INDEX
Explanations
references to cocktails and mixology
New Auto-Interp
Negative Logits
erli
-0.15
iece
-0.15
997
-0.14
Canter
-0.14
ittel
-0.14
lacak
-0.14
_CN
-0.14
бед
-0.14
bjerg
-0.14
éłĤ
-0.14
POSITIVE LOGITS
_AUD
0.17
conc
0.16
,buf
0.15
_FUN
0.15
iline
0.14
dens
0.14
eyes
0.14
erts
0.14
еÑĢÑĤа
0.14
ewise
0.13
Activations Density 0.029%