{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 51.336898395721924,
  "eval_steps": 400,
  "global_step": 2400,
  "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": 1.8558,
      "step": 1
    },
    {
      "epoch": 0.21390374331550802,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.0957,
      "step": 10
    },
    {
      "epoch": 0.42780748663101603,
      "grad_norm": 63.96761703491211,
      "learning_rate": 8e-08,
      "loss": 2.1274,
      "step": 20
    },
    {
      "epoch": 0.6417112299465241,
      "grad_norm": 56.0494270324707,
      "learning_rate": 4.4e-07,
      "loss": 2.0513,
      "step": 30
    },
    {
      "epoch": 0.8556149732620321,
      "grad_norm": 31.82117462158203,
      "learning_rate": 8.400000000000001e-07,
      "loss": 1.6275,
      "step": 40
    },
    {
      "epoch": 1.0695187165775402,
      "grad_norm": 27.546764373779297,
      "learning_rate": 1.2400000000000002e-06,
      "loss": 1.3922,
      "step": 50
    },
    {
      "epoch": 1.2834224598930482,
      "grad_norm": 27.145185470581055,
      "learning_rate": 1.6400000000000002e-06,
      "loss": 1.0424,
      "step": 60
    },
    {
      "epoch": 1.4973262032085561,
      "grad_norm": 33.02227020263672,
      "learning_rate": 2.04e-06,
      "loss": 1.0434,
      "step": 70
    },
    {
      "epoch": 1.7112299465240641,
      "grad_norm": 23.329740524291992,
      "learning_rate": 2.4400000000000004e-06,
      "loss": 1.0585,
      "step": 80
    },
    {
      "epoch": 1.9251336898395723,
      "grad_norm": 26.408071517944336,
      "learning_rate": 2.84e-06,
      "loss": 0.8849,
      "step": 90
    },
    {
      "epoch": 2.1390374331550803,
      "grad_norm": 17.790225982666016,
      "learning_rate": 3.2400000000000003e-06,
      "loss": 0.5427,
      "step": 100
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 29.846942901611328,
      "learning_rate": 3.6400000000000003e-06,
      "loss": 0.2784,
      "step": 110
    },
    {
      "epoch": 2.5668449197860963,
      "grad_norm": 20.77630043029785,
      "learning_rate": 4.04e-06,
      "loss": 0.2569,
      "step": 120
    },
    {
      "epoch": 2.7807486631016043,
      "grad_norm": 33.25271987915039,
      "learning_rate": 4.440000000000001e-06,
      "loss": 0.1836,
      "step": 130
    },
    {
      "epoch": 2.9946524064171123,
      "grad_norm": 66.23503875732422,
      "learning_rate": 4.84e-06,
      "loss": 0.3042,
      "step": 140
    },
    {
      "epoch": 3.2085561497326203,
      "grad_norm": 4.618898868560791,
      "learning_rate": 5.240000000000001e-06,
      "loss": 0.0455,
      "step": 150
    },
    {
      "epoch": 3.4224598930481283,
      "grad_norm": 5.856906890869141,
      "learning_rate": 5.64e-06,
      "loss": 0.0558,
      "step": 160
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 25.529407501220703,
      "learning_rate": 6.040000000000001e-06,
      "loss": 0.0835,
      "step": 170
    },
    {
      "epoch": 3.8502673796791442,
      "grad_norm": 16.971477508544922,
      "learning_rate": 6.440000000000001e-06,
      "loss": 0.1126,
      "step": 180
    },
    {
      "epoch": 4.064171122994653,
      "grad_norm": 9.28547191619873,
      "learning_rate": 6.8400000000000014e-06,
      "loss": 0.0838,
      "step": 190
    },
    {
      "epoch": 4.278074866310161,
      "grad_norm": 17.296798706054688,
      "learning_rate": 7.24e-06,
      "loss": 0.0518,
      "step": 200
    },
    {
      "epoch": 4.491978609625669,
      "grad_norm": 12.538177490234375,
      "learning_rate": 7.640000000000001e-06,
      "loss": 0.0532,
      "step": 210
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 24.446659088134766,
      "learning_rate": 8.040000000000001e-06,
      "loss": 0.1583,
      "step": 220
    },
    {
      "epoch": 4.919786096256685,
      "grad_norm": 25.139781951904297,
      "learning_rate": 8.44e-06,
      "loss": 0.0894,
      "step": 230
    },
    {
      "epoch": 5.133689839572193,
      "grad_norm": 2.361607074737549,
      "learning_rate": 8.84e-06,
      "loss": 0.0531,
      "step": 240
    },
    {
      "epoch": 5.347593582887701,
      "grad_norm": 4.768627643585205,
      "learning_rate": 9.240000000000001e-06,
      "loss": 0.0664,
      "step": 250
    },
    {
      "epoch": 5.561497326203209,
      "grad_norm": 37.02785873413086,
      "learning_rate": 9.640000000000001e-06,
      "loss": 0.1502,
      "step": 260
    },
    {
      "epoch": 5.775401069518717,
      "grad_norm": 48.693763732910156,
      "learning_rate": 1e-05,
      "loss": 0.1953,
      "step": 270
    },
    {
      "epoch": 5.989304812834225,
      "grad_norm": 37.84030532836914,
      "learning_rate": 1e-05,
      "loss": 0.1806,
      "step": 280
    },
    {
      "epoch": 6.2032085561497325,
      "grad_norm": 39.74201965332031,
      "learning_rate": 1e-05,
      "loss": 0.0866,
      "step": 290
    },
    {
