{
  "best_metric": 0.018313532695174217,
  "best_model_checkpoint": "runs/deepseek_lora_20240423-133229/checkpoint-2000",
  "epoch": 0.6266154930680661,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 1.5461480617523193,
      "learning_rate": 4.0000000000000003e-07,
      "loss": 1.5239,
      "step": 10
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.797250747680664,
      "learning_rate": 8.000000000000001e-07,
      "loss": 1.4966,
      "step": 20
    },
    {
      "epoch": 0.01,
      "grad_norm": 4.274460315704346,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 1.4788,
      "step": 30
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.9981650114059448,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 1.4001,
      "step": 40
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.2322547435760498,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.4683,
      "step": 50
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.250878930091858,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 1.1961,
      "step": 60
    },
    {
      "epoch": 0.02,
      "grad_norm": 4.917982578277588,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 1.4261,
      "step": 70
    },
    {
      "epoch": 0.03,
      "grad_norm": 3.885340929031372,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 1.2742,
      "step": 80
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.170296311378479,
      "learning_rate": 3.6000000000000003e-06,
      "loss": 1.253,
      "step": 90
    },
    {
      "epoch": 0.03,
      "grad_norm": 6.992724895477295,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.2521,
      "step": 100
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.9540568590164185,
      "learning_rate": 4.4e-06,
      "loss": 1.0828,
      "step": 110
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.9779148697853088,
      "learning_rate": 4.800000000000001e-06,
      "loss": 1.1255,
      "step": 120
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.2271130084991455,
      "learning_rate": 5.2e-06,
      "loss": 0.9496,
      "step": 130
    },
    {
      "epoch": 0.04,
      "grad_norm": 2.1939122676849365,
      "learning_rate": 5.600000000000001e-06,
      "loss": 0.953,
      "step": 140
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.5918376445770264,
      "learning_rate": 6e-06,
      "loss": 0.9348,
      "step": 150
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.772602379322052,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 0.7762,
      "step": 160
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.8226258158683777,
      "learning_rate": 6.800000000000001e-06,
      "loss": 0.6798,
      "step": 170
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.0945135354995728,
      "learning_rate": 7.2000000000000005e-06,
      "loss": 0.5936,
      "step": 180
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.6972842812538147,
      "learning_rate": 7.600000000000001e-06,
      "loss": 0.4161,
      "step": 190
    },
    {
      "epoch": 0.06,
      "grad_norm": 3.9560859203338623,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.3959,
      "step": 200
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4297197759151459,
      "learning_rate": 8.400000000000001e-06,
      "loss": 0.3166,
      "step": 210
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.4494841992855072,
      "learning_rate": 8.8e-06,
      "loss": 0.2056,
      "step": 220
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.6981032490730286,
      "learning_rate": 9.200000000000002e-06,
      "loss": 0.3024,
      "step": 230
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.5783653259277344,
      "learning_rate": 9.600000000000001e-06,
      "loss": 0.4392,
      "step": 240
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.33407163619995117,
      "learning_rate": 1e-05,
      "loss": 0.3708,
      "step": 250
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.3096323311328888,
      "learning_rate": 1.04e-05,
      "loss": 0.1637,
      "step": 260
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.6187375783920288,
      "learning_rate": 1.0800000000000002e-05,
      "loss": 0.1078,
      "step": 270
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.27954748272895813,
      "learning_rate": 1.1200000000000001e-05,
      "loss": 0.2668,
      "step": 280
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.4082590937614441,
      "learning_rate": 1.16e-05,
      "loss": 0.1843,
      "step": 290
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.2993428707122803,
      "learning_rate": 1.2e-05,
