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

    Neuronpedia

    INDEX
    Explanations

    phrases related to authority or rules

    oai_token-act-pair · gpt-3.5-turbo

    phrases that include the word "according to."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     appealed
    -0.77
    gage
    -0.76
    aukee
    -0.73
     reunited
    -0.72
    ocket
    -0.72
    icably
    -0.70
    oval
    -0.70
    rows
    -0.69
    ulence
    -0.68
    ruce
    -0.68
    POSITIVE LOGITS
     folklore
    0.75
    ÃĽ
    0.69
     Folk
    0.69
     tradition
    0.69
    Lenin
    0.66
     directions
    0.66
     Parables
    0.66
    Ô
    0.66
     scripture
    0.65
     Tradition
    0.65
    Act Density 0.071%

    No Known Activations