nttx commited on
Commit
1188cd9
·
verified ·
1 Parent(s): 1005230

Training in progress, step 100, checkpoint

Browse files
last-checkpoint/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5f560512508b1e8101260e7417815fcf548ffbea4a2f5964b1cbad55c1c7a2da
3
  size 50503544
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7acd314e7dbdbbdea46f52c386df7b5309c699ecea84275877231b956ab7cb65
3
  size 50503544
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c9a7195752e5790a3a110dddb6946dba23b886c22a5320bd47d461e29867fdfc
3
  size 101184122
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfdae3642c63b2bca1f00dfb24935cc05a120284bc2b572e3f4285a8a156cc27
3
  size 101184122
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3810bd4304c368d8c3912bf3de23792c39bc07e9866dbd3caa351faf25d687de
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2e2f4372fe2def646a495be7bf6bdbfcbbe51dd56796a745275c2592b7219c3
3
  size 14244
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:106f7f3431a1e35fa851f7ed4f625269613fcac6eee4e94ffe77a1ae094fbfee
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dcdf921ba93c40432f46569609c07042f6599be99ba80522cb2d9b52ee37d8be
3
  size 1064
last-checkpoint/trainer_state.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "best_metric": 1.7979708909988403,
3
- "best_model_checkpoint": "miner_id_24/checkpoint-50",
4
- "epoch": 0.5698005698005698,
5
  "eval_steps": 25,
6
- "global_step": 50,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
@@ -381,6 +381,372 @@
381
  "eval_samples_per_second": 29.433,
382
  "eval_steps_per_second": 4.121,
383
  "step": 50
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
384
  }
385
  ],
386
  "logging_steps": 1,
@@ -409,7 +775,7 @@
409
  "attributes": {}
410
  }
411
  },
412
- "total_flos": 1.2522914467282944e+16,
413
  "train_batch_size": 8,
414
  "trial_name": null,
415
  "trial_params": null
 
1
  {
2
+ "best_metric": 1.4987461566925049,
3
+ "best_model_checkpoint": "miner_id_24/checkpoint-100",
4
+ "epoch": 1.1424501424501425,
5
  "eval_steps": 25,
6
+ "global_step": 100,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
 
381
  "eval_samples_per_second": 29.433,
382
  "eval_steps_per_second": 4.121,
383
  "step": 50
384
+ },
385
+ {
386
+ "epoch": 0.5811965811965812,
387
+ "grad_norm": 1.1083546876907349,
388
+ "learning_rate": 0.0002570649169030708,
389
+ "loss": 1.9054,
390
+ "step": 51
391
+ },
392
+ {
393
+ "epoch": 0.5925925925925926,
394
+ "grad_norm": 1.1118803024291992,
395
+ "learning_rate": 0.0002550576202307026,
396
+ "loss": 1.5566,
397
+ "step": 52
398
+ },
399
+ {
400
+ "epoch": 0.603988603988604,
401
+ "grad_norm": 1.1292729377746582,
402
+ "learning_rate": 0.00025301269668202516,
403
+ "loss": 1.5418,
404
+ "step": 53
405
+ },
406
+ {
407
+ "epoch": 0.6153846153846154,
408
+ "grad_norm": 1.0905343294143677,
409
+ "learning_rate": 0.0002509308786559378,
410
+ "loss": 1.567,
