Wilsonwin commited on
Commit
21389ef
·
verified ·
1 Parent(s): b958b3a

Training in progress, step 4000, checkpoint

Browse files
last-checkpoint/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:738c2de85f0e51539b3b28ef14af7af58903d06b8cb198594333ab46070e708c
3
  size 328277848
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd5257ed25b3deedcdfbd77b311ce64f39ce97cab4262552a2cce890d0e1ed2f
3
  size 328277848
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:024ddd20febf80999241a3aa205a823795a590616e197f4bf659cd2cd012b820
3
  size 318646859
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38c6d9ddeda93bf2814232d10b3b4a6111c3ba43c271d5af0fe9ac07ad7bdf8f
3
  size 318646859
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:28f4baaa7c055b0898b0ca303ea04e20e5e8e6d3ba1fd3140d755b6a404cb08b
3
  size 14645
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8647979d889bb2b15d0a3e8961a7e547be28d07767d240f858bd959476bb870c
3
  size 14645
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:78485ac70c57dc668ae0e88817673f3f048988e00a751f9337313c02c006f1eb
3
  size 1465
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f34721a2fd924d02bdad3691f09e25bcb5ed140f7982be7b710c4ccbd2538c0
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": 0.5913161006926846,
6
  "eval_steps": 500,
7
- "global_step": 3500,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
@@ -2521,6 +2521,364 @@
2521
  "eval_samples_per_second": 278.824,
2522
  "eval_steps_per_second": 5.855,
2523
  "step": 3500
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2524
  }
2525
  ],
2526
  "logging_steps": 10,
@@ -2540,7 +2898,7 @@
2540
  "attributes": {}
2541
  }
2542
  },
2543
- "total_flos": 1.17059887300608e+17,
2544
  "train_batch_size": 48,
2545
  "trial_name": null,
2546
  "trial_params": null
 
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
+ "epoch": 0.6757898293630681,
6
  "eval_steps": 500,
7
+ "global_step": 4000,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
 
2521
  "eval_samples_per_second": 278.824,
2522
  "eval_steps_per_second": 5.855,
2523
  "step": 3500
2524
+ },
2525
+ {
2526
+ "epoch": 0.5930055752660922,
2527
+ "grad_norm": 0.6327577233314514,
2528
+ "learning_rate": 0.00028291929128998293,
2529
+ "loss": 4.818768310546875,
2530
+ "step": 3510
2531
+ },
2532
+ {
2533
+ "epoch": 0.5946950498395,
2534
+ "grad_norm": 0.5604032278060913,
2535
+ "learning_rate": 0.00028269662709221635,
2536
+ "loss": 4.822237777709961,
2537
+ "step": 3520
2538
+ },
2539
+ {
2540
+ "epoch": 0.5963845244129076,
2541
+ "grad_norm": 0.5533990859985352,
2542
+ "learning_rate": 0.00028247260974544037,
2543
+ "loss": 4.827470779418945,
2544
+ "step": 3530
2545
+ },
2546
+ {
2547
+ "epoch": 0.5980739989863153,
2548
+ "grad_norm": 0.613385796546936,
2549
+ "learning_rate": 0.00028224724153403015,
2550
+ "loss": 4.831082153320312,
2551
+ "step": 3540
2552
+ },
2553
+ {
2554
+ "epoch": 0.5997634735597229,
2555
+ "grad_norm": 0.5741211175918579,
2556
+ "learning_rate": 0.0002820205247561356,
2557
+ "loss": 4.8289031982421875,
2558
+ "step": 3550
2559
