INDEX
Explanations
references to individuals and social relationships, particularly focusing on the historical context of enslaved populations and their descendants
New Auto-Interp
Negative Logits
590
-0.16
atis
-0.16
apiro
-0.15
imprison
-0.15
-send
-0.15
stdClass
-0.14
.SC
-0.14
ventured
-0.14
recursive
-0.14
yk
-0.14
POSITIVE LOGITS
bec
0.20
become
0.17
BIT
0.17
cla
0.17
defect
0.16
becoming
0.16
ITCH
0.16
langu
0.15
Become
0.15
becomes
0.15
Activations Density 0.338%