{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 981,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.6546974384784698,
      "epoch": 0.3058103975535168,
      "grad_norm": 0.37161776423454285,
      "learning_rate": 9.931806517013612e-06,
      "loss": 2.1612,
      "mean_token_accuracy": 0.5265661662817002,
      "num_tokens": 41805.0,
      "step": 100
    },
    {
      "entropy": 0.7053046183288098,
      "epoch": 0.6116207951070336,
      "grad_norm": 0.14000245928764343,
      "learning_rate": 9.381208231670505e-06,
      "loss": 0.6959,
      "mean_token_accuracy": 0.8424646317958832,
      "num_tokens": 84002.0,
      "step": 200
    },
    {
      "entropy": 0.27937801614403723,
      "epoch": 0.9174311926605505,
      "grad_norm": 0.1330832690000534,
      "learning_rate": 8.336447938411616e-06,
      "loss": 0.2956,
      "mean_token_accuracy": 0.9407258796691894,
      "num_tokens": 125931.0,
      "step": 300
    },
    {
      "entropy": 0.2766046752035618,
      "epoch": 1.2232415902140672,
      "grad_norm": 0.25764134526252747,
      "learning_rate": 6.915365469768857e-06,
      "loss": 0.2872,
      "mean_token_accuracy": 0.9397406846284866,
      "num_tokens": 168641.0,
      "step": 400
    },
    {
      "entropy": 0.24791674077510834,
      "epoch": 1.529051987767584,
      "grad_norm": 0.17321506142616272,
      "learning_rate": 5.2782465115587796e-06,
      "loss": 0.2481,
      "mean_token_accuracy": 0.9486746054887771,
      "num_tokens": 210299.0,
      "step": 500
    },
    {
      "entropy": 0.24874767586588858,
      "epoch": 1.834862385321101,
      "grad_norm": 0.19254006445407867,
      "learning_rate": 3.609743778011684e-06,
      "loss": 0.2462,
      "mean_token_accuracy": 0.9466655850410461,
      "num_tokens": 252718.0,
      "step": 600
    },
    {
      "entropy": 0.23701502427458762,
      "epoch": 2.140672782874618,
      "grad_norm": 0.1829081028699875,
      "learning_rate": 2.098049799001508e-06,
      "loss": 0.2318,
      "mean_token_accuracy": 0.9517926776409149,
      "num_tokens": 294521.0,
      "step": 700
    },
    {
      "entropy": 0.249193025380373,
      "epoch": 2.4464831804281344,
      "grad_norm": 0.19727888703346252,
      "learning_rate": 9.136704470001101e-07,
      "loss": 0.2398,
      "mean_token_accuracy": 0.9476780182123185,
      "num_tokens": 337195.0,
      "step": 800
    },
    {
      "entropy": 0.22941243454813956,
      "epoch": 2.7522935779816513,
      "grad_norm": 0.21921473741531372,
      "learning_rate": 1.9019336445009918e-07,
      "loss": 0.222,
      "mean_token_accuracy": 0.9518460136651993,
      "num_tokens": 378540.0,
      "step": 900
    }
  ],
  "logging_steps": 100,
  "max_steps": 981,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8985572221907456e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}