llama-lora-fine-tuning icon indicating copy to clipboard operation
llama-lora-fine-tuning copied to clipboard

Could you please provide the code for merging the generated output file into the original model?

Open codezealot opened this issue 2 years ago • 1 comments

Very good project, I wonder if you can provide the code for merging the generated output file into the original model?

codezealot avatar Jun 18 '23 01:06 codezealot

It's same with fastchat,example:

python3 -m fastchat.model.apply_delta
--base-model-path /path/to/llama-7b
--target-model-path /path/to/checkpoints/...
--delta-path model/output

It should be noted that the final training is the PEFT model, and the model that needs to be merged needs to be found from the checkpoints

little51 avatar Jun 18 '23 14:06 little51