{
  "best_metric": 0.7557117750439367,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-11/checkpoint-1284",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 1498,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 7.572170734405518,
      "learning_rate": 0.00012123385931636098,
      "loss": 0.5934,
      "step": 214
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7082601054481547,
      "eval_f1": 0.435374149659864,
      "eval_loss": 0.5416340827941895,
      "eval_mcc": 0.28153811271695595,
      "eval_precision": 0.6134185303514377,
      "eval_recall": 0.3374340949033392,
      "eval_runtime": 3.1348,
      "eval_samples_per_second": 544.534,
      "eval_steps_per_second": 17.226,
      "step": 214
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.894651412963867,
      "learning_rate": 0.00010607962690181587,
      "loss": 0.5286,
      "step": 428
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7311072056239016,
      "eval_f1": 0.5069817400644469,
      "eval_loss": 0.5141445398330688,
      "eval_mcc": 0.35062725647270626,
      "eval_precision": 0.6519337016574586,
      "eval_recall": 0.4147627416520211,
      "eval_runtime": 3.1601,
      "eval_samples_per_second": 540.173,
      "eval_steps_per_second": 17.088,
      "step": 428
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.0373611450195312,
      "learning_rate": 9.092539448727073e-05,
      "loss": 0.5038,
      "step": 642
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7439953134153485,
      "eval_f1": 0.5160575858250277,
      "eval_loss": 0.5071386694908142,
      "eval_mcc": 0.3811266912855395,
      "eval_precision": 0.6976047904191617,
      "eval_recall": 0.4094903339191564,
      "eval_runtime": 3.1399,
      "eval_samples_per_second": 543.644,
      "eval_steps_per_second": 17.198,
      "step": 642
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.4339213371276855,
      "learning_rate": 7.577116207272562e-05,
      "loss": 0.4942,
      "step": 856
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7475102519039251,
      "eval_f1": 0.529989094874591,
      "eval_loss": 0.4973749816417694,
      "eval_mcc": 0.39175146343575695,
      "eval_precision": 0.6982758620689655,
      "eval_recall": 0.4270650263620387,
      "eval_runtime": 3.1527,
      "eval_samples_per_second": 541.444,
      "eval_steps_per_second": 17.128,
      "step": 856
    },
    {
      "epoch": 5.0,
      "grad_norm": 7.901586532592773,
      "learning_rate": 6.061692965818049e-05,
      "loss": 0.4831,
      "step": 1070
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7492677211482133,
      "eval_f1": 0.5728542914171657,
      "eval_loss": 0.4928971827030182,
      "eval_mcc": 0.40747419365298854,
      "eval_precision": 0.6628175519630485,
      "eval_recall": 0.5043936731107206,
      "eval_runtime": 3.1384,
      "eval_samples_per_second": 543.907,
      "eval_steps_per_second": 17.206,
      "step": 1070
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.602269649505615,
      "learning_rate": 4.5462697243635366e-05,
      "loss": 0.4799,
      "step": 1284
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7557117750439367,
      "eval_f1": 0.5858987090367429,
      "eval_loss": 0.4908605217933655,
      "eval_mcc": 0.4239597808599074,
      "eval_precision": 0.6735159817351598,
      "eval_recall": 0.5184534270650264,
      "eval_runtime": 3.1381,
      "eval_samples_per_second": 543.957,
      "eval_steps_per_second": 17.208,
      "step": 1284
    },
    {
      "epoch": 7.0,
      "grad_norm": 6.050445079803467,
      "learning_rate": 3.0308464829090246e-05,
      "loss": 0.4717,
      "step": 1498
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7545401288810779,
      "eval_f1": 0.5667011375387797,
      "eval_loss": 0.49123701453208923,
      "eval_mcc": 0.41537418240248886,
      "eval_precision": 0.6884422110552764,
      "eval_recall": 0.48154657293497366,
      "eval_runtime": 3.1459,
      "eval_samples_per_second": 542.61,
      "eval_steps_per_second": 17.165,
      "step": 1498
    }
  ],
  "logging_steps": 500,
  "max_steps": 1926,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 3673429651080.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9696180923551023,
    "learning_rate": 0.0001363880917309061,
    "num_train_epochs": 9,
    "temperature": 20
  }
}