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"
@@ -1553,4 +1553,508 @@ Non-default generation parameters: {'max_length': 200, 'early_stopping': True, '
 
 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
+
 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]
 46%|████▌     | 30066/65536 [5:09:41<6:23:42,  1.54it/s]
 46%|████▌     | 30067/65536 [5:09:41<6:19:00,  1.56it/s]
 46%|████▌     | 30068/65536 [5:09:42<6:25:26,  1.53it/s]
 46%|████▌     | 30069/65536 [5:09:43<6:27:20,  1.53it/s]
 46%|████▌     | 30070/65536 [5:09:43<6:23:55,  1.54it/s]
 46%|████▌     | 30071/65536 [5:09:44<6:17:41,  1.56it/s]
 46%|████▌     | 30072/65536 [5:09:44<6:08:34,  1.60it/s]
 46%|████▌     | 30073/65536 [5:09:45<6:14:19,  1.58it/s]
 46%|████▌     | 30074/65536 [5:09:46<6:22:43,  1.54it/s]
 46%|████▌     | 30075/65536 [5:09:46<6:24:21,  1.54it/s]
 46%|████▌     | 30076/65536 [5:09:47<6:34:44,  1.50it/s]
 46%|████▌     | 30077/65536 [5:09:48<6:45:22,  1.46it/s]
 46%|████▌     | 30078/65536 [5:09:49<6:45:57,  1.46it/s]
 46%|████▌     | 30079/65536 [5:09:49<6:42:58,  1.47it/s]
 46%|████▌     | 30080/65536 [5:09:50<6:27:56,  1.52it/s]
                                                         
{'loss': 1.653, 'learning_rate': 5.884045549161258e-07, 'epoch': 1856.79}
+
 46%|████▌     | 30080/65536 [5:09:50<6:27:56,  1.52it/s]
 46%|████▌     | 30081/65536 [5:09:50<6:18:21,  1.56it/s]
 46%|████▌     | 30082/65536 [5:09:51<6:10:51,  1.59it/s]
 46%|████▌     | 30083/65536 [5:09:52<6:07:21,  1.61it/s]
 46%|████▌     | 30084/65536 [5:09:52<6:18:48,  1.56it/s]
 46%|████▌     | 30085/65536 [5:09:53<6:08:01,  1.61it/s]
 46%|████▌     | 30086/65536 [5:09:54<6:06:23,  1.61it/s]
 46%|████▌     | 30087/65536 [5:09:54<6:02:23,  1.63it/s]
 46%|████▌     | 30088/65536 [5:09:55<5:54:52,  1.66it/s]
 46%|████▌     | 30089/65536 [5:09:55<5:57:14,  1.65it/s]
 46%|████▌     | 30090/65536 [5:09:56<6:11:15,  1.59it/s]
 46%|████▌     | 30091/65536 [5:09:57<6:16:35,  1.57it/s]
 46%|████▌     | 30092/65536 [5:09:57<6:13:54,  1.58it/s]
 46%|████▌     | 30093/65536 [5:09:58<6:13:03,  1.58it/s]
 46%|████▌     | 30094/65536 [5:09:59<6:21:35,  1.55it/s]
 46%|████▌     | 30095/65536 [5:09:59<6:19:20,  1.56it/s]
 46%|████▌     | 30096/65536 [5:10:00<6:24:04,  1.54it/s]
 46%|████▌     | 30097/65536 [5:10:00<6:15:05,  1.57it/s]
 46%|████▌     | 30098/65536 [5:10:01<6:06:42,  1.61it/s]
 46%|████▌     | 30099/65536 [5:10:02<6:00:48,  1.64it/s]
 46%|████▌     | 30100/65536 [5:10:02<6:11:00,  1.59it/s]
                                                         
{'loss': 1.7042, 'learning_rate': 5.881290559568997e-07, 'epoch': 1858.02}
+
 46%|████▌     | 30100/65536 [5:10:02<6:11:00,  1.59it/s]
 46%|████▌     | 30101/65536 [5:10:03<6:08:41,  1.60it/s]
 46%|████▌     | 30102/65536 [5:10:04<6:10:52,  1.59it/s]
 46%|████▌     | 30103/65536 [5:10:04<6:07:36,  1.61it/s]
 46%|████▌     | 30104/65536 [5:10:05<6:28:58,  1.52it/s]
 46%|████▌     | 30105/65536 [5:10:06<6:24:47,  1.53it/s]
 46%|████▌     | 30106/65536 [5:10:06<6:18:24,  1.56it/s]
 46%|████▌     | 30107/65536 [5:10:07<6:25:45,  1.53it/s]
 46%|████▌     | 30108/65536 [5:10:08<6:31:08,  1.51it/s]
 46%|████▌     | 30109/65536 [5:10:08<6:26:09,  1.53it/s]
 46%|████▌     | 30110/65536 [5:10:09<6:20:49,  1.55it/s]
 46%|████▌     | 30111/65536 [5:10:09<6:15:56,  1.57it/s]
 46%|████▌     | 30112/65536 [5:10:10<6:06:15,  1.61it/s]
 46%|████▌     | 30113/65536 [5:10:11<6:13:41,  1.58it/s]
 46%|████▌     | 30114/65536 [5:10:11<6:17:07,  1.57it/s]
 46%|████▌     | 30115/65536 [5:10:12<6:14:22,  1.58it/s]
 46%|████▌     | 30116/65536 [5:10:13<6:26:23,  1.53it/s]
 46%|████▌     | 30117/65536 [5:10:13<6:32:13,  1.51it/s]
 46%|████▌     | 30118/65536 [5:10:14<6:24:27,  1.54it/s]
 46%|████▌     | 30119/65536 [5:10:15<6:23:10,  1.54it/s]
 46%|████▌     | 30120/65536 [5:10:15<6:30:51,  1.51it/s]
                                                         
{'loss': 1.6651, 'learning_rate': 5.878535569976735e-07, 'epoch': 1859.26}
+
 46%|████▌     | 30120/65536 [5:10:15<6:30:51,  1.51it/s]
 46%|████▌     | 30121/65536 [5:10:16<6:27:29,  1.52it/s]
 46%|████▌     | 30122/65536 [5:10:17<6:18:23,  1.56it/s]
 46%|████▌     | 30123/65536 [5:10:17<6:16:42,  1.57it/s]
 46%|████▌     | 30124/65536 [5:10:18<6:23:48,  1.54it/s]
 46%|████▌     | 30125/65536 [5:10:19<6:25:02,  1.53it/s]
 46%|████▌     | 30126/65536 [5:10:19<6:21:20,  1.55it/s]
 46%|████▌     | 30127/65536 [5:10:20<6:33:50,  1.50it/s]
 46%|████▌     | 30128/65536 [5:10:21<6:35:03,  1.49it/s]
 46%|████▌     | 30129/65536 [5:10:21<6:33:33,  1.50it/s]
 46%|████▌     | 30130/65536 [5:10:22<6:27:43,  1.52it/s]
 46%|████▌     | 30131/65536 [5:10:22<6:15:44,  1.57it/s]
 46%|████▌     | 30132/65536 [5:10:23<6:04:30,  1.62it/s]
 46%|████▌     | 30133/65536 [5:10:24<6:25:44,  1.53it/s]
 46%|████▌     | 30134/65536 [5:10:24<6:26:32,  1.53it/s]
 46%|████▌     | 30135/65536 [5:10:25<6:25:09,  1.53it/s]
 46%|████▌     | 30136/65536 [5:10:26<6:21:06,  1.55it/s]
 46%|████▌     | 30137/65536 [5:10:26<6:14:36,  1.57it/s]
 46%|████▌     | 30138/65536 [5:10:27<6:21:08,  1.55it/s]
 46%|████▌     | 30139/65536 [5:10:28<6:10:28,  1.59it/s]
 46%|████▌     | 30140/65536 [5:10:28<6:19:13,  1.56it/s]
                                                         
{'loss': 1.6653, 'learning_rate': 5.875780580384474e-07, 'epoch': 1860.49}
+
 46%|████▌     | 30140/65536 [5:10:28<6:19:13,  1.56it/s]
 46%|████▌     | 30141/65536 [5:10:29<6:26:52,  1.52it/s]
 46%|████▌     | 30142/65536 [5:10:30<6:18:38,  1.56it/s]
 46%|████▌     | 30143/65536 [5:10:30<6:24:51,  1.53it/s]
 46%|████▌     | 30144/65536 [5:10:31<6:16:24,  1.57it/s]
 46%|████▌     | 30145/65536 [5:10:31<6:09:39,  1.60it/s]
 46%|████▌     | 30146/65536 [5:10:32<6:11:29,  1.59it/s]
 46%|████▌     | 30147/65536 [5:10:33<6:15:47,  1.57it/s]
 46%|████▌     | 30148/65536 [5:10:33<6:27:16,  1.52it/s]
 46%|████▌     | 30149/65536 [5:10:34<6:34:14,  1.50it/s]
 46%|████▌     | 30150/65536 [5:10:35<6:24:24,  1.53it/s]
 46%|████▌     | 30151/65536 [5:10:35<6:29:28,  1.51it/s]
 46%|████▌     | 30152/65536 [5:10:36<6:34:16,  1.50it/s]
 46%|████▌     | 30153/65536 [5:10:37<6:35:59,  1.49it/s]
 46%|████▌     | 30154/65536 [5:10:37<6:27:01,  1.52it/s]
 46%|████▌     | 30155/65536 [5:10:38<6:24:01,  1.54it/s]
 46%|████▌     | 30156/65536 [5:10:39<6:16:40,  1.57it/s]
 46%|████▌     | 30157/65536 [5:10:39<6:17:24,  1.56it/s]
 46%|████▌     | 30158/65536 [5:10:40<6:17:12,  1.56it/s]
 46%|████▌     | 30159/65536 [5:10:40<6:10:52,  1.59it/s]
 46%|████▌     | 30160/65536 [5:10:41<6:08:02,  1.60it/s]
                                                         
{'loss': 1.6434, 'learning_rate': 5.873025590792212e-07, 'epoch': 1861.73}
+
 46%|████▌     | 30160/65536 [5:10:41<6:08:02,  1.60it/s]
 46%|████▌     | 30161/65536 [5:10:42<6:15:53,  1.57it/s]
 46%|████▌     | 30162/65536 [5:10:42<6:08:42,  1.60it/s]
 46%|████▌     | 30163/65536 [5:10:43<6:01:15,  1.63it/s]
 46%|████▌     | 30164/65536 [5:10:44<6:09:24,  1.60it/s]
 46%|████▌     | 30165/65536 [5:10:44<6:20:20,  1.55it/s]
 46%|████▌     | 30166/65536 [5:10:45<6:12:41,  1.58it/s]
 46%|████▌     | 30167/65536 [5:10:46<6:16:49,  1.56it/s]
 46%|████▌     | 30168/65536 [5:10:46<6:04:36,  1.62it/s]
 46%|████▌     | 30169/65536 [5:10:47<5:56:59,  1.65it/s]
 46%|████▌     | 30170/65536 [5:10:47<5:54:28,  1.66it/s]
 46%|████▌     | 30171/65536 [5:10:48<6:04:46,  1.62it/s]
 46%|████▌     | 30172/65536 [5:10:49<5:58:45,  1.64it/s]
 46%|████▌     | 30173/65536 [5:10:49<6:11:43,  1.59it/s]
 46%|████▌     | 30174/65536 [5:10:50<6:12:52,  1.58it/s]
 46%|████▌     | 30175/65536 [5:10:50<6:13:58,  1.58it/s]
 46%|████▌     | 30176/65536 [5:10:51<6:11:01,  1.59it/s]
 46%|████▌     | 30177/65536 [5:10:52<6:10:45,  1.59it/s]
 46%|████▌     | 30178/65536 [5:10:52<6:14:02,  1.58it/s]
 46%|████▌     | 30179/65536 [5:10:53<6:18:12,  1.56it/s]
 46%|████▌     | 30180/65536 [5:10:54<6:21:18,  1.55it/s]
                                                         
{'loss': 1.6873, 'learning_rate': 5.870270601199951e-07, 'epoch': 1862.96}
+
 46%|████▌     | 30180/65536 [5:10:54<6:21:18,  1.55it/s]
 46%|████▌     | 30181/65536 [5:10:54<6:27:52,  1.52it/s]
 46%|████▌     | 30182/65536 [5:10:55<6:24:39,  1.53it/s]
 46%|████▌     | 30183/65536 [5:10:56<6:25:33,  1.53it/s]
 46%|████▌     | 30184/65536 [5:10:56<6:19:50,  1.55it/s]
 46%|████▌     | 30185/65536 [5:10:57<6:28:11,  1.52it/s]
 46%|████▌     | 30186/65536 [5:10:58<6:28:43,  1.52it/s]
 46%|████▌     | 30187/65536 [5:10:58<6:25:22,  1.53it/s]
 46%|████▌     | 30188/65536 [5:10:59<6:19:56,  1.55it/s]
 46%|████▌     | 30189/65536 [5:11:00<6:31:57,  1.50it/s]
 46%|████▌     | 30190/65536 [5:11:00<6:26:11,  1.53it/s]
 46%|████▌     | 30191/65536 [5:11:01<6:13:28,  1.58it/s]
 46%|████▌     | 30192/65536 [5:11:01<6:07:10,  1.60it/s]
 46%|████▌     | 30193/65536 [5:11:02<6:07:34,  1.60it/s]
 46%|████▌     | 30194/65536 [5:11:03<6:12:52,  1.58it/s]
 46%|████▌     | 30195/65536 [5:11:03<6:06:12,  1.61it/s]
 46%|████▌     | 30196/65536 [5:11:04<6:05:16,  1.61it/s]
 46%|████▌     | 30197/65536 [5:11:05<6:21:24,  1.54it/s]
 46%|████▌     | 30198/65536 [5:11:05<6:13:20,  1.58it/s]
 46%|████▌     | 30199/65536 [5:11:06<6:14:21,  1.57it/s]
 46%|████▌     | 30200/65536 [5:11:07<6:13:10,  1.58it/s]
                                                         
{'loss': 1.725, 'learning_rate': 5.86751561160769e-07, 'epoch': 1864.2}
+
 46%|████▌     | 30200/65536 [5:11:07<6:13:10,  1.58it/s]
 46%|████▌     | 30201/65536 [5:11:07<6:10:53,  1.59it/s]
 46%|████▌     | 30202/65536 [5:11:08<6:07:03,  1.60it/s]
 46%|████▌     | 30203/65536 [5:11:08<6:10:58,  1.59it/s]
 46%|████▌     | 30204/65536 [5:11:09<6:25:45,  1.53it/s]
 46%|████▌     | 30205/65536 [5:11:10<6:23:01,  1.54it/s]
 46%|████▌     | 30206/65536 [5:11:10<6:18:22,  1.56it/s]
 46%|████▌     | 30207/65536 [5:11:11<6:19:02,  1.55it/s]
 46%|████▌     | 30208/65536 [5:11:12<6:24:58,  1.53it/s]
 46%|████▌     | 30209/65536 [5:11:12<6:17:43,  1.56it/s]
 46%|████▌     | 30210/65536 [5:11:13<6:15:09,  1.57it/s]
 46%|████▌     | 30211/65536 [5:11:14<6:14:13,  1.57it/s]
 46%|████▌     | 30212/65536 [5:11:14<6:18:36,  1.56it/s]
 46%|████▌     | 30213/65536 [5:11:15<6:12:37,  1.58it/s]
 46%|████▌     | 30214/65536 [5:11:16<6:24:45,  1.53it/s]
 46%|████▌     | 30215/65536 [5:11:16<6:21:49,  1.54it/s]
 46%|████▌     | 30216/65536 [5:11:17<6:09:33,  1.59it/s]
 46%|████▌     | 30217/65536 [5:11:17<6:20:49,  1.55it/s]
 46%|████▌     | 30218/65536 [5:11:18<6:13:43,  1.58it/s]
 46%|████▌     | 30219/65536 [5:11:19<6:06:02,  1.61it/s]
 46%|████▌     | 30220/65536 [5:11:19<5:58:59,  1.64it/s]
                                                         
{'loss': 1.6511, 'learning_rate': 5.864760622015427e-07, 'epoch': 1865.43}
+
 46%|████▌     | 30220/65536 [5:11:19<5:58:59,  1.64it/s]
 46%|████▌     | 30221/65536 [5:11:20<5:59:25,  1.64it/s]
 46%|████▌     | 30222/65536 [5:11:20<5:55:59,  1.65it/s]
 46%|████▌     | 30223/65536 [5:11:21<5:50:36,  1.68it/s]
 46%|████▌     | 30224/65536 [5:11:22<6:06:48,  1.60it/s]
 46%|████▌     | 30225/65536 [5:11:22<6:10:32,  1.59it/s]
 46%|████▌     | 30226/65536 [5:11:23<6:03:41,  1.62it/s]
 46%|████▌     | 30227/65536 [5:11:24<5:58:28,  1.64it/s]
 46%|████▌     | 30228/65536 [5:11:24<5:55:15,  1.66it/s]
 46%|████▌     | 30229/65536 [5:11:25<5:58:37,  1.64it/s]
 46%|████▌     | 30230/65536 [5:11:25<6:09:16,  1.59it/s]
 46%|████▌     | 30231/65536 [5:11:26<6:10:52,  1.59it/s]
 46%|████▌     | 30232/65536 [5:11:27<6:10:14,  1.59it/s]
 46%|████▌     | 30233/65536 [5:11:27<6:03:11,  1.62it/s]
 46%|████▌     | 30234/65536 [5:11:28<6:08:45,  1.60it/s]
 46%|████▌     | 30235/65536 [5:11:29<6:07:31,  1.60it/s]
 46%|████▌     | 30236/65536 [5:11:29<6:03:31,  1.62it/s]
 46%|████▌     | 30237/65536 [5:11:30<6:10:56,  1.59it/s]
 46%|████▌     | 30238/65536 [5:11:30<6:06:19,  1.61it/s]
 46%|████▌     | 30239/65536 [5:11:31<6:09:30,  1.59it/s]
 46%|████▌     | 30240/65536 [5:11:32<6:05:48,  1.61it/s]
                                                         
{'loss': 1.6844, 'learning_rate': 5.862005632423166e-07, 'epoch': 1866.67}
+
 46%|████▌     | 30240/65536 [5:11:32<6:05:48,  1.61it/s]
 46%|████▌     | 30241/65536 [5:11:32<6:01:56,  1.63it/s]
 46%|████▌     | 30242/65536 [5:11:33<5:55:27,  1.65it/s]
 46%|████▌     | 30243/65536 [5:11:33<6:02:10,  1.62it/s]
 46%|████▌     | 30244/65536 [5:11:34<6:01:30,  1.63it/s]
 46%|████▌     | 30245/65536 [5:11:35<6:02:58,  1.62it/s]
 46%|████▌     | 30246/65536 [5:11:35<6:14:06,  1.57it/s]
 46%|████▌     | 30247/65536 [5:11:36<6:14:32,  1.57it/s]
 46%|████▌     | 30248/65536 [5:11:37<6:21:50,  1.54it/s]
 46%|████▌     | 30249/65536 [5:11:37<6:09:18,  1.59it/s]
 46%|████▌     | 30250/65536 [5:11:38<5:57:34,  1.64it/s]
 46%|████▌     | 30251/65536 [5:11:38<5:51:18,  1.67it/s]
 46%|████▌     | 30252/65536 [5:11:39<5:52:29,  1.67it/s]
 46%|████▌     | 30253/65536 [5:11:40<5:53:53,  1.66it/s]
 46%|████▌     | 30254/65536 [5:11:40<5:53:15,  1.66it/s]
 46%|████▌     | 30255/65536 [5:11:41<6:04:08,  1.61it/s]
 46%|████▌     | 30256/65536 [5:11:41<6:00:05,  1.63it/s]
 46%|████▌     | 30257/65536 [5:11:42<6:10:47,  1.59it/s]
 46%|████▌     | 30258/65536 [5:11:43<6:03:12,  1.62it/s]
 46%|████▌     | 30259/65536 [5:11:43<6:01:26,  1.63it/s]
 46%|████▌     | 30260/65536 [5:11:44<5:56:40,  1.65it/s]
                                                         
{'loss': 1.6549, 'learning_rate': 5.859250642830904e-07, 'epoch': 1867.9}
+
 46%|████▌     | 30260/65536 [5:11:44<5:56:40,  1.65it/s]
 46%|████▌     | 30261/65536 [5:11:45<6:05:05,  1.61it/s]
 46%|████▌     | 30262/65536 [5:11:45<6:14:10,  1.57it/s]
 46%|████▌     | 30263/65536 [5:11:46<6:11:17,  1.58it/s]
 46%|████▌     | 30264/65536 [5:11:46<6:07:46,  1.60it/s]
 46%|████▌     | 30265/65536 [5:11:47<5:59:43,  1.63it/s]
 46%|████▌     | 30266/65536 [5:11:48<5:58:00,  1.64it/s]
 46%|████▌     | 30267/65536 [5:11:48<6:05:20,  1.61it/s]
 46%|████▌     | 30268/65536 [5:11:49<6:02:50,  1.62it/s]
 46%|████▌     | 30269/65536 [5:11:50<5:58:32,  1.64it/s]
 46%|████▌     | 30270/65536 [5:11:50<5:53:02,  1.66it/s]
 46%|████▌     | 30271/65536 [5:11:51<5:52:42,  1.67it/s]
 46%|████▌     | 30272/65536 [5:11:51<5:54:13,  1.66it/s]
 46%|████▌     | 30273/65536 [5:11:52<6:12:38,  1.58it/s]
 46%|████▌     | 30274/65536 [5:11:53<6:12:08,  1.58it/s]
 46%|████▌     | 30275/65536 [5:11:53<6:01:39,  1.62it/s]
 46%|████▌     | 30276/65536 [5:11:54<5:58:25,  1.64it/s]
 46%|████▌     | 30277/65536 [5:11:54<5:58:37,  1.64it/s]
 46%|████▌     | 30278/65536 [5:11:55<6:07:13,  1.60it/s]
 46%|████▌     | 30279/65536 [5:11:56<6:15:25,  1.57it/s]
 46%|████▌     | 30280/65536 [5:11:56<6:13:44,  1.57it/s]
                                                         
{'loss': 1.6959, 'learning_rate': 5.856495653238644e-07, 'epoch': 1869.14}
+
 46%|████▌     | 30280/65536 [5:11:56<6:13:44,  1.57it/s]
 46%|████▌     | 30281/65536 [5:11:57<6:05:03,  1.61it/s]
 46%|████▌     | 30282/65536 [5:11:58<6:14:18,  1.57it/s]
 46%|████▌     | 30283/65536 [5:11:58<6:19:53,  1.55it/s]
 46%|████▌     | 30284/65536 [5:11:59<6:22:08,  1.54it/s]
 46%|████▌     | 30285/65536 [5:12:00<6:14:58,  1.57it/s]
 46%|████▌     | 30286/65536 [5:12:00<6:09:23,  1.59it/s]
 46%|████▌     | 30287/65536 [5:12:01<6:05:26,  1.61it/s]
 46%|████▌     | 30288/65536 [5:12:01<6:05:11,  1.61it/s]
 46%|████▌     | 30289/65536 [5:12:02<6:00:41,  1.63it/s]
 46%|████▌     | 30290/65536 [5:12:03<6:00:06,  1.63it/s]
 46%|████▌     | 30291/65536 [5:12:03<5:54:26,  1.66it/s]
 46%|████▌     | 30292/65536 [5:12:04<5:53:25,  1.66it/s]
 46%|████▌     | 30293/65536 [5:12:04<5:55:23,  1.65it/s]
 46%|████▌     | 30294/65536 [5:12:05<5:54:16,  1.66it/s]
 46%|████▌     | 30295/65536 [5:12:06<6:21:04,  1.54it/s]
 46%|████▌     | 30296/65536 [5:12:06<6:09:46,  1.59it/s]
 46%|████▌     | 30297/65536 [5:12:07<6:07:43,  1.60it/s]
 46%|████▌     | 30298/65536 [5:12:08<6:10:44,  1.58it/s]
 46%|████▌     | 30299/65536 [5:12:08<6:02:15,  1.62it/s]
 46%|████▌     | 30300/65536 [5:12:09<6:01:11,  1.63it/s]
                                                         
{'loss': 1.6674, 'learning_rate': 5.85374066364638e-07, 'epoch': 1870.37}
+
 46%|████▌     | 30300/65536 [5:12:09<6:01:11,  1.63it/s]
 46%|████▌     | 30301/65536 [5:12:09<6:01:40,  1.62it/s]
 46%|████▌     | 30302/65536 [5:12:10<5:54:52,  1.65it/s]
 46%|████▌     | 30303/65536 [5:12:11<5:59:18,  1.63it/s]
 46%|████▌     | 30304/65536 [5:12:11<6:00:07,  1.63it/s]
 46%|████▌     | 30305/65536 [5:12:12<5:55:40,  1.65it/s]
 46%|████▌     | 30306/65536 [5:12:12<5:53:12,  1.66it/s]
 46%|████▌     | 30307/65536 [5:12:13<5:49:27,  1.68it/s]
 46%|████▌     | 30308/65536 [5:12:14<5:51:22,  1.67it/s]
 46%|████▌     | 30309/65536 [5:12:14<6:08:58,  1.59it/s]
 46%|████▌     | 30310/65536 [5:12:15<6:02:21,  1.62it/s]
 46%|████▋     | 30311/65536 [5:12:16<6:17:03,  1.56it/s]
 46%|████▋     | 30312/65536 [5:12:16<6:24:51,  1.53it/s]
 46%|████▋     | 30313/65536 [5:12:17<6:12:02,  1.58it/s]
 46%|████▋     | 30314/65536 [5:12:17<6:08:07,  1.59it/s]
 46%|████▋     | 30315/65536 [5:12:18<6:11:55,  1.58it/s]
 46%|████▋     | 30316/65536 [5:12:19<6:04:35,  1.61it/s]
 46%|████▋     | 30317/65536 [5:12:19<6:00:38,  1.63it/s]
 46%|████▋     | 30318/65536 [5:12:20<5:58:14,  1.64it/s]
 46%|████▋     | 30319/65536 [5:12:21<6:00:22,  1.63it/s]
 46%|████▋     | 30320/65536 [5:12:21<5:52:57,  1.66it/s]
                                                         
{'loss': 1.6911, 'learning_rate': 5.85098567405412e-07, 'epoch': 1871.6}
+
 46%|████▋     | 30320/65536 [5:12:21<5:52:57,  1.66it/s]
 46%|████▋     | 30321/65536 [5:12:22<6:01:04,  1.63it/s]
 46%|████▋     | 30322/65536 [5:12:22<6:00:49,  1.63it/s]
 46%|████▋     | 30323/65536 [5:12:23<6:01:22,  1.62it/s]
 46%|████▋     | 30324/65536 [5:12:24<6:00:46,  1.63it/s]
 46%|████▋     | 30325/65536 [5:12:24<5:54:09,  1.66it/s]
 46%|████▋     | 30326/65536 [5:12:25<5:56:24,  1.65it/s]
 46%|████▋     | 30327/65536 [5:12:26<6:11:15,  1.58it/s]
 46%|████▋     | 30328/65536 [5:12:26<6:09:03,  1.59it/s]
 46%|████▋     | 30329/65536 [5:12:27<6:16:43,  1.56it/s]
 46%|████▋     | 30330/65536 [5:12:27<6:11:57,  1.58it/s]
 46%|████▋     | 30331/65536 [5:12:28<6:16:47,  1.56it/s]
 46%|████▋     | 30332/65536 [5:12:29<6:18:33,  1.55it/s]
 46%|████▋     | 30333/65536 [5:12:29<6:20:00,  1.54it/s]
 46%|████▋     | 30334/65536 [5:12:30<6:13:00,  1.57it/s]
 46%|████▋     | 30335/65536 [5:12:31<6:03:50,  1.61it/s]
 46%|████▋     | 30336/65536 [5:12:31<6:08:44,  1.59it/s]
 46%|████▋     | 30337/65536 [5:12:32<6:04:13,  1.61it/s]
 46%|████▋     | 30338/65536 [5:12:32<5:56:12,  1.65it/s]
 46%|████▋     | 30339/65536 [5:12:33<5:58:38,  1.64it/s]
 46%|████▋     | 30340/65536 [5:12:34<5:59:38,  1.63it/s]
                                                         
{'loss': 1.6571, 'learning_rate': 5.848230684461859e-07, 'epoch': 1872.84}
+
 46%|████▋     | 30340/65536 [5:12:34<5:59:38,  1.63it/s]
 46%|████▋     | 30341/65536 [5:12:34<5:53:00,  1.66it/s]
 46%|████▋     | 30342/65536 [5:12:35<5:49:00,  1.68it/s]
 46%|████▋     | 30343/65536 [5:12:36<6:08:44,  1.59it/s]
 46%|████▋     | 30344/65536 [5:12:36<6:07:52,  1.59it/s]
 46%|████▋     | 30345/65536 [5:12:37<6:17:01,  1.56it/s]
 46%|████▋     | 30346/65536 [5:12:37<6:08:34,  1.59it/s]
 46%|████▋     | 30347/65536 [5:12:38<5:59:36,  1.63it/s]
 46%|████▋     | 30348/65536 [5:12:39<5:53:00,  1.66it/s]
 46%|████▋     | 30349/65536 [5:12:39<5:59:39,  1.63it/s]
 46%|████▋     | 30350/65536 [5:12:40<5:57:17,  1.64it/s]
 46%|████▋     | 30351/65536 [5:12:40<6:01:33,  1.62it/s]
 46%|████▋     | 30352/65536 [5:12:41<5:59:48,  1.63it/s]
 46%|████▋     | 30353/65536 [5:12:42<5:57:35,  1.64it/s]
 46%|████▋     | 30354/65536 [5:12:42<5:53:51,  1.66it/s]
 46%|████▋     | 30355/65536 [5:12:43<6:04:23,  1.61it/s]
 46%|████▋     | 30356/65536 [5:12:43<6:01:47,  1.62it/s]
 46%|████▋     | 30357/65536 [5:12:44<5:58:56,  1.63it/s]
 46%|████▋     | 30358/65536 [5:12:45<6:02:09,  1.62it/s]
 46%|████▋     | 30359/65536 [5:12:45<6:12:15,  1.57it/s]
 46%|████▋     | 30360/65536 [5:12:46<6:06:29,  1.60it/s]
                                                         
{'loss': 1.7023, 'learning_rate': 5.845475694869597e-07, 'epoch': 1874.07}
+
 46%|████▋     | 30360/65536 [5:12:46<6:06:29,  1.60it/s]
 46%|████▋     | 30361/65536 [5:12:47<6:02:18,  1.62it/s]
 46%|████▋     | 30362/65536 [5:12:47<5:57:09,  1.64it/s]
 46%|████▋     | 30363/65536 [5:12:48<5:56:25,  1.64it/s]
 46%|████▋     | 30364/65536 [5:12:48<6:01:12,  1.62it/s]
 46%|████▋     | 30365/65536 [5:12:49<5:58:09,  1.64it/s]
 46%|████▋     | 30366/65536 [5:12:50<5:54:52,  1.65it/s]
 46%|████▋     | 30367/65536 [5:12:50<6:15:26,  1.56it/s]
 46%|████▋     | 30368/65536 [5:12:51<6:11:54,  1.58it/s]
 46%|████▋     | 30369/65536 [5:12:52<6:06:34,  1.60it/s]
 46%|████▋     | 30370/65536 [5:12:52<6:06:52,  1.60it/s]
 46%|████▋     | 30371/65536 [5:12:53<6:05:50,  1.60it/s]
 46%|████▋     | 30372/65536 [5:12:53<5:57:01,  1.64it/s]
 46%|████▋     | 30373/65536 [5:12:54<6:01:04,  1.62it/s]
 46%|████▋     | 30374/65536 [5:12:55<5:59:04,  1.63it/s]
 46%|████▋     | 30375/65536 [5:12:55<6:00:53,  1.62it/s]
 46%|████▋     | 30376/65536 [5:12:56<6:18:53,  1.55it/s]
 46%|████▋     | 30377/65536 [5:12:57<6:08:01,  1.59it/s]
 46%|████▋     | 30378/65536 [5:12:57<5:56:22,  1.64it/s]
 46%|████▋     | 30379/65536 [5:12:58<5:55:34,  1.65it/s]
 46%|████▋     | 30380/65536 [5:12:58<6:05:55,  1.60it/s]
                                                         
{'loss': 1.6806, 'learning_rate': 5.842720705277336e-07, 'epoch': 1875.31}
+
 46%|████▋     | 30380/65536 [5:12:58<6:05:55,  1.60it/s]
 46%|████▋     | 30381/65536 [5:12:59<5:55:37,  1.65it/s]
 46%|████▋     | 30382/65536 [5:13:00<5:58:21,  1.63it/s]
 46%|████▋     | 30383/65536 [5:13:00<6:03:56,  1.61it/s]
 46%|████▋     | 30384/65536 [5:13:01<6:03:53,  1.61it/s]
 46%|████▋     | 30385/65536 [5:13:01<5:54:50,  1.65it/s]
 46%|████▋     | 30386/65536 [5:13:02<5:55:31,  1.65it/s]
 46%|████▋     | 30387/65536 [5:13:03<5:53:15,  1.66it/s]
 46%|████▋     | 30388/65536 [5:13:03<6:00:38,  1.62it/s]
 46%|████▋     | 30389/65536 [5:13:04<5:58:20,  1.63it/s]
 46%|████▋     | 30390/65536 [5:13:05<6:11:52,  1.58it/s]
 46%|████▋     | 30391/65536 [5:13:05<6:13:33,  1.57it/s]
 46%|████▋     | 30392/65536 [5:13:06<6:30:26,  1.50it/s]
 46%|████▋     | 30393/65536 [5:13:07<6:18:31,  1.55it/s]
 46%|████▋     | 30394/65536 [5:13:07<6:20:29,  1.54it/s]
 46%|████▋     | 30395/65536 [5:13:08<6:22:39,  1.53it/s]
 46%|████▋     | 30396/65536 [5:13:08<6:13:29,  1.57it/s]
 46%|████▋     | 30397/65536 [5:13:09<6:09:21,  1.59it/s]
 46%|████▋     | 30398/65536 [5:13:10<6:10:26,  1.58it/s]
 46%|████▋     | 30399/65536 [5:13:10<6:08:25,  1.59it/s]
 46%|████▋     | 30400/65536 [5:13:11<5:58:27,  1.63it/s]
                                                         
{'loss': 1.6539, 'learning_rate': 5.839965715685074e-07, 'epoch': 1876.54}
+
 46%|████▋     | 30400/65536 [5:13:11<5:58:27,  1.63it/s]
 46%|████▋     | 30401/65536 [5:13:12<5:59:57,  1.63it/s]
 46%|████▋     | 30402/65536 [5:13:12<5:59:42,  1.63it/s]
 46%|████▋     | 30403/65536 [5:13:13<6:00:25,  1.62it/s]
 46%|████▋     | 30404/65536 [5:13:13<5:52:38,  1.66it/s]
 46%|████▋     | 30405/65536 [5:13:14<5:58:19,  1.63it/s]
 46%|████▋     | 30406/65536 [5:13:15<5:53:57,  1.65it/s]
 46%|████▋     | 30407/65536 [5:13:15<5:47:02,  1.69it/s]
 46%|████▋     | 30408/65536 [5:13:16<6:03:01,  1.61it/s]
 46%|████▋     | 30409/65536 [5:13:16<5:55:21,  1.65it/s]
 46%|████▋     | 30410/65536 [5:13:17<5:59:06,  1.63it/s]
 46%|████▋     | 30411/65536 [5:13:18<6:01:48,  1.62it/s]
 46%|████▋     | 30412/65536 [5:13:18<5:46:48,  1.69it/s]
 46%|████▋     | 30413/65536 [5:13:19<5:47:54,  1.68it/s]
 46%|████▋     | 30414/65536 [5:13:19<5:50:46,  1.67it/s]
 46%|████▋     | 30415/65536 [5:13:20<6:01:27,  1.62it/s]
 46%|████▋     | 30416/65536 [5:13:21<5:58:15,  1.63it/s]
 46%|████▋     | 30417/65536 [5:13:21<5:51:26,  1.67it/s]
 46%|████▋     | 30418/65536 [5:13:22<5:50:29,  1.67it/s]
 46%|████▋     | 30419/65536 [5:13:22<5:54:26,  1.65it/s]
 46%|████▋     | 30420/65536 [5:13:23<6:07:22,  1.59it/s]
                                                         
{'loss': 1.7039, 'learning_rate': 5.837210726092813e-07, 'epoch': 1877.78}
+
 46%|████▋     | 30420/65536 [5:13:23<6:07:22,  1.59it/s]
 46%|████▋     | 30421/65536 [5:13:24<6:04:21,  1.61it/s]
 46%|████▋     | 30422/65536 [5:13:24<5:59:51,  1.63it/s]
 46%|████▋     | 30423/65536 [5:13:25<5:54:02,  1.65it/s]
 46%|████▋     | 30424/65536 [5:13:26<6:03:24,  1.61it/s]
 46%|████▋     | 30425/65536 [5:13:26<6:18:12,  1.55it/s]
 46%|████▋     | 30426/65536 [5:13:27<6:13:06,  1.57it/s]
 46%|████▋     | 30427/65536 [5:13:27<6:09:35,  1.58it/s]
 46%|████▋     | 30428/65536 [5:13:28<6:03:31,  1.61it/s]
 46%|████▋     | 30429/65536 [5:13:29<5:58:48,  1.63it/s]
 46%|████▋     | 30430/65536 [5:13:29<5:56:14,  1.64it/s]
 46%|████▋     | 30431/65536 [5:13:30<5:54:11,  1.65it/s]
 46%|████▋     | 30432/65536 [5:13:31<6:02:01,  1.62it/s]
 46%|████▋     | 30433/65536 [5:13:31<6:13:06,  1.57it/s]
 46%|████▋     | 30434/65536 [5:13:32<6:03:42,  1.61it/s]
 46%|████▋     | 30435/65536 [5:13:32<5:55:35,  1.65it/s]
 46%|████▋     | 30436/65536 [5:13:33<5:54:29,  1.65it/s]
 46%|████▋     | 30437/65536 [5:13:34<5:52:51,  1.66it/s]
 46%|████▋     | 30438/65536 [5:13:34<5:48:10,  1.68it/s]
 46%|████▋     | 30439/65536 [5:13:35<5:47:47,  1.68it/s]
 46%|████▋     | 30440/65536 [5:13:35<6:03:12,  1.61it/s]
                                                         
{'loss': 1.6855, 'learning_rate': 5.834455736500552e-07, 'epoch': 1879.01}
+
 46%|████▋     | 30440/65536 [5:13:35<6:03:12,  1.61it/s]
 46%|████▋     | 30441/65536 [5:13:36<6:03:23,  1.61it/s]
 46%|████▋     | 30442/65536 [5:13:37<5:58:03,  1.63it/s]
 46%|████▋     | 30443/65536 [5:13:37<6:02:53,  1.61it/s]
 46%|████▋     | 30444/65536 [5:13:38<6:08:03,  1.59it/s]
 46%|████▋     | 30445/65536 [5:13:39<6:05:30,  1.60it/s]
 46%|████▋     | 30446/65536 [5:13:39<6:11:18,  1.58it/s]
 46%|████▋     | 30447/65536 [5:13:40<6:09:30,  1.58it/s]
 46%|████▋     | 30448/65536 [5:13:40<6:06:17,  1.60it/s]
 46%|████▋     | 30449/65536 [5:13:41<6:03:01,  1.61it/s]
 46%|████▋     | 30450/65536 [5:13:42<6:02:25,  1.61it/s]
 46%|████▋     | 30451/65536 [5:13:42<6:08:44,  1.59it/s]
 46%|████▋     | 30452/65536 [5:13:43<5:58:51,  1.63it/s]
 46%|████▋     | 30453/65536 [5:13:43<5:58:07,  1.63it/s]
 46%|████▋     | 30454/65536 [5:13:44<6:06:53,  1.59it/s]
 46%|████▋     | 30455/65536 [5:13:45<5:57:01,  1.64it/s]
 46%|████▋     | 30456/65536 [5:13:45<5:58:19,  1.63it/s]
 46%|████▋     | 30457/65536 [5:13:46<6:14:38,  1.56it/s]
 46%|████▋     | 30458/65536 [5:13:47<6:10:11,  1.58it/s]
 46%|████▋     | 30459/65536 [5:13:47<6:13:03,  1.57it/s]
 46%|████▋     | 30460/65536 [5:13:48<6:09:07,  1.58it/s]
                                                         
{'loss': 1.6723, 'learning_rate': 5.83170074690829e-07, 'epoch': 1880.25}
+
 46%|████▋     | 30460/65536 [5:13:48<6:09:07,  1.58it/s]
 46%|████▋     | 30461/65536 [5:13:49<6:00:38,  1.62it/s]
 46%|████▋     | 30462/65536 [5:13:49<6:05:53,  1.60it/s]
 46%|████▋     | 30463/65536 [5:13:50<5:56:43,  1.64it/s]
 46%|████▋     | 30464/65536 [5:13:50<5:52:30,  1.66it/s]
 46%|████▋     | 30465/65536 [5:13:51<5:56:01,  1.64it/s]
 46%|████▋     | 30466/65536 [5:13:52<5:55:47,  1.64it/s]
 46%|████▋     | 30467/65536 [5:13:52<6:01:39,  1.62it/s]
 46%|████▋     | 30468/65536 [5:13:53<5:59:30,  1.63it/s]
 46%|████▋     | 30469/65536 [5:13:53<5:57:41,  1.63it/s]
 46%|████▋     | 30470/65536 [5:13:54<5:52:30,  1.66it/s]
 46%|████▋     | 30471/65536 [5:13:55<5:55:04,  1.65it/s]
 46%|████▋     | 30472/65536 [5:13:55<6:02:39,  1.61it/s]
 46%|████▋     | 30473/65536 [5:13:56<6:15:21,  1.56it/s]
 46%|████▋     | 30474/65536 [5:13:57<6:11:14,  1.57it/s]
 47%|████▋     | 30475/65536 [5:13:57<6:16:04,  1.55it/s]
 47%|████▋     | 30476/65536 [5:13:58<6:15:23,  1.56it/s]
 47%|████▋     | 30477/65536 [5:13:58<6:09:35,  1.58it/s]
 47%|████▋     | 30478/65536 [5:13:59<6:06:02,  1.60it/s]
 47%|████▋     | 30479/65536 [5:14:00<6:08:43,  1.58it/s]
 47%|████▋     | 30480/65536 [5:14:00<6:03:41,  1.61it/s]
                                                         
{'loss': 1.6378, 'learning_rate': 5.828945757316029e-07, 'epoch': 1881.48}
+
 47%|████▋     | 30480/65536 [5:14:00<6:03:41,  1.61it/s]
 47%|████▋     | 30481/65536 [5:14:01<5:57:51,  1.63it/s]
 47%|████▋     | 30482/65536 [5:14:02<6:09:43,  1.58it/s]
 47%|████▋     | 30483/65536 [5:14:02<6:00:54,  1.62it/s]
 47%|████▋     | 30484/65536 [5:14:03<5:50:28,  1.67it/s]
 47%|████▋     | 30485/65536 [5:14:03<5:53:51,  1.65it/s]
 47%|████▋     | 30486/65536 [5:14:04<5:51:13,  1.66it/s]
 47%|████▋     | 30487/65536 [5:14:05<5:51:40,  1.66it/s]
 47%|████▋     | 30488/65536 [5:14:05<5:54:00,  1.65it/s]
 47%|████▋     | 30489/65536 [5:14:06<6:05:47,  1.60it/s]
 47%|████▋     | 30490/65536 [5:14:06<5:58:47,  1.63it/s]
 47%|████▋     | 30491/65536 [5:14:07<5:50:33,  1.67it/s]
 47%|████▋     | 30492/65536 [5:14:08<5:54:47,  1.65it/s]
 47%|████▋     | 30493/65536 [5:14:08<6:13:59,  1.56it/s]
 47%|████▋     | 30494/65536 [5:14:09<6:18:27,  1.54it/s]
 47%|████▋     | 30495/65536 [5:14:10<6:21:59,  1.53it/s]
 47%|████▋     | 30496/65536 [5:14:10<6:21:01,  1.53it/s]
 47%|████▋     | 30497/65536 [5:14:11<6:13:52,  1.56it/s]
 47%|████▋     | 30498/65536 [5:14:12<6:09:10,  1.58it/s]
 47%|████▋     | 30499/65536 [5:14:12<6:11:13,  1.57it/s]
 47%|████▋     | 30500/65536 [5:14:13<6:06:57,  1.59it/s]
                                                         
{'loss': 1.6862, 'learning_rate': 5.826190767723766e-07, 'epoch': 1882.72}
+
 47%|████▋     | 30500/65536 [5:14:13<6:06:57,  1.59it/s]
 47%|████▋     | 30501/65536 [5:14:13<5:55:25,  1.64it/s]
 47%|████▋     | 30502/65536 [5:14:14<5:49:19,  1.67it/s]
 47%|████▋     | 30503/65536 [5:14:15<5:52:48,  1.65it/s]
 47%|████▋     | 30504/65536 [5:14:15<5:55:28,  1.64it/s]
 47%|████▋     | 30505/65536 [5:14:16<6:11:24,  1.57it/s]
 47%|████▋     | 30506/65536 [5:14:16<6:07:44,  1.59it/s]
 47%|████▋     | 30507/65536 [5:14:17<6:02:39,  1.61it/s]
 47%|████▋     | 30508/65536 [5:14:18<6:02:12,  1.61it/s]
 47%|████▋     | 30509/65536 [5:14:18<6:07:24,  1.59it/s]
 47%|████▋     | 30510/65536 [5:14:19<6:03:42,  1.61it/s]
 47%|████▋     | 30511/65536 [5:14:20<6:08:15,  1.59it/s]
 47%|████▋     | 30512/65536 [5:14:20<5:55:38,  1.64it/s]
 47%|████▋     | 30513/65536 [5:14:21<5:57:35,  1.63it/s]
 47%|████▋     | 30514/65536 [5:14:21<5:49:55,  1.67it/s]
 47%|████▋     | 30515/65536 [5:14:22<5:49:05,  1.67it/s]
 47%|████▋     | 30516/65536 [5:14:23<5:47:34,  1.68it/s]
 47%|████▋     | 30517/65536 [5:14:23<5:39:12,  1.72it/s]
 47%|████▋     | 30518/65536 [5:14:24<5:49:02,  1.67it/s]
 47%|████▋     | 30519/65536 [5:14:24<5:50:23,  1.67it/s]
 47%|████▋     | 30520/65536 [5:14:25<5:46:11,  1.69it/s]
                                                         
{'loss': 1.7127, 'learning_rate': 5.823435778131505e-07, 'epoch': 1883.95}
+
 47%|████▋     | 30520/65536 [5:14:25<5:46:11,  1.69it/s]
 47%|████▋     | 30521/65536 [5:14:26<6:12:31,  1.57it/s]
 47%|████▋     | 30522/65536 [5:14:26<6:03:02,  1.61it/s]
 47%|████▋     | 30523/65536 [5:14:27<5:56:32,  1.64it/s]
 47%|████▋     | 30524/65536 [5:14:27<5:58:50,  1.63it/s]
 47%|████▋     | 30525/65536 [5:14:28<6:05:46,  1.60it/s]
 47%|████▋     | 30526/65536 [5:14:29<6:00:32,  1.62it/s]
 47%|████▋     | 30527/65536 [5:14:29<5:58:13,  1.63it/s]
 47%|████▋     | 30528/65536 [5:14:30<5:59:42,  1.62it/s]
 47%|████▋     | 30529/65536 [5:14:31<5:55:27,  1.64it/s]
 47%|████▋     | 30530/65536 [5:14:31<5:50:24,  1.66it/s]
 47%|████▋     | 30531/65536 [5:14:32<6:03:10,  1.61it/s]
 47%|████▋     | 30532/65536 [5:14:32<5:58:07,  1.63it/s]
 47%|████▋     | 30533/65536 [5:14:33<5:46:47,  1.68it/s]
 47%|████▋     | 30534/65536 [5:14:34<5:43:53,  1.70it/s]
 47%|████▋     | 30535/65536 [5:14:34<5:43:39,  1.70it/s]
 47%|████▋     | 30536/65536 [5:14:35<5:51:10,  1.66it/s]
 47%|████▋     | 30537/65536 [5:14:35<5:48:09,  1.68it/s]
 47%|████▋     | 30538/65536 [5:14:36<6:04:01,  1.60it/s]
 47%|████▋     | 30539/65536 [5:14:37<6:00:58,  1.62it/s]
 47%|████▋     | 30540/65536 [5:14:37<6:05:52,  1.59it/s]
                                                         
{'loss': 1.6837, 'learning_rate': 5.820680788539243e-07, 'epoch': 1885.19}
+
 47%|████▋     | 30540/65536 [5:14:37<6:05:52,  1.59it/s]
 47%|████▋     | 30541/65536 [5:14:38<5:59:40,  1.62it/s]
 47%|████▋     | 30542/65536 [5:14:38<6:01:12,  1.61it/s]
 47%|████▋     | 30543/65536 [5:14:39<5:57:01,  1.63it/s]
 47%|████▋     | 30544/65536 [5:14:40<5:55:00,  1.64it/s]
 47%|████▋     | 30545/65536 [5:14:40<5:52:28,  1.65it/s]
 47%|████▋     | 30546/65536 [5:14:41<5:52:05,  1.66it/s]
 47%|████▋     | 30547/65536 [5:14:41<5:57:44,  1.63it/s]
 47%|████▋     | 30548/65536 [5:14:42<5:57:41,  1.63it/s]
 47%|████▋     | 30549/65536 [5:14:43<5:57:33,  1.63it/s]
 47%|████▋     | 30550/65536 [5:14:43<6:07:36,  1.59it/s]
 47%|████▋     | 30551/65536 [5:14:44<6:08:03,  1.58it/s]
 47%|████▋     | 30552/65536 [5:14:45<6:09:18,  1.58it/s]
 47%|████▋     | 30553/65536 [5:14:45<6:07:00,  1.59it/s]
 47%|████▋     | 30554/65536 [5:14:46<6:15:34,  1.55it/s]
 47%|████▋     | 30555/65536 [5:14:47<6:11:53,  1.57it/s]
 47%|████▋     | 30556/65536 [5:14:47<6:13:30,  1.56it/s]
 47%|████▋     | 30557/65536 [5:14:48<6:03:47,  1.60it/s]
 47%|████▋     | 30558/65536 [5:14:48<6:09:48,  1.58it/s]
 47%|████▋     | 30559/65536 [5:14:49<6:03:52,  1.60it/s]
 47%|████▋     | 30560/65536 [5:14:50<6:01:19,  1.61it/s]
                                                         
{'loss': 1.6474, 'learning_rate': 5.817925798946981e-07, 'epoch': 1886.42}
+
 47%|████▋     | 30560/65536 [5:14:50<6:01:19,  1.61it/s]
 47%|████▋     | 30561/65536 [5:14:50<6:02:55,  1.61it/s]
 47%|████▋     | 30562/65536 [5:14:51<5:56:20,  1.64it/s]
 47%|████▋     | 30563/65536 [5:14:51<5:51:30,  1.66it/s]
 47%|████▋     | 30564/65536 [5:14:52<5:56:34,  1.63it/s]
 47%|████▋     | 30565/65536 [5:14:53<6:05:51,  1.59it/s]
 47%|████▋     | 30566/65536 [5:14:53<6:03:08,  1.60it/s]
 47%|████▋     | 30567/65536 [5:14:54<5:58:46,  1.62it/s]
 47%|████▋     | 30568/65536 [5:14:55<5:53:47,  1.65it/s]
 47%|████▋     | 30569/65536 [5:14:55<5:44:13,  1.69it/s]
 47%|████▋     | 30570/65536 [5:14:56<5:59:39,  1.62it/s]
 47%|████▋     | 30571/65536 [5:14:56<5:51:17,  1.66it/s]
 47%|████▋     | 30572/65536 [5:14:57<6:00:56,  1.61it/s]
 47%|████▋     | 30573/65536 [5:14:58<6:05:01,  1.60it/s]
 47%|████▋     | 30574/65536 [5:14:58<6:05:04,  1.60it/s]
 47%|████▋     | 30575/65536 [5:14:59<5:54:49,  1.64it/s]
 47%|████▋     | 30576/65536 [5:15:00<5:56:52,  1.63it/s]
 47%|████▋     | 30577/65536 [5:15:00<5:57:42,  1.63it/s]
 47%|████▋     | 30578/65536 [5:15:01<6:14:19,  1.56it/s]
 47%|████▋     | 30579/65536 [5:15:01<6:10:47,  1.57it/s]
 47%|████▋     | 30580/65536 [5:15:02<6:06:13,  1.59it/s]
                                                         
{'loss': 1.6561, 'learning_rate': 5.81517080935472e-07, 'epoch': 1887.65}
+
 47%|████▋     | 30580/65536 [5:15:02<6:06:13,  1.59it/s]
 47%|████▋     | 30581/65536 [5:15:03<6:05:59,  1.59it/s]
 47%|████▋     | 30582/65536 [5:15:03<6:00:01,  1.62it/s]
 47%|████▋     | 30583/65536 [5:15:04<6:00:28,  1.62it/s]
 47%|████▋     | 30584/65536 [5:15:05<6:02:44,  1.61it/s]
 47%|████▋     | 30585/65536 [5:15:05<5:51:32,  1.66it/s]
 47%|��███▋     | 30586/65536 [5:15:06<6:04:05,  1.60it/s]
 47%|████▋     | 30587/65536 [5:15:06<6:01:52,  1.61it/s]
 47%|████▋     | 30588/65536 [5:15:07<6:01:35,  1.61it/s]
 47%|████▋     | 30589/65536 [5:15:08<6:03:26,  1.60it/s]
 47%|████▋     | 30590/65536 [5:15:08<5:59:05,  1.62it/s]
 47%|████▋     | 30591/65536 [5:15:09<5:59:58,  1.62it/s]
 47%|████▋     | 30592/65536 [5:15:10<6:19:17,  1.54it/s]
 47%|████▋     | 30593/65536 [5:15:10<6:17:34,  1.54it/s]
 47%|████▋     | 30594/65536 [5:15:11<6:07:19,  1.59it/s]
 47%|████▋     | 30595/65536 [5:15:11<6:11:10,  1.57it/s]
 47%|████▋     | 30596/65536 [5:15:12<5:54:06,  1.64it/s]
 47%|████▋     | 30597/65536 [5:15:13<5:46:23,  1.68it/s]
 47%|████▋     | 30598/65536 [5:15:13<5:46:53,  1.68it/s]
 47%|████▋     | 30599/65536 [5:15:14<5:55:16,  1.64it/s]
 47%|████▋     | 30600/65536 [5:15:14<5:58:14,  1.63it/s]
                                                         
{'loss': 1.6545, 'learning_rate': 5.812415819762458e-07, 'epoch': 1888.89}
+
 47%|████▋     | 30600/65536 [5:15:14<5:58:14,  1.63it/s]
 47%|████▋     | 30601/65536 [5:15:15<5:54:40,  1.64it/s]
 47%|████▋     | 30602/65536 [5:15:16<6:07:14,  1.59it/s]
 47%|████▋     | 30603/65536 [5:15:16<5:55:24,  1.64it/s]
 47%|████▋     | 30604/65536 [5:15:17<6:12:37,  1.56it/s]
 47%|████▋     | 30605/65536 [5:15:18<6:07:21,  1.58it/s]
 47%|████▋     | 30606/65536 [5:15:18<6:03:46,  1.60it/s]
 47%|████▋     | 30607/65536 [5:15:19<6:15:07,  1.55it/s]
 47%|████▋     | 30608/65536 [5:15:19<6:03:30,  1.60it/s]
 47%|████▋     | 30609/65536 [5:15:20<6:07:31,  1.58it/s]
 47%|████▋     | 30610/65536 [5:15:21<5:58:53,  1.62it/s]
 47%|████▋     | 30611/65536 [5:15:21<5:55:49,  1.64it/s]
 47%|████▋     | 30612/65536 [5:15:22<5:49:38,  1.66it/s]
 47%|████▋     | 30613/65536 [5:15:22<5:51:21,  1.66it/s]
 47%|████▋     | 30614/65536 [5:15:23<6:00:36,  1.61it/s]
 47%|████▋     | 30615/65536 [5:15:24<5:56:32,  1.63it/s]
 47%|████▋     | 30616/65536 [5:15:24<5:56:01,  1.63it/s]
 47%|████▋     | 30617/65536 [5:15:25<5:51:15,  1.66it/s]
 47%|████▋     | 30618/65536 [5:15:26<5:46:53,  1.68it/s]
 47%|████▋     | 30619/65536 [5:15:26<6:05:27,  1.59it/s]
 47%|████▋     | 30620/65536 [5:15:27<5:59:00,  1.62it/s]
                                                         
{'loss': 1.6694, 'learning_rate': 5.809660830170197e-07, 'epoch': 1890.12}
+
 47%|████▋     | 30620/65536 [5:15:27<5:59:00,  1.62it/s]
 47%|████▋     | 30621/65536 [5:15:27<5:57:13,  1.63it/s]
 47%|████▋     | 30622/65536 [5:15:28<5:55:04,  1.64it/s]
 47%|████▋     | 30623/65536 [5:15:29<5:55:52,  1.64it/s]
 47%|████▋     | 30624/65536 [5:15:29<5:56:28,  1.63it/s]
 47%|████▋     | 30625/65536 [5:15:30<5:48:46,  1.67it/s]
 47%|████▋     | 30626/65536 [5:15:31<6:11:30,  1.57it/s]
 47%|████▋     | 30627/65536 [5:15:31<6:00:40,  1.61it/s]
 47%|████▋     | 30628/65536 [5:15:32<6:02:57,  1.60it/s]
 47%|████▋     | 30629/65536 [5:15:32<5:58:35,  1.62it/s]
 47%|████▋     | 30630/65536 [5:15:33<5:55:19,  1.64it/s]
 47%|████▋     | 30631/65536 [5:15:34<5:54:05,  1.64it/s]
 47%|████▋     | 30632/65536 [5:15:34<5:55:35,  1.64it/s]
 47%|████▋     | 30633/65536 [5:15:35<5:55:33,  1.64it/s]
 47%|████▋     | 30634/65536 [5:15:35<5:59:56,  1.62it/s]
 47%|████▋     | 30635/65536 [5:15:36<6:14:56,  1.55it/s]
 47%|████▋     | 30636/65536 [5:15:37<6:07:34,  1.58it/s]
 47%|████▋     | 30637/65536 [5:15:37<6:01:35,  1.61it/s]
 47%|████▋     | 30638/65536 [5:15:38<6:07:00,  1.58it/s]
 47%|████▋     | 30639/65536 [5:15:39<6:02:22,  1.61it/s]
 47%|████▋     | 30640/65536 [5:15:39<6:00:59,  1.61it/s]
                                                         
{'loss': 1.6651, 'learning_rate': 5.806905840577934e-07, 'epoch': 1891.36}
+
 47%|████▋     | 30640/65536 [5:15:39<6:00:59,  1.61it/s]
 47%|████▋     | 30641/65536 [5:15:40<5:52:27,  1.65it/s]
 47%|████▋     | 30642/65536 [5:15:40<5:57:15,  1.63it/s]
 47%|████▋     | 30643/65536 [5:15:41<5:57:35,  1.63it/s]
 47%|████▋     | 30644/65536 [5:15:42<6:06:56,  1.58it/s]
 47%|████▋     | 30645/65536 [5:15:42<6:07:51,  1.58it/s]
 47%|████▋     | 30646/65536 [5:15:43<5:57:06,  1.63it/s]
 47%|████▋     | 30647/65536 [5:15:43<5:50:47,  1.66it/s]
 47%|████▋     | 30648/65536 [5:15:44<5:54:56,  1.64it/s]
 47%|████▋     | 30649/65536 [5:15:45<5:48:33,  1.67it/s]
 47%|████▋     | 30650/65536 [5:15:45<5:57:04,  1.63it/s]
 47%|████▋     | 30651/65536 [5:15:46<6:15:14,  1.55it/s]
 47%|████▋     | 30652/65536 [5:15:47<6:06:00,  1.59it/s]
 47%|████▋     | 30653/65536 [5:15:47<6:01:41,  1.61it/s]
 47%|████▋     | 30654/65536 [5:15:48<6:04:28,  1.60it/s]
 47%|████▋     | 30655/65536 [5:15:48<5:52:38,  1.65it/s]
 47%|████▋     | 30656/65536 [5:15:49<6:06:22,  1.59it/s]
 47%|████▋     | 30657/65536 [5:15:50<6:06:32,  1.59it/s]
 47%|████▋     | 30658/65536 [5:15:50<6:11:21,  1.57it/s]
 47%|████▋     | 30659/65536 [5:15:51<6:06:36,  1.59it/s]
 47%|████▋     | 30660/65536 [5:15:52<6:06:12,  1.59it/s]
                                                         
{'loss': 1.6752, 'learning_rate': 5.804150850985674e-07, 'epoch': 1892.59}
+
 47%|████▋     | 30660/65536 [5:15:52<6:06:12,  1.59it/s]
 47%|████▋     | 30661/65536 [5:15:52<5:56:05,  1.63it/s]
 47%|████▋     | 30662/65536 [5:15:53<5:52:17,  1.65it/s]
 47%|████▋     | 30663/65536 [5:15:53<5:47:31,  1.67it/s]
 47%|████▋     | 30664/65536 [5:15:54<5:50:31,  1.66it/s]
 47%|████▋     | 30665/65536 [5:15:55<5:50:34,  1.66it/s]
 47%|████▋     | 30666/65536 [5:15:55<5:48:10,  1.67it/s]
 47%|████▋     | 30667/65536 [5:15:56<6:08:27,  1.58it/s]
 47%|████▋     | 30668/65536 [5:15:57<6:02:31,  1.60it/s]
 47%|████▋     | 30669/65536 [5:15:57<5:54:53,  1.64it/s]
 47%|████▋     | 30670/65536 [5:15:58<5:54:06,  1.64it/s]
 47%|████▋     | 30671/65536 [5:15:58<6:00:27,  1.61it/s]
 47%|████▋     | 30672/65536 [5:15:59<5:58:54,  1.62it/s]
 47%|████▋     | 30673/65536 [5:16:00<5:54:36,  1.64it/s]
 47%|████▋     | 30674/65536 [5:16:00<5:53:43,  1.64it/s]
 47%|████▋     | 30675/65536 [5:16:01<5:53:32,  1.64it/s]
 47%|████▋     | 30676/65536 [5:16:01<5:46:25,  1.68it/s]
 47%|████▋     | 30677/65536 [5:16:02<5:52:08,  1.65it/s]
 47%|████▋     | 30678/65536 [5:16:03<5:57:01,  1.63it/s]
 47%|████▋     | 30679/65536 [5:16:03<5:55:45,  1.63it/s]
 47%|████▋     | 30680/65536 [5:16:04<5:56:41,  1.63it/s]
                                                         
{'loss': 1.674, 'learning_rate': 5.801395861393412e-07, 'epoch': 1893.83}
+
 47%|████▋     | 30680/65536 [5:16:04<5:56:41,  1.63it/s]
 47%|████▋     | 30681/65536 [5:16:05<6:09:59,  1.57it/s]
 47%|████▋     | 30682/65536 [5:16:05<6:05:54,  1.59it/s]
 47%|████▋     | 30683/65536 [5:16:06<6:12:50,  1.56it/s]
 47%|████▋     | 30684/65536 [5:16:06<6:10:42,  1.57it/s]
 47%|████▋     | 30685/65536 [5:16:07<6:09:07,  1.57it/s]
 47%|████▋     | 30686/65536 [5:16:08<5:57:09,  1.63it/s]
 47%|████▋     | 30687/65536 [5:16:08<6:04:05,  1.60it/s]
 47%|████▋     | 30688/65536 [5:16:09<5:59:03,  1.62it/s]
 47%|████▋     | 30689/65536 [5:16:09<5:53:08,  1.64it/s]
 47%|████▋     | 30690/65536 [5:16:10<5:56:31,  1.63it/s]
 47%|████▋     | 30691/65536 [5:16:11<5:50:59,  1.65it/s]
 47%|████▋     | 30692/65536 [5:16:11<6:10:08,  1.57it/s]
 47%|████▋     | 30693/65536 [5:16:12<6:11:25,  1.56it/s]
 47%|████▋     | 30694/65536 [5:16:13<6:08:35,  1.58it/s]
 47%|████▋     | 30695/65536 [5:16:13<6:14:33,  1.55it/s]
 47%|████▋     | 30696/65536 [5:16:14<5:57:41,  1.62it/s]
 47%|████▋     | 30697/65536 [5:16:14<5:53:37,  1.64it/s]
 47%|████▋     | 30698/65536 [5:16:15<5:44:05,  1.69it/s]
 47%|████▋     | 30699/65536 [5:16:16<5:52:28,  1.65it/s]
 47%|████▋     | 30700/65536 [5:16:16<6:02:06,  1.60it/s]
                                                         
{'loss': 1.6675, 'learning_rate': 5.798640871801151e-07, 'epoch': 1895.06}
+
 47%|████▋     | 30700/65536 [5:16:16<6:02:06,  1.60it/s]
 47%|████▋     | 30701/65536 [5:16:17<5:58:29,  1.62it/s]
 47%|████▋     | 30702/65536 [5:16:18<6:09:38,  1.57it/s]
 47%|████▋     | 30703/65536 [5:16:18<6:02:15,  1.60it/s]
 47%|████▋     | 30704/65536 [5:16:19<5:51:17,  1.65it/s]
 47%|████▋     | 30705/65536 [5:16:19<5:49:32,  1.66it/s]
 47%|████▋     | 30706/65536 [5:16:20<5:56:36,  1.63it/s]
 47%|████▋     | 30707/65536 [5:16:21<5:56:29,  1.63it/s]
 47%|████▋     | 30708/65536 [5:16:21<5:51:09,  1.65it/s]
 47%|████▋     | 30709/65536 [5:16:22<5:47:11,  1.67it/s]
 47%|████▋     | 30710/65536 [5:16:22<5:56:45,  1.63it/s]
 47%|████▋     | 30711/65536 [5:16:23<5:51:28,  1.65it/s]
 47%|████▋     | 30712/65536 [5:16:24<5:50:12,  1.66it/s]
 47%|████▋     | 30713/65536 [5:16:24<5:52:53,  1.64it/s]
 47%|████▋     | 30714/65536 [5:16:25<5:52:24,  1.65it/s]
 47%|████▋     | 30715/65536 [5:16:25<6:00:49,  1.61it/s]
 47%|████▋     | 30716/65536 [5:16:26<6:20:21,  1.53it/s]
 47%|████▋     | 30717/65536 [5:16:27<6:11:55,  1.56it/s]
 47%|████▋     | 30718/65536 [5:16:27<5:59:00,  1.62it/s]
 47%|████▋     | 30719/65536 [5:16:28<5:53:40,  1.64it/s]
 47%|████▋     | 30720/65536 [5:16:29<5:53:17,  1.64it/s]
                                                         
{'loss': 1.698, 'learning_rate': 5.79588588220889e-07, 'epoch': 1896.3}
+
 47%|████▋     | 30720/65536 [5:16:29<5:53:17,  1.64it/s]
 47%|████▋     | 30721/65536 [5:16:29<5:58:19,  1.62it/s]
 47%|████▋     | 30722/65536 [5:16:30<5:54:54,  1.63it/s]
 47%|████▋     | 30723/65536 [5:16:30<5:53:27,  1.64it/s]
 47%|████▋     | 30724/65536 [5:16:31<5:57:43,  1.62it/s]
 47%|████▋     | 30725/65536 [5:16:32<5:59:54,  1.61it/s]
 47%|████▋     | 30726/65536 [5:16:32<6:05:58,  1.59it/s]
 47%|████▋     | 30727/65536 [5:16:33<5:57:31,  1.62it/s]
 47%|████▋     | 30728/65536 [5:16:33<5:48:00,  1.67it/s]
 47%|████▋     | 30729/65536 [5:16:34<5:47:50,  1.67it/s]
 47%|████▋     | 30730/65536 [5:16:35<5:57:44,  1.62it/s]
 47%|████▋     | 30731/65536 [5:16:35<6:07:21,  1.58it/s]
 47%|████▋     | 30732/65536 [5:16:36<6:14:14,  1.55it/s]
 47%|████▋     | 30733/65536 [5:16:37<6:06:05,  1.58it/s]
 47%|████▋     | 30734/65536 [5:16:37<6:08:17,  1.57it/s]
 47%|████▋     | 30735/65536 [5:16:38<5:57:12,  1.62it/s]
 47%|████▋     | 30736/65536 [5:16:39<5:56:50,  1.63it/s]
 47%|████▋     | 30737/65536 [5:16:39<5:51:01,  1.65it/s]
 47%|████▋     | 30738/65536 [5:16:40<6:02:43,  1.60it/s]
 47%|████▋     | 30739/65536 [5:16:40<5:56:49,  1.63it/s]
 47%|████▋     | 30740/65536 [5:16:41<5:55:17,  1.63it/s]
                                                         
{'loss': 1.6722, 'learning_rate': 5.793130892616628e-07, 'epoch': 1897.53}
+
 47%|████▋     | 30740/65536 [5:16:41<5:55:17,  1.63it/s]
 47%|████▋     | 30741/65536 [5:16:42<5:56:43,  1.63it/s]
 47%|████▋     | 30742/65536 [5:16:42<5:54:42,  1.63it/s]
 47%|████▋     | 30743/65536 [5:16:43<5:58:19,  1.62it/s]
 47%|████▋     | 30744/65536 [5:16:43<6:06:18,  1.58it/s]
 47%|████▋     | 30745/65536 [5:16:44<6:00:48,  1.61it/s]
 47%|████▋     | 30746/65536 [5:16:45<5:59:30,  1.61it/s]
 47%|████▋     | 30747/65536 [5:16:45<6:00:48,  1.61it/s]
 47%|████▋     | 30748/65536 [5:16:46<6:10:09,  1.57it/s]
 47%|████▋     | 30749/65536 [5:16:47<5:58:47,  1.62it/s]
 47%|████▋     | 30750/65536 [5:16:47<6:05:53,  1.58it/s]
 47%|████▋     | 30751/65536 [5:16:48<6:06:24,  1.58it/s]
 47%|████▋     | 30752/65536 [5:16:48<6:03:52,  1.59it/s]
 47%|████▋     | 30753/65536 [5:16:49<5:59:55,  1.61it/s]
 47%|████▋     | 30754/65536 [5:16:50<5:52:44,  1.64it/s]
 47%|████▋     | 30755/65536 [5:16:50<5:50:17,  1.65it/s]
 47%|████▋     | 30756/65536 [5:16:51<5:50:56,  1.65it/s]
 47%|████▋     | 30757/65536 [5:16:52<5:55:26,  1.63it/s]
 47%|████▋     | 30758/65536 [5:16:52<5:53:28,  1.64it/s]
 47%|████▋     | 30759/65536 [5:16:53<5:55:37,  1.63it/s]
 47%|████▋     | 30760/65536 [5:16:53<5:52:33,  1.64it/s]
                                                         
{'loss': 1.6899, 'learning_rate': 5.790375903024366e-07, 'epoch': 1898.77}
+
 47%|████▋     | 30760/65536 [5:16:53<5:52:33,  1.64it/s]
 47%|████▋     | 30761/65536 [5:16:54<5:56:21,  1.63it/s]
 47%|████▋     | 30762/65536 [5:16:55<5:50:07,  1.66it/s]
 47%|████▋     | 30763/65536 [5:16:55<6:00:23,  1.61it/s]
 47%|████▋     | 30764/65536 [5:16:56<6:10:13,  1.57it/s]
 47%|████▋     | 30765/65536 [5:16:57<6:12:09,  1.56it/s]
 47%|████▋     | 30766/65536 [5:16:57<6:04:34,  1.59it/s]
 47%|████▋     | 30767/65536 [5:16:58<5:56:43,  1.62it/s]
 47%|████▋     | 30768/65536 [5:16:58<5:48:11,  1.66it/s]
 47%|████▋     | 30769/65536 [5:16:59<5:48:29,  1.66it/s]
 47%|████▋     | 30770/65536 [5:17:00<5:57:51,  1.62it/s]
 47%|████▋     | 30771/65536 [5:17:00<5:50:46,  1.65it/s]
 47%|████▋     | 30772/65536 [5:17:01<6:07:05,  1.58it/s]
 47%|████▋     | 30773/65536 [5:17:01<6:07:58,  1.57it/s]
 47%|████▋     | 30774/65536 [5:17:02<5:52:29,  1.64it/s]
 47%|████▋     | 30775/65536 [5:17:03<6:03:53,  1.59it/s]
 47%|████▋     | 30776/65536 [5:17:03<5:57:37,  1.62it/s]
 47%|████▋     | 30777/65536 [5:17:04<5:54:24,  1.63it/s]
 47%|████▋     | 30778/65536 [5:17:04<5:51:49,  1.65it/s]
 47%|████▋     | 30779/65536 [5:17:05<5:48:31,  1.66it/s]
 47%|████▋     | 30780/65536 [5:17:06<5:50:01,  1.65it/s]
                                                         
{'loss': 1.6784, 'learning_rate': 5.787620913432104e-07, 'epoch': 1900.0}
+
 47%|████▋     | 30780/65536 [5:17:06<5:50:01,  1.65it/s]
 47%|████▋     | 30781/65536 [5:17:06<6:08:20,  1.57it/s]
 47%|████▋     | 30782/65536 [5:17:07<6:02:03,  1.60it/s]
 47%|████▋     | 30783/65536 [5:17:08<6:00:52,  1.61it/s]
 47%|████▋     | 30784/65536 [5:17:08<6:01:16,  1.60it/s]
 47%|████▋     | 30785/65536 [5:17:09<6:02:45,  1.60it/s]
 47%|████▋     | 30786/65536 [5:17:09<5:59:24,  1.61it/s]
 47%|████▋     | 30787/65536 [5:17:10<5:56:04,  1.63it/s]
 47%|████▋     | 30788/65536 [5:17:11<5:56:03,  1.63it/s]
 47%|████▋     | 30789/65536 [5:17:11<6:01:27,  1.60it/s]
 47%|████▋     | 30790/65536 [5:17:12<5:55:35,  1.63it/s]
 47%|████▋     | 30791/65536 [5:17:13<5:54:54,  1.63it/s]
 47%|████▋     | 30792/65536 [5:17:13<5:51:43,  1.65it/s]
 47%|████▋     | 30793/65536 [5:17:14<5:54:54,  1.63it/s]
 47%|████▋     | 30794/65536 [5:17:14<5:59:24,  1.61it/s]
 47%|████▋     | 30795/65536 [5:17:15<5:55:14,  1.63it/s]
 47%|████▋     | 30796/65536 [5:17:16<5:56:29,  1.62it/s]
 47%|████▋     | 30797/65536 [5:17:16<6:05:57,  1.58it/s]
 47%|████▋     | 30798/65536 [5:17:17<6:03:37,  1.59it/s]
 47%|████▋     | 30799/65536 [5:17:18<6:04:56,  1.59it/s]
 47%|████▋     | 30800/65536 [5:17:18<6:08:05,  1.57it/s]
                                                         
{'loss': 1.6472, 'learning_rate': 5.784865923839843e-07, 'epoch': 1901.23}
+
 47%|████▋     | 30800/65536 [5:17:18<6:08:05,  1.57it/s]
 47%|████▋     | 30801/65536 [5:17:19<6:09:43,  1.57it/s]
 47%|████▋     | 30802/65536 [5:17:19<6:07:16,  1.58it/s]
 47%|████▋     | 30803/65536 [5:17:20<5:59:25,  1.61it/s]
 47%|████▋     | 30804/65536 [5:17:21<5:58:10,  1.62it/s]
 47%|████▋     | 30805/65536 [5:17:21<5:58:06,  1.62it/s]
 47%|████▋     | 30806/65536 [5:17:22<5:45:30,  1.68it/s]
 47%|████▋     | 30807/65536 [5:17:22<5:48:14,  1.66it/s]
 47%|████▋     | 30808/65536 [5:17:23<5:46:41,  1.67it/s]
 47%|████▋     | 30809/65536 [5:17:24<5:51:23,  1.65it/s]
 47%|████▋     | 30810/65536 [5:17:24<5:47:34,  1.67it/s]
 47%|████▋     | 30811/65536 [5:17:25<5:52:02,  1.64it/s]
 47%|████▋     | 30812/65536 [5:17:25<5:53:31,  1.64it/s]
 47%|████▋     | 30813/65536 [5:17:26<6:05:19,  1.58it/s]
 47%|████▋     | 30814/65536 [5:17:27<5:55:33,  1.63it/s]
 47%|████▋     | 30815/65536 [5:17:27<5:59:30,  1.61it/s]
 47%|████▋     | 30816/65536 [5:17:28<5:50:43,  1.65it/s]
 47%|████▋     | 30817/65536 [5:17:29<5:49:32,  1.66it/s]
 47%|████▋     | 30818/65536 [5:17:29<5:57:52,  1.62it/s]
 47%|████▋     | 30819/65536 [5:17:30<6:14:10,  1.55it/s]
 47%|████▋     | 30820/65536 [5:17:31<6:08:29,  1.57it/s]
                                                         
{'loss': 1.6882, 'learning_rate': 5.782110934247582e-07, 'epoch': 1902.47}
+
 47%|████▋     | 30820/65536 [5:17:31<6:08:29,  1.57it/s]
 47%|████▋     | 30821/65536 [5:17:31<6:11:35,  1.56it/s]
 47%|████▋     | 30822/65536 [5:17:32<6:08:59,  1.57it/s]
 47%|████▋     | 30823/65536 [5:17:32<6:05:24,  1.58it/s]
 47%|████▋     | 30824/65536 [5:17:33<6:00:40,  1.60it/s]
 47%|████▋     | 30825/65536 [5:17:34<5:54:50,  1.63it/s]
 47%|████▋     | 30826/65536 [5:17:34<5:49:25,  1.66it/s]
 47%|████▋     | 30827/65536 [5:17:35<5:49:54,  1.65it/s]
 47%|████▋     | 30828/65536 [5:17:35<5:52:08,  1.64it/s]
 47%|████▋     | 30829/65536 [5:17:36<6:09:16,  1.57it/s]
 47%|████▋     | 30830/65536 [5:17:37<6:05:50,  1.58it/s]
 47%|████▋     | 30831/65536 [5:17:37<6:06:30,  1.58it/s]
 47%|████▋     | 30832/65536 [5:17:38<5:57:54,  1.62it/s]
 47%|████▋     | 30833/65536 [5:17:39<5:57:08,  1.62it/s]
 47%|████▋     | 30834/65536 [5:17:39<5:51:42,  1.64it/s]
 47%|████▋     | 30835/65536 [5:17:40<5:50:48,  1.65it/s]
 47%|████▋     | 30836/65536 [5:17:40<5:53:30,  1.64it/s]
 47%|████▋     | 30837/65536 [5:17:41<5:55:00,  1.63it/s]
 47%|████▋     | 30838/65536 [5:17:42<5:59:26,  1.61it/s]
 47%|████▋     | 30839/65536 [5:17:42<5:56:25,  1.62it/s]
 47%|████▋     | 30840/65536 [5:17:43<5:47:13,  1.67it/s]
                                                         
{'loss': 1.6812, 'learning_rate': 5.779355944655319e-07, 'epoch': 1903.7}
+
 47%|████▋     | 30840/65536 [5:17:43<5:47:13,  1.67it/s]
 47%|████▋     | 30841/65536 [5:17:43<5:56:48,  1.62it/s]
 47%|████▋     | 30842/65536 [5:17:44<5:59:23,  1.61it/s]
 47%|████▋     | 30843/65536 [5:17:45<5:57:14,  1.62it/s]
 47%|████▋     | 30844/65536 [5:17:45<5:55:16,  1.63it/s]
 47%|████▋     | 30845/65536 [5:17:46<6:18:14,  1.53it/s]
 47%|████▋     | 30846/65536 [5:17:47<6:05:37,  1.58it/s]
 47%|████▋     | 30847/65536 [5:17:47<6:08:02,  1.57it/s]
 47%|████▋     | 30848/65536 [5:17:48<6:05:28,  1.58it/s]
 47%|████▋     | 30849/65536 [5:17:49<6:12:03,  1.55it/s]
 47%|████▋     | 30850/65536 [5:17:49<6:05:48,  1.58it/s]
 47%|████▋     | 30851/65536 [5:17:50<6:02:47,  1.59it/s]
 47%|████▋     | 30852/65536 [5:17:50<5:55:08,  1.63it/s]
 47%|████▋     | 30853/65536 [5:17:51<5:49:14,  1.66it/s]
 47%|████▋     | 30854/65536 [5:17:52<5:41:53,  1.69it/s]
 47%|████▋     | 30855/65536 [5:17:52<5:36:51,  1.72it/s]
 47%|████▋     | 30856/65536 [5:17:53<5:41:32,  1.69it/s]
 47%|████▋     | 30857/65536 [5:17:53<5:48:44,  1.66it/s]
 47%|████▋     | 30858/65536 [5:17:54<5:45:46,  1.67it/s]
 47%|████▋     | 30859/65536 [5:17:55<5:54:52,  1.63it/s]
 47%|████▋     | 30860/65536 [5:17:55<6:06:21,  1.58it/s]
                                                         
{'loss': 1.6791, 'learning_rate': 5.776600955063058e-07, 'epoch': 1904.94}
+
 47%|████▋     | 30860/65536 [5:17:55<6:06:21,  1.58it/s]
 47%|████▋     | 30861/65536 [5:17:56<6:07:23,  1.57it/s]
 47%|████▋     | 30862/65536 [5:17:57<6:22:18,  1.51it/s]
 47%|████▋     | 30863/65536 [5:17:57<6:14:31,  1.54it/s]
 47%|████▋     | 30864/65536 [5:17:58<6:05:22,  1.58it/s]
 47%|████▋     | 30865/65536 [5:17:58<6:02:34,  1.59it/s]
 47%|████▋     | 30866/65536 [5:17:59<6:03:54,  1.59it/s]
 47%|████▋     | 30867/65536 [5:18:00<6:05:25,  1.58it/s]
 47%|████▋     | 30868/65536 [5:18:00<5:55:44,  1.62it/s]
 47%|████▋     | 30869/65536 [5:18:01<5:55:21,  1.63it/s]
 47%|████▋     | 30870/65536 [5:18:02<6:01:40,  1.60it/s]
 47%|████▋     | 30871/65536 [5:18:02<5:56:33,  1.62it/s]
 47%|████▋     | 30872/65536 [5:18:03<5:53:48,  1.63it/s]
 47%|████▋     | 30873/65536 [5:18:03<5:51:04,  1.65it/s]
 47%|████▋     | 30874/65536 [5:18:04<5:54:16,  1.63it/s]
 47%|████▋     | 30875/65536 [5:18:05<5:49:25,  1.65it/s]
 47%|████▋     | 30876/65536 [5:18:05<5:51:00,  1.65it/s]
 47%|████▋     | 30877/65536 [5:18:06<5:52:48,  1.64it/s]
 47%|████▋     | 30878/65536 [5:18:06<6:07:26,  1.57it/s]
 47%|████▋     | 30879/65536 [5:18:07<5:55:09,  1.63it/s]
 47%|████▋     | 30880/65536 [5:18:08<5:51:42,  1.64it/s]
                                                         
{'loss': 1.6917, 'learning_rate': 5.773845965470796e-07, 'epoch': 1906.17}
+
 47%|████▋     | 30880/65536 [5:18:08<5:51:42,  1.64it/s]
 47%|████▋     | 30881/65536 [5:18:08<5:55:42,  1.62it/s]
 47%|████▋     | 30882/65536 [5:18:09<5:48:57,  1.66it/s]
 47%|████▋     | 30883/65536 [5:18:10<5:57:42,  1.61it/s]
 47%|████▋     | 30884/65536 [5:18:10<5:57:16,  1.62it/s]
 47%|████▋     | 30885/65536 [5:18:11<5:46:47,  1.67it/s]
 47%|████▋     | 30886/65536 [5:18:11<6:04:09,  1.59it/s]
 47%|████▋     | 30887/65536 [5:18:12<6:01:58,  1.60it/s]
 47%|████▋     | 30888/65536 [5:18:13<6:02:58,  1.59it/s]
 47%|████▋     | 30889/65536 [5:18:13<6:05:27,  1.58it/s]
 47%|████▋     | 30890/65536 [5:18:14<6:02:16,  1.59it/s]
 47%|████▋     | 30891/65536 [5:18:15<5:59:54,  1.60it/s]
 47%|████▋     | 30892/65536 [5:18:15<5:54:35,  1.63it/s]
 47%|████▋     | 30893/65536 [5:18:16<5:50:14,  1.65it/s]
 47%|████▋     | 30894/65536 [5:18:16<6:05:58,  1.58it/s]
 47%|████▋     | 30895/65536 [5:18:17<6:10:00,  1.56it/s]
 47%|████▋     | 30896/65536 [5:18:18<6:03:47,  1.59it/s]
 47%|████▋     | 30897/65536 [5:18:18<6:01:22,  1.60it/s]
 47%|████▋     | 30898/65536 [5:18:19<5:56:32,  1.62it/s]
 47%|████▋     | 30899/65536 [5:18:20<6:08:12,  1.57it/s]
 47%|████▋     | 30900/65536 [5:18:20<6:01:36,  1.60it/s]
                                                         
{'loss': 1.6563, 'learning_rate': 5.771090975878535e-07, 'epoch': 1907.41}
+
 47%|████▋     | 30900/65536 [5:18:20<6:01:36,  1.60it/s]
 47%|████▋     | 30901/65536 [5:18:21<5:56:42,  1.62it/s]
 47%|████▋     | 30902/65536 [5:18:21<5:58:53,  1.61it/s]
 47%|████▋     | 30903/65536 [5:18:22<5:56:53,  1.62it/s]
 47%|████▋     | 30904/65536 [5:18:23<5:51:09,  1.64it/s]
 47%|████▋     | 30905/65536 [5:18:23<6:06:50,  1.57it/s]
 47%|████▋     | 30906/65536 [5:18:24<6:05:35,  1.58it/s]
 47%|████▋     | 30907/65536 [5:18:24<5:56:18,  1.62it/s]
 47%|████▋     | 30908/65536 [5:18:25<5:50:02,  1.65it/s]
 47%|████▋     | 30909/65536 [5:18:26<5:48:09,  1.66it/s]
 47%|████▋     | 30910/65536 [5:18:26<6:11:33,  1.55it/s]
 47%|████▋     | 30911/65536 [5:18:27<6:12:26,  1.55it/s]
 47%|████▋     | 30912/65536 [5:18:28<6:09:39,  1.56it/s]
 47%|████▋     | 30913/65536 [5:18:28<6:03:45,  1.59it/s]
 47%|████▋     | 30914/65536 [5:18:29<6:12:36,  1.55it/s]
 47%|████▋     | 30915/65536 [5:18:30<6:05:00,  1.58it/s]
 47%|████▋     | 30916/65536 [5:18:30<6:00:30,  1.60it/s]
 47%|████▋     | 30917/65536 [5:18:31<5:50:08,  1.65it/s]
 47%|████▋     | 30918/65536 [5:18:31<5:55:00,  1.63it/s]
 47%|████▋     | 30919/65536 [5:18:32<5:57:03,  1.62it/s]
 47%|████▋     | 30920/65536 [5:18:33<5:57:52,  1.61it/s]
                                                         
{'loss': 1.6402, 'learning_rate': 5.768335986286274e-07, 'epoch': 1908.64}
+
 47%|████▋     | 30920/65536 [5:18:33<5:57:52,  1.61it/s]
 47%|████▋     | 30921/65536 [5:18:33<5:50:59,  1.64it/s]
 47%|████▋     | 30922/65536 [5:18:34<5:53:28,  1.63it/s]
 47%|████▋     | 30923/65536 [5:18:34<5:46:51,  1.66it/s]
 47%|████▋     | 30924/65536 [5:18:35<5:51:12,  1.64it/s]
 47%|████▋     | 30925/65536 [5:18:36<5:54:09,  1.63it/s]
 47%|████▋     | 30926/65536 [5:18:36<6:03:22,  1.59it/s]
 47%|████▋     | 30927/65536 [5:18:37<5:58:01,  1.61it/s]
 47%|████▋     | 30928/65536 [5:18:38<6:00:18,  1.60it/s]
 47%|████▋     | 30929/65536 [5:18:38<5:56:22,  1.62it/s]
 47%|████▋     | 30930/65536 [5:18:39<5:55:24,  1.62it/s]
 47%|████▋     | 30931/65536 [5:18:39<5:54:11,  1.63it/s]
 47%|████▋     | 30932/65536 [5:18:40<5:54:20,  1.63it/s]
 47%|████▋     | 30933/65536 [5:18:41<5:49:24,  1.65it/s]
 47%|████▋     | 30934/65536 [5:18:41<5:44:56,  1.67it/s]
 47%|████▋     | 30935/65536 [5:18:42<6:00:31,  1.60it/s]
 47%|████▋     | 30936/65536 [5:18:43<6:05:06,  1.58it/s]
 47%|████▋     | 30937/65536 [5:18:43<6:06:51,  1.57it/s]
 47%|████▋     | 30938/65536 [5:18:44<5:56:20,  1.62it/s]
 47%|████▋     | 30939/65536 [5:18:44<5:53:08,  1.63it/s]
 47%|████▋     | 30940/65536 [5:18:45<5:52:23,  1.64it/s]
                                                         
{'loss': 1.6835, 'learning_rate': 5.765580996694013e-07, 'epoch': 1909.88}
+
 47%|████▋     | 30940/65536 [5:18:45<5:52:23,  1.64it/s]
 47%|████▋     | 30941/65536 [5:18:46<5:50:09,  1.65it/s]
 47%|████▋     | 30942/65536 [5:18:46<5:49:15,  1.65it/s]
 47%|████▋     | 30943/65536 [5:18:47<5:55:15,  1.62it/s]
 47%|████▋     | 30944/65536 [5:18:47<6:04:00,  1.58it/s]
 47%|████▋     | 30945/65536 [5:18:48<6:02:14,  1.59it/s]
 47%|████▋     | 30946/65536 [5:18:49<5:58:29,  1.61it/s]
 47%|████▋     | 30947/65536 [5:18:49<5:57:56,  1.61it/s]
 47%|████▋     | 30948/65536 [5:18:50<6:08:01,  1.57it/s]
 47%|████▋     | 30949/65536 [5:18:51<6:11:03,  1.55it/s]
 47%|████▋     | 30950/65536 [5:18:51<5:58:10,  1.61it/s]
 47%|████▋     | 30951/65536 [5:18:52<5:51:03,  1.64it/s]
 47%|████▋     | 30952/65536 [5:18:52<5:58:45,  1.61it/s]
 47%|████▋     | 30953/65536 [5:18:53<5:55:59,  1.62it/s]
 47%|████▋     | 30954/65536 [5:18:54<5:49:43,  1.65it/s]
 47%|████▋     | 30955/65536 [5:18:54<5:53:46,  1.63it/s]
 47%|████▋     | 30956/65536 [5:18:55<5:48:17,  1.65it/s]
 47%|████▋     | 30957/65536 [5:18:55<5:50:06,  1.65it/s]
 47%|████▋     | 30958/65536 [5:18:56<5:51:58,  1.64it/s]
 47%|████▋     | 30959/65536 [5:18:57<6:14:46,  1.54it/s]
 47%|████▋     | 30960/65536 [5:18:57<6:04:00,  1.58it/s]
                                                         
{'loss': 1.6657, 'learning_rate': 5.762826007101752e-07, 'epoch': 1911.11}
+
 47%|████▋     | 30960/65536 [5:18:57<6:04:00,  1.58it/s]
 47%|████▋     | 30961/65536 [5:18:58<6:07:26,  1.57it/s]
 47%|████▋     | 30962/65536 [5:18:59<6:08:36,  1.56it/s]
 47%|████▋     | 30963/65536 [5:18:59<5:57:37,  1.61it/s]
 47%|████▋     | 30964/65536 [5:19:00<5:50:34,  1.64it/s]
 47%|████▋     | 30965/65536 [5:19:00<5:49:54,  1.65it/s]
 47%|████▋     | 30966/65536 [5:19:01<5:45:11,  1.67it/s]
 47%|████▋     | 30967/65536 [5:19:02<5:43:55,  1.68it/s]
 47%|████▋     | 30968/65536 [5:19:02<5:48:35,  1.65it/s]
 47%|████▋     | 30969/65536 [5:19:03<5:44:18,  1.67it/s]
 47%|████▋     | 30970/65536 [5:19:03<5:43:23,  1.68it/s]
 47%|████▋     | 30971/65536 [5:19:04<5:40:21,  1.69it/s]
 47%|████▋     | 30972/65536 [5:19:05<5:54:45,  1.62it/s]
 47%|████▋     | 30973/65536 [5:19:05<5:47:42,  1.66it/s]
 47%|████▋     | 30974/65536 [5:19:06<5:54:07,  1.63it/s]
 47%|████▋     | 30975/65536 [5:19:07<6:10:41,  1.55it/s]
 47%|████▋     | 30976/65536 [5:19:07<6:06:40,  1.57it/s]
 47%|████▋     | 30977/65536 [5:19:08<5:54:35,  1.62it/s]
 47%|████▋     | 30978/65536 [5:19:08<5:53:26,  1.63it/s]
 47%|████▋     | 30979/65536 [5:19:09<6:00:23,  1.60it/s]
 47%|████▋     | 30980/65536 [5:19:10<5:56:44,  1.61it/s]
                                                         
{'loss': 1.7137, 'learning_rate': 5.76007101750949e-07, 'epoch': 1912.35}
+
 47%|████▋     | 30980/65536 [5:19:10<5:56:44,  1.61it/s]
 47%|████▋     | 30981/65536 [5:19:10<5:50:36,  1.64it/s]
 47%|████▋     | 30982/65536 [5:19:11<5:55:20,  1.62it/s]
 47%|████▋     | 30983/65536 [5:19:11<5:54:33,  1.62it/s]
 47%|████▋     | 30984/65536 [5:19:12<6:02:30,  1.59it/s]
 47%|████▋     | 30985/65536 [5:19:13<5:55:42,  1.62it/s]
 47%|████▋     | 30986/65536 [5:19:13<6:01:39,  1.59it/s]
 47%|████▋     | 30987/65536 [5:19:14<6:02:26,  1.59it/s]
 47%|████▋     | 30988/65536 [5:19:15<5:58:44,  1.61it/s]
 47%|████▋     | 30989/65536 [5:19:15<5:51:01,  1.64it/s]
 47%|████▋     | 30990/65536 [5:19:16<5:53:12,  1.63it/s]
 47%|████▋     | 30991/65536 [5:19:16<6:02:49,  1.59it/s]
 47%|████▋     | 30992/65536 [5:19:17<6:01:50,  1.59it/s]
 47%|████▋     | 30993/65536 [5:19:18<6:06:36,  1.57it/s]
 47%|████▋     | 30994/65536 [5:19:18<5:55:25,  1.62it/s]
 47%|████▋     | 30995/65536 [5:19:19<5:59:25,  1.60it/s]
 47%|████▋     | 30996/65536 [5:19:20<6:06:32,  1.57it/s]
 47%|████▋     | 30997/65536 [5:19:20<5:56:35,  1.61it/s]
 47%|████▋     | 30998/65536 [5:19:21<6:04:23,  1.58it/s]
 47%|████▋     | 30999/65536 [5:19:21<5:55:59,  1.62it/s]
 47%|████▋     | 31000/65536 [5:19:22<5:48:34,  1.65it/s]
                                                         
{'loss': 1.6696, 'learning_rate': 5.757316027917229e-07, 'epoch': 1913.58}
+
 47%|████▋     | 31000/65536 [5:19:22<5:48:34,  1.65it/s]
 47%|████▋     | 31001/65536 [5:19:23<5:57:10,  1.61it/s]
 47%|████▋     | 31002/65536 [5:19:23<5:55:35,  1.62it/s]
 47%|████▋     | 31003/65536 [5:19:24<5:48:06,  1.65it/s]
 47%|████▋     | 31004/65536 [5:19:25<5:48:58,  1.65it/s]
 47%|████▋     | 31005/65536 [5:19:25<5:50:54,  1.64it/s]
 47%|████▋     | 31006/65536 [5:19:26<5:56:23,  1.61it/s]
 47%|████▋     | 31007/65536 [5:19:26<6:06:58,  1.57it/s]
 47%|████▋     | 31008/65536 [5:19:27<6:10:22,  1.55it/s]
 47%|████▋     | 31009/65536 [5:19:28<6:00:39,  1.60it/s]
 47%|████▋     | 31010/65536 [5:19:28<5:58:06,  1.61it/s]
 47%|████▋     | 31011/65536 [5:19:29<6:05:43,  1.57it/s]
 47%|████▋     | 31012/65536 [5:19:30<6:02:00,  1.59it/s]
 47%|████▋     | 31013/65536 [5:19:30<5:58:43,  1.60it/s]
 47%|████▋     | 31014/65536 [5:19:31<5:51:39,  1.64it/s]
 47%|████▋     | 31015/65536 [5:19:31<5:42:20,  1.68it/s]
 47%|████▋     | 31016/65536 [5:19:32<5:41:03,  1.69it/s]
 47%|████▋     | 31017/65536 [5:19:33<5:55:37,  1.62it/s]
 47%|████▋     | 31018/65536 [5:19:33<5:59:07,  1.60it/s]
 47%|████▋     | 31019/65536 [5:19:34<5:48:59,  1.65it/s]
 47%|████▋     | 31020/65536 [5:19:34<5:45:51,  1.66it/s]
                                                         
{'loss': 1.6753, 'learning_rate': 5.754561038324967e-07, 'epoch': 1914.81}
+
 47%|████▋     | 31020/65536 [5:19:34<5:45:51,  1.66it/s]
 47%|████▋     | 31021/65536 [5:19:35<5:53:35,  1.63it/s]
 47%|████▋     | 31022/65536 [5:19:36<5:49:38,  1.65it/s]
 47%|████▋     | 31023/65536 [5:19:36<5:50:19,  1.64it/s]
 47%|████▋     | 31024/65536 [5:19:37<6:04:10,  1.58it/s]
 47%|████▋     | 31025/65536 [5:19:38<5:58:16,  1.61it/s]
 47%|████▋     | 31026/65536 [5:19:38<5:57:59,  1.61it/s]
 47%|████▋     | 31027/65536 [5:19:39<5:48:06,  1.65it/s]
 47%|████▋     | 31028/65536 [5:19:39<6:04:27,  1.58it/s]
 47%|████▋     | 31029/65536 [5:19:40<6:06:08,  1.57it/s]
 47%|████▋     | 31030/65536 [5:19:41<6:03:44,  1.58it/s]
 47%|████▋     | 31031/65536 [5:19:41<6:08:54,  1.56it/s]
 47%|████▋     | 31032/65536 [5:19:42<6:03:16,  1.58it/s]
 47%|████▋     | 31033/65536 [5:19:43<5:57:05,  1.61it/s]
 47%|████▋     | 31034/65536 [5:19:43<5:50:55,  1.64it/s]
 47%|████▋     | 31035/65536 [5:19:44<5:50:22,  1.64it/s]
 47%|████▋     | 31036/65536 [5:19:44<5:48:19,  1.65it/s]
 47%|████▋     | 31037/65536 [5:19:45<5:48:51,  1.65it/s]
 47%|████▋     | 31038/65536 [5:19:46<5:56:15,  1.61it/s]
 47%|████▋     | 31039/65536 [5:19:46<5:52:30,  1.63it/s]
 47%|████▋     | 31040/65536 [5:19:47<6:04:27,  1.58it/s]
                                                         
{'loss': 1.6665, 'learning_rate': 5.751806048732704e-07, 'epoch': 1916.05}
+
 47%|████▋     | 31040/65536 [5:19:47<6:04:27,  1.58it/s]
 47%|████▋     | 31041/65536 [5:19:47<5:57:40,  1.61it/s]
 47%|████▋     | 31042/65536 [5:19:48<5:52:19,  1.63it/s]
 47%|████▋     | 31043/65536 [5:19:49<5:53:15,  1.63it/s]
 47%|████▋     | 31044/65536 [5:19:49<5:50:21,  1.64it/s]
 47%|████▋     | 31045/65536 [5:19:50<6:08:48,  1.56it/s]
 47%|████▋     | 31046/65536 [5:19:51<6:04:17,  1.58it/s]
 47%|████▋     | 31047/65536 [5:19:51<5:58:11,  1.60it/s]
 47%|████▋     | 31048/65536 [5:19:52<5:51:58,  1.63it/s]
 47%|████▋     | 31049/65536 [5:19:52<6:00:07,  1.60it/s]
 47%|████▋     | 31050/65536 [5:19:53<5:52:48,  1.63it/s]
 47%|████▋     | 31051/65536 [5:19:54<5:56:12,  1.61it/s]
 47%|████▋     | 31052/65536 [5:19:54<5:58:33,  1.60it/s]
 47%|████▋     | 31053/65536 [5:19:55<5:54:06,  1.62it/s]
 47%|████▋     | 31054/65536 [5:19:56<5:51:52,  1.63it/s]
 47%|████▋     | 31055/65536 [5:19:56<5:48:31,  1.65it/s]
 47%|████▋     | 31056/65536 [5:19:57<6:01:01,  1.59it/s]
 47%|████▋     | 31057/65536 [5:19:57<5:52:48,  1.63it/s]
 47%|████▋     | 31058/65536 [5:19:58<5:49:16,  1.65it/s]
 47%|████▋     | 31059/65536 [5:19:59<6:02:20,  1.59it/s]
 47%|████▋     | 31060/65536 [5:19:59<5:57:53,  1.61it/s]
                                                         
{'loss': 1.6984, 'learning_rate': 5.749051059140442e-07, 'epoch': 1917.28}
+
 47%|████▋     | 31060/65536 [5:19:59<5:57:53,  1.61it/s]
 47%|████▋     | 31061/65536 [5:20:00<5:50:40,  1.64it/s]
 47%|████▋     | 31062/65536 [5:20:00<5:46:06,  1.66it/s]
 47%|████▋     | 31063/65536 [5:20:01<5:42:55,  1.68it/s]
 47%|████▋     | 31064/65536 [5:20:02<5:36:34,  1.71it/s]
 47%|████▋     | 31065/65536 [5:20:02<5:37:50,  1.70it/s]
 47%|████▋     | 31066/65536 [5:20:03<5:39:11,  1.69it/s]
 47%|████▋     | 31067/65536 [5:20:03<5:41:28,  1.68it/s]
 47%|████▋     | 31068/65536 [5:20:04<5:47:10,  1.65it/s]
 47%|████▋     | 31069/65536 [5:20:05<5:46:49,  1.66it/s]
 47%|████▋     | 31070/65536 [5:20:05<5:56:04,  1.61it/s]
 47%|████▋     | 31071/65536 [5:20:06<6:04:41,  1.58it/s]
 47%|████▋     | 31072/65536 [5:20:07<6:09:18,  1.56it/s]
 47%|████▋     | 31073/65536 [5:20:07<5:55:52,  1.61it/s]
 47%|████▋     | 31074/65536 [5:20:08<5:54:37,  1.62it/s]
 47%|████▋     | 31075/65536 [5:20:08<5:53:14,  1.63it/s]
 47%|████▋     | 31076/65536 [5:20:09<5:46:17,  1.66it/s]
 47%|████▋     | 31077/65536 [5:20:10<5:50:23,  1.64it/s]
 47%|████▋     | 31078/65536 [5:20:10<5:50:32,  1.64it/s]
 47%|████▋     | 31079/65536 [5:20:11<5:52:29,  1.63it/s]
 47%|████▋     | 31080/65536 [5:20:11<5:56:58,  1.61it/s]
                                                         
{'loss': 1.6797, 'learning_rate': 5.746296069548181e-07, 'epoch': 1918.52}
+
 47%|████▋     | 31080/65536 [5:20:11<5:56:58,  1.61it/s]
 47%|████▋     | 31081/65536 [5:20:12<6:05:43,  1.57it/s]
 47%|████▋     | 31082/65536 [5:20:13<5:59:42,  1.60it/s]
 47%|████▋     | 31083/65536 [5:20:13<6:03:24,  1.58it/s]
 47%|████▋     | 31084/65536 [5:20:14<6:05:59,  1.57it/s]
 47%|████▋     | 31085/65536 [5:20:15<6:07:34,  1.56it/s]
 47%|████▋     | 31086/65536 [5:20:15<5:57:17,  1.61it/s]
 47%|████▋     | 31087/65536 [5:20:16<6:02:53,  1.58it/s]
 47%|████▋     | 31088/65536 [5:20:17<6:10:57,  1.55it/s]
 47%|████▋     | 31089/65536 [5:20:17<6:00:10,  1.59it/s]
 47%|████▋     | 31090/65536 [5:20:18<6:04:11,  1.58it/s]
 47%|████▋     | 31091/65536 [5:20:18<6:02:18,  1.58it/s]
 47%|████▋     | 31092/65536 [5:20:19<6:05:16,  1.57it/s]
 47%|████▋     | 31093/65536 [5:20:20<6:01:48,  1.59it/s]
 47%|████▋     | 31094/65536 [5:20:20<5:55:20,  1.62it/s]
 47%|████▋     | 31095/65536 [5:20:21<5:57:14,  1.61it/s]
 47%|████▋     | 31096/65536 [5:20:22<5:55:34,  1.61it/s]
 47%|████▋     | 31097/65536 [5:20:22<5:48:57,  1.64it/s]
 47%|████▋     | 31098/65536 [5:20:23<5:44:49,  1.66it/s]
 47%|████▋     | 31099/65536 [5:20:23<5:41:29,  1.68it/s]
 47%|████▋     | 31100/65536 [5:20:24<5:49:38,  1.64it/s]
                                                         
{'loss': 1.6895, 'learning_rate': 5.74354107995592e-07, 'epoch': 1919.75}
+
 47%|████▋     | 31100/65536 [5:20:24<5:49:38,  1.64it/s]
 47%|████▋     | 31101/65536 [5:20:25<5:47:47,  1.65it/s]
 47%|████▋     | 31102/65536 [5:20:25<5:46:50,  1.65it/s]
 47%|████▋     | 31103/65536 [5:20:26<5:49:21,  1.64it/s]
 47%|████▋     | 31104/65536 [5:20:26<5:53:55,  1.62it/s]
 47%|████▋     | 31105/65536 [5:20:27<6:02:20,  1.58it/s]
 47%|████▋     | 31106/65536 [5:20:28<5:57:34,  1.60it/s]
 47%|████▋     | 31107/65536 [5:20:28<5:51:39,  1.63it/s]
 47%|████▋     | 31108/65536 [5:20:29<5:57:18,  1.61it/s]
 47%|████▋     | 31109/65536 [5:20:29<5:54:51,  1.62it/s]
 47%|████▋     | 31110/65536 [5:20:30<5:44:40,  1.66it/s]
 47%|████▋     | 31111/65536 [5:20:31<5:52:23,  1.63it/s]
 47%|████▋     | 31112/65536 [5:20:31<5:52:12,  1.63it/s]
 47%|████▋     | 31113/65536 [5:20:32<5:43:55,  1.67it/s]
 47%|████▋     | 31114/65536 [5:20:32<5:43:02,  1.67it/s]
 47%|████▋     | 31115/65536 [5:20:33<5:46:28,  1.66it/s]
 47%|████▋     | 31116/65536 [5:20:34<5:48:08,  1.65it/s]
 47%|████▋     | 31117/65536 [5:20:34<5:51:59,  1.63it/s]
 47%|████▋     | 31118/65536 [5:20:35<6:04:44,  1.57it/s]
 47%|████▋     | 31119/65536 [5:20:36<5:59:56,  1.59it/s]
 47%|████▋     | 31120/65536 [5:20:36<5:56:32,  1.61it/s]
                                                         
{'loss': 1.6677, 'learning_rate': 5.740786090363658e-07, 'epoch': 1920.99}
+
 47%|████▋     | 31120/65536 [5:20:36<5:56:32,  1.61it/s]
 47%|████▋     | 31121/65536 [5:20:37<6:06:16,  1.57it/s]
 47%|████▋     | 31122/65536 [5:20:38<6:02:14,  1.58it/s]
 47%|████▋     | 31123/65536 [5:20:38<5:53:05,  1.62it/s]
 47%|████▋     | 31124/65536 [5:20:39<6:06:14,  1.57it/s]
 47%|████▋     | 31125/65536 [5:20:39<5:56:41,  1.61it/s]
 47%|████▋     | 31126/65536 [5:20:40<6:03:18,  1.58it/s]
 47%|████▋     | 31127/65536 [5:20:41<6:06:52,  1.56it/s]
 47%|████▋     | 31128/65536 [5:20:41<5:58:16,  1.60it/s]
 47%|████▋     | 31129/65536 [5:20:42<5:57:51,  1.60it/s]
 48%|████▊     | 31130/65536 [5:20:43<6:02:12,  1.58it/s]
 48%|████▊     | 31131/65536 [5:20:43<5:56:07,  1.61it/s]
 48%|████▊     | 31132/65536 [5:20:44<5:44:35,  1.66it/s]
 48%|████▊     | 31133/65536 [5:20:44<5:47:47,  1.65it/s]
 48%|████▊     | 31134/65536 [5:20:45<5:47:07,  1.65it/s]
 48%|████▊     | 31135/65536 [5:20:46<5:47:22,  1.65it/s]
 48%|████▊     | 31136/65536 [5:20:46<5:49:28,  1.64it/s]
 48%|████▊     | 31137/65536 [5:20:47<6:04:06,  1.57it/s]
 48%|████▊     | 31138/65536 [5:20:47<5:58:35,  1.60it/s]
 48%|████▊     | 31139/65536 [5:20:48<6:04:34,  1.57it/s]
 48%|████▊     | 31140/65536 [5:20:49<6:01:37,  1.59it/s]
                                                         
{'loss': 1.6921, 'learning_rate': 5.738031100771397e-07, 'epoch': 1922.22}
+
 48%|████▊     | 31140/65536 [5:20:49<6:01:37,  1.59it/s]
 48%|████▊     | 31141/65536 [5:20:49<5:54:23,  1.62it/s]
 48%|████▊     | 31142/65536 [5:20:50<5:49:08,  1.64it/s]
 48%|████▊     | 31143/65536 [5:20:50<5:46:44,  1.65it/s]
 48%|████▊     | 31144/65536 [5:20:51<5:51:30,  1.63it/s]
 48%|████▊     | 31145/65536 [5:20:52<5:54:14,  1.62it/s]
 48%|████▊     | 31146/65536 [5:20:52<5:51:51,  1.63it/s]
 48%|████▊     | 31147/65536 [5:20:53<5:53:06,  1.62it/s]
 48%|████▊     | 31148/65536 [5:20:54<5:49:57,  1.64it/s]
 48%|████▊     | 31149/65536 [5:20:54<5:48:16,  1.65it/s]
 48%|████▊     | 31150/65536 [5:20:55<5:56:29,  1.61it/s]
 48%|████▊     | 31151/65536 [5:20:56<6:08:33,  1.55it/s]
 48%|████▊     | 31152/65536 [5:20:56<6:03:44,  1.58it/s]
 48%|████▊     | 31153/65536 [5:20:57<6:17:26,  1.52it/s]
 48%|████▊     | 31154/65536 [5:20:57<6:09:58,  1.55it/s]
 48%|████▊     | 31155/65536 [5:20:58<6:05:54,  1.57it/s]
 48%|████▊     | 31156/65536 [5:20:59<6:02:55,  1.58it/s]
 48%|████▊     | 31157/65536 [5:20:59<5:53:37,  1.62it/s]
 48%|████▊     | 31158/65536 [5:21:00<6:00:05,  1.59it/s]
 48%|████▊     | 31159/65536 [5:21:01<5:56:47,  1.61it/s]
 48%|████▊     | 31160/65536 [5:21:01<5:49:09,  1.64it/s]
                                                         
{'loss': 1.6702, 'learning_rate': 5.735276111179135e-07, 'epoch': 1923.46}
+
 48%|████▊     | 31160/65536 [5:21:01<5:49:09,  1.64it/s]
 48%|████▊     | 31161/65536 [5:21:02<5:41:57,  1.68it/s]
 48%|████▊     | 31162/65536 [5:21:02<5:40:39,  1.68it/s]
 48%|████▊     | 31163/65536 [5:21:03<5:47:05,  1.65it/s]
 48%|████▊     | 31164/65536 [5:21:04<5:45:55,  1.66it/s]
 48%|████▊     | 31165/65536 [5:21:04<5:44:51,  1.66it/s]
 48%|████▊     | 31166/65536 [5:21:05<5:45:08,  1.66it/s]
 48%|████▊     | 31167/65536 [5:21:05<5:43:53,  1.67it/s]
 48%|████▊     | 31168/65536 [5:21:06<5:55:48,  1.61it/s]
 48%|████▊     | 31169/65536 [5:21:07<6:05:02,  1.57it/s]
 48%|████▊     | 31170/65536 [5:21:07<6:01:05,  1.59it/s]
 48%|████▊     | 31171/65536 [5:21:08<6:00:53,  1.59it/s]
 48%|████▊     | 31172/65536 [5:21:09<6:02:11,  1.58it/s]
 48%|████▊     | 31173/65536 [5:21:09<5:55:43,  1.61it/s]
 48%|████▊     | 31174/65536 [5:21:10<5:48:53,  1.64it/s]
 48%|████▊     | 31175/65536 [5:21:10<5:52:42,  1.62it/s]
 48%|████▊     | 31176/65536 [5:21:11<5:48:58,  1.64it/s]
 48%|████▊     | 31177/65536 [5:21:12<5:45:50,  1.66it/s]
 48%|████▊     | 31178/65536 [5:21:12<5:45:09,  1.66it/s]
 48%|████▊     | 31179/65536 [5:21:13<5:46:07,  1.65it/s]
 48%|████▊     | 31180/65536 [5:21:13<5:48:48,  1.64it/s]
                                                         
{'loss': 1.6713, 'learning_rate': 5.732521121586874e-07, 'epoch': 1924.69}
+
 48%|████▊     | 31180/65536 [5:21:13<5:48:48,  1.64it/s]
 48%|████▊     | 31181/65536 [5:21:14<5:50:45,  1.63it/s]
 48%|████▊     | 31182/65536 [5:21:15<5:59:19,  1.59it/s]
 48%|████▊     | 31183/65536 [5:21:15<6:05:22,  1.57it/s]
 48%|████▊     | 31184/65536 [5:21:16<5:57:12,  1.60it/s]
 48%|████▊     | 31185/65536 [5:21:16<5:47:14,  1.65it/s]
 48%|████▊     | 31186/65536 [5:21:17<5:59:37,  1.59it/s]
 48%|████▊     | 31187/65536 [5:21:18<6:14:37,  1.53it/s]
 48%|████▊     | 31188/65536 [5:21:18<6:00:17,  1.59it/s]
 48%|████▊     | 31189/65536 [5:21:19<6:06:41,  1.56it/s]
 48%|████▊     | 31190/65536 [5:21:20<5:54:54,  1.61it/s]
 48%|████▊     | 31191/65536 [5:21:20<5:53:50,  1.62it/s]
 48%|████▊     | 31192/65536 [5:21:21<5:52:01,  1.63it/s]
 48%|████▊     | 31193/65536 [5:21:21<5:48:48,  1.64it/s]
 48%|████▊     | 31194/65536 [5:21:22<5:57:24,  1.60it/s]
 48%|████▊     | 31195/65536 [5:21:23<5:58:11,  1.60it/s]
 48%|████▊     | 31196/65536 [5:21:23<5:54:24,  1.61it/s]
 48%|████▊     | 31197/65536 [5:21:24<5:50:33,  1.63it/s]
 48%|████▊     | 31198/65536 [5:21:25<5:50:24,  1.63it/s]
 48%|████▊     | 31199/65536 [5:21:25<5:49:07,  1.64it/s]
 48%|████▊     | 31200/65536 [5:21:26<5:45:20,  1.66it/s]
                                                         
{'loss': 1.6473, 'learning_rate': 5.729766131994613e-07, 'epoch': 1925.93}
+
 48%|████▊     | 31200/65536 [5:21:26<5:45:20,  1.66it/s]
 48%|████▊     | 31201/65536 [5:21:26<5:47:36,  1.65it/s]
 48%|████▊     | 31202/65536 [5:21:27<5:58:48,  1.59it/s]
 48%|████▊     | 31203/65536 [5:21:28<5:56:50,  1.60it/s]
 48%|████▊     | 31204/65536 [5:21:28<5:47:33,  1.65it/s]
 48%|████▊     | 31205/65536 [5:21:29<5:52:54,  1.62it/s]
 48%|████▊     | 31206/65536 [5:21:30<5:52:40,  1.62it/s]
 48%|████▊     | 31207/65536 [5:21:30<5:49:50,  1.64it/s]
 48%|████▊     | 31208/65536 [5:21:31<5:46:41,  1.65it/s]
 48%|████▊     | 31209/65536 [5:21:31<5:50:13,  1.63it/s]
 48%|████▊     | 31210/65536 [5:21:32<5:44:00,  1.66it/s]
 48%|████▊     | 31211/65536 [5:21:33<5:54:11,  1.62it/s]
 48%|████▊     | 31212/65536 [5:21:33<5:56:35,  1.60it/s]
 48%|████▊     | 31213/65536 [5:21:34<5:52:53,  1.62it/s]
 48%|████▊     | 31214/65536 [5:21:34<5:51:27,  1.63it/s]
 48%|████▊     | 31215/65536 [5:21:35<5:47:36,  1.65it/s]
 48%|████▊     | 31216/65536 [5:21:36<5:54:36,  1.61it/s]
 48%|████▊     | 31217/65536 [5:21:36<6:03:53,  1.57it/s]
 48%|████▊     | 31218/65536 [5:21:37<6:06:51,  1.56it/s]
 48%|████▊     | 31219/65536 [5:21:38<5:55:34,  1.61it/s]
 48%|████▊     | 31220/65536 [5:21:38<5:50:56,  1.63it/s]
                                                         
{'loss': 1.6703, 'learning_rate': 5.727011142402351e-07, 'epoch': 1927.16}
+
 48%|████▊     | 31220/65536 [5:21:38<5:50:56,  1.63it/s]
 48%|████▊     | 31221/65536 [5:21:39<5:53:36,  1.62it/s]
 48%|████▊     | 31222/65536 [5:21:39<5:56:57,  1.60it/s]
 48%|████▊     | 31223/65536 [5:21:40<6:03:24,  1.57it/s]
 48%|████▊     | 31224/65536 [5:21:41<6:03:54,  1.57it/s]
 48%|████▊     | 31225/65536 [5:21:41<5:56:09,  1.61it/s]
 48%|████▊     | 31226/65536 [5:21:42<5:55:14,  1.61it/s]
 48%|████▊     | 31227/65536 [5:21:43<5:54:19,  1.61it/s]
 48%|████▊     | 31228/65536 [5:21:43<5:50:22,  1.63it/s]
 48%|████▊     | 31229/65536 [5:21:44<5:53:30,  1.62it/s]
 48%|████▊     | 31230/65536 [5:21:44<5:51:22,  1.63it/s]
 48%|████▊     | 31231/65536 [5:21:45<5:50:43,  1.63it/s]
 48%|████▊     | 31232/65536 [5:21:46<5:49:27,  1.64it/s]
 48%|████▊     | 31233/65536 [5:21:46<5:44:55,  1.66it/s]
 48%|████▊     | 31234/65536 [5:21:47<6:03:15,  1.57it/s]
 48%|████▊     | 31235/65536 [5:21:47<5:54:18,  1.61it/s]
 48%|████▊     | 31236/65536 [5:21:48<5:48:31,  1.64it/s]
 48%|████▊     | 31237/65536 [5:21:49<5:44:52,  1.66it/s]
 48%|████▊     | 31238/65536 [5:21:49<5:43:29,  1.66it/s]
 48%|████▊     | 31239/65536 [5:21:50<5:43:06,  1.67it/s]
 48%|████▊     | 31240/65536 [5:21:50<5:42:44,  1.67it/s]
                                                         
{'loss': 1.6477, 'learning_rate': 5.72425615281009e-07, 'epoch': 1928.4}
+
 48%|████▊     | 31240/65536 [5:21:50<5:42:44,  1.67it/s]
 48%|████▊     | 31241/65536 [5:21:51<5:50:30,  1.63it/s]
 48%|████▊     | 31242/65536 [5:21:52<5:55:29,  1.61it/s]
 48%|████▊     | 31243/65536 [5:21:52<5:54:04,  1.61it/s]
 48%|████▊     | 31244/65536 [5:21:53<5:51:36,  1.63it/s]
 48%|████▊     | 31245/65536 [5:21:54<5:54:04,  1.61it/s]
 48%|████▊     | 31246/65536 [5:21:54<5:59:45,  1.59it/s]
 48%|████▊     | 31247/65536 [5:21:55<6:16:19,  1.52it/s]
 48%|████▊     | 31248/65536 [5:21:56<6:25:15,  1.48it/s]
 48%|████▊     | 31249/65536 [5:21:56<6:22:48,  1.49it/s]
 48%|████▊     | 31250/65536 [5:21:57<6:26:12,  1.48it/s]
 48%|████▊     | 31251/65536 [5:21:58<6:21:33,  1.50it/s]
 48%|████▊     | 31252/65536 [5:21:58<6:07:46,  1.55it/s]
 48%|████▊     | 31253/65536 [5:21:59<6:04:56,  1.57it/s]
 48%|████▊     | 31254/65536 [5:21:59<5:58:39,  1.59it/s]
 48%|████▊     | 31255/65536 [5:22:00<6:04:42,  1.57it/s]
 48%|████▊     | 31256/65536 [5:22:01<5:58:21,  1.59it/s]
 48%|████▊     | 31257/65536 [5:22:01<5:50:43,  1.63it/s]
 48%|████▊     | 31258/65536 [5:22:02<5:55:20,  1.61it/s]
 48%|████▊     | 31259/65536 [5:22:03<5:52:57,  1.62it/s]
 48%|████▊     | 31260/65536 [5:22:03<5:45:01,  1.66it/s]
                                                         
{'loss': 1.6524, 'learning_rate': 5.721501163217828e-07, 'epoch': 1929.63}
+
 48%|████▊     | 31260/65536 [5:22:03<5:45:01,  1.66it/s]
 48%|████▊     | 31261/65536 [5:22:04<5:48:00,  1.64it/s]
 48%|████▊     | 31262/65536 [5:22:04<5:39:50,  1.68it/s]
 48%|████▊     | 31263/65536 [5:22:05<5:39:31,  1.68it/s]
 48%|████▊     | 31264/65536 [5:22:06<5:38:05,  1.69it/s]
 48%|████▊     | 31265/65536 [5:22:06<5:45:00,  1.66it/s]
 48%|████▊     | 31266/65536 [5:22:07<5:50:23,  1.63it/s]
 48%|████▊     | 31267/65536 [5:22:07<6:00:16,  1.59it/s]
 48%|████▊     | 31268/65536 [5:22:08<5:58:53,  1.59it/s]
 48%|████▊     | 31269/65536 [5:22:09<5:54:34,  1.61it/s]
 48%|████▊     | 31270/65536 [5:22:09<5:48:16,  1.64it/s]
 48%|████▊     | 31271/65536 [5:22:10<5:51:26,  1.62it/s]
 48%|████▊     | 31272/65536 [5:22:10<5:50:34,  1.63it/s]
 48%|████▊     | 31273/65536 [5:22:11<5:50:26,  1.63it/s]
 48%|████▊     | 31274/65536 [5:22:12<5:43:40,  1.66it/s]
 48%|████▊     | 31275/65536 [5:22:12<6:00:50,  1.58it/s]
 48%|████▊     | 31276/65536 [5:22:13<5:53:34,  1.61it/s]
 48%|████▊     | 31277/65536 [5:22:14<6:05:16,  1.56it/s]
 48%|████▊     | 31278/65536 [5:22:14<5:54:45,  1.61it/s]
 48%|████▊     | 31279/65536 [5:22:15<5:40:47,  1.68it/s]
 48%|████▊     | 31280/65536 [5:22:15<5:45:57,  1.65it/s]
                                                         
{'loss': 1.6835, 'learning_rate': 5.718746173625567e-07, 'epoch': 1930.86}
+
 48%|████▊     | 31280/65536 [5:22:15<5:45:57,  1.65it/s]
 48%|████▊     | 31281/65536 [5:22:16<5:49:24,  1.63it/s]
 48%|████▊     | 31282/65536 [5:22:17<5:46:51,  1.65it/s]
 48%|████▊     | 31283/65536 [5:22:17<5:58:23,  1.59it/s]
 48%|████▊     | 31284/65536 [5:22:18<5:51:48,  1.62it/s]
 48%|████▊     | 31285/65536 [5:22:19<5:50:02,  1.63it/s]
 48%|████▊     | 31286/65536 [5:22:19<5:45:14,  1.65it/s]
 48%|████▊     | 31287/65536 [5:22:20<5:50:29,  1.63it/s]
 48%|████▊     | 31288/65536 [5:22:20<5:50:26,  1.63it/s]
 48%|████▊     | 31289/65536 [5:22:21<5:47:05,  1.64it/s]
 48%|████▊     | 31290/65536 [5:22:22<5:57:13,  1.60it/s]
 48%|████▊     | 31291/65536 [5:22:22<5:54:13,  1.61it/s]
 48%|████▊     | 31292/65536 [5:22:23<5:48:02,  1.64it/s]
 48%|████▊     | 31293/65536 [5:22:23<5:47:05,  1.64it/s]
 48%|████▊     | 31294/65536 [5:22:24<5:41:48,  1.67it/s]
 48%|████▊     | 31295/65536 [5:22:25<5:44:59,  1.65it/s]
 48%|████▊     | 31296/65536 [5:22:25<6:00:29,  1.58it/s]
 48%|████▊     | 31297/65536 [5:22:26<5:58:31,  1.59it/s]
 48%|████▊     | 31298/65536 [5:22:27<6:09:09,  1.55it/s]
 48%|████▊     | 31299/65536 [5:22:27<6:15:24,  1.52it/s]
 48%|████▊     | 31300/65536 [5:22:28<6:05:43,  1.56it/s]
                                                         
{'loss': 1.7149, 'learning_rate': 5.715991184033304e-07, 'epoch': 1932.1}
+
 48%|████▊     | 31300/65536 [5:22:28<6:05:43,  1.56it/s]
 48%|████▊     | 31301/65536 [5:22:28<5:59:54,  1.59it/s]
 48%|████▊     | 31302/65536 [5:22:29<5:53:17,  1.61it/s]
 48%|████▊     | 31303/65536 [5:22:30<5:58:24,  1.59it/s]
 48%|████▊     | 31304/65536 [5:22:30<5:57:28,  1.60it/s]
 48%|████▊     | 31305/65536 [5:22:31<6:02:54,  1.57it/s]
 48%|████▊     | 31306/65536 [5:22:32<5:55:55,  1.60it/s]
 48%|████▊     | 31307/65536 [5:22:32<5:49:55,  1.63it/s]
 48%|████▊     | 31308/65536 [5:22:33<5:43:43,  1.66it/s]
 48%|████▊     | 31309/65536 [5:22:33<5:38:43,  1.68it/s]
 48%|████▊     | 31310/65536 [5:22:34<5:38:09,  1.69it/s]
 48%|████▊     | 31311/65536 [5:22:35<5:46:29,  1.65it/s]
 48%|████▊     | 31312/65536 [5:22:35<5:46:37,  1.65it/s]
 48%|████▊     | 31313/65536 [5:22:36<5:51:59,  1.62it/s]
 48%|████▊     | 31314/65536 [5:22:36<5:52:47,  1.62it/s]
 48%|████▊     | 31315/65536 [5:22:37<6:11:50,  1.53it/s]
 48%|████▊     | 31316/65536 [5:22:38<6:06:42,  1.56it/s]
 48%|████▊     | 31317/65536 [5:22:38<6:12:12,  1.53it/s]
 48%|████▊     | 31318/65536 [5:22:39<5:59:39,  1.59it/s]
 48%|████▊     | 31319/65536 [5:22:40<6:00:31,  1.58it/s]
 48%|████▊     | 31320/65536 [5:22:40<5:54:29,  1.61it/s]
                                                         
