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

    Neuronpedia

    INDEX
    Explanations

    names of people and their associations with various roles

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     retire
    -0.20
     widow
    -0.19
     decades
    -0.19
     Widow
    -0.17
     wives
    -0.17
     retired
    -0.17
     wid
    -0.17
     elderly
    -0.16
     veteran
    -0.16
     retirees
    -0.16
    POSITIVE LOGITS
     young
    0.21
    young
    0.20
     boyfriend
    0.18
     underage
    0.17
     girl
    0.17
     boy
    0.17
    @student
    0.17
     preco
    0.17
     jeune
    0.16
     Young
    0.16
    Act Density 0.682%

    No Known Activations