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

    Neuronpedia

    INDEX
    Explanations

    instances where a sentence includes contrasting information or exceptions ("however")

    oai_token-act-pair · gpt-3.5-turbo

    the word "however" indicating contrasts or exceptions in statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Flavoring
    -0.71
    urring
    -0.64
    mer
    -0.63
    cup
    -0.62
     Cosponsors
    -0.59
    enter
    -0.59
    iard
    -0.58
    ursed
    -0.58
    urrence
    -0.57
    cycle
    -0.57
    POSITIVE LOGITS
     insignificant
    0.70
     slight
    0.69
     reluctantly
    0.67
     interestingly
    0.65
     entertained
    0.65
     disposed
    0.64
     circumst
    0.63
     agre
    0.63
    soever
    0.63
     importantly
    0.62
    Act Density 0.028%

    No Known Activations