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

    Neuronpedia

    INDEX
    Explanations

    words related to a negative judgment or lack of respect towards someone or something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "contempt" and related expressions of disdain or mistrust

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     hemor
    -0.72
     helicop
    -0.64
     scen
    -0.64
    ramid
    -0.62
     encyclopedia
    -0.62
     Lans
    -0.61
     toget
    -0.61
     advoc
    -0.60
     enthusi
    -0.60
    NetMessage
    -0.60
    POSITIVE LOGITS
    uous
    1.27
    uously
    1.26
    fully
    1.06
    acy
    0.97
    ful
    0.95
    ible
    0.93
    ibly
    0.91
    fulness
    0.90
    uality
    0.87
    ibles
    0.86
    Act Density 0.025%

    No Known Activations