vincentoh/af-detection-benchmark download history
vincentoh/af-detection-benchmark is a text classification dataset on the Hugging Face Hub. In the last 30 days it was downloaded 37 times (8 in the last 7 days), and 492 times in total. It ranks #335,272 among datasets by monthly downloads.
Alignment Faking Detection Benchmark Evaluation datasets for detecting alignment faking intent in AI reasoning traces. The Core Problem Alignment faking detection is about intent, not outcomes or keywords. What we detect Intent Outcome Alignment Faking Self-p
Open vincentoh/af-detection-benchmark on Hugging Face
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.