411
+ "step": 54
412
+ },
413
+ {
414
+ "epoch": 0.6267806267806267,
415
+ "grad_norm": 1.1679775714874268,
416
+ "learning_rate": 0.00024881291176526903,
417
+ "loss": 1.478,
418
+ "step": 55
419
+ },
420
+ {
421
+ "epoch": 0.6381766381766382,
422
+ "grad_norm": 1.0859359502792358,
423
+ "learning_rate": 0.00024665955456973154,
424
+ "loss": 1.3628,
425
+ "step": 56
426
+ },
427
+ {
428
+ "epoch": 0.6495726495726496,
429
+ "grad_norm": 1.2914154529571533,
430
+ "learning_rate": 0.00024447157830424066,
431
+ "loss": 1.4341,
432
+ "step": 57
433
+ },
434
+ {
435
+ "epoch": 0.6609686609686609,
436
+ "grad_norm": 4.928417205810547,
437
+ "learning_rate": 0.00024224976660269302,
438
+ "loss": 1.7669,
439
+ "step": 58
440
+ },
441
+ {
442
+ "epoch": 0.6723646723646723,
443
+ "grad_norm": 1.7731802463531494,
444
+ "learning_rate": 0.0002399949152173043,
445
+ "loss": 1.3646,
446
+ "step": 59
447
+ },
448
+ {
449
+ "epoch": 0.6837606837606838,
450
+ "grad_norm": 1.7798770666122437,
451
+ "learning_rate": 0.00023770783173360704,
452
+ "loss": 1.32,
453
+ "step": 60
454
+ },
455
+ {
456
+ "epoch": 0.6951566951566952,
457
+ "grad_norm": 1.3400565385818481,
458
+ "learning_rate": 0.00023538933528120988,
459
+ "loss": 1.0693,
460
+ "step": 61
461
+ },
462
+ {
463
+ "epoch": 0.7065527065527065,
464
+ "grad_norm": 1.7840746641159058,
465
+ "learning_rate": 0.00023304025624042263,
466
+ "loss": 1.2359,
467
+ "step": 62
468
+ },
469
+ {
470
+ "epoch": 0.717948717948718,
471
+ "grad_norm": 3.244560956954956,
472
+ "learning_rate": 0.00023066143594485178,
473
+ "loss": 2.1066,
474
+ "step": 63
475
+ },
476
+ {
477
+ "epoch": 0.7293447293447294,
478
+ "grad_norm": 1.5248583555221558,
479
+ "learning_rate": 0.00022825372638007267,
480
+ "loss": 2.2681,
481
+ "step": 64
482
+ },
483
+ {
484
+ "epoch": 0.7407407407407407,
485
+ "grad_norm": 1.579367995262146,
486
+ "learning_rate": 0.0002258179898784871,
487
+ "loss": 2.1084,
488
+ "step": 65
489
+ },
490
+ {
491
+ "epoch": 0.7521367521367521,
492
+ "grad_norm": 1.5203423500061035,
493
+ "learning_rate": 0.00022335509881047497,
494
+ "loss": 2.0729,
495
+ "step": 66
496
+ },
497
+ {
498
+ "epoch": 0.7635327635327636,
499
+ "grad_norm": 1.4482111930847168,
500
+ "learning_rate": 0.00022086593527195062,
501
+ "loss": 2.0057,
502
+ "step": 67
503
+ },
504
+ {
505
+ "epoch": 0.7749287749287749,
506
+ "grad_norm": 1.1762452125549316,
507
+ "learning_rate": 0.00021835139076843623,
508
+ "loss": 1.811,
509
+ "step": 68
510
+ },
511
+ {
512
+ "epoch": 0.7863247863247863,
513
+ "grad_norm": 0.9151526093482971,
514
+ "learning_rate": 0.00021581236589576476,
515
+ "loss": 1.5962,
516
+ "step": 69
517
+ },
518
+ {
519
+ "epoch": 0.7977207977207977,
520
+ "grad_norm": 1.0164101123809814,
521
+ "learning_rate": 0.00021324977001752757,
522
+ "loss": 1.5083,
523
+ "step": 70
524
+ },
525
+ {
526
+ "epoch": 0.8091168091168092,
527
+ "grad_norm": 0.9672901630401611,
