INDEX
Explanations
phrases indicating physical or emotional experiences related to how people interact with video games or media
New Auto-Interp
Negative Logits
941
-0.16
imoto
-0.15
ixe
-0.15
ikat
-0.15
лада
-0.15
ognitive
-0.14
uden
-0.14
ereg
-0.14
art
-0.14
result
-0.14
POSITIVE LOGITS
iscard
0.15
arness
0.15
topics
0.15
topic
0.14
_topic
0.14
ipsoid
0.14
thag
0.14
nst
0.14
sparing
0.14
unde
0.14
Activations Density 0.033%