INDEX
Explanations
phrases or words with special characters such as symbols or accented letters
special characters or unusual text formatting
New Auto-Interp
Negative Logits
recl
-0.74
ostr
-0.71
othal
-0.71
sofa
-0.67
iferation
-0.66
inx
-0.65
aimon
-0.64
vow
-0.64
omnia
-0.64
pouch
-0.63
POSITIVE LOGITS
¢
1.12
��������
1.10
����
1.06
�
1.02
�
0.88
taboola
0.85
bold
0.84
SEE
0.82
âĤ¬
0.81
��
0.80
Activations Density 0.005%