{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.3827751196172249,
  "eval_steps": 500,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004784688995215311,
      "grad_norm": 6.923069477081299,
      "learning_rate": 4.917142857142858e-06,
      "loss": 0.4274,
      "mean_token_accuracy": 0.9153166055679322,
      "num_tokens": 562634.0,
      "step": 5
    },
    {
      "epoch": 0.009569377990430622,
      "grad_norm": 0.9214457273483276,
      "learning_rate": 1.106357142857143e-05,
      "loss": 0.1174,
      "mean_token_accuracy": 0.9728275179862976,
      "num_tokens": 1117508.0,
      "step": 10
    },
    {
      "epoch": 0.014354066985645933,
      "grad_norm": 0.4124715328216553,
      "learning_rate": 1.7210000000000003e-05,
      "loss": 0.0288,
      "mean_token_accuracy": 0.9882451295852661,
      "num_tokens": 1664035.0,
      "step": 15
    },
    {
      "epoch": 0.019138755980861243,
      "grad_norm": 0.30297964811325073,
      "learning_rate": 2.3356428571428574e-05,
      "loss": 0.0159,
      "mean_token_accuracy": 0.9927031397819519,
      "num_tokens": 2219488.0,
      "step": 20
    },
    {
      "epoch": 0.023923444976076555,
      "grad_norm": 0.1963028460741043,
      "learning_rate": 2.950285714285715e-05,
      "loss": 0.0135,
      "mean_token_accuracy": 0.9932918190956116,
      "num_tokens": 2776280.0,
      "step": 25
    },
    {
      "epoch": 0.028708133971291867,
      "grad_norm": 0.2711648941040039,
      "learning_rate": 3.5649285714285724e-05,
      "loss": 0.0118,
      "mean_token_accuracy": 0.993871009349823,
      "num_tokens": 3332826.0,
      "step": 30
    },
    {
      "epoch": 0.03349282296650718,
      "grad_norm": 0.2680172622203827,
      "learning_rate": 4.179571428571429e-05,
      "loss": 0.0129,
      "mean_token_accuracy": 0.9940371870994568,
      "num_tokens": 3885525.0,
      "step": 35
    },
    {
      "epoch": 0.03827751196172249,
      "grad_norm": 0.28794169425964355,
      "learning_rate": 4.302375119816257e-05,
      "loss": 0.0099,
      "mean_token_accuracy": 0.9953582048416137,
      "num_tokens": 4440176.0,
      "step": 40
    },
    {
      "epoch": 0.0430622009569378,
      "grad_norm": 0.16371244192123413,
      "learning_rate": 4.301867827201191e-05,
      "loss": 0.0098,
      "mean_token_accuracy": 0.9949951767921448,
      "num_tokens": 4987034.0,
      "step": 45
    },
    {
      "epoch": 0.04784688995215311,
      "grad_norm": 0.19273635745048523,
      "learning_rate": 4.300970439749554e-05,
      "loss": 0.0115,
      "mean_token_accuracy": 0.99526948928833,
      "num_tokens": 5545414.0,
      "step": 50
    },
    {
      "epoch": 0.05263157894736842,
      "grad_norm": 0.1035027876496315,
      "learning_rate": 4.299683174515572e-05,
      "loss": 0.0115,
      "mean_token_accuracy": 0.9954748630523682,
      "num_tokens": 6109344.0,
      "step": 55
    },
    {
      "epoch": 0.05741626794258373,
      "grad_norm": 0.13568036258220673,
      "learning_rate": 4.298006342854567e-05,
      "loss": 0.0101,
      "mean_token_accuracy": 0.9951767921447754,
      "num_tokens": 6669528.0,
      "step": 60
    },
    {
      "epoch": 0.06220095693779904,
      "grad_norm": 0.1464010775089264,
      "learning_rate": 4.295940350347647e-05,
      "loss": 0.009,
      "mean_token_accuracy": 0.9958411931991578,
      "num_tokens": 7236493.0,
      "step": 65
    },
    {
      "epoch": 0.06698564593301436,