      "loss": 0.1969,
      "step": 300
    },
    {
      "epoch": 0.1,
      "grad_norm": 3.0335004329681396,
      "learning_rate": 1.2400000000000002e-05,
      "loss": 0.122,
      "step": 310
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.0972365066409111,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 0.1354,
      "step": 320
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.26203760504722595,
      "learning_rate": 1.3200000000000002e-05,
      "loss": 0.17,
      "step": 330
    },
    {
      "epoch": 0.11,
      "grad_norm": 5.179258346557617,
      "learning_rate": 1.3600000000000002e-05,
      "loss": 0.1409,
      "step": 340
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.032190825790166855,
      "learning_rate": 1.4e-05,
      "loss": 0.076,
      "step": 350
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.02556140348315239,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 0.0468,
      "step": 360
    },
    {
      "epoch": 0.12,
      "grad_norm": 5.667189121246338,
      "learning_rate": 1.48e-05,
      "loss": 0.1345,
      "step": 370
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.07244217395782471,
      "learning_rate": 1.5200000000000002e-05,
      "loss": 0.0411,
      "step": 380
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.11255475878715515,
      "learning_rate": 1.5600000000000003e-05,
      "loss": 0.1907,
      "step": 390
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.9602845907211304,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.0717,
      "step": 400
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.013551376760005951,
      "learning_rate": 1.64e-05,
      "loss": 0.0808,
      "step": 410
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.14421305060386658,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 0.0604,
      "step": 420
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.01705547235906124,
      "learning_rate": 1.72e-05,
      "loss": 0.1742,
      "step": 430
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.021360402926802635,
      "learning_rate": 1.76e-05,
      "loss": 0.0886,
      "step": 440
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.04369840770959854,
      "learning_rate": 1.8e-05,
      "loss": 0.1063,
      "step": 450
    },
    {
      "epoch": 0.14,
      "grad_norm": 2.2816083431243896,
      "learning_rate": 1.8400000000000003e-05,
      "loss": 0.1061,
      "step": 460
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.00989863183349371,
      "learning_rate": 1.88e-05,
      "loss": 0.2317,
      "step": 470
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.014644264243543148,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 0.1054,
      "step": 480
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.13385985791683197,
      "learning_rate": 1.9600000000000002e-05,
      "loss": 0.1196,
      "step": 490
    },
    {
      "epoch": 0.16,
      "grad_norm": 2.099637985229492,
      "learning_rate": 2e-05,
      "loss": 0.1584,
      "step": 500
    },
    {
      "epoch": 0.16,
      "eval_loss": 0.05546905845403671,
      "eval_runtime": 62.1757,
      "eval_samples_per_second": 16.083,
      "eval_steps_per_second": 16.083,
      "step": 500
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.013259253464639187,
      "learning_rate": 1.9955555555555557e-05,
      "loss": 0.0373,
      "step": 510
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.1753215193748474,
      "learning_rate": 1.9911111111111112e-05,
      "loss": 0.0986,
      "step": 520
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.006998735014349222,
      "learning_rate": 1.9866666666666667e-05,
      "loss": 0.1372,
      "step": 530
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.3990759253501892,
      "learning_rate": 1.9822222222222226e-05,
      "loss": 0.127,
      "step": 540
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.8248735666275024,
      "learning_rate": 1.977777777777778e-05,
      "loss": 0.1239,
      "step": 550
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.005832708440721035,
      "learning_rate": 1.9733333333333336e-05,
      "loss": 0.1181,
      "step": 560
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.10705375671386719,
      "learning_rate": 1.968888888888889e-05,
      "loss": 0.0972,
      "step": 570
    },
    {
      "epoch": 0.18,
      "grad_norm": 3.7429046630859375,
      "learning_rate": 1.9644444444444447e-05,
      "loss": 0.1543,
      "step": 580
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.010058793239295483,
      "learning_rate": 1.9600000000000002e-05,
