mlfoundations-dev/oh_v1-2_only_alpaca

该数据集包含对话数据，每个对话包含from和value两个字段，均为字符串类型。数据集还包括一个shard_id字段，用于标识数据的分片。数据集包含一个训练分割，共有104,302个例子，总大小为43,903,046字节。

This dataset contains conversation data, with each conversation including from and value fields, both of string type. The dataset also includes a shard_id field to identify the shard of the data. It contains a training split with a total of 104,302 examples, amounting to 43,903,046 bytes.