INDEX
Explanations
medical advice related to mask-wearing and preventing virus transmission
New Auto-Interp
Negative Logits
ibur
-0.17
ibia
-0.15
Morrison
-0.15
dress
-0.15
aksi
-0.15
arkin
-0.15
Report
-0.15
_sink
-0.14
óc
-0.14
oman
-0.14
POSITIVE LOGITS
irket
0.18
ersen
0.14
èĹ
0.14
vette
0.14
pra
0.14
erdale
0.13
ascus
0.13
пÑĭ
0.13
HIR
0.13
errat
0.13
Activations Density 0.311%