0z5a 587b0ee046 fix: keep async rollouts version-consistent ...

- serialize shared-model optimizer updates with generation
- reject future or over-lagged rollout results after asynchronous scoring
- close cache publication races
- persist policy versions in online checkpoints

2026-09-03 12:01:24 +08:00

..

config

fix: keep async rollouts version-consistent

2026-09-03 12:01:24 +08:00

dataset

refactor: deduplicate low-risk code paths

2026-08-19 16:17:40 +08:00

extension

perf: dispatch linear gemv by decode batch size and unify extension style

2026-09-03 07:23:13 +08:00

inference

fix: keep async rollouts version-consistent

2026-09-03 12:01:24 +08:00

model

perf: tune bf16 gemv and add opt-in fused swiglu

2026-09-03 04:26:53 +08:00

optim

fix: use torch.optim.AdamW in ManoAdamW instead of NAdamW

2026-08-01 09:20:54 +08:00

parallel

fix: resolve audited training and inference bugs

2026-09-02 21:25:01 +08:00

preprocessing

fix : correct prefill sampling and record alignment

2026-08-05 22:20:29 +08:00

serialization

fix: publish checkpoints atomically

2026-09-02 15:29:22 +08:00

tokenize

refactor: simplify BaseFactory and separate ModelFactory from AutoModel

2026-07-30 09:38:20 +08:00

trainer

fix: keep async rollouts version-consistent

2026-09-03 12:01:24 +08:00

__init__.py

refactor: simplify training and inference interfaces

2026-08-19 20:55:13 +08:00

factory.py

refactor: simplify BaseFactory and separate ModelFactory from AutoModel

2026-07-30 09:38:20 +08:00

logging.py

fix: report true rank in logs via dist-aware helpers

2026-08-27 13:39:38 +08:00

protocols.py

refactor: 重构训练后端为 Executor 模式

2026-05-24 20:35:44 +08:00

signal_handler.py

refactor: move signal_handler from parallel/ to top-level for broader reuse

2026-07-28 10:36:17 +08:00