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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating personal experiences or interactions

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     accordingly
    -0.65
    lying
    -0.65
    arry
    -0.64
    interstitial
    -0.64
    umbing
    -0.63
    peak
    -0.62
    iter
    -0.62
    anwhile
    -0.62
     thats
    -0.61
    annel
    -0.60
    POSITIVE LOGITS
     opportunity
    1.39
     privilege
    1.28
     misfortune
    1.21
     pleasure
    1.13
     courage
    1.07
     chance
    1.06
     option
    1.03
     utmost
    0.99
     displeasure
    0.97
     guts
    0.97
    Act Density 0.097%

    No Known Activations