [4:34:31<6:38:34,  1.62it/s]
 41%|████      | 26681/65536 [4:34:32<6:39:49,  1.62it/s]
 41%|████      | 26682/65536 [4:34:32<6:53:29,  1.57it/s]
 41%|████      | 26683/65536 [4:34:33<6:40:24,  1.62it/s]
 41%|████      | 26684/65536 [4:34:34<6:43:45,  1.60it/s]
 41%|████      | 26685/65536 [4:34:34<6:37:14,  1.63it/s]
 41%|████      | 26686/65536 [4:34:35<6:30:27,  1.66it/s]
 41%|████      | 26687/65536 [4:34:35<6:36:29,  1.63it/s]
 41%|████      | 26688/65536 [4:34:36<6:35:57,  1.64it/s]
 41%|████      | 26689/65536 [4:34:37<6:23:36,  1.69it/s]
 41%|████      | 26690/65536 [4:34:37<6:17:31,  1.71it/s]
 41%|████      | 26691/65536 [4:34:38<6:32:20,  1.65it/s]
 41%|████      | 26692/65536 [4:34:38<6:31:46,  1.65it/s]
 41%|████      | 26693/65536 [4:34:39<6:38:27,  1.62it/s]
 41%|████      | 26694/65536 [4:34:40<6:34:28,  1.64it/s]
 41%|████      | 26695/65536 [4:34:40<6:29:21,  1.66it/s]
 41%|████      | 26696/65536 [4:34:41<6:32:54,  1.65it/s]
 41%|████      | 26697/65536 [4:34:42<6:39:00,  1.62it/s]
 41%|████      | 26698/65536 [4:34:42<6:52:52,  1.57it/s]
 41%|████      | 26699/65536 [4:34:43<6:47:27,  1.59it/s]
 41%|████      | 26700/65536 [4:34:43<6:43:15,  1.61it/s]
                                                         
{'loss': 1.7967, 'learning_rate': 6.34963879025346e-07, 'epoch': 1648.15}
+
 41%|████      | 26700/65536 [4:34:43<6:43:15,  1.61it/s]
 41%|████      | 26701/65536 [4:34:44<6:46:18,  1.59it/s]
 41%|████      | 26702/65536 [4:34:45<6:39:09,  1.62it/s]
 41%|████      | 26703/65536 [4:34:45<6:34:59,  1.64it/s]
 41%|████      | 26704/65536 [4:34:46<6:33:42,  1.64it/s]
 41%|████      | 26705/65536 [4:34:46<6:39:25,  1.62it/s]
 41%|████      | 26706/65536 [4:34:47<6:46:58,  1.59it/s]
 41%|████      | 26707/65536 [4:34:48<6:45:55,  1.59it/s]
 41%|████      | 26708/65536 [4:34:48<6:41:49,  1.61it/s]
 41%|████      | 26709/65536 [4:34:49<6:42:23,  1.61it/s]
 41%|████      | 26710/65536 [4:34:50<6:36:18,  1.63it/s]
 41%|████      | 26711/65536 [4:34:50<6:33:19,  1.65it/s]
 41%|████      | 26712/65536 [4:34:51<6:36:41,  1.63it/s]
 41%|████      | 26713/65536 [4:34:51<6:28:17,  1.67it/s]
 41%|████      | 26714/65536 [4:34:52<6:38:01,  1.63it/s]
 41%|████      | 26715/65536 [4:34:53<6:37:41,  1.63it/s]
 41%|████      | 26716/65536 [4:34:53<6:38:58,  1.62it/s]
 41%|████      | 26717/65536 [4:34:54<6:40:31,  1.62it/s]
 41%|████      | 26718/65536 [4:34:55<6:51:23,  1.57it/s]
 41%|████      | 26719/65536 [4:34:55<6:42:13,  1.61it/s]
 41%|████      | 26720/65536 [4:34:56<6:45:44,  1.59it/s]
                                                         
{'loss': 1.7317, 'learning_rate': 6.346883800661197e-07, 'epoch': 1649.38}
+
 41%|████      | 26720/65536 [4:34:56<6:45:44,  1.59it/s]
 41%|████      | 26721/65536 [4:34:56<6:37:54,  1.63it/s]
 41%|████      | 26722/65536 [4:34:57<6:40:44,  1.61it/s]
 41%|████      | 26723/65536 [4:34:58<6:47:09,  1.59it/s]
 41%|████      | 26724/65536 [4:34:58<6:47:17,  1.59it/s]
 41%|████      | 26725/65536 [4:34:59<6:41:44,  1.61it/s]
 41%|████      | 26726/65536 [4:34:59<6:32:40,  1.65it/s]
 41%|████      | 26727/65536 [4:35:00<6:31:40,  1.65it/s]
 41%|████      | 26728/65536 [4:35:01<6:20:33,  1.70it/s]
 41%|████      | 26729/65536 [4:35:01<6:19:45,  1.70it/s]
 41%|████      | 26730/65536 [4:35:02<6:17:11,  1.71it/s]
 41%|████      | 26731/65536 [4:35:03<6:53:26,  1.56it/s]
 41%|████      | 26732/65536 [4:35:03<6:48:02,  1.58it/s]
 41%|████      | 26733/65536 [4:35:04<6:41:04,  1.61it/s]
 41%|████      | 26734/65536 [4:35:04<6:39:16,  1.62it/s]
 41%|████      | 26735/65536 [4:35:05<6:43:08,  1.60it/s]
 41%|████      | 26736/65536 [4:35:06<6:34:35,  1.64it/s]
 41%|████      | 26737/65536 [4:35:06<6:36:11,  1.63it/s]
 41%|████      | 26738/65536 [4:35:07<6:38:08,  1.62it/s]
 41%|████      | 26739/65536 [4:35:07<6:36:42,  1.63it/s]
 41%|████      | 26740/65536 [4:35:08<6:30:55,  1.65it/s]
                                                         
{'loss': 1.766, 'learning_rate': 6.344128811068936e-07, 'epoch': 1650.62}
+
 41%|████      | 26740/65536 [4:35:08<6:30:55,  1.65it/s]
 41%|████      | 26741/65536 [4:35:09<6:27:03,  1.67it/s]
 41%|████      | 26742/65536 [4:35:09<6:18:32,  1.71it/s]
 41%|████      | 26743/65536 [4:35:10<6:17:05,  1.71it/s]
 41%|████      | 26744/65536 [4:35:10<6:16:50,  1.72it/s]
 41%|████      | 26745/65536 [4:35:11<6:21:44,  1.69it/s]
 41%|████      | 26746/65536 [4:35:12<6:29:55,  1.66it/s]
 41%|████      | 26747/65536 [4:35:12<6:49:44,  1.58it/s]
 41%|████      | 26748/65536 [4:35:13<7:02:05,  1.53it/s]
 41%|████      | 26749/65536 [4:35:14<6:48:06,  1.58it/s]
 41%|████      | 26750/65536 [4:35:14<6:38:10,  1.62it/s]
 41%|████      | 26751/65536 [4:35:15<6:32:05,  1.65it/s]
 41%|████      | 26752/65536 [4:35:15<6:22:06,  1.69it/s]
 41%|████      | 26753/65536 [4:35:16<6:33:57,  1.64it/s]
 41%|████      | 26754/65536 [4:35:17<6:36:14,  1.63it/s]
 41%|████      | 26755/65536 [4:35:17<6:34:57,  1.64it/s]
 41%|████      | 26756/65536 [4:35:18<6:32:01,  1.65it/s]
 41%|████      | 26757/65536 [4:35:18<6:26:11,  1.67it/s]
 41%|████      | 26758/65536 [4:35:19<6:33:50,  1.64it/s]
 41%|████      | 26759/65536 [4:35:20<6:31:50,  1.65it/s]
 41%|████      | 26760/65536 [4:35:20<6:35:07,  1.64it/s]
                                                         
{'loss': 1.7895, 'learning_rate': 6.341373821476675e-07, 'epoch': 1651.85}
+
 41%|████      | 26760/65536 [4:35:20<6:35:07,  1.64it/s]
 41%|████      | 26761/65536 [4:35:21<6:35:53,  1.63it/s]
 41%|████      | 26762/65536 [4:35:21<6:34:42,  1.64it/s]
 41%|████      | 26763/65536 [4:35:22<7:05:06,  1.52it/s]
 41%|████      | 26764/65536 [4:35:23<7:03:38,  1.53it/s]
 41%|████      | 26765/65536 [4:35:23<6:53:57,  1.56it/s]
 41%|████      | 26766/65536 [4:35:24<6:41:20,  1.61it/s]
 41%|████      | 26767/65536 [4:35:25<6:34:25,  1.64it/s]
 41%|████      | 26768/65536 [4:35:25<6:28:58,  1.66it/s]
 41%|████      | 26769/65536 [4:35:26<6:32:20,  1.65it/s]
 41%|████      | 26770/65536 [4:35:26<6:37:08,  1.63it/s]
 41%|████      | 26771/65536 [4:35:27<6:26:16,  1.67it/s]
 41%|████      | 26772/65536 [4:35:28<6:22:35,  1.69it/s]
 41%|████      | 26773/65536 [4:35:28<6:22:08,  1.69it/s]
 41%|████      | 26774/65536 [4:35:29<6:31:42,  1.65it/s]
 41%|████      | 26775/65536 [4:35:29<6:34:59,  1.64it/s]
 41%|████      | 26776/65536 [4:35:30<6:42:17,  1.61it/s]
 41%|████      | 26777/65536 [4:35:31<6:35:24,  1.63it/s]
 41%|████      | 26778/65536 [4:35:31<6:27:42,  1.67it/s]
 41%|████      | 26779/65536 [4:35:32<6:41:15,  1.61it/s]
 41%|████      | 26780/65536 [4:35:33<6:44:53,  1.60it/s]
                                                         
{'loss': 1.7515, 'learning_rate': 6.338618831884413e-07, 'epoch': 1653.09}
+
 41%|████      | 26780/65536 [4:35:33<6:44:53,  1.60it/s]
 41%|████      | 26781/65536 [4:35:33<6:39:31,  1.62it/s]
 41%|████      | 26782/65536 [4:35:34<6:43:45,  1.60it/s]
 41%|████      | 26783/65536 [4:35:34<6:40:00,  1.61it/s]
 41%|████      | 26784/65536 [4:35:35<6:27:40,  1.67it/s]
 41%|████      | 26785/65536 [4:35:36<6:25:46,  1.67it/s]
 41%|████      | 26786/65536 [4:35:36<6:28:34,  1.66it/s]
 41%|████      | 26787/65536 [4:35:37<6:21:21,  1.69it/s]
 41%|████      | 26788/65536 [4:35:37<6:28:31,  1.66it/s]
 41%|████      | 26789/65536 [4:35:38<6:21:35,  1.69it/s]
 41%|████      | 26790/65536 [4:35:39<6:31:25,  1.65it/s]
 41%|████      | 26791/65536 [4:35:39<6:45:04,  1.59it/s]
 41%|████      | 26792/65536 [4:35:40<6:37:00,  1.63it/s]
 41%|████      | 26793/65536 [4:35:40<6:41:13,  1.61it/s]
 41%|████      | 26794/65536 [4:35:41<6:40:01,  1.61it/s]
 41%|████      | 26795/65536 [4:35:42<6:44:47,  1.60it/s]
 41%|████      | 26796/65536 [4:35:42<6:44:18,  1.60it/s]
 41%|████      | 26797/65536 [4:35:43<6:41:45,  1.61it/s]
 41%|████      | 26798/65536 [4:35:44<6:49:46,  1.58it/s]
 41%|████      | 26799/65536 [4:35:44<6:38:59,  1.62it/s]
 41%|████      | 26800/65536 [4:35:45<6:41:12,  1.61it/s]
                                                         
{'loss': 1.7262, 'learning_rate': 6.335863842292151e-07, 'epoch': 1654.32}
+
 41%|████      | 26800/65536 [4:35:45<6:41:12,  1.61it/s]
 41%|████      | 26801/65536 [4:35:45<6:38:53,  1.62it/s]
 41%|████      | 26802/65536 [4:35:46<6:37:13,  1.63it/s]
 41%|████      | 26803/65536 [4:35:47<6:34:22,  1.64it/s]
 41%|████      | 26804/65536 [4:35:47<6:29:08,  1.66it/s]
 41%|████      | 26805/65536 [4:35:48<6:36:50,  1.63it/s]
 41%|████      | 26806/65536 [4:35:49<6:50:32,  1.57it/s]
 41%|████      | 26807/65536 [4:35:49<6:37:43,  1.62it/s]
 41%|████      | 26808/65536 [4:35:50<6:27:39,  1.67it/s]
 41%|████      | 26809/65536 [4:35:50<6:28:02,  1.66it/s]
 41%|████      | 26810/65536 [4:35:51<6:21:50,  1.69it/s]
 41%|████      | 26811/65536 [4:35:51<6:23:54,  1.68it/s]
 41%|████      | 26812/65536 [4:35:52<6:38:54,  1.62it/s]
 41%|████      | 26813/65536 [4:35:53<6:37:03,  1.63it/s]
 41%|████      | 26814/65536 [4:35:53<6:42:28,  1.60it/s]
 41%|████      | 26815/65536 [4:35:54<6:30:22,  1.65it/s]
 41%|████      | 26816/65536 [4:35:55<6:33:12,  1.64it/s]
 41%|████      | 26817/65536 [4:35:55<6:34:20,  1.64it/s]
 41%|████      | 26818/65536 [4:35:56<6:29:31,  1.66it/s]
 41%|████      | 26819/65536 [4:35:56<6:32:06,  1.65it/s]
 41%|████      | 26820/65536 [4:35:57<6:27:30,  1.67it/s]
                                                         
{'loss': 1.7339, 'learning_rate': 6.333108852699889e-07, 'epoch': 1655.56}
+
 41%|████      | 26820/65536 [4:35:57<6:27:30,  1.67it/s]
 41%|████      | 26821/65536 [4:35:58<6:21:32,  1.69it/s]
 41%|████      | 26822/65536 [4:35:58<6:23:41,  1.68it/s]
 41%|████      | 26823/65536 [4:35:59<6:23:13,  1.68it/s]
 41%|████      | 26824/65536 [4:35:59<6:25:35,  1.67it/s]
 41%|████      | 26825/65536 [4:36:00<6:35:45,  1.63it/s]
 41%|████      | 26826/65536 [4:36:01<6:53:18,  1.56it/s]
 41%|████      | 26827/65536 [4:36:01<6:42:38,  1.60it/s]
 41%|████      | 26828/65536 [4:36:02<6:56:21,  1.55it/s]
 41%|████      | 26829/65536 [4:36:03<6:58:00,  1.54it/s]
 41%|████      | 26830/65536 [4:36:03<6:42:59,  1.60it/s]
 41%|████      | 26831/65536 [4:36:04<6:42:35,  1.60it/s]
 41%|████      | 26832/65536 [4:36:04<6:40:38,  1.61it/s]
 41%|████      | 26833/65536 [4:36:05<6:41:51,  1.61it/s]
 41%|████      | 26834/65536 [4:36:06<6:50:25,  1.57it/s]
 41%|████      | 26835/65536 [4:36:06<6:43:19,  1.60it/s]
 41%|████      | 26836/65536 [4:36:07<6:40:57,  1.61it/s]
 41%|████      | 26837/65536 [4:36:07<6:33:54,  1.64it/s]
 41%|████      | 26838/65536 [4:36:08<6:33:15,  1.64it/s]
 41%|████      | 26839/65536 [4:36:09<6:28:26,  1.66it/s]
 41%|████      | 26840/65536 [4:36:09<6:30:04,  1.65it/s]
                                                         
{'loss': 1.721, 'learning_rate': 6.330353863107628e-07, 'epoch': 1656.79}
+
 41%|████      | 26840/65536 [4:36:09<6:30:04,  1.65it/s]
 41%|████      | 26841/65536 [4:36:10<6:26:51,  1.67it/s]
 41%|████      | 26842/65536 [4:36:11<6:29:20,  1.66it/s]
 41%|████      | 26843/65536 [4:36:11<6:33:57,  1.64it/s]
 41%|████      | 26844/65536 [4:36:12<6:45:13,  1.59it/s]
 41%|████      | 26845/65536 [4:36:12<6:44:44,  1.59it/s]
 41%|████      | 26846/65536 [4:36:13<6:33:56,  1.64it/s]
 41%|████      | 26847/65536 [4:36:14<6:38:48,  1.62it/s]
 41%|████      | 26848/65536 [4:36:14<6:47:34,  1.58it/s]
 41%|████      | 26849/65536 [4:36:15<6:43:14,  1.60it/s]
 41%|████      | 26850/65536 [4:36:16<6:38:46,  1.62it/s]
 41%|████      | 26851/65536 [4:36:16<6:31:48,  1.65it/s]
 41%|████      | 26852/65536 [4:36:17<6:28:28,  1.66it/s]
 41%|████      | 26853/65536 [4:36:17<6:30:22,  1.65it/s]
 41%|████      | 26854/65536 [4:36:18<6:29:08,  1.66it/s]
 41%|████      | 26855/65536 [4:36:18<6:29:06,  1.66it/s]
 41%|████      | 26856/65536 [4:36:19<6:40:01,  1.61it/s]
 41%|████      | 26857/65536 [4:36:20<6:43:18,  1.60it/s]
 41%|████      | 26858/65536 [4:36:20<6:41:36,  1.61it/s]
 41%|████      | 26859/65536 [4:36:21<6:29:49,  1.65it/s]
 41%|████      | 26860/65536 [4:36:22<6:45:25,  1.59it/s]
                                                         
{'loss': 1.757, 'learning_rate': 6.327598873515366e-07, 'epoch': 1658.02}
+
 41%|████      | 26860/65536 [4:36:22<6:45:25,  1.59it/s]
 41%|████      | 26861/65536 [4:36:22<6:37:56,  1.62it/s]
 41%|████      | 26862/65536 [4:36:23<6:41:13,  1.61it/s]
 41%|████      | 26863/65536 [4:36:23<6:34:09,  1.64it/s]
 41%|████      | 26864/65536 [4:36:24<6:32:52,  1.64it/s]
 41%|████      | 26865/65536 [4:36:25<6:27:27,  1.66it/s]
 41%|████      | 26866/65536 [4:36:25<6:20:02,  1.70it/s]
 41%|████      | 26867/65536 [4:36:26<6:32:18,  1.64it/s]
 41%|████      | 26868/65536 [4:36:26<6:23:39,  1.68it/s]
 41%|████      | 26869/65536 [4:36:27<6:26:07,  1.67it/s]
 41%|████      | 26870/65536 [4:36:28<6:24:04,  1.68it/s]
 41%|████      | 26871/65536 [4:36:28<6:29:21,  1.66it/s]
 41%|████      | 26872/65536 [4:36:29<6:26:13,  1.67it/s]
 41%|████      | 26873/65536 [4:36:29<6:35:53,  1.63it/s]
 41%|████      | 26874/65536 [4:36:30<6:41:08,  1.61it/s]
 41%|████      | 26875/65536 [4:36:31<6:45:20,  1.59it/s]
 41%|████      | 26876/65536 [4:36:31<6:56:49,  1.55it/s]
 41%|████      | 26877/65536 [4:36:32<6:41:49,  1.60it/s]
 41%|████      | 26878/65536 [4:36:33<6:41:10,  1.61it/s]
 41%|████      | 26879/65536 [4:36:33<6:44:16,  1.59it/s]
 41%|████      | 26880/65536 [4:36:34<6:42:32,  1.60it/s]
                                                         
{'loss': 1.7432, 'learning_rate': 6.324843883923105e-07, 'epoch': 1659.26}
+
 41%|████      | 26880/65536 [4:36:34<6:42:32,  1.60it/s]
 41%|████      | 26881/65536 [4:36:34<6:33:35,  1.64it/s]
 41%|████      | 26882/65536 [4:36:35<6:31:47,  1.64it/s]
 41%|████      | 26883/65536 [4:36:36<6:40:20,  1.61it/s]
 41%|████      | 26884/65536 [4:36:36<6:33:28,  1.64it/s]
 41%|████      | 26885/65536 [4:36:37<6:40:24,  1.61it/s]
 41%|████      | 26886/65536 [4:36:38<6:37:24,  1.62it/s]
 41%|████      | 26887/65536 [4:36:38<6:40:40,  1.61it/s]
 41%|████      | 26888/65536 [4:36:39<6:33:25,  1.64it/s]
 41%|████      | 26889/65536 [4:36:39<6:40:13,  1.61it/s]
 41%|████      | 26890/65536 [4:36:40<6:35:30,  1.63it/s]
 41%|████      | 26891/65536 [4:36:41<6:31:39,  1.64it/s]
 41%|████      | 26892/65536 [4:36:41<6:22:53,  1.68it/s]
 41%|████      | 26893/65536 [4:36:42<6:31:22,  1.65it/s]
 41%|████      | 26894/65536 [4:36:42<6:31:28,  1.65it/s]
 41%|████      | 26895/65536 [4:36:43<6:39:30,  1.61it/s]
 41%|████      | 26896/65536 [4:36:44<6:41:23,  1.60it/s]
 41%|████      | 26897/65536 [4:36:44<6:56:57,  1.54it/s]
 41%|████      | 26898/65536 [4:36:45<6:47:04,  1.58it/s]
 41%|████      | 26899/65536 [4:36:46<6:41:46,  1.60it/s]
 41%|████      | 26900/65536 [4:36:46<6:41:05,  1.61it/s]
                                                         
{'loss': 1.7564, 'learning_rate': 6.322088894330844e-07, 'epoch': 1660.49}
+
 41%|████      | 26900/65536 [4:36:46<6:41:05,  1.61it/s]
 41%|████      | 26901/65536 [4:36:47<6:40:58,  1.61it/s]
 41%|████      | 26902/65536 [4:36:47<6:34:20,  1.63it/s]
 41%|████      | 26903/65536 [4:36:48<6:32:48,  1.64it/s]
 41%|████      | 26904/65536 [4:36:49<6:28:02,  1.66it/s]
 41%|████      | 26905/65536 [4:36:49<6:25:10,  1.67it/s]
 41%|████      | 26906/65536 [4:36:50<6:17:17,  1.71it/s]
 41%|████      | 26907/65536 [4:36:50<6:27:43,  1.66it/s]
 41%|████      | 26908/65536 [4:36:51<6:24:29,  1.67it/s]
 41%|████      | 26909/65536 [4:36:52<6:41:47,  1.60it/s]
 41%|████      | 26910/65536 [4:36:52<6:39:12,  1.61it/s]
 41%|████      | 26911/65536 [4:36:53<6:38:01,  1.62it/s]
 41%|████      | 26912/65536 [4:36:54<6:43:05,  1.60it/s]
 41%|████      | 26913/65536 [4:36:54<6:41:49,  1.60it/s]
 41%|████      | 26914/65536 [4:36:55<6:38:14,  1.62it/s]
 41%|████      | 26915/65536 [4:36:55<6:38:20,  1.62it/s]
 41%|████      | 26916/65536 [4:36:56<6:32:47,  1.64it/s]
 41%|████      | 26917/65536 [4:36:57<6:43:07,  1.60it/s]
 41%|████      | 26918/65536 [4:36:57<6:33:41,  1.63it/s]
 41%|████      | 26919/65536 [4:36:58<6:39:53,  1.61it/s]
 41%|████      | 26920/65536 [4:36:59<6:36:34,  1.62it/s]
                                                         
{'loss': 1.7716, 'learning_rate': 6.319333904738582e-07, 'epoch': 1661.73}
+
 41%|████      | 26920/65536 [4:36:59<6:36:34,  1.62it/s]
 41%|████      | 26921/65536 [4:36:59<6:32:20,  1.64it/s]
 41%|████      | 26922/65536 [4:37:00<6:30:18,  1.65it/s]
 41%|████      | 26923/65536 [4:37:00<6:32:12,  1.64it/s]
 41%|████      | 26924/65536 [4:37:01<6:27:53,  1.66it/s]
 41%|████      | 26925/65536 [4:37:02<6:38:11,  1.62it/s]
 41%|████      | 26926/65536 [4:37:02<6:35:31,  1.63it/s]
 41%|████      | 26927/65536 [4:37:03<6:33:30,  1.64it/s]
 41%|████      | 26928/65536 [4:37:03<6:36:49,  1.62it/s]
 41%|████      | 26929/65536 [4:37:04<6:46:30,  1.58it/s]
 41%|████      | 26930/65536 [4:37:05<6:44:14,  1.59it/s]
 41%|████      | 26931/65536 [4:37:05<6:39:51,  1.61it/s]
 41%|████      | 26932/65536 [4:37:06<6:38:34,  1.61it/s]
 41%|████      | 26933/65536 [4:37:07<6:35:15,  1.63it/s]
 41%|████      | 26934/65536 [4:37:07<6:31:52,  1.64it/s]
 41%|████      | 26935/65536 [4:37:08<6:39:57,  1.61it/s]
 41%|████      | 26936/65536 [4:37:08<6:43:00,  1.60it/s]
 41%|████      | 26937/65536 [4:37:09<6:36:14,  1.62it/s]
 41%|████      | 26938/65536 [4:37:10<6:32:03,  1.64it/s]
 41%|████      | 26939/65536 [4:37:10<6:34:57,  1.63it/s]
 41%|████      | 26940/65536 [4:37:11<6:38:54,  1.61it/s]
                                                         
{'loss': 1.7049, 'learning_rate': 6.316578915146321e-07, 'epoch': 1662.96}
+
 41%|████      | 26940/65536 [4:37:11<6:38:54,  1.61it/s]
 41%|████      | 26941/65536 [4:37:12<6:47:23,  1.58it/s]
 41%|████      | 26942/65536 [4:37:12<7:01:34,  1.53it/s]
 41%|████      | 26943/65536 [4:37:13<6:48:25,  1.57it/s]
 41%|████      | 26944/65536 [4:37:13<6:38:35,  1.61it/s]
 41%|████      | 26945/65536 [4:37:14<6:29:53,  1.65it/s]
 41%|████      | 26946/65536 [4:37:15<6:36:23,  1.62it/s]
 41%|████      | 26947/65536 [4:37:15<6:35:27,  1.63it/s]
 41%|████      | 26948/65536 [4:37:16<6:41:40,  1.60it/s]
 41%|████      | 26949/65536 [4:37:16<6:31:11,  1.64it/s]
 41%|████      | 26950/65536 [4:37:17<6:32:17,  1.64it/s]
 41%|████      | 26951/65536 [4:37:18<6:28:50,  1.65it/s]
 41%|████      | 26952/65536 [4:37:18<6:41:35,  1.60it/s]
 41%|████      | 26953/65536 [4:37:19<6:45:09,  1.59it/s]
 41%|████      | 26954/65536 [4:37:20<6:39:35,  1.61it/s]
 41%|████      | 26955/65536 [4:37:20<6:38:51,  1.61it/s]
 41%|████      | 26956/65536 [4:37:21<6:33:27,  1.63it/s]
 41%|████      | 26957/65536 [4:37:21<6:39:08,  1.61it/s]
 41%|████      | 26958/65536 [4:37:22<6:40:00,  1.61it/s]
 41%|████      | 26959/65536 [4:37:23<6:34:32,  1.63it/s]
 41%|████      | 26960/65536 [4:37:23<6:26:21,  1.66it/s]
                                                         
{'loss': 1.7469, 'learning_rate': 6.313823925554059e-07, 'epoch': 1664.2}
+
 41%|████      | 26960/65536 [4:37:23<6:26:21,  1.66it/s]
 41%|████      | 26961/65536 [4:37:24<6:26:15,  1.66it/s]
 41%|████      | 26962/65536 [4:37:24<6:24:07,  1.67it/s]
 41%|████      | 26963/65536 [4:37:25<6:29:28,  1.65it/s]
 41%|████      | 26964/65536 [4:37:26<6:30:26,  1.65it/s]
 41%|████      | 26965/65536 [4:37:26<6:46:13,  1.58it/s]
 41%|████      | 26966/65536 [4:37:27<6:35:39,  1.62it/s]
 41%|████      | 26967/65536 [4:37:27<6:35:05,  1.63it/s]
 41%|████      | 26968/65536 [4:37:28<6:38:10,  1.61it/s]
 41%|████      | 26969/65536 [4:37:29<6:26:09,  1.66it/s]
 41%|████      | 26970/65536 [4:37:29<6:35:43,  1.62it/s]
 41%|████      | 26971/65536 [4:37:30<6:49:04,  1.57it/s]
 41%|████      | 26972/65536 [4:37:31<6:45:36,  1.58it/s]
 41%|████      | 26973/65536 [4:37:31<6:36:39,  1.62it/s]
 41%|████      | 26974/65536 [4:37:32<6:53:35,  1.55it/s]
 41%|████      | 26975/65536 [4:37:33<6:41:30,  1.60it/s]
 41%|████      | 26976/65536 [4:37:33<6:35:58,  1.62it/s]
 41%|████      | 26977/65536 [4:37:34<6:37:54,  1.62it/s]
 41%|████      | 26978/65536 [4:37:34<6:25:29,  1.67it/s]
 41%|████      | 26979/65536 [4:37:35<6:37:09,  1.62it/s]
 41%|████      | 26980/65536 [4:37:36<6:36:10,  1.62it/s]
                                                         
{'loss': 1.7347, 'learning_rate': 6.311068935961798e-07, 'epoch': 1665.43}
+
 41%|████      | 26980/65536 [4:37:36<6:36:10,  1.62it/s]
 41%|████      | 26981/65536 [4:37:36<6:38:33,  1.61it/s]
 41%|████      | 26982/65536 [4:37:37<6:36:18,  1.62it/s]
 41%|████      | 26983/65536 [4:37:37<6:38:13,  1.61it/s]
 41%|████      | 26984/65536 [4:37:38<6:37:19,  1.62it/s]
 41%|████      | 26985/65536 [4:37:39<6:43:45,  1.59it/s]
 41%|████      | 26986/65536 [4:37:39<6:37:07,  1.62it/s]
 41%|████      | 26987/65536 [4:37:40<6:34:33,  1.63it/s]
 41%|████      | 26988/65536 [4:37:40<6:26:36,  1.66it/s]
 41%|████      | 26989/65536 [4:37:41<6:31:47,  1.64it/s]
 41%|████      | 26990/65536 [4:37:42<6:46:46,  1.58it/s]
 41%|████      | 26991/65536 [4:37:42<6:38:41,  1.61it/s]
 41%|████      | 26992/65536 [4:37:43<6:37:00,  1.62it/s]
 41%|████      | 26993/65536 [4:37:44<6:39:13,  1.61it/s]
 41%|████      | 26994/65536 [4:37:44<6:34:14,  1.63it/s]
 41%|████      | 26995/65536 [4:37:45<6:47:09,  1.58it/s]
 41%|████      | 26996/65536 [4:37:45<6:42:11,  1.60it/s]
 41%|████      | 26997/65536 [4:37:46<6:33:18,  1.63it/s]
 41%|████      | 26998/65536 [4:37:47<6:39:54,  1.61it/s]
 41%|████      | 26999/65536 [4:37:47<6:29:23,  1.65it/s]
 41%|████      | 27000/65536 [4:37:48<6:28:28,  1.65it/s]
                                                         
{'loss': 1.7357, 'learning_rate': 6.308313946369535e-07, 'epoch': 1666.67}
+
 41%|████      | 27000/65536 [4:37:48<6:28:28,  1.65it/s]
 41%|████      | 27001/65536 [4:37:48<6:27:27,  1.66it/s]
 41%|████      | 27002/65536 [4:37:49<6:28:23,  1.65it/s]
 41%|████      | 27003/65536 [4:37:50<6:18:56,  1.69it/s]
 41%|████      | 27004/65536 [4:37:50<6:29:12,  1.65it/s]
 41%|████      | 27005/65536 [4:37:51<6:33:01,  1.63it/s]
 41%|████      | 27006/65536 [4:37:52<6:39:39,  1.61it/s]
 41%|████      | 27007/65536 [4:37:52<6:34:38,  1.63it/s]
 41%|████      | 27008/65536 [4:37:53<6:25:39,  1.67it/s]
 41%|████      | 27009/65536 [4:37:53<6:20:53,  1.69it/s]
 41%|████      | 27010/65536 [4:37:54<6:15:57,  1.71it/s]
 41%|████      | 27011/65536 [4:37:54<6:16:50,  1.70it/s]
 41%|████      | 27012/65536 [4:37:55<6:33:51,  1.63it/s]
 41%|████      | 27013/65536 [4:37:56<6:40:05,  1.60it/s]
 41%|████      | 27014/65536 [4:37:56<6:44:23,  1.59it/s]
 41%|████      | 27015/65536 [4:37:57<6:39:47,  1.61it/s]
 41%|████      | 27016/65536 [4:37:58<6:36:58,  1.62it/s]
 41%|████      | 27017/65536 [4:37:58<6:33:42,  1.63it/s]
 41%|████      | 27018/65536 [4:37:59<6:31:24,  1.64it/s]
 41%|████      | 27019/65536 [4:37:59<6:24:10,  1.67it/s]
 41%|████      | 27020/65536 [4:38:00<6:25:06,  1.67it/s]
                                                         
{'loss': 1.7542, 'learning_rate': 6.305558956777274e-07, 'epoch': 1667.9}
+
 41%|████      | 27020/65536 [4:38:00<6:25:06,  1.67it/s]
 41%|████      | 27021/65536 [4:38:01<6:41:00,  1.60it/s]
 41%|████      | 27022/65536 [4:38:01<6:54:58,  1.55it/s]
 41%|████      | 27023/65536 [4:38:02<6:44:17,  1.59it/s]
 41%|████      | 27024/65536 [4:38:03<6:37:19,  1.62it/s]
 41%|████      | 27025/65536 [4:38:03<6:38:34,  1.61it/s]
 41%|████      | 27026/65536 [4:38:04<6:35:11,  1.62it/s]
 41%|████      | 27027/65536 [4:38:04<6:39:03,  1.61it/s]
 41%|████      | 27028/65536 [4:38:05<6:31:46,  1.64it/s]
 41%|████      | 27029/65536 [4:38:06<6:37:13,  1.62it/s]
 41%|████      | 27030/65536 [4:38:06<6:32:29,  1.64it/s]
 41%|████      | 27031/65536 [4:38:07<6:35:04,  1.62it/s]
 41%|████      | 27032/65536 [4:38:07<6:31:27,  1.64it/s]
 41%|████      | 27033/65536 [4:38:08<6:35:04,  1.62it/s]
 41%|████▏     | 27034/65536 [4:38:09<6:39:20,  1.61it/s]
 41%|████▏     | 27035/65536 [4:38:09<6:37:11,  1.62it/s]
 41%|████▏     | 27036/65536 [4:38:10<6:39:38,  1.61it/s]
 41%|████▏     | 27037/65536 [4:38:11<6:37:36,  1.61it/s]
 41%|████▏     | 27038/65536 [4:38:11<6:49:05,  1.57it/s]
 41%|████▏     | 27039/65536 [4:38:12<6:44:12,  1.59it/s]
 41%|████▏     | 27040/65536 [4:38:13<6:59:51,  1.53it/s]
                                                         
{'loss': 1.6961, 'learning_rate': 6.302803967185013e-07, 'epoch': 1669.14}
+
 41%|████▏     | 27040/65536 [4:38:13<6:59:51,  1.53it/s]
 41%|████▏     | 27041/65536 [4:38:13<6:45:15,  1.58it/s]
 41%|████▏     | 27042/65536 [4:38:14<6:35:01,  1.62it/s]
 41%|████▏     | 27043/65536 [4:38:14<6:26:23,  1.66it/s]
 41%|████▏     | 27044/65536 [4:38:15<6:18:28,  1.70it/s]
 41%|████▏     | 27045/65536 [4:38:16<6:18:45,  1.69it/s]
 41%|████▏     | 27046/65536 [4:38:16<6:29:28,  1.65it/s]
 41%|████▏     | 27047/65536 [4:38:17<6:29:53,  1.65it/s]
 41%|████▏     | 27048/65536 [4:38:17<6:20:52,  1.68it/s]
 41%|████▏     | 27049/65536 [4:38:18<6:20:42,  1.68it/s]
 41%|████▏     | 27050/65536 [4:38:19<6:32:38,  1.63it/s]
 41%|████▏     | 27051/65536 [4:38:19<6:33:58,  1.63it/s]
 41%|████▏     | 27052/65536 [4:38:20<6:31:10,  1.64it/s]
 41%|████▏     | 27053/65536 [4:38:20<6:27:24,  1.66it/s]
 41%|████▏     | 27054/65536 [4:38:21<6:28:58,  1.65it/s]
 41%|████▏     | 27055/65536 [4:38:22<6:42:09,  1.59it/s]
 41%|████▏     | 27056/65536 [4:38:22<6:37:39,  1.61it/s]
 41%|████▏     | 27057/65536 [4:38:23<6:32:33,  1.63it/s]
 41%|████▏     | 27058/65536 [4:38:24<6:40:06,  1.60it/s]
 41%|████▏     | 27059/65536 [4:38:24<6:44:04,  1.59it/s]
 41%|████▏     | 27060/65536 [4:38:25<6:34:15,  1.63it/s]
                                                         
{'loss': 1.7537, 'learning_rate': 6.30004897759275e-07, 'epoch': 1670.37}
+
 41%|████▏     | 27060/65536 [4:38:25<6:34:15,  1.63it/s]
 41%|████▏     | 27061/65536 [4:38:25<6:39:51,  1.60it/s]
 41%|████▏     | 27062/65536 [4:38:26<6:26:36,  1.66it/s]
 41%|████▏     | 27063/65536 [4:38:27<6:25:33,  1.66it/s]
 41%|████▏     | 27064/65536 [4:38:27<6:22:49,  1.67it/s]
 41%|████▏     | 27065/65536 [4:38:28<6:32:59,  1.63it/s]
 41%|████▏     | 27066/65536 [4:38:28<6:27:59,  1.65it/s]
 41%|████▏     | 27067/65536 [4:38:29<6:35:59,  1.62it/s]
 41%|████▏     | 27068/65536 [4:38:30<6:31:47,  1.64it/s]
 41%|████▏     | 27069/65536 [4:38:30<6:29:03,  1.65it/s]
 41%|████▏     | 27070/65536 [4:38:31<6:28:17,  1.65it/s]
 41%|████▏     | 27071/65536 [4:38:31<6:39:36,  1.60it/s]
 41%|████▏     | 27072/65536 [4:38:32<6:31:48,  1.64it/s]
 41%|████▏     | 27073/65536 [4:38:33<6:34:35,  1.62it/s]
 41%|████▏     | 27074/65536 [4:38:33<6:41:24,  1.60it/s]
 41%|████▏     | 27075/65536 [4:38:34<6:39:17,  1.61it/s]
 41%|████▏     | 27076/65536 [4:38:35<6:40:40,  1.60it/s]
 41%|████▏     | 27077/65536 [4:38:35<6:43:35,  1.59it/s]
 41%|████▏     | 27078/65536 [4:38:36<6:30:54,  1.64it/s]
 41%|████▏     | 27079/65536 [4:38:36<6:28:51,  1.65it/s]
 41%|████▏     | 27080/65536 [4:38:37<6:23:55,  1.67it/s]
                                                         
{'loss': 1.7612, 'learning_rate': 6.297293988000489e-07, 'epoch': 1671.6}
+
 41%|████▏     | 27080/65536 [4:38:37<6:23:55,  1.67it/s]
 41%|████▏     | 27081/65536 [4:38:38<6:21:40,  1.68it/s]
 41%|████▏     | 27082/65536 [4:38:38<6:24:16,  1.67it/s]
 41%|████▏     | 27083/65536 [4:38:39<6:23:05,  1.67it/s]
 41%|████▏     | 27084/65536 [4:38:39<6:27:35,  1.65it/s]
 41%|████▏     | 27085/65536 [4:38:40<6:26:02,  1.66it/s]
 41%|████▏     | 27086/65536 [4:38:41<6:36:15,  1.62it/s]
 41%|████▏     | 27087/65536 [4:38:41<6:51:13,  1.56it/s]
 41%|████▏     | 27088/65536 [4:38:42<6:49:52,  1.56it/s]
 41%|████▏     | 27089/65536 [4:38:43<6:41:06,  1.60it/s]
 41%|████▏     | 27090/65536 [4:38:43<6:44:30,  1.58it/s]
 41%|████▏     | 27091/65536 [4:38:44<6:31:51,  1.64it/s]
 41%|████▏     | 27092/65536 [4:38:44<6:27:41,  1.65it/s]
 41%|████▏     | 27093/65536 [4:38:45<6:31:32,  1.64it/s]
 41%|████▏     | 27094/65536 [4:38:46<6:31:34,  1.64it/s]
 41%|████▏     | 27095/65536 [4:38:46<6:37:01,  1.61it/s]
 41%|████▏     | 27096/65536 [4:38:47<6:34:40,  1.62it/s]
 41%|████▏     | 27097/65536 [4:38:47<6:39:13,  1.60it/s]
 41%|████▏     | 27098/65536 [4:38:48<6:40:45,  1.60it/s]
 41%|████▏     | 27099/65536 [4:38:49<6:32:25,  1.63it/s]
 41%|████▏     | 27100/65536 [4:38:49<6:26:47,  1.66it/s]
                                                         
{'loss': 1.7779, 'learning_rate': 6.294538998408227e-07, 'epoch': 1672.84}
+
 41%|████▏     | 27100/65536 [4:38:49<6:26:47,  1.66it/s]
 41%|████▏     | 27101/65536 [4:38:50<6:31:58,  1.63it/s]
 41%|████▏     | 27102/65536 [4:38:51<6:33:02,  1.63it/s]
 41%|████▏     | 27103/65536 [4:38:51<6:37:50,  1.61it/s]
 41%|████▏     | 27104/65536 [4:38:52<6:37:06,  1.61it/s]
 41%|████▏     | 27105/65536 [4:38:52<6:36:17,  1.62it/s]
 41%|████▏     | 27106/65536 [4:38:53<6:34:23,  1.62it/s]
 41%|████▏     | 27107/65536 [4:38:54<6:30:59,  1.64it/s]
 41%|████▏     | 27108/65536 [4:38:54<6:25:44,  1.66it/s]
 41%|████▏     | 27109/65536 [4:38:55<6:19:57,  1.69it/s]
 41%|████▏     | 27110/65536 [4:38:55<6:16:01,  1.70it/s]
 41%|████▏     | 27111/65536 [4:38:56<6:16:05,  1.70it/s]
 41%|████▏     | 27112/65536 [4:38:57<6:22:47,  1.67it/s]
 41%|████▏     | 27113/65536 [4:38:57<6:28:28,  1.65it/s]
 41%|████▏     | 27114/65536 [4:38:58<6:49:06,  1.57it/s]
 41%|████▏     | 27115/65536 [4:38:58<6:42:30,  1.59it/s]
 41%|████▏     | 27116/65536 [4:38:59<6:42:35,  1.59it/s]
 41%|████▏     | 27117/65536 [4:39:00<6:38:28,  1.61it/s]
 41%|████▏     | 27118/65536 [4:39:00<6:35:15,  1.62it/s]
 41%|████▏     | 27119/65536 [4:39:01<6:43:46,  1.59it/s]
 41%|████▏     | 27120/65536 [4:39:02<6:29:03,  1.65it/s]
                                                         
