{
  "best_metric": 0.4980430528375734,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/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": 0.8380242586135864,
      "learning_rate": 8.6265645867868e-06,
      "loss": 0.6538,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4980430528375734,
      "eval_f1": 0.06215722120658134,
      "eval_loss": 0.6460775136947632,
      "eval_precision": 0.4722222222222222,
      "eval_recall": 0.033268101761252444,
      "eval_runtime": 32.7473,
      "eval_samples_per_second": 31.209,
      "eval_steps_per_second": 0.977,
      "step": 96
    }
  ],
  "logging_steps": 500,
  "max_steps": 576,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 235695197280.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9136100763812092,
    "learning_rate": 1.035187750414416e-05,
    "num_train_epochs": 6,
    "temperature": 5
  }
}