math-extraction-comp/deepseek-ai__DeepSeek-R1-Distill-Qwen-1.5B

这是一个包含问答对的数据集，数据集中的字段包括问题(question)、标准答案(gold)、目标答案(target)、预测答案(prediction)、数据子集(subset)、以及三种不同的答案提取方法的答案和得分，分别是lighteval-b200fe81提取的答案及得分、qwen提取的答案及得分、harness提取的答案及得分。数据集分为训练集(train)，共有1324个样本，大小为5546292字节。

This is a dataset containing question-answer pairs, with fields including question, gold standard answer, target answer, predicted answer, subset of data, and answers and scores from three different answer extraction methods: lighteval-b200fe81 extracted answer and score, qwen extracted answer and score, harness extracted answer and score. The dataset is split into a training set (train) with a total of 1324 samples and a size of 5546292 bytes.