{
  "best_metric": 0.6674509803921569,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-5/checkpoint-320",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 640,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9986019134521484,
      "learning_rate": 6.067732661916699e-05,
      "loss": 0.3474,
      "step": 160
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1": 0.0,
      "eval_loss": 0.33498698472976685,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.8603,
      "eval_samples_per_second": 685.372,
      "eval_steps_per_second": 21.502,
      "step": 160
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9157262444496155,
      "learning_rate": 4.045155107944466e-05,
      "loss": 0.3312,
      "step": 320
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6674509803921569,
      "eval_f1": 0.004694835680751173,
      "eval_loss": 0.3251466453075409,
      "eval_mcc": 0.03962144258751637,
      "eval_precision": 1.0,
      "eval_recall": 0.002352941176470588,
      "eval_runtime": 1.8638,
      "eval_samples_per_second": 684.076,
      "eval_steps_per_second": 21.461,
      "step": 320
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.6788557767868042,
      "learning_rate": 2.022577553972233e-05,
      "loss": 0.3247,
      "step": 480
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1": 0.049217002237136466,
      "eval_loss": 0.32314401865005493,
      "eval_mcc": 0.046847973719895464,
      "eval_precision": 0.5,
      "eval_recall": 0.02588235294117647,
      "eval_runtime": 1.8829,
      "eval_samples_per_second": 677.158,
      "eval_steps_per_second": 21.244,
      "step": 480
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6379467844963074,
      "learning_rate": 0.0,
      "loss": 0.3227,
      "step": 640
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_f1": 0.018475750577367205,
      "eval_loss": 0.3209414780139923,
      "eval_mcc": 0.028093878027715367,
      "eval_precision": 0.5,
      "eval_recall": 0.009411764705882352,
      "eval_runtime": 1.8827,
      "eval_samples_per_second": 677.207,
      "eval_steps_per_second": 21.246,
      "step": 640
    }
  ],
  "logging_steps": 500,
  "max_steps": 640,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 1167021751680.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.5122281577891578,
    "learning_rate": 8.090310215888932e-05,
    "num_train_epochs": 4,
    "temperature": 14
  }
}