{"total":99,"rows":[{"id":"2307.05782","title":"Large Language Models","created":"2023-07-11","cited_by_count":780,"area":null},{"id":"2305.01610","title":"Finding Neurons in a Haystack: Case Studies with Sparse Probing","created":"2023-05-02","cited_by_count":334,"area":null},{"id":"2304.05969","title":"Localizing Model Behavior with Path Patching","created":"2023-04-12","cited_by_count":195,"area":null},{"id":"2306.17844","title":"The Clock and the Pizza: Two Stories in Mechanistic Explanation of Neural Networks","created":"2023-06-30","cited_by_count":156,"area":null},{"id":"2307.15771","title":"The Hydra Effect: Emergent Self-repair in Language Model Computations","created":"2023-07-28","cited_by_count":156,"area":null},{"id":"2402.14811","title":"Fine-Tuning Enhances Existing Mechanisms: A Case Study on Entity Tracking","created":"2024-02-22","cited_by_count":149,"area":null},{"id":"2411.00816","title":"CycleResearcher: Improving Automated Research via Automated Review","created":"2024-10-28","cited_by_count":148,"area":null},{"id":"2309.02390","title":"Explaining grokking through circuit efficiency","created":"2023-09-05","cited_by_count":111,"area":null},{"id":"2303.07971","title":"A Theory of Emergent In-Context Learning as Implicit Structure Induction","created":"2023-03-14","cited_by_count":86,"area":null},{"id":"2311.18817","title":"Dichotomy of Early and Late Phase Implicit Biases Can Provably Induce Grokking","created":"2023-11-30","cited_by_count":80,"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":"2407.15017","title":"Knowledge Mechanisms in Large Language Models: A Survey and Perspective","created":"2024-07-22","cited_by_count":51,"area":null},{"id":"2401.12874","title":"From Understanding to Utilization: A Survey on Explainability for Large Language Models","created":"2024-01-23","cited_by_count":48,"area":null},{"id":"2501.04697","title":"Grokking at the Edge of Numerical Stability","created":"2025-01-08","cited_by_count":46,"area":null},{"id":"2402.07321","title":"Summing Up the Facts: Additive Mechanisms Behind Factual Recall in LLMs","created":"2024-02-11","cited_by_count":42,"area":null},{"id":"2503.21073","title":"Shared Global and Local Geometry of Language Model Embeddings","created":"2025-03-27","cited_by_count":36,"area":null},{"id":"2311.07568","title":"Feature emergence via margin maximization: case studies in algebraic tasks","created":"2023-11-13","cited_by_count":33,"area":null},{"id":"2312.01429","title":"Transformers are uninterpretable with myopic methods: a case study with bounded Dyck grammars","created":"2023-12-03","cited_by_count":30,"area":null},{"id":"2310.16441","title":"Grokking in Linear Estimators -- A Solvable Model that Groks without Understanding","created":"2023-10-25","cited_by_count":25,"area":null},{"id":"2312.03656","title":"Interpretability Illusions in the Generalization of Simplified Models","created":"2023-12-06","cited_by_count":25,"area":null},{"id":"2312.06581","title":"Grokking Group Multiplication with Cosets","created":"2023-12-11","cited_by_count":24,"area":null},{"id":"2512.05117","title":"The Universal Weight Subspace Hypothesis","created":"2025-12-04","cited_by_count":24,"area":null},{"id":"2312.08550","title":"Harmonics of Learning: Universal Fourier Features Emerge in Invariant Networks","created":"2023-12-13","cited_by_count":23,"area":null},{"id":"2402.15175","title":"Unified View of Grokking, Double Descent and Emergent Abilities: A Perspective from Circuits Competition","created":"2024-02-23","cited_by_count":23,"area":null},{"id":"2407.04690","title":"Missed Causes and Ambiguous Effects: Counterfactuals Pose Challenges for Interpreting Neural Networks","created":"2024-07-05","cited_by_count":23,"area":null},{"id":"2311.12997","title":"Compositional Capabilities of Autoregressive Transformers: A Study on Synthetic, Interpretable