Edoardo Procino

Hi write here since the pull-request page didn't work well. Hi, my name is Edoardo (from Italy!) and I used the adapter in my master thesis. The bug generated when...

Hi, Yes sorry, it was my first pull request and I had problems with GitHub since the page of the pull request lagged a lot. If you want I can...

I did inference on multiple GPUs on vast.ai and I didn't have problems. The only things I did was to give privileges to dist_test.sh. You can try either _chmod +x...

Same error, I opened an issue too. It seems that also the [example notebooks on colab](https://colab.research.google.com/github/unslothai/notebooks/blob/main/nb/Gemma2_(9B)-Alpaca.ipynb#scrollTo=QmUBVEnvCDJv) do not work. I tried vast, colab and to install in various ways (also...

Hi @Sweaterdog! On colab I think that you can try with model.save_pretrained_gguf( save_directory = "folder_name", quantization_type='Q8_0') (add repo id and token to push on HF)

@Sweaterdog Yesterday I wrote on discord to one of the developers. He said that they are working to implement into unsloth the new classes of llama.cpp I guess we have...

@shimmyshimmer Yes I did, but the val loss is still "No Log"

Follow up For apparently no reasons now the working code doesn't work anymore, I didn't change anything but now trainer.train() returns this error AssertionError Traceback (most recent call last) [/usr/local/lib/python3.11/dist-packages/torch/_dynamo/output_graph.py](https://localhost:8080/#)...