{
  "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": 0.5832258064516129,
      "eval_loss": 0.19133226573467255,
      "eval_runtime": 1.4535,
      "eval_samples_per_second": 2132.847,
      "eval_steps_per_second": 44.721,
      "step": 318
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.685534591194969e-05,
      "loss": 0.309,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.827741935483871,
      "eval_loss": 0.09374182671308517,
      "eval_runtime": 1.4509,
      "eval_samples_per_second": 2136.594,
      "eval_steps_per_second": 44.8,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8790322580645161,
      "eval_loss": 0.06333240121603012,
      "eval_runtime": 1.4496,
      "eval_samples_per_second": 2138.579,
      "eval_steps_per_second": 44.841,
      "step": 954
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.1078,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.896774193548387,
      "eval_loss": 0.04806679114699364,
      "eval_runtime": 1.4589,
      "eval_samples_per_second": 2124.857,
      "eval_steps_per_second": 44.553,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "learning_rate": 1.0566037735849058e-05,
      "loss": 0.0686,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9112903225806451,
      "eval_loss": 0.0393882654607296,
      "eval_runtime": 1.4641,
      "eval_samples_per_second": 2117.338,
      "eval_steps_per_second": 44.396,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9167741935483871,
      "eval_loss": 0.03456535562872887,
      "eval_runtime": 1.4626,
      "eval_samples_per_second": 2119.465,
      "eval_steps_per_second": 44.44,
      "step": 1908
    },
    {
      "epoch": 6.29,
      "learning_rate": 7.421383647798742e-06,
      "loss": 0.0538,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9216129032258065,
      "eval_loss": 0.030887985602021217,
      "eval_runtime": 1.4707,
      "eval_samples_per_second": 2107.905,
      "eval_steps_per_second": 44.198,
      "step": 2226
    },
    {
      "epoch": 7.86,
      "learning_rate": 4.276729559748428e-06,
      "loss": 0.0464,
      "step": 2500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9270967741935484,
      "eval_loss": 0.029196659103035927,
      "eval_runtime": 1.4685,
      "eval_samples_per_second": 2111.062,
      "eval_steps_per_second": 44.264,
      "step": 2544
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9309677419354838,
      "eval_loss": 0.028075959533452988,
      "eval_runtime": 1.4722,
      "eval_samples_per_second": 2105.684,
      "eval_steps_per_second": 44.151,
      "step": 2862
    },
    {
      "epoch": 9.43,
      "learning_rate": 1.1320754716981133e-06,
      "loss": 0.0428,
      "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": 0.3001695987245524,
    "num_train_epochs": 10,
    "temperature": 17
  }
}