{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.3911537128998726,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.92,
      "learning_rate": 0.0001934857749846208,
      "loss": 0.1563,
      "step": 500
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00017479180248950295,
      "loss": 0.0667,
      "step": 1000
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.00014635361737993667,
      "loss": 0.0389,
      "step": 1500
    },
    {
      "epoch": 3.7,
      "learning_rate": 0.00011187627439257638,
      "loss": 0.0217,
      "step": 2000
    },
    {
      "epoch": 4.62,
      "learning_rate": 7.58516369304635e-05,
      "loss": 0.012,
      "step": 2500
    },
    {
      "epoch": 5.54,
      "learning_rate": 4.2973156884111344e-05,
      "loss": 0.0058,
      "step": 3000
    },
    {
      "epoch": 6.47,
      "learning_rate": 1.752439059723171e-05,
      "loss": 0.0025,
      "step": 3500
    },
    {
      "epoch": 7.39,
      "learning_rate": 2.8209178689553083e-06,
      "loss": 0.0014,
      "step": 4000
    }
  ],
  "logging_steps": 500,
  "max_steps": 4328,
  "num_train_epochs": 8,
  "save_steps": 2000,
  "total_flos": 3.256255138554317e+18,
  "trial_name": null,
  "trial_params": null
}