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

    Neuronpedia

    INDEX
    Explanations

    long

    np_max-act · gemini-2.0-flash

    The neuron detects language signaling long‐standing states or enduring problems—especially phrases like “long been” or “suffered from.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     unstoppable
    -0.07
     функци
    -0.07
     Pil
    -0.06
     maximum
    -0.06
     appraisal
    -0.06
     Broadway
    -0.06
     případ
    -0.06
     тка
    -0.06
    /dd
    -0.06
    exclusive
    -0.06
    POSITIVE LOGITS
     longtime
    0.09
     longstanding
    0.08
     long
    0.08
    :self
    0.07
     hovered
    0.07
    msg
    0.07
    Smoke
    0.07
    _STOP
    0.07
     коп
    0.07
    身
    0.07
    Act Density 0.020%

    No Known Activations