correct prompt.json download hyperlink for xpu llm (#1921)

llm/xpu/index.html

@@ -183,7 +183,7 @@ <h4>Preparations<a class="headerlink" href="#preparations" title="Permalink to t
 
 <span class="c1"># Get the sample prompt.json</span>
 <span class="c1"># Make sure the downloaded prompt.json file is under the same directory as that of the python scripts mentioned above.</span>
-$<span class="w"> </span>wget<span class="w"> </span>https://intel-extension-for-pytorch.s3.amazonaws.com/miscellaneous/llm/prompt.txt
+$<span class="w"> </span>wget<span class="w"> </span>https://intel-extension-for-pytorch.s3.amazonaws.com/miscellaneous/llm/prompt.json
 </pre></div>
 </div>
 <p>The following environment variables are required to run llama on Intel® Data Center GPU Max Series.</p>