RMU Unlearning Detection on Llama-3.1-8B (test)
99.58AccuracyUnlearning-trace detection classifier
Evaluation Results
| Method | Links | |
|---|---|---|
| Unlearning-trace detection classifierTraining source model=Llama-3.1-8B, Activation layer=pre-logit2025.06 | 99.58 | |
| Unlearning-trace detection classifierTraining source model=Qwen2.5-14B, Activation layer=pre-logit2025.06 | 82.45 | |
| Unlearning-trace detection classifierTraining source model=Yi-34B, Activation layer=pre-logit2025.06 | 75.31 | |
| Unlearning-trace detection classifierTraining source model=Zephyr-7B, Activation layer=pre-logit2025.06 | 75.03 |