      "epoch": 6.4171122994652405,
      "grad_norm": 7.8511834144592285,
      "learning_rate": 1e-05,
      "loss": 0.0994,
      "step": 300
    },
    {
      "epoch": 6.6310160427807485,
      "grad_norm": 54.63271713256836,
      "learning_rate": 1e-05,
      "loss": 0.1712,
      "step": 310
    },
    {
      "epoch": 6.8449197860962565,
      "grad_norm": 44.596153259277344,
      "learning_rate": 1e-05,
      "loss": 0.1442,
      "step": 320
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 5.094964504241943,
      "learning_rate": 1e-05,
      "loss": 0.0526,
      "step": 330
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 12.109665870666504,
      "learning_rate": 1e-05,
      "loss": 0.03,
      "step": 340
    },
    {
      "epoch": 7.4866310160427805,
      "grad_norm": 34.3682975769043,
      "learning_rate": 1e-05,
      "loss": 0.0759,
      "step": 350
    },
    {
      "epoch": 7.7005347593582885,
      "grad_norm": 2.3397929668426514,
      "learning_rate": 1e-05,
      "loss": 0.0685,
      "step": 360
    },
    {
      "epoch": 7.9144385026737964,
      "grad_norm": 21.006542205810547,
      "learning_rate": 1e-05,
      "loss": 0.0637,
      "step": 370
    },
    {
      "epoch": 8.128342245989305,
      "grad_norm": 2.5418567657470703,
      "learning_rate": 1e-05,
      "loss": 0.057,
      "step": 380
    },
    {
      "epoch": 8.342245989304812,
      "grad_norm": 18.59604263305664,
      "learning_rate": 1e-05,
      "loss": 0.1096,
      "step": 390
    },
    {
      "epoch": 8.556149732620321,
      "grad_norm": 52.85743713378906,
      "learning_rate": 1e-05,
      "loss": 0.1422,
      "step": 400
    },
    {
      "epoch": 8.556149732620321,
      "eval_accuracy": 0.4,
      "eval_loss": 4.6875,
      "eval_runtime": 0.8724,
      "eval_samples_per_second": 11.462,
      "eval_steps_per_second": 1.146,
      "step": 400
    },
    {
      "epoch": 8.770053475935828,
      "grad_norm": 36.897674560546875,
      "learning_rate": 1e-05,
      "loss": 0.1633,
      "step": 410
    },
    {
      "epoch": 8.983957219251337,
      "grad_norm": 16.528366088867188,
      "learning_rate": 1e-05,
      "loss": 0.0336,
      "step": 420
    },
    {
      "epoch": 9.197860962566844,
      "grad_norm": 87.09658813476562,
      "learning_rate": 1e-05,
      "loss": 0.1071,
      "step": 430
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 4.741786479949951,
      "learning_rate": 1e-05,
      "loss": 0.0701,
      "step": 440
    },
    {
      "epoch": 9.62566844919786,
      "grad_norm": 12.09588623046875,
      "learning_rate": 1e-05,
      "loss": 0.0746,
      "step": 450
    },
    {
      "epoch": 9.83957219251337,
      "grad_norm": 32.604068756103516,
      "learning_rate": 1e-05,
      "loss": 0.0872,
      "step": 460
    },
    {
      "epoch": 10.053475935828876,
      "grad_norm": 4.408775806427002,
      "learning_rate": 1e-05,
      "loss": 0.1084,
      "step": 470
    },
    {
      "epoch": 10.267379679144385,
      "grad_norm": 22.274890899658203,
      "learning_rate": 1e-05,
      "loss": 0.0297,
      "step": 480
    },
    {
      "epoch": 10.481283422459892,
      "grad_norm": 58.223323822021484,
      "learning_rate": 1e-05,
      "loss": 0.0637,
      "step": 490
    },
    {
      "epoch": 10.695187165775401,
      "grad_norm": 2.7027640342712402,
      "learning_rate": 1e-05,
      "loss": 0.1206,
      "step": 500
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 43.47382736206055,
      "learning_rate": 1e-05,
      "loss": 0.045,
      "step": 510
    },
    {
      "epoch": 11.122994652406417,
      "grad_norm": 1.6665784120559692,
      "learning_rate": 1e-05,
      "loss": 0.045,
      "step": 520
    },
    {
      "epoch": 11.336898395721924,
      "grad_norm": 0.5738474130630493,
      "learning_rate": 1e-05,
      "loss": 0.008,
      "step": 530
    },
    {
      "epoch": 11.550802139037433,
      "grad_norm": 16.410573959350586,
      "learning_rate": 1e-05,
      "loss": 0.0995,
      "step": 540
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.7015154361724854,
      "learning_rate": 1e-05,
      "loss": 0.0313,
      "step": 550
    },
    {
      "epoch": 11.97860962566845,
      "grad_norm": 11.851508140563965,
      "learning_rate": 1e-05,
      "loss": 0.0724,
      "step": 560
    },
    {
      "epoch": 12.192513368983958,
      "grad_norm": 16.754575729370117,
      "learning_rate": 1e-05,
      "loss": 0.0272,
      "step": 570
    },
    {
      "epoch": 12.406417112299465,
      "grad_norm": 15.92441177368164,
      "learning_rate": 1e-05,
      "loss": 0.0241,
      "step": 580
    },
    {
      "epoch": 12.620320855614974,
      "grad_norm": 0.05043609067797661,
      "learning_rate": 1e-05,
      "loss": 0.042,
      "step": 590
    },
    {
      "epoch": 12.834224598930481,
      "grad_norm": 0.3494648337364197,
      "learning_rate": 1e-05,
