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

    Neuronpedia

    INDEX
    Explanations

    increases and changes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    uldu
    0.50
    fromParams
    0.44
     політи
    0.43
     ошибки
    0.43
    archiwizowane
    0.40
     আচরণ
    0.39
     уведом
    0.39
     антен
    0.39
    መ
    0.39
     полити
    0.39
    POSITIVE LOGITS
     rinse
    0.54
     levar
    0.48
     জলে
    0.47
     water
    0.46
     kitchens
    0.46
    廚
    0.45
     easily
    0.45
     Cooking
    0.44
     Hour
    0.44
    sh
    0.43
    Act Density 0.002%

    No Known Activations