INDEX
Explanations
questions or statements that express uncertainty about locations or origins
New Auto-Interp
Negative Logits
them
-0.20
ãģªãĤĵãģ¦
-0.15
yla
-0.15
eux
-0.15
sure
-0.15
scan
-0.14
redo
-0.14
same
-0.14
nya
-0.14
ëŀĢ
-0.14
POSITIVE LOGITS
else
0.42
exactly
0.42
/how
0.39
abouts
0.33
fore
0.29
precisely
0.29
/if
0.27
Exactly
0.27
they
0.26
ver
0.25
Activations Density 0.035%