SafetyBench: Evaluating the Safety of Large Language Models with Multiple Choice Questions

Open in new window