Do I still need to merge the llama7b model and vicuna7b model after I have fully tuned them? Why is there a problem of garbled code after training, whether merged or not?
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 39.5k
- Forks
- 4.8k
- PR merge metrics
- No merged PRs in 30d
Description
Hello everyone, I'm going to fine-tune the llama7b and vicuna7b models with full parameters. Do you still need to merge them after training? Why is there a problem of garbled code after training, whether merged or not?
Using the official sample data and the original train_ Vicuna_ 7b. sh goes to train the llama model, and after merging and testing, there will also be garbled code. May I ask what is the situation with all the experts?
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the official sample data and the original train_Vicuna_7b.sh script, then reproduce the reported output with and without merging the llama7b and vicuna7b models. Done means documenting whether merging is required after full-parameter tuning and identifying the cause of the garbled output.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python, shell
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100