modelscope / modelscope/ms-swift
模型蒸馏时的采样温度
Open
Nobody has claimed this yet.
stale
- Dominant language
- Python
- Stars
- 15.7k
- Forks
- 1.7k
- Avg merge
- 1d 16h
- Merged PRs (30d)
- 136
Description
请问，如果我要用教师模型去教小模型输出json的话，采样的温度设为0可以吗，还是默认值0.9呢，因为他内部还要算logits所以我感觉温度为0太极端了，但是任务是输出正确的json格式，温度高了加大了采样错误的可能性。该怎么平衡呢
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
No file, test, or entry point is named. First determine whether this question should become documentation or a concrete implementation request, then identify the relevant distillation and generation documentation. Done means providing a project-specific, reproducible recommendation for temperature settings and JSON-output validation.
Written by the indexing model from the issue text.
Assessment
- Domain
- machine-learning
- Issue type
- Documentation
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 15/100