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

    Neuronpedia

    INDEX
    Explanations

    parenthetical phrases and conjunctions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Это
    0.58
    Если
    0.57
    Якщо
    0.56
    Нет
    0.52
    Тех
    0.51
    यदि
    0.50
    abilirsiniz
    0.50
    큽
    0.49
    ():
    0.49
    यह
    0.49
    POSITIVE LOGITS
     whose
    0.61
     presumably
    0.50
     cuja
    0.48
     which
    0.47
     এবং
    0.47
     cujo
    0.47
    ซึ่ง
    0.46
     cuyo
    0.46
     and
    0.43
    ،
    0.40
    Act Density 0.185%

    No Known Activations