      "grad_norm": 0.11470970511436462,
      "learning_rate": 4.293485696703609e-05,
      "loss": 0.0088,
      "mean_token_accuracy": 0.9952039599418641,
      "num_tokens": 7794472.0,
      "step": 70
    },
    {
      "epoch": 0.07177033492822966,
      "grad_norm": 0.06180080771446228,
      "learning_rate": 4.2906429756380716e-05,
      "loss": 0.0075,
      "mean_token_accuracy": 0.996013879776001,
      "num_tokens": 8355629.0,
      "step": 75
    },
    {
      "epoch": 0.07655502392344497,
      "grad_norm": 0.15871667861938477,
      "learning_rate": 4.2874128747298694e-05,
      "loss": 0.0093,
      "mean_token_accuracy": 0.9954859972000122,
      "num_tokens": 8908291.0,
      "step": 80
    },
    {
      "epoch": 0.08133971291866028,
      "grad_norm": 0.09185042977333069,
      "learning_rate": 4.2837961752547505e-05,
      "loss": 0.008,
      "mean_token_accuracy": 0.9961334943771363,
      "num_tokens": 9467220.0,
      "step": 85
    },
    {
      "epoch": 0.0861244019138756,
      "grad_norm": 0.08072248101234436,
      "learning_rate": 4.2797937519964016e-05,
      "loss": 0.0094,
      "mean_token_accuracy": 0.9954643011093139,
      "num_tokens": 10029631.0,
      "step": 90
    },
    {
      "epoch": 0.09090909090909091,
      "grad_norm": 0.06886106729507446,
      "learning_rate": 4.2754065730348625e-05,
      "loss": 0.0072,
      "mean_token_accuracy": 0.9965373277664185,
      "num_tokens": 10594944.0,
      "step": 95
    },
    {
      "epoch": 0.09569377990430622,
      "grad_norm": 0.08072028309106827,
      "learning_rate": 4.270635699512376e-05,
      "loss": 0.0074,
      "mean_token_accuracy": 0.995949125289917,
      "num_tokens": 11157297.0,
      "step": 100
    },
    {
      "epoch": 0.10047846889952153,
      "grad_norm": 0.05935250222682953,
      "learning_rate": 4.265482285376721e-05,
      "loss": 0.0072,
      "mean_token_accuracy": 0.9966120004653931,
      "num_tokens": 11711248.0,
      "step": 105
    },
    {
      "epoch": 0.10526315789473684,
      "grad_norm": 0.29420608282089233,
      "learning_rate": 4.259947577102108e-05,
      "loss": 0.0074,
      "mean_token_accuracy": 0.9962902069091797,
      "num_tokens": 12258102.0,
      "step": 110
    },
    {
      "epoch": 0.11004784688995216,
      "grad_norm": 0.049765560775995255,
      "learning_rate": 4.254032913387685e-05,
      "loss": 0.0065,
      "mean_token_accuracy": 0.996790885925293,
      "num_tokens": 12819258.0,
      "step": 115
    },
    {
      "epoch": 0.11483253588516747,
      "grad_norm": 0.08367352932691574,
      "learning_rate": 4.2477397248337476e-05,
      "loss": 0.006,
      "mean_token_accuracy": 0.9973790884017945,
      "num_tokens": 13392933.0,
      "step": 120
    },
    {
      "epoch": 0.11961722488038277,
      "grad_norm": 0.08371472358703613,
      "learning_rate": 4.241069533595709e-05,
      "loss": 0.007,
      "mean_token_accuracy": 0.996220326423645,
      "num_tokens": 13952091.0,
      "step": 125
    },
    {
      "epoch": 0.12440191387559808,
      "grad_norm": 0.1115838810801506,
      "learning_rate": 4.2340239530159336e-05,
      "loss": 0.0081,
      "mean_token_accuracy": 0.9959458708763123,
      "num_tokens": 14515449.0,
      "step": 130
    },
    {
      "epoch": 0.1291866028708134,
      "grad_norm": 0.08440716564655304,
      "learning_rate": 4.226604687233514e-05,
      "loss": 0.0065,
