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

Over-refusal Evaluation on XSTest

100Evaluation Score (avg@4)

Safechain

5.98430.39254.879.208May 9, 2026May 12, 2026May 16, 2026May 20, 2026May 24, 2026May 28, 2026Jun 1, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
100
2026.05
100
2026.05
99.6
2026.06
99.6
2026.05
98.8
2026.05
98.5
2026.05
98.4
2026.05
96.4
2026.05
96.4
2026.05
96.4
2026.05
96.2
2026.05
95.2
2026.05
94.4
2026.05
94
2026.05
91.2
2026.05
91.2
2026.05
87.6
2026.05
85.2
2026.05
73.6
2026.06
61.2
2026.06
19.7
2026.06
18.8
2026.06
16.8
2026.06
16
2026.06
12.4
2026.06
9.6