INDEX
Explanations
phrases indicating possession or obligation
New Auto-Interp
Negative Logits
Efq
-1.16
InputBorder
-1.14
sidemargin
-1.13
myſelf
-1.12
ainfi
-1.12
purpoſe
-1.11
pleaſure
-1.11
ſelf
-1.11
auffi
-1.11
doubtnut
-1.10
POSITIVE LOGITS
<eos>
0.65
'
0.60
↵
0.60
''
0.59
’
0.57
Having
0.57
I
0.57
<i>
0.57
I
0.57
w
0.56
Activations Density 0.270%