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

    Neuronpedia

    INDEX
    Explanations

    negations or negative contexts

    oai_token-act-pair · gpt-3.5-turbo

    negations or phrases indicating the absence of something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     gears
    -0.67
    now
    -0.66
     newsletters
    -0.65
    inks
    -0.65
     loopholes
    -0.62
     grapple
    -0.61
    cano
    -0.61
    ilus
    -0.60
    ãģį
    -0.60
    plates
    -0.59
    POSITIVE LOGITS
     originally
    1.02
     consulted
    1.02
     initially
    0.95
    icably
    0.95
     harmed
    0.93
     intending
    0.92
    icable
    0.92
     intentional
    0.91
     adequately
    0.90
     able
    0.89
    Act Density 0.144%

    No Known Activations