{
  "best_metric": 0.8256880733944955,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-18/checkpoint-1054",
  "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,
      "learning_rate": 0.00014893639687145338,
      "loss": 0.8667,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8165137614678899,
      "eval_loss": 0.8296506404876709,
      "eval_runtime": 6.9326,
      "eval_samples_per_second": 125.782,
      "eval_steps_per_second": 1.01,
      "step": 527
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00013031934726252174,
      "loss": 0.4473,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8256880733944955,
      "eval_loss": 0.9700397849082947,
      "eval_runtime": 7.0917,
      "eval_samples_per_second": 122.961,
      "eval_steps_per_second": 0.987,
      "step": 1054
    }
  ],
  "logging_steps": 500,
  "max_steps": 4743,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 16162868129520.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7585638468345914,
    "learning_rate": 0.00016755344648038508,
    "num_train_epochs": 9,
    "temperature": 8
  }
}