fine-tune an LLM with chat logs

[mfreeman451]

Need to fine-tune an LLM with a chat log and use it in our QA sequence to see if that can help ground the LLM when performing question-answering.

https://docs.predibase.com/user-guide/getting-started/quickstart

Fine-tuning jobs:

• [ ] ??

Serving:

• [ ] Predibase/lorax
• [ ] #67