Dataset
SafetyBench
A comprehensive Chinese and English safety evaluation benchmark with 11,435 multiple-choice questions spanning 7 safety scenarios, including offensive content, bias, physical safety, and privacy. Designed to evaluate LLM safety in bilingual contexts.
Owner: Tsinghua CoAI Lab
Access: Available on Hugging Face without access restrictions. Dataset and evaluation scripts are publicly downloadable.
Safety use case
Multilingual safety evaluation of LLMs; identifying safety gaps across diverse harm categories in both Chinese and English.
Benchmark relevance
Relevant to: SafetyBench
Source
SafetyBench on Hugging FaceStatus: Verified. Last verified: 2026-06-13.
Want to list this dataset on the marketplace?
Data Canon takes no transaction fees. Contact us to update listing details or add new datasets.
Contact about this dataset