© 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-3-4B-IT
    gemma-3-4b-it
    Google Deepmind
    Jump to Source/SAE
    Jump to Feature
    INDEX

    Releases

    Gemma Scope 2: Suite of SAEs and Transcoders for Gemma 3
    December 2025
    Google DeepMind
    gemma-scope-2

    Attention Visualizer

    HeadVis (Luger, Kamath et al.)
    Find Head By Metric
    Metric & Number of Heads
    Top 8
    Click head to select
    Layer 5
    Layer 11
    Layer 23
    Layer 29
    Select Head Manually
    Layer
    Head Index

    Circuit Tracing

    Search Explanations

    Browse

    Features in GEMMA-3-4B-IT@17-gemmascope-2-res-16k
    1. Hover over a feature on the left to preview its details.
    2. Click a feature to lock it and interact with it.