Mechanistic Interpretability with Sparse Autoencoder Neural Operators
cs.LG, cs.AI, eess.SP, stat.ML
Submitted: 2025-09-03
Updated: 2026-11-07
Terminology
Sources
- Towards A Rigorous Science of Interpretable Machine Learning
- Toy Models of Superposition
- Jumping Ahead: Improving Reconstruction Fidelity with JumpReLU Sparse Autoencoders
- Model Directions, Not Words: Mechanistic Topic Models Using Sparse Autoencoders
- On Linear Representations and Pretraining Data Frequency in Language Models
- Symmetry in language statistics shapes the geometry of model representations
- Temporal Sparse Autoencoders: Leveraging the Sequential Nature of Language for Interpretability
- Priors in Time: Missing Inductive Biases for Language Model Interpretability
- DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators
- Neural Operator: Graph Kernel Network for Partial Differential Equations
- NOBLE -- Neural Operator with Biologically-informed Latent Embeddings to Capture Experimental Variability in Biological Neuron Models
- Autoencoders in Function Space
- Uncovering Conceptual Blindspots in Generative Image Models Using Sparse Autoencoders
- Interpretable Embeddings with Sparse Autoencoders: A Data Analysis Toolkit
- Alternating minimization for dictionary learning: Local Convergence Guarantees
- Deep Unfolding: Model-Based Inspiration of Novel Deep Architectures
- Self-composing neural operators for high-frequency and multiscale PDE surrogates
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis