{
  "best_metric": 0.6,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-7/checkpoint-2376",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2376,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.3391610383987427,
      "learning_rate": 6.940969436974916e-05,
      "loss": 0.4122,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5128712871287129,
      "eval_f1": 0.13986013986013984,
      "eval_loss": 0.4059968590736389,
      "eval_mcc": 0.047945551909639166,
      "eval_precision": 0.5882352941176471,
      "eval_recall": 0.07936507936507936,
      "eval_runtime": 0.9296,
      "eval_samples_per_second": 543.258,
      "eval_steps_per_second": 17.212,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5489808320999146,
      "learning_rate": 6.0733482573530516e-05,
      "loss": 0.4023,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5247524752475248,
      "eval_f1": 0.25465838509316774,
      "eval_loss": 0.4042968153953552,
      "eval_mcc": 0.06956267902824147,
      "eval_precision": 0.5857142857142857,
      "eval_recall": 0.1626984126984127,
      "eval_runtime": 0.9413,
      "eval_samples_per_second": 536.509,
      "eval_steps_per_second": 16.998,
      "step": 594
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.1429613828659058,
      "learning_rate": 5.205727077731187e-05,
      "loss": 0.3977,
      "step": 891
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5405940594059406,
      "eval_f1": 0.3255813953488372,
      "eval_loss": 0.40699630975723267,
      "eval_mcc": 0.10353804224648727,
      "eval_precision": 0.6086956521739131,
      "eval_recall": 0.2222222222222222,
      "eval_runtime": 0.9349,
      "eval_samples_per_second": 540.154,
      "eval_steps_per_second": 17.114,
      "step": 891
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6367562413215637,
      "learning_rate": 4.338105898109323e-05,
      "loss": 0.3948,
      "step": 1188
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5386138613861386,
      "eval_f1": 0.3206997084548105,
      "eval_loss": 0.39911898970603943,
      "eval_mcc": 0.09881732000604304,
      "eval_precision": 0.6043956043956044,
      "eval_recall": 0.21825396825396826,
      "eval_runtime": 0.9316,
      "eval_samples_per_second": 542.078,
      "eval_steps_per_second": 17.175,
      "step": 1188
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.9888297319412231,
      "learning_rate": 3.470484718487458e-05,
      "loss": 0.3917,
      "step": 1485
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5504950495049505,
      "eval_f1": 0.30581039755351686,
      "eval_loss": 0.3905428349971771,
      "eval_mcc": 0.14003878523542773,
      "eval_precision": 0.6666666666666666,
      "eval_recall": 0.1984126984126984,
      "eval_runtime": 0.9274,
      "eval_samples_per_second": 544.539,
      "eval_steps_per_second": 17.253,
      "step": 1485
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.473734974861145,
      "learning_rate": 2.6028635388655934e-05,
      "loss": 0.3871,
      "step": 1782
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5881188118811881,
      "eval_f1": 0.4285714285714286,
      "eval_loss": 0.38874551653862,
      "eval_mcc": 0.2107811676496078,
      "eval_precision": 0.6964285714285714,
      "eval_recall": 0.30952380952380953,
      "eval_runtime": 0.9277,
      "eval_samples_per_second": 544.378,
      "eval_steps_per_second": 17.248,
      "step": 1782
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.7756550908088684,
      "learning_rate": 1.735242359243729e-05,
      "loss": 0.3838,
      "step": 2079
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.598019801980198,
      "eval_f1": 0.5155131264916467,
      "eval_loss": 0.3883345127105713,
      "eval_mcc": 0.2076354401651176,
      "eval_precision": 0.6467065868263473,
      "eval_recall": 0.42857142857142855,
      "eval_runtime": 0.9379,
      "eval_samples_per_second": 538.42,
      "eval_steps_per_second": 17.059,
      "step": 2079
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.0732905864715576,
      "learning_rate": 8.676211796218645e-06,
      "loss": 0.3818,
      "step": 2376
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6,
      "eval_f1": 0.46276595744680843,
      "eval_loss": 0.38453981280326843,
      "eval_mcc": 0.23116659628812858,
      "eval_precision": 0.7016129032258065,
      "eval_recall": 0.34523809523809523,
      "eval_runtime": 0.9331,
      "eval_samples_per_second": 541.179,
      "eval_steps_per_second": 17.146,
      "step": 2376
    }
  ],
  "logging_steps": 500,
  "max_steps": 2673,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 5845610127840.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.5216419879912846,
    "learning_rate": 7.80859061659678e-05,
    "num_train_epochs": 9,
    "temperature": 30
  }
}