Tasks","created":"2023-11-21","cited_by_count":22,"area":null},{"id":"2407.15421","title":"Planning in a recurrent neural network that plays Sokoban","created":"2024-07-22","cited_by_count":18,"area":null},{"id":"2305.14699","title":"Can Transformers Learn to Solve Problems Recursively?","created":"2023-05-24","cited_by_count":17,"area":null},{"id":"2307.01189","title":"Trainable Transformer in Transformer","created":"2023-07-03","cited_by_count":17,"area":null},{"id":"2412.07942","title":"Neural Scaling Laws Rooted in the Data Distribution","created":"2024-12-10","cited_by_count":15,"area":null},{"id":"2401.12947","title":"Transformer-Based Models Are Not Yet Perfect At Learning to Emulate Structural Recursion","created":"2024-01-23","cited_by_count":14,"area":null},{"id":"2402.16726","title":"Towards Empirical Interpretation of Internal Circuits and Properties in Grokked Transformers on Modular Polynomials","created":"2024-02-26","cited_by_count":14,"area":null},{"id":"2506.06489","title":"Alternating Gradient Flows: A Theory of Feature Learning in Two-layer Neural Networks","created":"2025-06-06","cited_by_count":14,"area":null},{"id":"2310.17639","title":"In-Context Learning Dynamics with Random Binary Sequences","created":"2023-10-26","cited_by_count":13,"area":null},{"id":"2504.13292","title":"Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model","created":"2025-04-17","cited_by_count":13,"area":null},{"id":"2305.02469","title":"The System Model and the User Model: Exploring AI Dashboard Design","created":"2023-05-04","cited_by_count":12,"area":null},{"id":"2311.04354","title":"Uncovering Intermediate Variables in Transformers using Circuit Probing","created":"2023-11-07","cited_by_count":12,"area":null},{"id":"2312.01648","title":"Characterizing Large Language Model Geometry Helps Solve Toxicity Detection and Generation","created":"2023-12-04","cited_by_count":12,"area":null},{"id":"2410.04489","title":"Grokking at the Edge of Linear Separability","created":"2024-10-06","cited_by_count":12,"area":null},{"id":"2602.03655","title":"Sequential Group Composition: A Window into the Mechanics of Deep Learning","created":"2026-02-03","cited_by_count":10,"area":null},{"id":"2502.04700","title":"EigenLoRAx: Recycling Adapters to Find Principal Subspaces for Resource-Efficient Adaptation and Inference","created":"2025-02-07","cited_by_count":9,"area":null},{"id":"2601.19791","title":"To Grok Grokking: Provable Grokking in Ridge Regression","created":"2026-01-27","cited_by_count":8,"area":null},{"id":"2305.12544","title":"Has It All Been Solved? Open NLP Research Questions Not Solved by Large Language Models","created":"2023-05-21","cited_by_count":7,"area":null},{"id":"2405.17420","title":"Survival of the Fittest Representation: A Case Study with Modular Addition","created":"2024-05-27","cited_by_count":7,"area":null},{"id":"2505.17101","title":"A quantitative analysis of semantic information in deep representations of text and images","created":"2025-05-21","cited_by_count":7,"area":null},{"id":"2603.13331","title":"The Norm-Separation Delay Law of Grokking: A First-Principles Theory of Delayed Generalization","created":"2026-03-05","cited_by_count":7,"area":null},{"id":"2310.10899","title":"Instilling Inductive Biases with Subnetworks","created":"2023-10-17","cited_by_count":6,"area":null},{"id":"2307.09550","title":"The semantic landscape paradigm for neural networks","created":"2023-07-18","cited_by_count":4,"area":null},{"id":"2405.17209","title":"How Do Transformers \"Do\" Physics? Investigating the Simple Harmonic Oscillator","created":"2024-05-23","cited_by_count":4,"area":null},{"id":"2409.10967","title":"Relative Representations: Topological and Geometric Perspectives","created":"2024-09-17","cited_by_count":4,"area":null},{"id":"2509.06931","title":"Learning words in groups: fusion algebras, tensor ranks and