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

    Neuronpedia

    INDEX
    Explanations

    phrases related to human rights abuses and brutality

    oai_token-act-pair · gpt-3.5-turbo

    references to human rights violations and related terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    paren
    -0.80
    onym
    -0.77
    ramer
    -0.76
    ership
    -0.73
     ellipt
    -0.73
    aran
    -0.72
    adult
    -0.71
    ellar
    -0.71
    ode
    -0.70
    arten
    -0.70
    POSITIVE LOGITS
     atrocities
    0.95
     brutality
    0.89
     abuses
    0.87
     inflicted
    0.84
     perpetrated
    0.84
     massacres
    0.81
     dehuman
    0.79
     massac
    0.79
     killings
    0.73
     injustice
    0.73
    Act Density 0.064%

    No Known Activations