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

    Neuronpedia

    INDEX
    Explanations

    the word "perhaps" and its variations, indicating speculation or uncertainty

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     pleaſure
    -0.75
     houſe
    -0.67
     ſp
    -0.64
    TM
    -0.61
     Jefus
    -0.61
     tức
    -0.60
     Efq
    -0.60
     himſelf
    -0.59
     ſame
    -0.56
     poupée
    -0.56
    POSITIVE LOGITS
     perhaps
    1.90
    perhaps
    1.86
     Perhaps
    1.78
    Perhaps
    1.77
     maybe
    1.67
    maybe
    1.62
    Maybe
    1.62
     Maybe
    1.58
    anskje
    1.48
     quizás
    1.40
    Act Density 0.162%

    No Known Activations