pi05_so101_block_stack_sim

pi0.5 fine-tuned to stack one 30 mm block on another with an SO-101, in simulation. Trained by train/pi05_so101_lora_backbone.yaml in vla-test on revision v2 of drwahl/so101_block_stack_sim.

LoRA on lerobot/pi05_base's gemma_2b language backbone, with the action expert and the projection heads dense and SigLIP frozen. adapter/ holds the adapter and the normalization processors; train/merge.py folds it into the base weights for a checkpoint that loads without one.

hf download drwahl/pi05_so101_block_stack_sim --local-dir pi05_so101_block_stack_sim
uv run python train/merge.py pi05_so101_block_stack_sim/adapter pi05_so101_block_stack_sim/merged
uv run python sim/scripts/eval.py pi05_so101_block_stack_sim/merged

The eval rolls the checkpoint out on the 150 held-out layouts and scores it by the gates that screened the oracle. Chunks are stitched with Real-Time Chunking and read back off a shape-preserving cubic before they are executed.

drwahl/pi05_so101_block_stack is the same recipe trained on sim and hardware episodes together, and is the one that runs on the arm.