modelscope / modelscope/ms-swift
Does Swift support expanding not-special tokens
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 15.7k
- Forks
- 1.7k
- Avg merge
- 1d 16h
- Merged PRs (30d)
- 136
Description
The llama-factory supports an add_tokens parameter, allowing users to customize tokens without requiring model conversion for training.So I'd like to ask if Swift supports, I only found the new_special_token parameter.
Contributor guide
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 by comparing ms-swift's existing new_special_token handling with llama-factory's add_tokens parameter to determine whether non-special token expansion is supported. Trace the relevant training and tokenizer paths, then define the expected behavior and validation for supporting customized non-special tokens.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100