Open 12dc32d opened 3 months ago
same with you
If you could add more information about your problem, I may be able to help - I had a 'couldn't find path' issue today I had to debug
If you could add more information about your problem, I may be able to help - I had a 'couldn't find path' issue today I had to debug
Thank you bro. I already find the Original file and here has tokenizer model and checkpoint file. But the one more question is my tablet uses Intel integrated graphics, it can not support cuda (only nvdia can) driver. I downloaded the intel graphics driver but I don't know how to modify the code to make the llama3 model calculate on the intel graphics card. Every time it shows that cuda is required. May share any ideas as you want?
same with you
You may click and download the first file. It is original file include tokenizer_model, other 2 files are checkpoint_model.
If you could add more information about your problem, I may be able to help - I had a 'couldn't find path' issue today I had to debug
Thank you bro. I already find the Original file and here has tokenizer model and checkpoint file. But the one more question is my tablet uses Intel integrated graphics, it can not support cuda (only nvdia can) driver. I downloaded the intel graphics driver but I don't know how to modify the code to make the llama3 model calculate on the intel graphics card. Every time it shows that cuda is required. May share any ideas as you want?
This link may be helpful, and could adapt to this repository...unfortunately I'm not an expert with this stuff lol. With integrated graphics though, you may want to explore solutions like llama.cpp or ollama, as they are better at adapting to hardware, especially non-specialist graphics units.
Okay....thanks. My Gpu is Intel(R) UHD Graphics, this is an iGPU, all of videos resource about Intel gpu are dGPU (The famous one is Intel Arc). The Intel Arc series is Intel's recently launched high-performance graphics card series, designed to compete with NVIDIA and AMD's discrete graphics cards. Actually, what I want to ask is how to modify the llama3 code so that it can run on Intel GPU. Although I have disabled cuda on PyCharm using the statement [os.environ["CUDA_VISIBLE_DEVICES"] = ""], every time running, it prompts me that cuda is not available, instead of considering using Intel GPU. If you have solved similar problems before or have any ideas, please have a pleasant communication.
Okay....thanks. My Gpu is Intel(R) UHD Graphics, this is an iGPU, all of videos resource about Intel gpu are dGPU (The famous one is Intel Arc). The Intel Arc series is Intel's recently launched high-performance graphics card series, designed to compete with NVIDIA and AMD's discrete graphics cards. Actually, what I want to ask is how to modify the llama3 code so that it can run on Intel GPU. Although I have disabled cuda on PyCharm using the statement [os.environ["CUDA_VISIBLE_DEVICES"] = ""], every time running, it prompts me that cuda is not available, instead of considering using Intel GPU. If you have solved similar problems before or have any ideas, please have a pleasant communication.
Before submitting a bug, please make sure the issue hasn't been already addressed by searching through the FAQs and existing/past issues
Describe the bug
<Please provide a clear and concise description of what the bug is. If relevant, please include a minimal (least lines of code necessary) reproducible (running this will give us the same result as you get) code snippet. Make sure to include the relevant imports.>
Minimal reproducible example
<Remember to wrap the code in
```triple-quotes blocks```
>Output
<Remember to wrap the output in
```triple-quotes blocks```
>Runtime Environment
meta-llama-3-8b-instruct
]Additional context Add any other context about the problem or environment here.