      "mean_token_accuracy": 0.9969314575195313,
      "num_tokens": 15079636.0,
      "step": 135
    },
    {
      "epoch": 0.1339712918660287,
      "grad_norm": 0.15264005959033966,
      "learning_rate": 4.218813530772081e-05,
      "loss": 0.0072,
      "mean_token_accuracy": 0.9959210157394409,
      "num_tokens": 15636212.0,
      "step": 140
    },
    {
      "epoch": 0.13875598086124402,
      "grad_norm": 0.09543953090906143,
      "learning_rate": 4.210652368105759e-05,
      "loss": 0.0077,
      "mean_token_accuracy": 0.9961390972137452,
      "num_tokens": 16172065.0,
      "step": 145
    },
    {
      "epoch": 0.14354066985645933,
      "grad_norm": 0.05650537833571434,
      "learning_rate": 4.2021231732033596e-05,
      "loss": 0.0058,
      "mean_token_accuracy": 0.9970490217208863,
      "num_tokens": 16721332.0,
      "step": 150
    },
    {
      "epoch": 0.14832535885167464,
      "grad_norm": 0.09336641430854797,
      "learning_rate": 4.193228009050934e-05,
      "loss": 0.0072,
      "mean_token_accuracy": 0.9961395978927612,
      "num_tokens": 17277070.0,
      "step": 155
    },
    {
      "epoch": 0.15311004784688995,
      "grad_norm": 0.07291215658187866,
      "learning_rate": 4.183969027152788e-05,
      "loss": 0.0051,
      "mean_token_accuracy": 0.9973039507865906,
      "num_tokens": 17835974.0,
      "step": 160
    },
    {
      "epoch": 0.15789473684210525,
      "grad_norm": 0.06777418404817581,
      "learning_rate": 4.1743484670110926e-05,
      "loss": 0.0051,
      "mean_token_accuracy": 0.9970315933227539,
      "num_tokens": 18398462.0,
      "step": 165
    },
    {
      "epoch": 0.16267942583732056,
      "grad_norm": 0.06863659620285034,
      "learning_rate": 4.164368655584205e-05,
      "loss": 0.0063,
      "mean_token_accuracy": 0.9968485116958619,
      "num_tokens": 18961950.0,
      "step": 170
    },
    {
      "epoch": 0.1674641148325359,
      "grad_norm": 0.12660522758960724,
      "learning_rate": 4.154032006723842e-05,
      "loss": 0.0068,
      "mean_token_accuracy": 0.9970182418823242,
      "num_tokens": 19524145.0,
      "step": 175
    },
    {
      "epoch": 0.1722488038277512,
      "grad_norm": 0.08207586407661438,
      "learning_rate": 4.143341020591231e-05,
      "loss": 0.0069,
      "mean_token_accuracy": 0.996542227268219,
      "num_tokens": 20081330.0,
      "step": 180
    },
    {
      "epoch": 0.17703349282296652,
      "grad_norm": 0.0664004310965538,
      "learning_rate": 4.1322982830523904e-05,
      "loss": 0.0067,
      "mean_token_accuracy": 0.9966094136238098,
      "num_tokens": 20633071.0,
      "step": 185
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 0.05618370696902275,
      "learning_rate": 4.1209064650526745e-05,
      "loss": 0.006,
      "mean_token_accuracy": 0.9967063546180726,
      "num_tokens": 21194591.0,
      "step": 190
    },
    {
      "epoch": 0.18660287081339713,
      "grad_norm": 0.048017967492341995,
      "learning_rate": 4.1091683219707416e-05,
      "loss": 0.0067,
      "mean_token_accuracy": 0.9964720487594605,
      "num_tokens": 21758337.0,
      "step": 195
    },
    {
      "epoch": 0.19138755980861244,
      "grad_norm": 0.074529267847538,
      "learning_rate": 4.097086692952106e-05,
      "loss": 0.0053,
      "mean_token_accuracy": 0.9972187995910644,
      "num_tokens": 22329316.0,
      "step": 200
    },
    {
      "epoch": 0.19617224880382775,
