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

    Neuronpedia

    INDEX
    Explanations

    requires careful handling

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    }]$,
    0.52
     bát
    0.51
    但在
    0.50
    ל
    0.49
    astaan
    0.47
     فوټبال
    0.47
     funkce
    0.47
    "].
    0.46
    پيديا
    0.46
    ckiego
    0.45
    POSITIVE LOGITS
    eval
    0.57
    od
    0.55
    os
    0.52
    as
    0.49
    es
    0.48
    w
    0.47
    ad
    0.46
    m
    0.46
    icaria
    0.46
    recogn
    0.45
    Act Density 0.000%

    No Known Activations