INDEX
Explanations
prepositions and subordinators indicating time and place
New Auto-Interp
Negative Logits
ories
-0.15
forth
-0.15
ife
-0.15
SCAN
-0.15
ires
-0.15
hart
-0.15
oref
-0.14
Buch
-0.14
uki
-0.14
koc
-0.13
POSITIVE LOGITS
mis
0.16
peech
0.15
period
0.15
bids
0.15
peats
0.15
parallel
0.14
sut
0.14
çŁ¢
0.14
ληÏĤ
0.14
657
0.14
Activations Density 0.078%