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

    Neuronpedia

    INDEX
    Explanations

    phrases related to quotation marks and direct speech

    oai_token-act-pair · gpt-3.5-turbo

    usage of quotation marks or direct speech

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Archdemon
    -0.86
     division
    -0.76
     blasts
    -0.73
     greatly
    -0.72
     anguish
    -0.72
     Klu
    -0.71
     candles
    -0.71
     buoy
    -0.70
     accomp
    -0.70
     regard
    -0.70
    POSITIVE LOGITS
    normal
    1.73
    official
    1.57
    natural
    1.57
    safe
    1.55
    traditional
    1.54
    cheat
    1.54
    regular
    1.51
    clean
    1.50
    pure
    1.50
    neutral
    1.48
    Act Density 0.119%

    No Known Activations