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

    Neuronpedia

    INDEX
    Explanations

    T

    np_max-act · gemini-2.0-flash

    This neuron primarily activates on numeric tokens (e.g. years, article numbers, and other standalone numbers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _EVT
    -0.07
    ={`${
    -0.06
    -0.06
    App
    -0.06
     Chevy
    -0.06
    Bulletin
    -0.06
     Jackets
    -0.06
    Server
    -0.06
    Ce
    -0.06
     tease
    -0.06
    POSITIVE LOGITS
    itical
    0.07
    _machine
    0.07
     پرو
    0.06
     città
    0.06
    oteca
    0.06
    �
    0.06
     Esta
    0.06
     publi
    0.06
    selectors
    0.06
     bekl
    0.06
    Act Density 0.024%

    No Known Activations