INDEX
Explanations
plural nouns, particularly those that end in "ets" or similar sounds
New Auto-Interp
Negative Logits
ettings
-0.08
sworth
-0.07
ov
-0.07
etsk
-0.07
sg
-0.07
trs
-0.07
stown
-0.07
sson
-0.07
eturn
-0.07
etect
-0.07
POSITIVE LOGITS
uki
0.09
ãĤ¡
0.09
/stdc
0.08
gerald
0.08
ãĤ§
0.08
unami
0.07
chrift
0.07
ubishi
0.07
ch
0.07
ing
0.07
Activations Density 0.015%