INDEX
Explanations
proper nouns
the presence of the word "Tag" in various forms and contexts
New Auto-Interp
Negative Logits
obs
-0.69
cond
-0.67
evapor
-0.66
culus
-0.65
rec
-0.65
neum
-0.63
wings
-0.63
requ
-0.62
pine
-0.62
syndrome
-0.61
POSITIVE LOGITS
Tag
3.84
Tag
2.59
TAG
1.40
Tags
1.38
tag
1.38
Tags
1.31
Bag
1.30
tags
1.24
Breed
1.17
Beer
1.13
Activations Density 0.024%