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

    Neuronpedia

    INDEX
    Explanations

    phrases related to specific regions or nationalities, such as "Hong Kong" and "Cantonese."

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "ones" and related terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     exception
    -0.69
    ALLY
    -0.67
    LESS
    -0.63
    ECH
    -0.62
     grou
    -0.61
    ISON
    -0.60
    ãģ®å®
    -0.59
    istically
    -0.58
    BIL
    -0.58
     WC
    -0.57
    POSITIVE LOGITS
    pace
    1.71
    hift
    1.54
    paces
    1.52
    creen
    1.47
    ystem
    1.46
    ilver
    1.42
    peed
    1.41
    hip
    1.38
    cript
    1.33
    pring
    1.32
    Act Density 0.094%

    No Known Activations