nttx commited on
Commit
8359ba5
·
verified ·
1 Parent(s): e25c049

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 sha256:396beccc2e3df7c8d45082af215bb44e02467037ca7a1f80ff3e8fc1eabffdce
3
  size 167832240
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a021a576323171bd8dda11800b5d3112abb2fe6a9250ee98c71990d9330a022c
3
  size 167832240
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ee99aa715dcf6da1cb736110e1c0fd39d574168329e81f4d794a196a5c7b591b
3
  size 335922386
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9d07d7fac5431b641d71fddbefa114760c75c765f60fa88de3dc9c189cf5dec
3
  size 335922386
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8f6a6b7361968c99fcbbebddd48cbc234439c874c2cf299c0cb531c754d8a73c
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6f40c27f40f4db8063cf59f47e450371ab83bb30e2dc338471413b28f5fa986b
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": 0.760506808757782,
3
- "best_model_checkpoint": "miner_id_24/checkpoint-50",
4
- "epoch": 0.7120605251446372,
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": 11.961,
382
  "eval_steps_per_second": 1.675,
383
  "step": 50
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
384
  }
385
  ],
386
  "logging_steps": 1,
@@ -409,7 +775,7 @@
409
  "attributes": {}
410
  }
411
  },
412
- "total_flos": 7.88255238782976e+16,
413
  "train_batch_size": 1,
414
  "trial_name": null,
415
  "trial_params": null
 
1
  {
2
+ "best_metric": 0.6796185970306396,
3
+ "best_model_checkpoint": "miner_id_24/checkpoint-100",
4
+ "epoch": 1.4241210502892745,
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": 11.961,
382
  "eval_steps_per_second": 1.675,
383
  "step": 50
384
+ },
385
+ {
386
+ "epoch": 0.7263017356475301,
387
+ "grad_norm": 3.0040671825408936,
388
+ "learning_rate": 0.0002668315918143169,
389
+ "loss": 0.8039,
390
+ "step": 51
391
+ },
392
+ {
393
+ "epoch": 0.7405429461504228,
394
+ "grad_norm": 3.1810615062713623,
395
+ "learning_rate": 0.00026526016662852886,
396
+ "loss": 0.7996,
397
+ "step": 52
398
+ },
399
+ {
400
+ "epoch": 0.7547841566533156,
401
+ "grad_norm": 2.66048526763916,
402
+ "learning_rate": 0.00026365723046405023,
403
+ "loss": 0.7514,
404
+ "step": 53
405
+ },
406
+ {
407
+ "epoch": 0.7690253671562083,
408
+ "grad_norm": 2.6719470024108887,
409
+ "learning_rate": 0.0002620232215476231,
410
+ "loss": 0.7929,
411
+ "step": 54
412
+ },
413
+ {
414
+ "epoch": 0.783266577659101,
415
+ "grad_norm": 2.316591739654541,
416
+ "learning_rate": 0.0002603585866009697,
417
+ "loss": 0.7603,
418
+ "step": 55
419
+ },
420
+ {
421
+ "epoch": 0.7975077881619937,
422
+ "grad_norm": 1.115259051322937,
423
+ "learning_rate": 0.00025866378071866334,
424
+ "loss": 0.6665,
425
+ "step": 56
426
+ },
427
+ {
428
+ "epoch": 0.8117489986648865,
429
+ "grad_norm": 0.5784854292869568,
430
+ "learning_rate": 0.00025693926724370956,
431
+ "loss": 0.6798,
432
+ "step": 57
433
+ },
434
+ {
435
+ "epoch": 0.8259902091677792,
436
+ "grad_norm": 0.49967318773269653,
437
+ "learning_rate": 0.00025518551764087326,
438
+ "loss": 0.6836,
439
+ "step": 58
440
+ },
441
+ {
442
+ "epoch": 0.840231419670672,
443
+ "grad_norm": 0.8447709083557129,
444
+ "learning_rate": 0.00025340301136778483,
445
+ "loss": 0.7267,
446
+ "step": 59
447
+ },
448
+ {
449
+ "epoch": 0.8544726301735648,
450
+ "grad_norm": 0.5715835094451904,
451
+ "learning_rate": 0.00025159223574386114,
452
+ "loss": 0.6858,
453
+ "step": 60
454
+ },
455
+ {
456
+ "epoch": 0.8687138406764575,
457
+ "grad_norm": 1.0063656568527222,
458
+ "learning_rate": 0.0002497536858170772,
459
+ "loss": 0.706,
460
+ "step": 61
461
+ },
462
+ {
463
+ "epoch": 0.8829550511793502,
464
+ "grad_norm": 0.9968670606613159,
