David Chanin
Scholar

David Chanin

Google Scholar ID: xnGP_IcAAAAJ
University College London
Google Scholar↗
Citations & Impact
All-time
Citations
185
 
H-index
7
 
i10-index
4
 
Publications
11
 
Co-authors
4
list available
Contact
No contact links provided.
Publications
7 items
SynthSAEBench: Evaluating Sparse Autoencoders on Scalable Realistic Synthetic Data
2026
Cited
0
Biases in the Blind Spot: Detecting What LLMs Fail to Mention
2026
Cited
0
Sparse but Wrong: Incorrect L0 Leads to Incorrect Features in Sparse Autoencoders
2025
Cited
0
Feature Hedging: Correlated Features Break Narrow Sparse Autoencoders
2025
Cited
0
SAEBench: A Comprehensive Benchmark for Sparse Autoencoders in Language Model Interpretability
2025
Cited
2
A is for Absorption: Studying Feature Splitting and Absorption in Sparse Autoencoders
arXiv.org · 2024
Cited
29
Analyzing the Generalization and Reliability of Steering Vectors
arXiv.org · 2024
Cited
5
Resume (English only)
Co-authors
4 total
Joseph Isaac Bloom
Joseph Isaac Bloom
UK AI Safety Institute
Adrià Garriga-Alonso
Adrià Garriga-Alonso
Research Scientist, FAR AI
Anthony Hunter
Anthony Hunter
Professor of Artificial Intelligence, University College London
Oana-Maria Camburu
Oana-Maria Camburu
Assistant Professor at Imperial College London