bbytxt commited on
Commit
7ba1705
·
verified ·
1 Parent(s): d80b911

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 sha256:52f2e10836f31257cb3feab4b885c3e3885f464c61d340b444ff2b388e404861
3
  size 63592
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ccfafaf348148ea96c61473c7dacf00540d621e7ab619148a5d06d246e680396
3
  size 63592
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d83500d9b5cbda5304e4aaacadf2e490f608fea82f85ea2c8b7787722b8ad3a1
3
  size 136814
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bdb9dc1e52ca08b5670b0a206602bac85192fdb0aaef101ff006de78d3536b3
3
  size 136814
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:23c1344dcabbad12b9c4b38041604444af120fa9d028c0690033ba68b5cb0dc0
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1c9330007793a699f9fb95e5b45a1a7c86fb5e1b3f0ce05c20fc24f6f47d20b
3
  size 14244
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7ad2841b888ce0ae948634757c3fcacf0119c249e0fec8f3ca61ea266369ef92
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5d2a6c6aafc669cea03b9634666f204de949a3d45ce2f48a07e7e3eaf18c715
3
  size 1064
last-checkpoint/trainer_state.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "best_metric": 10.520312309265137,
3
- "best_model_checkpoint": "miner_id_24/checkpoint-50",
4
- "epoch": 0.03297609233305853,
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": 441.817,
382
  "eval_steps_per_second": 61.854,
383
  "step": 50
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
384
  }
385
  ],
386
  "logging_steps": 1,
@@ -409,7 +775,7 @@
409
  "attributes": {}
410
  }
411
  },
412
- "total_flos": 397737984000.0,
413
  "train_batch_size": 8,
414
  "trial_name": null,
415
  "trial_params": null
 
1
  {
2
+ "best_metric": 10.496562957763672,
3
+ "best_model_checkpoint": "miner_id_24/checkpoint-100",
4
+ "epoch": 0.06595218466611706,
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": 441.817,
382
  "eval_steps_per_second": 61.854,
383
  "step": 50
384
+ },
385
+ {
386
+ "epoch": 0.033635614179719706,
387
+ "grad_norm": 1.5558362007141113,
388
+ "learning_rate": 0.0002668315918143169,
389
+ "loss": 42.1289,
390
+ "step": 51
391
+ },
392
+ {
393
+ "epoch": 0.034295136026380874,
394
+ "grad_norm": 1.4583945274353027,
395
+ "learning_rate": 0.00026526016662852886,
396
+ "loss": 42.0273,
397
+ "step": 52
398
+ },
399
+ {
400
+ "epoch": 0.03495465787304204,
401
+ "grad_norm": 1.2193175554275513,
402
+ "learning_rate": 0.00026365723046405023,
403
+ "loss": 42.0605,
404
+ "step": 53
405
+ },
406
+ {
407
+ "epoch": 0.035614179719703216,
408
+ "grad_norm": 1.212590217590332,
409
+ "learning_rate": 0.0002620232215476231,
410
+ "loss": 42.0234,
411
+ "step": 54
412
+ },
413
+ {
414
+ "epoch": 0.036273701566364384,
415
+ "grad_norm": 0.9861271381378174,
416
+ "learning_rate": 0.0002603585866009697,
417
+ "loss": 42.0625,
418
+ "step": 55
419
+ },
420
+ {
