© 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. 85560
    Prev
    Next
    INDEX
    Explanations

    refused to communicate or act

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

    The neuron detects verbs and phrases expressing refusal or declining (e.g., “declined,” “refused,” “denies,” “declines,” “refuses”).

    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
     bemerken
    -2.03
     všechny
    -2.03
     všetky
    -1.95
     urbanas
    -1.85
     motivadoras
    -1.81
    ،
    -1.80
     ошибка
    -1.78
     protestas
    -1.78
     geniales
    -1.77
    autogui
    -1.77
    POSITIVE LOGITS
     is
    1.74
    ↵
    1.71
    啻
    1.70
     necessarily
    1.66
     in
    1.64
    ла
    1.62
    $
    1.62
     jacobs
    1.53
     bauer
    1.53
    <i>
    1.49
    Activations Density 0.004%

    No Known Activations