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

    Neuronpedia

    INDEX
    Explanations

    references to political leaders with dictatorial qualities

    oai_token-act-pair · gpt-3.5-turbo

    references to various forms of authoritarian leadership

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Sea
    -0.89
    LAN
    -0.85
    ttp
    -0.83
    atha
    -0.82
    Ward
    -0.81
    Lear
    -0.80
    bra
    -0.80
    awks
    -0.79
    lan
    -0.78
    Adv
    -0.78
    POSITIVE LOGITS
     dictator
    1.54
     dictatorship
    1.33
     dictators
    1.16
     regimes
    1.01
     nomine
    0.96
     regime
    0.93
    ocrats
    0.93
     tyrant
    0.91
     ruler
    0.87
     Saddam
    0.86
    Act Density 0.018%

    No Known Activations