INDEX
Explanations
instances of the verb "to be" in various conjugated forms
New Auto-Interp
Negative Logits
Ĥ
-0.15
ombo
-0.14
aze
-0.14
uba
-0.14
Bath
-0.14
inds
-0.14
xe
-0.14
Barth
-0.14
lane
-0.14
imony
-0.14
POSITIVE LOGITS
uron
0.16
nex
0.15
0.14
Bundle
0.14
rün
0.14
éĻ
0.14
ocker
0.14
اÙĤ
0.13
ekil
0.13
Hammer
0.13
Activations Density 0.003%