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

    Neuronpedia

    INDEX
    Explanations

    list items with quantities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ون
    0.56
    न
    0.53
    G
    0.50
    在
    0.47
    ન
    0.45
    K
    0.43
     in
    0.42
    The
    0.42
    S
    0.41
    υ
    0.41
    POSITIVE LOGITS
     is
    0.41
    ،
    0.40
     are
    0.38
    ется
    0.38
    como
    0.36
     tripped
    0.35
     Перейти
    0.34
     как
    0.34
     (\<
    0.34
     toasted
    0.33
    Act Density 2.323%

    No Known Activations