INDEX
Explanations
the word "capture".
New Auto-Interp
Negative Logits
]<<"
-0.71
ſeveral
-0.69
Shakspeare
-0.69
Theſe
-0.69
=?";
-0.68
YR
-0.66
―――――
-0.66
Kuru
-0.66
Duro
-0.65
Bep
-0.65
POSITIVE LOGITS
MLLoader
0.66
off
0.62
for
0.60
and
0.60
dieną
0.55
but
0.53
RenderAtEndOf
0.53
so
0.52
in
0.52
I
0.52
Activations Density 0.836%