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

    Neuronpedia

    INDEX
    Explanations

    use

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     reportedly
    -0.07
     ему
    -0.06
    zy
    -0.06
     želez
    -0.06
     Tb
    -0.06
    (MigrationBuilder
    -0.06
    iterals
    -0.06
    表
    -0.06
     кот
    -0.06
     crates
    -0.06
    POSITIVE LOGITS
     teg
    0.07
     تغ
    0.06
     Brilliant
    0.06
     З
    0.06
     نمی
    0.06
     TreeSet
    0.06
     ग
    0.06
     Friends
    0.06
     Val
    0.06
    film
    0.06
    Act Density 0.008%

    No Known Activations