Summary of Scaling Laws with Vocabulary: Larger Models Deserve Larger Vocabularies, by Chaofan Tao et al.
Scaling Laws with Vocabulary: Larger Models Deserve Larger Vocabularies
by Chaofan Tao, Qian Liu, Longxu Dou, Niklas Muennighoff, Zhongwei Wan, Ping Luo, Min Lin, Ngai Wong
First submitted to arxiv on: 18 Jul 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper investigates the impact of vocabulary size on large language model scaling laws. The authors propose three approaches to predict the optimal vocabulary size based on compute budget, and validate their predictions by training models with varying FLOPs budgets. The results show that increasing the vocabulary size can improve downstream performance, highlighting the importance of jointly considering tokenization and model scaling for efficient pre-training. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Large language models (LLMs) are getting bigger! But did you know that the number of words they know also matters? This paper looks at how many words LLMs should know to work well. They tried different approaches to figure out the right number, and found that bigger models need more words. The authors even tested their ideas by training some big models, and saw that using more words helped them do better on certain tasks. |
Keywords
» Artificial intelligence » Large language model » Scaling laws » Tokenization
