{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 25.668449197860962,
  "eval_steps": 400,
  "global_step": 1200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0213903743315508,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.9241,
      "step": 1
    },
    {
      "epoch": 0.21390374331550802,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 1.7796,
      "step": 10
    },
    {
      "epoch": 0.42780748663101603,
      "grad_norm": 43.9833984375,
      "learning_rate": 4e-08,
      "loss": 1.8211,
      "step": 20
    },
    {
      "epoch": 0.6417112299465241,
      "grad_norm": 40.75557327270508,
      "learning_rate": 4.4e-07,
      "loss": 1.733,
      "step": 30
    },
    {
      "epoch": 0.8556149732620321,
      "grad_norm": 33.69241714477539,
      "learning_rate": 8.400000000000001e-07,
      "loss": 1.4004,
      "step": 40
    },
    {
      "epoch": 1.0695187165775402,
      "grad_norm": 32.86851501464844,
      "learning_rate": 1.2400000000000002e-06,
      "loss": 1.3416,
      "step": 50
    },
    {
      "epoch": 1.2834224598930482,
      "grad_norm": 27.803329467773438,
      "learning_rate": 1.6400000000000002e-06,
      "loss": 1.081,
      "step": 60
    },
    {
      "epoch": 1.4973262032085561,
      "grad_norm": 31.654834747314453,
      "learning_rate": 2.04e-06,
      "loss": 1.05,
      "step": 70
    },
    {
      "epoch": 1.7112299465240641,
      "grad_norm": 29.529935836791992,
      "learning_rate": 2.4400000000000004e-06,
      "loss": 0.9944,
      "step": 80
    },
    {
      "epoch": 1.9251336898395723,
      "grad_norm": 29.650171279907227,
      "learning_rate": 2.84e-06,
      "loss": 0.8562,
      "step": 90
    },
    {
      "epoch": 2.1390374331550803,
      "grad_norm": 19.60784912109375,
      "learning_rate": 3.2400000000000003e-06,
      "loss": 0.5071,
      "step": 100
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 40.14349365234375,
      "learning_rate": 3.6400000000000003e-06,
      "loss": 0.2566,
      "step": 110
    },
    {
      "epoch": 2.5668449197860963,
      "grad_norm": 38.05447006225586,
      "learning_rate": 4.04e-06,
      "loss": 0.2548,
      "step": 120
    },
    {
      "epoch": 2.7807486631016043,
      "grad_norm": 48.719520568847656,
      "learning_rate": 4.440000000000001e-06,
      "loss": 0.2113,
      "step": 130
    },
    {
      "epoch": 2.9946524064171123,
      "grad_norm": 82.81842803955078,
      "learning_rate": 4.84e-06,
      "loss": 0.3946,
      "step": 140
    },
    {
      "epoch": 3.2085561497326203,
      "grad_norm": 20.48181915283203,
      "learning_rate": 5.240000000000001e-06,
      "loss": 0.09,
      "step": 150
    },
    {
      "epoch": 3.4224598930481283,
      "grad_norm": 5.467124938964844,
      "learning_rate": 5.64e-06,
      "loss": 0.07,
      "step": 160
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 27.81447410583496,
      "learning_rate": 6.040000000000001e-06,
      "loss": 0.0837,
      "step": 170
    },
    {
      "epoch": 3.8502673796791442,
      "grad_norm": 23.266380310058594,
      "learning_rate": 6.440000000000001e-06,
      "loss": 0.1036,
      "step": 180
    },
    {
      "epoch": 4.064171122994653,
      "grad_norm": 34.769287109375,
      "learning_rate": 6.8400000000000014e-06,
      "loss": 0.0438,
      "step": 190
    },
    {
      "epoch": 4.278074866310161,
      "grad_norm": 25.370189666748047,
      "learning_rate": 7.24e-06,
      "loss": 0.0541,
      "step": 200
    },
    {
      "epoch": 4.491978609625669,
      "grad_norm": 32.902244567871094,
      "learning_rate": 7.640000000000001e-06,
      "loss": 0.1416,
      "step": 210
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 47.98198699951172,
      "learning_rate": 8.040000000000001e-06,
      "loss": 0.0939,
      "step": 220
    },
    {
      "epoch": 4.919786096256685,
      "grad_norm": 16.503026962280273,
      "learning_rate": 8.44e-06,
      "loss": 0.1332,
      "step": 230
    },
    {
