{
  "best_global_step": 500,
  "best_metric": 1.6617584228515625,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 560,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.571428571428571,
      "grad_norm": 10.813451766967773,
      "learning_rate": 9.900000000000002e-06,
      "loss": 0.8466,
      "step": 100
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": 3.657083511352539,
      "learning_rate": 1.9900000000000003e-05,
      "loss": 0.2805,
      "step": 200
    },
    {
      "epoch": 10.714285714285714,
      "grad_norm": 11.072490692138672,
      "learning_rate": 2.9900000000000002e-05,
      "loss": 0.1182,
      "step": 300
    },
    {
      "epoch": 14.285714285714286,
      "grad_norm": 4.655518054962158,
      "learning_rate": 3.99e-05,
      "loss": 0.0968,
      "step": 400
    },
    {
      "epoch": 17.857142857142858,
      "grad_norm": 3.7923898696899414,
      "learning_rate": 4.99e-05,
      "loss": 0.0893,
      "step": 500
    },
    {
      "epoch": 17.857142857142858,
      "eval_loss": 1.6617584228515625,
      "eval_runtime": 0.2105,
      "eval_samples_per_second": 57.019,
      "eval_steps_per_second": 9.503,
      "step": 500
    }
  ],
  "logging_steps": 100,
  "max_steps": 560,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 147118903787520.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}