INDEX
Explanations
instances of numerical data or quantities in the text
New Auto-Interp
Negative Logits
imer
-0.17
er
-0.17
hap
-0.15
379
-0.15
ahl
-0.14
142
-0.14
duct
-0.14
ubern
-0.14
hil
-0.14
iff
-0.14
POSITIVE LOGITS
ponde
0.17
oldown
0.15
resa
0.15
ocuk
0.15
rophe
0.14
Äįan
0.14
ANNEL
0.14
šak
0.14
itudes
0.14
artz
0.14
Activations Density 0.047%