INDEX
Explanations
disjunctions or alternatives in sentences, particularly the word 'or.'
New Auto-Interp
Negative Logits
itzer
-0.17
enerator
-0.17
ylvania
-0.16
loub
-0.16
tdown
-0.15
asca
-0.15
ubern
-0.14
IDO
-0.14
endum
-0.14
./(
-0.14
POSITIVE LOGITS
indeed
0.16
Trial
0.15
************************************************************************
0.15
ss
0.15
beck
0.15
ooter
0.15
trial
0.14
Warwick
0.14
-r
0.14
maybe
0.14
Activations Density 0.048%