{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 27,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11347517730496454,
      "grad_norm": 0.7319265875933817,
      "learning_rate": 0.0001,
      "loss": 0.1236,
      "mean_token_accuracy": 0.9668329227715731,
      "num_tokens": 1033040.0,
      "step": 1
    },
    {
      "epoch": 1.1134751773049645,
      "grad_norm": 0.04641399621415606,
      "learning_rate": 0.0001,
      "loss": 0.0812,
      "mean_token_accuracy": 0.9677546548505201,
      "num_tokens": 10055949.0,
      "step": 10
    },
    {
      "epoch": 2.226950354609929,
      "grad_norm": 0.059588094492247355,
      "learning_rate": 0.0001,
      "loss": 0.0518,
      "mean_token_accuracy": 0.9785908912397494,
      "num_tokens": 20117738.0,
      "step": 20
    }
  ],
  "logging_steps": 10,
  "max_steps": 36,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 63842852274176.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}