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

    Neuronpedia

    INDEX
    Explanations

    people's names and physical attributes (such as owner, spokesman) in sentences

    oai_token-act-pair · gpt-3.5-turbo

    references to male speakers or subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    cial
    -0.66
    duc
    -0.64
    odder
    -0.63
     Gad
    -0.61
     Claire
    -0.61
     Fashion
    -0.60
     Mandatory
    -0.60
     Medicine
    -0.58
     Marriage
    -0.58
     Transactions
    -0.57
    POSITIVE LOGITS
    'd
    1.08
    'll
    0.90
     encount
    0.88
     suspic
    0.83
    resy
    0.83
    've
    0.82
    ©¶æ
    0.76
     couldn
    0.75
    ppard
    0.73
     agre
    0.73
    Act Density 0.155%

    No Known Activations