lemonilia

/

Limamono-Mistral-7B-v0.50

like 24

Merging the provided LoRA adapter to Mistral-7B-v0.1 should allow users to obtain HF FP16 model weights for Limamono. I don't generally upload FP16 HF weights due to to my limited upload bandwidth.

Here is a script that I used in the past for merging the LoRA adapter to the base model: https://github.com/artidoro/qlora/blob/d1765faa571a5c856bac34f02535544ce19d7b81/merge_peft.py