Update app.py

app.py

@@ -7,7 +7,7 @@ model = AutoModelForCausalLM.from_pretrained("LingoIITGN/ganga-1b")
 
 @spaces.GPU
 def greet(input_text):
-    input_token = tokenizer.encode(input_text, return_tensors="pt").to("cuda")
+    input_token = tokenizer.encode(input_text, return_tensors="pt")
     output = model.generate(input_token, max_new_tokens=100, num_return_sequences=1, do_sample=True, top_k=50, top_p=0.95, temperature=0.7)
     output_text = tokenizer.batch_decode(output)[0]
     return output_text