{'loss': 1.6778, 'learning_rate': 5.713236194441043e-07, 'epoch': 1933.33}
+
 48%|████▊     | 31320/65536 [5:22:40<5:54:29,  1.61it/s]
 48%|████▊     | 31321/65536 [5:22:41<5:55:50,  1.60it/s]
 48%|████▊     | 31322/65536 [5:22:41<5:46:29,  1.65it/s]
 48%|████▊     | 31323/65536 [5:22:42<5:43:32,  1.66it/s]
 48%|████▊     | 31324/65536 [5:22:43<5:46:45,  1.64it/s]
 48%|████▊     | 31325/65536 [5:22:43<5:41:16,  1.67it/s]
 48%|████▊     | 31326/65536 [5:22:44<5:48:47,  1.63it/s]
 48%|████▊     | 31327/65536 [5:22:45<5:53:43,  1.61it/s]
 48%|████▊     | 31328/65536 [5:22:45<5:46:57,  1.64it/s]
 48%|████▊     | 31329/65536 [5:22:46<5:52:23,  1.62it/s]
 48%|████▊     | 31330/65536 [5:22:46<5:49:22,  1.63it/s]
 48%|████▊     | 31331/65536 [5:22:47<5:59:45,  1.58it/s]
 48%|████▊     | 31332/65536 [5:22:48<6:01:00,  1.58it/s]
 48%|████▊     | 31333/65536 [5:22:48<5:53:40,  1.61it/s]
 48%|████▊     | 31334/65536 [5:22:49<5:48:39,  1.63it/s]
 48%|████▊     | 31335/65536 [5:22:50<5:52:04,  1.62it/s]
 48%|████▊     | 31336/65536 [5:22:50<5:50:24,  1.63it/s]
 48%|████▊     | 31337/65536 [5:22:51<5:43:23,  1.66it/s]
 48%|████▊     | 31338/65536 [5:22:51<5:59:44,  1.58it/s]
 48%|████▊     | 31339/65536 [5:22:52<5:58:44,  1.59it/s]
 48%|████▊     | 31340/65536 [5:22:53<5:54:39,  1.61it/s]
                                                         
{'loss': 1.676, 'learning_rate': 5.710481204848782e-07, 'epoch': 1934.57}
+
 48%|████▊     | 31340/65536 [5:22:53<5:54:39,  1.61it/s]
 48%|████▊     | 31341/65536 [5:22:53<5:52:30,  1.62it/s]
 48%|████▊     | 31342/65536 [5:22:54<5:45:05,  1.65it/s]
 48%|████▊     | 31343/65536 [5:22:54<5:53:37,  1.61it/s]
 48%|████▊     | 31344/65536 [5:22:55<5:45:01,  1.65it/s]
 48%|████▊     | 31345/65536 [5:22:56<5:48:24,  1.64it/s]
 48%|████▊     | 31346/65536 [5:22:56<5:50:51,  1.62it/s]
 48%|████▊     | 31347/65536 [5:22:57<5:45:39,  1.65it/s]
 48%|████▊     | 31348/65536 [5:22:58<5:56:16,  1.60it/s]
 48%|████▊     | 31349/65536 [5:22:58<5:49:16,  1.63it/s]
 48%|████▊     | 31350/65536 [5:22:59<5:44:52,  1.65it/s]
 48%|████▊     | 31351/65536 [5:22:59<5:47:55,  1.64it/s]
 48%|████▊     | 31352/65536 [5:23:00<5:44:11,  1.66it/s]
 48%|████▊     | 31353/65536 [5:23:00<5:39:24,  1.68it/s]
 48%|████▊     | 31354/65536 [5:23:01<5:46:44,  1.64it/s]
 48%|████▊     | 31355/65536 [5:23:02<5:43:44,  1.66it/s]
 48%|████▊     | 31356/65536 [5:23:02<5:43:24,  1.66it/s]
 48%|████▊     | 31357/65536 [5:23:03<5:42:16,  1.66it/s]
 48%|████▊     | 31358/65536 [5:23:04<5:39:53,  1.68it/s]
 48%|████▊     | 31359/65536 [5:23:04<6:03:52,  1.57it/s]
 48%|████▊     | 31360/65536 [5:23:05<6:03:32,  1.57it/s]
                                                         
{'loss': 1.6905, 'learning_rate': 5.70772621525652e-07, 'epoch': 1935.8}
+
 48%|████▊     | 31360/65536 [5:23:05<6:03:32,  1.57it/s]
 48%|████▊     | 31361/65536 [5:23:05<5:59:27,  1.58it/s]
 48%|████▊     | 31362/65536 [5:23:06<6:00:24,  1.58it/s]
 48%|████▊     | 31363/65536 [5:23:07<5:53:51,  1.61it/s]
 48%|████▊     | 31364/65536 [5:23:07<6:06:35,  1.55it/s]
 48%|████▊     | 31365/65536 [5:23:08<5:57:35,  1.59it/s]
 48%|████▊     | 31366/65536 [5:23:09<5:50:33,  1.62it/s]
 48%|████▊     | 31367/65536 [5:23:09<5:55:38,  1.60it/s]
 48%|████▊     | 31368/65536 [5:23:10<5:47:14,  1.64it/s]
 48%|████▊     | 31369/65536 [5:23:10<5:46:18,  1.64it/s]
 48%|████▊     | 31370/65536 [5:23:11<5:48:10,  1.64it/s]
 48%|████▊     | 31371/65536 [5:23:12<5:53:44,  1.61it/s]
 48%|████▊     | 31372/65536 [5:23:12<5:54:58,  1.60it/s]
 48%|████▊     | 31373/65536 [5:23:13<5:53:52,  1.61it/s]
 48%|████▊     | 31374/65536 [5:23:14<5:54:39,  1.61it/s]
 48%|████▊     | 31375/65536 [5:23:14<5:45:02,  1.65it/s]
 48%|████▊     | 31376/65536 [5:23:15<5:43:27,  1.66it/s]
 48%|████▊     | 31377/65536 [5:23:15<5:36:50,  1.69it/s]
 48%|████▊     | 31378/65536 [5:23:16<5:59:18,  1.58it/s]
 48%|████▊     | 31379/65536 [5:23:17<5:54:39,  1.61it/s]
 48%|████▊     | 31380/65536 [5:23:17<6:02:58,  1.57it/s]
                                                         
{'loss': 1.687, 'learning_rate': 5.704971225664258e-07, 'epoch': 1937.04}
+
 48%|████▊     | 31380/65536 [5:23:17<6:02:58,  1.57it/s]
 48%|████▊     | 31381/65536 [5:23:18<5:55:34,  1.60it/s]
 48%|████▊     | 31382/65536 [5:23:18<5:50:51,  1.62it/s]
 48%|████▊     | 31383/65536 [5:23:19<5:47:23,  1.64it/s]
 48%|████▊     | 31384/65536 [5:23:20<5:53:36,  1.61it/s]
 48%|████▊     | 31385/65536 [5:23:20<5:57:46,  1.59it/s]
 48%|████▊     | 31386/65536 [5:23:21<5:55:36,  1.60it/s]
 48%|████▊     | 31387/65536 [5:23:22<5:50:11,  1.63it/s]
 48%|████▊     | 31388/65536 [5:23:22<6:05:15,  1.56it/s]
 48%|████▊     | 31389/65536 [5:23:23<5:58:15,  1.59it/s]
 48%|████▊     | 31390/65536 [5:23:23<5:52:16,  1.62it/s]
 48%|████▊     | 31391/65536 [5:23:24<5:45:20,  1.65it/s]
 48%|████▊     | 31392/65536 [5:23:25<5:40:24,  1.67it/s]
 48%|████▊     | 31393/65536 [5:23:25<5:40:11,  1.67it/s]
 48%|████▊     | 31394/65536 [5:23:26<5:44:38,  1.65it/s]
 48%|████▊     | 31395/65536 [5:23:26<5:45:03,  1.65it/s]
 48%|████▊     | 31396/65536 [5:23:27<6:02:19,  1.57it/s]
 48%|████▊     | 31397/65536 [5:23:28<5:57:27,  1.59it/s]
 48%|████▊     | 31398/65536 [5:23:28<5:53:59,  1.61it/s]
 48%|████▊     | 31399/65536 [5:23:29<5:57:41,  1.59it/s]
 48%|████▊     | 31400/65536 [5:23:30<5:56:15,  1.60it/s]
                                                         
{'loss': 1.6852, 'learning_rate': 5.702216236071996e-07, 'epoch': 1938.27}
+
 48%|████▊     | 31400/65536 [5:23:30<5:56:15,  1.60it/s]
 48%|████▊     | 31401/65536 [5:23:30<5:56:57,  1.59it/s]
 48%|████▊     | 31402/65536 [5:23:31<5:53:58,  1.61it/s]
 48%|████▊     | 31403/65536 [5:23:32<5:54:21,  1.61it/s]
 48%|████▊     | 31404/65536 [5:23:32<5:51:50,  1.62it/s]
 48%|████▊     | 31405/65536 [5:23:33<5:54:21,  1.61it/s]
 48%|████▊     | 31406/65536 [5:23:33<5:47:32,  1.64it/s]
 48%|████▊     | 31407/65536 [5:23:34<5:50:23,  1.62it/s]
 48%|████▊     | 31408/65536 [5:23:35<5:42:07,  1.66it/s]
 48%|████▊     | 31409/65536 [5:23:35<5:39:17,  1.68it/s]
 48%|████▊     | 31410/65536 [5:23:36<5:48:47,  1.63it/s]
 48%|████▊     | 31411/65536 [5:23:36<5:57:44,  1.59it/s]
 48%|████▊     | 31412/65536 [5:23:37<6:03:10,  1.57it/s]
 48%|████▊     | 31413/65536 [5:23:38<6:02:04,  1.57it/s]
 48%|████▊     | 31414/65536 [5:23:38<6:03:05,  1.57it/s]
 48%|████▊     | 31415/65536 [5:23:39<5:57:57,  1.59it/s]
 48%|████▊     | 31416/65536 [5:23:40<5:50:37,  1.62it/s]
 48%|████▊     | 31417/65536 [5:23:40<5:45:34,  1.65it/s]
 48%|████▊     | 31418/65536 [5:23:41<5:45:03,  1.65it/s]
 48%|████▊     | 31419/65536 [5:23:41<5:42:06,  1.66it/s]
 48%|████▊     | 31420/65536 [5:23:42<5:43:20,  1.66it/s]
                                                         
{'loss': 1.6914, 'learning_rate': 5.699461246479735e-07, 'epoch': 1939.51}
+
 48%|████▊     | 31420/65536 [5:23:42<5:43:20,  1.66it/s]
 48%|████▊     | 31421/65536 [5:23:43<5:43:53,  1.65it/s]
 48%|████▊     | 31422/65536 [5:23:43<5:52:16,  1.61it/s]
 48%|████▊     | 31423/65536 [5:23:44<5:47:19,  1.64it/s]
 48%|████▊     | 31424/65536 [5:23:44<5:48:04,  1.63it/s]
 48%|████▊     | 31425/65536 [5:23:45<5:41:02,  1.67it/s]
 48%|████▊     | 31426/65536 [5:23:46<5:40:09,  1.67it/s]
 48%|████▊     | 31427/65536 [5:23:46<5:34:47,  1.70it/s]
 48%|████▊     | 31428/65536 [5:23:47<5:55:37,  1.60it/s]
 48%|████▊     | 31429/65536 [5:23:48<6:01:29,  1.57it/s]
 48%|████▊     | 31430/65536 [5:23:48<6:05:05,  1.56it/s]
 48%|████▊     | 31431/65536 [5:23:49<5:59:23,  1.58it/s]
 48%|████▊     | 31432/65536 [5:23:49<5:48:24,  1.63it/s]
 48%|████▊     | 31433/65536 [5:23:50<5:48:55,  1.63it/s]
 48%|████▊     | 31434/65536 [5:23:51<5:50:49,  1.62it/s]
 48%|████▊     | 31435/65536 [5:23:51<5:53:44,  1.61it/s]
 48%|████▊     | 31436/65536 [5:23:52<5:55:18,  1.60it/s]
 48%|████▊     | 31437/65536 [5:23:52<5:55:58,  1.60it/s]
 48%|████▊     | 31438/65536 [5:23:53<5:50:23,  1.62it/s]
 48%|████▊     | 31439/65536 [5:23:54<5:46:37,  1.64it/s]
 48%|████▊     | 31440/65536 [5:23:54<5:37:41,  1.68it/s]
                                                         
{'loss': 1.7071, 'learning_rate': 5.696706256887474e-07, 'epoch': 1940.74}
+
 48%|████▊     | 31440/65536 [5:23:54<5:37:41,  1.68it/s]
 48%|████▊     | 31441/65536 [5:23:55<5:38:11,  1.68it/s]
 48%|████▊     | 31442/65536 [5:23:55<5:44:27,  1.65it/s]
 48%|████▊     | 31443/65536 [5:23:56<5:42:07,  1.66it/s]
 48%|████▊     | 31444/65536 [5:23:57<5:40:43,  1.67it/s]
 48%|████▊     | 31445/65536 [5:23:57<5:52:16,  1.61it/s]
 48%|████▊     | 31446/65536 [5:23:58<5:41:07,  1.67it/s]
 48%|████▊     | 31447/65536 [5:23:58<5:44:06,  1.65it/s]
 48%|████▊     | 31448/65536 [5:23:59<5:38:53,  1.68it/s]
 48%|████▊     | 31449/65536 [5:24:00<5:46:31,  1.64it/s]
 48%|████▊     | 31450/65536 [5:24:00<5:46:55,  1.64it/s]
 48%|████▊     | 31451/65536 [5:24:01<5:52:33,  1.61it/s]
 48%|████▊     | 31452/65536 [5:24:02<5:57:06,  1.59it/s]
 48%|████▊     | 31453/65536 [5:24:02<6:03:06,  1.56it/s]
 48%|████▊     | 31454/65536 [5:24:03<5:57:09,  1.59it/s]
 48%|████▊     | 31455/65536 [5:24:04<6:04:23,  1.56it/s]
 48%|████▊     | 31456/65536 [5:24:04<5:51:45,  1.61it/s]
 48%|████▊     | 31457/65536 [5:24:05<5:59:41,  1.58it/s]
 48%|████▊     | 31458/65536 [5:24:05<5:54:18,  1.60it/s]
 48%|████▊     | 31459/65536 [5:24:06<5:52:10,  1.61it/s]
 48%|████▊     | 31460/65536 [5:24:07<5:50:50,  1.62it/s]
                                                         
{'loss': 1.6584, 'learning_rate': 5.693951267295212e-07, 'epoch': 1941.98}
+
 48%|████▊     | 31460/65536 [5:24:07<5:50:50,  1.62it/s]
 48%|████▊     | 31461/65536 [5:24:07<6:14:21,  1.52it/s]
 48%|████▊     | 31462/65536 [5:24:08<6:06:32,  1.55it/s]
 48%|████▊     | 31463/65536 [5:24:09<5:57:46,  1.59it/s]
 48%|████▊     | 31464/65536 [5:24:09<5:52:20,  1.61it/s]
 48%|████▊     | 31465/65536 [5:24:10<5:50:21,  1.62it/s]
 48%|████▊     | 31466/65536 [5:24:10<5:50:19,  1.62it/s]
 48%|████▊     | 31467/65536 [5:24:11<6:01:36,  1.57it/s]
 48%|████▊     | 31468/65536 [5:24:12<5:52:07,  1.61it/s]
 48%|████▊     | 31469/65536 [5:24:12<5:59:48,  1.58it/s]
 48%|████▊     | 31470/65536 [5:24:13<6:00:20,  1.58it/s]
 48%|████▊     | 31471/65536 [5:24:14<6:00:42,  1.57it/s]
 48%|████▊     | 31472/65536 [5:24:14<5:58:18,  1.58it/s]
 48%|████▊     | 31473/65536 [5:24:15<5:59:43,  1.58it/s]
 48%|████▊     | 31474/65536 [5:24:15<5:55:29,  1.60it/s]
 48%|████▊     | 31475/65536 [5:24:16<5:49:17,  1.63it/s]
 48%|████▊     | 31476/65536 [5:24:17<5:47:36,  1.63it/s]
 48%|████▊     | 31477/65536 [5:24:17<5:58:56,  1.58it/s]
 48%|████▊     | 31478/65536 [5:24:18<5:59:37,  1.58it/s]
 48%|████▊     | 31479/65536 [5:24:19<5:57:41,  1.59it/s]
 48%|████▊     | 31480/65536 [5:24:19<5:54:01,  1.60it/s]
                                                         
{'loss': 1.6533, 'learning_rate': 5.691196277702951e-07, 'epoch': 1943.21}
+
 48%|████▊     | 31480/65536 [5:24:19<5:54:01,  1.60it/s]
 48%|████▊     | 31481/65536 [5:24:20<5:56:46,  1.59it/s]
 48%|████▊     | 31482/65536 [5:24:20<5:56:43,  1.59it/s]
 48%|████▊     | 31483/65536 [5:24:21<5:50:21,  1.62it/s]
 48%|████▊     | 31484/65536 [5:24:22<5:42:21,  1.66it/s]
 48%|████▊     | 31485/65536 [5:24:22<5:42:51,  1.66it/s]
 48%|████▊     | 31486/65536 [5:24:23<5:42:13,  1.66it/s]
 48%|████▊     | 31487/65536 [5:24:23<5:46:17,  1.64it/s]
 48%|████▊     | 31488/65536 [5:24:24<5:49:18,  1.62it/s]
 48%|████▊     | 31489/65536 [5:24:25<5:39:11,  1.67it/s]
 48%|████▊     | 31490/65536 [5:24:25<5:41:44,  1.66it/s]
 48%|████▊     | 31491/65536 [5:24:26<5:38:57,  1.67it/s]
 48%|████▊     | 31492/65536 [5:24:27<5:46:53,  1.64it/s]
 48%|████▊     | 31493/65536 [5:24:27<6:05:20,  1.55it/s]
 48%|████▊     | 31494/65536 [5:24:28<5:57:14,  1.59it/s]
 48%|████▊     | 31495/65536 [5:24:28<5:51:34,  1.61it/s]
 48%|████▊     | 31496/65536 [5:24:29<5:50:14,  1.62it/s]
 48%|████▊     | 31497/65536 [5:24:30<5:59:33,  1.58it/s]
 48%|████▊     | 31498/65536 [5:24:30<6:02:31,  1.56it/s]
 48%|████▊     | 31499/65536 [5:24:31<6:01:56,  1.57it/s]
 48%|████▊     | 31500/65536 [5:24:32<6:01:53,  1.57it/s]
                                                         
{'loss': 1.646, 'learning_rate': 5.688441288110689e-07, 'epoch': 1944.44}
+
 48%|████▊     | 31500/65536 [5:24:32<6:01:53,  1.57it/s]
 48%|████▊     | 31501/65536 [5:24:32<6:03:09,  1.56it/s]
 48%|████▊     | 31502/65536 [5:24:33<5:53:12,  1.61it/s]
 48%|█���██▊     | 31503/65536 [5:24:33<5:49:07,  1.62it/s]
 48%|████▊     | 31504/65536 [5:24:34<5:52:54,  1.61it/s]
 48%|████▊     | 31505/65536 [5:24:35<5:49:59,  1.62it/s]
 48%|████▊     | 31506/65536 [5:24:35<5:44:23,  1.65it/s]
 48%|████▊     | 31507/65536 [5:24:36<5:42:16,  1.66it/s]
 48%|████▊     | 31508/65536 [5:24:36<5:41:16,  1.66it/s]
 48%|████▊     | 31509/65536 [5:24:37<5:39:40,  1.67it/s]
 48%|████▊     | 31510/65536 [5:24:38<6:00:11,  1.57it/s]
 48%|████▊     | 31511/65536 [5:24:38<5:53:57,  1.60it/s]
 48%|████▊     | 31512/65536 [5:24:39<5:48:03,  1.63it/s]
 48%|████▊     | 31513/65536 [5:24:40<5:45:40,  1.64it/s]
 48%|████▊     | 31514/65536 [5:24:40<5:57:07,  1.59it/s]
 48%|████▊     | 31515/65536 [5:24:41<5:58:35,  1.58it/s]
 48%|████▊     | 31516/65536 [5:24:41<5:50:44,  1.62it/s]
 48%|████▊     | 31517/65536 [5:24:42<5:57:23,  1.59it/s]
 48%|████▊     | 31518/65536 [5:24:43<5:51:40,  1.61it/s]
 48%|████▊     | 31519/65536 [5:24:43<5:56:43,  1.59it/s]
 48%|████▊     | 31520/65536 [5:24:44<5:51:58,  1.61it/s]
                                                         
{'loss': 1.6853, 'learning_rate': 5.685686298518428e-07, 'epoch': 1945.68}
+
 48%|████▊     | 31520/65536 [5:24:44<5:51:58,  1.61it/s]
 48%|████▊     | 31521/65536 [5:24:45<5:58:18,  1.58it/s]
 48%|████▊     | 31522/65536 [5:24:45<5:45:12,  1.64it/s]
 48%|████▊     | 31523/65536 [5:24:46<5:46:14,  1.64it/s]
 48%|████▊     | 31524/65536 [5:24:46<5:42:01,  1.66it/s]
 48%|████▊     | 31525/65536 [5:24:47<5:43:37,  1.65it/s]
 48%|████▊     | 31526/65536 [5:24:48<6:05:22,  1.55it/s]
 48%|████▊     | 31527/65536 [5:24:48<5:59:02,  1.58it/s]
 48%|████▊     | 31528/65536 [5:24:49<5:55:21,  1.60it/s]
 48%|████▊     | 31529/65536 [5:24:50<5:48:17,  1.63it/s]
 48%|████▊     | 31530/65536 [5:24:50<5:46:35,  1.64it/s]
 48%|████▊     | 31531/65536 [5:24:51<5:38:39,  1.67it/s]
 48%|████▊     | 31532/65536 [5:24:51<5:36:26,  1.68it/s]
 48%|████▊     | 31533/65536 [5:24:52<5:32:24,  1.70it/s]
 48%|████▊     | 31534/65536 [5:24:53<5:40:34,  1.66it/s]
 48%|████▊     | 31535/65536 [5:24:53<5:46:52,  1.63it/s]
 48%|████▊     | 31536/65536 [5:24:54<5:44:47,  1.64it/s]
 48%|████▊     | 31537/65536 [5:24:54<5:47:19,  1.63it/s]
 48%|████▊     | 31538/65536 [5:24:55<5:51:59,  1.61it/s]
 48%|████▊     | 31539/65536 [5:24:56<5:52:10,  1.61it/s]
 48%|████▊     | 31540/65536 [5:24:56<5:46:11,  1.64it/s]
                                                         
{'loss': 1.6666, 'learning_rate': 5.682931308926166e-07, 'epoch': 1946.91}
+
 48%|████▊     | 31540/65536 [5:24:56<5:46:11,  1.64it/s]
 48%|████▊     | 31541/65536 [5:24:57<5:50:08,  1.62it/s]
 48%|████▊     | 31542/65536 [5:24:58<5:59:43,  1.58it/s]
 48%|████▊     | 31543/65536 [5:24:58<5:51:07,  1.61it/s]
 48%|████▊     | 31544/65536 [5:24:59<5:48:14,  1.63it/s]
 48%|████▊     | 31545/65536 [5:24:59<5:45:15,  1.64it/s]
 48%|████▊     | 31546/65536 [5:25:00<5:46:34,  1.63it/s]
 48%|████▊     | 31547/65536 [5:25:01<5:48:15,  1.63it/s]
 48%|████▊     | 31548/65536 [5:25:01<5:58:17,  1.58it/s]
 48%|████▊     | 31549/65536 [5:25:02<5:54:59,  1.60it/s]
 48%|████▊     | 31550/65536 [5:25:02<5:59:18,  1.58it/s]
 48%|████▊     | 31551/65536 [5:25:03<5:58:14,  1.58it/s]
 48%|████▊     | 31552/65536 [5:25:04<5:42:28,  1.65it/s]
 48%|████▊     | 31553/65536 [5:25:04<5:50:51,  1.61it/s]
 48%|████▊     | 31554/65536 [5:25:05<5:44:15,  1.65it/s]
 48%|████▊     | 31555/65536 [5:25:06<5:45:03,  1.64it/s]
 48%|████▊     | 31556/65536 [5:25:06<5:49:03,  1.62it/s]
 48%|████▊     | 31557/65536 [5:25:07<5:47:22,  1.63it/s]
 48%|████▊     | 31558/65536 [5:25:07<5:56:30,  1.59it/s]
 48%|████▊     | 31559/65536 [5:25:08<5:50:28,  1.62it/s]
 48%|████▊     | 31560/65536 [5:25:09<5:48:47,  1.62it/s]
                                                         
{'loss': 1.6918, 'learning_rate': 5.680176319333905e-07, 'epoch': 1948.15}
+
 48%|████▊     | 31560/65536 [5:25:09<5:48:47,  1.62it/s]
 48%|████▊     | 31561/65536 [5:25:09<5:46:38,  1.63it/s]
 48%|████▊     | 31562/65536 [5:25:10<5:41:51,  1.66it/s]
 48%|████▊     | 31563/65536 [5:25:10<5:47:37,  1.63it/s]
 48%|████▊     | 31564/65536 [5:25:11<5:53:19,  1.60it/s]
 48%|████▊     | 31565/65536 [5:25:12<5:56:29,  1.59it/s]
 48%|████▊     | 31566/65536 [5:25:12<5:57:41,  1.58it/s]
 48%|████▊     | 31567/65536 [5:25:13<5:50:42,  1.61it/s]
 48%|████▊     | 31568/65536 [5:25:14<5:53:46,  1.60it/s]
 48%|████▊     | 31569/65536 [5:25:14<5:50:11,  1.62it/s]
 48%|████▊     | 31570/65536 [5:25:15<5:44:16,  1.64it/s]
 48%|████▊     | 31571/65536 [5:25:15<5:46:53,  1.63it/s]
 48%|████▊     | 31572/65536 [5:25:16<6:00:12,  1.57it/s]
 48%|████▊     | 31573/65536 [5:25:17<5:58:22,  1.58it/s]
 48%|████▊     | 31574/65536 [5:25:17<6:03:47,  1.56it/s]
 48%|████▊     | 31575/65536 [5:25:18<5:54:29,  1.60it/s]
 48%|████▊     | 31576/65536 [5:25:19<5:49:48,  1.62it/s]
 48%|████▊     | 31577/65536 [5:25:19<5:44:25,  1.64it/s]
 48%|████▊     | 31578/65536 [5:25:20<5:46:24,  1.63it/s]
 48%|████▊     | 31579/65536 [5:25:20<5:44:07,  1.64it/s]
 48%|████▊     | 31580/65536 [5:25:21<5:42:22,  1.65it/s]
                                                         
{'loss': 1.6719, 'learning_rate': 5.677421329741643e-07, 'epoch': 1949.38}
+
 48%|████▊     | 31580/65536 [5:25:21<5:42:22,  1.65it/s]
 48%|████▊     | 31581/65536 [5:25:22<5:37:34,  1.68it/s]
 48%|████▊     | 31582/65536 [5:25:22<5:50:28,  1.61it/s]
 48%|████▊     | 31583/65536 [5:25:23<5:47:19,  1.63it/s]
 48%|████▊     | 31584/65536 [5:25:24<5:58:38,  1.58it/s]
 48%|████▊     | 31585/65536 [5:25:24<5:48:07,  1.63it/s]
 48%|████▊     | 31586/65536 [5:25:25<5:45:02,  1.64it/s]
 48%|████▊     | 31587/65536 [5:25:25<5:45:13,  1.64it/s]
 48%|████▊     | 31588/65536 [5:25:26<5:43:03,  1.65it/s]
 48%|████▊     | 31589/65536 [5:25:27<5:45:57,  1.64it/s]
 48%|████▊     | 31590/65536 [5:25:27<5:43:41,  1.65it/s]
 48%|████▊     | 31591/65536 [5:25:28<5:51:55,  1.61it/s]
 48%|████▊     | 31592/65536 [5:25:28<5:49:12,  1.62it/s]
 48%|████▊     | 31593/65536 [5:25:29<5:54:06,  1.60it/s]
 48%|████▊     | 31594/65536 [5:25:30<5:58:01,  1.58it/s]
 48%|████▊     | 31595/65536 [5:25:30<5:59:14,  1.57it/s]
 48%|████▊     | 31596/65536 [5:25:31<5:45:26,  1.64it/s]
 48%|████▊     | 31597/65536 [5:25:31<5:49:07,  1.62it/s]
 48%|████▊     | 31598/65536 [5:25:32<5:47:35,  1.63it/s]
 48%|████▊     | 31599/65536 [5:25:33<5:51:44,  1.61it/s]
 48%|████▊     | 31600/65536 [5:25:33<5:52:54,  1.60it/s]
                                                         
{'loss': 1.6708, 'learning_rate': 5.674666340149381e-07, 'epoch': 1950.62}
+
 48%|████▊     | 31600/65536 [5:25:33<5:52:54,  1.60it/s]
 48%|████▊     | 31601/65536 [5:25:34<5:54:51,  1.59it/s]
 48%|████▊     | 31602/65536 [5:25:35<5:48:50,  1.62it/s]
 48%|████▊     | 31603/65536 [5:25:35<5:43:44,  1.65it/s]
 48%|████▊     | 31604/65536 [5:25:36<5:42:04,  1.65it/s]
 48%|████▊     | 31605/65536 [5:25:36<5:43:21,  1.65it/s]
 48%|████▊     | 31606/65536 [5:25:37<5:45:59,  1.63it/s]
 48%|████▊     | 31607/65536 [5:25:38<6:01:39,  1.56it/s]
 48%|████▊     | 31608/65536 [5:25:38<5:51:45,  1.61it/s]
 48%|████▊     | 31609/65536 [5:25:39<5:42:49,  1.65it/s]
 48%|████▊     | 31610/65536 [5:25:39<5:40:16,  1.66it/s]
 48%|████▊     | 31611/65536 [5:25:40<5:37:16,  1.68it/s]
 48%|████▊     | 31612/65536 [5:25:41<5:44:10,  1.64it/s]
 48%|████▊     | 31613/65536 [5:25:41<5:44:27,  1.64it/s]
 48%|████▊     | 31614/65536 [5:25:42<5:42:36,  1.65it/s]
 48%|████▊     | 31615/65536 [5:25:43<5:45:17,  1.64it/s]
 48%|████▊     | 31616/65536 [5:25:43<5:45:22,  1.64it/s]
 48%|████▊     | 31617/65536 [5:25:44<5:55:58,  1.59it/s]
 48%|████▊     | 31618/65536 [5:25:44<5:56:34,  1.59it/s]
 48%|████▊     | 31619/65536 [5:25:45<5:48:52,  1.62it/s]
 48%|████▊     | 31620/65536 [5:25:46<5:46:41,  1.63it/s]
                                                         
{'loss': 1.6474, 'learning_rate': 5.671911350557119e-07, 'epoch': 1951.85}
+
 48%|████▊     | 31620/65536 [5:25:46<5:46:41,  1.63it/s]
 48%|████▊     | 31621/65536 [5:25:46<5:49:28,  1.62it/s]
 48%|████▊     | 31622/65536 [5:25:47<5:49:09,  1.62it/s]
 48%|████▊     | 31623/65536 [5:25:48<6:13:03,  1.52it/s]
 48%|████▊     | 31624/65536 [5:25:48<6:04:16,  1.55it/s]
 48%|████▊     | 31625/65536 [5:25:49<5:59:03,  1.57it/s]
 48%|████▊     | 31626/65536 [5:25:49<5:49:21,  1.62it/s]
 48%|████▊     | 31627/65536 [5:25:50<5:50:13,  1.61it/s]
 48%|████▊     | 31628/65536 [5:25:51<5:44:22,  1.64it/s]
 48%|████▊     | 31629/65536 [5:25:51<5:48:29,  1.62it/s]
 48%|████▊     | 31630/65536 [5:25:52<5:43:38,  1.64it/s]
 48%|████▊     | 31631/65536 [5:25:52<5:45:07,  1.64it/s]
 48%|████▊     | 31632/65536 [5:25:53<5:46:33,  1.63it/s]
 48%|████▊     | 31633/65536 [5:25:54<5:48:34,  1.62it/s]
 48%|████▊     | 31634/65536 [5:25:54<5:47:52,  1.62it/s]
 48%|████▊     | 31635/65536 [5:25:55<5:43:28,  1.64it/s]
 48%|████▊     | 31636/65536 [5:25:56<5:44:55,  1.64it/s]
 48%|████▊     | 31637/65536 [5:25:56<5:46:11,  1.63it/s]
 48%|████▊     | 31638/65536 [5:25:57<5:44:53,  1.64it/s]
 48%|████▊     | 31639/65536 [5:25:57<6:04:06,  1.55it/s]
 48%|████▊     | 31640/65536 [5:25:58<5:56:08,  1.59it/s]
                                                         
{'loss': 1.6686, 'learning_rate': 5.669156360964857e-07, 'epoch': 1953.09}
+
 48%|████▊     | 31640/65536 [5:25:58<5:56:08,  1.59it/s]
 48%|████▊     | 31641/65536 [5:25:59<5:48:28,  1.62it/s]
 48%|████▊     | 31642/65536 [5:25:59<5:49:09,  1.62it/s]
 48%|████▊     | 31643/65536 [5:26:00<5:44:30,  1.64it/s]
 48%|████▊     | 31644/65536 [5:26:01<5:49:53,  1.61it/s]
 48%|████▊     | 31645/65536 [5:26:01<5:56:27,  1.58it/s]
 48%|████▊     | 31646/65536 [5:26:02<5:51:05,  1.61it/s]
 48%|████▊     | 31647/65536 [5:26:02<5:57:01,  1.58it/s]
 48%|████▊     | 31648/65536 [5:26:03<5:53:55,  1.60it/s]
 48%|████▊     | 31649/65536 [5:26:04<5:46:00,  1.63it/s]
 48%|████▊     | 31650/65536 [5:26:04<5:43:05,  1.65it/s]
 48%|████▊     | 31651/65536 [5:26:05<5:48:49,  1.62it/s]
 48%|████▊     | 31652/65536 [5:26:06<5:53:30,  1.60it/s]
 48%|████▊     | 31653/65536 [5:26:06<5:42:01,  1.65it/s]
 48%|████▊     | 31654/65536 [5:26:07<5:36:47,  1.68it/s]
 48%|████▊     | 31655/65536 [5:26:07<5:47:00,  1.63it/s]
 48%|████▊     | 31656/65536 [5:26:08<5:44:57,  1.64it/s]
 48%|████▊     | 31657/65536 [5:26:09<5:43:52,  1.64it/s]
 48%|████▊     | 31658/65536 [5:26:09<5:41:42,  1.65it/s]
 48%|████▊     | 31659/65536 [5:26:10<5:45:17,  1.64it/s]
 48%|████▊     | 31660/65536 [5:26:10<5:48:10,  1.62it/s]
                                                         
{'loss': 1.6534, 'learning_rate': 5.666401371372596e-07, 'epoch': 1954.32}
+
 48%|████▊     | 31660/65536 [5:26:10<5:48:10,  1.62it/s]
 48%|████▊     | 31661/65536 [5:26:11<5:44:45,  1.64it/s]
 48%|████▊     | 31662/65536 [5:26:12<5:38:50,  1.67it/s]
 48%|████▊     | 31663/65536 [5:26:12<5:43:35,  1.64it/s]
 48%|████▊     | 31664/65536 [5:26:13<5:52:14,  1.60it/s]
 48%|████▊     | 31665/65536 [5:26:13<5:57:45,  1.58it/s]
 48%|████▊     | 31666/65536 [5:26:14<5:53:54,  1.60it/s]
 48%|████▊     | 31667/65536 [5:26:15<5:42:35,  1.65it/s]
 48%|████▊     | 31668/65536 [5:26:15<5:49:05,  1.62it/s]
 48%|████▊     | 31669/65536 [5:26:16<5:48:11,  1.62it/s]
 48%|████▊     | 31670/65536 [5:26:16<5:36:39,  1.68it/s]
 48%|████▊     | 31671/65536 [5:26:17<5:37:11,  1.67it/s]
 48%|████▊     | 31672/65536 [5:26:18<5:50:45,  1.61it/s]
 48%|████▊     | 31673/65536 [5:26:18<5:43:08,  1.64it/s]
 48%|████▊     | 31674/65536 [5:26:19<5:49:59,  1.61it/s]
 48%|████▊     | 31675/65536 [5:26:20<5:44:43,  1.64it/s]
 48%|████▊     | 31676/65536 [5:26:20<5:53:30,  1.60it/s]
 48%|████▊     | 31677/65536 [5:26:21<5:52:39,  1.60it/s]
 48%|████▊     | 31678/65536 [5:26:21<5:56:42,  1.58it/s]
 48%|████▊     | 31679/65536 [5:26:22<5:57:37,  1.58it/s]
 48%|████▊     | 31680/65536 [5:26:23<5:52:48,  1.60it/s]
                                                         
{'loss': 1.6554, 'learning_rate': 5.663646381780334e-07, 'epoch': 1955.56}
+
 48%|████▊     | 31680/65536 [5:26:23<5:52:48,  1.60it/s]
 48%|████▊     | 31681/65536 [5:26:23<5:49:14,  1.62it/s]
 48%|████▊     | 31682/65536 [5:26:24<5:44:40,  1.64it/s]
 48%|████▊     | 31683/65536 [5:26:25<5:46:56,  1.63it/s]
 48%|████▊     | 31684/65536 [5:26:25<5:40:07,  1.66it/s]
 48%|████▊     | 31685/65536 [5:26:26<5:41:37,  1.65it/s]
 48%|████▊     | 31686/65536 [5:26:26<5:51:11,  1.61it/s]
 48%|████▊     | 31687/65536 [5:26:27<5:47:47,  1.62it/s]
 48%|████▊     | 31688/65536 [5:26:28<6:01:24,  1.56it/s]
 48%|████▊     | 31689/65536 [5:26:28<5:54:20,  1.59it/s]
 48%|████▊     | 31690/65536 [5:26:29<5:50:58,  1.61it/s]
 48%|████▊     | 31691/65536 [5:26:30<5:56:32,  1.58it/s]
 48%|████▊     | 31692/65536 [5:26:30<5:58:18,  1.57it/s]
 48%|████▊     | 31693/65536 [5:26:31<5:50:44,  1.61it/s]
 48%|████▊     | 31694/65536 [5:26:31<5:49:05,  1.62it/s]
 48%|████▊     | 31695/65536 [5:26:32<5:43:42,  1.64it/s]
 48%|████▊     | 31696/65536 [5:26:33<5:47:09,  1.62it/s]
 48%|████▊     | 31697/65536 [5:26:33<5:45:10,  1.63it/s]
 48%|████▊     | 31698/65536 [5:26:34<5:49:27,  1.61it/s]
 48%|████▊     | 31699/65536 [5:26:35<5:56:10,  1.58it/s]
 48%|████▊     | 31700/65536 [5:26:35<5:58:57,  1.57it/s]
                                                         
{'loss': 1.6135, 'learning_rate': 5.660891392188073e-07, 'epoch': 1956.79}
+
 48%|████▊     | 31700/65536 [5:26:35<5:58:57,  1.57it/s]
 48%|████▊     | 31701/65536 [5:26:36<6:04:23,  1.55it/s]
 48%|████▊     | 31702/65536 [5:26:36<5:56:21,  1.58it/s]
 48%|████▊     | 31703/65536 [5:26:37<5:50:09,  1.61it/s]
 48%|████▊     | 31704/65536 [5:26:38<5:52:27,  1.60it/s]
 48%|████▊     | 31705/65536 [5:26:38<5:50:51,  1.61it/s]
 48%|████▊     | 31706/65536 [5:26:39<5:45:18,  1.63it/s]
 48%|████▊     | 31707/65536 [5:26:39<5:40:52,  1.65it/s]
 48%|████▊     | 31708/65536 [5:26:40<5:44:41,  1.64it/s]
 48%|████▊     | 31709/65536 [5:26:41<5:41:52,  1.65it/s]
 48%|████▊     | 31710/65536 [5:26:41<5:38:07,  1.67it/s]
 48%|████▊     | 31711/65536 [5:26:42<5:37:49,  1.67it/s]
 48%|████▊     | 31712/65536 [5:26:42<5:41:29,  1.65it/s]
 48%|████▊     | 31713/65536 [5:26:43<5:39:05,  1.66it/s]
 48%|████▊     | 31714/65536 [5:26:44<5:35:03,  1.68it/s]
 48%|████▊     | 31715/65536 [5:26:44<5:46:56,  1.62it/s]
 48%|████▊     | 31716/65536 [5:26:45<5:42:38,  1.65it/s]
 48%|████▊     | 31717/65536 [5:26:46<5:44:59,  1.63it/s]
 48%|████▊     | 31718/65536 [5:26:46<5:54:48,  1.59it/s]
 48%|████▊     | 31719/65536 [5:26:47<6:04:03,  1.55it/s]
 48%|████▊     | 31720/65536 [5:26:48<6:07:36,  1.53it/s]
                                                         
{'loss': 1.6707, 'learning_rate': 5.658136402595812e-07, 'epoch': 1958.02}
+
 48%|████▊     | 31720/65536 [5:26:48<6:07:36,  1.53it/s]
 48%|████▊     | 31721/65536 [5:26:48<6:04:07,  1.55it/s]
 48%|████▊     | 31722/65536 [5:26:49<6:06:39,  1.54it/s]
 48%|████▊     | 31723/65536 [5:26:49<5:57:48,  1.58it/s]
 48%|████▊     | 31724/65536 [5:26:50<5:54:49,  1.59it/s]
 48%|████▊     | 31725/65536 [5:26:51<5:51:31,  1.60it/s]
 48%|████▊     | 31726/65536 [5:26:51<5:48:07,  1.62it/s]
 48%|████▊     | 31727/65536 [5:26:52<5:44:27,  1.64it/s]
 48%|████▊     | 31728/65536 [5:26:52<5:43:03,  1.64it/s]
 48%|████▊     | 31729/65536 [5:26:53<5:42:18,  1.65it/s]
 48%|████▊     | 31730/65536 [5:26:54<5:47:15,  1.62it/s]
 48%|████▊     | 31731/65536 [5:26:54<5:51:07,  1.60it/s]
 48%|████▊     | 31732/65536 [5:26:55<5:50:26,  1.61it/s]
 48%|████▊     | 31733/65536 [5:26:56<5:50:21,  1.61it/s]
 48%|████▊     | 31734/65536 [5:26:56<5:52:33,  1.60it/s]
 48%|████▊     | 31735/65536 [5:26:57<5:55:04,  1.59it/s]
 48%|████▊     | 31736/65536 [5:26:58<6:10:41,  1.52it/s]
 48%|████▊     | 31737/65536 [5:26:58<6:00:23,  1.56it/s]
 48%|████▊     | 31738/65536 [5:26:59<5:54:06,  1.59it/s]
 48%|████▊     | 31739/65536 [5:26:59<6:03:57,  1.55it/s]
 48%|████▊     | 31740/65536 [5:27:00<5:52:45,  1.60it/s]
                                                         
{'loss': 1.6569, 'learning_rate': 5.65538141300355e-07, 'epoch': 1959.26}
+
 48%|████▊     | 31740/65536 [5:27:00<5:52:45,  1.60it/s]
 48%|████▊     | 31741/65536 [5:27:01<5:58:00,  1.57it/s]
 48%|████▊     | 31742/65536 [5:27:01<5:55:07,  1.59it/s]
 48%|████▊     | 31743/65536 [5:27:02<5:51:51,  1.60it/s]
 48%|████▊     | 31744/65536 [5:27:03<5:48:59,  1.61it/s]
 48%|████▊     | 31745/65536 [5:27:03<5:42:36,  1.64it/s]
 48%|████▊     | 31746/65536 [5:27:04<5:49:20,  1.61it/s]
 48%|████▊     | 31747/65536 [5:27:04<5:39:36,  1.66it/s]
 48%|████▊     | 31748/65536 [5:27:05<5:43:17,  1.64it/s]
 48%|████▊     | 31749/65536 [5:27:06<5:43:05,  1.64it/s]
 48%|████▊     | 31750/65536 [5:27:06<5:49:20,  1.61it/s]
 48%|████▊     | 31751/65536 [5:27:07<5:48:37,  1.62it/s]
 48%|████▊     | 31752/65536 [5:27:07<5:48:33,  1.62it/s]
 48%|████▊     | 31753/65536 [5:27:08<5:59:37,  1.57it/s]
 48%|████▊     | 31754/65536 [5:27:09<5:53:37,  1.59it/s]
 48%|████▊     | 31755/65536 [5:27:09<5:49:28,  1.61it/s]
 48%|████▊     | 31756/65536 [5:27:10<5:40:17,  1.65it/s]
 48%|████▊     | 31757/65536 [5:27:11<5:46:54,  1.62it/s]
 48%|████▊     | 31758/65536 [5:27:11<5:40:05,  1.66it/s]
 48%|████▊     | 31759/65536 [5:27:12<5:39:39,  1.66it/s]
 48%|████▊     | 31760/65536 [5:27:12<5:34:37,  1.68it/s]
                                                         
{'loss': 1.7037, 'learning_rate': 5.652626423411289e-07, 'epoch': 1960.49}
+
 48%|████▊     | 31760/65536 [5:27:12<5:34:37,  1.68it/s]
 48%|████▊     | 31761/65536 [5:27:13<5:41:36,  1.65it/s]
 48%|████▊     | 31762/65536 [5:27:14<5:51:00,  1.60it/s]
 48%|████▊     | 31763/65536 [5:27:14<5:49:27,  1.61it/s]
 48%|████▊     | 31764/65536 [5:27:15<5:55:04,  1.59it/s]
 48%|████▊     | 31765/65536 [5:27:16<5:57:01,  1.58it/s]
 48%|████▊     | 31766/65536 [5:27:16<5:51:03,  1.60it/s]
 48%|████▊     | 31767/65536 [5:27:17<5:57:59,  1.57it/s]
 48%|████▊     | 31768/65536 [5:27:17<5:45:46,  1.63it/s]
 48%|████▊     | 31769/65536 [5:27:18<5:55:41,  1.58it/s]
 48%|████▊     | 31770/65536 [5:27:19<6:07:05,  1.53it/s]
 48%|████▊     | 31771/65536 [5:27:19<6:03:45,  1.55it/s]
 48%|████▊     | 31772/65536 [5:27:20<5:57:06,  1.58it/s]
 48%|████▊     | 31773/65536 [5:27:21<6:04:19,  1.54it/s]
 48%|████▊     | 31774/65536 [5:27:21<5:54:58,  1.59it/s]
 48%|████▊     | 31775/65536 [5:27:22<5:48:06,  1.62it/s]
 48%|████▊     | 31776/65536 [5:27:22<5:48:02,  1.62it/s]
 48%|████▊     | 31777/65536 [5:27:23<5:47:05,  1.62it/s]
 48%|████▊     | 31778/65536 [5:27:24<5:48:50,  1.61it/s]
 48%|████▊     | 31779/65536 [5:27:24<5:40:27,  1.65it/s]
 48%|████▊     | 31780/65536 [5:27:25<5:41:51,  1.65it/s]
                                                         
{'loss': 1.6324, 'learning_rate': 5.649871433819027e-07, 'epoch': 1961.73}
+
 48%|████▊     | 31780/65536 [5:27:25<5:41:51,  1.65it/s]
 48%|████▊     | 31781/65536 [5:27:25<5:44:58,  1.63it/s]
 48%|████▊     | 31782/65536 [5:27:26<5:40:06,  1.65it/s]
 48%|████▊     | 31783/65536 [5:27:27<5:41:37,  1.65it/s]
 48%|████▊     | 31784/65536 [5:27:27<5:38:36,  1.66it/s]
 49%|████▊     | 31785/65536 [5:27:28<5:58:34,  1.57it/s]
 49%|████▊     | 31786/65536 [5:27:29<5:57:09,  1.57it/s]
 49%|████▊     | 31787/65536 [5:27:29<6:08:38,  1.53it/s]
 49%|████▊     | 31788/65536 [5:27:30<6:08:33,  1.53it/s]
 49%|████▊     | 31789/65536 [5:27:31<6:03:16,  1.55it/s]
 49%|████▊     | 31790/65536 [5:27:31<5:52:52,  1.59it/s]
 49%|████▊     | 31791/65536 [5:27:32<5:47:28,  1.62it/s]
 49%|████▊     | 31792/65536 [5:27:32<5:43:10,  1.64it/s]
 49%|████▊     | 31793/65536 [5:27:33<5:46:20,  1.62it/s]
 49%|████▊     | 31794/65536 [5:27:34<5:38:41,  1.66it/s]
 49%|████▊     | 31795/65536 [5:27:34<5:38:03,  1.66it/s]
 49%|████▊     | 31796/65536 [5:27:35<5:39:28,  1.66it/s]
 49%|████▊     | 31797/65536 [5:27:35<5:35:04,  1.68it/s]
 49%|████▊     | 31798/65536 [5:27:36<5:27:56,  1.71it/s]
 49%|████▊     | 31799/65536 [5:27:37<5:29:30,  1.71it/s]
 49%|████▊     | 31800/65536 [5:27:37<5:36:07,  1.67it/s]
                                                         
{'loss': 1.666, 'learning_rate': 5.647116444226766e-07, 'epoch': 1962.96}
+
 49%|████▊     | 31800/65536 [5:27:37<5:36:07,  1.67it/s]
 49%|████▊     | 31801/65536 [5:27:38<6:02:07,  1.55it/s]
 49%|████▊     | 31802/65536 [5:27:39<6:03:12,  1.55it/s]
 49%|████▊     | 31803/65536 [5:27:39<5:52:29,  1.59it/s]
 49%|████▊     | 31804/65536 [5:27:40<5:57:09,  1.57it/s]
 49%|████▊     | 31805/65536 [5:27:40<5:48:12,  1.61it/s]
 49%|████▊     | 31806/65536 [5:27:41<5:47:15,  1.62it/s]
 49%|████▊     | 31807/65536 [5:27:42<5:47:22,  1.62it/s]
 49%|████▊     | 31808/65536 [5:27:42<5:47:25,  1.62it/s]
 49%|████▊     | 31809/65536 [5:27:43<5:42:31,  1.64it/s]
 49%|████▊     | 31810/65536 [5:27:43<5:44:23,  1.63it/s]
 49%|████▊     | 31811/65536 [5:27:44<5:39:56,  1.65it/s]
 49%|████▊     | 31812/65536 [5:27:45<5:38:24,  1.66it/s]
 49%|████▊     | 31813/65536 [5:27:45<5:37:47,  1.66it/s]
 49%|████▊     | 31814/65536 [5:27:46<5:45:36,  1.63it/s]
 49%|████▊     | 31815/65536 [5:27:46<5:44:05,  1.63it/s]
 49%|████▊     | 31816/65536 [5:27:47<5:53:10,  1.59it/s]
 49%|████▊     | 31817/65536 [5:27:48<6:08:15,  1.53it/s]
 49%|████▊     | 31818/65536 [5:27:48<6:02:02,  1.55it/s]
 49%|████▊     | 31819/65536 [5:27:49<6:08:14,  1.53it/s]
 49%|████▊     | 31820/65536 [5:27:50<6:07:45,  1.53it/s]
                                                         
{'loss': 1.651, 'learning_rate': 5.644361454634505e-07, 'epoch': 1964.2}
+
 49%|████▊     | 31820/65536 [5:27:50<6:07:45,  1.53it/s]
 49%|████▊     | 31821/65536 [5:27:50<5:58:44,  1.57it/s]
 49%|████▊     | 31822/65536 [5:27:51<5:51:21,  1.60it/s]
 49%|████▊     | 31823/65536 [5:27:52<5:45:22,  1.63it/s]
 49%|████▊     | 31824/65536 [5:27:52<5:49:35,  1.61it/s]
 49%|████▊     | 31825/65536 [5:27:53<5:42:38,  1.64it/s]
 49%|████▊     | 31826/65536 [5:27:53<5:45:49,  1.62it/s]
 49%|████▊     | 31827/65536 [5:27:54<5:44:45,  1.63it/s]
 49%|████▊     | 31828/65536 [5:27:55<5:48:33,  1.61it/s]
 49%|████▊     | 31829/65536 [5:27:55<5:44:42,  1.63it/s]
 49%|████▊     | 31830/65536 [5:27:56<5:42:26,  1.64it/s]
 49%|████▊     | 31831/65536 [5:27:56<5:42:38,  1.64it/s]
 49%|████▊     | 31832/65536 [5:27:57<5:51:00,  1.60it/s]
 49%|████▊     | 31833/65536 [5:27:58<5:49:43,  1.61it/s]
 49%|████▊     | 31834/65536 [5:27:58<6:00:55,  1.56it/s]
 49%|████▊     | 31835/65536 [5:27:59<5:56:01,  1.58it/s]
 49%|████▊     | 31836/65536 [5:28:00<6:02:16,  1.55it/s]
 49%|████▊     | 31837/65536 [5:28:00<6:01:16,  1.55it/s]
 49%|████▊     | 31838/65536 [5:28:01<5:57:08,  1.57it/s]
 49%|████▊     | 31839/65536 [5:28:02<5:54:48,  1.58it/s]
 49%|████▊     | 31840/65536 [5:28:02<5:55:47,  1.58it/s]
                                                         
{'loss': 1.6548, 'learning_rate': 5.641606465042243e-07, 'epoch': 1965.43}
+
 49%|████▊     | 31840/65536 [5:28:02<5:55:47,  1.58it/s]
 49%|████▊     | 31841/65536 [5:28:03<5:55:46,  1.58it/s]
 49%|████▊     | 31842/65536 [5:28:03<5:48:38,  1.61it/s]
 49%|████▊     | 31843/65536 [5:28:04<5:46:25,  1.62it/s]
 49%|████▊     | 31844/65536 [5:28:05<5:44:35,  1.63it/s]
 49%|████▊     | 31845/65536 [5:28:05<5:47:01,  1.62it/s]
 49%|████▊     | 31846/65536 [5:28:06<5:47:48,  1.61it/s]
 49%|████▊     | 31847/65536 [5:28:07<5:46:23,  1.62it/s]
 49%|████▊     | 31848/65536 [5:28:07<5:46:42,  1.62it/s]
 49%|████▊     | 31849/65536 [5:28:08<5:42:02,  1.64it/s]
 49%|████▊     | 31850/65536 [5:28:08<5:56:27,  1.58it/s]
 49%|████▊     | 31851/65536 [5:28:09<6:00:46,  1.56it/s]
 49%|████▊     | 31852/65536 [5:28:10<5:56:35,  1.57it/s]
 49%|████▊     | 31853/65536 [5:28:10<5:53:23,  1.59it/s]
 49%|████▊     | 31854/65536 [5:28:11<6:05:54,  1.53it/s]
 49%|████▊     | 31855/65536 [5:28:12<6:01:55,  1.55it/s]
 49%|████▊     | 31856/65536 [5:28:12<5:43:37,  1.63it/s]
 49%|████▊     | 31857/65536 [5:28:13<5:47:40,  1.61it/s]
 49%|████▊     | 31858/65536 [5:28:13<5:46:32,  1.62it/s]
 49%|████▊     | 31859/65536 [5:28:14<5:45:56,  1.62it/s]
 49%|████▊     | 31860/65536 [5:28:15<5:44:57,  1.63it/s]
                                                         
{'loss': 1.6696, 'learning_rate': 5.638851475449982e-07, 'epoch': 1966.67}
+
 49%|████▊     | 31860/65536 [5:28:15<5:44:57,  1.63it/s]
 49%|████▊     | 31861/65536 [5:28:15<5:40:46,  1.65it/s]
 49%|████▊     | 31862/65536 [5:28:16<5:43:59,  1.63it/s]
 49%|████▊     | 31863/65536 [5:28:16<5:40:34,  1.65it/s]
 49%|████▊     | 31864/65536 [5:28:17<5:39:42,  1.65it/s]
 49%|████▊     | 31865/65536 [5:28:18<5:34:16,  1.68it/s]
 49%|████▊     | 31866/65536 [5:28:18<5:41:40,  1.64it/s]
 49%|████▊     | 31867/65536 [5:28:19<5:45:12,  1.63it/s]
 49%|████▊     | 31868/65536 [5:28:20<5:47:53,  1.61it/s]
 49%|████▊     | 31869/65536 [5:28:20<5:55:54,  1.58it/s]
 49%|████▊     | 31870/65536 [5:28:21<5:48:36,  1.61it/s]
 49%|████▊     | 31871/65536 [5:28:21<5:35:24,  1.67it/s]
 49%|████▊     | 31872/65536 [5:28:22<5:37:04,  1.66it/s]
 49%|████▊     | 31873/65536 [5:28:23<5:44:40,  1.63it/s]
 49%|████▊     | 31874/65536 [5:28:23<5:42:48,  1.64it/s]
 49%|████▊     | 31875/65536 [5:28:24<5:37:28,  1.66it/s]
 49%|████▊     | 31876/65536 [5:28:24<5:36:59,  1.66it/s]
 49%|████▊     | 31877/65536 [5:28:25<5:32:36,  1.69it/s]
 49%|████▊     | 31878/65536 [5:28:26<5:30:14,  1.70it/s]
 49%|████▊     | 31879/65536 [5:28:26<5:48:33,  1.61it/s]
 49%|████▊     | 31880/65536 [5:28:27<5:45:53,  1.62it/s]
                                                         
{'loss': 1.7076, 'learning_rate': 5.636096485857721e-07, 'epoch': 1967.9}
+
 49%|████▊     | 31880/65536 [5:28:27<5:45:53,  1.62it/s]
 49%|████▊     | 31881/65536 [5:28:27<5:45:00,  1.63it/s]
 49%|████▊     | 31882/65536 [5:28:28<5:55:22,  1.58it/s]
 49%|████▊     | 31883/65536 [5:28:29<5:53:42,  1.59it/s]
 49%|████▊     | 31884/65536 [5:28:29<5:47:29,  1.61it/s]
 49%|████▊     | 31885/65536 [5:28:30<5:41:55,  1.64it/s]
 49%|████▊     | 31886/65536 [5:28:31<5:50:05,  1.60it/s]
 49%|████▊     | 31887/65536 [5:28:31<5:49:04,  1.61it/s]
 49%|████▊     | 31888/65536 [5:28:32<5:52:53,  1.59it/s]
 49%|████▊     | 31889/65536 [5:28:32<5:41:40,  1.64it/s]
 49%|████▊     | 31890/65536 [5:28:33<5:42:12,  1.64it/s]
 49%|████▊     | 31891/65536 [5:28:34<5:43:42,  1.63it/s]
 49%|████▊     | 31892/65536 [5:28:34<5:44:16,  1.63it/s]
 49%|████▊     | 31893/65536 [5:28:35<5:39:33,  1.65it/s]
 49%|████▊     | 31894/65536 [5:28:35<5:33:03,  1.68it/s]
 49%|████▊     | 31895/65536 [5:28:36<5:36:20,  1.67it/s]
 49%|████▊     | 31896/65536 [5:28:37<5:36:59,  1.66it/s]
 49%|████▊     | 31897/65536 [5:28:37<5:46:36,  1.62it/s]
 49%|████▊     | 31898/65536 [5:28:38<6:00:32,  1.56it/s]
 49%|████▊     | 31899/65536 [5:28:39<5:52:16,  1.59it/s]
 49%|████▊     | 31900/65536 [5:28:39<5:42:09,  1.64it/s]
                                                         
{'loss': 1.702, 'learning_rate': 5.633341496265459e-07, 'epoch': 1969.14}
+
 49%|████▊     | 31900/65536 [5:28:39<5:42:09,  1.64it/s]
 49%|████▊     | 31901/65536 [5:28:40<5:50:09,  1.60it/s]
 49%|████▊     | 31902/65536 [5:28:40<5:45:16,  1.62it/s]
 49%|████▊     | 31903/65536 [5:28:41<5:49:23,  1.60it/s]
 49%|████▊     | 31904/65536 [5:28:42<5:44:58,  1.62it/s]
 49%|████▊     | 31905/65536 [5:28:42<5:43:13,  1.63it/s]
 49%|████▊     | 31906/65536 [5:28:43<5:40:07,  1.65it/s]
 49%|████▊     | 31907/65536 [5:28:43<5:34:32,  1.68it/s]
 49%|████▊     | 31908/65536 [5:28:44<5:44:40,  1.63it/s]
 49%|████▊     | 31909/65536 [5:28:45<5:47:07,  1.61it/s]
 49%|████▊     | 31910/65536 [5:28:45<5:44:26,  1.63it/s]
 49%|████▊     | 31911/65536 [5:28:46<5:45:35,  1.62it/s]
 49%|████▊     | 31912/65536 [5:28:47<5:47:13,  1.61it/s]
 49%|████▊     | 31913/65536 [5:28:47<5:52:12,  1.59it/s]
 49%|████▊     | 31914/65536 [5:28:48<5:43:51,  1.63it/s]
 49%|████▊     | 31915/65536 [5:28:48<5:53:26,  1.59it/s]
 49%|████▊     | 31916/65536 [5:28:49<5:55:38,  1.58it/s]
 49%|████▊     | 31917/65536 [5:28:50<6:01:26,  1.55it/s]
 49%|████▊     | 31918/65536 [5:28:50<5:56:21,  1.57it/s]
 49%|████▊     | 31919/65536 [5:28:51<5:48:56,  1.61it/s]
 49%|████▊     | 31920/65536 [5:28:52<5:46:55,  1.61it/s]
                                                         
{'loss': 1.6458, 'learning_rate': 5.630586506673195e-07, 'epoch': 1970.37}
+
 49%|████▊     | 31920/65536 [5:28:52<5:46:55,  1.61it/s]
 49%|████▊     | 31921/65536 [5:28:52<5:51:47,  1.59it/s]
 49%|████▊     | 31922/65536 [5:28:53<5:52:17,  1.59it/s]
 49%|████▊     | 31923/65536 [5:28:53<5:43:09,  1.63it/s]
 49%|████▊     | 31924/65536 [5:28:54<5:39:04,  1.65it/s]
 49%|████▊     | 31925/65536 [5:28:55<5:36:31,  1.66it/s]
 49%|████▊     | 31926/65536 [5:28:55<5:41:05,  1.64it/s]
 49%|████▊     | 31927/65536 [5:28:56<5:42:01,  1.64it/s]
 49%|████▊     | 31928/65536 [5:28:56<5:41:19,  1.64it/s]
 49%|████▊     | 31929/65536 [5:28:57<5:32:15,  1.69it/s]
 49%|████▊     | 31930/65536 [5:28:58<5:35:58,  1.67it/s]
 49%|████▊     | 31931/65536 [5:28:58<5:49:21,  1.60it/s]
 49%|████▊     | 31932/65536 [5:28:59<5:51:50,  1.59it/s]
 49%|████▊     | 31933/65536 [5:29:00<5:54:55,  1.58it/s]
 49%|████▊     | 31934/65536 [5:29:00<5:52:03,  1.59it/s]
 49%|████▊     | 31935/65536 [5:29:01<5:51:00,  1.60it/s]
 49%|████▊     | 31936/65536 [5:29:01<5:47:34,  1.61it/s]
 49%|████▊     | 31937/65536 [5:29:02<5:43:23,  1.63it/s]
 49%|████▊     | 31938/65536 [5:29:03<5:44:39,  1.62it/s]
 49%|████▊     | 31939/65536 [5:29:03<5:50:34,  1.60it/s]
 49%|████▊     | 31940/65536 [5:29:04<5:44:32,  1.63it/s]
                                                         
{'loss': 1.6834, 'learning_rate': 5.627831517080934e-07, 'epoch': 1971.6}
+
 49%|████▊     | 31940/65536 [5:29:04<5:44:32,  1.63it/s]
 49%|████▊     | 31941/65536 [5:29:05<5:43:54,  1.63it/s]
 49%|████▊     | 31942/65536 [5:29:05<5:44:02,  1.63it/s]
 49%|████▊     | 31943/65536 [5:29:06<5:44:11,  1.63it/s]
 49%|████▊     | 31944/65536 [5:29:06<5:40:09,  1.65it/s]
 49%|████▊     | 31945/65536 [5:29:07<5:52:10,  1.59it/s]
 49%|████▊     | 31946/65536 [5:29:08<5:49:21,  1.60it/s]
 49%|████▊     | 31947/65536 [5:29:08<6:04:13,  1.54it/s]
 49%|████▊     | 31948/65536 [5:29:09<6:03:47,  1.54it/s]
 49%|████▉     | 31949/65536 [5:29:10<5:51:22,  1.59it/s]
 49%|████▉     | 31950/65536 [5:29:10<5:56:08,  1.57it/s]
 49%|████▉     | 31951/65536 [5:29:11<5:49:23,  1.60it/s]
 49%|████▉     | 31952/65536 [5:29:11<5:51:07,  1.59it/s]
 49%|████▉     | 31953/65536 [5:29:12<5:41:11,  1.64it/s]
 49%|████▉     | 31954/65536 [5:29:13<5:40:08,  1.65it/s]
 49%|████▉     | 31955/65536 [5:29:13<5:50:29,  1.60it/s]
 49%|████▉     | 31956/65536 [5:29:14<6:02:16,  1.54it/s]
 49%|████▉     | 31957/65536 [5:29:15<5:54:40,  1.58it/s]
 49%|████▉     | 31958/65536 [5:29:15<5:51:04,  1.59it/s]
 49%|████▉     | 31959/65536 [5:29:16<5:49:25,  1.60it/s]
 49%|████▉     | 31960/65536 [5:29:16<5:50:36,  1.60it/s]
                                                         
{'loss': 1.6328, 'learning_rate': 5.625076527488673e-07, 'epoch': 1972.84}
+
 49%|████▉     | 31960/65536 [5:29:16<5:50:36,  1.60it/s]
 49%|████▉     | 31961/65536 [5:29:17<5:46:48,  1.61it/s]
 49%|████▉     | 31962/65536 [5:29:18<5:44:43,  1.62it/s]
 49%|████▉     | 31963/65536 [5:29:18<5:52:36,  1.59it/s]
 49%|████▉     | 31964/65536 [5:29:19<5:48:12,  1.61it/s]
 49%|████▉     | 31965/65536 [5:29:20<5:45:14,  1.62it/s]
 49%|████▉     | 31966/65536 [5:29:20<5:36:23,  1.66it/s]
 49%|████▉     | 31967/65536 [5:29:21<5:37:45,  1.66it/s]
 49%|████▉     | 31968/65536 [5:29:21<5:39:10,  1.65it/s]
 49%|████▉     | 31969/65536 [5:29:22<5:44:44,  1.62it/s]
 49%|████▉     | 31970/65536 [5:29:23<5:52:29,  1.59it/s]
 49%|████▉     | 31971/65536 [5:29:23<5:47:56,  1.61it/s]
 49%|████▉     | 31972/65536 [5:29:24<5:47:21,  1.61it/s]
 49%|████▉     | 31973/65536 [5:29:25<5:50:40,  1.60it/s]
 49%|████▉     | 31974/65536 [5:29:25<5:50:58,  1.59it/s]
 49%|████▉     | 31975/65536 [5:29:26<5:51:15,  1.59it/s]
 49%|████▉     | 31976/65536 [5:29:26<5:45:17,  1.62it/s]
 49%|████▉     | 31977/65536 [5:29:27<5:51:56,  1.59it/s]
 49%|████▉     | 31978/65536 [5:29:28<5:44:52,  1.62it/s]
 49%|████▉     | 31979/65536 [5:29:28<5:55:14,  1.57it/s]
 49%|████▉     | 31980/65536 [5:29:29<5:52:08,  1.59it/s]
                                                         
{'loss': 1.6697, 'learning_rate': 5.622321537896412e-07, 'epoch': 1974.07}
+
 49%|████▉     | 31980/65536 [5:29:29<5:52:08,  1.59it/s]
 49%|████▉     | 31981/65536 [5:29:30<5:57:29,  1.56it/s]
 49%|████▉     | 31982/65536 [5:29:30<5:48:38,  1.60it/s]
 49%|████▉     | 31983/65536 [5:29:31<5:43:13,  1.63it/s]
 49%|████▉     | 31984/65536 [5:29:31<5:39:04,  1.65it/s]
 49%|████▉     | 31985/65536 [5:29:32<5:50:19,  1.60it/s]
 49%|████▉     | 31986/65536 [5:29:33<5:42:05,  1.63it/s]
 49%|████▉     | 31987/65536 [5:29:33<5:41:08,  1.64it/s]
 49%|████▉     | 31988/65536 [5:29:34<5:44:47,  1.62it/s]
 49%|████▉     | 31989/65536 [5:29:34<5:39:41,  1.65it/s]
 49%|████▉     | 31990/65536 [5:29:35<5:33:32,  1.68it/s]
 49%|████▉     | 31991/65536 [5:29:36<5:36:14,  1.66it/s]
 49%|████▉     | 31992/65536 [5:29:36<5:47:58,  1.61it/s]
 49%|████▉     | 31993/65536 [5:29:37<5:45:08,  1.62it/s]
 49%|████▉     | 31994/65536 [5:29:37<5:45:32,  1.62it/s]
 49%|████▉     | 31995/65536 [5:29:38<5:40:44,  1.64it/s]
 49%|████▉     | 31996/65536 [5:29:39<5:53:36,  1.58it/s]
 49%|████▉     | 31997/65536 [5:29:39<5:53:10,  1.58it/s]
 49%|████▉     | 31998/65536 [5:29:40<5:43:49,  1.63it/s]
 49%|████▉     | 31999/65536 [5:29:41<5:59:12,  1.56it/s]
 49%|████▉     | 32000/65536 [5:29:41<5:50:50,  1.59it/s]
                                                         
{'loss': 1.6517, 'learning_rate': 5.61956654830415e-07, 'epoch': 1975.31}
+
 49%|████▉     | 32000/65536 [5:29:41<5:50:50,  1.59it/s]
 49%|████▉     | 32001/65536 [5:29:42<5:57:20,  1.56it/s]
 49%|████▉     | 32002/65536 [5:29:43<5:56:23,  1.57it/s]
 49%|████▉     | 32003/65536 [5:29:43<5:59:03,  1.56it/s]
 49%|████▉     | 32004/65536 [5:29:44<5:54:00,  1.58it/s]
 49%|████▉     | 32005/65536 [5:29:44<5:50:26,  1.59it/s]
 49%|████▉     | 32006/65536 [5:29:45<5:40:44,  1.64it/s]
 49%|████▉     | 32007/65536 [5:29:46<5:41:36,  1.64it/s]
 49%|████▉     | 32008/65536 [5:29:46<5:35:15,  1.67it/s]
 49%|████▉     | 32009/65536 [5:29:47<5:36:50,  1.66it/s]
 49%|████▉     | 32010/65536 [5:29:47<5:37:21,  1.66it/s]
 49%|████▉     | 32011/65536 [5:29:48<5:33:24,  1.68it/s]
 49%|████▉     | 32012/65536 [5:29:49<5:46:03,  1.61it/s]
 49%|████▉     | 32013/65536 [5:29:49<5:44:04,  1.62it/s]
 49%|████▉     | 32014/65536 [5:29:50<5:41:14,  1.64it/s]
 49%|████▉     | 32015/65536 [5:29:51<5:43:16,  1.63it/s]
 49%|████▉     | 32016/65536 [5:29:51<5:47:34,  1.61it/s]
 49%|████▉     | 32017/65536 [5:29:52<5:40:17,  1.64it/s]
 49%|████▉     | 32018/65536 [5:29:52<5:41:51,  1.63it/s]
 49%|████▉     | 32019/65536 [5:29:53<5:38:43,  1.65it/s]
 49%|████▉     | 32020/65536 [5:29:54<5:34:33,  1.67it/s]
                                                         
{'loss': 1.6757, 'learning_rate': 5.616811558711889e-07, 'epoch': 1976.54}
+
 49%|████▉     | 32020/65536 [5:29:54<5:34:33,  1.67it/s]
 49%|████▉     | 32021/65536 [5:29:54<5:45:40,  1.62it/s]
 49%|████▉     | 32022/65536 [5:29:55<5:41:32,  1.64it/s]
 49%|████▉     | 32023/65536 [5:29:55<5:42:08,  1.63it/s]
 49%|████▉     | 32024/65536 [5:29:56<5:42:55,  1.63it/s]
 49%|████▉     | 32025/65536 [5:29:57<5:45:42,  1.62it/s]
 49%|████▉     | 32026/65536 [5:29:57<5:50:32,  1.59it/s]
 49%|████▉     | 32027/65536 [5:29:58<5:41:16,  1.64it/s]
 49%|████▉     | 32028/65536 [5:29:59<5:53:52,  1.58it/s]
 49%|████▉     | 32029/65536 [5:29:59<5:50:14,  1.59it/s]
 49%|████▉     | 32030/65536 [5:30:00<5:55:26,  1.57it/s]
 49%|████▉     | 32031/65536 [5:30:00<5:44:20,  1.62it/s]
 49%|████▉     | 32032/65536 [5:30:01<5:37:37,  1.65it/s]
 49%|████▉     | 32033/65536 [5:30:02<5:40:33,  1.64it/s]
 49%|████▉     | 32034/65536 [5:30:02<5:39:24,  1.65it/s]
 49%|████▉     | 32035/65536 [5:30:03<5:34:00,  1.67it/s]
 49%|████▉     | 32036/65536 [5:30:03<5:36:11,  1.66it/s]
 49%|████▉     | 32037/65536 [5:30:04<5:33:28,  1.67it/s]
 49%|████▉     | 32038/65536 [5:30:05<5:31:02,  1.69it/s]
 49%|████▉     | 32039/65536 [5:30:05<5:39:22,  1.64it/s]
 49%|████▉     | 32040/65536 [5:30:06<5:47:53,  1.60it/s]
                                                         
{'loss': 1.6457, 'learning_rate': 5.614056569119628e-07, 'epoch': 1977.78}
+
 49%|████▉     | 32040/65536 [5:30:06<5:47:53,  1.60it/s]
 49%|████▉     | 32041/65536 [5:30:06<5:52:30,  1.58it/s]
 49%|████▉     | 32042/65536 [5:30:07<5:44:00,  1.62it/s]
 49%|████▉     | 32043/65536 [5:30:08<5:46:56,  1.61it/s]
 49%|████▉     | 32044/65536 [5:30:08<5:57:40,  1.56it/s]
 49%|████▉     | 32045/65536 [5:30:09<5:53:19,  1.58it/s]
 49%|████▉     | 32046/65536 [5:30:10<5:58:11,  1.56it/s]
 49%|████▉     | 32047/65536 [5:30:10<5:44:48,  1.62it/s]
 49%|████▉     | 32048/65536 [5:30:11<5:45:03,  1.62it/s]
 49%|████▉     | 32049/65536 [5:30:12<5:58:42,  1.56it/s]
 49%|████▉     | 32050/65536 [5:30:12<5:51:44,  1.59it/s]
 49%|████▉     | 32051/65536 [5:30:13<5:48:14,  1.60it/s]
 49%|████▉     | 32052/65536 [5:30:13<5:53:37,  1.58it/s]
 49%|████▉     | 32053/65536 [5:30:14<6:06:59,  1.52it/s]
 49%|████▉     | 32054/65536 [5:30:15<6:05:31,  1.53it/s]
 49%|████▉     | 32055/65536 [5:30:15<5:56:49,  1.56it/s]
 49%|████▉     | 32056/65536 [5:30:16<5:45:23,  1.62it/s]
 49%|████▉     | 32057/65536 [5:30:17<5:48:22,  1.60it/s]
 49%|████▉     | 32058/65536 [5:30:17<5:59:11,  1.55it/s]
 49%|████▉     | 32059/65536 [5:30:18<5:52:36,  1.58it/s]
 49%|████▉     | 32060/65536 [5:30:19<6:03:56,  1.53it/s]
                                                         
{'loss': 1.6244, 'learning_rate': 5.611301579527367e-07, 'epoch': 1979.01}
+
 49%|████▉     | 32060/65536 [5:30:19<6:03:56,  1.53it/s]
 49%|████▉     | 32061/65536 [5:30:19<6:07:48,  1.52it/s]
 49%|████▉     | 32062/65536 [5:30:20<5:58:01,  1.56it/s]
 49%|████▉     | 32063/65536 [5:30:21<5:59:04,  1.55it/s]
 49%|████▉     | 32064/65536 [5:30:21<5:52:10,  1.58it/s]
 49%|████▉     | 32065/65536 [5:30:22<5:49:06,  1.60it/s]
 49%|████▉     | 32066/65536 [5:30:22<5:42:58,  1.63it/s]
 49%|████▉     | 32067/65536 [5:30:23<5:36:01,  1.66it/s]
 49%|████▉     | 32068/65536 [5:30:24<5:47:45,  1.60it/s]
 49%|████▉     | 32069/65536 [5:30:24<5:44:48,  1.62it/s]
 49%|████▉     | 32070/65536 [5:30:25<5:50:57,  1.59it/s]
 49%|████▉     | 32071/65536 [5:30:25<5:52:34,  1.58it/s]
 49%|████▉     | 32072/65536 [5:30:26<5:54:22,  1.57it/s]
 49%|████▉     | 32073/65536 [5:30:27<5:42:57,  1.63it/s]
 49%|████▉     | 32074/65536 [5:30:27<5:39:56,  1.64it/s]
 49%|████▉     | 32075/65536 [5:30:28<5:36:26,  1.66it/s]
 49%|████▉     | 32076/65536 [5:30:28<5:36:30,  1.66it/s]
 49%|████▉     | 32077/65536 [5:30:29<5:43:13,  1.62it/s]
 49%|████▉     | 32078/65536 [5:30:30<5:49:35,  1.60it/s]
 49%|████▉     | 32079/65536 [5:30:30<5:52:40,  1.58it/s]
 49%|████▉     | 32080/65536 [5:30:31<5:46:56,  1.61it/s]
                                                         
{'loss': 1.6646, 'learning_rate': 5.608546589935105e-07, 'epoch': 1980.25}
+
 49%|████▉     | 32080/65536 [5:30:31<5:46:56,  1.61it/s]
 49%|████▉     | 32081/65536 [5:30:32<5:43:00,  1.63it/s]
 49%|████▉     | 32082/65536 [5:30:32<5:42:11,  1.63it/s]
 49%|████▉     | 32083/65536 [5:30:33<5:44:04,  1.62it/s]
 49%|████▉     | 32084/65536 [5:30:33<5:42:19,  1.63it/s]
 49%|████▉     | 32085/65536 [5:30:34<5:41:02,  1.63it/s]
 49%|████▉     | 32086/65536 [5:30:35<5:48:58,  1.60it/s]
 49%|████▉     | 32087/65536 [5:30:35<5:38:55,  1.64it/s]
 49%|████▉     | 32088/65536 [5:30:36<5:37:45,  1.65it/s]
 49%|████▉     | 32089/65536 [5:30:37<5:42:44,  1.63it/s]
 49%|████▉     | 32090/65536 [5:30:37<5:45:09,  1.61it/s]
 49%|████▉     | 32091/65536 [5:30:38<5:39:15,  1.64it/s]
 49%|████▉     | 32092/65536 [5:30:38<5:42:01,  1.63it/s]
 49%|████▉     | 32093/65536 [5:30:39<5:58:46,  1.55it/s]
 49%|████▉     | 32094/65536 [5:30:40<5:57:36,  1.56it/s]
 49%|████▉     | 32095/65536 [5:30:40<5:51:27,  1.59it/s]
 49%|████▉     | 32096/65536 [5:30:41<5:47:05,  1.61it/s]
 49%|████▉     | 32097/65536 [5:30:42<5:44:03,  1.62it/s]
 49%|████▉     | 32098/65536 [5:30:42<5:50:16,  1.59it/s]
 49%|████▉     | 32099/65536 [5:30:43<5:44:04,  1.62it/s]
 49%|████▉     | 32100/65536 [5:30:43<5:46:12,  1.61it/s]
                                                         
{'loss': 1.6297, 'learning_rate': 5.605791600342844e-07, 'epoch': 1981.48}
+
 49%|████▉     | 32100/65536 [5:30:43<5:46:12,  1.61it/s]
 49%|████▉     | 32101/65536 [5:30:44<5:41:44,  1.63it/s]
 49%|████▉     | 32102/65536 [5:30:45<5:34:47,  1.66it/s]
 49%|████▉     | 32103/65536 [5:30:45<5:38:26,  1.65it/s]
 49%|████▉     | 32104/65536 [5:30:46<5:55:28,  1.57it/s]
 49%|████▉     | 32105/65536 [5:30:47<5:57:15,  1.56it/s]
 49%|████▉     | 32106/65536 [5:30:47<5:51:05,  1.59it/s]
 49%|████▉     | 32107/65536 [5:30:48<5:44:59,  1.61it/s]
 49%|████▉     | 32108/65536 [5:30:48<5:39:26,  1.64it/s]
 49%|████▉     | 32109/65536 [5:30:49<5:49:03,  1.60it/s]
 49%|████▉     | 32110/65536 [5:30:50<5:46:54,  1.61it/s]
 49%|████▉     | 32111/65536 [5:30:50<5:40:31,  1.64it/s]
 49%|████▉     | 32112/65536 [5:30:51<5:39:37,  1.64it/s]
 49%|████▉     | 32113/65536 [5:30:51<5:49:14,  1.60it/s]
 49%|████▉     | 32114/65536 [5:30:52<5:49:21,  1.59it/s]
 49%|████▉     | 32115/65536 [5:30:53<5:46:01,  1.61it/s]
 49%|████▉     | 32116/65536 [5:30:53<5:37:12,  1.65it/s]
 49%|████▉     | 32117/65536 [5:30:54<5:42:45,  1.63it/s]
 49%|████▉     | 32118/65536 [5:30:55<5:42:57,  1.62it/s]
 49%|████▉     | 32119/65536 [5:30:55<5:44:00,  1.62it/s]
 49%|████▉     | 32120/65536 [5:30:56<5:43:01,  1.62it/s]
                                                         
{'loss': 1.6971, 'learning_rate': 5.603036610750581e-07, 'epoch': 1982.72}
+
 49%|████▉     | 32120/65536 [5:30:56<5:43:01,  1.62it/s]
 49%|████▉     | 32121/65536 [5:30:56<5:48:20,  1.60it/s]
 49%|████▉     | 32122/65536 [5:30:57<5:42:55,  1.62it/s]
 49%|████▉     | 32123/65536 [5:30:58<5:53:35,  1.57it/s]
 49%|████▉     | 32124/65536 [5:30:58<5:48:58,  1.60it/s]
 49%|████▉     | 32125/65536 [5:30:59<5:51:04,  1.59it/s]
 49%|████▉     | 32126/65536 [5:31:00<5:44:23,  1.62it/s]
 49%|████▉     | 32127/65536 [5:31:00<6:00:29,  1.54it/s]
 49%|████▉     | 32128/65536 [5:31:01<5:58:32,  1.55it/s]
 49%|████▉     | 32129/65536 [5:31:01<5:49:16,  1.59it/s]
 49%|████▉     | 32130/65536 [5:31:02<5:42:41,  1.62it/s]
 49%|████▉     | 32131/65536 [5:31:03<5:40:16,  1.64it/s]
 49%|████▉     | 32132/65536 [5:31:03<5:30:40,  1.68it/s]
 49%|████▉     | 32133/65536 [5:31:04<5:28:11,  1.70it/s]
 49%|████▉     | 32134/65536 [5:31:04<5:33:52,  1.67it/s]
 49%|████▉     | 32135/65536 [5:31:05<5:33:45,  1.67it/s]
 49%|████▉     | 32136/65536 [5:31:06<5:39:23,  1.64it/s]
 49%|████▉     | 32137/65536 [5:31:06<5:49:39,  1.59it/s]
 49%|████▉     | 32138/65536 [5:31:07<5:44:27,  1.62it/s]
 49%|████▉     | 32139/65536 [5:31:08<5:46:22,  1.61it/s]
 49%|████▉     | 32140/65536 [5:31:08<5:39:07,  1.64it/s]
                                                         
{'loss': 1.6461, 'learning_rate': 5.60028162115832e-07, 'epoch': 1983.95}
+
 49%|████▉     | 32140/65536 [5:31:08<5:39:07,  1.64it/s]
 49%|████▉     | 32141/65536 [5:31:09<5:53:11,  1.58it/s]
 49%|████▉     | 32142/65536 [5:31:09<5:44:21,  1.62it/s]
 49%|████▉     | 32143/65536 [5:31:10<5:42:09,  1.63it/s]
 49%|████▉     | 32144/65536 [5:31:11<5:40:10,  1.64it/s]
 49%|████▉     | 32145/65536 [5:31:11<5:42:21,  1.63it/s]
 49%|████▉     | 32146/65536 [5:31:12<5:44:38,  1.61it/s]
 49%|████▉     | 32147/65536 [5:31:13<5:53:07,  1.58it/s]
 49%|████▉     | 32148/65536 [5:31:13<5:47:26,  1.60it/s]
 49%|████▉     | 32149/65536 [5:31:14<5:38:00,  1.65it/s]
 49%|████▉     | 32150/65536 [5:31:14<5:34:36,  1.66it/s]
 49%|████▉     | 32151/65536 [5:31:15<5:32:31,  1.67it/s]
 49%|████▉     | 32152/65536 [5:31:15<5:36:54,  1.65it/s]
 49%|████▉     | 32153/65536 [5:31:16<5:43:15,  1.62it/s]
 49%|████▉     | 32154/65536 [5:31:17<5:55:21,  1.57it/s]
 49%|████▉     | 32155/65536 [5:31:17<5:52:14,  1.58it/s]
 49%|████▉     | 32156/65536 [5:31:18<5:48:39,  1.60it/s]
 49%|████▉     | 32157/65536 [5:31:19<5:41:36,  1.63it/s]
 49%|████▉     | 32158/65536 [5:31:19<6:08:02,  1.51it/s]
 49%|████▉     | 32159/65536 [5:31:20<6:00:09,  1.54it/s]
 49%|████▉     | 32160/65536 [5:31:21<6:00:05,  1.54it/s]
                                                         
{'loss': 1.6862, 'learning_rate': 5.597526631566058e-07, 'epoch': 1985.19}
+
 49%|████▉     | 32160/65536 [5:31:21<6:00:05,  1.54it/s]
 49%|████▉     | 32161/65536 [5:31:21<5:50:54,  1.59it/s]
 49%|████▉     | 32162/65536 [5:31:22<5:49:48,  1.59it/s]
 49%|████▉     | 32163/65536 [5:31:22<5:41:36,  1.63it/s]
 49%|████▉     | 32164/65536 [5:31:23<5:37:08,  1.65it/s]
 49%|████▉     | 32165/65536 [5:31:24<5:37:22,  1.65it/s]
 49%|████▉     | 32166/65536 [5:31:24<5:30:04,  1.68it/s]
 49%|████▉     | 32167/65536 [5:31:25<5:31:53,  1.68it/s]
 49%|████▉     | 32168/65536 [5:31:25<5:41:13,  1.63it/s]
 49%|████▉     | 32169/65536 [5:31:26<5:42:14,  1.62it/s]
 49%|████▉     | 32170/65536 [5:31:27<5:39:25,  1.64it/s]
 49%|████▉     | 32171/65536 [5:31:27<5:38:17,  1.64it/s]
 49%|████▉     | 32172/65536 [5:31:28<5:39:13,  1.64it/s]
 49%|████▉     | 32173/65536 [5:31:29<5:49:22,  1.59it/s]
 49%|████▉     | 32174/65536 [5:31:29<6:00:23,  1.54it/s]
 49%|████▉     | 32175/65536 [5:31:30<5:51:21,  1.58it/s]
 49%|████▉     | 32176/65536 [5:31:30<5:37:41,  1.65it/s]
 49%|████▉     | 32177/65536 [5:31:31<5:37:41,  1.65it/s]
 49%|████▉     | 32178/65536 [5:31:32<5:37:50,  1.65it/s]
 49%|████▉     | 32179/65536 [5:31:32<5:43:44,  1.62it/s]
 49%|████▉     | 32180/65536 [5:31:33<5:35:57,  1.65it/s]
                                                         
{'loss': 1.6764, 'learning_rate': 5.594771641973796e-07, 'epoch': 1986.42}
+
 49%|████▉     | 32180/65536 [5:31:33<5:35:57,  1.65it/s]
 49%|████▉     | 32181/65536 [5:31:34<5:43:59,  1.62it/s]
 49%|████▉     | 32182/65536 [5:31:34<5:41:48,  1.63it/s]
 49%|████▉     | 32183/65536 [5:31:35<5:39:56,  1.64it/s]
 49%|████▉     | 32184/65536 [5:31:35<5:36:12,  1.65it/s]
 49%|████▉     | 32185/65536 [5:31:36<5:43:30,  1.62it/s]
 49%|████▉     | 32186/65536 [5:31:37<5:43:55,  1.62it/s]
 49%|████▉     | 32187/65536 [5:31:37<5:50:20,  1.59it/s]
 49%|████▉     | 32188/65536 [5:31:38<5:47:46,  1.60it/s]
 49%|████▉     | 32189/65536 [5:31:39<5:52:10,  1.58it/s]
 49%|████▉     | 32190/65536 [5:31:39<5:56:24,  1.56it/s]
 49%|████▉     | 32191/65536 [5:31:40<5:55:43,  1.56it/s]
 49%|████▉     | 32192/65536 [5:31:40<5:41:12,  1.63it/s]
 49%|████▉     | 32193/65536 [5:31:41<5:49:58,  1.59it/s]
 49%|████▉     | 32194/65536 [5:31:42<5:42:14,  1.62it/s]
 49%|████▉     | 32195/65536 [5:31:42<5:42:08,  1.62it/s]
 49%|████▉     | 32196/65536 [5:31:43<5:47:24,  1.60it/s]
 49%|████▉     | 32197/65536 [5:31:43<5:38:36,  1.64it/s]
 49%|████▉     | 32198/65536 [5:31:44<5:44:55,  1.61it/s]
 49%|████▉     | 32199/65536 [5:31:45<5:47:04,  1.60it/s]
 49%|████▉     | 32200/65536 [5:31:45<5:43:52,  1.62it/s]
                                                         
{'loss': 1.6503, 'learning_rate': 5.592016652381535e-07, 'epoch': 1987.65}
+
 49%|████▉     | 32200/65536 [5:31:45<5:43:52,  1.62it/s]
 49%|████▉     | 32201/65536 [5:31:46<5:50:09,  1.59it/s]
 49%|████▉     | 32202/65536 [5:31:47<5:57:06,  1.56it/s]
 49%|████▉     | 32203/65536 [5:31:47<5:53:24,  1.57it/s]
 49%|████▉     | 32204/65536 [5:31:48<5:45:42,  1.61it/s]
 49%|████▉     | 32205/65536 [5:31:48<5:44:55,  1.61it/s]
 49%|████▉     | 32206/65536 [5:31:49<5:53:32,  1.57it/s]
 49%|████▉     | 32207/65536 [5:31:50<5:42:00,  1.62it/s]
 49%|████▉     | 32208/65536 [5:31:50<5:51:36,  1.58it/s]
 49%|████▉     | 32209/65536 [5:31:51<5:39:37,  1.64it/s]
 49%|████▉     | 32210/65536 [5:31:52<5:38:58,  1.64it/s]
 49%|████▉     | 32211/65536 [5:31:52<5:38:15,  1.64it/s]
 49%|████▉     | 32212/65536 [5:31:53<5:33:27,  1.67it/s]
 49%|████▉     | 32213/65536 [5:31:53<5:50:08,  1.59it/s]
 49%|████▉     | 32214/65536 [5:31:54<5:41:06,  1.63it/s]
 49%|████▉     | 32215/65536 [5:31:55<5:43:12,  1.62it/s]
 49%|████▉     | 32216/65536 [5:31:55<5:40:05,  1.63it/s]
 49%|████▉     | 32217/65536 [5:31:56<5:39:07,  1.64it/s]
 49%|████▉     | 32218/65536 [5:31:56<5:40:07,  1.63it/s]
 49%|████▉     | 32219/65536 [5:31:57<5:41:11,  1.63it/s]
 49%|████▉     | 32220/65536 [5:31:58<5:48:51,  1.59it/s]
                                                         
{'loss': 1.6331, 'learning_rate': 5.589261662789273e-07, 'epoch': 1988.89}
+
 49%|████▉     | 32220/65536 [5:31:58<5:48:51,  1.59it/s]
 49%|████▉     | 32221/65536 [5:31:58<5:47:51,  1.60it/s]
 49%|████▉     | 32222/65536 [5:31:59<5:57:37,  1.55it/s]
 49%|████▉     | 32223/65536 [5:32:00<5:50:26,  1.58it/s]
 49%|████▉     | 32224/65536 [5:32:00<5:37:42,  1.64it/s]
 49%|████▉     | 32225/65536 [5:32:01<5:37:44,  1.64it/s]
 49%|████▉     | 32226/65536 [5:32:01<5:38:11,  1.64it/s]
 49%|████▉     | 32227/65536 [5:32:02<5:40:01,  1.63it/s]
 49%|████▉     | 32228/65536 [5:32:03<5:50:13,  1.59it/s]
 49%|████▉     | 32229/65536 [5:32:03<5:49:30,  1.59it/s]
 49%|████▉     | 32230/65536 [5:32:04<5:59:54,  1.54it/s]
 49%|████▉     | 32231/65536 [5:32:05<5:51:45,  1.58it/s]
 49%|████▉     | 32232/65536 [5:32:05<5:45:21,  1.61it/s]
 49%|████▉     | 32233/65536 [5:32:06<5:45:35,  1.61it/s]
 49%|████▉     | 32234/65536 [5:32:07<5:48:07,  1.59it/s]
 49%|████▉     | 32235/65536 [5:32:07<5:43:39,  1.62it/s]
 49%|████▉     | 32236/65536 [5:32:08<5:47:22,  1.60it/s]
 49%|████▉     | 32237/65536 [5:32:08<5:38:30,  1.64it/s]
 49%|████▉     | 32238/65536 [5:32:09<5:37:10,  1.65it/s]
 49%|████▉     | 32239/65536 [5:32:10<5:55:46,  1.56it/s]
 49%|████▉     | 32240/65536 [5:32:10<5:48:01,  1.59it/s]
                                                         
{'loss': 1.6758, 'learning_rate': 5.586506673197012e-07, 'epoch': 1990.12}
+
 49%|████▉     | 32240/65536 [5:32:10<5:48:01,  1.59it/s]
 49%|████▉     | 32241/65536 [5:32:11<5:47:04,  1.60it/s]
 49%|████▉     | 32242/65536 [5:32:11<5:42:23,  1.62it/s]
 49%|████▉     | 32243/65536 [5:32:12<5:43:08,  1.62it/s]
 49%|████▉     | 32244/65536 [5:32:13<5:41:22,  1.63it/s]
 49%|████▉     | 32245/65536 [5:32:13<5:36:48,  1.65it/s]
 49%|████▉     | 32246/65536 [5:32:14<5:36:35,  1.65it/s]
 49%|████▉     | 32247/65536 [5:32:14<5:34:43,  1.66it/s]
 49%|████▉     | 32248/65536 [5:32:15<5:34:41,  1.66it/s]
 49%|████▉     | 32249/65536 [5:32:16<5:34:04,  1.66it/s]
 49%|████▉     | 32250/65536 [5:32:16<5:31:12,  1.67it/s]
 49%|████▉     | 32251/65536 [5:32:17<5:29:27,  1.68it/s]
 49%|████▉     | 32252/65536 [5:32:18<5:50:34,  1.58it/s]
 49%|████▉     | 32253/65536 [5:32:18<5:45:39,  1.60it/s]
 49%|████▉     | 32254/65536 [5:32:19<5:46:35,  1.60it/s]
 49%|████▉     | 32255/65536 [5:32:20<6:02:36,  1.53it/s]
 49%|████▉     | 32256/65536 [5:32:20<5:56:36,  1.56it/s]
 49%|████▉     | 32257/65536 [5:32:21<5:44:29,  1.61it/s]
 49%|████▉     | 32258/65536 [5:32:21<5:47:57,  1.59it/s]
 49%|████▉     | 32259/65536 [5:32:22<5:40:13,  1.63it/s]
 49%|████▉     | 32260/65536 [5:32:23<5:44:24,  1.61it/s]
                                                         
{'loss': 1.6388, 'learning_rate': 5.58375168360475e-07, 'epoch': 1991.36}
+
 49%|████▉     | 32260/65536 [5:32:23<5:44:24,  1.61it/s]
 49%|████▉     | 32261/65536 [5:32:23<5:40:59,  1.63it/s]
 49%|████▉     | 32262/65536 [5:32:24<5:36:31,  1.65it/s]
 49%|████▉     | 32263/65536 [5:32:24<5:37:10,  1.64it/s]
 49%|████▉     | 32264/65536 [5:32:25<5:45:10,  1.61it/s]
 49%|████▉     | 32265/65536 [5:32:26<5:47:32,  1.60it/s]
 49%|████▉     | 32266/65536 [5:32:26<5:40:12,  1.63it/s]
 49%|████▉     | 32267/65536 [5:32:27<5:34:17,  1.66it/s]
 49%|████▉     | 32268/65536 [5:32:27<5:42:17,  1.62it/s]
 49%|████▉     | 32269/65536 [5:32:28<5:36:03,  1.65it/s]
 49%|████▉     | 32270/65536 [5:32:29<5:38:22,  1.64it/s]
 49%|████▉     | 32271/65536 [5:32:29<5:53:26,  1.57it/s]
 49%|████▉     | 32272/65536 [5:32:30<5:48:37,  1.59it/s]
 49%|████▉     | 32273/65536 [5:32:31<5:55:43,  1.56it/s]
 49%|████▉     | 32274/65536 [5:32:31<5:40:31,  1.63it/s]
 49%|████▉     | 32275/65536 [5:32:32<5:45:01,  1.61it/s]
 49%|████▉     | 32276/65536 [5:32:32<5:43:20,  1.61it/s]
 49%|████▉     | 32277/65536 [5:32:33<5:40:15,  1.63it/s]
 49%|████▉     | 32278/65536 [5:32:34<5:38:39,  1.64it/s]
 49%|████▉     | 32279/65536 [5:32:34<5:40:52,  1.63it/s]
 49%|████▉     | 32280/65536 [5:32:35<5:36:58,  1.64it/s]
                                                         
{'loss': 1.6492, 'learning_rate': 5.580996694012489e-07, 'epoch': 1992.59}
+
 49%|████▉     | 32280/65536 [5:32:35<5:36:58,  1.64it/s]
 49%|████▉     | 32281/65536 [5:32:36<5:39:26,  1.63it/s]
 49%|████▉     | 32282/65536 [5:32:36<5:46:57,  1.60it/s]
 49%|████▉     | 32283/65536 [5:32:37<5:42:27,  1.62it/s]
 49%|████▉     | 32284/65536 [5:32:37<5:39:54,  1.63it/s]
 49%|████▉     | 32285/65536 [5:32:38<5:33:59,  1.66it/s]
 49%|████▉     | 32286/65536 [5:32:39<5:38:28,  1.64it/s]
 49%|████▉     | 32287/65536 [5:32:39<5:46:03,  1.60it/s]
 49%|████▉     | 32288/65536 [5:32:40<5:44:19,  1.61it/s]
 49%|████▉     | 32289/65536 [5:32:40<5:45:38,  1.60it/s]
 49%|████▉     | 32290/65536 [5:32:41<5:43:40,  1.61it/s]
 49%|████▉     | 32291/65536 [5:32:42<5:40:26,  1.63it/s]
 49%|████▉     | 32292/65536 [5:32:42<5:37:04,  1.64it/s]
 49%|████▉     | 32293/65536 [5:32:43<5:33:48,  1.66it/s]
 49%|████▉     | 32294/65536 [5:32:43<5:35:34,  1.65it/s]
 49%|████▉     | 32295/65536 [5:32:44<5:35:54,  1.65it/s]
 49%|████▉     | 32296/65536 [5:32:45<5:43:15,  1.61it/s]
 49%|████▉     | 32297/65536 [5:32:45<5:41:46,  1.62it/s]
 49%|████▉     | 32298/65536 [5:32:46<5:50:22,  1.58it/s]
 49%|████▉     | 32299/65536 [5:32:47<5:53:26,  1.57it/s]
 49%|████▉     | 32300/65536 [5:32:47<5:48:47,  1.59it/s]
                                                         
{'loss': 1.635, 'learning_rate': 5.578241704420228e-07, 'epoch': 1993.83}
+
 49%|████▉     | 32300/65536 [5:32:47<5:48:47,  1.59it/s]
 49%|████▉     | 32301/65536 [5:32:48<5:48:49,  1.59it/s]
 49%|████▉     | 32302/65536 [5:32:49<5:43:31,  1.61it/s]
 49%|████▉     | 32303/65536 [5:32:49<5:50:50,  1.58it/s]
 49%|████▉     | 32304/65536 [5:32:50<5:51:52,  1.57it/s]
 49%|████▉     | 32305/65536 [5:32:50<5:48:35,  1.59it/s]
 49%|████▉     | 32306/65536 [5:32:51<5:45:54,  1.60it/s]
 49%|████▉     | 32307/65536 [5:32:52<5:50:27,  1.58it/s]
 49%|████▉     | 32308/65536 [5:32:52<5:45:51,  1.60it/s]
 49%|████▉     | 32309/65536 [5:32:53<5:41:28,  1.62it/s]
 49%|████▉     | 32310/65536 [5:32:54<5:47:22,  1.59it/s]
 49%|████▉     | 32311/65536 [5:32:54<5:45:06,  1.60it/s]
 49%|████▉     | 32312/65536 [5:32:55<5:46:57,  1.60it/s]
 49%|████▉     | 32313/65536 [5:32:55<5:47:32,  1.59it/s]
 49%|████▉     | 32314/65536 [5:32:56<5:54:16,  1.56it/s]
 49%|████▉     | 32315/65536 [5:32:57<5:49:54,  1.58it/s]
 49%|████▉     | 32316/65536 [5:32:57<5:41:19,  1.62it/s]
 49%|████▉     | 32317/65536 [5:32:58<5:44:44,  1.61it/s]
 49%|████▉     | 32318/65536 [5:32:58<5:36:18,  1.65it/s]
 49%|████▉     | 32319/65536 [5:32:59<5:33:36,  1.66it/s]
 49%|████▉     | 32320/65536 [5:33:00<5:51:37,  1.57it/s]
                                                         
{'loss': 1.65, 'learning_rate': 5.575486714827966e-07, 'epoch': 1995.06}
+
 49%|████▉     | 32320/65536 [5:33:00<5:51:37,  1.57it/s]
 49%|████▉     | 32321/65536 [5:33:00<5:48:15,  1.59it/s]
 49%|████▉     | 32322/65536 [5:33:01<5:42:02,  1.62it/s]
 49%|████▉     | 32323/65536 [5:33:02<5:46:26,  1.60it/s]
 49%|████▉     | 32324/65536 [5:33:02<5:48:17,  1.59it/s]
 49%|████▉     | 32325/65536 [5:33:03<5:43:07,  1.61it/s]
 49%|████▉     | 32326/65536 [5:33:03<5:41:06,  1.62it/s]
 49%|████▉     | 32327/65536 [5:33:04<5:43:12,  1.61it/s]
 49%|████▉     | 32328/65536 [5:33:05<5:37:11,  1.64it/s]
 49%|████▉     | 32329/65536 [5:33:05<5:45:47,  1.60it/s]
 49%|████▉     | 32330/65536 [5:33:06<5:46:04,  1.60it/s]
 49%|████▉     | 32331/65536 [5:33:07<5:41:07,  1.62it/s]
 49%|████▉     | 32332/65536 [5:33:07<5:45:17,  1.60it/s]
 49%|████▉     | 32333/65536 [5:33:08<5:40:19,  1.63it/s]
 49%|████▉     | 32334/65536 [5:33:08<5:32:24,  1.66it/s]
 49%|████▉     | 32335/65536 [5:33:09<5:26:40,  1.69it/s]
 49%|████▉     | 32336/65536 [5:33:10<5:37:14,  1.64it/s]
 49%|████▉     | 32337/65536 [5:33:10<5:43:55,  1.61it/s]
 49%|████▉     | 32338/65536 [5:33:11<5:45:26,  1.60it/s]
 49%|████▉     | 32339/65536 [5:33:11<5:39:30,  1.63it/s]
 49%|████▉     | 32340/65536 [5:33:12<5:33:58,  1.66it/s]
                                                         
{'loss': 1.6663, 'learning_rate': 5.572731725235705e-07, 'epoch': 1996.3}
+
 49%|████▉     | 32340/65536 [5:33:12<5:33:58,  1.66it/s]
 49%|████▉     | 32341/65536 [5:33:13<5:32:38,  1.66it/s]
 49%|████▉     | 32342/65536 [5:33:13<5:42:19,  1.62it/s]
 49%|████▉     | 32343/65536 [5:33:14<5:47:21,  1.59it/s]
 49%|████▉     | 32344/65536 [5:33:15<5:47:01,  1.59it/s]
 49%|████▉     | 32345/65536 [5:33:15<5:39:01,  1.63it/s]
 49%|████▉     | 32346/65536 [5:33:16<5:38:44,  1.63it/s]
 49%|████▉     | 32347/65536 [5:33:16<5:39:23,  1.63it/s]
 49%|████▉     | 32348/65536 [5:33:17<5:44:34,  1.61it/s]
 49%|████▉     | 32349/65536 [5:33:18<5:45:11,  1.60it/s]
 49%|████▉     | 32350/65536 [5:33:18<5:37:17,  1.64it/s]
 49%|████▉     | 32351/65536 [5:33:19<5:33:44,  1.66it/s]
 49%|████▉     | 32352/65536 [5:33:20<6:00:15,  1.54it/s]
 49%|████▉     | 32353/65536 [5:33:20<5:52:36,  1.57it/s]
 49%|████▉     | 32354/65536 [5:33:21<5:52:37,  1.57it/s]
 49%|████▉     | 32355/65536 [5:33:21<5:54:02,  1.56it/s]
 49%|████▉     | 32356/65536 [5:33:22<5:52:41,  1.57it/s]
 49%|████▉     | 32357/65536 [5:33:23<5:41:43,  1.62it/s]
 49%|████▉     | 32358/65536 [5:33:23<5:30:53,  1.67it/s]
 49%|████▉     | 32359/65536 [5:33:24<5:30:25,  1.67it/s]
 49%|████▉     | 32360/65536 [5:33:24<5:37:38,  1.64it/s]
                                                         
{'loss': 1.6435, 'learning_rate': 5.569976735643443e-07, 'epoch': 1997.53}
+
 49%|████▉     | 32360/65536 [5:33:24<5:37:38,  1.64it/s]
 49%|████▉     | 32361/65536 [5:33:25<5:39:44,  1.63it/s]
 49%|████▉     | 32362/65536 [5:33:26<5:29:03,  1.68it/s]
 49%|████▉     | 32363/65536 [5:33:26<5:26:46,  1.69it/s]
 49%|████▉     | 32364/65536 [5:33:27<5:29:42,  1.68it/s]
 49%|████▉     | 32365/65536 [5:33:27<5:22:21,  1.72it/s]
 49%|████▉     | 32366/65536 [5:33:28<5:37:51,  1.64it/s]
 49%|████▉     | 32367/65536 [5:33:29<5:37:22,  1.64it/s]
 49%|████▉     | 32368/65536 [5:33:29<5:53:54,  1.56it/s]
 49%|████▉     | 32369/65536 [5:33:30<5:53:18,  1.56it/s]
 49%|████▉     | 32370/65536 [5:33:31<5:53:10,  1.57it/s]
 49%|████▉     | 32371/65536 [5:33:31<5:46:25,  1.60it/s]
 49%|████▉     | 32372/65536 [5:33:32<5:36:12,  1.64it/s]
 49%|████▉     | 32373/65536 [5:33:33<5:46:57,  1.59it/s]
 49%|████▉     | 32374/65536 [5:33:33<5:35:42,  1.65it/s]
 49%|████▉     | 32375/65536 [5:33:34<5:42:01,  1.62it/s]
 49%|████▉     | 32376/65536 [5:33:34<5:39:16,  1.63it/s]
 49%|████▉     | 32377/65536 [5:33:35<5:43:11,  1.61it/s]
 49%|████▉     | 32378/65536 [5:33:36<5:40:09,  1.62it/s]
 49%|████▉     | 32379/65536 [5:33:36<5:39:47,  1.63it/s]
 49%|████▉     | 32380/65536 [5:33:37<5:43:38,  1.61it/s]
                                                         
{'loss': 1.6865, 'learning_rate': 5.567221746051181e-07, 'epoch': 1998.77}
+
 49%|████▉     | 32380/65536 [5:33:37<5:43:38,  1.61it/s]
 49%|████▉     | 32381/65536 [5:33:37<5:49:18,  1.58it/s]
 49%|████▉     | 32382/65536 [5:33:38<5:43:15,  1.61it/s]
 49%|████▉     | 32383/65536 [5:33:39<5:36:28,  1.64it/s]
 49%|████▉     | 32384/65536 [5:33:39<5:50:21,  1.58it/s]
 49%|████▉     | 32385/65536 [5:33:40<5:41:25,  1.62it/s]
 49%|████▉     | 32386/65536 [5:33:41<5:42:52,  1.61it/s]
 49%|████▉     | 32387/65536 [5:33:41<5:34:56,  1.65it/s]
 49%|████▉     | 32388/65536 [5:33:42<5:41:41,  1.62it/s]
 49%|████▉     | 32389/65536 [5:33:42<5:42:04,  1.61it/s]
 49%|████▉     | 32390/65536 [5:33:43<5:36:47,  1.64it/s]
 49%|████▉     | 32391/65536 [5:33:44<5:35:07,  1.65it/s]
 49%|████▉     | 32392/65536 [5:33:44<5:30:06,  1.67it/s]
 49%|████▉     | 32393/65536 [5:33:45<5:27:25,  1.69it/s]
 49%|████▉     | 32394/65536 [5:33:45<5:30:51,  1.67it/s]
 49%|████▉     | 32395/65536 [5:33:46<5:28:00,  1.68it/s]
 49%|████▉     | 32396/65536 [5:33:47<5:40:17,  1.62it/s]
 49%|████▉     | 32397/65536 [5:33:47<5:44:20,  1.60it/s]
 49%|████▉     | 32398/65536 [5:33:48<5:36:43,  1.64it/s]
 49%|████▉     | 32399/65536 [5:33:48<5:41:25,  1.62it/s]
 49%|████▉     | 32400/65536 [5:33:49<5:45:52,  1.60it/s]
                                                         
{'loss': 1.6806, 'learning_rate': 5.564466756458919e-07, 'epoch': 2000.0}
+
 49%|████▉     | 32400/65536 [5:33:49<5:45:52,  1.60it/s]
 49%|████▉     | 32401/65536 [5:33:50<5:51:20,  1.57it/s]
 49%|████▉     | 32402/65536 [5:33:50<5:47:08,  1.59it/s]
 49%|████▉     | 32403/65536 [5:33:51<5:50:51,  1.57it/s]
 49%|████▉     | 32404/65536 [5:33:52<5:46:23,  1.59it/s]
 49%|████▉     | 32405/65536 [5:33:52<5:54:33,  1.56it/s]
 49%|████▉     | 32406/65536 [5:33:53<5:54:54,  1.56it/s]
 49%|████▉     | 32407/65536 [5:33:54<5:42:41,  1.61it/s]
 49%|████▉     | 32408/65536 [5:33:54<5:39:45,  1.63it/s]
 49%|████▉     | 32409/65536 [5:33:55<5:36:23,  1.64it/s]
 49%|████▉     | 32410/65536 [5:33:55<5:33:36,  1.65it/s]
 49%|████▉     | 32411/65536 [5:33:56<5:32:35,  1.66it/s]
 49%|████▉     | 32412/65536 [5:33:57<5:41:01,  1.62it/s]
 49%|████▉     | 32413/65536 [5:33:57<5:42:06,  1.61it/s]
 49%|████▉     | 32414/65536 [5:33:58<5:35:15,  1.65it/s]
 49%|████▉     | 32415/65536 [5:33:58<5:40:15,  1.62it/s]
 49%|████▉     | 32416/65536 [5:33:59<5:44:44,  1.60it/s]
 49%|████▉     | 32417/65536 [5:34:00<5:52:58,  1.56it/s]
 49%|████▉     | 32418/65536 [5:34:00<5:45:25,  1.60it/s]
 49%|████▉     | 32419/65536 [5:34:01<5:35:59,  1.64it/s]
 49%|████▉     | 32420/65536 [5:34:01<5:34:28,  1.65it/s]
                                                         
{'loss': 1.664, 'learning_rate': 5.561711766866658e-07, 'epoch': 2001.23}
+
 49%|████▉     | 32420/65536 [5:34:01<5:34:28,  1.65it/s]
 49%|████▉     | 32421/65536 [5:34:02<5:34:16,  1.65it/s]
 49%|████▉     | 32422/65536 [5:34:03<5:29:28,  1.68it/s]
 49%|████▉     | 32423/65536 [5:34:03<5:34:32,  1.65it/s]
 49%|████▉     | 32424/65536 [5:34:04<5:45:02,  1.60it/s]
 49%|████▉     | 32425/65536 [5:34:05<5:47:07,  1.59it/s]
 49%|████▉     | 32426/65536 [5:34:05<5:44:27,  1.60it/s]
 49%|████▉     | 32427/65536 [5:34:06<6:00:00,  1.53it/s]
 49%|████▉     | 32428/65536 [5:34:07<5:57:48,  1.54it/s]
 49%|████▉     | 32429/65536 [5:34:07<5:50:47,  1.57it/s]
 49%|████▉     | 32430/65536 [5:34:08<5:50:07,  1.58it/s]
 49%|████▉     | 32431/65536 [5:34:08<5:42:21,  1.61it/s]
 49%|████▉     | 32432/65536 [5:34:09<5:35:25,  1.64it/s]
 49%|████▉     | 32433/65536 [5:34:10<5:53:26,  1.56it/s]
 49%|████▉     | 32434/65536 [5:34:10<6:03:27,  1.52it/s]
 49%|████▉     | 32435/65536 [5:34:11<5:57:51,  1.54it/s]
 49%|████▉     | 32436/65536 [5:34:12<5:51:05,  1.57it/s]
 49%|████▉     | 32437/65536 [5:34:12<5:42:35,  1.61it/s]
 49%|████▉     | 32438/65536 [5:34:13<5:41:56,  1.61it/s]
 49%|████▉     | 32439/65536 [5:34:13<5:47:56,  1.59it/s]
 49%|████▉     | 32440/65536 [5:34:14<5:42:55,  1.61it/s]
                                                         
{'loss': 1.583, 'learning_rate': 5.558956777274397e-07, 'epoch': 2002.47}
+
 49%|████▉     | 32440/65536 [5:34:14<5:42:55,  1.61it/s]
 50%|████▉     | 32441/65536 [5:34:15<5:37:24,  1.63it/s]
 50%|████▉     | 32442/65536 [5:34:15<5:32:33,  1.66it/s]
 50%|████▉     | 32443/65536 [5:34:16<5:35:07,  1.65it/s]
 50%|████▉     | 32444/65536 [5:34:16<5:28:26,  1.68it/s]
 50%|████▉     | 32445/65536 [5:34:17<5:29:35,  1.67it/s]
 50%|████▉     | 32446/65536 [5:34:18<5:26:50,  1.69it/s]
 50%|████▉     | 32447/65536 [5:34:18<5:26:58,  1.69it/s]
 50%|████▉     | 32448/65536 [5:34:19<5:29:59,  1.67it/s]
 50%|████▉     | 32449/65536 [5:34:20<5:53:27,  1.56it/s]
 50%|████▉     | 32450/65536 [5:34:20<5:49:59,  1.58it/s]
 50%|████▉     | 32451/65536 [5:34:21<5:35:54,  1.64it/s]
 50%|████▉     | 32452/65536 [5:34:21<5:45:15,  1.60it/s]
 50%|████▉     | 32453/65536 [5:34:22<5:38:41,  1.63it/s]
 50%|████▉     | 32454/65536 [5:34:23<5:41:27,  1.61it/s]
 50%|████▉     | 32455/65536 [5:34:23<5:41:35,  1.61it/s]
 50%|████▉     | 32456/65536 [5:34:24<5:32:41,  1.66it/s]
 50%|████▉     | 32457/65536 [5:34:24<5:31:40,  1.66it/s]
 50%|████▉     | 32458/65536 [5:34:25<5:32:38,  1.66it/s]
 50%|████▉     | 32459/65536 [5:34:26<5:36:06,  1.64it/s]
 50%|████▉     | 32460/65536 [5:34:26<5:38:52,  1.63it/s]
                                                         
{'loss': 1.7005, 'learning_rate': 5.556201787682135e-07, 'epoch': 2003.7}
+
 50%|████▉     | 32460/65536 [5:34:26<5:38:52,  1.63it/s]
 50%|████▉     | 32461/65536 [5:34:27<5:51:15,  1.57it/s]
 50%|████▉     | 32462/65536 [5:34:28<5:44:35,  1.60it/s]
 50%|████▉     | 32463/65536 [5:34:28<5:56:06,  1.55it/s]
 50%|████▉     | 32464/65536 [5:34:29<5:48:22,  1.58it/s]
 50%|████▉     | 32465/65536 [5:34:30<5:57:02,  1.54it/s]
 50%|████▉     | 32466/65536 [5:34:30<5:56:36,  1.55it/s]
 50%|████▉     | 32467/65536 [5:34:31<5:53:13,  1.56it/s]
 50%|████▉     | 32468/65536 [5:34:31<5:48:13,  1.58it/s]
 50%|████▉     | 32469/65536 [5:34:32<5:39:29,  1.62it/s]
 50%|████▉     | 32470/65536 [5:34:33<5:40:23,  1.62it/s]
 50%|████▉     | 32471/65536 [5:34:33<5:40:33,  1.62it/s]
 50%|████▉     | 32472/65536 [5:34:34<5:38:45,  1.63it/s]
 50%|████▉     | 32473/65536 [5:34:34<5:36:56,  1.64it/s]
 50%|████▉     | 32474/65536 [5:34:35<5:29:51,  1.67it/s]
 50%|████▉     | 32475/65536 [5:34:36<5:26:32,  1.69it/s]
 50%|████▉     | 32476/65536 [5:34:36<5:33:55,  1.65it/s]
 50%|████▉     | 32477/65536 [5:34:37<5:31:50,  1.66it/s]
 50%|████▉     | 32478/65536 [5:34:37<5:41:34,  1.61it/s]
 50%|████▉     | 32479/65536 [5:34:38<5:46:40,  1.59it/s]
 50%|████▉     | 32480/65536 [5:34:39<5:43:30,  1.60it/s]
                                                         
{'loss': 1.6248, 'learning_rate': 5.553446798089874e-07, 'epoch': 2004.94}
+
 50%|████▉     | 32480/65536 [5:34:39<5:43:30,  1.60it/s]
 50%|████▉     | 32481/65536 [5:34:39<5:41:12,  1.61it/s]
 50%|████▉     | 32482/65536 [5:34:40<5:55:39,  1.55it/s]
 50%|████▉     | 32483/65536 [5:34:41<5:50:45,  1.57it/s]
 50%|████▉     | 32484/65536 [5:34:41<5:40:00,  1.62it/s]
 50%|████▉     | 32485/65536 [5:34:42<5:34:28,  1.65it/s]
 50%|████▉     | 32486/65536 [5:34:42<5:35:21,  1.64it/s]
 50%|████▉     | 32487/65536 [5:34:43<5:38:01,  1.63it/s]
 50%|████▉     | 32488/65536 [5:34:44<5:41:05,  1.61it/s]
 50%|████▉     | 32489/65536 [5:34:44<5:36:34,  1.64it/s]
 50%|████▉     | 32490/65536 [5:34:45<5:39:41,  1.62it/s]
 50%|████▉     | 32491/65536 [5:34:46<5:36:24,  1.64it/s]
 50%|████▉     | 32492/65536 [5:34:46<5:32:18,  1.66it/s]
 50%|████▉     | 32493/65536 [5:34:47<5:38:17,  1.63it/s]
 50%|████▉     | 32494/65536 [5:34:47<5:41:01,  1.61it/s]
 50%|████▉     | 32495/65536 [5:34:48<5:32:50,  1.65it/s]
 50%|████▉     | 32496/65536 [5:34:49<5:44:14,  1.60it/s]
 50%|████▉     | 32497/65536 [5:34:49<5:40:26,  1.62it/s]
 50%|████▉     | 32498/65536 [5:34:50<5:50:05,  1.57it/s]
 50%|████▉     | 32499/65536 [5:34:50<5:44:26,  1.60it/s]
 50%|████▉     | 32500/65536 [5:34:51<5:40:09,  1.62it/s]
                                                         
{'loss': 1.6806, 'learning_rate': 5.550691808497612e-07, 'epoch': 2006.17}
+
 50%|████▉     | 32500/65536 [5:34:51<5:40:09,  1.62it/s]
 50%|████▉     | 32501/65536 [5:34:52<5:37:22,  1.63it/s]
 50%|████▉     | 32502/65536 [5:34:52<5:30:58,  1.66it/s]
 50%|████▉     | 32503/65536 [5:34:53<5:36:22,  1.64it/s]
 50%|████▉     | 32504/65536 [5:34:53<5:34:33,  1.65it/s]
 50%|████▉     | 32505/65536 [5:34:54<5:37:18,  1.63it/s]
 50%|████▉     | 32506/65536 [5:34:55<5:44:31,  1.60it/s]
 50%|████▉     | 32507/65536 [5:34:55<5:41:49,  1.61it/s]
 50%|████▉     | 32508/65536 [5:34:56<5:37:57,  1.63it/s]
 50%|████▉     | 32509/65536 [5:34:57<5:39:53,  1.62it/s]
 50%|████▉     | 32510/65536 [5:34:57<5:39:59,  1.62it/s]
 50%|████▉     | 32511/65536 [5:34:58<5:35:17,  1.64it/s]
 50%|████▉     | 32512/65536 [5:34:58<5:37:44,  1.63it/s]
 50%|████▉     | 32513/65536 [5:34:59<5:40:43,  1.62it/s]
 50%|████▉     | 32514/65536 [5:35:00<5:53:12,  1.56it/s]
 50%|████▉     | 32515/65536 [5:35:00<5:47:22,  1.58it/s]
 50%|████▉     | 32516/65536 [5:35:01<5:52:16,  1.56it/s]
 50%|████▉     | 32517/65536 [5:35:02<5:52:39,  1.56it/s]
 50%|████▉     | 32518/65536 [5:35:02<5:46:06,  1.59it/s]
 50%|████▉     | 32519/65536 [5:35:03<5:51:34,  1.57it/s]
 50%|████▉     | 32520/65536 [5:35:04<5:47:00,  1.59it/s]
                                                         
{'loss': 1.6522, 'learning_rate': 5.547936818905351e-07, 'epoch': 2007.41}
+
 50%|████▉     | 32520/65536 [5:35:04<5:47:00,  1.59it/s]
 50%|████▉     | 32521/65536 [5:35:04<5:43:26,  1.60it/s]
 50%|████▉     | 32522/65536 [5:35:05<5:38:36,  1.62it/s]
 50%|████▉     | 32523/65536 [5:35:05<5:32:46,  1.65it/s]
 50%|████▉     | 32524/65536 [5:35:06<5:33:15,  1.65it/s]
 50%|████▉     | 32525/65536 [5:35:07<5:37:17,  1.63it/s]
 50%|████▉     | 32526/65536 [5:35:07<5:33:39,  1.65it/s]
 50%|████▉     | 32527/65536 [5:35:08<5:26:56,  1.68it/s]
 50%|████▉     | 32528/65536 [5:35:08<5:31:04,  1.66it/s]
 50%|████▉     | 32529/65536 [5:35:09<5:32:36,  1.65it/s]
 50%|████▉     | 32530/65536 [5:35:10<5:48:03,  1.58it/s]
 50%|████▉     | 32531/65536 [5:35:10<5:41:26,  1.61it/s]
 50%|████▉     | 32532/65536 [5:35:11<5:40:13,  1.62it/s]
 50%|████▉     | 32533/65536 [5:35:12<5:46:12,  1.59it/s]
 50%|████▉     | 32534/65536 [5:35:12<5:38:38,  1.62it/s]
 50%|████▉     | 32535/65536 [5:35:13<5:30:12,  1.67it/s]
 50%|████▉     | 32536/65536 [5:35:13<5:29:31,  1.67it/s]
 50%|████▉     | 32537/65536 [5:35:14<5:33:20,  1.65it/s]
 50%|████▉     | 32538/65536 [5:35:15<5:36:30,  1.63it/s]
 50%|████▉     | 32539/65536 [5:35:15<5:36:06,  1.64it/s]
 50%|████▉     | 32540/65536 [5:35:16<5:37:33,  1.63it/s]
                                                         
{'loss': 1.6834, 'learning_rate': 5.54518182931309e-07, 'epoch': 2008.64}
+
 50%|████▉     | 32540/65536 [5:35:16<5:37:33,  1.63it/s]
 50%|████▉     | 32541/65536 [5:35:16<5:38:32,  1.62it/s]
 50%|████▉     | 32542/65536 [5:35:17<5:48:22,  1.58it/s]
 50%|████▉     | 32543/65536 [5:35:18<5:49:42,  1.57it/s]
 50%|████▉     | 32544/65536 [5:35:18<5:50:00,  1.57it/s]
 50%|████▉     | 32545/65536 [5:35:19<5:40:25,  1.62it/s]
 50%|████▉     | 32546/65536 [5:35:20<6:00:45,  1.52it/s]
 50%|████▉     | 32547/65536 [5:35:20<5:51:20,  1.56it/s]
 50%|████▉     | 32548/65536 [5:35:21<5:50:53,  1.57it/s]
 50%|████▉     | 32549/65536 [5:35:21<5:47:44,  1.58it/s]
 50%|████▉     | 32550/65536 [5:35:22<5:34:09,  1.65it/s]
 50%|████▉     | 32551/65536 [5:35:23<5:35:33,  1.64it/s]
 50%|████▉     | 32552/65536 [5:35:23<5:34:25,  1.64it/s]
 50%|████▉     | 32553/65536 [5:35:24<5:30:24,  1.66it/s]
 50%|████▉     | 32554/65536 [5:35:25<5:49:49,  1.57it/s]
 50%|████▉     | 32555/65536 [5:35:25<5:48:13,  1.58it/s]
 50%|████▉     | 32556/65536 [5:35:26<5:49:55,  1.57it/s]
 50%|████▉     | 32557/65536 [5:35:26<5:40:10,  1.62it/s]
 50%|████▉     | 32558/65536 [5:35:27<5:40:29,  1.61it/s]
 50%|████▉     | 32559/65536 [5:35:28<5:40:43,  1.61it/s]
 50%|████▉     | 32560/65536 [5:35:28<5:34:30,  1.64it/s]
                                                         
{'loss': 1.6222, 'learning_rate': 5.542426839720828e-07, 'epoch': 2009.88}
+
 50%|████▉     | 32560/65536 [5:35:28<5:34:30,  1.64it/s]
 50%|████▉     | 32561/65536 [5:35:29<5:35:18,  1.64it/s]
 50%|████▉     | 32562/65536 [5:35:29<5:40:40,  1.61it/s]
 50%|████▉     | 32563/65536 [5:35:30<5:50:03,  1.57it/s]
 50%|████▉     | 32564/65536 [5:35:31<5:39:58,  1.62it/s]
 50%|████▉     | 32565/65536 [5:35:31<5:34:26,  1.64it/s]
 50%|████▉     | 32566/65536 [5:35:32<5:36:44,  1.63it/s]
 50%|████▉     | 32567/65536 [5:35:33<5:41:03,  1.61it/s]
 50%|████▉     | 32568/65536 [5:35:33<5:38:50,  1.62it/s]
 50%|████▉     | 32569/65536 [5:35:34<5:30:25,  1.66it/s]
 50%|████▉     | 32570/65536 [5:35:34<5:43:08,  1.60it/s]
 50%|████▉     | 32571/65536 [5:35:35<5:43:28,  1.60it/s]
 50%|████▉     | 32572/65536 [5:35:36<5:45:57,  1.59it/s]
 50%|████▉     | 32573/65536 [5:35:36<5:51:04,  1.56it/s]
 50%|████▉     | 32574/65536 [5:35:37<5:42:55,  1.60it/s]
 50%|████▉     | 32575/65536 [5:35:38<5:47:14,  1.58it/s]
 50%|████▉     | 32576/65536 [5:35:38<5:43:53,  1.60it/s]
 50%|████▉     | 32577/65536 [5:35:39<5:40:35,  1.61it/s]
 50%|████▉     | 32578/65536 [5:35:39<5:34:02,  1.64it/s]
 50%|████▉     | 32579/65536 [5:35:40<5:41:43,  1.61it/s]
 50%|████▉     | 32580/65536 [5:35:41<5:35:56,  1.64it/s]
                                                         
{'loss': 1.6515, 'learning_rate': 5.539671850128567e-07, 'epoch': 2011.11}
+
 50%|████▉     | 32580/65536 [5:35:41<5:35:56,  1.64it/s]
 50%|████▉     | 32581/65536 [5:35:41<5:46:00,  1.59it/s]
 50%|████▉     | 32582/65536 [5:35:42<5:41:22,  1.61it/s]
 50%|████▉     | 32583/65536 [5:35:43<5:41:21,  1.61it/s]
 50%|████▉     | 32584/65536 [5:35:43<5:39:00,  1.62it/s]
 50%|████▉     | 32585/65536 [5:35:44<5:35:57,  1.63it/s]
 50%|████▉     | 32586/65536 [5:35:44<5:42:55,  1.60it/s]
 50%|████▉     | 32587/65536 [5:35:45<5:50:21,  1.57it/s]
 50%|████▉     | 32588/65536 [5:35:46<5:50:40,  1.57it/s]
 50%|████▉     | 32589/65536 [5:35:46<5:48:22,  1.58it/s]
 50%|████▉     | 32590/65536 [5:35:47<5:49:17,  1.57it/s]
 50%|████▉     | 32591/65536 [5:35:48<5:46:46,  1.58it/s]
 50%|████▉     | 32592/65536 [5:35:48<5:41:52,  1.61it/s]
 50%|████▉     | 32593/65536 [5:35:49<5:36:27,  1.63it/s]
 50%|████▉     | 32594/65536 [5:35:49<5:36:03,  1.63it/s]
 50%|████▉     | 32595/65536 [5:35:50<5:51:42,  1.56it/s]
 50%|████▉     | 32596/65536 [5:35:51<5:55:20,  1.54it/s]
 50%|████▉     | 32597/65536 [5:35:51<5:45:14,  1.59it/s]
 50%|████▉     | 32598/65536 [5:35:52<5:35:07,  1.64it/s]
 50%|████▉     | 32599/65536 [5:35:53<5:41:48,  1.61it/s]
 50%|████▉     | 32600/65536 [5:35:53<5:36:59,  1.63it/s]
                                                         
