 Tri Dao
|
dfe29f5e2b
[Gen] Don't use ft_attention, use flash_attn_with_kvcache instead
|
1 жил өмнө |
|
Tri Dao
|
8a733cbd53
[Gen] Fix calling update_graph_cache in tests
|
1 жил өмнө |
|
Tri Dao
|
913922cac5
[Gen] Refactor decoding function
|
1 жил өмнө |
|
Tri Dao
|
9b713872ea
[GPT] Move GPT and OPT generation tests to test_{gpt,opt}.py
|
1 жил өмнө |
|
Tri Dao
|
0e8c46ae08
Run isort and black on test files
|
1 жил өмнө |
|
Tri Dao
|
4d87e4d875
Implement GPT-J
|
1 жил өмнө |
|
Tri Dao
|
f68d41ec77
[Gen] Add OPT to generation test
|
2 жил өмнө |
|
Tri Dao
|
780e8eeabb
[ViT] Support timm checkpoint, add tests
|
2 жил өмнө |
|
Tri Dao
|
ff34123bd4
Reorder LN in Block, support OPT
|
2 жил өмнө |
