{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 4840,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.0661157024793386,
      "grad_norm": 4.206299781799316,
      "learning_rate": 4.4834710743801654e-05,
      "loss": 4.2802,
      "step": 500
    },
    {
      "epoch": 4.132231404958677,
      "grad_norm": 4.235241889953613,
      "learning_rate": 3.9669421487603306e-05,
      "loss": 3.9317,
      "step": 1000
    },
    {
      "epoch": 6.198347107438017,
      "grad_norm": 4.2089643478393555,
      "learning_rate": 3.4504132231404964e-05,
      "loss": 3.7086,
      "step": 1500
    },
    {
      "epoch": 8.264462809917354,
      "grad_norm": 3.783989191055298,
      "learning_rate": 2.9338842975206616e-05,
      "loss": 3.5136,
      "step": 2000
    },
    {
      "epoch": 10.330578512396695,
      "grad_norm": 3.957587242126465,
      "learning_rate": 2.4173553719008264e-05,
      "loss": 3.3546,
      "step": 2500
    },
    {
      "epoch": 12.396694214876034,
      "grad_norm": 4.028080463409424,
      "learning_rate": 1.900826446280992e-05,
      "loss": 3.2309,
      "step": 3000
    },
    {
      "epoch": 14.462809917355372,
      "grad_norm": 4.280130863189697,
      "learning_rate": 1.3842975206611573e-05,
      "loss": 3.1278,
      "step": 3500
    },
    {
      "epoch": 16.52892561983471,
      "grad_norm": 3.9069178104400635,
      "learning_rate": 8.677685950413224e-06,
      "loss": 3.0433,
      "step": 4000
    },
    {
      "epoch": 18.59504132231405,
      "grad_norm": 4.312071800231934,
      "learning_rate": 3.5123966942148763e-06,
      "loss": 3.0038,
      "step": 4500
    },
    {
      "epoch": 20.0,
      "step": 4840,
      "total_flos": 5058613739520000.0,
      "train_loss": 3.4317847007562308,
      "train_runtime": 1978.8511,
      "train_samples_per_second": 4.892,
      "train_steps_per_second": 2.446
    }
  ],
  "logging_steps": 500,
  "max_steps": 4840,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 5058613739520000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}