{"total":76,"rows":[{"id":"2404.14082","title":"Mechanistic Interpretability for AI Safety -- A Review","created":"2024-04-22","cited_by_count":422,"area":"mechinterp"},{"id":"2404.15255","title":"How to use and interpret activation patching","created":"2024-04-23","cited_by_count":214,"area":null},{"id":"2406.11944","title":"Transcoders Find Interpretable LLM Feature Circuits","created":"2024-06-17","cited_by_count":200,"area":"mechinterp"},{"id":"2404.16014","title":"Improving Dictionary Learning with Gated Sparse Autoencoders","created":"2024-04-24","cited_by_count":179,"area":"mechinterp"},{"id":"2406.19384","title":"The Remarkable Robustness of LLMs: Stages of Inference?","created":"2024-06-27","cited_by_count":146,"area":null},{"id":"2407.02646","title":"A Practical Review of Mechanistic Interpretability for Transformer-Based Language Models","created":"2024-07-02","cited_by_count":118,"area":"mechinterp"},{"id":"2403.00745","title":"AtP*: An efficient and scalable method for localizing LLM behaviour to components","created":"2024-03-01","cited_by_count":111,"area":"mechinterp"},{"id":"2401.12181","title":"Universal Neurons in GPT2 Language Models","created":"2024-01-22","cited_by_count":102,"area":"mechinterp"},{"id":"2401.07927","title":"Are self-explanations from Large Language Models faithful?","created":"2024-01-15","cited_by_count":101,"area":null},{"id":"2410.11414","title":"ReDeEP: Detecting Hallucination in Retrieval-Augmented Generation via Mechanistic Interpretability","created":"2024-10-15","cited_by_count":101,"area":"rag"},{"id":"2410.21272","title":"Arithmetic Without Algorithms: Language Models Solve Math With a Bag of Heuristics","created":"2024-10-28","cited_by_count":97,"area":"lr"},{"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":"2312.09230","title":"Successor Heads: Recurring, Interpretable Attention Heads In The Wild","created":"2023-12-14","cited_by_count":77,"area":null},{"id":"2409.03752","title":"Attention Heads of Large Language Models: A Survey","created":"2024-09-05","cited_by_count":67,"area":null},{"id":"2407.10827","title":"LLM Circuit Analyses Are Consistent Across Training and Scale","created":"2024-07-15","cited_by_count":63,"area":null},{"id":"2406.01563","title":"LoFiT: Localized Fine-tuning on LLM Representations","created":"2024-06-03","cited_by_count":57,"area":null},{"id":"2505.13898","title":"Do Language Models Use Their Depth Efficiently?","created":"2025-05-20","cited_by_count":50,"area":null},{"id":"2402.11655","title":"Competition of Mechanisms: Tracing How Language Models Handle Facts and Counterfactuals","created":"2024-02-18","cited_by_count":40,"area":null},{"id":"2403.05767","title":"Extending Activation Steering to Broad Skills and Multiple Behaviours","created":"2024-03-09","cited_by_count":40,"area":"mechinterp"},{"id":"2503.09543","title":"PolyPythias: Stability and Outliers across Fifty Language Model Pre-Training Runs","created":"2025-03-12","cited_by_count":35,"area":null},{"id":"2506.09944","title":"Query-Focused Retrieval Heads Improve Long-Context Reasoning and Re-ranking","created":"2025-06-11","cited_by_count":31,"area":"lr"},{"id":"2503.10996","title":"Taming Knowledge Conflicts in Language Models","created":"2025-03-14","cited_by_count":30,"area":null},{"id":"2405.16229","title":"No Two Devils Alike: Unveiling Distinct Mechanisms of Fine-tuning Attacks","created":"2024-05-25","cited_by_count":29,"area":"alignment"},{"id":"2410.09223","title":"The Same But Different: Structural Similarities and Differences in Multilingual Language Modeling","created":"2024-10-11","cited_by_count":28,"area":"mechinterp"},{"id":"2409.09951","title":"Optimal ablation for interpretability","created":"2024-09-16","cited_by_count":25,"area":null},{"id":"2502.11812","title":"Towards Understanding Fine-Tuning Mechanisms of LLMs via Circuit Analysis","created":"2025-02-17","cited_by_count":20,"area":null},{"id":"2502.20129","title":"Finite State Automata Inside Transformers with Chain-of-Thought: A Mechanistic Study on State