INDEX
Explanations
references to spatial location and positional dynamics
New Auto-Interp
Negative Logits
-has
-0.16
ÙĨدارد
-0.14
iÄħ
-0.14
anka
-0.14
каÑģ
-0.13
surrounds
-0.13
HAS
-0.13
yster
-0.13
ante
-0.13
ScrollBar
-0.13
POSITIVE LOGITS
are
0.38
there
0.30
lies
0.28
çļĦæĺ¯
0.27
is
0.25
lie
0.25
estão
0.24
were
0.23
sits
0.23
we
0.23
Activations Density 0.177%