INDEX
Explanations
emotional responses and strong feelings in narratives
New Auto-Interp
Negative Logits
orman
-0.16
.ManyToMany
-0.16
bilateral
-0.16
addock
-0.15
xOffset
-0.15
Sky
-0.15
Vak
-0.15
ĨĴ
-0.15
decorate
-0.14
959
-0.14
POSITIVE LOGITS
tinh
0.18
writ
0.15
apus
0.15
marsh
0.14
ockey
0.14
amic
0.14
ÙĤدر
0.14
b
0.13
asher
0.13
spine
0.13
Activations Density 0.209%