INDEX
Explanations
occurrences of the term "normal" and its variations in context
New Auto-Interp
Negative Logits
eling
-0.17
essel
-0.16
ile
-0.15
ILE
-0.15
eb
-0.14
igure
-0.14
uncture
-0.14
awner
-0.14
anical
-0.14
light
-0.14
POSITIVE LOGITS
cy
0.22
ity
0.21
mente
0.21
-normal
0.20
ities
0.18
ously
0.16
afen
0.16
ITY
0.15
idad
0.15
abwe
0.15
Activations Density 0.033%