Apple Intelligence is exciting and in their latest technical report, they talk about Apple’s Foundation Large Language models ( or LLM ), namely AFM-server and AFM-on-device. In this video, I break down all the algorithmic progress made in this paper – explaining concepts like Next word Prediction with Transformer Decoders, Reinforcement Learning with Human Feedback, Low Rank Adaptation (LoRA), Knowledge Distillation, Structured Pruning, Quantization with Palettization, Mirror Descent Policy Optimization with Leave One Out (MDLOO), and many more concepts!
#ai #apple #artificialintelligence
Youtube Members and Patrons will get access to write-ups, slides, notebooks, and bonus content from all videos on my channel!
Buy me a coffee at https://ko-fi.com/neuralavb !
Visit my Patreon link to see what else is available:
https://www.patreon.com/NeuralBreakdownwithAVB
Links:
Apple Paper: https://arxiv.org/pdf/2407.21075
Apple Blogpost: https://machinelearning.apple.com/research/introducing-apple-foundation-models
Palettization: https://apple.github.io/coremltools/docs-guides/source/opt-palettization-overview.html
Sheared LLama: https://arxiv.org/pdf/2310.06694
Structured Pruning: https://arxiv.org/pdf/1910.04732
Videos you may like:
The Full History of NLP Explained – https://youtu.be/uocYQH0cWTs
Attention to Transformers Playlist – https://www.youtube.com/playlist?list=PLGXWtN1HUjPfq0MSqD5dX8V7Gx5ow4QYW
Timestamps:
0:00 – Intro
1:32 – Chapter 1 – Overview
2:55 – Pretraining
4:12 – Structured Pruning
5:12 – Knowledge Distillation
5:55 – Post Training
7:36 – Iterative Teaching Committee
9:06 – Chapter 2 – Adapters
9:30 – LoRA (Low Rank Adapters)
11:55 – Quantization (Palettization)
13:27 – Chapter 3 – RLHF
14:11 – Reward Modelling
16:03 – Leave One Out
17:43 – Mirror Descent Policy and MDLOO
19:15 – Results
Note: Neural Breakdown with AVB is the original author of this video, we just embed it, if you have any questions please contact him via Youtube.
Amazing video with many important basic concepts all compressed into a short video. Very nice format.
Was not expecting you to explain all of LoRA, Quantization, RLHF, PPO, Distillation using Teaching Committee, Structured Pruning, etc all in this seemingly random video about Apple Intelligence
wonderful video
love your content!
💙
can we connect somewhere to have a chat?
Can I connect multiple lora adapters at the same time to the base model?😊