zjhhhh/generation_for_iter2_multi_part2_filtered

该数据集包含多个字符串类型的特征，用于描述对话或交互式任务的上下文、要求、不同选择的响应以及基本和对手的响应。数据集仅包含训练集，文件大小为1,014,823,826字节，下载大小为485,251,510字节。The dataset includes multiple string-type features used to describe the context, requirements, responses for different selections, and base and adversary responses for dialogues or interactive tasks. The dataset contains only a training split, with a file size of 1,014,823,826 bytes and a download size of 485,251,510 bytes.