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

    Neuronpedia

    INDEX
    Explanations

    choice of politeness/theft/legal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    鍘
    0.46
    க்கிறது
    0.45
    刄
    0.43
    ASSOCI
    0.41
    тат
    0.41
    )");
    0.40
    高
    0.40
     হইতেছে
    0.39
     Perimeter
    0.38
     ଓ
    0.37
    POSITIVE LOGITS
    i
    0.45
    rea
    0.44
    čiai
    0.44
    how
    0.43
    pract
    0.43
    frei
    0.42
    いが
    0.42
    ség
    0.41
    is
    0.41
    adie
    0.40
    Act Density 0.000%

    No Known Activations