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

    Neuronpedia

    INDEX
    Explanations

    conditional statement beginning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ственно
    0.42
    返回
    0.41
    하게
    0.41
    вшего
    0.40
    Seine
    0.39
     aisle
    0.38
     вместо
    0.38
    시키는
    0.38
    要素
    0.38
    儩
    0.38
    POSITIVE LOGITS
    if
    1.03
     যদি
    1.00
     nếu
    0.95
    如果
    0.94
     if
    0.93
     если
    0.93
     якщо
    0.93
     If
    0.92
    If
    0.91
     اگر
    0.91
    Act Density 0.495%

    No Known Activations