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

    Neuronpedia

    INDEX
    Explanations

    phrases with superlatives and extreme descriptions

    oai_token-act-pair · gpt-3.5-turbo

    phrases that emphasize the concept of "worst" or extreme negative circumstances

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Ô
    -0.71
    EStream
    -0.71
    ãĤ¼ãĤ¦ãĤ¹
    -0.68
    culosis
    -0.68
    YA
    -0.64
    ĸļ
    -0.62
    assetsadobe
    -0.62
    resy
    -0.62
     Cosponsors
    -0.61
    krit
    -0.61
    POSITIVE LOGITS
     course
    0.93
     ours
    0.85
     sorts
    0.82
     humankind
    0.75
     humanity
    0.69
     theirs
    0.69
    course
    0.69
    icial
    0.69
     hers
    0.65
     mankind
    0.61
    Act Density 0.118%

    No Known Activations