Metric	Score
Likelihood	0.00000
Accuracy	0.51464

	Likelihood	Accuracy
+H	0.00000	0.47250
+C	0.00000	0.50000
-C	0.00000	0.51510

Metric	Score
Likelihood	0.00000
Accuracy	0.51665

	Likelihood	Accuracy
+H	0.00000	0.00000
+C	0.00000	0.00000
-C	0.00000	0.51665

worst items

note: the gold standard is taken from the submission itself, not from the challenge data!

#	input	expected output	actual output	dev-0 Accuracy +C
1	zakończyłem jakiś czas temu. Potem dość długie lata śpiewałem w chórze for-humans contaminated	0	1	0.00000

Metric	Score
Likelihood	0.00000
Accuracy	0.51690

	Likelihood	Accuracy
+H	0.00000	0.51000
+C	0.00000	0.00000
-C	0.00000	0.51691