{
  "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": 0.6012903225806452,
      "eval_loss": 0.20320981740951538,
      "eval_runtime": 3.0228,
      "eval_samples_per_second": 1025.524,
      "eval_steps_per_second": 21.503,
      "step": 318
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.6164258122444153,
      "learning_rate": 1.650593990216632e-05,
      "loss": 0.3277,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8293548387096774,
      "eval_loss": 0.09816579520702362,
      "eval_runtime": 2.0519,
      "eval_samples_per_second": 1510.811,
      "eval_steps_per_second": 31.678,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8764516129032258,
      "eval_loss": 0.06465902924537659,
      "eval_runtime": 2.065,
      "eval_samples_per_second": 1501.211,
      "eval_steps_per_second": 31.477,
      "step": 954
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.5033944249153137,
      "learning_rate": 1.3011879804332637e-05,
      "loss": 0.1135,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9016129032258065,
      "eval_loss": 0.04934108257293701,
      "eval_runtime": 2.0912,
      "eval_samples_per_second": 1482.422,
      "eval_steps_per_second": 31.083,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "grad_norm": 0.37088626623153687,
      "learning_rate": 9.517819706498952e-06,
      "loss": 0.0718,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 2862,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 402837063777120.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9559918266969052,
    "num_train_epochs": 9,
    "temperature": 9
  }
}