      "loss": 0.0083,
      "step": 600
    },
    {
      "epoch": 13.04812834224599,
      "grad_norm": 23.260435104370117,
      "learning_rate": 1e-05,
      "loss": 0.026,
      "step": 610
    },
    {
      "epoch": 13.262032085561497,
      "grad_norm": 5.789357662200928,
      "learning_rate": 1e-05,
      "loss": 0.0224,
      "step": 620
    },
    {
      "epoch": 13.475935828877006,
      "grad_norm": 0.056954510509967804,
      "learning_rate": 1e-05,
      "loss": 0.0642,
      "step": 630
    },
    {
      "epoch": 13.689839572192513,
      "grad_norm": 72.28430938720703,
      "learning_rate": 1e-05,
      "loss": 0.0366,
      "step": 640
    },
    {
      "epoch": 13.903743315508022,
      "grad_norm": 0.5189899802207947,
      "learning_rate": 1e-05,
      "loss": 0.033,
      "step": 650
    },
    {
      "epoch": 14.117647058823529,
      "grad_norm": 0.07448354363441467,
      "learning_rate": 1e-05,
      "loss": 0.0088,
      "step": 660
    },
    {
      "epoch": 14.331550802139038,
      "grad_norm": 0.05018957331776619,
      "learning_rate": 1e-05,
      "loss": 0.0082,
      "step": 670
    },
    {
      "epoch": 14.545454545454545,
      "grad_norm": 33.61125183105469,
      "learning_rate": 1e-05,
      "loss": 0.0753,
      "step": 680
    },
    {
      "epoch": 14.759358288770054,
      "grad_norm": 0.023922119289636612,
      "learning_rate": 1e-05,
      "loss": 0.0267,
      "step": 690
    },
    {
      "epoch": 14.973262032085561,
      "grad_norm": 0.04713653773069382,
      "learning_rate": 1e-05,
      "loss": 0.096,
      "step": 700
    },
    {
      "epoch": 15.18716577540107,
      "grad_norm": 0.0457034632563591,
      "learning_rate": 1e-05,
      "loss": 0.0083,
      "step": 710
    },
    {
      "epoch": 15.401069518716577,
      "grad_norm": 0.011708029545843601,
      "learning_rate": 1e-05,
      "loss": 0.072,
      "step": 720
    },
    {
      "epoch": 15.614973262032086,
      "grad_norm": 0.3412806987762451,
      "learning_rate": 1e-05,
      "loss": 0.0032,
      "step": 730
    },
    {
      "epoch": 15.828877005347593,
      "grad_norm": 1.667181372642517,
      "learning_rate": 1e-05,
      "loss": 0.0512,
      "step": 740
    },
    {
      "epoch": 16.0427807486631,
      "grad_norm": 0.004597651772201061,
      "learning_rate": 1e-05,
      "loss": 0.0555,
      "step": 750
    },
    {
      "epoch": 16.25668449197861,
      "grad_norm": 0.01679002121090889,
      "learning_rate": 1e-05,
      "loss": 0.0294,
      "step": 760
    },
    {
      "epoch": 16.470588235294116,
      "grad_norm": 24.51487922668457,
      "learning_rate": 1e-05,
      "loss": 0.0979,
      "step": 770
    },
    {
      "epoch": 16.684491978609625,
      "grad_norm": 0.010406852699816227,
      "learning_rate": 1e-05,
      "loss": 0.0361,
      "step": 780
    },
    {
      "epoch": 16.898395721925134,
      "grad_norm": 4.567187786102295,
      "learning_rate": 1e-05,
      "loss": 0.1229,
      "step": 790
    },
    {
      "epoch": 17.112299465240643,
      "grad_norm": 1.7842605113983154,
      "learning_rate": 1e-05,
      "loss": 0.0121,
      "step": 800
    },
    {
      "epoch": 17.112299465240643,
      "eval_accuracy": 0.5,
      "eval_loss": 6.6640625,
      "eval_runtime": 0.8689,
      "eval_samples_per_second": 11.509,
      "eval_steps_per_second": 1.151,
      "step": 800
    },
    {
      "epoch": 17.32620320855615,
      "grad_norm": 0.061683282256126404,
      "learning_rate": 1e-05,
      "loss": 0.0079,
      "step": 810
    },
    {
      "epoch": 17.540106951871657,
      "grad_norm": 0.11749367415904999,
      "learning_rate": 1e-05,
      "loss": 0.0857,
      "step": 820
    },
    {
      "epoch": 17.754010695187166,
      "grad_norm": 0.02787230722606182,
      "learning_rate": 1e-05,
      "loss": 0.0051,
      "step": 830
    },
    {
      "epoch": 17.967914438502675,
      "grad_norm": 0.015304004773497581,
      "learning_rate": 1e-05,
      "loss": 0.0882,
      "step": 840
    },
    {
      "epoch": 18.181818181818183,
      "grad_norm": 0.41026026010513306,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "step": 850
    },
    {
      "epoch": 18.39572192513369,
      "grad_norm": 0.012662127614021301,
      "learning_rate": 1e-05,
      "loss": 0.0013,
      "step": 860
    },
    {
      "epoch": 18.609625668449198,
      "grad_norm": 0.10844732820987701,
      "learning_rate": 1e-05,
      "loss": 0.0062,
      "step": 870
    },
    {
      "epoch": 18.823529411764707,
      "grad_norm": 2.6394922733306885,
      "learning_rate": 1e-05,
      "loss": 0.0335,
      "step": 880
    },
    {
      "epoch": 19.037433155080215,
      "grad_norm": 0.28504815697669983,
      "learning_rate": 1e-05,
      "loss": 0.0092,
      "step": 890
    },
    {
      "epoch": 19.25133689839572,
      "grad_norm": 0.03371066227555275,
