Update app.py

app.py

@@ -9,9 +9,15 @@ login(token=hf_token)
 
 model_name = "meta-llama/Llama-3.2-3B-Instruct"
 
-# Load the tokenizer and model with the token
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name)
+rope_scaling = {
+    "type": "llama3",  # or another valid type
+    "factor": 32.0  # your scaling factor, can be adjusted based on needs
+}
+
+# Ensure the model loading process uses the corrected `rope_scaling`
+tokenizer = AutoTokenizer.from_pretrained(model_name, rope_scaling=rope_scaling)
+model = AutoModelForCausalLM.from_pretrained(model_name, rope_scaling=rope_scaling)
+
 
 # Define the generator function using the LLaMA model
 generator = pipeline("text-generation", model=model, tokenizer=tokenizer)