© 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. 100043
    Prev
    Next
    INDEX
    Explanations

    code comments and copyright notices

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

    closing tags or brackets in code, particularly in copyright headers and XML/markup documents.

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

    The neuron strongly activates on the literal “>” token—that is, on the closing bracket of XML/HTML‐style tags.

    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
     on
    -2.89
     in
    -2.69
     that
    -2.58
     or
    -2.50
     for
    -2.45
     as
    -2.41
     with
    -2.34
     This
    -2.34
     at
    -2.33
     not
    -2.33
    POSITIVE LOGITS
     pulseira
    2.11
     própri
    2.09
    饃
    2.06
     hemd
    2.00
     valla
    1.98
     alho
    1.98
     sprüche
    1.95
     zoude
    1.94
     klien
    1.92
    乁
    1.92
    Activations Density 0.176%

    No Known Activations