Agent skill · research science · nousresearch
saelens
Train sparse autoencoders to interpret model features.
Why this skill is useful
Provides domain-specific knowledge on training sparse autoencoders and their application in feature interpretability that the AI wouldn't reliably generate on its own.
What it needs
Requires sae-lens, torch, transformer-lens installed locally. About 7k tokens when loaded. Last updated 2026-08-07. 226,679 stars on the source repository.
What this skill does
SAELens: Sparse Autoencoders for Mechanistic Interpretability SAELens is the primary library for training and analyzing Sparse Autoencoders (SAEs) - a technique for decomposing polysemantic neural network activations into sparse, interpretable features. Based on Anthropic's groundbreaking research on monosemanticity. GitHub: jbloomAus/SAELens (1,100+ stars) The Problem: Polysemanticity & Superposition Individual neurons in neural networks are polysemantic - they activate in multiple, semantically distinct contexts. This happens because models use superposition to represent more features than they have neurons, making interpretability difficult. SAEs solve this by decomposing dense activations into sparse, monosemantic features - typically only a small number of features activate for any given input, and each feature corresponds to an interpretable concept. When to Use SAELens Use SAELens when you need to: Discover interpretable features in model activations Understand what concepts a model has learned Study superposition and feature geometry Perform feature-based steering or ablation Analyze safety-relevant features (deception, bias, harmful content) Consider alternatives when: You need basic activation analysis → Use TransformerLens directly You want causal intervention experiments → Use pyvene or TransformerLens You need production steering → Consider direct activation engineering Installation Requirements: Python 3.10+, transformer-lens>=2.0.0 Core Concepts What SAEs Learn SAEs are trained to reconstruct model activations through a sparse bottleneck: Loss Function: MSE(original, reconstructed) + L1coefficient × L1(features) Key Validation (Anthropic Research) In "Towards Monosemanticity", human evaluators found 70% of SAE features genuinely interpretable. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills nousresearch/saelens