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

    Neuronpedia

    INDEX
    Explanations

    program fit

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ف
    0.52
    imension
    0.44
    收入
    0.43
     ڈیز
    0.42
     infectious
    0.40
    ပ်
    0.40
     Numbers
    0.40
    SubType
    0.39
     Maß
    0.39
    فاف
    0.39
    POSITIVE LOGITS
     uph
    0.49
    ardom
    0.48
     cốt
    0.44
     cenderung
    0.44
    .=
    0.43
     samot
    0.42
     основной
    0.42
    edo
    0.41
     W
    0.40
    ҝ
    0.40
    Act Density 0.009%

    No Known Activations