Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

How can i do continue pretraining? #47

Open
hwaking opened this issue Mar 18, 2024 · 1 comment
Open

How can i do continue pretraining? #47

hwaking opened this issue Mar 18, 2024 · 1 comment

Comments

@hwaking
Copy link

hwaking commented Mar 18, 2024

can you share the continue pretraining scripts and sample data?

@GoneZ5
Copy link
Collaborator

GoneZ5 commented Mar 19, 2025

We apologize for the delayed response. To address your issue, please follow these steps:

  1. First, use the run_offline_tokenization.sh script to tokenize your data.
  2. Then, make the following modifications in the full_train_config.json file:
    • "data_paths": "[/workspace/data/data1,/workspace/data/data2]"
    • "data_weights": "[1.,1.]"
    • "tokenize_mode": "sst"
  3. In the ds_multinode_launch.sh file, modify the entry script from pefts/mft_accelerate.py to mpt/mpt_accelerate.py.

Let us know if you need any further clarification or have additional questions.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants