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

    Neuronpedia

    INDEX
    Explanations

    academic and code references

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    is
    0.65
     that
    0.63
     İş
    0.63
     îi
    0.61
    larını
    0.60
    chaft
    0.59
     freien
    0.58
    ella
    0.57
    çe
    0.57
    $,
    0.56
    POSITIVE LOGITS
    presidente
    0.65
    ม
    0.62
     년
    0.61
    ク
    0.61
    লে
    0.59
    Дата
    0.58
    αν
    0.57
     למ
    0.57
    م
    0.57
    ვ
    0.57
    Act Density 0.000%

    No Known Activations