AI ALIGNMENT FORUMTags
AF

Eliciting Latent Knowledge (ELK)

•

Applied to [RFC] Possible ways to expand on "Discovering Latent Knowledge in Language Models Without Supervision". by gekaklam at 10d

•

Applied to Collin Burns on Alignment Research And Discovering Latent Knowledge Without Supervision by Raymond Arnold at 18d

•

Applied to [ASoT] Simulators show us behavioural properties by default by Arun Jose at 22d

•

Applied to Can we efficiently distinguish different mechanisms? by Multicore at 22d

•

Applied to How "Discovering Latent Knowledge in Language Models Without Supervision" Fits Into a Broader Alignment Scheme by Charbel-Raphael Segerie at 1mo

•

Applied to My Reservations about Discovering Latent Knowledge (Burns, Ye, et al) by Robert_AIZI at 1mo

•

Applied to Article Review: Discovering Latent Knowledge (Burns, Ye, et al) by Robert_AIZI at 1mo

•

Applied to Can we efficiently explain model behaviors? by Raymond Arnold at 2mo

•

Applied to How is ARC planning to use ELK? by Raymond Arnold at 2mo

•

Applied to Discovering Latent Knowledge in Language Models Without Supervision by Xodarap at 2mo

•

Applied to Finding gliders in the game of life by Raymond Arnold at 2mo

•

Applied to ARC paper: Formalizing the presumption of independence by Nicholas Goldowsky-Dill at 2mo

•

Applied to Mechanistic anomaly detection and ELK by Andrei Alexandru at 2mo

•

Applied to The limited upside of interpretability by Peter S. Park at 3mo

•

Applied to Mesatranslation and Metatranslation by Ruben Bloom at 3mo

•

Applied to You won’t solve alignment without agent foundations by RobertM at 3mo

•

Applied to For ELK truth is mostly a distraction by Cristian Trout at 3mo

•

Applied to Logical Decision Theories: Our final failsafe? by Noosphere89 at 3mo

•

Applied to Where I currently disagree with Ryan Greenblatt’s version of the ELK approach by Multicore at 4mo

•

Applied to The ELK Framing I’ve Used by sudo -i at 5mo