      "epoch": 5.133689839572193,
      "grad_norm": 6.1169633865356445,
      "learning_rate": 8.84e-06,
      "loss": 0.0508,
      "step": 240
    },
    {
      "epoch": 5.347593582887701,
      "grad_norm": 6.8135552406311035,
      "learning_rate": 9.240000000000001e-06,
      "loss": 0.0915,
      "step": 250
    },
    {
      "epoch": 5.561497326203209,
      "grad_norm": 42.288963317871094,
      "learning_rate": 9.640000000000001e-06,
      "loss": 0.1364,
      "step": 260
    },
    {
      "epoch": 5.775401069518717,
      "grad_norm": 68.40443420410156,
      "learning_rate": 1e-05,
      "loss": 0.183,
      "step": 270
    },
    {
      "epoch": 5.989304812834225,
      "grad_norm": 18.801176071166992,
      "learning_rate": 1e-05,
      "loss": 0.1234,
      "step": 280
    },
    {
      "epoch": 6.2032085561497325,
      "grad_norm": 17.309553146362305,
      "learning_rate": 1e-05,
      "loss": 0.0899,
      "step": 290
    },
    {
      "epoch": 6.4171122994652405,
      "grad_norm": 39.96405792236328,
      "learning_rate": 1e-05,
      "loss": 0.1682,
      "step": 300
    },
    {
      "epoch": 6.6310160427807485,
      "grad_norm": 37.08085250854492,
      "learning_rate": 1e-05,
      "loss": 0.1817,
      "step": 310
    },
    {
      "epoch": 6.8449197860962565,
      "grad_norm": 41.293277740478516,
      "learning_rate": 1e-05,
      "loss": 0.1853,
      "step": 320
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 19.117578506469727,
      "learning_rate": 1e-05,
      "loss": 0.084,
      "step": 330
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 3.196685552597046,
      "learning_rate": 1e-05,
      "loss": 0.0796,
      "step": 340
    },
    {
      "epoch": 7.4866310160427805,
      "grad_norm": 34.18360900878906,
      "learning_rate": 1e-05,
      "loss": 0.0638,
      "step": 350
    },
    {
      "epoch": 7.7005347593582885,
      "grad_norm": 34.174346923828125,
      "learning_rate": 1e-05,
      "loss": 0.0499,
      "step": 360
    },
    {
      "epoch": 7.9144385026737964,
      "grad_norm": 30.022125244140625,
      "learning_rate": 1e-05,
      "loss": 0.0511,
      "step": 370
    },
    {
      "epoch": 8.128342245989305,
      "grad_norm": 5.687129020690918,
      "learning_rate": 1e-05,
      "loss": 0.0381,
      "step": 380
    },
    {
      "epoch": 8.342245989304812,
      "grad_norm": 8.238314628601074,
      "learning_rate": 1e-05,
      "loss": 0.0699,
      "step": 390
    },
    {
      "epoch": 8.556149732620321,
      "grad_norm": 48.88713836669922,
      "learning_rate": 1e-05,
      "loss": 0.1932,
      "step": 400
    },
    {
      "epoch": 8.556149732620321,
      "eval_accuracy": 0.6,
      "eval_loss": 4.55859375,
      "eval_runtime": 0.8621,
      "eval_samples_per_second": 11.599,
      "eval_steps_per_second": 1.16,
      "step": 400
    },
    {
      "epoch": 8.770053475935828,
      "grad_norm": 1.5831031799316406,
      "learning_rate": 1e-05,
      "loss": 0.1437,
      "step": 410
    },
    {
      "epoch": 8.983957219251337,
      "grad_norm": 1.2749309539794922,
      "learning_rate": 1e-05,
      "loss": 0.0762,
      "step": 420
    },
    {
      "epoch": 9.197860962566844,
      "grad_norm": 56.0820426940918,
      "learning_rate": 1e-05,
      "loss": 0.056,
      "step": 430
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 21.727741241455078,
      "learning_rate": 1e-05,
      "loss": 0.0458,
      "step": 440
    },
    {
      "epoch": 9.62566844919786,
      "grad_norm": 3.6833608150482178,
      "learning_rate": 1e-05,
      "loss": 0.044,
      "step": 450
    },
    {
      "epoch": 9.83957219251337,
      "grad_norm": 43.264564514160156,
      "learning_rate": 1e-05,
      "loss": 0.0981,
      "step": 460
    },
    {
      "epoch": 10.053475935828876,
      "grad_norm": 0.6142730712890625,
      "learning_rate": 1e-05,
      "loss": 0.0384,
      "step": 470
    },
    {
      "epoch": 10.267379679144385,
      "grad_norm": 1.119858980178833,
