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

    Neuronpedia

    INDEX
    Explanations

    names of specific characters

    oai_token-act-pair · gpt-3.5-turbo

    references to the character Elsa

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    upon
    -0.68
    score
    -0.62
    igious
    -0.62
    eal
    -0.62
    link
    -0.61
    abol
    -0.61
     IMAGES
    -0.61
    aka
    -0.61
    oor
    -0.61
    ochond
    -0.60
    POSITIVE LOGITS
    Elsa
    1.22
     Elsa
    1.20
    issance
    0.88
    ipeg
    0.84
    éĹĺ
    0.82
     Anna
    0.80
    Anna
    0.79
    theless
    0.79
    ette
    0.76
     "$:/
    0.71
    Act Density 0.007%

    No Known Activations