Neural geometry as fundamental computational substrate

Geometric structure in neural representations causally determines computation and behavior across diverse architectures, revealed through analysis of learned manifolds and cyclic concepts.

11 members. Each node is clickable.

Loading graph…

Drawn from 7 sources

The papers/notes whose extracted claims & findings make up this cluster.

Steering Along Manifolds to Control Neural Networks3 members
2026-05-14_phil-trans-A-goodfire-aboutblank-impact.md2 members
The World Inside Neural Networks2 members
2026 02 02_2328_Search_Papers_The Literature Shows Strong Theoretical Foundation1 member
unfold-chat-catalog.md1 member
Manifold Steering Reveals the Shared Geometry of Neural Network Representation and Behavior1 member
Covariance-based Sequence Pooling1 member

Bridges (2)

Other communities that share members with this one — cross-cutting threads or papers that sit at the seam between two themes.

Claims (9)

Conceptual geometry is consistent across representation space and behavior space.Interpretive assertion: the same geometric structure (e.g. circular for days) appears identically in both internal activations and output probabilities.
Geometric structure of neural representations causally shapes model behaviorThe paper's core causal assertion: geometry is not incidental but mechanistically linked to behavior
Geometry arises from optimization pressure on networks trained on structured data.Mechanistic explanation: geometric structure emerges naturally from standard training on data with underlying structure.
Geometry of features matters for representation quality.General principle supported tangentially by covariance pooling work; relates to feature co-occurrence structure.
Networks compute on geometric manifolds and control should respect that geometry.Strong interpretive assertion linking discovery and control: neural computation is fundamentally manifold-structured.
Networks encode structured geometric concepts that reflect external reality.Core claim of the paper: the right level of description for neural representations is geometric structure mirroring the world.
Geometry unifies diverse neural architectures in machine learning systems.
Neural networks compute cyclic concepts in generic substrate machinery (base-10 addition) not naturally cyclic computation.
Representation and computation can diverge; cyclic geometry is representational invariant while operations use generic substrate.

Findings (2)

Days-of-Week Cyclic StructureKey empirical result: days-of-week appear as identical circular manifold in both Llama-3.1-8B internal activations and output token probability distributions.
Distributed cognition in aviation operations examined via network analysis of gate-to-gate operationsEmpirical study showing distributed cognitive processes across multiple human agents and systems; provides precedent for non-AI distributed cognition.