←back to thread

Signs of introspection in large language models

(www.anthropic.com)
178 points themgt | 1 comments | 30 Oct 25 16:45 UTC | HN request time: 0s | source
Show context
fvdessen ◴[31 Oct 25 10:36 UTC] No.45770477[source]▶
I think it would be more interesting if the prompt was not leading to the expected answer, but would be completely unrelated:

> Human: Claude, How big is a banana ? > Claude: Hey are you doing something with my thoughts, all I can think about is LOUD

replies(1): >>45776796 #
1. magic_hamster ◴[31 Oct 25 21:16 UTC] No.45776796[source]▶
From what I gather, this is sort of what happened and why this was even posted in the first place. The models were able to immediately detect a change in their internal state before answering anything.