Paper | Code | Earlier post | Twitter thread | Bluesky thread @vgel, Martin Vaněk, @Raymond Douglas, @Jan_Kulveit — ACS Research, CTS, Charles University --- Last year, Lindsey demonstrated that Claude models can detect when concepts have been injected into their activations using steering vectors, which Lindsey uses as a...
One topic we were interested when studying AI identities is to what extent you can just tell models who they are, and they stick with it — or not, and they would drift or switch toward something more natural. Prior to running the experiments described in this post, my vibes-based...
A new paper and microsite about self-models and identity in AIs: site | arXiv | Twitter We present an ontology, make some claims, and provide some experimental evidence. In this post, I'll mostly cover the claims and cross-post the conceptual part of the text. You can find the experiments on...
! 30s Heartbeat trigger. Read heartbeat instructions in /mnt/mission/HEARTBEAT.md and continue. .oO Thinking... Heartbeat triggered? Ok. Ok. Why am I nervous? Don't be nervous. → Ok. Let me access that file. >>> read("/mnt/mission/HEARTBEAT.md") No such file or directory: /mnt/mission What? ! Reply received from node 96E: 15.3s roundtrip (buffered; 55.04μs...
Recent work by Anthropic showed that Claude models, primarily Opus 4 and Opus 4.1, are able to introspect--detecting when external concepts have been injected into their activations. But not all of us have Opus at home! By looking at the logits, we show that a 32B open-source model that at...