INDEX
Explanations
statements related to education and political issues
New Auto-Interp
Negative Logits
ãĤ´ãĥ³
-0.86
surprisingly
-0.78
©¶æ¥µ
-0.63
xtap
-0.62
ãĤ¨ãĥ«
-0.58
quir
-0.58
ometimes
-0.58
utterstock
-0.57
ËĪ
-0.56
NFC
-0.56
POSITIVE LOGITS
..."
1.99
â̦"
1.96
.")
1.86
,'"
1.78
%"
1.72
',"
1.70
,"
1.68
),"
1.68
â̦"
1.65
'"
1.65
Activations Density 8.528%