[User] I encountered a problem ..

[vivekhub]

Prerequisites

Please answer the following questions for yourself before submitting an issue.

• I am running the latest code. Development is very rapid so there are no tagged versions as of now.
• I carefully followed the README.md.
• I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
• I reviewed the Discussions, and have a new bug or useful enhancement to share.

Expected Behavior

Trying to run the model converter tool using

{'dim': 4096, 'multiple_of': 256, 'n_heads': 32, 'n_layers': 32, 'norm_eps': 1e-05, 'vocab_size': -1}
Namespace(dir_model='../llama.cpp/models/llama-2-7b/', ftype=1, vocab_only=0)
n_parts = 1

Processing part 0

Processing variable: tok_embeddings.weight with shape: torch.Size([32000, 4096]) and type: torch.bfloat16
Traceback (most recent call last):
  File "/Users/vivekv/software/llama-go/./convert-pth-to-ggml.py", line 181, in <module>
    main()
  File "/Users/vivekv/software/llama-go/./convert-pth-to-ggml.py", line 174, in main
    process_and_write_variables(fout, model, ftype)
  File "/Users/vivekv/software/llama-go/./convert-pth-to-ggml.py", line 109, in process_and_write_variables
    data = datao.numpy().squeeze()
           ^^^^^^^^^^^^^
TypeError: Got unsupported ScalarType BFloat16

• Physical (or virtual) hardware you are using, e.g. for Linux:
  running on an M1 Mac Pro

$ python3 --version

Python 3.11.4


$ make --version

GNU Make 3.81
Copyright (C) 2006  Free Software Foundation, Inc.
This is free software; see the source for copying conditions.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE.

This program built for i386-apple-darwin11.3.0

$ g++ --version

Apple clang version 14.0.3 (clang-1403.0.22.14.1)
Target: arm64-apple-darwin22.4.0
Thread model: posix
InstalledDir: /Library/Developer/CommandLineTools/usr/bin

[quanpinjie]

The same error, how can it be resolved?

[rahulnetsmartz]

On a Macbook Pro M2, getting the same issue. Any resolution?