Hugging Face Journal Club: Scaling Laws for Pre-training & RL
The Hugging Face research team discusses the paper "Understanding Reasoning from Pretraining to Post-Training" which discovers a joint scaling law for pre-training and reinforcement learning. Paper link:…