Daniel Han 6f3e52bbcf Prefer flex attention when available (#3979) ... * Enable flex attention by default * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Avoid dropping flex attention when SDPA unsupported --------- Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>		2026-02-05 03:19:04 -08:00
..
dataprep	[pre-commit.ci] auto fixes from pre-commit.com hooks	2026-01-08 11:35:21 +00:00
kernels	Apply suggestion from @danielhanchen	2026-01-12 21:32:20 -08:00
models	Prefer flex attention when available (#3979)	2026-02-05 03:19:04 -08:00
registry	Revert "[pre-commit.ci] auto fixes from pre-commit.com hooks"	2025-12-01 07:24:58 -08:00
utils	remove unused variable BlockDiagonalCausalMask	2026-01-04 09:21:44 -08:00
__init__.py	Add vLLM + torch < 2.9.0 + SM100 compatibility check (#3973)	2026-02-03 03:10:24 -08:00
_auto_install.py	Float8 GRPO, RL (#3640)	2025-11-25 07:23:26 -08:00
chat_templates.py	Fix indefinite article usage in comments and docstrings (#3648)	2025-11-26 18:15:27 -08:00
device_type.py	Add missing import of inspect (#3778)	2025-12-25 18:43:59 -08:00
import_fixes.py	Fix torchvision compatibility check for source builds and future torch versions (#3978)	2026-02-04 04:50:26 -08:00
ollama_template_mappers.py	Fix indefinite article usage in comments and docstrings (#3648)	2025-11-26 18:15:27 -08:00
save.py	Fix psutil.cpu_count() potentially returning None in save.py	2026-01-04 12:58:45 +00:00
tokenizer_utils.py	[fix] qwen3-guard tokenizer (#3959)	2026-02-01 22:09:15 -08:00
trainer.py	Improve TRL compatibility and GRPO state restore	2026-01-05 07:02:36 +00:00