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

    Neuronpedia

    INDEX
    Explanations

    Okay, let's break down

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    很好的
    0.24
     mixture
    0.24
     кор
    0.23
     veces
    0.23
     kupu
    0.23
    പക്ഷ
    0.23
     ballpark
    0.23
    pathetic
    0.22
    കം
    0.22
    䇞
    0.22
    POSITIVE LOGITS
    Ř
    0.25
     having
    0.25
     accessing
    0.24
     THIS
    0.23
    యాన్ని
    0.23
     waarin
    0.23
     iemand
    0.22
    ኖ
    0.22
     any
    0.22
    阎
    0.22
    Act Density 0.127%

    No Known Activations