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

    Neuronpedia

    INDEX
    Explanations

    se/si

    np_max-act · gemini-2.0-flash

    The neuron activates on Portuguese text—it flags words and phrases (e.g. “Se tens…”, “Você”) that mark the start of Portuguese clauses.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     jogo
    -0.07
     Longitude
    -0.07
     vznik
    -0.07
     Three
    -0.07
    ologists
    -0.07
    .lineWidth
    -0.07
    _issues
    -0.07
    3
    -0.07
     */)
    -0.06
    -hit
    -0.06
    POSITIVE LOGITS
     Se
    0.07
    Se
    0.07
    Khi
    0.07
     se
    0.07
    	se
    0.06
    σι
    0.06
     если
    0.06
    рование
    0.06
    .Se
    0.06
    Якщо
    0.06
    Act Density 0.028%

    No Known Activations