INDEX
Explanations
words related to Welsh culture and events
New Auto-Interp
Negative Logits
å·±
-0.17
uels
-0.16
åĪ
-0.16
eye
-0.15
LOPT
-0.15
Iz
-0.14
Highlander
-0.14
Ñĩе
-0.14
mote
-0.14
ë§¥
-0.14
POSITIVE LOGITS
wr
0.23
yr
0.22
'r
0.22
yn
0.22
wy
0.21
ar
0.21
dd
0.21
-dd
0.20
dd
0.20
y
0.20
Activations Density 0.005%