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

    Neuronpedia

    INDEX
    Explanations

    when clauses with dates

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     zonder
    0.50
     consigui
    0.47
     ohne
    0.45
    without
    0.43
    ्हे
    0.42
     senza
    0.42
     معنات
    0.40
    इसका
    0.40
     memulai
    0.39
    これが
    0.39
    POSITIVE LOGITS
     when
    0.83
    随着
    0.74
     когда
    0.73
     cuando
    0.70
     lorsque
    0.70
     ketika
    0.69
    ในการ
    0.69
     khi
    0.69
     quando
    0.68
     lorsqu
    0.68
    Act Density 0.010%

    No Known Activations