Lightning-AI / Lightning-AI/litgpt
LitGPT fine-tuning Dont Use GPU
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 13.7k
- Forks
- 1.5k
- Avg merge
- 15h 37m
- Merged PRs (30d)
- 1
Description
This is the first time I have used LitGPT for fine-tuning, and as an experiment, I chose Qwen2.5-0.5B for fine-tuning.
I am Runing this on windows server 2022dc
(.env) f:\litgpt>litgpt finetune Qwen/Qwen2.5-0.5B-Instruct --data JSON --data.json_path F:\litgpt\alpaca_zh_demo.json --data.val_split_fraction 0.1 --out_dir f:\out
I found that the whole process was done using the CPU, not my GPU
I confirm that cuda12.2 is installed correctly and I can use ML Studio to use the GGUF format model.
Did I miss any settings?
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 reproducing the exact `litgpt finetune` command on Windows Server 2022 with Qwen/Qwen2.5-0.5B-Instruct and the supplied dataset. Check whether the environment and CUDA setup allow the command to select the GPU; done means identifying the cause and documenting a reproducible configuration or fix.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100