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

    Neuronpedia

    INDEX
    Explanations

    phrases ending in a period

    oai_token-act-pair · gpt-3.5-turbo

    punctuations, specifically periods and exclamation points

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     quiet
    -0.81
     purse
    -0.79
     savage
    -0.73
     crocod
    -0.73
     wildlife
    -0.73
     paid
    -0.71
     wilderness
    -0.70
     gown
    -0.70
     soup
    -0.70
     coral
    -0.69
    POSITIVE LOGITS
     Additionally
    1.33
     Therefore
    1.25
     However
    1.24
     Furthermore
    1.20
     Consequently
    1.19
     Conversely
    1.17
     Accordingly
    1.17
     Moreover
    1.15
     Alternatively
    1.15
     Likewise
    1.12
    Act Density 0.885%

    No Known Activations