INDEX
Explanations
occurrences of the word "draft" and its variations
New Auto-Interp
Negative Logits
/he
-0.16
ÑĶм
-0.16
rael
-0.15
idade
-0.15
nd
-0.15
ial
-0.15
rvine
-0.15
ities
-0.14
ous
-0.14
igel
-0.14
POSITIVE LOGITS
ivism
0.25
sm
0.23
iness
0.23
ney
0.20
ers
0.18
y
0.18
eds
0.18
able
0.17
aroo
0.17
égor
0.17
Activations Density 0.013%