INDEX
Explanations
words related to heritage and family lineage
New Auto-Interp
Negative Logits
303
-0.14
rani
-0.14
ildren
-0.13
icont
-0.13
326
-0.13
onom
-0.13
nem
-0.13
withStyles
-0.13
idata
-0.13
leston
-0.13
POSITIVE LOGITS
from
0.16
from
0.14
Âłmiles
0.14
andes
0.14
gift
0.14
PIX
0.13
ump
0.13
PI
0.13
from
0.13
unspecified
0.13
Activations Density 0.078%