OAT Dense LIBERO-10 — Checkpoint Epoch 700
Hugging Face model repository for a dense cross-attention OAT policy trained on
LIBERO-10 (N500). This snapshot was taken at epoch 700 during a long run
(oat_dense_with_uid_long_0530_220204).
Files
| File | Description |
|---|---|
ep-0700.ckpt |
PyTorch workspace checkpoint (~729 MB) |
training_logs.jsonl |
Full training JSONL (train/val curves) |
training_metrics_dashboard.png |
Training loss dashboard |
overfit_watcher/ |
Counterfactual early-stop reports |
sim_eval/ |
Phase A screen eval (30 ep/task) |
sim_eval_phase_b/ |
Phase B confirm eval (50 ep/task, 3 exp) — ep-0700 only |
experiment_log_dense_visual_memory.md |
Experiment journal |
Metrics @ epoch 700
| Train loss | Val loss | Reconst MSE | Sim SR (Phase A) |
|---|---|---|---|
| 2.067359447479248 | 5.751441478729248 | 0.07376550883054733 | 51.7% |
Training (offline)
Sim eval (LIBERO-10, Phase A screen)
Mean success rate: 51.7% — 30 episodes/task, 300 total rollouts, seed 1000.
Details: sim_eval/eval_summary.md · sim_eval/eval_log.json
Sim eval Phase B confirm (50 ep/task, 3 seeds)
Mean success rate: 47.60% ± 1.75% — official-style protocol for comparison with OAT paper (~56.3%).
Details: sim_eval_phase_b/ (separate from Phase A sim_eval/ — not overwritten).
Model configuration (summary)
- Policy: OAT with
use_dense_visual_memory=true(spatial visual tokens + cross-attn) - State memory: enabled (
use_state_memory_tokens=true) - Task UID: enabled in state tokens
- Dataset:
libero10_N500.zarr - Embed dim: 256
Baseline reference
Paper OAT8 on LIBERO-10: ~56.3% mean success rate (external reference).
Citation
If you use this checkpoint, please cite OAT: Ordered Action Tokenization and specify epoch 700 of the dense LIBERO-10 ladder.