
Inside s1: An o1-Style Reasoning Model That Cost Under $50 to Train with Niklas Muennighoff - #721
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
Optimizing Model Performance
This chapter explores the intricacies of enhancing model performance through controlled weight adjustments during training, revealing optimal weight factors and their impact on performance variability. It also examines the challenges of evaluating language models with a focus on compute budgets, decontamination processes, and contrasting evaluation methods such as budget forcing and rejection sampling.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.