{"total":79,"rows":[{"id":"2502.16681","title":"Are Sparse Autoencoders Useful? A Case Study in Sparse Probing","created":"2025-02-23","cited_by_count":115,"area":"mechinterp"},{"id":"2505.20063","title":"SAEs Are Good for Steering -- If You Select the Right Features","created":"2025-05-26","cited_by_count":78,"area":"mechinterp"},{"id":"2503.05613","title":"A Survey on Sparse Autoencoders: Interpreting the Internal Mechanisms of Large Language Models","created":"2025-03-07","cited_by_count":76,"area":"mechinterp"},{"id":"2504.13151","title":"MIB: A Mechanistic Interpretability Benchmark","created":"2025-04-17","cited_by_count":56,"area":"mechinterp"},{"id":"2504.14218","title":"Understanding the Repeat Curse in Large Language Models from a Feature Perspective","created":"2025-04-19","cited_by_count":37,"area":"mechinterp"},{"id":"2502.11356","title":"SAIF: A Sparse Autoencoder Framework for Interpreting and Steering Instruction Following of Language Models","created":"2025-02-17","cited_by_count":34,"area":"mechinterp"},{"id":"2601.14004","title":"Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models","created":"2026-01-20","cited_by_count":27,"area":"mechinterp"},{"id":"2503.08200","title":"Route Sparse Autoencoder to Interpret Large Language Models","created":"2025-03-11","cited_by_count":26,"area":"mechinterp"},{"id":"2601.22594","title":"Language Model Circuits Are Sparse in the Neuron Basis","created":"2026-01-30","cited_by_count":25,"area":"mechinterp"},{"id":"2605.11887","title":"Qwen-Scope: Turning Sparse Features into Development Tools for Large Language Models","created":"2026-05-12","cited_by_count":21,"area":"mechinterp"},{"id":"2503.03601","title":"Feature-Level Insights into Artificial Text Detection with Sparse Autoencoders","created":"2025-03-05","cited_by_count":15,"area":"mechinterp"},{"id":"2502.03032","title":"Analyze Feature Flow to Enhance Interpretation and Steering in Language Models","created":"2025-02-05","cited_by_count":13,"area":"mechinterp"},{"id":"2505.05111","title":"Unveiling Language-Specific Features in Large Language Models via Sparse Autoencoders","created":"2025-05-08","cited_by_count":13,"area":"mechinterp"},{"id":"2508.12535","title":"CorrSteer: Generation-Time LLM Steering via Correlated Sparse Autoencoder Features","created":"2025-08-18","cited_by_count":12,"area":"mechinterp"},{"id":"2504.20938","title":"Towards Understanding the Nature of Attention with Low-Rank Sparse Decomposition","created":"2025-04-29","cited_by_count":9,"area":"mechinterp"},{"id":"2503.24277","title":"Evaluating and Designing Sparse Autoencoders by Approximating Quasi-Orthogonality","created":"2025-03-31","cited_by_count":8,"area":"mechinterp"},{"id":"2602.05027","title":"AudioSAE: Towards Understanding of Audio-Processing Models with Sparse AutoEncoders","created":"2026-02-04","cited_by_count":8,"area":"mechinterp"},{"id":"2602.06964","title":"Learning a Generative Meta-Model of LLM Activations","created":"2026-02-06","cited_by_count":8,"area":"mechinterp"},{"id":"2501.19406","title":"Low-Rank Adapting Models for Sparse Autoencoders","created":"2025-01-31","cited_by_count":7,"area":"mechinterp"},{"id":"2602.04428","title":"Fine-Grained Activation Steering: Steering Less, Achieving More","created":"2026-02-04","cited_by_count":7,"area":"mechinterp"},{"id":"2605.00842","title":"Understanding Emergent Misalignment via Feature Superposition Geometry","created":"2026-04-07","cited_by_count":7,"area":"mechinterp"},{"id":"2508.16929","title":"Dimensional Collapse in Transformer Attention Outputs: A Challenge for Sparse Dictionary Learning","created":"2025-08-23","cited_by_count":6,"area":"mechinterp"},{"id":"2509.18127","title":"Safe-SAIL: Towards a Fine-grained Safety Landscape of Large Language Models via Sparse Autoencoder Interpretation Framework","created":"2025-09-11","cited_by_count":6,"area":"mechinterp"},{"id":"2511.19265","title":"Unboxing the Black Box: Mechanistic Interpretability for Algorithmic Understanding of Neural Networks","created":"2025-11-24","cited_by_count":6,"area":"mechinterp"},{"id":"2601.05437","title":"Tracing Moral Foundations in Large Language Models","created":"2026-01-09","cited_by_count":6,"area":"mechinterp"},{"id":"2603.13277","title":"Learning Retrieval Models with Sparse Autoencoders","created":"2026-02-27","cited_by_count":6,"area":"mechinterp"},{"id":"2604.19974","title":"Are LLM Uncertainty and Correctness Encoded by the Same Features? A Functional Dissociation via Sparse Autoencoders","created":"2026-04-21","cited_by_count":6,"area":"mechinterp"},{"id":"2506.10920","title":"Constructing