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

    Neuronpedia

    INDEX
    Explanations

    numbers

    np_max-act · gemini-2.0-flash

    The neuron fires on numeric version tokens, i.e. digits in software dependency or package version strings.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     combust
    -0.07
     شمال
    -0.07
    :\
    -0.07
    ?>;↵
    -0.07
    样的
    -0.07
     Gtk
    -0.07
     свеж
    -0.06
     leaving
    -0.06
     ضد
    -0.06
     responsible
    -0.06
    POSITIVE LOGITS
    WI
    0.06
    DO
    0.06
    roj
    0.06
    igram
    0.06
     vhodné
    0.06
     stressing
    0.05
     Wheat
    0.05
    _past
    0.05
    ENTS
    0.05
    orado
    0.05
    Act Density 0.012%

    No Known Activations