Sveriges mest populära poddar
LessWrong (Curated & Popular)
Samhälle och kulturTeknologi

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

9 min•20 januari 2024

This is a linkpost for https://arxiv.org/abs/2401.05566

Support ongoing human narrations of LessWrong's curated posts:
www.patreon.com/LWCurated

Source:
https://www.lesswrong.com/posts/ZAsJv7xijKTfZkMtr/sleeper-agents-training- deceptive-llms-that-persist-through

Narrated for LessWrong by Perrin Walker.

Share feedback on this narration.

[Curated Post] ✓
[125+ Karma Post] ✓

Fler avsnitt av LessWrong (Curated & Popular)

Visa alla avsnitt av LessWrong (Curated & Popular)
"What is up with e/acc?" by KatjaGrace
•4 min
"Existential AI safety needs an effective social movement. PauseAI is building it" by Maxime Fournes, Espedair Street
•1 tim 3 min
"Surprising facts about the slave trade" by Joseph Miller
•13 min
"AI catastrophe: more like a genocide than a thought experiment" by KatjaGrace
•2 min
"AI pause: the case for ASAP" by KatjaGrace
•2 min
"The Invisible Side of AI Governance" by Charbel-Raphaël
•28 min
"A Theory of Prompt Injection (and why you should study roles)" by Charles Ye, softboiledheart
•32 min
"Machinic Psychopharmacology: Do LLMs Self-Medicate?" by Sid Black, Joseph Bloom
•53 min
"Can activation verbalizers surface an internal chain of thought?" by oakhu, ryan_greenblatt
•1 tim 20 min
"The LLM shoggoth meme is weirder than you think" by HedonicEscalator
•14 min

LessWrong (Curated & Popular) med LessWrong finns tillgänglig på flera plattformar. Informationen på denna sida kommer från offentliga podd-flöden.