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

    Neuronpedia

    INDEX
    Explanations

    mass and massive scale

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    the
    2.05
    thet
    1.12
    ties
    1.08
    on
    1.02
    tis
    1.02
    tas
    1.02
    jší
    1.00
    সংখ্যক
    0.96
    then
    0.94
    x
    0.93
    POSITIVE LOGITS
    ↵↵
    1.71
    ↵
    1.44
     as
    1.33
    ล
    1.30
    ీ
    1.20
    ס
    1.19
    س
    1.16
    是
    1.12
    กัน
    1.11
     który
    1.02
    Act Density 0.125%

    No Known Activations