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

    Neuronpedia

    INDEX
    Explanations

    positive or notable aspects within sentences

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate the best or worst aspects of experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Klux
    -0.86
    ternity
    -0.76
    berus
    -0.75
    elsius
    -0.73
     pione
    -0.73
     practition
    -0.73
    ãĤī
    -0.72
    lished
    -0.71
     cumbers
    -0.69
    arma
    -0.69
    POSITIVE LOGITS
    ials
    0.86
    uring
    0.72
    ially
    0.71
     Colbert
    0.70
    UID
    0.69
    ioned
    0.69
     Whedon
    0.65
    nered
    0.64
     (>
    0.63
     Hoo
    0.63
    Act Density 0.028%

    No Known Activations