trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0362694300518134,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.26,
      "learning_rate": 4.22279792746114e-05,
      "loss": 1.0121,
      "step": 250
    },
    {
      "epoch": 0.52,
      "learning_rate": 3.963730569948187e-05,
      "loss": 0.986,
      "step": 500
    },
    {
      "epoch": 0.78,
      "learning_rate": 3.704663212435233e-05,
      "loss": 0.987,
      "step": 750
    },
    {
      "epoch": 1.04,
      "learning_rate": 3.44559585492228e-05,
      "loss": 0.9468,
      "step": 1000
    }
  ],
  "max_steps": 4825,
  "num_train_epochs": 5,
  "total_flos": 2104654034016000.0,
  "trial_name": null,
  "trial_params": null
}