Update app.py

app.py

@@ -1,18 +1,27 @@
 import spaces
+
 import torch
+import torch._dynamo
+
 import gradio as gr
 
 from peft import AutoPeftModelForCausalLM
 from transformers import AutoTokenizer, BitsAndBytesConfig
 
+torch._dynamo.config.suppress_errors = True
+torch._dynamo.disable()
+
 max_seq_length = 2048
 dtype = (
     None
 )
 load_in_4bit = True
 
-quantization_config = BitsAndBytesConfig(load_in_4bit=True,
-                                         bnb_4bit_compute_dtype=torch.float16)
+
+quantization_config = BitsAndBytesConfig(
+    load_in_4bit=True,
+    # bnb_4bit_compute_dtype=torch.float16,
+)
 
 tokenizer = AutoTokenizer.from_pretrained("ua-l/gemma-2-9b-legal-uk")
 model = AutoPeftModelForCausalLM.from_pretrained(