nifty-researcher/memoir-assistant

该数据集是一个用于训练大型语言模型（LLMs）的回忆录和个人叙事辅导及写作指导的完整JSONL数据集集合。它包含多个文件，每个文件专注于回忆录写作的不同方面，如发现性辅导、高级技巧和文学分析。数据集来源于开放教育资源，格式为JSONL，便于与OpenAI和HuggingFace等微调管道集成。数据集覆盖了写作技巧和辅导方法，包括结构与组织、声音与视角、细节与描述、意义与结构、伦理写作等方面。此外，还提供了如何使用这些数据集的建议，包括通用回忆录辅导、发现性辅导、个人叙事教学、文学分析与写作技巧、混合媒体自我探索等。数据集质量经过手动编写、基于同行评审的出版来源、强调具体可操作的指导等多重保证。

This dataset is a complete collection of JSONL datasets for training large language models (LLMs) on memoir and personal narrative coaching and writing instruction. It includes various files, each focusing on different aspects of memoir writing, such as discovery coaching, advanced techniques, and literary analysis. The dataset is sourced from open educational resources and formatted in JSONL for easy integration with fine-tuning pipelines like OpenAI and HuggingFace. It covers writing techniques and coaching approaches, including structure & organization, voice & perspective, detail & description, meaning & structure, and ethical writing. The README also provides recommendations on how to use these datasets, such as for general memoir coaching, discovery-focused coaching, personal narrative teaching, literary analysis + writing craft, and mixed-media self-exploration. The datasets quality is assured through hand-written entries, peer-reviewed sources, and emphasis on specific, actionable guidance.