Edit readme

Changed blog link, edited analytics labels, minor formatting

README.md

@@ -9,7 +9,8 @@ pinned: false
 
 <div class="grid lg:grid-cols-3 gap-x-4 gap-y-7">
 	<p class="lg:col-span-3">
-	  Intel and Hugging Face are working together to democratize machine learning, making the latest and greatest models from Hugging Face run fast and efficiently on Intel devices. To make this acceleration accessible to the global AI community, Intel is proud to sponsor the free and accelerated inference of over 80,000 open source models on Hugging Face, powered by <a href="https://www.intel.fr/content/www/fr/fr/products/details/processors/xeon.html" class="underline" data-ga-category="intel-org" data-ga-action="clicked partner page" data-ga-label="partner page">Intel Xeon Ice Lake processors</a> in the Hugging Face Inference API. Intel Xeon Ice Lake provides up to <a href="https://huggingface.co/blog/openvino" class="underline" data-ga-category="intel-org" data-ga-action="clicked partner page" data-ga-label="partner page">34% acceleration for transformer model inference</a>. Try it out today on any Hugging Face model, right from the model page, using the Inference Widget.
+	  Intel and Hugging Face are working together to democratize machine learning, making the latest and greatest models from Hugging Face run fast and efficiently on Intel devices. To make this acceleration accessible to the global AI community, Intel is proud to sponsor the <strong>free and accelerated inference of over 80,000 open source models on Hugging Face</strong>, powered by <a href="https://www.intel.fr/content/www/fr/fr/products/details/processors/xeon.html" class="underline" data-ga-category="intel-org" data-ga-action="clicked partner link" data-ga-label="xeon link">Intel Xeon Ice Lake processors</a> in the Hugging Face Inference API. Intel Xeon Ice Lake provides up to <a href="https://huggingface.co/blog/infinity-cpu-performance" class="underline" data-ga-category="intel-org" data-ga-action="clicked partner link" data-ga-label="acceleration">34% acceleration for transformer model inference</a>.<br/><br/> 
+	  <strong>Try it out today</strong> on any Hugging Face model, right from the model page, using the Inference Widget!
 	</p>
 	<a
 		href="https://huggingface.co/blog/intel"