florian peyron

Results 9 comments of florian peyron

From your replies do I understand correctly that the merge doesn't add the llava features to the model ? Here are the steps I followed: _(from this and root readme)...

Hi, thanks for your reply. I have tried to follow the steps, but my folders does not match the ones from the examples, using the Qlora finetune config. Indeed, in...

Thanks, this worked well for me. I have a question however, the script read ``` freeze_llm=True, freeze_visual_encoder=True, ``` Why, if the llm is frozen, do I need to merge a...

Thanks very much for your time, this is very clear.

I am also interested in documentation about this. My use case is to add a way for the user to insert custom html tags in the input, so I want...

After digging through the code it seems repo path meant /organization/repoID, but never local path. I saw I can use the --cache option to save the models locally. But this...

I get the same error from a macbook M2, I am currently trying to find a commit where the build work as I was able to build it in the...

EDIT: I do not get this error when building normaly, but when using `docker buildx build --platform linux/amd64 -t llm-amd64 . `

I have the same error that triggered when playing with support gems on my build, I can't start the app now as it automatically try to resume my build I...