---
license: apache-2.0
task_categories:
- question-answering
language:
- zh
---

# Chinese Enthusiastic DPO

This dataset comprises conversational pairs characterized by an informal, colloquial style. Its primary objective is to train large language models (LLMs) to respond with greater enthusiasm, thereby enhancing their writing style in roleplay scenarios.

该数据集包含以非正式、口语化风格为特征的对话对。其主要目标是训练大型语言模型（LLM）以更热情的方式进行回应，从而提升其在角色扮演场景中的写作风格。