concept
active
concept:lookup-table-featuresLookup table features
Features translating between input digit properties and output properties in addition.
Neighborhood — ranked by edge-count
Papers (1)
paper
Methods (1)
method
- Operand plotsaboutVisualizations of feature activity on '=' token across 10,000 a+b prompts.
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.
- Interpretable concept representations hypothesized as basic units of computation.
- Method of optimizing input to cause a neuron to fire maximally, used to characterize what a neuron detects; establishes causal link
- The single primitive operation of looking up a value from a set of n keys.
- Features that activate before the action sequence they bring about, fulfilling functions of intention
- Transcoder feature that activates on p and produces p, memorizing something the model did not.
- Proposed attention analog of sparse monosemantic features for decomposing attention layers.
- Property that features activate on only a small fraction of inputs; enables compressed sensing and is what allows superposition to work
- Frequency-weighted spectral value πkλkj determining which features are retained.