{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.962962962962963,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.805763495862484,
      "epoch": 0.07407407407407407,
      "grad_norm": 0.6806204319000244,
      "learning_rate": 8.032786885245902e-05,
      "loss": 1.5484,
      "mean_token_accuracy": 0.6665007689595223,
      "num_tokens": 408149.0,
      "step": 50
    },
    {
      "entropy": 0.41144788280129435,
      "epoch": 0.14814814814814814,
      "grad_norm": 0.38455039262771606,
      "learning_rate": 9.990765991730485e-05,
      "loss": 0.3321,
      "mean_token_accuracy": 0.9129975068569184,
      "num_tokens": 816230.0,
      "step": 100
    },
    {
      "entropy": 0.16323913749307395,
      "epoch": 0.2222222222222222,
      "grad_norm": 0.29704713821411133,
      "learning_rate": 9.950545603782162e-05,
      "loss": 0.1447,
      "mean_token_accuracy": 0.9618216013908386,
      "num_tokens": 1224471.0,
      "step": 150
    },
    {
      "entropy": 0.1175146003998816,
      "epoch": 0.2962962962962963,
      "grad_norm": 0.35487300157546997,
      "learning_rate": 9.878674879048427e-05,
      "loss": 0.1071,
      "mean_token_accuracy": 0.9733556269109249,
      "num_tokens": 1632497.0,
      "step": 200
    },
    {
      "entropy": 0.1009879010822624,
      "epoch": 0.37037037037037035,
      "grad_norm": 0.17419321835041046,
      "learning_rate": 9.775613308830824e-05,
      "loss": 0.0925,
      "mean_token_accuracy": 0.9769376286864281,
      "num_tokens": 2041392.0,
      "step": 250
    },
    {
      "entropy": 0.09154447751119733,
      "epoch": 0.4444444444444444,
      "grad_norm": 0.20543242990970612,
      "learning_rate": 9.642019796948866e-05,
      "loss": 0.0836,
      "mean_token_accuracy": 0.9787026332318782,
      "num_tokens": 2450311.0,
      "step": 300
    },
    {
      "entropy": 0.08632300381548702,
      "epoch": 0.5185185185185185,
      "grad_norm": 0.17172595858573914,
      "learning_rate": 9.478748447168449e-05,
      "loss": 0.0812,
      "mean_token_accuracy": 0.9789653661847114,
      "num_tokens": 2858744.0,
      "step": 350
    },
    {
      "entropy": 0.08412999271415174,
      "epoch": 0.5925925925925926,
      "grad_norm": 0.1447569578886032,
      "learning_rate": 9.28684310265789e-05,
      "loss": 0.0805,
      "mean_token_accuracy": 0.9786932443082332,
      "num_tokens": 3265542.0,
      "step": 400
    },
    {
      "entropy": 0.08065679710358381,
      "epoch": 0.6666666666666666,
      "grad_norm": 0.19630704820156097,
      "learning_rate": 9.067530672382544e-05,
      "loss": 0.0773,
      "mean_token_accuracy": 0.9797722736001014,
      "num_tokens": 3674162.0,
      "step": 450
    },
    {
      "entropy": 0.07874332463368773,
      "epoch": 0.7407407407407407,
      "grad_norm": 0.08524929732084274,
      "learning_rate": 8.822213287104348e-05,
      "loss": 0.0762,
      "mean_token_accuracy": 0.9801681047677994,
      "num_tokens": 4082734.0,
      "step": 500
    },
    {
      "entropy": 0.07778633100911975,
      "epoch": 0.8148148148148148,
      "grad_norm": 0.10848797112703323,
      "learning_rate": 8.552459335135381e-05,
      "loss": 0.0753,
      "mean_token_accuracy": 0.9801766823232174,
      "num_tokens": 4491115.0,
      "step": 550
    },
    {
      "entropy": 0.07791180345229805,
      "epoch": 0.8888888888888888,
      "grad_norm": 0.12547598779201508,
      "learning_rate": 8.259993435156559e-05,
      "loss": 0.0752,
      "mean_token_accuracy": 0.9802092918753624,
      "num_tokens": 4899794.0,
      "step": 600
    },
    {
      "entropy": 0.07790746555663645,
      "epoch": 0.9629629629629629,
      "grad_norm": 0.0992884486913681,
      "learning_rate": 7.946685410208296e-05,
      "loss": 0.0759,
      "mean_token_accuracy": 0.979848040342331,
      "num_tokens": 5307342.0,
      "step": 650
    },
    {
      "entropy": 0.07524958597496152,
      "epoch": 1.037037037037037,
