Cyclotomic Fields comments

Results 15 comments of


                                            Cyclotomic Fields

[Bug]: Not able to do lora inference with phi-3

@Raibows thanks for your helpful python script! May I ask another question? I want to use Ollama with a fine tuned Phi3 model (using QLoRA), and now I have succeed...

Looking for help for using llama.cpp with Phi3 model and LoRA

I find that the structure of Phi2 and Phi3 are named in different way, in Phi2, llama.cpp works fine on transforming LoRA weights to GGML (where the layer is named...

Looking for help for using llama.cpp with Phi3 model and LoRA

hope this will be fixed as soon as possible

Looking for help for using llama.cpp with Phi3 model and LoRA

The reason is that llama.cpp treats phi3 as llama architecture, i.e., splitting the merged qkv_proj into q_proj, k_proj and v_proj layers. One way posted by @Raibows at https://github.com/vllm-project/vllm/issues/4715 is to...

Looking for help for using llama.cpp with Phi3 model and LoRA

anyone?

[BUG]cannot pip install on linux (Ubuntu)

I am so grateful! It works! > It looks like this is trying to install 23.12, an old release. Can you try to install the latest release? We have improved...

[BUG]cannot pip install on linux (Ubuntu)

Another question, I wonder how to install and use desk_cudf, I find that repository has been moved to CUDF.

Looking for SQL ORM for FastAPI

Thanks guys on your valuable sharing! The motivation for this question is that I don't know whether SQLAlchemy is good at asyncio 😂

Looking for SQL ORM for FastAPI

Thanks guys!

Solution for 'FlashAttention only supports Ampere GPUs or newer' Error on V100 GPUs

> Hi, > > I encountered an issue while trying to run the `InternVL2-8B` model on an NVIDIA V100 GPU, where I received the error `FlashAttention only supports Ampere GPUs...