INDEX
Explanations
references to windows and related concepts
New Auto-Interp
Negative Logits
composite
-0.18
ideographic
-0.17
krv
-0.16
intree
-0.15
aign
-0.15
anten
-0.15
urer
-0.15
Composite
-0.15
ayet
-0.15
Composite
-0.14
POSITIVE LOGITS
entlich
0.16
ss
0.16
olk
0.15
.mj
0.15
ALLED
0.15
mey
0.14
pane
0.14
lag
0.14
om
0.14
0.14
Activations Density 0.014%