{
  "best_metric": 0.9162844036697247,
  "best_model_checkpoint": "electra-distilled-sst\\run-37\\checkpoint-1581",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1581,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 29.725248336791992,
      "learning_rate": 5.633928647343514e-05,
      "loss": 1.3149,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9094036697247706,
      "eval_loss": 0.8187136054039001,
      "eval_runtime": 1.4446,
      "eval_samples_per_second": 603.645,
      "eval_steps_per_second": 4.846,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 28.089006423950195,
      "learning_rate": 4.226114329112109e-05,
      "loss": 0.6381,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9013761467889908,
      "eval_loss": 0.8704111576080322,
      "eval_runtime": 1.4438,
      "eval_samples_per_second": 603.964,
      "eval_steps_per_second": 4.848,
      "step": 1054
    },
    {
      "epoch": 3.0,
      "grad_norm": 12.265154838562012,
      "learning_rate": 2.818300010880705e-05,
      "loss": 0.4568,
      "step": 1581
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9162844036697247,
      "eval_loss": 0.8463584184646606,
      "eval_runtime": 1.4494,
      "eval_samples_per_second": 601.628,
      "eval_steps_per_second": 4.83,
      "step": 1581
    }
  ],
  "logging_steps": 500,
  "max_steps": 2635,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 561038813831964.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6809679700189717,
    "learning_rate": 7.039071591157021e-05,
    "num_train_epochs": 5,
    "temperature": 15
  }
}