INDEX
Explanations
references to firearms and their specifications
New Auto-Interp
Negative Logits
uden
-0.17
hire
-0.14
bomb
-0.14
.scalablytyped
-0.14
figur
-0.14
hire
-0.14
apons
-0.14
isque
-0.14
_unix
-0.14
inte
-0.14
POSITIVE LOGITS
556
0.24
suppressed
0.23
Upper
0.22
upper
0.22
chamber
0.22
suppress
0.21
Gas
0.20
suppress
0.20
AR
0.20
Suppress
0.19
Activations Density 0.027%