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

    Neuronpedia

    INDEX
    Explanations

    technical writing snippets

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric tokens—especially numbers and decimal values.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    the presence of numeric quantities (numbers and measurements such as distances, years, counts, temperatures).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    ो,
    -0.07
    ازل
    -0.06
    Interval
    -0.06
    cret
    -0.06
    Property
    -0.06
    _chk
    -0.06
     Digital
    -0.06
    ceased
    -0.06
    suffix
    -0.06
    网络
    -0.06
    POSITIVE LOGITS
     その他
    0.07
     Arg
    0.07
    achsen
    0.06
     Thesis
    0.06
     gra
    0.06
    sat
    0.06
    مر
    0.06
    ところ
    0.06
    ((↵
    0.06
    urrences
    0.06
    Act Density 0.676%

    No Known Activations