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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating non-compatibility or negation

    oai_token-act-pair · gpt-3.5-turbo

    negative statements or affirmations regarding compatibility and inclusion

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    TIME
    -0.80
    GREEN
    -0.71
    Nine
    -0.70
    Honest
    -0.69
    LOS
    -0.68
    Truth
    -0.66
     mouths
    -0.66
    Kings
    -0.66
     glances
    -0.65
    WAY
    -0.65
    POSITIVE LOGITS
     necessarily
    1.34
    icably
    1.09
    icable
    1.08
     recommended
    0.90
    epad
    0.89
     automatically
    0.85
    necess
    0.83
     permitted
    0.82
     included
    0.81
    ifies
    0.81
    Act Density 0.241%

    No Known Activations