{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 313,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.09649655148386956,
      "epoch": 0.032,
      "grad_norm": 0.2255859375,
      "learning_rate": 0.00018,
      "loss": 0.14522699117660523,
      "mean_token_accuracy": 0.9369595624506474,
      "num_tokens": 66476.0,
      "step": 10
    },
    {
      "entropy": 0.08880856377654708,
      "epoch": 0.064,
      "grad_norm": 0.173828125,
      "learning_rate": 0.0001995649347969019,
      "loss": 0.08154842257499695,
      "mean_token_accuracy": 0.9461333908140659,
      "num_tokens": 140569.0,
      "step": 20
    },
    {
      "entropy": 0.08434813401545398,
      "epoch": 0.096,
      "grad_norm": 0.169921875,
      "learning_rate": 0.0001980658652386421,
      "loss": 0.08437747955322265,
      "mean_token_accuracy": 0.9686024446040392,
      "num_tokens": 204086.0,
      "step": 30
    },
    {
      "entropy": 0.0869230936281383,
      "epoch": 0.128,
      "grad_norm": 0.1357421875,
      "learning_rate": 0.0001955135170885202,
      "loss": 0.07795352935791015,
      "mean_token_accuracy": 0.9808623071759939,
      "num_tokens": 271252.0,
      "step": 40
    },
    {
      "entropy": 0.07329072935972362,
      "epoch": 0.16,
      "grad_norm": 0.19140625,
      "learning_rate": 0.00019193530389822363,
      "loss": 0.08132061958312989,
      "mean_token_accuracy": 0.9672858133912087,
      "num_tokens": 337656.0,
      "step": 50
    },
    {
      "entropy": 0.0891664584632963,
      "epoch": 0.192,
      "grad_norm": 0.17578125,
      "learning_rate": 0.00018736965754556528,
      "loss": 0.09287956953048707,
      "mean_token_accuracy": 0.9650922071188688,
      "num_tokens": 410305.0,
      "step": 60
    },
    {
      "entropy": 0.07671013890067116,
      "epoch": 0.224,
      "grad_norm": 0.203125,
      "learning_rate": 0.00018186561545605054,
      "loss": 0.07794617414474488,
      "mean_token_accuracy": 0.9760642010718584,
      "num_tokens": 477217.0,
      "step": 70
    },
    {
      "entropy": 0.0767284018744249,
      "epoch": 0.256,
      "grad_norm": 0.123046875,
      "learning_rate": 0.00017548229391532572,
      "loss": 0.08190351128578185,
      "mean_token_accuracy": 0.9707993991672993,
      "num_tokens": 549230.0,
      "step": 80
    },
    {
      "entropy": 0.0770263057667762,
      "epoch": 0.288,
      "grad_norm": 0.171875,
      "learning_rate": 0.00016828825312940592,
      "loss": 0.06415510773658753,
      "mean_token_accuracy": 0.9708753030747175,
      "num_tokens": 616669.0,
      "step": 90
    },
    {
      "entropy": 0.06914123268215917,
      "epoch": 0.32,
      "grad_norm": 0.13671875,
      "learning_rate": 0.00016036076085226814,
      "loss": 0.07777998447418213,
      "mean_token_accuracy": 0.9653518848121166,
      "num_tokens": 686556.0,
      "step": 100
    },
    {
      "entropy": 0.06991175076691433,
      "epoch": 0.352,
      "grad_norm": 0.166015625,
      "learning_rate": 0.00015178496248983254,
      "loss": 0.07323018312454224,
      "mean_token_accuracy": 0.9576951999217271,
      "num_tokens": 759521.0,
      "step": 110
    },
    {
      "entropy": 0.09207692868367304,
      "epoch": 0.384,
      "grad_norm": 0.21484375,
      "learning_rate": 0.00014265296659384956,
      "loss": 0.09533499479293824,
      "mean_token_accuracy": 0.9764811851084232,
      "num_tokens": 826869.0,
      "step": 120
    },
    {
      "entropy": 0.08726739198318682,
      "epoch": 0.416,
      "grad_norm": 0.1904296875,
      "learning_rate": 0.00013306285556796495,
      "loss": 0.08955158591270447,
      "mean_token_accuracy": 0.956151831895113,
      "num_tokens": 898111.0,
      "step": 130
    },
    {
      "entropy": 0.06639394556113984,
      "epoch": 0.448,
      "grad_norm": 0.1533203125,
      "learning_rate": 0.000123117632211497,
      "loss": 0.06314496397972107,
      "mean_token_accuracy": 0.9615629483014345,
      "num_tokens": 969339.0,
      "step": 140
    },
    {
      "entropy": 0.0558678554254584,
      "epoch": 0.48,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0001129241134155949,
      "loss": 0.06386124491691589,
      "mean_token_accuracy": 0.9733367815613747,
      "num_tokens": 1036672.0,
      "step": 150
    },
    {
      "entropy": 0.060232574946712705,
      "epoch": 0.512,
      "grad_norm": 0.1455078125,
      "learning_rate": 0.00010259178289406011,
      "loss": 0.06906680464744568,
      "mean_token_accuracy": 0.9617645684629679,
      "num_tokens": 1108529.0,
      "step": 160
    },
    {
      "entropy": 0.06748775953310542,
      "epoch": 0.544,
      "grad_norm": 0.14453125,
