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

    Neuronpedia

    INDEX
    Explanations

    PDF.AddPage(Page)

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     potentially
    0.60
    视
    0.59
     nullable
    0.58
     birth
    0.54
     OUR
    0.54
    Ĭ
    0.54
     interpreted
    0.54
     \%)$.
    0.54
     ▪
    0.53
    ீடு
    0.52
    POSITIVE LOGITS
     Gleich
    0.64
     Luego
    0.60
    ಓ
    0.59
    addButton
    0.59
     Nachdem
    0.58
    了嗎
    0.57
     puis
    0.56
     hossz
    0.55
    這樣的
    0.55
     closeModal
    0.54
    Act Density 0.394%

    No Known Activations