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

    Neuronpedia

    INDEX
    Explanations

    defining abstract concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    кою
    0.56
    скоп
    0.55
    $,
    0.52
    сті
    0.51
    -$
    0.49
    িত
    0.48
    і
    0.48
    ט
    0.48
    ित्र
    0.48
    $:
    0.47
    POSITIVE LOGITS
    ed
    0.62
    os
    0.61
     درخواست
    0.56
    ak
    0.52
    e
    0.49
    em
    0.49
     exigences
    0.48
     conviv
    0.48
    ЕН
    0.47
    iya
    0.47
    Act Density 0.230%

    No Known Activations