Multilingual-Transfer
Collection
Pretraining models to find what allows multilingual transfer • 30 items • Updated • 2
llama3-7b-en-anchored-ar-v2
LLaMA-3 7B pretrained on English-Arabic bilingual data with anchored training, 134k steps (injection fix v2).
Validation Results
Final checkpoint: step 133,600
| Validation set | Cross-entropy loss (nats) | Perplexity |
|---|---|---|
| English (en) | 2.267 | 9.65 |
AR (ar) |
2.8797 | 17.81 |
Note: Rerun with corrected injection config (Buckwalter transliteration fix + injection count fix).
Evaluation Results
EEE-format evaluation results are stored under eval_results/eee/ in this repository.
Tasks: Global MMLU (EN/AR), PIQA, ECLeKTic, Fictive Entity (2-rate mix).
Citation
Part of the The-CoLab multilingual-transfer collection.
- Downloads last month
- 44
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support