INDEX
Explanations
code-related identifiers and structure elements
New Auto-Interp
Negative Logits
ertz
-0.15
rada
-0.15
department
-0.15
ereotype
-0.15
Resident
-0.14
spell
-0.14
Sek
-0.14
BCM
-0.14
å¯
-0.14
¶
-0.14
POSITIVE LOGITS
Thr
0.36
thrift
0.31
THR
0.30
thr
0.25
THR
0.24
thr
0.24
IDL
0.23
Thr
0.23
.scheme
0.20
thrill
0.19
Activations Density 0.002%