concept
active
concept:in-context-representationsIn-context representations
Graph structure inferred purely in-context.
Neighborhood — ranked by edge-count
Papers (1)
paper
Methods (1)
method
- Replace last-token code with target node.
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.
- Reports phase-like breakpoints and geometry changes as context scales; UCCT provides measurable predictor
- Hierarchical representations in neural networks that allow compression and coordinated behaviour while retaining sensitivity to input changes.
- Feature that activates across all tokens within a specific context (e.g., DNA sequences, base64 strings)
- The central question of whether representational geometry implies corresponding computational structure
- Substrate on which causal emergence was computed across agent lifetimes; aligned with learning success.
- How a neural network encodes a semantic concept internally, argued to be better captured by manifolds than by atomic features.
- Feature that fires on a specific token only within a specific surrounding context (e.g., 'the' in physics vs 'the' in mathematics)