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

    Neuronpedia

    INDEX
    Explanations

    sister brother relationship

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ي
    1.53
    સર
    1.47
     sebuah
    1.41
    लाइट
    1.38
    宝宝
    1.37
    आर
    1.36
    ках
    1.34
    что
    1.31
    需
    1.31
     وعلى
    1.30
    POSITIVE LOGITS
    er
    2.09
    erade
    1.46
    });
    1.45
    ‣
    1.45
    ঝি
    1.41
    neighbor
    1.41
    👬
    1.41
    。「
    1.38
    বোন
    1.38
    2
    1.34
    Act Density 0.013%

    No Known Activations