owenisas commited on
Commit
3504033
·
verified ·
1 Parent(s): a642fc9

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 sha256:1c302dc1f7d4b868ed2fec7fb599c56ab89a9be3b061d10a09c33f91bc884118
3
  size 3537299144
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd69768e6c7362e7034401e4891974deb19956c18b21ba3b31f81853efbb627b
3
  size 3537299144
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c82c2ea846bada76c6987cfb10fc7217cfd00b4b82d0021a138e9add209aaec9
3
  size 1798933287
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38440800300a79f86e4139287e715f395d397c23b6a41359b4d1bf2f7d2aa355
3
  size 1798933287
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9ccb8eeb935749fc43744e0a5eeacdf6f0f10253be15266a497cbca0ffaa2573
3
  size 14645
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec2cf52e4c3f15759e904ba93b5a10165bb850df54a226c7977bcfa5b79f76fb
3
  size 14645
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:83429aff07094f43f6ae84f250d5d91c95fca2dfaf4ecddce133674cbbfe1442
3
  size 1465
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2fa42c1f1d72d6fd251db37cabd50277c40b6993f9c2917ab544677f4cec64b3
3
  size 1465
last-checkpoint/trainer_state.json CHANGED
@@ -2,9 +2,9 @@
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
- "epoch": 0.3454231433506045,
6
  "eval_steps": 50,
7
- "global_step": 50,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
@@ -366,6 +366,364 @@
366
  "eval_samples_per_second": 2.414,
367
  "eval_steps_per_second": 0.606,
368
  "step": 50
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
369
  }
370
  ],
371
  "logging_steps": 1,
@@ -385,7 +743,7 @@
385
  "attributes": {}
386
  }
387
  },
388
- "total_flos": 1.0220481364790016e+18,
389
  "train_batch_size": 8,
390
  "trial_name": null,
391
  "trial_params": null
 
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
+ "epoch": 0.690846286701209,
6
  "eval_steps": 50,
7
+ "global_step": 100,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
 
366
  "eval_samples_per_second": 2.414,
367
  "eval_steps_per_second": 0.606,
368
  "step": 50
369
+ },
370
+ {
371
+ "epoch": 0.35233160621761656,
372
+ "grad_norm": 0.26529672741889954,
373
+ "learning_rate": 7.660160382576683e-06,
374
+ "loss": 2.0817,
375
+ "step": 51
376
+ },
377
+ {
378
+ "epoch": 0.3592400690846287,
379
+ "grad_norm": 0.2169308364391327,
380
+ "learning_rate": 7.564496387029532e-06,
381
+ "loss": 1.7478,
382
+ "step": 52
383
+ },
384
+ {
385
+ "epoch": 0.36614853195164077,
386
+ "grad_norm": 0.24268858134746552,
387
+ "learning_rate": 7.467541090321735e-06,
388
+ "loss": 1.8318,
389
+ "step": 53
390
+ },
391
+ {
392
+ "epoch": 0.37305699481865284,
393
+ "grad_norm": 0.2409997284412384,
394
+ "learning_rate": 7.369343312364994e-06,
395
+ "loss": 1.7587,
396
+ "step": 54
397
+ },
398
+ {
399
+ "epoch": 0.3799654576856649,
