run-2/checkpoint-246/trainer_state.json

{
  "best_metric": 0.5322896281800391,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/checkpoint-246",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 246,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.0383042097091675,
      "learning_rate": 2.0778074552979435e-05,
      "loss": 0.4799,
      "step": 123
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5,
      "eval_f1": 0.0,
      "eval_loss": 0.4709824323654175,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 29.6795,
      "eval_samples_per_second": 34.435,
      "eval_steps_per_second": 1.078,
      "step": 123
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.8932008743286133,
      "learning_rate": 1.780977818826809e-05,
      "loss": 0.4618,
      "step": 246
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5322896281800391,
      "eval_f1": 0.1758620689655172,
      "eval_loss": 0.44377976655960083,
      "eval_precision": 0.7391304347826086,
      "eval_recall": 0.09980430528375733,
      "eval_runtime": 29.7432,
      "eval_samples_per_second": 34.361,
      "eval_steps_per_second": 1.076,
      "step": 246
    }
  ],
  "logging_steps": 500,
  "max_steps": 984,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 25,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6164141606295046,
    "learning_rate": 2.3746370917690784e-05,
    "num_train_epochs": 8,
    "per_device_train_batch_size": 25,
    "temperature": 15
  }
}