© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 102622
    Prev
    Next
    INDEX
    Explanations

    XML tags and localization

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

    XML and structured markup tags, especially opening angle brackets in XML/HTML documents.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron responds strongly to XML/HTML‐style markup—angle brackets, tag names, and related structural tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     at
    -2.48
     on
    -2.17
     arctic
    -1.44
     over
    -1.37
     of
    -1.37
     and
    -1.36
     but
    -1.36
     también
    -1.32
     three
    -1.30
    钠
    -1.29
    POSITIVE LOGITS
     your
    2.50
    鳚
    1.88
     you
    1.78
    帮你
    1.66
    ただし
    1.61
     Inſ
    1.57
    ⃤
    1.54
    Ꭳ
    1.52
    为你
    1.48
     visant
    1.38
    Activations Density 0.003%

    No Known Activations