WRM_delta-all125

Complete checkpoint series for WRM_delta trained on the local WRM_grasp_cylinder_different_sizes_lero_tightness LeRobot dataset.

• GPULab job: e50cbdd4-cb5b-445b-93fc-a54de5ec3d46
• Training steps: 10000 through 100000, every 10000 steps
• Checkpoints: ten numbered milestones plus final EMA-enabled last.pt
• SHA-256: aeb41e8c69f67b81c89d16bbd4c44731aba92e15ff1d609f231d60e717d4fa54
• Source configuration: resolved_config.yaml
• Metrics: metrics.jsonl

The repository is private and intended for AIRO-Doffy real-robot evaluation.