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

    Neuronpedia

    INDEX
    Explanations

    news

    np_max-act · gemini-2.0-flash

    This neuron detects the presence of the Russian word “новости” (news/updates), i.e. it activates on text referring to news or announcements.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Blacks
    -0.07
     sánh
    -0.07
     Jain
    -0.06
    _issues
    -0.06
     deben
    -0.06
    _population
    -0.06
     बस
    -0.06
     Jame
    -0.06
     understand
    -0.06
     Ritual
    -0.06
    POSITIVE LOGITS
     grated
    0.07
    pad
    0.07
    edení
    0.07
    IMITER
    0.07
     GridLayout
    0.06
    _updates
    0.06
    ційний
    0.06
    _cate
    0.06
    mine
    0.06
     pä
    0.06
    Act Density 0.027%

    No Known Activations