INDEX
Explanations
occurrences of the word "of" in various contexts
of + article/noun
New Auto-Interp
Negative Logits
enderror
-0.41
hospitality
-0.38
ladder
-0.37
Mathias
-0.36
withCredentials
-0.35
cheese
-0.35
<bos>
-0.35
책
-0.34
TA
-0.33
Traditional
-0.33
POSITIVE LOGITS
informée
0.71
IBOutlet
0.66
awtextra
0.62
ImageContext
0.62
ніципалі
0.58
ValueStyle
0.58
odkazy
0.56
новништво
0.55
copies
0.54
ähn
0.53
Activations Density 0.113%