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

    Neuronpedia

    INDEX
    Explanations

    instances of actions performed without consent

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating actions taken without permission or consent

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    lyak
    -0.74
    =-=-=-=-=-=-=-=-
    -0.69
    rez
    -0.67
    ãģ®éŃĶ
    -0.67
     Reincarnated
    -0.67
    atile
    -0.66
    dominated
    -0.65
    raq
    -0.65
    hov
    -0.65
    esa
    -0.65
    POSITIVE LOGITS
     authorization
    1.25
     permission
    1.18
     informing
    1.10
     consent
    1.09
     specifying
    1.04
     explanation
    1.02
     reperc
    1.01
     disclosing
    1.01
     warning
    1.00
     supervision
    0.99
    Act Density 0.082%

    No Known Activations