Question Answering on Paralex
61.1AccuracyOur Trained Model
Evaluation Results
| Method | Links | |
|---|---|---|
| Our Trained ModelModel Size=52B, Data type=W8A8, Zero-shot=true2023.05 | 61.1 | |
| Our Trained ModelModel Size=52B, Data type=FP16, Zero-shot=true2023.05 | 61.05 | |
| Our Trained ModelModel Size=13B, Data type=W8A8, Zero-shot=true2023.05 | 60.9 | |
| Our Trained ModelModel Size=13B, Data type=FP16, Zero-shot=true2023.05 | 60.68 | |
| Our Trained ModelModel Size=52B, Data type=W4, Zero-shot=true2023.05 | 60.43 | |
| Our Trained ModelModel Size=6B, Data type=W8A8, Zero-shot=true2023.05 | 60.2 | |
| Our Trained ModelModel Size=6B, Data type=FP16, Zero-shot=true2023.05 | 60.1 | |
| Our Trained ModelModel Size=13B, Data type=W4, Zero-shot=true2023.05 | 59.74 | |
| Our Trained ModelModel Size=6B, Data type=W4, Zero-shot=true2023.05 | 59.01 | |
| Our Trained ModelModel Size=410M, Data type=FP16, Zero-shot=true2023.05 | 57.1 | |
| Our Trained ModelModel Size=410M, Data type=W8A8, Zero-shot=true2023.05 | 57 | |
| Our Trained ModelModel Size=410M, Data type=W4, Zero-shot=true2023.05 | 54.6 |