      "learning_rate": 1e-05,
      "loss": 0.0001,
      "step": 900
    },
    {
      "epoch": 19.46524064171123,
      "grad_norm": 0.01855628192424774,
      "learning_rate": 1e-05,
      "loss": 0.0046,
      "step": 910
    },
    {
      "epoch": 19.67914438502674,
      "grad_norm": 0.008938683196902275,
      "learning_rate": 1e-05,
      "loss": 0.0014,
      "step": 920
    },
    {
      "epoch": 19.893048128342247,
      "grad_norm": 0.0025397331919521093,
      "learning_rate": 1e-05,
      "loss": 0.0023,
      "step": 930
    },
    {
      "epoch": 20.106951871657753,
      "grad_norm": 0.0024771266616880894,
      "learning_rate": 1e-05,
      "loss": 0.001,
      "step": 940
    },
    {
      "epoch": 20.32085561497326,
      "grad_norm": 56.46254348754883,
      "learning_rate": 1e-05,
      "loss": 0.0533,
      "step": 950
    },
    {
      "epoch": 20.53475935828877,
      "grad_norm": 0.046234406530857086,
      "learning_rate": 1e-05,
      "loss": 0.0113,
      "step": 960
    },
    {
      "epoch": 20.74866310160428,
      "grad_norm": 0.03888735547661781,
      "learning_rate": 1e-05,
      "loss": 0.0001,
      "step": 970
    },
    {
      "epoch": 20.962566844919785,
      "grad_norm": 4.142932415008545,
      "learning_rate": 1e-05,
      "loss": 0.0078,
      "step": 980
    },
    {
      "epoch": 21.176470588235293,
      "grad_norm": 0.01926487125456333,
      "learning_rate": 1e-05,
      "loss": 0.0015,
      "step": 990
    },
    {
      "epoch": 21.390374331550802,
      "grad_norm": 0.03864801302552223,
      "learning_rate": 1e-05,
      "loss": 0.0032,
      "step": 1000
    },
    {
      "epoch": 21.60427807486631,
      "grad_norm": 0.40588998794555664,
      "learning_rate": 1e-05,
      "loss": 0.0003,
      "step": 1010
    },
    {
      "epoch": 21.818181818181817,
      "grad_norm": 0.2934487760066986,
      "learning_rate": 1e-05,
      "loss": 0.0081,
      "step": 1020
    },
    {
      "epoch": 22.032085561497325,
      "grad_norm": 0.004641254432499409,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1030
    },
    {
      "epoch": 22.245989304812834,
      "grad_norm": 14.084956169128418,
      "learning_rate": 1e-05,
      "loss": 0.0018,
      "step": 1040
    },
    {
      "epoch": 22.459893048128343,
      "grad_norm": 1.7572118043899536,
      "learning_rate": 1e-05,
      "loss": 0.001,
      "step": 1050
    },
    {
      "epoch": 22.67379679144385,
      "grad_norm": 2.7290031909942627,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "step": 1060
    },
    {
      "epoch": 22.887700534759357,
      "grad_norm": 0.005051525309681892,
      "learning_rate": 1e-05,
      "loss": 0.0007,
      "step": 1070
    },
    {
      "epoch": 23.101604278074866,
      "grad_norm": 0.002047200920060277,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1080
    },
    {
      "epoch": 23.315508021390375,
      "grad_norm": 0.0035107170697301626,
      "learning_rate": 1e-05,
      "loss": 0.0494,
      "step": 1090
    },
    {
      "epoch": 23.529411764705884,
      "grad_norm": 0.004472868517041206,
      "learning_rate": 1e-05,
      "loss": 0.0034,
      "step": 1100
    },
    {
      "epoch": 23.74331550802139,
      "grad_norm": 0.08594793826341629,
      "learning_rate": 1e-05,
      "loss": 0.0063,
      "step": 1110
    },
    {
      "epoch": 23.9572192513369,
      "grad_norm": 0.752574622631073,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "step": 1120
    },
    {
      "epoch": 24.171122994652407,
      "grad_norm": 0.021934740245342255,
      "learning_rate": 1e-05,
      "loss": 0.0019,
      "step": 1130
    },
    {
      "epoch": 24.385026737967916,
      "grad_norm": 0.007998108863830566,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1140
    },
    {
      "epoch": 24.59893048128342,
      "grad_norm": 0.010390130802989006,
      "learning_rate": 1e-05,
      "loss": 0.0072,
      "step": 1150
    },
    {
      "epoch": 24.81283422459893,
      "grad_norm": 0.002071298658847809,
      "learning_rate": 1e-05,
      "loss": 0.0013,
      "step": 1160
    },
    {
      "epoch": 25.02673796791444,
      "grad_norm": 0.0045625255443155766,
      "learning_rate": 1e-05,
      "loss": 0.002,
      "step": 1170
    },
    {
      "epoch": 25.240641711229948,
      "grad_norm": 0.017212701961398125,
      "learning_rate": 1e-05,
      "loss": 0.0048,
      "step": 1180
    },
    {
      "epoch": 25.454545454545453,
      "grad_norm": 12.734994888305664,
      "learning_rate": 1e-05,
      "loss": 0.0018,
      "step": 1190
    },
    {
      "epoch": 25.668449197860962,
      "grad_norm": 0.0035849984269589186,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1200
    },
    {
      "epoch": 25.668449197860962,
      "eval_accuracy": 0.8,
      "eval_loss": 4.71484375,
      "eval_runtime": 0.8712,
      "eval_samples_per_second": 11.478,
