INDEX
Explanations
references to Islam and Muslims
references to Islam and Muslim identity
New Auto-Interp
Negative Logits
testing
-0.81
billing
-0.79
Pike
-0.76
Clyde
-0.75
clut
-0.74
gru
-0.74
Hank
-0.70
sob
-0.68
delivery
-0.68
Bott
-0.68
POSITIVE LOGITS
Islam
3.68
Muslim
2.53
Islamic
2.44
Muslims
2.42
Pakistan
1.65
Jews
1.61
Allah
1.59
Arab
1.49
ISIS
1.44
Syrian
1.39
Activations Density 0.030%