      "grad_norm": 0.08167537301778793,
      "learning_rate": 4.0846645002224166e-05,
      "loss": 0.0062,
      "mean_token_accuracy": 0.9963092565536499,
      "num_tokens": 22891296.0,
      "step": 205
    },
    {
      "epoch": 0.20095693779904306,
      "grad_norm": 0.06797411292791367,
      "learning_rate": 4.071904748380657e-05,
      "loss": 0.0059,
      "mean_token_accuracy": 0.9966914176940918,
      "num_tokens": 23451406.0,
      "step": 210
    },
    {
      "epoch": 0.20574162679425836,
      "grad_norm": 0.046226996928453445,
      "learning_rate": 4.0588105236724024e-05,
      "loss": 0.0051,
      "mean_token_accuracy": 0.9973398685455322,
      "num_tokens": 24011087.0,
      "step": 215
    },
    {
      "epoch": 0.21052631578947367,
      "grad_norm": 0.07212501764297485,
      "learning_rate": 4.04538499324335e-05,
      "loss": 0.0058,
      "mean_token_accuracy": 0.9971255302429199,
      "num_tokens": 24566193.0,
      "step": 220
    },
    {
      "epoch": 0.215311004784689,
      "grad_norm": 0.07598569989204407,
      "learning_rate": 4.031631404373259e-05,
      "loss": 0.006,
      "mean_token_accuracy": 0.9966546177864075,
      "num_tokens": 25118212.0,
      "step": 225
    },
    {
      "epoch": 0.22009569377990432,
      "grad_norm": 0.08445984870195389,
      "learning_rate": 4.0175530836905304e-05,
      "loss": 0.0051,
      "mean_token_accuracy": 0.9975785493850708,
      "num_tokens": 25662573.0,
      "step": 230
    },
    {
      "epoch": 0.22488038277511962,
      "grad_norm": 0.05163101106882095,
      "learning_rate": 4.003153436367573e-05,
      "loss": 0.0049,
      "mean_token_accuracy": 0.9969427704811096,
      "num_tokens": 26223599.0,
      "step": 235
    },
    {
      "epoch": 0.22966507177033493,
      "grad_norm": 0.08453308045864105,
      "learning_rate": 3.988435945297194e-05,
      "loss": 0.0057,
      "mean_token_accuracy": 0.9975227952003479,
      "num_tokens": 26789167.0,
      "step": 240
    },
    {
      "epoch": 0.23444976076555024,
      "grad_norm": 0.047894496470689774,
      "learning_rate": 3.9734041702501684e-05,
      "loss": 0.0058,
      "mean_token_accuracy": 0.9967633724212647,
      "num_tokens": 27332532.0,
      "step": 245
    },
    {
      "epoch": 0.23923444976076555,
      "grad_norm": 0.047592829912900925,
      "learning_rate": 3.9580617470142356e-05,
      "loss": 0.0058,
      "mean_token_accuracy": 0.996880829334259,
      "num_tokens": 27887009.0,
      "step": 250
    },
    {
      "epoch": 0.24401913875598086,
      "grad_norm": 0.09274763613939285,
      "learning_rate": 3.942412386514689e-05,
      "loss": 0.0051,
      "mean_token_accuracy": 0.9974756479263306,
      "num_tokens": 28453010.0,
      "step": 255
    },
    {
      "epoch": 0.24880382775119617,
      "grad_norm": 0.06600073724985123,
      "learning_rate": 3.9264598739168083e-05,
      "loss": 0.0047,
      "mean_token_accuracy": 0.9972898960113525,
      "num_tokens": 29009936.0,
      "step": 260
    },
    {
      "epoch": 0.2535885167464115,
      "grad_norm": 0.06851529330015182,
      "learning_rate": 3.910208067710326e-05,
      "loss": 0.0048,
      "mean_token_accuracy": 0.9973468899726867,
      "num_tokens": 29566562.0,
      "step": 265
    },
    {
      "epoch": 0.2583732057416268,
      "grad_norm": 0.10837122052907944,
      "learning_rate": 3.8936608987761594e-05,
      "loss": 0.0071,
