-
-
Notifications
You must be signed in to change notification settings - Fork 10k
Description
📚 The doc issue
On page https://docs.vllm.ai/en/latest/serving/engine_args.html#engine-args
Regarding the flag --override-pooler-config
. The documentation provides the following example:
Override or set the pooling method for pooling models. e.g. {“pooling_type”: “mean”, “normalize”: false}.’
However this example does not work if copy-pasted into a UTF-8 aware text editor as it is not a valid JSON document. (The quotation marks are not ascii quotation marks, they are left-quote and right-quote.) This is an insidious error as it is nearly invisible to the naked eye.
In addition to --override-pooler-config
, this issue affects --override-neuron-config
, --rope-scaling
, and --mm-processor-kwargs
.
Suggest a potential alternative/fix
Change
Override or set the pooling method for pooling models. e.g. {“pooling_type”: “mean”, “normalize”: false}.’
to
Override or set the pooling method for pooling models. e.g.
{"pooling_type": "mean", "normalize": false}
.’
(Replace non-ascii quotes with ascii quotes and surround with a code block to ensure verbatim inclusion.)
Before submitting a new issue...
- Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.