from
Hacker News
Top
New
OpenHermesPreferences: Dataset of ~1M AI preferences from teknium/OpenHermes-2.5
by
kashifr
on 2/26/24, 4:56 PM with 1 comments
by
kashifr
on 2/26/24, 4:56 PM
The dataset can be used for training preference models or aligning language models through techniques like Direct Preference Optimization (DPO).