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

    Neuronpedia

    INDEX
    Explanations

    numerical values related to quantity, such as "one," "both," and "five."

    oai_token-act-pair · gpt-3.5-turbo

    numerical values and terms expressing quantity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     helicop
    -0.72
     Niet
    -0.70
    Mobil
    -0.68
     Seym
    -0.68
     Jagu
    -0.66
     Mobil
    -0.62
     Azerb
    -0.62
    ãĥ¼ãĥ³
    -0.61
     Vaugh
    -0.61
    edIn
    -0.60
    POSITIVE LOGITS
     ][
    0.86
     )]
    0.66
     )))
    0.65
     largeDownload
    0.64
     ));
    0.61
    nces
    0.61
     partName
    0.60
     ))
    0.60
     ]
    0.60
     ):
    0.60
    Act Density 0.087%

    No Known Activations