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

    Neuronpedia

    INDEX
    Explanations

    personalized messages asking for approval or support

    oai_token-act-pair · gpt-3.5-turbo

    direct addresses or references to the reader

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ipal
    -0.84
    asio
    -0.70
    inct
    -0.67
     Cook
    -0.67
    ape
    -0.67
    apes
    -0.67
     Verge
    -0.66
     Dispatch
    -0.66
    acular
    -0.66
    otom
    -0.65
    POSITIVE LOGITS
     guys
    1.24
    're
    1.17
    tub
    1.07
    've
    0.92
    'll
    0.89
     sir
    0.89
    RS
    0.87
     know
    0.85
     filthy
    0.84
     gentlemen
    0.80
    Act Density 0.255%

    No Known Activations