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

    Neuronpedia

    INDEX
    Explanations

    variable r

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ف
    2.31
    ください
    2.30
    ური
    2.20
    ج
    2.19
    mout
    2.14
    m
    2.13
    ت
    2.13
    milk
    2.08
    ו
    2.08
    yog
    2.05
    POSITIVE LOGITS
    ра
    2.23
    械
    2.03
    r
    2.02
    ments
    1.88
    IZATION
    1.88
    les
    1.80
     suspects
    1.79
    ेश
    1.78
    ATING
    1.77
    으로써
    1.77
    Act Density 0.188%

    No Known Activations