INDEX
Explanations
mentions of eyes being closed or the act of sleeping
New Auto-Interp
Negative Logits
ذ
-0.19
egl
-0.17
isoft
-0.15
trx
-0.15
ardless
-0.15
olate
-0.14
جر
-0.14
Lonely
-0.14
arily
-0.14
isman
-0.14
POSITIVE LOGITS
deo
0.17
raki
0.17
lew
0.16
bourg
0.16
elib
0.15
yro
0.15
abies
0.15
conform
0.14
/mock
0.14
лев
0.14
Activations Density 0.123%