tocmo0nlord/axolotl
1
0
Fork 0
Code Issues Pull Requests Actions 3 Packages Projects Releases Wiki Activity

duplicate code ignore

Browse Source
This commit is contained in:
Dan Saunders
2024-12-17 18:54:49 +00:00
parent bda1eed59e
commit 63b8e42c6b
1 changed files with 4 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
src/axolotl/integrations/differential_transformer/differential_attention.py
View File

@@ -262,6 +262,7 @@ class LlamaDifferentialSdpaAttention(LlamaDifferentialAttention):
dtype: Data type for the layer parameters
"""
# pylint: disable=duplicate-code
def forward(
self,
hidden_states: torch.Tensor, # [bsz, seq_len, hidden_size]
@@ -284,8 +285,8 @@ class LlamaDifferentialSdpaAttention(LlamaDifferentialAttention):
'but specifying the manual implementation will be required from Transformers version v5.0.0 onwards. This warning can be removed using the argument `attn_implementation="eager"` when loading the model.'
)
return super().forward(
hidden_states=hidden_states, # pylint: disable=duplicate-code
attention_mask=attention_mask, # pylint: disable=duplicate-code
hidden_states=hidden_states,
attention_mask=attention_mask,
position_ids=position_ids,
past_key_value=past_key_value,
output_attentions=output_attentions,
@@ -422,6 +423,7 @@ class LlamaDifferentialFlashAttention2(LlamaDifferentialAttention):
dtype: Data type for the layer parameters
"""
# pylint: disable=duplicate-code
def forward(
self,
hidden_states: torch.Tensor, # [bsz, seq_len, hidden_size]