bug fixed

app.py

@@ -15,7 +15,7 @@ device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
 @st.cache_resource
 def load_model():
     base_model = AutoModelForCausalLM.from_pretrained(base_model_id, use_auth_token=hf_token)
-    model = PeftModel.from_pretrained(base_model, model_id).to(device)
+    model = PeftModel.from_pretrained(base_model, model_id, use_auth_token=hf_token).to(device)
     tokenizer = AutoTokenizer.from_pretrained(base_model_id, use_auth_token=hf_token)
     tokenizer.pad_token = tokenizer.eos_token
     tokenizer.padding_side='left'

requirements.txt

@@ -1,2 +1,3 @@
 transformers
-torch
+torch
+peft