README: show ME epochs as /120 (training target) instead of /500 config cap

README: final epoch-120 eval results (mean SR 86.1%, NJD 0.209)

README: add me-full_25000_120 (final epoch-120 model, eval pending)

Add me-full_25000_120: final true-epoch-120 multi-embodiment checkpoint (step 302000)

Upload README.md with huggingface_hub

add me-full true-epoch-90 checkpoint entry to model card (eval pending)

add me-full true-epoch-90 checkpoint (folder locally labeled 150 due to an epoch-counting bug from the 3-GPU batch-size switch, corrected for the true dataset-pass count)

add me-full epoch-30 checkpoint eval to model card

add me-full epoch-30 checkpoint

unified model card: 3 checkpoints with eval results, incl. me-full epoch-5 per-gripper eval

restore original single-embodiment card; append me-full epoch-5 section

model card

epoch-5 checkpoint (full multi-embodiment config, MI300X)

Upload README.md with huggingface_hub

Upload folder using huggingface_hub

Upload folder using huggingface_hub

initial commit