{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00029267002299148163,
  "eval_steps": 10,
  "global_step": 26,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.1256539345826217e-05,
      "eval_loss": NaN,
      "eval_runtime": 1424.1245,
      "eval_samples_per_second": 26.266,
      "eval_steps_per_second": 13.133,
      "step": 1
    },
    {
      "epoch": 5.628269672913108e-05,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.00011256539345826216,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.00011256539345826216,
      "eval_loss": NaN,
      "eval_runtime": 1430.726,
      "eval_samples_per_second": 26.145,
      "eval_steps_per_second": 13.072,
      "step": 10
    },
    {
      "epoch": 0.00016884809018739323,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.00022513078691652433,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.00022513078691652433,
      "eval_loss": NaN,
      "eval_runtime": 1432.0472,
      "eval_samples_per_second": 26.121,
      "eval_steps_per_second": 13.06,
      "step": 20
    },
    {
      "epoch": 0.0002814134836456554,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 843175472136192.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}