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

    Neuronpedia

    INDEX
    Explanations

    prepositions followed by nouns that suggest hierarchy or a ranking

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the phrase "top of the" or similar phrases indicating a ranking or position

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Äĩ
    -0.84
    soever
    -0.75
    gans
    -0.69
    venants
    -0.66
    iances
    -0.65
    rw
    -0.65
    ary
    -0.63
    yy
    -0.63
    istas
    -0.63
    forts
    -0.63
    POSITIVE LOGITS
     earners
    0.83
     nowhere
    0.74
     Rampage
    0.68
     contention
    0.67
     charts
    0.67
     shelf
    0.66
     heap
    0.66
    paying
    0.66
     hill
    0.65
     shelves
    0.64
    Act Density 0.068%

    No Known Activations