Update README.md

README.md

@@ -17,8 +17,8 @@ A LoRA (Low-Rank Adaptation) decomposed from base model from base to instruct
 
 ## Training Configuration
 ### Training Hyperparameters
-- Rank (r): 16
-- Alpha: 16
+- Rank (r): 16 -> 16
+- Alpha: 1 -> 16
 
 ### LoRA Configuration
 - Rank (r): 16