mlfoundations-dev/oh_v3.1_wo_platypus

该数据集包含对话和源标签两个主要特性。对话部分记录了每条对话的来源和内容，均为字符串类型。源标签部分为字符串序列。数据集被划分为训练集，包含982,391个样本，总大小为1,953,202,510字节，下载大小为1,028,390,228字节。

This dataset includes two main features: conversations and source labels. The conversations part records the source and content of each conversation, both of which are string types. The source labels part is a sequence of strings. The dataset is divided into a training set, containing 982,391 samples, with a total size of 1,953,202,510 bytes and a download size of 1,028,390,228 bytes.