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

    Neuronpedia

    INDEX
    Explanations

    phrases related to various forms of abuse and misconduct

    oai_token-act-pair · gpt-3.5-turbo

    instances of various forms of abuse and violence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    zzo
    -0.82
    gue
    -0.82
    imester
    -0.80
    eport
    -0.78
    uve
    -0.77
    gio
    -0.76
    cade
    -0.75
    abase
    -0.73
    plates
    -0.73
    baum
    -0.72
    POSITIVE LOGITS
     vandalism
    1.26
     criminality
    1.25
     violence
    1.24
     persecution
    1.22
     intimidation
    1.21
     harassment
    1.21
     brutality
    1.18
     oppression
    1.18
     injustice
    1.18
     exploitation
    1.17
    Act Density 0.317%

    No Known Activations