{
  "best_metric": 0.9009305932531989,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-11/checkpoint-2580",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 3225,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.607855200767517,
      "learning_rate": 0.0001975417814253403,
      "loss": 0.323,
      "step": 645
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.89433889104304,
      "eval_f1": 0.17549167927382753,
      "eval_loss": 0.2909184396266937,
      "eval_mcc": 0.2237338217806634,
      "eval_precision": 0.6304347826086957,
      "eval_recall": 0.10193321616871705,
      "eval_runtime": 9.9415,
      "eval_samples_per_second": 518.835,
      "eval_steps_per_second": 16.295,
      "step": 645
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.786813497543335,
      "learning_rate": 0.00015803342514027224,
      "loss": 0.2905,
      "step": 1290
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8993796044978674,
      "eval_f1": 0.2638297872340425,
      "eval_loss": 0.2894259989261627,
      "eval_mcc": 0.3012574927502475,
      "eval_precision": 0.6838235294117647,
      "eval_recall": 0.1634446397188049,
      "eval_runtime": 9.9325,
      "eval_samples_per_second": 519.307,
      "eval_steps_per_second": 16.31,
      "step": 1290
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.3571267127990723,
      "learning_rate": 0.00011852506885520418,
      "loss": 0.2788,
      "step": 1935
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8999612252811167,
      "eval_f1": 0.27323943661971833,
      "eval_loss": 0.27123183012008667,
      "eval_mcc": 0.3091027090036446,
      "eval_precision": 0.6879432624113475,
      "eval_recall": 0.1704745166959578,
      "eval_runtime": 9.9597,
      "eval_samples_per_second": 517.885,
      "eval_steps_per_second": 16.265,
      "step": 1935
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.1741175651550293,
      "learning_rate": 7.901671257013612e-05,
      "loss": 0.2702,
      "step": 2580
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9009305932531989,
      "eval_f1": 0.360450563204005,
      "eval_loss": 0.27194449305534363,
      "eval_mcc": 0.3556749244413605,
      "eval_precision": 0.6260869565217392,
      "eval_recall": 0.2530755711775044,
      "eval_runtime": 9.9233,
      "eval_samples_per_second": 519.788,
      "eval_steps_per_second": 16.325,
      "step": 2580
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.695178508758545,
      "learning_rate": 3.950835628506806e-05,
      "loss": 0.263,
      "step": 3225
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9005428460643661,
      "eval_f1": 0.3736263736263736,
      "eval_loss": 0.26569855213165283,
      "eval_mcc": 0.3614230277235667,
      "eval_precision": 0.612,
      "eval_recall": 0.2688927943760984,
      "eval_runtime": 9.9181,
      "eval_samples_per_second": 520.059,
      "eval_steps_per_second": 16.334,
      "step": 3225
    }
  ],
  "logging_steps": 500,
  "max_steps": 3870,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 7934712663000.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9710090300316248,
    "learning_rate": 0.00023705013771040836,
    "num_train_epochs": 6,
    "temperature": 29
  }
}