mirror of
https://github.com/huggingface/text-generation-inference.git
synced 2025-04-19 22:02:06 +00:00
fix(server): fix flash causal (#219)
This commit is contained in:
parent
86bca365df
commit
1ffea36ec2
@ -225,6 +225,7 @@ class FlashCausalLMBatch(Batch):
|
|||||||
|
|
||||||
return FlashCausalLMBatch(
|
return FlashCausalLMBatch(
|
||||||
batch_id=self.batch_id,
|
batch_id=self.batch_id,
|
||||||
|
past_pad=self.past_pad,
|
||||||
requests=requests,
|
requests=requests,
|
||||||
requests_idx_mapping=requests_idx_mapping,
|
requests_idx_mapping=requests_idx_mapping,
|
||||||
input_ids=input_ids,
|
input_ids=input_ids,
|
||||||
@ -311,6 +312,7 @@ class FlashCausalLMBatch(Batch):
|
|||||||
|
|
||||||
return FlashCausalLMBatch(
|
return FlashCausalLMBatch(
|
||||||
batch_id=batches[0].batch_id,
|
batch_id=batches[0].batch_id,
|
||||||
|
past_pad=batches[0].past_pad,
|
||||||
requests=requests,
|
requests=requests,
|
||||||
requests_idx_mapping=requests_idx_mapping,
|
requests_idx_mapping=requests_idx_mapping,
|
||||||
input_ids=input_ids,
|
input_ids=input_ids,
|
||||||
|
Loading…
Reference in New Issue
Block a user