6 Commits

Author	SHA1	Message	Date
Tom Lu	2cd0007685	Add state-based training resume command	2026-07-23 19:48:41 +08:00
Tom Lu	54ac00152a	Use a fixed training container and runtime state	2026-07-23 19:39:07 +08:00
Tom Lu	98c159679b	Add Stage 2 curriculum and graceful checkpoint shutdown	2026-07-23 19:30:09 +08:00
yi_lu	2b0588ad17	Sync g0050 training and inference tooling	2026-07-09 08:02:25 +08:00
yi_lu	5609b1f8e4	Add Megatron data manifest and g0050 setup	2026-07-02 20:50:24 +08:00
yi_lu	dcabce8c07	Initialize rebalanced 2B MoE project	2026-07-01 16:41:54 +08:00