fxmarty
1da1302ec8
Flash Attention 2 support for RoCm (#27611)
* support FA2
* fix typo
* fix broken tests
* fix more test errors
* left/right
* fix bug
* more test
* typo
* fix layout flash attention falcon
* do not support this case
* use allclose instead of equal
* fix various bugs with flash attention
* bump
* fix test
* fix mistral
* use skiptest instead of return that may be misleading
* add fix causal arg flash attention
* fix copies
* more explicit comment
* still use self.is_causal
* fix causal argument
* comment
* fixes
* update documentation
* add link
* wrong test
* simplify FA2 RoCm requirements
* update opt
* make flash_attn_uses_top_left_mask attribute private and precise comment
* better error handling
* fix copy & mistral
* Update src/transformers/modeling_utils.py
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
* Update src/transformers/modeling_utils.py
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
* Update src/transformers/modeling_utils.py
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
* Update src/transformers/utils/import_utils.py
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
* use is_flash_attn_greater_or_equal_2_10 instead of is_flash_attn_greater_or_equal_210
* fix merge
* simplify
* inline args
---------
Co-authored-by: Felix Marty <felix@hf.co>
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
2023-12-04 21:52:17 +09:00
..
2023-11-16 11:44:36 -08:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-15 10:13:52 -08:00
2023-11-28 08:40:44 -08:00
2023-11-03 10:57:03 -04:00
2023-11-10 13:40:30 +00:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-06 19:45:03 +00:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-10 13:49:10 +00:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-23 17:44:08 +00:00
2023-11-18 17:22:52 +00:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-23 17:44:08 +00:00
2023-11-03 10:57:03 -04:00
2023-06-20 18:07:47 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-09-04 17:18:34 +01:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-10-19 15:36:41 +02:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-12-04 21:52:17 +09:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-10-30 21:42:19 +01:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-28 13:19:50 +00:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-06 19:45:03 +00:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-24 12:34:24 +01:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-06 19:45:03 +00:00
2023-11-13 14:20:54 +01:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-23 10:16:51 +00:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-29 13:36:38 +01:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-10 15:28:30 +00:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-07-24 15:34:19 +01:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-06-20 18:07:47 -04:00
2023-11-03 10:57:03 -04:00
2023-11-06 19:45:03 +00:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-07-13 11:46:54 -04:00
2023-11-06 19:45:03 +00:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 15:23:13 +01:00
2023-11-30 20:24:43 +01:00
2023-12-04 11:47:54 +00:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-08-25 09:20:37 +02:00
2023-08-03 14:12:07 +01:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-06-20 18:07:47 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-06 19:45:03 +00:00
2023-11-25 00:59:17 +09:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-06 19:45:03 +00:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-24 11:48:02 +01:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-23 17:02:16 +00:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-22 17:21:36 +01:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-20 17:36:48 +01:00
2023-06-20 18:07:47 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00
2023-11-03 10:57:03 -04:00