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

    Neuronpedia

    INDEX
    Explanations

    change or safety

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Att
    0.38
    Balls
    0.37
     thicker
    0.36
    balls
    0.36
    Hos
    0.35
    Hik
    0.35
     ব্যাক্ত
    0.35
    Fat
    0.35
    年
    0.35
     обла
    0.35
    POSITIVE LOGITS
     плохо
    0.42
    apos
    0.41
     pac
    0.39
     barely
    0.38
    ANCE
    0.38
    悪い
    0.37
    oida
    0.37
    anza
    0.37
    Application
    0.37
    astar
    0.36
    Act Density 0.000%

    No Known Activations