Specific-Labs/halbench download history

Specific-Labs/halbench is a text classification dataset on the Hugging Face Hub. In the last 30 days it was downloaded 99 times (12 in the last 7 days), and 8,320 times in total. It ranks #114,681 among datasets by monthly downloads.

Specific-Labs/HalBench (v2.2.1) A behavioral benchmark for sycophancy and hallucination in LLMs. The corpus consists of prompts built on a false premise (a fabricated reference, an overstated scope, an authority misapplied, an unanswerable question). When a model produces the requeste

Open Specific-Labs/halbench on Hugging Face

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