      "loss": 0.1146,
      "step": 590
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.012124909088015556,
      "learning_rate": 1.9555555555555557e-05,
      "loss": 0.1582,
      "step": 600
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.012700803577899933,
      "learning_rate": 1.9511111111111113e-05,
      "loss": 0.0711,
      "step": 610
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.2153177261352539,
      "learning_rate": 1.9466666666666668e-05,
      "loss": 0.0452,
      "step": 620
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.01289213914424181,
      "learning_rate": 1.9422222222222223e-05,
      "loss": 0.067,
      "step": 630
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.007021366152912378,
      "learning_rate": 1.9377777777777778e-05,
      "loss": 0.0776,
      "step": 640
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.005274974275380373,
      "learning_rate": 1.9333333333333333e-05,
      "loss": 0.2159,
      "step": 650
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.6235843896865845,
      "learning_rate": 1.928888888888889e-05,
      "loss": 0.0664,
      "step": 660
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.009710369631648064,
      "learning_rate": 1.9244444444444444e-05,
      "loss": 0.1125,
      "step": 670
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.0918051227927208,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 0.0755,
      "step": 680
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.012988925911486149,
      "learning_rate": 1.9155555555555558e-05,
      "loss": 0.095,
      "step": 690
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.09063850343227386,
      "learning_rate": 1.9111111111111113e-05,
      "loss": 0.0419,
      "step": 700
    },
    {
      "epoch": 0.22,
      "grad_norm": 3.1540229320526123,
      "learning_rate": 1.9066666666666668e-05,
      "loss": 0.0338,
      "step": 710
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.009780442342162132,
      "learning_rate": 1.9022222222222223e-05,
      "loss": 0.0935,
      "step": 720
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.011248274706304073,
      "learning_rate": 1.897777777777778e-05,
      "loss": 0.017,
      "step": 730
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.03275947645306587,
      "learning_rate": 1.8933333333333334e-05,
      "loss": 0.1128,
      "step": 740
    },
    {
      "epoch": 0.23,
      "grad_norm": 2.224815607070923,
      "learning_rate": 1.888888888888889e-05,
      "loss": 0.1213,
      "step": 750
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.9777376651763916,
      "learning_rate": 1.8844444444444444e-05,
      "loss": 0.1128,
      "step": 760
    },
    {
      "epoch": 0.24,
      "grad_norm": 7.6119561195373535,
      "learning_rate": 1.88e-05,
      "loss": 0.0557,
      "step": 770
    },
    {
      "epoch": 0.24,
      "grad_norm": 3.7092978954315186,
      "learning_rate": 1.8755555555555558e-05,
      "loss": 0.1196,
      "step": 780
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.08959495276212692,
      "learning_rate": 1.8711111111111113e-05,
      "loss": 0.0231,
      "step": 790
    },
    {
      "epoch": 0.25,
      "grad_norm": 2.716322183609009,
      "learning_rate": 1.866666666666667e-05,
      "loss": 0.1253,
      "step": 800
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.006216592621058226,
      "learning_rate": 1.8622222222222224e-05,
      "loss": 0.0377,
      "step": 810
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.005503402091562748,
      "learning_rate": 1.857777777777778e-05,
      "loss": 0.1072,
      "step": 820
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.4769572913646698,
      "learning_rate": 1.8533333333333334e-05,
      "loss": 0.0888,
      "step": 830
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.034854818135499954,
      "learning_rate": 1.848888888888889e-05,
      "loss": 0.0424,
      "step": 840
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.00536635797470808,
      "learning_rate": 1.8444444444444448e-05,
      "loss": 0.0161,
      "step": 850
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.002790963975712657,
      "learning_rate": 1.8400000000000003e-05,
      "loss": 0.025,
      "step": 860
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.003568426240235567,
      "learning_rate": 1.835555555555556e-05,
      "loss": 0.0597,
      "step": 870
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.005008562933653593,
      "learning_rate": 1.8311111111111114e-05,
      "loss": 0.1817,
      "step": 880