      "grad_norm": 0.086652472615242,
      "learning_rate": 7.614538333345735e-05,
      "loss": 0.0731,
      "mean_token_accuracy": 0.9808213406801224,
      "num_tokens": 5716156.0,
      "step": 700
    },
    {
      "entropy": 0.07525249728001654,
      "epoch": 1.1111111111111112,
      "grad_norm": 0.13188883662223816,
      "learning_rate": 7.265675721386285e-05,
      "loss": 0.0728,
      "mean_token_accuracy": 0.9810096868872642,
      "num_tokens": 6123905.0,
      "step": 750
    },
    {
      "entropy": 0.07601501471363008,
      "epoch": 1.1851851851851851,
      "grad_norm": 0.0819055363535881,
      "learning_rate": 6.902327958623736e-05,
      "loss": 0.0736,
      "mean_token_accuracy": 0.9805573572218418,
      "num_tokens": 6532143.0,
      "step": 800
    },
    {
      "entropy": 0.07429057988338172,
      "epoch": 1.2592592592592593,
      "grad_norm": 0.09344803541898727,
      "learning_rate": 6.526818037306228e-05,
      "loss": 0.0727,
      "mean_token_accuracy": 0.9813062380254268,
      "num_tokens": 6940424.0,
      "step": 850
    },
    {
      "entropy": 0.07440330957062542,
      "epoch": 1.3333333333333333,
      "grad_norm": 0.11321987956762314,
      "learning_rate": 6.14154670604355e-05,
      "loss": 0.0725,
      "mean_token_accuracy": 0.9811275874078274,
      "num_tokens": 7348719.0,
      "step": 900
    },
    {
      "entropy": 0.07393304943107068,
      "epoch": 1.4074074074074074,
      "grad_norm": 0.087185338139534,
      "learning_rate": 5.7489771210944564e-05,
      "loss": 0.0726,
      "mean_token_accuracy": 0.9807305666804313,
      "num_tokens": 7756710.0,
      "step": 950
    },
    {
      "entropy": 0.07433672657236456,
      "epoch": 1.4814814814814814,
      "grad_norm": 0.10077723860740662,
      "learning_rate": 5.351619098663021e-05,
      "loss": 0.0726,
      "mean_token_accuracy": 0.9807634821534157,
      "num_tokens": 8165027.0,
      "step": 1000
    },
    {
      "entropy": 0.07262381819076837,
      "epoch": 1.5555555555555556,
      "grad_norm": 0.05774468928575516,
      "learning_rate": 4.952013068883795e-05,
      "loss": 0.0717,
      "mean_token_accuracy": 0.9813279174268246,
      "num_tokens": 8573428.0,
      "step": 1050
    },
    {
      "entropy": 0.07313465082086623,
      "epoch": 1.6296296296296298,
      "grad_norm": 0.156134694814682,
      "learning_rate": 4.5527138340828776e-05,
      "loss": 0.0717,
      "mean_token_accuracy": 0.9811334984004497,
      "num_tokens": 8981661.0,
      "step": 1100
    },
    {
      "entropy": 0.07226949028670787,
      "epoch": 1.7037037037037037,
      "grad_norm": 0.060666773468256,
      "learning_rate": 4.156274235153189e-05,
      "loss": 0.071,
      "mean_token_accuracy": 0.9813579262793064,
      "num_tokens": 9390026.0,
      "step": 1150
    },
    {
      "entropy": 0.07272680706344545,
      "epoch": 1.7777777777777777,
      "grad_norm": 0.10639354586601257,
      "learning_rate": 3.765228830469794e-05,
      "loss": 0.0711,
      "mean_token_accuracy": 0.981099860817194,
      "num_tokens": 9798433.0,
      "step": 1200
    },
    {
      "entropy": 0.07212639102712273,
      "epoch": 1.8518518518518519,
      "grad_norm": 0.06637933105230331,
      "learning_rate": 3.3820776916908857e-05,
      "loss": 0.0711,
      "mean_token_accuracy": 0.9812221045792103,
      "num_tokens": 10206854.0,
      "step": 1250
    },
    {
      "entropy": 0.07320461552590132,
      "epoch": 1.925925925925926,
      "grad_norm": 0.10055620223283768,
      "learning_rate": 3.0092704200428058e-05,
      "loss": 0.0719,
      "mean_token_accuracy": 0.9807316599786282,
      "num_tokens": 10615186.0,
      "step": 1300
    },
    {
      "entropy": 0.07290558220818638,
      "epoch": 2.0,
      "grad_norm": 0.0677887499332428,
      "learning_rate": 2.649190485277792e-05,
      "loss": 0.0718,
      "mean_token_accuracy": 0.9811730526387692,
      "num_tokens": 11023650.0,
      "step": 1350
    },
    {
