OLMo-2-1124-7B-Instruct-ky-adalora

OLMo 2 1124 7B Instruct for Kyrgyz: AdaLoRA

This model is built on top of OLMo 2 1124 7B Instruct adapted for Kyrgyz using 200M target language tokens sampled from MADLAD-400. The model is adapted using the AdaLoRA approach. This is based on https://arxiv.org/abs/2303.10512 and was the best-performing LoRA-based method in the HFT paper.

Model Description

• Language: Kyrgyz
• License: Apache 2.0
• Fine-tuned from model: allenai/OLMo-2-1124-7B-Instruct

Model Sources

• Repository: https://github.com/gucci-j/ssu
• Paper: https://arxiv.org/abs/2512.04844

How to Get Started with the Model

Use the code below to get started with the model.

from transformers import AutoTokenizer, AutoModelForCausalLM

from peft import PeftModel

base_model = AutoModelForCausalLM.from_pretrained(
    "allenai/OLMo-2-1124-7B-Instruct",
)
model = PeftModel.from_pretrained(
    base_model,
    "ssu-project/OLMo-2-1124-7B-Instruct-ky-adalora",
)
model = model.merge_and_unload()
tokenizer = AutoTokenizer.from_pretrained(
    "allenai/OLMo-2-1124-7B-Instruct"
)

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}, 
}