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

    Neuronpedia

    INDEX
    Explanations

    code and data structures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ed
    0.50
    و
    0.45
    ே
    0.43
     mặc
    0.41
    at
    0.41
    ین
    0.40
    ic
    0.39
     чыныгы
    0.39
    ле
    0.39
     بیشتر
    0.39
    POSITIVE LOGITS
    0.57
     was
    0.48
    :
    0.40
    是
    0.40
    ➀
    0.38
    <h2>
    0.38
     are
    0.37
    ielle
    0.36
    </h2>
    0.35
    )))
    0.35
    Act Density 0.125%

    No Known Activations