{'loss': 1.7238, 'learning_rate': 6.291784008815966e-07, 'epoch': 1674.07}
+
 41%|████▏     | 27120/65536 [4:39:02<6:29:03,  1.65it/s]
 41%|████▏     | 27121/65536 [4:39:02<6:44:42,  1.58it/s]
 41%|████▏     | 27122/65536 [4:39:03<6:38:01,  1.61it/s]
 41%|████▏     | 27123/65536 [4:39:03<6:31:43,  1.63it/s]
 41%|████▏     | 27124/65536 [4:39:04<6:45:33,  1.58it/s]
 41%|████▏     | 27125/65536 [4:39:05<6:41:34,  1.59it/s]
 41%|████▏     | 27126/65536 [4:39:05<6:35:03,  1.62it/s]
 41%|████▏     | 27127/65536 [4:39:06<6:31:04,  1.64it/s]
 41%|████▏     | 27128/65536 [4:39:07<6:40:28,  1.60it/s]
 41%|████▏     | 27129/65536 [4:39:07<6:47:08,  1.57it/s]
 41%|████▏     | 27130/65536 [4:39:08<6:37:13,  1.61it/s]
 41%|████▏     | 27131/65536 [4:39:08<6:31:39,  1.63it/s]
 41%|████▏     | 27132/65536 [4:39:09<6:33:47,  1.63it/s]
 41%|████▏     | 27133/65536 [4:39:10<6:32:55,  1.63it/s]
 41%|████▏     | 27134/65536 [4:39:10<6:25:04,  1.66it/s]
 41%|████▏     | 27135/65536 [4:39:11<6:26:12,  1.66it/s]
 41%|████▏     | 27136/65536 [4:39:11<6:41:11,  1.60it/s]
 41%|████▏     | 27137/65536 [4:39:12<6:30:43,  1.64it/s]
 41%|████▏     | 27138/65536 [4:39:13<6:26:28,  1.66it/s]
 41%|████▏     | 27139/65536 [4:39:13<6:25:27,  1.66it/s]
 41%|████▏     | 27140/65536 [4:39:14<6:39:26,  1.60it/s]
                                                         
{'loss': 1.7107, 'learning_rate': 6.289029019223705e-07, 'epoch': 1675.31}
+
 41%|████▏     | 27140/65536 [4:39:14<6:39:26,  1.60it/s]
 41%|████▏     | 27141/65536 [4:39:14<6:27:07,  1.65it/s]
 41%|████▏     | 27142/65536 [4:39:15<6:27:06,  1.65it/s]
 41%|████▏     | 27143/65536 [4:39:16<6:42:58,  1.59it/s]
 41%|████▏     | 27144/65536 [4:39:16<6:34:42,  1.62it/s]
 41%|████▏     | 27145/65536 [4:39:17<6:39:11,  1.60it/s]
 41%|████▏     | 27146/65536 [4:39:18<6:38:07,  1.61it/s]
 41%|████▏     | 27147/65536 [4:39:18<6:34:42,  1.62it/s]
 41%|████▏     | 27148/65536 [4:39:19<6:41:41,  1.59it/s]
 41%|████▏     | 27149/65536 [4:39:19<6:36:09,  1.61it/s]
 41%|████▏     | 27150/65536 [4:39:20<6:26:14,  1.66it/s]
 41%|████▏     | 27151/65536 [4:39:21<6:23:51,  1.67it/s]
 41%|████▏     | 27152/65536 [4:39:21<6:31:06,  1.64it/s]
 41%|████▏     | 27153/65536 [4:39:22<6:23:50,  1.67it/s]
 41%|████▏     | 27154/65536 [4:39:22<6:28:55,  1.64it/s]
 41%|████▏     | 27155/65536 [4:39:23<6:41:22,  1.59it/s]
 41%|████▏     | 27156/65536 [4:39:24<6:43:39,  1.58it/s]
 41%|████▏     | 27157/65536 [4:39:24<6:32:28,  1.63it/s]
 41%|████▏     | 27158/65536 [4:39:25<6:29:44,  1.64it/s]
 41%|████▏     | 27159/65536 [4:39:26<6:29:22,  1.64it/s]
 41%|████▏     | 27160/65536 [4:39:26<6:37:44,  1.61it/s]
                                                         
{'loss': 1.7065, 'learning_rate': 6.286274029631443e-07, 'epoch': 1676.54}
+
 41%|████▏     | 27160/65536 [4:39:26<6:37:44,  1.61it/s]
 41%|████▏     | 27161/65536 [4:39:27<6:39:21,  1.60it/s]
 41%|████▏     | 27162/65536 [4:39:27<6:33:33,  1.63it/s]
 41%|████▏     | 27163/65536 [4:39:28<6:44:00,  1.58it/s]
 41%|████▏     | 27164/65536 [4:39:29<6:36:57,  1.61it/s]
 41%|████▏     | 27165/65536 [4:39:29<6:30:38,  1.64it/s]
 41%|████▏     | 27166/65536 [4:39:30<6:29:32,  1.64it/s]
 41%|████▏     | 27167/65536 [4:39:30<6:21:02,  1.68it/s]
 41%|████▏     | 27168/65536 [4:39:31<6:33:21,  1.63it/s]
 41%|████▏     | 27169/65536 [4:39:32<6:25:16,  1.66it/s]
 41%|████▏     | 27170/65536 [4:39:32<6:28:39,  1.65it/s]
 41%|████▏     | 27171/65536 [4:39:33<6:41:54,  1.59it/s]
 41%|████▏     | 27172/65536 [4:39:34<6:35:34,  1.62it/s]
 41%|████▏     | 27173/65536 [4:39:34<6:34:34,  1.62it/s]
 41%|████▏     | 27174/65536 [4:39:35<6:39:31,  1.60it/s]
 41%|████▏     | 27175/65536 [4:39:35<6:42:17,  1.59it/s]
 41%|████▏     | 27176/65536 [4:39:36<6:33:54,  1.62it/s]
 41%|████▏     | 27177/65536 [4:39:37<6:22:18,  1.67it/s]
 41%|████▏     | 27178/65536 [4:39:37<6:28:33,  1.65it/s]
 41%|████▏     | 27179/65536 [4:39:38<6:23:10,  1.67it/s]
 41%|████▏     | 27180/65536 [4:39:38<6:18:15,  1.69it/s]
                                                         
{'loss': 1.7517, 'learning_rate': 6.283519040039182e-07, 'epoch': 1677.78}
+
 41%|████▏     | 27180/65536 [4:39:38<6:18:15,  1.69it/s]
 41%|████▏     | 27181/65536 [4:39:39<6:23:49,  1.67it/s]
 41%|████▏     | 27182/65536 [4:39:40<6:29:27,  1.64it/s]
 41%|████▏     | 27183/65536 [4:39:40<6:21:04,  1.68it/s]
 41%|████▏     | 27184/65536 [4:39:41<6:43:18,  1.58it/s]
 41%|████▏     | 27185/65536 [4:39:42<6:40:14,  1.60it/s]
 41%|████▏     | 27186/65536 [4:39:42<6:31:18,  1.63it/s]
 41%|████▏     | 27187/65536 [4:39:43<6:25:41,  1.66it/s]
 41%|████▏     | 27188/65536 [4:39:43<6:34:25,  1.62it/s]
 41%|████▏     | 27189/65536 [4:39:44<6:23:19,  1.67it/s]
 41%|████▏     | 27190/65536 [4:39:45<6:19:34,  1.68it/s]
 41%|████▏     | 27191/65536 [4:39:45<6:25:51,  1.66it/s]
 41%|████▏     | 27192/65536 [4:39:46<6:40:13,  1.60it/s]
 41%|████▏     | 27193/65536 [4:39:46<6:35:32,  1.62it/s]
 41%|████▏     | 27194/65536 [4:39:47<6:23:19,  1.67it/s]
 41%|████▏     | 27195/65536 [4:39:48<6:28:48,  1.64it/s]
 41%|████▏     | 27196/65536 [4:39:48<6:25:12,  1.66it/s]
 41%|████▏     | 27197/65536 [4:39:49<6:28:13,  1.65it/s]
 42%|████▏     | 27198/65536 [4:39:49<6:24:37,  1.66it/s]
 42%|████▏     | 27199/65536 [4:39:50<6:37:06,  1.61it/s]
 42%|████▏     | 27200/65536 [4:39:51<6:43:41,  1.58it/s]
                                                         
{'loss': 1.7725, 'learning_rate': 6.28076405044692e-07, 'epoch': 1679.01}
+
 42%|████▏     | 27200/65536 [4:39:51<6:43:41,  1.58it/s]
 42%|████▏     | 27201/65536 [4:39:51<6:38:04,  1.60it/s]
 42%|████▏     | 27202/65536 [4:39:52<6:32:07,  1.63it/s]
 42%|████▏     | 27203/65536 [4:39:53<6:46:24,  1.57it/s]
 42%|████▏     | 27204/65536 [4:39:53<6:42:29,  1.59it/s]
 42%|████▏     | 27205/65536 [4:39:54<6:32:00,  1.63it/s]
 42%|████▏     | 27206/65536 [4:39:54<6:29:15,  1.64it/s]
 42%|████▏     | 27207/65536 [4:39:55<6:25:21,  1.66it/s]
 42%|████▏     | 27208/65536 [4:39:56<6:29:18,  1.64it/s]
 42%|████▏     | 27209/65536 [4:39:56<6:24:49,  1.66it/s]
 42%|████▏     | 27210/65536 [4:39:57<6:34:12,  1.62it/s]
 42%|████▏     | 27211/65536 [4:39:57<6:38:22,  1.60it/s]
 42%|████▏     | 27212/65536 [4:39:58<6:27:10,  1.65it/s]
 42%|████▏     | 27213/65536 [4:39:59<6:18:20,  1.69it/s]
 42%|████▏     | 27214/65536 [4:39:59<6:21:36,  1.67it/s]
 42%|████▏     | 27215/65536 [4:40:00<6:21:16,  1.68it/s]
 42%|████▏     | 27216/65536 [4:40:00<6:15:53,  1.70it/s]
 42%|████▏     | 27217/65536 [4:40:01<6:36:31,  1.61it/s]
 42%|████▏     | 27218/65536 [4:40:02<6:31:53,  1.63it/s]
 42%|████▏     | 27219/65536 [4:40:02<6:35:02,  1.62it/s]
 42%|████▏     | 27220/65536 [4:40:03<6:31:22,  1.63it/s]
                                                         
{'loss': 1.7428, 'learning_rate': 6.278009060854659e-07, 'epoch': 1680.25}
+
 42%|████▏     | 27220/65536 [4:40:03<6:31:22,  1.63it/s]
 42%|████▏     | 27221/65536 [4:40:04<6:30:38,  1.63it/s]
 42%|████▏     | 27222/65536 [4:40:04<6:16:40,  1.70it/s]
 42%|████▏     | 27223/65536 [4:40:05<6:31:58,  1.63it/s]
 42%|████▏     | 27224/65536 [4:40:05<6:39:06,  1.60it/s]
 42%|████▏     | 27225/65536 [4:40:06<6:40:04,  1.60it/s]
 42%|████▏     | 27226/65536 [4:40:07<6:34:09,  1.62it/s]
 42%|████▏     | 27227/65536 [4:40:07<6:33:12,  1.62it/s]
 42%|████▏     | 27228/65536 [4:40:08<6:34:51,  1.62it/s]
 42%|████▏     | 27229/65536 [4:40:08<6:40:54,  1.59it/s]
 42%|████▏     | 27230/65536 [4:40:09<6:37:28,  1.61it/s]
 42%|████▏     | 27231/65536 [4:40:10<6:35:37,  1.61it/s]
 42%|████▏     | 27232/65536 [4:40:10<6:30:19,  1.64it/s]
 42%|████▏     | 27233/65536 [4:40:11<6:50:56,  1.55it/s]
 42%|████▏     | 27234/65536 [4:40:12<6:40:32,  1.59it/s]
 42%|████▏     | 27235/65536 [4:40:12<6:36:54,  1.61it/s]
 42%|████▏     | 27236/65536 [4:40:13<6:33:50,  1.62it/s]
 42%|████▏     | 27237/65536 [4:40:13<6:36:38,  1.61it/s]
 42%|████▏     | 27238/65536 [4:40:14<6:39:49,  1.60it/s]
 42%|████▏     | 27239/65536 [4:40:15<6:35:22,  1.61it/s]
 42%|████▏     | 27240/65536 [4:40:15<6:29:45,  1.64it/s]
                                                         
{'loss': 1.7686, 'learning_rate': 6.275254071262399e-07, 'epoch': 1681.48}
+
 42%|████▏     | 27240/65536 [4:40:15<6:29:45,  1.64it/s]
 42%|████▏     | 27241/65536 [4:40:16<6:35:12,  1.61it/s]
 42%|████▏     | 27242/65536 [4:40:17<6:31:06,  1.63it/s]
 42%|████▏     | 27243/65536 [4:40:17<6:26:40,  1.65it/s]
 42%|████▏     | 27244/65536 [4:40:18<6:27:22,  1.65it/s]
 42%|████▏     | 27245/65536 [4:40:18<6:23:15,  1.67it/s]
 42%|████▏     | 27246/65536 [4:40:19<6:30:34,  1.63it/s]
 42%|████▏     | 27247/65536 [4:40:20<6:32:19,  1.63it/s]
 42%|████▏     | 27248/65536 [4:40:20<6:26:56,  1.65it/s]
 42%|████▏     | 27249/65536 [4:40:21<6:40:12,  1.59it/s]
 42%|████▏     | 27250/65536 [4:40:21<6:31:51,  1.63it/s]
 42%|████▏     | 27251/65536 [4:40:22<6:38:36,  1.60it/s]
 42%|████▏     | 27252/65536 [4:40:23<6:34:06,  1.62it/s]
 42%|████▏     | 27253/65536 [4:40:23<6:35:15,  1.61it/s]
 42%|████▏     | 27254/65536 [4:40:24<6:34:38,  1.62it/s]
 42%|████▏     | 27255/65536 [4:40:24<6:26:31,  1.65it/s]
 42%|████▏     | 27256/65536 [4:40:25<6:32:14,  1.63it/s]
 42%|████▏     | 27257/65536 [4:40:26<6:36:49,  1.61it/s]
 42%|████▏     | 27258/65536 [4:40:26<6:23:22,  1.66it/s]
 42%|████▏     | 27259/65536 [4:40:27<6:30:07,  1.64it/s]
 42%|████▏     | 27260/65536 [4:40:28<6:29:15,  1.64it/s]
                                                         
{'loss': 1.7229, 'learning_rate': 6.272499081670135e-07, 'epoch': 1682.72}
+
 42%|████▏     | 27260/65536 [4:40:28<6:29:15,  1.64it/s]
 42%|████▏     | 27261/65536 [4:40:28<6:27:56,  1.64it/s]
 42%|████▏     | 27262/65536 [4:40:29<6:30:40,  1.63it/s]
 42%|████▏     | 27263/65536 [4:40:29<6:29:26,  1.64it/s]
 42%|████▏     | 27264/65536 [4:40:30<6:25:14,  1.66it/s]
 42%|████▏     | 27265/65536 [4:40:31<6:33:28,  1.62it/s]
 42%|████▏     | 27266/65536 [4:40:31<6:37:00,  1.61it/s]
 42%|████▏     | 27267/65536 [4:40:32<6:28:28,  1.64it/s]
 42%|████▏     | 27268/65536 [4:40:32<6:20:49,  1.67it/s]
 42%|████▏     | 27269/65536 [4:40:33<6:17:28,  1.69it/s]
 42%|████▏     | 27270/65536 [4:40:34<6:18:27,  1.69it/s]
 42%|████▏     | 27271/65536 [4:40:34<6:23:17,  1.66it/s]
 42%|████▏     | 27272/65536 [4:40:35<6:20:20,  1.68it/s]
 42%|████▏     | 27273/65536 [4:40:35<6:34:38,  1.62it/s]
 42%|████▏     | 27274/65536 [4:40:36<6:28:41,  1.64it/s]
 42%|████▏     | 27275/65536 [4:40:37<6:21:34,  1.67it/s]
 42%|████▏     | 27276/65536 [4:40:37<6:18:23,  1.69it/s]
 42%|████▏     | 27277/65536 [4:40:38<6:10:35,  1.72it/s]
 42%|████▏     | 27278/65536 [4:40:38<6:25:54,  1.65it/s]
 42%|████▏     | 27279/65536 [4:40:39<6:27:27,  1.65it/s]
 42%|████▏     | 27280/65536 [4:40:40<6:21:52,  1.67it/s]
                                                         
{'loss': 1.7524, 'learning_rate': 6.269744092077874e-07, 'epoch': 1683.95}
+
 42%|████▏     | 27280/65536 [4:40:40<6:21:52,  1.67it/s]
 42%|████▏     | 27281/65536 [4:40:40<6:51:55,  1.55it/s]
 42%|████▏     | 27282/65536 [4:40:41<6:44:37,  1.58it/s]
 42%|████▏     | 27283/65536 [4:40:42<6:37:34,  1.60it/s]
 42%|████▏     | 27284/65536 [4:40:42<6:40:18,  1.59it/s]
 42%|████▏     | 27285/65536 [4:40:43<6:37:32,  1.60it/s]
 42%|████▏     | 27286/65536 [4:40:43<6:39:14,  1.60it/s]
 42%|████▏     | 27287/65536 [4:40:44<6:30:16,  1.63it/s]
 42%|████▏     | 27288/65536 [4:40:45<6:31:59,  1.63it/s]
 42%|████▏     | 27289/65536 [4:40:45<6:24:25,  1.66it/s]
 42%|████▏     | 27290/65536 [4:40:46<6:39:35,  1.60it/s]
 42%|████▏     | 27291/65536 [4:40:47<6:42:20,  1.58it/s]
 42%|████▏     | 27292/65536 [4:40:47<6:30:14,  1.63it/s]
 42%|████▏     | 27293/65536 [4:40:48<6:32:08,  1.63it/s]
 42%|████▏     | 27294/65536 [4:40:48<6:30:15,  1.63it/s]
 42%|████▏     | 27295/65536 [4:40:49<6:23:00,  1.66it/s]
 42%|████▏     | 27296/65536 [4:40:50<6:22:11,  1.67it/s]
 42%|████▏     | 27297/65536 [4:40:50<6:20:45,  1.67it/s]
 42%|████▏     | 27298/65536 [4:40:51<6:34:22,  1.62it/s]
 42%|████▏     | 27299/65536 [4:40:51<6:38:33,  1.60it/s]
 42%|████▏     | 27300/65536 [4:40:52<6:27:16,  1.65it/s]
                                                         
{'loss': 1.7512, 'learning_rate': 6.266989102485612e-07, 'epoch': 1685.19}
+
 42%|████▏     | 27300/65536 [4:40:52<6:27:16,  1.65it/s]
 42%|████▏     | 27301/65536 [4:40:53<6:35:47,  1.61it/s]
 42%|████▏     | 27302/65536 [4:40:53<6:28:02,  1.64it/s]
 42%|████▏     | 27303/65536 [4:40:54<6:28:22,  1.64it/s]
 42%|████▏     | 27304/65536 [4:40:54<6:22:25,  1.67it/s]
 42%|████▏     | 27305/65536 [4:40:55<6:14:19,  1.70it/s]
 42%|████▏     | 27306/65536 [4:40:56<6:21:58,  1.67it/s]
 42%|████▏     | 27307/65536 [4:40:56<6:21:49,  1.67it/s]
 42%|████▏     | 27308/65536 [4:40:57<6:26:01,  1.65it/s]
 42%|████▏     | 27309/65536 [4:40:57<6:24:15,  1.66it/s]
 42%|████▏     | 27310/65536 [4:40:58<6:27:52,  1.64it/s]
 42%|████▏     | 27311/65536 [4:40:59<6:39:22,  1.60it/s]
 42%|████▏     | 27312/65536 [4:40:59<6:30:10,  1.63it/s]
 42%|████▏     | 27313/65536 [4:41:00<6:29:46,  1.63it/s]
 42%|████▏     | 27314/65536 [4:41:01<6:37:10,  1.60it/s]
 42%|████▏     | 27315/65536 [4:41:01<6:33:12,  1.62it/s]
 42%|████▏     | 27316/65536 [4:41:02<6:24:26,  1.66it/s]
 42%|████▏     | 27317/65536 [4:41:02<6:31:21,  1.63it/s]
 42%|████▏     | 27318/65536 [4:41:03<6:24:15,  1.66it/s]
 42%|████▏     | 27319/65536 [4:41:04<6:21:44,  1.67it/s]
 42%|████▏     | 27320/65536 [4:41:04<6:26:16,  1.65it/s]
                                                         
{'loss': 1.7376, 'learning_rate': 6.264234112893351e-07, 'epoch': 1686.42}
+
 42%|████▏     | 27320/65536 [4:41:04<6:26:16,  1.65it/s]
 42%|████▏     | 27321/65536 [4:41:05<6:32:12,  1.62it/s]
 42%|████▏     | 27322/65536 [4:41:05<6:21:53,  1.67it/s]
 42%|████▏     | 27323/65536 [4:41:06<6:33:33,  1.62it/s]
 42%|████▏     | 27324/65536 [4:41:07<6:35:34,  1.61it/s]
 42%|████▏     | 27325/65536 [4:41:07<6:35:20,  1.61it/s]
 42%|████▏     | 27326/65536 [4:41:08<6:38:07,  1.60it/s]
 42%|████▏     | 27327/65536 [4:41:09<6:44:42,  1.57it/s]
 42%|████▏     | 27328/65536 [4:41:09<6:32:18,  1.62it/s]
 42%|████▏     | 27329/65536 [4:41:10<6:26:05,  1.65it/s]
 42%|████▏     | 27330/65536 [4:41:10<6:48:56,  1.56it/s]
 42%|████▏     | 27331/65536 [4:41:11<6:41:16,  1.59it/s]
 42%|████▏     | 27332/65536 [4:41:12<6:31:10,  1.63it/s]
 42%|████▏     | 27333/65536 [4:41:12<6:29:48,  1.63it/s]
 42%|████▏     | 27334/65536 [4:41:13<6:27:01,  1.65it/s]
 42%|████▏     | 27335/65536 [4:41:13<6:23:49,  1.66it/s]
 42%|████▏     | 27336/65536 [4:41:14<6:34:09,  1.62it/s]
 42%|████▏     | 27337/65536 [4:41:15<6:35:21,  1.61it/s]
 42%|████▏     | 27338/65536 [4:41:15<6:31:07,  1.63it/s]
 42%|████▏     | 27339/65536 [4:41:16<6:30:21,  1.63it/s]
 42%|████▏     | 27340/65536 [4:41:16<6:20:42,  1.67it/s]
                                                         
{'loss': 1.7337, 'learning_rate': 6.261479123301089e-07, 'epoch': 1687.65}
+
 42%|████▏     | 27340/65536 [4:41:16<6:20:42,  1.67it/s]
 42%|████▏     | 27341/65536 [4:41:17<6:27:40,  1.64it/s]
 42%|████▏     | 27342/65536 [4:41:18<6:29:01,  1.64it/s]
 42%|████▏     | 27343/65536 [4:41:18<6:21:48,  1.67it/s]
 42%|████▏     | 27344/65536 [4:41:19<6:21:46,  1.67it/s]
 42%|████▏     | 27345/65536 [4:41:19<6:25:12,  1.65it/s]
 42%|████▏     | 27346/65536 [4:41:20<6:45:56,  1.57it/s]
 42%|████▏     | 27347/65536 [4:41:21<6:43:47,  1.58it/s]
 42%|████▏     | 27348/65536 [4:41:21<6:32:35,  1.62it/s]
 42%|████▏     | 27349/65536 [4:41:22<6:30:00,  1.63it/s]
 42%|████▏     | 27350/65536 [4:41:23<6:28:22,  1.64it/s]
 42%|████▏     | 27351/65536 [4:41:23<6:37:32,  1.60it/s]
 42%|████▏     | 27352/65536 [4:41:24<6:33:36,  1.62it/s]
 42%|████▏     | 27353/65536 [4:41:24<6:22:34,  1.66it/s]
 42%|████▏     | 27354/65536 [4:41:25<6:18:56,  1.68it/s]
 42%|████▏     | 27355/65536 [4:41:26<6:26:33,  1.65it/s]
 42%|████▏     | 27356/65536 [4:41:26<6:20:06,  1.67it/s]
 42%|████▏     | 27357/65536 [4:41:27<6:16:06,  1.69it/s]
 42%|████▏     | 27358/65536 [4:41:27<6:18:43,  1.68it/s]
 42%|████▏     | 27359/65536 [4:41:28<6:27:11,  1.64it/s]
 42%|████▏     | 27360/65536 [4:41:29<6:35:11,  1.61it/s]
                                                         
{'loss': 1.7469, 'learning_rate': 6.258724133708828e-07, 'epoch': 1688.89}
+
 42%|████▏     | 27360/65536 [4:41:29<6:35:11,  1.61it/s]
 42%|████▏     | 27361/65536 [4:41:29<6:42:16,  1.58it/s]
 42%|████▏     | 27362/65536 [4:41:30<8:13:07,  1.29it/s]
 42%|████▏     | 27363/65536 [4:41:31<7:33:24,  1.40it/s]
 42%|████▏     | 27364/65536 [4:41:32<7:16:58,  1.46it/s]
 42%|████▏     | 27365/65536 [4:41:32<7:00:28,  1.51it/s]
 42%|████▏     | 27366/65536 [4:41:33<6:50:41,  1.55it/s]
 42%|████▏     | 27367/65536 [4:41:33<6:44:59,  1.57it/s]
 42%|████▏     | 27368/65536 [4:41:34<6:41:29,  1.58it/s]
 42%|████▏     | 27369/65536 [4:41:35<6:40:04,  1.59it/s]
 42%|████▏     | 27370/65536 [4:41:35<6:46:56,  1.56it/s]
 42%|████▏     | 27371/65536 [4:41:36<6:40:14,  1.59it/s]
 42%|████▏     | 27372/65536 [4:41:37<6:35:26,  1.61it/s]
 42%|████▏     | 27373/65536 [4:41:37<6:35:43,  1.61it/s]
 42%|████▏     | 27374/65536 [4:41:38<6:27:21,  1.64it/s]
 42%|████▏     | 27375/65536 [4:41:38<6:22:22,  1.66it/s]
 42%|████▏     | 27376/65536 [4:41:39<6:22:30,  1.66it/s]
 42%|████▏     | 27377/65536 [4:41:40<6:24:23,  1.65it/s]
 42%|████▏     | 27378/65536 [4:41:40<6:34:18,  1.61it/s]
 42%|████▏     | 27379/65536 [4:41:41<6:43:44,  1.58it/s]
 42%|████▏     | 27380/65536 [4:41:42<6:41:29,  1.58it/s]
                                                         
{'loss': 1.7249, 'learning_rate': 6.255969144116567e-07, 'epoch': 1690.12}
+
 42%|████▏     | 27380/65536 [4:41:42<6:41:29,  1.58it/s]
 42%|████▏     | 27381/65536 [4:41:42<6:37:37,  1.60it/s]
 42%|████▏     | 27382/65536 [4:41:43<6:48:00,  1.56it/s]
 42%|████▏     | 27383/65536 [4:41:43<6:38:54,  1.59it/s]
 42%|████▏     | 27384/65536 [4:41:44<6:31:51,  1.62it/s]
 42%|████▏     | 27385/65536 [4:41:45<6:30:54,  1.63it/s]
 42%|████▏     | 27386/65536 [4:41:45<6:36:28,  1.60it/s]
 42%|████▏     | 27387/65536 [4:41:46<6:38:01,  1.60it/s]
 42%|████▏     | 27388/65536 [4:41:47<6:37:13,  1.60it/s]
 42%|████▏     | 27389/65536 [4:41:47<6:30:41,  1.63it/s]
 42%|████▏     | 27390/65536 [4:41:48<6:35:42,  1.61it/s]
 42%|████▏     | 27391/65536 [4:41:48<6:47:34,  1.56it/s]
 42%|████▏     | 27392/65536 [4:41:49<6:32:19,  1.62it/s]
 42%|████▏     | 27393/65536 [4:41:50<6:37:32,  1.60it/s]
 42%|████▏     | 27394/65536 [4:41:50<6:27:57,  1.64it/s]
 42%|████▏     | 27395/65536 [4:41:51<6:37:47,  1.60it/s]
 42%|████▏     | 27396/65536 [4:41:51<6:26:01,  1.65it/s]
 42%|████▏     | 27397/65536 [4:41:52<6:26:34,  1.64it/s]
 42%|████▏     | 27398/65536 [4:41:53<6:19:56,  1.67it/s]
 42%|████▏     | 27399/65536 [4:41:53<6:20:01,  1.67it/s]
 42%|████▏     | 27400/65536 [4:41:54<6:24:31,  1.65it/s]
                                                         
{'loss': 1.728, 'learning_rate': 6.253214154524305e-07, 'epoch': 1691.36}
+
 42%|████▏     | 27400/65536 [4:41:54<6:24:31,  1.65it/s]
 42%|████▏     | 27401/65536 [4:41:54<6:23:35,  1.66it/s]
 42%|████▏     | 27402/65536 [4:41:55<6:26:08,  1.65it/s]
 42%|████▏     | 27403/65536 [4:41:56<6:24:52,  1.65it/s]
 42%|████▏     | 27404/65536 [4:41:56<6:29:06,  1.63it/s]
 42%|████▏     | 27405/65536 [4:41:57<6:37:32,  1.60it/s]
 42%|████▏     | 27406/65536 [4:41:58<6:32:38,  1.62it/s]
 42%|████▏     | 27407/65536 [4:41:58<6:41:24,  1.58it/s]
 42%|████▏     | 27408/65536 [4:41:59<6:35:52,  1.61it/s]
 42%|████▏     | 27409/65536 [4:41:59<6:34:07,  1.61it/s]
 42%|████▏     | 27410/65536 [4:42:00<6:31:31,  1.62it/s]
 42%|████▏     | 27411/65536 [4:42:01<6:45:57,  1.57it/s]
 42%|████▏     | 27412/65536 [4:42:01<6:54:07,  1.53it/s]
 42%|████▏     | 27413/65536 [4:42:02<6:49:57,  1.55it/s]
 42%|████▏     | 27414/65536 [4:42:03<6:33:44,  1.61it/s]
 42%|████▏     | 27415/65536 [4:42:03<6:32:20,  1.62it/s]
 42%|████▏     | 27416/65536 [4:42:04<6:23:34,  1.66it/s]
 42%|████▏     | 27417/65536 [4:42:04<6:24:02,  1.65it/s]
 42%|████▏     | 27418/65536 [4:42:05<6:19:51,  1.67it/s]
 42%|████▏     | 27419/65536 [4:42:06<6:15:56,  1.69it/s]
 42%|████▏     | 27420/65536 [4:42:06<6:17:55,  1.68it/s]
                                                         
{'loss': 1.7316, 'learning_rate': 6.250459164932044e-07, 'epoch': 1692.59}
+
 42%|████▏     | 27420/65536 [4:42:06<6:17:55,  1.68it/s]
 42%|████▏     | 27421/65536 [4:42:07<6:17:12,  1.68it/s]
 42%|████▏     | 27422/65536 [4:42:07<6:16:38,  1.69it/s]
 42%|████▏     | 27423/65536 [4:42:08<6:29:58,  1.63it/s]
 42%|████▏     | 27424/65536 [4:42:09<6:27:48,  1.64it/s]
 42%|████▏     | 27425/65536 [4:42:09<6:28:49,  1.63it/s]
 42%|████▏     | 27426/65536 [4:42:10<6:34:55,  1.61it/s]
 42%|████▏     | 27427/65536 [4:42:11<6:40:58,  1.58it/s]
 42%|████▏     | 27428/65536 [4:42:11<6:32:09,  1.62it/s]
 42%|████▏     | 27429/65536 [4:42:12<6:35:04,  1.61it/s]
 42%|████▏     | 27430/65536 [4:42:12<6:30:00,  1.63it/s]
 42%|████▏     | 27431/65536 [4:42:13<6:30:29,  1.63it/s]
 42%|████▏     | 27432/65536 [4:42:14<6:30:20,  1.63it/s]
 42%|████▏     | 27433/65536 [4:42:14<6:33:16,  1.61it/s]
 42%|████▏     | 27434/65536 [4:42:15<6:26:24,  1.64it/s]
 42%|████▏     | 27435/65536 [4:42:15<6:21:56,  1.66it/s]
 42%|████▏     | 27436/65536 [4:42:16<6:41:14,  1.58it/s]
 42%|████▏     | 27437/65536 [4:42:17<6:31:01,  1.62it/s]
 42%|████▏     | 27438/65536 [4:42:17<6:35:42,  1.60it/s]
 42%|████▏     | 27439/65536 [4:42:18<6:31:23,  1.62it/s]
 42%|████▏     | 27440/65536 [4:42:18<6:26:20,  1.64it/s]
                                                         
{'loss': 1.7038, 'learning_rate': 6.247704175339782e-07, 'epoch': 1693.83}
+
 42%|████▏     | 27440/65536 [4:42:18<6:26:20,  1.64it/s]
 42%|████▏     | 27441/65536 [4:42:19<6:26:27,  1.64it/s]
 42%|████▏     | 27442/65536 [4:42:20<6:43:40,  1.57it/s]
 42%|████▏     | 27443/65536 [4:42:21<7:09:42,  1.48it/s]
 42%|████▏     | 27444/65536 [4:42:21<7:01:25,  1.51it/s]
 42%|████▏     | 27445/65536 [4:42:22<6:48:24,  1.55it/s]
 42%|████▏     | 27446/65536 [4:42:22<6:38:26,  1.59it/s]
 42%|████▏     | 27447/65536 [4:42:23<6:42:38,  1.58it/s]
 42%|████▏     | 27448/65536 [4:42:24<6:41:30,  1.58it/s]
 42%|████▏     | 27449/65536 [4:42:24<6:31:03,  1.62it/s]
 42%|████▏     | 27450/65536 [4:42:25<6:27:19,  1.64it/s]
 42%|████▏     | 27451/65536 [4:42:25<6:28:12,  1.64it/s]
 42%|████▏     | 27452/65536 [4:42:26<6:33:32,  1.61it/s]
 42%|████▏     | 27453/65536 [4:42:27<6:27:41,  1.64it/s]
 42%|████▏     | 27454/65536 [4:42:27<6:23:03,  1.66it/s]
 42%|████▏     | 27455/65536 [4:42:28<6:24:17,  1.65it/s]
 42%|████▏     | 27456/65536 [4:42:28<6:26:10,  1.64it/s]
 42%|████▏     | 27457/65536 [4:42:29<6:30:13,  1.63it/s]
 42%|████▏     | 27458/65536 [4:42:30<6:28:02,  1.64it/s]
 42%|████▏     | 27459/65536 [4:42:30<6:26:35,  1.64it/s]
 42%|████▏     | 27460/65536 [4:42:31<6:37:16,  1.60it/s]
                                                         
{'loss': 1.7378, 'learning_rate': 6.244949185747521e-07, 'epoch': 1695.06}
+
 42%|████▏     | 27460/65536 [4:42:31<6:37:16,  1.60it/s]
 42%|████▏     | 27461/65536 [4:42:32<6:31:29,  1.62it/s]
 42%|████▏     | 27462/65536 [4:42:32<6:29:51,  1.63it/s]
 42%|████▏     | 27463/65536 [4:42:33<6:37:33,  1.60it/s]
 42%|████▏     | 27464/65536 [4:42:33<6:38:56,  1.59it/s]
 42%|████▏     | 27465/65536 [4:42:34<6:34:21,  1.61it/s]
 42%|████▏     | 27466/65536 [4:42:35<6:27:00,  1.64it/s]
 42%|████▏     | 27467/65536 [4:42:35<6:28:37,  1.63it/s]
 42%|████▏     | 27468/65536 [4:42:36<6:31:54,  1.62it/s]
 42%|████▏     | 27469/65536 [4:42:37<6:26:58,  1.64it/s]
 42%|████▏     | 27470/65536 [4:42:37<6:31:59,  1.62it/s]
 42%|████▏     | 27471/65536 [4:42:38<6:29:57,  1.63it/s]
 42%|████▏     | 27472/65536 [4:42:38<6:25:26,  1.65it/s]
 42%|████▏     | 27473/65536 [4:42:39<6:25:14,  1.65it/s]
 42%|████▏     | 27474/65536 [4:42:40<6:27:11,  1.64it/s]
 42%|████▏     | 27475/65536 [4:42:40<6:20:23,  1.67it/s]
 42%|████▏     | 27476/65536 [4:42:41<6:31:46,  1.62it/s]
 42%|████▏     | 27477/65536 [4:42:41<6:29:52,  1.63it/s]
 42%|████▏     | 27478/65536 [4:42:42<6:31:11,  1.62it/s]
 42%|████▏     | 27479/65536 [4:42:43<6:25:17,  1.65it/s]
 42%|████▏     | 27480/65536 [4:42:43<6:22:43,  1.66it/s]
                                                         
{'loss': 1.7121, 'learning_rate': 6.24219419615526e-07, 'epoch': 1696.3}
+
 42%|████▏     | 27480/65536 [4:42:43<6:22:43,  1.66it/s]
 42%|████▏     | 27481/65536 [4:42:44<6:29:29,  1.63it/s]
 42%|████▏     | 27482/65536 [4:42:44<6:26:12,  1.64it/s]
 42%|████▏     | 27483/65536 [4:42:45<6:30:31,  1.62it/s]
 42%|████▏     | 27484/65536 [4:42:46<6:23:45,  1.65it/s]
 42%|████▏     | 27485/65536 [4:42:46<6:20:30,  1.67it/s]
 42%|████▏     | 27486/65536 [4:42:47<6:25:59,  1.64it/s]
 42%|████▏     | 27487/65536 [4:42:47<6:20:34,  1.67it/s]
 42%|████▏     | 27488/65536 [4:42:48<6:26:13,  1.64it/s]
 42%|████▏     | 27489/65536 [4:42:49<6:23:35,  1.65it/s]
 42%|████▏     | 27490/65536 [4:42:49<6:13:42,  1.70it/s]
 42%|████▏     | 27491/65536 [4:42:50<6:25:25,  1.65it/s]
 42%|████▏     | 27492/65536 [4:42:51<6:38:22,  1.59it/s]
 42%|████▏     | 27493/65536 [4:42:51<6:50:31,  1.54it/s]
 42%|████▏     | 27494/65536 [4:42:52<6:40:02,  1.58it/s]
 42%|████▏     | 27495/65536 [4:42:52<6:39:51,  1.59it/s]
 42%|████▏     | 27496/65536 [4:42:53<6:26:09,  1.64it/s]
 42%|████▏     | 27497/65536 [4:42:54<6:38:53,  1.59it/s]
 42%|████▏     | 27498/65536 [4:42:54<6:40:58,  1.58it/s]
 42%|████▏     | 27499/65536 [4:42:55<6:28:41,  1.63it/s]
 42%|████▏     | 27500/65536 [4:42:56<6:24:32,  1.65it/s]
                                                         
{'loss': 1.7308, 'learning_rate': 6.239439206562998e-07, 'epoch': 1697.53}
+
 42%|████▏     | 27500/65536 [4:42:56<6:24:32,  1.65it/s]
 42%|████▏     | 27501/65536 [4:42:56<6:36:39,  1.60it/s]
 42%|████▏     | 27502/65536 [4:42:57<6:31:19,  1.62it/s]
 42%|████▏     | 27503/65536 [4:42:57<6:23:35,  1.65it/s]
 42%|████▏     | 27504/65536 [4:42:58<6:24:30,  1.65it/s]
 42%|████▏     | 27505/65536 [4:42:59<6:23:29,  1.65it/s]
 42%|████▏     | 27506/65536 [4:42:59<6:24:06,  1.65it/s]
 42%|████▏     | 27507/65536 [4:43:00<6:20:31,  1.67it/s]
 42%|████▏     | 27508/65536 [4:43:00<6:28:23,  1.63it/s]
 42%|████▏     | 27509/65536 [4:43:01<6:43:26,  1.57it/s]
 42%|████▏     | 27510/65536 [4:43:02<6:36:54,  1.60it/s]
 42%|████▏     | 27511/65536 [4:43:02<6:31:17,  1.62it/s]
 42%|████▏     | 27512/65536 [4:43:03<6:40:38,  1.58it/s]
 42%|████▏     | 27513/65536 [4:43:04<6:36:23,  1.60it/s]
 42%|████▏     | 27514/65536 [4:43:04<6:35:20,  1.60it/s]
 42%|████▏     | 27515/65536 [4:43:05<6:29:18,  1.63it/s]
 42%|████▏     | 27516/65536 [4:43:05<6:31:19,  1.62it/s]
 42%|████▏     | 27517/65536 [4:43:06<6:29:14,  1.63it/s]
 42%|████▏     | 27518/65536 [4:43:07<6:34:09,  1.61it/s]
 42%|████▏     | 27519/65536 [4:43:07<6:29:37,  1.63it/s]
 42%|████▏     | 27520/65536 [4:43:08<6:25:40,  1.64it/s]
                                                         
{'loss': 1.7108, 'learning_rate': 6.236684216970737e-07, 'epoch': 1698.77}
+
 42%|████▏     | 27520/65536 [4:43:08<6:25:40,  1.64it/s]
 42%|████▏     | 27521/65536 [4:43:08<6:20:13,  1.67it/s]
 42%|████▏     | 27522/65536 [4:43:09<6:10:22,  1.71it/s]
 42%|████▏     | 27523/65536 [4:43:10<6:12:12,  1.70it/s]
 42%|████▏     | 27524/65536 [4:43:10<6:36:44,  1.60it/s]
 42%|████▏     | 27525/65536 [4:43:11<6:29:48,  1.63it/s]
 42%|████▏     | 27526/65536 [4:43:11<6:29:51,  1.62it/s]
 42%|████▏     | 27527/65536 [4:43:12<6:31:07,  1.62it/s]
 42%|████▏     | 27528/65536 [4:43:13<6:28:36,  1.63it/s]
 42%|████▏     | 27529/65536 [4:43:13<6:25:36,  1.64it/s]
 42%|████▏     | 27530/65536 [4:43:14<6:26:57,  1.64it/s]
 42%|████▏     | 27531/65536 [4:43:15<6:24:15,  1.65it/s]
 42%|████▏     | 27532/65536 [4:43:15<6:26:06,  1.64it/s]
 42%|████▏     | 27533/65536 [4:43:16<6:29:59,  1.62it/s]
 42%|████▏     | 27534/65536 [4:43:16<6:35:38,  1.60it/s]
 42%|████▏     | 27535/65536 [4:43:17<6:26:54,  1.64it/s]
 42%|████▏     | 27536/65536 [4:43:18<6:22:48,  1.65it/s]
 42%|████▏     | 27537/65536 [4:43:18<6:26:20,  1.64it/s]
 42%|████▏     | 27538/65536 [4:43:19<6:23:26,  1.65it/s]
 42%|████▏     | 27539/65536 [4:43:19<6:15:35,  1.69it/s]
 42%|████▏     | 27540/65536 [4:43:20<6:21:08,  1.66it/s]
                                                         
