INDEX
Explanations
punctuation marks, particularly periods and quotation marks
New Auto-Interp
Negative Logits
ongo
-0.15
Continue
-0.15
SKU
-0.14
Continue
-0.14
plusplus
-0.14
cán
-0.14
annonces
-0.14
sson
-0.14
posted
-0.14
ologists
-0.13
POSITIVE LOGITS
Correction
0.25
Else
0.23
Meanwhile
0.23
Meanwhile
0.22
Else
0.21
elsewhere
0.20
Separ
0.19
COR
0.18
correction
0.18
separately
0.18
Activations Density 0.028%