INDEX
Explanations
terms related to race and racial issues
New Auto-Interp
Negative Logits
inition
-0.15
.metamodel
-0.15
Baz
-0.15
ãĥ«ãĥĪ
-0.14
osex
-0.14
yd
-0.14
ogl
-0.14
yun
-0.14
lication
-0.14
qv
-0.14
POSITIVE LOGITS
profiling
0.24
/color
0.20
cleansing
0.20
-prof
0.19
icious
0.18
profiler
0.17
Cleans
0.17
ized
0.17
/class
0.17
/E
0.17
Activations Density 0.031%