Tracking","created":"2025-02-27","cited_by_count":18,"area":"lr"},{"id":"2405.12522","title":"Sparse Autoencoders Enable Scalable and Reliable Circuit Identification in Language Models","created":"2024-05-21","cited_by_count":15,"area":"mechinterp"},{"id":"2405.14992","title":"Linking In-context Learning in Transformers to Human Episodic Memory","created":"2024-05-23","cited_by_count":15,"area":null},{"id":"2406.01352","title":"Position: An Inner Interpretability Framework for AI Inspired by Lessons from Cognitive Neuroscience","created":"2024-06-03","cited_by_count":11,"area":null},{"id":"2409.13710","title":"You can remove GPT2's LayerNorm by fine-tuning","created":"2024-09-06","cited_by_count":11,"area":"mechinterp"},{"id":"2510.06036","title":"Refusal Falls off a Cliff: How Safety Alignment Fails in Reasoning?","created":"2025-10-07","cited_by_count":11,"area":"lr"},{"id":"2605.06510","title":"Is One Layer Enough? Understanding Inference Dynamics in Tabular Foundation Models","created":"2026-05-07","cited_by_count":10,"area":null},{"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":"2507.02559","title":"Transformers Don't Need LayerNorm at Inference Time: Scaling LayerNorm Removal to GPT-2 XL and the Implications for Mechanistic Interpretability","created":"2025-07-03","cited_by_count":7,"area":"mechinterp"},{"id":"2406.13131","title":"When Parts Are Greater Than Sums: Individual LLM Components Can Outperform Full Models","created":"2024-06-19","cited_by_count":6,"area":null},{"id":"2502.13946","title":"Why Safeguarded Ships Run Aground? Aligned Large Language Models' Safety Mechanisms Tend to Be Anchored in The Template Region","created":"2025-02-19","cited_by_count":6,"area":"alignment"},{"id":"2602.18458","title":"The Story is Not the Science: Execution-Grounded Evaluation of Mechanistic Interpretability Research","created":"2026-02-05","cited_by_count":6,"area":"mechinterp"},{"id":"2408.12664","title":"Multilevel Interpretability Of Artificial Neural Networks: Leveraging Framework And Methods From Neuroscience","created":"2024-08-22","cited_by_count":5,"area":null},{"id":"2505.17630","title":"Correcting Gradient-Based Circuit Localization via Interaction-Aware Backpropagation","created":"2025-05-23","cited_by_count":5,"area":null},{"id":"2601.05328","title":"Bi-Orthogonal Factor Decomposition for Vision Transformers","created":"2026-01-08","cited_by_count":5,"area":null},{"id":"2312.08793","title":"Forbidden Facts: An Investigation of Competing Objectives in Llama-2","created":"2023-12-14","cited_by_count":4,"area":null},{"id":"2407.12161","title":"Interpretability in Action: Exploratory Analysis of VPT, a Minecraft Agent","created":"2024-07-16","cited_by_count":4,"area":"agents"},{"id":"2608.23651","title":"Feedback That Backfires: Why Small Language Model Agents Repeat the Call They Just Watched Fail","created":"2026-08-24","cited_by_count":4,"area":"agents"},{"id":"2503.01896","title":"Neuroplasticity and Corruption in Model Mechanisms: A Case Study Of Indirect Object Identification","created":"2025-02-27","cited_by_count":3,"area":null},{"id":"2509.24169","title":"Task Vectors, Learned Not Extracted: Performance Gains and Mechanistic Insight","created":"2025-09-29","cited_by_count":3,"area":null},{"id":"2509.22947","title":"Induction Signatures Are Not Enough: A Matched-Compute Study of Load-Bearing Structure in In-Context Learning","created":"2025-09-26","cited_by_count":2,"area":null},{"id":"2511.12381","title":"Don't Think of the White Bear: Ironic Negation in Transformer Models Under Cognitive Load","created":"2025-11-15","cited_by_count":2,"area":"lr"},{"id":"2603.03335","title":"Compressed Sensing for Capability Localization in Large Language Models","created":"2026-02-11","cited_by_count":2,"area":"lr"},{"id":"2603.06592","title":"Hierarchical Latent Structures in Data Generation Process Unify Mechanistic Phenomena across Scale","created":"2026-02-04","cited_by_count":2,"area":"mechinterp"},{"id":"2605.09314","title":"How LLMs Are Persuaded: A Few Attention Heads, Rerouted","created":"2026-05-10","cited_by_count":2,"area":null},{"id":"2606.07560","title":"Function-Vector