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

    Neuronpedia

    INDEX
    Explanations

    runs of characters starting with 'ev' as a prefix

    oai_token-act-pair · gpt-3.5-turbo

    references to evasion or avoidance

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    matically
    -0.73
    WAY
    -0.68
    manship
    -0.63
    WARE
    -0.63
     Franks
    -0.62
     Emirates
    -0.62
    matical
    -0.62
    cone
    -0.61
     spare
    -0.61
    WAYS
    -0.60
    POSITIVE LOGITS
    icted
    1.29
    idences
    1.27
    olve
    1.27
    asive
    1.24
    isions
    1.24
    olutions
    1.18
    ocation
    1.18
    ocative
    1.16
    idential
    1.13
    oked
    1.13
    Act Density 0.017%

    No Known Activations