465
+ "learning_rate": 0.00024788786422862526,
466
+ "loss": 0.6861,
467
+ "step": 62
468
+ },
469
+ {
470
+ "epoch": 0.897196261682243,
471
+ "grad_norm": 0.9589150547981262,
472
+ "learning_rate": 0.00024599528107549745,
473
+ "loss": 0.7194,
474
+ "step": 63
475
+ },
476
+ {
477
+ "epoch": 0.9114374721851357,
478
+ "grad_norm": 0.7447915077209473,
479
+ "learning_rate": 0.00024407645377103054,
480
+ "loss": 0.7086,
481
+ "step": 64
482
+ },
483
+ {
484
+ "epoch": 0.9256786826880284,
485
+ "grad_norm": 1.025489091873169,
486
+ "learning_rate": 0.00024213190690345018,
487
+ "loss": 0.6846,
488
+ "step": 65
489
+ },
490
+ {
491
+ "epoch": 0.9399198931909212,
492
+ "grad_norm": 1.9707061052322388,
493
+ "learning_rate": 0.00024016217209245374,
494
+ "loss": 0.7149,
495
+ "step": 66
496
+ },
497
+ {
498
+ "epoch": 0.954161103693814,
499
+ "grad_norm": 0.826673686504364,
500
+ "learning_rate": 0.00023816778784387094,
501
+ "loss": 0.6999,
502
+ "step": 67
503
+ },
504
+ {
505
+ "epoch": 0.9684023141967067,
506
+ "grad_norm": 1.027491807937622,
507
+ "learning_rate": 0.0002361492994024415,
508
+ "loss": 0.7006,
509
+ "step": 68
510
+ },
511
+ {
512
+ "epoch": 0.9826435246995995,
513
+ "grad_norm": 0.5436568260192871,
514
+ "learning_rate": 0.0002341072586027509,
515
+ "loss": 0.6732,
516
+ "step": 69
517
+ },
518
+ {
519
+ "epoch": 0.9968847352024922,
520
+ "grad_norm": 1.2011053562164307,
521
+ "learning_rate": 0.00023204222371836405,
522
+ "loss": 0.7055,
523
+ "step": 70
524
+ },
525
+ {
526
+ "epoch": 1.011125945705385,
527
+ "grad_norm": 0.7172683477401733,
528
+ "learning_rate": 0.00022995475930919905,
529
+ "loss": 1.1875,
530
+ "step": 71
531
+ },
532
+ {
533
+ "epoch": 1.0253671562082778,
534
+ "grad_norm": 0.8648049831390381,
535
+ "learning_rate": 0.00022784543606718227,
536
+ "loss": 0.6959,
537
+ "step": 72
538
+ },
539
+ {
540
+ "epoch": 1.0396083667111704,
541
+ "grad_norm": 1.807977557182312,
542
+ "learning_rate": 0.00022571483066022657,
543
+ "loss": 0.7421,
544
+ "step": 73
545
+ },
546
+ {
547
+ "epoch": 1.0538495772140632,
548
+ "grad_norm": 1.7105003595352173,
549
+ "learning_rate": 0.0002235635255745762,
550
+ "loss": 0.7546,
551
+ "step": 74
552
+ },
553
+ {
554
+ "epoch": 1.0680907877169559,
555
+ "grad_norm": 0.626197338104248,
556
+ "learning_rate": 0.00022139210895556104,
557
+ "loss": 0.6896,
558
+ "step": 75
559
+ },
560
+ {
561
+ "epoch": 1.0680907877169559,
562
+ "eval_loss": 0.719901442527771,
563
+ "eval_runtime": 4.188,
564
+ "eval_samples_per_second": 11.939,
565
+ "eval_steps_per_second": 1.671,
566
+ "step": 75
567
+ },
568
+ {
569
+ "epoch": 1.0823319982198487,
570
+ "grad_norm": 1.092750906944275,
571
+ "learning_rate": 0.00021920117444680317,
572
+ "loss": 0.7064,
573
+ "step": 76
574
+ },
575
+ {
576
+ "epoch": 1.0965732087227413,
577
+ "grad_norm": 0.5488213896751404,
578
+ "learning_rate": 0.00021699132102792097,
579
+ "loss": 0.6757,
580
+ "step": 77
581
+ },
582
+ {
583
+ "epoch": 1.1108144192256342,
584
+ "grad_norm": 1.0106656551361084,
585
+ "learning_rate": 0.0002147631528507739,
586
+ "loss": 0.6956,
587
+ "step": 78
588
+ },
589
+ {
590
+ "epoch": 1.125055629728527,
591
+ "grad_norm": 0.6059918999671936,
592
+ "learning_rate": 0.00021251727907429355,
593
+ "loss": 0.696,
594
+ "step": 79
595
+ },
596
+ {
597
+ "epoch": 1.1392968402314196,
598
+ "grad_norm": 0.7169640064239502,
599
+ "learning_rate": 0.0002102543136979454,
600
+ "loss": 0.7019,
601
+ "step": 80
602
+ },
603
+ {
604
+ "epoch": 1.1535380507343125,
605
+ "grad_norm": 0.8470028638839722,
606
+ "learning_rate": 0.0002079748753938678,
607
+ "loss": 0.6867,
608
+ "step": 81
609
+ },
610
+ {
611
+ "epoch": 1.167779261237205,
612
