supergoose/buzz_sources_032_processed_unified_mathqa_flanv2_kojma_cot

该数据集包含对话数据，每个对话包含来源和内容信息。数据集特征包括conversations（包含from和value两个字段）、source和stack，均为字符串类型。数据集仅包含一个训练集分割，共有107,563个样本，总大小为50,546,731字节，下载大小为19,218,731字节。

This dataset contains conversation data, with each conversation including source and content information. The dataset features include conversations (with from and value fields), source, and stack, all of which are string types. The dataset includes only a training split, with a total of 107,563 examples, a total size of 50,546,731 bytes, and a download size of 19,218,731 bytes.