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

Generation on Big-Bench Hard (test)

57.9Exact Match

FLAN-PaLM 540B

11.20423.32735.4547.573Dec 22, 2022
Updated 1mo ago

Evaluation Results

MethodLinks
2022.12
57.9
2022.12
52.8
2022.12
50.9
2022.12
49.1
2022.12
48.6
2022.12
47.5
2022.12
45.3
2022.12
37.4
2022.12
35.7
2022.12
33.6
2022.12
30.9
2022.12
30.2
2022.12
29.5
2022.12
28.4
2022.12
27.9
2022.12
26.5
2022.12
13