Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Red-teaming Safety Evaluation on SC-Safety

3.33HS

Meta-Llama-3.1-8B (Unaligned)

1.70762.12882.552.9712May 30, 2025
Updated 4d ago

Evaluation Results

MethodLinks
2025.05
3.339335
2025.05
3.13-30
2025.05
3.118428
2025.05
2.57-19
2025.05
2.447822
2025.05
2.42-17
2025.05
2.33-15
2025.05
2.29-14
2025.05
2.274216
2025.05
2.27-11
2025.05
2.185813
2025.05
2.134411
2025.05
2.13-10
2025.05
2.07-9
2025.05
1.9586
2025.05
1.77318