{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.020249746878164023,
  "eval_steps": 5,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0013499831252109348,
      "eval_loss": 4.265491008758545,
      "eval_runtime": 285.8199,
      "eval_samples_per_second": 4.366,
      "eval_steps_per_second": 0.546,
      "step": 1
    },
    {
      "epoch": 0.004049949375632805,
      "grad_norm": 35.99859619140625,
      "learning_rate": 3e-05,
      "loss": 17.0135,
      "step": 3
    },
    {
      "epoch": 0.006749915626054675,
      "eval_loss": 3.8347327709198,
      "eval_runtime": 287.9169,
      "eval_samples_per_second": 4.335,
      "eval_steps_per_second": 0.542,
      "step": 5
    },
    {
      "epoch": 0.00809989875126561,
      "grad_norm": 14.753717422485352,
      "learning_rate": 6e-05,
      "loss": 16.8055,
      "step": 6
    },
    {
      "epoch": 0.012149848126898413,
      "grad_norm": 19.439973831176758,
      "learning_rate": 9e-05,
      "loss": 15.2762,
      "step": 9
    },
    {
      "epoch": 0.01349983125210935,
      "eval_loss": 2.515470027923584,
      "eval_runtime": 287.8629,
      "eval_samples_per_second": 4.335,
      "eval_steps_per_second": 0.542,
      "step": 10
    },
    {
      "epoch": 0.01619979750253122,
      "grad_norm": 12.475789070129395,
      "learning_rate": 9.938441702975689e-05,
      "loss": 10.8456,
      "step": 12
    },
    {
      "epoch": 0.020249746878164023,
      "grad_norm": 11.72419261932373,
      "learning_rate": 9.619397662556435e-05,
      "loss": 8.899,
      "step": 15
    },
    {
      "epoch": 0.020249746878164023,
      "eval_loss": 1.7485647201538086,
      "eval_runtime": 287.8964,
      "eval_samples_per_second": 4.335,
      "eval_steps_per_second": 0.542,
      "step": 15
    }
  ],
  "logging_steps": 3,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.925000544531251e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}