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

    Neuronpedia

    INDEX
    Explanations

    statements ending with a period and containing numbers

    oai_token-act-pair · gpt-3.5-turbo

    sentences that signify the conclusion of thoughts or statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     volunte
    -0.91
     suspic
    -0.89
     gobl
    -0.88
     satell
    -0.86
    ŃĶ
    -0.83
     tremend
    -0.81
     abduct
    -0.81
     toget
    -0.80
     teasp
    -0.78
     affili
    -0.78
    POSITIVE LOGITS
     That
    1.45
     Instead
    1.40
     Similarly
    1.40
     Likewise
    1.34
     Nonetheless
    1.32
     Ironically
    1.30
     Nevertheless
    1.30
     Ultimately
    1.28
     Those
    1.28
     Consequently
    1.27
    Act Density 0.438%

    No Known Activations