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.006756258197128773,
      "learning_rate": 1.826666666666667e-05,
      "loss": 0.0084,
      "step": 890
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.3732385635375977,
      "learning_rate": 1.8222222222222224e-05,
      "loss": 0.072,
      "step": 900
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.003554440103471279,
      "learning_rate": 1.817777777777778e-05,
      "loss": 0.0507,
      "step": 910
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.06785377860069275,
      "learning_rate": 1.8133333333333335e-05,
      "loss": 0.0428,
      "step": 920
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.009546133689582348,
      "learning_rate": 1.808888888888889e-05,
      "loss": 0.0179,
      "step": 930
    },
    {
      "epoch": 0.29,
      "grad_norm": 4.595461368560791,
      "learning_rate": 1.8044444444444445e-05,
      "loss": 0.0506,
      "step": 940
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.310696601867676,
      "learning_rate": 1.8e-05,
      "loss": 0.0719,
      "step": 950
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.8953649997711182,
      "learning_rate": 1.7955555555555556e-05,
      "loss": 0.1718,
      "step": 960
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.706213116645813,
      "learning_rate": 1.791111111111111e-05,
      "loss": 0.0807,
      "step": 970
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.06022366136312485,
      "learning_rate": 1.7866666666666666e-05,
      "loss": 0.0792,
      "step": 980
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.033781811594963074,
      "learning_rate": 1.782222222222222e-05,
      "loss": 0.0864,
      "step": 990
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.014879009686410427,
      "learning_rate": 1.7777777777777777e-05,
      "loss": 0.0527,
      "step": 1000
    },
    {
      "epoch": 0.31,
      "eval_loss": 0.03033074364066124,
      "eval_runtime": 62.1716,
      "eval_samples_per_second": 16.085,
      "eval_steps_per_second": 16.085,
      "step": 1000
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.003431650809943676,
      "learning_rate": 1.7733333333333335e-05,
      "loss": 0.0425,
      "step": 1010
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.005604119040071964,
      "learning_rate": 1.768888888888889e-05,
      "loss": 0.0727,
      "step": 1020
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.058546032756567,
      "learning_rate": 1.7644444444444446e-05,
      "loss": 0.1304,
      "step": 1030
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.0267212390899658,
      "learning_rate": 1.76e-05,
      "loss": 0.0496,
      "step": 1040
    },
    {
      "epoch": 0.33,
      "grad_norm": 3.5381667613983154,
      "learning_rate": 1.7555555555555556e-05,
      "loss": 0.0935,
      "step": 1050
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.006851164624094963,
      "learning_rate": 1.751111111111111e-05,
      "loss": 0.0344,
      "step": 1060
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.005180824548006058,
      "learning_rate": 1.7466666666666667e-05,
      "loss": 0.0195,
      "step": 1070
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.586403489112854,
      "learning_rate": 1.7422222222222222e-05,
      "loss": 0.0868,
      "step": 1080
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.14490236341953278,
      "learning_rate": 1.737777777777778e-05,
      "loss": 0.0489,
      "step": 1090
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.005752285942435265,
      "learning_rate": 1.7333333333333336e-05,
      "loss": 0.0625,
      "step": 1100
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.8000261187553406,
      "learning_rate": 1.728888888888889e-05,
      "loss": 0.0923,
      "step": 1110
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.024827931076288223,
      "learning_rate": 1.7244444444444446e-05,
      "loss": 0.0355,
      "step": 1120
    },
    {
      "epoch": 0.35,
      "grad_norm": 3.424243927001953,
      "learning_rate": 1.72e-05,
      "loss": 0.0659,
      "step": 1130
    },
    {
      "epoch": 0.36,
      "grad_norm": 3.4055392742156982,
      "learning_rate": 1.7155555555555557e-05,
      "loss": 0.0681,
      "step": 1140
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.0077874124981462955,
      "learning_rate": 1.7111111111111112e-05,
      "loss": 0.106,
      "step": 1150
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.006299824919551611,
      "learning_rate": 1.706666666666667e-05,
      "loss": 0.0438,
      "step": 1160