421
+ "epoch": 0.03693322341302556,
422
+ "grad_norm": 0.9094260931015015,
423
+ "learning_rate": 0.00025866378071866334,
424
+ "loss": 42.0195,
425
+ "step": 56
426
+ },
427
+ {
428
+ "epoch": 0.03759274525968673,
429
+ "grad_norm": 0.9702939391136169,
430
+ "learning_rate": 0.00025693926724370956,
431
+ "loss": 42.0547,
432
+ "step": 57
433
+ },
434
+ {
435
+ "epoch": 0.038252267106347895,
436
+ "grad_norm": 0.8315839767456055,
437
+ "learning_rate": 0.00025518551764087326,
438
+ "loss": 42.0605,
439
+ "step": 58
440
+ },
441
+ {
442
+ "epoch": 0.03891178895300907,
443
+ "grad_norm": 0.9290674328804016,
444
+ "learning_rate": 0.00025340301136778483,
445
+ "loss": 42.0391,
446
+ "step": 59
447
+ },
448
+ {
449
+ "epoch": 0.03957131079967024,
450
+ "grad_norm": 1.1367566585540771,
451
+ "learning_rate": 0.00025159223574386114,
452
+ "loss": 41.9785,
453
+ "step": 60
454
+ },
455
+ {
456
+ "epoch": 0.04023083264633141,
457
+ "grad_norm": 0.842949390411377,
458
+ "learning_rate": 0.0002497536858170772,
459
+ "loss": 42.0508,
460
+ "step": 61
461
+ },
462
+ {
463
+ "epoch": 0.04089035449299258,
464
+ "grad_norm": 0.7557629346847534,
465
+ "learning_rate": 0.00024788786422862526,
466
+ "loss": 42.0332,
467
+ "step": 62
468
+ },
469
+ {
470
+ "epoch": 0.04154987633965375,
471
+ "grad_norm": 0.6981768608093262,
472
+ "learning_rate": 0.00024599528107549745,
473
+ "loss": 42.0215,
474
+ "step": 63
475
+ },
476
+ {
477
+ "epoch": 0.042209398186314924,
478
+ "grad_norm": 0.6374576687812805,
479
+ "learning_rate": 0.00024407645377103054,
480
+ "loss": 42.0039,
481
+ "step": 64
482
+ },
483
+ {
484
+ "epoch": 0.04286892003297609,
485
+ "grad_norm": 0.7283995151519775,
486
+ "learning_rate": 0.00024213190690345018,
487
+ "loss": 41.998,
488
+ "step": 65
489
+ },
490
+ {
491
+ "epoch": 0.04352844187963726,
492
+ "grad_norm": 0.7016935348510742,
493
+ "learning_rate": 0.00024016217209245374,
494
+ "loss": 41.9688,
495
+ "step": 66
496
+ },
497
+ {
498
+ "epoch": 0.044187963726298435,
499
+ "grad_norm": 0.7185855507850647,
500
+ "learning_rate": 0.00023816778784387094,
501
+ "loss": 41.9727,
502
+ "step": 67
503
+ },
504
+ {
505
+ "epoch": 0.0448474855729596,
506
+ "grad_norm": 0.5180225372314453,
507
+ "learning_rate": 0.0002361492994024415,
508
+ "loss": 41.9844,
509
+ "step": 68
510
+ },
511
+ {
512
+ "epoch": 0.04550700741962078,
513
+ "grad_norm": 0.7438845038414001,
514
+ "learning_rate": 0.0002341072586027509,
515
+ "loss": 41.959,
516
+ "step": 69
517
+ },
518
+ {
519
+ "epoch": 0.046166529266281946,
520
+ "grad_norm": 0.6051533818244934,
521
+ "learning_rate": 0.00023204222371836405,
522
+ "loss": 41.9551,
523
+ "step": 70
524
+ },
525
+ {
526
+ "epoch": 0.046826051112943114,
527
+ "grad_norm": 0.5988097190856934,
528
+ "learning_rate": 0.00022995475930919905,
529
+ "loss": 42.0098,
530
+ "step": 71
531
+ },
532
+ {
533
+ "epoch": 0.04748557295960429,
