INDEX
Explanations
references to or emphasis on auxiliary information like footnotes or asides
New Auto-Interp
Negative Logits
protoimpl
-0.81
enough
-0.57
للمعارف
-0.56
învă
-0.56
enough
-0.53
}))
-0.50
}\]
-0.49
Etwas
-0.48
ihnachts
-0.48
디오
-0.47
POSITIVE LOGITS
BTW
3.42
btw
3.23
BTW
2.98
btw
2.66
Btw
2.31
FYI
1.90
incidentally
1.82
FYI
1.68
übrigens
1.51
Incidentally
1.42
Activations Density 0.001%