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

    Neuronpedia

    INDEX
    Explanations

    references to nudity

    oai_token-act-pair · gpt-3.5-turbo

    references to nudity or nakedness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    riers
    -0.83
    soType
    -0.81
     Flavoring
    -0.77
    rador
    -0.72
    mental
    -0.72
    ppa
    -0.72
    SHIP
    -0.71
    ffee
    -0.70
    ãĥ£
    -0.70
    ãĥį
    -0.69
    POSITIVE LOGITS
     selfies
    1.06
     mole
    0.97
     selfie
    0.94
     photographs
    0.92
     nude
    0.91
     silhou
    0.89
     photos
    0.88
     pictures
    0.84
     breasts
    0.83
     bathing
    0.82
    Act Density 0.040%

    No Known Activations