Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Generation on Big-Bench Hard (test)
Loading...
57.9
Exact Match
FLAN-PaLM 540B
11.204
23.327
35.45
47.573
Dec 22, 2022
Exact Match
Updated 3d ago
Evaluation Results
Method
Method
Links
Exact Match
FLAN-PaLM 540B
# shots=3
2022.12
57.9
OpenAI code-davinci-002
# shots=3
2022.12
52.8
OpenAI text-davinci-003
# shots=3
2022.12
50.9
PaLM 540B
# shots=3
2022.12
49.1
OpenAI text-davinci-002
# shots=3
2022.12
48.6
FLAN-PaLM 62B
# shots=3
2022.12
47.5
FLAN-T5 11B
# shots=3
2022.12
45.3
PaLM 62B
# shots=3
2022.12
37.4
OPT-IML-Max 175B
# shots=3
2022.12
35.7
OpenAI davinci
# shots=3
2022.12
33.6
OPT-IML-Max 30B
# shots=3
2022.12
30.9
OPT 175B
# shots=3
2022.12
30.2
T5 11B
# shots=3
2022.12
29.5
OPT 30B
# shots=3
2022.12
28.4
OPT 1.3B
# shots=3
2022.12
27.9
OPT-IML-Max 1.3B
# shots=3
2022.12
26.5
T0pp 11B
# shots=3
2022.12
13
Feedback
Search any
task
Search any
task