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

    Neuronpedia

    INDEX
    Explanations

    conformal prediction, aspect ratio, js imports

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    k
    0.80
    ا
    0.76
    ون
    0.74
    in
    0.74
    f
    0.72
    In
    0.67
    IM
    0.66
    ok
    0.65
     In
    0.63
    ين
    0.61
    POSITIVE LOGITS
    মনসিংহ
    0.66
    ப்பத
    0.63
     ساختمان
    0.61
     яке
    0.61
    ycznych
    0.59
     Tunisie
    0.59
     Zeitung
    0.59
     sweetheart
    0.57
     მ
    0.57
    ⅾ
    0.56
    Act Density 0.006%

    No Known Activations