llama.cpp: Convert consolidated pth files to gguf / missing script?

I’m trying to run llama2 models on a mac M1 pro max

• I’ve downloaded the llama2 models with the link that was provided by email
• I’ve downloaded llama.cpp & compiled with make
• If I understand correctly llama.cpp needs the models to be converted to gguf (ggml?) format
• I see in the models only consolidated.00.pth files
• I’m looking for the script convert-pth-to-gguf.py but it’s not there

what am I missing?

I guess after this little step I can run make -j && ./main -m models/llama-13b-v2/ggml-model-q4_0.gguf -p "Building a website can be done in 10 simple steps:\nStep 1:" -n 400 -e

Version:

Python 3.11.6
GNU Make 3.81
Apple clang version 15.0.0 (clang-1500.0.40.1)

When i try to run:

python3 -m venv .env
source .env/bin/activate
pip install -r requirements.txt

python convert.py ../llama/llama-2-7b-chat

I get:

Loading model file ../llama/llama-2-7b-chat/consolidated.00.pth
Traceback (most recent call last):
  File "/Users/alex/dev/llama.cpp/convert.py", line 1207, in <module>
    main()
  File "/Users/alex/dev/llama.cpp/convert.py", line 1140, in main
    model_plus = load_some_model(args.model)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/alex/dev/llama.cpp/convert.py", line 1055, in load_some_model
    models_plus.append(lazy_load_file(path))
                       ^^^^^^^^^^^^^^^^^^^^
  File "/Users/alex/dev/llama.cpp/convert.py", line 740, in lazy_load_file
    raise ValueError(f"unknown format: {path}")
ValueError: unknown format: ../llama/llama-2-7b-chat/consolidated.00.pth