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

    Neuronpedia

    INDEX
    Explanations

    phrases signaling contrast or contradiction

    oai_token-act-pair · gpt-3.5-turbo

    the word "though" indicating contrasts or exceptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    lees
    -0.90
    lee
    -0.69
    anu
    -0.63
    è¦ļéĨĴ
    -0.62
    ultan
    -0.62
     bunny
    -0.62
    aru
    -0.60
    ragon
    -0.59
    ixties
    -0.59
     Cosponsors
    -0.59
    POSITIVE LOGITS
     circumst
    0.77
     admittedly
    0.73
    tons
    0.70
     curiously
    0.66
     interestingly
    0.65
    lihood
    0.65
    ,,
    0.64
     fortunately
    0.64
    tha
    0.64
     elusive
    0.62
    Act Density 0.032%

    No Known Activations