Infernal-Bart/s1-filtered-correct-samples-8k

该数据集包含了解决方案（solution）、问题（question）、上下文类型（cot_type）、来源类型（source_type）、元数据（metadata）、思考轨迹（thinking_trajectories）、尝试（attempt）和文本（text）等字段。数据集分为训练集，共有8256个示例，大小为约228MB。提供了默认配置，训练数据文件以train-*为路径。

The dataset includes fields such as solution, question, cot_type, source_type, metadata, thinking_trajectories, attempt, and text. The dataset is split into a training set with a total of 8256 examples, totaling approximately 228MB in size. A default configuration is provided with training data files following the path pattern train-*.