{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 96,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.4163570161908865,
      "epoch": 0.32,
      "grad_norm": 0.4453125,
      "learning_rate": 6e-05,
      "loss": 0.427,
      "mean_token_accuracy": 0.879430927336216,
      "num_tokens": 459017.0,
      "step": 10
    },
    {
      "entropy": 0.3372706573456526,
      "epoch": 0.64,
      "grad_norm": 0.447265625,
      "learning_rate": 6e-05,
      "loss": 0.3233,
      "mean_token_accuracy": 0.9003406524658203,
      "num_tokens": 912954.0,
      "step": 20
    },
    {
      "entropy": 0.32121987249702216,
      "epoch": 0.96,
      "grad_norm": 0.380859375,
      "learning_rate": 6e-05,
      "loss": 0.3098,
      "mean_token_accuracy": 0.9014183498919011,
      "num_tokens": 1368462.0,
      "step": 30
    },
    {
      "entropy": 0.27086804344041926,
      "epoch": 1.256,
      "grad_norm": 0.34375,
      "learning_rate": 6e-05,
      "loss": 0.2439,
      "mean_token_accuracy": 0.9211577402578818,
      "num_tokens": 1793628.0,
      "step": 40
    },
    {
      "entropy": 0.2517373651266098,
      "epoch": 1.576,
      "grad_norm": 0.302734375,
      "learning_rate": 6e-05,
      "loss": 0.2284,
      "mean_token_accuracy": 0.9240875385701657,
      "num_tokens": 2249448.0,
      "step": 50
    },
    {
      "entropy": 0.2411576334387064,
      "epoch": 1.896,
      "grad_norm": 0.310546875,
      "learning_rate": 6e-05,
      "loss": 0.2257,
      "mean_token_accuracy": 0.9256131462752819,
      "num_tokens": 2701085.0,
      "step": 60
    },
    {
      "entropy": 0.21875436301972415,
      "epoch": 2.192,
      "grad_norm": 0.310546875,
      "learning_rate": 6e-05,
      "loss": 0.1978,
      "mean_token_accuracy": 0.9344683112324895,
      "num_tokens": 3115785.0,
      "step": 70
    },
    {
      "entropy": 0.1890676412731409,
      "epoch": 2.512,
      "grad_norm": 0.271484375,
      "learning_rate": 6e-05,
      "loss": 0.1723,
      "mean_token_accuracy": 0.9410219177603721,
      "num_tokens": 3578352.0,
      "step": 80
    },
    {
      "entropy": 0.17951820120215417,
      "epoch": 2.832,
      "grad_norm": 0.291015625,
      "learning_rate": 6e-05,
      "loss": 0.166,
      "mean_token_accuracy": 0.9436265029013157,
      "num_tokens": 4034155.0,
      "step": 90
    }
  ],
  "logging_steps": 10,
  "max_steps": 96,
  "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": 3.961197133037568e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}