HerrHruby/simple_synthetic_programs

这是一个用于问答或推理任务的数据集，包含159,807个训练示例。每个示例具有以下特征：段落ID（唯一标识符）、段落文本（上下文）、类型（可能表示问题或任务类型）、问题（基于段落提出的问题）、答案（问题的答案）、cot（可能表示推理链或解释性文本）、family（可能表示数据类别或家族）、source（数据来源）。数据集适用于自然语言处理研究，如机器阅读理解、推理生成或问答系统训练。This is a dataset for question answering or reasoning tasks, containing 159,807 training examples. Each example includes the following features: passage_id (unique identifier), passage (context text), type (possibly indicating question or task type), question (question based on the passage), answer (answer to the question), cot (likely representing a chain of thought or explanation), family (possibly indicating data category or family), source (data origin). The dataset is suitable for natural language processing research, such as machine reading comprehension, reasoning generation, or training question answering systems.