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

    Neuronpedia

    INDEX
    Explanations

    terms related to race and mythology, particularly focusing on the word "Negro" and words associated with gods and goddesses

    oai_token-act-pair · gpt-3.5-turbo

    references to the term "Negro" and terms related to hierarchy or authority

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    izoph
    -0.73
    OTT
    -0.70
    rost
    -0.67
     LAW
    -0.67
     rh
    -0.66
     Finn
    -0.64
     thirds
    -0.64
    anners
    -0.63
    ©¶æ
    -0.63
     Heb
    -0.62
    POSITIVE LOGITS
    es
    2.01
    esville
    1.09
    edIn
    1.06
    engers
    1.04
    esian
    1.03
    ively
    1.02
    ername
    0.99
    ed
    0.98
    eson
    0.97
    esis
    0.94
    Act Density 0.038%

    No Known Activations