trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 1420,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.52112676056338,
      "grad_norm": 0.2243831753730774,
      "learning_rate": 1.9131455399061035e-05,
      "loss": 0.0913,
      "step": 500
    },
    {
      "epoch": 7.042253521126761,
      "grad_norm": 0.12650109827518463,
      "learning_rate": 1.3262910798122067e-05,
      "loss": 0.0111,
      "step": 1000
    }
  ],
  "logging_steps": 500,
  "max_steps": 2130,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7.940115037736141e+17,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}