trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.16632016632016633,
  "eval_steps": 100,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08316008316008316,
      "grad_norm": 5.463322309276886,
      "learning_rate": 1.1736481776669307e-05,
      "loss": 1.2877,
      "mean_token_accuracy": 0.6799418583512307,
      "step": 5
    },
    {
      "epoch": 0.16632016632016633,
      "grad_norm": 5.392606351436004,
      "learning_rate": 0.0,
      "loss": 1.279,
      "mean_token_accuracy": 0.6758172824978829,
      "step": 10
    },
    {
      "epoch": 0.16632016632016633,
      "step": 10,
      "total_flos": 34729623552.0,
      "train_loss": 1.2833720207214356,
      "train_runtime": 27.0374,
      "train_samples_per_second": 2.959,
      "train_steps_per_second": 0.37
    }
  ],
  "logging_steps": 5,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 34729623552.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}