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

    Neuronpedia

    INDEX
    Explanations

    conflict, war, espionage

    np_max-act · gemini-2.0-flash

    The neuron detects cautionary or warning language signaling avoidance of detection or attention (e.g. instructions to be “careful,” “avoid,” or not attract notice).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    됨
    -0.06
    ).[
    -0.06
     повідом
    -0.06
     cinemas
    -0.06
    陵
    -0.06
     Traditional
    -0.06
    .[
    -0.06
     zoo
    -0.06
     لن
    -0.06
     sordu
    -0.06
    POSITIVE LOGITS
    .parser
    0.07
    38
    0.07
    Serialization
    0.06
    .accuracy
    0.06
     triang
    0.06
     выбор
    0.06
     inoc
    0.06
    니아
    0.06
    Dispatcher
    0.06
     seedu
    0.06
    Act Density 0.018%

    No Known Activations