INDEX
Explanations
references to specific places, people, and organizations related to Welsh culture and recognition
New Auto-Interp
Negative Logits
LOPT
-0.18
bes
-0.17
lik
-0.17
ccak
-0.16
å·±
-0.15
kn
-0.15
üss
-0.15
Ye
-0.14
uels
-0.14
vere
-0.14
POSITIVE LOGITS
wr
0.25
yr
0.24
yn
0.23
wy
0.23
edd
0.22
'r
0.21
cy
0.20
ar
0.19
gw
0.19
gy
0.19
Activations Density 0.003%