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

    Neuronpedia

    INDEX
    Explanations

    we see

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     svoj
    0.45
     اپنا
    0.45
     svoju
    0.44
     свой
    0.43
     aims
    0.42
     свою
    0.42
     confers
    0.42
     easy
    0.40
     intends
    0.40
    Easy
    0.40
    POSITIVE LOGITS
     знаем
    0.81
     know
    0.69
     biết
    0.68
     sabemos
    0.67
     sappiamo
    0.66
    know
    0.66
     weten
    0.60
     увидеть
    0.59
    知道
    0.59
     شاهد
    0.58
    Act Density 0.047%

    No Known Activations