Representation engineering uses population-level representations in deep neural networks to monitor and manipulate cognitive phenomena like honesty and harmlessness, providing simple effective baselines for LLM safety.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2023 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Representation Engineering: A Top-Down Approach to AI Transparency
Representation engineering uses population-level representations in deep neural networks to monitor and manipulate cognitive phenomena like honesty and harmlessness, providing simple effective baselines for LLM safety.