DanielHesslow
607acd4fbd
Add Gated-SiLU to T5 ( #17420 )
...
* Add gated-silu to t5 architecture to support UL2
* Fix error message
* formatting
* formatting again
* refactor
* fix classnames in _init_weights
* remove is_gated
* add test
* fix test
* Try without the test?
* Add back the test.
* Improve error message.
Co-authored-by: Daniel Hesslow <daniel@lighton.ai >
2022-06-03 10:56:37 +02:00
..
2022-05-03 14:42:02 +02:00
2022-05-16 13:24:20 -04:00
2022-05-31 10:02:55 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-12 16:25:55 -04:00
2022-05-12 16:25:55 -04:00
2022-05-12 16:25:55 -04:00
2022-05-03 14:42:02 +02:00
2022-05-12 16:25:55 -04:00
2022-05-12 16:25:55 -04:00
2022-05-31 10:02:55 +02:00
2022-05-31 10:02:55 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-12 16:25:55 -04:00
2022-05-31 10:02:55 +02:00
2022-05-03 14:42:02 +02:00
2022-05-12 16:25:55 -04:00
2022-05-03 14:42:02 +02:00
2022-06-01 16:27:23 +02:00
2022-05-18 17:47:18 +02:00
2022-05-12 16:25:55 -04:00
2022-05-12 16:25:55 -04:00
2022-05-10 16:21:44 -04:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-16 21:14:25 +02:00
2022-05-12 16:25:55 -04:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-31 14:06:44 +01:00
2022-05-03 14:42:02 +02:00
2022-05-24 09:31:10 -04:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-12 16:25:55 -04:00
2022-05-03 14:42:02 +02:00
2022-05-24 12:30:46 +02:00
2022-05-31 10:02:55 +02:00
2022-05-24 09:53:45 +02:00
2022-05-24 09:53:45 +02:00
2022-06-01 16:26:37 +02:00
2022-05-12 16:25:55 -04:00
2022-06-01 17:06:20 +02:00
2022-05-03 14:42:02 +02:00
2022-06-01 10:03:06 -04:00
2022-05-03 14:42:02 +02:00
2022-05-31 10:02:55 +02:00
2022-05-31 10:02:55 +02:00
2022-05-03 14:42:02 +02:00
2022-05-31 10:02:55 +02:00
2022-05-12 16:25:55 -04:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-12 16:25:55 -04:00
2022-05-12 16:25:55 -04:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-06-02 18:52:46 +02:00
2022-05-31 10:02:55 +02:00
2022-05-12 16:25:55 -04:00
2022-05-03 14:42:02 +02:00
2022-05-31 10:02:55 +02:00
2022-05-03 14:42:02 +02:00
2022-05-12 16:25:55 -04:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-12 16:25:55 -04:00
2022-05-12 16:25:55 -04:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-17 14:33:13 +02:00
2022-05-03 14:42:02 +02:00
2022-05-04 10:04:10 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-31 10:02:55 +02:00
2022-05-31 10:02:55 +02:00
2022-05-17 23:42:14 +02:00
2022-05-03 14:42:02 +02:00
2022-05-31 10:02:55 +02:00
2022-06-03 10:56:37 +02:00
2022-06-01 21:01:32 +02:00
2022-05-03 14:42:02 +02:00
2022-05-17 19:07:43 -04:00
2022-05-03 14:42:02 +02:00
2022-05-31 10:02:55 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-12 16:25:55 -04:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-12 10:42:27 +02:00
2022-05-31 15:01:54 +02:00
2022-05-12 16:25:55 -04:00
2022-05-17 18:20:36 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-31 10:02:55 +02:00
2022-05-03 14:42:02 +02:00
2022-05-12 16:25:55 -04:00
2022-05-12 16:25:55 -04:00
2022-05-03 14:42:02 +02:00
2022-05-31 10:02:55 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00
2022-05-03 14:42:02 +02:00