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

    Neuronpedia

    INDEX
    Explanations

    complex topics

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     مثلا
    0.42
     Aynı
    0.39
     Schn
    0.39
     Include
    0.38
     Например
    0.38
    随意
    0.38
    があれば
    0.38
     Örneğin
    0.38
     อาท
    0.38
    比如
    0.37
    POSITIVE LOGITS
     nuanced
    0.64
     surprisingly
    0.62
     misunderstood
    0.62
     multifaceted
    0.62
    surprisingly
    0.61
     fascinating
    0.61
     niekoľ
    0.59
     tricky
    0.59
    several
    0.56
     several
    0.54
    Act Density 1.297%

    No Known Activations