mlfoundations-dev/oh_v1.3_alpaca_x4

该数据集包含对话数据，每个对话包含from和value两个字段，分别表示对话的来源和内容。数据集被分割为训练集，包含1,067,856个样本，总大小为1,543,131,875字节。数据集的下载大小为851,620,699字节。

This dataset contains conversational data, with each conversation including two fields: from and value, representing the source and content of the conversation, respectively. The dataset is split into a training set containing 1,067,856 samples, with a total size of 1,543,131,875 bytes. The download size of the dataset is 851,620,699 bytes.