AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders ｜ Neuronpedia

Neuronpedia

APISteer SAE Evals Blog/PodcastNEW Slack Privacy & Terms Contact

AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders

pyvene.ai, The Stanford NLP Group

·github.com ↗

axbench

Jump To

Jump to Source/SAE

Jump to Feature

INDEX

Random Feature

Search Explanations

Browse

Features in GEMMA-2-9B-IT@20-axbench-reft-r1-res-16k

Hover over a feature on the left to preview its details.
Click a feature to lock it and interact with it.

© Neuronpedia 2025

Privacy & Terms Blog/Podcast GitHub Slack Twitter Contact