INDEX
Explanations
occurrences of allowed runs in a sports context
New Auto-Interp
Negative Logits
Rouge
-0.17
è·¡
-0.16
letes
-0.14
è¸
-0.14
stery
-0.14
ettes
-0.14
kick
-0.14
견
-0.13
)init
-0.13
ÙĨص
-0.13
POSITIVE LOGITS
unei
0.16
anel
0.15
206
0.14
aptive
0.14
-bottom
0.14
iveness
0.14
enis
0.14
aan
0.14
hou
0.14
aniel
0.14
Activations Density 0.010%