lora or sft?

#4
by yuimo - opened

"This model is based on google/gemma-7b and hase been LoRA finetuned on 15 Indian languages and English language instruction datasets:"
the document above indicates that model is trained using lora, but in "files and versions", i can not find the lora weights.
so is this trained by lora or sft?

Sign up or log in to comment