INDEX
    Explanations

    descriptive features of home interiors and materials

    New Auto-Interp
    Negative Logits
    eren
    -0.15
    erb
    -0.15
    üs
    -0.14
    extr
    -0.14
    haar
    -0.14
    jo
    -0.14
     mort
    -0.14
    ennes
    -0.13
     Er
    -0.13
    auer
    -0.13
    POSITIVE LOGITS
    ê¸ī
    0.17
    ihn
    0.16
    IZE
    0.14
    izza
    0.14
    ocity
    0.14
    гов
    0.14
    inspace
    0.14
    SError
    0.14
    ogui
    0.14
    acias
    0.14
    Act Density 0.020%

    No Known Activations