{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.011131725417439703,
  "eval_steps": 3,
  "global_step": 3,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0037105751391465678,
      "grad_norm": 16.39884376525879,
      "learning_rate": 2e-05,
      "loss": 9.1969,
      "step": 1
    },
    {
      "epoch": 0.0037105751391465678,
      "eval_loss": 9.007013320922852,
      "eval_runtime": 26.0552,
      "eval_samples_per_second": 4.375,
      "eval_steps_per_second": 2.188,
      "step": 1
    },
    {
      "epoch": 0.0074211502782931356,
      "grad_norm": 18.28474235534668,
      "learning_rate": 4e-05,
      "loss": 8.8405,
      "step": 2
    },
    {
      "epoch": 0.011131725417439703,
      "grad_norm": 16.06081199645996,
      "learning_rate": 6e-05,
      "loss": 8.7897,
      "step": 3
    },
    {
      "epoch": 0.011131725417439703,
      "eval_loss": 8.362504005432129,
      "eval_runtime": 26.6541,
      "eval_samples_per_second": 4.277,
      "eval_steps_per_second": 2.139,
      "step": 3
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4451323701362688.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}