{'loss': 1.7702, 'learning_rate': 6.233929227378474e-07, 'epoch': 1700.0}
+
 42%|████▏     | 27540/65536 [4:43:20<6:21:08,  1.66it/s]
 42%|████▏     | 27541/65536 [4:43:21<6:33:29,  1.61it/s]
 42%|████▏     | 27542/65536 [4:43:21<6:26:07,  1.64it/s]
 42%|████▏     | 27543/65536 [4:43:22<6:21:06,  1.66it/s]
 42%|████▏     | 27544/65536 [4:43:22<6:23:11,  1.65it/s]
 42%|████▏     | 27545/65536 [4:43:23<6:21:03,  1.66it/s]
 42%|████▏     | 27546/65536 [4:43:24<6:22:29,  1.66it/s]
 42%|████▏     | 27547/65536 [4:43:24<6:23:05,  1.65it/s]
 42%|████▏     | 27548/65536 [4:43:25<6:22:16,  1.66it/s]
 42%|████▏     | 27549/65536 [4:43:26<6:33:54,  1.61it/s]
 42%|████▏     | 27550/65536 [4:43:26<6:35:32,  1.60it/s]
 42%|████▏     | 27551/65536 [4:43:27<6:35:40,  1.60it/s]
 42%|████▏     | 27552/65536 [4:43:27<6:44:03,  1.57it/s]
 42%|████▏     | 27553/65536 [4:43:28<6:46:15,  1.56it/s]
 42%|████▏     | 27554/65536 [4:43:29<6:44:27,  1.57it/s]
 42%|████▏     | 27555/65536 [4:43:29<6:40:50,  1.58it/s]
 42%|████▏     | 27556/65536 [4:43:30<6:32:08,  1.61it/s]
 42%|████▏     | 27557/65536 [4:43:31<6:42:59,  1.57it/s]
 42%|████▏     | 27558/65536 [4:43:31<6:35:20,  1.60it/s]
 42%|████▏     | 27559/65536 [4:43:32<6:35:07,  1.60it/s]
 42%|████▏     | 27560/65536 [4:43:32<6:36:36,  1.60it/s]
                                                         
{'loss': 1.7137, 'learning_rate': 6.231174237786213e-07, 'epoch': 1701.23}
+
 42%|████▏     | 27560/65536 [4:43:32<6:36:36,  1.60it/s]
 42%|████▏     | 27561/65536 [4:43:33<6:35:44,  1.60it/s]
 42%|████▏     | 27562/65536 [4:43:34<6:36:07,  1.60it/s]
 42%|████▏     | 27563/65536 [4:43:34<6:31:16,  1.62it/s]
 42%|████▏     | 27564/65536 [4:43:35<6:29:46,  1.62it/s]
 42%|████▏     | 27565/65536 [4:43:36<6:30:07,  1.62it/s]
 42%|████▏     | 27566/65536 [4:43:36<6:36:09,  1.60it/s]
 42%|████▏     | 27567/65536 [4:43:37<6:25:31,  1.64it/s]
 42%|████▏     | 27568/65536 [4:43:37<6:22:59,  1.65it/s]
 42%|████▏     | 27569/65536 [4:43:38<6:25:14,  1.64it/s]
 42%|████▏     | 27570/65536 [4:43:39<6:32:54,  1.61it/s]
 42%|████▏     | 27571/65536 [4:43:39<6:26:27,  1.64it/s]
 42%|████▏     | 27572/65536 [4:43:40<6:28:19,  1.63it/s]
 42%|████▏     | 27573/65536 [4:43:41<6:50:29,  1.54it/s]
 42%|████▏     | 27574/65536 [4:43:41<6:43:58,  1.57it/s]
 42%|████▏     | 27575/65536 [4:43:42<6:39:13,  1.58it/s]
 42%|████▏     | 27576/65536 [4:43:42<6:33:00,  1.61it/s]
 42%|████▏     | 27577/65536 [4:43:43<6:33:54,  1.61it/s]
 42%|████▏     | 27578/65536 [4:43:44<6:23:55,  1.65it/s]
 42%|████▏     | 27579/65536 [4:43:44<6:19:20,  1.67it/s]
 42%|████▏     | 27580/65536 [4:43:45<6:05:49,  1.73it/s]
                                                         
{'loss': 1.7127, 'learning_rate': 6.228419248193951e-07, 'epoch': 1702.47}
+
 42%|████▏     | 27580/65536 [4:43:45<6:05:49,  1.73it/s]
 42%|████▏     | 27581/65536 [4:43:45<6:08:16,  1.72it/s]
 42%|████▏     | 27582/65536 [4:43:46<6:15:28,  1.68it/s]
 42%|████▏     | 27583/65536 [4:43:47<6:14:58,  1.69it/s]
 42%|████▏     | 27584/65536 [4:43:47<6:23:54,  1.65it/s]
 42%|████▏     | 27585/65536 [4:43:48<6:27:14,  1.63it/s]
 42%|████▏     | 27586/65536 [4:43:48<6:18:53,  1.67it/s]
 42%|████▏     | 27587/65536 [4:43:49<6:23:27,  1.65it/s]
 42%|████▏     | 27588/65536 [4:43:50<6:40:15,  1.58it/s]
 42%|████▏     | 27589/65536 [4:43:50<6:48:00,  1.55it/s]
 42%|████▏     | 27590/65536 [4:43:51<6:44:55,  1.56it/s]
 42%|████▏     | 27591/65536 [4:43:52<6:36:57,  1.59it/s]
 42%|████▏     | 27592/65536 [4:43:52<6:28:36,  1.63it/s]
 42%|████▏     | 27593/65536 [4:43:53<6:19:44,  1.67it/s]
 42%|████▏     | 27594/65536 [4:43:53<6:20:57,  1.66it/s]
 42%|████▏     | 27595/65536 [4:43:54<6:28:52,  1.63it/s]
 42%|████▏     | 27596/65536 [4:43:55<6:33:44,  1.61it/s]
 42%|████▏     | 27597/65536 [4:43:55<6:26:08,  1.64it/s]
 42%|████▏     | 27598/65536 [4:43:56<6:41:22,  1.58it/s]
 42%|████▏     | 27599/65536 [4:43:56<6:33:25,  1.61it/s]
 42%|████▏     | 27600/65536 [4:43:57<6:28:36,  1.63it/s]
                                                         
{'loss': 1.7142, 'learning_rate': 6.225664258601689e-07, 'epoch': 1703.7}
+
 42%|████▏     | 27600/65536 [4:43:57<6:28:36,  1.63it/s]
 42%|████▏     | 27601/65536 [4:43:58<6:30:11,  1.62it/s]
 42%|████▏     | 27602/65536 [4:43:58<6:30:57,  1.62it/s]
 42%|████▏     | 27603/65536 [4:43:59<6:30:36,  1.62it/s]
 42%|████▏     | 27604/65536 [4:44:00<6:28:31,  1.63it/s]
 42%|████▏     | 27605/65536 [4:44:00<6:43:13,  1.57it/s]
 42%|████▏     | 27606/65536 [4:44:01<6:39:33,  1.58it/s]
 42%|████▏     | 27607/65536 [4:44:01<6:29:24,  1.62it/s]
 42%|████▏     | 27608/65536 [4:44:02<6:22:05,  1.65it/s]
 42%|████▏     | 27609/65536 [4:44:03<6:22:36,  1.65it/s]
 42%|████▏     | 27610/65536 [4:44:03<6:20:08,  1.66it/s]
 42%|████▏     | 27611/65536 [4:44:04<6:25:44,  1.64it/s]
 42%|████▏     | 27612/65536 [4:44:04<6:20:40,  1.66it/s]
 42%|████▏     | 27613/65536 [4:44:05<6:24:38,  1.64it/s]
 42%|████▏     | 27614/65536 [4:44:06<6:33:23,  1.61it/s]
 42%|████▏     | 27615/65536 [4:44:06<6:37:22,  1.59it/s]
 42%|████▏     | 27616/65536 [4:44:07<6:46:49,  1.55it/s]
 42%|████▏     | 27617/65536 [4:44:08<6:29:28,  1.62it/s]
 42%|████▏     | 27618/65536 [4:44:08<6:15:33,  1.68it/s]
 42%|████▏     | 27619/65536 [4:44:09<6:17:02,  1.68it/s]
 42%|████▏     | 27620/65536 [4:44:09<6:16:22,  1.68it/s]
                                                         
{'loss': 1.771, 'learning_rate': 6.222909269009427e-07, 'epoch': 1704.94}
+
 42%|████▏     | 27620/65536 [4:44:09<6:16:22,  1.68it/s]
 42%|████▏     | 27621/65536 [4:44:10<6:21:18,  1.66it/s]
 42%|████▏     | 27622/65536 [4:44:11<6:37:33,  1.59it/s]
 42%|████▏     | 27623/65536 [4:44:11<6:29:04,  1.62it/s]
 42%|████▏     | 27624/65536 [4:44:12<6:26:29,  1.63it/s]
 42%|████▏     | 27625/65536 [4:44:12<6:35:44,  1.60it/s]
 42%|████▏     | 27626/65536 [4:44:13<6:27:33,  1.63it/s]
 42%|████▏     | 27627/65536 [4:44:14<6:25:38,  1.64it/s]
 42%|████▏     | 27628/65536 [4:44:14<6:19:59,  1.66it/s]
 42%|████▏     | 27629/65536 [4:44:15<6:18:03,  1.67it/s]
 42%|████▏     | 27630/65536 [4:44:15<6:25:20,  1.64it/s]
 42%|████▏     | 27631/65536 [4:44:16<6:26:10,  1.64it/s]
 42%|████▏     | 27632/65536 [4:44:17<6:24:01,  1.65it/s]
 42%|████▏     | 27633/65536 [4:44:17<6:26:28,  1.63it/s]
 42%|████▏     | 27634/65536 [4:44:18<6:28:37,  1.63it/s]
 42%|████▏     | 27635/65536 [4:44:19<6:32:00,  1.61it/s]
 42%|████▏     | 27636/65536 [4:44:19<6:27:36,  1.63it/s]
 42%|████▏     | 27637/65536 [4:44:20<6:33:54,  1.60it/s]
 42%|████▏     | 27638/65536 [4:44:20<6:45:06,  1.56it/s]
 42%|████▏     | 27639/65536 [4:44:21<6:46:17,  1.55it/s]
 42%|████▏     | 27640/65536 [4:44:22<6:51:13,  1.54it/s]
                                                         
{'loss': 1.7219, 'learning_rate': 6.220154279417166e-07, 'epoch': 1706.17}
+
 42%|████▏     | 27640/65536 [4:44:22<6:51:13,  1.54it/s]
 42%|████▏     | 27641/65536 [4:44:22<6:46:44,  1.55it/s]
 42%|████▏     | 27642/65536 [4:44:23<6:28:45,  1.62it/s]
 42%|████▏     | 27643/65536 [4:44:24<6:20:44,  1.66it/s]
 42%|████▏     | 27644/65536 [4:44:24<6:19:03,  1.67it/s]
 42%|████▏     | 27645/65536 [4:44:25<6:24:33,  1.64it/s]
 42%|████▏     | 27646/65536 [4:44:25<6:31:19,  1.61it/s]
 42%|████▏     | 27647/65536 [4:44:26<6:32:33,  1.61it/s]
 42%|████▏     | 27648/65536 [4:44:27<6:55:30,  1.52it/s]
 42%|████▏     | 27649/65536 [4:44:27<6:58:51,  1.51it/s]
 42%|████▏     | 27650/65536 [4:44:28<6:46:06,  1.55it/s]
 42%|████▏     | 27651/65536 [4:44:29<6:38:05,  1.59it/s]
 42%|████▏     | 27652/65536 [4:44:29<6:31:38,  1.61it/s]
 42%|████▏     | 27653/65536 [4:44:30<6:23:51,  1.64it/s]
 42%|████▏     | 27654/65536 [4:44:31<6:39:43,  1.58it/s]
 42%|████▏     | 27655/65536 [4:44:31<6:46:59,  1.55it/s]
 42%|████▏     | 27656/65536 [4:44:32<6:39:44,  1.58it/s]
 42%|████▏     | 27657/65536 [4:44:32<6:37:43,  1.59it/s]
 42%|████▏     | 27658/65536 [4:44:33<6:33:55,  1.60it/s]
 42%|████▏     | 27659/65536 [4:44:34<6:29:35,  1.62it/s]
 42%|████▏     | 27660/65536 [4:44:34<6:22:16,  1.65it/s]
                                                         
{'loss': 1.7772, 'learning_rate': 6.217399289824905e-07, 'epoch': 1707.41}
+
 42%|████▏     | 27660/65536 [4:44:34<6:22:16,  1.65it/s]
 42%|████▏     | 27661/65536 [4:44:35<6:31:28,  1.61it/s]
 42%|████▏     | 27662/65536 [4:44:35<6:28:18,  1.63it/s]
 42%|████▏     | 27663/65536 [4:44:36<6:25:21,  1.64it/s]
 42%|████▏     | 27664/65536 [4:44:37<6:21:31,  1.65it/s]
 42%|████▏     | 27665/65536 [4:44:37<6:34:24,  1.60it/s]
 42%|████▏     | 27666/65536 [4:44:38<6:27:26,  1.63it/s]
 42%|████▏     | 27667/65536 [4:44:39<6:25:21,  1.64it/s]
 42%|████▏     | 27668/65536 [4:44:39<6:25:09,  1.64it/s]
 42%|████▏     | 27669/65536 [4:44:40<6:29:50,  1.62it/s]
 42%|████▏     | 27670/65536 [4:44:40<6:44:26,  1.56it/s]
 42%|████▏     | 27671/65536 [4:44:41<6:39:02,  1.58it/s]
 42%|████▏     | 27672/65536 [4:44:42<6:44:14,  1.56it/s]
 42%|████▏     | 27673/65536 [4:44:42<6:38:52,  1.58it/s]
 42%|████▏     | 27674/65536 [4:44:43<6:31:27,  1.61it/s]
 42%|████▏     | 27675/65536 [4:44:44<6:30:18,  1.62it/s]
 42%|████▏     | 27676/65536 [4:44:44<6:26:08,  1.63it/s]
 42%|████▏     | 27677/65536 [4:44:45<6:33:11,  1.60it/s]
 42%|████▏     | 27678/65536 [4:44:45<6:27:01,  1.63it/s]
 42%|████▏     | 27679/65536 [4:44:46<6:25:50,  1.64it/s]
 42%|████▏     | 27680/65536 [4:44:47<6:24:03,  1.64it/s]
                                                         
{'loss': 1.6827, 'learning_rate': 6.214644300232643e-07, 'epoch': 1708.64}
+
 42%|████▏     | 27680/65536 [4:44:47<6:24:03,  1.64it/s]
 42%|████▏     | 27681/65536 [4:44:47<6:28:39,  1.62it/s]
 42%|████▏     | 27682/65536 [4:44:48<6:26:48,  1.63it/s]
 42%|████▏     | 27683/65536 [4:44:48<6:26:49,  1.63it/s]
 42%|████▏     | 27684/65536 [4:44:49<6:34:06,  1.60it/s]
 42%|████▏     | 27685/65536 [4:44:50<6:25:49,  1.64it/s]
 42%|████▏     | 27686/65536 [4:44:50<6:32:49,  1.61it/s]
 42%|████▏     | 27687/65536 [4:44:51<6:39:37,  1.58it/s]
 42%|████▏     | 27688/65536 [4:44:52<6:34:34,  1.60it/s]
 42%|████▏     | 27689/65536 [4:44:52<6:50:08,  1.54it/s]
 42%|████▏     | 27690/65536 [4:44:53<6:40:16,  1.58it/s]
 42%|████▏     | 27691/65536 [4:44:54<6:35:46,  1.59it/s]
 42%|████▏     | 27692/65536 [4:44:54<6:26:10,  1.63it/s]
 42%|████▏     | 27693/65536 [4:44:55<6:27:47,  1.63it/s]
 42%|████▏     | 27694/65536 [4:44:55<6:25:25,  1.64it/s]
 42%|████▏     | 27695/65536 [4:44:56<6:31:18,  1.61it/s]
 42%|████▏     | 27696/65536 [4:44:57<6:20:54,  1.66it/s]
 42%|████▏     | 27697/65536 [4:44:57<6:18:09,  1.67it/s]
 42%|████▏     | 27698/65536 [4:44:58<6:25:10,  1.64it/s]
 42%|████▏     | 27699/65536 [4:44:58<6:22:45,  1.65it/s]
 42%|████▏     | 27700/65536 [4:44:59<6:32:56,  1.60it/s]
                                                         
{'loss': 1.7013, 'learning_rate': 6.211889310640382e-07, 'epoch': 1709.88}
+
 42%|████▏     | 27700/65536 [4:44:59<6:32:56,  1.60it/s]
 42%|████▏     | 27701/65536 [4:45:00<6:31:06,  1.61it/s]
 42%|████▏     | 27702/65536 [4:45:00<6:22:44,  1.65it/s]
 42%|████▏     | 27703/65536 [4:45:01<6:33:33,  1.60it/s]
 42%|████▏     | 27704/65536 [4:45:01<6:30:32,  1.61it/s]
 42%|████▏     | 27705/65536 [4:45:02<6:21:53,  1.65it/s]
 42%|████▏     | 27706/65536 [4:45:03<6:27:05,  1.63it/s]
 42%|████▏     | 27707/65536 [4:45:03<6:24:51,  1.64it/s]
 42%|████▏     | 27708/65536 [4:45:04<6:22:08,  1.65it/s]
 42%|████▏     | 27709/65536 [4:45:05<6:27:53,  1.63it/s]
 42%|████▏     | 27710/65536 [4:45:05<6:27:41,  1.63it/s]
 42%|████▏     | 27711/65536 [4:45:06<6:27:20,  1.63it/s]
 42%|████▏     | 27712/65536 [4:45:06<6:35:20,  1.59it/s]
 42%|████▏     | 27713/65536 [4:45:07<6:27:47,  1.63it/s]
 42%|████▏     | 27714/65536 [4:45:08<6:27:53,  1.63it/s]
 42%|████▏     | 27715/65536 [4:45:08<6:45:11,  1.56it/s]
 42%|████▏     | 27716/65536 [4:45:09<6:37:14,  1.59it/s]
 42%|████▏     | 27717/65536 [4:45:10<6:35:38,  1.59it/s]
 42%|████▏     | 27718/65536 [4:45:10<6:37:29,  1.59it/s]
 42%|████▏     | 27719/65536 [4:45:11<6:39:23,  1.58it/s]
 42%|████▏     | 27720/65536 [4:45:11<6:30:55,  1.61it/s]
                                                         
{'loss': 1.7148, 'learning_rate': 6.20913432104812e-07, 'epoch': 1711.11}
+
 42%|████▏     | 27720/65536 [4:45:11<6:30:55,  1.61it/s]
 42%|████▏     | 27721/65536 [4:45:12<6:29:35,  1.62it/s]
 42%|████▏     | 27722/65536 [4:45:13<6:33:46,  1.60it/s]
 42%|████▏     | 27723/65536 [4:45:13<6:27:31,  1.63it/s]
 42%|████▏     | 27724/65536 [4:45:14<6:29:40,  1.62it/s]
 42%|████▏     | 27725/65536 [4:45:15<6:37:58,  1.58it/s]
 42%|████▏     | 27726/65536 [4:45:15<6:40:26,  1.57it/s]
 42%|████▏     | 27727/65536 [4:45:16<6:45:43,  1.55it/s]
 42%|████▏     | 27728/65536 [4:45:16<6:31:55,  1.61it/s]
 42%|████▏     | 27729/65536 [4:45:17<6:19:05,  1.66it/s]
 42%|████▏     | 27730/65536 [4:45:18<6:17:27,  1.67it/s]
 42%|████▏     | 27731/65536 [4:45:18<6:17:27,  1.67it/s]
 42%|████▏     | 27732/65536 [4:45:19<6:19:40,  1.66it/s]
 42%|████▏     | 27733/65536 [4:45:19<6:12:31,  1.69it/s]
 42%|████▏     | 27734/65536 [4:45:20<6:21:55,  1.65it/s]
 42%|████▏     | 27735/65536 [4:45:21<6:39:30,  1.58it/s]
 42%|████▏     | 27736/65536 [4:45:21<6:32:12,  1.61it/s]
 42%|████▏     | 27737/65536 [4:45:22<6:30:14,  1.61it/s]
 42%|████▏     | 27738/65536 [4:45:22<6:20:20,  1.66it/s]
 42%|████▏     | 27739/65536 [4:45:23<6:20:45,  1.65it/s]
 42%|████▏     | 27740/65536 [4:45:24<6:19:33,  1.66it/s]
                                                         
{'loss': 1.7623, 'learning_rate': 6.206379331455859e-07, 'epoch': 1712.35}
+
 42%|████▏     | 27740/65536 [4:45:24<6:19:33,  1.66it/s]
 42%|████▏     | 27741/65536 [4:45:24<6:20:48,  1.65it/s]
 42%|████▏     | 27742/65536 [4:45:25<6:16:52,  1.67it/s]
 42%|████▏     | 27743/65536 [4:45:25<6:13:42,  1.69it/s]
 42%|████▏     | 27744/65536 [4:45:26<6:16:59,  1.67it/s]
 42%|████▏     | 27745/65536 [4:45:27<6:24:05,  1.64it/s]
 42%|████▏     | 27746/65536 [4:45:27<6:27:50,  1.62it/s]
 42%|████▏     | 27747/65536 [4:45:28<6:44:20,  1.56it/s]
 42%|████▏     | 27748/65536 [4:45:29<6:36:57,  1.59it/s]
 42%|████▏     | 27749/65536 [4:45:29<6:32:51,  1.60it/s]
 42%|████▏     | 27750/65536 [4:45:30<6:27:03,  1.63it/s]
 42%|████▏     | 27751/65536 [4:45:31<6:43:21,  1.56it/s]
 42%|████▏     | 27752/65536 [4:45:31<6:35:08,  1.59it/s]
 42%|████▏     | 27753/65536 [4:45:32<6:39:41,  1.58it/s]
 42%|████▏     | 27754/65536 [4:45:32<6:31:34,  1.61it/s]
 42%|████▏     | 27755/65536 [4:45:33<6:25:05,  1.64it/s]
 42%|████▏     | 27756/65536 [4:45:34<6:30:27,  1.61it/s]
 42%|████▏     | 27757/65536 [4:45:34<6:25:48,  1.63it/s]
 42%|████▏     | 27758/65536 [4:45:35<6:18:45,  1.66it/s]
 42%|████▏     | 27759/65536 [4:45:35<6:18:10,  1.66it/s]
 42%|████▏     | 27760/65536 [4:45:36<6:17:37,  1.67it/s]
                                                         
{'loss': 1.7249, 'learning_rate': 6.203624341863598e-07, 'epoch': 1713.58}
+
 42%|████▏     | 27760/65536 [4:45:36<6:17:37,  1.67it/s]
 42%|████▏     | 27761/65536 [4:45:37<6:22:11,  1.65it/s]
 42%|████▏     | 27762/65536 [4:45:37<6:22:34,  1.65it/s]
 42%|████▏     | 27763/65536 [4:45:38<6:21:51,  1.65it/s]
 42%|████▏     | 27764/65536 [4:45:38<6:26:28,  1.63it/s]
 42%|████▏     | 27765/65536 [4:45:39<6:35:27,  1.59it/s]
 42%|████▏     | 27766/65536 [4:45:40<6:39:31,  1.58it/s]
 42%|████▏     | 27767/65536 [4:45:40<6:46:22,  1.55it/s]
 42%|████▏     | 27768/65536 [4:45:41<6:38:26,  1.58it/s]
 42%|████▏     | 27769/65536 [4:45:42<6:31:17,  1.61it/s]
 42%|████▏     | 27770/65536 [4:45:42<6:30:06,  1.61it/s]
 42%|████▏     | 27771/65536 [4:45:43<6:33:17,  1.60it/s]
 42%|████▏     | 27772/65536 [4:45:43<6:33:06,  1.60it/s]
 42%|████▏     | 27773/65536 [4:45:44<6:34:23,  1.60it/s]
 42%|████▏     | 27774/65536 [4:45:45<6:24:33,  1.64it/s]
 42%|████▏     | 27775/65536 [4:45:45<6:23:59,  1.64it/s]
 42%|████▏     | 27776/65536 [4:45:46<6:27:10,  1.63it/s]
 42%|████▏     | 27777/65536 [4:45:47<6:28:31,  1.62it/s]
 42%|████▏     | 27778/65536 [4:45:47<6:23:06,  1.64it/s]
 42%|████▏     | 27779/65536 [4:45:48<6:16:13,  1.67it/s]
 42%|████▏     | 27780/65536 [4:45:48<6:13:46,  1.68it/s]
                                                         
{'loss': 1.7041, 'learning_rate': 6.200869352271336e-07, 'epoch': 1714.81}
+
 42%|████▏     | 27780/65536 [4:45:48<6:13:46,  1.68it/s]
 42%|████▏     | 27781/65536 [4:45:49<6:27:20,  1.62it/s]
 42%|████▏     | 27782/65536 [4:45:50<6:25:13,  1.63it/s]
 42%|████▏     | 27783/65536 [4:45:50<6:27:07,  1.63it/s]
 42%|████▏     | 27784/65536 [4:45:51<6:43:02,  1.56it/s]
 42%|████▏     | 27785/65536 [4:45:52<6:45:27,  1.55it/s]
 42%|████▏     | 27786/65536 [4:45:52<6:40:32,  1.57it/s]
 42%|████▏     | 27787/65536 [4:45:53<6:27:43,  1.62it/s]
 42%|████▏     | 27788/65536 [4:45:53<6:24:24,  1.64it/s]
 42%|████▏     | 27789/65536 [4:45:54<6:22:08,  1.65it/s]
 42%|████▏     | 27790/65536 [4:45:54<6:18:23,  1.66it/s]
 42%|████▏     | 27791/65536 [4:45:55<6:24:03,  1.64it/s]
 42%|████▏     | 27792/65536 [4:45:56<6:27:33,  1.62it/s]
 42%|████▏     | 27793/65536 [4:45:56<6:19:07,  1.66it/s]
 42%|████▏     | 27794/65536 [4:45:57<6:12:34,  1.69it/s]
 42%|████▏     | 27795/65536 [4:45:57<6:14:49,  1.68it/s]
 42%|████▏     | 27796/65536 [4:45:58<6:31:48,  1.61it/s]
 42%|████▏     | 27797/65536 [4:45:59<6:33:32,  1.60it/s]
 42%|████▏     | 27798/65536 [4:45:59<6:34:25,  1.59it/s]
 42%|████▏     | 27799/65536 [4:46:00<6:31:38,  1.61it/s]
 42%|████▏     | 27800/65536 [4:46:01<6:48:24,  1.54it/s]
                                                         
{'loss': 1.6833, 'learning_rate': 6.198114362679074e-07, 'epoch': 1716.05}
+
 42%|████▏     | 27800/65536 [4:46:01<6:48:24,  1.54it/s]
 42%|████▏     | 27801/65536 [4:46:01<6:40:34,  1.57it/s]
 42%|████▏     | 27802/65536 [4:46:02<6:32:24,  1.60it/s]
 42%|████▏     | 27803/65536 [4:46:03<6:36:09,  1.59it/s]
 42%|████▏     | 27804/65536 [4:46:03<6:49:03,  1.54it/s]
 42%|████▏     | 27805/65536 [4:46:04<6:33:33,  1.60it/s]
 42%|████▏     | 27806/65536 [4:46:04<6:23:39,  1.64it/s]
 42%|████▏     | 27807/65536 [4:46:05<6:21:17,  1.65it/s]
 42%|████▏     | 27808/65536 [4:46:06<6:16:28,  1.67it/s]
 42%|████▏     | 27809/65536 [4:46:06<6:10:41,  1.70it/s]
 42%|████▏     | 27810/65536 [4:46:07<6:23:04,  1.64it/s]
 42%|████▏     | 27811/65536 [4:46:07<6:19:45,  1.66it/s]
 42%|████▏     | 27812/65536 [4:46:08<6:24:42,  1.63it/s]
 42%|████▏     | 27813/65536 [4:46:09<6:19:05,  1.66it/s]
 42%|████▏     | 27814/65536 [4:46:09<6:21:15,  1.65it/s]
 42%|████▏     | 27815/65536 [4:46:10<6:26:45,  1.63it/s]
 42%|████▏     | 27816/65536 [4:46:11<6:46:57,  1.54it/s]
 42%|████▏     | 27817/65536 [4:46:11<6:43:12,  1.56it/s]
 42%|████▏     | 27818/65536 [4:46:12<6:37:35,  1.58it/s]
 42%|████▏     | 27819/65536 [4:46:12<6:28:47,  1.62it/s]
 42%|████▏     | 27820/65536 [4:46:13<6:22:20,  1.64it/s]
                                                         
{'loss': 1.7555, 'learning_rate': 6.195359373086812e-07, 'epoch': 1717.28}
+
 42%|████▏     | 27820/65536 [4:46:13<6:22:20,  1.64it/s]
 42%|████▏     | 27821/65536 [4:46:14<6:25:35,  1.63it/s]
 42%|████▏     | 27822/65536 [4:46:14<6:32:41,  1.60it/s]
 42%|████▏     | 27823/65536 [4:46:15<6:25:55,  1.63it/s]
 42%|████▏     | 27824/65536 [4:46:15<6:20:07,  1.65it/s]
 42%|████▏     | 27825/65536 [4:46:16<6:26:39,  1.63it/s]
 42%|████▏     | 27826/65536 [4:46:17<6:19:40,  1.66it/s]
 42%|████▏     | 27827/65536 [4:46:17<6:23:17,  1.64it/s]
 42%|████▏     | 27828/65536 [4:46:18<6:23:31,  1.64it/s]
 42%|████▏     | 27829/65536 [4:46:19<6:17:01,  1.67it/s]
 42%|████▏     | 27830/65536 [4:46:19<6:17:29,  1.66it/s]
 42%|████▏     | 27831/65536 [4:46:20<6:26:04,  1.63it/s]
 42%|████▏     | 27832/65536 [4:46:20<6:33:40,  1.60it/s]
 42%|████▏     | 27833/65536 [4:46:21<6:35:16,  1.59it/s]
 42%|████▏     | 27834/65536 [4:46:22<6:15:24,  1.67it/s]
 42%|████▏     | 27835/65536 [4:46:22<6:28:36,  1.62it/s]
 42%|████▏     | 27836/65536 [4:46:23<6:24:46,  1.63it/s]
 42%|████▏     | 27837/65536 [4:46:23<6:19:23,  1.66it/s]
 42%|████▏     | 27838/65536 [4:46:24<6:19:41,  1.65it/s]
 42%|████▏     | 27839/65536 [4:46:25<6:27:05,  1.62it/s]
 42%|████▏     | 27840/65536 [4:46:25<6:36:48,  1.58it/s]
                                                         
{'loss': 1.6939, 'learning_rate': 6.192604383494551e-07, 'epoch': 1718.52}
+
 42%|████▏     | 27840/65536 [4:46:25<6:36:48,  1.58it/s]
 42%|████▏     | 27841/65536 [4:46:26<6:36:48,  1.58it/s]
 42%|████▏     | 27842/65536 [4:46:27<6:36:04,  1.59it/s]
 42%|████▏     | 27843/65536 [4:46:27<6:23:54,  1.64it/s]
 42%|████▏     | 27844/65536 [4:46:28<6:18:58,  1.66it/s]
 42%|████▏     | 27845/65536 [4:46:28<6:13:43,  1.68it/s]
 42%|████▏     | 27846/65536 [4:46:29<6:19:16,  1.66it/s]
 42%|████▏     | 27847/65536 [4:46:30<6:22:31,  1.64it/s]
 42%|████▏     | 27848/65536 [4:46:30<6:40:26,  1.57it/s]
 42%|████▏     | 27849/65536 [4:46:31<6:34:57,  1.59it/s]
 42%|████▏     | 27850/65536 [4:46:31<6:25:28,  1.63it/s]
 42%|████▏     | 27851/65536 [4:46:32<6:24:43,  1.63it/s]
 42%|████▏     | 27852/65536 [4:46:33<6:25:32,  1.63it/s]
 43%|████▎     | 27853/65536 [4:46:33<6:28:52,  1.62it/s]
 43%|████▎     | 27854/65536 [4:46:34<6:22:00,  1.64it/s]
 43%|████▎     | 27855/65536 [4:46:35<6:24:20,  1.63it/s]
 43%|████▎     | 27856/65536 [4:46:35<6:33:10,  1.60it/s]
 43%|████▎     | 27857/65536 [4:46:36<6:36:17,  1.58it/s]
 43%|████▎     | 27858/65536 [4:46:37<6:45:47,  1.55it/s]
 43%|████▎     | 27859/65536 [4:46:37<6:40:19,  1.57it/s]
 43%|████▎     | 27860/65536 [4:46:38<6:31:07,  1.61it/s]
                                                         
{'loss': 1.6779, 'learning_rate': 6.18984939390229e-07, 'epoch': 1719.75}
+
 43%|████▎     | 27860/65536 [4:46:38<6:31:07,  1.61it/s]
 43%|████▎     | 27861/65536 [4:46:38<6:28:36,  1.62it/s]
 43%|████▎     | 27862/65536 [4:46:39<6:24:49,  1.63it/s]
 43%|████▎     | 27863/65536 [4:46:39<6:15:19,  1.67it/s]
 43%|████▎     | 27864/65536 [4:46:40<6:09:32,  1.70it/s]
 43%|████▎     | 27865/65536 [4:46:41<6:19:09,  1.66it/s]
 43%|████▎     | 27866/65536 [4:46:41<6:20:57,  1.65it/s]
 43%|████▎     | 27867/65536 [4:46:42<6:27:54,  1.62it/s]
 43%|████▎     | 27868/65536 [4:46:43<6:26:59,  1.62it/s]
 43%|████▎     | 27869/65536 [4:46:43<6:25:53,  1.63it/s]
 43%|████▎     | 27870/65536 [4:46:44<6:29:14,  1.61it/s]
 43%|████▎     | 27871/65536 [4:46:44<6:28:44,  1.61it/s]
 43%|████▎     | 27872/65536 [4:46:45<6:27:41,  1.62it/s]
 43%|████▎     | 27873/65536 [4:46:46<6:28:38,  1.62it/s]
 43%|████▎     | 27874/65536 [4:46:46<6:26:01,  1.63it/s]
 43%|████▎     | 27875/65536 [4:46:47<6:27:26,  1.62it/s]
 43%|████▎     | 27876/65536 [4:46:48<6:37:12,  1.58it/s]
 43%|████▎     | 27877/65536 [4:46:48<6:29:06,  1.61it/s]
 43%|████▎     | 27878/65536 [4:46:49<6:25:47,  1.63it/s]
 43%|████▎     | 27879/65536 [4:46:49<6:19:08,  1.66it/s]
 43%|████▎     | 27880/65536 [4:46:50<6:12:49,  1.68it/s]
                                                         
{'loss': 1.7628, 'learning_rate': 6.187094404310027e-07, 'epoch': 1720.99}
+
 43%|████▎     | 27880/65536 [4:46:50<6:12:49,  1.68it/s]
 43%|████▎     | 27881/65536 [4:46:51<6:28:30,  1.62it/s]
 43%|████▎     | 27882/65536 [4:46:51<6:33:41,  1.59it/s]
 43%|████▎     | 27883/65536 [4:46:52<6:49:30,  1.53it/s]
 43%|████▎     | 27884/65536 [4:46:53<6:32:06,  1.60it/s]
 43%|████▎     | 27885/65536 [4:46:53<6:27:24,  1.62it/s]
 43%|████▎     | 27886/65536 [4:46:54<6:18:54,  1.66it/s]
 43%|████▎     | 27887/65536 [4:46:54<6:10:30,  1.69it/s]
 43%|████▎     | 27888/65536 [4:46:55<6:09:47,  1.70it/s]
 43%|████▎     | 27889/65536 [4:46:55<6:13:43,  1.68it/s]
 43%|████▎     | 27890/65536 [4:46:56<6:22:21,  1.64it/s]
 43%|████▎     | 27891/65536 [4:46:57<6:27:24,  1.62it/s]
 43%|████▎     | 27892/65536 [4:46:57<6:18:36,  1.66it/s]
 43%|████▎     | 27893/65536 [4:46:58<6:19:23,  1.65it/s]
 43%|████▎     | 27894/65536 [4:46:59<6:28:43,  1.61it/s]
 43%|████▎     | 27895/65536 [4:46:59<6:22:10,  1.64it/s]
 43%|████▎     | 27896/65536 [4:47:00<6:16:48,  1.66it/s]
 43%|████▎     | 27897/65536 [4:47:00<6:37:32,  1.58it/s]
 43%|████▎     | 27898/65536 [4:47:01<6:32:02,  1.60it/s]
 43%|████▎     | 27899/65536 [4:47:02<6:25:14,  1.63it/s]
 43%|████▎     | 27900/65536 [4:47:02<6:26:12,  1.62it/s]
                                                         
{'loss': 1.7266, 'learning_rate': 6.184339414717766e-07, 'epoch': 1722.22}
+
 43%|████▎     | 27900/65536 [4:47:02<6:26:12,  1.62it/s]
 43%|████▎     | 27901/65536 [4:47:03<6:25:12,  1.63it/s]
 43%|████▎     | 27902/65536 [4:47:03<6:17:36,  1.66it/s]
 43%|████▎     | 27903/65536 [4:47:04<6:24:12,  1.63it/s]
 43%|████▎     | 27904/65536 [4:47:05<6:20:25,  1.65it/s]
 43%|████▎     | 27905/65536 [4:47:05<6:21:14,  1.65it/s]
 43%|████▎     | 27906/65536 [4:47:06<6:18:51,  1.66it/s]
 43%|████▎     | 27907/65536 [4:47:06<6:14:47,  1.67it/s]
 43%|████▎     | 27908/65536 [4:47:07<6:31:30,  1.60it/s]
 43%|████▎     | 27909/65536 [4:47:08<6:30:27,  1.61it/s]
 43%|████▎     | 27910/65536 [4:47:08<6:20:35,  1.65it/s]
 43%|████▎     | 27911/65536 [4:47:09<6:29:55,  1.61it/s]
 43%|████▎     | 27912/65536 [4:47:10<6:31:46,  1.60it/s]
 43%|████▎     | 27913/65536 [4:47:10<6:48:28,  1.54it/s]
 43%|████▎     | 27914/65536 [4:47:11<6:31:07,  1.60it/s]
 43%|████▎     | 27915/65536 [4:47:11<6:29:41,  1.61it/s]
 43%|████▎     | 27916/65536 [4:47:12<6:31:49,  1.60it/s]
 43%|████▎     | 27917/65536 [4:47:13<6:24:51,  1.63it/s]
 43%|████▎     | 27918/65536 [4:47:13<6:22:55,  1.64it/s]
 43%|████▎     | 27919/65536 [4:47:14<6:34:48,  1.59it/s]
 43%|████▎     | 27920/65536 [4:47:15<6:35:44,  1.58it/s]
                                                         
{'loss': 1.7034, 'learning_rate': 6.181584425125504e-07, 'epoch': 1723.46}
+
 43%|████▎     | 27920/65536 [4:47:15<6:35:44,  1.58it/s]
 43%|████▎     | 27921/65536 [4:47:15<6:36:09,  1.58it/s]
 43%|████▎     | 27922/65536 [4:47:16<6:39:47,  1.57it/s]
 43%|████▎     | 27923/65536 [4:47:17<6:32:34,  1.60it/s]
 43%|████▎     | 27924/65536 [4:47:17<6:24:30,  1.63it/s]
 43%|████▎     | 27925/65536 [4:47:18<6:12:33,  1.68it/s]
 43%|████▎     | 27926/65536 [4:47:18<6:13:31,  1.68it/s]
 43%|████▎     | 27927/65536 [4:47:19<6:11:14,  1.69it/s]
 43%|████▎     | 27928/65536 [4:47:19<6:16:02,  1.67it/s]
 43%|████▎     | 27929/65536 [4:47:20<6:27:05,  1.62it/s]
 43%|████▎     | 27930/65536 [4:47:21<6:24:56,  1.63it/s]
 43%|████▎     | 27931/65536 [4:47:21<6:31:46,  1.60it/s]
 43%|████▎     | 27932/65536 [4:47:22<6:29:45,  1.61it/s]
 43%|████▎     | 27933/65536 [4:47:23<6:27:20,  1.62it/s]
 43%|████▎     | 27934/65536 [4:47:23<6:21:58,  1.64it/s]
 43%|████▎     | 27935/65536 [4:47:24<6:14:06,  1.68it/s]
 43%|████▎     | 27936/65536 [4:47:24<6:17:08,  1.66it/s]
 43%|████▎     | 27937/65536 [4:47:25<6:19:01,  1.65it/s]
 43%|████▎     | 27938/65536 [4:47:26<6:29:28,  1.61it/s]
 43%|████▎     | 27939/65536 [4:47:26<6:25:42,  1.62it/s]
 43%|████▎     | 27940/65536 [4:47:27<6:33:01,  1.59it/s]
                                                         
{'loss': 1.758, 'learning_rate': 6.178829435533243e-07, 'epoch': 1724.69}
+
 43%|████▎     | 27940/65536 [4:47:27<6:33:01,  1.59it/s]
 43%|████▎     | 27941/65536 [4:47:27<6:31:54,  1.60it/s]
 43%|████▎     | 27942/65536 [4:47:28<6:29:19,  1.61it/s]
 43%|████▎     | 27943/65536 [4:47:29<6:29:16,  1.61it/s]
 43%|████▎     | 27944/65536 [4:47:29<6:17:39,  1.66it/s]
 43%|████▎     | 27945/65536 [4:47:30<6:12:13,  1.68it/s]
 43%|████▎     | 27946/65536 [4:47:31<6:29:17,  1.61it/s]
 43%|████▎     | 27947/65536 [4:47:31<6:26:24,  1.62it/s]
 43%|████▎     | 27948/65536 [4:47:32<6:31:09,  1.60it/s]
 43%|████▎     | 27949/65536 [4:47:32<6:21:18,  1.64it/s]
 43%|████▎     | 27950/65536 [4:47:33<6:25:34,  1.62it/s]
 43%|████▎     | 27951/65536 [4:47:34<6:18:40,  1.65it/s]
 43%|████▎     | 27952/65536 [4:47:34<6:25:41,  1.62it/s]
 43%|████▎     | 27953/65536 [4:47:35<6:19:03,  1.65it/s]
 43%|████▎     | 27954/65536 [4:47:35<6:14:04,  1.67it/s]
 43%|████▎     | 27955/65536 [4:47:36<6:15:45,  1.67it/s]
 43%|████▎     | 27956/65536 [4:47:37<6:28:46,  1.61it/s]
 43%|████▎     | 27957/65536 [4:47:37<6:16:08,  1.67it/s]
 43%|████▎     | 27958/65536 [4:47:38<6:27:06,  1.62it/s]
 43%|████▎     | 27959/65536 [4:47:39<6:37:09,  1.58it/s]
 43%|████▎     | 27960/65536 [4:47:39<6:36:23,  1.58it/s]
                                                         
