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

    Neuronpedia

    INDEX
    Explanations

    conditional logic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     이를
    0.41
     невозможно
    0.40
     изменить
    0.37
     ਨੂੰ
    0.37
     což
    0.37
    的角度
    0.37
    Acetic
    0.37
     aproxim
    0.36
     kev
    0.36
     поне
    0.36
    POSITIVE LOGITS
     if
    1.46
     nếu
    1.43
     если
    1.41
     If
    1.39
     اگر
    1.36
    if
    1.35
    如果
    1.35
     Если
    1.35
     якщо
    1.34
     eğer
    1.34
    Act Density 0.190%

    No Known Activations