© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 73563
    Prev
    Next
    INDEX
    Explanations

    html head and body tags

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

    This neuron detects references to file‐inclusion or resource‐linking in code, i.e. tokens naming file paths, filenames, and extensions (e.g. include/import tags, “.html”, “.jsp”, “common”, “head”, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     zwanzig
    -1.29
     welchen
    -1.27
     is
    -1.18
     crucial
    -1.12
     renowned
    -1.11
     astounding
    -1.11
    /*!
    -1.06
    ⎼
    -1.06
     ワイン
    -1.05
    elp
    -1.03
    POSITIVE LOGITS
    erobic
    1.21
    Гі
    1.17
    ﾛ
    1.13
    崭
    1.13
    آیا
    1.12
    Tú
    1.07
     comité
    1.07
    فیلم
    1.06
     pañ
    1.06
    אם
    1.05
    Activations Density 0.008%

    No Known Activations