{'loss': 1.7253, 'learning_rate': 6.176074445940982e-07, 'epoch': 1725.93}
+
 43%|████▎     | 27960/65536 [4:47:39<6:36:23,  1.58it/s]
 43%|████▎     | 27961/65536 [4:47:40<6:24:53,  1.63it/s]
 43%|████▎     | 27962/65536 [4:47:40<6:31:57,  1.60it/s]
 43%|████▎     | 27963/65536 [4:47:41<6:16:47,  1.66it/s]
 43%|████▎     | 27964/65536 [4:47:42<6:10:31,  1.69it/s]
 43%|████▎     | 27965/65536 [4:47:42<6:09:15,  1.70it/s]
 43%|████▎     | 27966/65536 [4:47:43<6:13:03,  1.68it/s]
 43%|████▎     | 27967/65536 [4:47:43<6:16:04,  1.66it/s]
 43%|████▎     | 27968/65536 [4:47:44<6:28:31,  1.61it/s]
 43%|████▎     | 27969/65536 [4:47:45<6:28:55,  1.61it/s]
 43%|████▎     | 27970/65536 [4:47:45<6:20:17,  1.65it/s]
 43%|████▎     | 27971/65536 [4:47:46<6:15:40,  1.67it/s]
 43%|████▎     | 27972/65536 [4:47:46<6:12:37,  1.68it/s]
 43%|████▎     | 27973/65536 [4:47:47<6:22:15,  1.64it/s]
 43%|████▎     | 27974/65536 [4:47:48<6:14:21,  1.67it/s]
 43%|████▎     | 27975/65536 [4:47:48<6:21:42,  1.64it/s]
 43%|████▎     | 27976/65536 [4:47:49<6:36:20,  1.58it/s]
 43%|████▎     | 27977/65536 [4:47:49<6:30:44,  1.60it/s]
 43%|████▎     | 27978/65536 [4:47:50<6:36:57,  1.58it/s]
 43%|████▎     | 27979/65536 [4:47:51<6:25:54,  1.62it/s]
 43%|████▎     | 27980/65536 [4:47:51<6:20:34,  1.64it/s]
                                                         
{'loss': 1.7962, 'learning_rate': 6.17331945634872e-07, 'epoch': 1727.16}
+
 43%|████▎     | 27980/65536 [4:47:51<6:20:34,  1.64it/s]
 43%|████▎     | 27981/65536 [4:47:52<6:27:19,  1.62it/s]
 43%|████▎     | 27982/65536 [4:47:53<6:27:09,  1.62it/s]
 43%|████▎     | 27983/65536 [4:47:53<6:33:51,  1.59it/s]
 43%|████▎     | 27984/65536 [4:47:54<6:21:30,  1.64it/s]
 43%|████▎     | 27985/65536 [4:47:54<6:24:26,  1.63it/s]
 43%|████▎     | 27986/65536 [4:47:55<6:17:41,  1.66it/s]
 43%|████▎     | 27987/65536 [4:47:56<6:25:57,  1.62it/s]
 43%|████▎     | 27988/65536 [4:47:56<6:18:24,  1.65it/s]
 43%|████▎     | 27989/65536 [4:47:57<6:14:05,  1.67it/s]
 43%|████▎     | 27990/65536 [4:47:57<6:19:43,  1.65it/s]
 43%|████▎     | 27991/65536 [4:47:58<6:20:59,  1.64it/s]
 43%|████▎     | 27992/65536 [4:47:59<6:23:48,  1.63it/s]
 43%|████▎     | 27993/65536 [4:47:59<6:25:11,  1.62it/s]
 43%|████▎     | 27994/65536 [4:48:00<6:35:50,  1.58it/s]
 43%|████▎     | 27995/65536 [4:48:01<6:37:28,  1.57it/s]
 43%|████▎     | 27996/65536 [4:48:01<6:40:27,  1.56it/s]
 43%|████▎     | 27997/65536 [4:48:02<6:36:09,  1.58it/s]
 43%|████▎     | 27998/65536 [4:48:02<6:31:35,  1.60it/s]
 43%|████▎     | 27999/65536 [4:48:03<6:39:00,  1.57it/s]
 43%|████▎     | 28000/65536 [4:48:04<6:19:43,  1.65it/s]
                                                         
{'loss': 1.7053, 'learning_rate': 6.170564466756459e-07, 'epoch': 1728.4}
+
 43%|████▎     | 28000/65536 [4:48:04<6:19:43,  1.65it/s]
 43%|████▎     | 28001/65536 [4:48:04<6:28:08,  1.61it/s]
 43%|████▎     | 28002/65536 [4:48:05<6:18:46,  1.65it/s]
 43%|████▎     | 28003/65536 [4:48:05<6:13:28,  1.67it/s]
 43%|████▎     | 28004/65536 [4:48:06<6:26:11,  1.62it/s]
 43%|████▎     | 28005/65536 [4:48:07<6:21:36,  1.64it/s]
 43%|████▎     | 28006/65536 [4:48:07<6:17:44,  1.66it/s]
 43%|████▎     | 28007/65536 [4:48:08<6:19:11,  1.65it/s]
 43%|████▎     | 28008/65536 [4:48:09<6:18:38,  1.65it/s]
 43%|████▎     | 28009/65536 [4:48:09<6:20:48,  1.64it/s]
 43%|████▎     | 28010/65536 [4:48:10<6:25:50,  1.62it/s]
 43%|████▎     | 28011/65536 [4:48:10<6:26:00,  1.62it/s]
 43%|████▎     | 28012/65536 [4:48:11<6:15:38,  1.66it/s]
 43%|████▎     | 28013/65536 [4:48:12<6:08:47,  1.70it/s]
 43%|████▎     | 28014/65536 [4:48:12<6:17:51,  1.66it/s]
 43%|████▎     | 28015/65536 [4:48:13<6:12:28,  1.68it/s]
 43%|████▎     | 28016/65536 [4:48:13<6:10:45,  1.69it/s]
 43%|████▎     | 28017/65536 [4:48:14<6:11:13,  1.68it/s]
 43%|████▎     | 28018/65536 [4:48:14<6:05:16,  1.71it/s]
 43%|████▎     | 28019/65536 [4:48:15<6:24:00,  1.63it/s]
 43%|████▎     | 28020/65536 [4:48:16<6:29:49,  1.60it/s]
                                                         
{'loss': 1.7582, 'learning_rate': 6.167809477164197e-07, 'epoch': 1729.63}
+
 43%|████▎     | 28020/65536 [4:48:16<6:29:49,  1.60it/s]
 43%|████▎     | 28021/65536 [4:48:16<6:19:53,  1.65it/s]
 43%|████▎     | 28022/65536 [4:48:17<6:20:47,  1.64it/s]
 43%|████▎     | 28023/65536 [4:48:18<6:27:04,  1.62it/s]
 43%|████▎     | 28024/65536 [4:48:18<6:23:43,  1.63it/s]
 43%|████▎     | 28025/65536 [4:48:19<6:23:09,  1.63it/s]
 43%|████▎     | 28026/65536 [4:48:20<6:35:43,  1.58it/s]
 43%|████▎     | 28027/65536 [4:48:20<6:46:58,  1.54it/s]
 43%|████▎     | 28028/65536 [4:48:21<6:33:42,  1.59it/s]
 43%|████▎     | 28029/65536 [4:48:21<6:36:58,  1.57it/s]
 43%|████▎     | 28030/65536 [4:48:22<6:25:05,  1.62it/s]
 43%|████▎     | 28031/65536 [4:48:23<6:21:37,  1.64it/s]
 43%|████▎     | 28032/65536 [4:48:23<6:19:34,  1.65it/s]
 43%|████▎     | 28033/65536 [4:48:24<6:15:51,  1.66it/s]
 43%|████▎     | 28034/65536 [4:48:24<6:16:49,  1.66it/s]
 43%|████▎     | 28035/65536 [4:48:25<6:21:14,  1.64it/s]
 43%|████▎     | 28036/65536 [4:48:26<6:21:19,  1.64it/s]
 43%|████▎     | 28037/65536 [4:48:26<6:22:09,  1.64it/s]
 43%|████▎     | 28038/65536 [4:48:27<6:24:45,  1.62it/s]
 43%|████▎     | 28039/65536 [4:48:28<6:35:42,  1.58it/s]
 43%|████▎     | 28040/65536 [4:48:28<6:23:12,  1.63it/s]
                                                         
{'loss': 1.7062, 'learning_rate': 6.165054487571936e-07, 'epoch': 1730.86}
+
 43%|████▎     | 28040/65536 [4:48:28<6:23:12,  1.63it/s]
 43%|████▎     | 28041/65536 [4:48:29<6:18:58,  1.65it/s]
 43%|████▎     | 28042/65536 [4:48:29<6:21:07,  1.64it/s]
 43%|████▎     | 28043/65536 [4:48:30<6:29:27,  1.60it/s]
 43%|████▎     | 28044/65536 [4:48:31<6:26:13,  1.62it/s]
 43%|████▎     | 28045/65536 [4:48:31<6:20:26,  1.64it/s]
 43%|████▎     | 28046/65536 [4:48:32<6:21:41,  1.64it/s]
 43%|████▎     | 28047/65536 [4:48:32<6:21:13,  1.64it/s]
 43%|████▎     | 28048/65536 [4:48:33<6:28:15,  1.61it/s]
 43%|████▎     | 28049/65536 [4:48:34<6:23:37,  1.63it/s]
 43%|████▎     | 28050/65536 [4:48:34<6:13:23,  1.67it/s]
 43%|████▎     | 28051/65536 [4:48:35<6:17:48,  1.65it/s]
 43%|████▎     | 28052/65536 [4:48:35<6:19:01,  1.65it/s]
 43%|████▎     | 28053/65536 [4:48:36<6:17:50,  1.65it/s]
 43%|████▎     | 28054/65536 [4:48:37<6:17:58,  1.65it/s]
 43%|████▎     | 28055/65536 [4:48:37<6:28:43,  1.61it/s]
 43%|████▎     | 28056/65536 [4:48:38<6:23:22,  1.63it/s]
 43%|████▎     | 28057/65536 [4:48:39<6:37:17,  1.57it/s]
 43%|████▎     | 28058/65536 [4:48:39<6:27:40,  1.61it/s]
 43%|████▎     | 28059/65536 [4:48:40<6:42:09,  1.55it/s]
 43%|████▎     | 28060/65536 [4:48:40<6:32:27,  1.59it/s]
                                                         
{'loss': 1.7443, 'learning_rate': 6.162299497979674e-07, 'epoch': 1732.1}
+
 43%|████▎     | 28060/65536 [4:48:40<6:32:27,  1.59it/s]
 43%|████▎     | 28061/65536 [4:48:41<6:22:53,  1.63it/s]
 43%|████▎     | 28062/65536 [4:48:42<6:34:16,  1.58it/s]
 43%|████▎     | 28063/65536 [4:48:42<6:25:48,  1.62it/s]
 43%|████▎     | 28064/65536 [4:48:43<6:19:52,  1.64it/s]
 43%|████▎     | 28065/65536 [4:48:44<6:35:31,  1.58it/s]
 43%|████▎     | 28066/65536 [4:48:44<6:28:45,  1.61it/s]
 43%|████▎     | 28067/65536 [4:48:45<6:24:56,  1.62it/s]
 43%|████▎     | 28068/65536 [4:48:45<6:16:22,  1.66it/s]
 43%|████▎     | 28069/65536 [4:48:46<6:17:34,  1.65it/s]
 43%|████▎     | 28070/65536 [4:48:47<6:10:12,  1.69it/s]
 43%|████▎     | 28071/65536 [4:48:47<6:24:45,  1.62it/s]
 43%|████▎     | 28072/65536 [4:48:48<6:19:47,  1.64it/s]
 43%|████▎     | 28073/65536 [4:48:48<6:20:06,  1.64it/s]
 43%|████▎     | 28074/65536 [4:48:49<6:11:06,  1.68it/s]
 43%|████▎     | 28075/65536 [4:48:50<6:27:46,  1.61it/s]
 43%|████▎     | 28076/65536 [4:48:50<6:19:19,  1.65it/s]
 43%|████▎     | 28077/65536 [4:48:51<6:19:42,  1.64it/s]
 43%|████▎     | 28078/65536 [4:48:51<6:13:26,  1.67it/s]
 43%|████▎     | 28079/65536 [4:48:52<6:06:13,  1.70it/s]
 43%|████▎     | 28080/65536 [4:48:53<6:14:55,  1.67it/s]
                                                         
{'loss': 1.7365, 'learning_rate': 6.159544508387412e-07, 'epoch': 1733.33}
+
 43%|████▎     | 28080/65536 [4:48:53<6:14:55,  1.67it/s]
 43%|████▎     | 28081/65536 [4:48:53<6:19:33,  1.64it/s]
 43%|████▎     | 28082/65536 [4:48:54<6:24:31,  1.62it/s]
 43%|████▎     | 28083/65536 [4:48:54<6:23:32,  1.63it/s]
 43%|████▎     | 28084/65536 [4:48:55<6:24:00,  1.63it/s]
 43%|████▎     | 28085/65536 [4:48:56<6:25:26,  1.62it/s]
 43%|████▎     | 28086/65536 [4:48:56<6:34:33,  1.58it/s]
 43%|████▎     | 28087/65536 [4:48:57<6:34:02,  1.58it/s]
 43%|████▎     | 28088/65536 [4:48:58<6:24:30,  1.62it/s]
 43%|████▎     | 28089/65536 [4:48:58<6:23:24,  1.63it/s]
 43%|████▎     | 28090/65536 [4:48:59<6:19:39,  1.64it/s]
 43%|████▎     | 28091/65536 [4:48:59<6:35:27,  1.58it/s]
 43%|████▎     | 28092/65536 [4:49:00<6:38:23,  1.57it/s]
 43%|████▎     | 28093/65536 [4:49:01<6:35:40,  1.58it/s]
 43%|████▎     | 28094/65536 [4:49:01<6:31:20,  1.59it/s]
 43%|████▎     | 28095/65536 [4:49:02<6:25:59,  1.62it/s]
 43%|████▎     | 28096/65536 [4:49:03<6:22:16,  1.63it/s]
 43%|████▎     | 28097/65536 [4:49:03<6:19:22,  1.64it/s]
 43%|████▎     | 28098/65536 [4:49:04<6:09:08,  1.69it/s]
 43%|████▎     | 28099/65536 [4:49:04<6:20:14,  1.64it/s]
 43%|████▎     | 28100/65536 [4:49:05<6:28:03,  1.61it/s]
                                                         
{'loss': 1.6809, 'learning_rate': 6.15678951879515e-07, 'epoch': 1734.57}
+
 43%|████▎     | 28100/65536 [4:49:05<6:28:03,  1.61it/s]
 43%|████▎     | 28101/65536 [4:49:06<6:29:13,  1.60it/s]
 43%|████▎     | 28102/65536 [4:49:06<6:24:34,  1.62it/s]
 43%|████▎     | 28103/65536 [4:49:07<6:23:07,  1.63it/s]
 43%|████▎     | 28104/65536 [4:49:07<6:24:24,  1.62it/s]
 43%|████▎     | 28105/65536 [4:49:08<6:17:44,  1.65it/s]
 43%|████▎     | 28106/65536 [4:49:09<6:23:04,  1.63it/s]
 43%|████▎     | 28107/65536 [4:49:09<6:16:08,  1.66it/s]
 43%|████▎     | 28108/65536 [4:49:10<6:26:17,  1.61it/s]
 43%|████▎     | 28109/65536 [4:49:11<6:32:33,  1.59it/s]
 43%|████▎     | 28110/65536 [4:49:11<6:35:01,  1.58it/s]
 43%|████▎     | 28111/65536 [4:49:12<6:25:29,  1.62it/s]
 43%|████▎     | 28112/65536 [4:49:12<6:23:06,  1.63it/s]
 43%|████▎     | 28113/65536 [4:49:13<6:17:06,  1.65it/s]
 43%|████▎     | 28114/65536 [4:49:14<6:29:44,  1.60it/s]
 43%|████▎     | 28115/65536 [4:49:14<6:39:52,  1.56it/s]
 43%|████▎     | 28116/65536 [4:49:15<6:35:28,  1.58it/s]
 43%|████▎     | 28117/65536 [4:49:16<6:27:07,  1.61it/s]
 43%|████▎     | 28118/65536 [4:49:16<6:25:47,  1.62it/s]
 43%|████▎     | 28119/65536 [4:49:17<6:27:28,  1.61it/s]
 43%|████▎     | 28120/65536 [4:49:17<6:16:33,  1.66it/s]
                                                         
{'loss': 1.7113, 'learning_rate': 6.15403452920289e-07, 'epoch': 1735.8}
+
 43%|████▎     | 28120/65536 [4:49:17<6:16:33,  1.66it/s]
 43%|████▎     | 28121/65536 [4:49:18<6:09:42,  1.69it/s]
 43%|████▎     | 28122/65536 [4:49:19<6:11:06,  1.68it/s]
 43%|████▎     | 28123/65536 [4:49:19<6:00:40,  1.73it/s]
 43%|████▎     | 28124/65536 [4:49:20<6:17:21,  1.65it/s]
 43%|████▎     | 28125/65536 [4:49:20<6:10:53,  1.68it/s]
 43%|████▎     | 28126/65536 [4:49:21<6:07:20,  1.70it/s]
 43%|████▎     | 28127/65536 [4:49:21<6:11:27,  1.68it/s]
 43%|████▎     | 28128/65536 [4:49:22<6:12:37,  1.67it/s]
 43%|████▎     | 28129/65536 [4:49:23<6:14:23,  1.67it/s]
 43%|████▎     | 28130/65536 [4:49:23<6:16:54,  1.65it/s]
 43%|████▎     | 28131/65536 [4:49:24<6:31:21,  1.59it/s]
 43%|████▎     | 28132/65536 [4:49:25<6:21:20,  1.63it/s]
 43%|█���██▎     | 28133/65536 [4:49:25<6:26:42,  1.61it/s]
 43%|████▎     | 28134/65536 [4:49:26<6:36:31,  1.57it/s]
 43%|████▎     | 28135/65536 [4:49:26<6:27:58,  1.61it/s]
 43%|████▎     | 28136/65536 [4:49:27<6:17:45,  1.65it/s]
 43%|████▎     | 28137/65536 [4:49:28<6:10:24,  1.68it/s]
 43%|████▎     | 28138/65536 [4:49:28<6:07:28,  1.70it/s]
 43%|████▎     | 28139/65536 [4:49:29<6:16:00,  1.66it/s]
 43%|████▎     | 28140/65536 [4:49:29<6:25:45,  1.62it/s]
                                                         
{'loss': 1.7568, 'learning_rate': 6.151279539610628e-07, 'epoch': 1737.04}
+
 43%|████▎     | 28140/65536 [4:49:29<6:25:45,  1.62it/s]
 43%|████▎     | 28141/65536 [4:49:30<6:30:31,  1.60it/s]
 43%|████▎     | 28142/65536 [4:49:31<6:25:23,  1.62it/s]
 43%|████▎     | 28143/65536 [4:49:31<6:17:24,  1.65it/s]
 43%|████▎     | 28144/65536 [4:49:32<6:15:28,  1.66it/s]
 43%|████▎     | 28145/65536 [4:49:32<6:13:01,  1.67it/s]
 43%|████▎     | 28146/65536 [4:49:33<6:06:32,  1.70it/s]
 43%|████▎     | 28147/65536 [4:49:34<6:02:31,  1.72it/s]
 43%|████▎     | 28148/65536 [4:49:34<6:06:09,  1.70it/s]
 43%|████▎     | 28149/65536 [4:49:35<6:19:16,  1.64it/s]
 43%|████▎     | 28150/65536 [4:49:36<6:26:13,  1.61it/s]
 43%|████▎     | 28151/65536 [4:49:36<6:24:25,  1.62it/s]
 43%|████▎     | 28152/65536 [4:49:37<6:22:19,  1.63it/s]
 43%|████▎     | 28153/65536 [4:49:37<6:22:59,  1.63it/s]
 43%|████▎     | 28154/65536 [4:49:38<6:19:07,  1.64it/s]
 43%|████▎     | 28155/65536 [4:49:39<6:21:46,  1.63it/s]
 43%|████▎     | 28156/65536 [4:49:39<6:39:27,  1.56it/s]
 43%|████▎     | 28157/65536 [4:49:40<6:25:05,  1.62it/s]
 43%|████▎     | 28158/65536 [4:49:40<6:25:25,  1.62it/s]
 43%|████▎     | 28159/65536 [4:49:41<6:27:35,  1.61it/s]
 43%|████▎     | 28160/65536 [4:49:42<6:16:20,  1.66it/s]
                                                         
{'loss': 1.7214, 'learning_rate': 6.148524550018366e-07, 'epoch': 1738.27}
+
 43%|████▎     | 28160/65536 [4:49:42<6:16:20,  1.66it/s]
 43%|████▎     | 28161/65536 [4:49:42<6:06:18,  1.70it/s]
 43%|████▎     | 28162/65536 [4:49:43<6:05:50,  1.70it/s]
 43%|████▎     | 28163/65536 [4:49:43<6:18:01,  1.65it/s]
 43%|████▎     | 28164/65536 [4:49:44<6:11:13,  1.68it/s]
 43%|████▎     | 28165/65536 [4:49:45<6:22:51,  1.63it/s]
 43%|████▎     | 28166/65536 [4:49:45<6:16:42,  1.65it/s]
 43%|████▎     | 28167/65536 [4:49:46<6:23:22,  1.62it/s]
 43%|████▎     | 28168/65536 [4:49:47<6:21:02,  1.63it/s]
 43%|████▎     | 28169/65536 [4:49:47<6:25:17,  1.62it/s]
 43%|████▎     | 28170/65536 [4:49:48<6:21:12,  1.63it/s]
 43%|████▎     | 28171/65536 [4:49:48<6:25:00,  1.62it/s]
 43%|████▎     | 28172/65536 [4:49:49<6:43:20,  1.54it/s]
 43%|████▎     | 28173/65536 [4:49:50<6:40:36,  1.55it/s]
 43%|████▎     | 28174/65536 [4:49:50<6:46:31,  1.53it/s]
 43%|████▎     | 28175/65536 [4:49:51<6:40:42,  1.55it/s]
 43%|████▎     | 28176/65536 [4:49:52<6:39:15,  1.56it/s]
 43%|████▎     | 28177/65536 [4:49:52<6:37:20,  1.57it/s]
 43%|████▎     | 28178/65536 [4:49:53<6:32:01,  1.59it/s]
 43%|████▎     | 28179/65536 [4:49:53<6:27:53,  1.61it/s]
 43%|████▎     | 28180/65536 [4:49:54<6:24:46,  1.62it/s]
                                                         
{'loss': 1.6456, 'learning_rate': 6.145769560426105e-07, 'epoch': 1739.51}
+
 43%|████▎     | 28180/65536 [4:49:54<6:24:46,  1.62it/s]
 43%|████▎     | 28181/65536 [4:49:55<6:19:25,  1.64it/s]
 43%|████▎     | 28182/65536 [4:49:55<6:12:09,  1.67it/s]
 43%|████▎     | 28183/65536 [4:49:56<6:25:53,  1.61it/s]
 43%|████▎     | 28184/65536 [4:49:57<6:23:50,  1.62it/s]
 43%|████▎     | 28185/65536 [4:49:57<6:17:07,  1.65it/s]
 43%|████▎     | 28186/65536 [4:49:58<6:14:35,  1.66it/s]
 43%|████▎     | 28187/65536 [4:49:58<6:23:05,  1.62it/s]
 43%|████▎     | 28188/65536 [4:49:59<6:20:26,  1.64it/s]
 43%|████▎     | 28189/65536 [4:50:00<6:36:39,  1.57it/s]
 43%|████▎     | 28190/65536 [4:50:00<6:32:08,  1.59it/s]
 43%|████▎     | 28191/65536 [4:50:01<6:27:40,  1.61it/s]
 43%|████▎     | 28192/65536 [4:50:02<6:35:36,  1.57it/s]
 43%|████▎     | 28193/65536 [4:50:02<6:25:55,  1.61it/s]
 43%|████▎     | 28194/65536 [4:50:03<6:15:44,  1.66it/s]
 43%|████▎     | 28195/65536 [4:50:03<6:17:24,  1.65it/s]
 43%|████▎     | 28196/65536 [4:50:04<6:21:13,  1.63it/s]
 43%|████▎     | 28197/65536 [4:50:05<6:34:38,  1.58it/s]
 43%|████▎     | 28198/65536 [4:50:05<6:29:37,  1.60it/s]
 43%|████▎     | 28199/65536 [4:50:06<6:25:35,  1.61it/s]
 43%|████▎     | 28200/65536 [4:50:06<6:24:18,  1.62it/s]
                                                         
{'loss': 1.707, 'learning_rate': 6.143014570833843e-07, 'epoch': 1740.74}
+
 43%|████▎     | 28200/65536 [4:50:06<6:24:18,  1.62it/s]
 43%|████▎     | 28201/65536 [4:50:07<6:28:07,  1.60it/s]
 43%|████▎     | 28202/65536 [4:50:08<6:29:06,  1.60it/s]
 43%|████▎     | 28203/65536 [4:50:08<6:27:14,  1.61it/s]
 43%|████▎     | 28204/65536 [4:50:09<6:18:33,  1.64it/s]
 43%|████▎     | 28205/65536 [4:50:10<6:27:56,  1.60it/s]
 43%|████▎     | 28206/65536 [4:50:10<6:28:28,  1.60it/s]
 43%|████▎     | 28207/65536 [4:50:11<6:35:26,  1.57it/s]
 43%|████▎     | 28208/65536 [4:50:11<6:31:19,  1.59it/s]
 43%|████▎     | 28209/65536 [4:50:12<6:35:14,  1.57it/s]
 43%|████▎     | 28210/65536 [4:50:13<6:27:11,  1.61it/s]
 43%|████▎     | 28211/65536 [4:50:13<6:23:28,  1.62it/s]
 43%|████▎     | 28212/65536 [4:50:14<6:22:43,  1.63it/s]
 43%|████▎     | 28213/65536 [4:50:15<6:18:30,  1.64it/s]
 43%|████▎     | 28214/65536 [4:50:15<6:23:09,  1.62it/s]
 43%|████▎     | 28215/65536 [4:50:16<6:27:15,  1.61it/s]
 43%|████▎     | 28216/65536 [4:50:16<6:23:04,  1.62it/s]
 43%|████▎     | 28217/65536 [4:50:17<6:13:54,  1.66it/s]
 43%|████▎     | 28218/65536 [4:50:18<6:12:46,  1.67it/s]
 43%|████▎     | 28219/65536 [4:50:18<6:13:02,  1.67it/s]
 43%|████▎     | 28220/65536 [4:50:19<6:15:31,  1.66it/s]
                                                         
{'loss': 1.6937, 'learning_rate': 6.140259581241582e-07, 'epoch': 1741.98}
+
 43%|████▎     | 28220/65536 [4:50:19<6:15:31,  1.66it/s]
 43%|████▎     | 28221/65536 [4:50:19<6:23:43,  1.62it/s]
 43%|████▎     | 28222/65536 [4:50:20<6:21:36,  1.63it/s]
 43%|████▎     | 28223/65536 [4:50:21<6:15:39,  1.66it/s]
 43%|████▎     | 28224/65536 [4:50:21<6:18:48,  1.64it/s]
 43%|████▎     | 28225/65536 [4:50:22<6:22:47,  1.62it/s]
 43%|████▎     | 28226/65536 [4:50:22<6:15:34,  1.66it/s]
 43%|████▎     | 28227/65536 [4:50:23<6:12:34,  1.67it/s]
 43%|████▎     | 28228/65536 [4:50:24<6:10:25,  1.68it/s]
 43%|████▎     | 28229/65536 [4:50:24<6:14:12,  1.66it/s]
 43%|████▎     | 28230/65536 [4:50:25<6:15:14,  1.66it/s]
 43%|████▎     | 28231/65536 [4:50:26<6:28:31,  1.60it/s]
 43%|████▎     | 28232/65536 [4:50:26<6:23:29,  1.62it/s]
 43%|████▎     | 28233/65536 [4:50:27<6:18:50,  1.64it/s]
 43%|████▎     | 28234/65536 [4:50:27<6:28:51,  1.60it/s]
 43%|████▎     | 28235/65536 [4:50:28<6:31:47,  1.59it/s]
 43%|████▎     | 28236/65536 [4:50:29<6:26:40,  1.61it/s]
 43%|████▎     | 28237/65536 [4:50:29<6:31:02,  1.59it/s]
 43%|████▎     | 28238/65536 [4:50:30<6:23:45,  1.62it/s]
 43%|████▎     | 28239/65536 [4:50:30<6:21:07,  1.63it/s]
 43%|████▎     | 28240/65536 [4:50:31<6:24:22,  1.62it/s]
                                                         
{'loss': 1.7067, 'learning_rate': 6.137504591649321e-07, 'epoch': 1743.21}
+
 43%|████▎     | 28240/65536 [4:50:31<6:24:22,  1.62it/s]
 43%|████▎     | 28241/65536 [4:50:32<6:15:31,  1.66it/s]
 43%|████▎     | 28242/65536 [4:50:32<6:27:32,  1.60it/s]
 43%|████▎     | 28243/65536 [4:50:33<6:33:36,  1.58it/s]
 43%|████▎     | 28244/65536 [4:50:34<6:26:07,  1.61it/s]
 43%|████▎     | 28245/65536 [4:50:34<6:22:19,  1.63it/s]
 43%|████▎     | 28246/65536 [4:50:35<6:20:27,  1.63it/s]
 43%|████▎     | 28247/65536 [4:50:35<6:27:19,  1.60it/s]
 43%|████▎     | 28248/65536 [4:50:36<6:27:55,  1.60it/s]
 43%|████▎     | 28249/65536 [4:50:37<6:19:46,  1.64it/s]
 43%|████▎     | 28250/65536 [4:50:37<6:17:24,  1.65it/s]
 43%|████▎     | 28251/65536 [4:50:38<6:14:06,  1.66it/s]
 43%|████▎     | 28252/65536 [4:50:39<6:33:53,  1.58it/s]
 43%|████▎     | 28253/65536 [4:50:39<6:40:15,  1.55it/s]
 43%|████▎     | 28254/65536 [4:50:40<6:40:31,  1.55it/s]
 43%|████▎     | 28255/65536 [4:50:40<6:33:22,  1.58it/s]
 43%|████▎     | 28256/65536 [4:50:41<6:22:28,  1.62it/s]
 43%|████▎     | 28257/65536 [4:50:42<6:20:15,  1.63it/s]
 43%|████▎     | 28258/65536 [4:50:42<6:29:10,  1.60it/s]
 43%|████▎     | 28259/65536 [4:50:43<6:23:21,  1.62it/s]
 43%|████▎     | 28260/65536 [4:50:43<6:19:00,  1.64it/s]
                                                         
{'loss': 1.726, 'learning_rate': 6.134749602057059e-07, 'epoch': 1744.44}
+
 43%|████▎     | 28260/65536 [4:50:43<6:19:00,  1.64it/s]
 43%|████▎     | 28261/65536 [4:50:44<6:15:36,  1.65it/s]
 43%|████▎     | 28262/65536 [4:50:45<6:13:18,  1.66it/s]
 43%|████▎     | 28263/65536 [4:50:45<6:28:12,  1.60it/s]
 43%|████▎     | 28264/65536 [4:50:46<6:26:19,  1.61it/s]
 43%|████▎     | 28265/65536 [4:50:47<6:26:59,  1.61it/s]
 43%|████▎     | 28266/65536 [4:50:47<6:15:10,  1.66it/s]
 43%|████▎     | 28267/65536 [4:50:48<6:11:48,  1.67it/s]
 43%|████▎     | 28268/65536 [4:50:48<6:12:17,  1.67it/s]
 43%|████▎     | 28269/65536 [4:50:49<6:23:05,  1.62it/s]
 43%|████▎     | 28270/65536 [4:50:50<6:34:27,  1.57it/s]
 43%|████▎     | 28271/65536 [4:50:50<6:25:04,  1.61it/s]
 43%|████▎     | 28272/65536 [4:50:51<6:25:44,  1.61it/s]
 43%|████▎     | 28273/65536 [4:50:51<6:29:01,  1.60it/s]
 43%|████▎     | 28274/65536 [4:50:52<6:24:14,  1.62it/s]
 43%|████▎     | 28275/65536 [4:50:53<6:30:06,  1.59it/s]
 43%|████▎     | 28276/65536 [4:50:53<6:22:10,  1.62it/s]
 43%|████▎     | 28277/65536 [4:50:54<6:16:59,  1.65it/s]
 43%|████▎     | 28278/65536 [4:50:55<6:18:53,  1.64it/s]
 43%|████▎     | 28279/65536 [4:50:55<6:15:23,  1.65it/s]
 43%|████▎     | 28280/65536 [4:50:56<6:16:46,  1.65it/s]
                                                         
{'loss': 1.7014, 'learning_rate': 6.131994612464798e-07, 'epoch': 1745.68}
+
 43%|████▎     | 28280/65536 [4:50:56<6:16:46,  1.65it/s]
 43%|████▎     | 28281/65536 [4:50:56<6:11:06,  1.67it/s]
 43%|████▎     | 28282/65536 [4:50:57<6:05:57,  1.70it/s]
 43%|████▎     | 28283/65536 [4:50:58<6:24:49,  1.61it/s]
 43%|████▎     | 28284/65536 [4:50:58<6:24:32,  1.61it/s]
 43%|████▎     | 28285/65536 [4:50:59<6:18:17,  1.64it/s]
 43%|████▎     | 28286/65536 [4:50:59<6:32:32,  1.58it/s]
 43%|████▎     | 28287/65536 [4:51:00<6:24:05,  1.62it/s]
 43%|████▎     | 28288/65536 [4:51:01<6:34:37,  1.57it/s]
 43%|████▎     | 28289/65536 [4:51:01<6:33:25,  1.58it/s]
 43%|████▎     | 28290/65536 [4:51:02<6:38:05,  1.56it/s]
 43%|████▎     | 28291/65536 [4:51:03<6:22:00,  1.62it/s]
 43%|████▎     | 28292/65536 [4:51:03<6:16:16,  1.65it/s]
 43%|████▎     | 28293/65536 [4:51:04<6:09:08,  1.68it/s]
 43%|████▎     | 28294/65536 [4:51:04<6:11:50,  1.67it/s]
 43%|████▎     | 28295/65536 [4:51:05<6:12:45,  1.67it/s]
 43%|████▎     | 28296/65536 [4:51:06<6:14:02,  1.66it/s]
 43%|████▎     | 28297/65536 [4:51:06<6:19:28,  1.64it/s]
 43%|████▎     | 28298/65536 [4:51:07<6:13:52,  1.66it/s]
 43%|████▎     | 28299/65536 [4:51:07<6:20:59,  1.63it/s]
 43%|████▎     | 28300/65536 [4:51:08<6:13:59,  1.66it/s]
                                                         
{'loss': 1.7578, 'learning_rate': 6.129239622872536e-07, 'epoch': 1746.91}
+
 43%|████▎     | 28300/65536 [4:51:08<6:13:59,  1.66it/s]
 43%|████▎     | 28301/65536 [4:51:09<6:23:09,  1.62it/s]
 43%|████▎     | 28302/65536 [4:51:09<6:40:26,  1.55it/s]
 43%|████▎     | 28303/65536 [4:51:10<6:40:50,  1.55it/s]
 43%|████▎     | 28304/65536 [4:51:11<6:36:24,  1.57it/s]
 43%|████▎     | 28305/65536 [4:51:11<6:27:44,  1.60it/s]
 43%|████▎     | 28306/65536 [4:51:12<6:23:19,  1.62it/s]
 43%|████▎     | 28307/65536 [4:51:12<6:19:48,  1.63it/s]
 43%|████▎     | 28308/65536 [4:51:13<6:15:00,  1.65it/s]
 43%|████▎     | 28309/65536 [4:51:14<6:15:10,  1.65it/s]
 43%|████▎     | 28310/65536 [4:51:14<6:25:08,  1.61it/s]
 43%|████▎     | 28311/65536 [4:51:15<6:28:09,  1.60it/s]
 43%|████▎     | 28312/65536 [4:51:15<6:21:31,  1.63it/s]
 43%|████▎     | 28313/65536 [4:51:16<6:28:45,  1.60it/s]
 43%|████▎     | 28314/65536 [4:51:17<6:25:01,  1.61it/s]
 43%|████▎     | 28315/65536 [4:51:17<6:22:05,  1.62it/s]
 43%|████▎     | 28316/65536 [4:51:18<6:24:09,  1.61it/s]
 43%|████▎     | 28317/65536 [4:51:19<6:19:05,  1.64it/s]
 43%|████▎     | 28318/65536 [4:51:19<6:37:00,  1.56it/s]
 43%|████▎     | 28319/65536 [4:51:20<6:26:58,  1.60it/s]
 43%|████▎     | 28320/65536 [4:51:20<6:28:14,  1.60it/s]
                                                         
{'loss': 1.6701, 'learning_rate': 6.126484633280275e-07, 'epoch': 1748.15}
+
 43%|████▎     | 28320/65536 [4:51:20<6:28:14,  1.60it/s]
 43%|████▎     | 28321/65536 [4:51:21<6:29:19,  1.59it/s]
 43%|████▎     | 28322/65536 [4:51:22<6:20:06,  1.63it/s]
 43%|████▎     | 28323/65536 [4:51:22<6:23:40,  1.62it/s]
 43%|████▎     | 28324/65536 [4:51:23<6:11:19,  1.67it/s]
 43%|████▎     | 28325/65536 [4:51:23<6:13:50,  1.66it/s]
 43%|████▎     | 28326/65536 [4:51:24<6:11:47,  1.67it/s]
 43%|████▎     | 28327/65536 [4:51:25<6:15:36,  1.65it/s]
 43%|████▎     | 28328/65536 [4:51:25<6:16:16,  1.65it/s]
 43%|██��█▎     | 28329/65536 [4:51:26<6:21:32,  1.63it/s]
 43%|████▎     | 28330/65536 [4:51:27<6:33:13,  1.58it/s]
 43%|████▎     | 28331/65536 [4:51:27<6:27:54,  1.60it/s]
 43%|████▎     | 28332/65536 [4:51:28<6:33:17,  1.58it/s]
 43%|████▎     | 28333/65536 [4:51:28<6:24:44,  1.61it/s]
 43%|████▎     | 28334/65536 [4:51:29<6:38:16,  1.56it/s]
 43%|████▎     | 28335/65536 [4:51:30<6:31:31,  1.58it/s]
 43%|████▎     | 28336/65536 [4:51:30<6:22:50,  1.62it/s]
 43%|████▎     | 28337/65536 [4:51:31<6:14:52,  1.65it/s]
 43%|████▎     | 28338/65536 [4:51:32<6:12:48,  1.66it/s]
 43%|████▎     | 28339/65536 [4:51:32<6:10:52,  1.67it/s]
 43%|████▎     | 28340/65536 [4:51:33<6:16:45,  1.65it/s]
                                                         
{'loss': 1.6948, 'learning_rate': 6.123729643688012e-07, 'epoch': 1749.38}
+
 43%|████▎     | 28340/65536 [4:51:33<6:16:45,  1.65it/s]
 43%|████▎     | 28341/65536 [4:51:33<6:19:57,  1.63it/s]
 43%|████▎     | 28342/65536 [4:51:34<6:19:50,  1.63it/s]
 43%|████▎     | 28343/65536 [4:51:35<6:10:21,  1.67it/s]
 43%|████▎     | 28344/65536 [4:51:35<6:20:42,  1.63it/s]
 43%|████▎     | 28345/65536 [4:51:36<6:24:13,  1.61it/s]
 43%|████▎     | 28346/65536 [4:51:36<6:20:00,  1.63it/s]
 43%|████▎     | 28347/65536 [4:51:37<6:27:19,  1.60it/s]
 43%|████▎     | 28348/65536 [4:51:38<6:19:19,  1.63it/s]
 43%|████▎     | 28349/65536 [4:51:38<6:26:10,  1.60it/s]
 43%|████▎     | 28350/65536 [4:51:39<6:17:08,  1.64it/s]
 43%|████▎     | 28351/65536 [4:51:40<6:32:14,  1.58it/s]
 43%|████▎     | 28352/65536 [4:51:40<6:36:51,  1.56it/s]
 43%|████▎     | 28353/65536 [4:51:41<6:47:18,  1.52it/s]
 43%|████▎     | 28354/65536 [4:51:42<6:31:52,  1.58it/s]
 43%|████▎     | 28355/65536 [4:51:42<6:29:55,  1.59it/s]
 43%|████▎     | 28356/65536 [4:51:43<6:22:37,  1.62it/s]
 43%|████▎     | 28357/65536 [4:51:43<6:20:52,  1.63it/s]
 43%|████▎     | 28358/65536 [4:51:44<6:25:13,  1.61it/s]
 43%|████▎     | 28359/65536 [4:51:45<6:18:11,  1.64it/s]
 43%|████▎     | 28360/65536 [4:51:45<6:12:55,  1.66it/s]
                                                         
{'loss': 1.699, 'learning_rate': 6.120974654095751e-07, 'epoch': 1750.62}
+
 43%|████▎     | 28360/65536 [4:51:45<6:12:55,  1.66it/s]
 43%|████▎     | 28361/65536 [4:51:46<6:13:29,  1.66it/s]
 43%|████▎     | 28362/65536 [4:51:46<6:17:23,  1.64it/s]
 43%|████▎     | 28363/65536 [4:51:47<6:16:25,  1.65it/s]
 43%|████▎     | 28364/65536 [4:51:48<6:19:43,  1.63it/s]
 43%|████▎     | 28365/65536 [4:51:48<6:19:55,  1.63it/s]
 43%|████▎     | 28366/65536 [4:51:49<6:15:35,  1.65it/s]
 43%|████▎     | 28367/65536 [4:51:49<6:32:29,  1.58it/s]
 43%|████▎     | 28368/65536 [4:51:50<6:39:24,  1.55it/s]
 43%|████▎     | 28369/65536 [4:51:51<6:33:17,  1.58it/s]
 43%|████▎     | 28370/65536 [4:51:51<6:26:13,  1.60it/s]
 43%|████▎     | 28371/65536 [4:51:52<6:24:29,  1.61it/s]
 43%|████▎     | 28372/65536 [4:51:53<6:23:23,  1.62it/s]
 43%|████▎     | 28373/65536 [4:51:53<6:23:21,  1.62it/s]
 43%|████▎     | 28374/65536 [4:51:54<6:16:37,  1.64it/s]
 43%|████▎     | 28375/65536 [4:51:54<6:25:34,  1.61it/s]
 43%|████▎     | 28376/65536 [4:51:55<6:14:43,  1.65it/s]
 43%|████▎     | 28377/65536 [4:51:56<6:24:56,  1.61it/s]
 43%|████▎     | 28378/65536 [4:51:56<6:17:53,  1.64it/s]
 43%|████▎     | 28379/65536 [4:51:57<6:14:15,  1.65it/s]
 43%|████▎     | 28380/65536 [4:51:57<6:06:48,  1.69it/s]
                                                         
