INDEX
Explanations
references to educational curricula
New Auto-Interp
Negative Logits
yor
-0.19
iona
-0.17
y
-0.17
iya
-0.16
yan
-0.15
yang
-0.15
i
-0.15
orado
-0.14
GPL
-0.14
idon
-0.14
POSITIVE LOGITS
abus
0.39
abi
0.31
syll
0.23
ables
0.22
abic
0.21
ab
0.19
abwe
0.18
ABI
0.18
лаб
0.18
bject
0.18
Activations Density 0.005%