© 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. 84895
    Prev
    Next
    INDEX
    Explanations

    wikipedia, wiktionary, wiki

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

    The neuron strongly activates on mentions of “Wiki” or “Wikipedia” (i.e. the substring “wiki”).

    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
     and
    -2.44
     they
    -1.82
     but
    -1.80
     when
    -1.80
     because
    -1.77
     from
    -1.73
     for
    -1.73
     some
    -1.68
     with
    -1.66
     as
    -1.61
    POSITIVE LOGITS
    냑
    1.65
    킁
    1.61
    asties
    1.61
    쐈
    1.60
     MEANS
    1.52
    vocations
    1.52
     HAPPENED
    1.52
     autorisation
    1.52
     boletos
    1.51
     IRELAND
    1.50
    Activations Density 0.005%

    No Known Activations