{'loss': 1.6875, 'learning_rate': 6.11821966450349e-07, 'epoch': 1751.85}
+
 43%|████▎     | 28380/65536 [4:51:57<6:06:48,  1.69it/s]
 43%|████▎     | 28381/65536 [4:51:58<6:01:17,  1.71it/s]
 43%|████▎     | 28382/65536 [4:51:59<6:08:16,  1.68it/s]
 43%|████▎     | 28383/65536 [4:51:59<6:23:37,  1.61it/s]
 43%|████▎     | 28384/65536 [4:52:00<6:26:13,  1.60it/s]
 43%|████▎     | 28385/65536 [4:52:01<6:24:31,  1.61it/s]
 43%|████▎     | 28386/65536 [4:52:01<6:18:37,  1.64it/s]
 43%|████▎     | 28387/65536 [4:52:02<6:17:13,  1.64it/s]
 43%|████▎     | 28388/65536 [4:52:02<6:13:08,  1.66it/s]
 43%|████▎     | 28389/65536 [4:52:03<6:15:12,  1.65it/s]
 43%|████▎     | 28390/65536 [4:52:04<6:12:17,  1.66it/s]
 43%|████▎     | 28391/65536 [4:52:04<6:10:01,  1.67it/s]
 43%|████▎     | 28392/65536 [4:52:05<6:20:58,  1.62it/s]
 43%|████▎     | 28393/65536 [4:52:05<6:12:10,  1.66it/s]
 43%|████▎     | 28394/65536 [4:52:06<6:27:25,  1.60it/s]
 43%|████▎     | 28395/65536 [4:52:07<6:23:57,  1.61it/s]
 43%|████▎     | 28396/65536 [4:52:07<6:27:51,  1.60it/s]
 43%|████▎     | 28397/65536 [4:52:08<6:18:06,  1.64it/s]
 43%|████▎     | 28398/65536 [4:52:08<6:16:49,  1.64it/s]
 43%|████▎     | 28399/65536 [4:52:09<6:30:17,  1.59it/s]
 43%|████▎     | 28400/65536 [4:52:10<6:26:25,  1.60it/s]
                                                         
{'loss': 1.7007, 'learning_rate': 6.115464674911228e-07, 'epoch': 1753.09}
+
 43%|████▎     | 28400/65536 [4:52:10<6:26:25,  1.60it/s]
 43%|████▎     | 28401/65536 [4:52:10<6:23:20,  1.61it/s]
 43%|████▎     | 28402/65536 [4:52:11<6:30:16,  1.59it/s]
 43%|████▎     | 28403/65536 [4:52:12<6:37:54,  1.56it/s]
 43%|████▎     | 28404/65536 [4:52:12<6:44:34,  1.53it/s]
 43%|████▎     | 28405/65536 [4:52:13<6:34:12,  1.57it/s]
 43%|████▎     | 28406/65536 [4:52:14<6:26:28,  1.60it/s]
 43%|████▎     | 28407/65536 [4:52:14<6:16:52,  1.64it/s]
 43%|████▎     | 28408/65536 [4:52:15<6:12:06,  1.66it/s]
 43%|████▎     | 28409/65536 [4:52:15<6:21:32,  1.62it/s]
 43%|████▎     | 28410/65536 [4:52:16<6:18:21,  1.64it/s]
 43%|████▎     | 28411/65536 [4:52:17<6:21:41,  1.62it/s]
 43%|████▎     | 28412/65536 [4:52:17<6:32:04,  1.58it/s]
 43%|████▎     | 28413/65536 [4:52:18<6:33:45,  1.57it/s]
 43%|████▎     | 28414/65536 [4:52:19<6:30:52,  1.58it/s]
 43%|████▎     | 28415/65536 [4:52:19<6:39:48,  1.55it/s]
 43%|████▎     | 28416/65536 [4:52:20<6:28:25,  1.59it/s]
 43%|████▎     | 28417/65536 [4:52:20<6:13:08,  1.66it/s]
 43%|████▎     | 28418/65536 [4:52:21<6:21:57,  1.62it/s]
 43%|████▎     | 28419/65536 [4:52:22<6:14:39,  1.65it/s]
 43%|████▎     | 28420/65536 [4:52:22<6:22:38,  1.62it/s]
                                                         
{'loss': 1.7035, 'learning_rate': 6.112709685318966e-07, 'epoch': 1754.32}
+
 43%|████▎     | 28420/65536 [4:52:22<6:22:38,  1.62it/s]
 43%|████▎     | 28421/65536 [4:52:23<6:17:40,  1.64it/s]
 43%|████▎     | 28422/65536 [4:52:23<6:19:56,  1.63it/s]
 43%|████▎     | 28423/65536 [4:52:24<6:20:16,  1.63it/s]
 43%|████▎     | 28424/65536 [4:52:25<6:19:58,  1.63it/s]
 43%|████▎     | 28425/65536 [4:52:25<6:15:50,  1.65it/s]
 43%|████▎     | 28426/65536 [4:52:26<6:28:38,  1.59it/s]
 43%|████▎     | 28427/65536 [4:52:27<6:23:47,  1.61it/s]
 43%|████▎     | 28428/65536 [4:52:27<6:34:04,  1.57it/s]
 43%|████▎     | 28429/65536 [4:52:28<6:29:38,  1.59it/s]
 43%|████▎     | 28430/65536 [4:52:28<6:15:20,  1.65it/s]
 43%|████▎     | 28431/65536 [4:52:29<6:09:24,  1.67it/s]
 43%|████▎     | 28432/65536 [4:52:30<6:27:55,  1.59it/s]
 43%|████▎     | 28433/65536 [4:52:30<6:28:15,  1.59it/s]
 43%|████▎     | 28434/65536 [4:52:31<6:24:57,  1.61it/s]
 43%|████▎     | 28435/65536 [4:52:31<6:18:01,  1.64it/s]
 43%|████▎     | 28436/65536 [4:52:32<6:25:09,  1.61it/s]
 43%|████▎     | 28437/65536 [4:52:33<6:24:10,  1.61it/s]
 43%|████▎     | 28438/65536 [4:52:33<6:20:00,  1.63it/s]
 43%|████▎     | 28439/65536 [4:52:34<6:10:49,  1.67it/s]
 43%|████▎     | 28440/65536 [4:52:35<6:15:09,  1.65it/s]
                                                         
{'loss': 1.6506, 'learning_rate': 6.109954695726704e-07, 'epoch': 1755.56}
+
 43%|████▎     | 28440/65536 [4:52:35<6:15:09,  1.65it/s]
 43%|████▎     | 28441/65536 [4:52:35<6:24:32,  1.61it/s]
 43%|████▎     | 28442/65536 [4:52:36<6:32:47,  1.57it/s]
 43%|████▎     | 28443/65536 [4:52:36<6:28:15,  1.59it/s]
 43%|████▎     | 28444/65536 [4:52:37<6:23:47,  1.61it/s]
 43%|████▎     | 28445/65536 [4:52:38<6:22:03,  1.62it/s]
 43%|████▎     | 28446/65536 [4:52:38<6:13:02,  1.66it/s]
 43%|████▎     | 28447/65536 [4:52:39<6:15:51,  1.64it/s]
 43%|████▎     | 28448/65536 [4:52:40<6:29:34,  1.59it/s]
 43%|████▎     | 28449/65536 [4:52:40<6:35:59,  1.56it/s]
 43%|████▎     | 28450/65536 [4:52:41<6:18:15,  1.63it/s]
 43%|████▎     | 28451/65536 [4:52:41<6:11:19,  1.66it/s]
 43%|████▎     | 28452/65536 [4:52:42<6:07:35,  1.68it/s]
 43%|████▎     | 28453/65536 [4:52:43<6:15:13,  1.65it/s]
 43%|████▎     | 28454/65536 [4:52:43<6:35:24,  1.56it/s]
 43%|████▎     | 28455/65536 [4:52:44<6:42:52,  1.53it/s]
 43%|████▎     | 28456/65536 [4:52:45<6:36:14,  1.56it/s]
 43%|████▎     | 28457/65536 [4:52:45<6:17:29,  1.64it/s]
 43%|████▎     | 28458/65536 [4:52:46<6:25:28,  1.60it/s]
 43%|████▎     | 28459/65536 [4:52:46<6:20:09,  1.63it/s]
 43%|████▎     | 28460/65536 [4:52:47<6:29:36,  1.59it/s]
                                                         
{'loss': 1.7308, 'learning_rate': 6.107199706134443e-07, 'epoch': 1756.79}
+
 43%|████▎     | 28460/65536 [4:52:47<6:29:36,  1.59it/s]
 43%|████▎     | 28461/65536 [4:52:48<6:24:19,  1.61it/s]
 43%|████▎     | 28462/65536 [4:52:48<6:26:43,  1.60it/s]
 43%|████▎     | 28463/65536 [4:52:49<6:28:15,  1.59it/s]
 43%|████▎     | 28464/65536 [4:52:50<6:30:27,  1.58it/s]
 43%|████▎     | 28465/65536 [4:52:50<6:34:55,  1.56it/s]
 43%|████▎     | 28466/65536 [4:52:51<6:35:15,  1.56it/s]
 43%|████▎     | 28467/65536 [4:52:51<6:23:41,  1.61it/s]
 43%|████▎     | 28468/65536 [4:52:52<6:34:51,  1.56it/s]
 43%|████▎     | 28469/65536 [4:52:53<6:28:07,  1.59it/s]
 43%|████▎     | 28470/65536 [4:52:53<6:20:02,  1.63it/s]
 43%|████▎     | 28471/65536 [4:52:54<6:16:24,  1.64it/s]
 43%|████▎     | 28472/65536 [4:52:54<6:11:04,  1.66it/s]
 43%|████▎     | 28473/65536 [4:52:55<6:12:05,  1.66it/s]
 43%|████▎     | 28474/65536 [4:52:56<6:12:27,  1.66it/s]
 43%|████▎     | 28475/65536 [4:52:56<6:15:18,  1.65it/s]
 43%|████▎     | 28476/65536 [4:52:57<6:18:21,  1.63it/s]
 43%|████▎     | 28477/65536 [4:52:57<6:14:42,  1.65it/s]
 43%|████▎     | 28478/65536 [4:52:58<6:07:59,  1.68it/s]
 43%|████▎     | 28479/65536 [4:52:59<6:15:02,  1.65it/s]
 43%|████▎     | 28480/65536 [4:52:59<6:28:16,  1.59it/s]
                                                         
{'loss': 1.7238, 'learning_rate': 6.104444716542182e-07, 'epoch': 1758.02}
+
 43%|████▎     | 28480/65536 [4:52:59<6:28:16,  1.59it/s]
 43%|████▎     | 28481/65536 [4:53:00<6:32:33,  1.57it/s]
 43%|████▎     | 28482/65536 [4:53:01<6:20:59,  1.62it/s]
 43%|████▎     | 28483/65536 [4:53:01<6:14:48,  1.65it/s]
 43%|████▎     | 28484/65536 [4:53:02<6:11:51,  1.66it/s]
 43%|████▎     | 28485/65536 [4:53:02<6:07:07,  1.68it/s]
 43%|████▎     | 28486/65536 [4:53:03<6:17:04,  1.64it/s]
 43%|████▎     | 28487/65536 [4:53:04<6:16:47,  1.64it/s]
 43%|████▎     | 28488/65536 [4:53:04<6:20:46,  1.62it/s]
 43%|████▎     | 28489/65536 [4:53:05<6:21:47,  1.62it/s]
 43%|████▎     | 28490/65536 [4:53:05<6:22:30,  1.61it/s]
 43%|████▎     | 28491/65536 [4:53:06<6:22:10,  1.62it/s]
 43%|████▎     | 28492/65536 [4:53:07<6:22:25,  1.61it/s]
 43%|████▎     | 28493/65536 [4:53:07<6:19:06,  1.63it/s]
 43%|████▎     | 28494/65536 [4:53:08<6:14:40,  1.65it/s]
 43%|████▎     | 28495/65536 [4:53:08<6:11:58,  1.66it/s]
 43%|████▎     | 28496/65536 [4:53:09<6:28:35,  1.59it/s]
 43%|████▎     | 28497/65536 [4:53:10<6:22:47,  1.61it/s]
 43%|████▎     | 28498/65536 [4:53:10<6:19:45,  1.63it/s]
 43%|████▎     | 28499/65536 [4:53:11<6:16:25,  1.64it/s]
 43%|████▎     | 28500/65536 [4:53:12<6:12:35,  1.66it/s]
                                                         
{'loss': 1.6749, 'learning_rate': 6.10168972694992e-07, 'epoch': 1759.26}
+
 43%|████▎     | 28500/65536 [4:53:12<6:12:35,  1.66it/s]
 43%|████▎     | 28501/65536 [4:53:12<6:18:47,  1.63it/s]
 43%|████▎     | 28502/65536 [4:53:13<6:17:05,  1.64it/s]
 43%|████▎     | 28503/65536 [4:53:13<6:13:02,  1.65it/s]
 43%|████▎     | 28504/65536 [4:53:14<6:14:06,  1.65it/s]
 43%|████▎     | 28505/65536 [4:53:15<6:21:09,  1.62it/s]
 43%|████▎     | 28506/65536 [4:53:15<6:19:07,  1.63it/s]
 43%|████▎     | 28507/65536 [4:53:16<6:19:39,  1.63it/s]
 43%|████▎     | 28508/65536 [4:53:17<6:26:48,  1.60it/s]
 44%|████▎     | 28509/65536 [4:53:17<6:21:29,  1.62it/s]
 44%|████▎     | 28510/65536 [4:53:18<6:21:05,  1.62it/s]
 44%|████▎     | 28511/65536 [4:53:18<6:21:41,  1.62it/s]
 44%|████▎     | 28512/65536 [4:53:19<6:35:03,  1.56it/s]
 44%|████▎     | 28513/65536 [4:53:20<6:39:11,  1.55it/s]
 44%|████▎     | 28514/65536 [4:53:20<6:31:13,  1.58it/s]
 44%|████▎     | 28515/65536 [4:53:21<6:27:11,  1.59it/s]
 44%|████▎     | 28516/65536 [4:53:22<6:20:40,  1.62it/s]
 44%|████▎     | 28517/65536 [4:53:22<6:21:38,  1.62it/s]
 44%|████▎     | 28518/65536 [4:53:23<6:23:33,  1.61it/s]
 44%|████▎     | 28519/65536 [4:53:23<6:19:26,  1.63it/s]
 44%|████▎     | 28520/65536 [4:53:24<6:31:39,  1.58it/s]
                                                         
{'loss': 1.6481, 'learning_rate': 6.098934737357659e-07, 'epoch': 1760.49}
+
 44%|████▎     | 28520/65536 [4:53:24<6:31:39,  1.58it/s]
 44%|████▎     | 28521/65536 [4:53:25<6:22:20,  1.61it/s]
 44%|████▎     | 28522/65536 [4:53:25<6:15:28,  1.64it/s]
 44%|████▎     | 28523/65536 [4:53:26<6:12:22,  1.66it/s]
 44%|████▎     | 28524/65536 [4:53:26<6:06:27,  1.68it/s]
 44%|████���     | 28525/65536 [4:53:27<6:25:44,  1.60it/s]
 44%|████▎     | 28526/65536 [4:53:28<6:18:28,  1.63it/s]
 44%|████▎     | 28527/65536 [4:53:28<6:20:24,  1.62it/s]
 44%|████▎     | 28528/65536 [4:53:29<6:16:53,  1.64it/s]
 44%|████▎     | 28529/65536 [4:53:30<6:25:08,  1.60it/s]
 44%|████▎     | 28530/65536 [4:53:30<6:27:02,  1.59it/s]
 44%|████▎     | 28531/65536 [4:53:31<6:20:39,  1.62it/s]
 44%|████▎     | 28532/65536 [4:53:31<6:17:34,  1.63it/s]
 44%|████▎     | 28533/65536 [4:53:32<6:20:48,  1.62it/s]
 44%|████▎     | 28534/65536 [4:53:33<6:13:50,  1.65it/s]
 44%|████▎     | 28535/65536 [4:53:33<6:09:00,  1.67it/s]
 44%|████▎     | 28536/65536 [4:53:34<6:09:13,  1.67it/s]
 44%|████▎     | 28537/65536 [4:53:34<6:13:18,  1.65it/s]
 44%|████▎     | 28538/65536 [4:53:35<6:10:00,  1.67it/s]
 44%|████▎     | 28539/65536 [4:53:36<6:22:45,  1.61it/s]
 44%|████▎     | 28540/65536 [4:53:36<6:27:45,  1.59it/s]
                                                         
{'loss': 1.7056, 'learning_rate': 6.096179747765397e-07, 'epoch': 1761.73}
+
 44%|████▎     | 28540/65536 [4:53:36<6:27:45,  1.59it/s]
 44%|████▎     | 28541/65536 [4:53:37<6:18:12,  1.63it/s]
 44%|████▎     | 28542/65536 [4:53:38<6:24:50,  1.60it/s]
 44%|████▎     | 28543/65536 [4:53:38<6:19:42,  1.62it/s]
 44%|████▎     | 28544/65536 [4:53:39<6:18:45,  1.63it/s]
 44%|████▎     | 28545/65536 [4:53:39<6:32:32,  1.57it/s]
 44%|████▎     | 28546/65536 [4:53:40<6:29:23,  1.58it/s]
 44%|████▎     | 28547/65536 [4:53:41<6:20:59,  1.62it/s]
 44%|████▎     | 28548/65536 [4:53:41<6:12:34,  1.65it/s]
 44%|████▎     | 28549/65536 [4:53:42<6:18:23,  1.63it/s]
 44%|████▎     | 28550/65536 [4:53:43<6:26:28,  1.60it/s]
 44%|████▎     | 28551/65536 [4:53:43<6:15:14,  1.64it/s]
 44%|████▎     | 28552/65536 [4:53:44<6:11:35,  1.66it/s]
 44%|████▎     | 28553/65536 [4:53:44<6:23:19,  1.61it/s]
 44%|████▎     | 28554/65536 [4:53:45<6:20:14,  1.62it/s]
 44%|████▎     | 28555/65536 [4:53:46<6:16:43,  1.64it/s]
 44%|████▎     | 28556/65536 [4:53:46<6:19:37,  1.62it/s]
 44%|████▎     | 28557/65536 [4:53:47<6:11:58,  1.66it/s]
 44%|████▎     | 28558/65536 [4:53:47<6:17:01,  1.63it/s]
 44%|████▎     | 28559/65536 [4:53:48<6:12:51,  1.65it/s]
 44%|████▎     | 28560/65536 [4:53:49<6:04:57,  1.69it/s]
                                                         
{'loss': 1.7139, 'learning_rate': 6.093424758173136e-07, 'epoch': 1762.96}
+
 44%|████▎     | 28560/65536 [4:53:49<6:04:57,  1.69it/s]
 44%|████▎     | 28561/65536 [4:53:49<6:23:45,  1.61it/s]
 44%|████▎     | 28562/65536 [4:53:50<6:32:26,  1.57it/s]
 44%|████▎     | 28563/65536 [4:53:51<6:34:00,  1.56it/s]
 44%|████▎     | 28564/65536 [4:53:51<6:28:21,  1.59it/s]
 44%|████▎     | 28565/65536 [4:53:52<6:20:03,  1.62it/s]
 44%|████▎     | 28566/65536 [4:53:52<6:12:12,  1.66it/s]
 44%|████▎     | 28567/65536 [4:53:53<6:09:06,  1.67it/s]
 44%|████▎     | 28568/65536 [4:53:53<6:09:50,  1.67it/s]
 44%|████▎     | 28569/65536 [4:53:54<6:08:16,  1.67it/s]
 44%|████▎     | 28570/65536 [4:53:55<6:03:07,  1.70it/s]
 44%|████▎     | 28571/65536 [4:53:55<6:19:31,  1.62it/s]
 44%|████▎     | 28572/65536 [4:53:56<6:17:56,  1.63it/s]
 44%|████▎     | 28573/65536 [4:53:57<6:14:00,  1.65it/s]
 44%|████▎     | 28574/65536 [4:53:57<6:21:51,  1.61it/s]
 44%|████▎     | 28575/65536 [4:53:58<6:15:06,  1.64it/s]
 44%|████▎     | 28576/65536 [4:53:58<6:12:38,  1.65it/s]
 44%|████▎     | 28577/65536 [4:53:59<6:25:56,  1.60it/s]
 44%|████▎     | 28578/65536 [4:54:00<6:24:02,  1.60it/s]
 44%|████▎     | 28579/65536 [4:54:00<6:26:03,  1.60it/s]
 44%|████▎     | 28580/65536 [4:54:01<6:22:58,  1.61it/s]
                                                         
{'loss': 1.6868, 'learning_rate': 6.090669768580873e-07, 'epoch': 1764.2}
+
 44%|████▎     | 28580/65536 [4:54:01<6:22:58,  1.61it/s]
 44%|████▎     | 28581/65536 [4:54:02<6:24:57,  1.60it/s]
 44%|████▎     | 28582/65536 [4:54:02<6:20:24,  1.62it/s]
 44%|████▎     | 28583/65536 [4:54:03<6:19:57,  1.62it/s]
 44%|████▎     | 28584/65536 [4:54:03<6:07:36,  1.68it/s]
 44%|████▎     | 28585/65536 [4:54:04<6:17:48,  1.63it/s]
 44%|████▎     | 28586/65536 [4:54:05<6:21:55,  1.61it/s]
 44%|████▎     | 28587/65536 [4:54:05<6:32:16,  1.57it/s]
 44%|████▎     | 28588/65536 [4:54:06<6:20:46,  1.62it/s]
 44%|████▎     | 28589/65536 [4:54:06<6:13:52,  1.65it/s]
 44%|████▎     | 28590/65536 [4:54:07<6:17:23,  1.63it/s]
 44%|████▎     | 28591/65536 [4:54:08<6:03:48,  1.69it/s]
 44%|████▎     | 28592/65536 [4:54:08<5:59:43,  1.71it/s]
 44%|████▎     | 28593/65536 [4:54:09<6:05:17,  1.69it/s]
 44%|████▎     | 28594/65536 [4:54:09<6:27:20,  1.59it/s]
 44%|████▎     | 28595/65536 [4:54:10<6:15:00,  1.64it/s]
 44%|████▎     | 28596/65536 [4:54:11<6:20:10,  1.62it/s]
 44%|████▎     | 28597/65536 [4:54:11<6:26:00,  1.59it/s]
 44%|████▎     | 28598/65536 [4:54:12<6:28:31,  1.58it/s]
 44%|████▎     | 28599/65536 [4:54:13<6:14:23,  1.64it/s]
 44%|████▎     | 28600/65536 [4:54:13<6:11:21,  1.66it/s]
                                                         
{'loss': 1.7287, 'learning_rate': 6.087914778988613e-07, 'epoch': 1765.43}
+
 44%|████▎     | 28600/65536 [4:54:13<6:11:21,  1.66it/s]
 44%|████▎     | 28601/65536 [4:54:14<6:16:41,  1.63it/s]
 44%|████▎     | 28602/65536 [4:54:14<6:20:54,  1.62it/s]
 44%|████▎     | 28603/65536 [4:54:15<6:15:02,  1.64it/s]
 44%|████▎     | 28604/65536 [4:54:16<6:12:04,  1.65it/s]
 44%|████▎     | 28605/65536 [4:54:16<6:03:42,  1.69it/s]
 44%|████▎     | 28606/65536 [4:54:17<6:09:08,  1.67it/s]
 44%|████▎     | 28607/65536 [4:54:17<6:13:57,  1.65it/s]
 44%|████▎     | 28608/65536 [4:54:18<6:17:27,  1.63it/s]
 44%|████▎     | 28609/65536 [4:54:19<6:11:56,  1.65it/s]
 44%|████▎     | 28610/65536 [4:54:19<6:27:31,  1.59it/s]
 44%|████▎     | 28611/65536 [4:54:20<6:19:47,  1.62it/s]
 44%|████▎     | 28612/65536 [4:54:20<6:10:10,  1.66it/s]
 44%|████▎     | 28613/65536 [4:54:21<6:14:04,  1.65it/s]
 44%|████▎     | 28614/65536 [4:54:22<6:29:56,  1.58it/s]
 44%|████▎     | 28615/65536 [4:54:22<6:22:01,  1.61it/s]
 44%|████▎     | 28616/65536 [4:54:23<6:11:49,  1.65it/s]
 44%|████▎     | 28617/65536 [4:54:24<6:20:00,  1.62it/s]
 44%|████▎     | 28618/65536 [4:54:24<6:24:27,  1.60it/s]
 44%|████▎     | 28619/65536 [4:54:25<6:36:19,  1.55it/s]
 44%|████▎     | 28620/65536 [4:54:25<6:24:50,  1.60it/s]
                                                         
{'loss': 1.676, 'learning_rate': 6.085159789396351e-07, 'epoch': 1766.67}
+
 44%|████▎     | 28620/65536 [4:54:25<6:24:50,  1.60it/s]
 44%|████▎     | 28621/65536 [4:54:26<6:23:10,  1.61it/s]
 44%|████▎     | 28622/65536 [4:54:27<6:17:39,  1.63it/s]
 44%|████▎     | 28623/65536 [4:54:27<6:17:06,  1.63it/s]
 44%|████▎     | 28624/65536 [4:54:28<6:21:38,  1.61it/s]
 44%|████▎     | 28625/65536 [4:54:28<6:14:12,  1.64it/s]
 44%|████▎     | 28626/65536 [4:54:29<6:31:25,  1.57it/s]
 44%|████▎     | 28627/65536 [4:54:30<6:36:06,  1.55it/s]
 44%|████▎     | 28628/65536 [4:54:30<6:33:12,  1.56it/s]
 44%|████▎     | 28629/65536 [4:54:31<6:17:24,  1.63it/s]
 44%|████▎     | 28630/65536 [4:54:32<6:21:39,  1.61it/s]
 44%|████▎     | 28631/65536 [4:54:32<6:18:57,  1.62it/s]
 44%|████▎     | 28632/65536 [4:54:33<6:11:53,  1.65it/s]
 44%|████▎     | 28633/65536 [4:54:33<6:14:49,  1.64it/s]
 44%|████▎     | 28634/65536 [4:54:34<6:12:41,  1.65it/s]
 44%|████▎     | 28635/65536 [4:54:35<6:11:55,  1.65it/s]
 44%|████▎     | 28636/65536 [4:54:35<6:19:09,  1.62it/s]
 44%|████▎     | 28637/65536 [4:54:36<6:14:23,  1.64it/s]
 44%|████▎     | 28638/65536 [4:54:37<6:14:30,  1.64it/s]
 44%|████▎     | 28639/65536 [4:54:37<6:08:22,  1.67it/s]
 44%|████▎     | 28640/65536 [4:54:38<6:05:22,  1.68it/s]
                                                         
{'loss': 1.7063, 'learning_rate': 6.082404799804089e-07, 'epoch': 1767.9}
+
 44%|████▎     | 28640/65536 [4:54:38<6:05:22,  1.68it/s]
 44%|████▎     | 28641/65536 [4:54:38<6:09:12,  1.67it/s]
 44%|████▎     | 28642/65536 [4:54:39<6:20:42,  1.62it/s]
 44%|████▎     | 28643/65536 [4:54:40<6:12:21,  1.65it/s]
 44%|████▎     | 28644/65536 [4:54:40<6:13:31,  1.65it/s]
 44%|████▎     | 28645/65536 [4:54:41<6:14:28,  1.64it/s]
 44%|████▎     | 28646/65536 [4:54:41<6:12:24,  1.65it/s]
 44%|████▎     | 28647/65536 [4:54:42<6:05:24,  1.68it/s]
 44%|████▎     | 28648/65536 [4:54:43<6:11:01,  1.66it/s]
 44%|████▎     | 28649/65536 [4:54:43<6:20:48,  1.61it/s]
 44%|████▎     | 28650/65536 [4:54:44<6:12:32,  1.65it/s]
 44%|████▎     | 28651/65536 [4:54:44<6:35:27,  1.55it/s]
 44%|████▎     | 28652/65536 [4:54:45<6:28:16,  1.58it/s]
 44%|████▎     | 28653/65536 [4:54:46<6:25:46,  1.59it/s]
 44%|████▎     | 28654/65536 [4:54:46<6:21:30,  1.61it/s]
 44%|████▎     | 28655/65536 [4:54:47<6:17:18,  1.63it/s]
 44%|████▎     | 28656/65536 [4:54:47<6:09:20,  1.66it/s]
 44%|████▎     | 28657/65536 [4:54:48<6:14:07,  1.64it/s]
 44%|████▎     | 28658/65536 [4:54:49<6:24:36,  1.60it/s]
 44%|████▎     | 28659/65536 [4:54:49<6:15:55,  1.63it/s]
 44%|████▎     | 28660/65536 [4:54:50<6:26:35,  1.59it/s]
                                                         
{'loss': 1.7094, 'learning_rate': 6.079649810211828e-07, 'epoch': 1769.14}
+
 44%|████▎     | 28660/65536 [4:54:50<6:26:35,  1.59it/s]
 44%|████▎     | 28661/65536 [4:54:51<6:13:15,  1.65it/s]
 44%|████▎     | 28662/65536 [4:54:51<6:09:19,  1.66it/s]
 44%|████▎     | 28663/65536 [4:54:52<6:11:19,  1.66it/s]
 44%|████▎     | 28664/65536 [4:54:52<6:18:43,  1.62it/s]
 44%|████▎     | 28665/65536 [4:54:53<6:16:22,  1.63it/s]
 44%|████▎     | 28666/65536 [4:54:54<6:14:23,  1.64it/s]
 44%|████▎     | 28667/65536 [4:54:54<6:12:11,  1.65it/s]
 44%|████▎     | 28668/65536 [4:54:55<6:09:26,  1.66it/s]
 44%|████▎     | 28669/65536 [4:54:55<6:20:00,  1.62it/s]
 44%|████▎     | 28670/65536 [4:54:56<6:15:36,  1.64it/s]
 44%|████▎     | 28671/65536 [4:54:57<6:13:21,  1.65it/s]
 44%|████▍     | 28672/65536 [4:54:57<6:13:01,  1.65it/s]
 44%|████▍     | 28673/65536 [4:54:58<6:26:00,  1.59it/s]
 44%|████▍     | 28674/65536 [4:54:59<6:16:30,  1.63it/s]
 44%|████▍     | 28675/65536 [4:54:59<6:25:02,  1.60it/s]
 44%|████▍     | 28676/65536 [4:55:00<6:14:32,  1.64it/s]
 44%|████▍     | 28677/65536 [4:55:00<6:18:21,  1.62it/s]
 44%|████▍     | 28678/65536 [4:55:01<6:15:04,  1.64it/s]
 44%|████▍     | 28679/65536 [4:55:02<6:18:38,  1.62it/s]
 44%|████▍     | 28680/65536 [4:55:02<6:14:02,  1.64it/s]
                                                         
{'loss': 1.7092, 'learning_rate': 6.076894820619566e-07, 'epoch': 1770.37}
+
 44%|████▍     | 28680/65536 [4:55:02<6:14:02,  1.64it/s]
 44%|████▍     | 28681/65536 [4:55:03<6:13:46,  1.64it/s]
 44%|████▍     | 28682/65536 [4:55:03<6:17:44,  1.63it/s]
 44%|████▍     | 28683/65536 [4:55:04<6:15:27,  1.64it/s]
 44%|████▍     | 28684/65536 [4:55:05<6:12:06,  1.65it/s]
 44%|████▍     | 28685/65536 [4:55:05<6:11:13,  1.65it/s]
 44%|████▍     | 28686/65536 [4:55:06<6:12:37,  1.65it/s]
 44%|████▍     | 28687/65536 [4:55:06<6:11:04,  1.66it/s]
 44%|████▍     | 28688/65536 [4:55:07<6:03:01,  1.69it/s]
 44%|████▍     | 28689/65536 [4:55:08<6:11:16,  1.65it/s]
 44%|████▍     | 28690/65536 [4:55:08<6:20:32,  1.61it/s]
 44%|████▍     | 28691/65536 [4:55:09<6:38:49,  1.54it/s]
 44%|████▍     | 28692/65536 [4:55:10<6:20:45,  1.61it/s]
 44%|████▍     | 28693/65536 [4:55:10<6:16:21,  1.63it/s]
 44%|████▍     | 28694/65536 [4:55:11<6:11:58,  1.65it/s]
 44%|████▍     | 28695/65536 [4:55:11<6:07:06,  1.67it/s]
 44%|████▍     | 28696/65536 [4:55:12<6:11:59,  1.65it/s]
 44%|████▍     | 28697/65536 [4:55:13<6:17:01,  1.63it/s]
 44%|████▍     | 28698/65536 [4:55:13<6:10:58,  1.65it/s]
 44%|████▍     | 28699/65536 [4:55:14<6:07:49,  1.67it/s]
 44%|████▍     | 28700/65536 [4:55:14<6:10:30,  1.66it/s]
                                                         
{'loss': 1.7185, 'learning_rate': 6.074139831027304e-07, 'epoch': 1771.6}
+
 44%|████▍     | 28700/65536 [4:55:14<6:10:30,  1.66it/s]
 44%|████▍     | 28701/65536 [4:55:15<6:20:49,  1.61it/s]
 44%|████▍     | 28702/65536 [4:55:16<6:23:34,  1.60it/s]
 44%|████▍     | 28703/65536 [4:55:16<6:17:32,  1.63it/s]
 44%|████▍     | 28704/65536 [4:55:17<6:16:04,  1.63it/s]
 44%|████▍     | 28705/65536 [4:55:18<6:19:56,  1.62it/s]
 44%|████▍     | 28706/65536 [4:55:18<6:25:15,  1.59it/s]
 44%|████▍     | 28707/65536 [4:55:19<6:36:46,  1.55it/s]
 44%|████▍     | 28708/65536 [4:55:19<6:32:42,  1.56it/s]
 44%|████▍     | 28709/65536 [4:55:20<6:28:59,  1.58it/s]
 44%|████▍     | 28710/65536 [4:55:21<6:25:21,  1.59it/s]
 44%|████▍     | 28711/65536 [4:55:21<6:16:04,  1.63it/s]
 44%|████▍     | 28712/65536 [4:55:22<6:18:46,  1.62it/s]
 44%|████▍     | 28713/65536 [4:55:23<6:12:57,  1.65it/s]
 44%|████▍     | 28714/65536 [4:55:23<6:18:21,  1.62it/s]
 44%|████▍     | 28715/65536 [4:55:24<6:06:38,  1.67it/s]
 44%|████▍     | 28716/65536 [4:55:24<6:12:33,  1.65it/s]
 44%|████▍     | 28717/65536 [4:55:25<6:09:45,  1.66it/s]
 44%|████▍     | 28718/65536 [4:55:26<6:12:13,  1.65it/s]
 44%|████▍     | 28719/65536 [4:55:26<6:21:04,  1.61it/s]
 44%|████▍     | 28720/65536 [4:55:27<6:15:02,  1.64it/s]
                                                         
{'loss': 1.666, 'learning_rate': 6.071384841435042e-07, 'epoch': 1772.84}
+
 44%|████▍     | 28720/65536 [4:55:27<6:15:02,  1.64it/s]
 44%|████▍     | 28721/65536 [4:55:27<6:11:46,  1.65it/s]
 44%|████▍     | 28722/65536 [4:55:28<6:07:37,  1.67it/s]
 44%|████▍     | 28723/65536 [4:55:29<6:30:56,  1.57it/s]
 44%|████▍     | 28724/65536 [4:55:29<6:31:13,  1.57it/s]
 44%|████▍     | 28725/65536 [4:55:30<6:25:11,  1.59it/s]
 44%|████▍     | 28726/65536 [4:55:31<6:22:01,  1.61it/s]
 44%|████▍     | 28727/65536 [4:55:31<6:19:35,  1.62it/s]
 44%|████▍     | 28728/65536 [4:55:32<6:15:59,  1.63it/s]
 44%|████▍     | 28729/65536 [4:55:32<6:06:21,  1.67it/s]
 44%|████▍     | 28730/65536 [4:55:33<6:08:27,  1.66it/s]
 44%|████▍     | 28731/65536 [4:55:33<6:02:05,  1.69it/s]
 44%|████▍     | 28732/65536 [4:55:34<6:10:28,  1.66it/s]
 44%|████▍     | 28733/65536 [4:55:35<6:19:42,  1.62it/s]
 44%|████▍     | 28734/65536 [4:55:35<6:26:50,  1.59it/s]
 44%|████▍     | 28735/65536 [4:55:36<6:37:39,  1.54it/s]
 44%|████▍     | 28736/65536 [4:55:37<6:33:55,  1.56it/s]
 44%|████▍     | 28737/65536 [4:55:37<6:24:15,  1.60it/s]
 44%|████▍     | 28738/65536 [4:55:38<6:12:17,  1.65it/s]
 44%|████▍     | 28739/65536 [4:55:39<6:19:55,  1.61it/s]
 44%|████▍     | 28740/65536 [4:55:39<6:09:37,  1.66it/s]
                                                         
{'loss': 1.7145, 'learning_rate': 6.068629851842781e-07, 'epoch': 1774.07}
+
 44%|████▍     | 28740/65536 [4:55:39<6:09:37,  1.66it/s]
 44%|████▍     | 28741/65536 [4:55:40<6:02:10,  1.69it/s]
 44%|████▍     | 28742/65536 [4:55:40<6:08:12,  1.67it/s]
 44%|████▍     | 28743/65536 [4:55:41<6:07:43,  1.67it/s]
 44%|████▍     | 28744/65536 [4:55:42<6:13:28,  1.64it/s]
 44%|████▍     | 28745/65536 [4:55:42<6:12:57,  1.64it/s]
 44%|████▍     | 28746/65536 [4:55:43<6:04:18,  1.68it/s]
 44%|████▍     | 28747/65536 [4:55:43<6:11:52,  1.65it/s]
 44%|████▍     | 28748/65536 [4:55:44<6:06:35,  1.67it/s]
 44%|████▍     | 28749/65536 [4:55:45<6:09:10,  1.66it/s]
 44%|████▍     | 28750/65536 [4:55:45<6:06:34,  1.67it/s]
 44%|████▍     | 28751/65536 [4:55:46<6:16:07,  1.63it/s]
 44%|████▍     | 28752/65536 [4:55:46<6:15:10,  1.63it/s]
 44%|████▍     | 28753/65536 [4:55:47<6:17:58,  1.62it/s]
 44%|████▍     | 28754/65536 [4:55:48<6:21:23,  1.61it/s]
 44%|████▍     | 28755/65536 [4:55:48<6:23:44,  1.60it/s]
 44%|████▍     | 28756/65536 [4:55:49<6:23:33,  1.60it/s]
 44%|████▍     | 28757/65536 [4:55:50<6:23:54,  1.60it/s]
 44%|████▍     | 28758/65536 [4:55:50<6:17:52,  1.62it/s]
 44%|████▍     | 28759/65536 [4:55:51<6:13:38,  1.64it/s]
 44%|████▍     | 28760/65536 [4:55:51<6:17:24,  1.62it/s]
                                                         
{'loss': 1.7243, 'learning_rate': 6.06587486225052e-07, 'epoch': 1775.31}
+
 44%|████▍     | 28760/65536 [4:55:51<6:17:24,  1.62it/s]
 44%|████▍     | 28761/65536 [4:55:52<6:11:42,  1.65it/s]
 44%|████▍     | 28762/65536 [4:55:53<6:22:30,  1.60it/s]
 44%|████▍     | 28763/65536 [4:55:53<6:29:24,  1.57it/s]
 44%|████▍     | 28764/65536 [4:55:54<6:23:54,  1.60it/s]
 44%|████▍     | 28765/65536 [4:55:54<6:25:59,  1.59it/s]
 44%|████▍     | 28766/65536 [4:55:55<6:31:02,  1.57it/s]
 44%|████▍     | 28767/65536 [4:55:56<6:23:24,  1.60it/s]
 44%|████▍     | 28768/65536 [4:55:56<6:24:21,  1.59it/s]
 44%|████▍     | 28769/65536 [4:55:57<6:16:54,  1.63it/s]
 44%|████▍     | 28770/65536 [4:55:58<6:13:28,  1.64it/s]
 44%|████▍     | 28771/65536 [4:55:58<6:07:45,  1.67it/s]
 44%|████▍     | 28772/65536 [4:55:59<6:29:32,  1.57it/s]
 44%|████▍     | 28773/65536 [4:55:59<6:28:33,  1.58it/s]
 44%|████▍     | 28774/65536 [4:56:00<6:15:09,  1.63it/s]
 44%|████▍     | 28775/65536 [4:56:01<6:15:18,  1.63it/s]
 44%|████▍     | 28776/65536 [4:56:01<6:11:10,  1.65it/s]
 44%|████▍     | 28777/65536 [4:56:02<6:16:06,  1.63it/s]
 44%|████▍     | 28778/65536 [4:56:03<6:20:16,  1.61it/s]
 44%|████▍     | 28779/65536 [4:56:03<6:15:33,  1.63it/s]
 44%|████▍     | 28780/65536 [4:56:04<6:07:23,  1.67it/s]
                                                         
{'loss': 1.7053, 'learning_rate': 6.063119872658258e-07, 'epoch': 1776.54}
+
 44%|████▍     | 28780/65536 [4:56:04<6:07:23,  1.67it/s]
 44%|████▍     | 28781/65536 [4:56:04<6:12:37,  1.64it/s]
 44%|████▍     | 28782/65536 [4:56:05<6:14:09,  1.64it/s]
 44%|████▍     | 28783/65536 [4:56:06<6:16:20,  1.63it/s]
 44%|████▍     | 28784/65536 [4:56:06<6:17:58,  1.62it/s]
 44%|████▍     | 28785/65536 [4:56:07<6:13:49,  1.64it/s]
 44%|████▍     | 28786/65536 [4:56:07<6:16:56,  1.62it/s]
 44%|████▍     | 28787/65536 [4:56:08<6:08:35,  1.66it/s]
 44%|████▍     | 28788/65536 [4:56:09<6:28:04,  1.58it/s]
 44%|████▍     | 28789/65536 [4:56:09<6:25:04,  1.59it/s]
 44%|████▍     | 28790/65536 [4:56:10<6:20:39,  1.61it/s]
 44%|████▍     | 28791/65536 [4:56:11<6:21:42,  1.60it/s]
 44%|████▍     | 28792/65536 [4:56:11<6:21:48,  1.60it/s]
 44%|████▍     | 28793/65536 [4:56:12<6:13:35,  1.64it/s]
 44%|████▍     | 28794/65536 [4:56:12<6:13:30,  1.64it/s]
 44%|████▍     | 28795/65536 [4:56:13<6:05:37,  1.67it/s]
 44%|████▍     | 28796/65536 [4:56:13<6:02:36,  1.69it/s]
 44%|████▍     | 28797/65536 [4:56:14<5:57:27,  1.71it/s]
 44%|████▍     | 28798/65536 [4:56:15<5:53:11,  1.73it/s]
 44%|████▍     | 28799/65536 [4:56:15<5:59:51,  1.70it/s]
 44%|████▍     | 28800/65536 [4:56:16<6:03:02,  1.69it/s]
                                                         
