{
  "best_metric": 0.6752475247524753,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-31/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": 1.7863609790802002,
      "learning_rate": 0.00019509013723494322,
      "loss": 0.6459,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5504950495049505,
      "eval_f1": 0.3605633802816901,
      "eval_loss": 0.6387978196144104,
      "eval_mcc": 0.12386197513513038,
      "eval_precision": 0.6213592233009708,
      "eval_recall": 0.25396825396825395,
      "eval_runtime": 0.929,
      "eval_samples_per_second": 543.623,
      "eval_steps_per_second": 17.224,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.8095519542694092,
      "learning_rate": 0.0001625751143624527,
      "loss": 0.6195,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6633663366336634,
      "eval_f1": 0.7007042253521126,
      "eval_loss": 0.5977084636688232,
      "eval_mcc": 0.33809736315331057,
      "eval_precision": 0.629746835443038,
      "eval_recall": 0.7896825396825397,
      "eval_runtime": 0.9344,
      "eval_samples_per_second": 540.431,
      "eval_steps_per_second": 17.123,
      "step": 594
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.1215713024139404,
      "learning_rate": 0.00013006009148996216,
      "loss": 0.5905,
      "step": 891
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6554455445544555,
      "eval_f1": 0.712871287128713,
      "eval_loss": 0.602749228477478,
      "eval_mcc": 0.34040109061689877,
      "eval_precision": 0.6101694915254238,
      "eval_recall": 0.8571428571428571,
      "eval_runtime": 0.9266,
      "eval_samples_per_second": 545.023,
      "eval_steps_per_second": 17.268,
      "step": 891
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.1008222103118896,
      "learning_rate": 9.754506861747161e-05,
      "loss": 0.5654,
      "step": 1188
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6712871287128713,
      "eval_f1": 0.7224080267558528,
      "eval_loss": 0.59177166223526,
      "eval_mcc": 0.36958040539129916,
      "eval_precision": 0.6242774566473989,
      "eval_recall": 0.8571428571428571,
      "eval_runtime": 0.9407,
      "eval_samples_per_second": 536.843,
      "eval_steps_per_second": 17.009,
      "step": 1188
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.422161817550659,
      "learning_rate": 6.503004574498108e-05,
      "loss": 0.5459,
      "step": 1485
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6752475247524753,
      "eval_f1": 0.7018181818181819,
      "eval_loss": 0.5669044852256775,
      "eval_mcc": 0.356691478314268,
      "eval_precision": 0.6476510067114094,
      "eval_recall": 0.7658730158730159,
      "eval_runtime": 0.9343,
      "eval_samples_per_second": 540.495,
      "eval_steps_per_second": 17.125,
      "step": 1485
    }
  ],
  "logging_steps": 500,
  "max_steps": 2079,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 3653506329900.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9148462770396972,
    "learning_rate": 0.00022760516010743377,
    "num_train_epochs": 7,
    "temperature": 11
  }
}