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

    Neuronpedia

    INDEX
    Explanations

    however, it's crucial to rule out

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     том
    0.39
    кологи
    0.33
    τό
    0.33
    flops
    0.33
     оказывается
    0.32
    ক্ট
    0.31
     достой
    0.31
     предназначен
    0.31
     यानी
    0.31
     gleiche
    0.31
    POSITIVE LOGITS
    년대
    0.40
     când
    0.39
     fazia
    0.39
    ัน
    0.39
    nál
    0.38
    nél
    0.38
    ्स
    0.37
     (\<
    0.37
    δου
    0.37
     aast
    0.37
    Act Density 0.363%

    No Known Activations