{"total":80,"rows":[{"id":"2404.03592","title":"ReFT: Representation Finetuning for Language Models","created":"2024-04-04","cited_by_count":197,"area":null},{"id":"2402.10376","title":"Interpreting CLIP with Sparse Linear Concept Embeddings (SpLiCE)","created":"2024-02-16","cited_by_count":119,"area":null},{"id":"2405.00208","title":"A Primer on the Inner Workings of Transformer-based Language Models","created":"2024-04-30","cited_by_count":85,"area":null},{"id":"2502.12892","title":"Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models","created":"2025-02-18","cited_by_count":73,"area":"mechinterp"},{"id":"2406.16778","title":"Finding Transformer Circuits with Edge Pruning","created":"2024-06-24","cited_by_count":60,"area":null},{"id":"2511.13653","title":"Weight-sparse transformers have interpretable circuits","created":"2025-11-17","cited_by_count":44,"area":"mechinterp"},{"id":"2406.09519","title":"Talking Heads: Understanding Inter-layer Communication in Transformer Language Models","created":"2024-06-13","cited_by_count":41,"area":null},{"id":"2507.08802","title":"The Non-Linear Representation Dilemma: Is Causal Abstraction Enough for Mechanistic Interpretability?","created":"2025-07-11","cited_by_count":37,"area":"mechinterp"},{"id":"2502.20914","title":"Everything, Everywhere, All at Once: Is Mechanistic Interpretability Identifiable?","created":"2025-02-28","cited_by_count":34,"area":"mechinterp"},{"id":"2504.11695","title":"Interpreting the linear structure of vision-language model embedding spaces","created":"2025-04-16","cited_by_count":32,"area":"mllm"},{"id":"2403.19521","title":"Interpreting Key Mechanisms of Factual Recall in Transformer-Based Language Models","created":"2024-03-28","cited_by_count":31,"area":null},{"id":"2409.09951","title":"Optimal ablation for interpretability","created":"2024-09-16","cited_by_count":25,"area":null},{"id":"2407.14662","title":"Relational Composition in Neural Networks: A Survey and Call to Action","created":"2024-07-19","cited_by_count":23,"area":null},{"id":"2407.08734","title":"Transformer Circuit Faithfulness Metrics are not Robust","created":"2024-07-11","cited_by_count":22,"area":"mechinterp"},{"id":"2507.22928","title":"How does Chain of Thought Think? Mechanistic Interpretability of Chain-of-Thought Reasoning with Sparse Autoencoding","created":"2025-07-24","cited_by_count":19,"area":"lr"},{"id":"2401.12631","title":"A Reply to Makelov et al. (2023)'s \"Interpretability Illusion\" Arguments","created":"2024-01-23","cited_by_count":17,"area":null},{"id":"2505.20254","title":"Position: Mechanistic Interpretability Should Prioritize Feature Consistency in SAEs","created":"2025-05-26","cited_by_count":14,"area":"mechinterp"},{"id":"2406.15955","title":"Beyond the Doors of Perception: Vision Transformers Represent Relations Between Objects","created":"2024-06-22","cited_by_count":13,"area":null},{"id":"2512.18792","title":"The Dead Salmons of AI Interpretability","created":"2025-12-21","cited_by_count":12,"area":null},{"id":"2411.10950","title":"Understanding Multimodal LLMs: the Mechanistic Interpretability of Llava in Visual Question Answering","created":"2024-11-17","cited_by_count":11,"area":"mllm"},{"id":"2602.15293","title":"The Information Geometry of Softmax: Probing and Steering","created":"2026-02-17","cited_by_count":11,"area":null},{"id":"2601.20834","title":"Linear representations in language models can change dramatically over a conversation","created":"2026-01-28","cited_by_count":10,"area":null},{"id":"2507.06445","title":"Can Interpretation Predict Behavior on Unseen Data?","created":"2025-07-08","cited_by_count":8,"area":null},{"id":"2508.01916","title":"Decomposing Representation Space into Interpretable Subspaces with Unsupervised Learning","created":"2025-08-03","cited_by_count":8,"area":"mechinterp"},{"id":"2511.04638","title":"Addressing divergent representations from causal interventions on neural networks","created":"2025-11-06","cited_by_count":7,"area":"mechinterp"},{"id":"2506.18852","title":"Mechanistic Interpretability Needs Philosophy","created":"2025-06-23","cited_by_count":6,"area":"mechinterp"},{"id":"2604.02608","title":"Steerable but Not Decodable: Function Vectors Operate Beyond the Logit