HALF OFF all print books, including all preorders in MEAP! Renew or get Manning Online Pro annual for just $199.99—SAVE $50!
Save 37% on everything else
Offer ends midnight PT July 19th. Only at manning.com.
Picture this. You're sitting on a lounge chair at the pool, sipping a cold beverage. You put on your sun hat and Wayfarers and pull a brand new copy of Build a Reasoning Model (From Scratch) out of your beach bag. This could be you! Save half on all print books this week at manning.com—or get it all with a Manning Online Pro annual subscription for just $199.99.
Now in print! Build a Reasoning Model (From Scratch)
In Build A Reasoning Model (From Scratch) you’ll learn how to build a working reasoning model from the ground up. Sebastian Raschka, the bestselling author of Build A Reasoning Model (From Scratch), is your guide on this exciting journey. Sebastian mentors you every step of the way with clear explanations, practical code, and a keen focus on what really matters.
Reasoning models break problems into steps, producing more reliable answers in math, logic, and code. Build A Reasoning Model (From Scratch) demystifies these complex models with a simple philosophy: the best way to learn how
something works is to build it yourself! You’ll begin with a pre-trained LLM, adding and improving its reasoning capabilities in ways you can see, test, and understand. [Read more]
Now in print! Reinforcement Learning from Human Feedback
When ChatGPT proved that Reinforcement Learning from Human Feedback (RLHF) could make AI production-ready, the technique transformed the field overnight. In Reinforcement Learning from Human Feedback, post-training expert Nathan Lambert offers an insider's guide to building AI that is safer, smarter, and aligned with human values. Drawing on his real-world experience building models like Olmo, Lambert shares hard-won insights on preventing over-optimization, crafting model character, and performing rigorous model evaluation.
Through hands-on experiments and mini-
implementations, Lambert walks you through the full post-training pipeline that emerged from RLHF, from foundational theory and landmark papers to the optimization tools that turn a generic base model into a capable AI assistant. You'll explore reward models, instruction tuning, direct alignment algorithms, and the mathematics behind the latest reinforcement learning techniques with LLMs. Then go further with advanced techniques including designing asynchronous systems for reinforcement learning with verifiable rewards (RLVR), constitutional AI, synthetic data, and tool-use. [Read more]
.png?width=460&upscale=true&name=Build%20a%20Reasoning%20Model%20(From%20Scratch).png)
