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

    Neuronpedia

    INDEX
    Explanations

    quantitative references, such as counts or numbers

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate quantities or frequencies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ħĭ
    -0.81
    uer
    -0.74
    arium
    -0.72
    gypt
    -0.72
    ĸļ
    -0.71
    ļéĨĴ
    -0.71
    é¾įåĸļ士
    -0.69
    soever
    -0.67
    ree
    -0.65
    resy
    -0.64
    POSITIVE LOGITS
     opportunities
    0.71
     individuals
    0.70
     instances
    0.70
     households
    0.68
     consecutive
    0.67
     people
    0.67
     resources
    0.66
     dollars
    0.66
     times
    0.66
     seats
    0.65
    Act Density 0.149%

    No Known Activations