+ "grad_norm": 0.5408539175987244,
613
+ "learning_rate": 0.0002056795873377331,
614
+ "loss": 0.6932,
615
+ "step": 82
616
+ },
617
+ {
618
+ "epoch": 1.182020471740098,
619
+ "grad_norm": 0.7373553514480591,
620
+ "learning_rate": 0.00020336907703837748,
621
+ "loss": 0.7023,
622
+ "step": 83
623
+ },
624
+ {
625
+ "epoch": 1.1962616822429906,
626
+ "grad_norm": 0.552183985710144,
627
+ "learning_rate": 0.00020104397616624645,
628
+ "loss": 0.7053,
629
+ "step": 84
630
+ },
631
+ {
632
+ "epoch": 1.2105028927458834,
633
+ "grad_norm": 0.7189547419548035,
634
+ "learning_rate": 0.00019870492038070252,
635
+ "loss": 0.6598,
636
+ "step": 85
637
+ },
638
+ {
639
+ "epoch": 1.224744103248776,
640
+ "grad_norm": 0.7592505812644958,
641
+ "learning_rate": 0.0001963525491562421,
642
+ "loss": 0.6908,
643
+ "step": 86
644
+ },
645
+ {
646
+ "epoch": 1.2389853137516689,
647
+ "grad_norm": 0.42698734998703003,
648
+ "learning_rate": 0.0001939875056076697,
649
+ "loss": 0.6838,
650
+ "step": 87
651
+ },
652
+ {
653
+ "epoch": 1.2532265242545617,
654
+ "grad_norm": 0.5881170034408569,
655
+ "learning_rate": 0.00019161043631427666,
656
+ "loss": 0.6908,
657
+ "step": 88
658
+ },
659
+ {
660
+ "epoch": 1.2674677347574543,
661
+ "grad_norm": 1.1356054544448853,
662
+ "learning_rate": 0.00018922199114307294,
663
+ "loss": 0.6936,
664
+ "step": 89
665
+ },
666
+ {
667
+ "epoch": 1.2817089452603472,
668
+ "grad_norm": 0.5578824281692505,
669
+ "learning_rate": 0.00018682282307111987,
670
+ "loss": 0.6895,
671
+ "step": 90
672
+ },
673
+ {
674
+ "epoch": 1.2959501557632398,
675
+ "grad_norm": 1.2620768547058105,
676
+ "learning_rate": 0.00018441358800701273,
677
+ "loss": 0.7067,
678
+ "step": 91
679
+ },
680
+ {
681
+ "epoch": 1.3101913662661326,
682
+ "grad_norm": 0.9779208898544312,
683
+ "learning_rate": 0.00018199494461156203,
684
+ "loss": 0.7215,
685
+ "step": 92
686
+ },
687
+ {
688
+ "epoch": 1.3244325767690253,
689
+ "grad_norm": 0.9356729388237,
690
+ "learning_rate": 0.000179567554117722,
691
+ "loss": 0.6979,
692
+ "step": 93
693
+ },
694
+ {
695
+ "epoch": 1.338673787271918,
696
+ "grad_norm": 0.8828136324882507,
697
+ "learning_rate": 0.00017713208014981648,
698
+ "loss": 0.7469,
699
+ "step": 94
700
+ },
701
+ {
702
+ "epoch": 1.352914997774811,
703
+ "grad_norm": 0.33228611946105957,
704
+ "learning_rate": 0.00017468918854211007,
705
+ "loss": 0.6815,
706
+ "step": 95
707
+ },
708
+ {
709
+ "epoch": 1.3671562082777036,
710
+ "grad_norm": 0.744838535785675,
711
+ "learning_rate": 0.00017223954715677627,
712
+ "loss": 0.696,
713
+ "step": 96
714
+ },
715
+ {
716
+ "epoch": 1.3813974187805964,
717
+ "grad_norm": 0.5581339597702026,
718
+ "learning_rate": 0.00016978382570131034,
719
+ "loss": 0.6878,
720
+ "step": 97
721
+ },
722
+ {
723
+ "epoch": 1.395638629283489,
724
+ "grad_norm": 0.6438109278678894,
725
+ "learning_rate": 0.00016732269554543794,
726
+ "loss": 0.7204,
727
+ "step": 98
728
+ },
729
+ {
730
+ "epoch": 1.4098798397863819,
731
+ "grad_norm": 0.4048614501953125,
732
+ "learning_rate": 0.00016485682953756942,
733
+ "loss": 0.682,
734
+ "step": 99
735
+ },
736
+ {
737
+ "epoch": 1.4241210502892745,
738
+ "grad_norm": 0.6994701027870178,
739
+ "learning_rate": 0.00016238690182084986,
740
+ "loss": 0.6491,
741
+ "step": 100
742
+ },
743
+ {
744
+ "epoch": 1.4241210502892745,
745
+ "eval_loss": 0.6796185970306396,
746
+ "eval_runtime": 4.1861,
747
+ "eval_samples_per_second": 11.944,
748
+ "eval_steps_per_second": 1.672,
749
+ "step": 100
750
  }
751
  ],
752
  "logging_steps": 1,
 
775
  "attributes": {}
776
  }
777
  },
778
+ "total_flos": 1.576510477565952e+17,
779
  "train_batch_size": 1,
780
  "trial_name": null,
781
  "trial_params": null