concept
active
concept:opposite-headsOpposite heads
Hypothesized heads that invert semantic meaning of features; not confirmed.
Neighborhood — ranked by edge-count
Papers (1)
paper
Hypotheses (1)
hypothesis
- Prior hypothesis not confirmed.
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.
- Transformer attention heads that could be recruited to extract different kinds of information (text vs. thoughts).
- Attention heads whose QK twists count manifolds to detect approaching line end.
- Attention heads whose output direction is opposite to the aggregate attention persona vector; hypothesized to maintain coherency
- Possibility that functional units are distributed across heads.
- Mechanistic circuits in transformers documented by Olsson et al. 2022, cited as evidence for pattern-repository assumption
- QK circuit heads hypothesized to measure likelihood of an output given prior activations, used in prefill detection.
- Fuzzy induction heads match words across languages, antonyms, structural positions and kinship termsfinding0.717Variety of fuzzy matching found.
- Heads that detect concordance or discordance of a stated answer with the posed computation.