forked from anwesh44/gpt-neo-fine-tuning-example
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathtraining_and_results_gpt_neo_27.txt
732 lines (728 loc) · 86.1 KB
/
training_and_results_gpt_neo_27.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
[2022-01-08 01:38:49,947] [INFO] [distributed.py:46:init_distributed] Initializing torch distributed with backend: nccl
Max length: 62
Using amp half precision backend
[2022-01-08 01:39:12,781] [INFO] [logging.py:69:log_dist] [Rank 0] DeepSpeed info: version=0.5.9+d0ab722, git-hash=d0ab722, git-branch=master
[2022-01-08 01:39:12,789] [INFO] [logging.py:69:log_dist] [Rank 0] initializing deepspeed groups
[2022-01-08 01:39:12,789] [INFO] [logging.py:69:log_dist] [Rank 0] initializing deepspeed model parallel group with size 1
[2022-01-08 01:39:12,790] [INFO] [logging.py:69:log_dist] [Rank 0] initializing deepspeed expert parallel group with size 1
[2022-01-08 01:39:12,790] [INFO] [logging.py:69:log_dist] [Rank 0] creating expert data parallel process group with ranks: [0]
[2022-01-08 01:39:12,790] [INFO] [logging.py:69:log_dist] [Rank 0] creating expert parallel process group with ranks: [0]
[2022-01-08 01:39:12,818] [INFO] [engine.py:277:__init__] DeepSpeed Flops Profiler Enabled: False
Adam Optimizer #0 is created with AVX2 arithmetic capability.
Config: alpha=0.000050, betas=(0.900000, 0.999000), weight_decay=0.000000, adam_w=1
[2022-01-08 01:39:12,994] [INFO] [engine.py:1107:_configure_optimizer] Using DeepSpeed Optimizer param name adamw as basic optimizer
[2022-01-08 01:39:13,010] [INFO] [engine.py:1115:_configure_optimizer] DeepSpeed Basic Optimizer = DeepSpeedCPUAdam
[2022-01-08 01:39:13,010] [INFO] [utils.py:43:is_zero_supported_optimizer] Checking ZeRO support for optimizer=DeepSpeedCPUAdam type=<class 'deepspeed.ops.adam.cpu_adam.DeepSpeedCPUAdam'>
[2022-01-08 01:39:13,010] [INFO] [logging.py:69:log_dist] [Rank 0] Creating fp16 ZeRO stage 2 optimizer
[2022-01-08 01:39:13,010] [INFO] [stage_1_and_2.py:113:__init__] Reduce bucket size 500000000
[2022-01-08 01:39:13,010] [INFO] [stage_1_and_2.py:114:__init__] Allgather bucket size 500000000.0
[2022-01-08 01:39:13,010] [INFO] [stage_1_and_2.py:115:__init__] CPU Offload: True
[2022-01-08 01:39:13,010] [INFO] [stage_1_and_2.py:116:__init__] Round robin gradient partitioning: False
Rank: 0 partition count [1] and sizes[(2651312640, False)]
[2022-01-08 01:39:18,093] [INFO] [utils.py:822:see_memory_usage] Before initializing optimizer states
[2022-01-08 01:39:18,093] [INFO] [utils.py:823:see_memory_usage] MA 5.3 GB Max_MA 10.61 GB CA 15.56 GB Max_CA 16 GB
[2022-01-08 01:39:18,093] [INFO] [utils.py:831:see_memory_usage] CPU Virtual Memory: used = 23.39 GB, percent = 18.6%
[2022-01-08 01:39:24,336] [INFO] [utils.py:822:see_memory_usage] After initializing optimizer states
[2022-01-08 01:39:24,336] [INFO] [utils.py:823:see_memory_usage] MA 5.3 GB Max_MA 5.3 GB CA 15.56 GB Max_CA 16 GB
[2022-01-08 01:39:24,336] [INFO] [utils.py:831:see_memory_usage] CPU Virtual Memory: used = 53.27 GB, percent = 42.4%
[2022-01-08 01:39:24,336] [INFO] [stage_1_and_2.py:483:__init__] optimizer state initialized
[2022-01-08 01:39:24,354] [INFO] [utils.py:822:see_memory_usage] After initializing ZeRO optimizer
[2022-01-08 01:39:24,354] [INFO] [utils.py:823:see_memory_usage] MA 5.3 GB Max_MA 5.3 GB CA 15.56 GB Max_CA 16 GB
[2022-01-08 01:39:24,354] [INFO] [utils.py:831:see_memory_usage] CPU Virtual Memory: used = 53.27 GB, percent = 42.4%
[2022-01-08 01:39:24,354] [INFO] [logging.py:69:log_dist] [Rank 0] DeepSpeed Final Optimizer = adamw
[2022-01-08 01:39:24,354] [INFO] [engine.py:797:_configure_lr_scheduler] DeepSpeed using configured LR scheduler = WarmupLR
[2022-01-08 01:39:24,354] [INFO] [logging.py:69:log_dist] [Rank 0] DeepSpeed LR Scheduler = <deepspeed.runtime.lr_schedules.WarmupLR object at 0x7fdc2afb6580>
[2022-01-08 01:39:24,354] [INFO] [logging.py:69:log_dist] [Rank 0] step=0, skipped=0, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:39:24,355] [INFO] [config.py:1058:print] DeepSpeedEngine configuration:
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] activation_checkpointing_config {
"partition_activations": false,
"contiguous_memory_optimization": false,
"cpu_checkpointing": false,
"number_checkpoints": null,
"synchronize_checkpoint_boundary": false,
"profile": false
}
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] aio_config ................... {'block_size': 1048576, 'queue_depth': 8, 'thread_count': 1, 'single_submit': False, 'overlap_events': True}
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] amp_enabled .................. False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] amp_params ................... False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] autotuning_config ............ {
"enabled": false,
"start_step": null,
"end_step": null,
"metric_path": null,
"arg_mappings": null,
"metric": "throughput",
"model_info": null,
"results_dir": null,
"exps_dir": null,
"overwrite": true,
"fast": true,
"start_profile_step": 3,
"end_profile_step": 5,
"tuner_type": "gridsearch",
"tuner_early_stopping": 5,
"tuner_num_trials": 50,
"model_info_path": null,
"mp_size": 1,
"max_train_batch_size": null,
"min_train_batch_size": 1,
"max_train_micro_batch_size_per_gpu": 1.024000e+03,
"min_train_micro_batch_size_per_gpu": 1,
"num_tuning_micro_batch_sizes": 3
}
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] bfloat16_enabled ............. False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] checkpoint_tag_validation_enabled True
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] checkpoint_tag_validation_fail False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] communication_data_type ...... None
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] curriculum_enabled ........... False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] curriculum_params ............ False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] dataloader_drop_last ......... False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] disable_allgather ............ False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] dump_state ................... False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] dynamic_loss_scale_args ...... {'init_scale': 4294967296, 'scale_window': 1000, 'delayed_shift': 2, 'min_scale': 1}
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] eigenvalue_enabled ........... False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] eigenvalue_gas_boundary_resolution 1
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] eigenvalue_layer_name ........ bert.encoder.layer
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] eigenvalue_layer_num ......... 0
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] eigenvalue_max_iter .......... 100
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] eigenvalue_stability ......... 1e-06
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] eigenvalue_tol ............... 0.01
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] eigenvalue_verbose ........... False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] elasticity_enabled ........... False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] flops_profiler_config ........ {
"enabled": false,
"profile_step": 1,
"module_depth": -1,
"top_modules": 1,
"detailed": true,
"output_file": null
}
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] fp16_enabled ................. True
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] fp16_master_weights_and_gradients False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] fp16_mixed_quantize .......... False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] global_rank .................. 0
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] gradient_accumulation_steps .. 1
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] gradient_clipping ............ 0.0
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] gradient_predivide_factor .... 1.0
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] initial_dynamic_scale ........ 4294967296
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] loss_scale ................... 0
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] memory_breakdown ............. False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] optimizer_legacy_fusion ...... False
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] optimizer_name ............... adamw
[2022-01-08 01:39:24,355] [INFO] [config.py:1062:print] optimizer_params ............. {'lr': 5e-05, 'betas': [0.9, 0.999], 'eps': 1e-08}
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] pipeline ..................... {'stages': 'auto', 'partition': 'best', 'seed_layers': False, 'activation_checkpoint_interval': 0}
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] pld_enabled .................. False
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] pld_params ................... False
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] prescale_gradients ........... False
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] quantize_change_rate ......... 0.001
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] quantize_groups .............. 1
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] quantize_offset .............. 1000
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] quantize_period .............. 1000
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] quantize_rounding ............ 0
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] quantize_start_bits .......... 16
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] quantize_target_bits ......... 8
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] quantize_training_enabled .... False
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] quantize_type ................ 0
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] quantize_verbose ............. False
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] scheduler_name ............... WarmupLR
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] scheduler_params ............. {'warmup_min_lr': 0, 'warmup_max_lr': 5e-05, 'warmup_num_steps': 50}
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] sparse_attention ............. None
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] sparse_gradients_enabled ..... False
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] steps_per_print .............. 10
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] tensorboard_enabled .......... False
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] tensorboard_job_name ......... DeepSpeedJobName
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] tensorboard_output_path ......
