4 posts tagged by Digg
AI
Goodfire says sparse autoencoders are useful for exploring a model when you don’t know what to look for. For a specific target, such as reward hacking, it recommends a probe.
AI
A researcher at the lab says the grant will support a search for evidence of world models and convergent representations in today’s agents, building on earlier theoretical work.
AI
Marius Hobbhahn shares two perspectives on building for-profit AI safety businesses.
AI
The AI interpretability company announces access to its research tool.
