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

    Neuronpedia

    INDEX
    Explanations

    percentage

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric figures—especially percentages—highlighting numerical data like “12.3%,” “11.7%,” etc.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     아이디
    -0.06
    ahrungen
    -0.06
     billionaire
    -0.06
    	unsigned
    -0.06
    妇
    -0.06
    impse
    -0.06
    _Release
    -0.06
    (Collision
    -0.06
    -Cds
    -0.06
    MOVED
    -0.06
    POSITIVE LOGITS
     shredded
    0.07
     relentless
    0.06
     Isles
    0.06
     contexts
    0.06
     향
    0.06
    			↵
    0.06
    雪
    0.06
    dan
    0.06
     months
    0.06
     Calcium
    0.06
    Act Density 0.004%

    No Known Activations