      "eval_steps_per_second": 1.148,
      "step": 1200
    },
    {
      "epoch": 25.88235294117647,
      "grad_norm": 0.005425190087407827,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1210
    },
    {
      "epoch": 26.09625668449198,
      "grad_norm": 0.007795536890625954,
      "learning_rate": 1e-05,
      "loss": 0.0034,
      "step": 1220
    },
    {
      "epoch": 26.310160427807485,
      "grad_norm": 0.00858351681381464,
      "learning_rate": 1e-05,
      "loss": 0.0032,
      "step": 1230
    },
    {
      "epoch": 26.524064171122994,
      "grad_norm": 0.0035222836304455996,
      "learning_rate": 1e-05,
      "loss": 0.0103,
      "step": 1240
    },
    {
      "epoch": 26.737967914438503,
      "grad_norm": 0.004781092517077923,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "step": 1250
    },
    {
      "epoch": 26.951871657754012,
      "grad_norm": 0.0031953530851751566,
      "learning_rate": 1e-05,
      "loss": 0.0013,
      "step": 1260
    },
    {
      "epoch": 27.165775401069517,
      "grad_norm": 0.00516524026170373,
      "learning_rate": 1e-05,
      "loss": 0.0007,
      "step": 1270
    },
    {
      "epoch": 27.379679144385026,
      "grad_norm": 0.004281027242541313,
      "learning_rate": 1e-05,
      "loss": 0.0106,
      "step": 1280
    },
    {
      "epoch": 27.593582887700535,
      "grad_norm": 0.0015023744199424982,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1290
    },
    {
      "epoch": 27.807486631016044,
      "grad_norm": 0.005484332330524921,
      "learning_rate": 1e-05,
      "loss": 0.0023,
      "step": 1300
    },
    {
      "epoch": 28.02139037433155,
      "grad_norm": 0.0035302124451845884,
      "learning_rate": 1e-05,
      "loss": 0.0001,
      "step": 1310
    },
    {
      "epoch": 28.235294117647058,
      "grad_norm": 0.0033441612031310797,
      "learning_rate": 1e-05,
      "loss": 0.0025,
      "step": 1320
    },
    {
      "epoch": 28.449197860962567,
      "grad_norm": 0.0035446221008896828,
      "learning_rate": 1e-05,
      "loss": 0.0014,
      "step": 1330
    },
    {
      "epoch": 28.663101604278076,
      "grad_norm": 0.010898416861891747,
      "learning_rate": 1e-05,
      "loss": 0.002,
      "step": 1340
    },
    {
      "epoch": 28.87700534759358,
      "grad_norm": 0.008727981708943844,
      "learning_rate": 1e-05,
      "loss": 0.1828,
      "step": 1350
    },
    {
      "epoch": 29.09090909090909,
      "grad_norm": 0.004256559070199728,
      "learning_rate": 1e-05,
      "loss": 0.0431,
      "step": 1360
    },
    {
      "epoch": 29.3048128342246,
      "grad_norm": 0.21178306639194489,
      "learning_rate": 1e-05,
      "loss": 0.0099,
      "step": 1370
    },
    {
      "epoch": 29.518716577540108,
      "grad_norm": 0.2352713793516159,
      "learning_rate": 1e-05,
      "loss": 0.0835,
      "step": 1380
    },
    {
      "epoch": 29.732620320855617,
      "grad_norm": 0.03791838884353638,
      "learning_rate": 1e-05,
      "loss": 0.0522,
      "step": 1390
    },
    {
      "epoch": 29.946524064171122,
      "grad_norm": 0.04343949630856514,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "step": 1400
    },
    {
      "epoch": 30.16042780748663,
      "grad_norm": 2.7350242137908936,
      "learning_rate": 1e-05,
      "loss": 0.003,
      "step": 1410
    },
    {
      "epoch": 30.37433155080214,
      "grad_norm": 0.030035167932510376,
      "learning_rate": 1e-05,
      "loss": 0.006,
      "step": 1420
    },
    {
      "epoch": 30.58823529411765,
      "grad_norm": 3.1150763034820557,
      "learning_rate": 1e-05,
      "loss": 0.0016,
      "step": 1430
    },
    {
      "epoch": 30.802139037433154,
      "grad_norm": 0.028651399537920952,
      "learning_rate": 1e-05,
      "loss": 0.0016,
      "step": 1440
    },
    {
      "epoch": 31.016042780748663,
      "grad_norm": 37.155609130859375,
      "learning_rate": 1e-05,
      "loss": 0.0128,
      "step": 1450
    },
    {
      "epoch": 31.22994652406417,
      "grad_norm": 0.002968219807371497,
      "learning_rate": 1e-05,
      "loss": 0.0141,
      "step": 1460
    },
    {
      "epoch": 31.44385026737968,
      "grad_norm": 5.24705696105957,
      "learning_rate": 1e-05,
      "loss": 0.0121,
      "step": 1470
    },
    {
      "epoch": 31.657754010695186,
      "grad_norm": 0.0024348751176148653,
      "learning_rate": 1e-05,
      "loss": 0.1083,
      "step": 1480
    },
    {
      "epoch": 31.871657754010695,
      "grad_norm": 0.5254661440849304,
      "learning_rate": 1e-05,
      "loss": 0.1179,
      "step": 1490
    },
    {
      "epoch": 32.0855614973262,
      "grad_norm": 72.65499114990234,
      "learning_rate": 1e-05,
      "loss": 0.1361,
      "step": 1500
    },
    {
      "epoch": 32.29946524064171,
      "grad_norm": 14.370610237121582,
      "learning_rate": 1e-05,
      "loss": 0.0346,
      "step": 1510