      "entropy": 0.07251124914735556,
      "epoch": 2.074074074074074,
      "grad_norm": 0.11045810580253601,
      "learning_rate": 2.3041399874302905e-05,
      "loss": 0.0709,
      "mean_token_accuracy": 0.9809805656969547,
      "num_tokens": 11430802.0,
      "step": 1400
    },
    {
      "entropy": 0.07181317125447094,
      "epoch": 2.148148148148148,
      "grad_norm": 0.07149960100650787,
      "learning_rate": 1.976324938794482e-05,
      "loss": 0.0708,
      "mean_token_accuracy": 0.9813783176243305,
      "num_tokens": 11839809.0,
      "step": 1450
    },
    {
      "entropy": 0.0717768538929522,
      "epoch": 2.2222222222222223,
      "grad_norm": 0.05869750306010246,
      "learning_rate": 1.667841160219835e-05,
      "loss": 0.0707,
      "mean_token_accuracy": 0.9814658354222775,
      "num_tokens": 12248612.0,
      "step": 1500
    },
    {
      "entropy": 0.0715598820708692,
      "epoch": 2.2962962962962963,
      "grad_norm": 0.08496281504631042,
      "learning_rate": 1.3806608818939203e-05,
      "loss": 0.0705,
      "mean_token_accuracy": 0.9812778060138225,
      "num_tokens": 12657077.0,
      "step": 1550
    },
    {
      "entropy": 0.07179237512871622,
      "epoch": 2.3703703703703702,
      "grad_norm": 0.06790705770254135,
      "learning_rate": 1.1166201342777438e-05,
      "loss": 0.0709,
      "mean_token_accuracy": 0.9812556092441082,
      "num_tokens": 13064930.0,
      "step": 1600
    },
    {
      "entropy": 0.0720329173374921,
      "epoch": 2.4444444444444446,
      "grad_norm": 0.06569824367761612,
      "learning_rate": 8.774070098071668e-06,
      "loss": 0.0711,
      "mean_token_accuracy": 0.9811322076618672,
      "num_tokens": 13472953.0,
      "step": 1650
    },
    {
      "entropy": 0.07238605052232742,
      "epoch": 2.5185185185185186,
      "grad_norm": 0.05740037187933922,
      "learning_rate": 6.645508704069003e-06,
      "loss": 0.0712,
      "mean_token_accuracy": 0.9810698322951794,
      "num_tokens": 13881339.0,
      "step": 1700
    },
    {
      "entropy": 0.07082363245077432,
      "epoch": 2.5925925925925926,
      "grad_norm": 0.06362631171941757,
      "learning_rate": 4.794125698167262e-06,
      "loss": 0.0703,
      "mean_token_accuracy": 0.9813959409296512,
      "num_tokens": 14289567.0,
      "step": 1750
    },
    {
      "entropy": 0.07266209500841797,
      "epoch": 2.6666666666666665,
      "grad_norm": 0.05623815581202507,
      "learning_rate": 3.231757532415458e-06,
      "loss": 0.0714,
      "mean_token_accuracy": 0.9809911704063415,
      "num_tokens": 14697205.0,
      "step": 1800
    },
    {
      "entropy": 0.07201593144796789,
      "epoch": 2.7407407407407405,
      "grad_norm": 0.05804692208766937,
      "learning_rate": 1.9683928994924385e-06,
      "loss": 0.071,
      "mean_token_accuracy": 0.9810844567418099,
      "num_tokens": 15105573.0,
      "step": 1850
    },
    {
      "entropy": 0.07237200179137289,
      "epoch": 2.814814814814815,
      "grad_norm": 0.07103168219327927,
      "learning_rate": 1.0121088719706296e-06,
      "loss": 0.0715,
      "mean_token_accuracy": 0.9812616856396198,
      "num_tokens": 15514180.0,
      "step": 1900
    },
    {
      "entropy": 0.07117950812913477,
      "epoch": 2.888888888888889,
      "grad_norm": 0.062420960515737534,
      "learning_rate": 3.6901926314575894e-07,
      "loss": 0.0703,
      "mean_token_accuracy": 0.981419977247715,
      "num_tokens": 15922315.0,
      "step": 1950
    },
    {
      "entropy": 0.07066867646761238,
      "epoch": 2.962962962962963,
      "grad_norm": 0.05154326930642128,
      "learning_rate": 4.323553957759629e-08,
      "loss": 0.07,
      "mean_token_accuracy": 0.9817947860062123,
      "num_tokens": 16331178.0,
      "step": 2000
    }
  ],
  "logging_steps": 50,
  "max_steps": 2025,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 6.821968917982372e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}