{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.12,
  "eval_steps": 1000,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002,
      "grad_norm": 0.5271317362785339,
      "learning_rate": 0.00066,
      "loss": 7.779569091796875,
      "step": 100
    },
    {
      "epoch": 0.004,
      "grad_norm": 0.6387752890586853,
      "learning_rate": 0.0009999976160274871,
      "loss": 6.24430908203125,
      "step": 200
    },
    {
      "epoch": 0.006,
      "grad_norm": 0.42202863097190857,
      "learning_rate": 0.000999977956590203,
      "loss": 5.698394165039063,
      "step": 300
    },
    {
      "epoch": 0.008,
      "grad_norm": 0.42532625794410706,
      "learning_rate": 0.0009999384399149713,
      "loss": 5.405557861328125,
      "step": 400
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.4566122889518738,
      "learning_rate": 0.0009998790675712448,
      "loss": 5.189588623046875,
      "step": 500
    },
    {
      "epoch": 0.012,
      "grad_norm": 0.4048032760620117,
      "learning_rate": 0.000999799841917069,
      "loss": 5.048131713867187,
      "step": 600
    },
    {
      "epoch": 0.014,
      "grad_norm": 0.33951064944267273,
      "learning_rate": 0.0009997007660989881,
      "loss": 4.923414611816407,
      "step": 700
    },
    {
      "epoch": 0.016,
      "grad_norm": 0.3192354440689087,
      "learning_rate": 0.0009995818440519195,
      "loss": 4.7589013671875,
      "step": 800
    },
    {
      "epoch": 0.018,
      "grad_norm": 0.4026377201080322,
      "learning_rate": 0.0009994430804989986,
      "loss": 4.643058166503907,
      "step": 900
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.3453679084777832,
      "learning_rate": 0.0009992844809513896,
      "loss": 4.579913330078125,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.269652397386301,
      "eval_loss": 4.505138874053955,
      "eval_runtime": 6.375,
      "eval_samples_per_second": 304.469,
      "eval_steps_per_second": 19.137,
      "step": 1000
    },
    {
      "epoch": 0.022,
      "grad_norm": 0.3148439824581146,
      "learning_rate": 0.0009991060517080686,
      "loss": 4.5075772094726565,
      "step": 1100
    },
    {
      "epoch": 0.024,
      "grad_norm": 0.3168841600418091,
      "learning_rate": 0.0009989077998555718,
      "loss": 4.395697326660156,
      "step": 1200
    },
    {
      "epoch": 0.026,
      "grad_norm": 0.277462899684906,
      "learning_rate": 0.0009986897332677144,
      "loss": 4.3901171875,
      "step": 1300
    },
    {
      "epoch": 0.028,
      "grad_norm": 0.27992936968803406,
      "learning_rate": 0.0009984518606052791,
      "loss": 4.348124084472656,
      "step": 1400
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.2656252980232239,
      "learning_rate": 0.0009981941913156701,
      "loss": 4.3058538818359375,
      "step": 1500
    },
    {
      "epoch": 0.032,
      "grad_norm": 0.36868757009506226,
      "learning_rate": 0.0009979167356325394,
      "loss": 4.28273681640625,
      "step": 1600
    },
    {
      "epoch": 0.034,
      "grad_norm": 0.25333118438720703,
      "learning_rate": 0.0009976195045753797,
      "loss": 4.249351501464844,
      "step": 1700
    },
    {
      "epoch": 0.036,
      "grad_norm": 0.2491893470287323,
      "learning_rate": 0.0009973025099490876,
      "loss": 4.22731689453125,
      "step": 1800
    },
    {
      "epoch": 0.038,
      "grad_norm": 0.28424108028411865,
      "learning_rate": 0.0009969657643434933,
      "loss": 4.169223022460938,
      "step": 1900
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.2666165232658386,
      "learning_rate": 0.0009966092811328612,
      "loss": 4.170299682617188,
      "step": 2000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.3056205014664501,
      "eval_loss": 4.103638648986816,
      "eval_runtime": 6.566,
      "eval_samples_per_second": 295.612,
      "eval_steps_per_second": 18.58,
      "step": 2000
    },
    {
      "epoch": 0.042,
      "grad_norm": 0.3338668942451477,
