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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "who" and variations that indicate questioning or identification of people

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ArrowToggle
    -0.67
     Eis
    -0.65
     Lyman
    -0.64
     Berech
    -0.63
    らう
    -0.63
     świą
    -0.61
     fré
    -0.61
     Merk
    -0.61
     paff
    -0.60
    work
    -0.59
    POSITIVE LOGITS
     who
    1.63
     Who
    1.29
     WHO
    1.27
    who
    1.24
    Who
    1.18
    Whoosh
    1.11
    WHO
    1.06
     whom
    1.05
     którzy
    1.04
     quienes
    1.01
    Act Density 0.077%

    No Known Activations