      "learning_rate": 1e-05,
      "loss": 0.0681,
      "step": 480
    },
    {
      "epoch": 10.481283422459892,
      "grad_norm": 0.2561619281768799,
      "learning_rate": 1e-05,
      "loss": 0.0087,
      "step": 490
    },
    {
      "epoch": 10.695187165775401,
      "grad_norm": 37.55344009399414,
      "learning_rate": 1e-05,
      "loss": 0.0371,
      "step": 500
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 25.553625106811523,
      "learning_rate": 1e-05,
      "loss": 0.1983,
      "step": 510
    },
    {
      "epoch": 11.122994652406417,
      "grad_norm": 3.476731777191162,
      "learning_rate": 1e-05,
      "loss": 0.0475,
      "step": 520
    },
    {
      "epoch": 11.336898395721924,
      "grad_norm": 32.92385482788086,
      "learning_rate": 1e-05,
      "loss": 0.0108,
      "step": 530
    },
    {
      "epoch": 11.550802139037433,
      "grad_norm": 14.769713401794434,
      "learning_rate": 1e-05,
      "loss": 0.0174,
      "step": 540
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 1.3393691778182983,
      "learning_rate": 1e-05,
      "loss": 0.0503,
      "step": 550
    },
    {
      "epoch": 11.97860962566845,
      "grad_norm": 0.19340232014656067,
      "learning_rate": 1e-05,
      "loss": 0.0393,
      "step": 560
    },
    {
      "epoch": 12.192513368983958,
      "grad_norm": 2.4283690452575684,
      "learning_rate": 1e-05,
      "loss": 0.0659,
      "step": 570
    },
    {
      "epoch": 12.406417112299465,
      "grad_norm": 0.5373936295509338,
      "learning_rate": 1e-05,
      "loss": 0.0014,
      "step": 580
    },
    {
      "epoch": 12.620320855614974,
      "grad_norm": 1.1644189357757568,
      "learning_rate": 1e-05,
      "loss": 0.0651,
      "step": 590
    },
    {
      "epoch": 12.834224598930481,
      "grad_norm": 0.6392247676849365,
      "learning_rate": 1e-05,
      "loss": 0.013,
      "step": 600
    },
    {
      "epoch": 13.04812834224599,
      "grad_norm": 0.13331516087055206,
      "learning_rate": 1e-05,
      "loss": 0.0215,
      "step": 610
    },
    {
      "epoch": 13.262032085561497,
      "grad_norm": 0.08020053058862686,
      "learning_rate": 1e-05,
      "loss": 0.0267,
      "step": 620
    },
    {
      "epoch": 13.475935828877006,
      "grad_norm": 6.262119770050049,
      "learning_rate": 1e-05,
      "loss": 0.0137,
      "step": 630
    },
    {
      "epoch": 13.689839572192513,
      "grad_norm": 0.7786157727241516,
      "learning_rate": 1e-05,
      "loss": 0.037,
      "step": 640
    },
    {
      "epoch": 13.903743315508022,
      "grad_norm": 6.777099132537842,
      "learning_rate": 1e-05,
      "loss": 0.0015,
      "step": 650
    },
    {
      "epoch": 14.117647058823529,
      "grad_norm": 0.0033113746903836727,
      "learning_rate": 1e-05,
      "loss": 0.0255,
      "step": 660
    },
    {
      "epoch": 14.331550802139038,
      "grad_norm": 13.173226356506348,
      "learning_rate": 1e-05,
      "loss": 0.0056,
      "step": 670
    },
    {
      "epoch": 14.545454545454545,
      "grad_norm": 0.021669812500476837,
      "learning_rate": 1e-05,
      "loss": 0.0026,
      "step": 680
    },
    {
      "epoch": 14.759358288770054,
      "grad_norm": 0.03204642981290817,
      "learning_rate": 1e-05,
      "loss": 0.0046,
      "step": 690
    },
    {
      "epoch": 14.973262032085561,
      "grad_norm": 0.27756085991859436,
      "learning_rate": 1e-05,
      "loss": 0.0131,
      "step": 700
    },
    {
      "epoch": 15.18716577540107,
      "grad_norm": 2.8744680881500244,
      "learning_rate": 1e-05,
      "loss": 0.0051,
      "step": 710
    },
    {
      "epoch": 15.401069518716577,
      "grad_norm": 6.371706962585449,
      "learning_rate": 1e-05,
      "loss": 0.0471,
      "step": 720
    },
    {
      "epoch": 15.614973262032086,
      "grad_norm": 0.01101295743137598,
      "learning_rate": 1e-05,
      "loss": 0.0057,
      "step": 730
    },
    {
      "epoch": 15.828877005347593,
