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

    Neuronpedia

    INDEX
    Explanations

    expressed as percentage

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    坚定
    0.54
    强大
    0.42
    国际
    0.41
    小说
    0.41
    বু
    0.40
    厨房
    0.40
    나무
    0.40
    任务
    0.40
    实施
    0.40
     স্মৃতি
    0.40
    POSITIVE LOGITS
     percentage
    2.36
     percentages
    2.09
    percentage
    2.08
     Percentage
    2.06
     porcentaje
    1.91
    Percentage
    1.90
     percentuale
    1.84
     percent
    1.80
     pourcentage
    1.80
     porcent
    1.77
    Act Density 0.078%

    No Known Activations