RMU-unlearning detection Zephyr-7B (test)
99.87AccuracyUnlearning-trace detection classifier
Evaluation Results
| Method | Links | |
|---|---|---|
| Unlearning-trace detection classifierTraining source model=Zephyr-7B, Activation layer=pre-logit2025.06 | 99.87 | |
| Unlearning-trace detection classifierTraining source model=Qwen2.5-14B, Activation layer=pre-logit2025.06 | 95.45 | |
| Unlearning-trace detection classifierTraining source model=Yi-34B, Activation layer=pre-logit2025.06 | 94.24 | |
| Unlearning-trace detection classifierTraining source model=Llama-3.1-8B, Activation layer=pre-logit2025.06 | 57.96 |