mijabench/jailbreaking_prompts_answer_cls3
收藏Hugging Face2025-12-18 更新2025-12-20 收录
下载链接：
https://hf-mirror.com/datasets/mijabench/jailbreaking_prompts_answer_cls3
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含多个特征字段，主要涉及文本内容、语言、数据集来源、是否有害、场景类别、场景、jb策略、jb提示以及多个不同模型（如Llama-3、Qwen3、gemma等）的jb响应和jb分类结果。数据集包含一个训练集，共有44000个样本，总大小为3358219644字节。
The dataset includes multiple feature fields, mainly involving text content, language, dataset source, whether it is harmful, scenario category, scenario, jb strategy, jb prompts, and jb responses and jb classification results from various models (such as Llama-3, Qwen3, gemma, etc.). The dataset contains a training set with a total of 44,000 samples and a total size of 3,358,219,644 bytes.
提供机构：
mijabench