    },
    {
      "epoch": 32.51336898395722,
      "grad_norm": 20.988916397094727,
      "learning_rate": 1e-05,
      "loss": 0.0025,
      "step": 1520
    },
    {
      "epoch": 32.72727272727273,
      "grad_norm": 0.002341994782909751,
      "learning_rate": 1e-05,
      "loss": 0.0004,
      "step": 1530
    },
    {
      "epoch": 32.94117647058823,
      "grad_norm": 2.404449701309204,
      "learning_rate": 1e-05,
      "loss": 0.013,
      "step": 1540
    },
    {
      "epoch": 33.155080213903744,
      "grad_norm": 0.005042242351919413,
      "learning_rate": 1e-05,
      "loss": 0.0011,
      "step": 1550
    },
    {
      "epoch": 33.36898395721925,
      "grad_norm": 0.00044189911568537354,
      "learning_rate": 1e-05,
      "loss": 0.0139,
      "step": 1560
    },
    {
      "epoch": 33.58288770053476,
      "grad_norm": 0.0014409746509045362,
      "learning_rate": 1e-05,
      "loss": 0.0284,
      "step": 1570
    },
    {
      "epoch": 33.79679144385027,
      "grad_norm": 2.608125686645508,
      "learning_rate": 1e-05,
      "loss": 0.015,
      "step": 1580
    },
    {
      "epoch": 34.01069518716577,
      "grad_norm": 0.0005788679700344801,
      "learning_rate": 1e-05,
      "loss": 0.0201,
      "step": 1590
    },
    {
      "epoch": 34.224598930481285,
      "grad_norm": 75.40068817138672,
      "learning_rate": 1e-05,
      "loss": 0.028,
      "step": 1600
    },
    {
      "epoch": 34.224598930481285,
      "eval_accuracy": 0.7,
      "eval_loss": 6.69921875,
      "eval_runtime": 0.8667,
      "eval_samples_per_second": 11.537,
      "eval_steps_per_second": 1.154,
      "step": 1600
    },
    {
      "epoch": 34.43850267379679,
      "grad_norm": 0.005270595662295818,
      "learning_rate": 1e-05,
      "loss": 0.0087,
      "step": 1610
    },
    {
      "epoch": 34.6524064171123,
      "grad_norm": 5.715099334716797,
      "learning_rate": 1e-05,
      "loss": 0.0131,
      "step": 1620
    },
    {
      "epoch": 34.86631016042781,
      "grad_norm": 7.6973066329956055,
      "learning_rate": 1e-05,
      "loss": 0.0071,
      "step": 1630
    },
    {
      "epoch": 35.080213903743314,
      "grad_norm": 0.04163156822323799,
      "learning_rate": 1e-05,
      "loss": 0.0432,
      "step": 1640
    },
    {
      "epoch": 35.294117647058826,
      "grad_norm": 0.8603066205978394,
      "learning_rate": 1e-05,
      "loss": 0.081,
      "step": 1650
    },
    {
      "epoch": 35.50802139037433,
      "grad_norm": 0.02197711355984211,
      "learning_rate": 1e-05,
      "loss": 0.0087,
      "step": 1660
    },
    {
      "epoch": 35.72192513368984,
      "grad_norm": 0.2183353304862976,
      "learning_rate": 1e-05,
      "loss": 0.0463,
      "step": 1670
    },
    {
      "epoch": 35.93582887700535,
      "grad_norm": 1.1545729637145996,
      "learning_rate": 1e-05,
      "loss": 0.0073,
      "step": 1680
    },
    {
      "epoch": 36.149732620320854,
      "grad_norm": 0.5573898553848267,
      "learning_rate": 1e-05,
      "loss": 0.0011,
      "step": 1690
    },
    {
      "epoch": 36.36363636363637,
      "grad_norm": 0.0014353781007230282,
      "learning_rate": 1e-05,
      "loss": 0.077,
      "step": 1700
    },
    {
      "epoch": 36.57754010695187,
      "grad_norm": 0.0148012051358819,
      "learning_rate": 1e-05,
      "loss": 0.0052,
      "step": 1710
    },
    {
      "epoch": 36.79144385026738,
      "grad_norm": 0.03192583844065666,
      "learning_rate": 1e-05,
      "loss": 0.086,
      "step": 1720
    },
    {
      "epoch": 37.00534759358289,
      "grad_norm": 82.85762023925781,
      "learning_rate": 1e-05,
      "loss": 0.0983,
      "step": 1730
    },
    {
      "epoch": 37.219251336898395,
      "grad_norm": 0.08266828954219818,
      "learning_rate": 1e-05,
      "loss": 0.0279,
      "step": 1740
    },
    {
      "epoch": 37.4331550802139,
      "grad_norm": 0.008177646435797215,
      "learning_rate": 1e-05,
      "loss": 0.0143,
      "step": 1750
    },
    {
      "epoch": 37.64705882352941,
      "grad_norm": 0.009183383546769619,
      "learning_rate": 1e-05,
      "loss": 0.0107,
      "step": 1760
    },
    {
      "epoch": 37.86096256684492,
      "grad_norm": 1.8703384399414062,
      "learning_rate": 1e-05,
      "loss": 0.0479,
      "step": 1770
    },
    {
      "epoch": 38.07486631016043,
      "grad_norm": 0.004155300557613373,
      "learning_rate": 1e-05,
      "loss": 0.0595,
      "step": 1780
    },
    {
      "epoch": 38.288770053475936,
      "grad_norm": 2.2654850482940674,
      "learning_rate": 1e-05,
      "loss": 0.033,
      "step": 1790
    },
    {
      "epoch": 38.50267379679144,
      "grad_norm": 0.20017670094966888,
      "learning_rate": 1e-05,
      "loss": 0.0078,
      "step": 1800
    },
    {
      "epoch": 38.716577540106954,
      "grad_norm": 53.791812896728516,
      "learning_rate": 1e-05,