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.004886701237410307,
      "learning_rate": 1.7022222222222226e-05,
      "loss": 0.0403,
      "step": 1170
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.0028305843006819487,
      "learning_rate": 1.697777777777778e-05,
      "loss": 0.0299,
      "step": 1180
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.36411717534065247,
      "learning_rate": 1.6933333333333336e-05,
      "loss": 0.0639,
      "step": 1190
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.004984024912118912,
      "learning_rate": 1.688888888888889e-05,
      "loss": 0.1363,
      "step": 1200
    },
    {
      "epoch": 0.38,
      "grad_norm": 2.4117214679718018,
      "learning_rate": 1.6844444444444447e-05,
      "loss": 0.018,
      "step": 1210
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.2183905839920044,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 0.0062,
      "step": 1220
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.05338621512055397,
      "learning_rate": 1.6755555555555557e-05,
      "loss": 0.0343,
      "step": 1230
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.009272832423448563,
      "learning_rate": 1.6711111111111112e-05,
      "loss": 0.073,
      "step": 1240
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.0033628540113568306,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.0272,
      "step": 1250
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.055562809109687805,
      "learning_rate": 1.6622222222222223e-05,
      "loss": 0.0821,
      "step": 1260
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.6196753978729248,
      "learning_rate": 1.6577777777777778e-05,
      "loss": 0.0153,
      "step": 1270
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.004278264008462429,
      "learning_rate": 1.6533333333333333e-05,
      "loss": 0.022,
      "step": 1280
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.044023897498846054,
      "learning_rate": 1.648888888888889e-05,
      "loss": 0.0652,
      "step": 1290
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.08157920837402344,
      "learning_rate": 1.6444444444444444e-05,
      "loss": 0.0745,
      "step": 1300
    },
    {
      "epoch": 0.41,
      "grad_norm": 2.1145012378692627,
      "learning_rate": 1.64e-05,
      "loss": 0.0497,
      "step": 1310
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.00495466822758317,
      "learning_rate": 1.6355555555555557e-05,
      "loss": 0.0688,
      "step": 1320
    },
    {
      "epoch": 0.42,
      "grad_norm": 2.055875778198242,
      "learning_rate": 1.6311111111111113e-05,
      "loss": 0.0456,
      "step": 1330
    },
    {
      "epoch": 0.42,
      "grad_norm": 7.260204315185547,
      "learning_rate": 1.6266666666666668e-05,
      "loss": 0.0968,
      "step": 1340
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.00300275138579309,
      "learning_rate": 1.6222222222222223e-05,
      "loss": 0.0031,
      "step": 1350
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.6148861050605774,
      "learning_rate": 1.617777777777778e-05,
      "loss": 0.1008,
      "step": 1360
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.005163553170859814,
      "learning_rate": 1.6133333333333334e-05,
      "loss": 0.0454,
      "step": 1370
    },
    {
      "epoch": 0.43,
      "grad_norm": 3.313544988632202,
      "learning_rate": 1.608888888888889e-05,
      "loss": 0.0843,
      "step": 1380
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.061028964817523956,
      "learning_rate": 1.6044444444444444e-05,
      "loss": 0.0467,
      "step": 1390
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.011183706112205982,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.0539,
      "step": 1400
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.00391357159242034,
      "learning_rate": 1.5955555555555558e-05,
      "loss": 0.007,
      "step": 1410
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.21193568408489227,
      "learning_rate": 1.5911111111111113e-05,
      "loss": 0.0723,
      "step": 1420
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.344232201576233,
      "learning_rate": 1.586666666666667e-05,
      "loss": 0.0093,
      "step": 1430
    },
    {
      "epoch": 0.45,
      "grad_norm": 2.125556707382202,
      "learning_rate": 1.5822222222222224e-05,
      "loss": 0.1089,
      "step": 1440
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.9358251094818115,
      "learning_rate": 1.577777777777778e-05,
      "loss": 0.129,
      "step": 1450
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.25835612416267395,