400
+ "grad_norm": 0.25278374552726746,
401
+ "learning_rate": 7.269952498697734e-06,
402
+ "loss": 1.9785,
403
+ "step": 55
404
+ },
405
+ {
406
+ "epoch": 0.38687392055267705,
407
+ "grad_norm": 0.23796814680099487,
408
+ "learning_rate": 7.169418695587791e-06,
409
+ "loss": 1.8941,
410
+ "step": 56
411
+ },
412
+ {
413
+ "epoch": 0.39378238341968913,
414
+ "grad_norm": 0.21832697093486786,
415
+ "learning_rate": 7.067792524832604e-06,
416
+ "loss": 1.7578,
417
+ "step": 57
418
+ },
419
+ {
420
+ "epoch": 0.4006908462867012,
421
+ "grad_norm": 0.2829532027244568,
422
+ "learning_rate": 6.965125158269619e-06,
423
+ "loss": 2.1754,
424
+ "step": 58
425
+ },
426
+ {
427
+ "epoch": 0.4075993091537133,
428
+ "grad_norm": 0.2926061451435089,
429
+ "learning_rate": 6.8614682920097265e-06,
430
+ "loss": 2.4897,
431
+ "step": 59
432
+ },
433
+ {
434
+ "epoch": 0.41450777202072536,
435
+ "grad_norm": 0.2706138491630554,
436
+ "learning_rate": 6.7568741204067145e-06,
437
+ "loss": 2.0711,
438
+ "step": 60
439
+ },
440
+ {
441
+ "epoch": 0.4214162348877375,
442
+ "grad_norm": 0.22621501982212067,
443
+ "learning_rate": 6.651395309775837e-06,
444
+ "loss": 1.8491,
445
+ "step": 61
446
+ },
447
+ {
448
+ "epoch": 0.4283246977547496,
449
+ "grad_norm": 0.2318510115146637,
450
+ "learning_rate": 6.545084971874738e-06,
451
+ "loss": 1.7088,
452
+ "step": 62
453
+ },
454
+ {
455
+ "epoch": 0.43523316062176165,
456
+ "grad_norm": 0.2396102398633957,
457
+ "learning_rate": 6.437996637160086e-06,
458
+ "loss": 1.8507,
459
+ "step": 63
460
+ },
461
+ {
462
+ "epoch": 0.4421416234887737,
463
+ "grad_norm": 0.26362478733062744,
464
+ "learning_rate": 6.330184227833376e-06,
465
+ "loss": 1.9226,
466
+ "step": 64
467
+ },
468
+ {
469
+ "epoch": 0.44905008635578586,
470
+ "grad_norm": 0.24076825380325317,
471
+ "learning_rate": 6.2217020306894705e-06,
472
+ "loss": 1.7639,
473
+ "step": 65
474
+ },
475
+ {
476
+ "epoch": 0.45595854922279794,
477
+ "grad_norm": 0.24980144202709198,
478
+ "learning_rate": 6.112604669781572e-06,
479
+ "loss": 1.911,
480
+ "step": 66
481
+ },
482
+ {
483
+ "epoch": 0.46286701208981,
484
+ "grad_norm": 0.2638571858406067,
485
+ "learning_rate": 6.002947078916365e-06,
486
+ "loss": 2.0759,
487
+ "step": 67
488
+ },
489
+ {
490
+ "epoch": 0.4697754749568221,
491
+ "grad_norm": 0.29171910881996155,
492
+ "learning_rate": 5.892784473993184e-06,
493
+ "loss": 2.0647,
494
+ "step": 68
495
+ },
496
+ {
497
+ "epoch": 0.47668393782383417,
498
+ "grad_norm": 0.2925715744495392,
499
+ "learning_rate": 5.782172325201155e-06,
500
+ "loss": 2.0447,
501
+ "step": 69
502
+ },
503
+ {
504
+ "epoch": 0.4835924006908463,
505
+ "grad_norm": 0.23386667668819427,
506
+ "learning_rate": 5.671166329088278e-06,
507
+ "loss": 1.852,
508
+ "step": 70
509
+ },
510
+ {
511
+ "epoch": 0.4905008635578584,
