INDEX
Explanations
references to visual observations or descriptions involving specific subjects, particularly concerning people or situations
New Auto-Interp
Head Attr Weights
0:0.06
1:0.02
2:0.10
3:0.09
4:0.05
5:0.11
6:0.03
7:0.02
8:0.17
9:0.17
10:0.08
11:0.04
Negative Logits
��
-1.02
anke
-1.01
Trust
-0.99
keeping
-0.98
owder
-0.96
withholding
-0.95
Habit
-0.94
hiring
-0.93
imester
-0.93
nominating
-0.92
POSITIVE LOGITS
VIDEOS
1.34
pics
1.25
resemblance
1.21
ONY
1.19
similarities
1.18
soDeliveryDate
1.17
ideos
1.15
unfold
1.13
pictures
1.13
RAW
1.11
Activations Density 0.103%