528
+ "learning_rate": 0.00021066452093938153,
529
+ "loss": 1.5546,
530
+ "step": 71
531
+ },
532
+ {
533
+ "epoch": 0.8205128205128205,
534
+ "grad_norm": 1.0712361335754395,
535
+ "learning_rate": 0.0002080575445803326,
536
+ "loss": 1.5119,
537
+ "step": 72
538
+ },
539
+ {
540
+ "epoch": 0.8319088319088319,
541
+ "grad_norm": 1.0604687929153442,
542
+ "learning_rate": 0.00020542977464111352,
543
+ "loss": 1.6416,
544
+ "step": 73
545
+ },
546
+ {
547
+ "epoch": 0.8433048433048433,
548
+ "grad_norm": 1.0448590517044067,
549
+ "learning_rate": 0.00020278215226977493,
550
+ "loss": 1.5965,
551
+ "step": 74
552
+ },
553
+ {
554
+ "epoch": 0.8547008547008547,
555
+ "grad_norm": 1.0953928232192993,
556
+ "learning_rate": 0.0002001156257246085,
557
+ "loss": 1.5314,
558
+ "step": 75
559
+ },
560
+ {
561
+ "epoch": 0.8547008547008547,
562
+ "eval_loss": 1.6335612535476685,
563
+ "eval_runtime": 1.6991,
564
+ "eval_samples_per_second": 29.427,
565
+ "eval_steps_per_second": 4.12,
566
+ "step": 75
567
+ },
568
+ {
569
+ "epoch": 0.8660968660968661,
570
+ "grad_norm": 1.022580862045288,
571
+ "learning_rate": 0.00019743115003452357,
572
+ "loss": 1.394,
573
+ "step": 76
574
+ },
575
+ {
576
+ "epoch": 0.8774928774928775,
577
+ "grad_norm": 1.150123119354248,
578
+ "learning_rate": 0.0001947296866569998,
579
+ "loss": 1.0808,
580
+ "step": 77
581
+ },
582
+ {
583
+ "epoch": 0.8888888888888888,
584
+ "grad_norm": 1.2115788459777832,
585
+ "learning_rate": 0.00019201220313373607,
586
+ "loss": 1.4386,
587
+ "step": 78
588
+ },
589
+ {
590
+ "epoch": 0.9002849002849003,
591
+ "grad_norm": 1.132363200187683,
592
+ "learning_rate": 0.00018927967274412098,
593
+ "loss": 1.1949,
594
+ "step": 79
595
+ },
596
+ {
597
+ "epoch": 0.9116809116809117,
598
+ "grad_norm": 1.3680353164672852,
599
+ "learning_rate": 0.00018653307415664877,
600
+ "loss": 1.2202,
601
+ "step": 80
602
+ },
603
+ {
604
+ "epoch": 0.9230769230769231,
605
+ "grad_norm": 1.075006127357483,
606
+ "learning_rate": 0.00018377339107840412,
607
+ "loss": 0.9873,
608
+ "step": 81
609
+ },
610
+ {
611
+ "epoch": 0.9344729344729344,
612
+ "grad_norm": 1.3361048698425293,
613
+ "learning_rate": 0.0001810016119027429,
614
+ "loss": 1.2343,
615
+ "step": 82
616
+ },
617
+ {
618
+ "epoch": 0.9458689458689459,
619
+ "grad_norm": 1.9565190076828003,
620
+ "learning_rate": 0.00017821872935529505,
621
+ "loss": 1.0822,
622
+ "step": 83
623
+ },
624
+ {
625
+ "epoch": 0.9572649572649573,
626
+ "grad_norm": 3.1362149715423584,
627
+ "learning_rate": 0.0001754257401384145,
628
+ "loss": 1.6614,
629
+ "step": 84
630
+ },
631
+ {
632
+ "epoch": 0.9686609686609686,
633
+ "grad_norm": 1.214998722076416,
634
+ "learning_rate": 0.00017262364457420608,
635
+ "loss": 1.8554,
636
+ "step": 85
637
+ },
638
+ {
639
+ "epoch": 0.98005698005698,
640
+ "grad_norm": 1.4069725275039673,
641
+ "learning_rate": 0.00016981344624625536,
642
