Python: Getting started jupyter notebook "Building Semantic with Embeddings" doesn't work out of the box

[gbm]

Running 05-memory-and-embeddings.ipynb, when I run cell

from semantic_kernel.connectors.in_memory import InMemoryStore

in_memory_store = InMemoryStore()

collection = in_memory_store.get_collection(record_type=SimpleModel)
await collection.ensure_collection_exists()
# Add records to the collection
await collection.upsert(records)

with text-embedding-ada-002 as my AZURE_OPENAI_EMBEDDING_DEPLOYMENT_NAME I get

BadRequestError: Error code: 400 - {'error': {'message': 'This model does not support specifying dimensions.', 'type': 'invalid_request_error', 'param': None, 'code': None}}

I guess this is the wrong embedding model to use.
There is no guidance over embedding model to use and I based my choice off of Tutorial: Explore Azure OpenAI in Azure AI Foundry Models embeddings and document search

[gbm]

text-embedding-3-small works as a deployment for this notebook. suggest docs / notebook is tightened up a bit on this. There is also a dependency from the same notebook onto a cognitive search deployment which isn't spelled out at the setup at the top.

[gbm]

@eavanvalkenburg I was thinking it might be better to add a 'getting setup' page to the 'getting started' dir with the jupyter notebooks. That was you are isolated from any changes (and so dependency) on the main docs. The setup instructions are a bit sparse right now.
CONFIGURING_THE_KERNEL.md is kind of not what I want. I want - deploy this model, deploy that model get that API key put it your .env, then tear it all down again once you are done.

[gbm]

@eavanvalkenburg re the ai search dependency in this notebook, the only current guidance is this line
"Please note you will need an AzureAI Search api_key or token credential and endpoint for the following example to work properly"
Would be better to offer more guidance here - maybe in the 'getting setup' page I suggest above about the need to go to Azure AI Search and create a service. And crucially here advice on creating a free tier search service as that service defaults to standard which in Sweden Central is €234.00 per month. You will not make new friends that way :)
Also note I had to use an admin API key to get the notebook code to run - a query API key didn't work. So if that's correct (and I haven't messed up somehow) then you should also make that clear. Thanks!