trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.3986013986013985,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17,
      "learning_rate": 2.8251748251748252e-05,
      "loss": 1.7367,
      "step": 500
    },
    {
      "epoch": 0.35,
      "learning_rate": 2.6503496503496503e-05,
      "loss": 1.6337,
      "step": 1000
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.4755244755244757e-05,
      "loss": 1.5123,
      "step": 1500
    },
    {
      "epoch": 0.7,
      "learning_rate": 2.300699300699301e-05,
      "loss": 1.4435,
      "step": 2000
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.125874125874126e-05,
      "loss": 1.4016,
      "step": 2500
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.951048951048951e-05,
      "loss": 1.3019,
      "step": 3000
    },
    {
      "epoch": 1.22,
      "learning_rate": 1.7762237762237765e-05,
      "loss": 1.0332,
      "step": 3500
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.6013986013986016e-05,
      "loss": 1.0454,
      "step": 4000
    }
  ],
  "logging_steps": 500,
  "max_steps": 8580,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 7348665076063200.0,
  "trial_name": null,
  "trial_params": null
}