INDEX
Explanations
elements related to comedy and drama in narratives
New Auto-Interp
Negative Logits
odd
-0.17
pure
-0.14
etal
-0.14
Mobil
-0.14
#!
-0.14
Ni
-0.14
åĺĽ
-0.14
Cast
-0.14
gou
-0.14
dou
-0.14
POSITIVE LOGITS
ɵ
0.18
pyx
0.16
idos
0.16
oppers
0.15
-dom
0.15
oley
0.15
TEGER
0.15
motion
0.15
rede
0.14
Ú¯ÙĨ
0.14
Activations Density 0.173%