mlfoundations-dev/oh_v1.3_evol_instruct_x8

该数据集包含对话数据，每条记录由对话内容（conversations）和分片ID（shard_id）组成。对话内容进一步细分为来源（from）和值（value）。数据集分为训练集（train），包含1,135,022个示例，总大小为2,209,368,989字节。

This dataset contains conversation data, with each record consisting of conversation content (conversations) and a shard ID (shard_id). The conversation content is further divided into source (from) and value (value). The dataset is divided into a training set (train), containing 1,135,022 examples, with a total size of 2,209,368,989 bytes.