mala-500-10b-v2

MaLA-500: Massive Language Adaptation of Large Language Models

MaLA-500 is a novel large language model designed to cover an extensive range of 534 languages. This model builds upon LLaMA 2 7B and integrates continued pretraining with vocabulary extension, with an expanded vocabulary size of 260,164, and LoRA low-rank adaptation.

• Continued Pretraining: Enhances the model's ability to adapt to a wide range of languages.
• LoRA Low-Rank Adaptation: LoRA low-rank adaptation refines the model's adaptation capabilities.
• Vocabulary Extension: MaLA-500 boasts an extended vocabulary size of 260,164.
• Multilingual Proficiency: Trained on Glot500-c, covering 534 languages.

With vocabulary extension and LoRA modules, the MaLA-500 introduces additional 2.1B trainable parameters, making the total parameters to be 10.7B.

Please refer to our paper for more details.

How to Get Started with the Model

Requirements:

transformers>=4.36.1
peft>=0.6.2

Use the code below to get started with the model.

from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import PeftModel

base_model = AutoModelForCausalLM.from_pretrained('meta-llama/Llama-2-7b-hf')
base_model.resize_token_embeddings(260164)
tokenizer = AutoTokenizer.from_pretrained('MaLA-LM/mala-500-10b')
model = PeftModel.from_pretrained(base_model, 'MaLA-LM/mala-500-10b')

Citation

@misc{lin2024mala500,
      title={MaLA-500: Massive Language Adaptation of Large Language Models}, 
      author={Peiqin Lin and Shaoxiong Ji and Jörg Tiedemann and André F. T. Martins and Hinrich Schütze},
      year={2024},
      eprint={2401.13303},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}