INDEX
Explanations
instances of the letter "i"
New Auto-Interp
Negative Logits
himſelf
-0.83
purpoſe
-0.80
entanto
-0.78
themſelves
-0.75
Chriftian
-0.75
Shakspeare
-0.74
houſe
-0.73
myſelf
-0.73
ſame
-0.72
canin
-0.71
POSITIVE LOGITS
và
1.08
и
1.04
και
1.02
InjectAttribute
0.98
\&
0.96
And
0.93
\&
0.90
力和
0.88
そして
0.87
och
0.85
Activations Density 0.020%