Files
tinygrad/extra
b1tg 0fbc551622 train bert with fp8 (#13874)
* fp8 train

* clean

* lint

* test fix from #13439

* skip first/last layer

* rm __init__, restore unroll <=32 check

* tests

* clean test, remove unused

* multi-gpu test, clean quantize_to_fp8

* remove bert contiguous

* run script

* test: better check

* run script search

* add seed in bert data shuffle

* move script to mi350x folder

---------

Co-authored-by: chenyu <chenyu@fastmail.com>
2026-01-09 09:21:59 -05:00
..
2026-01-05 13:10:56 +03:00
2025-12-19 17:14:56 -04:00
2026-01-09 09:21:59 -05:00
2026-01-08 15:47:16 +03:00
2025-12-03 16:40:43 +03:00
2025-11-30 16:46:55 +03:00
2026-01-01 10:25:08 -05:00
2025-08-06 14:00:34 +03:00
2025-11-24 18:07:32 -08:00
2026-01-06 10:19:47 +09:00
2025-09-10 15:15:48 -04:00
2025-09-10 15:15:48 -04:00
2025-02-20 19:20:01 +08:00
2025-05-28 20:48:20 -07:00
2025-06-08 08:42:22 -07:00
2025-02-13 12:26:15 +08:00
2025-11-20 12:35:57 -05:00
2025-09-10 15:15:48 -04:00
2025-12-14 00:45:57 -05:00