INDEX
Explanations
tokens in various Indic scripts or languages
New Auto-Interp
Negative Logits
reira
-0.16
adb
-0.16
990
-0.15
elling
-0.15
352
-0.14
et
-0.14
matt
-0.14
386
-0.14
102
-0.14
odo
-0.14
POSITIVE LOGITS
¸
0.32
²
0.31
¤
0.30
ķ
0.30
¬
0.30
®
0.29
ļ
0.28
¯
0.28
¦
0.28
Ĺ
0.28
Activations Density 0.004%