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

    Neuronpedia

    INDEX
    Explanations

    sequences of words that look like a special character followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    descriptions of performance and confidence in artistic settings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ertodd
    -0.83
    itionally
    -0.78
    osition
    -0.78
    hement
    -0.77
    iage
    -0.77
    terness
    -0.75
    etz
    -0.73
    ername
    -0.72
    erville
    -0.71
    ĸļ
    -0.71
    POSITIVE LOGITS
    ORN
    0.85
    WIND
    0.80
    BuyableInstoreAndOnline
    0.78
    é¾
    0.74
    Balt
    0.73
    Reward
    0.73
    voy
    0.70
    Virgin
    0.70
    âĵĺ
    0.68
    Mu
    0.68
    Act Density 0.181%

    No Known Activations