{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.091383949706429,
      "epoch": 2.0,
      "grad_norm": 0.143918976187706,
      "learning_rate": 6.909830056250527e-05,
      "logits/chosen": -2.6255019325219764,
      "logits/rejected": -2.659680873301937,
      "logps/chosen": -208.22477260388825,
      "logps/rejected": -238.7575555098684,
      "loss": 0.45360164642333983,
      "mean_token_accuracy": 0.47160694003105164,
      "num_tokens": 34910.0,
      "rewards/accuracies": 0.7302631578947368,
      "rewards/chosen": 0.3593159549391681,
      "rewards/margins": 0.8960761683747956,
      "rewards/rejected": -0.5367602110399228,
      "step": 10
    }
  ],
  "logging_steps": 10,
  "max_steps": 15,
  "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": 1447691195695104.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}