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

    Neuronpedia

    INDEX
    Explanations

    para, paragraph, method, marks, windows

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    of
    1.62
     של
    1.16
    på
    1.14
     pēc
    1.12
     נו
    1.09
    );
    1.09
     व
    1.09
     nyata
    1.09
     tinggi
    1.08
     powied
    1.08
    POSITIVE LOGITS
    at
    1.18
    para
    1.16
    ب
    1.14
    شون
    1.13
    Para
    1.06
    ו
    1.06
    itions
    1.04
    2
    1.04
    ди
    1.03
    缱
    1.03
    Act Density 0.032%

    No Known Activations