Language Understanding on MMLU (dev)
55.6AccuracyFLAN-UL2 20B
Evaluation Results
| Method | Links | |
|---|---|---|
| FLAN-UL2 20Bfew-shot=true, checkpoin t=best for both tasks2022.05 | 55.6 | |
| FLAN-UL2 20Bfew-shot=true, checkpoin t=individual task best2022.05 | 55.1 | |
| FLAN T5 XXLfew-shot=true2022.05 | 54.5 | |
| OPT IML-Max 175Bfew-shot=true2022.05 | 49.1 | |
| T0p p 11Bfew-shot=true2022.05 | 46.7 | |
| OPT IML-Max 30Bfew-shot=true2022.05 | 46.3 | |
| OPT 175Bfew-shot=true2022.05 | 27.3 | |
| OPT 30Bfew-shot=true2022.05 | 23.5 |