[HUMAN VOICE] "Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training" by evhub et al

LessWrong (Curated & Popular)
20/01/2024 8 min
[HUMAN VOICE] "Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training" by evhub et al

Listen "[HUMAN VOICE] "Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training" by evhub et al"

Descargar episodio

Episode Synopsis

This is a linkpost for https://arxiv.org/abs/2401.05566Support ongoing human narrations of LessWrong's curated posts:www.patreon.com/LWCuratedSource:https://www.lesswrong.com/posts/ZAsJv7xijKTfZkMtr/sleeper-agents-training- deceptive-llms-that-persist-throughNarrated for LessWrong by Perrin Walker.Share feedback on this narration.[Curated Post] ✓[125+ Karma Post] ✓

More episodes of the podcast LessWrong (Curated & Popular)

"Scientific breakthroughs of the year" by technicalities 17/12/2025
"A high integrity/epistemics political machine?" by Raemon 17/12/2025
"How I stopped being sure LLMs are just making up their internal experience (but the topic is still confusing)" by Kaj_Sotala 16/12/2025
“My AGI safety research—2025 review, ’26 plans” by Steven Byrnes 15/12/2025
“Weird Generalization & Inductive Backdoors” by Jorio Cocola, Owain_Evans, dylan_f 14/12/2025
“Insights into Claude Opus 4.5 from Pokémon” by Julian Bradshaw 13/12/2025
“The funding conversation we left unfinished” by jenn 13/12/2025
“The behavioral selection model for predicting AI motivations” by Alex Mallen, Buck 11/12/2025
“Little Echo” by Zvi 09/12/2025
“A Pragmatic Vision for Interpretability” by Neel Nanda 08/12/2025
© 2003-2025 ZARZA | AHEAD OF OUR TIME
privacy policy
In God we trust In God we trust
PreSales Questions? Click Here
or call our toll-free line at +1-888-5002068 Miami, Florida, USA