      "mean_token_accuracy": 0.9960312485694885,
      "num_tokens": 30107384.0,
      "step": 270
    },
    {
      "epoch": 0.2631578947368421,
      "grad_norm": 0.09420755505561829,
      "learning_rate": 3.876822369435639e-05,
      "loss": 0.0061,
      "mean_token_accuracy": 0.996771514415741,
      "num_tokens": 30665378.0,
      "step": 275
    },
    {
      "epoch": 0.2679425837320574,
      "grad_norm": 0.06199979409575462,
      "learning_rate": 3.859696552482449e-05,
      "loss": 0.006,
      "mean_token_accuracy": 0.9967570900917053,
      "num_tokens": 31233462.0,
      "step": 280
    },
    {
      "epoch": 0.2727272727272727,
      "grad_norm": 0.050011880695819855,
      "learning_rate": 3.8422875901975285e-05,
      "loss": 0.0067,
      "mean_token_accuracy": 0.9959561347961425,
      "num_tokens": 31777090.0,
      "step": 285
    },
    {
      "epoch": 0.27751196172248804,
      "grad_norm": 0.13513068854808807,
      "learning_rate": 3.824599693347163e-05,
      "loss": 0.0054,
      "mean_token_accuracy": 0.9972455739974976,
      "num_tokens": 32344683.0,
      "step": 290
    },
    {
      "epoch": 0.2822966507177033,
      "grad_norm": 0.06049080565571785,
      "learning_rate": 3.8066371401645134e-05,
      "loss": 0.0068,
      "mean_token_accuracy": 0.9970306634902955,
      "num_tokens": 32906164.0,
      "step": 295
    },
    {
      "epoch": 0.28708133971291866,
      "grad_norm": 0.06717154383659363,
      "learning_rate": 3.788404275314821e-05,
      "loss": 0.0064,
      "mean_token_accuracy": 0.9967247366905212,
      "num_tokens": 33465658.0,
      "step": 300
    },
    {
      "epoch": 0.291866028708134,
      "grad_norm": 0.10201875865459442,
      "learning_rate": 3.769905508844552e-05,
      "loss": 0.0057,
      "mean_token_accuracy": 0.996989393234253,
      "num_tokens": 34020090.0,
      "step": 305
    },
    {
      "epoch": 0.2966507177033493,
      "grad_norm": 0.03700055554509163,
      "learning_rate": 3.751145315114724e-05,
      "loss": 0.0061,
      "mean_token_accuracy": 0.9961345672607422,
      "num_tokens": 34571366.0,
      "step": 310
    },
    {
      "epoch": 0.3014354066985646,
      "grad_norm": 0.04184204339981079,
      "learning_rate": 3.732128231718674e-05,
      "loss": 0.0058,
      "mean_token_accuracy": 0.9969716668128967,
      "num_tokens": 35134075.0,
      "step": 315
    },
    {
      "epoch": 0.3062200956937799,
      "grad_norm": 0.04125404730439186,
      "learning_rate": 3.712858858384536e-05,
      "loss": 0.0057,
      "mean_token_accuracy": 0.9969127655029297,
      "num_tokens": 35682776.0,
      "step": 320
    },
    {
      "epoch": 0.31100478468899523,
      "grad_norm": 0.02419581450521946,
      "learning_rate": 3.693341855862689e-05,
      "loss": 0.0055,
      "mean_token_accuracy": 0.9965312957763672,
      "num_tokens": 36229883.0,
      "step": 325
    },
    {
      "epoch": 0.3157894736842105,
      "grad_norm": 0.0618126355111599,
      "learning_rate": 3.67358194479845e-05,
      "loss": 0.006,
      "mean_token_accuracy": 0.9970920801162719,
      "num_tokens": 36784716.0,
      "step": 330
    },
    {
      "epoch": 0.32057416267942584,
      "grad_norm": 0.04135734587907791,
      "learning_rate": 3.653583904590263e-05,
      "loss": 0.0056,
      "mean_token_accuracy": 0.997107458114624,
      "num_tokens": 37333651.0,
      "step": 335
    },
    {
      "epoch": 0.3253588516746411,
