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

    Neuronpedia

    INDEX
    Explanations

    breakdown for localization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    鸳
    0.38
    部位
    0.37
     Kear
    0.35
     Leila
    0.35
    不可以
    0.35
     Forester
    0.34
     Adele
    0.34
    Ⲇ
    0.34
    Ю
    0.34
    痂
    0.34
    POSITIVE LOGITS
     silhouettes
    0.40
     factory
    0.39
     summ
    0.38
    factory
    0.38
    equalTo
    0.38
     Responsibility
    0.38
     مسئول
    0.37
    收获
    0.37
     learned
    0.36
    روض
    0.36
    Act Density 0.001%

    No Known Activations