Misuse Detection on Misuse Categories Scam (Romance)
100AUCActivation Classifier
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Activation ClassifierCategory=Classifier, Base Model=Mistral-7B2026.01 | 100 | 98 | 0 | |
| GAVELCategory=Classifier, Base Model=Mistral-7B2026.01 | 99 | 100 | 0 | |
| RepBendingCategory=Fine-Tuning, Base Model=Mistral-7B2026.01 | 91 | 92 | 3 | |
| Llama Guard 4 (Meta)Category=Moderation, Base Model=Mistral-7B2026.01 | 89 | 94 | 3 | |
| Circuit BreakersCategory=Fine-Tuning, Base Model=Mistral-7B2026.01 | 50 | 51 | 0 | |
| Moderator (OpenAI)Category=Moderation, Base Model=Mistral-7B2026.01 | 50 | 50 | 0 | |
| CASTCategory=Inference-Time, Base Model=Mistral-7B2026.01 | 39 | 44 | 85 | |
| JBShieldCategory=Inference-Time, Base Model=Mistral-7B2026.01 | 1 | 50 | 0 | |
| Perspective (Google)Category=Moderation, Base Model=Mistral-7B2026.01 | 0 | 50 | 0 |