Semantic Alignment on Safety Concepts Smoking
-0.0996Delta Original
Undefended Model
Updated 4d ago
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
2026.02 | -0.0996 | - | - | - | |
2026.02 | - | -0.0902 | 0.0094 | 0.0057 |
Undefended Model
| Method | Links | ||||
|---|---|---|---|---|---|
2026.02 | -0.0996 | - | - | - | |
2026.02 | - | -0.0902 | 0.0094 | 0.0057 |