{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 5e-05,
      "loss": 0.1012,
      "step": 500
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.2857142857142856e-05,
      "loss": 0.1334,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.571428571428572e-05,
      "loss": 0.0914,
      "step": 1500
    },
    {
      "epoch": 4.0,
      "learning_rate": 2.857142857142857e-05,
      "loss": 0.0573,
      "step": 2000
    },
    {
      "epoch": 5.0,
      "learning_rate": 2.1428571428571428e-05,
      "loss": 0.031,
      "step": 2500
    },
    {
      "epoch": 6.0,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.022,
      "step": 3000
    },
    {
      "epoch": 7.0,
      "learning_rate": 7.142857142857143e-06,
      "loss": 0.0115,
      "step": 3500
    },
    {
      "epoch": 8.0,
      "learning_rate": 0.0,
      "loss": 0.0081,
      "step": 4000
    }
  ],
  "logging_steps": 500,
  "max_steps": 4000,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 3.3679424618496e+16,
  "trial_name": null,
  "trial_params": null
}