All New Mistakes
Ganesh RavichandranI'm fairly alarmed here. Previous founding engineer @ SOOT. Traveled around the world looking for the most beautiful hike. Writes about scaling AI infrastructure and mechanistic interpretability.
| Platform |  | Pricing | Only free issues | Publishes | Infrequently |
|---|---|---|---|---|---|
| Issues | 6 | Founded | 3 years ago | Last Issue | 9 months ago |
| Active | |||||
Read this Newsletter
thebatmanofbutler.substack.comLatest Issues
Distributed Data Parallelism in JAX
After building a quad-GPU cluster at home, attempting to train GPT-2 despite severe memory constraints, and doing some early memory optimizations, I wanted to take a step back and understand distributed training at a more fundamental level....
Trading Compute for Memory: Using Activation Recomputation with GPT2
One of the most obvious patterns that we can notice when running training on transformers is that activations take up the most memory:
I’m seeing a lot of red! :(
In my last post, I talked about training a GPT2 model and did some basic me...
Building Deep Learning Intuition on a consumer GPU: Memory Profiling GPT-2 Training
The Craigslist post
For Sale: DeepLearning workstation or miner 4 x 1070ti Nvidia frame. $500.
It all started when crypto went bust and all the bros had to sell off their rigs for cheap. For $500, I got four perfect condition 1070 Tis...
Why do we need more than PyTorch for LLM training?
Photo by Evan Dennis on Unsplash
One of my goals with creating this blog is to periodically externalize my own high level understanding of the distributed deep learning landscape, including the latest tools and research. Recently I was ask...
Freedom Mode
This is actually not that far off from how I felt building my deep learning workstation.
I’ve had many brushes with AI over the last decade, as a student, an amateur enthusiast, and as a software engineer. Whenever I would hear about the l...
Key Facts
Audience Metrics
Subscribers, engagement, traffic and sponsorship for All New Mistakes.
| Subscribers | Monthly Web Visits | Accepts Sponsors | |||
|---|---|---|---|---|---|
| Estimated Cost per Ad | |||||
Explore More
Authors
The writers behind this newsletter.
Frequently Asked Questions
How can I access the email archive for All New Mistakes?
You can find recent issues that have been published by All New Mistakes on Reletter by scrolling up to where it says Latest Issues. Tap on the link for any of the most recent emails or hit More Issues to see older ones.
How many subscribers does All New Mistakes have?
To see how many people subscribe to All New Mistakes, simply upgrade your Reletter account. We provide readership numbers and lots of other stats for this newsletter so you can decide if it's worth reaching out to.
How can I advertise in All New Mistakes?
Newsletter advertising can be extremely effective when it's done right. Before you pitch All New Mistakes as a potential sponsor or partner, make sure that you've done your research and checked its newsletter stats with Reletter.
Then, personalize one of our winning pitching templates and send it to the right person using the contact info provided.
How much does it cost to sponsor a publication like All New Mistakes?
Newsletter ad rates (or CPM) vary depending on many factors, including industry, number of subscribers, open rate, ad placement and more.
To find out how much an ad will cost, contact All New Mistakes using the contact information provided and ask for a copy of their media kit.
How can I find newsletters related to All New Mistakes?
Scroll up to where it says Related Newsletters to see other publications like All New Mistakes. You can also search our email newsletter directory to discover other newsletters that cover the topics you're interested in.
How do I contact All New Mistakes?
Reletter provides this newsletter's website URL above, where you will often find their contact information. We also provide links to associated social media accounts and pitching templates so you can reach out fast.