© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129820
    Prev
    Next
    INDEX
    Explanations

    The neuron flags mentions of user‐interface component types (e.g. Layout, Manager, Control, List, Browser, Label, Button) in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    UI elements with trailing numbers or separators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    It
    -1.56
    '
    -1.47
     What
    -1.37
    mathrm
    -1.30
    {
    -1.24
     will
    -1.20
    As
    -1.19
     Sometimes
    -1.14
     There
    -1.13
    For
    -1.12
    POSITIVE LOGITS
     ARISING
    1.22
     sés
    1.20
    лыми
    1.16
    dcmdata
    1.15
    yatı
    1.15
    shadowOffset
    1.13
    haviours
    1.12
     MATTER
    1.11
    Напомним
    1.09
     fundación
    1.09
    Activations Density 0.001%

    No Known Activations