updated

docs/8. Quantized-Low Rank Adaptation(Qlora).md

@@ -50,7 +50,7 @@ When we **quantize** the model, we convert the floating-point weights from a hig
   Example:
   - After scaling, **-3.75** might be quantized to an integer, say **8**, which fits into the 4-bit integer range [0, 15].
 
-### Steps in the Quantization Process:
+### Steps in the Quantization Process: (maybe using bitsandbytes library, we use)
 
 1. **Scale the Values**:  
    The range of the weights (e.g., from -5.0 to 5.0) is scaled to fit within the range of the target precision (e.g., 0 to 15 for 4-bit precision).