Saved in:
Bibliographic Details
Main Authors: Gu, Qingshui, Li, Shu, Zheng, Tianyu, Zhang, Zhaoxiang
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.06635
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916611527540736
author Gu, Qingshui
Li, Shu
Zheng, Tianyu
Zhang, Zhaoxiang
author_facet Gu, Qingshui
Li, Shu
Zheng, Tianyu
Zhang, Zhaoxiang
contents Steel-LLM is a Chinese-centric language model developed from scratch with the goal of creating a high-quality, open-source model despite limited computational resources. Launched in March 2024, the project aimed to train a 1-billion-parameter model on a large-scale dataset, prioritizing transparency and the sharing of practical insights to assist others in the community. The training process primarily focused on Chinese data, with a small proportion of English data included, addressing gaps in existing open-source LLMs by providing a more detailed and practical account of the model-building journey. Steel-LLM has demonstrated competitive performance on benchmarks such as CEVAL and CMMLU, outperforming early models from larger institutions. This paper provides a comprehensive summary of the project's key contributions, including data collection, model design, training methodologies, and the challenges encountered along the way, offering a valuable resource for researchers and practitioners looking to develop their own LLMs. The model checkpoints and training script are available at https://github.com/zhanshijinwat/Steel-LLM.
format Preprint
id arxiv_https___arxiv_org_abs_2502_06635
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Steel-LLM:From Scratch to Open Source -- A Personal Journey in Building a Chinese-Centric LLM
Gu, Qingshui
Li, Shu
Zheng, Tianyu
Zhang, Zhaoxiang
Computation and Language
Artificial Intelligence
Steel-LLM is a Chinese-centric language model developed from scratch with the goal of creating a high-quality, open-source model despite limited computational resources. Launched in March 2024, the project aimed to train a 1-billion-parameter model on a large-scale dataset, prioritizing transparency and the sharing of practical insights to assist others in the community. The training process primarily focused on Chinese data, with a small proportion of English data included, addressing gaps in existing open-source LLMs by providing a more detailed and practical account of the model-building journey. Steel-LLM has demonstrated competitive performance on benchmarks such as CEVAL and CMMLU, outperforming early models from larger institutions. This paper provides a comprehensive summary of the project's key contributions, including data collection, model design, training methodologies, and the challenges encountered along the way, offering a valuable resource for researchers and practitioners looking to develop their own LLMs. The model checkpoints and training script are available at https://github.com/zhanshijinwat/Steel-LLM.
title Steel-LLM:From Scratch to Open Source -- A Personal Journey in Building a Chinese-Centric LLM
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2502.06635