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

    Neuronpedia

    INDEX
    Explanations

    multilingual words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     है
    1.00
    ска
    0.96
    0.96
     ép
    0.93
     emp
    0.93
    व्र
    0.93
     пи
    0.89
     آ
    0.89
    طة
    0.88
    คือ
    0.88
    POSITIVE LOGITS
    க்ஸாண்ட
    1.23
     लकार
    1.23
     които
    1.11
     religions
    1.10
     allerlei
    1.09
     ያሉ
    1.09
     memberships
    1.08
    okhlov
    1.02
    ന്യാ
    1.02
     sexuales
    1.02
    Act Density 1.454%

    No Known Activations