INDEX
Explanations
questions beginning with "why."
New Auto-Interp
Negative Logits
lei
-0.18
erland
-0.17
AYER
-0.15
hoff
-0.15
arna
-0.15
lech
-0.14
msec
-0.14
ivre
-0.14
/dc
-0.14
quette
-0.14
POSITIVE LOGITS
suddenly
0.23
bother
0.23
à¤ĩतन
0.20
Suddenly
0.18
bothering
0.18
why
0.18
so
0.17
sudden
0.17
why
0.17
bothered
0.17
Activations Density 0.101%