{
  "best_metric": 0.8371559633027523,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-15/checkpoint-527",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1054,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 20.894563674926758,
      "learning_rate": 0.00025961167449966676,
      "loss": 1.6678,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8371559633027523,
      "eval_loss": 1.635141134262085,
      "eval_runtime": 2.8174,
      "eval_samples_per_second": 309.502,
      "eval_steps_per_second": 2.485,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 7.817446708679199,
      "learning_rate": 0.00021634306208305565,
      "loss": 0.738,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8188073394495413,
      "eval_loss": 1.8634682893753052,
      "eval_runtime": 2.8123,
      "eval_samples_per_second": 310.062,
      "eval_steps_per_second": 2.489,
      "step": 1054
    }
  ],
  "logging_steps": 500,
  "max_steps": 3689,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 16141525436100.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.18003409042088625,
    "learning_rate": 0.0003028802869162779,
    "num_train_epochs": 7,
    "temperature": 27
  }
}