INDEX
Explanations
emotional expressions related to humor and sarcasm
New Auto-Interp
Negative Logits
'\\;'
-0.77
ThemeData
-0.66
daß
-0.65
SourceChecksum
-0.62
SONY
-0.62
الصفحه
-0.62
‡
-0.61
Arkivert
-0.59
-0.58
ニュアル
-0.58
POSITIVE LOGITS
goddamn
0.79
fucking
0.71
weirdly
0.69
lmao
0.67
ibatis
0.66
tbh
0.65
idk
0.64
mierda
0.64
FUCK
0.63
tryna
0.63
Activations Density 0.300%