      "learning_rate": 1.5733333333333334e-05,
      "loss": 0.0225,
      "step": 1460
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.004413430579006672,
      "learning_rate": 1.5688888888888893e-05,
      "loss": 0.0184,
      "step": 1470
    },
    {
      "epoch": 0.46,
      "grad_norm": 2.6624996662139893,
      "learning_rate": 1.5644444444444448e-05,
      "loss": 0.0924,
      "step": 1480
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.023669052869081497,
      "learning_rate": 1.5600000000000003e-05,
      "loss": 0.0468,
      "step": 1490
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.9958231449127197,
      "learning_rate": 1.555555555555556e-05,
      "loss": 0.0602,
      "step": 1500
    },
    {
      "epoch": 0.47,
      "eval_loss": 0.023086324334144592,
      "eval_runtime": 62.044,
      "eval_samples_per_second": 16.118,
      "eval_steps_per_second": 16.118,
      "step": 1500
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.0457289032638073,
      "learning_rate": 1.5511111111111114e-05,
      "loss": 0.0725,
      "step": 1510
    },
    {
      "epoch": 0.48,
      "grad_norm": 2.0417847633361816,
      "learning_rate": 1.546666666666667e-05,
      "loss": 0.0446,
      "step": 1520
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.007491718512028456,
      "learning_rate": 1.5422222222222224e-05,
      "loss": 0.0693,
      "step": 1530
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.0023406276013702154,
      "learning_rate": 1.537777777777778e-05,
      "loss": 0.0398,
      "step": 1540
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.03808426856994629,
      "learning_rate": 1.5333333333333334e-05,
      "loss": 0.0827,
      "step": 1550
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.04878142476081848,
      "learning_rate": 1.528888888888889e-05,
      "loss": 0.0094,
      "step": 1560
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.002342456951737404,
      "learning_rate": 1.5244444444444447e-05,
      "loss": 0.0308,
      "step": 1570
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.002332750940695405,
      "learning_rate": 1.5200000000000002e-05,
      "loss": 0.0703,
      "step": 1580
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.027806026861071587,
      "learning_rate": 1.5155555555555557e-05,
      "loss": 0.0577,
      "step": 1590
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.853770136833191,
      "learning_rate": 1.5111111111111112e-05,
      "loss": 0.0872,
      "step": 1600
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.02809343859553337,
      "learning_rate": 1.5066666666666668e-05,
      "loss": 0.0651,
      "step": 1610
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.29635584354400635,
      "learning_rate": 1.5022222222222223e-05,
      "loss": 0.0784,
      "step": 1620
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.006036743987351656,
      "learning_rate": 1.497777777777778e-05,
      "loss": 0.0702,
      "step": 1630
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.0020562768913805485,
      "learning_rate": 1.4933333333333335e-05,
      "loss": 0.1133,
      "step": 1640
    },
    {
      "epoch": 0.52,
      "grad_norm": 2.0853707790374756,
      "learning_rate": 1.488888888888889e-05,
      "loss": 0.0717,
      "step": 1650
    },
    {
      "epoch": 0.52,
      "grad_norm": 2.024653434753418,
      "learning_rate": 1.4844444444444445e-05,
      "loss": 0.0978,
      "step": 1660
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.0029632914811372757,
      "learning_rate": 1.48e-05,
      "loss": 0.0269,
      "step": 1670
    },
    {
      "epoch": 0.53,
      "grad_norm": 3.2619192600250244,
      "learning_rate": 1.4755555555555556e-05,
      "loss": 0.088,
      "step": 1680
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.9590596556663513,
      "learning_rate": 1.4711111111111111e-05,
      "loss": 0.0348,
      "step": 1690
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.01058279536664486,
      "learning_rate": 1.4666666666666666e-05,
      "loss": 0.0473,
      "step": 1700
    },
    {
      "epoch": 0.54,
      "grad_norm": 4.378236293792725,
      "learning_rate": 1.4622222222222225e-05,
      "loss": 0.0839,
      "step": 1710
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.04728490114212036,
      "learning_rate": 1.457777777777778e-05,
      "loss": 0.1131,
      "step": 1720
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.03082725591957569,
      "learning_rate": 1.4533333333333335e-05,
      "loss": 0.058,
      "step": 1730
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.4159008264541626,
