Distilling Knowledge into Tiny LLMs

Описание: Finetune tiny LLMs to enable inference using less resources.

Link to notebook: https://github.com/neuml/txtai/blob/m...

00:00 Intro
00:59 Base LLM
02:09 Build finetuning dataset
03:16 Finetune model
04:36 Evaluate model
06:09 Wrapping up