ViperEkura a74e5b91a3 feat: add record-mode to Store for DPO/GRPO ...

- Store gains fetch_record/num_records alongside stream fetch/__len__
- save_bin/load_bin support per-record offsets via record_keys param
- H5Store/MmapStore/JsonlStore all support dual stream+record access
- DPODataset/GRPODataset use fetch_record, no cross-record concat
- dpo_collate_fn + collate_fn wired through TrainConfig
- fixes attention context leakage in DPO from windowed concatenation

2026-07-18 21:02:29 +08:00

..

__init__.py

feat: add record-mode to Store for DPO/GRPO

2026-07-18 21:02:29 +08:00

checkpoint.py

feat: add --resume flag to decouple weight loading from training resumption

2026-07-16 14:23:23 +08:00

dataset.py

feat: add record-mode to Store for DPO/GRPO

2026-07-18 21:02:29 +08:00