+ "loss": 1.6906,
643
+ "step": 86
644
+ },
645
+ {
646
+ "epoch": 0.9914529914529915,
647
+ "grad_norm": 1.7144482135772705,
648
+ "learning_rate": 0.0001669961516401905,
649
+ "loss": 1.095,
650
+ "step": 87
651
+ },
652
+ {
653
+ "epoch": 1.0056980056980056,
654
+ "grad_norm": 2.2465782165527344,
655
+ "learning_rate": 0.00016417276978320468,
656
+ "loss": 2.0945,
657
+ "step": 88
658
+ },
659
+ {
660
+ "epoch": 1.017094017094017,
661
+ "grad_norm": 1.1649279594421387,
662
+ "learning_rate": 0.00016134431188266851,
663
+ "loss": 1.9323,
664
+ "step": 89
665
+ },
666
+ {
667
+ "epoch": 1.0284900284900285,
668
+ "grad_norm": 1.0408798456192017,
669
+ "learning_rate": 0.00015851179096396112,
670
+ "loss": 1.8263,
671
+ "step": 90
672
+ },
673
+ {
674
+ "epoch": 1.03988603988604,
675
+ "grad_norm": 1.0593069791793823,
676
+ "learning_rate": 0.00015567622150765057,
677
+ "loss": 1.8236,
678
+ "step": 91
679
+ },
680
+ {
681
+ "epoch": 1.0512820512820513,
682
+ "grad_norm": 0.940461277961731,
683
+ "learning_rate": 0.00015283861908615284,
684
+ "loss": 1.4745,
685
+ "step": 92
686
+ },
687
+ {
688
+ "epoch": 1.0626780626780628,
689
+ "grad_norm": 0.8671309351921082,
690
+ "learning_rate": 0.00015,
691
+ "loss": 1.3637,
692
+ "step": 93
693
+ },
694
+ {
695
+ "epoch": 1.074074074074074,
696
+ "grad_norm": 0.7851026058197021,
697
+ "learning_rate": 0.00014716138091384716,
698
+ "loss": 1.3622,
699
+ "step": 94
700
+ },
701
+ {
702
+ "epoch": 1.0854700854700854,
703
+ "grad_norm": 0.8758441805839539,
704
+ "learning_rate": 0.00014432377849234946,
705
+ "loss": 1.4627,
706
+ "step": 95
707
+ },
708
+ {
709
+ "epoch": 1.0968660968660968,
710
+ "grad_norm": 0.9090460538864136,
711
+ "learning_rate": 0.00014148820903603888,
712
+ "loss": 1.245,
713
+ "step": 96
714
+ },
715
+ {
716
+ "epoch": 1.1082621082621082,
717
+ "grad_norm": 0.8923598527908325,
718
+ "learning_rate": 0.00013865568811733151,
719
+ "loss": 1.3085,
720
+ "step": 97
721
+ },
722
+ {
723
+ "epoch": 1.1196581196581197,
724
+ "grad_norm": 0.9927690625190735,
725
+ "learning_rate": 0.00013582723021679532,
726
+ "loss": 1.2563,
727
+ "step": 98
728
+ },
729
+ {
730
+ "epoch": 1.131054131054131,
731
+ "grad_norm": 1.0350650548934937,
732
+ "learning_rate": 0.0001330038483598095,
733
+ "loss": 1.1181,
734
+ "step": 99
735
+ },
736
+ {
737
+ "epoch": 1.1424501424501425,
738
+ "grad_norm": 1.156041145324707,
739
+ "learning_rate": 0.00013018655375374467,
740
+ "loss": 1.1825,
741
+ "step": 100
742
+ },
743
+ {
744
+ "epoch": 1.1424501424501425,
745
+ "eval_loss": 1.4987461566925049,
746
+ "eval_runtime": 1.8846,
747
+ "eval_samples_per_second": 26.531,
748
+ "eval_steps_per_second": 3.714,
749
+ "step": 100
750
  }
751
  ],
752
  "logging_steps": 1,
 
775
  "attributes": {}
776
  }
777
  },
778
+ "total_flos": 2.4663032422465536e+16,
779
  "train_batch_size": 8,
780
  "trial_name": null,
781
  "trial_params": null