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

    Neuronpedia

    INDEX
    Explanations

    programming markdown

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0
    0.75
    =(
    0.59
    or
    0.58
    2
    0.57
    5
    0.57
    6
    0.57
    8
    0.56
    (
    0.55
    4
    0.55
     for
    0.54
    POSITIVE LOGITS
    quele
    0.56
    pesar
    0.55
    bbero
    0.50
     लक्ष
    0.48
     افزود
    0.48
     beispielsweise
    0.47
    seits
    0.47
     Якщо
    0.46
    rocław
    0.46
     قلت
    0.46
    Act Density 0.061%

    No Known Activations