HanningZhang/scalebio_qwen_math_20k_uw1e-5_alpha100_lambda1e-2

该数据集是一个对话数据集，包含消息内容和发送者角色信息，每个对话都有一个唯一的标识ID。数据集被划分为训练集，共有21,400个示例，总大小约为35,236,306.08字节。

This dataset is a conversational dataset that includes message content and sender role information, with each conversation having a unique identifier ID. The dataset is split into a training set, containing a total of 21,400 examples, with a total size of approximately 35,236,306.08 bytes.