{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1590,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 5.9957,
      "eval_samples_per_second": 517.04,
      "eval_steps_per_second": 10.841,
      "step": 318
    },
    {
      "epoch": 1.5723270440251573,
      "grad_norm": 81175.15625,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.0,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 6.2003,
      "eval_samples_per_second": 499.98,
      "eval_steps_per_second": 10.483,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 7.5831,
      "eval_samples_per_second": 408.805,
      "eval_steps_per_second": 8.572,
      "step": 954
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 281952.03125,
      "learning_rate": 7.421383647798742e-06,
      "loss": 0.0,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 6.527,
      "eval_samples_per_second": 474.951,
      "eval_steps_per_second": 9.959,
      "step": 1272
    },
    {
      "epoch": 4.716981132075472,
      "grad_norm": 445468.96875,
      "learning_rate": 1.1320754716981133e-06,
      "loss": 0.0,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 1590,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 391368939443328.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.3407610050374421,
    "num_train_epochs": 5,
    "temperature": 9
  }
}