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

    Neuronpedia

    INDEX
    Explanations

    references to various groups or categories of people and their characteristics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     individuals
    -0.70
     människor
    -0.66
     mensen
    -0.66
     people
    -0.65
    individuals
    -0.63
    eseorang
    -0.61
     itself
    -0.61
     Itself
    -0.59
    itself
    -0.59
    somebody
    -0.59
    POSITIVE LOGITS
     whom
    0.98
    whom
    0.77
     whose
    0.63
     who
    0.58
    whose
    0.57
     opposing
    0.54
    Whom
    0.54
    who
    0.52
     professions
    0.52
     Whom
    0.51
    Act Density 0.690%

    No Known Activations