© 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-4-31B
    gemma-4-31b
    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 3
    Layer 5
    Layer 23
    Layer 28
    Layer 41
    Layer 59
    Select Head Manually
    Layer
    Head Index

    Search Explanations

    Search via Inference

    Inference is not currently not enabled for this model.

    Browse

    Features in GEMMA-4-31B@30-res-matryoshka-131k
    1. Hover over a feature on the left to preview its details.
    2. Click a feature to lock it and interact with it.