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

    Neuronpedia

    INDEX
    Explanations

    bullet points

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    wirkung
    0.40
     하나
    0.39
     ！
    0.39
    包括
    0.38
     stoichi
    0.38
     hauptsächlich
    0.38
    Generate
    0.37
     incluyen
    0.37
     находится
    0.37
     quantile
    0.37
    POSITIVE LOGITS
    下記
    0.39
    👇
    0.38
     LA
    0.38
     SA
    0.38
    <table>
    0.37
    :]
    0.37
     Αγ
    0.37
     below
    0.36
     Así
    0.36
    :</
    0.36
    Act Density 0.055%

    No Known Activations