{'loss': 1.7349, 'learning_rate': 6.060364883065997e-07, 'epoch': 1777.78}
+
 44%|████▍     | 28800/65536 [4:56:16<6:03:02,  1.69it/s]
 44%|████▍     | 28801/65536 [4:56:16<6:09:05,  1.66it/s]
 44%|████▍     | 28802/65536 [4:56:17<6:06:16,  1.67it/s]
 44%|████▍     | 28803/65536 [4:56:18<6:23:41,  1.60it/s]
 44%|████▍     | 28804/65536 [4:56:18<6:37:21,  1.54it/s]
 44%|████▍     | 28805/65536 [4:56:19<6:43:02,  1.52it/s]
 44%|████▍     | 28806/65536 [4:56:20<6:35:52,  1.55it/s]
 44%|████▍     | 28807/65536 [4:56:20<6:29:04,  1.57it/s]
 44%|████▍     | 28808/65536 [4:56:21<6:15:38,  1.63it/s]
 44%|████▍     | 28809/65536 [4:56:22<6:23:10,  1.60it/s]
 44%|████▍     | 28810/65536 [4:56:22<6:31:42,  1.56it/s]
 44%|████▍     | 28811/65536 [4:56:23<6:26:12,  1.58it/s]
 44%|████▍     | 28812/65536 [4:56:23<6:23:54,  1.59it/s]
 44%|████▍     | 28813/65536 [4:56:24<6:15:25,  1.63it/s]
 44%|████▍     | 28814/65536 [4:56:25<6:08:19,  1.66it/s]
 44%|████▍     | 28815/65536 [4:56:25<6:11:33,  1.65it/s]
 44%|████▍     | 28816/65536 [4:56:26<6:15:19,  1.63it/s]
 44%|████▍     | 28817/65536 [4:56:26<6:11:16,  1.65it/s]
 44%|████▍     | 28818/65536 [4:56:27<6:06:59,  1.67it/s]
 44%|████▍     | 28819/65536 [4:56:28<6:12:21,  1.64it/s]
 44%|████▍     | 28820/65536 [4:56:28<6:15:25,  1.63it/s]
                                                         
{'loss': 1.7217, 'learning_rate': 6.057609893473735e-07, 'epoch': 1779.01}
+
 44%|████▍     | 28820/65536 [4:56:28<6:15:25,  1.63it/s]
 44%|████▍     | 28821/65536 [4:56:29<6:22:00,  1.60it/s]
 44%|████▍     | 28822/65536 [4:56:30<6:18:58,  1.61it/s]
 44%|████▍     | 28823/65536 [4:56:30<6:19:36,  1.61it/s]
 44%|████▍     | 28824/65536 [4:56:31<6:21:30,  1.60it/s]
 44%|████▍     | 28825/65536 [4:56:31<6:18:35,  1.62it/s]
 44%|████▍     | 28826/65536 [4:56:32<6:15:47,  1.63it/s]
 44%|████▍     | 28827/65536 [4:56:33<6:11:49,  1.65it/s]
 44%|████▍     | 28828/65536 [4:56:33<6:03:34,  1.68it/s]
 44%|████▍     | 28829/65536 [4:56:34<6:06:34,  1.67it/s]
 44%|████▍     | 28830/65536 [4:56:34<6:07:39,  1.66it/s]
 44%|████▍     | 28831/65536 [4:56:35<6:16:25,  1.63it/s]
 44%|████▍     | 28832/65536 [4:56:36<6:15:13,  1.63it/s]
 44%|████▍     | 28833/65536 [4:56:36<6:21:15,  1.60it/s]
 44%|████▍     | 28834/65536 [4:56:37<6:10:51,  1.65it/s]
 44%|████▍     | 28835/65536 [4:56:37<6:05:23,  1.67it/s]
 44%|████▍     | 28836/65536 [4:56:38<6:16:55,  1.62it/s]
 44%|████▍     | 28837/65536 [4:56:39<6:36:01,  1.54it/s]
 44%|████▍     | 28838/65536 [4:56:39<6:35:26,  1.55it/s]
 44%|████▍     | 28839/65536 [4:56:40<6:32:25,  1.56it/s]
 44%|████▍     | 28840/65536 [4:56:41<6:23:06,  1.60it/s]
                                                         
{'loss': 1.6761, 'learning_rate': 6.054854903881474e-07, 'epoch': 1780.25}
+
 44%|████▍     | 28840/65536 [4:56:41<6:23:06,  1.60it/s]
 44%|████▍     | 28841/65536 [4:56:41<6:25:39,  1.59it/s]
 44%|████▍     | 28842/65536 [4:56:42<6:17:06,  1.62it/s]
 44%|████▍     | 28843/65536 [4:56:43<6:19:45,  1.61it/s]
 44%|████▍     | 28844/65536 [4:56:43<6:13:16,  1.64it/s]
 44%|████▍     | 28845/65536 [4:56:44<6:17:34,  1.62it/s]
 44%|████▍     | 28846/65536 [4:56:44<6:10:40,  1.65it/s]
 44%|████▍     | 28847/65536 [4:56:45<6:12:58,  1.64it/s]
 44%|████▍     | 28848/65536 [4:56:46<6:19:00,  1.61it/s]
 44%|████▍     | 28849/65536 [4:56:46<6:07:32,  1.66it/s]
 44%|████▍     | 28850/65536 [4:56:47<6:10:16,  1.65it/s]
 44%|████▍     | 28851/65536 [4:56:47<6:09:10,  1.66it/s]
 44%|████▍     | 28852/65536 [4:56:48<6:09:57,  1.65it/s]
 44%|████▍     | 28853/65536 [4:56:49<6:28:51,  1.57it/s]
 44%|████▍     | 28854/65536 [4:56:49<6:20:45,  1.61it/s]
 44%|████▍     | 28855/65536 [4:56:50<6:23:42,  1.59it/s]
 44%|████▍     | 28856/65536 [4:56:51<6:20:03,  1.61it/s]
 44%|████▍     | 28857/65536 [4:56:51<6:30:23,  1.57it/s]
 44%|████▍     | 28858/65536 [4:56:52<6:18:30,  1.62it/s]
 44%|████▍     | 28859/65536 [4:56:52<6:13:33,  1.64it/s]
 44%|████▍     | 28860/65536 [4:56:53<6:17:19,  1.62it/s]
                                                         
{'loss': 1.6951, 'learning_rate': 6.052099914289213e-07, 'epoch': 1781.48}
+
 44%|████▍     | 28860/65536 [4:56:53<6:17:19,  1.62it/s]
 44%|████▍     | 28861/65536 [4:56:54<6:21:35,  1.60it/s]
 44%|████▍     | 28862/65536 [4:56:54<6:21:32,  1.60it/s]
 44%|████▍     | 28863/65536 [4:56:55<6:21:02,  1.60it/s]
 44%|████▍     | 28864/65536 [4:56:55<6:17:30,  1.62it/s]
 44%|████▍     | 28865/65536 [4:56:56<6:14:12,  1.63it/s]
 44%|████▍     | 28866/65536 [4:56:57<6:13:17,  1.64it/s]
 44%|████▍     | 28867/65536 [4:56:57<6:09:30,  1.65it/s]
 44%|████▍     | 28868/65536 [4:56:58<6:05:46,  1.67it/s]
 44%|████▍     | 28869/65536 [4:56:59<6:24:52,  1.59it/s]
 44%|████▍     | 28870/65536 [4:56:59<6:20:39,  1.61it/s]
 44%|████▍     | 28871/65536 [4:57:00<6:26:11,  1.58it/s]
 44%|████▍     | 28872/65536 [4:57:00<6:27:27,  1.58it/s]
 44%|████▍     | 28873/65536 [4:57:01<6:29:22,  1.57it/s]
 44%|████▍     | 28874/65536 [4:57:02<6:21:19,  1.60it/s]
 44%|████▍     | 28875/65536 [4:57:02<6:31:41,  1.56it/s]
 44%|████▍     | 28876/65536 [4:57:03<6:36:45,  1.54it/s]
 44%|████▍     | 28877/65536 [4:57:04<6:28:26,  1.57it/s]
 44%|████▍     | 28878/65536 [4:57:04<6:24:13,  1.59it/s]
 44%|████▍     | 28879/65536 [4:57:05<6:14:16,  1.63it/s]
 44%|████▍     | 28880/65536 [4:57:05<6:11:01,  1.65it/s]
                                                         
{'loss': 1.7104, 'learning_rate': 6.049344924696951e-07, 'epoch': 1782.72}
+
 44%|████▍     | 28880/65536 [4:57:05<6:11:01,  1.65it/s]
 44%|████▍     | 28881/65536 [4:57:06<6:15:35,  1.63it/s]
 44%|████▍     | 28882/65536 [4:57:07<6:11:24,  1.64it/s]
 44%|████▍     | 28883/65536 [4:57:07<6:07:46,  1.66it/s]
 44%|████▍     | 28884/65536 [4:57:08<6:17:56,  1.62it/s]
 44%|████▍     | 28885/65536 [4:57:09<6:30:44,  1.56it/s]
 44%|████▍     | 28886/65536 [4:57:09<6:26:26,  1.58it/s]
 44%|████▍     | 28887/65536 [4:57:10<6:19:32,  1.61it/s]
 44%|████▍     | 28888/65536 [4:57:10<6:17:35,  1.62it/s]
 44%|████▍     | 28889/65536 [4:57:11<6:22:45,  1.60it/s]
 44%|████▍     | 28890/65536 [4:57:12<6:13:39,  1.63it/s]
 44%|████▍     | 28891/65536 [4:57:12<6:23:53,  1.59it/s]
 44%|████▍     | 28892/65536 [4:57:13<6:18:39,  1.61it/s]
 44%|████▍     | 28893/65536 [4:57:14<6:30:37,  1.56it/s]
 44%|████▍     | 28894/65536 [4:57:14<6:20:23,  1.61it/s]
 44%|████▍     | 28895/65536 [4:57:15<6:17:11,  1.62it/s]
 44%|████▍     | 28896/65536 [4:57:15<6:19:24,  1.61it/s]
 44%|████▍     | 28897/65536 [4:57:16<6:23:00,  1.59it/s]
 44%|████▍     | 28898/65536 [4:57:17<6:15:49,  1.62it/s]
 44%|████▍     | 28899/65536 [4:57:17<6:07:44,  1.66it/s]
 44%|████▍     | 28900/65536 [4:57:18<6:05:29,  1.67it/s]
                                                         
{'loss': 1.6784, 'learning_rate': 6.04658993510469e-07, 'epoch': 1783.95}
+
 44%|████▍     | 28900/65536 [4:57:18<6:05:29,  1.67it/s]
 44%|████▍     | 28901/65536 [4:57:18<6:20:43,  1.60it/s]
 44%|████▍     | 28902/65536 [4:57:19<6:16:16,  1.62it/s]
 44%|████▍     | 28903/65536 [4:57:20<6:16:39,  1.62it/s]
 44%|████▍     | 28904/65536 [4:57:20<6:10:44,  1.65it/s]
 44%|████▍     | 28905/65536 [4:57:21<6:12:21,  1.64it/s]
 44%|████▍     | 28906/65536 [4:57:22<6:12:52,  1.64it/s]
 44%|████▍     | 28907/65536 [4:57:22<6:11:05,  1.65it/s]
 44%|████▍     | 28908/65536 [4:57:23<6:05:58,  1.67it/s]
 44%|████▍     | 28909/65536 [4:57:23<5:59:17,  1.70it/s]
 44%|████▍     | 28910/65536 [4:57:24<6:04:53,  1.67it/s]
 44%|████▍     | 28911/65536 [4:57:24<6:03:25,  1.68it/s]
 44%|████▍     | 28912/65536 [4:57:25<6:09:32,  1.65it/s]
 44%|████▍     | 28913/65536 [4:57:26<6:16:31,  1.62it/s]
 44%|████▍     | 28914/65536 [4:57:26<6:19:34,  1.61it/s]
 44%|████▍     | 28915/65536 [4:57:27<6:28:28,  1.57it/s]
 44%|████▍     | 28916/65536 [4:57:28<6:27:16,  1.58it/s]
 44%|████▍     | 28917/65536 [4:57:28<6:21:16,  1.60it/s]
 44%|████▍     | 28918/65536 [4:57:29<6:28:33,  1.57it/s]
 44%|████▍     | 28919/65536 [4:57:30<6:25:47,  1.58it/s]
 44%|████▍     | 28920/65536 [4:57:30<6:28:33,  1.57it/s]
                                                         
{'loss': 1.6877, 'learning_rate': 6.043834945512428e-07, 'epoch': 1785.19}
+
 44%|████▍     | 28920/65536 [4:57:30<6:28:33,  1.57it/s]
 44%|████▍     | 28921/65536 [4:57:31<6:21:44,  1.60it/s]
 44%|████▍     | 28922/65536 [4:57:31<6:30:08,  1.56it/s]
 44%|████▍     | 28923/65536 [4:57:32<6:19:35,  1.61it/s]
 44%|████▍     | 28924/65536 [4:57:33<6:13:08,  1.64it/s]
 44%|████▍     | 28925/65536 [4:57:33<6:08:14,  1.66it/s]
 44%|████▍     | 28926/65536 [4:57:34<6:06:19,  1.67it/s]
 44%|████▍     | 28927/65536 [4:57:34<6:07:53,  1.66it/s]
 44%|████▍     | 28928/65536 [4:57:35<6:08:18,  1.66it/s]
 44%|████▍     | 28929/65536 [4:57:36<6:14:28,  1.63it/s]
 44%|████▍     | 28930/65536 [4:57:36<6:14:09,  1.63it/s]
 44%|████▍     | 28931/65536 [4:57:37<6:11:50,  1.64it/s]
 44%|████▍     | 28932/65536 [4:57:37<6:02:52,  1.68it/s]
 44%|████▍     | 28933/65536 [4:57:38<6:09:19,  1.65it/s]
 44%|████▍     | 28934/65536 [4:57:39<6:25:21,  1.58it/s]
 44%|████▍     | 28935/65536 [4:57:39<6:17:58,  1.61it/s]
 44%|████▍     | 28936/65536 [4:57:40<6:11:11,  1.64it/s]
 44%|████▍     | 28937/65536 [4:57:41<6:06:51,  1.66it/s]
 44%|████▍     | 28938/65536 [4:57:41<6:04:00,  1.68it/s]
 44%|████▍     | 28939/65536 [4:57:42<6:09:48,  1.65it/s]
 44%|████▍     | 28940/65536 [4:57:42<6:18:36,  1.61it/s]
                                                         
{'loss': 1.6964, 'learning_rate': 6.041079955920167e-07, 'epoch': 1786.42}
+
 44%|████▍     | 28940/65536 [4:57:42<6:18:36,  1.61it/s]
 44%|████▍     | 28941/65536 [4:57:43<6:20:37,  1.60it/s]
 44%|████▍     | 28942/65536 [4:57:44<6:10:29,  1.65it/s]
 44%|████▍     | 28943/65536 [4:57:44<6:14:47,  1.63it/s]
 44%|████▍     | 28944/65536 [4:57:45<6:15:21,  1.62it/s]
 44%|████▍     | 28945/65536 [4:57:45<6:17:43,  1.61it/s]
 44%|████▍     | 28946/65536 [4:57:46<6:19:21,  1.61it/s]
 44%|████▍     | 28947/65536 [4:57:47<6:22:48,  1.59it/s]
 44%|████▍     | 28948/65536 [4:57:47<6:19:01,  1.61it/s]
 44%|████▍     | 28949/65536 [4:57:48<6:20:56,  1.60it/s]
 44%|████▍     | 28950/65536 [4:57:49<6:34:28,  1.55it/s]
 44%|████▍     | 28951/65536 [4:57:49<6:29:29,  1.57it/s]
 44%|████▍     | 28952/65536 [4:57:50<6:24:37,  1.59it/s]
 44%|████▍     | 28953/65536 [4:57:51<6:18:18,  1.61it/s]
 44%|████▍     | 28954/65536 [4:57:51<6:16:26,  1.62it/s]
 44%|████▍     | 28955/65536 [4:57:52<6:28:18,  1.57it/s]
 44%|████▍     | 28956/65536 [4:57:52<6:23:56,  1.59it/s]
 44%|████▍     | 28957/65536 [4:57:53<6:16:51,  1.62it/s]
 44%|████▍     | 28958/65536 [4:57:54<6:07:58,  1.66it/s]
 44%|████▍     | 28959/65536 [4:57:54<6:19:55,  1.60it/s]
 44%|████▍     | 28960/65536 [4:57:55<6:17:41,  1.61it/s]
                                                         
{'loss': 1.6788, 'learning_rate': 6.038324966327904e-07, 'epoch': 1787.65}
+
 44%|████▍     | 28960/65536 [4:57:55<6:17:41,  1.61it/s]
 44%|████▍     | 28961/65536 [4:57:55<6:18:06,  1.61it/s]
 44%|████▍     | 28962/65536 [4:57:56<6:17:31,  1.61it/s]
 44%|████▍     | 28963/65536 [4:57:57<6:18:02,  1.61it/s]
 44%|████▍     | 28964/65536 [4:57:57<6:21:53,  1.60it/s]
 44%|████▍     | 28965/65536 [4:57:58<6:23:07,  1.59it/s]
 44%|████▍     | 28966/65536 [4:57:59<6:39:11,  1.53it/s]
 44%|████▍     | 28967/65536 [4:57:59<6:31:34,  1.56it/s]
 44%|████▍     | 28968/65536 [4:58:00<6:27:29,  1.57it/s]
 44%|████▍     | 28969/65536 [4:58:01<6:22:21,  1.59it/s]
 44%|████▍     | 28970/65536 [4:58:01<6:29:07,  1.57it/s]
 44%|████▍     | 28971/65536 [4:58:02<6:28:27,  1.57it/s]
 44%|████▍     | 28972/65536 [4:58:02<6:24:55,  1.58it/s]
 44%|████▍     | 28973/65536 [4:58:03<6:15:46,  1.62it/s]
 44%|████▍     | 28974/65536 [4:58:04<6:04:37,  1.67it/s]
 44%|████▍     | 28975/65536 [4:58:04<6:03:05,  1.68it/s]
 44%|████▍     | 28976/65536 [4:58:05<5:57:56,  1.70it/s]
 44%|████▍     | 28977/65536 [4:58:05<6:11:04,  1.64it/s]
 44%|████▍     | 28978/65536 [4:58:06<6:18:22,  1.61it/s]
 44%|████▍     | 28979/65536 [4:58:07<6:16:25,  1.62it/s]
 44%|████▍     | 28980/65536 [4:58:07<6:06:28,  1.66it/s]
                                                         
{'loss': 1.6828, 'learning_rate': 6.035569976735643e-07, 'epoch': 1788.89}
+
 44%|████▍     | 28980/65536 [4:58:07<6:06:28,  1.66it/s]
 44%|████▍     | 28981/65536 [4:58:08<6:10:58,  1.64it/s]
 44%|████▍     | 28982/65536 [4:58:09<6:17:38,  1.61it/s]
 44%|████▍     | 28983/65536 [4:58:09<6:16:56,  1.62it/s]
 44%|████▍     | 28984/65536 [4:58:10<6:09:05,  1.65it/s]
 44%|████▍     | 28985/65536 [4:58:10<6:07:56,  1.66it/s]
 44%|████▍     | 28986/65536 [4:58:11<6:09:33,  1.65it/s]
 44%|████▍     | 28987/65536 [4:58:12<6:11:13,  1.64it/s]
 44%|████▍     | 28988/65536 [4:58:12<6:05:09,  1.67it/s]
 44%|████▍     | 28989/65536 [4:58:13<6:10:01,  1.65it/s]
 44%|████▍     | 28990/65536 [4:58:13<6:18:11,  1.61it/s]
 44%|████▍     | 28991/65536 [4:58:14<6:22:28,  1.59it/s]
 44%|████▍     | 28992/65536 [4:58:15<6:32:00,  1.55it/s]
 44%|████▍     | 28993/65536 [4:58:15<6:32:56,  1.55it/s]
 44%|████▍     | 28994/65536 [4:58:16<6:29:14,  1.56it/s]
 44%|████▍     | 28995/65536 [4:58:17<6:15:37,  1.62it/s]
 44%|████▍     | 28996/65536 [4:58:17<6:19:15,  1.61it/s]
 44%|████▍     | 28997/65536 [4:58:18<6:16:43,  1.62it/s]
 44%|████▍     | 28998/65536 [4:58:18<6:12:25,  1.64it/s]
 44%|████▍     | 28999/65536 [4:58:19<6:24:20,  1.58it/s]
 44%|████▍     | 29000/65536 [4:58:20<6:17:23,  1.61it/s]
                                                         
{'loss': 1.6915, 'learning_rate': 6.032814987143382e-07, 'epoch': 1790.12}
+
 44%|████▍     | 29000/65536 [4:58:20<6:17:23,  1.61it/s]
 44%|████▍     | 29001/65536 [4:58:20<6:18:25,  1.61it/s]
 44%|████▍     | 29002/65536 [4:58:21<6:11:25,  1.64it/s]
 44%|████▍     | 29003/65536 [4:58:22<6:21:07,  1.60it/s]
 44%|████▍     | 29004/65536 [4:58:22<6:12:25,  1.63it/s]
 44%|████▍     | 29005/65536 [4:58:23<6:20:38,  1.60it/s]
 44%|████▍     | 29006/65536 [4:58:23<6:13:09,  1.63it/s]
 44%|████▍     | 29007/65536 [4:58:24<6:07:24,  1.66it/s]
 44%|████▍     | 29008/65536 [4:58:25<6:11:11,  1.64it/s]
 44%|████▍     | 29009/65536 [4:58:25<6:16:04,  1.62it/s]
 44%|████▍     | 29010/65536 [4:58:26<6:16:31,  1.62it/s]
 44%|████▍     | 29011/65536 [4:58:26<6:19:17,  1.60it/s]
 44%|████▍     | 29012/65536 [4:58:27<6:18:30,  1.61it/s]
 44%|████▍     | 29013/65536 [4:58:28<6:20:30,  1.60it/s]
 44%|████▍     | 29014/65536 [4:58:28<6:28:40,  1.57it/s]
 44%|████▍     | 29015/65536 [4:58:29<6:35:30,  1.54it/s]
 44%|████▍     | 29016/65536 [4:58:30<6:30:44,  1.56it/s]
 44%|████▍     | 29017/65536 [4:58:30<6:20:47,  1.60it/s]
 44%|████▍     | 29018/65536 [4:58:31<6:11:24,  1.64it/s]
 44%|████▍     | 29019/65536 [4:58:32<6:19:43,  1.60it/s]
 44%|████▍     | 29020/65536 [4:58:32<6:25:44,  1.58it/s]
                                                         
{'loss': 1.6709, 'learning_rate': 6.03005999755112e-07, 'epoch': 1791.36}
+
 44%|████▍     | 29020/65536 [4:58:32<6:25:44,  1.58it/s]
 44%|████▍     | 29021/65536 [4:58:33<6:23:40,  1.59it/s]
 44%|████▍     | 29022/65536 [4:58:33<6:24:43,  1.58it/s]
 44%|████▍     | 29023/65536 [4:58:34<6:21:54,  1.59it/s]
 44%|████▍     | 29024/65536 [4:58:35<6:19:04,  1.61it/s]
 44%|████▍     | 29025/65536 [4:58:35<6:09:17,  1.65it/s]
 44%|████▍     | 29026/65536 [4:58:36<6:17:40,  1.61it/s]
 44%|████▍     | 29027/65536 [4:58:36<6:11:00,  1.64it/s]
 44%|████▍     | 29028/65536 [4:58:37<6:16:13,  1.62it/s]
 44%|████▍     | 29029/65536 [4:58:38<6:21:01,  1.60it/s]
 44%|████▍     | 29030/65536 [4:58:38<6:16:29,  1.62it/s]
 44%|████▍     | 29031/65536 [4:58:39<6:20:22,  1.60it/s]
 44%|████▍     | 29032/65536 [4:58:40<6:21:35,  1.59it/s]
 44%|████▍     | 29033/65536 [4:58:40<6:09:30,  1.65it/s]
 44%|████▍     | 29034/65536 [4:58:41<6:16:08,  1.62it/s]
 44%|████▍     | 29035/65536 [4:58:41<6:19:59,  1.60it/s]
 44%|████▍     | 29036/65536 [4:58:42<6:09:52,  1.64it/s]
 44%|████▍     | 29037/65536 [4:58:43<6:05:28,  1.66it/s]
 44%|████▍     | 29038/65536 [4:58:43<6:03:25,  1.67it/s]
 44%|████▍     | 29039/65536 [4:58:44<6:08:45,  1.65it/s]
 44%|████▍     | 29040/65536 [4:58:44<6:07:41,  1.65it/s]
                                                         
{'loss': 1.7406, 'learning_rate': 6.027305007958859e-07, 'epoch': 1792.59}
+
 44%|████▍     | 29040/65536 [4:58:44<6:07:41,  1.65it/s]
 44%|████▍     | 29041/65536 [4:58:45<6:09:26,  1.65it/s]
 44%|████▍     | 29042/65536 [4:58:46<6:08:58,  1.65it/s]
 44%|████▍     | 29043/65536 [4:58:46<6:25:41,  1.58it/s]
 44%|████▍     | 29044/65536 [4:58:47<6:30:13,  1.56it/s]
 44%|████▍     | 29045/65536 [4:58:48<6:26:01,  1.58it/s]
 44%|████▍     | 29046/65536 [4:58:48<6:19:00,  1.60it/s]
 44%|████▍     | 29047/65536 [4:58:49<6:21:45,  1.59it/s]
 44%|████▍     | 29048/65536 [4:58:50<6:28:23,  1.57it/s]
 44%|████▍     | 29049/65536 [4:58:50<6:23:15,  1.59it/s]
 44%|████▍     | 29050/65536 [4:58:51<6:17:59,  1.61it/s]
 44%|████▍     | 29051/65536 [4:58:51<6:17:41,  1.61it/s]
 44%|████▍     | 29052/65536 [4:58:52<6:19:55,  1.60it/s]
 44%|████▍     | 29053/65536 [4:58:53<6:20:11,  1.60it/s]
 44%|████▍     | 29054/65536 [4:58:53<6:10:34,  1.64it/s]
 44%|████▍     | 29055/65536 [4:58:54<6:12:58,  1.63it/s]
 44%|████▍     | 29056/65536 [4:58:54<6:14:51,  1.62it/s]
 44%|████▍     | 29057/65536 [4:58:55<6:13:12,  1.63it/s]
 44%|████▍     | 29058/65536 [4:58:56<6:09:42,  1.64it/s]
 44%|████▍     | 29059/65536 [4:58:56<6:05:45,  1.66it/s]
 44%|████▍     | 29060/65536 [4:58:57<5:58:36,  1.70it/s]
                                                         
{'loss': 1.6912, 'learning_rate': 6.024550018366597e-07, 'epoch': 1793.83}
+
 44%|████▍     | 29060/65536 [4:58:57<5:58:36,  1.70it/s]
 44%|████▍     | 29061/65536 [4:58:57<6:09:25,  1.65it/s]
 44%|████▍     | 29062/65536 [4:58:58<6:20:41,  1.60it/s]
 44%|████▍     | 29063/65536 [4:58:59<6:28:00,  1.57it/s]
 44%|████▍     | 29064/65536 [4:58:59<6:14:05,  1.62it/s]
 44%|████▍     | 29065/65536 [4:59:00<6:10:24,  1.64it/s]
 44%|████▍     | 29066/65536 [4:59:01<6:14:30,  1.62it/s]
 44%|████▍     | 29067/65536 [4:59:01<6:08:20,  1.65it/s]
 44%|████▍     | 29068/65536 [4:59:02<6:03:36,  1.67it/s]
 44%|████▍     | 29069/65536 [4:59:02<6:01:41,  1.68it/s]
 44%|████▍     | 29070/65536 [4:59:03<5:58:22,  1.70it/s]
 44%|████▍     | 29071/65536 [4:59:03<5:59:00,  1.69it/s]
 44%|████▍     | 29072/65536 [4:59:04<6:06:28,  1.66it/s]
 44%|████▍     | 29073/65536 [4:59:05<6:06:58,  1.66it/s]
 44%|████▍     | 29074/65536 [4:59:05<6:12:37,  1.63it/s]
 44%|████▍     | 29075/65536 [4:59:06<6:23:41,  1.58it/s]
 44%|████▍     | 29076/65536 [4:59:07<6:19:10,  1.60it/s]
 44%|████▍     | 29077/65536 [4:59:07<6:13:33,  1.63it/s]
 44%|████▍     | 29078/65536 [4:59:08<6:23:08,  1.59it/s]
 44%|████▍     | 29079/65536 [4:59:09<6:21:39,  1.59it/s]
 44%|████▍     | 29080/65536 [4:59:09<6:27:01,  1.57it/s]
                                                         
{'loss': 1.6927, 'learning_rate': 6.021795028774336e-07, 'epoch': 1795.06}
+
 44%|████▍     | 29080/65536 [4:59:09<6:27:01,  1.57it/s]
 44%|████▍     | 29081/65536 [4:59:10<6:42:15,  1.51it/s]
 44%|████▍     | 29082/65536 [4:59:11<6:33:02,  1.55it/s]
 44%|████▍     | 29083/65536 [4:59:11<6:21:45,  1.59it/s]
 44%|████▍     | 29084/65536 [4:59:12<6:21:05,  1.59it/s]
 44%|████▍     | 29085/65536 [4:59:12<6:17:55,  1.61it/s]
 44%|████▍     | 29086/65536 [4:59:13<6:12:59,  1.63it/s]
 44%|████▍     | 29087/65536 [4:59:13<6:07:03,  1.65it/s]
 44%|████▍     | 29088/65536 [4:59:14<6:07:52,  1.65it/s]
 44%|████▍     | 29089/65536 [4:59:15<6:04:48,  1.67it/s]
 44%|████▍     | 29090/65536 [4:59:15<6:09:12,  1.65it/s]
 44%|████▍     | 29091/65536 [4:59:16<6:04:07,  1.67it/s]
 44%|████▍     | 29092/65536 [4:59:16<5:57:46,  1.70it/s]
 44%|████▍     | 29093/65536 [4:59:17<6:03:33,  1.67it/s]
 44%|████▍     | 29094/65536 [4:59:18<6:03:01,  1.67it/s]
 44%|████▍     | 29095/65536 [4:59:18<6:01:18,  1.68it/s]
 44%|████▍     | 29096/65536 [4:59:19<6:14:26,  1.62it/s]
 44%|████▍     | 29097/65536 [4:59:20<6:21:32,  1.59it/s]
 44%|████▍     | 29098/65536 [4:59:20<6:25:24,  1.58it/s]
 44%|████▍     | 29099/65536 [4:59:21<6:19:05,  1.60it/s]
 44%|████▍     | 29100/65536 [4:59:21<6:24:02,  1.58it/s]
                                                         
{'loss': 1.725, 'learning_rate': 6.019040039182075e-07, 'epoch': 1796.3}
+
 44%|████▍     | 29100/65536 [4:59:21<6:24:02,  1.58it/s]
 44%|████▍     | 29101/65536 [4:59:22<6:22:48,  1.59it/s]
 44%|████▍     | 29102/65536 [4:59:23<6:17:03,  1.61it/s]
 44%|████▍     | 29103/65536 [4:59:23<6:10:43,  1.64it/s]
 44%|████▍     | 29104/65536 [4:59:24<6:06:27,  1.66it/s]
 44%|████▍     | 29105/65536 [4:59:25<6:08:12,  1.65it/s]
 44%|████▍     | 29106/65536 [4:59:25<6:09:30,  1.64it/s]
 44%|████▍     | 29107/65536 [4:59:26<6:02:15,  1.68it/s]
 44%|████▍     | 29108/65536 [4:59:26<6:03:08,  1.67it/s]
 44%|████▍     | 29109/65536 [4:59:27<6:00:33,  1.68it/s]
 44%|████▍     | 29110/65536 [4:59:28<6:12:59,  1.63it/s]
 44%|████▍     | 29111/65536 [4:59:28<6:10:40,  1.64it/s]
 44%|████▍     | 29112/65536 [4:59:29<6:24:40,  1.58it/s]
 44%|████▍     | 29113/65536 [4:59:29<6:24:11,  1.58it/s]
 44%|████▍     | 29114/65536 [4:59:30<6:22:31,  1.59it/s]
 44%|████▍     | 29115/65536 [4:59:31<6:19:23,  1.60it/s]
 44%|████▍     | 29116/65536 [4:59:31<6:32:35,  1.55it/s]
 44%|████▍     | 29117/65536 [4:59:32<6:20:14,  1.60it/s]
 44%|████▍     | 29118/65536 [4:59:33<6:28:09,  1.56it/s]
 44%|████▍     | 29119/65536 [4:59:33<6:27:51,  1.56it/s]
 44%|████▍     | 29120/65536 [4:59:34<6:19:02,  1.60it/s]
                                                         
{'loss': 1.6578, 'learning_rate': 6.016285049589813e-07, 'epoch': 1797.53}
+
 44%|████▍     | 29120/65536 [4:59:34<6:19:02,  1.60it/s]
 44%|████▍     | 29121/65536 [4:59:34<6:12:03,  1.63it/s]
 44%|████▍     | 29122/65536 [4:59:35<6:08:11,  1.65it/s]
 44%|████▍     | 29123/65536 [4:59:36<6:12:38,  1.63it/s]
 44%|████▍     | 29124/65536 [4:59:36<6:07:39,  1.65it/s]
 44%|████▍     | 29125/65536 [4:59:37<6:02:27,  1.67it/s]
 44%|████▍     | 29126/65536 [4:59:37<5:58:50,  1.69it/s]
 44%|████▍     | 29127/65536 [4:59:38<5:56:02,  1.70it/s]
 44%|████▍     | 29128/65536 [4:59:39<6:13:44,  1.62it/s]
 44%|████▍     | 29129/65536 [4:59:39<6:09:56,  1.64it/s]
 44%|████▍     | 29130/65536 [4:59:40<6:14:44,  1.62it/s]
 44%|████▍     | 29131/65536 [4:59:41<6:13:03,  1.63it/s]
 44%|████▍     | 29132/65536 [4:59:41<6:09:50,  1.64it/s]
 44%|████▍     | 29133/65536 [4:59:42<6:15:29,  1.62it/s]
 44%|████▍     | 29134/65536 [4:59:42<6:09:56,  1.64it/s]
 44%|████▍     | 29135/65536 [4:59:43<6:15:10,  1.62it/s]
 44%|████▍     | 29136/65536 [4:59:44<6:06:21,  1.66it/s]
 44%|████▍     | 29137/65536 [4:59:44<6:15:22,  1.62it/s]
 44%|████▍     | 29138/65536 [4:59:45<6:11:50,  1.63it/s]
 44%|████▍     | 29139/65536 [4:59:45<6:08:17,  1.65it/s]
 44%|████▍     | 29140/65536 [4:59:46<6:09:14,  1.64it/s]
                                                         
{'loss': 1.7392, 'learning_rate': 6.013530059997552e-07, 'epoch': 1798.77}
+
 44%|████▍     | 29140/65536 [4:59:46<6:09:14,  1.64it/s]
 44%|████▍     | 29141/65536 [4:59:47<6:10:31,  1.64it/s]
 44%|████▍     | 29142/65536 [4:59:47<6:02:41,  1.67it/s]
 44%|████▍     | 29143/65536 [4:59:48<6:03:30,  1.67it/s]
 44%|████▍     | 29144/65536 [4:59:48<6:14:20,  1.62it/s]
 44%|████▍     | 29145/65536 [4:59:49<6:19:43,  1.60it/s]
 44%|████▍     | 29146/65536 [4:59:50<6:08:50,  1.64it/s]
 44%|████▍     | 29147/65536 [4:59:50<6:10:42,  1.64it/s]
 44%|████▍     | 29148/65536 [4:59:51<6:12:02,  1.63it/s]
 44%|████▍     | 29149/65536 [4:59:52<6:16:46,  1.61it/s]
 44%|████▍     | 29150/65536 [4:59:52<6:12:27,  1.63it/s]
 44%|████▍     | 29151/65536 [4:59:53<6:15:00,  1.62it/s]
 44%|████▍     | 29152/65536 [4:59:53<6:18:25,  1.60it/s]
 44%|████▍     | 29153/65536 [4:59:54<6:22:23,  1.59it/s]
 44%|████▍     | 29154/65536 [4:59:55<6:23:07,  1.58it/s]
 44%|████▍     | 29155/65536 [4:59:55<6:15:36,  1.61it/s]
 44%|████▍     | 29156/65536 [4:59:56<6:13:11,  1.62it/s]
 44%|████▍     | 29157/65536 [4:59:57<6:17:49,  1.60it/s]
 44%|████▍     | 29158/65536 [4:59:57<6:11:20,  1.63it/s]
 44%|████▍     | 29159/65536 [4:59:58<6:10:05,  1.64it/s]
 44%|████▍     | 29160/65536 [4:59:58<6:04:18,  1.66it/s]
                                                         
{'loss': 1.6798, 'learning_rate': 6.010775070405289e-07, 'epoch': 1800.0}
+
 44%|████▍     | 29160/65536 [4:59:58<6:04:18,  1.66it/s]
 44%|████▍     | 29161/65536 [4:59:59<6:12:26,  1.63it/s]
 44%|████▍     | 29162/65536 [5:00:00<6:11:51,  1.63it/s]
 44%|████▍     | 29163/65536 [5:00:00<6:15:54,  1.61it/s]
 45%|████▍     | 29164/65536 [5:00:01<6:11:26,  1.63it/s]
 45%|████▍     | 29165/65536 [5:00:01<6:13:33,  1.62it/s]
 45%|████▍     | 29166/65536 [5:00:02<6:09:42,  1.64it/s]
 45%|████▍     | 29167/65536 [5:00:03<6:16:12,  1.61it/s]
 45%|████▍     | 29168/65536 [5:00:03<6:13:32,  1.62it/s]
 45%|████▍     | 29169/65536 [5:00:04<6:12:33,  1.63it/s]
 45%|████▍     | 29170/65536 [5:00:05<6:28:11,  1.56it/s]
 45%|████▍     | 29171/65536 [5:00:05<6:24:23,  1.58it/s]
 45%|████▍     | 29172/65536 [5:00:06<6:19:34,  1.60it/s]
 45%|████▍     | 29173/65536 [5:00:06<6:16:14,  1.61it/s]
 45%|████▍     | 29174/65536 [5:00:07<6:15:21,  1.61it/s]
 45%|████▍     | 29175/65536 [5:00:08<6:08:44,  1.64it/s]
 45%|████▍     | 29176/65536 [5:00:08<6:04:39,  1.66it/s]
 45%|████▍     | 29177/65536 [5:00:09<6:14:31,  1.62it/s]
 45%|████▍     | 29178/65536 [5:00:10<6:21:46,  1.59it/s]
 45%|████▍     | 29179/65536 [5:00:10<6:15:08,  1.62it/s]
 45%|████▍     | 29180/65536 [5:00:11<6:16:21,  1.61it/s]
                                                         
{'loss': 1.6791, 'learning_rate': 6.008020080813028e-07, 'epoch': 1801.23}
+
 45%|████▍     | 29180/65536 [5:00:11<6:16:21,  1.61it/s]
 45%|████▍     | 29181/65536 [5:00:11<6:11:54,  1.63it/s]
 45%|████▍     | 29182/65536 [5:00:12<6:11:55,  1.63it/s]
 45%|████▍     | 29183/65536 [5:00:13<6:06:08,  1.65it/s]
 45%|████▍     | 29184/65536 [5:00:13<6:12:16,  1.63it/s]
 45%|████▍     | 29185/65536 [5:00:14<6:07:17,  1.65it/s]
 45%|████▍     | 29186/65536 [5:00:14<6:09:39,  1.64it/s]
 45%|████▍     | 29187/65536 [5:00:15<6:06:28,  1.65it/s]
 45%|████▍     | 29188/65536 [5:00:16<6:21:39,  1.59it/s]
 45%|████▍     | 29189/65536 [5:00:16<6:21:48,  1.59it/s]
 45%|████▍     | 29190/65536 [5:00:17<6:16:18,  1.61it/s]
 45%|████▍     | 29191/65536 [5:00:18<6:18:08,  1.60it/s]
 45%|████▍     | 29192/65536 [5:00:18<6:10:07,  1.64it/s]
 45%|████▍     | 29193/65536 [5:00:19<6:27:51,  1.56it/s]
 45%|████▍     | 29194/65536 [5:00:19<6:32:44,  1.54it/s]
 45%|████▍     | 29195/65536 [5:00:20<6:26:12,  1.57it/s]
 45%|████▍     | 29196/65536 [5:00:21<6:27:01,  1.56it/s]
 45%|████▍     | 29197/65536 [5:00:21<6:22:18,  1.58it/s]
 45%|████▍     | 29198/65536 [5:00:22<6:08:10,  1.64it/s]
 45%|████▍     | 29199/65536 [5:00:22<6:02:13,  1.67it/s]
 45%|████▍     | 29200/65536 [5:00:23<6:00:19,  1.68it/s]
                                                         
{'loss': 1.6692, 'learning_rate': 6.005265091220766e-07, 'epoch': 1802.47}
+
 45%|████▍     | 29200/65536 [5:00:23<6:00:19,  1.68it/s]
 45%|████▍     | 29201/65536 [5:00:24<6:12:49,  1.62it/s]
 45%|████▍     | 29202/65536 [5:00:24<6:15:04,  1.61it/s]
 45%|████▍     | 29203/65536 [5:00:25<6:17:08,  1.61it/s]
 45%|████▍     | 29204/65536 [5:00:26<6:12:33,  1.63it/s]
 45%|████▍     | 29205/65536 [5:00:26<6:19:40,  1.59it/s]
 45%|████▍     | 29206/65536 [5:00:27<6:20:36,  1.59it/s]
 45%|████▍     | 29207/65536 [5:00:27<6:16:28,  1.61it/s]
 45%|████▍     | 29208/65536 [5:00:28<6:04:11,  1.66it/s]
 45%|████▍     | 29209/65536 [5:00:29<6:16:17,  1.61it/s]
 45%|████▍     | 29210/65536 [5:00:29<6:17:35,  1.60it/s]
 45%|████▍     | 29211/65536 [5:00:30<6:14:13,  1.62it/s]
 45%|████▍     | 29212/65536 [5:00:31<6:17:04,  1.61it/s]
 45%|████▍     | 29213/65536 [5:00:31<6:17:48,  1.60it/s]
 45%|████▍     | 29214/65536 [5:00:32<6:14:55,  1.61it/s]
 45%|████▍     | 29215/65536 [5:00:32<6:16:45,  1.61it/s]
 45%|████▍     | 29216/65536 [5:00:33<6:11:34,  1.63it/s]
 45%|████▍     | 29217/65536 [5:00:34<6:03:39,  1.66it/s]
 45%|████▍     | 29218/65536 [5:00:34<6:02:41,  1.67it/s]
 45%|████▍     | 29219/65536 [5:00:35<5:58:04,  1.69it/s]
 45%|████▍     | 29220/65536 [5:00:35<6:06:38,  1.65it/s]
                                                         
