INDEX
Explanations
relationships and interactions among characters in a narrative
New Auto-Interp
Negative Logits
DT
-0.18
_firestore
-0.16
allis
-0.15
iferay
-0.15
ecess
-0.14
égor
-0.14
ertino
-0.14
577
-0.14
Tyler
-0.13
ñana
-0.13
POSITIVE LOGITS
cold
0.22
expression
0.18
pale
0.17
Cold
0.17
cold
0.17
ind
0.16
Cold
0.16
Expression
0.15
speech
0.15
turb
0.14
Activations Density 0.015%