{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.4705882352941178,
  "eval_steps": 500,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 3.6089425086975098,
      "epoch": 0.058823529411764705,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 6.292151927947998,
      "mean_token_accuracy": 0.0,
      "num_tokens": 15.0,
      "step": 1
    },
    {
      "entropy": 5.715427875518799,
      "epoch": 0.11764705882352941,
      "grad_norm": Infinity,
      "learning_rate": 0.0,
      "loss": 6.3516693115234375,
      "mean_token_accuracy": 0.0,
      "num_tokens": 22.0,
      "step": 2
    },
    {
      "entropy": 6.061905860900879,
      "epoch": 0.17647058823529413,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 4.695749282836914,
      "mean_token_accuracy": 0.3333333432674408,
      "num_tokens": 29.0,
      "step": 3
    },
    {
      "entropy": 5.138950824737549,
      "epoch": 0.23529411764705882,
      "grad_norm": 34.225440979003906,
      "learning_rate": 0.0,
      "loss": 4.924757957458496,
      "mean_token_accuracy": 0.375,
      "num_tokens": 41.0,
      "step": 4
    },
    {
      "entropy": 5.107641220092773,
      "epoch": 0.29411764705882354,
      "grad_norm": Infinity,
      "learning_rate": 4.000000000000001e-06,
      "loss": 4.769550800323486,
      "mean_token_accuracy": 0.0,
      "num_tokens": 47.0,
      "step": 5
    },
    {
      "entropy": 4.351223945617676,
      "epoch": 0.35294117647058826,
      "grad_norm": 26.4381046295166,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.740773916244507,
      "mean_token_accuracy": 0.4285714328289032,
      "num_tokens": 62.0,
      "step": 6
    },
    {
      "entropy": 4.605663299560547,
      "epoch": 0.4117647058823529,
      "grad_norm": 44.312320709228516,
      "learning_rate": 8.000000000000001e-06,
      "loss": 4.73244571685791,
      "mean_token_accuracy": 0.25,
      "num_tokens": 70.0,
      "step": 7
    },
    {
      "entropy": 4.652010440826416,
      "epoch": 0.47058823529411764,
      "grad_norm": 29.16073989868164,
      "learning_rate": 1.2e-05,
      "loss": 3.551349639892578,
      "mean_token_accuracy": 0.3333333432674408,
      "num_tokens": 83.0,
      "step": 8
    },
    {
      "entropy": 3.92993426322937,
      "epoch": 0.5294117647058824,
      "grad_norm": 24.797420501708984,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 3.086703300476074,
      "mean_token_accuracy": 0.5454545617103577,
      "num_tokens": 101.0,
      "step": 9
    },
    {
      "entropy": 4.687472343444824,
      "epoch": 0.5882352941176471,
      "grad_norm": 65.40557098388672,
      "learning_rate": 2e-05,
      "loss": 6.584832668304443,
      "mean_token_accuracy": 0.0,
      "num_tokens": 106.0,
      "step": 10
    },
    {
      "entropy": 5.460619926452637,
      "epoch": 0.6470588235294118,
      "grad_norm": 62.57985305786133,
      "learning_rate": 1.9975640502598243e-05,
      "loss": 5.282391548156738,
      "mean_token_accuracy": 0.3333333432674408,
      "num_tokens": 111.0,
      "step": 11
    },
    {
      "entropy": 5.458609580993652,
      "epoch": 0.7058823529411765,
      "grad_norm": 47.68376922607422,
      "learning_rate": 1.9902680687415704e-05,
      "loss": 3.4104318618774414,
      "mean_token_accuracy": 0.5,
      "num_tokens": 118.0,
      "step": 12
    },
    {
      "entropy": 5.297245979309082,
      "epoch": 0.7647058823529411,
      "grad_norm": 27.625783920288086,
      "learning_rate": 1.9781476007338058e-05,
      "loss": 2.5839743614196777,
      "mean_token_accuracy": 0.6000000238418579,
      "num_tokens": 126.0,
