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

    Neuronpedia

    INDEX
    Explanations

    creating options packed learning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    信息
    0.49
     to
    0.47
     information
    0.46
    CHREIBUNG
    0.45
    他的
    0.44
     historical
    0.43
     unthinkable
    0.41
    完整
    0.41
    อต
    0.41
     well
    0.40
    POSITIVE LOGITS
     благоприят
    0.53
     Цвет
    0.52
     BBBB
    0.51
     ಬಂಧ
    0.50
     신청
    0.49
     који
    0.48
    ћи
    0.48
    сий
    0.47
    딪
    0.47
     сопротив
    0.47
    Act Density 0.000%

    No Known Activations