      "learning_rate": 9.223161527109937e-05,
      "loss": 0.06768612861633301,
      "mean_token_accuracy": 0.9634172908961773,
      "num_tokens": 1184675.0,
      "step": 170
    },
    {
      "entropy": 0.06814730220357887,
      "epoch": 0.576,
      "grad_norm": 0.185546875,
      "learning_rate": 8.195488415592238e-05,
      "loss": 0.06328855752944947,
      "mean_token_accuracy": 0.9700407199561596,
      "num_tokens": 1256207.0,
      "step": 180
    },
    {
      "entropy": 0.07081062349316199,
      "epoch": 0.608,
      "grad_norm": 0.55859375,
      "learning_rate": 7.187196700608373e-05,
      "loss": 0.06718837618827819,
      "mean_token_accuracy": 0.9618636917322874,
      "num_tokens": 1328169.0,
      "step": 190
    },
    {
      "entropy": 0.06150082344538532,
      "epoch": 0.64,
      "grad_norm": 0.1044921875,
      "learning_rate": 6.209115961596208e-05,
      "loss": 0.05759283900260925,
      "mean_token_accuracy": 0.9679569214582443,
      "num_tokens": 1399956.0,
      "step": 200
    },
    {
      "entropy": 0.06286092348891544,
      "epoch": 0.672,
      "grad_norm": 0.197265625,
      "learning_rate": 5.271751296338823e-05,
      "loss": 0.06326322555541992,
      "mean_token_accuracy": 0.9615916445851326,
      "num_tokens": 1475214.0,
      "step": 210
    },
    {
      "entropy": 0.0519205622273148,
      "epoch": 0.704,
      "grad_norm": 0.169921875,
      "learning_rate": 4.385170490729712e-05,
      "loss": 0.05610005855560303,
      "mean_token_accuracy": 0.972290551662445,
      "num_tokens": 1546118.0,
      "step": 220
    },
    {
      "entropy": 0.058427285050856884,
      "epoch": 0.736,
      "grad_norm": 0.224609375,
      "learning_rate": 3.558895885496023e-05,
      "loss": 0.05968340039253235,
      "mean_token_accuracy": 0.9686542131006718,
      "num_tokens": 1620482.0,
      "step": 230
    },
    {
      "entropy": 0.06077302123012487,
      "epoch": 0.768,
      "grad_norm": 0.115234375,
      "learning_rate": 2.8018021012875994e-05,
      "loss": 0.05487210750579834,
      "mean_token_accuracy": 0.9765994187444449,
      "num_tokens": 1686508.0,
      "step": 240
    },
    {
      "entropy": 0.05762897890235763,
      "epoch": 0.8,
      "grad_norm": 0.1455078125,
      "learning_rate": 2.1220207206178688e-05,
      "loss": 0.06742877364158631,
      "mean_token_accuracy": 0.9620399903506041,
      "num_tokens": 1764710.0,
      "step": 250
    },
    {
      "entropy": 0.059268097168387615,
      "epoch": 0.832,
      "grad_norm": 0.2041015625,
      "learning_rate": 1.526852950422226e-05,
      "loss": 0.058321452140808104,
      "mean_token_accuracy": 0.9726373586803675,
      "num_tokens": 1832644.0,
      "step": 260
    },
    {
      "entropy": 0.05387882571376394,
      "epoch": 0.864,
      "grad_norm": 0.1796875,
      "learning_rate": 1.0226912032836611e-05,
      "loss": 0.04902423620223999,
      "mean_token_accuracy": 0.9753172263503075,
      "num_tokens": 1901629.0,
      "step": 270
    },
    {
      "entropy": 0.05652731657464756,
      "epoch": 0.896,
      "grad_norm": 0.322265625,
      "learning_rate": 6.149504395842087e-06,
      "loss": 0.0603355348110199,
      "mean_token_accuracy": 0.9647678934037686,
      "num_tokens": 1975257.0,
      "step": 280
    },
    {
      "entropy": 0.05044347028888296,
      "epoch": 0.928,
      "grad_norm": 0.1484375,
      "learning_rate": 3.0801000800333877e-06,
      "loss": 0.04434172809123993,
      "mean_token_accuracy": 0.970330499112606,
      "num_tokens": 2042770.0,
      "step": 290
    },
    {
      "entropy": 0.05634855758107733,
      "epoch": 0.96,
      "grad_norm": 0.123046875,
      "learning_rate": 1.0516660902673448e-06,
      "loss": 0.053922271728515624,
      "mean_token_accuracy": 0.9622644379734993,
      "num_tokens": 2111795.0,
      "step": 300
    },
    {
      "entropy": 0.06218618047278142,
      "epoch": 0.992,
      "grad_norm": 0.1806640625,
      "learning_rate": 8.598886661895788e-08,
      "loss": 0.06558623909950256,
      "mean_token_accuracy": 0.9811610858887434,
      "num_tokens": 2182727.0,
      "step": 310
    },
    {
      "epoch": 1.0,
      "eval_entropy": 0.10073611672315747,
      "eval_loss": 0.283896803855896,
      "eval_mean_token_accuracy": 0.931689467549324,
      "eval_num_tokens": 2202446.0,
      "eval_runtime": 135.118,
      "eval_samples_per_second": 3.7,
      "eval_steps_per_second": 3.7,
      "step": 313
    }
  ],
  "logging_steps": 10,
  "max_steps": 313,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.867113674568868e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}