train-llm-from-scratch · Issues· 7 open
Open on GitHubLocally synced open issues (discussions stay on GitHub)
- #36
Bug: torch.compile + DDP causes _orig_mod. prefix leakage in checkpoint keys
Updated Sep 13, 2026 - #38
Feature Request: Add a CPU-Friendly "Train an LLM from Scratch" Learning Path for Students
Updated Aug 10, 2026 - #34
Feature Request: Integrate GaaR Evolutionary Optimization Framework with Transformer Training
Updated Jun 29, 2026 - #8
help
enhancementUpdated Jun 16, 2026