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

    Neuronpedia

    INDEX
    Explanations

    retail outlets

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ’
    0.66
    ва
    0.59
    のは
    0.57
     is
    0.55
    ат
    0.54
    ем
    0.54
    ал
    0.53
    ために
    0.52
    টি
    0.51
    기
    0.51
    POSITIVE LOGITS
    é
    0.55
    T
    0.53
    ：
    0.53
    ler
    0.52
    َ
    0.48
    N
    0.48
    د
    0.48
    k
    0.47
    <
    0.46
    )
    0.45
    Act Density 0.003%

    No Known Activations