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

    Neuronpedia

    INDEX
    Explanations

    educational

    np_max-act · gemini-2.0-flash

    The neuron flags words that describe a piece of content’s tone or purpose—especially adjectives like “educational,” “entertaining,” or “informative.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    fell
    -0.06
    rottle
    -0.06
    >;↵↵
    -0.06
    /of
    -0.06
    ?>&
    -0.06
    �
    -0.06
     крем
    -0.06
    -fold
    -0.06
    /slick
    -0.06
    _gpu
    -0.06
    POSITIVE LOGITS
     frontline
    0.07
     thicker
    0.07
    ��
    0.07
     mission
    0.07
     monthly
    0.06
     educational
    0.06
     mong
    0.06
    营
    0.06
     adına
    0.06
    肃
    0.06
    Act Density 0.024%

    No Known Activations