Hello, I am trying to use llama.cpp to convert my fine-tuned unsloth/DeepSeek-R1-Distill-Qwen-7B model to gguf format. However, when doing so, I encountered the same issues as during fine-tuning—I didn't get the same responses, and sometimes the output even mixed in another language. I used the following code:

python convert_hf_to_gguf.py ./DeepSeek-R1-Medical-COT-Qwen-7B --outfile ./DeepSeek-R1-Medical-COT-Qwen-7B/DeepSeek-R1-Medical-COT-Qwen-7B.gguf --outtype f16

Am I possibly using the wrong Python file? Alternatively, is there any way to verify that the conversion to gguf has been done correctly?