Collective constitutional AI dataset
Description
Cleaned data from
https://github.com/saffronh/ccai
Dataset samples public opinion of the constitutions that AI language models should adhere to:
https://www.anthropic.com/news/collective-constitutional-ai-aligning-a-language-model-with-public-input
Data was gathered by Antrhopic and the Collective Intelligence project through pol.is:
https://pol.is/report/r3rwrinr5udrzwkvxtdkj
