INDEX
Explanations
phrases that emphasize a critical argument or main idea
New Auto-Interp
Negative Logits
eso
-0.16
.Apis
-0.15
dej
-0.15
accounts
-0.14
ijke
-0.14
ibal
-0.14
combe
-0.13
igel
-0.13
STEM
-0.13
InnerText
-0.13
POSITIVE LOGITS
point
0.36
point
0.29
-point
0.28
(point
0.27
.point
0.26
made
0.25
Point
0.25
POINT
0.25
points
0.24
POINT
0.23
Activations Density 0.023%