INDEX
Explanations
events involving conflict, resolution, and transitions in a narrative context
New Auto-Interp
Negative Logits
yourselves
-0.15
bert
-0.14
ond
-0.13
atak
-0.13
roller
-0.13
asy
-0.13
dew
-0.13
umni
-0.13
Fold
-0.13
Ren
-0.13
POSITIVE LOGITS
-bed
0.15
inesis
0.15
Ãľl
0.14
bedtime
0.14
redicate
0.14
iffies
0.14
475
0.13
ieee
0.13
ãĥ¼ãĥ©
0.13
rale
0.13
Activations Density 0.399%