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

    Neuronpedia

    INDEX
    Explanations

    to express purpose

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    應該
    0.43
     கொள்ளலாம்
    0.43
     should
    0.42
     இருக்கலாம்
    0.39
    াহারের
    0.38
    应该
    0.38
    should
    0.37
    是个
    0.36
     shouldn
    0.36
     yaşad
    0.36
    POSITIVE LOGITS
    하려면
    1.46
    려면
    1.44
     Чтобы
    1.43
     để
    1.38
    Чтобы
    1.38
     Để
    1.35
    为了
    1.34
     כדי
    1.26
     чтобы
    1.25
    Để
    1.24
    Act Density 0.028%

    No Known Activations