      "grad_norm": 0.021259065717458725,
      "learning_rate": 1e-05,
      "loss": 0.0822,
      "step": 740
    },
    {
      "epoch": 16.0427807486631,
      "grad_norm": 19.74135398864746,
      "learning_rate": 1e-05,
      "loss": 0.0141,
      "step": 750
    },
    {
      "epoch": 16.25668449197861,
      "grad_norm": 0.0036215828731656075,
      "learning_rate": 1e-05,
      "loss": 0.0158,
      "step": 760
    },
    {
      "epoch": 16.470588235294116,
      "grad_norm": 19.020896911621094,
      "learning_rate": 1e-05,
      "loss": 0.0107,
      "step": 770
    },
    {
      "epoch": 16.684491978609625,
      "grad_norm": 0.029464904218912125,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "step": 780
    },
    {
      "epoch": 16.898395721925134,
      "grad_norm": 0.0017767682438716292,
      "learning_rate": 1e-05,
      "loss": 0.0006,
      "step": 790
    },
    {
      "epoch": 17.112299465240643,
      "grad_norm": 4.318113327026367,
      "learning_rate": 1e-05,
      "loss": 0.0061,
      "step": 800
    },
    {
      "epoch": 17.112299465240643,
      "eval_accuracy": 0.6,
      "eval_loss": 5.87890625,
      "eval_runtime": 0.8622,
      "eval_samples_per_second": 11.598,
      "eval_steps_per_second": 1.16,
      "step": 800
    },
    {
      "epoch": 17.32620320855615,
      "grad_norm": 0.0028562787920236588,
      "learning_rate": 1e-05,
      "loss": 0.0066,
      "step": 810
    },
    {
      "epoch": 17.540106951871657,
      "grad_norm": 3.876565933227539,
      "learning_rate": 1e-05,
      "loss": 0.0172,
      "step": 820
    },
    {
      "epoch": 17.754010695187166,
      "grad_norm": 0.0012491046218201518,
      "learning_rate": 1e-05,
      "loss": 0.0033,
      "step": 830
    },
    {
      "epoch": 17.967914438502675,
      "grad_norm": 0.3624451756477356,
      "learning_rate": 1e-05,
      "loss": 0.0363,
      "step": 840
    },
    {
      "epoch": 18.181818181818183,
      "grad_norm": 0.9271169900894165,
      "learning_rate": 1e-05,
      "loss": 0.0219,
      "step": 850
    },
    {
      "epoch": 18.39572192513369,
      "grad_norm": 61.365291595458984,
      "learning_rate": 1e-05,
      "loss": 0.0444,
      "step": 860
    },
    {
      "epoch": 18.609625668449198,
      "grad_norm": 0.0029974980279803276,
      "learning_rate": 1e-05,
      "loss": 0.0118,
      "step": 870
    },
    {
      "epoch": 18.823529411764707,
      "grad_norm": 0.007906668819487095,
      "learning_rate": 1e-05,
      "loss": 0.0061,
      "step": 880
    },
    {
      "epoch": 19.037433155080215,
      "grad_norm": 0.0023081921972334385,
      "learning_rate": 1e-05,
      "loss": 0.0015,
      "step": 890
    },
    {
      "epoch": 19.25133689839572,
      "grad_norm": 0.005389372818171978,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "step": 900
    },
    {
      "epoch": 19.46524064171123,
      "grad_norm": 0.005591968540102243,
      "learning_rate": 1e-05,
      "loss": 0.0035,
      "step": 910
    },
    {
      "epoch": 19.67914438502674,
      "grad_norm": 0.014199817553162575,
      "learning_rate": 1e-05,
      "loss": 0.0014,
      "step": 920
    },
    {
      "epoch": 19.893048128342247,
      "grad_norm": 0.12109483778476715,
      "learning_rate": 1e-05,
      "loss": 0.0027,
      "step": 930
    },
    {
      "epoch": 20.106951871657753,
      "grad_norm": 0.12968279421329498,
      "learning_rate": 1e-05,
      "loss": 0.0089,
      "step": 940
    },
    {
      "epoch": 20.32085561497326,
      "grad_norm": 0.05300404876470566,
      "learning_rate": 1e-05,
      "loss": 0.0001,
      "step": 950
    },
    {
      "epoch": 20.53475935828877,
      "grad_norm": 0.6401832699775696,
      "learning_rate": 1e-05,
      "loss": 0.0357,
      "step": 960
    },
    {
      "epoch": 20.74866310160428,
      "grad_norm": 0.00800898764282465,
      "learning_rate": 1e-05,
      "loss": 0.0057,
      "step": 970
    },
    {
      "epoch": 20.962566844919785,
      "grad_norm": 0.02033822424709797,
