INDEX
Explanations
words related to variety and intensity
New Auto-Interp
Negative Logits
eki
-0.21
tember
-0.21
ey
-0.20
tober
-0.20
tele
-0.19
ted
-0.19
to
-0.19
tech
-0.19
tem
-0.18
tv
-0.18
POSITIVE LOGITS
ive
0.32
ions
0.31
iveness
0.28
ión
0.27
ely
0.26
cope
0.26
ively
0.26
es
0.26
ory
0.25
ible
0.25
Activations Density 0.064%