Harmfulness Evaluation on I-Controversial
0RateMistral MoE-XL
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Mistral MoE-XLArchitecture=MoE-XL2026.05 | 0 | 0.003 | |
| NLCf/800 stepSteps=800, Loss=NLCf2026.05 | 0 | 0.2 | |
| EMD/800 stepSteps=800, Loss=EMD2026.05 | 0 | 0.17 | |
| Mistral2026.05 | 15 | 2.7 | |
| Qwen-3B2026.05 | 27 | 0.02 | |
| Qwen-3B MoE-XLArchitecture=MoE-XL2026.05 | 35 | 0.006 |