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

    Neuronpedia

    INDEX
    Explanations

    normal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    具体的
    0.44
     specifically
    0.38
     אותו
    0.37
     μπορεί
    0.35
    ลอด
    0.35
     lacked
    0.34
     bleibt
    0.34
     বিরাট
    0.34
    特定的
    0.34
     изменения
    0.34
    POSITIVE LOGITS
     ordinary
    0.98
     normal
    0.95
     normale
    0.93
    ordinary
    0.90
     обы
    0.88
     സാധാരണ
    0.86
     சாதாரண
    0.85
     normalen
    0.83
     обычной
    0.82
    normal
    0.82
    Act Density 0.045%

    No Known Activations