INDEX
Explanations
emotional expressions and relationships in narratives
New Auto-Interp
Negative Logits
IFO
-0.17
æį·
-0.15
.gdx
-0.14
oblin
-0.14
ongyang
-0.14
RAY
-0.14
bn
-0.14
indsay
-0.14
许
-0.14
Wenger
-0.14
POSITIVE LOGITS
Bast
0.16
strtol
0.15
Ellis
0.14
thi
0.14
401
0.14
col
0.13
_den
0.13
Denver
0.13
brothers
0.13
Denver
0.13
Activations Density 0.207%