INDEX
Explanations
phrases expressing the consequences of conflict and emotional themes in storytelling
New Auto-Interp
Negative Logits
zwar
-0.14
_READONLY
-0.14
žen
-0.14
ICLE
-0.14
-sort
-0.13
yna
-0.13
orthy
-0.13
WindowText
-0.13
obili
-0.13
icle
-0.13
POSITIVE LOGITS
being
0.21
each
0.20
each
0.18
being
0.18
today
0.16
herent
0.16
æ¯ı
0.15
circumstance
0.15
our
0.15
yesterday
0.14
Activations Density 0.239%