© 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
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 12737
    Prev
    Next
    INDEX
    Explanations

    br

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron responds strongly to the substring “Bru” (as in “de Bruin,” “Bruzzone,” “Bruges,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Judi
    0.64
    Avi
    0.54
    ionized
    0.53
    ి
    0.53
    hence
    0.52
    ခ
    0.51
    ieren
    0.50
    endale
    0.50
     डालने
    0.50
    ेंद्रीय
    0.50
    POSITIVE LOGITS
     br
    0.75
     braces
    0.59
     Brü
    0.59
    </table>
    0.57
     brute
    0.56
    嗟
    0.56
     brutal
    0.54
     bruto
    0.54
    로는
    0.53
    ⏪
    0.53
    Activations Density 0.067%

    No Known Activations