{
  "best_metric": 0.7622309197651663,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-44/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": 3.694380760192871,
      "learning_rate": 0.00012850625937427086,
      "loss": 0.5721,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7622309197651663,
      "eval_f1": 0.7993393889347646,
      "eval_loss": 0.48739945888519287,
      "eval_precision": 0.6914285714285714,
      "eval_recall": 0.9471624266144814,
      "eval_runtime": 30.106,
      "eval_samples_per_second": 33.947,
      "eval_steps_per_second": 1.063,
      "step": 96
    }
  ],
  "logging_steps": 500,
  "max_steps": 768,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 235695197280.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8889440199459292,
    "learning_rate": 0.00014686429642773812,
    "num_train_epochs": 8,
    "temperature": 4
  }
}