[Doc]: fix typos in Python comments (#24001)
Signed-off-by: Didier Durand <durand.didier@gmail.com>
This commit is contained in:
@@ -1239,7 +1239,7 @@ class LLMEngine:
|
||||
|
||||
# Stop the execute model loop in parallel workers until there are
|
||||
# more requests to process. This avoids waiting indefinitely in
|
||||
# torch.distributed ops which may otherwise timeout, and unblocks
|
||||
# torch.distributed ops which may otherwise time out, and unblocks
|
||||
# the RPC thread in the workers so that they can process any other
|
||||
# queued control plane messages, such as add/remove lora adapters.
|
||||
logger.debug("Stopping remote worker execution loop.")
|
||||
|
||||
Reference in New Issue
Block a user