{'loss': 1.6526, 'learning_rate': 5.536916860536305e-07, 'epoch': 2012.35}
+
 50%|████▉     | 32600/65536 [5:35:53<5:36:59,  1.63it/s]
 50%|████▉     | 32601/65536 [5:35:54<5:39:23,  1.62it/s]
 50%|████▉     | 32602/65536 [5:35:54<5:39:16,  1.62it/s]
 50%|████▉     | 32603/65536 [5:35:55<5:34:40,  1.64it/s]
 50%|████▉     | 32604/65536 [5:35:56<5:39:04,  1.62it/s]
 50%|████▉     | 32605/65536 [5:35:56<5:38:08,  1.62it/s]
 50%|████▉     | 32606/65536 [5:35:57<5:38:20,  1.62it/s]
 50%|████▉     | 32607/65536 [5:35:57<5:33:48,  1.64it/s]
 50%|████▉     | 32608/65536 [5:35:58<5:38:52,  1.62it/s]
 50%|████▉     | 32609/65536 [5:35:59<5:33:38,  1.64it/s]
 50%|████▉     | 32610/65536 [5:35:59<5:38:05,  1.62it/s]
 50%|████▉     | 32611/65536 [5:36:00<5:47:41,  1.58it/s]
 50%|████▉     | 32612/65536 [5:36:01<5:33:37,  1.64it/s]
 50%|████▉     | 32613/65536 [5:36:01<5:30:23,  1.66it/s]
 50%|████▉     | 32614/65536 [5:36:02<5:32:50,  1.65it/s]
 50%|████▉     | 32615/65536 [5:36:02<5:30:09,  1.66it/s]
 50%|████▉     | 32616/65536 [5:36:03<5:27:56,  1.67it/s]
 50%|████▉     | 32617/65536 [5:36:04<5:30:12,  1.66it/s]
 50%|████▉     | 32618/65536 [5:36:04<5:46:46,  1.58it/s]
 50%|███���▉     | 32619/65536 [5:36:05<5:44:17,  1.59it/s]
 50%|████▉     | 32620/65536 [5:36:05<5:42:52,  1.60it/s]
                                                         
{'loss': 1.6346, 'learning_rate': 5.534161870944044e-07, 'epoch': 2013.58}
+
 50%|████▉     | 32620/65536 [5:36:05<5:42:52,  1.60it/s]
 50%|████▉     | 32621/65536 [5:36:06<5:39:58,  1.61it/s]
 50%|████▉     | 32622/65536 [5:36:07<5:41:50,  1.60it/s]
 50%|████▉     | 32623/65536 [5:36:07<5:43:00,  1.60it/s]
 50%|████▉     | 32624/65536 [5:36:08<5:30:47,  1.66it/s]
 50%|████▉     | 32625/65536 [5:36:09<5:39:15,  1.62it/s]
 50%|████▉     | 32626/65536 [5:36:09<5:36:57,  1.63it/s]
 50%|████▉     | 32627/65536 [5:36:10<5:47:55,  1.58it/s]
 50%|████▉     | 32628/65536 [5:36:10<5:47:00,  1.58it/s]
 50%|████▉     | 32629/65536 [5:36:11<5:37:40,  1.62it/s]
 50%|████▉     | 32630/65536 [5:36:12<5:27:43,  1.67it/s]
 50%|████▉     | 32631/65536 [5:36:12<5:40:34,  1.61it/s]
 50%|████▉     | 32632/65536 [5:36:13<5:32:04,  1.65it/s]
 50%|████▉     | 32633/65536 [5:36:13<5:29:51,  1.66it/s]
 50%|████▉     | 32634/65536 [5:36:14<5:26:40,  1.68it/s]
 50%|████▉     | 32635/65536 [5:36:15<5:28:16,  1.67it/s]
 50%|████▉     | 32636/65536 [5:36:15<5:35:48,  1.63it/s]
 50%|████▉     | 32637/65536 [5:36:16<5:45:26,  1.59it/s]
 50%|████▉     | 32638/65536 [5:36:17<5:37:21,  1.63it/s]
 50%|████▉     | 32639/65536 [5:36:17<5:41:13,  1.61it/s]
 50%|████▉     | 32640/65536 [5:36:18<5:42:21,  1.60it/s]
                                                         
{'loss': 1.6423, 'learning_rate': 5.531406881351782e-07, 'epoch': 2014.81}
+
 50%|████▉     | 32640/65536 [5:36:18<5:42:21,  1.60it/s]
 50%|████▉     | 32641/65536 [5:36:18<5:45:40,  1.59it/s]
 50%|████▉     | 32642/65536 [5:36:19<5:47:39,  1.58it/s]
 50%|████▉     | 32643/65536 [5:36:20<5:38:50,  1.62it/s]
 50%|████▉     | 32644/65536 [5:36:20<5:55:06,  1.54it/s]
 50%|████▉     | 32645/65536 [5:36:21<5:54:00,  1.55it/s]
 50%|████▉     | 32646/65536 [5:36:22<5:42:18,  1.60it/s]
 50%|████▉     | 32647/65536 [5:36:22<5:39:09,  1.62it/s]
 50%|████▉     | 32648/65536 [5:36:23<5:32:34,  1.65it/s]
 50%|████▉     | 32649/65536 [5:36:23<5:32:24,  1.65it/s]
 50%|████▉     | 32650/65536 [5:36:24<5:41:29,  1.61it/s]
 50%|████▉     | 32651/65536 [5:36:25<5:36:18,  1.63it/s]
 50%|████▉     | 32652/65536 [5:36:25<5:41:11,  1.61it/s]
 50%|████▉     | 32653/65536 [5:36:26<5:39:44,  1.61it/s]
 50%|████▉     | 32654/65536 [5:36:26<5:29:33,  1.66it/s]
 50%|████▉     | 32655/65536 [5:36:27<5:30:58,  1.66it/s]
 50%|████▉     | 32656/65536 [5:36:28<5:35:23,  1.63it/s]
 50%|████▉     | 32657/65536 [5:36:28<5:33:14,  1.64it/s]
 50%|████▉     | 32658/65536 [5:36:29<5:28:50,  1.67it/s]
 50%|████▉     | 32659/65536 [5:36:30<5:39:50,  1.61it/s]
 50%|████▉     | 32660/65536 [5:36:30<5:49:33,  1.57it/s]
                                                         
{'loss': 1.6667, 'learning_rate': 5.528651891759519e-07, 'epoch': 2016.05}
+
 50%|████▉     | 32660/65536 [5:36:30<5:49:33,  1.57it/s]
 50%|████▉     | 32661/65536 [5:36:31<5:48:02,  1.57it/s]
 50%|████▉     | 32662/65536 [5:36:32<5:57:54,  1.53it/s]
 50%|████▉     | 32663/65536 [5:36:32<5:56:49,  1.54it/s]
 50%|████▉     | 32664/65536 [5:36:33<5:50:15,  1.56it/s]
 50%|████▉     | 32665/65536 [5:36:33<5:41:50,  1.60it/s]
 50%|████▉     | 32666/65536 [5:36:34<5:40:51,  1.61it/s]
 50%|████▉     | 32667/65536 [5:36:35<5:35:33,  1.63it/s]
 50%|████▉     | 32668/65536 [5:36:35<5:36:28,  1.63it/s]
 50%|████▉     | 32669/65536 [5:36:36<5:34:48,  1.64it/s]
 50%|████▉     | 32670/65536 [5:36:36<5:40:31,  1.61it/s]
 50%|████▉     | 32671/65536 [5:36:37<5:34:32,  1.64it/s]
 50%|████▉     | 32672/65536 [5:36:38<5:27:24,  1.67it/s]
 50%|████▉     | 32673/65536 [5:36:38<5:31:05,  1.65it/s]
 50%|████▉     | 32674/65536 [5:36:39<5:30:11,  1.66it/s]
 50%|████▉     | 32675/65536 [5:36:39<5:29:21,  1.66it/s]
 50%|████▉     | 32676/65536 [5:36:40<5:45:57,  1.58it/s]
 50%|████▉     | 32677/65536 [5:36:41<5:36:15,  1.63it/s]
 50%|████▉     | 32678/65536 [5:36:41<5:32:12,  1.65it/s]
 50%|████▉     | 32679/65536 [5:36:42<5:29:50,  1.66it/s]
 50%|████▉     | 32680/65536 [5:36:42<5:26:14,  1.68it/s]
                                                         
{'loss': 1.6526, 'learning_rate': 5.525896902167258e-07, 'epoch': 2017.28}
+
 50%|████▉     | 32680/65536 [5:36:42<5:26:14,  1.68it/s]
 50%|████▉     | 32681/65536 [5:36:43<5:32:41,  1.65it/s]
 50%|████▉     | 32682/65536 [5:36:44<5:26:25,  1.68it/s]
 50%|████▉     | 32683/65536 [5:36:44<5:36:01,  1.63it/s]
 50%|████▉     | 32684/65536 [5:36:45<5:43:56,  1.59it/s]
 50%|████▉     | 32685/65536 [5:36:46<5:43:14,  1.60it/s]
 50%|████▉     | 32686/65536 [5:36:46<5:48:01,  1.57it/s]
 50%|████▉     | 32687/65536 [5:36:47<5:48:55,  1.57it/s]
 50%|████▉     | 32688/65536 [5:36:48<5:47:36,  1.57it/s]
 50%|████▉     | 32689/65536 [5:36:48<5:38:55,  1.62it/s]
 50%|████▉     | 32690/65536 [5:36:49<5:40:44,  1.61it/s]
 50%|████▉     | 32691/65536 [5:36:49<5:36:33,  1.63it/s]
 50%|████▉     | 32692/65536 [5:36:50<5:46:28,  1.58it/s]
 50%|████▉     | 32693/65536 [5:36:51<5:46:45,  1.58it/s]
 50%|████▉     | 32694/65536 [5:36:51<5:35:17,  1.63it/s]
 50%|████▉     | 32695/65536 [5:36:52<5:34:47,  1.63it/s]
 50%|████▉     | 32696/65536 [5:36:52<5:42:22,  1.60it/s]
 50%|████▉     | 32697/65536 [5:36:53<5:51:28,  1.56it/s]
 50%|████▉     | 32698/65536 [5:36:54<5:44:11,  1.59it/s]
 50%|████▉     | 32699/65536 [5:36:54<5:43:22,  1.59it/s]
 50%|████▉     | 32700/65536 [5:36:55<5:42:58,  1.60it/s]
                                                         
{'loss': 1.6486, 'learning_rate': 5.523141912574996e-07, 'epoch': 2018.52}
+
 50%|████▉     | 32700/65536 [5:36:55<5:42:58,  1.60it/s]
 50%|████▉     | 32701/65536 [5:36:56<5:44:21,  1.59it/s]
 50%|████▉     | 32702/65536 [5:36:56<5:38:57,  1.61it/s]
 50%|████▉     | 32703/65536 [5:36:57<5:34:17,  1.64it/s]
 50%|████▉     | 32704/65536 [5:36:57<5:34:14,  1.64it/s]
 50%|████▉     | 32705/65536 [5:36:58<5:32:41,  1.64it/s]
 50%|████▉     | 32706/65536 [5:36:59<5:40:29,  1.61it/s]
 50%|████▉     | 32707/65536 [5:36:59<5:39:29,  1.61it/s]
 50%|████▉     | 32708/65536 [5:37:00<5:47:47,  1.57it/s]
 50%|████▉     | 32709/65536 [5:37:01<5:42:29,  1.60it/s]
 50%|████▉     | 32710/65536 [5:37:01<5:38:52,  1.61it/s]
 50%|████▉     | 32711/65536 [5:37:02<5:37:02,  1.62it/s]
 50%|████▉     | 32712/65536 [5:37:02<5:35:49,  1.63it/s]
 50%|████▉     | 32713/65536 [5:37:03<5:33:39,  1.64it/s]
 50%|████▉     | 32714/65536 [5:37:04<5:32:05,  1.65it/s]
 50%|████▉     | 32715/65536 [5:37:04<5:37:57,  1.62it/s]
 50%|████▉     | 32716/65536 [5:37:05<5:32:35,  1.64it/s]
 50%|████▉     | 32717/65536 [5:37:05<5:34:49,  1.63it/s]
 50%|████▉     | 32718/65536 [5:37:06<5:30:18,  1.66it/s]
 50%|████▉     | 32719/65536 [5:37:07<5:31:44,  1.65it/s]
 50%|████▉     | 32720/65536 [5:37:07<5:38:25,  1.62it/s]
                                                         
{'loss': 1.6536, 'learning_rate': 5.520386922982735e-07, 'epoch': 2019.75}
+
 50%|████▉     | 32720/65536 [5:37:07<5:38:25,  1.62it/s]
 50%|████▉     | 32721/65536 [5:37:08<5:42:57,  1.59it/s]
 50%|████▉     | 32722/65536 [5:37:09<5:41:56,  1.60it/s]
 50%|████▉     | 32723/65536 [5:37:09<5:36:26,  1.63it/s]
 50%|████▉     | 32724/65536 [5:37:10<5:29:50,  1.66it/s]
 50%|████▉     | 32725/65536 [5:37:10<5:51:27,  1.56it/s]
 50%|████▉     | 32726/65536 [5:37:11<5:41:00,  1.60it/s]
 50%|████▉     | 32727/65536 [5:37:12<5:43:54,  1.59it/s]
 50%|████▉     | 32728/65536 [5:37:12<5:29:03,  1.66it/s]
 50%|████▉     | 32729/65536 [5:37:13<5:32:17,  1.65it/s]
 50%|████▉     | 32730/65536 [5:37:14<5:38:15,  1.62it/s]
 50%|████▉     | 32731/65536 [5:37:14<5:47:08,  1.57it/s]
 50%|████▉     | 32732/65536 [5:37:15<5:47:32,  1.57it/s]
 50%|████▉     | 32733/65536 [5:37:15<5:35:47,  1.63it/s]
 50%|████▉     | 32734/65536 [5:37:16<5:38:10,  1.62it/s]
 50%|████▉     | 32735/65536 [5:37:17<5:46:09,  1.58it/s]
 50%|████▉     | 32736/65536 [5:37:17<5:37:48,  1.62it/s]
 50%|████▉     | 32737/65536 [5:37:18<5:39:44,  1.61it/s]
 50%|████▉     | 32738/65536 [5:37:18<5:35:26,  1.63it/s]
 50%|████▉     | 32739/65536 [5:37:19<5:33:25,  1.64it/s]
 50%|████▉     | 32740/65536 [5:37:20<5:33:01,  1.64it/s]
                                                         
{'loss': 1.6297, 'learning_rate': 5.517631933390473e-07, 'epoch': 2020.99}
+
 50%|████▉     | 32740/65536 [5:37:20<5:33:01,  1.64it/s]
 50%|████▉     | 32741/65536 [5:37:20<5:46:16,  1.58it/s]
 50%|████▉     | 32742/65536 [5:37:21<5:45:05,  1.58it/s]
 50%|████▉     | 32743/65536 [5:37:22<5:45:06,  1.58it/s]
 50%|████▉     | 32744/65536 [5:37:22<5:49:42,  1.56it/s]
 50%|████▉     | 32745/65536 [5:37:23<5:41:42,  1.60it/s]
 50%|████▉     | 32746/65536 [5:37:23<5:36:44,  1.62it/s]
 50%|████▉     | 32747/65536 [5:37:24<5:40:12,  1.61it/s]
 50%|████▉     | 32748/65536 [5:37:25<5:39:44,  1.61it/s]
 50%|████▉     | 32749/65536 [5:37:25<5:36:20,  1.62it/s]
 50%|████▉     | 32750/65536 [5:37:26<5:45:41,  1.58it/s]
 50%|████▉     | 32751/65536 [5:37:27<5:44:48,  1.58it/s]
 50%|████▉     | 32752/65536 [5:37:27<5:41:56,  1.60it/s]
 50%|████▉     | 32753/65536 [5:37:28<5:39:14,  1.61it/s]
 50%|████▉     | 32754/65536 [5:37:28<5:38:04,  1.62it/s]
 50%|████▉     | 32755/65536 [5:37:29<5:33:00,  1.64it/s]
 50%|████▉     | 32756/65536 [5:37:30<5:32:36,  1.64it/s]
 50%|████▉     | 32757/65536 [5:37:30<5:50:32,  1.56it/s]
 50%|████▉     | 32758/65536 [5:37:31<5:45:00,  1.58it/s]
 50%|████▉     | 32759/65536 [5:37:32<5:37:27,  1.62it/s]
 50%|████▉     | 32760/65536 [5:37:32<5:36:55,  1.62it/s]
                                                         
{'loss': 1.648, 'learning_rate': 5.514876943798212e-07, 'epoch': 2022.22}
+
 50%|████▉     | 32760/65536 [5:37:32<5:36:55,  1.62it/s]
 50%|████▉     | 32761/65536 [5:37:33<5:37:49,  1.62it/s]
 50%|████▉     | 32762/65536 [5:37:33<5:34:29,  1.63it/s]
 50%|████▉     | 32763/65536 [5:37:34<5:40:06,  1.61it/s]
 50%|████▉     | 32764/65536 [5:37:35<5:30:51,  1.65it/s]
 50%|████▉     | 32765/65536 [5:37:35<5:42:06,  1.60it/s]
 50%|████▉     | 32766/65536 [5:37:36<5:44:35,  1.58it/s]
 50%|████▉     | 32767/65536 [5:37:37<5:40:38,  1.60it/s]
 50%|█████     | 32768/65536 [5:37:37<5:37:33,  1.62it/s]
 50%|█████     | 32769/65536 [5:37:38<5:40:21,  1.60it/s]
 50%|█████     | 32770/65536 [5:37:38<5:33:29,  1.64it/s]
 50%|█████     | 32771/65536 [5:37:39<5:36:07,  1.62it/s]
 50%|█████     | 32772/65536 [5:37:40<5:37:16,  1.62it/s]
 50%|█████     | 32773/65536 [5:37:40<5:46:51,  1.57it/s]
 50%|█████     | 32774/65536 [5:37:41<5:44:30,  1.59it/s]
 50%|█████     | 32775/65536 [5:37:42<5:39:55,  1.61it/s]
 50%|█████     | 32776/65536 [5:37:42<5:35:30,  1.63it/s]
 50%|█████     | 32777/65536 [5:37:43<5:33:53,  1.64it/s]
 50%|█████     | 32778/65536 [5:37:43<5:28:12,  1.66it/s]
 50%|█████     | 32779/65536 [5:37:44<5:43:01,  1.59it/s]
 50%|█████     | 32780/65536 [5:37:45<5:38:43,  1.61it/s]
                                                         
{'loss': 1.632, 'learning_rate': 5.51212195420595e-07, 'epoch': 2023.46}
+
 50%|█████     | 32780/65536 [5:37:45<5:38:43,  1.61it/s]
 50%|█████     | 32781/65536 [5:37:45<5:33:25,  1.64it/s]
 50%|█████     | 32782/65536 [5:37:46<5:35:22,  1.63it/s]
 50%|█████     | 32783/65536 [5:37:46<5:41:25,  1.60it/s]
 50%|█████     | 32784/65536 [5:37:47<5:35:59,  1.62it/s]
 50%|█████     | 32785/65536 [5:37:48<5:42:29,  1.59it/s]
 50%|█████     | 32786/65536 [5:37:48<5:36:56,  1.62it/s]
 50%|█████     | 32787/65536 [5:37:49<5:40:34,  1.60it/s]
 50%|█████     | 32788/65536 [5:37:50<5:35:43,  1.63it/s]
 50%|█████     | 32789/65536 [5:37:50<5:43:39,  1.59it/s]
 50%|█████     | 32790/65536 [5:37:51<5:32:37,  1.64it/s]
 50%|█████     | 32791/65536 [5:37:51<5:29:22,  1.66it/s]
 50%|█████     | 32792/65536 [5:37:52<5:32:24,  1.64it/s]
 50%|█████     | 32793/65536 [5:37:53<5:38:26,  1.61it/s]
 50%|█████     | 32794/65536 [5:37:53<5:37:41,  1.62it/s]
 50%|█████     | 32795/65536 [5:37:54<5:29:44,  1.65it/s]
 50%|█████     | 32796/65536 [5:37:54<5:33:58,  1.63it/s]
 50%|█████     | 32797/65536 [5:37:55<5:46:10,  1.58it/s]
 50%|█████     | 32798/65536 [5:37:56<5:42:41,  1.59it/s]
 50%|█████     | 32799/65536 [5:37:56<5:37:16,  1.62it/s]
 50%|█████     | 32800/65536 [5:37:57<5:28:49,  1.66it/s]
                                                         
{'loss': 1.6787, 'learning_rate': 5.509366964613689e-07, 'epoch': 2024.69}
+
 50%|█████     | 32800/65536 [5:37:57<5:28:49,  1.66it/s]
 50%|█████     | 32801/65536 [5:37:58<5:30:15,  1.65it/s]
 50%|█████     | 32802/65536 [5:37:58<5:38:17,  1.61it/s]
 50%|█████     | 32803/65536 [5:37:59<5:33:16,  1.64it/s]
 50%|█████     | 32804/65536 [5:37:59<5:36:32,  1.62it/s]
 50%|█████     | 32805/65536 [5:38:00<5:31:59,  1.64it/s]
 50%|█████     | 32806/65536 [5:38:01<5:43:36,  1.59it/s]
 50%|█████     | 32807/65536 [5:38:01<5:44:57,  1.58it/s]
 50%|█████     | 32808/65536 [5:38:02<5:39:42,  1.61it/s]
 50%|█████     | 32809/65536 [5:38:03<5:38:47,  1.61it/s]
 50%|█████     | 32810/65536 [5:38:03<5:43:12,  1.59it/s]
 50%|█████     | 32811/65536 [5:38:04<5:34:45,  1.63it/s]
 50%|█████     | 32812/65536 [5:38:04<5:24:52,  1.68it/s]
 50%|█████     | 32813/65536 [5:38:05<5:32:31,  1.64it/s]
 50%|█████     | 32814/65536 [5:38:06<5:37:03,  1.62it/s]
 50%|█████     | 32815/65536 [5:38:06<5:29:29,  1.66it/s]
 50%|█████     | 32816/65536 [5:38:07<5:27:24,  1.67it/s]
 50%|█████     | 32817/65536 [5:38:07<5:33:30,  1.64it/s]
 50%|█████     | 32818/65536 [5:38:08<5:33:15,  1.64it/s]
 50%|█████     | 32819/65536 [5:38:09<5:33:14,  1.64it/s]
 50%|█████     | 32820/65536 [5:38:09<5:43:34,  1.59it/s]
                                                         
{'loss': 1.6219, 'learning_rate': 5.506611975021428e-07, 'epoch': 2025.93}
+
 50%|█████     | 32820/65536 [5:38:09<5:43:34,  1.59it/s]
 50%|█████     | 32821/65536 [5:38:10<5:42:31,  1.59it/s]
 50%|█████     | 32822/65536 [5:38:11<5:47:35,  1.57it/s]
 50%|█████     | 32823/65536 [5:38:11<5:47:17,  1.57it/s]
 50%|█████     | 32824/65536 [5:38:12<5:45:07,  1.58it/s]
 50%|█████     | 32825/65536 [5:38:12<5:44:55,  1.58it/s]
 50%|█████     | 32826/65536 [5:38:13<5:44:20,  1.58it/s]
 50%|█████     | 32827/65536 [5:38:14<5:35:23,  1.63it/s]
 50%|█████     | 32828/65536 [5:38:14<5:33:29,  1.63it/s]
 50%|█████     | 32829/65536 [5:38:15<5:38:08,  1.61it/s]
 50%|█████     | 32830/65536 [5:38:16<5:36:53,  1.62it/s]
 50%|█████     | 32831/65536 [5:38:16<5:41:56,  1.59it/s]
 50%|█████     | 32832/65536 [5:38:17<5:41:31,  1.60it/s]
 50%|█████     | 32833/65536 [5:38:17<5:45:13,  1.58it/s]
 50%|█████     | 32834/65536 [5:38:18<5:38:28,  1.61it/s]
 50%|█████     | 32835/65536 [5:38:19<5:33:46,  1.63it/s]
 50%|█████     | 32836/65536 [5:38:19<5:31:51,  1.64it/s]
 50%|█████     | 32837/65536 [5:38:20<5:31:17,  1.64it/s]
 50%|█████     | 32838/65536 [5:38:20<5:41:05,  1.60it/s]
 50%|█████     | 32839/65536 [5:38:21<5:37:26,  1.61it/s]
 50%|█████     | 32840/65536 [5:38:22<5:32:30,  1.64it/s]
                                                         
{'loss': 1.6477, 'learning_rate': 5.503856985429166e-07, 'epoch': 2027.16}
+
 50%|█████     | 32840/65536 [5:38:22<5:32:30,  1.64it/s]
 50%|█████     | 32841/65536 [5:38:22<5:35:04,  1.63it/s]
 50%|█████     | 32842/65536 [5:38:23<5:31:44,  1.64it/s]
 50%|█████     | 32843/65536 [5:38:24<5:31:56,  1.64it/s]
 50%|█████     | 32844/65536 [5:38:24<5:30:46,  1.65it/s]
 50%|█████     | 32845/65536 [5:38:25<5:29:21,  1.65it/s]
 50%|█████     | 32846/65536 [5:38:25<5:33:52,  1.63it/s]
 50%|█████     | 32847/65536 [5:38:26<5:31:18,  1.64it/s]
 50%|█████     | 32848/65536 [5:38:27<5:41:28,  1.60it/s]
 50%|█████     | 32849/65536 [5:38:27<5:31:53,  1.64it/s]
 50%|█████     | 32850/65536 [5:38:28<5:38:24,  1.61it/s]
 50%|█████     | 32851/65536 [5:38:28<5:35:29,  1.62it/s]
 50%|█████     | 32852/65536 [5:38:29<5:40:49,  1.60it/s]
 50%|█████     | 32853/65536 [5:38:30<5:35:15,  1.62it/s]
 50%|█████     | 32854/65536 [5:38:30<5:42:27,  1.59it/s]
 50%|█████     | 32855/65536 [5:38:31<5:42:43,  1.59it/s]
 50%|█████     | 32856/65536 [5:38:32<5:38:23,  1.61it/s]
 50%|█████     | 32857/65536 [5:38:32<5:39:28,  1.60it/s]
 50%|█████     | 32858/65536 [5:38:33<6:00:34,  1.51it/s]
 50%|█████     | 32859/65536 [5:38:34<6:00:01,  1.51it/s]
 50%|█████     | 32860/65536 [5:38:34<5:52:19,  1.55it/s]
                                                         
{'loss': 1.6809, 'learning_rate': 5.501101995836905e-07, 'epoch': 2028.4}
+
 50%|█████     | 32860/65536 [5:38:34<5:52:19,  1.55it/s]
 50%|█████     | 32861/65536 [5:38:35<5:43:28,  1.59it/s]
 50%|█████     | 32862/65536 [5:38:36<5:51:53,  1.55it/s]
 50%|█████     | 32863/65536 [5:38:36<5:45:17,  1.58it/s]
 50%|█████     | 32864/65536 [5:38:37<5:49:52,  1.56it/s]
 50%|█████     | 32865/65536 [5:38:37<5:49:52,  1.56it/s]
 50%|█████     | 32866/65536 [5:38:38<5:44:38,  1.58it/s]
 50%|█████     | 32867/65536 [5:38:39<5:35:24,  1.62it/s]
 50%|█████     | 32868/65536 [5:38:39<5:27:39,  1.66it/s]
 50%|█████     | 32869/65536 [5:38:40<5:31:44,  1.64it/s]
 50%|█████     | 32870/65536 [5:38:40<5:46:03,  1.57it/s]
 50%|█████     | 32871/65536 [5:38:41<5:40:49,  1.60it/s]
 50%|█████     | 32872/65536 [5:38:42<5:44:54,  1.58it/s]
 50%|█████     | 32873/65536 [5:38:42<5:40:10,  1.60it/s]
 50%|█████     | 32874/65536 [5:38:43<5:34:48,  1.63it/s]
 50%|█████     | 32875/65536 [5:38:44<5:34:58,  1.63it/s]
 50%|█████     | 32876/65536 [5:38:44<5:30:29,  1.65it/s]
 50%|█████     | 32877/65536 [5:38:45<5:29:24,  1.65it/s]
 50%|█████     | 32878/65536 [5:38:45<5:29:38,  1.65it/s]
 50%|█████     | 32879/65536 [5:38:46<5:33:04,  1.63it/s]
 50%|█████     | 32880/65536 [5:38:47<5:37:50,  1.61it/s]
                                                         
{'loss': 1.6266, 'learning_rate': 5.498347006244643e-07, 'epoch': 2029.63}
+
 50%|█████     | 32880/65536 [5:38:47<5:37:50,  1.61it/s]
 50%|█████     | 32881/65536 [5:38:47<5:31:58,  1.64it/s]
 50%|█████     | 32882/65536 [5:38:48<5:29:11,  1.65it/s]
 50%|█████     | 32883/65536 [5:38:48<5:31:42,  1.64it/s]
 50%|█████     | 32884/65536 [5:38:49<5:38:06,  1.61it/s]
 50%|█████     | 32885/65536 [5:38:50<5:43:03,  1.59it/s]
 50%|█████     | 32886/65536 [5:38:50<5:33:15,  1.63it/s]
 50%|█████     | 32887/65536 [5:38:51<5:41:08,  1.60it/s]
 50%|█████     | 32888/65536 [5:38:52<5:33:43,  1.63it/s]
 50%|█████     | 32889/65536 [5:38:52<5:34:35,  1.63it/s]
 50%|█████     | 32890/65536 [5:38:53<5:33:22,  1.63it/s]
 50%|█████     | 32891/65536 [5:38:53<5:46:30,  1.57it/s]
 50%|█████     | 32892/65536 [5:38:54<5:47:13,  1.57it/s]
 50%|█████     | 32893/65536 [5:38:55<5:39:10,  1.60it/s]
 50%|█████     | 32894/65536 [5:38:55<5:34:24,  1.63it/s]
 50%|█████     | 32895/65536 [5:38:56<5:34:42,  1.63it/s]
 50%|█████     | 32896/65536 [5:38:57<5:42:13,  1.59it/s]
 50%|█████     | 32897/65536 [5:38:57<5:39:19,  1.60it/s]
 50%|█████     | 32898/65536 [5:38:58<5:37:41,  1.61it/s]
 50%|█████     | 32899/65536 [5:38:58<5:43:09,  1.59it/s]
 50%|█████     | 32900/65536 [5:38:59<5:32:04,  1.64it/s]
                                                         
{'loss': 1.6525, 'learning_rate': 5.495592016652381e-07, 'epoch': 2030.86}
+
 50%|█████     | 32900/65536 [5:38:59<5:32:04,  1.64it/s]
 50%|█████     | 32901/65536 [5:39:00<5:28:20,  1.66it/s]
 50%|█████     | 32902/65536 [5:39:00<5:24:31,  1.68it/s]
 50%|█████     | 32903/65536 [5:39:01<5:33:09,  1.63it/s]
 50%|█████     | 32904/65536 [5:39:01<5:40:33,  1.60it/s]
 50%|█████     | 32905/65536 [5:39:02<5:39:27,  1.60it/s]
 50%|█████     | 32906/65536 [5:39:03<5:41:51,  1.59it/s]
 50%|█████     | 32907/65536 [5:39:03<5:41:03,  1.59it/s]
 50%|█████     | 32908/65536 [5:39:04<5:38:33,  1.61it/s]
 50%|█████     | 32909/65536 [5:39:05<5:41:11,  1.59it/s]
 50%|█████     | 32910/65536 [5:39:05<5:37:23,  1.61it/s]
 50%|█████     | 32911/65536 [5:39:06<5:47:59,  1.56it/s]
 50%|█████     | 32912/65536 [5:39:07<5:41:32,  1.59it/s]
 50%|█████     | 32913/65536 [5:39:07<5:36:24,  1.62it/s]
 50%|█████     | 32914/65536 [5:39:08<5:42:44,  1.59it/s]
 50%|█████     | 32915/65536 [5:39:08<5:47:24,  1.56it/s]
 50%|█████     | 32916/65536 [5:39:09<5:44:27,  1.58it/s]
 50%|█████     | 32917/65536 [5:39:10<5:30:44,  1.64it/s]
 50%|█████     | 32918/65536 [5:39:10<5:21:55,  1.69it/s]
 50%|█████     | 32919/65536 [5:39:11<5:31:30,  1.64it/s]
 50%|█████     | 32920/65536 [5:39:11<5:30:10,  1.65it/s]
                                                         
{'loss': 1.6831, 'learning_rate': 5.49283702706012e-07, 'epoch': 2032.1}
+
 50%|█████     | 32920/65536 [5:39:11<5:30:10,  1.65it/s]
 50%|█████     | 32921/65536 [5:39:12<5:32:41,  1.63it/s]
 50%|█████     | 32922/65536 [5:39:13<5:28:36,  1.65it/s]
 50%|█████     | 32923/65536 [5:39:13<5:35:29,  1.62it/s]
 50%|█████     | 32924/65536 [5:39:14<5:35:07,  1.62it/s]
 50%|█████     | 32925/65536 [5:39:14<5:29:22,  1.65it/s]
 50%|█████     | 32926/65536 [5:39:15<5:34:49,  1.62it/s]
 50%|█████     | 32927/65536 [5:39:16<5:30:18,  1.65it/s]
 50%|█████     | 32928/65536 [5:39:16<5:27:20,  1.66it/s]
 50%|█████     | 32929/65536 [5:39:17<5:38:09,  1.61it/s]
 50%|█████     | 32930/65536 [5:39:18<5:39:10,  1.60it/s]
 50%|█████     | 32931/65536 [5:39:18<5:33:14,  1.63it/s]
 50%|█████     | 32932/65536 [5:39:19<5:32:23,  1.63it/s]
 50%|█████     | 32933/65536 [5:39:19<5:36:39,  1.61it/s]
 50%|█████     | 32934/65536 [5:39:20<5:33:30,  1.63it/s]
 50%|█████     | 32935/65536 [5:39:21<5:48:22,  1.56it/s]
 50%|█████     | 32936/65536 [5:39:21<5:43:01,  1.58it/s]
 50%|█████     | 32937/65536 [5:39:22<5:34:14,  1.63it/s]
 50%|█████     | 32938/65536 [5:39:23<5:40:43,  1.59it/s]
 50%|█████     | 32939/65536 [5:39:23<5:35:32,  1.62it/s]
 50%|█████     | 32940/65536 [5:39:24<5:26:48,  1.66it/s]
                                                         
{'loss': 1.6116, 'learning_rate': 5.490082037467858e-07, 'epoch': 2033.33}
+
 50%|█████     | 32940/65536 [5:39:24<5:26:48,  1.66it/s]
 50%|█████     | 32941/65536 [5:39:24<5:27:10,  1.66it/s]
 50%|█████     | 32942/65536 [5:39:25<5:32:56,  1.63it/s]
 50%|█████     | 32943/65536 [5:39:26<5:32:35,  1.63it/s]
 50%|█████     | 32944/65536 [5:39:26<5:27:16,  1.66it/s]
 50%|█████     | 32945/65536 [5:39:27<5:25:59,  1.67it/s]
 50%|█████     | 32946/65536 [5:39:27<5:33:09,  1.63it/s]
 50%|█████     | 32947/65536 [5:39:28<5:33:15,  1.63it/s]
 50%|█████     | 32948/65536 [5:39:29<5:29:03,  1.65it/s]
 50%|█████     | 32949/65536 [5:39:29<5:36:52,  1.61it/s]
 50%|█████     | 32950/65536 [5:39:30<5:41:26,  1.59it/s]
 50%|█████     | 32951/65536 [5:39:31<5:55:07,  1.53it/s]
 50%|█████     | 32952/65536 [5:39:31<5:47:45,  1.56it/s]
 50%|█████     | 32953/65536 [5:39:32<5:35:51,  1.62it/s]
 50%|█████     | 32954/65536 [5:39:32<5:32:46,  1.63it/s]
 50%|█████     | 32955/65536 [5:39:33<5:35:29,  1.62it/s]
 50%|█████     | 32956/65536 [5:39:34<5:42:27,  1.59it/s]
 50%|█████     | 32957/65536 [5:39:34<5:29:54,  1.65it/s]
 50%|█████     | 32958/65536 [5:39:35<5:27:08,  1.66it/s]
 50%|█████     | 32959/65536 [5:39:35<5:36:42,  1.61it/s]
 50%|█████     | 32960/65536 [5:39:36<5:48:25,  1.56it/s]
                                                         
{'loss': 1.6331, 'learning_rate': 5.487327047875596e-07, 'epoch': 2034.57}
+
 50%|█████     | 32960/65536 [5:39:36<5:48:25,  1.56it/s]
 50%|█████     | 32961/65536 [5:39:37<5:43:38,  1.58it/s]
 50%|█████     | 32962/65536 [5:39:37<5:43:05,  1.58it/s]
 50%|█████     | 32963/65536 [5:39:38<5:40:19,  1.60it/s]
 50%|█████     | 32964/65536 [5:39:39<5:36:31,  1.61it/s]
 50%|█████     | 32965/65536 [5:39:39<5:37:37,  1.61it/s]
 50%|█████     | 32966/65536 [5:39:40<5:40:58,  1.59it/s]
 50%|█████     | 32967/65536 [5:39:40<5:31:38,  1.64it/s]
 50%|█████     | 32968/65536 [5:39:41<5:47:46,  1.56it/s]
 50%|█████     | 32969/65536 [5:39:42<5:40:15,  1.60it/s]
 50%|█████     | 32970/65536 [5:39:42<5:42:52,  1.58it/s]
 50%|█████     | 32971/65536 [5:39:43<5:37:14,  1.61it/s]
 50%|█████     | 32972/65536 [5:39:44<5:37:06,  1.61it/s]
 50%|█████     | 32973/65536 [5:39:44<5:33:58,  1.63it/s]
 50%|█████     | 32974/65536 [5:39:45<5:29:00,  1.65it/s]
 50%|█████     | 32975/65536 [5:39:45<5:30:19,  1.64it/s]
 50%|█████     | 32976/65536 [5:39:46<5:31:52,  1.64it/s]
 50%|█████     | 32977/65536 [5:39:47<5:36:56,  1.61it/s]
 50%|█████     | 32978/65536 [5:39:47<5:31:50,  1.64it/s]
 50%|█████     | 32979/65536 [5:39:48<5:35:53,  1.62it/s]
 50%|█████     | 32980/65536 [5:39:49<5:35:09,  1.62it/s]
                                                         
{'loss': 1.6585, 'learning_rate': 5.484572058283335e-07, 'epoch': 2035.8}
+
 50%|█████     | 32980/65536 [5:39:49<5:35:09,  1.62it/s]
 50%|█████     | 32981/65536 [5:39:49<5:33:43,  1.63it/s]
 50%|█████     | 32982/65536 [5:39:50<5:24:37,  1.67it/s]
 50%|█████     | 32983/65536 [5:39:50<5:35:33,  1.62it/s]
 50%|█████     | 32984/65536 [5:39:51<5:43:03,  1.58it/s]
 50%|█████     | 32985/65536 [5:39:52<5:44:06,  1.58it/s]
 50%|█████     | 32986/65536 [5:39:52<5:49:21,  1.55it/s]
 50%|█████     | 32987/65536 [5:39:53<5:43:10,  1.58it/s]
 50%|█████     | 32988/65536 [5:39:53<5:31:51,  1.63it/s]
 50%|█████     | 32989/65536 [5:39:54<5:26:20,  1.66it/s]
 50%|█████     | 32990/65536 [5:39:55<5:34:22,  1.62it/s]
 50%|█████     | 32991/65536 [5:39:55<5:25:34,  1.67it/s]
 50%|█████     | 32992/65536 [5:39:56<5:28:37,  1.65it/s]
 50%|█████     | 32993/65536 [5:39:57<5:31:27,  1.64it/s]
 50%|█████     | 32994/65536 [5:39:57<5:34:27,  1.62it/s]
 50%|█████     | 32995/65536 [5:39:58<5:30:14,  1.64it/s]
 50%|█████     | 32996/65536 [5:39:58<5:26:26,  1.66it/s]
 50%|█████     | 32997/65536 [5:39:59<5:28:26,  1.65it/s]
 50%|█████     | 32998/65536 [5:40:00<5:26:02,  1.66it/s]
 50%|█████     | 32999/65536 [5:40:00<5:37:11,  1.61it/s]
 50%|█████     | 33000/65536 [5:40:01<5:43:18,  1.58it/s]
                                                         
{'loss': 1.6951, 'learning_rate': 5.481817068691074e-07, 'epoch': 2037.04}
+
 50%|█████     | 33000/65536 [5:40:01<5:43:18,  1.58it/s]
 50%|█████     | 33001/65536 [5:40:01<5:34:38,  1.62it/s]
 50%|█████     | 33002/65536 [5:40:02<5:34:23,  1.62it/s]
 50%|█████     | 33003/65536 [5:40:03<5:30:06,  1.64it/s]
 50%|█████     | 33004/65536 [5:40:03<5:25:43,  1.66it/s]
 50%|█████     | 33005/65536 [5:40:04<5:18:48,  1.70it/s]
 50%|█████     | 33006/65536 [5:40:04<5:24:58,  1.67it/s]
 50%|█████     | 33007/65536 [5:40:05<5:24:49,  1.67it/s]
 50%|█████     | 33008/65536 [5:40:06<5:30:34,  1.64it/s]
 50%|█████     | 33009/65536 [5:40:06<5:42:19,  1.58it/s]
 50%|█████     | 33010/65536 [5:40:07<5:39:41,  1.60it/s]
 50%|█████     | 33011/65536 [5:40:08<5:38:30,  1.60it/s]
 50%|█████     | 33012/65536 [5:40:08<5:48:45,  1.55it/s]
 50%|█████     | 33013/65536 [5:40:09<5:37:17,  1.61it/s]
 50%|█████     | 33014/65536 [5:40:09<5:43:28,  1.58it/s]
 50%|█████     | 33015/65536 [5:40:10<5:38:52,  1.60it/s]
 50%|█████     | 33016/65536 [5:40:11<5:47:18,  1.56it/s]
 50%|█████     | 33017/65536 [5:40:11<5:41:55,  1.59it/s]
 50%|█████     | 33018/65536 [5:40:12<5:47:35,  1.56it/s]
 50%|█████     | 33019/65536 [5:40:13<5:49:37,  1.55it/s]
 50%|█████     | 33020/65536 [5:40:13<5:39:48,  1.59it/s]
                                                         
{'loss': 1.7036, 'learning_rate': 5.479062079098811e-07, 'epoch': 2038.27}
+
 50%|█████     | 33020/65536 [5:40:13<5:39:48,  1.59it/s]
 50%|█████     | 33021/65536 [5:40:14<5:39:32,  1.60it/s]
 50%|█████     | 33022/65536 [5:40:15<5:34:27,  1.62it/s]
 50%|█████     | 33023/65536 [5:40:15<5:28:42,  1.65it/s]
 50%|█████     | 33024/65536 [5:40:16<5:32:16,  1.63it/s]
 50%|█████     | 33025/65536 [5:40:16<5:30:58,  1.64it/s]
 50%|█████     | 33026/65536 [5:40:17<5:30:41,  1.64it/s]
 50%|█████     | 33027/65536 [5:40:18<5:27:40,  1.65it/s]
 50%|█████     | 33028/65536 [5:40:18<5:35:14,  1.62it/s]
 50%|█████     | 33029/65536 [5:40:19<5:34:18,  1.62it/s]
 50%|█████     | 33030/65536 [5:40:19<5:31:54,  1.63it/s]
 50%|█████     | 33031/65536 [5:40:20<5:34:26,  1.62it/s]
 50%|█████     | 33032/65536 [5:40:21<5:42:36,  1.58it/s]
 50%|█████     | 33033/65536 [5:40:21<5:41:09,  1.59it/s]
 50%|█████     | 33034/65536 [5:40:22<5:36:45,  1.61it/s]
 50%|█████     | 33035/65536 [5:40:23<5:33:23,  1.62it/s]
 50%|█████     | 33036/65536 [5:40:23<5:26:55,  1.66it/s]
 50%|█████     | 33037/65536 [5:40:24<5:26:14,  1.66it/s]
 50%|█████     | 33038/65536 [5:40:24<5:24:53,  1.67it/s]
 50%|█████     | 33039/65536 [5:40:25<5:25:40,  1.66it/s]
 50%|█████     | 33040/65536 [5:40:25<5:27:11,  1.66it/s]
                                                         
{'loss': 1.6319, 'learning_rate': 5.47630708950655e-07, 'epoch': 2039.51}
+
 50%|█████     | 33040/65536 [5:40:26<5:27:11,  1.66it/s]
 50%|█████     | 33041/65536 [5:40:26<5:24:06,  1.67it/s]
 50%|█████     | 33042/65536 [5:40:27<5:27:57,  1.65it/s]
 50%|█████     | 33043/65536 [5:40:27<5:32:10,  1.63it/s]
 50%|█████     | 33044/65536 [5:40:28<5:41:08,  1.59it/s]
 50%|█████     | 33045/65536 [5:40:29<5:40:02,  1.59it/s]
 50%|█████     | 33046/65536 [5:40:29<5:33:36,  1.62it/s]
 50%|█████     | 33047/65536 [5:40:30<5:37:12,  1.61it/s]
 50%|█████     | 33048/65536 [5:40:31<5:41:31,  1.59it/s]
 50%|█████     | 33049/65536 [5:40:31<5:48:24,  1.55it/s]
 50%|█████     | 33050/65536 [5:40:32<5:39:06,  1.60it/s]
 50%|█████     | 33051/65536 [5:40:32<5:41:25,  1.59it/s]
 50%|█████     | 33052/65536 [5:40:33<5:46:17,  1.56it/s]
 50%|█████     | 33053/65536 [5:40:34<5:47:39,  1.56it/s]
 50%|█████     | 33054/65536 [5:40:34<5:44:01,  1.57it/s]
 50%|█████     | 33055/65536 [5:40:35<5:34:25,  1.62it/s]
 50%|█████     | 33056/65536 [5:40:36<5:37:34,  1.60it/s]
 50%|█████     | 33057/65536 [5:40:36<5:34:44,  1.62it/s]
 50%|█████     | 33058/65536 [5:40:37<5:37:32,  1.60it/s]
 50%|█████     | 33059/65536 [5:40:37<5:45:37,  1.57it/s]
 50%|█████     | 33060/65536 [5:40:38<5:35:37,  1.61it/s]
                                                         
{'loss': 1.6505, 'learning_rate': 5.473552099914289e-07, 'epoch': 2040.74}
+
 50%|█████     | 33060/65536 [5:40:38<5:35:37,  1.61it/s]
 50%|█████     | 33061/65536 [5:40:39<5:33:59,  1.62it/s]
 50%|█████     | 33062/65536 [5:40:39<5:31:04,  1.63it/s]
 50%|█████     | 33063/65536 [5:40:40<5:27:12,  1.65it/s]
 50%|█████     | 33064/65536 [5:40:40<5:27:21,  1.65it/s]
 50%|█████     | 33065/65536 [5:40:41<5:36:00,  1.61it/s]
 50%|█████     | 33066/65536 [5:40:42<5:34:53,  1.62it/s]
 50%|█████     | 33067/65536 [5:40:42<5:33:00,  1.63it/s]
 50%|█████     | 33068/65536 [5:40:43<5:39:42,  1.59it/s]
 50%|█████     | 33069/65536 [5:40:44<5:48:59,  1.55it/s]
 50%|█████     | 33070/65536 [5:40:44<5:49:53,  1.55it/s]
 50%|█████     | 33071/65536 [5:40:45<5:43:53,  1.57it/s]
 50%|█████     | 33072/65536 [5:40:46<5:41:10,  1.59it/s]
 50%|█████     | 33073/65536 [5:40:46<5:48:33,  1.55it/s]
 50%|█████     | 33074/65536 [5:40:47<5:46:02,  1.56it/s]
 50%|█████     | 33075/65536 [5:40:47<5:44:01,  1.57it/s]
 50%|█████     | 33076/65536 [5:40:48<5:39:23,  1.59it/s]
 50%|█████     | 33077/65536 [5:40:49<5:31:57,  1.63it/s]
 50%|█████     | 33078/65536 [5:40:49<5:25:04,  1.66it/s]
 50%|█████     | 33079/65536 [5:40:50<5:17:07,  1.71it/s]
 50%|█████     | 33080/65536 [5:40:50<5:20:08,  1.69it/s]
                                                         
{'loss': 1.6459, 'learning_rate': 5.470797110322027e-07, 'epoch': 2041.98}
+
 50%|█████     | 33080/65536 [5:40:50<5:20:08,  1.69it/s]
 50%|█████     | 33081/65536 [5:40:51<5:33:34,  1.62it/s]
 50%|█████     | 33082/65536 [5:40:52<5:29:43,  1.64it/s]
 50%|█████     | 33083/65536 [5:40:52<5:34:17,  1.62it/s]
 50%|█████     | 33084/65536 [5:40:53<5:36:56,  1.61it/s]
 50%|█████     | 33085/65536 [5:40:53<5:27:01,  1.65it/s]
 50%|█████     | 33086/65536 [5:40:54<5:24:07,  1.67it/s]
 50%|█████     | 33087/65536 [5:40:55<5:24:20,  1.67it/s]
 50%|█████     | 33088/65536 [5:40:55<5:40:13,  1.59it/s]
 50%|█████     | 33089/65536 [5:40:56<5:38:18,  1.60it/s]
 50%|█████     | 33090/65536 [5:40:57<5:32:02,  1.63it/s]
 50%|█████     | 33091/65536 [5:40:57<5:29:44,  1.64it/s]
 50%|█████     | 33092/65536 [5:40:58<5:37:13,  1.60it/s]
 50%|█████     | 33093/65536 [5:40:58<5:34:08,  1.62it/s]
 50%|█████     | 33094/65536 [5:40:59<5:35:43,  1.61it/s]
 50%|█████     | 33095/65536 [5:41:00<5:33:29,  1.62it/s]
 51%|█████     | 33096/65536 [5:41:00<5:30:10,  1.64it/s]
 51%|█████     | 33097/65536 [5:41:01<5:35:55,  1.61it/s]
 51%|█████     | 33098/65536 [5:41:02<5:37:14,  1.60it/s]
 51%|█████     | 33099/65536 [5:41:02<5:37:40,  1.60it/s]
 51%|█████     | 33100/65536 [5:41:03<5:27:22,  1.65it/s]
                                                         
{'loss': 1.6577, 'learning_rate': 5.468042120729766e-07, 'epoch': 2043.21}
+
 51%|█████     | 33100/65536 [5:41:03<5:27:22,  1.65it/s]
 51%|█████     | 33101/65536 [5:41:03<5:24:48,  1.66it/s]
 51%|█████     | 33102/65536 [5:41:04<5:18:43,  1.70it/s]
 51%|█████     | 33103/65536 [5:41:05<5:29:15,  1.64it/s]
 51%|█████     | 33104/65536 [5:41:05<5:27:50,  1.65it/s]
 51%|█████     | 33105/65536 [5:41:06<5:29:21,  1.64it/s]
 51%|█████     | 33106/65536 [5:41:06<5:22:11,  1.68it/s]
 51%|█████     | 33107/65536 [5:41:07<5:23:30,  1.67it/s]
 51%|█████     | 33108/65536 [5:41:08<5:37:37,  1.60it/s]
 51%|█████     | 33109/65536 [5:41:08<5:36:48,  1.60it/s]
 51%|█████     | 33110/65536 [5:41:09<5:39:01,  1.59it/s]
 51%|█████     | 33111/65536 [5:41:09<5:33:44,  1.62it/s]
 51%|█████     | 33112/65536 [5:41:10<5:43:09,  1.57it/s]
 51%|█████     | 33113/65536 [5:41:11<5:51:50,  1.54it/s]
 51%|█████     | 33114/65536 [5:41:11<5:41:40,  1.58it/s]
 51%|█████     | 33115/65536 [5:41:12<5:32:17,  1.63it/s]
 51%|█████     | 33116/65536 [5:41:13<5:49:44,  1.54it/s]
 51%|█████     | 33117/65536 [5:41:13<5:41:32,  1.58it/s]
 51%|█████     | 33118/65536 [5:41:14<5:40:40,  1.59it/s]
 51%|█████     | 33119/65536 [5:41:15<5:34:58,  1.61it/s]
 51%|█████     | 33120/65536 [5:41:15<5:33:00,  1.62it/s]
                                                         
{'loss': 1.6493, 'learning_rate': 5.465287131137504e-07, 'epoch': 2044.44}
+
 51%|█████     | 33120/65536 [5:41:15<5:33:00,  1.62it/s]
 51%|█████     | 33121/65536 [5:41:16<5:32:54,  1.62it/s]
 51%|█████     | 33122/65536 [5:41:16<5:31:48,  1.63it/s]
 51%|█████     | 33123/65536 [5:41:17<5:34:11,  1.62it/s]
 51%|█████     | 33124/65536 [5:41:18<5:27:04,  1.65it/s]
 51%|█████     | 33125/65536 [5:41:18<5:29:54,  1.64it/s]
 51%|█████     | 33126/65536 [5:41:19<5:30:58,  1.63it/s]
 51%|█████     | 33127/65536 [5:41:19<5:38:53,  1.59it/s]
 51%|█████     | 33128/65536 [5:41:20<5:39:04,  1.59it/s]
 51%|█████     | 33129/65536 [5:41:21<5:34:48,  1.61it/s]
 51%|█████     | 33130/65536 [5:41:21<5:45:07,  1.56it/s]
 51%|█████     | 33131/65536 [5:41:22<5:41:44,  1.58it/s]
 51%|█████     | 33132/65536 [5:41:23<5:32:02,  1.63it/s]
 51%|█████     | 33133/65536 [5:41:23<5:26:30,  1.65it/s]
 51%|█████     | 33134/65536 [5:41:24<5:28:46,  1.64it/s]
 51%|█████     | 33135/65536 [5:41:24<5:34:13,  1.62it/s]
 51%|█████     | 33136/65536 [5:41:25<5:35:50,  1.61it/s]
 51%|█████     | 33137/65536 [5:41:26<5:28:53,  1.64it/s]
 51%|█████     | 33138/65536 [5:41:26<5:34:50,  1.61it/s]
 51%|█████     | 33139/65536 [5:41:27<5:36:33,  1.60it/s]
 51%|█████     | 33140/65536 [5:41:28<5:32:44,  1.62it/s]
                                                         
{'loss': 1.6472, 'learning_rate': 5.462532141545243e-07, 'epoch': 2045.68}
+
 51%|█████     | 33140/65536 [5:41:28<5:32:44,  1.62it/s]
 51%|���████     | 33141/65536 [5:41:28<5:22:28,  1.67it/s]
 51%|█████     | 33142/65536 [5:41:29<5:29:45,  1.64it/s]
 51%|█████     | 33143/65536 [5:41:29<5:40:52,  1.58it/s]
 51%|█████     | 33144/65536 [5:41:30<5:39:23,  1.59it/s]
 51%|█████     | 33145/65536 [5:41:31<5:33:21,  1.62it/s]
 51%|█████     | 33146/65536 [5:41:31<5:47:14,  1.55it/s]
 51%|█████     | 33147/65536 [5:41:32<5:44:58,  1.56it/s]
 51%|█████     | 33148/65536 [5:41:33<5:39:24,  1.59it/s]
 51%|█████     | 33149/65536 [5:41:33<5:34:12,  1.62it/s]
 51%|█████     | 33150/65536 [5:41:34<5:28:24,  1.64it/s]
 51%|█████     | 33151/65536 [5:41:34<5:24:13,  1.66it/s]
 51%|█████     | 33152/65536 [5:41:35<5:29:08,  1.64it/s]
 51%|█████     | 33153/65536 [5:41:36<5:33:57,  1.62it/s]
 51%|█████     | 33154/65536 [5:41:36<5:24:52,  1.66it/s]
 51%|█████     | 33155/65536 [5:41:37<5:25:44,  1.66it/s]
 51%|█████     | 33156/65536 [5:41:37<5:24:21,  1.66it/s]
 51%|█████     | 33157/65536 [5:41:38<5:28:15,  1.64it/s]
 51%|█████     | 33158/65536 [5:41:39<5:28:05,  1.64it/s]
 51%|█████     | 33159/65536 [5:41:39<5:35:03,  1.61it/s]
 51%|█████     | 33160/65536 [5:41:40<5:43:35,  1.57it/s]
                                                         
{'loss': 1.6356, 'learning_rate': 5.459777151952982e-07, 'epoch': 2046.91}
+
 51%|█████     | 33160/65536 [5:41:40<5:43:35,  1.57it/s]
 51%|█████     | 33161/65536 [5:41:41<5:39:44,  1.59it/s]
 51%|█████     | 33162/65536 [5:41:41<5:42:56,  1.57it/s]
 51%|█████     | 33163/65536 [5:41:42<5:37:16,  1.60it/s]
 51%|█████     | 33164/65536 [5:41:42<5:39:27,  1.59it/s]
 51%|█████     | 33165/65536 [5:41:43<5:27:30,  1.65it/s]
 51%|█████     | 33166/65536 [5:41:44<5:29:46,  1.64it/s]
 51%|█████     | 33167/65536 [5:41:44<5:28:48,  1.64it/s]
 51%|█████     | 33168/65536 [5:41:45<5:34:54,  1.61it/s]
 51%|█████     | 33169/65536 [5:41:45<5:38:53,  1.59it/s]
 51%|█████     | 33170/65536 [5:41:46<5:33:52,  1.62it/s]
 51%|█████     | 33171/65536 [5:41:47<5:35:00,  1.61it/s]
 51%|█████     | 33172/65536 [5:41:47<5:30:41,  1.63it/s]
 51%|█████     | 33173/65536 [5:41:48<5:30:56,  1.63it/s]
 51%|█████     | 33174/65536 [5:41:49<5:35:37,  1.61it/s]
 51%|█████     | 33175/65536 [5:41:49<5:33:29,  1.62it/s]
 51%|█████     | 33176/65536 [5:41:50<5:40:30,  1.58it/s]
 51%|█████     | 33177/65536 [5:41:50<5:33:51,  1.62it/s]
 51%|█████     | 33178/65536 [5:41:51<5:39:13,  1.59it/s]
 51%|█████     | 33179/65536 [5:41:52<5:32:28,  1.62it/s]
 51%|█████     | 33180/65536 [5:41:52<5:31:04,  1.63it/s]
                                                         
{'loss': 1.647, 'learning_rate': 5.45702216236072e-07, 'epoch': 2048.15}
+
 51%|█████     | 33180/65536 [5:41:52<5:31:04,  1.63it/s]
 51%|█████     | 33181/65536 [5:41:53<5:39:49,  1.59it/s]
 51%|█████     | 33182/65536 [5:41:54<5:36:51,  1.60it/s]
 51%|█████     | 33183/65536 [5:41:54<5:33:08,  1.62it/s]
 51%|█████     | 33184/65536 [5:41:55<5:25:23,  1.66it/s]
 51%|█████     | 33185/65536 [5:41:55<5:29:23,  1.64it/s]
 51%|█████     | 33186/65536 [5:41:56<5:38:35,  1.59it/s]
 51%|█████     | 33187/65536 [5:41:57<5:36:16,  1.60it/s]
 51%|█████     | 33188/65536 [5:41:57<5:33:03,  1.62it/s]
 51%|█████     | 33189/65536 [5:41:58<5:37:48,  1.60it/s]
 51%|█████     | 33190/65536 [5:41:58<5:30:08,  1.63it/s]
 51%|█████     | 33191/65536 [5:41:59<5:30:53,  1.63it/s]
 51%|█████     | 33192/65536 [5:42:00<5:38:23,  1.59it/s]
 51%|█████     | 33193/65536 [5:42:00<5:33:16,  1.62it/s]
 51%|█████     | 33194/65536 [5:42:01<5:46:41,  1.55it/s]
 51%|█████     | 33195/65536 [5:42:02<5:50:38,  1.54it/s]
 51%|█████     | 33196/65536 [5:42:02<5:42:22,  1.57it/s]
 51%|█████     | 33197/65536 [5:42:03<5:36:27,  1.60it/s]
 51%|█████     | 33198/65536 [5:42:03<5:32:56,  1.62it/s]
 51%|█████     | 33199/65536 [5:42:04<5:29:47,  1.63it/s]
 51%|█████     | 33200/65536 [5:42:05<5:25:53,  1.65it/s]
                                                         
{'loss': 1.6356, 'learning_rate': 5.454267172768459e-07, 'epoch': 2049.38}
+
 51%|█████     | 33200/65536 [5:42:05<5:25:53,  1.65it/s]
 51%|█████     | 33201/65536 [5:42:05<5:25:36,  1.66it/s]
 51%|█████     | 33202/65536 [5:42:06<5:27:19,  1.65it/s]
 51%|█████     | 33203/65536 [5:42:07<5:27:55,  1.64it/s]
 51%|█████     | 33204/65536 [5:42:07<5:28:36,  1.64it/s]
 51%|█████     | 33205/65536 [5:42:08<5:22:53,  1.67it/s]
 51%|█████     | 33206/65536 [5:42:08<5:37:19,  1.60it/s]
 51%|█████     | 33207/65536 [5:42:09<5:33:40,  1.61it/s]
 51%|█████     | 33208/65536 [5:42:10<5:40:52,  1.58it/s]
 51%|█████     | 33209/65536 [5:42:10<5:48:08,  1.55it/s]
 51%|█████     | 33210/65536 [5:42:11<5:46:02,  1.56it/s]
 51%|█████     | 33211/65536 [5:42:12<6:01:14,  1.49it/s]
 51%|█████     | 33212/65536 [5:42:12<5:54:57,  1.52it/s]
 51%|█████     | 33213/65536 [5:42:13<5:45:07,  1.56it/s]
 51%|█████     | 33214/65536 [5:42:14<5:35:42,  1.60it/s]
 51%|█████     | 33215/65536 [5:42:14<5:39:02,  1.59it/s]
 51%|█████     | 33216/65536 [5:42:15<5:36:50,  1.60it/s]
 51%|█████     | 33217/65536 [5:42:15<5:41:45,  1.58it/s]
 51%|█████     | 33218/65536 [5:42:16<5:42:54,  1.57it/s]
 51%|█████     | 33219/65536 [5:42:17<5:42:08,  1.57it/s]
 51%|█████     | 33220/65536 [5:42:17<5:47:23,  1.55it/s]
                                                         
{'loss': 1.5994, 'learning_rate': 5.451512183176196e-07, 'epoch': 2050.62}
+
 51%|█████     | 33220/65536 [5:42:17<5:47:23,  1.55it/s]
 51%|█████     | 33221/65536 [5:42:18<5:38:34,  1.59it/s]
 51%|█████     | 33222/65536 [5:42:19<5:32:17,  1.62it/s]
 51%|█████     | 33223/65536 [5:42:19<5:32:41,  1.62it/s]
 51%|█████     | 33224/65536 [5:42:20<5:23:39,  1.66it/s]
 51%|█████     | 33225/65536 [5:42:20<5:17:44,  1.69it/s]
 51%|█████     | 33226/65536 [5:42:21<5:17:00,  1.70it/s]
 51%|█████     | 33227/65536 [5:42:22<5:41:06,  1.58it/s]
 51%|█████     | 33228/65536 [5:42:22<5:34:16,  1.61it/s]
 51%|█████     | 33229/65536 [5:42:23<5:27:24,  1.64it/s]
 51%|█████     | 33230/65536 [5:42:23<5:28:05,  1.64it/s]
 51%|█████     | 33231/65536 [5:42:24<5:33:35,  1.61it/s]
 51%|█████     | 33232/65536 [5:42:25<5:30:49,  1.63it/s]
 51%|█████     | 33233/65536 [5:42:25<5:29:53,  1.63it/s]
 51%|█████     | 33234/65536 [5:42:26<5:39:50,  1.58it/s]
 51%|█████     | 33235/65536 [5:42:27<5:47:54,  1.55it/s]
 51%|█████     | 33236/65536 [5:42:27<5:42:46,  1.57it/s]
 51%|█████     | 33237/65536 [5:42:28<5:40:22,  1.58it/s]
 51%|█████     | 33238/65536 [5:42:28<5:37:52,  1.59it/s]
 51%|█████     | 33239/65536 [5:42:29<5:34:18,  1.61it/s]
 51%|█████     | 33240/65536 [5:42:30<5:30:26,  1.63it/s]
                                                         
{'loss': 1.6527, 'learning_rate': 5.448757193583936e-07, 'epoch': 2051.85}
+
 51%|█████     | 33240/65536 [5:42:30<5:30:26,  1.63it/s]
 51%|█████     | 33241/65536 [5:42:30<5:23:31,  1.66it/s]
 51%|█████     | 33242/65536 [5:42:31<5:18:43,  1.69it/s]
 51%|█████     | 33243/65536 [5:42:32<5:50:37,  1.54it/s]
 51%|█████     | 33244/65536 [5:42:32<5:39:13,  1.59it/s]
 51%|█████     | 33245/65536 [5:42:33<5:35:53,  1.60it/s]
 51%|█████     | 33246/65536 [5:42:33<5:31:55,  1.62it/s]
 51%|█████     | 33247/65536 [5:42:34<5:32:03,  1.62it/s]
 51%|█████     | 33248/65536 [5:42:35<5:27:30,  1.64it/s]
 51%|█████     | 33249/65536 [5:42:35<5:31:16,  1.62it/s]
 51%|█████     | 33250/65536 [5:42:36<5:33:45,  1.61it/s]
 51%|█████     | 33251/65536 [5:42:37<5:39:25,  1.59it/s]
 51%|█████     | 33252/65536 [5:42:37<5:43:14,  1.57it/s]
 51%|█████     | 33253/65536 [5:42:38<5:30:48,  1.63it/s]
 51%|█████     | 33254/65536 [5:42:38<5:36:57,  1.60it/s]
 51%|█████     | 33255/65536 [5:42:39<5:40:40,  1.58it/s]
 51%|█████     | 33256/65536 [5:42:40<5:33:31,  1.61it/s]
 51%|█████     | 33257/65536 [5:42:40<5:27:19,  1.64it/s]
 51%|█████     | 33258/65536 [5:42:41<5:22:58,  1.67it/s]
 51%|█████     | 33259/65536 [5:42:41<5:31:01,  1.63it/s]
 51%|█████     | 33260/65536 [5:42:42<5:36:48,  1.60it/s]
                                                         
{'loss': 1.6733, 'learning_rate': 5.446002203991674e-07, 'epoch': 2053.09}
+
 51%|█████     | 33260/65536 [5:42:42<5:36:48,  1.60it/s]
 51%|█████     | 33261/65536 [5:42:43<5:41:20,  1.58it/s]
 51%|█████     | 33262/65536 [5:42:43<5:32:40,  1.62it/s]
 51%|█████     | 33263/65536 [5:42:44<5:30:21,  1.63it/s]
 51%|█████     | 33264/65536 [5:42:45<5:27:44,  1.64it/s]
 51%|█████     | 33265/65536 [5:42:45<5:24:27,  1.66it/s]
 51%|█████     | 33266/65536 [5:42:46<5:32:42,  1.62it/s]
 51%|█████     | 33267/65536 [5:42:46<5:31:11,  1.62it/s]
 51%|█████     | 33268/65536 [5:42:47<5:29:18,  1.63it/s]
 51%|█████     | 33269/65536 [5:42:48<5:22:46,  1.67it/s]
 51%|█████     | 33270/65536 [5:42:48<5:22:01,  1.67it/s]
 51%|█████     | 33271/65536 [5:42:49<5:30:13,  1.63it/s]
 51%|█████     | 33272/65536 [5:42:49<5:26:51,  1.65it/s]
 51%|█████     | 33273/65536 [5:42:50<5:27:55,  1.64it/s]
 51%|█████     | 33274/65536 [5:42:51<5:23:17,  1.66it/s]
 51%|█████     | 33275/65536 [5:42:51<5:48:03,  1.54it/s]
 51%|█████     | 33276/65536 [5:42:52<5:39:54,  1.58it/s]
 51%|█████     | 33277/65536 [5:42:53<5:33:42,  1.61it/s]
 51%|█████     | 33278/65536 [5:42:53<5:38:48,  1.59it/s]
 51%|█████     | 33279/65536 [5:42:54<5:33:26,  1.61it/s]
 51%|█████     | 33280/65536 [5:42:54<5:41:51,  1.57it/s]
                                                         
{'loss': 1.679, 'learning_rate': 5.443247214399413e-07, 'epoch': 2054.32}
+
 51%|█████     | 33280/65536 [5:42:54<5:41:51,  1.57it/s]
 51%|█████     | 33281/65536 [5:42:55<5:34:40,  1.61it/s]
 51%|█████     | 33282/65536 [5:42:56<5:32:18,  1.62it/s]
 51%|█████     | 33283/65536 [5:42:56<5:28:12,  1.64it/s]
 51%|█████     | 33284/65536 [5:42:57<5:24:07,  1.66it/s]
 51%|█████     | 33285/65536 [5:42:57<5:23:26,  1.66it/s]
 51%|█████     | 33286/65536 [5:42:58<5:26:25,  1.65it/s]
 51%|█████     | 33287/65536 [5:42:59<5:31:19,  1.62it/s]
 51%|█████     | 33288/65536 [5:42:59<5:37:13,  1.59it/s]
 51%|█████     | 33289/65536 [5:43:00<5:36:52,  1.60it/s]
 51%|█████     | 33290/65536 [5:43:01<5:38:01,  1.59it/s]
 51%|█████     | 33291/65536 [5:43:01<5:37:48,  1.59it/s]
 51%|█████     | 33292/65536 [5:43:02<5:46:18,  1.55it/s]
 51%|█████     | 33293/65536 [5:43:03<5:41:02,  1.58it/s]
 51%|█████     | 33294/65536 [5:43:03<5:35:38,  1.60it/s]
 51%|█████     | 33295/65536 [5:43:04<5:29:07,  1.63it/s]
 51%|█████     | 33296/65536 [5:43:04<5:35:41,  1.60it/s]
 51%|█████     | 33297/65536 [5:43:05<5:32:01,  1.62it/s]
 51%|█████     | 33298/65536 [5:43:06<5:26:02,  1.65it/s]
 51%|█████     | 33299/65536 [5:43:06<5:25:11,  1.65it/s]
 51%|█████     | 33300/65536 [5:43:07<5:35:46,  1.60it/s]
                                                         
{'loss': 1.6314, 'learning_rate': 5.44049222480715e-07, 'epoch': 2055.56}
+
 51%|█████     | 33300/65536 [5:43:07<5:35:46,  1.60it/s]
 51%|█████     | 33301/65536 [5:43:07<5:29:58,  1.63it/s]
 51%|█████     | 33302/65536 [5:43:08<5:29:56,  1.63it/s]
 51%|█████     | 33303/65536 [5:43:09<5:25:26,  1.65it/s]
 51%|█████     | 33304/65536 [5:43:09<5:24:28,  1.66it/s]
 51%|█████     | 33305/65536 [5:43:10<5:29:24,  1.63it/s]
 51%|█████     | 33306/65536 [5:43:10<5:24:29,  1.66it/s]
 51%|█████     | 33307/65536 [5:43:11<5:24:03,  1.66it/s]
 51%|█████     | 33308/65536 [5:43:12<5:50:56,  1.53it/s]
 51%|█████     | 33309/65536 [5:43:12<5:42:42,  1.57it/s]
 51%|█████     | 33310/65536 [5:43:13<5:48:18,  1.54it/s]
 51%|█████     | 33311/65536 [5:43:14<5:54:30,  1.51it/s]
 51%|█████     | 33312/65536 [5:43:14<5:39:25,  1.58it/s]
 51%|█████     | 33313/65536 [5:43:15<5:34:38,  1.60it/s]
 51%|█████     | 33314/65536 [5:43:16<5:29:59,  1.63it/s]
 51%|█████     | 33315/65536 [5:43:16<5:35:42,  1.60it/s]
 51%|█████     | 33316/65536 [5:43:17<5:27:44,  1.64it/s]
 51%|█████     | 33317/65536 [5:43:17<5:28:19,  1.64it/s]
 51%|█████     | 33318/65536 [5:43:18<5:30:29,  1.62it/s]
 51%|█████     | 33319/65536 [5:43:19<5:23:01,  1.66it/s]
 51%|█████     | 33320/65536 [5:43:19<5:17:24,  1.69it/s]
                                                         
{'loss': 1.6589, 'learning_rate': 5.437737235214889e-07, 'epoch': 2056.79}
+
 51%|█████     | 33320/65536 [5:43:19<5:17:24,  1.69it/s]
 51%|█████     | 33321/65536 [5:43:20<5:25:48,  1.65it/s]
 51%|█████     | 33322/65536 [5:43:20<5:26:39,  1.64it/s]
 51%|█████     | 33323/65536 [5:43:21<5:32:24,  1.62it/s]
 51%|█████     | 33324/65536 [5:43:22<5:38:18,  1.59it/s]
 51%|█████     | 33325/65536 [5:43:22<5:32:19,  1.62it/s]
 51%|█████     | 33326/65536 [5:43:23<5:37:50,  1.59it/s]
 51%|█████     | 33327/65536 [5:43:24<5:35:29,  1.60it/s]
 51%|█████     | 33328/65536 [5:43:24<5:30:05,  1.63it/s]
 51%|█████     | 33329/65536 [5:43:25<5:34:48,  1.60it/s]
 51%|█████     | 33330/65536 [5:43:25<5:42:43,  1.57it/s]
 51%|█████     | 33331/65536 [5:43:26<5:35:14,  1.60it/s]
 51%|█████     | 33332/65536 [5:43:27<5:36:13,  1.60it/s]
 51%|█████     | 33333/65536 [5:43:27<5:31:46,  1.62it/s]
 51%|█████     | 33334/65536 [5:43:28<5:38:47,  1.58it/s]
 51%|█████     | 33335/65536 [5:43:29<5:35:35,  1.60it/s]
 51%|█████     | 33336/65536 [5:43:29<5:30:57,  1.62it/s]
 51%|█████     | 33337/65536 [5:43:30<5:30:38,  1.62it/s]
 51%|█████     | 33338/65536 [5:43:30<5:29:24,  1.63it/s]
 51%|█████     | 33339/65536 [5:43:31<5:34:47,  1.60it/s]
 51%|█████     | 33340/65536 [5:43:32<5:47:29,  1.54it/s]
                                                         
{'loss': 1.6226, 'learning_rate': 5.434982245622627e-07, 'epoch': 2058.02}
+
 51%|█████     | 33340/65536 [5:43:32<5:47:29,  1.54it/s]
 51%|█████     | 33341/65536 [5:43:32<5:49:45,  1.53it/s]
 51%|█████     | 33342/65536 [5:43:33<5:53:51,  1.52it/s]
 51%|█████     | 33343/65536 [5:43:34<5:45:14,  1.55it/s]
 51%|█████     | 33344/65536 [5:43:34<5:36:08,  1.60it/s]
 51%|█████     | 33345/65536 [5:43:35<5:28:36,  1.63it/s]
 51%|█████     | 33346/65536 [5:43:35<5:26:50,  1.64it/s]
 51%|█████     | 33347/65536 [5:43:36<5:32:10,  1.62it/s]
 51%|█████     | 33348/65536 [5:43:37<5:35:42,  1.60it/s]
 51%|█████     | 33349/65536 [5:43:37<5:30:51,  1.62it/s]
 51%|█████     | 33350/65536 [5:43:38<5:29:20,  1.63it/s]
 51%|█████     | 33351/65536 [5:43:39<5:27:47,  1.64it/s]
 51%|█████     | 33352/65536 [5:43:39<5:30:35,  1.62it/s]
 51%|█████     | 33353/65536 [5:43:40<5:32:25,  1.61it/s]
 51%|█████     | 33354/65536 [5:43:40<5:30:45,  1.62it/s]
 51%|█████     | 33355/65536 [5:43:41<5:18:49,  1.68it/s]
 51%|█████     | 33356/65536 [5:43:42<5:31:04,  1.62it/s]
 51%|█████     | 33357/65536 [5:43:42<5:36:12,  1.60it/s]
 51%|█████     | 33358/65536 [5:43:43<5:36:35,  1.59it/s]
 51%|█████     | 33359/65536 [5:43:44<5:40:57,  1.57it/s]
 51%|█████     | 33360/65536 [5:43:44<5:38:16,  1.59it/s]
                                                         
{'loss': 1.6508, 'learning_rate': 5.432227256030365e-07, 'epoch': 2059.26}
+
 51%|█████     | 33360/65536 [5:43:44<5:38:16,  1.59it/s]
 51%|█████     | 33361/65536 [5:43:45<5:38:37,  1.58it/s]
 51%|█████     | 33362/65536 [5:43:45<5:50:02,  1.53it/s]
 51%|█████     | 33363/65536 [5:43:46<5:37:49,  1.59it/s]
 51%|█████     | 33364/65536 [5:43:47<5:35:11,  1.60it/s]
 51%|█████     | 33365/65536 [5:43:47<5:29:59,  1.62it/s]
 51%|█████     | 33366/65536 [5:43:48<5:28:52,  1.63it/s]
 51%|█████     | 33367/65536 [5:43:48<5:23:32,  1.66it/s]
 51%|█████     | 33368/65536 [5:43:49<5:26:44,  1.64it/s]
 51%|█████     | 33369/65536 [5:43:50<5:29:56,  1.62it/s]
 51%|█████     | 33370/65536 [5:43:50<5:25:24,  1.65it/s]
 51%|█████     | 33371/65536 [5:43:51<5:23:28,  1.66it/s]
 51%|█████     | 33372/65536 [5:43:51<5:24:05,  1.65it/s]
 51%|█████     | 33373/65536 [5:43:52<5:45:41,  1.55it/s]
 51%|█████     | 33374/65536 [5:43:53<5:40:28,  1.57it/s]
 51%|█████     | 33375/65536 [5:43:53<5:36:31,  1.59it/s]
 51%|█████     | 33376/65536 [5:43:54<5:33:39,  1.61it/s]
 51%|█████     | 33377/65536 [5:43:55<5:40:47,  1.57it/s]
 51%|█████     | 33378/65536 [5:43:55<5:33:58,  1.60it/s]
 51%|█████     | 33379/65536 [5:43:56<5:31:45,  1.62it/s]
 51%|█████     | 33380/65536 [5:43:57<5:28:20,  1.63it/s]
                                                         
{'loss': 1.6574, 'learning_rate': 5.429472266438104e-07, 'epoch': 2060.49}
+
 51%|█████     | 33380/65536 [5:43:57<5:28:20,  1.63it/s]
 51%|█████     | 33381/65536 [5:43:57<5:25:50,  1.64it/s]
 51%|█████     | 33382/65536 [5:43:58<5:22:11,  1.66it/s]
 51%|█████     | 33383/65536 [5:43:58<5:20:23,  1.67it/s]
 51%|█████     | 33384/65536 [5:43:59<5:18:55,  1.68it/s]
 51%|█████     | 33385/65536 [5:43:59<5:15:58,  1.70it/s]
 51%|█████     | 33386/65536 [5:44:00<5:25:20,  1.65it/s]
 51%|█████     | 33387/65536 [5:44:01<5:37:01,  1.59it/s]
 51%|█████     | 33388/65536 [5:44:01<5:28:38,  1.63it/s]
 51%|█████     | 33389/65536 [5:44:02<5:40:05,  1.58it/s]
 51%|█████     | 33390/65536 [5:44:03<5:38:09,  1.58it/s]
 51%|█████     | 33391/65536 [5:44:03<5:47:02,  1.54it/s]
 51%|█████     | 33392/65536 [5:44:04<5:43:26,  1.56it/s]
 51%|█████     | 33393/65536 [5:44:05<5:40:33,  1.57it/s]
 51%|█████     | 33394/65536 [5:44:05<5:34:53,  1.60it/s]
 51%|█████     | 33395/65536 [5:44:06<5:34:31,  1.60it/s]
 51%|█████     | 33396/65536 [5:44:07<5:40:32,  1.57it/s]
 51%|█████     | 33397/65536 [5:44:07<5:35:05,  1.60it/s]
 51%|█████     | 33398/65536 [5:44:08<5:44:15,  1.56it/s]
 51%|█████     | 33399/65536 [5:44:08<5:36:05,  1.59it/s]
 51%|█████     | 33400/65536 [5:44:09<5:30:19,  1.62it/s]
                                                         
{'loss': 1.6206, 'learning_rate': 5.426717276845842e-07, 'epoch': 2061.73}
+
 51%|█████     | 33400/65536 [5:44:09<5:30:19,  1.62it/s]
 51%|█████     | 33401/65536 [5:44:10<5:32:16,  1.61it/s]
 51%|█████     | 33402/65536 [5:44:10<5:30:16,  1.62it/s]
 51%|█████     | 33403/65536 [5:44:11<5:22:50,  1.66it/s]
 51%|█████     | 33404/65536 [5:44:11<5:20:03,  1.67it/s]
 51%|█████     | 33405/65536 [5:44:12<5:30:22,  1.62it/s]
 51%|█████     | 33406/65536 [5:44:13<5:44:29,  1.55it/s]
 51%|█████     | 33407/65536 [5:44:13<5:38:01,  1.58it/s]
 51%|█████     | 33408/65536 [5:44:14<5:36:30,  1.59it/s]
 51%|█████     | 33409/65536 [5:44:15<5:30:18,  1.62it/s]
 51%|█████     | 33410/65536 [5:44:15<5:30:24,  1.62it/s]
 51%|█████     | 33411/65536 [5:44:16<5:33:54,  1.60it/s]
 51%|█████     | 33412/65536 [5:44:16<5:25:44,  1.64it/s]
 51%|█████     | 33413/65536 [5:44:17<5:29:43,  1.62it/s]
 51%|█████     | 33414/65536 [5:44:18<5:28:19,  1.63it/s]
 51%|█████     | 33415/65536 [5:44:18<5:24:43,  1.65it/s]
 51%|█████     | 33416/65536 [5:44:19<5:29:44,  1.62it/s]
 51%|█████     | 33417/65536 [5:44:19<5:27:21,  1.64it/s]
 51%|█████     | 33418/65536 [5:44:20<5:26:02,  1.64it/s]
 51%|█████     | 33419/65536 [5:44:21<5:24:11,  1.65it/s]
 51%|█████     | 33420/65536 [5:44:21<5:27:33,  1.63it/s]
                                                         
{'loss': 1.6472, 'learning_rate': 5.423962287253581e-07, 'epoch': 2062.96}
+
 51%|█████     | 33420/65536 [5:44:21<5:27:33,  1.63it/s]
 51%|█████     | 33421/65536 [5:44:22<5:37:51,  1.58it/s]
 51%|█████     | 33422/65536 [5:44:23<5:34:27,  1.60it/s]
 51%|█████     | 33423/65536 [5:44:23<5:32:28,  1.61it/s]
 51%|█████     | 33424/65536 [5:44:24<5:29:45,  1.62it/s]
 51%|█████     | 33425/65536 [5:44:24<5:17:21,  1.69it/s]
 51%|█████     | 33426/65536 [5:44:25<5:25:39,  1.64it/s]
 51%|█████     | 33427/65536 [5:44:26<5:30:27,  1.62it/s]
 51%|█████     | 33428/65536 [5:44:26<5:30:38,  1.62it/s]
 51%|█████     | 33429/65536 [5:44:27<5:29:25,  1.62it/s]
 51%|█████     | 33430/65536 [5:44:27<5:31:31,  1.61it/s]
 51%|█████     | 33431/65536 [5:44:28<5:31:24,  1.61it/s]
 51%|█████     | 33432/65536 [5:44:29<5:35:39,  1.59it/s]
 51%|█████     | 33433/65536 [5:44:29<5:38:51,  1.58it/s]
 51%|█████     | 33434/65536 [5:44:30<5:30:34,  1.62it/s]
 51%|█████     | 33435/65536 [5:44:31<5:33:11,  1.61it/s]
 51%|█████     | 33436/65536 [5:44:31<5:26:24,  1.64it/s]
 51%|█████     | 33437/65536 [5:44:32<5:39:01,  1.58it/s]
 51%|█████     | 33438/65536 [5:44:32<5:37:14,  1.59it/s]
 51%|█████     | 33439/65536 [5:44:33<5:41:34,  1.57it/s]
 51%|█████     | 33440/65536 [5:44:34<5:36:04,  1.59it/s]
                                                         
{'loss': 1.675, 'learning_rate': 5.42120729766132e-07, 'epoch': 2064.2}
+
 51%|█████     | 33440/65536 [5:44:34<5:36:04,  1.59it/s]
 51%|█████     | 33441/65536 [5:44:34<5:32:56,  1.61it/s]
 51%|█████     | 33442/65536 [5:44:35<5:43:15,  1.56it/s]
 51%|█████     | 33443/65536 [5:44:36<5:33:29,  1.60it/s]
 51%|█████     | 33444/65536 [5:44:36<5:34:18,  1.60it/s]
 51%|█████     | 33445/65536 [5:44:37<5:30:37,  1.62it/s]
 51%|█████     | 33446/65536 [5:44:37<5:34:20,  1.60it/s]
 51%|█████     | 33447/65536 [5:44:38<5:33:03,  1.61it/s]
 51%|█████     | 33448/65536 [5:44:39<5:27:48,  1.63it/s]
 51%|█████     | 33449/65536 [5:44:39<5:34:16,  1.60it/s]
 51%|█████     | 33450/65536 [5:44:40<5:23:33,  1.65it/s]
 51%|█████     | 33451/65536 [5:44:41<5:24:37,  1.65it/s]
 51%|█████     | 33452/65536 [5:44:41<5:25:23,  1.64it/s]
 51%|█████     | 33453/65536 [5:44:42<5:25:32,  1.64it/s]
 51%|█████     | 33454/65536 [5:44:42<5:32:35,  1.61it/s]
 51%|█████     | 33455/65536 [5:44:43<5:32:15,  1.61it/s]
 51%|█████     | 33456/65536 [5:44:44<5:32:50,  1.61it/s]
 51%|█████     | 33457/65536 [5:44:44<5:35:51,  1.59it/s]
 51%|█████     | 33458/65536 [5:44:45<5:34:35,  1.60it/s]
 51%|█████     | 33459/65536 [5:44:46<5:33:19,  1.60it/s]
 51%|█████     | 33460/65536 [5:44:46<5:34:53,  1.60it/s]
                                                         
{'loss': 1.6319, 'learning_rate': 5.418452308069058e-07, 'epoch': 2065.43}
+
 51%|█████     | 33460/65536 [5:44:46<5:34:53,  1.60it/s]
 51%|█████     | 33461/65536 [5:44:47<5:36:21,  1.59it/s]
 51%|█████     | 33462/65536 [5:44:47<5:32:29,  1.61it/s]
 51%|█████     | 33463/65536 [5:44:48<5:34:42,  1.60it/s]
 51%|█████     | 33464/65536 [5:44:49<5:45:59,  1.54it/s]
 51%|█████     | 33465/65536 [5:44:49<5:38:03,  1.58it/s]
 51%|█████     | 33466/65536 [5:44:50<5:38:56,  1.58it/s]
 51%|█████     | 33467/65536 [5:44:51<5:32:08,  1.61it/s]
 51%|█████     | 33468/65536 [5:44:51<5:25:16,  1.64it/s]
 51%|█████     | 33469/65536 [5:44:52<5:28:07,  1.63it/s]
 51%|█████     | 33470/65536 [5:44:52<5:40:40,  1.57it/s]
 51%|█████     | 33471/65536 [5:44:53<5:35:54,  1.59it/s]
 51%|█████     | 33472/65536 [5:44:54<5:30:40,  1.62it/s]
 51%|█████     | 33473/65536 [5:44:54<5:26:21,  1.64it/s]
 51%|█████     | 33474/65536 [5:44:55<5:24:59,  1.64it/s]
 51%|█████     | 33475/65536 [5:44:55<5:20:08,  1.67it/s]
 51%|█████     | 33476/65536 [5:44:56<5:24:23,  1.65it/s]
 51%|█████     | 33477/65536 [5:44:57<5:28:24,  1.63it/s]
 51%|█████     | 33478/65536 [5:44:57<5:32:01,  1.61it/s]
 51%|█████     | 33479/65536 [5:44:58<5:40:13,  1.57it/s]
 51%|█████     | 33480/65536 [5:44:59<5:30:53,  1.61it/s]
                                                         
{'loss': 1.6193, 'learning_rate': 5.415697318476797e-07, 'epoch': 2066.67}
+
 51%|█████     | 33480/65536 [5:44:59<5:30:53,  1.61it/s]
 51%|█████     | 33481/65536 [5:44:59<5:27:21,  1.63it/s]
 51%|█████     | 33482/65536 [5:45:00<5:39:00,  1.58it/s]
 51%|█████     | 33483/65536 [5:45:00<5:30:56,  1.61it/s]
 51%|█████     | 33484/65536 [5:45:01<5:33:16,  1.60it/s]
 51%|█████     | 33485/65536 [5:45:02<5:30:01,  1.62it/s]
 51%|█████     | 33486/65536 [5:45:02<5:40:40,  1.57it/s]
 51%|█████     | 33487/65536 [5:45:03<5:34:25,  1.60it/s]
 51%|█████     | 33488/65536 [5:45:04<5:31:52,  1.61it/s]
 51%|█████     | 33489/65536 [5:45:04<5:36:57,  1.59it/s]
 51%|█████     | 33490/65536 [5:45:05<5:24:53,  1.64it/s]
 51%|█████     | 33491/65536 [5:45:05<5:28:34,  1.63it/s]
 51%|█████     | 33492/65536 [5:45:06<5:31:58,  1.61it/s]
 51%|█████     | 33493/65536 [5:45:07<5:32:33,  1.61it/s]
 51%|█████     | 33494/65536 [5:45:07<5:36:34,  1.59it/s]
 51%|█████     | 33495/65536 [5:45:08<5:31:05,  1.61it/s]
 51%|█████     | 33496/65536 [5:45:09<5:25:52,  1.64it/s]
 51%|█████     | 33497/65536 [5:45:09<5:22:43,  1.65it/s]
 51%|█████     | 33498/65536 [5:45:10<5:21:50,  1.66it/s]
 51%|█████     | 33499/65536 [5:45:10<5:25:38,  1.64it/s]
 51%|█████     | 33500/65536 [5:45:11<5:26:45,  1.63it/s]
                                                         
{'loss': 1.6383, 'learning_rate': 5.412942328884535e-07, 'epoch': 2067.9}
+
 51%|█████     | 33500/65536 [5:45:11<5:26:45,  1.63it/s]
 51%|█████     | 33501/65536 [5:45:12<5:25:58,  1.64it/s]
 51%|█████     | 33502/65536 [5:45:12<5:41:19,  1.56it/s]
 51%|█████     | 33503/65536 [5:45:13<5:52:01,  1.52it/s]
 51%|█████     | 33504/65536 [5:45:14<5:38:58,  1.57it/s]
 51%|█████     | 33505/65536 [5:45:14<5:30:59,  1.61it/s]
 51%|█████     | 33506/65536 [5:45:15<5:29:18,  1.62it/s]
 51%|█████     | 33507/65536 [5:45:15<5:27:32,  1.63it/s]
 51%|█████     | 33508/65536 [5:45:16<5:27:37,  1.63it/s]
 51%|█████     | 33509/65536 [5:45:17<5:36:26,  1.59it/s]
 51%|█████     | 33510/65536 [5:45:17<5:24:39,  1.64it/s]
 51%|█████     | 33511/65536 [5:45:18<5:23:26,  1.65it/s]
 51%|█████     | 33512/65536 [5:45:18<5:27:01,  1.63it/s]
 51%|█████     | 33513/65536 [5:45:19<5:26:07,  1.64it/s]
 51%|█████     | 33514/65536 [5:45:20<5:27:17,  1.63it/s]
 51%|█████     | 33515/65536 [5:45:20<5:28:41,  1.62it/s]
 51%|█████     | 33516/65536 [5:45:21<5:27:26,  1.63it/s]
 51%|█████     | 33517/65536 [5:45:21<5:22:05,  1.66it/s]
 51%|█████     | 33518/65536 [5:45:22<5:31:30,  1.61it/s]
 51%|█████     | 33519/65536 [5:45:23<5:34:13,  1.60it/s]
 51%|█████     | 33520/65536 [5:45:23<5:34:12,  1.60it/s]
                                                         
{'loss': 1.6552, 'learning_rate': 5.410187339292274e-07, 'epoch': 2069.14}
+
 51%|█████     | 33520/65536 [5:45:23<5:34:12,  1.60it/s]
 51%|█████     | 33521/65536 [5:45:24<5:32:43,  1.60it/s]
 51%|█████     | 33522/65536 [5:45:25<5:40:27,  1.57it/s]
 51%|█████     | 33523/65536 [5:45:25<5:33:54,  1.60it/s]
 51%|█████     | 33524/65536 [5:45:26<5:38:49,  1.57it/s]
 51%|█████     | 33525/65536 [5:45:27<5:33:16,  1.60it/s]
 51%|█████     | 33526/65536 [5:45:27<5:22:37,  1.65it/s]
 51%|█████     | 33527/65536 [5:45:28<5:16:03,  1.69it/s]
 51%|█████     | 33528/65536 [5:45:28<5:26:16,  1.64it/s]
 51%|█████     | 33529/65536 [5:45:29<5:25:21,  1.64it/s]
 51%|█████     | 33530/65536 [5:45:29<5:23:25,  1.65it/s]
 51%|█████     | 33531/65536 [5:45:30<5:27:55,  1.63it/s]
 51%|█████     | 33532/65536 [5:45:31<5:19:25,  1.67it/s]
 51%|█████     | 33533/65536 [5:45:31<5:16:01,  1.69it/s]
 51%|█████     | 33534/65536 [5:45:32<5:27:26,  1.63it/s]
 51%|█████     | 33535/65536 [5:45:33<5:40:17,  1.57it/s]
 51%|█████     | 33536/65536 [5:45:33<5:36:06,  1.59it/s]
 51%|█████     | 33537/65536 [5:45:34<5:35:29,  1.59it/s]
 51%|█████     | 33538/65536 [5:45:34<5:30:15,  1.61it/s]
 51%|█████     | 33539/65536 [5:45:35<5:29:04,  1.62it/s]
 51%|█████     | 33540/65536 [5:45:36<5:27:08,  1.63it/s]
                                                         
{'loss': 1.6484, 'learning_rate': 5.407432349700013e-07, 'epoch': 2070.37}
+
 51%|█████     | 33540/65536 [5:45:36<5:27:08,  1.63it/s]
 51%|█████     | 33541/65536 [5:45:36<5:22:36,  1.65it/s]
 51%|█████     | 33542/65536 [5:45:37<5:25:25,  1.64it/s]
 51%|█████     | 33543/65536 [5:45:38<5:33:47,  1.60it/s]
 51%|█████     | 33544/65536 [5:45:38<5:34:51,  1.59it/s]
 51%|█████     | 33545/65536 [5:45:39<5:29:21,  1.62it/s]
 51%|█████     | 33546/65536 [5:45:39<5:33:46,  1.60it/s]
 51%|█████     | 33547/65536 [5:45:40<5:36:14,  1.59it/s]
 51%|█████     | 33548/65536 [5:45:41<5:27:39,  1.63it/s]
 51%|█████     | 33549/65536 [5:45:41<5:29:52,  1.62it/s]
 51%|█████     | 33550/65536 [5:45:42<5:21:02,  1.66it/s]
 51%|█████     | 33551/65536 [5:45:43<5:39:42,  1.57it/s]
 51%|█████     | 33552/65536 [5:45:43<5:33:44,  1.60it/s]
 51%|█████     | 33553/65536 [5:45:44<5:39:11,  1.57it/s]
 51%|█████     | 33554/65536 [5:45:44<5:31:29,  1.61it/s]
 51%|█████     | 33555/65536 [5:45:45<5:28:21,  1.62it/s]
 51%|█████     | 33556/65536 [5:45:46<5:31:54,  1.61it/s]
 51%|█████     | 33557/65536 [5:45:46<5:28:14,  1.62it/s]
 51%|█████     | 33558/65536 [5:45:47<5:33:57,  1.60it/s]
 51%|█████     | 33559/65536 [5:45:47<5:26:13,  1.63it/s]
 51%|█████     | 33560/65536 [5:45:48<5:20:10,  1.66it/s]
                                                         
{'loss': 1.6406, 'learning_rate': 5.40467736010775e-07, 'epoch': 2071.6}
+
 51%|█████     | 33560/65536 [5:45:48<5:20:10,  1.66it/s]
 51%|█████     | 33561/65536 [5:45:49<5:16:36,  1.68it/s]
 51%|█████     | 33562/65536 [5:45:49<5:24:21,  1.64it/s]
 51%|█████     | 33563/65536 [5:45:50<5:29:43,  1.62it/s]
 51%|█████     | 33564/65536 [5:45:51<5:33:36,  1.60it/s]
 51%|█████     | 33565/65536 [5:45:51<5:32:35,  1.60it/s]
 51%|█████     | 33566/65536 [5:45:52<5:35:20,  1.59it/s]
 51%|█████     | 33567/65536 [5:45:52<5:38:37,  1.57it/s]
 51%|█████     | 33568/65536 [5:45:53<5:34:49,  1.59it/s]
 51%|█████     | 33569/65536 [5:45:54<5:40:39,  1.56it/s]
 51%|█████     | 33570/65536 [5:45:54<5:33:21,  1.60it/s]
 51%|█████     | 33571/65536 [5:45:55<5:28:23,  1.62it/s]
 51%|█████     | 33572/65536 [5:45:56<5:31:10,  1.61it/s]
 51%|█████     | 33573/65536 [5:45:56<5:28:36,  1.62it/s]
 51%|█████     | 33574/65536 [5:45:57<5:24:28,  1.64it/s]
 51%|█████     | 33575/65536 [5:45:57<5:31:32,  1.61it/s]
 51%|█████     | 33576/65536 [5:45:58<5:26:29,  1.63it/s]
 51%|█████     | 33577/65536 [5:45:59<5:23:00,  1.65it/s]
 51%|█████     | 33578/65536 [5:45:59<5:25:25,  1.64it/s]
 51%|█████     | 33579/65536 [5:46:00<5:28:52,  1.62it/s]
 51%|█████     | 33580/65536 [5:46:00<5:33:21,  1.60it/s]
                                                         
{'loss': 1.6387, 'learning_rate': 5.40192237051549e-07, 'epoch': 2072.84}
+
 51%|█████     | 33580/65536 [5:46:00<5:33:21,  1.60it/s]
 51%|█████     | 33581/65536 [5:46:01<5:28:48,  1.62it/s]
 51%|█████     | 33582/65536 [5:46:02<5:24:25,  1.64it/s]
 51%|█████     | 33583/65536 [5:46:02<5:34:43,  1.59it/s]
 51%|█████     | 33584/65536 [5:46:03<5:29:51,  1.61it/s]
 51%|█████     | 33585/65536 [5:46:04<5:22:47,  1.65it/s]
 51%|█████     | 33586/65536 [5:46:04<5:31:25,  1.61it/s]
 51%|█████     | 33587/65536 [5:46:05<5:25:15,  1.64it/s]
 51%|█████▏    | 33588/65536 [5:46:05<5:23:17,  1.65it/s]
 51%|█████▏    | 33589/65536 [5:46:06<5:29:48,  1.61it/s]
 51%|█████▏    | 33590/65536 [5:46:07<5:24:42,  1.64it/s]
 51%|█████▏    | 33591/65536 [5:46:07<5:29:40,  1.62it/s]
 51%|█████▏    | 33592/65536 [5:46:08<5:23:45,  1.64it/s]
 51%|█████▏    | 33593/65536 [5:46:08<5:22:26,  1.65it/s]
 51%|█████▏    | 33594/65536 [5:46:09<5:23:13,  1.65it/s]
 51%|█████▏    | 33595/65536 [5:46:10<5:24:42,  1.64it/s]
 51%|█████▏    | 33596/65536 [5:46:10<5:23:19,  1.65it/s]
 51%|█████▏    | 33597/65536 [5:46:11<5:33:35,  1.60it/s]
 51%|█████▏    | 33598/65536 [5:46:12<5:29:15,  1.62it/s]
 51%|█████▏    | 33599/65536 [5:46:12<5:41:36,  1.56it/s]
 51%|█████▏    | 33600/65536 [5:46:13<5:38:26,  1.57it/s]
                                                         
{'loss': 1.6443, 'learning_rate': 5.399167380923227e-07, 'epoch': 2074.07}
+
 51%|█████▏    | 33600/65536 [5:46:13<5:38:26,  1.57it/s]
 51%|█████▏    | 33601/65536 [5:46:13<5:36:18,  1.58it/s]
 51%|█████▏    | 33602/65536 [5:46:14<5:33:01,  1.60it/s]
 51%|█████▏    | 33603/65536 [5:46:15<5:45:03,  1.54it/s]
 51%|█████▏    | 33604/65536 [5:46:15<5:39:01,  1.57it/s]
 51%|█████▏    | 33605/65536 [5:46:16<5:38:08,  1.57it/s]
 51%|█████▏    | 33606/65536 [5:46:17<5:37:50,  1.58it/s]
 51%|█████▏    | 33607/65536 [5:46:17<5:36:51,  1.58it/s]
 51%|█████▏    | 33608/65536 [5:46:18<5:30:46,  1.61it/s]
 51%|█████▏    | 33609/65536 [5:46:18<5:23:05,  1.65it/s]
 51%|█████▏    | 33610/65536 [5:46:19<5:24:44,  1.64it/s]
 51%|█████▏    | 33611/65536 [5:46:20<5:24:06,  1.64it/s]
 51%|█████▏    | 33612/65536 [5:46:20<5:27:23,  1.63it/s]
 51%|█████▏    | 33613/65536 [5:46:21<5:23:08,  1.65it/s]
 51%|█████▏    | 33614/65536 [5:46:22<5:27:07,  1.63it/s]
 51%|█████▏    | 33615/65536 [5:46:22<5:23:06,  1.65it/s]
 51%|█████▏    | 33616/65536 [5:46:23<5:38:39,  1.57it/s]
 51%|█████▏    | 33617/65536 [5:46:23<5:38:22,  1.57it/s]
 51%|█████▏    | 33618/65536 [5:46:24<5:36:23,  1.58it/s]
 51%|█████▏    | 33619/65536 [5:46:25<5:31:27,  1.60it/s]
 51%|█████▏    | 33620/65536 [5:46:25<5:36:29,  1.58it/s]
                                                         
{'loss': 1.5999, 'learning_rate': 5.396412391330966e-07, 'epoch': 2075.31}
+
 51%|█████▏    | 33620/65536 [5:46:25<5:36:29,  1.58it/s]
 51%|█████▏    | 33621/65536 [5:46:26<5:27:14,  1.63it/s]
 51%|█████▏    | 33622/65536 [5:46:26<5:20:23,  1.66it/s]
 51%|█████▏    | 33623/65536 [5:46:27<5:17:42,  1.67it/s]
 51%|█████▏    | 33624/65536 [5:46:28<5:26:31,  1.63it/s]
 51%|█████▏    | 33625/65536 [5:46:28<5:29:57,  1.61it/s]
 51%|█████▏    | 33626/65536 [5:46:29<5:29:56,  1.61it/s]
 51%|█████▏    | 33627/65536 [5:46:30<5:24:06,  1.64it/s]
 51%|█████▏    | 33628/65536 [5:46:30<5:28:28,  1.62it/s]
 51%|█████▏    | 33629/65536 [5:46:31<5:29:39,  1.61it/s]
 51%|█████▏    | 33630/65536 [5:46:31<5:28:19,  1.62it/s]
 51%|█████▏    | 33631/65536 [5:46:32<5:20:18,  1.66it/s]
 51%|█████▏    | 33632/65536 [5:46:33<5:38:01,  1.57it/s]
 51%|█████▏    | 33633/65536 [5:46:33<5:35:50,  1.58it/s]
 51%|█████▏    | 33634/65536 [5:46:34<5:30:30,  1.61it/s]
 51%|█████▏    | 33635/65536 [5:46:35<5:28:48,  1.62it/s]
 51%|█████▏    | 33636/65536 [5:46:35<5:23:53,  1.64it/s]
 51%|█████▏    | 33637/65536 [5:46:36<5:22:48,  1.65it/s]
 51%|█████▏    | 33638/65536 [5:46:36<5:17:16,  1.68it/s]
 51%|█████▏    | 33639/65536 [5:46:37<5:16:14,  1.68it/s]
 51%|█████▏    | 33640/65536 [5:46:38<5:17:36,  1.67it/s]
                                                         
{'loss': 1.6797, 'learning_rate': 5.393657401738705e-07, 'epoch': 2076.54}
+
 51%|█████▏    | 33640/65536 [5:46:38<5:17:36,  1.67it/s]
 51%|█████▏    | 33641/65536 [5:46:38<5:20:51,  1.66it/s]
 51%|█████▏    | 33642/65536 [5:46:39<5:22:55,  1.65it/s]
 51%|█████▏    | 33643/65536 [5:46:39<5:27:59,  1.62it/s]
 51%|█████▏    | 33644/65536 [5:46:40<5:36:50,  1.58it/s]
 51%|█████▏    | 33645/65536 [5:46:41<5:32:05,  1.60it/s]
 51%|█████▏    | 33646/65536 [5:46:41<5:23:02,  1.65it/s]
 51%|█████▏    | 33647/65536 [5:46:42<5:34:34,  1.59it/s]
 51%|█████▏    | 33648/65536 [5:46:43<5:40:02,  1.56it/s]
 51%|█████▏    | 33649/65536 [5:46:43<5:38:42,  1.57it/s]
 51%|█████▏    | 33650/65536 [5:46:44<5:30:44,  1.61it/s]
 51%|█████▏    | 33651/65536 [5:46:44<5:31:52,  1.60it/s]
 51%|█████▏    | 33652/65536 [5:46:45<5:32:28,  1.60it/s]
 51%|█████▏    | 33653/65536 [5:46:46<5:39:05,  1.57it/s]
 51%|█████▏    | 33654/65536 [5:46:46<5:29:28,  1.61it/s]
 51%|█████▏    | 33655/65536 [5:46:47<5:31:30,  1.60it/s]
 51%|█████▏    | 33656/65536 [5:46:48<5:33:12,  1.59it/s]
 51%|█████▏    | 33657/65536 [5:46:48<5:32:08,  1.60it/s]
 51%|█████▏    | 33658/65536 [5:46:49<5:22:55,  1.65it/s]
 51%|█████▏    | 33659/65536 [5:46:49<5:24:06,  1.64it/s]
 51%|█████▏    | 33660/65536 [5:46:50<5:32:13,  1.60it/s]
                                                         
{'loss': 1.6671, 'learning_rate': 5.390902412146443e-07, 'epoch': 2077.78}
+
 51%|█████▏    | 33660/65536 [5:46:50<5:32:13,  1.60it/s]
 51%|█████▏    | 33661/65536 [5:46:51<5:25:55,  1.63it/s]
 51%|█████▏    | 33662/65536 [5:46:51<5:34:34,  1.59it/s]
 51%|█████▏    | 33663/65536 [5:46:52<5:52:14,  1.51it/s]
 51%|█████▏    | 33664/65536 [5:46:53<5:52:39,  1.51it/s]
 51%|█████▏    | 33665/65536 [5:46:53<5:36:52,  1.58it/s]
 51%|█████▏    | 33666/65536 [5:46:54<5:29:21,  1.61it/s]
 51%|█████▏    | 33667/65536 [5:46:54<5:28:59,  1.61it/s]
 51%|█████▏    | 33668/65536 [5:46:55<5:25:11,  1.63it/s]
 51%|█████▏    | 33669/65536 [5:46:56<5:34:37,  1.59it/s]
 51%|█████▏    | 33670/65536 [5:46:56<5:29:50,  1.61it/s]
 51%|█████▏    | 33671/65536 [5:46:57<5:25:08,  1.63it/s]
 51%|█████▏    | 33672/65536 [5:46:57<5:21:17,  1.65it/s]
 51%|█████▏    | 33673/65536 [5:46:58<5:19:38,  1.66it/s]
 51%|█████▏    | 33674/65536 [5:46:59<5:26:02,  1.63it/s]
 51%|█████▏    | 33675/65536 [5:46:59<5:20:57,  1.65it/s]
 51%|█████▏    | 33676/65536 [5:47:00<5:28:24,  1.62it/s]
 51%|█████▏    | 33677/65536 [5:47:01<5:20:02,  1.66it/s]
 51%|█████▏    | 33678/65536 [5:47:01<5:29:42,  1.61it/s]
 51%|█████▏    | 33679/65536 [5:47:02<5:33:50,  1.59it/s]
 51%|█████▏    | 33680/65536 [5:47:03<5:43:49,  1.54it/s]
                                                         
{'loss': 1.6938, 'learning_rate': 5.388147422554182e-07, 'epoch': 2079.01}
+
 51%|█████▏    | 33680/65536 [5:47:03<5:43:49,  1.54it/s]
 51%|█████▏    | 33681/65536 [5:47:03<5:35:45,  1.58it/s]
 51%|█████▏    | 33682/65536 [5:47:04<5:36:32,  1.58it/s]
 51%|█████▏    | 33683/65536 [5:47:04<5:32:19,  1.60it/s]
 51%|█████▏    | 33684/65536 [5:47:05<5:31:12,  1.60it/s]
 51%|█████▏    | 33685/65536 [5:47:06<5:33:50,  1.59it/s]
 51%|█████▏    | 33686/65536 [5:47:06<5:32:50,  1.59it/s]
 51%|█████▏    | 33687/65536 [5:47:07<5:34:50,  1.59it/s]
 51%|█████▏    | 33688/65536 [5:47:07<5:28:06,  1.62it/s]
 51%|█████▏    | 33689/65536 [5:47:08<5:24:56,  1.63it/s]
 51%|█████▏    | 33690/65536 [5:47:09<5:18:31,  1.67it/s]
 51%|█████▏    | 33691/65536 [5:47:09<5:17:15,  1.67it/s]
 51%|█████▏    | 33692/65536 [5:47:10<5:21:44,  1.65it/s]
 51%|█████▏    | 33693/65536 [5:47:10<5:20:54,  1.65it/s]
 51%|█████▏    | 33694/65536 [5:47:11<5:37:02,  1.57it/s]
 51%|█████▏    | 33695/65536 [5:47:12<5:32:05,  1.60it/s]
 51%|█████▏    | 33696/65536 [5:47:12<5:24:59,  1.63it/s]
 51%|█████▏    | 33697/65536 [5:47:13<5:39:33,  1.56it/s]
 51%|█████▏    | 33698/65536 [5:47:14<5:33:16,  1.59it/s]
 51%|█████▏    | 33699/65536 [5:47:14<5:22:08,  1.65it/s]
 51%|█████▏    | 33700/65536 [5:47:15<5:25:42,  1.63it/s]
                                                         
{'loss': 1.6288, 'learning_rate': 5.385392432961919e-07, 'epoch': 2080.25}
+
 51%|█████▏    | 33700/65536 [5:47:15<5:25:42,  1.63it/s]
 51%|█████▏    | 33701/65536 [5:47:15<5:22:55,  1.64it/s]
 51%|█████▏    | 33702/65536 [5:47:16<5:27:54,  1.62it/s]
 51%|█████▏    | 33703/65536 [5:47:17<5:28:04,  1.62it/s]
 51%|█████▏    | 33704/65536 [5:47:17<5:41:49,  1.55it/s]
 51%|█████▏    | 33705/65536 [5:47:18<5:39:44,  1.56it/s]
 51%|█████▏    | 33706/65536 [5:47:19<5:33:09,  1.59it/s]
 51%|█████▏    | 33707/65536 [5:47:19<5:32:35,  1.59it/s]
 51%|█████▏    | 33708/65536 [5:47:20<5:29:50,  1.61it/s]
 51%|█████▏    | 33709/65536 [5:47:20<5:25:01,  1.63it/s]
 51%|█████▏    | 33710/65536 [5:47:21<5:20:51,  1.65it/s]
 51%|█████▏    | 33711/65536 [5:47:22<5:16:29,  1.68it/s]
 51%|█████▏    | 33712/65536 [5:47:22<5:25:08,  1.63it/s]
 51%|█████▏    | 33713/65536 [5:47:23<5:49:02,  1.52it/s]
 51%|█████▏    | 33714/65536 [5:47:24<5:44:46,  1.54it/s]
 51%|█████▏    | 33715/65536 [5:47:24<5:48:30,  1.52it/s]
 51%|█████▏    | 33716/65536 [5:47:25<5:45:00,  1.54it/s]
 51%|█████▏    | 33717/65536 [5:47:26<5:38:15,  1.57it/s]
 51%|█████▏    | 33718/65536 [5:47:26<5:33:57,  1.59it/s]
 51%|█████▏    | 33719/65536 [5:47:27<5:37:47,  1.57it/s]
 51%|█████▏    | 33720/65536 [5:47:27<5:36:46,  1.57it/s]
                                                         
{'loss': 1.6256, 'learning_rate': 5.382637443369658e-07, 'epoch': 2081.48}
+
 51%|█████▏    | 33720/65536 [5:47:27<5:36:46,  1.57it/s]
 51%|█████▏    | 33721/65536 [5:47:28<5:33:47,  1.59it/s]
 51%|█████▏    | 33722/65536 [5:47:29<5:29:23,  1.61it/s]
 51%|█████▏    | 33723/65536 [5:47:29<5:20:13,  1.66it/s]
 51%|█████▏    | 33724/65536 [5:47:30<5:21:09,  1.65it/s]
 51%|█████▏    | 33725/65536 [5:47:30<5:20:50,  1.65it/s]
 51%|█████▏    | 33726/65536 [5:47:31<5:19:11,  1.66it/s]
 51%|█████▏    | 33727/65536 [5:47:32<5:19:51,  1.66it/s]
 51%|█████▏    | 33728/65536 [5:47:32<5:17:13,  1.67it/s]
 51%|█████▏    | 33729/65536 [5:47:33<5:39:22,  1.56it/s]
 51%|█████▏    | 33730/65536 [5:47:34<5:34:52,  1.58it/s]
 51%|█████▏    | 33731/65536 [5:47:34<5:29:57,  1.61it/s]
 51%|█████▏    | 33732/65536 [5:47:35<5:25:41,  1.63it/s]
 51%|█████▏    | 33733/65536 [5:47:36<5:39:13,  1.56it/s]
 51%|█████▏    | 33734/65536 [5:47:36<5:35:10,  1.58it/s]
 51%|█████▏    | 33735/65536 [5:47:37<5:28:57,  1.61it/s]
 51%|█████▏    | 33736/65536 [5:47:37<5:26:24,  1.62it/s]
 51%|█████▏    | 33737/65536 [5:47:38<5:24:28,  1.63it/s]
 51%|█████▏    | 33738/65536 [5:47:39<5:26:17,  1.62it/s]
 51%|█████▏    | 33739/65536 [5:47:39<5:21:07,  1.65it/s]
 51%|█████▏    | 33740/65536 [5:47:40<5:14:53,  1.68it/s]
                                                         
{'loss': 1.6537, 'learning_rate': 5.379882453777396e-07, 'epoch': 2082.72}
+
 51%|█████▏    | 33740/65536 [5:47:40<5:14:53,  1.68it/s]
 51%|█████▏    | 33741/65536 [5:47:40<5:14:54,  1.68it/s]
 51%|█████▏    | 33742/65536 [5:47:41<5:20:32,  1.65it/s]
 51%|█████▏    | 33743/65536 [5:47:42<5:25:07,  1.63it/s]
 51%|█████▏    | 33744/65536 [5:47:42<5:27:23,  1.62it/s]
 51%|█████▏    | 33745/65536 [5:47:43<5:39:53,  1.56it/s]
 51%|█████▏    | 33746/65536 [5:47:43<5:32:34,  1.59it/s]
 51%|█████▏    | 33747/65536 [5:47:44<5:30:35,  1.60it/s]
 51%|█████▏    | 33748/65536 [5:47:45<5:27:35,  1.62it/s]
 51%|█████▏    | 33749/65536 [5:47:45<5:25:37,  1.63it/s]
 51%|█████▏    | 33750/65536 [5:47:46<5:23:46,  1.64it/s]
 51%|█████▏    | 33751/65536 [5:47:47<5:25:05,  1.63it/s]
 52%|█████▏    | 33752/65536 [5:47:47<5:25:18,  1.63it/s]
 52%|█████▏    | 33753/65536 [5:47:48<5:20:12,  1.65it/s]
 52%|█████▏    | 33754/65536 [5:47:48<5:33:35,  1.59it/s]
 52%|█████▏    | 33755/65536 [5:47:49<5:30:23,  1.60it/s]
 52%|█████▏    | 33756/65536 [5:47:50<5:33:48,  1.59it/s]
 52%|█████▏    | 33757/65536 [5:47:50<5:37:15,  1.57it/s]
 52%|█████▏    | 33758/65536 [5:47:51<5:32:19,  1.59it/s]
 52%|█████▏    | 33759/65536 [5:47:52<5:27:53,  1.62it/s]
 52%|█████▏    | 33760/65536 [5:47:52<5:29:48,  1.61it/s]
                                                         
{'loss': 1.618, 'learning_rate': 5.377127464185135e-07, 'epoch': 2083.95}
+
 52%|█████▏    | 33760/65536 [5:47:52<5:29:48,  1.61it/s]
 52%|█████▏    | 33761/65536 [5:47:53<5:33:07,  1.59it/s]
 52%|█████▏    | 33762/65536 [5:47:53<5:29:16,  1.61it/s]
 52%|█████▏    | 33763/65536 [5:47:54<5:29:25,  1.61it/s]
 52%|█████▏    | 33764/65536 [5:47:55<5:35:20,  1.58it/s]
 52%|█████▏    | 33765/65536 [5:47:55<5:27:38,  1.62it/s]
 52%|█████▏    | 33766/65536 [5:47:56<5:22:26,  1.64it/s]
 52%|█████▏    | 33767/65536 [5:47:56<5:22:25,  1.64it/s]
 52%|█████▏    | 33768/65536 [5:47:57<5:28:02,  1.61it/s]
 52%|█████▏    | 33769/65536 [5:47:58<5:25:16,  1.63it/s]
 52%|█████▏    | 33770/65536 [5:47:58<5:26:01,  1.62it/s]
 52%|█████▏    | 33771/65536 [5:47:59<5:32:41,  1.59it/s]
 52%|█████▏    | 33772/65536 [5:48:00<5:33:07,  1.59it/s]
 52%|█████▏    | 33773/65536 [5:48:00<5:35:32,  1.58it/s]
 52%|█████▏    | 33774/65536 [5:48:01<5:37:03,  1.57it/s]
 52%|█████▏    | 33775/65536 [5:48:02<5:31:34,  1.60it/s]
 52%|█████▏    | 33776/65536 [5:48:02<5:32:53,  1.59it/s]
 52%|█████▏    | 33777/65536 [5:48:03<5:28:42,  1.61it/s]
 52%|█████▏    | 33778/65536 [5:48:03<5:38:11,  1.57it/s]
 52%|█████▏    | 33779/65536 [5:48:04<5:36:05,  1.57it/s]
 52%|█████▏    | 33780/65536 [5:48:05<5:29:34,  1.61it/s]
                                                         
{'loss': 1.6029, 'learning_rate': 5.374372474592874e-07, 'epoch': 2085.19}
+
 52%|█████▏    | 33780/65536 [5:48:05<5:29:34,  1.61it/s]
 52%|█████▏    | 33781/65536 [5:48:05<5:29:39,  1.61it/s]
 52%|█████▏    | 33782/65536 [5:48:06<5:21:18,  1.65it/s]
 52%|█████▏    | 33783/65536 [5:48:06<5:21:15,  1.65it/s]
 52%|█████▏    | 33784/65536 [5:48:07<5:28:19,  1.61it/s]
 52%|█████▏    | 33785/65536 [5:48:08<5:25:55,  1.62it/s]
 52%|█████▏    | 33786/65536 [5:48:08<5:25:14,  1.63it/s]
 52%|█████▏    | 33787/65536 [5:48:09<5:23:03,  1.64it/s]
 52%|█████▏    | 33788/65536 [5:48:10<5:17:29,  1.67it/s]
 52%|█████▏    | 33789/65536 [5:48:10<5:22:07,  1.64it/s]
 52%|█████▏    | 33790/65536 [5:48:11<5:28:38,  1.61it/s]
 52%|█████▏    | 33791/65536 [5:48:11<5:29:47,  1.60it/s]
 52%|█████▏    | 33792/65536 [5:48:12<5:27:44,  1.61it/s]
 52%|█████▏    | 33793/65536 [5:48:13<5:28:32,  1.61it/s]
 52%|█████▏    | 33794/65536 [5:48:13<5:36:03,  1.57it/s]
 52%|█████▏    | 33795/65536 [5:48:14<5:37:30,  1.57it/s]
 52%|█████▏    | 33796/65536 [5:48:15<5:40:51,  1.55it/s]
 52%|█████▏    | 33797/65536 [5:48:15<5:34:14,  1.58it/s]
 52%|█████▏    | 33798/65536 [5:48:16<5:33:38,  1.59it/s]
 52%|█████▏    | 33799/65536 [5:48:16<5:25:26,  1.63it/s]
 52%|█████▏    | 33800/65536 [5:48:17<5:26:19,  1.62it/s]
                                                         
{'loss': 1.6189, 'learning_rate': 5.371617485000612e-07, 'epoch': 2086.42}
+
 52%|█████▏    | 33800/65536 [5:48:17<5:26:19,  1.62it/s]
 52%|█████▏    | 33801/65536 [5:48:18<5:26:03,  1.62it/s]
 52%|█████▏    | 33802/65536 [5:48:18<5:26:55,  1.62it/s]
 52%|█████▏    | 33803/65536 [5:48:19<5:32:43,  1.59it/s]
 52%|█████▏    | 33804/65536 [5:48:20<5:27:42,  1.61it/s]
 52%|█████▏    | 33805/65536 [5:48:20<5:36:29,  1.57it/s]
 52%|█████▏    | 33806/65536 [5:48:21<5:29:26,  1.61it/s]
 52%|█████▏    | 33807/65536 [5:48:21<5:22:58,  1.64it/s]
 52%|█████▏    | 33808/65536 [5:48:22<5:18:53,  1.66it/s]
 52%|█████▏    | 33809/65536 [5:48:23<5:18:48,  1.66it/s]
 52%|█████▏    | 33810/65536 [5:48:23<5:29:49,  1.60it/s]
 52%|█████▏    | 33811/65536 [5:48:24<5:20:53,  1.65it/s]
 52%|█████▏    | 33812/65536 [5:48:24<5:18:21,  1.66it/s]
 52%|█████▏    | 33813/65536 [5:48:25<5:25:35,  1.62it/s]
 52%|█████▏    | 33814/65536 [5:48:26<5:27:04,  1.62it/s]
 52%|█████▏    | 33815/65536 [5:48:26<5:25:41,  1.62it/s]
 52%|█████▏    | 33816/65536 [5:48:27<5:32:41,  1.59it/s]
 52%|█████▏    | 33817/65536 [5:48:28<5:27:15,  1.62it/s]
 52%|█████▏    | 33818/65536 [5:48:28<5:26:19,  1.62it/s]
 52%|█████▏    | 33819/65536 [5:48:29<5:41:06,  1.55it/s]
 52%|█████▏    | 33820/65536 [5:48:30<5:40:33,  1.55it/s]
                                                         
{'loss': 1.6409, 'learning_rate': 5.368862495408351e-07, 'epoch': 2087.65}
+
 52%|█████▏    | 33820/65536 [5:48:30<5:40:33,  1.55it/s]
 52%|█████▏    | 33821/65536 [5:48:30<5:34:49,  1.58it/s]
 52%|█████▏    | 33822/65536 [5:48:31<5:26:05,  1.62it/s]
 52%|█████▏    | 33823/65536 [5:48:31<5:23:08,  1.64it/s]
 52%|█████▏    | 33824/65536 [5:48:32<5:21:45,  1.64it/s]
 52%|█████▏    | 33825/65536 [5:48:33<5:21:03,  1.65it/s]
 52%|█████▏    | 33826/65536 [5:48:33<5:35:27,  1.58it/s]
 52%|█████▏    | 33827/65536 [5:48:34<5:49:11,  1.51it/s]
 52%|█████▏    | 33828/65536 [5:48:35<5:46:10,  1.53it/s]
 52%|█████▏    | 33829/65536 [5:48:35<5:40:16,  1.55it/s]
 52%|█████▏    | 33830/65536 [5:48:36<5:41:46,  1.55it/s]
 52%|█████▏    | 33831/65536 [5:48:36<5:39:11,  1.56it/s]
 52%|█████▏    | 33832/65536 [5:48:37<5:29:53,  1.60it/s]
 52%|█████▏    | 33833/65536 [5:48:38<5:27:00,  1.62it/s]
 52%|█████▏    | 33834/65536 [5:48:38<5:26:18,  1.62it/s]
 52%|█████▏    | 33835/65536 [5:48:39<5:22:37,  1.64it/s]
 52%|█████▏    | 33836/65536 [5:48:39<5:24:18,  1.63it/s]
 52%|█████▏    | 33837/65536 [5:48:40<5:19:54,  1.65it/s]
 52%|█████▏    | 33838/65536 [5:48:41<5:17:11,  1.67it/s]
 52%|█████▏    | 33839/65536 [5:48:41<5:12:17,  1.69it/s]
 52%|█████▏    | 33840/65536 [5:48:42<5:27:13,  1.61it/s]
                                                         
{'loss': 1.6319, 'learning_rate': 5.366107505816088e-07, 'epoch': 2088.89}
+
 52%|█████▏    | 33840/65536 [5:48:42<5:27:13,  1.61it/s]
 52%|█████▏    | 33841/65536 [5:48:43<5:22:06,  1.64it/s]
 52%|█████▏    | 33842/65536 [5:48:43<5:29:50,  1.60it/s]
 52%|█████▏    | 33843/65536 [5:48:44<5:30:21,  1.60it/s]
 52%|█████▏    | 33844/65536 [5:48:44<5:29:58,  1.60it/s]
 52%|█████▏    | 33845/65536 [5:48:45<5:32:23,  1.59it/s]
 52%|█████▏    | 33846/65536 [5:48:46<5:30:12,  1.60it/s]
 52%|█████▏    | 33847/65536 [5:48:46<5:25:50,  1.62it/s]
 52%|█████▏    | 33848/65536 [5:48:47<5:20:17,  1.65it/s]
 52%|█████▏    | 33849/65536 [5:48:48<5:27:45,  1.61it/s]
 52%|█████▏    | 33850/65536 [5:48:48<5:23:31,  1.63it/s]
 52%|█████▏    | 33851/65536 [5:48:49<5:23:04,  1.63it/s]
 52%|█████▏    | 33852/65536 [5:48:49<5:31:01,  1.60it/s]
 52%|█████▏    | 33853/65536 [5:48:50<5:28:45,  1.61it/s]
 52%|█████▏    | 33854/65536 [5:48:51<5:20:24,  1.65it/s]
 52%|█████▏    | 33855/65536 [5:48:51<5:21:01,  1.64it/s]
 52%|█████▏    | 33856/65536 [5:48:52<5:21:44,  1.64it/s]
 52%|█████▏    | 33857/65536 [5:48:52<5:21:24,  1.64it/s]
 52%|█████▏    | 33858/65536 [5:48:53<5:20:07,  1.65it/s]
 52%|█████▏    | 33859/65536 [5:48:54<5:33:50,  1.58it/s]
 52%|█████▏    | 33860/65536 [5:48:54<5:32:41,  1.59it/s]
                                                         
