{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.782525118099818e-05,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 2.555440902709961,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 3.6397,
      "step": 20
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.218903064727783,
      "learning_rate": 6.666666666666667e-06,
      "loss": 3.6917,
      "step": 40
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.9262466430664062,
      "learning_rate": 1e-05,
      "loss": 3.5828,
      "step": 60
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.782036542892456,
      "learning_rate": 1.3166666666666665e-05,
      "loss": 3.5865,
      "step": 80
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.9482054710388184,
      "learning_rate": 1.65e-05,
      "loss": 3.3337,
      "step": 100
    }
  ],
  "logging_steps": 20,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 20,
  "total_flos": 543796744273920.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}