{
  "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": 0.5883870967741935,
      "eval_loss": 0.2186722308397293,
      "eval_runtime": 5.9856,
      "eval_samples_per_second": 517.908,
      "eval_steps_per_second": 10.859,
      "step": 318
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.4758909853249476e-05,
      "loss": 0.3455,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8174193548387096,
      "eval_loss": 0.10785821825265884,
      "eval_runtime": 6.0703,
      "eval_samples_per_second": 510.681,
      "eval_steps_per_second": 10.708,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8719354838709678,
      "eval_loss": 0.07385361194610596,
      "eval_runtime": 5.9422,
      "eval_samples_per_second": 521.697,
      "eval_steps_per_second": 10.939,
      "step": 954
    },
    {
      "epoch": 3.14,
      "learning_rate": 9.517819706498952e-06,
      "loss": 0.1246,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8896774193548387,
      "eval_loss": 0.05901511758565903,
      "eval_runtime": 6.0232,
      "eval_samples_per_second": 514.677,
      "eval_steps_per_second": 10.792,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "learning_rate": 4.276729559748428e-06,
      "loss": 0.0836,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 1908,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 391111083046572.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.16451015660323232,
    "num_train_epochs": 6,
    "temperature": 7
  }
}