claim
active
claim:superposition-is-in-some-sense-deliberate-the-model-converts-pure-neurons-into-polysemantic-neurons-to-store-more-features-in-fewer-neuronsSuperposition is in some sense deliberate: the model converts pure neurons into polysemantic neurons to store more features in fewer neurons.
Interpretation of the cars-in-superposition circuit finding as an intentional representational strategy
Source paper
extracted_from(2020) · Chris Olah · Nick Cammarata · Ludwig Schubert · Gabriel Goh +2
Neighborhood — ranked by edge-count
Findings (1)
finding
- Circuit-level evidence that polysemantic neurons arise deliberately through superposition rather than entangled computation
Claims (1)
claim
- Mechanistic explanation for why superposition is geometrically feasible
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.
- superposition is deeply linked to the ability to identify and enumerate over all features in a modelquote0.840Safety rationale.
- Superposition hypothesis: neural networks represent more features than dimensions using almost-orthogonal directions.hypothesis0.838Explanation for why dictionary learning can recover many more features than dimensions.
- If we don't give a model enough neurons to implement this solution, how does it do computation in superposition?question0.835Left to future work.
- Superposition hypothesis: networks represent more features than neurons as almost-orthogonal directionshypothesis0.831Central hypothesis.
- Supported by abs model.
- Main contribution for interpretability.
- Analogy to many ice phases.
- Explanation for neuron interpretability.