Update README.md

README.md

@@ -12,7 +12,7 @@ MIDI model (RWKV-4 sounds better):
 
 Training data: https://huggingface.co/datasets/breadlicker45/bread-midi-dataset
 
-ABC model (RWKV-5 sounds better):
+ABC model (RWKV-5 sounds better if you use topp=0.5 without topk):
 
 Training data: ```load_dataset("sander-wood/massive_abcnotation_dataset")["train"]```