{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.716981132075472,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5670967741935484,
      "eval_loss": 0.199788898229599,
      "eval_runtime": 1.8441,
      "eval_samples_per_second": 1681.04,
      "eval_steps_per_second": 35.248,
      "step": 318
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.3148,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7967741935483871,
      "eval_loss": 0.10518480092287064,
      "eval_runtime": 1.8481,
      "eval_samples_per_second": 1677.393,
      "eval_steps_per_second": 35.171,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8541935483870968,
      "eval_loss": 0.07700461894273758,
      "eval_runtime": 1.8502,
      "eval_samples_per_second": 1675.498,
      "eval_steps_per_second": 35.131,
      "step": 954
    },
    {
      "epoch": 3.14,
      "learning_rate": 7.421383647798742e-06,
      "loss": 0.1209,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8767741935483871,
      "eval_loss": 0.06455741822719574,
      "eval_runtime": 2.3247,
      "eval_samples_per_second": 1333.502,
      "eval_steps_per_second": 27.961,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "learning_rate": 1.1320754716981133e-06,
      "loss": 0.0873,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 1590,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 391111083046572.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6620514719547677,
    "num_train_epochs": 5,
    "temperature": 17
  }
}