Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    ready or states like alone

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Bedeutung
    1.01
     pleases
    1.01
    away
    0.98
     offs
    0.96
     sigmoid
    0.93
     viento
    0.93
     pleasant
    0.93
     importance
    0.92
    ಪ್ಪು
    0.92
     remar
    0.91
    POSITIVE LOGITS
    ன
    1.62
    ن
    1.60
    स
    1.36
    ną
    1.23
    िफिकेट
    1.23
    ності
    1.19
    د
    1.17
    de
    1.16
    ب
    1.15
    ри
    1.13
    Act Density 0.091%

    No Known Activations