+ },
2560
+ {
2561
+ "epoch": 0.6014529481331305,
2562
+ "grad_norm": 0.5649030804634094,
2563
+ "learning_rate": 0.0002817924617236587,
2564
+ "loss": 4.8354541778564455,
2565
+ "step": 3560
2566
+ },
2567
+ {
2568
+ "epoch": 0.6031424227065383,
2569
+ "grad_norm": 0.5777986645698547,
2570
+ "learning_rate": 0.00028156305476222966,
2571
+ "loss": 4.831108856201172,
2572
+ "step": 3570
2573
+ },
2574
+ {
2575
+ "epoch": 0.6048318972799459,
2576
+ "grad_norm": 0.5951328277587891,
2577
+ "learning_rate": 0.0002813323062111828,
2578
+ "loss": 4.814881896972656,
2579
+ "step": 3580
2580
+ },
2581
+ {
2582
+ "epoch": 0.6065213718533536,
2583
+ "grad_norm": 0.6194175481796265,
2584
+ "learning_rate": 0.0002811002184235334,
2585
+ "loss": 4.799482345581055,
2586
+ "step": 3590
2587
+ },
2588
+ {
2589
+ "epoch": 0.6082108464267613,
2590
+ "grad_norm": 0.6000851392745972,
2591
+ "learning_rate": 0.00028086679376595314,
2592
+ "loss": 4.825896072387695,
2593
+ "step": 3600
2594
+ },
2595
+ {
2596
+ "epoch": 0.609900321000169,
2597
+ "grad_norm": 0.5737459659576416,
2598
+ "learning_rate": 0.00028063203461874635,
2599
+ "loss": 4.819542312622071,
2600
+ "step": 3610
2601
+ },
2602
+ {
2603
+ "epoch": 0.6115897955735766,
2604
+ "grad_norm": 0.5582530498504639,
2605
+ "learning_rate": 0.0002803959433758254,
2606
+ "loss": 4.792166137695313,
2607
+ "step": 3620
2608
+ },
2609
+ {
2610
+ "epoch": 0.6132792701469842,
2611
+ "grad_norm": 0.6597942113876343,
2612
+ "learning_rate": 0.0002801585224446866,
2613
+ "loss": 4.817278671264648,
2614
+ "step": 3630
2615
+ },
2616
+ {
2617
+ "epoch": 0.614968744720392,
2618
+ "grad_norm": 0.6341045498847961,
2619
+ "learning_rate": 0.0002799197742463854,
2620
+ "loss": 4.801259613037109,
2621
+ "step": 3640
2622
+ },
2623
+ {
2624
+ "epoch": 0.6166582192937996,
2625
+ "grad_norm": 0.5537974834442139,
2626
+ "learning_rate": 0.0002796797012155118,
2627
+ "loss": 4.792682266235351,
2628
+ "step": 3650
2629
+ },
2630
+ {
2631
+ "epoch": 0.6183476938672073,
2632
+ "grad_norm": 0.5912588238716125,
2633
+ "learning_rate": 0.0002794383058001657,
2634
+ "loss": 4.815021514892578,
2635
+ "step": 3660
2636
+ },
2637
+ {
2638
+ "epoch": 0.620037168440615,
2639
+ "grad_norm": 0.5857402086257935,
2640
+ "learning_rate": 0.00027919559046193156,
2641
+ "loss": 4.807975006103516,
2642
+ "step": 3670
2643
+ },
2644
+ {
2645
+ "epoch": 0.6217266430140226,
2646
+ "grad_norm": 0.5567502379417419,
2647
+ "learning_rate": 0.0002789515576758536,
2648
+ "loss": 4.815755081176758,
2649
+ "step": 3680
2650
+ },
2651
+ {
2652
+ "epoch": 0.6234161175874303,
2653
+ "grad_norm": 0.5999053120613098,
2654
+ "learning_rate": 0.00027870620993041055,
2655
+ "loss": 4.76678237915039,
2656
+ "step": 3690
2657
+ },
2658
+ {
2659
+ "epoch": 0.6251055921608379,
2660
+ "grad_norm": 0.6173041462898254,
2661
+ "learning_rate": 0.00027845954972749004,
2662
+ "loss": 4.78552474975586,
2663
+ "step": 3700
2664
+ },
2665
+ {
2666
+ "epoch": 0.6267950667342457,
2667
+ "grad_norm": 0.620463490486145,
2668
+ "learning_rate": 0.0002782115795823633,
2669