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] train_batch_size ............. 15
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] train_micro_batch_size_per_gpu 15
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] use_quantizer_kernel ......... False
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] wall_clock_breakdown ......... False
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] world_size ................... 1
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] zero_allow_untested_optimizer False
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] zero_config .................. {
"stage": 2,
"contiguous_gradients": true,
"reduce_scatter": true,
"reduce_bucket_size": 5.000000e+08,
"allgather_partitions": true,
"allgather_bucket_size": 5.000000e+08,
"overlap_comm": false,
"load_from_fp32_weights": true,
"elastic_checkpoint": true,
"offload_param": {
"device": "cpu",
"nvme_path": null,
"buffer_count": 5,
"buffer_size": 1.000000e+08,
"max_in_cpu": 1.000000e+09,
"pin_memory": false
},
"offload_optimizer": {
"device": "cpu",
"nvme_path": null,
"buffer_count": 4,
"pin_memory": false,
"pipeline_read": false,
"pipeline_write": false,
"fast_init": false,
"pipeline": false
},
"sub_group_size": 1.000000e+09,
"prefetch_bucket_size": 5.000000e+07,
"param_persistence_threshold": 1.000000e+05,
"max_live_parameters": 1.000000e+09,
"max_reuse_distance": 1.000000e+09,
"gather_fp16_weights_on_model_save": false,
"ignore_unused_parameters": true,
"round_robin_gradients": false,
"legacy_stage1": false
}
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] zero_enabled ................. True
[2022-01-08 01:39:24,356] [INFO] [config.py:1062:print] zero_optimization_stage ...... 2
[2022-01-08 01:39:24,356] [INFO] [config.py:1064:print] json = {
"train_batch_size": 15,
"fp16": {
"enabled": true,
"min_loss_scale": 1,
"opt_level": "O2"
},
"zero_optimization": {
"stage": 2,
"offload_param": {
"device": "cpu"
},
"offload_optimizer": {
"device": "cpu"
},
"allgather_partitions": true,
"allgather_bucket_size": 5.000000e+08,
"contiguous_gradients": true
},
"optimizer": {
"type": "AdamW",
"params": {
"lr": 5e-05,
"betas": [0.9, 0.999],
"eps": 1e-08
}
},
"scheduler": {
"type": "WarmupLR",
"params": {
"warmup_min_lr": 0,
"warmup_max_lr": 5e-05,
"warmup_num_steps": 50
}
}
}
***** Running training *****
Num examples = 7008
Num Epochs = 5
Instantaneous batch size per device = 15
Total train batch size (w. parallel, distributed & accumulation) = 15
Gradient Accumulation steps = 1
Total optimization steps = 2013
0%| | 1/2013 [00:00<28:48, 1.16it/s][2022-01-08 01:39:25,216] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 4294967296, reducing to 4294967296
[2022-01-08 01:39:26,056] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 4294967296, reducing to 2147483648.0
0%| | 3/2013 [00:02<28:18, 1.18it/s][2022-01-08 01:39:26,898] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 2147483648.0, reducing to 1073741824.0
[2022-01-08 01:39:27,745] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 1073741824.0, reducing to 536870912.0
0%| | 5/2013 [00:04<28:17, 1.18it/s][2022-01-08 01:39:28,589] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 536870912.0, reducing to 268435456.0
0%| | 6/2013 [00:05<28:15, 1.18it/s][2022-01-08 01:39:29,433] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 268435456.0, reducing to 134217728.0
[2022-01-08 01:39:30,288] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 134217728.0, reducing to 67108864.0
0%| | 8/2013 [00:06<28:57, 1.15it/s][2022-01-08 01:39:31,194] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 67108864.0, reducing to 33554432.0
0%| | 9/2013 [00:07<29:15, 1.14it/s][2022-01-08 01:39:32,090] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 33554432.0, reducing to 16777216.0
[2022-01-08 01:39:32,936] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 16777216.0, reducing to 8388608.0
[2022-01-08 01:39:32,936] [INFO] [logging.py:69:log_dist] [Rank 0] step=10, skipped=10, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:39:32,936] [INFO] [timer.py:181:stop] 0/10, SamplesPerSec=17.476337057468704
1%| | 11/2013 [00:09<28:44, 1.16it/s][2022-01-08 01:39:33,786] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 8388608.0, reducing to 4194304.0
[2022-01-08 01:39:34,633] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 4194304.0, reducing to 2097152.0
1%| | 13/2013 [00:11<28:27, 1.17it/s][2022-01-08 01:39:35,480] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 2097152.0, reducing to 1048576.0
[2022-01-08 01:39:36,326] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 1048576.0, reducing to 524288.0
1%| | 15/2013 [00:12<28:18, 1.18it/s][2022-01-08 01:39:37,173] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 524288.0, reducing to 262144.0
[2022-01-08 01:39:38,017] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 262144.0, reducing to 131072.0
1%| | 17/2013 [00:14<28:15, 1.18it/s][2022-01-08 01:39:38,869] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 131072.0, reducing to 65536.0
[2022-01-08 01:39:39,719] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 65536.0, reducing to 32768.0
1%| | 19/2013 [00:16<28:14, 1.18it/s][2022-01-08 01:39:40,569] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 32768.0, reducing to 16384.0
[2022-01-08 01:39:41,419] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 16384.0, reducing to 8192.0
[2022-01-08 01:39:41,420] [INFO] [logging.py:69:log_dist] [Rank 0] step=20, skipped=20, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:39:41,420] [INFO] [timer.py:181:stop] 0/20, SamplesPerSec=17.6068113915733
1%| | 21/2013 [00:17<28:14, 1.18it/s][2022-01-08 01:39:42,271] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 8192.0, reducing to 4096.0
[2022-01-08 01:39:43,120] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 4096.0, reducing to 2048.0
1%| | 23/2013 [00:19<28:12, 1.18it/s][2022-01-08 01:39:43,971] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 2048.0, reducing to 1024.0
1%| | 24/2013 [00:20<28:10, 1.18it/s][2022-01-08 01:39:44,819] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 1024.0, reducing to 512.0
1%| | 25/2013 [00:21<28:09, 1.18it/s][2022-01-08 01:39:45,670] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 512.0, reducing to 256.0
[2022-01-08 01:39:46,518] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 256.0, reducing to 128.0
1%|▏ | 27/2013 [00:23<28:08, 1.18it/s][2022-01-08 01:39:47,370] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 128.0, reducing to 64.0
[2022-01-08 01:39:48,221] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 64.0, reducing to 32.0
1%|▏ | 29/2013 [00:24<28:08, 1.17it/s][2022-01-08 01:39:49,074] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 32.0, reducing to 16.0
[2022-01-08 01:39:49,923] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 16.0, reducing to 8.0
[2022-01-08 01:39:49,924] [INFO] [logging.py:69:log_dist] [Rank 0] step=30, skipped=30, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:39:49,924] [INFO] [timer.py:181:stop] 0/30, SamplesPerSec=17.62883249432792
2%|▏ | 32/2013 [00:30<50:21, 1.53s/it] [2022-01-08 01:39:54,841] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 8.0, reducing to 4.0
2%|▏ | 33/2013 [00:31<43:39, 1.32s/it][2022-01-08 01:39:55,691] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 4.0, reducing to 2.0
[2022-01-08 01:39:56,547] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 2.0, reducing to 1.0
2%|▏ | 36/2013 [00:37<55:42, 1.69s/it] [2022-01-08 01:40:01,475] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 1.0, reducing to 1
2%|▏ | 39/2013 [00:49<1:47:11, 3.26s/it][2022-01-08 01:40:17,759] [INFO] [logging.py:69:log_dist] [Rank 0] step=40, skipped=34, lr=[2.2900676539246968e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:40:17,789] [INFO] [timer.py:181:stop] 0/40, SamplesPerSec=11.031298344774584
2%|▏ | 43/2013 [01:02<1:35:40, 2.91s/it][2022-01-08 01:40:26,799] [INFO] [stage_1_and_2.py:1631:step] [deepscale] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 1, reducing to 1
2%|▏ | 49/2013 [01:26<2:09:07, 3.94s/it][2022-01-08 01:40:55,351] [INFO] [logging.py:69:log_dist] [Rank 0] step=50, skipped=35, lr=[3.4611890029080124e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:40:55,382] [INFO] [timer.py:181:stop] 0/50, SamplesPerSec=8.067930724877757
{'loss': 8.0555, 'learning_rate': 3.4611890029080124e-05, 'epoch': 0.11}
3%|▎ | 59/2013 [02:07<2:12:53, 4.08s/it][2022-01-08 01:41:36,220] [INFO] [logging.py:69:log_dist] [Rank 0] step=60, skipped=35, lr=[4.1140808993222106e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:41:36,249] [INFO] [timer.py:181:stop] 0/60, SamplesPerSec=6.68789572068323
3%|▎ | 69/2013 [02:48<2:12:24, 4.09s/it][2022-01-08 01:42:17,068] [INFO] [logging.py:69:log_dist] [Rank 0] step=70, skipped=35, lr=[4.544129797493744e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:42:17,097] [INFO] [timer.py:181:stop] 0/70, SamplesPerSec=5.96801566122986
4%|▍ | 79/2013 [03:29<2:11:40, 4.09s/it][2022-01-08 01:42:57,921] [INFO] [logging.py:69:log_dist] [Rank 0] step=80, skipped=35, lr=[4.8653375561549195e-05], mom=[[0.9, 0.999]]
4%|▍ | 80/2013 [03:33<2:11:37, 4.09s/it][2022-01-08 01:42:57,954] [INFO] [timer.py:181:stop] 0/80, SamplesPerSec=5.525526128208344
4%|▍ | 89/2013 [04:10<2:11:03, 4.09s/it][2022-01-08 01:43:38,801] [INFO] [logging.py:69:log_dist] [Rank 0] step=90, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
4%|▍ | 90/2013 [04:14<2:10:56, 4.09s/it][2022-01-08 01:43:38,831] [INFO] [timer.py:181:stop] 0/90, SamplesPerSec=5.225669499980795
5%|▍ | 99/2013 [04:51<2:10:22, 4.09s/it][2022-01-08 01:44:19,655] [INFO] [logging.py:69:log_dist] [Rank 0] step=100, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:44:19,689] [INFO] [timer.py:181:stop] 0/100, SamplesPerSec=5.009645748243033
{'loss': 3.5939, 'learning_rate': 5e-05, 'epoch': 0.21}
5%|▌ | 109/2013 [05:32<2:11:54, 4.16s/it][2022-01-08 01:45:00,985] [INFO] [logging.py:69:log_dist] [Rank 0] step=110, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
