“Fine-Tuning vs RAG

(Retrieval Augmented Generation)

I really enjoyed this podcast. Guest Demetrios of ML Ops took shots at my current favourite target, LLM Fine Tuning.

He discussed the use of fine tuning and it’s purpose versus how it is being used. He talked about over engineering and I took that as some vindication for my beliefs about LLM modelling being re engineered to achieve desired results.

I see Fine tuning as a means to measure such elements as cost effectiveness and computing needs for bespoke LLM models.

More to come on this.

– –

“Fine-Tuning vs RAG (Retrieval Augmented Generation) with Demetrios Brinkmann of the MLOps Community (Practical AI #238),” September 6, 2023. https://changelog.com/practicalai/238.

Demetrios Brinkmann – Twitter

Chris Benson – Twitter, GitHub, LinkedIn, Website

Daniel Whitenack – Twitter, GitHub, Website