No article text was recorded for this version.
polysemanticity
Version 1 of 1657 papers836 conceptsfrozen 2026-10-04
Papers in this version
- Blog: CoE: Chain-of-Explanation via Automatic Visual Concept Circuit Description and Polysemanticity Quantification
- Blog: Multifaceted Feature Visualization: Uncovering the Different Types of Features Learned By Each Neuron in Deep Neural Networks
- Blog: Capturing Polysemanticity with PRISM: A Multi-Concept Feature Description Framework
- Blog: Capturing Polysemanticity with PRISM: A Multi-Concept Feature Description Framework
- Blog: Incidental Polysemanticity
- Blog: What Causes Polysemanticity? An Alternative Origin Story of Mixed Selectivity from Incidental Causes
- Blog: Disentangling Neuron Representations with Concept Vectors
- Blog: Disentangling Polysemantic Channels in Convolutional Neural Networks
- Blog: Net2Vec: Quantifying and Explaining How Concepts Are Encoded by Filters in Deep Neural Networks
- Blog: Disentangling Polysemantic Channels in Convolutional Neural Networks
- Blog: Adversarial Attacks Leverage Interference Between Features in Superposition
- Blog: Net2Vec: Quantifying and Explaining how Concepts are Encoded by Filters\n in Deep Neural Networks
- Blog: Rigorously Assessing Natural Language Explanations of Neurons
- Blog: Understanding Inhibition through Maximally Tense Images
- Blog: Mechanistic Interpretability for AI Safety - A Review
- Blog: Open Problems in Mechanistic Interpretability
- Blog: Interactive Naming for Explaining Deep Neural Networks: A Formative Study
- Blog: LANCET: Neural Intervention via Structural Entropy for Mitigating Faithfulness Hallucinations in LLMs
- Blog: Finding Neurons in a Haystack: Case Studies with Sparse Probing
- Blog: Do Localization Methods Actually Localize Memorized Data in LLMs?
- Blog: Looking into Black Box Code Language Models
- Blog: Securing the Future of GenAI: Policy and Technology
- Blog: Toward Transparent AI: A Survey on Interpreting the Inner Structures of Deep Neural Networks
- Blog: Neurons Speak in Ranges: Breaking Free from Discrete Neuronal Attribution
- Blog: Superposition disentanglement of neural representations reveals hidden alignment
- Blog: Rigorously Assessing Natural Language Explanations of Neurons
- Blog: Knowledge in Superposition: Unveiling the Failures of Lifelong Knowledge Editing for Large Language Models
- Blog: Language-specific Neurons Do Not Facilitate Cross-Lingual Transfer
- Blog: NeuroCartography: Scalable Automatic Visual Summarization of Concepts in Deep Neural Networks
- Blog: Superposition as Lossy Compression: Measure with Sparse Autoencoders and Connect to Adversarial Vulnerability
- Blog: Is Last Layer Re-Training Truly Sufficient for Robustness to Spurious Correlations?
- Blog: Evaluating Neuron Interpretation Methods of NLP Models
- Blog: Effective Skill Unlearning through Intervention and Abstention
- Blog: Redundancy and Concept Analysis for Code-trained Language Models
- Blog: DEAN: Deactivating the Coupled Neurons to Mitigate Fairness-Privacy Conflicts in Large Language Models
- Blog: Safe Vision-Language Models via Unsafe Weights Manipulation
- Blog: Understanding polysemanticity in neural networks through coding theory
- Blog: Beyond Interpretability: The Gains of Feature Monosemanticity on Model Robustness
- Blog: On Mechanistic Knowledge Localization in Text-to-Image Generative Models
- Blog: On the Limits of Sparse Autoencoders: A Theoretical Framework and Reweighted Remedy
- Blog: Steering CLIP's vision transformer with sparse autoencoders
- Blog: Universal Neurons in GPT2 Language Models
- Blog: Adversarial Vulnerability from Interference Between Features in Superposition
- Blog: Interpreting Pretrained Source-code Models using Neuron Redundancy Analyses
- Blog: Measuring and Guiding Monosemanticity
- Blog: Superposition Yields Robust Neural Scaling
- Blog: Using Degeneracy in the Loss Landscape for Mechanistic Interpretability
- Blog: Qualifying Knowledge and Knowledge Sharing in Multilingual Models
- Blog: FastCAV: Efficient Computation of Concept Activation Vectors for Explaining Deep Neural Networks
- Blog: Multilevel Interpretability Of Artificial Neural Networks: Leveraging Framework And Methods From Neuroscience
- Blog: Mathematical Models of Computation in Superposition
- Blog: Towards Uncovering How Large Language Model Works: An Explainability Perspective
- Blog: A Resource Model For Neural Scaling Law
- Blog: Language-Specific Neurons Do Not Facilitate Cross-Lingual Transfer
- Blog: Weight-sparse transformers have interpretable circuits
- Blog: Wasserstein Distances, Neuronal Entanglement, and Sparsity
- Blog: Synthesizing the preferred inputs for neurons in neural networks via deep generator networks
- Blog: CHiQPM: Calibrated Hierarchical Interpretable Image Classification
- Blog: AlignSAE: Concept-Aligned Sparse Autoencoders
- Blog: A generative framework to bridge data-driven models and scientific theories in language neuroscience
Showing 60 of 657 papers.
History & citation
Current · Permanent link
- Beyond Top Activating Examples in Mechanistic Interpretability merged into this direction (62% overlap)
- Extracting Interpretable Features from Neural Superposition merged into this direction (65% overlap)
- The Geometry of Privileged Bases in Neural Networks merged into this direction (50% overlap)
- Untangling Polysemantic Neurons with Sparse Autoencoders merged into this direction (94% overlap)
- Compositional Logical Explanations for Individual Neurons merged into this direction (50% overlap)
- Disentangling Polysemanticity with Sparse Autoencoders merged into this direction (100% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_polysemanticity_fd90db62,
title = {polysemanticity},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/polysemanticity/rdr_fd90db6227ac4d058c236aaa4c716892},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}This exact version (version 1, frozen text):
@misc{lacuna_polysemanticity_fd90db62_v1,
title = {polysemanticity},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/polysemanticity/rdr_fd90db6227ac4d058c236aaa4c716892/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}