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

    Neuronpedia

    INDEX
    Explanations

    instances where something is powered by a specific source or technology

    oai_token-act-pair · gpt-3.5-turbo

    references to energy sources or technologies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Archbishop
    -0.70
     acquaintance
    -0.62
     transgress
    -0.62
     Kham
    -0.59
    say
    -0.59
    llah
    -0.58
    erent
    -0.58
     Meadows
    -0.58
    atz
    -0.58
     Tao
    -0.57
    POSITIVE LOGITS
     powered
    3.71
     Powered
    2.55
    powered
    2.49
     powering
    2.13
     fueled
    1.67
     propelled
    1.67
     fuelled
    1.57
    owered
    1.28
    Power
    1.16
     driven
    1.15
    Act Density 0.011%

    No Known Activations