BaichuanSEED: Sharing the Potential of ExtensivE Data Collection and Deduplication by Introducing a Competitive Large Language Model Baseline
Year: 2024
Type: preprint
Abstract: The general capabilities of Large Language Models (LLM) highly rely on the composition and selection on extensive pretraining datasets, treated as commercial secrets by several institutions. To mitigate this issue, we open-source the details of a universally applicable data processing pipeline and validate its effectiveness and potential by introducing a competitive LLM baseline. Specifically, the data processing pipeline consists of broad collection to scale up and reweighting to improve quality. We then pretrain a 7B model BaichuanSEED with 3T tokens processed by our pipeline without any deliberate downstream task-related optimization, followed by an easy but effective supervised fine-tuning stage. BaichuanSEED demonstrates consistency and predictability throughout training and achieves comparable performance on comprehensive benchmarks with several commercial advanced large language models, such as Qwen1.5 and Llama3. We also conduct several heuristic experiments to discuss the potential for further optimization of downstream tasks, such as mathematics and coding.
Source: arXiv (Cornell University)
Authors: Dong, Guosheng, Pan, Da, Sun, Yiding, Zhang, Shusen, Liang, Zheng, Wu, Xin, Shen, Yanjun, Yang, Fan, Sun, Haoze, Li, Tianpeng, Lin, Mingan, Xu, Jianhua, Zhang, Yufan, Nie, Xiaonan, Su, Lei, Wang, Bingning, Zhang, Wentao, Mao, Jiaxin, Zhou, Zenan, Chen, Weipeng
Institutions: Baichuan (China), Renmin University of China, Peking University
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4402705130