INDEX
Explanations
references to government initiatives and plans
New Auto-Interp
Head Attr Weights
0:0.08
1:0.04
2:0.01
3:0.16
4:0.07
5:0.19
6:0.05
7:0.01
8:0.13
9:0.19
10:0.00
11:0.01
Negative Logits
jug
-2.18
slightest
-1.85
decent
-1.81
screaming
-1.77
liter
-1.68
correctly
-1.68
whiff
-1.66
throats
-1.64
qus
-1.62
scream
-1.60
POSITIVE LOGITS
redevelopment
1.64
Expansion
1.62
Architecture
1.62
Enhance
1.61
reon
1.59
redes
1.59
��
1.56
solete
1.56
覚醒
1.54
crowdfunding
1.54
Activations Density 0.132%