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

    Neuronpedia

    INDEX
    Explanations

    select what best

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     წლის
    0.33
    预计
    0.31
    顿时
    0.31
    événement
    0.31
     dzięki
    0.31
     החד
    0.30
    すぎて
    0.30
     podczas
    0.29
    是因为
    0.29
    지난
    0.29
    POSITIVE LOGITS
     would
    0.39
    ’
    0.38
     they
    0.33
     you
    0.32
    我們可以
    0.31
     can
    0.30
     Would
    0.30
     They
    0.30
    可以
    0.30
     we
    0.30
    Act Density 0.001%

    No Known Activations