Image de couverture de livre

Linearity of Relation Decoding in Transformer Language Models

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Hernandez, Evan, Sharma, Arnab Sen, Haklay, Tal, Meng, Kevin, Wattenberg, Martin, Andreas, Jacob, Belinkov, Yonatan, Bau, David
Format:	Preprint
Publié:	2023
Sujets:	Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Position-aware Automatic Circuit Discovery
par: Haklay, Tal, et autres
Publié: (2025)

Fine-Tuning Enhances Existing Mechanisms: A Case Study on Entity Tracking
par: Prakash, Nikhil, et autres
Publié: (2024)

Language Models use Lookbacks to Track Beliefs
par: Prakash, Nikhil, et autres
Publié: (2025)

Locating and Editing Factual Associations in Mamba
par: Sharma, Arnab Sen, et autres
Publié: (2024)

REVS: Unlearning Sensitive Information in Language Models via Rank Editing in the Vocabulary Space
par: Ashuach, Tomer, et autres
Publié: (2024)

Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models
par: Marks, Samuel, et autres
Publié: (2024)

Large Language Models Generate Harmful Content Using a Distinct, Unified Mechanism
par: Orgad, Hadas, et autres
Publié: (2026)

Back Attention: Understanding and Enhancing Multi-Hop Reasoning in Large Language Models
par: Yu, Zeping, et autres
Publié: (2025)

ContraSim -- Analyzing Neural Representations Based on Contrastive Learning
par: Rahamim, Adir, et autres
Publié: (2023)

Function Vectors in Large Language Models
par: Todd, Eric, et autres
Publié: (2023)

Decomposing Query-Key Feature Interactions Using Contrastive Covariances
par: Lee, Andrew, et autres
Publié: (2026)

Measuring and Controlling Instruction (In)Stability in Language Model Dialogs
par: Li, Kenneth, et autres
Publié: (2024)

Backward Lens: Projecting Language Model Gradients into the Vocabulary Space
par: Katz, Shahar, et autres
Publié: (2024)

Elucidating Mechanisms of Demographic Bias in LLMs for Healthcare
par: Ahsan, Hiba, et autres
Publié: (2025)

Pitfalls in Evaluating Interpretability Agents
par: Haklay, Tal, et autres
Publié: (2026)

Arithmetic Without Algorithms: Language Models Solve Math With a Bag of Heuristics
par: Nikankin, Yaniv, et autres
Publié: (2024)

Have Faith in Faithfulness: Going Beyond Circuit Overlap When Finding Model Mechanisms
par: Hanna, Michael, et autres
Publié: (2024)

Inspecting and Editing Knowledge Representations in Language Models
par: Hernandez, Evan, et autres
Publié: (2023)

Concept-Best-Matching: Evaluating Compositionality in Emergent Communication
par: Carmeli, Boaz, et autres
Publié: (2024)

Linear Relational Decoding of Morphology in Language Models
par: Xia, Eric, et autres
Publié: (2025)

Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task
par: Li, Kenneth, et autres
Publié: (2022)

ReFACT: Updating Text-to-Image Models by Editing the Text Encoder
par: Arad, Dana, et autres
Publié: (2023)

Are formal and functional linguistic mechanisms dissociated in language models?
par: Hanna, Michael, et autres
Publié: (2025)

Measuring Chain of Thought Faithfulness by Unlearning Reasoning Steps
par: Tutek, Martin, et autres
Publié: (2025)

SAEs Are Good for Steering -- If You Select the Right Features
par: Arad, Dana, et autres
Publié: (2025)

Leveraging Prototypical Representations for Mitigating Social Bias without Demographic Information
par: Iskander, Shadi, et autres
Publié: (2024)

Planted in Pretraining, Swayed by Finetuning: A Case Study on the Origins of Cognitive Biases in LLMs
par: Itzhak, Itay, et autres
Publié: (2025)

Reasoning Models Know What's Important, and Encode It in Their Activations
par: Nikankin, Yaniv, et autres
Publié: (2026)

DEPTH: Discourse Education through Pre-Training Hierarchically
par: Bamberger, Zachary, et autres
Publié: (2024)

Follow the Flow: On Information Flow Across Textual Tokens in Text-to-Image Models
par: Kaplan, Guy, et autres
Publié: (2025)

Unified Concept Editing in Diffusion Models
par: Gandikota, Rohit, et autres
Publié: (2023)

Fast Forwarding Low-Rank Training
par: Rahamim, Adir, et autres
Publié: (2024)

Differentiable Faithfulness Alignment for Cross-Model Circuit Transfer
par: Shao, Shun, et autres
Publié: (2026)

Shared Global and Local Geometry of Language Model Embeddings
par: Lee, Andrew, et autres
Publié: (2025)

Growing a Tail: Increasing Output Diversity in Large Language Models
par: Shur-Ofry, Michal, et autres
Publié: (2024)

Confidence Regulation Neurons in Language Models
par: Stolfo, Alessandro, et autres
Publié: (2024)

Distinguishing Ignorance from Error in LLM Hallucinations
par: Simhi, Adi, et autres
Publié: (2024)

Constructing Benchmarks and Interventions for Combating Hallucinations in LLMs
par: Simhi, Adi, et autres
Publié: (2024)

Findings of the BlackboxNLP 2025 Shared Task: Localizing Circuits and Causal Variables in Language Models
par: Arad, Dana, et autres
Publié: (2025)

Will it Merge? On The Causes of Model Mergeability
par: Rahamim, Adir, et autres
Publié: (2026)