INDEX
Explanations
references to street names or avenues
New Auto-Interp
Negative Logits
Houſe
-1.21
pleaſure
-1.17
itſelf
-1.15
Monfieur
-1.14
houſe
-1.11
myſelf
-1.08
Theſe
-1.08
greateſt
-1.07
ſelves
-1.07
Reſ
-1.05
POSITIVE LOGITS
Ave
0.71
ave
0.69
,
0.60
Led
0.58
ave
0.57
Sa
0.57
W
0.56
нки
0.56
Col
0.55
forma
0.54
Activations Density 0.082%