INDEX
Explanations
the word "minute"
New Auto-Interp
Negative Logits
-0.93
I
-0.78
a
-0.75
the
-0.72
$
-0.71
<eos>
-0.69
is
-0.67
giphy
-0.67
.
-0.67
several
-0.66
POSITIVE LOGITS
Efq
0.96
Shakspeare
0.94
Shaksp
0.91
<bos>
0.88
myſelf
0.81
་་
0.81
moistur
0.81
CascadeType
0.79
armoured
0.79
PyTuple
0.79
Activations Density 1.016%