INDEX
Explanations
mentions of tears and expressions of emotional pain
tears and destruction
New Auto-Interp
Negative Logits
bibfield
-0.53
barbati
-0.49
Mifflin
-0.44
orgánico
-0.44
Middle
-0.44
Indonesian
-0.44
тную
-0.43
Northam
-0.41
ympä
-0.41
simov
-0.41
POSITIVE LOGITS
tears
1.83
Tears
1.70
Tears
1.70
tears
1.63
lágrimas
1.05
larmes
1.01
wept
0.89
涙
0.86
泪
0.71
menangis
0.70
Activations Density 0.006%