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

    Neuronpedia

    INDEX
    Explanations

    initialize or install

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <unused727>
    0.42
    he
    0.42
     খুব
    0.42
    uz
    0.41
    atual
    0.41
    一定要
    0.41
    ंच
    0.40
    这本书
    0.40
     아주
    0.39
    업
    0.39
    POSITIVE LOGITS
     L
    0.57
     F
    0.56
     B
    0.55
     M
    0.52
     R
    0.50
     P
    0.49
     D
    0.48
     G
    0.48
     H
    0.48
     T
    0.46
    Act Density 0.076%

    No Known Activations