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

    Neuronpedia

    INDEX
    Explanations

    action words related to actions done without permission or consent

    oai_token-act-pair · gpt-3.5-turbo

    actions that occur without consent or permission

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    raq
    -0.81
    late
    -0.78
    geist
    -0.78
    cow
    -0.73
    iers
    -0.72
    ahime
    -0.70
    =-=-=-=-=-=-=-=-
    -0.69
    atile
    -0.67
    mun
    -0.67
    lish
    -0.66
    POSITIVE LOGITS
     regard
    1.02
     necessarily
    0.97
     sacrificing
    0.97
     compromising
    0.94
     mentioning
    0.93
     knowing
    0.92
     hesitation
    0.88
     disclosing
    0.87
     exception
    0.86
     risking
    0.85
    Act Density 0.042%

    No Known Activations