molmoact2_fold_tea_towel_norm_fix

Fine-tuned MolmoAct2 (action-expert-only) on SO101 data, with SO101 gripper normalization fix.

Base model	allenai/MolmoAct2
Dataset	villekuosmanen/armnetbench_fold_tea_towel
Task	fold_tea_towel
Action dim	12 (bimanual)
Cameras	top, left_wrist, right_wrist
Normalization	MEAN_STD (action + state + gripper), IDENTITY (visual)
Training	Isambard GH200, batch 64, bf16, gradient checkpointing
W&B project	molmoact2_fold_tea_towel
W&B run	gugir64b

Checkpoints

Step	Path
002000	checkpoints/002000/pretrained_model/

The latest step (002000) is also copied to the repository root for direct loading.

Usage

from lerobot.policies.molmoact2.modeling_molmoact2 import MolmoAct2Policy
policy = MolmoAct2Policy.from_pretrained("pravsels/molmoact2_fold_tea_towel_norm_fix")