+ "loss": 4.800563812255859,
2670
+ "step": 3710
2671
+ },
2672
+ {
2673
+ "epoch": 0.6284845413076533,
2674
+ "grad_norm": 0.5800994038581848,
2675
+ "learning_rate": 0.0002779623020236594,
2676
+ "loss": 4.786049270629883,
2677
+ "step": 3720
2678
+ },
2679
+ {
2680
+ "epoch": 0.630174015881061,
2681
+ "grad_norm": 0.6238998174667358,
2682
+ "learning_rate": 0.00027771171959333976,
2683
+ "loss": 4.825713348388672,
2684
+ "step": 3730
2685
+ },
2686
+ {
2687
+ "epoch": 0.6318634904544687,
2688
+ "grad_norm": 0.5553771257400513,
2689
+ "learning_rate": 0.00027745983484667164,
2690
+ "loss": 4.7812854766845705,
2691
+ "step": 3740
2692
+ },
2693
+ {
2694
+ "epoch": 0.6335529650278763,
2695
+ "grad_norm": 0.5241145491600037,
2696
+ "learning_rate": 0.0002772066503522026,
2697
+ "loss": 4.793375396728516,
2698
+ "step": 3750
2699
+ },
2700
+ {
2701
+ "epoch": 0.635242439601284,
2702
+ "grad_norm": 0.6024186015129089,
2703
+ "learning_rate": 0.00027695216869173415,
2704
+ "loss": 4.763290786743164,
2705
+ "step": 3760
2706
+ },
2707
+ {
2708
+ "epoch": 0.6369319141746916,
2709
+ "grad_norm": 0.6140456199645996,
2710
+ "learning_rate": 0.0002766963924602953,
2711
+ "loss": 4.783917236328125,
2712
+ "step": 3770
2713
+ },
2714
+ {
2715
+ "epoch": 0.6386213887480994,
2716
+ "grad_norm": 0.58733731508255,
2717
+ "learning_rate": 0.00027643932426611647,
2718
+ "loss": 4.783477401733398,
2719
+ "step": 3780
2720
+ },
2721
+ {
2722
+ "epoch": 0.640310863321507,
2723
+ "grad_norm": 0.5765292048454285,
2724
+ "learning_rate": 0.0002761809667306022,
2725
+ "loss": 4.76646842956543,
2726
+ "step": 3790
2727
+ },
2728
+ {
2729
+ "epoch": 0.6420003378949147,
2730
+ "grad_norm": 0.6646907329559326,
2731
+ "learning_rate": 0.00027592132248830526,
2732
+ "loss": 4.773694229125977,
2733
+ "step": 3800
2734
+ },
2735
+ {
2736
+ "epoch": 0.6436898124683224,
2737
+ "grad_norm": 0.5674440264701843,
2738
+ "learning_rate": 0.00027566039418689905,
2739
+ "loss": 4.759212493896484,
2740
+ "step": 3810
2741
+ },
2742
+ {
2743
+ "epoch": 0.64537928704173,
2744
+ "grad_norm": 0.5485038161277771,
2745
+ "learning_rate": 0.00027539818448715124,
2746
+ "loss": 4.7634929656982425,
2747
+ "step": 3820
2748
+ },
2749
+ {
2750
+ "epoch": 0.6470687616151377,
2751
+ "grad_norm": 0.601775586605072,
2752
+ "learning_rate": 0.000275134696062896,
2753
+ "loss": 4.772266006469726,
2754
+ "step": 3830
2755
+ },
2756
+ {
2757
+ "epoch": 0.6487582361885453,
2758
+ "grad_norm": 0.5745565891265869,
2759
+ "learning_rate": 0.0002748699316010073,
2760
+ "loss": 4.750381851196289,
2761
+ "step": 3840
2762
+ },
2763
+ {
2764
+ "epoch": 0.6504477107619531,
2765
+ "grad_norm": 0.611356258392334,
2766
+ "learning_rate": 0.000274603893801371,
2767
+ "loss": 4.7574516296386715,
2768
+ "step": 3850
2769
+ },
2770
+ {
2771
+ "epoch": 0.6521371853353607,
2772
+ "grad_norm": 0.5577182173728943,
2773
+ "learning_rate": 0.000274336585376858,
2774
+ "loss": 4.7543701171875,
2775
+ "step": 3860
2776
+ },
2777
+ {
2778
+ "epoch": 0.6538266599087684,
2779
+ "grad_norm": 0.5428398251533508,
2780
