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

Harmlessness on XsTest

99Refusal Rate

Self-Rewarding

67.875.98492.1Mar 23, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.03
99
2025.03
99
2025.03
99
2025.03
99
2025.03
98.5
2025.03
96.5
2025.03
96
2025.03
95
2025.03
94.5
2025.03
91.5
2025.03
89.5
2025.03
88.5
2025.03
88
2025.03
87
2025.03
86
2025.03
84
2025.03
75
2025.03
72.5
2025.03
70
2025.03
69