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

    Neuronpedia

    INDEX
    Explanations

    references to sexual assault and related discussions on prevention

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     prune
    -0.15
     eccentric
    -0.15
    adius
    -0.14
    cação
    -0.14
     curses
    -0.14
     Sweat
    -0.14
     smoker
    -0.14
    lun
    -0.14
    ç¥ĸ
    -0.14
     Void
    -0.13
    POSITIVE LOGITS
     sexual
    0.46
     Sexual
    0.44
     rape
    0.44
     Rape
    0.40
    rape
    0.38
    Sex
    0.35
     survivors
    0.35
     survivor
    0.34
    sexual
    0.34
     victim
    0.34
    Act Density 0.095%

    No Known Activations