Interpretable Features from Compositional Neuron Groups","created":"2025-06-12","cited_by_count":5,"area":"mechinterp"},{"id":"2507.12990","title":"Teach Old SAEs New Domain Tricks with Boosting","created":"2025-07-17","cited_by_count":5,"area":"mechinterp"},{"id":"2509.17665","title":"Mechanistic Interpretability with SAEs: Probing Religion, Violence, and Geography in Large Language Models","created":"2025-09-22","cited_by_count":4,"area":"mechinterp"},{"id":"2512.02004","title":"AlignSAE: Concept-Aligned Sparse Autoencoders","created":"2025-12-01","cited_by_count":4,"area":"mechinterp"},{"id":"2603.23301","title":"Steering LLMs for Culturally Localized Generation","created":"2026-03-24","cited_by_count":4,"area":"mechinterp"},{"id":"2505.17769","title":"Inference-Time Decomposition of Activations (ITDA): A Scalable Approach to Interpreting Large Language Models","created":"2025-05-23","cited_by_count":3,"area":"mechinterp"},{"id":"2510.01048","title":"Interpreting Language Models Through Concept Descriptions: A Survey","created":"2025-10-01","cited_by_count":3,"area":"mechinterp"},{"id":"2604.06005","title":"Disentangling MLP Neuron Weights in Vocabulary Space","created":"2026-04-07","cited_by_count":3,"area":"mechinterp"},{"id":"2605.31183","title":"Steering LLMs? Actually, Sparse Autoencoders can outperform simple baselines","created":"2026-05-29","cited_by_count":3,"area":"mechinterp"},{"id":"2607.25270","title":"Where Steering Signals Come From: Activation Source Selection in Activation Steering","created":"2026-07-28","cited_by_count":3,"area":"mechinterp"},{"id":"2506.20040","title":"Cross-Layer Discrete Concept Discovery for Interpreting Language Models","created":"2025-06-24","cited_by_count":2,"area":"mechinterp"},{"id":"2507.14894","title":"SASFT: Sparse Autoencoder-guided Supervised Finetuning to Mitigate Unexpected Code-Switching in LLMs","created":"2025-07-20","cited_by_count":2,"area":"mechinterp"},{"id":"2507.15227","title":"Mammo-SAE: Interpreting Breast Cancer Concept Learning with Sparse Autoencoders","created":"2025-07-21","cited_by_count":2,"area":"mechinterp"},{"id":"2605.12874","title":"Descriptive Collision in Sparse Autoencoder Auto-Interpretability: When One Explanation Describes Many Features","created":"2026-05-13","cited_by_count":2,"area":"mechinterp"},{"id":"2506.07691","title":"Breaking the Block: Preserving Data Continuity to Train Superior SAEs for Instruct Models","created":"2025-06-09","cited_by_count":1,"area":"mechinterp"},{"id":"2509.20376","title":"ConceptViz: A Visual Analytics Approach for Exploring Concepts in Large Language Models","created":"2025-09-20","cited_by_count":1,"area":"mechinterp"},{"id":"2510.18462","title":"DePass: Unified Feature Attributing by Simple Decomposed Forward Pass","created":"2025-10-21","cited_by_count":1,"area":"mechinterp"},{"id":"2601.22447","title":"Beyond Activation Patterns: A Weight-Based Out-of-Context Explanation of Sparse Autoencoder Features","created":"2026-01-30","cited_by_count":1,"area":"mechinterp"},{"id":"2604.03532","title":"LangFIR: Discovering Sparse Language-Specific Features from Monolingual Data for Language Steering","created":"2026-04-04","cited_by_count":1,"area":"mechinterp"},{"id":"2604.25866","title":"Why Are Some Emotions Harder for LLMs? Uncovering the Causal Mechanisms of Emotion Inference via Sparse Autoencoders","created":"2026-04-28","cited_by_count":1,"area":"mechinterp"},{"id":"2605.24946","title":"Interpretability Transfer from Language to Vision via Sparse Autoencoders","created":"2026-05-24","cited_by_count":1,"area":"mechinterp"},{"id":"2605.27819","title":"ReSAE: Residualized Sparse Autoencoders for Multi-Layer Transformer Interventions","created":"2026-05-27","cited_by_count":1,"area":"mechinterp"},{"id":"2607.20596","title":"Are Single-Token Sparse Autoencoder Features Causally Necessary? Layer-Depth and SAE-Family Effects","created":"2026-07-22","cited_by_count":1,"area":"mechinterp"},{"id":"2608.28806","title":"Enhancing SAE-based Steering via Neighbor Integrated Feature Selection","created":"2026-08-28","cited_by_count":1,"area":"mechinterp"},{"id":"2506.21468","title":"TopK Language Models","created":"2025-06-26","cited_by_count":0,"area":"mechinterp"},{"id":"2509.20784","title":"Towards Atoms of Large Language Models","created":"2025-09-25","cited_by_count":0,"area":"mechinterp"},{"id":"2601.02978","title":"Mechanistic Knobs in LLMs: Retrieving and Steering High-Order Semantic Features via Sparse