Hanno-Labs/warrantbench download history

Hanno-Labs/warrantbench is a text ranking dataset on the Hugging Face Hub. In the last 30 days it was downloaded 48 times (13 in the last 7 days), and 231 times in total. It ranks #273,827 among datasets by monthly downloads.

WarrantBench A benchmark for programmable, opinion-free relational judgment — does a judge execute the rule you gave it, including rules it never trained on? From Hanno Labs. A judge is any system — a frontier LLM, an embedder, or a specialized rater — that, given a proposed edge betw

Models trained on warrantbench

2 models list it as training data.

Open Hanno-Labs/warrantbench on Hugging Face

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