Cornell-AGI/math_size_qwen2.5_3b_eval

该数据集包含文本数据，其中包括等级、类型、数据来源、提示内容、角色、能力、奖励模型、额外信息（包括答案、索引、问题和分割信息）、多个响应字段和评估分数。数据集有7500个训练示例。

The dataset consists of textual data, including fields such as level, type, data source, prompt content, role, ability, reward model, extra information (including answer, index, question, and split), multiple response fields, and evaluation scores. The dataset has 7500 training examples.