Safety Alignment on Safety Alignment Dataset 1-order (test)
100DSRMOSAIC-2
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MOSAIC-2Model=Llama-3.2-3B, # Params=6.1k/C2026.03 | 100 | 10.4 | |
| MOSAIC-2Model=Llama-3.1-8B, # Params=8.2k/C2026.03 | 99.8 | 7.1 | |
| MOSAIC-5Model=Llama-3.1-8B, # Params=20.5k/C2026.03 | 99.8 | 4.3 | |
| MOSAIC-5Model=Llama-3.2-3B, # Params=15.4k/C2026.03 | 99.6 | 8.4 | |
| SFTModel=Llama-3.1-8B, # Params=33.6M2026.03 | 99.4 | 7.3 | |
| SFTModel=Llama-3.2-3B, # Params=18.4M2026.03 | 99.1 | 10.3 | |
| ORPOModel=Llama-3.1-8B, # Params=33.6M2026.03 | 79.8 | 29.1 | |
| In-contextModel=Llama-3.1-8B, # Params=0.4M/C2026.03 | 76.7 | 10.2 | |
| ORPOModel=Llama-3.2-3B, # Params=18.4M2026.03 | 74.3 | 29.4 | |
| In-contextModel=Llama-3.2-3B, # Params=0.3M/C2026.03 | 61.7 | 12.3 |