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

    Neuronpedia

    INDEX
    Explanations

    ross

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of the word “cross” in legal document headings, especially as part of hyphenated terms like “cross-appeal” or “cross-appellant.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    KeySpec
    -0.06
     rootReducer
    -0.06
    sız
    -0.06
    核
    -0.06
     resultados
    -0.06
     Mädchen
    -0.06
     Egyptian
    -0.06
     anz
    -0.06
     descon
    -0.06
    /Subthreshold
    -0.06
    POSITIVE LOGITS
    шим
    0.06
    ("""↵
    0.06
    Δ
    0.06
    ''
    0.06
     transactions
    0.06
     diapers
    0.06
     unfit
    0.06
     skin
    0.06
     Κατηγορία
    0.06
    .jupiter
    0.06
    Act Density 0.002%

    No Known Activations