[example] add llama2 example (#4527)

* [example] transfer llama-1 example

* [example] fit llama-2

* [example] refactor scripts folder

* [example] fit new gemini plugin

* [cli] fix multinode runner

* [example] fit gemini optim checkpoint

* [example] refactor scripts

* [example] update requirements

* [example] update requirements

* [example] rename llama to llama2

* [example] update readme and pretrain script

* [example] refactor scripts
This commit is contained in:
Hongxin Liu
2023-08-28 17:59:11 +08:00
committed by GitHub
parent 839847b7d7
commit 0b00def881
17 changed files with 1087 additions and 19 deletions

View File

@@ -1,11 +0,0 @@
# Pretraining LLaMA: best practices for building LLaMA-like base models
<p id="ColossalChat-Speed" align="center">
<img src="https://raw.githubusercontent.com/hpcaitech/public_assets/main/examples/images/LLaMA_pretraining.png" width=600/>
</p>
- 65-billion-parameter large model pretraining accelerated by 38%
[[code]](https://github.com/hpcaitech/ColossalAI/tree/example/llama/examples/language/llama)
[[blog]](https://www.hpc-ai.tech/blog/large-model-pretraining)
> Since the main branch is being updated, in order to maintain the stability of the code, this example is temporarily kept as an [independent branch](https://github.com/hpcaitech/ColossalAI/tree/example/llama/examples/language/llama).