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

    Neuronpedia

    INDEX
    Explanations

    code structuring punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ны
    0.53
    ੦
    0.49
    usalem
    0.49
    imcoords
    0.49
     goû
    0.49
    াৎ
    0.48
     ทำให้
    0.48
    Strafpunkte
    0.48
     Suède
    0.47
     vuurp
    0.46
    POSITIVE LOGITS
     and
    0.58
     for
    0.54
    that
    0.53
     in
    0.51
    ↵
    0.49
     de
    0.48
    In
    0.47
    A
    0.47
    ki
    0.45
    B
    0.45
    Act Density 0.081%

    No Known Activations