Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix incorrect attention mask truncate in WhisperFlashAttention2 #36477

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Fix incorrect attention mask truncate in WhisperFlashAttention2 #36477

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
54feb17
Fix incorrect attention mask truncate in whisper flash attention
OliBomby Feb 28, 2025
28a4190
also fix incorrect attention mask truncate in qwen2 audio
OliBomby Feb 28, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion src/transformers/models/qwen2_audio/modeling_qwen2_audio.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -290,7 +290,7 @@ def forward(

causal_mask = attention_mask
if attention_mask is not None: # no matter the length, we just slice it
causal_mask = attention_mask[:, : key_states.shape[-2]]
causal_mask = attention_mask[:, : key_states.shape[-3]]

# In PEFT, usually we cast the layer norms in float32 for training stability reasons
# therefore the input hidden states gets silently casted in float32. Hence, we need
Expand Down
2 changes: 1 addition & 1 deletion src/transformers/models/whisper/modeling_whisper.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -421,7 +421,7 @@ def forward(

causal_mask = attention_mask
if attention_mask is not None: # no matter the length, we just slice it
causal_mask = attention_mask[:, : key_states.shape[-2]]
causal_mask = attention_mask[:, : key_states.shape[-3]]

# In PEFT, usually we cast the layer norms in float32 for training stability reasons
# therefore the input hidden states gets silently casted in float32. Hence, we need
Expand Down