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

    Neuronpedia

    INDEX
    Explanations

    high quality praise

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    пуляр
    0.86
     popular
    0.84
    бовать
    0.84
    щую
    0.83
     പ്രധാന
    0.83
     популярных
    0.83
     популяр
    0.81
     Предпо
    0.80
    ukkig
    0.79
    ртуа
    0.78
    POSITIVE LOGITS
     honestly
    1.09
     Seriously
    0.99
    Seriously
    0.98
     unmatched
    0.98
     flawlessly
    0.97
     perfectly
    0.97
     Perfectly
    0.96
     impecc
    0.94
    seriously
    0.92
     unparalleled
    0.91
    Act Density 0.189%

    No Known Activations