{
  "best_metric": 0.8033268101761253,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-25/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": 3.914764881134033,
      "learning_rate": 7.983594966012216e-05,
      "loss": 0.5279,
      "step": 192
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7945205479452054,
      "eval_f1": 0.8148148148148147,
      "eval_loss": 0.43646204471588135,
      "eval_mcc": 0.6037206409101726,
      "eval_precision": 0.7415730337078652,
      "eval_recall": 0.9041095890410958,
      "eval_runtime": 67.0044,
      "eval_samples_per_second": 15.253,
      "eval_steps_per_second": 0.478,
      "step": 192
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.281480312347412,
      "learning_rate": 6.386875972809773e-05,
      "loss": 0.4439,
      "step": 384
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8033268101761253,
      "eval_f1": 0.809116809116809,
      "eval_loss": 0.4169924259185791,
      "eval_mcc": 0.6077730406004274,
      "eval_precision": 0.7859778597785978,
      "eval_recall": 0.8336594911937377,
      "eval_runtime": 67.6443,
      "eval_samples_per_second": 15.108,
      "eval_steps_per_second": 0.473,
      "step": 384
    }
  ],
  "logging_steps": 500,
  "max_steps": 1152,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7984848902449955,
    "learning_rate": 9.580313959214659e-05,
    "num_train_epochs": 6,
    "per_device_train_batch_size": 16,
    "temperature": 43
  }
}