INDEX
Explanations
instances where a comparison or contrast is made between different entities
the preposition "with" indicating relationships or connections in various contexts
New Auto-Interp
Negative Logits
shire
-0.69
âĢİ
-0.68
BP
-0.68
pton
-0.67
Ø©
-0.63
ais
-0.62
ights
-0.62
hops
-0.61
winter
-0.60
unes
-0.60
POSITIVE LOGITS
stood
1.52
regard
1.47
regards
1.44
draw
1.36
drawn
1.29
standing
1.23
respect
1.20
impunity
1.07
holding
0.99
hindsight
0.95
Activations Density 0.178%