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

    Neuronpedia

    INDEX
    Explanations

    housing structures and dwellings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     reformulation
    0.53
     reformulated
    0.53
     outspoken
    0.52
     Elektrokh
    0.51
     paise
    0.48
     dysph
    0.48
     Stimme
    0.47
     disillusioned
    0.47
     Elektrokhimiya
    0.45
     Valores
    0.45
    POSITIVE LOGITS
    建築
    1.17
     thatched
    1.12
     건축
    1.11
    建筑
    1.08
    🛖
    1.05
    房屋
    1.03
     buildings
    1.00
     architectural
    1.00
    小屋
    0.99
     building
    0.98
    Act Density 0.055%

    No Known Activations