CarperAI / CarperAI/trlx

How to train LLaMA2 on the summarize_rlhf example?

Open
#559 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
4.8k
Forks
487
PR merge metrics
No merged PRs in 30d

Description

How to train LLaMA2 on the summarize_rlhf example?

I tried to modify the original code [1] and a different notebook [2], but failed due to an out of memory error.
[1] https://github.com/CarperAI/trlx/blob/main/examples/summarize_rlhf/trlx_gptj_text_summarization.py
[2] https://github.com/HumanSignal/RLHF/blob/master/tutorials/RLHF_with_Custom_Datasets.ipynb

Do you have any success stories with LLaMA2?

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.