Machine Translation on WMT23 En-Uk (test)
27.4BLEUX-ALMA-13B (Only SFT, MoE)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| X-ALMA-13B (Only SFT, MoE)Parameters=13B, Training Strategy=Only SFT, Architecture=MoE2025.02 | 27.4 | 85.3 | |
| NLLB-3.3BParameters=3.3B2025.02 | 25.5 | 82.8 | |
| Aya-23-35BParameters=35B2025.02 | 24.9 | 84 | |
| Aya-23-8BParameters=8B2025.02 | 24.3 | 84.3 | |
| X-ALMA-13B-DAT (MoE)Parameters=13B, Training Strategy=DAT, Architecture=MoE2025.02 | 23.5 | 84.6 | |
| X-ALMA-13B-DATM (MoE)Parameters=13B, Training Strategy=DATM, Architecture=MoE2025.02 | 23.5 | 84.6 | |
| LLaMAX3-Alpaca-8BParameters=8B2025.02 | 19.8 | 80.6 | |
| Aya-1012025.02 | 19.7 | 82.7 |