{'loss': 1.6823, 'learning_rate': 6.002510101628504e-07, 'epoch': 1803.7}
+
 45%|████▍     | 29220/65536 [5:00:35<6:06:38,  1.65it/s]
 45%|████▍     | 29221/65536 [5:00:36<6:07:41,  1.65it/s]
 45%|████▍     | 29222/65536 [5:00:37<5:56:28,  1.70it/s]
 45%|████▍     | 29223/65536 [5:00:37<6:13:07,  1.62it/s]
 45%|████▍     | 29224/65536 [5:00:38<6:14:23,  1.62it/s]
 45%|████▍     | 29225/65536 [5:00:39<6:23:58,  1.58it/s]
 45%|████▍     | 29226/65536 [5:00:39<6:29:50,  1.55it/s]
 45%|████▍     | 29227/65536 [5:00:40<6:29:39,  1.55it/s]
 45%|████▍     | 29228/65536 [5:00:40<6:22:15,  1.58it/s]
 45%|████▍     | 29229/65536 [5:00:41<6:27:00,  1.56it/s]
 45%|████▍     | 29230/65536 [5:00:42<6:12:54,  1.62it/s]
 45%|████▍     | 29231/65536 [5:00:42<6:15:44,  1.61it/s]
 45%|████▍     | 29232/65536 [5:00:43<6:05:54,  1.65it/s]
 45%|████▍     | 29233/65536 [5:00:43<6:08:44,  1.64it/s]
 45%|████▍     | 29234/65536 [5:00:44<6:09:14,  1.64it/s]
 45%|████▍     | 29235/65536 [5:00:45<6:02:56,  1.67it/s]
 45%|████▍     | 29236/65536 [5:00:45<6:04:11,  1.66it/s]
 45%|████▍     | 29237/65536 [5:00:46<6:06:28,  1.65it/s]
 45%|████▍     | 29238/65536 [5:00:46<6:04:50,  1.66it/s]
 45%|████▍     | 29239/65536 [5:00:47<6:05:49,  1.65it/s]
 45%|████▍     | 29240/65536 [5:00:48<6:00:11,  1.68it/s]
                                                         
{'loss': 1.6822, 'learning_rate': 5.999755112036243e-07, 'epoch': 1804.94}
+
 45%|████▍     | 29240/65536 [5:00:48<6:00:11,  1.68it/s]
 45%|████▍     | 29241/65536 [5:00:48<6:08:16,  1.64it/s]
 45%|████▍     | 29242/65536 [5:00:49<6:16:13,  1.61it/s]
 45%|████▍     | 29243/65536 [5:00:50<6:14:23,  1.62it/s]
 45%|████▍     | 29244/65536 [5:00:50<6:21:35,  1.59it/s]
 45%|████▍     | 29245/65536 [5:00:51<6:23:01,  1.58it/s]
 45%|████▍     | 29246/65536 [5:00:51<6:17:51,  1.60it/s]
 45%|████▍     | 29247/65536 [5:00:52<6:27:04,  1.56it/s]
 45%|████▍     | 29248/65536 [5:00:53<6:14:27,  1.62it/s]
 45%|████▍     | 29249/65536 [5:00:53<6:12:08,  1.63it/s]
 45%|████▍     | 29250/65536 [5:00:54<6:04:56,  1.66it/s]
 45%|████▍     | 29251/65536 [5:00:55<6:09:49,  1.64it/s]
 45%|████▍     | 29252/65536 [5:00:55<6:03:42,  1.66it/s]
 45%|████▍     | 29253/65536 [5:00:56<6:02:53,  1.67it/s]
 45%|████▍     | 29254/65536 [5:00:56<6:05:29,  1.65it/s]
 45%|████▍     | 29255/65536 [5:00:57<6:02:12,  1.67it/s]
 45%|████▍     | 29256/65536 [5:00:58<6:06:17,  1.65it/s]
 45%|████▍     | 29257/65536 [5:00:58<6:18:48,  1.60it/s]
 45%|████▍     | 29258/65536 [5:00:59<6:33:07,  1.54it/s]
 45%|████▍     | 29259/65536 [5:01:00<6:27:20,  1.56it/s]
 45%|████▍     | 29260/65536 [5:01:00<6:29:51,  1.55it/s]
                                                         
{'loss': 1.6831, 'learning_rate': 5.997000122443981e-07, 'epoch': 1806.17}
+
 45%|████▍     | 29260/65536 [5:01:00<6:29:51,  1.55it/s]
 45%|████▍     | 29261/65536 [5:01:01<6:34:10,  1.53it/s]
 45%|████▍     | 29262/65536 [5:01:01<6:22:48,  1.58it/s]
 45%|████▍     | 29263/65536 [5:01:02<6:20:02,  1.59it/s]
 45%|████▍     | 29264/65536 [5:01:03<6:23:04,  1.58it/s]
 45%|████▍     | 29265/65536 [5:01:03<6:13:36,  1.62it/s]
 45%|████▍     | 29266/65536 [5:01:04<6:11:57,  1.63it/s]
 45%|████▍     | 29267/65536 [5:01:05<6:12:51,  1.62it/s]
 45%|████▍     | 29268/65536 [5:01:05<6:05:41,  1.65it/s]
 45%|████▍     | 29269/65536 [5:01:06<6:18:08,  1.60it/s]
 45%|████▍     | 29270/65536 [5:01:06<6:18:12,  1.60it/s]
 45%|████▍     | 29271/65536 [5:01:07<6:11:38,  1.63it/s]
 45%|████▍     | 29272/65536 [5:01:08<5:59:22,  1.68it/s]
 45%|████▍     | 29273/65536 [5:01:08<5:52:07,  1.72it/s]
 45%|████▍     | 29274/65536 [5:01:09<6:05:16,  1.65it/s]
 45%|████▍     | 29275/65536 [5:01:09<5:59:07,  1.68it/s]
 45%|████▍     | 29276/65536 [5:01:10<5:58:42,  1.68it/s]
 45%|████▍     | 29277/65536 [5:01:10<5:55:48,  1.70it/s]
 45%|████▍     | 29278/65536 [5:01:11<5:57:50,  1.69it/s]
 45%|████▍     | 29279/65536 [5:01:12<6:01:47,  1.67it/s]
 45%|████▍     | 29280/65536 [5:01:12<6:08:08,  1.64it/s]
                                                         
{'loss': 1.7249, 'learning_rate': 5.99424513285172e-07, 'epoch': 1807.41}
+
 45%|████▍     | 29280/65536 [5:01:12<6:08:08,  1.64it/s]
 45%|████▍     | 29281/65536 [5:01:13<6:11:20,  1.63it/s]
 45%|████▍     | 29282/65536 [5:01:14<6:12:13,  1.62it/s]
 45%|████▍     | 29283/65536 [5:01:14<6:15:05,  1.61it/s]
 45%|████▍     | 29284/65536 [5:01:15<6:09:36,  1.63it/s]
 45%|████▍     | 29285/65536 [5:01:15<6:09:58,  1.63it/s]
 45%|████▍     | 29286/65536 [5:01:16<6:08:03,  1.64it/s]
 45%|████▍     | 29287/65536 [5:01:17<6:09:11,  1.64it/s]
 45%|████▍     | 29288/65536 [5:01:17<6:10:20,  1.63it/s]
 45%|████▍     | 29289/65536 [5:01:18<6:24:35,  1.57it/s]
 45%|████▍     | 29290/65536 [5:01:19<6:26:48,  1.56it/s]
 45%|████▍     | 29291/65536 [5:01:19<6:20:45,  1.59it/s]
 45%|████▍     | 29292/65536 [5:01:20<6:07:54,  1.64it/s]
 45%|████▍     | 29293/65536 [5:01:20<6:05:49,  1.65it/s]
 45%|████▍     | 29294/65536 [5:01:21<6:05:36,  1.65it/s]
 45%|████▍     | 29295/65536 [5:01:22<6:14:45,  1.61it/s]
 45%|████▍     | 29296/65536 [5:01:22<6:25:25,  1.57it/s]
 45%|████▍     | 29297/65536 [5:01:23<6:21:42,  1.58it/s]
 45%|████▍     | 29298/65536 [5:01:23<6:11:04,  1.63it/s]
 45%|████▍     | 29299/65536 [5:01:24<6:05:59,  1.65it/s]
 45%|████▍     | 29300/65536 [5:01:25<6:16:43,  1.60it/s]
                                                         
{'loss': 1.6903, 'learning_rate': 5.991490143259457e-07, 'epoch': 1808.64}
+
 45%|████▍     | 29300/65536 [5:01:25<6:16:43,  1.60it/s]
 45%|████▍     | 29301/65536 [5:01:25<6:29:20,  1.55it/s]
 45%|████▍     | 29302/65536 [5:01:26<6:28:05,  1.56it/s]
 45%|████▍     | 29303/65536 [5:01:27<6:19:49,  1.59it/s]
 45%|████▍     | 29304/65536 [5:01:27<6:08:55,  1.64it/s]
 45%|████▍     | 29305/65536 [5:01:28<5:59:55,  1.68it/s]
 45%|████▍     | 29306/65536 [5:01:28<6:07:29,  1.64it/s]
 45%|████▍     | 29307/65536 [5:01:29<6:03:56,  1.66it/s]
 45%|████▍     | 29308/65536 [5:01:30<6:23:32,  1.57it/s]
 45%|████▍     | 29309/65536 [5:01:30<6:21:15,  1.58it/s]
 45%|████▍     | 29310/65536 [5:01:31<6:20:20,  1.59it/s]
 45%|████▍     | 29311/65536 [5:01:32<6:15:14,  1.61it/s]
 45%|████▍     | 29312/65536 [5:01:32<6:10:46,  1.63it/s]
 45%|████▍     | 29313/65536 [5:01:33<6:08:39,  1.64it/s]
 45%|████▍     | 29314/65536 [5:01:33<6:08:11,  1.64it/s]
 45%|████▍     | 29315/65536 [5:01:34<6:08:00,  1.64it/s]
 45%|████▍     | 29316/65536 [5:01:35<6:04:10,  1.66it/s]
 45%|████▍     | 29317/65536 [5:01:35<5:58:00,  1.69it/s]
 45%|████▍     | 29318/65536 [5:01:36<6:10:21,  1.63it/s]
 45%|████▍     | 29319/65536 [5:01:36<6:13:56,  1.61it/s]
 45%|████▍     | 29320/65536 [5:01:37<6:17:03,  1.60it/s]
                                                         
{'loss': 1.6981, 'learning_rate': 5.988735153667197e-07, 'epoch': 1809.88}
+
 45%|████▍     | 29320/65536 [5:01:37<6:17:03,  1.60it/s]
 45%|████▍     | 29321/65536 [5:01:38<6:13:18,  1.62it/s]
 45%|████▍     | 29322/65536 [5:01:38<6:15:21,  1.61it/s]
 45%|████▍     | 29323/65536 [5:01:39<6:35:42,  1.53it/s]
 45%|████▍     | 29324/65536 [5:01:40<6:23:31,  1.57it/s]
 45%|████▍     | 29325/65536 [5:01:40<6:23:51,  1.57it/s]
 45%|████▍     | 29326/65536 [5:01:41<6:23:36,  1.57it/s]
 45%|████▍     | 29327/65536 [5:01:42<6:19:03,  1.59it/s]
 45%|████▍     | 29328/65536 [5:01:42<6:13:33,  1.62it/s]
 45%|████▍     | 29329/65536 [5:01:43<6:02:01,  1.67it/s]
 45%|████▍     | 29330/65536 [5:01:43<6:07:11,  1.64it/s]
 45%|████▍     | 29331/65536 [5:01:44<6:04:47,  1.65it/s]
 45%|████▍     | 29332/65536 [5:01:44<6:01:12,  1.67it/s]
 45%|████▍     | 29333/65536 [5:01:45<6:01:43,  1.67it/s]
 45%|████▍     | 29334/65536 [5:01:46<6:07:57,  1.64it/s]
 45%|████▍     | 29335/65536 [5:01:46<6:04:30,  1.66it/s]
 45%|████▍     | 29336/65536 [5:01:47<6:12:23,  1.62it/s]
 45%|████▍     | 29337/65536 [5:01:48<6:11:00,  1.63it/s]
 45%|████▍     | 29338/65536 [5:01:48<6:07:03,  1.64it/s]
 45%|████▍     | 29339/65536 [5:01:49<6:22:22,  1.58it/s]
 45%|████▍     | 29340/65536 [5:01:49<6:18:45,  1.59it/s]
                                                         
{'loss': 1.699, 'learning_rate': 5.985980164074936e-07, 'epoch': 1811.11}
+
 45%|████▍     | 29340/65536 [5:01:49<6:18:45,  1.59it/s]
 45%|████▍     | 29341/65536 [5:01:50<6:24:22,  1.57it/s]
 45%|████▍     | 29342/65536 [5:01:51<6:22:28,  1.58it/s]
 45%|████▍     | 29343/65536 [5:01:51<6:13:42,  1.61it/s]
 45%|████▍     | 29344/65536 [5:01:52<6:12:56,  1.62it/s]
 45%|████▍     | 29345/65536 [5:01:53<6:09:33,  1.63it/s]
 45%|████▍     | 29346/65536 [5:01:53<6:02:50,  1.66it/s]
 45%|████▍     | 29347/65536 [5:01:54<6:03:07,  1.66it/s]
 45%|████▍     | 29348/65536 [5:01:54<6:02:19,  1.66it/s]
 45%|████▍     | 29349/65536 [5:01:55<5:57:30,  1.69it/s]
 45%|████▍     | 29350/65536 [5:01:56<6:05:40,  1.65it/s]
 45%|████▍     | 29351/65536 [5:01:56<6:08:35,  1.64it/s]
 45%|████▍     | 29352/65536 [5:01:57<6:14:32,  1.61it/s]
 45%|████▍     | 29353/65536 [5:01:57<6:10:11,  1.63it/s]
 45%|████▍     | 29354/65536 [5:01:58<6:08:46,  1.64it/s]
 45%|████▍     | 29355/65536 [5:01:59<6:16:17,  1.60it/s]
 45%|████▍     | 29356/65536 [5:01:59<6:10:46,  1.63it/s]
 45%|████▍     | 29357/65536 [5:02:00<6:07:07,  1.64it/s]
 45%|████▍     | 29358/65536 [5:02:00<6:08:25,  1.64it/s]
 45%|████▍     | 29359/65536 [5:02:01<6:07:11,  1.64it/s]
 45%|████▍     | 29360/65536 [5:02:02<6:05:36,  1.65it/s]
                                                         
{'loss': 1.6921, 'learning_rate': 5.983225174482674e-07, 'epoch': 1812.35}
+
 45%|████▍     | 29360/65536 [5:02:02<6:05:36,  1.65it/s]
 45%|████▍     | 29361/65536 [5:02:02<6:10:14,  1.63it/s]
 45%|████▍     | 29362/65536 [5:02:03<6:07:22,  1.64it/s]
 45%|████▍     | 29363/65536 [5:02:04<6:25:33,  1.56it/s]
 45%|████▍     | 29364/65536 [5:02:04<6:20:48,  1.58it/s]
 45%|████▍     | 29365/65536 [5:02:05<6:23:20,  1.57it/s]
 45%|████▍     | 29366/65536 [5:02:06<6:27:22,  1.56it/s]
 45%|████▍     | 29367/65536 [5:02:06<6:15:58,  1.60it/s]
 45%|████▍     | 29368/65536 [5:02:07<6:18:14,  1.59it/s]
 45%|████▍     | 29369/65536 [5:02:07<6:07:51,  1.64it/s]
 45%|████▍     | 29370/65536 [5:02:08<5:58:51,  1.68it/s]
 45%|████▍     | 29371/65536 [5:02:09<6:33:40,  1.53it/s]
 45%|████▍     | 29372/65536 [5:02:09<6:27:41,  1.55it/s]
 45%|████▍     | 29373/65536 [5:02:10<6:13:42,  1.61it/s]
 45%|████▍     | 29374/65536 [5:02:10<6:12:01,  1.62it/s]
 45%|████▍     | 29375/65536 [5:02:11<6:12:45,  1.62it/s]
 45%|████▍     | 29376/65536 [5:02:12<6:18:30,  1.59it/s]
 45%|████▍     | 29377/65536 [5:02:12<6:25:28,  1.56it/s]
 45%|████▍     | 29378/65536 [5:02:13<6:17:48,  1.60it/s]
 45%|████▍     | 29379/65536 [5:02:14<6:10:44,  1.63it/s]
 45%|████▍     | 29380/65536 [5:02:14<5:58:58,  1.68it/s]
                                                         
{'loss': 1.69, 'learning_rate': 5.980470184890413e-07, 'epoch': 1813.58}
+
 45%|████▍     | 29380/65536 [5:02:14<5:58:58,  1.68it/s]
 45%|████▍     | 29381/65536 [5:02:15<5:57:25,  1.69it/s]
 45%|████▍     | 29382/65536 [5:02:15<5:56:34,  1.69it/s]
 45%|████▍     | 29383/65536 [5:02:16<6:08:13,  1.64it/s]
 45%|████▍     | 29384/65536 [5:02:17<6:10:55,  1.62it/s]
 45%|████▍     | 29385/65536 [5:02:17<5:59:35,  1.68it/s]
 45%|████▍     | 29386/65536 [5:02:18<6:04:25,  1.65it/s]
 45%|████▍     | 29387/65536 [5:02:18<6:13:31,  1.61it/s]
 45%|████▍     | 29388/65536 [5:02:19<6:06:44,  1.64it/s]
 45%|████▍     | 29389/65536 [5:02:20<5:57:20,  1.69it/s]
 45%|████▍     | 29390/65536 [5:02:20<6:07:06,  1.64it/s]
 45%|████▍     | 29391/65536 [5:02:21<6:09:03,  1.63it/s]
 45%|████▍     | 29392/65536 [5:02:21<6:06:06,  1.65it/s]
 45%|████▍     | 29393/65536 [5:02:22<6:08:09,  1.64it/s]
 45%|████▍     | 29394/65536 [5:02:23<6:04:12,  1.65it/s]
 45%|████▍     | 29395/65536 [5:02:23<6:07:05,  1.64it/s]
 45%|████▍     | 29396/65536 [5:02:24<5:59:43,  1.67it/s]
 45%|████▍     | 29397/65536 [5:02:24<6:01:51,  1.66it/s]
 45%|████▍     | 29398/65536 [5:02:25<6:00:05,  1.67it/s]
 45%|████▍     | 29399/65536 [5:02:26<6:06:27,  1.64it/s]
 45%|████▍     | 29400/65536 [5:02:26<6:13:23,  1.61it/s]
                                                         
{'loss': 1.6803, 'learning_rate': 5.977715195298151e-07, 'epoch': 1814.81}
+
 45%|████▍     | 29400/65536 [5:02:26<6:13:23,  1.61it/s]
 45%|████▍     | 29401/65536 [5:02:27<6:30:54,  1.54it/s]
 45%|████▍     | 29402/65536 [5:02:28<6:23:04,  1.57it/s]
 45%|████▍     | 29403/65536 [5:02:28<6:20:38,  1.58it/s]
 45%|████▍     | 29404/65536 [5:02:29<6:25:27,  1.56it/s]
 45%|████▍     | 29405/65536 [5:02:30<6:21:52,  1.58it/s]
 45%|████▍     | 29406/65536 [5:02:30<6:23:27,  1.57it/s]
 45%|████▍     | 29407/65536 [5:02:31<6:20:12,  1.58it/s]
 45%|████▍     | 29408/65536 [5:02:31<6:12:24,  1.62it/s]
 45%|████▍     | 29409/65536 [5:02:32<6:26:17,  1.56it/s]
 45%|████▍     | 29410/65536 [5:02:33<6:16:59,  1.60it/s]
 45%|████▍     | 29411/65536 [5:02:33<6:17:28,  1.60it/s]
 45%|████▍     | 29412/65536 [5:02:34<6:19:46,  1.59it/s]
 45%|████▍     | 29413/65536 [5:02:35<6:17:04,  1.60it/s]
 45%|████▍     | 29414/65536 [5:02:35<6:13:49,  1.61it/s]
 45%|████▍     | 29415/65536 [5:02:36<6:12:10,  1.62it/s]
 45%|████▍     | 29416/65536 [5:02:36<6:08:58,  1.63it/s]
 45%|████▍     | 29417/65536 [5:02:37<6:19:43,  1.59it/s]
 45%|████▍     | 29418/65536 [5:02:38<6:15:26,  1.60it/s]
 45%|████▍     | 29419/65536 [5:02:38<6:06:49,  1.64it/s]
 45%|████▍     | 29420/65536 [5:02:39<6:20:09,  1.58it/s]
                                                         
{'loss': 1.7028, 'learning_rate': 5.974960205705889e-07, 'epoch': 1816.05}
+
 45%|████▍     | 29420/65536 [5:02:39<6:20:09,  1.58it/s]
 45%|████▍     | 29421/65536 [5:02:39<6:08:32,  1.63it/s]
 45%|████▍     | 29422/65536 [5:02:40<6:07:39,  1.64it/s]
 45%|████▍     | 29423/65536 [5:02:41<6:19:47,  1.58it/s]
 45%|████▍     | 29424/65536 [5:02:41<6:18:43,  1.59it/s]
 45%|████▍     | 29425/65536 [5:02:42<6:15:31,  1.60it/s]
 45%|████▍     | 29426/65536 [5:02:43<6:12:19,  1.62it/s]
 45%|████▍     | 29427/65536 [5:02:43<6:08:41,  1.63it/s]
 45%|████▍     | 29428/65536 [5:02:44<6:04:15,  1.65it/s]
 45%|████▍     | 29429/65536 [5:02:44<6:18:14,  1.59it/s]
 45%|████▍     | 29430/65536 [5:02:45<6:11:17,  1.62it/s]
 45%|████▍     | 29431/65536 [5:02:46<6:14:35,  1.61it/s]
 45%|████▍     | 29432/65536 [5:02:46<6:13:07,  1.61it/s]
 45%|████▍     | 29433/65536 [5:02:47<6:07:56,  1.64it/s]
 45%|████▍     | 29434/65536 [5:02:48<6:11:23,  1.62it/s]
 45%|████▍     | 29435/65536 [5:02:48<6:04:25,  1.65it/s]
 45%|████▍     | 29436/65536 [5:02:49<6:17:32,  1.59it/s]
 45%|████▍     | 29437/65536 [5:02:49<6:07:47,  1.64it/s]
 45%|████▍     | 29438/65536 [5:02:50<6:07:11,  1.64it/s]
 45%|████▍     | 29439/65536 [5:02:51<6:06:26,  1.64it/s]
 45%|████▍     | 29440/65536 [5:02:51<6:08:15,  1.63it/s]
                                                         
{'loss': 1.6634, 'learning_rate': 5.972205216113627e-07, 'epoch': 1817.28}
+
 45%|████▍     | 29440/65536 [5:02:51<6:08:15,  1.63it/s]
 45%|████▍     | 29441/65536 [5:02:52<6:09:11,  1.63it/s]
 45%|████▍     | 29442/65536 [5:02:52<6:14:09,  1.61it/s]
 45%|████▍     | 29443/65536 [5:02:53<6:12:45,  1.61it/s]
 45%|████▍     | 29444/65536 [5:02:54<6:14:56,  1.60it/s]
 45%|████▍     | 29445/65536 [5:02:54<6:18:59,  1.59it/s]
 45%|████▍     | 29446/65536 [5:02:55<6:18:41,  1.59it/s]
 45%|████▍     | 29447/65536 [5:02:56<6:15:36,  1.60it/s]
 45%|████▍     | 29448/65536 [5:02:56<6:12:16,  1.62it/s]
 45%|████▍     | 29449/65536 [5:02:57<6:05:52,  1.64it/s]
 45%|████▍     | 29450/65536 [5:02:57<6:12:08,  1.62it/s]
 45%|████▍     | 29451/65536 [5:02:58<6:05:17,  1.65it/s]
 45%|████▍     | 29452/65536 [5:02:59<6:23:24,  1.57it/s]
 45%|████▍     | 29453/65536 [5:02:59<6:22:28,  1.57it/s]
 45%|████▍     | 29454/65536 [5:03:00<6:16:21,  1.60it/s]
 45%|████▍     | 29455/65536 [5:03:01<6:12:34,  1.61it/s]
 45%|████▍     | 29456/65536 [5:03:01<6:11:32,  1.62it/s]
 45%|████▍     | 29457/65536 [5:03:02<6:12:53,  1.61it/s]
 45%|████▍     | 29458/65536 [5:03:02<6:03:25,  1.65it/s]
 45%|████▍     | 29459/65536 [5:03:03<5:57:17,  1.68it/s]
 45%|████▍     | 29460/65536 [5:03:03<5:50:43,  1.71it/s]
                                                         
{'loss': 1.6827, 'learning_rate': 5.969450226521366e-07, 'epoch': 1818.52}
+
 45%|████▍     | 29460/65536 [5:03:03<5:50:43,  1.71it/s]
 45%|████▍     | 29461/65536 [5:03:04<6:08:26,  1.63it/s]
 45%|████▍     | 29462/65536 [5:03:05<6:12:20,  1.61it/s]
 45%|████▍     | 29463/65536 [5:03:05<6:10:21,  1.62it/s]
 45%|████▍     | 29464/65536 [5:03:06<6:08:26,  1.63it/s]
 45%|████▍     | 29465/65536 [5:03:07<6:26:15,  1.56it/s]
 45%|████▍     | 29466/65536 [5:03:07<6:28:24,  1.55it/s]
 45%|████▍     | 29467/65536 [5:03:08<6:20:13,  1.58it/s]
 45%|████▍     | 29468/65536 [5:03:09<6:25:38,  1.56it/s]
 45%|████▍     | 29469/65536 [5:03:09<6:26:54,  1.55it/s]
 45%|████▍     | 29470/65536 [5:03:10<6:24:05,  1.56it/s]
 45%|████▍     | 29471/65536 [5:03:11<6:25:27,  1.56it/s]
 45%|████▍     | 29472/65536 [5:03:11<6:22:40,  1.57it/s]
 45%|████▍     | 29473/65536 [5:03:12<6:22:08,  1.57it/s]
 45%|████▍     | 29474/65536 [5:03:12<6:13:56,  1.61it/s]
 45%|████▍     | 29475/65536 [5:03:13<6:22:33,  1.57it/s]
 45%|████▍     | 29476/65536 [5:03:14<6:09:44,  1.63it/s]
 45%|████▍     | 29477/65536 [5:03:14<6:09:08,  1.63it/s]
 45%|████▍     | 29478/65536 [5:03:15<6:03:45,  1.65it/s]
 45%|████▍     | 29479/65536 [5:03:15<6:04:30,  1.65it/s]
 45%|████▍     | 29480/65536 [5:03:16<6:07:09,  1.64it/s]
                                                         
{'loss': 1.6808, 'learning_rate': 5.966695236929104e-07, 'epoch': 1819.75}
+
 45%|████▍     | 29480/65536 [5:03:16<6:07:09,  1.64it/s]
 45%|████▍     | 29481/65536 [5:03:17<6:04:30,  1.65it/s]
 45%|████▍     | 29482/65536 [5:03:17<5:59:26,  1.67it/s]
 45%|████▍     | 29483/65536 [5:03:18<6:07:48,  1.63it/s]
 45%|████▍     | 29484/65536 [5:03:19<6:13:49,  1.61it/s]
 45%|████▍     | 29485/65536 [5:03:19<6:23:20,  1.57it/s]
 45%|████▍     | 29486/65536 [5:03:20<6:29:22,  1.54it/s]
 45%|████▍     | 29487/65536 [5:03:21<6:25:26,  1.56it/s]
 45%|████▍     | 29488/65536 [5:03:21<6:24:20,  1.56it/s]
 45%|████▍     | 29489/65536 [5:03:22<6:16:23,  1.60it/s]
 45%|████▍     | 29490/65536 [5:03:22<6:10:01,  1.62it/s]
 45%|████▍     | 29491/65536 [5:03:23<6:10:59,  1.62it/s]
 45%|████▌     | 29492/65536 [5:03:24<6:12:34,  1.61it/s]
 45%|████▌     | 29493/65536 [5:03:24<6:07:38,  1.63it/s]
 45%|████▌     | 29494/65536 [5:03:25<6:05:40,  1.64it/s]
 45%|████▌     | 29495/65536 [5:03:25<5:58:59,  1.67it/s]
 45%|████▌     | 29496/65536 [5:03:26<6:04:03,  1.65it/s]
 45%|████▌     | 29497/65536 [5:03:27<6:06:12,  1.64it/s]
 45%|████▌     | 29498/65536 [5:03:27<6:05:58,  1.64it/s]
 45%|████▌     | 29499/65536 [5:03:28<6:17:32,  1.59it/s]
 45%|████▌     | 29500/65536 [5:03:28<6:09:32,  1.63it/s]
                                                         
{'loss': 1.6874, 'learning_rate': 5.963940247336842e-07, 'epoch': 1820.99}
+
 45%|████▌     | 29500/65536 [5:03:28<6:09:32,  1.63it/s]
 45%|████▌     | 29501/65536 [5:03:29<6:23:06,  1.57it/s]
 45%|████▌     | 29502/65536 [5:03:30<6:26:25,  1.55it/s]
 45%|████▌     | 29503/65536 [5:03:30<6:29:13,  1.54it/s]
 45%|████▌     | 29504/65536 [5:03:31<6:17:47,  1.59it/s]
 45%|████▌     | 29505/65536 [5:03:32<6:08:15,  1.63it/s]
 45%|████▌     | 29506/65536 [5:03:32<6:12:42,  1.61it/s]
 45%|████▌     | 29507/65536 [5:03:33<6:15:22,  1.60it/s]
 45%|████▌     | 29508/65536 [5:03:33<6:07:15,  1.64it/s]
 45%|████▌     | 29509/65536 [5:03:34<6:02:28,  1.66it/s]
 45%|████▌     | 29510/65536 [5:03:35<6:10:48,  1.62it/s]
 45%|████▌     | 29511/65536 [5:03:35<6:10:23,  1.62it/s]
 45%|████▌     | 29512/65536 [5:03:36<6:08:44,  1.63it/s]
 45%|████▌     | 29513/65536 [5:03:37<6:05:22,  1.64it/s]
 45%|████▌     | 29514/65536 [5:03:37<6:04:33,  1.65it/s]
 45%|████▌     | 29515/65536 [5:03:38<6:16:23,  1.60it/s]
 45%|████▌     | 29516/65536 [5:03:38<6:16:04,  1.60it/s]
 45%|████▌     | 29517/65536 [5:03:39<6:20:47,  1.58it/s]
 45%|████▌     | 29518/65536 [5:03:40<6:12:08,  1.61it/s]
 45%|████▌     | 29519/65536 [5:03:40<6:12:16,  1.61it/s]
 45%|████▌     | 29520/65536 [5:03:41<6:05:34,  1.64it/s]
                                                         
{'loss': 1.7005, 'learning_rate': 5.961185257744581e-07, 'epoch': 1822.22}
+
 45%|████▌     | 29520/65536 [5:03:41<6:05:34,  1.64it/s]
 45%|████▌     | 29521/65536 [5:03:41<6:05:29,  1.64it/s]
 45%|████▌     | 29522/65536 [5:03:42<6:04:27,  1.65it/s]
 45%|████▌     | 29523/65536 [5:03:43<6:16:37,  1.59it/s]
 45%|████▌     | 29524/65536 [5:03:43<6:24:58,  1.56it/s]
 45%|████▌     | 29525/65536 [5:03:44<6:28:29,  1.54it/s]
 45%|████▌     | 29526/65536 [5:03:45<6:29:44,  1.54it/s]
 45%|████▌     | 29527/65536 [5:03:45<6:17:05,  1.59it/s]
 45%|████▌     | 29528/65536 [5:03:46<6:22:35,  1.57it/s]
 45%|████▌     | 29529/65536 [5:03:47<6:16:19,  1.59it/s]
 45%|████▌     | 29530/65536 [5:03:47<6:08:08,  1.63it/s]
 45%|████▌     | 29531/65536 [5:03:48<6:08:48,  1.63it/s]
 45%|████▌     | 29532/65536 [5:03:48<6:02:49,  1.65it/s]
 45%|████▌     | 29533/65536 [5:03:49<6:12:03,  1.61it/s]
 45%|████▌     | 29534/65536 [5:03:50<6:06:13,  1.64it/s]
 45%|████▌     | 29535/65536 [5:03:50<6:02:46,  1.65it/s]
 45%|████▌     | 29536/65536 [5:03:51<6:11:50,  1.61it/s]
 45%|████▌     | 29537/65536 [5:03:51<6:08:17,  1.63it/s]
 45%|████▌     | 29538/65536 [5:03:52<5:58:11,  1.67it/s]
 45%|████▌     | 29539/65536 [5:03:53<6:03:09,  1.65it/s]
 45%|████▌     | 29540/65536 [5:03:53<5:54:26,  1.69it/s]
                                                         
{'loss': 1.7055, 'learning_rate': 5.958430268152319e-07, 'epoch': 1823.46}
+
 45%|████▌     | 29540/65536 [5:03:53<5:54:26,  1.69it/s]
 45%|████▌     | 29541/65536 [5:03:54<5:58:16,  1.67it/s]
 45%|████▌     | 29542/65536 [5:03:54<5:59:21,  1.67it/s]
 45%|████▌     | 29543/65536 [5:03:55<6:12:57,  1.61it/s]
 45%|████▌     | 29544/65536 [5:03:56<6:07:50,  1.63it/s]
 45%|████▌     | 29545/65536 [5:03:56<6:05:12,  1.64it/s]
 45%|████▌     | 29546/65536 [5:03:57<6:04:21,  1.65it/s]
 45%|████▌     | 29547/65536 [5:03:58<6:05:14,  1.64it/s]
 45%|████▌     | 29548/65536 [5:03:58<6:24:35,  1.56it/s]
 45%|████▌     | 29549/65536 [5:03:59<6:38:31,  1.51it/s]
 45%|████▌     | 29550/65536 [5:04:00<6:25:25,  1.56it/s]
 45%|████▌     | 29551/65536 [5:04:00<6:15:01,  1.60it/s]
 45%|████▌     | 29552/65536 [5:04:01<6:03:18,  1.65it/s]
 45%|████▌     | 29553/65536 [5:04:01<6:14:18,  1.60it/s]
 45%|████▌     | 29554/65536 [5:04:02<6:07:36,  1.63it/s]
 45%|████▌     | 29555/65536 [5:04:03<6:03:24,  1.65it/s]
 45%|████▌     | 29556/65536 [5:04:03<6:09:55,  1.62it/s]
 45%|████▌     | 29557/65536 [5:04:04<6:16:52,  1.59it/s]
 45%|████▌     | 29558/65536 [5:04:04<6:09:12,  1.62it/s]
 45%|████▌     | 29559/65536 [5:04:05<6:07:36,  1.63it/s]
 45%|████▌     | 29560/65536 [5:04:06<6:09:45,  1.62it/s]
                                                         
{'loss': 1.6762, 'learning_rate': 5.955675278560058e-07, 'epoch': 1824.69}
+
 45%|████▌     | 29560/65536 [5:04:06<6:09:45,  1.62it/s]
 45%|████▌     | 29561/65536 [5:04:06<6:11:51,  1.61it/s]
 45%|████▌     | 29562/65536 [5:04:07<6:10:24,  1.62it/s]
 45%|████▌     | 29563/65536 [5:04:07<6:05:16,  1.64it/s]
 45%|████▌     | 29564/65536 [5:04:08<5:59:48,  1.67it/s]
 45%|████▌     | 29565/65536 [5:04:09<6:07:52,  1.63it/s]
 45%|████▌     | 29566/65536 [5:04:09<6:23:45,  1.56it/s]
 45%|████▌     | 29567/65536 [5:04:10<6:19:43,  1.58it/s]
 45%|████▌     | 29568/65536 [5:04:11<6:13:13,  1.61it/s]
 45%|████▌     | 29569/65536 [5:04:11<6:07:54,  1.63it/s]
 45%|████▌     | 29570/65536 [5:04:12<6:06:04,  1.64it/s]
 45%|████▌     | 29571/65536 [5:04:12<6:02:40,  1.65it/s]
 45%|████▌     | 29572/65536 [5:04:13<6:02:39,  1.65it/s]
 45%|████▌     | 29573/65536 [5:04:14<6:03:56,  1.65it/s]
 45%|████▌     | 29574/65536 [5:04:14<6:22:56,  1.57it/s]
 45%|████▌     | 29575/65536 [5:04:15<6:20:31,  1.58it/s]
 45%|████▌     | 29576/65536 [5:04:16<6:17:00,  1.59it/s]
 45%|████▌     | 29577/65536 [5:04:16<6:18:09,  1.58it/s]
 45%|████▌     | 29578/65536 [5:04:17<6:22:10,  1.57it/s]
 45%|████▌     | 29579/65536 [5:04:17<6:21:38,  1.57it/s]
 45%|████▌     | 29580/65536 [5:04:18<6:14:51,  1.60it/s]
                                                         
{'loss': 1.6896, 'learning_rate': 5.952920288967796e-07, 'epoch': 1825.93}
+
 45%|████▌     | 29580/65536 [5:04:18<6:14:51,  1.60it/s]
 45%|████▌     | 29581/65536 [5:04:19<6:22:16,  1.57it/s]
 45%|████▌     | 29582/65536 [5:04:19<6:28:35,  1.54it/s]
 45%|████▌     | 29583/65536 [5:04:20<6:30:47,  1.53it/s]
 45%|████▌     | 29584/65536 [5:04:21<6:16:36,  1.59it/s]
 45%|████▌     | 29585/65536 [5:04:21<6:15:49,  1.59it/s]
 45%|████▌     | 29586/65536 [5:04:22<6:09:19,  1.62it/s]
 45%|████▌     | 29587/65536 [5:04:23<6:19:05,  1.58it/s]
 45%|████▌     | 29588/65536 [5:04:23<6:18:02,  1.58it/s]
 45%|████▌     | 29589/65536 [5:04:24<6:13:39,  1.60it/s]
 45%|████▌     | 29590/65536 [5:04:24<6:12:34,  1.61it/s]
 45%|████▌     | 29591/65536 [5:04:25<6:09:18,  1.62it/s]
 45%|████▌     | 29592/65536 [5:04:26<6:11:57,  1.61it/s]
 45%|████▌     | 29593/65536 [5:04:26<6:10:08,  1.62it/s]
 45%|████▌     | 29594/65536 [5:04:27<6:13:38,  1.60it/s]
 45%|████▌     | 29595/65536 [5:04:27<6:06:56,  1.63it/s]
 45%|████▌     | 29596/65536 [5:04:28<6:06:47,  1.63it/s]
 45%|████▌     | 29597/65536 [5:04:29<6:09:54,  1.62it/s]
 45%|████▌     | 29598/65536 [5:04:29<6:22:51,  1.56it/s]
 45%|████▌     | 29599/65536 [5:04:30<6:20:08,  1.58it/s]
 45%|████▌     | 29600/65536 [5:04:31<6:33:48,  1.52it/s]
                                                         
{'loss': 1.6752, 'learning_rate': 5.950165299375535e-07, 'epoch': 1827.16}
+
 45%|████▌     | 29600/65536 [5:04:31<6:33:48,  1.52it/s]
 45%|████▌     | 29601/65536 [5:04:31<6:22:08,  1.57it/s]
 45%|████▌     | 29602/65536 [5:04:32<6:19:16,  1.58it/s]
 45%|████▌     | 29603/65536 [5:04:33<6:18:59,  1.58it/s]
 45%|████▌     | 29604/65536 [5:04:33<6:04:04,  1.64it/s]
 45%|████▌     | 29605/65536 [5:04:34<6:08:01,  1.63it/s]
 45%|████▌     | 29606/65536 [5:04:34<6:07:58,  1.63it/s]
 45%|████▌     | 29607/65536 [5:04:35<6:03:52,  1.65it/s]
 45%|████▌     | 29608/65536 [5:04:36<6:14:10,  1.60it/s]
 45%|████▌     | 29609/65536 [5:04:36<6:22:47,  1.56it/s]
 45%|████▌     | 29610/65536 [5:04:37<6:19:42,  1.58it/s]
 45%|████▌     | 29611/65536 [5:04:38<6:20:07,  1.58it/s]
 45%|████▌     | 29612/65536 [5:04:38<6:09:33,  1.62it/s]
 45%|████▌     | 29613/65536 [5:04:39<6:03:43,  1.65it/s]
 45%|████▌     | 29614/65536 [5:04:39<6:20:13,  1.57it/s]
 45%|████▌     | 29615/65536 [5:04:40<6:16:52,  1.59it/s]
 45%|████▌     | 29616/65536 [5:04:41<6:18:42,  1.58it/s]
 45%|████▌     | 29617/65536 [5:04:41<6:06:03,  1.64it/s]
 45%|████▌     | 29618/65536 [5:04:42<5:58:31,  1.67it/s]
 45%|████▌     | 29619/65536 [5:04:42<6:02:19,  1.65it/s]
 45%|████▌     | 29620/65536 [5:04:43<6:05:38,  1.64it/s]
                                                         
{'loss': 1.712, 'learning_rate': 5.947410309783275e-07, 'epoch': 1828.4}
+
 45%|████▌     | 29620/65536 [5:04:43<6:05:38,  1.64it/s]
 45%|████▌     | 29621/65536 [5:04:44<6:05:23,  1.64it/s]
 45%|████▌     | 29622/65536 [5:04:44<6:16:37,  1.59it/s]
 45%|████▌     | 29623/65536 [5:04:45<6:17:50,  1.58it/s]
 45%|████▌     | 29624/65536 [5:04:46<6:18:43,  1.58it/s]
 45%|████▌     | 29625/65536 [5:04:46<6:18:52,  1.58it/s]
 45%|████▌     | 29626/65536 [5:04:47<6:23:22,  1.56it/s]
 45%|████▌     | 29627/65536 [5:04:48<6:18:16,  1.58it/s]
 45%|████▌     | 29628/65536 [5:04:48<6:14:01,  1.60it/s]
 45%|████▌     | 29629/65536 [5:04:49<6:09:17,  1.62it/s]
 45%|████▌     | 29630/65536 [5:04:49<6:27:24,  1.54it/s]
 45%|████▌     | 29631/65536 [5:04:50<6:27:15,  1.55it/s]
 45%|████▌     | 29632/65536 [5:04:51<6:19:19,  1.58it/s]
 45%|████▌     | 29633/65536 [5:04:51<6:19:19,  1.58it/s]
 45%|████▌     | 29634/65536 [5:04:52<6:14:33,  1.60it/s]
 45%|████▌     | 29635/65536 [5:04:53<6:09:47,  1.62it/s]
 45%|████▌     | 29636/65536 [5:04:53<6:13:39,  1.60it/s]
 45%|████▌     | 29637/65536 [5:04:54<6:14:42,  1.60it/s]
 45%|████▌     | 29638/65536 [5:04:54<6:15:25,  1.59it/s]
 45%|████▌     | 29639/65536 [5:04:55<6:11:54,  1.61it/s]
 45%|████▌     | 29640/65536 [5:04:56<6:24:02,  1.56it/s]
                                                         
{'loss': 1.6502, 'learning_rate': 5.944655320191013e-07, 'epoch': 1829.63}
+
 45%|████▌     | 29640/65536 [5:04:56<6:24:02,  1.56it/s]
 45%|████▌     | 29641/65536 [5:04:56<6:26:15,  1.55it/s]
 45%|████▌     | 29642/65536 [5:04:57<6:14:15,  1.60it/s]
 45%|████▌     | 29643/65536 [5:04:58<6:21:22,  1.57it/s]
 45%|████▌     | 29644/65536 [5:04:58<6:20:04,  1.57it/s]
 45%|████▌     | 29645/65536 [5:04:59<6:15:04,  1.59it/s]
 45%|████▌     | 29646/65536 [5:04:59<6:07:51,  1.63it/s]
 45%|████▌     | 29647/65536 [5:05:00<6:30:01,  1.53it/s]
 45%|████▌     | 29648/65536 [5:05:01<6:27:16,  1.54it/s]
 45%|████▌     | 29649/65536 [5:05:01<6:26:12,  1.55it/s]
 45%|████▌     | 29650/65536 [5:05:02<6:16:37,  1.59it/s]
 45%|████▌     | 29651/65536 [5:05:03<6:07:21,  1.63it/s]
 45%|████▌     | 29652/65536 [5:05:03<6:10:14,  1.62it/s]
 45%|████▌     | 29653/65536 [5:05:04<6:05:06,  1.64it/s]
 45%|████▌     | 29654/65536 [5:05:05<6:15:48,  1.59it/s]
 45%|████▌     | 29655/65536 [5:05:05<6:10:34,  1.61it/s]
 45%|████▌     | 29656/65536 [5:05:06<6:13:02,  1.60it/s]
 45%|████▌     | 29657/65536 [5:05:06<6:13:17,  1.60it/s]
 45%|████▌     | 29658/65536 [5:05:07<6:20:07,  1.57it/s]
 45%|████▌     | 29659/65536 [5:05:08<6:10:40,  1.61it/s]
 45%|████▌     | 29660/65536 [5:05:08<6:07:59,  1.62it/s]
                                                         
