INDEX
Explanations
dialogue exchanges and emotional interactions among characters
New Auto-Interp
Negative Logits
olini
-0.16
.tom
-0.15
977
-0.15
821
-0.14
æĺİ
-0.14
isia
-0.14
Äĥm
-0.14
.uc
-0.13
ono
-0.13
gota
-0.13
POSITIVE LOGITS
acher
0.15
å³
0.15
addCriterion
0.15
oice
0.14
ÂŃi
0.14
theid
0.13
MOOTH
0.13
Dst
0.13
indre
0.13
reply
0.13
Activations Density 0.774%