Skip to content

GitLab

Unverified Commit 43e8848d authored by UnicornChan's avatar UnicornChan Committed by GitHub
Browse files

Merge pull request #86 from xhedit/main 

typo fix: KMisrtal -> KMistral
parents 49539ac4 234faf79
Hide whitespace changes
Inline Side-by-side
Showing with 2 additions and 2 deletions
ktransformers/operators/experts.py
View file @ 43e8848d
...@@ -727,7 +727,7 @@ class KDeepseekV2MoE(BaseInjectedModule, DeepseekV2MoE): ...@@ -727,7 +727,7 @@ class KDeepseekV2MoE(BaseInjectedModule, DeepseekV2MoE):
) )
return final_out return final_out
class KMisrtalSparseMoEBlock(BaseInjectedModule, MixtralSparseMoeBlock): class KMistralSparseMoEBlock(BaseInjectedModule, MixtralSparseMoeBlock):
def forward(self, hidden_states: torch.Tensor) -> torch.Tensor: def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
""" """ """ """
......
ktransformers/optimize/optimize_rules/Mixtral.yaml
View file @ 43e8848d
...@@ -19,7 +19,7 @@ ...@@ -19,7 +19,7 @@
name: "^model\\.layers\\..*\\.block_sparse_moe$" name: "^model\\.layers\\..*\\.block_sparse_moe$"
class: ktransformers.models.modeling_mixtral.MixtralSparseMoeBlock class: ktransformers.models.modeling_mixtral.MixtralSparseMoeBlock
replace: replace:
class: ktransformers.operators.experts.KMisrtalSparseMoEBlock class: ktransformers.operators.experts.KMistralSparseMoEBlock
- match: - match:
name: "^model\\.layers\\..*\\.block_sparse_moe\\.experts$" name: "^model\\.layers\\..*\\.block_sparse_moe\\.experts$"
replace: replace:
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment