{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 4855,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.51,
      "learning_rate": 4.9800000000000004e-05,
      "loss": 0.0343,
      "step": 500
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.840406074472585e-05,
      "loss": 0.0582,
      "step": 1000
    },
    {
      "epoch": 1.54,
      "learning_rate": 4.3796242912077306e-05,
      "loss": 0.0465,
      "step": 1500
    },
    {
      "epoch": 2.06,
      "learning_rate": 3.6769507651293344e-05,
      "loss": 0.0433,
      "step": 2000
    },
    {
      "epoch": 2.57,
      "learning_rate": 2.8228136337365285e-05,
      "loss": 0.0262,
      "step": 2500
    },
    {
      "epoch": 3.09,
      "learning_rate": 1.927133119001763e-05,
      "loss": 0.0202,
      "step": 3000
    },
    {
      "epoch": 3.6,
      "learning_rate": 1.1066727413011618e-05,
      "loss": 0.0139,
      "step": 3500
    },
    {
      "epoch": 4.12,
      "learning_rate": 4.637661900152143e-06,
      "loss": 0.0086,
      "step": 4000
    },
    {
      "epoch": 4.63,
      "learning_rate": 8.290570153929589e-07,
      "loss": 0.0065,
      "step": 4500
    }
  ],
  "logging_steps": 500,
  "max_steps": 4855,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 4.114702337052672e+16,
  "trial_name": null,
  "trial_params": null
}