INDEX
Explanations
topics related to environmental issues and activism
New Auto-Interp
Negative Logits
arious
-0.70
Staff
-0.64
brate
-0.61
onomic
-0.60
æł
-0.60
ideshow
-0.59
bris
-0.59
agent
-0.58
oho
-0.58
agency
-0.57
POSITIVE LOGITS
versus
0.98
vs
0.95
versions
0.88
outper
0.87
fared
0.86
tended
0.86
version
0.86
tends
0.84
Versus
0.83
differs
0.83
Activations Density 0.214%