{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.761904761904762,
  "eval_steps": 500,
  "global_step": 8,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 10.363198280334473,
      "learning_rate": 8.75e-05,
      "loss": 9.6121,
      "step": 2
    },
    {
      "epoch": 1.7619047619047619,
      "grad_norm": 11.121501922607422,
      "learning_rate": 6.25e-05,
      "loss": 10.106,
      "step": 4
    },
    {
      "epoch": 2.761904761904762,
      "grad_norm": 13.02573299407959,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 9.8856,
      "step": 6
    },
    {
      "epoch": 3.761904761904762,
      "grad_norm": 14.771383285522461,
      "learning_rate": 1.25e-05,
      "loss": 9.7431,
      "step": 8
    },
    {
      "epoch": 3.761904761904762,
      "step": 8,
      "total_flos": 41961945956328.0,
      "train_loss": 9.836710214614868,
      "train_runtime": 229.2884,
      "train_samples_per_second": 0.715,
      "train_steps_per_second": 0.035
    }
  ],
  "logging_steps": 2,
  "max_steps": 8,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 41961945956328.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}