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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating contrast or contradiction

    oai_token-act-pair · gpt-3.5-turbo

    the word "however" and its various contextual uses

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    è¦ļéĨĴ
    -0.74
    ULAR
    -0.62
    Glass
    -0.61
    aza
    -0.61
     suitcase
    -0.60
    rawl
    -0.59
    Cover
    -0.59
    Laughs
    -0.59
    AI
    -0.57
    icide
    -0.57
    POSITIVE LOGITS
    chery
    0.80
     unlike
    0.76
    nown
    0.76
     acknow
    0.76
     alas
    0.74
     according
    0.73
     none
    0.70
     depending
    0.69
     fortunately
    0.68
     excluding
    0.67
    Act Density 0.202%

    No Known Activations