INDEX
Explanations
references to stories inspired by real-life events and true stories
New Auto-Interp
Negative Logits
bine
-0.16
Processed
-0.14
iaux
-0.14
BOOT
-0.14
preferredStyle
-0.14
HORT
-0.13
rede
-0.13
IRST
-0.13
imat
-0.13
greg
-0.13
POSITIVE LOGITS
real
0.65
actual
0.57
actual
0.54
real
0.52
REAL
0.50
Actual
0.47
Actual
0.45
réal
0.44
Real
0.44
_real
0.43
Activations Density 0.186%