bridgequant-so101-76m-sdrop01

BridgeQuant 76M SO101 teacher retrained with proprio.state_dropout: 0.1 (everything else identical to twanghcmut/bridgequant-so101-76m): a trial to make the policy rely on the wrist camera instead of the joint trajectory after a missed grasp. Same three tasks, same data.

Files: model.safetensors, norm_stats.json, bridgequant_vla.yaml / config.json / metadata.json, train_config.yaml (the teacher paths inside are only needed for training; inference ignores them).

Code: https://github.com/VLAOpt/bridgequant-async — scripts/so101_policy.py, guide in docs/so101_deploy.md.