search

formentor-studio / magnolia-ai-training

Fine tuning of LLM models using contents in Magnolia CMS
0 stars 1 forks source link

legacy fine-tuning #1

Open rdlfperezIII opened 3 weeks ago

rdlfperezIII commented 3 weeks ago

Hello, I've noticed that the ai-training module JSONL format being sent is for legacy fine tuning models such as babbage-002and davinci-002. Will there be an OOB support for the current fine-tuning method ?

Thanks in advance

joaquin-alfaro commented 3 weeks ago

Thanks for your suggestion, yes we should support both legacy and new models of OpenAI. If you have any proposal of implementation, feel free to open a pull request.

Thanks @rdlfperezIII