{'loss': 1.7099, 'learning_rate': 5.941900330598752e-07, 'epoch': 1830.86}
+
 45%|████▌     | 29660/65536 [5:05:08<6:07:59,  1.62it/s]
 45%|████▌     | 29661/65536 [5:05:09<6:01:24,  1.65it/s]
 45%|████▌     | 29662/65536 [5:05:09<6:03:06,  1.65it/s]
 45%|████▌     | 29663/65536 [5:05:10<6:17:59,  1.58it/s]
 45%|████▌     | 29664/65536 [5:05:11<6:18:46,  1.58it/s]
 45%|████▌     | 29665/65536 [5:05:11<6:11:38,  1.61it/s]
 45%|████▌     | 29666/65536 [5:05:12<6:16:23,  1.59it/s]
 45%|████▌     | 29667/65536 [5:05:13<6:13:46,  1.60it/s]
 45%|████▌     | 29668/65536 [5:05:13<6:14:04,  1.60it/s]
 45%|████▌     | 29669/65536 [5:05:14<6:15:03,  1.59it/s]
 45%|████▌     | 29670/65536 [5:05:15<6:13:45,  1.60it/s]
 45%|████▌     | 29671/65536 [5:05:15<6:19:17,  1.58it/s]
 45%|████▌     | 29672/65536 [5:05:16<6:18:20,  1.58it/s]
 45%|████▌     | 29673/65536 [5:05:16<6:15:07,  1.59it/s]
 45%|████▌     | 29674/65536 [5:05:17<6:10:26,  1.61it/s]
 45%|████▌     | 29675/65536 [5:05:18<6:22:29,  1.56it/s]
 45%|████▌     | 29676/65536 [5:05:18<6:13:57,  1.60it/s]
 45%|████▌     | 29677/65536 [5:05:19<6:23:51,  1.56it/s]
 45%|████▌     | 29678/65536 [5:05:20<6:17:57,  1.58it/s]
 45%|████▌     | 29679/65536 [5:05:20<6:25:37,  1.55it/s]
 45%|████▌     | 29680/65536 [5:05:21<6:18:39,  1.58it/s]
                                                         
{'loss': 1.6656, 'learning_rate': 5.93914534100649e-07, 'epoch': 1832.1}
+
 45%|████▌     | 29680/65536 [5:05:21<6:18:39,  1.58it/s]
 45%|████▌     | 29681/65536 [5:05:21<6:12:39,  1.60it/s]
 45%|████▌     | 29682/65536 [5:05:22<6:06:25,  1.63it/s]
 45%|████▌     | 29683/65536 [5:05:23<6:15:36,  1.59it/s]
 45%|████▌     | 29684/65536 [5:05:23<6:18:28,  1.58it/s]
 45%|████▌     | 29685/65536 [5:05:24<6:25:16,  1.55it/s]
 45%|████▌     | 29686/65536 [5:05:25<6:19:15,  1.58it/s]
 45%|████▌     | 29687/65536 [5:05:25<6:12:20,  1.60it/s]
 45%|████▌     | 29688/65536 [5:05:26<6:06:56,  1.63it/s]
 45%|████▌     | 29689/65536 [5:05:26<6:06:58,  1.63it/s]
 45%|████▌     | 29690/65536 [5:05:27<6:14:09,  1.60it/s]
 45%|████▌     | 29691/65536 [5:05:28<6:32:42,  1.52it/s]
 45%|████▌     | 29692/65536 [5:05:28<6:21:54,  1.56it/s]
 45%|████▌     | 29693/65536 [5:05:29<6:19:40,  1.57it/s]
 45%|████▌     | 29694/65536 [5:05:30<6:15:09,  1.59it/s]
 45%|████▌     | 29695/65536 [5:05:30<6:29:34,  1.53it/s]
 45%|████▌     | 29696/65536 [5:05:31<6:32:51,  1.52it/s]
 45%|████▌     | 29697/65536 [5:05:32<6:19:57,  1.57it/s]
 45%|████▌     | 29698/65536 [5:05:32<6:20:56,  1.57it/s]
 45%|████▌     | 29699/65536 [5:05:33<6:16:10,  1.59it/s]
 45%|████▌     | 29700/65536 [5:05:34<6:20:15,  1.57it/s]
                                                         
{'loss': 1.6611, 'learning_rate': 5.936390351414228e-07, 'epoch': 1833.33}
+
 45%|████▌     | 29700/65536 [5:05:34<6:20:15,  1.57it/s]
 45%|████▌     | 29701/65536 [5:05:34<6:25:06,  1.55it/s]
 45%|████▌     | 29702/65536 [5:05:35<6:17:39,  1.58it/s]
 45%|████▌     | 29703/65536 [5:05:35<6:15:41,  1.59it/s]
 45%|████▌     | 29704/65536 [5:05:36<6:12:36,  1.60it/s]
 45%|████▌     | 29705/65536 [5:05:37<6:17:05,  1.58it/s]
 45%|████▌     | 29706/65536 [5:05:37<6:10:31,  1.61it/s]
 45%|████▌     | 29707/65536 [5:05:38<6:06:23,  1.63it/s]
 45%|████▌     | 29708/65536 [5:05:38<6:02:27,  1.65it/s]
 45%|████▌     | 29709/65536 [5:05:39<6:01:09,  1.65it/s]
 45%|████▌     | 29710/65536 [5:05:40<6:00:23,  1.66it/s]
 45%|████▌     | 29711/65536 [5:05:40<6:05:45,  1.63it/s]
 45%|████▌     | 29712/65536 [5:05:41<6:08:57,  1.62it/s]
 45%|████▌     | 29713/65536 [5:05:42<6:12:38,  1.60it/s]
 45%|████▌     | 29714/65536 [5:05:42<6:06:47,  1.63it/s]
 45%|████▌     | 29715/65536 [5:05:43<6:05:18,  1.63it/s]
 45%|████▌     | 29716/65536 [5:05:43<6:04:22,  1.64it/s]
 45%|████▌     | 29717/65536 [5:05:44<6:06:14,  1.63it/s]
 45%|████▌     | 29718/65536 [5:05:45<6:04:37,  1.64it/s]
 45%|████▌     | 29719/65536 [5:05:45<6:08:39,  1.62it/s]
 45%|████▌     | 29720/65536 [5:05:46<6:07:08,  1.63it/s]
                                                         
{'loss': 1.7461, 'learning_rate': 5.933635361821967e-07, 'epoch': 1834.57}
+
 45%|████▌     | 29720/65536 [5:05:46<6:07:08,  1.63it/s]
 45%|████▌     | 29721/65536 [5:05:46<6:07:47,  1.62it/s]
 45%|████▌     | 29722/65536 [5:05:47<6:06:03,  1.63it/s]
 45%|████▌     | 29723/65536 [5:05:48<6:20:44,  1.57it/s]
 45%|████▌     | 29724/65536 [5:05:48<6:25:51,  1.55it/s]
 45%|████▌     | 29725/65536 [5:05:49<6:17:03,  1.58it/s]
 45%|████▌     | 29726/65536 [5:05:50<6:16:30,  1.59it/s]
 45%|████▌     | 29727/65536 [5:05:50<6:13:54,  1.60it/s]
 45%|████▌     | 29728/65536 [5:05:51<6:26:21,  1.54it/s]
 45%|████▌     | 29729/65536 [5:05:52<6:15:35,  1.59it/s]
 45%|████▌     | 29730/65536 [5:05:52<6:19:49,  1.57it/s]
 45%|████▌     | 29731/65536 [5:05:53<6:17:53,  1.58it/s]
 45%|████▌     | 29732/65536 [5:05:53<6:08:40,  1.62it/s]
 45%|████▌     | 29733/65536 [5:05:54<6:06:57,  1.63it/s]
 45%|████▌     | 29734/65536 [5:05:55<6:07:14,  1.62it/s]
 45%|████▌     | 29735/65536 [5:05:55<6:13:43,  1.60it/s]
 45%|████▌     | 29736/65536 [5:05:56<6:11:43,  1.61it/s]
 45%|████▌     | 29737/65536 [5:05:57<6:15:10,  1.59it/s]
 45%|████▌     | 29738/65536 [5:05:57<6:21:24,  1.56it/s]
 45%|████▌     | 29739/65536 [5:05:58<6:23:14,  1.56it/s]
 45%|████▌     | 29740/65536 [5:05:58<6:14:59,  1.59it/s]
                                                         
{'loss': 1.7029, 'learning_rate': 5.930880372229705e-07, 'epoch': 1835.8}
+
 45%|████▌     | 29740/65536 [5:05:58<6:14:59,  1.59it/s]
 45%|████▌     | 29741/65536 [5:05:59<6:18:01,  1.58it/s]
 45%|████▌     | 29742/65536 [5:06:00<6:12:15,  1.60it/s]
 45%|████▌     | 29743/65536 [5:06:00<6:03:16,  1.64it/s]
 45%|████▌     | 29744/65536 [5:06:01<6:14:00,  1.59it/s]
 45%|████▌     | 29745/65536 [5:06:02<6:12:24,  1.60it/s]
 45%|████▌     | 29746/65536 [5:06:02<6:10:39,  1.61it/s]
 45%|████▌     | 29747/65536 [5:06:03<6:04:15,  1.64it/s]
 45%|████▌     | 29748/65536 [5:06:03<6:05:45,  1.63it/s]
 45%|████▌     | 29749/65536 [5:06:04<6:05:05,  1.63it/s]
 45%|████▌     | 29750/65536 [5:06:05<6:08:16,  1.62it/s]
 45%|████▌     | 29751/65536 [5:06:05<6:10:47,  1.61it/s]
 45%|████▌     | 29752/65536 [5:06:06<6:11:22,  1.61it/s]
 45%|████▌     | 29753/65536 [5:06:06<6:10:59,  1.61it/s]
 45%|████▌     | 29754/65536 [5:06:07<6:14:48,  1.59it/s]
 45%|████▌     | 29755/65536 [5:06:08<6:13:14,  1.60it/s]
 45%|████▌     | 29756/65536 [5:06:08<6:24:47,  1.55it/s]
 45%|████▌     | 29757/65536 [5:06:09<6:14:03,  1.59it/s]
 45%|████▌     | 29758/65536 [5:06:10<6:11:59,  1.60it/s]
 45%|████▌     | 29759/65536 [5:06:10<6:16:16,  1.58it/s]
 45%|████▌     | 29760/65536 [5:06:11<6:24:07,  1.55it/s]
                                                         
{'loss': 1.6585, 'learning_rate': 5.928125382637443e-07, 'epoch': 1837.04}
+
 45%|████▌     | 29760/65536 [5:06:11<6:24:07,  1.55it/s]
 45%|████▌     | 29761/65536 [5:06:12<6:16:41,  1.58it/s]
 45%|████▌     | 29762/65536 [5:06:12<6:11:32,  1.60it/s]
 45%|████▌     | 29763/65536 [5:06:13<6:10:02,  1.61it/s]
 45%|████▌     | 29764/65536 [5:06:13<6:09:35,  1.61it/s]
 45%|████▌     | 29765/65536 [5:06:14<6:05:48,  1.63it/s]
 45%|████▌     | 29766/65536 [5:06:15<6:06:44,  1.63it/s]
 45%|████▌     | 29767/65536 [5:06:15<6:04:28,  1.64it/s]
 45%|████▌     | 29768/65536 [5:06:16<6:07:33,  1.62it/s]
 45%|████▌     | 29769/65536 [5:06:17<6:21:29,  1.56it/s]
 45%|████▌     | 29770/65536 [5:06:17<6:16:13,  1.58it/s]
 45%|████▌     | 29771/65536 [5:06:18<6:18:25,  1.58it/s]
 45%|████▌     | 29772/65536 [5:06:18<6:14:25,  1.59it/s]
 45%|████▌     | 29773/65536 [5:06:19<6:04:06,  1.64it/s]
 45%|████▌     | 29774/65536 [5:06:20<6:15:39,  1.59it/s]
 45%|████▌     | 29775/65536 [5:06:20<6:15:36,  1.59it/s]
 45%|████▌     | 29776/65536 [5:06:21<6:27:27,  1.54it/s]
 45%|████▌     | 29777/65536 [5:06:22<6:23:19,  1.55it/s]
 45%|████▌     | 29778/65536 [5:06:22<6:24:44,  1.55it/s]
 45%|████▌     | 29779/65536 [5:06:23<6:27:43,  1.54it/s]
 45%|████▌     | 29780/65536 [5:06:24<6:23:12,  1.56it/s]
                                                         
{'loss': 1.6802, 'learning_rate': 5.925370393045181e-07, 'epoch': 1838.27}
+
 45%|████▌     | 29780/65536 [5:06:24<6:23:12,  1.56it/s]
 45%|████▌     | 29781/65536 [5:06:24<6:21:19,  1.56it/s]
 45%|████▌     | 29782/65536 [5:06:25<6:13:18,  1.60it/s]
 45%|████▌     | 29783/65536 [5:06:25<6:12:20,  1.60it/s]
 45%|████▌     | 29784/65536 [5:06:26<6:09:17,  1.61it/s]
 45%|████▌     | 29785/65536 [5:06:27<6:06:33,  1.63it/s]
 45%|████▌     | 29786/65536 [5:06:27<6:07:14,  1.62it/s]
 45%|████▌     | 29787/65536 [5:06:28<6:02:35,  1.64it/s]
 45%|████▌     | 29788/65536 [5:06:28<6:02:36,  1.64it/s]
 45%|████▌     | 29789/65536 [5:06:29<5:59:40,  1.66it/s]
 45%|████▌     | 29790/65536 [5:06:30<6:03:03,  1.64it/s]
 45%|████▌     | 29791/65536 [5:06:30<6:17:58,  1.58it/s]
 45%|████▌     | 29792/65536 [5:06:31<6:24:18,  1.55it/s]
 45%|████▌     | 29793/65536 [5:06:32<6:21:11,  1.56it/s]
 45%|████▌     | 29794/65536 [5:06:32<6:19:55,  1.57it/s]
 45%|████▌     | 29795/65536 [5:06:33<6:08:44,  1.62it/s]
 45%|████▌     | 29796/65536 [5:06:33<6:03:52,  1.64it/s]
 45%|████▌     | 29797/65536 [5:06:34<5:58:57,  1.66it/s]
 45%|████▌     | 29798/65536 [5:06:35<6:02:01,  1.65it/s]
 45%|████▌     | 29799/65536 [5:06:35<6:01:33,  1.65it/s]
 45%|████▌     | 29800/65536 [5:06:36<6:08:40,  1.62it/s]
                                                         
{'loss': 1.6939, 'learning_rate': 5.92261540345292e-07, 'epoch': 1839.51}
+
 45%|████▌     | 29800/65536 [5:06:36<6:08:40,  1.62it/s]
 45%|████▌     | 29801/65536 [5:06:37<6:14:18,  1.59it/s]
 45%|████▌     | 29802/65536 [5:06:37<6:19:03,  1.57it/s]
 45%|████▌     | 29803/65536 [5:06:38<6:13:45,  1.59it/s]
 45%|████▌     | 29804/65536 [5:06:38<6:11:00,  1.61it/s]
 45%|████▌     | 29805/65536 [5:06:39<6:13:52,  1.59it/s]
 45%|████▌     | 29806/65536 [5:06:40<6:13:11,  1.60it/s]
 45%|████▌     | 29807/65536 [5:06:40<6:08:34,  1.62it/s]
 45%|████▌     | 29808/65536 [5:06:41<6:17:03,  1.58it/s]
 45%|████▌     | 29809/65536 [5:06:42<6:22:52,  1.56it/s]
 45%|████▌     | 29810/65536 [5:06:42<6:15:31,  1.59it/s]
 45%|████▌     | 29811/65536 [5:06:43<6:12:17,  1.60it/s]
 45%|████▌     | 29812/65536 [5:06:43<6:02:26,  1.64it/s]
 45%|████▌     | 29813/65536 [5:06:44<6:17:32,  1.58it/s]
 45%|████▌     | 29814/65536 [5:06:45<6:08:28,  1.62it/s]
 45%|████▌     | 29815/65536 [5:06:45<6:14:27,  1.59it/s]
 45%|████▌     | 29816/65536 [5:06:46<6:34:07,  1.51it/s]
 45%|████▌     | 29817/65536 [5:06:47<6:21:13,  1.56it/s]
 45%|████▌     | 29818/65536 [5:06:47<6:21:02,  1.56it/s]
 46%|████▌     | 29819/65536 [5:06:48<6:23:32,  1.55it/s]
 46%|████▌     | 29820/65536 [5:06:49<6:18:53,  1.57it/s]
                                                         
{'loss': 1.6721, 'learning_rate': 5.919860413860658e-07, 'epoch': 1840.74}
+
 46%|████▌     | 29820/65536 [5:06:49<6:18:53,  1.57it/s]
 46%|████▌     | 29821/65536 [5:06:49<6:15:01,  1.59it/s]
 46%|████▌     | 29822/65536 [5:06:50<6:09:04,  1.61it/s]
 46%|████▌     | 29823/65536 [5:06:50<6:06:54,  1.62it/s]
 46%|████▌     | 29824/65536 [5:06:51<6:04:14,  1.63it/s]
 46%|████▌     | 29825/65536 [5:06:52<6:18:17,  1.57it/s]
 46%|████▌     | 29826/65536 [5:06:52<6:17:46,  1.58it/s]
 46%|████▌     | 29827/65536 [5:06:53<6:11:24,  1.60it/s]
 46%|████▌     | 29828/65536 [5:06:54<6:12:11,  1.60it/s]
 46%|████▌     | 29829/65536 [5:06:54<6:11:46,  1.60it/s]
 46%|████▌     | 29830/65536 [5:06:55<6:07:37,  1.62it/s]
 46%|████▌     | 29831/65536 [5:06:55<6:10:52,  1.60it/s]
 46%|████▌     | 29832/65536 [5:06:56<6:10:25,  1.61it/s]
 46%|████▌     | 29833/65536 [5:06:57<6:10:56,  1.60it/s]
 46%|████▌     | 29834/65536 [5:06:57<6:05:22,  1.63it/s]
 46%|████▌     | 29835/65536 [5:06:58<6:15:18,  1.59it/s]
 46%|████▌     | 29836/65536 [5:06:59<6:20:50,  1.56it/s]
 46%|████▌     | 29837/65536 [5:06:59<6:17:02,  1.58it/s]
 46%|████▌     | 29838/65536 [5:07:00<6:08:01,  1.62it/s]
 46%|████▌     | 29839/65536 [5:07:00<6:04:21,  1.63it/s]
 46%|████▌     | 29840/65536 [5:07:01<6:10:15,  1.61it/s]
                                                         
{'loss': 1.6698, 'learning_rate': 5.917105424268397e-07, 'epoch': 1841.98}
+
 46%|████▌     | 29840/65536 [5:07:01<6:10:15,  1.61it/s]
 46%|████▌     | 29841/65536 [5:07:02<6:22:29,  1.56it/s]
 46%|████▌     | 29842/65536 [5:07:02<6:12:29,  1.60it/s]
 46%|████▌     | 29843/65536 [5:07:03<6:07:44,  1.62it/s]
 46%|████▌     | 29844/65536 [5:07:04<6:09:28,  1.61it/s]
 46%|████▌     | 29845/65536 [5:07:04<6:05:37,  1.63it/s]
 46%|████▌     | 29846/65536 [5:07:05<6:10:11,  1.61it/s]
 46%|████▌     | 29847/65536 [5:07:05<6:14:13,  1.59it/s]
 46%|████▌     | 29848/65536 [5:07:06<6:09:08,  1.61it/s]
 46%|████▌     | 29849/65536 [5:07:07<6:11:10,  1.60it/s]
 46%|████▌     | 29850/65536 [5:07:07<6:21:00,  1.56it/s]
 46%|████▌     | 29851/65536 [5:07:08<6:28:53,  1.53it/s]
 46%|████▌     | 29852/65536 [5:07:09<6:19:13,  1.57it/s]
 46%|████▌     | 29853/65536 [5:07:09<6:12:34,  1.60it/s]
 46%|████▌     | 29854/65536 [5:07:10<6:10:47,  1.60it/s]
 46%|████▌     | 29855/65536 [5:07:10<5:59:28,  1.65it/s]
 46%|████▌     | 29856/65536 [5:07:11<6:02:28,  1.64it/s]
 46%|████▌     | 29857/65536 [5:07:12<6:21:37,  1.56it/s]
 46%|████▌     | 29858/65536 [5:07:12<6:11:39,  1.60it/s]
 46%|████▌     | 29859/65536 [5:07:13<6:07:55,  1.62it/s]
 46%|████▌     | 29860/65536 [5:07:14<6:05:40,  1.63it/s]
                                                         
{'loss': 1.7243, 'learning_rate': 5.914350434676136e-07, 'epoch': 1843.21}
+
 46%|████▌     | 29860/65536 [5:07:14<6:05:40,  1.63it/s]
 46%|████▌     | 29861/65536 [5:07:14<6:07:39,  1.62it/s]
 46%|████▌     | 29862/65536 [5:07:15<6:03:01,  1.64it/s]
 46%|████▌     | 29863/65536 [5:07:15<5:56:43,  1.67it/s]
 46%|████▌     | 29864/65536 [5:07:16<6:01:04,  1.65it/s]
 46%|████▌     | 29865/65536 [5:07:17<6:20:32,  1.56it/s]
 46%|████▌     | 29866/65536 [5:07:17<6:28:50,  1.53it/s]
 46%|████▌     | 29867/65536 [5:07:18<6:15:37,  1.58it/s]
 46%|████▌     | 29868/65536 [5:07:19<6:12:06,  1.60it/s]
 46%|████▌     | 29869/65536 [5:07:19<6:10:17,  1.61it/s]
 46%|████▌     | 29870/65536 [5:07:20<6:15:49,  1.58it/s]
 46%|████▌     | 29871/65536 [5:07:20<6:11:16,  1.60it/s]
 46%|████▌     | 29872/65536 [5:07:21<6:09:17,  1.61it/s]
 46%|████▌     | 29873/65536 [5:07:22<6:20:07,  1.56it/s]
 46%|████▌     | 29874/65536 [5:07:22<6:11:37,  1.60it/s]
 46%|████▌     | 29875/65536 [5:07:23<6:08:42,  1.61it/s]
 46%|████▌     | 29876/65536 [5:07:24<6:07:17,  1.62it/s]
 46%|████▌     | 29877/65536 [5:07:24<6:03:38,  1.63it/s]
 46%|████▌     | 29878/65536 [5:07:25<6:05:14,  1.63it/s]
 46%|████▌     | 29879/65536 [5:07:25<5:58:50,  1.66it/s]
 46%|████▌     | 29880/65536 [5:07:26<5:59:53,  1.65it/s]
                                                         
{'loss': 1.6764, 'learning_rate': 5.911595445083874e-07, 'epoch': 1844.44}
+
 46%|████▌     | 29880/65536 [5:07:26<5:59:53,  1.65it/s]
 46%|████▌     | 29881/65536 [5:07:27<6:10:07,  1.61it/s]
 46%|████▌     | 29882/65536 [5:07:27<6:10:56,  1.60it/s]
 46%|████▌     | 29883/65536 [5:07:28<6:18:11,  1.57it/s]
 46%|████▌     | 29884/65536 [5:07:29<6:19:58,  1.56it/s]
 46%|████▌     | 29885/65536 [5:07:29<6:11:23,  1.60it/s]
 46%|████▌     | 29886/65536 [5:07:30<6:05:47,  1.62it/s]
 46%|████▌     | 29887/65536 [5:07:30<6:14:39,  1.59it/s]
 46%|████▌     | 29888/65536 [5:07:31<6:17:26,  1.57it/s]
 46%|████▌     | 29889/65536 [5:07:32<6:14:13,  1.59it/s]
 46%|████▌     | 29890/65536 [5:07:32<6:30:28,  1.52it/s]
 46%|████▌     | 29891/65536 [5:07:33<6:21:51,  1.56it/s]
 46%|████▌     | 29892/65536 [5:07:34<6:18:28,  1.57it/s]
 46%|████▌     | 29893/65536 [5:07:34<6:22:29,  1.55it/s]
 46%|████▌     | 29894/65536 [5:07:35<6:31:27,  1.52it/s]
 46%|████▌     | 29895/65536 [5:07:36<6:22:40,  1.55it/s]
 46%|████▌     | 29896/65536 [5:07:36<6:10:23,  1.60it/s]
 46%|████▌     | 29897/65536 [5:07:37<6:01:28,  1.64it/s]
 46%|████▌     | 29898/65536 [5:07:37<6:01:30,  1.64it/s]
 46%|████▌     | 29899/65536 [5:07:38<6:03:29,  1.63it/s]
 46%|████▌     | 29900/65536 [5:07:39<5:55:21,  1.67it/s]
                                                         
{'loss': 1.6299, 'learning_rate': 5.908840455491613e-07, 'epoch': 1845.68}
+
 46%|████▌     | 29900/65536 [5:07:39<5:55:21,  1.67it/s]
 46%|████▌     | 29901/65536 [5:07:39<5:57:39,  1.66it/s]
 46%|████▌     | 29902/65536 [5:07:40<5:55:42,  1.67it/s]
 46%|████▌     | 29903/65536 [5:07:40<6:02:26,  1.64it/s]
 46%|████▌     | 29904/65536 [5:07:41<6:03:40,  1.63it/s]
 46%|████▌     | 29905/65536 [5:07:42<5:59:12,  1.65it/s]
 46%|████▌     | 29906/65536 [5:07:42<6:08:18,  1.61it/s]
 46%|████▌     | 29907/65536 [5:07:43<6:05:57,  1.62it/s]
 46%|████▌     | 29908/65536 [5:07:43<6:06:02,  1.62it/s]
 46%|████▌     | 29909/65536 [5:07:44<6:02:35,  1.64it/s]
 46%|████▌     | 29910/65536 [5:07:45<6:02:08,  1.64it/s]
 46%|████▌     | 29911/65536 [5:07:45<6:07:20,  1.62it/s]
 46%|████▌     | 29912/65536 [5:07:46<6:07:01,  1.62it/s]
 46%|████▌     | 29913/65536 [5:07:46<6:01:21,  1.64it/s]
 46%|████▌     | 29914/65536 [5:07:47<6:00:47,  1.65it/s]
 46%|████▌     | 29915/65536 [5:07:48<6:02:37,  1.64it/s]
 46%|████▌     | 29916/65536 [5:07:48<6:10:12,  1.60it/s]
 46%|████▌     | 29917/65536 [5:07:49<6:02:48,  1.64it/s]
 46%|████▌     | 29918/65536 [5:07:50<6:00:41,  1.65it/s]
 46%|████▌     | 29919/65536 [5:07:50<6:00:47,  1.65it/s]
 46%|████▌     | 29920/65536 [5:07:51<6:07:45,  1.61it/s]
                                                         
{'loss': 1.6761, 'learning_rate': 5.90608546589935e-07, 'epoch': 1846.91}
+
 46%|████▌     | 29920/65536 [5:07:51<6:07:45,  1.61it/s]
 46%|████▌     | 29921/65536 [5:07:51<6:01:37,  1.64it/s]
 46%|████▌     | 29922/65536 [5:07:52<6:12:49,  1.59it/s]
 46%|████▌     | 29923/65536 [5:07:53<6:15:12,  1.58it/s]
 46%|████▌     | 29924/65536 [5:07:53<6:05:01,  1.63it/s]
 46%|████▌     | 29925/65536 [5:07:54<6:01:04,  1.64it/s]
 46%|████▌     | 29926/65536 [5:07:54<5:57:02,  1.66it/s]
 46%|████▌     | 29927/65536 [5:07:55<6:02:04,  1.64it/s]
 46%|████▌     | 29928/65536 [5:07:56<5:56:10,  1.67it/s]
 46%|████▌     | 29929/65536 [5:07:56<5:52:03,  1.69it/s]
 46%|████▌     | 29930/65536 [5:07:57<5:54:56,  1.67it/s]
 46%|████▌     | 29931/65536 [5:07:57<6:04:22,  1.63it/s]
 46%|████▌     | 29932/65536 [5:07:58<6:00:50,  1.64it/s]
 46%|████▌     | 29933/65536 [5:07:59<6:03:56,  1.63it/s]
 46%|████▌     | 29934/65536 [5:07:59<6:01:59,  1.64it/s]
 46%|████▌     | 29935/65536 [5:08:00<6:01:04,  1.64it/s]
 46%|████▌     | 29936/65536 [5:08:01<6:10:00,  1.60it/s]
 46%|████▌     | 29937/65536 [5:08:01<6:12:11,  1.59it/s]
 46%|████▌     | 29938/65536 [5:08:02<6:21:15,  1.56it/s]
 46%|████▌     | 29939/65536 [5:08:03<6:15:03,  1.58it/s]
 46%|████▌     | 29940/65536 [5:08:03<6:14:15,  1.59it/s]
                                                         
{'loss': 1.689, 'learning_rate': 5.90333047630709e-07, 'epoch': 1848.15}
+
 46%|████▌     | 29940/65536 [5:08:03<6:14:15,  1.59it/s]
 46%|████▌     | 29941/65536 [5:08:04<6:03:34,  1.63it/s]
 46%|████▌     | 29942/65536 [5:08:04<6:13:05,  1.59it/s]
 46%|████▌     | 29943/65536 [5:08:05<6:11:48,  1.60it/s]
 46%|████▌     | 29944/65536 [5:08:06<6:10:37,  1.60it/s]
 46%|████▌     | 29945/65536 [5:08:06<6:07:33,  1.61it/s]
 46%|████▌     | 29946/65536 [5:08:07<6:04:00,  1.63it/s]
 46%|████▌     | 29947/65536 [5:08:07<5:59:21,  1.65it/s]
 46%|████▌     | 29948/65536 [5:08:08<6:11:35,  1.60it/s]
 46%|████▌     | 29949/65536 [5:08:09<6:10:57,  1.60it/s]
 46%|████▌     | 29950/65536 [5:08:09<6:04:19,  1.63it/s]
 46%|████▌     | 29951/65536 [5:08:10<6:03:23,  1.63it/s]
 46%|████▌     | 29952/65536 [5:08:11<6:05:34,  1.62it/s]
 46%|████▌     | 29953/65536 [5:08:11<5:58:06,  1.66it/s]
 46%|████▌     | 29954/65536 [5:08:12<6:05:02,  1.62it/s]
 46%|████▌     | 29955/65536 [5:08:12<6:11:04,  1.60it/s]
 46%|████▌     | 29956/65536 [5:08:13<5:59:47,  1.65it/s]
 46%|████▌     | 29957/65536 [5:08:14<6:03:36,  1.63it/s]
 46%|████▌     | 29958/65536 [5:08:14<5:57:59,  1.66it/s]
 46%|████▌     | 29959/65536 [5:08:15<6:01:14,  1.64it/s]
 46%|████▌     | 29960/65536 [5:08:15<6:08:23,  1.61it/s]
                                                         
{'loss': 1.6907, 'learning_rate': 5.900575486714828e-07, 'epoch': 1849.38}
+
 46%|████▌     | 29960/65536 [5:08:15<6:08:23,  1.61it/s]
 46%|████▌     | 29961/65536 [5:08:16<6:14:33,  1.58it/s]
 46%|████▌     | 29962/65536 [5:08:17<6:06:50,  1.62it/s]
 46%|████▌     | 29963/65536 [5:08:17<6:11:58,  1.59it/s]
 46%|████▌     | 29964/65536 [5:08:18<6:04:08,  1.63it/s]
 46%|████▌     | 29965/65536 [5:08:19<6:07:03,  1.62it/s]
 46%|████▌     | 29966/65536 [5:08:19<6:10:15,  1.60it/s]
 46%|████▌     | 29967/65536 [5:08:20<6:02:37,  1.63it/s]
 46%|████▌     | 29968/65536 [5:08:20<5:56:51,  1.66it/s]
 46%|████▌     | 29969/65536 [5:08:21<6:05:48,  1.62it/s]
 46%|████▌     | 29970/65536 [5:08:22<5:58:41,  1.65it/s]
 46%|████▌     | 29971/65536 [5:08:22<6:06:35,  1.62it/s]
 46%|████▌     | 29972/65536 [5:08:23<5:58:50,  1.65it/s]
 46%|████▌     | 29973/65536 [5:08:23<6:00:59,  1.64it/s]
 46%|████▌     | 29974/65536 [5:08:24<6:08:41,  1.61it/s]
 46%|████▌     | 29975/65536 [5:08:25<5:59:18,  1.65it/s]
 46%|████▌     | 29976/65536 [5:08:25<6:00:04,  1.65it/s]
 46%|████▌     | 29977/65536 [5:08:26<6:06:11,  1.62it/s]
 46%|████▌     | 29978/65536 [5:08:26<6:04:43,  1.62it/s]
 46%|████▌     | 29979/65536 [5:08:27<6:26:15,  1.53it/s]
 46%|████▌     | 29980/65536 [5:08:28<6:17:25,  1.57it/s]
                                                         
{'loss': 1.6604, 'learning_rate': 5.897820497122566e-07, 'epoch': 1850.62}
+
 46%|████▌     | 29980/65536 [5:08:28<6:17:25,  1.57it/s]
 46%|████▌     | 29981/65536 [5:08:28<6:07:02,  1.61it/s]
 46%|████▌     | 29982/65536 [5:08:29<6:08:12,  1.61it/s]
 46%|████▌     | 29983/65536 [5:08:30<6:01:18,  1.64it/s]
 46%|████▌     | 29984/65536 [5:08:30<5:57:58,  1.66it/s]
 46%|████▌     | 29985/65536 [5:08:31<5:58:22,  1.65it/s]
 46%|████▌     | 29986/65536 [5:08:31<5:53:39,  1.68it/s]
 46%|████▌     | 29987/65536 [5:08:32<6:10:21,  1.60it/s]
 46%|████▌     | 29988/65536 [5:08:33<6:10:25,  1.60it/s]
 46%|████▌     | 29989/65536 [5:08:33<6:12:08,  1.59it/s]
 46%|████▌     | 29990/65536 [5:08:34<6:03:23,  1.63it/s]
 46%|████▌     | 29991/65536 [5:08:35<6:04:22,  1.63it/s]
 46%|████▌     | 29992/65536 [5:08:35<6:01:05,  1.64it/s]
 46%|████▌     | 29993/65536 [5:08:36<5:56:43,  1.66it/s]
 46%|████▌     | 29994/65536 [5:08:36<6:03:04,  1.63it/s]
 46%|████▌     | 29995/65536 [5:08:37<6:11:34,  1.59it/s]
 46%|████▌     | 29996/65536 [5:08:38<6:11:26,  1.59it/s]
 46%|████▌     | 29997/65536 [5:08:38<6:14:04,  1.58it/s]
 46%|████▌     | 29998/65536 [5:08:39<6:07:04,  1.61it/s]
 46%|████▌     | 29999/65536 [5:08:39<5:58:26,  1.65it/s]
 46%|████▌     | 30000/65536 [5:08:40<5:57:43,  1.66it/s]
                                                         
{'loss': 1.6979, 'learning_rate': 5.895065507530305e-07, 'epoch': 1851.85}
+
 46%|████▌     | 30000/65536 [5:08:40<5:57:43,  1.66it/s]Some non-default generation parameters are set in the model config. These should go into a GenerationConfig file (https://huggingface.co./docs/transformers/generation_strategies#save-a-custom-decoding-strategy-with-your-model) instead. This warning will be raised to an exception in v4.41.
+Non-default generation parameters: {'max_length': 200, 'early_stopping': True, 'num_beams': 5, 'forced_eos_token_id': 2}
+
 46%|████▌     | 30001/65536 [5:08:59<61:38:07,  6.24s/it]
 46%|████▌     | 30002/65536 [5:09:00<44:54:24,  4.55s/it]/opt/conda/lib/python3.10/multiprocessing/popen_fork.py:66: RuntimeWarning: os.fork() was called. os.fork() is incompatible with multithreaded code, and JAX is multithreaded, so this will likely lead to a deadlock.
+  self.pid = os.fork()
+
 46%|████▌     | 30003/65536 [5:09:01<33:32:37,  3.40s/it]
 46%|████▌     | 30004/65536 [5:09:01<25:19:26,  2.57s/it]
 46%|████▌     | 30005/65536 [5:09:02<19:32:08,  1.98s/it]
 46%|████▌     | 30006/65536 [5:09:03<15:27:53,  1.57s/it]
 46%|████▌     | 30007/65536 [5:09:03<12:47:58,  1.30s/it]
 46%|████▌     | 30008/65536 [5:09:04<10:41:29,  1.08s/it]
 46%|████▌     | 30009/65536 [5:09:05<9:26:36,  1.05it/s] 
 46%|████▌     | 30010/65536 [5:09:05<8:27:03,  1.17it/s]
 46%|████▌     | 30011/65536 [5:09:06<7:45:18,  1.27it/s]
 46%|████▌     | 30012/65536 [5:09:06<7:14:28,  1.36it/s]
 46%|████▌     | 30013/65536 [5:09:07<6:51:39,  1.44it/s]
 46%|████▌     | 30014/65536 [5:09:08<6:42:40,  1.47it/s]
 46%|████▌     | 30015/65536 [5:09:08<6:34:08,  1.50it/s]
 46%|████▌     | 30016/65536 [5:09:09<6:34:45,  1.50it/s]
 46%|████▌     | 30017/65536 [5:09:10<6:40:32,  1.48it/s]
 46%|████▌     | 30018/65536 [5:09:10<6:32:11,  1.51it/s]
 46%|████▌     | 30019/65536 [5:09:11<6:36:24,  1.49it/s]
 46%|████▌     | 30020/65536 [5:09:12<6:22:52,  1.55it/s]
                                                         
{'loss': 1.6912, 'learning_rate': 5.892310517938041e-07, 'epoch': 1853.09}
+
 46%|████▌     | 30020/65536 [5:09:12<6:22:52,  1.55it/s]
 46%|████▌     | 30021/65536 [5:09:12<6:32:04,  1.51it/s]
 46%|████▌     | 30022/65536 [5:09:13<6:29:02,  1.52it/s]
 46%|████▌     | 30023/65536 [5:09:14<6:22:25,  1.55it/s]
 46%|████▌     | 30024/65536 [5:09:14<6:15:18,  1.58it/s]
 46%|████▌     | 30025/65536 [5:09:15<6:14:51,  1.58it/s]
 46%|████▌     | 30026/65536 [5:09:15<6:15:32,  1.58it/s]
 46%|████▌     | 30027/65536 [5:09:16<6:12:55,  1.59it/s]
 46%|████▌     | 30028/65536 [5:09:17<6:18:19,  1.56it/s]
 46%|████▌     | 30029/65536 [5:09:17<6:10:31,  1.60it/s]
 46%|████▌     | 30030/65536 [5:09:18<6:13:29,  1.58it/s]
 46%|████▌     | 30031/65536 [5:09:19<6:13:53,  1.58it/s]
 46%|████▌     | 30032/65536 [5:09:19<6:09:12,  1.60it/s]
 46%|████▌     | 30033/65536 [5:09:20<6:05:24,  1.62it/s]
 46%|████▌     | 30034/65536 [5:09:20<6:01:45,  1.64it/s]
 46%|████▌     | 30035/65536 [5:09:21<6:16:53,  1.57it/s]
 46%|████▌     | 30036/65536 [5:09:22<6:20:09,  1.56it/s]
 46%|████▌     | 30037/65536 [5:09:22<6:15:34,  1.58it/s]
 46%|████▌     | 30038/65536 [5:09:23<6:09:29,  1.60it/s]
 46%|████▌     | 30039/65536 [5:09:23<6:00:16,  1.64it/s]
 46%|████▌     | 30040/65536 [5:09:24<5:57:04,  1.66it/s]
                                                         
{'loss': 1.6986, 'learning_rate': 5.889555528345781e-07, 'epoch': 1854.32}
+
 46%|████▌     | 30040/65536 [5:09:24<5:57:04,  1.66it/s]
 46%|████▌     | 30041/65536 [5:09:25<6:02:26,  1.63it/s]
 46%|████▌     | 30042/65536 [5:09:25<6:08:24,  1.61it/s]
 46%|████▌     | 30043/65536 [5:09:26<6:18:41,  1.56it/s]
 46%|████▌     | 30044/65536 [5:09:27<6:05:53,  1.62it/s]
 46%|████▌     | 30045/65536 [5:09:27<6:08:24,  1.61it/s]
 46%|████▌     | 30046/65536 [5:09:28<6:16:06,  1.57it/s]
 46%|████▌     | 30047/65536 [5:09:29<6:12:43,  1.59it/s]
 46%|████▌     | 30048/65536 [5:09:29<6:12:28,  1.59it/s]
 46%|████▌     | 30049/65536 [5:09:30<6:17:36,  1.57it/s]
 46%|████▌     | 30050/65536 [5:09:30<6:20:20,  1.56it/s]
 46%|████▌     | 30051/65536 [5:09:31<6:17:25,  1.57it/s]
 46%|████▌     | 30052/65536 [5:09:32<6:25:16,  1.53it/s]
 46%|████▌     | 30053/65536 [5:09:32<6:22:12,  1.55it/s]
 46%|████▌     | 30054/65536 [5:09:33<6:23:30,  1.54it/s]
 46%|████▌     | 30055/65536 [5:09:34<6:14:28,  1.58it/s]
 46%|████▌     | 30056/65536 [5:09:34<6:14:10,  1.58it/s]
 46%|████▌     | 30057/65536 [5:09:35<6:22:37,  1.55it/s]
 46%|████▌     | 30058/65536 [5:09:36<6:13:49,  1.58it/s]
 46%|████▌     | 30059/65536 [5:09:36<6:12:28,  1.59it/s]
 46%|████▌     | 30060/65536 [5:09:37<6:09:39,  1.60it/s]
                                                         
{'loss': 1.6842, 'learning_rate': 5.886800538753519e-07, 'epoch': 1855.56}
+
 46%|████▌     | 30060/65536 [5:09:37<6:09:39,  1.60it/s]
 46%|████▌     | 30061/65536 [5:09:37<6:16:14,  1.57it/s]
 46%|████▌     | 30062/65536 [5:09:38<6:13:11,  1.58it/s]
 46%|████▌     | 30063/65536 [5:09:39<6:04:58,  1.62it/s]
 46%|████▌     | 30064/65536 [5:09:39<6:09:44,  1.60it/s]
 46%|████▌     | 30065/65536 [5:09:40<6:13:30,  1.58it/s]
\ No newline at end of file