Question Answering on Knowledge Tasks (test)
75.93ARC (easy) AccuracyBLOOM176B
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| BLOOM176Bshot=1-shot2023.03 | 75.93 | 50.85 | 64.21 | 77.04 | 67.01 | 0.67 | |
| BLOOMBERGGPTshot=1-shot2023.03 | 73.99 | 48.63 | 65.52 | 77.86 | 66.5 | 0.75 | |
| OPT66Bshot=1-shot2023.03 | 71.25 | 44.54 | 66.42 | 77.58 | 64.95 | 0.33 | |
| GPT-3shot=1-shot2023.03 | 71.2 | 53.2 | — | 80.5 | — | — | |
| GPT-NeoXshot=1-shot2023.03 | 70.79 | 45.39 | 60.36 | 75.84 | 63.1 | 0.08 |