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

    Neuronpedia

    INDEX
    Explanations

    meaning isn't pre-ordained

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    '
    0.62
    ’
    0.49
     harb
    0.45
    /
    0.44
     prese
    0.43
    '=
    0.43
     deewana
    0.42
     Fortune
    0.41
     Q
    0.41
     slew
    0.40
    POSITIVE LOGITS
    ﺭ
    0.52
     केन्द्र
    0.49
    สินค้า
    0.48
     '</
    0.47
     మార
    0.45
    苯
    0.45
    larında
    0.45
    कार्
    0.45
     উন্নতি
    0.44
     ನೀವು
    0.44
    Act Density 0.001%

    No Known Activations