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

    Neuronpedia

    INDEX
    Explanations

    than any other

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ки
    1.61
    ل
    1.45
    ε
    1.40
    ல்
    1.34
    களில்
    1.30
    ית
    1.29
    ১০
    1.27
    luss
    1.23
    ер
    1.22
    י
    1.22
    POSITIVE LOGITS
     относится
    1.09
     Heav
    1.05
     Strategy
    1.04
        
    1.03
    else
    1.02
     φορά
    1.01
    璀
    1.00
     سایر
    0.99
     Diameter
    0.99
    ↵
    0.98
    Act Density 0.061%

    No Known Activations