Deema/countdown-reasoning-dataset

这是一个包含不同难度级别训练和测试数据的数据集，适用于评估模型在处理具有不同难度的问题时的性能。数据集特征包括问题提示、索引、难度、数字列表、目标值、数据来源、奖励模型和额外信息。

This is a dataset with training and test data of different difficulty levels, which is suitable for evaluating the performance of models in dealing with problems of different difficulties. The dataset features include problem prompts, index, difficulty, list of numbers, target value, data source, reward model, and extra information.