      "learning_rate": 1e-05,
      "loss": 0.0317,
      "step": 980
    },
    {
      "epoch": 21.176470588235293,
      "grad_norm": 0.0029611990321427584,
      "learning_rate": 1e-05,
      "loss": 0.0019,
      "step": 990
    },
    {
      "epoch": 21.390374331550802,
      "grad_norm": 0.006123053841292858,
      "learning_rate": 1e-05,
      "loss": 0.0159,
      "step": 1000
    },
    {
      "epoch": 21.60427807486631,
      "grad_norm": 0.008170252665877342,
      "learning_rate": 1e-05,
      "loss": 0.0799,
      "step": 1010
    },
    {
      "epoch": 21.818181818181817,
      "grad_norm": 0.3321262001991272,
      "learning_rate": 1e-05,
      "loss": 0.0077,
      "step": 1020
    },
    {
      "epoch": 22.032085561497325,
      "grad_norm": 0.4511232376098633,
      "learning_rate": 1e-05,
      "loss": 0.0032,
      "step": 1030
    },
    {
      "epoch": 22.245989304812834,
      "grad_norm": 0.0032966439612209797,
      "learning_rate": 1e-05,
      "loss": 0.001,
      "step": 1040
    },
    {
      "epoch": 22.459893048128343,
      "grad_norm": 1.7022850513458252,
      "learning_rate": 1e-05,
      "loss": 0.0018,
      "step": 1050
    },
    {
      "epoch": 22.67379679144385,
      "grad_norm": 0.06403864175081253,
      "learning_rate": 1e-05,
      "loss": 0.0121,
      "step": 1060
    },
    {
      "epoch": 22.887700534759357,
      "grad_norm": 0.29701313376426697,
      "learning_rate": 1e-05,
      "loss": 0.1031,
      "step": 1070
    },
    {
      "epoch": 23.101604278074866,
      "grad_norm": 0.001655231462791562,
      "learning_rate": 1e-05,
      "loss": 0.0007,
      "step": 1080
    },
    {
      "epoch": 23.315508021390375,
      "grad_norm": 0.01081305555999279,
      "learning_rate": 1e-05,
      "loss": 0.0305,
      "step": 1090
    },
    {
      "epoch": 23.529411764705884,
      "grad_norm": 0.00689849816262722,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "step": 1100
    },
    {
      "epoch": 23.74331550802139,
      "grad_norm": 12.52665901184082,
      "learning_rate": 1e-05,
      "loss": 0.0294,
      "step": 1110
    },
    {
      "epoch": 23.9572192513369,
      "grad_norm": 0.004055026452988386,
      "learning_rate": 1e-05,
      "loss": 0.0383,
      "step": 1120
    },
    {
      "epoch": 24.171122994652407,
      "grad_norm": 0.3052468001842499,
      "learning_rate": 1e-05,
      "loss": 0.0829,
      "step": 1130
    },
    {
      "epoch": 24.385026737967916,
      "grad_norm": 83.22528076171875,
      "learning_rate": 1e-05,
      "loss": 0.0365,
      "step": 1140
    },
    {
      "epoch": 24.59893048128342,
      "grad_norm": 6.493893146514893,
      "learning_rate": 1e-05,
      "loss": 0.0087,
      "step": 1150
    },
    {
      "epoch": 24.81283422459893,
      "grad_norm": 0.8986703157424927,
      "learning_rate": 1e-05,
      "loss": 0.061,
      "step": 1160
    },
    {
      "epoch": 25.02673796791444,
      "grad_norm": 0.02195625938475132,
      "learning_rate": 1e-05,
      "loss": 0.004,
      "step": 1170
    },
    {
      "epoch": 25.240641711229948,
      "grad_norm": 0.13845649361610413,
      "learning_rate": 1e-05,
      "loss": 0.0078,
      "step": 1180
    },
    {
      "epoch": 25.454545454545453,
      "grad_norm": 0.05403750389814377,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "step": 1190
    },
    {
      "epoch": 25.668449197860962,
      "grad_norm": 0.06856454908847809,
      "learning_rate": 1e-05,
      "loss": 0.008,
      "step": 1200
    },
    {
      "epoch": 25.668449197860962,
      "eval_accuracy": 0.5,
      "eval_loss": 8.5,
      "eval_runtime": 0.8643,
      "eval_samples_per_second": 11.571,
      "eval_steps_per_second": 1.157,
      "step": 1200
    }
  ],
  "logging_steps": 10,
  "max_steps": 2500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 55,
  "save_steps": 400,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.907150614607954e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}