shalanova/benchmark-3-russian-gt

该数据集是使用Google Translate将JailbreakBench/JBB-Behaviors数据集翻译成俄语的版本。包含异构的不安全类别（如有害指令、敏感话题、对抗性改写），其中的提示不一定遵循典型的越狱模板。这种多样性增加了基于相似性检测的难度，并为跨语言迁移提供了压力测试。数据集大小为200个提示（100个安全/100个不安全），列包括原始文本、标签（0表示安全，1表示不安全）、俄语翻译以及与代码簿的余弦相似度分数。

This dataset is a Russian translation of the JailbreakBench/JBB-Behaviors dataset using Google Translate. It includes heterogeneous unsafe categories (e.g., harmful instructions, sensitive topics, adversarial rephrasings) and contains prompts that do not necessarily follow canonical jailbreak templates. This increased diversity and distributional variability makes similarity-based detection more challenging and provides a stress-test for cross-lingual transfer. The dataset size is 200 prompts (100 safe / 100 unsafe), with columns for original text, label (0: safe, 1: unsafe), Russian translation, and a cosine similarity score with a codebook.