INDEX
Explanations
characters and their relationships within the narrative
New Auto-Interp
Negative Logits
знаÑĩа
-0.15
kovi
-0.15
esz
-0.15
iral
-0.14
ui
-0.14
eco
-0.14
ccione
-0.14
sher
-0.14
agan
-0.13
TF
-0.13
POSITIVE LOGITS
novel
0.20
Mr
0.19
narrator
0.18
Becky
0.16
doktor
0.16
Novel
0.16
-enable
0.15
Mr
0.15
neur
0.15
Crunch
0.14
Activations Density 0.166%