© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    5. Gemma-4-E2B
    gemma-4-e4b
    Google Deepmind
    Jump to Source/SAE
    Jump to Feature
    INDEX

    Releases

    Gemma 4 SAEs
    June 2026
    David Chanin
    gemma-4-saes

    Attention Visualizer

    HeadVis (Luger, Kamath et al.)
    Find Head By Metric
    Metric & Number of Heads
    Top 8
    Click head to select
    Layer 2
    Layer 5
    Layer 23
    Layer 26
    Layer 29
    Layer 30
    Layer 34
    Layer 35
    Select Head Manually
    Layer
    Head Index

    Search Explanations

    Browse

    Features in GEMMA-4-E4B@21-matryoshka-res-65k
    1. Hover over a feature on the left to preview its details.
    2. Click a feature to lock it and interact with it.