Javier de la Rosa
Javier de la Rosa
I just tried to compile it myself, to see it I was able to also get the `stream` binary to work. But after I apply the patch, the `make` command...
After some tweaking, I was able to compile my own cosmocc and then use it to compile `main`, `quantize` and even `server` 🎉 . However, for `stream` there seems to...
Hi! I've trained GPT-J models in the past, but for some reason I'm now seeing this too. Did you manage to solve it?
@rinapch for me the key was to select the alpha version when creating the TPU. Stable releases seem to break the implementation, not sure why.
I had the same issue in Linux and as suggested by @markusgod, running the [next code](https://github.com/r4victor/afaligner/issues/2#issuecomment-1038149281) before invoking `syncabook` made it work for me. ```bash $ echo 1 | sudo...
I was looking at the outputs of pico and they look great, but was wondering how to add other languages and then found this issue. Any luck at all adding...
Are there plans for an official conversion script that also supports RoBERTa models as the text encoder?
I had the same issue in Linux and running the code suggested by @bhattarai333 made it work for me. ```bash $ echo 1 | sudo tee /proc/sys/vm/overcommit_memory ```
I also faced the same issue uploading a big model with [`multi_commit`](https://huggingface.co/mlx-community/zephyr-orpo-141b-A35b-v0.1-8bit/discussions/1), but it seems the discussion branches where the PRs are created are not directly accessible, so resuming/fixing the...
Hey! Did you find why it failed?