INDEX
Explanations
references to cooking and food-related activities
New Auto-Interp
Negative Logits
udi
-0.16
oli
-0.15
atically
-0.15
ewart
-0.15
udes
-0.14
uv
-0.14
rost
-0.14
Berk
-0.14
Coast
-0.14
acom
-0.14
POSITIVE LOGITS
sey
0.23
PAD
0.21
ery
0.21
ware
0.19
pad
0.17
oÅĻ
0.17
tha
0.17
erm
0.16
ERY
0.16
ingham
0.16
Activations Density 0.024%