INDEX
Explanations
phrases related to actions and processes involving cooking and food preparation
New Auto-Interp
Negative Logits
нен
-0.15
/Public
-0.14
anye
-0.14
aci
-0.14
APON
-0.13
isl
-0.13
Workspace
-0.13
-Tr
-0.13
/Peak
-0.13
agh
-0.13
POSITIVE LOGITS
differently
0.28
wrong
0.24
slow
0.24
according
0.23
with
0.21
slower
0.21
fast
0.21
iffer
0.21
without
0.20
exactly
0.19
Activations Density 0.663%