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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating upcoming elaboration or emphasis in speech

    oai_token-act-pair · gpt-3.5-turbo

    expressions of uncertainty or ambivalence about a situation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ascript
    -0.76
    cific
    -0.76
    ithing
    -0.73
    igent
    -0.71
    stone
    -0.71
    aintain
    -0.69
    imeter
    -0.65
    hower
    -0.64
    ourses
    -0.64
    atana
    -0.63
    POSITIVE LOGITS
     kidding
    0.90
     damned
    0.76
    oops
    0.73
     hilar
    0.73
     Surprise
    0.73
     Creep
    0.71
     downright
    0.71
     damn
    0.70
     Bucs
    0.70
     darn
    0.69
    Act Density 1.200%

    No Known Activations