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

    Neuronpedia

    INDEX
    Explanations

    Function)

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    လူ
    0.63
     सुप्रीम
    0.58
     cardinal
    0.58
    随
    0.58
    पिता
    0.57
     cmap
    0.55
     اغ
    0.55
     개
    0.55
     trainers
    0.53
     हरिद्वार
    0.53
    POSITIVE LOGITS
     sekä
    0.67
     Еще
    0.65
    简称
    0.65
     Pvt
    0.65
     kısmı
    0.62
     Edition
    0.61
     这个
    0.60
     (\"
    0.60
     heeft
    0.60
    谙
    0.59
    Act Density 1.551%

    No Known Activations