Merge pull request #1851 from ssalgadodev/patch-30

Update model-catalog-overview.md

Author: prmerger-automator[bot]

articles/ai-studio/how-to/deploy-models-llama.md

@@ -5,7 +5,7 @@ description: Learn how to use the Meta Llama family of models with Azure AI Foun
 ms.service: azure-ai-studio
 manager: scottpolly
 ms.topic: how-to
-ms.date: 08/08/2024
+ms.date: 12/04/2024
 ms.reviewer: shubhiraj
 reviewer: shubhirajMsft
 ms.author: ssalgado

articles/ai-studio/how-to/model-catalog-overview.md

@@ -10,7 +10,7 @@ ms.custom:
   - ai-learning-hub
   - ignite-2024
 ms.topic: how-to
-ms.date: 5/21/2024
+ms.date: 12/04/2024
 ms.reviewer: jcioffi
 ms.author: ssalgado
 author: ssalgadodev

articles/machine-learning/concept-model-catalog.md

@@ -11,7 +11,7 @@ author: ssalgadodev
 ms.reviewer: timanghn
 ms.custom: references_regions, build-2024
 ms.collection: ce-skilling-ai-copilot 
-ms.date: 05/02/2024
+ms.date: 12/11/2024
 #Customer intent: As a data scientist, I want to learn about models available in the model catalog.
 ---
 

articles/machine-learning/how-to-deploy-models-from-huggingface.md

@@ -10,7 +10,7 @@ ms.topic: how-to
 ms.reviewer: None
 author: ssalgadodev
 ms.author: ssalgado
-ms.date: 12/15/2023
+ms.date: 12/11/2024
 ms.collection: ce-skilling-ai-copilot
 ---
 
@@ -26,7 +26,6 @@ Microsoft has partnered with Hugging Face to bring open-source models from Huggi
 
 Managed online endpoints in Azure Machine Learning help you deploy models to powerful CPU and GPU machines in Azure in a turnkey manner. Managed online endpoints take care of serving, scaling, securing, and monitoring your models, freeing you from the overhead of setting up and managing the underlying infrastructure. The virtual machines are provisioned on your behalf when you deploy models. You can have multiple deployments behind and [split traffic or mirror traffic](./how-to-safely-rollout-online-endpoints.md) to those deployments. Mirror traffic helps you to test new versions of models on production traffic without releasing them production environments. Splitting traffic lets you gradually increase production traffic to new model versions while observing performance. [Auto scale](./how-to-autoscale-endpoints.md) lets you dynamically ramp up or ramp down resources based on workloads. You can configure scaling based on utilization metrics, a specific schedule or a combination of both. An example of scaling based on utilization metrics is to add nodes if CPU utilization goes higher than 70%. An example of schedule-based scaling is to add nodes based on peak business hours. 
 
-
 ## Deploy HuggingFace hub models using Studio 
 
 To find a model to deploy, open the model catalog in Azure Machine Learning studio. Select 'All Filters', then select 'HuggingFace' in the Filter by collections section. Select the model tile to open the model page.
@@ -89,7 +88,7 @@ ml_client.begin_create_or_update(endpoint).result()
 
 ### Test the deployed model
 
-Create a file with inputs that can be submitted to the online endpoint for scoring. Below code sample input for the `fill-mask` type since we deployed the `bert-base-uncased` model. You can find input format, parameters and sample inputs on the [Hugging Face hub inference API documentation](https://huggingface.co/docs/api-inference/detailed_parameters).
+Create a file with inputs that can be submitted to the online endpoint for scoring. The code sample in this section allows an input for the `fill-mask` type since we deployed the `bert-base-uncased` model. You can find input format, parameters and sample inputs on the [Hugging Face hub inference API documentation](https://huggingface.co/docs/api-inference/detailed_parameters).
 
 ```python
 import json