May 5, 202613 min read
Mapping the Concept Space Inside Language Models
How we used sparse features, auto-interpretability, and steering experiments to turn hidden activations into searchable, testable, and steerable signals.
Read article
TechnicalResearch
What we find inside.
Open research on mechanistic interpretability, internal model signals, and the foundations behind NeuronLens products.
Dec 14, 202513 min read
Training Sparse Autoencoders That Produce Useful Features: A Practical Guide
A practical guide to training sparse autoencoders across model families, covering frameworks, failure modes, evaluation metrics, and case studies.
Read article
Fundamental ResearchNov 22, 20256 min read
Labeling domain specific features of Sparse Autoencoders: Practical Strategies
From feature discovery to label fidelity in large language models (Nemotron + Llama case study)
Read article
Fundamental ResearchOct 7, 20259 min read
Debugging Tool Use in Agent Systems
Using internal activations to diagnose and fix tool-use failures in agentic AI systems.
Read article
Applied Research