grokking","created":"2025-09-08","cited_by_count":4,"area":null},{"id":"2405.16658","title":"Acceleration of Grokking in Learning Arithmetic Operations via Kolmogorov-Arnold Representation","created":"2024-05-26","cited_by_count":3,"area":null},{"id":"2405.17479","title":"A rationale from frequency perspective for grokking in training neural network","created":"2024-05-24","cited_by_count":3,"area":null},{"id":"2406.16988","title":"MD tree: a model-diagnostic tree grown on loss landscape","created":"2024-06-24","cited_by_count":3,"area":null},{"id":"2502.12717","title":"Learning the symmetric group: large from small","created":"2025-02-18","cited_by_count":3,"area":null},{"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":"2511.12421","title":"From Black Box to Bijection: Interpreting Machine Learning to Build a Zeta Map Algorithm","created":"2025-11-16","cited_by_count":3,"area":null},{"id":"2512.25060","title":"On the geometry and topology of representations: the manifolds of modular addition","created":"2025-12-31","cited_by_count":3,"area":null},{"id":"2601.21150","title":"Can Neural Networks Learn Small Algebraic Worlds? An Investigation Into the Group-theoretic Structures Learned By Narrow Models Trained To Predict Group Operations","created":"2026-01-29","cited_by_count":3,"area":null},{"id":"2603.24746","title":"Grokking as a Falsifiable Finite-Size Transition","created":"2026-03-25","cited_by_count":3,"area":null},{"id":"2604.13082","title":"The Long Delay to Arithmetic Generalization: When Learned Representations Outrun Behavior","created":"2026-03-30","cited_by_count":3,"area":null},{"id":"2307.12941","title":"On Privileged and Convergent Bases in Neural Network Representations","created":"2023-07-24","cited_by_count":2,"area":null},{"id":"2406.03947","title":"Weight-based Decomposition: A Case for Bilinear MLPs","created":"2024-06-06","cited_by_count":2,"area":null},{"id":"2411.07467","title":"Machines and Mathematical Mutations: Using GNNs to Characterize Quiver Mutation Classes","created":"2024-11-12","cited_by_count":2,"area":null},{"id":"2602.19533","title":"Grokking Finite-Dimensional Algebra","created":"2026-02-23","cited_by_count":2,"area":null},{"id":"2603.07323","title":"Norm-Hierarchy Transitions in Representation Learning: When and Why Neural Networks Abandon Shortcuts","created":"2026-03-07","cited_by_count":2,"area":null},{"id":"2604.13123","title":"Spectral Entropy Collapse as a Phase Transition in Delayed Generalisation: An Interventional and Predictive Framework for Grokkin","created":"2026-04-13","cited_by_count":2,"area":null},{"id":"2605.27078","title":"Two Speeds of Learning: A Representation-Readout Decomposition of Grokking and Double Descent","created":"2026-05-26","cited_by_count":2,"area":null},{"id":"2606.12966","title":"Learned Structure Shows Up Earlier in the Task's Own Basis: A Controlled Look at a Synchronisation Precursor in Grokking","created":"2026-06-11","cited_by_count":2,"area":null},{"id":"2607.07066","title":"Multiplication Beyond Groups: Stratified Fourier Mechanisms in Transformer Circuits","created":"2026-07-08","cited_by_count":2,"area":null},{"id":"2608.07436","title":"Post-Grokking Collapse at the Representation-Readout Interface in Muon-Trained Transformers","created":"2026-08-07","cited_by_count":2,"area":null},{"id":"2403.05713","title":"tsGT: Stochastic Time Series Modeling With Transformer","created":"2024-03-08","cited_by_count":1,"area":null},{"id":"2405.07436","title":"Can Language Models Explain Their Own Classification Behavior?","created":"2024-05-13","cited_by_count":1,"area":null},{"id":"2410.16443","title":"Improving Neuron-level Interpretability with White-box Language Models","created":"2024-10-21","cited_by_count":1,"area":null},{"id":"2411.07071","title":"Universal Response and Emergence of Induction in LLMs","created":"2024-11-11","cited_by_count":1,"area":null},{"id":"2504.02064","title":"From