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

    Neuronpedia

    INDEX
    Explanations

    words related to specific cultural or identity groups

    oai_token-act-pair · gpt-3.5-turbo

    references to specific social and identity categories, particularly those connected to gender, ethnicity, nationality, and societal roles

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    epad
    -0.83
    liga
    -0.80
    creen
    -0.78
    ournal
    -0.76
    oscope
    -0.76
    ogle
    -0.75
    rator
    -0.74
    package
    -0.74
    genre
    -0.73
    »Ĵ
    -0.73
    POSITIVE LOGITS
     involvement
    1.34
     sovereignty
    1.21
     participation
    1.19
     dominance
    1.11
     supremacy
    1.11
     discontent
    1.11
     complicity
    1.10
     ingenuity
    1.09
     hegemony
    1.08
     reliance
    1.07
    Act Density 0.353%

    No Known Activations