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

    Neuronpedia

    INDEX
    Explanations

    words related to outrageous or egregious behavior

    oai_token-act-pair · gpt-3.5-turbo

    words that indicate extreme or outrageous behavior or characteristics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    erves
    -0.81
    ovember
    -0.77
    ajor
    -0.75
    ulton
    -0.70
    maid
    -0.70
    yer
    -0.68
    avers
    -0.68
    rief
    -0.68
    mediate
    -0.67
    igate
    -0.67
    POSITIVE LOGITS
    ness
    0.98
     amounts
    0.86
     outrageous
    0.86
    nesses
    0.84
     antics
    0.83
     shenan
    0.82
     nonsense
    0.78
     bluff
    0.72
    NESS
    0.70
     hypocritical
    0.70
    Act Density 0.023%

    No Known Activations