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

    Neuronpedia

    INDEX
    Explanations

    sentences ending with a period followed by numbers

    oai_token-act-pair · gpt-3.5-turbo

    punctuation or pause indicators within the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    mith
    -0.77
    ppelin
    -0.68
    ori
    -0.65
    ãĥ´
    -0.59
    ogle
    -0.59
     infl
    -0.59
     Ranch
    -0.59
     aquarium
    -0.58
     patriarch
    -0.57
    essel
    -0.57
    POSITIVE LOGITS
     ."
    1.11
    shall
    0.93
    until
    0.92
     ,"
    0.90
    unless
    0.80
    FORE
    0.77
     Ibid
    0.75
    WHERE
    0.74
    relevant
    0.73
    being
    0.73
    Act Density 0.020%

    No Known Activations