INDEX
Explanations
various forms of punctuation and emoticons commonly used to express emotions in text
New Auto-Interp
Negative Logits
$_"
-1.25
myſelf
-1.24
―――――
-1.17
windowFixed
-1.12
Autoritní
-1.12
ſelves
-1.09
contentLoaded
-1.09
Савезне
-1.08
^(@)
-1.08
ंदीखरीदारी
-1.07
POSITIVE LOGITS
↵↵
1.04
↵
0.93
↵↵↵
0.81
<eos>
0.81
0.80
...
0.75
(
0.69
↵↵↵↵
0.69
…
0.68
And
0.65
Activations Density 0.029%