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

    Neuronpedia

    INDEX
    Explanations

    like introducing comparisons

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     дела
    0.46
     ske
    0.41
    从而
    0.40
     бара
    0.38
     caballos
    0.38
     akin
    0.38
     किये
    0.37
     spindles
    0.37
    হাম্ম
    0.36
     rams
    0.36
    POSITIVE LOGITS
     Suddenly
    0.49
     suddenly
    0.47
    kennt
    0.41
    上げた
    0.40
     SUD
    0.40
     unwittingly
    0.39
    Suddenly
    0.39
    頂いた
    0.38
    ṡ
    0.38
    Have
    0.38
    Act Density 0.018%

    No Known Activations