5%|▌ | 110/2013 [05:36<2:11:34, 4.15s/it][2022-01-08 01:45:01,021] [INFO] [timer.py:181:stop] 0/110, SamplesPerSec=4.8395545003806975
6%|▌ | 119/2013 [06:13<2:10:51, 4.15s/it][2022-01-08 01:45:42,347] [INFO] [logging.py:69:log_dist] [Rank 0] step=120, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
6%|▌ | 120/2013 [06:18<2:10:50, 4.15s/it][2022-01-08 01:45:42,389] [INFO] [timer.py:181:stop] 0/120, SamplesPerSec=4.706362155986512
6%|▋ | 129/2013 [06:54<2:07:29, 4.06s/it][2022-01-08 01:46:23,102] [INFO] [logging.py:69:log_dist] [Rank 0] step=130, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
6%|▋ | 130/2013 [06:58<2:07:17, 4.06s/it][2022-01-08 01:46:23,132] [INFO] [timer.py:181:stop] 0/130, SamplesPerSec=4.606489989123324
7%|▋ | 139/2013 [07:35<2:08:37, 4.12s/it][2022-01-08 01:47:04,151] [INFO] [logging.py:69:log_dist] [Rank 0] step=140, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
7%|▋ | 140/2013 [07:39<2:10:36, 4.18s/it][2022-01-08 01:47:04,181] [INFO] [timer.py:181:stop] 0/140, SamplesPerSec=4.521363810177863
7%|▋ | 149/2013 [08:16<2:07:19, 4.10s/it][2022-01-08 01:47:45,118] [INFO] [logging.py:69:log_dist] [Rank 0] step=150, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:47:45,148] [INFO] [timer.py:181:stop] 0/150, SamplesPerSec=4.450944835640465
{'loss': 2.9109, 'learning_rate': 5e-05, 'epoch': 0.32}
8%|▊ | 159/2013 [08:57<2:05:06, 4.05s/it][2022-01-08 01:48:25,604] [INFO] [logging.py:69:log_dist] [Rank 0] step=160, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
8%|▊ | 160/2013 [09:01<2:05:01, 4.05s/it][2022-01-08 01:48:25,633] [INFO] [timer.py:181:stop] 0/160, SamplesPerSec=4.395125942496921
8%|▊ | 169/2013 [09:37<2:04:25, 4.05s/it][2022-01-08 01:49:06,091] [INFO] [logging.py:69:log_dist] [Rank 0] step=170, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:49:06,121] [INFO] [timer.py:181:stop] 0/170, SamplesPerSec=4.347081921578749
9%|▉ | 179/2013 [10:18<2:03:42, 4.05s/it][2022-01-08 01:49:46,567] [INFO] [logging.py:69:log_dist] [Rank 0] step=180, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:49:46,596] [INFO] [timer.py:181:stop] 0/180, SamplesPerSec=4.30539080340622
9%|▉ | 189/2013 [10:58<2:03:01, 4.05s/it][2022-01-08 01:50:27,060] [INFO] [logging.py:69:log_dist] [Rank 0] step=190, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:50:27,094] [INFO] [timer.py:181:stop] 0/190, SamplesPerSec=4.268662845854158
10%|▉ | 199/2013 [11:39<2:02:24, 4.05s/it][2022-01-08 01:51:07,548] [INFO] [logging.py:69:log_dist] [Rank 0] step=200, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:51:07,578] [INFO] [timer.py:181:stop] 0/200, SamplesPerSec=4.236264567109609
{'loss': 2.5181, 'learning_rate': 5e-05, 'epoch': 0.43}
10%|█ | 209/2013 [12:19<2:01:39, 4.05s/it][2022-01-08 01:51:48,014] [INFO] [logging.py:69:log_dist] [Rank 0] step=210, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
10%|█ | 210/2013 [12:23<2:01:34, 4.05s/it][2022-01-08 01:51:48,043] [INFO] [timer.py:181:stop] 0/210, SamplesPerSec=4.2075055963236165
11%|█ | 219/2013 [13:00<2:00:59, 4.05s/it][2022-01-08 01:52:28,480] [INFO] [logging.py:69:log_dist] [Rank 0] step=220, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
11%|█ | 220/2013 [13:04<2:00:53, 4.05s/it][2022-01-08 01:52:28,509] [INFO] [timer.py:181:stop] 0/220, SamplesPerSec=4.181716400622012
11%|█▏ | 229/2013 [13:40<2:00:21, 4.05s/it][2022-01-08 01:53:08,956] [INFO] [logging.py:69:log_dist] [Rank 0] step=230, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:53:08,985] [INFO] [timer.py:181:stop] 0/230, SamplesPerSec=4.15841161249685
12%|█▏ | 239/2013 [14:21<1:59:40, 4.05s/it][2022-01-08 01:53:49,436] [INFO] [logging.py:69:log_dist] [Rank 0] step=240, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
12%|█▏ | 240/2013 [14:25<1:59:35, 4.05s/it][2022-01-08 01:53:49,465] [INFO] [timer.py:181:stop] 0/240, SamplesPerSec=4.1372710498922505
12%|█▏ | 249/2013 [15:01<1:59:01, 4.05s/it][2022-01-08 01:54:29,916] [INFO] [logging.py:69:log_dist] [Rank 0] step=250, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:54:29,945] [INFO] [timer.py:181:stop] 0/250, SamplesPerSec=4.118029453304923
{'loss': 2.2938, 'learning_rate': 5e-05, 'epoch': 0.53}
13%|█▎ | 259/2013 [15:42<1:58:19, 4.05s/it][2022-01-08 01:55:10,398] [INFO] [logging.py:69:log_dist] [Rank 0] step=260, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
13%|█▎ | 260/2013 [15:46<1:58:14, 4.05s/it][2022-01-08 01:55:10,427] [INFO] [timer.py:181:stop] 0/260, SamplesPerSec=4.100428112215623
13%|█▎ | 269/2013 [16:22<1:57:37, 4.05s/it][2022-01-08 01:55:50,879] [INFO] [logging.py:69:log_dist] [Rank 0] step=270, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
13%|█▎ | 270/2013 [16:26<1:57:34, 4.05s/it][2022-01-08 01:55:50,908] [INFO] [timer.py:181:stop] 0/270, SamplesPerSec=4.08427563519003
14%|█▍ | 279/2013 [17:02<1:56:57, 4.05s/it][2022-01-08 01:56:31,354] [INFO] [logging.py:69:log_dist] [Rank 0] step=280, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:56:31,383] [INFO] [timer.py:181:stop] 0/280, SamplesPerSec=4.069423588039075
14%|█▍ | 289/2013 [17:43<1:56:15, 4.05s/it][2022-01-08 01:57:11,823] [INFO] [logging.py:69:log_dist] [Rank 0] step=290, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:57:11,853] [INFO] [timer.py:181:stop] 0/290, SamplesPerSec=4.055719991558985
15%|█▍ | 299/2013 [18:23<1:55:35, 4.05s/it][2022-01-08 01:57:52,286] [INFO] [logging.py:69:log_dist] [Rank 0] step=300, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
15%|█▍ | 300/2013 [18:27<1:55:30, 4.05s/it][2022-01-08 01:57:52,316] [INFO] [timer.py:181:stop] 0/300, SamplesPerSec=4.043043675522679
{'loss': 2.2055, 'learning_rate': 5e-05, 'epoch': 0.64}
15%|█▌ | 309/2013 [19:04<1:54:54, 4.05s/it][2022-01-08 01:58:32,751] [INFO] [logging.py:69:log_dist] [Rank 0] step=310, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:58:32,780] [INFO] [timer.py:181:stop] 0/310, SamplesPerSec=4.031256843267691
16%|█▌ | 319/2013 [19:44<1:54:15, 4.05s/it][2022-01-08 01:59:13,224] [INFO] [logging.py:69:log_dist] [Rank 0] step=320, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:59:13,254] [INFO] [timer.py:181:stop] 0/320, SamplesPerSec=4.020242988154115
16%|█▋ | 329/2013 [20:25<1:53:32, 4.05s/it][2022-01-08 01:59:53,684] [INFO] [logging.py:69:log_dist] [Rank 0] step=330, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 01:59:53,713] [INFO] [timer.py:181:stop] 0/330, SamplesPerSec=4.010001540923997
17%|█▋ | 339/2013 [21:05<1:52:53, 4.05s/it][2022-01-08 02:00:34,158] [INFO] [logging.py:69:log_dist] [Rank 0] step=340, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
17%|█▋ | 340/2013 [21:09<1:52:49, 4.05s/it][2022-01-08 02:00:34,187] [INFO] [timer.py:181:stop] 0/340, SamplesPerSec=4.000367521535013
17%|█▋ | 349/2013 [21:46<1:52:14, 4.05s/it][2022-01-08 02:01:14,624] [INFO] [logging.py:69:log_dist] [Rank 0] step=350, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:01:14,653] [INFO] [timer.py:181:stop] 0/350, SamplesPerSec=3.9913515055541375
{'loss': 2.1103, 'learning_rate': 5e-05, 'epoch': 0.75}
18%|█▊ | 359/2013 [22:26<1:51:37, 4.05s/it][2022-01-08 02:01:55,112] [INFO] [logging.py:69:log_dist] [Rank 0] step=360, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:01:55,141] [INFO] [timer.py:181:stop] 0/360, SamplesPerSec=3.982814524697442
18%|█▊ | 369/2013 [23:07<1:50:52, 4.05s/it][2022-01-08 02:02:35,591] [INFO] [logging.py:69:log_dist] [Rank 0] step=370, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:02:35,620] [INFO] [timer.py:181:stop] 0/370, SamplesPerSec=3.9747992600976216
19%|█▉ | 379/2013 [23:47<1:50:11, 4.05s/it][2022-01-08 02:03:16,067] [INFO] [logging.py:69:log_dist] [Rank 0] step=380, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
19%|█▉ | 380/2013 [23:51<1:50:07, 4.05s/it][2022-01-08 02:03:16,096] [INFO] [timer.py:181:stop] 0/380, SamplesPerSec=3.967245180913496
19%|█▉ | 389/2013 [24:28<1:49:31, 4.05s/it][2022-01-08 02:03:56,529] [INFO] [logging.py:69:log_dist] [Rank 0] step=390, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:03:56,558] [INFO] [timer.py:181:stop] 0/390, SamplesPerSec=3.960144396097978
20%|█▉ | 399/2013 [25:08<1:48:48, 4.04s/it][2022-01-08 02:04:36,990] [INFO] [logging.py:69:log_dist] [Rank 0] step=400, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:04:37,019] [INFO] [timer.py:181:stop] 0/400, SamplesPerSec=3.953427795728032
{'loss': 2.0659, 'learning_rate': 5e-05, 'epoch': 0.85}
20%|██ | 409/2013 [25:49<1:49:40, 4.10s/it][2022-01-08 02:05:17,860] [INFO] [logging.py:69:log_dist] [Rank 0] step=410, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:05:17,889] [INFO] [timer.py:181:stop] 0/410, SamplesPerSec=3.946022334146505
21%|██ | 419/2013 [26:30<1:48:24, 4.08s/it][2022-01-08 02:05:58,738] [INFO] [logging.py:69:log_dist] [Rank 0] step=420, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:05:58,767] [INFO] [timer.py:181:stop] 0/420, SamplesPerSec=3.938975216922789
21%|██▏ | 429/2013 [27:11<1:47:11, 4.06s/it][2022-01-08 02:06:39,470] [INFO] [logging.py:69:log_dist] [Rank 0] step=430, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
21%|██▏ | 430/2013 [27:15<1:47:00, 4.06s/it][2022-01-08 02:06:39,500] [INFO] [timer.py:181:stop] 0/430, SamplesPerSec=3.9326352521937826
22%|██▏ | 439/2013 [27:51<1:46:08, 4.05s/it][2022-01-08 02:07:19,928] [INFO] [logging.py:69:log_dist] [Rank 0] step=440, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
22%|██▏ | 440/2013 [27:55<1:46:02, 4.05s/it][2022-01-08 02:07:19,957] [INFO] [timer.py:181:stop] 0/440, SamplesPerSec=3.927246369217363
22%|██▏ | 449/2013 [28:32<1:45:27, 4.05s/it][2022-01-08 02:08:00,388] [INFO] [logging.py:69:log_dist] [Rank 0] step=450, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
22%|██▏ | 450/2013 [28:36<1:45:22, 4.05s/it][2022-01-08 02:08:00,417] [INFO] [timer.py:181:stop] 0/450, SamplesPerSec=3.922104804486527
{'loss': 2.0305, 'learning_rate': 5e-05, 'epoch': 0.96}
