Anchen
Anchen
I think, Inline style always applies based on selection of editorstate.you can try to toggle block style see if it works
@Marais I think you can try the example below to create your custom content block, then add custom render logic in blockRendererFn to handle your custom contentblock as make it...
@mkozhukharenko :sob: Sorry about that, I updated the example..
Hi, I would like to work on this if you guys don't mind.
Qlora works with the quantized lm_head, but the issue is to merge it back into the original model. The model's performance is a lot worse after dequantizing the lm_head.
> > Qlora works with the quantized lm_head, but the issue is to merge it back into the original model. The model's performance is a lot worse after dequantizing the...
> > Qlora works with the quantized lm_head, but the issue is to merge it back into the original model. The model's performance is a lot worse after dequantizing the...
Yeah, I don't think the performance issue is caused by Lora. Personally, I feel that it's more like after using Lora, if we want to merge it back into the...
> Ok I think it's clear now, but just to be sure: > > 1. Download https://huggingface.co/mlx-community/Mistral-7B-v0.1-hf-4bit-mlx > 2. Update the LoRA script to be able to load the quantized...
Nice work! Just a quick question: for the best fine-tune result, would it be possible to provide a flag that allows people to apply Lora to all linear layers (except...