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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric tokens (e.g. standalone digits, multi‐digit numbers, array indices, timestamps, footnote labels).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ONGODB
    -0.07
     Milk
    -0.06
    YSTEM
    -0.06
    oltage
    -0.06
     Point
    -0.06
     heartbreaking
    -0.06
     Kir
    -0.06
     pornografia
    -0.06
    IEW
    -0.06
    COPY
    -0.06
    POSITIVE LOGITS
     rgb
    0.08
     itertools
    0.07
     sectors
    0.07
    積
    0.06
    -have
    0.06
     bat
    0.06
     chairs
    0.06
     ries
    0.06
     risky
    0.06
    standing
    0.06
    Act Density 0.108%

    No Known Activations