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

    Neuronpedia

    INDEX
    Explanations

    phrases containing intensifiers related to belief or conviction

    oai_token-act-pair · gpt-3.5-turbo

    the word "utterly" and other similar intensifiers or adverbs that emphasize totality or completeness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ioch
    -0.82
     Occupations
    -0.79
    hao
    -0.77
    anwhile
    -0.77
    imeters
    -0.77
    arters
    -0.75
    hops
    -0.74
     cigars
    -0.73
    cers
    -0.70
    yip
    -0.67
    POSITIVE LOGITS
     Alone
    0.77
     annihil
    0.76
     satisfied
    0.71
     convinced
    0.69
     delighted
    0.68
    bane
    0.68
     impressed
    0.67
     unp
    0.65
     Dracula
    0.64
     inco
    0.64
    Act Density 0.013%

    No Known Activations