Update models.py

models.py

@@ -144,7 +144,11 @@ class OpenSourceLlama2Model(BaseTCOModel):
             input_tokens_cost_per_token = 0.00052
             r = maxed_out / 100
             return input_tokens_cost_per_token * 0.65 / r, output_tokens_cost_per_token * 0.65/ r
-        
+
+        self.source = gr.Markdown("""<span style="font-size: 16px; font-weight: 600; color: #212529;">Source</span>""")
+        self.info = gr.Markdown("The cost per input and output tokens values below are from [these benchmark results](https://www.cursor.so/blog/llama-inference#user-content-fn-llama-paper) that were obtained using the following initial configurations.", 
+                                 interactive=False, 
+                                 visible=False)
         self.vm = gr.Textbox(value="2x A100 80GB NVLINK", 
                               visible=False,
                               label="Instance of VM with GPU",
@@ -171,11 +175,6 @@ class OpenSourceLlama2Model(BaseTCOModel):
                                            interactive=False
                                            )
         self.maxed_out.change(on_maxed_out_change, inputs=[self.maxed_out, self.input_tokens_cost_per_token, self.output_tokens_cost_per_token], outputs=[self.input_tokens_cost_per_token, self.output_tokens_cost_per_token])
-        self.source = gr.Markdown("""<span style="font-size: 16px; font-weight: 600; color: #212529;">Source</span>""")
-        self.info = gr.Markdown("The cost per input and output tokens values above are from [these benchmark results](https://www.cursor.so/blog/llama-inference#user-content-fn-llama-paper)", 
-                                 label="Source", 
-                                 interactive=False, 
-                                 visible=False)
         
         self.labor = gr.Number(10000, visible=False, 
                                 label="($) Labor cost per month",