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

    Neuronpedia

    INDEX
    Explanations

    conjunctions and elaborations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ：
    0.46
     appartenant
    0.39
     loại
    0.37
    的时候
    0.36
     của
    0.36
    这一次
    0.36
     tejto
    0.35
    ،
    0.35
    堣
    0.34
    بيع
    0.33
    POSITIVE LOGITS
     மேலும்
    0.37
    并通过
    0.37
    represents
    0.36
     it
    0.36
     hence
    0.35
     எனவே
    0.35
     zudem
    0.35
    और
    0.35
    is
    0.34
    thus
    0.34
    Act Density 0.004%

    No Known Activations