INDEX
Explanations
instances of the word "second" and its variations
New Auto-Interp
Negative Logits
bara
-0.16
una
-0.15
lic
-0.14
egin
-0.14
оÑĩной
-0.14
uent
-0.14
lica
-0.14
Mahon
-0.14
дал
-0.13
onse
-0.13
POSITIVE LOGITS
arily
0.29
nd
0.24
/th
0.22
aries
0.22
عشر
0.19
-largest
0.19
-tier
0.18
-generation
0.18
-feira
0.17
reno
0.17
Activations Density 0.045%