INDEX
Explanations
phrases related to obligation or necessity
New Auto-Interp
Negative Logits
rok
-0.18
vrier
-0.16
gaard
-0.16
doing
-0.15
atis
-0.15
اÙĦØ¥ÙĨجÙĦÙĬزÙĬØ©
-0.15
pone
-0.15
ud
-0.15
l
-0.15
Doing
-0.15
POSITIVE LOGITS
.direct
0.18
directly
0.17
with
0.17
irect
0.17
diret
0.16
-With
0.16
DIRECT
0.16
¶Į
0.15
about
0.15
994
0.15
Activations Density 0.009%