{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00043609405095032164,
  "eval_steps": 3,
  "global_step": 3,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00014536468365010721,
      "grad_norm": 12.12260913848877,
      "learning_rate": 2e-05,
      "loss": 5.4024,
      "step": 1
    },
    {
      "epoch": 0.00014536468365010721,
      "eval_loss": 4.953571796417236,
      "eval_runtime": 572.8822,
      "eval_samples_per_second": 5.057,
      "eval_steps_per_second": 2.529,
      "step": 1
    },
    {
      "epoch": 0.00029072936730021443,
      "grad_norm": 11.454992294311523,
      "learning_rate": 4e-05,
      "loss": 4.8248,
      "step": 2
    },
    {
      "epoch": 0.00043609405095032164,
      "grad_norm": 12.662338256835938,
      "learning_rate": 6e-05,
      "loss": 3.6741,
      "step": 3
    },
    {
      "epoch": 0.00043609405095032164,
      "eval_loss": 4.851102828979492,
      "eval_runtime": 557.8236,
      "eval_samples_per_second": 5.193,
      "eval_steps_per_second": 2.598,
      "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": 186369435500544.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}