{'loss': 1.6839, 'learning_rate': 5.363352516223827e-07, 'epoch': 2090.12}
+
 52%|█████▏    | 33860/65536 [5:48:54<5:32:41,  1.59it/s]
 52%|█████▏    | 33861/65536 [5:48:55<5:35:46,  1.57it/s]
 52%|█████▏    | 33862/65536 [5:48:56<5:35:10,  1.57it/s]
 52%|█████▏    | 33863/65536 [5:48:56<5:34:26,  1.58it/s]
 52%|█████▏    | 33864/65536 [5:48:57<5:33:39,  1.58it/s]
 52%|█████▏    | 33865/65536 [5:48:57<5:33:59,  1.58it/s]
 52%|█████▏    | 33866/65536 [5:48:58<5:29:15,  1.60it/s]
 52%|█████▏    | 33867/65536 [5:48:59<5:25:45,  1.62it/s]
 52%|█████▏    | 33868/65536 [5:48:59<5:20:44,  1.65it/s]
 52%|█████▏    | 33869/65536 [5:49:00<5:33:36,  1.58it/s]
 52%|█████▏    | 33870/65536 [5:49:01<5:25:19,  1.62it/s]
 52%|█████▏    | 33871/65536 [5:49:01<5:20:48,  1.65it/s]
 52%|█████▏    | 33872/65536 [5:49:02<5:21:41,  1.64it/s]
 52%|█████▏    | 33873/65536 [5:49:02<5:30:24,  1.60it/s]
 52%|█████▏    | 33874/65536 [5:49:03<5:20:40,  1.65it/s]
 52%|█████▏    | 33875/65536 [5:49:04<5:34:40,  1.58it/s]
 52%|█████▏    | 33876/65536 [5:49:04<5:30:17,  1.60it/s]
 52%|█████▏    | 33877/65536 [5:49:05<5:33:51,  1.58it/s]
 52%|█████▏    | 33878/65536 [5:49:06<5:28:24,  1.61it/s]
 52%|█████▏    | 33879/65536 [5:49:06<5:38:58,  1.56it/s]
 52%|█████▏    | 33880/65536 [5:49:07<5:27:31,  1.61it/s]
                                                         
{'loss': 1.6088, 'learning_rate': 5.360597526631567e-07, 'epoch': 2091.36}
+
 52%|█████▏    | 33880/65536 [5:49:07<5:27:31,  1.61it/s]
 52%|█████▏    | 33881/65536 [5:49:07<5:30:11,  1.60it/s]
 52%|█████▏    | 33882/65536 [5:49:08<5:19:15,  1.65it/s]
 52%|█████▏    | 33883/65536 [5:49:09<5:17:24,  1.66it/s]
 52%|█████▏    | 33884/65536 [5:49:09<5:14:27,  1.68it/s]
 52%|█████▏    | 33885/65536 [5:49:10<5:16:35,  1.67it/s]
 52%|█████▏    | 33886/65536 [5:49:10<5:15:17,  1.67it/s]
 52%|█████▏    | 33887/65536 [5:49:11<5:13:30,  1.68it/s]
 52%|█████▏    | 33888/65536 [5:49:12<5:24:00,  1.63it/s]
 52%|█████▏    | 33889/65536 [5:49:12<5:22:57,  1.63it/s]
 52%|█████▏    | 33890/65536 [5:49:13<5:31:03,  1.59it/s]
 52%|█████▏    | 33891/65536 [5:49:14<5:33:53,  1.58it/s]
 52%|█████▏    | 33892/65536 [5:49:14<5:32:43,  1.59it/s]
 52%|█████▏    | 33893/65536 [5:49:15<5:29:26,  1.60it/s]
 52%|█████▏    | 33894/65536 [5:49:15<5:25:15,  1.62it/s]
 52%|█████▏    | 33895/65536 [5:49:16<5:26:01,  1.62it/s]
 52%|█████▏    | 33896/65536 [5:49:17<5:26:55,  1.61it/s]
 52%|█████▏    | 33897/65536 [5:49:17<5:30:05,  1.60it/s]
 52%|█████▏    | 33898/65536 [5:49:18<5:22:43,  1.63it/s]
 52%|█████▏    | 33899/65536 [5:49:18<5:25:50,  1.62it/s]
 52%|█████▏    | 33900/65536 [5:49:19<5:15:52,  1.67it/s]
                                                         
{'loss': 1.6999, 'learning_rate': 5.357842537039305e-07, 'epoch': 2092.59}
+
 52%|█████▏    | 33900/65536 [5:49:19<5:15:52,  1.67it/s]
 52%|█████▏    | 33901/65536 [5:49:20<5:15:42,  1.67it/s]
 52%|█████▏    | 33902/65536 [5:49:20<5:21:00,  1.64it/s]
 52%|█████▏    | 33903/65536 [5:49:21<5:16:14,  1.67it/s]
 52%|█████▏    | 33904/65536 [5:49:21<5:14:26,  1.68it/s]
 52%|█████▏    | 33905/65536 [5:49:22<5:19:26,  1.65it/s]
 52%|█████▏    | 33906/65536 [5:49:23<5:25:42,  1.62it/s]
 52%|█████▏    | 33907/65536 [5:49:23<5:36:36,  1.57it/s]
 52%|█████▏    | 33908/65536 [5:49:24<5:37:44,  1.56it/s]
 52%|█████▏    | 33909/65536 [5:49:25<5:33:50,  1.58it/s]
 52%|█████▏    | 33910/65536 [5:49:25<5:22:05,  1.64it/s]
 52%|█████▏    | 33911/65536 [5:49:26<5:27:51,  1.61it/s]
 52%|█████▏    | 33912/65536 [5:49:27<5:38:14,  1.56it/s]
 52%|█████▏    | 33913/65536 [5:49:27<5:32:30,  1.59it/s]
 52%|█████▏    | 33914/65536 [5:49:28<5:34:44,  1.57it/s]
 52%|█████▏    | 33915/65536 [5:49:28<5:33:24,  1.58it/s]
 52%|█████▏    | 33916/65536 [5:49:29<5:23:53,  1.63it/s]
 52%|█████▏    | 33917/65536 [5:49:30<5:21:05,  1.64it/s]
 52%|█████▏    | 33918/65536 [5:49:30<5:30:25,  1.59it/s]
 52%|█████▏    | 33919/65536 [5:49:31<5:24:34,  1.62it/s]
 52%|█████▏    | 33920/65536 [5:49:31<5:24:12,  1.63it/s]
                                                         
{'loss': 1.6296, 'learning_rate': 5.355087547447044e-07, 'epoch': 2093.83}
+
 52%|█████▏    | 33920/65536 [5:49:31<5:24:12,  1.63it/s]
 52%|█████▏    | 33921/65536 [5:49:32<5:25:15,  1.62it/s]
 52%|█████▏    | 33922/65536 [5:49:33<5:19:00,  1.65it/s]
 52%|█████▏    | 33923/65536 [5:49:33<5:27:14,  1.61it/s]
 52%|█████▏    | 33924/65536 [5:49:34<5:33:37,  1.58it/s]
 52%|█████▏    | 33925/65536 [5:49:35<5:32:48,  1.58it/s]
 52%|█████▏    | 33926/65536 [5:49:35<5:23:02,  1.63it/s]
 52%|█████▏    | 33927/65536 [5:49:36<5:22:28,  1.63it/s]
 52%|█████▏    | 33928/65536 [5:49:36<5:15:35,  1.67it/s]
 52%|█████▏    | 33929/65536 [5:49:37<5:12:31,  1.69it/s]
 52%|█████▏    | 33930/65536 [5:49:38<5:20:59,  1.64it/s]
 52%|█████▏    | 33931/65536 [5:49:38<5:21:20,  1.64it/s]
 52%|█████▏    | 33932/65536 [5:49:39<5:30:03,  1.60it/s]
 52%|█████▏    | 33933/65536 [5:49:39<5:24:14,  1.62it/s]
 52%|█████▏    | 33934/65536 [5:49:40<5:22:04,  1.64it/s]
 52%|█████▏    | 33935/65536 [5:49:41<5:23:33,  1.63it/s]
 52%|█████▏    | 33936/65536 [5:49:41<5:20:09,  1.65it/s]
 52%|█████▏    | 33937/65536 [5:49:42<5:22:26,  1.63it/s]
 52%|█████▏    | 33938/65536 [5:49:43<5:29:12,  1.60it/s]
 52%|█████▏    | 33939/65536 [5:49:43<5:26:36,  1.61it/s]
 52%|█████▏    | 33940/65536 [5:49:44<5:34:45,  1.57it/s]
                                                         
{'loss': 1.6524, 'learning_rate': 5.352332557854782e-07, 'epoch': 2095.06}
+
 52%|█████▏    | 33940/65536 [5:49:44<5:34:45,  1.57it/s]
 52%|█████▏    | 33941/65536 [5:49:44<5:39:43,  1.55it/s]
 52%|█████▏    | 33942/65536 [5:49:45<5:31:46,  1.59it/s]
 52%|█████▏    | 33943/65536 [5:49:46<5:28:24,  1.60it/s]
 52%|█████▏    | 33944/65536 [5:49:46<5:22:51,  1.63it/s]
 52%|█████▏    | 33945/65536 [5:49:47<5:16:50,  1.66it/s]
 52%|█████▏    | 33946/65536 [5:49:47<5:18:14,  1.65it/s]
 52%|█████▏    | 33947/65536 [5:49:48<5:22:50,  1.63it/s]
 52%|█████▏    | 33948/65536 [5:49:49<5:24:07,  1.62it/s]
 52%|█████▏    | 33949/65536 [5:49:49<5:23:56,  1.63it/s]
 52%|█████▏    | 33950/65536 [5:49:50<5:22:59,  1.63it/s]
 52%|█████▏    | 33951/65536 [5:49:51<5:32:45,  1.58it/s]
 52%|█████▏    | 33952/65536 [5:49:51<5:38:42,  1.55it/s]
 52%|█████▏    | 33953/65536 [5:49:52<5:29:58,  1.60it/s]
 52%|█████▏    | 33954/65536 [5:49:52<5:22:01,  1.63it/s]
 52%|█████▏    | 33955/65536 [5:49:53<5:23:37,  1.63it/s]
 52%|█████▏    | 33956/65536 [5:49:54<5:40:56,  1.54it/s]
 52%|█████▏    | 33957/65536 [5:49:54<5:34:38,  1.57it/s]
 52%|█████▏    | 33958/65536 [5:49:55<5:38:02,  1.56it/s]
 52%|█████▏    | 33959/65536 [5:49:56<5:29:36,  1.60it/s]
 52%|█████▏    | 33960/65536 [5:49:56<5:19:37,  1.65it/s]
                                                         
{'loss': 1.6768, 'learning_rate': 5.349577568262519e-07, 'epoch': 2096.3}
+
 52%|█████▏    | 33960/65536 [5:49:56<5:19:37,  1.65it/s]
 52%|█████▏    | 33961/65536 [5:49:57<5:24:42,  1.62it/s]
 52%|█████▏    | 33962/65536 [5:49:57<5:24:12,  1.62it/s]
 52%|█████▏    | 33963/65536 [5:49:58<5:26:26,  1.61it/s]
 52%|█████▏    | 33964/65536 [5:49:59<5:29:22,  1.60it/s]
 52%|█████▏    | 33965/65536 [5:49:59<5:28:10,  1.60it/s]
 52%|█████▏    | 33966/65536 [5:50:00<5:23:59,  1.62it/s]
 52%|█████▏    | 33967/65536 [5:50:01<5:23:53,  1.62it/s]
 52%|█████▏    | 33968/65536 [5:50:01<5:29:07,  1.60it/s]
 52%|█████▏    | 33969/65536 [5:50:02<5:21:07,  1.64it/s]
 52%|█████▏    | 33970/65536 [5:50:02<5:16:39,  1.66it/s]
 52%|█████▏    | 33971/65536 [5:50:03<5:19:29,  1.65it/s]
 52%|█████▏    | 33972/65536 [5:50:04<5:25:58,  1.61it/s]
 52%|█████▏    | 33973/65536 [5:50:04<5:22:26,  1.63it/s]
 52%|█████▏    | 33974/65536 [5:50:05<5:30:11,  1.59it/s]
 52%|█████▏    | 33975/65536 [5:50:05<5:20:04,  1.64it/s]
 52%|█████▏    | 33976/65536 [5:50:06<5:17:04,  1.66it/s]
 52%|█████▏    | 33977/65536 [5:50:07<5:26:31,  1.61it/s]
 52%|█████▏    | 33978/65536 [5:50:07<5:24:28,  1.62it/s]
 52%|█████▏    | 33979/65536 [5:50:08<5:29:11,  1.60it/s]
 52%|█████▏    | 33980/65536 [5:50:09<5:23:58,  1.62it/s]
                                                         
{'loss': 1.6304, 'learning_rate': 5.346822578670257e-07, 'epoch': 2097.53}
+
 52%|█████▏    | 33980/65536 [5:50:09<5:23:58,  1.62it/s]
 52%|█████▏    | 33981/65536 [5:50:09<5:21:19,  1.64it/s]
 52%|█████▏    | 33982/65536 [5:50:10<5:23:51,  1.62it/s]
 52%|█████▏    | 33983/65536 [5:50:10<5:25:10,  1.62it/s]
 52%|█████▏    | 33984/65536 [5:50:11<5:18:57,  1.65it/s]
 52%|█████▏    | 33985/65536 [5:50:12<5:16:31,  1.66it/s]
 52%|█████▏    | 33986/65536 [5:50:12<5:18:38,  1.65it/s]
 52%|█████▏    | 33987/65536 [5:50:13<5:26:23,  1.61it/s]
 52%|█████▏    | 33988/65536 [5:50:14<5:34:41,  1.57it/s]
 52%|█████▏    | 33989/65536 [5:50:14<5:35:16,  1.57it/s]
 52%|█████▏    | 33990/65536 [5:50:15<5:28:24,  1.60it/s]
 52%|█████▏    | 33991/65536 [5:50:15<5:21:19,  1.64it/s]
 52%|█████▏    | 33992/65536 [5:50:16<5:21:41,  1.63it/s]
 52%|█████▏    | 33993/65536 [5:50:17<5:21:38,  1.63it/s]
 52%|█████▏    | 33994/65536 [5:50:17<5:24:27,  1.62it/s]
 52%|█████▏    | 33995/65536 [5:50:18<5:18:46,  1.65it/s]
 52%|█████▏    | 33996/65536 [5:50:18<5:22:42,  1.63it/s]
 52%|█████▏    | 33997/65536 [5:50:19<5:21:09,  1.64it/s]
 52%|█████▏    | 33998/65536 [5:50:20<5:12:06,  1.68it/s]
 52%|█████▏    | 33999/65536 [5:50:20<5:15:28,  1.67it/s]
 52%|█████▏    | 34000/65536 [5:50:21<5:18:48,  1.65it/s]
                                                         
{'loss': 1.6687, 'learning_rate': 5.344067589077996e-07, 'epoch': 2098.77}
+
 52%|█████▏    | 34000/65536 [5:50:21<5:18:48,  1.65it/s]
 52%|█████▏    | 34001/65536 [5:50:21<5:17:50,  1.65it/s]
 52%|█████▏    | 34002/65536 [5:50:22<5:17:30,  1.66it/s]
 52%|█████▏    | 34003/65536 [5:50:23<5:28:39,  1.60it/s]
 52%|█████▏    | 34004/65536 [5:50:23<5:49:45,  1.50it/s]
 52%|█████▏    | 34005/65536 [5:50:24<5:38:24,  1.55it/s]
 52%|█████▏    | 34006/65536 [5:50:25<5:35:58,  1.56it/s]
 52%|█████▏    | 34007/65536 [5:50:25<5:24:43,  1.62it/s]
 52%|█████▏    | 34008/65536 [5:50:26<5:26:07,  1.61it/s]
 52%|█████▏    | 34009/65536 [5:50:26<5:30:19,  1.59it/s]
 52%|█████▏    | 34010/65536 [5:50:27<5:26:20,  1.61it/s]
 52%|█████▏    | 34011/65536 [5:50:28<5:26:09,  1.61it/s]
 52%|█████▏    | 34012/65536 [5:50:28<5:22:29,  1.63it/s]
 52%|█████▏    | 34013/65536 [5:50:29<5:19:43,  1.64it/s]
 52%|█████▏    | 34014/65536 [5:50:30<5:17:40,  1.65it/s]
 52%|█████▏    | 34015/65536 [5:50:30<5:19:37,  1.64it/s]
 52%|█████▏    | 34016/65536 [5:50:31<5:18:34,  1.65it/s]
 52%|█████▏    | 34017/65536 [5:50:31<5:11:44,  1.69it/s]
 52%|█████▏    | 34018/65536 [5:50:32<5:19:49,  1.64it/s]
 52%|█████▏    | 34019/65536 [5:50:33<5:33:13,  1.58it/s]
 52%|█████▏    | 34020/65536 [5:50:33<5:34:00,  1.57it/s]
                                                         
{'loss': 1.639, 'learning_rate': 5.341312599485735e-07, 'epoch': 2100.0}
+
 52%|█████▏    | 34020/65536 [5:50:33<5:34:00,  1.57it/s]
 52%|█████▏    | 34021/65536 [5:50:34<5:35:18,  1.57it/s]
 52%|█████▏    | 34022/65536 [5:50:35<5:28:33,  1.60it/s]
 52%|█████▏    | 34023/65536 [5:50:35<5:34:27,  1.57it/s]
 52%|█████▏    | 34024/65536 [5:50:36<5:25:48,  1.61it/s]
 52%|█████▏    | 34025/65536 [5:50:36<5:30:35,  1.59it/s]
 52%|█████▏    | 34026/65536 [5:50:37<5:26:08,  1.61it/s]
 52%|█████▏    | 34027/65536 [5:50:38<5:27:25,  1.60it/s]
 52%|█████▏    | 34028/65536 [5:50:38<5:22:27,  1.63it/s]
 52%|█████▏    | 34029/65536 [5:50:39<5:20:57,  1.64it/s]
 52%|█████▏    | 34030/65536 [5:50:39<5:25:09,  1.61it/s]
 52%|█████▏    | 34031/65536 [5:50:40<5:22:46,  1.63it/s]
 52%|█████▏    | 34032/65536 [5:50:41<5:29:29,  1.59it/s]
 52%|█████▏    | 34033/65536 [5:50:41<5:22:32,  1.63it/s]
 52%|█████▏    | 34034/65536 [5:50:42<5:21:29,  1.63it/s]
 52%|█████▏    | 34035/65536 [5:50:43<5:24:18,  1.62it/s]
 52%|█████▏    | 34036/65536 [5:50:43<5:21:53,  1.63it/s]
 52%|█████▏    | 34037/65536 [5:50:44<5:25:51,  1.61it/s]
 52%|█████▏    | 34038/65536 [5:50:44<5:30:03,  1.59it/s]
 52%|█████▏    | 34039/65536 [5:50:45<5:30:39,  1.59it/s]
 52%|█████▏    | 34040/65536 [5:50:46<5:34:45,  1.57it/s]
                                                         
{'loss': 1.6512, 'learning_rate': 5.338557609893473e-07, 'epoch': 2101.23}
+
 52%|█████▏    | 34040/65536 [5:50:46<5:34:45,  1.57it/s]
 52%|█████▏    | 34041/65536 [5:50:46<5:28:03,  1.60it/s]
 52%|█████▏    | 34042/65536 [5:50:47<5:26:33,  1.61it/s]
 52%|█████▏    | 34043/65536 [5:50:48<5:30:16,  1.59it/s]
 52%|█████▏    | 34044/65536 [5:50:48<5:32:57,  1.58it/s]
 52%|█████▏    | 34045/65536 [5:50:49<5:36:56,  1.56it/s]
 52%|█████▏    | 34046/65536 [5:50:49<5:26:21,  1.61it/s]
 52%|█████▏    | 34047/65536 [5:50:50<5:28:03,  1.60it/s]
 52%|█████▏    | 34048/65536 [5:50:51<5:24:35,  1.62it/s]
 52%|█████▏    | 34049/65536 [5:50:51<5:21:31,  1.63it/s]
 52%|█████▏    | 34050/65536 [5:50:52<5:24:29,  1.62it/s]
 52%|█████▏    | 34051/65536 [5:50:53<5:22:29,  1.63it/s]
 52%|█████▏    | 34052/65536 [5:50:53<5:24:47,  1.62it/s]
 52%|█████▏    | 34053/65536 [5:50:54<5:31:28,  1.58it/s]
 52%|█████▏    | 34054/65536 [5:50:54<5:23:19,  1.62it/s]
 52%|█████▏    | 34055/65536 [5:50:55<5:21:40,  1.63it/s]
 52%|█████▏    | 34056/65536 [5:50:56<5:24:58,  1.61it/s]
 52%|█████▏    | 34057/65536 [5:50:56<5:22:05,  1.63it/s]
 52%|█████▏    | 34058/65536 [5:50:57<5:28:11,  1.60it/s]
 52%|█████▏    | 34059/65536 [5:50:58<5:38:06,  1.55it/s]
 52%|█████▏    | 34060/65536 [5:50:58<5:37:11,  1.56it/s]
                                                         
{'loss': 1.6303, 'learning_rate': 5.335802620301212e-07, 'epoch': 2102.47}
+
 52%|█████▏    | 34060/65536 [5:50:58<5:37:11,  1.56it/s]
 52%|█████▏    | 34061/65536 [5:50:59<5:31:32,  1.58it/s]
 52%|█████▏    | 34062/65536 [5:50:59<5:23:06,  1.62it/s]
 52%|█████▏    | 34063/65536 [5:51:00<5:21:16,  1.63it/s]
 52%|█████▏    | 34064/65536 [5:51:01<5:26:50,  1.60it/s]
 52%|█████▏    | 34065/65536 [5:51:01<5:26:07,  1.61it/s]
 52%|█████▏    | 34066/65536 [5:51:02<5:25:40,  1.61it/s]
 52%|█████▏    | 34067/65536 [5:51:03<5:22:52,  1.62it/s]
 52%|█████▏    | 34068/65536 [5:51:03<5:18:46,  1.65it/s]
 52%|█████▏    | 34069/65536 [5:51:04<5:27:54,  1.60it/s]
 52%|█████▏    | 34070/65536 [5:51:04<5:35:01,  1.57it/s]
 52%|█████▏    | 34071/65536 [5:51:05<5:29:55,  1.59it/s]
 52%|█████▏    | 34072/65536 [5:51:06<5:28:12,  1.60it/s]
 52%|█████▏    | 34073/65536 [5:51:06<5:24:30,  1.62it/s]
 52%|█████▏    | 34074/65536 [5:51:07<5:25:29,  1.61it/s]
 52%|█████▏    | 34075/65536 [5:51:08<5:25:22,  1.61it/s]
 52%|█████▏    | 34076/65536 [5:51:08<5:21:38,  1.63it/s]
 52%|█████▏    | 34077/65536 [5:51:09<5:19:19,  1.64it/s]
 52%|█████▏    | 34078/65536 [5:51:09<5:16:34,  1.66it/s]
 52%|█████▏    | 34079/65536 [5:51:10<5:18:14,  1.65it/s]
 52%|█████▏    | 34080/65536 [5:51:11<5:18:18,  1.65it/s]
                                                         
{'loss': 1.6472, 'learning_rate': 5.33304763070895e-07, 'epoch': 2103.7}
+
 52%|█████▏    | 34080/65536 [5:51:11<5:18:18,  1.65it/s]
 52%|█████▏    | 34081/65536 [5:51:11<5:24:07,  1.62it/s]
 52%|█████▏    | 34082/65536 [5:51:12<5:27:41,  1.60it/s]
 52%|█████▏    | 34083/65536 [5:51:12<5:17:11,  1.65it/s]
 52%|█████▏    | 34084/65536 [5:51:13<5:27:24,  1.60it/s]
 52%|█████▏    | 34085/65536 [5:51:14<5:40:37,  1.54it/s]
 52%|█████▏    | 34086/65536 [5:51:14<5:37:54,  1.55it/s]
 52%|█████▏    | 34087/65536 [5:51:15<5:29:30,  1.59it/s]
 52%|█████▏    | 34088/65536 [5:51:16<5:26:09,  1.61it/s]
 52%|█████▏    | 34089/65536 [5:51:16<5:26:09,  1.61it/s]
 52%|█████▏    | 34090/65536 [5:51:17<5:22:14,  1.63it/s]
 52%|█████▏    | 34091/65536 [5:51:17<5:22:45,  1.62it/s]
 52%|█████▏    | 34092/65536 [5:51:18<5:19:12,  1.64it/s]
 52%|█████▏    | 34093/65536 [5:51:19<5:24:55,  1.61it/s]
 52%|█████▏    | 34094/65536 [5:51:19<5:23:25,  1.62it/s]
 52%|█████▏    | 34095/65536 [5:51:20<5:28:42,  1.59it/s]
 52%|█████▏    | 34096/65536 [5:51:21<5:31:02,  1.58it/s]
 52%|█████▏    | 34097/65536 [5:51:21<5:30:54,  1.58it/s]
 52%|█████▏    | 34098/65536 [5:51:22<5:25:45,  1.61it/s]
 52%|█████▏    | 34099/65536 [5:51:22<5:30:23,  1.59it/s]
 52%|█████▏    | 34100/65536 [5:51:23<5:25:11,  1.61it/s]
                                                         
{'loss': 1.6132, 'learning_rate': 5.330292641116689e-07, 'epoch': 2104.94}
+
 52%|█████▏    | 34100/65536 [5:51:23<5:25:11,  1.61it/s]
 52%|█████▏    | 34101/65536 [5:51:24<5:26:54,  1.60it/s]
 52%|█████▏    | 34102/65536 [5:51:24<5:37:40,  1.55it/s]
 52%|█████▏    | 34103/65536 [5:51:25<5:31:20,  1.58it/s]
 52%|█████▏    | 34104/65536 [5:51:26<5:37:08,  1.55it/s]
 52%|█████▏    | 34105/65536 [5:51:26<5:27:53,  1.60it/s]
 52%|█████▏    | 34106/65536 [5:51:27<5:19:52,  1.64it/s]
 52%|█████▏    | 34107/65536 [5:51:27<5:14:07,  1.67it/s]
 52%|█████▏    | 34108/65536 [5:51:28<5:17:00,  1.65it/s]
 52%|█████▏    | 34109/65536 [5:51:29<5:15:55,  1.66it/s]
 52%|█████▏    | 34110/65536 [5:51:29<5:18:40,  1.64it/s]
 52%|█████���    | 34111/65536 [5:51:30<5:20:24,  1.63it/s]
 52%|█████▏    | 34112/65536 [5:51:30<5:20:56,  1.63it/s]
 52%|█████▏    | 34113/65536 [5:51:31<5:17:23,  1.65it/s]
 52%|█████▏    | 34114/65536 [5:51:32<5:13:46,  1.67it/s]
 52%|█████▏    | 34115/65536 [5:51:32<5:21:49,  1.63it/s]
 52%|█████▏    | 34116/65536 [5:51:33<5:25:28,  1.61it/s]
 52%|█████▏    | 34117/65536 [5:51:34<5:32:56,  1.57it/s]
 52%|█████▏    | 34118/65536 [5:51:34<5:40:35,  1.54it/s]
 52%|█████▏    | 34119/65536 [5:51:35<5:36:04,  1.56it/s]
 52%|█████▏    | 34120/65536 [5:51:35<5:33:28,  1.57it/s]
                                                         
{'loss': 1.6328, 'learning_rate': 5.327537651524427e-07, 'epoch': 2106.17}
+
 52%|█████▏    | 34120/65536 [5:51:35<5:33:28,  1.57it/s]
 52%|█████▏    | 34121/65536 [5:51:36<5:37:14,  1.55it/s]
 52%|█████▏    | 34122/65536 [5:51:37<5:25:44,  1.61it/s]
 52%|█████▏    | 34123/65536 [5:51:37<5:23:09,  1.62it/s]
 52%|█████▏    | 34124/65536 [5:51:38<5:26:50,  1.60it/s]
 52%|█████▏    | 34125/65536 [5:51:39<5:24:04,  1.62it/s]
 52%|█████▏    | 34126/65536 [5:51:39<5:19:36,  1.64it/s]
 52%|█████▏    | 34127/65536 [5:51:40<5:19:58,  1.64it/s]
 52%|█████▏    | 34128/65536 [5:51:40<5:16:31,  1.65it/s]
 52%|█████▏    | 34129/65536 [5:51:41<5:20:23,  1.63it/s]
 52%|█████▏    | 34130/65536 [5:51:42<5:24:48,  1.61it/s]
 52%|█████▏    | 34131/65536 [5:51:42<5:23:04,  1.62it/s]
 52%|█████▏    | 34132/65536 [5:51:43<5:22:23,  1.62it/s]
 52%|█████▏    | 34133/65536 [5:51:44<5:28:09,  1.59it/s]
 52%|█████▏    | 34134/65536 [5:51:44<5:45:26,  1.52it/s]
 52%|█████▏    | 34135/65536 [5:51:45<5:42:11,  1.53it/s]
 52%|█████▏    | 34136/65536 [5:51:45<5:32:19,  1.57it/s]
 52%|█████▏    | 34137/65536 [5:51:46<5:31:42,  1.58it/s]
 52%|█████▏    | 34138/65536 [5:51:47<5:24:05,  1.61it/s]
 52%|█████▏    | 34139/65536 [5:51:47<5:23:55,  1.62it/s]
 52%|█████▏    | 34140/65536 [5:51:48<5:24:16,  1.61it/s]
                                                         
{'loss': 1.6175, 'learning_rate': 5.324782661932166e-07, 'epoch': 2107.41}
+
 52%|█████▏    | 34140/65536 [5:51:48<5:24:16,  1.61it/s]
 52%|█████▏    | 34141/65536 [5:51:49<5:34:56,  1.56it/s]
 52%|█████▏    | 34142/65536 [5:51:49<5:28:47,  1.59it/s]
 52%|█████▏    | 34143/65536 [5:51:50<5:31:47,  1.58it/s]
 52%|█████▏    | 34144/65536 [5:51:50<5:21:26,  1.63it/s]
 52%|█████▏    | 34145/65536 [5:51:51<5:20:47,  1.63it/s]
 52%|█████▏    | 34146/65536 [5:51:52<5:18:41,  1.64it/s]
 52%|█████▏    | 34147/65536 [5:51:52<5:13:03,  1.67it/s]
 52%|█████▏    | 34148/65536 [5:51:53<5:16:12,  1.65it/s]
 52%|█████▏    | 34149/65536 [5:51:53<5:16:45,  1.65it/s]
 52%|█████▏    | 34150/65536 [5:51:54<5:28:55,  1.59it/s]
 52%|█████▏    | 34151/65536 [5:51:55<5:24:46,  1.61it/s]
 52%|█████▏    | 34152/65536 [5:51:55<5:32:37,  1.57it/s]
 52%|█████▏    | 34153/65536 [5:51:56<5:42:15,  1.53it/s]
 52%|█████▏    | 34154/65536 [5:51:57<5:37:16,  1.55it/s]
 52%|█████▏    | 34155/65536 [5:51:57<5:37:21,  1.55it/s]
 52%|█████▏    | 34156/65536 [5:51:58<5:25:56,  1.60it/s]
 52%|█████▏    | 34157/65536 [5:51:59<5:21:03,  1.63it/s]
 52%|█████▏    | 34158/65536 [5:51:59<5:15:21,  1.66it/s]
 52%|█████▏    | 34159/65536 [5:52:00<5:14:18,  1.66it/s]
 52%|█████▏    | 34160/65536 [5:52:00<5:26:27,  1.60it/s]
                                                         
{'loss': 1.6078, 'learning_rate': 5.322027672339905e-07, 'epoch': 2108.64}
+
 52%|█████▏    | 34160/65536 [5:52:00<5:26:27,  1.60it/s]
 52%|█████▏    | 34161/65536 [5:52:01<5:30:58,  1.58it/s]
 52%|█████▏    | 34162/65536 [5:52:02<5:23:28,  1.62it/s]
 52%|█████▏    | 34163/65536 [5:52:02<5:16:01,  1.65it/s]
 52%|█████▏    | 34164/65536 [5:52:03<5:18:09,  1.64it/s]
 52%|█████▏    | 34165/65536 [5:52:03<5:21:55,  1.62it/s]
 52%|█████▏    | 34166/65536 [5:52:04<5:32:45,  1.57it/s]
 52%|█████▏    | 34167/65536 [5:52:05<5:30:55,  1.58it/s]
 52%|█████▏    | 34168/65536 [5:52:05<5:18:57,  1.64it/s]
 52%|█████▏    | 34169/65536 [5:52:06<5:24:43,  1.61it/s]
 52%|█████▏    | 34170/65536 [5:52:07<5:19:16,  1.64it/s]
 52%|█████▏    | 34171/65536 [5:52:07<5:15:14,  1.66it/s]
 52%|█████▏    | 34172/65536 [5:52:08<5:15:10,  1.66it/s]
 52%|█████▏    | 34173/65536 [5:52:08<5:11:21,  1.68it/s]
 52%|█████▏    | 34174/65536 [5:52:09<5:17:57,  1.64it/s]
 52%|█████▏    | 34175/65536 [5:52:10<5:22:06,  1.62it/s]
 52%|█████▏    | 34176/65536 [5:52:10<5:19:48,  1.63it/s]
 52%|█████▏    | 34177/65536 [5:52:11<5:30:34,  1.58it/s]
 52%|█████▏    | 34178/65536 [5:52:11<5:22:43,  1.62it/s]
 52%|█████▏    | 34179/65536 [5:52:12<5:27:47,  1.59it/s]
 52%|█████▏    | 34180/65536 [5:52:13<5:26:28,  1.60it/s]
                                                         
{'loss': 1.6795, 'learning_rate': 5.319272682747643e-07, 'epoch': 2109.88}
+
 52%|█████▏    | 34180/65536 [5:52:13<5:26:28,  1.60it/s]
 52%|█████▏    | 34181/65536 [5:52:13<5:20:58,  1.63it/s]
 52%|█████▏    | 34182/65536 [5:52:14<5:20:22,  1.63it/s]
 52%|█████▏    | 34183/65536 [5:52:15<5:34:03,  1.56it/s]
 52%|█████▏    | 34184/65536 [5:52:15<5:33:40,  1.57it/s]
 52%|█████▏    | 34185/65536 [5:52:16<5:38:46,  1.54it/s]
 52%|█████▏    | 34186/65536 [5:52:17<5:31:15,  1.58it/s]
 52%|█████▏    | 34187/65536 [5:52:17<5:29:12,  1.59it/s]
 52%|█████▏    | 34188/65536 [5:52:18<5:31:48,  1.57it/s]
 52%|█████▏    | 34189/65536 [5:52:18<5:23:28,  1.62it/s]
 52%|█████▏    | 34190/65536 [5:52:19<5:33:08,  1.57it/s]
 52%|█████▏    | 34191/65536 [5:52:20<5:26:27,  1.60it/s]
 52%|█████▏    | 34192/65536 [5:52:20<5:23:53,  1.61it/s]
 52%|█████▏    | 34193/65536 [5:52:21<5:21:53,  1.62it/s]
 52%|█████▏    | 34194/65536 [5:52:21<5:15:03,  1.66it/s]
 52%|█████▏    | 34195/65536 [5:52:22<5:16:05,  1.65it/s]
 52%|█████▏    | 34196/65536 [5:52:23<5:15:55,  1.65it/s]
 52%|█████▏    | 34197/65536 [5:52:23<5:22:38,  1.62it/s]
 52%|█████▏    | 34198/65536 [5:52:24<5:14:53,  1.66it/s]
 52%|█████▏    | 34199/65536 [5:52:25<5:27:33,  1.59it/s]
 52%|█████▏    | 34200/65536 [5:52:25<5:24:24,  1.61it/s]
                                                         
{'loss': 1.6347, 'learning_rate': 5.316517693155383e-07, 'epoch': 2111.11}
+
 52%|█████▏    | 34200/65536 [5:52:25<5:24:24,  1.61it/s]
 52%|█████▏    | 34201/65536 [5:52:26<5:28:03,  1.59it/s]
 52%|█████▏    | 34202/65536 [5:52:26<5:33:04,  1.57it/s]
 52%|█████▏    | 34203/65536 [5:52:27<5:29:29,  1.58it/s]
 52%|█████▏    | 34204/65536 [5:52:28<5:20:20,  1.63it/s]
 52%|█████▏    | 34205/65536 [5:52:28<5:16:06,  1.65it/s]
 52%|█████▏    | 34206/65536 [5:52:29<5:16:39,  1.65it/s]
 52%|█████▏    | 34207/65536 [5:52:29<5:15:30,  1.65it/s]
 52%|█████▏    | 34208/65536 [5:52:30<5:23:47,  1.61it/s]
 52%|█████▏    | 34209/65536 [5:52:31<5:26:57,  1.60it/s]
 52%|█████▏    | 34210/65536 [5:52:31<5:22:51,  1.62it/s]
 52%|█████▏    | 34211/65536 [5:52:32<5:21:01,  1.63it/s]
 52%|█████▏    | 34212/65536 [5:52:33<5:19:55,  1.63it/s]
 52%|█████▏    | 34213/65536 [5:52:33<5:19:20,  1.63it/s]
 52%|█████▏    | 34214/65536 [5:52:34<5:16:51,  1.65it/s]
 52%|█████▏    | 34215/65536 [5:52:34<5:30:44,  1.58it/s]
 52%|█████▏    | 34216/65536 [5:52:35<5:33:41,  1.56it/s]
 52%|█████▏    | 34217/65536 [5:52:36<5:30:45,  1.58it/s]
 52%|█████▏    | 34218/65536 [5:52:36<5:17:49,  1.64it/s]
 52%|█████▏    | 34219/65536 [5:52:37<5:24:37,  1.61it/s]
 52%|█████▏    | 34220/65536 [5:52:38<5:26:16,  1.60it/s]
                                                         
{'loss': 1.6583, 'learning_rate': 5.31376270356312e-07, 'epoch': 2112.35}
+
 52%|█████▏    | 34220/65536 [5:52:38<5:26:16,  1.60it/s]
 52%|█████▏    | 34221/65536 [5:52:38<5:26:04,  1.60it/s]
 52%|█████▏    | 34222/65536 [5:52:39<5:18:37,  1.64it/s]
 52%|█████▏    | 34223/65536 [5:52:39<5:19:24,  1.63it/s]
 52%|█████▏    | 34224/65536 [5:52:40<5:20:26,  1.63it/s]
 52%|█████▏    | 34225/65536 [5:52:41<5:12:11,  1.67it/s]
 52%|█████▏    | 34226/65536 [5:52:41<5:11:39,  1.67it/s]
 52%|█████▏    | 34227/65536 [5:52:42<5:12:11,  1.67it/s]
 52%|█████▏    | 34228/65536 [5:52:42<5:10:31,  1.68it/s]
 52%|█████▏    | 34229/65536 [5:52:43<5:14:56,  1.66it/s]
 52%|█████▏    | 34230/65536 [5:52:44<5:25:36,  1.60it/s]
 52%|█████▏    | 34231/65536 [5:52:44<5:35:45,  1.55it/s]
 52%|█████▏    | 34232/65536 [5:52:45<5:27:07,  1.59it/s]
 52%|█████▏    | 34233/65536 [5:52:46<5:28:20,  1.59it/s]
 52%|█████▏    | 34234/65536 [5:52:46<5:19:58,  1.63it/s]
 52%|█████▏    | 34235/65536 [5:52:47<5:27:54,  1.59it/s]
 52%|█████▏    | 34236/65536 [5:52:47<5:26:57,  1.60it/s]
 52%|█████▏    | 34237/65536 [5:52:48<5:22:39,  1.62it/s]
 52%|█████▏    | 34238/65536 [5:52:49<5:26:52,  1.60it/s]
 52%|█████▏    | 34239/65536 [5:52:49<5:24:50,  1.61it/s]
 52%|█████▏    | 34240/65536 [5:52:50<5:21:50,  1.62it/s]
                                                         
{'loss': 1.649, 'learning_rate': 5.311007713970858e-07, 'epoch': 2113.58}
+
 52%|█████▏    | 34240/65536 [5:52:50<5:21:50,  1.62it/s]
 52%|█████▏    | 34241/65536 [5:52:51<5:26:43,  1.60it/s]
 52%|█████▏    | 34242/65536 [5:52:51<5:21:08,  1.62it/s]
 52%|█████▏    | 34243/65536 [5:52:52<5:16:50,  1.65it/s]
 52%|█████▏    | 34244/65536 [5:52:52<5:21:04,  1.62it/s]
 52%|█████▏    | 34245/65536 [5:52:53<5:15:39,  1.65it/s]
 52%|█████▏    | 34246/65536 [5:52:54<5:16:08,  1.65it/s]
 52%|█████▏    | 34247/65536 [5:52:54<5:27:20,  1.59it/s]
 52%|█████▏    | 34248/65536 [5:52:55<5:28:24,  1.59it/s]
 52%|█████▏    | 34249/65536 [5:52:55<5:23:08,  1.61it/s]
 52%|█████▏    | 34250/65536 [5:52:56<5:18:14,  1.64it/s]
 52%|█████▏    | 34251/65536 [5:52:57<5:20:38,  1.63it/s]
 52%|█████▏    | 34252/65536 [5:52:57<5:16:12,  1.65it/s]
 52%|█████▏    | 34253/65536 [5:52:58<5:17:17,  1.64it/s]
 52%|█████▏    | 34254/65536 [5:52:59<5:26:20,  1.60it/s]
 52%|█████▏    | 34255/65536 [5:52:59<5:39:09,  1.54it/s]
 52%|█████▏    | 34256/65536 [5:53:00<5:28:26,  1.59it/s]
 52%|█████▏    | 34257/65536 [5:53:00<5:21:39,  1.62it/s]
 52%|█████▏    | 34258/65536 [5:53:01<5:22:50,  1.61it/s]
 52%|█████▏    | 34259/65536 [5:53:02<5:18:48,  1.64it/s]
 52%|█████▏    | 34260/65536 [5:53:02<5:17:45,  1.64it/s]
                                                         
{'loss': 1.6458, 'learning_rate': 5.308252724378597e-07, 'epoch': 2114.81}
+
 52%|█████▏    | 34260/65536 [5:53:02<5:17:45,  1.64it/s]
 52%|█████▏    | 34261/65536 [5:53:03<5:16:52,  1.64it/s]
 52%|█████▏    | 34262/65536 [5:53:04<5:26:22,  1.60it/s]
 52%|█████▏    | 34263/65536 [5:53:04<5:22:58,  1.61it/s]
 52%|█████▏    | 34264/65536 [5:53:05<5:38:15,  1.54it/s]
 52%|█████▏    | 34265/65536 [5:53:05<5:31:21,  1.57it/s]
 52%|█████▏    | 34266/65536 [5:53:06<5:26:40,  1.60it/s]
 52%|█████▏    | 34267/65536 [5:53:07<5:22:54,  1.61it/s]
 52%|█████▏    | 34268/65536 [5:53:07<5:24:56,  1.60it/s]
 52%|█████▏    | 34269/65536 [5:53:08<5:25:29,  1.60it/s]
 52%|█████▏    | 34270/65536 [5:53:09<5:22:38,  1.62it/s]
 52%|█████▏    | 34271/65536 [5:53:09<5:28:17,  1.59it/s]
 52%|█████▏    | 34272/65536 [5:53:10<5:33:45,  1.56it/s]
 52%|█████▏    | 34273/65536 [5:53:10<5:26:52,  1.59it/s]
 52%|█████▏    | 34274/65536 [5:53:11<5:26:54,  1.59it/s]
 52%|█████▏    | 34275/65536 [5:53:12<5:22:40,  1.61it/s]
 52%|█████▏    | 34276/65536 [5:53:12<5:22:12,  1.62it/s]
 52%|█████▏    | 34277/65536 [5:53:13<5:21:47,  1.62it/s]
 52%|█████▏    | 34278/65536 [5:53:13<5:17:44,  1.64it/s]
 52%|█████▏    | 34279/65536 [5:53:14<5:13:30,  1.66it/s]
 52%|█████▏    | 34280/65536 [5:53:15<5:28:29,  1.59it/s]
                                                         
{'loss': 1.6741, 'learning_rate': 5.305497734786335e-07, 'epoch': 2116.05}
+
 52%|█████▏    | 34280/65536 [5:53:15<5:28:29,  1.59it/s]
 52%|█████▏    | 34281/65536 [5:53:15<5:22:10,  1.62it/s]
 52%|█████▏    | 34282/65536 [5:53:16<5:33:18,  1.56it/s]
 52%|█████▏    | 34283/65536 [5:53:17<5:23:17,  1.61it/s]
 52%|█████▏    | 34284/65536 [5:53:17<5:18:41,  1.63it/s]
 52%|█████▏    | 34285/65536 [5:53:18<5:18:20,  1.64it/s]
 52%|█████▏    | 34286/65536 [5:53:18<5:19:33,  1.63it/s]
 52%|█████▏    | 34287/65536 [5:53:19<5:20:25,  1.63it/s]
 52%|█████▏    | 34288/65536 [5:53:20<5:30:12,  1.58it/s]
 52%|█████▏    | 34289/65536 [5:53:20<5:22:53,  1.61it/s]
 52%|█████▏    | 34290/65536 [5:53:21<5:30:06,  1.58it/s]
 52%|█████▏    | 34291/65536 [5:53:22<5:28:29,  1.59it/s]
 52%|█████▏    | 34292/65536 [5:53:22<5:26:07,  1.60it/s]
 52%|█████▏    | 34293/65536 [5:53:23<5:22:24,  1.62it/s]
 52%|█████▏    | 34294/65536 [5:53:23<5:23:26,  1.61it/s]
 52%|█████▏    | 34295/65536 [5:53:24<5:17:30,  1.64it/s]
 52%|█████▏    | 34296/65536 [5:53:25<5:30:45,  1.57it/s]
 52%|█████▏    | 34297/65536 [5:53:25<5:22:11,  1.62it/s]
 52%|█████▏    | 34298/65536 [5:53:26<5:20:12,  1.63it/s]
 52%|█████▏    | 34299/65536 [5:53:27<5:17:07,  1.64it/s]
 52%|█████▏    | 34300/65536 [5:53:27<5:08:25,  1.69it/s]
                                                         
{'loss': 1.6458, 'learning_rate': 5.302742745194074e-07, 'epoch': 2117.28}
+
 52%|█████▏    | 34300/65536 [5:53:27<5:08:25,  1.69it/s]
 52%|█████▏    | 34301/65536 [5:53:28<5:19:39,  1.63it/s]
 52%|█████▏    | 34302/65536 [5:53:28<5:33:55,  1.56it/s]
 52%|█████▏    | 34303/65536 [5:53:29<5:35:05,  1.55it/s]
 52%|█████▏    | 34304/65536 [5:53:30<5:23:32,  1.61it/s]
 52%|█████▏    | 34305/65536 [5:53:30<5:27:38,  1.59it/s]
 52%|█████▏    | 34306/65536 [5:53:31<5:23:28,  1.61it/s]
 52%|█████▏    | 34307/65536 [5:53:31<5:17:03,  1.64it/s]
 52%|█████▏    | 34308/65536 [5:53:32<5:17:32,  1.64it/s]
 52%|█████▏    | 34309/65536 [5:53:33<5:21:24,  1.62it/s]
 52%|█████▏    | 34310/65536 [5:53:33<5:16:48,  1.64it/s]
 52%|█████▏    | 34311/65536 [5:53:34<5:16:50,  1.64it/s]
 52%|█████▏    | 34312/65536 [5:53:35<5:31:35,  1.57it/s]
 52%|█████▏    | 34313/65536 [5:53:35<5:31:26,  1.57it/s]
 52%|█████▏    | 34314/65536 [5:53:36<5:25:37,  1.60it/s]
 52%|█████▏    | 34315/65536 [5:53:36<5:18:14,  1.64it/s]
 52%|█████▏    | 34316/65536 [5:53:37<5:22:53,  1.61it/s]
 52%|█████▏    | 34317/65536 [5:53:38<5:22:40,  1.61it/s]
 52%|█████▏    | 34318/65536 [5:53:38<5:27:58,  1.59it/s]
 52%|█████▏    | 34319/65536 [5:53:39<5:22:20,  1.61it/s]
 52%|█████▏    | 34320/65536 [5:53:40<5:25:26,  1.60it/s]
                                                         
{'loss': 1.6725, 'learning_rate': 5.299987755601812e-07, 'epoch': 2118.52}
+
 52%|█████▏    | 34320/65536 [5:53:40<5:25:26,  1.60it/s]
 52%|█████▏    | 34321/65536 [5:53:40<5:21:28,  1.62it/s]
 52%|█████▏    | 34322/65536 [5:53:41<5:22:18,  1.61it/s]
 52%|█████▏    | 34323/65536 [5:53:41<5:23:47,  1.61it/s]
 52%|█████▏    | 34324/65536 [5:53:42<5:30:16,  1.58it/s]
 52%|█████▏    | 34325/65536 [5:53:43<5:19:42,  1.63it/s]
 52%|█████▏    | 34326/65536 [5:53:43<5:20:02,  1.63it/s]
 52%|█████▏    | 34327/65536 [5:53:44<5:21:15,  1.62it/s]
 52%|█████▏    | 34328/65536 [5:53:45<5:26:33,  1.59it/s]
 52%|█████▏    | 34329/65536 [5:53:45<5:23:06,  1.61it/s]
 52%|█████▏    | 34330/65536 [5:53:46<5:22:50,  1.61it/s]
 52%|█████▏    | 34331/65536 [5:53:46<5:17:29,  1.64it/s]
 52%|█████▏    | 34332/65536 [5:53:47<5:17:20,  1.64it/s]
 52%|█████▏    | 34333/65536 [5:53:48<5:15:31,  1.65it/s]
 52%|█████▏    | 34334/65536 [5:53:48<5:19:20,  1.63it/s]
 52%|█████▏    | 34335/65536 [5:53:49<5:17:48,  1.64it/s]
 52%|█████▏    | 34336/65536 [5:53:49<5:22:38,  1.61it/s]
 52%|█████▏    | 34337/65536 [5:53:50<5:21:18,  1.62it/s]
 52%|█████▏    | 34338/65536 [5:53:51<5:15:25,  1.65it/s]
 52%|█████▏    | 34339/65536 [5:53:51<5:14:36,  1.65it/s]
 52%|█████▏    | 34340/65536 [5:53:52<5:22:35,  1.61it/s]
                                                         
{'loss': 1.6372, 'learning_rate': 5.297232766009551e-07, 'epoch': 2119.75}
+
 52%|█████▏    | 34340/65536 [5:53:52<5:22:35,  1.61it/s]
 52%|█████▏    | 34341/65536 [5:53:53<5:20:20,  1.62it/s]
 52%|█████▏    | 34342/65536 [5:53:53<5:21:59,  1.61it/s]
 52%|█████▏    | 34343/65536 [5:53:54<5:27:03,  1.59it/s]
 52%|█████▏    | 34344/65536 [5:53:54<5:30:28,  1.57it/s]
 52%|█████▏    | 34345/65536 [5:53:55<5:35:12,  1.55it/s]
 52%|█████▏    | 34346/65536 [5:53:56<5:41:52,  1.52it/s]
 52%|█████▏    | 34347/65536 [5:53:56<5:36:29,  1.54it/s]
 52%|█████▏    | 34348/65536 [5:53:57<5:30:07,  1.57it/s]
 52%|█████▏    | 34349/65536 [5:53:58<5:22:16,  1.61it/s]
 52%|█████▏    | 34350/65536 [5:53:58<5:27:21,  1.59it/s]
 52%|█████▏    | 34351/65536 [5:53:59<5:29:42,  1.58it/s]
 52%|█████▏    | 34352/65536 [5:54:00<5:29:18,  1.58it/s]
 52%|█████▏    | 34353/65536 [5:54:00<5:21:37,  1.62it/s]
 52%|█████▏    | 34354/65536 [5:54:01<5:12:26,  1.66it/s]
 52%|█████▏    | 34355/65536 [5:54:01<5:15:20,  1.65it/s]
 52%|█████▏    | 34356/65536 [5:54:02<5:16:39,  1.64it/s]
 52%|█████▏    | 34357/65536 [5:54:03<5:15:30,  1.65it/s]
 52%|█████▏    | 34358/65536 [5:54:03<5:23:07,  1.61it/s]
 52%|█████▏    | 34359/65536 [5:54:04<5:21:24,  1.62it/s]
 52%|█████▏    | 34360/65536 [5:54:04<5:13:34,  1.66it/s]
                                                         
{'loss': 1.5928, 'learning_rate': 5.294477776417289e-07, 'epoch': 2120.99}
+
 52%|█████▏    | 34360/65536 [5:54:04<5:13:34,  1.66it/s]
 52%|█████▏    | 34361/65536 [5:54:05<5:24:27,  1.60it/s]
 52%|█████▏    | 34362/65536 [5:54:06<5:19:01,  1.63it/s]
 52%|█████▏    | 34363/65536 [5:54:06<5:15:32,  1.65it/s]
 52%|█████▏    | 34364/65536 [5:54:07<5:12:25,  1.66it/s]
 52%|█████▏    | 34365/65536 [5:54:07<5:15:36,  1.65it/s]
 52%|█████▏    | 34366/65536 [5:54:08<5:27:56,  1.58it/s]
 52%|█████▏    | 34367/65536 [5:54:09<5:37:04,  1.54it/s]
 52%|█████▏    | 34368/65536 [5:54:09<5:35:38,  1.55it/s]
 52%|█████▏    | 34369/65536 [5:54:10<5:30:21,  1.57it/s]
 52%|█████▏    | 34370/65536 [5:54:11<5:33:39,  1.56it/s]
 52%|█████▏    | 34371/65536 [5:54:11<5:27:58,  1.58it/s]
 52%|█████▏    | 34372/65536 [5:54:12<5:24:09,  1.60it/s]
 52%|█████▏    | 34373/65536 [5:54:13<5:24:28,  1.60it/s]
 52%|█████▏    | 34374/65536 [5:54:13<5:24:06,  1.60it/s]
 52%|█████▏    | 34375/65536 [5:54:14<5:25:39,  1.59it/s]
 52%|█████▏    | 34376/65536 [5:54:14<5:21:55,  1.61it/s]
 52%|█████▏    | 34377/65536 [5:54:15<5:29:32,  1.58it/s]
 52%|█████▏    | 34378/65536 [5:54:16<5:25:04,  1.60it/s]
 52%|█████▏    | 34379/65536 [5:54:16<5:20:33,  1.62it/s]
 52%|█████▏    | 34380/65536 [5:54:17<5:29:13,  1.58it/s]
                                                         
{'loss': 1.6509, 'learning_rate': 5.291722786825027e-07, 'epoch': 2122.22}
+
 52%|█████▏    | 34380/65536 [5:54:17<5:29:13,  1.58it/s]
 52%|█████▏    | 34381/65536 [5:54:18<5:26:18,  1.59it/s]
 52%|█████▏    | 34382/65536 [5:54:18<5:24:00,  1.60it/s]
 52%|█████▏    | 34383/65536 [5:54:19<5:17:38,  1.63it/s]
 52%|█████▏    | 34384/65536 [5:54:19<5:16:05,  1.64it/s]
 52%|█████▏    | 34385/65536 [5:54:20<5:13:38,  1.66it/s]
 52%|█████▏    | 34386/65536 [5:54:21<5:11:27,  1.67it/s]
 52%|█████▏    | 34387/65536 [5:54:21<5:14:29,  1.65it/s]
 52%|█████▏    | 34388/65536 [5:54:22<5:15:33,  1.65it/s]
 52%|█████▏    | 34389/65536 [5:54:22<5:19:07,  1.63it/s]
 52%|█████▏    | 34390/65536 [5:54:23<5:24:51,  1.60it/s]
 52%|█████▏    | 34391/65536 [5:54:24<5:16:34,  1.64it/s]
 52%|█████▏    | 34392/65536 [5:54:24<5:15:13,  1.65it/s]
 52%|█████▏    | 34393/65536 [5:54:25<5:22:16,  1.61it/s]
 52%|█████▏    | 34394/65536 [5:54:26<5:25:37,  1.59it/s]
 52%|█████▏    | 34395/65536 [5:54:26<5:19:57,  1.62it/s]
 52%|█████▏    | 34396/65536 [5:54:27<5:21:12,  1.62it/s]
 52%|█████▏    | 34397/65536 [5:54:27<5:21:53,  1.61it/s]
 52%|█████▏    | 34398/65536 [5:54:28<5:22:13,  1.61it/s]
 52%|█████▏    | 34399/65536 [5:54:29<5:19:40,  1.62it/s]
 52%|█████▏    | 34400/65536 [5:54:29<5:14:22,  1.65it/s]
                                                         
{'loss': 1.688, 'learning_rate': 5.288967797232766e-07, 'epoch': 2123.46}
+
 52%|█████▏    | 34400/65536 [5:54:29<5:14:22,  1.65it/s]
 52%|█████▏    | 34401/65536 [5:54:30<5:22:38,  1.61it/s]
 52%|█████▏    | 34402/65536 [5:54:30<5:20:41,  1.62it/s]
 52%|█████▏    | 34403/65536 [5:54:31<5:18:16,  1.63it/s]
 52%|█████▏    | 34404/65536 [5:54:32<5:21:46,  1.61it/s]
 52%|█████▏    | 34405/65536 [5:54:32<5:29:22,  1.58it/s]
 52%|█████▏    | 34406/65536 [5:54:33<5:24:27,  1.60it/s]
 53%|█████▎    | 34407/65536 [5:54:34<5:20:55,  1.62it/s]
 53%|█████▎    | 34408/65536 [5:54:34<5:11:56,  1.66it/s]
 53%|█████▎    | 34409/65536 [5:54:35<5:20:40,  1.62it/s]
 53%|█████▎    | 34410/65536 [5:54:35<5:28:10,  1.58it/s]
 53%|█████▎    | 34411/65536 [5:54:36<5:20:45,  1.62it/s]
 53%|█████▎    | 34412/65536 [5:54:37<5:19:52,  1.62it/s]
 53%|█████▎    | 34413/65536 [5:54:37<5:24:02,  1.60it/s]
 53%|█████▎    | 34414/65536 [5:54:38<5:20:49,  1.62it/s]
 53%|█████▎    | 34415/65536 [5:54:38<5:14:58,  1.65it/s]
 53%|█████▎    | 34416/65536 [5:54:39<5:18:19,  1.63it/s]
 53%|█████▎    | 34417/65536 [5:54:40<5:28:00,  1.58it/s]
 53%|█████▎    | 34418/65536 [5:54:40<5:27:33,  1.58it/s]
 53%|█████▎    | 34419/65536 [5:54:41<5:32:05,  1.56it/s]
 53%|█████▎    | 34420/65536 [5:54:42<5:25:53,  1.59it/s]
                                                         
{'loss': 1.6257, 'learning_rate': 5.286212807640504e-07, 'epoch': 2124.69}
+
 53%|█████▎    | 34420/65536 [5:54:42<5:25:53,  1.59it/s]
 53%|█████▎    | 34421/65536 [5:54:42<5:29:21,  1.57it/s]
 53%|█████▎    | 34422/65536 [5:54:43<5:24:52,  1.60it/s]
 53%|█████▎    | 34423/65536 [5:54:44<5:22:09,  1.61it/s]
 53%|█████▎    | 34424/65536 [5:54:44<5:17:08,  1.64it/s]
 53%|█████▎    | 34425/65536 [5:54:45<5:13:17,  1.66it/s]
 53%|█████▎    | 34426/65536 [5:54:45<5:21:00,  1.62it/s]
 53%|█████▎    | 34427/65536 [5:54:46<5:20:58,  1.62it/s]
 53%|█████▎    | 34428/65536 [5:54:47<5:19:21,  1.62it/s]
 53%|█████▎    | 34429/65536 [5:54:47<5:16:32,  1.64it/s]
 53%|█████▎    | 34430/65536 [5:54:48<5:17:50,  1.63it/s]
 53%|█████▎    | 34431/65536 [5:54:48<5:20:42,  1.62it/s]
 53%|█████▎    | 34432/65536 [5:54:49<5:22:11,  1.61it/s]
 53%|█████▎    | 34433/65536 [5:54:50<5:11:36,  1.66it/s]
 53%|█████▎    | 34434/65536 [5:54:50<5:08:00,  1.68it/s]
 53%|█████▎    | 34435/65536 [5:54:51<5:19:48,  1.62it/s]
 53%|█████▎    | 34436/65536 [5:54:51<5:16:31,  1.64it/s]
 53%|█████▎    | 34437/65536 [5:54:52<5:13:46,  1.65it/s]
 53%|█████▎    | 34438/65536 [5:54:53<5:08:05,  1.68it/s]
 53%|█████▎    | 34439/65536 [5:54:53<5:19:30,  1.62it/s]
 53%|█████▎    | 34440/65536 [5:54:54<5:23:39,  1.60it/s]
                                                         
{'loss': 1.6655, 'learning_rate': 5.283457818048243e-07, 'epoch': 2125.93}
+
 53%|█████▎    | 34440/65536 [5:54:54<5:23:39,  1.60it/s]
 53%|█████▎    | 34441/65536 [5:54:55<5:21:05,  1.61it/s]
 53%|█████▎    | 34442/65536 [5:54:55<5:25:22,  1.59it/s]
 53%|█████▎    | 34443/65536 [5:54:56<5:19:45,  1.62it/s]
 53%|█████▎    | 34444/65536 [5:54:56<5:19:34,  1.62it/s]
 53%|█████▎    | 34445/65536 [5:54:57<5:07:26,  1.69it/s]
 53%|█████▎    | 34446/65536 [5:54:58<5:07:04,  1.69it/s]
 53%|█████▎    | 34447/65536 [5:54:58<5:07:36,  1.68it/s]
 53%|█████▎    | 34448/65536 [5:54:59<5:07:43,  1.68it/s]
 53%|█████▎    | 34449/65536 [5:54:59<5:10:06,  1.67it/s]
 53%|█████▎    | 34450/65536 [5:55:00<5:16:11,  1.64it/s]
 53%|█████▎    | 34451/65536 [5:55:01<5:21:24,  1.61it/s]
 53%|█████▎    | 34452/65536 [5:55:01<5:18:42,  1.63it/s]
 53%|█████▎    | 34453/65536 [5:55:02<5:25:01,  1.59it/s]
 53%|█████▎    | 34454/65536 [5:55:03<5:23:40,  1.60it/s]
 53%|█████▎    | 34455/65536 [5:55:03<5:25:57,  1.59it/s]
 53%|█████▎    | 34456/65536 [5:55:04<5:23:24,  1.60it/s]
 53%|█████▎    | 34457/65536 [5:55:04<5:20:13,  1.62it/s]
 53%|█████▎    | 34458/65536 [5:55:05<5:32:32,  1.56it/s]
 53%|█████▎    | 34459/65536 [5:55:06<5:24:40,  1.60it/s]
 53%|█████▎    | 34460/65536 [5:55:06<5:20:01,  1.62it/s]
                                                         
{'loss': 1.6803, 'learning_rate': 5.280702828455981e-07, 'epoch': 2127.16}
+
 53%|█████▎    | 34460/65536 [5:55:06<5:20:01,  1.62it/s]
 53%|█████▎    | 34461/65536 [5:55:07<5:19:07,  1.62it/s]
 53%|█████▎    | 34462/65536 [5:55:08<5:29:23,  1.57it/s]
 53%|█████▎    | 34463/65536 [5:55:08<5:26:36,  1.59it/s]
 53%|█████▎    | 34464/65536 [5:55:09<5:16:20,  1.64it/s]
 53%|█████▎    | 34465/65536 [5:55:09<5:21:05,  1.61it/s]
 53%|█████▎    | 34466/65536 [5:55:10<5:18:41,  1.62it/s]
 53%|█████▎    | 34467/65536 [5:55:11<5:20:33,  1.62it/s]
 53%|█████▎    | 34468/65536 [5:55:11<5:22:43,  1.60it/s]
 53%|█████▎    | 34469/65536 [5:55:12<5:24:13,  1.60it/s]
 53%|█████▎    | 34470/65536 [5:55:12<5:22:16,  1.61it/s]
 53%|█████▎    | 34471/65536 [5:55:13<5:14:06,  1.65it/s]
 53%|█████▎    | 34472/65536 [5:55:14<5:09:59,  1.67it/s]
 53%|█████▎    | 34473/65536 [5:55:14<5:12:50,  1.65it/s]
 53%|█████▎    | 34474/65536 [5:55:15<5:28:31,  1.58it/s]
 53%|█████▎    | 34475/65536 [5:55:16<5:37:10,  1.54it/s]
 53%|█████▎    | 34476/65536 [5:55:16<5:32:05,  1.56it/s]
 53%|█████▎    | 34477/65536 [5:55:17<5:25:34,  1.59it/s]
 53%|█████▎    | 34478/65536 [5:55:17<5:21:31,  1.61it/s]
 53%|█████▎    | 34479/65536 [5:55:18<5:19:01,  1.62it/s]
 53%|█████▎    | 34480/65536 [5:55:19<5:14:42,  1.64it/s]
                                                         
{'loss': 1.5928, 'learning_rate': 5.27794783886372e-07, 'epoch': 2128.4}
+
 53%|█████▎    | 34480/65536 [5:55:19<5:14:42,  1.64it/s]
 53%|█████▎    | 34481/65536 [5:55:19<5:14:22,  1.65it/s]
 53%|█████▎    | 34482/65536 [5:55:20<5:15:18,  1.64it/s]
 53%|█████▎    | 34483/65536 [5:55:21<5:23:02,  1.60it/s]
 53%|█████▎    | 34484/65536 [5:55:21<5:19:24,  1.62it/s]
 53%|█████▎    | 34485/65536 [5:55:22<5:28:40,  1.57it/s]
 53%|█████▎    | 34486/65536 [5:55:22<5:20:16,  1.62it/s]
 53%|█████▎    | 34487/65536 [5:55:23<5:19:59,  1.62it/s]
 53%|█████▎    | 34488/65536 [5:55:24<5:15:47,  1.64it/s]
 53%|█████▎    | 34489/65536 [5:55:24<5:09:47,  1.67it/s]
 53%|█████▎    | 34490/65536 [5:55:25<5:20:41,  1.61it/s]
 53%|█████▎    | 34491/65536 [5:55:25<5:20:13,  1.62it/s]
 53%|█████▎    | 34492/65536 [5:55:26<5:26:38,  1.58it/s]
 53%|█████▎    | 34493/65536 [5:55:27<5:19:01,  1.62it/s]
 53%|█████▎    | 34494/65536 [5:55:27<5:18:30,  1.62it/s]
 53%|█████▎    | 34495/65536 [5:55:28<5:16:35,  1.63it/s]
 53%|█████▎    | 34496/65536 [5:55:29<5:13:49,  1.65it/s]
 53%|█████▎    | 34497/65536 [5:55:29<5:21:26,  1.61it/s]
 53%|█████▎    | 34498/65536 [5:55:30<5:15:14,  1.64it/s]
 53%|█████▎    | 34499/65536 [5:55:30<5:18:11,  1.63it/s]
 53%|█████▎    | 34500/65536 [5:55:31<5:16:14,  1.64it/s]
                                                         
{'loss': 1.6843, 'learning_rate': 5.275192849271457e-07, 'epoch': 2129.63}
+
 53%|█████▎    | 34500/65536 [5:55:31<5:16:14,  1.64it/s]
 53%|█████▎    | 34501/65536 [5:55:32<5:12:08,  1.66it/s]
 53%|█████▎    | 34502/65536 [5:55:32<5:11:00,  1.66it/s]
 53%|█████▎    | 34503/65536 [5:55:33<5:08:20,  1.68it/s]
 53%|█████▎    | 34504/65536 [5:55:33<5:10:05,  1.67it/s]
 53%|█████▎    | 34505/65536 [5:55:34<5:16:07,  1.64it/s]
 53%|█████▎    | 34506/65536 [5:55:35<5:11:18,  1.66it/s]
 53%|█████▎    | 34507/65536 [5:55:35<5:24:51,  1.59it/s]
 53%|█████▎    | 34508/65536 [5:55:36<5:21:50,  1.61it/s]
 53%|█████▎    | 34509/65536 [5:55:36<5:15:34,  1.64it/s]
 53%|█████▎    | 34510/65536 [5:55:37<5:14:55,  1.64it/s]
 53%|█████▎    | 34511/65536 [5:55:38<5:17:24,  1.63it/s]
 53%|█████▎    | 34512/65536 [5:55:38<5:14:18,  1.65it/s]
 53%|█████▎    | 34513/65536 [5:55:39<5:24:51,  1.59it/s]
 53%|█████▎    | 34514/65536 [5:55:40<5:17:34,  1.63it/s]
 53%|█████▎    | 34515/65536 [5:55:40<5:09:12,  1.67it/s]
 53%|█████▎    | 34516/65536 [5:55:41<5:11:53,  1.66it/s]
 53%|█████▎    | 34517/65536 [5:55:41<5:14:21,  1.64it/s]
 53%|█████▎    | 34518/65536 [5:55:42<5:18:08,  1.62it/s]
 53%|█████▎    | 34519/65536 [5:55:43<5:18:27,  1.62it/s]
 53%|█████▎    | 34520/65536 [5:55:43<5:28:46,  1.57it/s]
                                                         
{'loss': 1.6173, 'learning_rate': 5.272437859679196e-07, 'epoch': 2130.86}
+
 53%|█████▎    | 34520/65536 [5:55:43<5:28:46,  1.57it/s]
 53%|█████▎    | 34521/65536 [5:55:44<5:20:51,  1.61it/s]
 53%|█████▎    | 34522/65536 [5:55:44<5:17:32,  1.63it/s]
 53%|█████▎    | 34523/65536 [5:55:45<5:27:31,  1.58it/s]
 53%|█████▎    | 34524/65536 [5:55:46<5:26:53,  1.58it/s]
 53%|█████▎    | 34525/65536 [5:55:46<5:19:47,  1.62it/s]
 53%|█████▎    | 34526/65536 [5:55:47<5:15:00,  1.64it/s]
 53%|█████▎    | 34527/65536 [5:55:48<5:10:00,  1.67it/s]
 53%|█████▎    | 34528/65536 [5:55:48<5:06:07,  1.69it/s]
 53%|█████▎    | 34529/65536 [5:55:49<5:11:22,  1.66it/s]
 53%|█████▎    | 34530/65536 [5:55:49<5:20:33,  1.61it/s]
 53%|█████▎    | 34531/65536 [5:55:50<5:18:07,  1.62it/s]
 53%|█████▎    | 34532/65536 [5:55:51<5:19:48,  1.62it/s]
 53%|█████▎    | 34533/65536 [5:55:51<5:12:53,  1.65it/s]
 53%|█████▎    | 34534/65536 [5:55:52<5:22:49,  1.60it/s]
 53%|█████▎    | 34535/65536 [5:55:53<5:28:15,  1.57it/s]
 53%|█████▎    | 34536/65536 [5:55:53<5:27:48,  1.58it/s]
 53%|█████▎    | 34537/65536 [5:55:54<5:25:33,  1.59it/s]
 53%|█████▎    | 34538/65536 [5:55:54<5:17:21,  1.63it/s]
 53%|█████▎    | 34539/65536 [5:55:55<5:31:32,  1.56it/s]
 53%|█████▎    | 34540/65536 [5:55:56<5:23:47,  1.60it/s]
                                                         
{'loss': 1.6635, 'learning_rate': 5.269682870086935e-07, 'epoch': 2132.1}
+
 53%|█████▎    | 34540/65536 [5:55:56<5:23:47,  1.60it/s]
 53%|█████▎    | 34541/65536 [5:55:56<5:12:15,  1.65it/s]
 53%|█████▎    | 34542/65536 [5:55:57<5:13:24,  1.65it/s]
 53%|█████▎    | 34543/65536 [5:55:58<5:28:56,  1.57it/s]
 53%|█████▎    | 34544/65536 [5:55:58<5:24:18,  1.59it/s]
 53%|█████▎    | 34545/65536 [5:55:59<5:19:14,  1.62it/s]
 53%|█████▎    | 34546/65536 [5:55:59<5:18:29,  1.62it/s]
 53%|█████▎    | 34547/65536 [5:56:00<5:14:34,  1.64it/s]
 53%|█████▎    | 34548/65536 [5:56:01<5:19:09,  1.62it/s]
 53%|█████▎    | 34549/65536 [5:56:01<5:16:43,  1.63it/s]
 53%|█████▎    | 34550/65536 [5:56:02<5:16:21,  1.63it/s]
 53%|█████▎    | 34551/65536 [5:56:02<5:10:02,  1.67it/s]
 53%|█████▎    | 34552/65536 [5:56:03<5:17:06,  1.63it/s]
 53%|█████▎    | 34553/65536 [5:56:04<5:12:16,  1.65it/s]
 53%|█████▎    | 34554/65536 [5:56:04<5:08:46,  1.67it/s]
 53%|█████▎    | 34555/65536 [5:56:05<5:16:52,  1.63it/s]
 53%|█████▎    | 34556/65536 [5:56:05<5:18:46,  1.62it/s]
 53%|█████▎    | 34557/65536 [5:56:06<5:16:20,  1.63it/s]
 53%|█████▎    | 34558/65536 [5:56:07<5:18:11,  1.62it/s]
 53%|█████▎    | 34559/65536 [5:56:07<5:13:58,  1.64it/s]
 53%|█████▎    | 34560/65536 [5:56:08<5:08:17,  1.67it/s]
                                                         
{'loss': 1.651, 'learning_rate': 5.266927880494673e-07, 'epoch': 2133.33}
+
 53%|█████▎    | 34560/65536 [5:56:08<5:08:17,  1.67it/s]
 53%|█████▎    | 34561/65536 [5:56:08<5:07:35,  1.68it/s]
 53%|█████▎    | 34562/65536 [5:56:09<5:13:54,  1.64it/s]
 53%|█████▎    | 34563/65536 [5:56:10<5:11:48,  1.66it/s]
 53%|█████▎    | 34564/65536 [5:56:10<5:09:06,  1.67it/s]
 53%|█████▎    | 34565/65536 [5:56:11<5:12:42,  1.65it/s]
 53%|█████▎    | 34566/65536 [5:56:11<5:16:56,  1.63it/s]
 53%|█████▎    | 34567/65536 [5:56:12<5:16:57,  1.63it/s]
 53%|█████▎    | 34568/65536 [5:56:13<5:18:51,  1.62it/s]
 53%|█████▎    | 34569/65536 [5:56:13<5:21:38,  1.60it/s]
 53%|█████▎    | 34570/65536 [5:56:14<5:26:31,  1.58it/s]
 53%|█████▎    | 34571/65536 [5:56:15<5:29:22,  1.57it/s]
 53%|█████▎    | 34572/65536 [5:56:15<5:24:23,  1.59it/s]
 53%|█████▎    | 34573/65536 [5:56:16<5:26:28,  1.58it/s]
 53%|█████▎    | 34574/65536 [5:56:17<5:29:57,  1.56it/s]
 53%|█████▎    | 34575/65536 [5:56:17<5:23:53,  1.59it/s]
 53%|█████▎    | 34576/65536 [5:56:18<5:20:43,  1.61it/s]
 53%|█████▎    | 34577/65536 [5:56:18<5:26:12,  1.58it/s]
 53%|█████▎    | 34578/65536 [5:56:19<5:20:29,  1.61it/s]
 53%|█████▎    | 34579/65536 [5:56:20<5:21:15,  1.61it/s]
 53%|█████▎    | 34580/65536 [5:56:20<5:21:44,  1.60it/s]
                                                         
{'loss': 1.5994, 'learning_rate': 5.264172890902412e-07, 'epoch': 2134.57}
+
 53%|█████▎    | 34580/65536 [5:56:20<5:21:44,  1.60it/s]
 53%|█████▎    | 34581/65536 [5:56:21<5:13:47,  1.64it/s]
 53%|█████▎    | 34582/65536 [5:56:21<5:14:41,  1.64it/s]
 53%|█████▎    | 34583/65536 [5:56:22<5:18:33,  1.62it/s]
 53%|█████▎    | 34584/65536 [5:56:23<5:22:57,  1.60it/s]
 53%|█████▎    | 34585/65536 [5:56:23<5:19:38,  1.61it/s]
 53%|█████▎    | 34586/65536 [5:56:24<5:13:05,  1.65it/s]
 53%|█████▎    | 34587/65536 [5:56:24<5:07:31,  1.68it/s]
 53%|█████▎    | 34588/65536 [5:56:25<5:25:05,  1.59it/s]
 53%|█████▎    | 34589/65536 [5:56:26<5:22:16,  1.60it/s]
 53%|█████▎    | 34590/65536 [5:56:27<5:35:18,  1.54it/s]
 53%|█████▎    | 34591/65536 [5:56:27<5:26:17,  1.58it/s]
 53%|█████▎    | 34592/65536 [5:56:28<5:21:15,  1.61it/s]
 53%|█████▎    | 34593/65536 [5:56:28<5:16:44,  1.63it/s]
 53%|█████▎    | 34594/65536 [5:56:29<5:15:25,  1.63it/s]
 53%|█████▎    | 34595/65536 [5:56:30<5:13:18,  1.65it/s]
 53%|█████▎    | 34596/65536 [5:56:30<5:23:43,  1.59it/s]
 53%|█████▎    | 34597/65536 [5:56:31<5:21:37,  1.60it/s]
 53%|█████▎    | 34598/65536 [5:56:31<5:12:18,  1.65it/s]
 53%|█████▎    | 34599/65536 [5:56:32<5:13:51,  1.64it/s]
 53%|█████▎    | 34600/65536 [5:56:33<5:09:51,  1.66it/s]
                                                         
{'loss': 1.6398, 'learning_rate': 5.26141790131015e-07, 'epoch': 2135.8}
+
 53%|█████▎    | 34600/65536 [5:56:33<5:09:51,  1.66it/s]
 53%|█████▎    | 34601/65536 [5:56:33<5:04:44,  1.69it/s]
 53%|█████▎    | 34602/65536 [5:56:34<5:04:17,  1.69it/s]
 53%|█████▎    | 34603/65536 [5:56:34<5:08:39,  1.67it/s]
 53%|█████▎    | 34604/65536 [5:56:35<5:18:20,  1.62it/s]
 53%|█████▎    | 34605/65536 [5:56:36<5:16:49,  1.63it/s]
 53%|█████▎    | 34606/65536 [5:56:36<5:16:26,  1.63it/s]
 53%|█████▎    | 34607/65536 [5:56:37<5:16:21,  1.63it/s]
 53%|█████▎    | 34608/65536 [5:56:37<5:15:02,  1.64it/s]
 53%|█████▎    | 34609/65536 [5:56:38<5:10:24,  1.66it/s]
 53%|█████▎    | 34610/65536 [5:56:39<5:13:19,  1.65it/s]
 53%|█████▎    | 34611/65536 [5:56:39<5:11:33,  1.65it/s]
 53%|█████▎    | 34612/65536 [5:56:40<5:10:19,  1.66it/s]
 53%|█████▎    | 34613/65536 [5:56:40<5:16:22,  1.63it/s]
 53%|█████▎    | 34614/65536 [5:56:41<5:18:18,  1.62it/s]
 53%|█████▎    | 34615/65536 [5:56:42<5:12:33,  1.65it/s]
 53%|█████▎    | 34616/65536 [5:56:42<5:10:22,  1.66it/s]
 53%|█████▎    | 34617/65536 [5:56:43<5:17:00,  1.63it/s]
 53%|█████▎    | 34618/65536 [5:56:44<5:11:34,  1.65it/s]
 53%|█████▎    | 34619/65536 [5:56:44<5:13:39,  1.64it/s]
 53%|█████▎    | 34620/65536 [5:56:45<5:43:17,  1.50it/s]
                                                         
{'loss': 1.638, 'learning_rate': 5.258662911717889e-07, 'epoch': 2137.04}
+
 53%|█████▎    | 34620/65536 [5:56:45<5:43:17,  1.50it/s]
 53%|█████▎    | 34621/65536 [5:56:46<5:31:56,  1.55it/s]
 53%|█████▎    | 34622/65536 [5:56:46<5:26:28,  1.58it/s]
 53%|█████▎    | 34623/65536 [5:56:47<5:26:05,  1.58it/s]
 53%|█████▎    | 34624/65536 [5:56:47<5:26:27,  1.58it/s]
 53%|█████▎    | 34625/65536 [5:56:48<5:25:23,  1.58it/s]
 53%|█████▎    | 34626/65536 [5:56:49<5:21:29,  1.60it/s]
 53%|█████▎    | 34627/65536 [5:56:49<5:17:37,  1.62it/s]
 53%|█████▎    | 34628/65536 [5:56:50<5:18:29,  1.62it/s]
 53%|█████▎    | 34629/65536 [5:56:50<5:14:45,  1.64it/s]
 53%|█████▎    | 34630/65536 [5:56:51<5:19:13,  1.61it/s]
 53%|█████▎    | 34631/65536 [5:56:52<5:11:57,  1.65it/s]
 53%|█████▎    | 34632/65536 [5:56:52<5:15:22,  1.63it/s]
 53%|█████▎    | 34633/65536 [5:56:53<5:20:56,  1.60it/s]
 53%|█████▎    | 34634/65536 [5:56:54<5:16:40,  1.63it/s]
 53%|█████▎    | 34635/65536 [5:56:54<5:14:13,  1.64it/s]
 53%|█████▎    | 34636/65536 [5:56:55<5:24:15,  1.59it/s]
 53%|█████▎    | 34637/65536 [5:56:55<5:31:17,  1.55it/s]
 53%|█████▎    | 34638/65536 [5:56:56<5:26:51,  1.58it/s]
 53%|█████▎    | 34639/65536 [5:56:57<5:23:08,  1.59it/s]
 53%|█████▎    | 34640/65536 [5:56:57<5:19:48,  1.61it/s]
                                                         
{'loss': 1.6332, 'learning_rate': 5.255907922125627e-07, 'epoch': 2138.27}
+
 53%|█████▎    | 34640/65536 [5:56:57<5:19:48,  1.61it/s]
 53%|█████▎    | 34641/65536 [5:56:58<5:21:38,  1.60it/s]
 53%|█████▎    | 34642/65536 [5:56:59<5:30:59,  1.56it/s]
 53%|█████▎    | 34643/65536 [5:56:59<5:35:40,  1.53it/s]
 53%|█████▎    | 34644/65536 [5:57:00<5:29:36,  1.56it/s]
 53%|█████▎    | 34645/65536 [5:57:01<5:27:34,  1.57it/s]
 53%|█████▎    | 34646/65536 [5:57:01<5:21:15,  1.60it/s]
 53%|█████▎    | 34647/65536 [5:57:02<5:14:31,  1.64it/s]
 53%|█████▎    | 34648/65536 [5:57:02<5:06:42,  1.68it/s]
 53%|█████▎    | 34649/65536 [5:57:03<5:04:43,  1.69it/s]
 53%|█████▎    | 34650/65536 [5:57:03<4:58:46,  1.72it/s]
 53%|█████▎    | 34651/65536 [5:57:04<5:00:07,  1.72it/s]
 53%|█████▎    | 34652/65536 [5:57:05<5:13:36,  1.64it/s]
 53%|█████▎    | 34653/65536 [5:57:05<5:06:39,  1.68it/s]
 53%|█████▎    | 34654/65536 [5:57:06<5:08:17,  1.67it/s]
 53%|█████▎    | 34655/65536 [5:57:06<5:10:37,  1.66it/s]
 53%|█████▎    | 34656/65536 [5:57:07<5:07:08,  1.68it/s]
 53%|█████▎    | 34657/65536 [5:57:08<5:06:22,  1.68it/s]
 53%|█████▎    | 34658/65536 [5:57:08<5:02:23,  1.70it/s]
 53%|█████▎    | 34659/65536 [5:57:09<5:13:42,  1.64it/s]
 53%|█████▎    | 34660/65536 [5:57:09<5:12:36,  1.65it/s]
                                                         
{'loss': 1.6314, 'learning_rate': 5.253152932533364e-07, 'epoch': 2139.51}
+
 53%|█████▎    | 34660/65536 [5:57:09<5:12:36,  1.65it/s]
 53%|█████▎    | 34661/65536 [5:57:10<5:14:19,  1.64it/s]
 53%|█████▎    | 34662/65536 [5:57:11<5:14:56,  1.63it/s]
 53%|█████▎    | 34663/65536 [5:57:11<5:18:57,  1.61it/s]
 53%|█████▎    | 34664/65536 [5:57:12<5:21:57,  1.60it/s]
 53%|█████▎    | 34665/65536 [5:57:13<5:17:48,  1.62it/s]
 53%|█████▎    | 34666/65536 [5:57:13<5:14:02,  1.64it/s]
 53%|█████▎    | 34667/65536 [5:57:14<5:27:11,  1.57it/s]
 53%|█████▎    | 34668/65536 [5:57:14<5:16:48,  1.62it/s]
 53%|█████▎    | 34669/65536 [5:57:15<5:25:48,  1.58it/s]
 53%|█████▎    | 34670/65536 [5:57:16<5:29:23,  1.56it/s]
 53%|█████▎    | 34671/65536 [5:57:16<5:26:26,  1.58it/s]
 53%|█████▎    | 34672/65536 [5:57:17<5:22:46,  1.59it/s]
 53%|█████▎    | 34673/65536 [5:57:18<5:29:26,  1.56it/s]
 53%|█████▎    | 34674/65536 [5:57:18<5:19:58,  1.61it/s]
 53%|█████▎    | 34675/65536 [5:57:19<5:16:55,  1.62it/s]
 53%|█████▎    | 34676/65536 [5:57:19<5:17:08,  1.62it/s]
 53%|█████▎    | 34677/65536 [5:57:20<5:08:03,  1.67it/s]
 53%|█████▎    | 34678/65536 [5:57:21<5:03:18,  1.70it/s]
 53%|█████▎    | 34679/65536 [5:57:21<5:10:13,  1.66it/s]
 53%|█████▎    | 34680/65536 [5:57:22<5:06:04,  1.68it/s]
                                                         
{'loss': 1.6453, 'learning_rate': 5.250397942941104e-07, 'epoch': 2140.74}
+
 53%|█████▎    | 34680/65536 [5:57:22<5:06:04,  1.68it/s]
 53%|█████▎    | 34681/65536 [5:57:22<5:08:17,  1.67it/s]
 53%|█████▎    | 34682/65536 [5:57:23<5:16:24,  1.63it/s]
 53%|█████▎    | 34683/65536 [5:57:24<5:18:58,  1.61it/s]
 53%|█████▎    | 34684/65536 [5:57:24<5:11:26,  1.65it/s]
 53%|█████▎    | 34685/65536 [5:57:25<5:27:03,  1.57it/s]
 53%|█████▎    | 34686/65536 [5:57:26<5:27:55,  1.57it/s]
 53%|█████▎    | 34687/65536 [5:57:26<5:26:37,  1.57it/s]
 53%|█████▎    | 34688/65536 [5:57:27<5:20:42,  1.60it/s]
 53%|█████▎    | 34689/65536 [5:57:27<5:12:44,  1.64it/s]
 53%|█████▎    | 34690/65536 [5:57:28<5:11:31,  1.65it/s]
 53%|█████▎    | 34691/65536 [5:57:29<5:15:30,  1.63it/s]
 53%|█████▎    | 34692/65536 [5:57:29<5:21:18,  1.60it/s]
 53%|█████▎    | 34693/65536 [5:57:30<5:26:54,  1.57it/s]
 53%|█████▎    | 34694/65536 [5:57:31<5:19:18,  1.61it/s]
 53%|█████▎    | 34695/65536 [5:57:31<5:17:37,  1.62it/s]
 53%|█████▎    | 34696/65536 [5:57:32<5:12:52,  1.64it/s]
 53%|█████▎    | 34697/65536 [5:57:32<5:20:16,  1.60it/s]
 53%|█████▎    | 34698/65536 [5:57:33<5:13:42,  1.64it/s]
 53%|█████▎    | 34699/65536 [5:57:34<5:09:35,  1.66it/s]
 53%|█████▎    | 34700/65536 [5:57:34<5:13:30,  1.64it/s]
                                                         
{'loss': 1.6315, 'learning_rate': 5.247642953348842e-07, 'epoch': 2141.98}
+
 53%|█████▎    | 34700/65536 [5:57:34<5:13:30,  1.64it/s]
 53%|█████▎    | 34701/65536 [5:57:35<5:25:38,  1.58it/s]
 53%|█████▎    | 34702/65536 [5:57:35<5:15:47,  1.63it/s]
 53%|█████▎    | 34703/65536 [5:57:36<5:15:49,  1.63it/s]
 53%|█████▎    | 34704/65536 [5:57:37<5:07:07,  1.67it/s]
 53%|█████▎    | 34705/65536 [5:57:37<5:10:15,  1.66it/s]
 53%|█████▎    | 34706/65536 [5:57:38<5:11:36,  1.65it/s]
 53%|█████▎    | 34707/65536 [5:57:38<5:11:41,  1.65it/s]
 53%|█████▎    | 34708/65536 [5:57:39<5:20:05,  1.61it/s]
 53%|█████▎    | 34709/65536 [5:57:40<5:19:43,  1.61it/s]
 53%|█████▎    | 34710/65536 [5:57:40<5:14:23,  1.63it/s]
 53%|█████▎    | 34711/65536 [5:57:41<5:12:50,  1.64it/s]
 53%|█████▎    | 34712/65536 [5:57:42<5:19:31,  1.61it/s]
 53%|█████▎    | 34713/65536 [5:57:42<5:21:51,  1.60it/s]
 53%|█████▎    | 34714/65536 [5:57:43<5:15:28,  1.63it/s]
 53%|█████▎    | 34715/65536 [5:57:43<5:10:56,  1.65it/s]
 53%|█████▎    | 34716/65536 [5:57:44<5:19:35,  1.61it/s]
 53%|█████▎    | 34717/65536 [5:57:45<5:26:41,  1.57it/s]
 53%|█████▎    | 34718/65536 [5:57:45<5:20:47,  1.60it/s]
 53%|█████▎    | 34719/65536 [5:57:46<5:26:18,  1.57it/s]
 53%|█████▎    | 34720/65536 [5:57:47<5:29:10,  1.56it/s]
                                                         
{'loss': 1.6325, 'learning_rate': 5.244887963756581e-07, 'epoch': 2143.21}
+
 53%|█████▎    | 34720/65536 [5:57:47<5:29:10,  1.56it/s]
 53%|█████▎    | 34721/65536 [5:57:47<5:26:20,  1.57it/s]
 53%|█████▎    | 34722/65536 [5:57:48<5:26:08,  1.57it/s]
 53%|█████▎    | 34723/65536 [5:57:49<5:23:54,  1.59it/s]
 53%|█████▎    | 34724/65536 [5:57:49<5:16:46,  1.62it/s]
 53%|█████▎    | 34725/65536 [5:57:50<5:15:13,  1.63it/s]
 53%|█████▎    | 34726/65536 [5:57:50<5:12:16,  1.64it/s]
 53%|█████▎    | 34727/65536 [5:57:51<5:06:41,  1.67it/s]
 53%|█████▎    | 34728/65536 [5:57:51<5:07:34,  1.67it/s]
 53%|█████▎    | 34729/65536 [5:57:52<5:04:02,  1.69it/s]
 53%|█████▎    | 34730/65536 [5:57:53<5:00:00,  1.71it/s]
 53%|█████▎    | 34731/65536 [5:57:53<5:03:33,  1.69it/s]
 53%|█████▎    | 34732/65536 [5:57:54<5:09:59,  1.66it/s]
 53%|█████▎    | 34733/65536 [5:57:55<5:25:06,  1.58it/s]
 53%|█████▎    | 34734/65536 [5:57:55<5:29:58,  1.56it/s]
 53%|█████▎    | 34735/65536 [5:57:56<5:24:16,  1.58it/s]
 53%|█████▎    | 34736/65536 [5:57:56<5:24:43,  1.58it/s]
 53%|█████▎    | 34737/65536 [5:57:57<5:15:30,  1.63it/s]
 53%|█████▎    | 34738/65536 [5:57:58<5:11:26,  1.65it/s]
 53%|█████▎    | 34739/65536 [5:57:58<5:07:32,  1.67it/s]
 53%|█████▎    | 34740/65536 [5:57:59<5:09:57,  1.66it/s]
                                                         
{'loss': 1.6387, 'learning_rate': 5.242132974164319e-07, 'epoch': 2144.44}
+
 53%|█████▎    | 34740/65536 [5:57:59<5:09:57,  1.66it/s]
 53%|█████▎    | 34741/65536 [5:57:59<5:07:12,  1.67it/s]
 53%|█████▎    | 34742/65536 [5:58:00<5:03:58,  1.69it/s]
 53%|█████▎    | 34743/65536 [5:58:01<5:09:00,  1.66it/s]
 53%|█████▎    | 34744/65536 [5:58:01<5:13:54,  1.63it/s]
 53%|█████▎    | 34745/65536 [5:58:02<5:22:40,  1.59it/s]
 53%|█████▎    | 34746/65536 [5:58:03<5:19:23,  1.61it/s]
 53%|█████▎    | 34747/65536 [5:58:03<5:15:28,  1.63it/s]
 53%|█████▎    | 34748/65536 [5:58:04<5:11:44,  1.65it/s]
 53%|█████▎    | 34749/65536 [5:58:04<5:05:40,  1.68it/s]
 53%|█████▎    | 34750/65536 [5:58:05<5:18:58,  1.61it/s]
 53%|█████▎    | 34751/65536 [5:58:06<5:13:24,  1.64it/s]
 53%|█████▎    | 34752/65536 [5:58:06<5:17:03,  1.62it/s]
 53%|█████▎    | 34753/65536 [5:58:07<5:14:46,  1.63it/s]
 53%|█████▎    | 34754/65536 [5:58:07<5:20:20,  1.60it/s]
 53%|█████▎    | 34755/65536 [5:58:08<5:24:33,  1.58it/s]
 53%|█████▎    | 34756/65536 [5:58:09<5:12:41,  1.64it/s]
 53%|█████▎    | 34757/65536 [5:58:09<5:03:46,  1.69it/s]
 53%|█████▎    | 34758/65536 [5:58:10<5:04:19,  1.69it/s]
 53%|█████▎    | 34759/65536 [5:58:10<5:05:12,  1.68it/s]
 53%|█████▎    | 34760/65536 [5:58:11<5:11:00,  1.65it/s]
                                                         
{'loss': 1.6771, 'learning_rate': 5.239377984572058e-07, 'epoch': 2145.68}
+
 53%|█████▎    | 34760/65536 [5:58:11<5:11:00,  1.65it/s]
 53%|█████▎    | 34761/65536 [5:58:12<5:13:44,  1.63it/s]
 53%|█████▎    | 34762/65536 [5:58:12<5:10:42,  1.65it/s]
 53%|█████▎    | 34763/65536 [5:58:13<5:07:17,  1.67it/s]
 53%|█████▎    | 34764/65536 [5:58:13<5:12:03,  1.64it/s]
 53%|█████▎    | 34765/65536 [5:58:14<5:09:39,  1.66it/s]
 53%|█████▎    | 34766/65536 [5:58:15<5:20:27,  1.60it/s]
 53%|█████▎    | 34767/65536 [5:58:15<5:20:49,  1.60it/s]
 53%|█████▎    | 34768/65536 [5:58:16<5:22:56,  1.59it/s]
 53%|█████▎    | 34769/65536 [5:58:17<5:15:08,  1.63it/s]
 53%|█████▎    | 34770/65536 [5:58:17<5:07:16,  1.67it/s]
 53%|█████▎    | 34771/65536 [5:58:18<5:03:09,  1.69it/s]
 53%|█████▎    | 34772/65536 [5:58:18<5:15:18,  1.63it/s]
 53%|█████▎    | 34773/65536 [5:58:19<5:11:30,  1.65it/s]
 53%|█████▎    | 34774/65536 [5:58:20<5:14:45,  1.63it/s]
 53%|█████▎    | 34775/65536 [5:58:20<5:08:51,  1.66it/s]
 53%|█████▎    | 34776/65536 [5:58:21<5:14:44,  1.63it/s]
 53%|█████▎    | 34777/65536 [5:58:21<5:10:41,  1.65it/s]
 53%|█████▎    | 34778/65536 [5:58:22<5:21:07,  1.60it/s]
 53%|█████▎    | 34779/65536 [5:58:23<5:19:13,  1.61it/s]
 53%|█████▎    | 34780/65536 [5:58:23<5:10:58,  1.65it/s]
                                                         
{'loss': 1.6287, 'learning_rate': 5.236622994979797e-07, 'epoch': 2146.91}
+
 53%|█████▎    | 34780/65536 [5:58:23<5:10:58,  1.65it/s]
 53%|█████▎    | 34781/65536 [5:58:24<5:10:29,  1.65it/s]
 53%|█████▎    | 34782/65536 [5:58:25<5:22:59,  1.59it/s]
 53%|█████▎    | 34783/65536 [5:58:25<5:15:40,  1.62it/s]
 53%|█████▎    | 34784/65536 [5:58:26<5:17:39,  1.61it/s]
 53%|█████▎    | 34785/65536 [5:58:26<5:21:54,  1.59it/s]
 53%|█████▎    | 34786/65536 [5:58:27<5:19:09,  1.61it/s]
 53%|█████▎    | 34787/65536 [5:58:28<5:15:11,  1.63it/s]
 53%|█████▎    | 34788/65536 [5:58:28<5:18:11,  1.61it/s]
 53%|█████▎    | 34789/65536 [5:58:29<5:13:21,  1.64it/s]
 53%|█████▎    | 34790/65536 [5:58:29<5:12:41,  1.64it/s]
 53%|█████▎    | 34791/65536 [5:58:30<5:06:30,  1.67it/s]
 53%|█████▎    | 34792/65536 [5:58:31<5:03:17,  1.69it/s]
 53%|█████▎    | 34793/65536 [5:58:31<5:14:06,  1.63it/s]
 53%|█████▎    | 34794/65536 [5:58:32<5:17:39,  1.61it/s]
 53%|█████▎    | 34795/65536 [5:58:32<5:09:27,  1.66it/s]
 53%|█████▎    | 34796/65536 [5:58:33<5:10:03,  1.65it/s]
 53%|█████▎    | 34797/65536 [5:58:34<5:15:30,  1.62it/s]
 53%|█████▎    | 34798/65536 [5:58:34<5:31:45,  1.54it/s]
 53%|█████▎    | 34799/65536 [5:58:35<5:27:04,  1.57it/s]
 53%|█████▎    | 34800/65536 [5:58:36<5:30:16,  1.55it/s]
                                                         
{'loss': 1.6298, 'learning_rate': 5.233868005387535e-07, 'epoch': 2148.15}
+
 53%|█████▎    | 34800/65536 [5:58:36<5:30:16,  1.55it/s]
 53%|█████▎    | 34801/65536 [5:58:36<5:26:34,  1.57it/s]
 53%|█████▎    | 34802/65536 [5:58:37<5:18:35,  1.61it/s]
 53%|█████▎    | 34803/65536 [5:58:38<5:15:26,  1.62it/s]
 53%|█████▎    | 34804/65536 [5:58:38<5:21:53,  1.59it/s]
 53%|█████▎    | 34805/65536 [5:58:39<5:22:58,  1.59it/s]
 53%|█████▎    | 34806/65536 [5:58:39<5:24:51,  1.58it/s]
 53%|█████▎    | 34807/65536 [5:58:40<5:22:00,  1.59it/s]
 53%|█████▎    | 34808/65536 [5:58:41<5:14:35,  1.63it/s]
 53%|█████▎    | 34809/65536 [5:58:41<5:16:20,  1.62it/s]
 53%|█████▎    | 34810/65536 [5:58:42<5:13:54,  1.63it/s]
 53%|█████▎    | 34811/65536 [5:58:42<5:08:00,  1.66it/s]
 53%|█████▎    | 34812/65536 [5:58:43<5:10:52,  1.65it/s]
 53%|█████▎    | 34813/65536 [5:58:44<5:07:56,  1.66it/s]
 53%|█████▎    | 34814/65536 [5:58:44<5:19:35,  1.60it/s]
 53%|█████▎    | 34815/65536 [5:58:45<5:17:50,  1.61it/s]
 53%|█████▎    | 34816/65536 [5:58:46<5:11:52,  1.64it/s]
 53%|█████▎    | 34817/65536 [5:58:46<5:08:42,  1.66it/s]
 53%|█████▎    | 34818/65536 [5:58:47<5:05:42,  1.67it/s]
 53%|█████▎    | 34819/65536 [5:58:47<5:17:40,  1.61it/s]
 53%|█████▎    | 34820/65536 [5:58:48<5:16:28,  1.62it/s]
                                                         
{'loss': 1.6201, 'learning_rate': 5.231113015795274e-07, 'epoch': 2149.38}
+
 53%|█████▎    | 34820/65536 [5:58:48<5:16:28,  1.62it/s]
 53%|█████▎    | 34821/65536 [5:58:49<5:14:57,  1.63it/s]
 53%|█████▎    | 34822/65536 [5:58:49<5:18:44,  1.61it/s]
 53%|█████▎    | 34823/65536 [5:58:50<5:12:14,  1.64it/s]
 53%|█████▎    | 34824/65536 [5:58:50<5:12:00,  1.64it/s]
 53%|█████▎    | 34825/65536 [5:58:51<5:12:05,  1.64it/s]
 53%|█████▎    | 34826/65536 [5:58:52<5:06:55,  1.67it/s]
 53%|█████▎    | 34827/65536 [5:58:52<5:08:14,  1.66it/s]
 53%|█████▎    | 34828/65536 [5:58:53<5:11:48,  1.64it/s]
 53%|█████▎    | 34829/65536 [5:58:54<5:25:10,  1.57it/s]
 53%|█████▎    | 34830/65536 [5:58:54<5:16:47,  1.62it/s]
 53%|█████▎    | 34831/65536 [5:58:55<5:21:19,  1.59it/s]
 53%|█████▎    | 34832/65536 [5:58:55<5:16:39,  1.62it/s]
 53%|█████▎    | 34833/65536 [5:58:56<5:14:37,  1.63it/s]
 53%|█████▎    | 34834/65536 [5:58:57<5:08:32,  1.66it/s]
 53%|█████▎    | 34835/65536 [5:58:57<5:04:35,  1.68it/s]
 53%|█████▎    | 34836/65536 [5:58:58<5:21:57,  1.59it/s]
 53%|█████▎    | 34837/65536 [5:58:58<5:14:25,  1.63it/s]
 53%|█████▎    | 34838/65536 [5:58:59<5:17:38,  1.61it/s]
 53%|█████▎    | 34839/65536 [5:59:00<5:15:03,  1.62it/s]
 53%|█████▎    | 34840/65536 [5:59:00<5:07:19,  1.66it/s]
                                                         
{'loss': 1.64, 'learning_rate': 5.228358026203012e-07, 'epoch': 2150.62}
+
 53%|█████▎    | 34840/65536 [5:59:00<5:07:19,  1.66it/s]
 53%|█████▎    | 34841/65536 [5:59:01<5:09:07,  1.65it/s]
 53%|█████▎    | 34842/65536 [5:59:01<5:13:51,  1.63it/s]
 53%|█████▎    | 34843/65536 [5:59:02<5:09:46,  1.65it/s]
 53%|█████▎    | 34844/65536 [5:59:03<5:11:49,  1.64it/s]
 53%|█████▎    | 34845/65536 [5:59:03<5:12:26,  1.64it/s]
 53%|█████▎    | 34846/65536 [5:59:04<5:18:00,  1.61it/s]
 53%|█████▎    | 34847/65536 [5:59:05<5:35:39,  1.52it/s]
 53%|█████▎    | 34848/65536 [5:59:05<5:26:46,  1.57it/s]
 53%|█████▎    | 34849/65536 [5:59:06<5:25:16,  1.57it/s]
 53%|█████▎    | 34850/65536 [5:59:07<5:25:18,  1.57it/s]
 53%|█████▎    | 34851/65536 [5:59:07<5:14:03,  1.63it/s]
 53%|█████▎    | 34852/65536 [5:59:08<5:11:23,  1.64it/s]
 53%|█████▎    | 34853/65536 [5:59:08<5:16:06,  1.62it/s]
 53%|█████▎    | 34854/65536 [5:59:09<5:15:21,  1.62it/s]
 53%|█████▎    | 34855/65536 [5:59:10<5:10:55,  1.64it/s]
 53%|█████▎    | 34856/65536 [5:59:10<5:06:54,  1.67it/s]
 53%|█████▎    | 34857/65536 [5:59:11<5:14:04,  1.63it/s]
 53%|█████▎    | 34858/65536 [5:59:11<5:08:03,  1.66it/s]
 53%|█████▎    | 34859/65536 [5:59:12<5:09:16,  1.65it/s]
 53%|█████▎    | 34860/65536 [5:59:13<5:10:26,  1.65it/s]
                                                         
{'loss': 1.6608, 'learning_rate': 5.225603036610751e-07, 'epoch': 2151.85}
+
 53%|█████▎    | 34860/65536 [5:59:13<5:10:26,  1.65it/s]
 53%|█████▎    | 34861/65536 [5:59:13<5:11:39,  1.64it/s]
 53%|█████▎    | 34862/65536 [5:59:14<5:10:08,  1.65it/s]
 53%|█████▎    | 34863/65536 [5:59:14<5:17:32,  1.61it/s]
 53%|█████▎    | 34864/65536 [5:59:15<5:07:56,  1.66it/s]
 53%|█████▎    | 34865/65536 [5:59:16<5:04:56,  1.68it/s]
 53%|█████▎    | 34866/65536 [5:59:16<5:03:44,  1.68it/s]
 53%|█████▎    | 34867/65536 [5:59:17<5:02:37,  1.69it/s]
 53%|█████▎    | 34868/65536 [5:59:17<5:09:29,  1.65it/s]
 53%|█████▎    | 34869/65536 [5:59:18<5:09:46,  1.65it/s]
 53%|█████▎    | 34870/65536 [5:59:19<5:08:22,  1.66it/s]
 53%|█████▎    | 34871/65536 [5:59:19<5:05:18,  1.67it/s]
 53%|█████▎    | 34872/65536 [5:59:20<5:07:06,  1.66it/s]
 53%|█████▎    | 34873/65536 [5:59:20<5:10:15,  1.65it/s]
 53%|█████▎    | 34874/65536 [5:59:21<5:14:28,  1.63it/s]
 53%|█████▎    | 34875/65536 [5:59:22<5:20:58,  1.59it/s]
 53%|█████▎    | 34876/65536 [5:59:22<5:31:06,  1.54it/s]
 53%|█████▎    | 34877/65536 [5:59:23<5:30:58,  1.54it/s]
 53%|█████▎    | 34878/65536 [5:59:24<5:27:24,  1.56it/s]
 53%|█████▎    | 34879/65536 [5:59:24<5:30:42,  1.54it/s]
 53%|█████▎    | 34880/65536 [5:59:25<5:28:40,  1.55it/s]
                                                         
{'loss': 1.6253, 'learning_rate': 5.22284804701849e-07, 'epoch': 2153.09}
+
 53%|█████▎    | 34880/65536 [5:59:25<5:28:40,  1.55it/s]
 53%|█████▎    | 34881/65536 [5:59:26<5:26:10,  1.57it/s]
 53%|█████▎    | 34882/65536 [5:59:26<5:17:00,  1.61it/s]
 53%|█████▎    | 34883/65536 [5:59:27<5:16:17,  1.62it/s]
 53%|█████▎    | 34884/65536 [5:59:27<5:08:43,  1.65it/s]
 53%|█████▎    | 34885/65536 [5:59:28<5:18:41,  1.60it/s]
 53%|█████▎    | 34886/65536 [5:59:29<5:16:56,  1.61it/s]
 53%|█████▎    | 34887/65536 [5:59:29<5:08:10,  1.66it/s]
 53%|█████▎    | 34888/65536 [5:59:30<5:05:59,  1.67it/s]
 53%|█████▎    | 34889/65536 [5:59:30<5:09:22,  1.65it/s]
 53%|█████▎    | 34890/65536 [5:59:31<5:12:40,  1.63it/s]
 53%|█████▎    | 34891/65536 [5:59:32<5:07:57,  1.66it/s]
 53%|█████▎    | 34892/65536 [5:59:32<5:17:06,  1.61it/s]
 53%|█████▎    | 34893/65536 [5:59:33<5:10:47,  1.64it/s]
 53%|█████▎    | 34894/65536 [5:59:33<5:11:08,  1.64it/s]
 53%|█████▎    | 34895/65536 [5:59:34<5:25:11,  1.57it/s]
 53%|█████▎    | 34896/65536 [5:59:35<5:28:59,  1.55it/s]
 53%|█████▎    | 34897/65536 [5:59:35<5:21:41,  1.59it/s]
 53%|█████▎    | 34898/65536 [5:59:36<5:18:02,  1.61it/s]
 53%|█████▎    | 34899/65536 [5:59:37<5:17:16,  1.61it/s]
 53%|█████▎    | 34900/65536 [5:59:37<5:20:22,  1.59it/s]
                                                         
{'loss': 1.6644, 'learning_rate': 5.220093057426228e-07, 'epoch': 2154.32}
+
 53%|█████▎    | 34900/65536 [5:59:37<5:20:22,  1.59it/s]
 53%|█████▎    | 34901/65536 [5:59:38<5:22:43,  1.58it/s]
 53%|█████▎    | 34902/65536 [5:59:39<5:14:48,  1.62it/s]
 53%|█████▎    | 34903/65536 [5:59:39<5:15:59,  1.62it/s]
 53%|█████▎    | 34904/65536 [5:59:40<5:15:54,  1.62it/s]
 53%|█████▎    | 34905/65536 [5:59:40<5:13:53,  1.63it/s]
 53%|█████▎    | 34906/65536 [5:59:41<5:07:57,  1.66it/s]
 53%|█████▎    | 34907/65536 [5:59:42<5:15:28,  1.62it/s]
 53%|█████▎    | 34908/65536 [5:59:42<5:12:09,  1.64it/s]
 53%|█████▎    | 34909/65536 [5:59:43<5:09:51,  1.65it/s]
 53%|█████▎    | 34910/65536 [5:59:43<5:11:18,  1.64it/s]
 53%|█████▎    | 34911/65536 [5:59:44<5:13:57,  1.63it/s]
 53%|█████▎    | 34912/65536 [5:59:45<5:28:11,  1.56it/s]
 53%|█████▎    | 34913/65536 [5:59:45<5:22:00,  1.58it/s]
 53%|█████▎    | 34914/65536 [5:59:46<5:18:04,  1.60it/s]
 53%|█████▎    | 34915/65536 [5:59:47<5:16:09,  1.61it/s]
 53%|█████▎    | 34916/65536 [5:59:47<5:18:30,  1.60it/s]
 53%|█████▎    | 34917/65536 [5:59:48<5:16:57,  1.61it/s]
 53%|█████▎    | 34918/65536 [5:59:48<5:20:16,  1.59it/s]
 53%|█████▎    | 34919/65536 [5:59:49<5:17:04,  1.61it/s]
 53%|█████▎    | 34920/65536 [5:59:50<5:09:42,  1.65it/s]
                                                         
{'loss': 1.6186, 'learning_rate': 5.217338067833965e-07, 'epoch': 2155.56}
+
 53%|█████▎    | 34920/65536 [5:59:50<5:09:42,  1.65it/s]
 53%|█████▎    | 34921/65536 [5:59:50<5:04:44,  1.67it/s]
 53%|█████▎    | 34922/65536 [5:59:51<4:57:36,  1.71it/s]
 53%|█████▎    | 34923/65536 [5:59:51<5:02:03,  1.69it/s]
 53%|█████▎    | 34924/65536 [5:59:52<5:10:28,  1.64it/s]
 53%|█████▎    | 34925/65536 [5:59:53<5:23:01,  1.58it/s]
 53%|█████▎    | 34926/65536 [5:59:53<5:19:25,  1.60it/s]
 53%|█████▎    | 34927/65536 [5:59:54<5:09:49,  1.65it/s]
 53%|█████▎    | 34928/65536 [5:59:55<5:29:52,  1.55it/s]
 53%|█████▎    | 34929/65536 [5:59:55<5:22:12,  1.58it/s]
 53%|█████▎    | 34930/65536 [5:59:56<5:19:51,  1.59it/s]
 53%|█████▎    | 34931/65536 [5:59:56<5:07:31,  1.66it/s]
 53%|█████▎    | 34932/65536 [5:59:57<5:10:55,  1.64it/s]
 53%|█████▎    | 34933/65536 [5:59:58<5:12:05,  1.63it/s]
 53%|█████▎    | 34934/65536 [5:59:58<5:12:28,  1.63it/s]
 53%|█████▎    | 34935/65536 [5:59:59<5:10:54,  1.64it/s]
 53%|█████▎    | 34936/65536 [5:59:59<5:02:51,  1.68it/s]
 53%|█████▎    | 34937/65536 [6:00:00<5:03:39,  1.68it/s]
 53%|█████▎    | 34938/65536 [6:00:01<5:03:58,  1.68it/s]
 53%|█████▎    | 34939/65536 [6:00:01<5:00:08,  1.70it/s]
 53%|█████▎    | 34940/65536 [6:00:02<5:06:27,  1.66it/s]
                                                         
{'loss': 1.7043, 'learning_rate': 5.214583078241703e-07, 'epoch': 2156.79}
+
 53%|█████▎    | 34940/65536 [6:00:02<5:06:27,  1.66it/s]
 53%|█████▎    | 34941/65536 [6:00:02<5:05:10,  1.67it/s]
 53%|█████▎    | 34942/65536 [6:00:03<5:05:39,  1.67it/s]
 53%|█████▎    | 34943/65536 [6:00:04<5:15:18,  1.62it/s]
 53%|█████▎    | 34944/65536 [6:00:04<5:24:23,  1.57it/s]
 53%|█████▎    | 34945/65536 [6:00:05<5:23:18,  1.58it/s]
 53%|█████▎    | 34946/65536 [6:00:06<5:21:36,  1.59it/s]
 53%|█████▎    | 34947/65536 [6:00:06<5:15:57,  1.61it/s]
 53%|█████▎    | 34948/65536 [6:00:07<5:20:29,  1.59it/s]
 53%|█████▎    | 34949/65536 [6:00:07<5:14:52,  1.62it/s]
 53%|█████▎    | 34950/65536 [6:00:08<5:08:22,  1.65it/s]
 53%|█████▎    | 34951/65536 [6:00:09<5:16:13,  1.61it/s]
 53%|█████▎    | 34952/65536 [6:00:09<5:22:47,  1.58it/s]
 53%|█████▎    | 34953/65536 [6:00:10<5:13:45,  1.62it/s]
 53%|█████▎    | 34954/65536 [6:00:10<5:07:51,  1.66it/s]
 53%|█████▎    | 34955/65536 [6:00:11<5:01:42,  1.69it/s]
 53%|█████▎    | 34956/65536 [6:00:12<5:03:51,  1.68it/s]
 53%|█████▎    | 34957/65536 [6:00:12<5:06:30,  1.66it/s]
 53%|█████▎    | 34958/65536 [6:00:13<4:59:53,  1.70it/s]
 53%|█████▎    | 34959/65536 [6:00:13<5:09:44,  1.65it/s]
 53%|█████▎    | 34960/65536 [6:00:14<5:22:45,  1.58it/s]
                                                         
{'loss': 1.6333, 'learning_rate': 5.211828088649442e-07, 'epoch': 2158.02}
+
 53%|█████▎    | 34960/65536 [6:00:14<5:22:45,  1.58it/s]
 53%|█████▎    | 34961/65536 [6:00:15<5:21:22,  1.59it/s]
 53%|█████▎    | 34962/65536 [6:00:15<5:22:26,  1.58it/s]
 53%|█████▎    | 34963/65536 [6:00:16<5:20:57,  1.59it/s]
 53%|█████▎    | 34964/65536 [6:00:17<5:17:00,  1.61it/s]
 53%|█████▎    | 34965/65536 [6:00:17<5:28:10,  1.55it/s]
 53%|█████▎    | 34966/65536 [6:00:18<5:29:00,  1.55it/s]
 53%|█████▎    | 34967/65536 [6:00:19<5:23:13,  1.58it/s]
 53%|█████▎    | 34968/65536 [6:00:19<5:09:24,  1.65it/s]
 53%|█████▎    | 34969/65536 [6:00:20<5:02:40,  1.68it/s]
 53%|█████▎    | 34970/65536 [6:00:20<5:00:30,  1.70it/s]
 53%|█████▎    | 34971/65536 [6:00:21<5:00:32,  1.69it/s]
 53%|█████▎    | 34972/65536 [6:00:22<5:17:38,  1.60it/s]
 53%|█████▎    | 34973/65536 [6:00:22<5:11:56,  1.63it/s]
 53%|█████▎    | 34974/65536 [6:00:23<5:11:08,  1.64it/s]
 53%|█████▎    | 34975/65536 [6:00:23<5:12:41,  1.63it/s]
 53%|█████▎    | 34976/65536 [6:00:24<5:24:07,  1.57it/s]
 53%|█████▎    | 34977/65536 [6:00:25<5:25:29,  1.56it/s]
 53%|█████▎    | 34978/65536 [6:00:25<5:21:27,  1.58it/s]
 53%|█████▎    | 34979/65536 [6:00:26<5:15:36,  1.61it/s]
 53%|█████▎    | 34980/65536 [6:00:27<5:17:06,  1.61it/s]
                                                         
{'loss': 1.6256, 'learning_rate': 5.20907309905718e-07, 'epoch': 2159.26}
+
 53%|█████▎    | 34980/65536 [6:00:27<5:17:06,  1.61it/s]
 53%|█████▎    | 34981/65536 [6:00:27<5:14:11,  1.62it/s]
 53%|█████▎    | 34982/65536 [6:00:28<5:11:06,  1.64it/s]
 53%|█████▎    | 34983/65536 [6:00:28<5:13:14,  1.63it/s]
 53%|█████▎    | 34984/65536 [6:00:29<5:14:22,  1.62it/s]
 53%|█████▎    | 34985/65536 [6:00:30<5:13:25,  1.62it/s]
 53%|█████▎    | 34986/65536 [6:00:30<5:08:21,  1.65it/s]
 53%|█████▎    | 34987/65536 [6:00:31<5:10:04,  1.64it/s]
 53%|█████▎    | 34988/65536 [6:00:31<5:05:31,  1.67it/s]
 53%|█████▎    | 34989/65536 [6:00:32<5:14:50,  1.62it/s]
 53%|█████▎    | 34990/65536 [6:00:33<5:16:34,  1.61it/s]
 53%|█████▎    | 34991/65536 [6:00:33<5:16:48,  1.61it/s]
 53%|█████▎    | 34992/65536 [6:00:34<5:10:00,  1.64it/s]
 53%|█████▎    | 34993/65536 [6:00:35<5:18:59,  1.60it/s]
 53%|█████▎    | 34994/65536 [6:00:35<5:19:20,  1.59it/s]
 53%|█████▎    | 34995/65536 [6:00:36<5:13:38,  1.62it/s]
 53%|█████▎    | 34996/65536 [6:00:36<5:14:56,  1.62it/s]
 53%|█████▎    | 34997/65536 [6:00:37<5:12:50,  1.63it/s]
 53%|█████▎    | 34998/65536 [6:00:38<5:21:29,  1.58it/s]
 53%|█████▎    | 34999/65536 [6:00:38<5:26:05,  1.56it/s]
 53%|█████▎    | 35000/65536 [6:00:39<5:22:00,  1.58it/s]
                                                         
{'loss': 1.5845, 'learning_rate': 5.20631810946492e-07, 'epoch': 2160.49}
+
 53%|█████▎    | 35000/65536 [6:00:39<5:22:00,  1.58it/s]
 53%|█████▎    | 35001/65536 [6:00:40<5:17:25,  1.60it/s]
 53%|█████▎    | 35002/65536 [6:00:40<5:17:35,  1.60it/s]
 53%|█████▎    | 35003/65536 [6:00:41<5:16:50,  1.61it/s]
 53%|█████▎    | 35004/65536 [6:00:41<5:05:07,  1.67it/s]
 53%|█████▎    | 35005/65536 [6:00:42<5:04:12,  1.67it/s]
 53%|█████▎    | 35006/65536 [6:00:43<5:08:51,  1.65it/s]
 53%|█████▎    | 35007/65536 [6:00:43<5:06:08,  1.66it/s]
 53%|█████▎    | 35008/65536 [6:00:44<5:05:44,  1.66it/s]
 53%|█████▎    | 35009/65536 [6:00:44<5:16:32,  1.61it/s]
 53%|█████▎    | 35010/65536 [6:00:45<5:11:12,  1.63it/s]
 53%|█████▎    | 35011/65536 [6:00:46<5:17:20,  1.60it/s]
 53%|█████▎    | 35012/65536 [6:00:46<5:07:16,  1.66it/s]
 53%|█████▎    | 35013/65536 [6:00:47<5:03:25,  1.68it/s]
 53%|█████▎    | 35014/65536 [6:00:47<5:04:32,  1.67it/s]
 53%|█████▎    | 35015/65536 [6:00:48<5:07:08,  1.66it/s]
 53%|█████▎    | 35016/65536 [6:00:49<5:12:29,  1.63it/s]
 53%|█████▎    | 35017/65536 [6:00:49<5:06:15,  1.66it/s]
 53%|█████▎    | 35018/65536 [6:00:50<5:09:51,  1.64it/s]
 53%|█████▎    | 35019/65536 [6:00:50<5:14:28,  1.62it/s]
 53%|█████▎    | 35020/65536 [6:00:51<5:13:29,  1.62it/s]
                                                         
{'loss': 1.681, 'learning_rate': 5.203563119872659e-07, 'epoch': 2161.73}
+
 53%|█████▎    | 35020/65536 [6:00:51<5:13:29,  1.62it/s]
 53%|█████▎    | 35021/65536 [6:00:52<5:06:10,  1.66it/s]
 53%|█████▎    | 35022/65536 [6:00:52<5:07:52,  1.65it/s]
 53%|█████▎    | 35023/65536 [6:00:53<5:16:22,  1.61it/s]
 53%|█████▎    | 35024/65536 [6:00:54<5:15:42,  1.61it/s]
 53%|█████▎    | 35025/65536 [6:00:54<5:25:51,  1.56it/s]
 53%|█████▎    | 35026/65536 [6:00:55<5:16:53,  1.60it/s]
 53%|█████▎    | 35027/65536 [6:00:55<5:16:31,  1.61it/s]
 53%|█████▎    | 35028/65536 [6:00:56<5:18:09,  1.60it/s]
 53%|█████▎    | 35029/65536 [6:00:57<5:21:22,  1.58it/s]
 53%|█████▎    | 35030/65536 [6:00:57<5:15:43,  1.61it/s]
 53%|█████▎    | 35031/65536 [6:00:58<5:08:04,  1.65it/s]
 53%|█████▎    | 35032/65536 [6:00:58<5:04:32,  1.67it/s]
 53%|█████▎    | 35033/65536 [6:00:59<5:00:18,  1.69it/s]
 53%|█████▎    | 35034/65536 [6:01:00<5:03:34,  1.67it/s]
 53%|█████▎    | 35035/65536 [6:01:00<5:17:18,  1.60it/s]
 53%|█████▎    | 35036/65536 [6:01:01<5:20:20,  1.59it/s]
 53%|█████▎    | 35037/65536 [6:01:02<5:16:37,  1.61it/s]
 53%|█████▎    | 35038/65536 [6:01:02<5:05:49,  1.66it/s]
 53%|█████▎    | 35039/65536 [6:01:03<5:15:00,  1.61it/s]
 53%|█████▎    | 35040/65536 [6:01:03<5:12:26,  1.63it/s]
                                                         
{'loss': 1.6559, 'learning_rate': 5.200808130280396e-07, 'epoch': 2162.96}
+
 53%|█████▎    | 35040/65536 [6:01:03<5:12:26,  1.63it/s]
 53%|█████▎    | 35041/65536 [6:01:04<5:19:08,  1.59it/s]
 53%|█████▎    | 35042/65536 [6:01:05<5:16:53,  1.60it/s]
 53%|█████▎    | 35043/65536 [6:01:05<5:18:49,  1.59it/s]
 53%|█████▎    | 35044/65536 [6:01:06<5:17:15,  1.60it/s]
 53%|█████▎    | 35045/65536 [6:01:07<5:11:41,  1.63it/s]
 53%|█████▎    | 35046/65536 [6:01:07<5:02:27,  1.68it/s]
 53%|█████▎    | 35047/65536 [6:01:08<5:05:33,  1.66it/s]
 53%|█████▎    | 35048/65536 [6:01:08<5:09:52,  1.64it/s]
 53%|█████▎    | 35049/65536 [6:01:09<5:10:37,  1.64it/s]
 53%|█████▎    | 35050/65536 [6:01:10<5:08:31,  1.65it/s]
 53%|█████▎    | 35051/65536 [6:01:10<5:14:17,  1.62it/s]
 53%|█████▎    | 35052/65536 [6:01:11<5:10:15,  1.64it/s]
 53%|█████▎    | 35053/65536 [6:01:11<5:08:42,  1.65it/s]
 53%|█████▎    | 35054/65536 [6:01:12<5:06:56,  1.66it/s]
 53%|█████▎    | 35055/65536 [6:01:13<5:09:33,  1.64it/s]
 53%|█████▎    | 35056/65536 [6:01:13<5:19:01,  1.59it/s]
 53%|█████▎    | 35057/65536 [6:01:14<5:25:29,  1.56it/s]
 53%|█████▎    | 35058/65536 [6:01:15<5:17:58,  1.60it/s]
 53%|█████▎    | 35059/65536 [6:01:15<5:27:18,  1.55it/s]
 53%|█████▎    | 35060/65536 [6:01:16<5:19:31,  1.59it/s]
                                                         
{'loss': 1.6413, 'learning_rate': 5.198053140688135e-07, 'epoch': 2164.2}
+
 53%|█████▎    | 35060/65536 [6:01:16<5:19:31,  1.59it/s]
 53%|█████▎    | 35061/65536 [6:01:16<5:08:31,  1.65it/s]
 54%|█████▎    | 35062/65536 [6:01:17<5:05:06,  1.66it/s]
 54%|█████▎    | 35063/65536 [6:01:18<5:05:05,  1.66it/s]
 54%|█████▎    | 35064/65536 [6:01:18<5:03:15,  1.67it/s]
 54%|█████▎    | 35065/65536 [6:01:19<5:11:39,  1.63it/s]
 54%|█████▎    | 35066/65536 [6:01:19<5:08:36,  1.65it/s]
 54%|█████▎    | 35067/65536 [6:01:20<5:02:28,  1.68it/s]
 54%|█████▎    | 35068/65536 [6:01:21<5:07:26,  1.65it/s]
 54%|█████▎    | 35069/65536 [6:01:21<5:04:21,  1.67it/s]
 54%|█████▎    | 35070/65536 [6:01:22<5:12:38,  1.62it/s]
 54%|█████▎    | 35071/65536 [6:01:22<5:12:18,  1.63it/s]
 54%|█████▎    | 35072/65536 [6:01:23<5:18:47,  1.59it/s]
 54%|█████▎    | 35073/65536 [6:01:24<5:20:22,  1.58it/s]
 54%|█████▎    | 35074/65536 [6:01:24<5:23:00,  1.57it/s]
 54%|█████▎    | 35075/65536 [6:01:25<5:24:17,  1.57it/s]
 54%|█████▎    | 35076/65536 [6:01:26<5:25:07,  1.56it/s]
 54%|█████▎    | 35077/65536 [6:01:26<5:20:52,  1.58it/s]
 54%|█████▎    | 35078/65536 [6:01:27<5:08:18,  1.65it/s]
 54%|█████▎    | 35079/65536 [6:01:27<5:12:42,  1.62it/s]
 54%|█████▎    | 35080/65536 [6:01:28<5:08:20,  1.65it/s]
                                                         
{'loss': 1.6549, 'learning_rate': 5.195298151095873e-07, 'epoch': 2165.43}
+
 54%|█████▎    | 35080/65536 [6:01:28<5:08:20,  1.65it/s]
 54%|█████▎    | 35081/65536 [6:01:29<5:04:33,  1.67it/s]
 54%|█████▎    | 35082/65536 [6:01:29<5:16:44,  1.60it/s]
 54%|█████▎    | 35083/65536 [6:01:30<5:14:42,  1.61it/s]
 54%|█████▎    | 35084/65536 [6:01:31<5:11:36,  1.63it/s]
 54%|█████▎    | 35085/65536 [6:01:31<5:09:52,  1.64it/s]
 54%|█████▎    | 35086/65536 [6:01:32<5:11:22,  1.63it/s]
 54%|█████▎    | 35087/65536 [6:01:32<5:03:18,  1.67it/s]
 54%|█████▎    | 35088/65536 [6:01:33<5:18:23,  1.59it/s]
 54%|█████▎    | 35089/65536 [6:01:34<5:09:56,  1.64it/s]
 54%|█████▎    | 35090/65536 [6:01:34<5:16:50,  1.60it/s]
 54%|█████▎    | 35091/65536 [6:01:35<5:10:34,  1.63it/s]
 54%|█████▎    | 35092/65536 [6:01:36<5:27:53,  1.55it/s]
 54%|█████▎    | 35093/65536 [6:01:36<5:18:50,  1.59it/s]
 54%|█████▎    | 35094/65536 [6:01:37<5:17:47,  1.60it/s]
 54%|█████▎    | 35095/65536 [6:01:37<5:11:10,  1.63it/s]
 54%|█████▎    | 35096/65536 [6:01:38<5:07:50,  1.65it/s]
 54%|█████▎    | 35097/65536 [6:01:39<5:03:53,  1.67it/s]
 54%|█████▎    | 35098/65536 [6:01:39<5:06:55,  1.65it/s]
 54%|█████▎    | 35099/65536 [6:01:40<5:11:28,  1.63it/s]
 54%|█████▎    | 35100/65536 [6:01:40<5:11:30,  1.63it/s]
                                                         
