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

    Neuronpedia

    INDEX
    Explanations

    probability distribution exceeds threshold

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ে
    0.59
    يلا
    0.52
     huo
    0.48
    مین
    0.47
    認め
    0.46
    த்திய
    0.45
    丢
    0.45
    িগ্ন
    0.44
    աս
    0.44
    ؤ
    0.44
    POSITIVE LOGITS
    ons
    0.51
     иллюстрация
    0.49
    Ons
    0.47
     شامل
    0.46
    aris
    0.45
    uls
    0.45
    á
    0.45
     худ
    0.44
     똑
    0.44
     Airport
    0.43
    Act Density 0.000%

    No Known Activations