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

    Neuronpedia

    INDEX
    Explanations

    specific effects reduce mass letters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    د
    0.54
    سٹم
    0.49
    ස්
    0.48
    CQL
    0.47
     gdy
    0.45
    باز
    0.45
     methylated
    0.44
    予想
    0.42
    翌
    0.42
     olvid
    0.42
    POSITIVE LOGITS
     dangling
    0.47
    idamente
    0.47
    y
    0.46
     harmonica
    0.45
     Кана
    0.44
    лю
    0.44
    事项
    0.43
    нина
    0.42
    len
    0.42
    ীক
    0.41
    Act Density 0.001%

    No Known Activations