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

    Neuronpedia

    INDEX
    Explanations

    Megapixels

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens—especially measurements, counts, and decimal values.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    �ng
    -0.07
    LAN
    -0.07
     至
    -0.07
     Pra
    -0.06
    ',[
    -0.06
     첨부파일
    -0.06
     Thus
    -0.06
     заб
    -0.06
     Term
    -0.06
     conn
    -0.06
    POSITIVE LOGITS
    still
    0.07
    teří
    0.07
     Gibbs
    0.07
    apixel
    0.07
     такая
    0.07
     такой
    0.07
    xff
    0.07
    ault
    0.06
    臨
    0.06
     管
    0.06
    Act Density 0.002%

    No Known Activations