Gary Mulder
Gary Mulder
> I was Successfully able to build the model when was not in a virtual machine. But Now I am wondering were I can download the LLAMA model. https://github.com/facebookresearch/llama/issues/92
I can neither confirm nor deny that that link will work :smile: For verification, and as there are some suspect models floating around, I published the md5 sums and file...
Something weird here. What am I doing wrong? ``` $ cat /data/llama/7B/params.json;echo {"dim": 4096, "multiple_of": 256, "n_heads": 32, "n_layers": 32, "norm_eps": 1e-06, "vocab_size": -1} $ docker run -v models:/models llamacpp-converter...
> I believe you’d need to run `docker run -v /data/llama:/models llamacpp-converter "/models/7B" 1` That works. Thx.
Where's the quantization step occurring? Logically this should occur in the `tools` Dockerfile, which implies running `make` there too and having a wrapper script to call first `convert-pth-to-ggml.py` and then...
That's normal behaviour currently. People are working on chat interfaces and Python bindings.
Because it is not ChatGPT. The command line interface was provided as a simple way to get the model up and running. Note that there is an interactive mode, but...
> Using your font in a banner with my icon: > >  Perhaps shorten to "Make your LLaMAs run at speed C++"? (triple pun)
See my comments in issue #53 for md5 signatures of *.pth and *.bin files to verify your torrent downloads are legitimate and correct.