{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 258,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.3396112367510795,
      "epoch": 0.1171303074670571,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00019999210442038162,
      "loss": 1.5911,
      "mean_token_accuracy": 0.6371529325842857,
      "num_tokens": 81104.0,
      "step": 10
    },
    {
      "entropy": 1.1428312733769417,
      "epoch": 0.2342606149341142,
      "grad_norm": 0.578125,
      "learning_rate": 0.00019904614256966512,
      "loss": 1.1178,
      "mean_token_accuracy": 0.7238687098026275,
      "num_tokens": 162974.0,
      "step": 20
    },
    {
      "entropy": 0.8962877839803696,
      "epoch": 0.3513909224011713,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0001965381638833274,
      "loss": 0.9277,
      "mean_token_accuracy": 0.7669008579105139,
      "num_tokens": 244811.0,
      "step": 30
    },
    {
      "entropy": 0.9452402792870999,
      "epoch": 0.4685212298682284,
      "grad_norm": 0.421875,
      "learning_rate": 0.0001925077206834458,
      "loss": 0.9307,
      "mean_token_accuracy": 0.761869502440095,
      "num_tokens": 326663.0,
      "step": 40
    },
    {
      "entropy": 0.885918003693223,
      "epoch": 0.5856515373352855,
      "grad_norm": 0.400390625,
      "learning_rate": 0.0001870183754669526,
      "loss": 0.9136,
      "mean_token_accuracy": 0.7613564830273389,
      "num_tokens": 408546.0,
      "step": 50
    },
    {
      "entropy": 0.7956090953201056,
      "epoch": 0.7027818448023426,
      "grad_norm": 0.35546875,
      "learning_rate": 0.00018015669848708767,
      "loss": 0.7743,
      "mean_token_accuracy": 0.7908696968108415,
      "num_tokens": 490427.0,
      "step": 60
    },
    {
      "entropy": 0.7252638831734657,
      "epoch": 0.8199121522693997,
      "grad_norm": 0.44140625,
      "learning_rate": 0.0001720309024887907,
      "loss": 0.7452,
      "mean_token_accuracy": 0.7986918680369854,
      "num_tokens": 572268.0,
      "step": 70
    },
    {
      "entropy": 0.8875785615295172,
      "epoch": 0.9370424597364568,
      "grad_norm": 0.3359375,
      "learning_rate": 0.00016276913612907007,
      "loss": 0.8771,
      "mean_token_accuracy": 0.7722796119749546,
      "num_tokens": 654069.0,
      "step": 80
    },
    {
      "entropy": 0.7577637726068497,
      "epoch": 1.0468521229868228,
      "grad_norm": 0.318359375,
      "learning_rate": 0.0001525174629961296,
      "loss": 0.735,
      "mean_token_accuracy": 0.7980754868189494,
      "num_tokens": 730754.0,
      "step": 90
    },
    {
      "entropy": 0.8036635374650359,
      "epoch": 1.16398243045388,
      "grad_norm": 0.375,
      "learning_rate": 0.00014143755809932845,
      "loss": 0.8076,
      "mean_token_accuracy": 0.7897886544466018,
      "num_tokens": 812566.0,
      "step": 100
    },
    {
      "entropy": 0.8438609030097723,
      "epoch": 1.281112737920937,
      "grad_norm": 0.33203125,
      "learning_rate": 0.0001297041581577035,
      "loss": 0.8318,
      "mean_token_accuracy": 0.7828883446753025,
      "num_tokens": 893688.0,
      "step": 110
    },
    {
      "entropy": 0.7181106580421328,
      "epoch": 1.3982430453879942,
      "grad_norm": 0.404296875,
      "learning_rate": 0.00011750230589752762,
      "loss": 0.7207,
      "mean_token_accuracy": 0.8055280044674873,
      "num_tokens": 975564.0,
      "step": 120
    },
    {
      "entropy": 0.7229301311075688,
      "epoch": 1.5153733528550513,
      "grad_norm": 0.388671875,
      "learning_rate": 0.00010502443181797697,
      "loss": 0.7249,
