Author Search Results :: Library Catalog

1

Language Models Implement Simple Word2Vec-style Vector Arithmetic Autoría: Merullo, Jack, Eickhoff, Carsten, Pavlick, Ellie

Published 2023

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
2

Circuit Component Reuse Across Tasks in Transformer Language Models Autoría: Merullo, Jack, Eickhoff, Carsten, Pavlick, Ellie

Published 2023

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
3

Axiomatic Causal Interventions for Reverse Engineering Relevance Computation in Neural Retrieval Models Autoría: Chen, Catherine, Merullo, Jack, Eickhoff, Carsten

Published 2024

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
4

Talking Heads: Understanding Inter-layer Communication in Transformer Language Models Autoría: Merullo, Jack, Eickhoff, Carsten, Pavlick, Ellie

Published 2024

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
5

I Have No Mouth, and I Must Rhyme: Uncovering Internal Phonetic Representations in LLaMA 3.2 Autoría: McLaughlin, Oliver, Khurana, Arjun, Merullo, Jack

Published 2025

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
6

Transferring Linear Features Across Language Models With Model Stitching Autoría: Chen, Alan, Merullo, Jack, Stolfo, Alessandro, Pavlick, Ellie

Published 2025

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
7

From Memorization to Reasoning in the Spectrum of Loss Curvature Autoría: Merullo, Jack, Vatsavaya, Srihita, Bushnaq, Lucius, Lewis, Owen

Published 2025

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
8

Transformer Mechanisms Mimic Frontostriatal Gating Operations When Trained on Human Working Memory Tasks Autoría: Traylor, Aaron, Merullo, Jack, Frank, Michael J., Pavlick, Ellie

Published 2024

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
9

Dual Process Learning: Controlling Use of In-Context vs. In-Weights Strategies with Weight Forgetting Autoría: Anand, Suraj, Lepori, Michael A., Merullo, Jack, Pavlick, Ellie

Published 2024

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
10

On Linear Representations and Pretraining Data Frequency in Language Models Autoría: Merullo, Jack, Smith, Noah A., Wiegreffe, Sarah, Elazar, Yanai

Published 2025

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
11

Does CLIP Bind Concepts? Probing Compositionality in Large Image Models Autoría: Lewis, Martha, Nayak, Nihal V., Yu, Peilin, Yu, Qinan, Merullo, Jack, Bach, Stephen H., Pavlick, Ellie

Published 2022

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
12

$100K or 100 Days: Trade-offs when Pre-Training with Academic Resources Autoría: Khandelwal, Apoorv, Yun, Tian, Nayak, Nihal V., Merullo, Jack, Bach, Stephen H., Sun, Chen, Pavlick, Ellie

Published 2024

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
13

Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought Autoría: Boppana, Siddharth, Ma, Annabel, Loeffler, Max, Sarfati, Raphael, Bigelow, Eric, Geiger, Atticus, Lewis, Owen, Merullo, Jack

Published 2026

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
14

Features as Rewards: Scalable Supervision for Open-Ended Tasks via Interpretability Autoría: Prasad, Aaditya Vikram, Watts, Connor, Merullo, Jack, Gala, Dhruvil, Lewis, Owen, McGrath, Thomas, Lubana, Ekdeep Singh

Published 2026

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
15

Stories in Space: In-Context Learning Trajectories in Conceptual Belief Space Autoría: Bigelow, Eric, Sarfati, Raphaël, Wurgaft, Daniel, Lewis, Owen, McGrath, Thomas, Merullo, Jack, Geiger, Atticus, Lubana, Ekdeep Singh

Published 2026

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
16

The Shape of Beliefs: Geometry, Dynamics, and Interventions along Representation Manifolds of Language Models' Posteriors Autoría: Sarfati, Raphaël, Bigelow, Eric, Wurgaft, Daniel, Boppana, Siddharth, Merullo, Jack, Geiger, Atticus, Lewis, Owen, McGrath, Tom, Lubana, Ekdeep Singh

Published 2026

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
17

Do Sparse Autoencoders Capture Concept Manifolds? Autoría: Bhalla, Usha, Fel, Thomas, Rager, Can, Feucht, Sheridan, Haklay, Tal, Wurgaft, Daniel, Boppana, Siddharth, Kowal, Matthew, Shyam, Vasudev, Merullo, Jack, Geiger, Atticus, Lubana, Ekdeep Singh

Published 2026

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
18

Arithmetic in the Wild: Llama uses Base-10 Addition to Reason About Cyclic Concepts Autoría: Feucht, Sheridan, Haklay, Tal, Bhalla, Usha, Wurgaft, Daniel, Rager, Can, Sarfati, Raphaël, Merullo, Jack, McGrath, Thomas, Lewis, Owen, Lubana, Ekdeep Singh, Fel, Thomas, Geiger, Atticus

