INDEX
Negative Logits
Important
-0.08
(-
-0.07
dauer
-0.07
(...)
-0.07
долго
-0.07
argu
-0.07
ياتي
-0.07
missing
-0.07
longitudinal
-0.07
-present
-0.07
POSITIVE LOGITS
comedian
0.09
fragrant
0.09
gần
0.08
浅
0.08
chuyển
0.08
Sai
0.08
vocalist
0.08
guitarist
0.08
không
0.08
golden
0.08
Activations Density 0.003%