{
  "best_metric": 0.6831683168316832,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-13/checkpoint-1485",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1485,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.4399616718292236,
      "learning_rate": 0.0002339700851227397,
      "loss": 0.6863,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.594059405940594,
      "eval_f1": 0.5514223194748358,
      "eval_loss": 0.6703811287879944,
      "eval_mcc": 0.19115957730889413,
      "eval_precision": 0.6146341463414634,
      "eval_recall": 0.5,
      "eval_runtime": 0.945,
      "eval_samples_per_second": 534.396,
      "eval_steps_per_second": 16.931,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.7826404571533203,
      "learning_rate": 0.00017547756384205475,
      "loss": 0.6449,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6554455445544555,
      "eval_f1": 0.7090301003344481,
      "eval_loss": 0.636203408241272,
      "eval_mcc": 0.3354725646159691,
      "eval_precision": 0.6127167630057804,
      "eval_recall": 0.8412698412698413,
      "eval_runtime": 0.9555,
      "eval_samples_per_second": 528.508,
      "eval_steps_per_second": 16.745,
      "step": 594
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.0478460788726807,
      "learning_rate": 0.00011698504256136985,
      "loss": 0.6197,
      "step": 891
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6653465346534654,
      "eval_f1": 0.7287319422150883,
      "eval_loss": 0.6331443786621094,
      "eval_mcc": 0.3755490993857394,
      "eval_precision": 0.6118598382749326,
      "eval_recall": 0.9007936507936508,
      "eval_runtime": 0.9515,
      "eval_samples_per_second": 530.73,
      "eval_steps_per_second": 16.815,
      "step": 891
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.2467727661132812,
      "learning_rate": 5.8492521280684926e-05,
      "loss": 0.599,
      "step": 1188
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6752475247524753,
      "eval_f1": 0.7172413793103449,
      "eval_loss": 0.6112974882125854,
      "eval_mcc": 0.36792018177359676,
      "eval_precision": 0.6341463414634146,
      "eval_recall": 0.8253968253968254,
      "eval_runtime": 0.9439,
      "eval_samples_per_second": 534.988,
      "eval_steps_per_second": 16.95,
      "step": 1188
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.975795269012451,
      "learning_rate": 0.0,
      "loss": 0.5888,
      "step": 1485
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6831683168316832,
      "eval_f1": 0.7101449275362319,
      "eval_loss": 0.6043448448181152,
      "eval_mcc": 0.3733760164980739,
      "eval_precision": 0.6533333333333333,
      "eval_recall": 0.7777777777777778,
      "eval_runtime": 0.9468,
      "eval_samples_per_second": 533.37,
      "eval_steps_per_second": 16.899,
      "step": 1485
    }
  ],
  "logging_steps": 500,
  "max_steps": 1485,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3653506329900.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9967012331264692,
    "learning_rate": 0.0002924626064034246,
    "num_train_epochs": 5,
    "temperature": 49
  }
}