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

    Neuronpedia

    INDEX
    Explanations

    names or words with accents or special characters

    oai_token-act-pair · gpt-3.5-turbo

    the character "r" in various forms, especially within names and phrases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     rooting
    -0.88
    iating
    -0.68
    ollah
    -0.65
     sanctuary
    -0.65
    oys
    -0.62
    ****************
    -0.62
    owered
    -0.62
     dividing
    -0.61
    ously
    -0.60
    osite
    -0.60
    POSITIVE LOGITS
    ré
    1.12
     ré
    0.95
     é
    0.92
    ité
    0.89
    nant
    0.86
     Ré
    0.85
    lé
    0.84
     Qué
    0.82
    miah
    0.82
     propos
    0.82
    Act Density 0.007%

    No Known Activations