{
  "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": 0.5829032258064516,
      "eval_loss": 0.2045331597328186,
      "eval_runtime": 5.9047,
      "eval_samples_per_second": 525.004,
      "eval_steps_per_second": 11.008,
      "step": 318
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.550763701707098e-05,
      "loss": 0.326,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8151612903225807,
      "eval_loss": 0.10090682655572891,
      "eval_runtime": 6.2986,
      "eval_samples_per_second": 492.175,
      "eval_steps_per_second": 10.32,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8729032258064516,
      "eval_loss": 0.06894117593765259,
      "eval_runtime": 5.942,
      "eval_samples_per_second": 521.707,
      "eval_steps_per_second": 10.939,
      "step": 954
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.101527403414196e-05,
      "loss": 0.1165,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8941935483870967,
      "eval_loss": 0.05427885055541992,
      "eval_runtime": 6.0236,
      "eval_samples_per_second": 514.64,
      "eval_steps_per_second": 10.791,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "learning_rate": 6.522911051212939e-06,
      "loss": 0.0768,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 2226,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 391111083046572.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.5943120895153833,
    "num_train_epochs": 7,
    "temperature": 10
  }
}