{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 19.08411214953271,
  "eval_steps": 500,
  "global_step": 180,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.03,
      "learning_rate": 1.9626168224299065e-05,
      "loss": 2.7578,
      "step": 21
    },
    {
      "epoch": 4.06,
      "learning_rate": 3.925233644859813e-05,
      "loss": 2.029,
      "step": 42
    },
    {
      "epoch": 6.08,
      "learning_rate": 5.8878504672897196e-05,
      "loss": 1.3632,
      "step": 63
    },
    {
      "epoch": 9.03,
      "learning_rate": 7.850467289719626e-05,
      "loss": 1.1552,
      "step": 84
    },
    {
      "epoch": 11.06,
      "learning_rate": 9.813084112149533e-05,
      "loss": 0.954,
      "step": 105
    },
    {
      "epoch": 13.08,
      "learning_rate": 0.00011775700934579439,
      "loss": 0.6208,
      "step": 126
    },
    {
      "epoch": 16.03,
      "learning_rate": 0.00013738317757009346,
      "loss": 0.301,
      "step": 147
    },
    {
      "epoch": 18.06,
      "learning_rate": 0.00015700934579439252,
      "loss": 0.1942,
      "step": 168
    }
  ],
  "logging_steps": 21,
  "max_steps": 2140,
  "num_train_epochs": 20,
  "save_steps": 500,
  "total_flos": 3.41435505180672e+16,
  "trial_name": null,
  "trial_params": null
}