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

    Neuronpedia

    INDEX
    Explanations

    python code block starts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     class
    0.90
     सैफ
    0.73
     SRC
    0.70
    class
    0.69
     political
    0.67
     minder
    0.66
     cuisine
    0.66
    typal
    0.66
     Ethnic
    0.65
     servant
    0.64
    POSITIVE LOGITS
    <!
    0.93
    #!/
    0.86
     opposit
    0.86
    हां
    0.83
     !",
    0.82
    பாத
    0.81
    codehaus
    0.81
    <?
    0.80
    !);
    0.80
    chak
    0.80
    Act Density 0.063%

    No Known Activations