{
  "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": 0.5806451612903226,
      "eval_loss": 0.19493919610977173,
      "eval_runtime": 1.9862,
      "eval_samples_per_second": 1560.769,
      "eval_steps_per_second": 32.726,
      "step": 318
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.613143265247345,
      "learning_rate": 1.550763701707098e-05,
      "loss": 0.312,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8112903225806452,
      "eval_loss": 0.0981183648109436,
      "eval_runtime": 2.3793,
      "eval_samples_per_second": 1302.89,
      "eval_steps_per_second": 27.319,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.87,
      "eval_loss": 0.06753243505954742,
      "eval_runtime": 3.757,
      "eval_samples_per_second": 825.133,
      "eval_steps_per_second": 17.301,
      "step": 954
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.5029990077018738,
      "learning_rate": 1.101527403414196e-05,
      "loss": 0.1131,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8935483870967742,
      "eval_loss": 0.0535094179213047,
      "eval_runtime": 2.65,
      "eval_samples_per_second": 1169.792,
      "eval_steps_per_second": 24.528,
      "step": 1272
    },
    {
      "epoch": 4.72,
      "grad_norm": 0.37056881189346313,
      "learning_rate": 6.522911051212939e-06,
      "loss": 0.0753,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 2226,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 391368939443328.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.5340221221277321,
    "num_train_epochs": 7,
    "temperature": 16
  }
}