INDEX
Explanations
words related to expansion or growth
New Auto-Interp
Negative Logits
pagen
-0.17
onn
-0.17
cott
-0.17
ialized
-0.15
kup
-0.15
rob
-0.15
ispecies
-0.15
cum
-0.15
cuts
-0.15
nap
-0.15
POSITIVE LOGITS
/exp
0.21
(exp
0.20
exp
0.20
Exp
0.19
ansion
0.16
iture
0.15
.exp
0.15
GIT
0.15
érience
0.15
unge
0.15
Activations Density 0.042%