Update README.md

README.md

@@ -64,6 +64,9 @@ decoded_sequences = tokenizer.batch_decode(outputs, skip_special_tokens=True)
 
 # Print the decoded sequences
 print(decoded_sequences)
+
+# It is expected to observe non-sense decoded sequences (e.g., 'AAAAAA')
+# The input sequences are too short to provide sufficient context.
 ```
 
 ### Simple example2: embedding