INDEX
Explanations
punctuation marks and their context in sentences
New Auto-Interp
Negative Logits
.dylib
-0.15
má
-0.14
atest
-0.14
ifton
-0.14
olk
-0.14
atas
-0.14
"class
-0.14
astery
-0.14
reco
-0.13
Crud
-0.13
POSITIVE LOGITS
rush
0.15
šil
0.14
rub
0.14
nels
0.14
šak
0.14
ontology
0.14
Barb
0.14
hlen
0.14
onto
0.14
ernal
0.14
Activations Density 0.070%