Lens","created":"2026-04-03","cited_by_count":6,"area":"mechinterp"},{"id":"2407.03859","title":"Anthropocentric bias in language model evaluation","created":"2024-07-04","cited_by_count":4,"area":null},{"id":"2410.00340","title":"Sparse Attention Decomposition Applied to Circuit Tracing","created":"2024-10-01","cited_by_count":4,"area":null},{"id":"2603.14833","title":"Ablate and Rescue: A Causal Analysis of Residual Stream Hyper-Connections","created":"2026-03-16","cited_by_count":4,"area":null},{"id":"2607.08349","title":"Certified Interventional Fidelity: Anytime-Valid, Adaptive Evaluation of Causal Claims in Mechanistic Interpretability","created":"2026-07-09","cited_by_count":4,"area":"mechinterp"},{"id":"2501.06164","title":"Model Alignment Search","created":"2025-01-10","cited_by_count":3,"area":"alignment"},{"id":"2502.11806","title":"Exploring the Translation Mechanism of Large Language Models","created":"2025-02-17","cited_by_count":3,"area":null},{"id":"2605.18830","title":"In-Context Learning Operates as Concept Subspace Learning","created":"2026-05-12","cited_by_count":3,"area":null},{"id":"2410.18527","title":"Probing Ranking LLMs: A Mechanistic Analysis for Information Retrieval","created":"2024-10-24","cited_by_count":2,"area":null},{"id":"2411.10019","title":"Towards Utilising a Range of Neural Activations for Comprehending Representational Associations","created":"2024-11-15","cited_by_count":2,"area":null},{"id":"2505.01372","title":"Evaluating Explanations: An Explanatory Virtues Framework for Mechanistic Interpretability -- The Strange Science Part I.ii","created":"2025-05-02","cited_by_count":2,"area":"mechinterp"},{"id":"2511.00177","title":"Can SAEs reveal and mitigate racial biases of LLMs in healthcare?","created":"2025-10-31","cited_by_count":2,"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":"2602.22600","title":"Transformers converge to invariant algorithmic cores","created":"2026-02-26","cited_by_count":1,"area":null},{"id":"2603.20381","title":"The production of meaning in the processing of natural language","created":"2026-03-20","cited_by_count":1,"area":"mechinterp"},{"id":"2605.02234","title":"Bucketing the Good Apples: A Method for Diagnosing and Improving Causal Abstraction","created":"2026-05-04","cited_by_count":1,"area":null},{"id":"2605.08012","title":"Position: Mechanistic Interpretability Must Disclose Identification Assumptions for Causal Claims","created":"2026-05-08","cited_by_count":1,"area":null},{"id":"2605.09195","title":"The Geometry of Forgetting: Temporal Knowledge Drift as an Independent Axis in LLM Representations","created":"2026-05-09","cited_by_count":1,"area":null},{"id":"2607.20469","title":"DecodeShare: Tracing the Shared Subspace of LLM Decode-Time Decisions","created":"2026-05-21","cited_by_count":1,"area":"inference"},{"id":"2608.03620","title":"A Theory of Conditional Collapse under Low-Rank Weight-Space Ablations: I. The Single-Block Theory and Synthetic Validation","created":"2026-08-04","cited_by_count":1,"area":"mechinterp"},{"id":"2608.04183","title":"Test, then Route: How Language Models Execute In-Context Conditional Rules Across Models and Languages","created":"2026-08-04","cited_by_count":1,"area":null},{"id":"2608.06409","title":"Separating Decision-Rule Misalignment from Readout-Coverage Limitations in Speech Language Models","created":"2026-08-03","cited_by_count":1,"area":null},{"id":"2608.29458","title":"Reference-Grafting Matches Fine-Tuning at Eliciting Sandbagged Capabilities","created":"2026-08-29","cited_by_count":1,"area":null},{"id":"2610.05295","title":"Readable Before Actionable: Causal Tracing of Indirect Prompt Injection","created":"2026-10-04","cited_by_count":1,"area":"agents"},{"id":"2506.11088","title":"One SPACE to Rule Them All: Jointly Mitigating Factuality and Faithfulness Hallucinations in LLMs","created":"2025-06-05","cited_by_count":0,"area":null},{"id":"2512.06681","title":"Mechanistic Interpretability of GPT-2: Lexical and Contextual Layers in Sentiment Analysis","created":"2025-12-07","cited_by_count":0,"area":"mechinterp"},{"id":"2604.19083","title":"ProjLens: Unveiling the Role of Projectors in Multimodal Model Safety","created":"2026-04-21","cited_by_count":0,"area":"mllm"},{"id":"2605.07284","title":"Same Targets, Different Computation: How Post-Training Divides Work Across