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

    Neuronpedia

    INDEX
    Explanations

    webpage welcome messages

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Welcome."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    negie
    -0.80
    orius
    -0.68
    appropri
    -0.68
    river
    -0.67
    onel
    -0.65
    ieth
    -0.65
    ole
    -0.65
     forgiven
    -0.64
    riter
    -0.62
    rained
    -0.62
    POSITIVE LOGITS
    elcome
    0.93
    prise
    0.89
     Welcome
    0.89
    Reviewer
    0.86
    Welcome
    0.82
     Surprise
    0.81
    giving
    0.80
     Guest
    0.79
    ISSION
    0.79
    ISTER
    0.78
    Act Density 0.021%

    No Known Activations