question
active
question:are-concepts-encoded-linearlyAre concepts encoded linearly?
Mechanistic open question.
Source paper
extracted_from(2025) · Adam Karvonen
Neighborhood — ranked by edge-count
Papers (1)
paper
Questions (1)
question
- How is semantic meaning encoded in activations?associated_withMechanistic question AOs can't answer.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- The established finding that transformer LLMs encode many interpretable concepts as linear directions in activation space
- Correlative technique measuring the type of information encoded in distributed representations via linear predictability.
- Jackson's assertion that concepts encompass state and behavior that motivates them, beyond what programmers assign to datatypes
- The finding that interpretable concepts including character traits are encoded as linear directions in transformer residual streams
- The idea that features are encoded as directions in activation space.
- Open question A.1.1a.2.
- The central object of study — the idea that a concept like truth is encoded as a direction in the LLM's latent space
- Open question.