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

    Neuronpedia

    INDEX
    Explanations

    sentences of strong opinions or judgments

    oai_token-act-pair · gpt-3.5-turbo

    sentences that express strong negative sentiments or judgments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     previously
    -0.73
     intended
    -0.72
     distribut
    -0.71
     formally
    -0.70
     dynamically
    -0.69
     inaccur
    -0.67
     identical
    -0.67
     commissioned
    -0.67
     extensively
    -0.67
     unlawfully
    -0.66
    POSITIVE LOGITS
     Anyway
    1.56
     Especially
    1.32
     Besides
    1.31
     Luckily
    1.27
     Otherwise
    1.26
     Thankfully
    1.23
     Lastly
    1.19
     Regardless
    1.17
     Eventually
    1.17
     Hence
    1.16
    Act Density 0.491%

    No Known Activations