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

    Neuronpedia

    INDEX
    Explanations

    function return signatures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    에서도
    0.37
     rethink
    0.37
     NEWS
    0.36
    0.36
     hiatus
    0.36
     hands
    0.36
    ⑶
    0.36
     galore
    0.35
    ⓢ
    0.35
     plur
    0.35
    POSITIVE LOGITS
     соответствует
    0.49
    In
    0.42
     corresponds
    0.42
     می‌تواند
    0.42
     In
    0.41
    Tiene
    0.41
     ensures
    0.41
     می‌باشد
    0.39
    tiene
    0.39
     Represents
    0.39
    Act Density 0.014%

    No Known Activations