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

    Neuronpedia

    INDEX
    Explanations

    declarations of welcoming or inviting actions

    oai_token-act-pair · gpt-3.5-turbo

    the term "welcome."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    arcity
    -0.80
    pard
    -0.75
    ynasty
    -0.75
    iph
    -0.72
    ocrat
    -0.71
    sis
    -0.71
    uyomi
    -0.71
    ikuman
    -0.71
    iple
    -0.69
    ols
    -0.69
    POSITIVE LOGITS
     additions
    0.93
     welcome
    0.91
     aboard
    0.85
     guests
    0.82
     newcomers
    0.80
     welcomes
    0.78
     visitors
    0.77
     welcoming
    0.74
    ãĤī
    0.74
     Guest
    0.73
    Act Density 0.020%

    No Known Activations