INDEX
Explanations
proper nouns, specifically names and places
New Auto-Interp
Negative Logits
vier
-0.17
folded
-0.15
Pooling
-0.15
elfast
-0.15
rå
-0.14
Fold
-0.14
AppleWebKit
-0.14
ιο
-0.14
oken
-0.13
lom
-0.13
POSITIVE LOGITS
ols
0.15
imon
0.15
/**č↵
0.14
åĤ
0.14
gim
0.14
ason
0.14
å·¥
0.13
imde
0.13
essian
0.13
.bad
0.13
Activations Density 0.061%