INDEX
Explanations
phrases related to comparisons and contrasts
New Auto-Interp
Negative Logits
addon
-0.15
célib
-0.14
å´İ
-0.14
ä½ľä¸º
-0.14
acco
-0.14
Orm
-0.14
ale
-0.14
ãģ¨ãģĹãģ¦
-0.14
izzo
-0.14
eps
-0.14
POSITIVE LOGITS
having
0.32
being
0.24
having
0.24
Having
0.23
Having
0.23
ayant
0.19
being
0.18
sendo
0.18
follows
0.17
rades
0.17
Activations Density 0.102%