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

    Neuronpedia

    INDEX
    Explanations

    spectral, 2-in-1, routine, app

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Ͻ
    1.38
     ء
    1.31
    ました
    1.23
    <unused1046>
    1.18
    rootDir
    1.18
     broccoli
    1.16
    ки
    1.14
     suffered
    1.14
     वॉलेट
    1.14
     erupted
    1.13
    POSITIVE LOGITS
    P
    1.39
    S
    1.31
    N
    1.23
    T
    1.20
    A
    1.17
    V
    1.16
    ല്ലാം
    1.15
    L
    1.13
    M
    1.08
    Ai
    1.08
    Act Density 0.001%

    No Known Activations