training_config.json

{
  "model_type": "feedback_detector",
  "labels": [
    "SAT",
    "NEED_CLARIFICATION",
    "WRONG_ANSWER",
    "WANT_DIFFERENT"
  ],
  "label2id": {
    "SAT": 0,
    "NEED_CLARIFICATION": 1,
    "WRONG_ANSWER": 2,
    "WANT_DIFFERENT": 3
  },
  "id2label": {
    "0": "SAT",
    "1": "NEED_CLARIFICATION",
    "2": "WRONG_ANSWER",
    "3": "WANT_DIFFERENT"
  },
  "base_model": "llm-semantic-router/mmbert-32k-yarn",
  "max_length": 512,
  "use_lora": true,
  "lora_rank": 64,
  "lora_alpha": 128,
  "class_weights": [
    1.0,
    1.0,
    1.0,
    1.0
  ],
  "metrics": {
    "eval_loss": 0.07450290769338608,
    "eval_accuracy": 0.9882747068676717,
    "eval_f1_macro": 0.9824253474141716,
    "eval_f1_weighted": 0.9882717896311659,
    "eval_f1_SAT": 1.0,
    "eval_f1_NEED_CLARIFICATION": 0.9979919678714859,
    "eval_f1_WRONG_ANSWER": 0.967098703888335,
    "eval_f1_WANT_DIFFERENT": 0.9646107178968655,
    "eval_runtime": 4.499,
    "eval_samples_per_second": 663.488,
    "eval_steps_per_second": 20.894,
    "epoch": 5.361930294906166
  }
}