512
+ "grad_norm": 0.24327710270881653,
513
+ "learning_rate": 5.559822380516539e-06,
514
+ "loss": 1.8899,
515
+ "step": 71
516
+ },
517
+ {
518
+ "epoch": 0.49740932642487046,
519
+ "grad_norm": 0.2617882192134857,
520
+ "learning_rate": 5.448196544517168e-06,
521
+ "loss": 2.0904,
522
+ "step": 72
523
+ },
524
+ {
525
+ "epoch": 0.5043177892918825,
526
+ "grad_norm": 0.2401256114244461,
527
+ "learning_rate": 5.336345028060199e-06,
528
+ "loss": 1.8795,
529
+ "step": 73
530
+ },
531
+ {
532
+ "epoch": 0.5112262521588946,
533
+ "grad_norm": 0.2712593674659729,
534
+ "learning_rate": 5.224324151752575e-06,
535
+ "loss": 2.0559,
536
+ "step": 74
537
+ },
538
+ {
539
+ "epoch": 0.5181347150259067,
540
+ "grad_norm": 0.2670513391494751,
541
+ "learning_rate": 5.112190321479026e-06,
542
+ "loss": 2.0488,
543
+ "step": 75
544
+ },
545
+ {
546
+ "epoch": 0.5250431778929189,
547
+ "grad_norm": 0.2406473010778427,
548
+ "learning_rate": 5e-06,
549
+ "loss": 1.9381,
550
+ "step": 76
551
+ },
552
+ {
553
+ "epoch": 0.531951640759931,
554
+ "grad_norm": 0.2831808030605316,
555
+ "learning_rate": 4.887809678520976e-06,
556
+ "loss": 2.137,
557
+ "step": 77
558
+ },
559
+ {
560
+ "epoch": 0.538860103626943,
561
+ "grad_norm": 0.2615388333797455,
562
+ "learning_rate": 4.775675848247427e-06,
563
+ "loss": 2.0174,
564
+ "step": 78
565
+ },
566
+ {
567
+ "epoch": 0.5457685664939551,
568
+ "grad_norm": 0.2528943121433258,
569
+ "learning_rate": 4.663654971939802e-06,
570
+ "loss": 1.8625,
571
+ "step": 79
572
+ },
573
+ {
574
+ "epoch": 0.5526770293609672,
575
+ "grad_norm": 0.23635736107826233,
576
+ "learning_rate": 4.551803455482833e-06,
577
+ "loss": 1.8214,
578
+ "step": 80
579
+ },
580
+ {
581
+ "epoch": 0.5595854922279793,
582
+ "grad_norm": 0.27131161093711853,
583
+ "learning_rate": 4.4401776194834615e-06,
584
+ "loss": 1.9002,
585
+ "step": 81
586
+ },
587
+ {
588
+ "epoch": 0.5664939550949913,
589
+ "grad_norm": 0.25781673192977905,
590
+ "learning_rate": 4.3288336709117246e-06,
591
+ "loss": 1.9126,
592
+ "step": 82
593
+ },
594
+ {
595
+ "epoch": 0.5734024179620034,
596
+ "grad_norm": 0.2586870789527893,
597
+ "learning_rate": 4.217827674798845e-06,
598
+ "loss": 1.9539,
599
+ "step": 83
600
+ },
601
+ {
602
+ "epoch": 0.5803108808290155,
603
+ "grad_norm": 0.2633845806121826,
604
+ "learning_rate": 4.107215526006818e-06,
605
+ "loss": 1.9201,
606
+ "step": 84
607
+ },
608
+ {
609
+ "epoch": 0.5872193436960277,
610
+ "grad_norm": 0.2853638231754303,
611
+ "learning_rate": 3.997052921083637e-06,
612
+ "loss": 2.0815,
613
+ "step": 85
614
+ },
615
+ {
616
+ "epoch": 0.5941278065630398,
617
+ "grad_norm": 0.2256501466035843,
618
+ "learning_rate": 3.887395330218429e-06,
619
+ "loss": 1.6324,
620
+ "step": 86
621
+ },
622
+ {
623
+ "epoch": 0.6010362694300518,
624
