{
  "best_metric": 0.761252446183953,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-7/checkpoint-384",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 384,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.15159010887146,
      "learning_rate": 2.5124348703312046e-05,
      "loss": 0.6503,
      "step": 192
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6908023483365949,
      "eval_f1": 0.6623931623931624,
      "eval_loss": 0.6152858734130859,
      "eval_mcc": 0.38712657099283027,
      "eval_precision": 0.7294117647058823,
      "eval_recall": 0.6066536203522505,
      "eval_runtime": 66.1991,
      "eval_samples_per_second": 15.438,
      "eval_steps_per_second": 0.483,
      "step": 192
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.791621208190918,
      "learning_rate": 1.884326152748403e-05,
      "loss": 0.5656,
      "step": 384
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.761252446183953,
      "eval_f1": 0.7761467889908258,
      "eval_loss": 0.5115423202514648,
      "eval_mcc": 0.5271935842180233,
      "eval_precision": 0.7305699481865285,
      "eval_recall": 0.8277886497064579,
      "eval_runtime": 66.4101,
      "eval_samples_per_second": 15.389,
      "eval_steps_per_second": 0.482,
      "step": 384
    }
  ],
  "logging_steps": 500,
  "max_steps": 960,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9315469293191833,
    "learning_rate": 3.1405435879140055e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 16,
    "temperature": 39
  }
}