1
0
mirror of https://github.com/ROCm/jax.git synced 2025-04-22 23:16:06 +00:00
Adam Paszke 3649da56fc [Mosaic GPU] Make the s4 -> bf16 upcast more flexible when it comes to vector length
We can now perform the conversion in groups of 2, 4 or even 8 elements at a time.

PiperOrigin-RevId: 737626600
2025-03-17 08:37:17 -07:00
..
2024-12-18 07:46:14 +00:00
2025-03-07 04:52:58 +00:00
2024-05-25 17:46:01 +00:00
2024-10-07 12:27:35 -07:00
2024-11-05 13:28:17 -08:00
2025-02-18 09:44:16 +00:00
2025-03-14 11:48:21 -07:00
2025-03-12 18:15:14 -04:00