INDEX
Explanations
references to a specific saint or religious figure
New Auto-Interp
Negative Logits
usk
-0.17
Haram
-0.16
pire
-0.15
grim
-0.14
kontakte
-0.14
ÅĤÄħ
-0.14
pent
-0.13
éľĩ
-0.13
ibus
-0.13
omers
-0.13
POSITIVE LOGITS
/rss
0.16
TRS
0.15
hani
0.15
eki
0.15
aml
0.14
ults
0.14
FromBody
0.14
inspace
0.14
ForObject
0.14
owan
0.14
Activations Density 0.009%