      "grad_norm": 0.037036579102277756,
      "learning_rate": 3.6333525722337014e-05,
      "loss": 0.0049,
      "mean_token_accuracy": 0.9972279071807861,
      "num_tokens": 37890282.0,
      "step": 340
    },
    {
      "epoch": 0.33014354066985646,
      "grad_norm": 0.0746559426188469,
      "learning_rate": 3.612892841151522e-05,
      "loss": 0.0046,
      "mean_token_accuracy": 0.9977670907974243,
      "num_tokens": 38458216.0,
      "step": 345
    },
    {
      "epoch": 0.3349282296650718,
      "grad_norm": 0.029700152575969696,
      "learning_rate": 3.592209660010074e-05,
      "loss": 0.0046,
      "mean_token_accuracy": 0.9973897814750672,
      "num_tokens": 39007024.0,
      "step": 350
    },
    {
      "epoch": 0.3397129186602871,
      "grad_norm": 0.024702252820134163,
      "learning_rate": 3.5713080315223536e-05,
      "loss": 0.005,
      "mean_token_accuracy": 0.9972789406776428,
      "num_tokens": 39559535.0,
      "step": 355
    },
    {
      "epoch": 0.3444976076555024,
      "grad_norm": 0.1009126529097557,
      "learning_rate": 3.550193011237972e-05,
      "loss": 0.005,
      "mean_token_accuracy": 0.9972064256668091,
      "num_tokens": 40122595.0,
      "step": 360
    },
    {
      "epoch": 0.3492822966507177,
      "grad_norm": 0.048237863928079605,
      "learning_rate": 3.528869706320361e-05,
      "loss": 0.0047,
      "mean_token_accuracy": 0.9971843361854553,
      "num_tokens": 40689946.0,
      "step": 365
    },
    {
      "epoch": 0.35406698564593303,
      "grad_norm": 0.039489373564720154,
      "learning_rate": 3.5073432743114794e-05,
      "loss": 0.0042,
      "mean_token_accuracy": 0.9969287157058716,
      "num_tokens": 41250456.0,
      "step": 370
    },
    {
      "epoch": 0.3588516746411483,
      "grad_norm": 0.046250659972429276,
      "learning_rate": 3.485618921884343e-05,
      "loss": 0.0052,
      "mean_token_accuracy": 0.9969390511512757,
      "num_tokens": 41806451.0,
      "step": 375
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 0.04269276559352875,
      "learning_rate": 3.463701903583667e-05,
      "loss": 0.0056,
      "mean_token_accuracy": 0.9968794226646424,
      "num_tokens": 42362517.0,
      "step": 380
    },
    {
      "epoch": 0.3684210526315789,
      "grad_norm": 0.05071031302213669,
      "learning_rate": 3.441597520554933e-05,
      "loss": 0.0045,
      "mean_token_accuracy": 0.9973430275917053,
      "num_tokens": 42918924.0,
      "step": 385
    },
    {
      "epoch": 0.37320574162679426,
      "grad_norm": 0.04529614746570587,
      "learning_rate": 3.41931111926218e-05,
      "loss": 0.0047,
      "mean_token_accuracy": 0.9974654674530029,
      "num_tokens": 43477772.0,
      "step": 390
    },
    {
      "epoch": 0.37799043062200954,
      "grad_norm": 0.022656572982668877,
      "learning_rate": 3.3968480901948394e-05,
      "loss": 0.0039,
      "mean_token_accuracy": 0.9975798964500427,
      "num_tokens": 44041805.0,
      "step": 395
    },
    {
      "epoch": 0.3827751196172249,
      "grad_norm": 0.02202552929520607,
      "learning_rate": 3.374213866563915e-05,
      "loss": 0.0039,
      "mean_token_accuracy": 0.997993528842926,
      "num_tokens": 44611893.0,
      "step": 400
    }
  ],
  "logging_steps": 5,
  "max_steps": 1045,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.5505998488018616e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}