INDEX
Explanations
references to prominent figures and concepts related to the Armenian context
New Auto-Interp
Negative Logits
hence
-0.17
omon
-0.16
åĨĨ
-0.16
aub
-0.15
Hence
-0.15
iro
-0.15
rosse
-0.15
nie
-0.15
hips
-0.14
urovision
-0.14
POSITIVE LOGITS
essian
0.21
ERSHEY
0.20
praak
0.17
locker
0.17
quarter
0.16
iday
0.16
ilities
0.16
soever
0.15
reds
0.15
sapi
0.15
Activations Density 0.580%