INDEX
Explanations
references to companies and their characteristics
New Auto-Interp
Negative Logits
abbo
-0.15
ened
-0.15
qua
-0.14
door
-0.14
uten
-0.14
most
-0.14
assi
-0.14
xes
-0.14
748
-0.14
lings
-0.13
POSITIVE LOGITS
wide
0.26
-wide
0.19
eur
0.16
ัà¸Ĺ
0.16
eval
0.15
oldown
0.15
-name
0.15
oleans
0.14
/state
0.14
ëª
0.14
Activations Density 0.079%