Downloads · 30 days
0
pi05-build-block-tower-subtask-recap-mixed
pravsels/pi05-build-block-tower-subtask-recap-mixed
pi05-build-block-tower-subtask-recap-mixed is a robotics model from pravsels. Use it for the robotics task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Fine-tuned pi0.5 checkpoint for block tower building using mixed reward recap semantics with advantage conditioning, dropout, and hierarchical subtask loss.
Access
Public
Updated Apr 10, 2026
Repo size
12.4 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
Other12.4 GB · 100%
At a glance
- Task
- Robotics
- License
- apache-2.0
- Access
- Public
- Created
- Apr 10, 2026
- Updated
- Apr 10, 2026
- SHA
- 7af7f2e4
Model card
From the Hugging Face model README
pi0.5 Build Block Tower Subtask Reward Recap - Mixed
Fine-tuned pi0.5 checkpoint for block tower building using mixed reward recap semantics with advantage conditioning, dropout, and hierarchical subtask loss.
Experiment
- Config name:
pi05_build_block_tower_subtask_recap_mixed - Run type: experiment
- Objective: train block tower recap with mixed advantage conditioning and hierarchical subtask prompting (subtask_loss_weight=1.0) to compare against flat-prompt and positive-only variants
Dataset
- 6 HuggingFace datasets:
villekuosmanen/build_block_towerplus 5 DAgger rounds (1.0.0 through 1.4.0)
Uploaded Checkpoints
22000: final checkpoint (training stopped at 22k/50k steps), SHA-2567addc02dc6542c61b08b8f608d5784d7aa1d5bbfc47d85a4c3ba9c8bf2af29f6
Checkpoints are stored as params-only artifacts under checkpoints/<step>/params/.
To verify integrity after download:
cd checkpoints/22000 && find params -type f | sort | xargs sha256sum | sha256sum
Assets
assets/contains normalization stats and dataset metadata used by this run.
W&B
Repo Structure
checkpoints/22000/params/
checkpoints/22000/assets/
README.md
TRAINING_LOG.md
- Task
- Robotics
- License
- apache-2.0
- Created
- Apr 10, 2026
- Updated
- Apr 10, 2026