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

    Neuronpedia

    INDEX
    Explanations

    hearts (♥) followed by characters in a unique format

    oai_token-act-pair · gpt-3.5-turbo

    instances of potential negative outcomes or consequences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     estranged
    -0.64
    illac
    -0.62
     honoured
    -0.61
     orchestr
    -0.60
     unaff
    -0.60
     Classics
    -0.60
     extrad
    -0.59
     Compass
    -0.58
     diving
    -0.58
     orche
    -0.58
    POSITIVE LOGITS
    ³³³³³³³³
    1.25
    Posted
    1.24
    ³³³³
    1.20
    Anonymous
    1.18
    ³³³
    1.15
    ³³³³³³³³³³³³³³³³
    1.10
    posted
    1.04
    Anyway
    1.02
    ³³
    0.97
    Rated
    0.95
    Act Density 0.477%

    No Known Activations