INDEX
Explanations
punctuation and grammatical elements in the text
New Auto-Interp
Negative Logits
Conv
-0.16
иб
-0.16
Conv
-0.16
èĢ
-0.15
idor
-0.15
leyen
-0.15
uvo
-0.15
inan
-0.15
mse
-0.14
çĩ
-0.14
POSITIVE LOGITS
imoto
0.16
apol
0.16
ieren
0.16
strictly
0.15
oses
0.15
ognito
0.15
089
0.14
anta
0.14
Strict
0.14
ition
0.14
Activations Density 0.002%