INDEX
Explanations
mentions of Russia or things happening in Russia
New Auto-Interp
Negative Logits
^(@)
-0.88
cember
-0.84
/>\
-0.80
oporosis
-0.77
abetes
-0.74
дописавши
-0.73
$_"
-0.73
'\\;'
-0.72
inghouse
-0.70
>*/
-0.69
POSITIVE LOGITS
<bos>
0.85
poichè
0.69
vägen
0.66
pareti
0.60
estudos
0.60
"
0.60
ulei
0.60
preuves
0.59
relatifs
0.59
es
0.59
Activations Density 1.507%