© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 9141
    Prev
    Next
    INDEX
    Explanations

    The neuron is triggered by standalone UI‐style action or menu labels (e.g. “Scroll,” “Balance,” “Bal…”)—in other words, it spots in-text interface commands or navigation/menu terms.

    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
     পর্যায়ে
    0.58
    zeta
    0.57
    counting
    0.55
     ds
    0.54
    otas
    0.54
    今回は
    0.54
     અમ
    0.54
    쏠
    0.54
    class
    0.53
     Buc
    0.52
    POSITIVE LOGITS
    🏆
    0.56
     নির্মিত
    0.55
     dış
    0.55
    ൂരി
    0.55
     чыга
    0.54
     dirigeants
    0.54
     karane
    0.53
    hesi
    0.52
     össze
    0.52
     iniciativas
    0.52
    Activations Density 0.000%

    No Known Activations