INDEX
Explanations
countries or entities that are major exporters
words associated with exporters
New Auto-Interp
Negative Logits
usp
-0.71
pron
-0.64
aiman
-0.62
decaying
-0.61
sembly
-0.60
sg
-0.58
scl
-0.57
pt
-0.56
misc
-0.56
Instruct
-0.55
POSITIVE LOGITS
orter
1.34
orters
1.23
culosis
0.86
extraord
0.75
tenance
0.74
itbart
0.72
iety
0.72
sov
0.70
Leban
0.69
tyard
0.66
Activations Density 0.011%