Fix app

app.py

@@ -1,4 +1,5 @@
 # Import modules
+import os
 import torch
 import gradio as gr
 from langchain_community.llms import HuggingFacePipeline
@@ -7,7 +8,9 @@ from langchain_community.vectorstores import Chroma
 from langchain_core.runnables import RunnablePassthrough
 from langchain_core.prompts import PromptTemplate
 from peft import PeftModel
-from transformers import AutoModelForCausalLM, BitsAndBytesConfig, GenerationConfig, pipeline
+from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig, GenerationConfig, pipeline
+
+HUGGINGFACE_ACCESS_TOKEN = os.environ["HUGGINGFACE_ACCESS_TOKEN"]
 
 base_model = "microsoft/phi-2"
 
@@ -18,6 +21,14 @@ embedding_function = SentenceTransformerEmbeddings(
     model_kwargs={"device": "cuda"},  # Use the GPU
 )
 
+tokenizer = AutoTokenizer.from_pretrained(
+    base_model,
+    use_fast=True,
+    token=HUGGINGFACE_ACCESS_TOKEN,
+)
+tokenizer.pad_token = tokenizer.eos_token
+tokenizer.padding_side = "right"
+
 bnb_config = BitsAndBytesConfig(
     load_in_4bit=True,
     bnb_4bit_quant_type="nf4",