INDEX
Explanations
symbols and special characters within the text
New Auto-Interp
Head Attr Weights
0:0.11
1:0.09
2:0.11
3:0.07
4:0.07
5:0.09
6:0.09
7:0.07
8:0.09
9:0.04
10:0.05
11:0.07
Negative Logits
ogun
-1.62
endings
-1.26
20439
-1.26
-1.24
culosis
-1.23
iership
-1.22
aeda
-1.21
regor
-1.20
Replay
-1.16
omial
-1.15
POSITIVE LOGITS
NRS
1.40
REDACTED
1.17
SOURCE
1.14
█
1.13
liquid
1.13
Brilliant
1.12
occup
1.12
wise
1.11
reflect
1.10
trak
1.08
Activations Density 0.001%