INDEX
Explanations
measurements of time and counts
New Auto-Interp
Negative Logits
rike
-0.17
aptive
-0.16
endum
-0.16
841
-0.15
º
-0.14
883
-0.14
legg
-0.14
urent
-0.14
158
-0.14
łģ
-0.14
POSITIVE LOGITS
-and
0.54
½
0.30
_and
0.27
_AND
0.26
And
0.26
åįĬ
0.25
And
0.24
_And
0.22
anda
0.20
AND
0.20
Activations Density 0.097%