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

    Neuronpedia

    INDEX
    Explanations

    questions starting with the word "which"

    oai_token-act-pair · gpt-3.5-turbo

    typical sentence starters or phrases that introduce questions or topics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ARC
    -0.73
    greg
    -0.72
    affle
    -0.69
    ancock
    -0.68
    ourning
    -0.67
    Rated
    -0.67
    usk
    -0.65
    perty
    -0.65
    ancial
    -0.65
    Repl
    -0.64
    POSITIVE LOGITS
     brings
    1.52
     begs
    1.51
     leads
    1.26
     reminds
    1.15
     means
    1.15
     raises
    1.12
     leaves
    1.09
     explains
    1.05
     translates
    1.04
     makes
    1.03
    Act Density 0.049%

    No Known Activations