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

    Neuronpedia

    INDEX
    Explanations

    float or constant buffers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     выяс
    0.44
     खड़ी
    0.44
     прямоуго
    0.44
     экспери
    0.43
     энер
    0.43
     אך
    0.43
     wzglę
    0.42
     직선
    0.42
     }-\
    0.41
     ตาม
    0.41
    POSITIVE LOGITS
    イ
    0.45
     Kobayashi
    0.44
    设备
    0.43
    Peter
    0.42
    特殊
    0.42
    ikovsky
    0.41
     tom
    0.40
     sockets
    0.40
    何
    0.40
    判
    0.40
    Act Density 0.000%

    No Known Activations