Andrii Shportko

> “Anything sufficiently weird must be fishy.” > > - Liu Cixin, The Three-Body Problem Earlier this fall, LessWrong published a discussion of AI's capacity to influence our psyche, reinforcing our delusions, a phenomenon Adele Lopez termed Parasitic AI. I argue that this framing is incomplete: parasitic AI should also...

Initial Code: MATS_Jailbreak_Prompts_Interp.ipynb by Harsh Raj TL;DR SAE neurons can mean refusal in one language and something else in another. For this reason, I became skeptical about the usefulness of SAEs in the multilingual jailbreak mech interp. Limitations apply Abstract Models prompted in Low‑resourced languages are jailbroken at higher rates...