groupuser/phi-1_5
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update modeling_mixformer_sequential.py

Browse Source 
Removes print regarding attention_mask to prevent excessive information from being logged.
This commit is contained in:
Gustavo de Rosa 2023-10-26 20:01:15 +00:00 committed by huggingface-web
parent 8091327f9e
commit d38e6f954e
1 changed files with 0 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
modeling_mixformer_sequential.py
View File

@ -756,9 +756,6 @@ class MixFormerSequentialForCausalLM(MixFormerSequentialPreTrainedModel):
labels: Optional[torch.LongTensor] = None, labels: Optional[torch.LongTensor] = None,
**kwargs, **kwargs,
) -> CausalLMOutputWithPast: ) -> CausalLMOutputWithPast:
if attention_mask is not None and self.training:
print("`attention_mask` is not supported during training. Using it might lead to unexpected results.")
if past_key_values is None and attention_mask is None: if past_key_values is None and attention_mask is None:
lm_logits = self.layers(input_ids) lm_logits = self.layers(input_ids)
else: else: