INDEX
Explanations
personal pronouns combined with verbs related to programming or technical tasks
instances of the pronoun "I" in a personal narrative context
New Auto-Interp
Negative Logits
independence
-0.66
Constitutional
-0.64
ideological
-0.63
tnc
-0.62
Rockefeller
-0.61
malnutrition
-0.60
mutants
-0.60
careers
-0.60
apartheid
-0.60
Jarrett
-0.60
POSITIVE LOGITS
'm
1.23
've
1.08
EEE
1.03
WB
0.97
'll
0.97
WI
0.96
OTA
0.95
'd
0.90
nex
0.84
recommend
0.84
Activations Density 0.279%