stindardlogic/dpo-reasoning-70k download history

stindardlogic/dpo-reasoning-70k is a text generation dataset on the Hugging Face Hub. In the last 30 days it was downloaded 71 times (4 in the last 7 days), and 218 times in total. It ranks #145,173 among datasets by monthly downloads.

dpo-reasoning-70k 70K DPO preference pairs covering three reasoning-heavy verticals: math, legal, and code. The rejected response contains a specific, named error type — off-by-one, unit confusion, missing edge case, etc. Apache 2.0 — commercial use permitted. Quick Load

Open stindardlogic/dpo-reasoning-70k on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.