Sidsidney/UltraInteract_sft

UltraInteract是一个大规模、高质量的用于复杂推理任务的对齐数据集。每个指令包含一个偏好树，其中包括多样化的规划策略推理链、多轮交互轨迹和成对数据。数据集结构包括任务类型、数据集来源、指令描述和响应。数据集格式为JSONC，包含id、parent_id、task、dataset、instruction和response等字段。数据集提供了编码、数学和逻辑三个示例。

UltraInteract is a large-scale, high-quality alignment dataset specifically designed for complex reasoning tasks. For each instruction, it includes a preference tree consisting of reasoning chains with diverse planning strategies, multi-turn interaction trajectories, and pairwise data. The dataset structure includes task type, dataset source, instruction description, and response. The dataset format is JSONC, containing fields such as id, parent_id, task, dataset, instruction, and response. The dataset provides three examples: coding, math, and logic.