      "loss": 0.058,
      "step": 1810
    },
    {
      "epoch": 38.93048128342246,
      "grad_norm": 47.128692626953125,
      "learning_rate": 1e-05,
      "loss": 0.045,
      "step": 1820
    },
    {
      "epoch": 39.144385026737964,
      "grad_norm": 41.79620361328125,
      "learning_rate": 1e-05,
      "loss": 0.156,
      "step": 1830
    },
    {
      "epoch": 39.35828877005348,
      "grad_norm": 4.738333702087402,
      "learning_rate": 1e-05,
      "loss": 0.0736,
      "step": 1840
    },
    {
      "epoch": 39.57219251336898,
      "grad_norm": 0.09291240572929382,
      "learning_rate": 1e-05,
      "loss": 0.0921,
      "step": 1850
    },
    {
      "epoch": 39.786096256684495,
      "grad_norm": 11.098824501037598,
      "learning_rate": 1e-05,
      "loss": 0.0778,
      "step": 1860
    },
    {
      "epoch": 40.0,
      "grad_norm": 40.247833251953125,
      "learning_rate": 1e-05,
      "loss": 0.0217,
      "step": 1870
    },
    {
      "epoch": 40.213903743315505,
      "grad_norm": 2.459455728530884,
      "learning_rate": 1e-05,
      "loss": 0.008,
      "step": 1880
    },
    {
      "epoch": 40.42780748663102,
      "grad_norm": 0.03633593022823334,
      "learning_rate": 1e-05,
      "loss": 0.0053,
      "step": 1890
    },
    {
      "epoch": 40.64171122994652,
      "grad_norm": 0.0483514666557312,
      "learning_rate": 1e-05,
      "loss": 0.0566,
      "step": 1900
    },
    {
      "epoch": 40.855614973262036,
      "grad_norm": 53.63500213623047,
      "learning_rate": 1e-05,
      "loss": 0.0183,
      "step": 1910
    },
    {
      "epoch": 41.06951871657754,
      "grad_norm": 0.13120533525943756,
      "learning_rate": 1e-05,
      "loss": 0.0099,
      "step": 1920
    },
    {
      "epoch": 41.283422459893046,
      "grad_norm": 0.0008334398153237998,
      "learning_rate": 1e-05,
      "loss": 0.0007,
      "step": 1930
    },
    {
      "epoch": 41.49732620320856,
      "grad_norm": 0.028349481523036957,
      "learning_rate": 1e-05,
      "loss": 0.0011,
      "step": 1940
    },
    {
      "epoch": 41.711229946524064,
      "grad_norm": 72.34516906738281,
      "learning_rate": 1e-05,
      "loss": 0.0117,
      "step": 1950
    },
    {
      "epoch": 41.92513368983957,
      "grad_norm": 0.0020225769840180874,
      "learning_rate": 1e-05,
      "loss": 0.0097,
      "step": 1960
    },
    {
      "epoch": 42.13903743315508,
      "grad_norm": 43.7913703918457,
      "learning_rate": 1e-05,
      "loss": 0.0821,
      "step": 1970
    },
    {
      "epoch": 42.35294117647059,
      "grad_norm": 18.823827743530273,
      "learning_rate": 1e-05,
      "loss": 0.01,
      "step": 1980
    },
    {
      "epoch": 42.5668449197861,
      "grad_norm": 0.01237189769744873,
      "learning_rate": 1e-05,
      "loss": 0.013,
      "step": 1990
    },
    {
      "epoch": 42.780748663101605,
      "grad_norm": 0.00256702140904963,
      "learning_rate": 1e-05,
      "loss": 0.0041,
      "step": 2000
    },
    {
      "epoch": 42.780748663101605,
      "eval_accuracy": 0.4,
      "eval_loss": 7.45703125,
      "eval_runtime": 0.871,
      "eval_samples_per_second": 11.481,
      "eval_steps_per_second": 1.148,
      "step": 2000
    },
    {
      "epoch": 42.99465240641711,
      "grad_norm": 1.2592170238494873,
      "learning_rate": 1e-05,
      "loss": 0.0114,
      "step": 2010
    },
    {
      "epoch": 43.20855614973262,
      "grad_norm": 0.007054993882775307,
      "learning_rate": 1e-05,
      "loss": 0.0012,
      "step": 2020
    },
    {
      "epoch": 43.42245989304813,
      "grad_norm": 0.01072179526090622,
      "learning_rate": 1e-05,
      "loss": 0.0322,
      "step": 2030
    },
    {
      "epoch": 43.63636363636363,
      "grad_norm": 0.06874937564134598,
      "learning_rate": 1e-05,
      "loss": 0.0056,
      "step": 2040
    },
    {
      "epoch": 43.850267379679146,
      "grad_norm": 0.3548222482204437,
      "learning_rate": 1e-05,
      "loss": 0.0567,
      "step": 2050
    },
    {
      "epoch": 44.06417112299465,
      "grad_norm": 0.00958499126136303,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "step": 2060
    },
    {
      "epoch": 44.27807486631016,
      "grad_norm": 0.022511236369609833,
      "learning_rate": 1e-05,
      "loss": 0.0001,
      "step": 2070
    },
    {
      "epoch": 44.49197860962567,
      "grad_norm": 0.0012719137594103813,
      "learning_rate": 1e-05,
      "loss": 0.0056,
      "step": 2080
    },
    {
      "epoch": 44.705882352941174,
      "grad_norm": 0.0017934085335582495,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 2090
    },
    {
      "epoch": 44.919786096256686,
      "grad_norm": 0.001212795963510871,
      "learning_rate": 1e-05,
      "loss": 0.0043,
      "step": 2100
    },
    {
      "epoch": 45.13368983957219,
      "grad_norm": 0.001217988203279674,
      "learning_rate": 1e-05,
