© 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. 34-GEMMASCOPE-RES-131K
    7. 58106
    Prev
    Next
    INDEX
    Explanations

    ok followed by text

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

    The neuron strongly activates on occurrences of the substring “ok” (e.g. the standalone token “OK” or “Ok” within words).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    getLocation
    -0.89
    ques
    -0.85
    älv
    -0.82
     Wettbewerb
    -0.82
     الأمن
    -0.81
    ırken
    -0.77
    utzt
    -0.77
    Hoje
    -0.73
    構
    -0.70
    合
    -0.69
    POSITIVE LOGITS
    lahoma
    1.08
     Ok
    1.07
    ok
    0.96
     boomer
    0.92
     økonom
    0.88
     OK
    0.88
    ays
    0.87
    췌
    0.85
    Ok
    0.85
     ок
    0.82
    Activations Density 0.021%

    No Known Activations