INDEX
Explanations
objects associated with containers and storage items
New Auto-Interp
Negative Logits
953
-0.17
ihn
-0.15
aterno
-0.14
597
-0.14
tract
-0.14
ilight
-0.14
ABCDEFGHI
-0.14
204
-0.14
498
-0.14
utely
-0.14
POSITIVE LOGITS
ful
0.43
containing
0.39
full
0.38
-full
0.36
full
0.35
/container
0.32
contents
0.32
filled
0.29
FULL
0.29
lid
0.28
Activations Density 0.106%