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

    Neuronpedia

    INDEX
    Explanations

    locations or countries

    oai_token-act-pair · gpt-3.5-turbo

    references to Belgium

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hod
    -0.74
     Rays
    -0.74
     Gore
    -0.73
    HUD
    -0.73
     ram
    -0.69
     rays
    -0.68
    iffin
    -0.63
     scal
    -0.63
     Jenna
    -0.62
    mod
    -0.62
    POSITIVE LOGITS
     Belgium
    3.75
     Belgian
    3.27
     Belg
    2.56
     Brussels
    2.11
     Luxembourg
    1.89
     Belarus
    1.82
    Dutch
    1.67
     Netherlands
    1.67
     Cameroon
    1.60
    Bel
    1.58
    Act Density 0.015%

    No Known Activations