Update webui.py

webui.py

@@ -53,7 +53,6 @@ def load_model(path, n_ctx, n_gpu_layers, n_threads, verbose, f16_kv, logits_all
     except:
         return ""
 
-load_model("llama-2-7b-chat.ggmlv3.q2_K.bin", 2048, 0, 32, True, True, False, False, True, False, 512, 64, False, 10000, 1)
 
 def list_models(name):
     return os.listdir(f"{dir}/models")
@@ -80,6 +79,8 @@ If a question does not make any sense, or is not factually coherent, explain why
 """
 '''
 
+load_model("llama-2-7b-chat.ggmlv3.q2_K.bin", 2048, 0, 32, True, True, False, False, True, False, 512, 64, False, 10000, 1)
+
 #@spaces.GPU
 def generate_text(message, history, system_prompt, preset, temperature, max_tokens, top_p, top_k, repeat_penalty):
     temp = ""
@@ -115,7 +116,7 @@ def generate_text(message, history, system_prompt, preset, temperature, max_toke
 
 
     
-chatbot = gr.Chatbot(show_label=False, layout="panel", show_copy_button=True, height=500, min_width=180)
+chatbot = gr.Chatbot(show_label=False, layout="panel", show_copy_button=True, height=500, min_width=180, autofocus=False)
 
 with gr.Blocks(theme="theme-repo/STONE_Theme", title="TensorLM", css="style.css") as demo:
     with gr.Row():