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

    Neuronpedia

    INDEX
    Explanations

    phrases related to social classes

    oai_token-act-pair · gpt-3.5-turbo

    references to social and economic classes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Leaks
    -0.78
    sure
    -0.68
    reb
    -0.66
    aye
    -0.66
    ay
    -0.65
    oshenko
    -0.63
     Affairs
    -0.62
    oil
    -0.62
    areth
    -0.62
    ews
    -0.61
    POSITIVE LOGITS
     class
    3.82
    class
    2.92
     classes
    2.85
     Class
    2.67
    Class
    2.48
     Classes
    2.36
     CLASS
    2.25
    classes
    2.23
     subclass
    2.09
    CLASS
    1.81
    Act Density 0.024%

    No Known Activations