Support fused_attention_qkv for auto_parallel llama #8432
0.00% of diff hit (target 80.00%)
View this Pull Request on Codecov
0.00% of diff hit (target 80.00%)
Annotations
Check warning on line 275 in paddlenlp/transformers/llama/modeling_auto.py
codecov / codecov/patch
paddlenlp/transformers/llama/modeling_auto.py#L275
Added line #L275 was not covered by tests
Check warning on line 277 in paddlenlp/transformers/llama/modeling_auto.py
codecov / codecov/patch
paddlenlp/transformers/llama/modeling_auto.py#L277
Added line #L277 was not covered by tests
Check warning on line 416 in paddlenlp/transformers/llama/modeling_auto.py
codecov / codecov/patch
paddlenlp/transformers/llama/modeling_auto.py#L416
Added line #L416 was not covered by tests
Check warning on line 419 in paddlenlp/transformers/llama/modeling_auto.py
codecov / codecov/patch
paddlenlp/transformers/llama/modeling_auto.py#L419
Added line #L419 was not covered by tests
Check warning on line 425 in paddlenlp/transformers/llama/modeling_auto.py
codecov / codecov/patch
paddlenlp/transformers/llama/modeling_auto.py#L424-L425
Added lines #L424 - L425 were not covered by tests