✳flâneur — a map of the web's best reading
Tiny TPU
tinytpu.com · 9,673 words · saved by 10 readers
An attempt to understand and build a TPU—by complete novices.
Tiny TPU tinytpu.com Tiny-TPU: the why and how Aug 17th 2025 · Xander Chin, Kenny Guo, Evan Lin, Surya Sure Github Twitter Google Drive Nobody really understands how TPUs work…and neither do we! So we wanted to make this because we wanted to take a shot and try to guess how it works–from the perspective of complete novices! This slideshow depicts the forward pass and backward pass on our toy accelerator. clk 0 Why did we start this project? We wanted to do something very challenging to prove to ourselves that we can do anything we put our mind to. The reasoning for why we chose to build a TPU
Explore this link on the map →saved by
- abinaya dinesh
- Hardeep Gambhir
- Jianmin Chen
- Laerdon Kim
- Mackay G
- Jackson Mowatt Gok
- Eva L
- Caden Juang
- 6774
- Tom Zheng
related reading
- How to Think About TPUs | How To Scale Your Modeljax-ml.github.io
- Understanding Matrix Multiplication on a Weight-Stationary Systolic Architecture | Telesenstelesens.co
- TPU Deep Divehenryhmko.github.io
- How To Scale Your Modeljax-ml.github.io
- Touching the Elephant - TPUs | Consider the Bulldogconsiderthebulldog.com
- The Little Book of Deep Learningfleuret.org
- Inside NVIDIA GPUs: Anatomy of high performance matmul kernels - Aleksa Gordićaleksagordic.com
- Transformer Inference Arithmetic | kipply's blogkipp.ly
- Introduction to Cloud TPU | Google Cloud Documentationcloud.google.com
- All About Rooflines | How To Scale Your Modeljax-ml.github.io
- microgptkarpathy.github.io
- Neural networks and deep learningneuralnetworksanddeeplearning.com