Llama 3.1 8B Instruct for Sinhala: ElChat (Linear merging)

Model Details

• Vocabulary: This model has an additional target vocabulary of 10K.
• Target vocabulary initialization: The target weights of the embedding and LM head were initialized using mean initialization.
• Training: This model was continually pre-trained on 500M target language tokens sampled from MADLAD-400.
• Post-processing: The model was post-processed using the ElChat method with linear merging instead of SLERP.

Model Description

• Language: Sinhala
• License: Llama 3.1 Community License Agreement
• Fine-tuned from model: meta-llama/Llama-3.1-8B-Instruct

Model Sources

• Repository: https://github.com/gucci-j/chat-cve
• Paper: https://arxiv.org/abs/2412.11704

How to Get Started with the Model

Citation

@article{yamaguchi2025adapting,
      title={Adapting Chat Language Models Using Only Target Unlabeled Language Data}, 
      author={Atsuki Yamaguchi and Terufumi Morishita and Aline Villavicencio and Nikolaos Aletras},
      journal={Transactions on Machine Learning Research},
      issn={2835-8856},
      year={2025},
      url={https://openreview.net/forum?id=6IdoIKowfe},
      note={}
}