+ "learning_rate": 0.0002740680090532958,
2781
+ "loss": 4.768844223022461,
2782
+ "step": 3870
2783
+ },
2784
+ {
2785
+ "epoch": 0.655516134482176,
2786
+ "grad_norm": 0.5419198870658875,
2787
+ "learning_rate": 0.0002737981675694411,
2788
+ "loss": 4.7659965515136715,
2789
+ "step": 3880
2790
+ },
2791
+ {
2792
+ "epoch": 0.6572056090555837,
2793
+ "grad_norm": 0.5463002920150757,
2794
+ "learning_rate": 0.00027352706367695203,
2795
+ "loss": 4.753771591186523,
2796
+ "step": 3890
2797
+ },
2798
+ {
2799
+ "epoch": 0.6588950836289914,
2800
+ "grad_norm": 0.5442675948143005,
2801
+ "learning_rate": 0.00027325470014035965,
2802
+ "loss": 4.76721076965332,
2803
+ "step": 3900
2804
+ },
2805
+ {
2806
+ "epoch": 0.660584558202399,
2807
+ "grad_norm": 0.54521244764328,
2808
+ "learning_rate": 0.0002729810797370402,
2809
+ "loss": 4.7523548126220705,
2810
+ "step": 3910
2811
+ },
2812
+ {
2813
+ "epoch": 0.6622740327758068,
2814
+ "grad_norm": 0.5418686866760254,
2815
+ "learning_rate": 0.00027270620525718647,
2816
+ "loss": 4.739299774169922,
2817
+ "step": 3920
2818
+ },
2819
+ {
2820
+ "epoch": 0.6639635073492144,
2821
+ "grad_norm": 0.540998101234436,
2822
+ "learning_rate": 0.0002724300795037796,
2823
+ "loss": 4.779198455810547,
2824
+ "step": 3930
2825
+ },
2826
+ {
2827
+ "epoch": 0.665652981922622,
2828
+ "grad_norm": 0.5339457392692566,
2829
+ "learning_rate": 0.00027215270529256015,
2830
+ "loss": 4.735797882080078,
2831
+ "step": 3940
2832
+ },
2833
+ {
2834
+ "epoch": 0.6673424564960297,
2835
+ "grad_norm": 0.5907135009765625,
2836
+ "learning_rate": 0.00027187408545199977,
2837
+ "loss": 4.734383392333984,
2838
+ "step": 3950
2839
+ },
2840
+ {
2841
+ "epoch": 0.6690319310694374,
2842
+ "grad_norm": 0.5747363567352295,
2843
+ "learning_rate": 0.00027159422282327204,
2844
+ "loss": 4.746397399902344,
2845
+ "step": 3960
2846
+ },
2847
+ {
2848
+ "epoch": 0.6707214056428451,
2849
+ "grad_norm": 0.5700234770774841,
2850
+ "learning_rate": 0.0002713131202602238,
2851
+ "loss": 4.765536499023438,
2852
+ "step": 3970
2853
+ },
2854
+ {
2855
+ "epoch": 0.6724108802162527,
2856
+ "grad_norm": 0.5308473706245422,
2857
+ "learning_rate": 0.0002710307806293458,
2858
+ "loss": 4.714194107055664,
2859
+ "step": 3980
2860
+ },
2861
+ {
2862
+ "epoch": 0.6741003547896605,
2863
+ "grad_norm": 0.5844800472259521,
2864
+ "learning_rate": 0.0002707472068097435,
2865
+ "loss": 4.748070526123047,
2866
+ "step": 3990
2867
+ },
2868
+ {
2869
+ "epoch": 0.6757898293630681,
2870
+ "grad_norm": 0.5399011373519897,
2871
+ "learning_rate": 0.0002704624016931079,
2872
+ "loss": 4.742585754394531,
2873
+ "step": 4000
2874
+ },
2875
+ {
2876
+ "epoch": 0.6757898293630681,
2877
+ "eval_loss": 4.705080509185791,
2878
+ "eval_runtime": 3.6213,
2879
+ "eval_samples_per_second": 276.145,
2880
+ "eval_steps_per_second": 5.799,
2881
+ "step": 4000
2882
  }
2883
  ],
2884
  "logging_steps": 10,
 
2898
  "attributes": {}
2899
  }
2900
  },
2901
+ "total_flos": 1.33782728343552e+17,
2902
  "train_batch_size": 48,
2903
  "trial_name": null,
2904
  "trial_params": null