INDEX
Explanations
entities and roles related to individuals in narratives
New Auto-Interp
Negative Logits
aeda
-0.19
afil
-0.17
aldi
-0.16
ounter
-0.15
ละ
-0.15
uku
-0.15
ouve
-0.15
strup
-0.15
ÑĥÑĤи
-0.14
anded
-0.14
POSITIVE LOGITS
whom
0.17
cri
0.16
frequent
0.15
æĭħå½ĵ
0.15
)
0.15
.k
0.15
int
0.14
Nov
0.14
ovan
0.14
inde
0.14
Activations Density 0.257%