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

    Neuronpedia

    INDEX
    Explanations

    why how directly modifying

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Brand
    0.44
     Concurrent
    0.44
     Phyt
    0.41
     Educator
    0.40
    ):
    0.40
     пропа
    0.40
    可以通过
    0.39
     Markup
    0.39
    应对
    0.38
     Recursos
    0.38
    POSITIVE LOGITS
     現在
    0.54
    至今
    0.50
    Jetzt
    0.47
     all
    0.46
     obecnie
    0.46
    ทั้งหมด
    0.45
    už
    0.45
    โลก
    0.45
     现在
    0.45
    ⓞ
    0.45
    Act Density 0.001%

    No Known Activations