      "learning_rate": 0.0009962330744753604,
      "loss": 4.163348388671875,
      "step": 2100
    },
    {
      "epoch": 0.044,
      "grad_norm": 0.25480538606643677,
      "learning_rate": 0.0009958371593124988,
      "loss": 4.118557434082032,
      "step": 2200
    },
    {
      "epoch": 0.046,
      "grad_norm": 0.2351720631122589,
      "learning_rate": 0.0009954215513685335,
      "loss": 4.1006707763671875,
      "step": 2300
    },
    {
      "epoch": 0.048,
      "grad_norm": 0.23063364624977112,
      "learning_rate": 0.0009949862671498437,
      "loss": 4.084933166503906,
      "step": 2400
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.270161896944046,
      "learning_rate": 0.000994531323944276,
      "loss": 4.043960876464844,
      "step": 2500
    },
    {
      "epoch": 0.052,
      "grad_norm": 0.23590487241744995,
      "learning_rate": 0.0009940567398204583,
      "loss": 4.043964233398437,
      "step": 2600
    },
    {
      "epoch": 0.054,
      "grad_norm": 0.2604995369911194,
      "learning_rate": 0.000993562533627081,
      "loss": 4.023601379394531,
      "step": 2700
    },
    {
      "epoch": 0.056,
      "grad_norm": 0.21301409602165222,
      "learning_rate": 0.0009930487249921495,
      "loss": 4.006760864257813,
      "step": 2800
    },
    {
      "epoch": 0.058,
      "grad_norm": 0.21575306355953217,
      "learning_rate": 0.0009925153343222043,
      "loss": 3.99057861328125,
      "step": 2900
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.2405770868062973,
      "learning_rate": 0.00099196238280151,
      "loss": 4.004396667480469,
      "step": 3000
    },
    {
      "epoch": 0.06,
      "eval_accuracy": 0.3194532243250246,
      "eval_loss": 3.945106267929077,
      "eval_runtime": 6.283,
      "eval_samples_per_second": 308.927,
      "eval_steps_per_second": 19.417,
      "step": 3000
    },
    {
      "epoch": 0.062,
      "grad_norm": 0.22380003333091736,
      "learning_rate": 0.0009913898923912148,
      "loss": 3.9952774047851562,
      "step": 3100
    },
    {
      "epoch": 0.064,
      "grad_norm": 0.21692751348018646,
      "learning_rate": 0.0009907978858284775,
      "loss": 3.9829229736328124,
      "step": 3200
    },
    {
      "epoch": 0.066,
      "grad_norm": 0.22457344830036163,
      "learning_rate": 0.0009901863866255652,
      "loss": 3.9637200927734373,
      "step": 3300
    },
    {
      "epoch": 0.068,
      "grad_norm": 0.20820939540863037,
      "learning_rate": 0.0009895554190689191,
      "loss": 3.930269470214844,
      "step": 3400
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.1961025595664978,
      "learning_rate": 0.0009889050082181896,
      "loss": 3.91483154296875,
      "step": 3500
    },
    {
      "epoch": 0.072,
      "grad_norm": 0.23347830772399902,
      "learning_rate": 0.0009882351799052416,
      "loss": 3.8973043823242186,
      "step": 3600
    },
    {
      "epoch": 0.074,
      "grad_norm": 0.21384640038013458,
      "learning_rate": 0.0009875459607331285,
      "loss": 3.9072760009765624,
      "step": 3700
    },
    {
      "epoch": 0.076,
      "grad_norm": 0.21380040049552917,
      "learning_rate": 0.000986837378075035,
      "loss": 3.8974929809570313,
      "step": 3800
    },
    {
      "epoch": 0.078,
      "grad_norm": 0.2199064940214157,
      "learning_rate": 0.0009861094600731911,
      "loss": 3.884527587890625,
      "step": 3900
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.2245541661977768,
      "learning_rate": 0.000985362235637753,
      "loss": 3.863822021484375,
      "step": 4000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.3280885939521158,
      "eval_loss": 3.8499362468719482,
      "eval_runtime": 6.3172,
      "eval_samples_per_second": 307.255,
      "eval_steps_per_second": 19.312,
      "step": 4000
    },
    {
      "epoch": 0.082,
      "grad_norm": 0.20376908779144287,
      "learning_rate": 0.000984595734445656,
      "loss": 3.8685012817382813,
      "step": 4100
