© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 33094
    Prev
    Next
    INDEX
    Explanations

    auto-generated code or xmlns declarations

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

    The neuron strongly activates on recurring boilerplate tokens in generated source files (e.g. “This,” “code,” “Changes,” “function,” “xmlns”), so it’s detecting standard autogenerated‐code header elements.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     Roumanie
    -0.75
    cheidung
    -0.68
    ربية
    -0.66
     ministros
    -0.66
    lade
    -0.66
    Tables
    -0.65
     Shares
    -0.65
    Shares
    -0.64
     CHINA
    -0.64
     kesi
    -0.64
    POSITIVE LOGITS
    𝗁
    0.78
     sickly
    0.73
     lộ
    0.72
    enus
    0.69
    YW
    0.69
     underv
    0.69
    ogener
    0.69
     Livingstone
    0.68
    liet
    0.66
    འི་
    0.66
    Activations Density 0.076%

    No Known Activations