23%|██▎ | 459/2013 [29:12<1:44:47, 4.05s/it][2022-01-08 02:08:40,842] [INFO] [logging.py:69:log_dist] [Rank 0] step=460, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:08:40,872] [INFO] [timer.py:181:stop] 0/460, SamplesPerSec=3.9172138294729657
23%|██▎ | 469/2013 [29:52<1:43:18, 4.01s/it][2022-01-08 02:09:21,153] [INFO] [logging.py:69:log_dist] [Rank 0] step=470, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:09:21,182] [INFO] [timer.py:181:stop] 0/470, SamplesPerSec=3.912857902795985
24%|██▍ | 479/2013 [30:33<1:43:23, 4.04s/it][2022-01-08 02:10:01,600] [INFO] [logging.py:69:log_dist] [Rank 0] step=480, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:10:01,629] [INFO] [timer.py:181:stop] 0/480, SamplesPerSec=3.9084021261057664
24%|██▍ | 489/2013 [31:13<1:42:48, 4.05s/it][2022-01-08 02:10:42,068] [INFO] [logging.py:69:log_dist] [Rank 0] step=490, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
24%|██▍ | 490/2013 [31:17<1:42:44, 4.05s/it][2022-01-08 02:10:42,098] [INFO] [timer.py:181:stop] 0/490, SamplesPerSec=3.9040916476022516
25%|██▍ | 499/2013 [31:54<1:42:05, 4.05s/it][2022-01-08 02:11:22,528] [INFO] [logging.py:69:log_dist] [Rank 0] step=500, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
25%|██▍ | 500/2013 [31:58<1:42:00, 4.05s/it][2022-01-08 02:11:22,558] [INFO] [timer.py:181:stop] 0/500, SamplesPerSec=3.8999822731421663
{'loss': 1.7833, 'learning_rate': 5e-05, 'epoch': 1.07}
25%|██▌ | 509/2013 [32:34<1:41:25, 4.05s/it][2022-01-08 02:12:02,995] [INFO] [logging.py:69:log_dist] [Rank 0] step=510, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:12:03,024] [INFO] [timer.py:181:stop] 0/510, SamplesPerSec=3.8960307883299508
26%|██▌ | 519/2013 [33:15<1:40:44, 4.05s/it][2022-01-08 02:12:43,455] [INFO] [logging.py:69:log_dist] [Rank 0] step=520, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:12:43,485] [INFO] [timer.py:181:stop] 0/520, SamplesPerSec=3.892250043059626
26%|██▋ | 529/2013 [33:55<1:40:05, 4.05s/it][2022-01-08 02:13:23,923] [INFO] [logging.py:69:log_dist] [Rank 0] step=530, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:13:23,952] [INFO] [timer.py:181:stop] 0/530, SamplesPerSec=3.8886053812957786
27%|██▋ | 539/2013 [34:36<1:39:25, 4.05s/it][2022-01-08 02:14:04,387] [INFO] [logging.py:69:log_dist] [Rank 0] step=540, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:14:04,416] [INFO] [timer.py:181:stop] 0/540, SamplesPerSec=3.8851096882859215
27%|██▋ | 549/2013 [35:16<1:38:44, 4.05s/it][2022-01-08 02:14:44,857] [INFO] [logging.py:69:log_dist] [Rank 0] step=550, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:14:44,886] [INFO] [timer.py:181:stop] 0/550, SamplesPerSec=3.8817375437319637
{'loss': 1.5955, 'learning_rate': 5e-05, 'epoch': 1.18}
28%|██▊ | 559/2013 [35:56<1:38:03, 4.05s/it][2022-01-08 02:15:25,319] [INFO] [logging.py:69:log_dist] [Rank 0] step=560, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:15:25,348] [INFO] [timer.py:181:stop] 0/560, SamplesPerSec=3.878506793993267
28%|██▊ | 569/2013 [36:37<1:37:26, 4.05s/it][2022-01-08 02:16:05,801] [INFO] [logging.py:69:log_dist] [Rank 0] step=570, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:16:05,831] [INFO] [timer.py:181:stop] 0/570, SamplesPerSec=3.8753569323112336
29%|██▉ | 579/2013 [37:17<1:36:42, 4.05s/it][2022-01-08 02:16:46,257] [INFO] [logging.py:69:log_dist] [Rank 0] step=580, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
29%|██▉ | 580/2013 [37:21<1:36:37, 4.05s/it][2022-01-08 02:16:46,286] [INFO] [timer.py:181:stop] 0/580, SamplesPerSec=3.8723671205522794
29%|██▉ | 589/2013 [37:58<1:36:01, 4.05s/it][2022-01-08 02:17:26,713] [INFO] [logging.py:69:log_dist] [Rank 0] step=590, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:17:26,742] [INFO] [timer.py:181:stop] 0/590, SamplesPerSec=3.869484178092346
30%|██▉ | 599/2013 [38:38<1:35:20, 4.05s/it][2022-01-08 02:18:07,172] [INFO] [logging.py:69:log_dist] [Rank 0] step=600, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
30%|██▉ | 600/2013 [38:42<1:35:16, 4.05s/it][2022-01-08 02:18:07,201] [INFO] [timer.py:181:stop] 0/600, SamplesPerSec=3.866695607143853
{'loss': 1.6314, 'learning_rate': 5e-05, 'epoch': 1.28}
30%|███ | 609/2013 [39:19<1:34:41, 4.05s/it][2022-01-08 02:18:47,637] [INFO] [logging.py:69:log_dist] [Rank 0] step=610, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
30%|███ | 610/2013 [39:23<1:34:36, 4.05s/it][2022-01-08 02:18:47,667] [INFO] [timer.py:181:stop] 0/610, SamplesPerSec=3.863995294753564
31%|███ | 619/2013 [39:59<1:33:59, 4.05s/it][2022-01-08 02:19:28,096] [INFO] [logging.py:69:log_dist] [Rank 0] step=620, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:19:28,126] [INFO] [timer.py:181:stop] 0/620, SamplesPerSec=3.8613944667162796
31%|███ | 629/2013 [40:40<1:33:18, 4.05s/it][2022-01-08 02:20:08,549] [INFO] [logging.py:69:log_dist] [Rank 0] step=630, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
31%|███▏ | 630/2013 [40:44<1:33:15, 4.05s/it][2022-01-08 02:20:08,578] [INFO] [timer.py:181:stop] 0/630, SamplesPerSec=3.858889805647715
32%|███▏ | 639/2013 [41:20<1:32:39, 4.05s/it][2022-01-08 02:20:49,010] [INFO] [logging.py:69:log_dist] [Rank 0] step=640, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
32%|███▏ | 640/2013 [41:24<1:32:35, 4.05s/it][2022-01-08 02:20:49,039] [INFO] [timer.py:181:stop] 0/640, SamplesPerSec=3.856452547471229
32%|███▏ | 649/2013 [42:01<1:31:59, 4.05s/it][2022-01-08 02:21:29,483] [INFO] [logging.py:69:log_dist] [Rank 0] step=650, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:21:29,512] [INFO] [timer.py:181:stop] 0/650, SamplesPerSec=3.854075451842506
{'loss': 1.6386, 'learning_rate': 5e-05, 'epoch': 1.39}
33%|███▎ | 659/2013 [42:41<1:31:21, 4.05s/it][2022-01-08 02:22:09,966] [INFO] [logging.py:69:log_dist] [Rank 0] step=660, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:22:09,995] [INFO] [timer.py:181:stop] 0/660, SamplesPerSec=3.851759156810231
33%|███▎ | 669/2013 [43:22<1:30:38, 4.05s/it][2022-01-08 02:22:50,434] [INFO] [logging.py:69:log_dist] [Rank 0] step=670, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:22:50,463] [INFO] [timer.py:181:stop] 0/670, SamplesPerSec=3.8495358080187763
34%|███▎ | 679/2013 [44:02<1:29:56, 4.05s/it][2022-01-08 02:23:30,888] [INFO] [logging.py:69:log_dist] [Rank 0] step=680, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
34%|███▍ | 680/2013 [44:06<1:29:52, 4.05s/it][2022-01-08 02:23:30,918] [INFO] [timer.py:181:stop] 0/680, SamplesPerSec=3.8474007121496987
34%|███▍ | 689/2013 [44:42<1:29:16, 4.05s/it][2022-01-08 02:24:11,349] [INFO] [logging.py:69:log_dist] [Rank 0] step=690, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:24:11,379] [INFO] [timer.py:181:stop] 0/690, SamplesPerSec=3.84532033204538
35%|███▍ | 699/2013 [45:23<1:28:36, 4.05s/it][2022-01-08 02:24:51,815] [INFO] [logging.py:69:log_dist] [Rank 0] step=700, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
35%|███▍ | 700/2013 [45:27<1:28:31, 4.05s/it][2022-01-08 02:24:51,844] [INFO] [timer.py:181:stop] 0/700, SamplesPerSec=3.8432963755921055
{'loss': 1.6634, 'learning_rate': 5e-05, 'epoch': 1.5}
35%|███▌ | 709/2013 [46:03<1:27:57, 4.05s/it][2022-01-08 02:25:32,284] [INFO] [logging.py:69:log_dist] [Rank 0] step=710, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:25:32,314] [INFO] [timer.py:181:stop] 0/710, SamplesPerSec=3.8413256812881262
36%|███▌ | 719/2013 [46:44<1:27:17, 4.05s/it][2022-01-08 02:26:12,754] [INFO] [logging.py:69:log_dist] [Rank 0] step=720, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:26:12,784] [INFO] [timer.py:181:stop] 0/720, SamplesPerSec=3.8394109698718157
36%|███▌ | 729/2013 [47:24<1:26:35, 4.05s/it][2022-01-08 02:26:53,217] [INFO] [logging.py:69:log_dist] [Rank 0] step=730, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:26:53,246] [INFO] [timer.py:181:stop] 0/730, SamplesPerSec=3.8375600621192807
37%|███▋ | 739/2013 [48:05<1:25:55, 4.05s/it][2022-01-08 02:27:33,680] [INFO] [logging.py:69:log_dist] [Rank 0] step=740, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:27:33,709] [INFO] [timer.py:181:stop] 0/740, SamplesPerSec=3.8357602518968394
37%|███▋ | 749/2013 [48:45<1:25:14, 4.05s/it][2022-01-08 02:28:14,146] [INFO] [logging.py:69:log_dist] [Rank 0] step=750, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:28:14,175] [INFO] [timer.py:181:stop] 0/750, SamplesPerSec=3.8340076399035885
{'loss': 1.6479, 'learning_rate': 5e-05, 'epoch': 1.6}
38%|███▊ | 759/2013 [49:26<1:24:34, 4.05s/it][2022-01-08 02:28:54,609] [INFO] [logging.py:69:log_dist] [Rank 0] step=760, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:28:54,638] [INFO] [timer.py:181:stop] 0/760, SamplesPerSec=3.8323056392290065
38%|███▊ | 769/2013 [50:06<1:23:55, 4.05s/it][2022-01-08 02:29:35,089] [INFO] [logging.py:69:log_dist] [Rank 0] step=770, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:29:35,118] [INFO] [timer.py:181:stop] 0/770, SamplesPerSec=3.830627444246552
39%|███▊ | 779/2013 [50:47<1:23:14, 4.05s/it][2022-01-08 02:30:15,559] [INFO] [logging.py:69:log_dist] [Rank 0] step=780, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
39%|███▊ | 780/2013 [50:51<1:23:10, 4.05s/it][2022-01-08 02:30:15,589] [INFO] [timer.py:181:stop] 0/780, SamplesPerSec=3.8290056714847913
39%|███▉ | 789/2013 [51:27<1:22:32, 4.05s/it][2022-01-08 02:30:56,022] [INFO] [logging.py:69:log_dist] [Rank 0] step=790, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
39%|███▉ | 790/2013 [51:31<1:22:29, 4.05s/it][2022-01-08 02:30:56,051] [INFO] [timer.py:181:stop] 0/790, SamplesPerSec=3.827436712396159
40%|███▉ | 799/2013 [52:08<1:21:51, 4.05s/it][2022-01-08 02:31:36,485] [INFO] [logging.py:69:log_dist] [Rank 0] step=800, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
40%|███▉ | 800/2013 [52:12<1:21:47, 4.05s/it][2022-01-08 02:31:36,514] [INFO] [timer.py:181:stop] 0/800, SamplesPerSec=3.8259080607520604
{'loss': 1.6626, 'learning_rate': 5e-05, 'epoch': 1.71}
40%|████ | 809/2013 [52:48<1:21:12, 4.05s/it][2022-01-08 02:32:16,957] [INFO] [logging.py:69:log_dist] [Rank 0] step=810, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:32:16,986] [INFO] [timer.py:181:stop] 0/810, SamplesPerSec=3.8244084164060608
