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

    Neuronpedia

    INDEX
    Explanations

    mathematical summation notation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     protégé
    0.80
     moniker
    0.70
     SalesLT
    0.63
     उतारा
    0.63
     protég
    0.62
    からない
    0.60
     wizard
    0.60
    율
    0.59
     administrators
    0.58
     tackling
    0.57
    POSITIVE LOGITS
    substack
    1.25
    ୀ
    1.06
     만족
    0.95
    限
    0.94
    forever
    0.92
    től
    0.91
    infty
    0.88
     infinite
    0.88
    Fork
    0.87
     பரு
    0.85
    Act Density 0.037%

    No Known Activations