Matthew Gunton – Medium

Matthew Gunton

Matthew Gunton
in
Towards Data Science

How to Improve Model Quality Without Building Larger Models

Going into the Google DeepMind’s “Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters”

5d ago

How to Improve Model Quality Without Building Larger Models

5d ago

Matthew Gunton
in
Towards Data Science

Line-By-Line, Let’s Reproduce GPT-2: Section 3 — Training

This blog post will go line-by-line through the code in Section 3 of Andrej Karpathy’s “Let’s reproduce GPT-2 (124M)”

Sep 3

Line-By-Line, Let’s Reproduce GPT-2: Section 3 — Training

Sep 3

Matthew Gunton
in
Towards Data Science

Line-By-Line, Let’s Reproduce GPT-2: Section 2 — Hardware Optimization

This blog post will go line-by-line through the hardware optimizations in Section 2 of Andrej Karpathy’s “Let’s reproduce GPT-2 (124M)”

Jul 31

Line-By-Line, Let’s Reproduce GPT-2: Section 2 — Hardware Optimization

Jul 31

Matthew Gunton
in
Towards Data Science

Line By Line, Let’s Reproduce GPT-2: Section 1

This blog post will go line-by-line through the code in Section 1 of Andrej Karpathy’s “Let’s reproduce GPT-2 (124M)”

Jul 23

Line By Line, Let’s Reproduce GPT-2: Section 1

Jul 23

Matthew Gunton
in
Towards Data Science

Exploring Medusa and Multi-Token Prediction

This blog post will go into detail on the “MEDUSA: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads” paper

Jul 10

Exploring Medusa and Multi-Token Prediction

Jul 10

Matthew Gunton
in
Towards Data Science

Diving Deep into AutoGen and Agentic Frameworks

This blog post will go into the details of the “AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation” paper

Jun 28

Diving Deep into AutoGen and Agentic Frameworks

Jun 28

Matthew Gunton
in
Towards AI

Understanding Mamba and Selective State Space Models (SSMs)

This blog post will go in detail on the “Mamba: Linear-Time Sequence Modeling with Selective State Spaces” paper

Jun 24

Understanding Mamba and Selective State Space Models (SSMs)

Jun 24

Matthew Gunton
in
Towards Data Science

Understanding You Only Cache Once

This blog post will go in detail on the “You Only Cache Once: Decoder-Decoder Architectures for Language Models” Paper and its findings

Jun 4

Understanding You Only Cache Once

Jun 4

Matthew Gunton
in
Towards Data Science

Understanding Low Rank Adaptation (LoRA) in Fine Tuning LLMs

How LoRA works to fine-tune LLMs, following the methodology set out in the “LoRA: Low-Rank Adaptation of Large Language Models” paper

May 24

Understanding Low Rank Adaptation (LoRA) in Fine Tuning LLMs

May 24

Matthew Gunton
in
Towards Data Science

Understanding Long RoPE in LLMs

This blog post will go in detail about the new Long RoPE Methodology used to expand the context lengths LLMs can support without…

May 15

Understanding Long RoPE in LLMs

May 15

Matthew Gunton

Matthew Gunton

Hi, my name is Matthew Gunton. I like talking about the latest ways technology can be used to improve the world

Following

Help
Status
About
Careers
Press
Blog
Privacy
Terms
Text to speech
Teams