41%|████ | 819/2013 [53:29<1:20:31, 4.05s/it][2022-01-08 02:32:57,424] [INFO] [logging.py:69:log_dist] [Rank 0] step=820, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:32:57,454] [INFO] [timer.py:181:stop] 0/820, SamplesPerSec=3.8229513148389014
41%|████ | 829/2013 [54:09<1:19:51, 4.05s/it][2022-01-08 02:33:37,885] [INFO] [logging.py:69:log_dist] [Rank 0] step=830, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:33:37,915] [INFO] [timer.py:181:stop] 0/830, SamplesPerSec=3.821537705164049
42%|████▏ | 839/2013 [54:49<1:19:10, 4.05s/it][2022-01-08 02:34:18,351] [INFO] [logging.py:69:log_dist] [Rank 0] step=840, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
42%|████▏ | 840/2013 [54:54<1:19:06, 4.05s/it][2022-01-08 02:34:18,381] [INFO] [timer.py:181:stop] 0/840, SamplesPerSec=3.820152968516307
42%|████▏ | 849/2013 [55:30<1:18:30, 4.05s/it][2022-01-08 02:34:58,823] [INFO] [logging.py:69:log_dist] [Rank 0] step=850, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
42%|████▏ | 850/2013 [55:34<1:18:26, 4.05s/it][2022-01-08 02:34:58,852] [INFO] [timer.py:181:stop] 0/850, SamplesPerSec=3.8187951577261847
{'loss': 1.6684, 'learning_rate': 5e-05, 'epoch': 1.82}
43%|████▎ | 859/2013 [56:10<1:17:50, 4.05s/it][2022-01-08 02:35:39,295] [INFO] [logging.py:69:log_dist] [Rank 0] step=860, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:35:39,325] [INFO] [timer.py:181:stop] 0/860, SamplesPerSec=3.8174698476898015
43%|████▎ | 869/2013 [56:51<1:17:09, 4.05s/it][2022-01-08 02:36:19,765] [INFO] [logging.py:69:log_dist] [Rank 0] step=870, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:36:19,795] [INFO] [timer.py:181:stop] 0/870, SamplesPerSec=3.816178815977781
44%|████▎ | 879/2013 [57:31<1:16:29, 4.05s/it][2022-01-08 02:37:00,236] [INFO] [logging.py:69:log_dist] [Rank 0] step=880, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:37:00,265] [INFO] [timer.py:181:stop] 0/880, SamplesPerSec=3.8149162260617087
44%|████▍ | 889/2013 [58:12<1:15:47, 4.05s/it][2022-01-08 02:37:40,697] [INFO] [logging.py:69:log_dist] [Rank 0] step=890, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:37:40,727] [INFO] [timer.py:181:stop] 0/890, SamplesPerSec=3.8136928093326032
45%|████▍ | 899/2013 [58:52<1:15:07, 4.05s/it][2022-01-08 02:38:21,159] [INFO] [logging.py:69:log_dist] [Rank 0] step=900, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:38:21,188] [INFO] [timer.py:181:stop] 0/900, SamplesPerSec=3.8124973793214734
{'loss': 1.6559, 'learning_rate': 5e-05, 'epoch': 1.92}
45%|████▌ | 909/2013 [59:33<1:14:26, 4.05s/it][2022-01-08 02:39:01,612] [INFO] [logging.py:69:log_dist] [Rank 0] step=910, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:39:01,641] [INFO] [timer.py:181:stop] 0/910, SamplesPerSec=3.811339152155201
46%|████▌ | 919/2013 [1:00:13<1:13:46, 4.05s/it][2022-01-08 02:39:42,075] [INFO] [logging.py:69:log_dist] [Rank 0] step=920, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
46%|████▌ | 920/2013 [1:00:17<1:13:42, 4.05s/it][2022-01-08 02:39:42,105] [INFO] [timer.py:181:stop] 0/920, SamplesPerSec=3.8101944253176074
46%|████▌ | 929/2013 [1:00:54<1:13:05, 4.05s/it][2022-01-08 02:40:22,530] [INFO] [logging.py:69:log_dist] [Rank 0] step=930, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:40:22,559] [INFO] [timer.py:181:stop] 0/930, SamplesPerSec=3.809085397484642
47%|████▋ | 939/2013 [1:01:34<1:12:05, 4.03s/it][2022-01-08 02:41:02,820] [INFO] [logging.py:69:log_dist] [Rank 0] step=940, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:41:02,849] [INFO] [timer.py:181:stop] 0/940, SamplesPerSec=3.8081701776944348
47%|████▋ | 949/2013 [1:02:14<1:11:44, 4.05s/it][2022-01-08 02:41:43,278] [INFO] [logging.py:69:log_dist] [Rank 0] step=950, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
47%|████▋ | 950/2013 [1:02:18<1:11:40, 4.05s/it][2022-01-08 02:41:43,309] [INFO] [timer.py:181:stop] 0/950, SamplesPerSec=3.8071008520815415
{'loss': 1.5356, 'learning_rate': 5e-05, 'epoch': 2.03}
48%|████▊ | 959/2013 [1:02:55<1:11:04, 4.05s/it][2022-01-08 02:42:23,743] [INFO] [logging.py:69:log_dist] [Rank 0] step=960, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:42:23,772] [INFO] [timer.py:181:stop] 0/960, SamplesPerSec=3.8060521384459487
48%|████▊ | 969/2013 [1:03:35<1:10:23, 4.05s/it][2022-01-08 02:43:04,195] [INFO] [logging.py:69:log_dist] [Rank 0] step=970, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:43:04,225] [INFO] [timer.py:181:stop] 0/970, SamplesPerSec=3.805035727408843
49%|████▊ | 979/2013 [1:04:16<1:09:43, 4.05s/it][2022-01-08 02:43:44,657] [INFO] [logging.py:69:log_dist] [Rank 0] step=980, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:43:44,686] [INFO] [timer.py:181:stop] 0/980, SamplesPerSec=3.8040317669755774
49%|████▉ | 989/2013 [1:04:56<1:09:03, 4.05s/it][2022-01-08 02:44:25,116] [INFO] [logging.py:69:log_dist] [Rank 0] step=990, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
49%|████▉ | 990/2013 [1:05:00<1:08:59, 4.05s/it][2022-01-08 02:44:25,145] [INFO] [timer.py:181:stop] 0/990, SamplesPerSec=3.8030511553031
50%|████▉ | 999/2013 [1:05:37<1:08:22, 4.05s/it][2022-01-08 02:45:05,572] [INFO] [logging.py:69:log_dist] [Rank 0] step=1000, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
50%|████▉ | 1000/2013 [1:05:41<1:08:17, 4.05s/it][2022-01-08 02:45:05,602] [INFO] [timer.py:181:stop] 0/1000, SamplesPerSec=3.8020936727615156
{'loss': 1.1188, 'learning_rate': 5e-05, 'epoch': 2.14}
50%|█████ | 1009/2013 [1:06:17<1:07:42, 4.05s/it][2022-01-08 02:45:46,035] [INFO] [logging.py:69:log_dist] [Rank 0] step=1010, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
50%|█████ | 1010/2013 [1:06:21<1:07:38, 4.05s/it][2022-01-08 02:45:46,064] [INFO] [timer.py:181:stop] 0/1010, SamplesPerSec=3.8011496998307788
51%|█████ | 1019/2013 [1:06:58<1:07:00, 4.04s/it][2022-01-08 02:46:26,488] [INFO] [logging.py:69:log_dist] [Rank 0] step=1020, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
51%|█████ | 1020/2013 [1:07:02<1:06:56, 4.04s/it][2022-01-08 02:46:26,517] [INFO] [timer.py:181:stop] 0/1020, SamplesPerSec=3.8002330853204738
51%|█████ | 1029/2013 [1:07:38<1:06:21, 4.05s/it][2022-01-08 02:47:06,952] [INFO] [logging.py:69:log_dist] [Rank 0] step=1030, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:47:06,981] [INFO] [timer.py:181:stop] 0/1030, SamplesPerSec=3.799324762921159
52%|█████▏ | 1039/2013 [1:08:19<1:05:41, 4.05s/it][2022-01-08 02:47:47,417] [INFO] [logging.py:69:log_dist] [Rank 0] step=1040, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
52%|█████▏ | 1040/2013 [1:08:23<1:05:37, 4.05s/it][2022-01-08 02:47:47,446] [INFO] [timer.py:181:stop] 0/1040, SamplesPerSec=3.798433009902687
52%|█████▏ | 1049/2013 [1:08:59<1:05:00, 4.05s/it][2022-01-08 02:48:27,875] [INFO] [logging.py:69:log_dist] [Rank 0] step=1050, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
52%|█████▏ | 1050/2013 [1:09:03<1:04:55, 4.05s/it][2022-01-08 02:48:27,904] [INFO] [timer.py:181:stop] 0/1050, SamplesPerSec=3.7975656624936875
{'loss': 1.1015, 'learning_rate': 5e-05, 'epoch': 2.24}
53%|█████▎ | 1059/2013 [1:09:39<1:04:20, 4.05s/it][2022-01-08 02:49:08,340] [INFO] [logging.py:69:log_dist] [Rank 0] step=1060, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:49:08,370] [INFO] [timer.py:181:stop] 0/1060, SamplesPerSec=3.796709676940089
53%|█████▎ | 1069/2013 [1:10:20<1:03:39, 4.05s/it][2022-01-08 02:49:48,805] [INFO] [logging.py:69:log_dist] [Rank 0] step=1070, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
53%|█████▎ | 1070/2013 [1:10:24<1:03:35, 4.05s/it][2022-01-08 02:49:48,835] [INFO] [timer.py:181:stop] 0/1070, SamplesPerSec=3.7958689557314207
54%|█████▎ | 1079/2013 [1:11:00<1:02:59, 4.05s/it][2022-01-08 02:50:29,285] [INFO] [logging.py:69:log_dist] [Rank 0] step=1080, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:50:29,315] [INFO] [timer.py:181:stop] 0/1080, SamplesPerSec=3.795030473812278
54%|█████▍ | 1089/2013 [1:11:41<1:02:19, 4.05s/it][2022-01-08 02:51:09,752] [INFO] [logging.py:69:log_dist] [Rank 0] step=1090, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
54%|█████▍ | 1090/2013 [1:11:45<1:02:15, 4.05s/it][2022-01-08 02:51:09,782] [INFO] [timer.py:181:stop] 0/1090, SamplesPerSec=3.7942195484997527
55%|█████▍ | 1099/2013 [1:12:21<1:01:38, 4.05s/it][2022-01-08 02:51:50,220] [INFO] [logging.py:69:log_dist] [Rank 0] step=1100, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:51:50,249] [INFO] [timer.py:181:stop] 0/1100, SamplesPerSec=3.7934240124344063
{'loss': 1.0865, 'learning_rate': 5e-05, 'epoch': 2.35}
55%|█████▌ | 1109/2013 [1:13:02<1:00:57, 4.05s/it][2022-01-08 02:52:30,684] [INFO] [logging.py:69:log_dist] [Rank 0] step=1110, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:52:30,713] [INFO] [timer.py:181:stop] 0/1110, SamplesPerSec=3.7926460611632735
56%|█████▌ | 1119/2013 [1:13:42<1:00:17, 4.05s/it][2022-01-08 02:53:11,152] [INFO] [logging.py:69:log_dist] [Rank 0] step=1120, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:53:11,182] [INFO] [timer.py:181:stop] 0/1120, SamplesPerSec=3.7918782335755292
56%|█████▌ | 1129/2013 [1:14:23<59:37, 4.05s/it][2022-01-08 02:53:51,619] [INFO] [logging.py:69:log_dist] [Rank 0] step=1130, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:53:51,648] [INFO] [timer.py:181:stop] 0/1130, SamplesPerSec=3.791125838458544
57%|█████▋ | 1139/2013 [1:15:03<58:57, 4.05s/it][2022-01-08 02:54:32,088] [INFO] [logging.py:69:log_dist] [Rank 0] step=1140, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:54:32,117] [INFO] [timer.py:181:stop] 0/1140, SamplesPerSec=3.790385068891586
57%|█████▋ | 1149/2013 [1:15:44<58:16, 4.05s/it][2022-01-08 02:55:12,555] [INFO] [logging.py:69:log_dist] [Rank 0] step=1150, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
57%|█████▋ | 1150/2013 [1:15:48<58:12, 4.05s/it][2022-01-08 02:55:12,584] [INFO] [timer.py:181:stop] 0/1150, SamplesPerSec=3.789659158687227
{'loss': 1.0831, 'learning_rate': 5e-05, 'epoch': 2.46}
