INDEX
Explanations
mentions of institutions like Johns Hopkins University
mentions of specific universities and individuals associated with them
New Auto-Interp
Negative Logits
mble
-0.81
tto
-0.80
trap
-0.67
Spread
-0.65
anza
-0.63
wreck
-0.62
ment
-0.62
charg
-0.62
lier
-0.61
liest
-0.61
POSITIVE LOGITS
Hopkins
1.01
istry
0.96
sonian
0.86
Johns
0.85
insula
0.84
imore
0.83
otom
0.83
ons
0.81
arton
0.80
omen
0.77
Activations Density 0.029%