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

    Neuronpedia

    INDEX
    Explanations

    references to names or terms with 'hu' as part of them

    oai_token-act-pair · gpt-3.5-turbo

    the presence of names, particularly those containing the syllable "hu"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    bread
    -0.71
    lining
    -0.65
    cor
    -0.64
    crop
    -0.63
    papers
    -0.62
    turn
    -0.61
    cloth
    -0.61
    book
    -0.61
    orius
    -0.61
    nings
    -0.60
    POSITIVE LOGITS
    awei
    1.35
    isine
    1.01
    pta
    0.97
    ulkan
    0.96
    pload
    0.94
    isu
    0.94
    izen
    0.90
    ilty
    0.89
    ricanes
    0.88
    ionage
    0.86
    Act Density 0.025%

    No Known Activations