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

    Neuronpedia

    INDEX
    Explanations

    whom we/you

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    IN
    0.53
     భాగంగా
    0.53
     melibatkan
    0.50
    a
    0.50
    া
    0.49
    ।"
    0.48
    ড
    0.46
    inį
    0.45
     بیٹے
    0.44
     এলাক
    0.44
    POSITIVE LOGITS
    soever
    0.62
     we
    0.54
    é
    0.47
     cui
    0.46
     можно
    0.45
     можна
    0.45
    ми
    0.45
     också
    0.44
    有一些
    0.44
     you
    0.44
    Act Density 0.005%

    No Known Activations