Update app.py

app.py

@@ -3,7 +3,7 @@ from huggingface_hub import InferenceClient
 from pathlib import Path
 from typing import List
 from pdfplumber import open as open_pdf
-from transformers import GPT2LMHeadModel, AutoTokenizer
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
 """
 For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
@@ -19,9 +19,9 @@ with open_pdf(pdf_path) as pdf:
 chunk_size = 1000  # Adjust this value based on your needs
 text_chunks: List[str] = [text[i:i+chunk_size] for i in range(0, len(text), chunk_size)]
 
-# Load the GPT2LMHeadModel and tokenizer
-model = GPT2LMHeadModel.from_pretrained("gpt2")
-tokenizer = AutoTokenizer.from_pretrained("gpt2")
+# Load the AutoModelForCausalLM and tokenizer
+model = AutoModelForCausalLM.from_pretrained("tiiuae/falcon-tpu-automotive")
+tokenizer = AutoTokenizer.from_pretrained("tiiuae/falcon-tpu-automotive")
 
 def respond(
     message,
@@ -50,7 +50,7 @@ def respond(
     # Encode the context and user's message
     input_ids = tokenizer.encode(context + "\n\n" + message, return_tensors="pt")
 
-    # Generate the response using the GPT2LMHeadModel
+    # Generate the response using the AutoModelForCausalLM
     output = model.generate(
         input_ids,
         max_length=max_tokens,