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

    Neuronpedia

    INDEX
    Explanations

    rounding to the nearest

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    !”
    0.48
     воспа
    0.46
    ITUDE
    0.46
     تبع
    0.45
     довго
    0.45
     ампли
    0.45
    ?”
    0.45
     sputtered
    0.45
     программы
    0.44
     функциона
    0.44
    POSITIVE LOGITS
    petal
    0.42
    dua
    0.42
    ک
    0.41
     नहीं
    0.40
     yuè
    0.40
     chết
    0.40
     Lu
    0.39
     comparable
    0.39
    dus
    0.39
     fuch
    0.39
    Act Density 0.002%

    No Known Activations