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

    Neuronpedia

    INDEX
    Explanations

    references to royalty or princesses

    oai_token-act-pair · gpt-3.5-turbo

    references to princess characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     smoker
    -0.70
    sych
    -0.69
    olson
    -0.65
    ORN
    -0.63
    andem
    -0.62
     neur
    -0.61
    oker
    -0.61
    entimes
    -0.61
     revolving
    -0.60
    EFF
    -0.60
    POSITIVE LOGITS
     Leia
    1.20
     Princess
    1.09
     Bride
    0.97
     Peach
    0.93
     Diana
    0.92
    anova
    0.91
     Celest
    0.91
     princess
    0.85
    ette
    0.84
    cess
    0.79
    Act Density 0.007%

    No Known Activations