make the training process async

update README.md

Merge branch 'dev-2' of github.com:zetavg/llama-lora into dev-2

resolves #2 numpy version mismatch on colab

add wandb support

update README.md

update README.md

fix

update README.md

Merge branch 'dev-2'

rename project

update README.md

latest peft makes model.save_pretrained in finetune.py save a 443 B adapter_model.bin which is clearly incorrect (normally adapter_model.bin should > 16 MB)

Update README.md

update README.md

update README.md and LLaMA_LoRA.ipynb

update