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

    Neuronpedia

    INDEX
    Explanations

    dtype or cat

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     type
    1.95
    type
    1.79
     Type
    1.70
    Type
    1.60
     TYPE
    1.55
     tipo
    1.46
    TYPE
    1.35
     टाइप
    1.34
     типа
    1.33
     тип
    1.32
    POSITIVE LOGITS
    잡
    0.36
    ორ
    0.35
    擞
    0.34
    賃
    0.34
    擺
    0.34
     ১৫০
    0.34
    đu
    0.33
     unavoidable
    0.33
     Mã
    0.32
     asymmetries
    0.32
    Act Density 0.018%

    No Known Activations