torch-mlir/projects
Rob Suderman 25738b8c19
[linalg] Broadcast batch for mask on sdpa lowering (#3824)
Attention often broadcasts a mask across the batch dimension as masking
is usually performed the same across attention heads. Added this
materialization to the mask dimensions optionally.
2024-10-31 17:59:24 -07:00
..
jit_ir_common Additional information in error message (#2783) 2024-01-30 08:28:08 -08:00
ltc [NFC reformat] Run pre-commit on all files and format misc. 2024-04-27 14:08:09 -07:00
onnx_c_importer build: find Protobuf using config mode search (#2900) 2024-02-12 17:31:41 -06:00
pt1 [linalg] Broadcast batch for mask on sdpa lowering (#3824) 2024-10-31 17:59:24 -07:00
CMakeLists.txt [NFC reformat] Run pre-commit on all files and format misc. 2024-04-27 14:08:09 -07:00