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

    Neuronpedia

    INDEX
    Explanations

    can warn about problems

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    чами
    1.22
     scanf
    1.20
     ứng
    1.17
    郎
    1.10
    لط
    1.05
    ἦ
    1.04
    horn
    1.04
    uje
    1.00
     realizacji
    0.99
    ceğiz
    0.98
    POSITIVE LOGITS
    ्स
    1.33
    こちら
    1.22
    椋
    1.15
     Siena
    1.12
     thoughtful
    1.12
     мене
    1.11
    ات
    1.11
    রাও
    1.11
    ଈ
    1.09
     уг
    1.08
    Act Density 0.001%

    No Known Activations