      "loss": 0.0054,
      "step": 2110
    },
    {
      "epoch": 45.3475935828877,
      "grad_norm": 0.0004769123042933643,
      "learning_rate": 1e-05,
      "loss": 0.0076,
      "step": 2120
    },
    {
      "epoch": 45.56149732620321,
      "grad_norm": 0.04562573507428169,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "step": 2130
    },
    {
      "epoch": 45.775401069518715,
      "grad_norm": 0.009098031558096409,
      "learning_rate": 1e-05,
      "loss": 0.0072,
      "step": 2140
    },
    {
      "epoch": 45.98930481283423,
      "grad_norm": 0.0003972891136072576,
      "learning_rate": 1e-05,
      "loss": 0.0327,
      "step": 2150
    },
    {
      "epoch": 46.20320855614973,
      "grad_norm": 0.001338423229753971,
      "learning_rate": 1e-05,
      "loss": 0.0001,
      "step": 2160
    },
    {
      "epoch": 46.41711229946524,
      "grad_norm": 62.341766357421875,
      "learning_rate": 1e-05,
      "loss": 0.0439,
      "step": 2170
    },
    {
      "epoch": 46.63101604278075,
      "grad_norm": 0.00038688722997903824,
      "learning_rate": 1e-05,
      "loss": 0.0002,
      "step": 2180
    },
    {
      "epoch": 46.844919786096256,
      "grad_norm": 0.001497891847975552,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 2190
    },
    {
      "epoch": 47.05882352941177,
      "grad_norm": 0.013468620367348194,
      "learning_rate": 1e-05,
      "loss": 0.0018,
      "step": 2200
    },
    {
      "epoch": 47.27272727272727,
      "grad_norm": 0.00993560440838337,
      "learning_rate": 1e-05,
      "loss": 0.0029,
      "step": 2210
    },
    {
      "epoch": 47.48663101604278,
      "grad_norm": 0.0010626667644828558,
      "learning_rate": 1e-05,
      "loss": 0.0001,
      "step": 2220
    },
    {
      "epoch": 47.70053475935829,
      "grad_norm": 0.00042168135405518115,
      "learning_rate": 1e-05,
      "loss": 0.0052,
      "step": 2230
    },
    {
      "epoch": 47.9144385026738,
      "grad_norm": 0.0007084342068992555,
      "learning_rate": 1e-05,
      "loss": 0.0016,
      "step": 2240
    },
    {
      "epoch": 48.1283422459893,
      "grad_norm": 0.0010569526348263025,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 2250
    },
    {
      "epoch": 48.342245989304814,
      "grad_norm": 0.00023022103414405137,
      "learning_rate": 1e-05,
      "loss": 0.0008,
      "step": 2260
    },
    {
      "epoch": 48.55614973262032,
      "grad_norm": 0.005315893329679966,
      "learning_rate": 1e-05,
      "loss": 0.0091,
      "step": 2270
    },
    {
      "epoch": 48.77005347593583,
      "grad_norm": 0.007200363092124462,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 2280
    },
    {
      "epoch": 48.98395721925134,
      "grad_norm": 0.0005015671486034989,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 2290
    },
    {
      "epoch": 49.19786096256684,
      "grad_norm": 0.0022118561901152134,
      "learning_rate": 1e-05,
      "loss": 0.001,
      "step": 2300
    },
    {
      "epoch": 49.411764705882355,
      "grad_norm": 0.0009918130235746503,
      "learning_rate": 1e-05,
      "loss": 0.0028,
      "step": 2310
    },
    {
      "epoch": 49.62566844919786,
      "grad_norm": 0.0068803769536316395,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 2320
    },
    {
      "epoch": 49.839572192513366,
      "grad_norm": 0.0005201214225962758,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 2330
    },
    {
      "epoch": 50.05347593582888,
      "grad_norm": 0.003280508564785123,
      "learning_rate": 1e-05,
      "loss": 0.0021,
      "step": 2340
    },
    {
      "epoch": 50.26737967914438,
      "grad_norm": 0.0004493014421314001,
      "learning_rate": 1e-05,
      "loss": 0.005,
      "step": 2350
    },
    {
      "epoch": 50.481283422459896,
      "grad_norm": 0.0019184891134500504,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 2360
    },
    {
      "epoch": 50.6951871657754,
      "grad_norm": 0.000974490656517446,
      "learning_rate": 1e-05,
      "loss": 0.0009,
      "step": 2370
    },
    {
      "epoch": 50.90909090909091,
      "grad_norm": 0.0006326729780994356,
      "learning_rate": 1e-05,
      "loss": 0.0006,
      "step": 2380
    },
    {
      "epoch": 51.12299465240642,
      "grad_norm": 0.0008777513867244124,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 2390
    },
    {
      "epoch": 51.336898395721924,
      "grad_norm": 0.0005483902059495449,
      "learning_rate": 1e-05,
      "loss": 0.0021,
      "step": 2400
    },
    {
      "epoch": 51.336898395721924,
      "eval_accuracy": 0.4,
      "eval_loss": 7.8984375,
      "eval_runtime": 0.8716,
      "eval_samples_per_second": 11.474,
      "eval_steps_per_second": 1.147,
      "step": 2400
    }
  ],
  "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": 1.9814468053798748e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}