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

    Neuronpedia

    INDEX
    Explanations

    fantasy races

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <unused99>
    0.77
    <unused423>
    0.68
    <unused286>
    0.66
    <unused641>
    0.66
    il
    0.65
    <unused635>
    0.65
    <unused995>
    0.65
     outerwear
    0.64
    <unused752>
    0.64
    pz
    0.64
    POSITIVE LOGITS
    -
    0.95
    -]
    0.65
     ktorí
    0.64
    കൊ
    0.62
    -'
    0.61
    시
    0.61
    ريم
    0.60
    {
    0.59
    ON
    0.59
    TH
    0.58
    Act Density 0.146%

    No Known Activations