Junghwan Park

A Little guide to building Large Language Models in 2024

This is a post-recording of a 75min lecture I gave two weeks ago on how to train a LLM from scratch in 2024. I tried to keep it short and comprehensive – focusing on concepts that are crucial for training good LLM but often hidden in tech reports.

In the lecture, I introduce the students to all the important concepts/tools/techniques for training good performance LLM:
* finding, preparing and evaluating web scale data
* understanding model parallelism and efficient training
* fine-tuning/aligning models
* fast inference

There is of course many things and details missing and that I should have added to it, don't hesitate to tell me you're most frustrating omission and I'll add it in a future part. In particular I think I'll add more focus on how to filter topics well and extensively and maybe more practical anecdotes and details.

Now that I recorded it I've been thinking this could be part 1 of a two-parts series with a 2nd fully hands-on video on how to run all these steps with some libraries and recipes we've released recently at HF around LLM training (and could be easily adapted to your other framework anyway):
*datatrove for all things web-scale data preparation: https://github.com/huggingface/datatrove
*nanotron for lightweight 4D parallelism LLM training: https://github.com/huggingface/nanotron
*lighteval for in-training fast parallel LLM evaluations: https://github.com/huggingface/lighteval

Here is the link to watch the lecture on Youtube: https://www.youtube.com/watch?v=2-SPH9hIKT8
And here is the link to the Google slides: https://docs.google.com/presentation/d/1IkzESdOwdmwvPxIELYJi8--K3EZ98_cL6c5ZcLKSyVg/edit#slide=id.p

Enjoy and happy to hear feedback on it and what to add, correct, extend in a second part.

2 replies

liked a dataset over 2 years ago

MBZUAI/VideoInstruct-100K

Viewer • Updated Sep 29, 2023 • 100k • 282 • 49

liked a model over 2 years ago

vdo/Video-LLaMA-Series

Visual Question Answering • Updated Jun 14, 2023 • 10

reacted to wvaneaton's post with 👍 over 2 years ago

Post

We fine-tuned 25 mistral-7b models that outperform GPT-4 on task-specific use cases!

Check them out at LoRA Land: https://pbase.ai/3uFh7Qc

You can prompt them and compare their results to mistral-7b-instruct in real time!

They're also on HF for you to play with:

predibase

Let us know what you think!

upvoted a paper about 3 years ago

On the Origin of LLMs: An Evolutionary Tree and Graph for 15,821 Large Language Models

Paper • 2307.09793 • Published Jul 19, 2023 • 48

liked 5 models about 3 years ago

Junghwan Park

AI & ML interests

Recent Activity

Organizations

9bow's activity

Open Perflexity V2

Open VLM Leaderboard