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

    Neuronpedia

    INDEX
    Explanations

    numerical values and quantities related to grouping or categorization

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    the word "three" or "two"

    np_acts-logits-general · gemini-2.0-flash

    counting numbers and following nouns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     flere
    -0.69
     alguno
    -0.66
     daugiau
    -0.63
     flera
    -0.61
     altro
    -0.60
     âmes
    -0.59
     enemmän
    -0.59
     tudo
    -0.57
     ciascuno
    -0.57
     chacun
    -0.56
    POSITIVE LOGITS
     remaining
    1.15
     largest
    1.08
     oldest
    1.05
    iniest
    1.05
     newest
    1.04
    liest
    1.03
     biggest
    1.00
     closest
    1.00
    iest
    0.99
     highest
    0.98
    Act Density 0.210%

    No Known Activations