{'loss': 1.6315, 'learning_rate': 5.192543161503612e-07, 'epoch': 2166.67}
+
 54%|█████▎    | 35100/65536 [6:01:40<5:11:30,  1.63it/s]
 54%|█████▎    | 35101/65536 [6:01:41<5:06:32,  1.65it/s]
 54%|█████▎    | 35102/65536 [6:01:42<5:08:58,  1.64it/s]
 54%|█████▎    | 35103/65536 [6:01:42<5:14:42,  1.61it/s]
 54%|█████▎    | 35104/65536 [6:01:43<5:12:22,  1.62it/s]
 54%|█████▎    | 35105/65536 [6:01:43<5:05:57,  1.66it/s]
 54%|█████▎    | 35106/65536 [6:01:44<5:21:26,  1.58it/s]
 54%|█████▎    | 35107/65536 [6:01:45<5:17:57,  1.60it/s]
 54%|█████▎    | 35108/65536 [6:01:45<5:16:53,  1.60it/s]
 54%|█████▎    | 35109/65536 [6:01:46<5:18:08,  1.59it/s]
 54%|█████▎    | 35110/65536 [6:01:47<5:17:16,  1.60it/s]
 54%|█████▎    | 35111/65536 [6:01:47<5:21:41,  1.58it/s]
 54%|█████▎    | 35112/65536 [6:01:48<5:16:01,  1.60it/s]
 54%|█████▎    | 35113/65536 [6:01:49<5:24:35,  1.56it/s]
 54%|█████▎    | 35114/65536 [6:01:49<5:15:22,  1.61it/s]
 54%|█████▎    | 35115/65536 [6:01:50<5:11:32,  1.63it/s]
 54%|█████▎    | 35116/65536 [6:01:50<5:06:16,  1.66it/s]
 54%|█████▎    | 35117/65536 [6:01:51<5:12:55,  1.62it/s]
 54%|█████▎    | 35118/65536 [6:01:52<5:12:33,  1.62it/s]
 54%|█████▎    | 35119/65536 [6:01:52<5:06:55,  1.65it/s]
 54%|█████▎    | 35120/65536 [6:01:53<5:07:08,  1.65it/s]
                                                         
{'loss': 1.5978, 'learning_rate': 5.18978817191135e-07, 'epoch': 2167.9}
+
 54%|█████▎    | 35120/65536 [6:01:53<5:07:08,  1.65it/s]
 54%|█████▎    | 35121/65536 [6:01:53<5:10:06,  1.63it/s]
 54%|█████▎    | 35122/65536 [6:01:54<5:25:00,  1.56it/s]
 54%|█████▎    | 35123/65536 [6:01:55<5:15:15,  1.61it/s]
 54%|█████▎    | 35124/65536 [6:01:55<5:10:15,  1.63it/s]
 54%|█████▎    | 35125/65536 [6:01:56<5:12:39,  1.62it/s]
 54%|█████▎    | 35126/65536 [6:01:56<5:07:54,  1.65it/s]
 54%|█████▎    | 35127/65536 [6:01:57<5:12:50,  1.62it/s]
 54%|█████▎    | 35128/65536 [6:01:58<5:14:53,  1.61it/s]
 54%|█████▎    | 35129/65536 [6:01:58<5:13:21,  1.62it/s]
 54%|█████▎    | 35130/65536 [6:01:59<5:14:22,  1.61it/s]
 54%|█████▎    | 35131/65536 [6:02:00<5:08:32,  1.64it/s]
 54%|█████▎    | 35132/65536 [6:02:00<5:06:04,  1.66it/s]
 54%|█████▎    | 35133/65536 [6:02:01<5:07:45,  1.65it/s]
 54%|█████▎    | 35134/65536 [6:02:01<5:11:56,  1.62it/s]
 54%|█████▎    | 35135/65536 [6:02:02<5:13:12,  1.62it/s]
 54%|█████▎    | 35136/65536 [6:02:03<5:08:10,  1.64it/s]
 54%|█████▎    | 35137/65536 [6:02:03<5:08:14,  1.64it/s]
 54%|█████▎    | 35138/65536 [6:02:04<5:18:50,  1.59it/s]
 54%|█████▎    | 35139/65536 [6:02:04<5:11:23,  1.63it/s]
 54%|█████▎    | 35140/65536 [6:02:05<5:10:41,  1.63it/s]
                                                         
{'loss': 1.6748, 'learning_rate': 5.187033182319089e-07, 'epoch': 2169.14}
+
 54%|█████▎    | 35140/65536 [6:02:05<5:10:41,  1.63it/s]
 54%|█████▎    | 35141/65536 [6:02:06<5:07:52,  1.65it/s]
 54%|█████▎    | 35142/65536 [6:02:06<5:07:15,  1.65it/s]
 54%|█████▎    | 35143/65536 [6:02:07<4:59:32,  1.69it/s]
 54%|█████▎    | 35144/65536 [6:02:07<5:03:54,  1.67it/s]
 54%|█████▎    | 35145/65536 [6:02:08<5:06:34,  1.65it/s]
 54%|█████▎    | 35146/65536 [6:02:09<5:14:33,  1.61it/s]
 54%|█████▎    | 35147/65536 [6:02:09<5:16:46,  1.60it/s]
 54%|█████▎    | 35148/65536 [6:02:10<5:18:49,  1.59it/s]
 54%|█████▎    | 35149/65536 [6:02:11<5:17:46,  1.59it/s]
 54%|█████▎    | 35150/65536 [6:02:11<5:09:52,  1.63it/s]
 54%|█████▎    | 35151/65536 [6:02:12<5:06:05,  1.65it/s]
 54%|█████▎    | 35152/65536 [6:02:12<5:05:48,  1.66it/s]
 54%|█████▎    | 35153/65536 [6:02:13<5:13:36,  1.61it/s]
 54%|█████▎    | 35154/65536 [6:02:14<5:12:37,  1.62it/s]
 54%|█████▎    | 35155/65536 [6:02:14<5:27:33,  1.55it/s]
 54%|█████▎    | 35156/65536 [6:02:15<5:20:26,  1.58it/s]
 54%|█████▎    | 35157/65536 [6:02:16<5:21:02,  1.58it/s]
 54%|█████▎    | 35158/65536 [6:02:16<5:19:52,  1.58it/s]
 54%|█████▎    | 35159/65536 [6:02:17<5:15:20,  1.61it/s]
 54%|█████▎    | 35160/65536 [6:02:17<5:09:58,  1.63it/s]
                                                         
{'loss': 1.6301, 'learning_rate': 5.184278192726828e-07, 'epoch': 2170.37}
+
 54%|█████▎    | 35160/65536 [6:02:17<5:09:58,  1.63it/s]
 54%|█████▎    | 35161/65536 [6:02:18<5:05:03,  1.66it/s]
 54%|█████▎    | 35162/65536 [6:02:19<5:10:12,  1.63it/s]
 54%|█████▎    | 35163/65536 [6:02:19<5:02:05,  1.68it/s]
 54%|█████▎    | 35164/65536 [6:02:20<5:11:28,  1.63it/s]
 54%|█████▎    | 35165/65536 [6:02:20<5:04:25,  1.66it/s]
 54%|█████▎    | 35166/65536 [6:02:21<5:05:54,  1.65it/s]
 54%|█████▎    | 35167/65536 [6:02:22<5:09:04,  1.64it/s]
 54%|█████▎    | 35168/65536 [6:02:22<5:04:49,  1.66it/s]
 54%|█████▎    | 35169/65536 [6:02:23<5:05:54,  1.65it/s]
 54%|█████▎    | 35170/65536 [6:02:24<5:17:19,  1.59it/s]
 54%|█████▎    | 35171/65536 [6:02:24<5:22:55,  1.57it/s]
 54%|█████▎    | 35172/65536 [6:02:25<5:17:58,  1.59it/s]
 54%|█████▎    | 35173/65536 [6:02:25<5:10:27,  1.63it/s]
 54%|█████▎    | 35174/65536 [6:02:26<5:11:46,  1.62it/s]
 54%|█████▎    | 35175/65536 [6:02:27<5:22:26,  1.57it/s]
 54%|█████▎    | 35176/65536 [6:02:27<5:19:51,  1.58it/s]
 54%|█████▎    | 35177/65536 [6:02:28<5:21:42,  1.57it/s]
 54%|█████▎    | 35178/65536 [6:02:29<5:31:11,  1.53it/s]
 54%|█████▎    | 35179/65536 [6:02:29<5:30:52,  1.53it/s]
 54%|█████▎    | 35180/65536 [6:02:30<5:24:43,  1.56it/s]
                                                         
{'loss': 1.6186, 'learning_rate': 5.181523203134565e-07, 'epoch': 2171.6}
+
 54%|█████▎    | 35180/65536 [6:02:30<5:24:43,  1.56it/s]
 54%|█████▎    | 35181/65536 [6:02:31<5:24:08,  1.56it/s]
 54%|█████▎    | 35182/65536 [6:02:31<5:15:57,  1.60it/s]
 54%|█████▎    | 35183/65536 [6:02:32<5:19:24,  1.58it/s]
 54%|█████▎    | 35184/65536 [6:02:32<5:19:41,  1.58it/s]
 54%|█████▎    | 35185/65536 [6:02:33<5:18:36,  1.59it/s]
 54%|█████▎    | 35186/65536 [6:02:34<5:17:15,  1.59it/s]
 54%|█████▎    | 35187/65536 [6:02:34<5:22:38,  1.57it/s]
 54%|█████▎    | 35188/65536 [6:02:35<5:25:10,  1.56it/s]
 54%|█████▎    | 35189/65536 [6:02:36<5:21:54,  1.57it/s]
 54%|█████▎    | 35190/65536 [6:02:36<5:12:48,  1.62it/s]
 54%|█████▎    | 35191/65536 [6:02:37<5:14:54,  1.61it/s]
 54%|█████▎    | 35192/65536 [6:02:37<5:10:44,  1.63it/s]
 54%|█████▎    | 35193/65536 [6:02:38<5:02:37,  1.67it/s]
 54%|█████▎    | 35194/65536 [6:02:39<5:12:20,  1.62it/s]
 54%|█████▎    | 35195/65536 [6:02:39<5:07:27,  1.64it/s]
 54%|█████▎    | 35196/65536 [6:02:40<5:11:17,  1.62it/s]
 54%|█████▎    | 35197/65536 [6:02:40<5:08:34,  1.64it/s]
 54%|█████▎    | 35198/65536 [6:02:41<5:06:09,  1.65it/s]
 54%|█████▎    | 35199/65536 [6:02:42<5:03:39,  1.67it/s]
 54%|█████▎    | 35200/65536 [6:02:42<5:02:09,  1.67it/s]
                                                         
{'loss': 1.6626, 'learning_rate': 5.178768213542304e-07, 'epoch': 2172.84}
+
 54%|█████▎    | 35200/65536 [6:02:42<5:02:09,  1.67it/s]
 54%|█████▎    | 35201/65536 [6:02:43<5:11:00,  1.63it/s]
 54%|█████▎    | 35202/65536 [6:02:43<5:02:05,  1.67it/s]
 54%|█████▎    | 35203/65536 [6:02:44<5:13:27,  1.61it/s]
 54%|█████▎    | 35204/65536 [6:02:45<5:06:13,  1.65it/s]
 54%|█████▎    | 35205/65536 [6:02:45<5:10:17,  1.63it/s]
 54%|█████▎    | 35206/65536 [6:02:46<5:15:34,  1.60it/s]
 54%|█████▎    | 35207/65536 [6:02:47<5:17:59,  1.59it/s]
 54%|█████▎    | 35208/65536 [6:02:47<5:14:18,  1.61it/s]
 54%|█████▎    | 35209/65536 [6:02:48<5:16:40,  1.60it/s]
 54%|█████▎    | 35210/65536 [6:02:48<5:12:02,  1.62it/s]
 54%|█████▎    | 35211/65536 [6:02:49<5:05:07,  1.66it/s]
 54%|█████▎    | 35212/65536 [6:02:50<5:03:26,  1.67it/s]
 54%|█████▎    | 35213/65536 [6:02:50<5:04:32,  1.66it/s]
 54%|█████▎    | 35214/65536 [6:02:51<5:03:49,  1.66it/s]
 54%|█████▎    | 35215/65536 [6:02:51<5:10:45,  1.63it/s]
 54%|█████▎    | 35216/65536 [6:02:52<5:08:00,  1.64it/s]
 54%|█████▎    | 35217/65536 [6:02:53<5:06:11,  1.65it/s]
 54%|█████▎    | 35218/65536 [6:02:53<5:01:23,  1.68it/s]
 54%|█████▎    | 35219/65536 [6:02:54<5:20:42,  1.58it/s]
 54%|█████▎    | 35220/65536 [6:02:55<5:14:23,  1.61it/s]
                                                         
{'loss': 1.6176, 'learning_rate': 5.176013223950042e-07, 'epoch': 2174.07}
+
 54%|█████▎    | 35220/65536 [6:02:55<5:14:23,  1.61it/s]
 54%|█████▎    | 35221/65536 [6:02:55<5:17:04,  1.59it/s]
 54%|█████▎    | 35222/65536 [6:02:56<5:12:06,  1.62it/s]
 54%|█████▎    | 35223/65536 [6:02:56<5:14:44,  1.61it/s]
 54%|█████▎    | 35224/65536 [6:02:57<5:14:18,  1.61it/s]
 54%|█████▎    | 35225/65536 [6:02:58<5:23:42,  1.56it/s]
 54%|█████▍    | 35226/65536 [6:02:58<5:17:08,  1.59it/s]
 54%|█████▍    | 35227/65536 [6:02:59<5:11:58,  1.62it/s]
 54%|█████▍    | 35228/65536 [6:03:00<5:09:39,  1.63it/s]
 54%|█████▍    | 35229/65536 [6:03:00<5:09:02,  1.63it/s]
 54%|█████▍    | 35230/65536 [6:03:01<5:07:56,  1.64it/s]
 54%|█████▍    | 35231/65536 [6:03:01<5:08:44,  1.64it/s]
 54%|█████▍    | 35232/65536 [6:03:02<5:03:32,  1.66it/s]
 54%|█████▍    | 35233/65536 [6:03:03<5:06:16,  1.65it/s]
 54%|█████▍    | 35234/65536 [6:03:03<5:06:53,  1.65it/s]
 54%|█████▍    | 35235/65536 [6:03:04<5:12:23,  1.62it/s]
 54%|█████▍    | 35236/65536 [6:03:05<5:27:14,  1.54it/s]
 54%|█████▍    | 35237/65536 [6:03:05<5:18:16,  1.59it/s]
 54%|█████▍    | 35238/65536 [6:03:06<5:13:15,  1.61it/s]
 54%|█████▍    | 35239/65536 [6:03:06<5:12:26,  1.62it/s]
 54%|█████▍    | 35240/65536 [6:03:07<5:12:59,  1.61it/s]
                                                         
{'loss': 1.6184, 'learning_rate': 5.173258234357781e-07, 'epoch': 2175.31}
+
 54%|█████▍    | 35240/65536 [6:03:07<5:12:59,  1.61it/s]
 54%|█████▍    | 35241/65536 [6:03:08<5:21:37,  1.57it/s]
 54%|█████▍    | 35242/65536 [6:03:08<5:13:29,  1.61it/s]
 54%|█████▍    | 35243/65536 [6:03:09<5:08:18,  1.64it/s]
 54%|█████▍    | 35244/65536 [6:03:09<5:08:02,  1.64it/s]
 54%|█████▍    | 35245/65536 [6:03:10<5:04:45,  1.66it/s]
 54%|█████▍    | 35246/65536 [6:03:11<5:00:59,  1.68it/s]
 54%|█████▍    | 35247/65536 [6:03:11<4:56:46,  1.70it/s]
 54%|█████▍    | 35248/65536 [6:03:12<5:10:21,  1.63it/s]
 54%|█████▍    | 35249/65536 [6:03:12<5:05:36,  1.65it/s]
 54%|█████▍    | 35250/65536 [6:03:13<5:00:39,  1.68it/s]
 54%|█████▍    | 35251/65536 [6:03:14<5:01:25,  1.67it/s]
 54%|█████▍    | 35252/65536 [6:03:14<5:11:57,  1.62it/s]
 54%|█████▍    | 35253/65536 [6:03:15<5:11:38,  1.62it/s]
 54%|█████▍    | 35254/65536 [6:03:15<5:10:41,  1.62it/s]
 54%|█████▍    | 35255/65536 [6:03:16<5:13:41,  1.61it/s]
 54%|█████▍    | 35256/65536 [6:03:17<5:13:25,  1.61it/s]
 54%|█████▍    | 35257/65536 [6:03:17<5:06:41,  1.65it/s]
 54%|█████▍    | 35258/65536 [6:03:18<5:11:23,  1.62it/s]
 54%|█████▍    | 35259/65536 [6:03:19<5:20:59,  1.57it/s]
 54%|█████▍    | 35260/65536 [6:03:19<5:14:50,  1.60it/s]
                                                         
{'loss': 1.669, 'learning_rate': 5.17050324476552e-07, 'epoch': 2176.54}
+
 54%|█████▍    | 35260/65536 [6:03:19<5:14:50,  1.60it/s]
 54%|█████▍    | 35261/65536 [6:03:20<5:10:25,  1.63it/s]
 54%|█████▍    | 35262/65536 [6:03:20<5:05:41,  1.65it/s]
 54%|█████▍    | 35263/65536 [6:03:21<5:00:42,  1.68it/s]
 54%|█████▍    | 35264/65536 [6:03:22<5:04:23,  1.66it/s]
 54%|█████▍    | 35265/65536 [6:03:22<5:01:46,  1.67it/s]
 54%|█████▍    | 35266/65536 [6:03:23<5:05:48,  1.65it/s]
 54%|█████▍    | 35267/65536 [6:03:23<5:03:47,  1.66it/s]
 54%|█████▍    | 35268/65536 [6:03:24<5:19:19,  1.58it/s]
 54%|█████▍    | 35269/65536 [6:03:25<5:13:04,  1.61it/s]
 54%|█████▍    | 35270/65536 [6:03:25<5:14:08,  1.61it/s]
 54%|█████▍    | 35271/65536 [6:03:26<5:17:47,  1.59it/s]
 54%|█████▍    | 35272/65536 [6:03:27<5:21:03,  1.57it/s]
 54%|█████▍    | 35273/65536 [6:03:27<5:15:02,  1.60it/s]
 54%|█████▍    | 35274/65536 [6:03:28<5:10:07,  1.63it/s]
 54%|█████▍    | 35275/65536 [6:03:28<5:00:58,  1.68it/s]
 54%|█████▍    | 35276/65536 [6:03:29<5:03:43,  1.66it/s]
 54%|█████▍    | 35277/65536 [6:03:30<5:10:03,  1.63it/s]
 54%|█████▍    | 35278/65536 [6:03:30<5:12:26,  1.61it/s]
 54%|█████▍    | 35279/65536 [6:03:31<5:10:21,  1.62it/s]
 54%|█████▍    | 35280/65536 [6:03:31<5:08:21,  1.64it/s]
                                                         
{'loss': 1.6438, 'learning_rate': 5.167748255173257e-07, 'epoch': 2177.78}
+
 54%|█████▍    | 35280/65536 [6:03:31<5:08:21,  1.64it/s]
 54%|█████▍    | 35281/65536 [6:03:32<5:12:25,  1.61it/s]
 54%|█████▍    | 35282/65536 [6:03:33<5:09:36,  1.63it/s]
 54%|█████▍    | 35283/65536 [6:03:33<5:01:20,  1.67it/s]
 54%|█████▍    | 35284/65536 [6:03:34<5:09:56,  1.63it/s]
 54%|█████▍    | 35285/65536 [6:03:35<5:08:40,  1.63it/s]
 54%|█████▍    | 35286/65536 [6:03:35<5:05:50,  1.65it/s]
 54%|█████▍    | 35287/65536 [6:03:36<5:08:53,  1.63it/s]
 54%|█████▍    | 35288/65536 [6:03:36<5:09:40,  1.63it/s]
 54%|█████▍    | 35289/65536 [6:03:37<5:14:36,  1.60it/s]
 54%|█████▍    | 35290/65536 [6:03:38<5:06:01,  1.65it/s]
 54%|█████▍    | 35291/65536 [6:03:38<5:06:00,  1.65it/s]
 54%|█████▍    | 35292/65536 [6:03:39<5:07:38,  1.64it/s]
 54%|█████▍    | 35293/65536 [6:03:39<5:08:37,  1.63it/s]
 54%|█████▍    | 35294/65536 [6:03:40<5:06:57,  1.64it/s]
 54%|█████▍    | 35295/65536 [6:03:41<4:59:37,  1.68it/s]
 54%|█████▍    | 35296/65536 [6:03:41<5:14:09,  1.60it/s]
 54%|█████▍    | 35297/65536 [6:03:42<5:06:20,  1.65it/s]
 54%|█████▍    | 35298/65536 [6:03:42<5:08:15,  1.63it/s]
 54%|█████▍    | 35299/65536 [6:03:43<5:06:47,  1.64it/s]
 54%|█████▍    | 35300/65536 [6:03:44<5:13:15,  1.61it/s]
                                                         
{'loss': 1.6993, 'learning_rate': 5.164993265580997e-07, 'epoch': 2179.01}
+
 54%|█████▍    | 35300/65536 [6:03:44<5:13:15,  1.61it/s]
 54%|█████▍    | 35301/65536 [6:03:44<5:17:09,  1.59it/s]
 54%|█████▍    | 35302/65536 [6:03:45<5:11:42,  1.62it/s]
 54%|█████▍    | 35303/65536 [6:03:46<5:11:43,  1.62it/s]
 54%|█████▍    | 35304/65536 [6:03:46<5:19:42,  1.58it/s]
 54%|█████▍    | 35305/65536 [6:03:47<5:30:47,  1.52it/s]
 54%|█████▍    | 35306/65536 [6:03:48<5:23:30,  1.56it/s]
 54%|█████▍    | 35307/65536 [6:03:48<5:13:30,  1.61it/s]
 54%|█████▍    | 35308/65536 [6:03:49<5:03:43,  1.66it/s]
 54%|█████▍    | 35309/65536 [6:03:49<5:06:50,  1.64it/s]
 54%|█████▍    | 35310/65536 [6:03:50<5:00:55,  1.67it/s]
 54%|█████▍    | 35311/65536 [6:03:50<4:57:46,  1.69it/s]
 54%|█████▍    | 35312/65536 [6:03:51<4:56:59,  1.70it/s]
 54%|█████▍    | 35313/65536 [6:03:52<4:59:56,  1.68it/s]
 54%|█████▍    | 35314/65536 [6:03:52<4:57:36,  1.69it/s]
 54%|█████▍    | 35315/65536 [6:03:53<5:05:34,  1.65it/s]
 54%|█████▍    | 35316/65536 [6:03:53<5:04:48,  1.65it/s]
 54%|█████▍    | 35317/65536 [6:03:54<5:16:52,  1.59it/s]
 54%|█████▍    | 35318/65536 [6:03:55<5:07:36,  1.64it/s]
 54%|█████▍    | 35319/65536 [6:03:55<5:05:27,  1.65it/s]
 54%|█████▍    | 35320/65536 [6:03:56<5:12:29,  1.61it/s]
                                                         
{'loss': 1.6705, 'learning_rate': 5.162238275988734e-07, 'epoch': 2180.25}
+
 54%|█████▍    | 35320/65536 [6:03:56<5:12:29,  1.61it/s]
 54%|█████▍    | 35321/65536 [6:03:57<5:11:24,  1.62it/s]
 54%|█████▍    | 35322/65536 [6:03:57<5:11:03,  1.62it/s]
 54%|█████▍    | 35323/65536 [6:03:58<5:02:25,  1.66it/s]
 54%|█████▍    | 35324/65536 [6:03:58<5:04:33,  1.65it/s]
 54%|█████▍    | 35325/65536 [6:03:59<5:02:33,  1.66it/s]
 54%|█████▍    | 35326/65536 [6:04:00<5:01:40,  1.67it/s]
 54%|█████▍    | 35327/65536 [6:04:00<5:15:17,  1.60it/s]
 54%|█████▍    | 35328/65536 [6:04:01<5:10:53,  1.62it/s]
 54%|█████▍    | 35329/65536 [6:04:01<5:15:00,  1.60it/s]
 54%|█████▍    | 35330/65536 [6:04:02<5:21:33,  1.57it/s]
 54%|█████▍    | 35331/65536 [6:04:03<5:19:32,  1.58it/s]
 54%|█████▍    | 35332/65536 [6:04:03<5:13:08,  1.61it/s]
 54%|█████▍    | 35333/65536 [6:04:04<5:18:48,  1.58it/s]
 54%|█████▍    | 35334/65536 [6:04:05<5:15:44,  1.59it/s]
 54%|█████▍    | 35335/65536 [6:04:05<5:22:15,  1.56it/s]
 54%|█████▍    | 35336/65536 [6:04:06<5:15:02,  1.60it/s]
 54%|█████▍    | 35337/65536 [6:04:06<5:07:46,  1.64it/s]
 54%|█████▍    | 35338/65536 [6:04:07<5:08:27,  1.63it/s]
 54%|█████▍    | 35339/65536 [6:04:08<5:14:53,  1.60it/s]
 54%|█████▍    | 35340/65536 [6:04:08<5:12:32,  1.61it/s]
                                                         
{'loss': 1.6452, 'learning_rate': 5.159483286396473e-07, 'epoch': 2181.48}
+
 54%|█████▍    | 35340/65536 [6:04:08<5:12:32,  1.61it/s]
 54%|█████▍    | 35341/65536 [6:04:09<5:13:49,  1.60it/s]
 54%|█████▍    | 35342/65536 [6:04:10<5:09:31,  1.63it/s]
 54%|█████▍    | 35343/65536 [6:04:10<5:11:14,  1.62it/s]
 54%|█████▍    | 35344/65536 [6:04:11<5:06:01,  1.64it/s]
 54%|█████▍    | 35345/65536 [6:04:11<4:58:38,  1.68it/s]
 54%|█████▍    | 35346/65536 [6:04:12<5:07:23,  1.64it/s]
 54%|█████▍    | 35347/65536 [6:04:13<5:07:20,  1.64it/s]
 54%|█████▍    | 35348/65536 [6:04:13<4:59:40,  1.68it/s]
 54%|█████▍    | 35349/65536 [6:04:14<5:07:56,  1.63it/s]
 54%|█████▍    | 35350/65536 [6:04:14<5:10:20,  1.62it/s]
 54%|█████▍    | 35351/65536 [6:04:15<5:10:20,  1.62it/s]
 54%|█████▍    | 35352/65536 [6:04:16<5:15:35,  1.59it/s]
 54%|█████▍    | 35353/65536 [6:04:16<5:12:48,  1.61it/s]
 54%|█████▍    | 35354/65536 [6:04:17<5:06:19,  1.64it/s]
 54%|█████▍    | 35355/65536 [6:04:18<5:27:03,  1.54it/s]
 54%|█████▍    | 35356/65536 [6:04:18<5:19:06,  1.58it/s]
 54%|█████▍    | 35357/65536 [6:04:19<5:10:18,  1.62it/s]
 54%|█████▍    | 35358/65536 [6:04:19<5:08:34,  1.63it/s]
 54%|█████▍    | 35359/65536 [6:04:20<5:09:36,  1.62it/s]
 54%|█████▍    | 35360/65536 [6:04:21<5:07:03,  1.64it/s]
                                                         
{'loss': 1.6083, 'learning_rate': 5.156728296804211e-07, 'epoch': 2182.72}
+
 54%|█████▍    | 35360/65536 [6:04:21<5:07:03,  1.64it/s]
 54%|█████▍    | 35361/65536 [6:04:21<5:04:14,  1.65it/s]
 54%|█████▍    | 35362/65536 [6:04:22<5:04:11,  1.65it/s]
 54%|█████▍    | 35363/65536 [6:04:22<4:59:52,  1.68it/s]
 54%|█████▍    | 35364/65536 [6:04:23<4:57:46,  1.69it/s]
 54%|█████▍    | 35365/65536 [6:04:24<5:12:00,  1.61it/s]
 54%|█████▍    | 35366/65536 [6:04:24<5:10:03,  1.62it/s]
 54%|█████▍    | 35367/65536 [6:04:25<5:19:52,  1.57it/s]
 54%|█████▍    | 35368/65536 [6:04:26<5:10:22,  1.62it/s]
 54%|█████▍    | 35369/65536 [6:04:26<5:16:23,  1.59it/s]
 54%|█████▍    | 35370/65536 [6:04:27<5:08:33,  1.63it/s]
 54%|█████▍    | 35371/65536 [6:04:27<5:03:52,  1.65it/s]
 54%|█████▍    | 35372/65536 [6:04:28<5:05:02,  1.65it/s]
 54%|█████▍    | 35373/65536 [6:04:29<5:07:06,  1.64it/s]
 54%|█████▍    | 35374/65536 [6:04:29<5:15:10,  1.59it/s]
 54%|█████▍    | 35375/65536 [6:04:30<5:15:01,  1.60it/s]
 54%|█████▍    | 35376/65536 [6:04:31<5:09:15,  1.63it/s]
 54%|█████▍    | 35377/65536 [6:04:31<5:10:15,  1.62it/s]
 54%|█████▍    | 35378/65536 [6:04:32<5:07:33,  1.63it/s]
 54%|█████▍    | 35379/65536 [6:04:32<5:02:42,  1.66it/s]
 54%|█████▍    | 35380/65536 [6:04:33<5:04:01,  1.65it/s]
                                                         
{'loss': 1.6284, 'learning_rate': 5.15397330721195e-07, 'epoch': 2183.95}
+
 54%|█████▍    | 35380/65536 [6:04:33<5:04:01,  1.65it/s]
 54%|█████▍    | 35381/65536 [6:04:34<5:10:28,  1.62it/s]
 54%|█████▍    | 35382/65536 [6:04:34<5:12:35,  1.61it/s]
 54%|█████▍    | 35383/65536 [6:04:35<5:10:55,  1.62it/s]
 54%|█████▍    | 35384/65536 [6:04:35<5:06:58,  1.64it/s]
 54%|█████▍    | 35385/65536 [6:04:36<5:13:15,  1.60it/s]
 54%|█████▍    | 35386/65536 [6:04:37<5:15:08,  1.59it/s]
 54%|█████▍    | 35387/65536 [6:04:37<5:29:07,  1.53it/s]
 54%|█████▍    | 35388/65536 [6:04:38<5:17:51,  1.58it/s]
 54%|█████▍    | 35389/65536 [6:04:39<5:12:25,  1.61it/s]
 54%|█████▍    | 35390/65536 [6:04:39<5:06:31,  1.64it/s]
 54%|█████▍    | 35391/65536 [6:04:40<5:07:10,  1.64it/s]
 54%|█████▍    | 35392/65536 [6:04:40<4:55:04,  1.70it/s]
 54%|█████▍    | 35393/65536 [6:04:41<4:55:08,  1.70it/s]
 54%|█████▍    | 35394/65536 [6:04:42<4:57:48,  1.69it/s]
 54%|█████▍    | 35395/65536 [6:04:42<4:58:08,  1.68it/s]
 54%|█████▍    | 35396/65536 [6:04:43<5:07:17,  1.63it/s]
 54%|█████▍    | 35397/65536 [6:04:43<5:10:16,  1.62it/s]
 54%|█████▍    | 35398/65536 [6:04:44<5:24:50,  1.55it/s]
 54%|█████▍    | 35399/65536 [6:04:45<5:16:36,  1.59it/s]
 54%|█████▍    | 35400/65536 [6:04:45<5:05:47,  1.64it/s]
                                                         
{'loss': 1.6395, 'learning_rate': 5.151218317619689e-07, 'epoch': 2185.19}
+
 54%|█████▍    | 35400/65536 [6:04:45<5:05:47,  1.64it/s]
 54%|█████▍    | 35401/65536 [6:04:46<5:09:57,  1.62it/s]
 54%|█████▍    | 35402/65536 [6:04:46<5:04:12,  1.65it/s]
 54%|█████▍    | 35403/65536 [6:04:47<5:00:00,  1.67it/s]
 54%|█████▍    | 35404/65536 [6:04:48<5:00:29,  1.67it/s]
 54%|█████▍    | 35405/65536 [6:04:48<4:58:21,  1.68it/s]
 54%|█████▍    | 35406/65536 [6:04:49<5:03:28,  1.65it/s]
 54%|█████▍    | 35407/65536 [6:04:49<5:04:55,  1.65it/s]
 54%|█████▍    | 35408/65536 [6:04:50<5:00:36,  1.67it/s]
 54%|█████▍    | 35409/65536 [6:04:51<5:00:23,  1.67it/s]
 54%|█████▍    | 35410/65536 [6:04:51<5:04:04,  1.65it/s]
 54%|█████▍    | 35411/65536 [6:04:52<4:57:25,  1.69it/s]
 54%|█████▍    | 35412/65536 [6:04:52<5:06:14,  1.64it/s]
 54%|█████▍    | 35413/65536 [6:04:53<5:15:57,  1.59it/s]
 54%|█████▍    | 35414/65536 [6:04:54<5:27:05,  1.53it/s]
 54%|█████▍    | 35415/65536 [6:04:54<5:21:04,  1.56it/s]
 54%|█████▍    | 35416/65536 [6:04:55<5:17:23,  1.58it/s]
 54%|█████▍    | 35417/65536 [6:04:56<5:08:08,  1.63it/s]
 54%|█████▍    | 35418/65536 [6:04:56<5:08:32,  1.63it/s]
 54%|█████▍    | 35419/65536 [6:04:57<5:07:15,  1.63it/s]
 54%|█████▍    | 35420/65536 [6:04:57<5:07:51,  1.63it/s]
                                                         
{'loss': 1.6433, 'learning_rate': 5.148463328027427e-07, 'epoch': 2186.42}
+
 54%|█████▍    | 35420/65536 [6:04:58<5:07:51,  1.63it/s]
 54%|█████▍    | 35421/65536 [6:04:58<5:12:48,  1.60it/s]
 54%|█████▍    | 35422/65536 [6:04:59<5:17:14,  1.58it/s]
 54%|█████▍    | 35423/65536 [6:04:59<5:09:54,  1.62it/s]
 54%|█████▍    | 35424/65536 [6:05:00<5:04:49,  1.65it/s]
 54%|█████▍    | 35425/65536 [6:05:01<5:06:28,  1.64it/s]
 54%|█████▍    | 35426/65536 [6:05:01<5:07:07,  1.63it/s]
 54%|█████▍    | 35427/65536 [6:05:02<5:11:17,  1.61it/s]
 54%|█████▍    | 35428/65536 [6:05:02<5:03:49,  1.65it/s]
 54%|█████▍    | 35429/65536 [6:05:03<5:06:03,  1.64it/s]
 54%|█████▍    | 35430/65536 [6:05:04<5:18:22,  1.58it/s]
 54%|█████▍    | 35431/65536 [6:05:04<5:17:56,  1.58it/s]
 54%|█████▍    | 35432/65536 [6:05:05<5:18:52,  1.57it/s]
 54%|█████▍    | 35433/65536 [6:05:06<5:20:31,  1.57it/s]
 54%|█████▍    | 35434/65536 [6:05:06<5:11:13,  1.61it/s]
 54%|█████▍    | 35435/65536 [6:05:07<5:01:01,  1.67it/s]
 54%|█████▍    | 35436/65536 [6:05:07<5:00:47,  1.67it/s]
 54%|█████▍    | 35437/65536 [6:05:08<4:58:00,  1.68it/s]
 54%|█████▍    | 35438/65536 [6:05:09<5:02:55,  1.66it/s]
 54%|█████▍    | 35439/65536 [6:05:09<5:00:33,  1.67it/s]
 54%|█████▍    | 35440/65536 [6:05:10<5:15:31,  1.59it/s]
                                                         
{'loss': 1.6108, 'learning_rate': 5.145708338435166e-07, 'epoch': 2187.65}
+
 54%|█████▍    | 35440/65536 [6:05:10<5:15:31,  1.59it/s]
 54%|█████▍    | 35441/65536 [6:05:10<5:12:25,  1.61it/s]
 54%|█████▍    | 35442/65536 [6:05:11<5:07:29,  1.63it/s]
 54%|█████▍    | 35443/65536 [6:05:12<5:06:02,  1.64it/s]
 54%|█████▍    | 35444/65536 [6:05:12<5:01:32,  1.66it/s]
 54%|█████▍    | 35445/65536 [6:05:13<5:04:03,  1.65it/s]
 54%|█████▍    | 35446/65536 [6:05:14<5:14:18,  1.60it/s]
 54%|█████▍    | 35447/65536 [6:05:14<5:07:55,  1.63it/s]
 54%|█████▍    | 35448/65536 [6:05:15<5:06:41,  1.64it/s]
 54%|█████▍    | 35449/65536 [6:05:15<5:02:32,  1.66it/s]
 54%|█████▍    | 35450/65536 [6:05:16<4:59:17,  1.68it/s]
 54%|█████▍    | 35451/65536 [6:05:16<5:00:20,  1.67it/s]
 54%|█████▍    | 35452/65536 [6:05:17<5:02:48,  1.66it/s]
 54%|█████▍    | 35453/65536 [6:05:18<5:04:48,  1.64it/s]
 54%|█████▍    | 35454/65536 [6:05:18<4:55:17,  1.70it/s]
 54%|█████▍    | 35455/65536 [6:05:19<5:12:51,  1.60it/s]
 54%|█████▍    | 35456/65536 [6:05:20<5:15:29,  1.59it/s]
 54%|█████▍    | 35457/65536 [6:05:20<5:11:11,  1.61it/s]
 54%|█████▍    | 35458/65536 [6:05:21<5:13:58,  1.60it/s]
 54%|█████▍    | 35459/65536 [6:05:21<5:11:05,  1.61it/s]
 54%|█████▍    | 35460/65536 [6:05:22<5:04:09,  1.65it/s]
                                                         
{'loss': 1.6232, 'learning_rate': 5.142953348842904e-07, 'epoch': 2188.89}
+
 54%|█████▍    | 35460/65536 [6:05:22<5:04:09,  1.65it/s]
 54%|█████▍    | 35461/65536 [6:05:23<5:06:37,  1.63it/s]
 54%|█████▍    | 35462/65536 [6:05:23<5:15:30,  1.59it/s]
 54%|█████▍    | 35463/65536 [6:05:24<5:10:17,  1.62it/s]
 54%|█████▍    | 35464/65536 [6:05:25<5:07:31,  1.63it/s]
 54%|█████▍    | 35465/65536 [6:05:25<5:07:54,  1.63it/s]
 54%|█████▍    | 35466/65536 [6:05:26<5:07:31,  1.63it/s]
 54%|█████▍    | 35467/65536 [6:05:26<5:06:53,  1.63it/s]
 54%|█████▍    | 35468/65536 [6:05:27<4:59:59,  1.67it/s]
 54%|█████▍    | 35469/65536 [6:05:28<5:02:08,  1.66it/s]
 54%|█████▍    | 35470/65536 [6:05:28<5:09:08,  1.62it/s]
 54%|█████▍    | 35471/65536 [6:05:29<5:05:31,  1.64it/s]
 54%|█████▍    | 35472/65536 [6:05:29<5:04:37,  1.64it/s]
 54%|█████▍    | 35473/65536 [6:05:30<4:59:56,  1.67it/s]
 54%|█████▍    | 35474/65536 [6:05:31<4:55:13,  1.70it/s]
 54%|█████▍    | 35475/65536 [6:05:31<4:56:50,  1.69it/s]
 54%|█████▍    | 35476/65536 [6:05:32<5:06:08,  1.64it/s]
 54%|█████▍    | 35477/65536 [6:05:33<5:22:11,  1.55it/s]
 54%|█████▍    | 35478/65536 [6:05:33<5:14:57,  1.59it/s]
 54%|█████▍    | 35479/65536 [6:05:34<5:25:38,  1.54it/s]
 54%|█████▍    | 35480/65536 [6:05:34<5:17:18,  1.58it/s]
                                                         
{'loss': 1.6295, 'learning_rate': 5.140198359250643e-07, 'epoch': 2190.12}
+
 54%|█████▍    | 35480/65536 [6:05:34<5:17:18,  1.58it/s]
 54%|█████▍    | 35481/65536 [6:05:35<5:19:52,  1.57it/s]
 54%|█████▍    | 35482/65536 [6:05:36<5:20:23,  1.56it/s]
 54%|█████▍    | 35483/65536 [6:05:36<5:15:13,  1.59it/s]
 54%|█████▍    | 35484/65536 [6:05:37<5:12:27,  1.60it/s]
 54%|█████▍    | 35485/65536 [6:05:38<5:24:15,  1.54it/s]
 54%|█████▍    | 35486/65536 [6:05:38<5:19:02,  1.57it/s]
 54%|█████▍    | 35487/65536 [6:05:39<5:17:38,  1.58it/s]
 54%|█████▍    | 35488/65536 [6:05:39<5:14:02,  1.59it/s]
 54%|█████▍    | 35489/65536 [6:05:40<5:08:05,  1.63it/s]
 54%|█████▍    | 35490/65536 [6:05:41<5:05:26,  1.64it/s]
 54%|█████▍    | 35491/65536 [6:05:41<4:59:49,  1.67it/s]
 54%|█████▍    | 35492/65536 [6:05:42<4:57:24,  1.68it/s]
 54%|█████▍    | 35493/65536 [6:05:42<4:56:29,  1.69it/s]
 54%|█████▍    | 35494/65536 [6:05:43<4:57:45,  1.68it/s]
 54%|█████▍    | 35495/65536 [6:05:44<5:16:45,  1.58it/s]
 54%|█████▍    | 35496/65536 [6:05:44<5:12:37,  1.60it/s]
 54%|█████▍    | 35497/65536 [6:05:45<5:01:12,  1.66it/s]
 54%|█████▍    | 35498/65536 [6:05:45<5:02:40,  1.65it/s]
 54%|█████▍    | 35499/65536 [6:05:46<5:02:47,  1.65it/s]
 54%|█████▍    | 35500/65536 [6:05:47<5:04:22,  1.64it/s]
                                                         
{'loss': 1.6395, 'learning_rate': 5.137443369658382e-07, 'epoch': 2191.36}
+
 54%|█████▍    | 35500/65536 [6:05:47<5:04:22,  1.64it/s]
 54%|█████▍    | 35501/65536 [6:05:47<5:04:01,  1.65it/s]
 54%|█████▍    | 35502/65536 [6:05:48<5:02:34,  1.65it/s]
 54%|█████▍    | 35503/65536 [6:05:48<4:59:04,  1.67it/s]
 54%|█████▍    | 35504/65536 [6:05:49<5:12:07,  1.60it/s]
 54%|█████▍    | 35505/65536 [6:05:50<5:09:30,  1.62it/s]
 54%|█████▍    | 35506/65536 [6:05:50<5:12:41,  1.60it/s]
 54%|█████▍    | 35507/65536 [6:05:51<5:11:28,  1.61it/s]
 54%|█████▍    | 35508/65536 [6:05:52<5:12:06,  1.60it/s]
 54%|█████▍    | 35509/65536 [6:05:52<5:13:04,  1.60it/s]
 54%|█████▍    | 35510/65536 [6:05:53<5:11:58,  1.60it/s]
 54%|█████▍    | 35511/65536 [6:05:54<5:22:03,  1.55it/s]
 54%|█████▍    | 35512/65536 [6:05:54<5:22:19,  1.55it/s]
 54%|█████▍    | 35513/65536 [6:05:55<5:12:34,  1.60it/s]
 54%|█████▍    | 35514/65536 [6:05:55<5:08:04,  1.62it/s]
 54%|█████▍    | 35515/65536 [6:05:56<4:59:39,  1.67it/s]
 54%|█████▍    | 35516/65536 [6:05:57<4:57:55,  1.68it/s]
 54%|█████▍    | 35517/65536 [6:05:57<5:00:15,  1.67it/s]
 54%|█████▍    | 35518/65536 [6:05:58<5:08:56,  1.62it/s]
 54%|█████▍    | 35519/65536 [6:05:58<5:04:52,  1.64it/s]
 54%|█████▍    | 35520/65536 [6:05:59<5:01:50,  1.66it/s]
                                                         
{'loss': 1.637, 'learning_rate': 5.13468838006612e-07, 'epoch': 2192.59}
+
 54%|█████▍    | 35520/65536 [6:05:59<5:01:50,  1.66it/s]
 54%|█████▍    | 35521/65536 [6:06:00<5:13:30,  1.60it/s]
 54%|█████▍    | 35522/65536 [6:06:00<5:15:53,  1.58it/s]
 54%|█████▍    | 35523/65536 [6:06:01<5:08:42,  1.62it/s]
 54%|█████▍    | 35524/65536 [6:06:02<5:06:05,  1.63it/s]
 54%|█████▍    | 35525/65536 [6:06:02<5:03:54,  1.65it/s]
 54%|█████▍    | 35526/65536 [6:06:03<5:03:13,  1.65it/s]
 54%|█████▍    | 35527/65536 [6:06:03<5:21:41,  1.55it/s]
 54%|█████▍    | 35528/65536 [6:06:04<5:13:27,  1.60it/s]
 54%|█████▍    | 35529/65536 [6:06:05<5:08:43,  1.62it/s]
 54%|█████▍    | 35530/65536 [6:06:05<5:06:09,  1.63it/s]
 54%|█████▍    | 35531/65536 [6:06:06<5:25:08,  1.54it/s]
 54%|█████▍    | 35532/65536 [6:06:07<5:16:31,  1.58it/s]
 54%|█████▍    | 35533/65536 [6:06:07<5:08:37,  1.62it/s]
 54%|█████▍    | 35534/65536 [6:06:08<5:00:14,  1.67it/s]
 54%|█████▍    | 35535/65536 [6:06:08<5:02:09,  1.65it/s]
 54%|█████▍    | 35536/65536 [6:06:09<5:05:18,  1.64it/s]
 54%|█████▍    | 35537/65536 [6:06:10<5:07:03,  1.63it/s]
 54%|█████▍    | 35538/65536 [6:06:10<5:09:48,  1.61it/s]
 54%|█████▍    | 35539/65536 [6:06:11<5:06:33,  1.63it/s]
 54%|█████▍    | 35540/65536 [6:06:11<5:02:40,  1.65it/s]
                                                         
{'loss': 1.6075, 'learning_rate': 5.131933390473858e-07, 'epoch': 2193.83}
+
 54%|█████▍    | 35540/65536 [6:06:11<5:02:40,  1.65it/s]
 54%|█████▍    | 35541/65536 [6:06:12<5:08:24,  1.62it/s]
 54%|█████▍    | 35542/65536 [6:06:13<5:08:25,  1.62it/s]
 54%|█████▍    | 35543/65536 [6:06:13<5:15:37,  1.58it/s]
 54%|█████▍    | 35544/65536 [6:06:14<5:06:17,  1.63it/s]
 54%|█████▍    | 35545/65536 [6:06:14<5:05:57,  1.63it/s]
 54%|█████▍    | 35546/65536 [6:06:15<5:12:53,  1.60it/s]
 54%|█████▍    | 35547/65536 [6:06:16<5:08:41,  1.62it/s]
 54%|█████▍    | 35548/65536 [6:06:16<5:14:59,  1.59it/s]
 54%|█████▍    | 35549/65536 [6:06:17<5:08:32,  1.62it/s]
 54%|█████▍    | 35550/65536 [6:06:18<5:04:02,  1.64it/s]
 54%|█████▍    | 35551/65536 [6:06:18<5:02:49,  1.65it/s]
 54%|█████▍    | 35552/65536 [6:06:19<5:07:07,  1.63it/s]
 54%|█████▍    | 35553/65536 [6:06:19<5:08:38,  1.62it/s]
 54%|█████▍    | 35554/65536 [6:06:20<5:17:22,  1.57it/s]
 54%|█████▍    | 35555/65536 [6:06:21<5:12:13,  1.60it/s]
 54%|█████▍    | 35556/65536 [6:06:21<5:14:40,  1.59it/s]
 54%|█████▍    | 35557/65536 [6:06:22<5:11:00,  1.61it/s]
 54%|█████▍    | 35558/65536 [6:06:23<5:04:43,  1.64it/s]
 54%|█████▍    | 35559/65536 [6:06:23<5:01:45,  1.66it/s]
 54%|█████▍    | 35560/65536 [6:06:24<5:07:12,  1.63it/s]
                                                         
{'loss': 1.6256, 'learning_rate': 5.129178400881597e-07, 'epoch': 2195.06}
+
 54%|█████▍    | 35560/65536 [6:06:24<5:07:12,  1.63it/s]
 54%|█████▍    | 35561/65536 [6:06:24<5:08:05,  1.62it/s]
 54%|█████▍    | 35562/65536 [6:06:25<5:01:19,  1.66it/s]
 54%|█████▍    | 35563/65536 [6:06:26<5:13:27,  1.59it/s]
 54%|█████▍    | 35564/65536 [6:06:26<5:16:44,  1.58it/s]
 54%|█████▍    | 35565/65536 [6:06:27<5:10:46,  1.61it/s]
 54%|█████▍    | 35566/65536 [6:06:28<5:08:00,  1.62it/s]
 54%|█████▍    | 35567/65536 [6:06:28<5:06:39,  1.63it/s]
 54%|█████▍    | 35568/65536 [6:06:29<4:59:52,  1.67it/s]
 54%|█████▍    | 35569/65536 [6:06:29<5:12:23,  1.60it/s]
 54%|█████▍    | 35570/65536 [6:06:30<5:15:38,  1.58it/s]
 54%|█████▍    | 35571/65536 [6:06:31<5:08:46,  1.62it/s]
 54%|█████▍    | 35572/65536 [6:06:31<5:08:28,  1.62it/s]
 54%|█████▍    | 35573/65536 [6:06:32<5:08:51,  1.62it/s]
 54%|█████▍    | 35574/65536 [6:06:32<5:02:08,  1.65it/s]
 54%|█████▍    | 35575/65536 [6:06:33<4:56:51,  1.68it/s]
 54%|█████▍    | 35576/65536 [6:06:34<5:06:07,  1.63it/s]
 54%|█████▍    | 35577/65536 [6:06:34<5:06:09,  1.63it/s]
 54%|█████▍    | 35578/65536 [6:06:35<5:12:20,  1.60it/s]
 54%|█████▍    | 35579/65536 [6:06:35<5:04:36,  1.64it/s]
 54%|█████▍    | 35580/65536 [6:06:36<5:09:15,  1.61it/s]
                                                         
{'loss': 1.6463, 'learning_rate': 5.126423411289334e-07, 'epoch': 2196.3}
+
 54%|█████▍    | 35580/65536 [6:06:36<5:09:15,  1.61it/s]
 54%|█████▍    | 35581/65536 [6:06:37<5:07:19,  1.62it/s]
 54%|█████▍    | 35582/65536 [6:06:37<5:12:19,  1.60it/s]
 54%|█████▍    | 35583/65536 [6:06:38<5:09:42,  1.61it/s]
 54%|█████▍    | 35584/65536 [6:06:39<5:03:01,  1.65it/s]
 54%|█████▍    | 35585/65536 [6:06:39<5:09:08,  1.61it/s]
 54%|█████▍    | 35586/65536 [6:06:40<5:14:39,  1.59it/s]
 54%|█████▍    | 35587/65536 [6:06:40<5:12:33,  1.60it/s]
 54%|█████▍    | 35588/65536 [6:06:41<5:10:16,  1.61it/s]
 54%|█████▍    | 35589/65536 [6:06:42<5:28:21,  1.52it/s]
 54%|█████▍    | 35590/65536 [6:06:42<5:20:58,  1.55it/s]
 54%|█████▍    | 35591/65536 [6:06:43<5:15:30,  1.58it/s]
 54%|█████▍    | 35592/65536 [6:06:44<5:22:30,  1.55it/s]
 54%|█████▍    | 35593/65536 [6:06:44<5:14:46,  1.59it/s]
 54%|█████▍    | 35594/65536 [6:06:45<5:09:58,  1.61it/s]
 54%|█████▍    | 35595/65536 [6:06:46<5:05:50,  1.63it/s]
 54%|█████▍    | 35596/65536 [6:06:46<5:11:37,  1.60it/s]
 54%|█████▍    | 35597/65536 [6:06:47<5:13:53,  1.59it/s]
 54%|█████▍    | 35598/65536 [6:06:47<5:13:45,  1.59it/s]
 54%|█████▍    | 35599/65536 [6:06:48<5:09:33,  1.61it/s]
 54%|█████▍    | 35600/65536 [6:06:49<5:05:07,  1.64it/s]
                                                         
{'loss': 1.6383, 'learning_rate': 5.123668421697072e-07, 'epoch': 2197.53}
+
 54%|█████▍    | 35600/65536 [6:06:49<5:05:07,  1.64it/s]
 54%|█████▍    | 35601/65536 [6:06:49<5:05:52,  1.63it/s]
 54%|█████▍    | 35602/65536 [6:06:50<5:07:17,  1.62it/s]
 54%|█████▍    | 35603/65536 [6:06:50<5:07:48,  1.62it/s]
 54%|█████▍    | 35604/65536 [6:06:51<5:04:42,  1.64it/s]
 54%|█████▍    | 35605/65536 [6:06:52<5:05:37,  1.63it/s]
 54%|█████▍    | 35606/65536 [6:06:52<5:02:28,  1.65it/s]
 54%|█████▍    | 35607/65536 [6:06:53<5:12:46,  1.59it/s]
 54%|█████▍    | 35608/65536 [6:06:54<5:18:57,  1.56it/s]
 54%|█████▍    | 35609/65536 [6:06:54<5:14:41,  1.59it/s]
 54%|█████▍    | 35610/65536 [6:06:55<5:15:24,  1.58it/s]
 54%|█████▍    | 35611/65536 [6:06:55<5:13:16,  1.59it/s]
 54%|█████▍    | 35612/65536 [6:06:56<5:24:07,  1.54it/s]
 54%|█████▍    | 35613/65536 [6:06:57<5:17:08,  1.57it/s]
 54%|█████▍    | 35614/65536 [6:06:57<5:13:46,  1.59it/s]
 54%|█████▍    | 35615/65536 [6:06:58<5:07:45,  1.62it/s]
 54%|█████▍    | 35616/65536 [6:06:59<5:05:59,  1.63it/s]
 54%|█████▍    | 35617/65536 [6:06:59<5:06:27,  1.63it/s]
 54%|█████▍    | 35618/65536 [6:07:00<5:05:12,  1.63it/s]
 54%|█████▍    | 35619/65536 [6:07:00<4:55:52,  1.69it/s]
 54%|█████▍    | 35620/65536 [6:07:01<4:56:45,  1.68it/s]
                                                         
{'loss': 1.6451, 'learning_rate': 5.120913432104811e-07, 'epoch': 2198.77}
+
 54%|█████▍    | 35620/65536 [6:07:01<4:56:45,  1.68it/s]
 54%|█████▍    | 35621/65536 [6:07:02<4:56:04,  1.68it/s]
 54%|█████▍    | 35622/65536 [6:07:02<5:03:29,  1.64it/s]
 54%|█████▍    | 35623/65536 [6:07:03<5:02:35,  1.65it/s]
 54%|█████▍    | 35624/65536 [6:07:04<5:14:51,  1.58it/s]
 54%|█████▍    | 35625/65536 [6:07:04<5:08:23,  1.62it/s]
 54%|█████▍    | 35626/65536 [6:07:05<5:06:27,  1.63it/s]
 54%|█████▍    | 35627/65536 [6:07:05<5:05:34,  1.63it/s]
 54%|█████▍    | 35628/65536 [6:07:06<4:58:16,  1.67it/s]
 54%|█████▍    | 35629/65536 [6:07:06<4:57:43,  1.67it/s]
 54%|█████▍    | 35630/65536 [6:07:07<4:56:50,  1.68it/s]
 54%|█████▍    | 35631/65536 [6:07:08<4:55:41,  1.69it/s]
 54%|█████▍    | 35632/65536 [6:07:08<4:57:08,  1.68it/s]
 54%|█████▍    | 35633/65536 [6:07:09<5:00:42,  1.66it/s]
 54%|█████▍    | 35634/65536 [6:07:10<5:04:58,  1.63it/s]
 54%|█████▍    | 35635/65536 [6:07:10<5:05:20,  1.63it/s]
 54%|█████▍    | 35636/65536 [6:07:11<5:06:26,  1.63it/s]
 54%|█████▍    | 35637/65536 [6:07:11<5:24:46,  1.53it/s]
 54%|█████▍    | 35638/65536 [6:07:12<5:13:52,  1.59it/s]
 54%|█████▍    | 35639/65536 [6:07:13<5:04:01,  1.64it/s]
 54%|█████▍    | 35640/65536 [6:07:13<5:00:23,  1.66it/s]
                                                         
{'loss': 1.6669, 'learning_rate': 5.118158442512551e-07, 'epoch': 2200.0}
+
 54%|█████▍    | 35640/65536 [6:07:13<5:00:23,  1.66it/s]
 54%|█████▍    | 35641/65536 [6:07:14<5:12:12,  1.60it/s]
 54%|█████▍    | 35642/65536 [6:07:14<5:07:37,  1.62it/s]
 54%|█████▍    | 35643/65536 [6:07:15<4:57:08,  1.68it/s]
 54%|█████▍    | 35644/65536 [6:07:16<5:02:59,  1.64it/s]
 54%|█████▍    | 35645/65536 [6:07:16<4:58:13,  1.67it/s]
 54%|█████▍    | 35646/65536 [6:07:17<5:03:51,  1.64it/s]
 54%|█████▍    | 35647/65536 [6:07:18<5:11:53,  1.60it/s]
 54%|█████▍    | 35648/65536 [6:07:18<5:12:19,  1.59it/s]
 54%|█████▍    | 35649/65536 [6:07:19<5:07:19,  1.62it/s]
 54%|█████▍    | 35650/65536 [6:07:19<5:08:43,  1.61it/s]
 54%|█████▍    | 35651/65536 [6:07:20<5:07:46,  1.62it/s]
 54%|█████▍    | 35652/65536 [6:07:21<5:08:18,  1.62it/s]
 54%|█████▍    | 35653/65536 [6:07:21<5:06:32,  1.62it/s]
 54%|█████▍    | 35654/65536 [6:07:22<5:15:15,  1.58it/s]
 54%|█████▍    | 35655/65536 [6:07:23<5:09:46,  1.61it/s]
 54%|█████▍    | 35656/65536 [6:07:23<5:03:00,  1.64it/s]
 54%|█████▍    | 35657/65536 [6:07:24<5:11:50,  1.60it/s]
 54%|█████▍    | 35658/65536 [6:07:24<5:05:18,  1.63it/s]
 54%|█████▍    | 35659/65536 [6:07:25<5:07:38,  1.62it/s]
 54%|█████▍    | 35660/65536 [6:07:26<5:06:22,  1.63it/s]
                                                         
{'loss': 1.6535, 'learning_rate': 5.115403452920288e-07, 'epoch': 2201.23}
+
 54%|█████▍    | 35660/65536 [6:07:26<5:06:22,  1.63it/s]
 54%|█████▍    | 35661/65536 [6:07:26<5:11:13,  1.60it/s]
 54%|█████▍    | 35662/65536 [6:07:27<5:13:36,  1.59it/s]
 54%|█████▍    | 35663/65536 [6:07:27<5:12:08,  1.60it/s]
 54%|█████▍    | 35664/65536 [6:07:28<5:16:20,  1.57it/s]
 54%|█████▍    | 35665/65536 [6:07:29<5:12:56,  1.59it/s]
 54%|█████▍    | 35666/65536 [6:07:29<5:08:11,  1.62it/s]
 54%|█████▍    | 35667/65536 [6:07:30<5:08:21,  1.61it/s]
 54%|█████▍    | 35668/65536 [6:07:31<5:07:39,  1.62it/s]
 54%|█████▍    | 35669/65536 [6:07:31<5:16:49,  1.57it/s]
 54%|█████▍    | 35670/65536 [6:07:32<5:08:34,  1.61it/s]
 54%|█████▍    | 35671/65536 [6:07:32<5:04:56,  1.63it/s]
 54%|█████▍    | 35672/65536 [6:07:33<4:57:34,  1.67it/s]
 54%|█████▍    | 35673/65536 [6:07:34<5:07:44,  1.62it/s]
 54%|█████▍    | 35674/65536 [6:07:34<5:10:24,  1.60it/s]
 54%|█████▍    | 35675/65536 [6:07:35<5:03:01,  1.64it/s]
 54%|█████▍    | 35676/65536 [6:07:36<5:06:23,  1.62it/s]
 54%|█████▍    | 35677/65536 [6:07:36<5:01:19,  1.65it/s]
 54%|█████▍    | 35678/65536 [6:07:37<4:53:51,  1.69it/s]
 54%|█████▍    | 35679/65536 [6:07:37<5:01:05,  1.65it/s]
 54%|█████▍    | 35680/65536 [6:07:38<5:01:53,  1.65it/s]
                                                         
{'loss': 1.623, 'learning_rate': 5.112648463328027e-07, 'epoch': 2202.47}
+
 54%|█████▍    | 35680/65536 [6:07:38<5:01:53,  1.65it/s]
 54%|█████▍    | 35681/65536 [6:07:38<4:57:32,  1.67it/s]
 54%|█████▍    | 35682/65536 [6:07:39<5:12:38,  1.59it/s]
 54%|█████▍    | 35683/65536 [6:07:40<5:21:34,  1.55it/s]
 54%|█████▍    | 35684/65536 [6:07:40<5:12:02,  1.59it/s]
 54%|█████▍    | 35685/65536 [6:07:41<5:09:24,  1.61it/s]
 54%|█████▍    | 35686/65536 [6:07:42<5:03:31,  1.64it/s]
 54%|█████▍    | 35687/65536 [6:07:42<4:59:08,  1.66it/s]
 54%|█████▍    | 35688/65536 [6:07:43<5:02:04,  1.65it/s]
 54%|█████▍    | 35689/65536 [6:07:44<5:16:57,  1.57it/s]
 54%|█████▍    | 35690/65536 [6:07:44<5:13:52,  1.58it/s]
 54%|█████▍    | 35691/65536 [6:07:45<5:07:56,  1.62it/s]
 54%|█████▍    | 35692/65536 [6:07:45<5:06:25,  1.62it/s]
 54%|█████▍    | 35693/65536 [6:07:46<5:13:26,  1.59it/s]
 54%|█████▍    | 35694/65536 [6:07:47<5:10:48,  1.60it/s]
 54%|█████▍    | 35695/65536 [6:07:47<5:05:53,  1.63it/s]
 54%|█████▍    | 35696/65536 [6:07:48<5:14:55,  1.58it/s]
 54%|█████▍    | 35697/65536 [6:07:48<5:06:54,  1.62it/s]
 54%|█████▍    | 35698/65536 [6:07:49<5:01:57,  1.65it/s]
 54%|█████▍    | 35699/65536 [6:07:50<5:08:17,  1.61it/s]
 54%|█████▍    | 35700/65536 [6:07:50<5:04:28,  1.63it/s]
                                                         
{'loss': 1.6265, 'learning_rate': 5.109893473735765e-07, 'epoch': 2203.7}
+
 54%|█████▍    | 35700/65536 [6:07:50<5:04:28,  1.63it/s]
 54%|█████▍    | 35701/65536 [6:07:51<5:06:42,  1.62it/s]
 54%|█████▍    | 35702/65536 [6:07:52<5:06:14,  1.62it/s]
 54%|█████▍    | 35703/65536 [6:07:52<5:02:40,  1.64it/s]
 54%|█████▍    | 35704/65536 [6:07:53<5:00:53,  1.65it/s]
 54%|█████▍    | 35705/65536 [6:07:53<5:11:26,  1.60it/s]
 54%|█████▍    | 35706/65536 [6:07:54<5:12:27,  1.59it/s]
 54%|█████▍    | 35707/65536 [6:07:55<5:12:38,  1.59it/s]
 54%|█████▍    | 35708/65536 [6:07:55<5:14:17,  1.58it/s]
 54%|█████▍    | 35709/65536 [6:07:56<5:10:36,  1.60it/s]
 54%|█████▍    | 35710/65536 [6:07:57<5:12:36,  1.59it/s]
 54%|█████▍    | 35711/65536 [6:07:57<5:02:00,  1.65it/s]
 54%|█████▍    | 35712/65536 [6:07:58<4:57:29,  1.67it/s]
 54%|█████▍    | 35713/65536 [6:07:58<4:57:02,  1.67it/s]
 54%|█████▍    | 35714/65536 [6:07:59<5:05:06,  1.63it/s]
 54%|█████▍    | 35715/65536 [6:08:00<4:59:48,  1.66it/s]
 54%|█████▍    | 35716/65536 [6:08:00<5:00:53,  1.65it/s]
 54%|█████▍    | 35717/65536 [6:08:01<4:56:09,  1.68it/s]
 55%|█████▍    | 35718/65536 [6:08:01<5:04:52,  1.63it/s]
 55%|█████▍    | 35719/65536 [6:08:02<5:09:30,  1.61it/s]
 55%|█████▍    | 35720/65536 [6:08:03<5:04:40,  1.63it/s]
                                                         
{'loss': 1.6418, 'learning_rate': 5.107138484143504e-07, 'epoch': 2204.94}
+
 55%|█████▍    | 35720/65536 [6:08:03<5:04:40,  1.63it/s]
 55%|█████▍    | 35721/65536 [6:08:03<5:09:43,  1.60it/s]
 55%|█████▍    | 35722/65536 [6:08:04<5:31:04,  1.50it/s]
 55%|█████▍    | 35723/65536 [6:08:05<5:19:49,  1.55it/s]
 55%|█████▍    | 35724/65536 [6:08:05<5:09:55,  1.60it/s]
 55%|█████▍    | 35725/65536 [6:08:06<5:14:08,  1.58it/s]
 55%|█████▍    | 35726/65536 [6:08:06<5:12:32,  1.59it/s]
 55%|█████▍    | 35727/65536 [6:08:07<5:09:00,  1.61it/s]
 55%|█████▍    | 35728/65536 [6:08:08<5:05:27,  1.63it/s]
 55%|█████▍    | 35729/65536 [6:08:08<5:01:40,  1.65it/s]
 55%|█████▍    | 35730/65536 [6:08:09<4:58:34,  1.66it/s]
 55%|█████▍    | 35731/65536 [6:08:09<4:50:36,  1.71it/s]
 55%|█████▍    | 35732/65536 [6:08:10<4:57:16,  1.67it/s]
 55%|█████▍    | 35733/65536 [6:08:11<5:02:52,  1.64it/s]
 55%|█████▍    | 35734/65536 [6:08:11<5:03:00,  1.64it/s]
 55%|█████▍    | 35735/65536 [6:08:12<5:05:09,  1.63it/s]
 55%|█████▍    | 35736/65536 [6:08:13<5:09:54,  1.60it/s]
 55%|█████▍    | 35737/65536 [6:08:13<5:05:11,  1.63it/s]
 55%|█████▍    | 35738/65536 [6:08:14<5:19:34,  1.55it/s]
 55%|█████▍    | 35739/65536 [6:08:14<5:14:43,  1.58it/s]
 55%|█████▍    | 35740/65536 [6:08:15<5:08:16,  1.61it/s]
                                                         
{'loss': 1.6721, 'learning_rate': 5.104383494551243e-07, 'epoch': 2206.17}
+
 55%|█████▍    | 35740/65536 [6:08:15<5:08:16,  1.61it/s]
 55%|█████▍    | 35741/65536 [6:08:16<5:08:36,  1.61it/s]
 55%|█████▍    | 35742/65536 [6:08:16<5:03:06,  1.64it/s]
 55%|█████▍    | 35743/65536 [6:08:17<5:05:47,  1.62it/s]
 55%|█████▍    | 35744/65536 [6:08:17<5:04:13,  1.63it/s]
 55%|█████▍    | 35745/65536 [6:08:18<4:57:58,  1.67it/s]
 55%|█████▍    | 35746/65536 [6:08:19<5:05:26,  1.63it/s]
 55%|█████▍    | 35747/65536 [6:08:19<5:09:22,  1.60it/s]
 55%|█████▍    | 35748/65536 [6:08:20<5:17:56,  1.56it/s]
 55%|█████▍    | 35749/65536 [6:08:21<5:08:35,  1.61it/s]
 55%|█████▍    | 35750/65536 [6:08:21<5:04:53,  1.63it/s]
 55%|█████▍    | 35751/65536 [6:08:22<5:04:46,  1.63it/s]
 55%|█████▍    | 35752/65536 [6:08:22<5:01:17,  1.65it/s]
 55%|█████▍    | 35753/65536 [6:08:23<4:55:11,  1.68it/s]
 55%|█████▍    | 35754/65536 [6:08:24<5:05:39,  1.62it/s]
 55%|█████▍    | 35755/65536 [6:08:24<5:00:37,  1.65it/s]
 55%|█████▍    | 35756/65536 [6:08:25<4:59:36,  1.66it/s]
 55%|█████▍    | 35757/65536 [6:08:25<5:06:42,  1.62it/s]
 55%|█████▍    | 35758/65536 [6:08:26<5:05:54,  1.62it/s]
 55%|█████▍    | 35759/65536 [6:08:27<5:08:23,  1.61it/s]
 55%|█████▍    | 35760/65536 [6:08:27<5:06:32,  1.62it/s]
                                                         
{'loss': 1.6456, 'learning_rate': 5.101628504958981e-07, 'epoch': 2207.41}
+
 55%|█████▍    | 35760/65536 [6:08:27<5:06:32,  1.62it/s]
 55%|█████▍    | 35761/65536 [6:08:28<5:05:15,  1.63it/s]
 55%|█████▍    | 35762/65536 [6:08:29<5:11:52,  1.59it/s]
 55%|█████▍    | 35763/65536 [6:08:29<5:12:15,  1.59it/s]
 55%|█████▍    | 35764/65536 [6:08:30<5:08:44,  1.61it/s]
 55%|█████▍    | 35765/65536 [6:08:30<5:05:22,  1.62it/s]
 55%|█████▍    | 35766/65536 [6:08:31<4:55:08,  1.68it/s]
 55%|█████▍    | 35767/65536 [6:08:32<4:54:22,  1.69it/s]
 55%|█████▍    | 35768/65536 [6:08:32<4:55:25,  1.68it/s]
 55%|█████▍    | 35769/65536 [6:08:33<4:53:56,  1.69it/s]
 55%|█████▍    | 35770/65536 [6:08:33<5:05:36,  1.62it/s]
 55%|█████▍    | 35771/65536 [6:08:34<5:01:43,  1.64it/s]
 55%|█████▍    | 35772/65536 [6:08:35<4:59:08,  1.66it/s]
 55%|█████▍    | 35773/65536 [6:08:35<4:58:47,  1.66it/s]
 55%|█████▍    | 35774/65536 [6:08:36<5:00:45,  1.65it/s]
 55%|█████▍    | 35775/65536 [6:08:36<4:59:27,  1.66it/s]
 55%|█████▍    | 35776/65536 [6:08:37<5:01:00,  1.65it/s]
 55%|█████▍    | 35777/65536 [6:08:38<4:59:25,  1.66it/s]
 55%|█████▍    | 35778/65536 [6:08:38<4:56:01,  1.68it/s]
 55%|█████▍    | 35779/65536 [6:08:39<5:00:46,  1.65it/s]
 55%|█████▍    | 35780/65536 [6:08:39<5:02:51,  1.64it/s]
                                                         
{'loss': 1.6195, 'learning_rate': 5.09887351536672e-07, 'epoch': 2208.64}
+
 55%|█████▍    | 35780/65536 [6:08:39<5:02:51,  1.64it/s]
 55%|█████▍    | 35781/65536 [6:08:40<5:06:53,  1.62it/s]
 55%|█████▍    | 35782/65536 [6:08:41<5:15:25,  1.57it/s]
 55%|█████▍    | 35783/65536 [6:08:41<5:09:14,  1.60it/s]
 55%|█████▍    | 35784/65536 [6:08:42<5:18:13,  1.56it/s]
 55%|█████▍    | 35785/65536 [6:08:43<5:16:55,  1.56it/s]
 55%|█████▍    | 35786/65536 [6:08:43<5:15:12,  1.57it/s]
 55%|█████▍    | 35787/65536 [6:08:44<5:10:56,  1.59it/s]
 55%|█████▍    | 35788/65536 [6:08:44<5:04:57,  1.63it/s]
 55%|█████▍    | 35789/65536 [6:08:45<5:01:56,  1.64it/s]
 55%|█████▍    | 35790/65536 [6:08:46<5:04:04,  1.63it/s]
 55%|█████▍    | 35791/65536 [6:08:46<5:04:35,  1.63it/s]
 55%|█████▍    | 35792/65536 [6:08:47<5:03:52,  1.63it/s]
 55%|█████▍    | 35793/65536 [6:08:48<5:00:07,  1.65it/s]
 55%|█████▍    | 35794/65536 [6:08:48<5:00:20,  1.65it/s]
 55%|█████▍    | 35795/65536 [6:08:49<5:15:27,  1.57it/s]
 55%|█████▍    | 35796/65536 [6:08:49<5:08:14,  1.61it/s]
 55%|█████▍    | 35797/65536 [6:08:50<5:02:03,  1.64it/s]
 55%|█████▍    | 35798/65536 [6:08:51<4:55:17,  1.68it/s]
 55%|█████▍    | 35799/65536 [6:08:51<5:01:03,  1.65it/s]
 55%|█████▍    | 35800/65536 [6:08:52<5:00:58,  1.65it/s]
                                                         
{'loss': 1.6122, 'learning_rate': 5.096118525774458e-07, 'epoch': 2209.88}
+
 55%|█████▍    | 35800/65536 [6:08:52<5:00:58,  1.65it/s]
 55%|█████▍    | 35801/65536 [6:08:52<5:01:36,  1.64it/s]
 55%|█████▍    | 35802/65536 [6:08:53<5:04:22,  1.63it/s]
 55%|█████▍    | 35803/65536 [6:08:54<5:13:13,  1.58it/s]
 55%|█████▍    | 35804/65536 [6:08:54<5:13:09,  1.58it/s]
 55%|█████▍    | 35805/65536 [6:08:55<5:09:01,  1.60it/s]
 55%|█████▍    | 35806/65536 [6:08:56<5:11:54,  1.59it/s]
 55%|█████▍    | 35807/65536 [6:08:56<5:16:07,  1.57it/s]
 55%|█████▍    | 35808/65536 [6:08:57<5:21:03,  1.54it/s]
 55%|█████▍    | 35809/65536 [6:08:58<5:14:00,  1.58it/s]
 55%|█████▍    | 35810/65536 [6:08:58<5:07:38,  1.61it/s]
 55%|█████▍    | 35811/65536 [6:08:59<5:10:36,  1.60it/s]
 55%|█████▍    | 35812/65536 [6:08:59<5:03:54,  1.63it/s]
 55%|█████▍    | 35813/65536 [6:09:00<4:59:56,  1.65it/s]
 55%|█████▍    | 35814/65536 [6:09:01<4:54:35,  1.68it/s]
 55%|█████▍    | 35815/65536 [6:09:01<4:58:41,  1.66it/s]
 55%|█████▍    | 35816/65536 [6:09:02<4:57:52,  1.66it/s]
 55%|█████▍    | 35817/65536 [6:09:02<4:53:39,  1.69it/s]
 55%|█████▍    | 35818/65536 [6:09:03<4:55:35,  1.68it/s]
 55%|█████▍    | 35819/65536 [6:09:04<5:07:29,  1.61it/s]
 55%|█████▍    | 35820/65536 [6:09:04<5:03:42,  1.63it/s]
                                                         
{'loss': 1.6416, 'learning_rate': 5.093363536182197e-07, 'epoch': 2211.11}
+
 55%|█████▍    | 35820/65536 [6:09:04<5:03:42,  1.63it/s]
 55%|█████▍    | 35821/65536 [6:09:05<5:00:16,  1.65it/s]
 55%|█████▍    | 35822/65536 [6:09:05<5:03:30,  1.63it/s]
 55%|█████▍    | 35823/65536 [6:09:06<5:13:07,  1.58it/s]
 55%|█████▍    | 35824/65536 [6:09:07<5:08:27,  1.61it/s]
 55%|█████▍    | 35825/65536 [6:09:07<4:59:07,  1.66it/s]
 55%|█████▍    | 35826/65536 [6:09:08<5:00:32,  1.65it/s]
 55%|█████▍    | 35827/65536 [6:09:08<5:00:53,  1.65it/s]
 55%|█████▍    | 35828/65536 [6:09:09<4:57:01,  1.67it/s]
 55%|█████▍    | 35829/65536 [6:09:10<5:02:54,  1.63it/s]
 55%|█████▍    | 35830/65536 [6:09:10<5:04:51,  1.62it/s]
 55%|█████▍    | 35831/65536 [6:09:11<5:05:35,  1.62it/s]
 55%|█████▍    | 35832/65536 [6:09:12<5:12:40,  1.58it/s]
 55%|█████▍    | 35833/65536 [6:09:12<5:09:25,  1.60it/s]
 55%|█████▍    | 35834/65536 [6:09:13<5:09:02,  1.60it/s]
 55%|█████▍    | 35835/65536 [6:09:13<5:14:29,  1.57it/s]
 55%|█████▍    | 35836/65536 [6:09:14<5:12:51,  1.58it/s]
 55%|█████▍    | 35837/65536 [6:09:15<5:06:10,  1.62it/s]
 55%|█████▍    | 35838/65536 [6:09:15<5:15:41,  1.57it/s]
 55%|█████▍    | 35839/65536 [6:09:16<5:10:38,  1.59it/s]
 55%|█████▍    | 35840/65536 [6:09:17<5:05:07,  1.62it/s]
                                                         
{'loss': 1.6261, 'learning_rate': 5.090608546589935e-07, 'epoch': 2212.35}
+
 55%|█████▍    | 35840/65536 [6:09:17<5:05:07,  1.62it/s]
 55%|█████▍    | 35841/65536 [6:09:17<5:04:28,  1.63it/s]
 55%|█████▍    | 35842/65536 [6:09:18<5:05:33,  1.62it/s]
 55%|█████▍    | 35843/65536 [6:09:18<4:58:20,  1.66it/s]
 55%|█████▍    | 35844/65536 [6:09:19<4:59:10,  1.65it/s]
 55%|█████▍    | 35845/65536 [6:09:20<4:58:57,  1.66it/s]
 55%|█████▍    | 35846/65536 [6:09:20<5:12:00,  1.59it/s]
 55%|█████▍    | 35847/65536 [6:09:21<5:05:34,  1.62it/s]
 55%|█████▍    | 35848/65536 [6:09:22<5:08:16,  1.61it/s]
 55%|█████▍    | 35849/65536 [6:09:22<4:58:56,  1.66it/s]
 55%|█████▍    | 35850/65536 [6:09:23<5:05:00,  1.62it/s]
 55%|█████▍    | 35851/65536 [6:09:23<5:08:07,  1.61it/s]
 55%|█████▍    | 35852/65536 [6:09:24<5:13:52,  1.58it/s]
 55%|█████▍    | 35853/65536 [6:09:25<5:11:55,  1.59it/s]
 55%|█████▍    | 35854/65536 [6:09:25<5:01:59,  1.64it/s]
 55%|█████▍    | 35855/65536 [6:09:26<5:05:37,  1.62it/s]
 55%|█████▍    | 35856/65536 [6:09:26<5:09:26,  1.60it/s]
 55%|█████▍    | 35857/65536 [6:09:27<5:03:34,  1.63it/s]
 55%|█████▍    | 35858/65536 [6:09:28<4:59:46,  1.65it/s]
 55%|█████▍    | 35859/65536 [6:09:28<4:59:57,  1.65it/s]
 55%|█████▍    | 35860/65536 [6:09:29<4:57:22,  1.66it/s]
                                                         
{'loss': 1.6247, 'learning_rate': 5.087853556997673e-07, 'epoch': 2213.58}
+
 55%|█████▍    | 35860/65536 [6:09:29<4:57:22,  1.66it/s]
 55%|█████▍    | 35861/65536 [6:09:29<4:51:49,  1.69it/s]
 55%|█████▍    | 35862/65536 [6:09:30<4:55:12,  1.68it/s]
 55%|█████▍    | 35863/65536 [6:09:31<4:57:12,  1.66it/s]
 55%|█████▍    | 35864/65536 [6:09:31<5:04:37,  1.62it/s]
 55%|█████▍    | 35865/65536 [6:09:32<4:57:50,  1.66it/s]
 55%|█████▍    | 35866/65536 [6:09:32<4:56:19,  1.67it/s]
 55%|█████▍    | 35867/65536 [6:09:33<5:02:51,  1.63it/s]
 55%|█████▍    | 35868/65536 [6:09:34<5:00:29,  1.65it/s]
 55%|█████▍    | 35869/65536 [6:09:34<4:56:35,  1.67it/s]
 55%|█████▍    | 35870/65536 [6:09:35<4:56:37,  1.67it/s]
 55%|█████▍    | 35871/65536 [6:09:35<5:00:02,  1.65it/s]
 55%|█████▍    | 35872/65536 [6:09:36<5:08:07,  1.60it/s]
 55%|█████▍    | 35873/65536 [6:09:37<5:11:44,  1.59it/s]
 55%|█████▍    | 35874/65536 [6:09:37<5:10:58,  1.59it/s]
 55%|█████▍    | 35875/65536 [6:09:38<5:07:36,  1.61it/s]
 55%|█████▍    | 35876/65536 [6:09:39<5:04:42,  1.62it/s]
 55%|█████▍    | 35877/65536 [6:09:39<5:06:15,  1.61it/s]
 55%|█████▍    | 35878/65536 [6:09:40<5:09:04,  1.60it/s]
 55%|█████▍    | 35879/65536 [6:09:41<5:07:28,  1.61it/s]
 55%|█████▍    | 35880/65536 [6:09:41<5:02:59,  1.63it/s]
                                                         
{'loss': 1.6314, 'learning_rate': 5.085098567405412e-07, 'epoch': 2214.81}
+
 55%|█████▍    | 35880/65536 [6:09:41<5:02:59,  1.63it/s]
 55%|█████▍    | 35881/65536 [6:09:42<4:58:52,  1.65it/s]
 55%|█████▍    | 35882/65536 [6:09:42<4:58:55,  1.65it/s]
 55%|█████▍    | 35883/65536 [6:09:43<5:04:42,  1.62it/s]
 55%|█████▍    | 35884/65536 [6:09:44<5:12:31,  1.58it/s]
 55%|█████▍    | 35885/65536 [6:09:44<5:10:22,  1.59it/s]
 55%|█████▍    | 35886/65536 [6:09:45<5:11:51,  1.58it/s]
 55%|█████▍    | 35887/65536 [6:09:46<5:12:32,  1.58it/s]
 55%|█████▍    | 35888/65536 [6:09:46<5:04:55,  1.62it/s]
 55%|█████▍    | 35889/65536 [6:09:47<4:55:30,  1.67it/s]
 55%|█████▍    | 35890/65536 [6:09:47<4:58:39,  1.65it/s]
 55%|█████▍    | 35891/65536 [6:09:48<4:57:40,  1.66it/s]
 55%|█████▍    | 35892/65536 [6:09:48<4:55:13,  1.67it/s]
 55%|█████▍    | 35893/65536 [6:09:49<4:55:23,  1.67it/s]
 55%|█████▍    | 35894/65536 [6:09:50<4:55:50,  1.67it/s]
 55%|█████▍    | 35895/65536 [6:09:50<5:02:26,  1.63it/s]
 55%|█████▍    | 35896/65536 [6:09:51<5:00:16,  1.65it/s]
 55%|█████▍    | 35897/65536 [6:09:52<5:03:44,  1.63it/s]
 55%|█████▍    | 35898/65536 [6:09:52<4:59:35,  1.65it/s]
 55%|���████▍    | 35899/65536 [6:09:53<5:05:13,  1.62it/s]
 55%|█████▍    | 35900/65536 [6:09:53<5:14:27,  1.57it/s]
                                                         
{'loss': 1.6695, 'learning_rate': 5.082343577813151e-07, 'epoch': 2216.05}
+
 55%|█████▍    | 35900/65536 [6:09:53<5:14:27,  1.57it/s]
 55%|█████▍    | 35901/65536 [6:09:54<5:08:04,  1.60it/s]
 55%|█████▍    | 35902/65536 [6:09:55<5:05:36,  1.62it/s]
 55%|█████▍    | 35903/65536 [6:09:55<5:18:41,  1.55it/s]
 55%|█████▍    | 35904/65536 [6:09:56<5:15:32,  1.57it/s]
 55%|█████▍    | 35905/65536 [6:09:57<5:10:59,  1.59it/s]
 55%|█████▍    | 35906/65536 [6:09:57<5:07:15,  1.61it/s]
 55%|█████▍    | 35907/65536 [6:09:58<5:03:17,  1.63it/s]
 55%|█████▍    | 35908/65536 [6:09:58<4:58:53,  1.65it/s]
 55%|█████▍    | 35909/65536 [6:09:59<4:50:46,  1.70it/s]
 55%|█████▍    | 35910/65536 [6:10:00<4:58:12,  1.66it/s]
 55%|█████▍    | 35911/65536 [6:10:00<4:59:04,  1.65it/s]
 55%|█████▍    | 35912/65536 [6:10:01<4:51:44,  1.69it/s]
 55%|█████▍    | 35913/65536 [6:10:01<4:58:44,  1.65it/s]
 55%|█████▍    | 35914/65536 [6:10:02<4:52:03,  1.69it/s]
 55%|█████▍    | 35915/65536 [6:10:03<4:56:34,  1.66it/s]
 55%|█████▍    | 35916/65536 [6:10:03<5:03:05,  1.63it/s]
 55%|█████▍    | 35917/65536 [6:10:04<5:00:56,  1.64it/s]
 55%|█████▍    | 35918/65536 [6:10:04<4:56:39,  1.66it/s]
 55%|█████▍    | 35919/65536 [6:10:05<5:08:37,  1.60it/s]
 55%|█████▍    | 35920/65536 [6:10:06<5:08:34,  1.60it/s]
                                                         
{'loss': 1.7002, 'learning_rate': 5.079588588220889e-07, 'epoch': 2217.28}
+
 55%|█████▍    | 35920/65536 [6:10:06<5:08:34,  1.60it/s]
 55%|█████▍    | 35921/65536 [6:10:06<5:04:14,  1.62it/s]
 55%|█████▍    | 35922/65536 [6:10:07<5:11:06,  1.59it/s]
 55%|█████▍    | 35923/65536 [6:10:07<5:02:01,  1.63it/s]
 55%|█████▍    | 35924/65536 [6:10:08<5:11:35,  1.58it/s]
 55%|█████▍    | 35925/65536 [6:10:09<5:10:23,  1.59it/s]
 55%|█████▍    | 35926/65536 [6:10:09<5:11:50,  1.58it/s]
 55%|█████▍    | 35927/65536 [6:10:10<5:01:52,  1.63it/s]
 55%|█████▍    | 35928/65536 [6:10:11<5:01:14,  1.64it/s]
 55%|█████▍    | 35929/65536 [6:10:11<4:55:17,  1.67it/s]
 55%|█████▍    | 35930/65536 [6:10:12<4:58:14,  1.65it/s]
 55%|█████▍    | 35931/65536 [6:10:12<4:53:41,  1.68it/s]
 55%|█████▍    | 35932/65536 [6:10:13<4:57:57,  1.66it/s]
 55%|█████▍    | 35933/65536 [6:10:14<4:58:27,  1.65it/s]
 55%|█████▍    | 35934/65536 [6:10:14<5:05:20,  1.62it/s]
 55%|█████▍    | 35935/65536 [6:10:15<5:08:46,  1.60it/s]
 55%|█████▍    | 35936/65536 [6:10:16<5:25:04,  1.52it/s]
 55%|█████▍    | 35937/65536 [6:10:16<5:18:47,  1.55it/s]
 55%|█████▍    | 35938/65536 [6:10:17<5:16:25,  1.56it/s]
 55%|█████▍    | 35939/65536 [6:10:17<5:06:18,  1.61it/s]
 55%|█████▍    | 35940/65536 [6:10:18<5:09:27,  1.59it/s]
                                                         
{'loss': 1.6342, 'learning_rate': 5.076833598628627e-07, 'epoch': 2218.52}
+
 55%|█████▍    | 35940/65536 [6:10:18<5:09:27,  1.59it/s]
 55%|█████▍    | 35941/65536 [6:10:19<5:02:53,  1.63it/s]
 55%|█████▍    | 35942/65536 [6:10:19<4:55:43,  1.67it/s]
 55%|█████▍    | 35943/65536 [6:10:20<5:05:39,  1.61it/s]
 55%|█████▍    | 35944/65536 [6:10:20<5:00:21,  1.64it/s]
 55%|█████▍    | 35945/65536 [6:10:21<4:57:00,  1.66it/s]
 55%|█████▍    | 35946/65536 [6:10:22<4:56:40,  1.66it/s]
 55%|█████▍    | 35947/65536 [6:10:22<4:56:10,  1.67it/s]
 55%|█████▍    | 35948/65536 [6:10:23<5:07:25,  1.60it/s]
 55%|█████▍    | 35949/65536 [6:10:24<5:08:25,  1.60it/s]
 55%|█████▍    | 35950/65536 [6:10:24<5:03:51,  1.62it/s]
 55%|█████▍    | 35951/65536 [6:10:25<5:02:33,  1.63it/s]
 55%|█████▍    | 35952/65536 [6:10:25<5:07:08,  1.61it/s]
 55%|█████▍    | 35953/65536 [6:10:26<5:07:37,  1.60it/s]
 55%|█████▍    | 35954/65536 [6:10:27<5:11:03,  1.59it/s]
 55%|█████▍    | 35955/65536 [6:10:27<5:10:23,  1.59it/s]
 55%|█████▍    | 35956/65536 [6:10:28<5:08:00,  1.60it/s]
 55%|█████▍    | 35957/65536 [6:10:29<5:01:45,  1.63it/s]
 55%|█████▍    | 35958/65536 [6:10:29<5:01:19,  1.64it/s]
 55%|█████▍    | 35959/65536 [6:10:30<5:02:09,  1.63it/s]
 55%|█████▍    | 35960/65536 [6:10:30<4:53:20,  1.68it/s]
                                                         
{'loss': 1.6208, 'learning_rate': 5.074078609036366e-07, 'epoch': 2219.75}
+
 55%|█████▍    | 35960/65536 [6:10:30<4:53:20,  1.68it/s]
 55%|█████▍    | 35961/65536 [6:10:31<5:03:04,  1.63it/s]
 55%|█████▍    | 35962/65536 [6:10:32<4:58:54,  1.65it/s]
 55%|█████▍    | 35963/65536 [6:10:32<4:53:40,  1.68it/s]
 55%|█████▍    | 35964/65536 [6:10:33<4:56:11,  1.66it/s]
 55%|█████▍    | 35965/65536 [6:10:33<4:59:29,  1.65it/s]
 55%|█████▍    | 35966/65536 [6:10:34<5:08:36,  1.60it/s]
 55%|█████▍    | 35967/65536 [6:10:35<5:05:21,  1.61it/s]
 55%|█████▍    | 35968/65536 [6:10:35<5:02:59,  1.63it/s]
 55%|█████▍    | 35969/65536 [6:10:36<5:04:58,  1.62it/s]
 55%|█████▍    | 35970/65536 [6:10:36<5:00:03,  1.64it/s]
 55%|█████▍    | 35971/65536 [6:10:37<4:53:17,  1.68it/s]
 55%|█████▍    | 35972/65536 [6:10:38<5:04:29,  1.62it/s]
 55%|█████▍    | 35973/65536 [6:10:38<4:56:12,  1.66it/s]
 55%|█████▍    | 35974/65536 [6:10:39<5:00:24,  1.64it/s]
 55%|█████▍    | 35975/65536 [6:10:39<4:59:18,  1.65it/s]
 55%|█████▍    | 35976/65536 [6:10:40<5:00:39,  1.64it/s]
 55%|█████▍    | 35977/65536 [6:10:41<4:58:09,  1.65it/s]
 55%|█████▍    | 35978/65536 [6:10:41<4:56:24,  1.66it/s]
 55%|█████▍    | 35979/65536 [6:10:42<4:53:22,  1.68it/s]
 55%|█████▍    | 35980/65536 [6:10:43<5:01:22,  1.63it/s]
                                                         
{'loss': 1.648, 'learning_rate': 5.071323619444105e-07, 'epoch': 2220.99}
+
 55%|█████▍    | 35980/65536 [6:10:43<5:01:22,  1.63it/s]
 55%|█████▍    | 35981/65536 [6:10:43<5:12:55,  1.57it/s]
 55%|█████▍    | 35982/65536 [6:10:44<5:04:26,  1.62it/s]
 55%|█████▍    | 35983/65536 [6:10:44<5:00:37,  1.64it/s]
 55%|█████▍    | 35984/65536 [6:10:45<4:57:52,  1.65it/s]
 55%|█████▍    | 35985/65536 [6:10:46<5:04:43,  1.62it/s]
 55%|█████▍    | 35986/65536 [6:10:46<4:59:25,  1.64it/s]
 55%|█████▍    | 35987/65536 [6:10:47<4:58:08,  1.65it/s]
 55%|█████▍    | 35988/65536 [6:10:47<4:58:52,  1.65it/s]
 55%|█████▍    | 35989/65536 [6:10:48<5:11:08,  1.58it/s]
 55%|█████▍    | 35990/65536 [6:10:49<5:11:14,  1.58it/s]
 55%|█████▍    | 35991/65536 [6:10:49<5:06:13,  1.61it/s]
 55%|█████▍    | 35992/65536 [6:10:50<5:08:44,  1.59it/s]
 55%|█████▍    | 35993/65536 [6:10:51<5:04:01,  1.62it/s]
 55%|█████▍    | 35994/65536 [6:10:51<5:02:47,  1.63it/s]
 55%|█████▍    | 35995/65536 [6:10:52<4:58:29,  1.65it/s]
 55%|█████▍    | 35996/65536 [6:10:52<4:59:59,  1.64it/s]
 55%|█████▍    | 35997/65536 [6:10:53<5:10:13,  1.59it/s]
 55%|█████▍    | 35998/65536 [6:10:54<5:18:47,  1.54it/s]
 55%|█████▍    | 35999/65536 [6:10:54<5:11:01,  1.58it/s]
 55%|█████▍    | 36000/65536 [6:10:55<5:03:39,  1.62it/s]
                                                         
{'loss': 1.6647, 'learning_rate': 5.068568629851841e-07, 'epoch': 2222.22}
+
 55%|█████▍    | 36000/65536 [6:10:55<5:03:39,  1.62it/s]
 55%|█████▍    | 36001/65536 [6:10:56<5:09:26,  1.59it/s]
 55%|█████▍    | 36002/65536 [6:10:56<5:06:53,  1.60it/s]
 55%|█████▍    | 36003/65536 [6:10:57<4:59:53,  1.64it/s]
 55%|█████▍    | 36004/65536 [6:10:57<5:09:53,  1.59it/s]
 55%|█████▍    | 36005/65536 [6:10:58<5:01:41,  1.63it/s]
 55%|█████▍    | 36006/65536 [6:10:59<5:02:29,  1.63it/s]
 55%|█████▍    | 36007/65536 [6:10:59<4:57:16,  1.66it/s]
 55%|█████▍    | 36008/65536 [6:11:00<4:52:46,  1.68it/s]
 55%|█████▍    | 36009/65536 [6:11:00<4:49:38,  1.70it/s]
 55%|█████▍    | 36010/65536 [6:11:01<4:49:07,  1.70it/s]
 55%|█████▍    | 36011/65536 [6:11:02<4:48:54,  1.70it/s]
 55%|█████▍    | 36012/65536 [6:11:02<4:52:44,  1.68it/s]
 55%|█████▍    | 36013/65536 [6:11:03<5:00:29,  1.64it/s]
 55%|█████▍    | 36014/65536 [6:11:03<4:58:55,  1.65it/s]
 55%|█████▍    | 36015/65536 [6:11:04<5:01:08,  1.63it/s]
 55%|█████▍    | 36016/65536 [6:11:05<5:00:58,  1.63it/s]
 55%|█████▍    | 36017/65536 [6:11:05<5:00:32,  1.64it/s]
 55%|█████▍    | 36018/65536 [6:11:06<5:08:00,  1.60it/s]
 55%|█████▍    | 36019/65536 [6:11:07<5:06:14,  1.61it/s]
 55%|█████▍    | 36020/65536 [6:11:07<5:11:22,  1.58it/s]
                                                         
{'loss': 1.6099, 'learning_rate': 5.065813640259581e-07, 'epoch': 2223.46}
+
 55%|█████▍    | 36020/65536 [6:11:07<5:11:22,  1.58it/s]
 55%|█████▍    | 36021/65536 [6:11:08<4:58:57,  1.65it/s]
 55%|█████▍    | 36022/65536 [6:11:08<5:01:25,  1.63it/s]
 55%|█████▍    | 36023/65536 [6:11:09<4:57:12,  1.66it/s]
 55%|█████▍    | 36024/65536 [6:11:10<4:56:11,  1.66it/s]
 55%|█████▍    | 36025/65536 [6:11:10<4:53:12,  1.68it/s]
 55%|█████▍    | 36026/65536 [6:11:11<4:52:09,  1.68it/s]
 55%|█████▍    | 36027/65536 [6:11:11<4:57:06,  1.66it/s]
 55%|█████▍    | 36028/65536 [6:11:12<4:55:58,  1.66it/s]
 55%|█████▍    | 36029/65536 [6:11:13<5:13:55,  1.57it/s]
 55%|█████▍    | 36030/65536 [6:11:13<5:02:30,  1.63it/s]
 55%|█████▍    | 36031/65536 [6:11:14<5:00:19,  1.64it/s]
 55%|█████▍    | 36032/65536 [6:11:14<4:58:32,  1.65it/s]
 55%|█████▍    | 36033/65536 [6:11:15<5:01:03,  1.63it/s]
 55%|█████▍    | 36034/65536 [6:11:16<4:59:58,  1.64it/s]
 55%|█████▍    | 36035/65536 [6:11:16<5:04:44,  1.61it/s]
 55%|█████▍    | 36036/65536 [6:11:17<5:05:59,  1.61it/s]
 55%|█████▍    | 36037/65536 [6:11:18<5:05:15,  1.61it/s]
 55%|█████▍    | 36038/65536 [6:11:18<5:05:08,  1.61it/s]
 55%|█████▍    | 36039/65536 [6:11:19<5:07:47,  1.60it/s]
 55%|█████▍    | 36040/65536 [6:11:19<5:04:48,  1.61it/s]
                                                         
{'loss': 1.6258, 'learning_rate': 5.063058650667319e-07, 'epoch': 2224.69}
+
 55%|█████▍    | 36040/65536 [6:11:19<5:04:48,  1.61it/s]
 55%|█████▍    | 36041/65536 [6:11:20<5:05:12,  1.61it/s]
 55%|█████▍    | 36042/65536 [6:11:21<4:59:01,  1.64it/s]
 55%|█████▍    | 36043/65536 [6:11:21<4:54:59,  1.67it/s]
 55%|█████▍    | 36044/65536 [6:11:22<5:02:46,  1.62it/s]
 55%|█████▌    | 36045/65536 [6:11:22<4:59:01,  1.64it/s]
 55%|█████▌    | 36046/65536 [6:11:23<5:10:46,  1.58it/s]
 55%|█████▌    | 36047/65536 [6:11:24<5:08:50,  1.59it/s]
 55%|█████▌    | 36048/65536 [6:11:24<5:05:46,  1.61it/s]
 55%|█████▌    | 36049/65536 [6:11:25<5:17:47,  1.55it/s]
 55%|█████▌    | 36050/65536 [6:11:26<5:18:46,  1.54it/s]
 55%|█████▌    | 36051/65536 [6:11:26<5:12:06,  1.57it/s]
 55%|█████▌    | 36052/65536 [6:11:27<5:06:14,  1.60it/s]
 55%|█████▌    | 36053/65536 [6:11:27<5:02:01,  1.63it/s]
 55%|█████▌    | 36054/65536 [6:11:28<4:56:26,  1.66it/s]
 55%|█████▌    | 36055/65536 [6:11:29<4:58:41,  1.64it/s]
 55%|█████▌    | 36056/65536 [6:11:29<5:01:53,  1.63it/s]
 55%|█████▌    | 36057/65536 [6:11:30<4:58:17,  1.65it/s]
 55%|█████▌    | 36058/65536 [6:11:30<4:56:20,  1.66it/s]
 55%|█████▌    | 36059/65536 [6:11:31<4:50:41,  1.69it/s]
 55%|█████▌    | 36060/65536 [6:11:32<4:56:00,  1.66it/s]
                                                         
{'loss': 1.6306, 'learning_rate': 5.060303661075058e-07, 'epoch': 2225.93}
+
 55%|█████▌    | 36060/65536 [6:11:32<4:56:00,  1.66it/s]
 55%|█████▌    | 36061/65536 [6:11:32<5:04:13,  1.61it/s]
 55%|█████▌    | 36062/65536 [6:11:33<5:16:22,  1.55it/s]
 55%|█████▌    | 36063/65536 [6:11:34<5:12:41,  1.57it/s]
 55%|█████▌    | 36064/65536 [6:11:34<5:04:22,  1.61it/s]
 55%|█████▌    | 36065/65536 [6:11:35<5:04:38,  1.61it/s]
 55%|█████▌    | 36066/65536 [6:11:35<4:54:55,  1.67it/s]
 55%|█████▌    | 36067/65536 [6:11:36<5:01:58,  1.63it/s]
 55%|█████▌    | 36068/65536 [6:11:37<5:13:41,  1.57it/s]
 55%|█████▌    | 36069/65536 [6:11:37<5:08:39,  1.59it/s]
 55%|█████▌    | 36070/65536 [6:11:38<5:06:15,  1.60it/s]
 55%|█████▌    | 36071/65536 [6:11:39<5:04:58,  1.61it/s]
 55%|█████▌    | 36072/65536 [6:11:39<4:58:20,  1.65it/s]
 55%|█████▌    | 36073/65536 [6:11:40<4:56:17,  1.66it/s]
 55%|█████▌    | 36074/65536 [6:11:40<4:55:01,  1.66it/s]
 55%|█████▌    | 36075/65536 [6:11:41<4:54:50,  1.67it/s]
 55%|█████▌    | 36076/65536 [6:11:42<4:54:04,  1.67it/s]
 55%|█████▌    | 36077/65536 [6:11:42<4:56:29,  1.66it/s]
 55%|█████▌    | 36078/65536 [6:11:43<5:04:39,  1.61it/s]
 55%|█████▌    | 36079/65536 [6:11:43<4:58:17,  1.65it/s]
 55%|█████▌    | 36080/65536 [6:11:44<4:58:41,  1.64it/s]
                                                         
{'loss': 1.6182, 'learning_rate': 5.057548671482796e-07, 'epoch': 2227.16}
+
 55%|█████▌    | 36080/65536 [6:11:44<4:58:41,  1.64it/s]
 55%|█████▌    | 36081/65536 [6:11:45<5:05:25,  1.61it/s]
 55%|█████▌    | 36082/65536 [6:11:45<5:11:11,  1.58it/s]
 55%|█████▌    | 36083/65536 [6:11:46<5:14:16,  1.56it/s]
 55%|█████▌    | 36084/65536 [6:11:47<5:05:16,  1.61it/s]
 55%|█████▌    | 36085/65536 [6:11:47<4:58:56,  1.64it/s]
 55%|█████▌    | 36086/65536 [6:11:48<5:01:51,  1.63it/s]
 55%|█████▌    | 36087/65536 [6:11:48<4:56:03,  1.66it/s]
 55%|█████▌    | 36088/65536 [6:11:49<4:53:49,  1.67it/s]
 55%|█████▌    | 36089/65536 [6:11:50<5:01:35,  1.63it/s]
 55%|█████▌    | 36090/65536 [6:11:50<4:56:21,  1.66it/s]
 55%|█████▌    | 36091/65536 [6:11:51<5:11:23,  1.58it/s]
 55%|█████▌    | 36092/65536 [6:11:51<5:05:50,  1.60it/s]
 55%|█████▌    | 36093/65536 [6:11:52<5:10:44,  1.58it/s]
 55%|█████▌    | 36094/65536 [6:11:53<5:16:54,  1.55it/s]
 55%|█████▌    | 36095/65536 [6:11:53<5:18:25,  1.54it/s]
 55%|█████▌    | 36096/65536 [6:11:54<5:11:16,  1.58it/s]
 55%|█████▌    | 36097/65536 [6:11:55<5:06:16,  1.60it/s]
 55%|█████▌    | 36098/65536 [6:11:55<5:03:28,  1.62it/s]
 55%|█████▌    | 36099/65536 [6:11:56<5:09:54,  1.58it/s]
 55%|█████▌    | 36100/65536 [6:11:56<5:00:13,  1.63it/s]
                                                         
{'loss': 1.5924, 'learning_rate': 5.054793681890535e-07, 'epoch': 2228.4}
+
 55%|█████▌    | 36100/65536 [6:11:56<5:00:13,  1.63it/s]
 55%|█████▌    | 36101/65536 [6:11:57<5:03:04,  1.62it/s]
 55%|█████▌    | 36102/65536 [6:11:58<5:00:40,  1.63it/s]
 55%|█████▌    | 36103/65536 [6:11:58<4:53:44,  1.67it/s]
 55%|█████▌    | 36104/65536 [6:11:59<4:58:55,  1.64it/s]
 55%|█████▌    | 36105/65536 [6:12:00<4:59:53,  1.64it/s]
 55%|█████▌    | 36106/65536 [6:12:00<4:59:57,  1.64it/s]
 55%|█████▌    | 36107/65536 [6:12:01<4:56:30,  1.65it/s]
 55%|█████▌    | 36108/65536 [6:12:01<5:07:01,  1.60it/s]
 55%|█████▌    | 36109/65536 [6:12:02<5:07:28,  1.60it/s]
 55%|█████▌    | 36110/65536 [6:12:03<5:11:57,  1.57it/s]
 55%|█████▌    | 36111/65536 [6:12:03<5:06:16,  1.60it/s]
 55%|█████▌    | 36112/65536 [6:12:04<5:07:10,  1.60it/s]
 55%|█████▌    | 36113/65536 [6:12:05<5:03:29,  1.62it/s]
 55%|█████▌    | 36114/65536 [6:12:05<5:06:11,  1.60it/s]
 55%|█████▌    | 36115/65536 [6:12:06<5:03:32,  1.62it/s]
 55%|█████▌    | 36116/65536 [6:12:06<5:01:44,  1.63it/s]
 55%|█████▌    | 36117/65536 [6:12:07<5:02:17,  1.62it/s]
 55%|█████▌    | 36118/65536 [6:12:08<4:55:12,  1.66it/s]
 55%|█████▌    | 36119/65536 [6:12:08<4:51:35,  1.68it/s]
 55%|█████▌    | 36120/65536 [6:12:09<4:52:45,  1.67it/s]
                                                         
{'loss': 1.6585, 'learning_rate': 5.052038692298274e-07, 'epoch': 2229.63}
+
 55%|█████▌    | 36120/65536 [6:12:09<4:52:45,  1.67it/s]
 55%|█████▌    | 36121/65536 [6:12:09<4:55:29,  1.66it/s]
 55%|█████▌    | 36122/65536 [6:12:10<4:52:05,  1.68it/s]
 55%|█████▌    | 36123/65536 [6:12:11<4:49:02,  1.70it/s]
 55%|█████▌    | 36124/65536 [6:12:11<4:55:45,  1.66it/s]
 55%|█████▌    | 36125/65536 [6:12:12<5:01:25,  1.63it/s]
 55%|█████▌    | 36126/65536 [6:12:12<5:04:24,  1.61it/s]
 55%|█████▌    | 36127/65536 [6:12:13<5:17:55,  1.54it/s]
 55%|█████▌    | 36128/65536 [6:12:14<5:19:23,  1.53it/s]
 55%|█████▌    | 36129/65536 [6:12:14<5:16:37,  1.55it/s]
 55%|█████▌    | 36130/65536 [6:12:15<5:04:52,  1.61it/s]
 55%|█████▌    | 36131/65536 [6:12:16<5:02:28,  1.62it/s]
 55%|█████▌    | 36132/65536 [6:12:16<4:55:10,  1.66it/s]
 55%|█████▌    | 36133/65536 [6:12:17<4:50:07,  1.69it/s]
 55%|█████▌    | 36134/65536 [6:12:17<4:51:56,  1.68it/s]
 55%|█████▌    | 36135/65536 [6:12:18<4:48:55,  1.70it/s]
 55%|█████▌    | 36136/65536 [6:12:19<4:48:56,  1.70it/s]
 55%|█████▌    | 36137/65536 [6:12:19<4:49:43,  1.69it/s]
 55%|█████▌    | 36138/65536 [6:12:20<4:53:58,  1.67it/s]
 55%|█████▌    | 36139/65536 [6:12:20<4:55:22,  1.66it/s]
 55%|█████▌    | 36140/65536 [6:12:21<4:54:27,  1.66it/s]
                                                         
{'loss': 1.6297, 'learning_rate': 5.049283702706012e-07, 'epoch': 2230.86}
+
 55%|█████▌    | 36140/65536 [6:12:21<4:54:27,  1.66it/s]
 55%|█████▌    | 36141/65536 [6:12:22<5:05:47,  1.60it/s]
 55%|█████▌    | 36142/65536 [6:12:22<5:01:33,  1.62it/s]
 55%|█████▌    | 36143/65536 [6:12:23<5:20:01,  1.53it/s]
 55%|█████▌    | 36144/65536 [6:12:24<5:23:12,  1.52it/s]
 55%|█████▌    | 36145/65536 [6:12:24<5:19:04,  1.54it/s]
 55%|█████▌    | 36146/65536 [6:12:25<5:14:57,  1.56it/s]
 55%|█████▌    | 36147/65536 [6:12:25<5:06:08,  1.60it/s]
 55%|█████▌    | 36148/65536 [6:12:26<5:03:35,  1.61it/s]
 55%|█████▌    | 36149/65536 [6:12:27<4:55:12,  1.66it/s]
 55%|█████▌    | 36150/65536 [6:12:27<4:50:45,  1.68it/s]
 55%|█████▌    | 36151/65536 [6:12:28<4:46:37,  1.71it/s]
 55%|█████▌    | 36152/65536 [6:12:28<4:53:50,  1.67it/s]
 55%|█████▌    | 36153/65536 [6:12:29<4:56:27,  1.65it/s]
 55%|█████▌    | 36154/65536 [6:12:30<5:01:24,  1.62it/s]
 55%|█████▌    | 36155/65536 [6:12:30<5:03:20,  1.61it/s]
 55%|█████▌    | 36156/65536 [6:12:31<4:55:53,  1.65it/s]
 55%|█████▌    | 36157/65536 [6:12:31<4:54:39,  1.66it/s]
 55%|█████▌    | 36158/65536 [6:12:32<4:57:14,  1.65it/s]
 55%|█████▌    | 36159/65536 [6:12:33<5:14:32,  1.56it/s]
 55%|█████▌    | 36160/65536 [6:12:33<5:11:28,  1.57it/s]
                                                         
{'loss': 1.6579, 'learning_rate': 5.046528713113751e-07, 'epoch': 2232.1}
+
 55%|█████▌    | 36160/65536 [6:12:33<5:11:28,  1.57it/s]
 55%|█████▌    | 36161/65536 [6:12:34<5:06:25,  1.60it/s]
 55%|█████▌    | 36162/65536 [6:12:35<5:08:07,  1.59it/s]
 55%|█████▌    | 36163/65536 [6:12:35<4:57:43,  1.64it/s]
 55%|█████▌    | 36164/65536 [6:12:36<5:00:59,  1.63it/s]
 55%|█████▌    | 36165/65536 [6:12:36<4:57:54,  1.64it/s]
 55%|█████▌    | 36166/65536 [6:12:37<5:00:30,  1.63it/s]
 55%|█████▌    | 36167/65536 [6:12:38<5:07:00,  1.59it/s]
 55%|█████▌    | 36168/65536 [6:12:38<5:10:00,  1.58it/s]
 55%|█████▌    | 36169/65536 [6:12:39<5:05:58,  1.60it/s]
 55%|█████▌    | 36170/65536 [6:12:40<4:59:27,  1.63it/s]
 55%|█████▌    | 36171/65536 [6:12:40<5:03:43,  1.61it/s]
 55%|█████▌    | 36172/65536 [6:12:41<5:09:27,  1.58it/s]
 55%|█████▌    | 36173/65536 [6:12:41<5:00:03,  1.63it/s]
 55%|█████▌    | 36174/65536 [6:12:42<4:56:43,  1.65it/s]
 55%|█████▌    | 36175/65536 [6:12:43<5:02:54,  1.62it/s]
 55%|█████▌    | 36176/65536 [6:12:43<4:57:10,  1.65it/s]
 55%|█████▌    | 36177/65536 [6:12:44<4:51:24,  1.68it/s]
 55%|█████▌    | 36178/65536 [6:12:44<4:51:22,  1.68it/s]
 55%|█████▌    | 36179/65536 [6:12:45<4:55:48,  1.65it/s]
 55%|█████▌    | 36180/65536 [6:12:46<5:05:06,  1.60it/s]
                                                         
{'loss': 1.6233, 'learning_rate': 5.043773723521489e-07, 'epoch': 2233.33}
+
 55%|█████▌    | 36180/65536 [6:12:46<5:05:06,  1.60it/s]
 55%|█████▌    | 36181/65536 [6:12:46<5:01:20,  1.62it/s]
 55%|█████▌    | 36182/65536 [6:12:47<5:01:47,  1.62it/s]
 55%|█████▌    | 36183/65536 [6:12:48<5:01:13,  1.62it/s]
 55%|█████▌    | 36184/65536 [6:12:48<4:54:02,  1.66it/s]
 55%|█████▌    | 36185/65536 [6:12:49<4:54:44,  1.66it/s]
 55%|█████▌    | 36186/65536 [6:12:49<4:54:55,  1.66it/s]
 55%|█████▌    | 36187/65536 [6:12:50<4:53:08,  1.67it/s]
 55%|█████▌    | 36188/65536 [6:12:51<5:03:47,  1.61it/s]
 55%|█████▌    | 36189/65536 [6:12:51<5:04:26,  1.61it/s]
 55%|█████▌    | 36190/65536 [6:12:52<5:03:22,  1.61it/s]
 55%|█████▌    | 36191/65536 [6:12:52<5:10:45,  1.57it/s]
 55%|█████▌    | 36192/65536 [6:12:53<5:05:43,  1.60it/s]
 55%|█████▌    | 36193/65536 [6:12:54<4:57:07,  1.65it/s]
 55%|█████▌    | 36194/65536 [6:12:54<4:55:03,  1.66it/s]
 55%|█████▌    | 36195/65536 [6:12:55<5:07:12,  1.59it/s]
 55%|█████▌    | 36196/65536 [6:12:56<5:00:17,  1.63it/s]
 55%|█████▌    | 36197/65536 [6:12:56<5:00:30,  1.63it/s]
 55%|█████▌    | 36198/65536 [6:12:57<4:58:11,  1.64it/s]
 55%|█████▌    | 36199/65536 [6:12:57<4:57:09,  1.65it/s]
 55%|█████▌    | 36200/65536 [6:12:58<4:55:28,  1.65it/s]
                                                         
{'loss': 1.6651, 'learning_rate': 5.041018733929228e-07, 'epoch': 2234.57}
+
 55%|█████▌    | 36200/65536 [6:12:58<4:55:28,  1.65it/s]
 55%|█████▌    | 36201/65536 [6:12:59<5:02:41,  1.62it/s]
 55%|█████▌    | 36202/65536 [6:12:59<4:56:59,  1.65it/s]
 55%|█████▌    | 36203/65536 [6:13:00<4:52:04,  1.67it/s]
 55%|█████▌    | 36204/65536 [6:13:00<5:02:51,  1.61it/s]
 55%|█████▌    | 36205/65536 [6:13:01<5:04:10,  1.61it/s]
 55%|█████▌    | 36206/65536 [6:13:02<5:08:00,  1.59it/s]
 55%|█████▌    | 36207/65536 [6:13:02<5:01:54,  1.62it/s]
 55%|█████▌    | 36208/65536 [6:13:03<5:13:10,  1.56it/s]
 55%|█████▌    | 36209/65536 [6:13:04<5:11:14,  1.57it/s]
 55%|█████▌    | 36210/65536 [6:13:04<5:07:11,  1.59it/s]
 55%|█████▌    | 36211/65536 [6:13:05<4:59:13,  1.63it/s]
 55%|█████▌    | 36212/65536 [6:13:05<5:00:47,  1.62it/s]
 55%|█████▌    | 36213/65536 [6:13:06<5:04:55,  1.60it/s]
 55%|█████▌    | 36214/65536 [6:13:07<4:58:57,  1.63it/s]
 55%|█████▌    | 36215/65536 [6:13:07<4:55:39,  1.65it/s]
 55%|█████▌    | 36216/65536 [6:13:08<4:58:29,  1.64it/s]
 55%|█████▌    | 36217/65536 [6:13:08<4:52:59,  1.67it/s]
 55%|█████▌    | 36218/65536 [6:13:09<4:52:11,  1.67it/s]
 55%|█████▌    | 36219/65536 [6:13:10<4:55:41,  1.65it/s]
 55%|█████▌    | 36220/65536 [6:13:10<4:56:46,  1.65it/s]
                                                         
{'loss': 1.6159, 'learning_rate': 5.038263744336967e-07, 'epoch': 2235.8}
+
 55%|█████▌    | 36220/65536 [6:13:10<4:56:46,  1.65it/s]
 55%|█████▌    | 36221/65536 [6:13:11<4:58:15,  1.64it/s]
 55%|█████▌    | 36222/65536 [6:13:11<4:58:16,  1.64it/s]
 55%|█████▌    | 36223/65536 [6:13:12<5:10:13,  1.57it/s]
 55%|█████▌    | 36224/65536 [6:13:13<5:09:30,  1.58it/s]
 55%|█████▌    | 36225/65536 [6:13:13<5:06:46,  1.59it/s]
 55%|█████▌    | 36226/65536 [6:13:14<5:03:32,  1.61it/s]
 55%|█████▌    | 36227/65536 [6:13:15<5:03:49,  1.61it/s]
 55%|█████▌    | 36228/65536 [6:13:15<5:01:14,  1.62it/s]
 55%|█████▌    | 36229/65536 [6:13:16<5:01:32,  1.62it/s]
 55%|█████▌    | 36230/65536 [6:13:17<5:08:14,  1.58it/s]
 55%|█████▌    | 36231/65536 [6:13:17<5:07:04,  1.59it/s]
 55%|█████▌    | 36232/65536 [6:13:18<5:06:39,  1.59it/s]
 55%|█████▌    | 36233/65536 [6:13:18<4:57:41,  1.64it/s]
 55%|█████▌    | 36234/65536 [6:13:19<5:02:04,  1.62it/s]
 55%|█████▌    | 36235/65536 [6:13:20<5:06:10,  1.59it/s]
 55%|█████▌    | 36236/65536 [6:13:20<5:03:52,  1.61it/s]
 55%|█████▌    | 36237/65536 [6:13:21<5:04:29,  1.60it/s]
 55%|█████▌    | 36238/65536 [6:13:21<5:02:47,  1.61it/s]
 55%|█████▌    | 36239/65536 [6:13:22<4:56:38,  1.65it/s]
 55%|█████▌    | 36240/65536 [6:13:23<5:10:04,  1.57it/s]
                                                         
{'loss': 1.6339, 'learning_rate': 5.035508754744705e-07, 'epoch': 2237.04}
+
 55%|█████▌    | 36240/65536 [6:13:23<5:10:04,  1.57it/s]
 55%|█████▌    | 36241/65536 [6:13:23<5:17:29,  1.54it/s]
 55%|█████▌    | 36242/65536 [6:13:24<5:16:37,  1.54it/s]
 55%|█████▌    | 36243/65536 [6:13:25<5:15:25,  1.55it/s]
 55%|█████▌    | 36244/65536 [6:13:25<5:08:16,  1.58it/s]
 55%|█████▌    | 36245/65536 [6:13:26<5:06:24,  1.59it/s]
 55%|█████▌    | 36246/65536 [6:13:27<4:58:03,  1.64it/s]
 55%|█████▌    | 36247/65536 [6:13:27<4:56:30,  1.65it/s]
 55%|█████▌    | 36248/65536 [6:13:28<4:52:44,  1.67it/s]
 55%|█████▌    | 36249/65536 [6:13:28<4:46:47,  1.70it/s]
 55%|█████▌    | 36250/65536 [6:13:29<4:54:18,  1.66it/s]
 55%|█████▌    | 36251/65536 [6:13:29<4:49:32,  1.69it/s]
 55%|█████▌    | 36252/65536 [6:13:30<4:51:23,  1.67it/s]
 55%|█████▌    | 36253/65536 [6:13:31<4:58:00,  1.64it/s]
 55%|█████▌    | 36254/65536 [6:13:31<5:03:18,  1.61it/s]
 55%|█████▌    | 36255/65536 [6:13:32<5:02:53,  1.61it/s]
 55%|█████▌    | 36256/65536 [6:13:33<5:08:38,  1.58it/s]
 55%|█████▌    | 36257/65536 [6:13:33<4:59:29,  1.63it/s]
 55%|█████▌    | 36258/65536 [6:13:34<4:54:57,  1.65it/s]
 55%|█████▌    | 36259/65536 [6:13:34<4:57:04,  1.64it/s]
 55%|█████▌    | 36260/65536 [6:13:35<4:58:52,  1.63it/s]
                                                         
{'loss': 1.6332, 'learning_rate': 5.032753765152442e-07, 'epoch': 2238.27}
+
 55%|█████▌    | 36260/65536 [6:13:35<4:58:52,  1.63it/s]
 55%|█████▌    | 36261/65536 [6:13:36<4:53:22,  1.66it/s]
 55%|█████▌    | 36262/65536 [6:13:36<4:43:10,  1.72it/s]
 55%|█████▌    | 36263/65536 [6:13:37<4:57:01,  1.64it/s]
 55%|█████▌    | 36264/65536 [6:13:38<5:10:27,  1.57it/s]
 55%|█████▌    | 36265/65536 [6:13:38<5:06:03,  1.59it/s]
 55%|█████▌    | 36266/65536 [6:13:39<5:03:58,  1.60it/s]
 55%|█████▌    | 36267/65536 [6:13:39<5:06:03,  1.59it/s]
 55%|█████▌    | 36268/65536 [6:13:40<5:01:36,  1.62it/s]
 55%|█████▌    | 36269/65536 [6:13:41<5:07:10,  1.59it/s]
 55%|█████▌    | 36270/65536 [6:13:41<5:01:38,  1.62it/s]
 55%|█████▌    | 36271/65536 [6:13:42<4:56:21,  1.65it/s]
 55%|█████▌    | 36272/65536 [6:13:42<5:03:37,  1.61it/s]
 55%|█████▌    | 36273/65536 [6:13:43<4:58:43,  1.63it/s]
 55%|█████▌    | 36274/65536 [6:13:44<5:00:59,  1.62it/s]
 55%|█████▌    | 36275/65536 [6:13:44<4:59:44,  1.63it/s]
 55%|█████▌    | 36276/65536 [6:13:45<5:03:43,  1.61it/s]
 55%|█████▌    | 36277/65536 [6:13:46<5:04:24,  1.60it/s]
 55%|█████▌    | 36278/65536 [6:13:46<4:57:10,  1.64it/s]
 55%|█████▌    | 36279/65536 [6:13:47<4:57:10,  1.64it/s]
 55%|█████▌    | 36280/65536 [6:13:47<4:58:22,  1.63it/s]
                                                         
{'loss': 1.6244, 'learning_rate': 5.02999877556018e-07, 'epoch': 2239.51}
+
 55%|█████▌    | 36280/65536 [6:13:47<4:58:22,  1.63it/s]
 55%|█████▌    | 36281/65536 [6:13:48<5:00:49,  1.62it/s]
 55%|█████▌    | 36282/65536 [6:13:49<5:08:57,  1.58it/s]
 55%|█████▌    | 36283/65536 [6:13:49<5:02:46,  1.61it/s]
 55%|█████▌    | 36284/65536 [6:13:50<4:57:16,  1.64it/s]
 55%|█████▌    | 36285/65536 [6:13:50<4:53:01,  1.66it/s]
 55%|█████▌    | 36286/65536 [6:13:51<4:48:54,  1.69it/s]
 55%|█████▌    | 36287/65536 [6:13:52<4:48:50,  1.69it/s]
 55%|█████▌    | 36288/65536 [6:13:52<4:49:23,  1.68it/s]
 55%|█████▌    | 36289/65536 [6:13:53<4:58:42,  1.63it/s]
 55%|█████▌    | 36290/65536 [6:13:53<4:53:00,  1.66it/s]
 55%|█████▌    | 36291/65536 [6:13:54<4:58:44,  1.63it/s]
 55%|█████▌    | 36292/65536 [6:13:55<4:58:27,  1.63it/s]
 55%|█████▌    | 36293/65536 [6:13:55<4:56:00,  1.65it/s]
 55%|█████▌    | 36294/65536 [6:13:56<4:54:18,  1.66it/s]
 55%|█████▌    | 36295/65536 [6:13:57<5:08:16,  1.58it/s]
 55%|█████▌    | 36296/65536 [6:13:57<5:02:17,  1.61it/s]
 55%|█████▌    | 36297/65536 [6:13:58<4:57:45,  1.64it/s]
 55%|█████▌    | 36298/65536 [6:13:58<4:59:25,  1.63it/s]
 55%|█████▌    | 36299/65536 [6:13:59<4:58:53,  1.63it/s]
 55%|█████▌    | 36300/65536 [6:14:00<4:57:04,  1.64it/s]
                                                         
{'loss': 1.6199, 'learning_rate': 5.027243785967919e-07, 'epoch': 2240.74}
+
 55%|█████▌    | 36300/65536 [6:14:00<4:57:04,  1.64it/s]
 55%|█████▌    | 36301/65536 [6:14:00<4:58:19,  1.63it/s]
 55%|█████▌    | 36302/65536 [6:14:01<4:57:47,  1.64it/s]
 55%|█████▌    | 36303/65536 [6:14:01<5:01:42,  1.61it/s]
 55%|█████▌    | 36304/65536 [6:14:02<5:07:00,  1.59it/s]
 55%|█████▌    | 36305/65536 [6:14:03<5:13:26,  1.55it/s]
 55%|█████▌    | 36306/65536 [6:14:03<5:03:18,  1.61it/s]
 55%|█████▌    | 36307/65536 [6:14:04<5:00:24,  1.62it/s]
 55%|█████▌    | 36308/65536 [6:14:05<5:02:07,  1.61it/s]
 55%|█████▌    | 36309/65536 [6:14:05<4:56:07,  1.65it/s]
 55%|█████▌    | 36310/65536 [6:14:06<5:01:16,  1.62it/s]
 55%|█████▌    | 36311/65536 [6:14:06<4:55:04,  1.65it/s]
 55%|█████▌    | 36312/65536 [6:14:07<4:52:47,  1.66it/s]
 55%|█████▌    | 36313/65536 [6:14:08<4:49:20,  1.68it/s]
 55%|█████▌    | 36314/65536 [6:14:08<5:03:44,  1.60it/s]
 55%|█████▌    | 36315/65536 [6:14:09<5:04:37,  1.60it/s]
 55%|█████▌    | 36316/65536 [6:14:09<4:59:52,  1.62it/s]
 55%|█████▌    | 36317/65536 [6:14:10<5:03:07,  1.61it/s]
 55%|█████▌    | 36318/65536 [6:14:11<5:06:15,  1.59it/s]
 55%|█████▌    | 36319/65536 [6:14:11<5:08:37,  1.58it/s]
 55%|█████▌    | 36320/65536 [6:14:12<5:01:52,  1.61it/s]
                                                         
{'loss': 1.5973, 'learning_rate': 5.024488796375657e-07, 'epoch': 2241.98}
+
 55%|█████▌    | 36320/65536 [6:14:12<5:01:52,  1.61it/s]
 55%|█████▌    | 36321/65536 [6:14:13<5:10:45,  1.57it/s]
 55%|█████▌    | 36322/65536 [6:14:13<5:06:36,  1.59it/s]
 55%|█████▌    | 36323/65536 [6:14:14<4:55:41,  1.65it/s]
 55%|█████▌    | 36324/65536 [6:14:14<4:53:19,  1.66it/s]
 55%|█████▌    | 36325/65536 [6:14:15<4:56:54,  1.64it/s]
 55%|█████▌    | 36326/65536 [6:14:16<5:02:24,  1.61it/s]
 55%|█████▌    | 36327/65536 [6:14:16<5:04:01,  1.60it/s]
 55%|█████▌    | 36328/65536 [6:14:17<5:06:00,  1.59it/s]
 55%|█████▌    | 36329/65536 [6:14:18<5:02:31,  1.61it/s]
 55%|█████▌    | 36330/65536 [6:14:18<5:01:03,  1.62it/s]
 55%|█████▌    | 36331/65536 [6:14:19<5:00:50,  1.62it/s]
 55%|█████▌    | 36332/65536 [6:14:19<4:58:24,  1.63it/s]
 55%|█████▌    | 36333/65536 [6:14:20<4:55:36,  1.65it/s]
 55%|█████▌    | 36334/65536 [6:14:21<5:04:35,  1.60it/s]
 55%|█████▌    | 36335/65536 [6:14:21<4:58:14,  1.63it/s]
 55%|█████▌    | 36336/65536 [6:14:22<5:02:45,  1.61it/s]
 55%|█████▌    | 36337/65536 [6:14:22<5:03:21,  1.60it/s]
 55%|█████▌    | 36338/65536 [6:14:23<4:56:47,  1.64it/s]
 55%|█████▌    | 36339/65536 [6:14:24<4:56:17,  1.64it/s]
 55%|█████▌    | 36340/65536 [6:14:24<4:49:21,  1.68it/s]
                                                         
