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

    Neuronpedia

    INDEX
    Explanations

    follows a pattern

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     monomers
    0.40
    漑
    0.39
     alc
    0.39
     burg
    0.39
     celebrities
    0.39
     ngữ
    0.38
     capitalists
    0.38
     consp
    0.38
     fases
    0.37
     ***!
    0.36
    POSITIVE LOGITS
    Eureka
    0.38
    Vector
    0.37
     ተ
    0.37
    Allium
    0.37
    豢
    0.37
    ⇉
    0.37
    Lieutenant
    0.37
    Oke
    0.37
     प्रतिसाद
    0.37
     Sichuan
    0.37
    Act Density 0.000%

    No Known Activations