      "step": 13
    },
    {
      "entropy": 4.2264180183410645,
      "epoch": 0.8235294117647058,
      "grad_norm": 21.131492614746094,
      "learning_rate": 1.961261695938319e-05,
      "loss": 2.5414154529571533,
      "mean_token_accuracy": 0.5714285969734192,
      "num_tokens": 138.0,
      "step": 14
    },
    {
      "entropy": 4.7774658203125,
      "epoch": 0.8823529411764706,
      "grad_norm": 40.83763885498047,
      "learning_rate": 1.9396926207859085e-05,
      "loss": 2.8697211742401123,
      "mean_token_accuracy": 0.4000000059604645,
      "num_tokens": 145.0,
      "step": 15
    },
    {
      "entropy": 5.260831832885742,
      "epoch": 0.9411764705882353,
      "grad_norm": 40.107913970947266,
      "learning_rate": 1.913545457642601e-05,
      "loss": 2.11948561668396,
      "mean_token_accuracy": 0.3333333432674408,
      "num_tokens": 153.0,
      "step": 16
    },
    {
      "entropy": 5.279318332672119,
      "epoch": 1.0,
      "grad_norm": 34.499481201171875,
      "learning_rate": 1.8829475928589272e-05,
      "loss": 2.028179883956909,
      "mean_token_accuracy": 0.5,
      "num_tokens": 161.0,
      "step": 17
    },
    {
      "entropy": 3.6013762950897217,
      "epoch": 1.0588235294117647,
      "grad_norm": 45.223697662353516,
      "learning_rate": 1.848048096156426e-05,
      "loss": 4.955894470214844,
      "mean_token_accuracy": 0.3333333432674408,
      "num_tokens": 176.0,
      "step": 18
    },
    {
      "entropy": 4.529469966888428,
      "epoch": 1.1176470588235294,
      "grad_norm": 30.39960479736328,
      "learning_rate": 1.8090169943749477e-05,
      "loss": 1.4438493251800537,
      "mean_token_accuracy": 0.6000000238418579,
      "num_tokens": 184.0,
      "step": 19
    },
    {
      "entropy": 3.973909378051758,
      "epoch": 1.1764705882352942,
      "grad_norm": 18.265600204467773,
      "learning_rate": 1.766044443118978e-05,
      "loss": 1.7491637468338013,
      "mean_token_accuracy": 0.7142857313156128,
      "num_tokens": 196.0,
      "step": 20
    },
    {
      "entropy": 4.81941556930542,
      "epoch": 1.2352941176470589,
      "grad_norm": 37.519798278808594,
      "learning_rate": 1.7193398003386514e-05,
      "loss": 3.381870985031128,
      "mean_token_accuracy": 0.3333333432674408,
      "num_tokens": 202.0,
      "step": 21
    },
    {
      "entropy": 4.61383056640625,
      "epoch": 1.2941176470588236,
      "grad_norm": 24.117401123046875,
      "learning_rate": 1.6691306063588583e-05,
      "loss": 0.9651630520820618,
      "mean_token_accuracy": 0.75,
      "num_tokens": 210.0,
      "step": 22
    },
    {
      "entropy": 4.4701128005981445,
      "epoch": 1.3529411764705883,
      "grad_norm": 28.767288208007812,
      "learning_rate": 1.6156614753256583e-05,
      "loss": 0.8833061456680298,
      "mean_token_accuracy": 1.0,
      "num_tokens": 217.0,
      "step": 23
    },
    {
      "entropy": 4.853976726531982,
      "epoch": 1.4117647058823528,
      "grad_norm": 41.10981369018555,
      "learning_rate": 1.5591929034707468e-05,
      "loss": 3.586022138595581,
      "mean_token_accuracy": 0.3333333432674408,
      "num_tokens": 222.0,
      "step": 24
    },
    {
      "entropy": 4.533289909362793,
      "epoch": 1.4705882352941178,
      "grad_norm": Infinity,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 2.87251877784729,
      "mean_token_accuracy": 0.5,
      "num_tokens": 229.0,
      "step": 25
    }
  ],
  "logging_steps": 1,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 145923548544.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}