INDEX
Explanations
references to the word "pom."
New Auto-Interp
Negative Logits
.bt
-0.15
:async
-0.15
{@-0.15
mite
-0.14
strict
-0.14
yte
-0.14
ourcem
-0.14
honest
-0.14
yne
-0.14
ùng
-0.14
POSITIVE LOGITS
Pom
0.19
3
0.17
pos
0.16
oldt
0.16
Magn
0.16
etime
0.16
532
0.16
antz
0.16
acent
0.15
pom
0.15
Activations Density 0.008%