sclarkson
|
1feb711f46
Fix compilation with clang on ARM64 (#1285)
|
1 month ago |
Tri Dao
|
e45a46a5b7
[Rotary] Implement GPT-J style (interleaved) rotary
|
1 year ago |
Tri Dao
|
1e712ea8b0
Implement TensorParallel for MHA
|
2 years ago |
Tri Dao
|
ca81f32e04
Implement rotary embedding in CUDA
|
2 years ago |