Find answers from the community

Updated last year

Hey guys hugely important for not

Hey guys, hugely important for not wasting money, maybe I lost it, but when i fine tune an embedding model following the llama index guide, how can I save it? It’s my first fine tuning so pretty ignorant. Thx in advance!

4 comments

LLogan M

This is the full embedding fine-tuning right, not the adaptor?

You can just point the embed model to the output path, it should be already saved to the output path you gave

Plain Text

service_context = ServiceContext.from_defaults(embed_model="local:/path/to/output")

AAndreaSel93

Thx @Logan M ! This is for the usage.
But suppose i’m in colab and want to close my sessions. Then when I open colab again I think i lose the FT model no?
I wanted to know how to save it somewhere in order to be able to use it again in the future!

LLogan M

Yea, you can right click and download the model folder from the file explorer in colab

LLogan M

Attachment

Add a reply