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

    Neuronpedia

    INDEX
    Explanations

    value transformation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ীয়
    0.43
    меется
    0.43
    igenous
    0.41
    isnull
    0.40
    itele
    0.39
    ğer
    0.38
    тере
    0.38
    ocyanate
    0.38
    気に
    0.38
    卒業
    0.38
    POSITIVE LOGITS
     saves
    0.45
    dash
    0.44
     ഓരോ
    0.43
     savings
    0.42
     small
    0.42
     series
    0.40
     ташки
    0.40
     तीन
    0.38
     strategies
    0.38
     three
    0.38
    Act Density 0.000%

    No Known Activations