© 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. 22-GEMMASCOPE-RES-131K
    7. 64925
    Prev
    Next
    INDEX
    Explanations

    configuration file structures

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

    The neuron strongly activates on standalone XML/HTML element names (i.e. tag identifiers like “dict,” “body,” “Triggers,” “property,” 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
     powinny
    -1.23
     powinno
    -1.18
     Acesso
    -1.16
    どのような
    -1.13
    centerTitle
    -1.09
    </tfoot>
    -1.09
    の記事
    -1.06
     vinho
    -1.04
     powin
    -1.03
     zdjęcie
    -1.02
    POSITIVE LOGITS
    𝖆
    1.09
    1.05
    irical
    1.01
    icoot
    1.01
    Appellee
    1.00
    vantaged
    0.98
    фин
    0.98
    upuncture
    0.98
    ,
    0.98
    Algérie
    0.97
    Activations Density 0.001%
    No Known Activations