Note: The H800 is a variant of the H100 for the Chinese market
OUTLINE:
0:00 - Introduction
5:30 - Adding new blocks to LLaMA
15:00 - Block expansion
27:40 - Experiments
30:40 - Conclusion
Paper: https://arxiv.org/abs/2401.02415
Other Paper: https://pro…
Note: The H800 is a variant of the H100 for the Chinese market
OUTLINE:
0:00 - Introduction
5:30 - Adding new blocks to LLaMA
15:00 - Block expansion
27:40 - Experiments
30:40 - Conclusion
Paper: https://arxiv.org/abs/2401.02415
Other Paper: https://proceedings.mlr.press/v162/shen22f/shen22f.pdf
Abstract:
Humans generally acquire new skills without compromising the old; however, the opposite holds for Large Language Models (LLMs), e.g., from LLaMA to CodeLLaMA. To this end, we propose a new post-pretraining method for LLMs with an expansion of Transformer blocks. We tune the expanded blocks using only new corpus, efficiently and effectively improving the model's knowledge without catastrophic forgetting. In this paper, we experiment on the corpus of code and math, yielding LLaMA Pro-8.3B, a versatile foundation model initialized from LLaMA2-7B, excelling in general tasks, programming, and mathematics. LLaMA Pro and its instruction-following counterpart (LLaMA Pro-Instruct) achieve advanced performance among various benchmarks, demonstrating superiority over existing open models in the LLaMA family and the immense potential of reasoning and addressing diverse tasks as an intelligent agent. Our findings provide valuable insights into integrating natural and programming languages, laying a solid foundation for developing advanced language agents that operate effectively in various environments.
Authors: Chengyue Wu, Yukang Gan, Yixiao Ge, Zeyu Lu, Jiahao Wang, Ye Feng, Ping Luo, Ying Shan
Links:
Homepage: https://ykilcher.com
Merch: https://ykilcher.com/merch
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
Discord: https://ykilcher.com/discord
LinkedIn: https://www.linkedin.com/in/ykilcher
If you want to support me, the best thing to do is to share out the content :)
If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this):
SubscribeStar: https://www.subscribestar...