58%|█████▊ | 1159/2013 [1:16:24<57:35, 4.05s/it][2022-01-08 02:55:53,020] [INFO] [logging.py:69:log_dist] [Rank 0] step=1160, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
58%|█████▊ | 1160/2013 [1:16:28<57:31, 4.05s/it][2022-01-08 02:55:53,050] [INFO] [timer.py:181:stop] 0/1160, SamplesPerSec=3.7889476647509777
58%|█████▊ | 1169/2013 [1:17:05<56:53, 4.04s/it][2022-01-08 02:56:33,474] [INFO] [logging.py:69:log_dist] [Rank 0] step=1170, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
58%|█████▊ | 1170/2013 [1:17:09<56:49, 4.04s/it][2022-01-08 02:56:33,503] [INFO] [timer.py:181:stop] 0/1170, SamplesPerSec=3.7882581704339544
59%|█████▊ | 1179/2013 [1:17:45<56:14, 4.05s/it][2022-01-08 02:57:13,934] [INFO] [logging.py:69:log_dist] [Rank 0] step=1180, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
59%|█████▊ | 1180/2013 [1:17:49<56:10, 4.05s/it][2022-01-08 02:57:13,964] [INFO] [timer.py:181:stop] 0/1180, SamplesPerSec=3.7875747382137805
59%|█████▉ | 1189/2013 [1:18:26<55:33, 4.05s/it][2022-01-08 02:57:54,394] [INFO] [logging.py:69:log_dist] [Rank 0] step=1190, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
59%|█████▉ | 1190/2013 [1:18:30<55:29, 4.05s/it][2022-01-08 02:57:54,424] [INFO] [timer.py:181:stop] 0/1190, SamplesPerSec=3.7869031852927955
60%|█████▉ | 1199/2013 [1:19:06<54:52, 4.05s/it][2022-01-08 02:58:34,851] [INFO] [logging.py:69:log_dist] [Rank 0] step=1200, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:58:34,880] [INFO] [timer.py:181:stop] 0/1200, SamplesPerSec=3.7862461317087472
{'loss': 1.0837, 'learning_rate': 5e-05, 'epoch': 2.56}
60%|██████ | 1209/2013 [1:19:46<54:13, 4.05s/it][2022-01-08 02:59:15,319] [INFO] [logging.py:69:log_dist] [Rank 0] step=1210, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:59:15,349] [INFO] [timer.py:181:stop] 0/1210, SamplesPerSec=3.7855905492462787
61%|██████ | 1219/2013 [1:20:27<53:32, 4.05s/it][2022-01-08 02:59:55,786] [INFO] [logging.py:69:log_dist] [Rank 0] step=1220, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 02:59:55,815] [INFO] [timer.py:181:stop] 0/1220, SamplesPerSec=3.7849475984423036
61%|██████ | 1229/2013 [1:21:07<52:52, 4.05s/it][2022-01-08 03:00:36,249] [INFO] [logging.py:69:log_dist] [Rank 0] step=1230, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
61%|██████ | 1230/2013 [1:21:11<52:48, 4.05s/it][2022-01-08 03:00:36,280] [INFO] [timer.py:181:stop] 0/1230, SamplesPerSec=3.7843161726472756
62%|██████▏ | 1239/2013 [1:21:48<52:11, 4.05s/it][2022-01-08 03:01:16,712] [INFO] [logging.py:69:log_dist] [Rank 0] step=1240, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:01:16,742] [INFO] [timer.py:181:stop] 0/1240, SamplesPerSec=3.7836979350673303
62%|██████▏ | 1249/2013 [1:22:28<51:30, 4.05s/it][2022-01-08 03:01:57,170] [INFO] [logging.py:69:log_dist] [Rank 0] step=1250, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
62%|██████▏ | 1250/2013 [1:22:32<51:27, 4.05s/it][2022-01-08 03:01:57,199] [INFO] [timer.py:181:stop] 0/1250, SamplesPerSec=3.7830925652096035
{'loss': 1.0851, 'learning_rate': 5e-05, 'epoch': 2.67}
63%|██████▎ | 1259/2013 [1:23:09<50:50, 4.05s/it][2022-01-08 03:02:37,631] [INFO] [logging.py:69:log_dist] [Rank 0] step=1260, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:02:37,660] [INFO] [timer.py:181:stop] 0/1260, SamplesPerSec=3.782495514881184
63%|██████▎ | 1269/2013 [1:23:49<50:10, 4.05s/it][2022-01-08 03:03:18,092] [INFO] [logging.py:69:log_dist] [Rank 0] step=1270, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
63%|██████▎ | 1270/2013 [1:23:53<50:06, 4.05s/it][2022-01-08 03:03:18,121] [INFO] [timer.py:181:stop] 0/1270, SamplesPerSec=3.781907169837822
64%|██████▎ | 1279/2013 [1:24:30<49:30, 4.05s/it][2022-01-08 03:03:58,558] [INFO] [logging.py:69:log_dist] [Rank 0] step=1280, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:03:58,587] [INFO] [timer.py:181:stop] 0/1280, SamplesPerSec=3.7813247189176713
64%|██████▍ | 1289/2013 [1:25:10<48:49, 4.05s/it][2022-01-08 03:04:39,025] [INFO] [logging.py:69:log_dist] [Rank 0] step=1290, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
64%|██████▍ | 1290/2013 [1:25:14<48:45, 4.05s/it][2022-01-08 03:04:39,054] [INFO] [timer.py:181:stop] 0/1290, SamplesPerSec=3.780750847876627
65%|██████▍ | 1299/2013 [1:25:51<48:08, 4.05s/it][2022-01-08 03:05:19,489] [INFO] [logging.py:69:log_dist] [Rank 0] step=1300, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
65%|██████▍ | 1300/2013 [1:25:55<48:04, 4.05s/it][2022-01-08 03:05:19,519] [INFO] [timer.py:181:stop] 0/1300, SamplesPerSec=3.780187448579763
{'loss': 1.1161, 'learning_rate': 5e-05, 'epoch': 2.78}
65%|██████▌ | 1309/2013 [1:26:31<47:28, 4.05s/it][2022-01-08 03:05:59,952] [INFO] [logging.py:69:log_dist] [Rank 0] step=1310, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:05:59,982] [INFO] [timer.py:181:stop] 0/1310, SamplesPerSec=3.7796345096360033
66%|██████▌ | 1319/2013 [1:27:12<46:48, 4.05s/it][2022-01-08 03:06:40,422] [INFO] [logging.py:69:log_dist] [Rank 0] step=1320, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
66%|██████▌ | 1320/2013 [1:27:16<46:44, 4.05s/it][2022-01-08 03:06:40,452] [INFO] [timer.py:181:stop] 0/1320, SamplesPerSec=3.779084920941463
66%|██████▌ | 1329/2013 [1:27:52<46:08, 4.05s/it][2022-01-08 03:07:20,892] [INFO] [logging.py:69:log_dist] [Rank 0] step=1330, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:07:20,921] [INFO] [timer.py:181:stop] 0/1330, SamplesPerSec=3.778544062223671
67%|██████▋ | 1339/2013 [1:28:32<45:28, 4.05s/it][2022-01-08 03:08:01,365] [INFO] [logging.py:69:log_dist] [Rank 0] step=1340, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
67%|██████▋ | 1340/2013 [1:28:37<45:23, 4.05s/it][2022-01-08 03:08:01,394] [INFO] [timer.py:181:stop] 0/1340, SamplesPerSec=3.7780095812402856
67%|██████▋ | 1349/2013 [1:29:13<44:46, 4.05s/it][2022-01-08 03:08:41,829] [INFO] [logging.py:69:log_dist] [Rank 0] step=1350, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
67%|██████▋ | 1350/2013 [1:29:17<44:42, 4.05s/it][2022-01-08 03:08:41,858] [INFO] [timer.py:181:stop] 0/1350, SamplesPerSec=3.777488797080223
{'loss': 1.1174, 'learning_rate': 5e-05, 'epoch': 2.88}
68%|██████▊ | 1359/2013 [1:29:53<44:06, 4.05s/it][2022-01-08 03:09:22,310] [INFO] [logging.py:69:log_dist] [Rank 0] step=1360, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:09:22,339] [INFO] [timer.py:181:stop] 0/1360, SamplesPerSec=3.77696479003716
68%|██████▊ | 1369/2013 [1:30:34<43:27, 4.05s/it][2022-01-08 03:10:02,782] [INFO] [logging.py:69:log_dist] [Rank 0] step=1370, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:10:02,811] [INFO] [timer.py:181:stop] 0/1370, SamplesPerSec=3.7764540670625775
69%|██████▊ | 1379/2013 [1:31:14<42:46, 4.05s/it][2022-01-08 03:10:43,257] [INFO] [logging.py:69:log_dist] [Rank 0] step=1380, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
69%|██████▊ | 1380/2013 [1:31:18<42:42, 4.05s/it][2022-01-08 03:10:43,286] [INFO] [timer.py:181:stop] 0/1380, SamplesPerSec=3.775948567947394
69%|██████▉ | 1389/2013 [1:31:55<42:05, 4.05s/it][2022-01-08 03:11:23,726] [INFO] [logging.py:69:log_dist] [Rank 0] step=1390, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:11:23,756] [INFO] [timer.py:181:stop] 0/1390, SamplesPerSec=3.7754550081573566
69%|██████▉ | 1399/2013 [1:32:35<41:25, 4.05s/it][2022-01-08 03:12:04,201] [INFO] [logging.py:69:log_dist] [Rank 0] step=1400, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
70%|██████▉ | 1400/2013 [1:32:39<41:20, 4.05s/it][2022-01-08 03:12:04,230] [INFO] [timer.py:181:stop] 0/1400, SamplesPerSec=3.774964902791683
{'loss': 1.1311, 'learning_rate': 5e-05, 'epoch': 2.99}
70%|██████▉ | 1409/2013 [1:33:16<40:39, 4.04s/it][2022-01-08 03:12:44,503] [INFO] [logging.py:69:log_dist] [Rank 0] step=1410, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
70%|███████ | 1410/2013 [1:33:20<40:36, 4.04s/it][2022-01-08 03:12:44,532] [INFO] [timer.py:181:stop] 0/1410, SamplesPerSec=3.7745988851220726
70%|███████ | 1419/2013 [1:33:56<40:03, 4.05s/it][2022-01-08 03:13:24,971] [INFO] [logging.py:69:log_dist] [Rank 0] step=1420, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:13:25,000] [INFO] [timer.py:181:stop] 0/1420, SamplesPerSec=3.7741260600756092
71%|███████ | 1429/2013 [1:34:37<39:23, 4.05s/it][2022-01-08 03:14:05,447] [INFO] [logging.py:69:log_dist] [Rank 0] step=1430, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:14:05,477] [INFO] [timer.py:181:stop] 0/1430, SamplesPerSec=3.773655414403905
71%|███████▏ | 1439/2013 [1:35:17<38:43, 4.05s/it][2022-01-08 03:14:45,923] [INFO] [logging.py:69:log_dist] [Rank 0] step=1440, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
72%|███████▏ | 1440/2013 [1:35:21<38:38, 4.05s/it][2022-01-08 03:14:45,953] [INFO] [timer.py:181:stop] 0/1440, SamplesPerSec=3.773191004457191
72%|███████▏ | 1449/2013 [1:35:58<38:02, 4.05s/it][2022-01-08 03:15:26,398] [INFO] [logging.py:69:log_dist] [Rank 0] step=1450, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:15:26,427] [INFO] [timer.py:181:stop] 0/1450, SamplesPerSec=3.772734343995773
{'loss': 0.5773, 'learning_rate': 5e-05, 'epoch': 3.1}
72%|███████▏ | 1459/2013 [1:36:38<37:22, 4.05s/it][2022-01-08 03:16:06,871] [INFO] [logging.py:69:log_dist] [Rank 0] step=1460, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:16:06,901] [INFO] [timer.py:181:stop] 0/1460, SamplesPerSec=3.772284399769869
73%|███████▎ | 1469/2013 [1:37:18<36:41, 4.05s/it][2022-01-08 03:16:47,340] [INFO] [logging.py:69:log_dist] [Rank 0] step=1470, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:16:47,369] [INFO] [timer.py:181:stop] 0/1470, SamplesPerSec=3.7718439338319625
73%|███████▎ | 1479/2013 [1:37:59<36:00, 4.05s/it][2022-01-08 03:17:27,810] [INFO] [logging.py:69:log_dist] [Rank 0] step=1480, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
74%|███████▎ | 1480/2013 [1:38:03<35:56, 4.05s/it][2022-01-08 03:17:27,839] [INFO] [timer.py:181:stop] 0/1480, SamplesPerSec=3.7714084083670345
