How To Scale Your Model
An online book on efficient training and inference of large models.
BookLearning & booksinferencescalingllm
Recommended in 1 thread
Sebastian Raschka's from-scratch video course on building reasoning modelsmagazine.sebastianraschka.com ·
Mentioned as a source of topics asked in AI job interviews.
Similar tools
- Scaling Laws for Neural Language Models ↗Learning & books1 thread
A 2020 OpenAI paper relating language model performance to compute, data and parameter count.
- Build a Large Language Model (From Scratch) ↗Learning & books1 thread
Sebastian Raschka's book walking through implementing a GPT-style LLM in PyTorch.
- Designing Data-Intensive Applications Learning & books0 threads
A book by Martin Kleppmann on the architecture of data systems.
- Performance Analysis and Tuning on Modern CPUs ↗Learning & books1 thread
A book by Denis Bakhvalov on CPU performance analysis and tuning.
- 1 thread
- You Don't Need JavaScript ↗Learning & books1 thread
A book on replacing JavaScript patterns with modern CSS and HTML.
- Neural Networks: Zero to Hero Learning & books1 thread
Andrej Karpathy's video series building neural networks up to a GPT-2 equivalent.
- Crafting Interpreters Learning & books1 thread
A book by Robert Nystrom on building programming language interpreters from scratch.
Also in Learning & books
- Wikipedia3 threads
A free collaborative online encyclopedia.
- Careless People2 threads
A memoir by former Facebook executive Sarah Wynn-Williams about the company's leadership.
- 2 threads
- 2 threads
- 1960: The Year The Singularity Was Cancelled ↗1 thread
A 2019 Slate Star Codex essay by Scott Alexander on von Foerster's hyperbolic population model.
- A Little Bit of Slope Makes Up For a Lot of Y-Intercept ↗1 thread
A lecture by John Ousterhout on how rate of learning outweighs starting ability.
- A People's History of the United States1 thread
Howard Zinn's history of the US told from the perspective of ordinary and marginalised people.
- A Perfect Vacuum1 thread
Stanisław Lem's collection of reviews of nonexistent books.
- ASD-STE100 Simplified Technical English ↗1 thread
A controlled English standard originally developed for aerospace maintenance documentation.
- Accelerando1 thread
A Charles Stross novel about accelerating technological change and autonomous AI agents.
- Ashton Manual1 thread
A guide by Heather Ashton describing slow benzodiazepine tapering schedules, often via diazepam substitution.
- Atomic Physics for Everyone1 thread
An introduction to atomic physics, quantum mechanics and precision spectroscopy with no college prerequisites.