{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.433962264150944,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6451612903225806,
      "eval_loss": 0.2875964641571045,
      "eval_runtime": 1.4249,
      "eval_samples_per_second": 2175.625,
      "eval_steps_per_second": 45.618,
      "step": 318
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.685534591194969e-05,
      "loss": 0.4599,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8493548387096774,
      "eval_loss": 0.11622390896081924,
      "eval_runtime": 1.4485,
      "eval_samples_per_second": 2140.189,
      "eval_steps_per_second": 44.875,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8948387096774194,
      "eval_loss": 0.06665954738855362,
      "eval_runtime": 1.4577,
      "eval_samples_per_second": 2126.625,
      "eval_steps_per_second": 44.591,
      "step": 954
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.1359,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9083870967741936,
      "eval_loss": 0.047446902841329575,
      "eval_runtime": 1.46,
      "eval_samples_per_second": 2123.227,
      "eval_steps_per_second": 44.519,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "learning_rate": 1.0566037735849058e-05,
      "loss": 0.0755,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9206451612903226,
      "eval_loss": 0.038521502166986465,
      "eval_runtime": 1.4489,
      "eval_samples_per_second": 2139.512,
      "eval_steps_per_second": 44.861,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9267741935483871,
      "eval_loss": 0.033671800047159195,
      "eval_runtime": 1.4547,
      "eval_samples_per_second": 2130.965,
      "eval_steps_per_second": 44.682,
      "step": 1908
    },
    {
      "epoch": 6.29,
      "learning_rate": 7.421383647798742e-06,
      "loss": 0.0572,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.93,
      "eval_loss": 0.030534734949469566,
      "eval_runtime": 1.4524,
      "eval_samples_per_second": 2134.446,
      "eval_steps_per_second": 44.755,
      "step": 2226
    },
    {
      "epoch": 7.86,
      "learning_rate": 4.276729559748428e-06,
      "loss": 0.049,
      "step": 2500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9312903225806451,
      "eval_loss": 0.029032038524746895,
      "eval_runtime": 1.4509,
      "eval_samples_per_second": 2136.588,
      "eval_steps_per_second": 44.799,
      "step": 2544
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9341935483870968,
      "eval_loss": 0.027912627905607224,
      "eval_runtime": 1.456,
      "eval_samples_per_second": 2129.055,
      "eval_steps_per_second": 44.641,
      "step": 2862
    },
    {
      "epoch": 9.43,
      "learning_rate": 1.1320754716981133e-06,
      "loss": 0.0453,
      "step": 3000
    }
  ],
  "logging_steps": 500,
  "max_steps": 3180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 780331565067432.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6133894128091588,
    "num_train_epochs": 10,
    "temperature": 3
  }
}