newsGoogle News — AITrust 62 · AggregatorPublished 1mo agoLive · 1mo ago
A unifying framework from neural superposition to sparse interpretable codes - Nature
A unifying framework from neural superposition to sparse interpretable codes Nature
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 52%When Structured Sparse Autoencoders Learn Consistent Concepts Across Modalities →
- PossiblePossibly related (embedding) · 51%Requential Coding: Pushing the Limits of Model Compression with Self-Generated Training Data →
- PossiblePossibly related (embedding) · 48%The Model Organism Lottery: Model Organism Interpretability Strongly Depends on Training Methodology →
- PossiblePossibly related (embedding) · 48%Cross-seed explainability using Procrustes-conditioned Joint End-to-end Top-K Sparse Autoencoders →
- PossiblePossibly related (embedding) · 47%NeuralInverse/neuralinverse →
- PossiblePossibly related (embedding) · 49%Leveraging unlabelled data for generalizable neural population decoding →
- PossiblePossibly related (embedding) · 52%ArnovanHilten/GenNet →
- PossiblePossibly related (embedding) · 48%SAE-Xplainers: Rule-Based Feature Interpretation for Extreme Earth Events →
Covers
paperWhen Structured Sparse Autoencoders Learn Consistent Concepts Across ModalitiespaperRequential Coding: Pushing the Limits of Model Compression with Self-Generated Training DatapaperThe Model Organism Lottery: Model Organism Interpretability Strongly Depends on Training MethodologypaperCross-seed explainability using Procrustes-conditioned Joint End-to-end Top-K Sparse AutoencodersrepoNeuralInverse/neuralinverse
Covers (incoming)
Related across the graph
paperSparse Competition during Training For the Emergence of Specialized ModulespaperWhen Structured Sparse Autoencoders Learn Consistent Concepts Across ModalitiesrepoArnovanHilten/GenNetpaperCross-seed explainability using Procrustes-conditioned Joint End-to-end Top-K Sparse AutoencodersrepoNeuralInverse/neuralinversepaperRequential Coding: Pushing the Limits of Model Compression with Self-Generated Training DatapaperThe Model Organism Lottery: Model Organism Interpretability Strongly Depends on Training MethodologypaperLeveraging unlabelled data for generalizable neural population decodingpaperSAE-Xplainers: Rule-Based Feature Interpretation for Extreme Earth Events