atsuki-yamaguchi's picture
Upload README.md with huggingface_hub
0cb09ba verified
metadata
license: apache-2.0
datasets:
  - allenai/MADLAD-400
language:
  - ig
base_model:
  - allenai/OLMo-2-1124-7B-Instruct

OLMo 2 1124 7B Instruct for Igbo: SSU-Wanda (12.5% freezing)

This model is built on top of OLMo 2 1124 7B Instruct adapted for Igbo using 200M target language tokens sampled from MADLAD-400. The model is adapted using the SSU-Wanda approach with 12.5% freezing.

Model Description

Model Sources

How to Get Started with the Model

Use the code below to get started with the model.

from transformers import AutoTokenizer, AutoModelForCausalLM

model = AutoModelForCausalLM.from_pretrained(
    "ssu-project/OLMo-2-1124-7B-Instruct-ig-ssu_0.125"
)
tokenizer = AutoTokenizer.from_pretrained(
    "ssu-project/OLMo-2-1124-7B-Instruct-ig-ssu_0.125"
)

Citation

@misc{yamaguchi2025mitigatingcatastrophicforgettingtarget,
    title={Mitigating Catastrophic Forgetting in Target Language Adaptation of LLMs via Source-Shielded Updates}, 
    author={Atsuki Yamaguchi and Terufumi Morishita and Aline Villavicencio and Nikolaos Aletras},
    year={2025},
    eprint={2512.04844},
    archivePrefix={arXiv},
    primaryClass={cs.CL},
    url={https://arxiv.org/abs/2512.04844}, 
}