Published 2026

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:
19

Manifold Steering Reveals the Shared Geometry of Neural Network Representation and Behavior Autoría: Wurgaft, Daniel, Rager, Can, Kowal, Matthew, Shyam, Vasudev, Feucht, Sheridan, Bhalla, Usha, Haklay, Tal, Bigelow, Eric, Sarfati, Raphael, McGrath, Thomas, Lewis, Owen, Merullo, Jack, Goodman, Noah, Fel, Thomas, Geiger, Atticus, Lubana, Ekdeep Singh

Published 2026

Fuente: arXiv

Tipo de material: Preprint

Acceso al recurso
Enlace del recurso

Standalone Record
Save to List

Saved in:

Language Models Implement Simple Word2Vec-style Vector Arithmetic Autoría: Merullo, Jack, Eickhoff, Carsten, Pavlick, Ellie

Circuit Component Reuse Across Tasks in Transformer Language Models Autoría: Merullo, Jack, Eickhoff, Carsten, Pavlick, Ellie

Axiomatic Causal Interventions for Reverse Engineering Relevance Computation in Neural Retrieval Models Autoría: Chen, Catherine, Merullo, Jack, Eickhoff, Carsten

Talking Heads: Understanding Inter-layer Communication in Transformer Language Models Autoría: Merullo, Jack, Eickhoff, Carsten, Pavlick, Ellie

I Have No Mouth, and I Must Rhyme: Uncovering Internal Phonetic Representations in LLaMA 3.2 Autoría: McLaughlin, Oliver, Khurana, Arjun, Merullo, Jack

Transferring Linear Features Across Language Models With Model Stitching Autoría: Chen, Alan, Merullo, Jack, Stolfo, Alessandro, Pavlick, Ellie

From Memorization to Reasoning in the Spectrum of Loss Curvature Autoría: Merullo, Jack, Vatsavaya, Srihita, Bushnaq, Lucius, Lewis, Owen

Transformer Mechanisms Mimic Frontostriatal Gating Operations When Trained on Human Working Memory Tasks Autoría: Traylor, Aaron, Merullo, Jack, Frank, Michael J., Pavlick, Ellie

Dual Process Learning: Controlling Use of In-Context vs. In-Weights Strategies with Weight Forgetting Autoría: Anand, Suraj, Lepori, Michael A., Merullo, Jack, Pavlick, Ellie

On Linear Representations and Pretraining Data Frequency in Language Models Autoría: Merullo, Jack, Smith, Noah A., Wiegreffe, Sarah, Elazar, Yanai

Does CLIP Bind Concepts? Probing Compositionality in Large Image Models Autoría: Lewis, Martha, Nayak, Nihal V., Yu, Peilin, Yu, Qinan, Merullo, Jack, Bach, Stephen H., Pavlick, Ellie

$100K or 100 Days: Trade-offs when Pre-Training with Academic Resources Autoría: Khandelwal, Apoorv, Yun, Tian, Nayak, Nihal V., Merullo, Jack, Bach, Stephen H., Sun, Chen, Pavlick, Ellie

Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought Autoría: Boppana, Siddharth, Ma, Annabel, Loeffler, Max, Sarfati, Raphael, Bigelow, Eric, Geiger, Atticus, Lewis, Owen, Merullo, Jack

Features as Rewards: Scalable Supervision for Open-Ended Tasks via Interpretability Autoría: Prasad, Aaditya Vikram, Watts, Connor, Merullo, Jack, Gala, Dhruvil, Lewis, Owen, McGrath, Thomas, Lubana, Ekdeep Singh

Stories in Space: In-Context Learning Trajectories in Conceptual Belief Space Autoría: Bigelow, Eric, Sarfati, Raphaël, Wurgaft, Daniel, Lewis, Owen, McGrath, Thomas, Merullo, Jack, Geiger, Atticus, Lubana, Ekdeep Singh

The Shape of Beliefs: Geometry, Dynamics, and Interventions along Representation Manifolds of Language Models' Posteriors Autoría: Sarfati, Raphaël, Bigelow, Eric, Wurgaft, Daniel, Boppana, Siddharth, Merullo, Jack, Geiger, Atticus, Lewis, Owen, McGrath, Tom, Lubana, Ekdeep Singh

Do Sparse Autoencoders Capture Concept Manifolds? Autoría: Bhalla, Usha, Fel, Thomas, Rager, Can, Feucht, Sheridan, Haklay, Tal, Wurgaft, Daniel, Boppana, Siddharth, Kowal, Matthew, Shyam, Vasudev, Merullo, Jack, Geiger, Atticus, Lubana, Ekdeep Singh