534
+ "grad_norm": 0.6344655156135559,
535
+ "learning_rate": 0.00022784543606718227,
536
+ "loss": 42.0039,
537
+ "step": 72
538
+ },
539
+ {
540
+ "epoch": 0.04814509480626546,
541
+ "grad_norm": 0.6249358057975769,
542
+ "learning_rate": 0.00022571483066022657,
543
+ "loss": 41.998,
544
+ "step": 73
545
+ },
546
+ {
547
+ "epoch": 0.04880461665292663,
548
+ "grad_norm": 0.6966153383255005,
549
+ "learning_rate": 0.0002235635255745762,
550
+ "loss": 41.9434,
551
+ "step": 74
552
+ },
553
+ {
554
+ "epoch": 0.0494641384995878,
555
+ "grad_norm": 0.6041496396064758,
556
+ "learning_rate": 0.00022139210895556104,
557
+ "loss": 41.9766,
558
+ "step": 75
559
+ },
560
+ {
561
+ "epoch": 0.0494641384995878,
562
+ "eval_loss": 10.489999771118164,
563
+ "eval_runtime": 0.1147,
564
+ "eval_samples_per_second": 436.085,
565
+ "eval_steps_per_second": 61.052,
566
+ "step": 75
567
+ },
568
+ {
569
+ "epoch": 0.05012366034624897,
570
+ "grad_norm": 0.6852412223815918,
571
+ "learning_rate": 0.00021920117444680317,
572
+ "loss": 42.0117,
573
+ "step": 76
574
+ },
575
+ {
576
+ "epoch": 0.05078318219291014,
577
+ "grad_norm": 0.43333616852760315,
578
+ "learning_rate": 0.00021699132102792097,
579
+ "loss": 41.9805,
580
+ "step": 77
581
+ },
582
+ {
583
+ "epoch": 0.05144270403957131,
584
+ "grad_norm": 0.6226640343666077,
585
+ "learning_rate": 0.0002147631528507739,
586
+ "loss": 41.9941,
587
+ "step": 78
588
+ },
589
+ {
590
+ "epoch": 0.05210222588623248,
591
+ "grad_norm": 0.57357257604599,
592
+ "learning_rate": 0.00021251727907429355,
593
+ "loss": 42.0098,
594
+ "step": 79
595
+ },
596
+ {
597
+ "epoch": 0.05276174773289365,
598
+ "grad_norm": 0.4521426260471344,
599
+ "learning_rate": 0.0002102543136979454,
600
+ "loss": 42.0098,
601
+ "step": 80
602
+ },
603
+ {
604
+ "epoch": 0.05342126957955482,
605
+ "grad_norm": 0.4179379642009735,
606
+ "learning_rate": 0.0002079748753938678,
607
+ "loss": 41.9688,
608
+ "step": 81
609
+ },
610
+ {
611
+ "epoch": 0.054080791426215996,
612
+ "grad_norm": 0.421734094619751,
613
+ "learning_rate": 0.0002056795873377331,
614
+ "loss": 41.9785,
615
+ "step": 82
616
+ },
617
+ {
618
+ "epoch": 0.054740313272877164,
619
+ "grad_norm": 0.3774226903915405,
620
+ "learning_rate": 0.00020336907703837748,
621
+ "loss": 41.9785,
622
+ "step": 83
623
+ },
624
+ {
625
+ "epoch": 0.05539983511953833,
626
+ "grad_norm": 0.3708513379096985,
627
+ "learning_rate": 0.00020104397616624645,
628
+ "loss": 41.9727,
629
+ "step": 84
630
+ },
631
+ {
632
+ "epoch": 0.05605935696619951,
633
+ "grad_norm": 0.42900216579437256,
634
+ "learning_rate": 0.00019870492038070252,
635
+ "loss": 41.957,
636
+ "step": 85
637
+ },
638
+ {
639
+ "epoch": 0.056718878812860675,
640
+ "grad_norm": 0.46859726309776306,
641
+ "learning_rate": 0.0001963525491562421,
642
+ "loss": 41.9707,
643
+ "step": 86
644
+ },
645
+ {
646
+ "epoch": 0.05737840065952185,