74%|███████▍ | 1489/2013 [1:38:39<35:20, 4.05s/it][2022-01-08 03:18:08,282] [INFO] [logging.py:69:log_dist] [Rank 0] step=1490, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:18:08,311] [INFO] [timer.py:181:stop] 0/1490, SamplesPerSec=3.7709779720469756
74%|███████▍ | 1499/2013 [1:39:20<34:39, 4.04s/it][2022-01-08 03:18:48,742] [INFO] [logging.py:69:log_dist] [Rank 0] step=1500, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
75%|███████▍ | 1500/2013 [1:39:24<34:35, 4.05s/it][2022-01-08 03:18:48,771] [INFO] [timer.py:181:stop] 0/1500, SamplesPerSec=3.770560906286029
{'loss': 0.5255, 'learning_rate': 5e-05, 'epoch': 3.21}
75%|███████▍ | 1509/2013 [1:40:00<33:59, 4.05s/it][2022-01-08 03:19:29,202] [INFO] [logging.py:69:log_dist] [Rank 0] step=1510, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:19:29,231] [INFO] [timer.py:181:stop] 0/1510, SamplesPerSec=3.7701494417930532
75%|███████▌ | 1519/2013 [1:40:41<33:19, 4.05s/it][2022-01-08 03:20:09,667] [INFO] [logging.py:69:log_dist] [Rank 0] step=1520, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
76%|███████▌ | 1520/2013 [1:40:45<33:15, 4.05s/it][2022-01-08 03:20:09,696] [INFO] [timer.py:181:stop] 0/1520, SamplesPerSec=3.7697410296434013
76%|███████▌ | 1529/2013 [1:41:21<32:38, 4.05s/it][2022-01-08 03:20:50,141] [INFO] [logging.py:69:log_dist] [Rank 0] step=1530, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
76%|███████▌ | 1530/2013 [1:41:25<32:34, 4.05s/it][2022-01-08 03:20:50,170] [INFO] [timer.py:181:stop] 0/1530, SamplesPerSec=3.769331455261203
76%|███████▋ | 1539/2013 [1:42:02<31:58, 4.05s/it][2022-01-08 03:21:30,611] [INFO] [logging.py:69:log_dist] [Rank 0] step=1540, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:21:30,640] [INFO] [timer.py:181:stop] 0/1540, SamplesPerSec=3.7689303914076984
77%|███████▋ | 1549/2013 [1:42:42<31:17, 4.05s/it][2022-01-08 03:22:11,082] [INFO] [logging.py:69:log_dist] [Rank 0] step=1550, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
77%|███████▋ | 1550/2013 [1:42:46<31:13, 4.05s/it][2022-01-08 03:22:11,112] [INFO] [timer.py:181:stop] 0/1550, SamplesPerSec=3.768533280844485
{'loss': 0.517, 'learning_rate': 5e-05, 'epoch': 3.31}
77%|███████▋ | 1559/2013 [1:43:23<30:37, 4.05s/it][2022-01-08 03:22:51,545] [INFO] [logging.py:69:log_dist] [Rank 0] step=1560, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
77%|███████▋ | 1560/2013 [1:43:27<30:32, 4.05s/it][2022-01-08 03:22:51,574] [INFO] [timer.py:181:stop] 0/1560, SamplesPerSec=3.7681478650910862
78%|███████▊ | 1569/2013 [1:44:03<29:57, 4.05s/it][2022-01-08 03:23:32,016] [INFO] [logging.py:69:log_dist] [Rank 0] step=1570, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:23:32,046] [INFO] [timer.py:181:stop] 0/1570, SamplesPerSec=3.7677608605512707
78%|███████▊ | 1579/2013 [1:44:44<29:15, 4.05s/it][2022-01-08 03:24:12,479] [INFO] [logging.py:69:log_dist] [Rank 0] step=1580, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
78%|███████▊ | 1580/2013 [1:44:48<29:12, 4.05s/it][2022-01-08 03:24:12,509] [INFO] [timer.py:181:stop] 0/1580, SamplesPerSec=3.7673837218644195
79%|███████▉ | 1589/2013 [1:45:24<28:35, 4.05s/it][2022-01-08 03:24:52,947] [INFO] [logging.py:69:log_dist] [Rank 0] step=1590, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
79%|███████▉ | 1590/2013 [1:45:28<28:31, 4.05s/it][2022-01-08 03:24:52,976] [INFO] [timer.py:181:stop] 0/1590, SamplesPerSec=3.767008802632748
79%|███████▉ | 1599/2013 [1:46:05<27:55, 4.05s/it][2022-01-08 03:25:33,413] [INFO] [logging.py:69:log_dist] [Rank 0] step=1600, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:25:33,442] [INFO] [timer.py:181:stop] 0/1600, SamplesPerSec=3.7666400234585478
{'loss': 0.5351, 'learning_rate': 5e-05, 'epoch': 3.42}
80%|███████▉ | 1609/2013 [1:46:45<27:14, 4.05s/it][2022-01-08 03:26:13,877] [INFO] [logging.py:69:log_dist] [Rank 0] step=1610, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
80%|███████▉ | 1610/2013 [1:46:49<27:10, 4.05s/it][2022-01-08 03:26:13,906] [INFO] [timer.py:181:stop] 0/1610, SamplesPerSec=3.7662768067385883
80%|████████ | 1619/2013 [1:47:25<26:34, 4.05s/it][2022-01-08 03:26:54,343] [INFO] [logging.py:69:log_dist] [Rank 0] step=1620, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:26:54,372] [INFO] [timer.py:181:stop] 0/1620, SamplesPerSec=3.7659171247485923
81%|████████ | 1629/2013 [1:48:06<25:53, 4.05s/it][2022-01-08 03:27:34,812] [INFO] [logging.py:69:log_dist] [Rank 0] step=1630, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
81%|████████ | 1630/2013 [1:48:10<25:49, 4.05s/it][2022-01-08 03:27:34,842] [INFO] [timer.py:181:stop] 0/1630, SamplesPerSec=3.7655598188243307
81%|████████▏ | 1639/2013 [1:48:46<25:13, 4.05s/it][2022-01-08 03:28:15,282] [INFO] [logging.py:69:log_dist] [Rank 0] step=1640, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:28:15,311] [INFO] [timer.py:181:stop] 0/1640, SamplesPerSec=3.7652069853642303
82%|████████▏ | 1649/2013 [1:49:27<24:33, 4.05s/it][2022-01-08 03:28:55,749] [INFO] [logging.py:69:log_dist] [Rank 0] step=1650, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
82%|████████▏ | 1650/2013 [1:49:31<24:28, 4.05s/it][2022-01-08 03:28:55,778] [INFO] [timer.py:181:stop] 0/1650, SamplesPerSec=3.7648604344072325
{'loss': 0.5438, 'learning_rate': 5e-05, 'epoch': 3.53}
82%|████████▏ | 1659/2013 [1:50:07<23:52, 4.05s/it][2022-01-08 03:29:36,213] [INFO] [logging.py:69:log_dist] [Rank 0] step=1660, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:29:36,243] [INFO] [timer.py:181:stop] 0/1660, SamplesPerSec=3.7645192147909743
83%|████████▎ | 1669/2013 [1:50:48<23:12, 4.05s/it][2022-01-08 03:30:16,679] [INFO] [logging.py:69:log_dist] [Rank 0] step=1670, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:30:16,709] [INFO] [timer.py:181:stop] 0/1670, SamplesPerSec=3.764180955869947
83%|████████▎ | 1679/2013 [1:51:28<22:31, 4.05s/it][2022-01-08 03:30:57,144] [INFO] [logging.py:69:log_dist] [Rank 0] step=1680, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
83%|████████▎ | 1680/2013 [1:51:32<22:27, 4.05s/it][2022-01-08 03:30:57,173] [INFO] [timer.py:181:stop] 0/1680, SamplesPerSec=3.7638476574908815
84%|████████▍ | 1689/2013 [1:52:09<21:51, 4.05s/it][2022-01-08 03:31:37,615] [INFO] [logging.py:69:log_dist] [Rank 0] step=1690, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:31:37,645] [INFO] [timer.py:181:stop] 0/1690, SamplesPerSec=3.7635141535703847
84%|████████▍ | 1699/2013 [1:52:49<21:11, 4.05s/it][2022-01-08 03:32:18,093] [INFO] [logging.py:69:log_dist] [Rank 0] step=1700, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:32:18,122] [INFO] [timer.py:181:stop] 0/1700, SamplesPerSec=3.763182119064589
{'loss': 0.5523, 'learning_rate': 5e-05, 'epoch': 3.63}
85%|████████▍ | 1709/2013 [1:53:30<20:30, 4.05s/it][2022-01-08 03:32:58,563] [INFO] [logging.py:69:log_dist] [Rank 0] step=1710, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:32:58,592] [INFO] [timer.py:181:stop] 0/1710, SamplesPerSec=3.7628579234650497
85%|████████▌ | 1719/2013 [1:54:10<19:49, 4.05s/it][2022-01-08 03:33:39,036] [INFO] [logging.py:69:log_dist] [Rank 0] step=1720, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:33:39,065] [INFO] [timer.py:181:stop] 0/1720, SamplesPerSec=3.762535997206537
86%|████████▌ | 1729/2013 [1:54:51<19:09, 4.05s/it][2022-01-08 03:34:19,497] [INFO] [logging.py:69:log_dist] [Rank 0] step=1730, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:34:19,527] [INFO] [timer.py:181:stop] 0/1730, SamplesPerSec=3.762223845502526
86%|████████▋ | 1739/2013 [1:55:31<18:28, 4.05s/it][2022-01-08 03:34:59,961] [INFO] [logging.py:69:log_dist] [Rank 0] step=1740, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:34:59,990] [INFO] [timer.py:181:stop] 0/1740, SamplesPerSec=3.761914204829434
87%|████████▋ | 1749/2013 [1:56:12<17:48, 4.05s/it][2022-01-08 03:35:40,431] [INFO] [logging.py:69:log_dist] [Rank 0] step=1750, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
87%|████████▋ | 1750/2013 [1:56:16<17:44, 4.05s/it][2022-01-08 03:35:40,460] [INFO] [timer.py:181:stop] 0/1750, SamplesPerSec=3.7616048576112933
{'loss': 0.5605, 'learning_rate': 5e-05, 'epoch': 3.74}
87%|████████▋ | 1759/2013 [1:56:52<17:08, 4.05s/it][2022-01-08 03:36:20,905] [INFO] [logging.py:69:log_dist] [Rank 0] step=1760, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:36:20,935] [INFO] [timer.py:181:stop] 0/1760, SamplesPerSec=3.7612965120380717
88%|████████▊ | 1769/2013 [1:57:33<16:27, 4.05s/it][2022-01-08 03:37:01,377] [INFO] [logging.py:69:log_dist] [Rank 0] step=1770, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:37:01,406] [INFO] [timer.py:181:stop] 0/1770, SamplesPerSec=3.7609931824089133
88%|████████▊ | 1779/2013 [1:58:13<15:47, 4.05s/it][2022-01-08 03:37:41,845] [INFO] [logging.py:69:log_dist] [Rank 0] step=1780, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:37:41,874] [INFO] [timer.py:181:stop] 0/1780, SamplesPerSec=3.760695119732919
89%|████████▉ | 1789/2013 [1:58:53<15:06, 4.05s/it][2022-01-08 03:38:22,320] [INFO] [logging.py:69:log_dist] [Rank 0] step=1790, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:38:22,349] [INFO] [timer.py:181:stop] 0/1790, SamplesPerSec=3.7603970137415414
89%|████████▉ | 1799/2013 [1:59:34<14:26, 4.05s/it][2022-01-08 03:39:02,795] [INFO] [logging.py:69:log_dist] [Rank 0] step=1800, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
89%|████████▉ | 1800/2013 [1:59:38<14:21, 4.05s/it][2022-01-08 03:39:02,824] [INFO] [timer.py:181:stop] 0/1800, SamplesPerSec=3.760101944483825
{'loss': 0.5709, 'learning_rate': 5e-05, 'epoch': 3.85}
90%|████████▉ | 1809/2013 [2:00:14<13:45, 4.05s/it][2022-01-08 03:39:43,270] [INFO] [logging.py:69:log_dist] [Rank 0] step=1810, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
90%|████████▉ | 1810/2013 [2:00:18<13:41, 4.05s/it][2022-01-08 03:39:43,299] [INFO] [timer.py:181:stop] 0/1810, SamplesPerSec=3.759810848057715
90%|█████████ | 1819/2013 [2:00:55<13:05, 4.05s/it][2022-01-08 03:40:23,740] [INFO] [logging.py:69:log_dist] [Rank 0] step=1820, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
