checkpoint-10/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.3636363636363638,
  "eval_steps": 50,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.3636363636363638,
      "grad_norm": 14.496957778930664,
      "learning_rate": 4.7745751406263165e-06,
      "loss": 12.4921,
      "step": 10
    }
  ],
  "logging_steps": 10,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 50,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3413708441124864.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}