INDEX
Explanations
references to the definite article "the."
New Auto-Interp
Negative Logits
itſelf
-1.11
-1.07
myſelf
-1.06
Phry
-1.04
raiſ
-1.04
doubtnut
-1.03
Houſe
-1.01
Mahomet
-1.01
pleaſure
-1.00
Huguen
-0.99
POSITIVE LOGITS
the
1.93
The
1.47
The
1.33
THE
1.31
same
1.20
enthe
1.06
the
1.06
rethe
1.05
ethe
1.01
entire
0.99
Activations Density 2.966%