Update app.py

app.py

@@ -27,7 +27,7 @@ TOKEN=os.getenv('HF_TOKEN')
 subprocess.run(["huggingface-cli", "login", "--token", TOKEN, "--add-to-git-credential"])
 ######
 # set this key as an environment variable
-os.environ["HUGGINGFACEHUB_API_TOKEN"] = st.secrets["HF_TOKEN"]
+os.environ["HF_TOKEN"] = st.secrets["HF_TOKEN"]
 
 
 # Initialize tokenizer
@@ -37,11 +37,11 @@ def load_model():
     MODEL_NAME = "google/gemma-2b-it"
 
     model = AutoModelForCausalLM.from_pretrained(
-        MODEL_NAME,
+        MODEL_NAME,\
         # quantization_config=nf4_config, # add config
         # torch_dtype=torch.bfloat16, # save memory using float16
         # low_cpu_mem_usage=True,
-        token=st.secrets["HF_TOKEN"],
+        # token=st.secrets["HF_TOKEN"],
     ).to("cuda")
 
     tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)