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

    Neuronpedia

    INDEX
    Explanations

    keywords related to espionage and secret information

    oai_token-act-pair · gpt-3.5-turbo

    references to hidden or secret knowledge related to world events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Weld
    -0.70
     âī¡
    -0.68
     Ross
    -0.66
     Clyde
    -0.63
     Irwin
    -0.62
     Tenn
    -0.62
     Ripple
    -0.62
     Kathryn
    -0.61
     Conflict
    -0.60
     Kenny
    -0.60
    POSITIVE LOGITS
    gallery
    0.91
    hillary
    0.91
    recomm
    0.90
    photos
    0.90
    dead
    0.88
    cond
    0.88
    terms
    0.88
    committee
    0.87
    cook
    0.87
    warm
    0.87
    Act Density 0.074%

    No Known Activations