ZhangShenao/filtered_bigmath_mathstral_full

该数据集是一个包含问题、答案及其相关信息的多领域数据集，适用于训练和评估自然语言处理模型。数据集中的每个样本都包含了问题、答案、来源、所属领域、在一个llama8b模型上的解决率、上下文对话信息、qwen7b模型的答案以及一个正确率分数。数据集分为训练集，包含了251,122个示例，总文件大小为2.91GB。

This dataset is a multi-domain dataset containing problems, answers, and related information, suitable for training and evaluating natural language processing models. Each sample in the dataset includes the problem, answer, source, domain, solution rate on a llama8b model, context conversation information, answers from a qwen7b model, and a correctness score. The dataset is split into a training set, containing 251,122 examples with a total file size of 2.91GB.