INDEX
Explanations
monetary values and numerical data
New Auto-Interp
Negative Logits
/Instruction
-0.15
flush
-0.15
gor
-0.15
Įĵ
-0.14
Flush
-0.14
onas
-0.14
holes
-0.14
/el
-0.13
øj
-0.13
_THREAD
-0.13
POSITIVE LOGITS
<!--[
0.16
uos
0.16
585
0.15
RICT
0.15
bolt
0.15
stake
0.15
019
0.15
emek
0.14
341
0.14
jee
0.14
Activations Density 0.075%