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

    Neuronpedia

    INDEX
    Explanations

    software application context

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    3
    1.43
    我
    1.41
    ร้าน
    1.32
    ス
    1.23
     capaz
    1.22
    د
    1.18
     joka
    1.17
    㦣
    1.17
     був
    1.15
    ión
    1.13
    POSITIVE LOGITS
    is
    1.85
    n
    1.80
    at
    1.72
    m
    1.71
    ing
    1.68
    us
    1.62
    ol
    1.59
    ir
    1.59
    ა
    1.57
    ೇ
    1.55
    Act Density 0.297%

    No Known Activations