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

    Neuronpedia

    INDEX
    Explanations

    names containing the sequence "hel" along with high activation values

    oai_token-act-pair · gpt-3.5-turbo

    instances of the name "Hel" or words that commonly follow it

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Rated
    -0.77
    print
    -0.69
    GGGG
    -0.67
     nomine
    -0.61
    EED
    -0.60
     NTS
    -0.60
     infamous
    -0.59
    ItemTracker
    -0.59
     entangled
    -0.59
     Flight
    -0.58
    POSITIVE LOGITS
    tered
    1.11
    mand
    1.02
    iflower
    0.98
    ibrary
    0.91
    tering
    0.90
    ters
    0.90
    mes
    0.89
    iday
    0.87
    iths
    0.87
    brook
    0.86
    Act Density 0.013%

    No Known Activations