PaLM-62B (few-shot, k=5)
Reported on 5 benchmarks across 4 tasks · 2 papers
Note: results are matched by exact model name. Different papers may use the same name for different model variants.
Natural Language Processing3 results
- Accuracy· 2022-04-0559.4best: 71.9 (PaLM-540B (few-shot, k=5))
- Accuracy· 2022-04-0561best: 65.9 (PaLM-540B (few-shot, k=5))
- Accuracy· 2022-04-0577.7best: 95.4 (PaLM-540B (few-shot, k=5))
Methodology2 results
- Accuracy· 2023-05-1736.5best: 44 (Chinchilla-70B (few-shot, k=5))
- Accuracy· 2022-04-0565.4best: 73.9 (PaLM-540B (few-shot, k=5))