Lugh@futurology.todayM to Futurology@futurology.todayEnglish · 6 months agoAnthropic just published new research that successfully identified and mapped millions of human-interpretable concepts, called “features”, within the neural networks of Claude.www.anthropic.comexternal-linkmessage-square3fedilinkarrow-up140arrow-down11
arrow-up139arrow-down1external-linkAnthropic just published new research that successfully identified and mapped millions of human-interpretable concepts, called “features”, within the neural networks of Claude.www.anthropic.comLugh@futurology.todayM to Futurology@futurology.todayEnglish · 6 months agomessage-square3fedilink
minus-squareDiplomjodler@lemmy.worldlinkfedilinkEnglisharrow-up5·6 months agoWow. This is potentially huge.
Wow. This is potentially huge.