Text to Graph: Leveraging Graph Neural Networks for Enhanced Explainability in NLP","created":"2025-04-02","cited_by_count":1,"area":null},{"id":"2602.04548","title":"Gradient Flow Through Diagram Expansions: Learning Regimes and Explicit Solutions","created":"2026-02-04","cited_by_count":1,"area":null},{"id":"2602.22600","title":"Transformers converge to invariant algorithmic cores","created":"2026-02-26","cited_by_count":1,"area":null},{"id":"2609.12259","title":"The Rank the Task Demands: A Causal Rank Law for Matrix Memories Trained on Group Composition","created":"2026-09-10","cited_by_count":1,"area":null},{"id":"2310.02430","title":"Episodic Memory Theory for the Mechanistic Interpretation of Recurrent Neural Networks","created":"2023-10-03","cited_by_count":0,"area":null},{"id":"2406.01365","title":"From Feature Visualization to Visual Circuits: Effect of Adversarial Model Manipulation","created":"2024-06-03","cited_by_count":0,"area":null},{"id":"2410.12391","title":"Tracking Universal Features Through Fine-Tuning and Model Merging","created":"2024-10-16","cited_by_count":0,"area":null},{"id":"2412.10898","title":"Exploring Grokking: Experimental and Mechanistic Investigations","created":"2024-12-14","cited_by_count":0,"area":null},{"id":"2510.01252","title":"GPT and Prejudice: A Sparse Approach to Understanding Learned Representations in Large Language Models","created":"2025-09-24","cited_by_count":0,"area":null},{"id":"2510.19427","title":"Revisiting the Relation Between Robustness and Universality","created":"2025-10-22","cited_by_count":0,"area":null},{"id":"2604.00324","title":"The Persistent Vulnerability of Aligned AI Systems","created":"2026-03-31","cited_by_count":0,"area":null},{"id":"2605.07407","title":"Emergent Symbolic Structure in Health Foundation Models: Extraction, Alignment, and Cross-Modal Transfer","created":"2026-05-08","cited_by_count":0,"area":null},{"id":"2606.03085","title":"Multi-component Causal Tracing in Large Language Models","created":"2026-06-02","cited_by_count":0,"area":null},{"id":"2606.17399","title":"The Discrete-Log Clock: How a Transformer Learns Modular Multiplication","created":"2026-06-16","cited_by_count":0,"area":null},{"id":"2607.05113","title":"Rating the Pitch, Not the Product: User Evaluations of LLMs Reflect Expectations More Than Performance","created":"2026-07-06","cited_by_count":0,"area":null},{"id":"2607.19379","title":"Bayesian Wind Tunnels for Model Selection","created":"2026-07-01","cited_by_count":0,"area":null},{"id":"2607.21491","title":"What, Where, and How: Disentangling the Roles of Task, Language, and Model in Code Model Representations","created":"2026-07-23","cited_by_count":0,"area":null},{"id":"2609.15289","title":"Math for AI safety: an invitation for mathematicians","created":"2026-09-14","cited_by_count":0,"area":null},{"id":"2609.17926","title":"Symmetry without a manifold: intrinsic dimension on orbits","created":"2026-09-15","cited_by_count":0,"area":null},{"id":"2609.20166","title":"Small Enough to Know Everything: The Fully-Enumerable Transformer as an Instrument for the Science of Delayed Generalization","created":"2026-08-03","cited_by_count":0,"area":null},{"id":"2609.29951","title":"Tracking States or Tracking Cosets? An Algebraic Account of Learned State Tracking","created":"2026-09-24","cited_by_count":0,"area":null},{"id":"2609.36079","title":"A Polyphonic Conception of AI Understanding","created":"2026-09-28","cited_by_count":0,"area":null},{"id":"2609.36221","title":"How Language Models Differ in Redistributing Attention-Head Activity Under Serial Demand","created":"2026-09-28","cited_by_count":0,"area":null},{"id":"2610.00647","title":"Group-Invariant Statistics Determine Embedding Geometry: Harmonic Analysis of Representations from Bach to the Night Sky","created":"2026-09-30","cited_by_count":0,"area":null}],"next_cursor":null}