A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 06:00 UTC
Fix Typo in Documentation and Function Name (#19442)
leopardracer committed
Jun 11, 2025 at 08:44 UTC
7c644ab6d55d887b3fbc67250897a930e29e0591
2 files changed
+2
-2
tests/kernels/attention/test_encoder_decoder_attn.py
+1
-1
@@ -99,7 +99,7 @@ class TestResources(NamedTuple):
99
Attributes:
100
101
* scale: 1/sqrt(d) scale factor for attn
102
- * attn_backend: implementatino of abstraction
102
+ * attn_backend: implementations of abstraction
103
attention interface using
104
a particular kernel library
105
i.e. XFormers
tests/v1/sample/test_topk_topp_sampler.py
+1
-1
@@ -28,7 +28,7 @@ def reset_default_device():
28
torch.set_default_device(original_device)
29
30
31
-def test_topk_impl_equivalance():
31
+def test_topk_impl_equivalence():
32
33
torch.set_default_device(DEVICE)
34
generator = Generator(device=DEVICE).manual_seed(33)