Do unsloth work on non lora case?

[rangehow]

I have tried unsloth+qlora, it's cool and brings considerable speedup and vram reduction. But I am not sure if this repo is useful for full fine-tune after searching repo/website/benchmark. So can someone familiar with unsloth can grant a suggestion about this? Thanks for your wonderful work : )

[danielhanchen]

Sadly currently full finetuning isnt yet supported - some Unsloth community members have tried doing it, and it does converge, albeit the layernorms are not trained