{'loss': 1.641, 'learning_rate': 5.021733806783396e-07, 'epoch': 2243.21}
+
 55%|█████▌    | 36340/65536 [6:14:24<4:49:21,  1.68it/s]
 55%|█████▌    | 36341/65536 [6:14:25<4:58:09,  1.63it/s]
 55%|█████▌    | 36342/65536 [6:14:26<5:08:21,  1.58it/s]
 55%|█████▌    | 36343/65536 [6:14:26<5:13:54,  1.55it/s]
 55%|█████▌    | 36344/65536 [6:14:27<5:05:54,  1.59it/s]
 55%|█████▌    | 36345/65536 [6:14:27<4:56:37,  1.64it/s]
 55%|█████▌    | 36346/65536 [6:14:28<4:50:49,  1.67it/s]
 55%|█████▌    | 36347/65536 [6:14:29<4:51:00,  1.67it/s]
 55%|█████▌    | 36348/65536 [6:14:29<4:57:15,  1.64it/s]
 55%|█████▌    | 36349/65536 [6:14:30<5:05:08,  1.59it/s]
 55%|█████▌    | 36350/65536 [6:14:30<4:59:08,  1.63it/s]
 55%|█████▌    | 36351/65536 [6:14:31<5:05:28,  1.59it/s]
 55%|█████▌    | 36352/65536 [6:14:32<5:08:22,  1.58it/s]
 55%|█████▌    | 36353/65536 [6:14:32<5:13:59,  1.55it/s]
 55%|█████▌    | 36354/65536 [6:14:33<5:05:31,  1.59it/s]
 55%|█████▌    | 36355/65536 [6:14:34<5:01:14,  1.61it/s]
 55%|█████▌    | 36356/65536 [6:14:34<4:57:42,  1.63it/s]
 55%|█████▌    | 36357/65536 [6:14:35<4:53:48,  1.66it/s]
 55%|█████▌    | 36358/65536 [6:14:35<4:55:57,  1.64it/s]
 55%|█████▌    | 36359/65536 [6:14:36<4:59:06,  1.63it/s]
 55%|█████▌    | 36360/65536 [6:14:37<4:59:37,  1.62it/s]
                                                         
{'loss': 1.6018, 'learning_rate': 5.018978817191135e-07, 'epoch': 2244.44}
+
 55%|█████▌    | 36360/65536 [6:14:37<4:59:37,  1.62it/s]
 55%|█████▌    | 36361/65536 [6:14:37<4:58:03,  1.63it/s]
 55%|█████▌    | 36362/65536 [6:14:38<4:55:43,  1.64it/s]
 55%|█████▌    | 36363/65536 [6:14:38<4:52:46,  1.66it/s]
 55%|█████▌    | 36364/65536 [6:14:39<4:50:23,  1.67it/s]
 55%|█████▌    | 36365/65536 [6:14:40<4:46:34,  1.70it/s]
 55%|█████▌    | 36366/65536 [6:14:40<4:56:26,  1.64it/s]
 55%|█████▌    | 36367/65536 [6:14:41<5:12:42,  1.55it/s]
 55%|█████▌    | 36368/65536 [6:14:42<5:01:14,  1.61it/s]
 55%|█████▌    | 36369/65536 [6:14:42<5:01:28,  1.61it/s]
 55%|█████▌    | 36370/65536 [6:14:43<5:03:21,  1.60it/s]
 55%|█████▌    | 36371/65536 [6:14:43<5:00:08,  1.62it/s]
 55%|█████▌    | 36372/65536 [6:14:44<5:06:58,  1.58it/s]
 56%|█████▌    | 36373/65536 [6:14:45<5:00:36,  1.62it/s]
 56%|█████▌    | 36374/65536 [6:14:45<5:06:55,  1.58it/s]
 56%|█████▌    | 36375/65536 [6:14:46<5:09:45,  1.57it/s]
 56%|█████▌    | 36376/65536 [6:14:47<5:06:49,  1.58it/s]
 56%|█████▌    | 36377/65536 [6:14:47<5:06:54,  1.58it/s]
 56%|█████▌    | 36378/65536 [6:14:48<5:03:08,  1.60it/s]
 56%|█████▌    | 36379/65536 [6:14:48<4:56:59,  1.64it/s]
 56%|█████▌    | 36380/65536 [6:14:49<4:58:20,  1.63it/s]
                                                         
{'loss': 1.6101, 'learning_rate': 5.016223827598873e-07, 'epoch': 2245.68}
+
 56%|█████▌    | 36380/65536 [6:14:49<4:58:20,  1.63it/s]
 56%|█████▌    | 36381/65536 [6:14:50<4:59:11,  1.62it/s]
 56%|█████▌    | 36382/65536 [6:14:50<4:58:12,  1.63it/s]
 56%|█████▌    | 36383/65536 [6:14:51<5:00:23,  1.62it/s]
 56%|█████▌    | 36384/65536 [6:14:51<4:57:04,  1.64it/s]
 56%|█████▌    | 36385/65536 [6:14:52<4:59:49,  1.62it/s]
 56%|█████▌    | 36386/65536 [6:14:53<5:06:36,  1.58it/s]
 56%|█████▌    | 36387/65536 [6:14:53<4:58:23,  1.63it/s]
 56%|█████▌    | 36388/65536 [6:14:54<5:00:48,  1.62it/s]
 56%|█████▌    | 36389/65536 [6:14:55<4:56:14,  1.64it/s]
 56%|█████▌    | 36390/65536 [6:14:55<4:50:45,  1.67it/s]
 56%|█████▌    | 36391/65536 [6:14:56<5:05:33,  1.59it/s]
 56%|█████▌    | 36392/65536 [6:14:56<5:05:43,  1.59it/s]
 56%|█████▌    | 36393/65536 [6:14:57<4:56:07,  1.64it/s]
 56%|█████▌    | 36394/65536 [6:14:58<4:59:47,  1.62it/s]
 56%|█████▌    | 36395/65536 [6:14:58<4:54:18,  1.65it/s]
 56%|█████▌    | 36396/65536 [6:14:59<4:53:24,  1.66it/s]
 56%|█████▌    | 36397/65536 [6:14:59<4:56:55,  1.64it/s]
 56%|█████▌    | 36398/65536 [6:15:00<4:54:15,  1.65it/s]
 56%|█████▌    | 36399/65536 [6:15:01<4:59:21,  1.62it/s]
 56%|█████▌    | 36400/65536 [6:15:01<5:03:53,  1.60it/s]
                                                         
{'loss': 1.6013, 'learning_rate': 5.013468838006612e-07, 'epoch': 2246.91}
+
 56%|█████▌    | 36400/65536 [6:15:01<5:03:53,  1.60it/s]
 56%|█████▌    | 36401/65536 [6:15:02<5:03:36,  1.60it/s]
 56%|█████▌    | 36402/65536 [6:15:03<5:11:11,  1.56it/s]
 56%|█████▌    | 36403/65536 [6:15:03<5:07:34,  1.58it/s]
 56%|█████▌    | 36404/65536 [6:15:04<5:00:30,  1.62it/s]
 56%|█████▌    | 36405/65536 [6:15:05<5:02:35,  1.60it/s]
 56%|█████▌    | 36406/65536 [6:15:05<4:57:08,  1.63it/s]
 56%|█████▌    | 36407/65536 [6:15:06<4:59:05,  1.62it/s]
 56%|██���██▌    | 36408/65536 [6:15:06<4:52:59,  1.66it/s]
 56%|█████▌    | 36409/65536 [6:15:07<4:54:12,  1.65it/s]
 56%|█████▌    | 36410/65536 [6:15:07<4:51:19,  1.67it/s]
 56%|█████▌    | 36411/65536 [6:15:08<5:04:06,  1.60it/s]
 56%|█████▌    | 36412/65536 [6:15:09<5:10:42,  1.56it/s]
 56%|█████▌    | 36413/65536 [6:15:09<5:02:22,  1.61it/s]
 56%|█████▌    | 36414/65536 [6:15:10<4:59:02,  1.62it/s]
 56%|█████▌    | 36415/65536 [6:15:11<4:58:01,  1.63it/s]
 56%|█████▌    | 36416/65536 [6:15:11<4:52:54,  1.66it/s]
 56%|█████▌    | 36417/65536 [6:15:12<4:54:53,  1.65it/s]
 56%|█████▌    | 36418/65536 [6:15:12<4:59:51,  1.62it/s]
 56%|█████▌    | 36419/65536 [6:15:13<4:58:44,  1.62it/s]
 56%|█████▌    | 36420/65536 [6:15:14<4:56:18,  1.64it/s]
                                                         
{'loss': 1.6538, 'learning_rate': 5.01071384841435e-07, 'epoch': 2248.15}
+
 56%|█████▌    | 36420/65536 [6:15:14<4:56:18,  1.64it/s]
 56%|█████▌    | 36421/65536 [6:15:14<4:55:44,  1.64it/s]
 56%|█████▌    | 36422/65536 [6:15:15<5:06:19,  1.58it/s]
 56%|█████▌    | 36423/65536 [6:15:16<5:10:50,  1.56it/s]
 56%|█████▌    | 36424/65536 [6:15:16<5:06:46,  1.58it/s]
 56%|█████▌    | 36425/65536 [6:15:17<4:59:09,  1.62it/s]
 56%|█████▌    | 36426/65536 [6:15:17<5:00:00,  1.62it/s]
 56%|█████▌    | 36427/65536 [6:15:18<5:00:45,  1.61it/s]
 56%|█████▌    | 36428/65536 [6:15:19<4:59:03,  1.62it/s]
 56%|█████▌    | 36429/65536 [6:15:19<4:55:14,  1.64it/s]
 56%|█████▌    | 36430/65536 [6:15:20<4:53:56,  1.65it/s]
 56%|█████▌    | 36431/65536 [6:15:20<4:48:45,  1.68it/s]
 56%|█████▌    | 36432/65536 [6:15:21<4:49:34,  1.68it/s]
 56%|█████▌    | 36433/65536 [6:15:22<5:00:17,  1.62it/s]
 56%|█████▌    | 36434/65536 [6:15:22<5:07:42,  1.58it/s]
 56%|█████▌    | 36435/65536 [6:15:23<5:02:06,  1.61it/s]
 56%|█████▌    | 36436/65536 [6:15:24<4:58:17,  1.63it/s]
 56%|█████▌    | 36437/65536 [6:15:24<5:00:26,  1.61it/s]
 56%|█████▌    | 36438/65536 [6:15:25<4:59:51,  1.62it/s]
 56%|█████▌    | 36439/65536 [6:15:25<5:02:03,  1.61it/s]
 56%|█████▌    | 36440/65536 [6:15:26<4:54:55,  1.64it/s]
                                                         
{'loss': 1.6441, 'learning_rate': 5.007958858822089e-07, 'epoch': 2249.38}
+
 56%|█████▌    | 36440/65536 [6:15:26<4:54:55,  1.64it/s]
 56%|█████▌    | 36441/65536 [6:15:27<4:49:54,  1.67it/s]
 56%|█████▌    | 36442/65536 [6:15:27<4:52:27,  1.66it/s]
 56%|█████▌    | 36443/65536 [6:15:28<4:57:14,  1.63it/s]
 56%|█████▌    | 36444/65536 [6:15:28<4:55:23,  1.64it/s]
 56%|█████▌    | 36445/65536 [6:15:29<4:56:05,  1.64it/s]
 56%|█████▌    | 36446/65536 [6:15:30<4:53:12,  1.65it/s]
 56%|█████▌    | 36447/65536 [6:15:30<4:51:27,  1.66it/s]
 56%|█████▌    | 36448/65536 [6:15:31<4:56:32,  1.63it/s]
 56%|█████▌    | 36449/65536 [6:15:32<4:57:23,  1.63it/s]
 56%|█████▌    | 36450/65536 [6:15:32<5:01:22,  1.61it/s]
 56%|█████▌    | 36451/65536 [6:15:33<5:04:48,  1.59it/s]
 56%|█████▌    | 36452/65536 [6:15:33<5:10:10,  1.56it/s]
 56%|█████▌    | 36453/65536 [6:15:34<4:58:50,  1.62it/s]
 56%|█████▌    | 36454/65536 [6:15:35<4:53:40,  1.65it/s]
 56%|█████▌    | 36455/65536 [6:15:35<4:57:10,  1.63it/s]
 56%|█████▌    | 36456/65536 [6:15:36<4:57:17,  1.63it/s]
 56%|█████▌    | 36457/65536 [6:15:36<4:55:49,  1.64it/s]
 56%|█████▌    | 36458/65536 [6:15:37<4:55:29,  1.64it/s]
 56%|█████▌    | 36459/65536 [6:15:38<5:05:26,  1.59it/s]
 56%|█████▌    | 36460/65536 [6:15:38<5:01:21,  1.61it/s]
                                                         
{'loss': 1.5998, 'learning_rate': 5.005203869229827e-07, 'epoch': 2250.62}
+
 56%|█████▌    | 36460/65536 [6:15:38<5:01:21,  1.61it/s]
 56%|█████▌    | 36461/65536 [6:15:39<5:04:33,  1.59it/s]
 56%|█████▌    | 36462/65536 [6:15:40<5:00:48,  1.61it/s]
 56%|█████▌    | 36463/65536 [6:15:40<5:02:43,  1.60it/s]
 56%|█████▌    | 36464/65536 [6:15:41<4:58:48,  1.62it/s]
 56%|█████▌    | 36465/65536 [6:15:41<4:53:40,  1.65it/s]
 56%|█████▌    | 36466/65536 [6:15:42<4:48:42,  1.68it/s]
 56%|█████▌    | 36467/65536 [6:15:43<4:53:17,  1.65it/s]
 56%|█████▌    | 36468/65536 [6:15:43<4:56:51,  1.63it/s]
 56%|█████▌    | 36469/65536 [6:15:44<4:52:39,  1.66it/s]
 56%|█████▌    | 36470/65536 [6:15:44<4:55:05,  1.64it/s]
 56%|█████▌    | 36471/65536 [6:15:45<4:53:44,  1.65it/s]
 56%|█████▌    | 36472/65536 [6:15:46<4:53:29,  1.65it/s]
 56%|█████▌    | 36473/65536 [6:15:46<5:00:34,  1.61it/s]
 56%|█████▌    | 36474/65536 [6:15:47<4:55:13,  1.64it/s]
 56%|█████▌    | 36475/65536 [6:15:47<4:54:12,  1.65it/s]
 56%|█████▌    | 36476/65536 [6:15:48<4:52:55,  1.65it/s]
 56%|█████▌    | 36477/65536 [6:15:49<4:54:15,  1.65it/s]
 56%|█████▌    | 36478/65536 [6:15:49<4:52:00,  1.66it/s]
 56%|█████▌    | 36479/65536 [6:15:50<4:40:50,  1.72it/s]
 56%|█████▌    | 36480/65536 [6:15:50<4:45:53,  1.69it/s]
                                                         
{'loss': 1.6576, 'learning_rate': 5.002448879637566e-07, 'epoch': 2251.85}
+
 56%|█████▌    | 36480/65536 [6:15:50<4:45:53,  1.69it/s]
 56%|█████▌    | 36481/65536 [6:15:51<4:49:25,  1.67it/s]
 56%|█████▌    | 36482/65536 [6:15:52<5:04:42,  1.59it/s]
 56%|█████▌    | 36483/65536 [6:15:52<5:16:15,  1.53it/s]
 56%|█████▌    | 36484/65536 [6:15:53<5:12:27,  1.55it/s]
 56%|█████▌    | 36485/65536 [6:15:54<5:05:23,  1.59it/s]
 56%|█████▌    | 36486/65536 [6:15:54<4:59:54,  1.61it/s]
 56%|█████▌    | 36487/65536 [6:15:55<5:01:12,  1.61it/s]
 56%|█████▌    | 36488/65536 [6:15:56<4:59:23,  1.62it/s]
 56%|█████▌    | 36489/65536 [6:15:56<5:03:33,  1.59it/s]
 56%|█████▌    | 36490/65536 [6:15:57<5:00:06,  1.61it/s]
 56%|█████▌    | 36491/65536 [6:15:57<4:56:32,  1.63it/s]
 56%|█████▌    | 36492/65536 [6:15:58<4:51:33,  1.66it/s]
 56%|█████▌    | 36493/65536 [6:15:59<4:50:17,  1.67it/s]
 56%|█████▌    | 36494/65536 [6:15:59<4:53:26,  1.65it/s]
 56%|█████▌    | 36495/65536 [6:16:00<4:47:26,  1.68it/s]
 56%|█████▌    | 36496/65536 [6:16:00<4:49:13,  1.67it/s]
 56%|█████▌    | 36497/65536 [6:16:01<4:51:56,  1.66it/s]
 56%|█████▌    | 36498/65536 [6:16:02<5:01:34,  1.60it/s]
 56%|█████▌    | 36499/65536 [6:16:02<5:14:00,  1.54it/s]
 56%|█████▌    | 36500/65536 [6:16:03<5:06:01,  1.58it/s]
                                                         
{'loss': 1.6271, 'learning_rate': 4.999693890045305e-07, 'epoch': 2253.09}
+
 56%|█████▌    | 36500/65536 [6:16:03<5:06:01,  1.58it/s]
 56%|█████▌    | 36501/65536 [6:16:04<5:05:27,  1.58it/s]
 56%|█████▌    | 36502/65536 [6:16:04<5:05:23,  1.58it/s]
 56%|█████▌    | 36503/65536 [6:16:05<4:54:23,  1.64it/s]
 56%|█████▌    | 36504/65536 [6:16:05<4:52:40,  1.65it/s]
 56%|█████▌    | 36505/65536 [6:16:06<4:58:33,  1.62it/s]
 56%|█████▌    | 36506/65536 [6:16:07<4:52:34,  1.65it/s]
 56%|█████▌    | 36507/65536 [6:16:07<4:57:09,  1.63it/s]
 56%|█████▌    | 36508/65536 [6:16:08<4:58:20,  1.62it/s]
 56%|█████▌    | 36509/65536 [6:16:08<4:55:07,  1.64it/s]
 56%|█████▌    | 36510/65536 [6:16:09<4:51:26,  1.66it/s]
 56%|█████▌    | 36511/65536 [6:16:10<4:47:54,  1.68it/s]
 56%|█████▌    | 36512/65536 [6:16:10<4:51:22,  1.66it/s]
 56%|█████▌    | 36513/65536 [6:16:11<4:54:25,  1.64it/s]
 56%|█████▌    | 36514/65536 [6:16:11<4:57:34,  1.63it/s]
 56%|█████▌    | 36515/65536 [6:16:12<5:04:14,  1.59it/s]
 56%|█████▌    | 36516/65536 [6:16:13<5:07:22,  1.57it/s]
 56%|█████▌    | 36517/65536 [6:16:13<5:05:15,  1.58it/s]
 56%|█████▌    | 36518/65536 [6:16:14<4:56:37,  1.63it/s]
 56%|█████▌    | 36519/65536 [6:16:15<5:04:59,  1.59it/s]
 56%|█████▌    | 36520/65536 [6:16:15<5:04:36,  1.59it/s]
                                                         
{'loss': 1.6348, 'learning_rate': 4.996938900453043e-07, 'epoch': 2254.32}
+
 56%|█████▌    | 36520/65536 [6:16:15<5:04:36,  1.59it/s]
 56%|█████▌    | 36521/65536 [6:16:16<5:04:43,  1.59it/s]
 56%|█████▌    | 36522/65536 [6:16:16<5:01:52,  1.60it/s]
 56%|█████▌    | 36523/65536 [6:16:17<5:02:20,  1.60it/s]
 56%|█████▌    | 36524/65536 [6:16:18<4:57:24,  1.63it/s]
 56%|█████▌    | 36525/65536 [6:16:18<4:50:42,  1.66it/s]
 56%|█████▌    | 36526/65536 [6:16:19<4:48:28,  1.68it/s]
 56%|█████▌    | 36527/65536 [6:16:19<4:49:56,  1.67it/s]
 56%|█████▌    | 36528/65536 [6:16:20<4:50:55,  1.66it/s]
 56%|█████▌    | 36529/65536 [6:16:21<4:46:14,  1.69it/s]
 56%|█████▌    | 36530/65536 [6:16:21<4:44:38,  1.70it/s]
 56%|█████▌    | 36531/65536 [6:16:22<4:44:48,  1.70it/s]
 56%|█████▌    | 36532/65536 [6:16:22<4:56:25,  1.63it/s]
 56%|█████▌    | 36533/65536 [6:16:23<4:56:25,  1.63it/s]
 56%|█████▌    | 36534/65536 [6:16:24<4:59:05,  1.62it/s]
 56%|█████▌    | 36535/65536 [6:16:24<4:50:01,  1.67it/s]
 56%|█████▌    | 36536/65536 [6:16:25<4:50:45,  1.66it/s]
 56%|█████▌    | 36537/65536 [6:16:25<4:52:24,  1.65it/s]
 56%|█████▌    | 36538/65536 [6:16:26<4:50:55,  1.66it/s]
 56%|█████▌    | 36539/65536 [6:16:27<4:52:27,  1.65it/s]
 56%|█████▌    | 36540/65536 [6:16:27<4:50:42,  1.66it/s]
                                                         
{'loss': 1.6658, 'learning_rate': 4.994183910860781e-07, 'epoch': 2255.56}
+
 56%|█████▌    | 36540/65536 [6:16:27<4:50:42,  1.66it/s]
 56%|█████▌    | 36541/65536 [6:16:28<4:58:46,  1.62it/s]
 56%|█████▌    | 36542/65536 [6:16:29<4:59:54,  1.61it/s]
 56%|█████▌    | 36543/65536 [6:16:29<5:12:49,  1.54it/s]
 56%|█████▌    | 36544/65536 [6:16:30<5:06:44,  1.58it/s]
 56%|█████▌    | 36545/65536 [6:16:31<5:03:56,  1.59it/s]
 56%|█████▌    | 36546/65536 [6:16:31<5:00:18,  1.61it/s]
 56%|█████▌    | 36547/65536 [6:16:32<4:53:51,  1.64it/s]
 56%|█████▌    | 36548/65536 [6:16:32<5:05:38,  1.58it/s]
 56%|█████▌    | 36549/65536 [6:16:33<5:03:59,  1.59it/s]
 56%|█████▌    | 36550/65536 [6:16:34<4:59:49,  1.61it/s]
 56%|█████▌    | 36551/65536 [6:16:34<4:57:15,  1.63it/s]
 56%|█████▌    | 36552/65536 [6:16:35<5:02:52,  1.59it/s]
 56%|█████▌    | 36553/65536 [6:16:35<5:04:01,  1.59it/s]
 56%|█████▌    | 36554/65536 [6:16:36<5:00:54,  1.61it/s]
 56%|█████▌    | 36555/65536 [6:16:37<4:56:56,  1.63it/s]
 56%|█████▌    | 36556/65536 [6:16:37<5:04:16,  1.59it/s]
 56%|█████▌    | 36557/65536 [6:16:38<5:00:16,  1.61it/s]
 56%|█████▌    | 36558/65536 [6:16:39<4:51:39,  1.66it/s]
 56%|█████▌    | 36559/65536 [6:16:39<4:51:44,  1.66it/s]
 56%|█████▌    | 36560/65536 [6:16:40<4:47:27,  1.68it/s]
                                                         
{'loss': 1.5881, 'learning_rate': 4.991428921268519e-07, 'epoch': 2256.79}
+
 56%|█████▌    | 36560/65536 [6:16:40<4:47:27,  1.68it/s]
 56%|█████▌    | 36561/65536 [6:16:40<4:56:09,  1.63it/s]
 56%|█████▌    | 36562/65536 [6:16:41<4:53:59,  1.64it/s]
 56%|█████▌    | 36563/65536 [6:16:42<5:01:26,  1.60it/s]
 56%|█████▌    | 36564/65536 [6:16:42<5:08:48,  1.56it/s]
 56%|█████▌    | 36565/65536 [6:16:43<5:01:13,  1.60it/s]
 56%|█████▌    | 36566/65536 [6:16:43<4:53:36,  1.64it/s]
 56%|█████▌    | 36567/65536 [6:16:44<4:51:46,  1.65it/s]
 56%|█████▌    | 36568/65536 [6:16:45<4:53:10,  1.65it/s]
 56%|█████▌    | 36569/65536 [6:16:45<4:50:58,  1.66it/s]
 56%|█████▌    | 36570/65536 [6:16:46<4:56:37,  1.63it/s]
 56%|█████▌    | 36571/65536 [6:16:47<4:56:05,  1.63it/s]
 56%|█████▌    | 36572/65536 [6:16:47<4:59:52,  1.61it/s]
 56%|█████▌    | 36573/65536 [6:16:48<5:00:03,  1.61it/s]
 56%|█████▌    | 36574/65536 [6:16:48<4:52:17,  1.65it/s]
 56%|█████▌    | 36575/65536 [6:16:49<4:50:40,  1.66it/s]
 56%|█████▌    | 36576/65536 [6:16:50<4:58:11,  1.62it/s]
 56%|█████▌    | 36577/65536 [6:16:50<4:57:13,  1.62it/s]
 56%|█████▌    | 36578/65536 [6:16:51<5:02:35,  1.60it/s]
 56%|█████▌    | 36579/65536 [6:16:51<4:55:48,  1.63it/s]
 56%|█████▌    | 36580/65536 [6:16:52<5:01:25,  1.60it/s]
                                                         
{'loss': 1.6299, 'learning_rate': 4.988673931676258e-07, 'epoch': 2258.02}
+
 56%|█████▌    | 36580/65536 [6:16:52<5:01:25,  1.60it/s]
 56%|█████▌    | 36581/65536 [6:16:53<4:55:37,  1.63it/s]
 56%|█████▌    | 36582/65536 [6:16:53<4:55:10,  1.63it/s]
 56%|█████▌    | 36583/65536 [6:16:54<5:00:49,  1.60it/s]
 56%|█████▌    | 36584/65536 [6:16:55<4:58:16,  1.62it/s]
 56%|█████▌    | 36585/65536 [6:16:55<5:01:16,  1.60it/s]
 56%|█████▌    | 36586/65536 [6:16:56<5:00:06,  1.61it/s]
 56%|█████▌    | 36587/65536 [6:16:56<4:50:07,  1.66it/s]
 56%|█████▌    | 36588/65536 [6:16:57<5:00:37,  1.60it/s]
 56%|█████▌    | 36589/65536 [6:16:58<4:52:45,  1.65it/s]
 56%|█████▌    | 36590/65536 [6:16:58<4:44:41,  1.69it/s]
 56%|█████▌    | 36591/65536 [6:16:59<4:49:07,  1.67it/s]
 56%|█████▌    | 36592/65536 [6:16:59<4:50:04,  1.66it/s]
 56%|█████▌    | 36593/65536 [6:17:00<4:53:56,  1.64it/s]
 56%|█████▌    | 36594/65536 [6:17:01<4:56:50,  1.63it/s]
 56%|█████▌    | 36595/65536 [6:17:01<5:06:59,  1.57it/s]
 56%|█████▌    | 36596/65536 [6:17:02<5:23:10,  1.49it/s]
 56%|█████▌    | 36597/65536 [6:17:03<5:15:23,  1.53it/s]
 56%|█████▌    | 36598/65536 [6:17:03<5:13:11,  1.54it/s]
 56%|█████▌    | 36599/65536 [6:17:04<5:15:31,  1.53it/s]
 56%|█████▌    | 36600/65536 [6:17:05<5:08:34,  1.56it/s]
                                                         
{'loss': 1.6025, 'learning_rate': 4.985918942083997e-07, 'epoch': 2259.26}
+
 56%|█████▌    | 36600/65536 [6:17:05<5:08:34,  1.56it/s]
 56%|█████▌    | 36601/65536 [6:17:05<5:04:25,  1.58it/s]
 56%|█████▌    | 36602/65536 [6:17:06<5:00:54,  1.60it/s]
 56%|█████▌    | 36603/65536 [6:17:06<4:57:24,  1.62it/s]
 56%|█████▌    | 36604/65536 [6:17:07<4:54:11,  1.64it/s]
 56%|█████▌    | 36605/65536 [6:17:08<4:53:03,  1.65it/s]
 56%|█████▌    | 36606/65536 [6:17:08<4:59:59,  1.61it/s]
 56%|█████▌    | 36607/65536 [6:17:09<5:02:11,  1.60it/s]
 56%|█████▌    | 36608/65536 [6:17:09<4:57:23,  1.62it/s]
 56%|█████▌    | 36609/65536 [6:17:10<4:53:59,  1.64it/s]
 56%|█████▌    | 36610/65536 [6:17:11<4:49:17,  1.67it/s]
 56%|█████▌    | 36611/65536 [6:17:11<4:44:08,  1.70it/s]
 56%|█████▌    | 36612/65536 [6:17:12<4:40:38,  1.72it/s]
 56%|█████▌    | 36613/65536 [6:17:12<4:57:46,  1.62it/s]
 56%|█████▌    | 36614/65536 [6:17:13<4:58:19,  1.62it/s]
 56%|█████▌    | 36615/65536 [6:17:14<4:50:13,  1.66it/s]
 56%|█████▌    | 36616/65536 [6:17:14<4:50:03,  1.66it/s]
 56%|█████▌    | 36617/65536 [6:17:15<4:50:57,  1.66it/s]
 56%|█████▌    | 36618/65536 [6:17:15<4:51:27,  1.65it/s]
 56%|█████▌    | 36619/65536 [6:17:16<4:54:36,  1.64it/s]
 56%|█████▌    | 36620/65536 [6:17:17<4:51:27,  1.65it/s]
                                                         
{'loss': 1.6725, 'learning_rate': 4.983163952491735e-07, 'epoch': 2260.49}
+
 56%|█████▌    | 36620/65536 [6:17:17<4:51:27,  1.65it/s]
 56%|█████▌    | 36621/65536 [6:17:17<4:47:49,  1.67it/s]
 56%|█████▌    | 36622/65536 [6:17:18<4:56:01,  1.63it/s]
 56%|█████▌    | 36623/65536 [6:17:19<4:54:21,  1.64it/s]
 56%|█████▌    | 36624/65536 [6:17:19<4:48:40,  1.67it/s]
 56%|█████▌    | 36625/65536 [6:17:20<4:50:40,  1.66it/s]
 56%|█████▌    | 36626/65536 [6:17:20<4:58:14,  1.62it/s]
 56%|█████▌    | 36627/65536 [6:17:21<5:03:02,  1.59it/s]
 56%|█████▌    | 36628/65536 [6:17:22<4:57:39,  1.62it/s]
 56%|█████▌    | 36629/65536 [6:17:22<5:01:36,  1.60it/s]
 56%|█████▌    | 36630/65536 [6:17:23<4:56:08,  1.63it/s]
 56%|█████▌    | 36631/65536 [6:17:24<4:59:43,  1.61it/s]
 56%|█████▌    | 36632/65536 [6:17:24<4:58:51,  1.61it/s]
 56%|█████▌    | 36633/65536 [6:17:25<4:52:32,  1.65it/s]
 56%|█████▌    | 36634/65536 [6:17:25<4:55:38,  1.63it/s]
 56%|█████▌    | 36635/65536 [6:17:26<4:58:24,  1.61it/s]
 56%|█████▌    | 36636/65536 [6:17:27<4:54:20,  1.64it/s]
 56%|█████▌    | 36637/65536 [6:17:27<4:52:35,  1.65it/s]
 56%|█████▌    | 36638/65536 [6:17:28<4:55:34,  1.63it/s]
 56%|█████▌    | 36639/65536 [6:17:28<4:47:54,  1.67it/s]
 56%|█████▌    | 36640/65536 [6:17:29<4:52:57,  1.64it/s]
                                                         
{'loss': 1.6149, 'learning_rate': 4.980408962899474e-07, 'epoch': 2261.73}
+
 56%|█████▌    | 36640/65536 [6:17:29<4:52:57,  1.64it/s]
 56%|█████▌    | 36641/65536 [6:17:30<4:52:46,  1.64it/s]
 56%|█████▌    | 36642/65536 [6:17:30<4:50:59,  1.65it/s]
 56%|█████▌    | 36643/65536 [6:17:31<4:59:28,  1.61it/s]
 56%|█████▌    | 36644/65536 [6:17:31<5:02:56,  1.59it/s]
 56%|█████▌    | 36645/65536 [6:17:32<5:07:14,  1.57it/s]
 56%|█████▌    | 36646/65536 [6:17:33<5:21:42,  1.50it/s]
 56%|█████▌    | 36647/65536 [6:17:34<5:29:35,  1.46it/s]
 56%|█████▌    | 36648/65536 [6:17:34<5:24:11,  1.49it/s]
 56%|█████▌    | 36649/65536 [6:17:35<5:27:19,  1.47it/s]
 56%|█████▌    | 36650/65536 [6:17:36<5:20:00,  1.50it/s]
 56%|█████▌    | 36651/65536 [6:17:36<5:07:23,  1.57it/s]
 56%|█████▌    | 36652/65536 [6:17:37<5:00:27,  1.60it/s]
 56%|█████▌    | 36653/65536 [6:17:37<4:52:17,  1.65it/s]
 56%|█████▌    | 36654/65536 [6:17:38<4:47:47,  1.67it/s]
 56%|█████▌    | 36655/65536 [6:17:38<4:44:36,  1.69it/s]
 56%|█████▌    | 36656/65536 [6:17:39<4:45:20,  1.69it/s]
 56%|█████▌    | 36657/65536 [6:17:40<4:45:07,  1.69it/s]
 56%|█████▌    | 36658/65536 [6:17:40<4:45:34,  1.69it/s]
 56%|█████▌    | 36659/65536 [6:17:41<4:50:21,  1.66it/s]
 56%|█████▌    | 36660/65536 [6:17:42<4:57:19,  1.62it/s]
                                                         
{'loss': 1.6358, 'learning_rate': 4.977653973307211e-07, 'epoch': 2262.96}
+
 56%|█████▌    | 36660/65536 [6:17:42<4:57:19,  1.62it/s]
 56%|█████▌    | 36661/65536 [6:17:42<5:03:32,  1.59it/s]
 56%|█████▌    | 36662/65536 [6:17:43<5:00:55,  1.60it/s]
 56%|█████▌    | 36663/65536 [6:17:43<4:59:25,  1.61it/s]
 56%|█████▌    | 36664/65536 [6:17:44<4:54:33,  1.63it/s]
 56%|█████▌    | 36665/65536 [6:17:45<4:58:27,  1.61it/s]
 56%|█████▌    | 36666/65536 [6:17:45<4:57:10,  1.62it/s]
 56%|█████▌    | 36667/65536 [6:17:46<4:58:55,  1.61it/s]
 56%|█████▌    | 36668/65536 [6:17:46<4:53:26,  1.64it/s]
 56%|█████▌    | 36669/65536 [6:17:47<4:51:36,  1.65it/s]
 56%|█████▌    | 36670/65536 [6:17:48<4:50:19,  1.66it/s]
 56%|█████▌    | 36671/65536 [6:17:48<4:48:54,  1.67it/s]
 56%|█████▌    | 36672/65536 [6:17:49<4:52:42,  1.64it/s]
 56%|█████▌    | 36673/65536 [6:17:50<4:56:55,  1.62it/s]
 56%|█████▌    | 36674/65536 [6:17:50<4:59:26,  1.61it/s]
 56%|█████▌    | 36675/65536 [6:17:51<4:58:47,  1.61it/s]
 56%|█████▌    | 36676/65536 [6:17:51<4:56:47,  1.62it/s]
 56%|█████▌    | 36677/65536 [6:17:52<5:06:53,  1.57it/s]
 56%|█████▌    | 36678/65536 [6:17:53<5:05:21,  1.58it/s]
 56%|█████▌    | 36679/65536 [6:17:53<5:09:56,  1.55it/s]
 56%|█████▌    | 36680/65536 [6:17:54<4:57:43,  1.62it/s]
                                                         
{'loss': 1.6036, 'learning_rate': 4.97489898371495e-07, 'epoch': 2264.2}
+
 56%|█████▌    | 36680/65536 [6:17:54<4:57:43,  1.62it/s]
 56%|█████▌    | 36681/65536 [6:17:55<4:55:26,  1.63it/s]
 56%|█████▌    | 36682/65536 [6:17:55<4:51:24,  1.65it/s]
 56%|█████▌    | 36683/65536 [6:17:56<4:53:58,  1.64it/s]
 56%|█████▌    | 36684/65536 [6:17:56<4:50:45,  1.65it/s]
 56%|█████▌    | 36685/65536 [6:17:57<4:58:12,  1.61it/s]
 56%|█████▌    | 36686/65536 [6:17:58<4:53:05,  1.64it/s]
 56%|█████▌    | 36687/65536 [6:17:58<4:48:41,  1.67it/s]
 56%|█████▌    | 36688/65536 [6:17:59<4:52:27,  1.64it/s]
 56%|█████▌    | 36689/65536 [6:17:59<4:54:23,  1.63it/s]
 56%|█████▌    | 36690/65536 [6:18:00<4:50:17,  1.66it/s]
 56%|█████▌    | 36691/65536 [6:18:01<4:46:02,  1.68it/s]
 56%|█████▌    | 36692/65536 [6:18:01<4:46:16,  1.68it/s]
 56%|█████▌    | 36693/65536 [6:18:02<4:56:20,  1.62it/s]
 56%|█████▌    | 36694/65536 [6:18:03<5:10:28,  1.55it/s]
 56%|█████▌    | 36695/65536 [6:18:03<5:02:56,  1.59it/s]
 56%|█████▌    | 36696/65536 [6:18:04<5:03:43,  1.58it/s]
 56%|█████▌    | 36697/65536 [6:18:04<5:01:39,  1.59it/s]
 56%|█████▌    | 36698/65536 [6:18:05<4:55:54,  1.62it/s]
 56%|█████▌    | 36699/65536 [6:18:06<4:51:05,  1.65it/s]
 56%|█████▌    | 36700/65536 [6:18:06<4:51:15,  1.65it/s]
                                                         
{'loss': 1.6184, 'learning_rate': 4.972143994122688e-07, 'epoch': 2265.43}
+
 56%|█████▌    | 36700/65536 [6:18:06<4:51:15,  1.65it/s]
 56%|█████▌    | 36701/65536 [6:18:07<4:50:09,  1.66it/s]
 56%|█████▌    | 36702/65536 [6:18:07<4:50:26,  1.65it/s]
 56%|█████▌    | 36703/65536 [6:18:08<4:53:19,  1.64it/s]
 56%|█████▌    | 36704/65536 [6:18:09<4:51:36,  1.65it/s]
 56%|█████▌    | 36705/65536 [6:18:09<4:57:09,  1.62it/s]
 56%|█████▌    | 36706/65536 [6:18:10<5:00:28,  1.60it/s]
 56%|█████▌    | 36707/65536 [6:18:11<5:04:04,  1.58it/s]
 56%|█████▌    | 36708/65536 [6:18:11<4:58:13,  1.61it/s]
 56%|█████▌    | 36709/65536 [6:18:12<4:58:59,  1.61it/s]
 56%|█████▌    | 36710/65536 [6:18:12<5:05:01,  1.58it/s]
 56%|█████▌    | 36711/65536 [6:18:13<4:59:15,  1.61it/s]
 56%|█████▌    | 36712/65536 [6:18:14<4:54:05,  1.63it/s]
 56%|█████▌    | 36713/65536 [6:18:14<4:55:21,  1.63it/s]
 56%|█████▌    | 36714/65536 [6:18:15<4:47:51,  1.67it/s]
 56%|█████▌    | 36715/65536 [6:18:15<4:51:04,  1.65it/s]
 56%|█████▌    | 36716/65536 [6:18:16<4:54:18,  1.63it/s]
 56%|█████▌    | 36717/65536 [6:18:17<5:01:35,  1.59it/s]
 56%|█████▌    | 36718/65536 [6:18:17<5:07:26,  1.56it/s]
 56%|█████▌    | 36719/65536 [6:18:18<4:59:29,  1.60it/s]
 56%|█████▌    | 36720/65536 [6:18:19<4:57:14,  1.62it/s]
                                                         
{'loss': 1.6165, 'learning_rate': 4.969389004530427e-07, 'epoch': 2266.67}
+
 56%|█████▌    | 36720/65536 [6:18:19<4:57:14,  1.62it/s]
 56%|█████▌    | 36721/65536 [6:18:19<4:54:55,  1.63it/s]
 56%|█████▌    | 36722/65536 [6:18:20<4:52:40,  1.64it/s]
 56%|█████▌    | 36723/65536 [6:18:20<5:02:12,  1.59it/s]
 56%|█████▌    | 36724/65536 [6:18:21<4:54:39,  1.63it/s]
 56%|█████▌    | 36725/65536 [6:18:22<4:48:15,  1.67it/s]
 56%|█████▌    | 36726/65536 [6:18:22<5:06:37,  1.57it/s]
 56%|█████▌    | 36727/65536 [6:18:23<5:00:28,  1.60it/s]
 56%|█████▌    | 36728/65536 [6:18:24<5:08:15,  1.56it/s]
 56%|█████▌    | 36729/65536 [6:18:24<5:08:27,  1.56it/s]
 56%|█████▌    | 36730/65536 [6:18:25<4:57:16,  1.62it/s]
 56%|█████▌    | 36731/65536 [6:18:25<4:51:23,  1.65it/s]
 56%|█████▌    | 36732/65536 [6:18:26<4:52:09,  1.64it/s]
 56%|█████▌    | 36733/65536 [6:18:27<4:45:20,  1.68it/s]
 56%|█████▌    | 36734/65536 [6:18:27<4:44:30,  1.69it/s]
 56%|█████▌    | 36735/65536 [6:18:28<4:48:08,  1.67it/s]
 56%|█████▌    | 36736/65536 [6:18:28<4:50:04,  1.65it/s]
 56%|█████▌    | 36737/65536 [6:18:29<4:45:37,  1.68it/s]
 56%|█████▌    | 36738/65536 [6:18:30<4:58:36,  1.61it/s]
 56%|█████▌    | 36739/65536 [6:18:30<4:53:08,  1.64it/s]
 56%|█████▌    | 36740/65536 [6:18:31<4:51:34,  1.65it/s]
                                                         
{'loss': 1.6563, 'learning_rate': 4.966634014938166e-07, 'epoch': 2267.9}
+
 56%|█████▌    | 36740/65536 [6:18:31<4:51:34,  1.65it/s]
 56%|█████▌    | 36741/65536 [6:18:31<4:48:34,  1.66it/s]
 56%|█████▌    | 36742/65536 [6:18:32<4:59:02,  1.60it/s]
 56%|█████▌    | 36743/65536 [6:18:33<4:56:24,  1.62it/s]
 56%|█████▌    | 36744/65536 [6:18:33<4:57:18,  1.61it/s]
 56%|█████▌    | 36745/65536 [6:18:34<4:54:02,  1.63it/s]
 56%|█████▌    | 36746/65536 [6:18:34<4:45:24,  1.68it/s]
 56%|█████▌    | 36747/65536 [6:18:35<4:48:40,  1.66it/s]
 56%|█████▌    | 36748/65536 [6:18:36<4:50:13,  1.65it/s]
 56%|█████▌    | 36749/65536 [6:18:36<4:53:03,  1.64it/s]
 56%|█████▌    | 36750/65536 [6:18:37<4:52:08,  1.64it/s]
 56%|█████▌    | 36751/65536 [6:18:38<4:54:31,  1.63it/s]
 56%|█████▌    | 36752/65536 [6:18:38<4:48:02,  1.67it/s]
 56%|█████▌    | 36753/65536 [6:18:39<5:04:43,  1.57it/s]
 56%|█████▌    | 36754/65536 [6:18:39<4:55:58,  1.62it/s]
 56%|█████▌    | 36755/65536 [6:18:40<5:01:16,  1.59it/s]
 56%|█████▌    | 36756/65536 [6:18:41<5:01:41,  1.59it/s]
 56%|█████▌    | 36757/65536 [6:18:41<4:59:17,  1.60it/s]
 56%|█████▌    | 36758/65536 [6:18:42<5:06:57,  1.56it/s]
 56%|█████▌    | 36759/65536 [6:18:43<5:00:03,  1.60it/s]
 56%|█████▌    | 36760/65536 [6:18:43<4:58:03,  1.61it/s]
                                                         
{'loss': 1.6184, 'learning_rate': 4.963879025345904e-07, 'epoch': 2269.14}
+
 56%|█████▌    | 36760/65536 [6:18:43<4:58:03,  1.61it/s]
 56%|█████▌    | 36761/65536 [6:18:44<4:51:49,  1.64it/s]
 56%|█████▌    | 36762/65536 [6:18:44<4:54:08,  1.63it/s]
 56%|█████▌    | 36763/65536 [6:18:45<4:49:11,  1.66it/s]
 56%|█████▌    | 36764/65536 [6:18:46<4:46:32,  1.67it/s]
 56%|█████▌    | 36765/65536 [6:18:46<4:41:39,  1.70it/s]
 56%|█████▌    | 36766/65536 [6:18:47<4:48:59,  1.66it/s]
 56%|█████▌    | 36767/65536 [6:18:47<4:52:26,  1.64it/s]
 56%|█████▌    | 36768/65536 [6:18:48<4:58:24,  1.61it/s]
 56%|█████▌    | 36769/65536 [6:18:49<4:52:31,  1.64it/s]
 56%|█████▌    | 36770/65536 [6:18:49<4:56:37,  1.62it/s]
 56%|█████▌    | 36771/65536 [6:18:50<4:57:03,  1.61it/s]
 56%|█████▌    | 36772/65536 [6:18:50<4:58:04,  1.61it/s]
 56%|█████▌    | 36773/65536 [6:18:51<5:02:22,  1.59it/s]
 56%|█████▌    | 36774/65536 [6:18:52<4:58:39,  1.61it/s]
 56%|█████▌    | 36775/65536 [6:18:52<5:11:40,  1.54it/s]
 56%|█████▌    | 36776/65536 [6:18:53<4:59:30,  1.60it/s]
 56%|█████▌    | 36777/65536 [6:18:54<4:54:27,  1.63it/s]
 56%|█████▌    | 36778/65536 [6:18:54<4:44:22,  1.69it/s]
 56%|█████▌    | 36779/65536 [6:18:55<4:44:34,  1.68it/s]
 56%|█████▌    | 36780/65536 [6:18:55<4:53:27,  1.63it/s]
                                                         
{'loss': 1.6589, 'learning_rate': 4.961124035753643e-07, 'epoch': 2270.37}
+
 56%|█████▌    | 36780/65536 [6:18:55<4:53:27,  1.63it/s]
 56%|█████▌    | 36781/65536 [6:18:56<4:48:01,  1.66it/s]
 56%|█████▌    | 36782/65536 [6:18:57<4:45:36,  1.68it/s]
 56%|█████▌    | 36783/65536 [6:18:57<4:51:12,  1.65it/s]
 56%|█████▌    | 36784/65536 [6:18:58<4:46:38,  1.67it/s]
 56%|█████▌    | 36785/65536 [6:18:58<4:54:07,  1.63it/s]
 56%|█████▌    | 36786/65536 [6:18:59<4:55:41,  1.62it/s]
 56%|█████▌    | 36787/65536 [6:19:00<4:56:57,  1.61it/s]
 56%|█████▌    | 36788/65536 [6:19:00<4:51:06,  1.65it/s]
 56%|█████▌    | 36789/65536 [6:19:01<4:51:36,  1.64it/s]
 56%|█████▌    | 36790/65536 [6:19:01<4:51:46,  1.64it/s]
 56%|█████▌    | 36791/65536 [6:19:02<5:08:09,  1.55it/s]
 56%|█████▌    | 36792/65536 [6:19:03<5:04:18,  1.57it/s]
 56%|█████▌    | 36793/65536 [6:19:03<4:55:43,  1.62it/s]
 56%|█████▌    | 36794/65536 [6:19:04<4:52:09,  1.64it/s]
 56%|█████▌    | 36795/65536 [6:19:05<4:52:46,  1.64it/s]
 56%|█████▌    | 36796/65536 [6:19:05<4:51:31,  1.64it/s]
 56%|█████▌    | 36797/65536 [6:19:06<4:50:55,  1.65it/s]
 56%|█████▌    | 36798/65536 [6:19:06<4:56:55,  1.61it/s]
 56%|█████▌    | 36799/65536 [6:19:07<4:57:34,  1.61it/s]
 56%|█████▌    | 36800/65536 [6:19:08<5:09:02,  1.55it/s]
                                                         
{'loss': 1.6132, 'learning_rate': 4.95836904616138e-07, 'epoch': 2271.6}
+
 56%|█████▌    | 36800/65536 [6:19:08<5:09:02,  1.55it/s]
 56%|█████▌    | 36801/65536 [6:19:08<4:59:34,  1.60it/s]
 56%|█████▌    | 36802/65536 [6:19:09<4:55:17,  1.62it/s]
 56%|█████▌    | 36803/65536 [6:19:10<5:00:48,  1.59it/s]
 56%|█████▌    | 36804/65536 [6:19:10<4:54:12,  1.63it/s]
 56%|█████▌    | 36805/65536 [6:19:11<4:52:13,  1.64it/s]
 56%|█████▌    | 36806/65536 [6:19:11<4:50:52,  1.65it/s]
 56%|█████▌    | 36807/65536 [6:19:12<5:03:08,  1.58it/s]
 56%|█████▌    | 36808/65536 [6:19:13<4:51:56,  1.64it/s]
 56%|█████▌    | 36809/65536 [6:19:13<4:49:49,  1.65it/s]
 56%|█████▌    | 36810/65536 [6:19:14<4:46:29,  1.67it/s]
 56%|█████▌    | 36811/65536 [6:19:14<4:47:49,  1.66it/s]
 56%|█████▌    | 36812/65536 [6:19:15<4:54:25,  1.63it/s]
 56%|█████▌    | 36813/65536 [6:19:16<4:54:19,  1.63it/s]
 56%|█████▌    | 36814/65536 [6:19:16<4:59:40,  1.60it/s]
 56%|█████▌    | 36815/65536 [6:19:17<5:02:09,  1.58it/s]
 56%|█████▌    | 36816/65536 [6:19:18<5:02:52,  1.58it/s]
 56%|█████▌    | 36817/65536 [6:19:18<4:57:46,  1.61it/s]
 56%|█████▌    | 36818/65536 [6:19:19<4:55:56,  1.62it/s]
 56%|█████▌    | 36819/65536 [6:19:19<4:58:31,  1.60it/s]
 56%|█████▌    | 36820/65536 [6:19:20<4:51:15,  1.64it/s]
                                                         
{'loss': 1.6007, 'learning_rate': 4.955614056569119e-07, 'epoch': 2272.84}
+
 56%|█████▌    | 36820/65536 [6:19:20<4:51:15,  1.64it/s]
 56%|█████▌    | 36821/65536 [6:19:21<4:50:21,  1.65it/s]
 56%|█████▌    | 36822/65536 [6:19:21<4:55:06,  1.62it/s]
 56%|█████▌    | 36823/65536 [6:19:22<5:04:07,  1.57it/s]
 56%|█████▌    | 36824/65536 [6:19:23<5:03:50,  1.57it/s]
 56%|█████▌    | 36825/65536 [6:19:23<5:03:35,  1.58it/s]
 56%|█████▌    | 36826/65536 [6:19:24<4:59:49,  1.60it/s]
 56%|█████▌    | 36827/65536 [6:19:24<5:06:49,  1.56it/s]
 56%|█████▌    | 36828/65536 [6:19:25<4:53:46,  1.63it/s]
 56%|█████▌    | 36829/65536 [6:19:26<4:54:18,  1.63it/s]
 56%|█████▌    | 36830/65536 [6:19:26<4:51:21,  1.64it/s]
 56%|█████▌    | 36831/65536 [6:19:27<4:45:42,  1.67it/s]
 56%|█████▌    | 36832/65536 [6:19:27<4:45:44,  1.67it/s]
 56%|█████▌    | 36833/65536 [6:19:28<4:49:25,  1.65it/s]
 56%|█████▌    | 36834/65536 [6:19:29<4:46:53,  1.67it/s]
 56%|█████▌    | 36835/65536 [6:19:29<4:45:59,  1.67it/s]
 56%|█████▌    | 36836/65536 [6:19:30<4:42:32,  1.69it/s]
 56%|█████▌    | 36837/65536 [6:19:30<4:55:40,  1.62it/s]
 56%|█████▌    | 36838/65536 [6:19:31<4:52:51,  1.63it/s]
 56%|█████▌    | 36839/65536 [6:19:32<5:01:10,  1.59it/s]
 56%|█████▌    | 36840/65536 [6:19:32<4:57:57,  1.61it/s]
                                                         
{'loss': 1.6238, 'learning_rate': 4.952859066976857e-07, 'epoch': 2274.07}
+
 56%|█████▌    | 36840/65536 [6:19:32<4:57:57,  1.61it/s]
 56%|█████▌    | 36841/65536 [6:19:33<4:54:29,  1.62it/s]
 56%|█████▌    | 36842/65536 [6:19:34<5:01:51,  1.58it/s]
 56%|█████▌    | 36843/65536 [6:19:34<5:01:46,  1.58it/s]
 56%|█████▌    | 36844/65536 [6:19:35<4:57:25,  1.61it/s]
 56%|█████▌    | 36845/65536 [6:19:35<4:54:12,  1.63it/s]
 56%|█████▌    | 36846/65536 [6:19:36<4:53:20,  1.63it/s]
 56%|█████▌    | 36847/65536 [6:19:37<4:55:36,  1.62it/s]
 56%|█████▌    | 36848/65536 [6:19:37<4:59:15,  1.60it/s]
 56%|█████▌    | 36849/65536 [6:19:38<4:56:39,  1.61it/s]
 56%|█████▌    | 36850/65536 [6:19:39<4:52:48,  1.63it/s]
 56%|█████▌    | 36851/65536 [6:19:39<4:50:34,  1.65it/s]
 56%|█████▌    | 36852/65536 [6:19:40<4:49:01,  1.65it/s]
 56%|█████▌    | 36853/65536 [6:19:40<4:51:14,  1.64it/s]
 56%|█████▌    | 36854/65536 [6:19:41<4:53:32,  1.63it/s]
 56%|█████▌    | 36855/65536 [6:19:42<4:50:28,  1.65it/s]
 56%|█████▌    | 36856/65536 [6:19:42<4:59:11,  1.60it/s]
 56%|█████▌    | 36857/65536 [6:19:43<4:57:23,  1.61it/s]
 56%|█████▌    | 36858/65536 [6:19:43<4:56:46,  1.61it/s]
 56%|█████▌    | 36859/65536 [6:19:44<4:59:58,  1.59it/s]
 56%|█████▌    | 36860/65536 [6:19:45<5:05:23,  1.57it/s]
                                                         
{'loss': 1.6, 'learning_rate': 4.950104077384597e-07, 'epoch': 2275.31}
+
 56%|█████▌    | 36860/65536 [6:19:45<5:05:23,  1.57it/s]
 56%|█████▌    | 36861/65536 [6:19:45<5:04:16,  1.57it/s]
 56%|█████▌    | 36862/65536 [6:19:46<4:57:43,  1.61it/s]
 56%|█████▌    | 36863/65536 [6:19:47<4:54:46,  1.62it/s]
 56%|█████▋    | 36864/65536 [6:19:47<4:53:36,  1.63it/s]
 56%|█████▋    | 36865/65536 [6:19:48<4:54:30,  1.62it/s]
 56%|█████▋    | 36866/65536 [6:19:48<4:55:25,  1.62it/s]
 56%|█████▋    | 36867/65536 [6:19:49<4:54:17,  1.62it/s]
 56%|█████▋    | 36868/65536 [6:19:50<4:50:59,  1.64it/s]
 56%|█████▋    | 36869/65536 [6:19:50<4:50:12,  1.65it/s]
 56%|█████▋    | 36870/65536 [6:19:51<4:56:40,  1.61it/s]
 56%|█████▋    | 36871/65536 [6:19:51<4:48:28,  1.66it/s]
 56%|█████▋    | 36872/65536 [6:19:52<4:58:52,  1.60it/s]
 56%|█████▋    | 36873/65536 [6:19:53<4:54:50,  1.62it/s]
 56%|█████▋    | 36874/65536 [6:19:53<4:53:41,  1.63it/s]
 56%|█████▋    | 36875/65536 [6:19:54<4:52:54,  1.63it/s]
 56%|█████▋    | 36876/65536 [6:19:55<4:50:38,  1.64it/s]
 56%|█████▋    | 36877/65536 [6:19:55<4:54:40,  1.62it/s]
 56%|█████▋    | 36878/65536 [6:19:56<4:59:15,  1.60it/s]
 56%|█████▋    | 36879/65536 [6:19:56<4:57:54,  1.60it/s]
 56%|█████▋    | 36880/65536 [6:19:57<4:50:59,  1.64it/s]
                                                         
{'loss': 1.6028, 'learning_rate': 4.947349087792336e-07, 'epoch': 2276.54}
+
 56%|█████▋    | 36880/65536 [6:19:57<4:50:59,  1.64it/s]
 56%|█████▋    | 36881/65536 [6:19:58<4:48:58,  1.65it/s]
 56%|█████▋    | 36882/65536 [6:19:58<4:42:56,  1.69it/s]
 56%|█████▋    | 36883/65536 [6:19:59<4:40:45,  1.70it/s]
 56%|█████▋    | 36884/65536 [6:19:59<4:34:40,  1.74it/s]
 56%|█████▋    | 36885/65536 [6:20:00<4:36:11,  1.73it/s]
 56%|█████▋    | 36886/65536 [6:20:01<4:52:07,  1.63it/s]
 56%|█████▋    | 36887/65536 [6:20:01<4:51:10,  1.64it/s]
 56%|█████▋    | 36888/65536 [6:20:02<5:06:07,  1.56it/s]
 56%|█████▋    | 36889/65536 [6:20:03<5:02:57,  1.58it/s]
 56%|█████▋    | 36890/65536 [6:20:03<4:55:55,  1.61it/s]
 56%|█████▋    | 36891/65536 [6:20:04<4:52:25,  1.63it/s]
 56%|█████▋    | 36892/65536 [6:20:04<4:50:19,  1.64it/s]
 56%|█████▋    | 36893/65536 [6:20:05<4:49:10,  1.65it/s]
 56%|█████▋    | 36894/65536 [6:20:06<4:55:07,  1.62it/s]
 56%|█████▋    | 36895/65536 [6:20:06<4:45:49,  1.67it/s]
 56%|█████▋    | 36896/65536 [6:20:07<4:41:23,  1.70it/s]
 56%|█████▋    | 36897/65536 [6:20:07<4:45:01,  1.67it/s]
 56%|█████▋    | 36898/65536 [6:20:08<4:45:45,  1.67it/s]
 56%|█████▋    | 36899/65536 [6:20:09<4:47:56,  1.66it/s]
 56%|█████▋    | 36900/65536 [6:20:09<4:54:31,  1.62it/s]
                                                         
{'loss': 1.6577, 'learning_rate': 4.944594098200074e-07, 'epoch': 2277.78}
+
 56%|█████▋    | 36900/65536 [6:20:09<4:54:31,  1.62it/s]
 56%|█████▋    | 36901/65536 [6:20:10<4:49:23,  1.65it/s]
 56%|█████▋    | 36902/65536 [6:20:10<4:55:06,  1.62it/s]
 56%|█████▋    | 36903/65536 [6:20:11<4:56:24,  1.61it/s]
 56%|█████▋    | 36904/65536 [6:20:12<5:04:46,  1.57it/s]
 56%|█████▋    | 36905/65536 [6:20:12<5:02:36,  1.58it/s]
 56%|█████▋    | 36906/65536 [6:20:13<5:08:08,  1.55it/s]
 56%|█████▋    | 36907/65536 [6:20:14<5:00:05,  1.59it/s]
 56%|█████▋    | 36908/65536 [6:20:14<4:51:18,  1.64it/s]
 56%|█████▋    | 36909/65536 [6:20:15<4:44:49,  1.68it/s]
 56%|█████▋    | 36910/65536 [6:20:15<4:40:17,  1.70it/s]
 56%|█████▋    | 36911/65536 [6:20:16<4:42:29,  1.69it/s]
 56%|█████▋    | 36912/65536 [6:20:17<4:47:36,  1.66it/s]
 56%|█████▋    | 36913/65536 [6:20:17<4:48:10,  1.66it/s]
 56%|█████▋    | 36914/65536 [6:20:18<4:42:44,  1.69it/s]
 56%|█████▋    | 36915/65536 [6:20:18<4:42:15,  1.69it/s]
 56%|█████▋    | 36916/65536 [6:20:19<4:46:36,  1.66it/s]
 56%|█████▋    | 36917/65536 [6:20:19<4:44:31,  1.68it/s]
 56%|█████▋    | 36918/65536 [6:20:20<4:50:12,  1.64it/s]
 56%|█████▋    | 36919/65536 [6:20:21<4:58:07,  1.60it/s]
 56%|███��█▋    | 36920/65536 [6:20:21<5:07:54,  1.55it/s]
                                                         
{'loss': 1.6336, 'learning_rate': 4.941839108607813e-07, 'epoch': 2279.01}
+
 56%|█████▋    | 36920/65536 [6:20:21<5:07:54,  1.55it/s]
 56%|█████▋    | 36921/65536 [6:20:22<4:59:47,  1.59it/s]
 56%|█████▋    | 36922/65536 [6:20:23<4:51:34,  1.64it/s]
 56%|█████▋    | 36923/65536 [6:20:23<4:48:19,  1.65it/s]
 56%|█████▋    | 36924/65536 [6:20:24<4:56:52,  1.61it/s]
 56%|█████▋    | 36925/65536 [6:20:25<4:56:32,  1.61it/s]
 56%|█████▋    | 36926/65536 [6:20:25<5:00:06,  1.59it/s]
 56%|█████▋    | 36927/65536 [6:20:26<4:55:41,  1.61it/s]
 56%|█████▋    | 36928/65536 [6:20:26<4:50:07,  1.64it/s]
 56%|█████▋    | 36929/65536 [6:20:27<4:59:38,  1.59it/s]
 56%|█████▋    | 36930/65536 [6:20:28<4:56:09,  1.61it/s]
 56%|█████▋    | 36931/65536 [6:20:28<4:54:38,  1.62it/s]
 56%|█████▋    | 36932/65536 [6:20:29<4:50:13,  1.64it/s]
 56%|█████▋    | 36933/65536 [6:20:29<4:50:26,  1.64it/s]
 56%|█████▋    | 36934/65536 [6:20:30<4:48:12,  1.65it/s]
 56%|█████▋    | 36935/65536 [6:20:31<4:41:28,  1.69it/s]
 56%|█████▋    | 36936/65536 [6:20:31<4:44:22,  1.68it/s]
 56%|█████▋    | 36937/65536 [6:20:32<5:01:28,  1.58it/s]
 56%|█████▋    | 36938/65536 [6:20:32<4:54:27,  1.62it/s]
 56%|█████▋    | 36939/65536 [6:20:33<4:55:12,  1.61it/s]
 56%|█████▋    | 36940/65536 [6:20:34<4:52:15,  1.63it/s]
                                                         
{'loss': 1.6477, 'learning_rate': 4.939084119015549e-07, 'epoch': 2280.25}
+
 56%|█████▋    | 36940/65536 [6:20:34<4:52:15,  1.63it/s]
 56%|█████▋    | 36941/65536 [6:20:34<4:48:33,  1.65it/s]
 56%|█████▋    | 36942/65536 [6:20:35<4:45:11,  1.67it/s]
 56%|█████▋    | 36943/65536 [6:20:36<4:48:41,  1.65it/s]
 56%|█████▋    | 36944/65536 [6:20:36<4:51:35,  1.63it/s]
 56%|█████▋    | 36945/65536 [6:20:37<4:42:31,  1.69it/s]
 56%|█████▋    | 36946/65536 [6:20:37<4:45:02,  1.67it/s]
 56%|█████▋    | 36947/65536 [6:20:38<4:45:14,  1.67it/s]
 56%|█████▋    | 36948/65536 [6:20:39<4:48:38,  1.65it/s]
 56%|█████▋    | 36949/65536 [6:20:39<4:50:18,  1.64it/s]
 56%|█████▋    | 36950/65536 [6:20:40<4:53:22,  1.62it/s]
 56%|█████▋    | 36951/65536 [6:20:40<4:51:33,  1.63it/s]
 56%|█████▋    | 36952/65536 [6:20:41<4:53:38,  1.62it/s]
 56%|█████▋    | 36953/65536 [6:20:42<5:08:55,  1.54it/s]
 56%|█████▋    | 36954/65536 [6:20:42<4:58:23,  1.60it/s]
 56%|█████▋    | 36955/65536 [6:20:43<4:51:10,  1.64it/s]
 56%|█████▋    | 36956/65536 [6:20:43<4:52:24,  1.63it/s]
 56%|█████▋    | 36957/65536 [6:20:44<4:46:52,  1.66it/s]
 56%|█████▋    | 36958/65536 [6:20:45<4:44:15,  1.68it/s]
 56%|█████▋    | 36959/65536 [6:20:45<4:55:52,  1.61it/s]
 56%|█████▋    | 36960/65536 [6:20:46<4:53:07,  1.62it/s]
                                                         
{'loss': 1.6654, 'learning_rate': 4.936329129423288e-07, 'epoch': 2281.48}
+
 56%|█████▋    | 36960/65536 [6:20:46<4:53:07,  1.62it/s]
 56%|█████▋    | 36961/65536 [6:20:47<4:52:50,  1.63it/s]
 56%|█████▋    | 36962/65536 [6:20:47<4:52:11,  1.63it/s]
 56%|█████▋    | 36963/65536 [6:20:48<4:49:17,  1.65it/s]
 56%|█████▋    | 36964/65536 [6:20:48<4:59:57,  1.59it/s]
 56%|█████▋    | 36965/65536 [6:20:49<4:52:26,  1.63it/s]
 56%|█████▋    | 36966/65536 [6:20:50<4:47:02,  1.66it/s]
 56%|█████▋    | 36967/65536 [6:20:50<4:43:49,  1.68it/s]
 56%|█████▋    | 36968/65536 [6:20:51<4:44:19,  1.67it/s]
 56%|█████▋    | 36969/65536 [6:20:51<4:54:54,  1.61it/s]
 56%|█████▋    | 36970/65536 [6:20:52<5:01:40,  1.58it/s]
 56%|█████▋    | 36971/65536 [6:20:53<4:50:36,  1.64it/s]
 56%|█████▋    | 36972/65536 [6:20:53<4:52:15,  1.63it/s]
 56%|█████▋    | 36973/65536 [6:20:54<4:50:11,  1.64it/s]
 56%|█████▋    | 36974/65536 [6:20:55<4:52:48,  1.63it/s]
 56%|█████▋    | 36975/65536 [6:20:55<4:50:36,  1.64it/s]
 56%|█████▋    | 36976/65536 [6:20:56<4:54:23,  1.62it/s]
 56%|█████▋    | 36977/65536 [6:20:56<4:51:37,  1.63it/s]
 56%|█████▋    | 36978/65536 [6:20:57<4:51:04,  1.64it/s]
 56%|█████▋    | 36979/65536 [6:20:58<4:57:02,  1.60it/s]
 56%|█████▋    | 36980/65536 [6:20:58<4:51:01,  1.64it/s]
                                                         
{'loss': 1.5985, 'learning_rate': 4.933574139831027e-07, 'epoch': 2282.72}
+
 56%|█████▋    | 36980/65536 [6:20:58<4:51:01,  1.64it/s]
 56%|█████▋    | 36981/65536 [6:20:59<4:47:08,  1.66it/s]
 56%|█████▋    | 36982/65536 [6:20:59<4:51:41,  1.63it/s]
 56%|█████▋    | 36983/65536 [6:21:00<4:52:12,  1.63it/s]
 56%|█████▋    | 36984/65536 [6:21:01<4:58:47,  1.59it/s]
 56%|█████▋    | 36985/65536 [6:21:01<5:07:25,  1.55it/s]
 56%|█████▋    | 36986/65536 [6:21:02<4:54:19,  1.62it/s]
 56%|█████▋    | 36987/65536 [6:21:03<4:59:15,  1.59it/s]
 56%|█████▋    | 36988/65536 [6:21:03<5:00:13,  1.58it/s]
 56%|█████▋    | 36989/65536 [6:21:04<5:05:11,  1.56it/s]
 56%|█████▋    | 36990/65536 [6:21:05<5:05:21,  1.56it/s]
 56%|█████▋    | 36991/65536 [6:21:05<5:08:12,  1.54it/s]
 56%|█████▋    | 36992/65536 [6:21:06<4:59:44,  1.59it/s]
 56%|█████▋    | 36993/65536 [6:21:06<4:55:07,  1.61it/s]
 56%|█████▋    | 36994/65536 [6:21:07<4:58:08,  1.60it/s]
 56%|█████▋    | 36995/65536 [6:21:08<4:49:06,  1.65it/s]
 56%|█████▋    | 36996/65536 [6:21:08<4:45:59,  1.66it/s]
 56%|█████▋    | 36997/65536 [6:21:09<4:41:24,  1.69it/s]
 56%|█████▋    | 36998/65536 [6:21:09<4:42:49,  1.68it/s]
 56%|█████▋    | 36999/65536 [6:21:10<4:38:30,  1.71it/s]
 56%|█████▋    | 37000/65536 [6:21:10<4:37:47,  1.71it/s]
                                                         
{'loss': 1.6147, 'learning_rate': 4.930819150238765e-07, 'epoch': 2283.95}
+
 56%|█████▋    | 37000/65536 [6:21:10<4:37:47,  1.71it/s]
 56%|█████▋    | 37001/65536 [6:21:11<4:56:03,  1.61it/s]
 56%|█████▋    | 37002/65536 [6:21:12<4:53:43,  1.62it/s]
 56%|█████▋    | 37003/65536 [6:21:12<4:52:14,  1.63it/s]
 56%|█████▋    | 37004/65536 [6:21:13<4:49:02,  1.65it/s]
 56%|█████▋    | 37005/65536 [6:21:14<4:44:08,  1.67it/s]
 56%|█████▋    | 37006/65536 [6:21:14<4:50:27,  1.64it/s]
 56%|█████▋    | 37007/65536 [6:21:15<4:46:51,  1.66it/s]
 56%|█████▋    | 37008/65536 [6:21:15<4:50:49,  1.63it/s]
 56%|█████▋    | 37009/65536 [6:21:16<4:55:10,  1.61it/s]
 56%|█████▋    | 37010/65536 [6:21:17<4:58:01,  1.60it/s]
 56%|█████▋    | 37011/65536 [6:21:17<5:00:26,  1.58it/s]
 56%|█████▋    | 37012/65536 [6:21:18<4:54:07,  1.62it/s]
 56%|█████▋    | 37013/65536 [6:21:19<4:58:44,  1.59it/s]
 56%|█████▋    | 37014/65536 [6:21:19<4:50:44,  1.63it/s]
 56%|█████▋    | 37015/65536 [6:21:20<4:42:19,  1.68it/s]
 56%|█████▋    | 37016/65536 [6:21:20<4:43:39,  1.68it/s]
 56%|█████▋    | 37017/65536 [6:21:21<4:39:35,  1.70it/s]
 56%|█████▋    | 37018/65536 [6:21:22<4:50:51,  1.63it/s]
 56%|█████▋    | 37019/65536 [6:21:22<4:48:01,  1.65it/s]
 56%|█████▋    | 37020/65536 [6:21:23<4:42:16,  1.68it/s]
                                                         
{'loss': 1.6644, 'learning_rate': 4.928064160646504e-07, 'epoch': 2285.19}
+
 56%|█████▋    | 37020/65536 [6:21:23<4:42:16,  1.68it/s]
 56%|█████▋    | 37021/65536 [6:21:23<4:44:37,  1.67it/s]
 56%|█████▋    | 37022/65536 [6:21:24<4:48:59,  1.64it/s]
 56%|█████▋    | 37023/65536 [6:21:25<4:45:14,  1.67it/s]
 56%|█████▋    | 37024/65536 [6:21:25<4:52:45,  1.62it/s]
 56%|█████▋    | 37025/65536 [6:21:26<5:05:01,  1.56it/s]
 56%|█████▋    | 37026/65536 [6:21:27<5:02:17,  1.57it/s]
 56%|█████▋    | 37027/65536 [6:21:27<4:58:46,  1.59it/s]
 57%|█████▋    | 37028/65536 [6:21:28<4:54:34,  1.61it/s]
 57%|█████▋    | 37029/65536 [6:21:28<4:49:19,  1.64it/s]
 57%|█████▋    | 37030/65536 [6:21:29<4:55:09,  1.61it/s]
 57%|█████▋    | 37031/65536 [6:21:30<4:53:31,  1.62it/s]
 57%|█████▋    | 37032/65536 [6:21:30<4:54:02,  1.62it/s]
 57%|█████▋    | 37033/65536 [6:21:31<4:53:27,  1.62it/s]
 57%|█████▋    | 37034/65536 [6:21:31<4:57:24,  1.60it/s]
 57%|█████▋    | 37035/65536 [6:21:32<4:56:50,  1.60it/s]
 57%|█████▋    | 37036/65536 [6:21:33<4:50:06,  1.64it/s]
 57%|█████▋    | 37037/65536 [6:21:33<4:59:25,  1.59it/s]
 57%|█████▋    | 37038/65536 [6:21:34<4:52:35,  1.62it/s]
 57%|█████▋    | 37039/65536 [6:21:35<4:52:33,  1.62it/s]
 57%|█████▋    | 37040/65536 [6:21:35<4:54:44,  1.61it/s]
                                                         
{'loss': 1.6265, 'learning_rate': 4.925309171054242e-07, 'epoch': 2286.42}
+
 57%|█████▋    | 37040/65536 [6:21:35<4:54:44,  1.61it/s]
 57%|█████▋    | 37041/65536 [6:21:36<4:54:51,  1.61it/s]
 57%|█████▋    | 37042/65536 [6:21:36<4:52:10,  1.63it/s]
 57%|█████▋    | 37043/65536 [6:21:37<4:45:53,  1.66it/s]
 57%|█████▋    | 37044/65536 [6:21:38<4:47:49,  1.65it/s]
 57%|█████▋    | 37045/65536 [6:21:38<4:53:23,  1.62it/s]
 57%|█████▋    | 37046/65536 [6:21:39<4:49:10,  1.64it/s]
 57%|█████▋    | 37047/65536 [6:21:39<4:54:13,  1.61it/s]
 57%|█████▋    | 37048/65536 [6:21:40<4:53:02,  1.62it/s]
 57%|█████▋    | 37049/65536 [6:21:41<4:54:04,  1.61it/s]
 57%|█████▋    | 37050/65536 [6:21:41<5:06:39,  1.55it/s]
 57%|█████▋    | 37051/65536 [6:21:42<5:08:00,  1.54it/s]
 57%|█████▋    | 37052/65536 [6:21:43<5:03:57,  1.56it/s]
 57%|█████▋    | 37053/65536 [6:21:43<4:59:37,  1.58it/s]
 57%|█████▋    | 37054/65536 [6:21:44<4:52:53,  1.62it/s]
 57%|█████▋    | 37055/65536 [6:21:44<4:54:04,  1.61it/s]
 57%|█████▋    | 37056/65536 [6:21:45<4:50:30,  1.63it/s]
 57%|█████▋    | 37057/65536 [6:21:46<4:54:55,  1.61it/s]
 57%|█████▋    | 37058/65536 [6:21:46<4:54:17,  1.61it/s]
 57%|█████▋    | 37059/65536 [6:21:47<4:46:42,  1.66it/s]
 57%|█████▋    | 37060/65536 [6:21:48<4:56:09,  1.60it/s]
                                                         
{'loss': 1.6096, 'learning_rate': 4.922554181461981e-07, 'epoch': 2287.65}
+
 57%|█████▋    | 37060/65536 [6:21:48<4:56:09,  1.60it/s]
 57%|█████▋    | 37061/65536 [6:21:48<4:59:52,  1.58it/s]
 57%|█████▋    | 37062/65536 [6:21:49<4:51:21,  1.63it/s]
 57%|█████▋    | 37063/65536 [6:21:49<4:47:17,  1.65it/s]
 57%|█████▋    | 37064/65536 [6:21:50<4:51:42,  1.63it/s]
 57%|█████▋    | 37065/65536 [6:21:51<4:43:22,  1.67it/s]
 57%|█████▋    | 37066/65536 [6:21:51<4:53:03,  1.62it/s]
 57%|█████▋    | 37067/65536 [6:21:52<4:58:27,  1.59it/s]
 57%|█████▋    | 37068/65536 [6:21:52<4:50:59,  1.63it/s]
 57%|█████▋    | 37069/65536 [6:21:53<4:49:40,  1.64it/s]
 57%|█████▋    | 37070/65536 [6:21:54<4:56:23,  1.60it/s]
 57%|█████▋    | 37071/65536 [6:21:54<4:48:28,  1.64it/s]
 57%|█████▋    | 37072/65536 [6:21:55<4:52:13,  1.62it/s]
 57%|█████▋    | 37073/65536 [6:21:56<4:49:38,  1.64it/s]
 57%|█████▋    | 37074/65536 [6:21:56<4:49:30,  1.64it/s]
 57%|█████▋    | 37075/65536 [6:21:57<4:49:28,  1.64it/s]
 57%|█████▋    | 37076/65536 [6:21:57<4:49:36,  1.64it/s]
 57%|█████▋    | 37077/65536 [6:21:58<4:44:13,  1.67it/s]
 57%|█████▋    | 37078/65536 [6:21:59<4:43:29,  1.67it/s]
 57%|█████▋    | 37079/65536 [6:21:59<4:45:15,  1.66it/s]
 57%|█████▋    | 37080/65536 [6:22:00<5:00:49,  1.58it/s]
                                                         
{'loss': 1.6549, 'learning_rate': 4.91979919186972e-07, 'epoch': 2288.89}
+
 57%|█████▋    | 37080/65536 [6:22:00<5:00:49,  1.58it/s]
 57%|█████▋    | 37081/65536 [6:22:00<4:56:03,  1.60it/s]
 57%|█████▋    | 37082/65536 [6:22:01<5:03:55,  1.56it/s]
 57%|█████▋    | 37083/65536 [6:22:02<5:12:00,  1.52it/s]
 57%|█████▋    | 37084/65536 [6:22:03<5:12:56,  1.52it/s]
 57%|█████▋    | 37085/65536 [6:22:03<5:07:36,  1.54it/s]
 57%|█████▋    | 37086/65536 [6:22:04<5:08:46,  1.54it/s]
 57%|█████▋    | 37087/65536 [6:22:04<5:04:52,  1.56it/s]
 57%|█████▋    | 37088/65536 [6:22:05<4:59:14,  1.58it/s]
 57%|█████▋    | 37089/65536 [6:22:06<4:57:59,  1.59it/s]
 57%|█████▋    | 37090/65536 [6:22:06<4:54:14,  1.61it/s]
 57%|█████▋    | 37091/65536 [6:22:07<4:57:50,  1.59it/s]
 57%|█████▋    | 37092/65536 [6:22:07<4:55:44,  1.60it/s]
 57%|█████▋    | 37093/65536 [6:22:08<4:50:45,  1.63it/s]
 57%|█████▋    | 37094/65536 [6:22:09<4:44:22,  1.67it/s]
 57%|█████▋    | 37095/65536 [6:22:09<4:44:16,  1.67it/s]
 57%|█████▋    | 37096/65536 [6:22:10<4:38:50,  1.70it/s]
 57%|█████▋    | 37097/65536 [6:22:10<4:44:14,  1.67it/s]
 57%|█████▋    | 37098/65536 [6:22:11<4:41:35,  1.68it/s]
 57%|█████▋    | 37099/65536 [6:22:12<4:55:14,  1.61it/s]
 57%|█████▋    | 37100/65536 [6:22:12<4:50:08,  1.63it/s]
                                                         
{'loss': 1.6385, 'learning_rate': 4.917044202277458e-07, 'epoch': 2290.12}
+
 57%|█████▋    | 37100/65536 [6:22:12<4:50:08,  1.63it/s]
 57%|█████▋    | 37101/65536 [6:22:13<4:53:49,  1.61it/s]
 57%|█████▋    | 37102/65536 [6:22:14<4:55:35,  1.60it/s]
 57%|█████▋    | 37103/65536 [6:22:14<4:51:53,  1.62it/s]
 57%|█████▋    | 37104/65536 [6:22:15<4:56:11,  1.60it/s]
 57%|█████▋    | 37105/65536 [6:22:15<4:53:46,  1.61it/s]
 57%|█████▋    | 37106/65536 [6:22:16<4:54:52,  1.61it/s]
 57%|█████▋    | 37107/65536 [6:22:17<4:48:19,  1.64it/s]
 57%|█████▋    | 37108/65536 [6:22:17<4:54:00,  1.61it/s]
 57%|█████▋    | 37109/65536 [6:22:18<5:01:18,  1.57it/s]
 57%|█████▋    | 37110/65536 [6:22:19<4:55:22,  1.60it/s]
 57%|█████▋    | 37111/65536 [6:22:19<4:51:55,  1.62it/s]
 57%|█████▋    | 37112/65536 [6:22:20<4:55:38,  1.60it/s]
 57%|█████▋    | 37113/65536 [6:22:20<4:50:43,  1.63it/s]
 57%|█████▋    | 37114/65536 [6:22:21<4:48:08,  1.64it/s]
 57%|█████▋    | 37115/65536 [6:22:22<4:59:41,  1.58it/s]
 57%|█████▋    | 37116/65536 [6:22:22<4:57:05,  1.59it/s]
 57%|█████▋    | 37117/65536 [6:22:23<4:52:10,  1.62it/s]
 57%|█████▋    | 37118/65536 [6:22:23<4:48:07,  1.64it/s]
 57%|█████▋    | 37119/65536 [6:22:24<4:52:02,  1.62it/s]
 57%|█████▋    | 37120/65536 [6:22:25<4:56:28,  1.60it/s]
                                                         
{'loss': 1.6274, 'learning_rate': 4.914289212685197e-07, 'epoch': 2291.36}
+
 57%|█████▋    | 37120/65536 [6:22:25<4:56:28,  1.60it/s]
 57%|█████▋    | 37121/65536 [6:22:25<4:52:24,  1.62it/s]
 57%|█████▋    | 37122/65536 [6:22:26<4:47:47,  1.65it/s]
 57%|█████▋    | 37123/65536 [6:22:27<4:46:06,  1.66it/s]
 57%|█████▋    | 37124/65536 [6:22:27<4:49:56,  1.63it/s]
 57%|█████▋    | 37125/65536 [6:22:28<4:43:53,  1.67it/s]
 57%|█████▋    | 37126/65536 [6:22:28<4:50:23,  1.63it/s]
 57%|█████▋    | 37127/65536 [6:22:29<4:51:23,  1.62it/s]
 57%|█████▋    | 37128/65536 [6:22:30<4:50:16,  1.63it/s]
 57%|█████▋    | 37129/65536 [6:22:30<4:49:40,  1.63it/s]
 57%|█████▋    | 37130/65536 [6:22:31<4:51:27,  1.62it/s]
 57%|█████▋    | 37131/65536 [6:22:32<5:00:43,  1.57it/s]
 57%|█████▋    | 37132/65536 [6:22:32<4:49:19,  1.64it/s]
 57%|█████▋    | 37133/65536 [6:22:33<4:46:12,  1.65it/s]
 57%|█████▋    | 37134/65536 [6:22:33<4:50:36,  1.63it/s]
 57%|█████▋    | 37135/65536 [6:22:34<4:50:29,  1.63it/s]
 57%|█████▋    | 37136/65536 [6:22:34<4:48:03,  1.64it/s]
 57%|█████▋    | 37137/65536 [6:22:35<4:46:11,  1.65it/s]
 57%|█████▋    | 37138/65536 [6:22:36<4:47:15,  1.65it/s]
 57%|█████▋    | 37139/65536 [6:22:36<5:00:59,  1.57it/s]
 57%|█████▋    | 37140/65536 [6:22:37<4:50:24,  1.63it/s]
                                                         
{'loss': 1.6508, 'learning_rate': 4.911534223092935e-07, 'epoch': 2292.59}
+
 57%|█████▋    | 37140/65536 [6:22:37<4:50:24,  1.63it/s]
 57%|█████▋    | 37141/65536 [6:22:38<4:55:55,  1.60it/s]
 57%|█████▋    | 37142/65536 [6:22:38<4:49:45,  1.63it/s]
 57%|█████▋    | 37143/65536 [6:22:39<4:48:16,  1.64it/s]
 57%|█████▋    | 37144/65536 [6:22:39<4:45:19,  1.66it/s]
 57%|█████▋    | 37145/65536 [6:22:40<4:43:54,  1.67it/s]
 57%|█████▋    | 37146/65536 [6:22:41<4:41:08,  1.68it/s]
 57%|█████▋    | 37147/65536 [6:22:41<4:57:45,  1.59it/s]
 57%|█████▋    | 37148/65536 [6:22:42<4:58:53,  1.58it/s]
 57%|█████▋    | 37149/65536 [6:22:43<5:03:01,  1.56it/s]
 57%|█████▋    | 37150/65536 [6:22:43<5:04:25,  1.55it/s]
 57%|█████▋    | 37151/65536 [6:22:44<4:55:45,  1.60it/s]
 57%|█████▋    | 37152/65536 [6:22:44<4:51:55,  1.62it/s]
 57%|█████▋    | 37153/65536 [6:22:45<4:51:55,  1.62it/s]
 57%|█████▋    | 37154/65536 [6:22:46<4:48:08,  1.64it/s]
 57%|█████▋    | 37155/65536 [6:22:46<4:53:51,  1.61it/s]
 57%|█████▋    | 37156/65536 [6:22:47<4:51:23,  1.62it/s]
 57%|█████▋    | 37157/65536 [6:22:48<4:56:26,  1.60it/s]
 57%|█████▋    | 37158/65536 [6:22:48<4:48:35,  1.64it/s]
 57%|█████▋    | 37159/65536 [6:22:49<4:41:41,  1.68it/s]
 57%|█████▋    | 37160/65536 [6:22:49<4:46:25,  1.65it/s]
                                                         
{'loss': 1.6107, 'learning_rate': 4.908779233500674e-07, 'epoch': 2293.83}
+
 57%|█████▋    | 37160/65536 [6:22:49<4:46:25,  1.65it/s]
 57%|█████▋    | 37161/65536 [6:22:50<4:43:54,  1.67it/s]
 57%|█████▋    | 37162/65536 [6:22:50<4:39:33,  1.69it/s]
 57%|█████▋    | 37163/65536 [6:22:51<4:57:25,  1.59it/s]
 57%|█████▋    | 37164/65536 [6:22:52<4:50:41,  1.63it/s]
 57%|█████▋    | 37165/65536 [6:22:52<4:44:50,  1.66it/s]
 57%|█████▋    | 37166/65536 [6:22:53<4:46:26,  1.65it/s]
 57%|█████▋    | 37167/65536 [6:22:53<4:41:53,  1.68it/s]
 57%|█████▋    | 37168/65536 [6:22:54<4:53:03,  1.61it/s]
 57%|█████▋    | 37169/65536 [6:22:55<4:47:05,  1.65it/s]
 57%|█████▋    | 37170/65536 [6:22:55<4:54:12,  1.61it/s]
 57%|█████▋    | 37171/65536 [6:22:56<4:50:08,  1.63it/s]
 57%|█████▋    | 37172/65536 [6:22:57<4:50:32,  1.63it/s]
 57%|█████▋    | 37173/65536 [6:22:57<4:46:11,  1.65it/s]
 57%|█████▋    | 37174/65536 [6:22:58<4:45:33,  1.66it/s]
 57%|█████▋    | 37175/65536 [6:22:58<4:46:30,  1.65it/s]
 57%|█████▋    | 37176/65536 [6:22:59<4:49:40,  1.63it/s]
 57%|█████▋    | 37177/65536 [6:23:00<4:51:17,  1.62it/s]
 57%|█████▋    | 37178/65536 [6:23:00<4:45:26,  1.66it/s]
 57%|█████▋    | 37179/65536 [6:23:01<4:45:41,  1.65it/s]
 57%|█████▋    | 37180/65536 [6:23:01<4:52:34,  1.62it/s]
                                                         
{'loss': 1.6367, 'learning_rate': 4.906024243908412e-07, 'epoch': 2295.06}
+
 57%|█████▋    | 37180/65536 [6:23:02<4:52:34,  1.62it/s]
 57%|█████▋    | 37181/65536 [6:23:02<4:59:43,  1.58it/s]
 57%|█████▋    | 37182/65536 [6:23:03<4:58:27,  1.58it/s]
 57%|█████▋    | 37183/65536 [6:23:03<4:54:25,  1.61it/s]
 57%|█████▋    | 37184/65536 [6:23:04<4:50:09,  1.63it/s]
 57%|█████▋    | 37185/65536 [6:23:05<4:45:15,  1.66it/s]
 57%|█████▋    | 37186/65536 [6:23:05<4:45:51,  1.65it/s]
 57%|█████▋    | 37187/65536 [6:23:06<4:54:27,  1.60it/s]
 57%|█████▋    | 37188/65536 [6:23:06<4:51:52,  1.62it/s]
 57%|█████▋    | 37189/65536 [6:23:07<4:50:19,  1.63it/s]
 57%|█████▋    | 37190/65536 [6:23:08<4:45:57,  1.65it/s]
 57%|█████▋    | 37191/65536 [6:23:08<4:46:06,  1.65it/s]
 57%|█████▋    | 37192/65536 [6:23:09<4:44:19,  1.66it/s]
 57%|█████▋    | 37193/65536 [6:23:09<4:46:53,  1.65it/s]
 57%|█████▋    | 37194/65536 [6:23:10<4:47:21,  1.64it/s]
 57%|█████▋    | 37195/65536 [6:23:11<4:42:37,  1.67it/s]
 57%|█████▋    | 37196/65536 [6:23:11<4:50:44,  1.62it/s]
 57%|█████▋    | 37197/65536 [6:23:12<4:54:05,  1.61it/s]
 57%|█████▋    | 37198/65536 [6:23:13<4:51:35,  1.62it/s]
 57%|█████▋    | 37199/65536 [6:23:13<4:46:20,  1.65it/s]
 57%|█████▋    | 37200/65536 [6:23:14<4:47:36,  1.64it/s]
                                                         
{'loss': 1.6522, 'learning_rate': 4.903269254316149e-07, 'epoch': 2296.3}
+
 57%|█████▋    | 37200/65536 [6:23:14<4:47:36,  1.64it/s]
 57%|█████▋    | 37201/65536 [6:23:14<4:50:36,  1.63it/s]
 57%|█████▋    | 37202/65536 [6:23:15<4:53:16,  1.61it/s]
 57%|█████▋    | 37203/65536 [6:23:16<4:59:51,  1.57it/s]
 57%|█████▋    | 37204/65536 [6:23:16<5:08:00,  1.53it/s]
 57%|█████▋    | 37205/65536 [6:23:17<4:57:47,  1.59it/s]
 57%|█████▋    | 37206/65536 [6:23:18<4:52:51,  1.61it/s]
 57%|█████▋    | 37207/65536 [6:23:18<4:42:52,  1.67it/s]
 57%|█████▋    | 37208/65536 [6:23:19<4:47:03,  1.64it/s]
 57%|█████▋    | 37209/65536 [6:23:19<4:40:41,  1.68it/s]
 57%|█████▋    | 37210/65536 [6:23:20<4:45:26,  1.65it/s]
 57%|█████▋    | 37211/65536 [6:23:21<4:44:08,  1.66it/s]
 57%|█████▋    | 37212/65536 [6:23:21<5:01:43,  1.56it/s]
 57%|█████▋    | 37213/65536 [6:23:22<4:53:10,  1.61it/s]
 57%|█████▋    | 37214/65536 [6:23:22<4:48:17,  1.64it/s]
 57%|█████▋    | 37215/65536 [6:23:23<4:47:53,  1.64it/s]
 57%|█████▋    | 37216/65536 [6:23:24<4:47:37,  1.64it/s]
 57%|█████▋    | 37217/65536 [6:23:24<4:51:03,  1.62it/s]
 57%|█████▋    | 37218/65536 [6:23:25<4:49:39,  1.63it/s]
 57%|█████▋    | 37219/65536 [6:23:25<4:44:26,  1.66it/s]
 57%|█████▋    | 37220/65536 [6:23:26<4:50:40,  1.62it/s]
                                                         
{'loss': 1.5979, 'learning_rate': 4.900514264723889e-07, 'epoch': 2297.53}
+
 57%|█████▋    | 37220/65536 [6:23:26<4:50:40,  1.62it/s]
 57%|█████▋    | 37221/65536 [6:23:27<4:45:04,  1.66it/s]
 57%|█████▋    | 37222/65536 [6:23:27<4:44:07,  1.66it/s]
 57%|█████▋    | 37223/65536 [6:23:28<4:47:11,  1.64it/s]
 57%|█████▋    | 37224/65536 [6:23:28<4:43:11,  1.67it/s]
 57%|█████▋    | 37225/65536 [6:23:29<4:40:12,  1.68it/s]
 57%|█████▋    | 37226/65536 [6:23:30<4:37:10,  1.70it/s]
 57%|█████▋    | 37227/65536 [6:23:30<4:46:13,  1.65it/s]
 57%|█████▋    | 37228/65536 [6:23:31<4:59:23,  1.58it/s]
 57%|█████▋    | 37229/65536 [6:23:32<4:52:54,  1.61it/s]
 57%|█████▋    | 37230/65536 [6:23:32<4:56:55,  1.59it/s]
 57%|█████▋    | 37231/65536 [6:23:33<4:48:24,  1.64it/s]
 57%|█████▋    | 37232/65536 [6:23:33<4:41:43,  1.67it/s]
 57%|█████▋    | 37233/65536 [6:23:34<4:37:28,  1.70it/s]
 57%|█████▋    | 37234/65536 [6:23:34<4:35:08,  1.71it/s]
 57%|█████▋    | 37235/65536 [6:23:35<4:47:44,  1.64it/s]
 57%|█████▋    | 37236/65536 [6:23:36<4:51:23,  1.62it/s]
 57%|█████▋    | 37237/65536 [6:23:36<4:45:45,  1.65it/s]
 57%|█████▋    | 37238/65536 [6:23:37<4:40:12,  1.68it/s]
 57%|█████▋    | 37239/65536 [6:23:38<4:42:30,  1.67it/s]
 57%|█████▋    | 37240/65536 [6:23:38<4:43:32,  1.66it/s]
                                                         
{'loss': 1.684, 'learning_rate': 4.897759275131627e-07, 'epoch': 2298.77}
+
 57%|█████▋    | 37240/65536 [6:23:38<4:43:32,  1.66it/s]
 57%|█████▋    | 37241/65536 [6:23:39<4:42:27,  1.67it/s]
 57%|█████▋    | 37242/65536 [6:23:39<4:49:51,  1.63it/s]
 57%|█████▋    | 37243/65536 [6:23:40<4:50:59,  1.62it/s]
 57%|█████▋    | 37244/65536 [6:23:41<5:02:19,  1.56it/s]
 57%|█████▋    | 37245/65536 [6:23:41<5:02:49,  1.56it/s]
 57%|█████▋    | 37246/65536 [6:23:42<4:58:48,  1.58it/s]
 57%|█████▋    | 37247/65536 [6:23:43<4:59:50,  1.57it/s]
 57%|█████▋    | 37248/65536 [6:23:43<4:55:44,  1.59it/s]
 57%|█████▋    | 37249/65536 [6:23:44<4:51:50,  1.62it/s]
 57%|█████▋    | 37250/65536 [6:23:44<4:52:14,  1.61it/s]
 57%|█████▋    | 37251/65536 [6:23:45<4:50:47,  1.62it/s]
 57%|█████▋    | 37252/65536 [6:23:46<4:49:46,  1.63it/s]
 57%|█████▋    | 37253/65536 [6:23:46<4:49:37,  1.63it/s]
 57%|█████▋    | 37254/65536 [6:23:47<4:53:53,  1.60it/s]
 57%|█████▋    | 37255/65536 [6:23:47<4:48:14,  1.64it/s]
 57%|█████▋    | 37256/65536 [6:23:48<4:45:07,  1.65it/s]
 57%|█████▋    | 37257/65536 [6:23:49<4:47:43,  1.64it/s]
 57%|█████▋    | 37258/65536 [6:23:49<4:49:37,  1.63it/s]
 57%|█████▋    | 37259/65536 [6:23:50<4:47:37,  1.64it/s]
 57%|█████▋    | 37260/65536 [6:23:51<4:49:09,  1.63it/s]
                                                         
{'loss': 1.615, 'learning_rate': 4.895004285539366e-07, 'epoch': 2300.0}
+
 57%|█████▋    | 37260/65536 [6:23:51<4:49:09,  1.63it/s]
 57%|█████▋    | 37261/65536 [6:23:51<5:02:50,  1.56it/s]
 57%|█████▋    | 37262/65536 [6:23:52<4:55:34,  1.59it/s]
 57%|█████▋    | 37263/65536 [6:23:52<4:48:45,  1.63it/s]
 57%|█████▋    | 37264/65536 [6:23:53<4:50:53,  1.62it/s]
 57%|█████▋    | 37265/65536 [6:23:54<4:55:28,  1.59it/s]
 57%|█████▋    | 37266/65536 [6:23:54<4:50:43,  1.62it/s]
 57%|█████▋    | 37267/65536 [6:23:55<4:55:17,  1.60it/s]
 57%|█████▋    | 37268/65536 [6:23:56<4:54:33,  1.60it/s]
 57%|█████▋    | 37269/65536 [6:23:56<4:48:36,  1.63it/s]
 57%|█████▋    | 37270/65536 [6:23:57<4:54:55,  1.60it/s]
 57%|█████▋    | 37271/65536 [6:23:57<4:56:08,  1.59it/s]
 57%|█████▋    | 37272/65536 [6:23:58<4:55:20,  1.59it/s]
 57%|█████▋    | 37273/65536 [6:23:59<4:46:52,  1.64it/s]
 57%|█████▋    | 37274/65536 [6:23:59<4:36:23,  1.70it/s]
 57%|█████▋    | 37275/65536 [6:24:00<4:40:25,  1.68it/s]
 57%|█████▋    | 37276/65536 [6:24:00<4:45:49,  1.65it/s]
 57%|█████▋    | 37277/65536 [6:24:01<4:49:42,  1.63it/s]
 57%|█████▋    | 37278/65536 [6:24:02<4:53:01,  1.61it/s]
 57%|█████▋    | 37279/65536 [6:24:02<4:51:04,  1.62it/s]
 57%|█████▋    | 37280/65536 [6:24:03<4:40:23,  1.68it/s]
                                                         
{'loss': 1.6249, 'learning_rate': 4.892249295947104e-07, 'epoch': 2301.23}
+
 57%|█████▋    | 37280/65536 [6:24:03<4:40:23,  1.68it/s]
 57%|█████▋    | 37281/65536 [6:24:04<4:54:58,  1.60it/s]
 57%|█████▋    | 37282/65536 [6:24:04<4:57:03,  1.59it/s]
 57%|█████▋    | 37283/65536 [6:24:05<4:58:39,  1.58it/s]
 57%|█████▋    | 37284/65536 [6:24:05<4:48:22,  1.63it/s]
 57%|█████▋    | 37285/65536 [6:24:06<4:47:57,  1.64it/s]
 57%|█████▋    | 37286/65536 [6:24:07<4:42:55,  1.66it/s]
 57%|█████▋    | 37287/65536 [6:24:07<4:47:09,  1.64it/s]
 57%|█████▋    | 37288/65536 [6:24:08<4:48:41,  1.63it/s]
 57%|█████▋    | 37289/65536 [6:24:08<4:46:24,  1.64it/s]
 57%|█████▋    | 37290/65536 [6:24:09<4:44:33,  1.65it/s]
 57%|█████▋    | 37291/65536 [6:24:10<4:42:58,  1.66it/s]
 57%|█████▋    | 37292/65536 [6:24:10<4:42:40,  1.67it/s]
 57%|█████▋    | 37293/65536 [6:24:11<4:51:02,  1.62it/s]
 57%|█████▋    | 37294/65536 [6:24:12<4:54:42,  1.60it/s]
 57%|█████▋    | 37295/65536 [6:24:12<4:56:34,  1.59it/s]
 57%|█████▋    | 37296/65536 [6:24:13<4:53:46,  1.60it/s]
 57%|█████▋    | 37297/65536 [6:24:13<4:45:34,  1.65it/s]
 57%|█████▋    | 37298/65536 [6:24:14<4:45:54,  1.65it/s]
 57%|█████▋    | 37299/65536 [6:24:15<4:46:43,  1.64it/s]
 57%|█████▋    | 37300/65536 [6:24:15<4:44:04,  1.66it/s]
                                                         
{'loss': 1.6209, 'learning_rate': 4.889494306354843e-07, 'epoch': 2302.47}
+
 57%|█████▋    | 37300/65536 [6:24:15<4:44:04,  1.66it/s]
 57%|█████▋    | 37301/65536 [6:24:16<4:47:06,  1.64it/s]
 57%|█████▋    | 37302/65536 [6:24:16<4:44:54,  1.65it/s]
 57%|█████▋    | 37303/65536 [6:24:17<4:42:23,  1.67it/s]
 57%|█████▋    | 37304/65536 [6:24:18<4:44:39,  1.65it/s]
 57%|█████▋    | 37305/65536 [6:24:18<4:51:45,  1.61it/s]
 57%|█████▋    | 37306/65536 [6:24:19<5:01:36,  1.56it/s]
 57%|█████▋    | 37307/65536 [6:24:20<5:02:32,  1.56it/s]
 57%|█████▋    | 37308/65536 [6:24:20<4:52:57,  1.61it/s]
 57%|█████▋    | 37309/65536 [6:24:21<5:06:49,  1.53it/s]
 57%|█████▋    | 37310/65536 [6:24:21<4:58:51,  1.57it/s]
 57%|█████▋    | 37311/65536 [6:24:22<4:50:40,  1.62it/s]
 57%|█████▋    | 37312/65536 [6:24:23<4:53:53,  1.60it/s]
 57%|█████▋    | 37313/65536 [6:24:23<4:48:37,  1.63it/s]
 57%|█████▋    | 37314/65536 [6:24:24<4:53:15,  1.60it/s]
 57%|█████▋    | 37315/65536 [6:24:25<4:53:22,  1.60it/s]
 57%|█████▋    | 37316/65536 [6:24:25<4:52:19,  1.61it/s]
 57%|█████▋    | 37317/65536 [6:24:26<4:46:06,  1.64it/s]
 57%|█████▋    | 37318/65536 [6:24:26<4:43:49,  1.66it/s]
 57%|█████▋    | 37319/65536 [6:24:27<4:48:31,  1.63it/s]
 57%|█████▋    | 37320/65536 [6:24:28<4:41:10,  1.67it/s]
                                                         
{'loss': 1.6521, 'learning_rate': 4.886739316762581e-07, 'epoch': 2303.7}
+
 57%|█████▋    | 37320/65536 [6:24:28<4:41:10,  1.67it/s]
 57%|█████▋    | 37321/65536 [6:24:28<4:48:31,  1.63it/s]
 57%|█████▋    | 37322/65536 [6:24:29<4:50:57,  1.62it/s]
 57%|█████▋    | 37323/65536 [6:24:29<4:45:06,  1.65it/s]
 57%|█████▋    | 37324/65536 [6:24:30<4:42:44,  1.66it/s]
 57%|█████▋    | 37325/65536 [6:24:31<4:53:12,  1.60it/s]
 57%|█████▋    | 37326/65536 [6:24:31<5:01:03,  1.56it/s]
 57%|█████▋    | 37327/65536 [6:24:32<4:55:35,  1.59it/s]
 57%|█████▋    | 37328/65536 [6:24:33<4:51:29,  1.61it/s]
 57%|█████▋    | 37329/65536 [6:24:33<4:48:46,  1.63it/s]
 57%|█████▋    | 37330/65536 [6:24:34<4:46:42,  1.64it/s]
 57%|█████▋    | 37331/65536 [6:24:34<4:44:45,  1.65it/s]
 57%|█████▋    | 37332/65536 [6:24:35<4:42:34,  1.66it/s]
 57%|█████▋    | 37333/65536 [6:24:36<4:41:57,  1.67it/s]
 57%|█████▋    | 37334/65536 [6:24:36<4:41:23,  1.67it/s]
 57%|█████▋    | 37335/65536 [6:24:37<4:38:19,  1.69it/s]
 57%|█████▋    | 37336/65536 [6:24:37<4:37:19,  1.69it/s]
 57%|█████▋    | 37337/65536 [6:24:38<4:46:05,  1.64it/s]
 57%|█████▋    | 37338/65536 [6:24:39<4:43:57,  1.66it/s]
 57%|█████▋    | 37339/65536 [6:24:39<4:41:39,  1.67it/s]
 57%|█████▋    | 37340/65536 [6:24:40<4:39:31,  1.68it/s]
                                                         
{'loss': 1.6456, 'learning_rate': 4.883984327170319e-07, 'epoch': 2304.94}
+
 57%|█████▋    | 37340/65536 [6:24:40<4:39:31,  1.68it/s]
 57%|█████▋    | 37341/65536 [6:24:40<4:44:19,  1.65it/s]
 57%|█████▋    | 37342/65536 [6:24:41<4:57:27,  1.58it/s]
 57%|█████▋    | 37343/65536 [6:24:42<4:51:00,  1.61it/s]
 57%|█████▋    | 37344/65536 [6:24:42<4:41:50,  1.67it/s]
 57%|█████▋    | 37345/65536 [6:24:43<4:46:06,  1.64it/s]
 57%|█████▋    | 37346/65536 [6:24:43<4:40:32,  1.67it/s]
 57%|█████▋    | 37347/65536 [6:24:44<4:42:31,  1.66it/s]
 57%|█████▋    | 37348/65536 [6:24:45<4:39:26,  1.68it/s]
 57%|█████▋    | 37349/65536 [6:24:45<4:44:48,  1.65it/s]
 57%|█████▋    | 37350/65536 [6:24:46<4:44:45,  1.65it/s]
 57%|█████▋    | 37351/65536 [6:24:46<4:48:57,  1.63it/s]
 57%|█████▋    | 37352/65536 [6:24:47<4:57:37,  1.58it/s]
 57%|█████▋    | 37353/65536 [6:24:48<4:57:25,  1.58it/s]
 57%|█████▋    | 37354/65536 [6:24:48<4:47:43,  1.63it/s]
 57%|█████▋    | 37355/65536 [6:24:49<4:44:18,  1.65it/s]
 57%|█████▋    | 37356/65536 [6:24:50<4:52:00,  1.61it/s]
 57%|█████▋    | 37357/65536 [6:24:50<4:47:14,  1.64it/s]
 57%|█████▋    | 37358/65536 [6:24:51<4:58:41,  1.57it/s]
 57%|█████▋    | 37359/65536 [6:24:51<4:53:41,  1.60it/s]
 57%|█████▋    | 37360/65536 [6:24:52<4:49:35,  1.62it/s]
                                                         
{'loss': 1.64, 'learning_rate': 4.881229337578058e-07, 'epoch': 2306.17}
+
 57%|█████▋    | 37360/65536 [6:24:52<4:49:35,  1.62it/s]
 57%|█████▋    | 37361/65536 [6:24:53<4:51:45,  1.61it/s]
 57%|█████▋    | 37362/65536 [6:24:53<4:56:13,  1.59it/s]
 57%|█████▋    | 37363/65536 [6:24:54<4:56:39,  1.58it/s]
 57%|█████▋    | 37364/65536 [6:24:55<4:49:33,  1.62it/s]
 57%|█████▋    | 37365/65536 [6:24:55<4:49:00,  1.62it/s]
 57%|█████▋    | 37366/65536 [6:24:56<4:41:39,  1.67it/s]
 57%|█████▋    | 37367/65536 [6:24:56<4:41:23,  1.67it/s]
 57%|█████▋    | 37368/65536 [6:24:57<4:48:59,  1.62it/s]
 57%|█████▋    | 37369/65536 [6:24:58<4:44:43,  1.65it/s]
 57%|█████▋    | 37370/65536 [6:24:58<4:38:31,  1.69it/s]
 57%|█████▋    | 37371/65536 [6:24:59<4:43:04,  1.66it/s]
 57%|█████▋    | 37372/65536 [6:24:59<4:43:46,  1.65it/s]
 57%|█████▋    | 37373/65536 [6:25:00<4:53:10,  1.60it/s]
 57%|█████▋    | 37374/65536 [6:25:01<4:56:35,  1.58it/s]
 57%|█████▋    | 37375/65536 [6:25:01<4:51:38,  1.61it/s]
 57%|█████▋    | 37376/65536 [6:25:02<4:41:30,  1.67it/s]
 57%|█████▋    | 37377/65536 [6:25:02<4:47:23,  1.63it/s]
 57%|█████▋    | 37378/65536 [6:25:03<4:41:38,  1.67it/s]
 57%|█████▋    | 37379/65536 [6:25:04<4:37:58,  1.69it/s]
 57%|█████▋    | 37380/65536 [6:25:04<4:42:08,  1.66it/s]
                                                         
{'loss': 1.6339, 'learning_rate': 4.878474347985796e-07, 'epoch': 2307.41}
+
 57%|█████▋    | 37380/65536 [6:25:04<4:42:08,  1.66it/s]
 57%|█████▋    | 37381/65536 [6:25:05<4:50:44,  1.61it/s]
 57%|█████▋    | 37382/65536 [6:25:06<4:53:12,  1.60it/s]
 57%|█████▋    | 37383/65536 [6:25:06<4:44:48,  1.65it/s]
 57%|█████▋    | 37384/65536 [6:25:07<4:44:42,  1.65it/s]
 57%|█████▋    | 37385/65536 [6:25:07<4:43:24,  1.66it/s]
 57%|█████▋    | 37386/65536 [6:25:08<4:50:00,  1.62it/s]
 57%|█████▋    | 37387/65536 [6:25:09<4:52:54,  1.60it/s]
 57%|█████▋    | 37388/65536 [6:25:09<4:51:26,  1.61it/s]
 57%|█████▋    | 37389/65536 [6:25:10<4:51:09,  1.61it/s]
 57%|█████▋    | 37390/65536 [6:25:10<4:54:52,  1.59it/s]
 57%|█████▋    | 37391/65536 [6:25:11<4:48:32,  1.63it/s]
 57%|█████▋    | 37392/65536 [6:25:12<4:43:39,  1.65it/s]
 57%|█████▋    | 37393/65536 [6:25:12<4:43:51,  1.65it/s]
 57%|█████▋    | 37394/65536 [6:25:13<4:41:51,  1.66it/s]
 57%|█████▋    | 37395/65536 [6:25:13<4:46:09,  1.64it/s]
 57%|█████▋    | 37396/65536 [6:25:14<4:42:03,  1.66it/s]
 57%|█████▋    | 37397/65536 [6:25:15<4:38:07,  1.69it/s]
 57%|█████▋    | 37398/65536 [6:25:15<4:34:01,  1.71it/s]
 57%|█████▋    | 37399/65536 [6:25:16<4:37:44,  1.69it/s]
 57%|█████▋    | 37400/65536 [6:25:16<4:40:10,  1.67it/s]
                                                         
{'loss': 1.6251, 'learning_rate': 4.875719358393535e-07, 'epoch': 2308.64}
+
 57%|█████▋    | 37400/65536 [6:25:16<4:40:10,  1.67it/s]
 57%|█████▋    | 37401/65536 [6:25:17<4:43:42,  1.65it/s]
 57%|█████▋    | 37402/65536 [6:25:18<4:40:03,  1.67it/s]
 57%|█████▋    | 37403/65536 [6:25:18<4:47:52,  1.63it/s]
 57%|█████▋    | 37404/65536 [6:25:19<4:46:36,  1.64it/s]
 57%|█████▋    | 37405/65536 [6:25:19<4:50:16,  1.62it/s]
 57%|█████▋    | 37406/65536 [6:25:20<5:06:05,  1.53it/s]
 57%|█████▋    | 37407/65536 [6:25:21<5:03:02,  1.55it/s]
 57%|█████▋    | 37408/65536 [6:25:21<4:54:27,  1.59it/s]
 57%|█████▋    | 37409/65536 [6:25:22<4:54:30,  1.59it/s]
 57%|█████▋    | 37410/65536 [6:25:23<5:04:10,  1.54it/s]
 57%|█████▋    | 37411/65536 [6:25:23<4:57:28,  1.58it/s]
 57%|█████▋    | 37412/65536 [6:25:24<4:54:06,  1.59it/s]
 57%|█████▋    | 37413/65536 [6:25:25<4:50:28,  1.61it/s]
 57%|█████▋    | 37414/65536 [6:25:25<4:46:09,  1.64it/s]
 57%|█████▋    | 37415/65536 [6:25:26<4:48:32,  1.62it/s]
 57%|█████▋    | 37416/65536 [6:25:26<4:46:10,  1.64it/s]
 57%|█████▋    | 37417/65536 [6:25:27<4:36:26,  1.70it/s]
 57%|█████▋    | 37418/65536 [6:25:28<4:46:33,  1.64it/s]
 57%|█████▋    | 37419/65536 [6:25:28<4:45:07,  1.64it/s]
 57%|█████▋    | 37420/65536 [6:25:29<4:43:03,  1.66it/s]
                                                         
{'loss': 1.6307, 'learning_rate': 4.872964368801273e-07, 'epoch': 2309.88}
+
 57%|█████▋    | 37420/65536 [6:25:29<4:43:03,  1.66it/s]
 57%|█████▋    | 37421/65536 [6:25:29<4:39:03,  1.68it/s]
 57%|█████▋    | 37422/65536 [6:25:30<4:36:44,  1.69it/s]
 57%|█████▋    | 37423/65536 [6:25:31<4:53:03,  1.60it/s]
 57%|█████▋    | 37424/65536 [6:25:31<4:45:32,  1.64it/s]
 57%|█████▋    | 37425/65536 [6:25:32<4:49:31,  1.62it/s]
 57%|█████▋    | 37426/65536 [6:25:32<4:46:09,  1.64it/s]
 57%|█████▋    | 37427/65536 [6:25:33<4:41:35,  1.66it/s]
 57%|█████▋    | 37428/65536 [6:25:34<4:43:53,  1.65it/s]
 57%|█████▋    | 37429/65536 [6:25:34<4:39:28,  1.68it/s]
 57%|█████▋    | 37430/65536 [6:25:35<4:41:11,  1.67it/s]
 57%|█████▋    | 37431/65536 [6:25:35<4:46:01,  1.64it/s]
 57%|█████▋    | 37432/65536 [6:25:36<4:48:09,  1.63it/s]
 57%|█████▋    | 37433/65536 [6:25:37<4:49:39,  1.62it/s]
 57%|█████▋    | 37434/65536 [6:25:37<4:45:02,  1.64it/s]
 57%|█████▋    | 37435/65536 [6:25:38<4:45:21,  1.64it/s]
 57%|█████▋    | 37436/65536 [6:25:39<4:44:14,  1.65it/s]
 57%|█████▋    | 37437/65536 [6:25:39<4:43:08,  1.65it/s]
 57%|█████▋    | 37438/65536 [6:25:40<4:50:50,  1.61it/s]
 57%|█████▋    | 37439/65536 [6:25:40<4:59:54,  1.56it/s]
 57%|█████▋    | 37440/65536 [6:25:41<4:54:09,  1.59it/s]
                                                         
{'loss': 1.6564, 'learning_rate': 4.870209379209012e-07, 'epoch': 2311.11}
+
 57%|█████▋    | 37440/65536 [6:25:41<4:54:09,  1.59it/s]
 57%|█████▋    | 37441/65536 [6:25:42<4:48:06,  1.63it/s]
 57%|█████▋    | 37442/65536 [6:25:42<4:53:43,  1.59it/s]
 57%|█████▋    | 37443/65536 [6:25:43<4:44:31,  1.65it/s]
 57%|█████▋    | 37444/65536 [6:25:43<4:48:02,  1.63it/s]
 57%|█████▋    | 37445/65536 [6:25:44<4:44:16,  1.65it/s]
 57%|█████▋    | 37446/65536 [6:25:45<4:48:33,  1.62it/s]
 57%|█████▋    | 37447/65536 [6:25:45<4:57:45,  1.57it/s]
 57%|█████▋    | 37448/65536 [6:25:46<4:53:23,  1.60it/s]
 57%|█████▋    | 37449/65536 [6:25:47<4:48:18,  1.62it/s]
 57%|█████▋    | 37450/65536 [6:25:47<4:47:33,  1.63it/s]
 57%|█████▋    | 37451/65536 [6:25:48<4:38:53,  1.68it/s]
 57%|█████▋    | 37452/65536 [6:25:48<4:39:04,  1.68it/s]
 57%|█████▋    | 37453/65536 [6:25:49<4:44:17,  1.65it/s]
 57%|█████▋    | 37454/65536 [6:25:50<4:42:06,  1.66it/s]
 57%|█████▋    | 37455/65536 [6:25:50<4:48:24,  1.62it/s]
 57%|█████▋    | 37456/65536 [6:25:51<4:48:16,  1.62it/s]
 57%|█████▋    | 37457/65536 [6:25:51<4:47:31,  1.63it/s]
 57%|█████▋    | 37458/65536 [6:25:52<4:48:07,  1.62it/s]
 57%|█████▋    | 37459/65536 [6:25:53<4:44:21,  1.65it/s]
 57%|█████▋    | 37460/65536 [6:25:53<4:42:57,  1.65it/s]
                                                         
{'loss': 1.6458, 'learning_rate': 4.867454389616751e-07, 'epoch': 2312.35}
+
 57%|█████▋    | 37460/65536 [6:25:53<4:42:57,  1.65it/s]
 57%|█████▋    | 37461/65536 [6:25:54<4:43:14,  1.65it/s]
 57%|█████▋    | 37462/65536 [6:25:55<5:05:43,  1.53it/s]
 57%|█████▋    | 37463/65536 [6:25:55<4:54:12,  1.59it/s]
 57%|█████▋    | 37464/65536 [6:25:56<4:48:41,  1.62it/s]
 57%|█████▋    | 37465/65536 [6:25:56<4:50:34,  1.61it/s]
 57%|█████▋    | 37466/65536 [6:25:57<4:42:51,  1.65it/s]
 57%|█████▋    | 37467/65536 [6:25:58<4:42:47,  1.65it/s]
 57%|█████▋    | 37468/65536 [6:25:58<4:47:36,  1.63it/s]
 57%|█████▋    | 37469/65536 [6:25:59<4:42:49,  1.65it/s]
 57%|█████▋    | 37470/65536 [6:25:59<4:44:10,  1.65it/s]
 57%|█████▋    | 37471/65536 [6:26:00<5:00:53,  1.55it/s]
 57%|█████▋    | 37472/65536 [6:26:01<4:53:49,  1.59it/s]
 57%|█████▋    | 37473/65536 [6:26:01<4:53:46,  1.59it/s]
 57%|█████▋    | 37474/65536 [6:26:02<4:53:27,  1.59it/s]
 57%|█████▋    | 37475/65536 [6:26:03<4:51:43,  1.60it/s]
 57%|█████▋    | 37476/65536 [6:26:03<4:51:48,  1.60it/s]
 57%|█████▋    | 37477/65536 [6:26:04<4:52:09,  1.60it/s]
 57%|█████▋    | 37478/65536 [6:26:04<4:45:20,  1.64it/s]
 57%|█████▋    | 37479/65536 [6:26:05<4:45:24,  1.64it/s]
 57%|█████▋    | 37480/65536 [6:26:06<4:40:57,  1.66it/s]
                                                         
{'loss': 1.5916, 'learning_rate': 4.864699400024488e-07, 'epoch': 2313.58}
+
 57%|█████▋    | 37480/65536 [6:26:06<4:40:57,  1.66it/s]
 57%|█████▋    | 37481/65536 [6:26:06<4:41:19,  1.66it/s]
 57%|█████▋    | 37482/65536 [6:26:07<4:35:28,  1.70it/s]
 57%|█████▋    | 37483/65536 [6:26:07<4:34:21,  1.70it/s]
 57%|█████▋    | 37484/65536 [6:26:08<4:40:10,  1.67it/s]
 57%|█████▋    | 37485/65536 [6:26:09<4:41:46,  1.66it/s]
 57%|█████▋    | 37486/65536 [6:26:09<4:49:49,  1.61it/s]
 57%|█████▋    | 37487/65536 [6:26:10<4:57:47,  1.57it/s]
 57%|█████▋    | 37488/65536 [6:26:11<4:49:36,  1.61it/s]
 57%|█████▋    | 37489/65536 [6:26:11<4:50:07,  1.61it/s]
 57%|█████▋    | 37490/65536 [6:26:12<4:42:55,  1.65it/s]
 57%|█████▋    | 37491/65536 [6:26:12<4:50:50,  1.61it/s]
 57%|█████▋    | 37492/65536 [6:26:13<4:46:52,  1.63it/s]
 57%|█████▋    | 37493/65536 [6:26:14<4:41:34,  1.66it/s]
 57%|█████▋    | 37494/65536 [6:26:14<4:37:49,  1.68it/s]
 57%|█████▋    | 37495/65536 [6:26:15<4:34:48,  1.70it/s]
 57%|█████▋    | 37496/65536 [6:26:15<4:39:46,  1.67it/s]
 57%|█████▋    | 37497/65536 [6:26:16<4:40:56,  1.66it/s]
 57%|█████▋    | 37498/65536 [6:26:17<4:46:05,  1.63it/s]
 57%|█████▋    | 37499/65536 [6:26:17<4:58:10,  1.57it/s]
 57%|█████▋    | 37500/65536 [6:26:18<4:55:41,  1.58it/s]
                                                         
{'loss': 1.5901, 'learning_rate': 4.861944410432227e-07, 'epoch': 2314.81}
+
 57%|█████▋    | 37500/65536 [6:26:18<4:55:41,  1.58it/s]
 57%|█████▋    | 37501/65536 [6:26:18<4:51:13,  1.60it/s]
 57%|█████▋    | 37502/65536 [6:26:19<4:43:58,  1.65it/s]
 57%|█████▋    | 37503/65536 [6:26:20<4:46:44,  1.63it/s]
 57%|█████▋    | 37504/65536 [6:26:20<4:52:20,  1.60it/s]
 57%|█████▋    | 37505/65536 [6:26:21<4:50:27,  1.61it/s]
 57%|█████▋    | 37506/65536 [6:26:22<4:50:46,  1.61it/s]
 57%|█████▋    | 37507/65536 [6:26:22<4:56:50,  1.57it/s]
 57%|█████▋    | 37508/65536 [6:26:23<4:55:17,  1.58it/s]
 57%|█████▋    | 37509/65536 [6:26:24<5:01:24,  1.55it/s]
 57%|█████▋    | 37510/65536 [6:26:24<4:54:24,  1.59it/s]
 57%|█████▋    | 37511/65536 [6:26:25<4:55:38,  1.58it/s]
 57%|█████▋    | 37512/65536 [6:26:25<4:53:33,  1.59it/s]
 57%|█████▋    | 37513/65536 [6:26:26<4:48:23,  1.62it/s]
 57%|█████▋    | 37514/65536 [6:26:27<4:43:11,  1.65it/s]
 57%|█████▋    | 37515/65536 [6:26:27<4:41:35,  1.66it/s]
 57%|█████▋    | 37516/65536 [6:26:28<4:39:03,  1.67it/s]
 57%|█████▋    | 37517/65536 [6:26:28<4:38:10,  1.68it/s]
 57%|█████▋    | 37518/65536 [6:26:29<4:33:11,  1.71it/s]
 57%|█████▋    | 37519/65536 [6:26:30<4:39:05,  1.67it/s]
 57%|█████▋    | 37520/65536 [6:26:30<4:49:15,  1.61it/s]
                                                         
{'loss': 1.6709, 'learning_rate': 4.859189420839965e-07, 'epoch': 2316.05}
+
 57%|█████▋    | 37520/65536 [6:26:30<4:49:15,  1.61it/s]
 57%|█████▋    | 37521/65536 [6:26:31<4:52:47,  1.59it/s]
 57%|█████▋    | 37522/65536 [6:26:32<5:02:53,  1.54it/s]
 57%|█████▋    | 37523/65536 [6:26:32<5:00:53,  1.55it/s]
 57%|█████▋    | 37524/65536 [6:26:33<4:55:20,  1.58it/s]
 57%|█████▋    | 37525/65536 [6:26:33<4:56:44,  1.57it/s]
 57%|█████▋    | 37526/65536 [6:26:34<4:57:51,  1.57it/s]
 57%|█████▋    | 37527/65536 [6:26:35<4:52:05,  1.60it/s]
 57%|█████▋    | 37528/65536 [6:26:35<4:47:11,  1.63it/s]
 57%|█████▋    | 37529/65536 [6:26:36<4:49:03,  1.61it/s]
 57%|█████▋    | 37530/65536 [6:26:36<4:42:23,  1.65it/s]
 57%|█████▋    | 37531/65536 [6:26:37<4:42:09,  1.65it/s]
 57%|█████▋    | 37532/65536 [6:26:38<4:37:46,  1.68it/s]
 57%|█████▋    | 37533/65536 [6:26:38<4:36:25,  1.69it/s]
 57%|█████▋    | 37534/65536 [6:26:39<4:37:43,  1.68it/s]
 57%|█████▋    | 37535/65536 [6:26:39<4:36:56,  1.69it/s]
 57%|█████▋    | 37536/65536 [6:26:40<4:46:25,  1.63it/s]
 57%|█████▋    | 37537/65536 [6:26:41<4:46:35,  1.63it/s]
 57%|█████▋    | 37538/65536 [6:26:41<4:51:00,  1.60it/s]
 57%|█████▋    | 37539/65536 [6:26:42<4:44:50,  1.64it/s]
 57%|█████▋    | 37540/65536 [6:26:42<4:41:02,  1.66it/s]
                                                         
{'loss': 1.6112, 'learning_rate': 4.856434431247704e-07, 'epoch': 2317.28}
+
 57%|█████▋    | 37540/65536 [6:26:42<4:41:02,  1.66it/s]
 57%|█████▋    | 37541/65536 [6:26:43<4:45:30,  1.63it/s]
 57%|█████▋    | 37542/65536 [6:26:44<4:45:02,  1.64it/s]
 57%|█████▋    | 37543/65536 [6:26:44<4:41:48,  1.66it/s]
 57%|█████▋    | 37544/65536 [6:26:45<4:48:48,  1.62it/s]
 57%|█████▋    | 37545/65536 [6:26:46<4:51:38,  1.60it/s]
 57%|█████▋    | 37546/65536 [6:26:46<4:49:09,  1.61it/s]
 57%|█████▋    | 37547/65536 [6:26:47<4:45:59,  1.63it/s]
 57%|█████▋    | 37548/65536 [6:26:47<4:51:47,  1.60it/s]
 57%|█████▋    | 37549/65536 [6:26:48<4:49:08,  1.61it/s]
 57%|█████▋    | 37550/65536 [6:26:49<4:46:19,  1.63it/s]
 57%|█████▋    | 37551/65536 [6:26:49<4:44:17,  1.64it/s]
 57%|█████▋    | 37552/65536 [6:26:50<4:52:15,  1.60it/s]
 57%|█████▋    | 37553/65536 [6:26:51<4:44:27,  1.64it/s]
 57%|█████▋    | 37554/65536 [6:26:51<4:43:32,  1.64it/s]
 57%|█████▋    | 37555/65536 [6:26:52<4:40:06,  1.66it/s]
 57%|█████▋    | 37556/65536 [6:26:52<4:42:42,  1.65it/s]
 57%|█████▋    | 37557/65536 [6:26:53<4:51:22,  1.60it/s]
 57%|█████▋    | 37558/65536 [6:26:54<4:38:17,  1.68it/s]
 57%|█████▋    | 37559/65536 [6:26:54<4:47:39,  1.62it/s]
 57%|█████▋    | 37560/65536 [6:26:55<4:48:03,  1.62it/s]
                                                         
