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

    Neuronpedia

    INDEX
    Explanations

    heading tags and markdown structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    {
    1.35
     on
    1.23
    ти
    1.05
    #,
    0.99
    ༨
    0.90
    \,
    0.89
    هِ
    0.87
    {*
    0.87
    {:
    0.86
    성과
    0.86
    POSITIVE LOGITS
    在
    1.59
    m
    1.55
     in
    1.45
    1
    1.39
    も
    1.29
     في
    1.18
    도
    1.11
    ใน
    1.10
    ও
    1.10
    aj
    1.09
    Act Density 0.001%

    No Known Activations