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

    Neuronpedia

    INDEX
    Explanations

    at, a

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens (e.g. times, day numbers, and other digits) in travel-itinerary text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ary
    -0.07
     sinks
    -0.07
    _day
    -0.07
     fotos
    -0.07
     drink
    -0.06
     skill
    -0.06
     inputFile
    -0.06
     lung
    -0.06
    ์และ
    -0.06
    utowired
    -0.06
    POSITIVE LOGITS
    另一
    0.06
    utenberg
    0.06
     прави
    0.06
    -delete
    0.06
     Yuan
    0.06
    沈
    0.06
    __;
    0.06
    Translations
    0.06
    ريد
    0.06
     jeszcze
    0.06
    Act Density 0.013%

    No Known Activations