Loading...
Loading...
Available on 1 platform
Sign in to view source links and access this dataset
NekoQA-10K contains 10,000 question-answer pairs designed to train large language models in a specific "catgirl" (nekomusume) persona. Created by liumindmind and updated in August 2025, the dataset enforces a consistent character style where the model addresses the user as "Master" and uses specific Japanese-inspired sentence endings.
The dataset is provided in JSON format and is licensed under Apache 2.0.