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

    Neuronpedia

    INDEX
    Explanations

    redacted XXX codes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    {
    0.90
    -
    0.79
    >
    0.78
     manej
    0.77
    ’
    0.75
    ی
    0.75
     subir
    0.75
     adapt
    0.73
     accompagn
    0.72
     trova
    0.72
    POSITIVE LOGITS
    v
    1.04
     बच्चे
    0.80
    million
    0.73
     люди
    0.72
    stä
    0.68
    än
    0.67
    рили
    0.65
     व्यक्तियों
    0.64
     κά
    0.63
    thol
    0.63
    Act Density 0.000%

    No Known Activations