Dan Saunders
59cd472504
SP cu_seqlens fix, refactor (#2495)
* working on masking fix
* refactor and fix multipack seqlens
* pre-commit fix
* adding smoke test
* using existing packed seqlens util
* log warning re: logged losses / gradient scaling per rank
2025-04-07 14:47:57 -04:00
..
2025-03-21 11:02:43 -04:00
2025-03-31 13:40:12 +07:00
2025-04-07 14:47:57 -04:00
2023-12-12 09:39:22 -08:00
2025-03-21 11:02:43 -04:00
2025-03-21 11:02:43 -04:00
2025-03-21 12:26:47 -04:00
2025-03-31 13:40:12 +07:00
2025-03-31 13:40:12 +07:00
2025-03-31 13:40:12 +07:00
2025-04-05 01:25:44 -04:00
2024-12-02 08:47:10 -05:00
2025-04-05 01:25:44 -04:00
2025-03-31 13:40:12 +07:00
2025-04-01 08:47:50 -04:00
2025-03-21 11:02:43 -04:00
2025-04-01 08:47:50 -04:00
2025-03-21 11:02:43 -04:00
2024-03-14 11:05:42 -04:00
2025-03-21 11:02:43 -04:00
2025-03-21 11:02:43 -04:00
2025-04-01 08:47:50 -04:00
2025-03-31 13:40:12 +07:00
2025-03-29 08:30:06 -04:00
2025-03-21 11:02:43 -04:00
2025-04-01 08:47:50 -04:00
2024-08-22 11:46:57 -04:00
2025-03-21 11:02:43 -04:00
2025-04-05 01:25:44 -04:00
2025-04-01 09:39:12 -04:00