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

    Neuronpedia

    INDEX
    Explanations

    phrases related to essential foundations or origins

    oai_token-act-pair · gpt-3.5-turbo

    phrases that reference foundational concepts or elements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    UX
    -0.73
    PRESS
    -0.70
    Cent
    -0.67
    Pont
    -0.66
    bas
    -0.66
    Charg
    -0.62
    apt
    -0.61
    Problem
    -0.61
    marg
    -0.61
    DIT
    -0.61
    POSITIVE LOGITS
    soever
    1.11
    xual
    0.80
    izens
    0.66
     leases
    0.63
    adoes
    0.62
     gobl
    0.62
     defenses
    0.61
     specializes
    0.61
     liqu
    0.61
     governs
    0.60
    Act Density 0.037%

    No Known Activations