This repo contains the portable model checkpoint 0062000.pt and the DeepSpeed ZeRO-2 checkpoint shards needed for optimizer-state resume.

Important: because this run used DeepSpeed ZeRO-2, 0062000.pt intentionally stores optimizer: null. Optimizer state lives in checkpoints/0062000/.

To resume with optimizer/scheduler state in this codebase, keep the sibling layout:

Then pass --ckpt /path/to/0062000.pt with the checkpoints/0062000/ directory present; train_robot.py resolves the sibling ZeRO folder and calls DeepSpeed load_checkpoint(tag="0062000").