90%|█████████ | 1820/2013 [2:00:59<13:00, 4.05s/it][2022-01-08 03:40:23,769] [INFO] [timer.py:181:stop] 0/1820, SamplesPerSec=3.7595248648974176
91%|█████████ | 1829/2013 [2:01:35<12:24, 4.05s/it][2022-01-08 03:41:04,207] [INFO] [logging.py:69:log_dist] [Rank 0] step=1830, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:41:04,236] [INFO] [timer.py:181:stop] 0/1830, SamplesPerSec=3.7592437188094854
91%|█████████▏| 1839/2013 [2:02:16<11:44, 4.05s/it][2022-01-08 03:41:44,681] [INFO] [logging.py:69:log_dist] [Rank 0] step=1840, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:41:44,711] [INFO] [timer.py:181:stop] 0/1840, SamplesPerSec=3.7589622119151223
92%|█████████▏| 1849/2013 [2:02:56<11:03, 4.05s/it][2022-01-08 03:42:25,155] [INFO] [logging.py:69:log_dist] [Rank 0] step=1850, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
92%|█████████▏| 1850/2013 [2:03:00<10:59, 4.05s/it][2022-01-08 03:42:25,184] [INFO] [timer.py:181:stop] 0/1850, SamplesPerSec=3.7586840886245376
{'loss': 0.5853, 'learning_rate': 5e-05, 'epoch': 3.95}
92%|█████████▏| 1859/2013 [2:03:37<10:23, 4.05s/it][2022-01-08 03:43:05,629] [INFO] [logging.py:69:log_dist] [Rank 0] step=1860, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
92%|█████████▏| 1860/2013 [2:03:41<10:19, 4.05s/it][2022-01-08 03:43:05,659] [INFO] [timer.py:181:stop] 0/1860, SamplesPerSec=3.7584088253658683
93%|█████████▎| 1869/2013 [2:04:17<09:42, 4.05s/it][2022-01-08 03:43:46,100] [INFO] [logging.py:69:log_dist] [Rank 0] step=1870, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
93%|█████████▎| 1870/2013 [2:04:21<09:38, 4.05s/it][2022-01-08 03:43:46,130] [INFO] [timer.py:181:stop] 0/1870, SamplesPerSec=3.7581376962393236
93%|█████████▎| 1879/2013 [2:04:58<09:01, 4.04s/it][2022-01-08 03:44:26,386] [INFO] [logging.py:69:log_dist] [Rank 0] step=1880, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:44:26,415] [INFO] [timer.py:181:stop] 0/1880, SamplesPerSec=3.7579629767145355
94%|█████████▍| 1889/2013 [2:05:38<08:21, 4.05s/it][2022-01-08 03:45:06,858] [INFO] [logging.py:69:log_dist] [Rank 0] step=1890, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:45:06,888] [INFO] [timer.py:181:stop] 0/1890, SamplesPerSec=3.757696318318126
94%|█████████▍| 1899/2013 [2:06:18<07:41, 4.05s/it][2022-01-08 03:45:47,324] [INFO] [logging.py:69:log_dist] [Rank 0] step=1900, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
94%|█████████▍| 1900/2013 [2:06:22<07:37, 4.05s/it][2022-01-08 03:45:47,353] [INFO] [timer.py:181:stop] 0/1900, SamplesPerSec=3.75743633895728
{'loss': 0.4146, 'learning_rate': 5e-05, 'epoch': 4.06}
95%|█████████▍| 1909/2013 [2:06:59<07:00, 4.05s/it][2022-01-08 03:46:27,788] [INFO] [logging.py:69:log_dist] [Rank 0] step=1910, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:46:27,817] [INFO] [timer.py:181:stop] 0/1910, SamplesPerSec=3.7571801921643524
95%|█████████▌| 1919/2013 [2:07:39<06:20, 4.05s/it][2022-01-08 03:47:08,245] [INFO] [logging.py:69:log_dist] [Rank 0] step=1920, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
95%|█████████▌| 1920/2013 [2:07:43<06:16, 4.05s/it][2022-01-08 03:47:08,274] [INFO] [timer.py:181:stop] 0/1920, SamplesPerSec=3.7569294448435056
96%|█████████▌| 1929/2013 [2:08:20<05:39, 4.05s/it][2022-01-08 03:47:48,710] [INFO] [logging.py:69:log_dist] [Rank 0] step=1930, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
96%|█████████▌| 1930/2013 [2:08:24<05:35, 4.05s/it][2022-01-08 03:47:48,740] [INFO] [timer.py:181:stop] 0/1930, SamplesPerSec=3.7566775980522458
96%|█████████▋| 1939/2013 [2:09:00<04:59, 4.05s/it][2022-01-08 03:48:29,177] [INFO] [logging.py:69:log_dist] [Rank 0] step=1940, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
96%|█████████▋| 1940/2013 [2:09:04<04:55, 4.05s/it][2022-01-08 03:48:29,207] [INFO] [timer.py:181:stop] 0/1940, SamplesPerSec=3.7564275051520726
97%|█████████▋| 1949/2013 [2:09:41<04:18, 4.05s/it][2022-01-08 03:49:09,639] [INFO] [logging.py:69:log_dist] [Rank 0] step=1950, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:49:09,668] [INFO] [timer.py:181:stop] 0/1950, SamplesPerSec=3.756182735788761
{'loss': 0.2756, 'learning_rate': 5e-05, 'epoch': 4.17}
97%|█████████▋| 1959/2013 [2:10:21<03:38, 4.05s/it][2022-01-08 03:49:50,112] [INFO] [logging.py:69:log_dist] [Rank 0] step=1960, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
97%|█████████▋| 1960/2013 [2:10:25<03:34, 4.05s/it][2022-01-08 03:49:50,141] [INFO] [timer.py:181:stop] 0/1960, SamplesPerSec=3.75593523788703
98%|█████████▊| 1969/2013 [2:11:02<02:58, 4.05s/it][2022-01-08 03:50:30,581] [INFO] [logging.py:69:log_dist] [Rank 0] step=1970, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
98%|█████████▊| 1970/2013 [2:11:06<02:53, 4.05s/it][2022-01-08 03:50:30,610] [INFO] [timer.py:181:stop] 0/1970, SamplesPerSec=3.755692055378648
98%|█████████▊| 1979/2013 [2:11:42<02:17, 4.05s/it][2022-01-08 03:51:11,056] [INFO] [logging.py:69:log_dist] [Rank 0] step=1980, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:51:11,086] [INFO] [timer.py:181:stop] 0/1980, SamplesPerSec=3.7554480096485228
99%|█████████▉| 1989/2013 [2:12:23<01:37, 4.05s/it][2022-01-08 03:51:51,528] [INFO] [logging.py:69:log_dist] [Rank 0] step=1990, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
99%|█████████▉| 1990/2013 [2:12:27<01:33, 4.05s/it][2022-01-08 03:51:51,558] [INFO] [timer.py:181:stop] 0/1990, SamplesPerSec=3.7552084247483197
99%|█████████▉| 1999/2013 [2:13:03<00:56, 4.05s/it][2022-01-08 03:52:32,004] [INFO] [logging.py:69:log_dist] [Rank 0] step=2000, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
[2022-01-08 03:52:32,033] [INFO] [timer.py:181:stop] 0/2000, SamplesPerSec=3.7549697349900724
{'loss': 0.2783, 'learning_rate': 5e-05, 'epoch': 4.27}
100%|█████████▉| 2009/2013 [2:13:44<00:16, 4.08s/it][2022-01-08 03:53:12,904] [INFO] [logging.py:69:log_dist] [Rank 0] step=2010, skipped=35, lr=[5e-05], mom=[[0.9, 0.999]]
100%|█████████▉| 2010/2013 [2:13:48<00:12, 4.08s/it][2022-01-08 03:53:12,942] [INFO] [timer.py:181:stop] 0/2010, SamplesPerSec=3.754532461339503
100%|██████████| 2013/2013 [2:14:00<00:00, 4.09s/it]
{'train_runtime': 8040.9292, 'train_samples_per_second': 3.748, 'train_steps_per_second': 0.25, 'train_loss': 1.4952462921673186, 'epoch': 4.3}
Training completed. Do not forget to share your model on huggingface.co/models =)
100%|██████████| 2013/2013 [2:14:00<00:00, 3.99s/it]
Setting `pad_token_id` to `eos_token_id`:50256 for open-end generation.
0: While playing matchmaking games amid Argentine tradition crushing under the thumb of a father's radical ideologies, five students from divergent backgrounds find their paths crossing in this anthology film.
1: As a lonely woman awaits her high school girlfriend C quirks at the studentoves at the senior year, college separation becomes the most likely destination.
2: Hosts Sarah Thae-dies and Julian Mazzaro-Lopes gorge on pizza masters and masters of goblin Salem, Cesur, Tinker, and more in an all-new special.
3: Fowered by reuse chic urban style Gold artisaneer Gold embarks on an upscale quest for urban treasure in this reboot of the TV series Thwarted in the modern era.
4: From top-notch hangovers and rec center breakout Tomato, embark on comedic adventures across four countries in this Bollywood entertainment series.
5: An Afghanistan veteran's second attempt to kick-start his acting career while juggling a family affair puts him in the middle of a conflict involving political corruption, romantic rivalries and familiesubs.
6: Five hopeful friends journey south to meet royalty in the land of a royal consort and navigate crushing poverty in an area where development is impossible.
7: Amid the turbulent 1920s, five women launch a bold campaign against the man ordained to transform them into robotnic cab drivers. Each makes a perilous journey toward victory.
8: When the Irish MMA star gamer Colin Kennedy becomes sidelined by an accident, his political ambitions turn to private violence.
9: Join P. in his quest to find seven commonalities among humans living on Earth, to unearth the future of modern cults. Baking ventriloquist Syd Foster improvised characterizations for Pigskin parties in 12oos.
10: An off-color joke nearly got violent, triggering a deep obsession in an Argentine detective who’s involved with one victims of every cult at one peak. Reached epic — though comic Genoa Gray didn’t intend to get a big story.
11: A group of young college roomers gets caught up in a terrifying hook-up hell organized by a predatory vampire leader and cop Port nightclub gang Head vampire Squidgy. Faster episodes available on Netflix.
12: After waking up at a strange time in a barren town under a sky curses Rainbow city crewmember Aang Mok's weird adventures continue as he makes new friends and gets captured by an awful fate.
13: Amid last-minute fire in 1980 New York, the Neverending Que Delaps Mr. Fire Seguro and Fire Seguro, two rival shoe designers putting their visions for the webster sales strategy into action.
14: Trolls kidnap the king's daughters and overtake the kingdom of Messaba, desperate for affield with the monster Rani. Boon Pie becomes BFFs and BFFs fight for freedom.
15: Hosts Mike Huckle and Olivia Newton Oslo host the newest game show in which several players get a unique chance at victory. Host Emily Calandrelli explores the science behind life's many hidden riches.
16: From fangs to claws to venomous stings, it's no wonder how bad people manages to survive in the jungle, on the hunt for bad surprises.
17: Five individuals in 20th-century Korea find interconnected destinies as they pursue diverse path to happiness throughout the nation's Trojan horse Sen.
18: As communal violence erupts in his neighborhood, an undocumented immigrant set to return to Canada incurses the ire of two bitter rivals as internal war ignites.
19: When a father of two teen girls asks his students if they'd like sex, Mitsuba and Aotsuba Canary join host Mitsuhisa Neume to go out with these contestants.
Process finished with exit code 0