Arithmetic in the Wild: Llama uses Base-10 Addition to Reason About Cyclic Concepts Autoría: Feucht, Sheridan, Haklay, Tal, Bhalla, Usha, Wurgaft, Daniel, Rager, Can, Sarfati, Raphaël, Merullo, Jack, McGrath, Thomas, Lewis, Owen, Lubana, Ekdeep Singh, Fel, Thomas, Geiger, Atticus

Search Results - Merullo, Jack

Language Models Implement Simple Word2Vec-style Vector Arithmetic Autoría: Merullo, Jack, Eickhoff, Carsten, Pavlick, Ellie

Circuit Component Reuse Across Tasks in Transformer Language Models Autoría: Merullo, Jack, Eickhoff, Carsten, Pavlick, Ellie

Axiomatic Causal Interventions for Reverse Engineering Relevance Computation in Neural Retrieval Models Autoría: Chen, Catherine, Merullo, Jack, Eickhoff, Carsten

Talking Heads: Understanding Inter-layer Communication in Transformer Language Models Autoría: Merullo, Jack, Eickhoff, Carsten, Pavlick, Ellie

I Have No Mouth, and I Must Rhyme: Uncovering Internal Phonetic Representations in LLaMA 3.2 Autoría: McLaughlin, Oliver, Khurana, Arjun, Merullo, Jack

Transferring Linear Features Across Language Models With Model Stitching Autoría: Chen, Alan, Merullo, Jack, Stolfo, Alessandro, Pavlick, Ellie

From Memorization to Reasoning in the Spectrum of Loss Curvature Autoría: Merullo, Jack, Vatsavaya, Srihita, Bushnaq, Lucius, Lewis, Owen

Transformer Mechanisms Mimic Frontostriatal Gating Operations When Trained on Human Working Memory Tasks Autoría: Traylor, Aaron, Merullo, Jack, Frank, Michael J., Pavlick, Ellie

Dual Process Learning: Controlling Use of In-Context vs. In-Weights Strategies with Weight Forgetting Autoría: Anand, Suraj, Lepori, Michael A., Merullo, Jack, Pavlick, Ellie

On Linear Representations and Pretraining Data Frequency in Language Models Autoría: Merullo, Jack, Smith, Noah A., Wiegreffe, Sarah, Elazar, Yanai

Does CLIP Bind Concepts? Probing Compositionality in Large Image Models Autoría: Lewis, Martha, Nayak, Nihal V., Yu, Peilin, Yu, Qinan, Merullo, Jack, Bach, Stephen H., Pavlick, Ellie

$100K or 100 Days: Trade-offs when Pre-Training with Academic Resources Autoría: Khandelwal, Apoorv, Yun, Tian, Nayak, Nihal V., Merullo, Jack, Bach, Stephen H., Sun, Chen, Pavlick, Ellie

Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought Autoría: Boppana, Siddharth, Ma, Annabel, Loeffler, Max, Sarfati, Raphael, Bigelow, Eric, Geiger, Atticus, Lewis, Owen, Merullo, Jack

Features as Rewards: Scalable Supervision for Open-Ended Tasks via Interpretability Autoría: Prasad, Aaditya Vikram, Watts, Connor, Merullo, Jack, Gala, Dhruvil, Lewis, Owen, McGrath, Thomas, Lubana, Ekdeep Singh

Stories in Space: In-Context Learning Trajectories in Conceptual Belief Space Autoría: Bigelow, Eric, Sarfati, Raphaël, Wurgaft, Daniel, Lewis, Owen, McGrath, Thomas, Merullo, Jack, Geiger, Atticus, Lubana, Ekdeep Singh

The Shape of Beliefs: Geometry, Dynamics, and Interventions along Representation Manifolds of Language Models' Posteriors Autoría: Sarfati, Raphaël, Bigelow, Eric, Wurgaft, Daniel, Boppana, Siddharth, Merullo, Jack, Geiger, Atticus, Lewis, Owen, McGrath, Tom, Lubana, Ekdeep Singh

Do Sparse Autoencoders Capture Concept Manifolds? Autoría: Bhalla, Usha, Fel, Thomas, Rager, Can, Feucht, Sheridan, Haklay, Tal, Wurgaft, Daniel, Boppana, Siddharth, Kowal, Matthew, Shyam, Vasudev, Merullo, Jack, Geiger, Atticus, Lubana, Ekdeep Singh

Arithmetic in the Wild: Llama uses Base-10 Addition to Reason About Cyclic Concepts Autoría: Feucht, Sheridan, Haklay, Tal, Bhalla, Usha, Wurgaft, Daniel, Rager, Can, Sarfati, Raphaël, Merullo, Jack, McGrath, Thomas, Lewis, Owen, Lubana, Ekdeep Singh, Fel, Thomas, Geiger, Atticus

Search Tools: