INDEX
Explanations
names of influential female figures in comedy
names and titles
New Auto-Interp
Negative Logits
efforts
-0.41
捷
-0.40
ải
-0.37
ster
-0.37
HAY
-0.37
effort
-0.36
\]
-0.36
good
-0.36
ORD
-0.35
RAP
-0.35
POSITIVE LOGITS
Tina
2.11
Tina
1.98
tina
1.45
tina
1.05
ainfi
0.84
Tin
0.83
Tin
0.83
présidenti
0.74
inoxydable
0.74
Tint
0.72
Activations Density 0.002%