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

    Neuronpedia

    INDEX
    Explanations

    in

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    .Element
    -0.07
     dependable
    -0.06
    何
    -0.06
    ику
    -0.06
    Seat
    -0.06
    UGIN
    -0.06
    wh
    -0.06
    usi
    -0.06
    .less
    -0.06
     QCOMPARE
    -0.06
    POSITIVE LOGITS
     disrupted
    0.07
    ประม
    0.06
     protested
    0.06
     دام
    0.06
     masculinity
    0.06
     @@↵
    0.06
    ดำ
    0.06
     stratej
    0.06
    }*/↵
    0.06
    。あ
    0.06
    Act Density 0.033%

    No Known Activations