+ "grad_norm": 0.27014851570129395,
625
+ "learning_rate": 3.778297969310529e-06,
626
+ "loss": 1.8823,
627
+ "step": 87
628
+ },
629
+ {
630
+ "epoch": 0.6079447322970639,
631
+ "grad_norm": 0.2682372033596039,
632
+ "learning_rate": 3.669815772166625e-06,
633
+ "loss": 1.8645,
634
+ "step": 88
635
+ },
636
+ {
637
+ "epoch": 0.614853195164076,
638
+ "grad_norm": 0.2270493060350418,
639
+ "learning_rate": 3.562003362839914e-06,
640
+ "loss": 1.8234,
641
+ "step": 89
642
+ },
643
+ {
644
+ "epoch": 0.6217616580310881,
645
+ "grad_norm": 0.255436509847641,
646
+ "learning_rate": 3.4549150281252635e-06,
647
+ "loss": 1.9353,
648
+ "step": 90
649
+ },
650
+ {
651
+ "epoch": 0.6286701208981001,
652
+ "grad_norm": 0.2589050233364105,
653
+ "learning_rate": 3.3486046902241663e-06,
654
+ "loss": 1.998,
655
+ "step": 91
656
+ },
657
+ {
658
+ "epoch": 0.6355785837651122,
659
+ "grad_norm": 0.24356301128864288,
660
+ "learning_rate": 3.2431258795932863e-06,
661
+ "loss": 1.7773,
662
+ "step": 92
663
+ },
664
+ {
665
+ "epoch": 0.6424870466321243,
666
+ "grad_norm": 0.27413704991340637,
667
+ "learning_rate": 3.1385317079902743e-06,
668
+ "loss": 1.9753,
669
+ "step": 93
670
+ },
671
+ {
672
+ "epoch": 0.6493955094991365,
673
+ "grad_norm": 0.2753913700580597,
674
+ "learning_rate": 3.0348748417303826e-06,
675
+ "loss": 2.1045,
676
+ "step": 94
677
+ },
678
+ {
679
+ "epoch": 0.6563039723661486,
680
+ "grad_norm": 0.29331421852111816,
681
+ "learning_rate": 2.932207475167398e-06,
682
+ "loss": 2.2978,
683
+ "step": 95
684
+ },
685
+ {
686
+ "epoch": 0.6632124352331606,
687
+ "grad_norm": 0.24595493078231812,
688
+ "learning_rate": 2.83058130441221e-06,
689
+ "loss": 1.9159,
690
+ "step": 96
691
+ },
692
+ {
693
+ "epoch": 0.6701208981001727,
694
+ "grad_norm": 0.24188601970672607,
695
+ "learning_rate": 2.7300475013022666e-06,
696
+ "loss": 1.7877,
697
+ "step": 97
698
+ },
699
+ {
700
+ "epoch": 0.6770293609671848,
701
+ "grad_norm": 0.2542547285556793,
702
+ "learning_rate": 2.6306566876350072e-06,
703
+ "loss": 1.9244,
704
+ "step": 98
705
+ },
706
+ {
707
+ "epoch": 0.6839378238341969,
708
+ "grad_norm": 0.2622241973876953,
709
+ "learning_rate": 2.532458909678266e-06,
710
+ "loss": 1.7582,
711
+ "step": 99
712
+ },
713
+ {
714
+ "epoch": 0.690846286701209,
715
+ "grad_norm": 0.25373589992523193,
716
+ "learning_rate": 2.43550361297047e-06,
717
+ "loss": 2.0471,
718
+ "step": 100
719
+ },
720
+ {
721
+ "epoch": 0.690846286701209,
722
+ "eval_loss": 0.5006352066993713,
723
+ "eval_runtime": 214.2546,
724
+ "eval_samples_per_second": 2.399,
725
+ "eval_steps_per_second": 0.602,
726
+ "step": 100
727
  }
728
  ],
729
  "logging_steps": 1,
 
743
  "attributes": {}
744
  }
745
  },
746
+ "total_flos": 2.0400273584358912e+18,
747
  "train_batch_size": 8,
748
  "trial_name": null,
749
  "trial_params": null