INDEX
Explanations
instances of the word "buy" and its variations
New Auto-Interp
Negative Logits
auen
-0.15
thy
-0.15
pol
-0.15
RL
-0.14
ÑģÑĭлки
-0.14
tras
-0.14
ye
-0.14
dehy
-0.14
har
-0.14
had
-0.14
POSITIVE LOGITS
enance
0.17
/install
0.17
ngo
0.15
ipi
0.15
isto
0.15
boys
0.15
Cres
0.14
/download
0.14
zilla
0.14
anz
0.14
Activations Density 0.031%