Representation engineering: A top-down approach to AI transparency (Zou et al., 2023)

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Representational Transparencyconcept0.803
  Property of conscious representations: they do not contain information about the fact that they are representations at the level of the representation itself
• Representation Engineeringframework0.795
  A class of methods that modify how models internally process representations; SOO fine-tuning fits within this framework
• Representation engineering for large-language models: Survey and research challenges (Bartoszcze et al., 2025)concept0.794
  Survey of representation engineering methods cited as related work
• Representation engineering successfully quantifies deception via high-accuracy steering vectors, establishing it as a measurable property of model representationsclaim0.767
  Key interpretive claim that deception has a tractable geometric signature in activation space
• Representational abstraction of truth may emerge more clearly with model scaleclaim0.761
  Interpretation of weaker PCA separation and lower ASR in smaller models
• Representation Designconcept0.757
  The aspect of design dealing with data structures, modules, and implementation.
• Bai et al. 2022: Constitutional AI — harmlessness from AI feedbackconcept0.753
  Paper on AI-feedback fine-tuning as alternative to human-feedback RLHF; cited as ref 20
• The only human oversight is provided through a list of rules or principles, and so we refer to the method as 'Constitutional AI'.quote0.752
  Defines the core concept of the paper.