Autoencoders","created":"2026-01-06","cited_by_count":0,"area":"mechinterp"},{"id":"2601.20420","title":"Concept Component Analysis: A Principled Approach for Concept Extraction in LLMs","created":"2026-01-28","cited_by_count":0,"area":"mechinterp"},{"id":"2604.23829","title":"Domain-Filtered Knowledge Graphs from Sparse Autoencoder Features","created":"2026-04-26","cited_by_count":0,"area":"mechinterp"},{"id":"2604.25167","title":"From Insight to Action: A Novel Framework for Interpretability-Guided Data Selection in Large Language Models","created":"2026-04-28","cited_by_count":0,"area":"mechinterp"},{"id":"2604.26866","title":"MoRFI: Monotonic Sparse Autoencoder Feature Identification","created":"2026-04-29","cited_by_count":0,"area":"mechinterp"},{"id":"2605.01609","title":"Concepts Whisper: Spectral Anti-Concentration and the Dual Geometry of Transformer Representations","created":"2026-05-02","cited_by_count":0,"area":"mechinterp"},{"id":"2605.03160","title":"Steering grids for sparse-autoencoder features: when a top-context label names an activation regime rather than a causal axis","created":"2026-05-04","cited_by_count":0,"area":"mechinterp"},{"id":"2605.18808","title":"Compositional Literary Primitives in Instruction-Tuned LLMs: Cross-Architectural SAE Features for Self, Style, and Affect","created":"2026-05-11","cited_by_count":0,"area":"mechinterp"},{"id":"2605.28149","title":"Sign-Aware Gated Sparse Autoencoders: Modeling Anticorrelated Features with Bi-Jump-ReLU Activations","created":"2026-05-27","cited_by_count":0,"area":"mechinterp"},{"id":"2605.28649","title":"Interpretability-Guided Layer Selection over Subspace Projection: SAEs as Stethoscopes, Not Scalpels, for Raw Task Vector Model Editing","created":"2026-05-27","cited_by_count":0,"area":"mechinterp"},{"id":"2606.08365","title":"Pre-Intervention Prediction of Sparse Autoencoder Steering Side Effects","created":"2026-06-06","cited_by_count":0,"area":"mechinterp"},{"id":"2606.14040","title":"Decompose Sparsely Where You Should, Absorb Densely Where You Should No","created":"2026-06-12","cited_by_count":0,"area":"mechinterp"},{"id":"2606.26620","title":"Discovering Millions of Interpretable Features with Sparse Autoencoders","created":"2026-06-25","cited_by_count":0,"area":"mechinterp"},{"id":"2607.25907","title":"Minimizing Targeted Activations: Input-Only Suppression of Evaluation-Awareness Latents in Large Language Models","created":"2026-07-28","cited_by_count":0,"area":"mechinterp"},{"id":"2608.10172","title":"Intrinsic Structure: Spectral Identifiability for Mechanistic Interpretability","created":"2026-08-10","cited_by_count":0,"area":"mechinterp"},{"id":"2608.11197","title":"Beyond a Bag of Features: Set-Level Instability in Sparse Autoencoders","created":"2026-08-11","cited_by_count":0,"area":"mechinterp"},{"id":"2608.26136","title":"Reward-Informed Sparse Autoencoders and the Solution-Completeness Confound","created":"2026-06-25","cited_by_count":0,"area":"mechinterp"},{"id":"2609.08173","title":"Key Path Identification for Resolving Knowledge Conflicts via SAE-based Steering","created":"2026-09-08","cited_by_count":0,"area":"mechinterp"},{"id":"2609.08322","title":"Tracing Stereotypes from Representation to Output in Multilingual LLMs","created":"2026-09-08","cited_by_count":0,"area":"mechinterp"},{"id":"2609.15533","title":"The Misery of Mechanistic Interpretability: A Formal Perspective","created":"2026-09-14","cited_by_count":0,"area":"mechinterp"},{"id":"2609.32469","title":"PULSE: Identifying Demonstration-Utility Features with Sparse Autoencoders","created":"2026-09-26","cited_by_count":0,"area":"mechinterp"},{"id":"2609.34326","title":"Routing Without Embeddings: Fast And Interpretable Routing With Regular Expressions","created":"2026-09-28","cited_by_count":0,"area":"mechinterp"},{"id":"2609.36294","title":"When Trees Are Not Enough: Learning Mixed-Topology Feature Graphs with Adaptive Graph Sparse Autoencoders","created":"2026-09-28","cited_by_count":0,"area":"mechinterp"},{"id":"2610.04905","title":"ScopeSAE: Model-Scope Feature Discovery with Interpretable Layer Selection","created":"2026-10-04","cited_by_count":0,"area":"mechinterp"},{"id":"2610.19636","title":"Sparse Autoencoders Are a Shared Coordinate System for Symbolic Matching","created":"2026-10-14","cited_by_count":0,"area":"mechinterp"},{"id":"2610.20139","title":"Sparse Autoencoders for Mechanistic Interpretability in NLP: A Critical Survey of Recoverability, Interpretability, and Utility","created":"2026-10-14","cited_by_count":0,"area":"mechinterp"}],"next_cursor":null}