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

    Neuronpedia

    INDEX
    Explanations

    code, technical or descriptive terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    بحث
    0.46
     vault
    0.38
    𝗋
    0.38
     Defensa
    0.38
    باح
    0.37
    这也
    0.37
    боль
    0.36
    默
    0.36
    مكن
    0.36
     "\""
    0.36
    POSITIVE LOGITS
    weathermap
    0.47
     Bruce
    0.38
     Render
    0.38
    casualties
    0.37
     попу
    0.37
     renderings
    0.37
    aji
    0.36
     captures
    0.36
     일본
    0.36
     fortunes
    0.36
    Act Density 0.000%

    No Known Activations