math-extraction-comp/vicgalle__CarbonBeagle-11B-truthy

该数据集包含了问题、参考答案、模型预测答案、预测字符串、不同提取工具的答案和分数等字段。数据集分为训练集(train)，提供了字节数和示例数量信息。总体而言，这是一个用于文本匹配或答案预测的任务的数据集。

The dataset includes fields such as questions, reference answers, model predicted answers, prediction strings, answers and scores from different extraction tools. The dataset is split into training set (train), with information on the number of bytes and examples provided. Overall, this is a dataset for text matching or answer prediction tasks.