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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating uncertainty and the need for further information or investigation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Cabin
    -0.64
     Nurs
    -0.60
     Tea
    -0.60
     Converted
    -0.58
    hell
    -0.58
    ront
    -0.57
     Virtue
    -0.57
     angered
    -0.56
     Hort
    -0.56
    cellence
    -0.55
    POSITIVE LOGITS
     definitively
    1.16
     conclusive
    1.09
     specifics
    1.04
     definitive
    1.01
     confir
    0.99
     extrap
    0.94
     anecd
    0.91
     conjecture
    0.90
     estimates
    0.88
     estimate
    0.88
    Act Density 0.230%

    No Known Activations