INDEX
Explanations
references to immigration and relocation experiences
New Auto-Interp
Negative Logits
igne
-0.16
éľĬ
-0.15
eyer
-0.15
zier
-0.15
sobie
-0.15
oire
-0.15
CAUSED
-0.15
indle
-0.15
itur
-0.14
andoned
-0.14
POSITIVE LOGITS
intent
0.31
looking
0.30
via
0.30
accompanied
0.29
carrying
0.28
seeking
0.28
armed
0.25
bearing
0.25
hoping
0.25
to
0.24
Activations Density 0.548%