{
  "best_metric": 0.5587084148727984,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-384",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 384,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.0371571779251099,
      "learning_rate": 0.0001305282668790618,
      "loss": 0.3216,
      "step": 192
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5499021526418787,
      "eval_f1": 0.23841059602649006,
      "eval_loss": 0.2955350875854492,
      "eval_mcc": 0.17350853577425898,
      "eval_precision": 0.7741935483870968,
      "eval_recall": 0.14090019569471623,
      "eval_runtime": 67.6225,
      "eval_samples_per_second": 15.113,
      "eval_steps_per_second": 0.473,
      "step": 192
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.1835968494415283,
      "learning_rate": 0.00011421223351917907,
      "loss": 0.2979,
      "step": 384
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5587084148727984,
      "eval_f1": 0.2642740619902121,
      "eval_loss": 0.29196101427078247,
      "eval_mcc": 0.19586518668589045,
      "eval_precision": 0.7941176470588235,
      "eval_recall": 0.15851272015655576,
      "eval_runtime": 67.5618,
      "eval_samples_per_second": 15.127,
      "eval_steps_per_second": 0.474,
      "step": 384
    }
  ],
  "logging_steps": 500,
  "max_steps": 1728,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.4208671172864604,
    "learning_rate": 0.00014684430023894452,
    "num_train_epochs": 9,
    "per_device_train_batch_size": 16,
    "temperature": 48
  }
}