{'loss': 1.6341, 'learning_rate': 4.853679441655442e-07, 'epoch': 2318.52}
+
 57%|█████▋    | 37560/65536 [6:26:55<4:48:03,  1.62it/s]
 57%|█████▋    | 37561/65536 [6:26:55<4:55:30,  1.58it/s]
 57%|█████▋    | 37562/65536 [6:26:56<4:50:11,  1.61it/s]
 57%|█████▋    | 37563/65536 [6:26:57<4:51:08,  1.60it/s]
 57%|█████▋    | 37564/65536 [6:26:57<4:48:41,  1.61it/s]
 57%|█████▋    | 37565/65536 [6:26:58<4:41:14,  1.66it/s]
 57%|█████▋    | 37566/65536 [6:26:59<4:47:37,  1.62it/s]
 57%|█████▋    | 37567/65536 [6:26:59<4:44:51,  1.64it/s]
 57%|█████▋    | 37568/65536 [6:27:00<4:54:22,  1.58it/s]
 57%|█████▋    | 37569/65536 [6:27:00<4:59:09,  1.56it/s]
 57%|█████▋    | 37570/65536 [6:27:01<4:53:33,  1.59it/s]
 57%|█████▋    | 37571/65536 [6:27:02<4:49:09,  1.61it/s]
 57%|█████▋    | 37572/65536 [6:27:02<4:47:02,  1.62it/s]
 57%|█████▋    | 37573/65536 [6:27:03<4:49:36,  1.61it/s]
 57%|█████▋    | 37574/65536 [6:27:03<4:43:29,  1.64it/s]
 57%|█████▋    | 37575/65536 [6:27:04<4:46:45,  1.63it/s]
 57%|█████▋    | 37576/65536 [6:27:05<4:49:07,  1.61it/s]
 57%|█████▋    | 37577/65536 [6:27:05<4:59:11,  1.56it/s]
 57%|█████▋    | 37578/65536 [6:27:06<4:56:09,  1.57it/s]
 57%|█████▋    | 37579/65536 [6:27:07<4:53:15,  1.59it/s]
 57%|█████▋    | 37580/65536 [6:27:07<4:49:27,  1.61it/s]
                                                         
{'loss': 1.6095, 'learning_rate': 4.850924452063181e-07, 'epoch': 2319.75}
+
 57%|█████▋    | 37580/65536 [6:27:07<4:49:27,  1.61it/s]
 57%|█████▋    | 37581/65536 [6:27:08<4:45:39,  1.63it/s]
 57%|█████▋    | 37582/65536 [6:27:08<4:40:43,  1.66it/s]
 57%|█████▋    | 37583/65536 [6:27:09<4:37:03,  1.68it/s]
 57%|█████▋    | 37584/65536 [6:27:10<4:39:29,  1.67it/s]
 57%|█████▋    | 37585/65536 [6:27:10<4:58:20,  1.56it/s]
 57%|█████▋    | 37586/65536 [6:27:11<4:51:11,  1.60it/s]
 57%|█████▋    | 37587/65536 [6:27:12<4:47:48,  1.62it/s]
 57%|█████▋    | 37588/65536 [6:27:12<4:42:26,  1.65it/s]
 57%|█████▋    | 37589/65536 [6:27:13<4:43:05,  1.65it/s]
 57%|█████▋    | 37590/65536 [6:27:13<4:46:30,  1.63it/s]
 57%|█████▋    | 37591/65536 [6:27:14<4:45:15,  1.63it/s]
 57%|█████▋    | 37592/65536 [6:27:15<4:51:51,  1.60it/s]
 57%|█████▋    | 37593/65536 [6:27:15<4:50:16,  1.60it/s]
 57%|█████▋    | 37594/65536 [6:27:16<4:43:54,  1.64it/s]
 57%|█████▋    | 37595/65536 [6:27:16<4:48:27,  1.61it/s]
 57%|█████▋    | 37596/65536 [6:27:17<4:44:07,  1.64it/s]
 57%|█████▋    | 37597/65536 [6:27:18<4:40:23,  1.66it/s]
 57%|█████▋    | 37598/65536 [6:27:18<4:38:03,  1.67it/s]
 57%|█████▋    | 37599/65536 [6:27:19<4:40:29,  1.66it/s]
 57%|█████▋    | 37600/65536 [6:27:20<4:47:50,  1.62it/s]
                                                         
{'loss': 1.6618, 'learning_rate': 4.848169462470919e-07, 'epoch': 2320.99}
+
 57%|█████▋    | 37600/65536 [6:27:20<4:47:50,  1.62it/s]
 57%|█████▋    | 37601/65536 [6:27:20<4:52:21,  1.59it/s]
 57%|█████▋    | 37602/65536 [6:27:21<4:44:54,  1.63it/s]
 57%|█████▋    | 37603/65536 [6:27:21<4:49:44,  1.61it/s]
 57%|█████▋    | 37604/65536 [6:27:22<4:52:03,  1.59it/s]
 57%|█████▋    | 37605/65536 [6:27:23<4:52:32,  1.59it/s]
 57%|█████▋    | 37606/65536 [6:27:23<4:56:32,  1.57it/s]
 57%|█████▋    | 37607/65536 [6:27:24<4:54:47,  1.58it/s]
 57%|█████▋    | 37608/65536 [6:27:25<4:50:22,  1.60it/s]
 57%|█████▋    | 37609/65536 [6:27:25<4:51:07,  1.60it/s]
 57%|█████▋    | 37610/65536 [6:27:26<4:55:09,  1.58it/s]
 57%|█████▋    | 37611/65536 [6:27:26<4:47:25,  1.62it/s]
 57%|█████▋    | 37612/65536 [6:27:27<4:45:19,  1.63it/s]
 57%|█████▋    | 37613/65536 [6:27:28<4:35:58,  1.69it/s]
 57%|█████▋    | 37614/65536 [6:27:28<4:34:46,  1.69it/s]
 57%|█████▋    | 37615/65536 [6:27:29<4:38:54,  1.67it/s]
 57%|█████▋    | 37616/65536 [6:27:29<4:41:06,  1.66it/s]
 57%|█████▋    | 37617/65536 [6:27:30<4:53:08,  1.59it/s]
 57%|█████▋    | 37618/65536 [6:27:31<4:57:14,  1.57it/s]
 57%|█████▋    | 37619/65536 [6:27:31<4:55:51,  1.57it/s]
 57%|█████▋    | 37620/65536 [6:27:32<4:53:31,  1.59it/s]
                                                         
{'loss': 1.6223, 'learning_rate': 4.845414472878657e-07, 'epoch': 2322.22}
+
 57%|█████▋    | 37620/65536 [6:27:32<4:53:31,  1.59it/s]
 57%|█████▋    | 37621/65536 [6:27:33<4:48:56,  1.61it/s]
 57%|█████▋    | 37622/65536 [6:27:33<4:43:51,  1.64it/s]
 57%|█████▋    | 37623/65536 [6:27:34<4:40:58,  1.66it/s]
 57%|█████▋    | 37624/65536 [6:27:34<4:40:54,  1.66it/s]
 57%|█████▋    | 37625/65536 [6:27:35<4:40:39,  1.66it/s]
 57%|█████▋    | 37626/65536 [6:27:36<4:37:47,  1.67it/s]
 57%|█████▋    | 37627/65536 [6:27:36<4:36:22,  1.68it/s]
 57%|█████▋    | 37628/65536 [6:27:37<4:31:34,  1.71it/s]
 57%|█████▋    | 37629/65536 [6:27:37<4:36:31,  1.68it/s]
 57%|█████▋    | 37630/65536 [6:27:38<4:42:50,  1.64it/s]
 57%|█████▋    | 37631/65536 [6:27:39<4:43:43,  1.64it/s]
 57%|█████▋    | 37632/65536 [6:27:39<4:39:38,  1.66it/s]
 57%|█████▋    | 37633/65536 [6:27:40<4:49:14,  1.61it/s]
 57%|█████▋    | 37634/65536 [6:27:40<4:45:07,  1.63it/s]
 57%|█████▋    | 37635/65536 [6:27:41<4:38:05,  1.67it/s]
 57%|█████▋    | 37636/65536 [6:27:42<4:35:37,  1.69it/s]
 57%|█████▋    | 37637/65536 [6:27:42<4:35:13,  1.69it/s]
 57%|█████▋    | 37638/65536 [6:27:43<4:37:01,  1.68it/s]
 57%|█████▋    | 37639/65536 [6:27:43<4:41:00,  1.65it/s]
 57%|█████▋    | 37640/65536 [6:27:44<4:45:08,  1.63it/s]
                                                         
{'loss': 1.6658, 'learning_rate': 4.842659483286396e-07, 'epoch': 2323.46}
+
 57%|█████▋    | 37640/65536 [6:27:44<4:45:08,  1.63it/s]
 57%|█████▋    | 37641/65536 [6:27:45<4:54:20,  1.58it/s]
 57%|█████▋    | 37642/65536 [6:27:45<4:50:49,  1.60it/s]
 57%|█████▋    | 37643/65536 [6:27:46<4:46:46,  1.62it/s]
 57%|█████▋    | 37644/65536 [6:27:46<4:45:46,  1.63it/s]
 57%|█████▋    | 37645/65536 [6:27:47<4:43:24,  1.64it/s]
 57%|█████▋    | 37646/65536 [6:27:48<4:39:15,  1.66it/s]
 57%|█████▋    | 37647/65536 [6:27:48<4:47:56,  1.61it/s]
 57%|█████▋    | 37648/65536 [6:27:49<4:48:04,  1.61it/s]
 57%|█████▋    | 37649/65536 [6:27:50<5:04:00,  1.53it/s]
 57%|█████▋    | 37650/65536 [6:27:50<5:00:35,  1.55it/s]
 57%|█████▋    | 37651/65536 [6:27:51<4:53:11,  1.59it/s]
 57%|█████▋    | 37652/65536 [6:27:51<4:46:27,  1.62it/s]
 57%|█████▋    | 37653/65536 [6:27:52<4:41:03,  1.65it/s]
 57%|█████▋    | 37654/65536 [6:27:53<4:37:47,  1.67it/s]
 57%|█████▋    | 37655/65536 [6:27:53<4:38:45,  1.67it/s]
 57%|█████▋    | 37656/65536 [6:27:54<4:43:07,  1.64it/s]
 57%|█████▋    | 37657/65536 [6:27:54<4:43:21,  1.64it/s]
 57%|█████▋    | 37658/65536 [6:27:55<4:40:24,  1.66it/s]
 57%|█████▋    | 37659/65536 [6:27:56<4:48:23,  1.61it/s]
 57%|█████▋    | 37660/65536 [6:27:56<4:51:36,  1.59it/s]
                                                         
{'loss': 1.5882, 'learning_rate': 4.839904493694134e-07, 'epoch': 2324.69}
+
 57%|█████▋    | 37660/65536 [6:27:56<4:51:36,  1.59it/s]
 57%|█████▋    | 37661/65536 [6:27:57<4:56:31,  1.57it/s]
 57%|█████▋    | 37662/65536 [6:27:58<4:53:29,  1.58it/s]
 57%|█████▋    | 37663/65536 [6:27:58<4:54:11,  1.58it/s]
 57%|█████▋    | 37664/65536 [6:27:59<4:51:51,  1.59it/s]
 57%|█████▋    | 37665/65536 [6:28:00<4:45:38,  1.63it/s]
 57%|█████▋    | 37666/65536 [6:28:00<5:08:35,  1.51it/s]
 57%|█████▋    | 37667/65536 [6:28:01<4:59:29,  1.55it/s]
 57%|█████▋    | 37668/65536 [6:28:01<4:48:44,  1.61it/s]
 57%|█████▋    | 37669/65536 [6:28:02<4:46:25,  1.62it/s]
 57%|█████▋    | 37670/65536 [6:28:03<4:47:05,  1.62it/s]
 57%|█████▋    | 37671/65536 [6:28:03<4:45:29,  1.63it/s]
 57%|█████▋    | 37672/65536 [6:28:04<4:45:07,  1.63it/s]
 57%|█████▋    | 37673/65536 [6:28:04<4:43:11,  1.64it/s]
 57%|█████▋    | 37674/65536 [6:28:05<4:42:42,  1.64it/s]
 57%|█████▋    | 37675/65536 [6:28:06<4:43:40,  1.64it/s]
 57%|█████▋    | 37676/65536 [6:28:06<4:45:29,  1.63it/s]
 57%|█████▋    | 37677/65536 [6:28:07<4:38:23,  1.67it/s]
 57%|█████▋    | 37678/65536 [6:28:08<4:44:06,  1.63it/s]
 57%|█████▋    | 37679/65536 [6:28:08<4:38:14,  1.67it/s]
 57%|█████▋    | 37680/65536 [6:28:09<4:41:12,  1.65it/s]
                                                         
{'loss': 1.586, 'learning_rate': 4.837149504101873e-07, 'epoch': 2325.93}
+
 57%|█████▋    | 37680/65536 [6:28:09<4:41:12,  1.65it/s]
 57%|█████▋    | 37681/65536 [6:28:09<4:44:11,  1.63it/s]
 57%|█████▋    | 37682/65536 [6:28:10<4:55:09,  1.57it/s]
 57%|█████▋    | 37683/65536 [6:28:11<4:49:09,  1.61it/s]
 58%|█████▊    | 37684/65536 [6:28:11<4:45:50,  1.62it/s]
 58%|█████▊    | 37685/65536 [6:28:12<4:43:57,  1.63it/s]
 58%|█████▊    | 37686/65536 [6:28:12<4:42:28,  1.64it/s]
 58%|█████▊    | 37687/65536 [6:28:13<4:40:47,  1.65it/s]
 58%|█████▊    | 37688/65536 [6:28:14<4:42:31,  1.64it/s]
 58%|█████▊    | 37689/65536 [6:28:14<4:44:14,  1.63it/s]
 58%|█████▊    | 37690/65536 [6:28:15<4:42:05,  1.65it/s]
 58%|█████▊    | 37691/65536 [6:28:16<4:49:57,  1.60it/s]
 58%|█████▊    | 37692/65536 [6:28:16<4:53:54,  1.58it/s]
 58%|█████▊    | 37693/65536 [6:28:17<4:53:30,  1.58it/s]
 58%|█████▊    | 37694/65536 [6:28:17<4:57:23,  1.56it/s]
 58%|█████▊    | 37695/65536 [6:28:18<4:46:51,  1.62it/s]
 58%|█████▊    | 37696/65536 [6:28:19<4:42:58,  1.64it/s]
 58%|█████▊    | 37697/65536 [6:28:19<4:33:59,  1.69it/s]
 58%|█████▊    | 37698/65536 [6:28:20<4:43:23,  1.64it/s]
 58%|█████▊    | 37699/65536 [6:28:21<4:54:25,  1.58it/s]
 58%|█████▊    | 37700/65536 [6:28:21<4:54:11,  1.58it/s]
                                                         
{'loss': 1.6513, 'learning_rate': 4.834394514509611e-07, 'epoch': 2327.16}
+
 58%|█████▊    | 37700/65536 [6:28:21<4:54:11,  1.58it/s]
 58%|█████▊    | 37701/65536 [6:28:22<4:55:17,  1.57it/s]
 58%|█████▊    | 37702/65536 [6:28:22<4:58:07,  1.56it/s]
 58%|█████▊    | 37703/65536 [6:28:23<4:50:41,  1.60it/s]
 58%|█████▊    | 37704/65536 [6:28:24<4:45:50,  1.62it/s]
 58%|█████▊    | 37705/65536 [6:28:24<4:41:19,  1.65it/s]
 58%|█████▊    | 37706/65536 [6:28:25<4:38:06,  1.67it/s]
 58%|█████▊    | 37707/65536 [6:28:25<4:42:23,  1.64it/s]
 58%|█████▊    | 37708/65536 [6:28:26<4:39:04,  1.66it/s]
 58%|█████▊    | 37709/65536 [6:28:27<4:39:24,  1.66it/s]
 58%|█████▊    | 37710/65536 [6:28:27<4:41:03,  1.65it/s]
 58%|█████▊    | 37711/65536 [6:28:28<4:42:39,  1.64it/s]
 58%|█████▊    | 37712/65536 [6:28:28<4:42:05,  1.64it/s]
 58%|█████▊    | 37713/65536 [6:28:29<4:41:09,  1.65it/s]
 58%|█████▊    | 37714/65536 [6:28:30<4:57:42,  1.56it/s]
 58%|█████▊    | 37715/65536 [6:28:30<4:49:55,  1.60it/s]
 58%|█████▊    | 37716/65536 [6:28:31<4:49:18,  1.60it/s]
 58%|█████▊    | 37717/65536 [6:28:32<4:57:55,  1.56it/s]
 58%|█████▊    | 37718/65536 [6:28:32<4:52:42,  1.58it/s]
 58%|█████▊    | 37719/65536 [6:28:33<4:54:16,  1.58it/s]
 58%|█████▊    | 37720/65536 [6:28:34<4:56:40,  1.56it/s]
                                                         
{'loss': 1.602, 'learning_rate': 4.83163952491735e-07, 'epoch': 2328.4}
+
 58%|█████▊    | 37720/65536 [6:28:34<4:56:40,  1.56it/s]
 58%|█████▊    | 37721/65536 [6:28:34<4:48:54,  1.60it/s]
 58%|█████▊    | 37722/65536 [6:28:35<4:42:46,  1.64it/s]
 58%|█████▊    | 37723/65536 [6:28:35<4:35:44,  1.68it/s]
 58%|█████▊    | 37724/65536 [6:28:36<4:30:39,  1.71it/s]
 58%|█████▊    | 37725/65536 [6:28:36<4:31:31,  1.71it/s]
 58%|█████▊    | 37726/65536 [6:28:37<4:38:23,  1.66it/s]
 58%|█████▊    | 37727/65536 [6:28:38<4:35:34,  1.68it/s]
 58%|█████▊    | 37728/65536 [6:28:38<4:37:39,  1.67it/s]
 58%|█████▊    | 37729/65536 [6:28:39<4:37:44,  1.67it/s]
 58%|█████▊    | 37730/65536 [6:28:40<4:50:59,  1.59it/s]
 58%|█████▊    | 37731/65536 [6:28:40<4:55:18,  1.57it/s]
 58%|█████▊    | 37732/65536 [6:28:41<4:54:40,  1.57it/s]
 58%|█████▊    | 37733/65536 [6:28:41<4:45:09,  1.62it/s]
 58%|█████▊    | 37734/65536 [6:28:42<4:44:09,  1.63it/s]
 58%|█████▊    | 37735/65536 [6:28:43<4:39:14,  1.66it/s]
 58%|█████▊    | 37736/65536 [6:28:43<4:40:14,  1.65it/s]
 58%|█████▊    | 37737/65536 [6:28:44<4:33:24,  1.69it/s]
 58%|█████▊    | 37738/65536 [6:28:44<4:39:38,  1.66it/s]
 58%|█████▊    | 37739/65536 [6:28:45<4:40:31,  1.65it/s]
 58%|█████▊    | 37740/65536 [6:28:46<4:45:51,  1.62it/s]
                                                         
{'loss': 1.6642, 'learning_rate': 4.828884535325089e-07, 'epoch': 2329.63}
+
 58%|█████▊    | 37740/65536 [6:28:46<4:45:51,  1.62it/s]
 58%|█████▊    | 37741/65536 [6:28:46<4:42:27,  1.64it/s]
 58%|█████▊    | 37742/65536 [6:28:47<4:49:59,  1.60it/s]
 58%|█████▊    | 37743/65536 [6:28:48<4:46:58,  1.61it/s]
 58%|█████▊    | 37744/65536 [6:28:48<4:46:29,  1.62it/s]
 58%|█████▊    | 37745/65536 [6:28:49<4:50:26,  1.59it/s]
 58%|█████▊    | 37746/65536 [6:28:49<4:46:12,  1.62it/s]
 58%|█████▊    | 37747/65536 [6:28:50<4:50:24,  1.59it/s]
 58%|████���▊    | 37748/65536 [6:28:51<4:47:18,  1.61it/s]
 58%|█████▊    | 37749/65536 [6:28:51<4:49:40,  1.60it/s]
 58%|█████▊    | 37750/65536 [6:28:52<4:42:22,  1.64it/s]
 58%|█████▊    | 37751/65536 [6:28:52<4:42:27,  1.64it/s]
 58%|█████▊    | 37752/65536 [6:28:53<4:47:30,  1.61it/s]
 58%|█████▊    | 37753/65536 [6:28:54<4:50:32,  1.59it/s]
 58%|█████▊    | 37754/65536 [6:28:54<4:43:39,  1.63it/s]
 58%|█████▊    | 37755/65536 [6:28:55<4:48:26,  1.61it/s]
 58%|█████▊    | 37756/65536 [6:28:56<4:49:40,  1.60it/s]
 58%|█████▊    | 37757/65536 [6:28:56<4:39:00,  1.66it/s]
 58%|█████▊    | 37758/65536 [6:28:57<4:44:25,  1.63it/s]
 58%|█████▊    | 37759/65536 [6:28:57<4:43:20,  1.63it/s]
 58%|█████▊    | 37760/65536 [6:28:58<4:46:15,  1.62it/s]
                                                         
{'loss': 1.6289, 'learning_rate': 4.826129545732827e-07, 'epoch': 2330.86}
+
 58%|█████▊    | 37760/65536 [6:28:58<4:46:15,  1.62it/s]
 58%|█████▊    | 37761/65536 [6:28:59<4:40:41,  1.65it/s]
 58%|█████▊    | 37762/65536 [6:28:59<4:43:40,  1.63it/s]
 58%|█████▊    | 37763/65536 [6:29:00<4:53:45,  1.58it/s]
 58%|█████▊    | 37764/65536 [6:29:00<4:40:50,  1.65it/s]
 58%|█████▊    | 37765/65536 [6:29:01<4:41:40,  1.64it/s]
 58%|█████▊    | 37766/65536 [6:29:02<4:38:56,  1.66it/s]
 58%|█████▊    | 37767/65536 [6:29:02<4:37:46,  1.67it/s]
 58%|█████▊    | 37768/65536 [6:29:03<4:43:27,  1.63it/s]
 58%|█████▊    | 37769/65536 [6:29:04<4:45:18,  1.62it/s]
 58%|█████▊    | 37770/65536 [6:29:04<4:41:05,  1.65it/s]
 58%|█████▊    | 37771/65536 [6:29:05<4:52:57,  1.58it/s]
 58%|█████▊    | 37772/65536 [6:29:05<4:55:29,  1.57it/s]
 58%|█████▊    | 37773/65536 [6:29:06<4:52:29,  1.58it/s]
 58%|█████▊    | 37774/65536 [6:29:07<4:49:28,  1.60it/s]
 58%|█████▊    | 37775/65536 [6:29:07<4:51:47,  1.59it/s]
 58%|█████▊    | 37776/65536 [6:29:08<4:48:03,  1.61it/s]
 58%|█████▊    | 37777/65536 [6:29:09<4:42:16,  1.64it/s]
 58%|█████▊    | 37778/65536 [6:29:09<4:39:28,  1.66it/s]
 58%|█████▊    | 37779/65536 [6:29:10<4:49:01,  1.60it/s]
 58%|█████▊    | 37780/65536 [6:29:10<4:43:17,  1.63it/s]
                                                         
{'loss': 1.6072, 'learning_rate': 4.823374556140566e-07, 'epoch': 2332.1}
+
 58%|█████▊    | 37780/65536 [6:29:10<4:43:17,  1.63it/s]
 58%|█████▊    | 37781/65536 [6:29:11<4:49:54,  1.60it/s]
 58%|█████▊    | 37782/65536 [6:29:12<4:44:11,  1.63it/s]
 58%|█████▊    | 37783/65536 [6:29:12<4:42:33,  1.64it/s]
 58%|█████▊    | 37784/65536 [6:29:13<4:42:00,  1.64it/s]
 58%|█████▊    | 37785/65536 [6:29:13<4:38:59,  1.66it/s]
 58%|█████▊    | 37786/65536 [6:29:14<4:35:31,  1.68it/s]
 58%|█████▊    | 37787/65536 [6:29:15<4:35:19,  1.68it/s]
 58%|█████▊    | 37788/65536 [6:29:15<4:46:54,  1.61it/s]
 58%|█████▊    | 37789/65536 [6:29:16<4:57:43,  1.55it/s]
 58%|█████▊    | 37790/65536 [6:29:17<4:49:49,  1.60it/s]
 58%|█████▊    | 37791/65536 [6:29:17<4:46:00,  1.62it/s]
 58%|█████▊    | 37792/65536 [6:29:18<4:43:22,  1.63it/s]
 58%|█████▊    | 37793/65536 [6:29:18<4:41:06,  1.64it/s]
 58%|█████▊    | 37794/65536 [6:29:19<4:41:51,  1.64it/s]
 58%|█████▊    | 37795/65536 [6:29:20<4:48:42,  1.60it/s]
 58%|█████▊    | 37796/65536 [6:29:20<4:50:05,  1.59it/s]
 58%|█████▊    | 37797/65536 [6:29:21<4:50:33,  1.59it/s]
 58%|█████▊    | 37798/65536 [6:29:21<4:43:22,  1.63it/s]
 58%|█████▊    | 37799/65536 [6:29:22<4:50:10,  1.59it/s]
 58%|█████▊    | 37800/65536 [6:29:23<4:50:58,  1.59it/s]
                                                         
{'loss': 1.6095, 'learning_rate': 4.820619566548304e-07, 'epoch': 2333.33}
+
 58%|█████▊    | 37800/65536 [6:29:23<4:50:58,  1.59it/s]
 58%|█████▊    | 37801/65536 [6:29:24<5:05:13,  1.51it/s]
 58%|█████▊    | 37802/65536 [6:29:24<4:53:56,  1.57it/s]
 58%|█████▊    | 37803/65536 [6:29:25<4:43:56,  1.63it/s]
 58%|█████▊    | 37804/65536 [6:29:25<4:37:54,  1.66it/s]
 58%|█████▊    | 37805/65536 [6:29:26<4:43:15,  1.63it/s]
 58%|█████▊    | 37806/65536 [6:29:26<4:40:29,  1.65it/s]
 58%|█████▊    | 37807/65536 [6:29:27<4:36:06,  1.67it/s]
 58%|█████▊    | 37808/65536 [6:29:28<4:37:34,  1.66it/s]
 58%|█████▊    | 37809/65536 [6:29:28<4:32:12,  1.70it/s]
 58%|█████▊    | 37810/65536 [6:29:29<4:31:36,  1.70it/s]
 58%|█████▊    | 37811/65536 [6:29:29<4:40:41,  1.65it/s]
 58%|█████▊    | 37812/65536 [6:29:30<4:52:47,  1.58it/s]
 58%|█████▊    | 37813/65536 [6:29:31<4:44:30,  1.62it/s]
 58%|█████▊    | 37814/65536 [6:29:31<4:48:50,  1.60it/s]
 58%|█████▊    | 37815/65536 [6:29:32<4:43:12,  1.63it/s]
 58%|█████▊    | 37816/65536 [6:29:33<4:42:32,  1.64it/s]
 58%|█████▊    | 37817/65536 [6:29:33<4:44:14,  1.63it/s]
 58%|█████▊    | 37818/65536 [6:29:34<4:41:16,  1.64it/s]
 58%|█████▊    | 37819/65536 [6:29:34<4:36:55,  1.67it/s]
 58%|█████▊    | 37820/65536 [6:29:35<4:36:38,  1.67it/s]
                                                         
{'loss': 1.643, 'learning_rate': 4.817864576956043e-07, 'epoch': 2334.57}
+
 58%|█████▊    | 37820/65536 [6:29:35<4:36:38,  1.67it/s]
 58%|█████▊    | 37821/65536 [6:29:36<4:31:47,  1.70it/s]
 58%|█████▊    | 37822/65536 [6:29:36<4:38:17,  1.66it/s]
 58%|█████▊    | 37823/65536 [6:29:37<4:45:01,  1.62it/s]
 58%|█████▊    | 37824/65536 [6:29:37<4:38:57,  1.66it/s]
 58%|█████▊    | 37825/65536 [6:29:38<4:35:41,  1.68it/s]
 58%|█████▊    | 37826/65536 [6:29:39<4:35:32,  1.68it/s]
 58%|█████▊    | 37827/65536 [6:29:39<4:39:14,  1.65it/s]
 58%|█████▊    | 37828/65536 [6:29:40<4:48:02,  1.60it/s]
 58%|█████▊    | 37829/65536 [6:29:40<4:44:03,  1.63it/s]
 58%|█████▊    | 37830/65536 [6:29:41<4:43:49,  1.63it/s]
 58%|█████▊    | 37831/65536 [6:29:42<4:44:07,  1.63it/s]
 58%|█████▊    | 37832/65536 [6:29:42<4:45:33,  1.62it/s]
 58%|█████▊    | 37833/65536 [6:29:43<4:48:11,  1.60it/s]
 58%|█████▊    | 37834/65536 [6:29:44<4:42:52,  1.63it/s]
 58%|█████▊    | 37835/65536 [6:29:44<4:34:55,  1.68it/s]
 58%|█████▊    | 37836/65536 [6:29:45<4:30:52,  1.70it/s]
 58%|█████▊    | 37837/65536 [6:29:45<4:29:12,  1.71it/s]
 58%|█████▊    | 37838/65536 [6:29:46<4:40:28,  1.65it/s]
 58%|█████▊    | 37839/65536 [6:29:47<4:48:47,  1.60it/s]
 58%|█████▊    | 37840/65536 [6:29:47<4:43:03,  1.63it/s]
                                                         
{'loss': 1.617, 'learning_rate': 4.815109587363782e-07, 'epoch': 2335.8}
+
 58%|█████▊    | 37840/65536 [6:29:47<4:43:03,  1.63it/s]
 58%|█████▊    | 37841/65536 [6:29:48<4:48:23,  1.60it/s]
 58%|█████▊    | 37842/65536 [6:29:48<4:49:22,  1.60it/s]
 58%|█████▊    | 37843/65536 [6:29:49<4:46:53,  1.61it/s]
 58%|█████▊    | 37844/65536 [6:29:50<4:51:20,  1.58it/s]
 58%|█████▊    | 37845/65536 [6:29:50<4:43:45,  1.63it/s]
 58%|█████▊    | 37846/65536 [6:29:51<4:43:53,  1.63it/s]
 58%|█████▊    | 37847/65536 [6:29:51<4:45:08,  1.62it/s]
 58%|█████▊    | 37848/65536 [6:29:52<4:48:34,  1.60it/s]
 58%|█████▊    | 37849/65536 [6:29:53<4:41:05,  1.64it/s]
 58%|█████▊    | 37850/65536 [6:29:53<4:46:00,  1.61it/s]
 58%|█████▊    | 37851/65536 [6:29:54<4:50:08,  1.59it/s]
 58%|█████▊    | 37852/65536 [6:29:55<4:51:07,  1.58it/s]
 58%|█████▊    | 37853/65536 [6:29:55<4:51:03,  1.59it/s]
 58%|█████▊    | 37854/65536 [6:29:56<4:52:05,  1.58it/s]
 58%|█████▊    | 37855/65536 [6:29:57<4:51:43,  1.58it/s]
 58%|█████▊    | 37856/65536 [6:29:57<4:45:54,  1.61it/s]
 58%|█████▊    | 37857/65536 [6:29:58<4:41:37,  1.64it/s]
 58%|█████▊    | 37858/65536 [6:29:58<4:45:22,  1.62it/s]
 58%|█████▊    | 37859/65536 [6:29:59<4:40:39,  1.64it/s]
 58%|█████▊    | 37860/65536 [6:30:00<4:48:44,  1.60it/s]
                                                         
{'loss': 1.6112, 'learning_rate': 4.81235459777152e-07, 'epoch': 2337.04}
+
 58%|█████▊    | 37860/65536 [6:30:00<4:48:44,  1.60it/s]
 58%|█████▊    | 37861/65536 [6:30:00<4:44:20,  1.62it/s]
 58%|█████▊    | 37862/65536 [6:30:01<4:46:13,  1.61it/s]
 58%|█████▊    | 37863/65536 [6:30:01<4:44:58,  1.62it/s]
 58%|█████▊    | 37864/65536 [6:30:02<4:44:52,  1.62it/s]
 58%|█████▊    | 37865/65536 [6:30:03<4:51:01,  1.58it/s]
 58%|█████▊    | 37866/65536 [6:30:03<4:45:39,  1.61it/s]
 58%|█████▊    | 37867/65536 [6:30:04<4:39:17,  1.65it/s]
 58%|█████▊    | 37868/65536 [6:30:05<4:45:56,  1.61it/s]
 58%|█████▊    | 37869/65536 [6:30:05<4:45:05,  1.62it/s]
 58%|█████▊    | 37870/65536 [6:30:06<4:49:59,  1.59it/s]
 58%|█████▊    | 37871/65536 [6:30:06<4:51:24,  1.58it/s]
 58%|█████▊    | 37872/65536 [6:30:07<4:51:38,  1.58it/s]
 58%|█████▊    | 37873/65536 [6:30:08<4:45:54,  1.61it/s]
 58%|█████▊    | 37874/65536 [6:30:08<4:42:23,  1.63it/s]
 58%|█████▊    | 37875/65536 [6:30:09<4:31:41,  1.70it/s]
 58%|█████▊    | 37876/65536 [6:30:09<4:41:21,  1.64it/s]
 58%|█████▊    | 37877/65536 [6:30:10<4:46:00,  1.61it/s]
 58%|█████▊    | 37878/65536 [6:30:11<4:45:27,  1.61it/s]
 58%|█████▊    | 37879/65536 [6:30:11<4:47:04,  1.61it/s]
 58%|█████▊    | 37880/65536 [6:30:12<4:39:32,  1.65it/s]
                                                         
{'loss': 1.6165, 'learning_rate': 4.809599608179257e-07, 'epoch': 2338.27}
+
 58%|█████▊    | 37880/65536 [6:30:12<4:39:32,  1.65it/s]
 58%|█████▊    | 37881/65536 [6:30:13<4:36:56,  1.66it/s]
 58%|█████▊    | 37882/65536 [6:30:13<4:37:03,  1.66it/s]
 58%|█████▊    | 37883/65536 [6:30:14<4:38:58,  1.65it/s]
 58%|█████▊    | 37884/65536 [6:30:14<4:34:58,  1.68it/s]
 58%|█████▊    | 37885/65536 [6:30:15<4:38:04,  1.66it/s]
 58%|█████▊    | 37886/65536 [6:30:16<4:38:41,  1.65it/s]
 58%|█████▊    | 37887/65536 [6:30:16<4:36:38,  1.67it/s]
 58%|█████▊    | 37888/65536 [6:30:17<4:32:33,  1.69it/s]
 58%|█████▊    | 37889/65536 [6:30:17<4:31:13,  1.70it/s]
 58%|█████▊    | 37890/65536 [6:30:18<4:34:00,  1.68it/s]
 58%|█████▊    | 37891/65536 [6:30:18<4:36:55,  1.66it/s]
 58%|█████▊    | 37892/65536 [6:30:19<4:54:37,  1.56it/s]
 58%|█████▊    | 37893/65536 [6:30:20<4:57:56,  1.55it/s]
 58%|█████▊    | 37894/65536 [6:30:21<5:00:21,  1.53it/s]
 58%|█████▊    | 37895/65536 [6:30:21<4:52:55,  1.57it/s]
 58%|█████▊    | 37896/65536 [6:30:22<4:42:24,  1.63it/s]
 58%|█████▊    | 37897/65536 [6:30:22<4:41:24,  1.64it/s]
 58%|█████▊    | 37898/65536 [6:30:23<4:40:24,  1.64it/s]
 58%|█████▊    | 37899/65536 [6:30:24<4:47:42,  1.60it/s]
 58%|█████▊    | 37900/65536 [6:30:24<4:39:18,  1.65it/s]
                                                         
{'loss': 1.633, 'learning_rate': 4.806844618586995e-07, 'epoch': 2339.51}
+
 58%|█████▊    | 37900/65536 [6:30:24<4:39:18,  1.65it/s]
 58%|█████▊    | 37901/65536 [6:30:25<4:38:24,  1.65it/s]
 58%|█████▊    | 37902/65536 [6:30:25<4:41:29,  1.64it/s]
 58%|█████▊    | 37903/65536 [6:30:26<4:39:48,  1.65it/s]
 58%|█████▊    | 37904/65536 [6:30:27<4:46:25,  1.61it/s]
 58%|█████▊    | 37905/65536 [6:30:27<4:45:04,  1.62it/s]
 58%|█████▊    | 37906/65536 [6:30:28<4:42:47,  1.63it/s]
 58%|█████▊    | 37907/65536 [6:30:28<4:36:43,  1.66it/s]
 58%|█████▊    | 37908/65536 [6:30:29<4:36:02,  1.67it/s]
 58%|█████▊    | 37909/65536 [6:30:30<4:49:54,  1.59it/s]
 58%|█████▊    | 37910/65536 [6:30:30<4:46:52,  1.60it/s]
 58%|█████▊    | 37911/65536 [6:30:31<4:46:07,  1.61it/s]
 58%|█████▊    | 37912/65536 [6:30:32<4:58:39,  1.54it/s]
 58%|█████▊    | 37913/65536 [6:30:32<4:55:26,  1.56it/s]
 58%|█████▊    | 37914/65536 [6:30:33<4:51:39,  1.58it/s]
 58%|█████▊    | 37915/65536 [6:30:33<4:48:36,  1.60it/s]
 58%|█████▊    | 37916/65536 [6:30:34<4:39:41,  1.65it/s]
 58%|█████▊    | 37917/65536 [6:30:35<4:37:57,  1.66it/s]
 58%|█████▊    | 37918/65536 [6:30:35<4:38:10,  1.65it/s]
 58%|█████▊    | 37919/65536 [6:30:36<4:35:10,  1.67it/s]
 58%|█████▊    | 37920/65536 [6:30:36<4:31:43,  1.69it/s]
                                                         
{'loss': 1.6311, 'learning_rate': 4.804089628994734e-07, 'epoch': 2340.74}
+
 58%|█████▊    | 37920/65536 [6:30:36<4:31:43,  1.69it/s]
 58%|█████▊    | 37921/65536 [6:30:37<4:33:11,  1.68it/s]
 58%|█████▊    | 37922/65536 [6:30:38<4:34:38,  1.68it/s]
 58%|█████▊    | 37923/65536 [6:30:38<4:35:57,  1.67it/s]
 58%|█████▊    | 37924/65536 [6:30:39<4:45:53,  1.61it/s]
 58%|█████▊    | 37925/65536 [6:30:40<4:51:10,  1.58it/s]
 58%|█████▊    | 37926/65536 [6:30:40<4:50:15,  1.59it/s]
 58%|█████▊    | 37927/65536 [6:30:41<4:50:54,  1.58it/s]
 58%|█████▊    | 37928/65536 [6:30:41<4:44:15,  1.62it/s]
 58%|█████▊    | 37929/65536 [6:30:42<4:37:37,  1.66it/s]
 58%|█████▊    | 37930/65536 [6:30:43<4:38:42,  1.65it/s]
 58%|█████▊    | 37931/65536 [6:30:43<4:31:25,  1.70it/s]
 58%|█████▊    | 37932/65536 [6:30:44<4:37:45,  1.66it/s]
 58%|█████▊    | 37933/65536 [6:30:44<4:39:18,  1.65it/s]
 58%|█████▊    | 37934/65536 [6:30:45<4:37:27,  1.66it/s]
 58%|█████▊    | 37935/65536 [6:30:46<4:47:47,  1.60it/s]
 58%|█████▊    | 37936/65536 [6:30:46<4:45:30,  1.61it/s]
 58%|█████▊    | 37937/65536 [6:30:47<4:46:30,  1.61it/s]
 58%|█████▊    | 37938/65536 [6:30:48<4:46:08,  1.61it/s]
 58%|█████▊    | 37939/65536 [6:30:48<4:43:23,  1.62it/s]
 58%|█████▊    | 37940/65536 [6:30:49<4:41:13,  1.64it/s]
                                                         
{'loss': 1.5876, 'learning_rate': 4.801334639402472e-07, 'epoch': 2341.98}
+
 58%|█████▊    | 37940/65536 [6:30:49<4:41:13,  1.64it/s]
 58%|█████▊    | 37941/65536 [6:30:49<4:49:48,  1.59it/s]
 58%|█████▊    | 37942/65536 [6:30:50<4:51:11,  1.58it/s]
 58%|█████▊    | 37943/65536 [6:30:51<4:45:51,  1.61it/s]
 58%|█████▊    | 37944/65536 [6:30:51<4:53:22,  1.57it/s]
 58%|█████▊    | 37945/65536 [6:30:52<4:50:29,  1.58it/s]
 58%|█████▊    | 37946/65536 [6:30:53<4:46:55,  1.60it/s]
 58%|█████▊    | 37947/65536 [6:30:53<4:44:15,  1.62it/s]
 58%|█████▊    | 37948/65536 [6:30:54<4:42:14,  1.63it/s]
 58%|█████▊    | 37949/65536 [6:30:54<4:45:50,  1.61it/s]
 58%|█████▊    | 37950/65536 [6:30:55<4:46:34,  1.60it/s]
 58%|█████▊    | 37951/65536 [6:30:56<4:43:31,  1.62it/s]
 58%|█████▊    | 37952/65536 [6:30:56<4:42:46,  1.63it/s]
 58%|█████▊    | 37953/65536 [6:30:57<4:41:23,  1.63it/s]
 58%|█████▊    | 37954/65536 [6:30:57<4:40:00,  1.64it/s]
 58%|█████▊    | 37955/65536 [6:30:58<4:36:39,  1.66it/s]
 58%|█████▊    | 37956/65536 [6:30:59<4:31:25,  1.69it/s]
 58%|█████▊    | 37957/65536 [6:30:59<4:44:48,  1.61it/s]
 58%|█████▊    | 37958/65536 [6:31:00<4:40:42,  1.64it/s]
 58%|█████▊    | 37959/65536 [6:31:01<4:47:19,  1.60it/s]
 58%|█████▊    | 37960/65536 [6:31:01<4:44:26,  1.62it/s]
                                                         
{'loss': 1.6237, 'learning_rate': 4.798579649810211e-07, 'epoch': 2343.21}
+
 58%|█████▊    | 37960/65536 [6:31:01<4:44:26,  1.62it/s]
 58%|█████▊    | 37961/65536 [6:31:02<4:38:01,  1.65it/s]
 58%|█████▊    | 37962/65536 [6:31:02<4:34:20,  1.68it/s]
 58%|█████▊    | 37963/65536 [6:31:03<4:40:37,  1.64it/s]
 58%|█████▊    | 37964/65536 [6:31:04<4:50:02,  1.58it/s]
 58%|█████▊    | 37965/65536 [6:31:04<4:46:24,  1.60it/s]
 58%|█████▊    | 37966/65536 [6:31:05<4:44:07,  1.62it/s]
 58%|█████▊    | 37967/65536 [6:31:05<4:45:10,  1.61it/s]
 58%|█████▊    | 37968/65536 [6:31:06<4:43:36,  1.62it/s]
 58%|█████▊    | 37969/65536 [6:31:07<4:46:36,  1.60it/s]
 58%|█████▊    | 37970/65536 [6:31:07<4:41:35,  1.63it/s]
 58%|█████▊    | 37971/65536 [6:31:08<4:43:32,  1.62it/s]
 58%|█████▊    | 37972/65536 [6:31:09<4:47:05,  1.60it/s]
 58%|█████▊    | 37973/65536 [6:31:09<4:55:19,  1.56it/s]
 58%|█████▊    | 37974/65536 [6:31:10<4:53:04,  1.57it/s]
 58%|█████▊    | 37975/65536 [6:31:10<4:44:57,  1.61it/s]
 58%|█████▊    | 37976/65536 [6:31:11<4:41:37,  1.63it/s]
 58%|█████▊    | 37977/65536 [6:31:12<4:38:51,  1.65it/s]
 58%|█████▊    | 37978/65536 [6:31:12<4:36:10,  1.66it/s]
 58%|█████▊    | 37979/65536 [6:31:13<4:40:44,  1.64it/s]
 58%|█████▊    | 37980/65536 [6:31:13<4:44:23,  1.61it/s]
                                                         
{'loss': 1.6244, 'learning_rate': 4.795824660217951e-07, 'epoch': 2344.44}
+
 58%|█████▊    | 37980/65536 [6:31:13<4:44:23,  1.61it/s]
 58%|█████▊    | 37981/65536 [6:31:14<4:42:26,  1.63it/s]
 58%|█████▊    | 37982/65536 [6:31:15<4:40:47,  1.64it/s]
 58%|█████▊    | 37983/65536 [6:31:15<4:38:58,  1.65it/s]
 58%|█████▊    | 37984/65536 [6:31:16<4:42:58,  1.62it/s]
 58%|█████▊    | 37985/65536 [6:31:17<4:41:27,  1.63it/s]
 58%|█████▊    | 37986/65536 [6:31:17<4:46:01,  1.61it/s]
 58%|█████▊    | 37987/65536 [6:31:18<4:45:21,  1.61it/s]
 58%|█████▊    | 37988/65536 [6:31:18<4:44:01,  1.62it/s]
 58%|█████▊    | 37989/65536 [6:31:19<4:38:16,  1.65it/s]
 58%|█████▊    | 37990/65536 [6:31:20<4:54:29,  1.56it/s]
 58%|█████▊    | 37991/65536 [6:31:20<4:49:04,  1.59it/s]
 58%|█████▊    | 37992/65536 [6:31:21<4:45:09,  1.61it/s]
 58%|█████▊    | 37993/65536 [6:31:22<4:55:48,  1.55it/s]
 58%|█████▊    | 37994/65536 [6:31:22<4:48:28,  1.59it/s]
 58%|█████▊    | 37995/65536 [6:31:23<4:49:52,  1.58it/s]
 58%|█████▊    | 37996/65536 [6:31:23<4:44:30,  1.61it/s]
 58%|█████▊    | 37997/65536 [6:31:24<4:37:18,  1.66it/s]
 58%|█████▊    | 37998/65536 [6:31:25<4:40:59,  1.63it/s]
 58%|█████▊    | 37999/65536 [6:31:25<4:38:45,  1.65it/s]
 58%|█████▊    | 38000/65536 [6:31:26<4:34:43,  1.67it/s]
                                                         
{'loss': 1.586, 'learning_rate': 4.793069670625689e-07, 'epoch': 2345.68}
+
 58%|█████▊    | 38000/65536 [6:31:26<4:34:43,  1.67it/s]
 58%|█████▊    | 38001/65536 [6:31:26<4:35:21,  1.67it/s]
 58%|█████▊    | 38002/65536 [6:31:27<4:32:15,  1.69it/s]
 58%|█████▊    | 38003/65536 [6:31:28<4:32:18,  1.69it/s]
 58%|█████▊    | 38004/65536 [6:31:28<4:33:39,  1.68it/s]
 58%|█████▊    | 38005/65536 [6:31:29<4:33:53,  1.68it/s]
 58%|█████▊    | 38006/65536 [6:31:29<4:45:42,  1.61it/s]
 58%|█████▊    | 38007/65536 [6:31:30<4:37:48,  1.65it/s]
 58%|█████▊    | 38008/65536 [6:31:31<4:34:18,  1.67it/s]
 58%|█████▊    | 38009/65536 [6:31:31<4:30:13,  1.70it/s]
 58%|█████▊    | 38010/65536 [6:31:32<4:38:48,  1.65it/s]
 58%|█████▊    | 38011/65536 [6:31:32<4:47:36,  1.60it/s]
 58%|█████▊    | 38012/65536 [6:31:33<4:43:05,  1.62it/s]
 58%|█████▊    | 38013/65536 [6:31:34<4:35:24,  1.67it/s]
 58%|█████▊    | 38014/65536 [6:31:34<4:34:22,  1.67it/s]
 58%|█████▊    | 38015/65536 [6:31:35<4:38:50,  1.64it/s]
 58%|█████▊    | 38016/65536 [6:31:36<4:52:37,  1.57it/s]
 58%|█████▊    | 38017/65536 [6:31:36<5:01:43,  1.52it/s]
 58%|█████▊    | 38018/65536 [6:31:37<4:57:04,  1.54it/s]
 58%|█████▊    | 38019/65536 [6:31:38<4:55:18,  1.55it/s]
 58%|█████▊    | 38020/65536 [6:31:38<4:55:05,  1.55it/s]
                                                         
{'loss': 1.6345, 'learning_rate': 4.790314681033427e-07, 'epoch': 2346.91}
+
 58%|█████▊    | 38020/65536 [6:31:38<4:55:05,  1.55it/s]
 58%|█████▊    | 38021/65536 [6:31:39<4:50:03,  1.58it/s]
 58%|█████▊    | 38022/65536 [6:31:40<5:02:10,  1.52it/s]
 58%|█████▊    | 38023/65536 [6:31:40<4:52:15,  1.57it/s]
 58%|█████▊    | 38024/65536 [6:31:41<4:47:05,  1.60it/s]
 58%|█████▊    | 38025/65536 [6:31:41<4:55:05,  1.55it/s]
 58%|█████▊    | 38026/65536 [6:31:42<4:56:40,  1.55it/s]
 58%|█████▊    | 38027/65536 [6:31:43<4:43:04,  1.62it/s]
 58%|█████▊    | 38028/65536 [6:31:43<4:45:50,  1.60it/s]
 58%|█████▊    | 38029/65536 [6:31:44<4:41:58,  1.63it/s]
 58%|█████▊    | 38030/65536 [6:31:44<4:42:25,  1.62it/s]
 58%|█████▊    | 38031/65536 [6:31:45<4:41:58,  1.63it/s]
 58%|█████▊    | 38032/65536 [6:31:46<4:41:33,  1.63it/s]
 58%|█████▊    | 38033/65536 [6:31:46<4:34:44,  1.67it/s]
 58%|█████▊    | 38034/65536 [6:31:47<4:37:23,  1.65it/s]
 58%|█████▊    | 38035/65536 [6:31:47<4:42:30,  1.62it/s]
 58%|█████▊    | 38036/65536 [6:31:48<4:44:08,  1.61it/s]
 58%|█████▊    | 38037/65536 [6:31:49<4:37:43,  1.65it/s]
 58%|█████▊    | 38038/65536 [6:31:49<4:42:38,  1.62it/s]
 58%|█████▊    | 38039/65536 [6:31:50<4:38:09,  1.65it/s]
 58%|█████▊    | 38040/65536 [6:31:51<4:46:59,  1.60it/s]
                                                         
{'loss': 1.6393, 'learning_rate': 4.787559691441165e-07, 'epoch': 2348.15}
+
 58%|█████▊    | 38040/65536 [6:31:51<4:46:59,  1.60it/s]
 58%|█████▊    | 38041/65536 [6:31:51<4:42:53,  1.62it/s]
 58%|█████▊    | 38042/65536 [6:31:52<4:45:45,  1.60it/s]
 58%|█████▊    | 38043/65536 [6:31:52<4:44:50,  1.61it/s]
 58%|█████▊    | 38044/65536 [6:31:53<4:40:50,  1.63it/s]
 58%|█████▊    | 38045/65536 [6:31:54<4:42:30,  1.62it/s]
 58%|█████▊    | 38046/65536 [6:31:54<4:40:37,  1.63it/s]
 58%|█████▊    | 38047/65536 [6:31:55<4:44:38,  1.61it/s]
 58%|█████▊    | 38048/65536 [6:31:55<4:40:55,  1.63it/s]
 58%|█████▊    | 38049/65536 [6:31:56<4:38:33,  1.64it/s]
 58%|█████▊    | 38050/65536 [6:31:57<4:35:58,  1.66it/s]
 58%|█████▊    | 38051/65536 [6:31:57<4:35:54,  1.66it/s]
 58%|█████▊    | 38052/65536 [6:31:58<4:35:20,  1.66it/s]
 58%|█████▊    | 38053/65536 [6:31:58<4:36:04,  1.66it/s]
 58%|█████▊    | 38054/65536 [6:31:59<4:43:47,  1.61it/s]
 58%|█████▊    | 38055/65536 [6:32:00<4:43:09,  1.62it/s]
 58%|█████▊    | 38056/65536 [6:32:00<4:43:46,  1.61it/s]
 58%|█████▊    | 38057/65536 [6:32:01<4:42:38,  1.62it/s]
 58%|█████▊    | 38058/65536 [6:32:02<4:37:28,  1.65it/s]
 58%|█████▊    | 38059/65536 [6:32:02<4:42:54,  1.62it/s]
 58%|█████▊    | 38060/65536 [6:32:03<4:41:58,  1.62it/s]
                                                         
{'loss': 1.6173, 'learning_rate': 4.784804701848904e-07, 'epoch': 2349.38}
+
 58%|█████▊    | 38060/65536 [6:32:03<4:41:58,  1.62it/s]
 58%|█████▊    | 38061/65536 [6:32:03<4:44:51,  1.61it/s]
 58%|█████▊    | 38062/65536 [6:32:04<4:49:16,  1.58it/s]
 58%|█████▊    | 38063/65536 [6:32:05<4:42:32,  1.62it/s]
 58%|█████▊    | 38064/65536 [6:32:05<4:38:01,  1.65it/s]
 58%|█████▊    | 38065/65536 [6:32:06<4:48:58,  1.58it/s]
 58%|█████▊    | 38066/65536 [6:32:07<4:50:18,  1.58it/s]
 58%|███���█▊    | 38067/65536 [6:32:07<4:44:54,  1.61it/s]
 58%|█████▊    | 38068/65536 [6:32:08<4:39:53,  1.64it/s]
 58%|█████▊    | 38069/65536 [6:32:08<4:32:20,  1.68it/s]
 58%|█████▊    | 38070/65536 [6:32:09<4:34:35,  1.67it/s]
 58%|█████▊    | 38071/65536 [6:32:10<4:42:29,  1.62it/s]
 58%|█████▊    | 38072/65536 [6:32:10<4:42:04,  1.62it/s]
 58%|█████▊    | 38073/65536 [6:32:11<4:40:32,  1.63it/s]
 58%|█████▊    | 38074/65536 [6:32:11<4:42:37,  1.62it/s]
 58%|█████▊    | 38075/65536 [6:32:12<4:45:23,  1.60it/s]
 58%|█████▊    | 38076/65536 [6:32:13<4:44:30,  1.61it/s]
 58%|█████▊    | 38077/65536 [6:32:13<4:45:29,  1.60it/s]
 58%|█████▊    | 38078/65536 [6:32:14<4:36:45,  1.65it/s]
 58%|█████▊    | 38079/65536 [6:32:15<4:45:57,  1.60it/s]
 58%|█████▊    | 38080/65536 [6:32:15<4:42:35,  1.62it/s]
                                                         
{'loss': 1.6243, 'learning_rate': 4.782049712256643e-07, 'epoch': 2350.62}
+
 58%|█████▊    | 38080/65536 [6:32:15<4:42:35,  1.62it/s]
 58%|█████▊    | 38081/65536 [6:32:16<4:34:51,  1.66it/s]
 58%|█████▊    | 38082/65536 [6:32:16<4:31:06,  1.69it/s]
 58%|█████▊    | 38083/65536 [6:32:17<4:29:59,  1.69it/s]
 58%|█████▊    | 38084/65536 [6:32:18<4:37:45,  1.65it/s]
 58%|█████▊    | 38085/65536 [6:32:18<4:41:33,  1.62it/s]
 58%|█████▊    | 38086/65536 [6:32:19<4:36:13,  1.66it/s]
 58%|█████▊    | 38087/65536 [6:32:19<4:46:43,  1.60it/s]
 58%|█████▊    | 38088/65536 [6:32:20<4:59:17,  1.53it/s]
 58%|█████▊    | 38089/65536 [6:32:21<4:44:26,  1.61it/s]
 58%|█████▊    | 38090/65536 [6:32:21<4:40:03,  1.63it/s]
 58%|█████▊    | 38091/65536 [6:32:22<4:40:48,  1.63it/s]
 58%|█████▊    | 38092/65536 [6:32:22<4:35:05,  1.66it/s]
 58%|█████▊    | 38093/65536 [6:32:23<4:38:21,  1.64it/s]
 58%|█████▊    | 38094/65536 [6:32:24<4:35:18,  1.66it/s]
 58%|█████▊    | 38095/65536 [6:32:24<4:27:04,  1.71it/s]
 58%|█████▊    | 38096/65536 [6:32:25<4:31:49,  1.68it/s]
 58%|█████▊    | 38097/65536 [6:32:25<4:35:04,  1.66it/s]
 58%|█████▊    | 38098/65536 [6:32:26<4:33:40,  1.67it/s]
 58%|█████▊    | 38099/65536 [6:32:27<4:38:42,  1.64it/s]
 58%|█████▊    | 38100/65536 [6:32:27<4:35:08,  1.66it/s]
                                                         
{'loss': 1.6464, 'learning_rate': 4.779294722664381e-07, 'epoch': 2351.85}
+
 58%|█████▊    | 38100/65536 [6:32:27<4:35:08,  1.66it/s]
 58%|█████▊    | 38101/65536 [6:32:28<4:36:53,  1.65it/s]
 58%|█████▊    | 38102/65536 [6:32:29<4:38:52,  1.64it/s]
 58%|█████▊    | 38103/65536 [6:32:29<4:56:23,  1.54it/s]
 58%|█████▊    | 38104/65536 [6:32:30<4:49:28,  1.58it/s]
 58%|█████▊    | 38105/65536 [6:32:31<4:51:41,  1.57it/s]
 58%|█████▊    | 38106/65536 [6:32:31<4:45:12,  1.60it/s]
 58%|█████▊    | 38107/65536 [6:32:32<4:44:33,  1.61it/s]
 58%|█████▊    | 38108/65536 [6:32:32<4:41:48,  1.62it/s]
 58%|█████▊    | 38109/65536 [6:32:33<4:39:22,  1.64it/s]
 58%|█████▊    | 38110/65536 [6:32:33<4:34:04,  1.67it/s]
 58%|█████▊    | 38111/65536 [6:32:34<4:37:07,  1.65it/s]
 58%|█████▊    | 38112/65536 [6:32:35<4:36:00,  1.66it/s]
 58%|█████▊    | 38113/65536 [6:32:35<4:41:47,  1.62it/s]
 58%|█████▊    | 38114/65536 [6:32:36<4:38:57,  1.64it/s]
 58%|█████▊    | 38115/65536 [6:32:37<4:34:10,  1.67it/s]
 58%|█████▊    | 38116/65536 [6:32:37<4:34:10,  1.67it/s]
 58%|█████▊    | 38117/65536 [6:32:38<4:33:43,  1.67it/s]
 58%|█████▊    | 38118/65536 [6:32:38<4:45:36,  1.60it/s]
 58%|█████▊    | 38119/65536 [6:32:39<4:52:57,  1.56it/s]
 58%|█████▊    | 38120/65536 [6:32:40<4:49:22,  1.58it/s]
                                                         
{'loss': 1.6287, 'learning_rate': 4.77653973307212e-07, 'epoch': 2353.09}
+
 58%|█████▊    | 38120/65536 [6:32:40<4:49:22,  1.58it/s]
 58%|█████▊    | 38121/65536 [6:32:40<4:49:23,  1.58it/s]
 58%|█████▊    | 38122/65536 [6:32:41<4:44:59,  1.60it/s]
 58%|█████▊    | 38123/65536 [6:32:42<4:37:22,  1.65it/s]
 58%|█████▊    | 38124/65536 [6:32:42<4:37:52,  1.64it/s]
 58%|█████▊    | 38125/65536 [6:32:43<4:33:30,  1.67it/s]
 58%|█████▊    | 38126/65536 [6:32:43<4:39:50,  1.63it/s]
 58%|█████▊    | 38127/65536 [6:32:44<4:42:57,  1.61it/s]
 58%|█████▊    | 38128/65536 [6:32:45<4:35:50,  1.66it/s]
 58%|█████▊    | 38129/65536 [6:32:45<4:36:33,  1.65it/s]
 58%|█████▊    | 38130/65536 [6:32:46<4:48:46,  1.58it/s]
 58%|█████▊    | 38131/65536 [6:32:46<4:39:44,  1.63it/s]
 58%|█████▊    | 38132/65536 [6:32:47<4:41:46,  1.62it/s]
 58%|█████▊    | 38133/65536 [6:32:48<4:40:38,  1.63it/s]
 58%|█████▊    | 38134/65536 [6:32:48<4:42:24,  1.62it/s]
 58%|█████▊    | 38135/65536 [6:32:49<4:55:23,  1.55it/s]
 58%|█████▊    | 38136/65536 [6:32:50<4:48:53,  1.58it/s]
 58%|█████▊    | 38137/65536 [6:32:50<4:48:32,  1.58it/s]
 58%|█████▊    | 38138/65536 [6:32:51<4:39:38,  1.63it/s]
 58%|█████▊    | 38139/65536 [6:32:51<4:33:42,  1.67it/s]
 58%|█████▊    | 38140/65536 [6:32:52<4:42:04,  1.62it/s]
                                                         
{'loss': 1.6247, 'learning_rate': 4.773784743479857e-07, 'epoch': 2354.32}
+
 58%|█████▊    | 38140/65536 [6:32:52<4:42:04,  1.62it/s]
 58%|█████▊    | 38141/65536 [6:32:53<4:36:12,  1.65it/s]
 58%|█████▊    | 38142/65536 [6:32:53<4:32:52,  1.67it/s]
 58%|█████▊    | 38143/65536 [6:32:54<4:31:12,  1.68it/s]
 58%|█████▊    | 38144/65536 [6:32:54<4:33:36,  1.67it/s]
 58%|█████▊    | 38145/65536 [6:32:55<4:35:26,  1.66it/s]
 58%|█████▊    | 38146/65536 [6:32:56<4:46:51,  1.59it/s]
 58%|█████▊    | 38147/65536 [6:32:56<4:48:20,  1.58it/s]
 58%|█████▊    | 38148/65536 [6:32:57<4:38:38,  1.64it/s]
 58%|█████▊    | 38149/65536 [6:32:57<4:38:55,  1.64it/s]
 58%|█████▊    | 38150/65536 [6:32:58<4:38:52,  1.64it/s]
 58%|█████▊    | 38151/65536 [6:32:59<4:34:24,  1.66it/s]
 58%|█████▊    | 38152/65536 [6:32:59<4:43:16,  1.61it/s]
 58%|█████▊    | 38153/65536 [6:33:00<4:42:11,  1.62it/s]
 58%|█████▊    | 38154/65536 [6:33:01<4:34:10,  1.66it/s]
 58%|█████▊    | 38155/65536 [6:33:01<4:29:37,  1.69it/s]
 58%|█████▊    | 38156/65536 [6:33:02<4:42:43,  1.61it/s]
 58%|█████▊    | 38157/65536 [6:33:02<4:40:04,  1.63it/s]
 58%|█████▊    | 38158/65536 [6:33:03<4:47:18,  1.59it/s]
 58%|█████▊    | 38159/65536 [6:33:04<4:45:53,  1.60it/s]
 58%|█████▊    | 38160/65536 [6:33:04<4:42:07,  1.62it/s]
                                                         
{'loss': 1.6204, 'learning_rate': 4.771029753887596e-07, 'epoch': 2355.56}
+
 58%|█████▊    | 38160/65536 [6:33:04<4:42:07,  1.62it/s]
 58%|█████▊    | 38161/65536 [6:33:05<4:36:33,  1.65it/s]
 58%|█████▊    | 38162/65536 [6:33:05<4:40:14,  1.63it/s]
 58%|█████▊    | 38163/65536 [6:33:06<4:36:26,  1.65it/s]
 58%|█████▊    | 38164/65536 [6:33:07<4:34:37,  1.66it/s]
 58%|█████▊    | 38165/65536 [6:33:07<4:37:42,  1.64it/s]
 58%|█████▊    | 38166/65536 [6:33:08<4:35:48,  1.65it/s]
 58%|█████▊    | 38167/65536 [6:33:09<4:43:56,  1.61it/s]
 58%|█████▊    | 38168/65536 [6:33:09<5:06:34,  1.49it/s]
 58%|█████▊    | 38169/65536 [6:33:10<4:56:27,  1.54it/s]
 58%|█████▊    | 38170/65536 [6:33:11<4:55:00,  1.55it/s]
 58%|█████▊    | 38171/65536 [6:33:11<4:48:53,  1.58it/s]
 58%|█████▊    | 38172/65536 [6:33:12<4:45:46,  1.60it/s]
 58%|█████▊    | 38173/65536 [6:33:12<4:45:02,  1.60it/s]
 58%|█████▊    | 38174/65536 [6:33:13<4:38:25,  1.64it/s]
 58%|█████▊    | 38175/65536 [6:33:14<4:40:07,  1.63it/s]
 58%|█████▊    | 38176/65536 [6:33:14<4:43:19,  1.61it/s]
 58%|█████▊    | 38177/65536 [6:33:15<4:37:06,  1.65it/s]
 58%|█████▊    | 38178/65536 [6:33:15<4:36:51,  1.65it/s]
 58%|█████▊    | 38179/65536 [6:33:16<4:38:43,  1.64it/s]
 58%|█████▊    | 38180/65536 [6:33:17<4:40:28,  1.63it/s]
                                                         
{'loss': 1.6291, 'learning_rate': 4.768274764295335e-07, 'epoch': 2356.79}
+
 58%|█████▊    | 38180/65536 [6:33:17<4:40:28,  1.63it/s]
 58%|█████▊    | 38181/65536 [6:33:17<4:39:08,  1.63it/s]
 58%|█████▊    | 38182/65536 [6:33:18<4:39:51,  1.63it/s]
 58%|█████▊    | 38183/65536 [6:33:18<4:29:52,  1.69it/s]
 58%|█████▊    | 38184/65536 [6:33:19<4:44:59,  1.60it/s]
 58%|█████▊    | 38185/65536 [6:33:20<4:48:20,  1.58it/s]
 58%|█████▊    | 38186/65536 [6:33:20<4:40:55,  1.62it/s]
 58%|█████▊    | 38187/65536 [6:33:21<4:38:32,  1.64it/s]
 58%|█████▊    | 38188/65536 [6:33:22<4:41:52,  1.62it/s]
 58%|█████▊    | 38189/65536 [6:33:22<4:44:05,  1.60it/s]
 58%|█████▊    | 38190/65536 [6:33:23<4:45:48,  1.59it/s]
 58%|█████▊    | 38191/65536 [6:33:23<4:40:46,  1.62it/s]
 58%|█████▊    | 38192/65536 [6:33:24<4:35:09,  1.66it/s]
 58%|█████▊    | 38193/65536 [6:33:25<4:42:34,  1.61it/s]
 58%|█████▊    | 38194/65536 [6:33:25<4:41:22,  1.62it/s]
 58%|█████▊    | 38195/65536 [6:33:26<4:36:41,  1.65it/s]
 58%|█████▊    | 38196/65536 [6:33:26<4:37:24,  1.64it/s]
 58%|█████▊    | 38197/65536 [6:33:27<4:38:16,  1.64it/s]
 58%|█████▊    | 38198/65536 [6:33:28<4:36:27,  1.65it/s]
 58%|█████▊    | 38199/65536 [6:33:28<4:37:21,  1.64it/s]
 58%|█████▊    | 38200/65536 [6:33:29<4:39:13,  1.63it/s]
                                                         
{'loss': 1.6355, 'learning_rate': 4.765519774703073e-07, 'epoch': 2358.02}
+
 58%|█████▊    | 38200/65536 [6:33:29<4:39:13,  1.63it/s]
 58%|█████▊    | 38201/65536 [6:33:30<4:39:51,  1.63it/s]
 58%|█████▊    | 38202/65536 [6:33:30<4:33:33,  1.67it/s]
 58%|█████▊    | 38203/65536 [6:33:31<4:29:30,  1.69it/s]
 58%|█████▊    | 38204/65536 [6:33:31<4:28:36,  1.70it/s]
 58%|█████▊    | 38205/65536 [6:33:32<4:32:28,  1.67it/s]
 58%|█████▊    | 38206/65536 [6:33:33<4:35:23,  1.65it/s]
 58%|█████▊    | 38207/65536 [6:33:33<4:35:21,  1.65it/s]
 58%|█████▊    | 38208/65536 [6:33:34<4:32:55,  1.67it/s]
 58%|█████▊    | 38209/65536 [6:33:34<4:32:21,  1.67it/s]
 58%|█████▊    | 38210/65536 [6:33:35<4:37:48,  1.64it/s]
 58%|█████▊    | 38211/65536 [6:33:36<4:36:10,  1.65it/s]
 58%|█████▊    | 38212/65536 [6:33:36<4:43:08,  1.61it/s]
 58%|█████▊    | 38213/65536 [6:33:37<4:41:43,  1.62it/s]
 58%|█████▊    | 38214/65536 [6:33:37<4:36:29,  1.65it/s]
 58%|█████▊    | 38215/65536 [6:33:38<4:32:44,  1.67it/s]
 58%|█████▊    | 38216/65536 [6:33:39<4:49:03,  1.58it/s]
 58%|█████▊    | 38217/65536 [6:33:39<4:44:41,  1.60it/s]
 58%|█████▊    | 38218/65536 [6:33:40<4:40:46,  1.62it/s]
 58%|█████▊    | 38219/65536 [6:33:41<4:44:11,  1.60it/s]
 58%|█████▊    | 38220/65536 [6:33:41<4:45:38,  1.59it/s]
                                                         
{'loss': 1.6379, 'learning_rate': 4.7627647851108116e-07, 'epoch': 2359.26}
+
 58%|█████▊    | 38220/65536 [6:33:41<4:45:38,  1.59it/s]
 58%|█████▊    | 38221/65536 [6:33:42<4:37:55,  1.64it/s]
 58%|█████▊    | 38222/65536 [6:33:42<4:41:42,  1.62it/s]
 58%|█████▊    | 38223/65536 [6:33:43<4:35:25,  1.65it/s]
 58%|█████▊    | 38224/65536 [6:33:44<4:39:44,  1.63it/s]
 58%|█████▊    | 38225/65536 [6:33:44<4:33:35,  1.66it/s]
 58%|█████▊    | 38226/65536 [6:33:45<4:39:44,  1.63it/s]
 58%|█████▊    | 38227/65536 [6:33:45<4:40:33,  1.62it/s]
 58%|█████▊    | 38228/65536 [6:33:46<4:45:24,  1.59it/s]
 58%|█████▊    | 38229/65536 [6:33:47<4:40:24,  1.62it/s]
 58%|█████▊    | 38230/65536 [6:33:47<4:36:34,  1.65it/s]
 58%|█████▊    | 38231/65536 [6:33:48<4:40:11,  1.62it/s]
 58%|█████▊    | 38232/65536 [6:33:49<4:42:00,  1.61it/s]
 58%|█████▊    | 38233/65536 [6:33:49<4:47:15,  1.58it/s]
 58%|█████▊    | 38234/65536 [6:33:50<4:37:50,  1.64it/s]
 58%|█████▊    | 38235/65536 [6:33:50<4:43:33,  1.60it/s]
 58%|█████▊    | 38236/65536 [6:33:51<4:41:27,  1.62it/s]
 58%|█████▊    | 38237/65536 [6:33:52<4:39:40,  1.63it/s]
 58%|█████▊    | 38238/65536 [6:33:52<4:31:30,  1.68it/s]
 58%|█████▊    | 38239/65536 [6:33:53<4:30:07,  1.68it/s]
 58%|█████▊    | 38240/65536 [6:33:53<4:30:06,  1.68it/s]
                                                         
{'loss': 1.6773, 'learning_rate': 4.76000979551855e-07, 'epoch': 2360.49}
+
 58%|█████▊    | 38240/65536 [6:33:53<4:30:06,  1.68it/s]
 58%|█████▊    | 38241/65536 [6:33:54<4:32:36,  1.67it/s]
 58%|█████▊    | 38242/65536 [6:33:55<4:35:43,  1.65it/s]
 58%|█████▊    | 38243/65536 [6:33:55<4:41:27,  1.62it/s]
 58%|█████▊    | 38244/65536 [6:33:56<4:39:27,  1.63it/s]
 58%|█████▊    | 38245/65536 [6:33:56<4:49:26,  1.57it/s]
 58%|█████▊    | 38246/65536 [6:33:57<4:50:33,  1.57it/s]
 58%|█████▊    | 38247/65536 [6:33:58<4:41:09,  1.62it/s]
 58%|█████▊    | 38248/65536 [6:33:58<4:39:03,  1.63it/s]
 58%|█████▊    | 38249/65536 [6:33:59<4:50:16,  1.57it/s]
 58%|█████▊    | 38250/65536 [6:34:00<4:47:19,  1.58it/s]
 58%|█████▊    | 38251/65536 [6:34:00<4:40:44,  1.62it/s]
 58%|█████▊    | 38252/65536 [6:34:01<4:39:08,  1.63it/s]
 58%|█████▊    | 38253/65536 [6:34:01<4:36:25,  1.65it/s]
 58%|█████▊    | 38254/65536 [6:34:02<4:31:50,  1.67it/s]
 58%|█████▊    | 38255/65536 [6:34:03<4:37:54,  1.64it/s]
 58%|█████▊    | 38256/65536 [6:34:03<4:34:28,  1.66it/s]
 58%|█████▊    | 38257/65536 [6:34:04<4:29:24,  1.69it/s]
 58%|█████▊    | 38258/65536 [6:34:04<4:35:49,  1.65it/s]
 58%|█████▊    | 38259/65536 [6:34:05<4:37:06,  1.64it/s]
 58%|█████���    | 38260/65536 [6:34:06<4:38:00,  1.64it/s]
                                                         
{'loss': 1.5918, 'learning_rate': 4.757254805926289e-07, 'epoch': 2361.73}
+
 58%|█████▊    | 38260/65536 [6:34:06<4:38:00,  1.64it/s]
 58%|█████▊    | 38261/65536 [6:34:06<4:48:28,  1.58it/s]
 58%|█████▊    | 38262/65536 [6:34:07<4:43:10,  1.61it/s]
 58%|█████▊    | 38263/65536 [6:34:08<4:36:38,  1.64it/s]
 58%|█████▊    | 38264/65536 [6:34:08<4:35:35,  1.65it/s]
 58%|█████▊    | 38265/65536 [6:34:09<4:47:33,  1.58it/s]
 58%|█████▊    | 38266/65536 [6:34:09<4:45:59,  1.59it/s]
 58%|█████▊    | 38267/65536 [6:34:10<4:44:41,  1.60it/s]
 58%|█████▊    | 38268/65536 [6:34:11<4:39:02,  1.63it/s]
 58%|█████▊    | 38269/65536 [6:34:11<4:33:06,  1.66it/s]
 58%|█████▊    | 38270/65536 [6:34:12<4:35:32,  1.65it/s]
 58%|█████▊    | 38271/65536 [6:34:12<4:32:32,  1.67it/s]
 58%|█████▊    | 38272/65536 [6:34:13<4:38:48,  1.63it/s]
 58%|█████▊    | 38273/65536 [6:34:14<4:43:01,  1.61it/s]
 58%|█████▊    | 38274/65536 [6:34:14<4:40:40,  1.62it/s]
 58%|█████▊    | 38275/65536 [6:34:15<4:44:17,  1.60it/s]
 58%|█████▊    | 38276/65536 [6:34:16<4:39:31,  1.63it/s]
 58%|█████▊    | 38277/65536 [6:34:16<4:36:01,  1.65it/s]
 58%|█████▊    | 38278/65536 [6:34:17<4:36:05,  1.65it/s]
 58%|█████▊    | 38279/65536 [6:34:17<4:33:55,  1.66it/s]
 58%|█████▊    | 38280/65536 [6:34:18<4:38:18,  1.63it/s]
                                                         
{'loss': 1.6183, 'learning_rate': 4.7544998163340263e-07, 'epoch': 2362.96}
+
 58%|█████▊    | 38280/65536 [6:34:18<4:38:18,  1.63it/s]
 58%|█████▊    | 38281/65536 [6:34:19<4:44:37,  1.60it/s]
 58%|█████▊    | 38282/65536 [6:34:19<4:42:24,  1.61it/s]
 58%|█████▊    | 38283/65536 [6:34:20<4:39:31,  1.62it/s]
 58%|█████▊    | 38284/65536 [6:34:20<4:42:14,  1.61it/s]
 58%|█████▊    | 38285/65536 [6:34:21<4:36:43,  1.64it/s]
 58%|█████▊    | 38286/65536 [6:34:22<4:41:31,  1.61it/s]
 58%|█████▊    | 38287/65536 [6:34:22<4:40:37,  1.62it/s]
 58%|█████▊    | 38288/65536 [6:34:23<4:49:10,  1.57it/s]
 58%|█████▊    | 38289/65536 [6:34:24<4:45:10,  1.59it/s]
 58%|█████▊    | 38290/65536 [6:34:24<4:49:52,  1.57it/s]
 58%|█████▊    | 38291/65536 [6:34:25<4:44:41,  1.59it/s]
 58%|█████▊    | 38292/65536 [6:34:25<4:44:43,  1.59it/s]
 58%|█████▊    | 38293/65536 [6:34:26<4:39:18,  1.63it/s]
 58%|█████▊    | 38294/65536 [6:34:27<4:34:38,  1.65it/s]
 58%|█████▊    | 38295/65536 [6:34:27<4:31:44,  1.67it/s]
 58%|█████▊    | 38296/65536 [6:34:28<4:36:41,  1.64it/s]
 58%|█████▊    | 38297/65536 [6:34:29<4:42:08,  1.61it/s]
 58%|█████▊    | 38298/65536 [6:34:29<4:37:40,  1.63it/s]
 58%|█████▊    | 38299/65536 [6:34:30<4:44:28,  1.60it/s]
 58%|█████▊    | 38300/65536 [6:34:30<4:40:47,  1.62it/s]
                                                         
{'loss': 1.5934, 'learning_rate': 4.751744826741765e-07, 'epoch': 2364.2}
+
 58%|█████▊    | 38300/65536 [6:34:30<4:40:47,  1.62it/s]
 58%|█████▊    | 38301/65536 [6:34:31<4:38:49,  1.63it/s]
 58%|█████▊    | 38302/65536 [6:34:32<4:39:49,  1.62it/s]
 58%|█████▊    | 38303/65536 [6:34:32<4:36:57,  1.64it/s]
 58%|█████▊    | 38304/65536 [6:34:33<4:40:40,  1.62it/s]
 58%|█████▊    | 38305/65536 [6:34:33<4:34:02,  1.66it/s]
 58%|█████▊    | 38306/65536 [6:34:34<4:30:00,  1.68it/s]
 58%|█████▊    | 38307/65536 [6:34:35<4:38:33,  1.63it/s]
 58%|█████▊    | 38308/65536 [6:34:35<4:31:47,  1.67it/s]
 58%|█████▊    | 38309/65536 [6:34:36<4:35:35,  1.65it/s]
 58%|█████▊    | 38310/65536 [6:34:36<4:33:56,  1.66it/s]
 58%|█████▊    | 38311/65536 [6:34:37<4:47:23,  1.58it/s]
 58%|█████▊    | 38312/65536 [6:34:38<4:41:56,  1.61it/s]
 58%|█████▊    | 38313/65536 [6:34:38<4:34:29,  1.65it/s]
 58%|█████▊    | 38314/65536 [6:34:39<4:43:34,  1.60it/s]
 58%|█████▊    | 38315/65536 [6:34:40<4:37:34,  1.63it/s]
 58%|█████▊    | 38316/65536 [6:34:40<4:37:13,  1.64it/s]
 58%|█████▊    | 38317/65536 [6:34:41<4:31:51,  1.67it/s]
 58%|█████▊    | 38318/65536 [6:34:41<4:28:51,  1.69it/s]
 58%|█████▊    | 38319/65536 [6:34:42<4:29:23,  1.68it/s]
 58%|█████▊    | 38320/65536 [6:34:43<4:41:09,  1.61it/s]
                                                         
{'loss': 1.5774, 'learning_rate': 4.7489898371495035e-07, 'epoch': 2365.43}
+
 58%|█████▊    | 38320/65536 [6:34:43<4:41:09,  1.61it/s]
 58%|█████▊    | 38321/65536 [6:34:43<4:42:10,  1.61it/s]
 58%|█████▊    | 38322/65536 [6:34:44<4:43:29,  1.60it/s]
 58%|█████▊    | 38323/65536 [6:34:44<4:44:15,  1.60it/s]
 58%|█████▊    | 38324/65536 [6:34:45<4:37:36,  1.63it/s]
 58%|█████▊    | 38325/65536 [6:34:46<4:35:16,  1.65it/s]
 58%|█████▊    | 38326/65536 [6:34:46<4:38:25,  1.63it/s]
 58%|█████▊    | 38327/65536 [6:34:47<4:41:15,  1.61it/s]
 58%|█████▊    | 38328/65536 [6:34:48<4:41:41,  1.61it/s]
 58%|█████▊    | 38329/65536 [6:34:48<4:39:46,  1.62it/s]
 58%|█████▊    | 38330/65536 [6:34:49<4:44:50,  1.59it/s]
 58%|█████▊    | 38331/65536 [6:34:49<4:41:06,  1.61it/s]
 58%|█████▊    | 38332/65536 [6:34:50<4:43:35,  1.60it/s]
 58%|█████▊    | 38333/65536 [6:34:51<4:38:19,  1.63it/s]
 58%|█████▊    | 38334/65536 [6:34:51<4:32:49,  1.66it/s]
 58%|█████▊    | 38335/65536 [6:34:52<4:43:15,  1.60it/s]
 58%|█████▊    | 38336/65536 [6:34:53<4:46:34,  1.58it/s]
 58%|█████▊    | 38337/65536 [6:34:53<4:48:17,  1.57it/s]
 58%|█████▊    | 38338/65536 [6:34:54<4:45:10,  1.59it/s]
 59%|█████▊    | 38339/65536 [6:34:54<4:39:09,  1.62it/s]
 59%|█████▊    | 38340/65536 [6:34:55<4:38:42,  1.63it/s]
                                                         
{'loss': 1.6208, 'learning_rate': 4.7462348475572426e-07, 'epoch': 2366.67}
+
 59%|█████▊    | 38340/65536 [6:34:55<4:38:42,  1.63it/s]
 59%|█████▊    | 38341/65536 [6:34:56<4:34:28,  1.65it/s]
 59%|█████▊    | 38342/65536 [6:34:56<4:32:34,  1.66it/s]
 59%|█████▊    | 38343/65536 [6:34:57<4:33:03,  1.66it/s]
 59%|█████▊    | 38344/65536 [6:34:57<4:35:57,  1.64it/s]
 59%|█████▊    | 38345/65536 [6:34:58<4:35:21,  1.65it/s]
 59%|█████▊    | 38346/65536 [6:34:59<4:46:47,  1.58it/s]
 59%|█████▊    | 38347/65536 [6:34:59<4:42:48,  1.60it/s]
 59%|█████▊    | 38348/65536 [6:35:00<4:54:21,  1.54it/s]
 59%|█████▊    | 38349/65536 [6:35:01<4:47:58,  1.57it/s]
 59%|█████▊    | 38350/65536 [6:35:01<4:46:40,  1.58it/s]
 59%|█████▊    | 38351/65536 [6:35:02<4:43:23,  1.60it/s]
 59%|█████▊    | 38352/65536 [6:35:02<4:46:34,  1.58it/s]
 59%|█████▊    | 38353/65536 [6:35:03<4:41:05,  1.61it/s]
 59%|█████▊    | 38354/65536 [6:35:04<4:37:07,  1.63it/s]
 59%|█████▊    | 38355/65536 [6:35:04<4:31:44,  1.67it/s]
 59%|█████▊    | 38356/65536 [6:35:05<4:28:52,  1.68it/s]
 59%|█████▊    | 38357/65536 [6:35:05<4:28:04,  1.69it/s]
 59%|█████▊    | 38358/65536 [6:35:06<4:41:44,  1.61it/s]
 59%|█████▊    | 38359/65536 [6:35:07<4:44:08,  1.59it/s]
 59%|█████▊    | 38360/65536 [6:35:07<4:38:32,  1.63it/s]
                                                         
{'loss': 1.6088, 'learning_rate': 4.743479857964981e-07, 'epoch': 2367.9}
+
 59%|█████▊    | 38360/65536 [6:35:07<4:38:32,  1.63it/s]
 59%|█████▊    | 38361/65536 [6:35:08<4:36:48,  1.64it/s]
 59%|█████▊    | 38362/65536 [6:35:09<4:39:43,  1.62it/s]
 59%|█████▊    | 38363/65536 [6:35:09<4:40:09,  1.62it/s]
 59%|█████▊    | 38364/65536 [6:35:10<4:38:08,  1.63it/s]
 59%|█████▊    | 38365/65536 [6:35:10<4:36:40,  1.64it/s]
 59%|█████▊    | 38366/65536 [6:35:11<4:37:53,  1.63it/s]
 59%|█████▊    | 38367/65536 [6:35:12<4:43:06,  1.60it/s]
 59%|█████▊    | 38368/65536 [6:35:12<4:35:00,  1.65it/s]
 59%|█████▊    | 38369/65536 [6:35:13<4:40:42,  1.61it/s]
 59%|█████▊    | 38370/65536 [6:35:13<4:39:53,  1.62it/s]
 59%|█████▊    | 38371/65536 [6:35:14<4:40:32,  1.61it/s]
 59%|█████▊    | 38372/65536 [6:35:15<4:44:44,  1.59it/s]
 59%|█████▊    | 38373/65536 [6:35:15<4:37:46,  1.63it/s]
 59%|█████▊    | 38374/65536 [6:35:16<4:44:14,  1.59it/s]
 59%|█████▊    | 38375/65536 [6:35:17<4:39:40,  1.62it/s]
 59%|█████▊    | 38376/65536 [6:35:17<4:37:14,  1.63it/s]
 59%|█████▊    | 38377/65536 [6:35:18<4:37:47,  1.63it/s]
 59%|█████▊    | 38378/65536 [6:35:18<4:48:01,  1.57it/s]
 59%|█████▊    | 38379/65536 [6:35:19<4:40:22,  1.61it/s]
 59%|█████▊    | 38380/65536 [6:35:20<4:38:52,  1.62it/s]
                                                         
{'loss': 1.6248, 'learning_rate': 4.74072486837272e-07, 'epoch': 2369.14}
+
 59%|█████▊    | 38380/65536 [6:35:20<4:38:52,  1.62it/s]
 59%|█████▊    | 38381/65536 [6:35:20<4:43:22,  1.60it/s]
 59%|█████▊    | 38382/65536 [6:35:21<4:39:41,  1.62it/s]
 59%|█████▊    | 38383/65536 [6:35:22<4:34:50,  1.65it/s]
 59%|█████▊    | 38384/65536 [6:35:22<4:31:08,  1.67it/s]
 59%|█████▊    | 38385/65536 [6:35:23<4:28:59,  1.68it/s]
 59%|���████▊    | 38386/65536 [6:35:23<4:29:56,  1.68it/s]
 59%|█████▊    | 38387/65536 [6:35:24<4:37:21,  1.63it/s]
 59%|█████▊    | 38388/65536 [6:35:25<4:43:12,  1.60it/s]
 59%|█████▊    | 38389/65536 [6:35:25<4:40:44,  1.61it/s]
 59%|█████▊    | 38390/65536 [6:35:26<4:34:54,  1.65it/s]
 59%|█████▊    | 38391/65536 [6:35:26<4:39:35,  1.62it/s]
 59%|█████▊    | 38392/65536 [6:35:27<4:37:03,  1.63it/s]
 59%|█████▊    | 38393/65536 [6:35:28<4:37:10,  1.63it/s]
 59%|█████▊    | 38394/65536 [6:35:28<4:42:46,  1.60it/s]
 59%|█████▊    | 38395/65536 [6:35:29<4:53:49,  1.54it/s]
 59%|█████▊    | 38396/65536 [6:35:30<4:52:52,  1.54it/s]
 59%|█████▊    | 38397/65536 [6:35:30<4:44:55,  1.59it/s]
 59%|█████▊    | 38398/65536 [6:35:31<4:37:44,  1.63it/s]
 59%|█████▊    | 38399/65536 [6:35:31<4:37:38,  1.63it/s]
 59%|█████▊    | 38400/65536 [6:35:32<4:38:14,  1.63it/s]
                                                         
{'loss': 1.5924, 'learning_rate': 4.7379698787804584e-07, 'epoch': 2370.37}
+
 59%|█████▊    | 38400/65536 [6:35:32<4:38:14,  1.63it/s]
 59%|█████▊    | 38401/65536 [6:35:33<4:37:28,  1.63it/s]
 59%|█████▊    | 38402/65536 [6:35:33<4:40:43,  1.61it/s]
 59%|█████▊    | 38403/65536 [6:35:34<4:36:12,  1.64it/s]
 59%|█████▊    | 38404/65536 [6:35:34<4:34:09,  1.65it/s]
 59%|█████▊    | 38405/65536 [6:35:35<4:37:41,  1.63it/s]
 59%|█████▊    | 38406/65536 [6:35:36<4:36:40,  1.63it/s]
 59%|█████▊    | 38407/65536 [6:35:36<4:28:11,  1.69it/s]
 59%|█████▊    | 38408/65536 [6:35:37<4:26:31,  1.70it/s]
 59%|█████▊    | 38409/65536 [6:35:38<4:40:14,  1.61it/s]
 59%|█████▊    | 38410/65536 [6:35:38<4:34:04,  1.65it/s]
 59%|█████▊    | 38411/65536 [6:35:39<4:44:09,  1.59it/s]
 59%|█████▊    | 38412/65536 [6:35:39<4:39:25,  1.62it/s]
 59%|█████▊    | 38413/65536 [6:35:40<4:32:49,  1.66it/s]
 59%|█████▊    | 38414/65536 [6:35:41<4:38:18,  1.62it/s]
 59%|█████▊    | 38415/65536 [6:35:41<4:35:36,  1.64it/s]
 59%|█████▊    | 38416/65536 [6:35:42<4:38:58,  1.62it/s]
 59%|█████▊    | 38417/65536 [6:35:42<4:38:54,  1.62it/s]
 59%|█████▊    | 38418/65536 [6:35:43<4:37:50,  1.63it/s]
 59%|█████▊    | 38419/65536 [6:35:44<4:48:24,  1.57it/s]
 59%|█████▊    | 38420/65536 [6:35:44<4:44:15,  1.59it/s]
                                                         
{'loss': 1.6087, 'learning_rate': 4.735214889188196e-07, 'epoch': 2371.6}
+
 59%|█████▊    | 38420/65536 [6:35:44<4:44:15,  1.59it/s]
 59%|█████▊    | 38421/65536 [6:35:45<4:43:38,  1.59it/s]
 59%|█████▊    | 38422/65536 [6:35:46<4:52:37,  1.54it/s]
 59%|█████▊    | 38423/65536 [6:35:46<4:43:37,  1.59it/s]
 59%|█████▊    | 38424/65536 [6:35:47<4:33:17,  1.65it/s]
 59%|█████▊    | 38425/65536 [6:35:47<4:38:42,  1.62it/s]
 59%|█████▊    | 38426/65536 [6:35:48<4:36:46,  1.63it/s]
 59%|█████▊    | 38427/65536 [6:35:49<4:47:17,  1.57it/s]
 59%|█████▊    | 38428/65536 [6:35:49<4:50:52,  1.55it/s]
 59%|█████▊    | 38429/65536 [6:35:50<4:42:02,  1.60it/s]
 59%|█████▊    | 38430/65536 [6:35:51<4:34:44,  1.64it/s]
 59%|█████▊    | 38431/65536 [6:35:51<4:45:02,  1.58it/s]
 59%|█████▊    | 38432/65536 [6:35:52<4:41:52,  1.60it/s]
 59%|█████▊    | 38433/65536 [6:35:52<4:32:58,  1.65it/s]
 59%|█████▊    | 38434/65536 [6:35:53<4:28:31,  1.68it/s]
 59%|█████▊    | 38435/65536 [6:35:54<4:37:16,  1.63it/s]
 59%|█████▊    | 38436/65536 [6:35:54<4:36:32,  1.63it/s]
 59%|█████▊    | 38437/65536 [6:35:55<4:41:41,  1.60it/s]
 59%|█████▊    | 38438/65536 [6:35:55<4:40:42,  1.61it/s]
 59%|█████▊    | 38439/65536 [6:35:56<4:35:52,  1.64it/s]
 59%|█████▊    | 38440/65536 [6:35:57<4:35:44,  1.64it/s]
                                                         
{'loss': 1.5936, 'learning_rate': 4.7324598995959345e-07, 'epoch': 2372.84}
+
 59%|█████▊    | 38440/65536 [6:35:57<4:35:44,  1.64it/s]
 59%|█████▊    | 38441/65536 [6:35:57<4:36:51,  1.63it/s]
 59%|█████▊    | 38442/65536 [6:35:58<4:32:37,  1.66it/s]
 59%|█████▊    | 38443/65536 [6:35:59<4:40:25,  1.61it/s]
 59%|█████▊    | 38444/65536 [6:35:59<4:35:54,  1.64it/s]
 59%|█████▊    | 38445/65536 [6:36:00<4:34:36,  1.64it/s]
 59%|█████▊    | 38446/65536 [6:36:00<4:33:33,  1.65it/s]
 59%|█████▊    | 38447/65536 [6:36:01<4:30:00,  1.67it/s]
 59%|█████▊    | 38448/65536 [6:36:02<4:34:56,  1.64it/s]
 59%|█████▊    | 38449/65536 [6:36:02<4:42:23,  1.60it/s]
 59%|█████▊    | 38450/65536 [6:36:03<4:34:27,  1.64it/s]
 59%|█████▊    | 38451/65536 [6:36:03<4:32:51,  1.65it/s]
 59%|█████▊    | 38452/65536 [6:36:04<4:33:44,  1.65it/s]
 59%|█████▊    | 38453/65536 [6:36:05<4:35:45,  1.64it/s]
 59%|█████▊    | 38454/65536 [6:36:05<4:37:45,  1.63it/s]
 59%|█████▊    | 38455/65536 [6:36:06<4:42:24,  1.60it/s]
 59%|█████▊    | 38456/65536 [6:36:07<4:50:42,  1.55it/s]
 59%|█████▊    | 38457/65536 [6:36:07<4:43:57,  1.59it/s]
 59%|█████▊    | 38458/65536 [6:36:08<4:35:12,  1.64it/s]
 59%|█████▊    | 38459/65536 [6:36:08<4:43:01,  1.59it/s]
 59%|█████▊    | 38460/65536 [6:36:09<4:37:03,  1.63it/s]
                                                         
{'loss': 1.6532, 'learning_rate': 4.729704910003673e-07, 'epoch': 2374.07}
+
 59%|█████▊    | 38460/65536 [6:36:09<4:37:03,  1.63it/s]
 59%|█████▊    | 38461/65536 [6:36:10<4:38:30,  1.62it/s]
 59%|█████▊    | 38462/65536 [6:36:10<4:44:38,  1.59it/s]
 59%|█████▊    | 38463/65536 [6:36:11<4:38:39,  1.62it/s]
 59%|█████▊    | 38464/65536 [6:36:11<4:38:33,  1.62it/s]
 59%|█████▊    | 38465/65536 [6:36:12<4:33:40,  1.65it/s]
 59%|█████▊    | 38466/65536 [6:36:13<4:31:54,  1.66it/s]
 59%|█████▊    | 38467/65536 [6:36:13<4:27:50,  1.68it/s]
 59%|█████▊    | 38468/65536 [6:36:14<4:30:30,  1.67it/s]
 59%|█████▊    | 38469/65536 [6:36:14<4:34:56,  1.64it/s]
 59%|█████▊    | 38470/65536 [6:36:15<4:42:53,  1.59it/s]
 59%|█████▊    | 38471/65536 [6:36:16<4:41:34,  1.60it/s]
 59%|█████▊    | 38472/65536 [6:36:16<4:40:27,  1.61it/s]
 59%|█████▊    | 38473/65536 [6:36:17<4:50:43,  1.55it/s]
 59%|█████▊    | 38474/65536 [6:36:18<4:45:49,  1.58it/s]
 59%|█████▊    | 38475/65536 [6:36:18<4:37:32,  1.63it/s]
 59%|█████▊    | 38476/65536 [6:36:19<4:53:21,  1.54it/s]
 59%|█████▊    | 38477/65536 [6:36:20<4:44:22,  1.59it/s]
 59%|█████▊    | 38478/65536 [6:36:20<4:38:44,  1.62it/s]
 59%|█████▊    | 38479/65536 [6:36:21<4:39:25,  1.61it/s]
 59%|█████▊    | 38480/65536 [6:36:21<4:37:50,  1.62it/s]
                                                         
{'loss': 1.6262, 'learning_rate': 4.7269499204114117e-07, 'epoch': 2375.31}
+
 59%|█████▊    | 38480/65536 [6:36:21<4:37:50,  1.62it/s]
 59%|█████▊    | 38481/65536 [6:36:22<4:39:48,  1.61it/s]
 59%|█████▊    | 38482/65536 [6:36:23<4:35:36,  1.64it/s]
 59%|█████▊    | 38483/65536 [6:36:23<4:29:45,  1.67it/s]
 59%|█████▊    | 38484/65536 [6:36:24<4:36:59,  1.63it/s]
 59%|█████▊    | 38485/65536 [6:36:24<4:35:40,  1.64it/s]
 59%|█████▊    | 38486/65536 [6:36:25<4:30:34,  1.67it/s]
 59%|█████▊    | 38487/65536 [6:36:26<4:31:49,  1.66it/s]
 59%|█████▊    | 38488/65536 [6:36:26<4:31:37,  1.66it/s]
 59%|█████▊    | 38489/65536 [6:36:27<4:33:55,  1.65it/s]
 59%|█████▊    | 38490/65536 [6:36:27<4:38:09,  1.62it/s]
 59%|█████▊    | 38491/65536 [6:36:28<4:46:39,  1.57it/s]
 59%|█████▊    | 38492/65536 [6:36:29<4:49:19,  1.56it/s]
 59%|█████▊    | 38493/65536 [6:36:29<4:52:24,  1.54it/s]
 59%|█████▊    | 38494/65536 [6:36:30<4:42:36,  1.59it/s]
 59%|█████▊    | 38495/65536 [6:36:31<4:50:16,  1.55it/s]
 59%|█████▊    | 38496/65536 [6:36:31<4:45:01,  1.58it/s]
 59%|█████▊    | 38497/65536 [6:36:32<4:43:15,  1.59it/s]
 59%|█████▊    | 38498/65536 [6:36:33<4:42:17,  1.60it/s]
 59%|█████▊    | 38499/65536 [6:36:33<4:37:11,  1.63it/s]
 59%|█████▊    | 38500/65536 [6:36:34<4:35:40,  1.63it/s]
                                                         
{'loss': 1.6101, 'learning_rate': 4.72419493081915e-07, 'epoch': 2376.54}
+
 59%|█████▊    | 38500/65536 [6:36:34<4:35:40,  1.63it/s]
 59%|█████▊    | 38501/65536 [6:36:34<4:39:51,  1.61it/s]
 59%|█████▊    | 38502/65536 [6:36:35<4:37:39,  1.62it/s]
 59%|█████▉    | 38503/65536 [6:36:36<4:35:00,  1.64it/s]
 59%|█████▉    | 38504/65536 [6:36:36<4:31:12,  1.66it/s]
 59%|█████▉    | 38505/65536 [6:36:37<4:38:26,  1.62it/s]
 59%|█████▉    | 38506/65536 [6:36:37<4:36:47,  1.63it/s]
 59%|█████▉    | 38507/65536 [6:36:38<4:31:16,  1.66it/s]
 59%|█████▉    | 38508/65536 [6:36:39<4:40:06,  1.61it/s]
 59%|█████▉    | 38509/65536 [6:36:39<4:39:47,  1.61it/s]
 59%|█████▉    | 38510/65536 [6:36:40<4:37:49,  1.62it/s]
 59%|█████▉    | 38511/65536 [6:36:41<4:44:32,  1.58it/s]
 59%|█████▉    | 38512/65536 [6:36:41<4:39:38,  1.61it/s]
 59%|█████▉    | 38513/65536 [6:36:42<4:34:08,  1.64it/s]
 59%|█████▉    | 38514/65536 [6:36:42<4:42:40,  1.59it/s]
 59%|█████▉    | 38515/65536 [6:36:43<4:49:09,  1.56it/s]
 59%|█████▉    | 38516/65536 [6:36:44<4:43:03,  1.59it/s]
 59%|█████▉    | 38517/65536 [6:36:44<4:37:44,  1.62it/s]
 59%|█████▉    | 38518/65536 [6:36:45<4:35:49,  1.63it/s]
 59%|█████▉    | 38519/65536 [6:36:45<4:33:23,  1.65it/s]
 59%|█████▉    | 38520/65536 [6:36:46<4:30:13,  1.67it/s]
                                                         
{'loss': 1.634, 'learning_rate': 4.721439941226889e-07, 'epoch': 2377.78}
+
 59%|█████▉    | 38520/65536 [6:36:46<4:30:13,  1.67it/s]
 59%|█████▉    | 38521/65536 [6:36:47<4:38:07,  1.62it/s]
 59%|█████▉    | 38522/65536 [6:36:47<4:38:23,  1.62it/s]
 59%|█████▉    | 38523/65536 [6:36:48<4:33:59,  1.64it/s]
 59%|█████▉    | 38524/65536 [6:36:49<4:44:04,  1.58it/s]
 59%|█████▉    | 38525/65536 [6:36:49<4:39:43,  1.61it/s]
 59%|█████▉    | 38526/65536 [6:36:50<4:39:37,  1.61it/s]
 59%|█████▉    | 38527/65536 [6:36:51<4:46:54,  1.57it/s]
 59%|█████▉    | 38528/65536 [6:36:51<4:43:50,  1.59it/s]
 59%|█████▉    | 38529/65536 [6:36:52<4:47:17,  1.57it/s]
 59%|█████▉    | 38530/65536 [6:36:52<4:38:25,  1.62it/s]
 59%|█████▉    | 38531/65536 [6:36:53<4:40:31,  1.60it/s]
 59%|█████▉    | 38532/65536 [6:36:54<4:34:25,  1.64it/s]
 59%|█████▉    | 38533/65536 [6:36:54<4:34:06,  1.64it/s]
 59%|█████▉    | 38534/65536 [6:36:55<4:31:06,  1.66it/s]
 59%|█████▉    | 38535/65536 [6:36:55<4:32:41,  1.65it/s]
 59%|█████▉    | 38536/65536 [6:36:56<4:32:13,  1.65it/s]
 59%|█████▉    | 38537/65536 [6:36:57<4:27:12,  1.68it/s]
 59%|█████▉    | 38538/65536 [6:36:57<4:26:57,  1.69it/s]
 59%|█████▉    | 38539/65536 [6:36:58<4:30:55,  1.66it/s]
 59%|█████▉    | 38540/65536 [6:36:58<4:41:18,  1.60it/s]
                                                         
{'loss': 1.6157, 'learning_rate': 4.7186849516346274e-07, 'epoch': 2379.01}
+
 59%|█████▉    | 38540/65536 [6:36:58<4:41:18,  1.60it/s]
 59%|█████▉    | 38541/65536 [6:36:59<4:36:04,  1.63it/s]
 59%|█████▉    | 38542/65536 [6:37:00<4:35:56,  1.63it/s]
 59%|█████▉    | 38543/65536 [6:37:00<4:37:23,  1.62it/s]
 59%|█████▉    | 38544/65536 [6:37:01<4:40:21,  1.60it/s]
 59%|█████▉    | 38545/65536 [6:37:02<4:40:25,  1.60it/s]
 59%|█████▉    | 38546/65536 [6:37:02<4:34:21,  1.64it/s]
 59%|█████▉    | 38547/65536 [6:37:03<4:29:35,  1.67it/s]
 59%|█████▉    | 38548/65536 [6:37:03<4:28:14,  1.68it/s]
 59%|█████▉    | 38549/65536 [6:37:04<4:36:31,  1.63it/s]
 59%|█████▉    | 38550/65536 [6:37:05<4:34:46,  1.64it/s]
 59%|█████▉    | 38551/65536 [6:37:05<4:38:21,  1.62it/s]
 59%|█████▉    | 38552/65536 [6:37:06<4:34:39,  1.64it/s]
 59%|█████▉    | 38553/65536 [6:37:06<4:31:59,  1.65it/s]
 59%|█████▉    | 38554/65536 [6:37:07<4:29:04,  1.67it/s]
 59%|█████▉    | 38555/65536 [6:37:08<4:30:42,  1.66it/s]
 59%|█████▉    | 38556/65536 [6:37:08<4:25:41,  1.69it/s]
 59%|█████▉    | 38557/65536 [6:37:09<4:38:46,  1.61it/s]
 59%|█████▉    | 38558/65536 [6:37:09<4:32:59,  1.65it/s]
 59%|█████▉    | 38559/65536 [6:37:10<4:38:18,  1.62it/s]
 59%|█████▉    | 38560/65536 [6:37:11<4:30:37,  1.66it/s]
                                                         
{'loss': 1.6171, 'learning_rate': 4.715929962042365e-07, 'epoch': 2380.25}
+
 59%|█████▉    | 38560/65536 [6:37:11<4:30:37,  1.66it/s]
 59%|█████▉    | 38561/65536 [6:37:11<4:42:03,  1.59it/s]
 59%|█████▉    | 38562/65536 [6:37:12<4:36:10,  1.63it/s]
 59%|█████▉    | 38563/65536 [6:37:12<4:31:16,  1.66it/s]
 59%|█████▉    | 38564/65536 [6:37:13<4:33:54,  1.64it/s]
 59%|█████▉    | 38565/65536 [6:37:14<4:44:43,  1.58it/s]
 59%|█████▉    | 38566/65536 [6:37:14<4:38:34,  1.61it/s]
 59%|█████▉    | 38567/65536 [6:37:15<4:34:21,  1.64it/s]
 59%|█████▉    | 38568/65536 [6:37:16<4:33:01,  1.65it/s]
 59%|█████▉    | 38569/65536 [6:37:16<4:29:27,  1.67it/s]
 59%|█████▉    | 38570/65536 [6:37:17<4:26:06,  1.69it/s]
 59%|█████▉    | 38571/65536 [6:37:17<4:27:38,  1.68it/s]
 59%|█████▉    | 38572/65536 [6:37:18<4:26:11,  1.69it/s]
 59%|█████▉    | 38573/65536 [6:37:19<4:39:36,  1.61it/s]
 59%|█████▉    | 38574/65536 [6:37:19<4:46:06,  1.57it/s]
 59%|█████▉    | 38575/65536 [6:37:20<4:41:24,  1.60it/s]
 59%|█████▉    | 38576/65536 [6:37:20<4:36:16,  1.63it/s]
 59%|█████▉    | 38577/65536 [6:37:21<4:39:01,  1.61it/s]
 59%|█████▉    | 38578/65536 [6:37:22<4:34:19,  1.64it/s]
 59%|█��███▉    | 38579/65536 [6:37:22<4:31:56,  1.65it/s]
 59%|█████▉    | 38580/65536 [6:37:23<4:35:04,  1.63it/s]
                                                         
{'loss': 1.6804, 'learning_rate': 4.7131749724501036e-07, 'epoch': 2381.48}
+
 59%|█████▉    | 38580/65536 [6:37:23<4:35:04,  1.63it/s]
 59%|█████▉    | 38581/65536 [6:37:23<4:33:22,  1.64it/s]
 59%|█████▉    | 38582/65536 [6:37:24<4:26:13,  1.69it/s]
 59%|█████▉    | 38583/65536 [6:37:25<4:28:52,  1.67it/s]
 59%|█████▉    | 38584/65536 [6:37:25<4:35:32,  1.63it/s]
 59%|█████▉    | 38585/65536 [6:37:26<4:35:18,  1.63it/s]
 59%|█████▉    | 38586/65536 [6:37:26<4:32:55,  1.65it/s]
 59%|█████▉    | 38587/65536 [6:37:27<4:30:27,  1.66it/s]
 59%|█████▉    | 38588/65536 [6:37:28<4:25:57,  1.69it/s]
 59%|█████▉    | 38589/65536 [6:37:28<4:44:29,  1.58it/s]
 59%|█████▉    | 38590/65536 [6:37:29<4:41:09,  1.60it/s]
 59%|█████▉    | 38591/65536 [6:37:30<4:35:17,  1.63it/s]
 59%|█████▉    | 38592/65536 [6:37:30<4:36:36,  1.62it/s]
 59%|█████▉    | 38593/65536 [6:37:31<4:44:12,  1.58it/s]
 59%|█████▉    | 38594/65536 [6:37:31<4:41:38,  1.59it/s]
 59%|█████▉    | 38595/65536 [6:37:32<4:39:52,  1.60it/s]
 59%|█████▉    | 38596/65536 [6:37:33<4:36:36,  1.62it/s]
 59%|█████▉    | 38597/65536 [6:37:33<4:30:24,  1.66it/s]
 59%|█████▉    | 38598/65536 [6:37:34<4:28:06,  1.67it/s]
 59%|█████▉    | 38599/65536 [6:37:34<4:32:08,  1.65it/s]
 59%|█████▉    | 38600/65536 [6:37:35<4:36:18,  1.62it/s]
                                                         
{'loss': 1.596, 'learning_rate': 4.710419982857842e-07, 'epoch': 2382.72}
+
 59%|█████▉    | 38600/65536 [6:37:35<4:36:18,  1.62it/s]
 59%|█████▉    | 38601/65536 [6:37:36<4:31:34,  1.65it/s]
 59%|█████▉    | 38602/65536 [6:37:36<4:31:01,  1.66it/s]
 59%|█████▉    | 38603/65536 [6:37:37<4:25:57,  1.69it/s]
 59%|█████▉    | 38604/65536 [6:37:37<4:25:10,  1.69it/s]
 59%|█████▉    | 38605/65536 [6:37:38<4:36:40,  1.62it/s]
 59%|█████▉    | 38606/65536 [6:37:39<4:35:11,  1.63it/s]
 59%|█████▉    | 38607/65536 [6:37:39<4:35:12,  1.63it/s]
 59%|█████▉    | 38608/65536 [6:37:40<4:32:51,  1.64it/s]
 59%|█████▉    | 38609/65536 [6:37:41<4:26:33,  1.68it/s]
 59%|█████▉    | 38610/65536 [6:37:41<4:26:25,  1.68it/s]
 59%|█████▉    | 38611/65536 [6:37:42<4:35:04,  1.63it/s]
 59%|█████▉    | 38612/65536 [6:37:42<4:33:42,  1.64it/s]
 59%|█████▉    | 38613/65536 [6:37:43<4:29:49,  1.66it/s]
 59%|█████▉    | 38614/65536 [6:37:44<4:31:00,  1.66it/s]
 59%|█████▉    | 38615/65536 [6:37:44<4:26:07,  1.69it/s]
 59%|█████▉    | 38616/65536 [6:37:45<4:28:37,  1.67it/s]
 59%|█████▉    | 38617/65536 [6:37:45<4:31:58,  1.65it/s]
 59%|█████▉    | 38618/65536 [6:37:46<4:29:17,  1.67it/s]
 59%|█████▉    | 38619/65536 [6:37:47<4:34:59,  1.63it/s]
 59%|█████▉    | 38620/65536 [6:37:47<4:39:51,  1.60it/s]
                                                         
{'loss': 1.6372, 'learning_rate': 4.7076649932655807e-07, 'epoch': 2383.95}
+
 59%|█████▉    | 38620/65536 [6:37:47<4:39:51,  1.60it/s]
 59%|█████▉    | 38621/65536 [6:37:48<4:46:07,  1.57it/s]
 59%|█████▉    | 38622/65536 [6:37:49<4:44:08,  1.58it/s]
 59%|█████▉    | 38623/65536 [6:37:49<4:36:26,  1.62it/s]
 59%|█████▉    | 38624/65536 [6:37:50<4:45:32,  1.57it/s]
 59%|█████▉    | 38625/65536 [6:37:50<4:40:38,  1.60it/s]
 59%|█████▉    | 38626/65536 [6:37:51<4:40:09,  1.60it/s]
 59%|█████▉    | 38627/65536 [6:37:52<4:32:55,  1.64it/s]
 59%|█████▉    | 38628/65536 [6:37:52<4:37:22,  1.62it/s]
 59%|█████▉    | 38629/65536 [6:37:53<4:41:44,  1.59it/s]
 59%|█████▉    | 38630/65536 [6:37:54<4:46:01,  1.57it/s]
 59%|█████▉    | 38631/65536 [6:37:54<4:42:57,  1.58it/s]
 59%|█████▉    | 38632/65536 [6:37:55<4:41:22,  1.59it/s]
 59%|█████▉    | 38633/65536 [6:37:55<4:35:44,  1.63it/s]
 59%|█████▉    | 38634/65536 [6:37:56<4:40:18,  1.60it/s]
 59%|█████▉    | 38635/65536 [6:37:57<4:33:41,  1.64it/s]
 59%|█████▉    | 38636/65536 [6:37:57<4:34:34,  1.63it/s]
 59%|█████▉    | 38637/65536 [6:37:58<4:30:30,  1.66it/s]
 59%|█████▉    | 38638/65536 [6:37:58<4:38:38,  1.61it/s]
 59%|█████▉    | 38639/65536 [6:37:59<4:38:33,  1.61it/s]
 59%|█████▉    | 38640/65536 [6:38:00<4:40:11,  1.60it/s]
                                                         
{'loss': 1.6268, 'learning_rate': 4.7049100036733193e-07, 'epoch': 2385.19}
+
 59%|█████▉    | 38640/65536 [6:38:00<4:40:11,  1.60it/s]
 59%|█████▉    | 38641/65536 [6:38:00<4:31:57,  1.65it/s]
 59%|█████▉    | 38642/65536 [6:38:01<4:39:02,  1.61it/s]
 59%|█████▉    | 38643/65536 [6:38:02<4:36:55,  1.62it/s]
 59%|█████▉    | 38644/65536 [6:38:02<4:28:28,  1.67it/s]
 59%|█████▉    | 38645/65536 [6:38:03<4:36:35,  1.62it/s]
 59%|█████▉    | 38646/65536 [6:38:03<4:36:24,  1.62it/s]
 59%|█████▉    | 38647/65536 [6:38:04<4:29:19,  1.66it/s]
 59%|█████▉    | 38648/65536 [6:38:04<4:26:41,  1.68it/s]
 59%|█████▉    | 38649/65536 [6:38:05<4:36:01,  1.62it/s]
 59%|█████▉    | 38650/65536 [6:38:06<4:34:36,  1.63it/s]
 59%|█████▉    | 38651/65536 [6:38:06<4:35:05,  1.63it/s]
 59%|█████▉    | 38652/65536 [6:38:07<4:32:41,  1.64it/s]
 59%|█████▉    | 38653/65536 [6:38:08<4:30:08,  1.66it/s]
 59%|█████▉    | 38654/65536 [6:38:08<4:40:43,  1.60it/s]
 59%|█████▉    | 38655/65536 [6:38:09<4:37:01,  1.62it/s]
 59%|█████▉    | 38656/65536 [6:38:09<4:30:56,  1.65it/s]
 59%|█████▉    | 38657/65536 [6:38:10<4:34:18,  1.63it/s]
 59%|█████▉    | 38658/65536 [6:38:11<4:35:37,  1.63it/s]
 59%|█████▉    | 38659/65536 [6:38:11<4:32:04,  1.65it/s]
 59%|█████▉    | 38660/65536 [6:38:12<4:36:51,  1.62it/s]
                                                         
{'loss': 1.6041, 'learning_rate': 4.702155014081058e-07, 'epoch': 2386.42}
+
 59%|█████▉    | 38660/65536 [6:38:12<4:36:51,  1.62it/s]
 59%|█████▉    | 38661/65536 [6:38:13<4:35:39,  1.62it/s]
 59%|█████▉    | 38662/65536 [6:38:13<4:39:32,  1.60it/s]
 59%|█████▉    | 38663/65536 [6:38:14<4:31:56,  1.65it/s]
 59%|█████▉    | 38664/65536 [6:38:14<4:28:28,  1.67it/s]
 59%|█████▉    | 38665/65536 [6:38:15<4:29:09,  1.66it/s]
 59%|█████▉    | 38666/65536 [6:38:16<4:31:40,  1.65it/s]
 59%|█████▉    | 38667/65536 [6:38:16<4:37:10,  1.62it/s]
 59%|█████▉    | 38668/65536 [6:38:17<4:35:11,  1.63it/s]
 59%|█████▉    | 38669/65536 [6:38:17<4:33:53,  1.63it/s]
 59%|█████▉    | 38670/65536 [6:38:18<4:43:32,  1.58it/s]
 59%|█████▉    | 38671/65536 [6:38:19<4:46:05,  1.57it/s]
 59%|█████▉    | 38672/65536 [6:38:19<4:50:53,  1.54it/s]
 59%|█████▉    | 38673/65536 [6:38:20<4:41:59,  1.59it/s]
 59%|█████▉    | 38674/65536 [6:38:21<4:35:38,  1.62it/s]
 59%|█████▉    | 38675/65536 [6:38:21<4:42:34,  1.58it/s]
 59%|█████▉    | 38676/65536 [6:38:22<4:50:11,  1.54it/s]
 59%|█████▉    | 38677/65536 [6:38:23<4:43:05,  1.58it/s]
 59%|█████▉    | 38678/65536 [6:38:23<4:38:40,  1.61it/s]
 59%|█████▉    | 38679/65536 [6:38:24<4:39:05,  1.60it/s]
 59%|█████▉    | 38680/65536 [6:38:24<4:35:35,  1.62it/s]
                                                         
{'loss': 1.6394, 'learning_rate': 4.6994000244887954e-07, 'epoch': 2387.65}
+
 59%|█████▉    | 38680/65536 [6:38:24<4:35:35,  1.62it/s]
 59%|█████▉    | 38681/65536 [6:38:25<4:35:30,  1.62it/s]
 59%|█████▉    | 38682/65536 [6:38:26<4:29:50,  1.66it/s]
 59%|█████▉    | 38683/65536 [6:38:26<4:31:07,  1.65it/s]
 59%|█████▉    | 38684/65536 [6:38:27<4:35:55,  1.62it/s]
 59%|█████▉    | 38685/65536 [6:38:27<4:30:59,  1.65it/s]
 59%|█████▉    | 38686/65536 [6:38:28<4:40:26,  1.60it/s]
 59%|█████▉    | 38687/65536 [6:38:29<4:31:39,  1.65it/s]
 59%|█████▉    | 38688/65536 [6:38:29<4:32:30,  1.64it/s]
 59%|█████▉    | 38689/65536 [6:38:30<4:30:03,  1.66it/s]
 59%|█████▉    | 38690/65536 [6:38:30<4:33:07,  1.64it/s]
 59%|█████▉    | 38691/65536 [6:38:31<4:40:01,  1.60it/s]
 59%|█████▉    | 38692/65536 [6:38:32<4:41:54,  1.59it/s]
 59%|█████▉    | 38693/65536 [6:38:32<4:34:04,  1.63it/s]
 59%|█████▉    | 38694/65536 [6:38:33<4:30:37,  1.65it/s]
 59%|█████▉    | 38695/65536 [6:38:34<4:37:58,  1.61it/s]
 59%|█████▉    | 38696/65536 [6:38:34<4:40:02,  1.60it/s]
 59%|█████▉    | 38697/65536 [6:38:35<4:37:36,  1.61it/s]
 59%|█████▉    | 38698/65536 [6:38:35<4:33:17,  1.64it/s]
 59%|█████▉    | 38699/65536 [6:38:36<4:32:34,  1.64it/s]
 59%|█████▉    | 38700/65536 [6:38:37<4:28:07,  1.67it/s]
                                                         
{'loss': 1.6243, 'learning_rate': 4.6966450348965346e-07, 'epoch': 2388.89}
+
 59%|█████▉    | 38700/65536 [6:38:37<4:28:07,  1.67it/s]
 59%|█████▉    | 38701/65536 [6:38:37<4:30:21,  1.65it/s]
 59%|█████▉    | 38702/65536 [6:38:38<4:42:10,  1.58it/s]
 59%|█████▉    | 38703/65536 [6:38:38<4:36:02,  1.62it/s]
 59%|█████▉    | 38704/65536 [6:38:39<4:36:23,  1.62it/s]
 59%|█████▉    | 38705/65536 [6:38:40<4:37:02,  1.61it/s]
 59%|█████▉    | 38706/65536 [6:38:40<4:34:38,  1.63it/s]
 59%|█████▉    | 38707/65536 [6:38:41<4:37:13,  1.61it/s]
 59%|█████▉    | 38708/65536 [6:38:42<4:37:24,  1.61it/s]
 59%|█████▉    | 38709/65536 [6:38:42<4:31:12,  1.65it/s]
 59%|█████▉    | 38710/65536 [6:38:43<4:27:24,  1.67it/s]
 59%|█████▉    | 38711/65536 [6:38:43<4:28:20,  1.67it/s]
 59%|█████▉    | 38712/65536 [6:38:44<4:29:22,  1.66it/s]
 59%|█████▉    | 38713/65536 [6:38:45<4:30:53,  1.65it/s]
 59%|█████▉    | 38714/65536 [6:38:45<4:34:59,  1.63it/s]
 59%|█████▉    | 38715/65536 [6:38:46<4:32:12,  1.64it/s]
 59%|█████▉    | 38716/65536 [6:38:46<4:34:01,  1.63it/s]
 59%|█████▉    | 38717/65536 [6:38:47<4:33:59,  1.63it/s]
 59%|█████▉    | 38718/65536 [6:38:48<4:33:52,  1.63it/s]
 59%|█████▉    | 38719/65536 [6:38:48<4:47:55,  1.55it/s]
 59%|█████▉    | 38720/65536 [6:38:49<4:41:41,  1.59it/s]
                                                         
{'loss': 1.6462, 'learning_rate': 4.693890045304273e-07, 'epoch': 2390.12}
+
 59%|█████▉    | 38720/65536 [6:38:49<4:41:41,  1.59it/s]
 59%|█████▉    | 38721/65536 [6:38:50<4:42:27,  1.58it/s]
 59%|█████▉    | 38722/65536 [6:38:50<4:38:26,  1.61it/s]
 59%|█████▉    | 38723/65536 [6:38:51<4:36:46,  1.61it/s]
 59%|█████▉    | 38724/65536 [6:38:51<4:34:48,  1.63it/s]
 59%|█████▉    | 38725/65536 [6:38:52<4:33:26,  1.63it/s]
 59%|█████▉    | 38726/65536 [6:38:53<4:38:19,  1.61it/s]
 59%|█████▉    | 38727/65536 [6:38:53<4:38:24,  1.60it/s]
 59%|█████▉    | 38728/65536 [6:38:54<4:36:05,  1.62it/s]
 59%|█████▉    | 38729/65536 [6:38:54<4:33:28,  1.63it/s]
 59%|█████▉    | 38730/65536 [6:38:55<4:28:59,  1.66it/s]
 59%|█████▉    | 38731/65536 [6:38:56<4:33:35,  1.63it/s]
 59%|█████▉    | 38732/65536 [6:38:56<4:38:45,  1.60it/s]
 59%|█████▉    | 38733/65536 [6:38:57<4:34:24,  1.63it/s]
 59%|█████▉    | 38734/65536 [6:38:58<4:41:47,  1.59it/s]
 59%|█████▉    | 38735/65536 [6:38:58<4:48:15,  1.55it/s]
 59%|█████▉    | 38736/65536 [6:38:59<4:46:58,  1.56it/s]
 59%|█████▉    | 38737/65536 [6:39:00<4:43:55,  1.57it/s]
 59%|█████▉    | 38738/65536 [6:39:00<4:35:20,  1.62it/s]
 59%|█████▉    | 38739/65536 [6:39:01<4:41:39,  1.59it/s]
 59%|█████▉    | 38740/65536 [6:39:01<4:36:21,  1.62it/s]
                                                         
{'loss': 1.608, 'learning_rate': 4.6911350557120117e-07, 'epoch': 2391.36}
+
 59%|█████▉    | 38740/65536 [6:39:01<4:36:21,  1.62it/s]
 59%|█████▉    | 38741/65536 [6:39:02<4:32:54,  1.64it/s]
 59%|█████▉    | 38742/65536 [6:39:03<4:35:52,  1.62it/s]
 59%|█████▉    | 38743/65536 [6:39:03<4:34:01,  1.63it/s]
 59%|█████▉    | 38744/65536 [6:39:04<4:28:42,  1.66it/s]
 59%|█████▉    | 38745/65536 [6:39:04<4:37:22,  1.61it/s]
 59%|█████▉    | 38746/65536 [6:39:05<4:35:47,  1.62it/s]
 59%|█████▉    | 38747/65536 [6:39:06<4:34:37,  1.63it/s]
 59%|█████▉    | 38748/65536 [6:39:06<4:31:55,  1.64it/s]
 59%|█████▉    | 38749/65536 [6:39:07<4:33:47,  1.63it/s]
 59%|█████▉    | 38750/65536 [6:39:07<4:28:24,  1.66it/s]
 59%|█████▉    | 38751/65536 [6:39:08<4:39:54,  1.59it/s]
 59%|█████▉    | 38752/65536 [6:39:09<4:37:33,  1.61it/s]
 59%|█████▉    | 38753/65536 [6:39:09<4:31:43,  1.64it/s]
 59%|█████▉    | 38754/65536 [6:39:10<4:34:15,  1.63it/s]
 59%|█████▉    | 38755/65536 [6:39:11<4:31:27,  1.64it/s]
 59%|█████▉    | 38756/65536 [6:39:11<4:30:09,  1.65it/s]
 59%|█████▉    | 38757/65536 [6:39:12<4:26:24,  1.68it/s]
 59%|█████▉    | 38758/65536 [6:39:12<4:26:04,  1.68it/s]
 59%|█████▉    | 38759/65536 [6:39:13<4:30:16,  1.65it/s]
 59%|█████▉    | 38760/65536 [6:39:14<4:28:13,  1.66it/s]
                                                         
{'loss': 1.6236, 'learning_rate': 4.6883800661197503e-07, 'epoch': 2392.59}
+
 59%|█████▉    | 38760/65536 [6:39:14<4:28:13,  1.66it/s]
 59%|█████▉    | 38761/65536 [6:39:14<4:34:33,  1.63it/s]
 59%|█████▉    | 38762/65536 [6:39:15<4:32:33,  1.64it/s]
 59%|█████▉    | 38763/65536 [6:39:15<4:33:24,  1.63it/s]
 59%|█████▉    | 38764/65536 [6:39:16<4:30:38,  1.65it/s]
 59%|█████▉    | 38765/65536 [6:39:17<4:27:20,  1.67it/s]
 59%|█████▉    | 38766/65536 [6:39:17<4:28:45,  1.66it/s]
 59%|█████▉    | 38767/65536 [6:39:18<4:38:13,  1.60it/s]
 59%|█████▉    | 38768/65536 [6:39:18<4:41:33,  1.58it/s]
 59%|███���█▉    | 38769/65536 [6:39:19<4:37:00,  1.61it/s]
 59%|█████▉    | 38770/65536 [6:39:20<4:34:34,  1.62it/s]
 59%|█████▉    | 38771/65536 [6:39:20<4:32:59,  1.63it/s]
 59%|█████▉    | 38772/65536 [6:39:21<4:35:13,  1.62it/s]
 59%|█████▉    | 38773/65536 [6:39:22<4:34:23,  1.63it/s]
 59%|█████▉    | 38774/65536 [6:39:22<4:38:56,  1.60it/s]
 59%|█████▉    | 38775/65536 [6:39:23<4:40:01,  1.59it/s]
 59%|█████▉    | 38776/65536 [6:39:23<4:34:50,  1.62it/s]
 59%|█████▉    | 38777/65536 [6:39:24<4:33:48,  1.63it/s]
 59%|█████▉    | 38778/65536 [6:39:25<4:37:52,  1.60it/s]
 59%|█████▉    | 38779/65536 [6:39:25<4:34:23,  1.63it/s]
 59%|█████▉    | 38780/65536 [6:39:26<4:31:50,  1.64it/s]
                                                         
{'loss': 1.639, 'learning_rate': 4.685625076527489e-07, 'epoch': 2393.83}
+
 59%|█████▉    | 38780/65536 [6:39:26<4:31:50,  1.64it/s]
 59%|█████▉    | 38781/65536 [6:39:26<4:32:10,  1.64it/s]
 59%|█████▉    | 38782/65536 [6:39:27<4:40:40,  1.59it/s]
 59%|█████▉    | 38783/65536 [6:39:28<4:48:06,  1.55it/s]
 59%|█████▉    | 38784/65536 [6:39:28<4:44:09,  1.57it/s]
 59%|█████▉    | 38785/65536 [6:39:29<4:41:04,  1.59it/s]
 59%|█████▉    | 38786/65536 [6:39:30<4:36:57,  1.61it/s]
 59%|█████▉    | 38787/65536 [6:39:30<4:32:29,  1.64it/s]
 59%|█████▉    | 38788/65536 [6:39:31<4:33:20,  1.63it/s]
 59%|█████▉    | 38789/65536 [6:39:31<4:30:09,  1.65it/s]
 59%|█████▉    | 38790/65536 [6:39:32<4:29:56,  1.65it/s]
 59%|█████▉    | 38791/65536 [6:39:33<4:31:09,  1.64it/s]
 59%|█████▉    | 38792/65536 [6:39:33<4:27:59,  1.66it/s]
 59%|█████▉    | 38793/65536 [6:39:34<4:42:20,  1.58it/s]
 59%|█████▉    | 38794/65536 [6:39:35<4:38:03,  1.60it/s]
 59%|█████▉    | 38795/65536 [6:39:35<4:32:32,  1.64it/s]
 59%|█████▉    | 38796/65536 [6:39:36<4:30:44,  1.65it/s]
 59%|█████▉    | 38797/65536 [6:39:36<4:34:03,  1.63it/s]
 59%|█████▉    | 38798/65536 [6:39:37<4:31:51,  1.64it/s]
 59%|█████▉    | 38799/65536 [6:39:38<4:25:42,  1.68it/s]
 59%|█████▉    | 38800/65536 [6:39:38<4:43:50,  1.57it/s]
                                                         
