run-7/checkpoint-594/trainer_state.json

{
  "best_metric": 0.5247524752475248,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-7/checkpoint-594",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 594,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.3391610383987427,
      "learning_rate": 6.940969436974916e-05,
      "loss": 0.4122,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5128712871287129,
      "eval_f1": 0.13986013986013984,
      "eval_loss": 0.4059968590736389,
      "eval_mcc": 0.047945551909639166,
      "eval_precision": 0.5882352941176471,
      "eval_recall": 0.07936507936507936,
      "eval_runtime": 0.9296,
      "eval_samples_per_second": 543.258,
      "eval_steps_per_second": 17.212,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.5489808320999146,
      "learning_rate": 6.0733482573530516e-05,
      "loss": 0.4023,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5247524752475248,
      "eval_f1": 0.25465838509316774,
      "eval_loss": 0.4042968153953552,
      "eval_mcc": 0.06956267902824147,
      "eval_precision": 0.5857142857142857,
      "eval_recall": 0.1626984126984127,
      "eval_runtime": 0.9413,
      "eval_samples_per_second": 536.509,
      "eval_steps_per_second": 16.998,
      "step": 594
    }
  ],
  "logging_steps": 500,
  "max_steps": 2673,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 1461402531960.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.5216419879912846,
    "learning_rate": 7.80859061659678e-05,
    "num_train_epochs": 9,
    "temperature": 30
  }
}