Wild "And yet if you then ask Claude how it worked that out, it will say something like: “I added the ones, then carry [...] resulting in 95.” In other words, it gives you a common approach found everywhere online rather than what it actually did. Yep! LLMs are weird. (And not to be trusted.)"
technologyreview.comAnthropic can now track the bizarre inner workings of a large language modelWhat the firm found challenges some basic assumptions about how this technology really works.
0 reposts0 likes