      "mean_token_accuracy": 0.8056504145264626,
      "num_tokens": 1057403.0,
      "step": 130
    },
    {
      "entropy": 0.7823872465640307,
      "epoch": 1.6325036603221084,
      "grad_norm": 0.36328125,
      "learning_rate": 9.246731944720675e-05,
      "loss": 0.7791,
      "mean_token_accuracy": 0.7945993095636368,
      "num_tokens": 1139268.0,
      "step": 140
    },
    {
      "entropy": 0.6786214027553796,
      "epoch": 1.7496339677891655,
      "grad_norm": 0.39453125,
      "learning_rate": 8.002900194855932e-05,
      "loss": 0.6803,
      "mean_token_accuracy": 0.8125956118106842,
      "num_tokens": 1221128.0,
      "step": 150
    },
    {
      "entropy": 0.7029102340340614,
      "epoch": 1.8667642752562226,
      "grad_norm": 0.451171875,
      "learning_rate": 6.790563901927907e-05,
      "loss": 0.6941,
      "mean_token_accuracy": 0.8080258596688509,
      "num_tokens": 1302883.0,
      "step": 160
    },
    {
      "entropy": 0.8408648435026407,
      "epoch": 1.9838945827232797,
      "grad_norm": 0.421875,
      "learning_rate": 5.6288423334906735e-05,
      "loss": 0.843,
      "mean_token_accuracy": 0.7804605267941952,
      "num_tokens": 1384758.0,
      "step": 170
    },
    {
      "entropy": 0.7994472912947337,
      "epoch": 2.0937042459736457,
      "grad_norm": 0.3671875,
      "learning_rate": 4.5360565326573104e-05,
      "loss": 0.7804,
      "mean_token_accuracy": 0.7989432060718537,
      "num_tokens": 1461478.0,
      "step": 180
    },
    {
      "entropy": 0.7261181980371475,
      "epoch": 2.210834553440703,
      "grad_norm": 0.380859375,
      "learning_rate": 3.52944038430556e-05,
      "loss": 0.7032,
      "mean_token_accuracy": 0.8093580298125744,
      "num_tokens": 1543336.0,
      "step": 190
    },
    {
      "entropy": 0.6980411611497402,
      "epoch": 2.32796486090776,
      "grad_norm": 0.41015625,
      "learning_rate": 2.624868826418262e-05,
      "loss": 0.7014,
      "mean_token_accuracy": 0.8104535348713398,
      "num_tokens": 1625196.0,
      "step": 200
    },
    {
      "entropy": 0.6276400487869978,
      "epoch": 2.445095168374817,
      "grad_norm": 0.376953125,
      "learning_rate": 1.8366074928281607e-05,
      "loss": 0.63,
      "mean_token_accuracy": 0.8211923144757748,
      "num_tokens": 1707105.0,
      "step": 210
    },
    {
      "entropy": 0.7523287607356905,
      "epoch": 2.562225475841874,
      "grad_norm": 0.392578125,
      "learning_rate": 1.1770877356504683e-05,
      "loss": 0.7501,
      "mean_token_accuracy": 0.7998858217149973,
      "num_tokens": 1788910.0,
      "step": 220
    },
    {
      "entropy": 0.7270868588238955,
      "epoch": 2.679355783308931,
      "grad_norm": 0.3671875,
      "learning_rate": 6.5671057543387985e-06,
      "loss": 0.7306,
      "mean_token_accuracy": 0.8045763105154038,
      "num_tokens": 1869982.0,
      "step": 230
    },
    {
      "entropy": 0.64420210942626,
      "epoch": 2.7964860907759883,
      "grad_norm": 0.416015625,
      "learning_rate": 2.836826708532603e-06,
      "loss": 0.6385,
      "mean_token_accuracy": 0.8231038711965084,
      "num_tokens": 1951891.0,
      "step": 240
    },
    {
      "entropy": 0.7004209227859974,
      "epoch": 2.9136163982430454,
      "grad_norm": 0.41015625,
      "learning_rate": 6.388689479991605e-07,
      "loss": 0.7025,
      "mean_token_accuracy": 0.8102728992700576,
      "num_tokens": 2033696.0,
      "step": 250
    }
  ],
  "logging_steps": 10,
  "max_steps": 258,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5660925050880000.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}