Add perf numbers for perf_train_cpu (#20974)

* Update perf_train_cpu.mdx

* Update perf_train_cpu.mdx

* Update perf_train_cpu.mdx

* Update docs/source/en/perf_train_cpu.mdx
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update perf_train_cpu.mdx

* Update perf_train_cpu.mdx

* Update perf_train_cpu.mdx

* Update perf_train_cpu.mdx

---------
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

docs/source/en/perf_train_cpu.mdx

@@ -58,3 +58,6 @@ Take an example of the use cases on [Transformers question-answering](https://gi
 <b>--use_ipex \</b>
 <b>--bf16 --no_cuda</b></pre> 
 
+### Practice example
+
+Blog: [Accelerating PyTorch Transformers with Intel Sapphire Rapids](https://huggingface.co/blog/intel-sapphire-rapids)