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

    Neuronpedia

    INDEX
    Explanations

    Communication and requests

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on Cyrillic-script tokens (i.e. bits of Russian text).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    -icon
    -0.08
    _expiry
    -0.08
     airstrikes
    -0.06
     اق
    -0.06
     Hatch
    -0.06
    bundles
    -0.06
     derece
    -0.06
    ्रद
    -0.06
    _duplicates
    -0.06
     Střed
    -0.06
    POSITIVE LOGITS
     Perm
    0.07
    、『
    0.06
     `,
    0.06
    !',
    0.06
     školy
    0.06
     isi
    0.06
     Wort
    0.06
     ama
    0.06
    KHR
    0.06
     nearest
    0.06
    Act Density 0.049%

    No Known Activations