{
  "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": 0.6222580645161291,
      "eval_loss": 0.23082642257213593,
      "eval_runtime": 2.0656,
      "eval_samples_per_second": 1500.808,
      "eval_steps_per_second": 31.469,
      "step": 318
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.6480781435966492,
      "learning_rate": 1.606918238993711e-05,
      "loss": 0.37,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8335483870967741,
      "eval_loss": 0.10731751471757889,
      "eval_runtime": 2.0611,
      "eval_samples_per_second": 1504.028,
      "eval_steps_per_second": 31.536,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8783870967741936,
      "eval_loss": 0.06844896823167801,
      "eval_runtime": 4.6848,
      "eval_samples_per_second": 661.707,
      "eval_steps_per_second": 13.875,
      "step": 954
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.5467702150344849,
      "learning_rate": 1.2138364779874214e-05,
      "loss": 0.1247,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9016129032258065,
      "eval_loss": 0.051504459232091904,
      "eval_runtime": 2.316,
      "eval_samples_per_second": 1338.506,
      "eval_steps_per_second": 28.065,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "grad_norm": 0.4089033603668213,
      "learning_rate": 8.207547169811321e-06,
      "loss": 0.0771,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 2544,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 402837063777120.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.3788177808142403,
    "num_train_epochs": 8,
    "temperature": 5
  }
}