647
+ "grad_norm": 0.34871360659599304,
648
+ "learning_rate": 0.0001939875056076697,
649
+ "loss": 41.9727,
650
+ "step": 87
651
+ },
652
+ {
653
+ "epoch": 0.05803792250618302,
654
+ "grad_norm": 0.3781905472278595,
655
+ "learning_rate": 0.00019161043631427666,
656
+ "loss": 41.9902,
657
+ "step": 88
658
+ },
659
+ {
660
+ "epoch": 0.058697444352844186,
661
+ "grad_norm": 0.6564770936965942,
662
+ "learning_rate": 0.00018922199114307294,
663
+ "loss": 41.9258,
664
+ "step": 89
665
+ },
666
+ {
667
+ "epoch": 0.05935696619950536,
668
+ "grad_norm": 0.7199106812477112,
669
+ "learning_rate": 0.00018682282307111987,
670
+ "loss": 41.9258,
671
+ "step": 90
672
+ },
673
+ {
674
+ "epoch": 0.06001648804616653,
675
+ "grad_norm": 0.3589068651199341,
676
+ "learning_rate": 0.00018441358800701273,
677
+ "loss": 41.9805,
678
+ "step": 91
679
+ },
680
+ {
681
+ "epoch": 0.0606760098928277,
682
+ "grad_norm": 0.4372813403606415,
683
+ "learning_rate": 0.00018199494461156203,
684
+ "loss": 41.998,
685
+ "step": 92
686
+ },
687
+ {
688
+ "epoch": 0.06133553173948887,
689
+ "grad_norm": 0.4405193626880646,
690
+ "learning_rate": 0.000179567554117722,
691
+ "loss": 41.9453,
692
+ "step": 93
693
+ },
694
+ {
695
+ "epoch": 0.06199505358615004,
696
+ "grad_norm": 0.7273338437080383,
697
+ "learning_rate": 0.00017713208014981648,
698
+ "loss": 41.9199,
699
+ "step": 94
700
+ },
701
+ {
702
+ "epoch": 0.06265457543281121,
703
+ "grad_norm": 0.4150626063346863,
704
+ "learning_rate": 0.00017468918854211007,
705
+ "loss": 41.9609,
706
+ "step": 95
707
+ },
708
+ {
709
+ "epoch": 0.06331409727947239,
710
+ "grad_norm": 0.4945806860923767,
711
+ "learning_rate": 0.00017223954715677627,
712
+ "loss": 41.9844,
713
+ "step": 96
714
+ },
715
+ {
716
+ "epoch": 0.06397361912613356,
717
+ "grad_norm": 0.36564674973487854,
718
+ "learning_rate": 0.00016978382570131034,
719
+ "loss": 41.9512,
720
+ "step": 97
721
+ },
722
+ {
723
+ "epoch": 0.06463314097279473,
724
+ "grad_norm": 0.632477879524231,
725
+ "learning_rate": 0.00016732269554543794,
726
+ "loss": 41.918,
727
+ "step": 98
728
+ },
729
+ {
730
+ "epoch": 0.0652926628194559,
731
+ "grad_norm": 0.24244551360607147,
732
+ "learning_rate": 0.00016485682953756942,
733
+ "loss": 41.9531,
734
+ "step": 99
735
+ },
736
+ {
737
+ "epoch": 0.06595218466611706,
738
+ "grad_norm": 0.39558830857276917,
739
+ "learning_rate": 0.00016238690182084986,
740
+ "loss": 41.9551,
741
+ "step": 100
742
+ },
743
+ {
744
+ "epoch": 0.06595218466611706,
745
+ "eval_loss": 10.496562957763672,
746
+ "eval_runtime": 0.1143,
747
+ "eval_samples_per_second": 437.308,
748
+ "eval_steps_per_second": 61.223,
749
+ "step": 100
750
  }
751
  ],
752
  "logging_steps": 1,
 
775
  "attributes": {}
776
  }
777
  },
778
+ "total_flos": 795475968000.0,
779
  "train_batch_size": 8,
780
  "trial_name": null,
781
  "trial_params": null