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

    Neuronpedia

    INDEX
    Explanations

    circumstances

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    াড়
    1.63
     있으며
    1.59
    样的
    1.55
    ية
    1.51
     Còn
    1.49
    ির
    1.48
     الزمن
    1.48
    և
    1.46
     sánh
    1.45
    ר
    1.45
    POSITIVE LOGITS
    ne
    2.38
    ました
    2.22
    َ
    2.22
    llä
    2.14
    g
    2.05
    س
    2.02
    ly
    1.99
    d
    1.99
    th
    1.95
    duced
    1.93
    Act Density 0.012%

    No Known Activations