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

    Neuronpedia

    INDEX
    Explanations

    phrases related to crime and criminal activities involving people

    oai_token-act-pair · gpt-3.5-turbo

    possessive pronouns and references to familial relationships in the context of crime or violence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     largeDownload
    -0.88
    ãĥ´ãĤ¡
    -0.77
    inctions
    -0.75
    taboola
    -0.71
    ernels
    -0.70
    Show
    -0.70
    Wide
    -0.69
    eers
    -0.69
    GAN
    -0.68
    days
    -0.68
    POSITIVE LOGITS
     own
    1.26
     wife
    1.20
     girlfriend
    1.19
     fian
    1.12
     daughter
    1.11
     cousin
    1.10
     colleague
    1.04
     grandmother
    1.04
     father
    1.03
     estranged
    1.03
    Act Density 0.174%

    No Known Activations