INDEX
Explanations
references to console logging functions in code
New Auto-Interp
Negative Logits
icom
-0.16
ruk
-0.16
hall
-0.16
rc
-0.15
et
-0.15
otre
-0.15
etak
-0.15
etre
-0.15
ady
-0.14
etro
-0.14
POSITIVE LOGITS
.log
0.34
.dir
0.23
.warn
0.23
.table
0.22
.info
0.22
.group
0.21
_vlog
0.20
_log
0.20
.assert
0.19
log
0.19
Activations Density 0.004%