    },
    {
      "epoch": 0.084,
      "grad_norm": 0.18422065675258636,
      "learning_rate": 0.0009838099869394353,
      "loss": 3.853851318359375,
      "step": 4200
    },
    {
      "epoch": 0.086,
      "grad_norm": 0.19115814566612244,
      "learning_rate": 0.0009830050243260172,
      "loss": 3.8678164672851563,
      "step": 4300
    },
    {
      "epoch": 0.088,
      "grad_norm": 0.20425236225128174,
      "learning_rate": 0.0009821808785754791,
      "loss": 3.8287237548828124,
      "step": 4400
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.1944635510444641,
      "learning_rate": 0.0009813375824197811,
      "loss": 3.8451242065429687,
      "step": 4500
    },
    {
      "epoch": 0.092,
      "grad_norm": 0.17947296798229218,
      "learning_rate": 0.0009804751693514645,
      "loss": 3.845400695800781,
      "step": 4600
    },
    {
      "epoch": 0.094,
      "grad_norm": 0.20076608657836914,
      "learning_rate": 0.0009795936736223222,
      "loss": 3.8257626342773436,
      "step": 4700
    },
    {
      "epoch": 0.096,
      "grad_norm": 0.18554596602916718,
      "learning_rate": 0.0009786931302420386,
      "loss": 3.808670959472656,
      "step": 4800
    },
    {
      "epoch": 0.098,
      "grad_norm": 0.23913995921611786,
      "learning_rate": 0.000977773574976799,
      "loss": 3.810527038574219,
      "step": 4900
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.18414103984832764,
      "learning_rate": 0.0009768350443478686,
      "loss": 3.7990948486328127,
      "step": 5000
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.33500293894949945,
      "eval_loss": 3.7702300548553467,
      "eval_runtime": 6.4137,
      "eval_samples_per_second": 302.634,
      "eval_steps_per_second": 19.022,
      "step": 5000
    },
    {
      "epoch": 0.102,
      "grad_norm": 0.20944379270076752,
      "learning_rate": 0.0009758775756301432,
      "loss": 3.798116455078125,
      "step": 5100
    },
    {
      "epoch": 0.104,
      "grad_norm": 0.18056762218475342,
      "learning_rate": 0.0009749012068506673,
      "loss": 3.7997445678710937,
      "step": 5200
    },
    {
      "epoch": 0.106,
      "grad_norm": 0.20922255516052246,
      "learning_rate": 0.000973905976787125,
      "loss": 3.7765731811523438,
      "step": 5300
    },
    {
      "epoch": 0.108,
      "grad_norm": 0.19328580796718597,
      "learning_rate": 0.0009728919249662993,
      "loss": 3.729337463378906,
      "step": 5400
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.2031419277191162,
      "learning_rate": 0.0009718590916625022,
      "loss": 3.744057922363281,
      "step": 5500
    },
    {
      "epoch": 0.112,
      "grad_norm": 0.19743800163269043,
      "learning_rate": 0.0009708075178959756,
      "loss": 3.7054071044921875,
      "step": 5600
    },
    {
      "epoch": 0.114,
      "grad_norm": 0.1741236299276352,
      "learning_rate": 0.0009697372454312618,
      "loss": 3.699843444824219,
      "step": 5700
    },
    {
      "epoch": 0.116,
      "grad_norm": 0.1799750030040741,
      "learning_rate": 0.0009686483167755448,
      "loss": 3.7227572631835937,
      "step": 5800
    },
    {
      "epoch": 0.118,
      "grad_norm": 0.2036087065935135,
      "learning_rate": 0.000967540775176962,
      "loss": 3.6742709350585936,
      "step": 5900
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.21463191509246826,
      "learning_rate": 0.0009664146646228871,
      "loss": 3.6995980834960935,
      "step": 6000
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.34141821705074654,
      "eval_loss": 3.711214303970337,
      "eval_runtime": 6.3349,
      "eval_samples_per_second": 306.4,
      "eval_steps_per_second": 19.259,
      "step": 6000
    }
  ],
  "logging_steps": 100,
  "max_steps": 50000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 2000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.3519765200896e+17,
  "train_batch_size": 120,
  "trial_name": null,
  "trial_params": null
}