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

    Neuronpedia

    INDEX
    Explanations

    sentence

    np_max-act · gemini-2.0-flash

    This neuron responds to Cyrillic subword tokens, in particular fragments of Russian linguistic-terminology words (e.g. pieces of “синтагматика,” “парадигматика,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Fernandez
    -0.07
    gran
    -0.07
    sales
    -0.07
     Kul
    -0.07
     能
    -0.07
     Sons
    -0.07
     crane
    -0.06
     emphasizes
    -0.06
    Numer
    -0.06
    (Const
    -0.06
    POSITIVE LOGITS
    三个
    0.06
     Meadow
    0.06
     ثبت
    0.06
    .timedelta
    0.06
     $(
    0.06
    Ư�
    0.06
     토
    0.06
    -leaning
    0.06
     Ρ
    0.06
     onDestroy
    0.06
    Act Density 0.031%

    No Known Activations