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

    Neuronpedia

    INDEX
    Explanations

    lack of speed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     sowohl
    0.33
    のではなく
    0.30
    rather
    0.30
    不错
    0.29
    不錯
    0.28
     ഓരോ
    0.28
    而非
    0.28
     rather
    0.27
     both
    0.27
     både
    0.27
    POSITIVE LOGITS
     إلا
    0.59
     siquiera
    0.57
     unless
    0.56
     anymore
    0.53
    Unless
    0.48
    任何
    0.46
    unless
    0.46
     except
    0.45
    除非
    0.44
     nor
    0.44
    Act Density 1.979%

    No Known Activations