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

    Neuronpedia

    INDEX
    Explanations

    assault vs other issues

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Unlike
    0.47
     atelier
    0.42
     Atelier
    0.40
     Different
    0.39
     Ketika
    0.39
     deutscher
    0.39
     apabila
    0.39
     dibandingkan
    0.37
     whanne
    0.37
     ইহাই
    0.36
    POSITIVE LOGITS
     наличие
    0.48
     просто
    0.46
    avanja
    0.43
    之类的
    0.41
     인해
    0.40
     отношении
    0.40
     embarazo
    0.40
     cuestiones
    0.40
     необходимость
    0.40
    加密
    0.39
    Act Density 0.007%

    No Known Activations