INDEX
Explanations
references to entities named "Beast" from a range of contexts
New Auto-Interp
Negative Logits
bered
-0.76
PsyNetMessage
-0.74
encing
-0.69
enza
-0.69
psons
-0.68
ixel
-0.65
arning
-0.65
Rockies
-0.65
stanbul
-0.64
Winchester
-0.63
POSITIVE LOGITS
mode
0.93
ly
0.90
ie
0.90
lord
0.90
layer
0.89
roots
0.88
ies
0.87
lords
0.86
buster
0.86
ous
0.85
Activations Density 0.027%