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

    Neuronpedia

    INDEX
    Explanations

    record

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on numeric tokens—especially decimals and section or case citation numbers.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     IPT
    -0.07
     mannen
    -0.07
    392
    -0.06
    inction
    -0.06
    nilai
    -0.06
     Transformers
    -0.06
     ucz
    -0.06
    bing
    -0.06
    ceries
    -0.06
    enza
    -0.06
    POSITIVE LOGITS
    лся
    0.07
     서울특별시
    0.07
     record
    0.07
     Christina
    0.07
     pests
    0.06
    <hr
    0.06
     alongside
    0.06
     Veteran
    0.06
     Họ
    0.06
     occasional
    0.06
    Act Density 0.001%

    No Known Activations