Developing an LLM: Building, Training, Finetuning
REFERENCES:
1. Build an LLM from Scratch book: https://mng.bz/M96o
2. Build an LLM from Scratch repo: https://github.com/rasbt/LLMs-from-scratch
3. Slides: https://sebastianraschka.com/pdf/slides/2024-build-llms.pdf
4. LitGPT: https://github.com/Lightning-AI/litgpt
5. TinyLlama pretraining: https://lightning.ai/lightning-ai/studios/pretrain-llms-tinyllama-1-1b
DESCRIPTION:
This video provides an overview of the three stages of developing an LLM: Building, Training, and Finetuning. The focus is on explaining how LLMs work by describing how each step works.
OUTLINE:
00:00 – Using LLMs
02:50 – The stages of developing an LLM
05:26 – The dataset
10:15 – Generating multi-word outputs
12:30 – Tokenization
15:35 – Pretraining datasets
21:53 – LLM architecture
27:20 – Pretraining
35:21 – Classification finetuning
39:48 – Instruction finetuning
43:06 – Preference finetuning
46:04 – Evaluating LLMs
53:59 – Pretraining & finetuning rules of thumb