zhangyiss/mlx-examples
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx-examples.git synced 2025-09-03 23:14:34 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Length masking for batch inputs (#1173)

Browse Source 
* length masking

* add mask to mlx_lm model interface

* remove lengths

* fix test:

* comment + fix
This commit is contained in:
Alex Barron
2024-12-18 19:43:52 -08:00
committed by GitHub
parent db109184b7
commit d4ef909d4a
34 changed files with 191 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
llms/mlx_lm/models/stablelm.py
View File

@@ -199,7 +199,10 @@ class Model(nn.Module):
mask: mx.array = None,
cache=None,
) -> mx.array:
mask = create_attention_mask(x, cache)
if mask is None:
mask = create_attention_mask(x, cache)
y = self.model(x, mask, cache)
return self.lm_head(y)