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

    Neuronpedia

    INDEX
    Explanations

    the end of sentences

    oai_token-act-pair · gpt-3.5-turbo

    sentence-ending periods

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     tremend
    -1.04
     horrend
    -0.87
     gobl
    -0.79
     carbohyd
    -0.78
     psychiat
    -0.77
     thous
    -0.77
     dracon
    -0.76
     teasp
    -0.75
     desper
    -0.75
     advoc
    -0.75
    POSITIVE LOGITS
    ↵
    1.41
     Lastly
    1.36
    <|endoftext|>
    1.32
     Finally
    1.26
     Meanwhile
    1.20
     Eventually
    1.17
     Ultimately
    1.16
     Nonetheless
    1.13
     Earlier
    1.10
     Regardless
    1.10
    Act Density 0.532%

    No Known Activations