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

    Neuronpedia

    INDEX
    Explanations

    format specifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ly
    0.50
    s
    0.49
    ological
    0.46
    .
    0.45
     complemented
    0.45
    ore
    0.44
    ist
    0.43
    س
    0.43
     current
    0.43
    aren
    0.41
    POSITIVE LOGITS
     Lulu
    0.58
    стаў
    0.57
    ўні
    0.57
     Владимира
    0.53
    𝟎
    0.52
     forêts
    0.50
    вовано
    0.50
     Пі
    0.50
     അറ
    0.50
    чнай
    0.49
    Act Density 0.006%

    No Known Activations