Lora MoE Align Improvements (#29257)
Signed-off-by: gnovack <gnovack@amazon.com>
This commit is contained in:
@@ -1961,6 +1961,7 @@ def moe_lora_align_block_size(
|
||||
num_tokens_post_pad: torch.Tensor,
|
||||
adapter_enabled: torch.Tensor,
|
||||
lora_ids: torch.Tensor,
|
||||
expert_map: torch.Tensor | None = None,
|
||||
) -> None:
|
||||
torch.ops._moe_C.moe_lora_align_block_size(
|
||||
topk_ids,
|
||||
@@ -1975,6 +1976,7 @@ def moe_lora_align_block_size(
|
||||
num_tokens_post_pad,
|
||||
adapter_enabled,
|
||||
lora_ids,
|
||||
expert_map,
|
||||
)
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user