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

    Neuronpedia

    INDEX
    Explanations

    stunt performers and their work

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    art
    1.08
    I
    1.06
     in
    1.05
    O
    1.02
    its
    0.89
    W
    0.88
     a
    0.85
    า
    0.83
    ang
    0.82
    Z
    0.80
    POSITIVE LOGITS
     stunts
    1.13
     stunt
    1.04
    ແມ
    0.96
     поведение
    0.91
    н
    0.90
     antics
    0.89
     дії
    0.88
    に適
    0.87
     rollerskates
    0.86
     acrob
    0.83
    Act Density 0.001%

    No Known Activations