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

    Neuronpedia

    INDEX
    Explanations

    the word "hundreds" followed by a quantity or descriptor

    oai_token-act-pair · gpt-3.5-turbo

    the expression of large quantities or numbers of people or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ieu
    -0.67
     fixation
    -0.65
    hawks
    -0.64
    position
    -0.63
    swick
    -0.62
    llor
    -0.62
    atorium
    -0.61
    aker
    -0.60
    ax
    -0.60
    forward
    -0.60
    POSITIVE LOGITS
     thousands
    1.06
     tens
    0.98
     thous
    0.97
     hundreds
    0.95
    undreds
    0.95
    ousands
    0.92
     millions
    0.91
    ourcing
    0.88
     thousand
    0.88
    chool
    0.82
    Act Density 0.010%

    No Known Activations