INDEX
Explanations
names of specific political figures
references to political figures, particularly Hosni Mubarak and Khamenei
New Auto-Interp
Negative Logits
atu
-0.71
rient
-0.66
ãĤ´ãĥ³
-0.66
americ
-0.65
theless
-0.64
asure
-0.63
Clicker
-0.63
rification
-0.63
Unch
-0.63
UCK
-0.62
POSITIVE LOGITS
sein
0.92
ouri
0.84
ework
0.82
eni
0.76
elin
0.73
lers
0.72
Haf
0.70
nard
0.69
soever
0.67
ename
0.67
Activations Density 0.058%