Update app.py

app.py

@@ -47,7 +47,7 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
 # model_id = "mistralai/Mistral-7B-v0.3"
 
 model_id = "openchat/openchat-3.6-8b-20240522"
-model_id = "Qwen/Qwen2-7B-Instruct"
+model_id = "01-ai/Yi-1.5-34B-Chat"
 
 
 tokenizer = AutoTokenizer.from_pretrained(
@@ -61,10 +61,10 @@ accelerator = Accelerator()
 model = AutoModelForCausalLM.from_pretrained(model_id, token= token, 
                                                  # torch_dtype= torch.uint8, 
                                              torch_dtype=torch.bfloat16,
-                                              # load_in_8bit=True,
+                                              load_in_8bit=True,
                                              # #  # torch_dtype=torch.fl,
                                              attn_implementation="flash_attention_2",
-                                             # low_cpu_mem_usage=True,
+                                             low_cpu_mem_usage=True,
                                              # device_map='cuda',
                                              # device_map=accelerator.device_map,