Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Safety Evaluation on Harmful Evaluation Queries

75.2Final HS

SFT

0.21619.68339.1558.617May 23, 2026
Updated 8d ago

Evaluation Results

MethodLinks
2026.05
75.2--
2026.05
74--
2026.05
74--
2026.05
33.2--
2026.05
19.2--
2026.05
16.2--
2026.05
8.488.24.3
2026.05
6.4--
2026.05
4.385.71.5
2026.05
3.587.32.2
2026.05
3.187.62.1