{'loss': 1.6154, 'learning_rate': 4.682870086935227e-07, 'epoch': 2395.06}
+
 59%|█████▉    | 38800/65536 [6:39:38<4:43:50,  1.57it/s]
 59%|█████▉    | 38801/65536 [6:39:39<4:34:08,  1.63it/s]
 59%|█████▉    | 38802/65536 [6:39:39<4:31:58,  1.64it/s]
 59%|█████▉    | 38803/65536 [6:39:40<4:39:09,  1.60it/s]
 59%|█████▉    | 38804/65536 [6:39:41<4:36:10,  1.61it/s]
 59%|█████▉    | 38805/65536 [6:39:41<4:33:14,  1.63it/s]
 59%|█████▉    | 38806/65536 [6:39:42<4:32:15,  1.64it/s]
 59%|█████▉    | 38807/65536 [6:39:43<4:30:32,  1.65it/s]
 59%|█████▉    | 38808/65536 [6:39:43<4:35:58,  1.61it/s]
 59%|█████▉    | 38809/65536 [6:39:44<4:55:14,  1.51it/s]
 59%|█████▉    | 38810/65536 [6:39:45<4:49:46,  1.54it/s]
 59%|█████▉    | 38811/65536 [6:39:45<4:41:03,  1.58it/s]
 59%|█████▉    | 38812/65536 [6:39:46<4:35:26,  1.62it/s]
 59%|█████▉    | 38813/65536 [6:39:46<4:34:59,  1.62it/s]
 59%|█████▉    | 38814/65536 [6:39:47<4:33:57,  1.63it/s]
 59%|█████▉    | 38815/65536 [6:39:48<4:32:24,  1.63it/s]
 59%|█████▉    | 38816/65536 [6:39:48<4:39:28,  1.59it/s]
 59%|█████▉    | 38817/65536 [6:39:49<4:43:03,  1.57it/s]
 59%|█████▉    | 38818/65536 [6:39:49<4:34:53,  1.62it/s]
 59%|█████▉    | 38819/65536 [6:39:50<4:31:22,  1.64it/s]
 59%|█████▉    | 38820/65536 [6:39:51<4:32:49,  1.63it/s]
                                                         
{'loss': 1.6044, 'learning_rate': 4.680115097342965e-07, 'epoch': 2396.3}
+
 59%|█████▉    | 38820/65536 [6:39:51<4:32:49,  1.63it/s]
 59%|█████▉    | 38821/65536 [6:39:51<4:31:09,  1.64it/s]
 59%|█████▉    | 38822/65536 [6:39:52<4:27:42,  1.66it/s]
 59%|█████▉    | 38823/65536 [6:39:52<4:21:33,  1.70it/s]
 59%|█████▉    | 38824/65536 [6:39:53<4:25:37,  1.68it/s]
 59%|█████▉    | 38825/65536 [6:39:54<4:29:01,  1.65it/s]
 59%|█████▉    | 38826/65536 [6:39:54<4:37:54,  1.60it/s]
 59%|█████▉    | 38827/65536 [6:39:55<4:49:31,  1.54it/s]
 59%|█████▉    | 38828/65536 [6:39:56<4:49:14,  1.54it/s]
 59%|█████▉    | 38829/65536 [6:39:56<4:40:21,  1.59it/s]
 59%|█████▉    | 38830/65536 [6:39:57<4:37:38,  1.60it/s]
 59%|█████▉    | 38831/65536 [6:39:57<4:28:13,  1.66it/s]
 59%|█████▉    | 38832/65536 [6:39:58<4:37:12,  1.61it/s]
 59%|█████▉    | 38833/65536 [6:39:59<4:34:01,  1.62it/s]
 59%|█████▉    | 38834/65536 [6:39:59<4:36:25,  1.61it/s]
 59%|█████▉    | 38835/65536 [6:40:00<4:37:13,  1.61it/s]
 59%|█████▉    | 38836/65536 [6:40:01<4:39:32,  1.59it/s]
 59%|█████▉    | 38837/65536 [6:40:01<4:33:32,  1.63it/s]
 59%|█████▉    | 38838/65536 [6:40:02<4:33:16,  1.63it/s]
 59%|█████▉    | 38839/65536 [6:40:02<4:28:28,  1.66it/s]
 59%|█████▉    | 38840/65536 [6:40:03<4:31:21,  1.64it/s]
                                                         
{'loss': 1.6302, 'learning_rate': 4.6773601077507036e-07, 'epoch': 2397.53}
+
 59%|█████▉    | 38840/65536 [6:40:03<4:31:21,  1.64it/s]
 59%|█████▉    | 38841/65536 [6:40:04<4:39:26,  1.59it/s]
 59%|█████▉    | 38842/65536 [6:40:04<4:36:43,  1.61it/s]
 59%|█████▉    | 38843/65536 [6:40:05<4:34:33,  1.62it/s]
 59%|█████▉    | 38844/65536 [6:40:05<4:33:33,  1.63it/s]
 59%|█████▉    | 38845/65536 [6:40:06<4:29:26,  1.65it/s]
 59%|█████▉    | 38846/65536 [6:40:07<4:29:33,  1.65it/s]
 59%|█████▉    | 38847/65536 [6:40:07<4:35:33,  1.61it/s]
 59%|█████▉    | 38848/65536 [6:40:08<4:41:58,  1.58it/s]
 59%|█████▉    | 38849/65536 [6:40:09<4:38:22,  1.60it/s]
 59%|█████▉    | 38850/65536 [6:40:09<4:35:57,  1.61it/s]
 59%|█████▉    | 38851/65536 [6:40:10<4:40:21,  1.59it/s]
 59%|█████▉    | 38852/65536 [6:40:10<4:40:18,  1.59it/s]
 59%|█████▉    | 38853/65536 [6:40:11<4:37:40,  1.60it/s]
 59%|█████▉    | 38854/65536 [6:40:12<4:35:49,  1.61it/s]
 59%|█████▉    | 38855/65536 [6:40:12<4:39:18,  1.59it/s]
 59%|█████▉    | 38856/65536 [6:40:13<4:30:21,  1.64it/s]
 59%|█████▉    | 38857/65536 [6:40:14<4:30:23,  1.64it/s]
 59%|█████▉    | 38858/65536 [6:40:14<4:27:09,  1.66it/s]
 59%|█████▉    | 38859/65536 [6:40:15<4:28:34,  1.66it/s]
 59%|█████▉    | 38860/65536 [6:40:15<4:30:50,  1.64it/s]
                                                         
{'loss': 1.6219, 'learning_rate': 4.674605118158442e-07, 'epoch': 2398.77}
+
 59%|█████▉    | 38860/65536 [6:40:15<4:30:50,  1.64it/s]
 59%|█████▉    | 38861/65536 [6:40:16<4:32:33,  1.63it/s]
 59%|█████▉    | 38862/65536 [6:40:17<4:27:16,  1.66it/s]
 59%|█████▉    | 38863/65536 [6:40:17<4:34:38,  1.62it/s]
 59%|█████▉    | 38864/65536 [6:40:18<4:39:39,  1.59it/s]
 59%|█████▉    | 38865/65536 [6:40:18<4:34:44,  1.62it/s]
 59%|█████▉    | 38866/65536 [6:40:19<4:39:00,  1.59it/s]
 59%|█████▉    | 38867/65536 [6:40:20<4:36:01,  1.61it/s]
 59%|█████▉    | 38868/65536 [6:40:20<4:41:11,  1.58it/s]
 59%|█████▉    | 38869/65536 [6:40:21<4:43:53,  1.57it/s]
 59%|█████▉    | 38870/65536 [6:40:22<4:38:14,  1.60it/s]
 59%|█████▉    | 38871/65536 [6:40:22<4:34:57,  1.62it/s]
 59%|█████▉    | 38872/65536 [6:40:23<4:32:00,  1.63it/s]
 59%|█████▉    | 38873/65536 [6:40:23<4:31:27,  1.64it/s]
 59%|█████▉    | 38874/65536 [6:40:24<4:29:19,  1.65it/s]
 59%|█████▉    | 38875/65536 [6:40:25<4:23:21,  1.69it/s]
 59%|█████▉    | 38876/65536 [6:40:25<4:23:18,  1.69it/s]
 59%|█████▉    | 38877/65536 [6:40:26<4:33:47,  1.62it/s]
 59%|█████▉    | 38878/65536 [6:40:26<4:37:18,  1.60it/s]
 59%|█████▉    | 38879/65536 [6:40:27<4:36:47,  1.61it/s]
 59%|█████▉    | 38880/65536 [6:40:28<4:37:20,  1.60it/s]
                                                         
{'loss': 1.6178, 'learning_rate': 4.671850128566181e-07, 'epoch': 2400.0}
+
 59%|█████▉    | 38880/65536 [6:40:28<4:37:20,  1.60it/s]
 59%|█████▉    | 38881/65536 [6:40:28<4:42:40,  1.57it/s]
 59%|█████▉    | 38882/65536 [6:40:29<4:40:07,  1.59it/s]
 59%|█████▉    | 38883/65536 [6:40:30<4:37:50,  1.60it/s]
 59%|█████▉    | 38884/65536 [6:40:30<4:31:38,  1.64it/s]
 59%|█████▉    | 38885/65536 [6:40:31<4:31:33,  1.64it/s]
 59%|█████▉    | 38886/65536 [6:40:31<4:23:32,  1.69it/s]
 59%|█████▉    | 38887/65536 [6:40:32<4:28:52,  1.65it/s]
 59%|█████▉    | 38888/65536 [6:40:33<4:31:58,  1.63it/s]
 59%|█████▉    | 38889/65536 [6:40:33<4:29:15,  1.65it/s]
 59%|█████▉    | 38890/65536 [6:40:34<4:34:23,  1.62it/s]
 59%|█████▉    | 38891/65536 [6:40:34<4:34:45,  1.62it/s]
 59%|█████▉    | 38892/65536 [6:40:35<4:37:22,  1.60it/s]
 59%|█████▉    | 38893/65536 [6:40:36<4:40:00,  1.59it/s]
 59%|█████▉    | 38894/65536 [6:40:36<4:38:33,  1.59it/s]
 59%|█████▉    | 38895/65536 [6:40:37<4:36:29,  1.61it/s]
 59%|█████▉    | 38896/65536 [6:40:38<4:32:04,  1.63it/s]
 59%|█████▉    | 38897/65536 [6:40:38<4:43:08,  1.57it/s]
 59%|█████▉    | 38898/65536 [6:40:39<4:41:37,  1.58it/s]
 59%|█████▉    | 38899/65536 [6:40:39<4:36:12,  1.61it/s]
 59%|█████▉    | 38900/65536 [6:40:40<4:45:44,  1.55it/s]
                                                         
{'loss': 1.6197, 'learning_rate': 4.6690951389739194e-07, 'epoch': 2401.23}
+
 59%|█████▉    | 38900/65536 [6:40:40<4:45:44,  1.55it/s]
 59%|█████▉    | 38901/65536 [6:40:41<4:43:25,  1.57it/s]
 59%|█████▉    | 38902/65536 [6:40:41<4:44:28,  1.56it/s]
 59%|█████▉    | 38903/65536 [6:40:42<4:45:55,  1.55it/s]
 59%|█████▉    | 38904/65536 [6:40:43<4:39:40,  1.59it/s]
 59%|█████▉    | 38905/65536 [6:40:43<4:39:31,  1.59it/s]
 59%|█████▉    | 38906/65536 [6:40:44<4:37:34,  1.60it/s]
 59%|█████▉    | 38907/65536 [6:40:45<4:31:56,  1.63it/s]
 59%|█████▉    | 38908/65536 [6:40:45<4:33:27,  1.62it/s]
 59%|█████▉    | 38909/65536 [6:40:46<4:32:25,  1.63it/s]
 59%|█████▉    | 38910/65536 [6:40:46<4:30:33,  1.64it/s]
 59%|█████▉    | 38911/65536 [6:40:47<4:29:14,  1.65it/s]
 59%|█████▉    | 38912/65536 [6:40:48<4:21:13,  1.70it/s]
 59%|█████▉    | 38913/65536 [6:40:48<4:31:57,  1.63it/s]
 59%|█████▉    | 38914/65536 [6:40:49<4:34:26,  1.62it/s]
 59%|█████▉    | 38915/65536 [6:40:49<4:34:46,  1.61it/s]
 59%|█████▉    | 38916/65536 [6:40:50<4:35:13,  1.61it/s]
 59%|█████▉    | 38917/65536 [6:40:51<4:34:41,  1.62it/s]
 59%|█████▉    | 38918/65536 [6:40:51<4:47:13,  1.54it/s]
 59%|█████▉    | 38919/65536 [6:40:52<4:43:24,  1.57it/s]
 59%|█████▉    | 38920/65536 [6:40:53<4:35:35,  1.61it/s]
                                                         
{'loss': 1.6042, 'learning_rate': 4.666340149381658e-07, 'epoch': 2402.47}
+
 59%|█████▉    | 38920/65536 [6:40:53<4:35:35,  1.61it/s]
 59%|█████▉    | 38921/65536 [6:40:53<4:32:07,  1.63it/s]
 59%|█████▉    | 38922/65536 [6:40:54<4:28:14,  1.65it/s]
 59%|█████▉    | 38923/65536 [6:40:54<4:26:48,  1.66it/s]
 59%|█████▉    | 38924/65536 [6:40:55<4:26:06,  1.67it/s]
 59%|█████▉    | 38925/65536 [6:40:56<4:27:19,  1.66it/s]
 59%|█████▉    | 38926/65536 [6:40:56<4:30:44,  1.64it/s]
 59%|█████▉    | 38927/65536 [6:40:57<4:32:59,  1.62it/s]
 59%|█████▉    | 38928/65536 [6:40:57<4:39:16,  1.59it/s]
 59%|█████▉    | 38929/65536 [6:40:58<4:46:02,  1.55it/s]
 59%|█████▉    | 38930/65536 [6:40:59<4:41:13,  1.58it/s]
 59%|█████▉    | 38931/65536 [6:40:59<4:37:47,  1.60it/s]
 59%|█████▉    | 38932/65536 [6:41:00<4:31:28,  1.63it/s]
 59%|█████▉    | 38933/65536 [6:41:01<4:32:06,  1.63it/s]
 59%|█████▉    | 38934/65536 [6:41:01<4:27:58,  1.65it/s]
 59%|█████▉    | 38935/65536 [6:41:02<4:29:29,  1.65it/s]
 59%|█████▉    | 38936/65536 [6:41:02<4:40:27,  1.58it/s]
 59%|█████▉    | 38937/65536 [6:41:03<4:43:49,  1.56it/s]
 59%|█████▉    | 38938/65536 [6:41:04<4:42:17,  1.57it/s]
 59%|█████▉    | 38939/65536 [6:41:04<4:43:48,  1.56it/s]
 59%|█████▉    | 38940/65536 [6:41:05<4:36:01,  1.61it/s]
                                                         
{'loss': 1.591, 'learning_rate': 4.6635851597893966e-07, 'epoch': 2403.7}
+
 59%|█████▉    | 38940/65536 [6:41:05<4:36:01,  1.61it/s]
 59%|█████▉    | 38941/65536 [6:41:06<4:36:01,  1.61it/s]
 59%|█████▉    | 38942/65536 [6:41:06<4:28:54,  1.65it/s]
 59%|█████▉    | 38943/65536 [6:41:07<4:28:51,  1.65it/s]
 59%|█████▉    | 38944/65536 [6:41:07<4:26:58,  1.66it/s]
 59%|█████▉    | 38945/65536 [6:41:08<4:36:54,  1.60it/s]
 59%|█████▉    | 38946/65536 [6:41:09<4:38:52,  1.59it/s]
 59%|█████▉    | 38947/65536 [6:41:09<4:43:57,  1.56it/s]
 59%|█████▉    | 38948/65536 [6:41:10<4:40:32,  1.58it/s]
 59%|█████▉    | 38949/65536 [6:41:11<4:34:47,  1.61it/s]
 59%|█████▉    | 38950/65536 [6:41:11<4:40:20,  1.58it/s]
 59%|█████▉    | 38951/65536 [6:41:12<4:37:52,  1.59it/s]
 59%|█████▉    | 38952/65536 [6:41:12<4:39:27,  1.59it/s]
 59%|█████▉    | 38953/65536 [6:41:13<4:38:41,  1.59it/s]
 59%|█████▉    | 38954/65536 [6:41:14<4:32:13,  1.63it/s]
 59%|█████▉    | 38955/65536 [6:41:14<4:23:56,  1.68it/s]
 59%|█████▉    | 38956/65536 [6:41:15<4:26:30,  1.66it/s]
 59%|█████▉    | 38957/65536 [6:41:15<4:30:19,  1.64it/s]
 59%|█████▉    | 38958/65536 [6:41:16<4:30:47,  1.64it/s]
 59%|█████▉    | 38959/65536 [6:41:17<4:29:14,  1.65it/s]
 59%|█████▉    | 38960/65536 [6:41:17<4:28:45,  1.65it/s]
                                                         
{'loss': 1.6311, 'learning_rate': 4.660830170197134e-07, 'epoch': 2404.94}
+
 59%|█████▉    | 38960/65536 [6:41:17<4:28:45,  1.65it/s]
 59%|█████▉    | 38961/65536 [6:41:18<4:28:58,  1.65it/s]
 59%|█████▉    | 38962/65536 [6:41:19<4:46:06,  1.55it/s]
 59%|█████▉    | 38963/65536 [6:41:19<4:44:52,  1.55it/s]
 59%|█████▉    | 38964/65536 [6:41:20<4:37:05,  1.60it/s]
 59%|█████▉    | 38965/65536 [6:41:20<4:35:23,  1.61it/s]
 59%|█████▉    | 38966/65536 [6:41:21<4:31:19,  1.63it/s]
 59%|█████▉    | 38967/65536 [6:41:22<4:30:34,  1.64it/s]
 59%|█████▉    | 38968/65536 [6:41:22<4:29:30,  1.64it/s]
 59%|█████▉    | 38969/65536 [6:41:23<4:28:01,  1.65it/s]
 59%|█████▉    | 38970/65536 [6:41:24<4:34:11,  1.61it/s]
 59%|█████▉    | 38971/65536 [6:41:24<4:33:22,  1.62it/s]
 59%|█████▉    | 38972/65536 [6:41:25<4:38:49,  1.59it/s]
 59%|█████▉    | 38973/65536 [6:41:25<4:36:03,  1.60it/s]
 59%|█████▉    | 38974/65536 [6:41:26<4:35:21,  1.61it/s]
 59%|█████▉    | 38975/65536 [6:41:27<4:33:54,  1.62it/s]
 59%|█████▉    | 38976/65536 [6:41:27<4:28:59,  1.65it/s]
 59%|█████▉    | 38977/65536 [6:41:28<4:26:51,  1.66it/s]
 59%|█████▉    | 38978/65536 [6:41:29<4:41:12,  1.57it/s]
 59%|█████▉    | 38979/65536 [6:41:29<4:37:27,  1.60it/s]
 59%|█████▉    | 38980/65536 [6:41:30<4:39:59,  1.58it/s]
                                                         
{'loss': 1.644, 'learning_rate': 4.6580751806048727e-07, 'epoch': 2406.17}
+
 59%|█████▉    | 38980/65536 [6:41:30<4:39:59,  1.58it/s]
 59%|█████▉    | 38981/65536 [6:41:30<4:34:49,  1.61it/s]
 59%|█████▉    | 38982/65536 [6:41:31<4:29:30,  1.64it/s]
 59%|█████▉    | 38983/65536 [6:41:32<4:27:19,  1.66it/s]
 59%|█████▉    | 38984/65536 [6:41:32<4:29:58,  1.64it/s]
 59%|█████▉    | 38985/65536 [6:41:33<4:26:40,  1.66it/s]
 59%|█████▉    | 38986/65536 [6:41:33<4:28:06,  1.65it/s]
 59%|█████▉    | 38987/65536 [6:41:34<4:24:28,  1.67it/s]
 59%|█████▉    | 38988/65536 [6:41:35<4:20:02,  1.70it/s]
 59%|█████▉    | 38989/65536 [6:41:35<4:30:17,  1.64it/s]
 59%|█████▉    | 38990/65536 [6:41:36<4:26:51,  1.66it/s]
 59%|█████▉    | 38991/65536 [6:41:36<4:29:52,  1.64it/s]
 59%|█████▉    | 38992/65536 [6:41:37<4:29:19,  1.64it/s]
 59%|█████▉    | 38993/65536 [6:41:38<4:42:21,  1.57it/s]
 60%|█████▉    | 38994/65536 [6:41:38<4:43:58,  1.56it/s]
 60%|█████▉    | 38995/65536 [6:41:39<4:41:26,  1.57it/s]
 60%|█████▉    | 38996/65536 [6:41:40<4:38:14,  1.59it/s]
 60%|█████▉    | 38997/65536 [6:41:40<4:40:30,  1.58it/s]
 60%|█████▉    | 38998/65536 [6:41:41<4:34:34,  1.61it/s]
 60%|█████▉    | 38999/65536 [6:41:41<4:26:51,  1.66it/s]
 60%|█████▉    | 39000/65536 [6:41:42<4:24:46,  1.67it/s]
                                                         
{'loss': 1.6399, 'learning_rate': 4.6553201910126113e-07, 'epoch': 2407.41}
+
 60%|█████▉    | 39000/65536 [6:41:42<4:24:46,  1.67it/s]
 60%|█████▉    | 39001/65536 [6:41:43<4:24:30,  1.67it/s]
 60%|█████▉    | 39002/65536 [6:41:43<4:30:32,  1.63it/s]
 60%|█████▉    | 39003/65536 [6:41:44<4:29:55,  1.64it/s]
 60%|█████▉    | 39004/65536 [6:41:44<4:24:26,  1.67it/s]
 60%|█████▉    | 39005/65536 [6:41:45<4:26:04,  1.66it/s]
 60%|█████▉    | 39006/65536 [6:41:46<4:28:26,  1.65it/s]
 60%|█████▉    | 39007/65536 [6:41:46<4:29:35,  1.64it/s]
 60%|█████▉    | 39008/65536 [6:41:47<4:34:20,  1.61it/s]
 60%|█████▉    | 39009/65536 [6:41:47<4:31:30,  1.63it/s]
 60%|█████▉    | 39010/65536 [6:41:48<4:48:43,  1.53it/s]
 60%|█████▉    | 39011/65536 [6:41:49<4:40:50,  1.57it/s]
 60%|█████▉    | 39012/65536 [6:41:49<4:30:42,  1.63it/s]
 60%|█████▉    | 39013/65536 [6:41:50<4:31:43,  1.63it/s]
 60%|█████▉    | 39014/65536 [6:41:51<4:36:47,  1.60it/s]
 60%|█████▉    | 39015/65536 [6:41:51<4:33:48,  1.61it/s]
 60%|█████▉    | 39016/65536 [6:41:52<4:29:57,  1.64it/s]
 60%|█████▉    | 39017/65536 [6:41:52<4:25:15,  1.67it/s]
 60%|█████▉    | 39018/65536 [6:41:53<4:27:28,  1.65it/s]
 60%|█████▉    | 39019/65536 [6:41:54<4:33:01,  1.62it/s]
 60%|█████▉    | 39020/65536 [6:41:54<4:25:29,  1.66it/s]
                                                         
{'loss': 1.6386, 'learning_rate': 4.65256520142035e-07, 'epoch': 2408.64}
+
 60%|█████▉    | 39020/65536 [6:41:54<4:25:29,  1.66it/s]
 60%|█████▉    | 39021/65536 [6:41:55<4:22:53,  1.68it/s]
 60%|█████▉    | 39022/65536 [6:41:55<4:24:39,  1.67it/s]
 60%|█████▉    | 39023/65536 [6:41:56<4:30:57,  1.63it/s]
 60%|█████▉    | 39024/65536 [6:41:57<4:30:36,  1.63it/s]
 60%|█████▉    | 39025/65536 [6:41:57<4:24:46,  1.67it/s]
 60%|█████▉    | 39026/65536 [6:41:58<4:40:35,  1.57it/s]
 60%|█████▉    | 39027/65536 [6:41:59<4:35:59,  1.60it/s]
 60%|█████▉    | 39028/65536 [6:41:59<4:35:10,  1.61it/s]
 60%|█████▉    | 39029/65536 [6:42:00<4:36:55,  1.60it/s]
 60%|█████▉    | 39030/65536 [6:42:00<4:39:25,  1.58it/s]
 60%|█████▉    | 39031/65536 [6:42:01<4:33:29,  1.62it/s]
 60%|█████▉    | 39032/65536 [6:42:02<4:24:33,  1.67it/s]
 60%|█████▉    | 39033/65536 [6:42:02<4:19:37,  1.70it/s]
 60%|█████▉    | 39034/65536 [6:42:03<4:19:49,  1.70it/s]
 60%|█████▉    | 39035/65536 [6:42:03<4:17:51,  1.71it/s]
 60%|█████▉    | 39036/65536 [6:42:04<4:21:30,  1.69it/s]
 60%|█████▉    | 39037/65536 [6:42:05<4:29:05,  1.64it/s]
 60%|█████▉    | 39038/65536 [6:42:05<4:29:39,  1.64it/s]
 60%|█████▉    | 39039/65536 [6:42:06<4:39:08,  1.58it/s]
 60%|█████▉    | 39040/65536 [6:42:07<4:36:00,  1.60it/s]
                                                         
{'loss': 1.6388, 'learning_rate': 4.6498102118280884e-07, 'epoch': 2409.88}
+
 60%|█████▉    | 39040/65536 [6:42:07<4:36:00,  1.60it/s]
 60%|█████▉    | 39041/65536 [6:42:07<4:33:28,  1.61it/s]
 60%|█████▉    | 39042/65536 [6:42:08<4:34:57,  1.61it/s]
 60%|█████▉    | 39043/65536 [6:42:08<4:44:13,  1.55it/s]
 60%|█████▉    | 39044/65536 [6:42:09<4:44:19,  1.55it/s]
 60%|█████▉    | 39045/65536 [6:42:10<4:43:40,  1.56it/s]
 60%|█████▉    | 39046/65536 [6:42:10<4:37:47,  1.59it/s]
 60%|█████▉    | 39047/65536 [6:42:11<4:33:30,  1.61it/s]
 60%|█████▉    | 39048/65536 [6:42:12<4:29:41,  1.64it/s]
 60%|█████▉    | 39049/65536 [6:42:12<4:29:05,  1.64it/s]
 60%|█████▉    | 39050/65536 [6:42:13<4:25:11,  1.66it/s]
 60%|█████▉    | 39051/65536 [6:42:13<4:26:12,  1.66it/s]
 60%|█████▉    | 39052/65536 [6:42:14<4:27:36,  1.65it/s]
 60%|█████▉    | 39053/65536 [6:42:15<4:26:02,  1.66it/s]
 60%|█████▉    | 39054/65536 [6:42:15<4:23:39,  1.67it/s]
 60%|█████▉    | 39055/65536 [6:42:16<4:28:38,  1.64it/s]
 60%|█████▉    | 39056/65536 [6:42:16<4:26:56,  1.65it/s]
 60%|█████▉    | 39057/65536 [6:42:17<4:32:16,  1.62it/s]
 60%|█████▉    | 39058/65536 [6:42:18<4:33:20,  1.61it/s]
 60%|█████▉    | 39059/65536 [6:42:18<4:47:58,  1.53it/s]
 60%|█████▉    | 39060/65536 [6:42:19<4:41:51,  1.57it/s]
                                                         
{'loss': 1.6258, 'learning_rate': 4.647055222235827e-07, 'epoch': 2411.11}
+
 60%|█████▉    | 39060/65536 [6:42:19<4:41:51,  1.57it/s]
 60%|█████▉    | 39061/65536 [6:42:20<4:39:49,  1.58it/s]
 60%|█████▉    | 39062/65536 [6:42:20<4:37:36,  1.59it/s]
 60%|█████▉    | 39063/65536 [6:42:21<4:31:38,  1.62it/s]
 60%|█████▉    | 39064/65536 [6:42:21<4:32:05,  1.62it/s]
 60%|█████▉    | 39065/65536 [6:42:22<4:23:47,  1.67it/s]
 60%|█████▉    | 39066/65536 [6:42:23<4:25:36,  1.66it/s]
 60%|█████▉    | 39067/65536 [6:42:23<4:34:16,  1.61it/s]
 60%|█████▉    | 39068/65536 [6:42:24<4:29:09,  1.64it/s]
 60%|█████▉    | 39069/65536 [6:42:24<4:30:38,  1.63it/s]
 60%|█████▉    | 39070/65536 [6:42:25<4:24:10,  1.67it/s]
 60%|█████▉    | 39071/65536 [6:42:26<4:33:55,  1.61it/s]
 60%|█████▉    | 39072/65536 [6:42:26<4:31:38,  1.62it/s]
 60%|█████▉    | 39073/65536 [6:42:27<4:27:43,  1.65it/s]
 60%|█████▉    | 39074/65536 [6:42:27<4:32:25,  1.62it/s]
 60%|█████▉    | 39075/65536 [6:42:28<4:34:05,  1.61it/s]
 60%|█████▉    | 39076/65536 [6:42:29<4:31:14,  1.63it/s]
 60%|█████▉    | 39077/65536 [6:42:29<4:37:10,  1.59it/s]
 60%|█████▉    | 39078/65536 [6:42:30<4:44:55,  1.55it/s]
 60%|█████▉    | 39079/65536 [6:42:31<4:47:25,  1.53it/s]
 60%|█████▉    | 39080/65536 [6:42:31<4:43:59,  1.55it/s]
                                                         
{'loss': 1.6054, 'learning_rate': 4.6443002326435656e-07, 'epoch': 2412.35}
+
 60%|█████▉    | 39080/65536 [6:42:31<4:43:59,  1.55it/s]
 60%|█████▉    | 39081/65536 [6:42:32<4:45:11,  1.55it/s]
 60%|█████▉    | 39082/65536 [6:42:33<4:39:17,  1.58it/s]
 60%|█████▉    | 39083/65536 [6:42:33<4:31:42,  1.62it/s]
 60%|█████▉    | 39084/65536 [6:42:34<4:24:43,  1.67it/s]
 60%|█████▉    | 39085/65536 [6:42:34<4:27:26,  1.65it/s]
 60%|█████▉    | 39086/65536 [6:42:35<4:31:42,  1.62it/s]
 60%|█████▉    | 39087/65536 [6:42:36<4:31:09,  1.63it/s]
 60%|█████▉    | 39088/65536 [6:42:36<4:27:47,  1.65it/s]
 60%|█████▉    | 39089/65536 [6:42:37<4:21:34,  1.69it/s]
 60%|█████▉    | 39090/65536 [6:42:37<4:16:40,  1.72it/s]
 60%|█████▉    | 39091/65536 [6:42:38<4:24:39,  1.67it/s]
 60%|█████▉    | 39092/65536 [6:42:39<4:23:19,  1.67it/s]
 60%|█████▉    | 39093/65536 [6:42:39<4:23:55,  1.67it/s]
 60%|█████▉    | 39094/65536 [6:42:40<4:22:58,  1.68it/s]
 60%|█████▉    | 39095/65536 [6:42:40<4:34:30,  1.61it/s]
 60%|█████▉    | 39096/65536 [6:42:41<4:28:13,  1.64it/s]
 60%|█████▉    | 39097/65536 [6:42:42<4:18:32,  1.70it/s]
 60%|█████▉    | 39098/65536 [6:42:42<4:38:09,  1.58it/s]
 60%|█████▉    | 39099/65536 [6:42:43<4:31:01,  1.63it/s]
 60%|█████▉    | 39100/65536 [6:42:43<4:31:45,  1.62it/s]
                                                         
{'loss': 1.666, 'learning_rate': 4.6415452430513037e-07, 'epoch': 2413.58}
+
 60%|█████▉    | 39100/65536 [6:42:43<4:31:45,  1.62it/s]
 60%|█████▉    | 39101/65536 [6:42:44<4:28:27,  1.64it/s]
 60%|█████▉    | 39102/65536 [6:42:45<4:22:54,  1.68it/s]
 60%|█████▉    | 39103/65536 [6:42:45<4:25:44,  1.66it/s]
 60%|█████▉    | 39104/65536 [6:42:46<4:29:16,  1.64it/s]
 60%|█████▉    | 39105/65536 [6:42:46<4:25:52,  1.66it/s]
 60%|█████▉    | 39106/65536 [6:42:47<4:24:02,  1.67it/s]
 60%|█████▉    | 39107/65536 [6:42:48<4:34:16,  1.61it/s]
 60%|█████▉    | 39108/65536 [6:42:48<4:33:28,  1.61it/s]
 60%|█████▉    | 39109/65536 [6:42:49<4:33:15,  1.61it/s]
 60%|█████▉    | 39110/65536 [6:42:50<4:39:46,  1.57it/s]
 60%|█████▉    | 39111/65536 [6:42:50<4:37:23,  1.59it/s]
 60%|█████▉    | 39112/65536 [6:42:51<4:35:31,  1.60it/s]
 60%|█████▉    | 39113/65536 [6:42:51<4:31:44,  1.62it/s]
 60%|█████▉    | 39114/65536 [6:42:52<4:30:09,  1.63it/s]
 60%|█████▉    | 39115/65536 [6:42:53<4:31:28,  1.62it/s]
 60%|█████▉    | 39116/65536 [6:42:53<4:31:22,  1.62it/s]
 60%|█████▉    | 39117/65536 [6:42:54<4:36:19,  1.59it/s]
 60%|█████▉    | 39118/65536 [6:42:55<4:40:48,  1.57it/s]
 60%|█████▉    | 39119/65536 [6:42:55<4:37:04,  1.59it/s]
 60%|█████▉    | 39120/65536 [6:42:56<4:31:47,  1.62it/s]
                                                         
{'loss': 1.6082, 'learning_rate': 4.6387902534590423e-07, 'epoch': 2414.81}
+
 60%|█████▉    | 39120/65536 [6:42:56<4:31:47,  1.62it/s]
 60%|█████▉    | 39121/65536 [6:42:56<4:26:44,  1.65it/s]
 60%|█████▉    | 39122/65536 [6:42:57<4:24:50,  1.66it/s]
 60%|█████▉    | 39123/65536 [6:42:58<4:21:20,  1.68it/s]
 60%|█████▉    | 39124/65536 [6:42:58<4:27:42,  1.64it/s]
 60%|█████▉    | 39125/65536 [6:42:59<4:25:56,  1.66it/s]
 60%|█████▉    | 39126/65536 [6:42:59<4:25:09,  1.66it/s]
 60%|█████▉    | 39127/65536 [6:43:00<4:22:45,  1.68it/s]
 60%|█████▉    | 39128/65536 [6:43:01<4:24:35,  1.66it/s]
 60%|█████▉    | 39129/65536 [6:43:01<4:21:18,  1.68it/s]
 60%|█████▉    | 39130/65536 [6:43:02<4:28:29,  1.64it/s]
 60%|█████▉    | 39131/65536 [6:43:03<4:37:28,  1.59it/s]
 60%|█████▉    | 39132/65536 [6:43:03<4:38:55,  1.58it/s]
 60%|█████▉    | 39133/65536 [6:43:04<4:37:12,  1.59it/s]
 60%|█████▉    | 39134/65536 [6:43:04<4:40:24,  1.57it/s]
 60%|█████▉    | 39135/65536 [6:43:05<4:31:55,  1.62it/s]
 60%|█████▉    | 39136/65536 [6:43:06<4:39:50,  1.57it/s]
 60%|█████▉    | 39137/65536 [6:43:06<4:37:12,  1.59it/s]
 60%|█████▉    | 39138/65536 [6:43:07<4:31:45,  1.62it/s]
 60%|█████▉    | 39139/65536 [6:43:07<4:28:58,  1.64it/s]
 60%|█████▉    | 39140/65536 [6:43:08<4:41:19,  1.56it/s]
                                                         
{'loss': 1.6282, 'learning_rate': 4.636035263866781e-07, 'epoch': 2416.05}
+
 60%|█████▉    | 39140/65536 [6:43:08<4:41:19,  1.56it/s]
 60%|█████▉    | 39141/65536 [6:43:09<4:36:36,  1.59it/s]
 60%|█████▉    | 39142/65536 [6:43:09<4:34:19,  1.60it/s]
 60%|█████▉    | 39143/65536 [6:43:10<4:33:41,  1.61it/s]
 60%|█████▉    | 39144/65536 [6:43:11<4:33:51,  1.61it/s]
 60%|█████▉    | 39145/65536 [6:43:11<4:32:18,  1.62it/s]
 60%|█████▉    | 39146/65536 [6:43:12<4:27:25,  1.64it/s]
 60%|█████▉    | 39147/65536 [6:43:12<4:23:59,  1.67it/s]
 60%|█████▉    | 39148/65536 [6:43:13<4:19:04,  1.70it/s]
 60%|█████▉    | 39149/65536 [6:43:14<4:23:32,  1.67it/s]
 60%|█████▉    | 39150/65536 [6:43:14<4:20:11,  1.69it/s]
 60%|█████▉    | 39151/65536 [6:43:15<4:20:46,  1.69it/s]
 60%|█████▉    | 39152/65536 [6:43:15<4:21:27,  1.68it/s]
 60%|█████▉    | 39153/65536 [6:43:16<4:29:24,  1.63it/s]
 60%|█████▉    | 39154/65536 [6:43:17<4:27:35,  1.64it/s]
 60%|█████▉    | 39155/65536 [6:43:17<4:22:38,  1.67it/s]
 60%|█████▉    | 39156/65536 [6:43:18<4:46:38,  1.53it/s]
 60%|█████▉    | 39157/65536 [6:43:19<4:44:39,  1.54it/s]
 60%|█████▉    | 39158/65536 [6:43:19<4:36:08,  1.59it/s]
 60%|█████▉    | 39159/65536 [6:43:20<4:39:29,  1.57it/s]
 60%|█████▉    | 39160/65536 [6:43:20<4:38:42,  1.58it/s]
                                                         
{'loss': 1.6102, 'learning_rate': 4.633280274274519e-07, 'epoch': 2417.28}
+
 60%|█████▉    | 39160/65536 [6:43:20<4:38:42,  1.58it/s]
 60%|█████▉    | 39161/65536 [6:43:21<4:35:58,  1.59it/s]
 60%|█████▉    | 39162/65536 [6:43:22<4:29:49,  1.63it/s]
 60%|█████▉    | 39163/65536 [6:43:22<4:32:32,  1.61it/s]
 60%|█████▉    | 39164/65536 [6:43:23<4:23:45,  1.67it/s]
 60%|█████▉    | 39165/65536 [6:43:23<4:25:08,  1.66it/s]
 60%|█████▉    | 39166/65536 [6:43:24<4:24:12,  1.66it/s]
 60%|█████▉    | 39167/65536 [6:43:25<4:21:50,  1.68it/s]
 60%|█████▉    | 39168/65536 [6:43:25<4:22:40,  1.67it/s]
 60%|█████▉    | 39169/65536 [6:43:26<4:24:32,  1.66it/s]
 60%|█████▉    | 39170/65536 [6:43:26<4:25:50,  1.65it/s]
 60%|█████▉    | 39171/65536 [6:43:27<4:36:09,  1.59it/s]
 60%|█████▉    | 39172/65536 [6:43:28<4:49:10,  1.52it/s]
 60%|█████▉    | 39173/65536 [6:43:29<4:43:13,  1.55it/s]
 60%|█████▉    | 39174/65536 [6:43:29<4:37:31,  1.58it/s]
 60%|█████▉    | 39175/65536 [6:43:30<4:38:50,  1.58it/s]
 60%|█████▉    | 39176/65536 [6:43:30<4:35:43,  1.59it/s]
 60%|█████▉    | 39177/65536 [6:43:31<4:30:58,  1.62it/s]
 60%|█████▉    | 39178/65536 [6:43:32<4:27:30,  1.64it/s]
 60%|█████▉    | 39179/65536 [6:43:32<4:26:31,  1.65it/s]
 60%|█████▉    | 39180/65536 [6:43:33<4:29:34,  1.63it/s]
                                                         
{'loss': 1.6249, 'learning_rate': 4.630525284682258e-07, 'epoch': 2418.52}
+
 60%|█████▉    | 39180/65536 [6:43:33<4:29:34,  1.63it/s]
 60%|█████▉    | 39181/65536 [6:43:33<4:32:34,  1.61it/s]
 60%|█████▉    | 39182/65536 [6:43:34<4:34:00,  1.60it/s]
 60%|█████▉    | 39183/65536 [6:43:35<4:31:44,  1.62it/s]
 60%|█████▉    | 39184/65536 [6:43:35<4:27:58,  1.64it/s]
 60%|█████▉    | 39185/65536 [6:43:36<4:28:09,  1.64it/s]
 60%|█████▉    | 39186/65536 [6:43:36<4:26:18,  1.65it/s]
 60%|█████▉    | 39187/65536 [6:43:37<4:23:49,  1.66it/s]
 60%|█████▉    | 39188/65536 [6:43:38<4:28:49,  1.63it/s]
 60%|█████▉    | 39189/65536 [6:43:38<4:26:51,  1.65it/s]
 60%|█████▉    | 39190/65536 [6:43:39<4:22:05,  1.68it/s]
 60%|█████▉    | 39191/65536 [6:43:40<4:30:58,  1.62it/s]
 60%|█████▉    | 39192/65536 [6:43:40<4:28:08,  1.64it/s]
 60%|█████▉    | 39193/65536 [6:43:41<4:22:41,  1.67it/s]
 60%|█████▉    | 39194/65536 [6:43:41<4:26:21,  1.65it/s]
 60%|█████▉    | 39195/65536 [6:43:42<4:28:49,  1.63it/s]
 60%|█████▉    | 39196/65536 [6:43:43<4:26:50,  1.65it/s]
 60%|█████▉    | 39197/65536 [6:43:43<4:24:09,  1.66it/s]
 60%|█████▉    | 39198/65536 [6:43:44<4:36:58,  1.58it/s]
 60%|█████▉    | 39199/65536 [6:43:44<4:33:16,  1.61it/s]
 60%|█████▉    | 39200/65536 [6:43:45<4:27:35,  1.64it/s]
                                                         
{'loss': 1.6388, 'learning_rate': 4.6277702950899966e-07, 'epoch': 2419.75}
+
 60%|█████▉    | 39200/65536 [6:43:45<4:27:35,  1.64it/s]
 60%|█████▉    | 39201/65536 [6:43:46<4:30:50,  1.62it/s]
 60%|█████▉    | 39202/65536 [6:43:46<4:24:35,  1.66it/s]
 60%|█████▉    | 39203/65536 [6:43:47<4:26:12,  1.65it/s]
 60%|█████▉    | 39204/65536 [6:43:47<4:25:21,  1.65it/s]
 60%|█████▉    | 39205/65536 [6:43:48<4:33:08,  1.61it/s]
 60%|█████▉    | 39206/65536 [6:43:49<4:27:37,  1.64it/s]
 60%|█████▉    | 39207/65536 [6:43:49<4:31:26,  1.62it/s]
 60%|█████▉    | 39208/65536 [6:43:50<4:27:32,  1.64it/s]
 60%|█████▉    | 39209/65536 [6:43:50<4:24:26,  1.66it/s]
 60%|█████▉    | 39210/65536 [6:43:51<4:29:24,  1.63it/s]
 60%|█████▉    | 39211/65536 [6:43:52<4:21:52,  1.68it/s]
 60%|█████▉    | 39212/65536 [6:43:52<4:31:24,  1.62it/s]
 60%|█████▉    | 39213/65536 [6:43:53<4:29:49,  1.63it/s]
 60%|█████▉    | 39214/65536 [6:43:54<4:30:54,  1.62it/s]
 60%|█████▉    | 39215/65536 [6:43:54<4:28:38,  1.63it/s]
 60%|█████▉    | 39216/65536 [6:43:55<4:34:21,  1.60it/s]
 60%|█████▉    | 39217/65536 [6:43:55<4:31:31,  1.62it/s]
 60%|█████▉    | 39218/65536 [6:43:56<4:27:33,  1.64it/s]
 60%|█████▉    | 39219/65536 [6:43:57<4:27:50,  1.64it/s]
 60%|█████▉    | 39220/65536 [6:43:57<4:26:58,  1.64it/s]
                                                         
{'loss': 1.6135, 'learning_rate': 4.625015305497734e-07, 'epoch': 2420.99}
+
 60%|█████▉    | 39220/65536 [6:43:57<4:26:58,  1.64it/s]
 60%|█████▉    | 39221/65536 [6:43:58<4:39:37,  1.57it/s]
 60%|█████▉    | 39222/65536 [6:43:59<4:34:13,  1.60it/s]
 60%|█████▉    | 39223/65536 [6:43:59<4:26:14,  1.65it/s]
 60%|█████▉    | 39224/65536 [6:44:00<4:37:49,  1.58it/s]
 60%|█████▉    | 39225/65536 [6:44:00<4:32:05,  1.61it/s]
 60%|█████▉    | 39226/65536 [6:44:01<4:30:50,  1.62it/s]
 60%|█████▉    | 39227/65536 [6:44:02<4:39:20,  1.57it/s]
 60%|█████▉    | 39228/65536 [6:44:02<4:27:15,  1.64it/s]
 60%|█████▉    | 39229/65536 [6:44:03<4:27:48,  1.64it/s]
 60%|█████▉    | 39230/65536 [6:44:03<4:23:33,  1.66it/s]
 60%|█████▉    | 39231/65536 [6:44:04<4:29:22,  1.63it/s]
 60%|█████▉    | 39232/65536 [6:44:05<4:31:51,  1.61it/s]
 60%|█████▉    | 39233/65536 [6:44:05<4:36:11,  1.59it/s]
 60%|█████▉    | 39234/65536 [6:44:06<4:36:28,  1.59it/s]
 60%|█████▉    | 39235/65536 [6:44:07<4:30:39,  1.62it/s]
 60%|█████▉    | 39236/65536 [6:44:07<4:24:49,  1.66it/s]
 60%|█████▉    | 39237/65536 [6:44:08<4:32:55,  1.61it/s]
 60%|█████▉    | 39238/65536 [6:44:08<4:28:54,  1.63it/s]
 60%|█████▉    | 39239/65536 [6:44:09<4:30:38,  1.62it/s]
 60%|█████▉    | 39240/65536 [6:44:10<4:26:45,  1.64it/s]
                                                         
{'loss': 1.6742, 'learning_rate': 4.622260315905473e-07, 'epoch': 2422.22}
+
 60%|█████▉    | 39240/65536 [6:44:10<4:26:45,  1.64it/s]
 60%|█████▉    | 39241/65536 [6:44:10<4:24:23,  1.66it/s]
 60%|█████▉    | 39242/65536 [6:44:11<4:27:55,  1.64it/s]
 60%|█████▉    | 39243/65536 [6:44:11<4:27:07,  1.64it/s]
 60%|█████▉    | 39244/65536 [6:44:12<4:23:38,  1.66it/s]
 60%|█████▉    | 39245/65536 [6:44:13<4:26:05,  1.65it/s]
 60%|█████▉    | 39246/65536 [6:44:13<4:34:11,  1.60it/s]
 60%|█████▉    | 39247/65536 [6:44:14<4:28:06,  1.63it/s]
 60%|█████▉    | 39248/65536 [6:44:14<4:22:36,  1.67it/s]
 60%|█████▉    | 39249/65536 [6:44:15<4:36:23,  1.59it/s]
 60%|█████▉    | 39250/65536 [6:44:16<4:33:26,  1.60it/s]
 60%|█████▉    | 39251/65536 [6:44:16<4:31:05,  1.62it/s]
 60%|█████▉    | 39252/65536 [6:44:17<4:27:27,  1.64it/s]
 60%|█████▉    | 39253/65536 [6:44:18<4:32:12,  1.61it/s]
 60%|█████▉    | 39254/65536 [6:44:18<4:30:33,  1.62it/s]
 60%|█████▉    | 39255/65536 [6:44:19<4:32:40,  1.61it/s]
 60%|█████▉    | 39256/65536 [6:44:19<4:32:10,  1.61it/s]
 60%|█████▉    | 39257/65536 [6:44:20<4:25:17,  1.65it/s]
 60%|█████▉    | 39258/65536 [6:44:21<4:22:41,  1.67it/s]
 60%|█████▉    | 39259/65536 [6:44:21<4:21:38,  1.67it/s]
 60%|█████▉    | 39260/65536 [6:44:22<4:23:47,  1.66it/s]
                                                         
{'loss': 1.6415, 'learning_rate': 4.6195053263132113e-07, 'epoch': 2423.46}
+
 60%|█████▉    | 39260/65536 [6:44:22<4:23:47,  1.66it/s]
 60%|█████▉    | 39261/65536 [6:44:22<4:27:29,  1.64it/s]
 60%|█████▉    | 39262/65536 [6:44:23<4:25:06,  1.65it/s]
 60%|█████▉    | 39263/65536 [6:44:24<4:30:08,  1.62it/s]
 60%|█████▉    | 39264/65536 [6:44:24<4:32:46,  1.61it/s]
 60%|█████▉    | 39265/65536 [6:44:25<4:32:30,  1.61it/s]
 60%|█████▉    | 39266/65536 [6:44:26<4:36:28,  1.58it/s]
 60%|█████▉    | 39267/65536 [6:44:26<4:33:36,  1.60it/s]
 60%|█████▉    | 39268/65536 [6:44:27<4:26:56,  1.64it/s]
 60%|█████▉    | 39269/65536 [6:44:28<4:37:37,  1.58it/s]
 60%|█████▉    | 39270/65536 [6:44:28<4:39:07,  1.57it/s]
 60%|█████▉    | 39271/65536 [6:44:29<4:37:48,  1.58it/s]
 60%|█████▉    | 39272/65536 [6:44:29<4:30:48,  1.62it/s]
 60%|█████▉    | 39273/65536 [6:44:30<4:25:26,  1.65it/s]
 60%|█████▉    | 39274/65536 [6:44:31<4:26:26,  1.64it/s]
 60%|█████▉    | 39275/65536 [6:44:31<4:22:37,  1.67it/s]
 60%|█████▉    | 39276/65536 [6:44:32<4:32:09,  1.61it/s]
 60%|█████▉    | 39277/65536 [6:44:32<4:28:54,  1.63it/s]
 60%|█████▉    | 39278/65536 [6:44:33<4:28:43,  1.63it/s]
 60%|█████▉    | 39279/65536 [6:44:34<4:24:46,  1.65it/s]
 60%|█████▉    | 39280/65536 [6:44:34<4:33:42,  1.60it/s]
                                                         
{'loss': 1.6385, 'learning_rate': 4.61675033672095e-07, 'epoch': 2424.69}
+
 60%|█████▉    | 39280/65536 [6:44:34<4:33:42,  1.60it/s]
 60%|█████▉    | 39281/65536 [6:44:35<4:39:24,  1.57it/s]
 60%|█████▉    | 39282/65536 [6:44:36<4:46:28,  1.53it/s]
 60%|█████▉    | 39283/65536 [6:44:36<4:37:15,  1.58it/s]
 60%|█████▉    | 39284/65536 [6:44:37<4:34:33,  1.59it/s]
 60%|█████▉    | 39285/65536 [6:44:37<4:30:46,  1.62it/s]
 60%|█████▉    | 39286/65536 [6:44:38<4:38:04,  1.57it/s]
 60%|█████▉    | 39287/65536 [6:44:39<4:37:11,  1.58it/s]
 60%|█████▉    | 39288/65536 [6:44:39<4:31:42,  1.61it/s]
 60%|█████▉    | 39289/65536 [6:44:40<4:27:54,  1.63it/s]
 60%|█████▉    | 39290/65536 [6:44:41<4:33:16,  1.60it/s]
 60%|█████▉    | 39291/65536 [6:44:41<4:30:21,  1.62it/s]
 60%|█████▉    | 39292/65536 [6:44:42<4:24:00,  1.66it/s]
 60%|█████▉    | 39293/65536 [6:44:42<4:20:58,  1.68it/s]
 60%|█████▉    | 39294/65536 [6:44:43<4:25:09,  1.65it/s]
 60%|█████▉    | 39295/65536 [6:44:44<4:23:30,  1.66it/s]
 60%|█████▉    | 39296/65536 [6:44:44<4:30:18,  1.62it/s]
 60%|█████▉    | 39297/65536 [6:44:45<4:26:06,  1.64it/s]
 60%|█████▉    | 39298/65536 [6:44:45<4:24:32,  1.65it/s]
 60%|█████▉    | 39299/65536 [6:44:46<4:23:00,  1.66it/s]
 60%|█████▉    | 39300/65536 [6:44:47<4:32:43,  1.60it/s]
                                                         
{'loss': 1.6201, 'learning_rate': 4.6139953471286885e-07, 'epoch': 2425.93}
+
 60%|█████▉    | 39300/65536 [6:44:47<4:32:43,  1.60it/s]
 60%|█████▉    | 39301/65536 [6:44:47<4:31:59,  1.61it/s]
 60%|█████▉    | 39302/65536 [6:44:48<4:37:49,  1.57it/s]
 60%|█████▉    | 39303/65536 [6:44:49<4:31:25,  1.61it/s]
 60%|█████▉    | 39304/65536 [6:44:49<4:28:04,  1.63it/s]
 60%|█████▉    | 39305/65536 [6:44:50<4:28:53,  1.63it/s]
 60%|█████▉    | 39306/65536 [6:44:50<4:25:26,  1.65it/s]
 60%|█████▉    | 39307/65536 [6:44:51<4:27:32,  1.63it/s]
 60%|█████▉    | 39308/65536 [6:44:52<4:29:22,  1.62it/s]
 60%|█████▉    | 39309/65536 [6:44:52<4:24:16,  1.65it/s]
 60%|█████▉    | 39310/65536 [6:44:53<4:21:20,  1.67it/s]
 60%|█████▉    | 39311/65536 [6:44:53<4:30:51,  1.61it/s]
 60%|█████▉    | 39312/65536 [6:44:54<4:28:33,  1.63it/s]
 60%|█████▉    | 39313/65536 [6:44:55<4:29:57,  1.62it/s]
 60%|█████▉    | 39314/65536 [6:44:55<4:25:55,  1.64it/s]
 60%|█████▉    | 39315/65536 [6:44:56<4:32:40,  1.60it/s]
 60%|█████▉    | 39316/65536 [6:44:56<4:29:09,  1.62it/s]
 60%|█████▉    | 39317/65536 [6:44:57<4:27:35,  1.63it/s]
 60%|█████▉    | 39318/65536 [6:44:58<4:36:18,  1.58it/s]
 60%|█████▉    | 39319/65536 [6:44:58<4:34:08,  1.59it/s]
 60%|█████▉    | 39320/65536 [6:44:59<4:29:58,  1.62it/s]
                                                         
{'loss': 1.6216, 'learning_rate': 4.611240357536427e-07, 'epoch': 2427.16}
+
 60%|█████▉    | 39320/65536 [6:44:59<4:29:58,  1.62it/s]
 60%|█████▉    | 39321/65536 [6:45:00<4:27:29,  1.63it/s]
 60%|██████    | 39322/65536 [6:45:00<4:31:59,  1.61it/s]
 60%|██████    | 39323/65536 [6:45:01<4:34:01,  1.59it/s]
 60%|██████    | 39324/65536 [6:45:01<4:27:20,  1.63it/s]
 60%|██████    | 39325/65536 [6:45:02<4:28:21,  1.63it/s]
 60%|██████    | 39326/65536 [6:45:03<4:31:20,  1.61it/s]
 60%|██████    | 39327/65536 [6:45:03<4:30:01,  1.62it/s]
 60%|██████    | 39328/65536 [6:45:04<4:25:44,  1.64it/s]
 60%|██████    | 39329/65536 [6:45:04<4:24:29,  1.65it/s]
 60%|██████    | 39330/65536 [6:45:05<4:20:54,  1.67it/s]
 60%|██████    | 39331/65536 [6:45:06<4:16:28,  1.70it/s]
 60%|██████    | 39332/65536 [6:45:06<4:34:08,  1.59it/s]
 60%|██████    | 39333/65536 [6:45:07<4:49:54,  1.51it/s]
 60%|██████    | 39334/65536 [6:45:08<5:08:27,  1.42it/s]
 60%|██████    | 39335/65536 [6:45:09<4:58:49,  1.46it/s]
 60%|██████    | 39336/65536 [6:45:09<4:50:12,  1.50it/s]
 60%|██████    | 39337/65536 [6:45:10<4:44:59,  1.53it/s]
 60%|██████    | 39338/65536 [6:45:10<4:37:45,  1.57it/s]
 60%|██████    | 39339/65536 [6:45:11<4:33:34,  1.60it/s]
 60%|██████    | 39340/65536 [6:45:12<4:34:14,  1.59it/s]
                                                         
{'loss': 1.5967, 'learning_rate': 4.6084853679441657e-07, 'epoch': 2428.4}
+
 60%|██████    | 39340/65536 [6:45:12<4:34:14,  1.59it/s]
 60%|██████    | 39341/65536 [6:45:12<4:31:08,  1.61it/s]
 60%|██████    | 39342/65536 [6:45:13<4:26:49,  1.64it/s]
 60%|██████    | 39343/65536 [6:45:13<4:23:29,  1.66it/s]
 60%|██████    | 39344/65536 [6:45:14<4:24:19,  1.65it/s]
 60%|██████    | 39345/65536 [6:45:15<4:24:16,  1.65it/s]
 60%|██████    | 39346/65536 [6:45:15<4:22:11,  1.66it/s]
 60%|██████    | 39347/65536 [6:45:16<4:32:32,  1.60it/s]
 60%|██████    | 39348/65536 [6:45:17<4:33:30,  1.60it/s]
 60%|██████    | 39349/65536 [6:45:17<4:26:03,  1.64it/s]
 60%|██████    | 39350/65536 [6:45:18<4:37:50,  1.57it/s]
 60%|██████    | 39351/65536 [6:45:18<4:39:27,  1.56it/s]
 60%|██████    | 39352/65536 [6:45:19<4:41:05,  1.55it/s]
 60%|██████    | 39353/65536 [6:45:20<4:30:55,  1.61it/s]
 60%|██████    | 39354/65536 [6:45:20<4:32:17,  1.60it/s]
 60%|██████    | 39355/65536 [6:45:21<4:29:01,  1.62it/s]
 60%|██████    | 39356/65536 [6:45:21<4:29:42,  1.62it/s]
 60%|██████    | 39357/65536 [6:45:22<4:23:48,  1.65it/s]
 60%|██████    | 39358/65536 [6:45:23<4:23:53,  1.65it/s]
 60%|██████    | 39359/65536 [6:45:23<4:27:07,  1.63it/s]
 60%|██████    | 39360/65536 [6:45:24<4:26:45,  1.64it/s]
                                                         
{'loss': 1.6069, 'learning_rate': 4.605730378351903e-07, 'epoch': 2429.63}
+
 60%|██████    | 39360/65536 [6:45:24<4:26:45,  1.64it/s]
 60%|██████    | 39361/65536 [6:45:25<4:24:30,  1.65it/s]
 60%|██████    | 39362/65536 [6:45:25<4:24:34,  1.65it/s]
 60%|██████    | 39363/65536 [6:45:26<4:32:31,  1.60it/s]
 60%|██████    | 39364/65536 [6:45:26<4:30:17,  1.61it/s]
 60%|██████    | 39365/65536 [6:45:27<4:30:50,  1.61it/s]
 60%|██████    | 39366/65536 [6:45:28<4:28:01,  1.63it/s]
 60%|██████    | 39367/65536 [6:45:28<4:41:08,  1.55it/s]
 60%|██████    | 39368/65536 [6:45:29<4:37:26,  1.57it/s]
 60%|██████    | 39369/65536 [6:45:30<4:41:01,  1.55it/s]
 60%|██████    | 39370/65536 [6:45:30<4:35:05,  1.59it/s]
 60%|██████    | 39371/65536 [6:45:31<4:30:09,  1.61it/s]
 60%|██████    | 39372/65536 [6:45:31<4:31:02,  1.61it/s]
 60%|██████    | 39373/65536 [6:45:32<4:28:35,  1.62it/s]
 60%|██████    | 39374/65536 [6:45:33<4:26:12,  1.64it/s]
 60%|██████    | 39375/65536 [6:45:33<4:27:01,  1.63it/s]
 60%|██████    | 39376/65536 [6:45:34<4:23:44,  1.65it/s]
 60%|██████    | 39377/65536 [6:45:34<4:26:03,  1.64it/s]
 60%|██████    | 39378/65536 [6:45:35<4:26:54,  1.63it/s]
 60%|██████    | 39379/65536 [6:45:36<4:34:49,  1.59it/s]
 60%|██████    | 39380/65536 [6:45:36<4:30:31,  1.61it/s]
                                                         
{'loss': 1.6185, 'learning_rate': 4.602975388759642e-07, 'epoch': 2430.86}
+
 60%|██████    | 39380/65536 [6:45:36<4:30:31,  1.61it/s]
 60%|██████    | 39381/65536 [6:45:37<4:25:07,  1.64it/s]
 60%|██████    | 39382/65536 [6:45:38<4:22:09,  1.66it/s]
 60%|██████    | 39383/65536 [6:45:38<4:47:06,  1.52it/s]
 60%|██████    | 39384/65536 [6:45:39<4:52:58,  1.49it/s]
 60%|██████    | 39385/65536 [6:45:40<4:40:47,  1.55it/s]
 60%|██████    | 39386/65536 [6:45:40<4:36:00,  1.58it/s]
 60%|██████    | 39387/65536 [6:45:41<4:33:43,  1.59it/s]
 60%|██████    | 39388/65536 [6:45:41<4:28:35,  1.62it/s]
 60%|██████    | 39389/65536 [6:45:42<4:30:52,  1.61it/s]
 60%|██████    | 39390/65536 [6:45:43<4:40:24,  1.55it/s]
 60%|██████    | 39391/65536 [6:45:43<4:43:27,  1.54it/s]
 60%|██████    | 39392/65536 [6:45:44<4:36:17,  1.58it/s]
 60%|██████    | 39393/65536 [6:45:45<4:27:56,  1.63it/s]
 60%|██████    | 39394/65536 [6:45:45<4:27:40,  1.63it/s]
 60%|██████    | 39395/65536 [6:45:46<4:28:33,  1.62it/s]
 60%|██████    | 39396/65536 [6:45:46<4:34:02,  1.59it/s]
 60%|██████    | 39397/65536 [6:45:47<4:31:15,  1.61it/s]
 60%|██████    | 39398/65536 [6:45:48<4:34:37,  1.59it/s]
 60%|██████    | 39399/65536 [6:45:48<4:37:09,  1.57it/s]
 60%|██████    | 39400/65536 [6:45:49<4:35:18,  1.58it/s]
                                                         
{'loss': 1.6183, 'learning_rate': 4.6002203991673804e-07, 'epoch': 2432.1}
+
 60%|██████    | 39400/65536 [6:45:49<4:35:18,  1.58it/s]
 60%|██████    | 39401/65536 [6:45:50<4:38:36,  1.56it/s]
 60%|██████    | 39402/65536 [6:45:50<4:35:01,  1.58it/s]
 60%|██████    | 39403/65536 [6:45:51<4:32:06,  1.60it/s]
 60%|██████    | 39404/65536 [6:45:51<4:31:12,  1.61it/s]
 60%|██████    | 39405/65536 [6:45:52<4:29:28,  1.62it/s]
 60%|██████    | 39406/65536 [6:45:53<4:32:14,  1.60it/s]
 60%|██████    | 39407/65536 [6:45:53<4:30:19,  1.61it/s]
 60%|██████    | 39408/65536 [6:45:54<4:34:50,  1.58it/s]
 60%|██████    | 39409/65536 [6:45:55<4:40:04,  1.55it/s]
 60%|██████    | 39410/65536 [6:45:55<4:37:53,  1.57it/s]
 60%|██████    | 39411/65536 [6:45:56<4:32:30,  1.60it/s]
 60%|██████    | 39412/65536 [6:45:56<4:26:07,  1.64it/s]
 60%|██████    | 39413/65536 [6:45:57<4:26:01,  1.64it/s]
 60%|██████    | 39414/65536 [6:45:58<4:21:11,  1.67it/s]
 60%|██████    | 39415/65536 [6:45:58<4:37:50,  1.57it/s]
 60%|██████    | 39416/65536 [6:45:59<4:37:38,  1.57it/s]
 60%|██████    | 39417/65536 [6:46:00<4:41:18,  1.55it/s]
 60%|██████    | 39418/65536 [6:46:00<4:33:57,  1.59it/s]
 60%|██████    | 39419/65536 [6:46:01<4:31:18,  1.60it/s]
 60%|██████    | 39420/65536 [6:46:01<4:24:26,  1.65it/s]
                                                         
{'loss': 1.613, 'learning_rate': 4.597465409575119e-07, 'epoch': 2433.33}
+
 60%|██████    | 39420/65536 [6:46:01<4:24:26,  1.65it/s]
 60%|██████    | 39421/65536 [6:46:02<4:29:07,  1.62it/s]
 60%|██████    | 39422/65536 [6:46:03<4:26:23,  1.63it/s]
 60%|██████    | 39423/65536 [6:46:03<4:27:55,  1.62it/s]
 60%|██████    | 39424/65536 [6:46:04<4:30:24,  1.61it/s]
 60%|██████    | 39425/65536 [6:46:05<4:24:36,  1.64it/s]
 60%|██████    | 39426/65536 [6:46:05<4:24:24,  1.65it/s]
 60%|██████    | 39427/65536 [6:46:06<4:29:00,  1.62it/s]
 60%|██████    | 39428/65536 [6:46:06<4:22:56,  1.65it/s]
 60%|██████    | 39429/65536 [6:46:07<4:24:07,  1.65it/s]
 60%|██████    | 39430/65536 [6:46:08<4:25:16,  1.64it/s]
 60%|██████    | 39431/65536 [6:46:08<4:39:47,  1.55it/s]
 60%|██████    | 39432/65536 [6:46:09<4:33:26,  1.59it/s]
 60%|██████    | 39433/65536 [6:46:10<4:34:27,  1.59it/s]
 60%|██████    | 39434/65536 [6:46:10<4:27:41,  1.63it/s]
 60%|██████    | 39435/65536 [6:46:11<4:23:19,  1.65it/s]
 60%|██████    | 39436/65536 [6:46:11<4:27:26,  1.63it/s]
 60%|██████    | 39437/65536 [6:46:12<4:25:40,  1.64it/s]
 60%|██████    | 39438/65536 [6:46:13<4:36:12,  1.57it/s]
 60%|██████    | 39439/65536 [6:46:13<4:41:05,  1.55it/s]
 60%|██████    | 39440/65536 [6:46:14<4:45:09,  1.53it/s]
                                                         
{'loss': 1.6076, 'learning_rate': 4.5947104199828576e-07, 'epoch': 2434.57}
+
 60%|██████    | 39440/65536 [6:46:14<4:45:09,  1.53it/s]
 60%|██████    | 39441/65536 [6:46:15<4:36:33,  1.57it/s]
 60%|██████    | 39442/65536 [6:46:15<4:32:33,  1.60it/s]
 60%|██████    | 39443/65536 [6:46:16<4:36:42,  1.57it/s]
 60%|██████    | 39444/65536 [6:46:16<4:31:07,  1.60it/s]
 60%|██████    | 39445/65536 [6:46:17<4:25:28,  1.64it/s]
 60%|██████    | 39446/65536 [6:46:18<4:25:09,  1.64it/s]
 60%|██████    | 39447/65536 [6:46:18<4:26:05,  1.63it/s]
 60%|██████    | 39448/65536 [6:46:19<4:32:43,  1.59it/s]
 60%|██████    | 39449/65536 [6:46:19<4:28:52,  1.62it/s]
 60%|██████    | 39450/65536 [6:46:20<4:35:45,  1.58it/s]
 60%|██████    | 39451/65536 [6:46:21<4:36:38,  1.57it/s]
 60%|██████    | 39452/65536 [6:46:21<4:39:46,  1.55it/s]
 60%|██████    | 39453/65536 [6:46:22<4:32:56,  1.59it/s]
 60%|██████    | 39454/65536 [6:46:23<4:38:05,  1.56it/s]
 60%|██████    | 39455/65536 [6:46:23<4:35:50,  1.58it/s]
 60%|██████    | 39456/65536 [6:46:24<4:33:06,  1.59it/s]
 60%|██████    | 39457/65536 [6:46:25<4:27:35,  1.62it/s]
 60%|██████    | 39458/65536 [6:46:25<4:28:54,  1.62it/s]
 60%|██████    | 39459/65536 [6:46:26<4:26:08,  1.63it/s]
 60%|██████    | 39460/65536 [6:46:26<4:24:49,  1.64it/s]
                                                         
{'loss': 1.6154, 'learning_rate': 4.591955430390596e-07, 'epoch': 2435.8}
+
 60%|██████    | 39460/65536 [6:46:26<4:24:49,  1.64it/s]
 60%|██████    | 39461/65536 [6:46:27<4:24:58,  1.64it/s]
 60%|██████    | 39462/65536 [6:46:28<4:22:53,  1.65it/s]
 60%|██████    | 39463/65536 [6:46:28<4:16:00,  1.70it/s]
 60%|██████    | 39464/65536 [6:46:29<4:32:44,  1.59it/s]
 60%|██████    | 39465/65536 [6:46:29<4:26:21,  1.63it/s]
 60%|██████    | 39466/65536 [6:46:30<4:28:30,  1.62it/s]
 60%|██████    | 39467/65536 [6:46:31<4:24:52,  1.64it/s]
 60%|██████    | 39468/65536 [6:46:31<4:26:48,  1.63it/s]
 60%|██████    | 39469/65536 [6:46:32<4:31:26,  1.60it/s]
 60%|██████    | 39470/65536 [6:46:33<4:28:12,  1.62it/s]
 60%|████���█    | 39471/65536 [6:46:33<4:24:29,  1.64it/s]
 60%|██████    | 39472/65536 [6:46:34<4:20:21,  1.67it/s]
 60%|██████    | 39473/65536 [6:46:34<4:20:52,  1.67it/s]
 60%|██████    | 39474/65536 [6:46:35<4:22:15,  1.66it/s]
 60%|██████    | 39475/65536 [6:46:35<4:21:11,  1.66it/s]
 60%|██████    | 39476/65536 [6:46:36<4:23:05,  1.65it/s]
 60%|██████    | 39477/65536 [6:46:37<4:24:04,  1.64it/s]
 60%|██████    | 39478/65536 [6:46:37<4:35:56,  1.57it/s]
 60%|██████    | 39479/65536 [6:46:38<4:33:13,  1.59it/s]
 60%|██████    | 39480/65536 [6:46:39<4:32:26,  1.59it/s]
                                                         
{'loss': 1.6337, 'learning_rate': 4.5892004407983353e-07, 'epoch': 2437.04}
+
 60%|██████    | 39480/65536 [6:46:39<4:32:26,  1.59it/s]
 60%|██████    | 39481/65536 [6:46:39<4:32:03,  1.60it/s]
 60%|██████    | 39482/65536 [6:46:40<4:35:07,  1.58it/s]
 60%|██████    | 39483/65536 [6:46:41<4:32:58,  1.59it/s]
 60%|██████    | 39484/65536 [6:46:41<4:34:09,  1.58it/s]
 60%|██████    | 39485/65536 [6:46:42<4:28:37,  1.62it/s]
 60%|██████    | 39486/65536 [6:46:42<4:25:23,  1.64it/s]
 60%|██████    | 39487/65536 [6:46:43<4:20:37,  1.67it/s]
 60%|██████    | 39488/65536 [6:46:44<4:17:46,  1.68it/s]
 60%|██████    | 39489/65536 [6:46:44<4:23:30,  1.65it/s]
 60%|██████    | 39490/65536 [6:46:45<4:24:08,  1.64it/s]
 60%|██████    | 39491/65536 [6:46:45<4:21:06,  1.66it/s]
 60%|██████    | 39492/65536 [6:46:46<4:25:53,  1.63it/s]
 60%|██████    | 39493/65536 [6:46:47<4:32:07,  1.60it/s]
 60%|██████    | 39494/65536 [6:46:47<4:33:43,  1.59it/s]
 60%|██████    | 39495/65536 [6:46:48<4:32:32,  1.59it/s]
 60%|██████    | 39496/65536 [6:46:49<4:46:53,  1.51it/s]
 60%|██████    | 39497/65536 [6:46:49<4:42:32,  1.54it/s]
 60%|██████    | 39498/65536 [6:46:50<4:38:56,  1.56it/s]
 60%|██████    | 39499/65536 [6:46:51<4:41:34,  1.54it/s]
 60%|██████    | 39500/65536 [6:46:51<4:38:42,  1.56it/s]
                                                         
{'loss': 1.6416, 'learning_rate': 4.586445451206073e-07, 'epoch': 2438.27}
+
 60%|██████    | 39500/65536 [6:46:51<4:38:42,  1.56it/s]
 60%|██████    | 39501/65536 [6:46:52<4:29:44,  1.61it/s]
 60%|██████    | 39502/65536 [6:46:52<4:25:41,  1.63it/s]
 60%|██████    | 39503/65536 [6:46:53<4:20:11,  1.67it/s]
 60%|██████    | 39504/65536 [6:46:54<4:22:43,  1.65it/s]
 60%|██████    | 39505/65536 [6:46:54<4:25:55,  1.63it/s]
 60%|██████    | 39506/65536 [6:46:55<4:25:02,  1.64it/s]
 60%|██████    | 39507/65536 [6:46:55<4:26:07,  1.63it/s]
 60%|██████    | 39508/65536 [6:46:56<4:30:14,  1.61it/s]
 60%|██████    | 39509/65536 [6:46:57<4:30:22,  1.60it/s]
 60%|██████    | 39510/65536 [6:46:57<4:35:43,  1.57it/s]
 60%|██████    | 39511/65536 [6:46:58<4:32:53,  1.59it/s]
 60%|██████    | 39512/65536 [6:46:59<4:38:54,  1.56it/s]
 60%|██████    | 39513/65536 [6:46:59<4:35:39,  1.57it/s]
 60%|██████    | 39514/65536 [6:47:00<4:33:46,  1.58it/s]
 60%|██████    | 39515/65536 [6:47:00<4:31:04,  1.60it/s]
 60%|██████    | 39516/65536 [6:47:01<4:35:03,  1.58it/s]
 60%|██████    | 39517/65536 [6:47:02<4:31:36,  1.60it/s]
 60%|██████    | 39518/65536 [6:47:02<4:26:05,  1.63it/s]
 60%|██████    | 39519/65536 [6:47:03<4:29:14,  1.61it/s]
 60%|██████    | 39520/65536 [6:47:04<4:30:03,  1.61it/s]
                                                         
{'loss': 1.615, 'learning_rate': 4.583690461613811e-07, 'epoch': 2439.51}
+
 60%|██████    | 39520/65536 [6:47:04<4:30:03,  1.61it/s]
 60%|██████    | 39521/65536 [6:47:04<4:26:40,  1.63it/s]
 60%|██████    | 39522/65536 [6:47:05<4:24:59,  1.64it/s]
 60%|██████    | 39523/65536 [6:47:05<4:23:38,  1.64it/s]
 60%|██████    | 39524/65536 [6:47:06<4:29:34,  1.61it/s]
 60%|██████    | 39525/65536 [6:47:07<4:24:59,  1.64it/s]
 60%|██████    | 39526/65536 [6:47:07<4:25:59,  1.63it/s]
 60%|██████    | 39527/65536 [6:47:08<4:33:41,  1.58it/s]
 60%|██████    | 39528/65536 [6:47:09<4:27:48,  1.62it/s]
 60%|██████    | 39529/65536 [6:47:09<4:38:41,  1.56it/s]
 60%|██████    | 39530/65536 [6:47:10<4:29:36,  1.61it/s]
 60%|██████    | 39531/65536 [6:47:10<4:29:11,  1.61it/s]
 60%|██████    | 39532/65536 [6:47:11<4:23:29,  1.64it/s]
 60%|██████    | 39533/65536 [6:47:12<4:31:09,  1.60it/s]
 60%|██████    | 39534/65536 [6:47:12<4:28:58,  1.61it/s]
 60%|██████    | 39535/65536 [6:47:13<4:27:50,  1.62it/s]
 60%|██████    | 39536/65536 [6:47:13<4:29:06,  1.61it/s]
 60%|██████    | 39537/65536 [6:47:14<4:28:21,  1.61it/s]
 60%|██████    | 39538/65536 [6:47:15<4:26:04,  1.63it/s]
 60%|██████    | 39539/65536 [6:47:15<4:27:53,  1.62it/s]
 60%|██████    | 39540/65536 [6:47:16<4:26:32,  1.63it/s]
                                                         
{'loss': 1.6198, 'learning_rate': 4.58093547202155e-07, 'epoch': 2440.74}
+
 60%|██████    | 39540/65536 [6:47:16<4:26:32,  1.63it/s]
 60%|██████    | 39541/65536 [6:47:17<4:33:12,  1.59it/s]
 60%|██████    | 39542/65536 [6:47:17<4:28:04,  1.62it/s]
 60%|██████    | 39543/65536 [6:47:18<4:29:50,  1.61it/s]
 60%|██████    | 39544/65536 [6:47:18<4:26:06,  1.63it/s]
 60%|██████    | 39545/65536 [6:47:19<4:37:06,  1.56it/s]
 60%|██████    | 39546/65536 [6:47:20<4:34:53,  1.58it/s]
 60%|██████    | 39547/65536 [6:47:20<4:29:41,  1.61it/s]
 60%|██████    | 39548/65536 [6:47:21<4:25:16,  1.63it/s]
 60%|██████    | 39549/65536 [6:47:21<4:18:09,  1.68it/s]
 60%|██████    | 39550/65536 [6:47:22<4:17:11,  1.68it/s]
 60%|██████    | 39551/65536 [6:47:23<4:20:55,  1.66it/s]
 60%|██████    | 39552/65536 [6:47:23<4:17:28,  1.68it/s]
 60%|██████    | 39553/65536 [6:47:24<4:27:32,  1.62it/s]
 60%|██████    | 39554/65536 [6:47:25<4:28:08,  1.61it/s]
 60%|██████    | 39555/65536 [6:47:25<4:26:30,  1.62it/s]
 60%|██████    | 39556/65536 [6:47:26<4:24:46,  1.64it/s]
 60%|██████    | 39557/65536 [6:47:26<4:25:48,  1.63it/s]
 60%|██████    | 39558/65536 [6:47:27<4:26:42,  1.62it/s]
 60%|██████    | 39559/65536 [6:47:28<4:27:35,  1.62it/s]
 60%|██████    | 39560/65536 [6:47:28<4:33:02,  1.59it/s]
                                                         
{'loss': 1.6119, 'learning_rate': 4.5781804824292886e-07, 'epoch': 2441.98}
+
 60%|██████    | 39560/65536 [6:47:28<4:33:02,  1.59it/s]
 60%|██████    | 39561/65536 [6:47:29<4:38:41,  1.55it/s]
 60%|██████    | 39562/65536 [6:47:30<4:30:04,  1.60it/s]
 60%|██████    | 39563/65536 [6:47:30<4:26:43,  1.62it/s]
 60%|██████    | 39564/65536 [6:47:31<4:23:24,  1.64it/s]
 60%|██████    | 39565/65536 [6:47:31<4:25:51,  1.63it/s]
 60%|██████    | 39566/65536 [6:47:32<4:27:07,  1.62it/s]
 60%|██████    | 39567/65536 [6:47:33<4:22:43,  1.65it/s]
 60%|██████    | 39568/65536 [6:47:33<4:18:56,  1.67it/s]
 60%|██████    | 39569/65536 [6:47:34<4:24:35,  1.64it/s]
 60%|██████    | 39570/65536 [6:47:34<4:21:21,  1.66it/s]
 60%|██████    | 39571/65536 [6:47:35<4:29:49,  1.60it/s]
 60%|██████    | 39572/65536 [6:47:36<4:26:39,  1.62it/s]
 60%|██████    | 39573/65536 [6:47:36<4:33:32,  1.58it/s]
 60%|██████    | 39574/65536 [6:47:37<4:33:08,  1.58it/s]
 60%|██████    | 39575/65536 [6:47:38<4:29:07,  1.61it/s]
 60%|██████    | 39576/65536 [6:47:38<4:28:11,  1.61it/s]
 60%|██████    | 39577/65536 [6:47:39<4:39:27,  1.55it/s]
 60%|██████    | 39578/65536 [6:47:39<4:28:45,  1.61it/s]
 60%|██████    | 39579/65536 [6:47:40<4:27:44,  1.62it/s]
 60%|██████    | 39580/65536 [6:47:41<4:28:22,  1.61it/s]
                                                         
{'loss': 1.6009, 'learning_rate': 4.575425492837027e-07, 'epoch': 2443.21}
+
 60%|██████    | 39580/65536 [6:47:41<4:28:22,  1.61it/s]
 60%|██████    | 39581/65536 [6:47:41<4:24:10,  1.64it/s]
 60%|██████    | 39582/65536 [6:47:42<4:22:04,  1.65it/s]
 60%|██████    | 39583/65536 [6:47:42<4:25:03,  1.63it/s]
 60%|██████    | 39584/65536 [6:47:43<4:31:29,  1.59it/s]
 60%|██████    | 39585/65536 [6:47:44<4:28:26,  1.61it/s]
 60%|██████    | 39586/65536 [6:47:44<4:29:54,  1.60it/s]
 60%|██████    | 39587/65536 [6:47:45<4:22:25,  1.65it/s]
 60%|██████    | 39588/65536 [6:47:46<4:31:23,  1.59it/s]
 60%|██████    | 39589/65536 [6:47:46<4:25:55,  1.63it/s]
 60%|██████    | 39590/65536 [6:47:47<4:25:58,  1.63it/s]
 60%|██████    | 39591/65536 [6:47:47<4:21:30,  1.65it/s]
 60%|██████    | 39592/65536 [6:47:48<4:23:33,  1.64it/s]
 60%|██████    | 39593/65536 [6:47:49<4:31:06,  1.59it/s]
 60%|██████    | 39594/65536 [6:47:49<4:36:02,  1.57it/s]
 60%|██████    | 39595/65536 [6:47:50<4:33:33,  1.58it/s]
 60%|██████    | 39596/65536 [6:47:51<4:36:37,  1.56it/s]
 60%|██████    | 39597/65536 [6:47:51<4:31:23,  1.59it/s]
 60%|██████    | 39598/65536 [6:47:52<4:32:09,  1.59it/s]
 60%|██████    | 39599/65536 [6:47:52<4:25:16,  1.63it/s]
 60%|██████    | 39600/65536 [6:47:53<4:20:22,  1.66it/s]
                                                         
{'loss': 1.5824, 'learning_rate': 4.572670503244766e-07, 'epoch': 2444.44}
+
 60%|██████    | 39600/65536 [6:47:53<4:20:22,  1.66it/s]
 60%|██████    | 39601/65536 [6:47:54<4:18:36,  1.67it/s]
 60%|██████    | 39602/65536 [6:47:54<4:16:58,  1.68it/s]
 60%|██████    | 39603/65536 [6:47:55<4:13:56,  1.70it/s]
 60%|██████    | 39604/65536 [6:47:55<4:15:13,  1.69it/s]
 60%|██████    | 39605/65536 [6:47:56<4:25:33,  1.63it/s]
 60%|██████    | 39606/65536 [6:47:57<4:26:01,  1.62it/s]
 60%|██████    | 39607/65536 [6:47:57<4:24:02,  1.64it/s]
 60%|██████    | 39608/65536 [6:47:58<4:29:02,  1.61it/s]
 60%|██████    | 39609/65536 [6:47:58<4:22:37,  1.65it/s]
 60%|██████    | 39610/65536 [6:47:59<4:43:31,  1.52it/s]
 60%|██████    | 39611/65536 [6:48:00<4:37:03,  1.56it/s]
 60%|██████    | 39612/65536 [6:48:00<4:30:24,  1.60it/s]
 60%|██████    | 39613/65536 [6:48:01<4:22:00,  1.65it/s]
 60%|██████    | 39614/65536 [6:48:02<4:22:43,  1.64it/s]
 60%|██████    | 39615/65536 [6:48:02<4:21:25,  1.65it/s]
 60%|██████    | 39616/65536 [6:48:03<4:22:18,  1.65it/s]
 60%|██████    | 39617/65536 [6:48:04<4:30:42,  1.60it/s]
 60%|██████    | 39618/65536 [6:48:04<4:23:33,  1.64it/s]
 60%|██████    | 39619/65536 [6:48:05<4:20:35,  1.66it/s]
 60%|██████    | 39620/65536 [6:48:05<4:17:03,  1.68it/s]
                                                         
{'loss': 1.6549, 'learning_rate': 4.5699155136525033e-07, 'epoch': 2445.68}
+
 60%|██████    | 39620/65536 [6:48:05<4:17:03,  1.68it/s]
 60%|██████    | 39621/65536 [6:48:06<4:18:44,  1.67it/s]
 60%|██████    | 39622/65536 [6:48:07<4:26:39,  1.62it/s]
 60%|██████    | 39623/65536 [6:48:07<4:23:31,  1.64it/s]
 60%|██████    | 39624/65536 [6:48:08<4:18:43,  1.67it/s]
 60%|██████    | 39625/65536 [6:48:08<4:16:29,  1.68it/s]
 60%|██████    | 39626/65536 [6:48:09<4:28:58,  1.61it/s]
 60%|██████    | 39627/65536 [6:48:10<4:28:40,  1.61it/s]
 60%|██████    | 39628/65536 [6:48:10<4:27:31,  1.61it/s]
 60%|██████    | 39629/65536 [6:48:11<4:23:07,  1.64it/s]
 60%|██████    | 39630/65536 [6:48:11<4:31:37,  1.59it/s]
 60%|██████    | 39631/65536 [6:48:12<4:29:55,  1.60it/s]
 60%|██████    | 39632/65536 [6:48:13<4:26:00,  1.62it/s]
 60%|██████    | 39633/65536 [6:48:13<4:23:02,  1.64it/s]
 60%|██████    | 39634/65536 [6:48:14<4:26:52,  1.62it/s]
 60%|██████    | 39635/65536 [6:48:15<4:27:10,  1.62it/s]
 60%|██████    | 39636/65536 [6:48:15<4:27:50,  1.61it/s]
 60%|██████    | 39637/65536 [6:48:16<4:25:41,  1.62it/s]
 60%|██████    | 39638/65536 [6:48:16<4:23:18,  1.64it/s]
 60%|██████    | 39639/65536 [6:48:17<4:25:29,  1.63it/s]
 60%|██████    | 39640/65536 [6:48:18<4:25:28,  1.63it/s]
                                                         
{'loss': 1.636, 'learning_rate': 4.567160524060242e-07, 'epoch': 2446.91}
+
 60%|██████    | 39640/65536 [6:48:18<4:25:28,  1.63it/s]
 60%|██████    | 39641/65536 [6:48:18<4:20:18,  1.66it/s]
 60%|██████    | 39642/65536 [6:48:19<4:29:55,  1.60it/s]
 60%|██████    | 39643/65536 [6:48:19<4:32:54,  1.58it/s]
 60%|██████    | 39644/65536 [6:48:20<4:35:02,  1.57it/s]
 60%|██████    | 39645/65536 [6:48:21<4:33:00,  1.58it/s]
 60%|██████    | 39646/65536 [6:48:21<4:27:17,  1.61it/s]
 60%|██████    | 39647/65536 [6:48:22<4:22:05,  1.65it/s]
 60%|██████    | 39648/65536 [6:48:23<4:24:07,  1.63it/s]
 60%|██████    | 39649/65536 [6:48:23<4:22:15,  1.65it/s]
 61%|██████    | 39650/65536 [6:48:24<4:24:59,  1.63it/s]
 61%|██████    | 39651/65536 [6:48:24<4:24:04,  1.63it/s]
 61%|██████    | 39652/65536 [6:48:25<4:17:50,  1.67it/s]
 61%|██████    | 39653/65536 [6:48:26<4:18:53,  1.67it/s]
 61%|██████    | 39654/65536 [6:48:26<4:14:15,  1.70it/s]
 61%|██████    | 39655/65536 [6:48:27<4:13:45,  1.70it/s]
 61%|██████    | 39656/65536 [6:48:27<4:21:04,  1.65it/s]
 61%|██████    | 39657/65536 [6:48:28<4:19:18,  1.66it/s]
 61%|██████    | 39658/65536 [6:48:29<4:34:21,  1.57it/s]
 61%|██████    | 39659/65536 [6:48:29<4:28:32,  1.61it/s]
 61%|██████    | 39660/65536 [6:48:30<4:33:12,  1.58it/s]
                                                         
{'loss': 1.6012, 'learning_rate': 4.5644055344679804e-07, 'epoch': 2448.15}
+
 61%|██████    | 39660/65536 [6:48:30<4:33:12,  1.58it/s]
 61%|██████    | 39661/65536 [6:48:31<4:31:24,  1.59it/s]
 61%|██████    | 39662/65536 [6:48:31<4:27:48,  1.61it/s]
 61%|██████    | 39663/65536 [6:48:32<4:30:42,  1.59it/s]
 61%|██████    | 39664/65536 [6:48:32<4:34:20,  1.57it/s]
 61%|██████    | 39665/65536 [6:48:33<4:31:53,  1.59it/s]
 61%|██████    | 39666/65536 [6:48:34<4:25:15,  1.63it/s]
 61%|██████    | 39667/65536 [6:48:34<4:25:54,  1.62it/s]
 61%|██████    | 39668/65536 [6:48:35<4:21:36,  1.65it/s]
 61%|██████    | 39669/65536 [6:48:35<4:20:45,  1.65it/s]
 61%|██████    | 39670/65536 [6:48:36<4:22:42,  1.64it/s]
 61%|██████    | 39671/65536 [6:48:37<4:25:27,  1.62it/s]
 61%|██████    | 39672/65536 [6:48:37<4:21:36,  1.65it/s]
 61%|██████    | 39673/65536 [6:48:38<4:19:43,  1.66it/s]
 61%|██████    | 39674/65536 [6:48:39<4:32:10,  1.58it/s]
 61%|██████    | 39675/65536 [6:48:39<4:29:14,  1.60it/s]
 61%|██████    | 39676/65536 [6:48:40<4:29:51,  1.60it/s]
 61%|██████    | 39677/65536 [6:48:40<4:30:23,  1.59it/s]
 61%|██████    | 39678/65536 [6:48:41<4:27:16,  1.61it/s]
 61%|██████    | 39679/65536 [6:48:42<4:22:56,  1.64it/s]
 61%|██████    | 39680/65536 [6:48:42<4:20:15,  1.66it/s]
                                                         
{'loss': 1.6254, 'learning_rate': 4.561650544875719e-07, 'epoch': 2449.38}
+
 61%|██████    | 39680/65536 [6:48:42<4:20:15,  1.66it/s]
 61%|██████    | 39681/65536 [6:48:43<4:21:32,  1.65it/s]
 61%|██████    | 39682/65536 [6:48:43<4:25:05,  1.63it/s]
 61%|██████    | 39683/65536 [6:48:44<4:21:28,  1.65it/s]
 61%|██████    | 39684/65536 [6:48:45<4:18:57,  1.66it/s]
 61%|██████    | 39685/65536 [6:48:45<4:17:29,  1.67it/s]
 61%|██████    | 39686/65536 [6:48:46<4:21:34,  1.65it/s]
 61%|██████    | 39687/65536 [6:48:46<4:26:06,  1.62it/s]
 61%|██████    | 39688/65536 [6:48:47<4:32:03,  1.58it/s]
 61%|██████    | 39689/65536 [6:48:48<4:31:43,  1.59it/s]
 61%|██████    | 39690/65536 [6:48:48<4:28:34,  1.60it/s]
 61%|██████    | 39691/65536 [6:48:49<4:35:58,  1.56it/s]
 61%|██████    | 39692/65536 [6:48:50<4:30:57,  1.59it/s]
 61%|██████    | 39693/65536 [6:48:50<4:23:55,  1.63it/s]
 61%|██████    | 39694/65536 [6:48:51<4:29:40,  1.60it/s]
 61%|██████    | 39695/65536 [6:48:52<4:30:29,  1.59it/s]
 61%|██████    | 39696/65536 [6:48:52<4:34:53,  1.57it/s]
 61%|██████    | 39697/65536 [6:48:53<4:31:19,  1.59it/s]
 61%|██████    | 39698/65536 [6:48:53<4:27:22,  1.61it/s]
 61%|██████    | 39699/65536 [6:48:54<4:28:20,  1.60it/s]
 61%|██████    | 39700/65536 [6:48:55<4:28:27,  1.60it/s]
                                                         
{'loss': 1.5667, 'learning_rate': 4.5588955552834576e-07, 'epoch': 2450.62}
+
 61%|██████    | 39700/65536 [6:48:55<4:28:27,  1.60it/s]
 61%|██████    | 39701/65536 [6:48:55<4:23:10,  1.64it/s]
 61%|██████    | 39702/65536 [6:48:56<4:24:01,  1.63it/s]
 61%|██████    | 39703/65536 [6:48:56<4:24:18,  1.63it/s]
 61%|██████    | 39704/65536 [6:48:57<4:28:22,  1.60it/s]
 61%|██████    | 39705/65536 [6:48:58<4:19:10,  1.66it/s]
 61%|██████    | 39706/65536 [6:48:58<4:19:11,  1.66it/s]
 61%|██████    | 39707/65536 [6:48:59<4:31:42,  1.58it/s]
 61%|██████    | 39708/65536 [6:49:00<4:23:43,  1.63it/s]
 61%|██████    | 39709/65536 [6:49:00<4:23:50,  1.63it/s]
 61%|██████    | 39710/65536 [6:49:01<4:27:07,  1.61it/s]
 61%|██████    | 39711/65536 [6:49:01<4:23:15,  1.63it/s]
 61%|██████    | 39712/65536 [6:49:02<4:24:02,  1.63it/s]
 61%|██████    | 39713/65536 [6:49:03<4:17:20,  1.67it/s]
 61%|██████    | 39714/65536 [6:49:03<4:18:36,  1.66it/s]
 61%|██████    | 39715/65536 [6:49:04<4:24:17,  1.63it/s]
 61%|██████    | 39716/65536 [6:49:04<4:15:29,  1.68it/s]
 61%|██████    | 39717/65536 [6:49:05<4:16:58,  1.67it/s]
 61%|██████    | 39718/65536 [6:49:06<4:28:05,  1.61it/s]
 61%|██████    | 39719/65536 [6:49:06<4:31:20,  1.59it/s]
 61%|██████    | 39720/65536 [6:49:07<4:25:04,  1.62it/s]
                                                         
{'loss': 1.6614, 'learning_rate': 4.556140565691196e-07, 'epoch': 2451.85}
+
 61%|██████    | 39720/65536 [6:49:07<4:25:04,  1.62it/s]
 61%|██████    | 39721/65536 [6:49:07<4:25:13,  1.62it/s]
 61%|██████    | 39722/65536 [6:49:08<4:22:24,  1.64it/s]
 61%|██████    | 39723/65536 [6:49:09<4:28:47,  1.60it/s]
 61%|██████    | 39724/65536 [6:49:09<4:38:34,  1.54it/s]
 61%|██████    | 39725/65536 [6:49:10<4:32:42,  1.58it/s]
 61%|██████    | 39726/65536 [6:49:11<4:26:16,  1.62it/s]
 61%|██████    | 39727/65536 [6:49:11<4:28:35,  1.60it/s]
 61%|██████    | 39728/65536 [6:49:12<4:23:13,  1.63it/s]
 61%|██████    | 39729/65536 [6:49:12<4:24:31,  1.63it/s]
 61%|██████    | 39730/65536 [6:49:13<4:21:09,  1.65it/s]
 61%|██████    | 39731/65536 [6:49:14<4:18:46,  1.66it/s]
 61%|██████    | 39732/65536 [6:49:14<4:15:20,  1.68it/s]
 61%|██████    | 39733/65536 [6:49:15<4:20:03,  1.65it/s]
 61%|██████    | 39734/65536 [6:49:15<4:14:36,  1.69it/s]
 61%|██████    | 39735/65536 [6:49:16<4:18:00,  1.67it/s]
 61%|██████    | 39736/65536 [6:49:17<4:28:04,  1.60it/s]
 61%|██████    | 39737/65536 [6:49:17<4:23:22,  1.63it/s]
 61%|██████    | 39738/65536 [6:49:18<4:22:38,  1.64it/s]
 61%|██████    | 39739/65536 [6:49:19<4:28:28,  1.60it/s]
 61%|██████    | 39740/65536 [6:49:19<4:29:31,  1.60it/s]
                                                         
{'loss': 1.6136, 'learning_rate': 4.553385576098935e-07, 'epoch': 2453.09}
+
 61%|██████    | 39740/65536 [6:49:19<4:29:31,  1.60it/s]
 61%|██████    | 39741/65536 [6:49:20<4:23:49,  1.63it/s]
 61%|██████    | 39742/65536 [6:49:20<4:21:49,  1.64it/s]
 61%|██████    | 39743/65536 [6:49:21<4:18:32,  1.66it/s]
 61%|██████    | 39744/65536 [6:49:22<4:24:11,  1.63it/s]
 61%|██████    | 39745/65536 [6:49:22<4:28:16,  1.60it/s]
 61%|██████    | 39746/65536 [6:49:23<4:29:01,  1.60it/s]
 61%|██████    | 39747/65536 [6:49:23<4:25:56,  1.62it/s]
 61%|██████    | 39748/65536 [6:49:24<4:23:40,  1.63it/s]
 61%|██████    | 39749/65536 [6:49:25<4:28:22,  1.60it/s]
 61%|██████    | 39750/65536 [6:49:25<4:23:17,  1.63it/s]
 61%|██████    | 39751/65536 [6:49:26<4:20:28,  1.65it/s]
 61%|██████    | 39752/65536 [6:49:27<4:21:09,  1.65it/s]
 61%|██████    | 39753/65536 [6:49:27<4:20:08,  1.65it/s]
 61%|██████    | 39754/65536 [6:49:28<4:25:03,  1.62it/s]
 61%|██████    | 39755/65536 [6:49:28<4:30:13,  1.59it/s]
 61%|██████    | 39756/65536 [6:49:29<4:25:31,  1.62it/s]
 61%|██████    | 39757/65536 [6:49:30<4:20:11,  1.65it/s]
 61%|██████    | 39758/65536 [6:49:30<4:19:07,  1.66it/s]
 61%|██████    | 39759/65536 [6:49:31<4:24:23,  1.62it/s]
 61%|██████    | 39760/65536 [6:49:31<4:21:58,  1.64it/s]
                                                         
{'loss': 1.6299, 'learning_rate': 4.5506305865066723e-07, 'epoch': 2454.32}
+
 61%|██████    | 39760/65536 [6:49:31<4:21:58,  1.64it/s]
 61%|██████    | 39761/65536 [6:49:32<4:24:07,  1.63it/s]
 61%|██████    | 39762/65536 [6:49:33<4:28:20,  1.60it/s]
 61%|██████    | 39763/65536 [6:49:33<4:22:27,  1.64it/s]
 61%|██████    | 39764/65536 [6:49:34<4:21:10,  1.64it/s]
 61%|██████    | 39765/65536 [6:49:35<4:29:36,  1.59it/s]
 61%|██████    | 39766/65536 [6:49:35<4:27:35,  1.61it/s]
 61%|██████    | 39767/65536 [6:49:36<4:22:26,  1.64it/s]
 61%|██████    | 39768/65536 [6:49:36<4:27:46,  1.60it/s]
 61%|██████    | 39769/65536 [6:49:37<4:23:28,  1.63it/s]
 61%|██████    | 39770/65536 [6:49:38<4:19:42,  1.65it/s]
 61%|██████    | 39771/65536 [6:49:38<4:20:40,  1.65it/s]
 61%|██████    | 39772/65536 [6:49:39<4:35:18,  1.56it/s]
 61%|██████    | 39773/65536 [6:49:40<4:32:47,  1.57it/s]
 61%|██████    | 39774/65536 [6:49:40<4:37:57,  1.54it/s]
 61%|██████    | 39775/65536 [6:49:41<4:34:08,  1.57it/s]
 61%|██████    | 39776/65536 [6:49:41<4:21:01,  1.64it/s]
 61%|██████    | 39777/65536 [6:49:42<4:18:55,  1.66it/s]
 61%|██████    | 39778/65536 [6:49:43<4:18:58,  1.66it/s]
 61%|██████    | 39779/65536 [6:49:43<4:16:19,  1.67it/s]
 61%|██████    | 39780/65536 [6:49:44<4:26:25,  1.61it/s]
                                                         
{'loss': 1.5855, 'learning_rate': 4.547875596914411e-07, 'epoch': 2455.56}
+
 61%|██████    | 39780/65536 [6:49:44<4:26:25,  1.61it/s]
 61%|██████    | 39781/65536 [6:49:44<4:19:13,  1.66it/s]
 61%|██████    | 39782/65536 [6:49:45<4:21:06,  1.64it/s]
 61%|██████    | 39783/65536 [6:49:46<4:24:23,  1.62it/s]
 61%|██████    | 39784/65536 [6:49:46<4:27:32,  1.60it/s]
 61%|██████    | 39785/65536 [6:49:47<4:22:37,  1.63it/s]
 61%|██████    | 39786/65536 [6:49:47<4:24:38,  1.62it/s]
 61%|██████    | 39787/65536 [6:49:48<4:23:21,  1.63it/s]
 61%|██████    | 39788/65536 [6:49:49<4:28:36,  1.60it/s]
 61%|██████    | 39789/65536 [6:49:49<4:24:59,  1.62it/s]
 61%|██████    | 39790/65536 [6:49:50<4:23:48,  1.63it/s]
 61%|██████    | 39791/65536 [6:49:51<4:18:13,  1.66it/s]
 61%|██████    | 39792/65536 [6:49:51<4:27:36,  1.60it/s]
 61%|██████    | 39793/65536 [6:49:52<4:25:29,  1.62it/s]
 61%|██████    | 39794/65536 [6:49:52<4:18:41,  1.66it/s]
 61%|██████    | 39795/65536 [6:49:53<4:16:52,  1.67it/s]
 61%|██████    | 39796/65536 [6:49:54<4:25:32,  1.62it/s]
 61%|██████    | 39797/65536 [6:49:54<4:29:52,  1.59it/s]
 61%|██████    | 39798/65536 [6:49:55<4:26:40,  1.61it/s]
 61%|██████    | 39799/65536 [6:49:55<4:23:37,  1.63it/s]
 61%|██████    | 39800/65536 [6:49:56<4:26:33,  1.61it/s]
                                                         
{'loss': 1.6241, 'learning_rate': 4.5451206073221495e-07, 'epoch': 2456.79}
+
 61%|██████    | 39800/65536 [6:49:56<4:26:33,  1.61it/s]
 61%|██████    | 39801/65536 [6:49:57<4:19:36,  1.65it/s]
 61%|██████    | 39802/65536 [6:49:57<4:16:41,  1.67it/s]
 61%|██████    | 39803/65536 [6:49:58<4:20:53,  1.64it/s]
 61%|██████    | 39804/65536 [6:49:59<4:29:29,  1.59it/s]
 61%|██████    | 39805/65536 [6:49:59<4:34:02,  1.56it/s]
 61%|██████    | 39806/65536 [6:50:00<4:28:04,  1.60it/s]
 61%|██████    | 39807/65536 [6:50:00<4:22:12,  1.64it/s]
 61%|██████    | 39808/65536 [6:50:01<4:24:09,  1.62it/s]
 61%|██████    | 39809/65536 [6:50:02<4:23:03,  1.63it/s]
 61%|██████    | 39810/65536 [6:50:02<4:22:52,  1.63it/s]
 61%|██████    | 39811/65536 [6:50:03<4:22:18,  1.63it/s]
 61%|██████    | 39812/65536 [6:50:03<4:23:57,  1.62it/s]
 61%|██████    | 39813/65536 [6:50:04<4:24:13,  1.62it/s]
 61%|██████    | 39814/65536 [6:50:05<4:20:06,  1.65it/s]
 61%|██████    | 39815/65536 [6:50:05<4:19:14,  1.65it/s]
 61%|██████    | 39816/65536 [6:50:06<4:14:41,  1.68it/s]
 61%|██████    | 39817/65536 [6:50:07<4:20:34,  1.65it/s]
 61%|██████    | 39818/65536 [6:50:07<4:16:25,  1.67it/s]
 61%|██████    | 39819/65536 [6:50:08<4:13:58,  1.69it/s]
 61%|██████    | 39820/65536 [6:50:08<4:24:04,  1.62it/s]
                                                         
{'loss': 1.6204, 'learning_rate': 4.542365617729888e-07, 'epoch': 2458.02}
+
 61%|██████    | 39820/65536 [6:50:08<4:24:04,  1.62it/s]
 61%|██████    | 39821/65536 [6:50:09<4:28:49,  1.59it/s]
 61%|██████    | 39822/65536 [6:50:10<4:36:24,  1.55it/s]
 61%|██████    | 39823/65536 [6:50:10<4:24:55,  1.62it/s]
 61%|██████    | 39824/65536 [6:50:11<4:19:39,  1.65it/s]
 61%|██████    | 39825/65536 [6:50:11<4:20:53,  1.64it/s]
 61%|██████    | 39826/65536 [6:50:12<4:24:12,  1.62it/s]
 61%|██████    | 39827/65536 [6:50:13<4:22:07,  1.63it/s]
 61%|██████    | 39828/65536 [6:50:13<4:24:19,  1.62it/s]
 61%|██████    | 39829/65536 [6:50:14<4:23:53,  1.62it/s]
 61%|██████    | 39830/65536 [6:50:15<4:30:05,  1.59it/s]
 61%|██████    | 39831/65536 [6:50:15<4:23:29,  1.63it/s]
 61%|██████    | 39832/65536 [6:50:16<4:21:10,  1.64it/s]
 61%|██████    | 39833/65536 [6:50:16<4:21:57,  1.64it/s]
 61%|██████    | 39834/65536 [6:50:17<4:22:42,  1.63it/s]
 61%|██████    | 39835/65536 [6:50:18<4:19:34,  1.65it/s]
 61%|██████    | 39836/65536 [6:50:18<4:28:09,  1.60it/s]
 61%|██████    | 39837/65536 [6:50:19<4:32:21,  1.57it/s]
 61%|██████    | 39838/65536 [6:50:20<4:35:23,  1.56it/s]
 61%|██████    | 39839/65536 [6:50:20<4:33:41,  1.56it/s]
 61%|██████    | 39840/65536 [6:50:21<4:26:14,  1.61it/s]
                                                         
{'loss': 1.6197, 'learning_rate': 4.539610628137627e-07, 'epoch': 2459.26}
+
 61%|██████    | 39840/65536 [6:50:21<4:26:14,  1.61it/s]
 61%|██████    | 39841/65536 [6:50:21<4:21:29,  1.64it/s]
 61%|██████    | 39842/65536 [6:50:22<4:22:28,  1.63it/s]
 61%|██████    | 39843/65536 [6:50:23<4:17:53,  1.66it/s]
 61%|██████    | 39844/65536 [6:50:23<4:17:54,  1.66it/s]
 61%|██████    | 39845/65536 [6:50:24<4:14:34,  1.68it/s]
 61%|██████    | 39846/65536 [6:50:24<4:17:32,  1.66it/s]
 61%|██████    | 39847/65536 [6:50:25<4:14:20,  1.68it/s]
 61%|██████    | 39848/65536 [6:50:26<4:26:58,  1.60it/s]
 61%|██████    | 39849/65536 [6:50:26<4:27:12,  1.60it/s]
 61%|██████    | 39850/65536 [6:50:27<4:19:23,  1.65it/s]
 61%|██████    | 39851/65536 [6:50:27<4:17:59,  1.66it/s]
 61%|██████    | 39852/65536 [6:50:28<4:17:05,  1.67it/s]
 61%|██████    | 39853/65536 [6:50:29<4:27:55,  1.60it/s]
 61%|██████    | 39854/65536 [6:50:29<4:27:42,  1.60it/s]
 61%|██████    | 39855/65536 [6:50:30<4:22:22,  1.63it/s]
 61%|██████    | 39856/65536 [6:50:30<4:21:43,  1.64it/s]
 61%|██████    | 39857/65536 [6:50:31<4:17:48,  1.66it/s]
 61%|██████    | 39858/65536 [6:50:32<4:15:25,  1.68it/s]
 61%|██████    | 39859/65536 [6:50:32<4:17:53,  1.66it/s]
 61%|██████    | 39860/65536 [6:50:33<4:24:19,  1.62it/s]
                                                         
{'loss': 1.5769, 'learning_rate': 4.536855638545366e-07, 'epoch': 2460.49}
+
 61%|██████    | 39860/65536 [6:50:33<4:24:19,  1.62it/s]
 61%|██████    | 39861/65536 [6:50:34<4:28:05,  1.60it/s]
 61%|██████    | 39862/65536 [6:50:34<4:19:16,  1.65it/s]
 61%|██████    | 39863/65536 [6:50:35<4:30:11,  1.58it/s]
 61%|██████    | 39864/65536 [6:50:35<4:24:26,  1.62it/s]
 61%|██████    | 39865/65536 [6:50:36<4:27:12,  1.60it/s]
 61%|██████    | 39866/65536 [6:50:37<4:27:23,  1.60it/s]
 61%|██████    | 39867/65536 [6:50:37<4:21:27,  1.64it/s]
 61%|██████    | 39868/65536 [6:50:38<4:17:32,  1.66it/s]
 61%|██████    | 39869/65536 [6:50:39<4:27:31,  1.60it/s]
 61%|██████    | 39870/65536 [6:50:39<4:25:55,  1.61it/s]
 61%|██████    | 39871/65536 [6:50:40<4:22:40,  1.63it/s]
 61%|██████    | 39872/65536 [6:50:40<4:23:44,  1.62it/s]
 61%|██████    | 39873/65536 [6:50:41<4:23:25,  1.62it/s]
 61%|██████    | 39874/65536 [6:50:42<4:30:55,  1.58it/s]
 61%|██████    | 39875/65536 [6:50:42<4:29:05,  1.59it/s]
 61%|██████    | 39876/65536 [6:50:43<4:30:09,  1.58it/s]
 61%|██████    | 39877/65536 [6:50:44<4:30:32,  1.58it/s]
 61%|██████    | 39878/65536 [6:50:44<4:25:53,  1.61it/s]
 61%|██████    | 39879/65536 [6:50:45<4:22:37,  1.63it/s]
 61%|██████    | 39880/65536 [6:50:45<4:23:37,  1.62it/s]
                                                         
{'loss': 1.6498, 'learning_rate': 4.5341006489531044e-07, 'epoch': 2461.73}
+
 61%|██████    | 39880/65536 [6:50:45<4:23:37,  1.62it/s]
 61%|██████    | 39881/65536 [6:50:46<4:24:43,  1.62it/s]
 61%|██████    | 39882/65536 [6:50:47<4:21:39,  1.63it/s]
 61%|██████    | 39883/65536 [6:50:47<4:21:25,  1.64it/s]
 61%|██████    | 39884/65536 [6:50:48<4:19:59,  1.64it/s]
 61%|██████    | 39885/65536 [6:50:49<4:42:57,  1.51it/s]
 61%|██████    | 39886/65536 [6:50:49<4:42:05,  1.52it/s]
 61%|██████    | 39887/65536 [6:50:50<4:44:28,  1.50it/s]
 61%|██████    | 39888/65536 [6:50:50<4:32:09,  1.57it/s]
 61%|██████    | 39889/65536 [6:50:51<4:29:45,  1.58it/s]
 61%|██████    | 39890/65536 [6:50:52<4:25:54,  1.61it/s]
 61%|██████    | 39891/65536 [6:50:52<4:18:33,  1.65it/s]
 61%|██████    | 39892/65536 [6:50:53<4:20:07,  1.64it/s]
 61%|██████    | 39893/65536 [6:50:53<4:18:25,  1.65it/s]
 61%|██████    | 39894/65536 [6:50:54<4:27:18,  1.60it/s]
 61%|██████    | 39895/65536 [6:50:55<4:20:35,  1.64it/s]
 61%|██████    | 39896/65536 [6:50:55<4:24:48,  1.61it/s]
 61%|██████    | 39897/65536 [6:50:56<4:22:20,  1.63it/s]
 61%|██████    | 39898/65536 [6:50:57<4:23:11,  1.62it/s]
 61%|██████    | 39899/65536 [6:50:57<4:19:37,  1.65it/s]
 61%|██████    | 39900/65536 [6:50:58<4:20:23,  1.64it/s]
                                                         
{'loss': 1.59, 'learning_rate': 4.531345659360842e-07, 'epoch': 2462.96}
+
 61%|██████    | 39900/65536 [6:50:58<4:20:23,  1.64it/s]
 61%|██████    | 39901/65536 [6:50:58<4:26:44,  1.60it/s]
 61%|██████    | 39902/65536 [6:50:59<4:28:23,  1.59it/s]
 61%|██████    | 39903/65536 [6:51:00<4:21:30,  1.63it/s]
 61%|██████    | 39904/65536 [6:51:00<4:19:27,  1.65it/s]
 61%|██████    | 39905/65536 [6:51:01<4:23:35,  1.62it/s]
 61%|██████    | 39906/65536 [6:51:02<4:27:00,  1.60it/s]
 61%|██████    | 39907/65536 [6:51:02<4:27:16,  1.60it/s]
 61%|██████    | 39908/65536 [6:51:03<4:20:48,  1.64it/s]
 61%|██████    | 39909/65536 [6:51:03<4:20:07,  1.64it/s]
 61%|██████    | 39910/65536 [6:51:04<4:12:58,  1.69it/s]
 61%|██████    | 39911/65536 [6:51:05<4:18:55,  1.65it/s]
 61%|██████    | 39912/65536 [6:51:05<4:15:01,  1.67it/s]
 61%|██████    | 39913/65536 [6:51:06<4:20:34,  1.64it/s]
 61%|██████    | 39914/65536 [6:51:06<4:20:25,  1.64it/s]
 61%|██████    | 39915/65536 [6:51:07<4:16:00,  1.67it/s]
 61%|██████    | 39916/65536 [6:51:08<4:15:04,  1.67it/s]
 61%|██████    | 39917/65536 [6:51:08<4:19:33,  1.65it/s]
 61%|██████    | 39918/65536 [6:51:09<4:16:07,  1.67it/s]
 61%|██████    | 39919/65536 [6:51:09<4:12:14,  1.69it/s]
 61%|██████    | 39920/65536 [6:51:10<4:14:40,  1.68it/s]
                                                         
{'loss': 1.6261, 'learning_rate': 4.5285906697685805e-07, 'epoch': 2464.2}
+
 61%|██████    | 39920/65536 [6:51:10<4:14:40,  1.68it/s]
 61%|██████    | 39921/65536 [6:51:10<4:12:43,  1.69it/s]
 61%|██████    | 39922/65536 [6:51:11<4:07:19,  1.73it/s]
 61%|██████    | 39923/65536 [6:51:12<4:07:37,  1.72it/s]
 61%|██████    | 39924/65536 [6:51:12<4:10:22,  1.70it/s]
 61%|██████    | 39925/65536 [6:51:13<4:19:39,  1.64it/s]
 61%|██████    | 39926/65536 [6:51:13<4:17:41,  1.66it/s]
 61%|██████    | 39927/65536 [6:51:14<4:18:04,  1.65it/s]
 61%|██████    | 39928/65536 [6:51:15<4:20:51,  1.64it/s]
 61%|██████    | 39929/65536 [6:51:15<4:22:23,  1.63it/s]
 61%|██████    | 39930/65536 [6:51:16<4:21:59,  1.63it/s]
 61%|██████    | 39931/65536 [6:51:17<4:26:47,  1.60it/s]
 61%|██████    | 39932/65536 [6:51:17<4:27:09,  1.60it/s]
 61%|██████    | 39933/65536 [6:51:18<4:25:33,  1.61it/s]
 61%|██████    | 39934/65536 [6:51:18<4:28:28,  1.59it/s]
 61%|██████    | 39935/65536 [6:51:19<4:22:40,  1.62it/s]
 61%|██████    | 39936/65536 [6:51:20<4:26:04,  1.60it/s]
 61%|██████    | 39937/65536 [6:51:20<4:16:11,  1.67it/s]
 61%|██████    | 39938/65536 [6:51:21<4:24:33,  1.61it/s]
 61%|██████    | 39939/65536 [6:51:22<4:31:46,  1.57it/s]
 61%|██████    | 39940/65536 [6:51:22<4:28:35,  1.59it/s]
                                                         
{'loss': 1.6421, 'learning_rate': 4.525835680176319e-07, 'epoch': 2465.43}
+
 61%|██████    | 39940/65536 [6:51:22<4:28:35,  1.59it/s]
 61%|██████    | 39941/65536 [6:51:23<4:29:52,  1.58it/s]
 61%|██████    | 39942/65536 [6:51:23<4:24:54,  1.61it/s]
 61%|██████    | 39943/65536 [6:51:24<4:20:42,  1.64it/s]
 61%|██████    | 39944/65536 [6:51:25<4:25:10,  1.61it/s]
 61%|██████    | 39945/65536 [6:51:25<4:21:03,  1.63it/s]
 61%|██████    | 39946/65536 [6:51:26<4:21:45,  1.63it/s]
 61%|██████    | 39947/65536 [6:51:27<4:25:53,  1.60it/s]
 61%|██████    | 39948/65536 [6:51:27<4:21:49,  1.63it/s]
 61%|██████    | 39949/65536 [6:51:28<4:26:30,  1.60it/s]
 61%|██████    | 39950/65536 [6:51:28<4:31:51,  1.57it/s]
 61%|██████    | 39951/65536 [6:51:29<4:29:37,  1.58it/s]
 61%|██████    | 39952/65536 [6:51:30<4:37:52,  1.53it/s]
 61%|██████    | 39953/65536 [6:51:30<4:24:15,  1.61it/s]
 61%|██████    | 39954/65536 [6:51:31<4:24:32,  1.61it/s]
 61%|██████    | 39955/65536 [6:51:32<4:28:13,  1.59it/s]
 61%|██████    | 39956/65536 [6:51:32<4:23:59,  1.61it/s]
 61%|██████    | 39957/65536 [6:51:33<4:22:37,  1.62it/s]
 61%|██████    | 39958/65536 [6:51:33<4:20:26,  1.64it/s]
 61%|██████    | 39959/65536 [6:51:34<4:30:39,  1.58it/s]
 61%|██████    | 39960/65536 [6:51:35<4:39:23,  1.53it/s]
                                                         
{'loss': 1.5835, 'learning_rate': 4.5230806905840577e-07, 'epoch': 2466.67}
+
 61%|██████    | 39960/65536 [6:51:35<4:39:23,  1.53it/s]
 61%|██████    | 39961/65536 [6:51:35<4:27:17,  1.59it/s]
 61%|██████    | 39962/65536 [6:51:36<4:22:27,  1.62it/s]
 61%|██████    | 39963/65536 [6:51:37<4:24:56,  1.61it/s]
 61%|██████    | 39964/65536 [6:51:37<4:18:33,  1.65it/s]
 61%|██████    | 39965/65536 [6:51:38<4:17:03,  1.66it/s]
 61%|██████    | 39966/65536 [6:51:38<4:26:48,  1.60it/s]
 61%|██████    | 39967/65536 [6:51:39<4:26:03,  1.60it/s]
 61%|██████    | 39968/65536 [6:51:40<4:28:32,  1.59it/s]
 61%|██████    | 39969/65536 [6:51:40<4:25:17,  1.61it/s]
 61%|██████    | 39970/65536 [6:51:41<4:17:19,  1.66it/s]
 61%|██████    | 39971/65536 [6:51:41<4:15:31,  1.67it/s]
 61%|██████    | 39972/65536 [6:51:42<4:21:20,  1.63it/s]
 61%|██████    | 39973/65536 [6:51:43<4:25:54,  1.60it/s]
 61%|██████    | 39974/65536 [6:51:43<4:22:49,  1.62it/s]
 61%|██████    | 39975/65536 [6:51:44<4:19:20,  1.64it/s]
 61%|██████    | 39976/65536 [6:51:44<4:16:21,  1.66it/s]
 61%|██████    | 39977/65536 [6:51:45<4:15:50,  1.67it/s]
 61%|██████    | 39978/65536 [6:51:46<4:16:37,  1.66it/s]
 61%|██████    | 39979/65536 [6:51:46<4:24:58,  1.61it/s]
 61%|██████    | 39980/65536 [6:51:47<4:27:38,  1.59it/s]
                                                         
{'loss': 1.5877, 'learning_rate': 4.5203257009917963e-07, 'epoch': 2467.9}
+
 61%|██████    | 39980/65536 [6:51:47<4:27:38,  1.59it/s]
 61%|██████    | 39981/65536 [6:51:48<4:18:50,  1.65it/s]
 61%|██████    | 39982/65536 [6:51:48<4:29:24,  1.58it/s]
 61%|██████    | 39983/65536 [6:51:49<4:28:33,  1.59it/s]
 61%|██████    | 39984/65536 [6:51:49<4:19:52,  1.64it/s]
 61%|██████    | 39985/65536 [6:51:50<4:18:14,  1.65it/s]
 61%|██████    | 39986/65536 [6:51:51<4:19:25,  1.64it/s]
 61%|██████    | 39987/65536 [6:51:51<4:21:56,  1.63it/s]
 61%|██████    | 39988/65536 [6:51:52<4:11:19,  1.69it/s]
 61%|██████    | 39989/65536 [6:51:52<4:16:02,  1.66it/s]
 61%|██████    | 39990/65536 [6:51:53<4:28:31,  1.59it/s]
 61%|██████    | 39991/65536 [6:51:54<4:31:57,  1.57it/s]
 61%|██████    | 39992/65536 [6:51:54<4:36:25,  1.54it/s]
 61%|██████    | 39993/65536 [6:51:55<4:30:34,  1.57it/s]
 61%|██████    | 39994/65536 [6:51:56<4:32:20,  1.56it/s]
 61%|██████    | 39995/65536 [6:51:56<4:26:37,  1.60it/s]
 61%|██████    | 39996/65536 [6:51:57<4:27:47,  1.59it/s]
 61%|██████    | 39997/65536 [6:51:58<4:26:01,  1.60it/s]
 61%|██████    | 39998/65536 [6:51:58<4:31:01,  1.57it/s]
 61%|██████    | 39999/65536 [6:51:59<4:28:53,  1.58it/s]
 61%|██████    | 40000/65536 [6:51:59<4:27:53,  1.59it/s]
                                                         
{'loss': 1.6351, 'learning_rate': 4.517570711399535e-07, 'epoch': 2469.14}
+
 61%|██████    | 40000/65536 [6:51:59<4:27:53,  1.59it/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}
+
 61%|██████    | 40001/65536 [6:52:17<41:24:24,  5.84s/it]
 61%|██████    | 40002/65536 [6:52:18<30:15:12,  4.27s/it]
 61%|██████    | 40003/65536 [6:52:19<22:31:02,  3.17s/it]
 61%|██████    | 40004/65536 [6:52:19<17:01:54,  2.40s/it]
 61%|██████    | 40005/65536 [6:52:20<13:16:31,  1.87s/it]
 61%|██████    | 40006/65536 [6:52:21<10:35:59,  1.49s/it]
 61%|██████    | 40007/65536 [6:52:21<8:41:49,  1.23s/it] 
 61%|██████    | 40008/65536 [6:52:22<7:23:18,  1.04s/it]
 61%|██████    | 40009/65536 [6:52:22<6:30:55,  1.09it/s]
 61%|██████    | 40010/65536 [6:52:23<5:54:33,  1.20it/s]
 61%|██████    | 40011/65536 [6:52:24<5:37:34,  1.26it/s]
 61%|██████    | 40012/65536 [6:52:24<5:18:45,  1.33it/s]
 61%|██████    | 40013/65536 [6:52:25<5:08:41,  1.38it/s]
 61%|██████    | 40014/65536 [6:52:26<5:01:51,  1.41it/s]/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()
+
 61%|██████    | 40015/65536 [6:52:26<5:04:04,  1.40it/s]
 61%|██████    | 40016/65536 [6:52:27<4:57:24,  1.43it/s]
 61%|██████    | 40017/65536 [6:52:28<5:06:21,  1.39it/s]
 61%|██████    | 40018/65536 [6:52:29<5:02:25,  1.41it/s]
 61%|██████    | 40019/65536 [6:52:29<4:58:00,  1.43it/s]
 61%|██████    | 40020/65536 [6:52:30<4:56:09,  1.44it/s]
                                                         
{'loss': 1.5755, 'learning_rate': 4.5148157218072735e-07, 'epoch': 2470.37}
+
 61%|██████    | 40020/65536 [6:52:30<4:56:09,  1.44it/s]
 61%|██████    | 40021/65536 [6:52:31<4:56:47,  1.43it/s]
 61%|██████    | 40022/65536 [6:52:31<4:43:33,  1.50it/s]
 61%|██████    | 40023/65536 [6:52:32<4:41:32,  1.51it/s]
 61%|██████    | 40024/65536 [6:52:32<4:34:44,  1.55it/s]
 61%|██████    | 40025/65536 [6:52:33<4:35:02,  1.55it/s]
 61%|██████    | 40026/65536 [6:52:34<4:35:55,  1.54it/s]
 61%|██████    | 40027/65536 [6:52:34<4:35:35,  1.54it/s]
 61%|██████    | 40028/65536 [6:52:35<4:32:31,  1.56it/s]
 61%|██████    | 40029/65536 [6:52:36<4:36:55,  1.54it/s]
 61%|██████    | 40030/65536 [6:52:36<4:29:32,  1.58it/s]
 61%|██████    | 40031/65536 [6:52:37<4:39:52,  1.52it/s]
 61%|██████    | 40032/65536 [6:52:38<4:31:26,  1.57it/s]
 61%|██████    | 40033/65536 [6:52:38<4:26:56,  1.59it/s]
 61%|██████    | 40034/65536 [6:52:39<4:36:03,  1.54it/s]
 61%|██████    | 40035/65536 [6:52:40<4:38:25,  1.53it/s]
 61%|██████    | 40036/65536 [6:52:40<4:32:37,  1.56it/s]
 61%|██████    | 40037/65536 [6:52:41<4:29:07,  1.58it/s]
 61%|██████    | 40038/65536 [6:52:41<4:33:08,  1.56it/s]
 61%|██████    | 40039/65536 [6:52:42<4:32:40,  1.56it/s]
 61%|██████    | 40040/65536 [6:52:43<4:29:38,  1.58it/s]
                                                         
{'loss': 1.6251, 'learning_rate': 4.512060732215011e-07, 'epoch': 2471.6}
+
 61%|██████    | 40040/65536 [6:52:43<4:29:38,  1.58it/s]
 61%|██████    | 40041/65536 [6:52:43<4:30:42,  1.57it/s]
 61%|██████    | 40042/65536 [6:52:44<4:22:56,  1.62it/s]
 61%|██████    | 40043/65536 [6:52:45<4:28:29,  1.58it/s]
 61%|██████    | 40044/65536 [6:52:45<4:28:45,  1.58it/s]
 61%|██████    | 40045/65536 [6:52:46<4:30:49,  1.57it/s]
 61%|██████    | 40046/65536 [6:52:47<4:33:57,  1.55it/s]
 61%|██████    | 40047/65536 [6:52:47<4:41:38,  1.51it/s]
 61%|██████    | 40048/65536 [6:52:48<4:36:07,  1.54it/s]
 61%|██████    | 40049/65536 [6:52:49<4:37:46,  1.53it/s]
 61%|██████    | 40050/65536 [6:52:49<4:34:19,  1.55it/s]
 61%|██████    | 40051/65536 [6:52:50<4:33:30,  1.55it/s]
 61%|██████    | 40052/65536 [6:52:51<4:37:47,  1.53it/s]
 61%|██████    | 40053/65536 [6:52:51<4:39:10,  1.52it/s]
 61%|██████    | 40054/65536 [6:52:52<4:39:48,  1.52it/s]
 61%|██████    | 40055/65536 [6:52:53<4:40:53,  1.51it/s]
 61%|██████    | 40056/65536 [6:52:53<4:37:57,  1.53it/s]
 61%|██████    | 40057/65536 [6:52:54<4:30:36,  1.57it/s]
 61%|██████    | 40058/65536 [6:52:54<4:27:36,  1.59it/s]
 61%|██████    | 40059/65536 [6:52:55<4:28:56,  1.58it/s]
 61%|██████    | 40060/65536 [6:52:56<4:28:47,  1.58it/s]
                                                         
{'loss': 1.6108, 'learning_rate': 4.5093057426227496e-07, 'epoch': 2472.84}
+
 61%|██████    | 40060/65536 [6:52:56<4:28:47,  1.58it/s]
 61%|██████    | 40061/65536 [6:52:56<4:32:00,  1.56it/s]
 61%|██████    | 40062/65536 [6:52:57<4:33:06,  1.55it/s]
 61%|██████    | 40063/65536 [6:52:58<4:43:18,  1.50it/s]
 61%|██████    | 40064/65536 [6:52:58<4:40:05,  1.52it/s]
\ No newline at end of file