Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Sentence Completion on HellaSwag v1 (test)

81Normalized Accuracy

Mistral-7B

70.39273.14675.978.654Jul 11, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.07
81
2025.07
80.7
2025.07
80.7
2025.07
79.8
2025.07
79.7
2025.07
79.3
2025.07
79.1
2025.07
78.6
2025.07
77.9
2025.07
77.6
2025.07
77.1
2025.07
76.3
2025.07
76.3
2025.07
75.2
2025.07
75.1
2025.07
70.8