Heads Are Two Populations: Writers and Cancellers in In-Context Learning","created":"2026-05-25","cited_by_count":2,"area":null},{"id":"2608.19338","title":"Mechanistic Tomography: Designed Measurement for Control-Oriented Interpretability","created":"2026-08-19","cited_by_count":2,"area":"mechinterp"},{"id":"2409.19745","title":"PEAR: Position-Embedding-Agnostic Attention Re-weighting Enhances Retrieval-Augmented Generation with Zero Inference Overhead","created":"2024-09-29","cited_by_count":1,"area":"rag"},{"id":"2411.07071","title":"Universal Response and Emergence of Induction in LLMs","created":"2024-11-11","cited_by_count":1,"area":null},{"id":"2411.15281","title":"ElastiFormer: Learned Redundancy Reduction in Transformer via Self-Distillation","created":"2024-11-22","cited_by_count":1,"area":"lr"},{"id":"2502.20475","title":"Promote, Suppress, Iterate: How Language Models Answer One-to-Many Factual Queries","created":"2025-02-27","cited_by_count":1,"area":null},{"id":"2508.16109","title":"From Indirect Object Identification to Syllogisms: Exploring Binary Mechanisms in Transformer Circuits","created":"2025-08-22","cited_by_count":1,"area":"mechinterp"},{"id":"2510.09421","title":"On the Representations of Entities in Auto-regressive Large Language Models","created":"2025-10-10","cited_by_count":1,"area":null},{"id":"2605.12809","title":"Correcting Influence: Unboxing LLM Outputs with Orthogonal Latent Spaces","created":"2026-05-12","cited_by_count":1,"area":null},{"id":"2605.22870","title":"The Readout Shortcut: Positional Number Copying Dominates Arithmetic CoT Readout in Small Language Models","created":"2026-05-20","cited_by_count":1,"area":"lr"},{"id":"2606.28153","title":"Robust Harmful Features Under Jailbreak Attacks: Mechanistic Evidence from Attention Head Specialization in Large Language Models","created":"2026-06-26","cited_by_count":1,"area":"alignment"},{"id":"2609.03026","title":"ObserverBench: Testing Mechanistic Estimates for Intervention and Control","created":"2026-09-02","cited_by_count":1,"area":"mechinterp"},{"id":"2609.10210","title":"Through the Looking Glass: Directly Reading and Writing Transformers","created":"2026-09-09","cited_by_count":1,"area":null},{"id":"2407.03621","title":"The Mysterious Case of Neuron 1512: Injectable Realignment Architectures Reveal Internal Characteristics of Meta's Llama 2 Model","created":"2024-07-04","cited_by_count":0,"area":null},{"id":"2506.03434","title":"Time Course MechInterp: Analyzing the Evolution of Components and Knowledge in Large Language Models","created":"2025-06-03","cited_by_count":0,"area":null},{"id":"2602.17526","title":"The Anxiety of Influence: Bloom Filters in Transformer Attention Heads","created":"2026-02-19","cited_by_count":0,"area":null},{"id":"2602.22453","title":"Identifying Crucial Attention Heads for Multilingual Language Models: Retrieval and Retrieval-Transition Heads","created":"2026-02-25","cited_by_count":0,"area":null},{"id":"2603.12414","title":"SpectralGuard: Detecting Memory Collapse Attacks in State Space Models","created":"2026-03-12","cited_by_count":0,"area":"ssm"},{"id":"2604.27167","title":"How a Cooperative-Override Circuit Suppresses Nash Play in Large Language Models","created":"2026-04-29","cited_by_count":0,"area":"mechinterp"},{"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":"2607.01002","title":"Logit-Contribution Scoring Identifies Non-Literal Retrieval Heads","created":"2026-07-01","cited_by_count":0,"area":null},{"id":"2608.08536","title":"Can Graph Learning Learn Circuits?","created":"2026-08-09","cited_by_count":0,"area":"mechinterp"},{"id":"2609.32927","title":"The Geometry of Logic: Stratification Induces Semantic Structure and Robust Reasoning","created":"2026-09-26","cited_by_count":0,"area":"lr"},{"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":"2610.19113","title":"From Copying to Composition: How Language Models Develop In-Context Learning","created":"2026-10-13","cited_by_count":0,"area":null}],"next_cursor":null}