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

    Neuronpedia

    INDEX
    Explanations

    model response completion

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ሁሉም
    1.01
    кну
    0.98
    اء
    0.97
     muda
    0.96
    île
    0.95
    ὁ
    0.93
     Félix
    0.93
     and
    0.93
     κατά
    0.93
    ㍍
    0.93
    POSITIVE LOGITS
    ية
    1.27
    ian
    1.25
    ,
    1.20
    ли
    1.05
    ین
    1.05
    ർ
    1.04
    ال
    1.03
    ，「
    0.98
    us
    0.98
    ia
    0.98
    Act Density 0.477%

    No Known Activations