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

    Neuronpedia

    INDEX
    Explanations

    words related to refusal or defiance

    oai_token-act-pair · gpt-3.5-turbo

    instances of refusal or rejection

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    soDeliveryDate
    -1.04
    lined
    -0.80
    quickShipAvailable
    -0.77
    soType
    -0.75
    ////////////////////////////////
    -0.75
    nin
    -0.74
     largeDownload
    -0.71
    oward
    -0.68
    gradient
    -0.68
    retty
    -0.67
    POSITIVE LOGITS
     accept
    1.14
     cooperate
    1.14
     obey
    1.10
     heed
    1.09
     acknowledge
    1.09
     bud
    1.06
     comply
    1.05
     endorse
    1.02
     reproduce
    1.01
     recognize
    1.01
    Act Density 0.048%

    No Known Activations