INDEX
Explanations
references to humanitarian crises and protests
New Auto-Interp
Negative Logits
Templ
-0.19
Mario
-0.15
Hindu
-0.15
857
-0.15
andler
-0.15
oci
-0.15
Goa
-0.14
554
-0.14
Deniz
-0.14
Hindus
-0.14
POSITIVE LOGITS
Sudan
0.48
Kh
0.31
sud
0.30
Bash
0.29
سÙĪØ¯
0.29
Dar
0.28
.sd
0.27
SPL
0.27
Nile
0.27
Dar
0.26
Activations Density 0.014%