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

    Neuronpedia

    INDEX
    Explanations

    years, numbers, and foreign languages

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     de
    0.44
     l
    0.43
     optim
    0.42
    f
    0.40
     far
    0.39
     principalmente
    0.38
     Weil
    0.38
    cca
    0.38
    7
    0.38
     الو
    0.38
    POSITIVE LOGITS
    ,"@
    0.50
    毎年
    0.47
    ատ
    0.47
    初めて
    0.47
     മറ്റൊരു
    0.46
     AGAIN
    0.45
    某个
    0.44
     BOTH
    0.44
     каждую
    0.44
     আবার
    0.44
    Act Density 0.011%

    No Known Activations