diff --git a/nanochat/engine.py b/nanochat/engine.py
index 53fdec5..7976042 100644
--- a/nanochat/engine.py
+++ b/nanochat/engine.py
@@ -312,6 +312,7 @@ if __name__ == "__main__":
     generated_tokens = []
     torch.cuda.synchronize()
     t0 = time.time()
+    print(tokenizer.decode(prompt_tokens))
     stream = model.generate(prompt_tokens, **kwargs)
     with autocast_ctx:
         for token in stream:
@@ -326,6 +327,7 @@ if __name__ == "__main__":
     # generate tokens with Engine
     generated_tokens = []
     engine = Engine(model, tokenizer)
+    print(tokenizer.decode(prompt_tokens))
     stream = engine.generate(prompt_tokens, num_samples=1, **kwargs) # note: runs in fp32
     torch.cuda.synchronize()
     t0 = time.time()