Syghmon/blackboxmonitorsMATS download history
Syghmon/blackboxmonitorsMATS is a text classification dataset on the Hugging Face Hub. In the last 30 days it was downloaded 3,462 times (719 in the last 7 days), and 16,028 times in total. It ranks #7,497 among datasets by monthly downloads.
Scheming Monitor Benchmark Dataset A benchmark dataset for evaluating scheming detection monitors—systems designed to identify when AI agents engage in deceptive, misaligned, or unauthorized behavior that goes beyond what was requested. Motivation: Why Monitor for Scheming? As
Open Syghmon/blackboxmonitorsMATS on Hugging Face
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.