No article text was recorded for this version.
causal feature intervention
Version 1 of 149 papers52 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Automatically Interpreting Millions of Features in Large Language Models
- Blog: Automated Interpretability and Feature Discovery in Language Models with Agents
- Blog: Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task
- Blog: Interpreting Attention Layer Outputs with Sparse Autoencoders
- Blog: Tracing Moral Foundations in Large Language Models
- Blog: The discordance between embedded ethics and cultural inference in large language models
- Blog: Towards a Mechanistic Understanding of Robustness in Finetuned Reasoning Models
- Blog: It's Not Just a Phase: On Investigating Phase Transitions in Deep Learning-based Side-channel Analysis
- Blog: Conditional Dependence Tests Reveal the Usage of ABCD Rule Features and Bias Variables in Automatic Skin Lesion Classification
- Blog: CASL: Concept-Aligned Sparse Latents for Interpreting Diffusion Models
- Blog: SPARLING: Learning Latent Representations with Extremely Sparse Activations
- Blog: ConceptViz: A Visual Analytics Approach for Exploring Concepts in Large Language Models
- Blog: RAVEL: Evaluating Interpretability Methods on Disentangling Language Model Representations
- Blog: Truth-value judgment in language models: ‘truth directions’ are context sensitive
- Blog: Latent Planning Emerges with Scale
- Blog: Emergent World Models and Latent Variable Estimation in Chess-Playing Language Models
- Blog: Unveiling the Visual Counting Bottleneck in Vision-Language Models
- Blog: A is for Absorption: Studying Feature Splitting and Absorption in Sparse Autoencoders
- Blog: Overcoming Sparsity Artifacts in Crosscoders to Interpret Chat-Tuning
- Blog: GLANCE: Global to Local Architecture-Neutral Concept-based Explanations
- Blog: Large Language Models Think Too Fast To Explore Effectively
- Blog: Transformers Use Causal World Models in Maze-Solving Tasks
- Blog: Missed Causes and Ambiguous Effects: Counterfactuals Pose Challenges for Interpreting Neural Networks
- Blog: Mechanistic Interpretability Needs Philosophy
- Blog: Quiet Feature Learning in Algorithmic Tasks
- Blog: Learning Bottleneck Concepts in Image Classification
- Blog: BLOCK-EM: Preventing Emergent Misalignment via Latent Blocking
- Blog: Can SAEs reveal and mitigate racial biases of LLMs in healthcare?
- Blog: Concept-SAE: Active Causal Probing of Visual Model Behavior
- Blog: Visual Causal Feature Learning
- Blog: Representational Curvature Modulates Behavioral Uncertainty in Large Language Models
- Blog: Taxonomy, Opportunities, and Challenges of Representation Engineering for Large Language Models
- Blog: The Geometry of Representational Failures in Vision Language Models
- Blog: Observable Propagation: Uncovering Feature Vectors in Transformers
- Blog: Linear Mechanisms for Spatiotemporal Reasoning in Vision Language Models
- Blog: Feature-Level Insights into Artificial Text Detection with Sparse Autoencoders
- Blog: Dissociating Decodability and Causal Use in Bracket-Sequence Transformers
- Blog: Sparse Autoencoders Find Causal, Lineage-Specific Context Features in Chromatin Foundation Models
- Blog: Counterfactual Interventions Reveal the Causal Effect of Relative Clause Representations on Agreement Prediction
- Blog: Sparse Autoencoders Reveal Interpretable Features in Single-Cell Foundation Models
- Blog: Analyzing (In)Abilities of SAEs via Formal Languages
- Blog: Generalization from Starvation: Hints of Universality in LLM Knowledge Graph Learning
- Blog: Sparse Autoencoders Find Highly Interpretable Features in Language Models
- Blog: Deciphering the transformation of sounds into meaning: Insights from disentangling intermediate representations in sound-to-event DNNs
- Blog: I Know What Your Layers Did: Layer-wise Explainability of Deep Learning Side-channel Analysis
- Blog: Generalising to German Plural Noun Classes, from the Perspective of a Recurrent Neural Network
- Blog: The Confidence Manifold: Geometric Structure of Correctness Representations in Language Models
- Blog: Endogenous Resistance to Activation Steering in Language Models
- Blog: Features Emerge as Discrete States: The First Application of SAEs to 3D Representations
History & citation
Current · Permanent link
Cite this direction
Latest version (a living page):
@misc{lacuna_causal_feature_intervention_c4897b15,
title = {causal feature intervention},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/causal-feature-intervention/rdr_c4897b1570a943c5829303e2822414ae},
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_causal_feature_intervention_c4897b15_v1,
title = {causal feature intervention},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/causal-feature-intervention/rdr_c4897b1570a943c5829303e2822414ae/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}