INDEX
    Explanations

    mentions of familial and community structures

    belonging to each other

    New Auto-Interp
    Negative Logits
    GraphicsUnit
    -0.41
    __":
    
    -0.40
    __':
    
    -0.40
    OGND
    -0.40
    escalier
    -0.39
    󠁢
    -0.39
     meisje
    -0.38
    middels
    -0.38
     ممن
    -0.37
     käytet
    -0.36
    POSITIVE LOGITS
     CreateTagHelper
    0.52
     Infórmanos
    0.50
     kasarigan
    0.49
     Signalez
    0.47
    wiliwch
    0.45
    makeText
    0.44
    Alike
    0.44
    Giorgio
    0.43
     nakalista
    0.43
    styleType
    0.43
    Act Density 0.276%

    No Known Activations