Find answers from the community

Updated 3 months ago

intfloat/e5-mistral-7b-instruct · Huggin...

was looking at the mtbe benchmark and saw https://huggingface.co/intfloat/e5-mistral-7b-instruct
as number one... its an llm
https://huggingface.co/spaces/mteb/leaderboard
Can we use the fine-repo to fine-tune mistral?

3 comments

lol yeaaa... I would not use an LLM as an embedding model.

We don't have any support right now for fine-tuning LLMs to be embedding models

7B parameters and it just barely beats other models lol

good idea

Add a reply