Reading-Note - 标签 - vllbc02's blogs

Reading-Note

2026

LLM 训练教程 · Ch12 端到端 CPT（继续预训练） 09-01

Ultra-Scale Playbook · High-Level Overview 速查 08-26

SAO — Single-Rollout Asynchronous Optimization for Agentic RL 08-21

GLM-5.3 阅读笔记 — 后训练 scaling、SAO 算法与 Slime 框架 08-17

SFT、RL 与 OPD：从分布视角理解后训练 08-03

OPD (On-Policy Distillation) 07-31

Training extremely large neural networks across thousands of GPUs 07-30