{
  "best_metric": 0.7818003913894325,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-25/checkpoint-96",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 96,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.8755850791931152,
      "learning_rate": 0.00020053170483737637,
      "loss": 0.4801,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7818003913894325,
      "eval_f1": 0.8092386655260906,
      "eval_loss": 0.42955365777015686,
      "eval_precision": 0.7188449848024316,
      "eval_recall": 0.9256360078277887,
      "eval_runtime": 31.124,
      "eval_samples_per_second": 32.836,
      "eval_steps_per_second": 1.028,
      "step": 96
    }
  ],
  "logging_steps": 500,
  "max_steps": 960,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 235695197280.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6872599698926333,
    "learning_rate": 0.00022281300537486262,
    "num_train_epochs": 10,
    "temperature": 15
  }
}