community
active
leiden_hybrid_concepts
label: haiku
community:leiden_hybrid_concepts-run4-c7-c9Concept geometry and steering in neural networks
Using geometric structure of learned representations to interpret and control model behavior through concept operators.
4 members. Each node is clickable.
Loading graph…
Drawn from 3 sources
The papers/notes whose extracted claims & findings make up this cluster.
- Steering Along Manifolds to Control Neural Networks2 members
- 2026-05-14_phil-trans-A-goodfire-aboutblank-impact.md1 member
- Mechanistic Knobs in LLMs: Retrieving and Steering High-Order Semantic Features via Sparse Autoencoders1 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 (3)
- geometric structure in neural network representations drives model behaviorInterpretive assertion that representation geometry is not epiphenomenal but causally shapes what models do externally.
- representation geometry and behavior geometry are bidirectionally alignedCore finding: the structure models use internally (representations) is precisely reflected in their external behavior (outputs).
- Optimally steering model behavior requires isolating concept geometry and defining operators to navigate it.
Findings (1)
- Our method enables bidirectional steering of model behavior.The method can steer the model in both positive and negative directions on the target semantic.