INDEX
Explanations
references to theft or heist-related activities
New Auto-Interp
Negative Logits
_idxs
-0.16
LDS
-0.14
VRT
-0.14
erca
-0.14
Weber
-0.14
adera
-0.14
Jake
-0.14
_buckets
-0.13
Welsh
-0.13
AttributeName
-0.13
POSITIVE LOGITS
Bat
0.50
Batman
0.49
bat
0.46
bat
0.44
BAT
0.43
Gotham
0.43
Bat
0.43
Batman
0.41
BAT
0.40
.bat
0.35
Activations Density 0.024%