INDEX
Explanations
quantifiers indicating comparisons or amounts
New Auto-Interp
Negative Logits
\Collections
-0.18
ýš
-0.17
coni
-0.15
нен
-0.15
istar
-0.15
illes
-0.14
.experimental
-0.14
assis
-0.14
mess
-0.14
criptor
-0.14
POSITIVE LOGITS
anie
0.17
time
0.16
sayıda
0.16
room
0.16
money
0.15
ulin
0.15
ulet
0.15
اÙĦÙĪÙĤت
0.15
atro
0.15
of
0.14
Activations Density 0.117%