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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    This neuron does not respond to any ordinary text tokens—it remains silent on all shown content. In other words, it’s effectively inactive, finding no meaningful pattern in the input.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     жов
    -0.07
     finishes
    -0.07
     malignant
    -0.06
    ーン
    -0.06
     Fatal
    -0.06
    _fin
    -0.06
    CppMethodIntialized
    -0.06
     trọng
    -0.06
     cli
    -0.06
     مال
    -0.06
    POSITIVE LOGITS
    Bag
    0.07
     García
    0.07
    osta
    0.06
    EMA
    0.06
    odafone
    0.06
    ][/
    0.06
    řev
    0.06
    iera
    0.06
    _variance
    0.06
    净
    0.06
    Act Density 0.024%

    No Known Activations