INDEX
Explanations
phrases related to social issues and crises
phrases related to social and environmental crises
New Auto-Interp
Negative Logits
Offer
-0.71
beit
-0.67
propri
-0.66
nods
-0.63
Appearance
-0.63
Interesting
-0.62
iple
-0.62
endment
-0.61
oult
-0.59
Helpful
-0.59
POSITIVE LOGITS
worsened
1.30
worsen
1.23
worsening
1.21
exacerbated
1.20
devast
1.17
exacerbate
1.11
rav
1.11
epidemic
1.11
disproportionately
1.11
spir
1.10
Activations Density 0.575%