INDEX
Explanations
words related to promotion and marketing
New Auto-Interp
Negative Logits
icap
-0.18
liness
-0.16
lessly
-0.16
ern
-0.16
eenth
-0.15
/do
-0.15
-thirds
-0.15
ild
-0.15
nd
-0.15
zelf
-0.15
POSITIVE LOGITS
/prom
0.21
enade
0.16
adera
0.16
(prom
0.16
otional
0.15
/mark
0.15
inent
0.15
seudo
0.14
Ĭ
0.14
rax
0.14
Activations Density 0.037%