Model Layers","created":"2026-05-08","cited_by_count":0,"area":null},{"id":"2606.00926","title":"Refit the Probe: Single-Direction Ablation Is Not a Necessity Test","created":"2026-05-30","cited_by_count":0,"area":null},{"id":"2606.08292","title":"Necessary, Decodable and Reversible, Yet Not Transferable: A Stress Test for Attention-Head Role Claims","created":"2026-06-06","cited_by_count":0,"area":null},{"id":"2607.04167","title":"Geometry of Ordinal Representations in Language Models","created":"2026-07-05","cited_by_count":0,"area":null},{"id":"2607.17138","title":"Denoising Models Develop Human-Like Perceptual Illusion Representations Across Architectures","created":"2026-07-19","cited_by_count":0,"area":null},{"id":"2607.19618","title":"Causal dictionary learning reveals and validates transcription-factor binding features in genomic language models","created":"2026-07-21","cited_by_count":0,"area":null},{"id":"2607.28672","title":"LAWFUL: Law-Aligned Witness for Faithful Use of Latents","created":"2026-07-26","cited_by_count":0,"area":null},{"id":"2608.14717","title":"Local Gains and Fixed-Assignment Set Losses in Shared Set Decoders","created":"2026-08-12","cited_by_count":0,"area":null},{"id":"2608.15022","title":"Gathered, Not Admitted: How Attention Brings a Latent Variable into Verbalizable Form","created":"2026-08-15","cited_by_count":0,"area":null},{"id":"2608.21582","title":"Reading the Room: Implicit Confusion Encoding in Recurrent World Model States","created":"2026-08-21","cited_by_count":0,"area":"wm"},{"id":"2609.01106","title":"Hints Help But Do They Teach? Evaluating Skills Transfer in Code Generation","created":"2026-09-01","cited_by_count":0,"area":null},{"id":"2609.09902","title":"Contrastive Projection: Reading Transformer Internals by Differencing Logit Lenses","created":"2026-09-09","cited_by_count":0,"area":"mechinterp"},{"id":"2609.14754","title":"Calibrating Interpretability Instruments Before Trusting Their Verdicts","created":"2026-09-13","cited_by_count":0,"area":null},{"id":"2609.15980","title":"A Chosen Future Can Still Be Rewritten: Causal Writability in Video Models","created":"2026-09-14","cited_by_count":0,"area":null},{"id":"2609.22850","title":"Same Outcome, Different Readout: What Does a Steerable Valence Direction in LLMs Represent?","created":"2026-09-19","cited_by_count":0,"area":null},{"id":"2609.28150","title":"Exact Feedback Is Not Control: Evaluating Text-based Closed-Loop Revision in LLMs","created":"2026-09-23","cited_by_count":0,"area":null},{"id":"2609.32670","title":"What Would Falsify It? A Variable Specific Evidence Standard for Mechanistic Claims About Self Explanation","created":"2026-09-26","cited_by_count":0,"area":null},{"id":"2609.32942","title":"The Key Handoff: Retrieval in Hybrid Language Models","created":"2026-09-26","cited_by_count":0,"area":null},{"id":"2609.33298","title":"Direct Hidden-State Alignment: Mapping and Controlling Preference Expression in LLMs","created":"2026-09-27","cited_by_count":0,"area":"alignment"},{"id":"2609.36570","title":"CounterSteer: Suppressing Indirect Prompt Injection with Activation Steering","created":"2026-09-29","cited_by_count":0,"area":"agents"},{"id":"2609.39243","title":"Right Answer, Wrong Mechanism: Detecting Pernicious Divergence in Causal Interventions","created":"2026-09-30","cited_by_count":0,"area":"mechinterp"},{"id":"2610.01165","title":"Persistent Depth Ordering amid Shifting Block-Bypass Responses in Language Model Pretraining","created":"2026-10-01","cited_by_count":0,"area":null},{"id":"2610.02098","title":"Are We Recovering Mechanisms? Objective-Level Recovery Gaps in Mechanistic Interpretability","created":"2026-10-01","cited_by_count":0,"area":"mechinterp"},{"id":"2610.07798","title":"Thin Evidence, Thick Priors: How Language Models Substitute Identity for Missing Financial Facts","created":"2026-10-06","cited_by_count":0,"area":null},{"id":"2610.17117","title":"Gluing Local Contexts into Global Meaning: A Sheaf-Theoretic Decomposition of Transformer Representations","created":"2026-10-12","cited_by_count":0,"area":null},{"id":"2610.19359","title":"Steering Vectors for Competing Ethical Frameworks Are Interchangeable: A Competing-Target Test for Activation Steering","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}