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 16, 20:45 UTC
[doc][misc] doc update (#6439)
youkaichao committed
Jul 14, 2024 at 23:33 UTC
22e79ee8f3930c39f40f6a1529e41594a607c6b4
1 file changed
+4
docs/source/getting_started/debugging.rst
+4
@@ -50,6 +50,8 @@ Here are some common issues that can cause hangs:
50
value = cpu_data.mean().item()
51
assert value == world_size, f"Expected {world_size}, got {value}"
52
53
+ print("sanity check is successful!")
54
+
55
.. tip::
56
57
Save the script as ``test.py``.
@@ -62,4 +64,6 @@ Here are some common issues that can cause hangs:
64
- is reachable from all nodes
65
- is set before running the script.
66
67
+ If the script runs successfully, you should see the message ``sanity check is successful!``.
68
+
69
If the problem persists, feel free to `open an issue on GitHub <https://github.com/vllm-project/vllm/issues/new/choose>`_, with a detailed description of the issue, your environment, and the logs.