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

    Neuronpedia

    INDEX
    Explanations

    vivid descriptions and details

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ühe
    0.64
     animating
    0.62
     protein
    0.61
     increased
    0.60
    vised
    0.60
    gency
    0.60
    abouts
    0.60
     accrued
    0.59
     $=
    0.58
     aching
    0.58
    POSITIVE LOGITS
    ו
    0.80
    و
    0.77
    };
    0.74
    {
    0.72
    其他
    0.68
    ا
    0.67
    /
    0.66
    _
    0.64
    0.64
    رك
    0.64
    Act Density 0.001%

    No Known Activations