INDEX
Explanations
the definite article "the" across various contexts
New Auto-Interp
Negative Logits
iki
-0.15
mpp
-0.15
ĵ°
-0.15
vale
-0.15
&_
-0.14
ingen
-0.14
315
-0.14
æŀľ
-0.14
ungen
-0.13
afil
-0.13
POSITIVE LOGITS
sake
0.66
purposes
0.57
benefit
0.40
purpose
0.38
Benefit
0.30
purpose
0.28
-purpose
0.25
Purpose
0.25
duration
0.24
urpose
0.23
Activations Density 0.114%