      "learning_rate": 1.448888888888889e-05,
      "loss": 0.027,
      "step": 1740
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.09299231320619583,
      "learning_rate": 1.4444444444444446e-05,
      "loss": 0.0152,
      "step": 1750
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.0018398365937173367,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 0.0005,
      "step": 1760
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.0032970942556858063,
      "learning_rate": 1.4355555555555556e-05,
      "loss": 0.068,
      "step": 1770
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.9869989156723022,
      "learning_rate": 1.4311111111111111e-05,
      "loss": 0.0802,
      "step": 1780
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.4349244236946106,
      "learning_rate": 1.4266666666666668e-05,
      "loss": 0.0022,
      "step": 1790
    },
    {
      "epoch": 0.56,
      "grad_norm": 6.195014476776123,
      "learning_rate": 1.4222222222222224e-05,
      "loss": 0.075,
      "step": 1800
    },
    {
      "epoch": 0.57,
      "grad_norm": 4.658276557922363,
      "learning_rate": 1.4177777777777779e-05,
      "loss": 0.0341,
      "step": 1810
    },
    {
      "epoch": 0.57,
      "grad_norm": 2.3556618690490723,
      "learning_rate": 1.4133333333333334e-05,
      "loss": 0.0735,
      "step": 1820
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.00235918490216136,
      "learning_rate": 1.408888888888889e-05,
      "loss": 0.0133,
      "step": 1830
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.0019710322376340628,
      "learning_rate": 1.4044444444444445e-05,
      "loss": 0.0714,
      "step": 1840
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.0038756057620048523,
      "learning_rate": 1.4e-05,
      "loss": 0.058,
      "step": 1850
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.13870251178741455,
      "learning_rate": 1.3955555555555558e-05,
      "loss": 0.0598,
      "step": 1860
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.0035449981223791838,
      "learning_rate": 1.3911111111111114e-05,
      "loss": 0.0825,
      "step": 1870
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.005458540748804808,
      "learning_rate": 1.3866666666666669e-05,
      "loss": 0.0485,
      "step": 1880
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.0037788543850183487,
      "learning_rate": 1.3822222222222224e-05,
      "loss": 0.0279,
      "step": 1890
    },
    {
      "epoch": 0.6,
      "grad_norm": 2.4127650260925293,
      "learning_rate": 1.377777777777778e-05,
      "loss": 0.0414,
      "step": 1900
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.9315361976623535,
      "learning_rate": 1.3733333333333335e-05,
      "loss": 0.0541,
      "step": 1910
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.001827953034080565,
      "learning_rate": 1.368888888888889e-05,
      "loss": 0.0607,
      "step": 1920
    },
    {
      "epoch": 0.6,
      "grad_norm": 3.0561037063598633,
      "learning_rate": 1.3644444444444445e-05,
      "loss": 0.0489,
      "step": 1930
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.7366918325424194,
      "learning_rate": 1.3600000000000002e-05,
      "loss": 0.0107,
      "step": 1940
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.0025562909431755543,
      "learning_rate": 1.3555555555555557e-05,
      "loss": 0.0305,
      "step": 1950
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.006456757429987192,
      "learning_rate": 1.3511111111111112e-05,
      "loss": 0.0741,
      "step": 1960
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.003276415169239044,
      "learning_rate": 1.3466666666666668e-05,
      "loss": 0.0276,
      "step": 1970
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.0036139176227152348,
      "learning_rate": 1.3422222222222223e-05,
      "loss": 0.0886,
      "step": 1980
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.006649693939834833,
      "learning_rate": 1.3377777777777778e-05,
      "loss": 0.0179,
      "step": 1990
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.002991742454469204,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.0372,
      "step": 2000
    },
    {
      "epoch": 0.63,
      "eval_loss": 0.018313532695174217,
      "eval_runtime": 62.0835,
      "eval_samples_per_second": 16.107,
      "eval_steps_per_second": 16.107,
      "step": 2000
    }
  ],
  "logging_steps": 10,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 1000,
  "total_flos": 3.2204251987968e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}