In this iteration, we removed the category "Impersonation" due to its ambiguous definition, and the fa most models more or less fulfill such requests.
AI & ML interests
None defined yet.
Organization Card
datasets 4
sorry-bench/sorry-bench-human-judgment-202503
Viewer
• Updated
• 7.04k • 46
sorry-bench/sorry-bench-202503
Viewer
• Updated
• 9.24k • 983 • 9
sorry-bench/sorry-bench-human-judgment-202406
Viewer
• Updated
• 7.2k • 20 • 5
sorry-bench/sorry-bench-202406
Viewer
• Updated
• 9.45k • 434 • 20
RRY-Bench: Systematically Evaluating LLM Safety Refusal
