mlfoundations-dev/evalset_13f9

该数据集是一个训练集，包含了上下文信息（内容与角色）、生成参数（包括是否采样、最大新生成token数、随机种子和温度）、重复索引、请求索引、任务名称和元数据（期望答案、问题ID和参考解决方案）。数据集共有1594个训练样本，文件大小为1120514字节。

This dataset is a training set that includes context information (content and role), generation parameters (including whether to sample, the maximum number of new tokens to generate, random seed, and temperature), repeat index, request index, task name, and metadata (expected answer, problem ID, and reference solution). The dataset consists of 1594 training samples with a total file size of 1,120,514 bytes.