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

    Neuronpedia

    INDEX
    Explanations

    while not yet

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ジャン
    0.41
    enziali
    0.41
    述
    0.41
     coloqu
    0.40
     appartiennent
    0.40
    펄
    0.40
    argu
    0.40
     revolutionaries
    0.39
    changers
    0.39
    衍生
    0.39
    POSITIVE LOGITS
    ška
    0.55
     Time
    0.47
     Sau
    0.42
     Thời
    0.41
     Not
    0.41
     kul
    0.41
    يبة
    0.41
     gerekiyor
    0.41
     esimerkiksi
    0.41
     Init
    0.40
    Act Density 0.005%

    No Known Activations