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

    Neuronpedia

    INDEX
    Explanations

    island descriptions and contexts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ли
    1.08
    يت
    1.07
    લ
    1.03
    نية
    1.02
    }$
    1.02
    لي
    1.00
    لا
    0.99
    서는
    0.99
    ي
    0.98
    سي
    0.96
    POSITIVE LOGITS
     I
    1.47
     Island
    1.41
     It
    1.29
     ISLAND
    1.23
     for
    1.21
     A
    1.20
     island
    1.16
    Island
    1.16
     R
    1.14
     For
    1.13
    Act Density 0.004%

    No Known Activations