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

    Neuronpedia

    INDEX
    Explanations

    vulgarity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    LoopBlend
    0.53
     düğ
    0.48
    வரின்
    0.46
     ઉમે
    0.46
    Caution
    0.46
    ®,
    0.45
    0.45
    とっても
    0.45
     Оюн
    0.45
    ℗
    0.45
    POSITIVE LOGITS
     fuck
    0.95
     shit
    0.95
     fucking
    0.87
    좆
    0.86
     bullshit
    0.85
     Fuck
    0.84
    fuck
    0.82
     fucked
    0.81
     cunt
    0.80
     shitty
    0.77
    Act Density 0.048%

    No Known Activations