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

    Neuronpedia

    INDEX
    Explanations

    one-size-fits-all

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    nya
    1.18
     one
    1.14
     одна
    1.05
     ones
    0.95
    দের
    0.93
    mice
    0.92
    า
    0.91
    ด์
    0.90
    一些
    0.90
    отно
    0.89
    POSITIVE LOGITS
     hundred
    1.77
     liners
    1.63
    надцати
    1.60
     sided
    1.59
     heck
    1.57
    ToMany
    1.51
    hundred
    1.50
     lakh
    1.48
    жды
    1.45
    씩
    1.45
    Act Density 0.193%

    No Known Activations