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

    Neuronpedia

    INDEX
    Explanations

    sentences that end with punctuation marks like periods and question marks

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks used for ending statements and questions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     forwarding
    -0.66
     exclusively
    -0.65
    withstanding
    -0.64
     stout
    -0.63
    ensibly
    -0.63
     intended
    -0.62
     challeng
    -0.60
     extinguished
    -0.60
     pledged
    -0.60
     complete
    -0.60
    POSITIVE LOGITS
     Anyway
    1.28
     Lastly
    1.22
     Secondly
    1.19
     Therefore
    1.19
     Otherwise
    1.15
     However
    1.12
     Unfortunately
    1.07
     Moreover
    1.07
     But
    1.06
     Finally
    1.06
    Act Density 0.768%

    No Known Activations